THE ELEVENTH

MARCEL GROSSMANN MEETING

On Recent Developments in Theoretical and Experimental
General Relativity, Gravitation and Relativistic Field Theories
Also published by World Scientific:

PROCEEDINGS OF THE SIXTH

MARCEL GROSSMANN MEETING ON GENERAL RELATIVITY
PART A & PART B
Eds. Humitaka Sato and Takashi Nakamura
Series Ed. Remo Ruffini

PROCEEDINGS OF THE SEVENTH

MARCEL GROSSMANN MEETING ON GENERAL RELATIVITY
PART A & PART B
Eds. Robert T. Jantzen and G. Mac Keiser
Series Ed. Remo Ruffini

PROCEEDINGS OF THE EIGHTH

MARCEL GROSSMANN MEETING ON GENERAL RELATIVITY
PART A & PART B
Ed. Tsvi Piran
Series Ed. Remo Ruffini

PROCEEDINGS OF THE NINTH

MARCEL GROSSMANN MEETING ON GENERAL RELATIVITY
PART A, PART B & PART C
Eds. Vahe G. Gurzadyan and Robert T. Jantzen
Series Ed. Remo Ruffini

PROCEEDINGS OF THE TENTH

MARCEL GROSSMANN MEETING ON GENERAL RELATIVITY
PART A, PART B & PART C
Eds. Mario Novello and Santiago Perez Berg/iaffa
Series Ed. Remo Ruffini
 PARTS

THE ELEVENTH
MARCEL GROSSMANN MEETING
On Recent Developments in Theoretical and Experimental
General Relativity, Gravitation and Relativistic Field Theories

Proceedings of the MG11 Meeting

on General Relativity
Berlin, Germany 23-29 July 2006

Editors
Hagen Kleinert
Freie Universitat Berlin, Germany

Robert T Jantzen
Villanova University, USA

Series Editor
Remo Ruffini
University of Rome "La Sapienza"
Rome, Italy

1:0 World Scientific

NEW JERSEY· LONDON· SINGAPORE· BEIJING· SHANGHAI· HONG KONG· TAIPEI· CHENNAI
Published by
World Scientific Publishing Co. Pte. Ltd.
5 Toh Tuck Link, Singapore 596224
USA office: 27 Warren Street, Suite 401-402, Hackensack, NJ 07601
UK office: 57 Shelton Street, Covent Garden, London WC2H 9HE

British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library.

THE ELEVENTH MARCEL GROSSMANN MEETING (In 3 Volumes)

On Recent Developments in Theoretical and Experimental General Relativity,
Gravitation and Relativistic Field Theories

Copyright © 2008 by World Scientific Publishing Co. Pte. Ltd.

All rights reserved. This book, or parts thereof, may not be reproduced in any form or by any means, electronic or
mechanical, including photocopying, recording or any information storage and retrieval system now known or to
be invented, without written permission from the Publisher.

For photocopying of material in this volume, please pay a copying fee through the Copyright Clearance Center,
Inc., 222 Rosewood Drive, Danvers, MA 01923, USA. In this case permission to photocopy is not required from
the publisher.

ISBN-13 978-981-283-426-3 (Set)

ISBN-IO 981-283-426-5 (Set)
ISBN-13 978-981-283-427-0 (Vol. I)
ISBN-I0 981-283-427-3 (Vol. I)
ISBN-13 978-981-283-428-7 (Vol. 2)
ISBN-IO 981-283-428-1 (Vol. 2)
ISBN-13 978-981-283-429-4 (Vol. 3)
ISBN-IO 981-283-429-X (Vol. 3)

Printed in Singapore by Mainland Press Pte Ltd

 THE MARCEL GROSSMANN MEETINGS
The Marcel Grossmann Meetings were conceived with the aim of reviewing recent
developments in gravitation and general relativity, with major emphasis on math-
ematical foundations and physical predictions. Their main objective is to bring
together scientists from diverse backgrounds in order to deepen our understand-
ing of spacetime structure and review the status of experiments testing Einstein's
theory of gravitation.

Publications in the Series of Proceedings

Proceedings of the Eleventh Marcel Grossmann Meeting on General Relativity
(these volumes)
(Berlin, Germany, 2006)
Edited by H. Kleinert, R.T. Jantzen, R. Ruffini
World Scientific, 2008

Proceedings of the Tenth Marcel Grossmann Meeting on General Relativity

(Rio de Janiero, Brazil, 2003)
Edited by M. Novello, S. Perez-Bergliaffa, R. Ruffini
World Scientific, 2005

Proceedings of the Ninth Marcel Grossmann Meeting on General Relativity

(Rome, Italy, 2000)
Edited by V.G. Gurzadyan, R.T. Jantzen, R. Ruffini
World Scientific, 2002

Proceedings of the Eighth Marcel Grossmann Meeting on General Relativity

(Jerusalem, Israel, 1997)
Edited by T. Piran
World Scientific, 1998

Proceedings of the Seventh Marcel Grossmann Meeting on General Relativity

(Stanford, USA, 1994)
Edited by R.T. Jantzen and G.M. Keiser
World Scientific, 1996

Proceedings of the Sixth Marcel Grossmann Meeting on General Relativity

(Kyoto, Japan, 1991)
Edited by H. Sato and T. Nakamura
World Scientific, 1992

v
vi

Proceedings of the Fifth Marcel Grossmann Meeting on General Relativity

(Perth, Australia, 1988)
Edited by D.G. Blair and M.J. Buckingham
World Scientific, 1989

Proceedings of the Fourth Marcel Grossmann Meeting on General Relativity

(Rome, Italy, 1985)
Edited by R. Ruffini
World Scientific, 1986

Proceedings of the Third Marcel Grossmann Meeting on General Relativity

(Shanghai, People's Republic of China, 1982)
Edited by Hu Ning
Science Press - Beijing and North-Holland Publishing Company, 1983

Proceedings of the Second Marcel Grossmann Meeting on General Relativity

(Trieste, Italy, 1979)
Edited by R. Ruffini
North-Holland Publishing Company, 1982

Proceedings of the First Marcel Grossmann Meeting on General Relativity

(Trieste, Italy, 1976)
Edited by R. Ruffini
North-Holland Publishing Company, 1977

Series Editor: REMO RUFFINI

SPONSORS

Free University Berlin (FUB)

The German Research Foundation (DFG)
The German Academic Exchange Service (DAAD)
The Frankfurt Institute for Advanced Studies (FIAS)
The Frankfurt Institute for Advanced Studies (FIAS)
The International Union of Pure and Applied Physics (IUPAP)
The Center of Applied Space Technology and Microgravity (ZARM)
The Wilhelm and Else Heraus Foundation
International Center for Theoretical Physics (ICTP)
MPI for Gravitational Physics (Albert Einstein Institute)
MPI for Extraterrestrial Physics
 ORGANIZING BODIES OF THE
ELEVENTH MARCEL GROSSMANN MEETING:
INTERNATIONAL ORGANIZING COMMITTEE
David Blair, Yvonne Choquet-Bruhat, Demetrios Christodoulou, Thibault Damour,
Jurgen Ehlers, Francis Everitt, Fang Li-Zhi, Stephen Hawking, Yuval Ne'eman
Remo Ruffini (Chair), Humitaka Sato, Rashid Sunayev, Steven Weinberg

LOCAL ORGANIZING COMMITTEE

Bernd Brugmann, A. Chervyakov, Hansjorg Dittus, W. Janke, Hagen Kleinert
(chair), Jutta Kunz, Claus Laemmerzahl, Flavio Nogueira, Axel Pelster, Luciano
Rezzolla, Erwin Sedlmayr, Stefan Theisen, Thomas Thiemann

INTERNATIONAL COORDINATING COMMITTEE

Bob Jantzen (chair)

ALBANIA: Hafizi M., ARGENTINA: Jakubi A.S., Mirabel F., Nunez C.A., AR-
MENIA: Gurzadyan V., AUSTRALIA: Lun A., Manchester D., Scott S.M., Steele
J.D., Veitch P., AUSTRIA: Aichelburg P.C., Schindler S., BELGIUM: Henneaux
M., Surdej J., BELORUSSIA: Minkevich A.V., BOLIVIA: Aguirre C.B., BRAZIL:
Aguiar 0., Aldrovandi R., Novello M., Opher R, Perez Bergliaffa S.E., Villela T.,
CANADA: Cooperstock F., Page D.N., Papini G., Smolin L., CHILE: Bunster
Weitzman C., CHINA (Beijing): Feng L.-L., Gao J.-G., Lee D.-S, Lee W.OL., Li
M., Ni W.-T., Wu X.-P., Yip eng J., CHINA (Taepei): Lee D.S., Lee W.L., Ni W.T.,
COLOMBIA: Sepulveda H.A., Torres S., CROATIA: Milekovic M., CUBA: Quiros
1., CZECK REPUBLIC: Bicak J., DENMARK: Novikov I., EGYPT: Wanas M.L,
ESTONIA: Einasto J., FRANCE: Brillet A., Chardonnet P., Conllet P., de Fre-
itas Pacheco J.A., Deruelle N., Iliopoulos J., Mignard, F., GEORGIA: Lavrelashvili
G., GERMANY: Biermann P., DallZmann K, Fritzsch H., Genzel R, Greiner W.,
Hasinger G., Hehl F., Kiefer C., Neugebauer G., Nicolai H., Renn J., Ringwald
A., Ruediger A., Schutz B., GREECE: Batakis N., Cotsakis S., HUNGARY: Fodor
G., ICELAND: Bjornsson G., INDIA: Narlikar J., Sahni V., Vishveshwara C.V.,
IRAN: Mansouri R, Sobonti, Y, IRELAND: O'Murchada N., ISRAEL: Piran T.,
Sobouti, Y, ITALY: Belinsky V., Bianchi M., Ciufolini 1., Menotti P., Regge T.,
Stella L., Treves A., JAPAN: Fujimoto M.K, Makino J., Nakamura T., Sasaki M.,
Sato K, Tomimatsu A., KAZACHSTAN: Abdildin A.M., Mychelkin E.G., KOREA
(Pyeongyang): Kim J.S., Kim YG., KOREA (SeOUl): Lee Chul H., Lee Hyung W.,
Song .long D., KYRGYZSTAN: Gurovich V.Ts., LIBYA: Gadri M., LITVA: Piragas
KA., MEXICO: Garcia-Diaz A.A., Macias-Alvarez A., Mielke E.W., Rosenbaum
M., Ryan M.P., NETHERLANDS: 't Hooft G., NEW ZEALAND: Visser M., Wilt-
shire D., NORWAY: Knutsen H., POLAND: Demianski M., Nurowski P., Sokolowski

vii
viii

L., PORTUGAL: Costa M., Vargas Moniz P., ROMANIA: Visinescu M., RUS-
SIA: Bisnovatyi-Kogan G.S., Blinnikov, S., Chechetikin V.M., Cherepaschuk A.M.,
Khriplovich I.B., Kotov Y., Lipunov V.M., Lukash V., Melnikov V., Rudenko V.,
Starobinsky A.A., Tchetchetkine V. M., SERBIA: Sijacki D., SLOVENIA: Cadez
A., SOUTH AFRICA: Maharaj S., SPAIN: Ibanez J., Perez Mercader J., Verda-
guer E., SWEDEN: Marklund M., Rosquist K, SWITZERLAND: Durrer R, Jet-
zer P., TURKEY: Nutku Y., UK: Barrow J., Cruise A.M., Green M., Kibble T.,
Maartens R, USA: Ashtekar A., Bardeen J., Barish B., Chen P., Cornish N., Der-
mer C., DeWitt-Marette C., Drever R, Finkelstein D., Halpern L., Hellings RW.,
Jantzen RT., Klauder J., Kolb R, Lousto C., Mashhoon B., Matzner R, Melia F.,
Nordtvedt K, Parker L., Pullin J. Schwarz J., Shapiro I., Shoemaker D., Smoot
G., Thorne KS., van Nieuwenhuizen P., York J.W. Jr., UZBEKISTAN: Zalaletdi-
nov RM., VATICAN CITY: Stoeger W., VENEZUELA: Herrera L., Percoco U.,
VIETNAM: van Hieu N.

ACKNOWLEDGMENTS

We acknowledge the help of the following individuals before, during and after the
actual meeting itself: Michael Kleinert (meeting webmaster and local IT organizer),
Annemarie Kleinert (chief local organizer and finance manager), Flavio Nogueira
(local meeting point man), and the staff, students and post docs of Hagen Kleinert's
research group, and the ICRANet/ICRA secretarial support: Federica Di Berardino,
Veronica D'Angelo, Gilda Massa, Cesare Corsetti. We also acknowledge the gener-
ous assistance of the Italian Foreign Ministry and in particular of the Science and
Technology Attache of the Italian Embassy in Berlin Prof. Vincenzo Tov!.
In an age of increasing technological sophistication, this meeting could not have
functioned without the tireless dedication of ICRA system manager Vittorio Van-
nini, nor could these proceedings have been possible without his patient manage-
ment of the email and web communication and data handling necessary to produce
them. We also recognize the past contributions of the late system ICRA co-system
manager Maurizio Cosma whose friendship and valuable contributions to past MG
Meetings should not go unrecognized.
Finally we acknowledge the loss of our friend Leopold Halpern, a physicist,
humanitarian, environmentalist, naturalist, warld traveler and participant in every
MG Meeting whose advice to a young physicist (Remo Ruffini) at a key moment
influenced his choice to enter the field of general relativity and later cofound this
Meeting series.
 MARCEL GROSSMANN AWARDS

ELEVENTH MARCEL GROSSMANN MEETING

Institutional Award
Freie Universitiit Berlin

"for the successful endeavour of re-establishing

- in the spirit of the Humboldt tradition -
freedom of thinking and teaching within a democratic society
in a rapidly evolving cosmos"
-presented to Dr. Dieter Lenzen, President of FUB

Individual Awards
Roy Kerr

"for his fundamental contribution to Einstein's theory of

general relativity: The gravitational field of a
spinning mass as an example of algebraically special metrics"

George Coyne

"for' his committed support for the international development of

relativistic astrophysics and for his dedication to fostering
an enlightened relationship between science and religion"

Joachim Trumper

"for his outstanding scientific contributions to the physics of compact

astrophysical objects and for leading the highly successful ROBAT mission
which discovered more then 200,000 galactic and extragalactic X-ray sources:
a major step in the observational capabilities of X-ray astronomy
and in the knowledge of our universe"

Each recipient is presented with a silver casting of the TEST sculpture by the
artist A. Picrelli. The original casting was presented to His Holiness Pope John Paul
II on the first occasion of the Marcel Grossmann Awards.

ix
x

TENTH MARCEL GROSSMANN MEETING

Institutional Award
CBPF (BRAZILIAN CENTER FOR RESEARCH IN PHYSICS)
Individual Awards
YVONNE CHOQUET-BRUHAT, JAMES W. YORK, JR., YUVAL NE'EMAN

NINTH MARCEL GROSSMANN MEETING

Institutional Award
THE SOLVAY INSTITUTES
Individual Awards
RICCARDO GIACCONI, ROGER PENROSE

EIGHTH MARCEL GROSSMANN MEETING

Institutional Award
THE HEBREW UNIVERSITY OF JERUSALEM
Individual Awards
TULLIO REGGE, FRANCIS EVERITT

SEVENTH MARCEL GROSSMANN MEETING

Institutional Award
THE HUBBLE SPACE TELESCOPE INSTITUTE
Individual Awards
SUBRAHMANYAN CHANDRASEKHAR, JIM WILSON

SIXTH MARCEL GROSSMANN MEETING

Institutional Award
RESEARCH INSTITUTE FOR THEORETICAL PHYSICS (Hiroshima)
Individual Awards
MINORU ODA, STEPHEN HAWKING
 xi

FIFTH MARCEL GROSSMANN MEETING

Institutional Award
THE UNIVERSITY OF WESTERN AUSTRALIA
Individual Awards
SATIO HAYAKAWA, JOHN ARCHIBALD WHEELER

FOURTH MARCEL GROSSMANN MEETING

Institutional Award
THE VATICAN OBSERVATORY
Individual Awards
WILLIAM FAIRBANK, ABDUS SALAM
xii

TEST: sculpture by A. Pierelli.

 PREFACE

The Eleventh Marcel Grossmann Meeting on General Relativity (MGll) took place
during July 2:3~29, 2006 on the Campus of the Freie Universitat Berlin, an attractive
location for both practical and historical reasons. It is situated in the park-like
district of Berlin-Dahlem, where many famous German researchers of the early
20th century lived and worked, among them Planck and Einstein (Fig. 1). The
conference site lies close to the former Kaiser-Wilhelm-Institute of Physics where
Hahn, Meitner, and Strassmann discovered the fission of uranium in 19:)8 (Fig. 2).

Pig. 1 Fig. 2

Pig. 3

Otto Hahn's house is just around the corner. So is Einstein's apartment in Ehren-
bergstrasse 33 where he lived after moving from Zrich in 1914 (Fig. 3, with zoomed
bronze memorial plate at the entrance). Around 800 participants and accompany-

xiii
xiv

Fig. 4 Fig. 5

ing persons were present during a week of exceptionally warm summer weather in
Berlin. The meeting began with the Marcel Grossmann Awards ceremony on July
23. The institutional award went to Freie Universitt (FU) Berlin (Fig. 4) "for the
successful endeavor of re-establishing in the spirit of the Humboldt tradition -
freedom of thinking and teaching within a democratic society in a rapidly evolving
cosmos". Remo Ruffini handed the award to Dieter Lenzen, president of the FU
Berlin (Figs. 4 and 5).
Three individual awards were presented to Roy Kerr "for his fundamental contri-
bution to Einstein's theory of general relativity: The gravitational field of a spinning
mass as an example of algebraically special metrics" .
Three individual awards were presented to

Roy Kerr "for his fundamental contribution to Einstein's

theory of general relativity: The gravitational field of a
spinning mass as an example of algebraically special met-
rics" .

George Coyne "for his committed support for the interna-

tional development of relativistic astrophysics and for his
dedication to fostering an enlightened relationship between
science and religion" .
 xv

Joachim Triimper "for his outstanding scientific contribu-

tions to the physics of compact astrophysical objects and
for the leading successful ROSAT mission which discov-
ered more than 200,000 galactic and extragalactic X-ray
sources: a major step in the observational capabilities of
X-ray astronomy and in the knowledge of our universe".

Each laureate received a silver casting of the TEST sculpture by the artist
A. Pierelli. The original casting was presented on the first occasion of the Marcel
Grossmann Award to His Holiness Pope John Paul II.
After the prize ceremony the plenary program started with lectures by:
Thibault Damour (IHES, Bures-sur-Yvette) "Cosmology and string theory"
Sasha Polyakov (Princeton University) "The structure beyond spacetime"
Hermann Nicolai (Albert-Einstein-Inst. Potsdam) "Hidden symmetries and cos-
mological singularities"
They were continued each moruing from Tuesday to Saturday with the following
speakers:
Claes U ggla (Karlstaads University) "The nature of generic cosmological singu-
larities"
Eva Silverstein (Stanford University) "Cosmological singularities in string theory"
Igor Klebanov (Princeton University) "Gauge theories, strings and cosmology"
Joe Polchinski (UC Santa Barbara) "Cosmic superstrings"
Abhay Ashtekar (Pennsylvania State University) "Loop quantum gravity"
Dieter Luest (Humboldt Univ., Berlin) "String theory and the standard model of
particle physics"
Karsten Danzmann (Univ. Hannover) "LISA"
Marie Anne Bizouard (Univ. Paris XI) "VIRGO"
David Shoemaker (MIT) "LIGO: Status of instruments and observations"
Alessandra Buonanno (Univ. of Maryland) "Analytical approach to coalescing
binary black holes"
Francois Mignard (Observatoire Cote d'Azur) "Relativistic effects from HIPPAR-
COS and GAIA missions"
Michael Kramer (Univ. of Manchester) "Binary pulsars and general relativistic
effects"
Josh Grindlay (Harvard Univ.) "Globular clusters and millisecond pulsars"
xvi

Richard Mushotzky (NASA Goddard SFC) "Intermediate mass black holes and
X-ray sources"
Rashid Sunyaev (MPA Garching) "The sky in the hard x-ray spectrum"
Reinhard Genzel (MPE Garching) "The black hole in our galactic center"
George Djorgovski (CALTECH), "The origins of massive black holes and quasars
at high redshifts"
ReIllo Ruffini (ICRA, Roma) "Gamma ray bursts"
Francis Halzen (University of Wisconsin-Madison) "ICE CUBE"
Peter BierIllann (MPI for Radioastronomy, Bonn) "Sterile neutrinos in astro-
physics and cosmology"
Volker Springel (MPI for Astrophysics Garching) "Simulations of the formation,
evolution and clustering of galaxies and quasars"
Paolo De Bernardis (Univ. Roma La Sapienza) "CMB science from Boomerang
to PLANCK"
David Spergel (Princeton Center for Theoretical Physics) "WMAP and its cos-
mological implications"
Ethan J. Schreier (AUI, Washington, DC) "ALMA"
John Mester (Stanford University) "Equivalence principle from space"
Francis Everitt (Stanford University) "The NASA Gravity Probe B Mission: tech-
nical report"
Guy Monnet (Europ. South. Observatory, Garching) "Science and technology of
the European ELT"
Michael Garcia (Harvard-Smithsonian Ctr. for Astroph.) "Science from Chandra
to Constellation-X"
Nicholas White (HEASARC) "Beyond Einstein: from the big bang to black holes"
Theodor Haensch (Ludwig-Maximilian Univ. Munchen) "Precise clocks"
Juergen Renn (MPI for the History of Science, Berlin) "The genesis of general
relativity"

On Monday, Tuesday, Thursday, and Friday, public lectures were presented by

Hanns Ruder (University Tubingen) "Visualizations of relativistic effects"
Gunter Hasinger (MPE Garching) "The fate of the universe - new clues from
cosmology"
Bruno Leibundgut (Eur. Southern. Obs., Garching) "Das neue Weltbild der Kos-
mologie - Was ist Dunkle Energie?"
Christian Spiering (DESY Zeuthen) "Neutrinoastronomic- ein neues Fenster
zum Kosmos"

These lectures were well attended by Berlin citizens and conference participants
and found broad resonance in the media. Parallel sessions were held on the af-
ternoons in 20 lecture halls. Some 850 scientific papers were presented during 82
parallel sessions over four afternoons. A typical setting in front of one of the lecture
halls is shown below.
 xvii

Many speakers at MGll were accommodated in the famous Harnack House, a

place where much of the "Dahlem Legend" happened. The house was built during
the Weimar republic by the theologian Adolf von Harnack, the first president of
the Kaiser-Wilhelm Society. Many German Nobel Prize winners and their students
met here for social interaction and academic discussion. Here they held lectures and
colloquia, took lunch together, read the new international press, drank coffee in the
garden, engaged in sports, and played music. The list of former guests and lecturers
reads like a "Who's Who of Science": Albert Einstein, Peter Debye, Werner Heisen-
berg, .Fritz Haber, Adolf I3utenandt, Otto Hahn, Lise Meitner, Otto Meyerhof, Max
Planck, Max von Laue and Otto Warburg. One Nobel Prize winner, the biologist
Hans Fischer, even received the news of his award during his stay at the Harnack
xviii

House. Also great non-scientists stayed at this house, for instance Ricarda Huch,
the Swiss art historian Heinrich Wolffiin, and the Indian philosopher Rabindranath
Tagore. In 1935, in direct opposition to the government, Max Planck led an impres-
sive commemoration of Fritz Haber here. The Kaiser-Wilhelm Institutes were later
re-organized and renamed as the Max Planck Institutes.
During MG 11, a big beer tent was set up in the courtyard of the physics depart-
ment in the style of the famous Munich Oktoberfest, which was well frequented by
all participants since its informal atmosphere was very beneficial for social interac-
tions and the exchange of ideas.
A video stream exchange was set up with the Einstein Institute in Potsdam
so that its members were able to follow the Marcel Grossmann lectures and the
participants in Berlin could listen to lectures at the Einstein Institute if desired.
The opulent MG 11 conference banquet dinner was held at the Ritz Carlton Hotel
next to Potsdamer Platz. A Pruss ian 19th century type brass orchestra was there
to play music from the emperor's time.

On July 29 Remo Ruffini closed the meeting thanking all the speakers and
participants and sponsoring institutions.
These three volumes represent the proceedings of the meeting. The first vol-
ume contains articles by many of the plenary speakers together with some of the
review articles from the parallel sessions. The second and third volumes contain
the remaining contributions from the parallel sessions. The participant list and the
author index complete the third volume.
 INAUGURAL ADDRESS
Dear Mr. Ruffini,
Dear Mr. Sreenivasan,
Dear Mr. Umbach,
My Dear Colleague Mr. Kleinert,
Ladies and Gentlemen,
Honored Guests,

On the occasion of this year's Marcel Grossmann Conference at the Freie Uni-
versitat Berlin, it is a special honor for me to welcome all of you here to Dahlem,
one of Berlin's largest and most important centers of science and scholarship.
To honor the epochal achievements of Albert Einstein, who worked in Berlin-
Dahlem for nearly two decades as director of the Kaiser Wilhelm Institute for
Physics, is a central concern of the Freie U niversitat Berlin. Albert Einsteins time in
Berlin saw the emergence of contributions to physics that were so outstanding that
they have continued to be a source of fascination in the field of physics and beyond.
This morning, exceptional scientists received the distinction of the Marcel Gross-
mann Prize. The Freie Universitat Berlin too will be distinguished as an institution
that has rendered extraordinary services to unfettered science and scholarship ever
since its foundation. It is with great pleasure that I receive this honor in the name of
the Freie Universitat Berlin, for like virtually no other German institution of higher
learning, our university is closely associated with the concept of "freedom".
The establishment of this university in 1948 can be traced back to the struggle
for academic freedom. The impetus for its foundation emanated primarily from stu-
dents who--after bitter experiences with the National Socialist dictatorship-were
committed to freedom and democracy, and who rejected the relegation of Students
of Berlins Humboldt University in accordance with the worldview prevailing in the
East.
Through international material assistance, the university's members and numer-
ous sponsors among Berlin's citizens saw to it that students from the surrounding
regions and from Berlins eastern sector who had been refused the opportunity to
study for political reasons were able to complete their training at the Freie Univer-
sitat. All of this transpired against a backdrop of escalating conflict between the
Western allies and the Soviet Union concerning Europe's future political organiza-
tion. The founding of the Freie Universitat proceeded in the middle of the Soviet
blockade of West Berlin, which sealed the city off from the outside world from June
1948 to May 1949. Freie Universitat survived even the Berlin Blockade, because
international aid arrived, especially from the United States. Until 1961, when the
partition of the city was cemented by the erection of the Berlin Wall, the uni-
versity's founders had succeeded in establishing firm and supportive international
networks that won Freie Universitat a recognised position among German and in-

xix
xx

ternational universities. We continue to benefit from these international networks

today. The crimes of German fascism only began to be examined in earnest in 1968.
The student movement, which largely originated at Freie Universitiit, was also a
response to this need for unfettered scrutiny of our past. In the following years, the
traditional university under professorial governance was replaced by more account-
able structures, in which all members of the university are represented in university
governance. Our university continued to playa vital role after the breakdown of
the communist GDR and in the course of German reunification. The key challenge
then was to rebuild the universities that had lived under communist dictatorship
by giving them both financial and intellectual support. In the course of this internal
reform, our university was even able to raise its performance by some ten per cent
per annum since 2000. Now that this reform has been completed, we look forward
to addressing the universitys strategic globalization, based on its long tradition of
international networking. It is no coincidence that the Freie Universitiit is delighted
to host a series of events such as the Marcel Grossmann Conference, for intensive
exchange between the sciences has been a key priority of the Freie Universitat Berlin
since its foundation here in Dahlem in 1948. And here in particular, in Dahlem in
the south of Berlin, the Freie Universitiit Berlin perpetuates a scientific tradition
that provides ideal preconditions for an indispensable ingredient of contemporary
scientific and scholarly work: networked, interdisciplinary activities that transcend
subject areas and disciplinary boundaries. In light of this "Dahlem Myth" and of
the tradition of interdisciplinary exchange that is bound up with it, we can only re-
gard our own times-which demand so much readiness for change and reform in the
sphere of education and elsewhere--~as representing a new departure, one we must
use to our advantage, since for all of us, the future lies in science and education.
In this spirit, honored guests, I wish all of you a stimulating time at this year's
Marcel Grossmann Conference. Thanks to the organizing committee, to you, my
dear colleague Mr. Kleinert, and to the Department of Physics. And to all of you,
a warm welcome to the Freie Universitiit Berlin!

Dieter Lenzen
President of Freie Universitiit Berlin
 MARCEL GROSSMANN AWARD ESSAY
George Coyne, S.J.
Director Emeritus of the Vatican Observatory

I was deeply honored to have received a Marcel Grossman Award at the July
2006 meeting in Berlin, a city so rife with memories of discoveries in physics. The
citation noted my interest in the relationship between science and religion. Even
these few years since that meeting have seen many interesting developments in that
relationship. In fact, most recently some have even arrived at seriously posing the
question: Is God a mathematician? The background to that question harks back to
Albert Einstein's comment: "The most incomprehensible thing about the universe
is that it is comprehensible." But, in what way is it comprehensible? Here enters
the question as to God and mathematics. This question is, I think, at the core of
the intersection of the two cultures of science and religion in today's world.
Let us begin by marveling, as many others, including Einstein, did, that the
universe is comprehensible. In fact, I have co-authored with Michael Heller a book
entitled: "A Comprehensible Universe" (Springer Verlag, in press). \Ve see the com-
prehensibility of the universe as due to its mathematical structure. One can chal-
lenge the notion that physics is limited to the investigation of matter. In fact, in
much of the research in physics emphasis is placed on the fact that physics con-
structs mathematical models of the world and then confronts them with empirical
results. And such an approach has had an astonishing success because, indeed, the
world has a mathematical structure to it. And who set up that structure?
Science itself cannot find the WHO? But, that mathematical structure can serve
as an enticement, an invitation to go beyond the strict methodology of science to
the ultimate question: WHO? But let us look more closely at the concept of the
mathematical structures of the universe, which provide its comprehensibility and,
ultimately, the invitation to approach the WHO. At the birth of modern science
there was the persistent idea, as there had been for the Pythagoreans, that physi-
cists were discovering some grand transcendental design incarnate in the universe.
As to religious insights, the concept in St. John's Gospel of the logos becoming in-
carnate was particularly appropriate and hailed back in some way to Platonic and
Pythagorean concepts of the world of eternal ideas and of the transcendental char-
acter of mathematics. Indeed, Newton, Descartes, Kepler and others can be cited
as viewing physics and mathematics in this way. Kepler for instance, saw geometry
as providing God with a model for creation. He went so far as to see the circle as
transcendentally perfect, the straight line as the totally created and incarnate and
the ellipse as a combination of the two, an incarnation in this world of what would
have been the perfect geometry for the motion of the heavenly bodies in an ideal
world. The simple equations in which Newton expressed the law of gravity and the
laws of motion redirected for future centuries the role of mathematics in physics.
No longer was mathematics simply a description of what was observed; it was a

xxi
xxii

probe of the very nature of what was observed. This role of mathematics was only
enhanced as relativity theory, quantum mechanics and then quantum cosmology
came on the scene.
Leibniz once claimed that "When God calculates and thinks things through, the
world is made." Things thought through by God might be identified with mathe-
matical structures interpreted as structures of the visible universe. For God to plan
is the same as to implement the plan and thus to create. God has planned and,
thereby, created a structured world which participates, through the subtle random
events intrinsic to the structure, in the very creativity of God.
Will we eventually understand comprehensively the structure of the universe
and, therefore, the mind of the mathematician God? I suggest a definitive no. God
is mystery and the source of all that is mysterious in the universe. The search for
the ultimate mathematical structure is unending and that is what makes the search
being carried on by many scholars such a passionate adventure.
 CONTENTS

The Marcel Grossmann Meetings - Publications in this Series and Sponsors v

Organizing Committees Vll

Marcel Grossmann Awards ix

Preface xiii
Inaugural Address xix
Marcel Grossmann Award Essay XXI

PART A
PLENARY AND REVIEW TALKS

A Brief History of X-Ray Astronomy in Germany

Truemper, Joachim E. . . . . . . 3
On the Discovery of the Kerr metric
Kerr', Roy Patrick. . . . . . . . 9
Chaos and Symmetry in String Cosmology
DamolLr, ThibalLlt . . . . . . . . . . . . . . . . . 39
Hidden Symmetries, Cosmological Singularities and the
E lO / K(EIO) Sigma Model
Kleinschmidt, Axel; Nicolai, Herman . . . . . . . . . . . 49
The Nature of Generic Cosmological Singularities
Uggla, Claes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
QCD and String Theory
Klebanov, Igor R. . . . . . . . . . . . . . . . . . . . . . . 90
The Cosmic String Inverse Problem
Polchinski, Joseph . . . . . 105
Loop Quantum Gravity: Four Recent Advances and a Dozen
Frequently Asked Questions
Ashtekar, Abhay . . . . . 126
String Theory Landscape and the Standard Model of Particle
Physics
Lust, Dieter 148
The Status of the Virgo Gravitational Wave Detector
BizolLard, Marie-Anne; for the Virgo collaboration. . 177

xxiii
xxiv

Analytical Modeling of Binary Black Holes Coalescence

Buonanno, Alessandra. . . . . . . . . . . 197
Binary Pulsars and General Relativistic Effects
Kramer, Michael . . . . . . . . . 225
Space Astronometry and Relativity
Mignard, Fran<;ois; Klioner, Sergei A. 245
Neutrino Astronomy 2006
Halzen, Francis. . . . . 272
Dark Matter and Sterile Neutrinos
Biermann, Peter L.; lvlunyaneza, Faustin 291
Supercomputer Simulations of the Joint Formation and Evolution
of Galaxies and Quasars
Springel, Volker. . . . . . . . . . . . . . . . . . . . . . 309
CMB Observations: From BOOMERanG to Planck ... and Beyond
De Bernardis, Paolo. . . . . . . . . . . . . . . . . . . . 326
The Origins and the Early Evolution of Quasars and Supermassive
Black Holes
Djorgovski, S. George; Volonteri, Marta; Springel, Volker; Bromm,
Volker; Meylan, Georges . 340
On Gamma Ray Bursts
Ruffini, Remo; Bernardini, Maria Grazia; Bianco, Carlo Luciano;
Caito, Letizia; Chardonnet, Pascal; Cherubini, Christian; Dainotti,
Maria Giovanna; Fraschetti, Federico; Geralico, Andrea; Guida,
Roberto; Patricelli, Barbara; Rotondo, Michael; Rueda Hernandez,
Jorge Armando; Vereshchagin, Gregory; Xue, She-Sheng . . . . 368
Passion for Precision
Theodor, Hansch W. 506
The Genesis of General Relativity
Renn, Jiiergen 532
Superposition of Fields of Two Reissner-Nordstrom Sources
Alekseev, George A.; Belinski, V.A. . . . . . . . . . . . 543
Quasiperiodic Oscillations due to Axisymmetric and Non-
axisymmetric Shock Oscillations in Black Hole Accretion
Chakrabarti, Sandip K.; Debnath, D.; Pal, P.S.; Nandi, A.; Sarkar,
R.; Samanta, M.M.; Wiita, P.J.; Ghosh, H.; Som, D . . . . . . 569
Power Spectra of Black Holes (BH) and Neutron Stars (NS) as
a Probe of Hydrodynamical Structure of the Source: Diffusion
Theory and its Application to X-ray Observations of NS and
 xxv

BH Sources
Titarchuk, Lev; Shaposhnikov, Nikolai; Arefiev, Vadim . . . . . 589
Quark Matter in Compact Stars: Astrophysical Implications and
Possible Signatures
Bombaci, Ignazio . . . . . . . . . . 605
Gauge Gravity and Electroweak Theory
Hestenes, David 629
Black Holes in Higher Dimensions (Black Strings and Black Rings)
K unz, Jutta . . . . . . . . . . . . . . . . . . . . . . . 648
Some Remarks on Microlensing Towards LMC and M31
Jetzer, Philippe . . . . . . . . . . . . . . . . . . 663
OGLE-2005-BLG-390Lb - Gravity Reveals First Cool Rocky/Icy
Exoplanet
Dominik, Martin . . . . . . . . . . . . . . . . . . . . 670
Theoretical Gravitational Lensing - Beyond the Weak-Field
Small-Angle Approximation
Perlick, Volker . . . . . . . . . . . . . . . . . . . . . 680
Nonsingular Collapse of Spherically Symmetric Charged Dust
Krasinski, A ndrzej; Bolejko, K rzysztoJ 700
Quantum Cosmology Standpoint
Vargas Moniz, Paulo 708
Gamma Ray Burst Host Galaxies and the Link to Star-Formation
Fynbo, Johan P. U.; Hjorth, Jens; Malesani, Daniele; Sollerman,
Jesper; Watson, Damch J.; Jakobsson, Pall; Gorosabel, Javier;
Jaunsen, Andreas O. 726
Gamma-Ray Bursts with and without Supernova Fireworks
Della Valle, Massimo .............. . 736
Talking about Singularities
Cotsakis, Spiros. . . . . 758
Time Machines and Quantum Theory
Hadley, Mark 1. . . . . . . . . . 778
Slowly and Rigidly Rotating Perfect Fluid Balls of Petrov Type D
Bradley, Michael; Eriksson, Daniel; Fodor, Gyula; Racz, Istvan . 795
Numerical Wave Optics and the Lensing of Gravitational Waves
by Globular Clusters
Moylan, Andrew J.; McClelland, David E.; Scott, Susan M.;
Searle, Antony C.; Bicknell, Geoff V. 807
xxvi

Inflation, Bifurcations of Nonlinear Curvature Lagrangians and

Dark Energy
Mielke, Eckehard W.; Kusmartsev, Fjodor V.; Schunck, Franz E. 824
Virgo Data Analysis for C6 and C7 Engineering Runs
Cuoco, Elena et al. 844
Leopold Ernst Halpern and the Generalization of General
Relativity
Overduin, James M.; Plendl, Hans S. 870
Post-Newtonian Approximations, Compact Binaries, and Strong
Field Tests of Gravity
Blanchet, Luc; Grishchuk, L.P.; SchiiJer, Gerhard. . . . . 881
Tests of Lorentz Symmetry in the Photon Sector
Herrmann, Sven; Senger, Alexander; Moehle, Katharina;
Kovalchuk, Evgeny; Peters, Achim; . . . . . . . . . . . 895
OPTIS - High Precision Test of Special and General Relativity in
Space
Laemmerzahl, Claus; Dittus, Hansjorg; Hackmann, Eva;
Scheithauer, Silvia; Peters, Achim; Schiller, Stephan 905
Testing Special and General Relativity: Clocks and Trajectories
Dittus, Hansjorg; Lammerzahl, Claus; Peters, Achim;
Salomon, Christophe . . . . . . . . . . . . . . . . 916
Laboratory Limits for Temporal Variations of Fundamental
Constants: An Update
Peik, Ekkehard; Lipphardt, Burghard; Schnatz, Harald; Tamm,
Christian; Weyers, Stefan; Wynands, Robert . . . . . . . . 941
Some Old and Some New Opportunities for Quantum Gravity
Phenomenology
Amelino-Camelia, Giovanni 952
Visualization of Relativistic Effects
Ruder, Hanns; Nollert, Hans-Peter; Muller, Thomas; Borchers, Marc . . . 972

PARTB
PARALLEL SESSIONS

• Dark Matter
Chairperson: Biermann, Peter
Impact of Dark Matter on Reionization and Heating
Mapelli, Michela; Ripamonti, Emanuele 979
 xxvii

Impact of Dark Matter Decays and Annihilations on Structure

Formation
Ripamonti, Emanuele; Mapelli, Michela 982
Thermal and Chemical Evolution of the Primordial Clouds in
Warm Dark Matter Models with ke V Sterile Neutrinos in One-
Zone Approximation
Stasielak, Jaroslaw; Biermann, Peter Lo; Kusenko, Alexander 0 0 985
Restrictions on Sterile Neutrino Parameters from Astrophysical
Observations
Ruchayskiy, Oleg 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 988
Upper Limits on Density of Dark Matter in Solar System
Khriplovich, Iosi!; Pitjeva, Elena 0 0 0 0 0 0 0 0 0 0 0 991
The Observed Properties of Dark Matter on Small Astrophysical
Scales
Gilmore, Gerard 994
Is Dark Matter Futile on the Brane?
Gergely, Laszl6 Ao 0 0 0 0 0 0 0 0 997
Direct X-Ray Constraints on Sterile Neutrino Warm Dark Matter
Watson, Casey Ro 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1000
Limits on the Dark Matter Particle Mass from Black Hole Growth
in Galaxies
Munyaneza, Faustin 0 0 0 0 0 • 0 0 0 0 0 0 0 0 0 0 0 0 • 0 0 0 0 0 0 1003
Dark Matter: The Case of Sterile Neutrino
Shaposhnikov, Mikhail 0 0 0 0 • 0 0 0 0 0 • 0 0 0 0 0 0 0 0 0 0 0 1006
• Neutrino Masses: Experimental
Chairperson: Drexlin, Guido
Neutrino Background, Diffuse Backgrounds and CMB: Is the
Picture Consistent?
Popa, Lucia Aurelia; Vasile, Ana 0 0 0 0 0 0 0 0 0 • 0 0 0 • 0 0 0 0 1019
Constraining the Cosmological Lepton Asymmetry through Cosmic
Microwave Background Observations
Lattanzi, Massimiliano; Ruffini, Remo; Vereshchagin, Gregory V. 0 0 0 0 0 1022
Possible Neutrino-Antineutrino Oscillation under Gravity and its
Consequences
Mukhopadhyay, Banibrata o 0 0 0 0 0 1025
How Gravity Can Distinguish Between Dirac and Majorana
Neutrinos
Singh, Dinesh; Mobed, Nader; Papini, Giorgio o 0 0 0 0 0 0 0 0 0 0 1028
xxviii

• Cosmic Rays
Chairperson: Schlickeiser, Reinhard
Anisotropies of Ultra-High Energy Cosmic Rays
Serpico, Pasquale D. . . . . . . . . . . 1033
Recent Progress in Describing Cosmic Ray Transport
Tautz, Robert C. . . . . . . . . . . . . . . . . . . . . . 1036
Propagation of Ultra-High Energy Cosmic Rays: Towards a New
Astronomy
Mattei, Alvise; Chardonnet, P. . . 1039
• Astrophysics of Neutron Stars and Black Holes: Observations
Chairperson: Pian, Elena
Extragalactic X-Ray Jets
Worrall, Diana M. . . . . . . . . . . . . . . . . 1045
Initial Results from the Suzaku Satellite
Dotani, Tadayasu; The Suzaku Team . . . . . . . . . . . . . . . . 1048
Soft Gamma Repeaters and Magnetars
Hurley, Kevin C. . . . . . . . . . . 1051
• Theoretical Models of Observations from Black Hole Candidates
Chairperson: Chakrabarti, San dip K.
Epicyclic Frequencies and Resonant Phenomena Near Black Holes:
The Current Status
Aliev, Alikram N. . . . . . . . . . . . . . . . · . . . . . 1057
Humpy LNRF-Velocity Profiles in Accretion Discs Orbiting
Rapidly Rotating Kerr Black Holes
StuchHk, Zdenek; Slany, Petr; Torok, Gabriel . . . . . . . . . . . 1060
Standing Shocks in Pseudo-Kerr Geometry
Mondal, Soumen; Chakrabarti, Sandip K. · . . . . . 1063
Properties of Accretion Shock Waves in Viscous Flows with
Cooling Effects
Das, Santabrata; Chakrabarti, Sandip K. . . . . · . . . . . 1066
Model of Radiating Annuli near Black Holes for Iron Ko: Line
Profile Interpretations
Zakharov, Alexander F. . . . . . 1069
QPOs due to Centrifugally Supported Shocks around Stellar-Mass
and Supermassive Black Holes
Okuda, Toru; Teresi, Vincenzo; Molteni, Diego . . . . . . . . . . . . . 1072
 xxix

Observing the Flares of Sgr A * with the Very Large Telescope

Interferometer
Paumard, Thibaut; Muller, Thomas; Genzel, Reinhard; Eisen-
hauser, Frank; Gillessen, Stefan. . . . . . . . 1075

Simulating VLBI Images of Sgr A *

Noble, Scott C.; Leung, Po Kin; Gammie, Charles F.; Book, Laura G. . . . 1078
• Astrophysical Black Holes
Chairperson: Chakrabarti, Sandip K.

Astrophysical Black Holes - Do They Have Boundary Layers?

Chakrabar-ti, Sandip K.; Ghosh, Himadri; Som, Deboparn. . . . . . . . . 1085
Secondary Perturbation Effects in Keplerian Accretion Disks:
Elliptical Instability
Mukhopadhyay, Banibrata . . . . . . . . . . . . . . . . . . . . . . 1098
Gravitational Collapse of Population III Stars
Suwa, Yudai; Takiwaki, Tomoya; Kotake, Kei; Sato, Katsuhiko . . . . . . 1101

Near-Infrared Observations of Sagittarius A *

Trippe, Sascha; Paumard, Thibaut; Gillessen, Stefan; Otto,
Thomas; Eisenhauser, Frank; Martins, Fabrice; Genzel, Reinhard . . . . . 1104

Long-Term Monitoring of the Hard X-Ray/Gamma-Ray Emission

from Galactic Black Hole with BATSE
Case, Gary L.; Anzalone, Evan; Cherry, Michael L.; Rodi, James
C.; Ling, James G.; Radocinski, Robert G; Wells, Derek; Wheaton,
William A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1107

Marginally Stable Thick Discs Orbiting Kerr-de Sitter Black Holes

Slany, Petri StuchHk, Zdenek . . . . . . . . . . . . . . . . . . . . . 1110

Black Holes in Scalar Field or Quintessential Cosmology

Harada, Tomohiro . . . . . 1113

A New Solution for Einstein Field Equation in General Relativity

Mousavi, Sadegh . . . . . 1116

Pseudo-Kerr Geometry
Mondal, Soumen; Chakrabarti, Sandip K. . . . . . . . 1119

Extreme Gravitational Lensing by Supermassive Black Holes

Bozza, Valerio . . . . . . . 1122
xxx
• Spectral and Timing Appearances of the Galactic and
Extragalactic Black Holes
Chairperson: Titarchuk, Lev
Physical Characteristics of XTE J1650-500 and GRS 1915+105
with BeppoSAX
Montanari, Enrico; Titarchuk, Lev; Frontera, Filippo · . . . . 1127

Spectral and Timing Properties of Magnetized Advective Flows

with Standing Shocks
Mandal, Samir; Chakrabarti, San dip K. . . . . . . . . . . . . . . . . 1130

Estimating Black Hole Masses in ULXs

Soria, Roberto 1133

• Extreme Properties of Neutron Stars: Observations and Theory

Chairperson: Mendez, Mariano

Equation of State of Dense Matter in Neutron Stars

Cerny, Slavomir; Stone, Jii'ina, Rz1;;ovska; StuchUk, Zdenek;
Hled{k, Stanislav . . . . . . . . . . . . . . . . . . . · . . . . 1139

Detectability of Gravitational Waves from the r-Mode Instability

in Newly-Born Neutron Stars
Sa, Paulo M.; Tome, Brigitte. . . . . . . . . . . . . . . . . . . . . 1142

X-Ray Dim Isolated Neutron Stars: A Review of the Latest Timing

and Spectral Properties
Zane, Silvia · . . . . 1145

X-Ray Observations of Neutron Stars and the Equation of State

at Very High Densities
Triiemper, Joachim E. . . . . . . . . . . . . . . . . . . . . . . . . 1148

Eigenmodes of Rapidly Rotating Neutron Stars

Boutloukos, Stratos . . . . . . . . . . . . · . . . . 1152
Parameter Space Study of Magnetohydrodynamic Flows Around
Magnetized Compact Objects
Das, Santabrata; Chakrabarti, Sandip K. . . . . . . . . . . . . . . . . 1155

Gravitational Radiation from Accreting Millisecond Pulsars

Vigelius, Matthias; Payne, Donald; Melatos, Andrew · . . . . 1158

Dynamical Stability of Fluid Spheres in Spacetimes with a Nonzero

Cosmological Constant
Hled{k, Stanislav; Stuchl{k, Zdenek; Mrazova, Kristina . . . . . . . . . . 1161
 xxxi

• Strange Stars
Chairperson: Usov, Vladimir
Strangelets in Cosmic Rays
Madsen, Jes . . . . . . . . . . . . . . . . 1167
Can Strange Stars be Distinguished from Neutron Stars?
Harko, Tiberiu; Cheng, Kwong Sang. . . . . . . . 1177
Pair Winds in Schwarzschild Spacetime with Application to
Strange Stars
Aksenov, Alexey, G.; Milgram, Mordehai; Usov, Vladimir, V. . . . . . . . 1180
Evidence for White Dwarfs with Strange-Matter Cores
Mathews, Grant J.; Suh, In-Saeng; Lan, Nguyen Quynh; Otsuki,
Kaori; Weber, Fridolin · . . . . 1183
• Thermal Behavior of Compact Stars
Chairperson: Page, Dany
Magnetars: Internal Heating and Energy Budget
Yakovlev, Dmitry G.; Kaminker, Alexander D.; Potekhin, Alexan-
der Y.; Shternin, Peter S.; Chabrier, Gilles; Shibazaki, Noriaki · . . . . 1189
Trapping of Neutrinos in Extremely Compact Neutron Stars
Stuchlzk, Zdenek; Urbanec, Martin; Torok, Gabriel;
Hledzk, Stanislav; HladzK, Jan. . . . . . . . . . . . . . · . . . . 1192
A Self-Consistent Model of the Isolated Neutron Star RX J0720.4-
3125
Miralles, Juan A.; Pons, Jose A.; Perez-Azorin, J. Fernando;
Miniutti, Giovanni . . . . . . 1195
kHz QPO Pairs Expose the Neutron Star of Circinus X-I
Boutloukos, Stratos; van der Klis, Michiel; Altamirano, Diego;
Klein Wolt, Marc; Wijnands, Rudy . . . . . . . . . . . . . . . 1198
Neutron Star Atmospheres and X-Ray Spectra
Kundt, Wolfgang . . . . . . . . . . . . . . . . . . . . . . . . . . 1201
• Alternative Theories (A)
Chairperson: Schmidt, Hans-Juergen
Anisotropically Inflating Universes
Hervik, Sigbj¢rn; Barrow, John D. . . . . . . . . . . . . . . 1207
Thick Brane Solution with Two Scalar Fields
Dzhunushaliev, Vladimir; Schmidt, Hans-Jiiergen; MYTzakulov,
Kairat; Myrzakulov, Ratbay . . . . . . . . . . . . . . . . . . . 1210
Shear Dynamics in Bianchi I Cosmologies with Rn-Gravity
Leach, Jannie A.; Dunsby, Peter K.S.; Carloni, Sante . . . . . . . . . . 1213
xxxii

Spontaneous Lorentz Violation, Gravity and Nambu-Goldstone

Modes
Bluhm, Robert; . . . . . . . . . . . . . 1217
Spontaneous Lorentz Breaking, Nambu-Goldstone Modes, and
Gravity
Potting, Robertus . . . . . . · . . . . . . . 1220
The Significance of Matter Coupling in f(R) Gravity
Sotiriou, Thomas P.. . . . . . . . . . . . . . . . . . . . . 1223
Constraining Alternative Theories of Gravity with the Energy
Conditions
Perez Bergliaffa, Santiago Esteban . . . . . . . . . . . . 1226
An f(R) Gravitation for Galactic Environments
Sobouti, Youse!. . . . . . . . . . . . . . . . . . . . . . . . . . . 1230
Causality and Superluminal Fields
Bruneton, J ean- Philippe . . . . . . . . . . . . 1233
Gravitation as a Vacuum Nonlinear Electrodynamics Effect
Chernitskii, Alexander A. . . . . . . . . . . . . . . . · . . . . 1236
Asymptotic Flatness and Birkhoff's Theorem in Higher-Derivative
Theories of Gravity
Clifton, Timothy · . . . . . . . 1239
Cosmological Model with a Born-infield Type Scalar Field
Kerner, Richard; Serie, Emmanuel; Troisi, Antonio . · . . . . . . . 1242
A Teleparallel Representation of the Weyl Lagrangian
Vassiliev, Dmitri . . . . . . . . . . . . . . . . . . . . . . . 1245
Nonlinear Supersymmetric General Relativity
Shima, Kazunari; Tsuda, Motomu . . . . . . . . . . 1248
Black Hole Solutions in N > 4 Gauss-Bonnet Gravity
Alexeyev, Stanislav 0.; Popov, Nikolai. . . . . . . · . . . . 1251
Electrostatics and Confinement in Einstein's Unified Field Theory
Antoci, Salvatore; Liebscher, Dierck-Ekkehard; Mihich, Luigi . . . . . . . 1254
Galactic Disks in Theories with Yukawian Gravitational Potential
de Araujo, Jose Carlos N.; Miranda, Oswaldo D. . . . . . . . . . . . . 1257
On the Field Theoretic Description of Gravitation
Nieuwenhuizen, Theo M. . . . . . . . . . . . · . . . . 1260
De Sitter Stability in Theories with Second Order Curvature Terms
Toporensky, Alexey V.; Tretyakov, Petr V. . . . . . . . . . . . . . . . 1263
 xxxiii

Basic Relations of a Unified Theory of Electrodynamics, Quantum

Mechanics, and Gravitation
Ostermann, Peter. . . . . . . . . . . . . . . . . . . . . . . . . . 1266
Physical Interpretation and Viability of Various Metric Nonlinear
Gravity Theories
Sokolowski, Leszek M. . . . . . . . . . . . . . . . . . . . . . . . . 1269
• Alternative Theories (B)
Chairperson: Hammond, Richard
Are Active and Passive Electric Charges Equal?
Laemmerzahl, Claus; Macias,. Alfredo,: Miiller, Holger . . . . . . 1275
Charged Fluid Dynamics in Scalar-Tensor Theories of Gravity
with Torsion
Wang, Chih-Hung . . . . . . . . . . . . . . . . . . . . . . . . . 1278
Validation of the Weak Equivalence Principle in a Spatially-VSL
Gravitation Model
Bmekae'ri, Jan . . . . . . . . . . . . . . . . . . . . . . . . . . . 1281
• Higher Dimensional Theories
Chairperson: Coley, Alan
Exact Solution of the 5D Space-Time-Matter Universe and Their
Implications
Fukui, Takao . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1287
Hamiltonian Theory of Brane-World Gravity
Kovacs, Zolt6n; Gergely, Laszl6 A. . . . . . . . . . . . . . . 1290
Casimir Force Test of a 6D Brane World
Linares, Roman; Momles-Tecotl, Hugo A.; Pedmza, Omar . . . . . . . . 1293
Electro-Weak Model within a 5-Dimensional Lorentz Group
Theory
Lecian, Orchidea Maria; Montani, Giovanni . . . . . . . . . . . . . . 1296
Spacetimes with Constant Scalar Invariants
Hervik, Sigbj¢m; Coley, A.A.; Pelavas, Nicos . . . . . . . . . . . 1299
Higher Dimensional VSI Spacetimes and Supergravity
Fuster, Andrea; Pelavas,. Nicos . . . . . . . . . . . . . . . . . 1302
VSI & VSI i Spacetimes in Higher Dimensions
Pmvdova, Alena . . . . . . . 1305
The Electro-Weak Model as a Phenomenological Issue of
Multidimensions
Cianfmni, Fmncesco; Montani, Giovanni . . . . . . . . . . . . . . . 1308
xxxiv

Hamiltonian Formulation of the 5-D Kaluza-Klein Model and

Test-Particle Motion
Lacquaniti, Valentino; Montani, Giovanni · . . . . . 1311

Electromagnetism and Perfect Fluids Interplay in Multidimen-

sional Spacetimes
Mitskievich, Nikolai V. . . . . . . . . . . . . . . . . . . . . . . . 1314

Torsion Induces Gravity

Aros, Rodrigo · . . . . . 1317

Final Fate of Higher-Dimensional Spherical Dust Collapse in

Einstein-Gauss-Bonnet Gravity
Maeda, Hideki . . . . . . . . . . . . 1320

Classification of the Weyl Tensor in Higher Dimensions and its

Applications
Pravda, Vojtech. . . . . . . . . . . . . . . . . . . . . . . . . . . 1323

• Geometric Calculus in Gravity Theory

Chairperson: Hestenes, David

Geometrical and Kinematical Aspects of Rindler Observers

Romero, Carlos; Brasileiro Formiga, Jansen . . . . . 1329

On the Zeros of Spinor Fields and an Orthonormal Frame Gauge

Condition
Nester, James M. . . . . . . . . . . . . . . . . . . . . . . . . . . 1332

New Special Solutions of the Ricci Flow Equation in Two Dimen-

sions Using a Linearization Approach
Visinescu, Anca; Visinescu, Mihai . . . . . . . . . 1335

• Black Hole and Pair Creation in Strong Fields

Chairperson: Greiner, Walter

Pair Creation in Inhomogeneous Fields

Schubert, Christian . . . . . . . . . . . . . . . . . . . . . . . 1341

Monopole Decay in a Variable External Field

Mon'in, Alexander K.; Zayakin, Andrey V. . · . . . . . 1346

World-Making with Extended Gravity Black Holes for Cosmic

Natural selection in the Multiverse Scenario
Barrau, A urelien . . . . . . . . . . . . . . . . . . . . . . . 1349

Neutral Nuclear Core vs Super Charged One

Rotondo, Michael; Ruffini, Remo; X ue, She-Sheng. . . . . . . . . . . . 1352
 xxxv
• Black Holes in Higher Dimensions (Black Rings and Black Strings)
Chairperson: K unz, Jutta
Gravitational Perturbations of Higher Dimensional Rotating Black
Holes
Kunduri, Hari K.; Lucietti, James; Reall, Harvey S. . . . . . . . . . . . 1359
Gravitating Non-Abelian Solitons and Hairy Black Holes in 'Higher
Dimensions
Volkov, Mikhail S. . . . . . . . . . . . . . 1379
Derivation of the Dipole Black Ring Solutions
Yazadjiev, Stoytcho S. . . . . . . . . . . . . . . . . . . . . . 1397
Charged Rotating Black Holes in Higher Dimensions
Kunz, Jutta; Navarro-Lerida, Francisco; Viebahn, Jan;
Maison, Dieter . . . . . . . . · . . . . . 1400
Solitonic Generation of Solutions Including Five-Dimensional
Black Rings and Black Holes
Mishima, Takashi; Iguchi, Hideo . . . . . 1403
Kaluza-Klein Black Hole with Gravitational Charge in Einstein-
Gauss-Bonnet Gravity
Maeda, Hideki; Dadhich, Naresh K. . . . . . . . . . . . . . . . . . . 1406
Higher Dimensional Rotating Charged Black Holes
Aliev, Alikram N. . . . . . . . . . . . . . . . · . . . . . 1409
Perturbative Stability and Absorption Cross-Section in String
Corrected Black Holes
Moura, Filipe. . . . . . . . . . . . . . . . . . . . . . . . . . . . 1412
Ultrarelativistic Boost of the Black Ring
Ortaggio, Marcello; Krtous, Pavel; Podolsky, Jii'{ . . . . . . . . . . . . 1415
The Results of a New Solution of the Einstein Field Equations in
General Relativity and Black Hole New Movement
Mousavi, Sadegh . . . . . . . ..... . · . . . . . 1418
Hamiltonian Treatment of Static and Collapsing Spherically
Symmetric Charged Thin Shells in Lovelock Gravity
Dias, Gonr;alo A.B.; Lemos, Jose P.S.; Gao, Sijie . . . . . . . . . . . . 1421
New Nonuniform Black String Solutions
Kleihaus, Burkhard; Kunz, Jutta; Radu, Eugen . . . . . . . . . . . . . 1424
Lovelock Gravity and The Counterterm Method
Bostani, Neda; Dehghani, Mohammad Hossein; Sheykhi, Ahmad . . . . . 1427
LG(Landau-Ginzburg) in GL(Gregory-Laflamme)
Kol, Barak; Sorkin, Evgeny . . . . . . . . . . . . 1431
xxxvi

Causal Structure Around Spinning 5-Dimensional Cosmic Strings

Slagter, Reinoud Jan · . . . . 1434
Short Distances, Black Holes, and TeV Gravity
AguIl6, Ivan; Navarro-Salas, Jose; Olmo, Gonzalo J. · . . . . . . 1437
Black String Solutions with Negative Cosmological Constant
Mann, Robert; Radu, Eugen; Stelea, Cristian . . . . . · . . . . . . 1440
Matched Asymptotic Expansion for Caged Black Holes
Gorbonos, Dan; Kol, Bamk · . . . . 1443
Perturbatively Non-Uniform Charged Black Strings: A New Stable
Phase
Miyamoto, Umpe'l; Kudoh, Hideaki . . . , . . . . . . 1446
• Analog Models of and for General Relativity
Chairperson: Volovik, Grigor-y
From Quantum Hydrodynamics to Quantum Gravity
Volovik, Grigory . . . . . . . . . , . . . . . . , . . . . . . . . . 1451
Looking Beyond the Horizon
Babichev, Eugeny; Mukhanov, Viatcheslav; Vikman, Alexander . . . . , . 1471
A Dielectric Analogue Model of the Kerr Equatorial Plane
Rosq1List, KjeIl . . . . . . . . . . . . . . . . . . . · . . . . . . 1475
Bose-Einstein Condensates and QFT in Curved Space-Time
Fagnocchi, Serena. . . . . . . . . . . . . · . . . . . . 1479
Electromagnetic Light Rays in Local Dielectrics
De Lorenci, Vitorio A.; Klippert, Renato. . . . . . . . . . . . 1482
Scattering Problems on Rotating Acoustic Black Holes
Cherubini, Christian; Filippi, Simonetta . . . . . . . . . . 1485
• Black Hole Thermodynamics
Chairperson: Khriplovich, Iosi!
Thermodynamical Properties of Hairy Black Holes with Cosmo-
logical Constant
Nadalini, Mar-io; Vanzo. Luciano; Zer-bini, Ser-gio . . . . . . . . . . . . 1491
Radiation of Quantized Black Holes. Is it Observable'?
KhripIO'lJ'lch, Iosi! . . . . . . . . . . . . . . . . · . . . . 1494
Effects of Quantized Fields on the Spacetime Geometries of Static
Spherically Symmetric Black Holes
Anderson, Paul R.; Binkley, Mathew; Calderon, Hector,' Hiscock,
Will'lam A.; Mottola, Emil; Vaulin, Ruslan . . . . . . . . . . . . . . . 1497
 xxxvii

Thermodynamic Quantities of Kaluza-Klein Black Holes with

Squashed Horizons
Kurita, Yasunari; Ishihara, Hideki . . . . . . . . . . 1500
Hawking Radiation and Black Hole Thermodynamics
Page, Don N. ................ . . . . . . 1503
Entropy from Conformal Horizon States in D-Dimensional Spheri-
cal, Toroidal, and Hyperbolic Anti-de Sitter Black Holes
Lemos, Jose P.S.; Dias, Gon(:alo A.S. . . . . . . . . . . . . . . . . . 1508
Thermodynamic Geometries of Black Holes
Aman, Jan E.; Bengtsson, Ingemar-; Pidokrajt, Narit; Ward, John . . . . 1511
• Alternative Black Hole Models
Chair-person: Mazur, Pawel O.
On Quantum Effects in the Vicinity of Would-be Horizons
M arecki, Piotr . . . . . . . . 1517
Stable Dark Energy Stars: An Alternative to Black Holes?
Lobo, Francisco S.N. . . . . . 1520
Horizon News Function and Quasi-Local Energy-Momentum Flux
Near Black Hole
Wu, Yu-Huei. . . . . . . . . . . . . . . . . . . . . . . . . . . . 1523
Black Holes or Eternally Collapsing Objects?
Mitra, Abhas; Glendenning, Norman K. . . . . . . . . . . . . . . 1526
The Proposed Black Holes Around Us
Kundt, Wo~fqang . . . . . . . . . . . . . . . . . . 1529
Gravastars and Bifurcation in Quasistationary Accretion
Malec, Edward; Roszkowski, KrzysztoJ . . . . . . . . . . . 1537
• Numerical Relativity, Black Hole Collisions, and Algebraic
Computation
Chair-person: Husa, Sascha
Lifetime of Oscillons
Fodor, Gyula; Forgacs, Peter; Grandclement, Philippe; Racz, Istvan . . . . 1543
A Virtual Trip to the Schwarzschild-de Sitter Black Hole
Bakala, Pavel; Hledik, Stanislav; StuchHk, Zdenek; Truparova
Kamila; Cermak, Petr. . . . . . . . . . . . . . . 1546
Similarity Solutions Using DESOLV
Vu, Khai T.; Butcher, J.; Carminati, John. . . . . . . . . . . . . . . 1549
xTensor: A Free Fast Abstract Tensor Manipulator
Martin-Garcia, Jose M. . . . . . . . . . . . 1552
xxxviii

Tensor Computer Algebra

Martin-Garcia, Jose M. · . . . . 1555
• Simulations of Relativistic Flows and Compact Objects
Chairperson: Font, Jose A.
Relativistic MHD Simulations and Synthetic Synchrotron Emission
Maps: A Diagnostic Tool for Pulsar Wind Nebulae
Del Zanna, Luca; Volpi, Delia; Amato, Elena; Bucciantini,
Nicolo . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1561
GRMHD Simulations of Jet Formation with RAISHIN
Mizuno, Yosuke; Nishikawa, Ken-Ichi; Koide, Shinji; Hardee,
Philip; Fishman, Gerald J. . . . . . . . . . . . . . . . . . . . . . . 1564
Standing Shocks in Pseudo-Kerr Geometry
Mondal, Soumen; Chakrabarti, Sandip K. . . . . . . 1567
Evolving Relativistic Fluid Spacetimes Using Pseudospectral
Methods and Finite Differencing
Duez, Matthew D.; Kidder, Lawrence E.; Teukolsky, Saul A. · . . . . 1570
3D Relativistic MHD Simulation of a Tilted Accretion Disk Around
a Rapidly Rotating Black Hole
Fragile, P. Chris; Anninos, Peter; Blaes, Omer M.; Salmonson,
Jay D .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1573
Fragmentation of General Relativistic Quasi-Toroidal Polytropes
Zink, Burkhard Sebastian; Stergioulas, Nikolaos; Hawke, Ian; Ott,
Christian D.; Schnetter, Erik; Miiller, Ewald . . . . . . . . . . . . 1576
Adaptive Mesh Refinement and Relativistic MHD
Neilsen, David; Hirschmann, Eric W.; Anderson, Matthew;
Liebling, Steven L. . . . . . . . . . . . . . . . . . . . · . . . . 1579
3-D GRMHD and GRPIC Simulations of Disk-Jet Coupling and
Emission
Nishikawa, Ken-Ichi; Mizuno, Yosuke; Watson, Michael; Hardee,
Philip; Fuerst, Steve; Wu, Kinwah; Fishman, Gerald J. · . . . . 1582
Resistive General Relativistic MHD Simulations of Jet Formation
Around Kerr Black Hole
Koide, Shinji; Shibata, Kazunari; Kudoh, Takahiro · . . . . 1585
Making Up a Short GRB: The Bright Fate of Mergers of Compact
Objects
Aloy, Miguel Angel; Mimica, Petar . . . . . . . . . 1589
Spacetime Modes of Rapidly Rotating Relativistic Stars
Stergioulas, Nikolaos; Kokkotas, Kostas D.; Hawke, Ian . . . . . . . . . 1592
 xxxix

• Dynamics of Compact Binaries

Chairperson: Rezzolla, Luciano
Reducing Orbital Eccentricity in Binary Black Hole Simulations
Pfeiffer, Harald P.; Brown, Duncan A.; Kidder, Lawrence E.;
Lindblom, Lee; Lovelace, Geoffrey; Scheel, Mark A. . . . . . . . . 1597
Negative Komar Masses in Regular Stationary Spacetimes
Ansorg, Marcus; Petroff, David . . . . . . . . . . . . . . . . . . . . 1600
Constraint Relaxation
Marronetti, Pedro. . . . . . . . 1603
Relativistic Hydrodynamic Simulations of Multiple Orbits for
Close Neutron Star Binaries
Mathews, Grant J.; Haywood, J. Reese; Wilson, James R. · . . . . 1606
The Final Fate of Binary Neutron Star Systems: What Happens
After the Merger?
Duez, Matthew D.; Liu, Yuk Tung; Shapiro, Stuart L.; Stephens,
Branson C.; Shibata, Masaru . . . . . . . . . . . · . . . . 1609
Head-On Collisions of Different Initial Data
Sperhake, Ulrich; Brugmann, Bernd; Gonzalez, Jose A.; Hannam,
Mark D.; Husa, Sascha · . . . . 1612
Tearing instability in Relativistic Magnetically Dominated Plasmas
Barkov, Maxim V.; Komissarov, Serguei S.; Lyutikov, Maxim. · . . . . 1615
• Black Hole Collisions
Chairperson: Lousto, Carlos
Black Hole Bremsstrahlung in the Nonlinear Regime of General
Relativity
Oliveira, Henrique P.; Soares, 1. Damiao; Tonini, Eduardo V. . . . . . . 1621
Hyperboloidal Foliations with J-Fixing in Spherical Symmetry
Zenginoglu, Anil; Husa, Sascha . . . . . . . . . . . . . . . . . . . . 1624
High-Order Perturbations of a Spherical Spacetime
Brizuela, David; Martin-Garcia, Jose Maria; Mena Marugan,
Guillermo A. . . . . . . . . . . . . . . . . . . . . . . · . . . . 1627
Analytic Solutions of the Linearized Einstein Equations Used to
Test and Develop a Characteristic Code
Bishop, Nigel T. . . . . . . . . . ... . . . . . . . . . . . . . . . 1630
The Kerr Metric in Bondi-Sachs Form
Venter, Liebrecht R.; Bishop, Nigel T. . . . . . . . . . . . . . . . . . 1633
xl

Binary Black Hole Merger Waveforms in the Extreme Mass Ratio

Limit
Damour, Thibault; Nagar, Alessandro . . . . . . . . . . . . . . . . . 1636
• CMB Theory
Chairperson: Dore, Olivier
CMB Anomalies from Relic Anisotropy
Giimriik<;iioglu, A. Emir; Contaldi, Carlo, R.; Peloso, Marco . . . . . . . 1641
Can Extragalactic Foregrounds Explain the Large-Angle CMB
Anomalies?
Rakic, Aleksandar; Rasanen, Syksy; Schwarz, Dominik J. . . . . . . . . 1647
A New Realization of a Low Quadrupole Universe
Lee, Wo-Lung . . . . . . . . . . . . 1653
Perturbations of Dark Sectors from the CMB
Bashinsky, Sergei . . . . . . . . . . . . . . . . . . . . 1659
• CMB Experiment Space
Chairperson: Masi, Silvia
Observations of the CMB and Galactic Foregrounds at 11-17 GHz:
The COSMOSOMAS Experiment
Hildebrandt, Sergi R. . . . . . . . . . . . . . . . . . . . . . . 1667
• CMB Data Analysis
Chairperson: Natoli, Paolo
Probing Cosmic Dark Ages with the CMB Polarization Measure-
ments
Popa, L71cia Aurelia; Stefanescu, Petruta; Burigana, Carlo . . . . . . . . 1671
Dark Energy Constraints from Needlets Analysis of WMAP3 and
NVSS Data
Pietrobon, Davide; Balbi, Amedeo; Marinucci, Domenico. . . . . . . . . 1674
The Matter Power Spectrum as a Tool to Discriminate Dark
Matter-Dark Energy Interaction
Olivares, German; Pavon, Diego; Atrio-Barandela, Fernando . . . . . . . 1677
The BRAIN Experiment
Polenta, Gianluca; for the BRAIN collaboration. . . . . . . . . . . . . 1680
• Observational Gravitational Lensing
Chairperson: Jetzer, Philippe
Microlensing with the Radioastron Space Telescope
Zakharov, Alexander F. . . . . . . . . . . . . . . . . . . . . . . 1691
Microlensing Towards M31
Calchi Novati, Sebastiano . . . . . . . . . . . . . . . . . . . . . . 1694
 xli

Does the LMC Halo Contribute Significantly to the MACHO

Events?
Scarpetta, Gaetano o 0 0 0 0 0 0 0 0 1697
A New Analysis of the MEGA M31 Microlensing Events
Nucita, Achille Ao; Ingrosso, Gabriele; De Paolis, Francesco;
Strafella, Francesco; Calchi Novati, Sebastiano; Scarpetta, Gae-
tano; Jetzer, Philippe 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1700
On the Lens Nature in Microlensing Searches
De Paolis, Francesco; Ingrosso, Gabriele; Nucita, Achille Ao 0 0 0 0 0 0 0 1702
• Theoretical Gravitational Lensing
Chairperson: Perlick, Volker
Gravitational Lensing by Braneworld Black Holes
Whisker, Richard 0 0 0 0 0 0 0 0 0 0 0 0 0 0
o 0 0 0 0 1707
Gravitational Lensing of Stars Surrounding Supermassive Black
Holes
Bozza, Valerio; Mancini, Luigi 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1710
Kerr Black Holes Gravitational Lensing in the Strong Deflection
Limit: An Analytical Approach
De Luca, Fabiana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1713
On Gravitational Lensing by a Kerr Black Hole
Sereno, Mauro; De Luca, Fabiana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1716
Testing Theories of Gravity with Black Hole Lensing
Keeton, Charles Ro; Petters, Arlie 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1719
Gravitational Lensing by Higher Dimensional Black Holes
Majumdar, Archan So; Mukherjee, Nupur 0 0 0 0 0 0 0 o 0 0 0 0 0 1722
Iron Ka Line Profiles and Shadow Shapes as Evidences of a
Gravitational Lensing in a Strong Gravitational Field near BHs
Zakharov, Alexander F.; De Paolis, Francesco; Nucita, Achille Ao;
Ingrosso, Gabriele 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 o 0 0 0 0 1725
Lensing Effects on Gravitational Waves in a Clumpy Universe
Yoo, Chul-Moon; Nakao, Ken-ichi; Kozaki, Hiroshi; Takahashi, RY1tichi 0 0 1728
QSO Lensing
Miranda, Marco; Jetzer, Philippe; Jl;/accio, Andrea V. o 0 0 0 0 0 1731
JLenses and XFGLenses
Frutos-Alfaro, Francisco; Solis-Sanchez, Hugo 0 0 0 0 0 0 0 0 0 0 0 0 0 1734
Wave Fronts in General Relativity Theory
Grave, Frank; Frutos-Alfaro, Francisco; Miiller, Thomas; Adis, Daria 0 0 0 1737
xlii

• Galaxies and the Large-Scale Structure

Chairperson: Sheth, Ravi
Spherical Voids in a Newton-Friedmann Universe
Triay, Roland; Fliche, Henri H.. . . . . . . . . . . . . . . . . . . . 1743
• Dark Energy and Universe Acceleration
Chairper-son: Starobinsky, Alexei
Dark Energy and Universe Acceleration of Nonlinear Supersym-
metric General Relativity
Shima, Kazunari; Tsuda, Motomu · . . . . . 1749
Testing the Dark-Energy-Dominated Cosmology by the Solar-
System Experiments
Dumin, Yurii V. . . . . . . . . . . . . . . . . . . . . . . . . 1752
A Darkless Spacetime
Tartaglia, Angelo; Capone, Monica · . . . . . 1755
On Intrinsic Invariance in Gurzadyan-Xue Cosmological Models
Khachatrian, Harutyun; Vereshchagin, Gr-egory V.; Yegorian, Gegham . . . 1758
Phantom Dark Energy and its Cosmological Consequences.
Dgbrowski, Mariusz P. . . . . . . . . . . . . . . . . . . . . . 1761
The Generalized Second Law in Dark Energy Dominated Universes
Izquierdo, German; Pavon, Diego . . . . . . . . . . . . . . . . . . . 1764
Accelerated Expansion by Non-Minimally Coupled Scalar Fields
Bieli, Roger . . . . . 1767
Vacuum Energy Generating Mechanisms in Cosmic Expansion
with Natural UV Cutoff
Kempf, Achim . . . . . . . . . . 1770
New Kinematical Constraints on Cosmic Acceleration
Rapetti, David; Allen, Steve W.; Amin, Mustafa A.; Blandford,
Roger D. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1773
Dark Energy and Decaying Dark Matter
Mathews, Grant J.; Lan, Nguyen Quynh; Wilson, James R. . . . . . . . 1776
Gravitational Instanton-Solution to Cosmological Constant
X ue, She-Sheng. . . . . . . . . . . . . . . . . . . . . . . . . . . 1779
An Alternative Source for Dark Energy
Wanas, Mamdouh 1. ..... · . . . . . 1782
An Awesome Hypothesis for Dark Energy: The Abnormally
Weighting Energy
Fuzfa, AndTe; Alimi, Jean-Michel . . . . . . . . . . . . . . . . . . . 1785
 xliii

Perturbations of a Cosmological Constant Dominated Universe

Vasuth, Matyas; Czinner, Viktor · . . . . 1788
On the Gurzadyan-Xue Cosmological Models and the Dynamics
of Density Perturbation
Yegorian, Gegham. . . . . . . . . . . . . . . . . . . . . . . . . . 1791
Scalar-Tensor Dark Energy Models
Ganouji, Radouane; Polar-ski, David; Ranquet, Andre; Starobinsky,
Alexei A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1794
Reconstruction of Dark Energy Using Supernova and Other
Datasets
Alam, Ujjaini; Sahni, Varun; Starobinsky, Alexei A. . . . . . · . . . . 1797
feR) Dark energy: From the Time of Recombination till Present
Day
Gurovich, Viktor; Folomeev, Vladimir; Tokareva, Jya . . . . . . 1800
Probing Dynamical Dark Energy with Press-Schechter Mass
Functions
Le Delliou, Morgan · . . . . 1803
Broken Scale Invariance and Quintessence (A Quarter of a Century
Ago)
Venturi, Giovanni. . . . . . . . . . . . . . . . . . . . . . . . . . 1807
• Topology of the Universe
Chairperson: Demianski, Marek
An Axisymmetric Object-Based Search for a Flat Compact
Dimension
Mathews, Grant J.; Menzies, Dylan . . . . . . . . . . . . . . . . 1813
Topological Gravitation on Graph Manifolds
Mitskievich, Nikolai v.; Efr-emov, Vladimir N; Hernandez Magdaleno . . . 1816
Supernovae Constraints on Cosmological Density Parameters and
Cosmic Topology
Rebour;as, Marcelo J. . . . . . . 1819
Supernovae Constraints on DGP Model and Cosmic Topology
Rebour;as, Marcelo J. . . . . . . . . . . . . . . . . . . . . . . . . 1824
• Inhomogeneous Cosmology
Chairperson: Krasinski, A ndrzej
Reinterpreting Dark Energy Through Backreaction: The Minimally
Coupled Morphon Field
Larena, Julien; Bucher-t, Thomas; Alimi, Jean-Michel . . . . . . . . . . 1831
xliv

Initial Conditions for Primordial Black Hole Formation

Musco, Ilia; Polnarev, Alexander G. . . . . . . . . . · . . . . 1834
Is the Apparent Acceleration of the Universe Expansion Driven by
a Dark Energy-Like Component or by Inhomogeneities?
Marie-Noelle; Celerier. . . . . . . . . . . . . . . . . . . . . . . . 1837
Evolution of a Void and an Adjacent Galaxy Supercluster in the
Quasispherical Szekeres Model
Bolejko, KrzysztoJ . . . . . . . . . . . . . . . . . . . . . . . . . . 1847
Covariant Description of the Inhomogeneous Mixmaster Chaos
Benini, Riccardo; Montani, Giovanni . . . . . . 1857
The Mass and the Geometry of the Cosmos
Hellaby, Charles; Lu, Hui-Ching . . . . . . . . . . . . . . . . . . . 1860
• Nonsingular Cosmology - Inflation
Chairperson: Novello, Mario
Emergent Universe with Bulk Viscosity
Mukherjee, Sailo; Paul, Bikash C.; Dadhich, Naresh K.; Maharaj,
Sunil D.; Beesham, Aroonkumar · . . . . 1873
Bulk Viscosity Impact on the Scenario of Warm Inflation
Mimosa, Jose Pedro; Nunes, Ana; Pavon, Diego . . . . . . . . 1876
• Quantum Cosmology and Quantum Effects in the
Early Universe
Chairperson: Vargas lvloniz, Paulo
Scalar Field Phase Dynamics in Preheating
Charters, T.; Nunes, Ana; Mimosa, Jose Pedro . . . . . . . . 1881
Branch Wave Functions for Quasi-Classical Homogeneous
Universes
Craig, David . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1884
Quantum Phantom Cosmology
SandhoeJer, Barbara. . . . . . . . . . . . . . . . . . . . . 1887
Generic Evolutionary Quantum Universe
Battisti, Marco Valerio; Montani, Giovanni . . . . . . . . . 1890
Quantum Cosmology from Three Different Perspectives
Esposito, Giampiero . . . . . . . . . . . . . . . . · . . . . 1893
On the Thermal Boundary Condition of the Wave Function of the
Universe
Bouhmadi-Lopez, Mariam; Vargas Moniz, Paulo . . . . . . . . . . . . 1898
 xlv

Dark Energy from Quantum Wave Function Collapse of Dark

Matter
Majumdar, Archan S.; Home, D. · . . . . . 1901
Cosmological Perturbations in Quantum Cosmological Back-
grounds
Pinto-Neto, Nelson . . . . . . . . . . . . . . . . . . . . . . . . 1904
Classical and Quantum Aspects of the Inhomogeneous Mixmaster
Chaoticity
Benini, Riccardo; lv1ontani, Giovanni . . . . . . . . . . . 1909
Cosmological Dynamics with Vacuum Polarization
Toporensky, Alexey V.; Tretyakov, Petr V. . . . . . . . . . . . 1912
Some Cosmological Consequences of Loop Quantum Gravity
Mulryne, David J.; Tavakol, Reza . . . . . . . . . . . . . . 1915
Semiclassical Supersymmetric Quantum Gravity
Kiefer, Claus; Liick, Tobias; Vargas Moniz, Paulo . . . . . . . . . . . . 1920
Multigravity and Spacetime Foam
Gamttini, Remo · . . . . . 1925
Boundary Conditions and Predictions of Quantum Cosmology
Page, Don N. · . . . . . 1928
Quantum Cosmology with Nontrivial Topology
Fagundes, Helio V.; Vargas, Teofilo . . . . . . . . . . . . . . . 1933
Non-Singular Solutions in Loop Quantum Cosmology
Vereshchagin, Gregory V. . . . . . . . . . . . . · . . . . . 1936
Minimal Energy and Factor Ordering in Quantum Cosmology
Hinterieitner, Fmnz; Ste'igl, Roman . . . . . . . . · . . . . . 1939
On the False Vacuum Bubble Nucleation
Lee, Bum-Hoon; Lee, Chul Hoon; Lee, Wonwoo; Park, Chanyong . . . . . 1942

PARTC
PARALLEL SESSIONS
• The GRB - Supernova Connection
Chairperson: Chardonnet, Pascal
Swift Observations of GRB050712
De Pasquale, Massimiliano; Poole, Tmcey; Zane, Silvia; Page,
Mathew; Breeveld, Alice; O'Mason, Keith; Gmpe, Dicke; BUTTOWS,
David; Nousek, John; Roming, Peter; KTimm, Hans; GehTels,
Neil; Zhang, Bing; Kobayashi, Shiho. . . . . . . . . . . . . . . . . . 1947
xlvi

No Astrophysical Dyadospheres
Page, Don N. . . . . . 1950
Magnetized Hypermassive Neutron Star Collapse: A Candidate
Central Engine for Short-Hard GRBs
Stephens, Branson C.; Duez, Matthew D.; Liu, Yuk Tung; Shapiro,
Stuart L.; Shibata, Masaru . . . . . . . . . . . . . . . . . . . . . . 1953
Theoretical Interpretation of Luminosity and Spectral Properties
of GRB 031203
Bianco, Carlo Luciano; Bernardini, Maria Grazia; Chardonnet,
Pascal; Fraschetti, Federico; Ruffini, Remo; Xue, She-Sheng . . . . . 1956
GRB980425 and the Puzzling URCA1 Emission
Bernardini, Maria Grazia; Bianco, Carlo Luciano; Caito, Letizia;
Dainotti, Maria Giovanna; Guida Roberto; Ruffini, Remo J. . . . . . 1959

• The Afterglow, Short and Long GRBs

Chairperson: Arkhangelskaja, Irene
The Ep,i-Eiso Correlation and the Nature of Sub-Energetic GRBs
Amati, Lorenzo . . . . . . . . . . . . . . . . . . . . . . . . . . . 1965
The GRB Detected by AVS-F Apparatus Onboard CORONAS-F
Satellite in 2001-2005 Years
Arkhangelskaja, Irene v.; Arkhangelsky, Andrey I.; Glyanenko,
Alexander S.; Kotov, Yuri D.; Kuznetsov, Sergey N. . . . . . 1968
Special Relativistic Simulations of Magneto-Driven Jet from
Core-Collapse Supernovae
Takiwaki, Tomoya; Kotake, Kei; Yamada, Shoichi; Sato, Katsuhiko . . . . 1971
Theoretical Interpretation of "Long" and "Short" GRBs
Bianco, Carlo Luciano; Bernardini, Maria Grazia; Caito, Letizia;
Chardonnet, Pascal; Dainotti, Maria Giovanna; Fraschetti,
Francesca; Guida, Roberto; Ruffini, Remo; Xue, She-Sheng. . . . . . . . 1974
Theoretical Interpretation of GRBOl1121
Caito, Letizia; Bernardini, Maria Grazia; Bianco, Carlo Luciano;
Dainotti, Maria Giovanna; Guida, Roberto; Ruffini, Remo. . . . . . 1977
On GRB 060218 and the GRBs Related to Supernovae Ib/c
Dainotti, Maria Giovanna; Bernardini, Maria Grazia; Bianco,
Carlo Luciano; Caito, Letizia; Guida, Roberto; Ruffini, Remo. . . . . . . 1981
The "Fireshell" Model in the Swift Era
Bianco, Carlo Luciano; Ruffini, Remo . . . . . . . . . . . . . . . . . 1989
 xlvii

GRB970228 as a Prototype for the Class of GRBs with an Initial

Spikelike Emission
Bernardini, Maria Grazia; Bianco, Carlo Luciano; Caito, Letizia;
Dainotti, Maria Giovanna; Guida, Roberto; Ruffini, Remo . . . . . . . . 1992
Theoretical Interpretation of GRB060124: Preliminary Results
Guida, Roberto; Bernardini, Maria Grazia; Bianco, Carlo Luciano;
Caito, Letizia; Dainotti Maria Giovanna; Ruffini, Remo . . . . . . . . . 1995
• G RBs and Host Galaxies
Chairperson: Bjornsson, Gunnlaugur
Numerical Counterparts of GRB Host Galaxies
Courty, Stephanie; Bjornsson, Gunnlaugur; Gudmundsson, Einar H. '" 2003
The Host Galaxies of Long Gamma-Ray Bursts: The Mid-Infrared
view from the Spitzer Space Telescope
Le Floc 'h, Emeric . . . . . . . . . . . . . . . . . . . . . . . . . . 2006
Gamma-Ray Burst Host Galaxy Gas and Dust
Starling, Rhaana; Wijers, Ralph; Wiersema, Klaas . . . . . . . . . . . 2009
Low Redshift GRBs and their Host Galaxies
Tanvir, Nial R. . . . . . . . . . . . . . . . . . . . . . . . . . . . 2012
The Analysis of GRB Redshift Distribution
Arkhangelskaja, Irene V. . . . . . . . . . · . . . . 2015
Fundamental Properties of GRB-Selected Galaxies: A Swift/VLT
Legacy Survey
lakobsson, Pall; Hjorth, lens; Fynbo, lohan P. U.; Gomsabel,
lavier; launsen, Andreas O. . . . . . . . . . . 2019
• GRB Observations by SWIFT
Chairperson: Angelini, Lorella
The Swift XRT: Early X-Ray Afterglows
Tagliaferri, Gianpiem . . . . . . . . . . . . . . . . . . . . . 2025
Initial Results from SwiftjUVOT
Marshall, Francis E. . . . . . · . . . . 2030
Investigation of Jet Break Features in Swift Gamma-Ray Bursts
Sato, Gam et al. . . . . . . . . . . . . . . . . . . . . . . . . . . 2033
Recent Results from the Swift Burst Alert Telescope
Krimm, Hans A.; for the Swift/BAT team . . . . . · . . . . 2036
Optical Observations of Gamma-Ray Bursts at the First Russian
Robotic Telescope MASTER
Tyur'ina, Nataly; Lipunov, Vladimir M.; Kornilov, Victor G.;
Gorbovskoy, Evgeniy S.; K uvshinov, Dmirtiy A.. . . . . . . . 2039
xlviii

• Cosmological Singularities
Chairperson: Cotsakis, Spiros
Flat, Radiation Universes with Quadratic Corrections and Asymp-
totic Analysis
Cotsakis, Spiros; Tsokaros, Antonios . . . . . . 2045
The Recollapse Problem of Closed Isotropic Models in Second
Order Gravity Theory
Miritzis, John . . . . . . 2048
Big-Rip, Sudden Future and Other Exotic Singularities in the
Universe
Dgbrowski, Mariusz; Balcerzak, Adam . . . . . . . . . . . . . . . . . 2051
Braneworld Cosmological Singularities
Antoniadis, Ignatios; Cotsakis, Spiros; Klaoudatou, Ifigeneia · . . . . 2054
Generalized Puiseux Series Expansion for Cosmological Milestones
Cattoen, Celine; Visser, Matt. . . . . . . . . . . · . . . . 2057
• Chaos in General Relativity and Cosmology
Chairperson: Gurzadyan, Vahe
Chaos in the Yang-Mills Theory and Cosmology: Quantum Aspects
Matinyan, Sergei . . . . . . . . . . . . . . . · . . . . 2063
Chaos, Gravity and Wave Maps with Target SU(2)
Szybka, Sebastian Jan . . . . . . . . . . . . . . . . . . . . . 2078
Chaos in Core-Halo Gravitating Systems
Ghahramanyan, Tigran; Gurzadyan, Vahe G. . . . . . . . . . 2081
Transient Chaos in Scalar Field Cosmology on a Brane
Topor·ensky, Alexey . . . . . . . . . . . . . . . . · . . . . 2084
Toward a Holographic Origin of Cosmological Large Scale Structure
Mureika, Jonas R. · . . . . 2087
Vector Field Induced Chaos in Multi-dimensional Homogeneous
Cosmologies
Benini, Riccardo; Kirillov, A. Alexander; Montani, Giovanni . . . . . . . 2090
• Einstein-Maxwell Systems
Chairperson: Lee, Chul Hoon
Dynamo Action on Relativistic Spherical Stars
Nadiezhda, Montelongo-Garcia; Thomas, Zannias · . . . . 2095
External Electromagnetic Fields of a Slowly Rotating Magnetized
Star with Nonvanishing Gravitomagnetic Charge
Ahmedov, Bobomurat J.; Khugaev, Avas v.; Rakhmatov, Nemat I. . . . . 2098
 xlix

Aligned Electromagnetic Excitations of the Kerr-Schild Solution

Burinskii, Alexander o 0 0 0 0 2101
Static Perturbations of a Reissner-Nordstrom Black Hole by a
Charged Massive Particle
Bini, Donato; Geralico, Andrea; Ruffini, Remo 0 0 0 0 0 0 0 o 0 0 0 0 2104
Charged Black String Solutions of the Einstein-Maxwell Equations
in Higher Dimensions
Lee, Chul Hoon 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2107
On the Hypothesis of Gravimagnetism
Abdil'din, Meirkhan Mo; Abishev, Medeu Eo o 0 0 0 0 0 2110
Static Perturbations of a Reissner-Nordstrom Black Hole by a
Charged Massive Particle
Bini, Donato; Geralico, Andrea; Ruffini, Remo 0 0 0 0 0 0 0 0 0 0 0 0 0 2113
• Theoretical Issues in GR
Chairperson: Brill, Dieter
A Framework for the Discussion of Singularities in General
Relativity
Whale, Benjamin Eo; Scott, Susan Mo 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2119
Axial Symmetric Gravitomagnetic Monopole in Cylindrical
Coordinates
Kagramanova, Valeria Go; Ahmedov, Bobomurat J. o 0 0 0 0 2122
Optical Reference Geometry and Inertial Forces in Kerr-de Sitter
Spacetimes
Kovar, JiN; StuchHk, Zdenek 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2125
On the Construction of Syzgies of the Polynomial Invariants of the
Riemann Tensor
Lim, Allan EoK.; Carminati, John o 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2128
A General Covariant Stability Theory
Wanas, Mamdouh 1.,; Bakry, Mohamed Ao o 0 0 0 0 2131
Relativistic Generalization of the Inertial and Gravitational Masses
Equivalence Principle
Mitskievich, Nikolai V. o 0 0 0 0 0 0 0 0 2134
Static Perturbations by a Point Mass on a Schwarzschild
Black Hole
Bini, Donato; Geralico, Andrea; Ruffini, Remo 0 0 0 0 0 0 0 0 0 0 0 0 0 2137
Spatial Noncommutativity in a Rotating Frame
Beciu, Mircea 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2140
On Energy and Momentum of the Friedman and Some More
General Universes
Garecki, Janusz. . . . . . . . . . . . . . . . . . . . . . . . . . . 2143
Quasilocal Energy for an Unusual Slicing of Static Spherically
Symmetric Metrics
Chen, Chiang-Mei; Nester, James M. . . . . . . . . . . . . . . 2146
Quasilocal Energy for Cosmological Models
Nester, James M.; Chen, Chiang-Mei; Liu, Jian-Liang. . . . . . . . . . 2149
Relative Strains in General Relativity
Bini, Donato; de Felice, Fernando; Geralico, Andrea. . . . . . . . . . . 2152
Dyonic Kerr-Newman Black Holes, Complex Scalar Field and
Cosmic Censorship
Semiz, jbrahim . . . . . . . . . . . . . . . . . . . . . · . . . . 2155
The Ideas of GR, Quantization, Non-equilibrium Theormodynam-
ics and Gravimagnetism in Planetary Cosmogony
A bdil 'din, Meirkhan M.; Abishev, Medev E.; Beissen, Nurzada A. . . . . . 2158
• Wormholes, Energy Conditions and Time Machines
Chairperson: Hadley, Mark
N-Spheres: Regular Black Holes, Static Wormholes and Gravastars
with a Tube-Like Core
Zaslavskii, Oleg B. · . . . . 2169
Averaged Energy Inequalities for N on-Minimally Coupled Classical
Scalar Fields
Osterbrink, Lutz W. . . . . . . . . . . . . . . . . . . . . . . . . . 2172
Self-Sustained Traversable Wormholes and the Equation of State
Garattini, Remo · . . . . 2175
Classical and Quantum Wormholes in a Cosmology with Decaying
Dark Energy
Darabi, Farhad . . . . . . . . . . . . . . . . . . . . . . . . . . . 2178
Nariai-Bertotti-Robinson Spacetimes as a Building Material for
One-Way Wormholes with Horizons, but without Singularity
Mitskievich, Nikolai v.; Medina Guevara, Maria Guadalupe;
Rodriguez, Hector Vargas . . . . . . 2181
Cosmic Time Machines and Gamma Ray Bursts
De Felice, Fernando . . . . . . . . . . . . . . . . . . . . . . . 2184
Static and Dynamic Traversable Wormholes
Adamiak, Jaroslaw P. . . . . . . . . . . . . . . . . . . . . . . . . 2187
 Ii

Wormholes in the Accelerating Universe

Gonzalez-Diaz, Pedro F.; Martin-Moruno, Prado . . . . . . 2190
Traversable Wormholes Supported by Cosmic Accelerated Ex-
panding Equations of State
Lobo, Francisco S.N. . . . . . . . . . . . . . . . . . 2193
On Wormholes of Massless K-Essence
Estevez-Delgado, Joaquin; Zannias, Thomas . . . . . . . . . . 2196
Dynamic Wormhole Spacetimes Coupled to Nonlinear
Electrodynamics
Berrocal Arellano, Aaron V.; Lobo, Francisco S.N. . . . . . . . . . . . 2199
• Exact Solutions (Mathematical Aspects)
Chairperson: Alekseev, George
Robinson-Trautman Spacetimes in Higher Dimensions
Ortaggio, Marcello; Podolsky, JiN. . . . . . . . . · . . . . 2205
Solutions of Seiberg-Witten and Einstein-Maxwell-Dirac Equations
in Euclidean Signature
Cihan, Sar.;lioglu . . . . . . . . . . . . . . 2208
Euler Numbers on Cobordant Hypersurfaces
Harriott, Tina A.; Williams, Jeff G. . . . . . . . . . 2211
Symmetries of the Weyl Tensor in Bianchi V Spacetimes
Kashif, Abdul Rehman; Saifullah, Khalid; Shabbir, Ghulam S. . . . . . 2213
Classification of Spacetimes according to Conformal Killing Vectors
Saifullah, Khalid . . . . . . . . . . . . . . . . . . . . . . . . . . 2216
Exact Solutions for Radiating Relativistic Star Models
Misthry, Suryakumari S.; Maharaj, Sunil D. . . . . . . . . . . . . . . 2219
An EMP Model of Bianchi 1 Cosmology
Williams, Floyd L. . . . . . . . . . · . . . . 2222
Exact Static Solutions for Scalar Fields Coupled to Gravity in
(3+ 1)-Dimensions
Bilge, Ayse H.; Daghan, Durmus . . . . . . . . . . . . . . . . . . . 2225
Thermodynamic Description of Inelastic Collisions in General
Relativity
Neugebauer, Gernot; Hennig, Joerg . . . . . . . . . . . . · . . . . 2228
Distorted Killing Horizons and Algebraic Classification of Curva-
ture Tensors
Pravda, Vojtech; Zaslavskii, Oleg B. . . . . . . . . . . . . . . . . . . 2231
Quasi-Stationary Routes to the Kerr Black Hole
Meinel, Reinhard . . . . . . . . . . . . . . . . . . . . . . . . . . 2234
Iii

Classification Results on Purely Magnetic Perfect Fluid Models

WylIe man, Lode; Van den Bergh, Norbert . . . . . . . . . . . . 2237
Purely Electric Perfect Fluids of Petrov Type D
WylIeman, Lode . . . . . . . 2240
Self-Dual Fields on the Space of a Kerr-Taub-Bolt Instanton
Aliev, Alikram N.; Sar;lioglu, ethan . . . . . . . . . . . · . . . . 2243
The Kerr Theorem, Multisheeted Twistor Spaces and Multiparticle
Kerr-Schild Solutions.
Burinskii, Alexander · . . . . . 2246
Electrical Force Lines of a 2-Soliton Solution of the Einstein-
Maxwell Equations
Pizzi, Marco . . . . · . . . . . 2249
Monodromy Transform Approach in the Theory of Integrable
Reductions of Einstein's Field Equations and Some Applications
Alekseev, George · . . . . 2252
Closed Timelike Curves and Geodesics of Godel-Type Metrics
Sarioglu, Ozgiir. . . . . . . . . . . . . . · . . . . . 2255
Conformal Symmetries in Spherical Spacetimes
Maharaj, Sunil D.; Moopanar, Selvan . . . . · . . . . . 2258
A Theorem of Beltrami and the Integration of the Geodesic
Equations
Boccaletti, Dino; Catoni, Francesco; Cannata, Roberto; Zampetti, Paolo . . 2261
Gravitational Collapse and Horizon Formation in 2+ 1 Dimensional
Gravity
BTilI, Dieter; Khetarpal, Puneet. . . . . . . . . . . . . . . . . . . . 2264
Purely Magnetic Silent Universes Do Not Exist
Vu, Khai T.; Carminati, John . . . . . . . . . . . . . 2268
• Exact Solutions (Physical Aspects)
Chairperson: Scott, Susan M.
Zeeman-Type Dragging in the Kerr-Newman and NUT Spacetimes
Mitskievich, Nikolai v.; Lopez Benitez, Luis I. · . . . . 2273
Physical Implications for the Uniqueness of the Value of the
Integration Constant in the Vacuum Schwarzschild Solution
Mitra, Abhas . . . . . . . . . . . . . . . . . . . . . . . . . . . 2276
Singularity Analysis of Generalized Cylindrically Symmetric
Spacetimes
Konkowski, Deborah A.; Helliwell, Thomas M. . . . . . . . . . . . . . 2279
 liii

Some Properties of Kerr Geometry with a Repulsive Cosmological

Constant
Petrasek, Martin; Hled{k, Stanislav o 0 0 0 0 0 2282
Solution Generating Theorems: Perfect Fluid Spheres and the
TOV Equation
Boonserm, Petarpa; VisseT, Matt; Weinfurtner, Silke o 0 0 0 0 0 2285
Spherically Symmetric Gravitational Collapse of Perfect Fluids
Lasky, Paul; Lun, Anthony 0 0 0 0 0 0 0 0 0 0 0 0 o 0 0 0 0 0 2288
High-Speed Cylindrical Collapse of Two Dust Fluids
Sharif, Muhammad; Ahmad, Zahid 0 0 0 0 0 0 0 0 o 0 0 0 0 2291
Some Physical Consequences of the Multipole Structure of the
Kerr and Kerr-Newman Solutions
Rosquist, Kjell 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2294
Visualising Spacetimes via Embedding Diagrams
Hled{k, Stanislav; StuchHk, Zdenek; Cipko, Alois o 0 0 0 0 0 0 0 0 2299
Canonical Analysis of Radiating Atmospheres of Stars in
Equilibrium
Kovacs, Zolt6n; Gergely, Laszl6 Ao; Horvath, Zsolt 0 0 0 0 0 0 0 0 0 0 0 2302
• Self-Gravitating Systems
Chairperson: Mielke, Eckehard Wo
Platonic Sphalerons in Einstein-Yang-Mills and Yang-Mills-Dilaton
Theory
Kleihaus, Burkhard; Kunz, Jutta; Myklevoll, Kari 0 0 0 0 0 0 0 0 0 0 0 0 2307
Comment on "General Relativity Resolves Galactic Rotation
without Exotic Dark Matter" by FoI. Cooperstock and So Tieu
Fuchs, Burkhard; Phleps, Stefanie 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2310
Solitonic and Non-Solitonic Q-Stars
Verbin, Yosef 0 0 0 0 0 0 0 0 0 0 o 0 0 0 0 0 0 0 2313
Rotating Monopole-Antimonopole Pairs and Vortex Rings
Neemann, Ulrike; K1LnZ, Jutta; Kleihaus, Burkhard 0 0 0 0 0 0 0 0 2316
Sources of Static Cylindrical Spacetimes
Z6jka, Martin o 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2319
Gravitating Multi-Skyrmions
Kleihaus, Burkhard; Ioannidou, Theodora; Kunz, Jutta 0 0 0 0 0 0 0 0 0 0 2322
A New Exact Static Thin Disk with a Central Black Hole
Gonzalez, Guillermo Ao o 0 0 0 0 0 0 0 2325
liv

Bifurcations of Nonlinear Curvature Lagrangians in the Boson

Star Model
Schunck, Franz E. 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2328
Approximate Dynamics of Dark Matter Ellipsoids
Bisnovatyi-Kogan, Gennadyl So; Tsupko, Oley YUo o 0 0 0 0 2331
Nonextensive Statistical Theory of Density Distributions in Grav-
itationally Clustered Structures
Leubner, Manfred Po 0 0 0 0 o 0 0 0 0 0 0 0 0 0 0 0 2334
General Relativistic Accretion with Backreaction
Karkowski, Janusz; Kinasiewicz, Bogusz; Mach, Patryk; Malec,
Edward; SWierczyriski, Zdobyslaw 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2337
Non-Homogeneous Axisymmetric Models of Self-Gravitating
Systems
Cherubini, Christian; Filippi, Simonetta; Ruffini, Remo; Sepul-
veda, Alonso; Zuluaga, Jorge I. 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2340
Gravitational Wave Damping from a Self-Gravitating Vibrating
Ring of Matter around a Black Hole
Basu, Prasad; Chakrabarti, Sandip K. 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2343
Variational Principles and Hamiltonian Formulation of Spherical
Shell Dynamics
Kijowski, Jerzy; Magli, Giulio; Malafarina, Daniele 0 0 0 0 0 0 0 0 0 0 0 2346
• Operating G W Detectors
Chairperson: Bassan, Massimo
Virgo Commissioning Progress
Barsuglia, Matteo; for the Virgo Collaboration o 0 0 0 0 0 2351
Results from LIGO Observations: Stochastic Background and
Continuous Wave Signals
Christensen, Nelson; for the Ligo Scientific Collaboration o 0 0 0 0 2356
Explorer and Nautilus Gravitational Wave Detectors - A Status
Report
Bassan . Massimo; for the ROG Collaboration 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2359
AURIGA on the Air: Sensitivity, Calibration, Diagnostics and
Observations
Ortolan, Antonello; for the A URIGA collaboration 0 0 0 0 0 0 0 0 0 0 0 2365
• Advanced GW Detectors
Chairper-son: Blair, David
Optical Spring at Thermal Equilibrium
Di Virgilio, Angela 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2373
 Iv

Measurements of Electrical Charge Distribution Variations on

Fused Silica
Prokhorov, Leonid G.; Mitrofanov, Valery P. . . . . . . . . · . . . . 2376
Developments toward Monolithic Suspensions for Advanced Grav-
itational Wave Detectors
Heptonstall, Alastair; Cantley, Caroline; Crooks, David; Cumming,
Alan; Hough, James; Jones, Russell; Martin, lain; Rowan, Sheila;
Cagnoli, Gianpietro . . . . . . . . . . . . . . . . . . . . . . . . 2379
Concept Study of Yukawa-like Potential Tests Using Dynamic
Gravity-Gradients with Interferometric Gravitational-Wave
Detectors
Raffai, Peter; Mdrka, Szabolcs; Matone, Luca; Marka, Zsuzsa. . . . . . . 2382
Astrophysical Sources of the Gravitational Waves
Lipunov, Vladimir M. . . . . . . . . . . . . . . . . . . . . . . . 2385
• Space and Third Generation GW Detectors
Chairperson: Hough, Jim
DECIGO: The Japanese Space Gravitational Wave Antenna
Ando, Masaki; et al. . . . . . . . . . . . . . . . . . . · . . . . 2393
Design and Construction of the LISA Technology Package Optical
Bench Interferometer
Killow, Christian J.; Bogenstahl, Johanna; Perruer-Lloyd,
Michael; Ward, Henry; Robertson, David 1.; Guzman Cervantes,
Felipe; Steier, Frank. . . . . . . . . . . . . . . . . . . . . . . . 2398
Compact Binary Inspiral and the Science Potential of Third-
Generation Ground-Based Gravitational Wave Detectors
Van Den Broeck, Chris; Sengupta, Anand S. . . . . . · . . . . 2401
Discrete Sampling Variation Measurement Technique for Sub-SQL
Sensitivity Detection of Gravitational Waves
Danilishin, Stefan L.; Khalili, Farid Ya. . . . . . . . . . . . . . . . . 2404
The Detection of Gravitational Waves with Matter Wave
Interferometers
Delva, Pacome; Angonin, Marie-Christine; Tourrenc, Philippe . . . . . . 2407
• G W Data Analysis
Chairperson: Ricci, Fulvio
Detecting LISA Sources Using Time-Frequency Techniques
Gair, Jonathan R.; Jones, Gareth. . . . . . . . . . . · . . . . 2413
Determining the Neutron Star Equation of State using the Narrow-
Band Gravitational Wave Detector Schenberg
de Araujo, Jose Carlos N; Marranghello,. Guilherme F. . . . . . . . . . 2416
Ivi

Approximate Waveform Templates for Detection of Extreme Mass

Ratio Inspirals with LISA
Gair, Jonathan R. · . . . . . 2419
GW-Detector's Output Processing at the Non-Gaussian Noise
Background
Gusev, Andrei v.; Popov, Serghei M.; Rudenko, Valentin · . . . . 2422
Detecting a Stochastic Background of Gravitational Waves in the
Presence of Non-Gaussian Noise
Himemoto, Yoshiaki . . . . . . . . . . . . . . . . . . . . . . . . . 2426
Coincidences between the Gravitational Wave Detectors EX-
PLORER and NAUTILUS in the Years 1998, 2001, 2003 and
2004
Pizzella, Guido . . . . . · . . . . . 2429
Incoherent Strategies for the Network Detection of Periodic
Gravitational Waves
Astone, Pia; Frasca, Sergio; Palomba, Cristiano · . . . . 2438
Search for Continuous Gravitational Waves: Simple Criterion for
Optimal Detector Networks
Prix, Reinhard . . . . . . . . . . . . . . . . . . . . . . . . . . . 2441
First Coincidence Search among Periodic Gravitational Wave
Source Candidates Using Virgo Data
Palomba, Cristiano; for the Virgo Collaboration. . . . . . . · . . . . 2444
Primordial Black-Hole Gravitational Wave Background Noise in
the LISA, DECIGO and BBO Frequency Bands
de Araujo, Jose Carlos N.; Aguiar, Odylio D.; Miranda, Oswaldo P. . . . . 2448
• Recent Advances in the History of General Relativity
Chairperson: Renn, Juergen
The Einstein-Varicak Correspondence on Relativistic Rigid
Rotation
Sauer, Tilman . . . . . . . . . . . 2453
The History of the So-Called Lense-Thirring Effect
Pfister, Herbert . . . . . . . . . . . . . . . . . . . . . . . 2456
M.-A. Tonnelat's Research Concerning Unified Field Theory
Goenner, Hubert . . . . . . . . . . . . . . . . . · . . . . . 2459
Rosenfeld, Bergmann, Dirac and the Invention of Constrained
Hamiltonian Dynamics
Salisbury, Donald C. . . . . . . . . . . . . . . . . . . . . . . . . 2467
 Ivii

Stellar and Solar Positions in 1701-1703 Observed by Francesco

Bianchini at the Clementine Meridian Line in the Basilica of Santa
Maria degli Angeli in Rome, and its Calibration Curve
Sigismondi, Costantino . . . . . . . . . . . . . . . . . . . . . . . 2470
• Strong Gravity and Binaries
Chairperson: Blanchet, Luc
The Effacing Principle in the Post-Newtonian Mechanics
Kopeikin, Sergei; Vlasov, Igor . . . . . . . . . 2475
Gravitational Waves of a Lense-Thirring System
Vasuth, Matyas; Majrir, Janos . . . . . . . . . . . . 2478
York Map, Non-Inertial Frames and the Physical Interpretation of
the Gauge Variables of the Gravitational Field
Lusanna, Luca . . . . . . . . . . . . . . . . . . . . . . . . . . . 2481
• Post-Newtonian Dynamics in Binary Objects
Chairperson: Schaefer, Gerhard
Accurate and Efficient Gravitational Waveforms for Certain
Galactic Compact Binaries
Tessmer, Manuel; Gopakumar, Achamveedu · . . . . 2487
Dimensional Regularization of the Gravitational Interaction of
Point Masses in the ADM Formalism
Damour, Thibault; Jaranowski, Piotr; Schaefer, Gerhard . . . . . . . . . 2490
New Results at 3PN via an Effective Field Theory of Gravity
Porto, Rafael A. . . . . . . . . . . . . . . . . . . . . . . . . 2493
Orbital Phase in Inspiraling Compact Binaries
Vasuth, Matyas; Mikoczi, Balazs; Gergely, Laszlo A. . . . . . . 2497
Gravitational Wave Emission from a Stellar Companion Black
Hole in Presence of an Accretion Disk Around a Kerr Black Hole
Basu, Prasad; Chakrabarti, Sandip K.; Mondal, Soumen; Goswami,
K ushalendu . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2500
The Second Post-Newtonian Order Generalized Kepler Equation
Gergely, Laszlo A.; Keresztes, Zolton; Mikoczi, Balazs · . . . . 2503
• Tests of Local Lorentz Invariance
Chairperson: Peters, Achim
The Standard-Model Extension and Tests of Relativity
Russell, Neil . . . . . . . . . . . . . . . . . . . · . . . . 2509
New Measurements of the One-Way Speed of Light and its Relation
to Clock-Comparison Experiments
Unnikrishnan, C.S. . . . . . . . . . . . . . . . . . . . . . . . . . 2512
Iviii

Test of Time Dilation with a Two-Velocity Atomic Clock

Saathoff, Guido; Karpuk, Sergej; Reinhardt, Sascha; BuM, Hen-
rik; H iinsch, Theodor Wo ; Holzwarth, Ronald; Huber, Gerhard;
Novotny, Christian; Schwalm, Dirk; Udem, Thomas; Wolf, An-
dreas; Zimmermann, Marcus; Gwinner, Gerald 0 0 0 0 0 0 0 0 0 0 0 0 0 2515
• Laboratory Gravity Tests
Chairperson: Laemmerzahl, Claus
Atom Interferometry for Precision Tests of Gravity: Measurement
of G and Test of Newtonian Law at Micrometric Distances
Bertoldi, Andrea; Cacciapuoti, Luigi; de Angelis, Marella;
Drullinger, Robert E.; Ferrari, Gabriele; Lamporesi, Giacomo;
Poli, Nicola; Prevedelli, Marco; Sorrentino, Fiodor; Tino,
Guglielmo Mo 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 o 0 0 0 0 2519
Development of Accelerometer Prototype for Testing the Equiva-
lence Principle in Free Fall
Iafolla, Valerio; Lucchesi, David; Milyukov, Va dim; Nozzoli,
Sergio; Santoni, Francesco; Shapiro, Irvin 10; Lorenzini, Enrico
G.; Cosmo, Mario Lo; Ashenberg, loshua; Cheimets, Peter N.;
Glashow, Sheldon 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2530
Measurement of the Gravitational Constant G
Meyer, Hinrich; Kleinevoss, U~f; Piel, Helmut 0 0 0 0 0 0 0 0 0 0 0 0 0 2534
Solar Radius at Minimum of Cycle 23
Sigismondi, Costantino o 0 0 0 0 0 2537
The Newtonian Gravitational Constant: Modern Status and
Perspective of New Determination
Milyukov, Vadim; Luo, lun o 0 0 0 0 0 2540
Relativistic Astrometry with Gaia: Advances in the RAMOD
Project
Bucciarelli, Beatrice; Crosta, Maria Teresa; Lattanzi, Mario Go;
Vecchiato, Alberto; Preti, Giovanni; de FeZ,ice, Fernando 0 0 0 0 0 0 0 2543
• Clock and Space Tests of Gravity
Chairperson: Salomon, Christophe
Dynamical Clock Synchronization in Einstein's Theory: Implica-
tions for ACES mission of ESA
Lusanna, Luca 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2549
STEP Prototype Development Status
Mehls, Carsten et al. o 0 0 0 0 0 0 0 0 0 2553
On Stellar System Tests of the Cosmological Constant
Sereno, Mauro; letzer, Philippe 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2556
 lix

The Lense-Thirring Effect and the Pioneer Anomaly: Solar System

Tests
Iorio, Lorenzo · . . . . 2558
The Equivalence Principle and Its Tests in the Context of Gravity,
Quantum Mechanics and Cosmology
Unnikrishnan, C.S. . . . . . . . . . . . . . . . . . . . . . . . . . 2561
The Flyby-Anomaly
Liiemmerzahl, Claus; Dittus, Hansjoerg . . . . . . . . . . . . . . . . 2564
Gravity Tests and the Pioneer Anomaly
Jaekel, Marc-Thierry; Reynaud, Serge . . . . . . . . . . . . . . . . 2567
• Astrometry
Chairperson: Klioner, Sergei
A Nice Tool for Relativistic Astrometry: Synge's World Function
Teyssandier, Pierre; Le Poncin-Lafitte, Christophe · . . . . 2573
Lunar Laser Ranging: A Space Geodetic Technique to Test
Relativity
Muller, Jurgen . . . . . . . . . . . . . . . 2576
APOLLO: Next Generation Lunar Laser Ranging
Murphy, Thomas W. Jr.; Michelsen, Eric L.; Orin, Adam E.;
Battat, James B.; Stubbs, Christopher W.; Adelberger, Eric G.;
Hoyle, C.D.; Swanson, H. Erik . . . . . . . . . . . . . . . . . . . . 2579
Metric Extensions of General Relativity and Gravity Tests in the
Solar System
Reynaud, S erge; Jaekel, M arc- Thierry . . . . . . . . . . . . . . . . . 2582
Measurement of the PPN-, Parameter with a Space-Born Dyson-
Eddington-like Experiment
Vecchiato, Alberto; Gai, Mario; Lattanzi, Mario G.; Morbidelli, Roberto . . 2585
Relativistic Light Deflection near Giant Planets Using Gaia
Astrometry
Anglada-Escude, Guillem; Klioner, Sergei A.; Torra, Jordi . . . . . . . . 2588
Astrometrical Microlensing with Radioastron
Zakharov, Alexander F. . . . . . . . . . . . . . . . 2591
Asteroidal Occultation of Regulus: Differential Effect of Light
Bending
Sigismondi, Costantino; Troise, Davide · . . . . 2594
Testing General Relativity by Astrometric Measurements Close to
Jupiter, the Real GAREXPart II
Crosta, Maria Teresa; Gardiol, Daniele; Lattanzi, Mario G.;
Morbidelli, Roberto . . . . . . . . . . . . . . . . . . . . . . . 2597
Ix

Relativistic Tests from the Motion of Asteroids

Hestroffer, Daniel; Mouret, Serge; Berthier, Jerome; Mignard, Franr;ois . . 2600
• Quantum Gravity Phenomenology
Chairperson: Amelino-Camelia, Giovanni
Effective Vacuum Refractive Index from Gravity and Present
Ether-Drift Experiments
Consoli, Maurizio . . . . . . . . . . . . . . · . . . . . . . . . 2605
Quantum Gravity Effects in Rotating Black Holes
Reuter, Martin; Tuimn, Erick · . . . . . . . . . 2608
Lorentz Symmetry from Lorentz violation in the Bulk
Bertolami, Orfeu; Carvalho, Carla · . . . . . . . . . 2611
Quantum Gravity and Spacetime Symmetries
Lehnert, Ralf. . . . . . . . . . . . . . . . . . . . 2615
Lorentz Invariance Violation in Higher Order Electrodynamics
Lorek, Dennis; Liiemmerzahl, Claus . . . . . . . . . . . . . . . . 2618
Hubble Meets Planck: A Cosmic Peek at Quantum Foam
Ng, Y. Jack . . . . . . . '. . . . . . . . . . . . . . . . . . . . . 2621
Evolutionary Reformulation of Quantum Gravity
Montani, Giovanni . . . . . . . 2626
Kerr's Gravity as a Quantum Gravity on the Compton Level
Burinskii, Alexander . . . . . . . 2631
A Link between General Relativity and Quantum Mechanics
Rosquist, Kjell . . . . . . . . . . . . . . . . . . . . . . . . . . . 2634
Spacetime Fluctuations and Inertia
GoklU, Ertan; Liiemmerzahl, Claus; Camacho, Abel; Macias, Alfredo. . . . 2639
Quantum Gravity in Cyclic (Ekpyrotic) and Multiple (Anthropic)
Universes with Strings and/or Loops
Chung, T.J. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2642
• Quantum Fields
Chairperson: Belinski, Vladimir
Quantum Liouville Theory with Heavy Charges
Menotti, Pietro; Tonni, Erik . . . . . . . . . . . . . . . . . 2647
On the Path Integral for Non-Commutative (NC) QFT
Dehne, Christoph . . . . . . . . . . . . . . . . . . . . . 2650
An Irreducible Form for the Asymptotic Expansion Coefficients of
the Heat Kernel of Fermions
Yajima, Sat os hi; Higasida, Yoji; Fukuda, Makoto; Tokuo, Shoshi;
Kubota, Shin-Ichiro; Kamo, Yuki . . . . . . . . 2653
 Ix;

Quantum Anomalies for Generalized Euclidean Taub-Newman-

U nti-Tamburino Metrics
Visinescu, Mihai; Visinescu, Anca . . . . . . 2656
A New Expression for the Transition Rate of an Accelerated
Particle Detector
Louko, Jorma; Satz, Ale)andro . . . . . 2659
On the Geometrization of the Electromagnetic Interaction for a
Spinning Particle
Cia nfran i, Fr'ancesco; Milillo, Irene; Montani, Giovanni . . . . . . . . . 2662
Can EPR Correlations be Driven by an Effective Wormhole?
Santini, Eduardo Serg'to . . . . . . . . . . . . . . . . . . . . . . . 2665
Is Torsion a Fundamental Physical Field?
Lecian, Orchidea Maria; Mercuri, Simone; Montani, Giovanni . . . . . . 2668
Unitary Quantization of the Gowdy T3 Cosmology
Corichi, Alejandro; Cortez" Jeronimo; Mena Marugan, Guilter-mo A. . . . 2671
On the Interaction of the Gravitational Field of a Cosmic String
with Some Quantum Systems
Marques, Geusa; Bezerra, Valdir B. . . . . . . . . . . . . . . . . . . 2674
Einstein-Rosen Waves Coupled to Matter
Barbero Gonzalez, Jesus Fernando; Garay, Inaki; Villasenor,
Eduardo J.S. . . . . . . . . . .......... . . . . . . 2677
Electromagnetic Radiation from a Charge Rotating in
Schwarzschild Spacetime
Castine'tras, Jorge; Crispino, Luis C.B.; Murta, Rodrigo; Matsas,
George E.A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2680
Recent Developments in Quantum Energy Inequalities
Fewster-, Chr-istopher J. . . . . . . . . . . . . . . . . . . . . . . . 2683
Black Holes as Boundaries in 2D Dilaton Supergravity
Bergamin, Luzi; Grumiller, Daniel . . . . . . 2686
Quasinormal Modes for Arbitrary Spins in the Schwarzschild
Background
Khriplovich, Iosi!; Ruban, Gennady . . . . . . . . . . . . 2692
Can Quantum Mechanics Heal Classical Singularities?
H eltiwell, Thomas M.; K onkowski, Deborah A. . . . . . 2695
Quantizing Two-Dimensional Dilaton Gravity with Fermions: The
Vienna Way
Meyer, Rene . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2698
Ixii

Vacuum Polarization for a Spinor Massive Field in an Einstein-

Maxwell Spacetime
Bezerm, Valdir B.; Khusnutdinov, Nail R. . . . . . . . . . . . . . . . 2701
• Casimir Effect and Short-Range Gravity
Chairperson: Mostepanenko, Vladimir
The Casimir Effect in Relativistic Quantum Field Theories
Mostepanenko, Vladimir M. · . . . . 2707
Local and Global Casimir Energies in a Green's Function Approach
Milton, Kimbal A.; Cavera-Pelaez, Ines; Kirsten, Klaus · . . . . 2727
Boundary Induced Quantum Fluctuation Effects: From Moving
Mirror to Electron Coherence
Hsiang, Jen-Tsung; Lee, Da-Shin . . . . . . . · . . . . 2746
A Theory of Electromagnetic Fluctuations for Metallic Surfaces
and van der Waals Interactions between Metallic Bodies
Bimonte, Giuseppe · . . . . 2749
Theory of the Casimir Effect between Dielectric and Semiconductor
Plates
Klimchitskaya, Galina L.; Geyer, Bora · . . . . 2752
A Novel Experimental Approach for the Measure of the Casimir
Effect at Large Distances
Antonini, Piergiorgio; Bressi, Giacomo; Carugno, Giovanni;
Galeazzi, Giuseppe; Messineo, Giuseppe; Ruoso, Giuseppe · . . . . 2755
Measurement of the Casimir Force in the Range above 5 Microns
and Detection of the Finite Temperature Effect
Rajalakshm, Gurumukthy 1.; Suresh, Domvari; Cowsik, Ramanath;
Unnikrishnan, C.S. . . . . . . . . . . . . . . . . . . . . . . . 2758
Scalar Casimir Effect with Non-Local Boundary Conditions
Saharian, Amm; Esposito, Giampiera . . · . . . . . . 2761
Sample Dependence of the Casimir Force
Pirazhenko, Irina; Lambrecht, Astrid, Svetovoy, Vitaly B. · . . . . . . 2764
Casimir Interaction between Absorbing and Meta Materials
Intmvaia, Fmncesco; Henkel, Carsten . . . · . . . . . . 2767
Casimir Energy and a Cosmological Bounce
Herdeira, Carlos A.R. . . . . . . . . . . · . . . . 2770
Photon Generation from the Vacuum: An Experiment to Detect
the DCE
Bmggio, Caterina; Bressi, Giacomo; Carugno, Giovanni; Ruoso,
Giuseppe; Zanello, Dino . . . . . . . . . . . . . . . . . . . . . . . 2773
 Ixiii

• Loop Quantum Gravity, Quantum Geometry, Spin Foams

Chairperson: Lewandowski, J erzy
The Emergence of AdS z from Quantum Fluctuations
Ambj¢m, Jan; Janik, Romuald; Westra, Willem; Zohren, Stefan . . . . . 2779
The Ponzano-Regge Model and Reidemeister Torsion
Barrett, John W.; Naish-Guzman, Ileana. . . . . . . . . . . . 2782
The Proca-Field in Loop Quantum Gravity
Helesfai, Gabor . . . . . . . . . . . . . . . . . . . . 2785
Ambiguity of Black Hole Entropy in Loop Quantum Gravity
Tamaki, Takashi; Nomura, Hidefumi . . . . . 2788
Exploring the Diffeomorphism Invariant Hilbert Space of a Scalar
Field
Sahlmann, Hanno . . . . . . . . . . . . . . . . . . . . . . . . . . 2791
Nieh-Yan Invariant and Fermions in Ashtekar-Barbero-Immirzi
Formalism
Mercuri, Simone . . . . . . . . 2794
A Generalized Schroedinger Equation for Loop Quantum
Cosmology
Salisbury, Donald C.; Schmitz, Allison. . . . . . . . . . . . . . . 2797
Spectral Analysis of the Volume Operator in Loop Quantum
Gravity
Brunnemann, Johannes; Rideout, David . . . . . . . . . . . . . . . . 2800
Counting Entropy in Causal Set Quantum Gravity
Zohren, Stefan; Rideout, David . . . . . . . . . . . . . . . . . . 2803
Algebraic Approach to 'Quantum Spacetime Geometry'
Raptis, Ioannis; Wallden, Petros; Zapatrin, Roman . . . . . . . . . . . 2806
Noncummutative Translations and *-Product Formalism
Daszkiewicz, Marcin; LukieTski, JeTzy; Woronowicz, Mariusz . . . . . . . 2809
• Brane Worlds and String Motivated Cosmology
Chairperson: Galtsov, Dmitry
Black Holes on Cosmological Branes
GeTgely, Laszl6 A. . . . . . . . . . . . . . . . . 2815
Generalized Cosmological Equations for a Thick Brane
Khakshoumia, Samad . . . . . . . . . . . . . . . . . . . . . . . . 2818
Cerenkov Radiation from Collisions of Straight Cosmic
(Super )Strings
Melkumova, Elena; Gal'tsov, Dmitri v.; Salehi, KaTim . . . . . . . . . 2821
Ixiv

High-Energy Effects on the Spectra of Cosmological Perturbations

in Braneworld Cosmology
Hiramats1L, Takashi; Koyama, Kazuya; Tamya, Atsushi . . . . . . . . . 2824
Braneworlds and Quantum States of Relativistic Shells
Ansoldi, Stefano . . . . . . . . . . . . . . . . . . . . . . . . . . 2827
Rotating Braneworld Black Holes
Aliev, Alikram N. . . . . . . . · . . . . 2830
General Solution for Scalar Perturbations in Bouncing Cosmologies
Bozza, Valerio · . . . . 2833
Constraints on Accelerating Brane Cosmology with Exchange
between the Bulk and Bmne
Mathews, Grant 1.; Umezu, Ken-Ichi; Kajino, Toshitaka; Ichiki,
Kiyomoto; Nakamura, Ryoko; Yahiro, Masanobu · . . . . 2836
Testing DGP Modified Gravity in the Solar System
Iorio, Lorenzo · . . . . 2839
The Dynamics of Scalar-Tensor Cosmology from RS Two-Brane
Model
Kuusk, Piret; Jarv, Laur; Saal, Margus . . . . . . . . . . . . . . . . 2842
Self-T-Dual Brane Cosmology
Rinaldi, Massimiliano . . . . . . . . . . . . . . . . . . . . . . . . 2845
• Brane Worlds
Chairperson: Bianchi, Massimo
Catching Photons from Extra Dimensions
Dobado, Antorio; Maroto, Antorio L.; Cembranos, Jose A.R. . . . . . . . 2851
Lorentz Invariance Violation in Braneworld Models
KOToteev, Peter A. . . . . . . 2854
The Bazanski Approach in Brane-Worlds: A Brief Introduction
Kahil, Magd Elias. . . . . . . . . . . . . . . . . . . . . . . . . . 2857
• M-Theory and Dualities
Chairperson: Stelle, Kellogg
M-Theory and Dualities
Mac Conamhna, Oisin. . . . . . . . . . . . . . . . . . . . . . . . 2863
AdS Spacetimes in M-Theory
Gauntlett, Jerome P.; Mac Conamhna, Oisin A.P.; Mateos, Toni;
Waldram, Daniel . . . . . . . . . . . . . . . . . . . . . . . . . . 2875
Global Aspects of Seven-Brane Configurations
BergshoefJ, Eric A.; Hartong, Jelle; Ortin, Tomas; Roest, Diederik . . . . 2878
 Ixv

Duality and Black Hole Partition Functions

Mohaupt, Thomas 0 0 0 0 0 0 0 0 0 0 o 0 0 0 0 0 0 0 0 0 0 0 0 0 2881
M-Theory on Calabi-Yau Five-folds
Haupt, Alexander So; Stelle, Kellogg So 0 o 0 0 0 0 2884
Hagedorn Transition and Chronology Protection in String Theory
Herdeiro, Carlos AoRo 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2887
KK-Masses and Dipole Theories
Landsteiner, Karl; Montero, Sergio 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2890

List of Participants 2893

Author Index 0 0 0 2911
This page intentionally left blank
Parallel Sessions
This page intentionally left blank
Dark Matter
This page intentionally left blank
IMPACT OF DARK MATTER ON REIONIZATION AND HEATING

MICHELA MAPELLI
SISSA/ISAS, Via Beirut 2-4, Trieste 1-34014, Italy, [email protected]

EMANUELE RIPAMONTI
Kapteyn Astronomical Institute, University of Groningen, Postbus 800, 9700 A V,
Groningen, The Netherlands, [email protected]

We derived the evolution of the energy deposition in the intergalactic medium (IGM) by
different decaying (or annihilating) dark matter (DM) candidates. Heavy annihilating
DM particles (with mass larger than a few GeV) have no influence on reionization and
heating, even if we assume that all the energy emitted by annihilations is absorbed by
the IGM. In the case of lighter particles, the impact on reionization and heating depends
on the efficiency of energy absorption by the IGM. We calculated the fraction of energy
produced by decays and annihilations which is effectively absorbed by the IGM. We
found that this fraction is generally high at very high redshift (» 100), but drops at
more recent epochs.

1. Introduction

The reionization and heating history of the intergalactic medium (IGM) has not yet
been fully understood. In particular, the nature of the sources of these processes
is mostly unclear. An important contribution might arise from the first stars; but
other more exotic sources (e.g. intermediate mass black holes, decaying dark matter
particles, etc.) have been proposed as well. l In this proceeding, we focus OIl the
possible role of dark matter (DM) decays and annihilations on reionization and
heating of the IGM.

2. The energy absorbed fraction

The rate of energy released by DM decays/annihilations and absorbed (per baryon)
by the rGM can be expressed as
feZ) = fabs(Z) nDM(Z) mDMc 2, (1)
where fabs(Z) is the fraction of energy absorbed by the IGM, mDM is the mass of
the DM particle and il.DM(Z) is the decrease rate of the number of DM particles per
baryon. The expression of nDM(z) depends on the density of DM particles and on
the decays lifetime or on the thermally averaged annihilation cross-section, in the
case of decays and annihilations, respectively. 2
The most crucial parameter in equation (1) is fabs(Z), Previous studies 3 assumed
a complete and immediate energy absorption by the rGM (i.e. fabs(Z) = 1), deriving
upper limits to the effects of DM on reionization and heating. Annihilating heavy
DM particles (mDMc 2 ~ GeV) have no effects on such processes, even assuming
fabs(Z) = 1, because of the very small allowed interacting rate. Instead, under the
assumption of complete absorption, lighter particles (mDMc 2 ;S 100 MeV) could be
important sources of partial early reionization and heating. 3

979
980

fabs(Z) can be hardly calculated in the case of heavy parti-

cles (mDMc 2 »100 MeV), because of the uncertainties in modeling the cascade
associated with massive product particles a . On the other hand, for relatively light
(mDMc 2 ~ 100 MeV) DM candidates it is possible to derive the correct behaviour of
the absorbed fraction 2 fabs(Z). In fact, the possible decay/annihilation products of
these light particles are only photons, electron-positron pairs, or neutrinos (which
are assumed to have negligible interactions with matter). For photons the effects
of Compton scattering and photo-ionization must be considered; for pairs, the rele-
vant processes are inverse Compton scattering, collisional ionizations, and positron
annihilations.
If the decay products are both photons and pairs, fabs(Z) is found to be high at
early epochs (z » 100), when photo-ionization (considering photons) and inverse
Compton scattering or/and positron annihilations (considering pairs) are efficient
processes. In both cases, fabs(Z) drops at lower redshifts. As an example, in Fig. 1 we
show the behaviour of fabs(Z) in the case of sterile neutrinos which decay into pho-
tons (left panel) and light dark matter (LDM) particles which annihilate producing
pairs (right panel). Similar results are obtained for decaying LDM particles. 2

1.000 1.0
lV.e'-l

A \.,e'\J /,/
C C
0 0
°13 'B J Al
E: ev
""
'0
J:
'0
-e" "
~ 0.1 _'~<lfev
:;:
~
~~ 15
~ ~
0-"
0.001
.c"
.,~
!°Atev

5 10 2030 50 100 200 500 1000 5 10 2030 50 100 200 500 1000
z
(a) (b)
Fig. 1. Absorbed fraction as a function of redshifts for sterile neutrino decays (a) and LDM
annihilations (b).

3. Effects of decays and annihilations on reionization and heating

We can now calculate how the behaviour of fabs(Z) affects the impact of DM de-
cays/ annihilations on reionization and heating. 2 As an example, in Fig. 2 we again
consider the case of sterile neutrinos decaying into photons (left panel) and of LDM
particles annihilating into pairs (right panel). The thin and thick lines in both pan-
els refer to the case where we assume complete absorption and to the case in which

aModeling the cascade is not only difficult, but also not necessary for our purposes, since the
effects of massive particles on reionization are negligible,3 even if we assume fabs(Z) = 1.
 981

10 3
103
g 102 g 10 2
f- lO f-
lO
10- 3 0.1
10-' 10- 2
f-" 10- 5 - f-" 10- 3
10- 6 10-'
10- 2
10- 3
><"
><" 10- 3

10 100 1000 10 100 1000

z z
(a) (b)
Fig. 2. Ionized fraction (bottom panels), Thomson optical depth (central panels) and IGM tem-
perature (top panels) as a function of redshift due to sterile neutrino decays (a) and LDM anni-
hilations (b). The thin dot-dashed line represents, from bottom to top, the relic fraction of free
electrons, their contribution to Thomson optical depth and the IGM temperature without particle
decays. In the top panel, the thin dotted line represents the CMB temperature. In the left (right)
panel the thin and the thick solid (dashed) lines refer to 25-keV sterile neutrino decays (3-MeV
LDM annihilations), in the hypothesis of complete energy absorption and taking into account the
effective absorbed fraction fabs(Z) (Fig. 1), respectively.

fabs(Z) is the same as shown in Fig. 1, respectively. The effect of sterile neutrino
decays on reionization and heating is strongly suppressed, if we consider the correct
fabs(Z), A similar result can be found for LDM decays.2 Also in the case of LDM
annihilations this suppression is apparent, even if less dramatic.
In summary, we can conclude that the correct calculation of fabs is crucia1 2 : if
the correct values of fabs(Z) are taken into account, the impact of DM decays and
annihilations on reionization and heating is almost negligible, a factor ~ 2 to 1000
lower than previous estimates based on the hypothesis of complete and immediate
absorption.

Acknowledgments
MM acknowledges the organizers of the Eleventh Marcel Grossmann meeting for the
MGF grant. The authors thank P. L. Biermann for inviting them to the meeting.
ER acknowledges support from NWO grant 436016.

References
1. B. Ciardi and A. Ferrara, Space Science Reviews 116 625 (2005), and references therein.
2. E. Ripamonti, M. Mapelli and A. Ferrara, MNRAS, accepted, astro-ph/0606482.
3. M. Mapelli, A. Ferrara and E. Pierpaoli, MNRAS 369 1719 (2006), and references
therein.
 IMPACT OF DARK MATTER DECAYS AND ANNIHILATIONS
ON STRUCTURE FORMATION

EMANUELE RIPAMONTI
Kapteyn Astronomical Institute, University of Groningen, Postbus 800, 9700 AV,
Groningen, The Netherlands, [email protected]

MICHELA MAPELLI
SISSA/ISAS, Via Beirut 2-4, Trieste 1-34014, Italy, [email protected]

We derived the influence of dark matter (DM) decays and annihilations on structure
formation. The energy deposited by DM decays and annihilations into metal free halos
both increases the gas temperature and enhances the formation of molecules. Within
the primordial halos the temperature increase generally dominates over the molecular
cooling, slightly delaying the collapse. In fact, the critical mass for collapse is generally
higher than in the unperturbed case, when we consider the energy input from DM. In
presence of DM decays and/or annihilations the fraction of baryons inside collapsed
metal free halos should be slightly less (~ 0.4) than the expected cosmological value.

1. Introduction
The formation of the first luminous objects is heavily influenced both by the chem-
ical abundance of coolants and by any source of heating. In particular, higher tem-
peratures can prevent the collapse of halos, whereas a greater abundance of H2 and
HD molecules (which are the main coolants of the metal free Universe) enhances
the cooling of the gas, favouring the collapse. In principle, reionization sources can
both enhance the abundance of molecules, since free electrons act as catalysts of H2
and HD, and increase the temperature of the gas.

2. Heating and molecular abundance enhancement from DM

DM decays and annihilations can be sources of heating and partial early reioniza-
tionJ Thus, they are also expected to affect the abundance of molecules. Fig. l.a
shows the temperature and the fractional abundance of free electrons, H2 and HD
in the intergalactic medium (IGM) as a function of redshift in the unperturbed case
(solid line) and if we switch on different models a of DM decays and annihilations
(dashed line; from top to bottom: sterile neutrino decays, light dark matter (LDM)
decays and LDM annihilations). DM decays and annihilations both heat the gas 1
and increase the abundance of free electrons, which enhances the formation of H2
and HD.
To infer what is the net effect of DM decays and annihilations on structure for-
mation, we have to follow the evolution of a large grid of metal free halos. For this
purpose, we used a one-dimensional Lagrangian code,2 which simulates the grav-
itational and hydro-dynamical evolution of the gas, accounting for the evolution

aThe effects of DM decays and annihilation on heating and ionization have been calculated taking
into account the fraction of energy which is effectively absorbed by the gas. l

982
 983

of 12 chemical species, for the cooling/heating effects and for the gravitational in-
fluence of the DM halo. We included into the code the effects of DM decays and
annihilations. 1 Fig. 1. b shows the behaviour of gas density, temperature, ionization
fraction and H2 fractional abundances within a simulated halo of 6 x 10 5 M8 viri-
alizing at redshift 12. Also within the halo DM decays and annihilations enhance
both the temperature and the molecular abundance. However, from the plot of the
density, we can see that for the considered DM models the collapse is delayed by
decays/annihilations, and even prevented in the case of LDM decays.3

e
____ ~ __________ ~ .. _ 102~

~,~-~------ 10 1
10'
g

~ e 10'g
~ 10-' l!'
{l ---------H-- - 102~

~ 10-6 '-.,.~" ~
TI
~
10"
10 l-
~
LL 10'o~~'"_~.~~~___"_---"'=-:;::..:..:::>=~ "'"---~~~.L....:o---'-' 10'"---~~~~---'-'
1000 500 300200 100 50 30 20 10 20 16 14 12 10 9 8 20 16 14 12 10 9 8
Redshift Redshlft Redshifl

(a) (b)
Fig. 1. (a) Effects of decaying/annihilating DM on the IGM evolution. Left axis: fractional abun-
dances of free electrons (e-), H2 and HD as a function of the redshift. Right axis: matter tem-
perature as a function of the redshift. Top panel: Effect of decaying sterile neutrinos of mass 25
keV (dashed line). Central panel: decaying LDM of mass 10 MeV (dashed line). Bottom panel:
annihilating LDM of mass 1 MeV (dashed line). The dotted line is the CMB temperature and
the solid line represents the thermal and chemical evolution without DM decays/annihilations. (b)
Evolution of the central region of a 6 x 10 5 M0 halo virializing at Zvir = 12. From left to right
and from top to bottom: density, temperature, electron abundance and H2 abundance as function
of redshift. The solid line represents the unperturbed case. The dashed, dot-dashed and dotted
lines account for the contribution of 25-keV sterile neutrino decays, I-MeV LDM annihilations and
lO-MeV LDM decays, respectively.

3. Critical mass and gas content in metal free halos

In order to quantify these considerations, Fig. 2.a shows the critical mass merit
(i.e. the minimum halo mass for collapse at a given redshift) as a function of the
virialization redshift, in presence (from top to bottom) of sterile neutrino decays,
LDM decays and LDM annihilations. merit is generally increased by DM decays and
annihilations, confirming that these tend to delay the collapse of metal free halos.
However, the difference with respect to the unperturbed case is less than a factor
2-4 (depending on the redshift and on the model), indicating that the effect of DM
decays and annihilations on structure formation is quite negligible. 3
984

- :·:r
~O.6~
04 -

0.2

100t----~--~-~---___cJ ~:~fc.------~------~
0.8

0.6

1O"f----~--~-~------j

~ ." _ _ _ F ~._

~
0.2
0.0 "----_ _ _ _ _ _ _ _ _ _ _ _--.ej
100 50 30 20 10 10' 10" 10'
Vinalrzation redshlft Halo mass (M,un)

(a) (b)
Fig. 2. Top panel: Decaying sterile neutrinos of 4 (dot-dashed line), 15 (dotted) and 25 keV
(dashed). Central: Decaying LDM of 3 (dotted line) and 10 MeV (dashed). Bottom: Annihilating
LDM of 1 (dot-dashed line), 3 (dotted) and 10 MeV (dashed). The solid lines represent the un-
perturbed case. (a) meTit as a function of the virialization redshift Zvir. (b) Halo baryonic mass
fraction as a function of the halo mass for a fixed virialization redshift (Zvir = 10). Thick (thin)
lines indicate that the halo mass is larger (smaller) than merit.

On the other hand, DM decays and annihilations might have important effects
on the baryonic content of such small metal free halos. Fig. 2.b shows jgas, i.e.
the ratio between the amount of gas which is contained within the virial radius of
the simulated halos and the mass of gas which we should expect from cosmological
parameters, as a function of the halo mass. The solid line shows the unperturbed
case, while the dashed, dotted and dot-dashed lines indicate various models of DM
decays and annihilations. jgas is always smaller if we switch on the contribution
by DM decays and annihilations. In particular, in the case of LDM decays, jgas
drops to ~ 0.4 (if we consider only collapsed halos) or even to ~ 0.02 (if we include
smaller, non-collapsed halos). In conclusion, this means that, in presence of LDM
decays, small metal free halos can still collapse (merit being almost unperturbed);
but their baryonic content is smaller than in larger halos, and, as a consequence,
they can form a smaller total mass of stars than expected.

Acknowledgments
ER acknowledges support from NWO grant 436016. MM acknowledges the orga-
nizers of the Eleventh Marcel Grossmann meeting for the MGF grant. The authors
thank P. L. Biermann for inviting them to the meeting.

References
1. E. Ripamonti, M. Mapelli, and A. Ferrara, MNRAS, accepted, astro-ph/0606482.
2. E. Ripamonti, MNRAS, submitted
3. E. Ripamonti, M. Mapelli and A. Ferrara, MNRAS, submitted, astro-ph/0606483.
THERMAL AND CHEMICAL EVOLUTION OF THE PRIMORDIAL
CLOUDS IN WARM DARK MATTER MODELS WITH KEV
STERILE NEUTRINOS IN ONE-ZONE APPROXIMATION*

JAROSLAW STASIELAKt
Institute of Physics, Jagiellonian University, Krak6w, 30-059, Poland, and
Max-Planck Institute for Radioastronomy, Bonn. D-5312J. Germany, and
Department of Physics and Astronomy, University of Bonn, D-53121, Germany
[email protected]

PETER 1. BIERMANN
Max-Planck Institute for Radioastronomy, Bonn, D-53121, Germany, and
Department of Physics and Astronomy, University of Bonn, D-53121, Germany, and
Department of Physics and Astronomy, University of Alabama, AL 35487, Tuscaloosa, USA
[email protected]

ALEXANDER KUSENKO+
Department of Physics and Astronomy, University of California, CA 90095-1547, Los Angeles,
USA
[email protected]

We follow the evolution of the baryonic top-hat overdensity in a single-zone approx-

imation. Our goal is to juxtapose the evolution of the gas temperature in the primordial
clouds in the lambda cold dark matter model and the warm dark matter model with
keY sterile neutrinos and to check the effects of their decays, into one X-ray photon and
one active neutrino, on the structure formation. We find that, in all the cases we have
examined, the overall effect of sterile dark matter is to facilitate the cooling of gas and
to reduce the minimal mass of the halo able to collapse. Hence, we conclude that X-rays
from the decays of dark matter in the form of sterile neutrinos can help the early collapse
of gas clouds and the subsequent star formation.

Recent work has showed several inconsistencies between the predictions of N-

body simulations of collisionless cold dark matter (CDM) and the observations. 1
Perhaps, a better understanding of CDM on small scales will resolve these dis-
crepancies. It is true, however, that all these problems altogether can be solved by
suppression of the primordial power spectrum of scalar density fluctuations on small
scales. This can be done in warm dark matter (WDM) models via the non-negligible
kinetic energy of the dark matter particles.
One attractive candidate for WDM is sterile neutrino (SN) with mass of several
ke V and a small mixing angle with the ordinary neutrino. Such a particle would be a
natural part of minimal extension of the standard model (v MSM) in which existence

'The work of P. L. B. and J. S. was supported by the Pierre Auger grant O.SCU 5PD1/2 via
the Deutsches Elektronen-Synchrotron (DESY)/ Bundesministerium fuer Bildung und Forschung
(BMBF). The work of A. K. was supported in part by the DOE grant DE-FG 03- 91ER40662 and
by the NASA ATP grants NAG 05-10842 and NAG 05-13399.
t J. S. thanks the organizers of Marcel Grossmann meeting for financial support.
+A. K. thanks the CERN Theory unit and Ecole Poly technique Federale de Lausanne (EPFL) for
hospitality during his visit.

985
986
w' 10- 2

COM COM-
WDM3 WDM3
WDM2 WDM2
WDMl WDMl
W' 10-3

i
W· W"

5 10.5
10- '00 100

~1~1 =-
COM
w' WDM3···
WDM2
WDMl
W·

~ i
w' ':Ji.~ W·

W·
10'100
"
~"

Fig. 1. Top left, top right and bottom left: Evolution of ionization fraction, H2 fraction and
temperature with redshift for different models. In each case, the mass of the primordial cloud is
equal to M = 10 6 M8 and virialization redshift to Zvir = 12. Bottom right: Dependence of the
minimal mass of primordial halo able to collapse on its virialization redshift. Models we have used
in calculation are following: ms = 25 keY and sin 2 e = 3 x 10- 12 (WDM1), m" = 15 keY and
sin 2 e = 3 x 10- 12 (WDM2), ms = 3.3 keY and sin 2 e = 3 x 10- 9 (WDM3), and CDM.

of three SN s would be able to explain the masses of active neutrinos,2 the baryon
asymmetry of the universe 3 and the abundance of dark matter.4 Furthermore, SN
with mass in the ke V range can simultaneously solve another seemingly unrelated
astrophysical puzzle, namely, the origin of the rapid motions of pulsars. 5 It can also
bring the supernova calculation in better agreement with the observation 6 and help
the formation of super-massive black holes in the early universe. 7
Although dark-matter SNs are stable on cosmological time scales, they never-
theless decay. The most prominent "visible" mode is decay into one active neutrino
and one photon, Vs ---+ Val' It produces an X-ray background radiation that has
a two-fold effect on the collapsing clouds of hydrogen in the early universe. First,
the X-rays ionize gas and cause an increase in the fraction of molecular hydrogen, 9
which makes it easier for the gas to cool and to form stars. Second, the same X-rays
deposit a certain amount of heat, which could thwart the cooling of gas. Moreover,
it is well known that suppression on small scales in WDM models delays structure
formation and can lead, in principle, to inconsistency with WMAP measurements. s
It is, therefore, important to check cosmological implication of keY SNs via exami-
nation of their decays on star formation and re-ionization of the universe.
We follow the evolution of the baryonic top-hat overdensity in a single-zone
approximation,lO,11 assuming that the density of the collapsing cloud is constant
after virialization. In order to take into account the effects of SNs decays inside the
collapsing halo, we have solved radiative transfer equation for the spherically sym-
metric clouds with the uniform density. 12 In addition, we have included absorption
 987

of the X-rays from the SNs decays by both Hand He. These are improvements
with respect to 11, where one can find the details of our code.
Figure 1 shows our results. We can clearly see that the overall effect of SN decay
is to enhance ionization fraction, H2 fraction and to speed up the cooling of the
gas in the primordial halos. The minimal mass of the clouds able to collapse is
reduced in all WDM models we have used. It is worth to note that inclusion of SN
free-streaming length can change substantially these results and lead to the delay
of structure formation with respect to the CDM model.1 3 However, if one compares
two WDM models, namely the one with SNs which cannot decay and the second
one with decaying SNs, then the gas cooling should be enhanced in the latter case.
If this is true then our conclusion will not change. Our work shows the importance
of the dark matter decays in the structure formation in the early universe.
We thank G. Gilmore, M. Mapelli, E. Ripamonti and M. Shaposhnikov for very
helpful discussions and comments.

References
1. J. Sommer-Larsen and A. D. Dolgov, ApJ 551,608 (2001); B. Willman, F. Governato,
J. Wadsley and T. Quinn, MNRAS353, 639 (2004); F. Governato et al., ApJ607, 688
(2004); M. Fellhauer et al., ApJ 651, 167 (2006); V. Belokurov et al., ApJ 642, L137
(2006); B. Allgood, R. A. Flores, J. R. Primack, A. V. Kravtsov, R. H. Wechsler, A.
Faltenbacher and J. S. Bullock, MNRAS 367, 1781 (2006); G. Gilmore, M. Wilkinson,
J. Kleyna, A. Koch, Wyn Evans, R. F. G. Wyse and E. K. Grebel, astro-ph/0608528.
2. T. Asaka, S. Blanchet and M. Shaposhnikov, Phys. Lett. B 631, 151 (2005).
3. E. K. Akhmedov, V. A. Rubakov and A. Y. Smirnov, Phys. Rev. Lett. 81, 1359 (1998);
T. Asaka and M. Shaposhnikov, Phys. Lett. B 620, 17 (2005).
4. S. Dodelson and L. M. Widrow, Phys. Rev. Lett. 72, 17 (1994); K. Abazajian, G. M.
Fuller and M. Patel, Phys. Rev. D 64,023501 (2001); K. Abazajian, G. M. Fuller and
W. H. Tucker, ApJ 562,593 (2001); K. Abazajian, Phys. Rev. D 73,063506 (2006);
M. Shaposhnikov and I. Tkachev, Phys. Lett. B 639, 414 (2006).
5. A. Kusenko and G. Segre, Phys. Lett. B396, 197 (1997); Phys. Rev. D 59, 061302
(1999); G. M. Fuller, A. Kusenko, I. Mocioiu and S. Pascoli, Phys. Rev. D 68, 103002
(2003); M. Barkovich, J. C. D'Olivo and R. Montemayor, Phys. Rev. D 70, 043005
(2004); in Pulsars: New Research (Hauppauge: Nova Science, 2007), hep-ph/0503113;
A. Kusenko, IJMP D 13, 2065 (2004); hep-ph/0609081.
6. C. L. Fryer and A. Kusenko, ApJS 163, 335 (2006).
7. F. Munyaneza and P. L. Biermann, AA 458, L9 (2006); AA 436, 805 (2005); X. Fan
et al., AJ 122, 2833 (2001).
8. N. Yoshida, A. Sokasian, L. Hernquist and V. Springel, ApJL 591, 1 (2003).
9. P. L. Biermann and A. Kusenko, Phys. Rev. Lett. 96, 091301 (2006).
10. M. Tegmark, J. Silk, M. J. Rees, A. Blanchard, T. Abel and F. Palla, ApJ 474, 1
(1997).
11. J. Stasielak, P. L. Biermann and A. Kusenko, ApJ 654, 290 (2007).
12. D. Mihalas and B. W. Mihalas, Foundations of Radiation Hydrodynamics (New York:
Oxford Univ. Press, 1984); J. Stasielak, S. Stachniewicz and M. Kutschera, Acta Phys.
PoZ. B 36, 3897 (2005); the proceedings of the Island Universes conference held in
Terschelling, Netherlands, July 2005, ed. R. de Jong to appear, astro-ph/0612071.
13. E. Ripamonti, M. Mapelli and A. Ferrara, MNRASto appear (2007), astro-ph/0606483
 RESTRICTIONS ON STERILE NEUTRINO PARAMETERS FROM
ASTROPHYSICAL OBSERVATIONS

OLEG RUCHAYSKIY
Institut des Hautes Etudes Scientijiques, France

Adding 3 right-handed (sterile) neutrino to the Standard Model (SM) can solve sev-
eral "beyond the Standard Model" problems within one consistent framework: explain
neutrino oscillations and baryon asymmetry of the Universe and provide a dark matter
(DM) candidate. In this talk I will present current status of astrophysical searches for
the DM sterile neutrino.

1. Sterile neutrino as the DM candidate

It was noticed long ago that the sterile neutrino with the mass in the keY range
provides a valuable DM candidate. 1 Recently it was shown that the extensions of the
8M by 3 right-handed (sterile) neutrinos explains neutrino oscillations, allows for
baryogenesis and provides the DM candidate within one consistent framework. 2a the
usual see-saw models, the masses of all new particles in vM8M are below electroweak
scale, which makes this theory potentially testable. The baryogenesis requires two
sterile neutrinos to have masses 0(1 - 20) GeV and be quite degenerate. The third
neutrino should be much lighter and plays the role of the DM.
Any DM candidate should (1) be stable or cosmologically long-lived; (2) be
"dark" (interact very weakly with the 8M matter); and (3) be produce in the correct
amount in the early Universe. The sterile neutrino satisfies all these requirements.
The sterile neutrino interacts with the rest of the 8M only through mixing with
active neutrinos. The mixing is parameterized by e-
the ratio of Yukawa interac-
tion between left and right-handed neutrinos to the mass of the sterile neutrino. In
the vM8M this mixing can be made arbitrarily small. Therefore, the light sterile
neutrino is definitely "dark". However, it is not completely dark. Due to this inter-
action, the sterile neutrino can decay into three active neutrinos. The life-time of
such a decay exceeds the age of the Universe (T 26
= 5 x 10 sec x [ ~~] [1~~8]). The
5
sterile neutrino also has a (subdominant) decay channel into a photon and an active
neutrino. The energy of the photon is E, = ~s and the width of the decay line is
determined by the Doppler broadening: D.E / E, rv 10- 4 - 10- 2 . This means that
one can search for the narrow line of neutrino decay in the spectra of astrophysical
objects.
The mass of the sterile neutrino DM should be above 300 - 500 eV (Tremain-
Gunn bound 3 ), i.e. the lowest energy range to search for the sterile neutrino decay is
the X-ray.b The corresponding photon flux from the region with the DM overdcnsity

aThis extension has been called vMSM. For its review see the talk by M. Shaposhnikov.
bSterile neutrino with the mass in keY range has many interesting astrophysical applications. See
P. Biermann's contribution to these proceedings.4

988
 989

1e-04

le-06

1e-08
$'
N;
'iii
le-l0

le-12

1e-14
10 100
Ms [keY]

Fig. 1. Restrictions from X-ray observations. Combined exclusion plot of works. 6 •8- 10 ,14

is related to the parameters of the sterile neutrino as

FDM :::::;
M;': ]
6.38 [10 10 M0
[M P c]2. 2
DL
[M8]5 keV
sm (20) keV --;:--.s-'e-c

where M;~~ is the mass of DM within a telescope's FoV and DL is the luminous
distauce to the object, sterile neutrino has mass l'v1s and mixing angle 0 measure
of the interaction of the sterile neutrino with its active counterparts.
During the last year a number of works strengthened the bounds on parameters
of sterile neutrino by several orders of magnitude. 5- 12 Current exclusion region is
shown on FIG. 1.

2. Lyman-a forest constraints

Restrictions on the mass of the DM particles also come from the studies of the
details of structure formation in the Universe, containing in the Lyman-a: forest
data. Namely, by looking at the Lyman-a: absorption lines (absorption by the neutral
hydrogen at A = 1216A) in the quasar spectra at different red-shifts, and comparing
it with the results of numerical modeling of structure formation, one obtains a lower
bound on the DM particle mass MLyo;. The mass of the sterile neutrino is related
to this lower mass bound as M8 i~:~MLYO;' Here (Ps), (Pa) are average momenta
of sterile (active) neutrinos. The ratio «Ps»
Po.
depends on the production mechanism
of the DM sterile neutrino and on the physics beyond the vMSM. Results of 13 ,15
show that this ratio can be anywhere between rv 0.15 and 1. Therefore, the results
of Lya: constraint MLyo; > 14.5 keV from 16 imply that the DM mass can be as low
as M8 > 2.5 ke V (results of Lya: analysis of17 imply even lower bound M8 ~ 1 keV).
The scenarios with large lepton asymmetries 18 also provide (Ps) :::::; 0.2(Pa) and thus
comparable limits on the M8 ~ 2 keV.
990

While Lya method is potentially very powerful, it is also very indirect and
hinges on the ability to know the exact relation between Lya optical depth and local
gas density. This relation depends on local temperature, local (peculiar) velocities,
hydrogen overdensity and its neutral fraction. The knowledge of these quantities
requires a number of astrophysical assumptions.

3. Observational strategy
As shown in 6 the signal from almost all nearby objects (dwarf galaxies, Milky Way,
large elliptic galaxies, galaxy clusters) provide comparable (within an order of mag-
nitude) DM decay signal. Therefore, observation of any astrophysical object where
the underlying spectrum can be described by a convincing physical model is well
suited for the DM search. The best candidates are the dSph galaxies of the Milky
Way as they are expected to provide the strongest restrictions. Indeed, (i) they
have smaller velocity dispersion and thus Doppler broadening as compared to large
galaxies or galaxy clusters and (ii) they are very dark in X-ray, thus optimizing a
signal-to-noise ratio.

References
1. S. Dodelson and L. M. Widrow, Phys. Rev. Lett. 72, 17 (1994).
2. T. Asaka, S. Blanchet and M. Shaposhnikov, Phys. Lett. B631, 151 (2005).
3. S. Tremaine and J. E. Gunn, Phys. Rev. Lett. 42, 407 (1979).
4. P. L. Biermann and F. Munyaneza astro-ph/0702173 (2007).
5. A. Boyarsky, A. Neronov, O. Ruchayskiy and M. Shaposhnikov, MNRAS 370, 213
(July 2006).
6. A. Boyarsky, A. Neronov, O. Ruchayskiy, M. Shaposhnikov and I. Tkachev,
Phys. Rev. Lett. 97, p. 261302 (Dec 2007).
7. S. Riemer-Sorensen, S. H. Hansen and K. Pedersen, ApJ 644, L33 (June 2006).
8. C. R. Watson, J. F. Beacom, H. Yuksel and T. P. Walker, Phys. Rev. D74, p. 033009
(2006) .
9. A. Boyarsky, J. Nevalainen and O. Ruchayskiy, astro-ph/0610961, Submitted to A€9A
(2006).
10. A. Boyarsky, O. Ruchayskiy and M. Markevitch, astro-ph/0611168, Submitted to ApJ.
(2006).
11. A. Boyarsky, J. W. den Herder, A. Neronov and O. Ruchayskiy, astro-ph/0612219,
Submitted to Astropart. Phys. (2006).
12. K. N. Abazajian, M. Markevitch, S. M. Koushiappas and R. C. Hickox, astro-
ph/0611144, Submitted to Phys. Rev. D (November 2006).
13. M. Shaposhnikov and I. Tkachev, Phys. Lett. B639, 414 (2006).
14. A. Boyarsky, A. Neronov, O. Ruchayskiy and M. Shaposhnikov, Phys. Rev. D 74, p.
103506(nov 2006).
15. T. Asaka, A. Kusenko and M. Shaposhnikov, Phys. Lett. B638, 401 (2006).
16. U. Seljak, A. Makarov, P. McDonald and H. Trac, Phys. Rev. Lett. 97, p. 191303
(2006).
17. M. Viel, J. Lesgourgues, M. G. Haehnelt, S. Matarrese and A. Riotto, Phys. Rev. Lett.
97, p. 071301 (2006).
18. X.-d. Shi and G. M. Fuller, Phys. Rev. Lett. 82, 2832 (1999).
 UPPER LIMITS ON DENSITY OF DARK MATTER
IN SOLAR SYSTEM*

IOSIF KHRIPLOVICH
Budker Institute of Nuclear Physics,
Novosibirsk 630090, Russia
[email protected]

ELENA PITJEVA
Institute of Applied Astronomy,
St. Petersburg 191187, Russia
[email protected]

The analysis of observational data on secular perihelion precession of Earth and Mars,
based on the EPM2004 ephemerides, results in new upper limit on density of dark matter
in the Solar system: Pdm < 3 X 10- 19 g/cm 3 •

According to present observational data (see, for instance,l), the dark matter
density in the region of the Milky Way Galaxy, where our Solar system is situated,
is
P dm '" 0.5 X 10- 24 g/cm 3 . (1)
However, not so much is known about the local dark matter distribution inside our
Solar system itself. A direct information on this distribution, even rather modest
one as compared to the above number, is of certain interest.
Previous estimates of the dark matter density in the Solar system 2- 4 resulted in
the upper limits on the level of
P dm < 10- 16 g/cm3 . (2)
2
The bounds obtained in ,3 were based on the investigation of the possible effect of
dark matter on the orbit of Uranus. The limit derived in 4 resulted from the analysis
of the perihelion precession of the asteroid Icarus.
Here we present much stronger upper limits on the dark matter content in the
Solar system following from analysis of the perihelion precession of Mercury, Earth,
and Mars. Our limits are based on the precision EPM ephemerides constructed
in 5 by simultaneous numerical integration of the equations of motion for the nine
major planets, the Sun, and the Moon in the post-Newtonian approximation. Such
subtle effects as the influence of 301 large asteroids and of the ring of small aster-
oids, as well as the solar oblateness, were included into the calculation. The EPM
ephemerides resulted from a least squares adjustment to more than 317000 posi-
tion observations (1913-2003) of different types, including radiometric and optical
astrometric observations of spacecraft, planets, and their satellites.

*This research has been partially supported by the Russian Foundation for Basic Research grant
05/02/16627.

991
992

The conclusions of the present paper are based on the possible corrections to
the secular perihelion precession b¢ (i.e. on the deviations of the results of theo-
retical calculations from the observational data) of three planets 6 obtained from
about 250000 high-precision American and Russian ranging to planets and space-
craft (1961-2003), including in particular Viking-l,2, Pathfinder, Mars Global Syr-
veyor, and Odyssey.a They are presented in Table l.

Table 1. Corrections to secular perihelion precession of planets.

Mercury Earth Mars

/I per century - 0.0036 ± 0.0050 - 0.0002 ± 0.0004 0.0001 ± 0.0005

(8¢/27f) X 1011 -0.67±0.93 -0.15±0.31 0.14±0.73

Let us address now the possible influence of dark matter on the perihelion preces-
sion. To simplify our estimates, we will consider the dark matter as nonrelativistic
dust with spherically symmetric density p( r).
Then the correction bF(r) to the gravitational force acting upon a planet with
mass m situated at a distance r from the Sun is found easily by means of the Gauss
theorem:
fJ( r)
F(r) = - k m2- ' (3)
r
where

fJ( r) = 47f foT p( rdT~ drl

is the total mass of dark matter inside a sphere of radius r, and k is the Newton
gravitational constant. The corresponding correction to the gravitational potential
is

bU(r) = - r dr2F(r2) = km inr dT/L(~2)

io r
.
2
(4)

This correction shifts the perihelion of the planet orbit by angle

b¢ = ..!!...- 2m
dL L
r d¢r bU(r)
io
2
(5)

per period 7 ; here L is the planet angular momentum. With bU(r) given by formula
(4), and under the assumption (made also in2-4) that Pdm remains constant at the

aThe corresponding correction from 6 to the perihelion precession of Venus is not used here, since
it is much larger and less accurate. The reasons are as follows. On the one hand, Venus moves
around the Sun more slowly than Mercury. On the other hand, for Venus there are no high-
precision ranging, like those to martian landers or orbiting spacecraft, determining the accuracy
of the corrections for Mars and Earth.
 993

distances discussed, we arrive after some transformations at the result:

(6)

where M(') is the mass of the Sun and e is the eccentricity of the planet orbit. For
Mercury, Earth, and Mars e is small, about 0.21, 0.02, and 0.09, respectively, and
will be neglected in the estimates below.
Now the results of 6 presented in Table 1 can be interpreted as upper limits on
the mass of dark matter inside the orbits of corresponding planets. Using the data
in the last line of Table 1, we formulate these upper limits as follows:
,u(0.39 au) < 6 x 10- 12 M(') , (7)

,u(1 au) < 2 x 10- 12 M(') , (8)

/1(1.52 au) < 6 x 10- 12 M(') ; (9)

here we indicate in brackets the distance (in astronomical units, au) from the Sun
of the corresponding planet, Mercury, Earth, and Mars, respectively.
At last, we convert these upper limits into the limits on the dark matter density.
Then, the Mercury limit (7) results in the bound

Pdm < 10 -17 g / cm3 , (10)

somewhat better than (2).
The bounds following from the Earth and Mars limits, (8) and (9), are much
stronger. They practically coincide and constitute

Pdm <3 X 10- 19 g/cm 3 . (11)

The result (11), though being far from the global (Galaxy) estimate (1), improves
essentially the previous local (Solar system) limits (2).

References
1. G. Bertone, D. Merritt, Mod. Phys. Lett., A20, 1021 (2005); astro-ph/0504422.
2. J.D. Anderson, E.L. Lau, A.H. Taylor, D.A. Dicus, D.C. Teplitz, and V.L. Teplitz,
ApJ, 342, 539 (1989).
3. J.D. Anderson, E.L. Lau, T.P. Krisher, D.A. Dicus, D.C. Rosenbaum, and V.L. Teplitz,
ApJ, 448, 885 (1995); hep-ph/9503368.
4. 0. Gr0n and H.H. Soleng, ApJ, 456, 445 (1996); astro-ph/9507051.
5. E.V. Pitjeva, Solar System Research, 39, 176 (2005).
6. E.V. Pitjeva, Astronomy Letters, 31, 340 (2005).
7. L.D. Landau and E.M. Lifshitz, Mechanics, (Pergamon Press, 1958).
 THE OBSERVED PROPERTIES OF DARK MATTER ON SMALL
ASTROPHYSICAL SCALES

G GILMORE
Instite of Astronomy, Madingley Rd., Cambridge CB3 OHA, UK. [email protected]

There has long been evidence that low-mass galaxies are systematically larger, of lower
central mass density, and of lower central phase-space density, than are star clusters with
a similar number of stars. There is also suggestive evidence that there is a minimum mass
for the dark halos associated with small galaxies. Evidence to support all these relations is
becoming stronger. We reconcile the minimum size, minimum mass, low mass density and
low phase-space density relationships, showing they are aspects of a more fundamental
pair of intrinsic properties of dark matter itself. Dark matter clusters such that there
is a maximum volume mass density, which has the very low value of about 20GeV /c 2
cm- 3 (about O.5M0 pc- 3 ), and there is an associated shallow density profile (core) in
the mass distribution, with a characteristic length scale of order 5xl0 18 m (~ 150pc).
Even a cusp density profile leads to an interestingly low maximum central mass density
of ~ ITeV/c 2 cm- 3 .

Galaxies, Dark Matter, and Star Clusters

The smallest scales on which dark matter can form isolated gravitationally bound
systems depends on the nature of the dark matter particles. Plausible expected
values for the smallest halos are in the (rather wide) allowed range from mm to
parsecs. Astrophysically, the smallest (dwarf, dSph) galaxies are also of interest,
since it at the scales of these galaxies that there is the largest disagreement between
simulation predictions and observations. Among other issues, there is dispute about
the numbers of dSph galaxies, with very many more predicted than observed; there is
disagreement about the relevance of even those few observed to predictions, as their
stellar populations do not correspond to those seen in the halo of the Milky Way, as
simulations predict they should; and there is disagreement about the density profile
of dark matter on small scales, with simulations predicting steep density profiles,
which are not required by observations.
There is further information on small scales, namely that dark matter seems
not to concentrate in galactic disks or in galactic central regions. This became
well-established, and remains best-established, in the Solar neighbourhood. Detailed
analyses of the vertical gravitational potential gradient associated with the Galactic
disk near the Sun are feasible, and can be complemented by robust direct determina-
tions 10 of the baryonic contribution (stars, gas, dust ... ) to the mass generating that
potential. Robust analyses of the local potential (the 'Oort Limit') became available
in 1989. 3 ,11,12 These studies 9 showed that there is no dark matter distributed with
a small vertical scale height. Rather, the dark matter which dominates the Galactic
potential is distributed in a halo. Similar results 6 are found in the central regions of
low surface brightness galaxies, where cored distributions of dark mass are found.
That is, dark matter seems not to be found concentrated on (astrophysical) length
scales smaller than 100pc.

994
 995

At the same time kinematic studies of external galaxies, and of star clusters,
established a distinction between the lowest-mass galaxies and star clusters of sim-
ilar luminosity. The lowest mass galaxies overlap in stellar mass with the most
massive globular star clusters, and yet have fundamentally different kinematic and
structural properties. Globular star clusters are systematically smaller in half-light
radius than are galaxies of comparable luminosity (stellar mass), and star clusters
have kinematics which are well-described by the virial theorem for a purely stellar
system, while galaxies inhabit extended (dark matter) gravitational halos. Early no-
table studies 2 ,4 [ref 2, figure 2; ref 4 figs 13, 14 & 15]. These studies established the
differences in phase-space density between globular star clusters and dSph galax-
ies, and supported a gap in size distribution between the largest star clusters and
the smallest galaxies. These differences have been interpreted since their discovery
as being due to the presence of gravitationally-dominant extended dark matter in
dSph galaxies, and its absence from star clusters.
Some evidence for systematics in the distribution of dark matter on the scale of
dSphs was identified. The most significant was the suggestion 13 that there was an
apparent minimum dark halo mass of ~ 2 x 10 7 MG) seen in results from dynamical
studies. This was extended in an important study, 5 which also provided one of the
earliest robust studies that the dSph galaxy kinematics are unaffected by external
tides, so that the dynamical results are plausible. This minimum mass correlation
remained, with greater significance, in studies with larger kinematic datasets. 7 ,14
Much more massive star clusters have recently been studied, and all these are
compact. Very low luminosity galaxies have recently been discovered, and these
are all extended. No intermediate object is known,l suggesting that there is a real
physical minimum size associated with any stellar system which has dynamically
significant dark matter (~ 100pc), and that there is a real minimum physical scale
on which dark matter forms bound structures (~ 100 - 300pc). This deduction has
significant implications for the nature of the dominant form of dark matter.
A natural explanationS is that the dark matter has an intrinsic maximum density
and a minimum scale length. The justification for this reconciliation of the size and
phase-space density results above comes from available studies of the mass distribu-
tions in several dSph galaxies. Mass determinations involve the analysis of kinematic
and photometric length scale data, to provide solutions to the Collisionless Boltz-
mann Equation. There is a caveat: since stellar systems are truly collisionless, there
is no equation of state to relate density and pressure; anisotropic stress is a natural
feature of galaxies - and indeed provides shape for non-rotating galaxies. The ab
initio unknown stress function is degenerate with the desired mass profile, enriching
the literature in this subject. s However, recent results, based on the simplest as-
sumptions in each case, have produced the mass density distributions summarised
in Figure 1.
The results of figure 1 are remarkable: in each case, the mass density profile is
similar, as is the central normalisation of mass density. This similarity, if indica-
tive of a universal property of dark matter, naturally explains all the observational
996

'"
I

.:£
~ -------------
::2:0

Draco
--Carina
- - Ursa Minor
--Sextans
-1/r
106L=~~~==~--~------~~~--~
10-1 10°
r (kpc)

Fig. 1. Derived inner mass distributions from Jeans' equation analyses for four dSph galaxies.
Also shown is a predicted r- 1 density profile. The modelling is reliable in each case out to radii
of log (r)kpc~ 0.5. This figure is from. 14

correlations noted above. Figure 1 is also interesting in that the central mass den-
sity is so low, of order 20Ge V / c 2 em -3. Even if one assumes an inner cuspoed
mass profile, the kinematic data require a maximum central mass density of order
20Ge V / c 2 cm -3. In either case, this implies extremely low particle number densities
if the individual particles are indeed at the Higgs scale or above.

References
1. Belokurov, V. et al 2007 astroph-0608448
2. Bender,R., Burstein, D., Faber, S., 1992 ApJ 399462
3. Bienayme, 0., et al 2006 A&A 446 933
4. Burstein,D., Bender, R., Faber, S., & Nolthenius 1997 AJ 114
5. Cote, P., Mateo, M., Olszewski, E., & Cook, K, 1999 ApJ 526 147
6. Gentile, et al 2005 ApJL 634 L145
7. Gilmore, G., et al astroph-0608528
8. Gilmore, G., et al ApJ submitted
9. Gilmore, G., Wyse, R.F.G., & Kuijken, K, 1989 ARAA 27 555
10. Kroupa, P., Tout, C., & Gilmore, G., 1993 MNRAS 262 545
11. Kuijken, K, & Gilmore, G. 1989 MNRAS 239 605
12. Kuijken, K, & Gilmore, G. 1991 ApJL 367 L9
13. Mateo, M., Olszewski, E.W., Pryor, C., Welch, D.L., Fischer, P. 1993 AJ 105 510
14. Wilkinson, M. 1., Kleyna, J. T., Wyn Evans, N., Gilmore, G. F., Read, J. 1., Koch,
A., Grebel, E. K, & Irwin, M. J. 2006, EAS Publications Series, 20, 105
 IS DARK MATTER FUTILE ON THE BRANE? *

LAsZL6 A. GERGELY
Departments of Theoretical and Experimental Physics, University of Szeged,
D6m fer 9, H-6720 Szeged, Hungary
[email protected]

We investigate whether dark matter can be replaced by various source terms appearing
in the effective Einstein equation, valid on a brane embedded into a higher dimensional
space-time (the bulk). Such non-conventional source terms include a quadratic (ordinary)
matter source term, a geometric source term originating in the Weyl curvature of the
bulk, a source term arising from the possible asymmetric embedding, and finally the
pull-back to the brane of possible non-standard model bulk fields.

Keywords: brane-worIds, asymmetric embedding, brane-bulk energy exchange

According to modern preClSlon cosmologyl-3 our universe is described by a

Friedmann-Lemaitre-Robertson-Walker geometry with flat spatial sections. The
matter source for this geometry is however less well known. Ordinary, baryonic
matter gives a tiny R:: 3% fraction Pb of the critical energy density. In order to ex-
plain galactic cluster dynamics and galactic rotation curves, dark matter has been
introduced, and it is believed that nowadays there is approximately 10 times as
much dark matter as baryonic matter in the universe. Models of cold and warm
(but not hot) dark matter are compatible with observations. The remaining R:: 70%
of the critical energy density is given by a mysterious dark energy, a form of matter
represented in the simplest way by the famous cosmological constant A (a perfect
fluid with pressure PA = WPA , W = -1). Dark energy obeys Pde + 3Pde < 0, the
condition which (due to the Raychaudhuri equation) yields to an accelerated expan-
sion of the universe. The latter seems unavoidable in the light of the recent distant
supernovae surveys.4,5 Observations support very well the simplest possible dark
energy, the cosmological constant A and the basic question is not how different W is
from -1, but instead whether it is allowed to have any dynamics. But even if dark
energy would be simply a cosmological constant, what is its origin? The explana-
tion in terms of vacuum energy density turned into one of the biggest problems of
modern theoretical physics: the discrepancy is of 120 orders of magnitude.
With so much unknown matter and energy in our universe the question naturally
arises, whether gravitational dynamics, as we now it, has to be modified? Dark
matter is conveniently replaced by the so-called MOND gravitational theory.6,7
This however conflicts with cosmological observations. s The late-time acceleration
of the universe can be explained by the inverse curvature gravity theories. 9

*Research supported by OTKA grants no. T046939, TS044665 and the Janos Bolyai Fellowships
of the Hungarian Academy of Sciences. The author wishes to thank the organizers of the 11th
Marcel Grossmann Meeting for support.

997
998

The theory of gravitation can be also modified by returning to the pioneering

ideas of Kaluza and Klein, allowing for more than 4 space-time dimensions. Indeed
string / M-theory is multidimensional and in recent years a simpler class of effective
models,10-12 known as brane-worlds have been intensively studied. In these models
the observable universe is a 4-dimensional space-time continuum (the brane with
tension A) on which standard model fields act. Gravitation however is allowed to
leak out in a fifth, possibly non-compact dimension. Its 5-dimensional dynamics is
governed by the Einstein equation with cosmological constant term.
The apparent gravitational dynamics on our observable 4-dimensional universe
appears through a projection formalism. It consist of the twice contracted Gauss
equation, the Codazzi equation and an effective Einstein equation. 13 In the deriva-
tion of the latter the junction conditions across the brane are employed. The effec-
tive Einstein equation, first derived 14 for symmetric embedding and no bulk sources
was later supplemented by the pull-back to the brane of the bulk energy momentum
tensor IIab :

(1)

(with ~2 the bulk coupling constant and gab the induced metric on the brane) and
the asymmetry source term L~: which is the trace free part of the tensor
- - - -
Lab = Kab K - KacKb
-c gab
-""2 (-2 - -ab)
K - Kab K , (2)

(with K ab the mean extrinsic curvature). The effective Einstein equation is: 13
2 ~4 - -TF-
Gab=-Agab+rcTab+rc Sab-[ab+Lab +Pab· (3)
(with rc 2 the brane coupling constant). The function A = (~2 /2) (A-n c n d IT cd -L/4)
possibly varies due to both the normal projection of the bulk energy-momentum
tensor and the embedding. When constant, A represents the the brane cosmological
constant. The source term Sab is quadratic in the brane energy-momentum tensor
Tab, and it modifies early cosmology.15 The quantity
~ b d
[ae = Cabedn n

represents the electric part of the bulk Weyl tensor Cabed. In a cosmological context
[ab is known as dark radiation.
Modified gravitational dynamics leads therefore to four new source terms in the
effective Einstein equation. In what follows we discuss the potential of generating /
replacing dark matter, represented by these source terms.
From this point of view the less interesting of them is the non-linear source term
Sab. This is because it scales as p/ A, as compared to the energy-momentum tensor
and this ratio is infinitesimal, excepting the very early universe, due to the huge
value A > 138.59 Te V 4 of the brane tension. 16
The Weyl curvature of the bulk on the other hand can generate quite remarkable
electric part contributions [abo In a spherically symmetric brane-world metric the
 999

Schwarzschild mass parameter receives a new contribution due to Eab which is inter-
preted as the mass of dark matter. 17 When the cosmological constant is neglected,
the dark mass scales linearly with the radial distance, explaining the flatness of the
galactic rotation curves. The bending angle of light in such brane world models was
found much larger as compared to the predictions of the dark matter models, the
deviation increasing with the degree of compactness. This is however exactly the
regime where no observational data on gravitational lensing is available yet.
Properly chosen non-standard model bulk fields can replace dark matter in ex-
plaining structure formation,18 as the evolution of perturbations on the brane be-
comes similar to that of the Cold Dark Matter (CDM) model.
Most interestingly, the combined effect of asymmetry and a bulk radiation qual-
itatively can give both dark matter and dark energy.16,19 This has been investigated
in a closed brane-world universe model, with bulk black hole only on one side of
the brane. In this model the emitted Hawking radiation is partially absorbed on
the brane and partially transmitted through, but other models with radiation in
the bulk have been also considered. 20 In principle in such models there are two
competing effects: (a) the radiation pressure accelerates the brane, manifesting it-
self as dark energy, and (b) the absorbed radiation increases the energy density
of the bulk, appearing as CDM. These two effects compete with each other, and
with properly chosen initial data a critical-like behavior was observed. 16 ,19 Based
on the predictions of these toy models, it seems worth to work out more realistic
brane-world models in order to confront them with observations.

References
1. A. Dorashkevich, D. L. Tucker., S. Allam, and M. J. Way, Astron. Astraphys. 418, 7
(2004).
2. M. Tegmark, M.A. Strauss, M.R. Blanton, et al., Phys. Rev. D 69, 103501 (2004).
3. D. N. Spergel, R. Bean, O. Dore, et al., astro-ph/0603449 (2006).
4. A. G. Ries, L-G. Strolger, J. Tonry, et aI, Asrophys. J. 607, 665 (2005).
5. P. Astier, J. Guy, J. N. Regnault, et al., Astron. Astrophys. 447, 31 (2006).
6. M. Milgram, Astrophys. J. 270, 365 (1983), ibid. 371 (1983), ibid. 384 (1983).
7. P. D. Mannheim, Prog. Part. Nuci. Phys. 56,340 (2006).
8. A. Slosar, A. Melchiorri, and J. Silk, Phys. Rev. D72, 101301 (2005).
9. O. Mena, J. Santiago, and J. Weller, Phys. Rev. Lett. 96, 041103 (2006).
10. L. Randall and R. Sundrum, Phys. Rev. Lett. 83, 3370 (1999).
11. L. Randall and R. Sundrum, Phys. Rev. Lett. 83, 4690 (1999).
12. R. Maartens R, Living Rev. ReI. 7, 1 (2004).
13. L. A. Gergely, Phys. Rev. D 68, 124011 (2003).
14. T. Shiromizu T, K. Maeda, and M. Sasaki, Phys. Rev. D 62, 024012 (2000).
15. P. Binetruy, C. Deffayet, U. Ellwanger, and D. Langlois, Phys.Lett. B 477, 285 (2000).
16. L. A. Gergely and Z. Keresztes, JCAP 06(01), 022 (2006).
17. T. Harko and L. S. Cheng, Astrophys. J. 636, 8 (2006).
18. S. Pal, Phys. Rev. D 74, 024005 (2006).
19. Z. Keresztes, 1. Kepiro, and L. A. Gergely, JCAP 06(05), 020 (2006).
20. D. Jennings, 1. R. Vernon, A.-C. Davis, and C. van de Bruck, JCAP 05(04) 013 (2005).
DIRECT X-RAY CONSTRAINTS ON STERILE NEUTRINO WARM
DARK MATTER

CASEY R. WATSON
Department of Physics and Astronomy, Millikin University, Decatur, IL 62522, USA
[email protected]

We use the diffuse X-ray spectrum (total minus resolved point source emission) of the
Andromeda galaxy to constrain the radiative decay rate and mass, ms, of sterile neu-
trinos, an attractive dark matter candidate. Our findings demand ms < 3.5 keV (95%
C.L.) which is a significant improvement over the previous (95% C.L.) sterile neutrino
mass limits inferred from the X-ray emission of nearby clusters, ms < 8.2 keV (Virgo A)
and ms < 6.3 keV (Virgo A + Coma).

We consider direct constraints on the mass, m s , of sterile neutrinos, v s, based

on their radiative decays to active neutrinos and X-rays of energy E"s = m s /2
via Vs --+ Ve,!-"T + ,; see Watson et al. l for further details. In these proceedings,
we discuss the limits 1 imposed by XMM-Newton observations of the diffuse X-ray
spectrum of the Andromeda galaxy (M31).2
The line flux (at E"s = ms/2) from a sterile neutrino dark matter halo of mass
MDM at a distance D is 3- 5

<P
x,S
c:::: 5.1 x 1O-18 erg cm- 2 s- l D
--
( Mpc
)-2
xC:~;;8) (s;~~;:) (:~
The key parameters we used to estimate the sterile neutrino decay flux from An-
r (1)

dromeda and Virgo A are given in Table 1. 1 Note that for any value of m s ,
the expected sterile neutrino decay fluxes will be comparable, i.e., <p~:l /<p~~7 =
fov
D2 M
D~187 M~o':}·M31 c:::: 0.71 (Eqn. 1), but the sterile neutrino decay signals will be much
1\131 DM,M87
more pronounced above Andromeda's low astrophysical background, as shown in
Fig. 1a.

Table 1. Here we summarize the properties of Andromeda 1,2,6

and Virgo A.7-9 In rows (2) - (6), we show, respectively,
the distance to each object, the angular radius Bfav of the
XMM-Newton field of view (fov) for each observation, our es-
timates of the dark matter masses probed within each fov, the
XMM exposure times (in kiloseconds: ks), and the (95% C.L.)
upper bounds on ms.
Galaxy Name Andromeda (M31) Virgo A (M87)
Distance (M pc) 0.78 ± 0.02 15.8 ± 0.8
Bfav (arcminutes) 5.0' 8.5'
0.13 ± 0.02 75 ± 8
t exp (ks) 34.8 25.9
ms (keV) (95% C.L.) 3.5 8.2

1000
 1001

50

10- 12 10- 11 10- 10 10-9 IO-B 10.7

2
sin 2e

Fig. 1. Left: Here we compare the detectability of l/s decays in Andromeda 1,2 and Virgo A.7,s
The first statistically significant Vs decay peak relative to the measured spectrum of Andromeda
occurs at E,,{,s rns,iim/2 1.75 keV, which excludes rna > 3.5 keV (95% C.L.). The spectrum
of Virgo A excludes rns > 8.2 keV (95% C.L.),8 which would produce a decay signature like the
dashed histogram. Because Andromeda would produce a similar Va decay signal to Virgo A, but
over a much smaller background, the prospective decay signature of 8.2 keV sterile neutrinos in
Andromeda is enormous by comparison. As an intermediate case, we also show the decay peak
associated with a 6.3 keV sterile neutrino, the estimated Virgo A + Coma mass limit.1o,n The
vertical (10-) error bars reflect the Poisson statistics of the signal and background count rates mea-
sured during each observation. Right: Here we present constraints on rns as a function of mixing
angle, sin 2 2B, assuming that all dark matter is comprised of sterile neutrinos. For L = 0, the thick,
solid line corresponds to Os 0.24, 3 while the shaded region to the right corresponds to Os > 0.24.
Three density-production relationships associated with Os 0.3 and L » 10- 10 are also shown. 11
The two previous diTect radiative decay upper limits (both 95% C.L.) are based on measurements of
the Cosmic X-ray Background 12 and Virgo A (lVI87) and the Coma clnster. 8,10,11 The most strin-
gent direct limits, (also 95% C.L.), are based on XMM observations of the Andromeda galaxy. 1,2
The region bounded by the dashed line is excluded by the "I\, tot -scaling method", 1 while the
region above the solid, slightly jagged line is excluded by the more accurate "direct data method"
(see text). The indiTed lower limits: Lya(l), Lya(2),8 and Lya(3)1:l are based on analyses of the
Ly-a forest.

Our analysis of Andromeda yields an upper bound of ms < 3.5 keV (nominally
95% C.L., though, in reality, our limit corresponds to a Vs decay signal that is at
least 4 x the astrophysical flux in the relevant energy binI). Our limit is significantly
more restrictive than the upper bound determined from the spectrum of Virgo A
(ms < 8.2 keV, 95% C.L.).8 The key point is that nearby, low background objects
with modest dark matter masses can provide considerably stronger constraints than
very massive but more distant objects with large astrophysical backgrounds.
When we relax the constraint that Os 0.24, we can find the mixing angle
for which every value of ms satisfies our exclusion criterion by comparing Eqn. 1
to the spectral data in the energy bin E"s m s /2. We call this the "direct data
method". In Fig. lb, we compare our conservative (95% C.L.) exclusion region to
other direct and indirect limits. In conjunction with our di-rect constraint (ms < 3.5
1002

keY; 95% C.L.), the recent Seljak et al. (m s > 14 keY; 95% C.L.) and Viel et al.
(m s > 10 keY; 95% C.L.) Ly-a constraints 13 ,14 strongly suggest that the standard
L "': 0 production scenario of Abazajian et al. 3-5 is ruled out.
We have shown that nearby, low astrophysical background sources can provide
excellent direct constraints on sterile neutrino warm dark matter. The easiest way
to improve upon our limit would be to expose more of Andromeda's dark matter
halo to a larger field-of-view detector or a mosaic of XMM or Chandra pointings.

Acknowledgments
I would especially like to thank Peter Biermann for inviting me to the 11th Marcel
Grossman meeting (MG11) and for leading the interesting and fruitful sterile neu-
trino dark matter session during which this work was presented. I would also like
to thank the MG11 Local Organizing Committee, John Beacom, Terry Walker, and
The Ohio State University Department of Physics for funding the trip.

References
1. C. R. Watson, J. F. Beacom, H. Yiiksel, and T. P. Walker, Phys. Rev. D74, 033009
(2006); astro-ph/0605424
2. R. Shirey et al., Astron. & Astrophys. 365, 195 (2001); astro-ph/0011244.
3. K. Abazajian, Phys. Rev. D73, 063506 (2006); astro-ph/0511630.
4. K. Abazajian, G. M. Fuller, and M. Patel, Phys. Rev. D64, 023501 (2001) astro-
ph/0101524.
5. K. Abazajian, G. M. Fuller, and W. H. Tucker, Astrophys. J. 562, 593 (2001); astro-
ph/0106002.
6. K. Z. Stanek and P. M. Garnavich, Astrophys. J. 503, 131 (1998); astro-ph/9802121.
7. H. Bohringer et al., Astron. & Astrophys. 365, L181 (2001); astro-ph/0011459.
8. K. Abazajian Phys. Rev. D73, 063513 (2006); astro-ph/0512631.
9. H. Jerjen and B. Binggeli (2006); astro-ph/0605706.
10. A. Boyarsky, A. Neronov, O. Ruchayskiy, and M. Shaposhnikov, Phys. Rev. D74,
103506 (2006); astro-ph/0603368.
11. K. Abazajian and S. M. Koushiappas, Phys. Rev. D74, 023527 (2006); astro-
ph/0605271.
12. A. Boyarsky, A. Neronov, O. Ruchayskiy, and M. Shaposhnikov MNRAS 370, 213
(2006); astro-ph/0512509.
13. U. Seljak, A. Makarov, P. McDonald, and H. Trac, Phys. Rev. Lett. 97, 191303 (2006);
astro-ph / 0602430.
14. M. Viel, J. Lesgourgues, M. G. Haehnelt, S. Matarrese, and A. Riotto, Phys. Rev. Lett.
97, 071301 (2006); astro-ph/0605706.
 LIMITS ON THE DARK MATTER PARTICLE MASS FROM
BLACK HOLE GROWTH IN GALAXIES

FAUSTIN MUNYANEZA*
M ax-Planck- Institut fur Radioastronomie,
69 Auf Dem Hugel
D-53121 Bonn, Germany
[email protected]

I review the properties of degenerate fermion balls and investigate the dark matter
distribution at galactic centers using NFW, Moore and isothermal density profiles. I
show that dark matter becomes degenerate for particles masses of a few keY at distances
less than a few parsec from the center of our galaxy. To explain the galactic center black
hole of mass of ~ 3.5 x 106 Mo and a supermassive black hole of ~ 3 X 10 9 Mo at a
redshift of 6.41 in SDSS quasars, the mass of the fermion ball is assumed to be between
3 X 103 Mo and 3.5 x 10 6 Mo. This constrains the mass of the dark matter particle
between 0.6 keY and 82 keY. The lower limit on the dark matter mass is improved to
about 6 keY if exact solutions of Poisson's equation are used in the isothermal power law
case. The constrained dark matter particle could be interpreted as a sterile neutrino.

Keywords: Dark matter, sterile neutrinos, galaxies, black hole physics

1. Introduction
There is mounting evidence that most galaxies harbor supermassive black holes
(BRs) of masses from 10 6.5 to 10 9 . 5 M 8 . The typical case is of the Galactic black
hole of mass (3.1 ± 0.9)10 6M 8 . l ,2 It has also been established that the mass of
the central black hole is tightly correlated with the velocity dispersion (J of its host
bulge, where it is found that MBH rv (J4-5.3 In spite ofthe vast and tantalizing work
on black hole physics, their genesis and evolution are not well understood. Another
outstanding problem in modern astroparticle physics is the particle nature of dark
matter(DM). Recently, there has been a renewed interest in sterile neutrinos 4 as
candidates for dark matter as they could explain the baryon asymmetry of the
Universe, 5 the pulsar kicks,6 the early growth of black holes 7-9 and the reionisation
of the Universe. lO ,11 In this short communication, we review the properties fermion
balls, i.e. self gravitating systems of sterile neutrinos and study the implications of
NFW,12 Moore 13 and isothermal density profiles being degenerate near the galactic
centers.

2. Fermion balls and dark matter mass limits

We will assume that the DM particles obey a Fermi Dirac distribution function with
a non vanishing chemical potential. 7 Assuming only gravitational interaction, it has
been shown that these dark matter particles could condensate at galactic centers
forming degenerate fermi balls 14- 2l with a total mass MF that scales with their size

*Humboldt Fellow.

1003
1004

RF as MF ~ m;8 R F} where ms is the the sterile neutrino mass. Fermions have a

maximum mass Mov ~ m~lm;2 where mpl = (nc/G)1/2 is the Planck's mass.
It has been shown that the assumption of a degenerate fermion ball of mass
lv/F ~ 3 X 10 6 Mo at the center of a dark matter halo of 3 x 10 12 Mo with a density
scaling as 1/r2 in the outer edge of the halo constrains the fermion mass 7.9 to

2 3/4 (M )-1/2 ( 2 ) 1/4

ms:::::; 12 keV/c ( 156~m/s ) 310:~0 gs'
(1)

where (T is the velocity dispersion. Using (T = 100 ± 20 km/s 3 for the velocity
dispersion, and the mass of the black hole of MBH ~ (2.2 - 4)10 6 1\;10,1 then we
obtain a lower limit on the mass of the DM particle

(2)

The above limits lie within the range of sterile neutrino masses obtained from the
study of the origin of the high velocities up to 1000 km/s of pulsars. 6 Moreover,
similar limits were also obtained from X-ray background studies. 22 Recently, it
has been shown that the decay of such a sterile neutrino could help initiate star
formation in the early Universe. 10,11
The black hole grows exponentially with time from Eddington limited baryonic
matter accretion. 7 However, unless the seed black hole has a high mass of 10 3 - 4 M o ,
Eddington limited baryonic matter accretion cannot grow the seed black holes to
3 x 10 9 Mo black holes in SDSS quasars23 at z = 6.41. We have shown that the
Pauli's degeneracy principle helps feed the black hole with dark matter. 7 ,9 Stellar
seed black holes could be grown to 10 3- 4Mo in about 10 7 - 8 years. For this model
to work, the dark matter particle mass has been constrained to be in the order of
10 keV. 7 A further growth to 10 6 . 5 - 9 . 5 1\;/0 is achieved through Eddington baryonic
accretion.
Another constraint on the dark matter particle mass is obtained by studying
the NFW, Moore and isothermal density profiles of p ~ r- S with s being the
power law slope and s = 1 for NFW,12 S = 1.5 for Moore 13 and s = 2 for the
isothermal profiles. The mass enclosed within a radius r scales as Mr ~ r 3 - s and
the corresponding rotational velocity scales as Vrot ~ r1-s/2. We then study the
conditions under which these density profiles satisfy the Pauli degeneracy condition
msVrot ~ nn~/3, where ns is the sterile neutrino number density. It is found that for
all the three density profiles a dark matter becomes degenerate for particle masses
between 0.6 keY and 82 keV. 8 The lower limit could be improved to about 6 keY
if full solutions of Poisson's equation are used. 7 The mass of the degenerate core is
assumed to be between 3 x 10 3Mo and 3.5 x 10 6 Mo. The black holes grow then by

aFor NFW profile, the DM particle mass is in the range of 0.6 keY ;S ms ;S 6 keY. A range
of 1 keY ;S ms ;S 14 keY is found for Moore profile and finally we obtain a mass range of
2 keY ;S ms ;S 82 keY for the isothermal density profile.
 1005

consuming the whole mass from degenerate cores and then by Eddington baryonic
matter accretion at late stages of their evolution, i.e. at t ::, 10 7 - 8 years.

3. Conclusion
Fermionic dark matter of mass of the order of a few keY could help boost the
growth of supermassive black holes at galactic centers. Moreover, NFW, Moore and
isothermal density profiles become degenerate for particles mass of a few ke V at
distances of a few parsec from the galactic centers. The detection of an X-ray line
at half the sterile neutrino by XMM-Newton and CHANDRA satellites would be
the smoking gun for the existence of a such sterile neutrino.

Acknowledgments
It is a great pleasure to thank Peter L. Biermann for encouragement and support.
The author thanks the MG 11 organisers for invitation and the Humboldt foundation
for financial support to attend the conference.

References
l. R. Schadel et al., ApJ 596, 1015 (2003)
2. A. Chez et al., em ApJ 620, 744 (2005)
3. L. Ferrarese, and D. Merritt, ApJ 539, L9 (2000)
4. S. Dodelson, and L.M. Widrow, Phys. Rev. Lett. 72, 17 (1994)
5. T. Asaka and M. Shaposhnikov, Phys. Rev. Lett. B620, 17 (2005)
6. A. Kusenko, Int. Journ. Mod. Phys. D13, 2065 (2004)
7. F. Munyaneza and P.L. Biermann, Astr. €j Astrophy, 436, 805 (2005)
8. F. Munyaneza and P.L. Biermann, Astr. €j Astrophys. 458, L15-19 (2006)
9. F. Munyaneza and P. L. Biermann, Proceedings of the Galactic Center Workshop, Bad
Honnef, 18-22 April 2006, Ed. R. Schadel et al. Journ. of Physics: Conference Series,
54, 456 (2006)
10. P.L. Biermann and A. Kusenko, Phys. Rev.Lett. 96, 091301 (2006)
1l. J. Stasielak, A. Kusenko and P. L. Biermann, ApJ in press, astro-ph/0606435 (2006)
12. J.F. Navarro, C.S. Frenk and S.D.M White, ApJ 490,493 (1997)
13. B. Moore, S. Chigna, F. Covernato et al., ApJ 524, L19 (1999)
14. RD. Viollier, Prog. Part. Nucl. Phys 32, 51 (1994)
15. D. Tsiklauri and RD. Viollier, ApJ 500, 591 (1998)
16. F. Munyaneza, D. Tsiklauri and Viollier, RD, ApJ 509, L105 (1998)
17. F. Munyaneza, D. Tsiklauri and RD. Viollier, ApJ 526, 744 (1999)
18. N. Bilic, F. Munyaneza, C. B. Tupper and RD. Viollier, Prog. Part. Nucl. Phys. 48,
291 (2002)
19. N. Bilic, F. Munyaneza, C. B. Tupper and RD. Viollier, Proceedings of International
Conference DARK2002, Cape Town, 19-22 Feb. 2002, p. 46-52, ed. H.V. Klapdor-
Kleingrothaus, and R.D. Viollier, (Berlin Springer, 2002)
20. Bilic, N., Munyaneza, F., Viollier, R D. 1999, Phys. Rev. D. 59, 024003 (1999)
2l. F. Munyaneza and RD. Viollier, ApJ 564, 274 (2002)
22. M. Mappelli and A. Ferrara, MNRAS 364, 2 (2005)
23. C.J. Willot, RJ. McLure and M. Jarvis, ApJ 587, L15 (2003)
 DARK MATTER: THE CASE OF STERILE NEUTRINO

MIKHAIL SHAPOSHNIKOV
Institut de Theorie des Phenomfmes Physiques, Ecole Poly technique Federale de Lausanne,
CH-I015 Lausanne, Switzerland

An extension of the Standard Model by three right-handed neutrinos with masses smaller
than the electroweak scale (the vMSM) can explain simultaneously dark matter and
baryon asymmetry of the Universe, being consistent with the data on neutrino oscilla-
tions. A dark matter candidate in this theory is the sterile neutrino with the mass in keV
range. We discuss the constraints on the properties of this particle and mechanisms of
their cosmological production. Baryon asymmetry generation in this model is reviewed.
Crucial experiments that can confirm or rule out the v MSM are briefly discussed.

1. Introduction
There is compelling evidence that the Minimal Standard Model (MSM) of strong
and electroweak interactions is not complete. There are several experimental facts
that cannot be explained by the MSM. These are neutrino oscillations, the presence
of dark matter in the Universe, the baryon asymmetry of the Universe, its flatness,
and the existence of cosmological perturbations necessary for structure formation.
Indeed, in the MSM neutrinos are strictly massless and do not oscillate. The MSM
does not have any candidate for non-baryonic dark matter. Moreover, with the
present experimental limit on the Higgs mass, the high-temperature phase transi-
tion, required for electroweak baryogenesis, is absent. In addition, it is a challenge
to usc CP-violation in Kobayashi-Maskawa mixing of quarks to produce baryon
asymmetry in the MSM. Finally, the couplings of the single scalar field of the MSM
are too large for the Higgs boson to play the role of the inflaton. This means that
the MSM is unlikely to be a good effective field theory up to the Planck scale.
In l - 3 it was proposed that a simple extension of the MSM by three singlet right-
handed neutrinos and by a real scalar field (inflaton) with masses smaller than the
electroweak scale may happen to be a correct effective theory up to some high-
energy scale, which may be as large as the Planck scale. This model was called "the
vMSM", underlying the fact that it is the extension of the MSM in the neutrino
sector. Contrary to Grand Unified Theories, the vMSM does not have any internal
hierarchy problem, simply because it is a theory with a single mass scale. Moreover,
as the energy behaviour of the gauge couplings in this theory is the same as in
the MSM, the absence of gauge-coupling unification in it indicates that there may
be no grand unification, in accordance with our assumption of the validity of this
theory up to the Planck scale. As well as the MSM, the vMSM does not provide any
explanation why the weak scale is much smaller than the Planck scale. Similarly to
the MSM, all the parameters of the vMSM can be determined experimentally since
only accessible energy scales are present.
As we demonstrated in,1,2 the vMSM can explain simultaneously dark matter
and baryon asymmetry of the Universe being consistent with neutrino masses and

1006
 1007

mixings observed experimentally. Moreover, in 3 we have shown that inclusion of

an inflaton with scale-invariant couplings to the fields of the vMSM allows us to
have inflation and provides a common source for electroweak symmetry breaking
and Majorana neutrino masses of singlet fermions - sterile neutrinos. The role of
the dark matter is played by the lightest sterile neutrino with mass ms in the keY
range. In addition, the coherent oscillations of two other, almost degenerate, sterile
neutrinos lead to the creation of baryon asymmetry of the U niverse 2 through the
splitting of the lepton number between active and sterile neutrinos 4 and electroweak
sphalerons. 5 For other astrophysical applications of sterile neutrinos see. 6
In this talk I review the structure of the vMSM and discuss its dark matter
candidate - sterile neutrino. The baryogenesis in this model is briefly reviewed.

2. The vMSM
If three singlet right-handed fermions NJ are added to the Standard Model, the
most general renormalizable Lagrangian describing all possible interactions has the
form:
- - MJ -
L vMSM = LMS M + N Ji8p"p, NJ - FaJ LaN[if? - 2 N'jNJ + h.c., (1)

where LMs M is the Lagrangian of the MSM, if? and La (n = e, jL, T) are the Higgs
and lepton doublets, respectively, and both Dirac (M D = r(if?)) and Majorana
(MJ) masses for neutrinos are introduced. In comparison with the MSM, the vMSM
contains 18 new parameters: 3 Majorana masses of new neutral fermions N i , and
15 new Yukawa couplings in the leptonic sector (corresponding to 3 Dirac neutrino
masses, 6 mixing angles and 6 CP-violating phases).
Let us discuss in general terms what kind of scale for Majorana neutrino masses
MJ one could expect. If Dirac neutrino masses (MD)aJ = FaJv (where v = 174
GeV is the vacuum expectation value of the Higgs doublet) are much smaller than
the Majorana masses M J , the see-saw formula for active neutrino masses

(2)

is valid. Though it is known that the masses of act.ive neutrinos are smaller than
0(1) eV, it is clear that the scale of Majorana neutrino masses cannot be extracted.
This is simply because the total number of physical parameters describing mv is
equal to 9 (three absolute values of neutrino masses, three mixing angles and three
CP-violating phases), which is two times smaller than the number of new parameters
in the vMSM.
A most popular proposal7 is to say that the Yukawa couplings F in the active-
sterile interactions are of the same order of magnitude as those in the quark and
charged lepton sector. This choice is usually substantiated by aesthetic considera-
tions, but is not following from any experiment. Then one has to introduce a new
energy scale, MJ rv 10 10 _10 15 Ge V, which may be related to grand unification. The
model with this choice of MJ has several advantages in comparison with the MSM:
1008

it can explain neutrino masses and oscillations, and give rise to baryon asymme-
try of the Universe through leptogenesis 8 and anomalous electroweak number non-
conservation at high temperatures. s However, it cannot explain the dark matter as
the low energy limit of this theory is simply the MSM with non-zero active neutrino
masses corning from dimension five operators. On a theoretical side, as a model
with two very distinct energy scales it suffers from a fine-tuning hierarchy problem
NIr » A1w. Also, since the energy scale which appears in this scenario is so high,
it would be impossible to make a direct check of this conjecture by experimental
means.
Another suggestion is to fix the Majorana masses of sterile neutrinos in 1-10 eV
energy scale 9 to accommodate the LSND anomaly.lO The theory with this choice
of parameters, however, cannot explain the baryon asymmetry of the Universe and
docs not provide a candidate for dark matter particle.
Yet another paradigm is to determine the parameters of the I/MSM from avail-
able observations, i.e. from requirement that it should explain neutrino oscillations,
clark matter and baryon asymmetry of the universe in a unified way. It is this
choice that will be discussed below. It cloes not require introduction of any new en-
ergy scale, and NIr < Nl w . In this case the Yukawa couplings must be much smaller
than those in the quark sector. F < 10- 6 . The theory has a number of directly
testable predictions, which can confirm or reject it.

3. Dark matter
Though the 1/MSM does not offer any stable particle besides those already present
in the MSM, it contains a sterile neutrino with a life-time exceeding the age of the
Universe, provided the corresponding Yukawa coupling is small enough. The decay
rate of Nl to three active neutrinos and antineutrinos (assuming that N1 is the
lightest sterile neutrino) is given by

- CF
2 M5g2
r 1
(3)
3v - 967f3

where C F is the Fermi constant. For example, a choice of rno rv 0(1) eV and of
Nh rv 0(1) keV leads to a sterile neutrino life-time rv 10 17 years.ll
The mass of the sterile dark matter neutrino cannot be too small. An application
of the Tremaine-Cunn arguments 12 to the dwarf spheroidal galaxies 13 gives the
lower bOlll1d 14 Ivh > 0.3 keY. If the sterile neutrino mass is in the keY region, it
may playa role of warm dark matter .15.16 Sterile neutrino free streaming length an
matter-radiation equality is given by

AFS "::' 1Mp s (1 M1

keY) ((Ps/T;)
3.15 (4)

and the mass inside AFS is

3 3
MFS ,,::,3x1010M.(') (lkeV)
Ah
((Ps/T;)
3.15 '
 1009

10

1e-11

tii
en
.5
Ci
::l :>
0 1e-12 ~
0
Cl
~ 0.1
E
J:1::l
>-

1e-13
0.01

10 100

Fig. L Upper bound on Yukawa coupling constant (left vertical axis) and Dirac mass (right
vertical axis) of dark matter sterile neutrino, coming from X-ray observations of Large Magellanic
Cloud (LMC) and Milky Way (MW) by XMM-Newton and HEAO-l satellites.

where < Ps > « Pa » is an average momentum of sterile (active) neutrino at

the moment of structure formation, M8 is the solar mass. One normally defines
cold dark matter (CDM) as that corresponding to Alps < 10 5 lv18 ) hot DM as
the one with Mps > 10 14 lVI8 ) and warm DM as anything in between. Potentially)
WDM could solve some problems of the CDM scenario, such as the missing satellites
problem 17 ,lS and the problem of cuspy profiles in the CDM distributions. 19 ,20
Even stronger constraint on the mass of sterile neutrino comes from the analysis
of the cosmic microwave background and the matter power spectrum inferred from
Lyman-a forest data21 ,22 : Ml > }.ifo (~~:~). According to,23 lVlo = 14.5 keY)
whereas 24 gives }.ifo = 10 keY.
Yet another constraint on the parameters of dark matter sterile neutrino comes
from radiative decay Nl -+ V,(, suppressed in comparison with Nl 3v by a
factor O(CY) (a is a fine structure constant). This two body decay produces a line
in the spectrum of X-rays corning from dark matter in the Universe; corresponding
constraints are discussed in detail in other contributions to these proceedings) see
also Refs. 25 -35 To get an idea on admitted Yukawa coupling constants and of Dirac
mass of dark matter sterile neutrino see Fig. 1, based on results of Ref. 2s ,32
1010

4. Cosmological production of sterile neutrinos

Let us discuss now cosmological production of sterile neutrinos. In the region of
the parameter space admitted by X-ray observations sterile neutrinos were never
in thermal equilibrium in the early Universe. 15 This means that their abundance
cannot be predicted in the framework of the vMSM: 36 one should either fix the
concentration of sterile neutrinos at temperatures greater than 1 GeV, or specify
the physics beyond the vMSM.
One can address the question how many sterile neutrinos are produced due to the
vMSM interactions, eq. (1), i.e. because of the mixing with active neutrino flavours
characterized by parameter e. In fact, this mixing is temperature dependent: 37

e ---+ eM e
~ ----:---,--:=-~...,..-::-::--~-;;-;::-::-::-;::-::-~
- 1 + 2.4(T /200 MeV)6(keV /MI)2 '
r of sterile neutrino production is strong suppressed at T > 100
so that the rate
MeV, rex T- 7 . The rate peaks roughly at 15 Tpeak "-' 130 k~V
3
C
MeV, which
corresponds to the temperature of the QCD cross-over for ke V scale sterile neutri-
r/
nos. This fact makes an exact estimate of the number of produced sterile neutrinos
to be a very difficult task (see 36 for a discussion of the general formalism for com-
putation of sterile neutrino abundance), since Tpeak happens to be exactly at the
point where the quark-gluon plasma is strongly coupled and the dilute hadron gas
picture is not valid. The chiral perturbation theory works only at T < 50 MeV. The
perturbation theory in QCD works only at T» A QCD , and the convergence is very
slow. The lattice simulations work very well for pure gluodynamics. However, no
results with three light quarks and with reliable extrapolation to continuum limit
are available yet. Also, the treatment of hadronic initial and final states in reactions
v + q ---+ v + q, q + q ---+ vv is quite uncertain. In refs.u,15 the computation of
sterile neutrino production was done with the use of simplified kinetic equations
and without accounting for hadronic degrees of freedom. In 16 ,38 some effects related
to existence of quarks and hadrons in the media were included; the same type of ki-
netic equations were used. In 39 a computation of sterile neutrino production based
on first principles of statistical physics and quantum field theory has been done
and uncertainties related to hadronic dynamics were analyzed. The results are pre-
sented in Fig. 2. They correspond to the case when there is no entropy production
(S = 1) due to decay of heavier sterile neutrinos of the vMSM. 40 The area above
dotted line is certainly excluded: the amount of produced dark matter would lead
to over-closer of the universe. The region below dashed line is certainly allowed:
the amount of sterile neutrinos produced due to active-sterile transitions is smaller
than the amount of dark matter observed. Any point in the region between two solid
lines (corresponding to the "most reasonable" model for hadronic contribution39 )
can lead to dark matter generation entirely due to active-sterile transitions. Maxi-
mal variation of the hadronic model, defined in 39 extends this region to the space
between dotted and dashed lines. In the case of entropy production with S > 1 all
these four lines simply move up by a factor S.
 1011

mean value
absolute upper bound
absolute lower bound

Fig. 2. X-ray constraints from 28 •32 versus required mixing of sterile neutrino in Dodelson-Widrow
scenario. It is assumed that no entropy production from decays of heavier sterile neutrinos of the
uMSM is taking place. The area between two solid lines corresponds to all possible variations
of mixing angles to different leptonic families for "best choice" hadronic dynamics. 39 The area
between dotted and dashed lines corresponds to most conservative estimate of hadronic uncertain-
ties. 39

One can see that the active-sterile mixing can accommodate for all dark matter
only if Ml < 3.5 keY, if the "most reasonable" hadronic model is taken. The most
conservative limit would correspond to Nh < 6 keY, if all hadronic uncertainties are
pushed in the same direction and the uncertainty by a factor of 2 is admitted for
the X-ray bounds. Therefore, if Lyman-a constraints of 2:j ,24 are taken for granted,
the production of sterile neutrinos due to active-sterile neutrino transitions happens
to be too small to account for observed abundance of dark matter. In other words,
physics beyond the vMSM is likely to be required to produce dark matter sterile
neutrinos. Another option is to assume that the universe contained relatively large
lepton asymmetries. 41
In 3 it was proposed the the vMSM may be extended by a light inflaton in order to
accommodate inflation. To reduce the number of parameters and to have a common
source for the Higgs and sterile neutrino masses the inflatoIl-vMSM couplings can
1012

be taken to be scale invariant on the classical level:

LvMSM ---+ LvMSM[M~Ol + "21 (°I"X) 2 - 2II -C

NI NIX + h.c. - ( )
V ([>, X , (5)

where the Higgs-infiaton potential is given by:

V([>,X) = A ([>t([> - ~X2)2 + ~X4 - ~m~x2.

The requirement that the chaotic infiation 42 produces the correct amplitude for
scalar perturbations leads to the constraints:
(3 C::' 10- 13 , a:S 10- 7 , II:S 10- 3 .

For a > (3 infiaton mass is smaller than the Higgs mass, mI < M H .
One can show 3 that the infiaton with mass m [ > 300 MeV is in thermal equi-
librium thanks to reactions X ---+ e+ e-, X ---+ p,+ p,- down to T < mI. The sterile
neutrino abundance due to infiaton decays: X ---+ N N is given by
o r Moms 21T((5)
s C::' 0.26 my x 12 eV ((3)

So, for mI '"" 300 MeV (mI '"" 100 GeV) the correct Os is obtained for ms '"" 16 - 20
keY (ms '"" 0(10) MeV). A sterile neutrino in this mass range is perfectly consistent
with all cosmological and astrophysical observations. As for the bounds on mass
versus active-sterile mixing coming from X-ray observations of our galaxy and its
dwarf satellites, they are easily satisfied since the production mechanism of sterile
neutrinos discussed above has nothing to do with the active-sterile neutrino mixing
leading to the radiative mode of sterile neutrino decay.

5. Baryon Asymmetry of the Universe

The baryon (B) and lepton (L) numbers are not conserved in the vMSM. The lep-
ton number is violated by the Majorana neutrino masses, while B + L is broken by
the electroweak anomaly. As a result, the sphaleron processes with baryon number
non-conservation are in thermal equilibrium for 100 GeV < T < 10 12 GeV. As for
CP-breaking, the vMSM contains 6 CP-violating phases in the lepton sector and
a Kobayashi-Maskawa phase in the quark sector. This makes two of the Sakharov
conditions 43 for baryogenesis satisfied. Similarly to the MSM, this theory does not
have an electroweak phase transition with allowed values for the Higgs mass,44 mak-
ing impossible the electroweak baryogenesis, associated with the non-equilibrium
bubble expansion. However, the vMSM contains extra degrees of freedom - sterile
neutrinos - which may be out of thermal equilibrium exactly because their Yukawa
couplings to ordinary fermions are very small. The latter fact is a key point for the
baryogenesis in the vMSM, ensuring the validity of the third Sakharov condition.
In 4 it was proposed that the baryon asymmetry can be generated through CP-
violating sterile neutrino oscillations. For small Majorana masses the total lepton
number of the system, defined as the lepton number of active neutrinos plus the
 1013

total helicity of sterile neutrinos, is conserved and equal to zero during the Universe's
evolution. However, because of oscillations the lepton number of active neutrinos
becomes different from zero and gets transferred to the baryon number due to rapid
sphaleron transitions. Roughly speaking, the resulting baryon asymmetry is equal
to the lepton asymmetry at the sphaleron freeze-out.
The kinetics of sterile neutrino oscillations and of the transfers of leptonic num-
ber between active and sterile neutrino sectors has been worked out in. 2 The effects
to be taken into account include oscillations, creation and destruction of sterile and
active neutrinos, coherence in sterile neutrino sector and its lost due to interaction
with the medium, dynamical asymrnetries in active neutrinos and charged leptons.
The corresponding equations are written in terms of the density matrix for sterile
neutrinos and concentrations of active neutrinos and are rather lengthy and will
not be presented here due to the lack of space. They can be found in the original
work. 2 The corresponding equations are to be solved with the choice of the vMSM
parameters consistent with the experiments on neutrino oscillations and with the
requirement that dark matter neutrino has the necessary properties.
The value of baryon to entropy ratio n: can be found from the solution of the
kinetic equations and is given by2

where lVh,3 are the masses of the heavier sterile neutrinos, ~lVfj2 = Nfj - Nfi, and
the CP-breaking factor 6cp is expressed through the different mixing angles and
CP-violating phases, parameterizing the Dirac neutrino masses, and can be 0(1),
given the present experimental data on neutrino oscillations. This shows that the
correct baryon asymmetry of the Universe n: ~ (8.8 - 9.8) x 10- 11 is generated
when the heavier sterile neutrinos with the masses, say, 1 Ge V are degenerate to
one part in 10 5 . This looks like a strong fine tuning but may also indicate that there
exists some symmetry making the degeneracy automatic. 45
It is interesting to note that for masses of sterile neutrinos> 100 GeV the
mechanism does not work as the sterile neutrinos equilibrate. Also, the temperature
of baryogenesis is rather low, TL ~ (~M2 Mpl)~ > 10 2 GeV, i.e. validity of vMSM
is only required at the scales smaller than Mw or so.

6. Conclusions
The vMSM is, perhaps, the simplest and the most economical extension ofthe Mini-
mal Standard Model. It shares with the MSM its advantages (renormalizability and
agreement with most particle physics experiments) and its fine-tuning problems
(the gauge hierarchy problem, flavour problem, etc). However, unlike the MSM,
the vMSM can explain simultaneously three different phenomena, observed experi-
mentally, namely neutrino oscillations, dark matter, and baryon asymmetry of the
Universe. The parameter-space of the model is rather constrained: the dark matter
1014

neutrino should have a mass in the ke V region and be much lighter than two heav-
ier sterile neutrinos, which are required to be quite degenerate. The model has a
number of testable predictions. In astrophysics, one should search for X-rays from
decays of dark matter neutrinos, which could be achieved with a X-ray spectrome-
ter in Space with good energy resolution I5E / E rv 10- 3 ~ 10- 4 getting signals from
our Galaxy and its Dwarf satellites. 28 ,35 In particle physics, the vMSM predicts the
absolute values of active neutrino masses, 1,46 and existence of relatively light singlet
fermions 45 which can be searched for in decays of charmed, beauty and even K or
'if-mesons in experiments similar to. 47 ,48 The dark matter neutrino can be looked
for in p-decays of tritium and other isotopes. 49

Acknowledgments
This work was supported in part by the Swiss National Science Foundation. It
is a pleasure to thank Takehiko Asaka, Fedor Bezrukov, Steve Blanchet, Alexey
Boyarsky, Alexander Kusenko, Mikko Laine, Andrei Neronov, Oleg Ruchayskiy,
and Igor Tkachev for collaboration.

References
1. T. Asaka, S. Blanchet and M. Shaposhnikov, Phys. Lett. B 631 (2005) 151.
2. T. Asaka and M. Shaposhnikov, Phys. Lett. B 620 (2005) 17.
3. M. Shaposhnikov and I. Tkachev, Phys. Lett. B 639 (2006) 414.
4. E. Akhmedov, V. Rubakov and A. Smirnov, Phys. Rev. Lett. 81 (1998) 1359.
5. V. A. Kuzmin, V. A. Rubakov and M. E. Shaposhnikov, Phys. Lett. B 155 (1985) 36.
6. A. Kusenko and G. Segre, Phys. Rev. Lett. 77 (1996) 4872; A. Kusenko and G. Segre,
Phys. Lett. B 396 (1997) 197; G.M. Fuller, A. Kusenko, I. Mocioiu and S. Pascoli,
Phys. Rev. D 68 (2003) 103002; M. Barkovich, J.C. D'Olivo and R. Montemayor,
Phys. Rev. D 70 (2004) 043005; M. Mapelli, A. Ferrara and E. Pierpaoli, Mon.
Not. Roy. Astron. Soc. 369 (2006) 1719; E. Ripamonti, M. Mapelli and A. Fer-
rara, arXiv:astro-ph/0606482; E. Ripamonti, M. Mapelli and A. Ferrara, arXiv:astro-
ph/0606483; P.L. Biermann and A. Kusenko, Phys. Rev. Lett. 96 (2006) 091301;
J. Stasielak, P. L. Biermann and A. Kusenko, astro-ph/0606435; A. Kusenko, hep-
ph/0609081; F. Munyaneza and P.L. Biermann, astro-ph/0609388; J. Hidaka and
G.M. Fuller, astro-ph/0609425.
7. P. Minkowski, Phys. Lett. B 67 (1977) 421; T. Yanagida, Progr. Theor. Phys. 64
(1980) 1103 ; M. Gell-Mann, P. Ramond and R. Slansky, in Supergravity, North Hol-
land, Amsterdam 1980.
8. M. Fukugita and T. Yanagida, Phys. Lett. B 174 (1986) 45.
9. A. de Gouvea, Phys. Rev. D 72 (2005) 033005.
10. A. Aguilar et al. [LSND Collaboration]' Phys. Rev. D 64 (2001) 112007.
11. A. D. Dolgov and S. H. Hansen, Astropart. Phys. 16 (2002) 339.
12. S. Tremaine and J. E. Gunn, Phys. Rev. Lett. 42 (1979) 407.
13. D. N. C. Lin and S. M. Faber, Astrophys. J. 266 (1983) L21.
14. J. J. Dalcanton and C. J. Hogan, Astrophys. J. 561 (2001) 35.
15. S. Dodelson and L. M. Widrow, Phys. Rev. Lett. 72 (1994) 17 .
16. K. Abazajian, G. M. Fuller and M. Patel, Phys. Rev. D 64 (2001) 023501.
17. B. Moore et al., Astrophys. J. 524 (1999) L19.
 1015

18. P. Bode, J. P. Ostriker and N. Turok, Astrophys. J. 556 (2001) 93.

19. T. Goerdt et al., Mon. Not. Roy. Astron. Soc. 368 (2006) 1073.
20. G. Gilmore et al., arXiv:astro-ph/0608528.
21. S.H. Hansen, J. Lesgourgues, S. Pastor and J. Silk, Mon. Not. Roy. Astron. Soc. 333
(2002) 544.
22. M. Viel et al., Phys. Rev. D71 (2005) 063534.
23. U. Seljak et al., Phys. Rev. Lett. 97 (2006) 191303.
24. M. Viel et al., Phys. Rev. Lett. 97 (2006) 07130l.
25. K. Abazajian, G. M. Fuller and W. H. Tucker, Astrophys. J. 562 (2001) 593.
26. A. Boyarsky, A. Neronov, O. Ruchayskiy and M. Shaposhnikov, Mon. Not. Roy. As-
tron. Soc. 370 (2006) 213.
27. A. Boyarsky, A. Neronov, O. Ruchayskiy and M. Shaposhnikov, Phys. Rev. D74
(2006) 103506.
28. A. Boyarsky et al., arXiv:astro-ph/0603660.
29. S. Riemer-Sorensen, S. H. Hansen and K. Pedersen, Astrophys. J. 644 (2006) L33.
30. C. R. Watson, J. F. Beacom, H. Yuksel and T. P. Walker, Phys. Rev. D 74 (2006)
033009.
31. S. Riemer-Sorensen, K. Pedersen, S. H. Hansen and H. Dahle, astro-ph/0610034.
32. A. Boyarsky, J. Nevalainen and O. Ruchayskiy, astro-ph/061096l.
33. A. Boyarsky, O. Ruchayskiy and M. Markevitch, astro-ph/0611168.
34. K. N. Abazajian, M. Markevitch, S. M. Koushiappas and R. C. Hickox, arXiv:astro-
ph/0611144.
35. A. Boyarsky, J. W. d. Herder, A. Neronov and O. Ruchayskiy, arXiv:astro-ph/0612219.
36. T. Asaka, M. Laine and M. Shaposhnikov, JHEP 0606 (2006) 053.
37. R. Barbieri and A. Dolgov, Phys. Lett. B 237 (1990) 440.
38. K. Abazajian, Phys. Rev. D 73 (2006) 063513.
39. T. Asaka, M. Laine and M. Shaposhnikov, arXiv:hep-ph/0612182.
40. T. Asaka, M. Shaposhnikov and A. Kusenko, Phys. Lett. B 638 (2006) 40l.
41. X. Shi and G.M. Fuller, Phys. Rev. Lett. 82 (1999) 2832.
42. A. D. Linde, Phys. Lett. B 129 (1983) 177.
43. A. D. Sakharov, Pisma ZhETF 5 (1967) 32.
44. K. Kajantie et al., Phys. Rev. Lett. 77 (1996) 2887.
45. M. Shaposhnikov, arXiv:hep-ph/0605047.
46. A. Boyarsky, A. Neronov, O. Ruchayskiy and M. Shaposhnikov, JETP Lett. 83 (2006)
133.
47. G. Bernardi et al., Phys. Lett. B 203 (1988) 332.
48. P. Astier et al. [NOMAD Collaboration]' Phys. Lett. B 506 (2001) 27.
49. F. Bezrukov and M. Shaposhnikov, arXiv:hep-ph/0611352.
This page intentionally left blank
Neutrino Masses:
Experimental
This page intentionally left blank
 NEUTRINO BACKGROUND, DIFFUSE BACKGROUNDS AND
CMB: IS THE PICTURE CONSISTENT?

L.A. POPA * and A. VASILE**

Institute for Space Sciences,
Bucharest - Magurele, Ro-76900 Romania
* [email protected]
** [email protected]

Distortions of CMB temperature and polarization maps caused by gravitational lensing,

observable with high angular resolution and sensitivity, can be used to constrain the
sterile neutrino mass, ms, from CMB data alone. We forecast ms >1.75 keV from PLANCK
and ms >4.97 keV from INFLATION PROBE at 95% CL, by using the CMB weak lensing
extraction.

1. Introduction

The confluence of the most recent experimental data of the cosmic microwave back-
ground (CMB) anisotropies, large-scale structure (L88) galaxy surveys, supernovae
luminosity distance, Lyman-a forest and Hubble parameter, have lead in specifying
the ACDM model as the cosmological concordance model. 1 According to this model
the Universe is spatially nearly flat with energy densities of n CDM =O.27±O.07 in
cold dark matter (CDM) particles, nb =O.044±O.004 in baryons, nA=O.70±O.03 in
dark energy and Hubble constant H o=72±5 km S-l Mpc- 1 .
The direct confirmation of this theory was the detection of the acoustic Doppler
peaks structure of the CMB angular power spectrum. Further successes are related
to the correct prediction of the hierarchical structure formation via gravitational
instability, the abundance of clusters at small redshifts, the spatial distribution and
the number density of galaxies, the L88 matter power spectrum, the Lyman-a forest
amplitude and spectrum.
Despite its successes on large scales, the ACDM model produces too much power
on small scales. In general, the observed structures have softer cores, lower concen-
trations and are less clumped than those predicted by the ACDM mode1. 2
A possibility to alleviate the accumulating contradiction between ACDM pre-
dictions on small scales and observations is to add properties to the dark matter
sector, relaxing the hypothesis on dark matter as being cold. Free streaming due
to thermal motion of dark matter particles is the simplest known mechanism for
smearing out small scale structure. For example, the velocity dispersion of warm
dark matter (WDM) particles is sufficient to alleviate some of these problems. 3
Sterile neutrinos are considered the most promising WDM candidates. Lower
limits on sterile neutrino mass have been placed from various observational probes a .
The combination of CMB measurements, LS8 and Lyman-a forest power spectra
lead to rns > 1. 7 ke V with a further improvement to rns > 3 keV when high-

aThroughout the paper the sterile neutrino mass is quoted at 95% CL.

1019
1020

resolution Lyman-a spectra are considered. 4 The upper limit of sterile neutrino mass
is constrained by the limits on its radiative decay from Virgo cluster observations
and by the observations of the diffuse X-ray background. The combination of all
above constraints allows the range 1. 7 ke V < ms <8.2 ke V for the mass of this candi-
date. A significantly more stringent lower limit constraint, ms > 14 keY, was placed
by using the Lyman-a forest power spectrum and high-resolution spectroscopy ob-
servations in combination with CMB and galaxy clustering data, excluding sterile
neutrino as dark matter candidate. 5

2. Sterile neutrino mass from CMB lensing extraction

Like active neutrinos, sterile neutrinos can not cluster via gravitational instability
on scales below the free-streaming scale, with important implications for the growth
of density perturbations at late times.
The alteration of the gravitational potential changes the gravitational lensing of
the CMB photons. Week lensing introduces a deflection field so that the deflection
angle and the projected gravitational potential power spectra are related through
Cdd = 1(1 + l)C;<'<1>.
We modified the CMB anisotropy code CAMB to compute the lensed CMB
temperature and polarization anisotropy power spectra and the projected gravita-
tional potential power spectrum in the presence of a sterile neutrino component. We
include in the computation the momentum-dependent sterile neutrino phase-space
distribution function,1 its unperturbed and perturbed energy density and pressure,
energy flux and shear stress. The CMB weak lensing map can be reconstructed from
the statistical analysis of the CMB temperature and polarization anisotropy maps.

Table l. Experimental characteristics: v is the frequency of the channel, Bb is the FWHM,

aT and ap are the sensitivities per pixel for temperature and polarization maps.

Experiment Frequency(GHz) FWHM (arc-minutes) aT (/tK) ap(/tK)

100 9.5 6.8 10.9
PLA:-lCK 143 7.1 6.0 11.4
217 5.0 13.1 26.7
70 6.0 0.29 0.41
INFLATION PROBE 100 4.2 0.42 0.59
150 2.8 0.63 0.88
220 l.9 0.92 1.30

To evaluate the ability of the future CMB experiments PLANCK and INFLA-
TION PROBE to detect the sterile neutrino mass we employ the quadratic estimator
6
method to compute the expected noise power spectrum from lensing extraction.
For each experiment we construct the covariance matrix;

C=
 1021

where Of x with X {T, E} are the power spectra of primary anisotropies, and
Nf x are the corresponding noise power spectra, Ofd is deflection angle power spec-
trum, Nldd the noise power spectrum associated to the lensing extraction and oT d is
the power spectrum of the cross-correlation between the temperature and deflection
angle. The above power spectra, presented in the left panel of Fig.I, are obtained
for our fiducial model, the ACDM concordance model. We assume adiabatic ini-
tial conditions, primordial scalar density perturbations with scalar spectral index
71 8 =0.95 and three active neutrino flavors with the total mass m" =0.7eV.

10 100 1000
l 2
sin 28

Fig. 1. Left: The CMB temperature and polarization power spectra, CT T and CfE, and the
deflection angle power spectrum, d
ct
, for the fiducial model. Dashed lines (from top to bottom)
are the noise power spectra associated to temperature, polarization and lensing extraction for
PLANCK (blue dashed) and INFLATION PROBE (red dash-dotted). Right: Constraints on sterile
neutrino parameter space potentially obtained by the future PLANCK and INFLATION PROBE from
CMB weak lensing extraction compared with other experimental constraints (see also the text).

In the right panel of Fig.I. we present the constraints on sterile neutrino pa-
rameter space that can be potentially obtained by the future experiments PLANCK
and INFLATION PROBE from the CMB weak lensing extraction. We found ms >1.75
keY from PLANCK and ms >4.97 keY from INFLATION PROBE at 95% CL.

References
1. D.N. Spergel et al., astro-phj0603449.
2. P. Bode, J.P. Ostriker and N. Turok, ApJ 556, 93 (2001).
3. T. Padmanabhan, Structure Formation in the Universe (Camb. Univ. Press, 1995).
4. K. Abazajian, Phys. Rev. D73, 063513 (2006).
5. U. Seljak, A. Makarov, P. McDonald, H. Trac, Phys. Rev. Lett. 96, 191303 (2006).
6. J. Lesgourgues, L. Perotto, S. Pastor, M. Piat, Phys. Rev. D73, 045021 (2006).
7. K. Abazajian, G.M.Fuller, M. Patel, Phys. Rev. D64, 3501, (2001).
 CONSTRAINING THE COSMOLOGICAL LEPTON ASYMMETRY
THROUGH COSMIC MICROWAVE BACKGROUND
OBSERVATIONS

MASSIMILIANO LATTANZI
Department of Astrophysics, University of Oxford, Denys Wilkinson Building, Keble Road,
Oxford OX1 3RH, United Kingdom
INFN - Istituto Nazionale di Fisica Nucleare, Via E. Fermi 40, 00044 Frascati (Roma), Italy
E-mail: [email protected]

REMO RUFFINI and GREGORY V. VERESHCHAGIN

International Center for Relativistic Astrophysics and Dipartimento di Fisica, Universitd di
Roma "La Sapienza", P.le A. Moro 2, 1-00185 Roma, Italy

Introduction. -The presence in the Universe of a relic neutrino background (NB)

is one of the basic predictions of the hot big bang model of cosmology. In the
standard framework, these neutrinos follow a thermal distribution, with a present
day temperature T = 1.9 K, corresponding to a number density of order 100
particles/ cm 3 per species. The presence of the NB influences cosmological evolution,
and this means in turn that cosmological observations can be used to constraint its
properties; this is the basis of neutrino cosmology.
We know today from experiments on neutrino flavour oscillation that neutrinos
are massive, although oscillation experiments are unfortunately insensitive to the
absolute value of neutrino masses. The experimental results on tritium beta decayl
constraint the neutrino masses to be mv < 2.05 - 2.3 eV; the KATRIN project is
expected to improve this bound by an order of magnitude. 2 In the framework of a
ACDM cosmological model, the analysis of cosmological data constraint the sum
of neutrino masses Mv == 2:mv :s
0.5 - 2 eV, the exact result depending on the
assumed priors and on the data used (see Ref. 3 for a recent review).
Owing to the large amount and high precision recently reached by cosmological
data, it is worth to consider scenarios in which NB evolution deviates from the stan-
dard picture, and compare their predictions to current observations. A possibility
is represented by scenarios in which the present cosmological lepton asymmetry is
eventually much larger that the corresponding baryonic one, the latter being of or-
der 10- 10 . Several mechanisms, all deeply rooted in theoretical particle physics, that
lead to the production of large (in some cases of order unity) leptonic asymmetry
in the early Universe, have been discussed in the literature. 4
Cosmological Lepton Asymmetry.-The presence of an asymmetry in the lep-
tonic sector results in a non-zero chemical potential for neutrinos; this is parame-
terized in terms of the dimensionless degeneracy parameter ~, i.e., the ratio between
the chemical potential and the temperature at the time of neutrino decoupling. Big
Bang Nucleosynthesis (BBN) constraints -0.01 < ~e < 0.22 and I~!L,TI < 2.6. 5 These
limits are derived in the standard scenario with 3 active neutrino families and no
extra radiation; they are possibly different in the case of non-standard scenarios. 6
It has also been shown that, for the values of the neutrino mixing parameters cur-

1022
 1023

rently favoured by particle physics experiments, the oscillations between flavours

equalize the chemical potentials at BBN, thus resulting in the stringent bound
-0.05 < ~ < 0.07 that applies indistinctly to all flavours. 7 However, the presence of
extra relativistic degrees of freedom would make this limit relax, while it could be
possibly totally evaded, in the presence of a neutrino-majoron coupling. 8
Here we consider a class of cosmological models that extend the usual minimal
flat ACDM model to take into account the presence of 3 equally massive neutrinos
species with an eventual lepton asymmetry. Since this is constrained to be small if
only the standard model neutrinos are present, we also consider the presence of some
additional relativistic species, in the early Universe, in such a way that the BBN
bound can be evaded. The introduction of a non-zero neutrino chemical potential
alter the cosmological evolution mainly in three ways, namely: i) it increases the
radiation density in the early Universe; this makes matter-radiation equality happen
later; ii) it increases the present density of neutrinos; this changes the curvature
of the Universe, or, in models with fixed curvature, has to be compensed by a
smaller density in the other components; iii) it increases the average momentum
of neutrinos, thus making the Landau damping of neutrino density perturbations
more efficient. The combined effect on the cosmic microwave background (CMB)
and matter power spectra has been first studied in Ref. 9. Since the changes are
reflected on these cosmological observables, we can in principle use them to constrain
the models in consideration.
Method.-In order to do this, we used the CMBFast code IO to compute the spec-
trum of temperature and polarization CMB anisotropies for a grid of models with
the following parameters: baryon density Wb, total non-relativistic matter density
W m , scalar spectral index n, optical depth to reionization T, overall normalization A,
neutrino density W v , degeneracy parameter ~, effective number of relativistic species
Neff. We did not include as a free parameter the Hubble constant H Q , but instead we
put the prior h = 0.72, consistently with the recent measurements of Hubble Space
Telescope (HST) Key ProjectY Moreover, we restricted ourselves to the case of a
flat Universe. We sistematically compared the predicted CMB spectrum by each of
the models on our grid with the one that is actually observed, using the data from
the Wilkinson Microwave Anisotropy Probe (WMAP) first year mission;12 in this
way we constructed a likelihood function for our parameter space. We then obtained
95% confidence limits (CLs) for each free parameter using a bayesian approach.
Results and discussion.- The results are as follows (for a deeper discussion, see
Ref. 13). The 95% CLs for the five "standard" parameters (i.e., Wb, W m , n, T, A) do
not change significantly with respect to the ones found by the WMAP collaboration
in the case of a minimal ACDM model. The 95% CLs for the additional, neutrino-
related parameters are the following: 0 ~ Wv ~ 0.44, 0 ~ I~I ~ 1.10, -0.70 ~
D.Neff ~ 1.30, where we have introduced D.Neff == Neff - 3.04, measuring the
deviation of Neff from its standard value of 3.04. We also derived 95% CLs for the
total neutrino mass M v and for the lepton asymmetry L == (nv - nf)) / n"(; these read
o ~ ~v ~ 3.6 and 0 ~ ILl ~ 0.9. For a more complete information we show the
1024

/\
08 16
07

/\~
0.6 1.2
S
~
04
]!
:oJ 0.3 0.6
02
0.2
0 0 o /: :
0 0.5 1 15 2 25 3 35 4 45 0 02 04 0.6 0.8 12 -, -0.5 0 0.5 1 15 2 2.5
M, (eV) ILl .lN
efl

Fig. 1. Likelihood functions for !vIv , ILl, t:.N efI The dotted lines bound the 95% CLs.

likelihood curves for M v , ILl and 6.N eff in fig 1.

The results quoted above and the likelihood curves, show that WMAP data alone
cannot exclude the possibility of a large (i.e., of order unity) lepton asymmetry L,
and consequently do not rule out exotic leptogenesis scenarios where sucha a large
lepton number is produced. In effect, the standard scenario with L = 0 actually
lies outside the 68% confidence region. Even if this is of course not enough to
claim evidence, in the CMB anisotropy spectrum, of exotic physics, nevertheless we
think that it is however interesting that non-standard models are not ruled out but
actually preferred by the WMAP 1st year data. Another interesting results concerns
the limit on the neutrino mass Mv :S 3.6 eV, that is quite good although not as
stringent as the one found combining CMB and LSS data. This confirms similar
results that have been independently found, about the possibility of deriving eV
and possibly sub-e V neutrino mass limits by CMB experiments alone. 14

References
1. V.M. Lobashev, Nucl. Phys A 719 (2003) 153; C. Kraus, et al., Eur. Phys. J. C 40
(2005) 447.
2. A. Osipowicz, et al., hep-ex/0109033. KATRIN website: http://www-ik. fzk. del
katrinl
3. J. Lesgourgues, S. Pastor, Phys. Rep. 429 (2006) 307.
4. See for example, F. Takahashi, M. Yamaguchi, Phys. Rev. D69 (2004) 083506. For
further references, see Sec. 12.5 in A.D. Dolgov, Phys. Rep. 370 (2002) 333.
5. J.P. Kneller, et al., Phys. Rev. D64 (2001) 123506; S.H. Hansen, et al., Phys. Rev.
D65 (2002) 023511.
6. P. Di Bari, Phys. Rev. D65 (2002) 043509; P. Di Bari, Phys. Rev. D67 (2003) 127301.
7. A.D. Dolgov, et al., Nucl. Phys. B632 (2002) 363; K.N. Abazajian, J.F. Beacom, N.F.
Bell, Phys. Rev. D66 (2002) 013008; Y.Y.Y. Wong, Phys. Rev. D66 (2002) 025015.
8. A.D. Dolgov, F. Takahashi, Nuc!. Phys. B688 (2004) 189.
9. J. Lesgourgues, S. Pastor, Phys. Rev. D60 (1999) 103521.
10. U. Seljak, M. Zaldarriaga, Astrophys. J.469 (1996) 437.
11. W.L. Freedman, et al., Astrophys. J.553 (2001) 47.
12. D.N. Spergel, et al., Astrophys. J. Suppl. Ser.148 (2003) 175.
13. M. Lattanzi, R.J. Ruffini, G.V. Vereshchagin, Phys. Rev. D72 (2005) 063003.
14. K. Ichikawa, M. Fukugita, M. Kawasaki, Phys. Rev. D71 (2005) 043001; D.N. Spergel
et al., Astrophys. J., in press; pre-print available on astro-ph/0603449.
 POSSIBLE NEUTRINO-ANTINEUTRINO OSCILLATION UNDER
GRAVITY AND ITS CONSEQUENCES

BANIBRATA MUKHOPADHYAY
Department of Physics,Indian Institute of Science, Bangalore-560012, India
[email protected]

We show that under gravity the effective masses for neutrino and antineutrino are dif-
ferent which opens a possible window of neutrino-antineutrino oscillation even if the rest
masses of the corresponding eigenstates are same. This is due to CPT violation and pos-
sible to demonstrate if the neutrino mass eigenstates are expressed as a combination of
neutrino and anti neutrino eigenstates, as of the neutral kaon system, with the plausible
breaking of lepton number conservation. In early universe, in presence of various lepton
number violating processes, this oscillation might lead to neutrino-antineutrino asym-
metry which resulted baryogenesis from the B-L symmetry by electro-weak sphaleron
processes. On the other hand, for Majorana neutrinos, this oscillation is expected to
affect the inner edge of neutrino dominated accretion disks around a compact object
by influencing the neutrino sphere which controls the accretion dynamics, and then the
related type-II supernova evolution and the r-process nucleosynthesis.

1. Introduction
The neutrino oscillation, in the flat space, is due to difference in rest masses between
two mass eigenstates. However, in late eighties, it was first pointed outl that pres-
ence of gravitational field affects different neutrino flavors differently which violates
equivalence principle and thus governs oscillation, even if neutrinos are massless
or of degenerate mass. The neutrino oscillation with LSND data2 indeed can be
explained by degenerate or massless neutrinos with flavor non-diagonal gravita-
tional coupling. It was further argued 3 that the flavor oscillation is possible in weak
gravitational field with the probability phase proportional to the gravitomagnetic
field. The oscillation was also shown to be feasible when the maximum velocities of
different neutrino differ each other, even if they are massless. 4
All the above results are for flavor oscillation or/and without considering rigorous
general relativistic effects. However, properties of neutrino in curved spacetime have
already been discussed 5 - 7 in literature. Here we address the neutrino-antineutrino
oscillation, which violates lepton number conservation, focusing on the nature of
space-time curvature and its special effect.
While the neutrino-antineutrino oscillation under gravity is an interesting issue
on its own right, the present result is able to address two long-standing mysteries
in astrophysics and cosmology: (1) Source of abnormally large neutron abundance
to support the r-process nucleosynthesis in astrophysical site. (2) Possible origin of
baryogenesis.

2. Oscillation probability
Let us recall the fermion Lagrangian density in curved spacetime 5 ,7
£. = Fg1fi [(i--yaaa - m) + ,a'l BaJ 'Ij; = £.1 + £." (1)

1025
1026

where
(2)
the choice of unit is c = n = kB = l. £1 may be a CPT violating interaction and
thus the corresponding dispersion energy7 for neutrino and antineutrino in standard
model

(3)

Eq. (3) tells us that under gravity neutrino energy is split up from antineutrino
energy. The CPT status of £1 has been discussed in detail in our previous works. 7
Now motivated by the neutral kaon system, we consider two distinct orthonormal
eigenstates lEv> and lEv> for a neutrino and an anti neutrino type respectively.
Further we introduce a set of neutrino mass eigenstates at t = 0 as 8

Im1 >= case lEv> +sine lEv >, 1m2 >= -sine lEv> +cose lEv> . (4)
Therefore, in presence of gravity, the oscillation probability for Im1 (t) > at t = 0
to Im2(t) > at a later time t = tf can be found as

P12 = sin 2 2esin 2 6, 6 = (Ev -2

Edtf
= [(B o -lEI) + ~;2] tf, (5)

where we consider ultra-relativistic neutrinos. Normally, the rest mass difference of

particle and antiparticle is zero and thus possible 6 i= 0 is mostly due to Ba i= 0
i.e. due to gravitational coupling. Therefore, the neutrino-antineutrino oscillation
may be possible in presence of gravity provided there is a lepton number violating
process. If neutrinos exhibit Majorana mass, then lepton number violation is au-
tomatically taken care. Hence, the CPT violating nature of background curvature
coupling generates effective mass difference, while lepton number violating process
leads to oscillation between neutrino and antineutrino.
The oscillation probability is maximum at e = 7r / 4 and is zero at e = 0, 7r /2.
From eqn. (5), the oscillation length, Lose, by appropriately setting dimensions, is
obtained as
_ _ 7rnc 6.3 x 1O-19 G e V
L ose - C tf - -_- rv _ k m, (6)
B B
where B = Bo - lEI is expressed in GeV unit and the neutrino is considered to be
moving in the speed of light.

3. Consequence and Discussion

One of the situations where the gravity induced neutrino-antineutrino oscillation
may occur is the GUT era of anisotropic phase of early universe when B rv 105 GeV. 7
From eqn. (6), this leads to Lose rv 1O- 24 km which is 10 14 orders of magnitude larger
than the Planck length. This has an important implication as the size of universe
at the GUT era is within rv 10 26 times of the Planck. Therefore, the oscillation may
 1027

lead to leptogenesis and then to baryogenesis by electro-weak sphaleron processes

due to B - L conservation, what we see today.
Another plausible region for an oscillation of this kind to occur is the inner
accretion disk of the neutrino dominated accretion flow (NDAF)9 around a rotating
compact object which can be extended upto several thousand Schwarzschild radius.
From eqns. (2) and (6) we can obtain
=_4aJ!Viz 1.8x7/2Ms 1.2x 7/ 2
Bo ---;==, Lose rv km = M, (7)
/]2& aH aH
for the Kerr geometry, where 152 = 2r2 + a 2 - x 2 - y2 - z2. The detailed calculation
and discussion are presented elsewhere. 1o Here we choose the mass of the compact
object M = Ms Mev, radius and height of the disk orbit where oscillation takes place
respectively r rv 15 = x M and z = H M, and we assure B rv Bo. Any oscillation
at the inner edge of NDAF is expected to be influenced by gravity what affects
the accretion dynamics and outflow. From eqn. (7), Lose varies from a few factors
to several hundreds of Schwarzschild radii at x ::; 10 for a fast spinning compact
object.
Supernova is thought to be the astrophysical site of the r-process nucleosynthesis.
During supernova, neutron capture processes for radioactive elements take place in
presence of abnormally large neutron flux. However, how does the large neutron
flux arise is still an open question. There are two related reactions:

(8)
If De is over abundant than V e , then, from eqn. (8), neutron production is expected to
be more than proton production into the system. Therefore, the possible conversion
of Ve to De due to the gravity induced oscillation explains the overabundance of
neutron.

References
1. M. Gasperini, Phys. Rev. D 38, 2635 (1988).
2. R. B. Mann, & U. Sarkar, Phys. Rev. Lett. 76, 865 (1996).
3. D. V. Ahluwalia, & C. Burgard, Gen. Rel. Grav. 28, 1161 (1996); Phys. Rev. D57,
4724 (1998).
4. S. R. Coleman, & S. L. Glashow, Phys. Lett. B405, 249 (1999).
5. J. Schwinger, Particles, Sources and Fields, (Addison-Wesley, 1970).
6. J. F. Nieves, & P. B. Pal, Phys.Rev. D58, 096005 (1998).
7. P. Singh, & B. Mukhopadhyay, Mod. Phys. Lett. A18, 779 (2003); B. Mukhopadhyay,
Mod. Phys. Lett. A20, 2145 (2005); U. Debnath, B. Mukhopadhyay, & N. Dadhich,
Mod. Phys. Lett. A2l, 399 (2006).
8. G. Barenboim, J. F. Beacom, L. Borissov, & B. Kayser, Phys. Lett. B537, 227 (2002).
9. R. Narayan, T. Piran, & P. Kumar, Astmphys. J. 557, 949 (2001); K. Kohri, R.
Narayan, & T. Piran, Astrophys. J. 629,341 (2005).
10. B. Mukhopadhyay, Class. Quantum Grav. 24, 1433 (2007).
 HOW GRAVITY CAN DISTINGUISH BETWEEN
DIRAC AND MAJORANA NEUTRINOS

DINESH SINGH, NADER MOBED

Department of Physics, University of Regina, Regina, Saskatchewan, S4S OA2, Canada
[email protected], nader. [email protected]

GIORGIO PAPINI
Department of Physics, University of Regina, Regina, Saskatchewan, S4S OA2, Canada,
Prairie Particle Physics Institute, Regina, Saskatchewan, S4S OA2, Canada, and
International Institute for Advanced Scientific Studies, 89019 Vietri suI Mare (SA), Italy
[email protected]

We show Dirac and Majorana neutrino wave packets propagating in a Lense- Thirring
background can be identified via a spin-gravity interaction described by a gravita-
tional phase. From a perturbation Hamiltonian due to spin-gravity coupling and time-
independent perturbation theory, we note that the matrix elements for Dirac and Ma-
jorana neutrinos fundamentally differ and lead to observationally relevant corrections to
the neutrino oscillation length for a two-flavour system, as shown for SN1987 A.
The discovery of neutrino flavour oscillations in vacuum, suggesting the existence
of neutrino rest masses, indicates the first concrete evidence for physics beyond the
Standard Model. However, it is still an open question whether neutrinos are Dirac or
Majorana particles. Since neutrinos are subject to space-time curvature effects, grav-
ity provides an opportunity to generate interactions that may differentiate between
Dirac and Majorana particles, such as through a helicity flip due to spin-gravity
coupling. This is what we show1 by considering neutrinos as wave packets which
propagate in a Lense-Thirring (LT) space-time background, leading to differences
in the predicted gravitational corrections to their respective oscillation lengths.
From the covariant Dirac equation in -2 signature and the vierbein formalism,1,2
the Dirac Hamiltonian Ho to first order in M / r <R: 1 and MflR2 / r2 <R: 1 for
the gravitational source in terms of the LT metric 3 leads to the energy eigenvalue
t
E6±) ::::; V(f'tko)2 + m 2 - 2~[ (nko) + M~F (LZ ± ~), where m is the neutrino
mass, p = nko is the momentum eigenvalue, L Z is the orbital angular momentum
in the z-direction, and the upper (lower) sign refers to positive (negative) helicity.
In addition, for a weak field described by hf.lv = gf.lV - Tlf.l U <R: 1, there exists a gauge
invariant gravitational phase 4 ,5 <I>G which, according to the phase transformation
1j;(x) --> exp (i<I>G/n) 1/J(x) , results in a new Dirac Hamiltonian H = Ho + HiJ>G'
where

(1)

is treated as a perturbation of Ho that recovers all the expected no niner-

tial/ gravitational effects involving spin-1/2 particles. 6
We assume a wave function composed as a linear superposition of plane waves 1,2

11/J) = 1
(27f)3/2 J d 3 k~(k) IU(k)), (2)

1028
 1029

where IU(k) is a normalized solution of the free-particle Dirac equation, and ~(k)
is a Gaussian function in momentum space, of width o-p and centroid ko, with
ko = Ikol. The matrix element (1jJIH<I>cl1jJ) in spherical co-ordinates follows from
integration over all phase space within the vacuum region. From the Weyl repre-
sentation 7 for the gamma matrices, the Dirac four-spinor is

(3)
where IVL(R) is its left- (right)-handed chiral projection, while the Majorana four-
spinor 8 is
IW1(k)Maj. == e-ik,xlvL) + eik-xlvf),
IU(k)Ma j . = (4)
{ IW (k)Ma j . == e-ik-xl ) - eik,xlvIJ,
2 vR
and IWf(2)Maj. = j
±IW1(2)Ma . due to charge conjugation. 7 ,s By invoking the
Rayleigh plane wave expansion,9 we find that the orthonormality conditions trun-
cate the series, with a virtually exact expression for (1/JIH<I>c 11jJ) expressed as the
sum of spin-diagonal terms and spin-flip terms proportional to the Pauli spin ma-
trices a J• With the neutrino's spin quantization axis along its direction of prop-
agation, the helicity transition element 1,2 is (±Ial=t=) = [cos cos cp ± i sin cp] +e x
e
[cos sin cp =t= i cos cp] fj - sin ez,
where I±) are the two-component spinors which
define positive (negative) helicity for the neutrino.
By expanding (1/JIH<I>c 11jJ) with respect to in == m/(/j,ko) « 1, we have

(1jJIH<I>c 11jJ)Dirae = (lik o) [~ [Co + C1 m + C2 in 2 ]

+ MrlR2.
-r- [
2- sm e Do + D1 in + D2 in 2]] , (5)

for the Dirac case, while the Majorana counterpart is

(1jJ1(2) IH<I>c 11jJ1(2)Maj . = (1jJIH<I>c l'tP) Dirae

± (Ii k o ) sin e sin cp [ ~ (±lal=t=)Y [Cay + C1y in + C2y rn 2 ]

+ MrlR2
-7-,2- (±Ial=t=) i; [ DOi; + Dli; in + D 2 i; in 2]] , (6)

where Cj, Cjy, D j , and Dji; are dimensionless functions of ko, R, r, and q == ko/o- p.
The extra interaction terms in (6), with an overall factor of sin e sin cp, correspond
to the y-component of the neutrino beam, due to an extra a Y interaction following
from the Majorana neutrino's self-conjugation condition. 7 This anisotropy can be
removed by averaging over a complete cycle 1 in cp when evaluating the perturbation
energy corrections to the oscillation length, A spin-flip term still contributes to (6)
in the limit as n ----> 0, while no such term survives for (5).
From (5) and (6), the gravitational corrections in the neutrino oscillation length 7
are Lose. = 21f / ( E;:} - E:t;l) , where m2 > m1. We use the Brillouin-Wigner (BW)
 1030

method 10 of time-independent perturbation theory to eventually obtain 1

E~) - E~) = (hk o) [F1 (m2 - md +(F2 +~) (m.~ - mn] , (7)

where F1 and F2 are also dimensionless functions of ko, R, r, and q, and dependent
on a complicated combination of terms coupled to !v1/T and !v1DR 2 l,.2. Using data
from SN1987 A, 11 we present F1 below 1 as a function of q for varying orientations of
the neutrino beam angle e, where the neutrino wave packet has a mean momentum
of hko = 1 MeV. (Fig. 1) The F j are non-zero for the range of 10- 4 ;S q ;S 10 1 and
agrees well with reasonable choices of O"p for neutrinos produced inside a neutron
star, when compared to a mean free path analysis 2 from known stellar data.

1 ~OxIO'­

g ..\oxl(f
4

"@
! ~()xl(f'
20xlO'
2
'it ~",",,,' t- 10x10"

]()' 10"

(I (dimensionless) q (dimPllSionlf'ss)

(a) Fpirac (SN1987A) (b) Ft

1aj
(SN1987 A)

Fig. 1. Fl as a function of q due to the SN1987 A gravitational sourcc, for varying neutrino beam
angle e. Bcsides having opposite sign, Fpirac is two orders of magnitude larger than F~aj ..

References
1. D. Singh, N. Mobed, and G. Papini, Phys. Rev. Lett. 97,041101 (2006); gr-qc/0611016.
2. D. Singh, N. Mobed, and G. Papini, Phys. Lett. A351, 373 (2006).
3. J. Lense and H. Thirring, Z. Phys. 19, 156 (1918); (English translation: B. Mashhoon,
F.W. Hehl, and D.S. Theiss, Gen. Rei. Gmv. 16,711 (1984)).
4. Y.Q. Cai and G. Papini, Phys. Rev. Lett. 66, 1259 (1991); 68, 3811 (1992).
5. D. Singh and G. Papini, Nuovo Cimento Bll5, 223 (2000).
6. F.W. Hehl and W.-T. Ni, Phys. Rev. D42, 2045 (1990).
7. M. Fukugita and T. Yanagida, Physics of Neutrinos and Applications to Astrophysics,
(Springer-Verlag Press, 2003).
8. LJ.R. Aitchison and A.J.G. Hey, Gauge Theories in Particle Physics - Volume II:
QCD and the Electroweak Theory, 3rd Edition (rOp Publishing, 2004).
9. G.B. Artken and H.J. Weber, Mathematical Methods for Physicists, 4th Edition, (Aca-
demic Press, 1995).
10. L.E. Ballentine, Quantum Mechanics, (Prentice-Hall, 1990)
11. K. Soida et al., Class. Quantum Gmv. 20, S645 (2003).
Cosmic Rays
This page intentionally left blank
 ANISOTROPIES OF ULTRA-HIGH ENERGY COSMIC RAYS

PASQUALE D. SERPIC01,2
1 Particle Astrophysics Center,. Fermi National Accelerator Laboratory.
Batavia, IL 60510-0500 USA
2 Max-Planck-Institut fur Physik (Wemer-Heisenberg-Institut),
Fahringer Ring 6, 80805 Munich, Germany
[email protected]

We summarize the types of anisotropy expected in the arrival directions of ultra-high

cosmic rays (UHECRs) of extragalactic origin in several energy regimes. We also briefly
discuss the information on UHECRs encoded in these patterns which may be extracted
by forthcoming experiments.

Keywords: ultra-high energy cosmic rays; cosmology

1. Introduction

The origin of cosmic rays is still partially unknown, despite the fact that one century
has passed since the discovery of their extraterrestrial nature by Hess. The main
problem resides in the fact that they are charged particles: the directional infor-
mation is lost by repeated scattering in interstellar magnetic fields, thus preventing
the direct identification of the sources. Above rv 10 18 - 10 19 eV (ultra-high energy
cosmic rays, UHECRs) the rigidity of cosmic rays of galactic origin is high enough
that the deflection in the galactic magnetic field (GMF) should not completely wash
out correlations between their arrival directions and the galactic plane. The lack of
any correlation down to the percent level and the difficulty to find suitable galactic
candidates for acceleration up to 2-3 X 10 20 e V suggest an extragalactic origin for
UHECRs. This hypothesis immediately raises the possibility that UHECRs may be
messengers from deep space, and thus potential cosmic tracers of structures. Vice
versa, one may exploit the present knowledge of the universe to infer some informa-
tion on UHECR properties. In the following we summarize what can be learnt on
UHECRs from the expected anisotropies in their arrival directions. For the sake of
simplicity, we shalllirnit ourselves to "astrophysical" scenarios for UHECR acceler-
ation, although interesting and often distinctive directional signatures arise also in
exotic models for UHECR production. 1

2. UHECR anisotropy at large scales

The anisotropies expected for extragalactic cosmic rays can be classified into:

i) Small scale clustering.

ii) Anisotropies on medium-large scales.
iii) Proper motion dipole (Compton-Getting effect).
iv) Rigidity effect due to the Gl\lF.

i) At energies high enough that deflections in extragalactic and galactic magnetic

fields are sufficiently small, point sources may reveal themselves as small-scale clus-
1033
1034

ters of UHECR arrival directions. This requires additionally a rather low density of
UHECR sources so that the probability to observe several events of at least a subset
of especially bright sources is large enough. We shall not review here the numerous
studies that have been performed on this scenario especially after the AGA8A claim
of a statistically significant clustering of events;2 unfortunately, other experiments
with comparable or larger statistics have not yet confirmed this claim. 3 ,4
ii) Moving to lower energies, the energy-loss horizon of UHECRs and thereby
the number of visible sources increases. Also, the number of potential accelerators
should increase. Finally, deflections in magnetic fields become more important. As a
result, the identification of single sources is challenging if not impossible. However,
if UHECR sources trace to some degree the inhomogeneous distribution of matter
revealed by the observed large scale structure (L88) surveys within a few hundreds
Mpc, some anisotropies on medium-large scales should be detectable. In 5 it was
evaluated the expected anisotropy in the UHECR arrival distribution starting from
the IRA8 P8Cz astronomical catalogue of nearby galaxies, taking into account the
main selection effects in the catalogue as well as UHECR propagation effects. The
conclusion was that to confirm this scenario one needs about 300 - 400 events at
E ?: 4 - 5 X 10 19 eV, which should be attained by the Auger experiment within
a decade or so. Yet, by combining the 0(100) events at E ?: 4 - 5 X 10 19 eV
already collected by the previous generation of instruments, the authors 0[6 found
some evidence of a broad maximum of the cumulative two-point autocorrelation
function of UHECR arrival directions around 25 degrees. The authors suggested
that, given the energy dependence of the signal and its angular scale, it might be
interpreted as a first signature of the large-scale structure of UHECR sources and
of intervening magnetic fields. Recently, this claim was analyzed on the basis of sky
maps derived from the P8Cz catalogue. 7 It was found that the above mentioned
result is indeed inconsistent with a purely isotropic sky, but consistent within 2 (J
with the expectations for UHECRs tracing the L88. A more-than-linear bias with
overdensity improves the agreement. The still low statistics and the role of the
magnetic field deflections may explain why no significant C'f"Oss-correlation between
data and L88 overdensities is found.
iii) At even lower energies, also the L88 structure of sources disappears, both
because the inhomogeneities in the source distribution will be averaged out due
to the increased energy-loss horizon of UHECRs and because of deflections in the
extragalactic magnetic fields. Thus, if the Earth were in the cosmological rest frame
the CR sky would appear isotropic. The observation of the cosmic microwave back-
ground (CMB) dipole clearly shows that this is not the case, and a dipole anisotropy
of 0.6% in the cosmic ray intensity is expected if the CR flux is dominated by sources
at cosmological distance. The shift of the dipole as function of energy provides in-
formation about the mean charge of CRs and the GMF. A 3 (J detection of this
effect requires around 10 6 events in the considered energy range and is thus chal-
lenging but not impossible with present detectors, at least at energies below 10 19 eV.
A similar effect also allows one to constrain the fraction of the diffuse gamma-ray
 1035

background emitted by sources at cosmological distance, with promising detection

possibilities for the GLAST satellite. 8
iv) Finally, if the extragalactic flux is still the dominant component at sufficiently
low energy, the GMF may introduce blind regions on the external sky, which trans-
late into observable anisotropies for an Earth-based observer, even if the UHECR
flux is isotropic at the boundary of the Milky Way. Although the details depend
on the G MF model, anisotropies of this kind should be expected in scenarios in-
voking a dominant extragalactic proton component already at E ~ 4 X 10 17 eVor
extragalactic iron nuclei at E < 10 19 eV. 9

3. Conclusions
Anisotropies are an important tool to distinguish between different origin and pri-
mary models for the UHECRs, nicely complementing the information on the energy
spectrum and chemical composition. We have briefly summarized several signa-
tures which one expects to show up in the pattern of UHECR arrival directions
at different energies. Unfortunately, the numerous unknowns make it impossible at
present to determine precisely the exact energy ranges where the above-mentioned
regimes are realized in nature, although theoretical expectations put this window at
10 18 - 10 20 eV. Experiments like Auger and Telescope Array have realistic chances
to detect these features, which would ultimately allow astronomers to acquire new
and "ultra-energetic" tools to explore the cosmos.

Acknowledgments
I wish to thank Profs. H. M. Kleinert and R. Schlickeiser for their kind invitation
to the 11th Marcel Grossman Meeting in Berlin. It is a pleasure to thank A. Cuoco,
R. D' Abrusco, M. Kachelriei3, G. Longo, G. Miele, and M. Teshima for all discus-
sions and fruitful collaboration on the topics summarized here. At Fermilab, this
work was supported by the US Department of Energy and by NASA grant NAG5-
10842. I acknowledge support from the Max-Planck-Institut fiir Physik of Munich
in the initial stage of this work.

References
1. M. Kachelrief3, Comptes Rendus Physique 5, 441 (2004).
2. M. Takeda et al. [AGASA collaboration]' Astrophys. J. 522 225 (1999); See also
M. Takeda et al., Proc. 27th ICRC, Hamburg 2001.
3. R. U. Abbasi et al. [The HiRes Collaboration], Astrophys. J. 623, 164 (2005).
4. B. Revenu [Pierre Auger Collaboration], Proc. 29 th ICRC, Pune 2005.
5. A. Cuoco et al., JCAP 0601, 009 (2006).
6. M. Kachelrief3 and D. V. Semikoz, Astropart. Phys. 26, 10 (2006).
7. A. Cuoco, G. Miele and P. D. Serpico, Phys. Rev. D, in press. [astro-ph/0610374].
8. M. Kachelrief3 and P. D. Serpico, Phys. Lett. B 640, 225-229 (2006).
9. M. Kachelriei3, P. D. Serpico and M. Teshima, Astropart. Phys. 26/6, 378 (2007).
 RECENT PROGRESS IN DESCRIBING
COSMIC RAY TRANSPORT

ROBERT C. TAUTZ
Institut fur Theoretische Physik IV - Weltraum- und Astrophysik
Ruhr- Universitiit Bochum, 44780 Boch,um, Germany
rct@tp4·rub.de

The standard theory for the description of cosmic ray transport is the quasi-linear theory.
Whereas it was assumed for a long time that QLT is a reliable and useful tool to describe
transport of charged particles, recent test-particle simulations have demonstrated that for
certain parameter regimes a linear description is not appropriate. In this talk a rigorous
comparison between QLT and previous test-particle simulations will be presented. It
is demonstrated that QLT is invalid for parallel and perpendicular transport for the
most parameter regimes in magnetostatic as well as plasma wave turbulence. It is also
demonstrated that recent nonlinear theories like the NLGC theory are able to reproduce
at least some of the simulations.

Keywords: cosmic rays, turbulence, diffusion, transport

1. Introduction
As described in many previous papers, the transport of cosmic rays parallel and per-
pendicular to a large-scale magnetic field is an interesting and important problem. 1-4
Although this is regarded as one of the key problems of cosmic ray astrophysics, no
all-encompassing theory has been developed yet. As a standard tool, most of the
time the quasi-linear theory2 (QLT) has been used to calculate diffusion coefficients
and mean-free paths. In this theory, the true particle trajectories are replaced by
the unperturbed orbits in a homogeneous magnetic field, which makes inaccuracy
inevitable because of the perturbation theory character inherent in the QLT. How-
ever, it was expected for a long time that a linear descri ptiOll of particle transport
would be correct-at least qualitatively. Another problem arises because of the un-
certain input into transport theories. There is still no consensus about the correct
turbulence model and, therefore, many different models (e. g., magnetostatic, dy-
namical, or plasma wave turbulence with, e. g., slab, 2D, or isotropic geometry) have
been proposed up to now. 3 ,5

2. Isotropic turbulence
2.1. Quasi-linear results
Recently, it has been ShOWI1'5,6 that quasi-linear theory is invalid for isotropic tur-
bulence geometry. For magnetostatic turbulence, this invalidity is absolute because,
unlike in simulations,7 the calculations of the parallel and perpendicular mean-free
paths yield infinity. If one considers isotropic Alfvcn or isotropic fast magnetosonic
plasma wave turbulence, the mean-free paths are finite, but the comparison to sirn-
ulations 8 shows deviations by a (maximum) factor of 10 6.

1036
 1037

2.2. The non-linear guiding center theory

The non-linear guiding center (NLGC) theory9 is based upon the assumption that
the particles' guiding centers follow magnetic field lines even in a turbulent magnetic
field.
The resulting fraction of perpendicular to parallel mean-free path as a function
of the dimensionless rigidity R = RL/1o for 5B = Bo and as a function of the
fraction of turbulent to background magnetic field, (5B/Bo)2, are shown in Figs. 1

NLGC perpendicular mean free path

- • - NLGC isotroPicJ
10°. Sim. isotropic
1
. . .• Sim. composite

----.--------- -- .. - - ..... - --- .

•.. ..• .•
.

Fig. 1. The fraction A-L/ All as a function of the normalized Larmor radius R = RL/lo for 8E = Eo
for isotropic magnetostatic turbulence. Analytic results 5 are compared to simulations. 7

NLGC perpendicular mean free path

- · - NLGC isotroPiC, I
-1 Sim. isotropic
. . • ..
10 1
•.. Sim. composite

,
..•
••
•
•

Fig. 2. The fraction A-L/AII as a function of the fraction of turbulent to background magnetic
field for particles with rigidity R = 0.1.
1038

and 2, respectively, comparison to simulation data 7 is given. The analytic results 5

are overall too high, by an average (but variable) factor of 3 and 5.4, respectively.

3. Conclusion
Because of the well-known 90° problem, in isotropic turbulence QLT provides us
with infinitely large (magnetostatic turbulence) and dramatically deviating (plasma
wave turbulence) parallel and perpendicular mean-free paths, in disagreement with
simulations. To describe parallel transport in isotropic turbulence, a (nonlinear)
theory like the recently proposed second order QLTlO must be applied. For per-
pendicular transport, the recently proposed NLGC th eory 9 is able to reproduce the
simulations semi-quantitatively (to within about a factor 5 but with the correct
trend).

References
1. E. N. Parker, Planet. Space Sci. 13, 9 (1965)
2. J. R. Jokipii Astmphys. J. 146, 480 (1966)
3. J. W. Bieber, W. H. Matthaeus, C. W. Smith, W. Wanner, M.-B. Kallenrode, and G.
Wibberenz, Astmphys. J. 420, 294 (1994)
4. A. Shalchi, and R. Schlickeiser, Astmphys. J. 626, L97 (2005)
5. R. C. Tautz, A. Shalchi, and R. Schlickeiser, J. Phys. G 32, 809 (2006)
6. R. C. Tautz, A. Shalchi, and R. Schlickeiser, J. Phys. G 32, 1045 (2006)
7. J. Giacalone and J. Jokipii, Astmphys. J. 520, 204 (1999)
8. G. Michalek, Astmn. Astmphys., 376, 667 (2001)
9. W. H. Matthaeus, G. Qin, J. W. Bieber, and G. P. Zank, Astrophys. J. 590, L53
(2003)
10. A. Shalchi, Phys. Plasmas 12, 052905 (2005)
 PROPAGATION OF ULTRA-HIGH ENERGY COSMIC RAYS:
TOWARDS A NEW ASTRONOMY

ALVISE MATTEI
LAPTH - Laboratoire d'Annecy-le- Vieux de Physique Thorique
9, chemin de Bellevue, BP 110, 74941 Annecy-le- Vieux, France
[email protected]·fr

The study of propagation of Ultra High Energy Cosmic Rays (UHECR) is a key step in
order to unveil the secret of their origin. Up to now it was considered only the influence
of the galactic and the extragalactic magnetic fields. In this article we focus our analysis
on the influence of the magnetic field of the galaxies standing between possible UHECR
sources and us.

1. Introduction

The Ultra-High Energy Cosmic Rays (UHECR) puzzle is at stake since the discov-
ery, in 1962 by J. Linsley,1 of the first event with energy beyond 10 20 eV. After four
decades and several experiments, the main problems are still unsolved. Where are
the UHECR sources? Is there a GZK cutoff? Which is the UHECR composition?
During last 10 years, the AGASA and HiRes experiments were discussing the
cutoff in the cosmic ray spectrum at E ~ 5 X 10 19 eV. This cutoff is expected from
the interaction between the high energy particle and the cosmic background radia-
tion. In 2003 AGASA 2 claimed a 40" evidence of cutoff absence, while HiRes data 3
were in agreement with possible cutoff. Recently AGASA group has revised the
results, converging towards a weaker cutoff evidence. The new experiment Auger,
which combines the ground detector (AGASA) with the air fluorescence (HiRes)
techniques, still does not clarify this point, but indicates a spectrum flux 2 times
lower than in previous experiments. 4 This means that we shall wait a little longer,
maybe three years, to have strong results on cutoff.
For the composition, the best result was recently obtained by Auger experiment,
fixing the limit for photon component to < 18%.5 The discrimination between heavy
nuclei and protons is not possible at this time, because the simulations have large
incertitude on first impact models and consequently on MonteCarlo propagations
of extensive air showers.
The UHECR arrival direction are known with incertitude of ~ 10. At the ex-
tremely high energy, the UHECR propagation is expected to pass from diffusive to
ballistic, i. e. almost rectilinear. 6 A 10 20 e V proton or nucleus cannot travel for more
than 50-100 Mpc. This limits the UHECR sources research to a bubble of few tens
of Mpc radius. Some correlations were found between UHECR arrival directions
and the blazar class of AGN. 7
The mystery of sources is strictly connected with the knowledge of cosmic mag-
netic fields. The angular deviation for an UHECR is ~ 0.5 0 y'dMpcAMpcZBnG/ E 20 ,
where d i the source distance, A is the magnetic field mean coherence length, Z
the charge and E 20 the energy of UHECR expressed in units of 10 20 eV, BnG the

1039
1040

mean magnetic field in nanoGauss. The time delay between a UHECR and a photon
emitted at the same time is T c:::' e;cd rv 10 3 .-;- 10 4 yr.8 For these reasons, UHECR
can lose their correlation with sources not only spatially, but also temporally, if the
magnetic field is strong and regular. Then the source problem would shift to how
these fields are built.
However it is possible to reconstruct the trajectories of charged particles if we
are able to map the relevant cosmic magnetic fields. First of all, we have to know the
Galactic Magnetic Field (GMF). We used the results from a collection of measures 9
which corroborate previous theoretical models. 1o The general picture of GMF is a
spiral structure in the Galactic plane, a cylindric structure close to the plane, and a
simple dipolar structure in the Galactic halo. Then, with the help of a Fortran code,
the deviation of back-propagated protons and nuclei was calculated for each arrival
direction in the 47f sky. Protons are deviated from 1 to 10 degrees, while Fe nuclei
follow curly trajectories, suggesting that in the UHECR-heavy nuclei scenario the
problem of sources can be reduced to Galactic production by highly magnetized
pulsars. Outside the Milky Way, the local universe is well described by the catalog
HyperLEDA, which is a collection of galaxy catalogs. l l We extract a sample of
60130 galaxies, all inside a 120 Mpc radius from us. To model the magnetic field, we
apply a simple dipolar field to each galaxy. The strength is similar to that of Milky
·Way. To test the effects of these dipolar fields, we studied the back-propagation of
protons inside the Virgo Cluster. We launch 40000 trajectories inside a cone of 1°
radius centered on Virgo. The mean deviation is rv 1°, and increases to > 1.5° if we
take a frozen halo field, i. e. a lower value of 1 nG up to the external halo radius
(rv 100 kpc). This way of reconstructing galactic fields is conservative.
Once we established the importance of magnetic fields from other galaxies, we
studied the trajectories of real UHECR. We used AGASA data, with particular
attention to the triplet events, which are likely to come from the same source. We
injected 40000 trajectories for each event with the same technique used for the Virgo
Cluster test. The galaxies crossed by a trajectory were stored in a file. If the same
galaxy appeared in at least one trajectory for event, it was detected as a favorable
source. In this way, we recognize two interesting possible sources. NGC 3998 is an
elliptic galaxy at 25 Mpc, hosting an AGN type LINER. This would correspond, in
the unified scenario, to an AGN type 1 (oblique jets). AGN are likely to accelerate
particles to extremely high energies and are viable UHECR sources. The other
candidate, NGC 3992, is a well known spiral galaxy (M109) at 22 !VIpc, historically
present in literature because of a supernova type Ia explosion in 1956. 12 This is very
interesting in light of the recently discovered supernova-gamma ray burst (GRB)
connection. During the GRB explosion, particles can be accelerated up to UHE.
Unfortunately, the temporal distance between triplet events is not compatible with
a production at the same time in that galaxy. This work was published.? Later on,
a fourth event from HiRes was added, and the previous sources were confirmed.
Thus, in this scenario, NGC 3998 could be a long lasting source responsible for all
the four events clustered. In figure 1 we reported a slice of trajectories at 30 Mpc
 1041

with the position of possible sources. It is clear from the figure that some trajectories
suffer deviations of several degrees. All the trajectories are corrected for Milky Way
contribution to the deviation.
The results obtained are model dependent. The amount of uncertainty on GMF
direction and intensity (rv 10%) as well as the naive expression of galactic field for
other galaxies do not permit to claim that NGC 3992 is the only possible source.
However, we defined a precise strategy to solve the problem ofUHECR origin: model
the GMF with precision, reconstruct the local universe through a galaxy catalog,
then apply a magnetic field to local universe. The knowledge of these quantities can
improve in the next future, thanks to the starting SKA and LOFAR experiments.
The increase in data sample from AUGER, with eventual clustering, could give
further answers to the source quest.

References
1. J. Linsley, L. Scarsi and B. Rossi, Suppl. J. Phys. Soc. Japan 17, p. 91 (1962).
2. M. Takeda et al., Astmparticle Physics 19, 447 (2003).
3. R. U. Abbasi et al., Phys. Lett. B619, 271 (2005), astro-ph/050l317.
4. The Pierre Auger Collaboration (July 2005), arXiv:astro-ph/0507150.
5. The Pierre Auger Collaboration (June 2006), astro-ph/0606619.
6. R. Aloisio and V. Berezinsky, ApJ 612, 900 (2004), astro-ph/0403095.
7. D. S. Gorbunov et al., ApJ 577, L93 (2002), astro-ph/0204360.
S. E. Waxman and J. Miralda-Escude, ApJ 472, LS9 (1996) astro-ph/9607059.
9. R. Beck et al., Ann. Rev. Astmn. Astmphys. 34, 155 (1996).
10. T. Stanev, ApJ 479,290 (1997), astro-ph/96070S6.
11. G. Paturel et al., ABA 412, 45 (20m).
12. F. Zwicky and M. Karpowicz, Astron. J. 69, 759 (1964).
This page intentionally left blank
Astrophysics of Neutron
Stars and Black Holes:
Observation
This page intentionally left blank
 EXTRAGALACTIC X-RAY JETS

DIANA M. WORRALL
Dept. of Physics, University of Bristol, Tyndall Avenue, Bristol, BSB 1 TL, U.K.
d. [email protected]

1. Framework

There are important reasons for understanding the composition, speed and dynam-
ical processes of extragalactic radio jets. They are manifestations of supermassive
black holes that are undergoing significant accretion and releasing vast energy. Some
of the energy is radiated close to the nucleus, and some is transported to scales as
large as Mpc. Jets propagate through X-ray-emitting atmospheres which, as the
main reservoirs of baryonic mass in the Universe, provide the primary probes of
large-scale structure. The energy lost from jets that goes into gas heating can help
regulate the structures of the atmospheres, and understanding the processes and
measuring the amount of energy transfer is vital to a full understanding of the
structure and evolution of the Universe. By the mid 1990s the study of radio jets
reached something of a hiatus, but work has been revitalized as a result of Chandra
resolving jet X-ray emission in many tens of sources.
In what follows I will use the terms FRI and FRII to distinguish the two mor-
phological and luminosity classes of radio galaxies, although this is not a perfect
separator of physical properties for objects close to the boundary. FRls are broadly
of low total radio power and nearby, so there is good spatial resolution on their jets.
Broadening jets feed diffuse lobes, and it is believed that the jets are slowed by mass
loading. Indeed, where this has been modelled, there is excellent agreement between
jet pressure and that of the X-ray-emitting thermal atmosphere. l FRIIs are of high
power and are more distant. Under unified schemes FRIIs equate to quasars if one
of their twin jets points towards the observer. It is understood that FRII jets retain
high bulk-relativistic speeds out to larger distances than FRI jets. FRII jets are
low-efficiency radiators but efficient energy conveyors. They terminate at hotspots,
driving bow shocks which push old jet material back into edge-brightened cocoons.
Chandra has found that the X-ray cores of FRII radio galaxies generally show
higher absorption and a greater preponderance of Fe emission lines than FRIs,
supporting the idea that the FRI/FRII dichotomy is intrinsic, related to the cen-
tral structures and accretion, rather than resulting from jet/gas interactions as
the jet propagates from the nucleus. Recent work 2 takes into account the lumi-
nosity that could be hidden by a central torus in FRls in concluding that these
sources are undergoing radiatively inefficient accretion. FRIIs are consistent with
near-Eddington accretion onto a standard, geometrically thin and optically thick ac-
cretion disk (Shakura-Sunyaev or similar), and X-ray results for a sample containing
both quasars and radio galaxies matched in isotropic power support quasar /radio-

1045
1046

galaxy unification where all sources have both jet-related and accretion-related com-
ponents, but the mix in what is seen varies with source orientation due to anisotropic
emission and absorption. 3

2. Physical properties of jets

Magnetic field and particles dominate the internal pressure of radio structures, and
so it is crucial they be measured. Synchrotron radiation probes only the product of
their energy densities. However, if the electrons also Compton up-scatter a known
photon field into measured X-rays, the field strength and energy in relativistic elec-
trons can be inferred. X-ray data test the assumption that a radio source is at
minimum total energy, where B = B rne , the only handle on the magnetic field in
the absence of inverse Compton X-ray emission, and can probe jet composition
through pressure comparisons with the external X-ray-emitting gas. Measurements
with Chandra and XMM-Newton for the lobes and hotspots of several tens of FRII
sources 4- 7 typically find 0.3Bme < B < 1.3Brne on the assumption that only rela-
tivistic electrons and positrons provide particle energy. In several cases it has been
argued that such a lepton-dominated plasma has sufficient pressure to oppose the
pressure of the external X-ray-emitting gas 5 ,6 although, since FRII jets must carry
energy over millions of light years from core to lobe, it would be surprising if non-
radiating particles are not involved. Because the radiating particles are light, it is
possible that they react quickly to the fields, and thus that electron/magnetic-field
minimum-energy holds even in the presence of relativistic protons. For FRIs, new
data support findings from Einstein and ROSAT 8 ,9 that FRIs must have lobe pres-
sures beyond their minimum values to prevent collapse. This does not necessarily
imply that the electron/magnetic-field minimum energy is violated, however, since
extra lobe pressure is present if the particles do not precisely track the magnetic
fields (filling factors less than unity) or if protons keep the lobes inflated.
Two Chandra surveys have achieved a '" 60% X-ray detection rate of extended
jets in core-dominated quasars.lO,n Since quasar jets are beamed towards us, a likely
X-ray mechanism is inverse Compton scattering of CMB photons by low-energy
electrons that see boosted CMB and emit beamed X-rays in the observer's frame,
suggesting faster flow-speeds than inferred from earlier radio statistical analyses. 12
Alternative models are possible, and some quasars have broad-band morphologies
inconsistent with simple CMB-scattering models. Deep Chandra observations of
sources chosen to settle current controversies are under way. The CMB-scattering
model certainly doesn't work for FRII radio galaxies, where the X-rays are un-
beamed to the observer. The relative faintness of these jets is a problem, and indeed
if the CMB-scattering model holds for quasars there is no a priori reason why we
should expect to detect these jets at all in X-rays, apart from their bright terminal
hotspots. In a breakthrough, synchrotron X-ray emission has been detected in jet
knots of a few FRII radio galaxies. 13 -- 15 This means that particle acceleration, to
electron energies of Te V and more, must be proceeding along the length of the X-ray
 1047

jet even in fast FRII jets, as is already known for FRIs. Observations and modeling
aim to elucidate the mechanisms of particle acceleration. 13 ,16,17

3. Radio galaxies and their X-ray-emitting gaseous environments

It is a surprise that Chandra and XMM-Newton have not observed many instances
of X-ray brightening around radio lobes of FRII sources, indicative of the external
medium being heated and compressed as it passes through the bow shock of the
expanding lobe. The best example of such a structure is around a small young inner
radio lobe of Cen A where there is little evidence for current jet fuelling. 18 The
shell's kinetic energy exceeds both its thermal energy and the thermal energy of the
ambient X-ray-emitting gas. As the shell dissipates it will have a major effect on
the gas, providing distributed heating. There are several examples where FRIs in
nearby clusters are driving relatively weak shocks,19,20 but strong shocks appear to
be rare locally, although so too are FRIIs in rich environments.
A new discovery is the first example of a heated combined galaxy atmosphere,
resulting from a galaxy-galaxy merger, that is having a dynamical effect on the
radio plasma, rather than vice versa. 21 Here it is inferred that the radio-galaxy
lobes have been swept apart for'" 108 yrs by the pressure-driven expansion of the
X-ray bright gas, and that the gas may be re-energizing the relativistic particles in
the radio lobes through compression, rather than the radio source heating the gas
through shocks. This may provide a clue to the mechanism of particle acceleration
that is inferred to be present in cluster-embedded radio relic sources.

References
1. R.A. Laing, A.H. Bridle 2002, MNRAS, 336, 1161
2. D.A. Evans, D.M. Worrall, M.J. Hardcastle et al. 2006, ApJ, 642, 96
3. E. Belsole, D.M. Worrall, M.J. Hardcastle 2006, MNRAS, 366, 339
4. J.H. Croston, M.J. Hardcastle, D.E. Harris et al. 2005, ApJ, 626, 733
5. E. Belsole, D.M. Worrall, M.J. Hardcastle et al. 2004, MNRAS, 352, 924
6. J.H. Croston, M. Birkinshaw, M.J. Hardcastle, D.M. Worrall 2004, MNRAS, 353,879
7. M.J. Hardcastle, D.E. Harris, D.M. Worrall, M. Birkinshaw 2004, ApJ, 612, 729
8. R. Morganti, R. Fanti, 1.M. Gioia et al. 1988, A&A, 189, 11
9. D.M. Worrall, M. Birkinshaw 2000, ApJ, 530, 719
10. R.M. Sambruna, J.K. Gambill, L. Maraschi et al. 2004, ApJ, 608, 698
11. H.L. Marshall et al. 2005, ApJS, 156, 13
12. R.M. Sambruna, L. Maraschi, F. Tavecchio, et al. 2002, ApJ, 571, 206
13. D.M. Worrall, M. Birkinshaw 2005, MNRAS, 360, 926
14. R.P. Kraft, M.J. Hardcastle, D.M. Worrall, S.S. Murray 2005, ApJ, 622, 149
15. M.J. Hardcastle, 1. Sakelliou, D.M. Worrall 2005, MNRAS, 359, 1007
16. M.J. Hardcastle, D.M. Worrall, R.P. Kraft et al. 2003, ApJ, 593, 169
17. F. Dulwich, D.M. Worrall, M. Birkinshaw et al. 2007, MNRAS, 374, 1216
18. R.P. Kraft, S. Vazquez, W.R. Forman et al. 2003, ApJ, 592, 129
19. P.E.J. Nulsen, B.R. McNamara, M.W. Wise, L.P. David 2005, ApJ, 628, 629
20. W. Forman, P. Nulsen, S. Heinz et al. 2005, ApJ, 635, 894
21. D.M. Worrall, M. Birkinshaw, R.P. Kraft, M.J. Hardcastle 2007, ApJL, 658, in press
 INITIAL RESULTS FROM THE SUZAKU SATELLITE

TADAYASU DOTANI* and THE SUZAKU TEAM

Institute of Space and Astronautical Science,
Japan Aerospace Exploration Agency
3-1-1 Yoshinodai, Sagamihara, Kanagawa 229-8510, Japan
[email protected]

Suzaku is the 5th X-ray astronomy satellite of Japan developed under the Japan/US
collaboration. It was successfully launched on July 10, 2005. Suzaku is equipped with 4
sets of X-ray telescopes with the CCD cameras at the focal plane and the non-imaging
hard X-ray detector. Unique capabilities of Suzaku is a wide-band coverage (0.2-600 keY)
with low background and improved line spread function of the CCD below ~1 keY. High
lights of the initial results obtained with Suzaku are presented.

1. Introduction
The Suzaku satellite, formally known as Astro-E2, is the 5th X-ray astronomy satel-
lite of Japan developed under the Japan/US collaboration (figure 1).1 It is a retry
mission of Astro-E, which was lost during the launch in 2000 due to the failure of the
1st stage rocket motor. Thus, Suzaku succeeded the basic design of Astro-E includ-
ing the scientific payloads. Suzaku was successfully launched from the Uchinoura
Space Center with the M-V rocket on July 10, 2005. Its final orbit was selected to
be circular at an altitude of 570 km with an inclinat.ion of 32°.
In this paper, we review the scientific instruments on board Suzaku, and present
high lights of its initial results. Description of the payloads and the initial scientific
results is found in the special issue (vol. SP1, 2007) of Publ. Astron. Soc. Japan.

2. The Suzaku satellite

Suzaku was initially loaded with three scientific instruments; X-ray Imaging Spec-
trometers (XISs),2 X-Ray Spectrometer (XRS),3 both of which were located at the
foci of the X-Ray Telescopes (XRTs),4 and the Hard X-ray Detector (HXD).5 Con-
figuration of the instruments is shown in figure 2. In t.he initial stage of the mission,
XRS was lost due to the sudden loss of the liquid He. It is not operational now.
XISs consist of 4 sets of X-ray CCD cameras, each of which is located at the
focus of the XRT. The XRT utilizes a grazing incidence, thin-foil nested mirror
to achieve a large effective area with a moderate angular resolution (2 arcmin in
the half-power diameter). One of the XISs utilizes the back-illuminated (BI) CCD,
whereas the rest utilizes the front-illuminated (FI) CCD. The BI CCD has a superior
line-spread function at lower energies. XIS combined with XRT covers 0.2-12 ke V
(BI CCD) or 0.4-12 keY (FI CCD) with a field of view (fov) of 18' x 18'.

* Also for: Department of Physics, Tokyo Institute of Technology, and Department of Space and
Astronautical Science, the Graduate University for Advanced Studies.

1048
 1049

·----Sun shade

XRT-I
(4 units)

Star
tracker
'-...

scopes

XRS ,.

HXD
Base-
panel
units)

Fig. 1. Schematic view of the Suzaku satellite. The Fig. 2. Cross-section of the Suzaku
optical bench were extended and the solar paddles satellite is shown with the configu-
were deployed on orbit. ration of the main instruments.

HXD is a non-imaging, hard X-ray detector, which covers 10-600 keY with the
combinations of the Si PIN diodes and the GSO scintillators. The PIN and GSO
detectors are located inside the well-structure of the BGO active shield, which are
very effective in reducing the in-orbit background to an extremely low leve1. 6 The
fov of PIN is restricted to 34 arcmin (FWHM) with passive collimators, while that
of GSO is 4.5 degree defined by the active shield of BGO.

3. Initial results
In this section, we pick up several topics from the initial results from Suzaku, and
give a brief description of them.

Planetary nebula: Suzaku observed the brightest planetary nebula in X-ray

band, BD+3003639, on Sept. 20, 2005. The K lines from CVI, OVII and OVIII
were resolved for the first time with XIS.7 Using the XIS data, the C/O and
Ne/O abundance ratios were found to be larger than the solar values at least
by a factor of 30 and 5, respectively. This is the first determination of the
large abundance ratios in X-rays, which is much larger than those observed
in optical band. This result indicates that X-rays are mainly emitted by the
He shell-burning products.
Soft X-ray emission from geocorona: It is known that the charge exchange
process between the heavy ions in the solar wind and the neutrals in the
earth's geocorona account for a part of the soft X-ray background below
1 keY. This phenomenon was first reported during the ROSAT all sky
1050

survey as a long-term enhancement (LTE) of X-rays.8 During the Suzaku

observations of the north-ecliptic pole, CVI 4p to Is transition line was
detected in the LTE, which had comparable strength as the CVI 2p to Is
line. 9 This is a strong evidence of the charge exchange process, and the lines
are most likely originated from the solar-wind ions. This has a significant
impact on the observations of X-ray sources below 1 keY.
Cyclotron resonance features: Suzaku observed the transient X-ray pulsar
A0535+262 at the decay of the outburst. Clear cyclotron resonance feature
was detected at ",45 keY in spite of its low luminosity, 4 x 10 35 erg/sec,
thanks to the high sensitivity of HXD.lO The resonance energy was essen-
tially same as those measured when the source was more luminous by two
orders of magnitude.
Black hole candidate: Suzaku observed the black hole candidate, 4U1630-
472,6 times in the decay of the outburst in February and March, 2006. The
source stayed in the typical high/soft state, and clear absorption lines were
detected from He-like and H-like iron ions.u The ionization degree and the
outward velocity of the line-forming plasma showed a gradual decrease with
the decay of the outburst.
Seyfert galaxies: Several bright Seyfert galaxies, ego MCG-5-23-16 12 and
MCG-6-30-15,13 were observed in the early phase of the Suzaku mission.
These sources were detected well above ",50 keY, which was essential to
separate the reflection component from the continuum. The reflection com-
ponent was found to be much less variable compared to the underlying
power-law emission, which is consistent with the prediction of the light-
bending model.

4. Conclusion
Suzaku operation is now stable, and it is conducting the A02 observations smoothly.
We expect that a lot of new findings will be brought by Suzaku.

References
1. K. Mitsuda et al., Pubi. Astron. Soc. Japan 59, 81 (2007).
2. K. Koyama et ai., Publ. Astron. Soc. Japan 59, 823 (2007).
3. R. L. Kelley et ai., Pubi. Astron. Soc. Japan 59, 877 (2007).
4. P. J. 8erlemitsos et ai., Publ. Astron. Soc. Japan 59, 89 (2007).
5. T. Takahashi et ai., Pubi. Astron. Soc. Japan 59, 835 (2007).
6. M. Kokubun et ai., Publ. Astron. Soc. Japan 59, 853 (2007).
7. M. Murashima et ai., Astrophys. J. 647, L131 (2006).
8. 8. L. 8nowden et ai., Astrophys. J. 424, 714 (1994).
9. R. Fujimoto et ai., Pubi. Astron. Soc. Japan 59, 8133 (2007).
10. Y. Terada et ai., Astrophys. J. 648, L139 (2006).
11. A. Kubota et ai., Pubi. Astron. Soc. Japan 59, 8185 (2007).
12. J. N. Reeves et ai., Pubi. Astron. Soc. Japan 59, 8301 (2007).
13. G. Miniutti et ai., Publ. Astron. Soc. Japan 59, 8315 (2007).
 SOFT GAMMA REPEATERS AND MAGNETARS

K. C. HURLEY
u. C. Berkeley Space Sciences Laboratory
Berkeley, CA 94720-7450, USA
[email protected]

The soft gamma repeaters (SGRs) are one manifestation of magnetars. There is now a
solid body of multi-wavelength observations of SGRs spanning several decades. There are
also a number of fundamental unanswered questions, which will require better theory,
more sensitive observations, and many years to answer.

Keywords: neutron stars; magnetars; soft gamma repeaters.

1. Soft Gamma Repeaters

The SGRs are sources of short C·v 100 ms), repeating bursts of soft 'I-radiation
«100 keY). Figure 1 shows a series of bursts from SGR1900+14. When an SGR is
active, it can go through periods where hundreds of bursts are emitted in a period
of minutes. Active periods occur at apparently random intervals; outside of these
periods, it is common for SGRs to emit no detectable bursts at all for years.
Four SGRs are known. Three are in our galaxy (SGR1806-20, 1900+14, 1627-
41), and one is in the direction of the Large Magellanic Cloud (SGR0525-66, the
source of the March 5, 1979 burst). The fact that three are in the Galactic plane,
while one is in a young SNR, indicates that all the SGRs are probably young objects
(perhaps < 10,000 years old).
The SGRs are quiescent X-ray sources. Although their luminosities are somewhat
variable, they are generally strong sources (rv 1O- 11 ergcm- 2 s- 1 ) which can easily
be detected in soft X-rays, and in two cases, to energies up to 100 keY and above. 2
While there is a connection between the X-ray luminosity and bursting activity, the
quiescent X-ray emission is always present, even when there are no bursts.
The SGRs have rotation periods in the 5-8 s range, which are increasing mono-
tonically, although sometimes irregularly, with time, at rates rv 1O- 1 0s S-l. The
X-ray luminosity (rv 2 x 10 35 erg S-l) is much greater than the spin-down energy
(rv 1033 erg S-l ), which leads to an estimate of the magnetic field if dipole radiation
is assumed to be the cause of the spindown, and the particle wind is negligible. 3
Also, under these conditions, the spindown age P /2P rv 1500y, which is consistent
with the idea that SGRs are young objects. SGR1627-41 may be an exception. Its
periodicity is either undetectable, or its X-radiation is time-variable.
The SGRs occasionally emit long duration, hard spectrum giant flares, which
produce the most intense cosmic gamma-ray fluxes ever measured at Earth. Three
have been observed so far. The first was the March 5, 1979 event from SGR0525-
66 4 - 6 The second was the August 27,1998 event from SGR1900+14 7- 9 The third was
the December 27, 2004 burst from SGR1806-20, the most intense of the three. 10-12
Giant flares occur perhaps every 30 years on a given SGR (no SGR has yet been

1051
1052

observed to emit two giant flares). These bursts are intense (up to rv 3 x 1046erg at
the source, or 1 ergcm- 2 at Earth), last rv 5 minutes, and have very hard energy
spectra extending to MeV energies, at least. Their time histories are modulated
with the neutron star periodicity. SGRs are not quiescent radio emitters, but giant
flares create transient radio nebulae,13,14 and even produce dramatic ionospheric
disturbances. 15 Figure 1 shows an example. Some extragalactic giant flares could in
fact be mistaken for short gamma-ray bursts.
A recent review of SGRs and Anomalous X-ray Pulsars has appeared. 1

2. Interpretation
Several interpretations of SGR phenomenology have been proposed. Here the mag-
netar model will be outlined, as it has been elaborated by Thompson and Dun-
can. 16 ,17
In some rare supernova explosions, a neutron star is born with a fast rotation
period (rv ms) and a dynamo is established which creates or amplifies a strong
magnetic field. Differential rotation and magnetic braking quickly slow the period
down to the observed 5--10 s range. Magnetic diffusion and dissipation create hot
spots on the neutron star surface, which cause the star to be a quiescent, periodic
X-ray source. The strong magnetic field stresses the iron surface of the neutron
star, to which it is anchored. The surface undergoes localized cracking, shaking
the field lines and creating Alfven waves, which accelerate electrons to rv 100 keV;
they radiate their energy in short bursts with energies around 10 40 - 10 41 erg (in
earthquake terms, this can be thought of as a magnitude 19.5 crustquake). These
are the most common SGR bursts.
Localized cracking cannot relieve all the stress which the magnetic field exerts
on the surface, and it continues to build for decades. The built-up stress eventually
ruptures the surface of the star profoundly (a magnitude 23.2 starquake), resulting
in a giant flare. Magnetic field lines annihilate, accelerating electrons and positrons,

100000

rJ)
on 10000
~
f-
Z
=>
0
u
I
1000
10°1

100
100 200 300 400
TIME,s

Fig. 1. Left: bursts from SGR1900+14 during a particularly active phase. Right: time history of
a giant flare from SGR1900+14.
 1053

and filling the magnetosphere with a hot pair plasma. The initial spike in the giant
flare is radiation from the entire magnetosphere (B > 10 14 G is required to contain
the pair plasma). The 5 minute long, periodic component of the flare comes from
hot spots on the surface of the neutron star.

3. Open questions
Some of the outstanding questions about SGRs are the following.
1. What are the distances of the Galactic magnetars?
2. What is the number-intensity relation for giant magnetar flares?
3. What is the SGR birth rate?
4. What kind of supernova produces an SGR?
The progress in this field has been relatively slow up to now. The first SGR was
discovered about 30 years ago. In another 30 years, some of these questions may be
answered.

Acknowledgments
The author is grateful for support under .JPL Contract 1268385.

References
1. Woods, P., and Thompson, C., in Compact Stellar X-Ray Sources (Cambridge Uni-
versity Press Cambridge 2006)
2. Gotz, D. et al., Astron. Astrophys. 445, 313 (2006)
3. Kouveliotou, C. et al., Nature 393, 235 (1998)
4. Mazets, E. et al., Nature 282, 587 (1979)
5. Cline, T. et al., ApJ 237, Ll (1980)
6. Evans, W. et al., ApJ 237, L7 (1980)
7. Hurley, K. et al., Nature 397, 41 (1999)
8. Feroci, M. et al., ApJ 515, L9 (1999)
9. Mazets, E. et al., Astron. Lett. 25(10), 635 (1999)
10. Hurley, K. et al., Nature 434, 1098 (2005)
11. Mereghetti, S. et al., Ap. J. 624, L105 (2005)
12. Palmer, D. et al., Nature 434, 1107 (2005)
13. Frail, D., Kulkarni, S., and Bloom, J., Nature 398, 127 (1999)
14. Gaensler, B. et al., Nature 434, 1104 (2005)
15. Inan, U. et al., C.R.L. 26(22), 3357 (1999)
16. Thompson, C., and Duncan, R., Mon. Not. R. Astron. Soc. 275, 255 (1995)
17. Thompson, C., and Duncan, R., Ap. J. 473, 322 (1996)
This page intentionally left blank
 Theoretical Models of
Observation from Black
Hole Candidates
This page intentionally left blank
 EPICYCLIC FREQUENCIES AND RESONANT PHENOMENA
NEAR BLACK HOLES: THE CURRENT STATUS

ALlKRAM N. ALlEV
Feza Cursey Institute, P.K. 6 (}engelkoy, 34684 Istanbul, Turkey
[email protected]

We briefly discuss the theory of epicyclic motion near rotating black holes and its relation
to the origin of high frequency Quasi-Periodic Oscillations (QPOs) seen in many cases
of accreting black hole binaries. We also point out some new frequency ratios predicted
by the theory.
In the framework of general relativity the successive theory of epicyclic motion
around rotating black holes was first developed in Refs. 1 ,2(see also a review paper3).
As is known, the rotating black holes are uniquely described by the famous Kerr
solution
2
2.6.
ds =-~ (dt-asm ed¢) +~
. 2 2 ( dr
~+de2
2
) sin e 2
+~[adt-(r2+a2)d¢J ' (1)

where the metric functions

.6.=r2+a2 -2Mr,
The equation of motion for a test particle of mass j1. and charge e has the form
d 2x l" dx G dx f3 e dx V
- - +fl"af3 -
ds 2
- = -FI"
ds ds j1.
-
v ds .
(2)

From the symmetry properties of the metric (1) it follows that a cyclic motion of
the test particle is possible in the equatorial plane e = 1f /2. For the cyclic motion
we have zl"(s) = uos {I, 0, 0, wo}, where Wo is the angular velocity of the motion.
To describe the epicyclic motion of the test particle we employ a perturbative
expansion about the cyclic orbits, i.e. we assume that xl"(s) = zl"(s)+e(s). Substi-
tuting this into equation (2) and adding also the external force fl" = fl" vuv(UO)-l,
we arrive at the equation
d
2
e" iJ. d~a a
dt2 + la dt + ~ o a
UiJ. _ _
- j1.Uo
fiJ. NiJ.
+
e_ (c
<.."
d~)
dt ' a = 1,2 == r,e (3)

where NiJ. stands for the non-linear terms, the quantities I~ and oaUiJ. must be
taken on a cyclic orbit r = ro, e = 1f /2 and

1iJ. = 2fiJ. u f3 (U O)-l _ _ e_ FiJ.

a af3 j1.Uo a ,
UiJ. = ~2 [1iJ.Ua(UO)-l _ _ e_ FiJ. ua(uO)-l] (4)
a j1.Uo a
Performing an appropriate integration in (3) and restricting ourselves to the first

1057
1058

order terms in ~ we obtain the set of equations

d~A
ill + '"11A re
~ = /-L UD
Jf A
dt, A = 0,3 == t, ¢ (5)

(6)

d2~e 2C e _ _ e_ fe
dt2 + we" - /-L UD .
(7)

We see that in the absence of the external force the epicyclic motion is governed
by the two decoupled oscillatory-type equations. In other words, it amounts to free
radial and vertical oscillations with the associated frequencies

Wr = (
Our
or - I~ It
) 1/2
,
_(oU08e )
We -
1/2
(8)

The explicit expressions for these frequencies were first found in Ref.1 in the most
general case of black holes with an electric charge or with an external uniform
magnetic field. Further, the epicyclic frequencies were used to explore the maser
effect near black holes 2 as well as the effects of periastron precession and the Lense-
Thirring drag in the Schwarzschild and Kerr fields threaded by a cosmic string.4
For a "pure" Kerr field the frequencies are 1,2

2
W2 = W2 ( 6M
1- - - - -3a ± 8awK ) (9)
r D r r2 '

where Wo = ± WK / (1 ± aw K ), WK = viM /
r3 is the Kepler frequency and the two
signs refer to direct and retrograte motions. Later, in 1987 our expression for the
frequency of radial oscillations was re-derived in Ref.,5 while the formula for the
frequency of vertical oscillations re-appeared 6 only in 1990.
In ReU we have also pointed out that when taking into account the non-linear
terms on the right-hand-side of Eq. (3) a coupling between the oscillations occurs.
If the frequencies Wr and We are in the rational relation kr Wr = ke We, where kr
and ke are integers, the resonances must take place. We have plotted the positions
of low-order (k =1 kr 1+ 1ke I) resonances: k = 3,4,5, or (2: 1, 3: 1, 3: 2). In
recent developments, this theory has been extensively used to explain the frequency
ratios and the origin of high frequency Quasi-Periodic Oscillations (QPOs) observed
in many cases of accreting black hole or neutron star binaries (see papers 7 ,8 and
references therein). In some cases, even our plots for the positions of the non-linear
g
resonances were exactly reproduced (see for instance, Ref. ).
Our model also admits the appearance of the forced resonances under a per-
 1059

turbing force. Solving Eqs.(5)-(7) for Fourier transforms we obtain

CA ( ) _ __
i A Cr ( ) _ _ e_ fA(w, rn)
<" w,rn - 11 <" w,rn 0 2 ' (10)
Wm fLU Wm

Cr ( ) = _e_ .f r(w , rn) - ;:;:;:-

i 'VI
IA
fA(w , rn)
<" w,rn 0 2 2 ' (11 )
fLU Wr - Wm

Wm =W - rnwo . (12)

We see that at the frequencies w = rnwo, w = 0;

= rnwo ± wr , w = = Oi
rnwo ± We the system exhibits the resonant behaviour. 2 The similar type of the
resonant frequencies was discussed by Kato 10 within a model for kHz QPOs reso-
nantly induced in warped disks of X-ray binaries. In the latest developments, we
have performed a complete numerical analysis of the frequency ratios pertaining to
the orbits where the frequency of the radial oscillations approaches its maximum
value. Some results for direct orbits and for masses,...., 101\18 are given in the Table.

aiM rrnaxl M va(Hz) vr(Hz) ve(Hz) velvr valvr valve

0.00 8.00 141.42 70.71 141.42 2.00 2.00 1.00
0.10 7.58 152.60 75.73 151.17 2.00 2.02 1.Ol
0.20 7.1.5 165.68 81.52 162.37 1.99 2.03 1.02
0.30 6.70 181.26 88.27 175.45 1.99 2.05 1.03
0.40 6.24 199.97 96.29 190.74 1.98 2.08 1.05
0.50 5.76 223.29 105.99 209.23 1.97 2.11 1.07
0.60 5.26 252.90 118.03 231.78 1.96 2.14 1.09
0.70 4.71 292.71 133.53 260.59 1.95 2.19 1.12
0.80 4.11 349.83 154.57 298.90 1.93 2.26 1.17
0.90 3.42 442.93 185.95 353.95 1.90 2.38 1.25
0.99 2.45 662.11 243.45 447.52 1.84 2.72 1.48

Acknowledgements
The author thanks the Scientific and Technological Research Council of Turkey
(TlJBiTAK) for partial support under the Project 105T437.

References
1. A. N. Aliev and D. V. Gal'tsov, Gen. Relat. Gravit. 13, 899 (1981).
2. A. N. Aliev, D. V. Gal'tsov and V. 1. Petukhov, Astr. Space Sci. 124, 137 (1986).
3. A. N. Aliev and D. V. Gal'tsov, SOy. Phys. Usp. 32(1), 75 (1989).
4. A. N. Aliev and D. V. Galt'sov, SOy. Astron. Lett. 14(1), 48 (1988).
5. A. Okazaki, S. Kato and J. Fukue, Pub!. Astron. Soc. Japan 39, 457 (1987).
6. S. Kato , Pub!. Astron. Soc. Japan 42, 99 (1990).
7. A. Merloni, M. Vietri, L. Stella and D. Bini, Mon. Not. R. Ast. Soc. 304,155 (1999).
8. G. Torok, M. Abramowicz, W. Kluzniak and Z. Stuchlik, Ast.& Ap. 436, 1 (2005).
9. M. A. Abramowicz and W. Kluzniak, astro-ph/0312396
10. S. Kato, astro-ph/0506595.
 HUMPY LNRF-VELOCITY PROFILES IN ACCRETION DISCS
ORBITING RAPIDLY ROTATING KERR BLACK HOLES*

ZDENEK STUCHLIK, PETR SLANY and GABRIEL TOROK

Institute of Physics,
Silesian University in Opava,
Bezrucovo nam. 13, 74601 Opava, Czech Republic
zdenek. [email protected], petro [email protected], [email protected]

Humpy LNRF-velocity profiles were found for both Keplerian thin discs and marginally
stable thick discs orbiting almost extreme Kerr black holes, being located just above the
innermost stable circular orbit. In thick marginally stable discs, the global character of
this phenomenon is given by topology changes of equivelocity (von Zeipel) surfaces, as the
standard cylindrical von Zeipel surfaces are transformed into toroidal surfaces existing
around the circle corresponding to the local minimum of the equatorial humpy LNRF-
velocity profile. These phenomena could indicate an instability in the disc triggering
so-called "humpy" oscillations, the frequency of which is estimated by the maximum
positive gradient of the humpy velocity profile in terms of the proper radial distance. The
humpy oscillations could force epicyclic oscillations, if appropriate resonance conditions
are satisfied.

1. Introduction
In the Kerr spacetimes, the locally non-rotating frames (LNRF) corresponding to
zero angular momentum observers playa crucial role, as they give the most natural
framework for describing the accretion phenomena. 1 For matter orbiting a Kerr
black hole (of dimensionless spin a) with a 4-velocity cut, 0, 0, U'P), angular velocity
n = U'P jut and specific angular momentum f! = -U'PjUt , the orbital velocity is
given by the azimuthal component of its 3-velocity in the LNRF

U f1 ('1') A· e
V('P) = ~ = ~(n-w) (1)
UVeSt ) ~JX '
where ~ = ,2 - 2, + a 2 , ~ = ,2 + a 2 cos 2 e, A = (,2 + a 2 )2 - ~a2 sin 2 e and
w = 2a,jA is the angular velocity of the LNRF.
For the thin, Keplerian discs, there is

(2)

The orbital velocity

(,2 + a 2 )2 _ a2~ - 2a,(,3/2 + a)
V ('I' )(,. a) - -'--------'-------==------'- (3)
K ,- ,2(,3/2 + a)JX
then reveals a "humpy profile" in the field of Kerr black holes with a > ac(K) =
0.9953, see papers of Aschenbach. 2 ,3

*This research has been done as a part of the Research Project MSM 4781305903 and partly
supported by the Silesian university in Opava grants IGS SU 34/2006 (P.S.) and 28/2006 (T.G.).

1060
 1061

General formula relating the angular velocity and the specific angular momen-
tum,
[2 = _ Rgtt + gt<p (4)
Rgt'P + g'P'P '
implies for marginally stable thick discs, i.e. discs with uniform distribution of the
specific angular momentum R = R(r, e) = const, the orbital velocity in the form
VC'P) = A(D. - a 2 sin2 e) + 4a Zr2 sin2 eR
(5)
f=const ~v'K(A - 2aRr) sine .
Stuchlik et a1. 4 shown that it has a "humpy profile" in the field of Kerr black holes
with a > ac(T) ~ 0.99979.

2. Von Zeipel surfaces

The global character of the "humpy LNRF-velocity profile" phenomenon can be
manifested by the so-called von Zeipel surfaces, as the rotating fluid configurations
could be characterized by the von Zeipel radius 4
R
R == V('P)' (6)

It is related to the well known gyration radius 5 R by

R = (1- wR)R, (7)
where R = VC'P) /([2 - w).
In the case of marginally stable tori the von Zeipel surfaces R(r, e) = const
coincide with the equivelocity surfaces V('P) = const and usually have cylindrical
shape. In addition to the cylindrical open surface crossing itself under the inner edge
of the torus and existing for all values of a, for almost extreme Kerr black holes
with a > 0.99979 a second self-crossing surface and toroidal surfaces occur inside the
torus just above the innermost stable circular geodesic. 4 The toroidal surfaces are
centered around the circle corresponding to the minimum of the equatorial LNRF-
velocity profile (Fig. 1). The change of topology of the equivelocity surfaces from
the cylindrical to the toroidal one possibly indicate some instabilities of the tori
in both radial and vertical direction that, further, could excite so-called "humpy
oscillations" of the disc.

3. Humpy frequency
Frequency of "humpy oscillations" is estimated by the maximal positive rate of
change of the orbital LNRF velocity in terms of the proper radial distance. The
locally defined frequency has to be related to static distant observers. Therefore the
humpy frequency is given by 6

vh(rh) = V-(gtt + 2wgt'P + w g'P'P) a:;:-

8V('P) I
2 (8)
max
1062

3 0.6

2 0.4

0.2

""0en 0 ""'"
u 0
U
0
H H

-1 -0.2

-2
-0.4

-3
-0.6
7 1.4 1.6 1.8 2
r sine r sin Ii
(a) (b)
Fig. 1. Von Zeipel (equivelocity) surfaces in f!. const accretion discs (shaded region). Dashed
curve corresponds to the ergosphere, cross denotes the center of the disc. (a) just open (cylindrical)
surfaces exist in the disc. (b) toroidal surfaces exist in the region near the center of the disc.

where df y'9r;"dr = V'E,/ ~dr and all expressions are taken at the "humpy radius"
rh defined by the maximum rate of the velocity change, i.e., by the condition

OV(<P) I oV(<p) I (9)

of rh max

4. Conclusion
The humpy oscillations could excite epicyclic oscillations in both radial and verti-
cal direction, if proper resonant conditions between their frequencies are satisfied. 6
We can expect both the forced and parametric nOll-linear resonances giving rise to
a plenty of quasiperiodic oscillations which could be observed. We assume applica-
bility of this extended model of resonant phenomena in sources as GRS 1915+105.
Preliminary results were presented at the "VI Microquasar Workshop: Microquasars
and Beyond" being held in Como. 7

References
1. J. M. Bardeen, W. H. Press and S. A. Teukolsky, Astrophys. J. 178, 347 (1972).
2. B. Aschenbach, Astronomy and Astrophysics 425, 1075 (2004).
3. B. Aschenbach (2006), astro-ph/0603193.
4. Z. Stuchlfk, P. Slany, G. Torok and M. A. Abramowicz, Phys. Rev. D 71, 024037
(2005).
5. M. A. Abramowicz, J. Miller and Z. Stuchlik, Phys. Rev. D 47, 1440 (1993).
6. Z. Stuchlik, P. Slany and G. Torok (2006), astro-ph/0612439 (accepted by Astronomy
and Astrophysics).
7. Z. Stuchlik, P. Slany and G. Torok, in VI Microquasar Workshop: Microq'uasars and
Beyond, ed. T. Belloni (Proceedings of Science, 2006), http://pos . sissa. it.
 STANDING SHOCKS IN PSEUDO-KERR GEOMETRY

SOUMEN MONDAL
S.N. Bose National Centre for Basic Sciences, JD Block,
Salt Lake, Sector-III, Kolkata 700098, India
[email protected]

SANDIP K. CHAKRABARTI
S.N. Bose National Centre for Basic Sciences, JD Block,
Salt Lake, Sector-III, Kolkata 700098, India
and
Centre for Space Physics, Chalantika 43,
Garia Station Rd., Kolkata 700084
[email protected]

We present here the detailed properties of the standing, propagating or oscillating shock
waves which are very common to study the accretion and winds using our suitably
modified effective potential around compact objects. We specifically study the entire
parameter space spanned by the specific energy and angular momentum and divide it
depending on the nature of the solution topology. In future, these solutions can be used
as guide-lines to test numerical simulation provided the Kerr parameter remains within
-1::;a::;O.8.

1. Introduction
The two-component advective flows which include shock waves appear to be very
relevant to explain the oservational results from stellar and massive black holes (see,
e.g., Chakrabarti & Titarchuk, 1995; Chakrabarti et al. 1996; Ebisawa, Titarchuk
& Chakrabarti, 1996, Smith, Heindl & Swank, 2001; Smith et al. 2002). We wish
to follow up these studies around a Kerr black hole with the help of a pseudo-
Kerr potential (Chakrabarti & Mondal, 2006; Mondal & Chakrabarti, 2006). This
potential mimics the Kerr geometry very faithfully as long as the Kerr parameter
lies in the range -1 < a < 0.8 as far as the particle dynamics is concerned. For
fluid dynamical purposes, the pressure of the flow is affected by the red-shift factor.
In a Newtonian description, this boils down to changing the polytropic index of
the infallng gas in a manner so that the flow has similar properties as in Kerr
geometry. We carry out the sonic point analysis and explore the locations of the
sonic points and standing shock waves as functions of the conserved flow parameters,
namely the specific energy and the specific angular momentum As originally done
in Chakrabarti (1989) in the context of Schwarzschild black hole. We also carry out
our analysis of the solution topologies in Kerr geometry.

2. The Pseudo-Kerr geometry

We choose G = c = M = 1 and distances, specific angular momenta and times
are measured in units of GMjc 2 , GMjc and GMjc3 respectively. The potential we

1063
1064

propose is:

(1)

where, g is measured from the axis of rotation of the black hole and ro = 0.04 +
0.97 a + 0.085a 2 . The red-shift factor a (Eq. 1) and the dragging term, w have been
borrowed from general relativity,

a = ~JE and (2)

where, p2 = r2 + a 2cos 2g, ~2 = (r2 + a 2)2 - a 26.sin 2g and 6. = '('2 + a 2 - 2r.

l"¢ = -uq, = liJ>ejj, is the conserved specific angular momentum for the particle and
I is approximately the conserved specific angular momentum for fluid I = -ucp/Ut.

09 ' _L_ j L ________________

26 28 3 3.2 3.4 36 05 25

Fig. lea-b): The sub-division of the parameter space spanned by E and I in terms of
whether or not multiple sonic points or shocks may form. The solid curves represent pa-
rameter space for pseudo-Kerr potential and dotted curves for GTR values. The regions
marked 0, I, SA, SW, NSA, NSW contain flow parameters (E,l) such that the flows will
have only the outer sonic point, only the inner sonic point, shocks in accretion, shocks in
winds, 'no-shocks' in accretion and 'no-shocks' in winds respectively. In regions J* and 0*
only the inner a topology and outer a type topology are possible. In region N, no solution
is possible. The subscripts ms and mb refer to marginally bound and marginally stable
values respectively for a = 0.5. (b) The solid curves are for the accretion flow parameters
taken from the SA in Figs. lea). Only the shocks drawn in solid curves in (b) are stable.

3. Discussion
We presented a modified gravitational potential to study fluid dynamics around
a Kerr black hole, which, for all practical purposes, could be used as easily as a
Newtonian potential. This is found to reproduce characteristics of fluid dynamics
vcry accurately as long as the Kerr parameter -1 < a < 0.8. The errors in this
range, if any are found to be less than five percent or so.
 1065

120
H

100

80
x,
60

40

20

1 002 1 004 1 006 1 008 1 01 1 0 l;~

(5

Fig. 2(a-b): Variation of the (a-b) inner shock X s 2 and outer shock Xs:l (along y-axis) as
a function of energy (along the x-axis) and angular momentum I (marked on each curve).
The solid curves are for the accretion flow parameters taken from the SA in Figs. l(a),
while the dotted curves are for the wind flow parameters taken from the region SW in
those Figures. (a) a = 0 and (b) a = 0.5. Only the shocks drawn with dotted curves in
(a-b) and the shocks drawn in solid curves in (a-b) are stable.

_~~I_~---,,-
10- 5 2x\0 5 3x10-5 4xl0 5
.M

Fig. 3(a-b): Entropy accretion rate is plotted against the specific energy as the sonic
point is varied for a given specific angular momentum I: (a) a = 0.5. The sections AB and
CD are drawn for 'inner' and 'outer' saddle type sonic points while the section BC is drawn
for '0' type sonic points. The double arrowed curves represent the direction in which the
sonic point is increasing. Examples of energy preserving shock transitions are shown by
horizontal arrows 8152 in accretion and Wi W2 in winds respectively. (b) Classification of
the parameter space in the M vs. 1 plane a = 0.5. Subdivision of the parameter space
presented in terms of whether the shocks can form or not.

References
1. S. K. Chakrabarti, 1989, Astrophys. J., 347, 365
2. S. K. Chakrabarti, S. MandaI, 2006, MNRAS, 369, 976
3. S. K. Chakrabarti & L. G. Titarchuk, 1995, ApJ 455, 623
4. S. K. Chakrabarti, L. G. Titarchuk, D. Kazanas & K. Ebisawa 1996, A & A, Supp!.
Ser., 120, 163
5. K. Ebisawa, L.G. Titarchuk, & S. K. Chakrabarti, 1996, PASJ 48(1), 59
6. S. Mondal, S. K. Chakrabarti, 2006, MNRAS, 371,1418
7. D. M. Smith, W. A. Heindl, C. B. l\larkwardt, J. H. Swank, 2001, ApJ, 554, L41
8. D. M. Smith, W. A. Heindl, J. H. Swank, 2002, ApJ 569, 362
 PROPERTIES OF ACCRETION SHOCK WAVES IN VISCOUS
FLOWS WITH COOLING EFFECTS

SANTABRATA DASl and SANDIP K. CHAKRABARTI2,3

1 ARCSEC, Sejong University, 98 Gunja-Dong, Gwangjin-Gu, Seoul 143-747, South Korea
[email protected]

2S. N. Bose National centre for Basic Sciences, JD-Block, Sector III,
Salt Lake, Kolkata, 700 098, India. E-mail:[email protected]
3 Centre for Space Physics, Chalantika 43, Garia Station Rd., Kolkata 700084, India

We study the properties of the shock waves for a viscous accretion flow having low
angular momentum in presence of synchrotron cooling. We present all possible accretion
solutions in terms of flow parameters. We identify the region of the parameter space for
steady and oscillating shocks and show the effect of various energy dissipation processes
on it. We discuss the role of the shock waves while explaining the observations from black
hole candidates.

Keywords: accretion, accretion disc - black hole physics-shock waves.

1. Introduction
The shock induced accretion flow is currently one of the most promising self-
consistent hydrodynamical accretion disk model, since it explains the spectral states
of black holes as well as quasi-periodic oscillations (QPOs) most satisfactorily e,4
and references therein). Since then, various groups of workers extensively studied
the properties of the shock waves (e.g.,1,2) in different astrophysical contexts. A
fully self-consistent solutions of isothermal, viscous transonic flows was obtained
by Chakrabarti 3 considering sub-Keplerian flow at the inner part of the disc. In a
rotating accretion flow, centrifugal force acts as a barrier 4 ,.5,7 that triggers the flow
to undergo shock transition at a location where the Rankine-Hugoniot shock condi-
tions (hereafter RHCs) are satisfied. However, so far, the study of viscous flows in
presence of Synchrotron cooling has not been done. In the present paper, we concen-
trate on the study of a stationary, axisymmetric, viscous accretion solutions around
a Schwarzschild black hole in presence of Synchrotron cooling. The space-time ge-
ometry around a non-rotating black hole is approximated by the pseudo-Newtonian
potentia1. 10 We study all the relevant dynamical flow variables in terms of the in-
flow parameters. We identify the solution topologies which are essential for shock
formation. The effects of viscosity and cooling are expected to be different while
deciding the dynamical structure of the accretion flow since cooling reduces the flow
energy while viscosity not only tends to heat the flow but transports the angular
momentum from inner edge to the outer edge. We find that shock waves, standing
or oscillating, can form even at a very high dissipation limit. The hot and dense
post shock flow is the natural site of the hot radiation in the accretion disc and
is believed to be a powerful tool in understanding the spectral properties of black
holes,6 QPOs of the hard X-rays 9,1l and the formation of the accretion-powered
relativistic bipolar outflowsjjets. 8 In this paper, we discuss these issues.

1066
 1067

2. Results and Discussions

A set of classical shock solutions are presented in Fig. 1a in terms of viscosity (0;)
and/or cooling parameter ((3). We inject matter sub-sonically at the outer edge of
the disc Xinj = 145 with local energy f inj = 3.3663 X 10- 3 and angular momentum
Ainj = 1. 725. Solid vertical line represents the shock location for non-dissipative
(0; = 0 and (3 = 0) flow. When 0; # 0, the shock front moves inward depicted by the
dashed vertical line. In an accretion flow, viscosity transports angular momentum
outwards causing the reduction of the centrifugal pressure and at the same time,
the viscous heating increases flow energy. Since the shock moves forward as A(x) get
reduced, we can conclude that the centrifugal force is the primary cause for shock
formation. When (3 # 0, the shock location again proceeds towards the horizon.
In the hot and dense post-shock flow cooling is more efficient which reduces the
post-shock pressure causing the shock to move forward further. When both 0; # 0
and (3 # 0, shock location is predicted at Xs = 18.62 and is indicated by vertical
dot-dashed line. Here, the shock front is shifted significantly due to the combined
effects of viscosity and cooling.
In Fig. 1b, we separate the regions of the parameter space spanned by the specific
energy (fin) and specific angular momentum (Ain) at the inner sonic point (Xci)
according to the flow topologies for (3 0.00787 in inviscid limit. Solid boundary
represents the region for closed topologies passing through Xci. At the inset, all
possible solutions [Mach number vs. log(x)] with parameters chosen from different
region of the parameter space are presented. The box S represents the shock solution.
The box OS shows an accretion solution having oscillating shock. 1<'01' higher cooling,
we obtain a new solution topology as shown in box elM. We draw this solution
with dotted curve as it is obtained for higher cooling parameter. The solution I
passes directly through the ,Dei before entering into the black hole. The solution 0
represents a flow which passes through the outer sonic point (Xeo) only. The box

(a) (b) (c)

Fig.1. (a) Variation of Mach number with the logarithmic radial distance. Dissipation parameters
and corresponding shock locations are marked. (b) Sub-division of parameter space in the [in - Ain
plane according to the nature of solutions. (c) Variation of the parameter space for standing shocks
for different dissipation parameters (marked).
1068

CI shows a closed flow solution passing through Xci. This type of solution does
not extend to the outer edge to join smoothly with any Keplerian disc and flow is
expected to be unstable.
In Fig. Ic, we compair solutions with different dissipation parameters. Solid
boundary is the parameter space for standing shock for a = 0 and j3 = O. For
a i= 0, the effective region of the parameter space for standing shock separated
by dot-dashed curve shrinks and moves towards the higher [in and the lower Ain
regime. 5 When j3 i= 0, the parameter space is reduced and shifts to the lower energy
sides. 7 For flows with a i= 0 and j3 i= 0, the parameter space for the shock settles
down at an intermediate region (shaded part). This shows that the viscosity and
the synchrotron cooling act in opposite directions in deciding the parameter space.

3. Conclusion
We studied the properties of viscous accretion flow around a non-rotating BH in
presence of synchrotron cooling. We found that the flow can have shock waves even
when the viscosity and synchrotron cooling are high. The standing shocks form
closer to the BH when dissipation is increased and these shocks are centrifugal
pressure supported. We obtained the parameter space for the standing shocks for
various dissipation parameters and showed that the effective region of the parameter
space shrinks as dissipation rises. The parameter space for oscillating shock where
RHCs are not satisfied but the flow is likely to pass through a shock has been
identified. We also pointed out that the viscosity and the cooling have opposite
effects in deciding the parameter space for stationary shock waves. Moreover, since
the shocks form closer to the BH, QPO frequency increases with the enhancement
of accretion rate as observed in several BH candidates.
Acknowledgments: SD is thankful for financial support to KOSEF through ARC-
SEC, Korea.

References
1. Yang R., Kafatos M., 1995, A&A, 295, 238.
2. Fukumura K, Tsuruta S., 2004, ApJ, 611, 964.
3. Chakrabarti S. K, 1990, Theory of Transonic Astrophysical Flows. World Scientific
Publishing, Singapore.
4. Chakrabarti S. K, 1996, ApJ, 464, 664.
5. Chakrabarti S. K, Das S., 2004, MNRAS, 349, 649.
6. Chakrabarti S. K, Titarchuk L. G., 1995, ApJ, 455, 623.
7. Das S., 2007, MNRAS, 376 , 1659.
8. Das, S., Chattopadhyay I., 2007, New Astron. 12(6), 454.
9. Molteni, D., Sponholz, H., Chakrabarti, S.K 1996, ApJ, 457, 805
10. Paczynski B., Wiita P. J., 1980, A&A, 88, 23.
11. Ryu D., Chakrabarti S. K, Molteni D., 1997, ApJ, 474, 378.
 MODEL OF RADIATING ANNULI NEAR BLACK HOLES FOR
IRON Kcr. LINE PROFILE INTERPRETATIONS*

ALEXANDER F. ZAKHAROV
National Astronomical Observatories of Chinese Academy of Sciences, 20A Datun Road,
Chaoyang District, Beijing, 100012, China;
Institute of Theoretical and Experimental Physics, 117259, Moscow, Russia;
Bogoliubov Laboratory for Theoretical Physics, JINR, 141980 Dubna, Russia;
Center of Advanced Mathematics and Physics, National University of Science and Technology,
Rawalpindi, Pakistan
[email protected]

X-ray observations of microquasars and Seyfert galaxies reveal the broad emission
lines in their spectra, which can arise in the innermost parts of accretion disks. We
consider a radiating annulus model to simulate spectral line shapes. Recently Muller &
Camenzind (2004) presented results of their calculations and classified different types
of spectral line shapes and described their origin. We clarified their results and showed
that triangular spectral lines could arise for nearest annuli and high inclination angles.
\Ve discussed a possibility of appearance of narrow spectral line shapes as a result of
spiralling evolution of matter along quasi-circular orbits.
Comprehensive reviews summarize the detailed discussion of theoretical aspects
of possible scenarios for generation of broad iron lines in AGNs 2 (an influence of
microlensing on Fe Ka line shapes and spectra was discussed 3 ). Profiles of lines
emitted by AGNs and X-ray binary systems could have asymmetric double-peaked,
double horned or triangular shape according to classification. 1 Using a radiating
annulus model we checked the statements and clarified them for the case. We could
note here that in the framework of the model we do not use any assumptions about
an emissivity law, but we only assume that radiating region is a narrow circular ring
(annulus). Thus, below we do not use some specific model on surface emissivity of
accretion (we only assurne that the emitting region is narrow enough). But general
statements (which will be described below) could be generalized on a wide disk case
without any problem.
We used an ray shooting approach described earlier. 4 The approach was used in
particular to simulate spectral line shapes and it was based on results of qualitative
analysis. s Let us summarize results obtained using the simple modelY As it was
shown in the framework of the simple model the double peaked spectral line shape
arise almost for all parameters T and a except the case when radii are very small
T E (0.7,2) and inclination angles are in the band e
E [45°,90°] (for these parameters
the spectral line shape has triangular structure). The phenomenon could be easy
understood, since for this case the essential fraction of all photons emitted in the
opposite direction in respect to emitting segment of annulus is captured by a black
hole, therefore a red peak is strongly dumped. For other radii and angles spectral
line profiles have double peaked structure. Therefore, we clarify the statement 1

*This research has been partially supported by the National Natural Science Foundation of China
(NNSFC) (Grant # 10233050) and National Basic Research Program of China (2006CB806300).

1069
1070

that double peaked structure arises if radiation region is far enough. If we assume
that there is a weak dependence of emissivity function on radius, then a number of
photons characterizes relative intensity in the line (roughly speaking for r = 0.7 an
intensity (in counts) in about ten times lower then an intensity for r = 2) therefore
in observations for small radii one could detect only a narrow blue peak but another
part of spectra is non-distinguishable from a background.
One could note also that for fixed radius there is a strong monotone dependence
of intensity on inclination angle (maximal intensity corresponds to photon motion
near equatorial plane and only a small fraction of photons reach a distant observer
near the polar axis). That is a natural consequence of a photon boost due to a
circular motion of emitting fragment of annulus in the equatorial plane and an
influence of spin of a rotating black hole. In the framework of the simple model one
could understand that sometimes the Fe Ko line has only one narrow peak like in
observations of the Seyfert galaxy MCG-6-30-15 by the XMM-Newton satellite. If
radiating (or illuminating) region is a narrow annulus evolving along quasi-circular
orbits, then initially two peak structure of the spectral line profile transform in one
peaked (triangular) form. Moreover, an absolute intensity in the line is increased for
smaller radii since a significant fraction of emitted photons are captured by a black
hole during the evolution of emitting region toward to black hole in observations we
could detect only narrow blue peak and its height is essentially lower than its height
was before for larger radii. Another part of the triangular spectral line shape could
be non-distinguishable from a background. A relative low intensity for a triangular
spectral line shape could give a narrow single peak structure in observations. There
is a more description of the analysis. 6
As it was noted earlier, shadow detection is another possibility to check general
relativity in the strong gravitational field limit since there are non-negligible chances
to observe mirages (shadows) around the black hole at the Galactic Center and
in nearby AGNs in the radio-band (or in the nun-band) using Radioastron (or
Millimetron) facilities. Since a shadow size should be about 50 fLas for the black
hole in the Galactic Center and analyzing the shadow size and shape one could
evaluate the spin and charge of the black hole since fringe sizes for shortest wave
lengths are smaller than shadow diameters. 7

References
1. A. Muller & M. Camenzind, Astron. & Astrophys. 413, 861 (2004).
2. A.C. Fabian, K. Iwasawa, K.S. Reynolds, A.J. Young, Publ. Astron. Soc. Pacific 112,
1145-1161 (2000); V. Karas, Theoretical aspects of relativistic spectral features, astro-
ph/0609645.
3. L. Popovic, P. Jovanovic, E. Mediavilla et al. Astrophys. J. 637620 (2006).
4. A.F. Zakharov, Mon. Not. Roy. Astron. Soc. 269, 283 (1994); A.F Zakharov, On the
Hot Spot near a Kerr Black Hole, in Proc. of the 17th Texas Symposium on Relativistic
Astrophysics, eds. H. B6hringer, G.E. Morfill, .I.E. Trumper, (Ann. NY Academy of
Sciences, 1995), 759, 550; A.F. Zakharov, S.V. Repin, Astron. Rep. 43, 705 (1999).
A.F. Zakharov, S.V. Repin, The shape of the Ko-line as a possible indication of the
 1071

black hole existence, in Pmc. of the XXIV International Workshop on High Energy
Physics and Field Theory "Fundamental Problems of High Energy Physics and Field
Theory", ed. V.A. Petrov, (Institute for High Energy Physics, Protvino, Russia, 2001)
p. 99; A.F. Zakharov, S.V. Repin,Astmn. Rep. 46, 360 (2002); A.F. Zakharov, S.V. Re-
pin, The shape of the Ka line of iron as the evidence of the black hole existence, in
Pmc. of the Eleven Workshop on General Relativity and Gravitation in Japan, eds.
J. Koga, T. Nakamura, K. Maeda, K. Tomita (Waseda University, Tokyo, 2002), p. 68;
A.F. Zakharov, S.V. Repin, The shape of iron Ka line as the evidence for the black hole
existence in Seyfert Galaxies, 2002c, in Pmc. of the XXXVIIth Rencontres de Moriond
"The Gamma-ray Universe", eds. A. Goldwurm, D.N. Neumann, J. Tran Thanh Van
(The GIOI Publishers, 2002) p. 203; A.F. Zakharov, S.V. Repin, Astmn. fj Astrophys.
406, 7 (2003); A.F. Zakharov, Publ. of the Astron. Observatory of Belgrade, 76, 147
(2003); A.F. Zakharov, The iron Ka line as a tool for analysis of black hole param-
eters, in The Physics of Ionized Gases, eds. L. Hadzievski, T. Gvozdanov, N. Bibic,
(AlP Conf. Proc. 2004), 740, p. 398; A.F. Zakharov, Intern. J. Mod. Phys. A 20,
2321 (2005); A.F. Zakharov, Massive Black Holes: Theory vs. Observations, in Pmc.
of the Helmholtz International School and Workshop" Hot Points in Astrophysics and
Cosmology", eds. V.B. Belyaev, D. Blaschke, (JINR, Dubna, Russia, 2005), p. 332;
A.F. Zakharov, N.S. Kardashev, V.N. Lukash, S.V. Repin, Mon. Not. Roy. Astron.
Soc. 342, 1325 (2003).
5. A.F. Zakharov, Sov. Phys. - J. Exper. Theor. Phys. 64, 1 (1986); A.F. Zakharov, Sov.
Phys. - J. Exper. Theor. Phys. 68, 217 (1989).
6. A.F. Zakharov, S.V. Repin, Mem. S. A. It. Suppl. 7, 60 (2005); A.F. Zakharov,
S.V. Repin, New Astmn. 11, 405 (2006); A.F. Zakharov, Phys. Atom. Nucl., 70,
159 (2007).
7. A.F. Zakharov, A.A. Nucita, F. De Paolis, G. Ingrosso, New Astmnomy 10, 479
(2005); A.F. Zakharov, A.A. Nucita, F. De Paolis, G. Ingrosso, Retro gravitational
lensing for Sgr A* with Radioastron, in Pmc. of the 16th SIGRAV Conference on Gen-
eral Relativity and Gravitational Physics, eds. G. Vilasi, G. Esposito, G. Lambiase,
G. Marmo, G. Scarpetta, (AlP Conference Proceedings, 2005) 751, p. 227; A.F. Za-
kharov, A.A. Nucita, F. De Paolis, G. Ingrosso, Observational Features of Black Holes,
in Pmc. of the XXVII Workshop on the Fundamental Pmblems of High Energy and
Field Theory, ed. V.A. Petrov (Institute for High Energy Physics, Protvino, 2005)
p. 21; gr-qc/0507118; A.F. Zakharov, A.A. Nucita, F. De Paolis, G. Ingrosso, Mea-
suring parameters of supermassive black holes, in Pmc. of XXXXth Rencontres de
Moriond "Very High Energy Phenomena in the Universe", eds. J. Tran Thanh Van
and J. Dumarchez, (The GIOI Publishers, 2005) p. 223; A.F. Zakharov, A.A. Nucita,
F. De Paolis, G. Ingrosso, Shadows (Mirages) Around Black Holes and Retro Gravi-
tational Lensing, in Pmc. of the 22nd Texas Symposium on Relativistic Astmphysics
at Stanford University, SLAC-R-752, eds. P. Chen, E. Bloom, G. Madejski, V. Pet-
rosian, SLAC-R-752, eConf:C041213, http://www.slac.stanford.edu/econf/C041213,
paper 1226 (2005); A.F. Zakharov, F. De Paolis, G. Ingrosso, A.A. Nucita, Astron. fj
Astmphys. 442, 795 (2005); A.F. Zakharov, A.A. Nucita, F. De Paolis, G. Ingrosso,
Shadow Shapes around the Black Hole in the Galactic Centre, in Pmc. of" Dark M at-
ter in Astm- and Particle Physics" (DARK 2004), eds. H.V. Klapdor-Kleingrothaus
and D. Arnowitt, (Springer, Heidelberg, Germany, 2005), p. 77; A.F. Zakharov, F. De
Paolis, G. Ingrosso, A.A. Nucita, Measuring the black hole parameters from space,
in Gravity, Astmphysics, and Strings'05, Pmc. of the 3rd Advanced Workshop, eds.
P.P. Fiziev and M.D. Todorov, (St. Kliment Ohridski University Press, Sofia, 2006)
p.290.
 QPOS DUE TO CENTRIFUGALLY SUPPORTED SHOCKS
AROUND STELLAR-MASS AND SUPERMASSIVE BLACK HOLES

T. OKUDA
Hakodate Campus, Hokkaido University of Education,
Hakodate 040-0429, Japan
[email protected]

V. TERESI and D. MOLTENI*

Dipar-timento di Fisica e Tecnol09ie Relative, Universita di Palermo,
Viale delle Scienza, Palermo 90128, Italy
* [email protected]

Rotating adibatic accretion flows with adequate injection parameters around black holes
could form standing shock waves close to the black holes, after the flow passes through
the outer sonic point and can be virtually stopped by the centrifugal force. Taking
account of the cooling and heating of the gas and the radiation transport, we numerically
examine such shock waves in 2D accretion flows around a stellar mass (101\;10) and a
super massive (10 6 MO) black holes. The resultant 2D shocks are unstable with time and
the luminosities show the quasi-periodic oscillations (QPOs) with modulations of a factor
2 - 3 and with periods of a tenth seconds to several hours, depending on the black hole
masses. These shock oscillation models may not only explain the intermediate frequency
QPOs with 1-10 Hz observed in the black hole candidate GRS 1915+105 but also suggest
the existence of QPOs with the period of hours in AGNs.

Keywor-ds: Accretion, Accretion Disks; Black Hole Physics; Shock Waves

1. Introduction
Following the pioneer works of transonic problems of accretion and wind by Fukue 1
and Chakrabarti and hil:-> collaboratorl:->,2-4 it has been shown that these general-
ized accretion flows could be rel:->ponsible for the qual:->i-periodic oscillations (QPOs)
from the black hole candidates. 5- 7 Recent numerical 2D I:->imulations of the shocks 8 ,9
confirm such QPOs in stellar black holel:-> and supermassive black holes. Here we ex-
amine further shock oscillation models around a supermasl:->ive black hole with mass
of 10 6 1VI8 and with a wide range of the input accretion rate.

2. Model Parameters and Numerical Methods

Mass m
1.64 0.0751 0.0738 1.25 X 10- 6 30
1.875 0.0751 0.0654 10- 17 - 10- 6 30

For the central black hole, we consider a stellar-mass and a supermassive black
hole with mass of 10 and 10 6 ~M8, respectively. The basic equations and the numerical

1072
 1073

methods used here are given in Okuda et al. 8 Table 1 gives typical sets of injection
parameters, such as the specific angular momentum Aout, the radial velocity Vout, the
sound velocity aout, the ambient density Pout, and the accretion rate m normalized
to the Eddington critical accretion rate ME at an outer boundary radius Raub where
MEc 2 L E. Here, the velocities and distances are given in units of the speed oflight
c and the Schwartzschild radius R g , respectively. 1> is the subtended angle of the
central black hole to the initial disk at r Rout, that is, tan 1> = (h/r)out, where h
is the inflow thickness. These parameters are estimated from ID transonic solutions.
Under these flow parameters and adequate initial and boundary conditions, we solve
the initial value problem.

46,..------------,

0,17

40 1.7E-4

380.,...............~"'20"'OO~-'-"-"""'40"'00~-'-"-.....,.,60!;;;00.........
Time (Rg/c)

(a) (b)
Fig.1. (a) The luminosity curves for m 1.7 0.17,17.0, and 1.7 x 10 7 . (b) The power X 10- 4 ,
density spectra of the luminosity (red line) and shock position on the eqnatorial plane (blue line).

10'

10'

24 ~
~ 10'
.Il 0'

..
I
22<:1: !
~ 10'

"'-,
Masaloss flate 10'
18

16 10·;0 10 10 10 10
~ -4 ~ 0 2 10 10
logM/M_E MIM_sun

(a) (b)

Fig. 2. (a) Luminosity L and mass outflow rate Mloss versus mass accretion rate m. (b) OPO
period versus black hole mass.
1074

3. Numerical Results
In all of the above models, the centrifugally supported shocks are formed at the
region of 12 - 18Rg and 5 - 12 Rg for the stellar-mass and supermassive black holes,
respectively, depending weakly on the accretion rate, and oscillate quasi-periodically
around the shock position. As the results, for the stellar-mass black hole, we obtain
the QPO frequency of Vqpo = 4 - 7 Hz and the period P= 0.14 - 0.25 sec. 8 On the
other hand, for the cases of the supermassibe black hole, we find the QPO frequency
Vqpo = 3 X 10- 5 - 3 X 10- and the period P = 1 - 10 hours. The luminosities show
4

QPO behaviours with modulations of a factor 2 - 3 and with quasi-periods of a few

to several hours. Figure l(a) shows the luminosity curves for the latter cases with
m = 1.7 x 10- 4 , 0.17, 17.0, and 1.7 x 10 7 , respectively. The power density spectra
for the luminosities and the shock positions on the equatorial plane are given in
figure 1 (b). For the supermassive black hole with 10 6 M(,), we examined a wide rage
of accretion rates of m = 10- 4 - 10 7 . The luminosity increases in proportion to the
input accretion rate when it is low, but the luminosity tends to a saturated value
of", 3L E when the accretion rate exceeds considerably ME. Figure 2(a) shows the
luminosity L/ LE versus accretion rate it/ME, If we assume L = r;M c 2 , where r; is
the efficiency of graviatational energy into radiation, r; is estimated approximately
to be 1/16 when TTL is small. For very high accretion rates, the shock locations are
near 5 - 8 Rg but, for low accretion rates, the position moves outward in accordance
with the previous results in 1D flows. 8 Figure 2(a) also gives tha mass outflow rate
Mlo", from the system. The mass outflow rate increases in proportion to the input
accretion and it amounts to a few percent of the input accreting matter.
Together with other numerical results 9 obtained from the shock oscillation model
around the black holes with masses of 10 and 108 M(,), we plot the QPO period-mass
relation in figure 2(b). From the results, we obtain approximately

PQPO = 0.016M/M(,) sec. (1)

This suggests QPO periods expected from the shock oscilation models around the
black holes in various mass range.

References
1. J. Fukue, PASJ 39, 309(1987).
2. S. K. Chakrabarti,em ApJ, 347, 365(1989).
3. M. A. Abramowicz and S. K. Chakrabarti, ApJ, 350, 281(1990).
4. S. K. Chakrabarti and D. Molteni, ApJ, 417, 671(1993).
5. D. Molteni, H. Sponholz and S. K. Chakrabati, ApJ, 457, 805(1996)
6. D. Ryu, S. K. Chakrabarti and D. Molteni 1997, ApJ, 474, 378(1997)
7. G. Lanzafame, D. Molteni and S. K. Chakrabarti,MNRAS, 299, 799(1998)
8. T. Okuda, V. Teresi and D. Molteni, PASJ, 56, 547(2004)
9. S. K. Chakrabarti, K. Acharyya and D. Molteni,A&A, 421, 1(2004)
 OBSERVING THE FLARES OF Sgr A *
WITH THE VERY LARGE TELESCOPE INTERFEROMETER

THIBAUT PAUMARD1,t,*, T, MULLER2, R. GENZEU,3, F, EISENHAUERl and

S. GILLESSENl
1 MPE, GiefJenbachstmfJe, 85748 Garching, Germany
t thibaut. [email protected]

21nstitute of Astronomy and Astrophysics, Dept. of Theoretical Astrophysics,

University of Tubingen, Germany
3 Department of Physics, University of California, Berkeley, USA

Sgr A *, the super-massive compact object at the centre of the Galaxy, is known to exhibit
energetic events (" flares") in the near infrared and X-ray domains. These events are likely
due to processes very close to the central object, on a scale of a few Schwarzschild radii
(eo:: 10 j.Las). The Very Large Telescope Interferometer (VLTI) is in principle able to
deliver astrometry with this accuracy. We have used relativistic ray tracing to study
the possible apparent path of the centroid of flares given reasonable assumptions, and
shown that the VLTI will be able to resolve proper motion of flares. We also find that
the lightcurves already observed with adaptive optics techniques are well explained by a
hot spot being sheared on its orbit around the black hole.

Keywords: Galaxy: nucleus; Gravitational lensing; Techniques: interferometric.

Reaching the Schwarzschild scale using optical interferometry

The centre of our Galaxy is the closest galactic nucleus to the Earth, about 100
times closer than the Andromeda galaxy. The radio source Sgr A * lays motionless
(to within a few km/s) at the centre of gravity of the nuclear cluster. It has variable
counterparts in the infrared and X-ray domains. Dynamical evidence show that this
source is powered by a supermassive object, most likely a black-hole (BH).1-4 The
Schwarzschild radius for a massive object is given by Rs = 2QNl/c 2 , corresponding
to an angle (}s = Rs/ D at a distance D. Sgr A * has a mass N10 = 3.61±0.32 x 106M8
and distance Ro = 7.62±0.32 kpc,4 which yields (}s = Rs/ D ::::' 14R8/ Ro ::::' 8.6 f..LaS.
In contrast, for a stellar BH (Nl ::::' 1M 8 ) at 1 kpc, (}s ::::' 2 X 10- 5 /Las. Indeed, Sgr A *
is thc BH candidate with the largest apparent size, and therefore the candidate of
choice for resolving effects at the Schwarzschild scale.
This scale is still two orders of magnitudes too small for imaging in the near in-
frared, even using the most advanced interferometers (the four telescope ESO Very
Large Telescope Interferometer or VLTI, and the two telescope Keck interferome-
ter). However, it can be reached by these interferometers in differential astrome-
try between the target and a reference source. The Phase-Referenced Imaging and
Micro-arcsecond Astrometry (PRIMA)5 facility on the VLTI will reach a 10 /Las ac-
curacy for a typical observational setup: t = 30 min integration time, a separation
of 8 = 10" between the reference star and the science object, and a (projected)

*currently at LESIA, Observatoire de Paris/CNRS, 5 place Jules Janssen, 92190 Meudon, France

1075
1076

baseline B = 200 m. This accuracy scales as IJe ex: B-2/3er1/2.6 The GRAVITY
consortium exploits this scaling law in the concept for a new instrument 7 which will
allow 10-lLas-astrometry with minute time sampling. Such a high temporal resolu-
tion is warranted to resolve the Schwarzschild timescale of Sgr A *.
Sgr A * is very faint compared to its Eddington luminosity. However, it shows
variability at various timescales, ranging from minutes to years. In particular, it
exhibits outbursts of energy called 'flares' in both X rays8.9 and infrared. 10-12 These
flares last for for c::= 10-200 minutes and occur a few times a day. On these occasions,
the luminosity of Sgr A * increases by a factor of up to 100 in X and up to 20 at K,
making Sgr A * a mK c::= 15 source. 10 On top of these bright, hour-scale flares, quasi-
periodic variations with a typical timescale of 13-25 min are observed. 10 ,13 This
timescale corresponds to that of the period of the innermost stable corotating orbit
(ISCO) around the BH (27 min for a 3.6 x 10 6 M8 Schwarzschild BH), and is natural
if the flares involve material on these orbits, close to Sgr A *. For instance, Ref. 10
find a pseudo-periodic signal at c::= 17min. Assuming Sgr A * is a 3.6 x 10 6 M8 Kerr
BH and this pseudo-period corresponds to the period on the ISCO, they give an
estimate of its spin parameter (a = 0.52). Ref. 13 argue that flares happen at various
radii larger than the ISCO. From the shortest period ever observed (13 ± 2 min),
they argue that a ~ 0.70.

"
/
Cfj. I
!

/j
,
-~

\ , "
---I~ ___~ \_/)
-?5r---~-
'
----
,
\

, ,
,

j
~~~_~_ '-4_ ,J
c:. -4 -2 o
x (Rs)

Fig. 1. Apparent track on the sky of a red (vLl/ <X v- 3 ) blob orbiting a Schwarzschild BH seen
at an inclination of 45° (left) and 80° (right). The dashed (resp. dash-dotted) line corresponds to
the apparent track of the primary (resp. secondary) image of the blob.

We have used relativistic ray tracing 14 - 16 to study the possible apparent path of
the centroid of flares in Sgr A * given reasonable assumptions. Our models assume
the flare emission comes from matter orbiting the BH, and include the effects of
polarisation, shear, and finite temporal sampling. We have shown that 10 lLas as-
trometry, as will be provided by the GRAVITY instrument on ESO VLTI, had the
potential to uncover the nature of these flares, and to give access to strong lensing
effects. This work follows on Refs. 17,18 and will be published in further detail in
Ref. 19.
 1077

References
L R. Schodel, T. Ott, R. Genzel, R. Hofmann, M. Lehnert, A. Eckart, N. Mouawad,
T. Alexander, M. J. Reid, R. Lenzen, M. Hartung, F. Lacombe, D. Rouan, E. Gen-
dron, G. Rousset, A.-M. Lagrange, W. Brandner, N. Ageorges, C. Lidman, A. F. M.
Moorwood, J. Spyromilio, N. Hubin and K. M. Menten, Nature 419, 694 (October
2002).
2. R. Schodel, T. Ott, R. Genzel, A. Eckart, N. Mouawad and T. Alexander, ApJ 596,
1015 (October 2003).
3. A. M. Ghez, G. Duchene, K. Matthews, S. D. Hornstein, A. Tanner, J. Larkin, M. Mor-
ris, E. E. Becklin, S. Salim, T. Kremenek, D. Thompson, B. T. Soifer, G. Neugebauer
and I. McLean, ApJL 586, L127 (April 2003).
4. F. Eisenhauer, R. Genzel, T. Alexander, R. Abuter, T. Paumard, T. Ott, A. Gilbert,
S. Gillessen, M. Horrobin, S. Trippe, H. Bonnet, C. Dumas, N. Hubin, A. Kaufer,
M. Kissler-Patig, G. Monnet, S. Strobele, T. Szeifert, A. Eckart, R. Schodel and
S. Zucker, ApJ 628, 246 (July 2005).
5. A. Glindemann and S. Leveque, PRIMA - The VLTI Dual Feed System, in Prom
Extrasolar Planets to Cosmology: The VLT Opening Symposium, Proceedings of the
ESO Symposium held at Antofagasta. Chile, 1-4 March 1999, eds. J. Bergeron and
A. Renzini (Springer-Verlag, Berlin, 2000).
6. M. Shao and M. M. Colavita, ABA 262, 353 (August 1992).
7. F. Eisenhauer, G. Perrin, S. Rabien, A. Eckart, P. Lena, R. Genzel, R. Abuter and
T. Paumard, Astronomische Nachrichten 326, 561 (August 2005).
8. F. K. Baganoff, M. W. Bautz, W. N. Brandt, G. Chartas, E. D. Feigelson, G. P.
Garmire, Y. Maeda, M. Morris, G. R. Ricker, L. K. Townsley and F. Walter, Nature
413, 45(September 2001).
9. F. K. Baganoff, Y. Maeda, M. Morris, M. W. Bautz, W. N. Brandt, W. Cui, J. P.
Doty, E. D. Feigelson, G. P. Garmire, S. H. Pravdo, G. R. Ricker and L. K. Townsley,
ApJ 591,891 (July 2003).
10. R. Genzel, R. Schodel, T. Ott, A. Eckart, T. Alexander, F. Lacombe, D. Rouan and
B. Aschenbach, Nature 425, 934 (October 2003).
11. A. M. Ghez, S. A. Wright, K. Matthews, D. Thompson, D. Le Mignant, A. Tanner,
S. D. Hornstein, M. Morris, E. E. Becklin and B. T. Soifer, ApJL 601, L159 (February
2004).
12. Y. Clenet, D. Rouan, D. Gratadour, O. Marco, P. Lena, N. Ageorges and E. Gendron,
ABA 439, L9 (August 2005).
13. S. Trippe, T. Paumard, T. Ott, S. Gillessen, F. Eisenhauer, F. Martins and R. Genzel,
MNRAS 375, 764 (March 2007).
14. D. Weiskopf, Four-dimensional non-linear ray tracing as a visualization tool for grav-
itational physics, in IEEE Visualization 2000 Proceedings, eds. T. Ertl, B. Hamann
and A. Varshney (ACM Press, October 2000).
15. D. Weiskopf, Visualization of four-dimensional spacetimes, PhD thesis, Eberhard-
Karls- U niversitiit Tubingen, (Tubingen, Germany, 2001).
16. T. Muller, Visualisierung in der relativitiitstheorie, PhD thesis, Eberhard-Karls-
Universitiit Tubingen, (Tubingen, Germany, 2006).
17. F. Delplancke, M. Gai, R. Genzel, A. Glindemann, F. Eisenhauer, T. Ott, T. Paumard,
G. Perrin, J. Pott, J. Surdej and O. O. Von der Luhe, Reference missions for prima
ESO STC-362, (2004).
18. T. Paumard, G. Perrin, A. Eckart, R. Genzel, P. Lena, R. Schoedel, F. Eisenhauer,
T. Mueller and S. Gillessen, Astronomische Nachrichten 326, 568 (August 2005).
19. T. Paumard et al., ABA in prep. (2007).
 SIMULATING VLBI IMAGES OF Sgr A *
SCOTT C. NOBLE 1 • 2 ,t, PO KIN LEUNG3, CHARLES F. GAMMIE 2,3,4 and
LAURA G. BOOK2
1 Department of Physics and Astronomy, Johns Hopkins University, Baltimore, MD 21218
2 Department of Physics, University of Illinois at Urbana-Champaign, Urbana, IL 61801 USA
3 Department of Astronomy, University of Illinois at Urbana-Champaign, Urbana, IL 61801
USA
4 Institute for Advanced Study Einstein Drive, Princeton, NJ 08540
t [email protected]

Very Long Baseline Interferometry (VLBI) at sub-millimeter/millimeter wavelengths

shows promise at resolving the silhouette of the supermassive black hole at the Galactic
Center, Sagittarius A * (Sgr A *), in the near future. In order to accurately test theoretical
models of Sgr A * using these observations, a direct comparison of VLBI data to numerical
models must be made. We present calculated images and spectra of Sgr A * using accre-
tion disk simulation data from general relativistic magnetohydrodynamics (GRMHD)
evolutions. Synchrotron and bremsstrahlung emission models are considered in the op-
tically thin limit, which allows us to solve the radiative transfer equations using only
simulation and geodesic data in a post-processing step. We show predictions of millime-
ter observations at the expected angular resolution limit and the spectrum's variability.

Keywords: accretion, accretion disks; Galaxy:center; radiative transfer; MHD

1. Introduction
A particularly interesting object that is now receiving a great deal of attention is the
black hole at the center of our galaxy, the radio source known as Sgr A *. The black
hole's mass, M '::::' 4 X 106 M G , and distance from Earth, R = SkpC,1-3 are such that
it is the black hole with the largest angular size on our sky. This makes it the best
candidate for directly viewing the silhouette-or "shadow" 4-of an event horizon.
However, it is puzzling since it seems to accrete little matter and the matter it does
accrete radiates weakly.5,6
Current popular theories of Sgr A * fall into two categories: jet models 7,8 and
radiatively inefficient accretion flow (RIAF) models. 9 Each of these models are freely
specified by a number of degrees of freedom and, consequently, can predict the
spectrum quite well.
These two theories neglect spacetime curvature effects and do not account for dy-
namical variations of the spectrum self-consistently. More detailed calculations have
been performed but they either include general relativity (GR) and omit plasma
dynamics, 10, 11 or they include the dynamics but neglect GR effectsJ2 Here we in-
clude both aspects and present the first self-consistent optically thin calculations
of Sgr A * 's image and spectrum at about A = lmm, near the peak of its spectral
energy distribution (SED) and is approximately where the disk becomes optically
thin. 13 ,14 This band of radiation is particularly interesting since improvements in
millimeter and sub-millimeter VLBI will soon permit features at the scale of the
horizon to be resolved. 15

1078
 1079

2. Theory and Results

For Sgr A *, the material is well described by a well ionized plasma moving within the
background potential of the black hole. We thus evolve the ideal GRMHD equations
on a fixed, stationary metric-i.e. the Kerr spacetime-and omit radiation back-
reaction since we only consider optically thin emission.
The GRMHD equations of motion in axisymmetry are integrated using the flux-
conservative scheme known as HARM. 16.17 Our initial data consists of an orbiting
torus in hydrodynamic equilibrium. 18 A weak poloidal magnetic field is embedded
within the torus along density contours. The magnetorotational instability 19 drives
the development of turbulence in the disk and material accretes onto the black hole.
The disk's emission is calculated as a post-processing step using simulation data
over the time-steady phase of accretion. We consider non-polarized emission from a
thermal distribution of electrons, assuming that electron temperature is equal to the
baryon temperature. We include both synchrotron and bremsstrahlung radiation,
confirming that the former dominates. The radiative transfer equation in GR is
integrated along geodesics that penetrate the accretion flow. We use an anisotropic,
angle-dependent approximation for the synchrotron emissivity given by,20 which
yields results to an accuracy no worse than any of our other assumptions or approx-
imations. 21
Figure 1 shows spectra calculated from different snapshots of the simulation. The
filled circles with error bars represent observed flux values of Sgr A * during quies-
cence. 5,22-26 The exes are flux measurements during flare events,25 and the arrows
indicate upper limits at NIR/IR wavelengths. 22 Error bars indicate the measured
errors quoted in the references, which sometimes include intrinsic variability as well
as measurement uncertainty. We calculate Lv assuming isotropic emission. The ac-
cretion rate used for all SEDs was set by constraining the flux density at A = Imm
to 4Jy at t = I250GiVic- 3 . We find that variability near the power maximum is
at the same order of magnitude as what is observed. Further, we find that, for the
particular model considered here, the emission at higher frequencies is within upper
limits but can lie near observed fluxes at NIR frequencies. Departures from observa-
tions at low frequencies is the result of the finite extent of our simulation data and
the fact that the disk is optically thick there. We also find the spectrum becomes
harder as the inclination angle decreases or for simulations with faster spinning
black holes. 27
Shown in Figure 1 are both an "infinite" resolution image and an image sim-
ulating what an Earth-based VLBI observation would see at Imm. The simulated
VLBI image was created by convolving the original with a Gaussian beam the size
corresponding to a 8000km baseline observation. Since anisotropic scattering from
the interstellar medium becomes comparable to the theoretical diffraction limits of
VLBI at wavelengths greater than Imm, it is ignored. We find that the black hole
silhouette is noticeable in images calculated at VLBI resolutions for inclinations
greater than 30° (45 0 is shown here).27
1080

36

32

12 14
Log lO [ J/(Hz) 1

Fig. 1. Left frame: spectra taken at iinc 30° USiilg snapshots of the a* 0.94 disk at different
points along its evolution. Lines A-G are spectra from times 1150, 1250, 1326, 1434, 1500, 1560,
1666G M c- 3 , respectively. Middle frame: image of the accretion disk viewed at a wavelength of
Imm seen at an inclination angle of 45°. The frame is 601\l[ wide in the plane of the singularity.
Right frame: Gaussian convolution of the middle image simulating a diffraction limited, 8000km
baseline VLBI observation. Bottom: relative greyscale map used in each image.

Acknowledgments
This research was supported by NSF grants AST 00-93091 and PRY 02-05155.

References
1. A. M. Ghez, S. Salim, S. D. Hornstein, A. Tanner, J. R. Lu, M. Morris, E. E. Becklin
and G. Duch€me, AstTophys. J. 620, 744CFebruary 2005).
2. F. Eisenhauer, R. Genzel, T. Alexander, R. Abuter, T. Paumard, T. Ott, A. Gilbert,
S. Gillessen, M. Horrobin, S. Trippe, H. Bonnet, C. Dumas, N. Hubin, A. Kaufer,
M. Kissler-Patig, G. Monnet, S. Strabele, T. Szeifert, A. Eckart, R. Schadel and
S. Zucker, AstTOphys. J. 628, 246(July 2005).
3. A. M. Beloborodov, Y. Levin, F. Eisenhauer, R. Genzel, T. Paumard, S. Gillessen and
T. Ott, AstTOphys. J. 648, 405 (2006).
4. H ..F'alcke, F. Melia and E. Agol, AstTOphys. J. Lett. 528, L13 (2000).
5. J.-P. Macquart, G. C. Bower, M. C. H. Wright, D. C. Backer and H. Falcke, AstTO-
phys. J. Lett. 646, 1.,111 (2006).
6. D. P. Marrone, J. M. Moran, J.-H. Zhao and R. Rao, Jov,Tnal of Physics Conference
SeTies 354 (2006).
7. H. Falcke, AstTOphys. J. Lett. 464, 1.,67 (1996).
8. H. Falcke and S. Markoff, AstTon.€.1AstTOphys. 362, 113 (2000).
9. F. Yuan, E. Quataert and R. Narayan, AstTOphys. J. 598, 301 (2003).
10. A. E. Broderick and A. Loeb, Mon. Not. Roy. AstTOn. Soc. 363, 353 (2005).
11. A. E. Broderick and A. Loeb, Mon. Not. Roy. AstTOn. Soc. 367, 905 (2006).
12. J. E. Goldston, E. Quataert and 1. V. Igumenshchev, AstTOphys. J. 621, 785 (2005).
13. Z.-Q. Shen, K. Y. 1.,0, M.-C. Liang, P. T. P. Ho and J.-H. Zhao, Natv,Te 438,62 (2005).
14. G. C. Bower, W. M. Goss, H. Falcke, D. C. Baeker and Y. Lithwick, AstTOphys. J. Lett.
648, L127 (2006).
15. S. Doelema,n and G. Bower, Galactic Center Newsletter 18, 6 (2004).
16. C. F. Gamrnie, J. C. MeKinney and G. Toth, AstTOphys. J. 589, 444 (2003).
17. S. C. Noble, C. F. Gammie, J. C. MeKinney and L. Del Zanna, AstTOphys. J.641,
626 (2006).
 1081

18. L. C. Fishbone and V. Moncrief, Astmphys. 1. 207, 962 (1976).

19. S. A. Balbus and J. F. Hawley, Astmphys. 1.376, 214 (1991).
20. C. Wardziriski and A. A. Zdziarski, Mon. Not. Roy. Astron. Soc. 314, 183 (2000).
21. P. K. Leung, S. C. Noble and C. F. Cammie, in progress, (2007).
22. E. Serabyn, J. Carlstrom, O. Lay, D. C. Lis, T. R. Hunter and J. H. Lacy, Astm-
phys. 1. Lett. 490, L77 (1997).
23. H. Falcke, Vl . .M. Goss, H. Matsuo, P. Teuben, J.-H. Zhao and R. Zylka, Astmphys. 1.
499, 731 (1998).
24. S. D. Hornstein, A. M. Chez, A. Tanner, M. Morris, E. E. Becklin and P. vVizinowich,
Astmphys. J. Lett. 577, L9 (2002).
25. J.-H. Zhao, K. H. Young, R. M. Herrnstein, P. T. P. Ho, T. Tsutsumi, K. Y. Lo, W. M.
Coss and G. C. Bower, Astmphys. J. Lett. 586, L29 (2003).
26. D. P. Marrone, J. M. Moran, J.-H. Zhao and R. Rao, Astmphys. 1. 640, 308 (2006).
27. S. C. Noble, P. K. Leung, C. F. Cammie and L. C. Book, ArXiv, astro-ph/070I778
(2007).
This page intentionally left blank
Astrophysical Black Holes
This page intentionally left blank
 ASTROPHYSICAL BLACK HOLES - DO THEY HAVE
BOUNDARY LAYERS?

SANDIP K. CHAKRABARTI 1 ,2, H, GHOSH 1 and D, SOM 1

1 S, N. Bose National Centre for Basic Sciences, JD-Block, Sector III,
Salt Lake, Kolkata, 700098, India. E-mail:[email protected]
2 Centre for Space Physics, Chalantika 43, Garia Station Rd., Kolkata 700084, India.

We show that the black hole astrophysics is in reality the physics of the boundary layer
around the black hole. This boundary layer has first been proposed by our group and
is now universally used in various names. This layer is produced primarily due to the
centrifugal barrier that the accretion flow experiences. This layer is produced only in
flows where the angular momentum distribution is 'sub-Keplerian'. Keplerian disk and
the synchrotron emission in the non-Keplerian disk supply the requisite soft photons.
Comptonization of these soft photons produces the harder photons by the hot electrons
in this layer. The outflow is also driven by this layer. Quasi-periodic oscillations are
primarily due to the oscillations of this layer. The spectral states vary due to the varia-
tions of the Keplerian and sub-Keplerian components. All the observational results today
point to the existence of this boundary layer and two flow components of varying degree.
The same concept holds good for objects with a hard surface such as white dwarfs and
neutron stars, though the situation is more complex as there could be more than one
boundary layer.

Keywords: accretion, accretion disk - black hole physics-shock waves.

1. Introduction
Black holes are supposed to be the most mysterious of all objects. Yet, they are
the simplest, since they are characterized by at the most two parameters, namely
the mass (Nhh and the spin a). However, from an observational point of view, such
an object is not very useful, as it does not emit any radiation in the classical sense
to make itself visible. Thus the observers are more interested in those black holes
which are accreting, either from the companion as in a binary system, or through
the capture of nearby stars and stellar winds as in the galactic centers. This natu-
rally increases the number of parameters - the most important of them being the
mass accretion rate of the Keplerian eNId) and the sub-Keplerian (lvh) components,
the angular momentum distribution (>.(x)) of the component that is non-Keplerian
(which, in turn, is dictated by the viscosity parameter) and the strength of stochas-
tic magnetic field f3 which is measured by the ratio of the magnetic pressure to total
pressure. These parameters are independent and they determine the intrinsic spec-
tral states, rate of outflows and winds, frequency of quasi-periodic oscillations etc.
Complex behaviour of the emitted radiation out of these few parameters is possible
only if the governing equations are highly non-linear.
In this review, we will present how our solution configurations of the black hole
accretion perfectly explain all the observable features of the spectra. Our major
stress will be on the properties of the so-called 'CENtrifugal pressure supported
BOundary Layer' or CENBOL just outside the horizon.

1085
1086

2. Generic Properties of an Accretion Flow Around a Black Hole

There are as many number of 'models' in the present literature of the black hole
accretion as there are groups working on this subject. This is because, observers do
not necessarily require any solutions to explain observations. They try to explain by
drawing cartoons (from 'educated guess' or 'inspired guess'!) of what might be going
on. Fortunately, the governing equations have only one set of solutions for a given
boundary condition on the horizon. Thus, either al these guesses converge to the
solution or the solution inspires ones to make the guesses. In any case, it is fortunate
that all of the cartoons are converging to the two component advective flow solution
that was originally presented by our group in the context of the extragalactic black
holes in Chakrabarti (1994) and later elaborated in Chakrabarti & Titarchuk (1995)
in the context of the galactic black holes. Here we show how a unique picture of
black hole accretion is emerging. Before going further, we mention that we will use
the Schwarzschild radius r 9 = 2G Mbh/ c2 , the velocity of light c and Mbh to be the
unit of distance, the velocity and the mass respectively.

2.1. Why the Black Hole Accretion Picture Must be Unique

Black holes have a unique property that the velocity of matter at the event hori-
zon is the same as the velocity of light independent of the history of the matter
(Chakrabarti 1990a, 1996a). However, the sound velocity is at the most JI73 (in
units of the velocity of light). Thus the inflow is always supersonic. On the other
hand, a Keplerian disk is always sub-sonic. Thus, the matter must cross at least
one sonic point before leaving the Keplerian disk and entering a black hole. A most
logical location of this sonic point (where the Mach number is unity) is clearly be-
tween the marginally stable orbit and the horizon (these locations can vary between
r = 3 to 1 and r = 1 to 1/2 respectively depending on the spin parameter a of the
black hole). Indeed, the early study of the black hole accretion concentrated on this
aspect of the flow only (Paczynski & Czerny, 1980; Matsumoto et al. 1984). Mean-
while it was noted that the number of sonic points could be more than one (Liang
& Thompson, 1980) and similar to solar winds, one could find a shock solution also
(Fukue, 1987) in black hole accretion.
The accretion flow which includes a standing, propagating or oscillat-
ing shock wave which was extensively studied by Chakrabarti and his co-
workers (Chakrabarti, 1989; Abramowicz & Chakrabarti, 1990; Chakrabarti, 1990;
Chakrabarti 1996a) added a totally new dimension to the subject (see, later follow-
up of this work in, for example, Nobuta & Hanawa, 1994; Ryu, Brown, Ostriker
& Loeb, 1995; Yang & Kafatos, 1995; Yuan, Dong & Lu, 1996; Caditz & Tsuruta,
1998; Kovalenko, 1999; Takahashi, 2000; Le & Becker, 2005; Yu, Lou, Bian & Wu,
2006). The formation of the shock, one way or the other, was guaranteed by the
simple fact that the rate of angular momentum loss is very slow when the matter
is rushing towards the black hole and the resulting centrifugal force, which goes as
)..2 /r 3 , where, r is the radial distance from the black hole, becomes immensely high
 1087

and comparable to the gravitational force which goes as 1/r2. Thus, the matter
bounces from the centrifugal barrier. For a large 'Newtonian' object, this is not a
cause of concern since the matter never goes that close to the axis anyway. This
bounced matter slows down incoming matter and eventually piles up outside the
horizon to form a shock wave.
The uniqueness of the black hole picture, as far as the hydrodynamic properties
of the flow are concerned, is due to the transonic nature of the flow. Out of infinitely
many solutions, only one solution connects the horizon to infinity - the only one
which passes through the sonic point (the direct solution). However, there is another
incomplete solution which has higher entropy (Chakrabarti, 1989) and also passes
through a sonic point (the indirect solution). This incomplete and indirect solution
is always chosen whenever it is present simply because of entropy consideration.
The complete solution is thus partly from the direct solution and the rest from the
indirect solution. At the junction, some shock condition (such as Rankine-Hugoniot
condition for non-dissipative flow) may be satisfied and the shock would be a stand-
ing shock. Otherwise, it will exhibit oscillations indefinitely (Ryu, Chakrabarti &
Molteni, 1996), unless the flow parameters get changed dynamically.
In a viscous flow, the angular momentum is transported differently in the pre-
and the post-shock region if the incomplete prescription for viscous stress (w r ¢ =
-aP, where, wr¢ is the stress, P is the pressure and a < 1 is a constant.) is
used. Shocks do not form in such a case. However, one must recall that the radial
velocity is :::;j 1 close to the black hole, and the ram pressure becomes as important
as the thermal pressure. Thus, the correct and the most appropriate prescription
for viscous stress would be w r ¢ = -a(P + pv 2 ), where, p is the density and v is the
radial velocity. This prescription allows one to have shocks even in moderately high
viscous flows. Of course, when the viscosity is high, the flow topology is changed
totally and shock condition is not satisfied. The shock wave formed is driven by the
pressure difference and it eventually propagates to a large distance (Chakrabarti &
Molteni,1995).
When one considers a neutron star accretion, the centrifugal force dominated
shock may form also, but there is another shock transition, right next to the hard
surface of the star, which is commonly known as the boundary layer of the star. Some
study of neutron star boundary layer along this line has been made (Chakrabarti
& Sahu, 1997).

2.2. The Question of the Boundary Layer of a Black Hole

Ideally a boundary layer is the one where the flow dissipates its energy and angular
momentum in order to 'come to terms' with the object it needs to stay with, i.e., to
satisfy the boundary condition dictated by the object. For a star with a hard surface,
it is known to exist. However, for a black hole, the unique occurrence of the shock
wave, where matter virtually stops to dissipate the energy and possibly angular
momentum, the boundary layer is not something commonly thought of. Indeed, it
1088

is hardly a 'layer' which has a thickness ranging from a few to tens or even hundreds
of Schwarzschild radii, This region, in between the innermost sonic point and the
shock is termed as CENBOL - a CENtrifugal pressure supported BOundary Layer.
This layer may be oscillating or standing depending on the nature of the outer
boundary of the CENBOL, namely, the shock wave,
The hydrodynamic property of this boundary layer is very similar to that of a
thick accretion disk (e,g., Paczynski & Wiita, 1980), originally proposed to study
the collimation of the jets and outflows in active galaxies and quasars. The reason is:
in the post-shock region the flow virtually stops and the matter is basically rotating,
as in a thick disk. However, unlike a thick disk, where a cusp could also form to hold
matter inside permanently, here there is no cusp. The post-shock matter is destined
to move towards the black hole, or, if there is sufficient pressure, along the axis to
supply matter to jets.

2.3. The Generic Nature of This Picture

We conclude this section by pointing out that the picture presented so far is totally
independent of the mass of the black hole or spin of the black hole. In the absence
of any heating or cooling effects, shocks form exactly at the same non-dimensional
distance .T s for all the black holes, stellar, massive or supermassive. This the jets
can form in massive Quasars (~ Nhh ~ 10 9 ), less massive quasars or milliquasars
(~ Nh" ~ 10 7 ), intermediate mass quasars or micro quasars (~ l\hh ~ 10 4 ) or
just stellar mass black holes or nallO-quasars (~ lvIbh ~ 10). In absolute terms the
distance varies, being proportional to the mass of the black hole. Similarly, the time
dependent phenomenon occur in similar non-dimensional time scale, but in absolute
terms, it is also proportional to the mass of the black hole.

3. The Physics and Astrophysics of the Boundary Layer of a Black

Hole
More then twenty five years ago, thick accretion disks were considered to be 'toroidal
stars' and indeed extensive studies of nucleosynthesis were made (Chakrabarti, Jin &
Arnett, 1987; Jin, Arnett & Chakrabarti, 1988) as one would for spherical stars also.
As discussed above, the post-shock region, i.e., CENBOL, is also an astrophysically
important 'object' and one needs to study it accordingly.
Even though the puffed up inner region of the disk was investigated by Eard-
ley and Lightman (1974), the study was made in the context of a Keplerian disk.
When the radiation pressure supported or ion pressure supported thick accretion
disks were considered in early eighties, only the puffed up portion remained, the
Keplerian disk disappeared from the scene. It is only when the shock study started
(Chakrabarti, 1989; Chakrabarti, 1990) then only it was realized that the puffed up
'thick disk' is nothing but the post-shock region. This generalized accretion disk has
the convenient feature (Chakrabarti & Wiita 1992; Chakrabarti, 1993; Chakrabarti
& Titarchuk 1995) in that a cool part of the disk (pre-shock region) and a hot part
 1089

of the disk (post-shock region) reside side by side. This feature was found to be
generic in black holes, or neutron stars (Chakrabarti & Sahu, 1997) and was later
utilized to compute the spectrum of the black hole and neutron star candidates
(Chakrabarti & Titarchuk, 1995; Chakrabarti, 1997).
The transonic solutions of the governing equations not only produce the accre-
tion flow, but also produce the outflows and winds exactly as the Parker's solution
in the 1950s. Indeed, by early 90s, through extensive time dependent numerical
simulations (Chakrabarti 1993, Chakrabarti & Molteni, 1993; Molteni, Lanzafame
& Chakrabarti, 1994) it was already established that a significant fraction of the
incoming matter must leave fram CENBOL along the axis to form the jets (see,
Molteni, Ryu & Chakrabarti, 1996, for jets coming out between the centrifugal bar-
rier and funnel wall and Hawley & Kralik, 2006 and Kralik, Hawley & Hirose, 2007,
for similar results). First quantitative computation of the outflow rate under various
circumstances llsing so-called GIOS (Global Inflow Outflow Solution) was carried
out by Chakrabarti (1998ab, 1999), Das & Chakrabarti (1999) and Das, Chattopad-
hyay & Chakrabarti (2002) (see also, Blandford & Begelman, 1999 for a self-similar
ADIOS model). The major assumption, which was partly motivated by the numeri-
cal simulations mentioned above, was that the outflow is a direct result of the shock.
In other words, no shock no significant outflow! It was also shown (Chakrabarti &
Titarchuk, 1995) that the temperature difference between the pre-shock and the
post-shock flow disappears when the spectrum becomes soft. In other words, the
shock also disappears in soft states. All these lead to the logical conclusion that
the outflow rates are strongly guided by the spectral states of a black hole. Similar
conclusion can be trivially drawn for neutron stars as well. These predictions have
been vindicated by observations.
Another progress made in sharpening the paradigm picture is to study the role
of viscosity. Already in analytical work it was pointed out that beyond a critical vis-
cosity the shock solution was not tenable (Chakrabarti, 1990b; Chakrabarti 1996b;
Chakrabarti & Molteni 1994, Chakrabarti & Das, 2004). What this means is that if
there is a variation in viscosity in the vertical direction (which is most natural), the
disk would segregate itself into two flows: the high viscous flow along the equatorial
plane and the low viscous flow away from the equatorial plane. The latter flow in
turn forms shocks and outflows. Thus the two component advective flow (TCAF)
became the most general flow of all. It is therefore natural to believe that any ob-
servation of black holes should be explainable with this generalized disk solution.
Indeed the cartoons and suggestions present in the literature do not seem to require
any new component which is not already in our advective disk paradigm.

3.1. The Flow Configuration of the Advective Disk Paradigm

How would this new paradigm disk look like? Our solutions of the governing equa-
tions which supplied bits and pieces of the new advective flow paradigm can now be
combined to have a general idea of the flow paradigm. In Fig. l(a-i) we show how
1090

our solution is gradually tuned and used to various observations. We also present
in Fig. 2 (a-g) the diagrams drawn by observers after fitting the data. The conver-
gence of these pictures is an assurance that finally we have started understanding
the accretion of the compact objects in its entirety. The major changes from the
previous paradigms are that (a) the understanding that the CENBOL, which is
also the base of the jet, is indeed the elusive Compton cloud and (b) the jets need
not be forming from the entire disk. Since early 90s we have been able to show

(a)

(b)

(c)

Cd)

(e)

Fig. 1. Cartoon diagrams of advective disk paradigm around black hole and nentron stars by
our group based on theoretical considerations: (a) Chakrabarti 1990a; Abramowicz & Chakrabarti
(1990); (b) Chakrabarti & Titarchuk, 1995; (c) Ebisawa, Titarchuk & Chakrabarti, 1996; (d)
Chakrabarti & Sahn, 1997; (e) Chakrabarti, 1997; (f) Chakrabarti, 1995ab; (g) Chakrabarti &
Nandi 2000; (h) Nandi et al. 2001; (i) Chakrabarti et al. 2002.
 1091

that the CENBOL and the outflows are inseparable. That has lead to some more
cartoon figures presented in Fig. 2 above. Serious readers are encouraged to follow
the references to see how the presence picture emerged.

3.2. Physics of the CENBOL

The local height of a disk can be computed from a local vertical equilibrium con-
dition where the vertical component of the gravitational force is balanced by the
pressure. This is easily seen to be rv a('r )r 3 / 2 where, a(r) is the local sound speed and
r is the local radial distance. In the post-shock region, the gas is heated up and the

(al (e)

Hard Soft

(f)

(d)

(g)

Fig. 2. Cartoon diagrams of constraints on flow configurations from observational point of view
(a) Fender et al. 1999; (b) Zdziarski, 2000; (c) Nowak, 2008; (d) Zdziarski & Gierlinksi 2004; (e)
Markoff & Nowak 2004; (f) Titarchnk & Shaposhnikov, 2005; (g) Ferinelli, Titarchuk & Frontera,
2007.
1092

thus the gas behaves like a thick disk. This puffed up disk intercepts soft photons
from the pre-shock flow and reprocesses them through inverse-Compton scatter-
ing. The processed photons are of hard X-ray. Repeated inverse-Comptonization
increases the photon energy till the recoil effect becomes important. In presence
of the stochastic magnetic field, the synchrotron soft photons behave like discrete
sources and they are reprocessed similarly. The emitted spectrum from thermal
Comptonization has been computed under various accretion rates in Chakrabarti
& Titarchuk (1995) and we do not repeat them here. Needless to mention, from
purely physical point of view, it is obvious that when the accretion rate in the Ke-
plerian disk is high (::::; 0.5MEdd) while the accretion rate in the sub-Keplerian flow
is moderate « O.IMEdd) the CENBOL around a stellar black hole will always be
cooled down and the energy spectral index a (E rv v-a) becomes high (a ~ 1).
However, for very low Keplerian disk rate:::; 0.1 the electrons in CENBOL cannot
be cooled down and the energy spectral index remains low (a :::; 1). The CENBOL
boundary, namely, the shocks also accelerate electrons to a power-law distribution
and these photons may also be inverse-Comptonized. This has been worked out
in detail in MandaI & Chakrabarti (2005) and Chakrabarti & MandaI (2006) (see
also, Wardziriski & Zdziarski, 2001; Bednarek & Giovanelli, 2007). The latter paper,
included soft photons from the Keplerian disk photons as well as those emitted by
synchrotron radiation.
When the CENBOL is totally cooled down due to the excess soft photons, the
Comptonization is done by the bulk motion, the radial transonic flow between the
inner sonic point and the horizon (Chakrabarti & Titarchuk, 1995) starts Comp-
tonizing the soft photons to energies comparable to the electron mass which in-
creases as one approaches closer to the horizon. This is known as the bulk motion
Comptonization (BMC; CT95, Ebisawa, Titarchuk & Chakrabarti, 1996). Unless
there is a strong magnetic field which causes a strong synchrotron photon emission
which could wash-out the BMC, this power-law hard tail gives the surest signature
of the black holes.
The CENBOL size depends on the specific angular momentum and the spe-
cific energy of the flow (Chakrabarti, 1989) and thus on the dissipative properties.
Generally speaking, the CENBOL boundary (shock location) is very sensitive to
the angular momentum, since it is primarily angular momentum supported. If the
Keplerian accretion rate is high, the post-region will be cooled down and they may
not support the shock structure. The CENBOL may collapse and disappear.
In case the shock condition is not satisfied the shock may still form, but it starts
oscillating without knowing where to form (Ryu, Chakrabarti & Molteni, 1997). As
a result, the shock oscillates and therefore the CENBOL size expands and contracts.
When the cooling is present, the shocks can also oscillate when the cooling time scale
matches with the infall time scale inside the CENBOL. Expanding and contracting
CENBOL intercepts soft-photons differently and thus the hard photon counts are
also modulated. If the soft photons are from the Keplerian disk, the expansion and
contraction modulates soft photons in a way that is proportional to the variation of
 1093

the surface area of the CENBOL. On the other hand, if the soft photons are from
synchrotron photons the variation is proportional to the volume of the CENBOL.
These two effects should be easily distinguishable.
CENBOLs being hot, they naturally produce winds, which of course, have
to be collimated by some external processes, such as magnetic hoop stress. The
wind may also be accelerated by the pressure of the radiation. When the CEN-
BOL disappears, the outflow and jet will also be subsided. Thus when the
outflow rate is computed as a function of the inflow rate, the shock strength
naturally appears as a parameter (Chakrabarti 1998; Chakrabarti 1999; Das
& Chakrabarti 1999; Das & Chakrabarti, 2000; Chattopadhyay & Chakrabarti,
2002; Fabrika, 2004; Chattopadhyay, Das & Chakrabarti 2004; Das, Chattopad-
hyay, Nandi & Chakrabarti 2001; Chattopadhyay & Das, 2007). Jets can also
be blobby, either due to Compton cooling (Chakrabarti & Nandi, 2000; or due
to explosive magnetic events (Nandi et aI, 2001; Fender, Belloni & Gallo, 2005;
ohkawa , Kitamoto, Kohmura, 2005; Fender & Belloni, 2004; Vadawale et al.
2003)
The CENBOL properties in Kerr black hole have also been studied using full
general relativity (Chakrabarti, 1996ac; Mondal & Chakrabarti, 2006) as well as
using pseudo-Kerr potential. The results are similar. In general, the spin of the black
hole causes the shock to move closer to the horizon and therefore, the oscillation
frequencies of these CENBOLS are also higher.

3.3. Astrophysics of the CENBOL

The properties of CENBOL as stated above happen to be just right to explain the
observational results from galactic and extragalactic black hole candidates. Some
of the earlier reviews have already discussed many of these issues (Chakrabarti,
1996d; Chakrabarti 2000ab; Chakrabarti, 2002ab) and we do not repeat them here.
Needless of mention, this paradigm has successfully withstood the tests of all types of
observations ranging from spectral state change, high energy spectra in both hard
and soft states (Chakrabarti & Mondal, 2006), quasi-periodic oscillation (QPO)
and its properties (Molteni, Sponholz & Chakrabarti, 1996; Chakrabarti, Molteni
and Acharyya, 2004; Okuda, Teresi & Molteni, this volume), especially the energy
dependence of QPO frequencies to relationship (Chakrabarti & Manickam, 2000;
Rao et al. 2000; Muno et al. 2001; Trudolyubov, 2001; Okuda, Teresi, Toscano &
Molteni, 2004; Rodgriguez et a1. 2004; Remillard & McClintock, 2006; Sriram et al.
2007) between the spectral state and jet/outflow formation (Chakrabarti, 1998ab,
1999, 2001ab). The most interesting aspect is that the interactions between the
CENBOL and the jet, especially, the mass gain to and mass loss from the CENBOL
have also been observed in detail (Chakrabarti et al. 2002) and expected behaviours
have been observed.
1094

4. Concluding Remarks
In the last few years, the subject of the black hole astrophysics has progressed
tremendously. The solutions with shocks and CENBOL region and the TCAF so-
lution have been found to explain most of complex properties observed from the
black hole today (e.g., Kafatos, et al. 1996; Crary et al. 1996; Ebisawa et al. 1996;
Strohmayer et al. 1996; Barret et al. 1996; Zhang et al. 1996; 1997; Belloni et al.
1997; Morgan, Remilalrd & Greiner, 1997; Ling et al. 1997; Kuznetsov et al. 1997;
Paul et al. 1998; Shrader & Titarchuk, 1998; Laurent & Titarchuk, 1999; Main et al.
1999; Muno, Morgan & Remillard, 1999; Zhang et al. 2000; McConnell et al. 2000;
Reilly et al. 2001; Tomsick, 2001; Smith, Heindl, Markwardt & Swank, 2001; Smith,
Heindl, Swank, 2002; Wang & Netzer, 2003; Roberts, et al. 2005; Pottshmidt et al.
2006; Cen, 2007). Since the energy dissipated at the post-shock region is observed as
jets/outflows and hard X-rays, we naturally term this region as a boundary layer of
a black hole. This is not surprising, since hitting on the centrifugal barrier is similar
to its hitting of the flow on a hard surface. The nomenclature may be marginally
confusing, since the boundary layers of the stars are very thin (much thinner than
the radius at any rate) while the boundary layers of black holes are many times
larger than the black hole size itself. The details of spectral fitting using the advec-
tive paradigm is in progress and the results would be reported in due course.
Acknowledgments: SKC acknowledges kind hospitality of ICRA, Pescara where this
work was partly completed. HG acknowledges an ISRO RESPOND project.

References
1. Abramowicz M. A. & Chakrabarti S. K, 1990, ApJ, 350, 281.
2. BarretD., GrindlayJ.E., BloserP.F., ZhangS.N., FishmanG.J., HarmonB.A., Pa-
ciesasW.S., FordE., KaaretP. & TavaniM., 1996, A & A Supp., 120, 121.
3. BednarekW. & GiovannelliF., 2007, A&A, 464, 437.
4. BeUoniT., vanderKlisM., LewinW.H.G., vanParadijsJ., DotaniT., MitsudaK &
MiyamotoS., 1997, A&A, 322, 857.
5. Blandford R. D. and Begelman, M. c., 1999, MNRAS, 303, L1
6. Caditz D. M. & Tsuruta S., 1998, ApJ, 501, 242.
7. CenR., 2007, astro-ph/0702660.
8. Chattopadhyay,I. & Chakrabarti,S.K, 2002, BASI, 30, 313.
9. Chattopadhyay,I., Das,S. & Chakrabarti,S.K, 2004, MNRAS, 348, 846.
10. Chattopadhyay,I. & Das,S., 2007, New Astron., 12, 454.
11. Chakrabarti S. K, 1989, ApJ, 347, 365.
12. Chakrabarti S. K, 1990a, Theory of Transonic Astrophysical Flows. World Scientific
Publishing, Singapore.
13. Chakrabarti, S.K, 1993, in Numerical Simulations in Astrophysics, Eds. J. Franco et
al. (CUP, Cambridge), 301
14. Chakrabarti S. K, 1995 in Procedings of the Texas Symposium, NYASA, 759, 546
15. Chakrabarti,S.K, 1996a, ApJ, 471, 237.
16. Chakrabarti S. K, 1996b, ApJ, 464, 664.
17. Chakrabarti S. K, 1996c, MNRAS, 283, 325
18. Chakrabarti,S.K, 1996d, Phys. Rep., 266, 229.
 1095

19. Chakrabarti, S K., 1997, ApJ, 484, 313

20. ChakrabartiS.K., 1998a, Ind. J. Phys., 72B, 565 (astro-ph/9810412)
2l. ChakrabartiS.K., 1998b, in Black Holes: Theory and Observations (Eds. F.W. Hehl,
C. Keifer, RJ.K. Metzler), 80
22. ChakrabartiS.K., 1999, A&A, 351,185.
23. Chakrabarti, S. K., 2000, Nuov. Cim. B, 115, 897.
24. ChakrabartiS.K., 2001 a, in High Energy Gamma Ray Astronomy (Eds. F. A. Aharo-
nian, H.J. V olk) , 558, 246 (AIP:NY)
25. ChakrabartiS.K., 2001b, ApSSS, 276, 191
26. Chakrabarti,S.K., 2002a, Proceedings of the MGIXMM Meeting (Eds. V. G.
Gurzadyan, R T. Jantzen, R Ruffini) (Singapore: World Scientific Publishing), 1613.
27. Chakrabarti, S.K. 2002b in Frontiers in Astrophysics Ed. S.K. Chakrabarti (New
Delhi: Allied Publishers)
28. ChakrabartiS.K. & AcharyyaK. & MolteniD., 2004, A&A, 421, l.
29. Chakrabarti, S. K., Jin L. & Arnett W. D., 1987, ApJ, 313, 674.
30. Chakrabarti, S. K. & Molteni D., 1993, ApJ, 417, 67l.
3l. Chakrabarti, S. K. & Molteni D., 1995, MNRAS, 272, 80
32. Chakrabarti, S.K. & Das, S., 2004, 349, 649.
33. ChakrabartiS.K. & MandaIS., 2006, ApJ, 642, 49.
34. ChakrabartiS.K. & ManickamS.G., 2000, ApJ, 531, 4l.
35. Chakrabarti,S.K. & Nandi,A., 2000, Ind. J. Phys., 75(B), 1 (astro-ph/0012526)
36. ChakrabartiS.K., NandiA., ManickamS.G., MandalS. & RaoA.R., 2002 ApJ, 579, 2l.
37. Chakrabarti S. K. & Sahu S. A., 1997, A&A, 323, 382.
38. Chakrabarti S. K., Titarchuk L. G., 1995, ApJ, 455, 623 (CT95)
39. Chakrabarti, S.K. & Wiita P.J., 1992, ApJ, 387, 21
40. CraryD.J., KouveliotouC., vanParadijsJ., vanderHooftF., ScottD.M., PaciesasW.S.,
vanderKlisM., FingerM.H., HarmonB.A. & LewinW.H.G., 1996, ApJ, 462, 7l.
4l. Das T. K., Chakrabarti S. K., 1999, CQG, 16, 3879
42. DasT.K. & ChakrabartiS.K., 2000, Nuc. Phys. B, 80, Ol.
43. Das S., Chattopadhyay, 1. & Chakrabarti S.K., 2001, ApJ, 557, 983
44. Das,S., Chattopadhyay,1., Nandi,A. & Chakrabarti,S.K., 2001, A&A, 379, 683.
45. EbisawaK., TitarchukL. & ChakrabartiS.K., 1996, PASJ 48, 59.
46. Fabrika,S., 2004, Astrophys. Sp. Phys. Rev., 12, l.
47. Fender et al. 1999, ApJ, 518, 165
48. Fender,R & Belloni,T., 2004, Ann. Rev. Astron. & Astrophys., 42, 317.
49. Fender,R., Belloni,T. & Gallo,E., 2005, AP&SS, 300, l.
50. Fukue J., 1987, PASJ, 39, 309.
5l. Jin L., Arnett W. D., Chakrabarti, S. K., 1989, 336, 572
52. KafatosM., RamosE., BeckerP., SubramanianP. & YangR, 1996, ASPC, 110, 384.
53. Kovalenko 1. G. & Lukin D. V., 1999, Astron. Let., 25, 215.
54. KrolikJ.H. & HawleyJ., 2006, Proceedings of the VI Microquasar Workshop: Micro-
quasars and Beyond, 46.
55. KrolikJ.H., HawleyJ.F., Hirose,S., 2007, in Triggering Relativistic Jets (Eds. William
H. Lee & Enrico Ramrez-Ruiz), RevMAA, 27, l.
56. KuznetsovS. et al. 1997, 292, 651
57. LaurentP. & TitarchukL., 1999, ApJ, 511, 289.
58. Le T. & Becker P. A., 2005, ApJ, 632, 476.
59. Liang E. P. T., Thompson K. A., 1980, ApJ, 240, 27l.
60. Lightman A. P. & Eardley D. M., 1974, ApJ, 187, 1
61. LingJ.C., WheatonWm.A., Wallyn,P., MahoneyW.A., PaciesasW.S., HarmonB.A.,
1096

FishmanG.J., ZhangS.N. & HuaX.M., 1997, ApJ, 484, 375.

62. MainD.S., SmithD.M., HeindIW.A., SwankJ.H., 1999, ApJ, 525, 901
63. Mandai,S. & ChakrabartiS.K, 2005, A&A, 434, 839.
64. Markoff, S. & Nowak, M., 2004, ApJ, 609, 972
65. Matsumoto R., Kato S., Fukue J., Okazaki A. T., 1984, PASJ, 36, 71.
66. McConnellM.L. et al 2000, ApJ, 543, 928.
67. MolteniD., LanzafameG., ChakrabartiSK, 1994, ApJ, 425, 161.
68. MolteniD., RyuD., ChakrabartiS.K, 1996, ApJ, 470, 460.
69. MolteniD., SponholzH. & ChakrabartiS.K, 1996, ApJ, 457,805.
70. Mondal,S. & ChakrabartiS.K, 2006, MNRAS, 371, 1418.
71. MorganE.H., RemillardR.A., GreinerJ., 1997, ApJ, 482, 993.
72. MunoM.P., MorganE.H., RemillardR.A., 1999, ApJ, 527, 321.
73. MunoM.P., RemillardR.A., MorganE.H., WaltmanE.B., Dhawan V., HjelimingR.M.
& PooleyG., 2001, ApJ, 556, 515.
74. Nandi,A., Chakrabarti,S.K, Vadawale,S.V. & RaoA.R., 2001, A&A, 380, 245.
75. Nobuta K & Hanawa T., 1994, PASJ, 46, 257.
76. Nowak, M. 2003 in New Views on Microquasar, (Eds.) P. Durouchaux, et al., (Centre
for Space Physic:Kolkata)
77. Ohkawa,Y., Kitamoto,S. & Kohmura,T., 2005, ApJ, 621, 9510.
78. OkudaT., TeresiV., ToscanoE. & MolteniD., 2004, PASJ, 56, 547.
79. Okuda, T., Teresi, V. & Molteni, D., this volume (astro-ph/0703465)
80. Paczynski B. & Wiita P. J., 1980, A&A, 88, 23.
81. Paczynski B. & Schwarzenberg-Czerny A., 1980, Acta Astronomica, 30, 127.
82. PaulB., AgrawaIP.C., RaoA.R., VahiaM.N., YadavJ.S., SeethaS. & KasturiranganK,
1998, ApJ, 492, 63.
83. RaoA.R., YadavJ.S. & PaulB., 2000, ApJ, 544, 443.
84. ReillyKT. et al. 2001, ApJ, 561, 183.
85. RemillardR.A. & McClintockJ.E., 2006, ARA&A, 44, 49
86. RobertsT.P., Warwick, R.S., Ward, M.J., Goad, M.R., Jenkins, L.P., 2005, MNRAS
357, 13
87. RodriguezJ., Corbel, S., Kalemci, E., Tomsick, J.A., & Tagger, M. 2004, ApJ, 612,
988
88. Ryu D., Brown G. L., Ostriker J. P. & Loeb A., 1995, ApJ, 452, 364.
89. RyuD, ChakrabartiS.K & MolteniD., 1997, ApJ, 474, 378.
90. ShraderC. & TitarchukL., 1998, ApJ, 499, 31.
91. SmithD.M., HeindIW.A., SwankJ.H., 2002, ApJ, 569, 362.
92. SmithD.M., HeindIW.A., SwankJ.H., MarkwardtC.B., 2001, ApJ, 554, 41.
93. SriramK, AgrawaIV.K, PendharkarJ.K, RaoA.R., 2007, astro-ph/0703074
94. StrohmayerT.E., ZhangW., SwankJ.H., SmaleA., TitarchukL., DayC. & LeeU., 1996,
ApJ, 469,9.
95. Takahashi M., 2000, in Proceedings of the Astronomical Society of the Pacific, (San
Francisco), 233.
96. Titarchuk, L.G. & Shaposhnikov, N., 2005, ApJ, 626, 298
97. Ferinelli, R., Titarchuk, L.G. & Frontera, F., 2007, astro-ph/072624
98. TomsickJ., 2001, ApJ, 563, 229
99. Trudolyubov, S., 2001, ApJ, 558,276
100. Vadawale,S.V., Rao,A.R.,
Naik,S., Yadav,J.S., Ishwara-Chandra,C.H., PrameshRao,A. & Pooley,G.G., 2003,
ApJ, 597, 1023
101. Wardzinski, G. & Zdziarski, A.A. 2001, MNRAS 325, 963
 1097

102. WangJ.-M., NetzerH., 2003, A&A, 398, 927.

103. Yu C., Lou Yu-Qing, Bian Fu-Yan, Wu Y., 2006, MNRAS, 370, 121.
104. Yang R. & Kafatos M., 1995, A&A, 295, 238.
105. Yuan F., Dong S.& Lu Ju-Fu, 1996, AP&SS, 246, 197.
106. ZhangS.N. et aI., 1996, A & A Supp. 120, 227.
107. ZhangW., StrohmayerT.E., SwankJ.H., WhiteN.E. & Lapidus!., 1997, BAAS, 29,
1370
108. ZhangS.N., CuiW., ChenW., HarmonB.A., RobinsonC.R., SunX., YaoY., ZhangX.,
2000, Science, 287, 1239
109. Zdziarski, A.A., 2000, in proceedings of lAU symposium No. 195, Eds. P.C. Martens
et al. (San Francisco - Astronomical Society of the Pacific), 153
1l0. Zdziarski, A.A. & Gierlinski, M., 2004, Prog. Theo. Phys. Suppl., 155, 99
 SECONDARY PERTURBATION EFFECTS IN KEPLERIAN
ACCRETION DISKS: ELLIPTICAL INSTABILITY

BANIBRATA MUKHOPADHYAY
Department of Physics,Indian Institute of Science, Bangalore-560012, India
[email protected]

Origin of turbulence in cold accretion disks, particularly in 3D, which is expected to

be hydrodynamic but not magnetohydrodynamic, is a big puzzle. While the flow must
exhibit some turbulence in support of the transfer of mass inward and angular momentum
outward, according to the linear perturbation theory it should always be stable. We
demonstrate that the 3D secondary disturbance to the primarily perturbed disk which
exhibits elliptical vortices into the system solves the problem. This result is essentially
applicable to the outer region of accretion disks in active galactic nuclei where the gas
is significantly cold and neutral in charge and the magnetic Reynolds number is smaller
than 104 .

1. Introduction
Despite much effort devoted, the origin of hydrodynamic turbulence in Keplerian
accretion disks is still poorly understood. This is essentially important for accre-
tion disks around quiescent cataclysmic variables, proto-planetary and star-forming
disks, and the outer regions of disks in active galactic nuclei. 1
A Keplerian accretion disk flow having a very low molecular viscosity must
generate turbulence and successively diffusive viscosity to support transfer of mass
inward. However, theoretically this flow never exhibits any unstable mode. On the
other hand, laboratory experiments of Taylor-Couette systems, which are similar to
Keplerian disks, seem to indicate that although the Coriolis force delays the onset
of turbulence, the flow is ultimately unstable to turbulence. 2 However, some other
experiments say against it. 3 We believe that not finding hydrodynamic turbulence
is due to their choice of large aspect ratio and small Reynolds number.
Various kinds of secondary instability, such as the elliptical instability, are widely
discussed as a possible route to self-sustained turbulence in linearly perturbed shear
flows (see, e.g. 4 ). These effects have been proposed as a generic mechanism for the
breakdown of many two-dimensional high Reynolds number flows. However, such
effects have not been discussed properly in literatures for rotating Keplerian flows.
Therefore, we plan to show that the three-dimensional secondary perturbation
can generate large growth in the flow time scale and presumably trigger turbulence
in the Keplerian disk. Unlike the two-dimensional transient growth studied earlier 5
which were shown to be killed immediately in presence of vertical structure, in
the present case we demonstrate essentially three-dimensional growth. Possibility
of large growth in shear flows with 'rOtation by a three-dimensional perturbation
opens a new window to explain hydrodynamic turbulence which has an important
implication not only in astrophysics but also in general physics and fluid dynamics.

1098
 1099

2. Perturbation equations
We consider a small section of the Keplerian disk in the shearing box approximation 5
with the unperturbed velocity Op = (0, -x, 0). We also assume that the incompress-
ible flow extends from x = -1 to + 1 with no-slip boundary conditions and is un-
bounded with periodic boundary condition along y and z. Therefore the linearized
N avier-Stokes equations for a 2D primary perturbation such that Ur; --> Wx (x, y, z, t),
U: --> UP(x) + Wy(x, y, z, t), and pressure ji --> ji + p(x, y, z, t) are given by
dw x 8p 1 2 dwy 8p 1 2
-d
t
= 2Dw y - -8
x
+ -Re \7 WX, -
dt
= D(q - 2)w x - -8
y
+ -\7
Re
wy , (1)

with the equation of continuity 8wx/8x + 8wyl8y = 0, where the angular frequency
D = l/q and q = 3/2 for a Keplerian disk. When the Reynolds number Re is very
large, the solution of above equations is
_ ky . ( , _ kx . ( 2 _ 2
Wx - (f2 sm kxx + kyy), Wy - -( f2 sm k"x + kyy), l - kx + ky,2 (2)
where ( is the amplitude of vorticity perturbation. Under this primary perturbation,
the flow velocity and pressure modify to

O=Op+w=(Wx,-x+wy,O)=A.J, p'=ji+p(X,y,z,t), (3)

where d~ is the position vector and A is a tensor of rank 2.

Now we concentrate on a further small patch of the primarily perturbed flow
such that x < < 11 kx, y < < 11 k y. Then we consider a secondary perturbation to
this flow such that 0 --> 0 + il and pi --> pi + P with

(u;,p) = (Vi(t),p(t)) exp(ikm(t)x m ), i, m = 1,2,3. (4)

Therefore, we obtain the evolution of a linearized secondary perturbation
.
Vj + Akj Vk + 2 ErnkjH"m V k = -1P
. k
j -
Vj
-
k.2 , (5)
Re

(6)

3. Solution
We specifically concentrate on the How having low viscosity. Therefore, the general
solution6 of eqn. (5) can be written as a linear superposition of the Floquet modes

Vi(t) = exp(lTt) !i(¢), (7)

where ¢ = wt, fi(¢) is a periodic function having time-period T = 27rlw, and IT
is the Floquet exponent which is different at different E = (k x ll)2. Clearly, if IT is
positive then the system is unstable and plausibly turbulent.
When the secondary perturbation evolves much rapidly than the primary one
and ki(O) = (0,0,1), IT has an analytic solution given by

(8)
1100

Clearly, for a Keplerian disk (J = VCE - (4 - 3()/9. Therefore, a Keplerian flow is

hydrodynamically unstable under a vertical secondary perturbation if C> 1/3. For
other perturbations (J can be computed numerically described in detail elsewhere. 6
We now plan to quantify this by computing the corresponding turbulent viscos-
ity. For the isotropic disk fluid, the turbulent viscosity Vt = aC s h,7 where Cs and h
are local sound speed and disk thickness respectively. On the other hand, shearing
stress Txy =< Ul U2 >= -Vtq0,. Therefore, we obtain the Shakura-Sunyaev viscosity
parameter at a disk radius r
Txy 0,x
a= - where M=-. (9)
q0,2 (~) 3 M r2 Cs

For a disk with C = 0.35, kx = 3, ky = 0.7, a vertical secondary perturbation

evolving for time trn = 10 at a disk radius r = 15 and thickness h(r)/r = 0.05,
a ~ 0.02 (the detailed discussion is reported elsewhere 8 ). This is interesting as a
due to MRI computed by previous authors 9 is similar order of magnitude.

4. Discussion
Above results verify that the three-dimensional growth rate due to the secondary
perturbation in a Keplerian disk could be real and positive and corresponding
growth may be exponential and significant enough to trigger elliptical instability.
This eventually may trigger non-linearity and then plausible turbulence in the flow
time scale. As this growth is the result of a three-dimensional perturbation, underly-
ing perturbation effect should survive even in presence of viscosity. We also see that
the corresponding viscosity a to transport matter inward and angular momentum
outward is significant and comparable to that due to MRI.

References
1. C. Gammie, & K. Menou, Astrophys. J. 492, L75 (1998); O. Blaes, & S. Balbus,
Astrophys. J. 421, 163 (1994); K. Menou, & E. Quataert, Astrophys. J. 552, 204
(2001).
2. D. Richard, & J.-P. Zahn, Astron. Astrophys. 347, 734 (1999).
3. H. Ji, M. J. Burin, E. Schartman, & J. Goodman, Nature 444, 343 (2006).
4. R. Pierrehumbert, Phys. Rev. Lett. 57, 2157 (1986); B. Bayly, Phys. Rev. Lett. 57,
2160 (1986); A. Craik, J. Fluid Mech. 198,275 (1989).
5. B. Mukhopadhyay, N. Afshordi, & R. Narayan, Astrophys. J. 629, 383 (2005); N.
Afshordi, B. Mukhopadhyay, & R. Narayan, Astrophys. J. 629, 373 (2005).
6. B. Mukhopadhyay, Astrophys. J. 653, 503 (2006).
7. N. Shakura, & R. Sunyaev, Astron. Astrophys. 24, 337 (1973).
8. K. Saha, & B. MUkhopadhyay, "Turbulent viscosity in the Keplerian accretion disks
due to hydrodynamic instability" (submitted).
9. A. Brandenburg, A. Nordlund, R. F. Stein, & U. Torkelsson, Astrophys. J. 458, L45
(1996); J. M. Stone, J. F. Hawley, C. F. Gammie, & S. A. Balbus, Astrophys. J. 463,
656 (1996). J. C. B. Papaloizou, & R. P. Nelson, Mon. Not. Roy. Astron. Soc. 339,
983 (2003).
 GRAVITATIONAL COLLAPSE OF POPULATION III STARS

YUDAI SUWA1, TOMOYA TAKIWAKIl, KEI KOTAKE 2 and KATSUHIKO SAT0 1,3
1 Department of Physics, School of Science, The University of Tokyo, Tokyo 113-0033
2 National Astronomical Observatory of Japan, Mit aka, Tokyo 181-8588,Japan
3 Research Center for the Early Universe, School of Science, the University of Tokyo,
7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
[email protected]

We have simulated the collapse of rotating Population III stars with numerical simula-
tion and extracted the gravitational waves from the mass motions via the quadrupole
formula. In addition, we estimate the gravitational wave emission by anisotropic neutrino
radiation. With these wave forms, we calculate the gravitational wave background from
Population III stars. As a result, we have found that the gravitational wave background
from Population III stars is large enough to be detected by interferometers such as BBO
and DECIGO without correlation.

1. Method
The numerical methods employed in this paper are essentially the same as those
used in our previous paper. 1 We employed the ZEUS-2D code 2 as a base and added
major changes to include the microphysics. First we added an equation for electron
fraction to treat electron captures and neutrino transport by the so-called leakage
scheme. 3 Furthermore, we extend the scheme to include all 6 species of neutrino
(v e , De, vx), which is indispensable for the computations of the Pop III stars. Here
Vx means v IL ' DIL , V T and DT • As for the equation of state, we have incorporated the

tabulated one based on relativistic mean field theory.4 Spherical coordinates (r, e)
are employed with logarithmic zoning in the radial direction and regular zoning
in e. One quadrant of the meridian section is covered with 300 (r)x 30 (e) mesh
points. In our 2D calculations, axial symmetry and reflection symmetry across the
equatorial plane are assumed.
The initial condition are provided in the same manner of previous paper. 1 The
supposed rotation law is cylindrical rotation, which the strength of rotational energy
is 0.5% of gravitational energy in all models. We calculate some mass range from
300M8 to lOOOM8 .

2. Gravitational Wave of a single Population III Star Collapse

In this section, we consider the gravitational wave (GW) emission of a single 300M8
Population III star collapse. In this study, we estimate the GW from aspherical mass
motions in our models via the Newtonian quadrupole formalism. 5 In addition, we
compute the GW strain from anisotropic neutrino emission. 6 ,7
Figure 1 depicts the strain versus time after bounce of 300M8 star. First, the
matter contribution to changes in hTT dominates during first 70 ms. Afterward,
the anisotropic neutrino radiation begins to contribute because the thermal shock
occurs and the neutrino luminosity increases in the hot region, backward of the

1101
1102

o Total
() 10- 18 b---c~~
0.
S -1
-~
Neutrinos
Malter "'a"
"'"
!: - Tolal ~ 10- 19
is -2
I",
-3 <;::< 10- 20

-4 L-~L-~__~__~__~__~
-0.1 o 0.1 0.2 10' 10' 10'
Time[sec] f [Hz]

Fig. 1. The gravitational wave strain,h TT, Fig. 2. GW source spectra. The solid line is
times the distance to the Population Ill, D, the total GW, the dotted line is the neutrino
versus time after bounce (in seconds). The GW, and the dashed line is the matter GW
neutrino, matter, and total wave forms are for model 300M0 at D=lOkpc.
plotted with dotted, dashed, and solid lines,
respectively.

shock surface. 120 msec after bounce, the neutrino contribution converges to a
constant value and the matter contribution dumps to zero. This is due to the black
hole formation.
Figure 2 shows the Fourier transformation of the GW signal. The solid line,
which means total spectrum of signal, is larger than the wave form assumed in
the previous works 8 ,9 in low frequency region. The dotted line means the neutrino
contribution, which is dominant in the low frequency region, meanwhile, the dashed
line means the matter contribution, which dominates the high frequency region.
Current numerical simulations are done for 0(1) sec and do not cover the strain
spectrum below fractions of Hertz. However, for the low frequency region, we can
deduce the GW signal by applying the zero-frequency limit. 6

3. Gravitational Wave Background from Population III Stars

We are now in a position to discuss the contribution of GWs from the Population
III stars to the background radiation. The sum of the energy densities radiated by
a large number of independent Population III stars at each redshift is given by the
density parameter OGw(J) == p;;l(dPGw/dlogf) aslO
2
OGw(J) = 167f cD2 J ~ I dt I Jdrn ¢(rn)~(z)f'3Ih(J'W, (1)
15G N pc 1 + z dz

where Pc is the critical density (3H6/(87fG)), ¢(m) is the initial mass function
(IMF) of Population III, ~(z) is the star formation rate (SFR), and l' is the red
shifted frequency, (1 + z)f. We employ the model 2b of Sandick et al. (2006)9 as
the SFR, which is for the very massive stars, from 270M8 to 500M8 . As for the
IMF, we employ the same parameterization with Sandick et al. (2006)9 as ¢(m) ex
J
m- 2 . 3 , which is normalized by dm¢(m) = 1. The cosmological model enters with
Idt/dzl = [(1 + z)H(Z)]-l and, for a flat geometry, H(z) = HO[OA + Orn(l + z)3]1/2.
 1103

10-' .' ,
... I
GWB from,"Pop,1Ii
10- 10
. ,
,
,,
10- 12
r; ,
c: 10-1< ,
- - ~f~t:n~~E~/7 - - - - - --
~O
10- 16

lO- lB ultimate-DECIGO
(S/N~5)

10-20 '------'-~lillL~~"'--~llilll_'_~=----'-'-.LU.ll"---..L.::I
10-3 10- 2 10- 1 1 10 1
f [Hz]

Fig. 3. The energy density parameter of GWB. The horizontal dashed line shows the GWB from
the inflationary epoch constrained by COEE observations.

We use the parameters DA = 0.73, Dm = 0.27, and Ho = 100 hokm S-l Mpc- 1 with
ho = 0.71.
In Figure 3, the calculated DGW is plotted with the sensitivity curves of the
future detectors. The gravitational wave background (GWB) is broadly distributed
with peak amplitude h6DGW rv 10- 10 near 20 Hz. The GWB from Population III
can give a particularly contribution, masking completely the GWB generated in the
inflationary epoch constrained by COEE observations (the horizontal dashed line).
It can be seen that the GW from the anisotropic radiation of neutrinos, which are
dominated low frequency region, seem within the detection limit of the planning
detectors BBO and DECIGO without correlation.

References
1. Y. Suwa, T. Takiwaki, K. Kotake and K. Sato, submitted to PASJ (2007).
2. J. M. Stone and M. L. Norman, ApJS 80,753 (1992).
3. K. Kotake, S. Yamada and K. Sato, Phys. Rev. D68, 044023 (2003).
4. H. Shen, H. Toki, K. Oyamatsu and K. Sumiyoshi, Nucl. Phys. A637, 43.'") (1998).
5. R. Monchmeyer, G. Schaefer, E. Muller and R. E. Kates, Astron. Astrophys. 246, 417
(1991).
6. R. Epstein, ApJ 223, 1037 (1978).
7. E. Muller and H.-T. Janka, Astron. Astrophys. 317, 140 (1997).
8. A. Buonanno, G. Sigl, G. G. Raffelt, H.-T. Janka and E. Muller, Phys. Rev. D72,
084001 (2005).
9. P. Sandick, K. A. Olive, F. Daigne and E. Vangioni, Phys. Rev. D73, 104024 (2006).
10. E. S. Phinney, astro-ph/0108028 (2001).
 NEAR-INFRARED OBSERVATIONS OF SAGITTARIUS A *
S. TRIPPE, T. PAm/lARD * , S. GILLESSEN, T. OTT, F. EISENHAUER, F. MARTINS,
R. GENZELt
Max-Planck-Institut fur- extmter-r-estr-ische Physik, Postfach 1312, D-85741 Gar-ching, Ger-many
[email protected]

We present and discuss recent near infrared observations of Sagittarius A *, the supermas-
sive black hole in the centre of our Galaxy. Since 2003 this source has been observed pho-
tometrically, polarimetrically, and spectroscopically in H, K and L bands (1.5-4.1f.Lm).
The emission shows up in form of outbursts occuring few times per day and lasting typ-
ically 1-3 hours. These "flares" show quasi-periodic sub-structures at timescales of ~ 15
minutes. They are significantly polarised and show a variable spectral colour index. All
in all, the observations point towards decaying plasma hotspots orbiting the black hole
on relativistic orbits as a source of NIR synchrotron emission.

1. Introduction
The centre of our Milky Way hosts the 3.6-million-.iVleJ supermassive black holc 1
and radio source Sagittarius A* (SgrA*). Discovered in radio in 1974 2 it has since
then been observed extensively in a variety of wavelength bands ranging from x-
ray to radio. This black hole is generally invisible ill NIH wavelengths and was not
detected in this spectral range before 2002 when diffraction-limited observations at
optical 8-m-class telescopes became possible 3 ,4 (see Fig. 1 for an example). In the
last years we have continued and extended our near-infrared (NIH) observations in
order to understand the underlying physical emission mechanism; the main results
of these efforts are presented here.

2. Tracing the emission: Observations

Since 2002 we have regularly carried out observations on the 8.2-m Unit Telescope
(UT) 4 (Yepun) of the ESO Very Large Telescope (VLT) on Cerro Paranal, Chile.
Photornetr-ic observations were obtained using the 1024x 1024-pixel NIH camera
system NAOS/CONICA 5 ,6 (NACO). The diffraction-limited (resolutions of 40-60
mas) NACO images were obtained in Hand K bands. Polar-irnetr-ic observations
were carried out using NACO with a Wollaston prism separating the infalling light
into ordinary and extraordinary beams of perpendicular polarisations. Please note,
that throughout this article only linear polarisation is discussed. Spectmscopic data
were collected using SINFONI, an adaptive optics assisted integral field spectrom-
eter,7,8 in K band. SINFONI produced diffraction-limited data cubes with 64x32
pixels in the two spatial axes and 2048 pixels in the spectral axis.
All in all, these observations uncovered four main characteristics of the NIH
emission from SgrA *, out of which three arc illustrated in Fig. 2 using the example

'Present adress: Observatoire de Paris - Section de Meudon, 5 Place Jules Janssen, F-92195
Meudon Cedex, France
t Also: Department of Physics, University of California, CA 94720, Berkeley, USA

1104
 1105

Fig. 1. A flare from 8gr A * observed in April 2004 in H band. The position of 8grA * is encircled;
additionally, the photometric comparison stars 82 and 87 used in Fig. 2 are labeled. The time
difference between these two images is 44 minutes.

of an observation obtained in May 2006.

(1) Sgr A * emission is flaring. In NIR wavelengths it is regularly detected in
form of outbursts. These flares correspond to an increase of flux by factors up to
10 from the background level within some ten minutes. The typical length of a
r..;

flare is in the range of 1~-3 hours. The flare event rate (i.e. the nnmber of flares
per time) is in the order of few (about 3) events per day. In four cases NIR and
X-ray flares were observed to be simulaneous within the available time resolutions
(few minutes).9 Empirically (and within the limits of low-number statistics), these
outbursts show a general trend: flares are the more seldom, the more luminous they
are.
(2) SgrA * flares show a quasi-periodic sub-structure on time scales of minutes.
Quasi-periodic signals in NIR flares have now been found in the range of 13-30
minutes. The respective flare lightcurves show characteristic structures: an overall
profile (rise, maximum, decay) lasting about 1~~~2 hours is repeatedly modulated
in cycles of 13-30 minutes. This sub-structure is generally quite weak indeed
the "double peak" of the flare shown in Fig. 2 is the strongest case (in terms of
amplitude) seen so far - and detected only in a minority of all observed flares.
(3) SgrA * emission is polaTised. For three NIR flares so far observed polarimet-
rically we found polarisation degrees of 15-~20% and angles of 80° on sky at times
r..;

of maximum fluxes. Concerning the observed polarisation angle, it is important to

note that this angle was found repeatedly in three measurements covering a time
span of two years. This strongly suggests that the geometry of the emission region
is stable in time.
(4) SgrA * flares show a featuTeles8 spectTurn with colour indices a -3 ... + 2
(defined via v Lv ex va). The flare colour varies with the source flux: the stronger
the flare, the bluer the colour. lO

3. Understanding the physics: Interpretation

Using the observations described above as well as information gathered during the
last years from radio and X-ray observations, we see strong indication that the flare
1106

o deg
45 deg
90 deg

400 420 440 460 480 500 520

time [minutes]

Fig. 2. Emission from 8grA* as observed on May 31st, 2006. 8hown here are the light curves of
8grA * and two omparison stars (82 and 87) separately for four polarisation angles. The values for
82 and 87 are shifted along the flux axis. This data set nicely summarises three main characteristics
of the 8grA * NIR activity: (1) The emission occurs in form of an outburst lasting for at least 80
min and then vanishes again; (2) there is a short-time modulation of the flux on a time scale of ~15
min; and (3) the flare intensity depends on the polarisation angle, i.e. the emission is polarised.

emission in SgrA * is synchrotron radiation from material orbiting the black hole. 11 ,12
In this picture, a hot ("-' 10 12 K), relatively small (R < 0.3Rs) plasma bubble arises
from the accretion disk few times per day. This can occur due to violent stochastic
processes like magnetic reconnect ion or infall of matter. The plasma bubble orbits
SgrA * on or close to the innermost stable circular orbit, cools, and gets sheared along
its orbit due to tidal disruption. After few revolutions (each lasting "-'15-25 min)
the plasma has cooled down and the flare vanishes. Recent numerical simulations
based on this scenario are able to reproduce the observations well. 13

References
1. Schadel, R., et al. 2002a, Nature, 419, 694
2. Balick, B. & Brown, R.L. 1974, ApJ, 194, 265
3. Genzel, R., et al. 2003, Nature, 425, 934
4. Ghez, A., et al. 2004, ApJ, 601, L159
5. Rousset, G., et al. 2003, SPIE, 4839, 140
6. Lenzen, R., et al. 2003, SPIE, 4841, 944
7. Eisenhauer, F., et al. 2003a, SPIE, 4814, 1548
8. Bonnet, H., et al. 2003, SPIE, 4839, 329
9. Eckart, A., et al. 2006a, A&A, 450,535
10. Gillessen, S., et al. 2006, ApJ, 640, 163
11. Meyer, L., et al. 2006, A&A, 460, 1, 15
12. Trippe, S., et al. 2006, MNRAS, in press
13. Paumard, T., et al., in preperation
 LONG-TERM MONITORING OF THE HARD
X-RAYjGAMMA-RAY EMISSION FROM GALACTIC BLACK
HOLES WITH BATSE

G. L. CASE*, E. ANZALONE, M.L. CHERRY, and J. C. ROm

Department of Physics and Astronomy, Louisiana State University,
Baton Rouge, LA 70803, USA
* [email protected]

J. C. LING, R. G. RADOCINSKI, and D. WELLS

Jet Propulsion Laboratory, California Institute of Technology,
Pasadena, CA 91109, USA

W. A. WHEATON
Infrared Processing and Analysis Center, California Institute of Technology,
Pasadena, CA 91125, USA

The BATSE instrument aboard the Compton Gamma Ray Observatory monitored nu-
merous astronomical sources, including the Galactic black holes Cygnus X-I, GRO
J0422+32, GRO JI719-24, GRS 1915+105, and GRO JI655-40, almost continuously
from 1991 to 2000 in the energy range 23-1798 keY. We present results for these five
black holes from the full nine-year BATSE earth occultation database using the JPL
data analysis package EBOP (Enhanced BATSE Occultation Package). Light curves in
four broad energy bands with I-day resolution are presented showing flux histories for
each source over the nine years. Spectra during the high flux/outburst states and low
flux/quiescent states for these objects are also presented. The differences in the spectra
suggest that there may two different classes of Galactic black holes.

Keywords: Black holes; Gamma-ray observations; Cygnus X-I; GRO J0422+32; GRO
JI719-24; GRO J1655-40; GRS 1915+ 105

1. Discussion
The BATSE instrument aboard the Compton Gamma Ray ObseTvatoTY has been
used to monitor a number of known gamma-ray sources in the energy range 23- 1798
keY over a 9-year period from 1991 to 2000 using the earth occultation technique.
The JPL Enhanced BATSE Occultation Package (EBOP)l has been used to gen-
erate light curves with I-day resolution as well as time-resolved spectra for the five
Galactic black holes Cygnus X-I, GRO J0422+32, GRO 11719-24, GRS 1915+105,
and GRO JI655-40. Light curves and spectra shown here differ from previous re-
sults 2- 5 due to the removal of days in the present analysis where there is source
confusion, and also the revision of the edges of the energy bins, where now the most
up-to-date instrument response data are used. Figures 1-5 show the EBOP light
curves in four broad energy bands as well as spectra during the high/outburst states
and low/quiescent states for these five objects.

1107
1108

10- 1
IGH '(1,) STATE VS LOW (1.~'S~A~
10- 2 i

1I
;=2.51
COMPTON Model

~lO 3
kT = 51.1 keY
I
'til 10- 4

':5
~ 10- 5
o
:g
~10-6
x
o
;;::
10- 1

100 1000
Energy (keV)
TJD (Days)

Fig. L Cygnus X-1light curves and spectra for high (2) and low ho) states, with the COMPTEL
spectral data overlaid for comparison,

23-98 keY

TJD (Days)

Fig, 2, GRO J0422+32 light curves and spectra for outburst and low states.

:::t....,"" - .~ '. ---~~~

23-98 keY

HIGH STATE vs LOW STATE

eoce -.;...--~~ . 10- 2

~- .. ----. ,----'----,

..
'S.
~ 10- 3

.
':'S 10- 4
C
o
:g 1 0- 5
~
o
X Li: 10-6 • Low Stote
::J
G:: o High Stote

100 1000
TJD (Days) Energy (keV)

Fig. 3, GRO J1719-24 light curves and spectra for outburst and low states,
 1109

HIGH STATE vs. LOW STATE

10- 2

~ ~ BROKEN POWER LAW

;. J /~r:Qt~B= 282.90
'E
.
~ 10-

'i'~ 10- 4
r = 1.62

u
.
j:~::(~~~W.'.i" -~~~
C
o
:g 10-5
~
~ a~o'IG' yco_,r,' 9jL','O' 'OOKI{j' o
G: 10- 6 r", 2.17

• LOW STATE
1 0-7 OHIGH STATE

100 1000
TJD (Days) Energy (keV)

Fig. 4. GRS 1915+105 light curves and spectra for sum of high and low states.

HIGH STATE VS lOW STATE

~:::I.:~ r", 2 79

(/)
E1J '\
i : I\;
c:
o E 10-5 r = 2.34
-0
.s:: ~
~
x o HIGH State
:I
G: • Low Stote

10-7L_~
100 1000
Energy (keV)

Fig. 5. GRO J1655-40 light curves and spectra for sum of outburst and quiescent states.

When in the high flux state, Cygnus X-I, GRa J1719-24, and GRa J0422+32
show Comptonized spectra below rv 250 keY, while GRS 1915+ 105 and GRa .J1655-
40 show single power law or broken power law spectra. In all sources except GRa
JI655-40, a hardening of the spectrum above rv 250 keY is observed during most
high flux states. All five sources show similar hard power law spectra when in the low
flux state. This suggests that there may be at least two different classes of Galactic
black holes based on their spectra when in the high state: one characterized by a
Comptonized spectra below rv 250 keY and one characterized by a power law.

References
1. J. C. Ling et al., ApJS 127, 79 (2000).
2. M. L. McConnell et ai., ApJ 857, 984 (2002).
3. J. C. Ling and W. A. Wheaton, ApJ 584,399 (2003).
4. J. C. Ling and W. A. Wheaton, ApJ 622,492 (2005).
5. G. L. Case, M. L. Cherry, C. A. Fannin, J. C. Rodi, J. C. Ling and W. A. Wheaton,
Chinese J. Astron. CJ Astroph. 5, 341 (2005).
 MARGINALLY STABLE THICK DISCS
ORBITING KERR-DE SITTER BLACK HOLES*

PETR SLANY & ZDENEK STUCHLIK

Institute of Physics,
Silesian University in Opava,
Bezrucovo nam. 13, 74601 Opava, Czech Republic
petro [email protected], zdenek. [email protected]

Basic properties of equipotential (equipressure) surfaces in test barotropic perfect fluid

tori with uniform distribution of the specific angular momentum orbiting Kerr-de Sitter
black holes are summarized. The central mass-densities of adiabatic non-relativistic tori,
for which the approximation of test fluid is adequate, are given and compared with the
typical densities of Giant Molecular Clouds.

Presence of a repulsive cosmological constant, A > 0, changes the asymptotic struc-

ture of the black-hole backgrounds being de Sitter, not flat, and implies strong
consequences for the structure of thin and thick discs around Schwarzschild-de Sit-
ter (SdS) and Kerr-de Sitter (KdS) black holes, see, e.g., the review of Stuchlfk. 1
KdS spacetimes are characterized by three parameters, M, a, y, representing the
mass and spin of the black hole, and the dimensionless "cosmological parameter"
defined in geometrical units (c = G = 1) as y = AM 2 /3. For simplicity, we put
M = 1 hereafter. Gravitational attraction of the black hole is just compensated
by the cosmic repulsion at the "static radius", the only radius where the static
geodesic observers reside: rs = y-1/3 in the equatorial plane (8 = 7r/2). Stationary
discs exist only in the spacetimes admitting stable circular geodesics. Beside the
inner marginally stable (ms(i)) and the inner marginally bound (mb(i)) circular
geodesics located near the black-hole horizon, there are the outer marginally stable
(ms(o)) and marginally bound (mb(o)) circular geodesics near the static radius. 2
Basic properties of geometrically thick discs are determined by equilibrium con-
figurations of barotropic test perfect fluid orbiting a black hole. Marginally stable
thick discs are characterized by uniform distribution of the specific angular mo-
mentum, f(r, 8) == -U'P/Ut = const a . Solving the Euler equation for the fluid with
4-velocity U,., = (U t , 0, 0, U'P), the surfaces of constant pressure coincide with the
equipotential surfaces, W = const, given by the relation 3

~ Do,.Doe sin 82 ]1/2

W(r, 8) = InUt(r, 8) = In 2"
I [ Doe sin 8 (r2 + a - af)2 - Do,. (f - asin 2 8)
2 2
2 '

*This research has been done as a part of the Research Project MSM 4781305903.
a Angular velocity 12 = U'P jut is given by the relation with metric coefficients: 12 = _ ££9"
9tr.p
19t'l' .
9r.pr.p

1110
 1111

I.
(log r) sin 8 (log r) sin e (log r) sin e
(a) (b) (c)
Fig. 1. Typical behaviour of equipotential surfaces (meridional sections) in SdS and KdS black-
hole spacetimes. Light gray region contains closed equipotential surfaces. The last closed surface
is self-crossing in the cusp(s). Possible toroidal configurations correspond to: (a) accretion discs,
(b) marginally bound accretion discs and (c) excretion discs.

Boundary of any torus is given by corresponding closed equipotential surface, how-

ever, the last closed surface is critical, i.e. marginally closed, enabling the outflows of
matter from the disc through the cusp(s) due to violation of hydrostatic equilibrium,
if the critical surface is overfilled by the fluid. Stationary toroidal configurations ex-
ist only if R E (Rms(i)' Rms(o)), where Rms(i,o) is the Keplerian angular momentum on
the inner/outer ms-orbit. We can distinguish three kinds of discs (Fig. 1):

accretion discs: Toroidal equipotential surfaces are bounded by the critical sur-
face self-crossing in the inner cusp, rmb(i) < rin < rms(i), enabling outflow
of matter from the disc into the black hole. Another critical surface self-
crossing in the outer cusp is open. Matter filling the region between the
critical surfaces cannot remain in hydrostatic equilibrium and contributes
to the accretion flow along the inner cusp.
marginally bound accretion discs: Such configurations exist only for the spe-
cial distribution of R(r, B) = Rmb = const, where Rmb is the Keplerian angu-
lar momentum on mb-orbits. Toroidal equipotential surfaces are bounded
by the critical surface self-crossing in both cusps, Tin = Tmb(i), Tout = Tmb(o).
Any overfilling of the critical surface causes accretion inflow through the
inner cusp as well as the so-called excretion outflow through the outer cusp.
excretion discs: Toroidal equipotential surfaces are bounded by the critical sur-
face self-crossing in the outer cusp, T ms( 0) > rout> l' mb( 0)' enabling outflow
of matter from the disc into the outer space. The critical surface with the
inner cusp, if such surface exists, is open (cylindrical) and separated from
the critical surface with the outer cusp by other cylindrical surfaces which,
in fact, disable accretion into the black hole.

In accretion discs, in which the potential levels corresponding to the critical surfaces
are comparable, i.e., Wcrit(i) ;S Wcrit(o) , huge overfilling of the inner critical surface
1112

causing the accretion could be combined with the excretion (after overfilling of
the outer critical surface) having a capability to regulate the accretion flow and,
especially, stabilize the accretion discs against runaway instability, as shown by
Rezzolla et a1. 4 in the case of SdS backgrounds.
The repulsive cosmological constant is responsible for the existence of the outer
cusp determining the outer edge of the disc being located close to but bellow the
static radius, and strong collimation of open equipotential surfaces along the ro-
tational axis being evident near and behind the static radius. b For the current
value of the cosmological constant, Ao ;:::j 1.3 X 10- 56 cm -2, maximal dimensions of
test structures around supermassive black holes are smaller but comparable with
dimensions of large galaxies, as indicated by values of ro in Table 1.

Table 1. Mass parameter (M), the static radius (Ts) and the central mass-density (il) of
an adiabatic non-relativistic torus (with the adiabatic index r = (5/3, 7/5)), for which
mdisc "" MER, in SdS and KdS spacetimes with Ao "" 1.3 X 10-
56
cm- 2 .
M Ts SdS il5/3 il7/5 KdS il5/3 il7/5
[1118 ] [kpc] (a = 0) [kgm- 3 ] [kgm- 3 ] (a = 0.9) [kgm- 3 ] [kgm- 3 ]
106 10 10- 22 10- 13 - -

10 7 22 10- 22 10- 14 10- 21 10- 12

10 8 50 10- 22 10- 15 10- 22 10- 13
lOB 110 10- 23 10- 16 10- 22 10- 14

Relevance of test-fluid approximation can be tested by determining the central-

mass density (}e of the marginally bound accretion disc for which the masses of the
disc (with shape given by the critical equipotential surface of test-fluid limit) and
central black hole are comparable. For a non-relativistic adiabatic perfect fluid, i.e.
the fluid with equation of state P = K (}! where an adiabatic index r = 1 + lin,
the total mass-energy of the torus is given by the relation

m = 27r (}e
1[
disc
1 + Hl(r, e)] [exP{Win-W(r,e)}-l]n( r 2 +a 2 cos 2e)·
1 - {; rl(r, e) exp{Win - We} - 1
sm ed r de ,

where l¥ill (We) is the potential level on the surface (at the center) of the torus.
The results are presented in Table 1. Comparing with the typical densities of Giant
Molecular Clouds (GMC), eCMC '" 1O- 18 kgm- 3 , we see that for diatomic GMC
the limiting (}e » eCMC.

References
1. z. Stuchlik, Modern Phys. Lett. A 20, 561 (2005).
2. Z. Stuchlfk and P. Slany, Phys. Rev. D 69, 064001 (2004).
3. P. Slany and Z. Stuchlfk, Classical Quantum Gravity 22, 3623 (2005).
4. L. Rezzolla, o. Zanotti and J. A. Font, Astronomy and Astrophysics 412, 603 (2003).

bThis effect deserves further attention, as it may indicate some role of A > 0 for collimation of
jets far away from the active galaxy.
 BLACK HOLES IN SCALAR FIELD OR
QUINTESSENTIAL COSMOLOGY *

TOMOHIRO HARADA
Department of Physics, Rikkyo Unive'l'sity, Toshima, Tokyo 171-8501, Japan
[email protected]

For a stationary black hole in an asymptotically flat vacuum spacetime in the

Einsteinian gravity, the spacetime is uniquely given by the Kerr solution. This is
called the black hole uniqueness theorem. This means that under these assumptions,
the properties of the Kerr solution must be the properties of general black holes.
However, our universe is expanding and filled with matter fields. Hence, realistic
black holes would not be in vacuum. They exist in the expanding universe and hence
are not a8ymptotically flat. Moreover, they could accrete surrounding mass and
hence grow in time. The Kerr solution will give a very good approximation in some
cases but not in the other. Here, we focus on the effects of the expanding universe
filled with matter fields on black hole dynamics. We would call black holes in an
expanding universe cosmological black holes to put an emphasis on this particular
aspect of the astrophysical black holes.
Among cosmological black holes, primordial black holes have been intensively
studied because of their important roles as probes into the early universe, high
energy physics, quantum gravity and relativistic gravity in the expanding universe.
See a recent review 2 for details and references therein. It should be noted that the
primordial black holes will be born with the mass of the same order as the mass
contained within the cosmological horizon at the formation epoch.
The problem of accretion rate or growth rate of black holes in the expanding
universe was studied very long time ago. 3 This discussion based on the Newtonian
gravity was recently refined and generali8ed. 4 The mass growth rate is given by
dM GM 2
& = 41rar~CsPFRw(t) c:::' c3 t 2 ' (1)

where rA == G IvI / c~ is the accretion radius and a is a constant of order unity which
depends on the equation of state. This equation can be easily integrated. The, we
can find a critical maS8 of the same order as the horizon mass, which divides the
evolution of black hole mass qualitatively. If the initial mass is smaller than this
critical value, the accretion soon becomes insignificant. If the initial mass is equal to
the critical value, the solution is self-similar and the mass of the black hole grows in
proportion to the cosmological time due to the effective accretion. If the initial mass
is larger than the critical value, the mass accretion is so significant that the mass
of the black hole diverges in a finite time, which would be unphysical. It should be
noted that the existence of self-similar solution plays a key role for this analysis.
The general relativistic numerical simulation strongly suggests that the growth
of black holes in the universe containing a scalar field is not significant. 5 On the other

*This article is based on the collaboration 1 with H. Maeda and B. J. Carr

1113
1114

hand, the above quasi-Newtonian discussion of black hole growth has been applied
to a quintessence field, 6 a scalar field which slowly roles down its flat potential and
enables the universe to accelerate due to its large negative pressure. Then, based
on the equation similar to Eq. (1), it has been claimed that the accretion onto
black holes could be so significant in the quintessence-dominated era preceding
the radiation-dominated era that primordial black holes may provide the seeds for
supermassive black holes in the galactic nuclei.
Indeed, it was shown that there is no self-similar solution of the Einstein field
equation which contains a black hole event horizon in the expanding universe with
a dust, radiation fluid,7 a perfect fluid for p = kp (0 < k < 1)8 and a stiff fluid
(p = p).9 However, it has not yet been clear whether a self-similar black hole can
exist or not in an accelerating universe. Because the accelerating universe is strongly
favoured by the recent observational data, it would be very important to answer
this question.
The self-similar (homothetic) spacetime is defined by the existence of a vector
field ~ satisfying
(2)
We here consider a scalar field as a matter field because it very often appears in
many aspects of modern cosmology and also it provides one of the simplest and
physically reasonable model to accelerate the universe. It is remarkable that from
self-similarity assumption, the scalar field is restricted to be massless or with a
potential of the following exponential form:
V(¢) = VoeV81f.\¢, (3)
which is parametrised by),. Here we assume the potential is not negative. For the flat
Friedmann solution, the scalar field with the exponential potential can accelerate
the universe if 0 < ),2 < 2 and hence can be regarded as a quintessence model.
With the assumption of self-similarity and spherical symmetry, the line element
is given in the form:
ds 2 = -g(r/v)g(r/v)dv 2 + 2g(r/v)dvdr + r2(d8 2 + sin 2 8d¢2), (4)
in the Bondi coordinates. A similarity surface I: is defined as a surface on which
r / v=const. A similarity horizon is defined as a null similarity surface on which ~ is
also null. A similarity horizon can be identified with an event horizon or a particle
horizon.
The Einstein equation reduces to the set of ordinary differential equations with
respect to ~ = In( r / Iv I). The characteristic surfaces correspond to the singular points
of the ODEs, as do in the case of a perfect fluid of p = kp (0 < k < 1). They are
classified into nodes, saddles and foci as equilibrium points in dynamical systems
theory. Figure 1 illustrates a node and a saddle. The uniqueness may break down
at the singular points. The non-existence of black hole solution even with an exact
Friedmann exterior is therefore highly nontrivial. Moreover, the singular points also
correspond to similarity horizons for the scalar field case.
 1115

node saddle
~e.~

Fig. 1.
~
Node and saddle.

Fig. 2. Black hole in the I<riedmann universe.

We have proved the following nonexistence theorem. See l for the proof.

Theorem 1. Let a spherically symmetric self-similar spacetime with a scalar field

satisfy one of the following two conditions: (a) it coincides with the flat Friedmann
sol1tiion o'utside some finite r'adius; (b) it is asymptotic to the decelemting Friedmann
solution at spatial infinity. Then it has no black hole event horizon.

In summary, there is no self-similar black hole in a flat Friedmann universe con-

taining a scalar field or quintessence field. This strongly suggests that the accretion
onto black holes in the quintessence-dominant phase is not very effective even for
horizon-scale black holes. However, recall that we have here adopted two nontrivial
assumptions. One is that a scalar field is responsible for the acceleration of the uni-
verse and the other is that the universe is described exactly by the flat Friedmann
solution outside some finite radius for the accelerating case. It would be interesting
to relax these assumptions and see whether the conclusion changes or not.

References
1. T. Harada, H. Maeda and B. J. Carr, Phys. Rev. D74, 024024 (2006).
2. B. J. Carr, to appear in the Proceedings of "Inflating Horizon of Particle Astrophysics
and Cosmology", (Universal Academy Press Inc and Yamada Science Foundation,
Tokyo, 2005), Preprint astro-phj0511743.
3. Y. B. Zeldovich and I. D. Novikov Sov. Astron. 10, 602 (1967).
4. T. Harada and B. J. Carr, Phys. Rev. D71, 104009 (2005).
5. T. Harada and B. J. Carr, Phys. Rev. D71, 104010 (2005).
6. R. Bean and J. Magueijo, Phys. Rev. D66, 063505 (2002).
7. B. J. Carr and S. W. Hawking, Mon. Not. R. Astron. Soc. 168,339 (1974).
8. B. J. Carr, Ph.D. thesis, Cambridge University (1976).
9. G. V. Bicknell and R. N. Henriksen, Astrophys. J. 225, 237 (1978).
 A NEW SOLUTION FOR EINSTEIN FIELD EQUATION IN GENERAL
RELATIVITY

SADEGH MOUSA VI
Department of Physics, Amirkabir University of Technology
No 13, Unit 3, North Shahin St." West Hemmat Highway, Tehran, Iran
Postal Code: 14758-15835, Email: [email protected]

1. Introduction
There are different solutions for Einstein field equation in General Relativity, proposed
by different people, up to now. Of all the exact solutions, which are known, only a
limited class seems to have a real physical meaning, e.g. Schwarzschild [1], Reissner-
Nordstrom [2,3], Kerr[4] and Kerr-Newman [5]. However, these solutions are limited to
some special cases and Einstein field equation has a more complete solution which
contains all these solutions as its special forms.

2. General form ofthe new solution (Sadegh Metric)

The new solution we are considering for Einstein field equation is in the following
general form:

Where the coordinate system is chosen so that: x = [t, r,e,<p, u].

The important point here is the comparison of this metric with previous metrics
-about 130 metrics which exist in Gr-Tensor software web site [6]-. This comparison
shows that, this metric is not a coordinate transformation of any other metric. Therefore,
this metric is equation of a more comprehensive movement than a simple spinning
movement, discussed in Kerr metric [4], and it is a new metric.

3. Obtaining new metric functions

As seen in the metric equation (1), the metric functions are ( e 2V , e 21f/ ,w,w', ...... ). For
calculating the metric functions, we start from Einstein-Maxwell equations. After some
calculations on Einstein-Maxwell equations, which are governing to the space-time, one
could obtain Ernst equations [7]. With some calculations on these equations, we find
equations which determine 05, 05':
2aA(m(p2 - 2r2) + rQ2) 2acos8 sin8(2mr - Q2)(I1_ b 2 A')
m'2 = (l1-b2A' _a 2A)2 m'3= (l1-b2}.,' _a 2A)2

~, 2M'(m(p2 - 2r2) + rQ2)

(j)'2 = (l1-b2}.,' _a 2A)2

Where a comma denotes differentiation with respect to the next number and the number
e,
after comma denotes the number of coordinate: [t, r, <p, u] = [1,2,3,4,5].
From these equations it is obtained:

1116
 1117

and then:
a(2mr-Q2)
(J)=-----'---~-'-­
~2

and briefly after some calculations we find all metric functions:

e 2'1' = A~2 p2 Ji = cose

a=-
p2 t.
Ji' = COScp

n(m 2 _Q2) ill i b = jf)

v= n=- a=--'£..
n 2 +u 2 +nmu-Q2 m m m
Where m, a, b, and n are real constants and m is the Schwarzschild
mass. i rp and i e are angular momentums around q> and e respectively and ill is the
momentum in u direction.
Therefore we found all metric functions and, the complete fonn of the metric is:
d ' =[ (r' + a 'cos'o + b'cos'cp)(r' + a' + b' + Q' - 2mr) ]d ' _
s (r2 + a 2 + b 2 )' _ (a' sin' 0 +b 2 sin' cp)(r' + a' + b 2 + Q2 _ 2mr) !

[sin 'O(r' + a' + b')' -sin 'O(a' sin' 0 + b' sin' cp)(r' + a' + b' + Q' - 2mr)](E dcp _wei!)' _
r' + a 'cos'e + b'cos'cp 0

[sin' cp(r' + a' + b')' -sin' cp(a' sin',e + b', sin' cp)(r' + a' + b' + Q' - 2mr)](Ede _W'd!) , _
r' + a 'cos e + b-cos'cp

, , '0 b' , ( , Q')

r +a cos + cos cp dr' _[ nm - ]dudt
(r' +a' +b' +Q' -2mr) n' +u' +nmu-Q'
If we substitute b=O and u=O in this metric, the Kerr-Newman metric [5] will be
resulted and by substituting Q=O in that, the kerr metric will be obtained and by
substituting a=O in the obtained metric, the Schwarzschild metric [1] will be extracted.
Therefore this metric, results in previous metrics in its special fonns.
1118

4. Physical analysis of the new metric and actual movement of black holes
By analyzing the metric equation (1), it is seen that, in addition to the spinning movement
of black holes around qJ, discussed in Kerr metric [4], they have another spinning
movement around () and moreover they have a translational movement in u direction, all
at the same time.

Acknowledgments
I would like to thank Professor Y. Sobuti, for his encouragement during the work and for
the times that he spent to work on my problems and guide me.

References
l. Schwarzschild, Berl. Ber, 1916, p189.
2. Reissner, H. (1916). Ann. Physik, 50, 106.
3. Nordstrom, G. (1918). Proc. Kon. Ned. Akad. Wet., 20,1238.
4. R. P. Kerr, "Gravitational field of a spinning mass as an example of algebraically
special metrics," Phys. Rev. Lett. 11,237-238 (1963).
5. E. T. Newman, E. Couch, K. Chinnapared, A. Exton, A. Prakash and R. Torrence
"Metric of a rotating charged mass," J. Math.Phys.6,918-9 (1965).
6. [onlione]:cited on junI2,2006,http://grtensor\GRTensorIIhomepagesMainsite.htm.
7. S.Chandrasekhar, The Mathematical Theory of Black Holes (oxford u.P., New York
,1983), pp.564 -570.
 PSEUDO-KERR GEOMETRY

SOUMEN MONDAL
S.N. Bose National Centre for Basic Sciences, JD Block,
Salt Lake, Sector-III, Kolkata 700098, INDIA
[email protected]

SANDIP K. CHAKRABARTI
S.N. Bose National Centre for Basic Sciences, JD Block,
Salt Lake, Sector-III, Kolkata 700098, INDIA
and
Centre for Space Physics, Chalantika 43, Garia Station Rd., Kolkata 700084
[email protected]

We present here a simplified approach to the study of particle dynamics around rotating
black holes. We show that with a suitably modified effective potential of the central
gravitating rotating object, one can carry out these studies very accurately in Newtonian
formalism provided the Kerr parameter remains within -1 <::: a <::: O.S. We present the
geometrical quantities which are useful to study the particle dynamics from our potential
and also compare it to the general relativistic result.

1. Introduction
In the pseudo-Newtonian description of the space-time (which is originally proposed
by Paczynski & Wiita, 1980), the gravitational potential <l>N(r) = -GMbh/r is re-
placed by the pseudo-Newtonian (PN) potential <I>(r)PN = -GMbh/(r-2GMbh/c2).
Here, G and c are the universal gravitational constant and the velocity of light re-
spectively and Mbh is the mass of the black hole.
We extended this concept of using a 'pseudo-Newtonian' approach even for the
Kerr geometry and show that it is possible to mimic most os the properties by
a reasonably simple choice of an effective potential. Several attempts have been
made in this regard by e.g. Chakrabarti & Khanna (1992), Artemova, Bjoernsson
& Novikov (1996), and Mukhopadhyay (2002). Our approach here is very similar
to Chakrabarti & Khanna (1992) which is valid only for extreamly Kerr black hole
(Kerr parameter a > 0.9). Details of our results are in Chakrabarti & Mondal
(2006) .

2. The Pseudo-Kerr geometry

In what follows, we choose G = c = M = 1. Distances, specific angular momenta
and times are measured in units of GM/c 2 , GM/c and GM/c 3 respectively. The
potential we propose is:

(1)

where, () is measured from the axis of rotation of the black hole and TO = 0.04 +
0.97a + 0.085a 2 . The red-shift factor a (Eq. 1) and the dragging term, w have been

1119
1120

borrowed from general relativity,

a = fVE and (2)

where, p2 = 7'2 + a 2 cos 2 (j, ~2 = (7'2 + a 2 )2 - a 2 6.sin 2 (j and 6. = 7'2 + a 2 - 27'. The
specific angular momentum 1here has the same meaning as u¢ in GTR.

3. Comparison of the effective potentials

In Fig. la, we compare the potentials in the Kerr geometry (solid curves) given by

(3)

and that of the Pseudo-Kerr geometry for a = 0.5 (dotted curves). The specific
angular momentum, from the lowermost to the topmost curves, are (solid curves)
u¢ = 1 = 2.7,2.9029,3.1,3.3,3.4142 and 3.5 respectively and are compared with the
curves (dotted) for the Pseudo-Kerr potential for 1= 2.7,2.8978,3.1,3.33.4237 and
3.5 respectively.

4. The marginally stable and bound orbits

The radius at which the point of inflexion of the potential occurs is the marginally
stable orbit. Mathematically,

d<PPK
- -I
dr rrn. .'l
-0
- ,
(4)

The angular momentum for which these equations are satisfied is known as the
marginally stable angular momentum lms. For the marginally bound orbits, one
must have,

(5)

The angular momentum for which these equations are satisfied is known as
marginally bound angular momentum 1mb.

5. Discussion
We presented a modified gravitational potential to study particle dynamics around
a Kerr black hole, which, for all practical purposes, could be used as easily as a
Newtonian potential. This is found to reproduce characteristics of particle trajec-
tories very accurately as long as the Kerr parameter -1 < a < 0.8. The errors in
this range, if any, are found to be less than five percent or so.
 1121
104,----------------,

1.02 a=D.5

0.98 0.95

0.96
0.9
094
0.92

0.9
0.88 0.8
086
0.84 0.75
0.82 "---"1J..~_L_~___LI.~
6 8 10 0.7 L_~~~~~~~_~"___'
r -! -(J.8 -0.6 -0.4 -0.2 a 0.2 0.4 0.6 D.R I
a

Fig. lea-b): (a) Comparison of the General relativistic Kerr potential VK (solid), the
pseudo-Kerr potential <P P K for a = 0.5. See text for the parameters for which the curves
have been plotted. (b) Plot of marginally stable values of potential <Pms with a.

4.5
lms

2.5

1.5_Ll-.~U.-C8-.0~.6--C.O~.4-.~U.C-2-O~~O."-2-'0~.4--:-0~.6-0~.8c--'1
a

Fig. 2(a-b): (a) The locations of the marginally bound (rmb) and marginally stable (rms)
orbits at these orbits in both Kerr geometry (solid) and pseudo-Kerr geometry (dashed).
(b) The locations of the marginally bound angular momenta 1mb and lms at these orbits in
both Kerr geometry (solid) and pseudo-Kerr geometry (dashed). Results generally agree
for -1 < a < 0.8.

References
1. 1. V. Artemova, G. Bjoernsson, 1. D. Novikov, 1996, ApJ 461, 565
3. S. K. Chakrabarti, 1990, Theory of Transonic Astrophysical Flows, (Singapore: World
Scientific)
4. S. K. Chakrabarti, 1993, MNRAS 261, 625
5. S. K. Chakrabarti, R. Khanna, 1992, MNRAS 256, 300
8. S.K. Chakrabarti, S. Mondal, 2006, MNRAS 371, 1418
6. B. Mukhopadhyay, 2002, ApJ 581, 427
7. B. Paczynski, P. J. Wiita, 1980, A & A 88, 23
 EXTREME GRAVITATIONAL LENSING BY SUPERMASSIVE
BLACK HOLES

VALERIO BOZZA
Dipartimento di Fisica "E.R. Caianiello", Universita di Salerno,
Via S.Allende, 1-84081, Baronissi (SA), Italy
and
Istituto Nazionale di Fisica Nucleare, Sezione di Napoli, Italy
[email protected]

Photons grazing a supermassive black hole experience large deflections that cannot be
described in the weak field approximation. Such photons give rise to very faint additional
images of any given source, which would show up just outside the apparent shadow
of the black hole. We investigate the properties of these images through an analytical
expansion of the deflection angle around the minimal impact parameter, showing that
their positions and brightnesses are related to the parameters of the black hole and
the gravitational theory used to describe it. We present estimates for the position and
the luminosity of images for several kinds of potential sources orbiting around Sgr A *,
which indeed represents an ideal candidate lens, discussing the technical requirements
for their detection. In particular, we suggest that future space interferometry missions
will be able to detect these elusive witnesses of the strong gravitational fields generated
by black holes.

1. From weak to strong deflections

Light rays travelling very close to a black hole experience very high deflections.
General Relativity predicts that the deflection angle diverges when the impact pa-
rameter of the approaching photon trajectory reaches a minimum value Um. The
observable consequence of the divergence of the deflection angle is that an observer
having a telescope with enough spatial resolution would see two infinite sequences
of images on both sides of a black hole for every source in the sky. The outermost
image is formed by photons performing one loop around the black hole before reach-
ing the observer, the second image is formed by photons performing two loops and
so on. Of course, only sources very well aligned on the observer-lens axis would form
images bright enough to be taken in consideration for detection.

2. The Strong Deflection Limit for spherically symmetric black

holes
This extreme lensing phenomenology, first studied by Darwin 1 for the Schwarzschild
black hole, has been proved by us to be common to all spherically symmetric black
holes. 2 In fact, it is possible to write a universal expansion for the deflection angle
close to its divergence

Ct(u)=-alog(U: -l)+b+O(U-Um). (1)

1122
 1123

The divergence is always logarithmic and the first two terms of the expansion are
typically sufficient for a full description of the so-called relativistic images in the
extreme lensing regime. The coefficients a, band U m appearing in the universal
formula (1) depend on the specific form of the black hole metric and represent a
sort of identity card of the black hole. Since the position and the magnification of the
images depends on these coefficients, the detection of relativistic images would allow
to identify the actual metric of a black hole, opening the possibility of a new test of
general relativity in a strong field regime, where it is likely that alternative theories
of gravitation would deviate from Einstein's theory, with their own prediction for
the positions and the brightness of the relativistic images.

3. Spinning black holes

Recently, we have extended the Strong Deflection Limit methodology to the Kerr
black hole,3-5 representing a general relativity black hole endowed with an intrinsic
angular momentum. The loss of spherical symmetry complicates the investigation of
the problem. Nevertheless, we have managed to attain a full analytical description
for small values of angular momentum.
The most evident change induced by the presence of angular momentum is the
presence of finite-size caustics. In fact, for spherically symmetric black holes, the
relativistic images are infinitely magnified only if the source is perfectly aligned
behind the black hole. Then, one says that there is one caustic point of zero size
behind the lens. This caustic point is common to all relativistic images.
In Kerr black holes, each pair of relativistic images has its caustic in the form of
a closed curve, which is displaced from the observer-lens axis and acquires a finite
angular size. If the source crosses the first caustic, then the first pair of images will
be infinitely magnified, if the source crosses the second caustic, then the second
pair will be magnified and so on. The shape of the caustics is the classical astroid
shape that is usually met in weak deflection gravitational lensing as well. This shape
emerges every time the spherical symmetry is broken. In fact, astroid caustics are
also formed by Schwarzschild black holes embedded in an external field. 6
The existence of finite-size caustics has an important implication. In fact, when
the source enters a caustic, an additional pair of relativistic images appears, so that
the total number of magnified images becomes equal to four (Fig. 1).
An important fact emerging by our analysis is that, to the lowest order in the
black hole spin, all observables depend on the projection of the spin on a plane
orthogonal to the observer-lens axis. In order to disentangle the absolute value of
the spin from its inclination relative to the line of sight, it is necessary to go to the
next to leading order. 5

4. Observing relativistic images

The best candidate for the observation of images outside the weak deflection limit
is the black hole in the center of our galaxy, named Sgr A *. Recent estimates fix
1124

b
0.00
O
0.00
::!. -0.00
-0.00
<>
.00I • S:J

0.00aJO.002
b 8:88[Q-3.
0.00
::!. -0.00
-0.00
•

0.00aJO.002
!J.y !J.y

Fig. 1. Relativistic images for two positions of the source with respect to a Kerr caustic. On the
top left, the source is outside a caustic; the corresponding images, in the form of solid arclets are
shown in the bottom left panel. On the top right, the source is inside a caustic; the corresponding
images are shown in the bottom right panel, where it is evident that two additional relativistic
images appear.

its mass to 4 x 10 6 MC'). Several stars have been identified around Sgr A* and their
orbits have been tracked very accurately. With six of these stars as sources, we have
calculated the position and the brightness of the four brightest relativistic images
throughout their orbits. For S14 we have found that a favorable alignment will occur
in 2038, when the secondary image will reach the magnitude K = 24 in a regime
intermediate between weak and strong deflection. 7
The best possibility to detect relativistic images perhaps comes from the X-ray
band. In fact, several Low-Mass X-ray binaries around Sgr A * have been identi-
fied. Some of these sources are even more powerful than Sgr A * itself, so that one
could expect a high Signal-To-Noise ratio for the relativistic images in case of good
alignment. 5
Needless to say, these observations are very demanding, since a resolution of the
order of microarcsecond is necessary to distinguish a relativistic image from Sgr
A *. In the X-rays the best idea seems to be interferometry in space. The project
MAXIM (http://maxim.gsfc.nasa.gov) could reach the striking resolution of 0.3
microarcseconds, which would be largely sufficient for the observation of relativistic
images around Sgr A *. As soon as this observation is accomplished, a new door for
the study of gravitational theories will be opened.

References
1. C. Darwin, Pmc. of the Royal Soc. of London 249, 180 (1959).
2. V. Bozza, Phys. Rev. D66, 103001 (2002).
3. V. Bozza, Phys. Rev. D67, 103006 (2003).
4. V. Bozza, F. De Luca, G. Scarpetta, and M.Sereno, Phys. Rev. D72, 083003 (2005).
5. V. Bozza, F. De Luca, and G. Scarpetta, Phys. Rev. D74,063001 (2006).
6. V. Bozza and M. Sereno, Phys. Rev. D73, 103004 (2006).
7. V. Bozza and L. Mancini, ApJ 627, 790 (2005).
 Spectral and Timing
Appearances of the Galactic
and Extragalactic Black
Holes
This page intentionally left blank
 PHYSICAL CHARACTERISTICS OF XTE-J1650 AND GRS
1915+105 WITH BEPPOSAX

ENRICO MONTANARI
Dipartimento di Fisica, Universitii di Ferrara, via Saragat 1, 1-44100, Ferrara, Italy
[email protected]
and
IIS-Calvi, via Digione 20, 1-41034, Finale Emilia (MO), Italy

LEV TITARCHUK
George Mason University/Center for Earth Observing and Space Research, Fairfax, VA 22030
and
US Naval Research Laboratory, Code 7655, Washington, DC 20375-5352
[email protected]
and
Dipartimento di Fisica, Universitii di Ferrara, via Saragat 1, 1-44100, Ferrara, Italy
[email protected]
and
Goddard Space Flight Center, NASA, code 661, Greenbelt MD 20771
[email protected]

FILIPPO FRONTERA
Dipartimento di Fisica, Universitii di Ferrara, via Saragat 1, 1-44100, Ferrara, Italy
[email protected]
and
Istituto di Astrofisica Spaziale e Fisica Cosmica di Bologna, INAF, via Gobetti 101, 1-40129,
Bologna, Italy

We show preliminary results of a comparative spectral and timing analysis of Galactic

Black Holes XTE J1650-500 and GRS 1915+105, using data from BeppoSAX. A clear
relation is present between the evolution of energy spectra and power spectra. The central
frequency of a given QPO is increasing with the softening of the energy spectra. This
is consistent with the following picture: the spectral evolution from Hard to Soft is
occurring together with a decreasing of the dimension of the Comptonizing corona. This
also implies that the effect of bulk inflow vs thermal Comptonization is increasing its
importance during the softening of the source.

1. Introduction

XTE J1650-500 1 is a black hole binary with a mass function f(M) = 2.7 ± 0.6 M(')
with an upper limit for the black hole mass of rv 7.3l"vl(,) and a period of 7.63±0.02 h 2
(even if a different result was proposed in 3 ). Presence of 1 - 3 Hz Quasi Periodic
Oscillations (QPOs) was established. 4 ,5 Some authors 6 ,7 found evidence for an iron
line. However it appears that the need of this iron line is model dependent. In fact
using suitable continuum models no excess is present around 6 - 7 ke V.
GRS 1915+105 8 is a well known micro quasar. Evidence for radio-emitting
plasma jets was found. 9 The source mass function has been estimated to be
rv 9.5 M (') with a possible mass of rv 14 M (').10-12 The source shows a dramatic

variability of light curves and spectra that was described by 12 spectral states. 13

1127
1128

The observations were performed by BeppoSAX Narrow Field Instruments

(NFIs) described in. 15- IS For each source we have studied both spectral and timing
features, in order to obtain an organic and consistent picture of the evolution of the
sources.

2. Energy Spectra
Because of the spectral variability of the sources, the observations were divided
into shorter intervals during which spectra can safely be considered constant. Many
models can be used to fit the spectral shape (for XTE-J1650 see for instance 7 ,14,19)
Our aim is to compare the time evolution with a simple spectral parameter, the
high energy photon index. Therefore, for all the intervals we used

(1) A generic model describing both direct and Comptonized photons from a black
body-like source (a portion of the disk) and a corona. The Compton upscatter-
ing in the corona can be thermal or due to bulk inflow effect (XSPEC model
BMC 20 - 24 ).
(2) A black body-like source of direct soft photons with negligibly small interaction
with the corona (a portion of the disk at a greater distance to the central object)
(XSPEC model BB).
(3) A smeared edge describing the interaction of photons from the central object
with a wind (XSPEC model SMEDGE)
(4) For hard state only, a high energy cutoff describing the feedback from the coro-
nal plasma (XSPEC model HIGHECUT)
(5) Some spectra reveal a feature in the 15-30 keY range. The shape of this feature
can be described by XSPEC BB model.

In this analysis we are mainly interested in the high energy photon index, the
evolution of which will be compared to the timing evolution (this issue will be
addressed in the next section). Obviously the result of this comparison is reliable
if it is independent on the model used to fit the spectral shape. In fact, the use of
different models having the photon index as a parameter yields the same result.

3. Timing vs spectral evolution

We have used BeppoSAX data of all the NFIs to obtain Power Density Spectra
(PDS). QPOs were detected. The strength of these QPOs is stronger in the 4-15 ke V
range, so one is led to conclude that the process involved in their production occurs
in the corona, where high energy photons are generated.
Preliminary results regarding 4 - 10 ke V photons, show strong evidence that
when the source is harder the central frequency of QPOs is smaller than when the
source is softer.
This fact, together with the preceeding one (corona as QPOs formation region)
seems to imply that during the evolution from hard to soft the Compton cloud
 1129

becomes more compact: this is an indication that the role of bulk inflow effect is
increasing in the evolution from hard to soft.
A detailed discussion and final results of this investigation can be found in. 25

4. Conclusion
Our preliminary results show that

(1) A clear relation is present between the evolution of black hole energy spectra
and power spectra: power spectra become harder (high frequency of the QPO)
when energy spectra becom softer.
(2) There are indications for bulk inflow effect in Soft State of black holes: Compton
cloud gets more compact and thermal Comtonization only cannot explain data.
(3) For the first time power spectral shape has been determined in fUIlction of
energy in a very broad energy range (from 0.1 to 200 ke V). This allows us to
understand where photons of a given energy arc produced.

References
1. R. Remillard, IAU Oire. No. 7707 (2001).
2. J. A. Orosz, J. E. McClintock, R. A. Remillard, S. Corbel, Astrophys. J. 616, 376
(2004).
3. C. Sanchez-Fernandez, C. Zurita, J. Casares, A. J. Castro-Tirado, I. Bond, S. Brandt,
N. Lund, IAU Cire. No. 7989 (2002).
4. M. Revnivtsev, R. Sunyaev, IAU Cire. No. 7715 (2001).
5. R. Wijnands, J. M. Miller, W. H. G. Lewin, IAU Cire. No. 7715 (2001).
6. J. M. Miller, et al., Astraphys. J. 570, L69 (2002).
7. G. Miniutti, A. C. Fabian, J. M. Miller, Mon. Not. R. Astr. Soc. 351, 466 (2004).
8. A. J. Castro-Tirado, S. Brandt, N. Lund, IAU Cire. No. 5590 (1992).
9. 1. F. Mirabel, L. F. Rodriguez, Nature 371, 46 (1994).
10. K. Borozdin, M. Revnivtsev, S. Trudolyubov, C. R. Shrader, L. Titarchuk, Astraphys.
J. 517, 367 (1999).
11. J. Greiner, J. G. Cuby, M. J. McCaughrean, Nature 414,522 (2001).
12. C. R. Shrader, L. Titarchuk, Astraphys. J. 598, 168 (2003).
13. T. Belloni, M. Klein-Wolt, M. Mendez, M. van der Klis, J. van Paradijs, Astran.
Astrophys. 355, 271 (2000).
14. E. Montanari, F. Frontera, L. Amati, Nucl. Phys. B Suppi. 132,412 (2004).
15. A. N. Parmar et al., Astran. Astraphys. Suppi. 122, 309 (1997).
16. G. BoeHa et al., Astran. Astraphys. Suppi. 122, 327 (1997).
17. G. Manzo et al., Astran. Astrophys. Suppi. 122,341 (1997).
18. F. Frontera et al., Astran. Astraphys. Suppi. 122, 357 (1997).
19. E. Montanari, L. Titarchuk, F. Frontera in preparation.
20. L. Titarchuk, A. Mastichiadis, N. D. Kylafis, Astraphys. J. 487, 834 (1997).
21. L. Titarchuk, T. Zannias, Astraphys. J. 493, 863 (1998);
22. P. Laurent, L. Titarchuk, Astrophys. J. 511, 289 (1999);
23. T. Zannias, K. Borozdin, M. Revnivtsev, S. Trudolyubov, C. R. Shrader, L. Titarchuk,
Astraphys. J. 517, 367 (1999);
24. C. R. Shrader, L. Titarchuk, Astraphys. J. 521, L21 (1999).
25. E. Montanari, L. Titarchuk, F. Frontera, in preparation.
 SPECTRAL AND TIMING PROPERTIES OF MAGNETIZED
ADVECTIVE FLOWS WITH STANDING SHOCKS

SAMIR MANDAU,* and SANDIP K. CHAKRABARTI 1,2,t

1 Centre for Space Physics,
43 Chalantika, Garia Station Road, Kolkata 700084, India
2 S. N. Bose National Centre for Basic Sciences
JD Block, Sector III, Salt Lake, Kolkata 700098, India
* [email protected]
t [email protected]

Advective flows are known to have standing, oscillating or propagating shocks in the disk
which are responsible for the quasi-periodic oscillations. We have calculated the QPO
frequency from shock locations (x s ) in presence of cooling effects. We have shown the
change in the ratio of total integrated power (high/low) with shock location as well as
with the accretion rates. These results generally agree with observations.

1. Introduction

Observations of black hole candidates are best understood when both the Kep-
lerian and the sub-Keplerian components are assumed to be present. A standard
Shakura-Sunyaev Keplerian disk cannot explain most of the observational aspects
including the spectral state transitions, variability class transitions, quasi-periodic
oscillations etc. Sub-Keplerian component as a major component of the flow has
been proven by several theoretical and numerical works. In particular, Chakrabarti
& Titarchuk (1995, hereafter CT95)3 showed that the state transition of black
hole candidates is possible due to a change in accretion rates in a two-component
advective flow. However, CT95 did not include the synchrotron emission. Mandai
& Chakrabarti (2005, hereafter MC05)4 included synchrotron radiation assuming
stochastic magnetic fields. Subsequently, Chakrabarti & Mandai (2006)1 included
the Keplerian disk also as in CT95, and showed that observations of Cyg X-I can
be easily explained by variation of the Keplerian and sub-Keplerian rates. In the
present paper, we shall discuss some properties of the post-shock region, i.e., the
CENBOL (centrifugal pressure supported boundary layer). The important parame-
ters of our models are the halo rate (mh), the disk rate (md) and the shock location
(xs). We have taken the black hole mass to be M = 10MC'). We measure the mass
accretion rates in units of the Eddington rate and the radial distances in units of
the Schwarzschild radius.

2. Shock Signature
Post-shock region in between the horizon and the shock in a sub-Keplerian flow
serves as the 'boundary layer' of a black hole which produces the high energy radi-
ation. The QPO seen in the power density spectra is due to the oscillation of the
CENBOL region. QPO frequency is computed by dividing the infall velocity by the

1130
 1131

1.5

Non -thermal

Thermal

(:$ 1

0.5
o 100 200

Fig. 1. Variation of the energy spectral index (ex) for thermal (solid) as well as non-thermal
(dotted) electrons with the QPO frequency. For details see text.

shock location 5 though sometimes cost ant velocity due to turbulence yielded better
result. 2 In Fig. 1, we plot the QPO frequency vs. energy spectral index (a) for both
thermal electrons (solid) as well as non-thermal (dotted) electrons. The parame-
ters for the figures are in" = 0.3, in'd = 0.2. QPO frequency v is small for large
Xs' There is also a dip. This is expected since the supply of soft photons decrease

but the density of the flow also decreases thereby decreasing local cooling by syn-
chrotron, Comptonization etc. Before the dip, i.e., for very large shock location the
spectral slope increase since density is so small that even local synchrotron emission
is sufficient to cool the flow though the spectrum remains hard. For non-thermal
electrons, the spectral slope monotonically decreases with QPO frequency since it
does not depend on the temperature of the flow 1 but depends only on the optical
depth.
In Fig. 2 we plot the ratio (q) between the total integrated powers in high energy
(10keV-10Mev) and low energy (O.lkeV-lOkeV) with shock locations. Here, we vary
inh while keeping the sum (inh + ind = 0.5) fixed. This figure shows the effect of
changing one type of accretion rate to the other. The contribution in high energy
increases with the increase of shock location. A striking feature is that for a large
Xs this contribution decreases with increase in in". This is because the increase in

density means increase in self-cooling. For small X s , high energy radiation increases
with in" as the heating due to geometrical compression is high and increase in in,h
1132

750 /
/
/
/
/
/
/
/
. /
ffi h =O.2/
500
/
/
/
/
/
/
/
/
/
250 /
/

/
/

o
o 10 20 30 40 50

Fig. 2. Change in ratio (high/low) of the total integrated power with the location of the shock
(xs). The different values of mh are marked on the individual curves.

means decrease in md which reduces the Keplerian contribution.

This work is partially supported by ICRANet and ISRO-RESPOND project.

References
1. Chakrabarti, S. K and MandaI, S., ApJ 642, L49 (2006).
2. Chakrabarti, S. K and Manickam, S. G., ApJ 531, L41 (2000).
3. Chakrabarti, S. K and Titarchuk, L. G., ApJ 455, 623 (1995).
4. MandaI, S. and Chakrabarti, S. K, A&A 434,839 (2005).
5. Molteni, D., Sponholz, H., Chakrabarti, S.K, ApJ 457, 805 (1996).
 ESTIMATING BLACK HOLE MASSES IN ULXs

ROBERTO SORIA
Harvard-Smithsonian CfA, 60 Garden st, Cambridge, MA 02138, USA, and
Mullard Space Science Lab (UCL), Holmbury St Mary, Dorking, Surrey RH5 6NT, UK
[email protected]

1. Direct and indirect methods

Much of the enigma of ultraluminous X-ray sources (ULXs) stems from our current
inability to estimate the masses of the accreting black holes (BHs) powering those
sources. Several models have been suggested, with wildly discrepant mass ranges,
going from ordinary stellar-mass BHs (M ~ 10M8 ) to intermediate-mass BHs (M ~
10 3 M 8 ), and correspondingly diverse geometries of emission and physical origin.
Here, we summarize some of the direct and indirect methods that may be used to
estimate BH masses in ULXs.
Kinematic mass function
This is the most direct method, also used for Galactic BHs. Knowing the binary
period and the projected orbital velocity, one can put a lower limit to the BH mass. It
requires phase resolved spectroscopic observations of a disk emission line--assuming
that the accretion disk shares the same orbital motion as the BH. Typically, Bell
'>-4686 is used for Galactic BHs. Unfortunately, the optical counterparts of ULXs
are all fainter than V ~ 23 mag, making such observations very challenging even
for 8-m telescopes. Pioneering work is being carried out by M. Pakull's tell-m.l
Spectral fits to the thermal disk emission
In Galactic BHs, modelling the accretion disk emission with a multicolour blackbody
provides the second most reliable method for mass determination. 2 In the disk-
dominated high/soft state, the disk luminosity Ld is simply related to the peak
temperature Tin and the inner-disk radius R in , as L d ;::::; 471'RfnTi!' Both Ld and Tin
can be inferred from X-ray spectral data, leading to an estimate of Rin ~ 50 km
in Galactic BBs. Assuming that Rin ;::::; RISCO::; 6G 1M/ c 2 , one can turn that into a
mass estimate;::::; 10M8 . Further fine-tuning of this estimate is obtained by taking
into account the zero-torque condition, the hardening factor and the viewing angle. 2
Applying the same argument to ULX spectra is fraught with danger. It is still not
clear what contribution (if any) comes from the accretion disk. This is partly because
only the 0.3 -10 ke V band has been monitored, making it difficult to distinguish
thermal and non-thermal components. Some models predict that the disk dominates
at the high-energy end of the observed band (k~n ~ 1.5-2 keY); others claim the
disk may not be visible at all; other models suggest that the disk produces the
"soft-excess" at the low-energy end of the observed spectra (kTin ~ 0.15-0.20 keY).
Let us consider the third possibility as the most likely: in this case, the inferred
radius is Rin ~ 5000-10000 km. If this radius corresponded to the innermost stable
orbit, it would be evidence of intermediate-mass BHs with NI ~ 1000N18 . 3

1133
1134

However, this large value of R in may not represent the innermost stable orbit,
but a characteristic transition radius Re between the standard outer disk and an
inner region dominated by non-thermal processes, Poynting flux, and/or massive
outflowS 456 .7 Most of the accretion power from the inner region is not radiated by
the disk; this also explains why the peak temperature of the disk is lower. If so,
one additional constraint is needed to relate the BH mass to the fitted value Re.
Empirically, Re may be constrained by the ratio between thermal and non-thermal
spectral components. Typically, the thermal component contributes rv 5-30% of the
luminosity in the 0.3-10 keY band; this suggests that Re rv 10-20RISCO . Physically,
what is the transition at Re? In one scenario, the outer disk (R :2: Re) has a
luminosity Ld = LEdd, Re is the spherization radius, and the inner region (RISCO <
R <:::: Re) is dominated by massive outflows. In other scenarios, R <:::: Re is where most
of the accretion power is drained from the disk by magnetic processes; this allows
Ld < L Edd from the outer disk. Most scenarios predict that Re/ RISCO rv M / MEdd ;
the observed luminosities and temperatures are consistent with BH masses rv 40-
100MC') , accreting at a rate M ;;:, lOMEdd , with a power-law contribution (mostly
from the inner inflow region) Lpo rv a few xL Edd (or ;S L Edd but beamed by a
factor of a few). A transition radius Re > RISCO is also occasionally detected in the
bright, power-law dominated very-high-state of some Galactic BHs4. 7
Characteristic time variability
Low-frequency quasi-periodic oscillations (QPOs) have been observed in the non-
thermal component of three ULXs: M82 X-I (with frequency varying between 50
mHz and 166 mHz), Holmberg IX X-I (at 200 mHz) and NGC 5408 X-I (at 20 mHz).
Such frequencies are rv 20-100 times lower than those seen from stellar-mass BHs in
the very-high-state. Empirical observations and theoretical arguments 8 suggest that
frequencies scale rv 1/M across the mass range of compact objects, from neutron
stars to AGN. Taken at face value, this would suggest BH masses rv 1000MC').
However, as in the case of large inner-disk radii and low disk temperatures, low
QPO frequencies cannot be taken as evidence in favour of intermediate-mass BHs.
Characteristic QPO frequencies may depend on the size of the Comptonization
region, and on the values of Re and T(Re) at the disk/corona interface. Thus, the
lower frequencies seen in ULXs may be due to a combination of moderately higher
BH masses and much higher accretion rates. Similarly, spectral and timing studies
of the stellar-mass BH XTE J1550-564 show that the QPO frequency decreases
when the fitted value of Re » RISCO and the inner-disk temperature gets cooler. 7
Eddington limit
The X-ray luminosity function of ULXs seems to be a high-luminosity extension
of the high-mass X-ray binary luminosity function. 9 It has a break or turnover at
Lx :::::: 2 x 10 40 erg S-l, although a few sources have been found with apparent
luminosities up to Lx :::::: 10 41 erg s-l. It was suggested that 2 x 1040 erg S-l
represents the Eddington limit of the most massive EHs in ULXs: this would imply
BH masses:::::: 150MC'). More luminous sources could be transients or mildly beamed.
However, theoretical models allow for the possibility of persistent super-Eddington
 1135

emission (by a factor of a few): if so, the upper mass limit could be even < 150M8 .
Spectral state transitions
Stellar-mass BHs are found in the low/hard spectral state for accretion rates (and
corresponding luminosities) :S a few percent of their Eddington values. They tend
to be in the high/soft (thermal-dominant) state for X-ray luminosities ~ O.l-lLEdd,
and switch again to a power-Iaw-dominated state for higher accretion rates and lu-
minosities. If ULXs underwent the same pattern of state transitions, it would be
easy to constrain their Eddington luminosities. However, no ULX has been conclu-
sively found to switch between thermal and non-thermal states. This could be due
to their consistently higher accretion rates, from Roche-Iobe-filling B stars.
Radio-X-ray correlations
In the low/hard state, an empirical relation exists between core radio luminosity
(interpreted as synchrotron emission from a compact jet), core X-ray emission and
BH mass ("fundamental plane of BH accretion" 10). It is not yet clear whether a
similar correlation also holds in the very high state. Although radio emission has
been found associated with some ULXs, the fundamental-plane relation cannot be
applied to estimate their BH masses, because: a) ULXs are almost certainly not
in the low/hard state; b) their radio counterparts are not flat-spectrum core radio
sources. They are consistent with optically-thin synchrotron emission from a jet-
inflated bubble and/or an underlying supernova remnant (like in SS433). The typical
flux limit of current radio surveys is ~ 0.1 mJy. If ULXs also have unresolved core
radio emission, we expect its flux density to be an order of magnitude fainter.

2. Conclusions
There is increasing indirect evidence that most ULXs can be explained as accreting
BHs with masses :S 100M8 rather than ~ 10 3 M8 as initially thought. Thus, ULX
masses may be consistent with the highest BH mass that can be formed from in-
dividual stellar progenitors (M ~ 60M8 ), removing the need for exotic formation
processes such as primordial remnants or runaway coalescence of 0 stars in the core
of super star clusters. Low inner-disk temperatures, large apparent inner-disk radii,
low QPO frequencies and a dominant non-thermal X-ray component are consistent
with ULXs being in the very-high-state at accretion rates ~ 10MEdd. 7
References
1. M. Pakull, F. Grise and C. Motch, Proc. IAU Symp. 230, astro-ph/0603771.
2. K. Makishima et al., ApJ 535, 632 (2000).
3. J. M. Miller, A. C. Fabian and M. C. Miller, ApJ 614, L117 (2004).
4. C. Done and A. Kubota, MNRAS 371, 1216 (2006).
5. M. C. Begelman, A. R. King and J. E. Pringle, MNRAS 370, 399 (2006).
6. J. Poutanen, G. Lipunova, S. Fabrika, et al., MNRAS in press (2007).
7. R. Soria and Z. Kuncie, MNRAS submitted (2007).
8. L. Titarehuk and R. Fiorito, ApJ 612, 988 (2004).
9. D. A. Swartz, K. K. Ghosh, A. Tennant and K. Wu, ApJS 154, 519 (2004).
10. A. Merloni, S. Heinz and T. di Matteo, MNRAS 345, 1057 (2003).
This page intentionally left blank
 Extreme Properties of
Neutron Stars: Observations
and Theory
This page intentionally left blank
 EQUATION OF STATE OF DENSE MATTER IN NEUTRON
STARS*

SLAVOMIR CERNY
Institute of Physics,. Faculty of Philosophy and Science, Silesian University in Opava.
Bezrucovo nam, 13, Opava, CZ-746 01, Czech Republic
E-mail: [email protected]

JIIUNA RtKOVSKA STONE

Department of Physics, University of Oxford, Oxford, OX1 3PU, UK

ZDENEK STUCHLIKt and STANISLAV HLEDIK+

Institute of Physics, Faculty of Philosophy and Science, Silesian University in Opava,
Bezrucovo nam. 13, Opava, CZ-746 01, Czech Republic
E-mail: [email protected]@fpf.slu.cz

Keywords: Neutron star; Equation of state (EoS); Mean field model; Kaon; Pion; Hype-
ron.

1. Introduction
Interior of a neutron star consists of various types of matter;1,2 namely (from sur-
face towards the centre): iron crust described by Feynman-Metropolis-Teller EoS,
neutron rich nuclei described by Baym-Pethick-Sunderland EoS, neutron drip to
dissolve nuclei described by Baym-Bethe-Pethick EoS, and the very centre of a
neutron star described by various types of EoS with kaons, pions and hyperons.

2. Neutron Star Model

The neutron star models describe non-rotating, cold (T = 0) neutron stars without
magnetic field. The line element in standard Schwarzschild coordinates (t, T, ¢) e,
reads
(1)
where <P and \II are functions only of T.
To get the equation of hydrostatic equilibrium, Einstein field equation have to
be taken into account 3

(2)

where G",v is Einstein tensor, R",v is Ricci tensor, R is Ricci scalar and TILV is
stress-energy tensor. For a perfect fluid, the stress-energy tensor has simple form
(3)

*This research has been supported by Czech grant MSM 4781305903.

1139
1140

The conservation of energy and momentum is then expressed by

(4)
The equation of hydrostatic equilibrium can be derived from Eqs. (2)-(4) in the form
of Tolman-Oppenheimer-Volkoff (TOV) equation of hydrostatic equilibrium

dp (2 ) §-m(r) + ~pr3
-d = - pc +p [ 20 ()] ,
(5)
r r r -""C'2m r
where

m(r) = !aT 47rr,2 p(r') dr' . (6)

For obtaining neutron-star models, Eq. (6) needs to be integrated from the centre
out to the stellar surface, where the pressure vanishes. For given central conditions
(central energy density Po and the corresponding pressure Po), one can obtain a
neutron-star model and its gross properties (mass M, radius R, total baryon number
A etc.)

3. Relativistic Mean Field Model with Kaons (Pions)

Neither pions nor kaons exist as a component of nuclear matter under normal cir-
cumstances. The pion has negative parity; consequently, its expectation value in the
normal ground state, which is diagonal matrix element, vanishes. The kaon carries
strangeness and its expectation likewise vanishes. 4
However, conditions may arise at higher density that favour a change in the
structure of the ground state, yielding finite expectation values and permitting the
appearance of a Bose condensate. The electron chemical potential increases with the
density of charge neutral matter to compensate the growing density of protons. The
growth of the electron chemical potential can be arrested if hyperons of negative
charge become an important component of baryon populations.
A crucial question with regard to the occurrence of kaon (pion) condensation is
how well neutrality can be archived among baryon carrying species. The Kaon is
coupled to the mean field using minimal coupling
LK = D*I" K* DI" K - m*K
2 K K*
, (7)
where the vector fields are coupled by defining

(8)
The Lagrangians for the kaons describe the kaon-kaon interaction as well as the
kaon-nucleon interaction.
The form of equation of state depends sensitively on the chosen optical potential
of the kaon Udpo), which reflects its interaction with nuclear medium. The value
vary between -80 MeV and -140 MeV. The radius and especially the limiting mass
are sensitive functions of UK (Po). For the preferred value of UK (Po) < -120 MeV, the
 1141

radii are similar to neutron stars without the condensate. However, the behaviour
is actually continuous, but depends sensitively on UK(pO): a pure kaon condensed
core develops with decreasing values of the optical potential below ~ -120 MeV and
this causes the change of the radius from R ;:::::j 12.5 km to R;:::::j 8 km for Udpo) =
-140 MeV.

4. Conclusions
The presence of kaons in neutron stars occurs in the inner region up to 5 km.
Equation of state with hyperons is softened and the maximum mass of NS lowered.
(Some models without hyperons show maximum mass ~ 2 solar masses - but
there are still no observations.) We know that, according to present observations,
it appears that the lower bound on the maximum NS mass is ~ 1.5 solar masses.

Bibliography
1. E. 0stgaard, Internal structure of neutron stars, in Proceedings of RA Gtime 2/3: Work-
shops on black holes and neutron stars, Opava, 11-13/8-10 October 2000/2001, eds.
S. Hledik and Z. Stuchlik (Silesian University in Opava, Opava, 2001).
2. M. Urbanec, J. Miller and J. R. Stone, Matching of equations of state: influence on
calculated neutron star models, in Proceedings of RAGtime 6/7: Workshops on black
holes and neutron stars, Opava, 16-18/18-20 September 2004/2005, eds. S. Hledik and
Z. Stuchlik (Silesian University in Opava, Opava, 2005).
3. C. W. Misner, K. S. Thorne and J. A. Wheeler, Gravitation (Freeman, San Francisco,
1973).
4. N. K. Glendenning, Compact Stars: Nuclear Physics, Particle Physics, and General
Relativity (Springer-Verlag, New York, 2000).
 DETECTABILITY OF GRAVITATIONAL WAVES FROM THE
R-MODE INSTABILITY IN NEWLY-BORN NEUTRON STARS

PAULO M. sA * and BRIGITTE TOMEt

Centro Multidisciplinar de Astrojisica - CENTRA, Departamento de Fisica,
Faculdade de Ciencias e Tecnologia, Universidade do Algarve,
Campus de Gambelas, 8005-139 Faro, Portugal
* [email protected]
t [email protected]

Recently, it was shown that differential rotation is an unavoidable feature of r-modes. We

investigate its influence on the detectability of gravitational waves emitted by a newly-
born, hot, rapidly-rotating neutron star, as it spins down due to the r-mode instability.
We conclude that gravitational radiation could be detected by advanced laser interfer-
ometer detectors if the amount of differential rotation at the time the r-mode instability
becomes active is not too high.

Keywords: neutron stars; gravitational waves; r-mode instability.

R-modes are non-radial oscillating modes of rotating stars that have the Coriolis
force as their restoring force. In perfect-fluid stars these modes are driven unstable
by gravitational radiation for any value of the rotation frequency. 1,2 In newly-born,
hot, rapidly-rotating neutron stars the instability is active for a range of relevant
temperatures and angular velocities of the star. 3,4 As a consequence, the neutron
star spins down and strongly emits gravitational waves.
Owen et ai. 5 were the first to study the possibility of detecting gravitational
waves emitted by newly-born neutron stars due to the r-mode instability, arriving at
the conclusion that the advanced version of the kilometer-scale laser interferometer
detector LIGO could detect waves from sources located as far as 20 Mpc away
from Earth. However, a deeper understanding of this issue requires considering
the nonlinear theory. Taking into account nonlinear mode-mode interactions 6 and
the nonlinear mode self-interaction, 7 it was shown the initial results concerning
detect ability of gravitational waves from r-modes of newly-born neutron stars were,
most probably, too optimistic. Indeed, nonlinear calculations show that r-modes
may saturate at amplitudes much smaller than unity, making it more difficult to
detect gravitational waves.
In this paper we analyze the influence of differential rotation, arising within the
nonlinear theory at second order in the mode amplitude cx, on the detect ability of
gravitational waves from r-modes of newly-born, hot, rapidly-rotating neutron stars.
Differential rotation, which produces large-scale drifts of fluid elements along stellar
latitudes, is an unavoidable feature of nonlinear r-modes 8 and plays a fundamental
role in the evolution of the mode instability, as it provides a natural mechanism
of saturation of the r-mode amplitude. 9 The saturation amplitude depends on the
amount of differential rotation at the time the instability sets in, namely, CXsat ex
(K + 2)-1/2, where K, fixed by the choice of initial data, is related to the initial

1142
 1143

amount of differential rotation associated with the r-mode. 9

During the nonlinear evolution of the r-mode instability, part of the initial an-
gular momentum of the neutron star is carried away by gravitational waves. The
amplitude of these waves is given by 5

[h(t)[ = 1.3 x 1O-24a(t) (~;) 3 20 ~pc, (1)

where D is the angular velocity of the star, DK is the Keplerian angular velocity
at which the star starts shedding mass at the equator and D is the distance to the
star. The corresponding frequency-domain gravitational-wave amplitude is

[h(f)[ = 4.6 x 10-

VK +2
25 J
fmax
f
(20M P
D
C) Hz- 1
'
(2)

where the frequency f of the emitted gravitational wave decreases, due to the r-
mode instability, from f max ~ 1200 Hz (corresponding to the initial angular velocity
of the star, D K ) to fmin ~ 80 Hz (corresponding to final angular velocity of the
star,0.067D K ).7
The frequency of the gravitational waves emitted by newly-born neutron stars
due to the r-mode instability lies exactly in the range of frequencies in which the
laser interferometer gravitational-wave detectors LIGO and Virgo are most sensitive
and, therefore, a detection of these waves could be possible. Assuming matched
filtering, the signal-to-noise ratio for the initial LIGO, Virgo and the advanced
LIGO detectors is 7
0.9 (initial LIGO)
S 1 20 Mpc
0.7 (Virgo) (3)
N ~ VK +2 D x {
12.9 (advanced LIGO)
This expression shows clearly that a detection with the initial LIGO and Virgo
detectors is not probable, even for small values of the initial differential rotation
(K ~ 0). Indeed, if sources of gravitational waves are located about 20 Mpc away
from Earth the signal-to-noise ratio is not significant; if the distance is smaller, the
signal-to-noise ratio increases, but the number of neutron stars born in supernovae
within this distance is also smaller, making it improbable that an event takes place
during the operation of the gravitational wave detectors LIGO and Virgo. For higher
values of the initial differential rotation (K » 1), the situation is even worse.
Therefore, in what follows, we turn our attention to the advanced LIGO detector.
Let us assume that S / N = 4 is enough to detect a signal with the advanced LIGO
detector. Then, from Eq. (3) we obtain that a signal from a given source can be
detected if the distance to this source is less than 64.5 Mpc/VK + 2 (see Fig. 1). For
small values of the initial differential rotation (K < 10), a signal could be detected
if the distance to the source is less than about 20 Mpc. Within a sphere of such
radius, which includes the Virgo cluster of galaxies, one or two supernova events
per year are expected; therefore, in this case, one could hope to detect gravitational
1144

5
10

4
10
S/N<4
103
"
Q.
S/N>4
2l 2
10

10

103 105 10 6
8
10 102 104 107 10
K

Fig. 1. For values of K and D corresponding to points below the line, the signal-to-noise ratio
assuming matched filtering is SIN> 4 for the advanced LIGO detector.

waves from T-modes of newly-born neutron stars. On the other hand, if the star is
born with substantial differential rotation (for example, K > 10 6 ), then a detection
would require that the star is less than about 60 kpc away from Earth. Within a
sphere of such radius, only one or two supernova events per century are expected;
therefore, in this case, the probability of detecting gravitational waves from the
T-mode instability of newly-born neutron stars is quite small.

Acknowledgments
This work was supported in part by the Fundar;ao pam a CienC'ia e a Tecnologia,
Portugal.

References
1. N. Andersson, Astmphys. J. 502, 708 (1998).
2. J. L. Friedman and S. M. Morsink, Astrophys. J. 502, 714 (1998).
3. L. Lindblom, B. J. Owen, and S. M. Morsink, Phys. Rev. Lett. 80,4843 (1998).
4. N. Andersson, K. D. Kokkotas, and B. F. Schutz, Astrophys. J. 510,846 (1999).
5. B. J. Owen, L. Linblom, C. Cutler, B. F. Schutz, A. Vecchio, and N. Andersson, Phys.
Rev. D 58, 084020 (1998).
6. P. Arras, E. E. Flanagan, S. M. Morsink, A. K. Schenk, S. A. Teukolsky, and I. Wasser-
man, Astmphys. J. 591, 1129 (2003).
7. P. M. Sa and B. Tome, Phys. Rev. D 74, 044011 (2006).
8. P. M. Sa, Phys. Rev. D 69, 084001(2004).
9. P. M. Sa and B. Tome, Phys. Rev. D 71, 044007 (2005).
 X-RAY DIM ISOLATED NEUTRON STARS: A REVIEW OF THE
LATEST TIMING AND SPECTRAL PROPERTIES*

SILVIA ZANE
Mullard Space Science Laboratory, University College London,
Holmbury St Mary, Dorking, Surrey, RH56NT, UK
sz@mssl. ucl. ac. uk

During the meeting, I have been reviewing the current status of X-ray Dim Isolated
Neutron Stars observat.ions, presenting some of the more recent results. A brief summary
is reported here, together with a discussion of some of the proposed interpretations for
the observed phenomena.

1. Introduction

One of the major outcome of the ROBAT mission has been the discovery of a small
group of seven close-by X-ray Dim Isolated Neutron Stars (XDINSs in the following,
see Ref.1,2 for reviews). They are characterized by clean thermal emission at a few
hundred eVs without any trace of contamination from a surrounding supernova
remnant or magnetospheric activity. Therefore they are important targets to study
the direct emission from the neutron star surface layers. Over the last couple of
years, new discoveries have deeply changed our understanding of these sources.

2. Spectral features
Dedicated deep observations with XMM-Newton and Chandra have now shown
that the XDINSs X-ray continuum is well represented by a simple blackbody, and
cannot be reproduced by available atmospheric models (which, however, are all
computed under the simplifying assumption that temperature and magnetic field
strength/ orientation are constant over the star surface). 1,2 No features have been
found whatsoever in the spectrum of the prototype of the class, RXJ1856.5-3754,
not even after 500 ks of Chandra observations. 3 On the other hand, absorption
features have then been detected in the spectra of all the other members of the
XDINSs class. 4- 9 They appear at energies rv 200 - 700 eV and, in some sources,
their characteristics vary with spin phase. Very interestingly, new XMM-Newton
observations have now revealed the presence of multiple lines in a few sources (there
is evidence for three lines in the spectrum of RX .Jl605.3+3249,10 and probably for
two lines in both, RBS1223 11 and RX J0806.4-4123 1O ). Spectral features provide
a powerful tool to test the surface properties in detail. The evidence for multiple
lines gives further support the current idea that they are likely due to either proton
cyclotron resonances and/or bound-free, bound-bound transitions in H, H-like and
He-like atoms in presence of a relatively high magnetic field B ~ 5 X 10 13 _10 14 G.
In fact, a large fine tuning between the atmospheric parameters should be required

*SZ acknowledges PPARC for support through and Advanced Fellowship.

1145
1146

to make these lines (all of them being predicted by the simple current theoretical
models) clustering at roughly the same energy.

3. Timing studies and long terms variations

Evidence for a complex B-field geometry and patchy surface thermal map arises
from the study of the X-ray light curve of XDINSs. 12 ,13 So far, pulsations have been
detected in six sources (with periods in the range 3-11 S),2,14 often with a relatively
large pulsed fraction (rv 1.6%-35% in amplitude). Quite surprisingly, and contrary
to what one would expect in a simple dipolar geometry, often the hardness ratio is
minimum at the pulse maximum. 4,15 In addition, the X-ray light curves exhibit a
certain asymmetry and marked deviations from a pure sinusoidal shape. The most
evident example of this is RBS 1223, that shows a peculiar double-peaked X-ray
pulse profile, in which the separation between the two maxima is different from that
between the two minima. 4,16 As proposed by Ref. 16, this can be explained if the
emission originates from two hot caps which are not exactly antipodal (minimum
spot separation rv 130°16).
A further and possibly stronger evidence for the existence of not antipodal hot
spots comes from the spectacular example of RX J0720.4-3125. Although XDINSs
were unanimously believed to be steady sources, recently XMM-Newton observa-
tions have unexpectedly revealed a substantial change in the spectral shape and
pulse profile of this source over the last few years. 1O ,17,18 Possible variations in the
pulse profile of RX J0420.0-5022 over a similar timescale (rvO.5 years) have also
been reported,7 although at a low significance level. As we have recently shown, 10
precession of a neutron star with two hot spots of different temperature and size
may account for the variations in the RX J0720.4-3125's X-ray spectrum, pulsed
fraction, light curve and soft/hard phase-lag, provided that the two spots are not
antipodal, but located at rv 160° from each other. We inferred a precession period
of 7.1 ± 0.5 yrs, consistent within the errors with that found by an independent
sinusoidal fit to pulse timing residuals from a coherent analysis covering rv 12 yrs
(rv 7.7 ±0.6 yrs). Since the further evolution of spectrum and pulse shape largely de-
pends on the asymmetry of the surface thermal map, these new spectacular results
make the need of a continuous monitoring of XDINSs even more acute.

4. Evidence for a patchy temperature profile?

The large pulsed fraction, pulse asymmetry, long-term variations and the possi-
ble existence of hot spots in non-symmetrical positions, seem difficult to explain
by assuming that thermal emission originates at the star surface, at least if the
thermal surface distribution is induced by a simple core-centered dipolar magnetic
field. 12 ,13,19-21 The appearance of non-antipodal caps with different properties re-
quires a complex and non-axisymmetric thermal surface map. This can be induced
either, by a complex (multipolar) topology in the B-field outside the star, 12,13 or
by the presence of toroidal components in the crust. 22 ,23 Ref.22,23 investigated the
 1147

latter scenario, and found that, if the meridional component of the crustal field
dominates over the radial one, it is relatively easy to generate small hot regions,
with different size and temperature, strongly confined toward the poles (the exter-
nal field is assumed to be a dipole). The main limitation of this model is that, so
far, it has been computed by assuming axisymmetry, therefore it can not mimic the
non-antipodality which seems to be required by observations.
As mentioned before, an alternative way to produce a patchy surface tempera-
ture distribution is to consider a complex topology in the external field. We recently
explored this possibility, 12, 13 by assuming a star-centered dipolar+quadrupolar field.
With no exception, we have found that the broad characteristics of all the XDINSs
light curves can be successfully reproduced by a suitable combination of quadrupo-
lar magnetic field components and viewing angles (although, due to the multidi-
mensionality of the problem, the fit is not necessarily unique). In summary, both
approaches seems to have capabilities (and limitations) in reproducing the observed
data. Deep observations of XDINSs spread over the star spin period (or, in the
case of RX J0720.4-3125, complemented by studies of long terms spectral varia-
tions) may provide a unique discriminating tool, ultimately allowing to unveil the
complex thermal and magnetic surface map of Galactic neutron stars.

References
1. A. Treves et al. PASP 112, 297 (2000)
2. F. Haberl Adv. Sp. Research 33, 638 (2004)
3. V. Burwitz, et al. ACfA , 399, 1109 (2003)
4. F. Haberl, et al. ACfA , 403, L19 (2003)
5. F. Haberl, et al. ACfA , 419, 1077 (2004a)
6. M. H. Van Kerkwjik, et al. ApJ, 608, 423 (2004)
7. F. Haberl, et al. ACfA ,424,635 (2004b)
8. S. Zane et al. ApJ, 627, 397 (2005)
9. M. Cropper, et al. in press in Isolated Neutron Stars: from the Interior to the Surface
(Astroph. & Space Science, eds Zane S., Turolla R. & Page D., 2007)
10. F. Haberl, et al. ABA, 451, L17 (2006)
11. A. Schwope et al. astro-ph/0609705 (2006)
12. S. Zane & R. Throlla MNRAS, 366, 727 (2005)
13. S. Zane in press in Isolated Neutron Stars: from the Interior to the Surface (Astroph.
& Space Science, eds Zane S., Turolla R. & Page D., 2007)
14. A. Tiengo & S. Mereghetti ApjL, submitted, astro-ph/0612501
15. M. Cropper, et al. ABA, 365, L302 (2001)
16. A. Schwope et al. ACfA, 441, 597 (2005)
17. C.P. De Vries, et al. ACfA, 415, L31 (2004)
18. J. Vink, et al. ApJL, 609, L75 (2004)
19. D. Page ApJ, 442, 273 (1995)
20. G.G. Pavlov, et al. ACfA, 289, 837 (1994)
21. D. Page, & A. Sarmiento ApJ, 473, 1067 (1996)
22. U. Geppert, M. Kueker, & D. Page ABA, 426, 267 (2004)
23. U. Geppert, M. Kueker, & D. Page ABA, submitted, (astro-ph/0512530) (2005)
 X-RAY OBSERVATIONS OF NEUTRON STARS AND THE
EQUATION OF STATE AT VERY HIGH DENSITIES

JOACHIM TRUMPER
Max-Planck Instihd Jur- extmterrestTische Physik, Giessenbachstmsse 1, 85741 Gar-ching

Neutron stars (NS) represent observable matter in its densest form. Actually,
the density in the cores of neutron stars exceeds the density of atomic nuclei by
a considerable factor due to the additional compression provided by gravity. The
equation of state (EOS) is poorly known at these high densities. E.g. the EOS
models discussed by Lattimer & Prakash8 show a dispersion of pressures at a fixed
density by a factor of rv 7 for "normal" nuclear matter and even much larger dis-
persion if in addition strange quark matter is taken into account. In principle these
uncertainties can be reduced by nuclear collision experiments (e.g. Danielewicz et
a1. 5 ) or observations of neutron stars. In the following we are dealing with the latter
aspect.
For each EOS model a mass-radius relation can be computed. Fig 1. shows the
results obtained by Lattimer & Prakash. 8 Since the different curves cross each other,
in principle two pairs of masses/radii have to be measured in order to fix the EOS.
Unfortunately, there is not a single neutron star, for which both mass and radius
have been measured. For binary neutron stars masses have been determined with
high precision, giving in M rv l.4M(') which is close to the Chandrasekhar mass of
a white dwarf. The measurement of neutron star radii is much more difficult. There
are many results in the literature, but usually they are rather crude estimates, and
in addition they are model dependent. On the other hand, in order to fix the EOS in
a meaningful way, or to distinguish between neutron stars and strange quark stars,
one needs a precision of rv 10%. l<ortunately, there has been a lot of progress during
the last dozen of years, mainly due to observations by ROSAT, RXTE, BeppoSAX,
Chandra and XMM-Newton. In this paper we want to summarize results from
those observations which in our opinion are most constraining and most reliable
with respect to the models used for interpreting the observational data.

2.5

2.0
"0
~ 1.5
~
::;;
1.0

10 12 14 16 8 10 12 14 16 18
Radius (km) Radius (km)

Fig. 1. Mass-Radius Relations of Neutron Stars and Constraints from Observations.

1148
 1149

1. Coherent Burst Oscillations

The shape of the light curve of coherent X-ray pulsations observed during X-ray
bursts can be modelled by assuming a circular hot spot at the stellar surface, its
geographical latitude, the inclination of the spin axis with respect to the line of
sight, the angular size of the hot spot as well as the width of the emitted beam. The
emitted X-ray beam is then propagated through the space-time including effects
gravitational light bending and frame dragging. Comparing the model light curve
with the observed one for XTE J1814-338 showing 314 Hz pulsations (Bhattar-
charyya et al. l ) find Rsl R < 0.48. The corresponding limit is marked by "1" in fig.
l. We note that similar results have been obtained by Poutanen. 15

2. High Frequency QPO in Low Mass X-Ray Binaries

A generally accepted type of models identifies the high frequency QPO's observed
in accreting NS with the Kepler frequency of the rotating matter at the inner edge
of the accretion disk (Miller et a1. 9 ). Thus this frequency gives an upper limit of the
NS radius. The highest QPO frequency has been observed by van Straaten et al. 17
in 4U 0614+09 at 1330 Hz. This immediately translates into a lower bound in the
M-R diagram marked by "2" in fig. 1 (Miller IO ).

3. Gravitational Redshift of Spectral Features

Several papers have claimed the detection oflines in X-ray burst which are supposed
to originate at the surface of neutron stars (e.g. Nakamura et at,ll Cottam et
al. 4). Unfortunately, the real existence of these features have remained in doubt
because they could not be confirmed by later observations. Therefore we do not use
this method. But we want to point out, that it has a great potential, because in
principle it allows to determine three important parameters: the mass, the radius
and the distance of the neutron star (OzeI 13 ).

4. Radiation Radii of Thermally Emitting Neutron Stars

Blackbody-like spectra have been observed for a number of radio pulsars (e.g. Pavlov
& Zavlin14), for central compact sources in SNR (CeO), and for the radio quiet
isolated neutron stars discovered with ROSAT which have been dubbed "The Mag-
nificent Seven" (c.f. table, for a recent review see Haberl. 6 In order to derive a
radiation radius from observed thermal spectra one needs to know the radiative
properties of the atmosphere (composition, magnetic fields, etc.) and the distance
to the source. The best candidate for such an analysis is RX JI856-3754, the bright-
est of the Magnificent Seven, since its astrometric distance has been measured with
the Hubble Space Telescope, yielding 117 ± 12 pc (Walter & Lattimer l8 ) or more
recently 140 ± 40 pc (Ho 7 ). The X-ray spectrum of the neutron star measured with
high resolution and large photon statistics with the Chandra LETG can be very
well fit by a blackbody with kT = 0.63 ± 0.2 keY, devoid of any spectral features
1150

(Burwitz et al. 3); the opticaljUV-data measured by HST and VLT shows an v 2
dependence typical for a Rayleigh-Jeans spectrum. However, its intensity is a factor
rv 5 higher than the extrapolation of the X-ray blackbody, indicating that the latter

comes from a small hot spot (with a radius of 4.4 ± 0.1 km x dj1l7 pc) (Burwitz
et al. 3) while the opticaljUV is emitted by the cooler bulk surface. A double com-
ponent fit to the overall spectrum yields a blackbody radius Rbb = 16.5 km, while
a fit with a continuous temperature distribution across the stellar surface gives
Rbb = 16.8 km, both for a distance of 117 pc. It has been argued that these figures
represent lower limits to real radiation radius using the argument that a blackbody
is the most efficient radiator for a given temperature distribution (Braje et al., 2
Trumper et aP6). Thus we conclude that the radiation radius is R > 16.5 km. A
model with a thin (lgjcm 2 ), magnetic (3 - 4 x 10 12 G), partially ionized hydrogen
atmosphere on top of a condensed matter surface gives a radiation radius of 17 km
assuming a distance of 140 pc (Ho 7 ). However, this model is not fully self-consistent
and assumes a magnetic field which is one order of magnitude below the estimate
based on observations (c.f. table).

Table: X-ray and optical properties of nearby radio-quiet isolated neutron stars
Object kT Period Amplitude Optical dPjdt Bdb Bcyc
eV s % mag 10- 13 ss-l 10 13 G 10 13 G

RX J0420.0-5022 44 3.45 13 B = 26.6 <92 <18 6.6?

RX J0720.4-3125 85-95 8.39 8-15 B = 26.6 0.698(2) 2.4 5.2
RX J0806.4-4123 96 11.37 6 B > 24 <18 <14 8.0-9.1
RXS J1308.5+2127 86 10.31 18 ffi50ccd = 28.6 <9 <10 2-6
RX J1605.3+3249 96 B = 27.2 9.1-9.7
RX J1856.5-3754 63 7.06 1.5 V = 25.7 <19 4.2(*)
RXS J2143.0+0654 101 9.44 4 R> 23 ~14

(*) based on an age of ~ 5 x 10 5 years

From the radiation radius R (seen by an observer at inifinity) the true stellar
radius Ro can be calculated using the relation R = Ro(l - Rsj Ra)-1/2, where
Rs = 2G M j c 2 is the Schwarzschild radius of the neutron star. The resulting lower
limit to Ro as a function of mass is shown in fig 1 (marked with" 3"), using the
lower limit for Ro of 16.5 km. This implies a rather stiff EOS and tends to exclude
the possibility of RX J1856-3754 being a strang quark star. We note that another
strong support for a stiff equation of state comes from the observation of a high
mass neutron star (M = 2.1 ± 0.2M8) orbiting a white dwarf companion (Nice et
aP2), c.f. fig 1. In summary, we conclude that the observations of neutron stars
provide strong evidence for a stiff equation of state of nuclear matter at super-high
densities.
 1151

References
1. Bhattarcharyya, S., et al.: ApJ 619, 483-491 (2005)
2. Braje, T.M. & Romani, R.W.: ApJ 580, 1043-1047 (2002)
3. Burwitz, V., et al.: ABA 399, 1109-114 (2003)
4. Cottam, J., et al.: Nature 420, 6911, 51-54 (2002)
5. Danielewicz, P., et al.: Science 298, 5598, 1592-1596 (2002)
6. Haberl, F.: Astrophys. Space Science in press (2007)
7. Ho, w.e.G.: MNRAS 375, 7, 821 (2007)
8. Lattimer, J.M. & Prakash, M.: ApJ 550, 426-442 (2001)
9. Miller, M.C., et al.: ApJ 508, 791-830 (1998)
10. Miller, M.C.: in Proceedings of X-Ray Timing 2003, Rossi and Beyond (2003)
11. Nakamura, N., et al.: Publ.Astron.Soc.Japan 40, 209 (1988)
12. Nice, D.J., et al.: ApJ 634, 1242 (2005)
13. Ozel, F.: Nature 441, 1115 (2006)
14. Pavlov, G.G. & Zavlin, V.E.: astro-ph 0305435 (2003)
15. Poutanen, J.: astro-ph 0401209 (2004)
16. Triimper, J., et al.: Nuclear Physics B, Proc.Suppl., Vol. 132C, 560-565 (2004)
17. van Straaten, W., et al.: ApJ 540, 1049 (2000)
18. Walter, F.M. & Lattimer, J.M.: ApJ 576, L145-148 (2002)
 EIGENMODES OF RAPIDLY ROTATING NEUTRON STARS

STRATOSBOUTLOUKOS
Center for Theoretical Astrophysics, Department of Physics, University of Illinois at
Urbana-Champaign, 1110 W Green, 61801, Urbana, IL, USA
[email protected]

Motivated by the perspective of detecting gravitational waves, we use linearized pertur-

bation theory to formulate the eigenvalue problem of rapidly rotating neutron stars in
relativistic gravity and solve for both axisymmetric and non-axisymmetric modes. Next
to the - only bound from below in frequency - pressure mode spectrum, we find an infi-
nite set of inertial modes confined to a finite frequency range; its width depends linearly
on the rotation rate of the star, and is smaller for increasing compactness. The search for
the fundamental r-mode is hindered by the density of the modes and the mixed character
of the mode, but still revealed its presence at about the expected frequency.

1. Introduction
Oscillations of neutron stars are very important for astrophysics as they cary in-
formation about their inner structure and equation of state. 1 Excited in various
scenarios, they could grow in amplitude due to the CFS instability,2 which for the
T-modes is present for all rotation rates of an ideal fluid. The fundamental T-mode
(and possibly the f -mode) are the most relevant for gravitational wave emission.
Despite the various mechanisms that were proposed to limit the growth of such
modes (magnetic fields,3 hyperon viscosity,4mode-mode coupling 5 ), what the satu-
ration amplitude would be and how long a mode will be oscillating at that, is an
unresolved issue that determines the detectability (see Ref. 6 for an overview). The
slow-rotation approximation for studying the problem ignores the flattening of the
star due to rotation and has given contradictory results. 7-9' The computation of
oscillations modes of rapidly rotating relativistic stars has now become feasible1o,1l
and we study here for the first time their complete oscillation mode spectrum.

2. Formulation of the problem

We work on relativistic gravity for a rapidly and uniformly rotating neutron star,
taking the flattening due to rotation into account. We describe the fluid as a poly-
trope with an equation of state given by p = kpb' We solve the equilibrium config-
uration by using the RNS code 12 and in order to compare with results of Ref. 13
we use a sequence of models with I = 2, k=217.86 km 2, fc = 9.4 10 14 grjcm 3 and
rotation frequencies from 0 up to 62% of the breakup limit. For studying small
oscillations on this background it is sufficient to use linearized perturbation theory.
Since the perturbations of the metric do not couple strongly with the fluid pertur-
bations we set 6g/l v = 0 and work only with the pressure perturbation 6p and the
perturbations of the 4-velocity 6Ui. For those we assume an e im ¢ behavior and an
harmonic time-dependence eiut . The equations of motion arise from the perturbed
form of the conservation of energy momentum 6 (T/::/l) = 0 and lead to four first
order partial differential equations.

1152
 1153

In order to numerically solve this system we redifine our variables in such a way
that we eventually form an eigensystem for four variables, AX = iO" X
Due to the ill form of the equations on the rotation axis, we restrict our numerical
e
grid to 0 ~ r ~ Re and 0 < < 7r. We use one-sided differentiation for the equations
inside the grid while at the boundaries we make use of the appropriate conditions:
at the center the variables have to be regular, at the surface the Lagrangian pressure
perturbation vanishes, and close to the axis we use the symmetries of the variables.
We then solve the eigensystem for all possible eigenvalues with the QR algorithm.

3. Testing the code

We first solve the problem in the limit of no rotation where results are available
in the literature. 13 Due to numerical limitations we solve the problem for several
resolutions in the radial direction (n r ) and extrapolate to nr --+ 00. Since even for
a fixed angular resolution (ne) the eigenfrequencies converge roughly as linn we
only increase the radial resolution. The results agree within 10% with the results of
Ref. 13. This gives us confidence for proceeding to rotating stars.

4. Rapid rotation
4.1. General properties of the spectrum
For non-zero rotation the equations are much more lengthy (the full system can be
found in Ref. 14) and the results more complicated. Next to the pressure modes
above rv2 kHz, the spectrum of solutions consists also of inertial modes at lower
frequencies (Fig. 1). As resolution increases, the spectrum of pressure modes extends
to higher frequencies, though bound from below, but the inertial modes stay confined
to a frequency range that for this model is about 600 Hz wide. The total number of
modes though increase linearly with resolution, revealing an infinite inertial mode
spectrum in this finite frequency range. The behavior of both groups of solutions
with resolution can be seen in Fig. 1. For a model rotating with 793 Hz the frequency
width is 1300 Hz, and by extending also to lower rotation rates we find a linear
dependence O"max = 1.67 D a. In the general case of non-axisymmetric perturbations
(m i=- 0), this becomes O"maxlD = 1.64 -1.06m. For less relativistic models (cellO)
this changes to 1.93 -1.03m, close to 2 - m that is known for the Newtonian case. 5

4.2. The fundamental r-mode

The m = 2 case is specially interesting because of the presence of the R = 2 r-
mode, most important for gravitational waves. The search for the mode is not as
straightforward as for pressure modes, since the frequencies are much more densely
spaced and the eigenfunctions have a larger contribution from higher R. We perform

aTypical errors of the coefficients in these fits are 0.1.

1154

6 4

5
3 nr = 101
~ 4 ~
::c
~
::c
2
----
.:::::'...
3 ----
.:::::'...

< 2 <

Fig. 1. A histogram of eigenmodes for a model with 348 Hz rotation in the high (left panel) and
low (right panel) frequency domain for several resolutions in the radial direction.

our search by confining to a frequency range 1.2vs - 1.5vs , where the mode is
expected to lie, and look for eigenfunctions of bU(I in both coordinates that resample
the ones from the slow-rotation analysis,8 i.e. with no nodes in the radial direction
e-
(at the last angular point before the axis) and with a behavior (close to the center)
resampling the pi associated Legandre polynomia.l. For consistency we also check
the eigenfunction of bp and the linT convergence of the frequency. We find such
a mode at 1.41vs and 1.48vs for Vs of 348 Hz and 793 Hz respectively, which is
consistent with the values of (1.2 - 1.5)vs found in Ref. 10.

Acknowledgments
This work was mainly supported by the german science foundation (DFG) through
the Transregio Sonderforschungsbereich 7, and in part by NASA grant NAG 5-
12030, NSF grant AST 0098399, and the funds of the Fortner Endowed Chair at
the University of Illinois.

References
1. N. Andersson and K. Kokkotas, Mon. Not. R. Astron. Soc. 299, 1059 (1998)
2. J. L. Friedman and B. F. Schutz, Astroph. J. 221, 937 (1978)
3. L. Rezzolla, F. K. Lamb, and S. L. Shapiro, Astroph. J. 531, L139 (2000)
4. L. Lindblom and B. J. Owen, Phys. Rev. Lett. 65, 063006 (2002)
5. J. Brink, S. A. Teukolsky, and I. Wasserman, Phys. Rev. D 70, 124017 (2004)
6. N. Andersson and K. Kokkotas, Int. J. Mod. Phys. DlO, 381-442 (2001)
7. Y. Kojima, Mon. Not. R. Astron. Soc. 293, 49 (1998)
8. K. H. Lockitch, J. L. Friedman, and N. Andersson, Ph. Rev. D 68, 124010 (2003)
9. J. Ruoff, A. Stavridis, and K. Kokkotas, Mon. Not. R. Astron. Soc. 339, 1170 (2003)
10. S.'I. Yoshida, S. Yoshida, and Y. Eriguchi, Mon. Not. R. Astron. Soc. 356, 217 (2005)
11. H. Dimmelmeier, N. Stergioulas, and J. A. Font, Mon. Not. R. Astron. Soc. 368 1609
(2006)
12. N. Stergioulas and J. L. Friedman, Astroph. J. 444, 306 (1995)
13. J. A. Font, H. Dimmelmeier, A. Gupta, and N. Stergioulas, Mon. Not. R. Astron. Soc.
325, 1463 (2001)
14. S. Boutloukos and H.-P. NoUert, Phys. Rev. D 75, 043007 (2007)
 PARAMETER SPACE STUDY OF MAGNETOHYDRODYNAMIC
FLOWS AROUND MAGNETIZED COMPACT OBJECTS

SANTABRATA DAS1,* and SAN DIP K. CHAKRABARTI 2 ,3,t

1 A RCSEC, Sejong University, Seoul 143-747, Korea
2 SNBNCBS,JD-Block, Sector III, Salt Lake, Kolkata-98, India
3 Centre for Space Physics, Chalantika 43, Garia Station Rd" Kolkata 700084, India
* [email protected]
t [email protected]

We solve the magnetohydrodynamic (MHD) equations governing axisymmetric flows

around neutron stars and black holes and found all possible solution topologies for adi-
abatic accretion. We divide the parameter space spanned by the conserved energy and
angular momentum of the flow in terms of the flow topologies. We also study the possi-
bility of the formation of the MHD shock waves.

Keywords: accretion, accretion disc - black hole physics-magentohydrodynamics - shock

waves.

1. Introduction

In recent years, the study of magnetohydrodynamic (MHD) accretion flow around

compact objects has become very important since the magnetic field in ubiquitous
in the universe and it should play a part in the black hole accretion, especially
close to the inner disk. The pioneering works of Mestel (1967)1 and Weber & Davis
(1967)2 etc. were further extended by Chakrabarti (1990)3 for rotating compact
stars and obtained a few global solutions for accretion and winds. The study of
MHD flow in Kerr geometry is also carried out by Takahashi et al. (1990).4 They
discussed the properties of smooth trans-Alfvenic flows and focused on the possible
ways to extract energies out of the BH. Nitta et al. (1991)5 presented an analytical
solution of general relativistic Grad-Safranov equation around a rotating BH and
obtained few solution topologies. So far, a fully self-consistent study of the global
solutions around a compact star in terms of flow parameters was not explored. In this
paper, we wish to provide a complete study of the trans-magnetosonic accretion flow
around compact objects in presence of both the radial and toroidal magnetic fields
using Paczynski-Wiita 6 pseudo-Newtonian potential. We discuss how the nature
of the solution topologies depend on the parameter space spanned by the energy
(E) and the angular momentum (L) of the flow. We identify eighteen types of
solution topologies and divide the parameter space accordingly. We also find that
the standing magneto-hydrodynamic shock wave forms which could give rise to
the high energy particles ( few Me V) through the shock acceleration mechanism of
the electrons. These accelerated particles produce power-law synchrotron radiations
which could explain some of the spectral properties of BH candidates. 7 Moreover,
soft photons are energized by the hot electrons of post-shock flow through inverse
Comptonization. In this paper, we provide all possible solutions of MHD flow with
or without shocks and classify the parameter space according to the nature of the
flow solution.

1155
1156

2. Behaviour of MHD shocks, flow solutions and parameter space

An example of complete MHD solution with standing, slow-magnetosonic shock is
presented in Fig. 1a. We consider the Alfven velocity and the radius to be {)a = 1010
cm s~l and Ta = 10 7 cm respectively. The central mass was chosen to be 10M8
for illustration purpose. The flow parameters are (£, L) = (1.5, 1.45). The shock
location is predicted at Ts = 1.153 and denoted by the vertical dashed line. The
solution towards the BH is depicted by a single-arrowed curve while the sub-Alfvenic
double-arrowed curve is appropriate for a shock free neutron star solution. The
dash-dotted vertical line with triple-arrow represents a typical shock transition at
the boundary of the neutron star. The hot and dense post-shock flow is the most
important region since it is responsible for generating the hard X rays.
In Fig. 1b, we draw the parameter space in £-L plane and divide it into various
regions according to the nature of the solution topologies. We obtain the curve ADE
considering the special case where the radial velocity at all the magnetosonic points

(a) (b)

(c) (d)
Fig. 1. (a) A complete solution topology with standing magnetohydrodynamic shock wave. s
(b) Division of the parameter space according to the nature of solution topologies 8 (c-d) MHD
flow solutions for different (E, L) pair. Radial distance is plotted along horizontal axis while radial
velocity is plotted along vertical axis. Greek alphabets mark various types of solutions drawn with
parameters from different regions marked Fig. lb. 8
 1157

vanishes. The region shaded with dot-dashed lines is forbidden for the flow solution.
The curve PG is obtained when flow velocity is identical to the Alfven velocity at
the magnetosonic point. The regions bounded by PBEGHP, ABCFGUA, RDEQR
and QGUQ possess Bondi-like slow, Rotational-slow, Rotational-fast and Bondi-
like fast magnetosonic points respectively. The parameters from the region shaded
with dots (SKTS) allow magnetosonic shock transition. The region surrounded by
PCKOTUQP may produce non-steady shock since shock conditions are not satisfied
but entropy at the inner sonic point is higher than the outer sonic point.
We further sub-divide the parameter space according to the solution topologies.
We identified eighteen distinct types of solutions as shown in Fig. 1 (c-d). The ac-
cretion solutions for BH and neutron star differs only through the inner boundary
conditions. Therefore, for a set of flow parameters, one needs to assign the proper
boundary conditions to decide whether the solution is for BH or for neutron star.
We marked each solution by a Greek alphabet a, (3, .. etc which denote the pa-
rameters in £-L plane (Fig. Ib) for which the solutions are drawn. In Fig. l(c-d),
the Bondi-like (slow/fast), Rotational slow and Rotational fast magnetosonic points
are denoted by ml, m2, and m3 respectively. For L = 0, the solution reduce to the
Bondi flow solution.

3. Concluding Remarks
We identified all possible solutions including those containing standing shocks in
a MHD accretion flow around compact objects and divided the parameter space
according to the nature of flow solutions. We identified another region where the
flow may exhibit non-steady shocks very similar to those in a hydrodynamic flow.
The post-shock flow inverse Comptonizes soft photons either from the cooled Kep-
lerian disk or from the synchrotron radiation and re-emits them as hard X-rays.7 In
addition, high frequency QPOs of hard X-ray could be obtained since MHD shocks
form closer to the BH horizon.

Acknowledgments: This work is partly supported by a project (Grant No. SP /S2/K-

15/2001) funded by DST, India. SD is thankful for financial support to KOSEF
through ARCSEC, Korea.

References
1. Mestel, L., 1967, Plasma Astrophysics, ed. Sturrock, P. A., Academic Press, New York.
2. Weber, E. J. & Davis, L. Jr., 1967, ApJ 148, 217.
3. Chakrabarti, S. K., 1990, MNRAS 246, 134.
4. Takahashi, M., Nitta, S., Tatematsu, Y. & Tomimatsu, A., 1990, ApJ 363, 206.
5. Nitta, S., Takahashi, M. & Tomimatsu, A., 1991, Phys. Rev. D 44, 2295.
6. Paczynski E., Wiita P.J., 1980, A & A 88, 23.
7. Chakrabarti, S. K. and S. MandaI, 2006, ApJ, 642, L49.
8. Das, S., Chakrabarti, S. K., 2007, MNRAS, 374, 729.
 GRAVITATIONAL RADIATION FROM ACCRETING
MILLISECOND PULSARS

MATTHIAS VICELIUS

DONALD PAYNE

ANDREW MELATOS
School of Physics,
University of Melbourne,
Parkville, VIC 3010, Australia
[email protected]

It is widely assumed that the observed reduction of the magnetic field of millisecond
pulsars can be connected to the accretion phase during which the pulsar is spun up by
mass accretion from a companion. A wide variety of reduction mechanisms have been
proposed, including the burial of the field by a magnetic mountain, formed when the
accreted matter is confined to the poles by the tension of the stellar magnetic field. A
magnetic mountain effectively screens the magnetic dipole moment. On the other hand,
observational data suggests that accreting neutron stars arc sources of gravitational
waves, and magnetic mountains are a natural source of a time-dependent quadrupole
moment. We show that the emission is sufficiently strong to be detectable by current
and next generation long-baseline interferometers. Preliminary results from fully three-
dimensional magnetohydrodynamic (MHD) simulations are presented. We find that the
initial axisymmetric state relaxes into a nearly axisymmetric configuration via toroidal
magnetic modes. A substantial quadrupole moment is still present in the final state,
which is stable (in ideal MHD) yet highly distorted.

1. Introduction
Despite considerable effort, an unequivocal direct detection of gravitational waves
(GW) is yet to be achieved. The expected wave strain is several orders of magnitude
weaker than the sensitivity of current interferometric detectors. 1 One possibility is
to coherently integrate the signal of a continuous source. In this case, the signal-
to-noise ratio increases with the square root of the observation time. 2 A variety
of physical mechanisms for the generation of continuous gravitational waves have
been proposed,3,4 among them nonaxisymmetric distortions of the neutron star
crust, either due to temperature variations 5,6 or strong magnetic fields,7 r-mode
instabilities,8-11 or free precession.1 2 ,13
A promising GW source was recently suggested by two of us. 14 Matter accreting
onto a neutron star in a low-mass X-ray binary (LMXB) accumulates at the mag-
netic poles until the latitudinal pressure gradient overcomes the magnetic tension
and the plasma spreads equatorwards. The frozen-in magnetic field is carried along
with the spreading matter, and is therefore compressed, until the magnetic tension
is again able to counterbalance the thermal pressure. This configuration is termed a
magnetic mountain. 15 During this process, the magnetic dipole moment decreases
with accreted mass, consistent with observational data. 15 ,16
We examine the prospect of detection of GW from magnetic mountains in section
2. In section 3, we present preliminary results from fully three-dimensional magneto-

1158
 1159

hydrodynamic (MHD) simulations to test the mountains stability.

2. Gravitational radiation
A typical mountain with Ma ~ 1O- 4M 0 and pre-accretion dipolar magnetic field
of B = 10 12 G can provide 14 a gravitational ellipticity E = Ih - hl/h ~ 10- 5,
where hand 13 are the principal moments of inerta. E is considerably higher than
the deformation a conventional neutron star could sustain (E ~ 10- 7) via its free
elastic response and is only surpassed by exotic solid strange stars 17 (E ~ 10- 4).
The characteristic GW strain 18 is defined as he = (1281T 4 /15)1/2G1 zz j2E/(dc4 ),
where 1zz ~ 1045 g cm 2 is the principal moment of inertia, f the spin frequency,
and d the distance to the object. Fig. 1 (left) shows he as a function of f at a
distance of d = 1kpc and a mountain mass of 10- 8M0 ::; Ma ::; 10- 2 M0 together
with the design sensitivities of LIGO and advanced LIGO for a coherent integration
time a of 10 7 s. A mountain with Ma ~ 10- 6 M0 should be detectable by LIGO for
f > 200 Hz. This is also consistent with the observed cutoff at '" 700 Hz in the spin
frequency distribution of LMXBs - much slower than the breakup frequency.5

3. Three-dimensional hydromagnetic stability

Surprisingly, the distorted magnetic configuration is stable to axisymmetric
modes. 19 However, the full three-dimensional stability is yet to be examined. We

aThis is currently too optimistic due to computational limitations. The 82 run managed to reduce
a five hour stream of data. 3 Improvements are expected using added computational resources and
hierarchical search strategies.

10 100 1000
f [Hz]

(a) (b)

Fig. 1. (a) Characteristic wave strain he for mountain masses 10- 8 Mev :::; Ma :::; 1O- 2 1i1ev along
with the design sensitivity of LIGO and Advanced LIGO for a 99 per cent confidence and an
integration time of 10 7 s. (b) Time evolution of the quadrupole tensor. The left and righ axes
scale the diagonal and off-diagonal components, respectively. The time base is the radial Alfven
crossing time TA = 5.4 X 10- 7 s.
1160

perform three-dimensional simulations by loading the axisymmetric configuration

into the ideal MHD-code ZEUS-MP. A preliminary result is displayed in Fig. 1 (right).
Shown is the time evolution of the three cartesian quadrupole moments Q22, Q33,
and Q12, defined as Qij = J d3x' (3x~xj - r,25ij )p(x'), where p denotes the den-
sity. After readjusting initially, the system settles down into a state that still has a
sizeable quadrupole moment. Furthermore, the small magnitude of the off-diagonal
element Q12 suggests that the mountain is still nearly (within :::::1 %) symmetric
about the magnetic axis.
We tentatively interpret these results as a preliminary proof of three-dimensional
stability. However, the influence of resistivity still needs to be examined. Resistive
ballooning and resistive Rayleigh-Taylor modes may allow plasma slippage on a
short timescale. 2o Non-ideal MHD simulations to investigate these effects are cur-
rently under way.

References
1. J. A. Lobo, LNP Vol. 617: Current Trends in Relativistic Astrophysics 617,210 (2003).
2. P. Jaranowski, A. Krolak and B. F. Schutz, Phys. Rev. D 58, 063001 (1998).
3. The LIGO Scientific Collaboration, Preprint(gr-qc/0605028j (2006).
4. B. J. Owen, Classical and Quantum Gravity 23, I(April 2006).
5. L. Bildsten, ApJ 501, L89+ (1998).
6. G. Ushomirsky, C. Cutler and L. Bildsten, MNRAS 319, 902 (2000).
7. C. Cutler, Phys. Rev. D 66, 084025 (2002).
8. N. Stergioulas, Living Reviews in Relativity 6, 3 (2003).
9. B. J. Owen, L. Lindblom, C. Cutler, B. F. Schutz, A. Vecchio and N. Andersson,
Phys. Rev. D 58,084020 (1998).
10. N. Andersson, K. D. Kokkotas and N. Stergioulas, ApJ 516,307 (1999).
11. M. Nayyar and B. J. Owen, Phys. Rev. D 73, 084001(April 2006).
12. D. 1. Jones and N. Andersson, MNRAS 331, 203(March 2002).
13. C. Van Den Broeck, Classical and Quantum Gravity 22, 1825(May 2005).
14. A. Melatos and D. J. B. Payne, ApJ 623, 1044(April 2005).
15. D. J. B. Payne and A. Melatos, MNRAS 351, 569(June 2004).
16. A. Melatos and E. S. Phinney, Publications of the Astronomical Society of Australia
18, 421 (2001).
17. B. J. Owen, Physical Review Letters 95,211101 (2005).
18. P. R. Brady, T. Creighton, C. Cutler and B. F. Schutz, Phys. Rev. D 57, 21Ol(February
1998).
19. D. J. B. Payne and A. Melatos, MNRAS, in press (2007).
20. Arons, private communication.
 DYNAMICAL STABILITY OF FLUID SPHERES IN SPACETIMES
WITH A NONZERO COSMOLOGICAL CONSTANT*

STANISLAV HLEDIKt, ZDENEK STUCHLtK+ and KRISTINA MRAzovAO

Institute of Physics, Faculty of Philosophy and Science, Silesian University in Opava,
Bezruco'Vo nam. 13, Opava, CZ-'l46 01, Czech Republic
E-mail: [email protected]@[email protected]

The Sturm-Liouville eigenvalue equation for eigenmodes of the radial oscillations is de-
termined for spherically symmetric perfect fluid configurations in spacetimes with a
nonzero cosmological constant and applied in the cases of configurations with uniform
distribution of energy density and polytropic spheres. It is shown that a repulsive cosmo-
logical constant rises the critical adiabatic index and decreases the critical radius under
which the dynamical instability occurs.

Keywords: Perfect fluid configurations; Cosmological constant; Dynamical stability.

1. Introduction

The internal Schwarz schild spacetimes with nonzero cosmological constant (A i=- 0)
and uniform distribution of energy density were given for star-like configurations l
and extended to more general situations. 2 The polytropic and adiabatic spheres were
preliminary treated and compared,3,4 neutron star models with regions of nuclear
matter described by different relativistic equations of state that are matched were
also treated. 5 Their stability can be considered on energetic grounds 6 or it can be
treated in dynamical way.7 Here we determine the dynamical stability conditions
for the uniform density and polytropic spheres using the approach of Misner et a1. 8

2. Sturm-Liouville Equation
Interior of a spherically symmetric configuration is described (in standard Schwarz-
schild coordinates) by the line element

(1)

with metric coefficients taken in the general form

\It = \It(T,t) , <P = <p(r, t). (2)

The perfect fluid distribution is given by energy density and pressure profiles p(T, t)
and p(T, t). The static equilibrium configuration is given by <PO(T), \(fO(T), POeT),
PaCT) satisfying the Euler equations. The radially pulsating configuration is then
determined by

q(T, t) = qO(T) + Oq(T, t), oq == (O<P, OW, Op. op, on), (3)

*This research has been supported by Czech grant MSM 4781305903.

1161
1162

where n is the number density. The pulsation is given by the radial displacement
~ = ~(r, t). The Euler perturbations I5q are related to the Lagrangian perturbations
measured by an observer who moves with the pulsating fluid by the relation

liq(r, t) = q(r + ~(r, t), t) - qo(r) ~ I5q + qb~. (4)

Introducing a new variable ( == r2e-<[>o~, the radial pulsations are governed by

W( = (P(')' + Q( (5)
with the functions W(r), P(r), Q(r) determined for the equilibrium configuration
1
W == (Po + Po) 2" e 3 'lT o +<l>o , (6)
r
1
P == ,Po 2" e'lTo+3<l>o , (7)
r
(p'?
o _ _4p'
[ Po + _0 1 - (Po + Po) (87rG)
--Po - A - -
2
e 'lT
O
]
(8)
Po r2 r3 c4 r2
The linear stability analysis can be realized by the standard assumption of the
displacement decomposition

((r, t) = ((r)e iwt . (9)

Then the dynamic equation reduces to the Sturm-Liouville equation and the related
boundary conditions in the form

(10)
e<l>o
finite as r ---+ 0, ,Po - 2 (' ---+ 0 as r ---+ R. (11)
r
The Sturm-Liouville equation (10) and the boundary conditions determine eigen-
frequencies Wj and corresponding eigenmodes (i(r), where i = 1,2, ... , n. The ei-
genvalue Sturm-Liouville (SL) problem can be expressed in the variational form 8
as the extremal values of
W 2 = foR (P('2 - Q(2) dr (12)
foR W(2dr
determine the eigenfrequencies Wi and the corresponding functions (i (r) are the
eigenfunctions that have to satisfy the orthogonality relation

foR e3'lTo-<l>O(po + po)r2~(i)~(j) dr = O. (13)

3. Results and Conclusions

We applied the Sturm-Liouville approach to spheres with uniform energy density, 1
and polytropic spheres. 3 The case of uniform spheres can be properly taken as a test
bed of the dynamical instability problem - although these solutions of the Einstein
equations are of rather artificial character, they reflect quite well the basic properties
 1163

2 ----------
-----
o LLLL _ _ _ _ _ _ _ _ _ _ _ _ _- '

19/8 2 3 4
R/rg

Figure 1. Dependence of the critical value of adiabatic index IC on sphere radius R. Full curve:
vanishing cosmological constant A = 0; then IC diverges as R ---+ 9r g /8 from above. Dashed curve:
positive cosmological constant A = 0.1, the point of divergence is shifted to 1.18421 > 9/8, and
IC,\>o > IC,\=o, Dashed-dotted curve: negative cosmological constant A = ~0.1, the point of
divergence is shifted to 1.07143 < 9/8, and Ic,.\<o < Ic,.\=o,

of very compact objects. 9 The dependence of the critical value of adiabatic index
r 0= (8Inp/8Inn)s = (n/p)(f),p/f:m) on configuration radius R is for the uniform
case given in Fig. 1 (see also 12 ). The polytropic case is treated in details by Stuchlik
and Hledik. 10

Bibliography
1. Z. Stuchlik, Acta Phys. Slovaca 50, 219 (2000).
2. C. G. Bohmer, Gen. Relativity Gravitation 36, p. 1039 (2004).
3. Z. Stuchlik and S. Hledik, General Relativistic Poly tropes with a Nonzero Cosmolo-
gical Constant, In preparation, (2006).
4. S. Hledik, Z. Stuchlik and K. Mrazovi, General relativistic adiabatic fluid spheres
with a repulsive cosmological constant, In preparation, (2006).
5. M. Urbanec, J. Miller and J. R. Stone, Matching of equations of state: influence on
calculated neutron star models pp. 357-362. In Hledik and StuchlikY
6. R. F. Tooper, Astrophys. J. 140, 434 (1964).
7. S. Chandrasekhar, Astrophys. J. 140, p. 417 (1964).
8. C. W. Misner, K. S. Thorne and J. A. Wheeler, Gravitation (Freeman, San Francisco,
1973).
9. N. K. Glendenning, Phys. Rev. C 37, p. 2733 (1988).
10. Z. Stuchlik and S. Hledik, Radial pulsations and dynamical stability of spherically
symmetric perfect fluid configurations in spacetimes with a nonzero cosmological con-
stant pp. 209-222. In Hledik and Stuchlik. l l
11. S. Hledik and Z. Stuchlik (eds.), Proceedings of RAGtime 6/1: Workshops on black ho-
les and neutron stars, Opava, 16-18/18-20 September 2004/2005 (Silesian University
in Opava, Opava, 2005).
12. C. G. Bohmer and T. Harko, Dynamical instability of fluid spheres in the presence of
a cosmological constant, Phys. Rev. D 71, p. 084026 (2005), arXiv: gr-qc/0504075.
This page intentionally left blank
Strange Stars
This page intentionally left blank
 STRANGELETS IN COSMIC RAYS

JES MADSEN
Department of Physics and Astronomy, University of Aarhus,
DK-8000 Arhus C, Denmark
[email protected]

The properties of strangelets are reviewed and two experiments searching for them in
cosmic rays are described. The prospects for strangelets as ultra-high energy cosmic rays
beyond the classical GZK-cutoff are discussed.

1. Bulk quark mattera

Quark matter composed of u and d-quarks is unstable except for 3-quark baryons
(otherwise nuclei would decay to quark matter). An additional Fermi sea with a third
flavor, s, makes it possible to reduce the energy by increasing the spatial density of
quarks. The s-quark mass is high compared to that of u and d, so stability is most
likely for low s-quark mass. 2- 5
The Fermi momentum, PFi, of a non-interacting, massless quark-flavor, i, equals
the chemical potential, f..li. The number density is ni = f..lU 7r 2, the energy density
Ei = 3f..lt/(47r 2), and the pressure Pi = f..lt/(47r 2). The sum of the quark pressures is
balanced by the confining bag pressure, B; Li Pi = B; the total energy density is E =
Li Ei + B = 3 Li p;. + B = 4B, and the baryon number density is nB = Li ni/3. For
a gas of u and d-quarks charge neutrality requires nd = 2n u , or f..l2 == f..lu = 2- 1/ 3f..ld.
Thecorrespondingtwo-flavorquarkpressureisP2 = Pu+Pd = (1+2 4/ 3)f..li/(47r 2) =
B, the total energy density E2 = 3P2 + B = 4B, and the baryon number density
nB2 = (nu + nd)/3 = f..lV7r 2, giving an energy per baryon of

EdnB2 = (1 + 24/ 3)3/4(47r 2)1/4 B 1/ 4 = 6.441B 1/ 4 ~ 934MeVBi~:, (1)

where Bi~: == B 1 / 4 /145MeV; 145MeV being the lowest possible choice that avoids
instability of ordinary nuclei.
A three-flavor quark gas is electrically neutral for nu = nd = n s, i. e. f..l3 ==
f..lu = f..ld = f..ls· For fixed bag constant the three-quark gas should exert the same
pressure as the two-quark gas (leaving also the energy density, E3 = 3P3 + B = 4B,
unchanged). That happens when f..l3 = [(1 + 2 4/ 3)/3j1/4f..l2, giving a baryon number
density of nB3 = f..lV7r 2 = [(1 + 24/ 3)/3j3/4nB2 . The energy per baryon is then

E3/nB3 = 3f..l3 = 33/4(47r2)1/4B1/4 = 5.714B 1/ 4 ~ 829MeVBi~:; (2)

lower than in the two-quark case by a factor nB2/nB3 = (3/(1 + 24/ 3))3/4 ~ 0.89,
so one may gain of order 100 MeV per baryon by introducing an extra flavor.
For massive s-quarks weak equilibrium gives f..ls = f..ld = f..lu + f..le, where f..le > 0
corresponds to a small but non-zero density of electrons required for local charge

aSections 1 and 2 are partly based on Ref. 1.

1167
1168

neutrality. For reasonable values of Tns of order 100 MeV stable strange quark matter
remains possible for a range of B.
Quark matter at asymptotically high density has an interesting property called
color superconductivity.5,7 Even the weakest attraction (and such attractions exist
in QCD) leads quarks of different colors and flavors to form pairs, much like Cooper
pairs in a superconductor, except that the binding in QCD is caused by a direct
attraction channel rather than via indirect phonon interaction. The binding energy
of a pair, Do, can be very large, ranging from a few MeV to over 100 MeV. In general
these systems are called color superconductors. If all colors and flavors pair in an
equal manner one talks about color-flavor locking.
While color-flavor locking seems generic in the infinite density limit, the proper-
ties of strange quark matter at densities of order or somewhat higher than nuclear
matter density is at the focus of much current research and discussion. This is the
density regime of relevance for strangelets, strange stars, and for quark matter cores
in hybrid stars (the analogs of neutron stars if quark matter is metastable so that it
forms above a certain density in compact star interiors). An additional binding en-
ergy per baryon of approximately -3Do 2 / f-L is introduced in these systems, meaning
that extra binding of 10 - 100 MeV is not unrealistic, thus significantly increasing
the likelihood of absolutely stable strange quark matter and strangelets.

2. Strangelets
For A « 10 7 the quark part of strange quark matter is smaller than the Comp-
ton wavelength of electrons, so electrons no longer ensure local charge neutrality.
Therefore Coulomb effects have to be included, though the fortuitous cancellation
of qu + qd + qs = ~ - ~ - ~ = 0 means that Coulomb energy is much less important
for strangelets than for nuclei. In particular strangelets do not fission. For A < 10 3
other finite size effects such as surface tension and curvature have to be taken into
account.
Several strange let searches with relativistic heavy-ion collisions as well as cosmic
ray searches have been carried out, and others are planned for the future. 8 - 10 Most
of these searches are sensitive only to low A-values, so it is important to know the
properties of small lumps of strange quark matter (strangelcts).
Mode-filling for large numbers of quarks in a spherical MIT-bag l l was performed
for ud-systems in Ref. 12, and for strangelets in Refs. 5,13-17. All of these calcula-
tions were performed for strong fine-structure constant as = O.
The energy per baryon smoothly approaches the bulk limit for A --+ 00, whereas
the energy grows significantly for low A. For low s-quark mass shells are found for
A = 6 (3 colors and 2 spin orientations per flavor), and less conspicuous ones for
A = 18, 24, 42, 54, 60, 84, 102 etc. As Tns increases it becomes more and more
favorable to use u and d rather than s-quarks, and the "magic numbers" change;
for instance the first closed shell is seen for A = 4 rather than 6.
Strangelet mode-filling calculations can be modified by inclusion of Coulomb
 1169

energy and 2ero-point fluctuation energy. The Coulomb energy is generally small.
The zero-point energy is normally included as a phenomenological term of the form
- Zo/ R, where fits to light hadron spectra indicate the choice Zo = l.84. 15 Roughly
half of this phenomenological term is due to center-of-mass motion. The proper
choice of as and Zo is tricky. As discussed by Farhi and Jaffe 5 the values are
intimately coupled to Band Tn s , and it is not obvious that values deduced from bag
model fits to ordinary hadrons are to be preferred in the case of strangelets. This
uncertainty may have an important effect for A < 5~ 10, but the zero-point energy
quickly becomes negligible for increasing A as it decreases like A ~1/3. It means,
however, that it is difficult to match strangelet calculations to experimental data
concerning ordinary hadrons or limits on the putative A = 2 H-dibaryon.
Mode-filling calculations give the "correct" results as far as the model can be
trusted, but for many applications a global mass-formula analogous to the liquid
drop model for nuclei is of great use and also gives further physical insight.
Berger and Jaffe 18 made such a detailed analysis within the MIT bag model.
They included Coulomb corrections and surface tension effects stemming from the
depletion in the surface density of states due to the mass of the strange quark. Both
effects were treated as perturbations added to a bulk solution with the surface con-
tribution derived from a multiple reflection expansion. A self-consistent treatment
including also the very important curvature energy was given in Refs. 16,19,20.
In the ideal Fermi-gas approximation the energy (mass) of a system composed
of quark flavors i is given by

(3)

Here ni , Hi and JLi denote thermodynamic potentials, total number of quarks, and
chemical potentials, respectively. B is the bag constant, V is the bag volume, and
ECoul is the Coulomb energy.
In the multiple reflection expansion framework of Balian and Bloch,21 the ther-
modynamical quantities can be derived from a density of states of the form
2
dH i { k V
dk = 6 27T2 + fs (Tni
k ) kS + fe ( kTni ) C + .... } , (4)

where area S = 1dS (= 47T R2 for a sphere) and extrinsic curvature C =

J' (k + JJ dS (= 87T R for a sphere). Curvature radii are denoted Rl and R2·
For a spherical system RI = R2 = R. The functions 1s and 1e are given by 16,18

fs ( k
Tn)
=- 1{ (2)
87T 1- ;: tan ~I Tn
k } ' (5)

Tn) =
fe (k 1 { 3 (7T
127T2 1 - '2 Tn '2
k - tan
~I k)}
Tn .
(6)
1170

In terms of volume-, surface-, and curvature-densities, ni,v, ni,S, and ni,e, the
number of quarks of flavor i is
rkF'dNi
Ni = Jo dk dk = ni,vV + ni,sS + ni,e C , (7)

with Fermi momentum kFi = (IL; - m7)1/2.

The corresponding thermodynamic potentials are related by

!li = !li,V V + !li,SS + !li,e C , (S)

where O!li/OlLi = -Ni' and o!li,j/OlLi = -ni,j' For massless quarks !li,S = ni,S = 0;
!li,e = ILUS7r 2 ; ni,e = - lLi/ 47r2 .
Minimizing the total energy at fixed Ni gives the pressure equilibrium constraint
2 2
B =- "
Lt! l v - -R "
2,
!l S
Lt', - - ! l " e.
R2"Lt (9)

The optimal composition for fixed baryon number, A, can be found by minimiz-
ing the energy with respect to Ni giving

O -- ,
Lt, (IL, + OECOU1)
oN
0 dN,. (10)
, o ,

For uncharged bulk quark matter one arrives at the usual energy per baryon

(11)

where superscript 0 denotes bulk values. The energy minimization, Eq. (9), corre-
sponds to

(12)

where the last equality assumes massless quarks. In the bulk limit the baryon num-
ber density is given by

(13)

and one may define a bulk radius per baryon as

R O = (3/47rn~)1/3. (14)
For quark matter composed of massless u, d, and s-quarks, the Coulomb energy
vanishes at equal number densities due to the fact that the sum of the quark charges
is zero. Thus it is energetically most favorable to have equal chemical potentials
for the three flavors. From the equations above one may derive the following bulk
expressions for 3-flavor quark matter:

° (47r2
lLi =-3-
B) 1/4 -1.905B
_ 1/4 _
-
1/4.
276.2MeVB 145 , (15)
 1171

(16)

R O = (3/47fn~)1/3 = 0.699B- 1/ 4 . (17)

And the energy per baryon is

(18)
in agreement with Eq. (2).
To first order one may regard Coulomb, surface, and curvature energies as per-
turbations on top of the bulk solution. IS In this approach one gets a strange let mass
M (= E)

;::::: [829AMeV + 351MeVA 1/ 3 ] Bi~i. (19)

For ms > 0 and ,\ == ms/ fLs the energy minimization, Eq. (12), changes to

B = " (fL?)4 + (fL~)4 [(1- ,\2)1/2(1 _ ~,\2) + ~,\4ln 1 + (1- ,\2)1/2], (20)
~ 47f2 47f2 2 2 ,\
i=u,d

and the baryon number density is now given by

n~ ~3
= ["
~
(fL?)3
7f2
+ (fL~)3
7f2
(1 - ,\2)3/2] (21)
i=u,d

A bulk radius per baryon is still defined by Eq. (14).

In bulk equilibrium the chemical potentials of the three quark flavors are equal,
fL~ = fL~ = fL~ == flO = EO /3. Neglecting Coulomb energy one may approximate the
mass of small strangelets as a sum of bulk, surface and curvature terms, using the
chemical potential calculated in bulk:
(22)

where SO = 47f(R°)2 A 2 /3 and CO = 87f(RO )A 1/ 3 . Masses in MeV for Bl/4 = 145MeV

are (with s-quark mass in MeV given in parenthesis)

M(O) = 829A + OA 2 / 3 + 351A 1 / 3 (23)

M(50) = 835A + 61A 2 / 3 + 277 A 1/ 3 (24)
M(150) = 874A + 77 A 2 / 3 + 232A 1 / 3 (25)
M(200) = 896A + 53A 2 / 3 + 242A 1 / 3 (26)
M(250) = 9IlA + 22A 2 / 3 + 266A 1 / 3 (27)
M(300) = 917 A + + 295A 1 / 3
0.3A 2 / 3 (28)
M(350) = 917 A + OA 2 / 3 + 296A 1/ 3 (29)
1172

The lack of a significant Coulomb energy is due to the fortuitous cancellation

of charge +2e/3 up quarks and charge -e/3 down and strange quarks in strange
quark matter with equal numbers of the three quark flavors. Because of the non-zero
s-quark mass the cancellation is not perfect. Typical strangelets have slightly fewer
strange quarks compared to up and down, and therefore the net charge is slightly
positive. A typical model result (to be compared to Z ::::; 0.5A for nuclei) is 22

ms )2
Z = 0.1 ( 150MeV A (30)

Z=8( 150MeV
ms )2Al/3 (31)

for A « 700 and A » 700 respectively (the slower growth for higher A is a conse-
quence of charge screening).
Thus a unique experimental signature of strange lets is an unusually high mass-
to-charge ratio compared to nuclei.
Cooper pairing involves quarks with equal (but opposite) momenta, so the nat-
ural ground state of a color-flavor locked system has equal Fermi momenta for up,
down, and strange quarks and therefore equal number densities. Thus the total
net quark charge is zero for a bulk system. 23 A finite strangelet has a surface sup-
pression of massive strange quarks relative to the almost massless ups and downs
(massive particle wave functions are suppressed at a surface), so the total charge of
a color-flavor locked strangelet is positive and proportional to the surface area: 24 ,25

Z = 0.3 (
ms ) A2/3. (32)
150MeV
This phenomenon persists even for very large bags, such as strange stars, so
color-flavor locked strange stars also have a positive quark charge. 24 ,26,27
Writing IvI == EDA + csurfA2/3 + ccurvA1/3, with Csurf ;:::::j lOOMeV and Ccurv ~
300MeV, the stability condition M < Am n may be written as A > A~~, where

Anbs = (CSUrf + [c;urf + 4ccurv(mn - fO)]1/2) 3

(33)
ffilll 2(mn _ EO)
Stability at baryon number 30 requires a bulk binding energy in excess of 65 MeV,
which is barely within reach in unpaired strange quark matter for ms > 100MeV
if, at the same time, ud-quark matter shall be unstable. For color-flavor locked
strangelets, stability is more likely. Long-lived metastability with respect to neutron
emission is possible if dEcurv/dA + dEsurr/dA < mn - fO, or A larger than

Affieta = (CHurf + [C;urf + 3ccurv (mn - fO)]1/2)3

(34)
ffilll 3(mn - EO)

To have A~f~a < 30 requires mn - fO > 30 MeV.

Shell effects can have a stabilizing effect not taken into account in the liquid drop
model approach above. As stressed by Gilson and Jaffe 15 the fact that the slope of
 1173

E / A versus A becomes very steep near magic numbers can lead to strangelets that
are metastable (stable against single baryon emission) even for EO > 930MeV.

3. The AMS and LSSS experiments

Strangelet (meta)stability is a theoretical possibility as demonstrated above, but the
existence of small baryon number strangelets is ultimately an experimental issue.
Several experiments have searched for strangelets in cosmic rays. While some
interesting events have been found that are consistent with the predictions for
strangelets, none of these have been claimed as real discoveries. Whether interpreted
as flux limits or as detections these results are consistent with the flux predictions
from strange star collisions of a few strangelets per year per square meter per stera-
dian given in Ref. 28 (see also Refs. 29-33). More specifically the integrated flux of
mass A, charge Z strangelets reaching the inner Solar System was predicted to be 28

(35)
per year per square meter per steradian under the conservative assumption of 10- 10
solar masses per year of strangelets being released in our Galaxy from binary strange
star collisions. The major unknown is the mass spectrum of strangelets released,
but it seems plausible within quite different scenarios that some strangelets have
sufficiently low mass to be detectable at a reasonably high rate. 34 ,35
Two experiments that are currently underway will reach sensitivities that would
provide a definitive strange let detection or rule out a significant part of parameter
space.
AMS-02: The Alpha Magnetic Spectrometer (AMS) is a space-based particle
physics experiment involving several hundred physicists from more than 50 institu-
tions in 16 countries, led by Samuel Ting of MIT. A prototype (AMS-01) flew in
June 1998 aboard the Space Shuttle Discovery,36 and AMS-02 is currently scheduled
to fly to the International Space Station (ISS) in 2008. Once on the ISS AMS-02
will remain active for at least three years. Equipped with a superconducting mag-
net, time-of-flight detectors, trackers, calorimeter, a ring imaging Cerenkov counter,
etcetera, AMS-02 will provide data with unprecedented accuracy on cosmic ray elec-
trons, positrons, protons, nuclei, anti-nuclei and gammas in the GV- TV range and
probe issues such as antimatter, dark matter, cosmic ray formation and propaga-
tion. In addition it will be uniquely suited to discover strangelets characterized by
extreme rigidities for a given velocity compared to nuclei.g,lo AMS-02 will have ex-
cellent charge resolution up to Z ~ 26, and should be able to probe a large mass
range for strangelets. A reanalysis of data from the 1998 AMS-Ol mission has given
hints of some interesting events, such as one with Z = 2, A = 16 37 and another
with Z = 8, but with the larger AMS-02 detector running for 3 years or more, real
statistics is achievable.
LSSS: The Lunar Soil Strange let Search (LSSS) is a search for Z = 8 strangelets
using the tandem accelerator at the Wright Nuclear Structure Laboratory at Yale.
1174

The experiment involves a dozen people from Yale, MIT, and Arhus, led by Jack
Sandweiss of Yale and studies a sample of 15 grams of lunar soil from Apollo II.
lt will reach a sensitivity of 10- 17 over a wide mass range,38,39 sufficient to provide
detection according to the estimates in Ref. 28 if strangelets have been trapped in
the lunar surface layer. In contrast to the deep oceanic and geological mixing on
Earth, the effective mixing depth of the lunar surface is only around one meter,
determined largely by micrometeorite impacts. Therefore the surface layer has an
effective cosmic ray exposure time of around 500 million years. Combined with the
lack of a shielding magnetosphere, this results in an expected strange let concentra-
tion in lunar soil which is four orders of magnitude larger than the concentration
on Earth, where the search for strangelets so far has resulted in upper limits only.

4. Strangelets as ultra-high energy cosmic rays

Strangelets may even provide an explanation for one of the most interesting mys-
teries in cosmic ray physics: The existence of cosmic rays with energies well beyond
10 1g eV, with measured energies as high as 3 x 10 20 eV. 40 ,41 lt is almost impossible to
find a mechanism to accelerate cosmic rays to these energies. Furthermore ultra-high
energy cosmic rays lose energy in interactions with cosmic microwave background
photons, and only cosmic rays from nearby (unidentified) sources would reach us
with the energies measured. Strangelets circumvent both problems, and therefore
provide a possible mechanism for cosmic rays beyond the socalled Greisen-Zatsepin-
Kuzmin (GZK) cutoff: 42
Acceleration: All astrophysical "accelerators" involve electromagnetic fields,
and the maximal energy of a charged particle is proportional to its charge. The
charge of massive strangelets has no upper bound in contrast to nuclei, so highly
charged strangelets are capable of reaching energies much higher than those of
cosmic ray protons or nuclei using the same "accelerator" .42
The GZK-cutoffis a consequence of ultra relativistic cosmic rays hitting a 2.7K
background photon with a Lorentz-factor 'I large enough to boost the 7 x 10- 4 e V
photon to energies beyond the threshold of energy loss processes, such as photo-
pion production or photo-disintegration. The threshold for such a process has a
fixed energy, EThr, in the frame of the cosmic ray, e.g., EThr :=:::; 10MeV for photo-
disintegration of a nucleus or a strangelet, corresponding to '1Thr = EThrl E 2 .7K :=:::;
10 10 , or a cosmic ray total energy

(36)

Since strangelets can have much higher A-values than nuclei, this pushes the GZK-
cutoff energy well beyond the current observational limits for ultra-high energy
cosmic rays.42,43
 1175

5. Conclusion
Strange quark matter may be absolutely stable in bulk, and smaller lumps
(strangelets) can be stable down to some small limiting baryon number. Strangelets
may form in a first-order cosmological quark-hadron phase transition (unlikely),
or in processes related to ccompact stars (more likely). Flux estimates for lumps
reaching our neighborhood of the Galaxy as cosmic rays are in a range that makes
it realistic to either detect them in upcoming experiments like AMS-02 or LSSS,
or place significant limits on the existence of stable strange quark matter. Should
strangelets be discovered it would have profound consequences for our understand-
ing of the strong interactions and for a variety of astrophysical phenomena.

Acknowledgments
This work was supported by the Danish Natural Science Research Council.

References
1. J. Madsen, Lect. Notes Phys. 516, 162 (1999).
2. A. R. Bodmer, Phys. Rev. D 4 1601 (1971).
3. S. A. Chin and A. K. Kerman, Phys. Rev. Lett. 43, 1292 (1979).
4. E. Witten, Phys. Rev. D 30, 272 (1984).
5. E. Farhi and R. L. Jaffe, Phys. Rev. D 30, 2379 (1984).
6. M. G. Alford, K. Rajagopal and F. Wilczek, Phys. Lett. B 422, 247 (1998).
7. R. Rapp, T. Schafer, E. V. Shuryak and M. Velkovsky, Phys. Rev. Lett. 81, 53 (1998).
8. R. Klingenberg, J. Phys. G 27, 475 (2001).
9. J. Sandweiss, 1. Phys. G 30, S51 (2004).
10. E. Finch, 1. Phys. G 32, S251 (2006).
11. A. Chodos, R. L. Jaffe, K. Johnson, C. B. Thorn and V. F. Weisskopf, Phys. Rev. D
9, 3471 (1974).
12. D. Vasak, W. Greiner, and L. Neise, Phys. Rev. C 34, 1307 (1986).
13. C. Greiner, D.-H. Rischke, H. Stocker, and P. Koch, Phys. Rev. D 38, 2797 (1988).
14. K. Takahashi and R. N. Boyd, Astrophys. 1. 327, 1009 (1988).
15. E. P. Gilson and R. L. Jaffe, Phys. Rev. Lett. 71, 332 (1993).
16. J. Madsen, Phys. Rev. D 50, 3328 (1994).
17. J. Schaffner-Bielich, C. Greiner, A. Diener, and H. Stocker, Phys. Rev. C 55, 3038
(1997) .
18. M. S. Berger and R. L. Jaffe, Phys. Rev. C 35, 213 (1987).
19. J. Madsen, Phys. Rev. Lett. 70, 391 (1993).
20. J. Madsen, Phys. Rev. D 47, 5156 (1993).
21. R. Balian and C. Bloch, Ann. Phys. 60, 401 (1970).
22. H. Heiselberg, Phys. Rev. D 48, 1418 (1993).
23. K. Rajagopal and F. Wilczek, Phys. Rev. Lett. 86, 3492 (2001).
24. J. Madsen, Phys. Rev. Lett. 87, 172003 (2001).
25. J. Madsen, Phys. Rev. Lett. 85, 4687 (2000).
26. V. V. Usov, Phys. Rev. D 70, 067301 (2004).
27. M. Stejner and J. Madsen, Phys. Rev. D 72, 123005 (2005).
28. J. Madsen, Phys. Rev. D 71,014026 (2005).
29. J. Madsen, Phys. Rev. Lett. 61, 2909 (1988).
1176

30. J. L. Friedman and R. R. Caldwell, Phys. Lett. B 264, 143 (1991).

31. O. G. Benvenuto and J. E. Horvath, Mod. Phys. Lett. A 4, 1085 (1989).
32. G. A. Medina-Tanco and J. E. Horvath, Astmphys. J. 464, 354 (1996).
33. K. S. Cheng and V. V. Usov, Phys. Rev. D 74, 127303 (2006).
34. J. Madsen, J. Phys. G 28, 1737 (2002).
35. P. Jaikumar, S. Reddy and A. W. Steiner, Phys. Rev. Lett. 96, 041101 (2006);
M. G. Alford, K. Rajagopal, S. Reddy and A. W. Steiner, Phys. Rev. D 73, 114016
(2006).
36. M. Aguilar et al. [AMS Collaboration], Phys. Rept. 366, 331 (2002) [Erratum-ibid.
380, 97 (2003)].
37. V. Choutko, in 28th International Cosmic Ray Conference (Universal Academy
Press, 2003) 1765.
38. K. Han, J. Phys. G 32, S465 (2006).
39. B. Monreal, nucl-ex/0506012.
40. K. Greisen, Phys. Rev. Lett. 16, 748 (1966).
41. G. T. Zatsepin and V. A. Kuzmin, JETP Lett. 4, 78 (1966) [Pisma Zh. Eksp. Tear.
Fiz. 4, 114 (1966)].
42. J. Madsen and J. M. Larsen, Phys. Rev. Lett. 90, 121102 (2003).
43. M. Rybczynski, Z. Wlodarczyk and G. Wilk, Acta Phys. Polan. B 33, 277 (2002).
 CAN STRANGE STARS BE DISTINGUISHED
FROM NEUTRON STARS?

TIBERIU HARKO
Department of Physics and Center for Theoretical and Computational Physics, The University
of Hong Kong, Pok Fu Lam Road, Hong Kong
[email protected]

KWONG SANG CHENG

Department of Physics and Center for Theoretical and Computational Physics, The University
of Hong Kong, Pok Fu Lam Road, Hong Kong
[email protected]

We consider the astrophysical implications of the BCS-BEC crossover in the interior of

quark stars. If the attractive interaction is strong enough, at some critical temperature
the quarks may condense into the bosonic zero mode, forming a Bose-Einstein quark
condensate. By numerically integrating the general relativistic structure equations we
obtain the basic physical propert.ies of t.he gravitationally bounded condensed quark
matter configurations.

1. Introduction
The recent rcalization 1 of new superfiuid states in the ultracold gas of fermionic
alkali atoms 40K and 6Li opened the possibility of the laboratory study of the
long-standing idea of the crossover from the BCS state to the Bose-Einstein con-
densation (BEC). The basic concept of the BCS-BEC crossover is as follows: As
long as the attractive interaction between fermions is weak, the system exhibits the
superfiuidity characterized by the energy gap in the BCS mechanism. On the other
hand, if the attractive interaction is strong enough, the fermions first form bound
bosonic states. Then at some critical temperature they start to condense into the
bosonic zero mode. These two situations are smoothly connected, without a phase
transition. 2 ,3
When compressed, baryonic matter is expected to be deconfined to quark matter,
as a result of a hadron-quark phase transition at high densities and/or tempera-
tures. 4 At ultra-high density, matter is expected to form a degenerate Fermi gas
of quarks in which the Cooper pairs of quarks condensate near the Fermi surface
(color superconductor). If the attractive interaction is strong enough, at some crit-
ical temperature the fermions may condense into the bosonic zero mode, forming a
Bose-Einstein quark condensate. 2,3
It is the purpose of the present paper to consider the possible implications of the
BCS-BEC crossover on the structure and properties of the quark stars. To describe
the physical properties of the condensate we use the Gross-Pitaevskii equation with
quartic non-linearity.

2. The Bose-Einstein quark condensate

The equation describing the Bose-Einstein condensate is the Schrodinger equation
for the condensate wave function 1jJ with a quartic non-linear term: 5
1177
1178

in :t 1/J (T, t) = [- ; : \7
2 + Vext (1") + g' (11/J (T, t)12) ] 1/J (T, t), (1)

where we denoted g' = dg/dp, g (p) = Uo 11/J 14 /2 = uop2/2, Uo = 47rn 2a/m, a

is the coherent scattering length, and Vext is the external potential. The mass of
the condensed particles is denoted by m. As for Vext (f), we assume that it is the
gravitational potential V, Vext = V, and it satisfies the Poisson equation \72V =
47rGpm, where Pm = mp is the mass density inside the Bose-Einstein condensate.
The corresponding equation of state of the quark condensate is a polytrope with
index n = 1, P(Pm) = UoP;", with Uo = 27rn 2a/m 3, where P is the pressure of the
condensate.

3. General relativistic Bose-Einstein quark condensate stars

In the case of the quark condensate with quartic non-linearity the mass continuity
equation and the Tolman-Oppenheimer-Volkoff equation are given by

dMo _ 0 2 dO (1 + .\0) (.\0 21]3 + Mo)

(2)
d1] - 1], d1]
2.\1]2 (1 _ 2~o )

where we have introduced a set of dimensionless variables 1], Mo and 0, respectively,

defined as T = To1], M = m*Mo, Pm = PemO, P = Pemc2.\02, where Pem is the
central (T = 0) value of the energy density of the Bose-Einstein quark condensate,
TO = c/V47rGpem, m* = c3/J47rG3 Pcm and.\ = UOPcm/c2. M and R are the mass
and the radius of the star, respectively.
The mass-radius relation for Bose-Einstein condensate stars with quartic non-
linearity, m = 2m; = 1000 MeV, where m; is the effective mass of the s quark,
and different values of the scattering length a is represented, together with the
mass-radius relation for quark stars obeying the bag model equation of state, in
Fig. 1.

4. Discussions and final remarks

In the present paper we have considered the possibility of the existence of a BCS-
BEC crossover in dense stellar quark matter, and we have analyzed the properties of
the gravitationally bounded Bose-Einstein quark condensates. To obtain the physi-
cal characteristics of the system we have used the hydro dynamical model, in which
quark condensates can be modelled as a gas, whose density and pressure are related
by a barotropic equation of state.
In the framework of the general relativistic approach one must numerically in-
tegrate the structure equations of the star. In this way we obtain a large class of
stable condensed astrophysical objects, whose basic parameters (mass and radius)
sensitively depend on the two parameters (a and m) of our model. The global (like
the mass-radius ratio) properties of the condensed quark stars are very different
 1179

6 8 10 12
R (km)

Fig. 1. Comparison of the mass-radius relation for quark stars obeying the bag model equation
of state (B = 4 X 10 14 g/cm 3 ) (thick curve) and condensate quark stars with m = 2m; = 1000
MeV and different values of a: a = 0.5 fm (solid curve), a = 0.6 fm (dotted curve), a = 0.7 fm
(short dashed curve), a = 0.8 fm (dashed curve), a = 0.9 fm (long dashed curve) and a = 1 fm
(ultra-long dashed curve). For each case the range of pcrn is chosen so that Pcrn 2' 4 x pn and the
causality condition C s ::: c is automatically satisfied.

from the properties of the usual quark stars. Condensed quark stars are bounded
by gravity, and not by the exterior vacuum, as is the case of the "normal" quark
stars. However, condensed stars can still be very small if at very high densities the
s quark is very heavy.
Due to the very low surface photon emissivity of quark matter, quark condensate
stars may be very dark objects. The photon emission properties of quark matter
in the CFL/BEC phase are expected to resemble those of bare strange stars. Since
pairing effects should appear in the plasma frequency wp through the baryon number
density as a correction of order f.L~ 2 . Hence the plasma frequency wp will not differ
much of the typical 20 MeV frequency of unpaired quark matter. Thus we expect
that the equilibrium photon radiation in the quark condensate will be suppressed.
Therefore the very small photon luminosity makes condensate stars very difficult to
detect. Hence observation of quark condensate stars may prove to be an extremely
difficult task.

References
1. c. A. Regal, M. Greiner, and D. S. Jin, Phys. Rev. Lett. 92, 040403 (2004).
2. B. Kerbikov, Phys. Atom. Nucl. 65, 1918 (2002).
3. Y. Nishida and H. Abuki, Phys. Rev. D 72, 096004 (2005).
4. E. Witten, Phys. Rev. D 30, 272 (1984).
5. F. Dalfovo, S. Giorgini, L. P. Pitaevskii and S. Stringari, Rev. Mod. Phys. 71, 463
(1999).
 PAIR WINDS IN SCHWARZSCHILD SPACETIME WITH
APPLICATION TO STRANGE STARS

A.G. AKSENOY
Institute of Theoretical and Experimental Physics,
B. Cheremushkinskaya, 25, Moscow 117218, Russia
alexei. [email protected]

M. MILGROM* and Y.Y. USOyt

Center for Astrophysics, Weizmann Institute,
Rehovot 76100, Israel
* moti. [email protected]
t vladimir. [email protected]

We present the results of numerical simulations of stationary, spherically outfiowing, e±

pair winds, with total luminosities in the range 10 34 - 10 42 ergs s-l. In the concrete
example described here, the wind injection source is a hot, bare, strange star, predicted
to be a powerful source of e± pairs created by the Coulomb barrier at the quark surface.
We find that photons dominate in the emerging emission, and the emerging photon
spectrum is rather hard and differs substantially from the thermal spectrum expected
from a neutron star with the same luminosity. This might help distinguish the putative
bare strange stars from neutron stars.

1. Introduction
For an electron-positron (e±) wind out-flowing spherically from a surface of
radius R there is a maximum pair luminosity, L±ax = 47fmeC3 Rr 2 / O"T
10 36 (R/10 6cm)r 2 ergs S-l, beyond which the pairs annihilate significantly before
they escape, where r is the pair bulk Lorentz factor, and O"T the Thomson cross
section. Recently we developed a numerical code for solving the relativistic ki-
netic Boltzmann equations for pairs and photons in Schwarzschild geometry. Us-
ing this we considered a spherically out-flowing, non-relativistic (r '" 1) pair
winds with injected pair luminosity L± in the range 10 34 - 10 42 ergs S-l, that
is '" (10- 2 - 106)L±ax (Aksenov et al. 2003, 2004, 2005). While our numerical code
can be more generally employed, the results presented in this paper are for a hot,
bare, strange star as the wind injection source. Such stars are thought to be power-
ful (up to '" 10 51 ergs S-l) sources of pairs created by the Coulomb barrier at the
quark surface (Usov 1998, 2001).

2. Formulation of the problem

We consider an e± pair wind that flows away from a hot, bare, unmagnetized, non-
rotating, strange star. Space-time outside the star is described by Schwarzschild's
metric with the line element
ds 2 = _e 2q, c2dt 2 + e- 2q, dT2 + T2 (drJ 2 + sin2 rJ di.p2) , (1)
where eq, = (1 - Tg/T)1/2 and Tg = 2GM/c2 '::::' 2.95 x 105(M/Mc:)) cm.

1180
 1181

We use the general relativistic Boltzmann equations for the pairs and photons,
whereby the distribution function for the particles of type i, fi(p, JL, T, t), satisfies

(2)

Here, JL is the cosine of the angle between the radius and the particle momentum
p, P = ipi, Pe = ve/c, P"( = 1, and Ve is the velocity of electrons and positrons.
Also, iff is the emission coefficient for the production of a particle of type i via the
physical process labelled by q, and xi is the corresponding absorption coefficient.
The processes we include are listed in the following Table.

Basic Two-Body Radiative

Interaction Variant

M0ller and Bhaba

scattering Bremsstrahlung
ee ----+ ee ee +-+ eel
Compton scattering Double Compton scattering
,e ----+ ,e ,e +-+ Tel
Pair annihilation Three photon annihilation
e+e- ----+" e+e- +-+ I I I
Photon-photon
pair production
,,----+ e+e-

3. Numerical results
For injected pair luminosity L± higher than'" 10 34 ergs S-1, the emerging emission
consists mostly of photons (see Fig. 1, left panel). This simply reflects the fact that
in this case the pair annihilation time tann '" (n e O"Tc)-1 is less than the escape time
tese '" Ric. There is an upper limit to the rate of emerging pairs f.r:;ax ':::' 10 43 S-1
(see Fig. 1, right panel).
As L± increases from'" 10 34 to 10 42 ergs S-1, the mean energy of emergent
photons decreases from '" 400 ke V to 40 ke V, as the spectrum changes in shape
from that of a wide annihilation line to nearly a blackbody spectrum with a high
energy (> 100 keY) tail (see Fig.2).
 1182

10° 1043
y

10-'

10"
10-'
~~
::_r 10-3 -H
·z 10"
10-4

10-5 1040

1~ 1~ 1~ 1~ 1~ 1~ 1~ 1~ 1~ 1~ 1~ 1~ 1~ 1~ 1~ 1~ 1~ 1~

L± [ergs s-'] L± [ergs s-']

Fig. 1. LEFT: The fractional emerging luminosities in pairs (dashed line) and photons (solid
line) as functions of the injected pair luminosity, L±. RIGHT: Number rate of emerging pairs as
functions of L± (solid line). The case where gravity has been neglected is shown by the dashed
line.
1040 500
1 "
1039 1041---
"", 1038 40 400
10
), 1037
OJ 39
.>< 10
1036 >OJ 300
'"~ 1035 10" .><
~
co'
"C
1034 1\
IJ- 200
37
1033 10 e
.f 36
"C 32
10 10
100
10" 10"
1030 10"
0 "'
y
10"
20 100 500 1034 1035 1036 1037 1038 1039 1040 1041 10"
" [keV] L± [ergs 5']

Fig. 2. LEFT: The energy spectrum of emerging photons for different values of L±, as marked
on the curves. The dashed line is the spectrum of blackbody emission. RIGHT: The mean energy
of the emerging photons (thick solid line) and electrons (thin solid line) as a function of L±.
For comparison, we show as the dotted line the mean energy of blackbody photons for the same
energy density as that of the photons at the photosphere. Also shown as the dashed line is the
mean energy of the emerging photons in the case when only two particle processes are taken into
account.

Acknowledgments
The research was supported by the Israel Science Foundation of the Israel Academy
of Sciences and Humanities.

References
1. A.G. Aksenov, M. Milgrom and V.V. Usov, Mon. Not. RAS 343, L69 (2003).
2. A.G. Aksenov, M. Milgrom and V.V. Usov, Astrophys. J. 609, 363 (2004).
3. A.G. Aksenov, M. Milgrom and V.V. Usov, Astrophys. J. 623, 567 (2005).
4. V.V. Usov, Phys. Rev. Lett. 80, 230 (1998).
5. V.V. Usov, Astrophys. J. Lett. 550, L179 (2001).
 EVIDENCE FOR WHITE DWARFS WITH STRANGE-MATTER
CORES*

GRANT J. MATHEWS, INSAENG SUH and NGUYEN Q. LANt

Center for Astrophysics, University of Notre Dame, Department of Physics, Notre Dame, IN
46556 USA
gmathews@nd. edu

KAORI OTSUKI
University of Chicago, Chicago, IL 60637 USA

FRIDOLIN WEBER
Department of Physics, San Diego State University, San Diego, CA 92182 USA

We summarize masses and radii for a number of white dwarfs as deduced from a com-
bination of proper motion studies, Hipparcos parallax distances, effective temperatures,
and binary or spectroscopic masses. A puzzling feature of these data, however, is that
some stars appear to have radii which are significantly smaller than that expected for a
standard electron-degenerate white-dwarf equations of state. We construct a projection
of white-dwarf radii for fixed effective mass and conclude that there is at least marginal
evidence for bimodality in the radius distribution for white dwarfs. We argue that if such
compact white dwarfs exist it is unlikely that they contain an iron core. We propose an
alternative of strange-quark matter within the white-dwarf core. We also discuss the
impact of the so-called color-flavor locked (CFL) state in strange-matter core associated
with color superconductivity. We show that the data exhibit several features consistent
with the expected mass-radius relation of strange dwarfs. We identify eight nearby white
dwarfs which are possible candidates for strange matter cores and suggest observational
tests of this hypothesis.

We have explored l the possible existence of white dwarfs with strange-matter

cores. Such strange dwarfs could gradually form 2- 4 during the progenitor main-
sequence by the accretion of a strange-matter nugget which pre-exists either as a
relic of the early universe or as an ejected fragment from the merger/coalescence
of strange-matter neutron stars. Once captured by a star, strange-matter nuggets
would gravitationally settle to the center and begin to convert normal matter to
strange matter. This would eventually lead to the formation of an extended strange-
quark-matter core in the white dwarf remnant.
The additional degrees of freedom for strange-quark matter lower the degeneracy
pressure and Fermi energy and allow the matter to be more compact. A strange
white dwarf is expected l - 3 to consist of three distinct regions, a crust, a core-crust
boundary, and a core.
The crust of the star is composed of normal degenerate matter 5 taken here

* Work supported in part by the US Department of Energy under Nuclear Theory grant DE-FG02-
95ER40934. K.O. and N.Q.L. also supported in part by NSF grant PHY 02-16783 for the Joint
Institute for Nuclear Astrophysics (JINA). F. W. is supported in part by the National Science
Foundation under Grant PHY-0457329, and by the Research Corporation.
tPresent Address: Dept. of Physics, Hanoi University of Education, Hanoi Vietnam

1183
1184

to be 12C. Within the crust the density increases until the neutron drip density,
Pd,.ip = 4.3 X 1011 g cm -3, above which free neutrons are released which gravitate
to the core where they are absorbed and converted into strange matter. 2- 4 A sharp
boundary between the inner core and the outer crust develops 6 due to a Coulombic
repulsion. The finite mass of the strange quark prevents the achievement of neutral
strange matter. Degenerate electrons are unable to fully neutralize the positive
charge of the inner strange-matter core because they are not bound by the strong
interaction. Hence, a dipole layer of very high voltage develops between the crust
and the core which isolates the core. This gap prevents the further growth of the
core beyond the radius associated with the neutron drip density and defines the size
to which the strange-quark core can grow.
The core of the star is taken to be comprised of noninteracting :::::: massless u
and d quarks and gluons plus s quarks with m :::::: 150 MeV. The MIT bag model
equation of state is adequate for our purposes whereby, P = (p - 4B)/3, with P the
mass-energy density and B the bag constant constrained from hadronic properties
to be, B1j4 ~ 145 to 160 MeV. Within the quark core, the density reaches ~ 2 - 3
times nuclear matter density (~4 - 6 X 10 14 g cm- 3 ). Although the density jumps
discontinuously at the boundary, the star maintains pressure equilibrium and the
pressure varies continuously through the star.
Adopted masses and radii for 22 nearby white dwarfs 7 ,s are shown in Fig. 1. The
astrometric mass was taken 1 to be better than the spectroscopic or gravitational
mass. Radii are based upon the Stefan-Boltzmann radius inferred from the observed
luminosity and Hipparcos distances or the gravitational redshift.
The curves for strange dwarfs agree surprisingly well with several of the best
determined data points. Of particular importance are C2S8-44, EC 50 and CD
140. These stars have well determined masses and they lie very close to the strange
dwarf mass radius relation. The other stars which minimize X2 by identification as
a strange-dwarf are C156-64, EC21, C181-B5B, CD 279 and WD2007-S0S. The
total X2 if all stars are associated with a normal carbon white-dwarf EOS is 119
(reduced X;' = 5.2 for 23 degrees of freedom). If we allow an iron-core population
with members chosen to minimize X2 we obtain 105 (X;' = 4.4 with 24 degrees of
freedom). This is to be compared with a value of 78 (X;' = 3.2) when these eight
stars are identified with a strange-matter EOS. Hence, we have a ,0.X2 = 41 (~
60") preference for the presence of a strange-dwarf population over a normal EOS
versus a ,0.X2 = 14 (~ 40") improvement by identifying these stars with an iron-core
population.
We also note l that two of the stars C181-B5B and CD279 have temperatures
(T :::::: 13,500 K) which are tantalizingly close the the DAV (ZZ Ceti) pulsation
instability at T :::::: 12,000 K. We have estimated 1 that the buoyancy frequency will
be affected by the steep density gradient of the core. This will lead to mode filtering
and easier excitation of those pulsation modes which probe the inner strange-matter
core. Preliminary observations have begun to search for the relevant pulsations and
determine the cooling rate in these stars.
 1185

0.012
~
::l
~ 0.01
e<:
0.008

0.006

0.004

0.002

1.5
Mass

Fig. 1. Comparison of the theoretical mass-radius relationships for strange dwarfs (solid curves)
and normal white dwarfs (dot-dash curves) with the data (points) of Provencal et a1. 7 ,8

References
1. G. J. Mathews, 1. Suh, B. O'Gorman, N. Q. Lan. W. Zech, K. Otsuki, and F. Weber,
J. Phys. G32, 747 (2006).
2. N. K. Glendenning, Ch. Kettner, & F. Weber, Phys. Rev. Lett., 74, 3519 (1995).
3. N. K. Glendenning, Ch. Kettner, & F. Weber, ApJ, 450, 253 (1995).
4. N. K. Glendenning & F. Weber, ApJ, 400, 647 (1992).
5. T. Hamada & E. E. Salpeter, ApJ 134, 683 (1961).
6. C. Alcock, E. Farhi, & A. Olinto, ApJ, 310, 261(1986).
7. J. L. Provencal, H. L. Shipman, , E. Hog, & P. Thejll, ApJ 494, 759 (1998).
8. J. L. Provencal et al., ApJ 568, 324 (2002).
This page intentionally left blank
Thermal Behavior of
Compact Stars
This page intentionally left blank
 MAGNETARS: INTERNAL HEATING AND ENERGY BUDGET*

D. G. YAKOVLEV, A. D. KAMINKER, A.Y. POTEKHIN, P. S. SHTERNIN

Ioffe Physico- Technical Institute, Politekhnicheskaya 26
194021, Saint-Petersburg, Russia
[email protected]

G. CHABRIER
Ecole Normale Superieure de Lyon, CRAL (UMR 5574 CNRS) , 46 allee d'Italie
69364 Lyon, France
chabrier@ens-lyon·fr

N. SHIBAZAKI
Rikkyo University, Tokyo 171-8501, Japan
[email protected]

We develop models of magnetars as cooling neutron stars with an additional heating

in a spherical internal layer. We show that in order to explain high observable thermal
luminosities of magnetars and be consistent with the energy budget of neutron stars the
heat source should be located in the outer neutron star crust and should have the heat
intensity ~ 3 X 10 20 erg cm- 3 s-l

Magnetars form a special group of neutron stars1 which contain soft gamma
repeaters (SGRs) and anomalous X-ray pulsars (AXPs). They seem to be hot,
isolated, slowly spinning neutron stars of age t ;S 10 5 yr with extremely strong
magnetic fields, B ;(, 10 14 G. There is no strict theory of magnetar activity and
evolution. Many theoretical models 1,2 are divergent and assume that magnetars are
powered either by internal energy sources, or by magnetospheric processes, or by
combination of both. We consider a possibility that magnetars are cooling neutron
stars with internal heating (see Ref. 3 for details).
Cooling of magnetars has been simulated with our cooling code. The base of
the neutron-star heat blanketing envelope, where the main temperature gradient is
located in an ordinary cooling neutron star, was placed at a density of Ph = 1010
g cm -3 (a few hundred meters under the surface). The envelope was assumed to
consist of iron and possess a strong dipole magnetic field which affects the thermal
structure of the envelope. Neutrino emission from the blanketing envelope was ne-
glected; the effects of the magnetic field (others than producing internal heating)
in deeper layers were neglected as well. The cooling code calculates the thermal
surface luminosity (or, equivalently, the effective surface temperature TsCXJ properly
averaged over a stellar surface and redshifted for a distant observer) as a function of
the stellar age t. In Fig. 1 (from Ref. 3 ) we compare the results with estimated values

'This research has been partially supported by the Russian Foundation for Basic Research
(grants 05-02-16245, 05-02-22003), by the Federal Agency for Science and Innovations (grant NSh
9879.2006.2), by the Rikkyo University Invitee Research Associate Program, CNRS French-Russian
program (grant PICS 3202), and by the Dynasty Foundation.

1189
1190

9.2
M=1.4 M0
6.8
9
Q
----
h 8.8
.:=Jl

8.6 6.4

8.4
11 12 13 14 15 2 3 4 5 6
19 P (g cm- 3 ) 19 t (yrs)

Fig. 1. Cooling of a nonsuperfluid 1.4 A18 neutron star with a nucleon core and equation of state
proposed in Ref4 The magnetic field in the heat blanketing envelope is dipole (B = 5 x 10 14 G
at the poles); the heating duration is T = 5 X 10 4 years. Left: Temperature profiles within the star
of age t = 1000 years with four different positions I-IV of the heating layer (indicated by hatched
rectangles) and two levels of the heat intensity Ho = 3x 10 19 (thin lines) and 3 X 10 20 erg cm- 3 s-l
(thick lines). Right: Cooling curves for these models compared with the observations. The shaded
rectangle is the "magnetar box" to be explained by the theory.

of Tsoo for two SGRs (l=SGR 1900+14; 2=SGR 0526-66) and five AXPs (3=lE
1841-045;4=CXOU J010043.1-721134; 5=lRXS J170849-400910; 6=4U 0142+61;
7=lE 2259+586). The data are mostly taken from ReU and are highly uncertain. 3
If these values of T,oo really refer to the surface radiation emergent from stellar in-
teriors, magnetars are much hotter than ordinary cooling neutron stars of the same
age. 5 ,6
Calculations show that hot thermal states of magnetars can be explained, within
our model, only by assuming an additional heat source which we suggest to operate
in a spherical layer within the star. We have introduced this heating into the code
in a phenomenological way. The heating rate H [erg cm- 3 S-I] (possibly associated
with the magnetic field evolution) has been taken in the form

(1)

where Ho is the maximum heat intensity, 8(pl' P2) is a step-like function (8 ::::0 1
within some heating layer, PI < P < P2; 8 ::::0 0 outside this layer), and T is the
heating duration. A specific form of H is not important for our main conclusions:
(1) We can construct heating layers consistent with the data, but with strongly
restricted parameters. We obtain that hot magnetar states are solely supported
by an additional heat. Once the heating source is switched off, the star quickly
transforms into a much colder ordinary cooling neutron star. Accordingly, we must
have T rv 10 4 -10 5 years. Shorter T cannot explain older magnetars; longer T would
require too much heating energy.
 1191

(2) The heating layer should be located only in the outer magnetar crust (p ;S
4 X 1011 g cm- 3 ). Otherwise, even for a very high Ho and a bulky heating layer
(huge total integrated heat generation rate) the heat does not flow to the surface
but, instead, is radiated by neutrinos from the stellar interior, leaving the surface
much colder than necessary (Fig. 1). Even if the heat sources are placed in the outer
crust, maximum rv 1% of the heating energy is radiated by photons from the surface.
(3) The heat intensity in the outer crust should be Ho rv 3 X 10 20 erg cm- 3 s-l.
It makes magnetar interiors highly nonisothermal (Fig. 1), with T rv 2 X 10 9 K in the
heating layer (contrary to ordinary cooling neutron stars). Lower H 0 is insufficient to
heat the surface to the "magnetar box." Higher Ho will not help to heat the surface
(the heating layer will become too hot and the extra heat will be radiated away
by neutrinos); in addition, one will need too much energy to support magnetar's
activity during its life ( ~ 1050 ergs, which a neutron star cannot afford).
(4) The heated outer magnetar crust is thermally decoupled from deeper interior
and highly insensitive to the physics of the inner crust and the neutron star core
(to the equation of state, composition of matter, neutrino emission mechanisms,
superfluidity). This is in sharp contrast to ordinary cooling neutron stars whose
thermal states strongly depend on the physics of internal layers. 5 ,6 On the other
hand, thermal states of magnetars are very sensitive to the physics of outer crust 7
(first of all to neutrino emission there).
(5) In our scenario, magnetars differ from ordinary neutron stars by the presence
of some (probably magnetic) internal energy (rv 1049 - 1050 erg). It can be stored
in the whole magnetar body (e.g., B rv (1 - 3) x 10 16 G in the core) and released
in the outer crust during T rv 10 4 - 10 5 years. The storage and release mechanisms
remain to be explored. The release may be transient, leading to observed transient
activity of magnetars.
Our results should be further elaborated, particularly, by a careful treatment of
neutrino emission and heat transport in a magnetic field in the whole outer crust. 7
However, our conclusions seem rather insensitive to details of calculations. 3 ,7 New
observations and modeling will hopefully reveal the magnetar nature in near future.
DGY is grateful to the organizers of MGll for financial support.

References
1. P. M. Woods and C. Thompson, in: Compact Stellar X-ray Sources, eds. W. H. G.
Lewin and M. van der Klis (Cambridge Univ., 2006) p. 547.
2. A. M. Beloborodov and C. Thompson, astro-ph/0602417.
3. A. D. Kaminker, D. G. Yakovlev, A. Y. Potekhin, N. Shibazaki, P. S. Shternin and
O. Y. Gnedin, Mon. Not. Roy. Astron. Soc. 371, 477 (2006); Astrophys. Space Sci.,
in press (2007).
4. A. Akmal, V. R. Pandharipande and D. G. Ravenhall, Phys. Rev. C 58, 1804 (1998).
5. D. G. Yakovlev and C. J. Pethick, Ann. Rev. Astron. Astrophys. 42, 169 (2004).
6. D. Page, U. Geppert and F. Weber, Nucl. Phys. A 777, 497 (2006).
7. A. Y. Potekhin, G. Chabrier and D. G. Yakovlev, astro-ph/0611014.
TRAPPING OF NEUTRINOS IN EXTREMELY COMPACT STARS*

ZDENEK STUCHLIK, MARTIN URBANEC, GABRIEL TOROK, STANISLAV HLEDIK

AND JAN HLADIK
Institute of Physics, Faculty of Philosophy and Science, Silesian University in Opava,
Bezrucovo nam. 13, Opava 746 01, Czech Republic
zdenek. [email protected]

Trapping of neutrinos in extremely compact stars containing trapped null geodesics is

studied. We calculated the ratio of produced to trapped neutrinos in the simplest model
of uniform density stars. This gives the upper limit on trapping coefficients in real objects.

1. Introduction
Trapped null geodesics are concentrated around the stable circular geodesic. 2 ,3 We
suppose that neutrinos have zero mass and can move along null geodesics in the
situations when the neutrino mean frec path exceeds the star radius R. Considering
the internal Schwarzschild spacetimes, we use the effective potential related to the
impact parameter to calculate the ratios of trapped to produced neutrinos.

2. Internal Schwarzschild spacetime, effective potential and impact

parameter
The line element of internal Schwarzschild spacetimes of uniform density1 reads

(1)

The temporal and radial components of the metric tensor are given by the formulae

(2)
and in geometric units c = G = 1 there is

2) 1/2 R2) 1/2 1 2M

Y(T) = (1 - :.2 Y1 == Y(R) = (1 - """"d2
R3 '
(3)

where R is the radius and !vI is the mass of the object. The parameter a represents
the curvature of the internal Schwarz schild spacetime.
Due to the existence of two Killing vector fields, the temporal a/at onc, and the
azimuthal a/ a¢ one, two conserved components of the 4-momentum must exist:

E = -Pt (energy), L = P¢ (axial angular momentum). (4)

Because the motion plane is central, one can set e = 7r /2 = const, choosing the equa-
torial plane. The motion along null-geodesics is independent of energy (frequency)

*This research has been supported by Czech grants MSM 4781305903 and LC 06014 (M.U.).

1192
 1193

35

~ 20
>
15

10

Fig.1. Behavior of the effective potential for R = 2.5GlvI/c 2 . Neutrinos produced in the region
Tb(e) < T < R are trapped if e > le(e) and escape when e < le(e)'

and can conveniently be described in terms of the impact parameter £ = L/ E. The

radial motion is restricted by an effective potential related to £ by
. 4a 2 [1 - y2(T)]
V mt - --o---=----,-----'-~ for T' ::; R
£2 ::; Veff = elf - [3~~ - Y(T)j2 (5)
{ V ext _ _ __
for T > R,
elf - T - 2M

V;lr (Ve'fft) is the effective potential of the null-geodetical motion in the internal
(external) Schwarzschild spacetime 4 (see Fig. 1).

3. Escaped to produced neutrinos ratio

We assume that neutrinos are locally produced by isotropically emitting sources.
Then escaped-to-produced-neutrinos ratio depends on a geometrical argument only
and it is determined through directional escaping angle We given by

(6)

Let N p , Ne and Nb denote the number of produced, escaped and trapped neu-
trinos per unit. time of an external static observer at infinity. In order to determine
the global correction factors
Nb(R)
B(R) == Np(R) = 1 - [(R), (7)

it is necessary to introduce the local correction factor for escaping neutrinos at a

given radius T' E (Tb(e), R). The escaping solid angle is given by

\ji c 2,,-

ne(W e ) =.1 .I sin wdwd¢ = 27f(1 - cos We) (8)

o 0
1194

0.8

0.7
0.8
,
Ii
0.6

0.6 0.5
.£ :B,
0.4
"
"'- 0.4

0.2

0.2 0.4 0,6 0,8 2.3 2.4 2.5 2.6 2.7 2,8 2.9
,IR RIM

Fig, 2. Behavior of the coefficient of local trapping (3(r, R) (left) for several values of R/ M and of
the total coefficient of trapping B( R) (right). QL and XL determine trapping of internal (external)5
neutrinos; BL = QL + XL

and the escaping (trapping) correction factors f(r, R) ((3(r, R)) are given by
dNe(r) 20('lj;e(r, R))
f(r, R) = 1- (3(r, R) = dNp(r) = 47r = 1 - cos 'lj;e(r, R). (9)

The coefficient (3( r, R) determines local efficiency of the neutrino trapping (see Fig. 2
for behavior of local (left) and global (right) trapping coefficients). The global es-
caping (trapping) coefficients are given by integrating the local production rate
and escaping (trapping) coefficients through the star, and its trapping zone, respec-
tively.5 We assume local production rate uniformly distributed through the star
from the point of view of local observers but it is not uniform as viewed by distant
static observers. 5

4. Conclusions
We have shown in the simplest case of uniform density stars that trapping of neutri-
nos plays a significant role in cooling mechanism. The local effect given by (3(r, R)
is 10% for RjM = 2.95 while the global effect described by B(R) is 10% for
RjM = 2.87. We expect this trapping efficiency could be reached in some real stars
with specific but realistic equations of state.

References
l. K. Schwarzschild. Sitzber. Deut. Akad. Wiss. Berlin, Kl. Math.-Phys. Tech., pages
424-434, 1916.
2. Z. Stuchlik, S. Hledik, J. Soltes, and E. 0stgaard. Phys. Rev. D, 64(4):044004
(17 pages), 200l.
3. M. A. Abramowicz, J. Miller, and Z. Stuchlik. Phys. Rev. D, 47(4):1440-1447, 1993.
4. C. W. Misner, K. S. Thorne, and J. A. Wheeler. Gravitation. Freeman, San Francisco,
1973.
5. Z. Stuchlik, Gabriel Torok and Stanislav Hledik. In S. Hledik and Z. Stuchlik, editors,
Proceedings of RAGtime 4/5: Workshops on black holes and neutron stars, Opava,
14-16/13-15 October 2002/03, Opava, 2004. Silesian University in Opava.
 A SELF-CONSISTENT MODEL OF THE ISOLATED NEUTRON
STAR RX J0720.4-3125 *

JUAN A. MIRALLES, JOSE A. PONS, J. FERNANDO PEREZ-AZORIN

Departament de Fisica Aplicada, Universitat d'Alacant, Ap. Correus 99, 03080 Alacant, Spain

GIOVANNI MINIUTTI
Institute of Astronomy, University of Cambridge, Madingley Road, Cambridge, CB30HA, UK

Observation and analysis of thermal radiation from neutron stars can provide very valu-
able information on the properties of matter at high densities and can shed light on
the physical phenomena occurring at the surface layers of neutron stars. In this talk
we present a model for the isolated neutron star RX J0720.4-3125 that accounts simul-
taneously for the observed X-ray emission, the optical excess, the pulsed fraction, the
observed spectral feature at 0.3 keV and the long-term variability. The model is based
on the presence of a high magnetic field that induces a condensation of the surface layers
and modifies the surface temperature distribution and the local emissivity with respect
to a black body. We have explored different magnetic field configurations, including
both poloidal and toroidal components, to fit observations. Our results indicate that this
model can explain relatively well all the observed properties of RX J0720.4-3125 and,
although we cannot rule out other models, this one offers a self-consistent explanation
of the observed data. We are studying other neutron stars that we expect to be well
described by this model and preliminary results are promising.

Since the discovery, a decade ago, of the first radio-quiet X-ray isolated neutron
star (INS) by Walter et al. l , a new class of objects sharing common properties has
emerged, but a consistent theoretical model to explain all observations is still under
debate. The seven INSs observed up to date by the X-ray satellites have thermal
X-ray emission well described by a Planckian spectrum with temperatures between
50 and 100 eV. They show low interstellar absorption (hydrogen column density
nH ~ 10 20 cm- 2 ), what means that they are nearby objects, a few hundred parsecs
from us. The lack of association with supernova remnants implies that the ages
of these objects are greater than ~ 105 years. Another property shared by five of
these objects and very recently observcd in a sixth one (RX J1856.5-3754)2 is the
pulsation in the X-ray light curve, with periods ranging from 3.5 to 11 seconds and
pulsed fractions of thc order of ~ 10%. Observations in the optical band show that
some of these objects present an optical flux larger than the extrapolation to the
optical band of the best blackbody (BB) fit to the X-ray emission. The inferred
magnetic fields for these objects, either coming from the spin down measurement
or from interpretation of the observed features in their spectra as proton cyclotron
lines, are between 1013 to 10 14 G.
Concerning RX J0720.4-3125, it was discovered with ROSAT 3 and its X-ray
spectrum was soon found to be well described by a Planckian spectrum with tem-

*This work has been supported by the Spanish Ministerio de Ciencia y Tecnologfa grant AYA
2004-08067-C03-02.

1195
1196

perature cv 82 eV. Similarly to the rest of INS, it is a nearby object (~300 pc)4 and
shows low interstellar absorption (nH = 1-1.5 x 10 20 cm- 2). More interestingly, it
is a confirmed X-ray pulsar3 with a period of 8.391 s and it is one of the two INSs
with a reliable measure of the period derivative 5 P = 7 X 10- 14 , which implies a
magnetic field of about B = 2.4 X 10 13 G. The optical flux measured for this object
is about a factor of 6 larger than the predicted from the BB fit to the X-ray band.
This apparent optical excess flux of INSs, first observed in RX J1865-3754 can be
explained with the existence of large temperature anisotropies over the surface. 6
In the case of RX J0720.4-3125, the evidence of anisotropic temperature distribu-
tions is strongly supported by the observed X-ray pulsations with relatively large
amplitude (cv 11 %).
The theoretical model to explain observations is based on the existence of a
strong magnetic field which modifies the thermal structure of the crust and enve-
lope 7- 9 and provokes the condensation of the surface layers changing the emissivity
properties of the neutron star. 1O- 13 In previous papers 8 ,12 we have presented the
results of detailed calculations of the temperature distribution in the crust and
condensed envelopes of neutron stars in the presence of strong magnetic fields, by
obtaining axisymmetric, stationary solutions of the heat diffusion equation with
anisotropic thermal conductivities. Having explored a variety of magnetic field
strengths and configurations, we concluded that variations in the surface tempera-
ture of factors 2-10 are easily obtained with B ~ 10 13 _10 14 G whereas the average
luminosity (and therefore the inferred effective temperature) depends only weakly
on the strength of the magnetic field, but is drastically affected by the field geome-
try, in particular by the existence of a toroidal component. Moreover, if the magnetic
field is strong enough to induce the condensation of the surface,14 the condensed
surface models also predict the existence of a spectral edge that for B ~ 10 13 _10 14
falls in the range 0.2-0.5 keY and that can be consistent with an absorption line
such as that reported in RX J0720.4-3125. 15 We refer to the interested reader to
the previous work 8 for details about the calculations.
Using this model for a given magnetic field configuration (force free quadrupolar,
crust confined with a small dipolar component), in Perez-Azorfn et al.,16 we were
able to fit the parameters of our model (column density, temperature and magnetic
field strength at the pole, neutron star radius over the distance to the star, angle
between the angular momentum and the line of sight, 0, and angle between the
angular momentum and the magnetic field axis, B), to the XMM-Newton observa-
tions of the INS RX J0720.4-3125, showing that realistic models of neutron stars
with strong magnetic fields are consistent with the observed X-ray and optical spec-
tra, the observed deviation from a pure thermal spectrum in the X-ray band, and
the long term variability (Table 1). Although we do not exclude the presence of a
resonant proton cyclotron absorption or bound-bound transitions in H, we do not
need to invoke it to explain the observed spectral edge.
In addition to the above properties, long-term spectral variation, observed in
this object, can also be explained using this model in terms of neutron star preces-
 1197

Table 1. Joint fits to EPIC-pn and EPIC-MOSI FF mode observations for a crust con-
fined, quadrupole dominated magnetic field with 0 = 11 0 , forcing the hydrogen column
density, the normalization constant and the magnetic field intensity to be the same for
all observations.

Rev. nH B kTpole R/D300 Bp x 2 / d .o .f.

(10 2O cm- 2 ) (eV) (km/300 pc) (10 12 G)
175 1.23::gg1 73.9::~401 104.6::g:! 12.7::t:~ 18.2::i~
533 58.6::~:~ 107.5::g~
534 57.6::~~ 107.5::g~
815 26.9::6~ 118.2::g:3
1086 31.8::~:i 116.3::g:~ 1192/819

sion. If our interpretation is correct, we predict that the pulsed fraction, effective
temperature, and absorption line equivalent width as obtained with phenomenolog-
ical models, will decrease in subsequent observations of the source, until reaching
similar values to ones observed in the satellite revolution number 175. This begins to
be apparent in recent observation (revolution 1086). Similar conclusions have been
reached independently by other groups.17 We notice, however, that our fits still
show some variability of the pole temperature correlated to that of the orientation
angle.
We are also working to study other INSs that we expect to be well described by
these models. Our final goal is to find a general solution pointing towards a unified
picture in which INSs are just old, cold magnetars whose magnetic fields are a few
times smaller than usual, maybe because they have decayed during their lifetime.

References
1. Walter, F.M, Wolk, S.J., Neuhauser, R., Nature, 379, 233 (1996)
2. Tiengo, A., Mereghetti, S., ArXiv Astrophysics, astro-ph/0612501 (2006)
3. Haberl, F. Motch, C. Buckley, D. A. H., Zickgraf, F. J., & Pietsch, W., A&A, 326,
662 (1997)
4. Kaplan, D.L. et al., ApJ, 590, 1008 (2003)
5. Kaplan, D.L. & van Kerkwijk, M.H., ApJ, 628, L45 (2005)
6. Pons, J.A, Walter, F.M., Lattimer, J.M. et al., ApJ, 564, 981 (2002)
7. Geppert, U., Kuker, M., & Page, D., A&A, 426, 267 (2004)
8. Perez-Azorin, J.F., Miralles J.A., & Pons J.A., A&A, 451, 1009 (2006)
9. Geppert, U., Kuker, M., & Page, D., A&A, 457, 937 (2006)
10. Brinkmann, W., A&A, 82, 352 (1980)
11. Turolla, R., Zane, S., & Drake, J.J., ApJ, 603, 265 (2004)
12. Perez-Azorin, J.F., Miralles J.A., & Pons J.A., A&A, 433, 275 (2005)
13. van Adelsberg, M., Lai, D., & Potekhin, A., ApJ, 628, 902 (2005)
14. Lai, D., Rev. of Mod. Phys., 73, 629 (2001)
15. Haberl, F., Zavlin, V.E., Trumper, J., Burwitz, V., A&A, 419, 1077 (2004)
16. Perez-Azorin, J.F., Pons, J.A., Miralles, J.A., & Miniutti, G., A&A, 459,175 (2006)
17. Haberl, F., Turolla, R., de Vries, C. P., Zane, S., Vink, J., Mndez, M., Verbunt, A&A,
451, L17 (2006)
 kHz QPO PAIRS EXPOSE THE NEUTRON STAR
OF CIRCINUS X-I

S. BOUTLOUKOS
Center for Theoretical Astrophysics, Department of Physics, University of Illinois at
Urbana-Champaign, 1110 W Green, 61801, Urbana, IL, USA
[email protected]

M. VAN DER KLIS, D. ALTAMIRANO, M. KLEIN-WOLT and R. WIJNANDS

Astronomical Institute, University of Amsterdam, Kruislaan 403,
1098 SJ, Amsterdam, the Netherlands

We discovered kHz QPOs in 80 archived RXTE observations from the peculiar low-mass
X-ray binary (LMXB) Circinus X-I. In 11 cases these appear in pairs in the frequency
range of ~230 Hz to ~500 Hz for the upper kHz QPO and ~56 Hz to ~225 Hz for the
lower kHz QPO. Their correlation with each other, which is similar to that of frequencies
of kHz QPO pairs in other LMXBs containing a neutron star, and their variation by
a factor two confirm that the central object is a neutron star. These are the lowest
frequencies of kHz QPO pairs discovered so far and extend the above correlation over a
frequency range of factor four. In this new frequency range the frequency difference of
the two kHz QPOs increases monotonically by more than ~170 Hz with increasing kHz
QPO frequency, challenging theoretical models.

1. Introduction
Quasi periodic-oscillations (QPOs) are thought to reflect the motion of matter in
the curved spacetime around neutron stars and black holes. The high frequency
QPOs have proven to be a good indicator of the nature of the compact object.
If the LMXB includes a neutron star, pairs of kilohertz (kHz) QPOs often appear
simultaneously with frequencies of several hundred Hz. Their frequencies vary as the
system evolves but their difference appears to reflect the spin frequency of the star
or half of it, in case where this is known from X-ray pulsations. Their coherences
and amplitudes range up to about 200 and 20% respectively. On the other hand,
black holes only rarely show high frequency QPOs a and then mostly one at a time;
the frequency ratio of the higher to lower frequency QPO appears to be about 3:2
and they have small amplitudes and coherences. See Ref. 8 for a review.
The frequencies of the two ('lower' and 'upper') kHz QPOs follow a tight rela-
tion for all neutron stars from which they have been observed. 5 Most models for the
generation of the kHz QPO pairs relate at least one of them to an orbital motion in
the accretion disc. In the sonic-point spin-resonance modeI,3 the upper kHz QPO
is generated by a bright spot on the stellar surface generated by gas from a clump
orbiting at the sonic radius, whereas the lower kHz QPO is produced by illumina-
tion of gas at the spin-resonance radius. In the relativistic precession model,6 the
upper and lower kHz QPO are identified respectively as the nodal and periastron

aThe term kHz QPOs does not apply to them since they never reach 1 kHz.

1198
 1199

10000 5oor---~----~--~~--~~P~BK=99~-'
CirX-1

1000
400

100

~ 10

100
0.1

PBK99 >- 0-----<

CirX-1 >-----A------<
0.01 o~--~----~--~----~----~~
10 100 1000 200 400 600 800 1000 1200
v~(HZ) vOu(Hz)

Fig.1. Left: the frequencies of the upper kHz QPO (top) and of the low-frequency QPO (bottom)
against that of the lower kHz QPO for our results from Circinus X-I as also for other sources from
Ref. 5. Right: The frequency separation of the kHz QPO pairs for the same sources as before, as
well as the best fit for a straight line (with a slope of O.6±O.1) and the relativistic precession model
to the Circinus X-I data. The slight misplacement of the latter in Ref. 1 is corrected here.

precession of gas clumps at the inner edge of the disc. Relativistic resonance mod-
els 2 suggest resonances between orbital frequencies in the disc, which would produce
kHz QPO frequencies at a fixed ratio, similar to what is occasionally seen in black
hole systems.
Circinus X-I is an LMXB whose nature was long disputed. Although Type-
I X-ray bursts were observed from the field of the object,7 its X-ray, radio, and
spectroscopic properties are more typical of a black hole. Here we summarize the
discovery of kHz QPO pairs reported in Ref. 1. These show it is a neutron star.

2. Observations
We have analyzed ~2 Msec of archival data taken during 1996-2005 using the Rossi
X-ray Timing Explorer in the 3-60 keY energy range. We divided the observations
into segments of 16, 48, 128, or 256 sec, Fourier-transformed them, and then aver-
aged each one to get power spectra. We made a precise estimate of the deadtime
values of the detector by fitting the Zhang function lO to observations where Circi-
nus X-I does not seem to contribute, and used those to subtract the Poisson noise
power from all observations. After renormalizing the power spectra to rms squared
we fitted a multi-Lorentzian function for the characteristic frequencies. By plotting
the frequencies of the fitted QPOs from each observation against each other as in
Ref. 9, we found two groups of correlated frequencies extending over several hundred
Hz. Based on the high frequencies and the over an order of magnitude frequency
variability we identified those as lower and upper kHz QPOs. We found in total 80
observations with kHz QPOs. In 11 cases both appeared simultaneously. All their
characteristics can be found in Ref. 1; in general they have relatively low coherences
and amplitudes.
In the left panel of Fig. 1 we plot the two frequencies. They follow the same
relation that was found for other neutron stars in Ref. 5, but extend this relation to
frequencies lower by a factor 4. It is of particular interest to see how the frequency
1200

separation behaves in this new frequency range for kHz QPOs. We see in the right
panel of Fig. 1 that it increases over a factor 2 with increasing frequency of the lower
kHz QPO. This is different from what is seen in other sources that have shown
kHz QPO pairs at higher frequencies. Mathematically speaking, the relativistic
precession model and to a lesser extent the relativistic resonance model, can explain
the observed behavior for a neutron star mass of 2.2±0.3.i\I(o) or a 1:3 resonance
respectivelyb. The former is produced for slightly inclined, infinitesimally eccentric
geodesic orbits around non-rotating compact objects, a framework that has been
challenged. 4 Oscillations in the disk require an energy comparable to the binding
energy of the star to produce relative amplitudes similar to those observed. The
sonic-point spin-resonance model can explain the generation of kHz QPOs, but is
challenged by the strongly increasing frequency separation with kHz QPO frequency
and the low frequencies observed in Cir X-I.

Acknowledgments
This work was accomplished under the Marie-Curie Training Program HPMT-CT-
2001-00245. S.B. acknowledges support by grants NSF AST 00-98399 and NASA
NAG 5-12030 and the funds of the Fortner Endowed Chair at the University of
Illinois.

References
1. Boutloukos, S., van der Klis, M., Altamirano, D., Klein-Wolt, M., Wijnands, R.,
Jonker, P., & Fender, R, Astroph. 1. 653, 1435 (2006); astro-ph/0608089
2. Kluzniak W. & Abramowicz M. (2003); astro-ph/0304345
3. Lamb, F. K., & Miller, M. C., ApJL. submitted (2003); astro-ph/0308179
4. Markovic, D .. & Lamb, F.K., Mon. Not. Ro. Astr. S., submitted (2000); astro-
ph/00091169
5. Psaltis, D., Belloni, T., & van der Klis, M., Astroph. J. 520, 262 (1999); astro-
ph/9902013
6. Stella, L. & Vietri, M., Phys.Rev.L, 82, 17 (1999); astro-ph/980:3278
7. Tennant, A. F., Mon. Not. Ro. Astr. S. 226, 971 (1987)
8. van der Klis, M., in Compact Stellar X-ray Sources, Cambridge University Press, 39
(2006); astro-ph/0410551
9. van Straaten, S.,van der Klis, M., Wijnands, R, Astroph. 1. 619, 455 (2005); astra-
ph/0410505
10. Zhang, W., Jahoda, K., Swank, J. H., Morgan, E. H., Giles, A. B., Astroph. J. 449,
930 (1995)
11. Zhang, W., Morgan, E. H., Jahoda, K., Swank, J. H., Strohmayer, T. E., Jernigan,
G .. Klein, R 1., Astroph. J. 469, 29 (1996)

bNote that the line for the relativistic precession model is slightly different than in Ref.l because
of an error there in plotting the corresponding function.
 NEUTRON STAR ATMOSPHERES AND X-RAY SPECTRA

WOLFGANG KUNDT
Argelander Institute of Astronomy,
Auf dem Hugel 71,
D-53121 Bonn, Germany
[email protected]

Ever since the onset of X-ray spectroscopy of neutron-star sources, an attempt has
been made to prohe their chemical surface composition, so far without success: The
thermal components of the spectra have turned out to be perfect blackbodies. This
finding is likely due to the presence of an optically thick pair corona, created hy a steady
bombardment with relativistic electrons and positrons.

Keywords: pulsar atmospheres, blackbody radiation, pulsar magnetospheres

1. Neutron Star Magnetospheres

Pulsars are thought to be spinning, strongly magnetized neutron stars, with surface
magnetic fieldstrengths between ~ 10 11 G and;; 10 14 G, [e.g. Kundt 1998]: Weaker
fieldstrengths would exclude the Erber mechanism (of pair formation above the polar
caps), and stronger fieldstrengths are difficult to anchor (even in the deep interior of
a neutron star); the 'magnetars' would form exceptions, if they existed [Kundt 2006j.
These 'surface magnetic fieldstrengths' must be distinguished from their often much
weaker 'transverse dipole components', inferred from their spindown torque, which
fieldstrengths can be even lower than lOsG. This large seeming discrepancy between
measured components and expected total fields can be understood, after Flowers &
Ruderman [1977], by the necessity of anchoring the star's magnetic moment in the
interior of a fluid star where it would decay dynamically- by a relative rotation
through some 180 0 of two halfs of it - unless it was stabilized by a toroidal bandage.
Such a toroidal magnetic bandage can be provided during the birth of a neu-
tron star, when its progenitor's degenerate core spins up during collapse inside the
evolved star that goes supernova. It thereby winds the core's protruding flux rela-
tive to the (slower) overlying shell, towards saturation, and thereby squeezes it into
deeper regions, of higher conductivity, as hypothesized in fig. l.
Independently of any detail, strong magnetic surface fields induce strong trans-
verse electric fields if = iJ x 13 above a pulsar's surface, where j3 denotes the local
corotation speed in units of c, so that any unscreened, free magnetospheric charge
is almost instantaneously boosted into the relativistic regime, upward or downward
along 13, depending on the sign of its charge. Pulsar winds are thought to con-
sist almost exclusively of electrons and pm;itrons, for which the relativistic runway
measures in cm:
(1)
Even if most of them are boosted outward, a significant number of them will be
repeatedly seized by their downward field strength, and forced to bombard the

1201
1202

Fig. 1. Example of a crosscut through a neutron star's magnetosphere, obtained by Chang [1994]
by adding six times a 90° tilted (normalized) oct up ole to a 50° tilted dipole. It is believed that
realistic magnetospheres result when a (stabilizing) toroidal magnetic field is wrapped around an
inclined (approximate) dipole, both frozen into a conducting sphere, and that they share with pure
dipoles the property of having two antipodal polar caps.

surface. When these extremely relativistic downward currents hit the surface, they
cannot help creating relativistic pair plasma.
Quantitative estimates of this quasi-steady bombardment of a pulsar's surface
- consistent with estimates of its wind strength via observed standoff bowshocks -
have assured Schaaf and me that surface bombardment should easily reach satura-
tion in producing a steady-state pair corona, of scale height some 10 2 m, temperature
T ~ 10 6 .5 K [Kundt & Schaaf 1993]:
T = (e~,flBllmeC2 /27rerYSB) 1/4 = 106.5K[c1.6(~,h(flBh4]1/4. (2)
The steady state realizes a balance between pair production by bombardment and
losses via annihilation, with column density Ne ~ O";;~n' where O"ann is the anni-
hilation cross section. The pair corona is essential for the functioning of a pulsar
because it reduces the work function of the surface to zero, for charges of either sign:
the corona is (electrically) unbound, as opposed to the atoms in the (magnetized)
neutron-star surface.
 1203

A bit harder to get is an estimate of the optical depth of the pair corona. In
[Kundt 2002] I estimated the pair annihilation cross section in a strong magnetic
field to be distinctly smaller than the Thomson cross section, (J" ann :::::; 1TT'6 /4 ;::;
lO-l(J"T, with the implication that active neutron-star surfaces are Thomson opaque.

We should therefore not be surprised to measure pure blackbody X-ray spectra,

within 1%, from the (seven) old nearby pulsars called 'the seven musketeers' [Bur-
witz et al 2001].

Acknowledgments
It is a great pleasure to thank Gunter Lay and Ole Marggraf for electronic help.

References
1. V. Burwitz, V. E. Zavlin, R. Neuhauser, P. Predehl, J. Trumper, A. C. Brinkman,
Astron. fj Astrophys. 379, L35-L38 (2001)
2. H.- K Chang, dissertation, Bonn, May (1994)
3. E. Flowers, and M. A. Ruderman, Astrophys. 1. 215, 302 (1977)
4. W. Kundt, Fundamentals of Cosmic Physics 20, 1-119 (1998)
5. W. Kundt, in: 270. WE-Hemeus Seminar on Neutron Stars, Pulsars and Supernova
Remnants, MPE-Report 278, eds. W. Becker, H. Lesch and J. Trumper, 206-208
(Garching, 2002)
6. W. Kundt, in: 363. WE-Hemeus Seminar on Neutron Stars and Pulsars 40 years after
the discovery, MPE-Report 291, eds. W. Becker, and H. H. Huang, 193-196 (Garching,
2006)
7. W. Kundt, and R. Schaaf, Astrophys. and Space Science 200, 251-270 (1993)
This page intentionally left blank
Alternative Theories (A)
This page intentionally left blank
 ANISOTROPICALLY INFLATING UNIVERSES*

SIGBJ0RN HERVIK
Department of Mathematics 8 Statistics, Dalhousie University,
Halifax, Nova Scotia, Canada B3H 3J5
[email protected]

We show that in theories of gravity that add quadratic curvature invariants to the
Einstein-Hilbert action there exist expanding vacuum cosmologies with positive cosmo-
logical constant which do not approach the de Sitter universe. Exact solutions are found
which inflate anisotropically. This behaviour is driven by the Ricci curvature invariant
and has no counterpart in the general relativistic limit. These examples show that the
cosmic no-hair theorem does not hold in these higher-order extensions of general rela-
tivity and raises new questions about the ubiquity of inflation in the very early universe
and the thermodynamics of gravitational fields.

1. Introduction

The inflationary universe is the central cosmological paradigm which astronomi-

cal observations aim to test, and by which we seek to understand how the uni-
verse might have evolved from a general initial conditions into its present state
of large-scale isotropy and homogeneity together with an almost flat spectrum of
near-Gaussian fluctuations. The essential feature of this inflationary picture is a pe-
riod of accelerated expansion during the early stages of the universe. 2 The simplest
physically-motivated inflationary scenario drives the acceleration by a scalar field
with a constant potential, and the latter can also be described by adding a positive
cosmological constant to the Einstein equations. In order to understand the gen-
erality of this scenario it is important to determine whether universal acceleration
and asymptotic approach to the de Sitter metric always occurs. A series of cosmic
no-hair theorems of varying strengths and degrees of applicability have been proved
to demonstrate some necessary and sufficient conditions for its occurrence. 3 ,4 Here
we will confine our discussion to the situation that occurs when there is a positive
cosmological constant, A > O. In this respect, the cosmic no-hair theorem by Wald
is of relevance: 4

The cosmic no-hair theorem (Wald 1983): Consider a spatially homogeneous

universe model of Bianchi type 1- VIII. Then, if A > 0, Ho > 0, and matter obeys
the strong and dominant energy conditions, the universe will eventually approach
an isotropic de Sitter state.
Specifically, this means that the Hubble parameter, H --+ j!i; the shear, uIi --+

0; and the matter density, !l2 --+ O. Similar conclusions result when we consider
inflation in those generalisations of general relativity in which the Lagrangian is a
function only of the scalar curvature, R, of spacetime.

* Based on work done in collaboration with J.D. Barrow. 1

1207
1208

2. Higher-curvature theories of Gravity

Consider a general quadratic Lagrangian:
£ = R + aR2 + (3R/LV R/Lv - 2A.
Variation leads to the following generalised vacuum Einstein equation:

G/LL' + <P/LV + Ag/Lv = 0,

where
1
G/LV == R/Lv - 2Rg/Lv , (1)

<P/LV == 2aR ( R/l v - ~ R9/Lv) + (2a + (3) (g/LvD - \l /L \l v) R

+(30 ( R 1w - ~R9/Lv) + 2(3 ( R/LCYvp - ~9/LvRCYP) RCYP. (2)

In the limit (a, (3) --> (0,0) this reduces to the regular Einstein equation.

A natural question would now be: Does the no-hair theorem apply to these
quadratic theories of gravity; i.e., does the universe isotropise in the presence of
a A > 0 for these kinds of theories? The (local) answer to this question depends
on the following: (i) The existence of exact inflating solutions. (ii) The stability of
these solutions (if they exist). It is therefore of interest to seek exact inflationary
solutions to these theories.
Indeed, in addition to the de Sitter solution, several other types of inflation
solutions can be found (depending on the parameters a, beta, A):
Bianchi type I anisotropically inflating solutions:

d 2s 1 = -dt 2 + e 2bt [e- 4CY +t dx 2 + e 2(CY++v'3 CY -)td y 2 + e 2(CY+-v'3 CY -)tdz 2] ,

b2=1+SA(a+(3) (a 2 ( 2 )=_1+2A(4a+(3).
9(3 , ++ - lS(3
Bianchi type II anisotropically inflating solutions:
2 2
d 8II = -dt + e
2bt [dx +

a 2 = 11 + SA(l1a + 3(3)
~(zdy - ydz) r + e bt (dy2 + dz 2 ),
b2 = SA(a + 3(3) + 1.
30(3' 30(3
Bianchi type VI h anisotropically inflating solutions:
d 2 8Vl h = -dt 2 + dx 2 + e 2(rt+ax) [e- 2(st+ahX)d y 2 + e+ 2(st+ahX)dz 2] ,

2 S(38 2 + (3 + h2 )(1 + SAa) + SA(3(l + 17.2) 2 S(38 2 + SA(3a + (3) + 3

r = S(3h 2 ' a = S(3h 2 .

All of these solutions are inflating in the sense that the Hubble scalar H = Ho.
They are also space-time homogeneous. For the type I, II and VI h solutions the
 1209

inflation does not result in approach to isotropy or to asymptotic evolution close to

the de Sitter metric. Interestingly, even in the case of a vanishing A the universe
inflates exponentially but anisotropically. We also note from the solutions that the
essential term in the action causing this solution to exist is the f3Rf.'URIlv -term and
the distinctive behaviour occurs when 0 = o.

3. Discussion
The solutions that we have found raise new questions about the thermodynamic
interpretation of spacetimes. We are accustomed to attaching an entropy to the
geometric structure created by the presence of a cosmological constant, for example
the event horizon of de Sitter spacetime. Do these anisotropically inflating solutions
have a thermodynamic interpretation? If they are stable they may be related to
dissipative structures that appear in non-equilibrium thermodynamics and which
have appeared been identified in situations where de Sitter metrics appear in the
presence of stresses which violate the strong energy condition. 5 They also provide a
new perspective on the physical interpretation of higher-order gravity terms in the
gravitational Lagrangian.
In summary: we have found exact cosmological solutions of a gravitational the-
ory that generalises Einstein's by the addition of quadratic curvature terms to the
action. These solutions display the new phenomenon of anisotropic inflation when
A > O. They do not approach the de Sitter spacetime asymptotically and provide
examples of new outcomes for inflation that is driven by a p = -p stress and begins
from 'general' initial conditions.

References
1. J.D. Barrow and S. Hervik, Phys.Rev. D 73, 02300 (2006).
2. A. Guth, Phys. Rev. D 23, 347 (1981); A.D. Linde, Phys. Lett. B129, 177 (1983).
3. J.D. Barrow, Perturbations of a De Sitter Universe, In The Very Early Universe, eds. G.
Gibbons, S.W. Hawking and S.T.C Siklos, (Cambridge UP, Cambridge, p.267, 1983);
W. Boucher and G.W. Gibbons, In The Very Early Universe, ed. G. Gibbons, S.W.
Hawking and S.T.C Siklos, (Cambridge UP, Cambridge, p. 273, 1983); A.A. Starobin-
skii, Sov. Phys. JETP Lett. 37, 66 (1983); L.G. Jensen and J. Stein-Schabes, Phys.
Rev. D 35, 1146 (1987);
4. R. Wald, Phys. Rev. D 28, 2113 (1983).
5. J.D. Barrow, Phys. Lett. B 180,335 (1987); J.D. Barrow, Nue!. Phys. B 310, 743 (1988);
J.D. Barrow, Phys. Lett. B 183, 285 (1987); J.D. Barrow, Phys. Lett. B, 187, 12 (1987).
 THICK BRANE SOLUTION WITH TWO SCALAR FIELDS

VLADIMIR DZHUNUSHALIEV *
Dept. Phys. and Microel. Engineer., Kyrgyz-Russian Slavic University, Bishkek,
Kievskaya Str. 44, 720021, Kyrgyz Republic
[email protected]

HANS-JURGEN SCHMIDT
Inst. Math., University Potsdam, Am Neuen Palais 10, D-14469 Potsdam, Germany
[email protected]

KAIRAT MYRZAKULOV
Institute of Physics and Technology, 050032, Almaty, Kazakhstan

RATBAY MYRZAKULOV
Institute of Physics and Technology, 050032, Almaty, Kazakhstan
[email protected]

A new 5D thick brane solution is presented. We conjecture that the deduced thick brane
is a plane defect in a bulk gauge condensate.

1. Introduction
In recent years there has been a revived interest in theories having a larger number
of spatial dimensions than the three that are observed. In contrast to the original
Kaluza-Klein theories of extra dimensions, the recent development of extra dimen-
sional theories allow the extra dimensions to be large and even infinite in size (in
the original Kaluza-Klein theories the extra dimensions were curled up or com-
pactified to the experimentally unobservable small size of the Planck length: 10- 33
cm). These new extra dimensional theories have opened up new avenues to explain-
ing some of the open questions in particle physics (the hierarchy problem, nature
of the electro-weak symmetry breaking, explanation of the family structure) and
astrophysics (the nature of dark matter, the nature of dark energy). In addition
they predict new experimentally measurable phenomena in high precision gravity
experiments, particle accelerators, and in astronomical observations.
We consider 5D gravity + two interacting fields. 1 The key for the existence of a
regular solution here is that the scalar fields potential have to have local and global
minima, and at infinity the scalar fields tend to a local but not to global minimum.
The 5D metric is
ds 2 = a(yhl-'vdxl-'dxV - dy2. (1)
The Lagrangian for the scalar fields ¢ and X is
1 A 1 A
£=';j,'VA¢V ¢+2VAXV X-V(¢,X), (2)

*Senior Associate of the Abdus Salam ICTP.

1210
 1211

where A 0,1,2,3,5. The potential V(¢, X) is

Ai (2
V(¢,X ) =4 ¢ -ml2) 2 +4
A2 (2
X -m22) 2 +¢X
2 2
-Vo, (3)
where Vo is a constant which can be considered as a 5D cosmological constant A.
The profile of the potential V (¢, X) is presented in Fig. 1. The corresponding field

0.5

0.4

0.3

0.2

0.1

Fig. 1. The profile of the potential V(¢, X).

equations are
a" a,2 1
2 (¢'2 + X,2) , (4)
a

~ [¢'2 + X,2 _ ~1 (¢2 _ mD2 _ ~2 (X2 m~)2 ¢2X2 + 2VoJ ,(5)

,
¢" + 4~¢' ¢ [X2 + Ai (¢2 mD], (6)
a
a'
X" + 4-X' X [¢2 + A2 (X2 - m~)] . (7)
a
The boundary conditions are
a(O) ao, a'(O) 0, ¢(O) ¢o, ¢'(O) = 0, X(O) = Xo, X'(O) = O. (8)
The boundary condition (8) and Eq. (5) give us the following constraint

Va 4Ai (2
¢o
2)2
ml + A24 (2
Xo
2)2
m2
1 2 2
+ 2¢oXo, (9)
This means that the constant Vo is not an arbitrary constant but it is defined by
the Z2·symmetry of the thick brane.
The mathematical problem for solving these equations is that not for all values of
the parameters ml,2 a regular' solution exists. Thus the problem of finding a regular
solution of these equations is a non-linear eigenvalue pr'oblem for the parameters
ml,2 and the eigenfunctions 4), X. In Fig. 2 we present the regular solution which
1212

describes the thick brane solution for the 5D gravity. The dimensionless energy
density is presented in Fig. 3. The thickness of the brane depends on all parameters
which are included in the equations: '\1,2, ¢(O), X(O).

2,2 0,10
2,0

1,8

1,6

1,4
~ ,'(y)
0,05

1,2 a(y)

~)
1,0

0,8

0,6
•'" 10

0,4
-0,05
0,2

10
-0,10

Fig, 2, The functions a*(y),¢*(y),X*(Y) Fig, 3. The profile of dimensionless energy

density,

2. Main features of the presented solution

• At the infinity the scalar fields tends to a local minimum not to global.
• The first item leads to the fact that the solution is topologically trivial.
• Mathematically the solution of Einstein-scalar fields equations is a non-
linear eigenvalue problem.
• There are arguments e)
that these scalar fields present non-perturbatively
quantized SU(3) gauge field. In this case the brane world is a plain defect
in 5D spacetime filled with a gauge condensate.

More background material can be found in Refs. 3 and. 4

References
l. V. Dzhunushaliev, "Thick brane solution in the presence of two interacting scalar
fields", gr-qc/0603020,
2. V. Dzhunushaliev, "Color defects in a gauge condensate", hep-ph/0605070.
3. V. Dzhunushaliev, H.-J. Schmidt, Phys. Rev. D 62 (2000) 044035; gr-qc/991l080.
4. Juan Maldacena, Liat Maoz, JHEP 0402 (2004) 053, hep-th/040102,
 SHEAR DYNAMICS IN BIANCHI I COSMOLOGIES
WITH Rn-GRAVITY

J. A. LEACH 1 *, P. K. S. DUNSBy12 and S. CARLONIl

1. Department of Mathematics and Applied Mathematics,
University of Cape Town, Rondebosch, 7701, South Africa
2. South African Astronomical Observatory,
Observatory, Cape Town, South Africa
* E-mail: [email protected]

We consider the case of Rn-gravity and perform a detailed analysis of the dynamics
in Bianchi I cosmologies which exhibit local rotational symmetry (LRS). We find exact
solutions and study their behaviour and stability in terms of the values of the parameter
n. In particular, we found a set of cosmic histories in which the universe is initially
isotropic, then develops shear anisotropies which approaches a constant value.

Keywords: Higher Order Gravity; f(R)-theories; Isotropisation; Bianchi I spacetimes.

1. Introduction

The dynamical systems approach 1 has been used with great success over the last
30 years, to a gain (qualitative) description of the global dynamics of cosmologi-
cal models. This method provides a useful tool for finding exact solutions which
correspond to fixed points of the system.
Carloni et al 2 have recently used this method to study the dynamics of Rn_
theories in Friedmann-Lemaltre-Robertson-Walker (FLRW) universes. Clifton and
Barrow 3 used the dynamical systems approach to determine the extent to which
exact solutions can be considered as attractors of spatially flat universes at late
times. They compared the predictions of these results with a range of observations
and argued that the parameter n in FLRW may only deviate from GR by a very
small amount (n - 1 ~ 10- 19 ).
The main aim of this paper 4 is to see how the shear behaves in LRS Bianchi I
cosmologies in Rn_ gravity and whether these models isotropises at early and late
times. To achieve this goal we use the theory of dynamical systems 1 to analyse the
system of equations governing the evolution of this model with and without matter.

2. Dynamics in LRS Bianchi I cosmologies

The cosmological equations that we require for our analysis are:
. 1221 R f.L
(1)
8 + :3 8 + 20- - 2n R - (n - 1)R 8 + nRn-1 = 0,
1
-8
2
-(J"
2 R
+(n-I)-8-
(n - 1) f.L
R---=O (2)
3 R 2n nRn-1'

a= - (8 + (n -1)~) (J", (3)

/1, = -(1 + w)f.L8, (4)

1213
1214

where 8 = 3ali is the volume expansion and cr is the shear (cr 2 = ~crabcrab).
In order to convert the above equations into a system of autonomous first order
differential equations, we define the following set of expansion normalised variables;
3R
y = --en -1) (5)
2n8 2 '

whose equations are

~'= 2 (-2 + 2~ - - y - - 2x + z)~,

n-1
x' = y(2 + x) - -Y-(2 + nx) - 2x - 2X2 + XZ + (1 - 3w)z + 2x~, (6)
n-1
y' = - y - [(3 - 2n)x - 2y + 2(n - l)z + 4(n - 1)~ + 2(n - l)J,
n-1
z' = Z [2Z - (1 + 3w) - 3x - -'!:JL
n-1
+ 4~] ,
1- ~ +x- Y- Z = 0,
where primes denote derivatives with respect to a new time variable T = In a.
The solutions associated to the fixed points can be obtained from:
.
8=-8 2
3a '
1
a= (2+~.1. _~y.)~l
n-l 1. , (7)
Cr (3
-=--8 (3 = 2 + ~i + Yi + Zi· (8)
cr 3'
Under the condition that n =1= 1 and the terms inside the brackets of (7), do not add
up to zero, these equations may be integrated to give the following solutions

a = ao (t - to t, (9)
In terms of our expansion normalised variables (5), the energy density is given
by

(10)
From this relation it can be seen that when Z = 0 and Y =1= 0 the energy density
is zero. However when y = 0 and Z =1= 0 the behaviour of J.l does depend on the
value of n. In this case the energy density is zero when n > 1 but is divergent when
n < 1. When both y and Z are equal to zero and n < 1, one can only determine the
behaviour of J.l by direct substitution into the cosmological equations.

2.1. Vacuum analysis

The vacuum case is characterised by Z = O. In this case we can obtain the fixed
points of (6) by setting ~' = 0 and y' = O. We find one isotropic fixed point
(A): (0, ~~=~) and a line of fixed points (.cd: (~*' 0), with non-vanishing shear.
The isotropic fixed point is an attractor (stable node) for values of the parameter
n in the ranges n < 1/2, 1/2 < n < 1 and n > 5/4. In the range 1 < n < 5/4 this
 1215

(a) (b)
Fig. 1. Phase space of the vacuum model. (a) 1/2 < n < 1. (b) 1 < n < 5/4. The shaded regions
represents the regions of initial conditions for which the shear will always evolve faster than in
GR.

point is a repeller (unstable node) and therefore may be seen as a past attractor.
However, we also have attractors for (a/ H)* < < 1 on £1. Therefore inflation may
not be needed since the shear anisotropy approaches a constant value which may
be chosen as the expansion normalised shear observed today 5-7 ((a/H)* < 10- 9 ),
provided that other observational constraints such as nucleosynthesis are satisfied.
The phase space is divided into two regions by the line y = 1- I:, which represents
all points for which the shear dissipate at the same rate as in GR. The region
y < 1 - I: represents a fast shear dissipation (FSD) regime where shear dissipates
faster than in GR, and the region y > 1 - I:, is a slow shear dissipation (SSD)
regime where the shear dissipates slower than in GR (see Figure 1).

2.2. Matter analysis

Setting I:' = 0, y' = 0 and z' = 0 we obtain three isotropic fixed points and a line
of fixed points with non-vanishing shear.
We observe the same kind of behaviour as in the vacuum case; the phase space
is however 3-dimensional, but is similarly divided into two regions, by the plane
1 = I: + y + z. The space above the plane is a SSD region and below a FSD region.

3. Conclusions
In conclusion we have shown that Rn_ gravity modifies the dynamics of the shear in
LRS Bianchi I cosmologies by altering the rate at which the shear dissipates. There
are cases in which the shear always dissipates slower or faster than in GR, and there
are ones which make the transition from first evolving faster and then slower (and
vice versa) than in GR.

References
1. Wainwright J and Ellis G F R (ed) 1997 Dynamical systems in cosmology (Cambridge:
Cambridge University Press) (see also references therein)
1216

2. Carloni S, Dunsby P K S, Capozziello S and Troisi A 2005 Class. Quantum Grav. 22

4839
3. Clifton T and Barrow J D 2005 Phys. Rev. D72 103005
4. Leach J A, Carloni Sand Dunsby P K S 2006 Class. Quantum Grav. 234915
5. Bunn E F, Ferreira P G and Silk J 1996 Phys. Rev. Lett. 772883
6. Kogut A, Hinshaw G and Banday A J 1997 Phys. Rev. D55 1901
7. Jaffe T R, Banday A J, Eriksen H K, Gorski K M and Hansen F K 2005 Astrophys.
1. 629 L1
 SPONTANEOUS LORENTZ VIOLATION, GRAVITY
AND NAMBU-GOLDSTONE MODES

ROBERT BLUHM
Department of Physics, Colby College Waterville, ME 04901 USA
[email protected]

A brief summary is presented of recent work examining the fate of the N ambu-Goldstone
modes in gravitational theories with spontaneous Lorentz violation.

1. Introduction

The Goldstone theorem states that when a continuous global symmetry is spon-
taneously broken, massless Nambu-Goldstone (NG) modes appear. On the other
hand, if the symmetry is local, then a Higgs mechanism can occur in which the
gauge fields acquire mass. In this work, these processes are examined for the case
where the symmmetry is Lorentz symmetry. 1

2. Spontaneous Lorentz Breaking

Lorentz symmetry is spontaneously broken when a local tensor field acquires a vac-
uum expectation value (vev). In curved spacetime, the Lorentz group acts locally at
each spacetime point. In addition to being locally Lorentz invariant, a gravitational
theory is also invariant under diffeomorphisms. There are therefore two relevant
symmetries, and it is important to consider them both.
One can show in general that when a vacuum expectation value spontaneously
breaks Lorentz symmetry, then diffeomorphisms are also spontaneously broken. The
spontaneous breaking of these symmetries implies that NG modes should appear
(in the absence of a Higgs mechanism). In general, there can be up to as many NG
modes as there are broken symmetries. Since the maximal symmetry-breaking case
would yield six broken Lorentz generators and four broken diffeomorphisms, there
can therefore be up to ten NG modes. One natural gauge choice is to put all the
NG modes into the vierbein, as a simple counting argument shows is possible. The
vierbein has 16 components. With no spontaneous Lorentz violation, the six Lorentz
and four diffeomorphism degrees of freedom can be used to reduce the vierbein
down to six independent degrees of freedom. Thus, in a theory with spontaneous
Lorentz breaking, up to ten NG modes can appear, and all of them can naturally
be incorporated as degrees of freedom in the vierbein.
The simplest example of a theory with spontaneous Lorentz breaking is a bum-
blebee modeJ.2,3 These are defined as theories in which a vector field EM acquires
a vev, < BM > = bw The vev can be induced by a potential V in the Lagrangian
that has a minimum for nonzero values of the vector field. A simple example of a
bumblebee model has the form L = LC + L8 + LM, where LC describes the pure-
gravity sector, LM describes the matter sector, and (choosing a Maxwell form for

1217
1218

the kinetic term)

(1)

describes the bumblebee field. Here, P' is a matter current, and the bumblebee field
strength is Bl"v = DI"Bv - DvBI"' which in a Riemann spacetime (with no torsion)
reduces to Bl"v = 0I"Bv -ovB/"" (For simplicity, we are neglecting additional possible
interactions between the curvature tensor and B 1").
Among the possible choices for the potential are a sigma-model potential V =
A(BI"Bv ± b2 ), where A is a Lagrange-multiplier field, and a squared potential V =
~f£(BI"BV ± b2)2, where f£ is a constant (of mass dimension zero). In the former
case, only excitations that stay within the potential minimum (the NG modes) are
allowed by the constraint imposed by A. However, in the latter case, excitations
out of the potential minimum are possible as well. In either of these models, three
Lorentz symmetries and one diffeomorphism are broken, and therefore up to four
NG modes can appear. However, the diffeomorphism NG mode does not propagate. 1
It drops out of all of the kinetic terms and is purely an auxiliary field. In contrast,
the Lorentz NG modes do propagate. They comprise a massless vector, with two
independent transverse degrees of freedom (or polarizations). Indeed, they are found
to propagate just like a photon.
We find that the NG modes resulting from spontaneous local Lorentz violation
can lead to an alternative explanation for the existence of massless photons (besides
that of U(l) gauge invariance). Prevoius links between QED gauge fields, fermion
composites, and the NG modes had been uncovered in flat spacetime (with global
Lorentz symmetry). Here, we propose a theory with just a vector field (but no U(l)
gauge symmetry) giving rise to photons in the context of a gravitational theory
where local Lorentz symmetry is spontaneously broken. Defining BI" - bl" = AI"'
we find at lowest order that the Lorentz NG excitations propagate as transverse
massless modes obeying an axial gauge condition, bl"AI" = O. Hence, we conclude
that spontaneous local Lorentz violation may provide an alternative explanation for
massless photons. In the bumblebee model, the photon fields couple to the current
JI" as conventional photons, but also have additional Lorentz-violating background
interactions like those appearing in the Standard-Model Extension. By studying
these additional interactions, signatures can be searched for that might ultimately
distinguish between a photon theory based on local Lorentz breaking from that of
conventional Einstein-Maxwell theory.

3. Higgs Mechanisms
Since there are two sets of broken symmetries (Lorentz and diffeomorphisms) there
are potentially two associated Higgs mechanisms. However, in addition to the usual
Higgs mechanism (in which a gauge-covariant-derivative term generates a mass term
in the Lagrangian), it was shown 2 that an alternative Higgs mechanism can occur
due to the gravitational couplings that appear in the potential V.
 1219

First, it was shown that the usual Higgs mechanism involving the metric does
not occur. 2 This is because the mass term that is generated by covariant derivatives
involves the connection, which consists of derivatives of the metric and not the
metric itself. As a result, no mass term for the metric is generated following the
usual Higgs prescription. However, it was also shown that because of the form of
the potential, e.g., V = V(B/",gIW Bv + b2 ), quadratic terms for the metric can
arise, resulting in an alternative form of the Higgs mechanism. 2 These can lead to
mass terms that can potentially modify gravity in a way that avoids the van Dam,
Veltmann, and Zakharov discontinuity.
In contrast, for the case of Lorentz symmetry, it is found that a conventional
Higgs mechanism can occur. 1 In this case the relevant gauge field (for the Lorentz
symmetry) is the spin connection. This field appears directly in covariant derivatives
acting on local tensor components, and for the case where the local tensors acquire
a vev, quadratic mass terms for the spin connection can be generated following a
similar prescription as in the usual Higgs mechanism. For example. in the bumblebee
model, using a unitary gauge, the kinetic terms involving B/",v generate quadratic
mass terms for the spin connection. However, a viable Higgs mechanism involving
the spin connection can only occur if the spin connection is a dynamical field. This
then requires that there is nonzero torsion and that the geometry is Riemann-
Cartan.

4. Summary and Conclusions

In theories with spontaneous Lorentz violation, up to ten NG modes can appear.
They can all be incorporated naturally in the vierbein. For the vector bumblebee
model, the Lorentz NG modes propagate like photons in an axial gauge. In principle,
two Higgs mechanisms can occur, one associated with broken diffeomorphisms, the
other with Lorentz symmetry. While a usual Higgs mechanism (for diffeomorphisms)
does not occur involving the metric field, an alternative Higgs mechanism can lead
to the appearance of quadratic metric terms in the Lagrangian. If in addition the
geometry is Riemann-Cartan, then a Higgs mechanism (for the Lorentz symmetry)
can occur in which the spin connection acquires a mass.

Acknowledgments
This research has been supported by NSF grant PHY-0554663.

References
1. R. Bluhm and V.A. Kostelecky, Phys. Rev. D 71, 0065008 (2005).
2. V.A. Kostelecky and S. Samuel, Phys. Rev. D 40, 1886 (1989); Phys. Rev. D 39, 683
(1989); Phys. Rev. Lett. 63, 224 (1989).
3. V.A. Kostelecky and R. Lehnert, Phys. Rev. D 63, 065008 (2001); V.A. Kostelecky,
Phys. Rev. D 69, 105009 (2004); V.A. Kostelecky and R. Potting, Gen. ReI. Grav. 37,
1675 (2005); B. Altschul and V.A. Kostelecky, Phys. Lett. B 628, 106 (2005); V.A.
Kostelecky and Q.G. Bailey, gr-qc/0603030.
 SPONTANEOUS LORENTZ BREAKING, NAMBU-GOLDSTONE
MODES, AND GRAVITY

ROBERTUS POTTING
CENTRA and Physics Department,
FCT, University of the Algarve, Campus de Gambelas, 8005-139 Faro, Portugal
[email protected]

In general relativity, masslessness of gravitons can be traced to symmetry under diffeo-

morphisms. However, another elegant possibility exists: masslessness can instead arise
from spontaneous violation of local Lorentz invariance. We construct the corresponding
theory of gravity. It reproduces the Einstein-Hilbert action of general relativity at low
energies and temperatures. We discuss its merits and implications.

1. Introduction
For vector particles masslessness is generally taken to arise from gauge invariance
of the action. Another possibility that has been considered is that massless ness can
arise from the spontaneous symmetry breaking of Lorentz symmetry, 1 as discussed
in the talk by Robert Bluhm in this session. To this effect, one includes a scalar
potential for the vector field which has a minimum for nonzero field value. Massless
vector bosons appear that correspond to the broken Lorentz generators, and the
resulting model is, at low energy, equivalent to the corresponding gauge theory in
an axial gauge.
Recently this idea has been applied to the case of Lorentz 2-tensors,2 where it
was shown how massless gravitons arise by spontaneous breaking of local Lorentz
invariance. The resulting model corresponds to linearized general relativity in a
fixed axial-like gauge. In this talk, a report on ongoing work with Alan Kostelecky, 3
we show this model can be extended to a nonlinear form in a self-consistent way,
yielding the Einstein-Hilbert action of general relativity at low energies and tem-
peratures.
Note that this construction goes beyond incorporating Lorentz violation into a
theory of gravity.4

2. Linearized cardinal theory

Starting point for the linearized "cardinal" model is the lagrangian 2

I: = ~ CfLVJ{fL va (3C a(3 + V (CfLV, rtfL V) (1)

where CfL V is a 2-tensor, J{fL va (3 the quadratic kinetic operator for a massless a
spin-2 field and rtfLv a fiat background metric. V is a potential function built out of
the 4 independent scalars that can be built out of OfL V and rtfL v , namely OfLVrtvfL'
(0· rt . 0 . rt)~, (0· rt . 0 . rt . 0 . rl)~ and Crt· C . rt . 0 . rt . 0 . rt)~. Assuming V
has its minimum for OfL V = cfLV #- 0, this will force OfL V to acquire an expectation

1220
 1221

value (01,V) = A~cpv with A~ some Lorentz matrix (which we will assume to be
space-time independent), thus spontaneously breaking Lorentz symmetry.
At low energies, fluctuations in the directions normal to the minimum of V
will be negligible, and V can be approximated by sum of delta-functions that fix
the 4 independent scalars multiplied by Lagrange multipliers Ai (i = 0, .. ,3). The
low-energy equation of motion becomes

K pva /3h a /3 - AoTjpv - Al (TjCTj)pv - A2(TjCTjCTj)f1,V - A3(TjCTjCTjCTj)pv =0 (2)

with the constraints h~ = cpvh pv = (cTjc)pVhpv = (C7]c7]c)pVhpv = O. The resulting
low-energy dynamics of fluctuations around the vacuum expectation value equals
to linearized general relativity in the gauge defined by the constraints. 2

3. Exponentiation
As it stands, the linearized theory of gravity described above, while consistent,
has no matter or self-energy couplings. Inclusion of a coupling to matter can be
accomplished by inclusion of the term hpvTpv in the Lagrangian, where Tpv is the
trace-inversed energy-momentum tensor, yielding the linearized Einstein equation
K pva/3 h a/3 -
= RL -
pv -
T
pv' (3)
The energy-momentum tensor not only consists of contributions of matter, but
there is a contribution T~l) I"V of the gravitons themselves as well, quadratic in hl"v'
This requires the inclusion of a cubic term in the Lagrangian. This, in turn, implies
a cubic contribution T~2) I"V to the energy-momentum tensor, corresponding to a
quartic term in the Lagrangian. This process continues indefinitely,5 yielding in the
limit the full Einstein-Hilbert action RI"v (ca/3).
This process of "exponentiation" can be done in one step by using a procedure
used by Deser for general relativity.6 Here one rewrites the kinetic term in (1) in
first-order form with the aid of connection coefficients r~v as additional variables.
The exponentiation of the kinetic term then terminates after one step, yielding the
Einstein-Hilbert action RI"v(ca/3), where ca/3 naturally assumes the role of curved-
space metric density.
Exponentiation of the flat-space matter energy-momentum tensor yields the
usual curved-space matter lagrangian with metric density Cpv.
The exponentiation procedure can be applied consistently to the potential V
if the latter depends on scalars constructed out of Cpv (upper indices) and the
background Minkowski metric Tj I"V (lower indices).

4. Fluctuations in the non-flat directions

At low energies, the potential constrains CI"v-fluctuations to be strictly along the
flat directions of its minimum. This is no longer strictly the case at high energies or
temperatures, when fluctuations in the perpendicular directions, corresponding to
positive curvature, contribute to and affect the solutions to the equations of motion.
1222

These effects can be investigated systematically, for instance, by integrating out

the fluctuations in the perpendicular directions, leaving an effective action for the
remaining degrees of freedom.
Also significantly affected by, and sensitive to, the precise form of the cardinal
potential will be the quantum properties of the model. This may shed light on the
perennial problem of the consistent quantization of gravity. For example, recent
results for vector fields 7 can be adapted for the new theory to show that nonpoly-
nomial and superficially unrenormalizable potentials V can become renormalizable
and stable when quantum corrections are included.

5. Discussion
We described the construction of a theory of gravity, the cardinal theory, with
massless gravitons arising from spontaneous breaking of Lorentz symmetry by a
potential. The full nonlinear form of the Lagrangian is fixed by consistent coupling
of gravity to the total energy-momentum tensor. At low energy, the Lagrangian
corresponds to the Einstein-Hilbert action. At higher energies and temperatures,
the dynamics, both on the classical and the quantum level, becomes modified by
the potential, obviating the fact that the physics of the cardinal model is different
from that of General Relativity. To quantify these differences, a careful analysis of
the effect of the perpendicular fluctuations at high energy or temperature (early
Universe) and near singularities is necessary. We will report on these issues in the
near future. 3

Acknowledgments
It is a pleasure to thank Alan Kostelecky for collaboration. Partial support by the
Funda<;iio para a Ci€mcia e a Tecnologia is gratefully acknowledged.

References
1. R Bluhm and V.A. Kostelecky, Phys. Rev. D 71, 065008 (2004).
2. V.A. Kostelecky and R Potting, Gen. Rei. Grav. 37, 1675 (2005).
3. V.A. Kostelecky and R Potting, in preparation.
4. V.A. Kostelecky, Phys. Rev. D 69, 105009 (2004).
5. R Kraichnan, MIT thesis, 1947; Phys. Rev. 98, 1118 (1955); A. Papapetrou, Proc. Roy.
Irish Acad. 52A, 11 (1948); S.N. Gupta, Proc. Phys. Soc. London A65, 608 (1952);
RP. Feynman, Chapel Hill Conference (1956).
6. S. Deser, Gen. ReI. Grav. 1, 9 (1970).
7. B. Altschul and V.A. Kostelecky, Phys. Lett. B 628, 106 (2005).
THE SIGNIFICANCE OF MATTER COUPLING IN f(R) GRAVITY

THOMAS P. SOTIRIOU
SISSA-International School for Advanced Studies, via Beirut 2-4, 34014, Trieste, Italy and
INFN, Sezione di Trieste
[email protected]

The way one chooses to couple gravity to matter is an essential characteristic of any grav-
itational theory. In theories where the gravitational field is allowed to have more degrees
of freedom than those of General Relativity (e.g. scalar-tensor theory, f(R) gravity) this
issue often becomes even more important. We concentrate here on f(R) gravity treated
within the Palatini variational principle and discuss how the coupling between matter
and the extra degrees of freedom of gravity (the independent connections in our case)
affects not only the resulting phenomenology but even the geometrical meaning of fun-
damental fields.

Some of the most prominent questions in Physics are nowadays related to grav-
ity. These questions are relevant for High Energy Physics, since finding a quantum
theory of gravity has proved to be a difficult task and theories currently consid-
ered for the description of gravity at small scales seems to imply that non trivial
corrections with respect to the Einstein-Hilbert term should be included in the
gravitational action. 1 They are also relevant for Cosmology and Astrophysics, since
current observations indicate that approximately 70% of the energy density of the
universe is due to an unkown form of energy, which is often called dark energy and
is considered to be responsible for the late time accelerated expansion of the uni-
verse. 2 It is reasonable to examine the possibility that these problems are actually
related: corrections in the effective low energy gravitational action coming from our
high energy theories can lead to modified gravity even at large scales, which might
account for dark energy.
One of the easiest modifications of gravity comes from generalizing the Einstein-
Hilbert action by assuming that the gravitational Lagrangian is a general function
of the scalar curvature, feR). But is feR) gravity really the answer? Possibly not,
but this is actually the wrong question to ask! Since we still have little evidence
about which corrections to the gravitational action are most likely to appear -- this
is theory dependent and there are numerous candidates for "quantum" gravity-
we seem to be more in need of a toy low energy gravitational theory to understand
how different corrections influence our picture for the gravitational interaction. And
feR) gravity seems to be very suitable for these purposes.
If SM is used to denote the matter action, the action for feR) gravity is

(1)

For feR) = R it reduces to the Einstein-Hilbert action and metric variation yields
the Einstein equations. As can be found in textbooks,3 an alternative is to use the

1223
1224

Palatini variation, i. e. an independent variation of the metric and connection which

gives the same field equations but also the expression for the connection. For doing
this we take R = g/-t v R/-tv (r), where R/-tv (r) is the Ricci tensor constructed with the
independent connection r\v' What needs to be stressed here is that in order to
derive the Einstein equations with the Palatini variation one has to make an extra
assumption: that 5 M does not depend of the connection, and hence

155M = 0 (2)
K A .
ur /-tv
Under this assumption, the Palatini variation yields the field equations 5

j'(R)R(/-tv) - ~f(R)gJ1v = 87fGT1w , (3)

\7 A (v0jj'(R)g/-t V ) = 0, (4)

which for f(R) = R reduce to Einstein's equations and the definition of the Levi-
Civita connection respectively. The Palatini variation together with the assumption
stated in eq. (2) constitute was is called the Palatini formalism or Palatini f(R)
gravity, if the action has the form of (1). For a more general choice of f, eq. (4)
implies that r\v is the Levi-Civita connection of the metric h/-tv = l' (R)g/-tv, and
therefore eq. (3) can be written as an equation for g/-tv, which, however, is different
from both Einstein's equations and the field equations of f(R) gravity that one
derives by standard metric variation. 4
What I want to focus on here is that the assumption (2) is physically meaningful
and not trivially satisfied. The matter Lagrangians for scalar fields or the electro-
magnetic field do not depend on the connection. However, this is not true for all
matter fields, for example fermions. Therefore, forcing this assumption can mean
only two things: either only certain matter fields are included in the theory, or, for
some reason, Dirac fields or other matter fields that generally couple to the connec-
tion, couple to the Levi-Civita connection of the metric instead of the independent
one. The first option does not seem suitable for a theory describing the gravitational
interaction, since such a theory would be very limited. The second option is not very
appealing either. If the independent connection r\v
is to have the usual geometri-
cal properties, such as defining parallel transport and the covariant derivative, then
indeed this is the connection that Dirac fields should be coupled to. Actually in this
case, it is even more appropriate if this connection is allowed to be non-symmetric.
The resulting theory is then a metric-affine theory of gravity,6 whose field equations
resemble eqs. (3) and (4) but torsion terms and more importantly the matter term
!J./ v == -h !/~:v are present in the equivalent of eq. (4).
The fact, however, that assuming a priori that matter fields are coupled to the
Levi-Civita connection instead of the independent one is not very appealing, does
not mean that it is unfeasible. On the other hand, it does raise questions about the
geometrical meaning of r\v which is obviously not related with parallel transport
or the covariant derivative (unlike in metric-affine gravity). One can also show that
 1225

in the Palatini formalism it is not even related to matter conservation laws, since
these are expressed using the covariant derivate related to the metric. 7 Actually, if
r\v does not couple to matter, then the action (1) is dynamically equivalent to
the action of scalar-tensor theory with Brans-Dicke parameter Wo = -3/2. 8 Notice,
that this means that there is only one extra scalar degree of freedom besides the
metric, even thought r Altv has 64 component, 40 of which are initially independent
if it is assumed to be symmetric! Therefore, applying the Palatini variation whilst
assuming (2), is different from assuming that the metric and the true connection of
spacetime are independent. The true connection in this case is the Levi-Civita one
a prioTi and r\v is an auxiliary field whose introduction serves only to add a scalar
degree of freedom contrary to what one might think by examining the action.
Summarizing, it is worth mentioning the following: Claiming that the Pala-
tini variation leads to the Einstein equation directly is imprecise. Even when the
Einstein-Hilbert action is used, one needs the extra assumption that the indepen-
dent connection should not be coupled to the matter. The physical meaning of this
assumption is that this connection does not define the covariant derivative and is
therefore not the true connection of spacetime but an auxiliary field void of geomet-
rical meaning. The above can also be seen through the equivalence of Palatini f(R)
gravity with scalar tensor theory. If no such assumption is made then one gets a
metric-affine theory of gravity, which will have the same phenomenology as Palatini
f(R) gravity in cases where only matter fields that naturally do not couple to the
connection are considered, such as in cosmology.

Acknowledgements
The author wishes to thank Stefano Liberati and John Miller for fruitful discussions
and valuable comments.

References
1. 1. L. Buchbinder, S. D. Odintsov, and 1. L. Shapiro, Effective Action in Quantum
Gmvity (lOP, Bristol, 1992);
2. D. N. Spergel et al., arXiv: astro-ph/0603449.
3. R. M. Wald, Geneml Relativity (The University of Chicago Press, Chicago, 1984);
C. W. Misner, K. S. Thorne, J. A. Wheeler, Gmvitation (W. H. Freeman and Co., San
Francisco, 1973).
4. H. A. Buchdahl, Mon. Not. R. astr·. Soc. 150,1 (1970).
5. D. N. Vollick, Phys. Rev. D 68, 063510 (2003).
6. T. P. Sotiriou and S. Liberati, to appear in Ann. Phys. [arXiv: gr-qc/0604006];
T. P. Sotiriou and S. Liberati, arXiv: gr-qc/0611040.
7. T. Koivisto, Class. Quant. Gmu. 23, 4289 (2006).
8. T. P. Sotiriou, Class. Quant. Gmv. 23,5117 (2006).
CONSTRAINING ALTERNATIVE THEORIES OF GRAVITY WITH
THE ENERGY CONDITIONS*

SANTIAGO E. PEREZ BERGLIAFFA

Departamento de Fisica TeoTica, Instituto de Fisica, Universidade do Estado de Rio de Janeiro,
CEP 20550-013, Rio de Janeiro, Brazil
[email protected]

A new method to constrain gravitational theories depending on the Ricci scalar is pre-
sented. It is based on the weak energy condition and yields limits on the parameters of
a given theory through the current values of the derivatives of the scale factor of the
Friedmann-Robertson-Walker geometry. A further constraint depending on the current
value of the snap is also given. Actual constraints (and the corresponding error propa-
gation analysis) are calculated for two examples, which show that the method is useful
in limiting the possible feR) theories.

1. Introduction
It follows from several observations l that the universe is currently expanding with
positive acceleration. The many models that have been advanced to explain this
situation can be classified in two classes. The first class contains those models that
incorporate modifications to the matter side of Einstein's equations. This matter
(known as "dark energy") must violate the strong energy condition in order to
accelerate the universe in General Relativity (GR). The models in the second class
have normal matter as a source but assume that gravitation is not described by
GR at low curvatures. As an example of this latter class we can mention theories
depending on the Ricci scalar 2 ,3 (the so-called f(R) theories). Since many f(R) give
rise to accelerated expansion, how to reduce the theory space using observations is
an issue that requires attention. Constrains have been obtained from many different
sources. 4 Here a new criterion based on model-independent data shall be given, that
helps in deciding, without solving the EOM or making frame transformations or
assumptions about H, whether a given f(R) theory is appropriate to describe the
universe. The basic premise will be that the acceleration is due solely to a modified
gravitational theory with normal matter as a source. The criterion will then be
obtained by imposing the energy conditions on matter, yielding conditions on f(R)
and its derivatives w.r.t R in terms of the current value of the derivatives of the
scale factor. These conditions are to be satisfied if the theory given by f(R) is to
describe the current state of the universe, and they set limits on the parameters
that enter the theory under consideration.

2. Equations of Motion and Energy Conditions

Let us remind the reader that for a theory given by f(R), it follows from the the
EOM6 that the energy density and the pressure of the fluid can be expressed in

*This research was supported by UERJ and ICRANET.

1226
 1227

terms of the scale factor and its derivatives:

,
P = - f R tt - "2f + 3f"aR
--;;:, (1)

l' f"
p=--(Rtt+R)+--f (R 2aR)
" - - (2)
3 2 a

df 3' (...2)
where l' == dR' etc, R tt = aa, R = -6 ~ + ~2 ,and we have assumed a fiat
universe. Before proceeding to build with these equations the inequalities that define
the energy conditions,5 let us remark that observations show that the current matter
content of the universe (assumed here to be normal matter, as opposed to dark
energy) is pressureless. In this case the EC reduce to the inequality Po ~ 0 plus the
equation Po = 0, where the subindex 0 means that the quantity is evaluated today.
We shall express these conditions in terms of following kinematical parameters: the
Hubble and deceleration parameters, the jerk, and the snap, respectively given by 7
1 it; 1 .0,
H-~
- , j = H3~' s=--
H4 a'
a
While the current value of the first three parameters is Ho = 72±8km/(secMpc),8
qo = -0.81 ±0.14, jo = 2.16:::g:~~,g no measurements ofthe snap have been reported
yet. By writing Po ~ 0 in terms of the parameters we get

3qoH6f~ - ~o - 18H~ft(jo - qo - 2) ~ O. (3)

This inequality gives a relation between the derivatives of a given f(R) and it limits
the possible values of the parameters of the theory (see below).
Notice that Eqn.(2) involves the snap (through R). If we had a measurement of
So, we could use the equation Po = 0 to obtain another constraint on f(R). Since
this is not the case, we shall express Po = 0 in such a way that it gives the possible
current values of the snap for a given f(R):

So
E
= 6H~ f" (qo - 2) + 6H 5~
:" (-qo+ jo - 2)2 - [qo(qo +6) +2(1 + jo)]- 12H4 f'"
k (4)
OJO JO 0

3. An example
To see how Eqns.(3) and (4) can be used to put constraints on a given f(R) let us
examine an example, given bylO

(5)

with n EN, which can be taken as the n = 1 low-curvature limit of f(R) = R+ TI'
a model studied in. 2 Substituting Eqn.(5) in Eqn.(3) for a > 0 and n even we get

2
-3qoHonRo 1 2 - 18H4 n(n
- 2Ro + 1 )(Jo. - qo - 2 ) ~ O. (6)
o
1228

Replacing in this equation the numerical values of the parameters and using Ro =
6H5 (qO - I) we get an inequality that must be satisfied by n:

¢ == -17.64n 2 - 44.50n - 59.62 :2: O. (7)

Since this equation cannot satisfied by any real n, we conclude that n cannot be
even for a > O. The same analysis with odd n reverses the sign of the inequality
(7), so only odd values of n are allowed for a > O. This result generalizes that
obtained in 2 for n = I a. In the same way, we obtain that only even n are allowed for
a < O. These conclusions are valid even when the error coming from the kinematical
parameters is taken into account (see Fig.I). If we knew the value of So, we could

1 2 3 4 5 6 7 8 9 10
10 10

-500

cfJ, 6~~ -100 0 s,6s

-1500 -2 -2
-4
-2000 -6 -6
-8 -8
1 2 3 4 5 6 7 8 9 10

Fig. 2. Snap for f(R) as in Eqn.(5)

Fig. 1. ¢ (solid curve) and the associ-
(solid curve) and the associated error
ated error o¢ in terms of n.
in terms of n.

get a further constraint for the possible values of n using Eqn.(4). This equation
will be taken instead as giving the current value of the snap as a function of n (see
fig. (2)) b. We have also plotted in the figure the error associated with So (dashed
line), which grows as n for large n. The plot shows that with the current error of
the kinematical parameters the method outlined here is helpful in determining So
for the theory given by Eqn.(5) only for n = 1.

4. Discussion
A new method to restrict gravitational theories described by functions of the Ricci
scalar has been introduced. It is based essentially in the assumption that normal
matter composes the universe, the acceleration being caused by new gravitational
dynamics in the low curvature regime, described by f(R). By imposing that the
matter satisfy the weak energy condition, we obtain an inequality that constrain
the parameters in the theory. We have shown by way of an example how the method
can be used, and how it conduces to restrictive limits on the parameters, having
taken the error into account. We also obtained an equation that depends on the
snap, the fourth derivative of the scale factor. Had we any measurements of So, this

aNote the difference in the sign of R between our convention and the one used in. 2
bNote that Eqn.(4) is independent of the sign of a.
 1229

equation would furnish yet another condition on the parameters of the theory. Since
the current value of the snap has not been determined yet, we take this equation as
forecasting, for a given f(R), the current value of the snap. The method presented
here could be combined with other approach·es,12 to restrict the f(R) theories that
are being presented as candidates to model the acceleration of the universe.

References
1. See for instance S. Perlmutter et al., Astrophys. J. 511, 565(1999), astro-ph/9812133;
A. Riess et al., Astron. J. 116, 1009 (1998), astro-ph/9805201.
2. S. Carroll, V. Duvvuri, M. Trodden, M. Turner, Phys. Rev. D 10, 043528 (2004),
astro-ph/0306438.
3. S. Capozziello, S. Carloni, and A. Troisi (2003), astro-ph/0303041.
4. See references in Shin'ichi Nojiri, Sergei D. Odintsov, hep-th/0608008.
5. Lorentzian Wormholes, M. Visser, AlP Press, New York (1996), The Large Scale Struc-
ture of Spacetime, S. Hawking and G. Ellis, Cambridge University Press, England
(1973), C. Molina-Paris and M. Visser, Phys. Lett. B 455, 90 (1999), gr-qc/9810023,
C. Cattoen, M. Visser, Class. Quant. Grav. 22 4913 (2005), gr-qc/0508045. See
also M. Visser, Science 276, 88 (1997), J.H. Kung, Phys. Rev. D 53 3017 (1996),
gr-qc/9510008, B. Tippett, K. Lake, gr-qc/0409088, J. Santos, J.S. Alcaniz, M.J.
Reboucas, astro-ph/0608031.
6. See for instance R. Kerner, Gen. ReI. & Grav. 14, 453 (1982).
7. M. Visser, Class. Quant. Grav. 21, 2603 (2004), gr-qc/0309109.
8. W. Freedman et al. Ap. J. 553,47 (2001).
9. D. Rapetti, S. W. Allen, M. A. Amin, R. D. Blandford, astro-ph/0605683.
10. S. Capozziello et al., Int. J. Mod. Phys. D 12:1969-1982,2003, astro-ph/0307018.
11. S. Nojiri and S. D. Odintsov, Gen. ReI. & Grav. 36, 1765 (2004), hep-th/0308176.
12. A. De Felice, M. Hindmarsh, M. Trodden, astro-ph/0604154, S. Capozziello, A. Sta-
bile, A. Troisi, gr-qc/0603071.
 An feR) GRAVITATION FOR GALACTIC ENVIRONMENTS

Y. SOBOUTI
Institute for Advanced Studies in Basic Sciences, P. O. B. 45195-1159, Zanjan, Iran
[email protected]

We propose an action-based feR) modification of Einstein's gravity which admits of a

modified Schwarzschild metric. In the weak field limit this amounts to adding a small
logarithmic correction to the newtonian potential. A test star moving in such a spacetime
acquires a constant asymptotic speed at large distances. This speed, calibrated empir-
ically, is proportional to the fourth root of the mass of the central body in compliance
with the Tully-Fisher relation. A variance of MOND's gravity emerges as a consequence
of the formalism. It has also been shown (Mendoza et al. 2006) that a) the gravitational
waves in this spacetime propagate with the speed of light in vacuum and b) there is a
lensing additional to what one finds in the classic GR.

1. Introduction
Dark matter hypotheses, intended to explain the fiat rotation curves of spirals
have raised more questions than answers. Alternatives to newtonian dynamics have
been proposed but have had their own critics. The foremost among such theories,
the Modified Newtonian Dynamics of Milgrom (MOND) explains the fiat rotation
curves and the 1\llly-Fisher relation with considerable success. But it is often criti-
cized for the lack of an axiomatic foundation.
Here, we suggest to look for a modified Einstein gravity tailored to suit galactic
environments. We design an action integral, different but close to that of Einstein-
Hilbert, and find a spherically symmetric static solution to it. Next we analyze the
orbits of test objects moving in the ensuing modified spacetime and demonstrate
the kinship of the so obtained dynamics with that of MOND.

2. A modified field equation

We consider an isolated mass point endowed with the following action integral:

(1)

where R is the Ricci scalar and feR) is an, as yet, unspecified function of R. Vari-
ations of S with respect to the metric tensor leads to the following field equation

R/ll/ - ~g/LV ~ = (F;/LV - F;/,g/Lv) ~, (2)

where F = df I dR. For the purpose of galactic studies we envisage a spherically

symmetric static metric,

(3)

1230
 1231

We are interested in those solutions of Eqs (2) and (3) that differ from those of
the classic GR by small amounts. For the latter one has F = 1 and A(r)B(r) = 1.
Here, let us assume F(r) = (r/s)CX ~ 1 +aln(r/s), where a is a small dimensionless
parameter and s is a length scale of the system to be identified shortly. To the first
order in a, one finds
1 1 [ _ (~) (1-CX/2)] = (!:.)CX ~ 1 1 (r)CX
A (I-a) 1 r ,B s A' R=3a r2 , f=3a r2 ~ .(4)

For a = 0, Eq (4) is recognized as the Schwarzschild metric. Therefore, s is identified

with the Schwarzschild radius of a central body, 2G M / c2 . The Ricci scalar of the
Schwarzschild space is zero. For non zero a, however, it decreases as l/r2.

3. Application to galaxies
Solving the geodesic equations for the speed of a test star circling the central body
gives

(5)

where we have substituted s = 2GM/c 2 ) and kept only the first order terms in a.
A plot of 112 as a function of r has the horizontal asymptot ~ac2. To determine
this asymptot we argue that a cannot be a universal constant. For it will beat the
intuition to imagine that a galaxy and a speck of dust dictate the same speed on
faraway passerby objects. The parameter a should depend on the mass of the grav-
itating body residing at the origin. For any localized matter seen from far enough
distances will betray no characteristics other than its mass. To find this depen-
dence we resort to observations. From Sanders and Verheijn (1998) and Sanders
and McGaugh (2002) we have compiled a list of thirty one spirals for which total
masses, asymptotic orbital speeds, the actual velocity curves and curves derived the
observed luminous masses are reported. We have selected those objects which, a)
have a noticeable horizontal asymptot, b) have fairly reduced newtonian speeds by
the time the flat asymptot is approached, and c) do not possess anomalously high
HI content to hinder estimates of the total mass and the size of the galaxy. We have
further made the assumption that the total HI and stellar mass is distributed spher-
ically symmetrically and mimics a point mass if observed from faraway distances.
A power law fit to a = 211?:xo/ c2 and the total mass of the galaxy gives

(6)

It should be noted that Eq(6) is not a consequence of the present theory. It is

an empirical relation dictated by observations and is based on the masses and the
asymptotic speeds of a selected list of galaxies. Together with the popularly accepted
rule that the masses and the luminosities of spirals are linearly related, it leads to
a Thlly-Fisher relation, Luminosity ex: 1100 4.05.
1232

4. Kinship with MOND

The weak field limit of the present dynamics is derived by approximating
B(r) = 1 + 2¢(r)/c 2 = 1 + a + aln(r/s) - sir, (7)
and considering ¢(r) as the gravitational potential. From this and Eq(6) (with a
slight tolerance) the gravitational acceleration becomes
9 = Id¢/drl = (aog n )1/2 + gn, ao = a6cA/4GM8 gn = GM/r 2. (8)
The limiting behaviors of 9 are the same as those of MOND. One may, therefore,
comfortably identify ao as MOND's characteristic acceleration and calculate 000
anew. For ao = 1.2 x 1O- 8 cm/sec 2 , one finds a = 2.8 x 10- 12 (M/M8 )1/2. It is
gratifying how close this value of a is to the one in Eq (6) and how similar the low
and high acceleration limits of MOND and the present formalism are, in spite of
their totally different and independent starting points. It should also be noted that
there is no counterpart to the interpolating function of MOND here.

5. Concluding remarks
We have developed an f(R) ex R 1 - a / 2 gravitation which is essentially a logarithmic
modification of the Einstein-Hilbert action. In spherically symmetric static situa-
tions, the theory admits of a modified Schwarzschild metric. The latter in the limit
of weak fields gives a logarithmic correction to the newtonian potential. From the
observed asymptotic speeds and masses of spirals we learn that the correction is pro-
portional to almost the square root of the mass of the central body. Flat rotation
curves, the Tully-Fisher relation, and a version of MOND emerge as consequences.
Since the appearance of an earlier version of this paper in arXiv (Sobouti, 2006),
Mendoza et al. (2006) have investigated gravitational waves and lensing effects in
the proposed spacetime. They find: a) in any f(R) = R n gravitation, including the
present one, gravitational waves travel with the speed of light, and b) in the present
spacetime, there is a lensing additional to what one finds in the classical GR. Their
ratio of the additional deflection angle of a light ray, 5(3, to that in GR, (3CR, can
be reduced to

(9)
where rm is the impact parameter of the impinging light. The increase of the addi-
tionallensing with increasing rm should not be surprising. For the theory is designed
to highlight unexpected features at faraway rather than nearby distances.

References
1. Mendoza, S., & Rosas-Guevara, Y. M., 2006, arXiv:astro-phj0610390
2. Sanders, R. H., & Verheijen, M. A. W., 1998, arXiv:astro-phj9802240
3. Sanders, R. H., & McGough, S. S., 2002, arXiv:astro-phj0204521
4. Sobouti, Y., 2006, arXiv:astro-phj0603302 and Ai:>tron. Astrophys. 2007, to appear
 CAUSALITY AND SUPERLUMINAL FIELDS

JEAN-PHILIPPE BRUNETON
QIftc:co, Institut d'Astrophysique de Paris, UMR 7095-CNRS,
Un'iversite Pierre et Marie Ou'rie - Paris 6, 98 bis boulevard Arago F-75014, Paris, France
[email protected]

The expression of causality depends on an underlying choice of chronology, Since

a chronology is provided by any Lorentzian metric in relativistic theories, there are as
many expressions of causality as there are non-conformally related metrics over space-
time, Although tempting, a definitive choice of a preferred metric to which one may refer
to is not satisfying. It would indeed be in great conflict with the spirit of general covari-
ance. Moreover, a theory which appear to be non causal with respect to (hereafter, w.r.t)
this metric, may well be causal w.r.t another metric. In a theory involving fields that
propagate at different speeds (e,g. due to some spontaneous breaking of Lorentz invari-
ance) , spacetime is endowed with such a finite set of non-conformally related metrics. In
that case one must look for a new notion of causality, such that l. no particular metric
is favored and 2, there is an unique answer to the question: "is the theory causal?".
This new causality is unique and defined w.r.t the metric drawing the wider cone in the
tangent space of a given point of the manifold. Moreover, which metric defines the wider
cone may depend on the location on spacetime. In that sense, superluminal fields are
generically causal, provided that some other basic requirements are met,

1. Introduction
Many authors argue that superluminal fields are not causal l - 4 (but see Refs. S ). This
is not true, unless one refers to an indefensible notion of causality, Indeed, as the
notion of causality evolves from Newtonian dynamics to Special Relativity (SR), one
must as well reconsider the notion of causality from Special or General Relativity
(GR), in which spacetime is only endowed with the fiat (resp. gravitational) metric,
to the case where it is endowed with a finite set of Lorentzian metrics (notably then,
if there are superluminal fields).
In this short communication based on the more detailed paper,6 we thus look
for an expression of causality in such a multi-metric framework. The gravitational
metric field is denoted by g, and M is a four-dimensional differentiable manifold.

2. Causality and chronology in field theories

The analysis of the notion of causality leads in particular to the following:
Observation 1: Since causes must precede effects, causally connected events must
be time-ordered. Causality thus needs a notion of chronology to be expressed.
Observation 2: Any Lorentzian metric over M defines a local chronology (in the
tangent space), through the special relativistic notions of absolute future and past.
Gluing these two points together, we get the following
Main point: In relativistic field theories, there are as many notions of causality
as there are non-conformally related metrics over M. These metrics hi are the one
along which the various fields 'ljJi propagate, with velocities Ci f. Cj, Vi f. j.

1233
1234

This plurality of the notion of causality is the crucial feature of multi-metric

theories. Indeed, it does not make any sense to assert that a given theory is -or not-
causal, if one does not define to which metric (i.e. to which chronology) he refers
to. A theory which appear to be non causal w.r.t some metric may be causal w.r.t
another metric.
To face this issue, one may be tempted to assume that there exists a preferred
metric field over M. In other words, one may fix a preferred chronology and its
associated causal structure. Most of the literature on causality and super luminal
fields is based -often implicitly- on this first approach. In their famous textbook,7
Hawking and Ellis recognize explicitly that their notion of causality is defined w.r.t
the gravitational metric. This constitutes a "postulate which sets the metric g apart
from the other fields on M and gives it its distinctive geometrical character" 7 (p.60).
As a consequence, fields that propagate faster than gravitons are not causal. Thus,
"the null cones of the matter equations [must] coincide or lie within the null cone
of the spacetime metric g"7 (p.255).
Although such an attitude does not pose any problem when spacetime is endowed
with only one metric, as is the case of GR plus matter fields that couple to g, it
becomes highly problematic in the multi-metric case. First, indeed, there is no way
to find which metric should be favored, and which should not. Thus, by invoking
causality, different authors may find opposite requirements on the theory.6
Second, let us consider two fields 7jJi propagating along the metrics hi (i = 1,2),
such that 7/J2 travels faster than 7/JI. Following the above reasonning, we can define
causality w.r.t the metric hI. Then two observers that are spacelike related w.r.t hI
(and hence, non time-ordered) but timelike related w.r.t h2 must be considered as
causally disconnected, whereas they can interact thanks to the field 7/J2. The only
way to avoid so an absurd conclusion is to define causality w.r.t to the metric that
defines the wider cone in the tangent space (see below).
Third, any choice of a preferred metric is equivalent to a choice of preferred
coordinates which, locally, diagonalize it. But the existence of preferred coordinates,
or equivalently, of preferred rods and clocks, 6 is in great conflict with the whole spirit
of GR, namely diffeomorphism invariance; coordinates are meaningless in GR.
The above attitude is thus irrelevant in the multi-metric case. As an application,
one should not invoke such a notion of causality to put constraints on the theory
(notably in order to fix various signs), contrary to what is done in the literature. 4

3. An extended notion of causality and superluminal behaviors

There is only one relevant notion of (extended) chronology that does not refer to
a given metric. This consists 6,8 in defining the extended future of a point P as the
union of the futures of P defined by each metric hi. The corresponding (extended)
notion of causality is thus in accordance with the notion of interactiona . It is very

aThe other possibility is to define an extended future as the meet of each future of P. It would
however allow non-causally connected observers to interact, as in the previous section.
 1235

permissive in the sense that, by construction, any field theory is a pr'iorz causal
provided that the various fields propagate along Lorentzian metrics, so that the
(extended) spacelike region is never empty. Moreover, interactions cannot threaten
this causal behavior, since, by construction, the extended future and past are defined
at each point of M. Which metric defines the wider cone may thus depend on the
location on spacetime. In particular, superluminal fields are a priori causal.
Of course, this construction is not sufficient. Causality also requires, first, that
the whole theory has an initial value formulation. This is generically the case if
the field equations form a quasilincar, diagonal and second order hyperbolic sys-
tem. 6 ,7 Beware however that initial data must be assigned on hypersurfaces that
are spacelike in the extended sense, that is spacelike w.r.t to all metrics hi. All
the difficulties in the Cauchy problem of superluminal fields found in the literature
arise from an irrelevant choice of initial data surfaces. 1,2 Second, a local chronology
is not enough. We must have at hand a global chronology over spacetime, in order
to prevent, e.g. the existence of closed timelike curves. In the multi-metric case,
we shall also require that our extended chronology is a global one, that is that no
closed extended-timelike curves exist.
It has been shown 2 that a particular superluminal scalar field may suffer from
such a global pathology. This is however not enough to kill this theory, for the very
reason that GR itself may suffer from such causal anomalies. Therefore, difficulties
at a global level do not signal an intrinsic disease of superluminal fields. Rather,
they originate from the fact that the global topology of the Universe is not imposed
by local field equations. It is therefore necessary to assume that spacetime does not
involve any closed (extended) timelike curves to ensure causality.

References
1. Y. Aharonov, A. Komar and L. Susskind, Phys. Rev. 182, 1400 (1969)
2. A. Adams, N. Arkani-Hamed, S. Dubovsky, A. Nicolis and R. Rattazzi, lHEP 0610,
014 (2006)
3. N. Straumann, Mod. Phys. Lett. A21, 1083 (2006); G. Calcagni, B. Carlos and A. De
Felice, Nucl. Phys. B752, 404 (2006); C. Bonvin, C. Caprini and R. Durrer, Phys.
Rev. Lett. 97, 081303 (2006)
4. V. A. Rubakov, hep-th/0604153 ; A. De Felice, M. Hindmarsh and M. Trodden, lCAP
0608,005 (2006); A. Jenkins and D. O'Connell, hep-th/0609159 ; G. Gabadadze and
A. Iglesias Phys. Lett. B639, 88 (2006)
5. B. A. Bassett, S. Liberati, C. Molina-Paris and M. Visser, Phys. Rev. D62, 103518
(2000); C. Armendariz-Picon, T. Damour and V. Mukhanov, Phys. Lett. B458, 219
(1999); C. Armendariz-Picon and E. A. Lim, lCAP 0508, 007 (2005) ; A. D. Rendall
Class. Quant. Grav. 23, 1557 (2006)
6. J. P. Bruneton, gr-qc/0607055
7. S. W. Hawking and G. F. R. Ellis, The large scale structure of spacetime (Cambridge
Univ. Press, 1973)
8. M. A. Clayton and J. W. Moffat, Int. l. Mod. Phys. Dll, 187 (2002)
 GRAVITATION AS A VACUUM NONLINEAR
ELECTRODYNAMICS EFFECT

ALEXANDER A. CHERNITSKII
A. Friedmann Laboratory for Theoretical Physics, St.-Petersburg, Russia
and
State University of Engineering and Economics, Marata str. 27, St.-Petersburg, Russia, 191002
[email protected]

Effective Riemann space effect of vacuum nonlinear electrodynamics is considered in

the context of theory for unified gravitation and electromagnetism. The electromagnetic
four-vector potential in the scope of Born-Infeld nonlinear electrodynamics model is
considered as the unified field. The electromagnetic interaction appears naturally in the
first perturbation order by the small field of distant material objects. The gravitational
interaction appears naturally in the appropriate second order. In this case the effective
metric components contain the corresponding energy-momentum tensor components for
quick-oscillating electromagnetic field of the distant objects.

1. Introduction
The problem of unification for all interactions of material particles remains one of
the most important problems in modern theoretical physics. Particularly, unification
for the two known long-range interactions, viz., electromagnetism and gravitation,
must be considered as a priority problem in theoretical investigation of nature.
The approach to this problem considered here is connected with a consistent
application of the idea of nonlinear local unified field theory.
Spatially localized solutions in this theory correspond to solitary material par-
ticles. These solutions can be designated particle solutions. The many-particle
world configuration corresponds to a complicated many-particle world solution. Any
many-particle solution contains the appropriate particle solutions in the following
sense.
Each particle solution has at least ten free parameters for space-time rotation
and shift. Because of the nonlinearity, a sum of particle solutions is not a solution for
the model. But we can consider the free parameters of particle solutions to be weakly
time-dependent. This method is well known in the theory of nonlinear dynamics. A
sum of particle solutions with time-dependent free parameters can be considered as
an initial approximation to a many-particle solution. The time dependence of the
free parameters of the particle solutions corresponds to the interaction between the
particles.
This method, applied to a suitable model, must give electromagnetic and gravita-
tional interactions for the case when the interacting partieles are sufficiently distant
from each other.
For the first time, this approach to the problem of unification of the gravita-
tional and electromagnetic interactions appeared in the context of some nonlinear
electrodynamics model. 1 Later on, the approach was developed for another nonlin-
ear electrodynamic model (Born-Infeld). 2

1236
 1237

A distinguishing characteristic of this approach is that the gravitational interac-

tion must appear through an effective Riemannian space for propagation of particle
under consideration. This effective Riemannian space is induced by the electromag-
netic field of distant particles.

2. Model field equations

A variational principle of the model considered here is similar to the one proposed
by A.S. Eddington in the context of general relativity ideas 3 and investigated by
A. Einstein in the context of his unified field theory.4 Afterwards, M. Born and
L. Infeld used it in the context of nonlinear electrodynamics. 5
The model set of equations is 6
2
I ,W =- ~
£. (F{LV _ X
2 v
'7 E{LvuP F )
up, (1)

where £. == ViI - X2 I - X4 .12 1, I == F{Lv FV{L /2, .1 == E{Lvup F{LV FUP /8, E0l23 ==
Jfgj, E0123 = -1/ Jfgj, F{Lv == o{LAv - ovA{L' A{L is the electromagnetic potential.
The model energy-momentum tensor is

(2)
With the definitions Ei == FiO, Bi == _EDijk F jk /2, Fij = EOijk Bk, Di == fO i ,
ijk
Hi == EOijk pk /2, Iij = _EO Hk (Latin indexes take the values 1,2,3), the model
equations can be written as nonlinear equations for the electromagnetic field (see
also 7 ). This system has the characteristic equationS
-{LV o<f> o<f> _ 0
9 ox{L oxv - ,
(3)

where <f>(x{L) = 0 is the equation of a characteristic surface, and the induced metric
g{LV has the following very notable form, which is specific for the model under
consideration:

(4)
Here T{LV is the energy-momentum tensor defined in (2).

3. Electromagnetic and gravitational interactions

The model to be considered is an unusual electrodynamics not only because the
appropriate equations are nonlinear. This model does not contain the postulated
trajectory equation for charged particle in external electromagnetic field, i.e., the
postulated electromagnetic interaction. This interaction appears naturally as a man-
ifestation of the nonlinearity of' the model.
The electromagnetic interaction appears in this model as an electromagnetic
force acting on a massive charged particle 2 and a moment of force acting on a
particle with an electric or magnetic dipole moment and spin. 9 The appropriate
1238

dynamical equations follow from integral conservation laws for the field energy-
momentum and angular momentum, respectively (for details see 2 ,9). These obtained
equations, characterizing the electromagnetic interaction, have the corresponding
classical form.
The force and the moment of force contain the electromagnetic field of distant
particles in the first power. Thus we can say that the electromagnetic interaction
appear in the first order in the small field of distant particles.
An explanation of the gravitational interaction in the scope of this model based
on the effective Riemannian space with the metric 1l'" (4) induced by the elec-
tromagnetic field. According to the general method stated in the Introduction, dis-
tant particles modify the propagation conditions for particle under consideration by
means of this effective Riemannian space induced by the field of distant particles.
The effective metric includes the electromagnetic field components in even powers.
Thus we can say that the gravitational interaction appear in the second order in
the small field of the distant particles.
The cause of the gravitational interaction in this approach is the energy density
of the distant particle field. But to have the real behaviour of the gravitational
potential, i.e. l/r, we must take into account a quick-oscillating part of the distant
particle field with an electromagnetic wave background. In this case, an averaging
can give the necessary behaviour of the energy density, 1/r.10,11
For additional details see also. 12

4. Conclusion
Thus the present approach, based on a consistent application of the idea of a nonlin-
ear local unified four-vector field, can really unify electromagnetism and gravitation.

References
1. A. A. Chernitskii, Theoret. and Math. Phys. 90, 260 (1992).
2. A. A. Chernitskii, J. High Energy Phys. 12, Paper 10 (1999), hep-th/9911093.
3. A. Eddington, The Mathematical Theory of Relativity (Cambridge, 1924).
4. A. Einstein, Zur allgemeinen Relativitiitstheorie, in Sitzungsber. preuss. Akad. Wiss.,
phys.-math. (Kluwer Academic Publishers, 1923).
5. M. Born and L. Infeld, Proc. Roy. Soc. A 144,425 (1934).
6. A. A. Chernitskii, Helv. Phys. Acta 71, 274 (199S), hep-th/9705075.
7. A. A. Chernitskii, Born-Infeld equations, in Encyclopedia of Nonlinear Science, ed.
A. Scott (Routledge, New York and London, 2004). hep-th/05090S7.
S. A. A. Chernitskii, J. High Energy Phys. 11, Paper 15 (1998), hep-th/9S09175.
9. A. A. Chernitskii, Mass, spin, charge, and magnetic moment for electromagnetic par-
ticle, in XI Advanced Research Workshop on High Energy Spin Physics (DUBNA-
SPIN-05) Proceedings, eds. A. V. Efremov and S. V. Goloskokov (JINR, Dubna, 2006).
hep- th / 0603040.
10. A. A. Chernitskii, Gravitation Ei Cosmology 8, Supplement, 157 (2002), gr-qc/0211034.
11. A. A. Chernitskii, hep-th/0602079.
12. A. A. Chernitskii, Gravitation Ei Cosmology 12, 130 (2006), hep-th/0609204.
 ASYMPTOTIC FLATNESS AND BIRKHOFF'S THEOREM IN
HIGHER-DERIVATIVE THEORIES OF GRAVITY

T. CLIFTON
Physics Department, Stanford University, CA, USA
T. [email protected]

The spherically symmetric solutions to gravitational theories generated from La-

grangians of the form j(R) are considered. The asymptotic form of these solutions is
discussed, as is the lack of validity of Birkhoff's theorem. Exact solutions are presented
which illustrate these points, and their stability and geodesics investigated.

It has been known for some time that Birkhoff's theorem and asymptotic flatness
are not ensured in the vacuum solutions of generalised fourth-order theories of grav-
ity, as they are in general relativity1,2 (GR). The standard approach to overcoming
this problem is to consider theories which contain in their generating Lagrangian
an Einstein-Hilbert term which dominates the field equations in the low-curvature
limit. In this way the usual behaviour of approaching Minkowski space at asymptot-
ically large distances from sources is assumed to occur and perturbative expansions
about a Minkowski background are possible. The weak-field limit of the theory
in question can then be investigated in a straightforward way and compared with
experiment.
Whilst the existence of a Minkowski background is a great simplification, and
very useful in investigations of the weak-field, it is unclear whether or not a theory
should be disregarded solely on the grounds of this limit not existing. There are a
number of gravitational theories that can be conceived of which do not always ad-
mit Minkowski space as a solution. These include theories derived from Lagrangians
of the from Rn (where n i- 1),3-7 as well as those derived form R + a/ Rn (where
n > 0).9,10 The latter has caused considerable debate as to whether or not it is
compatible with solar system experiments. Studies on this subject usually follow
the prescribed analysis for computing the weak field geometry in GR; picking a
highly symmetric background* and calculating the form of spherically symmetric
perturbations by linear ising the field equations and solving them, to first order in
perturbations. However, it is often unclear how an appropriate background to ex-
pand about should be chosen, and how the lack of Birkhoff's theorem or asymptotic
flatness should be taken into account when performing these analyses. We will dis-
cuss these points here and attempt to make some progress into understanding the
spherically symmetric situation by using explicit exact vacuum solutions of the field
equations. Due to the need for brevity in these proceedings, the mathematical detail
will not be presented here; instead the reader is referred to a seperate publication. 11
We will now present two exact spherically symmetric vacuum solutions to grav-
itational theories derived from a Lagrangian of the form 3- 7 f(R) = R1+8. These

'Often Minkowski space, even though it is not always a solution of the field equations.

1239
1240

theories reduce to GR in the limit 6 ----+ 0 and will be the effective theory at low
curvatures for f(R) = R + a/ Rn. The first solution is 3
2
( )dt2 + (1 - 26 + 4( )(1 - 26(1 + 6)) d 2 + 2d0 2
Solutl"on 1 d S 2 -- -A 11' (1-6)2Bl(1') l' l'

20 (1+28) (1-40) (1-28+48 2 )

where Al(1') = l' (1-8) + C 11'- (1-0) , Bl(1') = 1 + C 11'- (1 8) and C 1 is a
constant. This solution is conformally related to the Q = 0 limit of a solution found
by Chan, Horne and Mann 12 and reduces to the Schwarzschild solution in the limit
6 ----+ O. The second solution is given by

Solution 2 ds 2 = -A2(1')dt2 + a2(t) (1 + ~2 ) 4 A(1')q+2b-l (d1'2 + 1' 2 d0 2)

.? 0 (1+20)
where A2(1') = ((1- C 21'- 1)/(1 + C 2 1'- 1)) q, a(t) = t (1-8), q2 = 1- 26 + 46 2 and
C 2 is a constant. This solution is conformally related to one found by Fonarev 13
and again reduces to the Schwarzschild solution in the limit 6 ----+ O.
These two solutions can be seen to exhibit features not present in the
Schwarzschild solution. Both are strongly curved, but each displays this curvature
in a different way. Solution 1 is static and does not reduce to an l' independent form
in the limit T ----+ 00 (despite the Ricci scalar approaching 0 in this limit). Solution 2
displays more conventional behaviour in the limit l' ----+ 00, but shows explicitly the
lack of validity of Birkhoff's theorem. This solution becomes l' independent in the
limit l' ----+ 00, but still displays strong curvature in this limit as the metric reduces
to a spatially fiat vacuum Friedmann-Robertson-Walker cosmology.3,6,7
Perturbative analyses in the literature are often performed about Minkowski or
de Sitter space. This is perfectly acceptable practise in GR and fourth-order theories
in which an Einstein-Hilbert. term dominates in the low curvature regime. In other
fourth-order theories, of the type considered here, in which the Einstein-Hilbert
term does not dominate the low curvature regime, there is good reason to consider
perturbing about other backgrounds. By considering spherically symmetric pertur-
bations to the backgrounds (1' ----+ 00 limit) of the two space-times given above it can
be seen that there exist extra modes which are not excited in the exact solutions.
However, the modes corresponding to the linearisation of these exact solutions are
the most interesting ones for comparison with gravitational experiments.
Once perturbative solutions have been obtained it is a straightforward calcula-
tion to find the equations of motion (EoM) of test particles following geodesics of
these space-times. Such calculations show that the EoM contain terms due to the
linear perturbations about the background, as well as terms due to the background
itself.
For solution 1 it is found that the EoM of test particles are modified from their
usual form in GR, both in the pre-multiplicative factors of the terms with GR
counterparts, as well as modifications to the powers of l' and the existence of an
entirely new terms which vanishes in the limit 6 ----+ O. The first order corrections
due to small, but non-zero, 6 are primarily due to the new term arising in these
 1241

equations. This term corresponds to an extra force which drops off as T- 1 and can be
used to impose upon 6 the tight constraint 3 6 = 2.7±4.5 X 10- 19 , from observations
of the perihelion precession of Mercury. 14
The EoM for test particles in solution 2 are simpler than those of solution I, but
still differ from those of GR in significant ways. All of the terms except one in the
EoM have GR counterparts, and the powers of T in these terms are all the same as
in the GR case. The pre-multiplicative factors of these terms are, however, modified
and can be described adequately within the frame-work of the parameterised post-
Newtonian approach 15 by assigning f3 = 1 and'Y = 1- 26. Making this identification
allows the constraints on I' from observations of the Shapiro time delay of radio
signals from the Cassini space probe 16 to impose upon 6 the constraint 6 = -l.1 ±
l.2 x 10- 5 . As well as the usual effects associated with 'Y - 1 being non-zero there
are extra effects in this space-time due to the term with no GR counterpart. This
term is proportional to the velocity of test-particle (when v « c) and is zero for
photons. For this reason we identify it as a friction term, with the friction coefficient
being given by a. This 'friction' is a purely gravitational effect and is not due to
any non-gravitational interaction of test particles with any other matter.
These considerations show explicitly that there are a number of different back-
grounds about which one may choose to perform a perturbative expansion, in fourth-
order theories, and that these backgrounds can display behaviour which is not per-
mitted in GR. Moreover, it has been shown that the choice of background is highly
non-trivial when performing a perturbative expansion. The form of the perturba-
tions are background dependent and, correspondingly, so are the geodesics of the
space-times.

References
1. E. Pechlaner and R. Sexl, Comm. Math. Phys. 2, 165 (1966).
2. K. S. Stelle, Phys. Rev. D 16, 953 (1977).
3. T. Clifton and J. D. Barrow, Phys. Rev. D 72, 103005 (2005).
4. J. D. Barrow and T. Clifton, Class. Quant. Grav. 23, Ll (2006).
5. T. Clifton and J. D. Barrow, Class. Quant. Grav. 23, 2951 (2006).
6. S. Carloni, P. K. S. Dunsby, S. Capoziello and A. Troisi, Class. Quant. Grav. 22, 4839
(2005).
7. H. J. Schmidt, Astron. Nachr. 311, 165 (1990).
8. T. Clifton and J. D. Barrow, Phys. Rev. D 72, 123003 (2005).
9. S. Capozziello, F. Occhionero and L. Amendola, Int. J. Math. Phys. D 1, 615 (1993).
10. S. M. Carroll, A. De Felice, V. Duvvuri, D. A. Easson, M. Trodden and M. S. Turner,
Phys. Rev. D 71, 063513 (2005).
11. T. Clifton, Classical and Quantum Gravity 23, 7445 (2006).
12. K. C. K. Chan, J. H. Horne and R. B. Mann, Nucl. Phys. B447, 441 (1995).
13. O. A. Fonarev, Class. Quant. Grav. 12, 1739 (1995).
14. 1. 1. Shapiro, C. C. Counselman III and R. W. King, Phys. Rev. Lett. 36, 555 (1976).
15. C. M. Will, Theory and Experiment in Gravitational Physics (rev. edn.), Cambridge
University Press, Cambridge (1993).
16. B. Bertotti, L. less and P. Tortora, Nature 425, 374 (2003).
 COSMOLOGICAL MODEL WITH A BORN-INFELD TYPE
SCALAR FIELD

R. KERNER*, E. SERIE*·t and A. TROISI", *

* Laboratoire de Physique Theorique de la Matiere Condensee, Universite Pierre-et-Marie-Curie
- CNRS UMR 7600 Tour 22, 4-eme etage, Boite 121, 4, Place Jussieu, 75005 Paris, France
* [email protected]

Laboratoire de Physique TMorique (UMR 8627) Universite Paris XI, Biitiment 210, 91405
Orsay Cedex, Prance

" Dipartimento di Scienze Fisiche Univ. di Napoli "Federico II", Istituto Nazionale di Fisica
Nucleare, sez. di Napoli Compl. Univ. di Monte S. Angelo Ed.N. Via Cinthia 80121 - Napoli,
Italy

Non-abelian generalization of the Born-Infeld non-linear Lagrangian is extended to the

non-commutative geometry of matrices on a manifold. We then analyze some cosmologi-
cal models with this Lagrangian, leading to primordial inflation with varying speed, with
possible early deceleration ruled by the relative strength of the Higgs field.

Keywords: Primordial Cosmology, Born-Infeld Lagrangian, Non-Commutative Geometry.

1. The Non-Commutative Born-Infeld Lagrangian

In currently considered cosmological models with scalar field as a source of pri-

mordial energy and the subsequent inflation there is usually no limit on the field
strength: the scalar field <I> can take on arbitrarily high values. This is true in
particular for the simplest model with V(cjJ) = AcjJ2 (cjJ2 - ,2) "double well" po-
tential. However, one may point out that principles of quantum field theory, when
combined with General Relativity, should lead to the existence of a cut-off for all
fields, because at certain intensity the very notion of space and time will not be
valid anymore. More precisely, when applied to the gravitational field Heisenberg's
uncertainty principle induces a spontaneous birth of black holes, which destroy the
very possibility of measuring any field's intensity. This is why it seems reasonable to
investigate theories in which such a cut-off is incorporated from the very beginning.
In many modern versions of the Kaluza-Klein theories and in relativistic string
models certain generalizations of the Born-Infeld lagrangian appear quite naturally
in variational principles ased on purely geometrical considerations.
The original version of non-linear electrodynamics proposed by Born and Infeld 1
was based on a Lorentz-invariant lagrangian density (when constructed on a pseudo-
Riemannian manifold, it is also invariant under diffeomorphisms), defined as follows:

LBI(g, F) = LBI(g, F)Jl9T = {32 (VI det(gJLv)l- Vi det(gJLv + (3-1 FJLv) I)

= {32 (1 -)1 + ;2 (B2 - E2) - ;4 (E· B)2) Jl9T. (1)

1242
 1243

The constant f3 appears for dimensional reasons and plays the role of the upper
limit of the field strength.
The non-abelian generalizations of Born-Infeld lagrangian were proposed since
the end of the seventies by Hagiwara;2 however, he did not try to find soliton-like
solutions. In 1997 Tseytlin3 argued in favour of the symmetrized trace prescription
which reproduced in the first four orders the string effective action for gauge poten-
tial. Finally Park4 introduced yet another non-abelian generalization and investi-
gated qualitative behavior of instanton-like solutions, considered also by Galtsov et
al. 7 8 In the framework of non-commutative geometry introduced ing 10 all geomet-
rical objects acquire non-commutative values realized as complex N x N matrices.
In 5 and 6 the following Lagrangian was proposed:

where j is a 2 x 2 unitary matrix whose square is equal to -1 2. The field strength

F:v takes its values in the matrix algebra. It can be generalized so as to contain also
the non-commutative exterior forms, which give rise to scalar multiplets, because
the generalized 2-form F contains also the components F:a = Df.l ¢~ and FIb =
Cfd¢~ <I>g - C~b <I>~.

2. Cosmological model with Born-Infeld scalar field

The Lagrangian (2) contains the contribution of two types of fields: the classical
Yang-Mills potential, A = A~eft )..b, corresponding to the usual space-time compo-
nents of the connection one-form, and the scalar multiplet coming from its matrix
components ¢ = ¢aea = ¢~)..bea. In the case when ¢ = 0, this Lagrangian coincides
with the one studied in. 5 For cosmological considerations, we will restrict ourselves
to a qualitative analysis of the case when the space time components of F do vanish
Fftv = 0, leaving only the contribution of scalar multiplet degrees of freedom.
The Lagrangian 1 can be reduced to a form in which it contains only a multiplet
of N 2 scalar fields, and further on, assuming a full invariance under matrix coordi-
nation choice, to one scalar field only. For cosmological purposes, we assume that
the scalar field is a function of time t alone. This yields the following expression for
the variational principle:

J + ~fJ-2
1 ( -V3, + ¢(t)) 2 ¢2(t) } (3)

where the parameter fJ is the analog of the BI cutoff for the non linear electromag-
netic theory while, gives account of the mass-shell of the scalar field and depends
1244

on the non commutative algebra considered for the gauge group. This model con-
tains in itself the ¢4 theory so we expect to obtain a standard behavior when the
Born-Infeld parameter (3 tends to infinity. In presence of gravity, we shall assume
the minimal coupling with the Hilbert-Einstein Lagrangian, which reads:

A =
Jd4 x A 1
mpz2 R + (32 ((
1-
.
1 + Y[¢(t), ¢(t)]
) 1/4
1+ ~IP[¢(t)]) . (4)

The two functions Y[¢(t), ¢(t)] and II[¢(t)] are defined as follows:

Y[¢(t), ¢(t)] = ((1 - (3-2¢2(t)? + ~6 II2 [( ¢(t)]) (5)

II[¢(t)] = (3-1¢(t)(¢(t) - V3,) (6)

The phase space analysis of cosmological models was introduced by Belinskii and
Kalatnikov in. l l Similar analysis of the full dynamical system for the scalar field
¢(t) and the FRW scale factor a(t) was performed by us in. 12 The properties of
the differential system resulting from (4) are ruled by two dimensionless parameters
which are the independent ratios between the three constants appearing in the
Lagrangian: the gravitational constant G, the parameter, determining the mass of
the scalar field, and the cut-off parameter (3.
The most important feature of the model is the fact that the scalar field is
confined by the condition I ¢ 1< (3. Various behaviors can be obtained with different
choices of parameters. They include initial inflation with subsequent decleration and
a possibility of further acceleration. 12

References
1. Born M., Infeld L., Pmc. Roy. Soc. Lond. A 144 425-451 (1934).
2. Hagiwara T., J. Phys. A 143059 (1981)
3. Tseytlin A.A., Nucl. Phys. B 501, 41-52 (1997).
4. Park J.H., Phys. Lett. B 458, 471-476, (1999).
5. Serie E., Masson T., Kerner R., Phys. Rev. D 68, 125003, (2003).
6. Serie E., Masson T., Kerner R., Phys. Rev. D, (2004).
7. Galtsov et al., Phys. Rev. D 65,084007 (2002).
8. Kerner R., Barbosa A.L., Galtsov D.V., Proceedings of XXXVII Karpacz Winter
School edited (American Mathematical Society), editors J. Lukierski and J. Rem-
bielinski, arXiv: preprint hep-th/0l08026.
9. Dubois-Violette M., Kerner R., Madore J., J. Math. Phys. 31, 323, (1990); ibid 31,
331, (1990).
10. Dubois-Violette M., Kerner R., Madore J., Phys.Lett. B B217 ,485 (1989)
11. Belinski V.A., Khalatnikov LM., Grishchuk L.P., Zeldovich Y.B., Phys. Lett. B 155,
232 (1985)
12. Troisi A., Serie E. and Kerner, R., , to appear in the Proceedings of 42nd Karpacz Win-
ter School, Intern. Journ. of Geoometrical Methods in Mod. Phys? Eds. A. Borowiec
and M. Francaviglia; arXIv: gr-qc/0607105
 A TELEPARALLEL REPRESENTATION
OF THE WEYL LAGRANGIAN

DMITRI VASSILIEV
Dept of Mathematics, University College London, Gower Street, London WC1E 6BT, UK
D. [email protected]

The main result of the paper is a new representation of the Weyl Lagrangian (massless
Dirac Lagrangian). As the dynamical variable we use the coframe, i.e. an orthonormal
tetrad of covector fields. We write down a simple Lagrangian - wedge product of axial
torsion with a lightlike element of the coframe - and show that variation of the resulting
action with respect to the coframe produces the Weyl equation. The advantage of our
approach is that it does not require the use of spinors, Pauli matrices or covariant dif-
ferentiation. The only geometric concepts we use are those of a metric, differential form,
wedge product and exterior derivative. Our result assigns a variational meaning to the
tetrad representation of the Weyl equation suggested by J. B. Griffiths and R. A. Newing.

1. Traditional model for the neutrino

Throughout this paper we work on a 4-manifold M equipped with prescribed
Lorentzian metric g.
The accepted mathematical model for a neutrino field is the following linear
partial differential equation on M know as the Weyl equation:

The corresponding Lagrangian is

(2)
Here lJo., ex = 0,1,2,3, are Pauli matrices, ~ is the unknown spinor field, and {\7}
is the covariant derivative with respect to the Levi-Civita connection: {\7}o.~a :=
oo.~a +~IJi3ac( oo.lJi3 bc+{r}i3 a.-ylJ' bc)~b where {r}i3a.,
are Christoffel symbols uniquely
determined by the metric.

2. Teleparallel model for the neutrino

The purpose of our paper is to give an alternative representation of the Weyl equa-
tion (1) and the Weyl Lagrangian (2). To this end, we follow 1 in introducing instead
of the spinor field a different unknown - the so-called cofmme. A coframe is a quartet
of real covector fields {)j, j = 0,1,2,3, satisfying the constraint
. k
9 = Ojk {)J ® {) (3)
where Ojk = oJk := diag(l, -1, -1, -1). In other words, the coframe is a field of
orthonormal bases with orthonormality understood in the Lorentzian sense.
Vle define an affine connection and corresponding covariant derivative 1\71 from
the conditions
(4)

1245
1246

Let us emphasize that we employ holonomic coordinates, so in explicit form condi-

tions (4) read aa 79~ - WI'Y af379~ = 0 giving a system of linear algebraic equations for
the unknown connection coefficients Irl'Yaf3. The connection defined by the system
of equations (4) is called the teleparallel or Weitzenbock connection.
Let l be a nonvanishing real light like teleparallel covector field (l·l = 0, IVll = 0).
Such a covector field can be written down explicitly as l = lj79 j where the lj are real
constants (components of the covector l in the basis 79 j ), not all zero, satisfying

(5)
We define our Lagrangian as

(6)
i j
where d stands for the exterior derivative. Note that Ojk 79 /\ d79 k is the axial (to-
tally antisymmetric) piece of torsion of the teleparallel connection. Let us emphasize
that formula (6) does not explicitly involve connections or covariant derivatives.
The Lagrangian (6) is a rank 4 covariant antisymmetric tensor so it can be viewed
as a 4-form and integrated over the manifold M to give the action S(79 j ,lj) :=
J L( 79 j , lj). Variation with respect to the coframe 79 j subject to the constraint (3)
produces an Euler-Lagrange equation which we write symbolically as

(7)
The explicit form of the field equation (7) is given in Griffiths' and Newing's paper. 1

3. Equivalence of the two models

Let us define the spinor field ~ as the solution of the system of equations

IVI~ = 0,

where IVla~a := aa~a+~af3aC(aaaf3bc+lrlf3 a'Ya'Ybc)e. The above system determines

the spinor field ~ uniquely up to a complex constant factor of modulus 1. This non-
uniqueness is acceptable because we will be substituting ~ into the Weyl equation (1)
and Weyl Lagrangian (2) which are both U(l)-invariant. We will call ~ the spinor
field associated with the coframe 79 j .
The main result of our paper is the following

Theorem 3.1. For any coframe 79 j we have L(79 j ,lj) = ±4LWey1 (O where ~ is the
associated spinor field. The coframe satisfies the field equation {7} if and only if the
associated spinor field satisfies the Weyl equation {1}.

The proof2 of Theorem 3.1 is based on the observation that our Lagrangian (6)
is invariant under the action of a certain class of local (i.e. with variable coefficients)
transformations of the coframe, the class in question being the subgroup B2 of the
Lorentz group. 3 This means that coframes come in equivalence classes (cosets) and
the nature of these co sets is such that they can be identified with spinors.
 1247

4. Discussion
Our Lagrangian (6) has the unusual feature that it depends on a quartet of real pa-
rameters lj which can be chosen arbitrarily as long as they satisfy the condition (5).
This parameter dependence requires an explanation. One possible explanation of
the physical nature of the lj's is sketched out below.
Consider the Lagrangian

(8)
("axial torsion squared"). Putting S(1'J j ) := L(1'J j ) and varying with respect to the
J
coframe 1'Jj subject to the constraint (3) we get an Euler-Lagrange equation which
we write symbolically as

(9)
It turns out that in the case of Minkowski metric one can construct an explicit
solution of (9) as follows. Let {}j be a constant reference coframe and let l "lObe
a constant real lightlike covector; here "constant" means "parallel with respect to
the Levi-Civita connection induced by the Minkowski metric". Of course,

(10)
for some real constants lj satisfying the condition (5). Perform a rigid Lorentz
transformation {}j f--+ iJj = Aj k{}k so that l = w( iJo + iJ3) for some w "I 0, put

o o
cosw(X + x 3)
O
sinw(x O + x 3 )
- sinw(x O + x 3 ) cosw(X O + x 3)
o o
where x j := J iJj . dy (here yc> are arbitrary local coordinates on the mani-
fold l'vl and x j are Minkowskian coordinates on M associated with the constant
coframe iJj), and, finally, set 1'J j := (A-1)jkiJ k . Straightforward calculations show
that the coframe 1'J j is indeed a solution of (9). We call this solution plane wave
with momentum l. Note that for a plane wave Ojk 1'J j 1\ d1'J k = ±2 * land l = lj1'Jj
where the lj are the original constants from (10).
Now consider a perturbation of a plane wave. This perturbation can be the result
of either a) us looking for a wider class of solutions or b) the metric ceasing to be
Minkowski. Application of a formal perturbation argument to the Lagrangian (8)
with 0jk 1'J j 1\ d1'J k =f 2l j * 1'Jj as small parameter gives the Lagrangian (6).

References
1. J. B. Griffiths and R. A. Newing, J. Phys. A: Gen. Phys. 3, 269 (1970).
2. D. Vassiliev, Phys. Rev. D75, 025006 (2007); gr-qc!0604011.
3. A. L. Besse, Einstein manifolds (Springer, 1987).
 NONLINEAR SUPERSYMMETRIC GENERAL RELATIVITY *

K. SHIMA * and M. TSUDA t

Laboratory of Physics, Saitama Institute of Technology,
Fukaya, Saitama 369-0293, Japan
* [email protected]
t [email protected]

We introduce Nonlinear Supersymmetric General Relativity (NLSUSY GR) byextend-

ing the geometrical arguments of Einstein GR on Riemann spacetime to a new spacetime
just inspired NLSUSY. We construct an Einstein-Hilbert (EH)-type NLSUSY GR ac-
tion in the new spacetime and explain spontaneous decays (Big Decay) of spacetime
to the ordinary Riemann spacetime described by the EH action coupled with mass-
less Nambu-Goldstone fermions (supemn matter). Spontaneously broken linear SUSY
(renormalizable interacting) theories which are equivalent to NLSUSY GR in asymp-
totic Riemann-flat spacetime are shown explicitly and the cosmological implications of
NLSUSY GR are discussed.

Keywords: Nonlinear supersymmetry; Nambu-Goldstone fermion; General relativity;

Spontaneously broken linear super symmetry.

1. Basics of Nonlinear Supersyrnrnetric General Relativity

Towards a (spontaneously broken) supersymmetric (SUSY) unified model of space-
time and matter, a new Einstein-Hilbert (EH)-type action in Nonlinear Supersym-
metric General Relativity (NLSUSY GR) was constructed in the superon-graviton
model (SGM) from composite viewpoints.1,2 The EH-type NLSUSY GR action is
defined on a new spacetime (called SCM spacetime), where the tangent spacetime
posesses SL(2,C) Grassmann coordinates 'lj; as local NLSUSY degrees of freedom
(d.o.f) for the coset space, superGL(4,R)jGL(4,R), besides the ordinary SO(3,1)
Minkowski coordinates xa. The Grassmann coordinates 1/J turn subsequently to the
Nambu-Goldstone (NG) fermion (called superon hereafter) dynamical d.oJ. indicat-
ing a spontaneous SUSY breaking (SSB)3 and the NLSUSY GR action describes
NLSUSY gravitational interaction of superon.
Indeed, the NLSUSY GR (N = 1 SGM) action 2 on empty SGM spacetime (of
everything) is defined as follows:

(1)

where

(2)

In Eq.(l) wal" is the (composite) unified vierbein on SGM spacetime, O(w) is the
scalar curvature in terms of w a 1"' and (0, A) are the Newtonian gravitational con-
stant and the (small) cosmological constant. In Eq.(2) eal" is the ordinary vierbein

*Talk given by M. Tsuda.

1248
 1249

and t a p is the mimic analogue for SL(2,C) whose dynamical parallel alignments for
the spacetime orientations of SGM spacetime are encoded in w a p- The K, is the ar-
bitrary constant whose dimension is (length)2 in the Volkov-Akulov (VA) model of
NLSUSY,4 which is now related to (C, A) by 1'£2 = (~:~)-l in NLSUSY GR (SGM)
scenario. The action (1) is invariant under NLSUSY transformations generated by
global spinor parametars ( as
1 - -
oQt/J = -( - if£((,P'ljJ)fJp1jJ, oQe ap = 2if£((,v1jJ)fJ[pe avl, (3)
1'£
which induce the GL( 4,R) transformation of the unified vierbein w a I"' while under
the following local Lorentz transformations 5 with local parameters Cab = C[ab] ex)

(4)
which induce the ordinary local Lorentz transformations of wap- Both Eqs.(3) and
(4) are form closed algebras, respectively.5 These arguments show that the local
Lorentz invariance is realized as a whole in the nonminimal and the highly non-
linear way and that EGR for the empty spacetime is generalized naturally. We
can summarize 5.6 the NLSUSY GR action (1) is invariant at least under the fol-
lowing spacetime symmetries [new NLSUSY]0 [local GL(4, R)]0 [local Lorentz] 0
[local spinor translation (LST)] which is locally isomorophic to super-Poincare(SP),
and the following internal symmetries for N-superons V;i (i = 1,2"" , N) extended
NLSUSY GR, [global SO(N)] 0 [local U(l)N] 0 chiral. Here we just mention that
the superon d.o.f. can be gauged away neither by the ordinary GL( 4,R) transforma-
tions of eap(x) nor by the LST, e.g. o1jJ(x) = ((x), oeap(x) = -if£2(((xhafJp1jJ(x) +
i;;(xhafJp((x)) which is nothing but a translation (redefinition) of the spinor coor-
dinates on SGM spacetime. Therefore, the action (1) is a nontrivial generalization
of the EH action in EGR.
NLSUSY GR is unstable due to the NLSUSY structure of tangent spacetime and
decays spontaneously (called Big Decay of spacetime) to ordinary Riemann space-
time described by the EH action in EGR coupled with massless superon (matter)
1jJ, which is written formally as follows,

S(e,1jJ) = c
167fC
4
Jd4 xlel{R(e) - A + T(e, 1jJ)}, (5)

where T( e, 1jJ) is the kinetic term and the gravitational interaction of superon. The
second and the third terms produce (N-extended) NLSUSY VA action 4 in asymp-
totic Riemann-flat (eap(x) - 7 t5 ap ) spacetime with ",2 = (~:~)-l, which is different
from SGM-flat (wafi(x) - 7 oafi) spacetime for (1).

2. Linearization of Nonlinear Supersymmetry

The linearization of NLSUSY is an interesting problem and is inevitable to obtain
an equivalent local field theory to NLSUSY GR, which is renormalizable and de-
scribes the observed low energy (SM) physics. We discuss these problems and the
cosmological implications of NLSUSY GR in this section.
1250

It is well known that by linearizing NLSUSY the VA model is recasted into

various spontaneously broken linear (L) SUSY theories, 7 which is renormarizable.
Indeed, N = 1 NLSUSY VA model describes a scalar supermultiplet 8 or a U(l) axial
vector one. 9 N = 2 NLSUSY VA model is (algebraically) equivalent to a SU(2) x
U(l) vector supermultiplet. lO
As usual in the linearization of NLSUSY the Fayet-Iliopoulos (FI) pI terms
indicating the SSB with the vev < pI >= ~ are generated automatically. And all
component fields of the LSUSY multiplet can be constructed as the composites of
superons'l/Ji (called SUSY invariant relations 8 ), in such a way as the familiar LSUSY
transformations defined on the elementary supermultiplet are reproduced on the
abovementiond composite supermultiplet in terms of the NLSUSY transformations
on the constituent superons.
Recently, we have found in D = 2 N = 2 that the ordinary LSUSY invariant
mass terms as well as the Yukawa interaction terms vanish identically under the
SUSY invariant relations in the linearization of NLSUSY. (We anticipate that this
is also the case in D = 4, which is yet to be proved.) Therefore, we can add
these terms to a linearized free action without violating the equivalence and the
invariance. l l Once they are added to the free action with the FI term (i.e. thrown
into the true vacuum), they become nontrivial, break SUSY and produce mass for
the composite fields automatically, which depends on the vev
12
< pI >= V~:~e,
as partly realized. This result predicts the spacetime origin in NLSUSY for the
tiny values of neutrino mass and of the energy density of the universe, which are
the direct evidence of SUSY breaking in the spontaneous phase transition of SCM
spacetime (Big Dicay) preceding the Big Bang. 12

References
1. K. Shima, Z. Phys. C18, 25 (1983); European Phys. J. C7, 341 (1999).
2. K. Shima, Phys. Lett. B501, 237 (2001) 237.
3. P. Fayet and J. Iliopoulos, Phys. Lett. B51, 461 (1974); L. O'Raifeartaigh, Nucl. Phys.
B96, 331 (1975).
4. D.V. Volkov and V.P. Akulov, JETP Lett. 16,438 (1972); Phys. Lett. B46, 109 (1973).
5. K. Shima and M. Tsuda, Phys. Lett. B507, 260 (2001).
6. K. Shima and M. Tsuda, Class. Quantum Gmv. 19,5101 (2002).
7. J. Wess and B. Zumino, Phys. Lett. B49, 52 (1974); Nucl. Phys. B78, 1 (1974); P.
Fayet, Nucl. Phys. B1l3,135 (1976).
8. E.A. Ivanov and A.A. Kapustnikov, Relation between linear and nonlinear realizations
of supersymmetry, JINR Dubna Report No. E2-10765, 1977 (unpublished); J. Phys.
All, 2375 (1978); J. Phys. G8, 167 (1982); M. Rocek, Phys. Rev. Lett. 41, 451 (1978);
T. Uematsu and C.K. Zachos, Nucl. Phys. B201 (1982) 250.
9. K. Shima, Y. Tanii and M. Tsuda, Phys. Lett. B525 (2002) 183.
10. K. Shima, Y. Tanii and M. Tsuda, Phys. Lett. B546 (2002) 162.
11. K. Shima and M. Tsuda, hep-th/0611051.
12. K. Shima and M. Tsuda, PoS HEP2005, 011 (2006); hep-th/0611072; See the talk by
K. Shima.
 BLACK HOLE SOLUTIONS IN N >4 GAUSS-BONNET GRAVITY

8.0. ALEXEYEV
Sternberg Astronomical Institute of Lomonosov J'.;[oscow State University,
Universitetsky Prospekt, 13, Moscow 11 9992, Russia
[email protected]

N.N. POPOV
Computer Center of Russian Academy of Sciences,
Vavilova Street, 21, Afoscow 119991, Russia

Gauss-Bonnet gravity provides one of the most promising frameworks to study curva-
ture corrections to the Einstein action in supersymmetric string theories, while avoiding
ghosts and keeping second order field equations. Although Schwarzschild-type solutions
for Gauss-Bonnet black holes have been known for long, the Kerr-Gauss-Bonnet met-
ric was missing. In this paper, a five dimensional Gauss-Bonnet solution is derived for
spinning black holes and the related thermodynamical properties are briefly outlined.
In any attempt to perturbatively quantize gravity as a field theory, higher-
derivative interactions must be included in the action. Such terms also arise in the
effective low-energy action of string theories. Furthermore, higher-derivative gravity
theories are intrinsically attractive as in many cases they display features of renor-
malizability and asymptotic freedom. Among such approaches, Lovelock gravityl is
especially interesting as the resulting equations of motion contain no more than sec-
ond derivatives of the metric, include the self interaction of gravitation, and are free
of ghosts when expanding around flat space. The four-derivative Gauss-Bonnet term
is most probably the dominant correction to the Einstein-Hilbert action 2 when con-
sidering the dimensionally extended Euler densities used in the Lovelock Lagrangian.
The action therefore reads as:

where a is a coupling constant of dimension (length) 2, and G the D-dimensional

Newton's constant defined as G = 1/MP-2 in terms of the fundamental Planck
scale M*. Gauss-Bonnet gravity was shown to exhibit a very rich phenomenology
in cosmology (see, e.g.,3 and references therein), high-energy physics (see, e.g.,4
and references therein) and black hole theory (see, e.g.,5 and references therein). It
also provides interesting solutions to the dark energy problem,6 offers a promising
framework for inflation,7 allows useful modification of the Randall-Sundrum mode1 8
and, of course, solves most divergences associated with the endpoint of the Hawking
evaporation process. 9
Either in D-dimensions or in 4-dimensions with a dilatonic coupling (required
to make the Gauss-Bonnet term dynamical), Gauss-Bonnet black holes and their
rich thermodynamical properties lO have only been studied in the non-spinning (i. e.
Schwarzschild-like) case. Although some general features can be derived in this
framework, it remains mostly unrealistic as both astrophysical black holes and mi-

1251
1252

croscopic black holes possibly formed at colliders l l - 13 are expected to be rotating

(i.e. Kerr-like). Of course, the latter -which should be copiously produced at the
Large H ardon Collider if the Planck scale is in the Te V range as predicted by some
large extra-dimension models 14 - are especially interesting for Gauss-Bonnet gravity
as they could be observed in the high-curvature region of General Relativity and
allow a direct measurement of the related coupling constant. 4 The range of impact
parameters corresponding to the formation of a non-rotating black hole being of zero
measure, the Schwarzschild or Schwarzschild-Gauss-Bonnet solutions are mostly ir-
relevant. This is also of experimental importance as only a few quanta should be
emitted by those light black holes, evading the Gibbons 15 and Page 16 arguments
usually used to neglect the angular momentum of primordial black holes. To in-
vestigate the detailed properties of black holes in Lovelock gravity, it is therefore
mandatory to derive the general solution, i.e the metric for the spinning case. Unlike
the numerical attempts that were presented in 17 for degenerated angular momenta,
the present paper focuses on the exact solution in 5 dimensions.
Einstein equations in Gauss-Bonnet gravity with a cosmological constant A read
as

RJ.LV - ~gJ.LvR = AgJ.Lv + a (~gJ.LV ( RJ.LvQj3RJ.LVQ(3 - Q

4RQ(3R (3 + R2)

- 2RRJ.Lv + 4RWyR~ + 4R"Y8RZ~ - 2RJ.L"Y8)..R~8)"), (2)

and the 5-dimensional metric in the spherically-symmertic Kerr-Schild type can be

written as

ds 2 = dt 2 - dr 2 - (r2 + a 2) sin 2 Bd¢i - (r2 + b2) cos 2 Bd¢~

2
+b ) cos
2
Bd¢~ - p2dB2 - 2dr ( a sin 2 Bd¢l + b cos 2 Bd¢2 )

-I) ( dt - dr - a ,in' Od¢, - b ea,' Odf, )', (3)

where p2 = r2 + a 2 cos e +b 2 sin 2 e and gtt = 1 - (3.

The analytical solution was obtained using (Be) component (analogously to 1S )
and it is:

p2 _ . / p4 _ 4aM - 'd a Ar 2(2 p2 - r2)

(3 = V 3 (4)
2a
Some improvements and developments can be foreseen. First, it should be very
welcome to obtain the same kind of solutions for any number of dimensions. Un-
fortunately the method introduced in this article is not easy to generalize and a
specific study should be made for each case. Then, it would be interesting to com-
pute the greybody factors for those black holes. Following the techniques of,19 it is
 1253

possible (although not strait forward) to obtain a numerical solution as soon as the
metric is known, at least in the A = 0 case. The Kerr-Gauss-Bonnet-(Anti)-de-Sitter
situation is more intricate as the metric is nowhere fiat, requiring a more detailed
investigation, as suggested in. 2o

References
1. D. Lovelock, J. Math. Phys. 12 (1971) 498; J. Math. Phys. 13 (1972) 874.
2. B. Zwiebach, Phys. Lett. B 156 (1985) 315.
3. C. Charmousis & J.-F. Dufaux, Class. Quant. Grav. 19 (2002) 4671.
4. A. Barrau, J. Grain & S.O. Alexeyev, Phys. Lett. B 584 (2004) 114
5. S. Alexeyev & M.V. Pomazanov, Phys. Rev. D 55 (1997) 2110
6. S. Nojiri, S.D. Odintsov & M. Sasaki, Phys. Rev. D 71 (2005) 123509
7. J.E. Lidsey & N.J. Nunes, Phys .Rev. D 67 (2003) 103510
8. J.E. Kim, B. Kyae & H.M. Lee, Nue!. Phys. B 582 (2000) 296
9. S. Alexeyev et al., Class. Quant. Grav. 19 (2002) 4431
10. R.C. Myers & J.Z. Simon, Phys. Rev. D 38 (1988) 2434
11. T. Banks & W. Fisher, hep-th/9906038
12. S. Dimopoulos & G. Landsberg, Phys. Rev. Lett. 87 (2001) 161602
13. S.B. Giddings & S. Thomas, Phys. Rev. D 65 (2002) 056010
14. N. Arkani-Hamed, S. Dimopoulos & G.R. Dvali, Phys. Lett. B 429 (1998) 257
15. G.W. Gibbons, Comm. Math. Phys. 44 (1975) 245
16. D.N. Page, Phys. Rev. D 16 (1977) 2402
17. S. Alexeyex et al., J. Phys. Conf. Ser. 33 (2006) 343
18. R.C. Myers and N.J. Perry, Ann. Phys. 174 (1986) 304.
19. J. Grain, A. Barrau & P. Kanti, Phys. Rev. D 72 (2005) 104016
20. P. Kanti, J. Grain & A. Barrau Phys. Rev. D 71 (2005) 104002
 ELECTROSTATICS AND CONFINEMENT IN EINSTEIN'S
UNIFIED FIELD THEORY

S. ANTOCI
Dipartimento di Fisica "A. Volta" and IPCF of CNR, Pavia, Italy

D-E. LIEBSCHER
Astrophysikalisches Institut Potsdam, Potsdam, Germany

L. MIHICH
Dipartimento di Fisica "A. Volta", Pavia, Italy

Einstein's unified field theory was devised already in 1925 for unifying gravitation
and electromagnetism through the use of a nonsymmetric fundamental tensor gik
and of a nonsymmetric affine connection f kl ; since 1945 it was intensely pursued
by both Einstein and Schrodinger, who assumed it to be a complete field theory,
not allowing for phenomenological source terms. However in 1955 a disappointed
Erwin Schrodinger wrote: "It is a disconcerting situation that ten years endeavour
of competent theorists has not yielded even a plausible glimpse of Coulomb's law."
It occurred to us, through the study of exact solutions found in the meantime, that
physically relevant results instead appear as soon as sources are allowed for, like it
happens with the sources of gravity in the general relativity of 1915. We consider
here the Hermitian version of the theory. Let

(1)

be both Hermitian, i.e. endowed with real symmetric and purely imaginary anti-
symmetric parts. We pose

(2)

where g == det (gik) is real. Then the original field equations without sources read:
gik,l - gnkfJ - gin f l1 = 0, (3)
g lis],8 = °, (4)
R(ik) (f) = 0, (5)
R[[ik],l] (f) = 0, (6)

where Rik(r) is the usual Ricci tensor, Hermitian thanks to (3) and (4).
A class of solutions to these equations was found,l that depend intrinsically on
three coordinates. These solutions show that there is merit in appending sources
at the right-hand sides of all the field equations, while preserving the Hermitian
symmetry through the symmetrised Ricci tensor Rik (r) of Borchsenius,2 that is
equal to Rik (f) wherever the sources vanish. Works by Lichnerowicz on the Cauchy
problem 3 and by Hely4 show that the Bianchi identities look physically meaningful 5

1254
 1255

when the metric is chosen to be a symmetric tensor Sik such that

(7)
il . .
where S Ski = 5L and S == det (Sik), and sources are appended at the nght-hand
sides of (3)-(6), in the form of a symmetric energy tensor Tik, and of two conserved
currents 41fji = g[i~~ and 81f Kikl = R[[ik],l]' Then the Bianchi identities read:

T is
;8
-
-
~2 S lk (oiR-. K.
J [hI + ,k8g
lSi]) . (8)

The fundamental tensor for the general electrostatic solution 6 reads

-~0 -~0 -1~~)

9ik =
( -a -b -c d
c ,where: (9)

d = 1 + a 2 + b2 + c,
2
a= .
zX,x, b ·C ·
= ZX,Y' = ZX,Z, X,xx + X,YY + X,zz = 0 . (10)

The imaginary part 9[ik] looks like the general electrostatic solution of Maxwell's
theory, because the "potential" X must obey Laplace's equation, and

g[i~~ = 0, 9[[ikJ,l] = 0,
happen to be satisfied. The squared interval reads:

If only one point charge h is present in the "Bildraum" x, y, Z, t, say, at the origin
of the coordinates, the "potential" X is

(12)

For this occurrence to happen, a net charge must appear at the right-hand side
of the equation g[i~~ = O. The surface dX = 0, d = 0 is the inner border of the
manifold. In the metric sense, this border is a point with a spherically symmetric
neighbourhood.
A solution for n point charges at equilibrium can be built 6 by considering, in
the "Bildraum", n closed surfaces possessing net charges, whose charge distribution
reproduces the one occurring on n conductors at rest, due to their mutual induc-
tion. By changing the shapes and the positions of the surfaces in order to obtain
that both dX = 0 and d = 0 on each of them, one gets, in the metric sense, n
point charges whose infinitesimal neighbourhood is spherically symmetric. When
this occurs, even approximately, the position of the charges mimics the equilibrium
positions prescribed by Coulomb's law. 6
1256

Axially symmetric solutions allowing for n charged poles at the right-hand side
of equation R[[ikJ,IJ = 0 are easily found.1 In cylindrical coordinates Xl = r, .1: 2 = Z,
x 3 = cp, x4 = t one writes the fundamental tensor:

gik = (~~ =~ ~ ~)
o 0 -7 1
(13)

The square of its line element reads:

ds 2 = sikdxidxk = gr (-dr 2 - dz 2 - r 2dcp2 + dt 2 ) + ~(d1j;)2. (15)

g
A static solution with n aligned poles built with Kikl is found by requiring:
n
"i _ _ ' " K 1 Pq +Z - Zq (16)
<!J- ~ qn r '
q=l

Kq and are constants. Let us consider the particular case when n = 3, and
Zq
Kl = K3 = K, K2 = -K, with Zl < Z2 < Z3·
The charges are always pointlike in the metric sense; moreover, with the choice
shown above, the metric happens to be spherically symmetric severally in the in-
finitesimal neighbourhood of each of the charges. 7
If chosen in this way, the three "magnetic" charges are always in equilibrium,
like it would happen if they would interact mutually with forces independent of dis-
tance. The same conclusion was already drawn by Treder in 1957 from approximate
calculations,8 while looking for electromagnetism in the theory. In 1980 Treder rein-
terpreted 9 his result as accounting for the confinement of quarks: in the Hermitian
theory two "magnetic" poles with unlike signs are confined entities, because they
are permanently bound by central forces of constant strength.

References
1. S. Antoci, Ann. Phys. (Leipzig), 44, 297 (1987); also: gr-qc!0l08042.
2. K. Borchsenius, Nuovo Cimento, 46A, 403 (1978).
3. A. Lichnerowicz, J. Rat. Meeh. Anal., 3, 487 (1954).
4. J. Hely, Comptes Rend. Acad. Sci. (Paris), 239, 385, 747 (1954).
5. S. Antoci, Gen. ReI. Gmv., 23, 47 (1991); also: gr-qc!0108052.
6. S. Antoci, D.-E. Liebscher, L. Mihich, Gen. Rei. Gmv., 37, 1191 (2005).
7. S. Antoci, D.-E. Liebscher, L. Mihich, gr-qc!0604003.
8. H.-J. Treder, Ann. Phys. (Leipzig), 19,369, (1957).
9. H.-J. Treder, Ann. Phys. (Leipzig), 37, 250 (1980).
 GALACTIC DISKS IN THEORIES WITH YUKAWIAN
GRAVITATIONAL POTENTIAL

J. C. N. de ARAUJO* and O. D. MIRANDAt

Divisao de Astmfisica - Instituto Nacional de Pesquisas Espaciais
Avenida dos Astmnautas 1758 - Sao Jose dos Campos - 12227-010 SP - Brazil
* [email protected]
t [email protected]

We present. a new solution for the rotation curves of galactic disks wit.h gravitational
potential of the Yukawa type. In particular, to obtain a galactic disk with a scale length
of b ~ 10 kpc. we should have a massive graviton of mg < < 10- 59 g. This result is much
more restrictive than those inferred from solar system constraints.

1. Introduction

A first motivation of the present paper is to present a mathematical tool that can be
used, independently of the particular theory of gravity, for the study of the rotation
curves of galaxies. In this case, there is only one constrain: the potential must be
Yukawa-like. Thus, for a point mass m we have

¢ = _ Gm e-r/A, (1)
r
where G is the gravitational constant. Note that, for ). ---+ 00 this potential becomes
identical to the Newtonian potential.
The field equation reads

(\72- :2) ¢ = -47fGp, (2)

where the constant). is the Compton wavelength of the exchange particle of mass
mg. Using this interpretation for the present case one can think of the exchange
particle is a massive graviton. In particular, the second motivation of the present
paper is to verify if it is possible to constrain the mass (mg) of this particle.

2. The rotation curve for a Yukawian gravitational potential

Toomre 1 showed that the potential of a thin disk can be written as follows

¢(r, z) = 27fG 10= Jo(kr)S(k)e-klzldk, (3)

where S(k) IS related to the surface density, tt(r), through the following Bessel
integral

f.L(r) = rco Jo(kr)S(k)kdk

.fo
where S(k) = 10= Jo(ku.)Jl.(u)udu, (4)

comes from the Fourier-Bessel integral theorem.

Once f.L(r) is given one can obtain the potential of the disk, and from the
centrifugal-equilibrium condition, namely,

1257
1258

g(r) =
2
v (r) = _ (8¢>,) (5)
r 8r z=O

one obtains the rotation curve of the disk. Freeman2 applied the above equations to
obtain the Newtonian rotation curve for an exponential disk, whose surface density,
/-L(r) , in cylindrical coordinates is given by /-L(r) = /-Loe- r / b , where b is the scale
length of the disk.
In particular, we apply the same technique to obtain the rotation curve for the
Yukawian potential. 3 After some manipulation and an eventual change of variables
the rotation curve reads

(6)

In Fig. 1 we compare the Newtonian rotation curve with different Yukawian

rotation curves. We plot the velocity, in terms of "jG /-Lo b, which has dimension of
velocity, versus the distance to the center of the disk in units of the scale length, b.
The Newtonian rotation curve has a maximum velocity around r /b '" 2, and for
larger values of r the velocity decreases monotonically. This is a well known result.
On the other hand, the behavior of the Yukawian rotation curves is the following:
the velocity rises up to a maximum value and then decreases; the greater the ratio
"b / A" , the lower the maximum velocity is. The size of the disk is strongly dependent
on the value of the ratio "b/ A". The greater the ratio "b/ A", the smaller the disk
is. As expected for b/ A « 1 the rotation curve becomes Newtonian.
These behaviors strongly constrain the value of A. As a result, for theories where
A is the Compton wavelength, strong constrains can be imposed in the putative mass
of the graviton. To the Yukawian rotation curve be approximately a Newtonian one,
we need b/ A « 1. Recall that the Compton wavelength, Ac , of a particle of mass
rng reads Ac = h/rngc, where c is the velocity of light and h is the Planck constant.
For b = 10 kpc we should have rng « 1O-59 g .

3. Conclusions
We present a new solution for the rotation curves of a thin disk for a Yukawian
gravitational potential. This thin disk can represent the disks present in spiral and
SO galaxies. It is important to stress that the mathematical tool developed here can
be used to study rotation curves in any model or theories of gravity which produce
a gravitational potential like the Yukawa one.
The behaviors of the Yukawian rotation curves strongly constrain the value of A.
As a result, for theories where A is the Compton wavelength of a exchange particle,
say a graviton, strong constraints can be imposed in its putative mass.
 1259

a = bI A
1.6

1.4

1.2
Newtonian
/ a = 0.01
1.0 I
~
:0- 0.8
=f:
~
~ 0.6
'>
a =1
0.4

0.2

0.0
0 2 3 4 5 6 7 8 9 10

rib

Fig. 1. Comparison between the Newtonian rotation curve with different Yukawian rotation
curves parameterized in terms of "b / A" .

It is worth stressing that the best bound on the graviton mass from planetary
motion surveys is obtained by using Kepler's third law to compare the orbits of
Earth and Mars, yielding mg < 1O-54 g .4 With such a mass and for b = 10 kpc one
would obtain that b/)... ~ 10 5 , which would imply that galactic discs could not exit
(see Fig. 1) for a Yukawian gravitational potential with this value of mg.

Acknowledgments
JCNA would like to thank the Brazilian agencies CNPq and FAPESP for partial
support.

References
1. A. Toomre, Astrophys. J. 138, 385 (1963).
2. K.C. Freeman, Astrophys. J. 160, 811 (1970).
3. J.C.N. de Araujo and O.D. Miranda, Gen. Relativ. Gravit., submitted for publication
(2006).
4. S.L. Larson, and W.A. Hiscock, Phys. Rev. D 61, 104008 (2000).
 ON THE FIELD THEORETIC DESCRIPTION OF GRAVITATION

THEO M. NIEUWENHUIZEN
Institute for Theoretical Physics, University of Amsterdam
Valckenierstraat 65, 1018 XE Amsterdam, The Netherlands
[email protected]

Maxwell started to describe gravitation as a field in Minkowski space. Such an approach

brought Babak and Grishchuk in 1999 the gravitational energy-momentum tensor. Simple
manipulations allow the Einstein equations to take the form Al'v = (87rG / c 4 )8I'V, where
A is the acceleration tensor and 8, the total energy momentum tensor, is conserved
in Minkowski space. For non-curved cosmology, the total energy density vanishes: the
gravitational part is negative and cancels the one of matter due to the Friedman equation.
In 1989 Logunov added to the Lagrangian a bimetric coupling between the Minkowski
and Riemann metrics. The case of a positive cosmological constant allows a regularization
of the interior of the Schwarzschild black hole: The components of the metric tensor donot
change sign, so, inside the black hole, time keeps its usual role. Cosmology generalizes
the ACDM model, while gravitational and matter energy cancel again at all times.

Keywords: gravitational energy momentum tensor, total energy of matter and gravita-
tion, bimetric theory, cosmology, black hole, grey hole

1. The gravitational energy momentum tensor in general relativity

Let the Minkowski metric 7JJ1.V = diag(l, -1, -1, -1) be denoted by 'J1.V = (r-l )J1.V
in arbitrary coodinates, and let hJ1.V be the gravitational tensor field. The Riemann
metric gJ1.V = (g-l)J1.v is then connected by kJ1.V == ,J1.V + hJ1.v = /gl, gJ1.v.
Define the acceleration tensor AJ1.V = ~[kJ1.vka;3 -kJ1. a k vj1 ]:a:j1, where the: denote
covariant differentiation in Minkowski space. The Riemann tensor and the Einstein
tensor GJ1.V = RJ1.V - ~gJ1.V R do not change when one replaces partial derivatives by
covariant derivatives in Minkowski space. 3 In TJ1.V == (c 4187rG) [(r 19 )AJ1.V - GJ1.v], all
second order covariant derivatives of gJ1.V or, equivalently, of hJ1.V, cancel, leaving
4
TJ1.V = c 'VI
167fG 9
[ hlt.V... ,hAP _ hJ1.A hVP
A:P . ':A .. :p
+ hJ1.A:Ph v. A:P _ hAP:J1.hV
. A:P
_ hJ1.A:Ph .. :v
AP

lowering (raising) of indices of the h a(3:, being performed with gJ1.V (gJ1.V). In Carte-
sian coodinates, T coincides with the Landau-Lifshitz quasi-tensor. 1 In general,
tJ1.V = (g I, )TJ1.V coincides with the Babak and Grishchuk result,2 obtained in two
ways independent of ours. Matter can be included simpler than in .2 Inserting the
Einstein equation GJ1.V = 87rGc- 4T'W in AJ1.V = (gl,)(GJ1.V + 87fGC 4T J1.V) , one gets

(1)

8J1.V is the total energy momentum tensor of gravitation and matter. It is conserved
in Minkowski space, 8J1.v: v = 0, since AJ1.v: v = 0 because covariant Minkowski

1260
 1261

derivatives commute. This approach allows, e.g., a faitful description of the energy
exchange between a gravitational wave and a detector, or the definition of the four-
J
momentum of a stationary localized object, pf.l = d 3 xy!="1 8f.l 0 .
For Robertson-Walker cosmology ds 2 = U(t)c 2dt 2 - V(t)[dr2 / (1 - kr 2) +r 2 d0 2 ]
with k = 0, U = 1, V = a 2 (t), it holds that ADO = 0, so also 8 00 = 0: the total energy
vanishes, the gravitational energy density rv _0,2/ a 2 cancels the one of matter, p. In
Minkowski space, galaxies are fixed at their coordinates, while the gravitational field
changes in time. 4 In Riemann space, increase of aCt) is interpreted as expansion.

2. The Relativistic Theory of Gravitation (RTG)

Rosen replaces partial derivatives by covariant ones in Minkowski space,

which makes G a tensor. Discarding a total derivative, he expresses the Hilbert-

Einstein Lagrangian as LR = (c4/167fG)gf.lV(G~vG~0" - G~O"G~)..), which involves
first order derivatives only. Logunov and coworkers continue on this. In the action
5 g = J d 4 xAL g they consider the Lagrangian 4 ,5

(3)

PA relates to a cosmological constant, Po is a shift of the zero level of energy. The

bimetric term Pbi couples the Minkowski and the Riemann metrics. It breaks general
coordinate invariance, causing a big difference with GR and bringing an analogy
to massive electrodynamics. Logunov then imposes the relation PA = Pbi = Po,
so in the absence of matter, cplW = 0, gf.lV = ,f.lV and also Lg = 0. His choice
°
PA == -m 2 c4 /(327fG) < leads to an inverse length m and, in quantum language,
a graviton mass fim/(c,f2), but this theory has drawbacks. We consider a positive
cosmo Ioglca
4
. 1 cons t ant , PA = 8ITG
c A = 8ITG
2
3c Hv,O
n H20 = 8ITG'
3c
2
°
74( 9.78Gyr
0.71 )2 . Now Pbi >

0, so the graviton has an "imaginary mass": Gravitational waves are unstable at

today's Hubble scale. This is of no concern: On that scale, not single gravitational
waves but the whole Universe comes in, and it is unstable (expanding) anyhow.
The Einstein equations involve Tto~ = Tf.l V + PAgf.l V + Pbi/pO" (gf.lPgO"v _ ~gf.lV gPO").
Conservation Tto~:v = 0, imposes, due to the Pbi terms, the only allowed gauge: the
harmonic gauge (/!iF; gf.lV):v = 0, equivalent to the Lorentz gauge hf.l :v = 0.
V
The Pbi-tenns are only of Hubble size, and have no effect on solar or galaxial
scale, or for double pulsars. 6 Their role is restricted to non-small gravitational fields,
e.g., near singularities and in cosmology.
Black holes. The line element ds 2 = U(r)c 2dt 2 - V(r)dr 2 - W2(r)d0 2 describes
a Schwarzschild black hole for Us = I/Vs = (r - rh)/(T + rh), Ws = r + rh, with
rh = GM/c 2. We scale r --+ TTh, and define U = eU, V = eV , W = 2rhe w . The
dimensionless small parameter is very small, fl == ,j2A rh = 2.3810- 23 NI/ NI8 . For
1262

small p, a parametric solution of the source free Einstein equations reads

1 + 71(el; + ~ + log 71 + ro)
r= u=~+log71, (4)
1 - 71( el; + ~ + log 71 + ro) ,
v = ~ - In 71 - 210g(e/; + 1), w = rye/; + p2(~ + log 71 + wo).
where ~ is the running variable and 71 == ap2 is a small scale. In the interior,
u'(r) = {exp[2a(r - 1)]- 71 2 - p2}/(2ry). Energy momentum conservation implies,
as in GR, (p + p)u' + 2p' = O. Since p == 0 is needed to have all mass at the origin,
but p(O) =J 0, energy conservation imposes u'(O) = 0, so a = I log pi, 71 = p2/llnpl·
The solution coincides with Schwarzschild's for ~ » 1. In the regime ~ = 0(1),
there is a transition towards the interior ~ « -1. Both U = rye/; and V = e/; /71 are
very small there, but, contrary to the Schwarzschild solution, they remain positive:
In the interior of our black hole, time keeps its normal role. The transition layer
~ = 0(1) acts like 't Hooft's brick wall, having a very small width, fI* == TJTh ~
10- 9(M/M8)3f1Planck. In general, the Pbi term may regularize singularities of GR.
Spherical shells of matter fall in with speed close to c, despite the claims of .4,5
In Logunov's situation, one has Pbi < 0, so in ry > 0 one gets a < O. Then the
condition u'(O) = 0 has no solution, depriving that theory of a proper black hole.
This condition can neither be obeyed in GR: If the central mass is slightly smeared,
the Schwarzschild black hole cannot obey energy-momentum conservation in GR.
Cosmology. The Lorentz gauge now brings two relations: U(t) rv V 3 (t) and
k = 0. 4 Putting U = a 6 (t), V = a~a2(t) and going from cosmic time t to conformal
time T = J a 3 dt, gives the Einstein equations in a familiar shape,
Ii 4nG Ac 2 1
- =-- (p + 3p) + - (1 - -). (5)
a 3c2 3 a6
For a » 1 and a~a » 1, this coincides with the ACDM model. At large T, there
is the exponential growth a( T) rv exp H OCJ T with HCXJ = /lc/ v'6. In cosmic time this
reads aCt) rv [Hoo(t o - t)]-1/3, where to is "the end of time". For details, see .7

References
1. L. D. Landau and E. M. Lifshitz, The Classical Theory of Fields, (Pergamon, Oxford,
U.K., 1951; revised 1979).
2. S. V. Babak and L. P. Grishchuk, Phys. Rev. D61, 024038 (1999).
3. N. Rosen, Phys. Rev. 57, 147 (1940); ibid 150; Ann. Phys. 22, 11 (1963).
4. A.A. Logunov, The Theory of Gravity, (Nauka, Moscow, 2001). gr-qc/0210005.
5. 8.8. Gershtein, A.A. Logunov and M.A.Mestvirishvili, gr-qc/0602029.
6. C.M. Will, Theory and experiment in 9ravitational physics, (Cambridge Univ. Press,
New York, 1993), chapter 12.3, rules out a more recent bimetric theory of Rosen.
7. Th.M. Nieuwenhuizen, unpublished.
 DE SITTER STABILITY IN THEORIES WITH SECOND ORDER
CURVATURE TERMS

A.Y. TOPORENSKY* and P.Y. TRETYAKOyt

Sternberg Astronomical Institute, Universitetsky prospect, 13, Moscow 119992, Russia
* [email protected]
t [email protected]

Quadratic curvature corrections to Einstein-Hilbert action lead in general to higher-

order equations of motion, which can induce instability of some unperturbed solutions of
General Relativity. We compare conditions for stability of de Sitter cosmological solution
for FRW and Bianchi I metrics.

Recently theories of higher-order gravity have become a matter of intensive in-

vestigations. Some of them are motivated by string theory,l other proposals have
been put forward in order to explain present accelerated expansion of the Uni-
verse. 2 ,3 One of the main features of these theories is the fact that they generally
lead to equations of motion which contain higher-order derivatives. As the number
of time derivatives in the equations of motions increases in comparison with the
situation in classical General relativity, the dimensionality of the phase space of the
corresponding dynamical system increases as well. This may result in a dynamical
behavior qualitatively different from that known in GR: some important GR so-
lutions can loose their stability. Though higher-derivative terms may be small on
classical solution, perturbations from higher-order corrections can grow in direction
of "additional" dimensions of the phase space, and a point which is stable in GR
may become unstable in higher-order theory. In the present communication we study
stability of de Sitter solution with respect to second order curvature corrections in
their general form.
A general form of second order curvature corrections is

82 = J R(aR ik1m Riklm + fJRik Rik + ,R2) d4 x. (1)

However, in 3 + 1 dimensions the Gauss-Bonnet term

G B = Riklm Riklm - 4Rik Rik + R2
does not contribute to equations of motion, and, then, introducing B = fJ + 4a and
C = , - a we can write the following action for the theory with a cosmological
constant term

8 = J R(R + BRik Rik + CR 2 - A) d 4 x. (2)

From the action (2) we obtain the following analog of Friedmann equation (we
denote D = 12B + 36C):
6H 2 -A+2DH2(H +H2) -3DH4_D(H +H2)2+2DH(H +3HH +H 3 ) = 0, (3)

where H == ~m is the Hubble parameter.

1263
1264

Note that the Friedmann equations, being algebraic in GR, becomes a differential
one in the theory with second order curvature corrections when D i- O. In the D = 0
case the Friedmann equation does not change at all, and we exclude this case from
the further analysis.
For stability studies this equation should be written in the form of a system of
two first order equations. The form of de Sitter solution (H = Ho = canst) give us
the equations for corresponding stable points:

~Ho
3
+ - -1( A ~ 6Ho2 + 2DHo)
4
= 0,
2DHo

its solution is Ho = ±/"{. It is remarkable, that this de Sitter solution is exactly the
same as in the pure GR, so higher-order corrections in the theory under investigation
do not shift the location of the fixed point. However, its stability can be affected by
the higher-order terms.
After the standard procedure we get the condition for stability of the de Sitter
solution in the form D = 12(0 + f3 + 31 ) = 12(B + 3C) > O. If, instead of the form
(2), we express the corrections through the Weil tensor and the curvature scalar, this
condition become even simpler: the coefficient before the curvature scalar should be
positive. 4
The situation become less simple if we allow anisotropic perturbations of the de
Sitter solution, which means that the Weil tensor has a non-zero contribution to
the equations of motion .
Consider a flat homogeneous anisotropic Universe with the Bianchi I metrics in
the form

(4)
.It should be noted that in the presence of general quadratic curvature corrections
the diagonal form of the metrics (4) is not the general one. 5 We restrict ourself by
this special form for simplicity. We show that even in this less general case we have
additional restrictions for stability conditions of de Sitter solution.
Starting from the action (2) we have three equations of motion. After evaluating
the corresponding system of equation of motion, we find the following eigenvalues
(details can be found in, 4 see als0 6 )

P,1,2,3 = ~3Ho, (5)

~18CHo ~ 6BHo ± 2V81C2 H;5 + 54BCH;5 + 9B2H;5 ~ 6C ~ 2B

IL4,5 = 2(6C + 2B) (6)

~3BHo ± V33B2H;5 + 4B + 96BCH;5

P,6,7,8,9 = 2B (7)
In expanding Universe we have always P,1,2,3 < O. The condition 114,5 < 0 leads to
the already known result B + 3C > O. However, we have now additional restrictions
arising from the condition P,6,7,8,9 < O. It leads to AB + 4CA + 1 < 0 if B > 0
 1265

and AIBI - 4CA - I < 0 if B < O. Note that these stability conditions now depend
not only on coefficients before the higher-order terms in the action (2), but also on
the value of the cosmological constant A. These additional restrictions arise from
instability with respect to anisotropic perturbations of the de Sitter metrics. It is
interesting that independently of A the whole quadrant B > 0, C > 0, which always
satisfies the stability conditions in a FRW Universe, is excluded in anisotropic case
(see Fig.l)

Fig. 1. Zones of de Sitter stability for the action (2) (shaded) The de Sitter solutions in zone
below the line 1 are unstable on isotropic background, the solutions in unstable zone above the
lines 1 and 2 are unstable due to anisotropic perturbations.

References
1. R. Metsaev and A.Tseytlin, Phys. Lett. BI85, 52 (1987).
2. S.Nojiri, S.Odintsov and M.Sami, Phys. Rev. D74 046004 (2006)
3. G.Cognola, E.Elizade, S.Nojiri, S.Odintsov and S.Zerbini, Phys. Rev. D73 084007
(2006)
4. A.Toporensky and P.Tretyakov, gr-qc/0611068.
5. H.-J. Schmidt, "Lectures on Mathematical Cosmology", gr-qc/0407095.
6. J.Barrow and S.Hervik, gr-qc/0610013.
 BASIC RELATIONS OF A UNIFIED THEORY OF
ELECTRODYNAMICS, QUANTUM MECHANICS,
AND GRAVITATION*

OSTERMANN, PETER
[email protected]

A first consistent variational principle of electrodynamics, quantum mechanics,

and gravitation yields the basic equations of a unified scalar-vector-tensor theory
which deals with extended particles of variable shape. It is compatible to solid prin-
ciples of classical physics and, in particular, to special and general relativity. There is
a direct way leading to a natural quantization of Maxwell's equations, to the Klein-
Gordon equation, or to Einstein's equations including a quantum energy-momentum
tensor of matter now. As an exemplification, Planck's fundamental energy-frequency
relation is derived for transitions of charged particles in an external electric field
and verified by calculating the energy integral of the H-atom.
A) Notation - Besides the well-known Landau-Lifshitz 1 notation there are used
some self-explanatory abbreviations and the natural constants e, c, Ii, '" == 8wG / c 4 .
Capital italic letters like in Ai will denote macroscopically smoothed quantities in
contrast to their microscopic counterparts like in at--.
K is a particle index, whereas
K implies an immediate summation over all particles except for K. A tilde like in
8i K == SiK + eK/c· at.. indicates a combination of components. mK,eK,/-LK,PK are
rest mass rna, charge eo, and their densities /-Lo, Po with respect to particle K.
B) The key to a necessary extension of electrodynamics - Because of the local
relativity of simultaneity some prototypes of quantum mechanics such as particles
in a box, rotator, or oscillator turn out to be dynamic paradoxa 2 of special rela-
tivity theory (SRT). To avoid contradictions, it seems necessary to presuppose an
energy-momentum postulate with respect to arbitrarily chosen inertial systems

(1)

i.e. the conservation laws

fhT/ = O. (2)
Therefore, neglecting internal gravitation, particles together with their carrier of
interaction have to fulfill (2) within freely falling local frames. Since any objects are
composed of charged particles, each physical process is subject to electrodynamics.
At a closer look*, inserting the conventional energy-momentum tensor

Tk - ( - F il Fkl
i (conventional) = + 4"1 ili F rl Frl) + /-LOC 2Ui Uk , (3)

phenomenologically combined of two independent parts so far, equation (2) implies

a 'classical' relation between current density and four-potential

* The original pdf-sheets of this talk (15 pages) are available from author's website, http://peter-
ostermann. de as well as the detailed e-print "Principles and aspects of an open theory of electro-
dynamics, gravitation, quantum mechanics" (115 pages, in German, 2 tables, 10 figures).

1266
 1267

JI := POUI Jo _~Q2
moc
(0 15+ eoc AI) (4)

as the key to a necessary extension of electrodynamics. The derivation of (4) is

based on the simplest ansatz JL
Po := eo~ == eoQ2, (5)
m,o
where a scalar Q is defined, which in its microscopic form qK determines the shape
of particle K. This new shape scalar immediately refers to quantum mechanics.
C) A jir-st consistent variational principle of electrodynamics, quantum mechan-
ics, and gravitation - The basic relations of the unified scalar-vector-tensor theory
are derived from the variational principle

15 J( (j5 + 'L)I>K)F9df? =
K
0 (6)

which is composed of the action densities

,
cJ>K
1 rl
== 4" fK
k 1
frl - 2mK qK
2 (I
sK
eK I)
+ --;;-ak (K
sl
eK k) 1 mKc 2 2
+ --;;-a1 +2 n2 I K
qK - 2mK qKql (7)

of each particle K, wherein qF, sF are the partial derivatives O[qK, O[SK. The index
R (non-K) is defined by ..K == L .. (v) , (8)
v#K
and Einstein's overall gravitational action density is
}i;._ G -
2fl:':¥ .- == 9 urn 9 8V 9 rw (rv,ur r w,ms -
r v,um r w,sr ) . (9)
The expression (7) assigned to a single particle K includes a familiar looking scalar
of the electromagnetic field which, however, means field products of different parti-
cles only. This prevents from any self energy or renormalization problems, though
the actual reason for this approach is the impossibility otherwise to get a consistent
energy-momentum tensor satisfying the conservation laws (2). Modified correspond-
ingly, a new model of electromagnetic waves seems compatible to photons now.
D) The quantized Maxwell equations -- While the 1st pair is fulfilled as an iden-
tity by definition, fi~;1 + fk~i + fl~k == 0, the variation of (6) with respect to all
microscopic electromagnetic potentials al yields the 2nd pair in the form

(10)

where the electromagnetic potential of the respective particle K does not appear
itself, but the potentials of all other particles R instead.
E) The covariant real wave equation - Variation of (6) with respect to the shape
scalar qK yields
2
1t qf;1 = qK [(s1< + ak) (S[K + e: a/<) - m~c2] . e: (11)

It is this variable shape of extended particles (and all fields) that implies indeter-
minism. In the formal approximation It ~ 0, however, equation (11) reduces to that
of Hamilton-Jacobi for charged mass points in a predefined electromagnetic field.
1268

F) Continuity equation for rest mass and charge - Variation of (6) with respect
to the microscopic action scalar SK results in the covariant continuity equation
jk.;l = O. (12)
This continuity equation holds not only for the charge density but for the rest-mass
density, too, since the latter is presupposed to be directly proportional to the charge
density in (5). It is of decisive importance that (12) also results from (10).
G) The Einstein equations with a quantum energy-momentum tensor of matter
- Variation of (6) with respect to the fundamental tensor gik yields
1
Eik == Rik - - Rgik = ,,"Tik, (13)
2
where the quantum tensor of matter and non-gravitational fields

Tik == LTi~ (14)

K
is composed of
Ti~ = (15)
2 2
1 ( K Kl
--2 fil fk
1 2 _ K _K
+ fklK fi Kl) + -qKsi n K K 1 [rl K n ( qK2);1]
sk + - q i qk + -4 gik fK frl - - ·1 .
mK mK mK '
This derivation of Einstein's equations includes a consistent completion by a quan-
tum energy-momentum tensor now. Its classical approximation - implying the
'geodesic' law of motion - has no longer to be added phenomenologically only.
H) Derivation of the complex Klein-Gordon equation - A mere substitution into
(11), (12) of qK == J 7/JK7/J'K , SK == _ i; In (7/JKN'K) (16)
yields the Klein-Gordon equation

( ZnUI -
'::J eK
--;:azK) [ r-::.
-g (.zno
v I
- eK z)]
--;:a K 7/JK = 2 c2
mK 7/JK (17)

in a covariant form. The complex quantum scalar 7/JK is normalized by the natural
requirement that the integral of j~ equals eK now. Only this unconventional nor-
maliziation proves consistent with the energy-frequency relation for photons, too.
l) The energy-frequency relation - Thus, for example, using (14) and the basic
relations above, a straightforward calculation of the energy of the H-atom results in
E(stationar y )
(H)= = cev
+ mpc,
2
(18)
where 00 may indicate a fixed proton here. Therefore, what has been a mysterious
quantum mechanical parameter cev so far, actually turns out to be the true energy
integral of the bound electron in stationary states (rie = 0). With w/w the well-
known oscillation frequency of the charge distribution during the transition tL ---> lJ,
now IcelL -Cev I = ftw/ w is derived as an energy quantum of radiation, i.e. of a photon.

References
1. Landau L. D., Lifschitz E. M.: Lehrbuch der theoretischen Physik; Berlin 1992
2. Ostermann, P.: Phys.u.D. 1 (1985) 23-37 (available from author's website, too)
 PHYSICAL INTERPRETATION AND VIABILITY OF VARIOUS
METRIC NONLINEAR GRAVITY THEORIES

LESZEK M. SOKOLOWSKI
Astronomical Observatory, Jagellonian University, Orla 171, 30-244 Krakow, Poland
[email protected]

We discuss physical foundations of metric gravity theories considered as gravi-

tational alternatives to dark energy models in cosmology. We investigate a generic
metric nonlinear gravity (NLG) theory with a Lagrangian L = f(gJ1l/' RJ1l/Q(3) being
any scalar function of C 2 class. A popular idea underlying the use of NLG theories
to account for the accelerating universe is following. Let L = R + a/ R + bR 2 and
Robertson-Walker (RW) spacetime. In the early universe, when the curvature was
large, the R2 term was dominant generating some kind of inflation, for most time
of the galactic era the scalar R had intermediate values so that the linear term in L
was leading preserving all the successes of the standard cosmology while at present
R is small and the R- 1 term is dominant giving hopefully rise to the accelerated
expansion. This argument is misleading for three reasons.
Firstly, the corrections 1/ R+ R2 (and any other ones) generate fourth-order field
equations, i.e. greatly modify the dynamical structure of the theory by introducing
new qualitatively different classes of solutions. As an example take a harmonic
oscillator perturbed by a third derivative term,
E i +i + w 2 x = 0
with lEI « 1. It has two solutions corresponding to the periodic unperturbed solu-
tions which are slowly damped or amplified by the factor exp(w 2 Et/2) and a third
non-analytic at E = 0 solution exp(-t/E).
Secondly, the Lagrangian is not a physical observable and there is no direct
relationship between value and variability of various terms in it and properties of
solutions of the field equations. It is a great success of the standard Friedmann
cosmology that irrespectively of the form of the equation of state for the cosmic
fluid (assuming that the strong energy condition holds) all solutions start from the
initial singularity, so that there must have been the small, dense and hot early
universe and then monotonically expand (for the spatially flat RW model, k = 0).
This uniqueness is lost if nonlinear terms appear in L: new classes of solutions exist
and if one finds a solution which well fits the observational data (deceleration in
the past and acceleration at present epoch) a problem arises of why this particular
solution rather than any other describes the real universe.
Thirdly, all NLG theories are solely applied to and tested on the spatially flat
RW metric. This spacetime is so simple that each NLG theory admits it as a solu-
tion, i.e. for each L there is a class of solutions a( t) for the cosmic scale factor in
the vacuum case. And conversely, for any cosmic evolution a( t) there are infinity of
Lagrangians f (g J1l/' RJ1 l/Q (3 ) having it as a solution. Thus the fact that a given spe-
cific L fits in this sense the cosmological data tells almost nothing about physical

1269
1270

correctness of the theory.

Before applying a given NLG theory to cosmology one should first establish its
physical features and check whether it satisfies general rules of classical field theory.
These features should be established for the vacuum case and only after checking
that it is a tenable theory one may study interactions of gravity with matter sources.
The first step is to determine the particle spectrum of the theory. It was found long
ago by studying the Cauchy problem that a generic NLG theory has eight physical
degrees of freedom. To find out the particle content of these degrees of freedom one
views the tensor field g!-'v in L = f(g!-'v, R!-'va(3) as a kind of unifying field mixing
various particles (fields) with different spins and thus admits that g!-'v need not be
the metric tensor of the physical world. To decompose g!-'v into a multiplet of fields
with definite masses and spins one applies appropriate Legendre transformations
(in classical mechnics these correspond to transformations from a tangent bundle to
a cotangent bundle to the spacetime). This formalism is as general and powerful as
the canonical one in mechanics and classical field theory; we stress that it involves
no ad hoc tricks and uses no "auxiliary" fields-what is popular in recent papers on
NLG theories in cosmology. In general one finds a gravitational triplet consisting of
a massless graviton (interpreted as a spacetime metric), a massive spin-2 field and
a massive scalar field. The crucial point is that both the initial unifying Lagrangian
f and the resulting triplet of fields do not determine the true metric of the physical
world since all the fields may be subject to further Legendre transformations. Which
frame is physical should be determined by experiment. Each triple forms a "frame"
of variables and in each frame the corresponding equations of motion are of second
order and the graviton in any frame satisfies Einstein field equations. Thus an
NLG theory is not an inherently higher derivative theory, this fact together with
dynamical equivalence and physical inequivalence of different frames gives rise to a
lot of misconceptions in the current literature.
We focus our attention on simplest NLG theories, L = f(R). In their case the
unifying field has three degrees of freedom and is a doublet of the graviton and a
scalar. Different frames for the doublet are related by degenerate Legendre trans-
formations being merely conformal mappings of the graviton. These (and general
NLG) theories have multiple ground state solutions (vacua) being Minkowski, de
Sitter or anti-de Sitter spacetimes. A necessary condition of viability of a theory
is that at least one of these vacuum states is stable. Stability of a vacuum state is
independent of which frame (Jordan, Einstein or any other) is regarded as physical.
There is a subtle problem of relationship between dynamical stability of a solu-
tion and positivity of total (ADM) energy. We investigate stability of the vacua in
Einstein frame employing the methods developed in general relativity; according to
these a vacuum state is stable if the matter source satisfies the dominant energy
condition (DEC). We prove that the scalar component of gravity satisfies DEC if
f(R) satisfies a differential inequality at the vacuum-or the potential of the scalar
has a nonnegative minimum at this state. From this point of view we investigate a
number of L = f(R) functions appearing in recent literature.
 1271

Examples.
1) Odd powers of R- 1 , for n = 0 the case most frequently studied,
f.l4n+4
L=R~ R2n+l'
n = 0,1,2, .... There are two possible vacua: de Sitter space which is classically
unstable and anti-de Sitter space which is a stable ground state for the theory (this
state was not studied in the literature).
2) Even powers of R- 1 ,
f.l 4n + 2
L = R~ R2n '
n = 1,2, .... The only possible vacuum is de Sitter space which turns out unstable,
thus the theory is unphysical.
3)
2
L=R+~
R·
This theory has no maximally symmetric spacetime as a solution (for f.l # 0) and
must be rejected on these grounds.
4)
f.l6
L = R+ R2.
Anti-de Sitter space is the only vacuum state and it is classically stable, thus the
theory is tenable. The same holds for
f.l 4n + 2
L=R+--
R2n
for any natural n.
This page intentionally left blank
Alternative Theories (B)
This page intentionally left blank
 ARE ACTIVE AND PASSIVE ELECTRIC CHARGES EQUAL?

CLAUS LAMMERZABL
ZARM, University of Bremen, Am Falturm, 28359 Bremen, Germany
[email protected]

ALFREDO MACIAS
Dept. de Fysica, Universidad Aut6noma Metropolitana Iztapalapa, A.P. 55534, C.P. 09340,
Mexico, D.F., Mexico
[email protected]. mx

BOLGER MULLER
Physics Department, Stanford University, Stanford, CA 94305-4060, USA
[email protected]

We explore the consequences of a hypothetical difference between active charges, which

generate electric fields, and passive charges, which respond to them. A confrontation
to experiments using atoms, molecules, or macroscopic matter yields limits on their
fractional difference at levels down to 10- 21 .
In electrodynamics, one may distinguish between two types of charge: The active
charge qa is the source of the electric field, V . E = 47rqa O(x) , whereas the passive
charge qp reacts to it, mJi; = qpE. Here, mi is the (inertial) mass and x the location
of the particle. In gravitational physics, a hypothetical difference between active and
passive mass has been considered and confronted with laboratory and astrophysical
observations. However, as yet nothing similar has been done for the electric and
magnetic analogue, in spite of a long history of precision experiments that includes
tests of the 11r Coulomb potential, searches for an electrostatic fifth force,1,2 the
photon mass,3 and violations of Lorentz invariance. 4 Any inequality would thus
have serious consequences throughout science, from the standard model of particle
physics to practical applications like precision metrology. We show 5 that limits as
low as 10- 21 can be derived for protons and electrons by introducing the concepts
of active and passive neutrality.
The dynamics of two particles located at Xl,2 in their mutual electric fields are
described by the equations
..
mliXl = qlpq2a 1
X2 - Xl
1
X2 - Xl 3
+ qlp E()
Xl,

..
m2i X 2 = q2pQla 1Xl
Xl -
-
X2
1
X2 3
+ qlp E()
X2, (1)

where E denotes an external electric field, and qlp, qla, q2p, q2a are the respective
passive and active charges. We introduce the center of mass and relative coordinate
where Mi = mli + m2i is the total inertial mass. We find
.. Qlpq2p X 1
X = ~C2llxl3 + Mi (qlp + q2p) E, (2)

where C 21 = q2a -
Q2p
!ll..;!..
Qlp
Thus, if active and passive charges are different, the center

1275
1276

of mass shows a self-acceleration along the direction of x, in addition to the accel-

eration caused by the external field E. The dynamics of the relative coordinate is
given by

(3)

where
- m1i q1a
D 21---
+ rn2i q2a _
----
q1a + m2i C
- 21 (4)
Mi q1p Mi q2p q1p Mi

and mred is the reduced mass of the system. In the standard framework, D21 = 1.
The analogous case of active and passive masses has first been discussed by
Bondi. 6 The relative acceleration between two masses has been measured in a lab-
oratory experiment,7 the self-acceleration of the center of mass has been tested by
LLR. 8
Contrary to the gravitational case, electric charges can have different signs.
Therefore, we can define active neutrality q1a + q2a = 0 as well as passive neutrality
q1p + q2p = O. This allows us to find alternative tests of the equality of active and
passive charges: An actively neutral system may not be passively neutral and vice
versa. Both definitions of neutrality are compatible, but a system can be actively
and passively neutral at the same time only if C 21 = O. Therefore, a self-acceleration
of the center of mass occurs if and only if the system possesses a nonzero total active
or passive charge.
In order to interpret tests of the neutrality of atoms and molecules as tests of the
equality of active and passive charge, we study compound particles that are actively
or passively neutral. The motion of the center of mass of an atom in a constant
external electric field is given by Eq. (2). If we assume that the total passive charge
vanishes, q1p + Q2p = 0, there will be no acceleration due to the external field E
and C 21 = (Q2a + Q1a) / Q2p' As already explained, the average self-acceleration of
the center of mass will vanish. The difference between active and passive charges
is now related to the neutrality of the composed system with respect to the active
charges: a passively neutral system may still generate an electric field according to
rI-.(x) = Q1a + Q2a _ Q1a + Q2a + ~C Q2p (5)
'I' Ix - xli Ix - x21 - Ixl ... ~ 211xl
(where the dots denote dipole and higher order multipole contributions that are
neglected here). On the other hand, an actively neutral atom in an external electric
field feels a force

(6)

which vanishes if the ratios of active and passive charges are the same for all bodies.
Thus, we can distinguish two types of tests of neutrality: (i) Tests of active neutral-
ity, which measure the electric monopole field created by a passively neutral system,
and (ii) tests of passive neutrality, which measure the force imposed by an external
field onto an actively neutral system. The neutrality of atoms can be tested at levels
 1277

down to 10- 21 elementary charges. 9 If there are no unfortituous cancellations, we

conclude [C21 [ ::;. 10- 21 .
A further limit can be derived from atomic spectroscopy. Therefore, we study
the shift of atomic transition frequencies due to an inequality of active and passive
charges. Although the center of mass motion of the two-particle system cannot be
quantized in general, the relative motion can. The Hamiltonian is
2
H = _P- + D21 q1pq2p . (7)
2m re d Ix[
The energy levels for a single electron atom are proportional to a modified fine
structure constant
_
0:12 -
q1pq2 p D 12 _
-
1
-Q1pq2p
(q1a +
-
m2i
-
C21 ) . (8)
lie lie Q1p Mi
Thus, the spacing between the energy levels will in general deviate from the ex-
pected scaling with the charges. Therefore, a comparison of the energy levels of,
e.g., Hydrogen and ionized Helium He+. For Hydrogen, we have one proton Q1 = Qp
and one electron Q2 = Qe; for He+, we have Q1 = 2Qp and Q2 = Qe. Then for the
transition frequencies lJH and lJHe+, we have

(9)

If the He+ spectroscopy reaches an accuracy at the 10- 13 level and no discrepancy
like Eq. (9) is found, we could deduce a limit of [C21 [ ::;. 4 x 10- 16 . This limit would
not be as precise as the ones derived from bulk matter experiments, however it
would be a particularly clean one.
The above analysis can be extended to active and passive magnetic moments. 5
We like to thank S. Chu, H. Dittus, F.vV. Hehl, K. Jungmann, and e.S. Unnikr-
ishnan for fruitful discussions. Financial support from the German Space Agency
DLR, and CONACYT Grant 42191-F is acknowledged. H.M. thanks the Alexander
von Humboldt Foundation for their support.

References
1. E. Fischbach and C.L. Talmadge, The Search for Non Newtonian Gravity (SpringerVer-
lag, New York, 1999).
2. G. Spavieri at al., Metrologia 41, S159 (2004).
3. L.-C. Th, J. Luo, and G.T. Gillies, Rep. Prog. Phys. 68, 77 (2005).
4. G. Amelino-Camelia et al. In: A.Macias, C. Lammerzahl, and D. Nunez (eds), Gravita-
tion and Cosmology, page 30. AlP Conference Proceedings 758, Melville, N.Y. (2005).
5. C. Lammerzahl, A. Macias, and H. Miiller, Phys. Rev. A 75, 052104 (2007).
6. H. Bondi, Rev. Mod. Phys. 29,423 (1957).
7. L.B. Kreuzer, Phys. Rev. 169, 1007 (1968).
8. D.F. Bartlett and D. van Buren, Phys. Rev. Lett. 57, 21 (1986).
9. C.S. Unnikrishnan and G.T. Gillies, Metrologia 41, S125 (2004).
 CHARGED FLUID DYNAMICS IN SCALAR-TENSOR THEORIES
OF GRAVITY WITH TORSION

CHIH-HUNG WANG
Physics Department, Lancaster University, Lancaster, LAl 4 YE, UK
c. [email protected]
[email protected]

In scalar-tensor theories of gravity with torsion, the gravitational field is described

in terms of a symmetric metric tensor g, a metric-compatible connection \7 with torsion,
and a scalar field 'P. The main aim is to explore an interaction of a charged perfect fluid
and a scalar field 'P in a background electromagnetic and gravitational field described
by {g, \7, 'P}. The interaction is based on an action functional Se of a charged perfect
fluid that is invariant under global conformal rescalings. Using a variational principle, we
obtain equations of motion for the charged perfect fluid. Moreover, we verify that these
equations of motion are equivalent to the gauge identities obtained from the invariance
of an action functional under spacetime dffeomorphisms and a local Uri) gauge group.

1. Action Functionals of Charged Perfect Fluids

In Brans-Dicke theory 1 a direct interaction between the Brans-Dicke scalar field and
matter is often said to violate the weak equivalence principle. 7 However, as Dirac 4
has shown one may consider gravity in a Weyl geometry which naturally induces
such interactions. But little exploration has been made of gravitational scalar field
interactions in a Cartan geometry. Dereli and Tucker 3 have noted that the motion
of an electrically neutral spinless particle in the background geometry of Dirac's
theory can also be reformulated in terms of an autoparallel of a Cartan geometry
with torsion determined by the gradient of the Brans-Dicke scalar field. 2 In this
note, we examine the motion of a relativistic charged fluid in a background Cartan
geometry determined by a scalar field. We shall motivate a particular coupling of
such a field to the fluid and derive the equations of motion from an action principle.
In general relativity (GR), the theory of a relativistic perfect fluid with internal
energy c( n), number density n, 4-velocity field V composed of particles ofrest mass
m and electric charge q can be determined from the action

SCR = fM AcR(ea,A;V,n, Al,A2)

= fM(-mn(I+E(n))*I+QnAI\*V+Al d (n*V)+A2(VI\*V+*I)),

where the background metric of spacetime is 9 = e a ® e a and the background

Maxwell field is F = dA. Latin indices run over 0 to 3. The signature is (-, +, +, +)
and c = 1. The Lagrange multiplier fields AI, A2 ensure the conservation equation
d (n * V) = 0 and normalization of 4-velocity g(V, V) = -=-1. We use; in ACR to
separate background fields e a , A from the fluid's variables. V denotes the metric dual
of V, i.e. V = g(V, -), and * is the Hodge map. For any I-form a, = g-l(oo, -), a
where g-1 is the inverse of g. In a non-Riemannian Cart an background with specified
metric 9 and torsion forms T a , we suppose that the gravitational scalar 'P couples

1278
 1279

to the relativistic fluid to modify the action to

a
Se = iM Ac(e , A, 'P; V, n, AI, A2)

= iM(-X'P n (1+c(n))*1+ Q nAI'I*V+A l d(n*V)+A2(VA*V+*1)),

where X is a coupling constant. We have motivated the action by demanding that

Sel'l'=""- = SCR and that it be invariant under the following global rescalings
x

A.=A, <p = n- l 'P,

n= n- 3
n, E(n) = c,

where D = constant. Here, the conformal weights of V and n are obtained by requir-
°
ing that d (n* V) = and g(V, V) = -1 remain invariant under D transformations.
Furthermore, we assume that X and Q are also invariant under n transformations.

2. Equations of Motion for the Charged Perfect Fluid

By varying Se with respect to V, n, AI, and A2, we obtain

Qn * A - n * d Al + 2A2 * V = 0, (1)
-X 'P f * 1 + Q A A *V - dAI A *V = 0, (2)
d(n* V) = 0, (3)
V A*V +*1 = 0, (4)
where f = 1 + c(n) + n g~ is the index of the fluid. 5 By solving for Al and A2, we
can derive the equations of motion for V and n by substituting them into the rest
of equations. There is no obvious way to know which two equations are best used
for solving Al and A2 since they both appear in the 3-form equation (1). Using (1),
(2) and (4), we can first solve for A2 to obtain A2 = ~X'Pnf. Then, we eliminate
Al from (1) and take the interior derivative iv to get
(5)

where the solution of A2 has been used and F = dA. Because of the identity
°
iv ivO == for any p-form 0, one recognizes that the I-form equation (5) has only 3
independent scalar equations. We consider the 3 independent equations (5) with (3)
and (4) as equations of motion for V and n. It is interesting to notice that equations
(5), (3) and (4) do not involve any direct connection couplings. We can express (5)
in terms of the torsion free Levi-Civita connection ~ as

(6)

where p = n 2 g~ is the pressure per mass and IIv is a projection operator defined by
IIv == 1 + V ® V. From (6), we find that 'P has a similar effect as the pressure per
mass p acting on the motion of the fluid. Moreover, a neutral dust-like (q = p = 0)
1280

fluid does not follow geodesic flow due to the scalar field effect. If we consider the
background torsion field Ta = e a /\ 3L,2 (6) becomes
'P

nxcp f\lvV + qnivF + XcpIIvV p = 0, (7)

where the non-Riemannian connection involves cpo From (7), it follows that the
integral curves of V for a neutral dust-like fluid are autoparallels of \l rather than
V.
Invariance of Se under spacetime diffeomorphism, local SO(3,1), and U(I) gauge
transformations yield the gauge identities 6

DTa + Tc /\ ixaTc + Sb c /\ iXaRbc + j /\ iXaF ~ p(ixadcp) == 0, (8)

bIb b _
DSa ~2(Ta/\e ~T /\e a ) =0, (9)
dj == 0, (10)
where 3-forms Ta , Sa b, j, and 4-form p can be recognized as source currents of e a ,
connection I-forms Wab with torsion, F, and cp respectively. D is a covariant exterior
derivative with respect to Wab and Rb c are the associated curvature 2-forms. From
Se, these source currents become

Ta =xcp(Va (n(I+E)+p)*V+p*e a ),
j=qn*V,
p= ~xn(I+E)*I,

and Sab = 0. 6 By substituting these source currents into (8), (9) and (10), we find
that (3) is equivalent to (10) and using the first structure equation

(11)
and (4), Eq. (5) becomes Eq. (8). Eq. (9) is automatically satisfied.
I am very grateful to Robin Tucker and David Burton for helpful discussions
during my PhD study and to Faculty of Science and Technology and Physics De-
partment at Lancaster University for a travel grant to attend MG 11.

References
1. Brans, C. & Dicke, R. R., Phys. Rev. 124,925, (1961).
2. Dereli, T. & Tucker, R. W., Phys. Letts. B110, 206, (1982).
3. Dereli, T. & Tucker, R. W., Mod. Phys. Lett. A 17, 421-428, (2002).
4. Dirac, P. A. M., Proc. Roy. Soc. Land. A 333, 419, (1973).
5. Lichnerowicz, A. Relativisitic Hydrodynamics and Magnetohydrodynamics (Benjamin,
New York, 1967).
6. Wang, C. R., Ph.D. Thesis, Lancaster, 2006 (unpublished).
7. Will, C. M. Living Rev. Relativ. 9, 3, (2006).
 VALIDATION OF THE WEAK EQUIVALENCE PRINCIPLE IN A
SPATIALLY-VSL GRAVITATION MODEL

JAN BROEKAERT
Philosophy of Physics Group, University of Oxford (temp. affil.)
8 eLEA-FUND, Vrije Universiteit Brussel
[email protected]

The WEP is proven to hold in a scalar-vector gravitation model with spatially vari-
able speed of light and Lorentz-Poincare type interpretation for the configuration of a
non-stationary source. 2 In this model the spatially-VSL, c(r) = c'q,2 + W, is a coor-
dinate velocity in contrast to the invariant velocity c', which is locally observable due
to the application of the gravitationally modified Lorentz Transformations for space
and time intervals 4 The Hamiltonian mechanics for particles and photons derived from
the corresponding energy and momentum transformations agrees at least till I-PN with
the dynamics of General Relativity Theory.5 Now in a local inertial frame and for a
non-stationary source the validity of the WEP is proven by deriving the nil relative
acceleration of a locally coincident free-falling particle. This is established by adapting
the derivative to incorporate the conventional procedure of localizing its arguments by
means of pamllel tmnsport. This transport over a free-fall trajectory can be naturally
expressed using a composition of gravitationally modified Lorentz transformations.

Keywords: Scalar-vector gravitation, Weak Equivalence Principle.

1. The Lorentz-Poincare model with spatial VSL

The LP gravitational model renders explicit the effects of gravitational length short-
ening and time dilation proper to the Lorentz-Poincare interpretation of Special Rel-
ativity. This interpretation of SRT is "dynamically" based; the Lorentz-covariance
of the electromagnetic interaction of the micro-constituents in rods and clocks re-
sults in the appearance of Minkowski "spacetime" .1,7 The LP interpretation distin-
guishes gravitationally affected observers -denoted S'- and gravitationally unaf-
fected "observations", similar to "unrenormalized" and "renormalized" coordinates
which are distinguished in field approaches. 8,9,12,14 The unaffected perspective -
here in the case with non-stationary source denoted Sw- corresponds to the co-
ordinate space description in GRT. The gravitational effects on space and time
observations were developed as gravitationally modified Lorentz Transformations
(GMLT) for space and time intervals. It was also shown that the elimination of
the unaffected perspective from the GMLT between two local observers restores the
local Lorentz covariance of the relations even if in Sw the spatial-variable speed of
light is present. 3 ,5 In the present case of a non-stationary source the fundamental
time-space transformations, conveying the LP properties, are composed of a gravita-
tionally modified Lorentz transformation, A, and a Galilean transformation, G; i.e.
S' <--- A---+ SO <--- G ---+ Sw. The GMLT now involve the vector potential field w - the
induced velocity -- caused by source movement 5 and, the Galilean transformation
in coordinate space expresses the effective velocity in the dynamics:

lit' )
( ox' = A(uo, r) (otoXoo ) /'0<1> -uoc02/'0<1>
,A == ( -UO/'O<1>-l 1<1>-1 + llO:gOj (/'0 - 1)<1>-1
)
(1)

1281
1282

where iP = iP(r) is related to the scalar part of the gravitational field, Co = c'iP 2
is the coordinate spatially variable speed of light and TO = (1 - u6/ C6)-1/2 the
kinematical affecting factor. The additional Galilean transformation expresses a
velocity translation by the field w on the (5to, 5xo)-frame;

Ji o )
( 5xo = G(w) (5tw)
5xw
,G -= (1 0) -w 1
(2)

We will now verify the Weak Equivalence Principle in this model by showing that
a free-falling observer crossed by some free-falling system with a non-zero relative
velocity, has a zero relative acceleration only while meeting at the intersection point
of their trajectories (see also 2 ).

2. Relative acceleration in the LP model

In order to calculate the acceleration the observer requires the initial and final
velocity over an infinitesimal time interval. Let the observer measure the final -
local- value of the particle's velocity at the intersection of their free-fall trajectories.
Then the initial velocity, an instance dt prior to intersection, should be rendered
local to the intersection instance by parallel transport (e.g., 10 sec. 6.1).

a'(PT) -= dt'lim
..... O (V'/I local - v'il l oca I) /dt' (3)

The transport frame coincides with the free-falling particle itself; at initial space
and time instance 1 the Sw space and time intervals {dtl' dXI} are transformed to
the free-falling particle's frame with velocity VI which evolves to space and time
instance 2. The affected intervals are given by A(VOI' l)G(wt)(dtl' dxt). Should
the Weak Equivalence Principle be applicable then during free-fall these affected
quantities remain invariant and, at final instance 2 the inverse transformation
G- I (W2)A -1 (V02, 2) gives the transported intervals in coordinate perspective of Sw,
with V2 = VI +a(l, vI)dt and a(r, v) is the (iP, w)-dependent free-fall acceleration
in coordinate space. In the unaffected perspective the time intervals are all taken
identical; then following eq. (3) the observed relative acceleration is given by;

,k r A(U02,2)G(2)\, ((1,V2)1" G- I (2)A-l(2)A(1)G(1)1"0" (l,Vt)O") 1

a (PT) = dt~~O A(U02, 2)G(2)OT (1, V2)T - G-l(2)A-l(2)A(1)G(1)Tw (1, Vl)w dt'
We will prove this relative acceleration is zero, when the transporting frame is free-
falling and irrespective the velocity U2 of the observer. The latter is easily observed
since the vanishing of the term brackets is independent of A(U02, 2) and expresses
the fact that all locally coincident LIF's will observe the zero relative acceleration
of any coinciding free-falling particle.
Using the explicit form of the transformations A and G we observe that the remote
velocity VI is transformed proportional to the local value V2;
 1283

Straightforward inspection of the relative acceleration expression shows that the

proportionality factor is cancelled in the final fraction: the bracketed term turns
out identically zero. Notice that this property will only be achieved when the initial
and final velocities are related by free-fall acceleration and as such can the transfor-
mation be identified as "parallel" transport. The transported quantities do therefor
remain constant during free-fall and we conclude that the relative acceleration of a
free-falling particle in the local coordinates of the observer's LIF-frame -and all
coinciding LIF frames- is zero:

a/CPT) = O. (5)
The Weak Equivalence Principle is endorsed in the LP model also for the configu-
ration of a non-stationary source.

Acknowledgments
This work was supported by FWO-Vlaanderen project F6/15-VC. A87.

References
1. Bell J. S., Speakable and Unspeakable in Quantum Mechanics, Cambridge U. P., 1987
2. Broekaert J., A Lorentz-Poincare type interpretation of the Weak Equivalence Prin-
ciple., gr-qc/0604107 (to appear in Int. Journ. Phys.)
3. Broekaert J., A Lorentz-Poincare-Type Interpretation of Relativistic Gravitation,
Proc. ESA SP-605, (ed.) M. Cruise, 2005 a, gr-qc/0510017 (to appear)
4. Broekaert J., On a modified-Lorentz-transformation based gravity model confirming
basic GRT experiments, Foundations of Physics, 35, 5, 839, 2005, gr-qc/0309023.
5. Broekaert J., A spatially-VSL gravity model with 1-PN limit of GRT, conference paper
GR17, 18-23 July 2004, Dublin, Arxiv ref: gr-qc/0405015 (submitted) 2004 b
6. Broekaert J., Verification of the 'essential' GRT experiments in a scalar Lorentz-
covariant gravitation, in PIRT VIII proc., (ed.) Duffy M.C., PD Pub., 37, 2002.
7. Brown H. R., Physical Relativity, Space-time structure from a dynamical perspective,
Oxford U.P., 2005.
8. Dehnen H., Honl H., West pfahl K., Ein heuristischer Zugang zur allgemeinen Rela-
tivitatstheorie, Annalen der Physik, 7, 6, 370, 1960.
9. Dicke R.H., Gravitation without a Principle of Equivalence, Reviews of Modern
Physics, 29, 363, 1957.
10. Kenyon I.R., General Relativity, Oxford U.P., 1990
11. Stephani H., Relativity. An Introduction to Special an General Relativity (Third edi-
tion), Cambridge U. P., 2004
12. Thirring W.E., An Alternative Approach to the Theory of Gravitation, Annals of
Physics, 16, 96, 1961.
13. Weinberg S., Gravitation and Cosmology. Principles and applications of the General
Theory of Relativity, Wiley, London, 1972.
14. Wilson H.A., An Electromagnetic Theory of Gravitation, Phys. Review, 17, 54, 1921
This page intentionally left blank
Higher Dimensional Theories
This page intentionally left blank
 EXACT SOLUTIONS OF THE 5D SPACE-TIME-MATTER
UNIVERSE AND THEIR IMPLICATIONS

TAKAO FUKUI
Dokkyo University, 1-1 Gakuencho, Soka, Saitama, 340-0042, Japan
* [email protected]

What can the exact solutions of the 5D STM Universe imply when they are compared
to the field equations of a Corrected Metric Tensor Universe? The comparison implies
the possibility of clarifying the meaning of the cosmological term.

Keywords: 5D STM theory, 4D CMT theory, Cosmological term

1. Introduction

The 5D STM theory predicts that the cosmological term is a possibly variable mea-
sure of the properties of the vacuum in N ?: 5D theories. l Nonseparable solutions
of a vacuum 5D STM Universe are obtained in the Robertson-Walker metric,2 and
are used for showing the prediction.
The cosmological term is explored in the 4D corrected metric tensor theory 3
which has been studied since 1960s by taking acccount of the quantum corrections, 4
and is reproduced either in a constsnt or in a variable.
We study in the present work whether or not we can elucidate the meaning of
the term by comparing the solutions to the field equations with the cosmological
term obtained from the corrected metric tensor theory.
The 5D STM and the 4D CMT Universes enable us to study an origin of the
term in the 4D Universe from a geometrical consideration of a higher dimensional
Universe.

2. 5D STM Universe
The field equations of the 5D STM theory are taken by analogy with the 4D Einstein
equations to be given by

RAE = 0 (A, B = 0,123,4). (1)

Then the 5D field equations can be splitted into the terms proper to the 4D and to
the other,

5DCafJ = 4DCafJ +5thD Extra terms = 0 (00,(3 = 0, 123), (2)

5DCa5 = 0, 5DC55 = O. (3)

In the 5D STM theory, the geometrical properties of the 5th coordinate are regarded
as the physical properties of the 4D Universe, that is

4DCafJ = _5thD Extra terms =4D Physical properties. (4)

1287
1288

The field equations are applied to the following metric dB where the RW metric ds
is characterized by adding the fifth coordinate and by a gauge factor 0"0,

(5)

where E = ±l.
Solutions of Eq.(l) with the RW metric Eq.(5), for the scale factor a(t, l) and
for the scale factor b( t, l) of the fifth dimention are

2 (2Ft-g)2+K,
a = - 4F ' (6)

and

b = _ 4wof jFt 2 - 2gF2t + ggF - wof j(g2 + K,)

(7)
4afF2 '

where F(l) == WOf2 + k, f = f(l) and g = g(l). K, is a constant with the physical
dimensions of (length) 2 .
The conventional 4D Universe is embedded in the 5D manifold and is on the
hypersurface where b = 0 is satsified. That is, from the relations j == df / dl = 0 and
g == dg / dl = 0, I is fixed on the hypersurface. Therefore according to the present
scenario, I should not be variable in the 4D Universe.

3. Corrected Metric Tensor Universe

The Lagrangian density of the 4D CMT theory reads as follows,

(8)

where An is a constant, Lm the Lagrangian density of matter and n runs over 0,1
and 2. It seems to be more complicated than that of Einstein's theory, but the case
of n = 2 which concerns with the renormalization program of the quantum theory,
is not studied here because it does not contribute to the cosmological term. 3 Here
the cosmological term is defined as follows,

(9)

The field equations are obtained by applying the variational principle to the La-
grangian density Eq.(8) as follows,

4D
G a (3 - Aga (3 + ,\ n Extra terms = -87rG
4 - Ta (3. (10)
C

This field equations are employed for obtaining the RW model of the Universe
which metric ds is shown in Eq.(5) above.
 1289

4. 5D STM and 4D CMT

From Eqs.(4) and (10),

(11)

where p = (r - 1)pc2 . The case An = 0, a vanishing cosmological term, is studied

by Fukui et a1. 2 Eq.(l1) can be compared with another variable cosmological term
which is a function of a cosmological quantity, as of the Hubble parameter. 5
As is easily observed especially in the case of 3, - 4 = 0 a radiation era, the LHS
of Eq. (11) directly contributes to an appearance of the cosmological term. In general,
Eq.(l1) shows that the 5th dimension plays a big role both in a material contents
and in the variable cosmological terrrl. That is, the origin of the cosmological term
might be traced to a geometrical property of the fifth dimension.

5. Comments
In the present work, we show the role of the 5th dimension to understand the
origin of the cosmological term which has remained to be seen in the conventional
4D Universe. In this sense, geometrical properties of higher dimensions, in general,
might give a clue to elucidate physical properties unresolved in the 4D Universe.
Recent report that new Hubble Space Telescope discoveries remain consistent
with a cosmological constant,6 supports A = A(l) modeV because in the model, A
is constant in the 4D Universe, on a hypersurface with a fixed l. Then how to cope
with the cosmological constant problem?
As far as we assume a new physical property to understand an unresolved prop-
erty of physics, a final goal will not come into view and a hierachical consideration
will continue further within the limit of our recognition. That is one of the reasons
why a higher dimensional vacuum Universe is studied. A geometrical property of
extra dimensions will enable us to catch a glimpse of a real world which lies beyond
our physical perspective.

References
1. P. S. Wesson, Five-Dimensional Physics (World Scientific, Singapore, 2006).
2. T. Fukui, S. S. Seahra and P. S. Wesson, 1. Math. Phys.42, 5195 (2001).
3. T. Fukui, Int. 1. Mod. Phys.l0, 893 (2001).
4. B. S. Dewitt, Theory of Groups and Fields (Gordon and Breach, New York, 1965).
5. J. M. Overduin and F. 1. Cooperstock, Phys. Rev.58, 043506 (1998).
6. A. G. Riess et al., arXiv.org;astro-ph/0611572.
 HAMILTONIAN THEORY OF BRANE-WORLD GRAVITY *

ZOLTAN KovAcstt and LAsZLO A. GERGELY:!:

t Max-Planck-Institut fur Radioastronomie,
Auf dem Hugel 69, D-53121 Bonn, Germany
+Departments of Theoretical and Experimental Physics, University of Szeged,
D6m ter 9, H-6720 Szeged, Hungary
[email protected], [email protected]

A brane-world universe consists of a 4-dimensional brane embedded into a 5-dimensional

space-time (bulk). We apply the Arnowitt-Deser-Misner decomposition to the brane-
world, which results in a 3+1+1 break-up of the bulk. We present the canonical theory
of brane cosmology based On this decomposition. The Hamiltonian equations allow for
the study of any physical phenomena in brane gravity. This method gives new prospects
for studying the initial value problem, stability analysis, brane black holes, cosmological
perturbation theory and canonical quantization in brane-worlds.

Keywords: canonical gravity, brane-world, embedding variables

The Hamiltonian theory of the brane-world scenario is based on the foliation of

the 4-dimensional (4d) world-sheet (the brane) which is embedded into the 5-
dimensional (5d) space-time manifold (the bulk B). Since the 3-dimensional (3d)
space-like slices of the foliation admit tangent bundles of co-dimension 2 with re-
spect to B, the slices form a two-parameter family of 3-spaces 2: tx with t, X E 1R
embedded in B. While the parameter t represents the many-fingered time in the
canonical formalism, a new parameter X defines the position of the brane in the
bulk. A common choice is at X = o.
The 3+ 1+ 1 decomposition of the 5d brane-world geometry allows one to express
the 5d field equations in terms of 3d quantities. These gravitational variables in
these picture are the three metric gab describing the intrinsic geometry of the slices
2: t x=o = 2:t, a vector field Ma and a scalar field M. The vector and scalar quantities
describe the contribution of the bulk-gravity. The extrinsic curvature of the leaves
embedded in B is given by the second fundamental forms Kab and Lab, the normal
fundamental forms Ka = .c a, and scalars K and .c associated with the two time-like
and space-like normal vector fields n a and la of 2: t . The quantities K ab , Ka and K
are equivalent with the time-derivatives of gab, M a and M, respectively, whereas
Lab and .c contain only pure spatial derivatives of them. l
As a result of the decomposition, the 5d Einstein-Hilbert action

SG[(5)gab] = Jd5 xL G = J d5 x v _(5)g(5) R (1)

*Research supported by OTKA grants no. T046939 and TS044665, the Janos Bolyai Fellowships
of the Hungarian Academy of Sciences, the Pierre Auger grant 05 CU 5PD1/2 via DESY /BMF
and the EU Erasmus Collaboration between the University of Szeged and the University of Bonn.
Z.K. and L.A.G. thank the organizers of the 11th Marcel Grossmann Meeting for support.

1290
 1291

(5) gab is the bulk metric and (5) R its scalar curvature) can be expressed in terms
of the set (gab, Ma, M; Kab, Ka, K, Lab, L), the lapse function N and the non-
vanishing components of the shift vector N a . The component of the shift vector
associated with the extra dimension is set to zero by the condition of the confinement
of the matter fields on the brane. With the set (gab, Ma, A1) chosen as canonical
coordinates of the vacuum bulk gravity, we can express the Lagrangian in the action
(1) solely in the terms of canonical coordinates and their time derivatives. Then we
introduce the momenta (nab, Pa, p) conjugated to the canonical coordinates, such
that the phase space of the 5d vacuum gravity is the set

(2)
with the abstract index A defined as gl = gab, g2 = lilIa, etc. In order to spec-
ify the possible states in the phase space, we first need the evolution equations,
then the vacuum constraint equations which restrict these solutions. (In fact they
restrict only the initial data. Once they are imposed, the dynamics preserves the
constraints. )
The Legendre transformation of the decomposed vacuum Lagrangian yields to
the Hamiltonian HG of the bulk gravity. This is a linear combination of the super-
Hamiltonian constraint HI and supermonentum constraint H:!:
(3)
When inserting the Lagrangian
LG[gA,nA;N,N a] = nAgA - HG[gA,nA;N,N a] ,
into the action (1) and extremizing it with respect to the canonical variables (2),
the lapse function and the shift vector, we obtain the equations of motion and the
constraints of the 5d vacuum gravity.
The Poisson brackets of any pair of functions on the phase space, as in the field
theories, can be defined with the help of the functional derivatives of the functions
with respect to the canonical variables or merely via the canonical commutation
relations. Then the dynamical equations of the bulk gravity lead to the forms

gA(X, X) = {gA(X, X), HG[N]} ,

1i"A(X, X) = {nA(x,x),HG[N]}
for any x E ~tx and X E IR, where the notation HG [N] includes the smearing of
the Hamiltonian density (3) with Nand N a .
When matter fields couple to gravity on the brane, we enlarge the phase space
of the 5d geometry with the canonical variables of the matter source. After decom-
posing the stress-energy tensor in the matter action, the Hamiltonian of the matter
fields can be derived as well. Then the dynamical and the constraint equations of
gravity must be supplemented with the contributions from matter, a procedure re-
sulting in the time-evolution equations and the constraints of the total system of
gravity and matter.
1292

The Hamiltonian formalism for brane-world scenarios gives a suitable starting

point for canonical quantization. In this approach the quantum state of gravity
should be described by a state functional W(t, x, x; gAl over the configuration space
(gA). This functional incorporates not only the intrinsic 3-geometries of the leaves
I: tx , but also the brane-off contributions of the bulk gravity. Dirac constraint quanti-
zation imposes either the vacuum constraints Hf and H;:, or the constraints of the
gravity coupled to matter on the state functional as operator equations, restricting
the possible states of the system:

itf(t,x,X;gA,7r A]W(t,x,X;gA] = 0,
it,;(t,X,X;gA,7r A]W(t,x,X;gA] = o. (4)
By inserting the explicit form of the super-Hamiltonian constraint into the first
equation, where the canonical momenta 7r A are represented by the operators ITA =
-iJ / J g A, we obtain a second order functional differential equation of the state
functional with respect to the canonical coordinates. The latter is the Wheeler-
deWitt equation of the brane-world gravity, which may be simplified by applying
the operator restriction of the super-momentum constraint from Eqs. (4).
We expect that the quantization of simple models already done in the con-
text of general relativity, such as the mixmaster universe 2 or the Einstein-Rosen
waves,3 is possible in the brane-world scenario either. The latter general relativis-
tic model illustrates, together with other examples,4,5 that various procedures can
be found which transform the constraint equations into new constraints, such that
the momentum canonically conjugated to the time variable enters the new super-
Hamiltonian only linearly. Hence the quantization of these systems leads to a func-
tional Schrodinger equation instead of the second order functional differential equa-
tion. There is hope that similar procedures based on the Hamiltonian formulation
of brane-world gravityl will be successful for various simple brane-world models as
well.

References
1. G. A. Gergely, Z. Kovacs, Phys. Rev. D 72, 064015 (2005).
2. C.W. Misner, Phys. Rev. Letters 22, 1071 (1969); Phys. Rev. 186, 1319 (1969); 186,
1328 (1969).
3. K.V. Kuchar, Phys. Rev. D 4, 955 (1971).
4. J.D. Brown, K.V. Kuchar, Phys. Rev. D 51, 5600 (1995).
5. Zs. Horvath, Z. Kovacs, G. A. Gergely, Phys. Rev. D 74, 084034 (2006).
 CASIMIR FORCE TEST OF A 6D BRANE WORLD"

ROMAN LINARES, HUGO A. MORALES-TECOTL AND OMAR PEDRAZA

Departamento de Fisica, Universidad Aut6noma Metropolitana Iztapalapa,
San Rafael Atlixco 186, CP 09340, Mexico D.F., Mexico.
[email protected],[email protected],[email protected]

Recently Randjbar-Daemi and Shaposhnikov put forward a 4D effective QED coming

from a Nielsen-Olesen vortex solution of an abelian Higgs model with fermions coupled
to gravity in 6D. In this contribution we study the corresponding effective Casimir effect.
We find that the extra dimensions yield fifth and third inverse powers in the separation
between plates for the effective Casimir force which are in conflict with known experi-
ments, thus reducing the phenomenological viability of the 6D model.

1. Introduction
Following the old idea that our observable 4D universe may be a brane extended in
some higher-dimensional space-time, many interesting field-theoretical models have
been constructed which have led to new insights in some long-standing problems
of particle theory such as the hierarchy problem and the cosmological constant
problem. Recently, starting from a Higgs model with fermions coupled to gravity in
6D, Randjbar-Daemi and Shaposhnikov constructed an effective QED in 4D, with
fermionic and gauge functions spread on the transverse direction in a small region
in the vicinity of the core of a Nielsen-Olesen vortex. l Amongst the various test
for QED, the Casimir force is as prominent as surprising. In 1948 Casimir showed
that ordinary QED predicts that two uncharged perfectly conducting flat plates,
placed in vacuum and separated by a distance I, should attract each other with a
7
force per unit area A given by F(I)jA = -2~~ 4c. This force is a purely quantum
effect caused by the alteration of the zero-point oscillators of the electromagnetic
field due to the plates. 2 In this contribution we use the Casimir effect to ask for
the phenomenological viability of the effective QED. The strategy is to start up
from the equations of motion for the effective 4D gauge field and to determine its
modified dispersion relations w(k), due to the extra dimensions. The computation
of the Casimir force follows directly from the knowledge of w(k).

2. The effective 4D QED and its dispersion relations

The effective 4D QED of Randjbar-Daemi and Shaposhnikovl emerges from consid-
ering a Nielsen-Olesen vortex-type solution of the abelian Higgs model with fermions
coupled to gravity in 6D. The various field configurations of the solution are 3 ,4
ds 2 = eA(r)T)/-wdxfld.Tv + dr 2 + e B (TJ a 2 drP ,
1> = J(r)e inlJ , aqAIJ = (P(r) - n)d8, (1)

*This research has been partially supported by grants CONACyT-SEP-2004-COl-47597 and

CONACyT-SEP-2005-COl-51132F.

1293
1294

where "71"1/ is the 4D flat metric, q is the 6D gauge coupling and a is the radius
e
of 8 1 covered by the coordinate. The boundary conditions that satisfy the field
functions 1(r), P(r), and the metrical functions A(r) and B(r), near the core of
the vortex are discussed in. 5 It should be stressed that solution (1) localize fields
of spin 0, 1/2 and 1. The effective QED action in this background results from a
specific mixture of the fluctuation of the 6D vector potential and the eM component
of the metric. Its explicit form is

8(W) = --;. f= drF(r)jd 4 x [(8I"Wv )2 + e- A 8r WI"8r WI"] , (2)

aq Jo

a:g
with F(r) = e B (r)/2(p 2 (r) +
2
eB(r)) and K, the 6D gravitational constant. Notice
that the quotient R == K,I q has dimension of length. The equations of motion for the
gauge fields are

(3)

A necessary requirement to have an effective 4D QED is that the spectrum for the
gauge field could contain a normalizable zero (or small mass) mode for a plane wave
function WI"(x, r) = RI"(r) eiPvxv, where PI" is the 4D wave vector. This is indeed
the case and such a wave function produces the dispersion relations

w2 ~2 2 2 2 1
2C = k + -e kr + -en 2 -t-112 ' (4)

where k is the 3D wave vector. Notice that the contribution of the extra dimensions
to the dispersion relations enters in two different ways. There is a continuum con-
tribution kr that comes from the radial extra dimension and there is one discrete
contribution that goes like n- 2 coming from the vortex number. 6

3. The effective Casimir force

Because of the presence of the plates, we impose the Dirichlet boundary condition
on the wave vector in the direction restricted by the plates: kN = 7r Nil, where l is
the distance between the plates. The Casimir energy between the plates is obtained
by summing up the zero-point energy per unit area, where the frequency of the
vacuum fluctuations is, according to (4)

(5)

with kJ.. = Jkr + k§. k1 and k2 are the wave vector components in the direction of
the unbounded space coordinates along the plates. Each of these modes contributes
an energy hw/2. Therefore the energy between plates reads

(6)
 1295

where L2 is the area of the plates and the parameter a, measuring the size of the
vortex'es core, appears associated to the integration in k r . The factor p indicates
the possible polarization of the photon. In our case p = 4. Using dimensional reg-
ularization we obtain that the exact Casimir energy per unit area in the vortex
scenario reads 6

= [Casimir f (l, a, £) , where,

nC7f 2 and
[vortex [Casimir =- 720[3 ,

f(l,a,£)=
45p (V1) (a) (l) ~
7f~r(-~) £
3
C n~l
00 (1) ~ K~ (fi
mn
2l m)
V~£-:;; . (7)

Here we recognize [Casimir as the standard 4D Casimir energy between parallel

plates. Moreover, the correction within the vortex scenario is encoded in the function
f (l, a, £) in the form of a factor rather than an additive term. This correction
depends on both parameters of the vortex scenario, namely the size of the core a
and the length £. To compare more neatly [vortex with the standard result it is
easier to get an approximate form of it. Taking l I £ « 1 in the argument of the
Bessel function K5/2 produces, to leading order in ll£
a a f)[vortex a a
[vortex ~ -a l4 + (3 £2[2 Fvortex = - 8l ~ -4a
ZS + 2(3 £2[3· (8)
where

a = 3nc((5) r (~) G and (3 = nc((2) ( (~) r (~) (~) 3/2

2 (47f)1/2 2 V~' 7 2
327f / 2 2 e
We conclude that the effective Casimir force contains both an attractive and a
repulsive contributions with inverse powers of the separation between plates l-5
and [-3, respectively. Demanding agreement of this force with the experiment to
set bounds for the parameters of the effective QED does not work due to the fact
that the correction is multiplicative yielding a different power in l with respect
to the standard case which goes as l-4. This result limits the phenomenological
implications of the effective QED here considered and hence of the 6D model.

References
1. S. Randjbar-Daemi and M. Shaposhnikov, QED from six-dimensional vortex and gauge
anomalies, JHEP 04 (2003) 016, hep-th/0303247.
2. H. B. G. Casimir, On the attraction between two perfectly conducting plates, Kon. Ned.
Akad. Wetensch. Proc. 51 (1948) 793-795.
3. M. Giovannini, H. Meyer and M. E. Shaposhnikov, Warped compactification on Abelian
vortex in six dimensions, Nuc!. Phys. B619 (2001) 615-645, hep-th/Ol04118.
4. M. Giovannini, Gauge field localization on Abelian vortices in six dimensions, Phys.
Rev. D66 (2002) 044016, hep-th/0205139.
5. T. Gherghetta and M. E. Shaposhnikov, Localizing gravity on a string-like defect in
six dimensions, Phys. Rev. Lett. 85 (2000) 240-243, hep-th/0004014.
6. R. Linares, H. A. Morales-Tecotl, and O. Pedraza, Casimir effect in a six-dimensional
vortex scenario, Phys. Lett. B633 (2006) 362-367, hep-ph/0505109.
 ELECTRO-WEAK MODEL WITHIN A 5-DIMENSIONAL
LORENTZ GROUP THEORY

O. M. LECIAN 1 •2 ,., G. MONTANI 1 ,2,3,t

lICRA - International Center for Relativistic Astrophysics.
2 Dipartimento di Fisica, Universitii di Roma "La Sapienza", P.le Aida Mora 5,
00185 Roma, Italy
3 ENEA C.R. Frascati (U. T.S. Fusione), Via Enrico Fermi 45,
00044 Frascati, Roma, Italy
• [email protected]
t [email protected]

The Electroweak model will be geometrized in a 5-D Riemann-Cartan framework: U(l)

weak hyper-charge group will be worked out in a Klauza-Klein scheme, while SU(2)
weak isospin group will be identified to suitable bein projections of the contortion field.
The possibility of introducing Ashtekar formalism in 5-D Klauza-Klein theories will be
investigated.

Keywords: kaluza-Klein; Electro-Weak Model; Lorentz Gauge Theory

1. Lorentz gauge theory

In general relativity, the expression of a generic isometric infinitesimal diffeomor-
phism and that of an infinitesimal local Lorentz transformation can be easily iden-
tified, i.e.

(1)
if spinor fields are introduced, the previous identification is no longer true, for
spinor fileds V) transform according to a spinor representation of the Lorentz group
'~) ----> S(A)V), and behave like scalars under diffeomorphisms. For this reasons,
local Lorentz transformations can be treated as an independent gauge group for
spinors a , i.e. 1jJ(x) ----> S(A)(x)1jJ(x) by the introduction of a gauge covariant deriva-
tive, i.e. 01-''1/) ----> DI-'V) = (01-' - ~~abA~b)1jJ and of a bosonic Lagrangian density
ex Fab(A)Fab(A) for the pertinent non-Abelian gauge field A~b, which transforms
like C~b ----> S(A)(x)ca bS-l(A)(x)+S(A)(x)OI-'S-l(A)(x), where structure constants
are defined as [~ab' ~cdl = D~{cd~ab'
In a fiat 4-D manifold, equipped of the set of bein vectors e~, the Lagrangian desity
is made invariant under diffeomorphisms by the presence of the bein vectors, i.e.
L = -~1jY'/le~ol-'1jJ + h.c .. In curved space-time, on the other hand, the request that
Dirac algebra be still valid in a non-fiat metric leads to the definition of a geomteri-
cal covariant derivative o,,1jJ ----> DI-'l/J = (0" - ~~abR~b - ~~abK~b)1jJ, where the bein
projections of Ricci coefficients and of the contortion field are introduced, the affine
connection of the manifold being r~p = { vp
/L } - Kf:p .Since Ricci coefficients are not

aWe emphasize that under a non-isometric transformation spinors behave like scalars.

1296
 1297

primitive objects, and cannot be taken as gauge fields, the contortion field, which
is a non-vanishing quantity even in flat space-time, can be identified to this gauge
field. Conserved quantities can be found by the identification of the conserved gauge
charge with the bein projection of the conserved spin angular momentum tensor:

(2)

2. 5-D model
In a 5-D Kaluza-Klein scheme V 5 = V 4 EEl Sl, the metric tensor components are

{ ~::: ~/k13~
j~v - (ilk? 13~13v = g~v,
(3)

where k is a constant introduced for dimensional reasons, 9' and 13v are a constant
and a (gauge) field that will be illustrated to be related with the weak hyper-charge
coupling constant and gauge field, respectively.
It is necessary to introduce matter fields and to extend the local Lorentz group to
this scenario.
Spinor fields can be here generalized by appending a suitable functional de-
pendence on the extra-coordinate and a normalization factor, i.e. 5 Xi (x P , x 5 ) =
Jr 41/Ji (xP)e--
i,27fN i.x 5
L-, where L is the length of the extra-ring. If ,5
is chosen as the
fifth Dirac matrix, 5-D Dirac equations show a natural chirality for right- and left-
handed spinors.
After the implementation of KK paradigm, the generators and the gauge fields of
the 5-D Lorentz group are no longer given by ~AB = i [rA, ,B]; as suggested by
the set of allowed KK transforations, 4-D generators are still valid, and SU(2) weak
isospin group can be looked for in the extra-D generators: from (2), it follows that
f;5~ == J~~I =? ~'5 ex: -'ig ;' PL. Weak_ i_sospin gauge fields are identified with the
pertinent extra-D gauge fields, while A~B must vanish in order to assure the correct
world transformations for the 4-D fields.
Accordingly, left-handed doublets

XL == (XXtiLL) == y7W
Vi
1 i2rrN··
e-r;-
iL ,LX
5

f,

and right-handed singlets XfiR are the natural spinor bases of this model.
Structure constants are given by the commutators of the generators: the only non-
vanishing commutators are those defined in section 1 and [~'5' ~J5] = Cf5~5~k5; ~05
is a redundant degree of freedom, and must be set equal to zero. The commutators
[~iib' ~'5] = 0, which vanish because the generators act on different spaces, define
the splitting of the 5-D Lorentz group after the implementation of KK paradigma,
SO(4, 1) -+ SO(3, 1) Q9 SU(2), and will be the starting point for the introduction of
Ashtekar formalism in this scheme.
1298

Conserved quantities are the weak isospin charge, according to (2), and the weak
hyper-charge, whereas SU(2) and U(l) gauge transformations for spinors are ob-
tained by the extra-D KK coordinate transformation.

3. Restoration of the Electro-Weak model

Collecting all the terms together,l after dimensional reduction, the Lagrangian den-
sity consists of the Lagrangian density of the Electro-Weak model and that of the
Lorentz gauge theory, as can be easily worked out by taking into account the proper-
ties of the connections and by eliminating the dependence on the extra-coordinate.
As in usual KK theories, relations between the extra-components of the metric
tensor and hyper-charge gauge objects can be found.

4. Ashtekar's formalism?
In 5 D, the total group we are dealing with is SO(4,1) ...... SO(3, 1) (>9 SU(2) ......
SU(2) (>9 SU(2) (>9 SU(2). An ADM splitting on the 5-D manifold gives a 4 + 1
reduction: on the 4-D Euclidean resulting manifold, where the index 5 plays a role
analogous to that of 0 in standard scenarios, a further splitting can be performed. 2
Here, SO(4) ...... SO(3)(>9SU(2), and Ashtekar variables 3 can be defined: C~ = Ejkw~k
for SO(3), and ± A~ = ± A~ = w~ ± ~C~ for SU(2); here, w~k denote the Euclidean
spin connections, and Eijk == Ei5jk (we recall that, in 5-D KK theory, the index 5 is
scalar). Bein vectors are evolutionary variables while connections are not: they can
be mixed up in order to find suitable variables, but, as a perspective, a more general
theory can be looked for, such that the SU(2) (>9 SU(2) group can be built making
use of 4-D objects only, thus recasting the original Ashtekar formalism, without
mixing up + A~ and e~.

References
1. O.M.Lecian, G.Montani Int. Journ. Mod. Phys. D 15,717 (2006) and the references
therein.
2. V.Lacquaniti, G.Montani, Int. J. Mod. Phys., D 15, 559 (2006).
3. c.Rovelli, Class. Quantum Gmv., 8, 1613 (1991).
 SPACETIMES WITH CONSTANT SCALAR INVARIANTS*

8IGBJ0RN HERVIK
Department of Mathematics & Statistics, Dalhousie University,
Halifax, Nova Scotia, Canada B3H 3J5
[email protected]

In this talk we will discuss spacetimes with constant scalar invariants (C8I spacetimes).
There are many examples of such spacetimes, among them spacetimes with vanishing
curvature invariants and homogeneous spaces. Special emphasis will be put on a certain
class of spacetimes to which all known inhomogeneous CSI spacetimes belong. The role
of this class of spacetimes will be discussed, as well as the mathematical and physical
properties of the spacetimes belonging to this class.

1. Introduction
Let us denote by Ik the set of all scalar invariants constructed from curvature
tensors of M up to order k.
Definition (CSI k spacetimes): M is called CSI k if for any invariant I E I k ,
°11 1 = 0 over M.
Moreover, if a spacetime is VSh or CSh for all k, we will simply call the
spacetime V S I or C S I, respectively. The set of all locally homogeneous spacetimes
will be denoted by H. Clearly VSI C CSI and H C CSI.
In the Riemannian case (i.e., signature (+ + + ... ) ) we have:
Theorem (Riemannian CSI):A Riemannian space is CSI if and only ifit is locally
homogeneous. Moreover, the set of curvature invariants, I, uniquely determines the
metric up to isometries.
Proof: See Prufer, Tricerri and Vanhecke. 2 Proof uses a theorem by Singer 3 and
compactness of O(n).

1.1. Lorentzian case

The Lorentzian case is considerably more difficult to deal with. However, all known
CSIs (to us) are of the following two forms:

(1) Locally homogeneous spaces

(2) A subclass of the higher-dimensional Kundt spacetimes.

The locally homogenenous spaces are trivially eSIs. For the non-homogeneous
space times it is advantageous to consider the following boost-weight decomposi-
tion: For an arbitrary tensor T:

T = 2:)T)b,
b

where (Th means the projection onto the vector space of boost weight b.

*Based on a work done in collaboration with A.A. Coley and N. Pelavas. 1

1299
1300

For the (locally) homogeneous spaces the boost-weight decomposition can be of

the most general type:

2+k
V(k) R= L (V(k) R) b .
b=~(2+k)

However, for the non-homogeneous eSIs, a frame can be chosen such that
a
V(k) R= L (V(k) R) b .
b=~(2+k)

In particular, this means that (R)b = 0, b > O. Moreover, all components of (R)a
are constants. We note that there are no restrictions on the components of negative
boost order since these do not contribute to the curvature invariants.

2. Kundt metrics
Assume that M admits a geodesic, non-twisting, non-expanding, shear-free null-
vector e. Then

(1)

We will refer to these metrics as K undt metrics.

Using these Kundt metrics we can construct eSI metrics by requiring that there
exists a frame (e, n, mi) such that the components of (V(k) R)a are constants and
(V(k)R)b = 0 for b > O.
This requirement implies that:

2 - -
d sH-gij (
Ua,X
k)d x id x j - s::
-Uijmmi 3

is a locally homogeneous space. Moreover,

and different types of solutions can be categorised in terms of W,(l) = W'v

t "

2.1. Kundt CS! examples:

For all cases, ds~ = Oijmim J is a locally homogeneous space, and
2
H( v, u, :r k ) = Vs [40" + (WP))(W(l)i)] + vH(l) (u, xk) + H(a) (u, xk),

, (u, xi) =
wP) 0: The space is now automatically a eSI space. Special Brinkmann
metrics and plane waves belong to this class.
 1301

wP)(u, Xk) = constant. This case is CSI if and only if the simpler metric, M:
•
ds 2 = 2du (dV + v; a-du + vw?)m i ) + 6iJmimJ, (2)

is a homogeneous space. The curvature invarian~of the Kundt metrics in this class
will have the same invariants as M; i.e. I = I(M).

W~1)(U,Xk) = non - constant, Will(u,xk) = constant: As an example of

this class we give a generalisation of" AdS-waves" (AdS waves 4 belong to this class):

ds 2 = e- 2 ¢(xU) [2du (dv + H du + Wadx a + Wowo) + 6a b dx adx b ] + 6o(3W o W(3,

where
wa = VW(l)(u
a'
x a ) + W(O)(u
a'"
xa X O
) W 0: = W(O)(u
a'"
xa XO )

¢ = ¢(xC», ¢oconstant,
2
H = Vs (WP»)(w(1)a) + vH(1l(u, Xa , XC» + H(O)(u, Xa , XO).
Here, for these examples, the metric inside the square brackets are VSI space-
times 5 while the metric 6o (3w Oo w(3 is a (locally) homogeneous space. This spaces have
been in the focus of attention as possible candidates for supersymmetric solutions
of supergravity.

The CSI spaces are related to the problem of distinguishing metrics using the
curvature invariants. As we have pointed out, in Lorentzian geometry, many in-
equivalent metrics can have the same curvature invariants. This is perhaps most
manifest for the VSI case;5 however, the CSI case also shows a number of such
cases. In spite of this 'degeneracy' of the CSI spacetimes, the investigation supports
the following conjecture:
The CSI conjecture: All Lorentzian CSI space times are either locally homoge-
neous, or a subclass of the Kundt spacetirnes.
This conjecture indicates that there is a deep connection between Kundt and CSI
spacetimes.

References
1. A.A. Coley, S. Hervik and N. Pelavas, Class. Quantum Grav. 23 (2006) 3053
2. F. Priifer, F. Tricerri, L. Vanhecke, 1996, Trans. Am. Math. Soc. 348, 4643.
3. 1. M. Singer, 1960, Comm. Pure and App. Math. 13, 685.
4. S.T.C. Siklos, 1985, Lobatchevski plane wave gravitational waves, in Galaxies, Axisym-
metric systems and Relativity, ed. M. MacCallum (CUP, Cambridge).
5. V. Pravda, A. Pravdova, A. Coley and R. Milson, Class. Quant. Grav. 19,6213 (2002);
A. Coley, A. Fuster, S. Hervik and N. Pelavas, to appear in Class. Quant. Grav.
 HIGHER DIMENSIONAL VSI SPACETIMES AND
SUPERGRAVITY

ANDREA FUSTER
NIKHEF, Kruislaan 409, 1098 SJ Amsterdam, The Netherlands
fuster@nikhef. nl

NICOS PELAVAS
Department of Mathematics and Statistics, Dalhousie University, Halifax, Nova Scotia B3H
3J5, Canada
[email protected]

We present the explicit form of higher dimensional VSI spacetimes in arbitrary number
of dimensions. We discuss briefly the VSI's in the context of supergravity/strings.

1. Introduction
We are interested in higher dimensional spacetimes for which curvature invariants
of all orders vanish (VSI spacetimes), contained within this class are the higher
dimensional pp-waves. In general, the higher dimensional VSI spacetimes have Ricci
and Weyl type III. I However, it is desirable to obtain explicit metric functions as
has been done in four dimensions. 2 We present metrics for Ricci type N, Weyl type
III VSI spacetimes.

2. Higher dimensional VSI spacetimes

Any N-dimensional VSI metric can be written in the form 3 ,4

ds 2 = 2du [dv + H(u, V, xk) du + Wi(u, V, xk) dx i ] + dxidx i (1)

where u, v are light-cone coordinates and xi, i = 1, ... , N - 2, are real spatial
coordinates. The functions H, Wi are real-valued. Note that (1) is a subclass of the
higher dimensional Kundt metrics. 3 Restricting (1) to be of Ricci type N results
in the Einstein equation Ruu = <P, where <P is determined by the matter field (see
Appendix). Two distinct cases arise depending on whether the functions Wi depend
on the light-cone coordinate v (E = 1) or not (E = 0):
_ 2E (0) k (0) k
WI - - 1 v+ WI (u,x), Wj = Wj (u,x)
x
Here j = 2, ... ,N - 2; the superscript (0) indicates functions without v-dependence.
The case E = Wi(O) = 0 corresponds to higher dimensional pp-waves. It is perhaps
surprising that only one of the Wi functions is allowed to depend on v, as in the 4-D
case. The resulting spacetimes are summarized in Table 1. An important remark
is that the function wiO) can be gauged away by a coordinate transformation, the
corresponding results have been recently presented. 5 In this note we treat all W?)
functions equally, thereby revealing additional algebraic symmetries found in the

1302
 1303

metric functions of solutions. This complements the paper by Coley et aI, wherein
further details and explanation may be found.

E Weyl Metric functions Eq.

Wi = Wi(O)(u,x k )
III A.1
H = H(D) (u, xi) + ~ (F _ W(D)i ,i) V, F(u, xi) defined by F,i = tl.Wi(O)

Wi = xkBki(U), BkiCU) antisymmetric

0 N A,2
H=H(O)(u,x i )

Wi as in type N
A.2
0 A.3
H(O) = ~WiWi + xifi(U) + HiD)

W, =- x2, V + W~O) (u, xk)

Wj = W,cD)(U,x k )
III A.4
H + 1.
= H(O)(u 'xi)2 (F _ W(D); . _
,'I-
2W,(o») v + ~v_2_
xl 2(x1)2'

F(u,xi)definedbyF,,= ,,2, W(D)i ,i+tl.W~O), F,j=tl.WjO)

W, = -2 xv, + x j Bjl Cu) + C , (u)

1 N Wj = xiB;j(u) + Cj(u), Bjl (u), Bij (u) antisymmetric A.5
w(O) 2
H - H(O)(u Xi) _ _
- ,
l_v
xl
+ ~v~
2(xl)2

Wi as in type N

0 H(O) = ~CW~O»)2 + ~L(Wj - x, Blj)2 + x'gOCU) + x'Xigi(U) - feiPDCu)x'XiXi, A.5

j
iP = iPD(U)X '

Table 1: All higher dimensional VSI spacetimes of Ricci type N. Ricci type 0 (vacuum) for <I> = 0
in (A.l)-(A.5).

3. Outlook: VSI's in supergravity

We can consider the embedding of VSI spacetimes in supergravity. The idea is to
construct bosonic solutions of the supergravity equations of motion where the met-
ric is that of a VSI. 6- 8 This has been extensively studied in the literature in the case
of (Weyl type N) pp-waves. These are exact string solutions to all orders in a' .9,10
The corresponding proof relies on the pp-waves invariants property. Furthermore,
1304

supergravity pp wave solutions preserve supersymmetry. The VSI metrics presented

here generalize the pp-waves while the invariants property is maintained. Therefore
it is natural to expect VSI supergravity solutions to have some of the pp-wave so-
lutions' features. In fact, a certain class of VSI supergravity solutions has already
been proved to be exact to all orders in a'. 7 Supersymmetry properties have only
been studied in the case of VSI's with a covariantly constant null vector. s In con-
clusion, these results provide an opportunity to explore VSI supergravity solutions,
and their associated supersymmetries.

Appendix A. Einstein equations

In the equations below Wik = W(Ok)
z,
- Wk(O)
,z
and H(1) is the coefficient of v in the
metric function H.
~H(O) - ~ Wik Wik - 2H(1) ,i W(O)i - H(l)W(O)i . - W(O)i . + <I> = 0 (A.I)
4 ,1,. ,2U

(A.2) (A.3)
i<j

Xl ~ ~~O») _(X~)2
( ((W;O)? + 2)Wj - Xl Blj)2) - LB~j 2 L:
- B;j + <I> = 0
J J t<J
(A.5)

References
1. A. Coley, R. Milson, V. Pravda and A. Pravdova, Class. Q. Grav. 21, 5519 (2004).
2. V. Pravda, A. Pravdova, A. Coley and R. Milson, Class. Q. Grav. 19, 6213 (2002).
3. A. Coley, R. Milson, N.Pelavas, V. Pravda, A. Pravdova and R. Zalaletdinov, Phys.
Rev. D 67, 104020 (2003).
4. A. Coley, S. Hervik and N. Pelavas, Class. Q. Grav. 23, 3053 (2006).
5. A. Coley, A. Fuster, S. Hervik and N. Pelavas, Class. Q. Grav. 23, 7431 (2006).
6. V. Frolov and F. Lin, Phys. Rev. D 73, 104028 (2006).
7. A. Coley, Phys. Rev. Lett. 89, 281601 (2002).
8. E. Bergshoeff, R. Kallosh and T. Ortin, Phys. Rev. D 47, 5444 (1993).
9. G. Horowitz and A. Steif, Phys. Rev. Lett. 64, 260 (1990).
10. G. Horowitz and A. Steif, Phys. Rev. D 42, 1950 (1990).
 VSI & VSIi SPACETIMES IN HIGHER DIMENSIONS·

ALENA PRAVDOvA
Mathematical Institute
Academy of Sciences
Zitna 25
11567 Prague 1
Czech Republic
[email protected]

We briefly summarize some properties of spacetimes with vanishing curvature invariants in four
and higher dimensions.

1. VSI spacetimes

A curvature invariant of order n is a scalar obtained by contraction from a poly-

nomial in the Riemann tensor and its covariant derivatives up to the order n.
Kretschmann scalar RabcdRabcd is an example of a zeroth order invariant.
In the Riemannian geometry vanishing of zeroth order invariants implies flat
spacetime. This is not true in the Lorentzian case. An example of spacetimes with
vanishing curvature invariants of all orders is known since 1960 - pp waves. 1
Here we are interested in all such spacetimes (VSI spacetimes) in arbitrary
dimension. VSI spacetimes in four dimensions have the following properties:
Theorem: 2 All curvature invariants of all orders vanish if and only if the following
two conditions are satisfied:
(A) The spacetime possesses a non-diverging, shear-free, geodesic null congruence.
(B) Relative to the above null congruence, all curvature scalars with non-negative
boost-weight vanish.
The analytic forms of conditions (A), (B) are

K,= p = = 0 (=} Kundt's class)

(J"

Wo = Wl = W2 = 0 (=} Petrov types III, N, 0)

<Poo = <POl = <P02 = <Pll = 0 (=} Segre types {(31)}, {(211)}, {(1111)})
R=O.

All such 4D metrics belong to the Kundt class.

The most general form of the Kundt metric in adapted coordinates u, v, (, ( is

ds 2 = 2du[Hdu + dv + Wd( + Wd(]- 2P- 2 d(d( ,

where the metric functions

H=H(u,v,(,(), W=W(u,v,(,(), P=P(u,(,()

-This research has been supported by research plan No AVOZ10190503 and research grant
KJBI019403.

1305
1306

satisfy the Einstein equations. For the VSI spacetimes we can put P = 1 and
determine the form of Wand H. Explicit forms of a1l4D VSI metrics are given in. 2
VSI metrics can be further classified according to their Petrov type, Segre type and
vanishing or non-vanishing of NP coefficient T. This leads to 16 non-trivial classes,
with pp-waves being the Petrov type N, vacuum, T = 0 class.
For higher dimensional generalization of the above theorem the generalization
of the Petrov classification3 ,4 to higher dimensions and the Bianchi identities 5 are
needed. The higher dimensional VSI theorem turns out to be surprisingly similar
to the 4D case.
Theorem: 6 All curvature invariants of all orders vanish in aD-dimensional
Lorentzian spacetime if and only if there exists an aligned non-expanding (Sij = 0),
non-twisting (Aij = 0) geodesic null direction £a along which the Riemann tensor
has negative boost order.
These spacetimes thus again belong to the higher-dimensional Kundt class 6 (see
also 7 ,8 for further development).
VSI spacetimes also satisfy an interesting property:
Theorem: 9 (The E property of VSI spacetimes)
For and only for VSI spacetimes (in arbitrary dimension D) one can find for ar-
bitrarily large N and arbitrarily small E a frame in which all components of the
Riemann tensor and its derivatives up to order N are smaller then E.
Thus we can make, by an appropriate boosting of the frame, all components of the
Riemann tensor and its derivatives up to an arbitrary N essentially undetectable.
This is very different from a general spacetime, where we can only locally transform
away Christoffel symbols but not the Riemann tensor.

2. VS1i spacetimes
For Riemannian metrics, vanishing of the zeroth order invariants implies flat space-
time. For studying peculiar properties of Lorentzian metrics, it is also of interest
to consider VSI i spacetimes,9 spacetimes with vanishing curvature invariants up to
order i. Obviously they do not have a Riemannian analog. From the proof of VSI
theorem it follows that VSI 2 is equivalent with VSI.
It turns out that in 4D all VSI 1 spacetimes that are not VSI are of Petrov type
N and in vacuum they have non-vanishing expansion and/or twist p = e + iw =I- 0
(see9). In the case with w = 0 they thus belong to the Robinson-Trautman class.
In higher dimensions, proper VSI 1 vacuum spacetimes are again of type N with
non-vanishing expansion and/or twist. This, however, also implies non-vanishing
shear 5 ,lO and consequently 4D vacuum VSI 1 space times do not have shearfree coun-
terpart in higher dimensions. Explicit form of all higher dimensional vacuum VSI 1
metrics is not known, though simple examples can be readily constructed.
 1307

References
1. P. Jordan, J. Ehlers, W. Kundt, Abh. Akad. Wiss. Mainz, Math. Nat., No 2 (1960)
p.77
2. V. Pravda, A. Pravdova, A. Coley, R. Milson, All spacetimes with vanishing curvature
invariants, Class. Quantum Grav. 19 (2002) 6213
3. R. Milson, A. Coley, V. Pravda, A. Pravdova, Alignment and algebraically special
tensors in Lorentzian geometry, Int. J. Geom. Meth. Mod. Phys. 2 (2005) 41
4. A. Coley, R. Milson, V. Pravda, A. Pravdova, Classification of the Weyl Tensor in
Higher Dimensions, Class. Quantum Grav. 21 (2004) L35
5. V. Pravda, A. Pravdova, A. Coley, R. Milson, Bianchi identities in higher dimensions,
Class. Quantum Grav. 21 (2004) 2873
6. A. Coley, R. Milson, V. Pravda, A. Pravdova, Vanishing Scalar Invariant Space times
in Higher Dimensions, Class. Quantum Grav. 21 (2004) 5519
7. A. Coley, A. Fuster, S. Hervik, N. Pelavas, Higher dimensional VSI space times, Class.
Quantum Grav. 23 (2006) 7431
8. A. Fuster, N. Pelavas, Higher dimensional VSI spacetimes and supergravity, hep-
th/0611103
9. N. Pelavas, A. Coley, R. Milson, V. Pravda, A. Pravdova, VSli spacetimes and the
E-property, J. Math. Phys. 46 (2005) 063501
10. J. Podolsky, M. Ortaggio, Robinson-Trautman spacetimes in higher dimensions, Class.
Quantum Grav. 23 (2006) 5785
 THE ELECTRO-WEAK MODEL AS A PHENOMENOLOGICAL
ISSUE OF MULTIDIMENSIONS

FRANCESCO CIANFRANI* and GIOVANNI MONTANPt

* ICRA -International Center for Relativistic Astrophysics
Dipartimento di Fisica (Gg), Universita di Roma, "La Sapienza",
Piazzale Aida Moro 5, 00185 Rome, Italy.

t ENEA-C.R. Frascati (U. T.S. Fusione),

via Enrico Fermi 45, 00044 Frascati, Rome, Italy.

E-mail: [email protected]
[email protected]

We consider Kaluza-Klein theories as candidates for the unification of gravity and the
electro-weak model. In particular, we fix how to reproduce geometrically the interaction
between fermions and gauge bosons, in the low energy limit.

Keywords: Kaluza-Klein theories; Electro-Weak Model.

One of the main task of modern theoretical physics is an explanation for the differ-
ent kind of interactions in nature. Among them, gravity plays a special role, being
described by a classical and geometrical theory (General Relativity); therefore, the
most difficult obstacle seems to be its unification with other forces. While it is a
common opinion that in a quantum formulation the gravitational field could look
more similar to gauge ones, no evidence that Quantum Gravity will imply a uni-
fication exists (for example, in Loop Quantum Gravity no unification happens 1 ).
Furthermore, a different perspective exists, that is to move the formulation of fun-
damental interactions close to that for gravity, by a geometrical interpretation for
gauge fields degrees of freedom. A possibility is to arrange bosons into the metric
tensor; it implies space-time to have more than four-dimensions. The additional
space must be compactified to distances not yet accessible to experiments. Such
kind of models are known as Kaluza-Klein theories. 2 ,3
In these models, the main assumption is the following form for the metric tensor

9!"v (x P ) + rmn (x P ; yr)Cl'r (yr)C~/yr)A~f (xP)A;J (x P ) rmn (x P; yT)/;;& (yT)A~-f (X P ))

JAB =
(
rmn(x P; y' )/;fj.(Y' )A;! (x P ) rmn(xP; yT)

being XfL and gfLv the four-dimensional coordinates and metric, respectively, while
ym and rmn the analogous ones on the extra-dimensional space, endowed with
Killing vectors ~;g;..
The possibility to geometrize a gauge theory is encoded in the existence of a ho-
mogeneous manifold, whose isometries reproduce the algebra of the gauge group in

1308
 1309

the following way

(1)

with C~ M structure constant of the Lie group.

It is possible to interpret A~I as gauge bosons since, by the dimensional reduction
of the Einstein-Hilbert action, the Yang-Mills Lagrangian density comes out.
However, the geometrization to be complete requires that fermions have to be intro-
duced in such a scheme and at this point Kaluza-Klein models failed. In particular,
it was not satisfied the attempt to reproduce the spectrum of Standard Model
fermions starting from multidimensional spinors, by identifying their gauge proper-
ties with extra-dimensional degrees of freedom.4
To overcome some of these difficulties, here we propose a phenomenological ap-
proach: 4-fields arise after an average of the equations of motion on the extra-
dimensional space, because additional dimensions are undetectable;5 as a conse-
quence, geometrical properties related to them are seen in a non-trivial way in a
4-dimensional perspective.
To geometrize the 8U(2) sector of the Electro-Weak model, we consider spinor on
8 3 . According with our approach, low-energy particles are solutions of the Dirac
equation averaged on the full space, with a unit weight because of its homogeneity.
Let us consider the following spinor
1 -la >.(p)e(q)(~)
\IIr(x; Y) = Xrs(Y)'lj)s(x) = JVe 2 (p)rs (q) Y 'l/Js(x) (2)

being V the volume of 8 3 , (J (p) Pauli matrices and>" a constant matrix satisfying

(>--l)(P)
(q)
=~
V
r V"4e om S (p)d3 y.
iS3
m
(q)
(3)

while for functions S we take

1
Sip) = -c(p)e-(31] r; > O. (4)
p
In this way, \II is an approximate solution of the averaged Dirac equation, with cor-
rections of order of p-1;6 we take it as a good approximation since in the low-energy
limit a weak dependence on extra-coordinates, thus a big value for p, is expected.
Moreover, by inserting \II into the Dirac action, we achieve the geometrization of
8U(2) gauge connections; otherwise, some additional and, in general, gauge violat-
ing p-1 terms come out. Therefore, the price to pay for the geometrization is the
appearance of some non-renormalizable interactions; however, it is not a problem,
because we are performing a low-energy effective theory and gauge violating terms
are supressed by the order parameter governing the low-energy limit.
Hence, the Electro-Weak model model can be inferred from the geometry of a space-
time manifold V 4 ®8 1 ®8 3 , where by the 8 1 manifold the U(l) sector is reproduced.
Since in a 8-dimensional manifold spinors have got 16 components, we suggest to
recast a quark generation and a fermionic family into the same geoemtrical object,
1310

thus explaining the equality of their number. At the end, we reproduce the full
fermionic spectrum of the Standard Model from the following two multidimensional
spinors for each fermionic family/quark generation

where the coefficients nr are related to particles hypercharges, Yr = n6 .

In this scenario, a lower bound for (3 can be obtained from experimental limits for
processes that violate the conservation of gauge charges. For example, the upper
bound on the cross section for the decay of a neutron into a proton plus a couple
neutrino-antineutrino gives (3 ;:::: 10 14 .
In order to give masses to particles, we introduce a field whose dimensional reduction
gives the Higgs field. By assuming the same dependence on extra-coordinates as for
spinors, two troubles of the standard Higgs formulation are overwhelmed:
(1) the appearance of the standard Kaluza-Klein mass term can explain the fine
tuning we have to perform, on the parameter of the potential, to stabilize Higg's
mass;
(2) the possibility to assign opposite hypercharges to the two components of the
field allows to reproduce a mass term for neutrinos by a standard Yukawa
coupling.

Finally, gauge violating terms give corrections to bosons masses; in particular, from
upper bound on the photon mass we get (3 > 10 28 .

References
1. C. Rovelli, Quantum Gravity, Cambridge University Press, (2004)
2. T. Appelquist, A. Chodos, P. Frund, Modern Kaluza-Klein theories, Addison Wesley
Publishing, Inc., (1987)
3. F. Cianfrani, A. Marrocco, G. Montani, Int. J. Mod. Phys, D14, 7, (2005), 1195
4. D. Bailin, A. Love, Rep. Prog. Phys., 50, (1987), 1087
5. F. Cianfrani, G. Montani, Mod. Phys. Lett., A21, 3, (2006), 265
6. F. Cianfrani, G. Montani, submitted to Classical and Quantum Gravity
 HAMILTONIAN FORMULATION OF THE 5-D KALUZA-KLEIN
MODEL AND TEST-PARTICLE MOTION

VALENTINO LACQUANITI*o and GIOVANNI MONTANI*t

* ICRA -International Center for Relativistic Astrophysics
Dipartimento di Fisica (G9) , Universitii di Roma, "La Sapienza",
Piazzale Aida Mora 5, 00185 Rome, Italy.

o Dipartimento di Fisica "E.Amaldi", Universit6. di Roma "RomaTre",

Via della Vasca Navale 84, 00146 Rome, Italy.

t ENEA-C.R. Frascati (U. T.S. Fusione),

via Enrico Fermi 45, 00044 Frascati, Rome, Italy.

E-mail: [email protected]@icra.it
[email protected]

We examine the ADM reformulation of the 5-D KK model: the dimensional reduction
is provided to commute with the ADM splitting and we show how the time component
of the gauge vector is given by combination of the Lagrangian multipliers for the 5-
D gravitational field. We consider 5D particles motion and after dimensional reduction
the definition of charge is recovered within electrodynamic coupling. A time-varying fine
structure constant is recognized because an extra scalar field is present in the 4-D theory.

Keywords: Kaluza-Klein theories; ADM splitting; Hamiltonian formulation.

KK 5-D model provides a unified geometrical picture of gravitation with electro-

magnetism plus an extra scalar field ¢ . This picture is achieved via the assumption
of some restrictives hyphotesis on the 5-D space-time that break the 5-D Poincare
symmetryl ,2 ,3 So a first goal is to understand if the KK reduction is compatible
with AMD splitting, that is usually requested in canonical approach, in order to
compute the Hamiltonian of the model. 4 Due to the symmetry breaking we could
have not equivalent dynamics wheter or not the ADM splitting is performed before
we implement the KK reduction. We have two possible procedures (we label these
as KK-ADM and ADM-KK ). In KK-ADM procedure we perform as first the usual
KK reduction of the metrics, and then a 3 + 1 ADM splitting of the gravitational
tensor and the abelian gauge vector:
-Si - ¢2AoAi
-{}ij - ¢2 AiAj
-¢2A i
Here JAB is the 5D metrics, AI-' the gauge vector, N, Si, {}ij the Lapse function,
the 3D Shift vector and the 3D induced metrics (A, B = 0,1,2,3,5; I-l, v =
0,1,2,3; i, j = 1,2,3). In this way we have a not complete space-time slicing,
due to the fact that we are doing a 3 + 1 splitting in a 5-D ambient, so the extra
dimension is not included. In the ADM-KK procedure we deal as first with a 4 + 1

1311
1312

splitting that includes the extra dimension and after consider the KK reduction
related to the pure spatial manifold:

-Ni
-7'J ij _2 q} AiAj
-¢2A i
Here N j and hjj are the 4D Shift vector and the 4D spatial induced metrics
(i, j = 1,2,3,5). Now we have a complete slicing but the set of variables lacks the
component Ao. Hence, both of procedures are unsatisfactory and we don't know if
they commute. Despite the outcomes of the metrics seem to be differents , never-
theless we are dealing with objects that must show well defined properties under
pure spatial KK diffeomorphism. This allow us to look for "conversion formulas"
between our two metrics. Indeed we can implement the KK reduction on Nj; by this
we recognize that N j is not a pure 4D spatial vector neither simple gauge vector
but a mixture of them. A detailed study of the tetradic structure allow us to state
the following formulas for N j which map metrics into each other.

Ni = Si + ¢2 AoAi
{ N5 = ¢ 2A O

The real physical meaning of these formulas is provided once we compare the two
Lagrangian. By our formulas we can recast the lagrangians in the same set of vari-
ables and we get:
1
Ladm-kk = Lo + --...fiJNKoT}¢
87rG
Here K is the trace of the 3D extrinsic curvature and we have aT} = (l/N)(oo-Si oi ).
Lo is a term, egual for both Lagrangians, that contains no derivatives of ¢; the re-
maining contributes are found to be equivalent apart from surface ternl. Hence, by
assuming proper boundary conditions, in order to discard these terms, we conclude
that we are dealing with equivalent dynamics and that our conversion formulas have
a real physical meaning. Therefore we can conclude that ADM splitting commutes
with KK reduction; we are able in both cases to have a complete space time slicing
and to rebuild the time component of the gauge vector. Moreover, due to the com-
mutation, we can have a unique, well-defined, Hamiltonian H; it can be proved that
H can be casted in an explicit form like the following one: H = N HN +SiHi+AoHo.
Here H N is the Superhamiltonian, Hi the Supermomentum, HO the electrodynam-
ical contribute; Ao appears like a lagrangian multiplier and it is predicted by the
conversion formula AoooN5 ; this means that after the KK reduction one of the
geometrical constraints of the model becomes a gauge one. So the proof of commu-
tation is a positive check for the consistency of the KK model, is necessary in order
to consider the hamiltonian formulation and provides insight in the understanding
of the gauge simmetry generation. Finally, it can be a first step in the Ashtekar
reformulation of KK model. For an accurate discussion about these results see ref. 5
 1313

Another point of interest is the presence of ¢ in the model and its role in the dynam-
ics. Indeeed there are some reasons to think that ¢ can be a good time variable in
the relational point of view and hence solve the problem of the frozen formalism. An
insight on the role of ¢ is provided by the study of test-particles motion. By study-
ing the 5D geodesic equations we are able to reproduce the usual electrodynamics
for a test-particle in a 4D space-time, where the charge-mass ratio is ruled out as
2
follows: qjm = u5(1 + ~ )~1. In this formula U5 is the fifth covariant component of
the 5D velocity and can be proved that it is a constant of motion and a scalar under
KK transformations. In general, the charge is not conserved due to the presence of
¢, and at the same time we cannot be sure that our test particle has a realistic
qjm. A large scalar field (¢ > 10 21 for the electron) allow us to have realistic value
for the charge mass ratio avoiding the problem of planckian mass, and, moreover,
allow us to restore the conservation of charge at a satisfactory degree of approxima-
tion. Actually, a time-varying charge is very interesting; an isotropic, slow varying ¢
can explain the time-variation of the fine structure constant over cosmological scale
which seems to be inferred by recent analysis. 6 Other features of ¢ arise from the
study of FriedmannjDeSitter-like models; we deal only with ¢(t) and the 3D scale
factor a(t) ; our results show that in both cases ¢ mimics the behaviour of the scale
factor and can be increasing or decreasing in time with power law (Friedmann) or
exponential law (DeSitter). Hence, the scalar field shows some classical feature that
we can link to a possible role of it as a time -like variable. Seems interesting to deal
with induced matter theory (and with a detailed cosmological solution ) andj or
with Ashtekar reformulation of variables in order to deep understand the role of ¢.

References
1. T.Kaluza On the Unity Problem of Physics. Sitzungseber.Press.Akad.Wiss.Phys.Math.
(1921) 966-972
2. O.Klein, Z.F.Physik 37, 895. (1926)
3. T. Appelquist, A. Chodos, P. Frund, Modern Kaluza-Klein theories, Addison Wesley
Publishing, Inc., (1987)
4. R. Arnowitt, S. Deser, C.W.Misner, Gravitation: An Introduction to current Research.
L. Witten (ed.), Wiley, New York,p.227 (1962)
5. V.Lacquaniti, G.Montani, Int.Journ.Mod.Phys., D15,4,(2006), pp.559-581
6. M.Kozlov, V.Korol, J.Berengut, V.Dzuba, V.Flambaum, PhysRev, A 70,062108,(2004)
ELECTROMAGNETISM AND PERFECT FLUIDS INTERPLAY IN
MULTIDIMENSIONAL SPACETIMES

NIKOLAI V. MITSKIEVICH
Department of Physics, GUGEl, Universidad de Guadalajara
Guadalajara, Jalisco, Mexico,
Apartado Postal 1-2011, G.P. 44100, Guadalajara, Jalisco, Mexico
[email protected]

We consider fields in (D > 2)-dimensional spacetime, whose potential is r-form (skew-

symmetric tensor of rank '1'), the field tensor F being its exterior derivative and the
Lagrangian, a function of the quadratic invariant I of this tensor. It is shown that vector
field ('I' = 1) describes electromagnetic field only for D = 4. In particular, for D = 3 and
the Lagrangean L as any function of the above-mentioned invariant, the ('I' = I)-field has
energy-momentum tensor identical with that of a perfect fluid whose equation of state
depends on the choice of L(1).
Taking D = n + 1 (see Ref. 3) and supposing n to be an odd integer, we find
that there exist new natural (r of. I)-generalizations of the electromagnetic field.
These generalizations involve analogues of the dual conjugation of the field tensor
F interchanging electric and magnetic parts of the field, with its energy-momentum
tensor invariant under this conjugation. We consider such a field whose potential A
is r-form, denoting this fields's invariant quadratic in the field intensity F (which has
rank r + 1) as I. For L ~ I, the trace of the energy-momentum tensor identically
vanishes (as this was the case for D = 4, the Maxwell field), so that the field is
intrinsically relativistic (see also my talk on the equivalence principle in Session
GT4 where only D = 4 was considered, and Ref. 9). In these cases r = ~ - 1. For
D = 4 we have the usual electromagnetic field with its (not intrinsically relativistic)
nonlinear generalizations. D = 6 yields r = 2, and F is skew-symmetric tensor of
rank 3, etc.
As to the multidimensional generalizations of perfect fluids, the skew (axial =
pseudo-) potentials have ranks r = D - 2 and the field tensors are their exte-
rior derivatives. 7- 9 The equations of state are similar to those in D = 4, and the
propagation of sound in multidimensional fluids is easily described as behaviour of
perturbations in these fluids closely copying the situation in D = 4. For example,
there is the case of D-dimensional stiff matter where sound propagates with the
velocity of light like in D = 4. Moreover, in the special relativistic limit it is easy
to perform the second quantization of small perturbations on the background of
homogeneous fluid (this was done in the unpublished Master of Sc. thesis of my
student H. Vargas Rodriguez in 1998) yielding phonons. This occurs not only in
D = 4 case, but also in general, in particular in D = 3 which shows that the vector
field describes there perfect fluids and not electromagnetism (see final comments in
this talk).
The stress-energy tensor of an r-form field follows from the Noether theorem. 6 ,lO

1314
 1315

In general it takes the form

-r/3.-
et .- 55£ gJlV 1/a3 -= !j5£
gJlV
JlV g
JlV 1/3 _
g
a -
5£ 1)
5 ( IgI2(r+l) gfiV
(lgI2(r't-1) gJlv) I: '
~
(1)

since the Lagrangian density, as well as the function L = £/ rv ]k, depend vT.9I
I: r::
on gJlV only algebraically (the r-form potentials are considered to be independent
of the metric tensor), and (lgI2(r't-l) g"v) is the Trautman coefficientY Then the
intrinsically relativistic property condition = 0 yields

( lgI2(r't-l) gfiV) leta = IgI2(r't-1) gJlV (2 _ D

k(r+l)
) = 0 =} k = D
2(r+l)
. (2)

When k = 1, only space-times of even number of dimensions D can fit this condition:
D = 2(r + 1). The same condition determines the conformal invariance property of
the fields. Thus in the intrinsically relativistic case it is necessary and sufficient to
use the simplest nonlinear Lagrangian densities (see the Table 1),

k- D (3)
- 2(r + 1)"

Table 1. Values of k versus rand D describing intrinsically relativistic fields.

r\D 2 3 4 5 6 7 8 9 10 11 12
0 1 3/2 2 5/2 3 7/2 4 9/2 5 11/2 6
1 1/2 3/4 1 5/4 3/2 7/4 2 9/4 5/2 11/4 3
2 1/2 2/3 5/6 1 7/6 4/3 3/2 5/3 11/6 2
3 1/2 5/8 3/4 7/8 1 9/8 5/4 11/8 3/2
4 1/2 3/5 7/10 4/5 9/10 1 11/10 6/5
5 1/2 7/12 2/3 3/4 5/6 11/12 1
6 1/2 4/7 9/14 5/7 11/14 6/7
7 1/2 9/16 5/8 11/16 3/4
8 1/2 5/9 11/18 2/3
9 1/2 11/20 3/5
10 1/2 6/11
11 1/2

This Table simply gives values of k; since 0 S; r S; D -1, the lower left corner consists
of blank spaces only: the "missing" r-form field potentials are either trivially exact
ones, or equal to zero. D = 2 is introduced here, of course, only formally since the
left-hand ("geometric") side of Einstein's equations then identically vanishes.
In particular, these results yield a
Theorem: (Generalized) fundamental electromagnetic fields exist only in even-D-
dimensional spacetimes, then being (r = D /2 -I)-form fields (see boldface 1 's in the
Table). They possess all essential properties of the 3+1 Maxwell fields (are linear,
intrinsically relativistic, conformally invariant, and subject to the D-dimensional
dual conjugation relations).
1316

However, like in D = 4, one has to introduce a new fundamental field (alongside

with the electromagnetic one) for any D in order to describe rotating perfect fluids
(as well as incompressible matter like that filling the interior Schwarzschild solution).
We call it the Machian field since it also determines the cosmological term which
vanishes in the intrinsically relativistic case (phantom field). The Machian field also
has a skew potential, now with r = D - l. As we see in the Table 1 above, its k in
the intrinsically relativistic case is equal to 1/2.
Thus we consider two fields (Machian, r = D -1, and Maxwellian, r = D /2 -1)
as fundamental physical fields. All other r-fields in the Table seem to be of less
general importance; for example, the (r = D - 2)-field models perfect fluids in the
respective spacetimes, and its Lagrangian needs to be chosen as such a function of
the field invariant which yields the desired equation of state (thus the relativistic
property does not have a deep meaning in the purely phenomenological case of
fluids). A plausible conjecture then is that both fundamental r-form fields are bound
to be intrinsically relativistic (the free Machian field then yields the cosmological
constant A = 0 7,8). A peculiar property of perfect fluids in any D is that in their field
theoretical description the inhomogeneity term in the corresponding field equations
has the sense of rotation and not of a source as this is the case for electromagnetic
and gravitational fields. This fact mathematically follows from the definition of
rotation as u 1\ du, the polar covector u = *F/ /1/2 being four (or D-) velocity of
the fluid. Precisely this fact is completely overlooked in "standard" consideration
of (r = I)-fields in D = 3 where they are erroneously treated as "electromagnetic"
ones 1 ,2,4,5 (to mention only few of a vast number of publications).

References
1. M. Banados, M. Henneaux, C. Teitelboim and J. Zanelli, Phys. Rev. D 48, 1506
(1993).
2. S. Carlip, Lectures on (2+1)-dimensional gravity. gr-qc/9503024 (1995).
3. H. van Dam and Y.J. Ng, Why 3+ 1 metric rather than 4+0 or 2+ 2? hep-th/0108067
(2001).
4. A. Garcia Diaz, On the rotating charged BTZ metric. hep-th/9909111 (1999).
5. M. Kamata and T. Koikawa, 2+1 dimensional charged black hole with (anti-)self dual
Maxwell fields. hep-th/9605114 (1996).
6. N.V. Mitskievich (Mizkjewitsch), Ann. Phys. (Leipzig), 1,319 (1958). In German. See
a modernized version of the Noether theorem in Ref. 10.
7. N.V. Mitskievich, Int. J. Theor. Phys. 38, 997 (1999).
8. N.V. Mitskievich, Gen. Rei. Grav. 31, 713 (1999).
9. N.V. Mitskievich, Revista Mexicana de Fisica, Vol. 49 Suplemento 2, 39 (2003); see
also gr-qc/0202032.
10. N.V. Mitskievich, Relativistic Physics in Arbitrary Reference Frames (Nova Science
Publishers, 2006). See also the early book preprint arXive gr-qc/9606051 (without the
Noether theorem).
11. A. Trautman, Bull. Acad. Polan. Sci., ser. III 4, 665 & 671 (1956).
 TORSION INDUCES GRAVITY

RODRIGO AROS
Departamento de Ciencias Fisicas,
Universidad Andres Bello, Av. Republica 252, Santiago, Chile

The degrees of freedom of torsion are usually ignored in gravity. In this work it is shown
that those degrees of freedom allow to obtain known solutions in higher dimensional the-
ories of gravity which manifestly differ from Einstein theory. In particular 5 dimensional
Schwarzschild solution in recover within Gauss Bonnet (Chern Simon) theory.

Keywords: Gravity in five dimensions

1. Introduction

In higher dimensions the premise of second order equation of motion for the metric
does not restrict the action to EH. One the the simplest extension of EH action give
rise to what is called Lovelock gravities. 1-3 In odd dimensions there is a subfamily
of Lovelock gravities, known as Chern Simons gravities, that permits non vanishing
torsion solution in principle,4 although few are known.

2. Poincare
The Poincare Chern Simons action in five dimensions 6 reads

I = 1M R
ab
RCdel Sabcdf· (1)

yielding the equations of motion Rab Rcdsabcdl = 0 and RabTcsabcdl = O.

Considering a spherical symmetric solution one is led to the ansatz
1
eO = f(r)dt e = g(r)dr ern = riY", (2a)

W01 = c(r)dt wmn = wrnn ,

wOrn = a(r)ern w 1m = b(r)e rn , (2b)
where ern is a dreivein (with m = 2,3,4) for the spherical transverse section and
wmn is its associated Levi Civita (torsion free)
connection.
As boundary condition is imposed that M be asymptotically flat. This deter-
mines that the functions in Eq.(2) satisfy
f(r)x-+E!Eoc ---> 1, c(r)x--+E!E= ---> 0,
g(r)x--+E!E= ---> 1, a(r)x--+E!Eoc ---> 0, (3)
b(r)x--+E!Eoc ---> -1.
A solution can be found by imposing that a(T)2 = b(T)2 - 1 and

fer) = )1- r~,

r2
geT) = f(1')-l, (4)
r2 r3
c(1') = ~-
l' r
!, beT) = -f(T) +b2 (1'),

1317
1318

where

T+ -1
F(r)=re-;;:-(r+r+) .

It must be stressed that the vanishing of the torsion at o"L. eo determines that

1 d
l
eo
C1 = - dPF()df(P).
r=r+ P P

3. Negative cosmological constant

With a negative cosmological constant the equation of motion 6 are

R- abR-cd Sabcdj = 0 an d R-abT c

Sabcdj = 0, (5)

where Rab = Rab + l-2e a e b .

A solution asymptotically AdS with spherical symmetric is given by a(r)2
b(r)2 - 1 - ~;

2 ~ r2
f (r) = 1 - r2 + [2
(r-r+)(r+r+)
r2
g(r) = f(r)-l (6)
c(r) = r~
r3
(1 + r~) + ~ _ ~ (1 + 2r~)
[2 l2 r l2
b(r) = -f(r)+b2 (r)

with

(7)

where

and

F(r) _ rl (8)
- r2 + r~ + [2'
 1319

4. A horizon
The existence of a horizon in a space with vanishing torsion is determined by
geodesic curves structure. In general one can argue that r = r + is a genuine horizon
if there no light like outward curves connecting r < r + to r > r +. In both the solu-
tions above either parallel transport or geodesic light curves are parallel, therefore
in this case the presence of torsion is irrelevant for the definition of a horizon, thus
r = r + is a genuine horizon.

5. Conclusions and discussion

In this work is proven that Chern Simons gravity in five dimensions has solutions
that resemble Schwarzschild black holes. These solutions have non vanishing torsion,
but the standard geodesic structure of black hole is preserved, though. In both
solutions (Eqs. (4,6)) torsion vanishes fast enough, in particular for A < 0, that its
influence can be neglected in a region still far from lR x 82:,00, thus both solutions
represent interesting generalizations of Schwarzschild solution, preserving its general
behavior.

Acknowledgments
R.A. would like to thank Abdus Salam International Centre for Theoretical Physics
(ICTP) for its support. This work was partially funded by grants FONDECYT
1040202 and DI 06-04. (UNAB).

References
1. D. Lovelock, The einstein tensor and its generalizations, J. Math. Phys. 12 (1971)
498-501.
2. B. Zumino, Gravity theories in more than four-dimensions, Phys. Rept. 137 (1986)
109.
3. J. Zanelli, (super)-gravities beyond 4 dimensions,
http://arXiv.org/abs/hep-th/0206169.
4. A. H. Chamseddine, Topological gauge theory of gravity in five-dimensions and all
odd dimensions, Phys. Lett. B233 (1989) 291.
5. S. Aminneborg, 1. Bengtsson, S. Holst, and P. Peldan, Making anti-de sitter black
holes, Class. Quant. Grav. 13 (1996) 2707-2714, [gr-qc/9604005j.
6. M. Banados, R. Troncoso, and J. Zanelli, Higher dimensional chern-simons
supergravity, Phys. Rev. D54 (1996) 2605-2611, [gr-qc/9601003j.
 FINAL FATE OF HIGHER-DIMENSIONAL SPHERICAL DUST
COLLAPSE IN EINSTEIN-GAUSS-BONNET GRAVITY

HIDEKI MAEDA
Graduate School of Science and Engineering, Waseda University, Tokyo 169-8555, Japan,
Department of Physics, Rikkyo University, Tokyo 171-8501, Japan,
and
Department of Physics, International Christian University, 3-10-2 Osawa, Mitaka-shi, Tokyo
181-8585, Japan
[email protected]

We study the n(? 5)-dimensional spherically symmetric gravitational collapse of dust

in Einstein-Gauss-Bonnet gravity. The solution having a general relativistic limit is only
considered. The whole picture and the final fate of the gravitational collapse differ greatly
between the cases with n = 5 and n ? 6. We show that naked singularities are massless
for n ? 6, while massive naked singularities are possible for n = 5. In the inhomogeneous
collapse with smooth initial data, the strong cosmic censorship hypothesis holds for
n ? 10 and for n = 9 depending on the parameters in the initial data, while a naked
singularity is always formed for 5 <::: n <::: 8. This paper is based on Ref. 1.

1. Model and basic equations

We begin with the following n-dimensional (n 2:: 5) action:

S = JdnXrl[2~~ (R - 2A + aLOE)] + Smatter. (1)

The Gauss-Bonnet term LOE is given by

L OE -- R2 - 4R f.LV Rf.L V + R f.Lvpu Rf.LVpu . (2)
a is the coupling constant of the Gauss-Bonnet ternl. This type of action is derived
in the low-energy limit of heterotic superstring theory.2 In that case, a is regarded
as the inverse string tension and positive definite, and thus we assume a > 0 in this
paper. We consider a perfect fluid as a matter field, whose action is represented by
Smatter in Eq. (1). We do not consider the case with n ::; 4, in which the Gauss-
Bonnet term does not contribute to the field equations.
The gravitational equation of the action (1) is
Gf.L v + aHf.L v + Abf.L v = I\;;[(p + p)uf.L uv + pbf.L v], (3)
where
1
Gf.LV == Rf.Lv - "2gf.LvR, (4)

Hf.Lv == 2 [RRILV - 2R1L c>RC>v - 2Rc>(3 Rf.Lc>v(3 + R 1Lc>(3, R VC>(3,] (5)

Hereafter we assume p cf -fL·
We consider the spacetime as a product manifold Mn r:::; JvJ2 X Kn-2, where
K n - 2 is the (n - 2)-dimensional space of constant curvature, with the line element
ds 2 = _e 2<f>(t,r)dt 2 + e 2W (t,r)dr 2 + S(t, r)2rijdxidxj, (6)

1320
 1321

where lij is the unit metric on Kn-2. k denotes the curvature of K n - 2 and takes
1, 0, and -1. We adopt the co moving coordinates such that the n-velocity of the
fluid element is ul-' = (e-<I>,O,O,O).
Here we give a definition of a scalar on Af2 with the dimension of mass as
(n - 2)Vk
m == 2K:~
n-2 [_Asn-I + sn-3(k - S S'I-')
.1-'
+ aS n - 5 (k - S S'I-')2]
,/,'
(7)

where a == (n - 3)(n - 4)0, A == 2A/[(n - 1)(n - 2)], and a comma denotes the
partial differentiation. A constant V;_2 is the surface area of the (n - 2 )-dimensional
unit space of constant curvature if it is compact. In the four-dimensional spherically
symmetric case without a cosmological constant, m is reduced to the Misner-Sharp
mass.
Then, the field equations are written in the following simple form:
p' = -(p + p)<])', (8)

P= - (p + p) [\ji + (n - 2) ~1' (9)

m' = V;_2PS' sn-2, (10)

rn n-2P s'sn-2 ,
. = - Vk (11)
° = - Ei' + <])' 8 + \ji S', (12)
m = (n - 2)V;_2 [_Asn-I + sn-3(k + e- 2 <1> 82 _ e- 2'lI S,2)
2K:~
+ aS n - 5 (k + e- 2 <1> 82 _ e- 2'lI S,2)2], (13)
where a dot and a prime denote a/at and 0/01', respectively.

2. Master equation for the spherically symmetric dust

After this, we only consider the spherical collapse of a dust fluid without a cosmo-
logical constant, i.e., P = 0 and A = 0, for simplicity. Then, the system reduces to
the following master equation:

82 = S
j' _ 2;,
L<
(1 . .,
...,
8aK:~m) '
1 + (n - 2)VL2sn-I
(14)

where m = m(T) and f(T)(> -1) are arbitrary functions of T. The metric and the
energy density of dust are given by
,2
S
ds,2 -_ -dt 2 + ( )dT 2
l+f1'
+ S(t,r) l'Jdx i dx j ,
. 2 ..
'
(15)

m'
P = VI s,Sn-2' (16)
n-2
From Eq. (16), we find that there may exist both shell-crossing singularities,
where S' = 0, and shell-focusing singularities, where S = O. We assume the positive
energy density, i.e., P > 0.
1322

There are two families of solutions which correspond to the sign in front of
the square root in Eq. (14). We call the family with the minus (plus) sign the
minus-branch (plus-branch) solution. The minus-branch solution has the general
relativistic limit 0: --+ 0, while the plus-branch solution does not.

3. Results
We only consider the minus-branch solution and assume that (i) a > 0, (ii) m > 0,
(iii) S' > 0, and (iv) smooth initial data around the symmetric center.
Then, in generic collapse, it is shown that the singularity appeared in the final
state is censored for n ;::: 10, while a naked singularity is formed for 5 :::; n :::; 8. For
n = 9, it is censored depending on the parameters in the initial data. The naked
singularity is massless for n ;::: 6, while it is massive for n = 5.
In general relativistic case, the final singularity is censored for n ;::: 6. 3 The
present result implies that the effects of the Gauss-Bonnet term worsen the situa-
tion from the viewpoint of cosmic censorship rather than prevent naked singularity
formation. (See also Ref. 4.)
The formation of massive naked singularities in Einstein-Gauss-Bonnet grav-
ity for n = 5 is quite remarkable. In four dimensions with spherical symmetry, it
has been shown by Lake under very generic situations without using the Einstein
equations that massive singularities formed from regular initial data are censored
by adopting the Misner-Sharp mass as a quasi-local mass. 5 As was already pointed
out in Ref. 4, the higher-dimensional counterpart of the Misner-Sharp mass is not
an appropriate quasi-local mass in Einstein-Gauss-Bonnet gravity, and therefore
his result does not conflict with ours. In Einstein-Gauss-Bonnet gravity, our defi-
nition of the quasi-local mass is preferable to the higher-dimensional Misner-Sharp
mass. 1 Lake's result will be extended in Einstein-Gauss-Bonnet gravity for n ;::: 6
by adopting our quasi-local mass.

Acknowledgments
This research has been partially supported by the Grant-in-Aid for Scientific Re-
search Fund of the Ministry of Education, Culture, Sports, Science and Technology,
Japan (Young Scientists (B) 18740162).

References
1. H. Maeda, Phys. Rev. D73, 104004 (2006).
2. D.J. Gross and J.H. Sloan, Nucl. Phys. B291, 41 (1987); M.C. Bento and O. Bertolami,
Phys. Lett. B368, 198 (1996).
3. A. Banerjee, U. Debnath, and S. Chakraborty, Int. 1. Mod. Phys. D12, 1255 (2003);
K.D. Patil, Phys. Rev. D67, 024017 (2003); R. Goswami and P.S. Joshi, Phys. Rev.
D69, 044002 (2004); Phys. Rev. D69, 104002 (2004).
4. H. Maeda, Class. Quantum Grav. 23,2155 (2006).
5. K. Lake, Phys. Rev. Lett. 68, 3129 (1992).
 CLASSIFICATION OF THE WEYL TENSOR IN HIGHER
DIMENSIONS AND ITS APPLICATIONS*

VOJTECH PRAVDA
Mathematical Institute
Academy of Sciences
Zitna 25
115 67 Prague 1
Czech Republic
[email protected]

The algebraic classification of the Weyl tensor in higher dimensions is a natural gener-
alization of the Petrov classification in four dimensions. We overview basic properties of
the classification and discuss various applications.

Recently a higher dimensional generalization of the well known Petrov classifica-

tion of the Weyl tensor has been developed. 1 ,2 This generalization leads to a dimen-
sion independent classification scheme for dimensions D > 4 and it also reproduces
the Petrov classification for D = 4. The classification is based on the (possible) exis-
tence of preferred null directions - Weyl aligned null directions (WANDs a ) and their
multiplicity (related to order of alignment) for a given spacetime. One can also say
that the Weyl tensor is of alignment type 1 (n, m) if the spacetime possesses two dis-
tinct WANDs of mUltiplicity nand m, respectively. All possible cases for the Weyl
tensor are summarized in the following Table. We refer to 1 ,2 for rigorous definitions
and to 3 for a brief overview of the classification. Many higher-dimensional metrics
are also classified in. 4 In this extended abstract we rather focus on simply listing
various properties of the classification, applications and appropriate references.

D>4 dimensions 4 dimensions

Weyl type alignment type Petrov type
G G
I (1)
Ii (1,1) I
II (2)
IIi (2,1 ) II
D (2,2) D
III (3)
IIIi (3,1) III
N (4) N

*This research has been partially supported by research plan No AVOZlO190503 and research grant
KJBlO19403.
aln four dimensions WANDs correspond to principal null directions of the Weyl tensor.

1323
1324

Properties of spacetimes in higher dimensions that are identical or similar to the

four-dimensional case:

• It was shown in 5 that static spherically symmetric spacetimes are "boost

invariant" (this corresponds to the type D) in arbitrary dimension. This in
fact also holds in the non-static case. 6
• Isolated horizons of higher dimensional black holes are locally of type II. 7
• All static spacetimes in arbitrary dimension are of types G, Ii or D.6 Explicit
examples for each case are charged black ring of type G,8 vacuum black ring
of type I i g and Myers-Perry black hole in five dimensions of type D. 1O ,12
• Peeling theorem holds (under certain assumptions) in arbitrary even di-
mension. 13
• A spacetime in arbitrary dimension has vanishing curvature invariants (VSI
spacetime) if and only if it admits an aligned shearfree, non-expanding,
non-twisting, geodetic null congruence along which the Riemann tensor has
negative boost order (Weyl tensor is thus of type III, Nor 0).

Properties of spacetimes in higher dimensions that are distinct from the four-
dimensional case:

• In higher dimensions there are spacetimes with no WANDs (type G).1.2,8

• All expanding vacuum spacetimes of types III and N in higher dimensions
are necessarily shearing. 10 Furthemore in odd dimensions twisting geodetic
WAND is also shearingY Therefore straightforward generalization of the
Goldberg-Sachs theorem to higher dimensions is not possible. b
• One consequence of the previous point is that the class of Robinson-
Trautmann spacetimes (defined as spacetimes admitting hypersurface or-
thogonal, non-shearing and expanding geodetic null congruence) in four
dimensions is much richer than in higher dimensions. In fact Robinson-
Trautman spacetimes in higher dimensions are only of type D 15 while in
four dimensions in addition to type D spacetimes there are also classes of
type N, type III and type II solutions.
• It was also pointed out 16 ,17 that VSI spacetimes admitting a covariantly
constant null vector field are of type III or N in higher dimensions while in
four dimensions they are necessarily of type N.

As we have seen, some properties of algebraically special higher dimensional

spacetimes are similar to their four dimensional counterpart, while other differ. At
this stage many related questions remain unanswered and further work is necessary
for complete understanding of the differences between the four dimensional and
higher dimensional case.

bGoldberg-Sachs theorem in four dimensions implies that algebraically special vacuum spacetimes
are shearfree.
1325

References
1. R. Milson, A. Coley, V. Pravda, A. Pravdova, Alignment and algebraically special
tensors in Lorentzian geometry, Int. J. Geom. Meth. Mod. Phys. 2 (2005) 41
2. A. Coley, R. Milson, V. Pravda, A. Pravdova, Classification of the Weyl Tensor in
Higher Dimensions, Class. Quantum Grav. 21 (2004) L35
3. V. Pravda, On the algebraic classification of spacetimes, J. Phys. Conf. Ser. 33 (2006)
463 [gr-qc/0512087]
4. A. Coley, N. Pelavas, Classification of Higher Dimensional Spacetimes, Gen. ReI. Grav.
38 (2006) 445
5. G. T. Horowitz, S. F. Ross, Properties of naked black holes, Phys. Rev. D 57 (1998)
1098
6. V. Pravda, A. Pravdova, M. Ortaggio, in preparation
7. J. Lewandowski, T. Pawlowski, Quasi-local rotating black holes in higher dimension:
geometry, Class. Quantum Grav. 22 (2005) 1573
8. M. Ortaggio, V. Pravda, Black rings with a small electric charge: Gyromagnetic ratios
and algebraic alignment, JHEP 0612 (2006) 054
9. V. Pravda, A. Pravdova, WANDs of the black ring, Gen. ReI. Grav. 37 (2005) 1277
10. V. Pravda, A. Pravdova, A. Coley, R. Milson, Bianchi identities in higher dimensions,
Class. Quantum Grav. 21 (2004) 2873
11. M. Ortaggio, V. Pravda, A. Pravdova, Ricci identities in higher dimensions, Class.
Quantum Grav. 24 (2007) 1657
12. V. Frolov and D. Stojkovic, Particle and light motion in space-time of a five-
dimensional rotating black hole, Phys. Rev. D 68 (2003) 064011
13. A. Pravdova, V. Pravda, A. Coley, A note on the peeling theorem in higher dimensions,
Class. Quantum Grav. 22 (2005) 2535
14. A. Coley, R. Milson, V. Pravda, A. Pravdova, Vanishing Scalar Invariant Space times
in Higher Dimensions, Class. Quantum Grav. 21 (2004) 5519
15. J. Podolsky, M. Ortaggio, Robinson-Trautman spacetimes in higher dimensions, Class.
Quantum Grav. 23 (2006) 5785
16. A. Coley, A. Fuster, S. Hervik, N. Pelavas, Higher dimensional VSI spacetimes, Class.
Quantum Grav. 23 (2006) 7431
17. A. Coley, A. Fuster, D. McNutt, N. Pelavas, in preparation
This page intentionally left blank
Geometric Calculus in
Gravity Theory
This page intentionally left blank
 GEOMETRIC AND KINEMATICAL ASPECTS OF RINDLER
OBSERVERS

C. ROMERO and J. B. FORMIGA

Departamento de Fisica, Universidade Federal da Paraiba,
Caixa Postal 5008, loao Pessoa, Pb, 58059-970, Brazil
[email protected]

We discuss some aspects of the differential geometry of timelike curves in Minkowski

spacetime and look at Rindler particles as a particular case of a more general class of
accelerated motion. We then point out that Rindler observers can in their proper frame
detect redshift and time delay effects, which provides a clear and simple illustration of
the principle of equivalence.

1. The Serret-Frenet equations in Minkowski spacetime

To extend the Serret-Frenet formalism to Minkowski spacetime we need to replace
the Euclidean metric by Minkowski metric TJa(3 = diag(l, -1, -1, -l)a, define a
second binormal and a second torsion T2 = T2 (s) . We define our set of orthonormal
four-vectors (a tetrad) by concomitantly requiring them to satisfy a four-dimensional
extension of the Frenet-Serret equations, which then governs the evolution of the
tetrad. 6 It is also convenient to restrict ourselves to timelike curves x a = XC'(s), i.e.
;" = 1, where s denotes the arc lengt h parameter in the
dx" d (J
those for which TJa(3d;;-
sense of Minkowski metric TJa(j. Accordingly, if we denote the tetrad vectors by urAl
(A = 0, ... 3) b, then the orthonormality conditions read ufAl U(B)a = TJa(3UfA) UfB) =
,
TJAB. It can be shown l that if we chose Ufo) = d~:;' i.e. Ufo) being the components
of the unit tangent vector , then we can easily construct an orthonormal basis of
vectors {urA)} ,defined along the curve, which obeys the following four-dimensional
Serret-Frenet equations, given in matrix representation by C

(1)

aWe follow the convention in which Greek indices range from 0 to 3, and use the so-called Einstein
summation rule, i.e. in the case of repeated index summation over the appropriate range is implied.
b From now on Latin capital indices (A, E, ... ) are labels for distinguishing the particular vector of
the tetrad, while (a, p, ... ) denotes its components. For convenience, however, the vectors of the
canonical coordinate basis will be denoted by e( Dc).
C We know that in Minkowski spacetime vectors which are orthogonal to timelike vectoTs aTe space-

like. Thus, as ----cr:s-

du(o)
is orthogonal to U(O), then ----cr:s-
du(O) .
is spacelike. We convemently define'
k( s )
d n i (I

to be positive and given by k(s) == (- '7c.[3 ~l~J) ':~O) )-1/2 We also conventionally define 71 to
be non-negative.

1329
1330

2. The fundamental theorem of curves in Minkowski spacetime

A most important result in the local theory of curves in Euclidean space R 3 , known
as the fundamental theorem of curves, may be straightforward extended to Minkowski
spacetime and states the following:
Theorem 1 Given differentiable functions k(8) > 0, 71(8) and 72(8),
there exists a regular parametrized time like d curve r such that k( 8) is the
curvature, 71 (8) and 72 (8) are, respectively, the first and second torsion of
r
r. Any other curve satisfying the same conditions, differs from r by a
Poincare transformation, i. e. by a transformation of the type X'M = A~vxv
+ aM , where A~v represents a proper Lorentz matrix and aM is a constant
four-vector.
Other extensions of known theorems of the differential geometry of curves in
Euclidian space R3 can easily be carried over into Minkowski spacetime. As an
example, let us consider the following theorem:
Theorem 2 A time like curve r, with non-vanishing curvature, lies in
a hyperplane if and only if the second torsion vanishes identically.
For the sake of space we omit the proofs of the above theorems and refer the
reader to 3 •

3. Torsionless timelike paths in Minkowski spacetime

Let us now turn out attention to the case of torsionless timelike paths in Minkowski
spacetime. For simplicity, let us choose our coordinate system such that the motion
takes place in the (x O, xl )-plane. Then, the Serret-Frenet equations yield d~~o) =
k U(l), d'S - k u(O), d'S
dU(1) - dU(2) - dU(3) -
- d'S - . °
Choosing U(2) and U(3) as the usual unit vectors e(2) and e(3) in the x 2 , x 3 _
directions, respectively, it remains to solve the two-dimensional system of differential
equations for u(O) and U(l)' As it can easily be verified, the general solution of the
Serret-Frenet equations is given by

X"(8) = (1 8

cosh(8(8))d8 + a, 1
8

sinh(8(8))d8 + b, 0, 0) , (2)

where the function 8(8) is given in terms of the curvature k = k( 8) by 8(8) =

J; k( 8)d8 + cp, with a, band cp being arbitrary constants.
Let us consider the case of constant curvature. If we choose, as initial conditions
of the motion, x"(o) = (0,1,0,0) and U CO )(8) = (1,0,0,0) then from (2) we have
X"'(8) = (cosh(k8), sinh(ks), 0, 0), which describes a hyperbolic motion in Minkowski
spacetime, and may be interpreted as a Rindler observer, i.e. an accelerated particle

d Indeveloping the Serret-Ji'renet formalism in Minkowski space we are ultimately interested

in applications to the motion of physical particles. Therefore we shall restrict ourselves to
timelike curves, although the formalism may also work for spacelike curves. Considemtion
of null curves should introduce some difficulty.
 1331

whose proper acceleration (i.e. acceleration relative to its instantaneous rest frame)
is constant.
Rindler observers deserve a detailed analysis as far as their kinematical proper-
ties are concerned. Due to lack of space we shall just mention two of these properties
and refer the reader to (5).

4. Kinematical properties of Rindler observers

At least two optical phenomena that in the context of the general theory of relativity
are usually associated with the presence of gravitational fields may also appear to
a set of Rindler observers. These are the spectral-shift and the time delay of light.
Being essentially geometrical, these effects have an interpretation in terms of the
bending of the spacetime due to the presence of matter. Nevertheless, they also
may appear in fiat Minkowski spacetime as a purely kinematical effect when we have
accelerated observers. This should not surprise us since from the Einstein's principle
of equivalence gravitational fields can be mimicked by non-inertial reference frames.
In ref.(5) we show how Rindler observers would detect and measure the spectral and
time delay of light. The analysis can be easily carried out by defining the so-called
Rindler coordinates (r,~) and applying the usual techniques of special relativity
theory.

Acknowledgments
C. Romero acknowledges CNPq and FAPESQ-PBjCNPq(PRONEX) for partial fi-
nancial support.

References
1. C. Lanczos, Space through the Ages (Academic, London, 1970).
2. See, for instance, M. P. do Carmo, Differential geometry of curves and surfaces,
Prentice-Hall, New Jersey, 1976, Ch.1.
3. J. B. Formiga and C. Romero, Am. J. Phys. 74, 1012 (2006).
4. E. Honig, E. Schucking and C. Vishveshwara, J. Math. Phys. 15, 774 (1974).
5. J. B. Formiga and C. Romero, On the spectral shift and the time delay of light in a
Rindler accelerated frame, Int. J. Mod. Phys. (to appear)
6. See, for example, J. L. Synge and A. Schild, Tensor Calculus, Dover, New York, 1978,
Ch.2
 ON THE ZEROS OF SPINOR FIELDS AND AN
ORTHONORMAL FRAME GAUGE CONDITION

JAMES M. NESTER
Department of Physics and Institute of Astronomy, National Central University,
Chung Ii, Taiwan 32054, R. O. C.
E-mail: [email protected]

We proposed (in connection with a positive energy proof/localization) certain rotational

gauge conditions for a preferred orthonormal frame on a Riemannian manifold. For cer-
tain physically interesting cases (dimensions 2, 3 and 4) such frames can be constructed
from spinors (regarded as rotation-dilations) which solve the Dirac or Witten equations-
with the important exception of points at which the spinor field vanishes. Are such zero
points a serious problem? Here, using explicit examples, we show that spinor zeros really
can happen. However we also argue that solutions with zeros are quite exceptional; for
generic metrics and boundary data there are no zeros. We conclude that the "special
orthonormal frame" gauge conditions can be used essentially without reservations.

Keywords: Spinor Zeros; Special Orthonormal Frame.

1. The SOF gauge condition

Given a parallelizable Riemannian manifold, one can choose an orthonormal
(co ) frame field {ja (dual to a frame eb). The choice is not unique: {jb' = R b' a{P'
We proposed in connection with certain gravitational energy investigations 1 the
Special Orthonormal Frame (SOF) rotational gauge condition. 2 ,3
A one-form and a 3-form characterize the rotational state of a frame:

ij := ea.J diP == b
-raba-o , q := (lj2)o ab -o a 1\ d-o b == ~r[abe]-oa 1\ -ob 1\ -oe. (1)
(i.e. the trace and anti-symmetric part of the orthonormal frame connection coef-
ficients). The SOF condition requires ij to be closed and q co-closed. Equivalent
(modulo cohomology) to the two integrability conditions:
dij = 0 = oq. (2)
which combine to a 2nd order equation dij+oq = 0 (0 is the co differential, the adjoint
of d: 0 ::: ± * d*) a quasi-linear 2nd order elliptic (hyperbolic) PDE system for a
rotation taking an initial frame to an SOF. From simple linearization arguments
one concludes that unique solutions should exist at least for nearly fiat metrics.
Dimakis & Miiller-Hoissen4 noted a simple connection between the non-linear SOF
condition and the linear 3D Dirac equation. The key: factor the spinoI' (recall the
Hestenes spinor as a rotation-dilation) into a magnitude and an SU(2) rotation.
The idea extends: 3 in any dimension the Dirac operator explicitly contains only
the rotational state parts of the connection:

I aD ao/'-
ni,._ a ( ni,
I eao/ + red a:4l[cld]o/
1 ni,) _ ani,
- I eao/
+ 21 r a bal bni<jJ:4
+ 1 r[abe] I[alble] ni,0/. (3)
In 3 and 4 D this provides a direct connection between a spinor field (via a unitary
transformation and a rotation) to an SOY-except where the spinor vanishes.

1332
 1333

2. Spinor zeros
The SU(2) element, and hence the rotation, is not well defined at points where
the spinor field vanishes. We know of only one published work 5 aimed at proving
that under certain conditions spinor zeros could not occur, however the argument
presented there has problems.
Consider a spinor field on a 3-manifold. Locally it can be represented by a 2D
complex vector. It vanishes only at points where both complex components, and
hence 4 real function all vanish. Generically the zero sets of 4 real functions will not
have any common points. The spin or field of interest, however, satisfies a certain
equation, so its components are not independent.
We note certain solutions to the 3-dim (massless) Dirac equation akEh'ljJ = on °
flat ]R3 that have zeros (the a k are the Pauli matrices). Consider the special class
o
of solutions with 1/J = am m ¢1/Jo (where 1/Jo is a constant spinor) and ¢ satisfies the
Laplace equation. ¢ can have saddle type critical points in the interior of a region.
At such points 1/J vanishes.
One instructive solution to the Dirac equation on ]R3 is

1/J = ( 2z + iE ). (4)
-(x + iy)
The corresponding SU(2) element obtained from'l/J = 11/JIU(6) has the form
ex .. ex k
U = cos-lI+zsm-nka , (5)
2 2

cos ~ = 2z/I1/JI, sin ~nl = -y/I1/JI, sin ~n2 = x/I1/JI, sin ~n3 = E/I1/JI, (6)

where 11/J12 = 4z2 + E2 + x 2 + y2. The associated SO(3) rotation has axis nk and
magnitude ex; it gives a solution to the original gauge conditions.
°
For E = the spin or field vanishes at (0,0,0). The associated SU(2) and SO(3)
rotations are not defined there, the limit as one approaches (0,0,0) does not exist.
°
As E passes from small negative values through to small positive values, the spinor
field undergoes a very small change, but the associated SU(2) and SO(3) elements
exhibit unstable behavior because of the amplification factor 11/J1- 1 . For most values
the rotation is stable. For small values of E it is not stable.
Other examples have several zeros. It seems that for any region generic boundary
data is ok, but there will be a discrete set of boundary data for which the gauge
conditions will have no solution, and near these particular boundary values the
behavior will be unstable. An analogue is the geodesic equation on a sphere. The
solution is not unique if the two given boundary points happen to be antipodes.
A small perturbation of the end points gives a small perturbation of the geodesic
except when the points are nearly antipodes. Another analogue is to consider a
frame aligned along the principle axes of a nearly spherically symmetric ellipsoid.
Exactly at spherical symmetry there is no unique solution. For nearby shapes small
perturbations of shape can cause large changes in the frame.
1334

The above example diverged at spatial infinity. From ¢ = z + l/r we found

'ljJ = ( 1 - z / r3 ). (7)
-(x + iy)/r3
This spinor field is constant at infinity but has a zero at (0,0,1). Generalizing to
the Schwarzschild geometry, we find a spinor solution to the Witten equation with
a zero point. In isotropic coordinates a co-frame is f)o = Ndt, f)k = rp 2 dx k , where
rp = 1 + m/2r. The spinor field
2 2 3
'ljJ -2 (1+iE-b m z/r ) (8)
= rp -b 2m 2(x + iy)/r3
is an asymptotically constant solution to the Witten equation. For E = it has a °
zero at (x,y,z) = (O,O,bm). The zero is in the "outer" r > m/2 region for b> 1/2.
For very large b the zero is in the asymptotic "weak field" region.

3. General conclusion
What happens here for the SOF conditions and the spinor zeros on flat IR.3 should
happen more generally. For any orthonormal frame gauge condition, for a given
region in a Riemannian manifold, unique solutions should exist for generic bound-
ary values. However for certain boundary data there will be no solution. A small
perturbation of the boundary data (or the metric) suffices to remove the difficulty,
but near these special choices the behavior will be unstable.

Acknowledgements
This work was supported by the National Science Council of the R.O.C. under
the grants NSC 94-2112-M-008-038 and NSC 95-2119-M008-027 and in part by
National Center of Theoretical Sciences and the (NCU) Center for Mathematics
and Theoretical Physics. Discussions over the years with L. B. Szabados, Y. K. Lau,
X.-n. Wu, R. Bartnik and C.-M. Chen were very helpful and much appreciated.

References
1. J. M. Nester, Int. J. Mod. Phys. A 4,1755 (1989); Phys. Lett A 139,112 (1989); Class.
Quant. Grav. 8 L19 (1991).
2. J. M. Nester, J. Math. Phys. 30, 624 (1989).
3. J. M. Nester, J. Math. Phys. 33, 910 (1992).
4. A. Dimakis and F. Miiller-Hoissen, Phys. Lett. A 142, 73 (1989).
5. V. Pelykh, J. Math Phys. 41, 5550 (2000).
NEW SPECIAL SOLUTIONS OF THE RICCI FLOW EQUATION IN
TWO DIMENSIONS USING A LINEARIZATION APPROACH*

ANCA VlSlNESCU and MIHAl VlSlNESCU

Department of Theoretical Physics,
Institute for Physics and Nuclear Engineering,
Magurele, P. a.Box MG-6, Bucharest, Romania
avisin, [email protected]

Two dimensional Ricci flow equation arises in many areas of physics including lower-
dimensional spacetimes models for gravity and cosmology. The linearization approach
proves a powerful procedure to generate explicit solutions of the Ricci flow equation. We
use a direct non-linear substitution, then split the resulting non-linear equation. On the
other hand, adding supplementary assumptions connected with the symmetries of the
concrete physical systems it is possible to get effectively large class of solutions. Some
special reductions are also discussed.

1. Introduction

Ricci flow is an important geometric evolution equation in Riemannian geometry. It

was introduced by R. Hamilton in 1982 1 for producing Einstein metrics of positive
scalar curvature and constant positive sectional curvature.
The Ricci flows are second order non-linear parabolic differential equations for
the components of the metric g/.1.1/ of an n-dimensional Riemannian manifold which
are driven by the Ricci curvature tensor RJ.Lv:

(1)

The Ricci flow equations arose independently in physics in the early '80s. Since
then they have become a major tool for addressing a variety of problems in the
quantum theory of fields, gravitation and cosmology.
In two dimensions, the Ricci flow equation written in a local system of confor-
mally flat coordinates has been studied in detail by Bakas 2 ,3 from an algebraic point
of view. It was considered as a continual version of the general Toda-type equation
for a given Lie algebra. It was found a formal power series solution by expand-
ing the path-ordered exponentials. Although the proposed general solution provides
a formal complete solution for the Ricci flow equation, its form is quite intricate
and difficult to handle. In Ref. 4 we used a direct non-linear substitution, then split
the resulting non-linear equation. This results in a class of special solutions, which
recover practically all known solution in the literature. On the other hand, adding
supplementary assumptions connected with the symmetries of the concrete physical
systems or assuming special dependence on parameter of deformation it is possible
to get effectively large class of solutions.

*This research has been partially supported by a CNCSlS (Romania) grant.

1335
1336

2. Linearization approach
In what follows we shall consider in a 2-dimensional space a local system of confor-
mally flat coordinates in which the metric has the form

ds; = ~e<P(x,y;t)(dx2 + dy2) = 2e<p(z+,z-;t)dz+dz - (2)

2
using Cartesian coordinates x, y or the complex conjugate variables 2z± = y ± ix.
Having in mind that the only non-vanishing component of the Ricci tensor is
(3)
the Ricci flow equation (1) becomes

:te<P(z+,z-;t) = o+(L<I>(z+,z-;t). (4)

It is useful to use the substitution
v(z+, z_; t) = e<p(z+,z-;t) (5)
writing Eq. (4) in the form
(6)
Starting with (6) and using the substitution v = rpz+, after integrating once with
respect to z+ we get
(7)
where C should be a function of z_ and t. Now making the substitution rp = In F
we will end up with the following quadratic equation:

(8)
Defining

(9)

then (7) will have the form

1/Jz+1/Jt = 'I/)Z+2_ . (10)

Using the same nonlinear substitution '1/) = InF we get

Fz+ (Ft + Fz-) = F Fz+z- . (11)

Eq. (11) can be split in some linear or nonlinear solvable equations. Of course,
all the possibilities we are going to analyze will give only special solutions and not
general ones.
The most interesting splitting of Eq. (11) into a system of linear equations is
the following:

Fz+z- = 0
Ft + Fz- = 0 (12)
 1337

with the general solution

(13)
where 1, g are arbitrary functions. The solution of (6) is
1'(z+ )
u(z+,z-;t) = (14)
f (z+) + g (t - z- )
which is in fact the generalization of the multi-shock solution.
Other linearizations of this type arc presented in Ref. 4
A different approach to Eq. (11) can be done using a special combination of the
variables z+ and z_ as a new independent variable. For example the well known
solutions of cigar-type, or rational type, discussed in 3 can be obtained by introducing
the following substitution:

(15)

Then (6) becomes

Vt = (In v)~~. (16)

which has been extensively studied in. 5

Finally, let us remark that in all the above searches for solutions we did not
assume a particular dependence on the parameter t. However, looking for a special
dependence on t, we are able to produce new solutions of special interest. 6
We also note for completeness some recent applications of the Ricci flow equa-
tions in the context of General Relativity?-9

References
1. R. Hamilton, J. Diff. Geom. 17, 255 (1982).
2. 1. Bakas, JHEP 0308,013 (2003).
3. 1. Bakas, Fortsch. Phys. 52,464 (2004).
4. A. S. Carstea and M. Visinescu, Mod. Phys. Lett. A 20, 2993 (2005).
5. P. Rosenau, Phys. Rev. Lett. 74 (1995), 1056.
6. A. Visinescu and M. Vi"inescu, in preparation.
7. W.Graf, gr-qc/0602054.
8. M. Headrick and T. Wiseman, hep-th/0606086.
9. S.Vacaru and M. VisinesclI, gr-qc/ /0609085; Int. J. Mod. Phys. A, (in press).
This page intentionally left blank
 Black Hole and Pair
Creation in Strong Fields
This page intentionally left blank
 PAIR CREATION IN INHOMOGENEOUS FIELDS

CHRISTIAN SCHUBERT
Instituto de Fisica y Matematicas, Universidad Michoacana de San Nicolas de Hidalgo,
Edificio C-3, Ciudad Universitaria, C.P. 58040 Morelia, Michoacan, Mexico
[email protected]

It has been known since the early days of QED that spontaneous electron-positron pair
creation from vacuum should occur in a strong electric field. A simple formula for the
pair creation rate in the constant field case was obtained in 1951 by Schwinger using
effective action methods. However, many potential applications such as in laser or heavy
ion physics involve electric fields which are strongly inhomogeneous. I will present a
recently developed approach to this problem based on a stationary phase approximation
of Feynman's relativistic worldline path integral representing the one-loop QED effective
action in the field.

1. Pair creation in electric fields: overview

As is well-known, QED predicts the creation of electron-positron pairs from vacuum
in electric fields of sufficient strength and extent. Schwinger showed in 1951 1 that
the pair production rate relates directly to the imaginary part of the QED effective
action in the field, Pproduction = 1 - e- 2Imr . For the case of a constant electric
field, and in the one-loop approximation, Schwinger was able to obtain the effective
Lagrangian in explicit form,

m
ImLspiuor(E) = 87r3f3
4
2 ~
~
1
n2 exp
[7rn]
-71 (1)

(f3 = eElm 2 ). Here the nth term relates to the coherent production of n pairs by
the field. The appearance of the field strength in the denominator of the exponent
indicates that the pair creation effect is nonperturbative in the field. It also suggests
an interpretation as a tunneling process in which a virtual electron-positron pair
separates out along the electric field lines and extracts a sufficient amount of energy
from the field to turn real. The corresponding formula for Scalar QED differs only
by signs and a global factor: 1

ImLscalar(E)
m 4
2
= -167r3f3 ~ ~ exp
(Xl (_l)n [7rn]
-71 (2)

From (1), (2) it is clear that the production rate is exponentially small for E below
a critical value Ecrit = m 2 Ie. For the case of spinor QED with m the electron
mass, this value is Ecrit = 1.3 X 10 18 V 1m. Creating a macroscopic electric field of
this strength was considered hopeless just a few years ago. However, due to recent
advances in laser technology it seems now conceivable that pair production might
be seen in laser fields in the near future. The optical laser POLARIS is projected
to reach a field strength of Emax ::::0 2 X 10 14 V Im,2 and the X-ray free electron laser
TESLA may even reach Emax ::::0 1.2 X 10 16 V 1m. 3

1341
1342

For realistic laser fields the constant field approximation is not usually justified;
it would be preferable to have generalizations of Schwinger's formula (1) to inho-
mogeneous and time varying fields. This is a subject on which important progress
has been made over the years by many authors (see, e.g.,4-11). Still, considering
this large body of work there is a surprising dearth of variety in the calculational
approaches; apart from a few special field configurations, all the cited works are
using WKB, or some variant of it. The "worldline instanton" technique which I will
sketch in the following is similar in spirit to WKB, but technically quite different,
and apparently more general. Its basic idea was proposed already in 1982 by Affleck,
Alvarez, and Manton,12 but applied only to the constant field case in scalar QED.
The generalization to spinor QED and to general electric fields was developed only
recently. 13-15
2. The worldline instanton technique
The worldline instanton approach is based on Feynman's "worldline representation"
of the one-loop effective action r[AJ. For Scalar QED, this representation is 16

r scalar [AJ -100

-
o T
dT e _m 2T J
V X e - foT dT (i x2 +ieA-x) (3)

Here T denotes the proper-time of the scalar particle in the loop, m its mass,
and IX(T)=x(O) Vx( T) a path integral over all closed loops in spacetime with fixed
periodicity x(T) = x(O). The representation (3) provides an alternative to standard
second-quantized field theory which has gained some popularity only during the
last fifteen years. Presently a number of different ways of evaluating this type of
path integral are in use. This includes the "string-inspired' formalism", based on
worldline correlators (see 17 and refs. therein), and the numerical "worldline Monte
Carlo formalism" .18 The approach initiated by Affleck et al. 12 is less general, since
it involves an approximation which is reasonable only for the imaginary part of the
effective action. They proposed to do both the T integral and the path integral in
(3) by a stationary phase (semiclassical) approximation. The stationarity condition
for the path integral simply yields the classical Lorentz force equation,

(4)
with a free parameter a to be adjusted so as to fulfill the periodicity condition. Such
a periodic extremal action trajectory xc! will be called a "worldline instant on" in the
following. Once the instanton is found, determining the semiclassical approximation
involves calculating the instanton action, and the determinant of fluctuations around
the instanton. The final result reached in 14 for the semiclassical approximation to
the path integral at fixed T is the following (see I3 ,14 for the details),

Idet[7]1~iree(T)] I
(5)
Idet [7]<;) (T)] I
 1343

In this formula, the exponential factor involving the worldline action evaluated on
the worldline instanton is familiar for a semiclassical approximation. Less obvious
is the form of the prefactor, which relates to the determinant of the operator of
fluctuations around this stationary trajectory. Although this determinant initially
is an infinite-dimensional one, in 14 it was, using a theorem by Levit and Smilansky,19
reduced to a 4 x 4 determinant of zero modes TJ1>-') (T) of this operator, normalized
by the corresponding free determinant. The factor eil} relates to the Morse index
of the same operator. The final T integral can again be done by a stationary point
approximation. It can be seen that this double stationary point approximation also
corresponds to a weak-field approximation. 13 Although it is not possible in general
to find a closed-form expression for xc! (T), the rhs of (5) is well-suited for a numerical
evaluation 15 since both the stationarity condition (4) and the equations determining
the zero modes are ordinary differential equations.

3. Examples
Constant field: The case considered by Affleck et al. 12 was a constant E field in
the z direction. They showed that for this case there is a discrete set of worldline
instantons which are circles in the z ~ t plane, parametrized by a winding number
n, with a field strength dependent radius,

(6)

The worldline action evaluated on the nth instanton produces just the nth term
in Schwinger's formula (2). Moreover, they were able to obtain the prefactor by a
direct calculation of the spectrum of the fluctuation operator.
Planar fields: The worldline instanton equations (4) are particularly easy to
solve for planar fields, defined by vector potentials of the form (0,0,A 3(X4),0))
or (0,0,0, A4(X3)). Particularly nice examples in this category are the single-pulse
time dependent electric field E(t) = Esech 2 (wt) and the analogous single-bump
space dependent electric field E(z) = Esech 2 (kz). In both cases the worldline in-
stantons can be found explicitly in terms of trigonometric functions,13 leading to
simple explicit formulas for the semiclassical actions. Fig. 1 shows the form of the
instanton trajectories in the (X3, X4) plane for various values of the "adiabaticity
parameter" "y == mwjeE resp. the "inhomogeneity parameter" l' == mkjeE.
Note that the single-pulse trajectories shrink for increasing values of the adia-
baticity parameter, while the single-bump trajectories grow with increasing inho-
mogeneity. We believe that this is just an instance of a general fact: Inhomogeneity
in time tends to shrink the size of the wordline instantons, leading to an increase
in the pair production rate; spatial inhomogeneity increases the instanton size and
decreases the pair production rate. The single-bump instantons cease to exist for
l' > 1, meaning that for insufficient spatial extension of the bump there is no pair
production at all, at least in this semiclassical approximation.
The single-bump field also allows us to test the validity of the semiclassical
approximation, since for this case an exact integral representation has been obtained
1344

~l

'( ~ 0.1

Fig. 1. Plot of the worldline instanton paths in the (X3, X4) plane for the single-pulse case (left
panel) and the single-bump case (right panel). Xa,4 have been expressed in units of m/eE.

by Nikishov,6 suitable for numerical evaluation, and moreover there is a worldline

Monte Carlo result. 2o As shown in fig. 2, all three results are in close agreement
over the whole range of the inhomogeneity parameter :y.14

0.2 0.4 06

Fig. 2. Plot of the imaginary part of the effective action for a space-dependent field E(x) =
Esech2(kx) as a function of the inhomogeneity parameter i, normalized by the weak field limit of
the "locally constant field approximation" .14 The dotted line shows the result obtained from the
worldline instanton approximation. The dashed line is the same ratio using a numerical integration
of Nikishov's exact expressionf; The circles represent the worldline Monte Carlo results of,20
evaluated for eE /m 2 = 1. Note that the agreement is excellent even far from the weak-field limit.

Nonplanar fields: Very recently, Dunne and Wang 15 have applied (5) to a more
general classes of fields defined by a vectorpotential whose nonvanishing component
is of the form A4(X) = -if: j(x). For a number of genuinely multi-dimensional
examples they have obtained Im.C(E) by a numerical evaluation of (5).
 1345

4. Generalization to Spinor QED

In Feynman's worldline path integral representation the transition from Scalar to
Spinor QED for the effective action can, up to the global normalization, be made
by inserting under the path integral a spin factor S[x, Ap6

(7)
Here trr denotes the Dirac trace and P the path ordering along the worldline. In
the semiclassical approximation this spin factor has to be evaluated on the extremal
trajectory. Surprisingly, it turns out that for all nonplanar fields the result is just
the same sign factor which makes already the difference between the two Schwinger
formulas (1), (2), namely S[xcl, A] = 4( _1)" with n the winding number.
5. Outlook
Let me conclude with a short summary of the advantages of the instanton approach
as compared to WKB: (i) as a pure x-space approach it completely circumvents the
usual momentum integrals; (ii) since it uses proper-time, not time, it is manifestly
relativistic from the beginning; (iii) it is in principle known how to include radiative
corrections. Some open questions are the following: (i) what is the importance of
corrections to the weak field limit, and how to calculate them; (ii) whether it is
only in the planar case that the spin factor reduces to a mere sign factor; (iii) the
generalization to nonabelian and gravitational fields.

References
1. J. Schwinger, Phys. Rev. 82, 664 (1951).
2. T. Heinzl et al., hep-ph/0601076.
3. TESLA technical design report.
4. L.V. Keldysh, JETP 20, 1307 (1965).
5. E. Brezin and C. Itzykson, Phys. Rev. D2, 1191 (1970).
6. A.I. Nikishov, Nucl. Phys. B21, 346 (1970).
7. N.B. Narozhnyi and A.I. Nikishov, Yad. Fiz. 11, 1072 (1970).
8. V.S. Popov, JETP 34, 709 (1971); V.S. Popov and M.S. Marinov, Yad. Fiz. 16, 809
(1972).
9. W. Greiner, B. Muller and J. Rafelski, Quantum Electmdynamics of Stmng Fields,
Springer, 1985.
10. G.V. Dunne and T. Hall, Phys. Rev. D58, 105022 (1998).
11. S.P. Kim and D. Page, Phys. Rev. D73, 065020 (2006).
12. I.K. Affleck, O. Alvarez and N.S. Manton, Nucl. Phys. B197, 509 (1982).
13. G.V. Dunne and C. Schubert, Phys. Rev. D72, 105004 (2005).
14. G.V. Dunne, H. Gies, C.Schubert and Q.-h. Wang, Phys. Rev. D73, 065028 (2006).
15. G.V. Dunne and Q.-h. Wang, Phys. Rev. D74, 065015 (2006).
16. R. P. Feynman, Phys. Rev. 80, 440 (1950); 84, 108 (1951).
17. C. Schubert, Phys. Rep. 355, 73 (2001).
18. H. Gies and K. Langfeld, Nucl. Phys. B613, 353 (2001).
19. S. Levit and U. Smilansky, Ann. Phys. 103, 198 (1977).
20. H. Gies and K. Klingmuller, Phys. Rev. D72, 065001 (2005).
 MONOPOLE DECAY IN A VARIABLE EXTERNAL FIELD

A. K. MONIN and A. V. ZAYAKIN

Institute for Theoretical and Experimental Physics
117259, Moscow, B.Cheremushkinskaya 25, Russia

Physics Department, Lomonosov Moscow State University,

Moscow, 119992, Russia
E-mail: [email protected]@itep.ru

The rate of monopole decay into a dyon and an electron in an inhomogeneous external
electric field is calculated by semiclassical methods. Comparison is made to an earlier
result where this quantity was calculated for a constant field. Experimental and cosmo-
logical tests are suggested.

Keywords: Monopoles, Schwinger processes, induced vacuum decay.

1. Spontaneous and Induced Decay in External Fields

Spontaneous non-perturbative processes of particle production in QFT (Schwinger
processes), or vacuum decay processes have been studied since the papers of Euler-
Heisenberg and Schwinger on e+e- generation in a constant electromagnetic field.
False vacuum decay in a scalar field theory with a stable and a metastable vacuum
states was first treated by Kobzarev et al. l Later this problem 2 ,3 was given a 1-
loop treatment, with both the exponent and the preexponential factor of decay
probability calculated.
Consideration of induced 4 ,5 Schwinger processes was a natural extension of the
scope of the problems described above. The term "induced" denotes the situation
in which the initial state is not vacuum, but rather contains some particle(s).
On the other hand, generalization of Schwinger processes can be thought of
as extending the class of fields in which the appropriate process takes place. The
original Euler and Heisenberg calculation in QED was performed for a constant
field. For harmonic plane waves calculations had first been done by Schwinger. A
semiclassical treatment of a broad class of fields was given by Popov. 6 Semiclassi-
cal methods were further developed basing upon WKB approximation and the so
called "worldline instanton method" .7 In the present short paper a combination of
the both generalizations of Schwinger processes is suggested. The possibility of an
induced monopole decay into a dyon and an electron was first suggested by Gorsky
et al. 8 Later it was calculated in a constant electric field 9 up to the leading classical
exponential factor.

1346
 1347

2. Semiclassical Approximation to Path Integral

We are going to consider 't Hooft - Polyakov monopole and dyon. Monopole mass
Mm is of the order of the scale ~w where Mw is W-boson mass, a coupling
constant. At the same time, monopole and dyon sizes are of the order of magnitude
Mil/, therefore in the weak coupling limit these particles are essentially classical
objects.
On the other hand, 't Hooft - Polyakov monopole does not possess a well-
established local field-theoretical description. That is why it is reasonable to treat
monopole and dyon in terms of a I-particle theory, evaluating Feynman path inte-
grals semi classically with restriction to the trajectories of classical motion in imag-
inary (Euclidean) time, satisfying T » M;\ where T is the typical trajectory size.
It is essential to take into account only closed trajectories' contributions because
only finite classical action configurations are relevant.
Consider a classical Euclidean worldline of a monopole with an insertion of one
electron-dyon loop. The first correction to the Green function of a monopole due to
the electron-dyon loop is given as

C(1)(T, O) = J Dye-M=J #dTDxDze-S[x,z,Al, (1)

where action S [x (u), z (v ), A J is the action for the charged particles (the dyon
and the electron) in the external field Ail' x(t) and z(t) are respectively elec-
tron and dyon coordinates, their worldlines being parameterized by u, v; action
J J J
is Six, z, A] = m Rdu + ie Aext(x)xdu + Md v'z'idv - ie Aext(z)idv. ThisJ
path integral evaluated, the Green function correction is compared to variation of
the free monopole Green function, finally yielding the imaginary mass correction
in terms of classical action Sel on the worldline Im(m) = Ke- Scl . The prefactor
K consists of Jacobians, corresponding to collective coordinates integration, and of
negative modes' contributions. 9 Below we show the results of a calculation for this
expression in different settings.

3. Results
The knowledge of Sel allows us to semiclassically estimate decay rate r = 2Imm rv
e- Scl , provided Euclidean worldlines exist and are calculable. Here is an example
of the appropriate calculation for E(t) = cosSCOt). For convenience, the results are
expressed n terms of Keldysh parameter, = r;:J}
The action on the extremal trajectory is Selh) = Seh) + Sdh) - 2Mm X4(UO),
where, if MJ < m 2 + M;,

Seh) = 27Tm: (2u oJ I

eE,
+ ,2 - ~ arctan (~)) ,
7T 1 +,2

Sdh) = 27Tm: (2VoVI+ ': -

eE, a 7T
~arctan (R))'
+ 1 L
",2
(2)
1348

1
_ -1 {(1+--y2)(CM=+m)2- MJ )(MJ-CM=-m)2)} 2"
an d 27rUo - tan (m2+M;;,-Md2)2 ,
1

-1 { (1+~) (CM=+m)2- MJ )(MJ-CM=-m)2)} 2" m .

27rVo = tan ( )2 , a = !VI' It follows Imme-
M;;,+M~-m2 d

diately that for the boundary conditions to be properly satisfied, dyon mass should
lie within the interval Mm - m < Md < Mm + m.
If the temporally-inhomogeneous field is substituted for a spatially-
inhomogeneous, the formulae above should be analytically continued in the imagi-
nary direction: I -) if·
The results of this simple calculation are easily visualized, which is not done
here due to lack of space. Comparing to the stationary field configuration, it can be
concluded that monopole decay is enhanced by the temporally-inhomogeneous field
and suppressed by the spatially inhomogeneous field.

4. Conclusion
In this short contribution the description of the induced monopole decay has been
generalized to the non-stationary field case. Despite being non-perturbatively sup-
pressed, this process might take place under some exotic conditions, e.g. in Reissner-
Nordstrom black holes and in gamma or x-ray pulse lasers, where a temporal non-
stationarity would play the key role in enhancing the process.

5. Acknowledgements
The authors are grateful to A. S. Gorsky for fruitful discussions. This work is sup-
ported in part by RFBR 05-01-00992 grant, Scientific School grant NSh-2339.2003.2
(A.Z.); RFBR 04-01-00646 and Scientific School grant NSh-8065.2006.2 (A.M.).

References
1. 1. Y. Kobzarev, L. B. Okun, M. B. Voloshin, Sov. J. Nue!. Phys. 20, 644 (1975).
2. S. R. Coleman, Phys. Rev. D 15, 2929 (1977).
3. C. G. Callan and S. R. Coleman, Phys. Rev. D 16, (1977) 1762.
4. 1. K. Affieek and F. De Lueeia, Phys. Rev. D 20, 3168 (1979).
5. K. Selivanov, M. Voloshin, ZHETP Lett, 42, 422 (1985).
6. V. S. Popov, M. S. Marinov, Yad. Fiz. 16, 809 (1972).
7. G.V.Dunne and C.Sehubert, Phys. Rev. D 72, 105004 (2005).
8. A. S. Gorsky, K. A. Saraikin and K. G. Selivanov, Nue!. Phys. B 628, 270 (2002).
9. A. K. Monin, JHEP 0510, 109 (2005).
 WORLD-MAKING WITH EXTENDED GRAVITY BLACK HOLES
FOR COSMIC NATURAL SELECTION IN THE
MULTIVERSE SCENARIO

AURELIEN BARRAU
Laboratory for Subatomic Physics and Cosmology, CNRSjIN2P3, Grenoble Universites
53, avenue de Martyrs, 38026 Grenoble cedex, Prance
[email protected]

Physics is facing contingency. Not only in facts but also in laws (the frontier becoming
extremely narrow). Cosmic natural selection is a tantalizing idea to explain the appar-
ently highly improbable structure of our Universe. In this brief note I will study the
creation of Universes by black holes in -string inspired- higher order curvature gravity.

I will focus in this note on the proposal from Lee Smolin 1 (which has already been
objected some pertinent arguments 2 ) where black holes are considered as creators
of new Universes. This alternative to the more popular view where chaotic inflation
plays this role is motivated by its ability to allow for inherited characters (what
I will call a Lamarck Universe), to rely on better understood physical grounds
and to develop a causal structure for selection effects to occur. In this scenario,
our Universe would have been born from a black hole in another Universe and
would have generated at least 10 15 Universes through the black holes it certainly
contains (this could have interesting implications on the usual question about the
surprisingly small value of the entropy of our Universe). The main idea relies on
bouncing which is quite well established for cosmological models but remains quite
controversial in the black hole sector. Frolov, Markov & Mukhanov 4 have shown
how the Schwarzschild metric inside a black hole can be attached to the de Sitter
one at some spacelike junction surface.
In addition to controversial postulates about the equation of state close to the
central singularity (R2 = 4R~R~), the basic assumption is the existence of a limiting
curvature: RMV)..pRMV)..P < 0:/l4 where l has the dimension of a length (expected to
lie close to the Planck size) and 0: rv 1.
This ad hoc assumption can be investigated in the framework of string gravity,
which is expected to account for some quantum effect in the gravitational sector.
Let us consider first the general 4-dimensional effective string action with second
order curvature corrections:

s = 1~7f .I xA[
d4 -R + 28fl¢8 M¢ + A2 e-2<P(R,w)..pRMV)..P - 4R
MV r+
MV 2
R )].

This Lagrangian is well known for appearing in the limit of different string
theories. 5 It can also be viewed as a second order truncature of Lovelock gravity
-made dynamical in 4 dimensions through the dilatonic coupling- which can be
seen as the sum of dimensionally extended Euler densities. The full Lagrangian
density (which corresponds to the most general tensor that can be equaled to the
1349
1350

stress-energy tensor while keeping the mathematical structure of the theory) reads
as 6

where Ci are constants, (j is totally antisymmetric in both sets of indices and Ln is

the Euler density of a 2n-dimensional manifold.
In any attempt to perturbatively quantize gravity as a field theory, higher-
derivative interactions must be included in the action. Furthermore, higher-
derivative gravity theories are intrinsically attractive as in many cases they dis-
play features of renormalizability and asymptotic freedom. Among such approaches,
Lovelock gravity 7 is especially interesting as the resulting equations of motion con-
tain no more than second derivatives of the metric, include the self interaction of
gravitation, and are free of ghosts when expanding around flat space. The four-
derivative Gauss-Bonnet term is most probably the dominant correction to the
Einstein-Hilbert action. s Gauss-Bonnet gravity was shown to exhibit a very rich
phenomenology in cosmology (see, e.g.,9 and references therein), high-energy physics
(see, e.g.,ID and references therein) and black hole theory (see, e.g.,ll and references
therein). It also provides interesting solutions to the dark energy problem,12 offers
a promising framework for inflation,13 allows useful modification of the Randall-
Sundrum model 14 and, of course, solves most divergences associated with the end-
point of the Hawking evaporation process. 15
Those Gauss-Bonnet black-holes can be studied with the choice of metric:
(J2
ds 2 = fj,dt 2 - fj, dr 2 - j2(d()2 + sin 2 ()dip2).

Integrating numerically (using an additional parameter) inside the event horizon rh

allows to show ll that a solution exists only up to a (lower) value ofthe radius r = r s'
A new branch exists between rs and a singular horizon rx but its physical meaning
is unclear. The main point I want to underline is that the curvature invariant can
be computed 11 around the minimal radius r s with the parameterization x = (J - (J s :

R R
I"VPA
I"VPA _
- 21
X
(4d
--2-6 + -d2)
s
r2rs(Js
26 + 0 (-1 )
2
r (Js x
-7 00
2
where d 2 and d s are expansion coefficients.
This means that although the correct topology 8 2 x R1 (an infinite -in
the t direction- tube of radius rs) naturally arises in this approach, the
expected regularization of the Riemann invariant does not occur. From
this viewpoint, Gauss-Bonnet gravity disfavours black holes as Universe-makers. Of
course, this theory remains a low-energy effective description and one can still hope
that a full quantum gravitational treatment changes this conclusion.
The point made here is nevertheless quite robust for two reasons. First, because
it remains true when a Maxwell term e- 2 <.!> FI"vFJLv is added to the action, at least
up to a critical value qcr (of the order of::::; O.3rh in Planck units) which was numer-
ically determined in. 16 When q < qcr, the topology of the solution is unchanged and
 1351

R?,v)"pR?'V)"P ----+ 00 (the case gel' < g < mV2 is interesting and should be studied
in this perspective). Then, because it remains stable under higher order curvature
corrections of string gravity. It can, in any case (expect for SUSY-II were the sit-
uation is still unclear), be explicitly shown 17 that the structure of the singularities
is unchanged and the Riemann invariant still diverges. In a way, the situation is
therefore even worst than in usual general relativity as this occurs at finite distance
from the center of the black hole.
The main point here is that perturbative string corrections seem to disfavor the
hypothesis of a limiting curvature which would have allowed for a junction layer
between a Schwarzschild space and a de Sitter space within black holes. Other-
wise stated, Gauss-Bonnet black holes are not very promising Universes progenitors
although they seem to be free of the most dramatic problems associated with quan-
tum gravity effects (in particular when considering the endpoint of the Hawking
evaporation process).
The multiverse hypothesis -whether falsifiable or not- offers a very rich frame-
work to elaborate both on the world(s) and on our scientific creations about the
world(s). It is not even clear that it makes sense to distinguish between both pro-
posals. The structure of time itslef could appear as radically modified (see, e.g.,18
and references therein). Using the string landscape to argue in favor of this model
is probably a bit naIve as this still relies on an assumed single meta-theory. This
"pluriverse" is maybe not that pluralistic ... As usual this could tell us much more
about the schemes of our mental activities than about the intrinsic structure of
reality. But it could very well be that physics is now taking distances with a kind of
implicit ontological commitment. Physics faces contingency. An exciting situation!

References
l. L. Smolin, Contribution to "Universe or Multiverse", ed. by Bernard Carr et. al., to
be published by Cambridge University Press [hep-th/0407213]
2. A. Vilenkin, hep-th/0610051
3. L. Susskind, hep-th/0407266
4. V. Frolov, M. Markov & V. Mukhanov, Phys. Lett. B, 216, 272 (1989)
5. S. Mignemi & N.R. Stewart, Phys. Rev. D, 47, 5259 (1993)
6. R. Myers & J. Simon, Phys. Rev. D, 38, 2434 (1988)
7. D. Lovelock, J. Math. Phys. 12 (1971) 498; J. Math. Phys. 13 (1972) 874.
8. B. Zwiebach, Phys. Lett. B 156 (1985) 315.
9. C. Charmousis & J.-F. Dufaux, Class. Quant. Grav. 19 (2002) 4671.
10. A. Barrau, J. Grain & S.O. Alexeyev, Phys. Lett. B 584 (2004) 114
11. S. Alexeyev & M.V. Pomazanov, Phys. Rev. D 55 (1997) 2110
12. S. Nojiri, S.D. Odintsov & M. Sasaki, Phys. Rev. D 71 (2005) 123509
13. J.E. Lidsey &: N.J. Nunes, Phys .Rev. D 67 (2003) 103510
14. J.E. Kim, B. Kyae & H.M. Lee, Nucl. Phys. B 582 (2000) 296
15. S. Alexeyev et al., Class. Quant. Grav. 19 (2002) 4431
16. S. Alexeyev, Grav. Cosm., 3, 161 (1997)
17. S. Alexeyev, M. Sazhin & M. Pomazonov, Inter. J. Mod. Phys. D 10 (2001) 225
18. G. Sparling, gr-qc/0606129
 NEUTRAL NUCLEAR CORE VS SUPER CHARGED ONE

M. ROTONDO, R. RUFFINI and S.-S. XUE

ICRA and Physics Department, University of Rome "La Sapienza", P.le A. Mom 5, 00185
Rome, Italy
ICRANei,. Piazzale della Repubblica 10, 65122 Pes cam, Italy

Based on the Thomas-Fermi approach, we describe and distinguish the electron distri-
butions around extended nuclear cores: (i) in the case that cores are neutral for electrons
bound by protons inside cores and proton and electron numbers are the same; (ii) in
the case that super charged cores are bare, electrons (positrons) produced by vacuum
polarization are bound by (fly into) cores (infinity).

1. Equilibrium of electron distribution in neutral cores

In Refs., 1-3 the Thomas-Fermi approach was used to study the electrostatic equi-
librium of electron distributions ne(r) around extended nuclear cores, where total
proton and electron numbers are the same Np = N c . Proton's density np(r) is
constant inside core r :::: Rc and vanishes outside the core r > Rc,
(1)
where Rc is the core radius and np proton density. Degenerate electron density,

(2)

where electron Fermi momentum p{, Fermi-energy £e(p{) and Coulomb potential
energy Vcou1(r) are related by,
[.e(Pt') = [(p[ c)2 + m;c4 ]1/2 - m ec2 - Vcou1(r). (3)
The electrostatic equilibrium of electron distributions is determined by
[.e(Pt') = 0, (4)
which means the balance of electron's kinetic and potential energies m Eq. (3)
and degenerate electrons occupy energy-levels up to +mec2. Eqs. (2,3,4) give the
relationships:

PeF = -1 [ ~oul
2 ()
r + 2mec2 VcouJ(r) ] 1/2 ; (5)
c
1 [2 2 ]3/2
ne(r) = 37f2(cn)3 ~oul(r) + 2mec VcouJ(r) . (6)

The Gauss law leads the following Poisson equation and boundary conditions,

These equations describe a Thomas-Fermi model for neutral nuclear cores, and
have numerically solved together with the empirical formula 1,2 and p-equilibrium
equation3 for the proton number Np and mass number A = Np + N n , where N n is
the neutron number.

1352
 1353

me c 2
20

-150

-200

-250

Fig. 1. Potential energy-gap ±me c2 - Veon! (r) and electron mass-gap ±mec2 in the unit of mec2
are plotted as a function of (r - Rc) / (lOA" ). The potential depth inside core (r < Re) is about pion
mass m"c 2 » m e c2 and potential energy-gap and electron mass-gap are indicated. The radius ro
where electron distribution ne(ro) vanishes in super charged core case is indicated as ro-, since
it is out of plotting range.

2. Equilibrium of electron distribution in super charged cores

In Ref.,4,5 assuming that super charged cores of proton density (1) are bare, electrons
(positrons) produced by vacuum polarization fall (fly) into cores (infinity), one
studied the equilibrium of electron distribution when vacuum polarization process
stop. When the proton density is about nuclear density, super charged core creates
a negative Coulomb potential well - Vcaul(T), whose depth is much more profound
than -m e c2 (see Fig. [1]), production of electron-positron pairs take places, and
electrons bound by the core and screen down its charge. Since the phase space of
negative energy-levels E(p)

E(P) -_ [()2
pc 2 4]1/2 -
+ meC V;caul (co)
r , (8)

below -m e c2 for accommodating electrons is limited, vacuum polarization process

completely stops when electrons fully occupy all negative energy-levels up to -m e c2 ,
even electric field is still critical. Therefore an equilibrium of degenerate electron
distribution is expected when the following condition is satisfied,

f (P ) _- [()2 2 4]1/2 - V;coul ()

pc + 7neC r _ .
- -rnee 2, (9)

and Fermi-energy

(10)
1354

which is rather different from Eq. (4). This equilibrium condition (10) leads to
electron's Fermi-momentum and number-density (2),

PeF = -1 [~oul(r)
2 - 2mec 2Vcoul (r )]1/2 ; (11)
c

ne(r) = 37r 2tcn)3 [~~ul(r) - 2meC2Vcoul(r)] 3/2 . (12)

which have a different sign contracting to Eqs. (5,6). Eq. (7) remains the same.
However, contracting to the neutrality condition Ne = Np and ne(r)lr-HX) --* 0 in
the case of neutral cores, the total number of electrons is given by

N~on = l ro
47rr2drne(r) < Np, (13)

where ro is the finite radius at which electron distribution ne(r) (12) vanishes:
ne(ro) = 0, i.e., Vcoul(ro) = 2mec2, and ne(r) == 0 for the range r > ro. Nion < Np
indicates that such configuration is not neutral. These equations describe a Thomas-
Fermi model for super charged cores, and have nurnerically 4 and analytically5 solved
with assumption Np = A/2.

3. Ultra-relativistic solution
In analytical approach,5,6 the ultra-relativistic approximation is adopted for
Vcou1(r) » 2meC2, the term 2mec2Vcoul(r) in Eqs. (5,6,11,12) is neglected. It turns
out that approximated Thomas-Fermi equations are the same for both cases of
neutral and charged cores, and solution Vcoul(r) = nC(37r 2n p)1/3¢(x),

_ { 1 - 3 [1
( )-
¢x v'2
+ 2- 1/ 2 sinh(3.44 - yi3x)]-1 , for x < O,}, (14)
(x+1.89) , for x> 0,

where x = 2(7r /3)1/6o)/2n~/3(r-Re) rv O.l(r- R e )/),7[ and the pion Compton length
),7[ = n/(m7[c). At the core center r = O(x --* -00), Vcoul(O) = nC(37r 2n p )1/3 rv
m7[c 2. On core surface r = Re(x = 0), Vcoul(R e ) = 3/4Vcou1 (0) »m ec2, indicating
that the ultra-relativistic approximation is applicable for r ;S Re. This approxima-
tion breaks down at r 2, ro. Clearly, it is impossible to determine the value ro out
of ultra-relativistically approximated equation, and full Thomas-Fermi equation (7)
with source terms Eq. (6) for the neutral case, and Eq. (12) for the charged case
have to be solved.
For r < ro where Vcoul(r) > 2mec2, we treat the term 2mec2Vcoul(r) in
Eqs. (6,12) as a small correction term, and find the following inequality is always
true

(15)

where n~eutral(r) and n~harged(r) stand for electron densities of neutral and super
charged cores. For the range r > ro, n~harged(r) == 0 in the case of super charged
 1355

core, while n~eutral (r) -+ 0 in the case of neutral core, which should be calculated
in non-relativistic approximation: the term Vc~ul(r) in Eq. (6) is neglected.
In conclusion, the physical scenarios and Thomas-Fermi equations of neutral
and super charged cores are slightly different. When the proton density np of cores
is about nuclear density, ultra-relativistic approximation applies for the Coulomb
potential energy Vcoul(r) » m e c 2 in 0 < r < ro and ro > R e , and approximate
equations and solutions for electron distributions inside and around cores are the
same. As relativistic regime r rv ro and non-relativistic regime r > TO (only applied
to neutral case) are approached, solutions in two cases are somewhat different, and
need direct integrations.

References
1. J. Ferreirinho, R. Ruffini and L. Stella, Phys. Lett. B 91, (1980) 314.
2. R. Ruffini and L. Stella, Phys. Lett. B 102 (1981) 442.
3. R. Ruffini, M. Rotondo and S.-S. Xue, Int. Journal of Modern Phys. D Vol. 16, No.1
(2007) 1-9.
4. B. Muller and J. Rafelski, Phys. Rev. Lett., Vol. 34, (1975) 349.
5. A. B. Migdal, D. N. Voskresenskii and V. S. Popov, JETP Letters, Vol. 24, No.3 (1976)
186, Sov. Phys. JETP 45 (3), (1977) 436.
6. M. Rotondo, R. Ruffini and S.-S. Xue, in preparation.
This page intentionally left blank
 Black Holes in Higher
Dimensions (Black Rings
and Black Strings)
This page intentionally left blank
 GRAVITATIONAL PERTURBATIONS OF HIGHER
DIMENSIONAL ROTATING BLACK HOLES

HARI K. KUNDURI*
School of Physics and Astronomy, University of Nottingham, NG72RD, UK
h. k. [email protected]

JAMES LUCIETTI
Centre for Particle Theory, Department of Mathematical Sciences,
University of Durham, South Road, Durham, DHl 3£E, UK
[email protected]

HARVEY S. REALL t
School of Physics and Astronomy, University of Nottingham, NG72RD, UK
harvey. [email protected]

Assessing the stability of higher-dimensional rotating black holes requires a study of

linearized gravitational perturbations around such backgrounds. We study perturbations
of Myers-Perry black holes with equal angular momenta in an odd number of dimensions
(greater than five), allowing for a cosmological constant. Such black exhibit enhanced
symmetry: they are cohomogeneity-one solutions. This allows gravitational perturbations
to be decomposed into scalar, vector and tensor types. The equations of motion for tensor
perturbations reduce to a single radial equation. In the asymptotically flat case we find
no evidence of any instability associated with tensor perturbations. In the asymptotically
anti-de Sitter case, we demonstrate the existence of a superradiant instability that sets in
precisely when the angular velocity of the black hole exceeds the speed of light from the
point of view of the conformal boundary. We suggest that the endpoint of the instability
may be a stationary, nonaxisymmetric black hole.

1. Introduction

Exact solutions describing higher-dimensional rotating black holes have been known
for a long time l but the question of their classical stability is still unresolved. There
are arguments suggesting that a Myers-Perry (MP) black hole will be unstable for
sufficiently large angular momentum in five 2 and higher 3 dimensions. However, a
convincing demonstration of this requires a study of linearized gravitational fluctu-
ations around such backgrounds. This has only been done in the limit of vanishing
angular momentum, i.e., for higher-dimensional Schwarzschild black holes. 4 - 6
Another context in which stability of higher dimensional rotating black holes has
been discussed is the AdS/eFT correspondence. 7 The MP solutions have been gen-
eralized to include a cosmological constant. 8 ,g There is a qualitative argument that
rotating, asymptotically AdS black holes might exhibit a superradiant instability.lO
The idea (inspired by the corresponding instability of a Kerr black hole in the pres-
ence of a massive scalar field l1 - 13 ) is that superradiant perturbations are trapped

*PPARC Postdoctoral Fellow

tRoyal Society University Research Fellow

1359
1360

by the AdS potential barrier at infinity and reflected towards the black hole where
they get amplified and the process repeats. It can be proved that no such instability
is present for black holes rotating at, or slower than, the speed of light relative to
the conformal boundary 1o (essentially because superradiant modes do not fit into
the AdS" box" 14) but an instability may well be present for more rapidly rotating
holes. The only way to find out is to study perturbations of such black holes. This
has been done for scalar field perturbations of small four-dimensional Kerr-AdS
black holes 15 but not for gravitational perturbations, large black holes, or higher
dimensions.
In four dimensions, gravitational perturbations of rotating black holes can be
studied analytically. The tractability of the problem arises from two miraculous
properties of the Kerr metric. Firstly, the equations of motion for gravitational
perturbations can be decoupled and reduced to a PDE for a single scalar quantity. 16
Secondly, this equation can be reduced to ODEs governing the radial and angular
behaviour by separation of variables. It is known that this separability property of
the Kerr metric docs extend to some of the higher-dimensional MP metrics, 17-19
and MP metrics with a cosmological constant. 20 This makes the study of scalar field
perturbations in such backgrounds tractable. However, so far no-one has succeeded
in decoupling the equations of motion governing gravitational perturbations of MP
black holes.
We have made progress with this problem 21 by considering the subclass of MP
black holes for which the number of space-time dimensions is odd and the angular
momenta are all equal, allowing for a cosmological constant. Such black holes are
cohomogeneity-l: the metric depends only on a radial coordinate. In D = 2N + 3
dimensions, the isometry group is enhanced from R x U(1)N+1 to R x U(N + 1)
where R denotes time translations. The horizon is a homogeneously squashed S2N+1
viewed as a Sl bundle over C pN. The symmetry enhancement allows us to classify
gravitational pertubations into scalar, vector and tensor types according to how
they transform under isometries of C pN. We consider the tensor perturbations.
These do not exist for N = 1 so we restrict ourselves to N > 1, i.e., spacetime di-
mensionality seven or greater. After separation of variables, the equations of motion
for tensor perturbations reduce to a single ODE governing the radial behaviour. For
completeness, we also consider massive scalar field perturbations of these black holes
(for N ;::: 1). We shall present a unified form for the radial equation that applies
both to scalar fields and to tensor gravitational perturbations.
Asymptotically flat, cohomogeneity-1 Myers-Perry black holes exhibit an upper
bound on their angular momentum for a given mass. Solutions saturating this bound
have a regular but degenerate horizon. This means that the black holes considered
here behave rather differently from ones for which some of the angular momenta
vanish, which are the ones expected to exhibit a gravitational instability.2,3 So
there is no a priori reason to expect any instability to exist for the MP black holes
considered in this paper and indeed we find no instability.
In the asymptotically anti-de Sitter case, there is also an upper bound on angular
 1361

momentum for given mass and black holes saturating this bound have a regular but
degenerate horizon. These extremal solutions always rotate faster than light and
can be arbitrarily large compared with the AdS radius. The argument of Ref. 10
suggests that black holes near to extremality might be unstable to losing energy and
angular momentum into superradiant gravitational and scalar field perturbations.
We shall demonstrate that this is indeed the case. Moreover, we shall show that
this instability appears as soon as the angular velocity of the hole exceeds the speed
of light, i.e., as soon as the stability argument of Ref. 10 fails. The instability is a
short distance instability in the sense that unstable modes exist for all wavelengths
below a certain critical value determined mainly by the angular velocity of the hole.
However, amongst unstable modes, the shortest wavelength modes are the least
unstable.
Having demonstrated the existence of an instability, it is natural to ask what
the endpoint of the instability iii. We propose that a black hole that suffers from
this superradiant instability will evolve to a stationary, nonaxisymmetric black hole
solution. The motivation behind our proposal will be explained at the end of this
paper.
This paper summarizes the results that we obtained in Ref. 2l. The reader is
urged to consult that paper for further details. However, section 3.1 contains new
material that will be explained in more detail in a future paper. 22

2. The background solution

The higher-dimensional generalization of the Kerr solution was obtained by Myers
and P erry 1 and subsequently generalized to include a cosmological constant in fiveS
and higher 9 dimensions. It is parameterized by a mass parameter !vI and [(D -1) /2]
angular momentum parameters ai. In D = 2N + 3 dimensions with equal rotation
parameters ai = a the solution is cohomogeneity l. The metric can be written as:

r2
g(r)2 = ( 1 + -
2M2
- -
2Ma 2
- +r 2N
-+-2
)-1 h(r) 2 =r 2 (
1+ r.2Ma
2

2N + 2
)
' (2)
f2 r 2N

r 2Ma
f(r) = g(r)h(r) [2(r) = r 2N h 2 ' (3)

where gab is the Fubini-Study metric on CpN with Ricci tensor Hab = 2(N + 1)Yab,
and A = Aadxa is a I-form such that J = ~dA is the Kahler form on CpN. This
way of writing the metric arises from the fact that S2N+1 can be written as an Sl
fibre over C pN. The fibre is parameterized by the coordinate VJ, which has period
21f.
The spacetime metric satisfies R"v = -e- 2 (D -1)g"v. Asymptotically, the solu-
tion approaches anti-de Sitter space with radius of curvature e. An asymptotically
1362

flat Myers-Perry black hole can be recovered by taking £ ---> 00. The event horizon
located at r = r + (the largest real root of g-2) is a Killing horizon of'; = at + OHa'!j; ,
where the angular velocity of the horizon is:

OH - 2Ma (4)
- r!N+2 + 2Ma2·

The mass E and angular momentum J (defined with respect to a'!j;) are 23

E = A
2N 1
+ M (N
41rC
+~ +~)
2 2£2 '
J =
2
A4 ::C+1 (N
"
+
l)Ma (5)

where A 2N + 1 is the area of a unit 2N + 1 sphere.

As written the metric is parameterised by (M, a). We shall assume a :2: 0, which
can always be achieved by t ---> -t if necessary. Sometimes it will be convenient to
work with more "physical" variables (OH, r +). Fortunately one can easily invert for
(M,a) in terms of (OH,r+):

(6)

For given r +, existence of a regular event horizon imposes an upper bound on OH:

1 N£2
OH <-
- £ 1+(N+1)r~· (7)

The extremal solution saturating this bound has a regular but degenerate horizon.
In the asymptotically AdS case, the" co-rotating" Killing vector field'; is timelike
everywhere outside the horizon if OH ::; 1/£ but becomes spacelike in a neighbour-
hood of infinity otherwise. With respect to the metric on the conformal boundary,
.; is timelike if OH < 1/£, null if OH = 1/£ and spacelike otherwise. For this reason,
black holes with OH > 1/£ are said to be rotating faster than light. Note that the
extremal black holes always rotate faster than light, and that such black holes can
be arbitrarily large compared with the AdS length.

3. Perturbation equations
3.1. Scalar, vector, tensor
As explained in the introduction, we can make progress with the study of grav-
itational perturbations of these cohomogeneity-one black holes by exploiting the
large symmetry group. Specifically, we can decompose a general gravitational per-
turbation into scalar, vector and tensor perturbations on C pN. This is familar from
perturbations of spherically symmetric black holes, where one performs an analo-
gous decomposition on a sphere. 5 However, there are two additional complications
present here.
It is natural to decompose any perturbation of our black hole into Fourier modes
around the Sl fibre parameterized by 'lj;, i.e., we assume that the perturbation is
proportional to exp(im'lj;) with m an integer. To understand the resulting equations,
 1363

it is convenient to imagine performing a dimensional reduction on this Sl, i.e.,

reduce S2N+1 to CpN. This reduction gives rise to a Kaluza-Klein magnetic field
A, and the perturbation carries charge m with respect to this field. Hence we will
need to understand charged perturbations on CpN. This means that we will need
to work with the gauge-covariant derivative V which acts on a field of charge m as
Va == 'Va - imAa, (8)
with 'V the Levi-Civita connection on C pN.
The second complication comes from the fact that on a sphere (of dimension
higher than two) there is only one second rank tensor invariant under the isometry
group, namely the metric. However, on C pN there are two such tensors: the metric
and the Kahler form J. Hence in defining what we mean by scalar, vector and
tensor, we have to consider contractions with both of these objects.
Consider a general metric perturbation:
(9)
Obviously, htt, htT') ht'lj; etc. transform as scalars on CpN. More interesting are
components such as Va == h ta , which carry a vector index a. It can be shown 22 that
Va can be decomposed as
Va = Wa + V a¢ + JabVb'IjJ, (10)
where ¢ and 'IjJ are scalars, and Wa obeys the" doubly transverse" conditions:
Vav a = JabVaVb = O. (11)
where we raise indices with the metric on C pN. We can therefore decompose hta
into a "vector part" Wa and a "scalar part" corresponding to ¢, 'IjJ. The same can be
done for h'lj;a and h ra . The important thing to note is that the vector parts obey two
transversality conditions, arising from the fact that we have two invariant tensors.
Finally, consider hab. It can be shown 22 that hab can be decomposed as
hab = Yab + ... , (12)
where the ellipsis denotes vector and scalar parts of hab, i.e., terms that can be
written in terms of derivatives of vectors (analagous to w a ) and scalars, and Yab is
the "tensor part" of the perturbation, defined by the (traceless, doubly transverse)
conditions
(13)
The reason for performing this decomposition of a general perturbation into scalar,
vector and tensor types is that. t.he linearized Einstein equations respect the decom-
position, i.e., the equations of motion for scalar, vector and tensor perturbations
decouple from each other. The decomposition of a general metric perturbation will
give rise to many scalars, and several vectors. However, t.here is only a single ten-
sor, so tensor perturbations are the simplest to study. We shall restrict attention
to tensor pert.urbations henceforth. Tensor perturbations do not exist for N = 1
(recall Cp1 = S2) so we shall assume N > 1.
1364

3.2. Tensor equation of motion

A general separable tensor perturbation takes the form

where Yab(x) is a charge m tensor (in the above sense, i.e., obeying equation 13)
on C pN. We shall assume that Yab can be expanded in eigenfunctions of a suit-
able Laplace-type operator, namely the gauge-covariant version of the Lichnerowicz
operator on C pN :

(15)
where Rabcd is the Riemann tensor of C pN. Note that bot
maps tensors to tensors.
There is no loss of generality in assume that Yab is an eigenfunction of bot
with
eigenvalue A, i.e., a tensor harmonic.
Define a map .:J by

(16)
This maps tensors to tensors, and has eigenvalues 1, -1 whose eigenvectors we re-
fer to as hermitian or anti-hermitian respectively. It commutes with bot.Hence we
can simultaneously diagonalize .:J and bot.
This implies that we can classify eigen-
functions of bot into hermitian and anti-hermitian. Further, in the antihermitian
case, one can distinguish between Yab of type (2,0) and (0,2), which we define by
J a cYcb = EiYab , where E = +1, -1 for (2,0) and (0,2) harmonics respectively.
Using these results, the linearized Einstein equation for a tensor perturbation
reduces to the radial equation 2l

_l~
9 dr
(l9 d1J!)
dr
+ V1J! = 0, (17)

where the "potential" V(r) is defined by

2
V(r) = Vo(r) _ (w-mD)2+ f2 (m2 + 4(1 - a)h + A - 4(N + 1) - 2E(1 + a)m) ,
h 2 r 4 r2
(18)
with a = 1 or -1 for anti-hermitian and hermitian Yab respectively, A is the eigen-
val ue of bot, and Va is defined by
(19)

The next step is to determine the eigenvalues A of tensor harmonics.

3.3. Tensor harmonics

Recall that there are no tensor harmonics on Cpl. Uncharged tensor harmonics
on CP2 have been considered previously.24,25 It is straightforward to generalize
 1365

the results to include charge. 21 The result is that doubly transverse charged tensor
harmonics must be anti-hermitian (0" = 1). The eigenvalue spectrum is

N=2: A = l(l + 4) + 12 - m2 + 4Em, l = 2k + 16 - Eml (20)

where
_ { 0,1,2,... if Em:S; 4
k - 1,2,3. . . if Em = 5 (21)
2,3,4, . . . if Em:::: 6

This implies that the minimum value of lis lmin = 6-Em if Em :s; 3 and lmin = Em-2
for Em:::: 4. Conversely, for given l, the allowed values of m are given by

Em = 6 - l, 6 - l + 2, 6 - l + 4, ... , l, 1+2, 1= 2,3,4, ... (22)

These charged tensor harmonics on C pN can be lifted to (uncharged) eigenfunc-

tions of the Lichnerowicz operator on S2N+l where l is the usual "total angular
momentum" quantum number. a Note I:::: 2, as expected for tensor harmonics.
We can obtain a general formula for the charged tensor eigenvalues for N :::: 2
by comparing our results with those for non-rotating black holes. If we set a = °
then we have a Schwarzschild(-AdS) black hole and our tensor perturbations should
form a subset of the tensor perturbations (on the sphere) considered in Refs 4-6.
Demanding agreement between our results and those of Ref. 6 yields

A = l(l + 2N) + 4N + 40" - m

2
+ 2E(1 + O")m. (23)

This is consistent with the above results for N = 2. The limitation of this approach is
that it does not tell us which values of I are permitted beyond the obvious restriction
l :::: 2.

3.4. Unified potential

To summarize, the equation of motion for tensor gravitational perturbations can be
separated and reduced to a single radial equation (17). The same thing can be done
for scalar field perturbations, and it is interesting to compare the results. Consider
a scalar field obeying the Klein-Gordan equation
(24)

Take a separable Ansatz:

<I> = e-iwt+im"vh(r)-1/2r-Nw(r)Y(x), (25)
where Y is a charged scalar eigenfunction of - V2 on C pN. Then the Klein-Gordan
equation reduces to a radial equation of exactly the same form as (17). The only
difference between scalar field and tensor gravitational perturbations shows up in

aNote that m = 2.:i. mi where (ml, m2, ... ) is the weight vector of SO(2N + 2) corresponding to
the harmonic in question.
1366

the potential V(r). This can be written in a unified form so that it applies to both
types of perturbation 21

V=Va+f21L2-(w-mO)2+~: 2
[l(l+2N)-m (1- ~:) +4(1-0") (~: -1)],
(26)
where Va is defined by (19). For scalar field perturbations, N ~ 1, 0" = 1 and
l = 2k + Iml, k = 0,1,2, .... For gravitational pertubations, N ~ 2, IL = 0 and the
permissible values of 0", l are discussed above.
Note that anti-hermitian (0" = 1) gravitational perturbations bey exactly the
same equation as a massless scalar field. Therefore, one might think that results
concerning the stability of (asymptotically flat) MP black holes against massless
scalar field perturbations 18 ,26,27 would imply stability with respect to 0" = 1 grav-
itational perturbations. However, these results concern black holes in five 18 and
six 26 ,27 dimensions with a single non-vanishing angular momentum whereas we are
interested in black holes in seven or more odd dimensions with all angular momenta
equal and non-vanishing. It appears that scalar field perturbations of such black
holes have not been considered previously. Furthermore, there is a difference be-
tween scalar field and gravitational perturbations: the lower bound on l is different
for the two cases. (For N = 2, the lower bound for scalars can be either less than, or
greater than, the lower bound for doubly transverse tensors, according to the value
of m.)
Note that by introducing a "tortoise" coordinate x(r) defined by
dx 9
(27)
dr J'
we can convert the radial equation (17) into a I-dimensional time-independent
Schrodinger equation:
d2
--1Ji
2
+ VIJi = O. (28)
dx

4. Stability analysis
4.1. Boundary conditions
The horizon is located at r = r +, which corresponds to x ---7 -00:

x rv - 1 log (r-r+)
--- , (29)
a r+
where
a = r+(g-2)'(r+)
(30)
h(r+) .
At the horizon, V ---7 -(w-mOH)2 so the solutions behave as exp (±i(w - mOH )x).
Regularity on the future horizon requires that we choose the lower sign, so we have
the boundary condition
IJi = exp (-i(w - mOH)x) <P, (31)
 1367

where <I>(T) is smooth at T = T+. Substituting this into the Schrodinger equation
and expanding around T = T+ gives (choosing <I>(T+) = 1)

"'-1
'±'- + 0'2 V'(T+)(T-T+) + 0(( T-T+ )2) . (32)
- 2iO'(w - mD H )

In the asymptotically flat case, we have

x = T +0 (T-2N+1) as T ----> 00. (33)
The general solution as T ----> 00 is a superposition of outgoing and incoming waves
(proportional to eikr and e- ikr respectively):

N '2 2 (34)

where H(i) are Hankel functions and k = Jw

2 - f-L2. In the N = 1 case, the solution

is as above except the order of the Hankel functions is now [(I + I? + 2M(f-L2 -
2w 2)j1/2. In the asymptotically flat case, we are mainly interested in gravitational
perturbations for which f-L = 0 and k = w.
In the asymptotically AdS case, we have x ~ -fi 2 /T ----> 0 as T ----> 00. The
asymptotic behaviour of the potential is

V rv Va
T2 f-L2
+- -
fi2
~ -T2
fi4
(
(N
1)
+ I? - -4+"p2fi2 (35)

with corresponding asymptotic solutions \jJ rv T- 1 / 2 ±y'(N+l)2+Jl2£2. Choosing the

normalizable solution corresponds to the boundary condition

as T ----> 00. (36)

For stability of the AdS background we demand that a scalar field obeys the Bre-
itenlohner-Freedman bound 31

(37)
A linearized instability of the black hole would correspond to a solution of the
radial equation that is regular on the future horizon and vanishing at infinity, with
Im(w) > O. In the asymptotically flat case this requires Zin = O. Note that such a
solution vanishes exponentially at the horizon.

4.2. The case m = 0

It is easy to show that there can be no instability (whether asymptotically flat or
asymptotically AdS) when m = 0 and f-L2 '2 O. If m = 0 then it is natural to consider
the potential V = V +w 2 , which does not depend on w. The radial equation becomes

(m = 0). (38)

Assume that \jJ describes an unstable mode, so Im(w) > 0 and \jJ vanishes at the
horizon and at infinity as described above. The differential operator on the LHS
1368

of (38) is self-adjoint on such functions and hence w 2 must be real so w is pure

imaginary and w 2 is negative. A simple argument 2l reveals that Vo is positive. If
we assume /L 2 :::> 0 then the remaining terms in 11 are non-negative so 11 is positive.
Hence w 2 must be positive, which is a contradiction.
For AdS black holes, it would be interesting to see if this conclusion could
be extended to tachyonic scalars satisfying the Breitenlohner-Freedman bound by
combining our argument with that of Ref. 31.

4.3. Strategy
We can look for unstable modes using the strategy adopted by Press and Teukolsky
for the Kerr black hole. 28 We expect the black hole to be stable for small angular
momentum because we know that the higher-dimensional Schwarzschild black hole is
stable. 6 Hence, for small angular velocity, the only admissible solutions of the radial
equation must have negative imaginary part, i.e., they are quasi-normal modes. If
an instability is to appear as the angular velocity increases then one of these quasi-
normal modes must cross the real axis in the complex w plane. b Hence we expect
the onset of instability to be indicated by the appearance of a real frequency mode
at a critical value of the angular velocity. The strategy is to look for such a mode.
By continuity it must have Zin = 0 in the asymptotically flat case, i.e., it must
be purely outgoing at infinity. In the AdS case it must obey the "normalizable"
boundary condition (36).
Note that the radial equation is invariant under w -+ -wand m -+ -m and, for
tensors, f -+ - f . e Hence we can assume w :::> O. Following Press and Teukolsky,30
consider the Wronskian of \[! and Wfor real w. This must be constant so we obtain

1m [Wox \[! l:~ = 0, (39)

for any Xl and :r;2. Taking Xl at the horizon and X2 at infinity and using the boundary
condition at the horizon and infinity gives, for the (massless) asymptotically flat
case,

(mrlH - w) = -2w ( IZout 12 -IZinl 2) . (40)

7r

Hence a purely outgoing mode must have

(asymptotically flat). (41)
In other words, the mode must be superradiant. In the AdS case, the LHS is un-
changed but the term at infinity vanishes and we must have
(asymptotically AdS). ( 42)

bVarious mathematical subtleties such as modes coming in from infinity might invalidate this
statement but such subtleties do not occur for Kerr 29 and we shall ignore this possibility here.
cThis just corresponds to complex conjugation of the solution. This invariance arises from a discrete
symmetry of the background which, in the coordinates of Ref. 9 is t -+ -t, ¢i -+ -¢i. In our
coordinates this amounts to t --t -t, 1j; -+ -1j; and A --t -A.
 1369

Physically, this is simply the statement that there cannot be a constant fiux of
radiation through the horizon if the flux at infinity vanishes. Note that for both
cases, we must have Tn ?:: 0 since DH ?:: O.
We shall discuss the asymptotically AdS case first because the fact that we only
have to consider a single value of w makes this case simpler to analyse than the
asymptotically fiat case, for which we have to consider a range of values for w.

5. Asymptotically anti-de Sitter black holes

5.1. Form of the potential and behaviour of solutions
Set w = mD H > O. The potential vanishes at the horizon and is monotonically
increasing just outside the horizon. For large r, the potential increases d proportional
to r2. What happens ill between depends on the values of the parameters. For small
w, V is positive everywhere. However, for sufficiently large w, there is a "classically
allowed"e region in which 1/ is negative. In more detail: v has roots at r = r1, 7'2.
T

The potential is positive for 7'+ < 7' < 7'1, negative for r1 < 7' < 7'2 and positive
for T' > 7'2, i.e., there is a potential barrier separating the classically allowed region
where V is negative from the horizon.
Note that, in the AdS case considered here, the initial data \[I(T' +) and \[1'(7'+)
are real and positive (since w = mDH)' Hence \[I is real everywhere. It is easy to
see that \[I will simply increase monotonically if V is positive everywhere. Hence we
need V to be negative somewhere for an acceptable solution of the radial equation to
exist. The solution \[I will increase monotonically in the potential barrier, oscillate in
the classically allowed region, and then match onto a sum of growing and decayingf
terms at large 7'. We need to tune DH until the coefficient of the growing mode
vanishes, i.e., until we obtain a "bound state" solution of the radial equation.

5.2. Small AdS black holes

Consider the case 7'+ « £. In this case, one can use matching techniques to solve the
radial equation. One finds that a solution obeying the boundary conditions exists
only for discrete values of w£. Since w = mDH, this translates into a condition on
DH:21

p = 0,1,2, ... ( 43)

The solutions with p > 0 correspond to "excited states" for which the solution of
the radial equation oscillates (with p + 1 extrema) before approaching zero at large

dIf fL2 is close to the Breitenlonher-Freedman bound then the coefficient of proportionality is
negative but we shan't worry about this and our results for small black holes suggest that it
doesn't change the qualitative behaviour of solutions.
eOf course, everything we are doing is classical but since we have written the radial equation in the
form of a Schrodinger equation, we can borrow terminology such as "classically allowed", "bound
state" etc. from quantum mechanics.
fMore precisely: non-normalizable and normalizable.
1370

r. We are interested in the onset of instability, corresponding to the smallest value

of DH£ for which a solution exists, so we are mainly interested in p = O.
For scalars, we have l ::0: m so, for given m, the smallest value of DH for which
we have a solution is

(44)

This is the critical value of DH beyond which modes with angular quantum number
m become unstable. Note that it always exceeds 1/£, consistent with the proof of
stability for DH ::; 1/£ given in Ref. 10. However, this proof has been criticized32
because it assumes the dominant energy condition, which is violated if f-L2 < O.
Our result shows that, for small black holes, this does not matter so long as the
Breitenlohner-Freedman bound is satisfied.
Note that the critical value for DH tends to 1/£ from above as m ----+ 00. This
proves that, for small black holes at least, the instability sets in as soon as DH
exceeds 1/£, with the shortest wavelength modes becoming unstable first.
Now consider gravitational pertubations, for which the threshold of stability
occurs at

(45)

For N = 2, taking l = lmin(m) and f = 1, this evaluates to 11,5,3 for m = 1,2,3

and 1 + 4/m for m ::0: 4. (Taking f = -1 just makes lmin bigger.) So the conclusion
is the same as for scalar field perturbations: a superradiant gravitational instability
sets in as soon as the angular velocity exceeds the speed of light, with the shortest
wavelength modes becoming unstable first.

5.3. Numerical results

In the context of the AdS/eFT correspondence, we have to consider a higher di-
mensional spacetime consisting of the product of the black hole space-time with a
compact internal space (e.g. a sphere). It is believed that small AdS black holes are
unstable with respect to the Gregory-Laflamme instability33 under which they are
expected to localize on the internal space. This means that, although small black
holes with DH£ ::; 1 do not suffer from a superradiant instability, they are never-
theless unstable. In order to eliminate the GL instability we have to extend our
results to large AdS black holes, i.e., r + > £. This can be done by solving the radial
equation numerically.21
Our numerical results show that the critical value of DH£ is always greater than
1, and tends to 1 as m ----+ 00. The interpretation is exactly the same as for small
black holes: for any given r +, if we start from DH = 0 and increase DH then as
soon as DH£ exceeds 1, the black hole will become unstable to all perturbations for
which m exceeds some critical value. In the next section we shall demonstrate this
analytically.
 1371

5.4. WKB analysis

In this section we consider both scalar field and gravitational perturbations gov-
erned by the effective Schrodinger equation with potential (26). As before, we are
interested in modes at the threshold of instability so W = mD H . The strategy is to
look at the potential for large m (and hence large w). In this limit, WKB techniques
can be used. g The result is that the critical value of DH is given by21

DH£::::; 1 + ~ (z - m + N+ 1 + V(N + 1)2 - ~ + fL2£2 + 2P) . (46)

This formula is valid for large m with Z - m = 0(1). P is a non-negative integer.

Once again, there are" excited state" solutions corresponding to positive p, p = 0
corresponds to the threshold of instability.
The coefficient of 11m in (46) is positive (at least in all cases for which we know
Zmin) so once again we see DH£ ---+ 1+ as m ---+ 00 so, irrespective of the size of the
black hole, once its angular velocity exceeds the speed of light it becomes unstable
to perturbations of arbitrarily short wavelength.

5.5. WKB calculation of unstable modes

So far, we have been looking for real frequency modes, whose existence indicates
the onset of instability. However, the WKB approach can also be used to determine
unstable modes directly. This has been used previously in a study of the superradiant
instability of the Kerr black hole in the presence of a massive scalar field. 12 We allow
w to be complex: w = w R +iw I and look for a suitable solution of the radial equation
in the WKB approximation, which we expect to be valid for large m.
Assuming WR < mD H and Z- m = 0(1) we obtain the quantization condition 21

w~£ ::::; 1 + ~ (Z- m + N+ 1 + V(N + 1)2 - ~ + fL2£2 + 2P) , P = 0,1,2, ...

(47)
and we can bound 21

0< WI£ < o:exp( -(3m) (48)

for some positive constants 0:, (3. We see that, although large m modes are the
first to become unstable when DH£ exceeds 1, the growth time of the instability
is exponentially large in m so these modes are the least unstable. This suggests
that the most unstable modes will be those for which m is not particularly large. It
would be interesting to calculate WI for such modes.

gStrictly speaking, the WKB approximation should only be valid for large /1-. For gravitational
perturbations we might not expect WKB to work very well. However, the same remark applies to
the calculation of black hole quasi-normal modes, where WKB has been found to be accurate 84
We shall see that the WKB results are in good agreement with our numerical results for /1- = O.
In any case, the WKB method is certainly reliable for sufficiently massive scalar fields.
1372

6. Asymptotically flat black holes

6.1. Introduction
We know that a mode at the threshold of instability must obey 0 :S w :S mOH,
i.e., it is superradiant. The only known way that superradiant modes can actually
lead to an instability is if they can be trapped by the potential at infinity, i.e., they
must be bound states. This would require a local minimum in V, as in the AdS
case, or for a massive scalar field in four dimensions. h In all cases that we have
examined, the qualitative form of the potential for JL = 0 is: V --+ -(w - mOH j2 as
x --+ -00, then V increases to a positive maximum and decreases to _w 2 as x --+ 00.
It appears that a local minimum in V is not possible so there is no obvious sign of
any gravitational instability apparent from our radial equation.
This qualitative argument is no substitute for a quantitative study. We shall
analyse the radial equation in two cases: first for large m and w using the WKB
method and then numerically for D = 7.

6.2. WKB approximation

Consider large m with r +wlm fixed and l rv lmin so lim --+ 1. Qualitatively, the
form of V is as followS. 21 It takes the value -(w - mn H )2 at the horizon, increases
to a positive maximum and then decreases to _w 2 at infinity. In other words, there
is a potential barrier of height proportional to m 2 separating the classically allowed
region near infinity from the classically allowed region near the horizon. The WKB
method will then give IZout!Zinl : : :; 1. However, as argued above, a mode at the
threshold of instability will have IZout!Zinl --+ 00. We conclude that no such mode
exists for large m and w.

6.3. Numerical results: asymptotically flat case

We shall only consider gravitational perturbations in D = 7 so JL = 0, N = 2.
For given l, m, our strategy (following Ref. 28) is to start with small OH and
search the interval 0 :S w :S mOH for a solution of the radial equation that is regular
on the future horizon and outgoing at infinity. This is then repeated for increasing
values of OH up to the maximum value.
A convenient object to consider is the ratio

(49)

We are looking for w for which this ratio diverges, corresponding to a purely outgoing
solution.

hIn higher dimensions, it appears that even a mass term for a scalar is not enough to lead to a
superradiant instability, at least for MP black holes with a single non-vanishing angular momen-
tum. 35
 1373

In all cases we have examined, the qualitative form of the potential is the same
as we found in the WKB analysis above, i.e., a potential barrier with a positive
maximum separates the classically allowed regions near the horizon and far from
the black hole. The corresponding behaviour of the solution \If is: oscillation near
the horizon, exponential growth in the potential barrier region and then oscillation
out to infinity. If the potential barrier is large then this implies that the amplitude
of oscillation far from the black hole will be large. However, we are looking for a
mode with Zin = O. From (40), such a mode obeys

lim 1\lf12 = mD H - w. (50)

T----l-(X) W

Hence, unless w is very small, such a mode will not have a large amplitude. Hence
it seems very unlikely that we will find a suitable mode when the potential barrier
is large. Phrasing the argument slightly differently, if \If is large then Zin andj or
Zout must be large compared with the left hand side of equation (40), which implies
Z ~ 1. More physically, if the potential barrier is large then one expects almost
perfect reflection and very little transmission, so the amplitude of \If is much greater
far from the black hole than near the horizon.
This argument suggests that we should examine the case for which the potential
barrier is smallest. The potential barrier is minimized when I is as small as possible
and m as large as possible. The most favourable case (using (22)) is therefore likely
to be I = 2, E = 1, m = 4. Our numerical results are shown in figure 1, where
we plot Z against wj(mDH) for DH jDmax = 0.5,0.7,0.9,0.99,0.999 where Dmax =
JNj(N + 1)/,+ is the upper bound on DH . The curves have the same qualitative
shape as for the Kerr black hole,28 i.e., Z is very close to 1 for small w, then increases
to a maximum near w = mD H and decreases back to 1 at w = mD H (the latter
property follows from equation (40)). The position of the maximum tends towards
w = mD H as DH ---> Dmax. The largest value for Z is Z = 1.115 so there is no
sign of Z diverging anywhere, as would be required for an instability. Note that
the amplification of energy flux in superradiant scattering is given by Z2 so the
maximum amplification apparent in our data is about 24%, and is achieved as the
black hole tends to extremality and w ---> mDH-. This is just as for Kerr, although
for Kerr, the maximum amplification is much greater: 138%.30
We have repeated our analysis for other values of (I, m). The results are qualita-
tively similar to the case we have just discussed. For DH IDmax = 0.99, the largest
value of Z obtained for E = 1 and (I, m) = (3,5), (3,3) was 1.056,1.000 respectively,
reflecting the fact that decreasing m tends to increase the potential barrier. In figure
2 we exhibit how Z varies with I with m = mmax(l) = 1+ 2. The largest value of Z
occurs for the I = 2, m = 4 case discussed above, and Z decreases monotonically to
1 as l increases, in agreement with our WKB analysis.
1374

1.1

1.075

Z
1.05

1.025

0.0 0.25 0.5 0.75 1.0

Fig. 1. Plots of Z againt wl(mOH) for (from bottom to top) OH lOmax = 0.5,0.7,0.9,0.99,0.999
with E = 1, l = 2, m = 4. Note that Z = 1 for w = mOH but this point has been deleted from the
topmost two curves to make the figure clearer.

<>

1.08

1.06

Z 1.04
<>

<>
1.02
<>
<>
<>
<> <>
1.0 <>

2.5 5.0 7.5 10.0 12.5

Fig. 2. Plot of Z against l = 2,3, ... for m = mmax = l+2, E = 1, 0H lOmax = 0.99, wl(mOH) =
0.99.
 1375

7. Discussion
We have shown that there exists a class of higher-dimensional rotating black hole
solutions for which linearized gravitational perturbations can be studied analyti-
cally. We have concentrated on the particular case of tensor modes. The equation
of motion for such modes reduces to a single ODE governing the behaviour in the
radial direction. In the asymptotically flat case, our analysis of this equation shows
no evidence of any instability of the black holes.
We have studied the case of asymptotically anti-de Sitter black holes in some
detail. Our main result is that the superradiant instability of such black holes dis-
cussed in Ref. 10 occurs precisely when the angular velocity of the black hole exceeds
the speed of light (in the sense that the co-rotating Killing field becomes space-like
on the conformal boundary). In other words, the instability occurs precisely when
the stability argument of Ref. 10 fails. Our results for small AdS black holes also
enable us to address a loop-hole in the proof of Ref. 10, namely that it could be
invalidated by the presence of a tachyonic scalar field (obeying the Breitenlohner-
Freedman bound), which violates the dominant energy condition. For small black
holes at least, such scalars do not behave any differently from more massive scalars,
i.e., the threshold of instability is unaffected. It would be interesting to see whether
the methods of Refs 10 and 31 could be combined to give a more general proof of
this.
Something that has not been addressed in the literature is the end-point of the
instability of Ref. 10. For the super-radiant instability of a Kerr black hole in the
presence of a massive scalar, the evolution is clear: the black hole gradually loses
energy and angular momentum to bound superradiant modes of the scalar field.
These couple non-linearly to gravitational waves (and thereby to non-superradiant
modes of the scalar field) so this energy and angular momentum is ultimately radi-
ated to infinity. After a very long time the black hole will have lost all of its angular
momentum this way.
In the AdS case, the evolution of the instability must be rather different. A
black hole that is initially rotating faster than light will lose energy and angular
momentum into superradiant modes of all fields of the theory under consideration.
However, in AdS this cannot be radiated to infinity so instead the energy and
angular momentum in fields outside the black hole must accumulate over time and
backreaction will become important. If the system ultimately settles down to an
equilibrium state then this must be described by a new stationary, asymptotically
AdS black hole solution. Presumably the angular velocity of this new solution will
not exceed the speed of light. Note that this argument does not depend on the
details of the theory: it would be valid if gravity were the only field present, so
there should even exist new vacuum black hole solutions.
Further evidence for the existence of new solutions comes from our analysis of
modes at the threshold of instability. Since these have w = mDH, they are preserved
by the co-rotating Killing vector field ~. However they break the symmetries gener-
1376

ated by 8/ 8t and 8/ 81jJ. The existence of these modes could indicate the presence of
new branches of solutions that bifurcate from the solutions of Refs 8,9 at the critical
(m-dependent) value of DH . The new solutions would not be invariant with respect
to 8/ 8t or 8/ 8'1j). In other words, they would llOt be stationary or axisymmetric.
However, ~ would still describe a symmetry. Since the bifurcation point occurs when
the original black hole is rotating faster than light, t; would be spacelike near in-
finity but timclike near the horizon. So, near the bifurcation point, the new branch
of solutions would correspond to black holes that are not stationary in the usual
sense but nevertheless admit a Killing field that is timelike ncar the horizon. This
Killing field becomes spacclike outside an "ergosphere" (this is what happens for
the co-rotating Killing field of the Kerr black hole). However, if such solutions do
exist, they are themselves rotating faster than light and therefore seem very likely
to be unstable.
What happens as Olle moves further away from a bifurcation point? Obviously
we can only speculate, but one possibility is that, if one moves sufficiently far along
one of the new branches of solutions, one reaches solutions for which ~ is timelike
everywhere outside the horizon. These would correspond to genuinely stationary
black holes which are nevertheless nonaxisymmetric. There would be no violation
of the theorem that a stationary black hole must be axisymmetric 37 because this
theorem assumes that the stationary Killing field is not normal to the event horizon
whereas ~ is normal to the horizon of all the black holes we have been discussing.
If such black holes exist then it is natural to guess that these should be the new
solutions describing the endpoint of the superradiant instability.
The possibility of a black hole being stationary with respect to a Killing field that
does not approach the "usual" generator of global AdS time translations deserves
further comment. Consider AdS5 in global coordinates:

ds 2 = - (1 + ~~) dt + (1 + ~~)
2
-1 dr 2 + r2 (de 2 + sin 2 Bd4>r + cos 2 Bd4>~). (51)

The generator of global time translations 8/ 8t is a globally timelike Killing field

whose norm diverges at infinity. Now consider the Killing field
8 1 8 1 8
- + - - +£-84>2
8t £ 84>1
-. (52)

This is globally timelike with constant norm. The same construction works in any
odd-dimensional AdS spacetime. Therefore there are (at least) two qualitatively
different ways that an odd-dimensional asymptotically AdS space-time can be sta-
tionary: the generator of time-translations could have either unbounded norm or
bounded norm. This does not appear to have been discussed before. The known
AdS black hole solutions are stationary in both senses because they admit Killing
fields that have the same asymptotic behaviour as 8/ 8t and 8/ 84>i above. However
there may well exist AdS black holes with less symmetry that are stationary only
with respect to a Killing field of bounded norm.
 1377

We do not know whether the stationary nonaxisymmetric black holes discussed

above must be of this form. If they are, then, since the stationary Killing field must
be normal to the horizon, and since a Killing field of bounded norm is null on
the conformal boundary, such black holes must be rotating at the speed of light.i
Superficially, this makes sense because the superradiant instability "switches off"
when the rotation of a black hole no longer exceeds the speed of light so one might
expect the black hole to evolve to a final state rotating precisely at the speed of
light. However, the evolution of the instability is a time-dependent process, during
which the notion of angular velocity is not even defined, so we see no reason why
the final time-independent state could not be rotating slower than light.
Finally, we note that supersymmetric black holes exist in AdS5 . 38 Supersymme-
try guarantees the existence of a non-spacelike Killing vector field that is normal
to the event horizon 36 and timelike with bounded norm at infinity,39 i.e., these so-
lutions have precisely the behaviour that we have just discussed and hence rotate
at the speed of light. However, these solutions admit extra Killing fields analagous
to [) / [)¢i (and hence also [) / 8t) above so they are also stationary in the usual sense
(i.e. with respect to a Killing field of unbounded norm). The existence of these ex-
tra Killing fields appears unrelated to supersymmetry, which raises the question of
whether there exist more general supersymmetric black hole solutions without these
extra symmetries. Such black holes would be nonaxisymmetric, and stationary only
in the new sense that we have been discussing.

References
1. R. C. Myers and M. J. Perry, Annals Phys. 172, 304 (1986).
2. R. Emparan and H. S. Reali, Phys. Rev. Lett. 88 (2002) 101101 [arXiv:hep-
th/0110260].
3. R. Emparan and R. C. Myers, JHEP 0309, 025 (2003) [arXiv:hep-th/0308056].
4. G. Gibbons and S. A. Hartnoll, Phys. Rev. D 66 (2002) 064024 [arXiv:hep-
th/0206202].
5. H. Kodama and A. Ishibashi, Prog. Theor. Phys. 110, 701 (2003) [arXiv:hep-
th/0305147j.
6. A. Ishibashi and H. Kodama, Prog. Theor. Phys. 110, 901 (2003) [arXiv:hep-
th/0305185j.
7. J. M. Maldacena, Adv. Theor. Math. Phys. 2, 231 (1998) [Int. J. Theor. Phys. 38,
1113 (1999)] [arXiv:hep-th/9711200], S. S. Gubser, 1. R. Klebanov and A. M. Polyakov,
Phys. Lett. B 428, 105 (1998) [arXiv:hep-th/9802109], E. Witten, Adv. Theor. Math.
Phys. 2, 253 (1998) [arXiv:hep-th/9802150j.
8. S. W. Hawking, C. J. Hunter and M. M. Taylor-Robinson, Phys. Rev. D 59 (1999)
064005 [arXiv:hep-th/9811056].

iThe terminology here may be a little confusing. If one defines angular velocity for such black
holes in the usual way then it will vanish because the stationary Killing field is normal to the
horizon. However, this is angular velocity defined with respect to a stationary bulk observer. Since
the stationary Killing field is null on the conformal boundary, such an observer actually rotates at
the speed of light with respect to the boundary and hence so does the black hole.
1378

9. G. W. Gibbons, H. Lu, D. N. Page and C. N. Pope, J. Geom. Phys. 53 (2005) 49

[arXiv:hep-th/0404008], Phys. Rev. Lett. 93, 171102 (2004) [arXiv:hep-th/0409155J.
10. S. W. Hawking and H. S. Reali, Phys. Rev. D 61, 024014 (2000) [arXiv:hep-
th/9908109j.
11. T. Damour, N. Deruelle and R. Ruffini, Lett. Nuovo Cim. 15 (1976) 257.
12. T. J. M. Zouros and D. M. Eardley, Annals Phys. 118 (1979) 139.
13. S. Detweiler, Phys. Rev. D 22, 2323 (1980).
14. V. Cardoso, O. J. C. Dias, J. P. S. Lemos and S. Yoshida, Phys. Rev. D 70, 044039
(2004) [Erratum-ibid. D 70, 049903 (2004)j [arXiv:hep-th/0404096j.
15. V. Cardoso and O. J. C. Dias, Phys. Rev. D 70, 084011 (2004) [arXiv:hep-th/0405006j.
16. S. A. Teukolsky, Phys. Rev. Lett. 29, 1114 (1972), Astrophys. J. 185,635 (1973).
17. V. P. Frolov and D. Stojkovic, Phys. Rev. D 67,084004 (2003) [arXiv:gr-qc/0211055j.
18. D. Ida, Y. Uchida and Y. Morisawa, Phys. Rev. D 67, 084019 (2003) [arXiv:gr-
qc/0212035j.
19. M. Vasudevan, K. A. Stevens and D. N. Page, Class. Quant. Grav. 22 (2005) 1469
[arXiv:gr-qc/0407030j.
20. M. Vasudevan, K. A. Stevens and D. N. Page, Class. Quant. Grav. 22 (2005) 339
[arXiv:gr-qc/0405125], H. K. Kunduri and J. Lucietti, Phys. Rev. D 71 (2005) 104021
[arXiv:hep-th/0502124], M. Vasudevan and K. A. Stevens, Phys. Rev. D 72 (2005)
124008 [arXiv:gr-qc/0507096j.
21. H. K. Kunduri, J. Lucietti and H. S. Reali, Phys. Rev. D 74, 084021 (2006) [arXiv:hep-
th/0606076j.
22. H. K. Kunduri, J. Lucietti and H. S. Reali, in preparation.
23. G. W. Gibbons, M. J. Perry and C. N. Pope, Class. Quant. Grav. 22 (2005) 1503
[arXiv:hep-th/0408217j.
24. N. P. Warner, Proc. Roy. Soc. Lond. A 383,217 (1982).
25. C. N. Pope, J. Phys. A 15, 2455 (1982).
26. Y. Morisawa and D. Ida, Phys. Rev. D 71, 044022 (2005) [arXiv:gr-qc/0412070j.
27. V. Cardoso, G. Siopsis and S. Yoshida, Phys. Rev. D 71, 024019 (2005) [arXiv:hep-
th/0412138j.
28. W.H. Press and S.A. Teukolsky, Astrophys. J. 185, 649 (1973).
29. J.B. Hartle and D.C. Wilkins, Commun. Math. Phys. 38, 47 (1974).
30. S.A. Teukolsky and W.H. Press, Astrophys. J. 193, 443 (1974).
31. P. Breitenlohner and D. Z. Freedman, Annals Phys. 144, 249 (1982).
32. S. S. Gubser and I. Mitra, JHEP 0108, 018 (2001) [arXiv:hep-th/0011127j.
33. R. Gregory and R. Laflamme, Phys. Rev. Lett. 70, 2837 (1993) [arXiv:hep-
th/9301052j.
34. B.F. Schutz and C.M. Will, Astrophys. J. 291, L33 (1985).
35. V. Cardoso and S. Yoshida, JHEP 0507, 009 (2005) [arXiv:hep-th/0502206j.
36. H. S. Reall, Phys. Rev. D 68, 024024 (2003) [Erratum-ibid. D 70, 089902 (2004)J
[arXiv:hep-th/0211290j.
37. S. Hollands, A. Ishibashi and R. M. Wald, arXiv:gr-qc/0605106.
38. J. B. Gutowski and H. S. Reall, JHEP 0402, 006 (2004) [arXiv:hep-th/0401042],
JHEP 0404 (2004) 048 [arXiv:hep-th/0401129], Z. W. Chong, M. Cvetic, H. Lu and
C. N. Pope, Phys. Rev. D 72, 041901 (2005) [arXiv:hep-th/0505112], Phys. Rev. Lett.
95, 161301 (2005) [arXiv:hep-th/0506029], H. K. Kunduri, J. Lucietti and H. S. Reali,
JHEP 0604, 036 (2006) [arXiv:hep-th/0601156j.
39. J. P. Gauntlett and J. B. Gutowski, Phys. Rev. D 68, 105009 (2003) [Erratum-ibid.
D 70, 089901 (2004)] [arXiv:hep-th/0304064].
 GRAVITATING NON-ABELIAN SOLITONS AND HAIRY BLACK
HOLES IN HIGHER DIMENSIONS

MIKHAIL S. VOLKOV
Laboratoire de Mathematiques et Physique Theorique, Universite Franr;ois-Rabelais Tours,
Federation Denis Poisson - CNRS, Pare de Grandmont, 37200 Tours, FRANCE
[email protected]·fr

This is a short review of classical solutions with gravitating Yang-Mills fields in D > 4
spacetime dimensions. The simplest SO(4) symmetric particlelike and SO(3) symmetric
vortex type solutions in the Einstein-Yang-Mills theory in D = 5 are considered, and
their various generalizations with or without an event horizon, for other symmetries, in
more general theories, and also in D > 5 are described. In addition, super symmetric
solutions with gravitating Yang-Mills fields in string theory are discussed.

1. Introduction

In this mini-review the following two questions will be addressed:

• What is known about solutions with gravity-coupled non-Abelian gauge

fields in D > 4 spacetime dimensions ?
• Why should one study such solutions?

In answering these questions we shall briefly consider the simplest gravitating non-
Abelian solutions in D = 5 and shall then discuss their known generalizations.

2. Einstein-Yang-Mills theory in D=4

Before going to higher dimensions, it is worth reminding the situation in D = 4. It
is well known that neither pure gravity nor pure Yang-Mills theory in D = 4 admit
solitons, by which we mean stationary and globally regular (but not necessarily
stable) solutions with finite total energy.
In the case of pure Einstein gravity with the action

SE = ~GJRF9d4x
167f
(1)

the statement is the content of the theorem l of Lichenrowitcz. The vacuum gravity
admits however solutions describing localized stationary objects with finite mass,
but these are not globally regular: black holes.
The action for the pure Yang-Mills (YM) theory with a compact and simple
gauge group Q is given by

S YM -- - 1
4Ke 2
J ;--:::d4 x,
trFp,v F p,v v-g (2)

where Fp,v = op,Av-ovAp, -i[Ap" Avl. Here Ap, = A~Ta and Ta are the gauge group
generators normalized such that tr(T aT b ) = K bab' If Q=SU(2) then a = 1,2,3 and
T a = ~ Ta where Ta are the Pauli matrices. The gauge coupling constant is denoted

1379
1380

bye. Since this theory is scale invariant, the non-existence of stationary solitons,
expressed by the statement "there are no classical glue balls" , essentially follows 2 - 4
from the scaling arguments.
The physical reason for the non-existence of solitons in the above two cases is
clear: one cannot have equilibrium objects in systems with only attractive (gravity)
or only repulsive (Yang-Mills) interactions. However, in systems with both gravity
and Yang-Mills fields described by the Einstein-Yang-Mills (EYM) action

(3)

both attractive and repulsive forces are present, and so the existence of solitons is
not excluded. At the same time, the existence of solitons in this case is not guaran-
teed either, and so it was a big surprise when such solutions were constructed 5 by
Bartnik and McKinnon. As this was actually the first known example of solitons in
self-gravitating systems, a lot of interest towards the Einstein-Yang-Mills system (3)
has been triggered. This interest was further increased by the discovery in the same
system of the first known example 6 - 8 of hairy black holes which are not uniquely
characterized by their conserved charges and so violate manifestly the no-hair con-
jecture. 9 Further surprising discoveries followed when these static EYM solitons and
black holes were generalized lO- 12 to the non-spherically symmetric and non-static
cases. This has revealed that the EYM solutions provide manifest counter-examples
also to a number of classical theorems originally proven for the Einstein-Maxwell
system but believed to be generally true, as for example the theorems of staticity,
circularity, the uniqueness theorem, Israel's theorem an so on (see 13 for a review).
As a result, the EYM theory has become an interesting topic of studies.

3. Pure gravity and pure Yang-Mills in D=5

Going to higher dimensions, let us first consider separately some simplest solutions
for pure gravity and pure Yang-Mills theory in D = 5. We shall denote the spacetime
coordinates by x M where M = 0,1,2,3,4.

3.1. Pure gravity

Vacuum gravity in D > 4 has been much studied and many interesting solutions
have been obtained (see 14 for a recent review), but we shall just mention a couple
of the simplest ones. If gffl/(xO"), where /-L, v, a = 1,2,3,4, is a Euclidean Ricci flat
metric (gravitational instanton), then the 5D geometry

ds 2 = -de + gEMl/ dxMdxl/

will also be Ricci flat. One obtains in this way static and globally regular vacuum
solutions in D = 5. However, since there are no 15 asymptotically Euclidean gravita-
tional instantons, these 5D solutions will not be asymptotically flat (although they
can be asymptotically locally flat). There exist also localized and asymptotically
 1381

flat solutions - black holes - but these are not globally regular. In the simplest case
with the SO (4) spatial symmetry this is the Schwarzschild black hole,

N=l- C:f
If 8/ 8x 4 is a symmetry and 9/-w is a Ricci fiat Lorentzian metric / JL = 0,1,2,3/
then the 5D metric

will also be Ricci flat. Choosing 9flv to be a 4D black hole metric gives then a
one-dimensional linear mass distribution in D = 5 - black string.

3.2. Pure Yang-Mills theory: particles and vortices

Since the Yang-Mills theory in D = 5 is not scale invariant, one can have localized
and globally regular object made of pure gauge field. We shall call them 'Yang-Mills
particles', these are essentially the 4D YM instantons uplifted to D=5. Specifically,
if A~(xV) is a solution of the 4D Euclidean YM equations then

/JL= 1,2,3,4/

will be a solution of the 5D YM equations describing a static object whose 5D

energy coincides with the 4D instanton action 16

The equality here is attained for self-dual configurations, with PflV = *PflV , in which
case n E ;Z is the number of the 'YM particles'. Their mutual interaction forces being
exactly zero, the particles can be located anywhere in the 4-space.
If 8/ 8x 4 is a symmetry, then choosing

/i,k = 1,2,3/
the YM energy per unit x4,

E=-2
1 /((8i H a +Eabc Abi H c)2 +-21 (Faik )0)d
~ 47flnl
3x~--2-'
2e e

coincides with the energy17 of the D=3 YM-Higgs system. Its absolute minima for
a given n are the BPS monopoles, and when lifted back to D=5 they become one-
dimensional objects - 'YM vortices'. The number of vortices, n, can be arbitrary,
and they can be located anywhere in the 3-space of xk.
\Ve shall now describe the self-gravitating analogs 18 for the YM particles and
also for the YM vortices. Practically all other known solutions with gravitating
gauge fields in higher dimensions are generalizations of these two simplest types.
1382

4. Gravitating YM particles 18
The EYM theory with gauge group 8U(2) in D=5 is defined by the action

5 = j£5y<5)gd 5x = j (_1_ 167fG

(5)R __ 1_
4e 2
a F aMN ) y<5)gd 5x
FM
N ,
(4)

where F7:1N = 8MA'N - 8 N A'M + EabcA~AN (a = 1,2,3). The Newton constant

G and gauge coupling e being both dimensionful, [G 1 / 3 ] = [e 2 ] = [length], one can
define the dimensionless coupling parameter as

Let us try to construct the 5D counterparts of the 4D solitons5 of Bartnik and

McKinnon: particle-like objects localized in 4 spatial dimensions. In the simplest
case they are static and purely magnetic and have the 80(4) spatial symmetry. The
fields are then given by

ds
2
= e
2
{ -cr(r)2 N(r)dt
2
+ ~~:) + r2 dO~ } , A a = (1 + w (r ) ) ga , (5)

where ga are the invariant forms on 53. With N == 1 - K,m(r)/r2, the EYM field
equations for the action (4) reduce to
Wi
r 2Nw" +rw' +K,(m- (w 2 _1)2)_ = 2(w 2 -l)w, (6)
r
(7)

(8)

If K, = 0 then (J = N = 1 and the geometry is flat. The solution of Eq.(6) is the 5D

YM particle, which is the same as the 4D one-instanton solution:
1 - br 2
w=-----::---,:::- (9)
1 + br 2 '
Here b > 0 is an integration constant that determines the size of the particle. This
solution is everywhere regular and its energy is the ADM mass: E = MADM =
m(oo) = ~.
The next question is what happens if K, =I 0 ? Let us consider asymptotically flat
solutions with finite ADM mass. If they are globally regular, then they will have
a regular origin at r = 0, and for small r the local solution of Eqs.(6)-(8) will be
given by

(10)
One can also consider black hole solutions, in which case

(11)
 1383

In both cases, using the boundary condition (10) if rh = 0 and (11) for rh > 0 and
integrating (7) gives the ADM mass functional
r2
MADM[w(r)] = m(oo) = ---.l!:.
K,
+ 1- Th
00
dr (r 2 w'2
r
+ (w 2 - 1)2)a-(r). (12)

This shows that one should have w( (0) = ±1 for the mass to be finite. Let us suppose
that there is a solutions with MADM < 00 and let w(r) be the corresponding gauge
field amplitude. The first variation of the mass functional (12) should then vanish
under all variations w(r) ---t w(r) + ow(r) respecting the boundary conditions. In
particular, if we consider special variations w(r) ---t w(Ar) then one should have
ddAM[w(Ar)] = 0 for A = 1. However, (12) implies that ddAM[w(Ar)] < 0 for any A.
As a result, there are no solutions with finite ADM mass.
We therefore conclude that the Bat space YM particles do not admit self-
gravitating generalizations with finite energy. One can say that they get com-
pletely destroyed by gravity, as in some sense they resemble dust, since their energy
momentum tensor is

and in addition they can be scaled to an arbitrary size. As a result, when gravity is
on, the repulsion and attraction are not balanced and so equilibrium states are not
possible. If one nevertheless tries integrating Eqs.(6)-(8) starting from the regular
origin, say, to see what happens, one finds a peculiar quasi-periodic behavior shown
in Fig.1. This can be explained qualitatively: integrating Eq.(6) with the regular

8 -

b=l

'--_~~---L_~'_'~~l
f--Lf\J- 2 10 ' -.~,~~410::-_~~~~60

In{r)

Fig. 1. Globally regular solution to Eqs.(6)-(8),(1Q) with b = 1, f£ = 10- 8 .

boundary condition (10) gives

W '2 - (w 2 - 1)2 = - iToo (In(a-2N))'w '2 dT, (13)

1384

with' == lr lr'
== rVH This describes a particle moving with friction in the inverted
double-well potential U = -(w 2 - 1)2. At T = -CXJ the particle starts at the local
maximum of the potential at w = 1, but then it looses energy due to the dissipation
and cannot reach the second maximum at w = -1. Hence it bounces back. For
large T the dissipative term tends to zero, and the particle ends up oscillating in
the potential well with a constant energy. After each oscillation the mass function
m increases in a step-like fashion, and for large r one has mer) ~ T ~ In r. As a
result, there emerges an infinite sequence of static spherical shells of the YM energy
in the D=5 spacetime. These arguments easily generalize to the black hole case.

4.1. Other particle like solutions 19- 26

A number of generalizations of the above results have been considered. It turns out
that the divergence of the mass is the generic property of the static EYM solutions
with the spherical symmetry in D > 4. In particular, it has been shown that for
all static EYM solutions with gauge group Q =SO( d) and with the spatial SOC d)
symmetry in D > 4 dimensions, where d = D if D is even and d = D - 1 if D is
odd, the ADM mass diverges. This concerns both globally regular and black hole
solutions, and also solutions with a A-term.
It turns out, however, that one can nevertheless have finite mass particlelike
solutions in D > 4 at the expense of adding the higher order curvature corrections to
the EYM Lagrangian. The gravitational term of the action, R, is then 'corrected' by
adding all possible Gauss-Bonnet invariants, while the Yang-Mills term, tr(F 1',* F) ,
is augmented by adding terms of the type tr(F /\ *F), where F = F /\ ... /\ F.
It is important that after taking these corrections into account the field equations
remain the second order. These equations admit localized solutions with a finite
mass which have been studied 19 ,2126 quite extensively.

4.2. Gravitating Yang monopoles 27

Flat space 4D YM instantons with Q=SU(2) can also be used to construct self-
gravitating solutions in D = 6. The YM instanton lives in this case on the 54
angular part of the SO(5)-invariant 5-space, the ansatz for the metric and gauge
field being
dr2
ds 2 = -0'(r)2 N(r)dt 2 + N(r) + r2(de + sin 2 x dr2 §) , A a = (1 + w(x)) aa.
YM equations decouple and admit a solution w(x) = cos X, whose energy mo-
mentum tensor does not depend on r and determines the source for the Einstein
equations. The latter give
2Gm(T)
0' = 1, N=l- 3 ' rn' = 81f,
r
and so the mass is linearly divergent for large T. This solution can be generalized 27
also to D = 2k + 2 dimensions, choosing SO(2k) as the gauge group. In this case
 1385

5. Gravitating YM vortices 18
Let us return to the 5D EYM model defined by Eq.(4) and assume the existence of
ox
a hypersurface orthogonal Killing vector 0/ 4 , such that the metric is

gl\1Ndxl\1dx N = e-(g,wdxf1dxV + e2((dx4)2,

/JL,v=0,1,2,3j. (14)

Inserting this to (4) reduces the 5D EYM model to the 4D EYM-Higgs-dilaton

theory,

Let us assume the 80(3) symmetry for the 4D fields,

e-((r)ds 2 = _e 2v (r)dt 2 + dr 2 + R 2 (r) dO~,

(15)

where n k is the unit normal to the 2-sphere, dO~ = dnkdnk. The independent EY1VI
equations can be rewritten 18 in the form of a seven-dimensional dynamical system

(16)

with Yk = {w, Wi, h, hi, Z = (" R, R'}. Solutions of these equations show interesting
features which can be qualitatively understood by studying the fixed points of the
system. The system has the following fixed points.
I. The origin: (w, h, Z, R) = (1,0,0,0). Assuming t.hat this fixed point is attained
for r = 0, the local behavior of the solution for small r is

II. Infinity: (w, h, Z, I/R) = (0,1,0,0). This fixed point. can be reached for
r -> CXJ, in which case

w = Arce- r + o(e- r ), Z = ""Qr- 2 + O(r- 3 InT),

h = 1- Cr- 1 + O(r- 2 In r), (18)
R = r - mIn T + m 2 r- 1 ln T - ro + ~fr-l + O(r- 2 Iwr).

In these expressions a, b, A, v, C, Q, rOC)) I' are eight free parameters. The ADM
mass is MADl\! = 3(C + (2 + ",,)Q).
1386

III. "Warped" Ad83 x 8 2 . The fixed point values of the functions are expressed
in this case in terms of roots of the cubic equation 4q3 + 7q2 + llq = 1 as
2 (llq-l)(I-q) 2 1 -q 2 4q2-13q+l
(19)
R = /l, (4q2 _ 13q + 1)' h = Ji2"' z = - (4q + I)R2 .

Evaluating,
1.27 0.31 t=
w=0.29, h= %' Z=± %' R=0.75 y /l" (20)

which gives an exact non-Abelian solution with the geometry

2
ds 2 = e 2(l+Kh )Zr dt 2 _ dr 2 _ e 2Zr (dx4)2 _ R2 df2~. (21)

Linearizing Eqs.(16) around this fixed point gives the characteristic eigenvalues
2.77 2.47 2.12 0.61 .1.24 0.88 1.54)
( - %,- %,- Vii,- Vii ±z Vii'+ Vii'+ Vii . (22)

II II II II
•
exteri9'~ solution

;,",~oo
•
III

I I I I .
K=O strong gravity
BPS monopole K«l K-l
weakly gravitating strongly gravitating
BPS monopole BPS monopole

Fig. 2. Schematic behavior of solutions of Eqs.16 interpolating between fixed points I and II.

Global solutions of Eqs.(16) can then be viewed as trajectories in the phase

space interpolating between fixed point I (origin) and fixed point II (infinity). Their
behavior is schematically sketched in Fig.2. For /l, = 0 the 4D solution is the flat
space BPS monopole and its 5D analog is the flat space YM vortex. For /l, ~ 1 the
solution is a weakly gravitating 5D vortex whose field configuration is only slightly
deformed as compared to the /l, = 0 case. The YM vortices therefore do generalize
to curved space, unlike the YM particles.

5.1. Strong gravity limit

Further increasing the gravitational coupling /l, deforms the YM vortex more and
more and the phase space trajectory approaches closer and closer the third fixed
point. Since the latter has complex eigenvalues with negative real part, the phase
trajectory gets attracted by this fixed point and oscillates around it for a while before
 1387

IC-O.310

Il~-~
-1
I / / - /w
O~~_--~_~ - , ::::J -2
-2 0 05
In(r)

Fig. 3. Left: For strongly gravitating solutions the field amplitudes oscillate around the values (20)
corresponding to the third fixed point. Right: The spiraling behavior of the solution parameters
in the strong gravity limit. One sees that for a given value of /'i, E [0.11,3.22] there exist several
(at least two) different solutions.

going to infinity. These oscillations manifest themselves in the solution profiles; see
Fig.3. A further increase of the gravitational force is not accompanied by an increase
of the value of K, but can rather be achieved by increasing the value of ~ = K,h'(O).
The parameters K,(~), MADM(~)' etc. then start developing spiraling oscillations
around some limiting values; see Fig.3.
Such strongly gravitating solutions have a regular core connected to the asymp-
totic region by a long throat - the region where the radius of the two-sphere, R,
is approximately constant and all other field amplitudes oscillate around the con-
stant values (20). Increasing the value of ~ = K,h'(O) this throat becomes longer and
longer, and finally the solution splits up into the two independent solutions: interior
and exterior, schematically shown in Fig.4.
The interior solution has a regular central core and approaches asymptotically
the 'warped ADS' geometry (21). In Fig.2 this solution corresponds to the phase tra-
jectory that starts at the fixed point I and after infinitely many oscillations around
the fixed point III ends up there. The exterior solution, shown in Fig.4, interpolates
between the 'warped ADS' and infinity. In Fig.2 this solution corresponds to the
phase trajectory that interpolates between the fixed points III and II. It does not
oscillate, since all 'repelling' eigenvalues (22) with positive real part are real. This
solution is somewhat similar to an extreme non-Abelian black string, since passing
to the Schwarzschild gauge where T" = R one has

-2
grr rv ( T" - T"h ) , gtt rv g44 = e 2( rv
(
T" - T"h
)2.02

where the event horizon radius is T"h = 0.42.

 1388

limiting solutions

strongly gravitating solutio

\ , I
(
I i

II i -+ interior + exterior I

I I
I I

\ I
~~
In(r)

Fig. 4. Left: Schematic behavior of solutions in the strong gravity limit. Right: The profiles of
the exterior part of the limiting solution.

5.2. Conclusions
Summarizing the above discussion, although the static and SO(4) symmetric YM
particles in D=5 get destroyed by gravity, in view of their scaling behavior, the static
and SO(3) symmetric solutions - YM vortices - do admit non-trivial curved space
generalizations, since their scale invariance is broken by the asymptotic value of the
'Higgs field' H a = A 4. These gravitatiI].g YM vortices comprise a one-parameter
family of globally regular solutions - the fundamental branch - that interpolates
between the flat space BPS monopole and the extreme non-Abelian black string.
For all these solutions the YM field amplitude w is positive definite. In addition,
there exist also excitations over the fundamental branch. These are solutions for
which the YM field amplitude w oscillates around zero value, these solutions do
not have the flat space limit.

6. Generalized YM vortices
The gravitating YM vortex solutions of the fundamental branch (but not the excited
ones) have been generalized in a number of ways.

6.1. YM black sirings 29

From the 4D viewpoint the YM vortices are regular gravitating solitons. It has
been known for quite a long time 28 that gravitating solitons can often be general-
ized to include a small black hole in the center. Technically this requires replacing
the boundary conditions at the regular origin (conditions of the type (10)) by the
boundary conditions at the regular event horizon (conditions of the type (11)). Such
a procedure has been used 29 to promote the regular YM vortices to black strings.
For a given K, there can be several YM vortex solutions corresponding to different
parts of the spiraling curves in Fig.3. This property generalizes also to the black
 1389

string case, where one also finds several black string solutions for a given ~, provided
that their even horizon radius Th is small enough. As Th increases, these solutions
approach each other and finally merge for some maximal value Thax(~). There are no
black strings with Th > Thax(~), and so black strings exist only in a finite domain
of the (~, Th) parameter plane. Black strings and regular vortex solutions with a
A_term 30,31 have also been considered.

6.2. 'Twisted'solutions 32 ,33

When performing the dimensional reduction from D=5 to D=4 in Eqs.(14) and (15)
it was assumed that the Killing vector CJ/CJx 4 is hypersurface orthogonal. Relaxing
this condition, Eq.(14) generalizes to
N
gMNd:rMdx = e-c,glwdxl-'dx V + e2C,(dx4 + Wl-'dxl-')2,

where the twist ]iVI-' can be viewed as a 4D vector field. Inserting this to the action
(4) gives instead of (15) a four dimensional EYM-Higgs-dilaton+ U(l) model. The
solutions of this ·twisted' model carry an additional U(l) charge under the vector
field WI-" This charge can be of electric 32 or magnetic 33 type. When this charge
vanishes, the solutions reduce to the YM vortices/black strings.

6.3. Deformed and stationary solutions 34 - 36

The 5D YM vortices/black strings. with or without twist, have been also general-
ized 34- 36 to the case where, after the dimensional reduction to D=4, the fields are
chosen to be static and axially symmetric, rather than spherically symmetric. The
ansatz for the gauge field A~ and H a contains in this case two integer winding num-
bers, n, TTL If n = 1, rn = 0 then the solutions are spherically symmetric monopoles.
For n > 1, Tn = 0 one obtains axially symmetric solutions of the multimonopole
type. Solutions with Tn = 1 describe monopole-antimonopole pairs. Both regular
solutions 35 and black strings 36 have been considered within this approach. In all
cases the existence of several solutions for a given value of ~ has been detected.
A further generalization is achieved 35 ,36 by returning back to D=5 an performing
a Lorentz boost along the x4 direction. This operation has a non-trivial effect on
the 4D configurations: it produces stationary spinning solutions.

6.4. Non-Abelian braneworlds 37,39-4 2

Solutions with gravitating gauge fields can also be considered in the context of the
braneworld models. As was discussed above, the non-Abelian monopole describes a
one-dimensional object (vortex) in D=5. It will therefore describe a two-dimensional
object (domain wall) in D=6 and a three-dimensional object (3-brane) in D=7. In
1390

all cases the SU(2) Yang-Mills field and the triplet Higgs fields can be chosen to be
static and spherically symmetric,

(23)

with r2 = xkxk. For the 3-brane in D=7 one chooses xk /k = 1,2,3/ to be coordi-
nates orthogonal to the brane, the 7D metric being

(24)

where yl" / JL = 0,1,2,3/ are coordinates on the brane. Integrating the coupled
EYM-Higgs-A equations reveals 37 that there are solutions satisfying the regularity
condition at the brane, A(O) = B(O) = w(O) = 1, h(O) = 0, for which A(r) --> 0
for r --> 00. Such solutions describe globally regular braneworlds confined in the
monopole core with gravity localized on it. Solitonic braneworlds have also been
studied in systems with a gravitating global monopole 38 and also for the local and
global monopoles 39 coupled to each other and to gravity in D=7.
More general solutions of the n-brane type have been obtained 4o - 42 by oxidizing
the D = 5 YM vortices to D = 4 + n dimensions. The metric is then chosen to be
n
ds 2 = _et,o(r l (d y O)2 +L ec'k(rl(dyk)2 + B(r)<5ik dx i dx k (25)
k=l

with the gauge field given by

(26)

What is interesting, the analog of the fixed point (19) can be obtained 4o within this
approach for any n, also expressed in terms of roots of a cubic polynomial.

7. Non-Abelian solitons in string theory

Coming to the question of why one should study gravitating Yang-Mills fields in
higher dimensions, one can say that (apart form pure curiosity) the motivation for
this is provided by string theory. Gravitating Yang-Mills fields enter supersymmetry
multiplets of the supergravity (SUGRA) theories to which string theory reduces at
low energies. The knowledge of the basic solutions for gravitating gauge fields can
therefore be useful for constructing solutions in low energy string theory. However,
unlike solutions of the pure vacuum gravity, solutions of the EYM theory will not
directly solve equations of SUGRA, since the latter generically contain additional
fields, as for example the dilaton field. Constructing supergravity solutions with
Yang-Mills fields is thus more complicated. However, imposing the supersymmetry
conditions one can sometimes reduce the problem to solving first order Bogomol'nyi
equations and not the second order field equations.
 1391

7.1. Heterotic solitons 43 ,44

The first example of supersymmetric solutions with gravitating Yang-Mills fields was
obtained by Strominger43 in the heterotic string theory. The low energy limit of the
latter is a super gravity whose bosonic sector contains a Yang-Mills field already
in D=lO. Strominger considers a 5-brane in D=lO and makes the 6+4 split of the
metric,

ds 2 = A(xJL)T]MNdyMdyN + B(X")JJLldxJLdxV,
where yM / M, M = 0,1,2,3,4,5/ are coordinates on the brane while the coordi-
nated of the 4-space orthogonal to the brane are xJL / JL = 1,2,3, 4j. He puts the
Yang-Mills field to the orthogonal Euclidean 4-space,

AM =0,
Since the Yang-Mills field in D=4 is conformally invariant and the relevant 4D
part of the metric is conformally fiat, the scale factor B(X") drops out from the
Yang-Mills equations. As a result, any fiat space self-dual Yang-Mills instanton in
D=4,

(27)

will solve the Yang-Mills part of the SUGRA equations. The metric functions A, B
as well as the axion and dilaton will then satisfy the Poisson equation on JE4 with the
source determined by FJLlI' and so the solution can be expressed in quadratures. As
there are many solutions of Eqs.(27), this gives a large family of heterotic 5-branes
in D=10. Performing then dimensional reductions, one can construct 44 yet many
more different solutions living in D < 10. All these are called heterotic solitons.

7.2. Non-Abelian vacua in gauged BUGRA s 48 ,50,51,55,57,58

Type I and type II string theories and also M theory reduce at low energies to su-
pergravities in D=10 and D=l1 whose multiplets do not contain Yang-Mills fields.
However, the latter appear when one performs dimensional reductions on internal
manifolds with non-Abelian isometries. This gives gauged SUGRAs in lower dimen-
sions whose gauge group is related to the isometry group of the internal manifold.
For example, the reduction of type II string theory on 8 5 gives the gauged SUGRA
in D=5 whose solutions are used in the AdS/eFT correspondence. Another exam-
ple is the reduction of M theory on 8 7 , which gives 45 the N=8 gauged SUGRA in
D=4 with the local SO(8)xSU(8) invariance. One can then study solutions in these
SUGRAs, and lifting them back to D = 10,11 will give vacua of string or M theory.
The field content of a gauged SUGRA model can be quite complicated, which is
why the gauge fields are often set to zero to find solutions. This gives, for example,
black holes with scalar hair 46 ,47 in N=8 SUGRA. However, in some cases one can
construct solutions also with gravitating gauge fields. Let us consider a SUGRA
1392

mode1 48 in D=4 which is a consistent truncation of the maximal N=8 SUGRA,

£4 = ~R - ~81"¢(YI¢ - ~ e2¢ F;vFalw + ~ (e- 2¢ + ee 2¢ + 4~). (28)

This theory contains a gravity-coupled Yang-Mills field with gauge group SU(2)
and the dilaton, whose potential depends on a real parameter f In the static and
spherically symmetric case the fields are given by
dSZ4 ) = _e 2V (p)dt 2 + e 2A (P)dp2 + r2(p)dD2,
A'kdxk = (w(p) - l)e a ikn i dnk, ¢ = ¢(p).
For one-shell configurations the amplitudes V, A, r, w, ¢ satisfy a system of second
order ODEs whose solutions can be studied 49 numerically. Instead of solving these
equations, one can also consider the conditions for the fields to have unbroken
supersymmetries. These conditions require the existence of a non-trivial spinor 10
satisfying the linear equations

0= _1_ rylla "10 + ~ eq)Fe + ~ (e-¢ - Cc¢)e.

V2 I 1"'1' 2 4 ".
_ 1 ¢ 1 _¢ ¢
0- Vl"e + ;7) e F,IJE + ;7) (e + ~e hl"f, (29)
2y 2 4y 2
where F = ~Ta F::{3,D,{3. In general these equations are inconsistent, since there
are 80 equations for 16 components of E, and so the only solution is E = O. However,
one can show 48 that if the background fields are such that the following conditions
are fulfilled,

"I = YL'2 BP
'I' N e A,

Wi rwB e -¢+A
=-N , N == PA' +1= j w 2 + p2, r' = Ne\ (30)

where Q is a constant and

1- w 2 r P 1_¢
P = e¢-- + -(e-¢ + ~e¢) B = --- + -e .
V2r 2V2 ' V2r 2 .
then there exist four independent solutions of Eqs.(29), which corresponds to the
N=l supersymmetry. Solutions 48 of the Bogomol'nyi equations (30) with the regular
boundary condition at the origin, p = r = 0, comprise a continuous family labeled
by ~ and show three completely different types of behavior, depending on the sign
on ~. For ~ > 0 the dilaton is everywhere bounded and the solutions approach the
AdS metric in the asymptotic region. For ~ < 0 the solutions are of the 'bag of
gold' type, since they have compact spatial sections with the topology of 53. The
geometry is generically singular at one pole of the 53. However, for ~ = - 2 the
solution is globally regular, the dilaton is constant, the geometry is that of JR.l x 53,
and the gauge field potential coincides with the invariant forms on 53. All these
solutions can be uplifted 48 to D = 11 to become vacua of M theory.
 1393
-----r
~-- ';0-0001
';---0

t= -0 5

t-O, ~-() 001

t;= JD

-
G
~---

o
-- ~~-~-

4
L L~~ _,_~ ~_ I _ __

Fig. 5. The asymptotically AdS and compact solutions of the Bogomol'nyi equations (30).

For ~ = 0 the solution is given by 50

ds 2 = 2 e2 4> { - dt 2 + dp2 + R2(p) ( d'l9 2 + sin2 'l9dcp2)} ,

W = ±-p-
sinhp' e
2(4)-4>0) - sinhp
- 2 R(p) ,
R(p) = V2 pcothp - w2 - 1. (31 )

This solution can be uplifted 51 to D=lO, the string frame metric and the three-form
in D=10 then being given by

ds 2 = -dt 2 + dyi + dy~ + dy~ + dp2 + R2(p) dQ~ + eae a,

H =
1 3
a a
- - e- 4 4>(F /\ e + E b e a /\ e /\ e
b C
(32)
2.J2 a c ,
)

where Fa is the gauge field 2-form and a = Aa - a with a being the invariant e e e
forms on 53. This solution describes a 3-brane in D=10.
This solution was originally obtained So ,51 by Chamseddine and Volkov, both in
D=4 and D=lO. A very interesting holographic interpretation for this solution was
then proposed 52 by Maldacena and Nunez, according to which this solution describes
the NS-NS 5-brane wrapped on 52, in which case it effectively becomes a 3-brane.
As a result, the solution provides a dual SUGRA description for the wrapped brane
worldvolume theory, which is the (deformed) N=l Super Yang-Mills in D=4. Since
this theory is confining, one can say that the solution (31),(32) provides the dual
SUGRA description for the phenomenon of confinement. For example, the value of
the Wilson loop, the beta-function and other parameters of the confining gauge field
theory can be obtained by simply computing some purely geometrical parameters of
the solution, like areas of spheres and the values of the H-fluxes through them. As a
result, this solution with self-gravitating Yang-Mills field has found quite interesting
and serious applications. Generally known as solution of Maldacena and Nunez (the
names of Chamseddine and Volkov seem now to be completely forgotten) it plays
an important role in the analysis of string theory (see 53 ,S4 for recent reviews).
1394

Non Abelian vacua have also been studied 55 in the context of a gauged SUGRA
in D=5. The gravitational and gauge fields in this case are given by the same SO(4)
invariant expressions (5) as for the YM particle, and there are also the dilaton and
axion. The solutions are not asymptotically flat, they have two supercharges, and
they can be used 56 for a dual SUGRA description of the confining N=l Super-Yang-
Mills theory in D=3.
The discussed above super symmetric solutions 50 ,55 have also been general-
ized 57 ,58 to the case where the spatial part of the metric is JR.l x 2:, where 2: is
a maximal symmetry space, that is sphere, hyperboloid or Euclidean space. Their
black hole generalizations 59 ,6o have been used for a holographic description of the
confinement/deconfinement phase transition in the dual gauge theory.

References
1. A. Lichenrowitcz, Theories Relativistes de la Gravitation et de l'Electromagnetisme
(Masson, Paris, 1955).
2. S. Deser, Absence of static solutions in source-free Yang-Mills theory. Phys.Lett. B
64, 463 (1976).
3. S. Coleman, There are no classical glueballs. Comm. Math. Phys. 55, 113 (1977).
4. S. Coleman and L. Smarr, Are there geon analogs in sourceless gauge field theories?
Comm. Math. Phys. 56, 1 (1977).
5. R. Bartnik and J. McKinnon, Particlelike solutions of the Einstein- Yang-Mills equa-
tions. Phys. Rev. Lett. 61, 141 (1988).
6. M.S. Volkov and D.V. Gal'tsov, Non-Abelian Einstein- Yang-Mills black holes. JETP
Lett. 50, 346 (1989).
7. H. Kunzle and A.K.M. Masood ul Alam, Spherically symmetric static SU(2) Einstein-
Yang-Mills fields. J.Math.Phys. 31, 928 (1990).
8. P. Bizon, Colored black holes. Phys.Rev.Lett. 64, 2844 (1990).
9. R. Ruffini and J.A. Wheeler, Introducing the black hole. Phys. Today 24, 30 (1971).
10. B. Kleihaus and J. Kunz, Static axially symmetric solutions of Einstein Yang-Mills
dilaton theory. Phys.Rev.Lett. 78,2527 (1997); [arXiv: hep-th/9612101j.
11. B. Kleihaus and J. Kunz, Static black hole solutions with axial symmetry.
Phys.Rev.Lett. 79, 1595 (1997); [arXiv: gr-qc/9704060j.
12. B. Kleihaus and J. Kunz, Rotating hairy black holes. Phys. Rev. Lett. 86, 3704 (2001);
[arXiv: gr-qc/0012081j.
13. M.S. Volkov and D.V. Gal'tsov, Gravitating non-Abelian solitons and black holes with
Yang-Mills fields. Phys.Rep. 319, 1 (1999); [arXiv: hep-th/9810070j.
14. R. Emparan and H.S. Reall, Black rings. Class. Quant. Grav. 23, R169 (2006); [arXiv:
hep-th/0608012j.
15. T. Eguchi, P.B. Gilkey and A.J. Hanson, Gravitation, gauge theories and differential
geometry. Phys.Rep. 66, 213 (1980).
16. A.A. Belavin, A. M. Polyakov, A.S. Shvarts, Yu.S. Tyupkin, Pseudoparticle solutions
of the Yang-Mills equations. Phys.Lett. B59, 85 (1975).
17. KB. Bogomol'nyi, Stability of classical solutions. Sov.J.Nucl.Phys. 24, 676 (1976).
18. M.S. Volkov, Gravitating Yang-Mills vortices in 4+1 spacetime dimensions. Phys. Lett.
B524, 369 (2002); [arXiv: hep-th/0103038j.
19. Y. Brihaye, A.Chakrabarti and D.H. Tchrakian, Particle-like solutions to higher order
curvature Einstein- Yang-Mills systems in d-dimensions. Class. Quant. Grav. 20, 2765
(2003); [arXiv: hep-th/0202141j.
 1395

20. N. Okuyama and Kei-ichi-Maeda, Five-dimensional black hole and particle so-
lution with non-Abelian gauge field. Phys. Rev. D 67, 104012 (2003); [arXiv:
hep-th/0212022j.
21. Y. Brihaye, A.Chakrabarti, B. Hartmann and D.H. Tchrakian, Higher order curva-
ture generalizations of Bartnick-McKinnon and colored black hole solutions in D = 5.
Phys.Lett. B 561, 161 (2003); [arXiv: hep-th/02122SSJ.
22. E. Radu and D.H. Tchrakian, No hair conjecture, non-Abelian hierarchies and Anti-de
Sitter spacetime. Phys.Rev. D 73,024006 (2006); [arXiv: gr-qc/050S033j.
23. E. Radu, C. Stelea and D.H. Tchrakian, Features of gravity- Yang-Mills hierarchies in
d-dimensions. Phys.Rev. D 73, 084015 (2006); [arXiv: gr-qc/060109Sj.
24. Y. Brihaye, E. Radu and D.H. Tchrakian, Einstein- Yang-Mills solutions in higher
dimensional de Sitter spacetime. CarXi v: gr-qc/ 061 00S7J.
25. P. Breitenlihner, D. Maison and D.H. Tchrakian, Regular solutions to higher order
curvature Einstein- Yang-Mills systems in higher dimensions. Class. Quant. Grav. 22,
5201 (2005); [arXi v: gr-qc/050S027J.
26. E. Radu, Ya. Shnir and D.H. Tchrakian, Particle-like solutions to the Yang-Mills-
dilaton system in d=4+1 dimensions. [arXiv: hep-th/0611270J.
27. G.W. Gibbons and P. Townsend, Self-gravitating Yang monopoles in all dimensions.
Class. Quant. Grav. 23, 4873 (2006); [arXiv: hep-th/0604024J.
28. D. Kastor, J.B. Traschen, Horizon inside classicall'umps. Phys.Rev. D 46, 5399 (1992);
[arXiv: hep-th/9207070J.
29. B. Hartmann, Non-Abelian black strings. Phys.Lett. B 602, 231 (2004); [arXiv:
hep-th/0409006J.
30. F. Bakkalo Taheri and y, Brihaye, Non-Abelian black strings and cosmological con-
stant. [arXiv: gr-qc/0605040j.
31. Y. Brihaye and T. Delsate, Black strings and solitons in five dimensional space-time
with positive cosmological constant. [arXiv: hep-th/0611195j.
32. Y. Brihaye and E. Radu, Gravitating Yang-Mills dyon vortices in 4+1 spacetime di-
mensions. Phys.Lett. B 605, 190 (2005); [arXiv: hep-th/0409065J.
33. Y. Brihaye and E. Radu, Kaluza-Klein black holes with squashed horizons and d=4
superposed monopoles. Phys.Lett. B 641, 212 (2006); [arXiv: hep-th/060622SJ .
. 34, Y. Brihaye, B. Hartmann, Dilatonic monopoles from (4 +1)-dimensional vortices.
Phys, Lett. B 534, 137 (2002); [arXiv: hep-th/0202054j.
35. Y. Brihaye, B. Hartmann and E. Radu, Deformed vortices in (4+1)-dimensional
Einstein - Yang - Mills theory, Phys,Rev, D 71, 085002 (2005). [arXiv:
hep-th/0502131j.
36. Y. Brihaye, B, Hartmann and E. Radu, Black strings in {4 +1)-dimensional Einstein-
Yang-Mills theory. Phys.Rev. D 72, 104008 (2005); [arXiv: hep-th/050S02SJ.
37. E. Roessl and M, Shaposhnikov, Localizing gravity on a 't Hooft-Polyakov monopole
in seven-dimensions, Phys.Rev. D 66,084008 (2002); [arXiv: hep-th/0205320j.
38, T. Gherghetta, E. Roessl and M. Shaposhnikov, Living inside a hedgehog: Higher
dimensional solutions that localize gravity, Phys.Lett. B 491, 353 (2000); [arXi v:
hep-th/0006251 J.
39. E. R. Bezerra de Mello and B. Hartmann, Localizing gravity in composite monopole
brane worlds without bulk cosmological constant. Phys, Lett. B 639, 546 (2006);
[arXi v: hep-th/0603077J.
40. Y. Brihaye, F. Clement, B, Hartmann, Spherically symmetric Yang-Mills solu-
tions in a {4+n)-dimensional space-time, Phys.Rev. D 70, 084003 (2004); [arXiv:
hep-th/0403041j.
41. Y. Brihaye, B. Hartmann, Spherically symmetric solutions of a (4+n) - dimensional
1396

Einstein- Yang-Mills model with cosmological constant. Class. Quant. Grav. 22, 183
(2005); [arXi v: hep-th/0403296].
42. B. Hartmann, Y. Brihaye, B. Bertrand, Spherically symmetric Yang-Mills solutions in
a five-dimensional anti-de Sitter space-time. Phys.Lett. B 570, 137 (2003); [arXiv:
hep-th/030706S].
43. A. Strominger, Heterotic solitons. Nucl.Phys. B 343, 167 (1990).
44. M.J. Duff, R.R. Khuri and J.X. Lu, String solitons. Phys.Rep. 259, 213 (1995);
[arXiv: hep-th/9412184].
45. B. de Wit and H. Nicolai, N=8 Supergravity. Nucl.Phys. B 208, 323 (1982).
46. M.J. Duff and J.T. Liu, Anti-de Sitter black holes in gauged N = 8 supergravity.
Nucl.Phys. B 554, 237 (1999); [arXiv: hep-th/9901149].
47. T. Hertog and K. Maeda, Black holes with scalar hair and asymptotics in N = 8
supergravity. JHEP 0407: 051 (2004); [arXiv: hep-th/0404261].
48. A. Chamseddine and M.S. Volkov, Non-Abelian solitons in N=4 gauged super-
gravity and leading order string theory. Phys.Rev. D 70, 086007 (2004); [arXiv:
hep-th/ 04041 71].
49. R.B. Mann, E. Radu and D.H. Tchrakian, Non-Abelian solutions in AdS(4) and d=ll
supergravity. Phys.Rev. D 74,064015 (2006); [arXiv: hep-th/0606004].
50. A. Chamseddine and M.S. Volkov, Regular non-Abelian vacua in N = 4, SO(4) gauged
supergravity. Phys. Rev. Lett. 79,3343 (1997); [arXiv: hep-th/9711181].
51. A. Chamseddine and M.S. Volkov, Non-Abelian BPS monopoles in N=4 gauged su-
pergravity. Phys.Rev. D 57,6242 (1998); [arXiv: hep-th/9707176].
52. J.M. Maldacena and C. Nunez, Towards the large N limit of pure N=l super- Yang-
Mills. Phys.Rev.Lett. 86, 588 (2001); [arXiv: hep-th/0008001].
53. M. Grana, Flux compactifications in string theory: A Comprehensive review. Phys.Rep.
423,91 (2006); [arXiv: hep-th/OS09003].
54. J.D. Edelstein, R. Portugues, Gauge/string duality in confining theories. Fortsch.Phys.
54,525 (2006); [arXiv: hep-th/0602021].
55. A. Chamseddine and M.S. Volkov, Non-Abelian vacua in D = 5, N=4 gauged super-
gravity. JHEP 0104,023 (2001); [arXiv: hep-th/0101202].
56. J.M. Maldacena and H. Nastase, The Supergravity dual of a theory with dynamical
supersymmetry breaking. JHEP 0109,024 (2001); [arXiv: hep-th/010S049].
57. E. Radu, New non-Abelian solutions in D = 4, N=4 gauged supergravity. Phys.Lett.
B 542, 275 (2002); [arXi v: gr-qc/0202103].
58. E. Radu, Non-Abelian solutions in N=4, D=5 gauged supergravity. Class. Quant. Grav.
23,4369 (2006); [arXiv: hep-th/060113S].
59. S.S. Gubser, A.A. Tseytlin and M.S. Volkov, Non-Abelian 4-d black holes,
wrapped five-branes, and their dual descriptions. JHEP 0109, 017 (2001); [arXiv:
hep-th/010820S].
60. G. Bertoldi, 5-D black holes, wrapped fivebranes and 3-D Chern-Simons super Yang-
Mills. JHEP 0210,042 (2002); [arXiv: hep-th/0210048].
 DERIVATION OF THE DIPOLE BLACK RING SOLUTIONS

STOYTCHO S. YAZADJIEV
Department of Theoretical Physics,
Sofia University, 5 James Bourchier Blvd, 1164 Sofia, Bulgaria
[email protected]

We present a solution generating method in 5D Einstein-Maxwell gravity. As an illus-

tration of the method we derive explicitly the solutions describing rotating dipole black
rings.

Keywords: Higher dimensional gravity; black rings.

We consider the 5D Einstein-Maxwell (EM) equations

R ILV -- "21 (FIL.\Fv.\ - (jFa.\F

1 17.\)
gILv n
,v ILF
ILV_
- 0, (1)

where R ILv is the Ricci tensor with respect to the spacetime metric gIL v and Ffiv
is the Maxwell tensor. We will consider spacetiems with three commuting Killing
vectors: one timelike Killing vector T and two spacelike Killing vectors KI and K 2 .
We also require that the Killing vector K2 is hypersurface orthogonal. Then, in
adapted coordinates T = 8/at, KI = 8/8X, K2 = 8j8Y the 5D metric can be
written in the form
ds 2 = gyydy 2 + goo (dt + AdX)2 + gxxdX2 + gpp(dp2 + dz 2 ), (2)
where all metric functions depend only on the canonical coordinates p and z.
We take the electromagnetic field in the form
F = dAy /\ dY, Ay = Ay(p, z). (3)
Let us note that this form of the electromagnetic field is compatible with the space-
time symmetries.
The following method for generating 5D EM solutions was developed: l
Proposition. Let us consider two solutions of the vacuum 5D Einstein equa-
tions
ds 2E(i) -- gE(i)dy 2 + gE(i) (dt + A(i)dX)2 + g-E(i)dX 2 + gE(i) (dp2 + dz 2 )
yy 00 E xx pp .
Then the following give a solution to the 5D EM equations

ds 2 = [lgE(I)IVgE(I)]
00 yy
2 dy2 + [ m
E(l)
Igoo IV
rE(l)
gyy'
1[gE(2) (dt + A(2)dX)
00 E
2

+gxx
-E(2)dX 2 + (I E(I)1 E(l) E(I))3
goo gyy gpp E(2) (dp2 + dz2)
1
2n(1)+2n(2) gPP ,
e E 3 E
1
Ay = ±2.j3f1 ) + canst,
1397
1398

where fJi) is a solution to the system

E(1»)2 E(l)
[} f(l) - _~ (goo gyy [) A(l)
PE- 2 p ZE'

E(1»)2 E(l)
!) f(l) - ~ (goo gyy [) A(l)
Uz E - 2 p P E'

and O~) satisfy

The presented proposition gives us a tool to generate new 5D EM solutions in a

simple way from known solutions to the vacuum 5D Einstein equations.
Through the use of the proposition we can generate the "5D EM images" of
all known solutions of the vacuum 5D Einstein equations with the symmetries we
consider here. We shall demonstrate the application of the proposition on the case
of rotating neutral black rings2 generating in this way the EM rotating dipole black
ring solutions.
The metric of the neutral black ring is given b y 2,3

where

F(x) = 1 + AX, G(x) = (1 - x 2 )(1 + vx), C(v, A) = VIA(A - l+A

v) 1 _ A· (4)

The coordinates X, y and the parameters v, A satisfy

-1::; X ::; 1, -00 < y ::; -1, 0 < v ::; A < 1. (5)

The dipole black ring solutions in 5D Einstein-Maxwell gravity can be generated

through the use of the proposition via the scheme

{neutral black ring} + {neutral black ring} -t {EM dipole black ring}. (6)

The first solution is with parameters {AI, v, R} while the second is parameterized
by {A2, v, R}. As a result we find
 1399

In order to exclude pathological behavior of the metric we must consider only neg-
ative AI, i.e. Al = -jL, 0 ~ jL < 1 and positive A2 and v satisfying 0 < v ~ A2 < 1.
One can easily see that the generated 5D EM solution is just the EM rotating dipole
black ring solution. 3 Let us also recall that in order to avoid the possible conical
singularities at x = ±1 and y = -1 we must impose

~(~)3 =(~)2 (8)

1 + A2 1- jL 1+ v
It would be interesting to find the solutions which can be generated through the
use of the proposition via the schemes

{neutral black ring} + {neutral black hole} ---+ {?}, (9)

{neutral black hole} + {neutral black hole} ---+ {?}, (10)

{appropriate solution} + {neutral black hole/ring} ---+ {black solution ?}. (11)
I am grateful to the organizers of the Marcel Grossmann Meeting for the financial
support. My thanks go especially to Prof. J. Kunz for her kind help.
This work was also partially supported by the Bulgarian National Science Fund
under Grant MUF04/05 (MU 408) and the Sofia University Research Fund under
Grant No.50.

References
1. s. S. Yazadjiev, Phys. Rev. D13, 104007 (2006).
2. R. Emparan and H. Reall, Phys. Rev. Lett. 88, 101101 (2002).
3. R. Emparan, JHEP 0403, 064 (2004).
 CHARGED ROTATING BLACK HOLES IN HIGHER
DIMENSIONS*

JUTTA KUNZ, FRANCISCO NAVARRO-LERIDA, JAN VIEBAHN

Institut fur Physik, Universitiit Oldenburg, Postfach 2503, D-26111 Oldenburg, Germany
[email protected], [email protected],
[email protected]

DIETER MAISON
Max-Planck-Institut fur Physik, Fahringer Ring 6, D-80805 Munchen, Germany
[email protected]

We consider charged rotating black holes in higher odd dimensions in theories such as
Einstein-Maxwell, Einstein-Maxwell-dilaton, and Einstein-Maxwell-Chern-Simons. Ret-
ricted to stationary axisymmetric black holes with equal-magnitude angular momenta
and a horizon of spherical topology, we present an Ansatz for the metric and the mat-
ter fields where the angular dependence is explicitly given and all the unknowns are
functions of the radial coordinate only. As a consequence, the field equations reduce to
ordinary differential equations. These black holes resemble, in general, their uncharged
counterparts, the Myers-Perry black holes. However, when the Chern-Simons term is
present new surprising effects appear, such as counterrotation, negative horizon mass, or
violation of uniqueness.

Higher dimensional black holes received much interest in recent years, in par-
ticular in the context of string theory, and with the advent of brane-world theories,
raising the possibility of direct observation in future high energy colliders. 1
Static charged asymptotically fiat black hole solutions of Einstein-Maxwell (EM)
theory exist for all spacetime dimensions D ~ 4. 2 ,3 The generalization of the Kerr
metric to higher dimensions was obtained by Myers and Perry,3 while the higher
dimensional generalization of the Kerr-Newman metric is still not known analyti-
cally.
In contrast to pure EM theory, exact solutions of higher dimensional charged ro-
tating black holes are known in theories with more symmetries or fields. For instance,
if a dilaton field is included, with the Kaluza-Klein value for the coupling constant,
one may produce black hole solutions to the corresponding Einstein-Maxwell-dilaton
(EMD) field equations by using the Myers and Perry solutions as seeds. 4
Another example would be the presence of a Chern-Simons (CS) term. Surpris-
ingly, for D = 5 the addition of that term allows to solve the Einstein-Maxwell-
Chern-Simons (EMCS) field equations exactly in the special case when the CS cou-
pling constant is set to the supergravity value,5 and analytic solutions describing
charged, rotating black holes are known. 6 - 8
For arbitrary values of the coupling constants of these theories (EM, EMD,
EMCS) no analytic solution is known. The numerical approach is the only possibility
then to study these black hole configurations. 9 ,lo

* This research has been partially supported by MEC under grant EX2005-007S.

1400
 1401

\Ve here concentrate on charged stationary axisymmetric black holes with equal-
magnitude angular momenta and a horizon of spherical topology in odd dimensional
EM, EMD, and EMCS theories. Under these assumptions it is possible to give
a general Ansatz for the metric and the matter fields, involving only unknown
functions of the radial coordinate.
Denoting the dimension of the spacetime by D( = 2N + 1), the metric for these
equal-magnitude angular momenta black holes reads

where eo == 0, ei E [0,7f/2j for i = 1, ... ,N -1, eN == 7f/2, lPk E [0,27fj for

k = 1, ... , N, and ck = ±1 denotes the sense of rotation in the k-th orthogonal
plane of rotation. An adequate parametrization for the gauge potential is given by

(2)

The dilaton field ¢ is a function of r only, like all the other unknown functions
f, rn, n, w, ao, and a'P'
To obtain asymptotically fiat solutions, the unknowns have to satisfy fl,,== =
m,I,,== = nl,,=oc = 1 , wl r== = 0, aol,,== = a'PI,,== = 0, ¢Ir=oo = O. The horizon
is located at rH, and is characterized by the condition f(rH) = 0. 9 ,10 Requiring
the horizon to be regular, the following boundary conditions must hold flr=nr =
rnlr=rH = nlr=r'H = 0, wlr=r'H = rHrl, <PH = (0,0 + rla'P)lr=T'H' da'P/drlr='f'H =
0, d¢/drlr=rll = 0, where rl is (related to) the horizon angular velocity and <PH is
the horizon electrostatic potential.
Subject to the above set of boundary conditions, the system of ordinary differ-
ential equations resulting from the substitution of the Ansatze in the field equations
posseSf:les regular black hole solutionf:l in El'vl, E.MD, and EMCS theories. In general,
those solutions share most of the properties of their uncharged counterparts, namely,
the Myers-Perry solutions.:3 However, in EMCS theory new surprising phenomena
appear for certain parameter ranges, to which we now turn.
To describe these new features we focus on D = 5 EMCS black holes, denoting
the CS coupling constant by A. For an appropriate normalization, A = 1 corresponds
to the supergravity value. Depending on the value of A, new types of black holes
appear, which can be classified by their total angular momentum J and horizon
angular velocity n. Black holes with A ::0: 1 may possess a static horizon, while their
1402

5D, IJ 11=IJ 21, 1.=3, rH=0.10, Q=-10

9 6

\
TypeI ~~

Typell Typel ~~
6 Typell! 4 Typell
Type IV Type Il! -
Type IV

-. :r:
3 :::s 2

0 \0/~ 0

-3 L-____ ~ ____ ~ ____ ~ ____ ~ -2 L-____ ~ ____- L_ _ _ _ ~ _ _ _ _~

-0.50 -0.25 0.00 0.25 0.50 -0.50 -0.25 0.00 0.25 0.50
Fig. 1a Q Fig, 1b Q

Fig. 1 Properties of 5D non-extremal A = 3 EMCS black holes with charge Q = -10 and
horizon radius rH = 0.1. a) Angular momentum J, b) horizon mass MH versus horizon
angular velocity n,

total angular momentum is non-vanishing (type II), When). > 1, counterrotating

configurations appear, for which the horizon angular velocity and the total angular
momentum have opposite signs, OJ < 0 (type III), When). 2: 2, black holes with
rotating horizon but vanishing total angular momentum appear (type IV), For ). 2: 2
thus four types of rotating black holes are present (Fig, 1a),
In addition, EMCS black holes with negative horizon mass (Fig. 1b) arise, their
total mass remaining always positive. Moreover, when). > 2 EMCS black holes are
not completely characterized by their global charges, giving rise to a violation of
uniqueness, and their gyromagnetic ratio may take any real value, including zero.
Further details of these intriguing new features may be found in Ref. [10].

References
1. S. Dimopoulos and G. Landsberg, Phys. Rev. Lett. 87, 161602 (2001); S. B. Giddings
and S. Thomas, Phys. Rev. D65, 056010 (2002).
2. F. R. Tangherlini, Nuovo Cimento 77, 636 (1963).
3. R. C. Myers, and M. J. Perry, Ann. Phys. (N. Y.) 172, 304 (1986) 304.
4. J. Kunz, D. Maison, F. Navarro-Lerida, and J. Viebahn, Phys. Lett. B639, 95 (2006).
5. G. T. Horowitz, gr-qc/0507080.
6. J. C. Breckenridge, D. A. Lowe, R. C. Myers, A. W. Peet, A. Strominger, C. Vafa,
Phys. Lett. B381, 423 (1996).
7. J. C. Breckenridge, R. C. Myers, A. W. Peet, C. Vafa, Phys. Lett. B391, 93 (1997).
8. M. Cvetic, Donam Youm, Nucl.Phys. B476, 118 (1996); M. Cvetic, H. Lii, C. N.
Pope, Phys. Lett. B598, 273 (2004); Phys. Rev. D70, 081502 (2004); R. Kallosh, A.
Rajaraman, W. K. Wong, Phys. Rev. D55, R3246 (1997).
9. J. Kunz, F. Navarro-Lerida, and A. K. Petersen, Phys. Lett. B614, 104 (2005); J.
Kunz, F. Navarro-Lerida, and J. Viebahn, Phys. Lett. B639, 362 (2006).
10. J. Kunz and F. Navarro-Lerida, Phys. Rev. Lett. 96, 081101 (2006); J. Kunz and F.
Navarro-Lerida, Phys. Lett. B643, 55 (2006); J. Kunz and F. Navarro-Lerida, Mod.
Phys. Lett. A21 2621 (2006).
 SOLITONIC GENERATION OF SOLUTIONS INCLUDING
FIVE-DIMENSIONAL BLACK RINGS AND BLACK HOLES

TAKAHSI MISHIMA * and HID EO IGUCHI

Laboratories of Physics, College of Science and Technology, Nihon University,
Narashinodai, Punabashi, Chiba 274-8501, Japan
E-mail: *[email protected]
[email protected]

We present a solitonic solution-generating method to construct single-rotational, axisym-

metric and stationary vacuum solutions in five-dimensional Einstein equations. Using this
method we give further some new five-dimensional solutions corresponding to 5 1 -rotating
black multi-ring.

Inspired by the remarkable discovery of a S1-rotating black ring solution by Em-

paran and Reall,l numerous investigations have been concentrated on the five-
dimensional black objects (strings, holes and rings). As further steps towards system-
atic construction of new solutions in five-dimensional case, we previously introduced
the new solitonic solution-generating method in five dimensions and succeeded in
obtaining the S2-rotating black ring solution,2,3 which is complementary to the
Sl-rotating black ring solution. Successively we also rederived Sl-rotating black
ring solution. 4 In this article, we give brief introduction of the solution-generating
method developed by us and present how to superpose black rings and a black hole.
Then we give the full metric of di-ring as an interesting example.
We concentrate on the spacetimes which satisfy the following conditions: (c1)
five dimensions, (c2) asymptotically flat spacetimes, (c3) the solutions of vacuum
Einstein equations, (c4) having three commuting Killing vectors including time
translational invariance and (c5) having a single nonzero angular momentum com-
ponent. Under the conditions (c1) - (c5), the following Weyl-Papapetrou metric
form can be adopted as a starting point, 5
ds 2 = _e2Uo (dx O _ wd¢)2 + e 2U1 p2 (d¢)2 + e2U2 (d'l/J)2 + e2h +U') (dp2 + dz 2) (1)
where Uo, U1 , U2 , wand 'Yare functions of p and z and also U 1 is set to -(Uo + U2 ).
After introducing new functions S := 2Uo + U2 and T := U2 , the metric form (1) is
rewritten into
ds 2 = _e- T [eS(dx O _ Wd¢)2 + e- s p2(d¢)2 + e2h-S) (dp2 + dz 2)] + e 2T (d1j;)2.
(2)
Then the five-dimensional Einstein equations are reduced to the following set of
equations,

1403
1404

where <P is defined through the equation (v) and the function Es is defined by
Es := eS + i <P. The equation (iii) is exactly the same as the well-known Ernst
equation, so that we call Es Ernst potential. The most crucial point to obtain new
metrics is to solve the equation (iii) because of it's non-linearity. For the actual
analysis in the following, we take some generalized Weyl solutions 5 as seeds and use
the formulas shown in the paper by Castejon-Amened and Manko. 6
The procedure to generate a new metric is the following: (a) choose an ap-
propriate generalized Weyl solution as a seed; (b) extract seed functions 5(0) =
2U6°) + UJO) and T(O) = UJO) from the seed metric; (c) determine the auxiliary
functions a and b by solving simple first-order linear differential equations 2 ; (d)
introduce further the following new functions

A := (x 2 - 1)(1 + ab)2 - (1 - y2)(b - aj2

B := [(x + 1) + (x - 1) ab]2 + [( 1 + y) a + (1 - y) W
{
G := (x 2 - 1)(1 + ab)[(l - y)b - (1 + y)a] + (1 - y2)(b - a)[x + 1- (x - l)ab] ,

where x and yare defined by (R,. + R-cr) /(2(J) and (R-cr - Rcr) / (2(J) with R±cr =
vip2 + (z =f (J)2, respectively; (e) give the corresponding Ernst potential

E = eS(O) x(l + ab) + iy(b - a) - (1 - ia)(l - ib)

S x(l+ab)+iy(b-a)+(l-ia)(l-ib)'

(f) integrate (ii) and (iv) to determine the function, (see the previous paper3 for
the explicit expression). Finally we can obtain the full metric with the following
formulas
S s(O) A
e = e - (3)
B'
_s(O) G
w = 2(Je A + G1 , (4)
e2i = G2(X 2 _ 1)-1 Ae2i', (5)

where,' have been already given in the step (f) and G 1 and G2 are fixed from some
physical conditions.
It is very helpful to use the viewpoint of the rod-structure analysis 5 ,7 to describe
what the method mentioned above does. By applying our method to a seed, the
finite space like rods in the interval -(J < Z < (J is lifted from the ¢-axis to the 'axis'
of time-translation (i.e. horizon) and also ¢-rotation is added to the timelike rods
which correspond to the static black rings and holes around the interval.
Hence we can generate the spacetime of a black multi-ring and a black hole with
the ¢-rotation if the generalized Weyl solution corresponding to the spacetime with
several static black rings (or a black hole) and also a local Kaluza-Klein bubble are
appropriately prepared as a seed. It is noticed that the K-K bubble should be put
on the interval -(J < Z < (J, as a necessary condition to obtain a regular solution.
 1405

In the rest we mention the black di-ring system as the simplest case of black
multi-ring and a hole system. The rod structure of the seed for the di-ring is given
in the left figure of Fig.I. The resultant rod structure is given as in the right figure
after applying the solitonic method.

Fig. 1. Schematic pictures of rod structure of black di-ring and its seed.

The seed functions of the black di-ring can be easily extracted from Fig.I,
(0) _ - - - - -
T - (U)..(J - U7)2(J + U7)1(J) - (U02 (J - U01cr ),
(0) _ - - - - -
S - (U)..(J - U7)2(J + U7)]fJ) + (US 2Cf - UO]U)'
where
Ud := ~ In [ J p2 + (z - d)2 + (z - d) ] and Ud := ~ In [ J p2 + (z - d)2 - (z - d) ].
The auxiliary functions a and b appeared in the procedure (c) are given systemati-
cally by the formulas in the previous work. 3 Then we obtain the corresponding full
metric following the procedure described above,
To extract the regular di-ring solutions, the further conditions must be intro-
duced. We comment the conditions briefly. The constants C 1 and C 2 of Eq.(4) and
(5) are fixed to eliminate the global rotation and adjust the periods of ¢ and '1jJ
to 27f at the infinity. To assure the regularity of the metric component g¢¢ and
to eliminate the closed timelike curves, some parameters a and (J of the auxiliary
functions a and b must be adjusted. Another condition comes from curing conical
singularities. Here we just say that some continuous set of parameters satisfies these
conditions simultaneously and can be confirmed to assure the global regularity of
the corresponding spacetime at least by numerical methods. The detailed analysis
and discussion are given in the recent work. s

References
1. R. Emparan and H. S. Reall, Phys. Rev. Lett. 88, 101101 (2002).
2. T. Mishima and H. Iguchi, Phys. Rev. D73, 044030 (2006).
3. H. Iguchi and T. Mishima, Phys. Rev. D74, 024029 (2006).
4. H. Iguchi and T. Mishima, Phys. Rev. D73, 121501(R) (2006).
5. R. Emparan and H. S. Reall, Phys. Rev. D65, 084025 (2002).
6. J. Castejon-Amenedo and V. S. Manko, Phys. Rev. D41, 2018 (1990).
7. T. Harmark, Phys. Rev. D70, 124002 (2004).
8. H. Iguchi and T. Mishima, hep-th/0701043.
 KALUZA-KLEIN BLACK HOLE WITH GRAVITATIONAL
CHARGE IN EINSTEIN-GAUSS-BONNET GRAVITY

HIDEKI MAEDA
Graduate School of Science and Engineering, Waseda University, Tokyo 169-8555, Japan,
Department of Physics, Rikkyo University, Tokyo 171-8501, Japan,
and
Department of Physics, International Christian University, Tokyo 181-8585, Japan
[email protected]

NARESH K. DADHlCH
Inter- University Centre for Astronomy €:f Astrophysics, Post Bag 4, Pune 411 007, India
[email protected]

We obtain a new exact black-hole solution in Einstein-Gauss-Bonnet gravity with a cos-

mological constant which bears a specific relation to the Gauss-Bonnet coupling constant.
The spacetime is a product of the usual4-dimensional manifold with a (n-4)-dimensional
space of constant negative curvature, i.e., its topology is locally Mn ~ M4 X Hn-4. The
solution has two parameters and asymptotically approximates to the field of a charged
black hole in anti-de Sitter spacetime. The most interesting and remarkable feature is
that the Gauss-Bonnet term acts like a Maxwell source for large r while at the other end
it regularizes the metric and weakens the central singularity. It is a pure gravitational
creation including Maxwell field in four-dimensional vacuum spacetime. This paper is
based on the results in Ref. 1.

1. Model and basic equation

We write action of Einstein-Gauss-Bonnet gravity with a cosmological constant for
n?: 5,

s= Jdnxycg[2~~ (R - 2A + aLCB)] ' (1)

where a is the Gauss-Bonnet (GB) coupling constant and all other symbols having
their usual meaning. The GB Lagrangian is given by

(2)
where the Greek indices run JL = 0,1, ... ,n - 1. This form of action follows from
low-energy limit of heterotic superstring theory.2 In that case, a is identified with
the inverse string tension and is positive definite, so we assume a ?: in this paper. °
It should be noted that LCB makes no contribution in the field equations for n :S 4.
The gravitational equation following from the action (1) is given by

(}J1. v == GJ1. v + aHJ1. v + A8J1. v = 0, (3)

where
1
G/1oV == R/1ov - 2g /1ovR, (4)

HJ1.v == 2 [RRJ1.v - 2R w",R cxv - 2R cx (3 R/1ocxv(3 + RJ1.cx (3, R Va(3,] - ~9J1.vLCB' (5)

1406
 1407

We consider the n-dimensional spacetime locally homeomorphic to M4 X Kn-4

with the metric, gf-tV = diag(gAB' r5'"Yab) , A, B = 0,··· ,3; a, b = 4,··· ,n - 1.
Here gAB is an arbitrary Lorentz metric on M 4 , ro is a constant and '"Yab is the
unit metric on the (n - 4)-dimensional space of constant curvature Kn-4 with its
curvature k = ±1, O. Then, we obtain the following no-go theorem on M4:

Theorem 1.1. If (i) r6 = -2ka(n-4)(n-5) and (ii) aA = -(n 2 -5n-2)/[8(n-

4)(n - 5)], then gA B = 0 for n 2: 6 and k and A being non-zero.

The conditions (i) and (ii) imply for a > 0, k = -1 and A < o. Hereafter we
set k = -1, i.e., the local topology of the extra dimensions is H n - 4 , and obtain the
vacuum solution satisfying the conditions (i) and (ii). The governing equation for
gAB is then a single scalar equation on M 4 , ga b = 0, which is given by
1 (4)R a (4)L 2n - 11
- -
n-4
+-2 GB+ a(n-4)2(n-5)
=0
'
(6)

where superscript (4) means the geometrical quantity on M4.

2. Kaluza-Klein black-hole solution

We seek a static solution with the metric on M4 ~ M2 X K2 reading as:

gABdxAdx
B
= - f(r)de + ftr) dr 2 + r2d~~(k)' (7)

where d~~(k) is the unit metric on K2 and k = ±1, o. Then, Eq. (6) yields the
general solution for the function fer):

r2 [
fer) = k + 2(n _ 4)a 1 =F

where JL and q are arbitrary dimensionless constants. The solution does not have the
general relativistic limit a ---> O. There are two branches of the solution indicated
by sign in front of the square root in Eq. (8), which we call the minus- and plus-
branches.
There exists a central curvature singularity at r = 0 as well as the branch
singularity at r = rb > 0 where the term inside the square root in Eq. (8) is
zero. This solution can represent a black hole depending on the parameters. The n-
dimensional black hole with (n - 4)-dimensional compact extra dimensions is called
the Kaluza-Klein black hole. The warp-factor of the submanifold r6 is proportional
to GB parameter a which is supposed to be very small. Thus, compactifying H n - 4
by appropriate identifications, we obtain the Kaluza-Klein black-hole spacetime
with small and compact extra dimensions.
The function fer) is expanded for ". ---> 00 as

al/2f.tV3(n - 4)(n - 5) aqV3(n - 4)(n - 5) ".2 ( ~-

4
() k
jrr:::J=F
r
± r
2 + 2(n - 4)a 1=f
3(n - 5) .

(9)
1408

This is the same as the Reissner-Nordstrom-anti-de Sitter (AdS) spacetime for k = 1

in spite of the absence of the Maxwell field. This suggests that ~ is the mass of the
central object and q is the charge-like new parameter. Further, the solution (8)
agrees with the solution in the Einstein-GB-Maxwell-A system having the topology
of l\1 n :::::; Jv1 2 X Kn-2 although it does not admit n = 4. 3 The new "gravitational
charge" q is generated by our choice of the topology of spacetime, splitting it into a
product of the usual 4-spacetime and a space of constant curvature. Thus, the solu-
tion (8) manifests gravitational creation of the Maxwell field, i.e., "matter without
rnatter" .
It is noted that 1(0) = k =t= .j=(j, which produces a solid angle deficit and it
represents a spacetime of global monopole. This means that at r = 0 curvatures
will diverge only as 1/r 2 and so would be density which on integration over volume
will go as r and would therefore vanish. This indicates that singularity is weak as
curvatures do not diverge strongly enough.

3. Summary
We have thus found a new Kaluza-Klein vacuum black hole solution (8) of Einstein-
Gauss-Bonnet gravity with topology of product of the usual 4-spacetime with a
negative constant curvature space. In this solution we have brought the GB effects
down on four dimensional black hole. This solution manifests gravitational creation
of the Maxwell field and asymptotically resembles a charged black hole in AdS back-
ground. What really happens is that GB term regularizes the metric and weakens
the singularity while the presence of extra dimensional hyperboloid space generates
the Kaluza-Klein modes giving rise to the Weyl charge. This class exact solutiom,
is further investigated in Ref. 4.

Acknow ledgrnents
This research has been partially supported by the Grant-in-Aid for Scientific Re-
search Fund of the Ministry of Education, Culture, Sports, Science and Technology,
Japan (Young Scientists (B) 18740162).

References
1. H. Maeda and N. Dadhich, Phys. Rev. D74, 021501(R) (2006).
2. D.J. Gross and J.H. Sloan, Nucl. Phys. B291, 41 (1987);
M.C. Bento and O. Bertolami, Phys. Lett. B368, 198 (1996).
3. R. -G. Cai, Phys. Rev. D65, 084014 (2002); D. L. Wiltshire, Phys. Lett. B169, 36
(1986); D. G. Boulware and S. Deser, Phys. Rev. Lett. 55, 2656 (1985); J. T. Wheeler,
Nucl. Phys. B268, 737 (1986).
4. H. Maeda and N. Dadhich, hep-th/0611188, accepted for publication in Phys. Rev. D.
 HIGHER DIMENSIONAL ROTATING CHARGED BLACK HOLES

ALIKRAM N. ALlEV*
Feza Giirsey Institute, P.K. 6 gengelkoy, 34684 Istanbul, Turkey

We discuss a new analytic solution to the Einstein-Maxwell field equations that describes
electrically charged black holes with a slow rotation and with a single angular momentum
in all higher dimensions. We also compute the gyromagnetic ratio of these black holes.

1. Introduction

Black holes remain to be one of the most intriguing and puzzling object of study in
higher dimensional spacetimes. It is widely believed that the remarkable features of
black holes in four dimensions, such as the equilibrium and uniqueness properties
as well as quantum properties of evaporation of mini-black holes might have played
a profound role in understanding the nature of fundamental theories in higher di-
mensions. Therefore, of particular interest is the study of black hole solutions in
higher-dimensional gravity theories as well as in string/M-theory.
The first higher dimensional solutions for static black holes with the spherical
topology of the horizon have been discussed by Tangherlini.l These solutions gen-
eralize the spherically symmetric Schwarzschild and Reissner-Nordstrom solutions
of four-dimensional general relativity. For the static black holes the uniqueness and
the stability properties still survive 2 in higher dimensions, however the situation is
different for rotating black holes. The exact solution for the rotating black holes was
found by Myers and Perry.3 The solution is not unique, unlike its four dimensional
counterpart, the Kerr solution. There exists a rotating black ring solution 4 in five
dimensions with the horizon topology of 52 x 51 which may have the same mass
and angular momentum as the Myers-Perry solution. However, the counterparts of
the Myers-Perry black holes in higher dimensional Einstein-Maxwell theory have
not been found yet. A numerical treatment of the problem for some special cases
in five dimensions was given in Ref. 5 Here, as a first step towards the desired exact
metric, we shall discuss the intermediate case of slow rotation and give a new ana-
lytical solution that describes electrically charged black holes with a single angular
momentum in all higher dimensions.

2. Metric ansatz
The strategy of obtaining the familiar Kerr-Newman solution in general relativity
is based on either using the metric ansatz in the Kerr-Schild form or applying the
method of complex coordinate transformation to a non-rotating charged black hole.
When employing in N + 1 dimensional spacetime both approaches lead us to the

* [email protected]

1409
1410

following metric ansatz

N 2 2
2 ( m q2) 2 2a(mr - - q 2)sin e
ds =- 1 - r N - 4 'E + r 2(N-3) 'E dt - r2(N-3) 'E dtd¢

(1)
where the parameters m , a and q are related to the mass, angular momentum and
electric charge of the black hole. The metric function
~ = r N - 2 (r 2 + a 2) - mr2 +l r 4- N (2)
and dO'Jy -3 is the metric on a unit (N - 3)-sphere.
It is straightforward to verify that the source-free Maxwell equations in the
background of the metric (1) admit the potential one-form field 6

A= ( i
N-2r N - 4 'E (dt - a sin e d¢ ) .
2
(3)

where Q is the electric charge of the black hole. For N = 3, inspecting the simulta-
neous system of the Einstein-Maxwell equations with this potential form and with
the metric (1) shows that it is satisfied for q2 = GQ2. This is the case of a Kerr-
Newman black hole in four dimensions. However, for N 2': 4, the consistent solution
to the system of equations is obtained only when we restrict ourselves to a slow
rotation.

3. Results
The metric for slowly rotating and charged black holes with a single angular mo-
mentum in all higher dimensions has the form

ds
2
= - ( m
1 - r N- 2 + r2(N-2)
q 2) dt
2
+
( m
1 - r N- 2 + r2(N-2)
q 2 )-1 dr
2

where the parameter q is given by

87rG ] 1/2
q=±Q [ (5)
(N - 2) (N - 1) A N - 1
and AN -1 is the area of a unit (N - 1)-sphere. The associated electromagnetic field
is described by the potential one-form

~ N-2
A=-( N-2r
2
(dt-asin ed¢) . (6)
 1411

The metric (4) generalizes the higher dimensional Schwarzschild-Tangherlini 1 so-

lution to include an arbitrarily small angular momentum of the black holes. For
N = 4 we have the metric in five dimensions. 7
It is clear that a rotating charged black hole must also have a magnetic dipole
moment. In our case, it is determined from the far distant behaviour of the magnetic
field in the metric (4). We find that the associated magnetic (N - 2) -form field has
the following orthonormal components 6

(7)

which give the value of the magnetic dipole moment JL = Qa. From the asymptotic
behavior of the metric (4) we find the specific angular momentum j = am of the
black hole. The latter allows us to rewrite the magnetic dipole moment in terms of
total mass and total angular momentum of the black hole as follows
Qj QJ
JL = - = (N -1) - . (8)
m 2M
This expression shows that a slowly rotating charged black hole in N + 1 dimensions
must have the gyromagnetic ratio
g=N-1. (9)
The value of the gyro magnetic ratio agrees with that obtained earlier 8 for the Myers-
Perry black hole with a test (small) electric charge in five dimensions. It also agrees
with the numerical analysis of paper. 9
Acknowledgements: The author thanks the Scientific and Technological Research
Council of Turkey (TUBiTAK) for partial support under the Research Project
105T437.

References
1. F. R. Tangherlini, Nuovo Cimento 77, 636 (1963).
2. G. W. Gibbons, D. Ida and T. Shiromizu, Phys. Rev. Lett. 89, 041101 (2002).
3. R. C. Myers and M. J. Perry, Ann. Phys. (N.Y.) 172,304 (1986).
4. R. Emparan and H. S. Reall, Phys. Rev. Lett. 88, 101101 (2002).
5. J. Kunz, F. Navarro-Lerida, and A. K. Petersen, Phys. Lett. B 614, 104 (2005).
6. A. N. Aliev, Phys. Rev. D 74,024011 (2006).
7. A. N. Aliev, Mod. Phys. Lett. A 21, 751 (2006).
8. A. N. Aliev and V. P. Frolov, Phys. Rev. D 69, 084022 (2004).
9. J. Kunz, F. Navarro-Lerida, and J. Viebahn Phys. Lett. B 639, 362 (2006).
 PERTURBATIVE STABILITY AND ABSORPTION CROSS
SECTION IN STRING CORRECTED BLACK HOLES

F. MOURA*
SQIG - IT, Instituto Superior Tecnico, Departamento de Matematica
Av. Rovisco Pais, 1049-001 Lisboa, Portugal
[email protected]

We consider a d-dimensional spherically symmetric dilatonic n2 string corrected black

hole solution. We study its stability under tensor type gravitational perturbations and
compute the absorption cross section for low frequency gravitational waves.

We are interested in studying the behavior of a string-corrected dilatonic black hole

solution under perturbations in d spacetime dimensions,l setting any tensorial or
fermionic fields to zero and taking as background metric

(1)

with dOL 2 = lij (B) dB i dBj, lij = 9ij / r2 being the metric of a (d - 2 )-sphere Sd- 2 .
We take, in the effective action, only the leading R2 a' correction: 2

_1_
2/"£2
J v
C"::g
-.'I d _ 2 (81"A-.)
[R - _4_ 'I'
A-. + e2~d¢~RI"I/PUR
81"'1' 2
] dd x
I"l/pU' (2)

with A = ~/, ~ , 0 for bosonic, heterotic and superstrings, respectively.

The corrected field equation for the graviton is, to this order,

R 1"1/ \ 2~d¢
+ /\e (R I"PUT R 1/pUT - 2(d 1_ 2) gl"1/ R PUAT RPUAT) -- 0 . (3)

Here we only take tensorial perturbations to the metric (by using the dilaton
field equation, we show that we can set 5¢ = 0), given by hl"1/ = 5gl"I/:3

hij = 2r 2 HT(r)T;j ( Bi) ,hir = hit = 0, hrr = htr = htt = O. (4)

Di is the Sd-2 covariant derivative; T;j are the eigentensors of the Sd-2 laplacian,
with eigenvalues -kT = 2 - £ (£ + d - 3), £ = 2,3,4 ... , satisfying

(5)
Using the explicit form of the Riemann tensor for the metric (1) and its varia-
tions, computed from (4), and perturbing (3), we determine the equation for HT,
which we write in the form of a master equation (with r * defined by dr * / dr = 1/1)
8 2 <1) 8 2 <1)
-8 2 - -8 2 =: VT<I). (6)
r* t
*Work supported by Funda<;iio para a Ciencia e a Tecnologia through Centro de L6gica e Com-
puta<;iio (CLC) and fellowship BPD/14064/2003, in collaboration with Ricardo Schiappa.

1412
 1413

As explained in ref.4, we derive our master function and potential:

<P
H ex
= ~
v7 p
(1 Lf + d-2
r
+ '±'(d
r3
- 4)>'(1 - f) - '±'>.fl - .2.>.fI2
2- ~>.f'
r2 rf dr )
'

1 )2( 4>. ) [d-4 ( 4 ) . )

VT[j(r)] = ( 1 _ 2>.f 1 + ~ (1 - f) 4r2 1 + ~ (1 - f) + 2>'1"-1]
2r2

+ 1 I[(kT + 2) L
+ 2(d _ 3/(1 - f) + d - 8 ff' _ ->'-f (f")2
1 - 2>' Lr r2 r2 2 r d- 2

+ 4>. (kT + 2) f(1 - f) + 2(d _ 3)>' f(1 - f)2 + 2(d _ 4)>' f(1 - f)fl]
r4 r4 r3

+ f f' + (d - 4) p. (7)
r r2
To study the stability of a solution, we use the "S-deformation approach". 3 After
having obtained the potential VT(f), we assume that its solutions are of the form
<P(r*, t) = eiwt¢(r*), such that 8<p/Ot = iw<p. The master equation is then written in
the Schrodinger form A<p = w 2 <p, and a solution to the field equation is then stable
if the operator A is positive definite with respect to the following inner product:

(¢, A¢) = i:= 4)(r*) [- ::; + V] ¢(r*) dr* = i:= (ID¢12 + ; 1¢1 2 ) dr* (8)

(see ref. 4 for the details), with D = d~* - f~T ir (:T) and

Qf = 1 l' 12 [(kT
1 - 2>'- r
+ 2) (1 + 4; (1 - f)) + (2d - 6)(1 - f) (1 +
r r
~ (1 - f))
r

- 2r l' - ->.- (f")2 r2] . (9)

d-2
All that is necessary to guarantee the stability is to check the positivity of 1.
In ref. 4 we considered the R 2 -corrected black hole solution of the type of (1)
studied in ref. 2, taking a coordinate system in which the horizon radius RH is not
changed. Assuming RH » J):, for this solution f(r) is given by

_ ( 1- (RH)d-3)
f(r)- [ 1->' (d-3)(d-4)R'~-5rd-l-R'k-ll
-----~
-
r 2 rd - 1 rd - 3 _ R d - 3 .
(10)
H

We showed 4 that 1
> 0; therefore this solution is stable under tensor perturbations.
In Einstein-Hilbert gravity, for any spherically symmetric black hole in arbitrary
dimension the absorption cross section of minimally coupled massless scalar fields
equals the area of the black hole horizon, 5 a result which suggests a universality of
the low-frequency absorption cross sections of generic black holes. Since the equation
describing gravitational perturbations to a black hole solution allows for a study of
scattering in this spacetime geometry, we tried to extend this result to the higher-
derivative corrected black hole (10), focusing only on the leading contribution to
1414

e
the scattering process: the s-wave, with = O. The low-frequency regime RHW <t: 1
allows us to fully analytically solve the problem by using the technique of matching
solutions near the event horizon to solutions at asymptotic infinity. In both these
regions the potential VT[J(r)] vanishes, and the master equation reduces to a free-
field equation whose solutions are plane waves in the tortoise coordinate.
Near the event horizon, r :::' R H , since we are computing the absorption cross
section, we shall consider the general solution for an incoming plane wave HT(r*) =
AneareiwT*; after expanding VT(r) and r*(r) this solution becomes
.RHW ( (d-l)(d-4),) (r-RH)) (11)
HT(r):::'Anear ( 1+z d _ 3 1+ 2 R'k log RH .

Close to infinity, one must consider a superposition of incoming and outgo-

ing waves which becomes, expressed in the original radial coordinate, HT(r)
(rw)(3-d)/2 [Aas J(d-3)/2 (rw) + Bas N(d-3)/2 (rw)] ; at low frequencies, rw <t: 1,
1 2 d;3 r (d-3)
HT(r) :::' Aas + Bas d~3 +0 (rw). (12)
(y)
d-3 _
2 -2 r 7f (rw )
In order to compute the absorption cross section, one needs the fluxes per unit area
J = :li (HJ,(r*)dd~~ -HT(r*)dd~~). Near the horizon this quantity is given by
J near = w IAnearl2; close to infinity we analogously have J as = ~r2-dw3-d IAasBasl.
In order to match the coefficients A near , Aas and BaS) one needs to interpolate be-
tween the solutions near the event horizon and at asymptotic infinity. This requires
solving the master equation in the intermediate region between the horizon and
infinity. The full computation can be found in ref. 4, where it is shown that

Aas = 2-r
d-3
2- - 1)
(d-2- A near ,

B _ i7f(R Hw)d-2 ( (d-1)(d-4)

(13)
as-- 2 Y(d_3)r(d;3) 1+ 2 ;'k) Anear ·
With these results one obtains the scattering cross section
£=0 Jrd-2JasdfJd_2 A ( (d-1)(d-4),)
O'T = = H 1+ -2 . (14)
~M 2 RH
We conclude that the absorption cross section is increased due to the 0/ correc-
tions, although it is still proportional to the area of the event horizon. The same
happens to the black hole entropy, although its 0/ correction has a different value. 2 ,4

References
1. H. Kodama, A. Ishibashi and O. Seto, Phys. Rev. D62 (2000) 064022.
2. C. G. Callan, R. C. Myers and M. J. Perry, Nucl. Phys. B311 (1989) 673.
3. A. Ishibashi and H. Kodama, Prog. Theor. Phys. 110 (2003) 901.
4. F. Moura and R. Schiappa, Class. Quant. Grav. 24 (2007) 36l.
5. S. R. Das, G. Gibbons and S. D. Mathur, Phys. Rev. Lett. 78 (1997) 417.
 ULTRARELATIVISTIC BOOSTS OF BLACK RINGS

MARCELLO ORTAGGIO
Dipartimento di Fisica, Universita degli Studi di Trento and INFN, Gruppo Collegato di Trento,
Via Sommarive 14, 38050 Povo (Trento), Italy
marcello.ortaggio AT comune. reo it

PAVEL KRTOUS and JIRI PODOLSKY

Institute of Theoretical Physics, Faculty of Mathematics and Physics,
Charles University in Prague, V HoleSovickach 2, 180 00 Prague 8, Czech Republic
Pavel. K [email protected] and Jiri. [email protected]

We summarize the main results of recent studies of Aichelburg-Sexl ultrarelativistic

limits of five dimensional vacuum and charged black rings.

1. Introduction

In 1959 Pirani argued that the geometry associated with a fast moving mass resem-
bles a "plane" gravitational wave. 1 Later on, Aichelburg and Sexl (AS)2 considered
a limiting ("ultrarelativistic") boost of the Schwarzschild line element to determine
the exact impulsive pp-wave 3 generated by a light like particle. In higher dimensions
D ~ 4, the AS limit of static black holes 4 has been known for some time 5

(i=2, ... ,D-1), (1)

r.; 37rV2 pb
where H= -8v2pMlnp- - - - (D = 4), (2)
2 p
2
PM PQ
H = CM pD-4 - CQ p2(D-3)-1 (D > 4), (3)

167rV2 (2D - 9)!! 2D - 5 7rV2

with CM = (D _ 4)DD-3 ' CQ = (D _ 3)! (D - 2)(D - 3) 2D - 4 '
p2 = ZiZi, J(u) is the Dirac delta, PM/PQ constants related to the mass/charge of
the original spacetime 4 and DD-3 the area of a unit (D - 3)-sphere. Such pp-waves
have been employed in studies of classical formation of black holes in high energy
collisions. 6 - 11 In the case of zero charge they can be straightforwardly generalized
to include an external magnetic field. 12 The AS boost of rotating black holes 13 has
been studied in. 14 Here we focus on D = 5 black rings (cU 5- 17 for more details).

2. Boost of black rings

Ultrarelativistic boosts of spacetimes rely on first identifying a notion of Lorentz
boost (e.g., with respect to asymptotic infinity). Then one applies such a transfor-
mation to the metric and takes the singular limit when the boost parameter tends
to the speed of light. Simultaneously, the mass is appropriately rescaled to zero. 2
The final metric depends on the boost direction. We shall use spatial "cartesian"

1415
1416

coordinates (Xl, X2, Yl, Y2) such that the 2-plane of the ring circle is (Yl, Y2), and
study boosts along the axes Xl and Yl, "orthogonal" and "parallel" to it.

Vacuum black ring In the orthogonal case, the ultrarelativistic boost of the vac-
uum black ring of18 results 15 ,16 in the following pp-wave propagating along Xl

(4)

4~L
in which k=
p= (~+L)2'

and e (L - ~) denotes the step function (cf. the appendix of15 for definitions of the
elliptic integrals K, E and II). The null coordinates u and v are defined by
-u+v
t= V2 ' (7)

and p).., Pv and L are constants related to the mass, angular momentum and radius
of the original ring. For black rings "in equilibrium" set p).. = 2pv in eq. (5).
In the case of a parallel boost the final pp-wave, now propagating along Y1, is

(8)

k' = (a 2 - 1'/2 - y'§ + L2)1/2 (a2 _ y,§)2

where
V2a 4a 2 y'§
a = [(rl + yg - L2)2 + 41]2 L2] 1/4, 7/ = yI'-xi-+-x-§, (10)
and the null coordinates are now defined by
-u+v
t= V2 ' (11)
 1417

Static charged black ring Static charged black rings were found in 19 (up to a
misprint in F)1v) in the Einstein-Maxwell theory, see also. 2o - 22 After an orthogonal
boost of such solutions, one obtains 17 a pp-wave (4) with

Vip),. [ (3L
21 + e2 + e + x 2 c' + L) K(k)
1-e 2 2(_L J(E,+L)2+x~

- . I(e, + L)2 + x~ E(k) - ; + LL x~

jE, - L)2 + 2 I1(po, k) + ~21X211 ,(12)
V ." - (( + L)2 + x 2
k and I; as in eq. (6) and Po = -(I; - L)2 /x~. The parameter e is related 17 to the
electric charge of the original static spacetime.
For a parallel boost, one finds a metric (8) with

HC =
II
Vip),. [(2L21+2e22
1- e
+a2+a?L:+7)~)~K(k')-2aE(k')
a - Y2 a
7",2 + L2 a 2 + y2
II(p', k') + ~21Y21 ,
]
- 2 2 ; (13)
a a - Y2
with k', pi, a and T/ as in eq. (10).
See 16 for the AS limit of the supersymmetric black ring of. 23

References
1. F. A. E. Pirani, Pmc. R. Soc. A 252, 96 (1959).
2. P. C. Aichelburg and R. U. Sexl, Gen. Rei. Gmv. 2, 303 (1971).
3. R. Penrose, Int. J. Theor. Phys. 1, 61 (1968).
4. F. R. Tangherlini, Nuovo Cimento 27, 636 (1963).
5. C. O. Loust6 and N. Sanchez, Int. J. Mod. Phys. A 5, 915 (1990).
6. D. M. Eardley and S. B. Giddings, Phys. Rev. D 66, 044011 (2002).
7. E. Kohlprath and G. Veneziano, JHEP 06, 057 (2002).
8. H. Yoshino and Y. Nambu, Phys. Rev. D 66, 065004 (2002).
9. S. B. Giddings and V. S. Rychkov, Phys. Rev. D 70, 104026 (2004).
10. H. Yoshino and V. S. Rychkov, Phys. Rev. D 71, 104028 (2005).
11. H. Yoshino and R. B. Mann, Phys. Rev. D 74,044003 (2006).
12. M. Ortaggio, JHEP 05, 048 (2005).
13. R. C. Myers and M. J. Perry, Ann. Phys. (N. Y) 172, 304 (1986).
14. H. Yoshino, Phys. Rev. D 71,044032 (2005).
15. M. Ortaggio, P. Krtous and J. Podolsky, Phys. Rev. D 71, 124031 (2005).
16. M. Ortaggio, J. Podolsky and P. Krtous, JHEP 12,001 (2005).
17. M. Ortaggio, J. Phys. Conf. Ser. 33, 386 (2006), gr-qc/0601093.
18. R. Emparan and H. S. Reali, Phys. Rev. Lett. 88, 101101 (2002).
19. D. Ida and Y. Uchida, Phys. Rev. D 68, 104014 (2003).
20. H. K. Kunduri and J. Lucietti, Phys. Lett. B 609, 143 (2005).
21. S. S. Yazadjiev (2005), hep-th/0507097.
22. M. Ortaggio and V. Pravda, JHEP 12, 054 (2006).
23. H. Elvang, R. Emparan, D. Mateos and H. S. Reali, Phys. Rev. Lett. 93, 211302
(2004).
 THE RESULTS OF A NEW SOLUTION OF THE EINSTEIN FIELD
EQUATIONS IN GENERAL RELATIVITY AND BLACK HOLE NEW
MOVEMENT

SADEGH, MOUSAVI
Department 0/ Physics, Amirkabir University o/Technology
No 13, Unit 3, North Shahin St., West Hemmat Highway, Tehran, Iran
Postal Code: 14758-15835, Email: [email protected]

1. Introduction
We have found a new solution for Einstein field equation in general relativity, which is
more complete than previous. solutions and contains the previous solutions as its special
forms. There are different solutions for Einstein field equation, but of all the exact
solutions which are known, a limited class seems to have a real physical meaning.
In the following, first the new metric and its calculated functions are presented briefly
then, based on that, some physical results are extracted and discussed.

2. Complete form of the new solution (Sadegh metric)

The new solution, which is obtained for Einstein field equation is:
2 2 2
di =[P2~]dt2 _(AL )~(r2 +b ,ut2)(asinO+b)]dIp_(a(2mr- Q2 »dt}2 _(AL? )~(r2 +a ,u2)(a+b)]dO_
L2 p2 ;(a+b) L2 p2 ;'(asinqJ+b)
Q2
(b(2mr-
L2
»dt}2 -cL)dr
~
-Vdudt

Where: L2=(r2+a2+b2)2_(a2sin28+b2sin2cp)A, A=1-,u2, ,u=cos8

;2=[(r2+b 2)2_b 2A'(r2+b 2 +Q 2 _2mr)]A' A'=1-J./ 2 , J1'=coscp

p2 = r2 + a 2cos 28 + b 2cos 2cp ;'2 = (r2 + a 2)2 _a 2A(r 2 +a 2 + Q2 - 2mr) (1)

A = r2 + a 2 + b 2 + Q2 - 2mr
Where m, a, and b are real constants. If j(J and jrp are angular momentums
around 8 andq> , respectively, then a and b are defined as the related momentums on unit
mass: b = j e / m, a = j cp / m where m is the Schwarzschild mass. V is the velocity
function in u direction. The coordinate system is chosen so that: x= [t, r, 8,cp ,u].

3. Physical analysis of the new metric

By analyzing the new metric, it is seen that, in addition to the spinning movement of black
holes around cp (Kerr metric[ 1]), they have another spinning movement around () and
moreover they have a translational movement in u direction, all at the same time.

1418
 1419

4. Two interesting theories

For better physical imagination of Sadegh metric (which shows the actual movement of
black holes), and for study of the physical movement route, we choose a voluminal
element of a black hole and study just its movement. Based on Sadegh metric, the
movement route of this element is a helix which is similar to the movement of an electric
charged particle in a uniform and constant magnetic field. From Electrodynamics [2], we
know, if a charged particle q with a relativistic speed venters into a uniform and constant
magnetic field B, this charged particle will move on a helical path, which the average
radius of the helix is: r = m}tv/ qB (2)
In equation 2, with starting to increase the magnitude of B, the radiuses of helix, r
start to decrease. If this increment continues, finally we will reach a limit that in this limit,
the radius of the helix is equal to the radius of the particle. Therefore, in this limit, the
charged particle starts spinning around itself and at the same time, it moves on a
translational path, in a special condition. So by solving the Einstein field equation in the
form of Sadegh metric, we have found the existence of a uniform and constant magnetic
field in all over the universe. Subsequently, two new and interesting theories are
extracted:
1. There is a uniform and constant magnetic field in whole universe, which according to
the calculations, its magnitude is: B = 5.25*10 8 (gauss).This magnetic field is the
reason of all spinning movements or all spin-orbital movements of charged objects in
the nature; e.g. the movement of an electron around itself and around proton in atom,
the movement of the moon around itself and around the earth, and the movement of
the earth or other planets, black holes, and etc.
2. If a charged object does a free spinning movement in the nature, it should certainly
have a translational movement at the same time, unless there is an opposite force that
resists against this translational motion that in this case, the opposite force needs to
be determined.

4.1. Some experimental evidences about the new theories

To examine the truth of these two theories numerically in the nature, we choose two
different scales: microscopic scale, e.g. electron, proton, and etc., and astronomical scale,
e.g. the earth, the moon, the sun and etc.
In microscopic scale, we want to calculate the magnitude of electric charge in
electron and mass of proton by using their linear spinning speed (v) and radius. First, the
magnitude of electric charge in an electron is calculated. If this theory is right, it will be
equal to the amount which is found from experiments. For an electron we have [3]:
m =9.10939*10-31 (kg), r = 2.14*10- 18 (m), and v =5.94*10 8 (m/s) where v is the spinning
speed. After substituting these amounts into Eq.(2), we obtain e=J.605 *10- 19(coulombs).
As seen this amount is equal to the actual amount which is obtained from experiments: e =
1.6 *10- 19 (c). In second example, the mass of proton is calculated by this theory. For a
proton, we have [4]: r = 1.96* 10- 15 (m), v = 2.97* 108 (m/s), and q = 1.6 * 10- 19 (c). If these
amounts are substituted into Eq.(2), we obtain: m=1.66303*10- 27 (kg). As seen, the
calculated proton mass based on this theory is equal to the actual mass of proton, which is
obtained from experiments. These two examples in microscopic scale are strong evidence
1420

and support for the proposed theory and one can test the theory for other elementary
particles.
A good example in macroscopic and astronomical scale, is the calculation of the
electric charge of the earth planet with this theory. For the earth, we have (in Gaussian
unit): r= 6.37*10 8(cm), v = 4.8* lO\cm/s) , and m=8.77*10 27 (gr). Now if these
amounts are substituted into Eq.(2), we obtain q= 1.2587*10 15 (st. coulombs). If it is
calculated the electric field of the earth with Gauss law, using this amount for the earth
charge, it is obtained: E=94 (volts/m). This amount is equal to the actual amount of the
earth electric field (on the surface) obtained by experiment [5]: E=100 (volts/meter). Note
that there is a potential difference of 300,000 Volts between the earth's surface and the
electrosphere, corresponding to an electric field strength of 6 (volts/meter) on the earth
surface. Thus we have: E=94+6=100 (volts/meter), which is the same amount that is
found from experiment.
Another example is the calculation of the electric charge of the moon using this
theory. For the moon, we have [6]: m =7.347*10 25 gr ,v =462.638 cm/s, r =1.736*10 8cm.
After substituting these amounts into Eq.(2), we obtain q =3.729428*1011(st.coloumbs).
Once again, by using the Gauss law, the electric field on the moon's surface with this
amount for electric charge is obtained: E=0.3727 (volts/m).As seen, This amount is equal
to the actual amount which is obtained for the moon's electric field from experiment.
An additional example is the calculation of the electric charge of the Sun with this
theory. For the Sun, we have [7]: m=1.988*10 33 (gr), v =1.9927*10 5 (cm/s), and
r =6.963*10 1o(cm). After substituting these amounts into Eq.(2), we obtain:
19
q =1.08368*10 (st.coloumbs). By using the Gauss law, with this amount for electric
charge, the electric field of the Sun is obtained: E=67.3322 (volts/m). As seen again,This
amount is equal to the actual amount which is obtained from experiments.
As a result, it is proved that these theories are correct about elementary particles and
astronomy and one can test them for other planets, stars, black holes, galaxies and etc.
Therefore we find two new laws in physics that it covers a wide range of phenomena,
from microscopic scale to astronomical and macroscopic scales; e.g. from a proton or
electron to planets, stares, black holes and etc.

Acknowledgments
I am very grateful to Professors Remo Rafiny, Hagen Kleinert and other organizers of the
Marcel Grossman II conference, for the possibility to attend this interesting conference
and to give this talk.

References
I. G.c. Debney, R.P. Kerr, and A. Schild, J. Math. Phys.lO, 1842 (1969).
2. J.D. Jackson, Classical Electrodynamics (University of Cali fomi a, Berkeley, 1974), pp.580-592.
3. [online]: cited on Nov 15,2006, http://en.wikipedia.org/wiki/Electron.
4. [online]: cited on Nov 15,2006, http://en.wikipedia.org/wiki/Proton.
5. D.R. Buehler, Journal of Space Mixing 2 (2004) 1-22.
6. [online]: cited on Nov 12,2006, http://en.wikipedia.org/wiki/moon.
7. [online]: cited on Nov 14,2006, http://en.wikipedia.org/wiki/sun.
 HAMILTONIAN TREATMENT OF STATIC AND COLLAPSING
SPHERICALLY SYMMETRIC CHARGED THIN SHELLS IN
LOVELOCK GRAVITY

GONQALO A. S. DIAS and JOSE P. S. LEMOS

Centro Multidiseiplinar de Astrojisica - CENTRA
Departamento de Fisica, Instituto Superior Teenico,
Universidade Teeniea de Lisboa,
Av. Roviseo Pais 1, 1049-001 Lisboa, Portugal
[email protected], [email protected]

SIJrE GAO
Department of Physics, Beijing Normal University,
Beijing 100875, China
[email protected]

Through a Hamiltonian treatment, charged thin shells in spherically symmetric space-

times, containing black holes, or other specific type of solutions, in d dimensional
Lovelock-Maxwell theory, are studied.

The total action of the theory I is the sum of the gravitational Lovelock action 1 ,
plus the electrodynamic and the matter actions

I = K, J
M
dd x
[d/2]
~
~
p=O
ClO
P
c-::. 2- P ;/"
g
V -\:J Vj
... ·.·!L2 p
V2p
R V/l.1!L2
I V2 •.• R f1,2p-lf1,2p
V2p-lV2p
(1)

4
f
~
d-2.
I M
dd X H Ff1,vF!LV - ~
f
J M
dd X H J!L Av + J
M
dd x H Lm .

Here K, is proportional to Newton's constant G, M is the d-dimensional space-

time manifold, 9 is the determinant of the spacetime metric gf1,V' 6~,1'''''~2: is totally
anti-symmetric in the upper and lower indices, Rf,~ is the Riemann tensor, E is pro-
portional to the vacuum electric permitivity EO, D d - 2 is the surface area of the d - 2
unit sphere, F!Lv is the Maxwell tensor, Jf1, is the electromagnetic current, Af1, is the
vector potential, and Lm is the matter Lagrangian. The energy-momentum tensor
derived from Lm will be that of a general perfect fluid. The free coefficients ClOp in
the Lovelock theory are chosen to obtain a sensible theory, with a negative cosmo-
logical constant appearing naturally. After writing the action and the Lagrangian
for a total spacetime comprised of an interior and an exterior regions, with a thin
shell as a boundary in between, one finds the Hamiltonian using the ADM decom-
position of the spacetime metric ds 2 = _(N.l)2dt 2 + gij(Nidt + dxi)(Njdt + dx j ),
where the gij are the canonical coordinates intrinsic to the (d - 1 )-dimensional time
foliation, and 7r ij are the respective canonical momenta. The consequent ADM de-
scription of the action is 1= J dt J dd-l x (7r ij gij - 7-i), with 7-i = N .l7-i.l + N i 7-i i
and 7-i.l = 1lt) + 7-i~) + 7-i~n), 7-ii = 7-i;9) + 7-i;e) + 7-i~m), wherc N.l, N i are the
Lagrange multipliers, and 7-i.l, 7-ii are the respective constraints 2,3 . To these one
adds E<p == P~~ - JO = 0, the electrodynamic constraint 4 , where pr is proportional

1421
1422

to the radial component of the electric field, its only non-zero component due to the
symmetry of the geometric set-up, and is the radial component of the conjugate mo-
mentum to the electrodynamic canonical coordinate Ai. The respective Lagrange
multiplier is cp == Ao. Upon particularizing to spherically symmetric spacetimes,
one reduces the relevant constraints to 1{~ = 0, 1{~ being the time translation gen-
erator, and E<p = 0, the electrodynamic constraint. Variation of the Hamiltonian
with respect to the canonical coordinates and conjugate momenta, and the relevant
Lagrange multipliers, yields the dynamic and constraint equations. The vacuum
solutions 4,5

2 r2 (2GkM + Jd- 2k ,1 E Gk Q2 ) 11k

I (r) = 1 + 12 - X rd-2k-l - (d _ 3) r2(d-k-2) , (2)

Q E Q
N = Noo = 1, per) = E r(d-2) , cp(r) = (d _ 3) r(d-3) , (3)

yield a division of the theory into two branches, namely d - 2k - 1 > 0 (which
includes general relativity, Born-Infeld type theories, and other generic gravities)
and d - 2k - 1 = 0 (which includes Chern-Simons type theories), where k is the
parameter giving the highest power of the curvature in the Lagrangian (cf. (2)).
There appears an additional parameter X = (-1 )k+l, which gives the character of
the vacuum solutions. For X = 1 the solutions, being of the type found in general
relativity, have a black hole character. For X = -1 the solutions, being of a new
type not found in general relativity, have a totally naked singularity character. Since
there is a negative cosmological constant, the spacetimes are asymptotically anti-
de Sitter (AdS), and AdS when empty. The integration of the constraint equation
1{~ = 0 from the interior to the exterior vacuum regions, through the thin shell,
takes care of the smooth junction, yielding the shell equation directly

(4)

where m == eYf"ld_2Rd-2, and M±, Q± are the masses and charges of the exter-
nal/internal vacuum solutions, resp., and 'Y± == JIf + R2 are generalized Lorentz'
factors, with If
being the metric functions of the inner and outer spacetimes, re-
spectively. The integration of E<p = 0 yields charge conservation, Q+ - Q_ = q,
with q == eYe f"ld_ 2 R d - 2 , eYe being surface charge density. The uncharged case is
treated in 6 . It is interesting to note that Eq. (4) is formally the same as that of
d-dimensional general relativity 7, however, the functions I± are different in general
relativity and Lovelock gravity. Differentiating (4) with respect to the proper time
of the thin shell, T, one obtains the equation of the acceleration of the thin shell

m,2
--x (5)
4E
 1423

where P is the pressure of the perfect fluid thin shell, and E is the right hand side
of (4). One can now study the static configurations of the thin-shell, determining
the pressure at which the shell is held in stable static equilibrium, that is R = R =
o, P -
- -(d-2)Rda
1
3n
d-2
~,+ ~,-
[(Q~-Q:'h+'Y-m
2ERd 2
a -
m
2

4E
(
'1-
df~
dR
I
Ro +'1+
df:'I)]
dR Ro '

which can be reduced to the known result for general relativity in four dimensions
P = (Gm 2 )/(167f R6(Ro - m))8.
Apart from static configurations, one can also study the collapse of the thin
shell, in particular when it is made of dust, P = 0 (note that expanding shells are
the time reversal of the collapsing shells). Following the division of the solutions
into the two branches and the two possible characters, one concludes from (4) and
(5) that the cosmic censorship holds, when the collapse is into an empty interior.
More generally, for a collapsing charged shell into initially non singular spacetimes
with generic character or empty interiors, it is proved that the cosmic censorship is
definitely upheld. Also, when the spacetimes in question have the same character
of those spacetimes provided by general relativity, (X = 1), the collapse of the
thin shells in the backgrounds, black hole or otherwise, of each different type of
Lovelock theory is in many ways similar to the collapse in general relativity itself,
and when the spacetimes in question have the opposite character (X = -1), some
other new behaviour shows up. This implies that if there are extra dimensions with
a relative large size, as proposed in large extra dimension scenarios, then differences
in the collapse of a thin shell in spacetimes with different characters can provide
the signature to uncovering not only of actual spacetime dimension d, but also the
value of the parameter k. For a detailed analysis of the above see 9 .
Acknowledgements
This work was partially funded by FCT, Portugal, through project
POCTI/FIS/57552/2004. GASD is supported by grant SFRH/BD/2003 from FCT.
SG was supported by NSFC Grants No. 10605006 and No. 10373003 and the Sci-
entific Research Foundation for the Returned Overseas Chinese Scholars, State Ed-
ucation Ministry, China.

References
1. D. Lovelock, J. Math. Phys. 12, 498 (1971).
2. C. Teitelboim and J. Zanelli, Constraint Theory and Relativistic Dynamics, edited by
G. Longhi and L. Lusanna (World Scientific, Singapure, 1987).
3. C. Teitelboim and J. Zanelli, Class. Quantum Grav. 4, L125 (1987).
4. M. Baiiados, C. Teitelboim, and J. Zanelli, Phys. Rev. D 49, 975 (1994).
5. J. Crisostomo, R. Troncoso, and J. Zanelli, Phys. Rev D 62, 084013 (2000).
6. J. Crisostomo, S. del Campo, and J. Saavedra, Phys. Rev. D 70, 064034 (2004).
7. J. Crisostomo and R. Olea, Phys. Rev. D 69, 104023 (2004).
8. K. Kuchar, Czech. J. Phys. B 18, 435 (1968).
9. G. A. S. Dias, S. Gao, and J. P. S. Lemos, Phys. Rev. D in press (2007); gr-qc/0612072.
 NEW NONUNIFORM BLACK STRING SOLUTIONS *

BURKHARD KLEIHAUS, JUTTA KUNZ

Institut fur Physik, Universitiit Oldenburg, Postfach 2503
D-26111 Oldenburg, Germany
[email protected]. de, [email protected]

EUGEN RADU
Department of Mathematical Physics, National University of Ireland
Maynooth, Ireland
[email protected]

We present nonuniform vacuum black strings in five and six spacetime dimensions. We
find qualitative agreement of the physical properties of nonuniform black strings in five
and six dimensions. Our results offer further evidence that the black hole and the black
string branches merge at a topology changing transition. The basic features of Einstein-
Maxwell-dilaton black strings can be derived from those of the vacuum black strings
after performing a Harrison transformation.

Black strings, present for D ::::: 5 spacetime dimensions, have horizon topology
5D~3 x 51. Uniform black strings possess a translational symmetry along the ex-
tracoordinate direction. As shown by Gregory and Laflamme,l these solutions are
unstable below a critical value of the mass, where at the marginally stable uniform
string a branch of nonuniform black strings arises. 2- 5
In addition to black string solutions there are also caged black hole solutions with
an event horizon of topology 5D~2. The numerical results presented in 6 suggest that,
for D = 6, the black hole and the nonuniform string branches merge at a topology
changing transition.? By numerically constructing nonuniform black strings in D =
5 dimensions, we here give evidence, that this is also true in D = 5. 8
We consider the Einstein action in a D-dimensional spacetime, where the black
string solutions approach asymptotically the D - 1 dimensional Minkowski-space
times a circle MD~l x 51. The nonuniform black string solutions are found within
the metric ansatz

d,52 = _e 2A (T,Z) f(r)dt 2 + e 2B (T.Z) (;~:) + dz 2) + e2C(T,Z)r2dnb~3' (1)

where f = 1- (TOr )D~4. The asymptotic form of the relevant metric components 9 ,10

(2)

yield mass and tension of the string solutions

nD~3L nD 3
(3)
M= 161fG ((D-3)Ct- cz), T= 161fC(Ct-(D-3)cz),

*This research has been partially supported by DFG under contract KU612/9-1 and by Enterprise-
Ireland under contract SC/2003/390

1424
 1425

where rl D - 3 is the area of the unit SD-3 sphere. The relative tension ll n = TL/l'v1.
is bounded, 0'::; n .::; D - 3, where uniform string solutions have no = l/(D - 3).
A measure of the deformation of the solutions is given by the nonuniformity
parameter .:\2

2(Rmax
.:\ = ~
Rmin _1) , (4)

where Rmax and Rmin represent the maximum radius of a (D - 3)-sphcre on the
horizon and the minimum radius, being the radius of the 'waist'. Thus for uniform
black strings .:\ = 0, while the conjectured horizon topology changing transition
should be appr.oached for .:\ ---+ 00. 4 •12
In Figure 1 we show the spatial embedding of the horizon into 3-dimensional
space for the D = 5 nonuniform black string solutions. In these embeddings the
proper radius of the horizon is plotted against the proper length along the compact
direction, yielding a geometrical view of the nonuniformity of the solutions.

A=2.0 A=9.0

2 2
1.5 1.5
1
0.5 0.5
0 0
-0.5 -0.5
-1 -1
-1.5 -1.5
-~ -~
-1.5 -1.5
-1 -1
-0.5 -0.5
0 0
0.5 0.5
1 1
1.5 1.5
2 0 2 0 4
-6 -4 -2 -4 -2
Z -6 Z

Fig. 1. The spatial embedding of the horizon of D = 5 nonuniform black string solutions with
horizon coordinate ro = 1 and asymptotic length of the compact direction L = Lcrit = 7.1713, is
shown for two values of the nonuniformity parameter, ..\ = 2, 9.

We exhibit in Figure 2 the mass of D = 5 and D = 6 nonuniform strings

and black holes. Although we see a backbending of the nonuniform string branch in
both D = 5 and D = 6 dimensions, not observed previously, because the nonuniform
string branch had not been continued to sufficiently high deformation, all our data
are consistent with the assumption, that the nonuniform string branch and the black
hole branch merge at such a topology changing transition. In fact, extrapolation
of the black hole branch towards this transition point appears to match well the
(extrapolated) endpoint of the (backbending) part of the nonuniform string branch.
For the phase diagram this would mean that we would have a region 0 < n < nb
with one branch of black hole solutions, then a region nb < n < n* with one
1426

D=6

extrapolation -
2.5 2.5
us' US

~j
BH BH

~
~
1.5
f 15 NBS

0.5 0.5
,
0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
nIno nIno

Fig. 2. The mass M of the D = 5 (a) and D = 6 (b) nonuniform string and black hole branches
are shown versus the relative string tension n (in units of the uniform string quantities Mo and
no). The black hole data are from. 6 The D = 6 black hole branch is extrapolated towards the
anticipated critical value n •.

branch of black hole solutions and two branches of nonuniform string solutions, the
ordinary one and the backbending one, and finally a region n* < n < no with only
one branch of nonuniform string solutions. Thus the topology changing transition
would be associated with n*, and nb < n < n* would represent a middle region
where three phases would coexist, one black hole and two nonuniform strings. This
is strongly reminiscent of the phase structure of the rotating black ring-rotating
black hole system in D = 5. 13
Black string solutions of the Einstein-Maxwell-dilaton theory can be obtained
via a Harrison transformation. 14 The basic features of these solutions can be derived
from those of the vacuum black string configurations. s

References
1. R. Gregory and R. Laflamme, Phys. Rev. Lett. 70, 2837 (1993) .
2. S. S. Gubser, Class. Quant. Grav. 19, 4825 (2002).
3. T. Wiseman, Class. Quant. Grav. 20, 1137 (2003).
4. T. Wiseman, Class. Quant. Grav. 20, 1177 (2003).
5. E. Sorkin, Phys. Rev. Lett. 93, 031601 (2004).
6. H. Kudoh and T. Wiseman, Phys. Rev. Lett. 94, 161102 (2005).
7. B. Kol, JHEP 0510, 049 (2005).
8. B. Kleihaus, J. Kunz and E. Radu, JHEP 0606, 016 (2006)
9. B. Kol, E. Sorkin and T. Piran, Phys. Rev. D69, 064031 (2004).
10. T. Harmark and N. A. Obers, Class. Quant. Grav. 21, 1709 (2004).
11. T. Harmark and N. A. Obers, JHEP 0405, 043 (2004).
12. B. Kol and T. Wiseman, Class. Quant. Grav. 20, 3493 (2003).
13. R. Emparan, H. S. Reall, Phys. Rev. Lett. 88, 101101 (2002).
14. D. V. Gal'tsov and O. A. Rytchkov, Phys. Rev. D58, 122001 (1998).
 LOVELOCK GRAVITY AND THE COUNTERTERM METHOD

NEDA BOSTANI

M. H. DEHGHANI

A. SHEYKHI
Physics Department and Biruni Observatory,
College of Sciences, Shiraz University, Shiraz 71454, Iran

In this paper we, first, generalize the quasi local definition of the stress energy tensor
of Einstein gravity to the case of Lovelock gravity, by introducing the tensorial form
of surface terms that make the action well-defined. In order to compute the conserved
quantities of the solutions of Lovelock gravity, We introduce the boundary counterterm
that removes the divergences of the action with flat boundary at constant t and r.

1. Introduction

The most natural extension of general relativity in higher dimensional spacetimes

with the assumption of Einstein - that the left hand side of the field equations
is the most general symmetric conserved tensor containing no more than second
derivatives of the metric - is Lovelock theory.l Since the Lovelock tensor contains
metric derivatives no higher than second order, the quantization of the linearized
Lovelock theory is ghost-free. 2
Our aim in this paper is to generalize the definition of the quasilocal stress en-
ergy tensor for computing the conserved quantities of a solution of Lovelock gravity.
The concepts of action and energy-momentum play central roles in gravity. How-
ever there is no good local notion of energy for a gravitating system. A quasilocal
definition of the energy and conserved quantities for Einstein gravity can be found
in. 3 They define the quasilocal stress energy tensor through the use of the well-
defined gravitational action of Einstein gravity with the surface term of Gibbons
and Hawking.4 Therefore the first step is to find the surface terms for the action of
Lovelock gravity that make the action well-defined. Of course, as in the case of Ein-
stein gravity, the action and conserved quantities diverge when the boundary goes
to infinity.3 For asymptotically AdS solutions, one can deal with these divergences
via the counterterm method inspired by AdS/eFT correspondence. 5 This conjec-
ture relates the low energy limit of string theory in asymptotically anti de-Sitter
spacetime and the quantum field theory on its boundary. In the present context
this conjecture furnishes a means for calculating the action and conserved quan-
tities intrinsically by adding additional terms on the boundary that are curvature
invariants of the induced metric.
The outline of our paper is as follows. In Sec. 2, we give the tensorial form of the
surface terms that make the action well-defined, generalize the Brown York energy-
momentum tensor for Lovelock gravity, and introduce the counterterm method for
calculating the finite action and conserved quantities of solutions of Lovelock gravity
with flat boundary. We finish our paper with some concluding remarks.

1427
1428

2. Lovelock Gravity and the Counterterm Method

We consider a D-dimensional spacetime manifold M with metric gJ.Lv. In this space-
time, The gravitational action satisfying the assumption of Einstein is precisely of
the form proposed by Lovelock: l

Ie = '" Jd
D
XM t
p=o
ap.c p (1)

where n == [(D - 1)/2] and [z] denotes the integer part of z, ap is an arbitrary
constant and .cp is the Euler density of a 2p-dimensional manifold

(2)

In Eq. (2) c5~1'~~ :::~:~~ is the generalized totally anti-symmetric Kronecker delta and
RJ.L/a is the Riemann tensor of the Manifold M.
The Einstein-Hilbert action (with a p = 0 for p ;:::: 2) does not have a well-
defined variational principle, since one encounters a total derivative that produces
a surface integral involving the derivative of c5gJ.Lv normal to the timelike boundary
8M. These normal derivative terms are canceled by the variation of the Gibbons-
Hawking surface term 4

IiI) = 2", r
JaM
dD-Ix.j=Ye (3)

where lab is induced metric on the boundary r = const. and e is trace of extrinsic
curvature of this boundary. The surface terms that make the variational principle of
Lovelock gravity well-defined are known in terms of differential forms. 6 The tensorial
form of these surface terms may be written as 7

n p-l ( l)P-S
h = -2",
iaM
dD-Ix.j=YL L
p=o 8=0
- pap 1-{(p)
2S(2p - 28 - 1)
(4)

where O'p is the Lovelock coefficients and 1{(p) is

1{(p) = c5[a, ... a2V-']Rb,b2
[b 1 ... b2p -
···R b2 ,-lb 2s eblal ···eb2V -1
(5)
1) ala2 'a2s-1a2s a2p-l

In Eq. (5) Rabcd(g)'s are the boundary components of the Riemann tensor of the
Manifold M, which depend on the velocities through the Gauss-Codazzi equations. 7
The explicit form of the second and third surface terms of Eq. (4) have been written
in. s
In general I = Ie+ Ib is divergent when evaluated on solutions, as is the Hamilto-
nian and other associated conserved quantities. In Einstein gravity, one can remove
the non logarithmic divergent terms in the action by adding a counterterm action
let which is a functional of the boundary curvature invariants. 9 The issue of deter-
mination of boundary counterterms with their coefficients for higher-order Lovelock
theories is at this point an open question. However for the case of a boundary with
 1429

zero curvature [Rabedh) = 0], it is quite straightforward. This is because all curva-
ture invariants are zero except for a constant, and so the only possible boundary
counterterm is one proportional to the volume of the boundary regardless of the
number of dimensions:

let = 2K:Aao r
JaM"",
dD-1xF (6)

where A is a constant which should be chosen such that the divergences of the action
is removed.
Having the total finite action, one can use the quasilocal definition of Brown
and York 3 to construct a divergence free stress-energy tensor as

n p-l ( l)P-S }
Ta = -2K: Aa "fa + '"""' '"""' - pap H(p,s)a (7)
bO { b ~~ 2s(2p _ 25 _ 1) b
p=Os=O

where H(p,s)a is
b

H(p,s)a _ 5[a, ... a2P-,a]Rb,b2 ... Rb2S-lb2s eb2s+l ... e b2p - l (8)
b - [b, ... b2p _ l b] a,a2 a2s-1a2S a2s+l a2p-l'
To compute the conserved mass of the spacetime, one should choose a spacelike
surface B in 8M with metric aij, and write the boundary metric in Arnowitt-Deser-
Misner (ADM) form:
"fabdxadxa = -N 2 dt 2 + (J"ij (dcpi + Nidt) (dcpj + Njdt)
where the coordinates cpi are the angular variables parameterizing the hypersurface
of constant r around the origin, and Nand N i are the lapse and shift functions
respectively. When there is a Killing vector field ~ on the boundary, then the quasilo-
cal conserved quantities associated with the stress tensors of Eq. (7) can be written
as

Q(O = l dD-2cpyf(iTabnae (9)

where (J" is the determinant of the metric (J"ab, ~ and n a are the Killing vector field
and the unit normal vector on the spacelike boundary B.

3. CLOSING REMARKS
The Lovelock action does not have a well-defined variational principle, since one en-
counters a total derivative that produces a surface integral involving the derivative
of 5g!-'v normal to the boundary 8M. These normal derivative terms in Lovelock
gravity are canceled by the variation of the surface terms that depend on the ex-
trinsic and intrinsic curvature of the boundary 8M. we wrote down the tensorial
form of these surface terms, and generalized the stress energy momentum tensor of
Brown and York 3 to the case of Lovelock gravity. As in the case of Einstein gravity,
1430

the action is divergent when evaluated on the solutions. We, therefore, introduced
a counterterm dependent only on the boundary volume which removed the diver-
gences of the action and conserved quantities of the solutions of Lovelock gravity
with zero curvature boundary.
We found that the counterterm (6) has only one term, since the boundaries of
our spacetimes are curvature-free.

References
1. D. Lovelock, J. Math. Phys. 12, 498 (1971); N. Deruelle and L. Farina-Busto, Phys.
Rev. D 41, 3696 (1990); G. A. MenaMarugan, ibid. 46, 4320 (1992); 4340 (1992).
2. B. Zwiebaeh, Phys. Lett. B156, 315 (1985); B. Zumino, Phys. Rep. 137, 109 (1986).
3. J. D. Brown and J. W. York, Phys. Rev. D47, 1407 (1993).
4. G. W. Gibbons and S. W. Hawking, Phys. Rev. D 15, 2752 (1977).
5. J. Maldaeena, Adv. Theor. Math. Phys., 2, 231 (1988); E. Witten, ibid. 2, 253 (1998);
O. Aharony, S. S. Gubser, J. Maldaeena, H. Ooguri and Y. Oz, Phys. Rept., 323, 183
(2000).
6. R. C. Myers, Phys. Rev. D 36, 392 (1987).
7. M. H. Dehghani, N. Bostani and A. Sheykhi, Phys. Rev. D 73, 104013, (2006).
8. S. C. Davis, Phys. Rev. D 67, 024030 (2003); M. H. Dehghani and R. B. Mann,
hep-th/0602243.
9. P. Kraus, F. Larsen, and R. Siebelink, Nue!. Phys. B563, 259 (1999).
 LG (LANDAU-GINZBURG) in GL (GREGORY-LAFLAMME)

BARAK KOU and EVGENY SORKIN 2

1 Racah Institute of Physics, Hebrew University, Jerusalem 91904, Israel
2 Department of Physics and Astronomy, University of British Columbia
6224 Agricultural Road, Vancouver V6G 1Z1, Canada

We report a study of the Gregory-Laflamme instability of black strings, or more pre-

cisely of the order of the transition, being either first or second order, and the critical
dimension which separates the two cases. First, we describe a novel method based on the
Landau-Ginzburg perspective for the thermodynamics that somewhat improves the ex-
isting techniques. Second, we generalize the comput.at.ion from a circle compactification
to an arbit.rary torus compactifications. We explain that the critical dimension cannot
be lowered in this way, and moreover in all cases studied the transit.ion order depends
only on t.he number of extended dimensions. We discuss the richer phase structure that
appears in the torus case.
In the presence of a compact dimension Gregory and Laflamme (GL) discov-
ered that uniform black strings are perturbatively unstable below a certain critical
dimensionless mass density.1 The order of the transition can be computed by fol-
lowing perturbatively the branch of non-uniform solutions which emanates from
the critical GL string, as first shown by Gubser in the case of a five-dimensional
spacetime 2 where the transition is first order. That calculation was generalized by
one of us (ES) to arbitrary spacetime dimensions with the surprising result that the
transition is first order only for D < D* = "13.5" while it is second order for higher
dimensions. 3 Here first order means a transition between two distinct configurations,
while a second order transition is smooth - the uniform string changes smoothly
into a slightly non-uniform string. Kudoh and Miyamot0 4 repeated the calculation
in the economical Harmark-Obers coordinates,5 confirmed previous results and ob-
served that in the canonical ensemble the critical dimension actually changes from
D* = "13.5" to D~an = "12.5". All this data is crucial in the construction of the
phase diagram for this system (see 6 and 7 for a review).
The present report includes two main results. First, we show how to somewhat
improve the existing method of calculating the transition order by employing a
Landau-Ginzburg perspective (the basic idea was described already in Appendix
A of8 ). Secondly, we generalize from the usual S1 == T1 compactification to an
arbitrary torus compactification TP.
Landau-Ginzburg improvement to the method. In the Landau-Ginzburg (LG)
theory of phase transitions one expands the free energy of the system around the
critical point in powers of order parameters. In particular, it is known that as long
as the coefficient of a certain cubic term in the free energy is non-vanishing then
the transition is first order. If the cubic term vanishes, for instance due to a parity
symmetry such as in our case, then it is the sign of the coefficient of a certain quartic
term, which we denote by C, that determines whether the transition is first order
or higher (of course if this term vanishes one has to go to higher terms).
Before we can compare the Landau-Ginzburg method with Gubser's method,

1431
1432

we recall the features of the latter. There one computes order by order the metric
of the static non-uniform string branch emanating from the critical GL string. The
first order is nothing but the GL mode. At the second order one computes the
back-reaction. Finally the third order is computed, or more precisely only the first
harmonic along the compact dimension, from which one can finally compute the
leading coefficients of the changes in mass and entropy, T)1, CJ2 of the new branch.
The sign of these two quantities determines the order of thetransition.
At first sight the two methods look quite different. However, we show that in
the LG method one also needs to precisely compute the second order back-reaction
to the metric. The third order however is not required in LG (thereby avoiding the
sol'ution of a set of linear differential equations with sources). Rather one needs only
to expand the action to an appropr'iate quartic order, to substitute in the results from
the first and second orders and perform certain integrals that add up to the constant
whose sign determines the order.
A way to understand the simplification is the following: in Gubser's method
one computes the third order, but it turns out that all that is really needed is the
projection of the third order onto the GL mode. That is precisely the reason why the
first harmonic sufficed (as the GL mode is in the first harmonic). Our substitution
into the quartic order of the free-energy achieves exactly that, without the need to
compute other properties of the third order. We perform the "Landau-Ginzburg"
calculation for an Sl compactification in various dimensions and verify that we get
the same bottom-line coefficients T)1, CJ2 as in the previous method.
Torus compactification. It is interesting to generalize the compactifying manifold,
and the simplest option beyond the circle Sl == T1, is a product of circles, or more
generally a p-torus TP. The number of extended spacetime dimensions is denoted
by d and the total spacetime dimension is D = d + p. The critical GL density for
such a torus compactification is easily found to be given in terms of the shortest
vector in the reciprocallattice. 9
Before we proceed to analyze the transition order we note that it sufficies to
restrict to square torii. Basically, we view the space of torii as having two boundaries
- on the one hand highly asymmetrical torii, where one (or more) dimensions are
much larger than the rest, and on the other hand highly symmetrical torii such as
the square torus. Since the limit of a highly asymmetrical torus reduces to the case of
a lower dimensional torus (mostly the well-understood case of Sl compactification),
we argue that by studying the opposite limit of a highly symmetrical torus, we
achieve an understanding of both limits and thereby also some understanding of
the intermediate region of general torii.
For a square TP torus compactification, p modes turn marginally tachyonic at
the same (GL) point. We find that the constant C is replaced by a p x p quadratic
form Cij, in order to allow for the various possible directions in the (marginally)
"tachyon space", and that the transition is second order iff cij is positive for all
directions. Namely, it is enough that there is a single direction in tachyon space
which sees a first order transition for the transition to be one. Taking into account
 1433

the Tl results we may immediately deduce that the critical dimension cannot be
lower than in the Tl case with the same number of extended dimensions, d.
Due to the high degree of symmetry of the square torii Cij consists only of two
independent entries: all the diagonal entries are the same, denoted by C=, and all
the off-diagonal entries are the same, denoted by Ctf. Since the diagonal term is
precisely the one computed in the Tl compactification, we set to compute the off-
diagonal ternl. Due to the symmetry Ctf- is the same for all p and for that purpose
it suffices to consider p = 2, namely we consider the square T2 torus. The only
parameter remaining is the number of extended dimensions. Once we have chosen
the diagonal direction in tachyon space we are not bothered any longer by the
presence of several (marginally) tachyonic modes. However, the number of metric
components involved in the calculation (back reaction and quartic coefficient) is
larger than in the Tl case. Certain discrete symmetries are found to be helpful in
simplifying the calculation.
As a result we find that for all the studied values of d where the Tl transition is
second order, the TP transition is also second order for all p. Combining this result
with observation mentioned above we conclude that the transition order for square
tOTii shows some robustness in that it depends only on d, the number of extended
dimensions, and not on p.
In addition we note some subtler implications, including the finding that for
almost all d the diagonal direction in tachyon space is disfavored relative to turning
on a tachyon in a single compact dimension, and in this sense we have spontaneous
symmetry breaking.
Acknowledgments. This research is supported in part by The Israel Science Foun-
dation (grants no 228/02, 607/05) and by the Binational Science Foundation BSF-
2002160, BSF-2004117. ES is supported in part by the CIAR Cosmology and Grav-
ity Program and by the NSERC of Canada.

References
1. R. Gregory and R. Laflamme, "Black Strings And P-Branes Are Unstable," Phys.
Rev. Lett. 70,2837 (1993).
2. S. S. Gubser, "On non-uniform black branes,", Class. Quant. Grav. 19,4825 (2002).
3. E. Sorkin, "A critical dimension in the black-string phase transition," Phys. Rev. Lett.
93, 031601 (2004) [arXiv:hep-th/0402216].
4. H. Kudoh and U. Miyamoto, "On non-uniform smeared black branes," Class. Quant.
Grav. 22, 3853 (2005) [arXiv:hep-th/0506019].
5. T. Harmark and N. A. Obers, "Black holes on cylinders," JHEP 0205, 032 (2002).
6. B. Kol, "The phase transition between caged black holes and black strings: A review,"
Phys. Rept. 422, 119 (2006) [arXiv:hep-th/0411240j.
7. T. Harmark and N. A. Obers, "Phases of Kaluza-Klein black holes: A brief review,"
arXiv:hep-th/0503020.
8. B. Kol, "Topology change in general relativity and the black-hole black-string transi-
tion," JHEP 0510, 049 (2005) [arXiv:hep-th/0206220j.
9. B. Kol and E. Sorkin, "On black-brane instability in an arbitrary dimension," Class.
Quant. Grav. 21,4793 (2004) [arXiv:gr-qc/0407058j.
 CAUSAL STRUCTURE AROUND SPINNING 5-DIMENSIONAL
COSMIC STRINGS

REINOUD JAN SLAGTER

A SF YON, Astronomisch Fysisch Onderzoek Nederland,
1405EP Bussum, The Netherlands,
[email protected]

We present a numerical solution of a stationary 5-dimensional spinning cosmic string in

the Einstein-Yang-Mills (EYM) model, where the extra bulk coordinate 'IjJ is periodic.
It turns out that when g",,,, approaches zero, i.e., a closed time-like curve (CTC) would
appear, the solution becomes singular. We also investigated the geometrical structure of
the static 5D cosmic string. Two opposite moving 5D strings could, in contrast with the
4D case, fulfil the Gott condition for CTC formation.

Keywords: Bulk space time; Einstein Yang-Mills; Cosmic strings; Closed timelike curves.

In recent years higher dimensional gravity is attracting much interest. The possibil-
ity that spacetime may have more than four dimensions is initiated by high energy
physics and inspired by D-brane ideology in string theory. Our 4-dimensional space-
time (brane) is embedded in the 5-dimensional spacetime (bulk). It is assumed that
all the standard model degrees of freedom reside on the brane, where as gravity
can propagate into the bulk. The effect of string theory on classical gravitational
physics is investigated by the low-energy effective action. In General Relativity(GR),
gravitating non-Abelian gauge field, i.e., the Yang-Mills(YM) field, can be regarded
as the most natural generalization of Einstein-Maxwell(EM) theory. In particular,
particle-like, soliton-like and black hole solutions in the combined EYM models,
shed new light on the complex features of compact object in these models.
Gravitating cosmic strings became of interest, when it was discovered that in-
flat ion airy cosmological models solved many shortcomings in the standard model.
Inflation is triggered by a Higgs field (<I» on the right hand side of the equations of
Einstein. The Higgs field is also responsible for the formation of topological defects
during the fabric of space time at the early stages of the universe. The last decades
many physicians studied the consequences of topological defects in general relativ-
ity. One of the earliest investigation was the gravitational Aharonov-Bohm effect in
a conical space time. The cosmic string can also be described as point particle in
(2+ 1) dimensional space time. An interesting example of the richness of the (2+1)
dimensional gravity, is the Gott space time. l An isolated pair of point particles,

1434
 1435

moving with respect to each other, may generate a surrounding region where close
timelike curves (CTC) occur. It was not a surprise that one can prove that the Gott
space time has unphysical features. 2
It is quite natural to consider as a next step the non-Abelian EYM situation is
context with cosmic string solutions and its unusual features like the formation of
CTC's. There is some evidence that, by suitable choice of the gauge, CTC's will not
emerge dynamically.3,4 Let us now consider a 5-dimensional space time with action

s = 1~1T Jd5xV-g5 [~5 (R - A) + K,(Rl-'va(3Rl-'va(3 - 4R a(3R a(3 + R2) - g12 TrF2] ,

(1)
with G 5 the gravitational constant, A the cosmological constant, K, the Gauss-
Bonnet coupling and g the gauge coupling. The coupled set of equations of the
EYM-GB system will then become

(2)
with GI-'V the Einstein tensor, GBI-'v the Gauss-Bonnet tensor and F/:v = 81-'A~ -
8 v AaI-' + gE abc AbI-' ACv·
Consider now the stationary axially symmetric 5-dimensional space time

ds 2 = -F(r)(dt + w(r)d'lj!)2 + dr 2 + dz 2 + A(r)2r 2dip2 + B(r)2d'lj!2, (3)

with the YM parameterization only in the brane (A~a) = A~a) = A~) = 0):

A~a) = (<I>(r)cosip,<I>(r)sinip,O), A~a) = (O,O,W(r)-l). (4)

The equations are easily solved numerically. From a combination of the YM equa-
tions, we obtain also for the angular momentum component w a first order expression
Bw(2F B' - BF')
w' = --~--~---=~
2 (5)
F(Fw + B2)
which can be used in the numerical code. We also have a constraint equation from
the Einstein equations

(6)

So a negative cosmological constant will keep g1jJ1jJ positive, which is desirable. Fur-
ther, from \7I-'Tl-'v = 0, we obtain in combination with the YM equations the same
equation as Eq.(5), which proves the consistency of the system. Without A, the
solution becomes singular when g1jJ1jJ approaches zero. So it seems that CTC forma-
tion is prevented by singular behaviour. In figure 1 we plotted a typical solution for
negative A (no GB contribution) . We see that g1jJ1jJ remains positive and no singular
behaviour is observed. So again no CTC will form. An interesting question is what
will happen with the reduced (3+ 1) dimensional model when the dz 2 term in the
metric is omitted, on the analogy of the Gott pair in (2+1) dimensional space time.
1436

A B

Fig.1 Typical solution for negative A

The effectively (3+1) dimensional point particles can be Lorentz-boosted in
opposite directions and one can close space by two rotations about the two angle
variables. The matching condition then yields

(cos 2 2(3 - 1)(1- 2cosh

2
0 < ~J3. (7)
with (3 = 4n-Gm = n(l - 8nGa), m the mass, a the angle deficit and tanh(O = v
the velocity of the point particles. Further we took equal angle deficits in the cp and
'IjJ variables. The Gott pair will produce a CTC when 1 cosh ~ sin (3 > 1 If we choose
the minimal possible value sin (3 = co~h E;' we obtain

/110 + 22V3
1 < cosh~ < 11 ~ 1.11 (8)
So in the static geometrical approach, there could be, in principle, a CTC. We
conjecture that in the more realistic situation, where a matter field is present, this
approach is not valid anymore.

References
1. J. R. Gott, Phys. Rev. Lett. 66, 1126 (1990).
2. S. Deser, R. Jackiw and G. 't Rooft, Phys. Rev. Lett. 68, 267 (1992).
3. R. J. Slagter, in Proceedings of Eigth Marcel Grossmann Meeting, edited by T. Piran
and R. Ruffini (World Scientific, Singapore, 2006) pp. 602-604.
4. R. J. Slagter, Phys. Rev. D 54, 4873 (1996), gr-qc/0609003.
 SHORT DISTANCES, BLACK HOLES, AND TeV GRAVITY

IV AN AGULLO and JOSE NAVARRO-SALAS

Dept. Fisica Teorica and IFIC, Centro Mixto Universidad de Valencia-CSIC,
Universidad de Valencia, Burjassot-461 00, Valencia, Spain
ivan. [email protected]
[email protected]

GONZALO J. OLMO
Physics Department, University of Wisconsin-A1ilwaukee,
P.O. Box 413, Milwaukee, Wisconsin 53201 USA
[email protected]

The Hawking effect can be rederived in terms of two-point functions and in such a
way that it makes it possible to estimate, within the conventional semiclassical theory,
the contribution of ultrashort distances at 1+ to the Planckian spectrum. Thermality is
preserved for black holes with f'(,/p « l. However, deviations from the Planckian spectrum
can be found for mini black holes in TeV gravity scenarios, even before reaching the
Planck phase.

1. Introduction

In 1974 Hawking predicted the thermal emission of quanta by black holes 1 using
semiclassical gravity. The deep connection of this result with thermodynamics and,
in particular, with the generalized second law, strongly support its robustness and
its interpretation as a low-energy effect, not affected by the particular underlying
theory of quantum gravity2 . However 3 , ultrahigh frequencies (or ultrashort dis-
tances) seem to playa crucial role in the derivation of the Hawking effect. Any
emitted quanta, even those with very low frequency at future infinity, will suf-
fer a divergent blueshift when propagated backwards in time and measured by a
freely falling observer. The exponential redshift of the event horizon provides, to
the external observer, a glimpse of the world at very short-distance scales, where
semiclassical tools are not well justified. All derivations of Hawking radiation seem
to invoke Planck-scale physics in a fundamental way, which makes it unclear the way
to parameterize the contribution of transplanckian physics in black hole radiation.
We propose an alternative to the standard approach in terms of Bogolubov
coefficients to derive the Hawking effect. In our approach, the correlation functions
of the matter fields are used to compute the spectrum of the emitted particles. This
provides an explicit way to evaluate the contribution of ultrashort distances (Planck-
scale) to the spectrum of Hawking quanta within the semiclassical approach.
Let us assume, for the sake of simplicity, that ¢ is a massIes, neutral, and min-
imally coupled scalar field. One can easily verify that the number operator can be
obtained from the following projection

aoutJajut = r dL:idL:2' [Ui ut (Xl)8/L][ujut* (x2)8v] {¢(xI)¢(x2) - (outl¢(Xl)¢(X2)lout)} ,

J~ (1)

1437
1438

where ufut (x) is a normalized positive frequency mode with respect to the inertial
time at future infinity, and I; represents a suitable initial value hypersurface. There-
fore, the number of particles in the ith mode measured by the "out" observer in the
"in" vacuum is given by (inJNrtJin) == (inJNiiutJin), where Nijut == n- 1a out Jajut
can be easily worked out using the above expression. Let us now apply (1) to
the formation process of a Schwarzschild black hole and restrict the "out" re-
gion to future null infinity (I+). The "in" region is, as usual, defined by past
null infinity (I-). At f+ we can consider the normalized radial plane-wave modes
u':nl;,,(t,r,e,r/J) = u w (u)r- 1 Yzm(e,r/J), where uw(u) = ~ and u is the null re-
tarded time. Using these modes in (1) one finds 4 ,5 (for simplicity we omit the
factor 01, b oml m2)

where z = Ul -U2 represents the "distance" between the points Ul and U2 and tim (w)
are the transmission coefficients. To get the Planckian spectrum, there remains to
perform the integration in z

(inJNoutJin) = -Jt1m(w)i2j+CXJ dze- iwz [ ",2 e -K,z _ ~] = Jt lm (W)J2 .

w 271'w -CXJ (e-K,Z - 1)2 z2 e27fWK, 1 - 1
The interesting aspect of the above expression is that it allows us to explicitly
evaluate the contribution of distances to the thermal spectrum. To be more explicit,
the integral

f(w, "', E)
1
= ---
j+E dze- tWZ
. [",2
(_
e -K,Z
)2 -
1]
2"
271'W -E e I<Z - 1 z
can be regarded as the contribution coming from distances z E [-E, E] to the full
spectrum. This integral can be solved analytically. For details and the case of a
massless spin s = 1/2 field see 4 ,5 . Obviously, in the limit E ----> 00, we recover the
Planckian result f(w, "', (0) = (e 27fW K,-1 _1)-1. For a rotating black hole the result
is similar with the usual replacement of w by w == w - mD H (m is the axial angular
momentum quantum number of the emitted particle and DH the angular velocity
of the horizon).
On the other hand, if we take E of order of the Planck length [p = 1.6 x 1O- 33 cm,
we obtain that the contribution to the thermal spectrum at the typical emission
frequency, Wtypical ~ ",/271' == T H , due to transplanckian scales is of order ",[p. This
contribution is negligible for macroscopic black holes with typical size much bigger
than the microscopic Planck length. In fact, for three solar masses black holes the
contribution to the total spectrum, (e 27fW ",-1 -1)-1, at Wtypical is of order 10- 38 %.
We need to look at high frequencies, W/Wtypical ~ 96, to get contributions of the
same order as the total spectrum itself. This is why Hawking thermal radiation is
very robust, as it has been confirmed in analysis based on acoustic black holes 6 .
Our results, in addition, indicate that when the product "'[p is of order unity,
 1439

the contribution of short distances to the Planckian spectrum is not negligible.

The integral J(w, Ii, E) gives values similar to (e 27fW ,,-1 - 1)-1 when W/Wtypical is
not very high. This happens in the case of black holes predicted by Te V gravity
scenarios 7,8 . For detailed and recent results see9 . Assuming a drastic change of
the strength of gravity at short distances due to n extra dimensions (a Planck
mass M TeV of 1 TeV) and for a (4 + n)-dimensional Kerr black hole with surface
gravity Ii rv 0.6 - 1 TeV- 1 (this means M rv 5 - 10 TeV when a=O), we obtain
that, at w = 1i/27r = TH, around the 20% of the spectrum comes from distances
shorter than the new Planck length lTeV rv 10- 17 em, for 17, = 2 - 6 and for spin
zero particles. Moreover, at frequencies w RO 3TH the contribution of ultra-short
distances is of order of the total spectrum itself. For massless s = 1/2-particles the
results can be obtained from the formulaes of5 . In this case the contribution from
ultrashort distances is smaller than for spin zero and it is around the 0.2% of the
spectrum at w = 1i/27r = TH. In addition we find that, for Ii = 0.9 - 1, and 17, = 6
we need to go to frequencies w RO 5.5TH and w RO 5.6TH, respectively, to find short-
distance contributions of order of the fermionic thermal spectrum (e27fWK-l + 1)-1.
For Ii = 0.6 - 0.8, and 17, = 2 we obtain w RO 6.2TH and w RO 6.9TH, respectively.
Therefore, in Te V gravity scenarios the spectrum of Hawking quanta is sensitive to
transplanckian physics and significant deviations from the thermal spectrum can
emerge in the "semiclassical" phase of the evaporation.
Acknowdelgements. We thank L. Parker for collaboration on the topic of this work.
J. N-S also thanks M. Casals for interesting discussions on rotating black holes and
Te V gravity. 1. A. also thanks R.M. Wald for useful discussions.

References
1. S.W. Hawking, Comm. Math. Phys. 43199 (1975)
2. R. M. Wald, Quantum field theory in curved spacetime and black hole thermodynamics,
CUP, Chicago (1994). V.P. Frolov and 1.D. Novikov, Black hole physics, Kluwer Aca-
demic Publishers, Dordrecht (1998). A. Fabbri and J.Navarro-Salas, Modeling black
hole evaporation, ICP-World Scientific, London (2005)
3. T. Jacobson, Phys. Rev. D 44 1731 (1991); Phys. Rev D 48 728 (1993)
4. 1. Agullo, J. Navarro-Salas and G.J. Olmo, Phys. Rev. Lett. 97, 041302 (2006)
5. 1. Agullo, J. Navarro-Salas, G.J. Olmo and L. Parker, hep-thj0611355.
6. W.G. Unruh, Phys. Rev. D 51 2827 (1995); R. Braut, S. Massar, R. Parentani and P.
Spindel, Phys. Rev. D 52 4559 (1995); S. Corley and T. Jacobson, Phys. Rev. D 54
1568 (1996). R. Balbinot, A. Fabbri, S. Fagnocchi and R. Parent ani, gr-qcj0601079
7. N. Arkani-Hamed, S. Dimopoulos and G.R. Dvali, Phys. Lett. B 429, 263 (1998). L.
Randall and R. Sundrum, Phys. Rev. Lett. 83, 3370 (1999)
8. S.B. Giddings and S. Thomas, Phys. Rev. D 65, 056010 (2002). S. Dimopoulos and
G. Landsberg, Phys. Rev. Lett. 87, 161602 (2001).
9. M. Casals, S.R. Dolan, P. Kanti and E. Winstanley, hep-thj0608193. M. Casals, P.
Kanti and E. Winstanley, JHEP 0602, 051 (2006). G. Duffy, C. Harris, P. Kanti and
E. Winstanley, JHEP 0509, 049 (2005).
 BLACK STRING SOLUTIONS WITH NEGATIVE
COSMOLOGICAL CONSTANT

ROBERT MANN
Perimeter Institute for Theoretical Physics, Waterloo, Ontario N2L 2Y5, Canada and
Department of Physics, University of Waterloo Waterloo

EUGEN RADU
Department of Mathematical Physics, National University of Ireland, Maynooth, Ireland

CRISTIAN STELEA
Department of Physics, University of Waterloo Waterloo

We present arguments for the existence of new black string solutions with negative
cosmological constant. These higher-dimensional configurations have no dependence on
the 'compact' extra dimension, and their conformal infinity is the product of time and
Sd-3 X R or H d - 3 X R. The configurations with an event horizon topology Sd-2 X Sl
have a nontrivial, globally regular limit with zero event horizon radius.

The physics of asymptotically Anti-de Sitter (AdS) black hole solutions is of

particular interest due to the AdS/eFT conjecture. The thermodynamic properties
of black objects in AdS offers the possibility of studying the nonperturbative aspects
of certain conformal field theories living on the AdS boundary.
Here we present arguments for the existence of a class of configurations, which
we interpret as the AdS counterparts of the A = 0 uniform black string solutions. l
For such solutions the topological structure of the AdS boundary is the product
of time and Sd-3 x R or H d- 3 X R and they correspond to black strings with
the horizon topology Sd-3 x Sl or H d- 3 X R respectively (here the black string is
wrapping the Sl circle).
We consider the following parametrization of the d-dimensional line element
(with d ;:::: 5)
2
2 2 dr 2 2 2
ds = a(r)dz + f(r) + r dL,k,d-3 - b(r)dt (1)

where the (d-3)-dimensional metric dL,%,d_3 is

df2 L 3 fork=+l
dL,%,d_3 = L~:; dx; for k = 0 (2)
{
d3L3 for k = -1

where df2~_3 is the unit metric on Sd-3. By H d- 3 we will understand the (d-3)-
dimensional hyperbolic space, whose unit metric d3~_3 can be obtained by analytic
continuation of that on Sd-3. The direction z is periodic with period L.
The event horizon is taken at constant r = rh where the metric functions b(r)
and f(r) are vanishing while a(r) takes positive values, b(r) = b1 (r - rh) + O(r-

1440
 1441

rh)2, f(r) = 11 (r - rh) + O(r - rh)2, b(r) = ah + al (r - rh) + O(r - rh)2. For even
d, the black string solutions admit at large r a power series expansion of the form:

(3)

where aj, fj are constants depending on the index k and the spacetime dimension
only and A = -(d - l)(d - 2)/(2£2) is the cosmological constant. A similar expan-
sion can be written for an odd number of spacetime dimensions, the main difference
being the presence of logarithmic terms in the corresponding expressions. The mass
M and the tension T of the black strings are found by using a boundary countert-
erm prescription, 2 being determined by the constants Ct and C z in the asymptotic
expansion at infinity
£d-4
M=Mo+M~k,d), Mo= 161TC[Cz-(d-2)Ct]LVk,d-3, (4)
£d-4
T = To + ~(k,d), To = ~C [(d - 2)c z - Ct]Vk d-3 , (5)
161T '
where Vk,d-3 is the total area of the angular sector. Here MJk,d) and rc(k,d) are
Casimir-like terms which appear for an odd spacetime dimension only.2
These solutions have a nonzero temperature and an entropy

TH = 4~ Vr~~2 [(d - l)r~ + k(d - 4)£2], S = 4~r~-3Vk,d-3Ly'ah, (6)

and satisfy the Smarr relation

(7)

We have found black string numerical solutions with AdS asymptotics in all di-
mensions between five and twelve. We conjecture that they exist for any d and,
in the case k = 1, for any value of the event horizon rho For all the solutions we
studied, the metric functions a(r), b(r) and f(r) interpolate monotonically between
the corresponding values at r = rh and the asymptotic values at infinity, without
presenting any local extrema. The metric functions a(r), b(r) and f(r) of a typical
solution are presented in Figure 1.
The condition for a regular event horizon implies in the k = -1 case the existence
of a minimal value of rh for a given A. The k = 0 solution appears to be unique,
with only one free parameter, corresponding to the known planar topological black
hole, a = r2, f = l/b = -2M/r d- 3 + r2/£2.
1442

70

2.5
60 f(r)/

50
1.5
,;»</;;;'///.
40

0.5
30
0
0.5 0.75 1.25

20

10

Fig. 1. The profiles of the metric functions a( r), b( r) and f (r) are shown for a typical k = 1 black
string solution with d = 6, k = 1 and rh = 0.5.

As with the spherically symmetric Schwarzschild-AdS solutions, the temperature

of the k = 1 black string solutions is bounded from below. The free energy of these
solutions is positive for small rh and negative for large rho The phase transition
found in 3 occurs here as well. The k = 1 solutions have a nontrivial zero event
horizon radius limit corresponding to AdS vortices. As rh ---+ 0 we find e.g. Ct(d =
6) ~ -0.0801, Ct(d = 7) ~ -0.0439, Ct(d = 8) ~ 0.0403, Ct(d = 9) ~ 0.0229, while
Ct(d = 10) ~ -0.0246.
More details on these solutions are presented in. 4 Five dimensional AdS black
string with an event horizon topology 52 x 51 are also discussed in. 5

Acknowledgments
This work was carried out in the framework of Enterprise-Ireland Basic Science
Research Project SC/2003/390.

References
1. G. T. Horowitz and A. Strominger, Nucl. Phys. B 360 (1991), 197.
2. V. Balasubramanian and P. Kraus, Commun. Math. Phys. 208 (1999) 413 [arXiv:hep-
th/9902121].
3. S. W. Hawking and D. N. Page, Commun. Math. Phys. 87 (1983) 577.
4. R. B. Mann, E. Radu and C. Stelea, JHEP 0609 (2006) 073 [arXiv:hep-th/0604205].
5. K. Copsey and G. T. Horowitz, JHEP 0606 (2006) 021 [arXiv:hep-th/0602003].
 MATCHED ASYMPTOTIC EXPANSION
FOR CAGED BLACK HOLES

DAN GORBONOS and BARAK KOL

Racah Institute of Physics, Hebrew University
Jerusalem 9190{, Israel
[email protected], [email protected]

We present an analytic perturbation theory for a small (static) black hole in a space-
time with a compact dimension. We set up a general procedure for an arbitrary order
in the perturbation series (including regularization of divergences) based on an asymp-
totic matched expansion between two coordinate patches: the near horizon zone and the
asymptotic zone. Using this method we obtain corrections to the thermodynamic quan-
tities of the black hole as well as its deformation due to the presence of the compact
dimension.
The black-string black-hole transition (which includes the Gregory-Laflamme
instability 1 ) is a phase transition in General Relativity which occurs at higher di-
mensions d > 4 where black hole uniqueness fails. It is known to raise deep issues
including topology change and critical dimensions as well as naked singularities and
thunderbolts (see the reviews 2 ,3 and references therein).
In order to test these issues, black object solutions were sought both analytically
and numerically in jRd-2,1 x Sl. One limit which is amenable to analytic study is the
small black hole limit, where Po, the black hole radius, is much smaller than L, the
size of the compact dimension. Here we present an analytic (though perturbative)
procedure to obtain solutions for small black holes. This procedure is based on the
famous "Matched Asymptotic Expansion" . Some other methods were used to obtain
similar results 4- 6 and a full numerical solution for the black hole branch in d = 5,6. 7
We denote by z, r the "cylindrical" coordinates, where z is the coordinate along
the compact dimension, and r is the radial coordinate in the extended jRd-2 spatial
dimensions. In addition we introduce "spherical" coordinates p, X.
In this method we consider two zones:

• The asymptotic zone where P » Po and Po is the small parameter. The

zeroth order solution is flat space with a periodic coordinate z rv Z + Land
the point at the origin (r, z) = (0,0) removed .
• The near zone where P « Land L -1 is the small parameter. The zeroth
order solution is the Schwarzschild black hole with radius Po.

In each zone the procedure is ordinary perturbation expansion in a dimensionful

parameter. Thus in the asymptotic zone we expand the metric in Po and in the near
zone we expand it in L -1. The metric in the two regions must be consistent over the
overlap region Po « p « L, in a double expansion in Po, L. This is the "matching
procedure". For example, the first match is between the Schwarzschild-Tangherlini
solution in the near zone (order L -0) and the Newtonian approximation (order
pg-3). The perturbation in the asymptotic zone is the post-Newtonian expansion
when in the near zone we generalize to d dimensions the Regge and Wheeler analysis

1443
1444

of static black hole perturbations. 8

At each order in the expansion one needs to solve a non-homogenous linear
equation ~ the linear equation being the same as the one which appears at first
order and the non-homogeneous source term being constructed from lower order
metric functions. The solution to this equation is determined up to a solution of
the homogeneous equation. This indeterminacy for each zone on its own reflects
the freedom of adding external field multipoles ~ in the asymptotic zone they
are situated at the origin while in the near zone they are at infinity. These external
multipoles must be determined by matching with the other zone, a procedure which
requires to identify (after matching the gauge) the leading terms in the metric on
both zones. We call this process "a dialogue of multipoles".
A priori it is not obvious that the required terms from the other zone are already
available at the right time, namely that the method is well-posed (that there are
sufficient boundary conditions). Hence it is interesting to study at any given order
in a specific zone which orders must be already available for matching from the
other one, and thereby describe the pattern of the dialogue ~ the orders at which
one should alternate between the zones. This pattern can be determined by a simple
dimensional analysis of the multipole coefficients,9 and indeed we find the system
to be well-posed.
In the next to leading order in the expansion of the metric a new qualitative
phenomenon appears - divergences, which are familiar from Post-Newtonian stud-
ies. We obtained a clear regularization in terms of a "cut-off and match" method,
which relies on Hadamard's "Partie finie" regularization. 10 Alternatively, a certain
subtraction of self-interaction is shown to be an equivalent method at this order.
The perturbation method for the metric (in both patches) may be carried in
principle to an arbitrarily high order in the small parameter. From the expansion
of the metric we calculated to the next to leading order some physical quantities
in each patch independently (although the thermodynamical quantities are related
via the first law and its integrated form - Smarr's formula).
The thermodynamic quantities are obtained from matching monopole correc-
tions to the metric between the two zones. From the metric in the near zone we
calculated the entropy and temperature:

(1)

(2)

where

.= (d-2)((d-3) (PO)d-3
E. d-3 L '

Od-2 is the area of a unit Sd-2 and ((0:) is the Riemann zeta function. In the
 1445

asymptotic zone the mass and the tension are:

M = (d - 2) Dd - 2 pg-3 (1 + d - 3 E) + 0 (E2) (3)

167rG d 2(d-2)
(d - 3? Dd - 2 p~-3 0 ( 2)
T= 327fG E+ E (4)
d
Higher multipole corrections correspond to the change in the shape of the black
hole in response to the field (multipoles) created by its periodic "mirrors" .. For
example the quadrupole match gives the first deviation from spherical symmetry
from which the eccentricity is given by
(d - 3)4 (r(2 + ~))2 ((d - 1) pg-l p8
E= 8(d-2)r(d~3) . Ld-l +O(Ld)' (5)

Another interesting physical quantity, for which we have to integrate over the
metric in both zones, is the inter-polar distance. It is defined to be the proper
distance between the "poles" of the black hole measured around the compact di-
mension. It was found that the black hole repels the spacetime around it and for
d = 5 the effect is the strongest, where to leading order in the small parameter the
inter-polar distance does not decrease at all. We term that "a black hole Archimedes
effect" .

References
l. R. Gregory and R. Laflamme, "Black strings and p-branes are unstable," Phys. Rev.
Lett. 70, 2837 (1993) [arXiv:hep-th/9301052j.
2. B. Kol, "The phase transition between caged black holes and black strings: A review,"
Phys. Rept. 422, 119 (2006) [arXiv:hep-th/0411240j.
3. T. Harmark, V. Niarchos and N. A. Obers, "Instabilities of black strings and branes,"
arXiv:hep-th/0701022.
4. T. Harmark, 'Small black holes on cylinders," Phys. Rev. D 69, 104015 (2004)
[arXiv:hep-th/0310259j.
5. Y. Z. Chu, W. D. Goldberger and I. Z. Rothstein, "Asymptotics of d-dimensional
Kaluza-Klein black holes: Beyond the newtonian approximation," JHEP 0603, 013
(2006) [arXiv:hep-th/0602016j.
6. D. Karasik, C. Sahabandu, P. Suranyi and L. C. R. Wijewardhana, "Analytic approx-
imation to 5 dimensional black holes with one compact dimension," Phys. Rev. D 71,
024024 (2005) [arXiv:hep-th/0410078j.
7. H. Kudoh and T. Wiseman, "Connecting black holes and black strings," Phys. Rev.
Lett. 94, 161102 (2005) [arXiv:hep-th/0409111j.
8. T. Regge and J. A. Wheeler, "Stability Of A Schwarzschild Singularity," Phys. Rev.
108, 1063 (1957).
9. D. Gorbonos and B. Kol, "A dialogue of multipoles: Matched asymptotic expansion
for caged black JHEP 0406, 053 (2004) [arXiv:hep-th/0406002j.
10. D. Gorbonos and B. Kol, "Matched asymptotic expansion for caged black holes:
Regularization of the post-Newtonian order," Class. Quant. Grav. 22, 3935 (2005)
[arXiv:hep-th/0505009j.
 PERTURBATIVELY NON-UNIFORM CHARGED BLACK
STRINGS: A NEW STABLE PHASE

UMPEI MIYAMOTO
Department of Physics, Waseda University, Okubo 3-4-1, Tokyo 169-8555, Japan
umpei at gravity.phys.waseda.ac.jp

HIDEAKI KUDOH
Department of Physics, UCSB, Santa Barbara, CA 93106, USA
and
Department of Physics, The University of Tokyo, Tokyo 113-0033, Japan
kudoh at utap.phys.s.u-tokyo.ac.jp

The final fate of the Gregory-Laflamme instability is one of the most interesting problems
in the black hole physics. We investigate the thermodynamical stability of gauge-charged
non-uniform black strings with higher-order perturbations. We find that there exists a
new stable state of non-uniform black strings, which can be the final fate of the dynamical
instability in any dimensions.

1. Introduction
It is known that black objects with translational symmetries, such as black branes
and black strings, suffer from the Gregory-Laflamme (GL) dynamical instability,
breaking the translational symmetries. 1 To know the final fate of GL instability,
it is useful to construct black objects, such as localized black holes, black strings
and so on, in Kaluza-Klein spacetimes. 2 The wllOle phase structure in a vacuum
spacetime has been clarified recently. 3 One of the most interesting features is the
phase structure, therefore the final fate of GL instability, will depend on spacetime
dimensions. 4 Since the gravity inevitably couples to other fundamental fields, such
as gauge fields and dilaton, it is important to generalize these analyses to those
of black objects with charges. As the first step, we construct non-uniform charged
black strings perturbatively in various dimensions in this article a.

2. Static perturbations of magnetic black strings

Let us consider D = d + 1 dimensional spacetime with the gravity coupled to (d - 2)-
form field Fd-2. The governing equations are
_ 1 d-3 2
R I-'V - 2(d _ 3)! F I-'
"2 ... "d-2
r r :FVI-'2,,·l-'d-2 - 2(d _l)!gl-'v F ,
V"FI-'I-'2"'l-'d-2
•
= O.
In addition, the form field satisfies the Bianchi identity, dFd - 2 = O. We construct
non-uniform black strings perturbatively with a metric ansatz given by
2
ds 2
d+l
= _e 2a (r,z)j dt2
+
+ e2b(r,z)j- (dr
i+j~
+ dz 2 ) + e2c (r,z)r 2 dn d-2'
2

aThis article is based on our paper 5, although a new result, which is related to the stability of
non-uniform black strings in a grand canonical ensemble, is added.

1446
 1447

where f±(r) = 1- (r±/r)d~3 and Fd~2 = Qm Voln d _ 2 • Here, r = r+ and r = r~

correspond to an outer and an inner horizons, respectively, and Qm is proportional
to a magnetic charge. Setting a = b = c = 0 gives the background uniform black
strings. The physical property of the background solution to be noted is that the
specific heat, which controls the thermodynamical stability, is negative for small
charge (0 :s; Q < QGM) and positive for large charge (QGM < Q < M) , where
QGM is a certain critical charge. The Gubser-Mitra conjecture 6 asserts that the GL
instability exists iff black objects are (locally) thermodynamically unstable. Indeed,
we will see the GL instability does not exist for Q ;:0: QGM as a realization of the
conjecture.
Now, we expand the metric functions X(r,z) (X = a,b,c) around the uniform
solution,
00 00 00

X(r,z) = LEnXn(r)cos(nKz), Xn(r) = LE 2P X n,p(r), K = LE 2q kq,

n=O p=o q=O
where Xo,o(r) = 0 is imposed. Here K is the GL critical wavenumber and E is an
expansion parameter. Substituting these expansions into the Einstein equations, we
obtain ODEs for Xn,p(r), to be solved order by order.

3. Results and Discussion

Solving the first order perturbations, the GL critical mode for each given charge Q
is obtained. The charge dependence of the critical wavenumber is shown in Fig. 1
(a). One can see that the critical wavenumber vanishes at Q = QGM. In addition, we
find that the vanishing of the wavenumber obeys a power law near the GM point,
ko ex IQ - QGM 111 , irrespective of dimensions. The universal "critical exponent" f3 is
nearly 1/2, which resembles a second-order phase transition. It would be interesting
to investigate the universal behavior in this line.
Solving the higher-order perturbations (up to third order), we can compare the
entropy S, Helmholtz free energy F and Gibbs free energy G between Uniform and
Non-Uniform black strings in suitable ensembles. We denote the differences ofthese
thermodynamical functions by

The charge dependence of the coefficients, (a2' P2, K,2), are shown in Figs. 1 (b)-
(d) for D = 6,10,14. Here let us focus on the entropy a2. For D :s; 13, a2, being
negative initially at Q = 0, increases as the charge increases, and it becomes positive
at some critical charge Q = Qr,cr' Increasing the charge furthermore, a2 falls off
and becomes negative at a second critical charge Q = QIl,cr' For D ;:0: 14, since a2
is initially positive even at Q = 0,4 the first critical charge is absent, and only the
"second" critical charge exists. For Qr,cr < Q < QIl,cn the non-uniform string have
a larger entropy than the uniform one, and the transition would be higher order.
1448

D=6
3 • D=15

.. ··~J~'lltll~~
-5

o 0.2 0.4 0.6 0.8 o 0.2 0.4 0.6 0.8

QIQ GM QIQGM

(a) (b)
D=1O
30

-100

-30

o 0.2 0.4 0.6 0.8 o ~2 0.4 ~6 ~8

QIQGM QIQGM

(c) (d)
Fig. 1. Cal: The charge dependance of the GL critical mode. (b)-(d): The difference of thermo-
dynamical functions between uniform and non-uniform strings for D = 6, 10 and 14.

Next, let us focus on the Gibbs free energy. For 5 :s; D :s; 12, ""2 is always
positive, showing the non-uniform blanch is unstable. While, for D ~ 13, ""2 is
negative around Q = 0 and monotonically increases to be positive for Q > Qm,cr'
The sign change of (0"2, P2, ,,"2) is an interesting phenomenon. We can say that
the charge controls the stability of non-uniform strings as well as that of uniform
strings. Fully non-linear construction of non-uniform strings, including dilaton, and
the application to the gauge theory via the gauge/gravity dual will be interesting
generalization of this work.

References
1. R. Gregory and R. Laflamme, Phys. Rev. Lett. 70, 2837 (1993); G. T. Horowitz and
K. Maeda, Phys. Rev. Lett. 87, 131301 (2001); M. W. Choptuik et aI., Phys. Rev. D
68, 044001 (2003).
2. S. S. Gubser, Class. Quant. Grav. 19,4825 (2002); Troels Harmark. Phys. Rev. D 69,
104015 (2004); D. Gorbonos and B. Kol, J. High Energy Phys. 06 (2004) 053.
3. T. Wiseman, Class. Quant. Grav. 20, 1137 (2003); H. Kudoh and T. Wiseman,
Phys. Rev. Lett. 94, 161102 (2005).
4. E. Sorkin, Phys. Rev. Lett. 93, 031601 (2004); H. Kudoh and U. Miyamoto,
Class. Quant. Grav. 22, 3853 (2005).
5. U. Miyamoto and H. Kudoh, J. High Energy Phys. 12 (2006) 048.
6. S. S. Gubser and 1. Mitra, hep-th/0009126; J. High Energy Phys. 08 (2001) 018.
Analog Models of and for
General Relativity
This page intentionally left blank
FROM QUANTUM HYDRODYNAMICS TO QUANTUM GRAVITY

GRIGORY VOLOVIK
Low Temperature Laboratory, Helsinki University of Technology
P.G.Box 2200, FIN-02015 HUT, Finland
and
L.D. Landau Institute for Theoretical Physics, Moscow 119334, Russia
[email protected]·fi

We discuss some lessons from quantum hydrodynamics to quantum gravity.

1. Introduction

In the presentations at the Session 'Analog Models of and for General Relativity'
at 11 Marcel Grossmann Meeting, general relativity has been considered as emer-
gent phenomenon. General approaches to emergent relativity have been analyzed. 1
Particular example when gravity is induced in the low-energy corner of quantum
condensed matter of the proper universality class has been presented. 2 It was sug-
gested that induced metric for scalar field may lead to super luminal propagation
of scalar field and escape from the black hole without violation of Lorentz invari-
ance. 3 On the kinematic level the metric field emerges in many different systems,
and this allows us to simulate (at least theoretically) effects of relativistic quantum
field theory (QFT) in curved space. At the moment the most promising media for
simulations are Bose-Einstein condensate (BEC), where the propagation of phonons
is identical to propagations of a massless scalar field on a curved space-time. In par-
ticular, it was suggested to use the renormalization techniques developed in QFT
to study the depletion of BEC;4 in other presentation the stability of sonic horizons
in BEC 5 and the scattering problems on rotating acoustic black holes have been
discussed. 6 Effective metric appears for light propagating in non-linear dispersive
dielectric media 7 and in moving media;8 for surface waves - rip pIons - propagating
on the surface of quantum liquids or at the interface between two superfluids. 9 The
latter allows us to study experimentally the instability of the quantum vacuum in
ergoregion.
Probably our experience with superfluids and BEC will give us some hints for
solution of the fundamental problems in gravity, such as quantum gravity and grav-
itating vacuum energy. Here we shall discuss the quantum hydrodynamics of BEC
and superfluids. Both hydrodynamics and general relativity are perfect classical the-
ories. General relativity can be viewed as the theory of hydrodynamic type where
the collective variables are the metric fields giJ. v . 10 At the quantum level, quantum
hydrodynamics and quantum gravity also share many common features, e.g. both
have quadratic divergences. This is the reason why the problem of quantization of
hydrodynamics is at least 65 years old (see quantization of the macroscopic dy-
namics of liquid in the first Landau paperll on super fluidity of 4He); it is almost
as old as the problem of quantization of gravity.12 Thus the lessons from quantum

1451
1452

hydrodynamics could be useful for quantum gravity.

2. Classical hydrodynamics
2.1. Classical hydrodynamics
The first quantization scheme for hydrodynamics was suggested by Landau in 1941
when he developed the theory of superfluidity in liquid 4He. l l In his approach
Landau separated liquid 4He into two parts: the ground state (which we now call
the vacuum) and quasiparticles - excitations above the ground state (which we call
matter). Such separation into vacuum and matter is generic and is applicable to
relativistic quantum fields (RQF). The Landau approach was essentially different
from that of Tisza, 13 who suggested to separate liquid 4He into the Bose condensate
and the non-condensed atoms. Tisza's approach does make sense, especially for the
dilute Bose gases, where the condensed fraction can be easily detected. However, it is
important that the dynamics of the Bose condensate and the exchange of energy and
atoms between the condensed and non-condensed fractions, belong to high-energy
microscopic physics. On the other hand, the low-energy behavior of the superfluid
liquids and gases is governed by the Landau hyrodynamics picture. In particular,
at zero temperature both condensed and non-condensed atoms participate in the
coherent motion of the quantum vacuum with the total mass density p. This is
because at T = 0 the whole liquid is in the coherent state described by a single
many-body wave function,14 and thus the whole liquid is involved in the superfluid
motion in agreement with Landau ideas.
According to Landau, the Hamiltonian of quantum hydrodynamics is the clas-
sical energy of liquid where the classical fields, velocity v and mass density p, are
substituted by the corresponding quantum operators v and (J. So let us start with
the classical hydrodynamic energy of the liquid:

Hhydro(p, v) = J d x
3 (~pV2 + E'(p)) , E'(p) = E(p) ~ lip. (1)

Here E(p) is the energy of static liquid which only depends on p: since we consider the
vacuum of the liquid (i.e. without excitations which will appear after quatization)
it is assumed that the temperature T = o. We added here the term with Lagrangian
multiplier - the constant chemical potential JL. This term does not change the hy-
drodynamic equations, but it allows us to study thermodynamics of the liquid. For
example, the equilibrium mass density of static liquid is obtained by minimization
of the energy with taking into account the conservation of the total mass of the
liquid, which gives:
dE
- = JL , or (2)
dp
The pressure of the liquid in equilibrium at T = 0 is
p = ~ d(VE(M/V)) = ~E, (3)
dV
 1453

where M is the total mass of liquid. This suggests that the relation between the
pressure P and energy f can be considered as the equation of state for vacuum,
and this is true. Such equation of state P = -f is applicable to the ground state
(vacuum) of any system, relativistic or non-relativistic; it follows from the general
thermodynamic arguments and does not depend on the microscopic physics of the
vacuum state. It is also applicable to the vacuum of RQF. SO, further on we shall
treat the quantities Evac == f and Pvac == -f as vacuum energy density and vacuum
pressure correspondingly.
There is no satisfactory description of classical hydrodynamics in terms of La-
grangian: this requires introduction of either artificial variables or extra dimension.
The hydrodynamic equations can be obtained using the Hamiltonian formalism of
Poisson brackets. The Poisson brackets between the hydrodynamic variables are
universal, they are determined by the symmetry of the system and do not depend
on the Hamiltonian (c£.15) , that is why it is not necessary to use the microscopic
quantum theory for their derivation. For classical hydrodynamic variables v and p
one has the following Poisson brackets: 15,16

{p(rd, p(r2)} = 0 , (4)

{v(rl),p(r2)} = -\75(rl - r2) , (5)
1
{vi(rI), Vj(r2)} = --eijk(\7 x vh5(r1 - r2) . (6)
p

The same Poisson brackets are obtained from the commutation relations for the
corresponding quantum operators p and v derived by Landau 11 which follow from
microscopic physics. Using the Poisson brackets (4)-(6) and the Hamiltonian in
Eq.(l), one obtains the hydrodynamic equations:

OtP = {H,p} = -\7. (pv) (7)

dE
Ot V = {H, v} = -(v· \7)v - \7 dp . (8)

There are no fundamental parameters in classical hydrodynamics. But there

are dimensional variables which enter the classical hydrodynamics: mass density
p, and energy density E(p); the speed of sound c is c 2 = p(d 2Eldp2). In principle,
in liquids one can construct the "fundamental" parameters, the values of c = Co
and p = Po under two conditions, when the liquid is: (i) static and in equilibrium;
and (ii) at zero external pressure. These two conditions give dEl dp/ po = /10 and
P = /1oPo - E(PO) = -fi(p) = 0 correspondingly. At zero external pressure, i.e. 111
the absence of external environment, one has

Evac = -Pvac = 0 . (9)

The nullification of vacuum energy occurs for any non-disturbed equilibrium

vacuum.
1454

2.2. Vortex-free classical hydrodynamics

If one is interested in the vortex-free flow only, v = \lB, the hydrodynamic equations
for e and p can be obtained using the Lagrangian formalism. The corresponding
hydrodynamic Lagrangian is:

The constant chemical potential J.L is absorbed here by Gte.

The Poisson brackets for the vortex degrees of freedom have been discussed in
Refs. 16 ,17
In linear approximation the Lagrangian (10) describes sound waves. Sound waves
propagating over background flow of the inhomogeneous liquid can be obtained from
the hydrodynamic equations (7) and (8); the rigorous procedure can be found in
Refs. 18 ,19 As was first found by Unruh 20 the flow of liquid has the same effect on
propagation of sound waves as the metric in general relativity on propagation of
a massless relativistic particle. The effective metric for sound waves generated by
vCr, t) and per, t) is

goo = --p (c2 - v

2)
,gij
P
= -l5 ij , gOi = .
-gijV] , F9 = p2
- . (11)
c c c
This is the half of general relativity, since the effective metric obeys the hydrody-
namic equations rather than Einstein equations. However, this is enough for sim-
ulations of aspects of general relativity which do not depend on Eistein equations.
For example, effects related to behavior of quantum fields in curved space can be
reproduced. 20 ,21
The full general relativity can be generated in fermionic vacua near the Fermi
points. 2 ,22,23 Fermi point is a generic singularity in the Green's function which is
protected by topology in momentum space. Expansion near the Fermi point leads to
chiral fermions, gauge fields and gravity as effective fields in the low-energy corner.

2.3. Extended classical hydrodynamics

The most general classical hydrodynamics is obtained when one introduces correc-
tions to classical hydrodynamics by adding the gradient terms. For static liquids
and gases the important modification is the dependence of energy on the gradient
of mass density:

The other possible terms are ex (\l.v)2 and ex (\l x v)2, which we do not discuss here.
While the Hamiltonian can be extended, the Poisson brackets for hydrodynamics
variables remain intact. It was also stressed by Landau that hydrodynamic equations
are less general than the commutation relations for hydrodynamic operators.
 1455

2.4. Classical superfluid hydrodynamics

Let us introduce the quantity

(13)
which has dimension of circulation of velocity. Then the lassical superfluid hydro-
dynamics is obtained if one considers within the extended classical hydrodynamics
the class of the potential velocity fields:

v = f{\7e . (14)
In this normalization the flow potential e is dimensionless. This allows us to in-
troduce instead of P and e the classical complex field where the dimensionless e
plays the role of the phase: W = ..;pee. In terms of W the extended version of the
hydrodynamic Lagrangian in Eq.(lO) becomes:

Lcp(W) = z;
.-
(w*8t W - w8t w*) +~
-2
\7W*\7W + E(p) - IlP , P = IwI 2 . (15)

The Equation (15) is the Lagrangian of the famous Gross-Pitaevskii (GP) theory
generalized to the arbitrary function E(p). In the original Gross-Pitaevskii theory
the non-linear term is quadratic, E(p) = (1/2)gp2, and the variation of Lcp(W)
leads to the nonlinear Schrodinger equation. Note that this nonlinear Schrodinger
equation (or the more general equation obtained using the general form E(p» is
the classical equation, since the Planck constant It does not enter Eq.(15). Instead
one has the parameter", (or f{ = "'/2n) which has the dimension of circulation
of velocity [",J = [v][rJ. Circulation 1dr· v is the adiabatic invariant in classical
hydrodynamics, and thus should be quantized in quantum theory. Another invariant
in hydrodynamics is J d 3 x (v· (\7 x v)). It is also quantized in quantum theory, see
Ref.24
The superfluid hydrodynamics (SH) has three dimensional parameters (c, p and
"'), and thus the characteristic length, energy and frequency scales are now deter-
mined:
f{
aSH =- , (16)
c
The superfluid hydrodynamics is classical. The corresponding hydrodynamic
Hamiltonian is expressed in terms of the classical velocity and mass density fields
as in Eq.(12):

HcP{p, v} = !d3x(~pV2+E(P)-IlP+:;(\7P)2). (17)

However, compared to the conventional classical hydrodynamics the classical super-

fluid hydrodynamics described by Eq.(15) has three modifications:
(i) The so-called quantum pressure term (f{2 /8p) (\7 p)2 is added. In principle,
this term can be of the classical origin. This term leads to the correction to the
linear dispersion relation for sound waves: w(k) = ckJ1 + ",2k 2/4c 2.
1456

(ii) The rotational degrees of freedom are involved in this description. Since the
phase e is not single-valued, the superfluid hydrodynamics (SH) contains vortices
with quantized circulation :f dr . v = n", = 27fnft, where n is integer.
(iii) Outside the vortex cores the velocity field is potential, \7 x v = o.
The energy required to excite the vortex degrees of freedom is the energy of the
vortex loop Evr rv pft2 r of minimal size r rv aSH = ft/c in Eq.(16). Thus there is the
gap for vortex excitations of order

(18)

The advantage of Lagrangian Eq.(15) with the general function c(p) compared to
the conventional Ginzburg-Pitaevskii (GP) Lagrangian which describes superfluid
hydrodynamics in a dilute Bose condensate is as follows. In a dilute Bose gas almost
all the atoms are in the Bose condensate, the depletion - the difference between the
total density of atoms and the density of condensate is small and can be neglected
in the main approximation. As a result, the equation for the condensate practi-
cally coincides with the hydrodynamics equations. It should be mentioned that the
Ginzburg-Pitaevskii equation is not applicable to Bose condensate if the depletion
is not small, since there is no conservation law for the condensate density.
For strongly interacting liquids the depletion is not small. For example, in su-
perfluid 4He the condensate comprises only the small fraction of the total density.
Nevertheless, even in this case, the Eq.(15) remains reasonable, since the function
2
\[I is normalized to the total density: 1\[11 = p. This reflects the fact that the super-
fluid hydrodynamics describes not the dynamics of the condensate density, but the
dynamics of the whole superfluid liquid at T = o.
The Lagrangian in Eq.(15) leads to correct hydrodynamic equations and to cor-
rect energy of quantized vortex lines both in the dilute Bose gases and strongly
interacting liquids. This implies that the extended GP Lagrangian gives the rea-
sonable description of the classical hydrodynamics of superfluids at T = 0, which
includes the hydrodynamics of super fluid component at T = 0 and the classical
dynamics of vortices with quantized circulation. The normal component made of
quanta of sound waves - phonons - is absent in this approach. It is included at the
stage of quantization to obtain the two fluid hydrodynamics at T # O. The draw-
back of this description is that as distinct from the GP equation for the dilute Bose
gases, the general Lagrangian in Eq.(15) gives only the model description of the
vortex core region; however, in many cases such model is sufficient since it allows us
to consider the core effects consistently without ambiguous cut-off procedure. The
further extension of the model with incorporation of the non-local interaction can
be found in Ref. 25
In conclusion, the model (17) simulates superfluid hydrodynamics not only in
weakly interacting Bose gas, but also in real quantum liquids, in which the Bose
condensate is either absent or is a small fraction of the total density. It is also im-
portant, that as distinct from the Bose gas, liquids can be stable even in the absence
 1457

of environment, i.e. at zero external pressure. This is important for the consider-
ation of the problems of vacuum energy and the related problems of cosmological
constant 26 ,27 using the ground state of an isolated quantum liquid as the physical
example of the quantum vacuum in which the nullification of the vacuum energy in
equilibrium occurs without any fine tuning. 28

3. Quantum hydrodynamics
3.1. Landau quantum hydrodynamics
Landau introduced quantum Hamiltonian expressing the cla8sical energy in Eq. (1)
it in terms of the corresponding non-commuting quantum operators V and p:

HAhydro (A
p,VA) = J (1d3. X AAA
2VPV+f (PA) -!lP') . (19)

The commutation relations for the components of velocity field operator v, and
between v and (J are

[(J(rd, (J(r2)] =0, (20)

[v(rI), (J(r2)] = ~\75(rl - r2) , (21 )
~

[vi(rd, Vj(r2)] = ~eijd\7 x vh5(rl -

zp
r2) , (22)

have been derived by Landau from the microscopics. They can also be obtained
from the Poisson brackets (4)-(6) for the classical variables.
Quantum hydrodynamics i8 characterized by three dimensional quantities. In
addition to equilibrium values of p and c, the really fundamental Planck constant
Ii enters the quantum hydrodynamics through the commutation relations (21) and
(22).
Using three dimensional quantities one can construct the characteristic 'Planck'
scales for the energy EQH, mass l'vIQH , length UQH, frequency WQH and energy
density EQH:

fQH '" E(p) '" pc 2 .

(23)

3.2. Rotational modes

Landau 8uggested that the only low frequency modes of quantum hydrodynamics
are quanta of sound waves - phonons, while the rotational modes (vortex degrees of
freedom) are separated by the gap. One may suggest that if the gap exists in quan-
tum hydrodynamics, it is given by the characteristic energy scale EQH in Eq.(23).
However, there are some arguments against that. Since the operators of vorticity
1458

\7 x V and density p are commuted, the Hamiltonian which governs the rotational
degrees of freedom is

Htransverse(v J..) = ~ Jd xp
3
vl , (24)

where vJ.. is the transverse (non-potential) part of velocity field. The above vortex
contribution to quantum hydrodynamics contains only two parameters: nand p.
Using these two quantities only one cannot construct the quantity with the dimen-
sion of the energy gap: the "Planck" energy scale EQH of quantum hydrodynamics
in Eq,(23) contains c which is irrelevant for transverse degrees of freedom.
This was probably the reason why Landau proposed different estimate for the
rotational gap which did not contain c, but contained the mass m of 4He atomY
n2 p2/3
6.L = m 5/ 3 ' (25)
The atomic mass m is the microscopic parameter, which is beyond the quantum
hydrodynamics. Incidentally or not, but since in superfiuid 4He the atomic mass m
and the quantum hydrodynamic mass MQH in Eq.(23) are of the same order, the
Landau estimation in Eq.(25) coincides with the estimation for the energy of the
elementary vortex excitation in superfiuid 4He - the smallest possible vortex ring -
in Eq.(18).
That quantum hydrodynamics alone cannot describe the superfiuid liquid has
been later emphasized by Feynman. 14 The main reason is that the classical hydrody-
namics lacks (has lost) the information on the important microscopic properties of
the underlying system, such as quantum statistics of atoms. It is the Bose statistics
of atoms which leeds to the gap in the spectrum of quantum vorticity.14 However,
such gap is not present in Fermi liquids (unless the Cooper pairing occurs). More-
over, Fermi liquids are not described by classical or quantum hydrodynamics.
All this demonstrates that in order to describe the real systems, the quantum
hydrodynamics requires the extension. As the starting point for quantization one can
choose the extended classical hydrodynamics discussed in Sec. 2.4. In this approach
the vortex degrees of freedom would have the gap already at the classical level (see
Eq.(18)). This will be discussed later in Sec. 3.4.

3.3. Quantization of phonon field

If for some reasons the rotational degrees of freedom are separated by the gap, then
the only low-energy degrees of freedom are represented by the vortex-free hydro-
dynamics and sound waves. In linear regime (and in the absence of the rotational
degrees of freedom) the Landau quantum hydrodynamics leads to quantization of
sound waves. Quanta of sound waves are phonons with linear spectrum Ek = nck.
The nonlinear terms in quantum hydrodynamic Hamiltonian describe interac-
tion of phonon fields, and lead to modification of phonon spectrum at large k. One
may expect that the linear dispersion of phonon spectrum (analog of Lorentz in-
variance) is violated at the Planck scale kQH = l/aQH in Eq.(23), provided that no
 1459

microscopic physics intervenes earlier. In principle, the correction to the spectrum of

phonons can be computed within the quantum hydrodynamics, however the diverg-
ing Feynman diagrams makes this procedure rather ambiguous. In the literature,
people used the inverse inter-atomic distance ka cv 11a as the natural ultraviolet
cut-off for diverging diagrams (see e.g. 29 ). However, such parameter characterizes
the microscopic physics beyond the quantum hydrodynamics. Whether it is possible
to make regularization in such a way that the natural cut-off is determined by the
quantum hydrodynamics itself, i.e. by kQH = l/aQH' is an open question.
If such regularization procedure exists, the first guess would be that the spectrum
is modified by the next order term:

(26)
The quantum hydrodynamic correction in the form of Eq.(26) is naturally obtained
in the 1+1 case, when p is the one-dimensional mass density and one has a~H =
lil(pc). In case of the 3+1 quantum hydrodynamics, where according to Eq.(23) one
has atH = iii (pc) , the correction in Eq.(26) is non-analytic in Ii, and the proper
correction would be the higher order term which is linear in Ii:

w
2 2
= c k
2
(1 + latHk4 In (at~k4) + ...) , III cv 1. (27)

The above guess is supported by the temperature corrections to the spectrum of

phonons 3o after substitution T cv lick.
The density of zero point energy of quantum phonon field can be estimated using
the Planck energy cut-off kQH = 1I aQH:

Ezp = -21 J- (
3
d k
2rr
)3 lick cv
4
lick QH
EQH
cv - 3-
a QH
cv
2
pc . (28)

This gives the correct estimation of at least the magnitude of the energy density of
the liquid (the sign of E(p) - /-1P is negative if the external pressure is positive). Note
that the result is classical, i.e. it does not depend on Ii. This is not very surprising
because the energy density constructed from Ii, c and p does not contain Ii (see
Eq.(23)).
From the modern point of view, the classical hydrodynamics as well as classical
gravity, is the classical output of the quantum system in the low-energy corner.
The 'initial classical' energy density E(p) is not only the starting point for 'quantum
hydrodynamics' but also is the final classical macroscopic result: it contains all the
quantum contributions to the energy density of the liquid. This also means that
the contribution of zero point energy of phonons to vacuum energy has already
been included from the very beginning and should not be counted again. Thus
the phonon Hamiltonian in the quadratic approximation must be written without
zero-point energy of phonons:

(29)
1460

where at and ak are operators of creation and annihilation of phonons, and V is

the volume of liquid.

3.4. Quantum superfluid hydrodynamics

There are several ways of quantization of superfiuid hydrodynamics: (i) One can
perform the full Landau quantization of Eq.(12), expressing the extended hydrody-
namic Hamiltonian in terms of the quantum fields P and v. (ii) One can perform
quantization starting with the superfluid hydrodynamics with quantized vortices de-
scribed by the Lagrangian in Eq.(15) by expressing it in terms of the non-commuting
fields Wand wt. Let us consider the last case. The quantum counterpart of classical
Lagrangian Eq.(15) is the Hamiltonian

HGP(W) = J d3x (~2 VWtVW + E(p) - ILP) , P= WtW , (30)

which is supplemented by commutation relations for quantum fields

(31)

If one identifies the parameter filF;, with the mass of an atom of the liquid, one
obtains that the extended quantum hydrodynamics is nothing but the microscopic
quantum mechanics of a system of identical bosonic atoms with mass m = filF;, and
with a special type of interaction term E(p) which only depends on density.
The superfluid quantum hydrodynamics (SQH) contains four parameters fi, m,
speed of sound c, and equilibrium density p. One can introduce the dimensionless
mass parameter mSQH:
m
mSQH = JI;[QH (32)

One may suggest that this dimensionless parameter characterizes microscopically

different systems, which have the common macroscopic (low-energy, hydrodynamic)
properties. In dilute Bose gases one has mSQH « 1, while in superfiuid liquid 4He
and in superfluid liquid 3He this parameter is of order unity, mSQH '" 1.
However, if one compares the quantum hydrodynamic Hamiltonian (30) with
the Hamiltonian of exact microscopic theory

(33)

one finds that the difference in the interaction term is enormous. In other words,
the prescribed down-up route from classical to quantum theory (see Fig. 1) does
not lead in general to the true microscopic theory.
And this is not the only drawback of quantum hydrodynamics. One may suggest
that inspite of disagreement with exact microscopic theory, the 'microscopic' Hamil-
tonian in Eq.(30) may serve as a relevant microscopic model. In principle, starting
 1461

with this Hamiltonian, one may obtain in the long-wave limit (i.e. in the up-down
route in Fig. 1) the classical hydrodynamic Hamiltonian for super fluid liquid state.
However, the emerging function I' will essentially deviate from I' in the original
classical hydrodynamics, i.e. 1'2 (p) =1= 1'1 (p). Moreover this function E(p) cannot be
expressed in terms of the renormalized coupling g. That is why this procedure ~
down- up (quantization), up-down (emergence of effective theory in the low-energy
corner of quantum theory), down-up, etc. in Fig. 1 ~ in general does not converge.

Quantum Theory Quantum Theory

Classical
/ \ Classical
/ \
Hydrodynamics Hydrodynamics

t1 (p) t2(P)
Fig. 1. From classical to quantum hydrodynamics (quantization) and back (to low-energy limit
of quantum system).

3.5. Weak coupling limit

The only case in which the exact theory and extended quantum hydrodynamics fit
each other is when the energy density is quadratic function of p:

(34)

and the parameter mSQH in Eq.(32) is small: mSQH « 1. This corresponds to small
coupling g
g _ 8/3 n2 a n2 (35)
go - m SQH «1 , go = m3 = p1/3 m 8/3

where a = (p/m)~1/3 is the interatomic distance. In liquid 4He one has m rv MQH,
and thus quantization of the hydrodynamics does not make sense.
The limit of small coupling g corresponds to the model of weakly interacting Bose
gas, which has been solved by Bogoliubov. When one starts with the microscopic
theory

(36)
1462

with small g, one obtains in the long-wave-length limit (i.e. on the up-down route in
Fig. 1) the classical hydrodynamics, where in the main approximation the function
E(p) = (g/2)p2 coincides with that in microscopic theory. In the next approximation
the function E(p) is modified by the "quantum" correction, as follows from the
Bogoliubov theory:

E(p) = flp2 1 + __
16 ( !L )3 /2) , g« go . (37)
2 ( 157f 2 go

This means that after the first iteration (down-up and up-down in Fig. 1) the
coupling constant is renormalized:

/
g=
16
9 ( 1 + 157f2
(
:0 )3 2) , g« go . (38)

There is a temptation to consider the correction to E(p) = (g/2)p2 in Eq.(37) as

the back reaction of the quantum vacuum to quantum fields of phonons. 31 At first
glance, one may identify this correction with the properly regularized zero point
energy of phonon field:

(39)

(40)

However, such interpretation is only valid at small 9 when the microscopic Bogoli-
ubov theory provides three counter-terms in Eq.(39).
Moreover, this correction contains the Planck constant h in the denominator
(since F;; = him). This means that the weakly interacting Bose gas (the system with
small g) actually corresponds to the ultra-quantum limit, in which the contribu-
tion of zero point motion of phonon field is small compared to the main quantum
contribution (1/2)gp2 to the vacuum energy.

3.5.1. Energy Scales

Because of the dimensionless quantity g/gO (or m/MQH) , the extended quantum
hydrodynamics in Eq. (36) contains different physically interesting scales for each
dimensional quantity. In addition to the hydrodynamic energy scale EQH = MQHC 2
in Eq.(23), there is the scale EL = mc 2 « E QH , where the Lorentz violation occurs
in a dilute Bose condensate.
Another energy scale was introduced by Landau l l in Eq.(25). This corresponds
to the energy of the smallest vortex ring in superfiuid 4He: Evr '" pF;; 2a '" /::"'L, where
a is the interatomic spacing (p '" m/ a 3 ) which determines the smallest possible
radius of the vortex ring in superfiuid 4He.
 1463

There are several important length scales in dilute Bose gases in addition to the
hydrodynamic length aH: interatomic spacing a and the coherence length ~ = fi/mc:

aH go 1/8 <"t ( go ) 1/2

-:=
(
-
)
1-==- (41)
a g ag

4. Back reaction of quantum vacuum

4.1. Depletion of condensate
The term 'depletion' means that because of interaction (i.e. for g # 0) the non-
vanishing number of atoms is not in the Bose-condensate. For a weakly interaction
dilute Bose gas, the atoms in the condensate are prevailing. The relative values of
the condensate and non-condensate mass densities are
3/2 3/2
Pcond = 1 _ _
1_ JL , Pnon-cond = _1_ JL . (42)
P 37[2 ( go ) P 37[2 ( go )

In strongly interacting 4He liquid the fraction of the non-condensate atoms is pre-
vailing, with Pcond/ P < 0.1. Nevetheless, at T = 0 the whole liquid is in the coherent
superfluid state - the quantum vacuum - with the superfluid component density
Ps = p. The same occurs in two dimensional systems, where the condensate is
completely depleted even in the absence of interaction, Pcond = O.
The depletion of the Bose-condensate is not in the framework of Landau quan-
tum hydrodynamics. It is in the framework of the Tisza description of superfluids
and is fully microscopic phenomenon, which is beyond the low-energy hydrodynam-
ics. Let us stress again that the Landau description in terms of vacuum and matter
(quasiparticles) is applicable for superfluids in the low-energy regime. In this regime
the hydrodynamics with its Euler and continuity equations has no information on
the separation of the liquid into the Bose condensate and atoms above the con-
densate caused by interaction, since at T = 0 both these fractions participate in a
single coherent flow of the quantum vacuum. The Tisza picture of condensed and
non-condensed fractions requires the microscopic description of the particle and en-
ergy exchange between the two fractions; this is the high-energy phenomenon which
is certainly beyond the responsibility of hydrodynamics.
In general, the depletion of the Bose-condensate is also beyond the quantum su-
perfluid hydrodynamics, except for the limit g « go, where the superfluid quantum
hydrodynamics coincides with the microscopic Bogoliubov model, and the depletion
can be studied using perturbation theory. This is the reason why the calcuations
of the depletion using the quantum fluctuations of phonon field (see e.g. Ref. 4) or
other back reaction effects (see e.g. Ref. 31 ) cannot be considered as generic.
However, there are some problems which are within the responsibility of Landau
quantum hydrodynamics. One of them is the depletion of the mass density caused
by phonons. This is the back reaction of quanta of sound waves onto the 'classical'
quantum vacuum (let us stress again that in the low-energy limit the superfluid
quantum vacuum behaves as classical liquid).
1464

4.2. Back reaction of vacuum density to quantum matter

At non-zero temperature the liquid consists of the vacuum (the ground state) with
density P and excitations (quanta of sound waves - phonons) in Eq.(29). Lets us
find how thermal phonons modify the mass density P of the quantum vacuum. This
is the back reaction of the vacuum to the quanta of sound waves. We assume that
temperature is small, T « EQH' so that only low-frequency phonons with linear
spectrum w = ek contribute to the thermal energy, and consider fixed external
pressure. The correction can be obtained by minimization of the free energy density
of the liquid F = E - T S over p. The free energy is the sum of the energy of ground
state (quantum vacuum) and the free energy of the phonon gas (matter). For the
phonons with linear dispersion relation the free energy density is the radiation
pressure with minus sign:

(43)

where Emat is the energy density of the gas of thermal phonons (radiation energy).
Since the vacuum does not contribute to the entropy of the system, the total
free energy density of a liquid is
1
F(T,p) = E(p) - J-lP - 3"E mat(P) . (44)

Let Po be the equilibrium density at T = 0 and J-l = J-lo, then considering Emat
as perturbation one obtains the following expansion in terms of bp = P - Po and
bJ-l = J-l- J-lo:
182Evac )2 1 8Emat
F ( T, P) = F ( T, Po ) +"2 8p2 (bp - 3" ----apb p - bJ-lbp . (45)

For phonon gas the dependence of the radiation energy on p in Eq.(43) only comes
from the speed of sound,

- - - - 3-
8Emat _
- - - - 3 uEmat
Emat 8e _
-- (46)
8p e 8p p
Here we introduced the function u
8lne
u = 8lnp , (47)

which is the fluid-state analogue of Griineisen parameter, see e.g. 30

Then we must take into account that the chemical potential J-l must be changed
to support the fixed external pressure. The total change of the pressure of the liquid,
which is the sum of the vacuum pressure of the liquid and the radiation pressure of
phonons, must be zero, bPvac + P mat = O. This gives
1
bPvac = -Pmat = -3"E mat , ( 48)

As a result the change in the chemical potential is

bJ-l = bPvac = _~ Emat (49)
Po 3 Po
 1465

Introducing Eqs. (46) and (49) into free energy (45) and minimizing over 6pone
obtains the response of the density of the liquid to the phonon gas:

6; = _ f;~t (~ + u) . (50)

The result in Eq.(50) can be also obtained from the analysis of classical hy-
drodynamic equations made by Stone in Refs. 18 ,19 The second term on the rhs of
Eq.(50) comes from the second order correction to the density of the liquid induced
by the sound wave. This is the Eq.(4.13) of Ref. 18 integrated over thermal quanta
of sound waves -- phonons. The first term in the rhs of Eq.(50), which is due to the
change in the vacuum pressure, can be also obtained using Stone's formalism.
Note that the depletion of liquid density 6p ex: T 4 , while the temperature cor-
rection to the depletion of the condensate is ex: T2 (see e.g. 4 ) . The reason for such
difference is that the density p is conserved quantity, while the condensate density is
not because of the Josephson coupling between the condensate and non-condensate
atoms. In conclusion, the depletion of the mass density is universal and is completely
determined by hydrodynamics, while the depletion of the condensate is beyond the
quantum hydrodynamics and strongly depends on the microscopic physics.

4.3. Response of dark (vacuum) energy to matter

Let us consider the back reaction of vacuum energy to thermal phonons. According
to Eq. (3) the analog of the vacuum energy density in liquids is Evac = E = E(p) - fLP.
It obeys the correct equation of state for quantum vacuum

Pvac = fLP - E(p) = -E vac . (51)

The correction to the 'vacuum energy' density due to thermal phonons is

6fvac = 6(E(p) - fLP) = (~; - fLO) 6p - Po6fL = -Po6fL = ~Emat , (52)

where we used Eq.(49) for 6fL.

Let us consider an equilibrium liquid in the absence of environment, i.e. when
the external pressure is zero. Then in the absence of phonons the vacuum energy
and pressure are zero according to Eq.(9), Pvac = -fvac = O. At T "I 0, thermal
phonons produce radiation pressure which must be compensated by the pressure of
the vacuum. As a result the vacuum energy density becomes non-zero:

(53)

This is the back reaction of the vacuum to relativistic matter. The same relation
between the dark energy and hot matter is applicable for such Universes in which
gravity is absent, i.e. in which the Newton constant G = 0 (see Refs. 2 ,32). Note that
in liquids, where the effective gravity obeys hydrodynamic equations rather than
Einstein general relativity, the vacuum energy is naturally of the order of matter
density. For Universes with gravity, situation is more compicated, since the vacuum
1466

energy responds also to gravitating matter, curvature, expansion and other pertur-
bations of the vacuum state. However, the main result is that the vacuum energy is
naturally determined by macroscopic quantities, rather than by huge microscopic
Planck energy scale. 28

5. Lessons for quantum gravity

5.1. From quantum gravity to quantum hydrodynamics
The results in Eqs.(50) and (52) for the back reaction of the vacuum are expressed
completetely in terms of quantum hydrodynamics, i.e. in terms of the function
c(p) and Planck constant h. These results are generic and do not depend on the
microscopic physics, so that the extension to quantum superfiuid hydrodynamcs
(with its extra parameter K, = him) is not required. The only role ofthe microscopic
physics is to supply us with the macroscopic function c(p).
In general relativity, there are also examples of the universal behavior of the back
reaction, such as universal temperature corrections to Einstein equations and to
Newton constant G. The temperature correction to the free energy of gravitational
field induced by N F massless fermionic quantum fields and Ns scalar quantum fields
is 33

(54)

Here R is the Ricci curvature of gravitational field and w 2 - w!-'w!-, , where

w!-' = ~8!-, In goo is 4-acceleration. This result also does not depend on the micro-
scopic Planck physics. It is expressed in terms of the Planck constant h and integral
numbers - numbers of species N F and Ns (actually one should also add contribu-
tion of the vector fields and gravitons). Thus the only role of the microscopic Planck
physics is to supply us with the definite number of fermionic and bosonic quantum
fields in the low energy corner.
Moreover, the temperature correction to the gravitational action in Eq.(54) is
applicable not only to general relativity but also to the effective gravity emerging in
quantum liquids. 34 In quantum liquids, the dominating contribution to the 'gravi-
tational action' is provided by hydrodynamics, while the subdominant corrections
are within responsibility of the QFT in curved space. For superfiuid 4He and for
Bose condensate of single atomic species the microscopic physics gives us N F = 0
and Ns = 1. Expressing R, g and w in Eq.(54) in terms of the effective metric
g!-'v experienced by phonon field in Eq.(ll), one obtains the correct subdominant
contribution to the hydrodynamic free energy of the liquid 4He or Bose gas. In
case of effective gravity in superfiuid 3He-A with gapless fermions, the microscopic
physics gives us N F = 2 and Ns = 0, and using Eq.(54) one obtains the correct
subdominant contribution to the gradient energy. These are examples when general
relativity helps us to solve some problems in superfiuids.
Another example is provided by the universal quantum correction to Newton law
(see e.g. 35 ). It has exact analog in quantum hydrodynamics and gives rise to the uni-
 1467

versal quantum correction to the classical hydrodynamic action caused by effective

QFT in effective curved space of acoustic metric (see Refs. 36 ). As an illustration let
us write one of the typical terms generated by the quantum hydrodynamics - the
contribution to the quantum pressure caused by quantum fluctuations of phonon
field in effective curved acoustic space obtained by Seeley-De Witt expansion: 36

(55)

This leads to the quantum correction to the spectrum of phonons in Eq.(27) which
is proportional to n. The infra-red (IR) logarithmic divergence of the quantum hy-
drodynamic corrections suggests that they may describe the creation of phonons
(matter) by the time dependent flow (gravitational field) in exact analogy with
particle production in gravitational field (see e.g. Ref. 37). In a similar way, in su-
perfluid 3He-A the logarithmically divergent action for the effective electromagnetic
field leads to the Schwinger-type production of fermionic quasiparticles by the time-
dependent order parameter. 38
The quantum pressure produced by quantum fluctuations of phonon field in
effective curved acoustic space-time is more pronounced in the 1+ 1 quantum hy-
drodynamics. The effect is related to the gravitational trace anomaly in the 1+ 1
space-time,39 and leads to the quantum correction to the phonon spectrum which
is also proportional to n: the factor in Eq.(26) is l'a~H = -n/(4Sn:pc).
The Hawking radiation also does not distinguish between gravity obeying the
general relativity and effective gravity in liquids obeying hydrodynamic equa-
tions. 9,2o In both cases, Hawking radiation from an astronomical or acoustic black
hole is described as the process of semi-classical tunneling between (quasi)particle
trajectories inside and outside the horizon. 40 ,41

5.2. From quantum hydrodynamics to quantum gravity

We considered some cases when the quantum hydrodynamics and quantum gravity
allow us to obtain the true corrections to hydrodynamics or/and to general relativ-
ity. There are some other examples of such kind, when the quantum hydrodynamics
and quantum gravity work. However, it is not the general case. Quantum hydro-
dynamics and quantum gravity reproduce only those (mostly subdominant) terms
in the action or in free energy which do not contain dimensional parameters, such
as Eqs. (54) and (55). In general, the down-up route from classical to quantum hy-
drodynamics (see Fig. 1) leads to the theory which does not coincide with the true
microscopic theory. This reflects the main property of the emegent physics: there
are only very few up-down ways, i.e. from the high energy microscopic theory to
the low-energy macroscopic hydrodynamic theory. The way depends on the univer-
sality class and is unique for given universality class. But there are infinitely many
down-up routes from macroscopics to microscopics. This is the main message for
those who would like to quantize gravity and hydrodynamics.
1468

One can quantize sound waves in hydrodynamics to obtain quanta of sound

waves - phonons.ll Similarly one can quantize gravitational waves in general rela-
tivity to obtain gravitons. 12 But one should not use the low-energy quantization for
calculation of the radiative corrections which contain Feynman diagrams with inte-
gration over high momenta. In particular, the effective field theory is not appropriate
for calculations of the vacuum energy in terms of the zero-point energy of quantum
fields. Such attempts lead to the cosmological constant problem in gravity,26,27 and
to the similar paradox for the vacuum energy in quantum hydrodynamics: in both
cases the vacuum energy estimated using the effective theory is by many orders
of magnitude too big. We know how this paradox is solved in quantum liquids,28
and we may expect that the same general arguments based on the thermodynamic
stability of the ground state of the quantum liquid are applicable to the vacuum of
relativistic quantum fields.
Another hint from hydrodynamics is that the underlying microscopic theory of
quantum gravity must contain additional parameter to n, c and C. Then one has the
dimensionless parameter, which distinguishes between different microscopic theories
with the same macroscopic phenomenology. Example of such parameter in quantum
hydrodynamics is rnSQH in Eq.(32). It appears that properly formulated quantum
hydrodynamics makes sense only in the limit when this parameter is small, i.e. for
the case of dilute Bose gases. The necessity of the small parameter for the emergent
general relativity and/or gauge fields is emphasized by Bjorken: 'the emergence can
only work if there is an extremely small expansion parameter in the game'.42 The
role of the small parameter could be played by the ratio EPlanck/ ELorentz between
the Planck energy scale and the energy scale above which the Lorentz invariance is
violated (see e.g. discussion in Ref. 43 ).
As follows from the experience with different quantum condensed matter sys-
tems, the metric field g{tv may naturally emerge in the low-energy corner of quantum
vacuum. It is important that in some systems gravity emerges as effective geom-
etry, rather than the spin-2 field. Even in such caricature gravity as the effective
gravity for sound waves propagating in inhomogeneous moving liquids, the acoustic
metric g{tv in Eq.(ll) is the emerging geometrical object, which has nothing to do
with the spin-2 field. Depending on the hierarchy of parameters of the underlying
microscopic system (quantum vacuum), the geometry (metric field) may obey the
nonlinear hydrodynamic equations, or the nonlinear equations of general relativity,
or Gross-Pitaevskii equations, etc.
In some vacua gravity emerges together with all the ingredients of Standard
Model: relativistic chiral fermions and quantum gauge fields. This is the general low-
energy property of vacua with the so called Fermi point in momentum space,2,22,23
which demonstrates that gravity is the natural part of physics, and it should not be
separated from the other fermionic and bosonic classical and quantum fields. The
separation only occurs at low energy, because of the difference between the running
couplings for gauge fields and gravity. This means that if gravity is the energent
phenomenon, it should naturally emerge together and simultaneously with the other
 1469

physical fields and physical laws. This is the main requirement for the future theory
of quantum gravity.

Acknowledgments
This work was supported in part by the Russian Ministry of Education and Science,
through the Leading Scientific School grant #2338.2003.2, by ESF COSLAB Pro-
gramme and by the Russian Foundation for Basic Research (grant 06-02-16002-a).

References
1. O. Dreyer, Emergent General Relativity, gr-qc/0604075.
2. G. E. Volovik, The Universe in a Helium Droplet (Clarendon Press, Oxford, 2003).
3. E. Babichev, V. Mukhanov and A. Vikman, Escaping from the black hole? JHEP
0609, 061 (2006); hep-th/0604075.
4. R Balbinot, S. Fagnocchi and A. Fabbri, The depletion in Bose Einstein condensates
using Quantum Field Theory in curved space, cond-mat/0610367.
5. C. Barcelo, A. Cano, L.J. Garay, G. Jannes, Stability analysis of sonic horizons in
Bose-Einstein condensates, Phys. Rev. D74, 024008, (2006).
6. C. Cherubini, F. Federici, S. Succi, M. P. Tosi, Excised acoustic black holes: the
scattering problem in the time domain, Phys.Rev. D72 084016 (2005).
7. V.A. De Lorenci, R Klippert, and D.H. Teodoro, Birefringence in nonlinear
anisotropic dielectric media, Phys. Rev. D 70, 124035 (2004).
8. K. Rosquist, A moving medium simulation of Schwarzschild black hole optics, Gen.
Rei. Gmv. 36, 1977-1982 (2004); gr-qc/0309104.
9. G.E. Volovik, Horizons and ergoregions in superfluids, J. Low Temp. Phys. 145, 337-
356 (2006); gr-qc/0603093.
10. B. L. Hu, New View on Quantum Gravity and the Origin of the Universe, gr-
qc/0611058.
11. L.D. Landau, Theory of superfluidity of helium-II, J. Phys. USSR, 5, 71 (1941).
12. M. Bronstein, Quantentheorie schwacher Gravitationsfelder, Phys. Ztschr. der Sowje-
tuion, 9, 140-157 (1936); see also G. Gorelik, Matvei Bronstein and quantum gravity:
70th anniversary of the unsolved problem, Physics-Uspekhi 48, 1039-1053 (2005).
13. L. Tisza, J. Phys. Radium 1, 164 (1940); ibid. 1, 350 (1940).
14. RP. Feynman, Statistical Mechanics (Benjamin, Massachusetts, 1972).
15. I.E. Dzyaloshinskii and G.E. Volovick, Poisson brackets in condensed matter, Ann.
Phys. 125, 67-97 (1980);
16. G.E. Volovik and V.S. Dotsenko, Poisson brackets and continual dynamics of the
vortex lattice in rotating Hen, JETP Lett. 29, 576 - 579 (1979); G.E. Volovik, Poisson
brackets scheme for vortex dynamics in superfluids and superconductors and effect of
band structure of crystal, JETP Lett. 64 ,845-852 (1996); cond-mat/9610157.
17. M. Rasetti and T. Regge, Physica A80, 217 (1975).
18. M. Stone, Acoustic energy and momentum in a moving medium, Phys. Rev. E62,
1341-1350 (2000).
19. M. Stone, Phonons and forces: momentum versus pseudomomentum in moving fluids,
in: Artificial Black Holes, eds. M. Novello, M. Visser and G. Volovik (World Scientific,
2002), pp. 335-364.
20. W.G. Unruh, Experimental black-hole evaporation? Phys. Rev. Lett. 46, 1351-1354
(1981); Sonic analogue of black holes and the effects of high frequencies on black hole
evaporation, Phys. Rev. D51, 2827-2838 (1995).
1470

21. C. Barcelo, S. Liberati and M. Visser, Analogue Gravity, Living Rev. Re!. 8, 12 (2005);
gr-qc/0505065.
22. C.D. Froggatt and H.B. Nielsen, Origin of Symmetry (World Scientific, Singapore,
1991).
23. P. Horava, Stability of Fermi surfaces and K theory, Phys. Rev. Lett. 95, 016405
(2005).
24. G.E. Volovik and V.P. Mineev, Investigation of singularities in superfluid 3He and
liquid crystals by homotopic topology methods, JETP 45 1186 - 1196 (1977); Particle
like solitons in superfluid 3He phases, JETP 46, 401-404 (1977).
25. N.G. Berloff and P.H. Roberts, Motions in a bose condensate: VI. Vortices in a nonlocal
model, J. Phys. A: Math. Gen. 32, 5611-5625 (1999).
26. S. Weinberg, The cosmological constant problem, Rev. Mod. Phys. 61, 1 (1989).
27. T. Padmanabhan, Cosmological constant - the weight of the vacuum, Phys. Rept. 380,
235-320 (2003).
28. G.E. Volovik, Vacuum Energy: Myths and Reality, prepared for the special issue of
Int. J. Mod. Phys. devoted to dark energy and dark matter; gr-qc/0604062.
29. S. Eckstein and B.B. Varga, Dispersion of phonons in 4He, Phys. Rev. Lett. 21, 1311-
1314 (1968).
30. I.M. Khalatnikov, An Introduction to the Theory of Superfluidity, (Benjamin, New
York, 1965).
31. U.R. Fischer, Dynamical aspects of analogue gravity: The backreaction of quantum
fluctuations in dilute Bose-Einstein condensates, cond-mat/0512537.
32. G.E. Volovik, Vacuum energy and Universe in special relativity, JETP Lett. 77, 639-
641 (2003); gr-qc/0304103.
33. Yu.V. Gusev and A.I. Zelnikov, Finite temperature non local effective action for quan-
tum fields in curved space, Phys. Rev. D59, 024002 (1998).
34. G.E. Volovik and A.I. Zelnikov, Universal temperature corrections to the free energy
for the gravitational field, JETP Lett. 78, 751-756 (2003); gr-qc/0309066.
35. G. Kirilin and I. Khriplovich, Quantum power correction to the Newton law, JETP
95,981-986 (2002).
36. K.N. Ilinski and A.S. Stepanenko, From Bose condensation to quantum gravity and
back, J. Phys. Stud. 2, 155-159 (1998); cond-mat/9803233; Hydrodynamics of a Bose
condensate: beyond the mean field approximation (II), cond-mat/9612117; First quan-
tum corrections for a hydrodynamics of a nonideal Bose gas, cond-mat/9607202.
37. A. Dobado and A.L. Maroto, Particle production from nonlocal gravitational effective
action, Phys. Rev. D60, 104045 (1999).
38. G.E. Volovik, Exotic Properties of Superfluid 3He" (World Scientific, Singapore, 1992),
Chapter 6.
39. R. Balbinot, S. Fagnocchi and A. Fabbri, Quantum effects in acoustic black holes: The
backreaction, Phys. Rev. D71, 064019 (2005).
40. G. E. Volovik, Simulation of Painleve-Gullstrand black hole in thin 3He-A film, JETP
Lett. 69, 705-713 (1999); gr-qc/9901077.
41. M.K. Parikh and F. Wilczek, Hawking radiation as tunneling, Phys. Rev. Lett. 85,
5042-5045 (2000).
42. J.D. Bjorken, private communications; J.D. Bjorken, Cosmology and the standard
model, Phys. Rev. D67, 043508 (2003).
43. F.R. Klinkhamer and G.E. Volovik, Merging gauge coupling constants without Grand
Unification, JETP Lett. 81, 551-555(2005); hep-ph/0505033.
 LOOKING BEYOND THE HORIZON

EUGENY BABICHEV
INFN - Laboratori Nazionali del Gran Sasso, S.S. 17bis, 67010 Assergi (L'Aquila), Italy
and Institute for Nuclear Research of the Russian Academy of Sciences, 60th October
Anniversary Prospect 7a, 117312 Moscow, Russia
[email protected]

VIATCHESLAV MUKHANOV
Arnold-Sommerfeld-Center for Theoretical Physics, Department fur Physik,
Ludwig-Maximilians-Universitiit Munchen, Theresienstr. 37, D-B0333, Munich, Germany
and Perimeter Institute for Theoretical Physics, 31 Caroline Street North Waterloo, Ontario,
N2L 2Y5, Canada
[email protected]

ALEXANDER VIKMAN
Arnold-Sommerfeld-Center for Theoretical Physics, Department fur Physik,
Ludwig-Maximilians- Universitiit Munchen, Theresienstr. 37, D-B0333, Munich, Germany
[email protected]

In our previous work 1 we have shown that it is principally possible to send infor-
mation from the interior of a black hole (BH) without modifying Hilbert-Einstein
action a. This may happen provided a special type of scalar field theory with non-
canonical kinetic term is realized in nature. In our approach we have considered the
accretion of a particular noncanonical field ¢ onto a black hole b. The scalar field
flow forms an acoustic black hole similar to the well known "dumb hole". 12 This
background dynamically breaks Lorentz invariance and serves as a "new ether" . Due
to the nonlinearity of the equation of motion, the" new ether" allows a superluminal
propagation of small perturbations 8¢ of the field ¢ c. The field theories with non-
standard kinetic terms have been subjects of investigation since a long time ago. 4
In cosmology they were first introduced in the context of k-inflation and k-essence
models. 5 The letter seems to require a super luminal sound speed during a period of
cosmic evolution. 7 The models with superluminal sound speed may have other in-
teresting applications in cosmology.8,9 In this short talk we review our results from
the paper 1 with the main stress on the issues of causality and acoustic metric in
eikonal approximation. In addition we correct the formula for the redshift of sound
signals from our original work. 1 In our paper1 we considered a scalar field ¢ with
the generally covariant and Lorentz invariant action

s= J 4
d xyCgp(X) , where p(X) = a
2
[VI + ~~ ~ 1]. (1)

aIn bimetric theories this possibility can be easily achieved. 2

bThe accretion of other noncanonnical scalar fields were studied in e.g. lO ,11
cor of other kinetically coupled to <p fields as in. 3 In this case it is also possible to get information
from the interior of BH.

1471
1472

The Lagrangian p(X) depends only on X == ~ \7 fJ¢\7fJ¢, and a is a free parameter

of the theory. d The kinetic part of the action is the same as in 9 and for small
derivatives, that is, in the limit 2X « a 2 , it describes the usual massless free scalar
field. In the case of arbitrary p(X) the equation of motion for ¢ is

GfJ V\7 Ji \7 v¢ = 0, where the induced metric GJiV == glW + p,xx \7Ji¢\7 v ¢, (2)
p,x
and p,x == op/ox. This equation is hyperbolic and its solutions are stable with
respect to high frequency perturbations provided (1 + 2Xp,xx/p,x) > 0. 8 ,14-16
The propagation vectors NJi are tangent to characteristic surface and define the
influence cone:
- 1 NfJN V --
G Ji1/ ° h
, were G- 1
fJV -_ gJiV _
p,x
P.XX
+ 2 X p,xx
" ~" ~
v I"'+'V v'+' (3)

is inverse matrix to Gl"v. The influence cone is larger than the light cone if
p,xx/p,x < 0. 14- 16 In this case the front of small perturbations of ¢ propagates
faster than light. If the background ¢(x) is trivial, \71"¢ = 0, then perturbations
(small discontinuities) propagate with the speed of light. Therefore only nontriv-
ial backgrounds ¢(x) spontaneously break the Lorentz invariance. Despite the fact
that the action (1) is manifestly Lorentz invariant the action for perturbations 5¢
around a non-trivial background solution is only generally covariant but not Lorentz
invariant anymore. This background can be considered as a medium or "new ether" .
Observers moving differently with respect to this medium may disagree in the re-
sults of some measurements. Moreover in the case of superluminal propagation there
is no Lorentz invariant notion of causality. 13,16 However, by virtue of the hyperbol-
icity of the system even in this case there may exist some Cauchy hypersurfaces 15
and therefore observers for which the causality is well defined. 17 Nevertheless, there
are backgrounds 13 where closed time like curves (CTC) exist. However, in the stan-
dard GR 18 it is also the case. The so-called chronology protection conjecture 19
may preclude the existence of CTC. For a more detailed discussion of causality
in the theories with spontaneously broken Lorentz-invariance see paper. 6 For the
energy-momentum tensor we have Tl"v = p,x \71"¢\7 v¢-pgl"v, Thus the Null Energy
Condition Tl"vnl"nv ?:: 0 e is satisfied if p,x ?:: 0 . This is always the case for our
model (1) and hence the black hole area theorem 2o holds. It is well known that, if
\7 v¢ is timelike (that is, X > 0 in our convention), then the system with general
p(X) is formally equivalent to a perfect fluid with the pressure p = p(X), energy
density c(X) = 2Xp,x(X) -p(X), the four-velocity ul" = \71"¢/V2X and the sound
speed c; == op/8c = p,x/c,x, Specializing to the case of the Lagrangian (1) we have
2 2X c -1 P \7 ¢\7 ¢
cs = 1 + - 2 ?:: 1, 2 = (1 - Cs ), 2 = (c s - 1), G;;} = gl"v + I" 2 1/ • (4)
a a a a

dThroughout the paper \1lt denotes the covariant derivative and we use the natural units in which
G=n=c=1.
en!' is null vector in g!'1/
 1473

Here we sketch how to find a stationary spherically symmetric background solu-

tion for the scalar field falling onto a Schwarzschild black hole. In the Eddington-
Finkelstein coordinates f the metric takes the form:
g
ds 2 =f(r)dV 2 -2dVdr-r 2 d0 2 , where f(r)=l- r , r g =2M. (5)
r
Inl we verified that there is a broad range offree parameter a 2 for which the infalling
field has a negligible influence on the black hole, that is, we consider an accretion
of the test fluid in the given gravitational field. The stationarity and cosmological
boundary conditions at spatial infinity imply the following ansatz for the solution:

¢(v, r) = a/c; - 1 (V + JT F(r')dr') , where Ci is the speed of sound at infinity.

(6)
For every solution F(r) the induced acoustic line element (in eikonal approximation)
is
=
dS 2 G~~dxMdxv = (c; - r:) dV 2 -2dVdT (1 - (c; - 1)F)+(c;-1)F 2dr2-r 2d0 2 .
(7)
In this acoustic metric the coordinate V is timelike. Therefore from (7) it follows
that there exists sonic horizon at r* = TgJC; S- r g . Substituting (6) into (2) and (7)
one can obtain that the only physical solution which satisfies all boundary conditions
and for which the acoustic space-time (7) is not singular for r :::0: r*, is given by:

F(x) = f(~) ( fe)

c;+f(X)-l) _
g
c i - 1) -1 , where x=rJr .
c i + (2
x x 48
(8)

And for the sound speed one obtains c;(x) = X3c~ J [1 + c;(x - 1)(1 + xc; + x 2c;)].
The acoustic spacetime (7) with the function (8) describes an analogue black hole
with the horizon which is inside the Schwarzschild horizon. Therefore it is possi-
ble to use perturbations o¢ around this background (6),(8) as signals and to send
information from the region r * < r < r 9 between two horizons, see Fig. 2 from. 1
On the background solution \lM¢ is a time like non-vanishing vector field well de-
fined for r :::0: T*. Thus in accordance with 21 the acoustic space time is stably (and
therefore strongly) causal for T :::0: r *. Suppose that a spacecraft moves together
with the falling background field and sends the acoustic signals with the frequency
W em . After simple calculations one can obtain that an observer at rest at the spatial

infinity will detect these signals at the frequency Win]:

(9)

This expression corrects our result from. 1 Note that the ratio Wem JWin] is finite for
any T > r * and it vanishes for r = r *. In particular for the moment of crossing the
Schwarzschild horizon we have WemJWin] = c;J1 + c; + c; + cV(c; - 1).

fnote that these coordinates are regular at Schwarzschild horizon

1474

Acknowledgements
We are very thankful to C. Bonvin, C. Caprini, S. Dubovsky, R. Durrer, V. Frolov,
S. Liberati, A. Rendall, S. Sibiryakov, A. Starobinsky, L. Susskind, R. Woodard
and especially Sergei Winitzki for very useful discussions. A. V. would like to thank
the theory group of Laboratori Nazionali del Gran Sasso, INFN and organizers and
staff of Les Houches Summer School for hospitality during the preparation of this
manuscript. E.B. thanks Alexander von Humboldt foundation for support on the
early stage of this project and INFN for support during the preparation of the
manuscript.

References
1. E. Babichev, V. Mukhanov, A. Vikman, JHEP 0609,061 (2006) [hep-th/0604075].
2. D. Blas, C. Deffayet, J. Garriga, Class. and Quant. Grav. 23 (2006) 1697 [hep-
th/050S163].
3. S. Dubovsky, S. Sibiryakov, Phys.Lett.B63S:509-514 (2006), [hep-th/060315S].
4. M. Born, L. Infeld, Proc.Roy.Soc.Lond.A144:425-451,(1934).
5. C. Armendariz-Picon, T. Damour, V. Mukhanov, Phys.Lett. B458, 209-21S
(1999) [hep-th/9904075]; C. Armendariz-Picon, V. Mukhanov, P. J. Stein-
hardt, Phys.Rev.Lett.85,443S-4441 (2000) [astro-ph/0004134]; C. Armendariz-Picon,
V. Mukhanov, Paul J. Steinhardt, Phys. Rev. D63:10351O (2001) [astro-ph/0006373].
6. E. Babichev, V. Mukhanov, A. Vikman in preparation.
7. C. Bonvin, C. Caprini, R. Durrer, Phys.Rev.Lett.97:0S1303 (2006), [astro-
ph/06065S4].
8. J. Garriga, V. F. Mukhanov, Phys.Lett.B458, 219-225,(1999) [hep-th/9904176].
9. V. Mukhanov, A. Vikman, (2006) JCAP 0602:004, [astro-ph/0512066]; A. Vikman
astro-ph/0606033.
10. S. Mukohyama, Phys.Rev.D71, 104019 (2005) [hep-th/0502189].
11. E. Babichev, V. Dokuchaev, Yu. Eroshenko, Phys.Rev.Lett. 93, 021102 (2004) [gr-
qc/0402089]; Zh.Eksp.Teor.Fiz.l00, 597, (2005) [astro-ph/050561S]
12. W.G. Unruh, Phys.Rev.Lett.46:1351-1353,(1981).
13. A. Adams, N. Arkani-Hamed, S. Dubovsky, A. Nicolis, R. Rattazzi,JHEP
0610:014,(2006), [hep-th/060217S].
14. C. Armendariz-Picon, E. A. Lim, JCAP 050S:007, (2005),[astro-ph/0505207].
15. A. D. Rendall, Class.Quant.Grav. 23, 1557-1570 (2006), [gr-qc/0511158].
16. Y. Aharonov, A. Komar, L. Susskind, Phys.Rev.182:1400-1403,(1969).
17. Jean-Philippe Bruneton, gr-qc/0607055, hep-th/0612113.
IS. W. B. Bonnor, Int. J. Mod. Phys. D12:1705-1708, (2003), [gr-qc/0211051]; Class.
Quant. Grav.19:5951-5957, (2002); Amos Ori, gr-qc/0701024; J.Richard Gott, Phys.
Rev. Lett.66:1126-1129, (1991).
19. S. W. Hawking, Phys. Rev. D 46, 603 - 611 (1992).
20. S. W. Hawking, G. F. R. Ellis, The large scale structure of space-time, Cambridge,
1973.
21. R. M. Wald, General relativity, The Univ. of Chicago Press, 1984.
 A DIELECTRIC ANALOGUE MODEL OF THE KERR
EQUATORIAL PLANE

KJELL ROSQUIST
Stockholm University, AlbaNova University Center, 10691 Stockholm, Sweden
[email protected]

An optical analogue solution of the Kerr equatorial plane is given. To find the solution a
special notation for calculations with two different metrics has been used. The possibility
of extending the solution to the complete Kerr geometry is also discussed.

1. The Schwarzschild analogue and the ergo region

The motion of light in a dielectric medium can be described as geodesic motion in
the Gordon geometry. For a medium with refractive index n the Gordon metric is
given byl
g=7]+ku@u, (1)
2
where 7] is the Minkowski metric of the laboratory, k = l/n and u is the 4-velocity
I-form of the medium (we are using the tilde to distinguish the 4-velocity u itself
from its dual I-form u). The geometry represented by the effective metric 9 will be
referred to as the target space. By an appropriate choice of medium, it is possible
to construct analogue models of solutions of the Einstein equations. Such analogue
models can be used to illuminate properties of the solutions. In principle, although
difficult in practice, they can also be used as blueprints for constructing laboratory
simulations of background gravitational fields.
In this note we will describe a dielectric analogue of the Kerr equatorial plane
starting from the Schwarzschild analogue model given in Ref.2. The standard form
of the Schwarzschild solution is
(2)
where f = 1 - 2MJr and d0 2 = de 2 + sin 2 e d¢ is the 2-sphere metric. The metric
function f can be expressed in the covariant form f = -e"~a where ~ = aJaT
is static Killing vector field. In the analogue formulation the laboratory time t is
related to the Schwarzschild time T by
dt = ndT + Fdr , (3)
where F = f~lJ(1- j)(n 2 - j). The analogue metric can be written in the "re-
fracted" form

where sa = (dt, dr, rde, r sin e d¢) is a laboratory orthonormal frame and u= uaS a
with the nonzero 4-velocity components given by

uo = -un =
~
-2-f
n
-- 2 -
1 '
u1 = Ul = - ~-f
-2- - .
n - 1
(5)

1475
1476

The laboratory 4-velocity I-form is given by ill = -So. It follows that the fluid
gamma factor r = -uaw a = U O is given by

~ (6)
r=V~'
and the magnitude of the fluid 3-velocity can be written in the form
1
v = ---;;====== (7)
/1 + (n 2 -1)~
V 2M
The relation (6) can also be expressed in the form
1- V /c 2
2

v 2/ cr; ,
~a ~a =- 1 (8)
-
where Cm = c/n is the speed of light in the medium. It follows that the superluminal
boundary v = Cm corresponds exactly to the ergo surface condition ~a ~a = o. The
original definition of the ergo region involves an observer at infinity (see e.g. Ref.3).
We see from (8) that in the optical analogue context we can define the ergo region by
the condition v > Cm. In general, we may define the ergo surface as the boundary of
the stationary/static region of spacetime. In the Schwarzschild case, the ergo region
according to our definition is r < 2M and hence coincides with the region inside
the black hole. Since this region cannot be related to an observer at infinity, it is
usually said that there is no ergo region in the Schwarzschild geometry. However,
by using the local definition of the ergo region it makes perfect sense to consider
also parts of the spacetime which are hidden behind the horizon.

2. The two assumptions

Another important consideration which can give us a clue for finding an analogue
formulation concerns the static observers. By definition, the static observers appear
static to observers at infinity. The 4-velocity of a static observer is given by 'f) =
1-1/2~. On the other hand, the lab 4-velocity is given by w = 8/8t = n- 1 8/8T =
n- 1 ( can be related to the static observers 'f)a in the Schwarzschild target space by

(9)
This relation shows that the static observer word lines coincide with the world lines
of the lab. The factor 1- 1 / 2 reflects the different normalizations required depending
on whether one is taking the perspective of the lab or the target space. A crucial part
of the derivation of the Kerr optical analogue was The static observer assumption:
The lab 4-velocity world world lines coincide with the static observer world lines.
We willl now use the relation (6) to formulate the second assumption which will
lead us to the Kerr analogue formulation. Note first that the function 1 can be
defined invariantly by 1 = _~a~a where ~ = 8/aT is the stationary Killing field.
From (9) we see that the static observers are only defined if 1 > O. It follows that
 1477

the ergo surface (outside the horizon) is given by f = O. We can now formulate our
second crucial assumption as The ergo function assumption: The relation between
the fluid gamma factor and the ergo function f is given by (6).

3. The Kerr equatorial analogue

The Kerr metric can be written in the following generalized Boyer-Lindquist form
gK = -h(Mo - \liMl)2 + h-l(Ml)2 + (M2)2 + (M 3 )2 , (10)
where \Ii = \Ii(r) is an arbitrary function and Me> is a Minkowski frame given by
3 sinB 2
M O = Po (dT - asin 2 0d¢) , Ml = £.dr , M2 = pdB, M = -(-adT+pod¢) ,
P Po P
(11)
and where
h = 1- 2~r
(12)
Po
The metric form (10) is similar to the form (4) of the Schwarzschild metric. The
important difference is the appearance of the refractive index in the latter. The
problem to obtain an optical analogue of Kerr is solved if we can write the Kerr
metric in a refracted form as in (4). We will show how this can be done for the
equatorial part of the geometry.
To derive the analogue formulation we look for a relation between the lab time
t and the Killing time T of the target space. The relation should take the form
dt = ndT + Fdr + aPde + l/ad¢ , (13)
where l/ is a dimensionless constant and the likewise dimensionless functions F =
F(r, e) and P = P(r, e) are constrained by the integrability condition F,e = aP,T".
Considering only the equatorial plane greatly simplifies the problem since = 11"/2 e
implies that we can set P = 0 and F = F(r). By letting l/ = -nk a further
simplification occurs since the fluid 4-velocity can then be taken to lie in the plane
of the principal null directions. Using the two assumptions mentioned above, the
fluid 4-covector takes the following form (in the lab geometry)
u =_ ,r
Po
KO _ F Po Kl.
nk,r
(14)

The normalization requirement 'i]",(3u"'u(3 = -1 leads to

F = 1,- 1 y!(n 2 - 1,)(1 - h) , (15)

where
1, _ hP6 a = a/n . (16)
- -2 '
Po
The refracted form of the equatorial Kerr metric is then given by
gKeq = -n- 21,(Mo - FMl)2 + 1,-l(Ml)2 + (M 3 )2 , (17)
1478

where
1
]\;[0 = Po (dt - ad¢) , ]\;[1 = ~dr - 3
M =
2
-(-adt + Pod¢)
-
(18)
r Po ' r
and where ¢ is defined by the relation

d¢ = d¢ + aPo2 Fdr . (19)

The fluid velocity is given by

u= - J 2
n - h ]\;[0 -
n 2 -1
J1- h
n 2 -1
]\;[1 (20)

The relation to the lab radial coordinate, denoted f, is f = Po. We not the close
resemblance in form of the fluid velocity in this case compared to the Schwarzschild
analogue case (5). In performing the calculations leading up to the above results,
it is indispensable to use a certain notation to distinguish the operations of raising
and lowering indices with respect to the lab and target metrics respectively. Gen-
eralizing to the full Kerr geometry is rather more complicated. In particular, the
condition v = -nk cannot be used to make the fluid velocity lie in the plane of the
principal null directions in that case. However, one can still use the general frame-
work indicated above and analyze if the assumptions about the static observers and
the ergo function are sufficient to produce an analogue model in the general Kerr
case.

Acknowledgement
This work was carried out with support from the ICRANet network.

References
1. W. Gordon, Ann. Phys. (Berlin) 72, p. 421 (1923).
2. K. Rosquist, Gen. Rei. Grav. 36, p. 1977 (2004), (related online version:
gr-qc/0309104) .
3. C. W. Misner, K. S. Thorne and J. A. Wheeler, Gravitation (Freeman, San Francisco,
USA,1973).
 BOSE-EINSTEIN CONDENSATES AND QFT
IN CURVED SPACE-TIME

SERENA FAGNOCCHI
Centro Studi e Ricerche "Enrico Fermi", Compendio Viminale, 00184 Roma, Italy
INFN sezione di Bologna, via Irnerio 46, 40126 Bologna, Italy
[email protected]

It is breafiy shown how QFT in curved space can be used to gain an analytical expression
for the depletion in non-homogeneous Bose-Einstein condensate.

It is known that weakly interacting Bose-Einstein condensates (BEe) in hy-

drodynamical approximation can be described by Quantum Field Theory (QFT)
in curved space,l since the fluctuations in hydrodynamical systems behave exactly
as a massless scalar field propagating on a fictitious curved spacetime, whose line
element reads 2

ds 2 = ~[-e2dt2 + (dx - iJdt) . (dx - iJdt)] , (1)

me
with iJ the velocity of the fluid, e the sound speed related to the density n via the
BEe equation of state me 2 = gn, g = 47rti2 a/m and a the scattering length. For a
homogeneous BEe the corresponding acoustic line element will be flat, since in these
systems the" curvature" is induced by inhomogeneities. This analogy holds in the so
called hydrodynamical regime, i.e. when the scale L over which the inhomogeneities
ofthe condensate vary is much bigger than the healing length (L » ~ = til (2g p) 1/2).
In this case, the usual Gross-Pitaevskii equation (GPE) for the order parameter
M m
I}J = y'rie / C~B = miJ) assumes the hydro dynamical form, namely a continuity
and a Bernoulli equation for the density n and the velocity potential B:

(2)

Within this approximation, the density fluctuation will be linked to the phase fluc-
tuation in an algebraic way, while the equation for the phase fluctuation 1 can e
be written via the gravitational formalism as Del = 0, where D is the covari-
ant D'Alembertian calculated on the curved background of Eq. (1). As a conse-
quence, for modes whose wavelength A » ~ the usual dispersion relation for BEe
w = ekJ1 + ek2 can be approximated with the relativistic one w = ek. In this
regime the condensate behaves as a fluid and QFT in curved space can be safely
used to describe the propagation of these modes. It is worth stressing that this is a
purely geometrical analogy, directly involving only kinematical features: black holes,
horizons, trapped surfaces and Hawking radiation, being geometrical objects, will
have an acoustic counterpart in particular BEe systems. Nevertheless this amazing
analogy can be useful also in addressing non-kinematical issues. In fact manipulat-
ing inhomogeneities with the standard techniques in hydrodynamical BEe is not a
trivial task. In some sense the analogy can try to overtake this kind of problems,

1479
1480

providing an alternative powerful tool to handle them. An example for this treat-
ment can be the estimation of the corrections on the non-condensed particle number
(the so .called depletion) the inhomogeneities induce on the standard result,3 which,
for low temperature (T .z::: /1g), reads:

8 3 2 K~BT 2]
no = 3K 1 / 2 (noa) / [1 + ( 2nog ) (3)

So far, the only attempt to take inhomogeneities into account is the so called lo-
cal density approximation (lda), 4 consisting in replacing into Eq. (3) the constant
density no with the actual spatial varying nIx). This way cannot take care of spa-
tial derivatives effects. From Eq. (3) it is evident that BEe show a non-vanishing
depletion even at zero temperature: unlike the temperature term which is com-
pletely determined by the relativistic part of the spectrum, this quantum part is
linked to the high frequency modes that cannot be described by the relativistic
dispersion relation, and therefore it can in no way be reproduced within hydrody-
namical approximation. Acting as a sort of ground state, our purpose is to evaluate
the correction to this induced by the inhomogeneities within hydro dynamical ap-
proximation. As for the thermal depletion we argue that corrections to the ground
state result (which is sensitive to the microphysics of the system6 ) can be estimated
through a low energy theory. What we have in mind is that curvature is felt by those
modes with wavelength comparable to its scale (,\ "-' L » 0, which are therefore
described by the phononic low-energy theory. The high frequency ones (,\ .z::: L) will
see the fluid as homogeneous and will not contribute to the curvature corrections.
If this were the case, QFT in curved space techniques can be used to analytically
value these terms. As shown in details in,7 the depletion is reduced to a (¢2) cal-
culation in curved space, with ¢ a massless scalar field propagating on the acoustic
background described in Eq. (1). This quantity is UV divergent and point-spilling
renormalization technique has been used to regularize it. 8 For a static spherically
symmetric configuration (n = n( r), if = 0), for very low T the result reads

_ ~~T2 (m3)1/2 (mgn)1/2 [7n'2 5n" 5n / ] 2g1/2n3/2

nin = -- - - --- + -- + -- 111/1 +
12h3 gn 96K 2h 8n 2 2n r n m 3/ 2

+ (mgn)1/2 [_ 39 n'2 + ~ n" + ~ n/].

48K 2h 16 n 2 4 n 2r n (4)

Note that this procedure is able to exactly reproduce the thermal depletion term in
Eq. (3). In addition to this, Eq. (4) gives in analytical form the contributions to the
fraction of non condensed particles coming from inhomogeneities for any density
function nCr) solution of the GPE.
It is possible to extend this treatment also to a non-static case. Adding source
terms which do not depend on the order parameter in the continuity equation a,

aThese terms are necessary in order to conserve the mass. They can be added without invalidating
our treatment, since the equations for the perturbations are left unchanged.
 1481

one can create stationary non-static spherically symmetric configurations: n = n(r),

'If = v(r)T/lrl For such situations is still possible to estimate the quantum depletion
induced by this radial flux. It reads:

ii q
m
=- Tn
961T 2nc 2 2
[~CC'2 + 5c2c" + -3v 2c" - 2cv'2 - 2cvv" + ~v2C'2
2c
2 2
-4vc'v' + ~(5C2C' - 4cvv' _ v 2c') _ 22V2C] Inp,2 (c - v ) +
c
T' T' g

+ 481TTn2nc 2 {
- 2c(c 21_ v 2) (3 C2C, - v 2c, - 2cvv ')(3 c2c, - 2v 2 c, - cvv ')

15 '2 - -1v 2 c,2 - -vc

--cc 5 "v 3 2"
+ +-c c - 1 2 c" - cv 12 - cvv " +
-v
4 4c 2 2 2
+~(3c2cl - v 2c' - 2cvv' ) } . (5)

Eq. (5) generalizes Eq. (4) for a stationary radial velocity of the fluid. It is worth
pointing out that this result is valid just for Ivl « c. The divergences appearing
in Eq. (5) for v 2 = c2 (the so called sonic horizon) are in fact due to the singular
behavior of the modes on the horizon (see 8 ). This kind of divergence is not due to
UV effect already renormalized and desappears if the black hole formation process
is taken into account (i.e. considering the quantum field in the so called Unruh
state 9 ). We expect the same happens in the acoustic settillg.

Acknowledgements
gratefully thank Roberto Balbinot for the fundamental contributions to this paper
and Grisha Volovik for stimulating comments. This work is supported by "E. Fermi"
Center, Rome.

References
1. L.G. Garay, J.R. Anglin, J.I. Cirac and P. Zoller, Phys. Rev. Lett. 85, 4643 (2000);
C. Barcelo, S. Liberati and M. Visser, Living Rev. ReI. 8, 12 (2005)
2. W.G. Unruh, Phys. Rev. Lett. 46, 1351 (1981); R. Balbinot, A. Fabbri, S. Fagnocchi,
R. Parentani, La Rivista del Nuovo Cimento 28, 1 (2005)
3. N.N. Bogoliubov, J. Phys. (USSR) 11 , 23 (1947); A.L. Fetter and J.D. Walecka,
Quantmn Theory of Many-Particle Systems, McGraw-Hill, New York, (1971)
4. L.P. Pitaevskii, S. Stringari, Bose-Einstein condensation, Clarendon, Oxford (2003)
5. S. Fagnocchi, gr-qc/0611096 (2006)
6. See for ex. Volovik's contribution in Artificial black holes, M. Novello, M. Visser, G.E.
Volovik, World Scientific, River Edge, USA (2002)
7. R. Balbinot, S. Fagnocchi, A. Fabbri, cond-mat/0610367 (2006)
8. P.R. Anderson, W.A. Hiscock and D.A. Samuel, Phys. Rev. D51, 4337 (1995)
9. A complete treatment of the behavior of modes in the quantum states associated to
gravitational black holes can be found in Black Hole Physics: basic concepts and new
developments, V.P. Frolov and I.D. Novikov, Kluwer Ed., Dordrecht, The Netherlands
(1997)
 ELECTROMAGNETIC LIGHT RAYS IN LOCAL DIELECTRICS*

VITORIO A. DE LORENCI

RENATO KLIPPERT
Universidade Federal de ltajuba (UNIFEI),
Av. BPS 1303, ltajuba 37500-903, Brazil
klippert@uni/ei. edu. br

An approach to deal with the limit of geometrical optics of electromagnetic waves which
propagate in moving nonlinear local dielectric media in the context of Maxwellian elec-
trodynamics is here developed in order to apply to quite general material media. Fresnel
equations for the light rays are generically found, and its solutions are intrinsically ob-
tained. The multi-refringence problem is addressed, and no more than four monochro-
matic polarization modes are found to propagate there. This work does not provide an
analog model of General Relativity, but instead relies in the fundamentals for an analog
model to be built in geometrical optics.
Maxwell theory naturally yields electromagnetic wave propagation phenomena.
This problem is fully solved in vacuum, since those equations become linear. Inside
material media, however, polarization and magnetization fields may lead to nonlin-
ear differential equations which are more difficult to deal with. Light propagation
is also studied in the context of nonlinear Lagrangians for electrodynamics,2,3 the
vacuum properties of which being similar to those of material media considered
here. Some time ago, electrodynamics in material media has been considered as a
possible scenario to investigate analogue models for gravitational phenomena (see
Ref. 4 and references therein). This work intends to obtain and discuss the gen-
eral Fresnel equations which completely describe the propagation of light rays in a
non-homogeneous non-isotropic nonlinear moving local dielectric medium. Material
discontinuities were omitted here due to the lack of space, but it can be shown l
that they do agree with the essential results discussed below for smooth media.
For the sake of generality, equations are written in a covariant form which ac-
counts for electrodynamics under the influence of the gravitational field 91LV in an
arbitrary causal orientable 4-dimensional Lorentzian manifold (i.e., a causal space-
time). We adopt the metric with signature (+, -, -, -). Although this approach
makes Maxwell equations rather involved, it is shown that the propagation of the
field discontinuities looks just the same as if working in the flat 9J.Lv = T/J.LV case (for
which the Levi-Civita tensor is given in Cartesian coordinates by T/0123 = +1). We
also adopt the standard decomposition for the covariant gradient of the observer's
normalized velocity vector congruence VJ.L as VJ.L;v = (e/3)h~+(TJ.Lv+wJ.Lv+aJ.LVvwith
all aJ.L, h~, (TJ.LV = (TVJ.L (traceless), and wJ.LV = -wJ.LV being spacelike tensors. We write
the electromagnetic tensors 5 of field strength FJ.Lv = V[J.L EV] - T/J.LV a(3 vaB(3 and
field excitation pJ.LV = V[J.L DV] - T/J.LV a(3va H(3. An electromagnetic current source
vector ]J.L may also be present, and the velocity of light in vacuum is c = 1. Only

*This work is based upon the homonimous recently published paper 1 by the same authors, and
was supported by the two brazilian agencies CNPq and FAPEMIG.

1482
 1483

monochromatic waves are considered here, avoiding subtleties on their velocity.

All the electromagnetic properties of usual dielectric materials are encoded in
the phenomenological constitutive relations which give pJLV as a function of Fa(3 in
the 3 + 1 form by means of the local expressions 5
D a =c(1)a(3E(3 +c(2t(3B(3, (la)
H a = M(lt (3B(3 + M(2) 0: (3E(3, (lb)
where all the parameters C(i) 0: (3 and M(it (3 may be dependent on position as well
as on the field strengths. Maxwell equations can be covariantly written either in
compact form as pJLV;v = 47r JJL and 'f]JLv)..p FJLv;)" = 0, or explicitly in terms of the
3+1 electromagnetic field strengths EJL and BJL with the aid of Eqs. (1).
In order to describe electromagnetic wave propagation in the limit of small
wavelenghts (geometrical optics), we adopt the Hadamard method of dealing with
field discontinuities. 6 Applying this procedure to the electromagnetic fields EJL and
BJL we assume that [EJL]L; = 0 = [BJL]L;, while [EJL;V]L; = eJL Kv and [BJL;V]L; = bJL K v ,
with all other fields being supposed to be smooth across ~, where eJL and bJL are the
spacelike polarization vectors of the light ray whose wave-front normal is KJl" The
discontinuity over ~ of Maxwell field equations yields

ba Ka = 0, (2a)
(d(1) a (3e(3 + d(2) a (3b(3)Ka = 0, (2b)
= 0,
(Kv VV)bJL - 'f]JLVPUVvepKu (3a)
(d(lt (3e(3 + d(2t (3b(3) (KvVV) - 'f]JLVP Vv(h(1)u (3b(3 + h(2t (3e(3)Kp = 0,
U (3b)
where d(i)JL (3 and h(i)JL (3 are known functions of cCi)(3;v and MCi)(3;v and their deriva-
tives. Eqs. (2) can be obtained from Eqs. (3) upon contraction with KJl" Propagation
of light rays in smooth media is thus completely described by Eqs. (3).
Let us consider 7 the 3-spatial projection qv ~ h~K).. of the wave-vector Kv.
Then, q2 ~ -ha(3qo:q(3 -:::: 0 and v<p ~ w/q is the phase velocity of the light ray along
the direction of q)... We assume q2 = 1 everywhere, with no loss of generality. Thus,
qv is a unitary spatial vector which identifies the direction of propagation of the
light ray if w > 0 or the opposite direction if w < O. In terms of qv, Eqs. (3) take
the form of a linearly polarized bJL = V;;;l'f])..JL vu VV qJLe u light ray determined by a
polarization vector eJL such that za (3e(3 = 0, where

Z o: (3 =. d (1) 0: (3 + v<p-2h (l)JL vI[JLI a ] + -1 (d a p).. + h p a).. ) VV

(3 v V<p (2) p'f] v(3 (2) (3'f] vp q).., (4)
where 10: (3 ~ h3 + qa q(3 is the symmetric projector onto the 2-dimensional space of
3-spatial vectors orthogonal to the 3-wave-vector q)...
Instead of expanding the eigenvector eJL with the aid of a suitable basis of vec-
tors,3 we rely here in the algebraic structure of the eigenvalue problem. Non-trivial
eigenvectors e(3 #- 0 can be obtained if and only if the 3-spatial determinant of
za(3 vanishes, a relation covariantly expressed 8 as Z 1 3 - 3Zl Z2 + 2Z3 = O. It is
convenient to express the Fresnel matrix za (3 as a polynomial function of l/v<p in a
compact form za (3 = a (3 + ba (3 / v<p + ea (3 / v~, where a, b, e are matrices independent
Q
1484

of v<p and can be read of from Eq. (4). One finds that the two higher order terms in
l/v<p (namely, the fifth and sixth order terms) are identically zero due to algebraic
identities. s The Fresnel equation det Z = 0 takes the form

[a1 3 - 3ala2 + 2a3] + ...£..[a 2 - ala + ai -2af n] : b + -%[(a 2 - ala + a/ 2-a 2 n) : c

v~ v~

2n )·. a] + "'£"[1(b 2
+ (b2 - bI b+ bi -b
2 v~ 3
2 - b b+ b, -b 2n )· b+
I 2' (5)
+ (abc + acb) - (albc + blac + clab) + alblcl]+
2
+ Jl,{[b 2 - bIb + b 1 2-b 2 n] : c + [c 2 - CIC + C1 ;C2 n] : a} = O.
2

V<p

This generically is an algebraic equation of fourth order for the velocity of the ray,
as expected. 3 ,g Material media with sufficiently general dielectric properties may
thus possibly give rise to four distinct positive velocities v~), corresponding to the
four propagating polarization modes (eo<, bO<)(j) for any given direction g)". That is,
an unpolarized light beam may split into up to four polarized beams, although the
most entangled situation known so far is trirefringence. lo Optical analog models
require a special factorization of Eq. (5) into two algebraic factors of second order.

References
1. V. A. De Lorenci and R. Klippert, Phys. Lett. A 357, 61 (2006).
2. W. Dittrich and H. Gies, Phys. Rev. D 58, 025004 (1998); ibid, Phys. Lett. B 431,
420 (1998); G. W. Gibbons and C. A. R. Herdeiro, Phys. Rev. D 63, 064006 (2003);
Yu. N. Obukhov and G. F. Rubilar, 64, 024042 (2002).
3. V. A. De Lorenci et ai, Phys. Lett. B 482, 134 (2000); M. Novello et ai, Phys. Rev. D
61, 045001 (2000).
4. C. Barcelo, S. Liberati, and M. Visser, gr-qc/0505065 (2005).
5. F. W. Hehl and Yu. N. Obukhov, physics/0005084 (2000); L. Landau and E. Lifchitz,
Electrodynamique des milieux continus (Mir, Moscow, 1969).
6. J. Hadamard, in Ler;ons sur la propagation des ondes et les equations de l'hydro-
dynamique (Ed. Hermann, Paris, 1903); see also Y. Choquet-Bruhat, C. De Witt-
Morette, and M. Dillard-Bleick, Analysis, Manifolds and Physics, p. 455 (North-
Holland Publishing, New York, 1977); J. Plebanski, Lectures on non-linear electro-
dynamics (Nordita, Copenhagen, 1968); G. Boillat, J. Math. Phys. 11, 941 (1970).
7. R. Wald, General Relativity (Univ. Chicago Press, Chicago, 1984).
8. R. R. Silva, J. Math. Phys. 39, 6206 (1998).
9. V. A. De Lorenci and M. A. Souza, Phys. Lett. B 512, 417 (2001); V. A. De Lorenci
and R. Klippert, Phys. Rev. D 65, 064027 (2002); G. F. Rubillar (PhD thesis), Ann.
Phys. (Leipzig) 9, 1 (2002); Yu. N. Obukhov and G. F. Rubilar, Phys. Rev. D 66,
024042 (2002).
10. M. C. Netti et ai, Phys. Rev. Lett. 86, 1526 (2001).
 SCATTERING PROBLEMS ON ROTATING
ACOUSTIC BLACK HOLES

CHRISTIAN CHERUBINI and SIMONETTA FILIPPI

Facolta di Ingegneria, Universita Campus Bio-Medico, Via E. Longoni 83, 1-00155 Rome, Italy,
ICRA, Universita di Roma "La Sapienza," 1-00185 Rome, Italy.
[email protected],[email protected]

Scattering problems on rotating acoustic black holes can be studied in the time domain
by using numerical methods only. In order to obtain high quality results in simulations,
tools developed by numerical relativists in the case of rotating Kerr black holes must be
adopted, i.e. constrained evolution schemes for strongly symmetric and non symmetric
hyperbolic systems of PDEs and excision techniques in horizon penetrating coordinates.
In this paper such a program is presented and analogies between astrophysical and
acoustic black holes are discussed in relation with Press and Teukolsky's relativistic
"black hole bomb."

The draining bathtub is a solution of irrotational barotropic and incompressible

Euler's equations,l being described by a velocity potential 1jJ( r, ¢) = -cp log( r / p) +
Op2¢, where c is the constant speed of sound, p is a certain length associated to the
"radius" of the vortex and 0 is the constant rotation frequency. The z-extruded
acoustic metric associated to this object is
p2C2 + p402) 2cp
ds 2 = - ( c 2 - r2 dt 2 + --;:-dtdr - 20p 2dtd¢ + dr 2 + r 2d¢2 + dz 2 (1)

as a consequence of the fact that perturbations of the velocity potential1jJ(1) == \[I

°
satisfy the scalar wave equation \lll'\l /-L \[I = on this background, i.e.
2 2 2 2 2
1 8 2p 8 2p 0 8 ( p2) 8
[ - c 2 8t 2 + cr 8tOr - c 2r2 8t8¢ + 1 - r2 8r2 + (2)

2p 30 8 2 c2r2 - p402 8 2 82 r2 + p2 8 2p 30 8 ]
+----+
cr 3 8r8¢ c2r 4 8¢2
+-+ 8z 2 r3
-
8r
- - - - \[1=0.
cr 4 8¢
The PDE above is strongly hyperbolic: in order to perform a successful integration,
numerical or analytic, the typical tools of differential geometry (and consequently
of General Relativity) can be fruitfully used. The line element (1) can be better
analysed if rewritten in 3 + 1 form:
(3)
The three metric is simply the fiat Euclidean space one in cylindrical coordi-
nates '1ij = Diag(l, r2, 1). The spatial indices are here denoted with latin let-
ters and are raised and lowered with '1ij and its inverse '1 ij = Diag(1,r- 2 ,1).
Greek letters denote four-dimensional quantities instead. The shift is given by
/3 i = (pcr-1,-p 20 r -2,0), so /3i = (pcr- 1 ,-p20,O). The lapse is simply a = c.
The slices t = const are fiat, consequently this effective space-time is written in
Painleve-Gullstrand like form. The presence of a shift vector manifests the well
know phenomenon of frame dragging. The analysis of the previous section allows
us to adapt for our equation (2) some of the mathematical prescriptions developed
1485
1486

in ref.2 for the numerical integration of the massless scalar field perturbations on a
rotating Kerr black hole background. The starting point is to introduce the quan-
tities <l>i = g;: ,II = -~ (~~ - ,Bi<l>i) , and insert them into eq.(2), obtaining -by
construction- an hyperbolic system. In order to avoid large computational requests,
we introduce the decomposition i]! = 'l/Jl (t, T)eimq, eikz ,II = 7rl (t, T)eim </> eikz ,<1>1 =
cPl (t, T)eim<Peikz ,<1>2 = imi]! ,<1>3 = iki]! . The initial wave equation is replaced now
by this first order system of PDEs
p2
8'l/Jl _ c~ (f!.'l/Jl) =(PC - im r1,) 'l/Jl - C7rl
8t 8T T T2
p2
87rl + c~ (cPl _ f!.7r 1 ) = (PC - im r1,) 7rl + C (k2 + m2) 11'1 - 9.cPl (4)
8t 8T T T2 T2 T
2
8cPl + c8- ( 7rl - -P '/-'1
-- "')
=
2ip mrl, 'l/Jl - - imrl,p2 '"
- - '/-'1 .
8t 8T T T3 T2
Associated on these variables there is a constraint Co = 18r 'l/Jl - cPll which must be
always very close to zero numerically. This quantity will be a useful indicator of the
goodness of the numerical system evolution (this is the well known "constraint vio-
lation problem"). Regarding the initial data, following the standard prescription for
scattering processes in Kerr black holes,3 we choose a gaussian pulse modulated by a
monochromatic wave 11'1 (0, T) = Ao exp[-(T - TO + ct? /b 2 - ia(T - TO + ct)/ cllt=o,
together with the associated initial conditions for 7rl and cPl, easily obtained using
their definitions in terms of 'l/Jl. The corresponding power spectrum of this initial
data for 'l/Jl is P(w) = Pmax exp [-(w - a)2b 2/( 4c2)]. The modulated pulse leads to
a Gaussian frequency distribution which peaks around w = a. In this way it is pos-
sible to enter in the superradiant regime by choosing 0 < a < mrl, or remain outside
this interval. Concerning the boundary conditions we use "excision"technique, due
to the horizon penetrating nature of the used coordinate system. Inside the hori-
zon in fact, light cones point towards the singularity so nothing from the interior
region of the hole can affect the region outside the horizon. Consequently we set
no boundary conditions on the inner boundary Tin, with 0 < Tin < p. Regarding
the outer boundary Tout the standard criterion is to have conditions which will not
violate constraints, will not generate spontaneously unphysical waves and will not
reflect back significatively any incoming signal, allowing in this way the wave to
escape towards infinity. No boundary condition for 'l/Jl and equation cPl = 7rl are
the proper ones. 4 These boundary conditions have given no significant constraint
violations during simulations. The problem formulated in this way has been recently
exhaustively analyzed 4 ,. 5 On the other hand one can oppositely impose reflecting
boundary conditions on the outer boundary (i.e. place a mirror)
2
8'ljh = 8 'l/Jl = 0 8'ljJl _ '" (5)
at 8t2 ' 8T - '/-'1 ,
and using the field equations (4), re-express these time dependent boundary condi-
tions in terms variables 'l/Jl, cPl and 7rl and their spatial derivatives only. Remaining
in the superradiant sector, an instability analogous to the one known for the Kerr
 1487

solution as the "black hole bomb"found by Press and Teukolsky 6 and analysed in
the 1 + 2 dimensional acoustic case in ref. 7 arises. Both these studies were performed
in the frequency domain. We have preferred to analyze the 3 + 1 case in the time
domain instead. In figure 1 it is shown the amplitude A = J1l1J at a given point
outside the horizon during time (details will be presented in ref. 8 ). As expected the
incoming wave is superradiantly amplified by the black hole, perfectly reflected by
the mirror then, superamplified, reflected again and so on, leading to the growing
black hole bomb instability. Possible applications in hydrodynamical contexts will
be analyzed in the future.

0.5

0.4

0.3

A
0.2

0.1

50 100 150 200 250 300 350 400 450 500

Figure 1. Amplitude of the field at a given point outside the horizon during time.

Bibliography
1. M. Visser Class. Quant. Grav. 15, 1767 (1998).
2. M. A. Scheel, A. L. Erickcek, L. M. Burko, L. E. Kidder, H. P. Pfeiffer, and S. A.
Teukolsky Phys. Rev. D 69, 104006 (2004).
3. N. Andersson, P. Laguna, and P. Papadopoulos Phys. Rev. D 58, 087503 (1998).
4. C. Cherubini, F. Federici, S. Succi, and M. P. Tosi Phys. Rev. D 72, 084016 (2005).
5. F. Federici, C. Cherubini, S. Succi, and M. P. Tosi Phys. Rev. A 73, 033604 (2006).
6. W.H. Press and S.A. Teukolsky, Nature, 238, 211 (1972).
7. V. Cardoso, O. J. C. Dias, J. P. S. Lemos, and S. Yoshida Phys. Rev. D 70, 044039
(2004).
8. C. Cherubini and S. Filippi, in preparation (2007).
This page intentionally left blank
Black Hole Thermodynamics
This page intentionally left blank
 THERMODYNAMICAL PROPERTIES
OF HAIRY BLACK HOLES
WITH COSMOLOGICAL CONSTANT

M. NADALINI* and L. VANZO** and S. ZERBINI***

Department of Physics, University of Trento,
via Sommarive,14 , 38100, Trento, Italy
and INFN, Gruppo Collegato di Trento, Italy
* E-mail: [email protected]
** E-mail: [email protected]
*** E-mail: [email protected]

We present an analysis of solutions of General Relativity with a scalar field conformally

coupled to the gravity sector. These are interesting thanks to the possibility, in principle,
of obtaining black hole solutions with negative thermodynamical entropy. Then, we move
on to a thermodynamical analysis of those solutions and give a prescription for how
to deal with such entropies and focus our attention to a somewhat unexpected result
regarding the AdS/eFT correspondence.

1. THE THEORY
The action of pure gravity in n space-time dimensions with a conformally coupled
scalar field ¢(x) and cosmological constant A = ±(n-2)(n-1)/21 2, looks as follows:

1=
J ~
d n Xy-g [1'2 ( 1 1
R-2A)-'2Cv¢) 2 -'2~cR¢ 2 -V(¢) ] , (1)

where ~c = (n - 2)/4(n - 1) is the conformal coupling, V(¢) = 0:¢2n/(n-2) is the

conformal potential, and we have choosen units Ii = f£B = C = 87rG = 1. Such a
choice for the coupling and the potential guarantees the invariance of the scalar
field equations under the group of conformal transformations of the metric.

2. SEARCHING FOR SOLUTIONS

We now search for solutions in this Schwartzschild-like form:
2 2 dr2 2
ds = -A(r)dt + A(r) + r dI',2, ¢(x) = ¢(r). (2)

What we leave unspecified is the topology of the (n - 2) dimensional transversal

manifold I',.
We can, by inspecting directly the resulting equation of motion, find a solution
for the scalar field, ¢(r) = c(r + ro)-(n-2)/2, and then we get two other equations
for the metric. One has a general analytic solution, and the other was used as a
consistenct check and for tuning the physical parameters like A, 0: and the two
integration constants arising in the scalar field solution.
In 3 spacetime dimension, where the topology of the transverse manifold becomes
trivial, we find solutions defined by A(r) = -Ar2+3r6(A+5120:)+2r5(A+5120:)/r,
c = ±~, 0:, ro > O. When 0: :s; -A/512 and A < 0, one can have a black hole

1491
1492

with a single event horizon. This solution was first found here. 1 In the positive A
case, one can find a solution with a single cosmological horizon.
In 4 spacetime dimensions the topology of the transverse manifold can become
richer. Defining k = 1 for spherical topology, k = 0 for euclidean topology, and
k = -1 for the hyperbolic one, a general solution of the system can be presented
as A(r) = k(I + ro/r)2 - Ar 2/3, with a = -A/36 if k #- O. This represents a black
hole in the de Sitter case if k = 1, 2 and in the anti de Sitter case if k = _1. 3 In the
latter case, in a certain range of the parameters, one has an allowed static region
inside the outer black hole horizon, sort of a black hole inside a black hole.
Another interesting solution can be obtained with some manipulation in the
euclidean case in order to obtain a time dependent solution for the scalar field on a
de Sitter background in all dimensions. It is
c
¢(t, r) = n/2-1. (3)
(ro + Toe t / l JI - r2 /[2)
Regarding other black hole solution, we have checked by direct calculation until
n = 11, and found none. We conjecture that, given the particular form of the
constraints on the physical parameters, there are no black hole solutions for n ~ 5
with that simple ansatz.

3. THERMODYNAMICS OF SOLUTIONS
We have analysed the thermic behaviour of the above solutions, using the tunneling
method for computing the temperature 45 and Wald's Ni:ither charge method 678 for
computing the entropy, yelding
T = IdA/drlr=rH (4)
41T '
where r H is the radius of the horizon being considered. The explicitly negative term
-~c¢(rH? is responsible for the possibility of having negative entropies.
Since we are dealing with a theory that has a conformally invariant degree of
freedom, we may recall that the Hawking temperature is a conformally invariant
quantity,9 as well as we may note that the entropy is invariant in value (but not
in form) at least for the transformations that bring the lagrangian from the Jordan
frame (nonminimal coupling of the field with the metric) to the Einstein frame
(minimal coupling).
Coming to the n = 4 de Sitter solution, one finds immediately that when ro < 0
both a cosmological and an event horizon are present. Their temperatures are the
same (lukewarm case), TE = Te = (JI - 4ro/l) /21Tl, while the entropies are
equal in value but opposite in sign: Be = -BE ex: T E . One can note that in the limit
ro ~ 0 the scalar field vanishes the solution collapse to the usual de Sitter one.
In our opinion, negative entropies are hardly acceptable. Thus, our proposal is
to exploit a shift ambiguity in Wald's differential method in order to obtain, in the
above limit, a null entropy for the event horizon and the usual de Sitter one for the
 1493

cosmological horizon. This can be accomplished with an appropriate shift in the

values of the entropy, and the results are perfectly acceptable (Vp is the measure of
the fundamental domain of the horizon manifold): SE = 27fVp 12 (1 - JI - roll)
The generic dimensional solution with the scalar field depending on both the t
and r coordinates, it doesn't present any problem: the entropy on the cosmologi-
cal horizon is time independent, and it is less than the standard de Sitter one as
expected from general arguments.
We now would like to focus our attention to the 4-dimensional asymptotically
anti de Sitter black hole. Performing an analysis of the mass with the first law of
B.H. thermodynamics, dM = TdS, one finds out that MBH = 2Vpro + Vp ex Tk. !
The AdS/CFT correspondence would suggest, at least for large mass, M ex Tk Is
this an example of violation of AdS/CFT correspondence? Possibly, this is not a
violation, but it could represent a perhaps new aspect of the correspondence that
deserves a little more inspection in order to be properly understood. This could be
related to the fact that the conformal weights of the conformally coupled scalar field
are different from the usual case of a free scalar field.

References
1. M. Henneaux, C. Martinez, R. Troncoso and J. Zanelli, "Black holes and asymptotics
of 2+1 gravity coupled to a scalar field," Phys. Rev. D 65 (2002) 104007 [arXiv:hep-
th/0201170j.
2. C. Martinez, R. Troncoso and J. Zanelli, "Exact black hole solution with a minimally
coupled scalar field," Phys. Rev. D 70 (2004) 084035 [arXiv:hep-th/0406111j.
3. C. Martinez and R. Troncoso, "Electrically charged black hole with scalar hair," Phys.
Rev. D 74 (2006) 064007 [arXiv:hep-th/0606130j.
4. M. Angheben, M. Nadalini, L. Vanzo and S. Zerbini, "Hawking radiation as tun-
neling for extremal and rotating black holes," JHEP 0505 (2005) 014 [arXiv:hep-
th/0503081 J.
5. M. Nadalini, L. Vanzo and S. Zerbini, "Hawking radiation as tunneling: The D-
dimensional rotating case," J. Phys. A 39 (2006) 6601 [arXiv:hep-th/0511250j.
6. R. M. Wald, "Black hole entropy in the Noether charge," Phys. Rev. D 48 (1993)
3427 [arXiv:gr-qc/9307038j.
7. V. Iyer and R. M. Wald, "Some properties of Noether charge and a proposal for
dynamical black hole Phys. Rev. D 50 (1994) 846 [arXiv:gr-qc/9403028j.
8. V. Iyer and R. M. Wald, "A Comparison of Noether charge and Euclidean methods
for computing the entropy of stationary black holes," Phys. Rev. D 52 (1995) 4430
[ar Xiv:gr-qc /9503052J.
9. T. Jacobson and G. Kang, "Conformal invariance of black hole temperature," Class.
Quant. Grav. 10 (1993) L201 [arXiv:gr-qc/9307002J.
 RADIATION OF QUANTIZED BLACK HOLES.
IS IT OBSERVABLE?*

IOSIF KHRIPLOVICH
Budker Institute of Nuclear Physics,
Novosibirsk 630090, Russia
[email protected]

The maximum entropy of a quantized surface is proportional to the surface area in

the classical limit. The general structure of the black hole horizon spectrum is found.
The value of the Barbero - Immirzi parameter in loop quantum gravity is derived. The
discrete spectrum of thermal radiation of a black hole fits the Wien profile. The natural
widths of the lines are much smaller than their separation. If primordial black holes
(PBH) with discrete radiation spectrum saturate the present upper limit on the dark
matter density in our Solar system, the sensitivity of modern detectors is close to that
necessary for detecting this radiation.

1. Quantized Horizon of Black Hole

The horizon area A consists of patches of typical size Planck mass squared l~ . Each
patch is characterized by a quantum number j, and its contribution a to the area
depends on j, a = a(j). Besides, a patch can possess a quantum number m, such
that a is independent of it. So, the horizon area is

A = 81fil; L a(j) Vjm ; (1)

jm

Vjm is the number of patches of given j and m, 81fi is a dimensionless factor.

To derive general relations for the "occupation numbers" Vjm, we use the Beken-
stein - Hawking relation between the horizon area and black hole entropy S:

(2)
and the so-called holographic bound: the entropy S of any system confined inside a
sphere of area A is bounded by relation

(3)
with the equality attained only for a system which is a black hole. In other words,
among the spherical surfaces of a given area, the surface of a black hole horizon has
the largest entropy.
We will consider the "micro canonical" entropy S of a quantized surface defined
as the logarithm of the number of states of this surface for a fixed area A. This
number of states K depends on the assumption concerning the distinguishability of
the patches. The only reasonable assumption, which may comply both with (2) and
(3) is as follows. 2 Only the patches of same j and same m are indistinguishable; the

* This research has been partially supported by the Russian Foundation for Basic Research grant
05/02/16627.

1494
 1495

patches of different j, or different m, or different both j and m are distinguishable.

Then, truly back-of-the-envelope calculations l - 4 result in

A = 81f'Y I; v L g(j) a(j) e- 27ria (j) , (4)

j

where g(j) is the number of possible values of m for a given j (i.e. the statistical
weight); the value of'Y is found from equation

L g(j) e- 27ri a(j) = 1. (5)

j

In particular, in Loop Quantum Gravity (LQG), where a(j) = jj(j + 1) and

g(j) = 2j+ 1, one arrives at (the so-called Barbero - Immirzi parameter) 'Y = 0.274.a

2. Quantized Radiation of Black Hole 4

A quantum is radiated when a patch of a given value of j disappears, so that

Wj = T/1>a(j) , /1> = 21f'Y' (6)

The radiation spectrum is discrete with a finite number of lines. It starts at Wmin =
T/1>a min, where amin is the minimum value of a(j), and terminates at Wmax = Tlnv.
The probability of radiation of a quantum with frequency Wj is proportional to
the occupation number Vj. It leads, at least in the Wien limit wiT » 1, to the
following expression for the discrete radiation spectrum of a black hole:

I ] . ~W]. -- I ] 8a(j)
8j
~. _ AT4 _1_ 4 3(.) 8a(j) -I"a(j) ~.
J - 41f2 /1> a J 8j e J. (7)

Here ~j is the difference between two successive values of j.

The width of a line with frequency Wj = T /1> a(j) is

(8)

The ratio of this natural line width to the line separation ~Wj is

(9)

With [/1> a(j)]2 e-I" a(j) -s; 0.54, this ratio is very small numerically: -s; 10- 3 . Thus,
the radiation spectrum of an isolated black hole is really discrete.

a Attempts to calculate the area spectrum in LQG had started long ago, but only recently5 led to
the correct result.
1496

3. Is Radiation of Quantized Black Holes Observable?6

Observational data for the secular perihelion precession of Earth and Mars dictate
the following upper limit on the density Pdm of dark matter in the Solar system: 7
(10)
Our quantitative estimates for the expected signal from PBHs (see Table) are
performed under the optimistic assumption that their density P rv 10- 19 g/ cm3 .
This assumption is compatible with other searches for PBHs. A PBH with the

2 X 10 15 5 X 10- 35 2.7 X lOll 5 1.5 X 10 19 1.6 X 10- 5

10 16 10- 35 0.5 X 10 12 1 3 X 10 18 10- 6

initial mass ma;oS; 5 x 10 14 g could not survive till our time due to the same
radiation. And for masses larger than 10 16 g, the signal gets hopelessly small. In
Table 1, n = p/m is the density of number of primordial black holes, f = n- 1 / 3 is
the typical distance between two of them; N ~ c/ (647f 4 r g) is the number of quanta
per second emitted in a line by a black hole with a gravitational radius r g; v is the
typical expected flux of quanta in a line at the distance f = n- 1 / 3 . The temperature
T roughly corresponds to the energy of the first line in the discrete spectrum of a
black hole. One may expect 2~ 3 relatively strong r lines with energies about 5,
10, 15 MeV in the spectra of PBHs with mass 2 x 10 15 g; about 1, 2, 3 MeV in
the spectra of PBHs with mass 10 16 g. Few such narrow lines in the spectrum of a
source of radiation is essential for its identification with a quantized black hole.
This identification requires a r ray telescope capable of fine spectroscopy for
point sources. One possibility is to use the SPI spectrometer on board INTEGRAL,
which has good sensitivity for discrete spectrum. 8 To observe the line around 5 MeV
which belongs to a PBH with m rv 2 X 10 15 g, would require to analyze very deep
sky images in a large set of fine energy bands. This is doable with existing data.

References
1. R.Y. Korkin, LB. Khriplovich, Sov. Phys. JETP, 95, 1 (2002); gr-qc/0112074.
2. LB. Khriplovich, Sov. Phys. JETP, 99, 460 (2004); gr-qc/0404083; gr-qc/0409031.
3. LB. Khriplovich, Sov. Phys. JETP, 100, 1075 (2005); gr-qc/0412121.
4. LB. Khriplovich, Gribov Memorial Volume (World Scientific, 2006); gr-qc/0506082.
5. A. Corichi, J. Diaz-Polo, E. Fernandez-Borja, gr-qc/0605014; gr-qc/0609122.
6. LB. Khriplovich, N. Produit, Int. J. Mod. Phys., D, in press; astro-ph/0604003.
7. LB. Khriplovich, E.Y. Pitjeva, Int. J. Mod. Phys., D15, 615 (2006); astro-ph/0601422.
8. J.P.Roques, ABA, 411, L91 (2003).
 EFFECTS OF QUANTIZED FIELDS ON THE SPACETIME
GEOMETRIES OF STATIC SPHERICALLY SYMMETRIC
BLACK HOLES

PAUL R. ANDERSON*,t,t, MATHEW BINKLEY*, HECTOR CALDERON§, WILLIAM A.

HISCOCK§, EMIL MOTTOLA' and RUSLAN VAULINII
* Department of Physics, Wake Forest University, P.O. Box 1501, Winston-Salem, NC, 21109,
USA and
t Racah Institute of Physics, Hebrew University of Jerusalem, Givat Ram, Jerusalem, 91904,
Israel and
t Departamento de Fisica Te6rica and IFIC, Universidad de Valencia-CSIC, C. Dr. Moliner 50,
Burjassot-46100, Valencia, Spain
§Department of Physics, Montana State University, Bozeman, MT, 59111, USA
, Theoretical Division, T8, Los Alamos National Laboratory, Los Alamos, NM, 81545, USA
II Department of Physics, Florida Atlantic University, 111 Glades Road, Boca Raton, FL, 33431,
USA

Analytic approximations for the stress-energy of quantized fields in the Hartle·Hawking

state l in static black hole spacetimes predict divergences on the event horizon of the black
hole for a number of important cases. Such divergences, if real, could .substantially alter
the spacetime geometry near the event horizon, possibly preventing the black hole from
existing. The results of three investigations of these types of effects are presented. The
first involves a new analytic approximation for conform ally invariant fields in Reissner-
Nordstrom (RN) spacetimes which is finite on the horizon. The second focuses on the
stress-energy of massless scalar fields in Schwarzschild-de Sitter black holes. The third
focuses on the stress-energy of massless scalar fields in zero temperature black hole
geometries that could be solutions to the semiclassical backreaction equations near the
event horizon of the black hole.

It is well known that previous analytical approximations for quantized massless

fields in the Hartle-Hawking state in RN spacetimes 2 ,3 predict that one component
of the stress-energy tensor diverges logarithmically on the event horizon in the
nonextreme case. For extreme Reissner-Nordstrom (ERN) black holes there is both
a powerlaw and a logarithmic divergence predicted. On the other hand numerical
calculations for massless spin 0 and spin 1/2 fields 3 - 5 show no evidence for such
divergences.
For conformally invariant fields there is an approximate effective action, some-
times called the anomaly action, which when varied with respect to the metric
results in a stress-energy tensor whose trace is equal to the trace anomaly.6-9 The
behavior of this stress-energy tensor has been studied for Schwarz schild and other
spacetimes.lO,ll It can be written in terms of two auxiliary fields which both obey
fourth order differential equations. This results in numerous solutions which, at least
to some extent, correspond to various possible states for the quantized fields.1o,1l
An investigation of this action for the case of RN and ERN spacetimes 12 shows
that there exist solutions to the auxiliary field equations which give a finite stress-
energy on the event horizons of both types of black holes. The best approximation
can be obtained by fitting the value of Ttt on the horizon and at large r in RN.
This results in one free parameter which can be varied to give the best fit to the
stress-energy at intermediate locations. The approximation for ERN is not as good
because Tt t is fixed on horizon and the value is not close to the correct one. However,

1497
1498

it is still possible to fix the value of (Trr - Ttt)/(-gtt) on the horizon and to fix
Tt t at large r. While not perfect, these approximations have the advantage of being
the only known approximations for massless quantized fields which do not have
divergences on the horizon for RN and ERN.
The analytic approximation 3 for massless scalar fields with arbitrary coupling ~
to the scalar curvature R makes the somewhat surprising prediction 13 that if the
black hole temperature is nonzero and if R =I 0 on the horizon (e.g. in Schwarzschild-
de Sitter spacetimes), then near the horizon
(T t t ) rv -(Trr) rv ~(~ - 1/6) R(rh) .
rh(r - rh)
These components are finite on the horizon in both RN and ERN spacetimes where
R = O. Given the fact that divergences predicted for a different component of the
stress-energy tensor in RN and ERN turned out not to be real, one might expect
that the same would be true for these apparent divergences associated with R =I O.
To test this conjecture, numerical computations of the stress-energy tensor for
massless scalar fields with various values of ~ have been computed on the horizon
for Schwarzschild-de Sitter spacetimes. 13 In these spacetimes there is both an event
horizon for the black hole and a cosmological horizon and the two are at different
temperatures. However, the fields can be put in the Hartle-Hawking state if the black
hole is surrounded by a perfectly reflecting mirror. The numerical calculations have
been done using a variation of a method 14 developed for Schwarzschild spacetime
that gives the values of (Tt t), (Tr r), and (TOo) on the horizon. The numerical
evidence indicates that these components are finite on the event horizon.
The existence of the trace anomaly, which is nonzero for Schwarzschild and
RN spacetimes, means that R =I 0 for self-consistent solutions to the semiclassical
backreaction equations. The case of zero temperature black holes is a particularly
natural one to investigate because there is no mirror needed to contain the radiation
as there is for other black holes when the fields are in the Hartle-Hawking state.
The existence of a divergence in the stress-energy of a potential solution to the
semiclassical equations would likely prevent that solution from existing. However,
that does not necessarily mean that no zero temperature black hole could exist.
Recent results regarding zero temperature black holes in two dimensions 15,16 show
that it is possible that the divergences could imply that no static state exists but that
there is no divergence if the black hole forms from gravitational collapse with the
fields in a particular nonstatic state. Nevertheless, this is still a significant deviation
from one's intuition that static solutions to the semiclassical backreaction equations
should exist that describe zero temperature black holes.
To determine the effects of the fields on the spacetime geometry in as complete a
way as possible, it is useful to work in the context of a large N expansion where N is
the number of identical quantized scalar fields. Unlike the usual loop expansion, the
leading order terms in the large N expansion allow for the possibility that quantum
effects can significantly alter the spacetime geometry. It is also consistent to neglect
the graviton stress tensor since this comes in at next to leading order.
If only conformally invariant fields are present then it is possible to argue that
static spherically symmetric zero temperature black hole metrics near the event
horizon should have the general form
ds 2 = -[a2(r - rh)2 + ... ]dt 2 + [b 2(r - rh)2 + ... r1
+ r 2d0 2
 1499

with rh the radius of the event horizon and a2 and b2 positive constants. The
trace equation has been solved in this case and it has been found that b2 is a
function of rh and that b2 > 1. 17 There is good evidence that (TM v) depends only
on geometry near the horizon. Numerical results have been obtained for the exact
metric -gtt = (r - n,? /r2 and grT = b2 gtt for various values of b2.
In all cases considered so far, there is preliminary evidence that the stress-energy
of the conformally invariant scalar field diverges on the horizon. The divergence
appears to occur for all values of b2 > 1. However this is not conclusive since the
value of the component which diverges on the horizon depends on a3, b3 and possibly
higher terms in the expansion. In the cases where the divergence is present it occurs
for all ro > 0 and thus for macroscopic as well as microscopic black holes.
In summary, a new analytic approximation has been found for Reissner-
Nordstrom and extreme Reissner-Nordstrom spacetimes which is finite on the event
horizon. No evidence has been found for the existence of divergences predicted by
an analytical approximation for the stress-energy of quantized massless scalar fields
in nonzero temperature black hole spacetimes when R -I=- O. There is evidence that
spacetime geometries exist for which the stress-energy of the conformally invariant
scalar field diverges on the event horizon of a zero temperature black hole. How-
ever the possibility that there are self-consistent static zero temperature black hole
solutions to the semiclassical backreaction equations has not been ruled out.

Acknowledgments
P.R.A. would like to thank Jacob Bekenstein, Alessandro Fabbri, Sara Farese, and
Jose Navarro-Salas for helpful conversations. This research has been partially sup-
ported by grant numbers PHY-9800971, PHY-0070981, and PHY-0556292 from the
National Science Foundation. P.R.A. thanks the Gravitation Group at the Univer-
sity of Maryland for hospitality and acknowledges the Einstein Center at Hebrew
University, the Forchheimer Foundation, and the Spanish Ministerio de Educacion
y Ciencia for financial support.

References
1. J. B. Hartle and S. W. Hawking, Phys. Rev. D 13, 2188 (1976).
2. V. P. Frolov and A. 1. Zel'nikov, Phys. Rev. D 35, 3031 (1987).
3. P.R. Anderson, W.A. Hiscock and D.A. Samuel, Phys. Rev. Lett. 70, 1739 (1993); Phys. Rev.
D 51, 4337 (1995).
4. P. R. Anderson, W. A. Hiscock, and D. J. Loranz, Phys. Rev. Lett. 74,4365 (1995).
5. E. D. Carlson, W. H. Hirsch, B. Obermayer, P. R. Anderson, and P. B. Groves, Phys. Rev.
Lett. 91, 051301 (2003).
6. R. J. Riegert. Phys. Lett. 134B, 56 (1984).
7. E. S. Fradkin and A. A. Tseytlin. Phys. Lett. 134B, 187 (1984).
8. S. D. Odintsov and 1. L. Shapiro. Class. Quant. Grav. 8, L57 (1991).
9. P. O. Mazur and E. Mottola. Phys. Rev. D 64, 104022 (2001).
10. R. Balbinot, A. Fabbri, and 1. Shapiro. Phys. Rev. Lett. 83, 1494 (1999); Nucl. Phys. B559
301 (1999).
11. E. Mottola and R. Vaulin, Phys. Rev. D 74, 064004 (2006).
12. P. R. Anderson, E. Mottola, and R. Vaulin, manuscript in preparation.
13. H. Calderon, W. A. Hiscock, and P. R. Anderson, manuscript in preparation.
14. P. Candelas, Phys. Rev. D 21, 2185 (1980).
15. R. Balbinot, S. Fagnocchi, A. Fabbri, S. Farese, J. Navarro-Salas, Phys. Rev. D 70, 064031
(2004).
16. S. Fagnocchi and S. Farese, Phys. Rev. D72, 024015 (2005).
17. P. R. Anderson and C. D. Mull, Phys. Rev. D59, 044007 (1999); gr-qc/9711063; gr-
qc/9707026.
 THERMODYAMIC QUANTITIES OF KALUZA-KLEIN BLACK
HOLES WITH SQUASHED HORIZONS

YASUNARI KURITA * and HIDEKI ISHIHARA

Department of Physics, Osaka City University
Sugimoto 3-3-138 Sumiyoshi-ku 558-8585 Osaka, Japan
* Email:[email protected]
[email protected]

We investigate thermodynamics of 5-dimensional charged static black hole with squashed

horizons. Taking the Kaluza-Klein monopole spacetime as a background spacetime, we
find a new non-trivial work term for the first law of the Komar mass, which represents
the squashing of the horizon. This thermodynamic formulation includes thermodynamics
of 5-dimensional Reissner-Nordstrom black hole as a limit, and thus, this is a generalized
thermodynamic formulation for five-dimensional charged black holes in Einstein-Maxwell
system.

1. Introduction and Summary

It has a long history to formulate black hole thermodynamics in terms of quantum
theory of gravity. Especially, the Euclidean path integral for gravitational field in
saddle-point approximation is successful and gives Bekenstein-Hawking formula for
black hole entropy. 1 This Euclidean path integral method is consistent formulation
for black hole thermodynamics and has a feature that it gives thermodynamic free
energy.
Recently, spacetime structure of charged static black holes with squashd 53
horizons in five-dimensional Einstein-Maxwell theory was studied by one of the
authors and his collaborator. 2 The spacetime is asymptotically locally fiat, i.e.,
asymptotes to a twisted 51 bundle over four-dimensional Minkowski spacetime.
The size of 51 is asymptotically constant and, if its size is adequately small, the
spacetime is effectively asymptotically four-dimensional. Furthermore, the solution
includes the Kaluza-Klein monopole spacetime 4 ,5 as a no horizon limit. Therefore,
the black hole looks like a Kaluza-Klein black hole (KKBH).
Cai et.a1 3 investigated the thermodynamics of KKBH mensioned above and
found a thermodynamic first law for Abbott-Deser mass on a product 51 bundle
over four-dimensional Minkowski spacetime background. The obtained first law is
satisfied only when a parameter characterizing the squashing is fixed. The KKBH
has three parameters which roughly characterize mass, electric charge and squashing
of the spacetime. Thus, it is natural to consider that its thermodynamic variables
include these three quantities.
In the present paper, we consider the Komar mass as an alternative thermo-
dynamic mass. On the Kaluza-Klein monopole background, we find a set of ther-
modynamic quantities including non-trivial work term, representing the squashing
of outer horizon. Furthermore, it will be shown that this thermodynamic formu-
lation on the Kaluza-Klein monopole background includes thermodynamics of 5-
dimensional Reissner-Nordstrom black hole, which is asymptotically fiat, as a limit.

1500
 1501

In this sense, this formulation is a generalized thermodynamics for five-dimensional

charged black holes in Einstein-Maxwell system.

2. Kaluza-Klein black hole with squashed horizons

The five-dimensional Einstein-Maxwell theory has a black hole solution with the
following metric:?
2 2 k(r)2 2 r2 [ 2 2]
ds = - f(r)dt + f(r) dr + 4" k(r)d0,s2 +X , (1)

f(r) = (1- r~) (1-

r2
r:'),
r2
k(r) = (r~ - r~)(r~ -
(r~ - r2)2
r:'),

d0,12 = de 2 + sin e 2dq} , X = d'ljJ + cos edT,

where r +, r _ and roo are constants, and 0 ::; e ::; 7f, 0::; T ::; 27f and 0 ::; 'ljJ ::; 47f are
angular coordinates. The solution is parametrized by three parameters: r +, r _ and
roo· The inner and outer horizons are at r = r _ and r = r + respectively. r -+ roo is
a spatial infinity. The electric potential is given by A = =f r~~_ dt. V;
In the limit r + -+ 0 and r _ -+ 0, the solution reduces to the Kaluza-Klein
monopole solution 4 ,5. It is a 5-dimensional vacuum solution of Einstein gravity
while, in four-dimensional point of view, it becomes a monopole solution in the
sense of Kaluza-Klein theory.
A time slice ofthe spacetime is foliated by three-dimensional surfaces r = canst.,
say I:. Each surface I: can be thought of as a twisted Sl fiber over S2 base space.
Now, the aspect ratio of the size of S2 base space to Sl fiber is given by the function
k (r). If k (r) is unity, then the surface I: r is round S3, and if it is not unity, then I: r
is squashed S3. If one considers the limit roo -+ 00, then the function k (r) becomes
to unity, and thus the surface I: r becomes round S3 and the spacetime becomes the
5-dimensional Reissner-Nordstrom spacetime.

3. Thermodynamic quantities
The free energy F of the Kaluza-Klein black hole on the Kaluza-Klein monopole
background can be calculated as

F = - - r -r 2 - zr 2]
7f1[2 (2)
4G z 00 - 00'

where z := Jf(roo). We consider the Komar mass as a thermodynamic mass be-

cause it satisfies Smarr's formula,6

_ 3
MK - - 327fG
J J.l v _ 37f
dSJ.ll/D ~ - 8Gz
(2r+ + r 2- - r~ r~ )
2 r~ . (3)

Now, we introduce a couple of quantities X and Y;

X = 4Cr!o, Y = _1_~ [r~ + -

--'--;:---
87fGC z r~
r:' 2 (1 - z) , ] (4)
1502

where C is a constant. Then the first law for the Komar mass is found to be
dMK = Td8 + ipdQ + X dY, (5)
where T is the Hawking temperature, 8 is entropy given by the Bekenstein-Hawking
formula, ip is the electric potential difference between that of the outer horizon and
of the spatial infinity r = roo. Q is the electric charge. With these thermodynamic
variables, the Komar mass can be related to the free energy on the Kaluza-Klein
monopole background by Legendre transformations
F = MK - T 8 - Qip - XY. (6)
This formulation includes usual thermodynamic formulation for 5-dimensional
Reissner-Nordstrom black hole as a limit roo ---> 00. Thus, it is a generalized formu-
lation for thermodynamics of 5-dimensional charged black holes. From this point of
view, the new work term is thought of as a thermodynamic value corresponding to
deformation from spherical symmetry. Actually, the variable Y can be rewritten as

Y = -
'if 2 (r +
i6GC R+r+ 2 - R+
)2 , (7)

where R+ = r; V(r'?x, -
r:") / (r'?x, - r~) is the circumference radius of 8 2 base space
on the outer horizon and r +/2 is the circumference radius of 8 1 fiber at the outer
horizon. Thus, Y vanishes when the size of 8 2 equals that of 8 1 and the horizon is
a full-orbed three sphere. In this way, Y measures the squashing of the horizon.
In the new work term, there exists an undetermined constant C. At present, we
do not know how to determine it. The situation is similar to the case of entropy in
early '70s; the coefficient of entropy was determined by the theoretical discovery of
Hawking radiation. Thus, also in this case, some unknown physics might determine
the constant C.

Acknowledgments
We are grateful to colleagues in the astrophysics and gravity group of Osaka City
University for helpful discussion and criticism. YK thanks to the Yukawa memorial
foundation for its support in early stage of this work.

References
1. G. W. Gibbons and S. W. Hawking, Phys. Rev. D 15, 2752 (1977).
2. H. Ishihara and K. Matsuno, Prog. Theor. Phys. 116, 417 (2006).
3. R. G. Cai, L. M. Cao and N. Ohta, Phys. Lett. B 639, 354 (2006).
4. D. J. Gross and M. J. Perry, Nue!. Phys. B 226, 29 (1983)
5. R. Sorkin, Phys. Rev. Lett. 51, 87 (1983)
6. J. P. Gauntlett, R. C. Myers and P. K. Townsend, Class. Quant. Grav. 16, 1 (1999).
 HAWKING RADIATION AND BLACK HOLE
THERMODYNAMICS *

DON N. PAGE
Theoretical Physics Institute, University of Alberta, Edmonton, Alberta, Canada T6G 2G'l,
[email protected]

A brief review of Hawking radiation and black hole thermodynamics is given.

1. Introduction

Black holes are perhaps the most perfectly thermal objects in the universe, and yet
their thermal properties are not fully understood. They are described very accu-
rately by a small number of macroscopic parameters (e.g., mass, angular momen-
tum, and charge), but the microscopic degrees of freedom that lead to their thermal
behavior have not yet been adequately identified.
Strong hints of the thermal properties of black holes came from the behavior
of their macroscopic properties that were formalized in the (classical) four laws
of black hole mechanics,l which have analogues in the corresponding four laws of
thermodynamics:
The zeroth law of black hole mechanics is that the surface gravity K, of a station-
ary black hole is constant over its event horizon. 1 ,2 This is analogous to the zeroth
law of thermodynamics, that the temperature T is constant for a system in thermal
equilibrium.
The first law of black hole mechanics expresses the conservation of energy by
relating the change in the black hole mass M to the changes in its area A, angular
momentum J, and electric charge Q in the following way:
1
8M = -K,8A + [MJ + <I>8Q, (1)
8n
where an extended form of the zeroth law implies that not only the surface gravity
K" but also the angular velocity n and the electrostatic potential <I> are constant

over the event horizon of any stationary black hole. This first law is essentially the
same as the first law of thermodynamics.
The second law of black hole mechanics is Hawking's area theorem,3 that the
area A of a black hole horizon cannot decrease. This is obviously analogous to
the second law of thermodynamics, that the entropy S of a closed system cannot
decrease.
The third law of black hole mechanics is that the surface gravity K, cannot be
reduced to zero by any finite sequence of operations. 4 This is analogous to the
weaker (Nernst) form of the third law of thermodynamics, that the temperature T
of a system cannot be reduced to absolute zero in a finite number of operations.

*This research has been partially supported by Natural Sciences and Engineering Research Council
of Canada.

1503
1504

However, the classical third law of black hole mechanics is not analogous to the
stronger (Planck) form of the third law of thermodynamics, that the entropy of a
system goes to zero when the temperature goes to zero.
Thus the four laws of black hole mechanics are analogous to the four laws of
thermodynamics if one makes an analogy between temperature T and some multiple
of the black hole surface gravity K, and between entropy S and some inversely
corresponding multiple of the black hole area A. That is, one might say that T = EK
and S = 7]A, with 87rE7] = 1, so that the KoA/(87r) term in the first law of black hole
mechanics becomes the heat transfer term ToS in the first law of thermodynamics.
Nevertheless, by a quite independent line of reasoning that was not directly moti-
vated by Bekenstein's proposal that he had rejected,l Hawking made the remarkable
discovery that black holes are not completely black but instead emit radiation. 5 ,6
Once he found that the radiation had a thermal spectrum, he realized that it did
make Bekenstein's idea consistent, of a finite black hole entropy proportional to
area, though not Bekenstein's conjectured value for 7]. In fact, Hawking found that
the black hole temperature was T = K/(27r), so E = 1/(27r) and hence 7] = 1/4. This
gives the famous Bekenstein-Hawking formula for the entropy of a black hole:
1
Sbh = SBH == -A. (2)
4
Here the subscript bh stands for "black hole," and the subscript BH stands for
"Bekenstein-Hawking."

2. Hawking Emission Formulae

For the Kerr-Newman metrics,7,8 which are the unique asymptotically flat station-
ary black holes in Einstein-Maxwell theory,9-13 one can get explicit expressions 14
for the area A, surface gravity K, angular velocity n, and electrostatic potential <P
of the black hole horizon in terms of the macroscopic conserved quantities of the
mass lvI, angular momentum J == Ma == M 2a*, and charge Q == MQ* of the hole,
using the value r + of the radial coordinate r at the event horizon as an auxiliary
parameter:

r+ = M + (M2 - a 2 - Q2)1/2 = M[l + (1 - a; _ Q;)1/2],

A = 47r(r! + a 2) = 47rM2[2 - Q; + 2(1 - a; _ Q;)1/2],

K = 47r(r+ - M) = ~M-l[l + (1- ~Q;)(l- a; _ Q;)-1/2j-l,

A
n= 4:a = a*M- 1 [2 - Q; + 2(1 - a; _ Q;)1/2r 1,
47rQr+ 1 + (1 - a: _ Q:)1/2
<P = - A - = Q* 2 _ Q; + 2(1 _ a; _ Q;)l/2 . (3)

Here a* = a/ M = J / Jl.1 2 and Q* = Q / M are the dimensionless angular mo-

mentum and charge parameters in geometrical units. For a nonrotating uncharged
 1505

stationary black hole (described by the Schwarzschild metric), a* = Q* = 0, so

T+ = 2M, A = 167fM 2 , /'i, = M/T~ = 1/(4M), f2 = 0, and if> = O.
Then Hawking's black hole emission calculation 5 ,6 for free fields gives the ex-
pected number of particles of the jth species with charge qj emitted in a wave mode
labeled by frequency or energy w, spheroidal harmonic l, axial quantum number or
angular momentum m, and polarization or helicity p as

(4)
Here the upper sign (minus above) is for bosons, and the lower sign (plus above)
is for fermions, and rjwl mp is the absorption probability for an incoming wave of
the mode being considered.
From the mean number Njwlmp and the entropy Sjwlmp per mode, one can sum
and integrate over modes to get the emission rates of energy, angular momentum
(the component parallel to the black hole spin axis), charge, and entropy by the
black hole:

(5)

dJrad
dt
dJ
dt
1
27f
L J
j,l,m,p
mNjwlmpdw, (6)

dQrad
dt
dQ
dt
1
27f
L J qjNjwlmpdw, (7)
j,l,m,p

dSrad
dt 2~ L
jJ,ffi,P
J Sjwlmpdw. (8)

Here M, J, and Q (without subscripts) denote the black hole's energy, angular mo-
mentum, and charge. By the conservation of the total energy, angular momentum,
and charge, the black hole loses these quantities at the same rates that the radiation
gains them.
This is not so for the total entropy, which generically increases. The black hole
entropy changes at the rate

-
dt
bh
dS- = - 1 L
27f j,l,m,p
J [27f/'i, -1 (w - mf2 - qj<P)]Njwlmpdw, (9)

and by using Eq. (4), one can show that the total entropy S = Sbh + Srad (black
hole plus radiation) changes at the rate

dS = 1
-d
t
- L "
27f.],l,ffi,p
J [ dw ±In(I±Njwlmp)+Njwlmpln l+ ( · 1 - r±N
r Jwlmp
jwlmp
Jwlmp
)] .

(10)
1506

For the emission of ns species of two-polarization massless particles of spin s

from a Schwarzschild black hole (nonrotating and uncharged) into empty space,
numerical calculations 14- 16 gave

dErad
-- =
dM
-- d = 10- 5M- 2( 8.1830nl/2 + 3.3638nl + 0.3836n2 ) , (11)
dt t

dSrad
- - = 10 -3 M -1( 3.371Onl/2 + 1.2684 nl + 0 .1300n2,) (12)
dt

dSbh
- - = -10- 3 M- 1 ( 2.0566nl/2 + 0.8454nl + 0.0964n2 ) . (13)
dt

3. The Generalized Second Law

Even if a black hole is not emitting into empty space, there are strong arguments
that the total entropy of the black hole plus its environment cannot decrease. This is
the Generalized Second Law (GSL). Bekenstein first conjectured it when he proposed
that black holes have finite entropy proportional to their area,17 and he gave various
arguments on its behalf, though it would have been violated by immersing a black
hole in a heat bath of sufficiently low temperature if the black hole could not emit
radiation. 1
Once Hawking found that black holes radiate,5,6 he showed that the GSL held
for a black hole immersed in a heat bath of arbitrary temperature, assuming that
the radiation thermalized to the temperature of the heat bath. Zurek and Thorne,18
and Thorne, Zurek, and Price,19 gave more general arguments for the GSL without
this last assumption. Their arguments were later fleshed out in a mathematical proof
of the GSL for any process involving a quasistationary semiclassical black hole. 2o
Other proofs of the GSL have also been given. 21 - 24
With some exceptions,22,24 these proofs so far generally have two key assump-
tions: (1) The black hole is assumed to be quasistationary, changing only slowly
during its interaction with an environment. It has been conjectured 19 that the GSL
also holds, using the Bekenstein-Hawking AI4 formula for the black hole entropy,
even for rapid changes in the black hole, but this has not been rigorously proved.
(2) The semiclassical approximation holds, so that the black hole is described by
a classical metric which responds only to some average or expectation value of the
quantum stress-energy tensor. This allows the black hole entropy to be represented
by AI4 of its classical horizon.
Other reviews 25- 34 have given more details of Hawking radiation and black hole
thermodynamics.
 1507

References
1. J. M. Bardeen, B. Carter, and S. W. Hawking, Comm. Math. Phys. 31, 161 (1973).
2. B. Carter, in Black Holes, eds. C. DeWitt and B. S. DeWitt (Gordon and Breach,
New York, 1973), p. 57.
3. S. W. Hawking, Phys. Rev. Lett. 26, 1344 (1971).
4. W. Israel, Phys. Rev. Lett. 57, 397 (1971).
5. S. W. Hawking, Nature 248, 30 (1974).
6. S. W. Hawking, Comm. Math. Phys. 43, 199 (1975).
7. R. P. Kerr, Phys. Rev. Lett. 11, 237 (1963).
8. E. T. Newman, E. Couch, K. Chinnapared, A. Exton, A. Prakash, and R. Torrence,
J. Math. Phys. 6, 918 (1965).
9. W. Israel, Phys. Rev. 164, 1776 (1967).
10. B. Carter, Phys. Rev. Lett. 26, 331 (1971).
11. S. W. Hawking and G. F. R. Ellis, The Large Scale Structure of Space-Time (Cam-
bridge University Press, Cambridge, 1973).
12. D. C. Robinson, Phys. Rev. Lett. 34, 905 (1975).
13. P. O. Mazur, J. Phys. A15, 3173 (1982); hep-th/0101012.
14. D. N. Page, Phys. Rev. D13, 198 (1976).
15. D. N. Page, Phys. Rev. D14, 3260 (1976).
16. D. N. Page, Phys. Rev. Lett. 50, 1013 (1983).
17. J. D. Bekenstein, Ph.D. thesis, Princeton University (1972); Lett. Nuovo Cimento 4,
737 (1972); Phys. Rev. D7, 2333 (1973); Phys. Rev. D9, 3292 (1974).
18. W. H. Zurek and K. S. Thorne, Phys. Rev. Lett. 54, 2171 (1985).
19. K. S. Thorne, W. H. Zurek, and R. H. Price, in Black Holes: The Membrane Paradigm,
eds. K. S. Thorne, R. H. Price, and D. A. MacDonald (Yale University Press, New
Haven, 1986), p. 280.
20. V. P. Frolov and D. N. Page, Phys. Rev. Lett. 71, 3902 (1993).
21. R. D. Sorkin, in Tenth International Conference on General Relativity and Gravitation,
Padua, 4-9 July, 1983, Contributed Papers, vol. II, eds. B. Bertotti, F. de Felice, and
A. Pascolini (Roma, Consiglio Nazionale Delle Ricerche, 1983), p. 734.
22. R. D. Sorkin, Phys. Rev. Lett 56, 1885 (1986).
23. R. M. Wald, Quantum Field Theory in Curved Spacetime and Black Hole Thermody-
namics (University of Chicago Press, Chicago, 1994).
24. R. D. Sorkin, in Black Holes and Relativistic Stars, ed. R. M. Wald (University of
Chicago Press, Chicago, 1998), p. 177; gr-qc/9705006.
25. R. M. Wald, Living Rev. Rel. 4, 6 (2001).
26. P. Majumdar, Pramana 55, 511 (2000), hep-th/0009008; hep-th/0011284; hep-
th/0110198; gr-qc/0604026.
27. C. Kiefer, in The Galactic Black Hole, eds. H. Falcke and F. W. Hehl (lOP Publishing,
Bristol, 2002), astro-ph/0202032.
28. W. Israel, Lect. Notes Phys. 617, 15 (2003).
29. T. Jacobson, gr-qc/0308048.
30. T. Damour, hep-th/0401160.
31. S. Das, Pramana 63, 797 (2004), hep-th/0403202.
32. D. V. Fursaev, Phys. Part. Nucl. 36, 81 (2005), gr-qc/0404038.
33. D. N. Page, New J. Phys. 7, 203 (2005), hep-th/0409024.
34. S. F. Ross, hep-th/0502195.
 ENTROPY FROM CONFORMAL HORIZON STATES IN
D-DIMENSIONAL SPHERICAL, TOROIDAL, AND HYPERBOLIC
ANTI-DE SITTER BLACK HOLES

JOSE P. S. LEMOS and GONQALO A. S. DIAS

Centro Multidisciplinar de Astrofisica - CENTRA
Departamento de Fisica, Instituto Superior Tecnico,
Universidade Tecnica de Lisboa,
Av. Rovisco Pais 1, 1049-001 Lisboa, Portugal
[email protected], [email protected]

A new method consisting in conformal techniques is presented in order to compute the

entropy of static, electrically charged, black holes with spherical, toroidal, and hyperbolic
compact and oriented horizons, in D spacetime dimensions.

The entropy of a black hole, given by the Bekenstein-Hawking 1,2 formula

S = A/(4G) (kBc 3 /n), has had no statistical description so far 3 . Following Solo-
dukhin's work for computing the entropy for spherical black holes in general rela-
tivity 4 , the entropy of Gauss-Bonnet black holes was determined in 5. Here a gen-
eralization of Solodukhin's method is given, for static, electrically charged, black
holes with spherical, toroidal, and hyperbolic compact and oriented horizons, in D
spacetime dimensions (see 6 for more details). These black holes live in an anti-de
Sitter spacetime, i.e., a spacetime with negative cosmological constant. The method
consists in a redefinition of the variables in the metric, by considering the radial co-
ordinate as a scalar field. Then its fluctuations are taken as the dynamical variable of
the theory. Thus, a possible candidate for the horizon states, in spherical symmetric
black holes, is indeed the set of degrees of freedom of spherically symmetric fluctu-
ations of the (D - 2)-dimensional horizon sphere. Hence one realizes a scalar theory
on the horizon by performing a 2 + (D - 2) dimensional reduction, where the (D - 2)
dimensions are in the angular coordinates, obtaining a 2-dimensional effective scalar
field theory. This theory is then seen to be a conformal theory in an infinitesimally
small vicinity of the horizon. The corresponding conformal symmetry will then have
conserved charges, associated with its infinitesimal conformal generators, which will
generate a classical Poisson algebra of the Virasoro type. Shifting the charges and
replacing Poisson brackets by commutators, one recovers the usual form of the Vi-
rasoro algebra, obtaining thus the level zero conserved charge eigenvalue Lo, and a
nonzero central charge c. The entropy is then obtained via the Cardy formula 7 . In
this analysis 6 one sees that one of the assumptions by Solodukhin 4 , namely that
the horizon topology is that of the D - 2-dimensional sphere, SD-2, can be lifted,
and that all the previous outline can be applied to this general class of black holes
of spherical, toroidal, and hyperbolic topologies. Spacetime is assumed to be time
dependent, due to fluctuations, but still keeping its original topology. The metric is

1508
 1509

written generally as 6
l
d(}"
2 ° 2 + (dx
= -f(x 1 )(dx) )2 2 ( k )2
f(x l ) +r (x) drl D_ 2 , (1)

where x stands collectively for the time and spatial coordinates x = (XO, Xl), or,
when convenient, for the null coordinates x = (X+, x-), where x+ = xO + Xl, and
x- = xO - xl, and the function f(x) has a behaviour near the horizon, given by
f(Xh) = 0, of the kind f(x l ) = ~: (xl - x~) + 0 ((Xl - x~)2), and drl'D_2 is the
line element of the D - 2 dimensional horizon surface, where k gives the topology,
k = 0 spherical, k = 1 toroidal (planar), and k = -1 hyperbolic. The reduction of
the Einstein-Maxwell action using (1) is given by,

1= - ~t~ J d2x F [rD-2 R + (D - 3)(D - 2)r D- 4 (V'r)2 +

k(D - 3)(D - 2)r D- 4 - 2Ar D- 2 + 81l' Gp2 r D-2] , (2)

where R is the scalar curvature in two dimensions, I is the detrminant of the

2-dimensional metric lab, A is the cosmological constant, "E. is the area of lJ-2
the corresponding unitary surface, a factor that comes from the integration of
V
the angular coordinates 6, and P = - ~ (D- 3dSD - 2 ) rDQ-2 is the modulus of
the radial electric field. After redefining the coordinate r as the dilaton field ¢,
¢ = 2lq (g=~) C~ (r2/rh) D:;2, where q is a constant to be determined later,

and C = ~€c/ (g=~), and performing the conformal transformation lab =

D-3 1 1>
(¢¢h) D-2
q2
e (E=n 1>h 1ab, where ¢h = 2lqC~ (g=n (rh)D:;2 is the classical value
of the field on the horizon. Then the action (2) becomes a Liouville type action,

where '\7 is the derivative associated with the new metric 1ab, R is the scalar curva-
ture associated with 1, derived from the original scalar R, and the potential U(¢)
is a function of the fields and the parameters of theory 6 . The trace of the energy
momentum tensor Tab, T::: = 1 ab T ab, is given by T::: = ~ q2 (g=~) ¢h 0 ¢ - U( ¢),
showing it is not conformally invariant everywhere. With a change of coordinates
given by z = rl /(~) = ~~ In(xl - x~), and identifying xO = t, the equation for ¢
becomes, at z ---+ -00,

(4)

a simple d'Alembert wave equation, which in light cone coordinates, z+ = t + z

and z- = t - z, can be written as 8_ 8+ ¢ = 0, which is conformally invariant.
When the equation of motion is obeyed, this implies that the trace of Tab is now
T_+ + T+_ = O. Thus we have a conformal field theory on the horizon. Due to
1510

the energy-momentum tensor conservation ELT++ = 0, one can define infinite con-
served charges on the horizon Qf, = f dz ~(z) T++, which respect a Poisson algebra
{Qf,,,Q6} = Q[e",6l + (q2th f fC[6,6]dz, where [6,6] = 6~; - 6~~, with
, == oz, and C[6, 6] == (~~ + ,Bhl~d(~~ + ,Bh 6)' - (~~ + ,Bhl~d'(~~ + ,Bh 6). How-
1 1

ever, one finds necessary to compactify in a circle of length L the coordinate z, so

that it lies in the interval -~ ::; z ::; ~, with Qn = 2~ f~D2 dzei¥nz T++, where
also the ~(z) = L an e ;;nz are split into Fourier modes 4,6 , and T++ is evaluated
i2

at t = O. Setting L = 27r,Bh, the generators Ln = Qn + lC2 6n ,a are defined. Swap-

ping Poisson brackets for commutators through [ , ] = in{ , }, the usually called
Virasoro operators Ln satisfy the usual Virasoro algebra,

(5)

with c 127rq4 (g=~) 2 ¢~. Having determined the zero mode operators as
4 6
La = c/12 - , one finds, through the Cardy formula 7 (see also 8) Sconf

27r J~ (La - 24) , the conformal entropy of the

C
black hole
2
2 4 D-3 2
Sconf = 27r q ( D _ 2) ¢h' (6)

Comparing with the gravitational, S = 27r q2 (g=n ¢~, the equality Sconf =
[q2 (g=~)7r] S demands q2 = (g=~) ~.
One sees, thus, that the contributions of the different topologies are absorbed
by the potential of the equivalent field theory, becoming irrelevant. This classical
conformal field theory has a non-zero central charge, which upon quantization,
through Cardy formula gives the entropy of the black hole.
Acknowledgments
This work was partially funded by FCT, Portugal, through project
POCTI/FIS/57552/2004. GASD is supported by grant SFRH/BD/2003 from FCT.

References
1. J. Bekenstein, Phys. Rev. D 7, 2333 (1973).
2. S. Hawking, Comm. Math. Phys. 43, 199 (1975).
3. J. P. S. Lemos, in Proceedings of the Fifth International Workshop on New Worlds in
Astroparticle Physics, eds. Ana Mounio et a!. (World Scientific, Singapore 2006), p.
71; gr-qc/0507101.
4. S. N. Solodukhin, Phys. Lett. B 454, 213 (1999).
5. M. Cvitan, S. Pallua, and P. Prester, Phys. Lett. B 555, 248 (2003).
6. G. A. S. Dias and J. P. S. Lemos, Phys. Rev. D 74, 044024 (2006).
7. J. Cardy, Nuc!. Phys. B 270, 186 (1986).
8. A. Strominger, JHEP 9802, 009 (1998).
 THERMODYNAMIC GEOMETRIES OF BLACK HOLES

JAN E. AMAN, INGEMAR BENGTSSON and NARIT PIDOKRAJT

Fysikum AlbaNova,
Stockholm University,
SE-106 91 Stockholm, Sweden
[email protected], [email protected], [email protected]

JOHN WARD
Department of Physics,
Queen Mary, University of London,
Mile End Road, London E1 4NS, United Kingdom
and
Theory Division, CERN,
CH-1211 Geneve 23, Switzerland
[email protected]

We report outcomes from the geometrical studies (Ruppeiner and Weinhold geome-
tries) of thermodynamics of certain black hole families in various spacetime dimensions
e.g. BTZ, general relativity and higher-dimensional Einstein-Maxwell gravity. So far the
most physically suggestive result occurs for the Kerr black hole in D :::> 5 in which ther-
modynamic stability is found to be encoded in the curvature of the Ruppeiner metric.
This is consistent with the known result in the literature.

1. Introduction
Some 27 years ago George Ruppeiner proposed a geometrical a way of studying
thermodynamics of equilibrium systems. 1 He considers the Hessian of the Gibbs
surface to be the metric on thermodynamic state space of the system in equilibrium.
We call such a metric the Ruppeiner metric, defined as
g~=-8J)jS(X), X=(U,Na); a=1,2, ... ,n (1)
where U is the system's internal energy (which is the mass in black hole (BH) ther-
modynamics), S(X) is an entropy function of the thermodynamic system one wishes
to consider and Na stand for other extensive variables, or-in our application-
mechanically conserved charges of the system. Remarkably, certain aspects of ther-
modynamics and statistical mechanics of the system under consideration are en-
coded in the Ruppeiner metric in that its curvature is flat for a non-interacting
statistical mechanical system like the ideal gas, whereas curvature singularities are
a signal of critical behavior. Furthermore the signature of the metric corresponds to
the positivity (negativity) of the specific heats of the system under consideration.
More details on the application of the Ruppeiner geometry to numerous systems
can be found in Ref. 2 Originally Ruppeiner geometry was developed in the context
of thermodynamic fluctuation theory, for systems in canonical ensembles. However
most BHs have negative specific heats and are described microcanonically. Despite

aWe will refer to it as the Ruppeiner geometry.

1511
1512

this, we have found that the Ruppeiner geometry of BHs is often surprisingly simple
and elegant. So far the most physically suggestive case in our investigations is the
Kerr BH in arbitary dimension (D) where the curvature singularities are a signal of
thermodynamic instability as found earlier by Emparan and Myers 4 using another
method. We present this briefly in Section 2.
The Ruppeiner geometry is conformal to the so-called Weinhold geometry via a
simple relation

(2)
where dsW is the line element on the space of the Weinhold geometry and T is the
temperature of the system. Intrinsically, the Weinhold metric 3 is defined as
gIr = E)JJj U(5, Na). (3)
This geometry does not have the same physical meaning as its counterpart but it
is useful when computations in Ruppeiner coordinates are not easily tractable.
In this paper we list results obtained from several Ruppeiner studies by us
including the 4D general relativity BH plus the BTZ BH,5 an extension to higher
dimensional spacetimes,6 the flatness theorem in thermodynamic geometries 7 and
an ongoing work on 4D Einstein-Maxwell dilaton BH.8 Due to the limitation on
the size of this article we cannot list all the references in this field, only the most
relavant ones to this article. We choose to present only the Kerr BH in a somewhat
brief manner and the rest of the outcomes in the table in the last section.

2. Application to Kerr BH
In D dimensions the Kerr BH has the mass function
D - 2 D-3 ( 4J2) 1/(D-2)
M = - - 5 D-2 1+ -52 (4)
4
The Weinhold metric of this BH is flat, whereas the Ruppeiner metric is curved
with the following curvature scalar
D-5 J2
1-12---
R = _.!. D - 3 52 (5)
5(1 _4 D - 5J )
D - 3 52
2
(1 + 4D - 5J )
D - 3 52
2
.

It is easy to see that in 4D the curvature scalar diverges along the curve 4J 2 = 52
consistent with the previous result. 5 Remarkably, in D > 5 we have a curvature
divergence but not in the limit of extremality, rather at
4J2 = D - 3 52 . (6)
D-5
This is where Emparan and Myers4 suggest that the Kerr BH becomes unstable and
changes its behavior to be like a black membrane. This is also where the temperature
of the higher-dimensional (D ?: 6) Kerr BH reaches its minimum. Full details on
the Ruppeiner study of Kerr and Kerr double-spin BHs can be found in Ref. 6
 1513

3. Summary and Outlook

Geometrical (Ruppeiner) theory of thermodynamics gives an alternative and elegant
route to obtain information of thermodynamics through Riemannian geometry, as
has been proven in various contexts. Since statistical mechanics of BHs are yet to be
established, application of the Ruppeiner theory to BH thermodynamics provides
a new perspective on this subject. Although we have so far obtained few physically
suggestive outcomes from the investigations, the geometrical patterns we have ob-
served (see table below) may become important when quantum theory of gravity is
better understood in the future.

BH family Ruppeiner geometry Weinhold geometry

BTZ curved flat
D-Dim Reissner-Nordstrom flat curved
D-Dim Kerr curved flat
5-Dim Kerr (double-spin) curved curved
4-Dim Kerr-Newman curved curved
4-Dim Einstein-Maxwell Dilaton flat curved

Acknowledgments
NP is supported by Doktorandtjanst of Stockholm University. The organizers of
MGll are appreciated for their efforts behind the scence. NP is also thankful to
Isabella Malmnas for making the trip to Berlin a very exciting and memorable one.

References
1. G. Ruppeiner, Phys. Rev. A20, 1608 (1979).
2. G. Ruppeiner, Rev. Mod. Phys. 67, 605 (1995).
3. F. Weinhold, J. Chern. Phys. 63, 2479 (1975).
4. R. Ernparan and R. C. Myers JHEP 0309, 025 (2003).
5. J. E. Arnan, 1. Bengtsson and N. Pidokrajt, Gen. ReI. Grav. 35,1733 (2003).
6. J. E. Arnan and N. Pidokrajt, Phys. Rev. D73, 024017 (2006).
7. J. E. Arnan, 1. Bengtsson and N. Pidokrajt, Gen. ReI. Grav. 38, 1305 (2006).
8. J. E. Arnan, 1. Bengtsson, N. Pidokrajt and J. Ward, in preparation
This page intentionally left blank
Alternative Black
Hole Models
This page intentionally left blank
 ON QUANTUM EFFECTS IN THE VICINITY OF WOULD-BE
HORIZONS*

PIOTR MARECKI
IT?, Universitiit Leipzig
D-0400g Leipzig, Germany [email protected]

We present a method based on the so-called Quantum Energy Inequalities, which allows
to compare, and bound, the expectation values of energy-densities of ground states of
quantum fields in spacetimes possessing isometric regions. The method supports the
conclusion, that the Boulware energy density is universal both: at modest (and far)
distances from compact spherical objects, and close to the would-be horizons of the
gravastar /QBHO spacetimes. It also provides a natural consistency check for concrete
(approximate, numerical) calculations of the expectation values of the energy-momentum
tensors.

1. Introduction
One of the most important aspects of Quantum Field Theory in curved spacetimes is
the existence of non-vanishing expectation values of the energy-momentum-tensor
operator in virtually all states of the quantum fields. Yet in concrete cases, for
specific spacetimes and states believed to be of physical interest, the calculation of
these expectation values is notoriously difficult. Not only the classical solutions of
the wave equation, FJ (x, t), need to be known a , but also sums of (absolute) squares
of these solutions over their index I need to be known at every point. Only in few
precious cases has this been possible, often in approximate sense only. Here we report
on a method, which allows to use the result known for one spacetime to approximate
the energy densities on another spacetime, which is isometric to the initial one in
some region only. As an application, the known properties of the Boulware state
in the Schwarzschild geometry provide a bound on the energy densities of ground
states in spacetimes of compact, spherically symmetric objects.
Let us take the spherically symmetric spacetimes, with the (general) metric
2
2 2 dr 2 2 2
(1)
ds = f(r)dt - h(r) - r (de + sin ed¢),

with the suitable functions f and h. The main object of interest is the (temporally
smeared) energy density operator,

p(w, x) = J dt Tab (x, t) uau b W 2 (t). (2)

where the operator Tab has been constructed by point-splitting w.r.t. a Hadamard
parametrixb . The function, w(t) is smooth, compactly supported, and normalized

*This research has been supported by the Kosicuszko Foundation

aWhich already exceeds our capabilities, for instance, in the Schwarzschild spacetime
bWe mean the standard construction employed in most concrete calculations, and note, that such
an operator is locally covariant.1. 2

1517
1518

as a probability density, J dt w 2 (t) = 1, while u a denotes the tangent vector of the

observer measuring the energy density.
We shall employ the method of Quantum Energy Inequalities,2-4 which asserts,
that all Hadamard states 1/J must necessarily fulfill

(g(w, x))1jJ - (g(w, x))c ~ -Qc[w, x], (3)

where G indicates the ground state of the quantum field, and

1
Q c = :;.fo roo dw 1 00

-00 dt ds-iw(t-s)
e w () t w ()
s [atas
fer) - ('t7X) ~ s, y~)I y=x'
('t7Y)i] W2c( t, x,
v i v

(4)
Here, wf(t, x, s, if) stands for the two-point function of the ground state C • The func-
tionals Qc[w, x] can conveniently be expressed by the respective mode-sums:

Qc[w, xl
~ 1
= 167r 3
[00
J
[00 dp [p2
dw Jo P f - H 2][00
r2 ar(Hr ar ) ~(21 + 1)IR 1p (r)1 2] . g(w + p),
o
(5)

where Fl(x) = ~Rlw(T) Yim(B, ¢) are the solutions of the wave equation corre-

sponding to the frequency w, and g(w) = IJ~CXJ dt w(t)eiwf. Here also H(T) = VTFi,
is derived from the metric functions.
The functional Qc, which is finite and always positive, is independent of the
quantum state 1/J, which is remarkable, as the variety of quantum states is enormous.
Even more surprising are the (typical) properties of Qc: its value falls off rapidly
as the time of measurement gets longer, i.e. as the support of wet) gets larger.
Thus, substantial sub-ground-state energy densities can be produced by quantum
fields only for short durations of time, and it is impossible to keep these densities
sub-ground for longer time periods.

2. Energy-densities in locally isometric spacetimes

Suppose that two static spacetimes are isometric in a region outside of a sphere
at T = R. If we focus on operators supported in the isometric region, than we can
establish two QEI's:1,2 firstly with G1 as the reference state (with G2 playing the
role of 1/J), and secondly with the exchanged role of the states. This provides bounds
from above and below on the difference of the energy densities:

QGl[w,x] ~ (g(x))Gl - (g(X))C2 ~ -QG2[w,.i]. (6)

If we were able to streach the support of w( t) arbitrarily, then the above inequalities
would force the difference (g(X))C2 - (O(x))Gl to vanish! However, there is a QFT

CThe QEI are proved to be true also with respect to any other Hadamard state,2 that is, we can
replace the ground state G by any other Hadamard state (although such a replacement changes
also the right-hand side of the inequality).
 1519

subtelty: a priori the states denoted by 1jJ in (3) should be expressible as density ma-
trices in the Fock space constructed upon G. In no way is this possible for 1jJ = G2
and G = GI, as these are states providing expectation values of observables from
completely incompatible algebras. Nonetheless, the inequalities hold (with expec-
tation values calculated in the usual way) if the smallest double cone containing
the support of the operators (in this case: the segment of the curve on which we
measure the energy-density) does remain in the isometric region d .
x
Thus, the distance between and the boundary of the isometric region limits
the measurement duration e , T, and allows a non-vanishing \Q(X))G2 - \Q(x))G1.

3. Remarks and outlook

The bounds (6) can be directly employed as a new consistency check to be fulfilled
in concrete calculations of energy-momentum tensors of quantum fields. 5 ,6 They are
especially tight if the allowed times of measurement are long, which happens either:
far away (in terms of the geodesic distance) from the boundary of the isometric
region, or if the boundary is close to an horizon. In these cases, the functionals
QG[w, x] depend only on the low frequency asymptotics of the mode-sums. In turn
this hints at universality of certain results of QFT in curved spacetimes, notably,
our bounds indicate l that the energy density of the Boulware state is universal
both: far away from the horizon (always) and close to the horizon, for spacetimes
isometric with Schwarzschild up to small distances from the horizon, such as the
gravastar or QBHO spacetimes. 7 ,8

Acknowledgments
The author would like to thank Professor Pawel O. Mazur for the opportunity to
present this research at the BHT5 section of the 11th Marcel Grossman Meeting.
The financial support of the Kosciuszko Foundation is also gratefully acknowledged.

References
1. P. Marecki, Phys. Rev. D73, 124009 (2006); gr-qc/0507089.
2. C.J. Fewster and M.J. Pfenning J. Math. Phys. 47,082303 (2006).
3. T. A. Roman Pmc. 10th Marcel Grossmann Meeting; gr-qc/0409090.
4. M.J. Pfenning and L. H. Ford Phys. Rev. D57, 3489 (1998).
5. P. R. Anderson, W.A. Hiscock, and D.A. Samuel Phys. Rev. D5l, 4337 (1995).
6. J. G. Jensen, B.P. McLaughlin and A.C. Ottewill Phys. Rev. D45, 3002 (1992).
7. G. Chapline, et al. Int. J. Mod. Phys. A18, 3587 (2003).
8. P. O. Mazur and E. Mottola Pmc. Nat. Acad. Sci. 111,9545 (2004).

dIn other words: all causal curves, with their ends attached to events of the beginning and the end
of the energy measurement, should not leave the isometric region.
x
eThe maximal duration, T, is dependent, and grows together with the distance from the bound-
ary, so that the bounds (6) get tighter.
 STABLE DARK ENERGY STARS:
AN ALTERNATIVE TO BLACK HOLES?

FRANCISCO S. N. LOBO
Centro de Astronomia e Astrofisica da Universidade de Lisboa,
Campo Grande, Ed. C8 1749-016 Lisboa. Portu9al
fiobo@cosmo·fis.fc.ul.pt

In this work, a generalization of the Mazur-Mottola gravastar model is explored,

by considering a matching of an interior solution governed by the dark energy equation
of state, W '= pip < -1/3, to an exterior Schwarzschild vacuum solution at a junction
interface, situated near to where the event horizon is expected to form. The motivation
for implementing this generalization arises from the fact that recent observations have
confirmed an accelerated cosmic expansion, for which dark energy is a possible candidate.

Keywords: Gravastars; dark energy.

Although evidence for the existencc of black hole8 is very convincing, a certain
amount of sceptici8m regarding the physical reality of singularities and event hori-
zons is still encountered. In part, due to this scepticism, an alternative picture for
the final state of gravitational collapse has emerged, where an interior compact ob-
ject is matched to an exterior Schwarzschild vacuum spacetime, at or near where
the event horizon is expected to form. Therefore, these alternative models do not
posse88 a singularity at the origin and have no event horizon, as its rigid 8urface is
located at a radius slightly greater than the Schwarzschild radius. In particular, the
gravastar (gravitational vacuum star) picture, proposed by Mazur and Mottola,l has
an effective phase transition at/near where the event horizon is expected to form,
and the interior is replaced by a de Sitter condensate. In this work, a generalization
of the gravastar picture is explored, by considering a matching of an interior solu-
tion governed by the dark energy equation of state, w == p/ p < -1/3, to an exterior
Schwarzschild vacuum solution at a junction interface. 2 This new emerging picture
consisting of a compact object resembling ordinary spacetime, in which the vac-
uum energy is much larger than the cosmological vacuum energy, shall be denoted
as a "dark energy star".3 We emphasize that the motivation for implementing thi8
generalization arises from the fact that recent observations have confirmed an accel-
erated expansion of the Universe, for which dark energy is a possible candidate. The
dynamical stability of the transition layer of these dark energy stars to linearized
spherically symmetric radial perturbations about static equilibrium solutions was
further explored. 2
Consider the interior spacetime, without a loss of generality, given by the fol-
lowing metric, in curvature coordinates

(1 )

1520
 1521

where 1>(r) and mer) are arbitrary functions of the radial coordinate, r. The function
mer) is the quasi-local mass, and is denoted as the mass function.
The Einstein field equation, G"v = 87rT"v provides the following relationships

(2)

(3)

where' = d/dr. per) is the energy density, Pr(r) is the radial pressure, and pt(r)
is the tangential pressure. Equation (3) corresponds to the anisotropic pressure
Tolman-Oppenheimer-Volkoff (TOV) equation. The factor 1>' (r) may be considered
the "gravity profile" as it is related to the locally measured acceleration due to grav-
ity, through the following relationship: A = )1 - 2m(r)/T 1>'(r). The convention
used is that 1>' (1') is positive for an inwardly gravitational attraction, and negative
for an outward gravitational repulsion.
Now, using the dark energy equation of state, Pr = wp, and taking into account
Eqs. (2), we have the following relationship

1>'(1') = m + wTm' . (4)

l' (1' - 2m)
One now has at hand four equations, namely, the field Eqs. (2)-(3) and Eq. (4),
with five unknown functions of 1', i.e., peT), Pr(T), Pt(T), 1>(1') and meT). We shall
consider the approach by choosing a specific choice for a physically reasonable mass
function meT), thus closing the system.
This interior solution is now matched to an exterior Schwarzschild solution at a
junction interface, a. The surface stresses of the thin shell are given by

2M
(J = -
1 (/
47ra V
1 - -;;: + a. 2 - V
. /
1-
2m
---;;: + a. 2 ) , (5)

p _ __ 1(1 - M
a
+ a.2 +..
aa _ 1+ wm' - !II:.
a
2
+ a2 + au + it 1-2m/a
m'(l+w) )
(6)
-87ra .fr_2M+a2 11_2m+a2 '
V a V a

where the overdot denotes a derivative with respect to proper time, T. (J and P
are the surface energy density and the tangential pressure, respectively.2,4 The
dynamical stability of the transition layer of these dark energy stars to linearized
spherically symmetric radial perturbations about static equilibrium solutions was
further explored using Eqs. (5)-(6) (sec 2 for details). Large stability regions were
found that exist sufficiently close to where the event horizon is expected to form, so
that it would be difficult to distinguish the exterior geometry of these dark energy
stars from astrophysical black holes.
Several relativistic dark energy stellar configurations may be analyzed by im-
posing specific choices for the mass function. 2 For instance, consider the following
1522

mass function
bor 3
m(r) = 2(1 + 2bor2) , (7)

where bo is a non-negative constant. The latter may be determined from the regu-
larity conditions and the finite character of the energy density at the origin r = 0,
and is given by bo = 87fPc/3, where Pc is the energy density at r = 0. This choice
of the mass function represents a monotonic decreasing energy density in the star
interior. Now the function ([>(r) may be deduced from Eq. (4), and is given by

(8)

It can be shown that the gravity profile is negative, ([>' (r) < 0, for w < - (1 +
2b or2)/(3 + 2b or2), indicating an outwardly gravitational repulsion,2 which is a
fundamental ingredient for gravastar models.
In concluding, a generalization of the gravastar picture was explored, by consid-
ering a matching of an interior solution governed by the dark energy equation of
state, w == p/ P < -1/3, to an exterior Schwarzschild vacuum solution at a junction
interface (An analogous case was analyzed in the phantom regime w < -1, in the
context of wormhole physics 5 ). Despite of the repulsive nature of the interior solu-
tion, large stability regions exist, with the transition layer placed sufficiently close
to where the event horizon is supposed to have formed. 2 It has also been argued
that there is no way of distinguishing a Schwarzschild black hole from a gravastar,
or a dark energy star, from observational data. 6 The dark energy stars outlined in
this paper may possibly have an origin in a density fluctuation in the cosmological
background, possibly resulting in the nucleation of a dark energy star through a
density perturbation. We would also like to state our agnostic position relatively
to the existence of dark energy stars,7 however, it is important to understand their
general properties to further understand the observational data of astrophysical
black holes.

References
1. P. O. Mazur and E. Mottola, "Gravitational Condensate Stars: An Alternative to
Black Holes," [arXiv:gr-qc/0l09035]; Proc. Nat. Acad. Sci. 111, 9545 (2004).
2. F. S. N. Lobo, Class. Quant. Grav. 23, 1525 (2006).
3. G. Chapline, "Dark energy stars," [arXiv:astro-ph/0503200].
4. F. S. N. Lobo and P. Crawford, Class. Quant. Grav. 22, 4869 (2005).
5. F. S. N. Lobo, Phys. Rev. D 71, 084011 (2005); Phys. Rev. D 71, 124022 (2005).
6. M. A. Abramowicz, W. Kluzniak and J. P. Lasota, Astron. Astrophys. 396 L31 (2002).
7. C. Cattoen, T. Faber and M. Visser, Class. Quant. Grav. 22 4189-4202 (2005).
 HORIZON NEWS FUNCTION AND QUASI-LOCAL ENERGY-MOMENTUM
FLUX NEAR BLACK HOLE

WU, YU-HUEI
School of Math, University o.f Southampton, SOI7 IB], UK
yw [email protected], [email protected]

From the 'quasi-loca\' definition of horizons, e.g. isolated horizon and dynamical horizon, the con-
sequence quasi-local energy-momentum near horizons can be observed by using the idea of frame
alignment. In particular, we find the horizon news function from the asymptotic expansion near hori-
zons and use this to describe the gravitational flux and change of mass of a black hole.

1. Introduction
In this paper we apply a similar method of asymptotic expansion near null infi nity
(Newman-Unti 6 ) to the geometry near a black hole to study gravitational radiation. The
defi nition of 'quasi-local' horizons (isolated horizon and dynamical horizon) are given by
Ashtekar and Krishnan. l,2 The covariant expression of quasi-local energy expression is
given by Nester-Witten two form 4 which refers to an implicit reference frame when ob-
serving the quasi-local quantities. What is a good reference frame near a strong gravitating
fi eld such as a black hole? To tackle this problem, we use the concept and idea of frame
alignment to fi nd the asymptotically constant spinors which is done by Bramsori for null
infi nity and try to move this work on the horizon in Section 2. We then be able to use the
spin frame ZAA = (AA' JiA) to give a defi nition of the energy-momentum for a rotating
dynamical horizon in Section 3. Particularly, we fi nd the horizon news function dominates
the gravitational radiation of a black hole in Section 4.

2. Constant spinors near quasi-local horizons

We fi nd the compatible conditions for asymptotically constant spinor AA mainly include the
Dougan-Mason's holomorphic conditions and a time-related condition. Different types of
quasi-local horizons require different conditions. 7 For example, the compatible conditions
of a rotating dynamical horizon are
o .0 0
1:>0Ao = 0, =? AO - fOAO = ° (1)
OOAg + (JOA? = 0, OOA? - JiOAg = 0, (2)
o - 0 (3)
PoAr = -OOAO
where include a time-related condition (1) and Dougan-Mason's holomorphic conditions
(2) and an extra condition (3). Next, we will focus on the issues of a rotating dynamical
horizon.

3. Quasi-local energy-momentum of a rotating dynamical horizon

By using the compatible constant spinor conditions for a rotating dynamical horizon, i.e.,
the Dougan-Mason's holomorphic conditions (2) in Section 2, we can get the quasi-local

1523
1524

momentum integral near a rotating dynamical horizon based on Nester-Witten two form is

I(r') = 4~ fs(p' ).,0>'0' + p).,l>'ddS (4)

= 4~ fs[-tLo)"8>'8, + O(r')JdSl;.
where we use p = O(r'), tL = tLo + O(r'), d Sl;. = limr,--->o dS r , and dSdv). By using
the result of the asymptotic expansion (NR7),7 we get the quasi-local momentum integral
on a rotating dynamical horizon

I(r l;.) = - 4~ f 2!o [\lfg - {Lo + TdtL6 + 0-' oa' 0) + oo 7T o + 7To7fo - 0-00-' oJ)"8>'8,dSl;..
(5)

4. Energy-momentum flux near a rotating dynamical horizon

Now, in order to make our calculation easier and use the approximate Kerr in Bondi co-
ordinate and the slow rotating Kerr-Vaidya as our basis. We make a coordinate choice
r l;. (v) = _.l/Lo
and make the assumption that 0-'0 = K,o = 0 to simplify our calculation of
flux formula. Here we will also need the time related condition (I) of constant spinor of
dynamical horizon in Section 2 and re-scale it. Therefore >-8 = O. We note that this rescal-
ing of spin frame is to chose a 'permissible' time so that the energy flux can only depend
on the news function. It's very tedious but straightforward to calculate the flux expression.
It largely depends on the non-radial NP equations and the second order NP coefficients.
We substitute these equations back into the energy-momentum flux formula to simplify our
flux expression (See 7 for the detail).
Apply time derivative on (4), we get

j(rl;.) = ~
47T
j{Lo)"8>'8,dSl;.. (6)

Here we can see the fact that {Lo is related with the mass loss of gain, hence it is the news
function. Integrate the above equation with respect to v and use {Lo = Tl;. (2) R/2, we then
have

(7)

Ashtekar's total flux formula 2 is

Fmatter + F grav = _1_

167T
r
LH
(2) RN d 3 V (8)

where F matter + F grav is equal to flux dI and d3 V = dr l;. dS on horizon. Therefore,

if N = 2)"8>.g" then our flux formula from equation (4) is completely the same with
Ashtekar-Krishnan's formula (8).
Apply time derivative on (5) and integrate the equation with respect to v, we have

(9)
 1525

where dv = dr A
rA
• The total flux of Ashtekar-Krishnan 2 in terms of NP in our gauge 7 is

Ftotal = -
47r
1 j [Icrl 2
+ 17r1 2 + <poo]N d3 V. (10)

In order to compare with Ashtekar's expression, therefore, if we choose N = 2>.g>-g, and

- ~2 ~ = 2, then (9) becomes
Eor 6.

4 j{[croao + 4r/:l7r07fo + <p8o] + ~[cro7r6 + ao 7f 6]}NdS/:ldr/:l. (11)

1
dI(r/:l) =
7r ~

However, here we have an extra term which is the coupling of the shear cro and 7ro.

5. Conclusion
The Nester-Witten two-form with the compatible conditions of constant spinors on 'quasi-
local' horizons and together with the results from the asymptotic expansion near' quasi-
local' horizons give us the quasi-local energy-momentum and flux expressions on 'quasi-
local' horizons. Dougan-Mason's holomorphic conditions tell us how to gauge fix the
quasi-local expression on each cross section of ' quasi-local' horizons. The time related
condition tells us how the quasi-local expressions change with time along horizon. The
news function which dominates the gravitational radiation near a rotating dynamical hori-
zon can be understood as the time derivative of the expansion of the incoming null tetrad
ito in equation (6). The gravitational flux for a rotating dynamical horizon is obviously
positive (mass gain) from our formula if the dominate energy condition holds. The shear
square term in equation (9) which is related with the gravitational radiation near dynamical
horizon in our formula matches the results of energy flux cross event horizon by using the
perturbation method [Hawking and Hartle]. The 7r square term is related with angular mo-
mentum contribution [See Ashtekar and Krishnan 2 ]. There's a shear and 7r coupling term
in my flux formula for a rotating dynamical horizon which is an extra term in Ashtekar-
Krishnan's formula.

References
1. Ashtekar, A., Beetle, C. and Lewandowski, J., 'Geometry of Generic Isolated Horizon', Class.
Quant. Grav. 19 1195-1225 (gr-qc/0I1067) (2002).
2. Ashtekar, A. and Krishnan, B., 'Dynamical Horizons: Energy, Angular Momentum, Fluxes and
Balance Laws', Class.Quant.Grav. 20 1031-1062 (gr-qc/0207080) (2003).
3. Bramson, B. D., 'The alinement of frames of reference at null infinity for asymptotically flat
Einstein-Maxwell manifolds', Proc. Roy. Soc. London A341, 451-461 (1975).
4. Dougan, A. J. and Mason, L. J., 'Quasi-local mass constructions with positive energy', Phys.
Rev. Lett. 67, 2119-22 (1991).
5. Krishnan, B., PhD thesis, Pennsylvania State University, (2002).
6. Newman, E. T. and Unit, T. W. J., 'Behavior of Asymptotically Flat Empty Spaces', J. Math.
Phys. 3, 891-901 (1962).
7. Wu, Yu-Huei, PhD thesis, University of Southampton, (2006).
 BLACK HOLES OR ETERNALLY COLLAPSING OBJECTS?

ABHAS MITRA
Theoretical Astrophysics, Bhabha Atomic Research Centre,
Mumbai -400085, India
* E-mail: [email protected]

NORMAN K. GLENDENNING
Nuclear Science Division 8 Institute for Nuclear and Particle Astrophysics
Lawrence Berkeley Laboratory, Berkeley, CA. 94720, U.S.A .
• E-mail: [email protected]

Formation of a Black Hole (BH) demands that the surface gravitational redshift of the
collapsing object, z -+ 00. We point out that as the collapse generated radiation quanta
would move in almost closed circular orbits for z » 1, they would get almost trapped
by the strong self-gravity. As the outward flux of the trapped radiation would attain its
Eddington value, by definition, catastrophic collapse would degenerate into a radiation
supported quasistatic state though, theoretically, the collapsing object would asymptot-
ically approach the z = 00 exact BH state.

Keywords: Black Holes; Eternally Collapsing Objects

1. Introduction
The strength of the self-gravitation of a collapsing object is indicated by
z = (1- R s /R)-1/2_ 1 (1)
where Rs = 2G M / c2 , G is the gravitational constant and c is the speed of light.
Physical General Relativistic (GR) gravitational collapse must be accompanied by
emission of radiation. 1 However, as long as z < V3 - 1, the emitted quanta nev-
ertheless manage to evade entrapment and move away to infinity. But when the
body would be so compact as to lie within its "photon sphere", i.e., R < (3/2)R s
or z > V3 - 1, then only the radiation emitted within a cone defined by semi-angle
Be:

. Be -_ -2-
sm v'27( 1 - Rs / R )1/2( Rs / R ) (2)

will be able to escape eventually.2,3 Radiation emitted in the rest of the hemisphere
would eventually return within the compact object. In fact, it may be seen that,
this gravitational trapping of the quanta is synonymous with the fact that they
may move in closed circular orbits for for z > V3 - 1. At high z, R ~ Rs and from
Eq.(2), one can see that, sin Be ---t Be ~ ()27/2)(1 + z)-l .2,3 Hence the solid angle
of escaping radiation is
271r
ne ~ 7rB2e ~ -(1
4 + z)-2 (3)

The chance of escape of radiation therefore decreases as Oe/27r ~ (27/8)(1 + z) -2.

Consequently, as the collapse generates internal heat/radiation, the energy density

1526
 1527

of trapped radiation and associated outward heat flux

(4)
In the following, we show that this outward heat flux must dynamically halt the
catastrophic collapse.

2. Eddington Luminosity
For an object of mass M, the locally defined Eddington luminosity is
47rCMc
Led = (l+z) (5)
""
But the observed value of the same is L';d = (1 + Z)-2 Led. Essentially, Led corre-
sponds to a critical comoving outward heat flux of
Led CM
qed = 47rR2 = ""R2(1+z) (6)

Using Eqs. (4) and (5), we see that, in this regime of z » 1,

qtrap
a=--~---
(1 + z) (7)
qed RM
Initially, of course, a « l. But during collapse both Rand M would decrease, and
even otherwise, Eq.(7) shows that, as z ---t 00, a would increase dramatically and
at a sufficiently high z, one must have a = qtrap / qed ---t l. At this stage the collapse
would degenerate into a secular quasistatic contraction by the very definition of
L e d. 1- 4 In a very strict sense, however, the body would still be contracting on
extremely long time-scales! Since L oo = -c2dM / du, the time scale associated with
this phase is 3 ,4
Mc 2 Mc 2 ""c(l + z)
u = -c2dM/du = -L-~- = -4'-7r-C-::---'- (8)

Obviously, u ---t 00 irrespective of the value of "" as the BH stage (z = 00) would
be arrived. Thus the Eddington-limited contracting phase actually becomes eternal
and the object in this phase may be called an Eternally Collapsing Object (ECO).1-4
Therefore the observed black hole candidates (in the present epoch) should be in
this ECO phase (z = finite» 1) rather than in the exact BH phase (z = 00).5 In
particular, with """( ~ 0.4 cm 2 /g, one has u"( = 4.lO s (1 + z)yr» Hubble Time.

3. Discussions
Unlike a Schwarzschild BH, an ECO will have a physical surface and strong intrinsic
magnetic field (B). In fact Robertson & Leiter 6- s have shown that the stellar mass
black-hole candidates (BHCs) do not have any EH (z = 00) though they have an
extremely large but finite z ~ 107-s. They have also inferred that for the stellar
mass ECOs B ~ 10 16 - 17 G. Note, the observed field would be reduced by a factor
1528

of (1 + z). Several previously unexplained features of Radio and X-ray jets of the so-
called BH binaries have already been explained by realizing that the BH Candidates
are actually magnetized ECOs. 6- 8 Moreover, direct observations have inferred that
the central compact object of one of the best studied quasar Q0957+561 is an
ultramagnetized ECO rather than a BH.9
Cuesta, Salim and Santos lO have found that collapse of (Newtonian) Supermas-
sive Stars first produces an ECO rather than a static BH. Several other authors
have found that continued GR collapse indeed results in a radiation supported hot
quasistatic phase as described here ll - 13 if the effect of heat flow would be appropri-
ately taken into consideration. There is an exact GR collapse solution which shows
that effect of heat flow prevents BH formation. 14
Mathematically, however, the ECO is asymptotically approaching the static con-
figuration of a z = 00 BH to honour the exact vacuum Schwarzschild solution.

References
1. A. Mitra, Phys. Rev. D 74,024010 (2006) (gr-qc/0605066)
2. A. Mitra, MNRAS Lett. 367, 367 (2006), gr-qc/0601025
3. A. Mitra, MNRAS 369,492 (2006), (gr-qc/0603055)
4. A. Mitra, New Astronomy 12, 146 (2006)
5. A. Mitra, Adv. Sp. Sc. 38(12), 2917 (2006)
6. S.L. Robertson and D. Leiter, Astrophys. J., 565, 447, (2002)
7. S.L. Robertson and D. Leiter, strophys. J.L., 569, L203 (2004)
8. S.L. Robertson and D. Leiter, Mon. Not. Roy. Astr. Soc., 350, 1391 (2004)
9. R.E. Schild, D.J. Leiter and S.L. Robertson, Astron. J., 132, 420 (2006)
10. H.J.M. Cuesta, J.M. Salim and N.O. Santos, in 100 Years of Relativity, Sao Paulo,
Brazil, 2005, see http://www.biblioteca.cbpf.br/apub/nf/NF-2005.html
11. L. Herrera and N.O. Santos, Phys. Rev. D70, 084004 (2004)
12. L. Herrera, A. Di Prisco, and W. Barreto, Phys. Rev. D73, 024008 (2006)
13. L. Herrerea, A. Di Prisco and J. Ospino, Phys. Rev. D74, 044001 (2006)
14. A. Banerjee, S. Chatterjee, & N. Dadhich, Mod. Phys. Lett. A 17, 2335 (2002)
 THE PROPOSED BLACK HOLES AROUND US

WOLFGANG KUNDT
Argelander Institute of Astronomy,
Auf dem Hugel 71,
D-53121 Bonn, Germany
[email protected]

This chapter combines two of my contributions to MG 11: an analysis of the proposed

supermassive black holes at the centers of galaxies which are thought to give rise to the
quasar phenomenon, and an analysis of the stellar-mass black-hole candidates (BHCs) -
typically a million times smaller in mass - which behave similarly to neutron stars, and
are often called 'microquasars'. For both source types, I shall argue that the black-hole
interpretation meets with serious inconsistencies, and that instead, we deal with non-
collapsed, rotating, strongly magnetized concentrations of matter, similar to ordinary
stars and neutron stars.

Keywords: black holes, supermassive black holes, black-hole candidates

1. Introduction

In the abstract of perhaps the leading present-day review on galactic n\lclei, by

Laura Ferrarese and Rolland Ford [2005], you can read: "Since the early '90s, rapid
technological advances ... have not only given us incontrovertible proof of the exis-
tence of supermassive black holes, but have unveiled fundamental connections be-
tween the mass of the central singularity and the global properties of the host galaxy.
It is thanks to these observations that we are now, for the first time, in a position to
understand the origin, evolution and cosmic relevance of these fascinating objects."
This review is quite carefully written; but all you can find in it is methods, and their
results, to determine the unresolved core masses of dozens of well-sampled galaxies.
No 'singularity' is offered, nor any black-hole (BR) property indicated. The review
holds its promises if you replace the two words 'BR' and 'singularity' by the word
'object'.
In the following two sections (on rotation curves and central disks), I will argue
from first principles that the AGN phenomenon can be understood as an expected
behaviour of massive gaseous disks governed by Newtonian dynamics. The mostly
unresolved central engines are the central parts of the (gaseous) galactic disks,
of solar-system size, whose mass densities take stellar values, and which therefore
behave like fiat stars with frequent nuclear detonations involving gigantic mass ejec-
tions, the latter seen successively as the BLR, NLR, and ESR (broad-line, narrow-
line, and extended-shell region, [Kundt 1987, 1996b]). Quantitative estimates sug-
gest that the central disk is considerably screened by its escaping wind matter.
The final section (on stellar-mass BRCs) will remind the reader of the virtual
indistinguishability of the class of BRCs from the class of neutron-star binaries, in
all observed properties other than the mass of the compact component. This higher
mass of the BRCs may well be present in the form of a massive accr·etion disk around

1529
1530

the neutron star, acquired from its companion, of degenerate mass density, whose
outer parts rotate rigidly. The mass of the disk effects the slight difference in its
feeding behaviour ~ with dim epochs of steadiness interspaced by epochs of flaring
outbursts ~ but does not strongly modify the radiative behaviour of its enclosed
neutron star.
At this point, it comforts me to quote Mitch Begelman's [2003] review 'Evi-
dence for Black Holes', in which he writes on page 1901: "Despite the overwhelming
circumstantial evidence for black holes, the measurements discussed so far do not
establish that the dark masses and compact objects we detect are the black holes
whose properties are predicted so precisely by general relativity." And he does not
mention the various shortcomings of the (axially symmetric!) BHs that one meets
when one tries to model all the observations; they will be discussed below.
The present understanding of the proposed BH family can also be found in Kim-
berly Weaver [2003], or Jeffrey McClintock [1988], with fresh claims in McClintock
et al [2006]. A new approach to its uniformity has been made by a study of vari-
abilities, whose (almost white) power spectra of fluctuations have broken-power-law
shape, with high-frequency breakpoints which are apparently reliable mass indica-
tors [McHardy et al 2006]. The break-frequencies also correlate with the widths of
certain emission lines. These new mass indicators suggest a uniform activity mech-
anism of all BH sources, spanning a range of more than 8 orders of magnitude
in mass. A similar uniformity is signalled by all the (high- and low-mass) jet en-
gines [Kundt 1996b, 2001, 2005, Kundt & Krishna 2004]: The sources of the jets
involve new-born stars, (our Sun not excepted), binary neutron stars or forming
white dwarfs, and the central engines of active galactic nuclei. Consequently, if the
jet engine is of uniform type, a black hole does not qualify. A magnetized rotator
will turn out to be a convincing choice.

2. Galactic Rotation Curves

Galactic rotation curves, v(r) versus r, are approximately flat: They are usually
plotted (known) for r between 1 Kpc and 20 Kpc, and show the rotational velocity
v(r) steeply rising at small r, from values of order 10 2 Km/s to asymptotic values
which are 2 to 3 times larger. According to Kepler's law ~ generalized from a central
point mass to the flattish mass distribution of a disk ~ a constant v(r) signals a
mass distribution which grows in proportion to r: M(r) "-' r. Standard disk theory
then predicts a constant mass spiral-in rate towards the center for the gaseous
component, because of (viscous, or rather magnetic [Balbus 2006]) shear forces, for
an angular velocity w(r) which decreases with r. For a massive galaxy, this rate is
of the order of one solar mass per year, i.e. leads to a fast accumulation of matter
at its center, some 10 6 M8 per Myr. How does such a mass accumulation behave?
Will it turn into a nuclear reactor, like every star? Will it even show non-thermal
behaviour, like every new-born star?
Present-day observations cannot resolve the deep interiors of galaxies; yet they
 1531

have advanced a lot farther towards the center than so far mentioned, from 1 Kpc
= 1021.5 cm down to some 10 15 cm in the case of the Milky Way, i.e. through more
than 6 orders of magnitude in r. This vastly extended (radial) range asks for a
double-logarithmic presentation, as is used in fig.1 below. In it, the average surface-
mass density u := M(r)/7fr 2 has been plotted instead of vCr), because it allows
us to coarsely predict the whole (extended) rotation curve: At large radii, gaseous
disks are marginally unstable to star formation (inside of condensing clouds), i.e.
have surface mass densities near that of the onset of Jeans instability, u ,<, U Jeans =
2
v'2WC s /7fG ::::0 1O-1.3 g cm- /r22.4 [Tassoul, 1978], where Cs := sound speed, G :=
Newton's constant, and where rx stands short for r/10 xdim(r).
For the rotation law u = U Jeans, gas in centrifugal balance with gravity is known
to spiral inward because of Kepler's law

(1)
in which the factor 7] E (0.3,1) takes care of a non-spherical mass distribution in
Newton's force law: GM(r)/r = 7]V 2 ; clearly, W := vir decreases outward as long as
u/r does, and matter spirals inward. Such inward motion comes to a halt, however,
when matter piles up at the center, and reaches typical observed values on the
timescale of several Myrs. Such a pileup gives rise to a quasi plane-parallel layering
- many stars squeezed into a thin sheet, a burning pancake, or 'burning disk' (BD)
- symmetrical to its midplane, of stellar thickness, and radial extent of order 10
a.u., whose local state has been analytically approximated, e.g., in [Kundt 1996,
2002].
Once matter has piled up at the center of a galactic disk, in the form of a solar-
system sized BD, it represents a density spike such that the surrounding (feeding)
gas has a negligible mass in comparison, hence approximates the law M(r) = const,
i.e. u rv r- 2 , the steepest possible radial drop of the average surface mass density
(because mass is positive), until it joins its asymptotic branch u = u Jeans, on radial
scales between a pc and a Kpc. According to these considerations, the function u(r)
has a broken-power-Iaw shape throughout, starting with exponent 0 at the center,
dropping with exponent -2 beyond, and reaching exponent -1+2E asymptotically in
the outer parts for the rotation law vCr) rv rE with E small, ::::0 ± 0.1. Figure 1 shows
that these expectations are consistent with the known data for all sampled galaxies.
The deduced broken-power-Iaw shape of u(r) in fig.1 implies a corresponding
broken-power-Iaw shape of vCr), with successive exponents {1/2, -1/2, E}, so that
vCr) assumes its maximum value at the outer edge of the nuclear-burning core, be-
tween 10 13 cm and 10 16 cm. This maximal rotation speed must not exceed the speed
of light, in conformity with the data: Fig.1 shows that a violation would happen
for nuclear masses in excess of a few 10 10 Mo, whilst the present-day record (of
Cyg A) reads 109.4Mo [Ferrarese & Ford 2005]. At the radial distance of V max ,
the shear of galactic matter is extreme, implying extreme toroidal magnetic field-
strengths, reconnect ions, pair formations, and spiral-in thrusting (into the BD).
Clearly, whenever this small radius can be resolved, it should stand out of the map;
1532

....... 8
<;'
E
u
C'I

1"2::1
Or-------------------------------~~~~------~

-4
12 14 log (r I em) 20 22

Fig. 1. Average surface-mass density a(r) M(r)j1rr2 versus radial distanee r for gaseous
galaetic disks, updated from [Kundt 1996, 2002]' with new data from [Ott et al 2003], [Ferrarese
& Ford 2005], [Marconi & Hunt 2003]; see also [Davies et al 2004, 20061. An upper limit to 0" is
set by stellar column densities 0". and by the relativistic barrier VK 2: c. The horizontal line T 2: 1
marks Thomson opaqueness, and 0" 2: 0" Jeans denotes onsetting Jeans instability. All well-sampled
galaxies have core masses between some 10 5 MG and 109 . 4 MG, typically above 106 MG if active.
Note that the eurves of eOllstant M(r) have slope -2.

has it been detected by Schild et al [2006], for Q0957 +561 ? Inside of this critical
radius or rather already (slightly) outside of it the BD is expected to switch on
main-sequence nuclear burniug, followed by explosive nuclear burning with increas-
ing temperature and metal enhancement further in: an AGN starts its activity.
As has been mentioned above, the maximal rotation speed '0(7') of a galactic
disk can get close to the speed of light c, but has so far uowhere ever reached it.
This insight implies that galactic nuclei have never reached mass densities sufficient
for gravitational collapse, not even at their centers: the supermassive BH has been
a premature speculation [K undt 2002]. The next section will collect independent
evidence for this claim. Right now, a fourth expected power-law exponent '1' of
'0(7') is worth mentioning, near the very center of a galactic disk, of stellar extent:
here the gas pressure of a BD is expected to exceed its centrifugal pull, and the
Keplerian law of equ.1 is replaced by rigid rotation, v(r') rv 7'. This rigidly rotating
innermost part of a BD should be its highest-pressure, hottest, most explosive core,
ready to be (piecewise) ejected during subsequent gigantic mass ejections. If it was
 1533

not, large present-day galaxies would all harbour SBHs above 10 10 M8 in their
nuclei, composed of the accumulated ashes of all their past quasar fires.

3. Properties of a Central Disk

The galactic rotation structure highlighted in the past section has a number of im-
portant consequences. It predicts a faster than linear increase of its average surface-
mass density a(r) with r- 1 , on approach of the center, and a consequent increase
of its average volume-mass density p( r) faster than r- 2 , for a constant opening
angle e of the disk. At solar-system distances from the center, therefore, we should
not be surprised to encounter mass densities p in excess of 1O- 18 g/cm 3 , (instead of
the local Galactic 1O- 2S g/cm 3 , for similar temperatures). expected (full) height h
of the inner disk, we infer from Newtonian gravity for a 2-d gas layer the I-sided
acceleration 9 = 27rGa, a = ph/2, and a midplane pressure gph/2 = p = nkT with
p =: nm, m := mass of the hydrogen atom, whence

(2)
and its surface-mass density

a* = 2ph = VSnkT/7rG = 1011.Sgcm-2VT7.Sn24' (3)

For an average mass density p near that in the Sun, n = 10 24 cm- 3 , the mass M(r)
of the central rotator is found to be
6 7
M(r) = 7rr 2 a = 10 . M8rr4all.5, (4)
with a peripheral rotation speed v of
v ~ v10Gm = I0 9S cms- 1 fti4 (5)
which would reach the speed of light for an (excessive!) radius of I0 16 cm, corre-
sponding to an enclosed mass of M = I0 10 .7M8 .
Such a central burning disk, of hypothetical surface temperature T ;S I0 6 K, is
expected to radiate a largely super-Eddington power L given by

(6)
where aSB:= Stefan-Boltzmann constant, aT:= Thomson cross section, and to blow
off a hot wind of mass rate Mout which on average equals its accreted mass rate
Min for typical parameters:
(7)
where N ~ I0 24 ,s cm- 2 denotes the wind's absorption column density, and E ~
10- 2 . 3 denotes the conversion efficiency of accreted mass into radiated power L
for incomplete nuclear hydrogen burning. In other words: unlike a BH, a BD can
re-eject all its accreted mass to infinity via nuclear power.
Can our telescopes hope to disclose (several of) the innermost galactic disks?
Or are the 30 magnitudes of visual absorption towards Sgr A * typical of central
1534

opaqueness? Does the broad iron line at the center of NGC 4258 discussed by
Bromley et al [1998] map Keplerian rotation? I do not think so: When focussing
at a galaxy's rotation center, our sight line must cross at least (i) the confining
outer wall material of the BLR, (ii) the radially coasting cloudlets of the BLR,
(iii) the volume-filling, escaping relativistic pair plasma (created in magnetospheric
reconnection events), and (iv) the disk's fluctuating corona.
The first three of these listed screens have (estimated) respective opacities T
of (i) CJTNwali ;::::; 10- 2 , (ii) CJTNclouds ;::::; 10°·8 M(o)/ ,13-1.51"16, and (iii) CJTNe ;::::;
CJTLjet/rm e c3,B, ;::::; L 44 /r16(,B')0.5, with M(o) := M/100M8 yc 1, ,13 := vic, Ne:=
column density of outgoing relativistic e±-plasma, of Lorentz factor " and bulk
velocity c,B, and where a typical BLR radius 1" of 10 16 cm has been assumed through-
out. This latter radius follows from: (j) 'reverberation', as c!1t/2 for !1t = 105 .8s,
(jj) emission-line strength, as VLline/Sline for VL44/S12 = 1, (jjj) BLR confine-
ment, for M(8)P-18/PO = 1, (jv) luminous blackbody area, as vL bb /4nCJsBTb for t
VL46/T1.3 = 1, (v) being distinctly larger than the expected speed-of-light radius
of the innermost disk, c/w = VTJR/nGCJ = 10 15 crn VR14/CJl1.5, and (vi) it agrees
with the observed nozzle size of the M 87 jet [Junor et al 2000].
In the last paragraph, I have shown that the innermost disk of an AGN lies
behind a thermal absorption screen of optical depth at least .<; unity, and in addi-
tion behind a relativistic pairplasma screen whose inverse-Compton boosting should
broaden all X-ray lines. Additional X-ray line intensity is expected to be superim-
posed further out - from the heated channel-wall material of the forming jet - so
that in particular the iron Ka line cannot be used to obtain reliable information
about the kinematics of an inner galactic disk.
From what has been shown, it should have become clear that I face the following
nine difficulties with the SBH interpretation of (active) galactic nuclei, cf. Kundt
[1996a, 2002, 2005]: • the missing mass at the centers of present-day galaxies, where
e.g. the Milky Way should now harbour some 10 10 . 5 M 8 , by having duely partici-
pated in the AGN phenomenon; • the powerful AGN winds, which eject as much
matter during active epochs as is on average supplied by the disk's feeding; • the
high ,-my compactness, which would cause the in-situ destruction of the pairplasma
required to blow the jets of (even) the (10%) radio-loud subpopulation; • the high
metal abundance of their ejecta, .<; 10 2 times solar, like the ashes of exhaustive nu-
clear burning; • their hard spectm, often peaking above Te V photon energy, which
clash with predicted BH temperatures of KeV(M8/M)1/4; • the nuclear-burning
hurdle, or missing birth events: their best birth sites, the centers of galaxies, are
distinctly underdense for BH formation; • the inverted evolution of the AGN phe-
nomenon, whose Eddington power would have to grow with cosmic time; • the
scaling in mass of the central engine with that of its host galaxy, [Marconi & Hunt
2003]; • the universality of the jet phenomenon, which is common to all fast magne-
tized rotators: young stars, binary neutron stars, and binary forming white dwarfs,
(whose QPO can be understood via magnetospheric oscillations).
In his Bonn colloquium of 16 November 2007, Frank Eisenhauer reported on new
 1535

results by the Garching team on the central engine of our Galaxy. Alarming was
a non-closing (by 30) Kepler ellipse of the nearest monitored star around Sgr A *,
S2, in conflict with a spherical gravitational potential. He also reported an effective
mass of 3.8 x 106 M(,), 1.3 times larger than an earlier determination (of 2.9 x 10 6 M('))
in the Messenger 111 of March 2003. Both results are expected for a massive central
disk, instead of a BH.

4. Neutron Stars versus Black-Hole Candidates

There are not only serious problems with the proposed supermassive BH engines in
the centers of galaxies, but equally with their (smaller) stellar-mass cousins fed by a
surrounding accretion disk. Such problems have been independently highlighted by
Kundt [2001, 2005], Robertson & Leiter [2006], and Mazur & Mottola [2004], and
confronted with alternative models: Newtonian ones, MECOs (:= Magnetospheric
Eternally Collapsing Objects), and gravastars, respectively.
When compared with neutron-star sources, binary BHs lack (i) a solid surface
(for thermalization, and for anchoring a magnetic flux), Oi) a (strong) inclined,
corotating magnetosphere (for torquing, and pair production), (iii) high transient
surface temperatures (during clumpy accretion), (iv) two extended windzones (shap-
ing their ~ non-periodic ~ lightcurves), and (v) non-swallowed accretion flows, with
Min;:::; M out [Miller et aI, 2006], guaranteeing strong (and variable) winds.
These five properties, in which neutron stars are superior (as active sources) to
BHs, cannot be dispensed with, so I think, when one wants to explain the various
peculiarities of the BHCs: (j) their broad, line-rich, and often hard spectra, (from
radio frequencies to TeV energies), (jj) their multiply structured lightcurves, (often
with noisy periodicities, yet showing precession, bimodality, flaring, bursting, dip-
ping, flickering, QPOs, polarizations), (jjj) occasional super-Eddington intensities
(by ;S 103-fold), (jv) twin-jet formation (indistinguishable, in detail, from all the
other jet sources), and (v) with masses and ages that allow an interpretation as the
(abundant) stellar class of binaries which has evolved from not-too-wide, high-mass
progenitor systems, different only by the super-solar mass in their accretion disk,
and which fills presently the mass gap between the high-mass and the low-mass
systems.

Acknowledgments
It is a great pleasure to thank Ole Marggraf for fig. 1, Gunter Lay for electronic
help, and Gernot Thuma and Hans Baumann for the manuscript.

References
1. S. A. Balbus, Nature 444, 281-283 (2006)
2. M. C. Begelman, Science 300, 1898-1903 (2003)
3. B. C. Bromley, W. A. Miller, V. 1. Pariev, Nature 391, 54 (1998)
4. R. 1. Davies, L. J. Tacconi and R. Genzel, Astrophys. J. 602, 148-161 (2004)
1536

5. R. 1. Davies, J. Thomas, R. Genzel, F. Mueller Sanchez, L. J. Tacconi, A. Sternberg,

F. Eisenhauer, R. Abuter, R. Saglia, and R. Bender, Astrophys. J. 646, 754-773 (2006)
6. L. Ferrarese, and H. Ford, Space Science Reviews 116, 523-624 (2005)
7. W. Junor, J.A. Biretta, and M. Livio, Nature 401, 891-892 (2000)
8. W. Kundt, in: Thirteenth Texas Symposium on Relativistic Astrophysics, ed. M.P.
Ulmer (World Scientific, Singapore), 314-315 (1987)
9. W. Kundt, Astrophys. and Space Science 235, 319-327 (1996a)
10. W. Kundt, in: Jets from Stars and Galactic Nuclei, Lecture Notes in Physics 471,
1-18 (1996b)
11. W. Kundt, in: MICROQUASARS, Granada Workshop on Galactic Relativistic Jet
Sources, eds. A. J. Castro-Tirado, J. Greiner, and J. M. Paredes, 273-277 (Kluwer,
2001)
12. W. Kundt, New Astronomy Reviews 46, 257-261 (2002)
13. W. Kundt, Astrophysics, A New Approach (Springer, 2005)
14. W. Kundt, G. Krishna, J. Astrophys. Astr. 25, 115-127 (2004); also: astro-ph/0406318
v2
15. A. Marconi, and L. K. Hunt, Astrophys. J. 589, L21-L24 (2003)
16. P. O. Mazur, and E. Mottola, PNAS 101, 9545-9550 (2004)
17. J. E. McClintock, Sky 8 Telescope January, 28-33 (1988)
18. J. E. McClintock, R. Shafee, R. Narayan, R. A. Remillard, S. W. Davis, and L.-X. Li,
Astrophys. J. 652, 518-539 (2006)
19. I. M. McHardy, E. Koerding, C. Knigge, P. Uttley, and R.P. Fender, Nature 444,
730-732 (2006)
20. J. M. Miller, J. Raymond, A. Fabian, D. Steeghs, J. Homan, C. Reynolds, M. van der
Klis, and R. Wijnands, Nature 441, 953-955 (2006)
21. T. Ott, et al (28 authors), The Messenger 111, 1-8 (2003)
22. S. L. Robertson, and D. J. Leiter, astro-ph/0602453 vI
23. R. E. Schild, D.J. Leiter, and S. L. Robertson, Astronom. J. 132, 420-432 (2006)
24. J.- L. Tassoul, Theory of Rotating Stars, (Princeton University Press, 1978)
25. K. Weaver, Scientific American, July, 26-33 (2003)
 GRAVASTARS AND BIFURCATION
IN QUASISTATIONARY ACCRETION

EDWARD MALEC and KRZYSZTOF ROSZKOWSKI

M. Smoluchowski Institute of Physics, Jagiellonian University,
Reymonta 4, 30-059 Krak6w, Poland

We investigate the newtonian stationary accretion of a polytropic perfect fluid onto a

central body with a hard surface. The selfgravitation of the fluid and its interaction with
luminosity is included in the model. We find that for a given luminosity, asymptotic mass
and temperature of the fluid there exist two solutions with different cores.

Keywords: gravastars, nonuniqueness of solutions, stationary accretion

1. Introduction

The question we want to address in this paper is the following inverse problem:
having a complete set of data describing a compact body immersed in a spherically
symmetric accreting fluid, find the mass of the central body. We assume that we
know the total mass, luminosity, asymptot.ic temperat.ure, the equation of st.at.e of
the accreting gas and the gravitational potential at the surface of the core.
The fundamental question is whether observers can distinguish between gravas-
tars [1] versus black holes as engines of luminous accreting systems (see a controversy
in [2, 3]). While we do not address this problem here, we show a related ambiguity
in a simple newtonian model.

2. The Shakura model

The first investigation of stationary accretion of spherically symmet.ric fluids, includ-
ing luminosity close to the Eddington limit, was provided by Shakura [4]. It was
lat.er extended to models including the gas pressure, its selfgravity and relativistic
effects [5-8].
In the following we will denote the areal velocity by U(r, t) = atR (where t is
co moving time and R the areal radius), the local, Eddington and tot.alluminosities
by L(R), LE and L o, quasilocal mass by m(R) and total by M, pressure by p,
the baryonic mass density by (J (the polytropic equation of state will be p = K (Jr,
1 < r ::; 5/3) and the gravitat.ional potential by ¢(R). The radius of the central
body is Ro while its "modified radius" is defined by Ro = GM/I¢(Ro)l. Under the
assumption that at the outer boundary of the fluid the following holds true:

U~ « Gm~~oo) «a~, (1)

we have t.he following set of equations:

1M = -47rR2(JU, (2)

UaRU = - Gm(R) _ ~aRP + a L(R) (3)

R2 (J R2 '

1537
1538

(4)

2 2 2
Lo - L(R) = M. ( ~
a - -a - - -U - ¢(R) ) . (5)
r-1 r-1 2
a is a dimensional constant a = aT /47fm p c. The details of solving the system are
provided elsewhere [9, 10] and we will present only the main results here.
We assume that the accretion is critical, i. e., there exists a sonic point, where
the speed of accreting gas U is equal to the speed of sound a. All values measured
at that point will be denoted with an asterisk. We define:

Lo
X= LE'

and obtain the total luminosity:

M3 ( 2 ) (5-3r)/2(r-1)
Lo = ¢oXooG 7f2 a~ (1 - y)[y - xexp( -x/,W
2
5 _ 3r (6)

Xoo is approximately the inverse of the volume of the gas located outside the sonic
point. In sake of brevity we will use
_ 2 M2 ( 2 ) (5-3r)/2(r-1)
(3 - a¢oXooG7f ~ 5 - 3r
00

to obtain Eq. (6) in a form using the relative luminosity:

x = (3(1 - y)[y - x exp( -x/,)f (7)

3. Bifurcation
For the relative luminosity, fulfilling Eq. (7) we proved the following theorem:

(i) For the functional F(x, y) = x - (3(1- y)[y -x exp( -x/,)]2 there exists a critical
point x = a, y = b such that F(a, b) = 0 and oyF(a,b) = 0 with 0 < a < b < 1
and a = 4(3(1 - b)3, b = [2 + aexp( -a/,)l!3.
(ii) For any 0 < x < a there exist two solutions y(x)! bifurcating from (a, b). They
are locally approximated by:
1/2
y+ = b ± (a - x)[b + aexp( -a/,)(l - 2a/,)]
( (8)
- (3[b - aexp(-a/,)][l- aexp(-a/,)] )
(iii) The relative luminosity x is extremized at the critical point (a, b).

4. Discussion
In the paper we have assumed the existence of an accreting system which satis-
fies certain conditions. Under those assumptions the complicated set of integro-
differential nonlinear equations (2-5) can be simplified to an algebraic one (7). We
 1539

checked numerically that the performed simplification causes errors of the order of
10- 3 (see [9J for details).
The analysis of Eq. (7) shows that there exist two different solutions, having the
same total luminosity and total mass, but different masses of the core objects. One
can also conclude that for sufficiently large f3 the maximal relative luminosity a can
get close to 1, i.e., the total luminosity approaches the Eddington limit.
As the two solution branches bifurcate from the point (a, b), there is no much
difference between the central masses of bright objects (see [9, 10] for plots). How-
ever, when luminosity is small (Lo « L E ), this difference can become arbitrarily
large. This can be understood intuitively, because the radiation is small for test
fluids (since the layer of gas is thin), or when the central object is light (therefore
weakly attracting surrounding gas).
The results obtained here are consistent with relativistic analysis neglecting
interaction between the gas and the radiation [11, 12J.

Acknowledgments
This paper has been partially supported by the MNII grant IP03B 01229.

References
[1] P. O. Mazur and E. Mottola, Proc. Nat. Acad. Sci. 101, p. 9545 (2004).
[2] M. Abramowicz, W. Kluiniak and J.-P. Lasota, Astron. Astrophys. 396, p. L31
(2002).
[3] A. E. Broderick and R. Narayan, Astrophys. J. 638, p. L21 (2006).
[4] N. 1. Shakura, Astr. Zh. 18, p. 441 (1974).
[5] T. Okuda and S. Sakashita, Astroph. Space Sc. 52, p. 350 (1977).
[6] K. S. Thorne, R. A. Flammang and A. N. Zytkow, MNRAS 194, p. 475 (1981).
[7] L. Rezzolla and J. C. Miller, Class. Quant. Grav. 11, p. 1815 (1994).
[8] M. G. Park and G. S. Miller, Astrophys. J. 371, p. 708 (1991).
[9] J. Karkowski, E. Malec and K. Roszkowski, Luminosity, selfgravitation and
nonuniqueness of stationary accretion, Submitted to Astrophys. J., http://arxiv .
org/ps/astro-ph/0611393, (2006).
[10] E. Malec and K. Roszkowski, Nonuniqueness of stationary accretion in the Shakura
model, in Proceedings of the XXIX Spanish Relativity Meeting (ERE 2006): Ein-
stein's Legacy: From the Theoretical Paradise to Astrophysical Observations, (Palma
de Mallorca, Spain, 2006).
[11] J. Karkowski, B. Kinasiewicz, P. Mach, E. Malec and Z. Swierczynski, Phys. Rev. D
73, p. 021503(R) (2006).
[12] B. Kinasiewicz, P. Mach and E. Malec, Selfgravitation in a general-relativistic ac-
cretion of steady fluids, in Proceedings the 42 Karpacz Winter School of Theoretical
Physics, (Ladek Zdroj, Poland, 2006).
This page intentionally left blank
Numerical Relativity, Black
Hole Collisions, and
Algebraic Computation
This page intentionally left blank
 LIFETIME OF OSCILLONS
GYULA FODORl, PETER FORGACS 1 ,2, PHILIPPE GRANDCLEMENT2,3
and ISTVAN RAcz 1
1 MTA RMKI, H-1525 Budapest 114, P. O.Box 49, Hungary,
2 LMPT,CNRS- UMR 6083, Universite de Tours, Parc de Grandmont, 37200 Tours, France,
3LUTH, CNRS-UMR 8102, Observatoire de Paris-Meudon, place Jules Janssen, 92195 Meudon
Cedex, France

Oscillons are remarkably long living, localized and nearly periodic configurations that
form in a variety of nonlinear theories. In this short note we review oscillons in the spher-
ically symmetric sector of ¢4 theory in 3 + 1 dimensions. Oscillons can be characterized
as the core part of strictly periodic weakly localized solutions, called quasi-breathers.

The aim of this short report is to present some of the main properties of oscillons
in ¢4 theory and to give their interpretation as the localized part of the exactly
periodic quasi-breather solutions. A more detailed discussion and fuller references
are presented in Ref. l. For a related study in D + 1 dimension see Ref. 2.
The field equation of a spherically symmetric nonlinear scalar field in ¢4 theory
on a 3 + 1 dimensional fiat background is given by
2
-¢,tt + 6.¢ = ¢ (¢2 - 1) , where 6. = 8; + -8r (1)
T"

In order to have finite energy the field must to tend to its vacuum value at infinity,
which is chosen to be ¢(XJ = -l. Although oscillons evolve form a variety of localized
initial field configurations, because of their invariance on the type of initial data, it
is customary to specify Gaussian initial profile

(2)

with ¢c giving the central amplitude of the field and T"o determining the width of the
pulse. For a large subset of the possible initial values ¢c and T"o, after the emission
of some surplus energy, an oscillon state forms in the central region. 3 - 5 In general,
although these states are long living (up to 7000 in natural units when ¢c = 1),
they clearly emit energy slowly, and after a certain time they suddenly disintegrate.
On the upper plot of Fig. 1 the upper envelope of ¢ at the center T" = 0 for two
typical oscillon states is shown. The frequency of the underlying oscillations follows
a slowly increasing tendency in the range l.2 < w < l.365.
Approaching closely a finite number of special initial parameters, instead of
the quick final disintegration a near-periodic state forms at the late stage of the
evolution. 6 The lifetime of these near-periodic states increases according to a scaling
law as one approaches the critical value with the initial parameter. For the examples
given on the lower plot of Fig. 1 the fine tuning is performed to 15 digits, and the
obtained lifetime of the near-periodic state is close to 1000. One of the results
presented in Ref. 1 is that these apparently periodic states are very slowly losing
energy, and consequently cannot have infinite lifetime even if the fine tuning is

1543
1544

0.8
0.6
0.4
0.2
x
0
",0
E!!. -0.2
-e-
-0.4 ro = 2.39375 ~-

-0.6 ro = 2.3925 ----

-0.8
-1
0 500 1000 1500 2000 2500 3000

0.8
0.6
0.4
0.2
0
ttio
E.!! -0.2
-e-
• -15
-0.4 rO = r27 + 3·10
-0.6 • -15
rO = r27 - 3·10
-0.8
-1
0 500 1000 1500 2000 2500 3000

Fig. l. On the upper plot the time dependence of the upper envelope of the central value of the
field for two typical oscillon states with CPc = 1 is presented. On the lower plot the parameter ro
is fine tuned close to a critical value denoted by r 27 , and a near-periodic state forms.

performed to exceedingly high precision. From the energy loss the estimated lifetime
of typical near periodic states are of the order 10 6 . Other important difference
between near-periodic states and the general non fine-tuned oscillons is in their
higher frequency range, all near-periodic states being in the interval 1.365 < w <
V2.
The time dependence of the field was calculated using a fourth-order "method
of lines" numerical evolution code employing spatial compactification. In addition,
a direct search was carried out for exactly periodic oscillon solutions using a Fourier
decomposition method. expanding the field in the form
00

¢(t, r) = -1 + ¢o(r) + L ¢n(r) cos(nwt) (3)

n=l

and solving the resulting system of ordinary differential equations for ¢n (r) by a
multidomain spectral method. Although the n = 0 and n = 1 modes are decay-
ing exponentially for large radii, the higher modes have the asymptotic structure
sin(AnT + 1n)/r, which is a too slow falloff to have finite energy. The numerical
analysis shows that it is not possible to suppress these modes completely, and con-
sequently there is no localized finite energy periodic solution. The most one can
 1545

Mode decomposition - -
Real part of Fourier transform -

0.01

n=O
1e-04

c
-e- 1e-06
'"
~

1e-08

n=5

1e-10

1e-12
0 5 10 15 20 25 30 35 40

Fig. 2. Comparison of the modes of the quasi-breather solution with frequency w = 1.398665 to
the Fourier decomposition of the corresponding near-periodic state.

do is to minimize the amplitudes, and for each frequency obtain a configuration

with a minimal energy density tail, called quasi-breather. The comparison of the
obtained modes to that of the Fourier decomposition of a near-periodic state calcu-
lated by the evolution code is shown on Fig. 2. Good agreement also holds for the
non fine-tuned lower frequency states. Consequently, the localized part of an oscil-
Ion can be interpreted as the core of the quasi-breather with the same frequency.
The evolution of an oscillon corresponds to an adiabatic evolution through a family
of quasi-breathers with slowly changing frequency w(t).

Acknowledgements
This research has been supported by OTKA Grants No. T034337, TS044665,
K61636 and by the Pole Numerique Meudon-Tours_ G_ F. and L R would like
to thank the Bolyai Foundation for financial support.

References
1. G_ Fodor, P_ Forgacs, P. GrandcJement, 1. Racz, Phys. Rev. D74, 124003 (2006).
2. P. M. Saffin, A. Tranberg, hep-th/0610191.
3. I. L. Bogoluvskii, and V. G. Makhankov, JETP Letters 24, 12 (1976), ibid. 25, 107
(1977)
4. M. Gleiser, Phys. Rev. D49, 2978-2981 (1994),
5. E. J. Copeland, M. Gleiser and H.-R. Muller, Phys. Rev. D52, 1920-1933 (1995).
6. E. P. Honda and M. W. Choptuik, Phys. Rev. D65, 084037 (2002).
 A VIRTUAL TRIP TO THE SCHWARZSCHILD-DE SITTER
BLACK HOLE*
PAVEL BAKALA"", STANISLAV HLEDIK"
ZDENEK STUCHLIK*, KAMILA TRUPARovAt
Institute of Physics, Faculty of Philosophy and Science,Silesian University in Opava
Bezrucovo nam. 13, CZ-74601 Opava, Czech Republic
"" pavel. [email protected], "stanislav. [email protected]
* [email protected], t kamila. [email protected]

PETRCERMAK
Institute of Computer Science, Faculty of Philosophy and Science, Silesian University in Opava
Bezrucovo nam. 13, CZ-74601 Opava, Czech Republic
petro [email protected]

We developed realistic fully general relativistic computer code for simulation of optical
projection in a strong, spherically symmetric gravitational field. Standard theoretical
analysis of optical projection for an observer in the vicinity of a Schwarzschild black
hole is extended to black hole spacetimes with a repulsive cosmological constant, i.e,
Schwarzschild-de Sitter (SdS) spacetimes. Influence of the cosmological constant is in-
vestigated for static observers and observers radially free-falling from static radius. Simu-
lation includes effects of gravitational lensing, multiple images, Doppler and gravitational
frequency shift, as well as the amplification of intensity. The code generates images of
static observers sky and a movie simulations for radially free-falling observers. Tech-
niques of parallel programming are applied to get high performance and fast run of the
simulation code.

1. Introduction
Recent observations indicate the cosmological expansion accelerated by a dark en-
ergy that can be described by a repulsive cosmological constant, A > 0. 4 We in-
vestigate influence of A > 0 on the appearance of distant universe for observers in
close vicinity of the SdS black hole. Visualization outputs of the simulation code,
images and movies, can be downloaded from our web site. 4

2. Optical projection in Schwarzschild-de Sitter spacetimes

Construction of relativistic optical projection consists in finding all null geodesics
connecting the source and the observer. In SdS spacetimes, the null geodesics are
characterised by the impact parameter b, defined as the ratio of motion constants
by the relation b == i!> / [;.5,6 The motion of photons in SdS spacetimes is governed
by the Binet formula 1 ,6
d¢ = ± 1
(1)
du y'b- 2 - u2 + 2u3 + y'
where u = r -1 and dimensionless cosmological parameter reads y = AM2. Here M i
is mass of the central black hole and A rv 10- 56 cm- 2 is the repulsive cosmological

-The present work was supported by the Czech grants MSM 4781305903 and LC06014 (P.B.)

1546
 1547

constant. 5 ,6 The critical impact parameter be = corresponds to

the circular photon geodesic. 5 ,6 Photons coming from distant universe with b < be
finish in the central singularity, while photons with b > be return to the cosmological
horizont. 1 ,6 An image will be observed in the direction determined by the space part
of 4-momentum of photon, tangent to the photon trajectory. In order to obtain the
optical projection for a given observer, the 4-momentum of the photons has to be
transformed into observer's local frame. The solution expressing b as a function of
the source angular coordinate and of the image order 1 has to be done numerically.

3. Apparent angular size of the black hole.

3.14,-----,----,-------,----,
Pure Schwarzschild
A=5.10- 3
A =10-2
A=lO""l

2.36

:g'" '"
(5
.c
'2
:0
'2
:0
'0 '0
:<l :<l
'w 'w 1.57
@ @
~ s

E
ffi ::
E
~ i!1
[
~'" ~

0.79

0.00 '-----'~----'-------'----'
10 20 o Photon orbit 10 20
Radial coordinate Radial coordinate

Fig. 1. Apparent angnlar size of the black hole as function of observer's radial coordinate. Left
panel: Static observers. Right panel: Radially free-falling observers

The apparent angular size S of the black hole can be naturally defined as the
observed angular size of the circular black region on the observer sky, in which
no images of distant objects exist and only radiation originated under the circular
photon orbit can be observed. 1 ,2,6 .For observers located above the circular photon
orbit the boundary of the black region corresponds to the outgoing geodesics with
impact parameter approaching be from above, while for observers located under the
circular photon orbit the boundary corresponds to the ingoing geodesics with the
impact parameter approaching be from below.
1548

In the case of static observers we have 1

S = 2 arccos A(robs, y; b) A(r,y;b) =±V1- :: (1- ~ - yr2). (2)

Here +' and '- 's signs correspond to observers located above and under the circular
photon orbit, respectively. Above the circular photon orbit increasing cosmological
constant causes shrinking of the black region, whereas under the circular photon
orbit the black region grows with increasing cosmological constant. In the limit
case of observers located just on the circular photon orbit, S is independent of the
cosmological constant. It is invariably 'iT, i.e., the black region always occupies just
one half of the observer sky.
In the case of observers radially free-falling from the static radius,1,5,6 the ap-
parent angular size of the black hole reads 1

(Z(r O bS ' y) + )1 - 3y 1/3 A(robs, y; b))

S = 2 arccos , (3)
( )1 - 3y 1/3 + Z(robs, y)A(robs, y; b))
where

Z(r, y) =V~ + yr 2 - 3y 1/3. (4)

Dependency on the cosmological constant is qualitatively different. For radially free-

falling observers S grows with increasing cosmological constant at all values of the
radial coordinate except the central singularity, where S is invariably 'iT, similarly to
the case of static observers located on the circular photon orbit. Consequently, the
radially free-falling observer will always observe smaller S then the static observer
on the same radial coordinate.

References
1. P.Bakala, P. Cermak, S. Hledik, Z. Stuchlik, K. Truparova Plskova : A virtual trip
to the Schwarzschild-de Sitter black hole. Proceedings of RAGtime 6/7: Workshop on
blackholes and neutron stars, Opava, September 2004/2005
2. C.T. Cunningham: Optical Appearance of Distant Observers near and inside a
Schwarzschild Black Hole. Phys .Rev. D.12, 323-328, 1975.
3. R. J. N emiroff: Visual distortion near a neutron star a and black hole. arXi v:
astro-ph/9312003, 1993.
4. Relativistic and particle physics and its astrophysical aplications, Czech research
project MSM 4781305903 http://www.physics.cz/research/.
5. Z. Stuchlik and. S. Hledfk : Some properties of the Schwarzschild-de Sitter and
Schwarzschild-anti-de Sitter spacetimes. Phys. Rev. D, 60(4):0044006(15 pages), 1999
6. Z. Stuchlik and K. Plskova : Optical apperance of isotropic ally radiating sphere in the
Schwarzschild-de Sitter spacetime. Proceedings of RAGtime4/5 , 167-185, 2004
 SIMILARITY SOLUTIONS USING DESOLV

K. T. VU*, J. BUTCHER and J. CARMINATI**

* F.I.S Card Services Asia Pacific,
Melbourne, Australia
** Mathematics and Computational Theory Group,
School of Information Technology,
Deakin University, Australia

We present and describe new reduction routines included in DESOLV which may allow
the complete automation of the determination of similarity solutions of partial differential
equations.

1. Introduction
The Maple routines within DESOLV should prove quite useful to the large com-
munity of Maple users who wish to investigate invariant solutions of differential
equations. Specifically, in many "simpler" cases, the determination of the similarity
groups and the subsequent reduction and integration of partial differential equa-
tions can be almost fully automated with, essentially, just the use of liesolve and
(possibly) dsolve or pdsolve. However, we wish to point out that finding the full
loS-parameter symmetry group of the Vaidya system is quite difficult and in a com-
parative study, it was shown that DESOLV was the only package that was able
to accomplish this without any intermediate user intervention. 1 In this article, we
present and describe new functions in DESOLV which automate the reduction of
variables of a system of PDEs given that it admits a Lie symmetry. We then show
how this leads directly to invariant solutions with very little to no user intervention.

2. Applications: Burgers' and Nonlinear Wave equations

Burgers' equation, U xx + UU x + Ut = 0, admits the five symmetry vector fields: 8 x ,
tax + 8 u , 8 t , x8x + 2tat - u8'L) and xtax + t 2 8 t + (x - tu)8u . These can be readily
obtained using pre-existing functions in DESOLV. We shall focus on the symmetry
xta x + t 2 8 t + (x - tu)8u and in a highly automated fashion determine the invariant
solutions.
> de:= [diff(u(x,t),t) + u(x,t)*diff(u(x,t),x) =
diff(u(x,t) ,x,x)];

8 8 82
de := [(8t u(x, t)) + u(x, t) (8x u(x, t)) = 8x 2 u(x, t)]

> sol:=liesolve(de,u(x,t),switch='selectsym');

symmetry #1

1549
 1550

8
symmetry #3 =, at P-

> select symmetry?(l .. 5,or c for combine)c;

a comb'lnation symmetry is in the form

al * SYM#l + a2 * SYM#2 + a3 * SYM#3 +

a4 * SYM#4 + a5 * SYM#5

> enter al?O;

> enter a2?O;
> enter a3?O;
> enter a4?O;
> enter a5?1;

a combination symmetr-y is

,r+_C2t)
-tanh (
C + x _Cl
sol := u(:1:, t) = 2 1t
_Cl t
A form of a nonlinear wave equation is, with c(u) i- 0,
2

In the group classification process (which can be totally done by DESOLV 's classify
routine), three important forms of c( u) occur, of which one is c( u) = a( u + b) k.
 1551

2.1. c(u}=a(u+bl ; a i:- 0, constant

In the Maple work below, we have arbitrarily chosen one of the symmetries to find
the corresponding invariant solutions.
> alias(U=u(x,t»:
> del:=Diff(U,t,t) =Diff(a-2*(U+b)-(2*k)*Diff(U,x) ,x);

del :=~U=~(a2(U+b)(2k)(~U))
at
2 ax ax
> sol:=liesolve([value(del)] ,[U] ,switch='selectsym');

symmetry #1

a
symmetry #3 =, ax P-

> select symmetry?(l .. 4,or c for combine)2;

&where

W _l(K -1) ka )
(J -
V_a(2+2k) + _Cl _a(-2k)
d_a - K _1 - _C2 = 0 &where

([K_l ~x, W_l ~(U+b)tmlJ}

References
1. J. Butcher, J. Carminati and K.T. Vu. A comparative study of some computer algebra
packages which determine the Lie point symmetries of differential equations. Comput.
Phys. Comm. 155 (2003) 92-114
2. W. F. Ames, R. J. Lohner, and E. Adams. Group properties of Utt = [f(u)uxl x . Int.
J. Nonlin. Mech. 16, (1981) 439-447.
 x TENSOR: A FREE FAST ABSTRACT TENSOR MANIPULATOR

JOSE M. MARTIN-GARCIA
Instituto de Estructura de la Materia, CSIC, Cj Serrano 123, Madrid 28006, Spain
[email protected]

The package xTensor is introduced, a very fast and general manipulator of tensor expres-
sions for Mathematica. Manifolds and vector bundles can be defined containing tensor
fields with arbitrary symmetry, connections of any type, metrics and other objects. Based
on the Penrose abstract-index notation, xTensor has a single canonicalizer which fully
simplifies all expressions, using highly efficient techniques of computational group theory.
A number of companion packages have been developed to address particular problems in
General Relativity, like metric perturbation theory or the manipulation of the Riemann
tensor.

1. Design of x Tensor
xTensor 1 has two main advantages for computations in General Relativity: its
adapted mathematical structure and its efficiency, defined as effective polynomic
scaling of timings in the number of indices at hand (naive algorithms are exponen-
tial). Being constructed on top of Mathematica,2 xTensor allows the users to take
advantage of the nearly 2000 already available built-ins, becoming both a compati-
ble and complementary extension of that widespread computer algebra framework.
This short contribution presents some basic ideas on the structure of the system.
Data types: .TTensor defines a coherent mathematical structure following the
usual cycle "let X be a type with properties props; perform this computation with
it". This is implemented through type declarations as

DefType[ X, props]
computation involving X

where the allowed types of objects are currently: ConstantSymbol (a symbol act-
ing as a constant under any derivative); Parameter (a real parameter, like time in
Newtonian physics); Manifold (a smooth differential manifold); VBundle (a real or
complex vector bundle, not necessarily the tangent bundle of a manifold); Tensor (a
tensor field on the manifolds, with indices on the vector bundles, including metrics
and scalars); CovD (a covariant derivative on any vector bundle, with or without tor-
sion/curvature, metric-compatible or not); ScalarFunction (a function of a number
of scalar expressions, like Sin or Power ); InertHead (a general concept of modifier
of a tensor expressions, like TraceFreePart). Other types can be included if required.
A generic expression is any consistent polynomial of those basic expressions.
Abstract indices: x TensoT" follows strictly the convention of abstract indices,3
which allows fully geometrical computations, without loosing the computational
power of indices. 4 Any vector bundle has its own collection of abstract indices. A
tensor field Tab is represented as the Mathernatica expression T [a, -b], both inter-
nally and as user input, with no parser in between, so that the user always knows

1552
 1553

how to interact with such an expression. The printed expressions are formatted with
sub- and superindices as expected. Covariant derivatives (including the particular
case of partial or ordinary derivatives, which are treated as derivatives with zero
torsion and zero curvature 5 ) are represented as operators acting on tensor expres-
sions: CD [-a] [ w [b] ] encodes the expression \7 awb where the symbol CD with type
CovD represents the connection \7 a. Several tensor fields, connections and metrics
can coexist in the same computation.
Canonicalization of expressions in xTensor is performed using a single algo-
rithm, called ToCanonical, which brings any expression into its unique canonical
form with respect to all permutation symmetries of all tensors and connections
involved. It uses highly efficient algorithms (with effectively polynomic scaling in
time) from Computational Group Theory, being much faster than the canonicalizer
of previous tensor computer algebra systems. For example, an increase by a factor 20
with respect to Math Tensor has been reported in checking that the general source of
a standard form in harmonic gauge for the Einstein equations in post- Minkowskian
theory has zero divergence at the 5PM order; the 6PM check, which had not been
possible before, has been achieved with xTensor for the first time. 6 Other examples
are given in the previous contribution by the author to these proceedings.
Finally, we shall mention that xTensor is well documented in several
Mathematica-executable notebooks and has been already tested in a number of
projects by several users.

2. The xAct project

The package xTensor is the flagship of a bigger project, called xAct, formed by a
suite of packages dedicated to different applications of tensor calculus. They are all
free software (GPL license) and can be downloaded from the author's webpage. 1
xAct can be separated into two parts: there is a computational kernel formed by:

xPerm: This is the central core for permutation-related computations. It imple-

ments algorithms for construction of strong generating sets of any subgroup
of the symmetric group Sn of permutations of n elements and the algorithms
for canonicalization of members of cosets and double cosets, both required
for canonicalization of tensor expressions with respect to monoterm sym-
metries. 7 To increase its speed, the most critical part of the package has
been recoded in some 2000 lines of C code, and compiled. The function
ToCanonical heavily relies on this package.
xTensor: This is the kernel for abstract tensor manipulations, as described in the
previous section, and the basis for any tensor computation in the system.
xCoba: Developed in collaboration with David Yllanes. Contains a large number of
tools to perform component computations, using both coordinate charts and
anholonomic frames, but all based on the abstract framework provided by
x Tensor. This ensures full compatibility between abstract and component
computations.
1554

Then there are other packages for special problems and applications:

xPert: Developed in collaboration with David Brizuela and Guillermo Mena

Marugan. 8 This package implements fast algorithms for high-order metric
perturbation theory in General Relativity. In particular, the expression of
the nth-order perturbation of any relevant tensor (inverse metric, Christof-
fel, Riemann, Ricci, RicciScalar, Einstein, Weyl) can be given (in seconds
for n s:5) in terms of perturbations of the metric.
Harmonics: Also developed in collaboration with David Brizuela and Guillermo
Mena Marugan. 8 A number of different forms of tensor spherical harmonics
have been coded up, and an algorithm for fast expansion of the product
any two of them is provided.
Invar: Developed in collaboration with Renato Portuga1. 9 This package is a generic
manipulator of a metric-compatible Riemann tensor. The permutation sym-
metries of the Riemann tensor are handled in real time using ToCanonical;
Concerning multiterm symmetries, a database with relations among poly-
nomic expressions of the Riemann tensor, the metric, and the totally an-
tisymmetric tensor is used. Currently the database contains all relations
among all algebraic (non-differential) Riemann scalars up to degree 7, in-
cluding all dimensionally dependent relations for dimension 4, with any
metric signature. Extensions to differential invariants and expressions with
free indices are currently under development. There is also a Maple version. 9
Spinors: Under development in collaboration with Alfonso Garcia-Parrado. This
package encodes a number of tools for efficient manipulation of four-
dimensional spinor theory, allowing both abstract computations (through
xTensor) and component-based computations (through xCoba).

Acknowledgments
This work has been supported by the I3P framework of CSIC and the European
Social Fund, and by the Spanish MEC Project No. FIS2005-05736-C03-02.

References
1. xTensor: a fast manipulator of abstract tensor expressions in Mathematica, J. M.
Martin-Garcia, GPL 2002-2007, http://metric. iem. csic. es/Martin-Garcia/xAct/
2. S. Wolfram, The Mathematica Book, 5th ed. (Wolfram Media, 2003).
3. R. Penrose. In Battele Rencontres, ed. C. DeWitt and J. A. Wheeler, New Your 1969,
Benjamin.
4. A. Ashtekar, G. T. Horowitz and A. Magnon-Ashtekar, Gen. Rei. Grav. 14,411 (1982).
5. General Relativity, R. M. Wald, Chicago 1984, The University of Chicago Press.
6. Guillaume Faye, private communication.
7. R. Portugal et al., Internat. J. Modern Phys. C 13 (2002) 859.
8. D. Brizuela, J. M. Martin-Garcia and G. A. Mena Marugan, Phys. Rev. D 74 (2006)
044039.
9. J. M. Martin-Garcia, R. Portugal and L. Manssur, to be published in Camp. Phys.
Comm. (2007).
 TENSOR COMPUTER ALGEBRA

JOSE M. MARTIN-GARCIA
Instituto de Estructura de la Materia, CSIC., Cj Serrano 123, Madrid 28006, Spain
[email protected]

Tensor Computer Algebra is an essential tool in those areas of General Relativity deal-
ing with long and/or complicated expressions. Very fast algorithms have been recently
introduced to simplify tensor expressions, based on the use of efficient techniques of
Computational Group Theory, allowing us to address a much wider class of problems.

1. Computer Algebra and its limitations

Roughly speaking, Computer Algebra (CA from now on) can be defined as the ma-
nipulation of mathematical expressions in a computer. (See the book by Gathen and
Gerhard 1 for an comprehensive review.) Its most distinctive feature is the avoidance
of truncation of information, as opposed to Numerical Computing where floating
point numbers are frequently truncated to fit the allocated memory. This often leads
to a fast growth in size of the expressions being manipulated in CA (with an as-
sociated increase of computing time to manipulate such large expressions), setting
then a harsh limit on the problems that we can address.
A second type of limitation, of theoretical origin, is given by the Richardson's
theorem 2 : Let R be the class of expressions generated by a) the rational numbers and
the two real numbers 'if and log 2, b) the identity function f(x) = x, c) the sin x, eX
and Ixl functions, and d) the operations of addition, multiplication and composition;
then there are expressions E in the class R for which E == 0 is undecidable. Because
those generators are contained in most CA systems, this theorem might seem to
predict absolute failure for these systems. Fortunately there are large classes of
expressions that can always be brought to some canonical form, such that equality of
two different expressions can be easily decided. For example, the set of polynomials is
clearly one of those classes. The algorithm or program in charge of canonicalization
is called the canon'icalizeT and is one of the main components of a CA system.
Note that canonicalization does not necessarily imply simplification, in the sense of
reducing the size of an expression. The latter is usually a much more difficult task.

2. Tensor Computer Algebra

Several areas of General Relativity involve large tensor computations which can
be done neither safely nor efficiently by hand: high-order perturbative expansions
(including Post-Newtonian expansions), the manipulation of polynomials of the Rie-
mann tensor and its derivatives, the classification of metrics, or the expansion of the
Einstein equations in component form without symmetry assumptions (required in
Numerical Relativity), to name a few. The use of CA techniques for tensor computa-
tions in GR (TCA in short) started with the pioneering work of R. d'Inverno in the
late 60's, who created the system ALAM, later extended by him and other authors

1555
1556

during the 70's and renamed as SHEEP, the basis for many other developments,3
all LISP based. Now there are many TCA systems (too many to list here), with
different strengths and weaknesses, but clearly the most used systems in the GR
community are GRTensor 4 (available for both Maple and Mathematica) for com-
ponent computations, and MathTensor 5 (which runs only on top of Mathematica)
for symbolic manipulations.
There are two different sources of complexity in tensor computations. We may
need to work with long expressions formed by the sum of many terms, each one
to be canonicalized independently; or we may have that one of those terms is a
tensor product of several tensors with complicated symmetry properties. The former
problem requires timings and memory sizes which are proportional to the number of
terms, and hence can be dealt with efficiently. The latter problem is harder because
canonicalizing a term requires permuting its indices, what frequently involves huge
subgroups of the symmetric group 5 n , where n is the number of indices participating
in the symmetry properties of the term. For instance a scalar formed by product
of 7 Riemann tensors requires the use of a subgroup of 5 28 with more than 1010
elements (though still small in comparison with the 28! ;::::; 3.10 29 elements of 5 28 ).
Most TCA systems use simple canonicalization algorithms which list all equivalent
permutations of indices to find which is the canonical one, what makes them very
slow. It is possible, however, to use very efficient algorithms to deal with a restricted
type of symmetry. There are two types of symmetry to consider:

• Permutation or monoterm symmetries: the tensor is invariant or changes

sign under permutations of indices. Examples for the Riemann tensor are:

Rbacd = - Rabcd, Rcdab = Rabcd. (1)

• Cyclic or multiterm symmetries: A linear combination of (more than two)

tensors with different index configurations is zero. The Bianchi identities
for the Riemann tensor are examples of this type:

Rabcd + Racdb + Radbc = O. (2)

Multiterm symmetries can be properties of a given tensor, as in the pre-
vious example, or dimension-dependent identities (also known as Lovelock
identities), valid for all tensors but only in certain dimensions. 6

R. Portugal and collaborators 7 have recently shown how to use highly efficient
algorithms of Computational Group Theory to solve any canonicalization problem
involving only permutation symmetries. Their algorithms have been implemented
in the TCA systems Canon 8 for Maple and xTensor 9 for Mathematica, both free
software. Figure 1 shows timings (computed with xTensor) of canonicalization of
scalars formed by product of seven Riemann tensors. As an example, the scalar
n
R bjdk R n c el Rhg ia Rk 9 de Rl j j c Rmaj h R m ib (3)
9
takes 0.0310s to canonicalize in xTensor and 19596s in MathTensor. 5
 1557

There is no known algorithm, however, to solve efficiently nontrivial problems

with multiterm symmetries. The TCA systems Tools of Tensor Calculus 10 and
cadabm ll contain general methods to manipulate multiterm symmetries, but with
no automatic implementation of dimension-dependent identities.

150000 6
A million
4
A
125000 Riemann 7 2
3.2 GHz 0.5 1 1.5 2
100000
Time (s)

75000
Nonzero: 558673
bin .005 s
50000 Zero: 441324

25000

0
0 0.02 0.04 0.06
Time (s)
0.08 ------ 0.1 0.12

Fig. 1. Histogram of canonicalization timings of one million random scalar products of 7 Riemann
tensors, computed with xTensor. 9 We have separated those cases which canonicalize to zero from
those which do not. There are only 2 cases taking longer than 1 second, as shown in the zoom box.

Acknowledgments
This work has been supported by the 13P framework of CSIC and the European
Social Fund, and by the Spanish MEC Project No. FIS2005-05736-C03-02.

References
1. Modern Computer Algebra, J. von zur Gathen and J. Gerhard, Cambridge, CUP 2003.
2. D. Richardson, The Journal of Symbolic Logic 33 (1968) 514
3. Algebraic Computing in General Relativity, M. MacCallum et al. eds., OUP 1995.
4. GRTensorII, P. Musgrave et al., http://grtensor.phy.queensu.ca/
5. MathTensor, L. Parker and S. Christensen, http://smc.vnet.net/MathTensor.html
6. S. B. Edgar and A. Hoglund, J. Math. Phys. 43 (2002) 659
7. R. Portugal et al., Internat. J. Modern Phys. C 13 (2002) 859 ..
8. Canon, R. Portugal, http://www .lnce. br / "-'portugal/Canon. html
9. xTensor, J. M. Martin-Garcia, http://metrie. iem. esie. es/Martin-Garcia/xAet/
10. TTC, A. Balfag6n, P. Castellvi and X. Jaen, http://baldufa.upe.es/xjaen/ttc/
11. cadabra, K. Peeters, http://www.aeLmpg.de/rvpeekas/eadabra/
This page intentionally left blank
 Simulations of Relativistic
Flows and Compact Objects
This page intentionally left blank
 RELATIVISTIC MHD SIMULATIONS AND SYNTHETIC
SYNCHROTRON EMISSION MAPS: A DIAGNOSTIC TOOL FOR
PULSAR WIND NEBULAE

L. DEL ZANNA * and D. VOLPI

Dipartimento di Astranomia e Scienza dello Spazio, Universitti di Firenze,
Largo E. Fermi 2, Firenze, 50125, Italy
* E-mail: [email protected]

E. AMATO
INAF - Osservatorio Astrafisico di Arcetri,
Largo E. Fermi 5, 50125, Firenze, Italy

N. BUCCIANTINI
Astronomy Department, University of California at Berkeley,
601 Campbell Hall, Berkeley, CA 94720-3411, USA

Numerical simulations in the relativistic regime have certainly become a powerful inves-
tigation tool for High Energy Astrophysics. Here we show a simple method to provide
any numerical code with a complete set of diagnostic tools aimed at producing synthetic
synchrotron emissivity, polarization, and spectral index maps for comparison with ob-
servations. The method is based on the addition to the numerical code of an evolution
equation for the maximum energy of the emitting particles, and this takes into account
both adiabatic and radiative losses along fluid streamlines. As an application, we con-
sider here the case of Pulsar Wind Nebulae (PWNe), for which the method is applied
on top of an axisymmetric MHD model. Results are presented and discussed.

Keywords: Relativity; Magnetohydradynamics (MHD); Pulsars: general; Radiation mech-

anisms: non-thermal.

1. The synchrotron emission recipes

Non-thermal radiation, and synchrotron emission from relativistic particles in mag-
netized plasmas in particular, is a Ubiquitous signature in high energy astrophysics.
For optically thin sources, the emission spectrum, typically a power law, is related
to the particles' distribution function at injection through well known transport
and radiation physics. However, the connection may be not easy to establish quan-
titatively in the presence of multi-dimensional flow patterns, making it difficult
to extract from observations direct information on the most basic physics of the
sources. Nowadays, multi-dimensional simulations provide a fundamental help to-
wards the understanding of the complex physical phaenomena involving relativistic
plasmas in many astrophysical situations. Here we summarize the main results by
Del Zanna et al} where a simple diagnostic tool to compute synchrotron bright-
ness, polarization and spectrum index maps associated with the plasma dynamics
resulting from a given MHD simulation was descibed. The method proposed can be
applied to any scheme for relativistic MHD (not necessarily in conservative form,
either in full 3-D or under particular symmetries) and it just requires the numerical

1561
1562

solution of an additional evolution equation.

Let us suppose an injection of emitting particles at some location (e.g. shock
fronts) with an initial distribution function which is power law in energy and
isotropic in momentum. Under the assumptions of quasi-stationarity and of neg-
ligible synchrotron losses (from the point of view of the overall dynamics), the local
emission coefficient can be written as
piE' x if'I<>+lD<>+2 v -<>, v oo :2 v,
jv (v, if) ex: (1)
{ 0, Voo < v.
where v is the frequency of observation, ii the direction of the observer, a the spec-
tral index, p is the thermal pressure, 13 the magnetic field vector, D the Doppler
boosting factor, and a prime indicates quantities measured in the local frame co-
moving with the fluid. Synchrotron burn-off is taken into account via the cut-off
frequency

Voo - - I~'
= D 3e B x n~/I Eoo
2
' (2)
4nmc
where Eoo is the the maximum particle's energy (in units of the rest mass energy
mc2 ) attainable locally. This quantity must be evolved directly by the numerical
scheme via integration of an additional equation (here written in conservative form,
as needed by shock-capturing schemes) which takes into account both adiabatic and
synchrotron losses

a
-("(pE) + \7. ("(pEv)~ = -p
2/3 4e
4
--B' E
2 2
(3)
at 9m3c5 00'

with E = Eool pl/3, where p is the mass density in the comoving frame, iJ is the
flow velocity, "( the Lorentz factor. Given the local emissivity, surface brightness,
polarization, and spectral index maps are easily computed via integration along the
line of sight.
As a first application of our techniques we have studied in detail the emission
properties of Pulsar Wind Nebulae (PWNe).l These are computed following the
prescriptions of a previously presented axisymmetric model, 2 based on a relativis-
tic, magnetized pulsar wind with anisotropic energy flux (increasing towards the
equator) interacting with expanding supernova remnants. Simulations have been
performed with the special relativistic code by Del Zanna,3,4 recently extended to
GRMHD and named ECHO (Eulerian Conservative High Order scheme 5 ). This
work follows our previous investigations on the physics and emission properties
of rotating neutron stars and their nebulae. 6 - 12 In our model, emitting particles
are produced continuously at the termination shock of the pulsar wind and then
advected by the nebular, mildly relativistic flow. In Fig. 1 an example of surface
brightness map in the X-rays is shown, to be compared with Chandra images of the
Crab Nebula. 13 Note the presence of an equatorial torus and of two polar jets, the
latter being the result of flows induced by magnetic hoop stresses, and of Doppler
 1563

-3.0 -2.5

Fig. 1. Simulated synchrotron surface brightness map of the Crab Nebula in the X-ray band.

boosted features like the main aTe and the knot, also found in real images. The
best agreement between simulations and data is obtained for a wind magnetization
(J ;::::;; 0.02, a value much higher than the constraint imposed by radial models, peak-

ing near the equator. Such result confirms the importance of providing relativistic
MHD simulations with synchrotron diagnostic tools.

References
1. L. Del Zanna, D. Volpi, E. Amato and N. Bucciantini, ABA 453, 621 (2006).
2. L. Del Zanna, E. Amato and N. Bucciantini, A€1A 421, 1063 (2004).
3. L. Del Zanna and N. Bucciantini, ABA 390,1177 (2002).
4. L. Del Zanna, N. Bucciantini and P. Londrillo, ABA 400, 397 (2003).
5. L. Del Zanna, O. Zanotti, N. Bucciantini and P. Londrillo, ABA submitted (2007).
6. N. Bucciantini, J. M. Blondin, L. Del Zanna and E. Amato, ABA 405,617 (2003).
7. N. Bucciantini, R. Bandiera, J. M. Blondin, E. Amato and L. Del Zanna, ABA 422,
609 (2004).
8. N. Bucciantini, E. Amato, R. Bandiera, J. M. Blondin and L. Del Zanna, ABA 423,
253 (2004).
9. N. Bucciantini, E. Amato and L. Del Zanna, ABA 434, 189 (2005).
10. N. Bucciantini, L. del Zanna, E. Amato and D. Volpi, ABA 443, 519 (2005).
11. N. Bucciantini, T. A. Thompson, J. Arons, E. Quataert and L. Del Zanna, MNRAS
368,1717 (2006).
12. N. Bucciantini and L. Del Zanna, ABA 454, 393 (2006).
13. M. C. Weisskopf et. al., ApJL 536, L81 (2000).
 GRMHD SIMULATIONS OF JET FORMATION WITH RAISHIN

Y. MIZUNO and K.-I. NISHIKAWA

National Space Science and Technology Center, Huntsville, AL 35805, USA
[email protected]

S. KOIDE
Department of Physics, Kumamoto University, Kurokami, Kumamoto 860-8555, Japan

P. HARDEE
The University of Alabama, Tuscaloosa, AL 35847, USA

G. J. FISHMAN
NASA/MSFC, NSSTC, Huntsville, AL 35805, USA

We have performed several simulations of black hole systems (non-rotating, black hole
spin parameter a = 0.0 and rapidly rotating, a = 0.95) with a geometrically thin Kep-
lerian disk using the newly developed RAISHIN code. The simulation results show the
formation of jets driven by the Lorentz force and the gas pressure gradient. It appears
that a rotating black hole creates an additional, faster, and more collimated matter-
dominated inner outflow formed and accelerated by the twisted magnetic field resulting
from frame-dragging in the black hole ergosphere. This is the first known simulation
confirming the formation of an inner magnetically-driven, matter-dominated jet by the
frame-dragging effect from a black hole co-rotating with a thin Keplerian disk threaded
by a vertical magnetic field. This result indicates that jet kinematic structure depends
on black hole rotation and on the initial magnetic field configuration and strength.

1. Introduction
Relativistic jets have been observed or postulated in various astrophysical objects,
including active galactic nuclei (AGNs), microquasars in our galaxy, and gamma-ray
bursts (GRBs). The most promising mechanisms for producing the relativistic jets
involve magnetohydrodynamic centrifugal acceleration and/or magnetic pressure
driven acceleration from an accretion disk around the compact objects, or involve
the extraction of rotating energy from a rotating black hole.
In this paper we report on new simulations showing jet formation from the
black hole magnetosphere co-rotating with a thin Keplerian disk threaded by a
vertical magnetic field using the recently developed three-dimensional, GRMHD
code RAISHIN which uses a conservative, high-resolution shock-capturing schemes
(Mizuno et al. 2006a 1 ).

2. Initial Condition
In order to study the formation of relativistic jets from a geometrically thin Kep-
lerian disk, we use newly developed 3-dimensional GRMHD code "RAISHIN" with
Boyer-Lindquist coordinates (r, (), ¢).1 In our present simulations: a geometrically
thin Keplerian disk rotates around a black hole (non-rotating, a = 0.0 or rapidly

1564
 1565

a=O.O, bo=O.05 a=O.95, bo=O.05

time/,s = 275. time/,s = 200.
.0 .0 Vt
14 14 0.6

12 12 0.4
10 10 0.2
8 8
N N 0.0
6 6
-0.2
4 4
2 2 -0.4
0
0 2 4 6 8 10 12 14 2 4 6 8 10 12 14
x x

Fig. 1. Snapshots of total velocity of the non-rotating black hole case (a = 0.0) and the rapidly
rotating black hole case (a = 0.95) at the applicable terminal simulation time. The color scales
show the total velocity. A negative velocity indicates inflow towards the black hole.

rotating, a = 0.95), where the disk density is 100 times higher than the coronal
density. The thickness of the disk is Hlr rv 0.06 at r = lOrs. In the rotating black
hole case the disk is co-rotating with the black hole. The background corona is
free-falling into the black hole (Bondi flow). The initial magnetic field is assumed to
be uniform and parallel to the rotational axis i.e., the Wald solution. These initial
conditions are similar to Koide et al. (1999, 2000)3,4 except with initial magnetic
field strength. The simulations are performed in the region 1.1(0.75)r8 S r S 20.0r8
and 0.03 S () S 1f12. We use 128 x 128 computational zones with logarithmic zone
spacing in the radial direction. 3 We assume axisymmetry with respect to the z-
axis and mirror symmetry with respect to the equatorial plane. We employ a free
boundary condition at the inner and outer boundaries in the radial direction.

3. Results
Figure 1 shows snapshots of the total velocity distribution for the non-rotating black
hole case, a = 0.0 (left panels); and the rapidly rotating black hole case, a = 0.95
(right panels); at each simulation's terminal time (non-rotating: t = 275T8 and
rotating: t 200T8)'
The numerical results show that matter in the disk loses angular momentum
through the magnetic field and falls into the black hole. A centrifugal barrier de-
celerates the falling matter and makes a shock around T 2r8. The matter near
the shock region is accelerated by the J x B force and the gas pressure increased
by the shock, resulting in jet formation. These results are similar to the previous
work of several authors. 3- 5 The jets in both cases have velocities somewhat greater
than O.4c. The result of rapidly rotating black hole case shows a formation of a
two-component jet. The outer jet is similar to that of the non-rotating black hole
case but the inner jet is not seen in the non-rotating black hole case. The inner jet
1566

is faster than the outer jet (over 0.5c).

The magnetic field is twisted near the black hole region strongly and propagates
outward with the jets as an Alfven wave. In the non-rotating black hole case the
magnetic field is twisted by the rotation of the accreting matter and forms an outer
jet. On the other hand, in the rapidly rotating black hole case the magnetic field
is mainly twisted by the frame-dragging effect of the rotating black hole rather
than the rotation of the accreting matter near the black hole region and forms an
additional jet component (inner jet) near the black hole region.
From a simple analysis to confirm the jet accleration mechanism, the outer jet is
accelerated mainly by the gas pressure gradient. The inner part of the outer jet may
be accelerated partially by the Lorentz force but the Lorentz force in the outer part
of the outer jet in the rapidly rotating black hole case is lower than the gas pressure
gradient by an order of magnitude. The inner jet is accelerated by the Lorentz force.
Therefore the acceleration mechanism is different in the inner and outer jets (see
more detail in Mizuno et al. 2006b 2 ).
A two component jet structure has also been seen in GRMHD simulations of a
black hole co-rotating with a thick torus. 6 ,7 One component is a matter-dominated
outflow (funnel wall jet) with a mildly relativistic velocity (~ 0.3c) along the cen-
trifugal barrier accelerated and collimated by magnetic and gas pressure forces in
the inner torus and the surrounding corona. This formation mechanism and jet
are the same as the outer jet seen in our simulations. Their other component is a
highly-relativistic Poynting flux dominated jet that is produced from the formation
of a large scale radial magnetic field within thefunnel. In our simulation, such a
highly-relativistic Poynting flux dominated jet is not seen. This is likely caused by
the difference in the initial magnetic field configuration.

Acknowledgments
Y. M. is a NASA Postdoctoral Program fellow at NASA Marshall Space Flight
Center. The simulations have been performed on the IBM p690 at the National
Center for Supercomputing Applications (NCSA) which is supported by the NSF
and Altix3700 BX2 at YITP in Kyoto University.

References
1. Y. Mizuno, K.-I. Nishikawa, S. Koide, P.Hardee, and G. J. Fishman, APJS, (2006a),
submitted, (Astra-ph/069004)
2. Y. Mizuno, K.-I. Nishikawa, S. Koide, P.Hardee, and G. J. Fishman, APJL, (2006b),
submitted, (Astro-ph/069344)
3. S. Koide, K. Shibata, and T. Kudoh, ApJ 522, 727, (1999)
4. S. Koide, D. L., Meier, K., Shibata, and T. Kudoh, ApJ 536, 668, (2000)
5. K.-I., Nishikawa et al., ApJ 625, 60, (2005)
6. J. F., Hawley and J. Kralik, ApJ 641,103, (2006)
7. J. C., McKinney, MNRAS 368, 1561, (2006)
 STANDING SHOCKS IN PSEUDO-KERR GEOMETRY

SOUMEN MONDAL
S.N. Bose National Centre for Basic Sciences, JD Block, Salt Lake, Sector-III, K olkata 700098,
India
[email protected]

SANDIP K. CHAKRABARTI
S.N. Bose National Centre for Basic Sciences, JD Block, Salt Lake, Sector-Ill, Kolkata 700098,
India
and
Centre for Space Physics, Chalantika 4-3, Caria Station Rd., Kolkata 700084-
[email protected]

We present here a simplified approach to the study of particle dynamics and fluid dynam-
ics around rotating black holes. We show that with a suitably modified effective potential
of the central gravitating rotating object, one can carry out these studies very accurately
in Newtonian formalism provided the Kerr parameter remains within -1 <:; a <:; 0.8. We
present the geometrical quantities useful to study the particle dynamics and the en-
tire parameter space spanned by the specific energy and angular momentum from our
potential and also compare it to the general relativistic result.

1. Introduction

We study the particle dynamics and fluid dynamics around a Kerr black hole
with the help of a pseudo-Kerr potential (Chakrabarti & Mondal, 2006; Mondal &
Chakrabarti, 2006). This potential mimics the Kerr geometry very faithfully as long
as the Kerr parameter lies in the range -1 < a < 0.8 as far as the particle dynamics
is concerned. For fluid dynamical purposes, the pressure of the flow is affected by
the red-shift factor. In a Newtonian description, this boils down to changing the
polytropic index of the infallng gas in a manner so that the flow has similar prop-
erties as in Kerr geometry. We carry out the sonic point analysis and explore the
locations of the sonic points and standing shock waves as functions of the conserved
flow parameters, namely the specific energy and the specific angular momentum As
originally done in Chakrabarti (1989) in the context of Schwarzschild black hole.
We also carry out our analysis of the solution topologies in Kerr geometry.

2. The Pseudo-Kerr geometry

We choose G = c = M = 1 and distances, specific angular momenta and times
are measured in units of GMJc 2 , GMJc and GMJc 3 respectively. The potential we
propose is:

1 2a (T + wr2) (T + wr2) 2
fP(r,O) = 1- - - + +0: 2 , (1)
r - ro r3 sinO 2r2 sin 20
where, 0 is measured from the axis of rotation of the black hole and ro = 0.04 +
0.97 a + 0.085a 2 . The red-shift factor 0: (Eq. 1) and the dragging term, w have been

1567
1568

borrowed from general relativity,

a= £~ and (2)
I;

where, p2 = r2 + a 2cos 2e, I;2 = (r2 + a2 )2 - a 2t::..sin 2e and t::.. = r2 + a 2 - 2r.

l¢ = -u¢ = l if> ef f' is the conserved specific angular momentum for the particle and
l is approximately the conserved specific angular momentum for fluid l = -Uip/Ut.

1.04 ~-----------~

1.02 a=O.5

Ln 098
§O.96
0.9
~O.94
~o 92
.e: 0.9
~o 88 0.8
0.86 _GTR
Pseudo-Kerr
0.84 0.75

10 L
O.7_ , ---:-:0.8;:---~O.-'-6--:'-0~.4--'-0~.2~O---="0.2;:---'-0.-'-4-C0~.6--'O~.8---.J,
r
a

Fig. lea-b): (a) Comparison of the General relativistic Kerr potential VK (solid), the
pseudo-Kerr potential if> P K for a = 0.5. See text for the parameters for which the curves
have been plotted. (b) Plot of marginally stable values of potential if>ms with a.

Sr-~~-~~~~-~~~-.
Im6
4.5
In/S

3.5

2.5

1.5 '--,'-;::---:e-,....--::'-c--c--c-~--::-'-:--:c"-c---:-'-c~:c--'
-I -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1
a a

Fig. 2(a-b): (a) The locations of the marginally bound (rmb) and marginally stable (rms)
orbits at these orbits in both Kerr geometry (solid) and pseudo-Kerr geometry (dashed).
(b) The locations of the marginally bound angular momenta 1mb and Ims at these orbits in
both Kerr geometry (solid) and pseudo-Kerr geometry (dashed). Results generally agree
for -1 < a < 0.8.
 1569

a=O 5

10~ L~~~~s~~wt~
e~TI/2
1~2 9
8~ 10075
8
M
0.9S 15

0.96

0.94

0.5
0.92
(8"".I m .)
0.9 ~ ..L~L ~--"--L~,_,.~I~"-,-,I~J

2.6 2.8 3 3.2 3.4 36 0.5 1.5 25

Fig. 3(a-b): The sub-division of the parameter space spanned by E and l in terms of
whether or not multiple sonic points or shocks may form. The solid curves represent pa-
rameter space for pseudo-Kerr potential and dotted curves for GTR values. The regions
marked 0, I, SA, SW, NSA, NSW contain flow parameters (E, l) such that the flows will
have only the outer sonic point, only the inner sonic point, shocks in accretion, shocks in
winds, 'no-shocks' in accretion and 'no-shocks' in winds respectively. In regions [* and O'
only the inner Cl' topology and outer Cl' type topology are possible. In region N, no solution
is possible. The subscripts ms and mb refer to marginally bound and marginally stable
values respectively for a = 0.5. (b) The solid curves are for the accretion flow parameters
taken from the SA in Figs. l(a). Only the shocks drawn in solid curves in (b) are stable.

3. Discussion
We presented a modified gravitational potential to study particle and fluid dynamics
around a Kerr black hole, which, for all practical purposes, could be used as easily as
a Newtonian potential. This is found to reproduce characteristics of fluid dynamics
very accurately as long as the Kerr parameter -1 < a < 0.8. The errors in this
range, if any are found to be less than five percent or so.

References
1. S. K. Chakrabarti, 1989, Astrophys. J., 347, 365
2. S. K. Chakrabarti, 1993, MNRAS 261, 625
3. S. K. Chakrabarti, S. Mondal, 2006, MNRAS, 369, 976
4. S. K. Chakrabarti & L. G. Titarchuk, 1995, ApJ 455, 623
5. S. K. Chakrabarti, L. G. Titarchuk, D. Kazanas & K. Ebisawa 1996, A & A, Supp!.
Ser., 120, 163
6. K. Ebisawa, L.G. Titarchuk, & S. K. Chakrabarti, 1996, PASJ 48(1), 59
7. S. Mondal, S. K. Chakrabarti, 2006, MNRAS, 371,1418
8. D. M. Smith, W. A. Heindl, J. H. Swank, 2002, ApJ 569, 362
9. 1. V. Artemova, G. Bjoernsson, 1. D. Novikov, 1996, ApJ 461, 565
10. S. K. Chakrabarti, 1990, Theory of Transonic Astrophysical Flows, (Singapore: World
Scientific)
 EVOLVING RELATIVISTIC FLUID SPACETIMES
USING PSEUDO SPECTRAL METHODS AND FINITE
DIFFERENCING

MATTHEW D. DUEZ, LAWRENCE E. KIDDER, SAUL A. TEUKOLSKY

Department of Astronomy, Cornell University, Ithaca, NY 14850 USA
[email protected]

We present a new code for solving the coupled Einstein-hydrodynamics equations to

evolve relativistic, self-gravitating fluids. The Einstein field equations are solved on one
grid using pseudospectral methods, while the fluids are evolved on another grid by finite
differencing. We discuss implementation details, such as the communication between the
grids and the treatment of stellar surfaces, and present code tests.

1. Introduction
Numerical relativity has the potential to make indispensable contributions to our
understanding of hydrodynamic compact object phenomena, such as neutron star-
neutron star (NSNS) binary merger, black hole-neutron star (BHNS) binary merger,
and stellar core collapse. In such systems, both the spacetime metric and the fluid
are dynamical, and they are strongly coupled.
The most common approach to numerically solving the coupled Einstein-
hydrodynamics equations is by finite differencing. Finite difference (FD) techniques
have been successfully used to simulate NSNS binaries, stellar collapse, and other
interesting phenomena. FD algorithms usually converge to the exact solution as
some power of the grid spacing. In addition, techniques have been developed which
can evolve fluids with discontinuities stably and accurately. Unfortunately, FD codes
usually require very large grids in order to obtain accurate results.
Einstein's equations can also be evolved using pseudospectral (PS) methods. For
smooth functions, PS methods converge exponentially to the exact solution as the
number of collocation points is increased. This allows PS methods to get accurate
results with much smaller grids than those used by FD codes. A PS code for solving
the Einstein equations has been developed by the Cornell-Cal tech relativity groupl,2
and successfully used to carry out binary black hole inspiral simulations3 which are
both the most accurate and computationally cheapest of their kind.
There is a difficulty in evolving non-vacuum spactimes spectrally, however. Be-
cause of the possibility of stellar surfaces and shocks, the evolved variables are not
always smooth at all derivatives. In these cases, spectral representations display
Gibbs oscillations near the discontinuity which converge away only like a power of
the number of collocation points, the order of convergence given by the order of the
discontinuity. In some cases, the problem can be avoided by placing domain bound-
aries at discontinuities, but this is not practical for complicated shocks or strongly
deformed stellar surfaces.
Another possibility would be a mixed approach: to evolve the metric fields, which
are much smoother, using PS methods and evolve the hydrodynamic fields using

1570
 1571

shock-capturing FD methods. This would seem to utilize the strongest features of

each method. This approach has been used successfully in a conformal gravity code
to perform stellar collapse computations. 4 Here we extend this approach to full GR.

2. Numerical Algorithm
We integrate the hydrodynamic equations in conservative form. using piecewise
parabolic reconstruction 5 together with a high-resolution central scheme. 6 We use
uniform grids in three or two dimensions (the latter for axisymmetric systems).
The vacuum outside the stars is handled by introducing a tenuous "atmosphere",
together with a density floor and an internal energy ceiling, in these regions. The
Einstein equations are evolved in the generalized harmonic system. 2 We find that
filtering the metric variables is sufficient to stabilize the PS code in the presence of
discontinuities. The interpolation from the spectral to the fluid grid would be very
expensive if done directly, but we make the process much quicker using a technique
introduced by Boyd. 7 Both codes use dual coordinate frames, which allow the grid
to dynamically adjust to the motion of the system. 3
Our finite difference code currently has no adaptive griding capability, but our
two-grid approach gives us a few similar advantages. The finite difference grid need
only cover the region containing the matter- a huge savings in many binary appli-
cations. The dual coordinate frame system allows the finite difference grid to move
with the stars. Also, separate coordinate mappings can be applied to the two grids
so that their resolutions can be controlled separately.

3. Tests
Our FD code has been successfully tested by evolving multi-dimensional shocks. In
order to test the full FD plus PS code, we evolve equilibrium poly tropes. We choose
the domain decomposition of the PS grid to consist of a filled sphere (a "ball")
centered on the star surrounded by several concentric spherical shells. Angular basis
functions are spherical harmonics. For radial basis functions, we use Chebyshev
polynomials on the shells and an appropriate set of functions 8 on the ball.
In Figure 1, we show results for four n = 1 poly tropes. Star A is a stable TOV
star with central density Pel Perit = 0.67, star B is an unstable TOV star with
Pel Per it = 1.33, star C has the same rest mass as star A but rotates uniformly with
an angular velocity 80% of the mass shedding limit, and star C is a hypermassive
rapidly differentially rotating star evolved by other groups.9 We use 45 2 FD grids.
For stars A, B, and C, we choose our PS grids to have two shells, with the inner one
containing the stellar surface. We use spherical harmonics up to L = 7, and we use
7, 9, and 7 radial collocation points in the ball and the two shells, respectively. We
find that the error in the PS grid is dominated by the shell containing the stellar
surface, and it decreases quadratically with the grid spacing in this shell. We add a
1% pressure depletion to test stability. For star D, we choose the ball with spherical
harmonics up to L = 14 and 18 radial points to cover the whole star.
1572

10
- Star A
-- Star B
I" Stare
-- Star D
p/pci

-.-- -,-~~

0 10 20 30 40
1/2
tP ci

Fig. 1. Central density pc as a function of time t for four equilibrium stars, where Pci = Pc(t = 0).

We find that our code can accurately evolve equilibrium stars and distinguish
stable from unstable configurations. We have also successfully evolved moving stars
using the dual coordinate system to track the star's center of mass. These tests
encourage us to think that our code might be able to produce accurate simulations
of NSNS and BHNS binaries.

4. Acknowledgments
This work was supported in part by a grant from the Sherman Fairchild Foundation,
by NSF grants DMS-0553677, PHY-0354631, and NASA grant NNG05GG51G.

References
1. L. E. Kidder, M. A. Scheel, S. A. Teukolsky, E. D. Carlson, and G. B. Cook, Phys.
Rev. D 62, 084032 (2000).
2. L. Lindblom, M. A. Scheel, L. E. Kidder, R. Owen, and O. Rinne, Class. Quant. Grav.
23, S447 (2006).
3. M. A. Scheel, et ai, gr-qc/0607056.
4. H. Dimmelmeier, J. Novak, J. A. Font, J. M. Ibanez, and E. Muller, Phys. Rev. D 71,
064023, (2005).
5. P. Colella and P. R. Woodward, J. Comput. Phys. 54, 174 (1984).
6. A. Kurganov and E. Tadmor J. Comput. Phys. 160, 241 (2000).
7. J. P. Boyd, J. Comput. Phys. 103, 243 (1992).
8. T. Matsushima and P. S. Marcus, J. Comput. Phys. 120, 365 (1995).
9. T. W. Baumgarte, S. L. Shapiro, and M. Shibata, Astrophys. Lett. 528, L29 (2000).
 3D RELATIVISTIC MHD SIMULATION OF A TILTED
ACCRETION DISK AROUND A RAPIDLY ROTATING BLACK
HOLE*

P. CHRIS FRAGILE
Dept of Physics (3 Astronomy, College of Charleston, Charleston, SC 29424 USA
[email protected]

PETER ANNINOS
Lawrence Livermore National Laboratory, Livermore, CA 94550 USA

OMER M. BLAES
University of California, Santa Barbara, CA 93106 USA

JAY D. SALMONSON
Lawrence Livermore National Laboratory, Livermore, CA 94550 USA

We posit that accreting compact objects, including stellar mass black holes and neutron
stars as well as supermassive black holes, may undergo extended periods of accretion dur-
ing which the angular momentum of the disk at large scales is misaligned with that of
the compact object. In such a scenario, Lense-Thirring precession caused by the rotating
compact object can dramatically affect the disk. In this presentation we describe results
from a three-dimensional relativistic magnetohydrodynamic simulation of an MRI tur-
bulent disk accreting onto a tilted rapidly rotating black hole. For this case, the disk does
not achieve the commonly described Bardeen-Petterson configuration; rather, it remains
nearly planar, undergoing a slow global precession. Accretion from the disk onto the
hole occurs predominantly through two opposing plunging streams that start from high
latitudes with respect to both the black-hole and disk midplanes. This is a consequence
of the non-sphericity of the gravitational spacetime of the black hole.

1. Introduction
Recently, our group has been working on three-dimensional, fully relativistic simu-
lations of tilted thick disks, first in the hydrodynamic limit 1 and here in full general
relativistic MHD (GRMHD). A general relativistic treatment ensures that all im-
portant relativistic features, such as the cusp in the potential and the Einstein and
Lense-Thirring precessions of the orbits, as well as any higher order couplings of
these features, are treated properly. The inclusion of magnetic fields is important
because it is now widely believed that local "viscous" stresses are generated by
turbulence that results from the magnetorotational instability (MRI).2
This work is carried out using the Cosmos++ astrophysical magnetohydrody-
namics code. 3 For this work, the key features of Cosmos++ are the flexible mesh

• This research has been partially supported by a REAP grant from the South Carolina Space Grant
Consortium. It was also supported under the following NSF programs: Partnerships for Advanced
Computational Infrastructure, Distributed Terascale Facility (DTF) and Terascale Extensions:
Enhancements to the Extensible Terascale Facility.

1573
1574

structure, adaptive gridding, and three-dimensional general relativistic MHD capa-

bilities. This simulation is carried out on a spherical-polar mesh with an equivalent
resolution of 64 3 zones. The initial conditions of the disk are the same as model
KDP of De Villiers, Hawley, & Krolik: 4 aiM 0.9 and TPmax 25Tc. The princi-
ple difference is that in this work the black hole is tilted by 15°. The full set up of
this problem and higher resolution simulations will be described in a forthcoming
paper (Fragile et al., in preparation).

2. Results
Perhaps the most striking feature of the simulated disk at late times are the two
opposing plunging streams that start from high latitudes both with respect to the
black-hole symmetry plane and the disk midplane. Figure 1 shows the structure of
the full disk and a zoomed in view of the plunging region with the plunging streams.
Note that one stream remains entirely above the black-hole symmetry plane, while
the other remains below. As material passes through this plunging stream it under-
goes strong differential precession. The precession totals approximately 180°; thus,
the material enters the black hole from the opposite azimuth from which it began
its plunge.

Fig.1. Logarithmic density plot from the final data dump (t 10 orbits at rpma ",). This figure
is oriented such that the black-hole spin axis points straight up. The z-axis of the triad gives the
original orientation of the disk angular momentum vector. The main body of the disk shows very
little warping or realignment toward the symmetry plane of the black hole. The plunging region
(shown in inset) shows two opposing, high-latitude streams of material connecting the disk to the
horizon (indicated by arrows).

The interesting question from Figure 1 is why do these opposing accretion

streams start from such high latitude with respect to the black-hole symmetry
plane and disk midplane? Fundamentally this is due to the aspherical nature of the
gravitational spacetime of the rotating black hole. A nice way to illustrate this is
 1575

to consider the dependence of rISCO, the coordinate radius of the innermost stable
circular orbit, on inclination. Figure 2 illustrates this dependence for a few selected
black-hole spins. The key point of the plot is that orbital stability around a rotat-
ing black hole is strongly dependent on the inclination of the orbit. Notice that the
unstable region increases monotonically for increasing inclination.

160

140

120

.~ 100
~ SO

60

40
,
i
20

4 5 6
2
risco (GMfc )

Fig. 2. Plot of the inclination dependence Fig. 3. Meridional plot (rp 0) through the
of rISCO for black-hole spins aiM 0, 0.5, final dump of the simulation showing the log-
0.9, and 0.998. Inclinations 0 :s; i :s; 90° rep- arithm of density overlaid with a polar plot of
resent prograde orbits, whereas inclinations the "ISCO surface" for an aiM = 0.9 black
90° :s; i :s; 180° represent retrograde orbits. hole. Notice that the plunging streams start
near the largest cylindrical radius (r cos'19) of
this surface (indicated by arrows).

We can make better use of the information in Figure 2 by converting it to a polar

plot and overlaying it on a plot of data from the simulation, as is done in Figure 3.
The polar plot creates an "ISCO surface" , which gives a clear indication of where the
most unstable regions are. Note that the plunging orbits highlighted previously start
near where the disk first encounters the ISCO surface. More precisely, the streams
start near the largest cylindrical radius (r cos '!9) of the ISCO surface, measured
with respect to the angular momentum axis of the disk. The increase in rISCO
with inclination then explains why these plunging orbits start at high inclinations
relative to the black-hole symmetry plane and the disk midplane.

References
1. P. C. Fragile and P. Anninos, ApJ 623, 347 (2005).
2. S. A. Balbus and J. F. Hawley, ApJ 376, 214 (1991).
3. P. Anninos, P. C. Fragile and J. D. Salmonson, ApJ 635, 723 (2005).
4. J. P. De Villiers, J. F. Hawley and J. H. Krolik, ApJ 599, 1238 (2003).
 FRAGMENTATION OF GENERAL RELATIVISTIC
QUASI-TOROIDAL POLYTROPES

BURKHARD SEBASTIAN ZINK

Center for Computation and Technology
and
Horace Hearne Jr. Institute for Theoretical Physics
Louisiana State University, Baton Rouge, LA 70803, U.S.A.
[email protected]

NIKOLAOS STERGIOULAS*, IAN HAWKEt, CHRISTIAN D. OTT", ERIK SCHNETTER§,

and EWALD MULLER'

How do black holes form from relativistic stars? This question is of great fun-
damental and practical importance in gravitational physics and general relativistic
astrophysics. On the fundamental level, black holes are genuinely relativistic objects,
and thus the study of their production involves questions about horizon dynamics,
global structure of spacetimes, and the nature of the singularities predicted as a
consequence of the occurrence of trapped surfaces. On the level of astrophysical
applications, systems involving black holes are possible engines for highly energetic
phenomena like AGNs or gamma-ray bursts, and also likely a comparatively strong
source of gravitational radiation.
The most simple model of black hole formation from, say, cold neutron stars,
is a fluid in spherically symmetric polytropic equilibrium moving on a sequence of
increasing mass due to accretion. 1 This assumes that (i) the stellar structure and
dynamics are represented reasonably by the ideal fluid equation of state and the
polytropic stratification, (ii) accretion processes are slow compared to the dynamical
timescales of the star, and (iii) rotation is negligible. Our focus has been to study
the effects of relaxing the third assumption.
In spherical symmetry, the sequence of equilibrium poly tropes has a maximum
in the mass function M(pc), where Pc denotes the central rest-mass density of the
poly trope. This maximum is connected to a change in the stability of the funda-

*Department of Physics, Aristotle University of Thessaloniki, Thessaloniki 54124, Greece

tSchool of Mathematics, University of Southampton, Southampton S017 IBJ, UK
+Max-Planck-Institut fur Gravitationsphysik, Albert-Einstein-Institut, 14476 Golm, Germany
§Center for Computation and Technology, Louisiana State University, Baton Rouge, LA 70803,
U.S.A.
'Max-Planck-Institut fur Astrophysik, Karl-Schwarzschild-Str. 1, 85741 Garching bei Munchen,
Germany

1576
 1577

mental mode of oscillation, 1 and thus collapse sets in via a dynamical instability to
radial deformations. During the subsequent evolution, a trapped tube forms at the
center which traverses the stellar material entirely. 2
How much of this behaviour is preserved when rotation is taken into account?
Rotation is known to change the equilibrium structure of the star, and, in con-
sequence, its modes of oscillation and set of unstable perturbations. The collapse
might also lead to the formation of a massive disk around the new-born black hole,
and finally only systems without spherical symmetry can be a source of gravitational
radiation.
Numerical simulations have been used to study the collapse and black hole for-
mation of general relativistic rotating polytropic stars.3 For the uniformly and mod-
erately differentia.!ly rotating models investigated in those studies, the dynamical
process is described by the instability of a quasi-radial mode and subsequent col-
lapse of the star up to the formation of an accreting Kerr black hole at the star's
center.
Will strong differential rotation modify this picture? Even before our study,
there was evidence that this should be the case. (i) Strong differential rotation can
deform the high-density regions of a star into a toroidal shape, thus changing the
equilibrium structure considerably. (ii) It admits stars of high normalized rotational
energy T/IWI 1 which are stable to axisymmetric perturbations. (iii) It admits non-
axisymmetric instabilities, for example by the occurrence of corotation points,4 at
low values of T/IWI.5 (iv) A bar-mode instability of the type found in Maclaurin
spheroids 6 would likely express itself by the formation of two orbiting fragments if
the initial high-density region has toroidal shape.
This last property has motivated us to ask this question: Can a bar deformation
transform a strongly differentially rotating star into a binary black hole merger with
a massive accretion disk? If so, this process might occur in supermassive stars if
the timescales associated with angular momentum transport are too large to enforce
uniform rotation.
We have investigated black hole formation in strongly differentially rotating,
quasi-toroidal models of supermassive stars,7.8 and found that a non-axisymmetric
instability can lead to the off-center formation of a trapped surface (see figure).
An extensive parameter space study of this fragmentation instability8 reveals that
many quasi-toroidal stars of this kind are dynamically unstable in this manner, even
for low values of T / IWI, and we have found evidence that the corotation mechanism
observed by Watts et a1. 4 might be active in these models. Since, on a sequence of
increasing T/lvVI, one of the low order m = 1 modes becomes dynamically unstable
before m = 2 and higher order modes, one would not expect a binary black hole
system to form in many situations (although this may depend on the rotation law
and details of the pre-collapse evolution as well). Rather, the off-center production
of a single black hole with a massive accretion disk appears more likely.
Since the normalized angular momentum J / M2 of the initial model is greater
than unity, there is another interesting consequence of this formation process: the
1578

Fig. 1. Development of the fragmentation instability in a model of a strongly differentially rotat-

ing supermassive star. The darker shades of grey indicate higher density. The closed white line in
the last plot is a trapped surface.

resulting black hole, unless it is ejected from its shell, may very well be rapidly
rotating, spun up by accretion of the material remaining outside the initial location
of the trapped surface. Investigating the late time behaviour of this accretion pro-
cess, estimating possible kick velocities of the resulting black hole, and finding the
mass of the final accretion disk is, however, beyond our present-day capabilities and
subject of future study.

References
1. S. Shapiro and S. Teukolsky, Black Holes, White Dwarfs and Neutr-on Stars (Wiley
1983).
2. S. Shapiro and S. Teukolsky, Astr-ophys. J. 235, 199 (1980).
3. M. Shibata, T. Baumgarte and S. Shapiro, Phys. Rev. D 61, 044012 (2000). L. Baiotti,
I. Hawke, P. Montero, F. Lomer, L. Rezzolla, N. Stergioulas, .J. A. Font and E. Seidel,
Phys. Rev. D 71, 024035 (2005), and references therein.
4. A. Watts, N. Andersson and D. Jones, kstr-ophys. J. L37 (2005).
5. J. Centrella, K. New, L. Lowe and J. Brown, Astr-ophys. J. 550 (2001).
6. S. Chandrasekhar, Ellipsoidal Figures of Equilibrium (Yale UP 1969).
7. B. Zink, N. Stergioulas, I. Hawke, C. D. Ott, E. Schnetter and E. Milller, Phys. Rev.
Letter-s 96, 161101 (2006).
8. B. Zink, N. Stergioulas, I. Hawke, C. D. Ott, E. Schnetter and E. Milller, 8.'3tro-
ph/0611601 (2006).
 ADAPTIVE MESH REFINEMENT AND RELATIVISTIC MHD

DAVID NEILSEN, ERIC W. HIRSCHMANN

Department of Physics and Astronomy, Brigham Young University,
Provo, UT 84602, USA

MATTHEW ANDERSON
Department of Physics and Astronomy, Louisiana State University,
Baton Rouge, LA 70803-4001, USA

STEVEN L. LIEBLING
Department of Physics, Long Island University-C. W. Post Campus,
Brookville, NY 11548, USA

We solve the general relativistic magnetohydrodynamics equations using distributed

parallel adaptive mesh refinement. We discuss strong scaling tests of the code, and present
evolutions of Michel accretion and a TOV star.

1. Introduction
Compact objects combine a wide array of fascinating physics, and gravitational
waves may open new ways to probe these objects. We are interested in systems
where gravitational and magnetic fields are dynamically important. One challenge
in simulating astrophysical compact objects is that these systems require a range of
important length and time scales. Adaptive mesh refinement (AMR) thus becomes
an increasingly important tool for large scale computations. Furthermore, large
computational problems on today's computers must be able to effectively utilize a
large number of distributed processors.
To address some of the challenges in studying compact objects with numeri-
cal relativity, we have developed a code to solve the general relativistic magneto-
hydrodynamics (GRMHD) equations with AMR.. We use the HAD infrastructure,
a modular code for solving hyperbolic and elliptic differential equations with dis-
tributed parallel AMR.. HAD uses Berger-Oliger style AMR with sub-cycling in time.
Refinement criteria may be problem specific, or a shadow hierarchy allows one to
easily estimate the truncation error dynamically for use in specifying refinement
criteria. The equations to be solved for a specific problem are isolated in equation
modules, which may be used independently or combined with other modules. For
example, the MHD and GR equations are in separate modules, which may be used
independently or combined for the GRMHD code.
The MHD equations are solved using the Convex Essentially Non-Oscillatory
(CENO) method, a third order scheme for smooth fluid flow. Although our AMR
driver can accommodate both finite difference and finite volume discretization meth-
ods, we choose a finite difference high-resolution shock-capturing method for the
fluid equations to simplify the combined GRMHD code. We use hyperbolic diver-
gence cleaning to control the \7 . B = 0 constraint for the magnetic field. Communi-

1579
1580

cation between coarse and fine grids uses WENO interpolation, a scheme designed
for discontinuous functions. Finally, the method of lines is used for the temporal
discretization, and we use a TVD-preserving, third-order Runge-Kutta scheme to
integrate the equations. In this paper, we briefly summarize some of our results.
The details of our method and more extensive tests are presented elsewhere. 1 ,2

2. Results
Astrophysical simulations of compact objects require that a large number of proces-
sors can be used efficiently. A rigorous measure of such performance is the strong
scaling test, where a model problem of fixed size is run on increasing numbers of
processors. These tests indicate that the GRMHD code uses the distributed paral-
lel computing environment relatively efficiently, as our code scales approximately
linearly as the number of processors is increased by a factor of ten. 2 (See Figure 1.)
Note, since the problem size is fixed, the scaling can not be linear indefinitely.
To verify that the equations are implemented correctly, we have compared results
with exact solutions, and we discuss two of these tests here: the Michel solution,
and Tolman-Oppenheimer-Volkoff (TOV) solutions. The Michel solution describes
the continuous spherical accretion of a fluid onto a Schwarzschild black hole in the
presence of a radial magnetic field. We use in-going Eddington-Finkelstein coordi-
nates in our calculation, and excise a centrally located cubical region of half width
0.3 M to remove the singularity. In this test, the fluid is initially set to the Michel
solution for radius r > 2.5lvI while for r :s: 2.5M a constant pressure and density are
chosen. The system is then evolved until a steady state is reached. The refinement
criterion is based on the estimation of the truncation error provided by the shadow
hierarchy. Figure 1 shows the AMR grid structures at time t = 50M.
A stable TOV solution is used for our second test, which we evolve in the Cowling
approximation (fixed geometry) for over 400 light-crossing times. The star oscillates
as expected, and the oscillations of the density at half the stellar radius (R/2) are
shown in Figure 2. The equation of state for the initial data is P = "'P6, with
r = 5/3 and", = 4.349. Similar runs on dynamic backgrounds have also been
performed and show similar results, though for slightly shorter periods of time.

Acknowledgments
We are pleased to thank Luis Lehner, Patrick Motl, and Ignacio Olabarrieta for
numerous suggestions and discussions during the course of this work. We also
thank Bruno Giacomazzo, Carlos Palenzuela, Oscar Reula, Luciano Rezzolla, and
Joel Tohline for helpful discussions. This work was supported by the NSF through
the grants PHY-0326311, PHY-0244699, PHY-0326378, PHY-0502218, and PHY-
0325224.
 1581

Mesh Refinement Scaling

10~'---'-------r---m

8 16 32 64 80
Processors

Fig. 1. The left frame shows a strong scaling test of our MHD code using AMR. In this test,
thirty iterations were performed on a coarse grid of size 81 3 and a single level of refinement. The
right frame shows the rest density po and the AMR grid structures for the Michel solution at
t = 50M in the x-y plane. The domain of simulation is {x,y,z} E [-15M,15MJ. The cubical
excision region is highlighted in the center of the grid on the left.

200 400 600 800

Fig. 2. The variation in the density at Rj2 for a stable TOV solution. The evolution is performed
using 129 3 points on a cubical domain {x,y,z} E [-l1AI, 11M]. The central density is 8.1 x 10- 4 ,
the stellar radius R 9.279, and mass M = 0.5659.

References
1. D. Neilsen, E. W. Hirschmann and R. 8. Millward, Class. Quantum Gmv. 23, 8505
(2006).
2. M. Anderson, E. Hirschmann, 8. L. Liebling and D. Neilsen, Class. Quantum Gmv. 23,
6503 (2006).
3-D GRMHD and GRPIC SIMULATIONS OF DISK-JET COUPLING
AND EMISSION

K-I. NISHIKAWA and Y. MIZUNO

National Space Science and Technology Center, 320 Sparkman Drive, Huntsville, AL 35805 USA
K en- I chi. [email protected]

M. WATSON
Fisk University, Department of Physics, 1000 17th Ave North, Nashville, TN 37208 USA
[email protected]

P. HARDEE
The University of Alabama, Tuscaloosa, AL 35487, USA

S. FUERST
KIPAC, Stanford Linear Accelerator Center, 2575 Sand Hill Road, MS 29, Menlo Park, CA
94025 USA

K WU
MSSL, University College London, Holmbury St. Mary, Surrey RH5 6NT, UK

G. J. FISHMAN
NASA/MSFC 320 Sparkman Drive, VP62 Huntsville, AL 35805 USA

We investigate jet formation in black-hole systems using 3-D General Relativistic

Particle-In-Cell (GRPIC) and 3-D GRMHD simulations. GRPIC simulations, which al-
low charge separations in a collisionless plasma, do not need to invoke the frozen condition
as in GRMHD simulations. 3-D GRPIC simulations show that jets are launched from
Kerr black holes as in 3-D GRMHD simulations, but jet formation in the two cases may
not be identical. Comparative study of black hole systems with GRPIC and GRMHD
simulations with the inclusion of radiate transfer will further clarify the mechanisms that
drive the evolution of disk-jet systems.

1. GRPIC numerical simulations and initial results

So far the black hole system has been investigated only by GRMHD simulations,
which ignore various important kinetic effects, in particular, charge separation due
to different motions of electrons and positrons/ions in a magnetic field. 1- 4 We in-
vestigated magnetic coupling in disk-jet system with proper treatment of kinetic
physics, using our general relativistic particle-in-cell (GRPIC) code (for details, see
Watson et al. 2006 5 ). Our objective is to identify the essential micro-physics for jet
formation.
The underlying physics of the particle motion is the contravariant form of the
Newton-Lorentz equation. This form provides the equation for the acceleration of
the particle. The acceleration is a function of the spacetime curvature defined by
the metric and the Lorentz force due to the electromagnetic field. The local field

1582
 1583

is described by the Maxwell field tensor. The components of the tensor are cal-
culated using the contravariant general relativistic form of Maxwell's equations.
Using these three equations, the particles are moved and the fields and currents
are calculated self-consistently. The individual kinematics of the system governs
the evolution of the simulation. The algorithm uses the relativistic PIC method. 6
The particle motion is calculated by integrating the equation of motion using a
fourth order Runge-Kutta. Similarly to the RPIC, the electric and magnetic field
components of Maxwell's field tensor are offset in space and time.
We present results of our 3-D simulations of jet formation using general relativis-
tic plasma particle dynamics in Kerr metric (angular momentum a = J/ J max = 0.9).
The initial setting of the simulation is as follow: a background plasma (8 electron-
positron pairs/cell), a free falling corona, and a Keplerian disk as that in GRMHD
Simulations. 1- 4 The black hole is located at the origin. The particle number of the
corona is 1/100 of the disk. The Keplerian disk is located at T > TD == 3Ts 1cos 81 < 0,
where 0 = 1/8, TD is the disk radius and TS == 2GM/c2 is the Schwarzschild radius.
In this region the particle number is 100 times that of the corona. The orbital veloc-
ity of the particles in the disk is v¢ = VK == cJTs/(2T), where VK is the Keplerian
velocity, T = V -a 2 /2 + R2/2 + ~ J(a 2 - R2)2 + 4a 2 z 2 , and R2 = x 2 + y2 + Z2.7
There are no disk particles initially at T < TD. The initial magnetic field is taken
to be uniform in the z direction. The magnitude of the field is 10 4 gauss. This field
component is the contravariant z component of the field.
Figure 1 shows GRPIC simulations of co-rotating jets launched from the disk
along the y - z plane at x = O. The angular momentum of particles shearing towards
the black hole is transferred to the jet, causing it to spiral. The particles are ejected
away from the black hole, which leads to formation of bipolar jets. Figure 1 also
shows a 3-D view of particle velocities. The most remarkable phenomenon is that the
jet is actually composed of streams of particles with different velocities. Such streams
are kinematically unstable, easily giving rise to non-linear and collective behaviors,
which leads to bunching, clumping and shock formation. Also noticeable is the
development of spiral structure (color). We note that the jet structure depends upon
the plasma particle species assumed in the simulation. At this time there is a need
to quantify the parameters required for jet formation. Our study of particle ejection
has demonstrated that jet material consists of particle pairs, i.e. not restricted to
baryons as assumed in many theoretical models. The particles in the jet maintain
some angular momentum and continue to spiral around the central axis.
The simulations do not show significant magnetic field deformation. This may
be a consequence of the number and the strength of particles prescribed in the
simulation. More particles may cause a stronger disk magnetic field and hence a
stronger electric field, due to the particles falling towards the horizon. Another
possible explanation for the lack of significant field deformation is that the frozen-in
condition is relaxed. In GRMHD simulations, as the frozen-in condition E = -vx B
is used, the magnetic fields are dragged by the fluid motion. The relative significance
1584

,-
I

Fig. 1. The 3D views of disk and corona particles are shown at tiTs = 0 and 1134 (TS rs/c).
Particle pairs are moving through the jet at different velocities. The jet has a structure which
forms spirals around the z (central) axis.

of each of these effects is crucial for proper understanding of jet formation and needs
to be quantified.
The inclusion of radiation transfer (Fuerst et a12006 8 ) is essential to make ob-
servational predictions and hence to verify the findings in the GRPIC and GRMHD
simulations.

Acknowledgments
K. N. acknowledges partial support by National Science Foundation award
AST-0506719 and the National Aeronautic and Space Administration awards
NNG05GK73G and HST-AR-I0966.01-A to the University of Alabama in
Huntsville. M. W. acknowledges partial support by a NASA Faculty Fellowship
Program at National Space and Technology Center.

References
1. K-I. Nishikawa, C. Richardson, S. Koide, K Shibata, T. Kudoh, P. Hardee, and C. J.
Fishman, ApJ 625, 60 (2005)
2. Y. Mizuno, K-I. Nishikawa, S. Koide, P. Hardee, and C. J. Fishman, ApJS, submitted,
(2006a) (astro-ph/0609004)
3. Y. Mizuno, K-I. Nishikawa, S. Koide, P. Hardee, and C. J. Fishman, ApJL, submitted,
(2006b) (astro-ph/0609344)
4. Y. Mizuno, K-I. Nishikawa, S. Koide, P. Hardee, and C. J. Fishman, in this proceeding,
(2006c)
5. M. Watson, K-I. Nishikawa, and Y. Mizuno, ApJ, in preparation, (2006)
6. K.-I. Nishikawa, P. Hardee, C. B. Hededal, and C. J. Fishman, ApJ 1267 (2006)
7. S. C. Bell, Compt. in Phys. 9 281 (1995)
8. S. Fuerst, Y. Mizuno, K.-l. Nishikawa, and K, Wu, ApJL, ill preparation (2006)
 RESISTIVE GENERAL RELATIVISTIC MHD SIMULATIONS OF
JET FORMATION AROUND KERR BLACK HOLE

SHINJI KOIDE
Faculty of Science, Kumamoto University, 2-39-1, Kurokami, Kumamoto 860-8555, Japan
[email protected]]J

KAZUNARI SHIBATA
Kwasan and Hida Observatory, Kyoto University, Yamashina, Kyoto 607-8417, Japan
[email protected]

TAKAHIRO KUDOH
Division of Theoretical Astronomy, National Astronomical Observatory of Japan,
Mitaka, Tokyo 181-8588, Japan
[email protected]

Recent general relativistic magnetohydrodynamic (GRMHD) simulations of jet forma-

tion show inevitable formation of anti-parallel magnetic field in black hole magneto-
spheres. In such situation, magnetic reconnect ions should take place and influence the
jet acceleration drastically. However, all of present GRMHD simulations assumed elec-
tric resistivity to be zero, where the magnetic reconnection is forbidden. To investigate
the magnetic reconnect ion near the black hole, we develop the numerical method of the
resistive GRMHD.

1. Introduction

Recently, general relativistic magnetohydrodynamic (GRMHD) simulations have

become popular.l~ll They show drastic, complex phenomena in black hole magne-
tospheres. Especially, simulations of magnetic bridges (magnetic fluxes) between an
ergosphere and a disk around a very rapidly rotating black hole show that the mag-
netic bridges never be stationary. That is, if the magnetic bridges are not swallowed
by the black hole, they expand explosively and form a jet. ll Furthermore, they
show that inside the jet, anti-parallel magnetic field is formed generally. The non-
relativistic MHD simulations l2 ,13 indicate that magnetic reconnections are caused
by the strong current sheets of the anti-parallel magnetic field and influence the jet
acceleration drastically. Only with the magnetic reconnection, magnetic islands are
formed in the jet. Sometimes, the present GRMHD simulations show the forma-
tion of many magnetic islands. 8 ,IO,1l However, all of these simulations assume the
electric resistivity to be zero (ideal GRMHD) and thus the magnetic reconnect ion
should not happen; the formation of the magnetic islands is artificial. The numerical
magnetic reconnection may influence the bulk motion of the plasma unphysically.
To investigate the magnetic reconnection around the black hole properly, we de-
velop the numerical method of the resistive GRMHD (7]GRMHD). In this report,
we show a brief explanation of the 7]GRMHD.

1585
1586

2. Method of 1JGRMHD
The difference of the basic equations between 1]GRMHD and ideal GRMHD is only
the resistive term in Ohm law; Ohm law with resistivity 1]
1 1]
E +v x B - - 2 (v . E) = - (J - Pe v) (1)
c I
is used for 1]GRMHD, while for the ideal GRMHD the right hand side (the resistive
term) is assumed to be zero. Here v, B, E, J are the velocity, magnetic flux density,
electric field strength, and current density observed by Zero-Angular-Momentum-
Observer (ZAMO) frame, and c, I == (1 - v2/c2)~1/2, and Pe are the light speed,
Lorentz factor, and charge density, respectively.
According to Watanabe & Yokoyama (2006),14 we considered the time boosting
of the electric field E and the charge density Pe with the equations of Ampere-
Maxwell law and charge conservation law to evaluate the displacement current and
the charge separation. When we solve the time evolutionary equations explicitly,
the scheme is stable only on the condition of the time interval tlt,
21]
tlt < -2-' (2)
c a
where a is the lapse function. This means that in the limit of ideal GRMHD, the
time interval tlt becomes zero to stabilize the numerical instability. Furthermore, the
explicit scheme does not converge into the numerical scheme of the ideal GRMHD
when we take the limit of ideal GRMHD. To avoid this numerical instability and the
discontinuity to the ideal GRMHD, we use the implicit method of the equations. To
get the physical variables at the next step, we have to solve the simultaneous alge-
braic equations. The algebraic equations are the extension of those of the primitive
variable recovery for the ideal GRMHD. These algebraic equations are difficult to
solve, but the numerical scheme is absolutely numerically stable. Furthermore, in
the limit of the ideal GRMHD, the algebraic equations become the exact expression
of the primitive variable recovery in the ideal GRMHD simulations.

3. Test calculations and discussion

Here we report a simple test calculation of 1]GRMHD to show the validity of the
implicit scheme. Figure 1 shows the results of the ideal GRMHD and 1]GRMHD
calculations of magnetic bridges between an ergosphere and a disk around a rapidly
rotating black hole (rotation parameter a = 0.99995). The initial condition and
numerical setting are given by those of "current loop + quasi-equilibrium plasma
case" of Koide, Kudoh, & Shibata (2006).11 These results show the rapid expansion
of the magnetic bridges between the ergosphere and the disk around the rapidly
rotating black hole. Panel (a) shows the ideal GRMHD case, which is essentially
identical to Fig. 10 (b) in Koide et al. (2006) 11 while the time is different. Panel
(b) shows the 1]GRMHD case with the resistivity 1] = 1O~4rsc using the implicitly
method (rs is the Schwartzschild radius). The time interval tlt is tlt = O.01tltcFL'
 1587

where ~tCFL is the critical time interval of the CFL condition. When we used
the explicit method of T/GRMHD with the same time step, the calculation stopped
due to the numerical instability. The resistivity is so small ('I] 1O-4 rsc ), thus
the influence of the resistivity to the plasma dynamics is negligibly small. The
results of ideal and resistive GRMHDs should be identical because there is no strong
current sheet. The slight difference between the results comes from the improper
approximation of the simplified implicit scheme. The complete implicit treatment
for the 'l]GRMHD is demanded.
After the long-term simulation of the T/GRMHD, the anti-parallel magnetic field
will form and the magnetic reconnect ions will happen. Then the drastic difference
between the results of ideal and resistive GRMHD should be shown.

Acknowledgments
One of authors (S.K.) thanks Mika Koide for her useful comments on this
manuscript. This work was supported in part by the Scientific Research Fund of
the Japanese Ministry of Education, Culture, Sports, Science, and Technology.

(a) :'0 (b) 50

4.0 4.0

:3.0 :3.0

N
20 20

1.0 1(1

III 2ll :JO 1.0 5.0

E/r s

Fig. 1. Comparison of the results of (a) ideal GRMHD calculation and (b) 'l)GRMHD calculation
with the resistivity 'I) 1O-4 r·sc (by simplified implicit method) at t = 15Ts, where TS
The black region at the origin indicates the black hole horizon. The dotted line very near the
horizon is the inner boundary of the calculation region. The broken line indicates the surface of
J
the ergosphere. The gray-scale shows the azimuthal component of the magnetic field B,p/ pc 2 (p
is the density). The white solid lines are the magnetic surfaces and the arrows show the poloidal
velocity of the plasma.

References
1. S. Koide, K. Shibata, and T. Kudoh, Astrophys. J. 495, 1063 (1998).
2. S. Koide, K. Shibata, and T. Kudoh, Astrophys. J. 522, 727 (1999).
3. S. Koide, D. L. Meier, K. Shibata, and T. Kudoh, Astrophys. J. 536, 668 (2000).
4. S. Koide, K. Shibata, T. Kudoh, and D. L. Meier, Science 295, 1688 (2002).
5. S. Koide, Physical Review D 67, 104010 (2003).
1588

6. S. Koide, Astraphys. 1. Lett. 606, L45 (2004).

7. S. Hirose, J. H. Krolik, J. F. Hawley, and J.-P. De Villiers, Ast7'ophys. 1. 606, 1083
(2004).
8. J. C. McKinney and C. F. Gammie, Astraphys. 1. 611, 977 (2004).
9. S. Komissarov, Mon. Not. Royal Astra. Soc. 359, 801 (2005).
10. J. C. McKinney, astro-phj0603045.
11. S. Koide, T. Kudoh, and K. Shibata, Physical Review D 74,044005 (2006).
12. M. R. Hayashi, K. Shibata, and R. Matsumoto, Astraphys. 1. Lett. 468, L37 (1996).
13. Y. Kato, S. Mineshige, and K. Shibata, Ast7'ophys. 1. 605, 307 (2004).
14. N. Watanabe and T. Yokoyama, Astraphys. 1. 647, L123 (2006).
 MAKING UP A SHORT GRB: THE BRIGHT FATE OF MERGERS
OF COMPACT OBJECTS*

MIGUEL-ANGEL ALOyt and P. MIMICA+

Departamento de Astronomia y Astrofisica, Universidad de Valencia
46100, Burjassot, Spain
Miguel. A .Aloy@uv. es

We show some of the most important reasons why the likely fate of the merger of a
neutron star with another compact object may be to yield a short gamma-ray burst
(sGRB). Emphasis is made on some robust results that general relativistic (mag-
neto)hydrodynamic simulations have established regarding the aforementioned subject.

1. What do we know about the progenitor systems of sGRBs?

Our knowledge about the progenitors of short gamma-ray bursts is relatively small.
Any proposed progenitor system should be able of releasing rv 10 49 erg in the form of
thermal energy or Poynting flux. Furthermore, the progenitor is requested to yield
outflows collimated into cones of half-opening angle OJ rv 4° - 25°, as inferred from
the observed breaks in the light curves of some short GRBs, 6 or the lower limits on
OJ deduced from the absence of such breaks. 7 ,lo Additionally, flow variability down
to a few milliseconds over timescales rv 0.1- 2 s has to be produced. Also, any viable
progenitor has to satisfy the fact that it can be generated in any of the typical hosts
observed for a handful of burst,16 as well as to occur at rates equal or larger than
the rate at which sGRBs are produced.

2. Why are mergers of compact objects good candidates to be

progenitors of sGRBs?
A family of systems that has been proposed as likely progenitors of sGRBs is the
remnant left by the merger of a binary system of compact objects formed by either
two neutron stars (NSs) or a NS and a BH.15 Such systems consist of a BH girded
by a thick gas torus from which it swallows matter at a hypercritical rate. In this
situation the cooling is dominated by the emission of neutrinos. These neutrinos
might either be the primary energy source blowing a fireball of e+e- pairs and
photons or to load with pairs a Poynting dominated outflow. 13 The duration of the
produced outflows is, in part, limited by the time needed by the BH to engulf most
of the matter of the accretion disk, namely, a few lOOms (but see below). Thereby,
a merger of compact objects (MeO) may release energy during a sufficiently long
time to fuel a short duration sGRB event. This limit on the time scale, set by the ON
time of the source, holds for both v-powered jets and for MHD-generated outflows. 4

* This research has been partially supported by the Spanish Ministerio de Ciencia y Tecnologfa
grant AYA2001-3490-C02-COl and through the EU grant MEIF-CT-2005-009395.
tRam6n y Cajal Fellow of the Spanish Ministry of Education and Science
+Marie Curie Fellow

1589
1590

According to the state-of-the-art numerical simulations of MCOs including real-

istic microphysics 12 it is possible to release a few 10 49 erg above the poles of a stellar
mass BH in a region of nearly vacuum as a result of the process of vD-annihilation
in such region. Detailled simulations of such process suggest that even more en-
ergy could be deposited in the system,5 although time-dependent numerical models
including energy transport are needed to give more reliable numbers. Therefore,
MCOs provide an energy budged which may satisfy the energy requirements to
produce sGRBs. Numerical simulations have also shown the ability of the system
to tap the energy from the BH and potentially fuel a GRB.14
In the standard merger scenario, initially the spiral-in process may take rv 1 Gy.
Since it is likely that the newly born binary system may receive a natal kick (of
a few rv 100kms- 1), it is expected that, when the two compact objects merge,
they have travelled to the outer skirts of the galaxy or, even, to the intergalactic
medium. This evolutionary path is compatible with observing sGRBs outside of
their putative host galaxies (e.g., GRB 050813).1 6 Alternatively, the evolutionary
tracks of MCOs might be much shorter (rv 1My),18 which has the implication that
many mergers could happen inside of their hosts galaxies and, hence, this would
accommodate the small offsets with respect to the galactic center observed for a
few sGRBs (e.g., GRB 050724).16
The number of well identified galaxies hosting sGRBs is still small to have a
good statistical sample. Nevertheless, it seems that sGRBs are associated with both
(young) start-forming and (old) elliptical galaxies. 8,16 It is very appealing (but
still quite speculative) to establish an association between sGRBs detected in old
galaxies with mergers happening after long evolutionary paths and, on the other
side, between sGRBs found in young galaxies with faster evolving mergers. However,
taking together the information provided by the typical galactic offsets and the
typical host galaxies of sGRBs one may infer only circumstantial evidences about
the nature of the progenitors of sGRBs.
Self-consistent (magneto) hydrodynamic modeling is needed to address issues
like:
Collimation. The generated outflows are either collimated by the accretion disk 3 or
self-collimated by the magnetic field. 14 The typical outflow half-opening angles are
rv 3° - 25° (i.e., compatible with observations; see § 1). The baryon-poor outflows
develop a transverse structure. Particularly, the transverse profile of the Lorentz
factor could be roughly fit by a Gaussian function, but more complicated functions
are required to provide accurate fits. 3
Variability. Even injecting energy at constant rates, the produced outflows are
highly variable. The interaction of the newborn fireball with the accretion torus
yields the growth of Kelvin-Helmholtz 3 instabilities. In case of MHD jets the vari-
ability is imprinted by pinch instabilities. 14 All computed axisymmetric models seem
to be either stable or marginally stable. It is not yet numerically verified whether
3D jets emerging from hyper accreting BHs are stable.
Influence of the environment. Mergers in low density environments may fuel
 1591

ultrarelativistic outflows with the potential to produce normal sGRBs, while in

case that the merger occurs in high density media, the observational signature is
not a sGRB.3 The fact that depending on the environmental density an sGRB can
be produced or not has direct implications for the estimates9 of the true rates of
sGRBs and compared with the rates of NS+NS mergers.
Asymptotic Lorentz factor. Saturation values of the bulk Lorentz factor,
r co 7::.- 500 - 1000 are obtained for both thermally3 and magneticallyI4 generated
outflows. For thermally produced outflows, there is a trend to produce much higher
values of roo for sGRBs (r co 7::.- 500 - 1000) than for IGRBs (roo rv 100). This
difference in Lorentz factor might be the reason for the paucity of soft sGRBs. l l
Duration of the events. In addition to the ON time of the source, the other
factor that sets the duration of a GRB event is the radial stretching of the fireball
which results from the differential acceleration of the forward and rear edges of the
fireball. 3,1l A prolonged activity could also be produced by the fall back onto the
BH of a fraction of the matter ejected during the early stages of the merger. 17

Acknowledgments
It is a pleasure to thank H.-Th Janka and E. Muller for encouragement and support.
The author thanks Max-Planck-Institut fur Astrophysik for hospitality. This work
has been partially supported by SFB 375 "Astroparticle Physics" and SFB-TR 7
"Gravitational Wave Astronomy" of the Deutsche Forschungsgemeinschaft.

References
1. M. A. Alloy, et. al., ApJL 531, L119 (2000).
2. M. A. Aloy, J. M. Ibanez, J.-A. Miralles and V. Urpin, A&'A 396, 693 (2002).
3. M. A. Aloy, H.-Th. Janka and E. Miiller, ApJ 436, 273 (2005).
4. M. A. Aloy and M. Obergaulinger, astro-phj0701187
5. R. Birkl, M. A. Aloy, H.-Th. Janka and E. Mueller, astro-ph/0608543.
6. D. N. Burrows, et. al., ApJ, 653, 468 (2006)
7. De Ugarte Postigo, A., et al., ApJL, 648, 83 (2006)
8. D. B. Fox, et. al., Nat. 437, 845 (2005).
9. D. Guetta and T. Piran, ApJ 435, 421 (2005).
10. D. Grupe, et. al., ApJ, 653, 462 (2006)
11. H.-Th. Janka, M. A. Aloy, P. A. Mazzali and E. Pian, ApJ 645, 1305 (2006).
12. H.-Th. Janka et. al., ApJ 527, L39 (1999); M. Ruffert and H.-Th. Janka, ApJ 344,
573 (1999); S. Rosswog and E. Ramirez-Ruiz, MNRAS 336, L 7 (2002); S. Rosswog, E.
Ramirez-Ruiz and M. B. Davies, MNRAS 345, 1077 (2003); R. Oechslin and H.-Th.
Janka, MNRAS 368, 1489 (2006).
13. A. Levinson and D. Eichler, ApJ 418, 386 (1993).
14. J. C. McKinney, MNRAS 368, 1561 (2006)
15. B. Paczynski, ApJL 308, L43 (1986); J. Goodman, ApJL 308, L47 (1986); D. Eichler,
et. al., Nat. 340, 126 (1989); R. Mochkovitch, et. al., Nat. 361, 236 (1993).
16. J. X. Prochaska, et. al., Api, 642, 989 (2006)
17. S. Rosswog, astro-ph/0611440
18. A. V. Tutukov and L. R. Yungelson, MNRAS 260, 675 (1993); K. Belczynski, V.
Kalogera and T. Bulik, ApJ bf 572, 407 (2002).
 SPACETIME MODES OF RAPIDLY ROTATING RELATIVISTIC
STARS *

NIKOLAOS STERGIOULAS
Department of Physics,
Aristotle University of Thessaloniki, Thessaloniki 54124, Greece
[email protected]

KOSTAS D. KOKKOTAS
Department of Physics,
Aristotle University of Thessaloniki, Thessaloniki 54124, Greece
[email protected]

IAN HAWKE
School of Mathematics,
University of Southampton, Southampton S017 lEJ, UK
[email protected]

We present w-mode frequencies of rapidly rotating relativistic stars, obtained through

nonlinear evolutions of the full Einstein equations.

The dynamics of compact stars is closely related to the emission of gravita-

tional waves. For example, a prime source of gravitational waves is the coalescence
of two neutron stars, while the formation and subsequent dynamical evolution of
proto-neutron stars, which involves oscillations and/or rotational instabilities, is
also a phase during which strong gravitational waves signals are emitted 1 . The os-
cillations and instabilities of newly born compact stars have typically been studied
using perturbation theory, i.e. using the linearized version of the Einstein equations.
Through this analysis many oscillation modes have been identified, most of them
being related to oscillations of the fluid (j-, p-, g-modes, etc.). However, during
the past decade the study of oscillations modes characterizing the perturbations
of spacetime itself (spacetime modes) has also become possible. This showed the
existence of three families of spacetime modes, the so called w-modes 2 , the WII-
modes or scattering modes 3 and (for extremely compact stars) the trapped modes 4 ,5
. Some of these families of modes could in principle become unstable due to the
Chandrasekhar-Friedman-Schutz (CFS) instability 6,7 if the star is rotating very
rapidly and is very compact 8 .
So far, the spectrum of spacetime oscillations is well studied and understood
for nonrotating or slowly rotating stars, but for the most interesting case, which is
the case of rapid rotation, theoretical and numerical difficulties have prevented the
computation of pulsation modes using perturbative techniques (for zero-frequency
modes or when the Cowling approximation is used, see 9 for a review). Recently, a

*This research has been partially supported by the European Network of Theoretical Astroparticle
Physics (ENTApP) as part of the ILIAS Network (RII3-CT-2004-506222).

1592
 1593

160
--_. empirical (full)
G--EJ ICAN==I,O
140 G"El ICA(AKS96)
'-.:..-.::-' leA N=l.O (rotating)

120

~ 100

~ 80 ""o":n..

'~."': ...
60

40

0.1 0.15 0.2 0.25

M/R

Fig. 1. Scaling relation for frequencies of w-modes in a sequence of uniformly rotating relativistic
poly tropes (solid line).

different approach in studying pulsations of rapidly rotating compact stars (relying

on the time-evolution of the fully nonlinear equations) has been employed in obtain-
ing frequencies of axisymmetric fluid pulsations in the Cowling approximation10,1l
as well as for dynamical spacetimes. 12 ,13 What is still missing is the oscillation spec-
trum of the spacetime modes. Ideally, both the fluid and spacetime must be evolved
in order to obtain an accurate description of the spacetime modes of compact stars.
However, the coupling of these modes to the fluid is very weak, so that in practice
one could neglect the evolution of the fluid and only consider the dynamics of the
spacetime. This approximation, called the Inverse Cowling Approximation (ICA)14
has been shown to yield quite accurate spacetime mode frequencies for nonrotating
stars. In the slow rotation approximation the spectrum of w-modes has been studied
only recently, in the ICA approximation 15 . Here, we extend this approach to rapid
rotation, using fully nonlinear dynamical evolutions.
For our simulations, we use a 3-dimensional, nonlinear, general-relativistic evo-
lution code that was presented in detail in 18-20 . To summarize, the code has been
constructed within the framework of the Cactus Computational Toolkit 21 , which
provides the basic grid structure, parallelization and spacetime evolution routines.
The spacetime is evolved in the BSSN formulation (see 22,23) of the Einstein equa-
tions. The spacetime differencing is fourth order accurate in both space and time,
while the method of lines is used. We use mesh refinement, provided by the Carpet
module 24 . The initial data are constructed using stationary equilibrium configura-
tions of rotating relativistic polytropes9 ,17 . The initial data are then perturbed by
adding a small quadrupolar disturbance in the density (of the order of 1%). This
disturbance appears as a nonstationary source term in the Einstein equations and
triggers the excitation of spacetime modes. Far from the source, we extracted the
emitted gravitational waves by matching the dynamical spacetime to non-spherical
perturbations of a Schwarzschild background, using the gauge-invariant Moncrief
1594

variables 16 (see 25 for a recent review). We have studied both nonrotating and ro-
tating equilibrium models with polytropic index N = 1 and polytropic constant
K = 100 (with c = G = Mo = 1). The rotating models coincide with the uniformly
rotating, constant rest-mass sequence in l l .
Figure 1 shows the scaling relation f R vs. M / R (where f is the frequency of the
mode and M, R are the gravitational mass and the equatorial circumferential radius
of the star) for the nonrotating sequence and the rotating sequence, compared to
the results of 14 and to the expected scaling relation in full GR (without the lCA
approximation)26 . Due to rotational flattening of the star at high rotation rates,
the scaling relation for the rotating sequence is drastically altered, compared to
the nonrotating limit. A complete study of w-mode frequencies for a larger set of
equilibrium sequences, including quasi-radially unstable models, will be presented
elsewhere.

References
1. K. D. Kokkotas and N. Stergioulas, Proceedings of 5th International Workshop "New
Worlds in Astroparticle Physics" World Scientific (2005), gr-qc/0506083
2. K. D. Kokkotas and B. F. Schutz M.N.R.A.S. 255, 119-128 (1992)
3. M. Leins, H.-P Nollert and M. H. Soffel, Phys.Rev. D 48, 3467-3472 (1993)
4. S. Chandrasekhar and V. Ferarri Proc.R.Soc. London A434, 449 (1991).
5. K. D. Kokkotas M.N.R.A.S. 268, 1015-1018 (1994)
6. S. Chandrasekhar Phys. Rev. Lett. 24 611 (1970)
7. J. L. Friedman and B. F. Schutz Astrophys. J. 221 937 (1978)
8. K. D. Kokkotas, J. Ruoff and N. Andersson, Phys. Rev. D 70 043003 (2004)
9. N. Stergioulas, Living Rev. Relativity, 6 3 (2003)
10. J. A. Font, H. Dimmelmeier, A. Gupta and N. Stergioulas M.N.R.A.S. 325 1463
(2001)
11. N. Stergioulas, Th. Apostolatos and J. A. Font M.N.R.A.S. 352 1089 (2004)
12. J. A. Font et a!., Phys.Rev. D 65 084024 (2002)
13. H. Dimmelmeier, N. Stergioulas and J. A. Font M.N.R.A.S. 386, 1609 (2006)
14. N. Andersson, K. D. Kokkotas and B. F. Schutz M.N.R.A.S. 280, 1230-1234 (1996)
15. A. Stavridis and K. D. Kokkotas Int.J.Mod.Phys. D14 543 (2005)
16. V. Moncrief, Annals of Physics 88 323 (1974)
17. N. Stergioulas and J. L. Friedman, Astrophys. J. 444 306 (1995)
18. L. Baiotti, I. Hawke, P. Montero and L. Rezzolla, in Computational Astrophysics
in Italy: Methods and Tools, edited by R. Capuzzo-Dolcetta, Mem. Soc. Astron. It.
Supp!., Trieste, 2003, vo!' 1, p. 327.
19. L. Baiotti, I. Hawke, P. J. Montero, F. Lamer, L. Rezzolla, N. Stergioulas, J. A. Font
and E. Seidel, Phys. Rev. D 71, 024035 (2005).
20. L. Baiotti, I. Hawke, L. Rezzolla and E. Schnetter, Phys. Rev. Lett. 94, 131101 (2005).
21. Cactus Computational Toolkit www. cactus code . org
22. M. Shibata and T. Nakamura, Phys. Rev. D 52, 5428 (1995).
23. T. W. Baumgarte and S. L. Shapiro, Phys. Rev. D 59, 024007 (1999).
24. E. Schnetter, S. H. Hawley and I. Hawke, Class. Quantum Grav. 21, 1465-1488 (2004).
25. A. Nagar and L. Rezzolla, Classical and Quantum Gravity 22, R167-R192 (2005).
26. K. D. Kokkotas, Th. Apostolatos and N. Andersson, M.N.R.A.S. 320 307 (2001)
 Dynamics of
Compact Binaries
This page intentionally left blank
 REDUCING ORBITAL ECCENTRICITY IN BINARY BLACK
HOLE SIMULATIONS
HARALD P. PFEIFFERl, DUNCAN A. BROWN 1 ,2, LAWRENCE E. KIDDER3,
LEE LINDBLOM 1 , GEOFFREY LOVELACE 1 and MARK A. SCHEEU
1 Theoretical Astrophysics, California Institute of Technology, Pasadena, California 91125
2 LIGO Laboratory, California Institute of Technology, Pasadena, California 91125
3 Center for Radiophysics and Space Research, Cornell University, Ithaca, New York, 14853

Binary black hole simulations starting from quasi-circular (i.e., zero radial velocity)
initial data have orbits with small but non-zero orbital eccentricities. Nonzero radial
velocities are added to the quasi-equilibrium formalism for constructing binary black
hole initial data, and it is demonstrated how the orbital frequency and radial velocities
can be tuned to obtain inspiral trajectories with significantly reduced orbital eccentricity.

1. Introduction
The orbits of inspiraling compact objects such as black holes circularize during the
inspiral due to the emission of gravitational waves,l so that binaries formed from
stellar evolution (rather than dynamical capture) are expected to have very small
eccentricities as they approach merger. For this reason and because the inspiral
timescale is much longer than the orbital timescale, the assumption of a quasi-
circular orbit (i. e., zero radial velocities) has been widely used in the construction
of binary black hole initial data. 2- 6 Inspiraling compact objects must, however, have
a small inward radial velocities. Neglecting these radial velocities in the initial data
leads to eccentricity in the subsequent evolution, as found here and in Refs. 7,8. It
is therefore necessary to devise a formalism for constructing initial data that allows
radial velocities on the individual black holes to be specified.

2. Binary black hole initial data with nonzero radial velocities

Incorporating radial velocities in puncture initial data 9 is straightforward because
the full linear momentum of each hole can be specified. For quasi-equilibrium initial
data4 - 6 this is more difficult, because the assumption of quasi-circularity is fun-
damentally tied to a corotating coordinate system. However, we have shown 8 that
corotating coordinates are not central to the scheme and identical initial data can
be obtained in an asymptotically inertial coordinate system: one merely requires
that the black holes move on circular coordinate trajectories rather than remaining
fixed. This idea can be generalized by requiring that the black holes move on inspi-
ral trajectories. 8 For corotating black holes, this is accomplished by adopting the
following shift boundary condition on each of the black hole excision surfaces:
.. . ri
(3' = as' - (no x r)' + V,..-. (1)
ro
Here a is the lapse, si the outward-pointing spatial unit normal to the excision
surface, and ro is the coordinate distance between the center of the excision surface
and the origin. The term (no x r)i causes the horizons to rotate about the origin

1597
1598

with orbital frequency no; the last term in Eq. (1) encodes the radial velocity Vr
of the black hole with respect to the origin. Besides Eq. (1), the only other change
to the standard quasi-equlibrium equations of Refs. 5,6 appears in the outer shift
boundary condition, which becomes (3i ----+ 0 as r ----+ 00. A non-zero radial velocity
is consistent with the horizon being in quasi-equilibrium, and does not prevent the
addition of any desired spin on the black holes. 8

3. Choice of orbital frequency and radial velocity

To demonstrate the orbital eccentricity of quasi-circular initial data, we construct
such data for corotating black holes (coordinate separation d = 20 of Table IV in
Ref. 5; MADMnO = 0.029792, Vr = 0) and evolve it for about 5 orbits with the
computational methods of Refs. 10-12. The dotted line on the left panel of Fig. 1 is
ds(t)/dt, where s(t) is the proper separation between the apparent horizons along
the coordinate line connecting the centers of the holes. This quantity clearly exhibits
oscillatory behavior due to the orbital eccentricity of the binary.

1.88
", , .,'

....... ><_/
1.84
~-0.02
....-
'--'
Vi
"0
1.8
-0.04 .... QC
-- QC t ---> t+llT
-LE
1.76

o 250 500 750 o 250 500 750

tlMADM tlMADM
Fig. 1. Left: Radial velocity for evolutions of quasi-circular (QC; dotted line) and low-eccentricity
(LE; solid line) initial data. The dashed line results from time-shifting the quasi-circular run by
l:!.T. Right: Difference in orbital phase between LE and time-shifted QC.

In Newtonian gravity, no and Vr influence the orbit in a straightforward way: Vr

is the initial radial velocity, and no is related to the initial radial acceleration, i.e.
increasing no makes d 2s / dt 2 larger. This correspondence also holds approximately
for binary black holes. Given initial guesses for no and V r , we construct initial data
and evolve these data to produce a plot of ds/dt versus time. We then proceed
iteratively, computing the next guesses for no and Vr to reduce the oscillations in
ds / dt. For the results presented here, we adjusted either no or Vr in each iteration,
starting from a simple guess in the first iteration and subsequently using linear
interpolation between the previous iterations. We obtain the final parameters no =
0.029963/MADM and Vr = -0.0015, resulting in the evolution shown by the solid
 1599

line in the left panel of Fig. 1, which has barely discernible oscillations.
The black holes appear to approach each other more quickly and merge ear-
lier in the evolution of the quasi-circular initial data. This corresponds to a simple
coordinate time shift t --t t + 6.T with 6.T = 59MADM to the quasi-circular evolu-
tion, resulting in the dashed curve in Fig. 1. This curve now oscillates around the
low-eccentricity curve, with no discernible secular drift. The right panel of Fig. 1
presents the orbital phase difference b¢ between the low-eccentricity run and the
(time-shifted) quasi-circular run. The offset of b¢ from zero corresponds to the or-
bital phase acquired by the low-eccentricity run up to t = 6.T; b¢ oscillates around
its mean without a noticeable secular drift.
A more comprehensive analysis of these evolutions 8 confirms the trends seen here
using a variety of additional diagnostics, like orbital phase and frequency, and phase
and frequency of gravitational waves. Because no coherent phase difference builds
up, this more comprehensive study also demonstrates very high overlaps between
gravitational waveforms from quasi-circular and low-eccentricity evolutions.

Acknowledgments
We thank Gregory Cook for inspiring discussions. This work was supported in part
by grants from the Sherman Fairchild Foundation and from the Brinson Founda-
tion; by NSF grants PHY-0099568, PHY-0244906, PHY-0601459, DMS-0553302 and
NASA grants NAG5-12834, NNG05GG52G at Caltech; and by NSF grants PHY-
0312072, PHY-0354631, and NASA grant NNG05GG51G at Cornell. Some of the
simulations discussed here were produced with LIGO Laboratory computing facil-
ities. LIGO was constructed by Caltech and MIT with funding from the NSF and
operates under cooperative agreement PHY-0107417.

References
1. P. C. Peters, Phys. Rev. 136, B1224(Nov 1964).
2. E. Gourgoulhon, P. Grandclement and S. Bonazzola, Phys. Rev. D 65,044020 (2002).
3. P. Grandclement, E. Gourgoulhon and S. Bonazzola, Phys. Rev. D 65,044021 (2002).
4. G. B. Cook, Phys. Rev. D 65, 084003 (2002).
5. G. B. Cook and H. P. Pfeiffer, Phys. Rev. D 70, 104016 (2004).
6. M. Caudill, G. B. Cook, J. D. Grigsby and H. P. Pfeiffer, Phys. Rev. D 74, 064011
(2006).
7. A. Buonanno, G. B. Cook and F. Pretorius, gr-qc/0610122 (2006).
8. H. P. Pfeiffer, D. A. Brown, L. E. Kidder, L. Lindblom, G. Lovelace and M. A. Scheel,
gr-qc/0702106 (2007).
9. S. Brandt and B. Briigmann, Phys. Rev. Lett. 78, 3606 (1997).
10. H. P. Pfeiffer, L. E. Kidder, M. A. Scheel and S. A. Teukolsky, Comput. Phys. Com-
mun. 152, 253 (2003).
11. M. A. Scheel, H. P. Pfeiffer, L. Lindblom, L. E. Kidder, O. Rinne and S. A. Teukolsky,
Phys. Rev. D 74, 104006 (2006).
12. L. Lindblom, M. A. Scheel, L. E. Kidder, R. Owen and O. Rinne, Class. Quantum
Grav. 23, S447 (2006).
 NEGATIVE KOMAR MASSES IN REGULAR STATIONARY
SPACETIMES

MARCUS ANSORG
Max-Planck-Institut fur Gravitationsphysik, Albert-Einstein-Institut,
D-144'l6 Golm, Germany
[email protected]

DAVID PETROFF
Theoretisch- Physikalisches Institut, University of J ena, M ax- Wien- Platz 1,
D-O'l'l43 Jena, Germany
D. [email protected]

A highly accurate multi-domain spectral method is used to study axially symmetric and
stationary spacetimes containing a black hole or disc of dust surrounded by a ring of
matter. It is shown that the matter ring can affect the properties of the central object
drastically. In particular, by virtue of the ring's frame dragging, the so-called Komar
mass of the black hole or disc can become negative. A continuous transition from such
discs to such black holes can ue found.

We study self-gravitating systems in equilibrium, consisting of a uniformly ro-

tating, homogeneous perfect fluid ring surrounding a central object which is either a
black hole or a rigidly rotating disc of dust. The corresponding space-time is charac-
terized by two Killing vectors rJ and ~ which describe axisymmetry and stationarity
respectively. For such configurations, Bardeen 1 assigns to each of the two objects
a mass based on the Komar integral. 2 In a similar way, he assigns an angular mo-
memtum to each of the objects. This definition of mass is applicable both to matter
and black holes and can be used for single components of a many-body problem
in stationary spacetimes. We will refer to it here as the Komar mass even when
applied to single objects.
If one deals with the Komar mass, a natural questions concerns its positive
definiteness. We address this question by analyzing the following formulae

(1)

and

(2)
valid for central black hole and central disc configurations respectively.
In Eqn. (1), the central black hole's Komar mass Mh is related to (i) its surface
gravity K" (ii) its horizon area A, (iii) the angular velocity Oh of the horizon and
(iv) the black hole's (Komar) angular momentum J h . For single black holes Eqn.
(1) was given by Smarr 3 but it holds true even in the presence of a surrounding
ring 1 (see also Ref. 4 ) .

1600
 1601

A similar expression can be derived for rigidly rotating discs of dm,t with
and without a surrounding ring of matter (cf. III.15 in Ref. 5 ). In Eqn. (2) the
d
disc's Komar mass Md is given in terms of (i) a constant e vo that is related
to the relative red shift Z8 of photons emitted from the center of the disc and
observed at infinity (Z8 = e- vod - 1), (ii) the baryonic mass Mo of the cen-
tral disc, (iii) its angular velocity Od and (iv) its (Komar) angular momentum
Jd·
The first summands on the right hand sides of formulae (1) and (2) are always
positive. However, either of these terms can become small if we assume the horizon
area and the baryonic mass to be finite and consider the central object to be close
to a degenerate black hole. In particular we find a continuous transition from the
central disc to the central black hole configurations,6 and at the transition point
the central object is a degenerate black hole for which the first terms in Eqns. (1,
2) vanish.
For the discussion of the sign of the second summands, a 'frame dragging'-effect
of the central object caused by the surrounding ring is important. If the torus is
highly relativistic and quickly rotating, it creates a large ergosphere (a portion of
space in which the Killingvector ~ is spacelike). In this case, a counter-rotating
central object (i.e. the sign of its angular momentum is opposite to that of the
torus) inside the ergo sphere is dragged along the direction of the motion of the
ring's fluid elements. As a consequence, the corresponding angular velocity of the
central object can assume the same sign as that of the surrounding ring, and thus
the second summand becomes negative.
Combining the two arguments, it is possible to identify negative Komar masses
by considering central objects close to a degenerate black hole and counter-rotating
with respect to the torus. Note that only highly relativistic and quickly rotating tori
will exert a sufficiently large frame dragging effect to bring this about. Moreover, the
specific rate of counter-rotation must be limited since very strong counter-rotation
would lead to opposite signs of the two angular velocities, Oh/d and Oring, and hence
to a positive second summand.
In Ref.6 we construct sequences of both central black hole and central disc
configurations along which the Komar mass of the central object becomes negative.
In addition, we show that the Komar mass can become negative on either side of
the continuous parametric transition from matter to a black hole.
The results presented reveal clearly that the Komar mass is not an intrinsic
property of a gravitational source but rather a feature of an object within a specific
highly relativistic spacetime geometry. The interesting question about the maxi-
mally attainable ratio

_ Mnegative / MPositive, (3)

where Mnegative and MPositive are the sums of all negative and positive Komar mass
1602

components, respectively, in a given stationary space-time, will be the subject of a

future publication. a

References
1. J. M. Bardeen, in Black Holes, Les astres occlus, eds. C. DeWitt and B.S. DeWitt,
pages 241-289, Gordon and Breach Science Publishers, New York (1973).
2. A. Komar, Phys. Rev. 113, 934 (1959).
3. L. Smarr, Phys. Rev. Lett. 30 71 (1973).
4. B. Carter, in Black Holes, Les astres occlus, eds. C. DeWitt and B.S. DeWitt, pages
57-214, Gordon and Breach Science Publishers, New York (1973).
5. J. M. Bardeen and R. V. Wagoner, Astrophys. J. 167, 359 (1971).
6. M. Ansorg and D. Petroff, Class. Quantum Gmv. 23, L81-L87 (2006).
7. G. W. Gibbons, S. W. Hawking, G. T. Horowitz and M. J. Perry, Commun. Math.
Phys. 88, 295 (1983).

aNote that by virtue of the positive mass theorem (see e. g. Ref. 7 and references therein), the ratio
(3) is always less than 1 for regular physically relevant space-times obeying the dominant energy
condition.
 CONSTRAINT RELAXATION

PEDRO MARRONETTI
Florida Atlantic University,
777 Glades Rd, Boca Raton, FL 33431, USA
[email protected]

Full relativistic simulations in three dimensions are known to develop runaway modes
that grow exponentially and are accompanied by violations of the Hamiltonian and
momentum constraints. We present here a method that controls the violation of these
constraints and is tested with simulations of binary neutron stars in circular orbits. We
show that this technique improves the overall quality of the simulations.

1. Introduction
For most of the past decade, the main theoretical thrust in gravitational research has
been directed toward obtaining stable and accurate numerical models of compact-
object binary systems. One of the most difficult problems to tackle has been the
control of exponentially growing instabilities that degrade the quality of any sim-
ulation and, eventually, terminate it. We propose in Refs. 1,2 an evolution scheme
(Constraint relaxation or CR) where the Hamiltonian and momentum constraints
are only approximately solved at every time step, gently steering the evolution to-
ward the space of solutions of the Einstein Field Equations without completely
forcing their numerical satisfaction. This method utilizes the conformal decompo-
sition of the spatial metric and extrinsic curvature, which has traditionally been
used to solve the initial value problem for binary systems. In this decomposition, a
conformal factor 'ljJ factored out of the spatial metric and a longitudinal addition to
the extrinsic curvature generated from a vector potential Wi are used to satisfy the
Hamiltonian and the three components of the momentum constraint respectively.
CR drives 'ljJ to the solution space of the Hamiltonian constraint by means of a
parabolic equation for the conformal factor. Similarly, the momentum constraint is
controlled by the use of Wi to push the simulation toward the space of solutions of
the momentum constraint. In both cases, a full relaxation of 'ljJ and Wi would lead
to the numerical solution of the constraints. However, the stability of the relaxation
methods relies on gently updating these fields during the evolution. We showed in
Ref. 2 that a full relaxation scheme becomes unstable rather quickly when used in
combination with BSSN.

2. Results
We tested the CR scheme with simulations of an irrotational BNS system. The
details of the initial data set are provided in table 1 of Ref. 1 and the corresponding
convergence tests are presented in Ref. 2 The plots show curves that, for clarity,
have been normalized to their corresponding initial values.
The evolution of the L2 norm of the Hamiltonian constraint residual He and the
three components of the momentum constraint residual Mi across the numerical

1603
1604

100.0,,----r----,-----,----,--.,..----,------,----,

-----
. . . ,/v------ ......... --------- ----- ---
10.0

1.0 .......................................................................... .

0.1

0.0 0.5 1.0 1.5 2.0

tiP

Fig. 1. Evolution of the L2 norm of the Hamiltonian constraint violation.

><
100.0
10.0 ~ "../----,/
---_ ..... /
--
;;E _....., __ __1'- - - - - -
1.0
0.1

'" --
100.0
,.., 10.0
;;E
1.0
0.1
/
100.0
N 10.0 . . __ --J ---,-,;" '" '"
;;E
1.0
0.1
0 0.5 1 1.5 2
tiP

Fig. 2. Evolution of the L2 norm of the momentum constraint violation.

grid are shown in Figs. 1 and 2 respectively. CR achieves an impressive reduction

of the Hamiltonian contraint violation (Fig. 1). Note that the curves are plotted
in logarithmic scale to highlight the more than two orders of magnitude difference
between the CR (dashed) and BSSN (solid) results. CR not only suppresses the
constraint violation modes, but also reduces the violation present in the ID set by
 1605

a factor of about 5. The difference between the CR (solid) and the BSSN (dashed)
results in the case of the momentum constraint (Fig. 2) is not that impressive. At
the end of the simulation, the momentum constraint violation was about four times
smaller than in the BSSN runs. The spikes present in the BSSN curves at t ::::' O.4P
occur on the stellar surface and are related to matter displacement in the grid, a
side-effect of using a frozen shift vector. Note, however, that those spikes disappear
when using CR.

rJ
US I
I
I

1.10
I
....., I

LOS

1.00
......
0.9S
0.0 O.S 1.0 1.S 2.0
tiP
Fig. 3. Evolution of the total angular momentum J.

Figure 3 shows the evolution of the total angular momentum. The plot shows the
CR (solid), BSSN (dashed) curves, as well as the PN estimation (dotted line) of the
angular momentum loss for a point-mass binary with the same mass and angular
momentum as the BNS in consideration (see Appendix B of Ref. 1). The CR based
run agrees with the PN prediction for about 1.5 orbital periods. The inset of figure
3 zooms in on the first half of the period, showing the reduced level of noise in the
CR curve.

Acknowledgments
This work was partially supported by the National Science Foundation under
grant PHY-0555644 and the National Computational Science Alliance under grants
PHY020007N. PHY050010T, and PHY050015N.

References
1. P. Marronetti, Class. Quant. Cmv. 22,2433 (2005). gr-qc/0501043.
2. P. Marronetti, Class. Quant. Cmv. 23,2681 (2006). gr-qc/0511143.
RELATIVISTIC HYDRODYNAMIC SIMULATIONS OF MULTIPLE
ORBITS FOR CLOSE NEUTRON STAR BINARIES*

G. J. MATHEWS
University of Notre Dame, Center for Astrophysics
Notre Dame, IN 46556, USA
E-mail: [email protected]

J. R. HAYWOOD
Indiana University Southeast
New Albany, IN 47150, USA
E-mail: jrhaywoo@ius. edu

J. R. WILSON
University of California,
Lawrence Livermore National Laboratory,
Livermore, CA 94550 USA
E-mail: [email protected]

We discuss numerical general relativistic hydrodynamic simulations for close neutron

star binary systems. The hydrodynamics and field equations are solved at each time slice
with a spatial 3-metric chosen to be conformally flat. By using an adaptive mesh we are
able to reliably numerically integrate many orbits of the binary system. The gravitational
radiation signal is extracted via a multipole expansion to the I = 4 order including both
mass and current moments and a correction for the slow motion approximation. We
compute quasi-equilibrium circular orbit conditions for two equal-mass neutron stars as
a function of total angular momentum from the post-Newtonian regime to near the last
stable circular orbit. We have considered a variety of neutron star equations of state.
From the angular momentum and power loss rate we reconstruct the gravitational wave
form and find that there is a detectable equation of state dependence of the computed
gravity-wave signal.

The detection of gravitational radiation near the last stable orbit of a close
compact binary requires the ability to numerically compute a reliable template for
thousands of orbits between the post-newtonian and strong-field regime. This has
proven to be a challenge for modern numerical codes which have only successfully
evolved a pair of neutron stars for a few orbits at best. 1 ,2 Here, we report on the
development and successful application of a method 3 to compute the evolution of
hundreds of orbits with high stability.
For these simulations we use the usual ADM (3 + 1) split of the 4D spacetime.
We also adopt a metric which is conformally flat. This split and approximation puts
the equations of motion into a set of Poisson-like elliptic equations which are easily
solvable using fast numerical techniques 4 ,5 and are very stable.

*Work at the University of Notre Dame supported by the US Department of Energy under Nu-
clear Theory grant DE-FG02-95ER40934. Work at the Lawrence Livermore National Laboratory
performed in part under the auspices of the U. S. Department of Energy under contract W-7405-
ENG-48 and NSF grant PHY-9401636.

1606
 1607

Another key to the problem of stable orbits, however, is in the treatment of

the hydrodynamics. One important element in our approach is that we utilize an
adaptive mesh which increases the resolution around the stars themselves. We have
found it necessary to have at least 40 zones across each star to accurately compute
the structure. We have also performed careful studies of the optimum Courant
condition to guarantee accuracy of the orbits. Most importantly, however, we adopt
a grid velocity which follows the fluid frame precisely. [Note, that this is not the
same as the Eulerian frame in which the ADM equations are solved.] By minimizing
the fluid motion with respect to the hydro grid we have evolved stable quasi-static
orbits for hundreds of orbits without a hint of instability. We have also tested
our 3D results against exact relativistic simulations6 of the axisymmetric head on
collision of 2 neutron stars. We find the same variation of the star configuration
with separation as noted in those simulations.
In order to evaluate the physical evolution of the quasi-static orbits it is also
necessary to include the power loss from gravitational radiation. Our method of
solving the field equations does not explicitly evolve the gravitational radiation.
Rather, we use a multipole expansion as described in [4,5]. We include terms out to
I = 4 including both the mass and current moments as well as corrections for the
"slow motion" approximation. These formulae apply to strong-field sources as well
as to weak field sources as long as the relevant components of the effective stress
energy tensor can be identified.
As part of this study we have evaluated the equation of state dependence of the
observable properties of the orbits. There is a surprisingly strong equation of state
dependence which may ultimately be utilized to better understand the EOS once
a signal is detected. We use five different equations of state. The first is the zero
temperature, zero neutrino chemical potential equation of state used to model super-
novae 5 and neutron star binaries. 4 We also used two versions of the low-temperature
relativistic EOS developed by Lattimer and Swesty,7 one having K = 220, the other
having K = 375 MeV. The fourth EOS has been developed by Ref. [8]. This EOS
has K = 240 MeV, which is close to the present experimental value. 9 The final EOS
is a polytropic r = 2 EOS. Although this EOS is frequently employed in numer-
ical simulations there are many quantitative differences between results obtained
with this EOS and the realistic versions. In particular, the realistic EOS's are softer
at high density and much more sensitive to general relativistic effects on the self
gravity of the stars.
After selecting an initial configuration for the binary, we have run the hydro-
dynamics calculation with viscous damping for sufficiently long time to relax to a
steady state. This is a crucial first step as the stars will typically oscillate around
equilibrium for the first few orbits. This tends to mask the physics of the binary. As
an illustration we have compared with two recent simulations 1 ,2 who studied close
neutron star binaries in exact BSSN relativity, but only for a few orbits. Our results
for similar initial conditions are nearly indistinguishable from those results. The
configurations of the stars as they evolve, however, are quite different than those
1608

reported by those authors.

We have examined stars at large separations which are in quasi-equilibrium cir-
cular orbits and stable hydrodynamic configurations. These correspond to the last
several minutes of orbit inspiral. By the time 10 orbits are completed the stars have
achieved a very stable configuration which does not change for hundreds of orbits
when the gravitational back reaction is removed from the dynamics. The configu-
rations of the stars, however, are consistently more compact and the gravitational
binding energy greater when the stars achieve true equilibrium.
One of our goals is to find where the strong field orbit dynamics computed
here connects with the post-Newtonian regime. Our method of setting up initial
configurations is designed so as to produce initially corotating stars, i.e. no spin in
the corotating frame. This minimizes initial matter motion on the grid. The system
then evolves to a stable hydrodynamic simulation.
An often utilized parameter characterizing the last stable orbit in the post-
Newtonian regime is the ratio of coordinate separation to total mass (in isolation)
dh/rn. The analogous quantity in our non-perturbative calculation is proper sep-
aration to gravitational mass, d p /rn~. This is not an exact comparison, however.
The separation corresponding to the last stable orbit in the post-Newtonian anal-
ysis does not occur until the stars have approached dh/rn = 6.03. In the results
reported here the last stable orbit occurs somewhere just below d p /rn~ = 7 for
both the polytropic star and the stars with a realistic EOS. As the stars approach
one another, the frequency varies more slowly than PN with separation distance.
Also, the orbit becomes unstable to plunge at a larger separation. At least part of
this difference can be attributed 4 to the effects of the finite size of the stars as well
as the slowing of clocks and stretching of distance in the strong field calculations
presented here. We find that we begin to match the Post-Newtonian point mass
limit at a separation of rv 200 km, corresponding to a gravity wave frequency of
rv 50 Hz. When comparing with gauge invariant quantities, however, the deviation

from PN2.5 results is not as dramatic.

References
1. Duez, M. D. and Marronetti, P. and Shapiro, S. L. and Baumgarte, T. W., Phys. Rev.
D67, 024004 (2003).
2. Miller, M. gr-qe/0510020
3. J. Reese Haywood, PhD Thesis, Univ. Notre Dame (2006).
4. J.R. Wilson, G.J. Mathews, & P. Marronetti, Phys. Rev. D54, 1317 (1996).
5. Wilson, J. R. & Mathews, G. J., 2003, Relativistic Numerical Hydrodynamics, (Cam-
bridge University Press, Cambridge).
6. Wilson, J. R. 2000, PRD, 66, 084015
7. Lattimer, J. M. and Swesty, F. D., Nue!. Phys., A535, 331 (1991).
8. Glendenning, N. K., in Compact stars. Nuclear physics, particle physics, and general
relativity, (Springer; Berlin) 1997.
9. Garg, U., Nue!. Phys. A731, 3, (2004).
 THE FINAL FATE OF BINARY NEUTRON STARS:
WHAT HAPPENS AFTER THE MERGER?

MATTHEW D. DUEZ
Department of Astmnomy, Cornell University, Ithaca, NY 14850 USA
[email protected]. edu

YUK TUNG LID, STUART L. SHAPIRO and BRANSON C. STEPHENS

Department of Physics, University of Illinois at Urbana-Champaign, Urbana, IL 61801-3080
USA

MASARU SHIBATA
Graduate School of Arts and Sciences, University of Tokyo, Komaba, Megum, Tokyo 153-8902,
Japan

The merger of two neutron stars usually produces a remnant with a mass significantly
above the single (nonrotating) neutron star maximum mass. In some cases, the remnant
will be stabilized against collapse by rapid, differential rotation. MHD-driven angular
momentum transport eventually leads to the collapse of the remnant's core, resulting
in a black hole surrounded by a massive accretion torus. Here we present simulations
of this process. The plausibility of generating short duration gamma ray bursts through
this scenario is discussed.

1. Introduction

The merger of binary neutron stars is now one of the favored hypotheses for explain-
ing short gamma ray bursts (GRBs). According to this scenario, after the merger,
a stellar-mass black hole (BH) is formed with an ambient accretion torus with rv 1-
10% of the total mass. Energy extracted from this system by MHD processes or
neutrino radiation powers the GRB fireball. The viability of this model depends on
the presence of a significantly massive accretion disk after the collapse of the rem-
nant following merger, and the presence of a baryon-poor region above the accretion
disk.
A typical binary neutron star system has total mass 2.6-2.81v18 ,1 much larger
than the spherical neutron star maximum mass M TOY . However, the remnant is
also rapidly and differentially rotating. Mass limits for nonrotating stars and for
rigidly rotating stars (the sl1pmrnassive limit, 1v/sup :::::: 1.21vhoy) can be significantly
exceeded when differential rotation is present. 2 Stars with masses greater than lv1sup
are called hyperrnassive stars. Thus, the remnant may form a hypermassive neutron
star (HMNS). General relativistic hydrodynamic simulations have shown that just
after the merger, either a black hole or a HMNS is formed. 3 A BH forms promptly
if the total mass of the system, lvI, is larger than a threshold mass lvlthr :::::: 2.81v18 .
In this case, far less than 1% of the matter remains outside the horizon, which is
unfavorable for GRBs. On the other hand, for !v! < !v!tlm a HMNS forms.
These HMNSs may survive for many orbital periods. However, on longer
timescales magnetic fields will transport angular momentum and may trigger

1609
1610

gravitational collapse. Two important mechanisms which transport angular mo-

mentum are magnetic braking 2 ,4 and the magnetorotational instability (MRI).5
Magnetic breaking transports angular momentum on the Alfven time scale,2,4
TA rv R/VA rv 102(B/1012 G)-l s, where R is the radius of the HMNS. MRloccurs
wherever angular velocity D decreases with cylindrical radius w. This instability
grows exponentially with an e-folding time of TMRI = 4 (8D/81nw)-1,5 indepen-
dent of the field strength. For the HMNS model considered here, TMRI rv 1 ms. The
length scale of the fastest growing unstable MRI modes, AMRI, does depend on the
field strength: AMRI rv 3 cm (D/4000s- 1)-1 (B/10 12 G)« R. When the MRI satu-
rates, turbulence consisting of small-scale eddies often develops, leading to angular
momentum transport on a timescale much longer than TMRI. 5

2. Simulations
To determine the final fate of the HMNS, it is necessary to carry out magneto-
hydrodynamic simulations in full general relativity. Such simulations have only re-
cently become possible. Duez et al. 6 and Shibata and Sekiguch? have developed new
codes to evolve the full Einstein-Maxwell-MHD system of equations self-consistently.
These codes have since been used to simulate the evolution of magnetized hyper-
massive neutron stars,8,9 and implications for short GRBs have been investigated. lD
We assume axisymmetry and equatorial symmetry in all our simulations. We
use uniform computational grids with sizes up to 500x500. To model the remnant
formed in binary merger simulations, we use as our initial data an equilibrium
HMNS, a r = 2 poly trope, with mass M = 1. 71Vhov = 1.5Msup and a rotation
profile chosen so that the ratio of equatorial to central D is rv 1/3. (We find that
an HMNS with a more realistic equation of state evolves similarly.9,lD) We add a
poloidal magnetic field with strength proportional to the gas pressure. The initial
magnetic pressure is set much smaller than the gas pressure, but not so small that
AMRI cannot be resolved. Therefore, we set AMRI ~ R/10, corresponding to B ~ 10 16
G and max( B2 / P) rv 10- 3 .
In our evolutions, the effects of magnetic winding are observed in the generation
of a toroidal B field which grows linearly with time during the early phase of the
evolution, and saturates on the Alfven timescale. The effects of MRI are observed
in an exponential growth of the poloidal field on the AMRI scale, a growth which
saturates after a few rotation periods. The magnetic fields cause angular momentum
to be transported outward, so that the core of the star contracts while the outer
layers expand. After about 66 rotation periods, the core collapses to a black hole.
Using singularity excision,l1 we continue the evolution to a quasi-stationary state.
The final state consists of a black hole of irreducible mass O.9M surrounded by
a hot accreting torus with rest mass O.lM and a collimated magnetic field near
the polar region. At its final accretion rate, the torus should survive rv lOms. The
torus is optically thick to neutrinos, and we estimate that it will emit rv 1050ergs in
neutrinos before being accreted. We also find that the region above the black hole is
 1611

very baryon-poor. All these properties make this system a promising central engine
for a short-hard GRB.
In order to study the evolution of the magnetic field more realistically, it will be
necessary to redo the evolutions in three dimensions. Also, realistic merger remnants
are usually expected to have much smaller magnetic fields than the ones studied
here. In an earlier analysis, we modeled the effects of small-scale MRI turbulence
as a shear viscosity12 and found that, if this model is valid, the evolution of the
HMNS is qualitatively similar to that shown here.

3. Acknowledgments
Numerical computations were performed at the National Center for Supercomputing
Applications at the University of Illinois at Urbana-Champaign (UIUC), and on the
FACOM xVPP5000 machine at the data analysis center ofNAOJ and the NEC SX6
machine in ISAS, JAXA. This work was in part supported by NSF Grants PHY-
0205155 and PHY-0345151, NASA Grants NNG04GK54G and NNG046N90H at
UIUC, and Japanese Monbukagakusho Grants (Nos. 17030004 and 17540232).

References
1. 1. H. Stairs, Science 304, 547 (2004).
2. T. W. Baumgarte, S. L. Shapiro, and M. Shibata, Astrophys. J. Lett. 528, L29 (2000).
3. M. Shibata, K. Taniguchi, and K. Uryu, Phys. Rev. D 71, 084021 (2005).
4. S. L. Shapiro, Astrophys. J. 544, 397 (2000); J. N. Cook, S. L. Shapiro, and B. C.
Stephens, Astrophys. J. 599, 1272 (2003); Y. T. Liu and S. L. Shapiro, Phys. Rev. D
69,044009 (2004).
5. S. A. Balbus and J. F. Hawley, Astrophys. 1. 376, 214 (1991); Rev. Mod. Phys. 70, 1
(1998).
6. M. D. Duez, Y. T. Liu, S. L. Shapiro, and B. C. Stephens, Phys. Rev. D 72, 024028
(2005).
7. M. Shibata and Y.-I. Sekiguchi, Phys. Rev. D 72, 044014 (2005).
8. M. D. Duez, Y. T. Liu, S. L. Shapiro, M. Shibata, and B. C. Stephens, Phys. Rev.
Lett. 96, 031101 (2006).
9. M. D. Duez, Y. T. Liu, S. L. Shapiro, M. Shibata, and B. C. Stephens, Phys. Rev. D
73, 104015 (2006).
10. M. Shibata, M. D. Duez, Y. T. Liu, S. L. Shapiro, and B. C. Stephens, Phys. Rev.
Lett. 96, 031102 (2006).
11. M. D. Duez, S. L. Shapiro, and H.-J. Yo, Phys. Rev. D 69, 104016 (2004).
12. M. D. Duez, Y. T. Liu, S. L. Shapiro, and B. C. Stephens, Phys. Rev. D 69, 104030
(2004).
 HEAD-ON COLLISIONS OF DIFFERENT INITIAL DATA
ULRICH SPERHAKE, BERND BRUGMANN, JOSE A. GONzALEZ,
MARK D. HANNAM, SASCHA HUSA
Theoretical Physics Institute,
University of Jena, D-07743 Jena, Germany

We discuss possible origins for discrepancies observed in the radiated energies in head-on
collisions of non-spinning binaries starting from Brill-Lindquist and superposed Kerr-
Schild data. For this purpose, we discuss the impact of different choices of gauge param-
eters and a small initial boost of the black holes.

1. Introduction
The area of numerical relativity has made dramatic progress in the last two years 3 ,5,9
and numerical simulations of black hole binaries performed by various groups have
resulted in a wealth of literature on astrophysical topics and those related to gravi-
tational wave data analysis. At the same time, laser-interferometric GW detectors,
LIGO, GE0600, TAMA, VIRGO, have started collecting data at design sensitivity.
The area of GW physics has thus entered a very exciting stage with vast potential
for astrophysics, our understanding of the early universe and fundamental physics.
From the viewpoint of numerical relativity, though, a number of important ques-
tions still remain to be addressed. These largely concern the accuracy of the pro-
duced waveforms, the dependency of the results on numerical techniques, their de-
tailed matching with results predicted by approximation theories as well as the
mass production of waveforms covering the complete parameter space for use in
GW observations and parameter estimation.
The purpose of this study is to address the dependency of the numerical results
on the choice of black-hole binary initial data. In contrast to spacetimes containing
single stationary black holes, there exist no uniqueness theorems guaranteeing that
two data sets for binary black holes using different data types represent the same
physical configuration. Indeed, such data sets are known to generally differ in the
amount of gravitational radiation inherent in the initial data.
The dependency on initial data parameters (though not data type) has been
studied in the case of binary black hole coalescence in Refs. 2,4,6. Using the moving
puncture technique, the merger waveforms are found to agree well for different
initial separations and algorithms to produce quasi-circular initial configurations. A
comparison of GWs produced in the evolution of Cook-Pfeiffer and puncture data
using different evolution techniques has been presented in Ref. 1 and shown good
agreement. There remains a difficulty in the identification of free initial parameters
in this case, however (cf. the non-vanishing spin in the Cook-Pfeiffer data set in
this comparison). This identification of parameters represents a simpler and cleaner
task in the case of head-on collisions of non-spinning black holes which has been
studied in Ref. 11. That study observed systematically larger amplitudes by about
10 % in the merger waveform resulting from Kerr-Schild data compared with those

1612
 1613

of Brill-Lindquist data. Here we investigate two possible causes for this discrepancy:
the dependency of the results on the gauge trajectories in the case of Kerr-Schild
data and the impact of deviations from time symmetry of the initial data.

2. Results
The simulations presented in this work have been obtained with the LEAN code 7 ,1l
which uses the BSSN formulation of the Einstein equations together with the moving
puncture approach. 3 ,5 It is based on the CACTUS 8 computational toolkit and the
CARPET lO mesh-refinement package. For a detailed description of the code as well
as the construction of initial data we refer the reader to Ref. 11.
We first discuss the gauge trajectories used in Ref. 11 for the Kerr-Schild data.
There, algebraic gauge conditions are constructed which require trajectories for the
(approximate) black hole positions (see 11 for details). These are prescribed as
polynomials ±xi(t) = Xo + vQt + ait 2 /2 + j i t 3 /6 + qi t 4 /24 which are smoothly (up
to fourth derivatives) matched to the static function Xi(t) = 0 in a time interval
tl < t < t2. Here xb, vb, ab, qb, hand t2 are free parameters which need to be
chosen carefully to avoid numerical instabilities. In Table 1 we list the values for
each (Kerr-Schild) model. In order to assess the impact of the particular choice of
these parameters, we have evolved the initial data of model 1 with alternative gauge
parameters as listed in the second row of the table. This alternative gauge trajectory
is motivated by the initial coordinate velocity v = -0.08 of the central position of
the apparent horizon as measured using Thornburg's AHFinderDirect. 12 ,13
The resulting waveforms are shown in the left panel of Fig. 1. Both the waveforms
and the radiated energies thus obtained for model 1 show excellent agreement. The
differences in radiated energy are about 1.5 % and thus substantially smaller than
the discrepancies between EKs and E BL .
Second, we assess the impact of deviations from exact time symmetry of the
initial superposed Kerr-Schild data. These deviations manifest themselves in a small
but non-vanishing initial coordinate velocity of the superposed Kerr-Schild holes as
measured by the central position of the apparent horizon. For the case of model 2

Table 1. Parameters for the black hole models. The gauge parameters in columns 5 to 9 are
only used for the Kerr-Schild simulations. There we also use q'O = -0.000278, -0.000165 and
-0.000104 M3 respectively for models 2a, 3 and 4. Here M is the ADM mass of the system.
The two rightmost columns list the energy radiated in the £ = 2, m = 0 mode for both data
types, ignoring contributions due to the spurious initial burst.

Model DKS DBL %E,b V 0Z aiS M 1'0 M2 iL .E2 %EKS %E~L

1a 10 8.6 2.8 0 -0.037 0.0038 10 35 0.066 0.051

1b 10 2.8 -0.08 -0.0061 -0.0002 20 40 0.065
2a 12 10.2 2.4 0 -0.029 0.0040 25 50 0.067 0.052
2b 10.2 2.4 (with initial physical boost v = 0.067) 0.0525
3 14 12.5 2.0 0 -0.022 0.0027 25 57 0.067 0.052
4 16 14.6 1.6 0 -0.018 0.0020 34.5 84.7 0.086 0.054
1614

Mrc,\IIn
O.02r--~~-,-~~-,----~~-,-~~,

0.01
0.01

-o.oJ

-0.Q2

-O.030·~~-5':;-O~~~'()(J;--~----;;150;;--~---c;;.ZOO -O.020';--~~5':;-O~~-I,*(J(J'-~----;;150;;--~---c;;.200
tiM tiM

Fig. 1. The jI = 2, m = 0 mode of the scaled Newman Penrose scalar Mrex'lj;4 extracted at
rex = 50 M for models la, lb (left panel) and models 2a, 2b (right panel).

we have measured this velocity to be v = 0.067. In order to estimate what impact

such an initial velocity has on the resulting waveforms, we have applied an initial
linear momentum pz = m v to the Brill-Lindquist version of this model, where m is
the irreducible mass of a single hole. The resulting waveform is compared with its
non-boosted counterpart in the right panel of Fig. 1. Again, the wave amplitudes
show good agreement, as do the resulting values for the radiated energy in Table 1.
In summary, we find the observed differences in radiated energy resulting from
modifications of the gauge trajectories and a possible initial boost of the black
holes to be of the order of 1 % and thus substantially below the differences of about
> 20 % observed in Ref. 11 between the two types of initial data.
Acknowledgments
This work was supported by DFG grant SFB/Transregio 7 "Gravitational Wave As-
tronomy", and the DEISA Consortium (co-funded by the ED, FP6 project 508830).
computations were performed at LRZ Munich and HLRS, Stuttgart.
References
1. J. G. Baker, M. Campanelli, F. Pretorius, and Y. Zlochower. 2007. gr-qc/0701016.
2. J. G. Baker et al. Phys. Rev. D, 73:104002, 2006.
3. J. G. Baker, J. Centrella, D.-I. Choi, M. Koppitz, and J. van Meter. Phys. Rev. Lett.,
96:111102, 2006.
4. B. Briigmann et al. 2006. gr-qc/0610128.
5. M. Campanelli, C. O. Lousto, P. Marronetti, and Y. Zlochower. Phys. Rev. Lett.,
96:111101, 2006.
6. M. Campanelli, C. O. Lousto, and Y. Zlochower. Phys. Rev. D, 74:041501, 2006.
7. J. A. Gonzalez, M. D. Hannam, U. Sperhake, B. Briigmann, and S. Husa. 2007. gr-
qc/0702052.
8. T. Goodale et al. In Vector and Parallel Processing - VECPAR '2002, 5th International
Conference, Lecture Notes in Computer Science, Berlin, 2003. Springer.
9. F. Pretorius. Phys. Rev. Lett., 95:121101, 2005.
10. E. Schnetter, S. H. Hawley, and 1. Hawke. Class. Quantum Grav., 21:1465-1488,2004.
11. U. Sperhake. 2006. gr-qc/0606079v1.
12. J. Thornburg. Phys. Rev. D, 54:4899-4918, 1996.
13. J. Thornburg. AlP Conference Proceedings, 686:247-252, 2003.
 TEARING INSTABILITY IN RELATIVISTIC MAGNETICALLY
DOMINATED PLASMAS
M.Y. BARKOy 1 ,2, S.S. KOMISSAROy 1 , M. LYUTIKOy 3
1 Department of Applied Mathematics, the University of Leeds, Leeds, LS2 9GT, UK. e-mail:
[email protected]
2 Space Research Institute, 84/32 Profsoyuznaya Str, Moscow 117997, Russia,e-mail:
[email protected]
3 University of British Columbia, 6224 Agricultural Road, Vancouver, BC, Y6T lZl, Canada

Many astrophysical sources of high energy emission, such as black hole magnetospheres,
superstrongly magnetized neutron stars (magnetars), and probably relativistic jets in
Active Galactic Nuclei and Gamma Ray Bursts involve relativistically magnetically dom-
inated plasma. In such plasma the energy density of magnetic field greatly exceeds the
thermal and the rest mass energy density of particles. Therefore the magnetic field is the
main reservoir of energy and its dissipation may power the bursting emission from these
sources, in close analogy to Solar flares. One of the principal dissipative instabilities that
may lead to release of magnetic energy is the tearing instability. In this paper we study
(see also 10 ), both analytically and numerically, the development of tearing instability in
relativistically magnetically-dominated plasma using the framework of resistive magne-
todynamics. We confirm and elucidate the previously obtained result on the growth rate
of the tearing mode: the shortest growth time is the same as in the case of classical non-
relativistic MHD, namely T = y'TaTd where Ta is the Alfven crossing time and Td is the
resistive time of a current layer. The reason for this coincidence is the close sim-
ilarity between the governing equations, especially in the quasi-equilibrium
approximation. In particular, the role of the mass density of nonrelativistic
MHD is played by the mass-energy density of the magnetic field, p = B2 /87rc2 •

Keywords: stars:pulsars - black hole physics - MHD - methods:analytical - meth-

ods:numerical

Dissipation of magnetic energy may power high energy emission in a variety of

relativistic astrophysical phenomena, e.g. pulsar wind nebulae,2,9,1l,19 jets of active
galactic nuclei,1,15 gamma-ray bursts,3,4 and magnetars. 12 ,18
The principal resistive instability in a conventional, non-relativistic plasma is a
so-called tearing mode. It is one of the principle unstable resistive modes, which
plays the main role in various TOKAMAK discharges like sawtooth oscillations
and major disruptions,8 unsteady reconnect ion in Solar fiares 1,16 and Earth mag-
netotail. 6 Qualitatively, the most important property of the tearing mode is that a
current layer of thickness l may dissipate on time scales much shorter than it the
resistive time scale Td rv l2/ry. In addition, tearing mode may be an initial stage of
the development of the (steady-state) reconnect ion layers.
Development of the tearing mode in a relativistic, strongly magnetized plasma
(urn» 1) is the principal topic ofthis work. lO Initially, this problem has been con-
sidered by Lyutikov,12 who found that in resistive magneto dynamics the current
layers are unstable towards formation of resistive small-scale current sheets, that's
to development of tearing mode. He also found that the growth rate of tearing mode
is intermediate between the short Alfven time scale Ta rv l/v a (which in a Urn » 1

1615
1616

plasma is the light crossing time scale Tc rv llc) and a long resistive time scale Td:
T rv (TdTa)1/2. Surprisingly, this is exactly the same expression as the one found
in the non-relativistic framework of incompressible magnetohydrodynamics. In the
paper Ref. 10 we uncover the reasons for this coincidence. We show that
the equations describing slow evolution of relativistic magnetically dom-
inated plasma are very similar to those of non-relativistic MHD. In full
agreement with basic ideas of the theory of relativity the magnetic field
has inertia and the role of mass density is played by B2 181fc2 • This mass
moves with the drift speed V = cE x B I B2 under the action of magnetic
stresses. Moreover, the linearized equations describing the evolution of
tearing mode can be written in exactly the same form as those of incom-
pressible MHD. We have carried out such simulations (see Fig.1 Right panel)
and they fully confirmed Lyutikov's12 conclusion on the growth rate in magneti-
cally dominated regime - it indeed coincides with that found earlier for the tearing
instability in non-relativistic incompressible MHD, the shortest growth time being
equal to the geometric mean of the Alfven and resistive timescales. 5 ,14 Moreover,
like in the non-relativistic case we observed formation of magnetic islands in the
non-linear phase of the instability. This is accompanied by the appearance of notice-
able O-type and X-type neutral points (or lines, see Fig.1left panel). The develop-
ment of an X-point may be considered as a first step in setting up of reconnection
layer where dissipated magnetic energy and magnetic flux are constantly replen-
ished due to plasma inflow. 17 In our numerical experiments in the non-liniar phase,
the grow rate reachs the maximal possible value rv (TdTa)-1/2 for k < £";//4, here
Lv. = V2T,dTc is relativistic Lundquist number, k is the dimensionless wavenumber.

,,
,,
.,,
,
,,
,

~ 4 ~ 0 2 -1 -0.5 _ 1/4 0 0.5

X 10910 k Lu

Fig. 1. Left panel: Multiple magnetic islands at the nonlinear stage of the model with k 0.0898
0.185, 'I) 10- 3 ). Right panel: Growth rate as a function of the wavelength. Dashed
line: 1) 10- 3 solid hne: 1) = 10- 4 .

Our results support the possibility of magnetar flares being magnetically driven
reconnection-like events that occur in magnetar magnetospheres 12 ,13 in a similar
fashion to flares and coronal mass ejections of Sun's magnetosphere. More com-
 1617

plicated magnetic field geometries, e.g. those with nonvanishing magnetic tension,
should also be subject to tearing instability on time scales intermediate between the
resistive and Alfven time scales. 14 The growth rate typically scales as T rv T~Td~a,
where 0 < 0: < 1 is some coefficient (0: = 1/2 in our case).

Acknowledgments
This research was funded by PPARC under the rolling grant "Theoretical Astro-
physics in Leeds"

References
1. Aschwanden M. J., Space Science Reviews, 101, 1, (2002).
2. Coroniti F. V.,ApJ, 349, 538, (1990).
3. Drenkhahn G.,ACfA, 387, 714, (2002).
4. Drenkhahn G., Spruit H.C.,ACfA, 391, 1141, (2002).
5. Furth H.P, Killeen J., Rosenbluth M.N.,Phys.Fluids, 6,459, (1963).
6. Galeev A. A., Coroniti F. V., Ashour-Abdalla, M.,Geophys.Research Lett., 5, 707,
(1978).
7. Jaroschek C.H., Lesch H., Treumann RA., ApJ.Lett., 605, L9, (2004).
8. Kadomtsev B.B., Soviet Journal of Plasma Physics, 1, 710, (1975).
9. Kirk J.G., Skjaraasen 0., ApJ, 591, 366,(2003).
10. Komissarov S.S., Barkov M.V., Lyutikov M., acepted by MNRAS, astro-ph/0606375,
(2006).
11. Lyubarsky Y.E., Kirk J.G., ApJ, 547, 437, (2001).
12. Lyutikov M.,MNRAS, 346,540, (2003).
13. Lyutikov M.,MNRAS, 367, 1594, (2006).
14. Priest E., Forbes T., Magnetic Reconnection. MHD Theory, Applications, Cambridge
Uni.Press, Cambridge, (2000).
15. Romanova M.M., Lovelace RV.E.,ACfA, 262, 26, (1992).
16. Shivamoggi B.K.,Astrophys.Sp.Sci., 114, 15, (1985).
17. Syrovatskii S.I.,Ann.Rev.Astron.Astrophys., 19, 163, (1981).
18. Thompson C., Duncan R C.,MNRAS, 275, 255, (1995).
19. Usov V.V.,MNRAS, 267, 1035, (1994).
This page intentionally left blank
Black Hole Collisions
This page intentionally left blank
 BLACK HOLE BREMSSTRAHLUNG IN THE NONLINEAR
REGIME OF GENERAL RELATIVITY

H. P. de OLIVEIRA 1, 1. DAMIAo SOARES 2 and E. V. TONINI3

1 Universidade do Estado do Rio de Janeiro
R. Sao Franscisco Xavier, 524, Maracana, CEP 20550-013, Rio de Janeiro, RJ, Brazil
2 Centro Brasileiro de Pesquisas Fisicas/MCT
R. Dr. Xavier Sigaud, 150, Urea, CEP 22290-180, Rio de Janeiro, RJ, Brazil
3 Centro Federal de Educa<;iio Tecnologiea do Espirito Santo
Avenida Vitoria, 1729, Jucutuquara, CEP 29040-780, Vitoria, ES, Brazil
1 [email protected]
2 [email protected]
3 [email protected]

We examine the gravitational wave emission process of a black hole in decelerated motion.
The deceleration of the black hole is realized by nonlinear gravitational perturbations
in the realm of Robinson-Trautmann geometries. In the nonlinear regime of General
Relativity a radiative transfer process is set up through which the black hole loses kinetic
energy and part of its rest mass by the emission of gravitational waves. The angular
pattern of the radiation emitted is typical of bremsstrahlung. The process has an initial
dominant pulse corresponding to a burst with a high power output.

Keywords: gravitational bremsstrahlung; black hole dynamics; gravitational waves.

An important astrophysical situation in which gravitational radiation IS pro-

duced is that of a black hole with arbitrarily large initial velocity being decel-
erated by a nearby large mass. The gravitational radiation produced is typically
bremstrahlung, in close analogy with electromagnetic bremstrahlung. However a
fundamental distinction from the electromagnetic case is that in the full General
Relativity dynamics the radiation extracts mass-energy from the black hole. Gravi-
tational bremsstrahlung produced when two stars fly past each other was originally
discussed in Refs. [1-4], using linear/postlinear and post-Newtonian approximations.
Here we examine a black hole initially in uniform motion that is perturbed as it
flies past a massive body, in an unbounded low-deflection orbit. As a consequence
of the gravitational interaction the black hole undergoes deceleration and starts to
emit gravitational waves. The perturbation is modelled in the realm of Robinson-
Trautmann spacetimes 5 and use is made of the full General Relativity dynamics.
Our results are given in the frame of an asymptotic inertial observer relative to
which the black hole is initially in uniform motion. Our approach relies on a recent
treatment of the dynamics of the collapse of an isolated bounded source emitting
gravitational waves in the nonlinear regime. 6 We use units such that 87rG = c = 1.
Robinson-Trautman (RT) metric" are described by

ds 2 = ().(u, e) - 2 ~LO + 2r ~)du2 + 2dudr _ r2 K2(U, 8)(de 2 + sin 2 ed<p2), (1)

1 Kee K~ Ke
).(u,e) = K2 -, K3 + K4 - K3 cote, (2)

1621
1622

k (Ae sine)e
-6 m o K + 2K2 sine = O. (3)

In the above, mo corresponds to the Schwarzschild mass when K = 1 and the

dynamics of the gravitational field is governed by Eq. (3), the RT equation.
In the semi-null local Lorentz frame of Ref. [6], the curvature tensor components
contributing to the gravitational degrees of freedom transversal to the direction of
propagation of the wave are
D(u, e) 1
R 0303 = -R0202 = -
r
+ O( 2")'
r
(4)

where

D(u e) = - 0
1 Kee Ke
- - cot e - 2 ( - )
Ke 2]
, 2K2 U [- K KK (5)

The function D contains all the information of the angular, and time dependence
of the gravitational wave amplitudes in the wave zone.
Now we consider the static solution (that yields ,\ = 1 in Eq.2),
1
K(e) - - - - - - - (6)
cosh I + cos e sinh I
This solution can be interpreted 7 as a black hole boosted along the negative z-axis,
with velocity parameter v = tanh I and mass function m( e) = moK3 (e) and total
mass-energy M = mo cosh I = mo/~.
A perturbation of the above solution (6) is done by expanding it in a Legendre
polynomial infinite series and truncating this expansion - at u = 0, say - into
1 N
K(e,O) = exp (-2" LbkPk(COSe)) (7)
k=O
where N < 00. This perturbation corresponds to decelerate the black hole with
3
initial deceleration proportional to 6.v(O) = cosh I - ~ J~l K (x, O)dx. The sytem
becomes then nonstationary starting to evolve according RT Eq. (3) and emitting
gravitational waves. We evolve the above initial conditions (7) via the RT equation
from u = 0, using the Galerkin projection method (cf. Ref. [6]) corresponding to a
truncation N = 7. During deceleration the configuration starts to emit gravitational
waves losing rest mass and kinetic energy. The curvature tensor of the initially static
configuration has now a radiative term proportional to D( u, e) that defines the time
and angular pattern of the amplitudes of gw in the wave zone.
Exhaustive numerical experiments show that for very large u c::: 300,000,
all modal coefficients bk become constant (up to 10-6 ). From these modal co-
efficients bk(final), evaluated at u = 300,000, we reconstruct K(e, final) =
exp (~ L.~=o bk(final)Pk( cos e)) that results, in all cases

Ko
K(e, final) c::: - - - - - - - - : - - - - (8)
cosh If inal + cose sinh If inal
 1623

0(U,-08)

(a) (b)
Fig. 1. (a) Angular distribution pattern D(u, e) of the curvature tensor at the wave zone in
(u = 100 - 400) for v ~ 0.76, showing that gw are predominantly emitted in a cone in the
direction of the motion, typical of the electromagnetic bremsstrahlung of a decelerated charge along
its direction of motion. (b) The initial pulse D(u, x = -0.8) occuring in the interval u = (0,1000)
for 'Y = 1.0; the horizontal line is a measure of the width of the pulse, evaluated at lie of the
maximum of D.

where now Ko < 1, contrary to the initial Ko = 1 adopted, so that the final
configuration corresponds to a boosted black hole with smaller velocity parameter
'final < , and final rest mas,s Mfinal = moKij < mo. Numerical results yield the
0.395
Mass Loss Function Mfinal = 0.999mo ( 1_V 5 . 19 ) , where v is the initial velocity
of the black hole. Further Figs. 1 display the angular and time patterns in the wave
zone. The angular pattern is typical of bremsstrahlung and the temporal signal is
that of an initial intense pulse, corresponding to a burst of gravitational waves. For
, = 1.0 and considering the rest mass of the black hole of the order of two solar
masses, the power output of the initial pulse is about 10 40 Gev 2 (not taking into
account the kinetic energy loss), a power typical of a gamma-ray burst.

References
1. P. C. Peters, Phys. Rev. D1, 1559 (1970).
2. R. V. Wagoner and C. M. Will, Astrophys. J. 210, 764 (1976).
3. M. TUrner, Astrophys. J. 216, 610 (1977); M. TUrner and C. M. Will, Astrophys. J.
220, 1107 (1978).
4. S. J. Kovacs and K. S. Thorne, Astrophys. J. 224, 62 (1978).
5. I. Robinson and A. Trautman, Phys. Rev. Lett. 4, 431 (1960); Proc. Roy. Soc. A265,
463 (1962).
6. H. P. Oliveira and I. Damiao Soares, Phys. Rev. D70, 084041 (2004).
7. H. Bondi, M. G. J. van der Berg, and A. W. K. Metzner, Proc. R. Soc. London A 269,
21 (1962); R. K. Sachs, Phys. Rev. 128, 2851 (1962).
 HYPERBOLOIDAL FOLIATIONS WITH J'-FIXING IN
SPHERICAL SYMMETRY

ANIL ZENGINOGLUl, SASCHA HUSA2

1 Max-Planck-Institut
fur Gmvitationsphysik (AEI)
D-14476 Potsdam, Germany
[email protected]

2 Universitiit Jena, Theoretisch-Physikalisches Institut,

Max- Wien-Platz 1, 07743 Jena, Germany
sascha. [email protected]

We study in spherical symmetry the conformal compactification for hyperboloidal folia-

tions with nonvanishing constant mean curvature. The conformal factor and the coordi-
nates are chosen such that null infinity is at a fixed radial coordinate location.

Hyperboloidal surfaces in asymptotically flat spacetimes have first been used for
an initial value formulation of the Einstein equations by Friedrich. 4 Instead of ap-
proaching spatial infinity as Cauchy surfaces do, they reach null infinity, J', which
makes them suitable for radiation extraction. Contrary to characteristic surfaces,
these spacelike surfaces are as flexible as Cauchy surfaces and they can be used
in numerical calculations with the 3+1 approach based on a hyperboloidal initial
value problem. 3 ,5-7 We want to study the conformal compactification ll of hyper-
boloidal foliations in spherical symmetry. It has been suggested 1 ,2,8 that conformal
compactifications in which J' is kept at a fixed spatial coordinate location might
be useful for testing new ideas in numerical calculations. Here we explicitly discuss
the simplest cases, namely the Minkowski and Schwarzschild spacetimes.
The physical line element in spherical symmetry can be written as

(1)
d(J"2 is the standard metric on S2 and the lapse a, the shift /3, and the spatial metric
function It are functions of the coordinates (t, f) only. We assume that the metric (1)
admits a regular conformal compactification and that the time coordinate t is such
that t=const. hypersurfaces are hyperboloidal hypersurfaces. We do not compactify
the time direction. The conformal compactification, 9 = D2g, can be done such that

(2)
with respect to a compactifying radial coordinate r. Note that we have some freedom
here. One can require for example h = 1 which leads to r being the proper distance,
however, then the radial coordinate transformation can not, in general, be written in
explicit form. lO By keeping h, we have the freedom to prescribe the conformal factor
in terms of a compactifying radial coordinate r and the coordinate tranformation
is explicit. The relation (2) implies for a given conformal factor D(r) a coordinate

1624
 1625

transformation f = D- 1 r so that df = (D - r D')D- 2 dr. Then the spatial metric

function transforms as h = (D - r D')D- 1 h. For the regularity of this conformal
compactification, h(t, f) needs to have a specific asymptotic fall-off behaviour for
f -> CXJ on the hyperboloidal surfaces of constant t. A simple choice for the conformal
factor that we will study is D = (1 - r), which implies h = D- 1 h. This is not a good
choice at the origin, but we are interested in the asymptotic region.
A simple example for hyperboloidal foliations are constant mean curvature
(CMC) foliations. We write the line element in Minkowski spacetime with standard
coordinates (l, f) as i) = -dP + df2 + f2 da 2 . We introduce t(l, f) = l - 2 + f2 va
as a new time coordinate. The constant a E lR is related to the constant mean
extrinsic curvature K of the level sets of t(l, f) by a = 3/ K. We use the convention
in which positive K means increasing volume to the future. The sign of K deter-
mines whether the surfaces reach J'+ or YO-. To get a feeling for these surfaces, we
analyse them in the familiar compactification of the Minkowski spacetime given by
the transformation 3
- 1 1
t(V, U) = 2(tan V + tanU), f(V, U) = 2(tan V - tan U).
The subsequent rescaling with the conformal factor D = cos V cos U leads to

71 = D2i) = -dUdV + sin2(~ - U) da 2 .

J'+ is at V = 7r /2 in these coordinates. Embedding our hyperboloidal surfaces into

the conformally extended Minkowski spacetime leads to

t(V, U) = ~(tan
2
V + tanU) - !a 2 + ~(tan V -
V 4 tanU)2.

Writing a series expansion in cot V near J'+,

t(V, U) :::::: tan U - a2 cot V - a2 tan U cot 2 V + O( cot 3 V), for V -> 7r /2,
we see that the cut at yo + depends on the value of t via U (t, V) Iyr + = arctan t, but
does not depend on the mean extrinsic curvature which determines the angle of the
cut. Fig. 1 shows three foliations in the Penrose diagram of Minkowski spacetime
for the same set of values of t, but different values of K. Null surfaces in a Penrose
diagram have an angle of 45 degrees to the horizontal. As each plotted surface is
spacelike, their angle is smaller. By the choice of the mean extrinsic curvature, we
can control the behaviour of the surfaces in the interior without changing their
asymptotics. For example, by choosing IKI small, we can make the hyperboloidal
surfaces behave, in a certain sense, more similar to Cauchy slices.
In Penrose diagrams of Minkowski and Schwarzschild, the null generators of ,)1+
converge. To avoid the corresponging loss of resolution in numerical calculations,
we would like to fix the radial coordinate location of J'+. The Minkowski metric in
the time coordinate t of a CMC-foliation reads
2f a2
= -dt 2 - + -2 df2 + da 2 .
i)
va 2 +f2
dtdf 2
a +r
f2
1626

v
/
/

Fig. 1. CMC-foliations of the Minkowski spacetime for K = 15,3,1.5

Conformal compactification, rJ = [22fJ, using (2) with [2 = 1 - r results in

2
2 2 2r a 2 2 2
rJ=-(l-r) dt - dtdr+ 2( )2 2 dr +r dCJ. (3)
yla 2(1-r)2+ r 2 a 1-r +r
For the Schwarzschild spacetime 9s, the general family of spherically symmetric
constant mean curvature surfaces has been constructed in. 9 Conformal compactifi-
cation, gs = [229s, with [2 = 1 - r results in
2m(1-r)) 2 2 2(C(1-r)3-Kr 3/3) r4 222
gs = - ( 1 - r (l-r) dt - P(r) dtdr+ p2(r) dr +r dCJ ,
(4)
where m, K, and C are constants and
1

P(r) := ((C(l _ r)3 _ Kr 3/3))2 + (1 _ 2m(~ - r)) (1- r)2r4) "2

As seen in the examples (3,4), to keep yr+ at a fixed radial coordinate location we
need an inward pointing shift vector in the asymptotic region.

Acknowledgments
The authors thank Helmut Friedrich for discussions. This work was supported in
part by the SFB /Transregio 7 of the German Science Foundation.

References
l. Andersson L 2002 Lect. Notes Phys. 604 183-194, gr-qc/0205083
2. Frauendiener J 1998 Phys.Rev.D 58 064003, gr-qc/9712052
3. Frauendiener J 2004 Living Rev. Relativity 7 253
4. Friedrich H 1983 Commun. Math. Phys. 91 445-472
5. Hubner P 2001 Class. Quant. Grav. 18 1871-1884, gr-qc/0010069
6. Husa S 2002 Lect. Notes Phys. 604 239-260, gr-qc/0204043
7. Husa S 2003 Lect. Notes Phys. 617 159-192, gr-qc/0204057
8. Husa S, et. al. 2006 AlP Conf. Proc. (ERE 2005) 841 306-313, gr-qc/0512033
9. Malec E and Murchadha N 2003 Phys. Rev. D 68 124019, gr-qc/0307046
10. Moncrief V, talk given in "From geometry to numerics", Paris, 20-24 November 2006
1l. Penrose R 1963 Phys. Rev. Lett. 10 66-68
 HIGH-ORDER PERTURBATIONS OF SPHERICAL SPACETIMES

DAVID BRIZUELA

JOSE MARIA MARTIN-GARCIA

GUILLERMO A. MENA MARUGAN

Instituto de Estructura de la Materia (CSIC)
Serrano 121-123, 28006-Madrid, Spain

We generalize the Gerlach and Sengupta (GS) formalism for linear perturbations of a
spherical background to high orders. With this aim, we define tensor harmonics of any
rank that reduce to the Regge-Wheeler-Zerilli (RWZ) harmonics for the low-rank cases.
In addition, we provide a closed formula for the product of any pair of these tensor
harmonics. We analyze the gauge invariance and describe an iterative procedure for
the construction of high-order gauge-invariant objects. At second order, gauge-invariant
variables and Einstein equations are explicitly obtained.

1. Introduction

Perturbation theory plays a prominent role in understanding physical processes and

proving the stability of solutions in General Relativity. Studying high orders has
been impracticable until very recently because of the complexity of the expressions
under consideration. Nowadays, it has become feasible thanks to the development
of the tensor computer algebra tools. High orders will provide more precision, as
well as error bars, to the results obtained so far, and delimitate regions of validity
for the first-order calculations. Another important improvement will be that the
nonlinearity of the theory will be reflected in the coupling of the first-order modes.
The aim of this work is to generalize the as
formalism to cope with the problem of
high-order perturbation theory for a generic spherical background.

2. High-order perturbation theory in General Relativity

We will consider a parametric family of spacetimes {M(c), gp;v(c), Tp;v(c)}, where c
is a dimensionless parameter, M a four-dimensional manifold, gp;v the metric tensor
and Tp;v the stress-energy tensor. The background spacetime will correspond to the
particular case c = 0 and all the background objects will be denoted without tilde,
that is, {M(O),gp;v(O), Tw'(O)} == {M,gp;v, Tw}' This background is assumed to be
an exact solution of the Einstein equations. Supposing a smooth dependence in c,
we write gp;v(E) = gp;v + 2.:::=1 ~~ {n}hp;v and Tp;v(E) = Tp;v + 2.:::=1 ~~ {n}yp;v' It is
then possible to deduce closed formulas that express the nth perturbation of all the

1627
1628

curvature tensors of interest in terms of the metric perturbations {=}h!"v (m :s: n).l
These formulas involve sums over all the sorted partitions of the number n.

3. Spherical background
3.1. Gerlach and Sengupta notation
Since the background metric is spherically symmetric, we decompose it as 2

where Greek indices, capital Latin indices and lowercase Latin indices take values
in the ranges (0, 1,2,3), (0, 1) arid (2,3) respectively. lab is the round metric on the
sphere and its associated covariant derivative will be denoted with a semicolon. A
block decomposition similar to (1) is done for the stress-energy tensor T!"v'

3.2. Spherical tensor harmonics

The so-called RWZ tensor harmonics 3 ,4 provide bases for expanding scalars, vectors
and symmetric rank-two tensors on the sphere. These harmonics have a well-defined
parity and are tangent to the sphere. They are classified into two polarity families:
under a parity transformation the polaT and axial harmonics change sign as (-1)1
and (_1)1+1, respectively, where the integer I is the corresponding harmonic label.
In order to face high-order perturbation theory, we generalize the RWZ harmon-
ics to any rank defining the polar harmonic Zr
a, ... a s == (Y[m :a, ... aJ
ST
and the axial
harmonic Xra, ... as == E(a,bZrba2 ... aS)' where Y[m is the scalar harmonic and the
superscript ST means the symmetric and trace-free part. In this way, a complete
basis for tensors of rank 8 is composed by Zr m
a, ... a s , X l a, ... a s and independent
linear combinations of products between ,ab and Eab with the basis of rank (8 - 2).
We have also obtained 1 a closed formula that expresses the product between any
pair of RWZ generalized tensor harmonics as linear combination of harmonics.

4. Non-spherical perturbations
We expand the perturbations of the metric using the RWZ basis of harmonics:
{nlHmAB zm {n}HmA zm +(n}h m xm )
{nih = ""' I I I I b I A I b (2)
!"v - ~(
I,m
Sym
.
{n}K rn 1'2", zm
I lab I
+ {n}C m 1'2 zm
I I ab
+{n}h'" Xrn
I I ab
.

A similar decomposition is done for the matter perturbations {n}T!"v' But these per-
turbations have some gauge (nonphysical) degrees of freedom. The gauge freedom of
the nth-order perturbation of a background tensor is encoded in n vector fields. 5 We
have demostrated that choosing vanishing harmonics coefficients {n}Hlm A, {n}C z
z
and {n}h fixes the gauge at any order. For the case n = 1 this gauge was intro-
duced by Regge and Wheeler. 3 Another way of removing these nonphysical degrees
of freedom is by using gauge invariant quantities. These invariants were introduced
 1629

by Moncrief 6 for Schwarzschild at linear order. GS 2 generalized them for any spher-
ically symmetric background and, taking linear combinations of the matter and
metric harmonic coefficients (2), they constructed ten matter invariants, denoted
{{l}I]i}, and six metric invariants, given by the polar objects {{l}Ki AB, {l}Ki} and
the axial vector {1}Kr A. We have constructed 7 an iterative method to generalize
these objects up to any order, under certain restrictions on the harmonic labels.
At second-order, we have explicitly calculated 7 the invariants that arise from the
same linear combinations that appear at first order, but supplemented with some
first-order quadratic terms. The second-order Einstein equations are schematically
given (e. g. in the polar case) byl
L[ {2}K Im j +'" '"
L..-L..- (c1 5 I711'rj>m
II
= 81l' {2}l]im
I' (3)
r,f m,m
where L are the GS linear differential operators2 acting on the polar metric invari-
r
ants {2}Ki, and {2 hp represents the second-order polar matter invariants. The
5 sources depend on the harmonic labels of the two first-order modes, (l, in) and
(I, m), that couple to give rise to a mode with labels (l, m), and we have defined
the sign E == (_1)/+[-1. The (+15 sources are composed by polar x polar and axial
x axial terms with real coefficients, whereas the (-15 sources contain polar x axial
terms with purely imaginary coefficients.

5. Conclusions
We have generalized the RWZ harmonics to any rank and obtained a closed formula
for their products. We have also found an iterative procedure to construct the high-
order gauge-invariant objects. This has permitted us to generalize the GS formalism
to high orders, achieving a covariant and gauge-invariant framework that has been
algebraically implemented.

Acknowledgments
D.B. acknowledges financial support from the FPI program of the Regional Gov-
ernment of Madrid and a Marcel Grossman Grant. J.M.M.-G. acknowledges the fi-
nancial aid provided by the 13P framework of CSIC and the European Social Fund.
This work was supported by the Spanish MEC Project No. FIS2005-05736-C03-02.

References
1. D. Brizuela, J. M. Martin-Garcia and G. A. Mena Marugan, Phys. Rev. D 74,044039
(2006).
2. U. H. Gerlach and U. K. Sengupta, Phys. Rev. D 19, 2268 (1979).
3. T. Regge and J. A. Wheeler, Phys. Rev. 108, 1063 (1957).
4. F. J. Zerilli, Phys. Rev. Lett. 24,737 (1970).
5. S. Son ego and M. Bruni, Commun. Math. Phys. 193, 209 (1998).
6. V. Moncrief, Annals of Physics 88, 323 (1973).
7. D. Brizuela, J. M. Martin-Garcia and G. A. Mena Marugan, In preparation.
 ANALYTIC SOLUTIONS OF THE LINEARIZED EINSTEIN
EQUATIONS USED TO TEST AND DEVELOP A
CHARACTERISTIC CODE

NIGEL T. BISHOP
Department of Mathematical Sciences, University of South Africa, Pretoria, 0003, South Africa
[email protected]

We describe the use of recently-obtained exact solutions of the linearized Einstein equa-
tions to test various versions of a characteristic code, thereby improving the code. We also
describe how the characteristic code could be used, in combination with an ADM-type
code, to obtain an accurate gravitational waveform.

Keywords: Numerical relativity; Bondi-Sachs metric

1. Introduction
Characteristic numerical relativity has many positive features - stability, conver-
gence, and inclusion of the whole spacetime. However, it has been unable to compute
gravitational radiation of, for example, a star in orbit around a black hole. Here,
we describe the use of recently-obtained exact solutions of the linearized Einstein
equations 1 to test various versions of a characteristic code, thereby leading to an
improved code. We also describe how the code could be used in combination with
an ADM-type code to obtain an accurate gravitational waveform.

2. Background material
The Bondi-Sachs metric uses coordinates based upon a family of outgoing null
hypersurfaces. Let u label these hypersurfaces, x A (A = 2,3), label the null rays
and r be a surface area coordinate. Then the metric takes the Bondi-Sachs form
ds 2 = - (e 2i3 (1 + Wcr) - r2hA B U A U B ) du 2
-2e 2i3 dudr - 2r2hA B U B dudxA + r 2hABdx A dx B , (1)
where det(hAB) = det(qAB), with qAB a unit sphere metric. We represent qAB
by means of a complex dyad qA, and then hAB by its complex dyad component
°
J = hABqAqB /2, with J = characterizing spherical symmetry. We introduce the
complex field U = U A qA , and the (complex differential) eth operators a and An a.
evolution problem is normally formulated as a solution to Einstein's equations in the
region of spacetime between a timelike or null worldtube r and future null infinity,
with (free) initial data J given on u = 0, and with boundary data for (3, U, We, J
satisfying the Einstein constraint equations given on r.
We use spin-weighted spherical harmonics sZ£m, where the suffix s denotes the
spin-weight, as basis functions. In the case s = 0, we omit the spin-weight index, i.e.
oZ£m = Z£m. The Z£m are the same as the familiar Yem except that they are real, i.e.
the ¢ dependence is cos m¢ or sin m¢ rather than e±i¢. The effect of the a operator
acting on Z£m is oZ£m = y'£(£ + 1) lZ£m, 0 2 Z£m = y'(£ - 1)£(£ + 1)(£ + 2) 2Z£m.

1630
 1631

3. Ansatz
Physically, we consider a thin, low-density, shell around empty space. The shell is
spherical at r =constant, but the density p can vary around the shell and with time.
The density and metric quantities are small, i.e. p, J,/3, U, We = 0(1':). We linearize
Einstein's equations, and use a standard separation of variables ansatz
J = V(£ - 1)£(£ + 1)(£ + 2) 2Zgm~(Jg(r)eivu),
U = V£(£+ 1) lZgm~(Ug(r)eiVU),
/3 = Zgm~(/3geivu), We = Zgm~(wg(r)eiVu). (2)
We solve (alllD) Einstein equations and obtain analytic formulas, in terms of simple
polynomials in r- 1 (up to order £ + 2), for Jg(r), Ug(r), Wce(r).l

4. The constraint equations

If the boundary data satisfies the constraints Roc< = 0 (here we restrict attention
to the vacuum case), and provided the hypersurface and evolution equations are
satisfied, the Bianchi identities guarantee that the constraints are satisfied through-
out the domain. In a numerical simulation constraint evaluation may provide useful
information concerning the reliability of the computation. We have written code
that evaluates the quantities Roo, ROl and qA ROA.

5. Comparative peformance of characteristic code

We have constructed versions of the characteristic code that handle the angular
coordinates in different ways, (a) stereographic, (b) six-patch 2nd order, and (c)
six-patch 4th order. The terms 2nd order and 4th order refer to the order of accuracy
of finite differencing of angular derivatives, and we have been able to implement 4th
order for the six-patch method because the ghost points are located on grid lines,
so only I-dimensional interpolation is needed. Detailed results of the comparison
are available elsewhere,2 and here we summarize the main conclusions.
All schemes give (approximate) 2nd order convergence of the metric quantities.
However, the news and the constraints (all of which contain 2nd derivatives of metric
quantities) exhibited, in some cases, degradation of the order of convergence. The
six-patch schemes were approximately 2nd order convergent throughout the run,
but the stereographic scheme exhibits convergence degradation that is periodic in
time. The analytic solution has period 27f, and it is interesting that degradation
peaks when the analytic solution is zero. The comparative behaviour of the error
norm is particularly interesting. On average, the error norm of 2nd order six-patch
was smaller than that of stereographic by a factor of order two (although there
were cases in which the error was slightly larger). However, the 4th order six-patch
scheme exhibited a dramatic reduction in the error norm, by a factor of up to 47
compared to that of the stereographic case.
We have also investigated how the resolution requirement (to be in a convergent
regime) scales with increasing £. The details are given elsewhere in this volume. 3
1632

6. Cauchy-characteristic extraction (CCE) and matching (CCM)

First, a worldtube is defined and the Cauchy metric data is used to generate charac-
teristic metric data on the worldtube. 4 Then the characteristic Einstein equations
are used to find metric data between the worldtube and null infinity, and compute
the radiation at null infinity. In CCE, a standard outer boundary condition for the
Cauchy evolution is imposed at some surface well outside the world tube; whereas in
CCM, the characteristic metric data is used to provide an outer boundary condition
for the Cauchy evolution.
Whereas CCM is clearly the desired goal, it is much more difficult than CCE
because the feedback may introduce an instability. CCM has been successfully imple-
mented for the nonlinear scalar wave equation, 5 and for linearized gravity. 6 However,
in the case of the full Einstein equations, to date only CCE has been implemented. 7
The important point about CCM, and under certain circumstances CCE, is that it
leads unambiguously to the gravitational radiation at future null infinity, and is the
most efficient way8 to find the waveform to high accuracy.

7. Conclusion
We have described the use of linearized eigen-solutions of the Einstein equations
to validate and develop a characteristic numerical code. We have also reviewd the
methods of Cauchy-characteristic matching and extraction, and shown that a suc-
cessful implementation is important because it can be expectedto lead to precise
statements about the error of radiation extraction at finite 'r.

Acknow ledgments
Parts of the work were done in collaboration with C.W. Lai, C. Reisswig, B. Szilagyi
and J. Thornburg. I thank Max-Planck-Institut fUr Gravitationsphysik for hospital-
ity. The work was supported by the National Research Foundation, South Africa,
under Grant number 2053724.

References
1. N. T. Bishop, Class. Quantum Gmv. 22, p. 2393 (2005).
2. C. Reisswig, N. T. Bishop, C. W. Lai, J. Thornburg and B. Szilagyi, (2006), www.
arxiv.org/abs/gr-qc/0610019.
3. C. W. Lai, In these proceedings.
4. N. T. Bishop, Class. Quantum Gmv. 10, p. 333 (1993).
5. N. T. Bishop, R. Gomez, P. R. Holvorcem, R. A. Matzner, P. Papadopoulos and J. Wini-
cour, Jnl. Compo Phys. 136, p. 140 (1997).
6. B. Szilagyi, (2000), www.arxiv.org/abs/gr-qc/0006091.
7. M. Babiuc, B. Szilagyi, 1. Hawke and Y. Zlochower, Class. Quantum Gmv. 22, p. 5089
(2005).
8. N. T. Bishop, R. Gomez, L. Lehner and J. Winicour, Phys. Rev. D 54, p. 6153 (1996).
 THE KERR METRIC IN BONDI-SACHS FORM*

LIEBRECHT R. VENTER and NIGEL T. BISHOP

Department of Mathematical Sciences, P.O. Box 392, Unisa 0003, RSA
[email protected] [email protected]

The most powerful source of gravitational radiation is the merger of a black hole with
another black hole (or other compact object such as a neutron star). Such processes
need to be calculated numerically, and the field of numerical relativity has been de-
veloped to tackle such problems. The most popular approach to numerical relativity is
based on the ADM formalism, and variations thereof. We however take the alternative
approach to numerical relativity based on an evolution of the metric variables in the
Bondi-Sachs form. To date, applications involving black holes have been restricted to
the Schwarzschild case, because of the lack of initial and boundary data for a rotating
black hole. We obtained a Kerr metric in Bondi-Sachs form which can provide such ini-
tial data. Our point of departure, was the Kerr geometry as obtained by Pretorius and
Israel. 3 We made coordinate transformations on this metric, to bring it into Bondi-Sachs
form. We confirmed elementary flatness of this metric, and we also confirmed that it is
asymptotic to the Bondi-Sachs form of the Schwarzschild geometry. The results obtained
in our paper l are needed so that numerical relativity codes based on the characteristic
formalism, can be applied to a situation that contains a rotating black hole. In order to
validate the metric, we evaluate it numerically on a regular grid of the new coordinates.
The Ricci tensor is then computed, for different discretizations, and was found to be
convergent to zero.

1. Transformation of the Kerr-Metric to Bondi-Sachs Form

Our transformation from standard Boyer-Lindquist coordinates x I3L (t,r,e,¢) to
Bondi-Sachs coordinates XES = (u, rs, e*, ¢*) is

u
2 2
=t_Jr +a d _1=r'2+ a2- Q (r',e*)d ' -
~(r) r r ~(r') r
e
}e,
p(e' e )de'
,* ,
r (1)

rs = y!2L cos e* sine, (2)

e* defined implicitly by
1 r
= dr'
Q(r',e*)
-
l()
e• de'
p(e',e*)
= 0
,
(3)

= 2mas
¢* = ¢+ 1
r ~(s)Q(s,e*)ds. (4)

We have used computer algebra to construct the Jacobian. The non-zero components
of the Jacobian are

1, j01 ~, j02 = -P, (5)

*The work was supported in part by the National Research Foundation, South Africa, under Grant
number 2053724.

1633
1634

J11 = - ~ 3/2 (2P 2 P?QCOS2 e* (a 2 COS2 e* (m - r) - 2r 3 - ra 2 - a2m)

4Q (L COSe*)

+ p 2Q2 (/-L,B, cot e* - /-L) + /-La2 COS 2 e* (Q2 - 2D.P 2) ), (6)

J12 = V23/2 (2/-LQ 2 cos 2 e* (a 2 sine + /-LPCOSe (p 2 - a2 sin 2 e))

4P (L cos e*) vsin e

- /-Lp 2a2 cos 2 e* sineD. + p 2Q2 sine (/-L,B, cote* - /-L)), (7)

where
8H) 3
H1 == ( 8e* r = ma sin(2e*)
100 Q(r',r'e*)3 dr'.
r
(9)

Note that the integral in Eq. (9) will need to be evaluated numerically. By substi-
tution of the transformation into the Kerr metric in Boyer-Lindquist coordinates,
we arrive at the contra-variant Bondi-Sachs metric representing the Kerr geometry.
01 j11 Q + j12 P 11 jf1 D. + jf2 12 h2Q - j11 P D.
gBS = - ~ ,gBS = ~ gBS = ~Lsin(2e*) ,

(10)

where

(11)

2. Numerical Evaluation
In order to obtain the Ricci tensor of the implicit metric numerically to O(h2), a
13-point regular grid 2 was used. We find metric first derivatives using the central
difference formula, then we calculate the Christoffel symbols, then we calculate first
derivatives of the Christoffel symbols using the central difference formula, and finally
we calculate the various components of the Ricci tensor. For given (r *, e*) we obtain
the Boyer-Lindquist coordinates (r, e) as follows. We make an estimate for r and use
Eq. (3) to find e. Once e has been obtained in this manner, (r, e, e*) is substituted
into the equation for the radial coordinate to obtain a value for r*. Of course, in
general this will not be the desired value of r*, so we repeat the calculation with a
 1635

different estimate of T, and then use the bisection method to find, to an accuracy
of 10- 12 the value of T which leads to the correct value of T*. Now, having obtained
(T,8) for given (T*,8*), we find all the metric coefficients, including L, at the point
(T*,8*).
The numerical code was validated by using it to find the Ricci tensor of the Kerr
metric in standard Boyer-Lindquist coordinates and the Fletcher-Lun 4 metric. Sec-
ond order convergence to zero was obtained in both cases. The Fletcher-Lun metric
is in Bondi-Sachs form, but not a viable candidate for numerical coding due to a co-
ordinate anomaly where the axis of symmetry is dependent on the radial coordinate.
During the procedure constructing our transformation, intermediate metrics were
obtained. To further validate our work, these metrics were tested for Ricci-flatness
over three discretizations namely (h = 0.01,0.02,0.04). What follows are brief data
extracts from the Fletcher-Lun [FL] test case at grid-point (m, a, T, 8)=(1, 0.1, 0.5,
0.9), and the implicit Pretorius-Israel [PI] and Bishop-Venter metrics [BV] at the
grid-point (m, a, T*, 8*)=(1, 0.1,0.5,0.9). The latter is equivalent to the metric (10)
except for a straight forward transformation to a radial surface coordinate.

3. Convergence of the Ricci Tensors

h = 0.01 h = 0.02 h = 0.04 n I Rc'/3 [Olll
FL .0000004274473 .0000017097918 .0000068391881 2.0 .0000000000012
PI .0000389173773 .0001556781866 .0006228653942 2.0 .0000000130977
BV .0000389167189 .0001556755506 .0006228548385 2.0 .0000000130975

As can be seen, the program shows O(h2) convergence to zero for the Ricci Tensors
for all three metrics, where n is the order of convergence, IIRo:/3[Olll the convergence
limit.

4. Conclusion
We constructed an implicit Kerr metric in Bondi-Sachs form which is Ricci-flat
and which can serve as initial data to evolve solutions describing massive rotating
black-hole mergers. We constructed programs to obtain the metric components of
this metric and to verify that the geometry is Ricci-flat. Our metric also satisfy
elementary flatness at the axis of symmetry and is asymptotic to the Bondi-Sachs
form of the Schwarzschild solution of which the details can be found in our paper. 1

References
1. N.T. Bishop and L.R. Venter, Phys. Review D. 73, 084023 (2006).
2. N.T. Bishop and L.R. Venter, arXiv Preprint, 0506077.
3. F. Pretorius and W. Israel, Class. Quantum Grav 15, 2289 (1998).
4. S.J. Fletcher and A.W.C. Lun, Class. Quantum Grav. 20, 4153 (2003).
 BINARY BLACK HOLE MERGER WAVEFORMS IN THE
EXTREME MASS RATIO LIMIT

THIBAULT DAMOUR
Institut des Hautes Etudes Scientifiques, 35 route de Chartres, 91440 Bures-sur- Yvette, France

ALESSANDRO NAGAR
Dipartimento di Fisica, Politecnico di Torino, and INFN, sezione di Torino, Torino, Italy

We discuss the transition from quasi-circular inspiral to plunge of a system of two non-
rotating black holes of masses ml and m2 in the extreme mass ratio limit mlm2 «
(ml + m2)2. In this limit, we compare the merger waveforms obtained by two differ-
ent methods: a numerical (Regge-Wheeler-Zerilli) one, and an analytical (Effective One
Body) one. This is viewed as a contribution to the matching between analytical and
numerical methods.

1. Introduction
The last months have witnessed a decisive advance in Numerical Relativity, with
different groups being able to simulate the merger of two black holes of compara-
ble masses. 1 Since such binary black holes systems (of a total mass~ 30M(')) are
believed to be among the most promising sources of gravitational waves for the
ground based detectors like LIGO and VIRGO, this breakthrough raises the hope
to have, for the first time, a reliable estimate of the complete waveform by join-
ing together Post-Newtonian (PN) and Numerical Relativity results. We recall that
PN techniques have provided us with high-order results for describing the motion
and radiation 2 of binary systems, and that further techniques have been proposed
for resumming the PN results,3,4 thereby allowing an analytical description of the
gravitational waveform emitted during the transition between inspiral and plunge,
and even during the subsequent merger and ringdown phases. We now face the
important task of constructing accurate complete waveforms by matching together
the information contained in Post-Newtonian and Numerical Relativity results. We
view the present work as a contribution towards this goal (see also Refs. 5 ,6).
The present work belongs to a scientific lineage which was started by Regge and
Wheeler,7 Zerilli,8 Davis, Ruffini, Press and Price 9 and Davis, Ruffini and Tiomno. lO
Refs. 9 ,10 studied the gravitational wave emission due to the radial plunge (from in-
finity) of a particle into a Schwarzschild black hole, as a model for the head-on
collision of two black holes in the extreme mass ratio limit. Here we consider, for
the first time, the transition from the quasi-circular adiabatic inspiral phase to the
plunge phase in extreme-mass-ratio binary black hole systems. We shall be able
to do that by getting round a limitation of the original Regge-Wheeler-Zerilli test-
particle approach: that of requiring the test particle to follow an exact geodesic
of the Schwarzschild background. We bypass this stumbling block by using an im-
proved form of PN theory: the Effective One Body (EOB) approach to the general
relativistic two body dynamics. This approach has been recently proposed to study

1636
 1637

221c_==:;z,,=:;rill~i_MC=o=","2.,,=====r-=--~j
025 20 ~ EOB-type QC approximation with F~2 Pade

0.2
I:l 18 "'"
OIS " "
lil\ I
I I
01 " :,::11 ,
:f
I I
"
0.05 I : I I I
" I

,.
I

2~ 0 ;:
"

I
I I
I
I I
I
I
I
I
I
I
r
I
I

,
I
I

I
" I I
-0.05 : ~ t I I
I I
I I
I I I
I

',~
-{I.I "

\(
uI(2M)
I

-015
"
"
I

\1
::t"
-{J2

-0.25

50 100 150 200 250 3(1) 350 400 100 150 200 250 300 350
U/(2M) U/(2M)

Fig. 1. Left panel: Gravitational waveforms (real and imaginary part) for e = m = 2 generated
by a plunge from an initial separation r = 7 M. Right panel: Comparison between the corre-
sponding gravitational-wave phase and that obtained from analytically matching a 3PN improved
quadrupole-type formula to a superposition of quasi-normal modes.

the transition from inspiral to plunge in the comparable-mass case. 4 It describes the
dynamics of a binary system in terms of two separate ingredients: (i) a Hamilto-
nian HEoB(M, J.L) describing the conservative part of the relative dynamics, and (ii)
a non-Hamiltonian supplementary force FEOB(M, J.L) approximately describing the
reaction to the loss of energy and angular momentum along quasi-circular orbits.
[Here, M == ml + m2, and J.L = mlm2/M.] The badly convergent PN-Taylor series
giving the angular momentum flux is resummed by means of Pade approximants. 3

2. Results
We summarize here our tools and main results (see Refs. 12 ,13 for more informa-
tion). We describe, in the extreme mass ratio limit, the relative dynamics of the
binary system of two non-rotating black holes by that of a "particle" (of mass
J.L « M) that follows a non-geodesic inspiral driven by radiation-reaction (until a
quasi-geodesic plunge) on a (quasi-)Schwarzschild background of mass M. We then
use the multipolar Regge-Wheeler-Zerilli perturbation theory around such a black
hole to compute the gravitational wave emission in the approximation J.L « M.
Einstein equations then lead to two (even/odd) decoupled wave-like equations (with
source terms S~:!,o) IX J.L) for two (even/odd) master functions \[!~:!,o).u
Fig. 1 shows the f! = m = 2 waveform generated by a binary system with initial
relative separation r = 7M a. After a "chirp-like" increase of the frequency and
modulus during the inspiral, the latter reaches a maximum when the particle crosses
the light ring (r = 3M) and eventually the waveform decays in a quasi-normal mode
ringing phase.

aNote that the other multipoles give a non-neglible contribution, due to the strong asymmetry of
the system.
1638

The so-constructed "exact" numerical waveforms are then compared with semi-
analytical ones, constructed within the EOB framework and philosophy. We recall
that the basic idea of the EOB framework is to produce quasi-analytical waveforms
by patching together a quadrupole-type waveform during the inspiral and plunge
to a QNM-type waveform after merger. We give an example of this numerical-
analytical comparison in the right panel of Fig. 1. The gravitational wave phases
¢~~ obtained by two different methods, one numerical and the other semi-analytical,
are compared. In the first method, ¢~~ is given from time-integration of the instan-
taneous gravitational wave frequency obtained as Mw~~ = -'S(W~~ /1}f~~). The
second method computes an approximate waveform in the following way: before
crossing the light ring one uses a (Pade resummed) 3PN-improved quadrupole-type
formula to compute the waveform from the EOB dynamics. After crossing the light
ring, the previous quadrupole-type signal, taken in the quasi-circular (QC) approx-
imation, is matched to a superposition of the first five QNMs of the black hole.
Then one computes the phase of the matched analytical waveform by integrating
the corresponding instantaneous gravitational wave frequency. The maximum dif-
ference between the "exact" phase and the "effective one body" phase turns out to
be less than 3% of a cycle (see inset).

References
l. F. Pretorius, Phys. Rev. Lett. 95, 121101 (2005); M. Campanelli, C.O. Lousto, P. Mar-
ronetti, and Y. Zlochower, Phys. Rev. Lett. 96 111101 (2006); J.G. Baker, J. Centrella,
D.l. Choi, M. Koppitz, and J. van Meter Phys. Rev. Lett. 96 111102 (2006); M. Cam-
panelli, C.O. Lousto, and Y. Zlochower, Phys. Rev. D 74, 084023 (2006); J. Gonzalez,
U. Sperhake, B. Bruegmann, M. Hannam, S. Husa, arXiv:gr-qc/061 0154.
2. T. Damour, P. Jaranowski and G. Schafer, Phys. Lett. B 513, 147 (2001); L. Blanchet,
T. Damour, and G. Esposito-Farese, Phys. Rev. D 69, 124007 (2004); L. Blanchet,
T. Damour, G. Esposito-Farese and B.R. Iyer, Phys. Rev. D 71, 124004 (2005);
L. Blanchet, Living Rev. Relativity 9,4. URL: http://www.livingreviews.org/lrr-2006-
4 (2006).
3. T. Damour, B.R. Iyer and B.S. Sathyaprakash, Phys. Rev. D, 57, 885 (1998).
4. A. Buonanno and T. Damour, Phys. Rev. D 59, 084006 (1999); ibid., 62, 064015
(2000).
5. A. Buonanno, G.B. Cook, and F. Pretorius, arXiv:gr-qc/0610122 (2006).
6. J. G. Baker, J. R. van Meter, S. T. McWilliams, J. Centrella and B. J. Kelly, arXiv:gr-
qc/0612024·
7. T. Regge and J.A. Wheeler, Phys. Rev. 108, 1063 (1957).
8. F.J. Zerilli, Phys. Rev. Lett. 24, 737 (1970).
9. M. Davis, R. Ruffini, W.H. Press, and R.H. Price, Phys. Rev. Lett. 27, 1466 (1971).
10. M. Davis, R. Ruffini, and J. Tiomno, Phys. Rev. D 5, 2932 (1972).
1l. A. Nagar and L. Rezzolla, Class. Q. Grav. 22, R167 (2005); erratum, 23, 4297 (2006).
12. A. Nagar, T. Damour and A. Tartaglia, in Class. Q. Grav. special issue based on New
Frontiers in Numerical Relativity conference, Golm (Germany), July 17-22 2006.
13. T. Damour and A. Nagar, in preparation.
CMB Theory
This page intentionally left blank
 CMB ANOMALIES FROM RELIC ANISOTROPY*

A. EMIR GUMRUKC;UOGLU
School of Physics and Astronomy, University of Minnesota, Minneapolis, MN 55455, USA

CARLO R. CONTALDI
Theoretical Physics, Blackett Laboratory, Imperial College, London, SW'l 2BZ, UK

MARCO PELOSO
School of Physics and Astronomy, University of Minnesota, Minneapolis, MN 55455, USA

Most of the analysis of the Cosmic Microwave Background relies on the assumption of
statistical isotropy. However, given some recent evidence pointing against isotropy, as
for instance the observed alignment of different multipoles on large scales, it is worth
testing this assumption against the increasing amount of available data. As a pivot model,
we assume that the spectrum of the primordial perturbations depends also on their
directionality (rather than just on the magnitude of their momentum, as in the standard
case). We explicitly compute the correlation matrix for the temperature anisotropies in
the simpler case in which there is a residual isotropy between two spatial directions. As
a concrete example, we consider a different initial expansion rate along one direction,
and the following isotropization which takes place during inflation. Depending on the
amount of inflation, this can lead to broken statistical isotropy on the largest observable
scales.

1. Introduction and discussion

The release of the WMAP 3-year results 1 has now placed our knowledge of the
largest scales of the CMB on a firm footing. An intriguing aspect of the maps
produced by WMAP is the presence of a number of seemingly related anomalies
on the largest scales. They comprise the lack of power in the lowest multipoles,
the alignment of the power in a number of the lowest multipoles and an apparent
asymmetry in the maps. The lack of power and its possible origin has been addressed
copiously in the literature. 2 ,4,5 The difficulty of calculating the posterior probability
of the measured values of the quadrupole and octupole has led to a spread in the
claims of the significance of the effect which now appears to be marginal. 5
The significance of the anomalies which represent a breaking of the statistical
isotropy of the maps appears much stronger. 6 ,7 Indeed, if the signal is statistically
anisotropic then much of the information content of the maps lies in the anisotropic
correlation of the spherical harmonic modes with (aRrnae'rn') rj::. OWornrn'. Obvious
explanations for the observed anomalies are that an unknown systematic is not
being accounted for in the analysis or that foreground signals, which are naturally
anisotropic in nature, are not being modeled and subtracted properly. However a
conclusive explanation along these lines has not been put forward yet.
Another, albeit more speculative, possibility is that the explanation lies in more

* Talk given by Carlo Contaldi .

1641
1642

fundamental aspects of the cosmological model. It seems advantageous therefore to

attempt to model the possible anisotropies in the sky from first principles from a
primordial perspective with a view to constrain departures from the standard cos-
mological picture. Here we assume that the broken statistical isotropy is imprinted
in the primordial spectrum Pip (k) of the cosmological perturbations, leading to a
specific and falsifiable pattern for the covariance matrix of the CMB fluctuations a.
The present discussion is organized as follows. We first compute the general cor-
relations that are obtained in this case. We then present a simple model which gives
rise to the above correlations. The background evolution of this model is axisymmet-
ric, with an expansion rate which is initially much faster along a single direction.
For simplicity, we assume isotropy in the remaining two directions. A period of
inflation than leads to full isotropy. Such a situation may emerge in models with
extra dimensions (as for instance in string theory). In this case, one faces the diffi-
cult problem to explain why only three directions undergo cosmological expansion,
while the other remain stabilized on a small size. s It is plausible to imagine that
the expansion of the three (now) large dimensions did not start precisely with the
same rate, but that isotropy was only reached afterwards in the first fewe-foldings
of inflation. With this approach, matter and radiation perturbations propagate in
an isotropically expanding background. The breaking of the statistical isotropy is
therefore entirely due to the initial conditions of the model (as opposed to a full
anisotropic evolution of perturbations9 ,1O). We conclude by showing the spectrum
of perturbations of scalar modes having the momentum in the anisotropic direction
(as we discuss, the computation is considerably simpler in this case).
Details of the computations, and the comparison with the observations, will be
presented elsewhere. 11

2. CMB covariance matrix

We want to calculate

CU'mm' == (a€maC'm') = J dOpdOp,(oT(p, 1]0, xo)oT(p', 1]0, xO))Y£~(P)Y£'m'(P')'

(1)
where oT is the temperature perturbation in the direction p, as measured by an
observer at position Xo and at time 1]0; it can be expressed as

OT(p, 1]0, xo) = J(~:~3 <I>(k, 1]i)b.(k, k· p, 1]o)ei k-x o , (2)

We assume that the anisotropy is only imprinted in the primordial curvature

power spectrum. Specifically, (27r/k)3 yf - k')Pip(k) == (<I>(k)<I>*(k)) depends also
o
aThis is in contrast to the opposite approach of constraining general modifications to the covari-
ance l ,7 such as (a cm a;'m') ex 6CC'6 mm , (Cc + ECm)·
 1643

on the orientation of the wavevectors, and not only on the magnitude. In con-
trast, the calculation of the later time transfer function tle (k, Tla) for the radiation
perturbation proceeds as the standard case by the line of sight integration of the
Einstein-Boltzmann system. Note that the transfer functions are isotropic in that
they only depend on the magnitude of the wavevector k. However the final integra-
tion over wavevectors k must be modified due to the directionality dependence of
Pcp (k).
For definiteness, we assume a residual symmetry along two directions. We denote
by ~ the cosine of the angle between the wavevector and the axis of symmetry.
Working the expression (1) out, we find

6mm , . c-c' (21' + 1)(21" + 1)(1' - m)!(1" - m)!

GCf'rnm' = -- (-z)
1f (1' + m)!(1" + m)!

x Jd: tlc (k, Tlo) tl£ (k, Tla) [11 d~pr (~) PP (0 Pcp (k, 0 (3)

In the isotropic case, Pcp (k, ~) == Pcp (k) , we recover the standard result

(4)

In general, we obtain a correlation between different multiples. Here, we focus on a

axisymmetric model, for which Pcp(k, -~) = Pcp(k,~). In this case, the correlation
is nonvanishing whenever the difference between [ and [' is even (as it follows from
the parity properties of of the associated legendre polynomials).

3. Background evolution
As a specific example, we consider a homogeneous but anisotropic background, with
an expansion rate along the x-direction different from the other two,
ds 2 = -dt 2 + a (t)2 dx 2 + b (t)2 (dy2 + dz 2) (5)
A scalar field in this background leads to the equations
if + 3H 2 = V,
~ + 3 H ¢ + V' = 0 ,
3H 2 - h 2 = ~i2 + V
2'1-' , (6)

where H == (Ha + 2 H b ) /3 and h == (Ha - H b ) / v'3 (we work in units of Mp = 1,

while prime denotes differentiation with respect to rP). The first two equations are
identical to the ones obtained in the isotropic case, in terms of the "average" expan-
sion rate H. The third equation, appearing as a "modified" Friedmann equation,
can then be used as an algebraic equation for the difference h between the two
expansion rates.
We consider a scalar field potential, and an initial scalar field value capable of
. 2
sustaining inflation, and a large initial anisotropy, h5 » Va »rPo . The system
1644

20 ~~~----~----~----~--~----~----~----,

15

'"
Q)
iii
a:
Q) 10
:0
..Q
::::J
I
.... ------------------

5 / / /....

/"
o ~/____~____ L __ _~_ _ _ __ L_ _ _ __ L_ _ _ _~_ _ _ _L __ _~

o 0.5 1.5 2 2.5 3 3.5 4

t [(3NO)1f2]

Fig. 1. Evolution of the Hubble rates and the difference in e-foldings during the anisotropic
and the following isotropic inflationary stages. We have chosen a chaotic inflationary model, with
quadratic potential V = m~<p2 /2 with initial conditions <Po = 16, ho = 104 m¢. This leads to
about 60 e-foldings during the isotropic inflationary stage.

then undergoes a quick anisotropic inflationary stage (with both a, b> 0), during
which ¢ is practically static, i.e. V c:::: Va, and the two Hubble parameters behave
as Ha c:::: lit and Hb c:::: O. To follow the evolution of h, it is useful to combine
eqs. (6) into h + 3 H h = O. This leads to a very rapid decrease of h during the
anisotropic stage, as the average rate approaches the standard (slow roll) value
H c:::: JVoI3. The timescale for the decrease is tiso == J31Vo (we can also see this
by requiring the continuity of Ha between the early lit behaviour and the later
constant value). From this moment on, h can be neglected, and we have a second
stage of standard (isotropic) inflation. Figure 1 shows, for a specific choice of initial
conditions, the evolution of the Hubble parameters along with the difference in the
number of e-foldings between the expansions in the x and y, and z axes, defined as
i1N == J~ dt (Ha - Hb). This increases during the first stage and remains constant
after tiso'

4. Perturbations
The standard FRW background is symmetric under spatial transformations. It is
therefore convenient to classify the perturbations according to how they transform
under spatial change of coordinates: modes which transform differently are not
coupled at the linearized level. One then finds that 3 physical modes are present.
They are contained in one scalar perturbation, and in the two polarizations of a
tensor mode, which are also decoupled from each other. Also in the present context
there are three physical modes. Two of them are coupled to each other, while the
third one is decoupled, due to the residual symmetry in the y - z plane. The first
 1645

Longitudinal - -
10

0.1

0.01

0.Q1 0.1 10 100 1000

klk iso

Fig. 2. Power spectrum for longitudinal scalar perturbations. The initial conditions for the back-
ground are as in figure 1, while ¢ = 3 at the moment shown here.

two modes decouple when the universe become isotropic, and they become a scalar
and a tensor polarization.
Therefore, we expect that the coupling is proportional to the difference between
the two Hubble rates. We also expect that the coupling vanishes when the momen-
tum of the modes is along the privileged direction x. We denote these modes as
longitudinal. Indeed, the directionality of a mode always breaks the full 3d isotropy;
when the momentum is oriented along the anisotropic direction x we have the same
symmetries as in the standard case. Therefore, the computation is much simpler for
these modes, that we denote these modes as longitudinal. Here, we focus only on
this simpler case. The complete computation is given in.ll
We proceed as in the standard case, by writing down the most general scalar
perturbations of the metric and of the inflaton field, and by choosing a gauge which
completely fixes the freedom of coordinates transformations (see l2 for a review).
In the longitudinal case, the linearized Einstein equations for the perturbations
can be reduced to a unique equation in terms of a single variable Q, which is the
generalization of the Mukhanov-Sasaki variable,13 and which obeys the equation

Q+(Ha+2Hb)Q+ [:~ +M2] Q=O

M == V" + 2¢V'/Hb - ¢4/ (2H;) + 2Ha¢2/Hb + ¢2 (7)

where k is the comoving momentum, oriented in the x direction. This equation

coincides with the standard one when Ha = Hb = H. We note that the scale
factor a is entering in the physical momentum k / a, since the mode is moving along
the x-direction. The initial conditions are given as in the standard case, since k / a
dominates the last term of eq. (7) at early times, leading to a standard adiabatic
initial vacuum.
1646

In figure 2, we show the power spectrum towards the end of inflation. k iso is
the comoving momentum of the modes which left the horizon when the universe
became isotropic. Modes with k > k iso leave the horizon during the later isotropic
stage of inflation. The standard result is recovered for this modes, with the typical
spectral index of massive chaotic inflation. However, modes with smaller momenta
left the horizon when the universe was still anisotropic, and a nonstandard result is
found in that case.

Acknowledgments
We thank Joao Magueijo for useful discussions. The work of A. E. G. was supported
by the Hoff Lu Fellowship in Physics at the University of Minnesota. The work of
M. P. was supported in part by the Department of Energy under contract DE-FG02-
94ER40823, and by a grant from the Office of the Dean of the Graduate School of
the University of Minnesota.

References
1. Spergel, D. N., et a!. 2006, arXiv:astro-phj0603449
2. Bielewicz, P., Gorski, K. M., & Banday, A. J. 2004, Mon. Not. Roy. Astron. Soc., 355,
1283
3. Bridle, S. L., Lewis, A. M., Weller, J., & Efstathiou, G. 2003, Mon. Not. Roy. Astron.
Soc., 342, L72; Schwarz, D. J., Starkman, G. D., Huterer, D., & Copi, C. J. 2004,
Physical Review Letters, 93, 221301; Slosar, A., Seljak, U., & Makarov, A. 2004,
Phys. Rev. D, 69, 123003.
4. Contaldi, C. R, Peloso, M., Kofman, L., & Linde, A. 2003, JCAP 0307, 002 (2003).
5. Efstathiou, G. 2004, Mon. Not. Roy. Astron. Soc., 348, 885; de Oliveira-Costa, A.,
Tegmark, M., Zaldarriaga, M., & Hamilton, A. 2004, Phys. Rev. D, 69, 063516.
6. Copi, C. J., Huterer, D., Schwarz, D. J., & Starkman, G. D. 2006, arXiv:astro-
phj0605135; Eriksen, H. K., Hansen, F. K., Banday, A. J., Gorski, K. M., & Lilje,
P. B. 2004, Astrophys. Journal, 609, 1198; Eriksen, H. K., Banday, A. J., Gorski,
K. M., & Lilje, P. B. 2005, Astrophys. Journal, 622, 58; Hansen, F. K., Banday, A. J.,
& Gorski, K. M. 2004, Mon. Not. Roy. Astron. Soc., 354, 641; Hansen, F. K., Cabella,
P., Marinucci, D., & Vittorio, N. 2004, ApJL, 607, L67; Jaffe, T. R, Banday, A. J.,
Eriksen, H. K., Gorski, K. M., & Hansen, F. K. 2005, ApJL, 629, L1; Land, K., &
Magueijo, J. 2005, Physical Review Letters, 95, 071301; Land, K., & Magueijo, J.
2005, Mon. Not. Roy. Astron. Soc., 357, 994; Land, K., & Magueijo, J. 2006, Mon.
Not. Roy. Astron. Soc., 367,1714; Ralston, J. P., & Jain, P. 2004, International Jour-
nal of Modern Physics D, 13, 1857; Vielva, P., Martinez-Gonzalez, E., Barreiro, R. B.,
Sanz, J. L., & Cayon, L. 2004, Astrophys. Journal, 609, 22.
7. Magueijo, J., & Sorkin, R D. 2006, arXiv:astro-phj060441O.
8. R. H. Brandenberger and C. Vafa, Nuc!. Phys. B 316, 391 (1989).
9. Bunn, E. F., Ferreira, P. G., & Silk, J. 1996, Physical Review Letters, 77, 2883
10. Gordon, C., Hu, W., Huterer, D., & Crawford, T. 2005, arXiv:astro-phj0509301
11. Contaldi, Giimriik!;iioglu, Peloso, work in progress.
12. V. F. Mukhanov, H. A. Feldman and R H. Brandenberger, Phys. Rept. 215, 203
(1992).
13. V. F. Mukhanov, JETP Lett. 41, 493 (1985) [Pisma Zh. Eksp. Teor. Fiz. 41, 402
(1985)]; M. Sasaki, Prog. Theor. Phys. 76, 1036 (1986).
 CAN EXTRAGALACTIC FOREGROUNDS EXPLAIN THE
LARGE~ANGLE CMB ANOMALIES?

ALEKSANDAR RAKIC 1, SYKSY RAsANEN 2 and DOMINIK J. SCHWARZ 1

1 Fakultiit fur Physik, Universitiit Bielefeld, Postfach 100131, D-33501 Bielefeld Germany
2 CERN Physics Department Theory Unit, CH-1211 Geneva 23, Switzerland
email: rakic at physik dot uni-bielefeld dot de, syksy dot rasanen at iki dot ft.,
dschwarz at physik dot uni-bielefeld dot de

We address the effect of an extended local foreground on the low-/' anomalies found in the
CMB. Recent X-ray catalogues point us to the existence of very massive superstructures
at the 100 h -1 Mpc scale that contribute significantly to the dipole velocity profile. Being
highly non-linear, these structures provide us a natural candidate to leave an imprint
on the CMB sky via a local Rees-Sciama effect. We show that the Rees-Sciama effect
of local foregrounds can induce CMB anisotropy of b.T IT ~ 10- 5 and we analyse its
impact on multi pole power as well as the induced phase pattern on largest angular scales.

1. Motivation and Overview

At largest angular scales which correspond to small multipole moments j! there exist
puzzling features in the Cosmic Microwave Background (CMB). The near vanishing
of the two-point angular correlation function in all wavebands for angular scales
between 60° and 170° is one of the longest known anomalies, already detected in
the data of the Cosmic Background Explorer's Differential Radiometer (COBE-
DMR). It has been confirmed and persists in the three-year Wilkinson Microwave
Anisotropy Probe data [WMAP(3yr)].1,2 Among the two-point angular correlation
functions it has been shown that none of the almost vanishing cut-sky wavebands
matches the full sky and again neither one of these is in accordance with the best
fit A cold dark matter (ACDM) mode1. 3 The disagreement turned out to be even
more distinctive in the WMAP(3yr) data than in WMAP(lyr) and is unexpected
at 99.97% C.L. for the updated Internal Linear Combination map [ILC(3yr)].3
Besides the lack of power, there are a number of remarkable anomalies regarding
the phase relationships of the quadrupole and octopole within the WMAP data. 4 - 6
In order to be able to make distinct statements with respect to a phase analysis of
multipoles we make use of the multipole vectors formalism. 7 Looking at quadrupole
plus octopole vectors from WMAP(3yr) the alignment with the equinox (EQX) and
with the ecliptic is found to be unlikely at 99.8% C.L. and 96% C.L. respectively. 3
The correlation with the dipole direction and with the galactic plane is found to
be odd at 99.7% C.L. and 99% C.L. respectively. Moreover from the combined full
sky map of i! = 2 + 3 one infers that the octopole is quite planar and that the
ecliptic strongly follows a zero line of the map, leaving the two strongest extrema
in the southern hemisphere and the two weakest in the northern hemisphere. Some
of these effects are statistically dependent, e.g. given the observed quadrupole-
octopole alignment, the significance of alignment with the galactic plane is reduced
to unremarkable 88% C.L.
These findings support the conclusion that either the Universe as seen by WMAP

1647
1648

is not statistically isotropic on largest scales, or that the observed features are due to
unexpected foregrounds, hidden systematics or new physics challenging the standard
cosmological model. Diverse attempts for explanation can be found in the litera-
ture: considering anisotropic or inhomogeneous models [Bianchi family, Lemaitre~
Tolman~Bondi (LTB) models],8-13 Solar system foreground,14 lensing of the CMB 15
and moving foregrounds,16 Sunyaev~Zel'dovich (SZ) effect 17 ,18 and Rees-Sciama
(RS) effect,12,19 considering a non~trivial topology of the Universe,20,21 considering
modifications and refinements of the standard simplest scenario of inflation,22-28
considering possible phenomenology of loop quantum gravity.29,30 In this talk we
update and expand our previous work 12 in the light of the WMAP(3yr) data release.

2. Local Structures and Rees-Sciama Effect

Recent X-ray catalogues of our neighborhood show that a major contribution to
the dipole velocity profile originates from the Shapley Supercluster (SSC) and other
density concentrations at a distance of of around 130~ 180 h~l MpC. 31 -34 The SSC is
a massive concentration centered around the object A3558 with a density contrast
of b ~ 5 over the inner 30 h ~ 1Mpc region. 35
We will show that the CMB displays correlations between the dipole and higher
mUltipoles after passing through non~linear structures, due to the RS effect. 36 The
physics of the RS effect is that in the non~linear regime of structure formation, the
gravitational potential changes with time, so photons climb out of a slightly different
potential well than the one they fell into. Following ref. 37 the CMB anisotropy pro-
duced by a spherical superstructure is estimated by the integral of the gravitational
potential perturbation ¢ ~ bMI d along the path of the photon: 6.T(e, rp )IT ~ ¢ Ve ,
where d is the physical size of the structure and bM is the mass excess. Here we
assumed a structure collapsing at velocity Ve and let the evolution time of the
structure te be the matter crossing time diVe (using c == 1 == G). We estimate
the typical collapse velocity from the energy balance condition v~ ~ ¢ and get:
6.T(e,rp)IT rv ¢3/2 rv (bMld)3/2. We model the non~linear structure by a spher-
ically symmetric LTB model embedded in a flat (0 = 1) Friedmann~Robertson~
Walker Universe. Substituting the expression for the mass excess within this model
we arrive at: 37

(1)

where t is the cosmic time at which the CMB photons crossed the structure.
Inserting the characteristics of the SSC it follows that a CMB anisotropy of 10~5
due to a local RS effect is reasonable. For simplicity we picture the local Universe
as a spherically symmetric density distribution, with the Local Group (LG) falling
towards the core of the overdensity at the centre. The line between our location and
the centre defines a preferred direction Z, which in the present case corresponds
to the direction of the dipole. This setup exhibits rotational symmetry w.r.t. the
z
axis (neglecting transverse components of our motion). Consequently, only zonal
 1649

0.1 0.1(; 0.2

C2 in [0.1 mK]2

Fig. 1. Likelihood of quadrupole and octopole power for increased axial contributions. Vertical
lines denote experimental data: WMAP(lyr) cut-sky and WMAP(3yr) maximum likelihood es-
timate. Considering the quadrupole adding any multipole power was excluded at > 99% C.L.
w.r.t WMAP(lyr) but it is possible to add up to 60jlK within the same exclusion level w.r.t. the
WMAP(3yr) value. The octopole is more resistant against axial contaminations as it is possible
to add a whole lOOjlK before reaching the same exclusion level w.r.t the updated WMAP data.

harmonics (rn 0 in the i-frame) are generated. Note that any other effect with
axial symmetry would also induce anisotropy only in the zonal harmonics.

3. Multipole Analysis
We study how maps of the CMB are affected by the anisotropy induced by additional
axisymmetric contributions affiial added to the quadrupole and octopole by using
Monte Carlo (MC) methods. As predicted by the simplest inflationary models, we
assume that the aRm are fully characterised only by angular power, for which we
use the values from the best fit ACDM temperature spectrum to the WMAP data. 2
We produced 105 MC realisations of £ = 2 and £ 3 for the statistical analysis.
The angular power spectrum is estimated by Ce = 1/(2£ + l):Z::m lac-rnI 2 • In fig.
1 we show how the histograms for the quadrupole and octopole power compare with
the measured values from WMAP(lyr,3yr). Considering the WMAP(lyr) cut--sky,
adding any power to the quadrupole was already excluded at > 99% C.L. whereas
the WMAP(3yr) data allows for adding up to a~oial = 60flK in order to reach the
same exclusion level. The octopole is quite robust against axial contaminations as
it lies better on the fit: in order to reach the same exclusion level of > 99% C.L.
it is necessary to add a30ial 80flK w.r.t. the WMAP(lyr) cut-sky and a whole
a3Qial = lOOflK w.r.t. the WMAP(3yr) value. Considering only the WMAP(3yr)
maximum likelihood estimate and increasing the effect of local structures up to
aeoial 70flK leads to an exclusion of 99.5% C.L. for C 2 and 92.9% C.L. for C 3 .
The next question is what kind of phase pattern the contribution aeo ia1 will
induce on the CMB sky. Using the multipole vector formalism 7 a (temperature)
multipole on a sphere can be alternatively decomposed as:
e
Te = aCmYem(f}, <p) (2)
m=-C
1650

where e(B,ip) = (sinBcosip,sinBsinip,cosB) is a radial unit vector. With the de-

composition (2) it is possible to obtain an unique factorisation of a multipole into a
scalar part A(e) which measures its total power and £ unit vectors v(e,i) that contain
all the directional information. The signs of the multipole vectors can be absorbed
into the scalar quantity A (e), and are thus unphysical.
Introducing the £(£ - 1)/2 oriented areas n(e;i,j) == vee,;) x v(e,j) Ilv(R,;) x v(R,j)1
we are ready to define a statistic in order to probe alignment of the normals n(R;;,j)
with a given physical direction x:

Snx == ~ L L In(e;i,j). xl· (3)

R=2,3 i<j

We test for alignment with three natural directions x: the north ecliptic pole
(NEP), EQX and the north galactic pole (NGP). The results of the correlation
z
analysis are shown in fig. 2: in the first row the preferred direction coincides with
the direction of local motion, the dipole. 38 Here the anomaly becomes worse when
increasing the amplitude of the axial contribution. But for x = NEP the exclusion
becomes somewhat milder; e.g. aR~ial = 40tLK leads to an exclusion of 99.2% C.L.
for ILC(lyr) but only 98.2% C.L. for the updated ILC map. Finding an alignment
with the EQX though is strongly excluded at > 99.2%C.L. even with a vanishing
axial contribution for both one- and three-year data.
In the second row of fig. 2 we let the preferred direction point to the NEP as
a complementary test. Here the probability to find an ecliptic alignment becomes
dramatically increased: with a aR~ial = 70tLK it is 17% and 10% for the ILC(3yr)
and ILC(lyr) values respectively. Regarding the three-year data the probability
for finding an EQX alignment increases from 1% to 3% for aR~ial = 70tLK. The
alignment with the NGP remains quite stable for both tested directions of z.

4. Conclusion
Recent astrophysical data cataloguing our neighborhood in the X-ray band point
us to the existence of massive non-linear structures like the SSC at distances
1
rv 100 h- Mpc. Besides its significant contribution to the dipole velocity profile

such a structure is able to induce anisotropies rv 10- 5 via its RS effect. Regarding
CMB modes, the spherical symmetry (LTB) which we use to approximate the local
superstructure reduces to an axial symmetry along the line connecting our position
and the centre of the superstructure where we locate the SSC. We produced statisti-
cally isotropic and gaussian MC maps of the CMB and computed their S-statistics
(3) for alignment with generic astrophysical directions like the NEP, EQX and NGP.
The additional zonal harmonics have been added with increasing strength (see ref.39
for full-sky maps). When gauging the preferred axis to the direction of local mo-
tion (WMAP dipole) the consistency of the data with theory becomes even worse,
albeit with less significance w.r.t. WMAP(3yr). On the other hand an orthogonally
directed (Solar system) effect would be more consistent with the three-year data.
 1651

Fig. 2. WMAP one- and three-year ILC maps compared to the alignment (3) of quadrupole
and octopole normals with physical directions (NEP, EQX, NGP in columns) for two orthogonal
realisations of the preferred directionz (WMAP dipole, NEP in rows). The bold histograms
represent statistically isotropic and gaussian skies. Increasing the axial contribution makes the
anomalies worse for z = WMAP dipole, but with the exclusions being less significant for the
ILC(3yr) than for the ILC(lyr). At the same time a Solar system effect is preferred by the data.

Considering extended local foregrounds Abramo et al. 17 recently proposed that

a cold spot in the direction of the local Supercluster could account for the cross
alignments of quadrupole and octopole. The cold spot would be realised by the SZ
effect of CMB photons scattering of the hot intracluster gas. On the other hand
Inoue and Silk19 suggest a certain geometrical pattern of two identical voids to
account for the cross alignment as well as for the octopole planarity via the RS
effect. Each of the latter approaches alone is not fully satisfactory. Nevertheless a
combined approach enfolding the RS effect as well as the SZ effect from extended
foregrounds seems promising for the future. Moreover, since the local RS effect can
contribute up to 10- 5 to the temperature anisotropies on large angular scales, a
detailed study is important for cross--correlating CMB data (including upcoming
Planck data) with astrophysical observations on the locallarge---scale structure.

Acknowledgment
It is a pleasure to thank the organisers of the 11th Marcel Grossmann meeting for their
effort and the opportunity to speak. We acknowledge the use of the Legacy Archive for
Microwave Background Data Analysis (LAMBDA) provided by the NASA Office of Space
Science. The work of AR is supported by the DFG grant GRK 881.

References
1. G. Hinshawet al., astro-ph!0603451.
2. WMAP data products at http://lambda . gsf c . nasa. gov /
3. C. Copi, D. Huterer, D. J. Schwarz and G. Starkman, astro-ph!0605135.
4. D. J. Schwarz, G. D. Starkman, D. Huterer and C. J. Copi, PRL 93, 221301 (2004).
1652

5. H. K. Eriksen, F. K. Hansen, A. J. Banday, K. M. Gorski and P. B. Lilje, ApJ 605,

14 (2004); (Erratum) 609, 1198 (2004).
6. A. de Oliveira-Costa, M. Tegmark, M. Zaldarriaga and M. Hamilton, Phys. Rev. D
69,063516 (2004);A. de Oliveira-Costa and M. Tegmark, astro-ph/0603369.
7. C. J. Copi, D. Huterer and G. D. Starkman, Phys. Rev. D 70, 043515 (2004).
8. T. Ghosh, A. Hajian and T. Souradeep, astro-ph/0604279.
9. T. R. Jaffe, A. J. Banday, H. K. Eriksen, K. M. Gorski and F. K. Hansen, astro-
ph/0606046 and references therein.
10. H. Alnes and M. Amarzguioui, astro-ph/0607334.
11. J. W. Moffat JCAP 0510, 012 (2005).
12. A. Raki6, S. Riisiinen and D. J. Schwarz, MNRAS 369, L27 (2006).
13. K. Tomita Phys. Rev. D 72, 043526 (2005), Phys. Rev. D 72 103506; (Erratum) D
73029901.
14. P. C. Frisch, ApJ 632, L143 (2005).
15. C. Vale, astro-ph/0509039.
16. A. Cooray and N. Seto, JCAP 0512, 004 (2005).
17. L. R. Abramo and L. Sodre Jr., astro-ph/0312124;
L. R. Abramo, L. Sodre Jr. and C. A. Wuensche, astro-ph/0605269.
18. F. K. Hansen, E. Branchini, P. Mazzotta, P. Cabella and K. Dolag, MNRAS 361,753
(2005).
19. K. T. Inoue and J. Silk, ApJ648, 23 (2006); K. T. Inoue and J. Silk, astro-ph/0612347.
20. A. Riazuelo, J. Weeks, J. P. Uzan, R. Lehoucq and J. P. Luminet, Phys. Rev. D 69,
103518 (2004); J. P. Luminet, J. Weeks, A. Riazuelo, R. Lehoucq and J. P. Uzan,
Nature 425, 593 (2003).
21. J. Shapiro Key, N. J. Cornish, D. N. Spergel and G. D. Starkman, astro-ph/0604616;
N. J. Cornish, D. N. Spergel, G. D. Starkman and E. Komatsu, Phys. Rev. Lett. 92,
201302 (2004).
22. D. Boyanovsky, H. J. de Vega and N. G. Sanchez, astro-ph/0607508; astro-ph/0607487.
23. L. Campanelli, P. Cea and L. Tedesco, astro-ph/0606266.
24. C. R. Contaldi, M. Peloso, L. Kofman and A. Linde, JCAP 0307, 002 (2003).
25. F. Ferrer, S. Riisiinen and J. Viiliviita, JCAP 0410, 010 (2004).
26. C. Gordon and W. Hu, Phys. Rev. D 70, 083003 (2004).
27. A. E. Giimriik<;iioglu, C. R. Contaldi and M. Peloso, astro-ph/0608405.
28. C.-H. Wu, K.-W. Ng, W. Lee, D.-S. Lee and Y.-Y. Charng, astro-ph/0604292.
29. S. Hofmann and O. Winkler, gr-qc/0411124.
30. S. Tsujikawa, P. Singh and R. Maartens, CQG 21, 5767 (2004).
31. D. D. Kocevski, C. R. Mullis, H. Ebeling, ApJ 608, 721 (2004).
32. D. D. Kocevski, H. Ebeling, ApJ 645, 1043 (2006).
33. M. J. Hudson, R. J. Smith, J. R. Lucey, E. Branchini, MNRAS 352, 61 (2004).
34. J. Lucey, D. Radburn-Smith, M. Hudson, astro-ph/0412329.
35. D. Proust et al., A&A 447,133 (2006).
36. M. J. Rees, D. W. Sciama, Nature 217, 511 (1968).
37. M. Panek, ApJ 388, 225 (1992).
38. C. L. Bennett et al., ApJS 148, 1 (2003).
39. Full-sky maps at http://www.physik.uni-bielefeld.de/cosmology/rs .html
 A NEW REALIZATION OF A LOW QUADRUPOLE UNIVERSE

LEE, WO-LUNG*
Department of Physics, National Taiwan Normal University,
Taipei, Taiwan 116, R. O. C.
* E-mail: [email protected]

We propose that cosmological density perturbation may originate from passive fluctua-
tions of the inflaton, which are induced by colored quantum noise due to the coupling
of the inflaton to the quantum environment. At small scales, the fluctuations grow with
time to become nearly scale-invariant. However, the larger-scale modes cross out the
horizon earlier and do not have enough time to grow, thus resulting in a suppression of
the density perturbation. This may explain the observed low quadrupole in the CMB
anisotropy data.

Keywords: Cosmic microwave background; Inflationary cosmology; Large-scale structure

of the Universe.

1. Introduction

The recently released data of the Wilkinson Microwave Anisotropy Probe (WMAP)
confirmed the earlier COBE-DMR's observation about the deficiency in fluctuation
power at the largest angular scales. 1 ,2 The amount of the quadrupole mode of the
CMB temperature fluctuations is anomalously low if compared to the prediction of
the ACDM model. The usual explanation to this low quadruple moment is to invoke
the so-call "cosmic variance", i.e. at large scales the CMB experiments are limited
by the fact that we only have one sky to measure and so cannot pin down the
cosmic average to infinite precision no matter how good the experiment is. Given
the large uncertainties due to the cosmic variance, we might never know whether the
phenomenon of the low CMB quadrupole constitutes a truly significant deviation
from the standard cosmological expectations.
Though the cosmic variance implies that we simply live in a universe with a low
quadrupole moment for no special reason, the lack of power on the largest scale can
be treated as a physical effect that demands a sensible explanation. There are many
proposed solutions for this anomaly. Basically, all the efforts can be classified into
three categories: (1) a possible non-trivial topology(for example, see J.-P. Luminet
et al. 3 and the references therein); (2) a cut-off due to causality (for example, see
A. Berera et al. 4 and the references therein); (3) initial hybrid fluctuations (for
example, see Lee & Fang 5 ). Recently the colored noise has been considered to explain
the anomaly in the context of stochastic inflation. 6 Instead of a Heaviside window
function used in the stochastic approach to inflation,7 Ref. [6] adopts a Gaussian
window function with a width characterizing the size of the coarse-grained domainS
which is then arbitrarily chosen to be comparable to the Hubble radius. This smooth
window function eventually yields a blue tilt of the power spectrum on large scales
which can be tuned to fit the WMAP large-scale anisotropy data. Nevertheless,
in this classical approach they inevitably resort to an ad hoc smoothing window

1653
1654

function and therefore the width of the window function remains undetermined.

2. Our model
To mimic the quantum environment, we consider a slow-rolling inflaton ¢ coupled
to a quantum massive scalar field (J, with a Lagrangian given by 9

1:: = ~gl.w
2
(] -+. (] -+. + ~gJLV (] (J (] (J _
JL'P v'P 2 JL v
V(-+') _
'P
m;
2
(J2 _ g2 -+.2(J2
2 'P ,
(1)

where V( ¢) is the inflaton potential that complies with the slow-roll conditions and
g is a coupling constant. Thus, we can approximate the space-time during inflation
by a de Sitter metric given by

(2)
where rJ is the conformal time and a(rJ) = -lj(HrJ) with H being the Hubble
parameter. Here we rescale a = 1 at the initial time of the inflation era, rJi = -1 j H.
Following the influence functional approach,lO,11 we trace out (J up to the one-loop
level and thus obtain the semiclassical Langevin equation for ¢, given by

¢ + 2aH¢ - V2¢ + a2 (V'(¢) + l((J2)¢) - g4 a2¢ x

J d4x'a 4(rJ')B(rJ-rJ')iG_(x,x')¢2(x') = ~~, (3)

where the dot and prime denote respectively differentiation with respect to rJ and
¢. The effects from the quantum field on the inflaton are given by the dissipation
via the kernel G_ as well as a stochastic force induced by the multiplicative colored
noise ~ with

(4)
The kernels G± in Eqs. (3) and (4) can be constructed from the Green's function
of the (J field with respect to a particular choice of the initial vacuum state which
we will specify later, given by

(5)
To solve Eq. (3), let us first drop the dissipative term which we will discuss
later. Then, after decomposing ¢ into a mean field and a classical perturbation:
¢(rJ, x) = ¢(rJ) + 'P(rJ, x), we obtain the linearized Langevin equation,

(6)
where the effective mass is m~eff = V" (¢) + g2 ((J2) and the time evolution of ¢
is governed by V (¢). The equation of motion for (J from which we construct its
Green's function can be read off from its quadratic terms in the Lagrangian (1) as

(J.. + 2 aH·(J - '\"72

v (J + a2 mueff(J
2
= 0, (7)
 1655

The solution to Eq. (6) is obtained 9 as

'Ph = ~i 11)
1), dr/ ¢(77')~h(r7') ['P},(7/)'P~(7]) ~ 'P~(7]/)'Pt(7])J ' (8)

where the homogeneous solutions 'Pi,2 are given by

'Pi,2 = 2a (nl7]I)1/2 H~1),(2) (k7]).

1
(9)

Here H~l) and HP) are Hankel functions of the first and second kinds respectively
and v 2 = 9/4 ~ m~efd H2. In addition, we have from Eq. (7) that

(10)

where the constants Cl and C2 are subject to the normalization condition, IC212 ~
1c11 2 = 1, and /1>2 = 9/4 ~ m;efrlH2.
Now, the power spectrum of the perturbation 'P can be obtained by employing
Eqs. (4) and (8) as

(11)

where the noise-driven power spectrum is given by

~~(1]) = g84 Z42

n
r }Zir
}Zi
dZ 1 dZ 2¢(7]d¢(1]2) sinz- [z-=:lsin 2Akz-
ZlZ2 Z-
~ 1] F(zdF(Z2)'
(12)
where z_ = Z2 ~ Zl, Z = k1], Zi = k7]i = ~ k / H, A is the momentum cutoff introduced
in the evaluation of the ultraviolet divergent Green's function in Eq. (5), and F(z)
is some trignometric function of z. 9
We plot ~~(7]) at the horizon-crossing time given by Z = ~2n versus k/H in
Fig. 1. The figure shows that the noise-driven fluctuations depend on the onset
time of inflation and approach asymptotically to a scale-invariant power spectrum
~~ ~ 0.2g4¢6/(4n2) at large k. Note that if 94¢6 ~ H2, ~~ will be comparable
to that of the known scale-invariant quantum fluctuation of the inflation which is
given by ~k = H2/(4n 2).12
In contrast to the scale-invariant density power spectrum induced by ~(k, the
noise-driven density power spectrum P~ induced by ~~ is blue-tilted on large scales.
Assuming a dominant P~ and using the set of cosmological parameters measured
by WMAP,2 we have run the CMBFAST numerical codes 13 to compute the CMB
anisotropy power spectrum as shown in Fig. 2. We can see that the known result
of the scale-invariant power spectrum at small scales can be achieved by properly
choosing the initial time of inflation. Meanwhile, the suppressed large-scale density
perturbation can account for the low CMB low-l multipoles.
The above results do not include the intrinsic fluctuation of inflaton. To take
that into account, an additional white noise term r;w in the free field case must be
involved on the right hand side of the Langevin equation Eq. (3). Furthermore, the
1656
c-~-~--·

0.2 ~

0.1

0.0 ---~--~--
o 100 200 300 400
kJH

Fig. 1. Power spectrum of the noise-driven inflaton fluctuations 8~ == 411"2 ~~/ g416. The starting
point, k/H = 211", corresponds to the k-mode that leaves the horizon at the start of inflation.

5000 i

4000 :
I
's3..: 3000
~

Q- 2000
:!:
~
i
1000
E~-~43~~- --~- __/
~----

-1000
10 100 1000

Fig. 2. CMB anisotropy in the ACDM model with the noise-driven density power spectrum P~.
The curves from below represent respectively different P~'s with k/ H = 20011", 50011", 100011", and
200011" corresponding to the physical scale of 0.05Mpc- 1 . The top curve is the ACDM model with
a scale-invariant power spectrum. We normalize all the anisotropy spectra at the first Doppler
peak. Also shown are the three-year WMAP data including error bars and the first year WMAP
data denoted by stars. 2

effective mass of the u-field in Eq. 7 must be modified as = m;eff m;

+g2( (1)2 + (<p~)),
where (<p~) represents the active quantum fluctuation with a scale-invariant power
spectrum given by 6.% = H2 j(47r 2 ).12 The total density power spectrum then is
given by

(13)

where X (¢) is determined by the slow-roll kinematics. Under the circumstances, the
power spectrum of the noise-driven inflaton fluctuations is not affected. However,
 1657

the CMB anisotropy in the ACDM model are modified accordingly, as shown in
Fig. 3.

6000

- IICDM
5000

.... IICDM+lloise (k/H=5(mj

4000

~
~ 3000
t'
~

?:f 2000
..:!;
"="
1000

-1000
1 10

Fig. 3. CMB anisotropy in the ACDM model with the density power spectrum Pk = P: + p}.
The solid curve is the ACDM model with a scale-invariant P:
induced by quantum fluctuations.
The dashed and dotted curves represent respectively noise-driven p}'s with k/H = 5007r and
k/H = 507r corresponding to 0.05Mpc~1

3. Conclusion
we have proposed a new scenario in which the inflaton is viewed as a phenomenolog-
ical classical field. In this model, the mean field drives the kinematics of inflation.
while its fluctuations are dynamically generated by the colored quantum noise as
a result of the coupling of the inflaton to a quantum environment. This model is
a better fit to the WMAP anisotropy data than the ACDM model. Moreover, the
observed low CMB quadrupole may unveil the initial condition for inflation. These
results are derived by solving approximately the Langevin equation (3) of the in-
flaton which is coupled to a massive quantum field. The noise term then generates
a stochastic force which drives the inflaton fluctuations. It is worth studying the
overall effect of the corrections induced by the dissipation, the static mass approxi-
mation, and the static mean field approximation to the spectral index of the density
power spectrum. Furthermore, the density power spectrum driven by the colored
noise is non-Gaussian.
Although we are based on a simple inflaton-scalar model, our results should be
generic in any interacting model. We have considered a quantum scalar field with
mass of the order of the Hubble parameter during inflation. In fact, one can consider
a scalar field with much higher mass. So it is mandatory to perform a full analysis
of the cosmological effects due to the quantum environment in a viable interacting
inflation model such as the hybrid or O(N) model.
1658

Acknowledgments
This presentation is based on the work astro-phj0604292. The author thanks Chun-
Hsien Wu, Kin-Wang Ng, Da-Shin Lee and Yeo-Yie Charng for the closed collabo-
ration on this subject.

References
1. C. L. Bennett et al., Astrophys. J. 464, Ll (1996).
2. C. L. Bennett et al., Astrophys. J. Suppl. Ser. 148, 1 (2003); G. Hinshaw et al.,
astro-ph/0603451.
3. J.-P. Luminet et al., Nature 425, 593 (2003).
4. L.-Z. Fang and Y. P. Jing, Mod. Phys. Lett. A 11, 1531 (1996); A. Berera, L.-Z. Fang,
G. Hinshaw, Phys. Rev. D 57, 2207 (1998).
5. W. Lee and L.-Z. Fang, Class. Quantum Gra. 17,4467 (2000); W. Lee and L.-Z. Fang,
Phys. Rev. D 69, 023514 (2004).
6. S. Mat arrese , M. A. Musso, and A. Riotto, J. Cosmol. Astropart. Phys. 05 (2004) 8;
M. Liguori et al., J. Cosmol. Astropart. Phys. 08 (2004) 11.
7. A. A. Starobinsky, in Field Theory, Quantum Gravity and Strings, edited by H. J. de
Vega and N. Sanchez (Springer, Berlin, 1986); A. S. Goncharov, A. D. Linde, and V.
F. Mukhanov, Int. J. Mod. Phys. A 2, 561 (1987).
8. S. Winitzki and A. Vilenkin, Phys. Rev. D 61, 084008 (2000).
9. C. H. Wu, N. K. Ng, W. Lee, D. S. Lee, and Y. Y. Charng, astro-ph/0604292.
10. M. Morikawa, Phys. Rev. D 42, 1027 (1990); B. L. Hu, J. P. Paz, and Y. Zhang, in
The Origin of Structure in the Universe, edited by E. Gunzig and P. Nardone (Kluwer,
Dordrecht, 1993); E. Calzetta and B. L. Hu, Phys. Rev. D 49, 6636 (1994); B. L. Hu
and S. Sinha, Phys. Rev. D 51, 1587 (1995); E. Calzetta and B. L. Hu, Phys. Rev.
D 52, 6770 (1995); C. Greiner and B. Muller, Phys. Rev. D 55, 1026 (1997); E. A.
Calzetta and S. Gonorazky, Phys. Rev. D 55, 1812 (1997); H. Kubotani, T. Uesugi,
M. Morikawa, and A. Sugamoto, Prog. Theor. Phys. 98, 1063 (1997); F. C. Lombardo
and D. L. Nacir, Phys. Rev. D 72 063506 (2005).
11. R. P. Feynman and F. L. Vernon, Ann. Phys. 24, 118 (1963).
12. S. W. Hawking, Phys. Lett. B 115, 295 (1982); A. A. Starobinsky, Phys. Lett. B 117,
175 (1982); A. H. Guth and S. Y. Pi, Phys. Rev. Lett. 49, 1110 (1982).
13. U. Seljak and M. Zaldarriaga, Astrophys. J. 469, 437 (1996).
 PERTURBATIONS OF DARK SECTORS FROM THE CMB

SERGEI BASHINSKY
Theoretical Division, T-8, Los Alamos National Laboratory, Los Alamos, NM 87545, USA

The kinetic properties of dark cosmological sectors are imprinted on the CMB through
the gravitational impact of the dark sectors' perturbations. The impact occurs primarily
during horizon entry, when general relativity is essential. By using a formalism which is
technically simpler than traditional ones and in which the gravitational forcing of large-
scale perturbations is concurrent with the responsible local dynamics of dark species, we
can alleviate gauge artifacts and map the observable characteristics of CMB fluctuations
to the properties of dark species which affect them. As an example, we show that the
order-of-magnitude enhancement of the CMB power above e cv 100 is entirely due to the
gravitational suppression of the CMB modes at the lower R's by matter inhomogeneities;
the frequently invoked resonant "radiation driving" at e 2; 100 is a gauge artifact. Con-
sequently, this feature strongly constrains dark dynamics and laws of gravity during the
matter but not radiation era.

1. Introduction

Galactic and cluster dynamics, cosmic structure, type Ia supernovae, the cosmic
microwave background (CMB), and the primordial abundances of light elements
have provided solid evidence that dark species constitute a significant energy frac-
tion of the universe at any of the redshifts z :s
10 10 . The dark species, presumably
decoupled from photons and luminous baryonic matter, influence the visible probes
by affecting both the background expansion and metric perturbations. Of the two
effects, the metric perturbations encode by far more independent cues about the
dark species' internal properties by offering new information on every spatial scale.
Correct interpretation of this information is essential for testing our views about
neutrinos and dark matter or for revealing the nature of the mysterious dark energy.
For a specific cosmological model and predetermined set of cosmological param-
eters it is generally straightforward to compute the CMB power spectra and matter
transfer functions with standard codes, if necessary, modified to include new dy-
namics. Yet for typical models, with close to ten unknown nuisance parameters,
it becomes intricate to establish numerically which of the observable changes in-
troduced by the new physics are nondegenerate with adjustments of the nuisance
parameters. The most unique and discriminative signatures may be tiny and easily
overlooked among large yet degenerate effects. As importantly, a black-box compu-
tation does not answer which aspects of a model are responsible for its observable
signatures. It, therefore, does not separate observationally favored from believed
true yet untested features of popular models.
The observables that reflect various characteristics of the dark species, as well as
the mechanisms through which these observables are impacted can be identified by
tracking evolution of the species' perturbations. Establishing the mechanism of the
impact is important for judging the robustness of the constraints from data. This
allows us to answer when the mechanism may fail or produce a different outcome.

1659
1660

There is, however, an important subtlety in making such identifications. Consid-

erable freedom exists in representing inhomogeneous evolution on scales comparable
to and exceeding the acoustic horizon. This descriptional freedom results in apparent
ambiguity in mapping observable features to specific physical mechanisms operating
on large scales. The explanations of physics underlying same features differ among
authors who use dissimilar albeit formally equivalent descriptions. We will argue in
Sec. 2 that some of the different explanations should be incorrect: there is a distinc-
tion between a connection of dark kinetics to an observable that can be identified
as an objective causal relation and as a gauge or descriptional artifact. In Sec. 3, we
outline a particularly simple formalism in which the apparent gravitational driving
of CMB or matter inhomogeneities on large scales coincides with the responsible
inhomogeneous dynamics in the dark sectors. By thus reducing gauge artifacts, we
are more likely to link dark kinetics and observables correctly. An example is given
in Sec. 4, with a comprehensive study in preparation.

2. Identifying interaction of species at horizon entry

To decipher the signatures of the dark kinetics in observational data, it is essen-
tial to understand gravitational interaction of perturbations during horizon entr-y.
During the entry and only then perturbations of all abundant dark species are
gravitationally imprinted on the visible species without suppression. On subhorizon
scales, the perturbations' gravitational interaction is suppressed, as it is evident from
the Poisson equation ([> = -k- 247rGa 2 5p ~ (Hlk)2 5pl p, where k is the (comoving)
wavevector of a perturbation mode and H is the expansion rate (in conformal time).
Back at the horizon entry, the factor (HI k)2 approaches unity and does not cause
suppression. In particular, only during the entry the visible species are influenced by
perturbations of dark radiation and dark energy, for which Jeans length is expected
to be comparable to the Hubble scale.
In study of evolution in the metric perturbed on large scales (k ;S H) descrip-
tional ambiguities arise. Of course, the predicted observables should be identical in
any description derived from same fundamental action. Yet, in different formalisms
features in the CMB and large scale structure appear to be generated in different
epochs and by different mechanisms. Notwithstanding, we can draw a distinction
between a connection of past local dynamics to a presently observable feature that
can be identified either as an objective causal relation or as a gauge artifact.
We consider a class of models with dark and visible sectors which interact only
through Einstein's gravity. It has been shown 1 (for generalization and applicabil-
ity beyond multifluid models see Ref. 2) that, when quantified by certain variables,
perturbations in separate sectors remain time-independent during superhorizon evo-
lution. a Therefore, during superhorizon evolution the species' perturbations do not
collect any information about the composition and dynamics of the universe.

aThese conservation laws may break down when gravitational decays of species into other species
are appreciable a This, however, is unlikely to endanger the post-BBN evolution, explored here.
 1661

On the other hand, the abundances and kinetic properties that dark sectors
have since the horizon entry do influence the observable cosmological anisotropies
and inhomogeneities. Because of the aforementioned freedom in representing the
evolution of large-scale inhomogeneities, in many formalisms, including the most
popular ones, the properties of dark species at horizon entry appear to affect the
CMB (or matter) perturbations long before of after the entry. This can mislead one
into assignment of an observable feature to an entirely different epoch and physical
process. This source of existing and potential misassignments can be eliminated
systematically, and with additional technical benefits.
A complete description of linearly perturbed Einstein-Boltzmann cosmological
dynamics in which the gravitational forcing of perturbations is concurrent with
the responsible local interactions and the above (as argued, unambiguous) causal
relations are manifest was suggested in Ref. 4. This description, based on canonical
coordinate variables, is simpler technically and practically than popular alternatives,
based on perturbations of proper quantities. As an example, in the next section we
apply this description to the evolution of CMB acoustic modes.

3. Dynamics of the CMB perturbations

When the CMB is tightly coupled to electrons by Thompson scattering, a Fourier
mode of photon (coordinate) overdensity d, == 15n,/n, evolves as
. + (R
d, Ifit 7-{ + 2Tdk 2)'d, + csk
2 2(d, - D) = O. (1)

Here and below Rb == 3pb/(4p,), Td determines the Silk damping rate,5 and
c; = [3(1 + Rb)]-l gives the sound speed in the photon-baryon plasma. Only scalar
perturbations are considered and the perturbed metric is described in the Newto-
J.
nian gauge ds 2 = a 2 [- (1 + 2<1> )dT2 + (1 - 2W)dx 2 The gravitational driving of
the CMB modes in eq. (1) is mediated through instantaneous equilibrium overden-
sity
D(T, k) = -3(<1> + 1]1 + Rb<1». (2)
Unlike proper number or energy overdensities, the coordinate overdensity d, is
strictly frozen on superhorizon scales. For adiabatic perturbations, the initial value
of d == 15ncoo/ncoo is related to the conserved Bardeen's curvature (6 as da , in = 3(in
for all species a. The variable d provides a reasonable parameterization of overden-
sity in the sense that it reduces to the overdensity 15n/n in an unperturbed FRW
metric, where 15n/n is unambiguous. For locally thermalized photons. the d, on
subhorizon scales gives the observable 315T,/T,. For a general nonthermal distri-
bution, it is also straightforward to quantify perturbation of CMB intensity by a
variable that is frozen beyond the horizon and becomes the measurable perturbation
of intensity after the horizon entry.4 Another convenient feature of eq. (1), or its
generalization 4 beyond the tight coupling, is independence of the gravitational driv-
ing from <i> and \ji, which, if present, would bring da through the Einstein equations
into the driving term.
1662

Eq. (2) shows that in the radiation era, when Rb <:: 1, the impact of dark per-
turbations on the CMB is mediated on any scale only through the sum <]) + \fI. 2 ,7
The same sum alone drives the CMB photons after their decoupling from baryons.
[Then this driving is well known as the integrated Sachs-Wolfe effect (ISW), 8 which
for subhoriz<J:II. phenomena can be usefully presented in terms of 1> + ~.l In con-
trast, matter perturbations on all scales respond only to the potential <]).2,4 Thus
a combined analysis of both the CMB and matter probes is particularly useful for
constraining the deviation of \fI 1<]) from unity, produced by free-streaming species
or generally expected in modified gravity.9

4. Application example: Dynamics responsible for the suppression

of the Sachs-Wolfe plateau relative to the acoustic CMB peaks
Systematic mapping of various kinetic properties of dark sectors to the observable
characteristics of the CMB and cosmic structure will be presented in a forthcoming
paper. Below, we consider one example of assigning a physical origin to a prominent
feature in the CMB temperature anisotropy.
First, consider the modes which enter the horizon after radiation-matter equality
but long before dark energy domination. (These modes provide the main contribu-
tion to the CMB multipoles with f! ;S 200.) In the epoch dominated by pressureless
CDM and baryons that have decoupled from the CMB, the gravitational potentials
<]) and \fI are generated entirely by pressureless matter and remain time-independent
throughout all the linear evolution. For these modes, the Doppler, polarization, and
ISW contributions to the observed CMB anisotropy AT(n)IT are negligible. For
clarity, let us also disregard CMB scattering on the reionized plasma, which is easily
accounted for by standard formulas. 1o The remaining contribution to the observed
temperature anisotropy is

AT
bTl
T. +<])+W, (3)
T In

where 8] lin == ~ d,,(, in. The quantity 8] lin is the perturbation to be observed for
the decoupled CMB photons if the metric would become homogeneous at any time
on superhorizon scales and remain homogeneous during the subsequent evolution. 4
For adiabatic perturbations and the evolution dominated by pressureless matter we
i
have <]) = \fI = - din, e.g. Ref. 2. Thus the Sachs-Wolfe result ATIT = <])/3 8 in this
description follows from cancellation of the primordial CMB density perturbation
bTITI in = - ~ <]) against the gravitational redshift <]) + \fI = 2<]) in proportion -5 : 6.

The 5-fold suppression of the primordial temperature perturbation Ii] lin results
in the suppression of the angular power Cz of CMB temperature by a factor of 25
(somewhat reduced by neutrino perturbations 2,12). This large suppression would not
be observed in models in which the potential decays early during horizon entry or
gravity is modified so that matter clustering does not generate metric perturbations.
 1663

1.5~~-~~~~-~

~- Adiabatic initial conditions

1.
- - -dCDMlnit=O
§ 0.5
c 0.5 \
~8
'"'"
~-'YwithCDM I
0.4
""-- ' _ ISW, early
-0.5
-1
\ - - -y,set<b+\fI:=ql

,---~_'~'~~_-_D~=_-3_(<I>~+_"-',--)/--.J
<£
~
1
0.3 Impact of COM perts.
-yi\
/ \ \
2
k S(t) In 0.2 \ \ baryon

~
9
..
:g 0.5
Mode with k = 0.002 h Mpc
Adiabatic i.e.
0.1 rr-_-Y.._. \1/\ loa:n
ro
- - - d CDM init= 0
~ 0 -------------
100 10

Fig. 1. Left top: Evolution of the CMB temperature perturbation d,(T, k) and the gravitational
driving term D = -3(<1> + w) [eq. (2) with Rb neglected] during matter domination (solid curves).
The driving is illustrated by a spring connecting the d, and D curves. Evolution of the same
primordial perturbation if the metric become homogeneous before the horizon entry (dashed curve).
Right: Suppression of the CMB temperature power C 1 for R ;S 100 by CDM inhomogeneities.
The solid curve shows Cz in the concordance ACDM model l l with adiabatic initial conditions.
The dashed curve describes the same model with changed initial CDM perturbations: dCDM is
artificially set to zero on superhorizon scales (the superhorizon d" d v , and db are unchanged).
Then the CDM inhomogeneities and the generated by them potential in the matter era are reduced.
The smoother metric suppresses less the CMB power at R ;S 100. Left bottom: The plots of <I> and
w confirm that the last model has smaller <i> + ,j:" hence, the enhancement of its power at low R
cannot be attributed to the ISW effect.

Traditional explanations of the Sachs-Wolfe result usually interpret the rela-

tion 6..T IT = <P 13 as - 2 : 3 cancellation between intrinsic temperature anisotropy
6T(Newt) ITlin = -~<P and gravitational redshift <p. 13- 15 With the smaller suppres-
sion of primordial anisotropy ascribed to dark matter, the rest of the jump in the
CMB power through £ rv 100 is explained, partly, by the small reduction of the
large-scale potential during the radiation-matter transition and, predominantly, by
close to a factor of 9 enhancement of the CMB power at R 2: 100 during the horizon
entry in the radiation era. 12 ,14,15 This conclusion is based on studying large-scale
evolution of traditional perturbation variables. The enhancement is interpreted as
a "resonant boost" of the CMB perturbations by their own gravitational potential,
whose decay soon after the entry is said to be "timed to leave the photon fluid
maximally compressed" .15
This explanation is inadequate because the suppression of large relatively to
small scales is not sensitive to the timing of the potential decay and even does not
require the metric to be perturbed in the radiation era. On the other hand, the
effect is sensitive to the evolution of matter perturbations in the matter era. The
incorrect assignment of this prominent observable feature to the kinetics of radi-
ation at high redshifts leads to wrong expectations whenever this view is applied
to anticipate the signatures of non-standard phenomena, such as non-minimal in-
teractions in the neutrino, dark matter, or dark energy sectors, modified gravity or
nonadiabatic initial conditions. The description presented here predicts the induced
1664

~:
III
"0

Fig. 2. Evolution of the CMB perturbation d"( and its driving D in the radiation era; impact of
perturbations in decoupled neutrinos (left) and quintessence (right). On both panels, the dashed
d"( and D curves describe a model without neutrinos and quintessence. On the left panel, the solid
curves are for the standard model with three neutrino flavors. On the right panel, the solid curves
show a fictitious model with the same 59% energy fraction in photons but the remaining 41 %
energy now carried by a scalar field cP, with a tracking background equation of state w¢ = 1/3.
Quantitatively, a change of N v from 3 to 4 changes the amplitude of d"( oscillations by _1.7%2
Equal addition of energy in tracking quintessence enhances the amplitude by 1.6%.4 The phase of
the CMB acoustic oscillations is shifted in both cases.

changes correctly. 4

Acknow ledgrnents
I thank organizers and participants of the MG 11 meeting and, particularly, of the
CMB Theory sessions for the interesting program and encouraging presentations.
I am grateful to S. Habib, K. Heitmann and G. Jungman for stimulating discussions,
and to R. Gupta for logistical help during proceedings preparation. This work was
supported by the us Department of Energy via the LDRD program of Los Alamos.

References
1. D. Wands, K. A. Malik, D. H. Lyth and A. R. Liddle, Phys. Rev. D 62, 043527 (2000).
2. S. Bashinsky and U. Seljak, Phys. Rev. D 69, 083002 (2004).
3. A. Linde and V. Mukhanov, JCAP 0604, 009 (2006).
4. S. Bashinsky, Phys. Rev. D 74, 043007 (2006).
5. J. Silk, Astrophys. J. 151,459 (1968); N. Kaiser, MNRAS 202, 1169 (1983).
6. J. M. Bardeen, P. J. Steinhardt and M. S. Turner, Phys. Rev. D 28, 679 (1983).
7. R. Durrer, Fund. Cosmic Phys. 15, 209 (1994).
8. R. K. Sachs and A. M. Wolfe, Astrophys. J. 147, 73 (1967).
9. E. Bertschinger, Astrophys. J. 648, 797 (2006).
10. U. Seljak and M. Zaldarriaga, Astrophys. J. 469, 437 (1996)
11. D. N. Spergel et al. [WMAP Collaboration], arXiv:astro-ph/0603449.
12. W. Hu and N. Sugiyama, Astrophys. J. 471, 542 (1996).
13. W. Hu and N. Sugiyama, Phys. Rev. D 51, 2599 (1995).
14. W. Hu, N. Sugiyama and J. Silk, Nature 386, 37 (1997).
15. W. Hu and S. Dodelson, Ann. Rev. Astron. Astrophys. 40, 171 (2002).
CMB Experiment Space
This page intentionally left blank
OBSERVATIONS OF THE CMB AND GALACTIC FOREGROUNDS
AT 11-17 GHZ: THE COSMOSOMAS EXPERIMENT

SERGI R. HILDEBRANDT'
Instituto de Astrofisica de Canarias. C/Via Lactea sin. La Laguna, Tenerife 38200
[email protected]

In this talk, I present the recent data analysis of the eOSMOSOMAS experiment. The
eOSMOSOMAS experiment is placed at the Observatorio del Teide (Tenerife, 2400 m).
It observes the sky with 5 channels between 10 and 17 GHz, covering a daily sky region
of ca. 7000 squared degrees, with an angular resolution of 1 degree. Its aim is the study
of eMB and Galactic foregrounds. The lower range frequency of the eOSMOSOMAS
experiment compared to the WMAP mission Or the forthcoming Planck Surveyor mission,
provides a natural complement for lower frequency behaviour of Galactic and extra
Galactic foregrounds. In particular, the eOSMOSOMAS experiment has given proofs
for the existence of a new foreground emission: anomalous dust emission. In this poster,
we also discuss whether the eOSMOSOMAS data are consistent with a spinning dust
model and whether this emission is also present at high Galactic latitudes. Finally, we
present first measurements of the polarization of this anomalous dust emission in the
Perseus molecular cloud which amounts to II = 3.4 ± 2% at 10.9 GHz.

1. Correlations of COSMOSOMAS data with other experiments.

The description of the COSMOSOMAS experiment can be found in. 1 ,2 Here I will
center the contribution in the main results: the correlation of COSMOSOMAS maps
with other experiments and observations.
To analyze the presence of Galactic foregrounds, we have cross-correlated 2 the
COSMOSOMAS and WMAP maps with several templates at lower and higher
frequencies. We use a technique similar to the one proposed by.3 It has been
detected a clear correlation between COSMOSOMAS and the 23 and 33 GHz maps
of WMAP, with the templates at 408 and 1420 MHz. 4 ,5 The spectral index of
this signal is in the range 2.87-3.15. We have found significant correlation between
COSMOSOMAS or WMAP with the He> map.6

2. High Galactic latitude anomalous microwave emission.

The existence of a fourth Galactic diffuse component that is correlated with the
thermal dust emission, but whose spectrum rises at lower frequencies (see 3 and
references therein) has received a considerable aattention in recent literature. The
COSMOSOMAS data detect an excess of corelation with Dust maps, which depends
on Galactic latitude and is not produced by synchrotron or free-free emission. Fur-
themore:

• On behalf of the eOSMOSOMAS colaboration (http://www.iac.es/project/ cmb/ cosmosomas/)

1667
1668

2.1. Direct detection of spinning dust in the Perseus molecular

complex.
We find evidence for anomalous microwave emission in the Perseus molecular cloud, 7
between IC348 and NGC1333, which shows a clear rising spectrum from 11 to 17
GHz. By extending the frequency coverage using WMAP maps convolved with the
COSMOSOMAS scanning pattern, we reveal a peak at near 22 GHz. The flux that
we measure at this frequency ( 40 Jy) is at least one order of magnitude higher
than can be explained in terms of normal Galactic emission processes (synchrotron,
free-free and thermal dust). No bright unresolved source (ultracompact HII region
or gigahertz peaked source) could be identified as responsible for this emission.
An adequate fit can be achieved by including a very significant contribution from
electric dipole emission from small spinning dust grains.

2.2. Polarization observations of the anomalous microwave

emission in Perseus.
Q results from the difference between the radiation intensity collected in the 0 plane
(North-South) and the 90 one. U is the difference between the orientation -45 0 and
+45 0 . For the Perseus complex we obtain Q = -0.2 ± 1.0% and U = -3.4 ± 2%
(both 95% c.l.) driving to an overall polarization parameter II = 3.4 ± 2%.8 The
maps are calibrated to the nearby California Nebula (free-free dominated) which is
assumed unpolarized (systematic error < 1%).

References
l. Gallegos, J. E., Macias-Perez, J. F., Gutierrez, C. M., Rebolo, R, Watson, R. A.,
Hoyland, R J., & Fernandez-Cerezo, S. 2001, 327, 1178
2. Fernandez-Cerezo, S. et al., 2006, MNRAS, 370, 15-24
3. de Oliveira-Costa, A., Kogut, A., Devlin, M.J., Netterfield, C. B., Page L. A., &
Wallack, E. J. 1997, ApJ, 582, L17
4. Haslam, C. G. T., Stoffel, H., Salter, C. J., Wilson, W. E. 1982, A&ASS, 47, 1
5. Reich, P. & Reich, W. 1986, A&ASS 163, 205
6. Finkbeiner, D. P. 2003, ApJS, 146, 407
7. Watson, RA., Rebolo, R., Rubino-Martin, J.A., Hildebrandt, S.R, Gutierrez, C.M.,
Hoyland, R.J., & Battistelli, E.S. 2005. ApJ, 624, L89
8. Battistelli et al. 2006, ApJ, 645, L141
CMB Data Analysis
This page intentionally left blank
 PROBING COSMIC DARK AGES WITH CMB POLARIZATION
MEASUREMENTS

L.A. POPA and P. STEFANESCU

Institute for Space Sciences,
Ro-077125 Bucharest, Romania
[email protected], [email protected]

C. BURIGANA
INAF/IASF Bologna,
via Gobetti 101, 40129 Bologna, Italy
[email protected]

We analyze the fur her potential returns from high sensitivity temperature and polariza-
tion anisotropy observations regarding the re-ionization history. They include improved
limits on cosmological parameters and the possibility to prove of a detailed ionization
history beyond the single optical depth estimate.

1. Introduction

Polarization is generated by Thomson scattering of the CMB anisotropic radiation

on free electrons coming mainly from ionzed hydrogen atoms, irrespective if the
anisotropy is produced by the density perturbations (E-mode) or by the gravita-
tional waves (B-mode).
The hydrogen recombination happens rather faster as the density of electrons
drops significantly in short time, making the Thomson scaterring extremely rare
and Universe transparent to the CMB photons. Later in the evolution of the Uni-
verse the first luminous sources start shining and their light re-ionize the hydrogen
atoms. The exact redshift of the epoch of hydrogen re-ionization, Zre, as well as the
properties of the sources driving this process are challenged by a variety of observa-
tional probes. A powerful observational probe is the value of the electron scattering
optical depth, T es , inferred from the CMB anisotropy measurements. With 3-year
data on the temperature anisotropy of the CMB and its polarization, the WMAP
experiment 1 ,2 produced an accurate determination of the electron optical depth to
the last scattering, Tes c:::: 0.09, consistent with an abrupt re-ionization at redshift
Zre c:::: 11. On the other hand, the Lyo: absorption spectra of high redshift quasars3
show that all known quasars at Z > 6 have a complete Gunn-Peterson trough and
a rapid evolving hydrogen neutral fraction compatible with the final stage of re-
ionization.
In this paper we analyze the furher potential returns from high sensitivity tem-
perature and polarization anisotropy observations regarding the re-ionization his-
tory of the Universe. They include the improved limits on cosmological parameters,
in particular the lifting of several degeneracies among the inflationary observables
that involve the optical depth through re-ionization and the possibility to prove of
a detailed ionization history beyond the single optical depth estimate. a

aThroughout is assumed Om = 0.24, Ob = 0.044, OA = 0.76 and Ho=72 km s-lMpc- 1 .

1671
1672

2. Improved science regarding the re-ionization history

The location of the re-ionization peak gives information about the epoch when re-
ionization happened while the height of the re-ionization peak is proportional to
the total amount of scattering, therefore with Tes. 4,5
The large angle E-mode polarization power spectrum is more sensitive to the re-
ionization history (the height of the re-ionization peak scales as the square of Tes)
than its cross-correlation with temperature (for which the height of the re-ionization
peak scales as Tes). The ability of high sensitivity large-angle E-mode polarization
observations to constrain Tes would drastically reduce the errors on the parameters
of primordial power spectrum of density perturbations: the scalar spectral index n s ,
the normalization of the scalar density perturbations As, and the running of the
scalar spectral index, n run .
The total amount of information that can be extracted at large angular scales
is challenged by some limiting factors: the cosmic variance, the uncertainties in
the determination of the angular power spectra of the polarized foregrounds, the
uncertainties in the determination of the re-ionization history of the Universe (the
evolution with redshift of ionization fraction).
Table 1 presents 3-a errors on the estimates of the parameters of primordial power
spectrum and electron scattering optical depth that can be potentially measured by
the future CMB projects BPOL and PLANCK at 143 GHz.
For this computation we use the Fisher matrix analysis taking into account the
contributions from cosmic variance and detector noise. For the case of BPOL we
assume an angular resolution of 20 arc min FWHM, a detector noise amplitude
of 5/LK-arcmin and two years for the mission duration. In the case of 143 GHz
channel of PLANCK we assume a resolution of 7.1 arcmin FWHM, a detector noise
amplitude of SOfLK-arcmin and 14 months for the mission duration. In both cases the
sky coverage is fsky = O.S. The computation is done for two distinct re-ionization
histories that have the same electron optical depth to the last scattering of the
CMB photons Tes = 0.09: an abrupt re-ionization at Zre ~ 11 and a more physical
inhomogeneous re-ionization history that assumes the re-ionization produced by a
Population III stellar cluster with mass of rv 100M8 and a heavy Larson initial
mass function. 6
Although BPOL has a coarser angular resolution than PLANCK at this frequency, its
higher sensitivity contribute to lift several degeneracies of inflationary observables
with optical depth through re-ionization b .
The foreground contamination and the residuals from foreground substraction
are ultimately the limiting factors in detecting the CMB linear polarization. We
assume that the polarized foregrounds can be substracted correctly down to a given
level and propagate the effects of foreground residuals into the errors for cosmolog-
ical parameter by using a similar procedure with that presented in. 7

b A; denotes the variation of the normalization of the scalar power spectrum at the pivot scale
ko = O.05Mpc- 1 compared to As = 2.95 X 10- 9 .
 1673

Fig. 1 presents the dependence of 3-a errors potentially detectable by BPOL

at 70 GHz and 143 GHz as function on foreground relative error. BPOL will can
easily measure OTes c:::o 1.5 X 10- 3 at 70 GHz and OTes c:::o 3.4 X 10- 3 at 143 GHz if
foregrounds are substracted at 5% level, but foreground contamination is ultimately
the limiting factor for the detection of r rv 10- 3 .

Fig. 1. 3-a errors on the estimates of the primordial power spectrum parameters and electron
optical depth that can be potentially detected by as function of foreground residual relative error.

FWHM=30orcmin a= 10fLKorcmin FWHM=20arcmin a=5fLKorcmin

BPol: 70 GHz BPol: 143 GHz or

0.010 . or ./
0.010
GA' ./
g -5----
cu on
b b
I ,
"' "' OT/ /' /'
."---.

--------.5------ 7 =-'- /' /es on .-/

. ..---S
0.001 0.001

0.01 010 1.00 10.00 100.00 0.01 0.10 1.00 10.00 100.00
Freground rei. error (%) Frcground rei. error (%)

Table 1. 3-a errors on the estimates of the primordial power spectrum parameters and electron
optical depth from Fisher matrix analysis.
Parameter Fiducial model Abrupt reion. Inhomogeneous reion.
BPOL PLANCK BPOL PLANCK
ns 0.95 8.55 x 10 -4 3.91 x 10 2.06 x 10 -J 3.92 X 10 -J
T 0.09 9.82 x 10- 4 4.60 X 10- 3 1.04 X 10- 3 4.64 X 10- 3
nrun 0.00 2.64 X 10- 3 7.53 X 10- 3 4.04 X 10- 3 7.56 X 10- 3
r = 2i- 1.00 X 10- 4 9.69 X 10- 4 2.56 X 10- 2 1.11 X 10- 3 2.58 X 10- 2
A*s 0.78 4.78 x 10- 3 2.18 X 10- 2 4.88 X 10- 3 2.19 X 10- 3

References
1. L. Page, et al., astro-ph/0603450.
2. D.N. Spergel, astro-ph/0603449.
3. X. Fan et al. AJ 128, 515 (2004).
4. L. Knox, New.A. Rev. 47, 883 (2003).
5. M. Kaplinghat et al. ApJ 583, 24 (2003).
6. L.A. Pop a astro-ph/0605358.
7. L. Verde, H.V. Peiris, R. Jimenez JCAP 01, 019 (2006).
 DARK ENERGY CONSTRAINTS FROM NEED LETS ANALYSIS
OF WMAP3 AND NVSS DATA

DAVIDE PIETROBON
Dipartimenta di Fisica, Universita di Rama "Tar Vergata",
V. della Ricerca Scientifica 1, 1-00133 Rama, Italy
davide. [email protected]

AMEDEO BALBI
Dipartimenta di Fisica, Universita di Rama "Tar Vergata" and
INFN Seziane di Rama "Tar Vergata", V. della Ricerca Scientifica 1, 1-00133 Rama, Italy
amedea. [email protected]

DOMENICO MARINUCCI
Dipartimenta di Matematica, Universita di Rama "Tar Vergata",
V. della Ricerca Scientifica 1, 1-00133 Rama, Italy
[email protected]

We cross-correlate the new 3 year Wilkinson Microwave Anistropy Probe (WMAP3)

cosmic microwave background (CMB) data with the NRAO VLA Sky Survey (NVSS)
radio galaxy data, and find further evidence of late integrated Sachs-Wolfe (ISW) effect
taking place at late times in cosmic history. Our detection makes use of a novel statistical
method based on a new construction of spherical wavelets, called needlets. The null
hypothesis (no ISW) is excluded at more than 99.7% confidence. When we compare the
measured cross-correlation with the theoretical predictions of standard, flat cosmological
models with a generalized dark energy component parameterized by its density, ODE,
equation of state wand speed of sound c;, we find 0.3 :s; ODE :s; 0.8 at 95% c.l.,
independently of c; and w. If dark energy is assumed to be a cosmological constant
(w = -1), the bound on density shrinks to 0.41 :s; ODE :s; 0.79. Models without dark
energy are excluded at more than 40". The bounds on w depend rather strongly on the
assumed value of c;.
1. Addressing the problem
The most outstanding problem in modern cosmology is understanding the mecha-
nism that led to a recent epoch of accelerated expansion of the universe. The evi-
dence that we live in an accelerating universe is now compelling. The most recent
cosmological data sets, luminosity distance measured from type Ia supernovae, 1 the
amount of clustered matter in the universe, detected from redshift surveys, clusters
of galaxies, etc.,2 and the pattern of anisotropy in the CMB radiation 3 have shown
that the total density of the universe is very close to its critical value, suggesting a
flat universe geometry. Matter, characterized by ordinary actractive gravity, covers
only'" 1/3 of the total density, while the remaining'" 2/3, res pons able for the ac-
celeration epoch, show a repulsive gravity. The precise nature of this cosmological
term, however, remains mysterious. The favoured working hypothesis is to consider
a dynamical, almost homogeneous component (termed dark energy).4
One key indication of an accelerated phase in cosmic history is the signature from
the integrated Sachs-Wolfe (ISW) effect 5 in the CMB angular power spectrum. A

1674
 1675

detection of a ISW signal in a flat universe is, in itself, a direct evidence of dark
energy. Furthermore, the details of the ISW contribution depend on the physics of
dark energy, in particular on its clustering properties, and are therefore a powerful
tool to better understand its nature. Unfortunately, because of the cosmic variance,
making the extraction of the ISW signal is a difficult task. A useful way to separate
the ISW contribution from the total signal is to cross-correlate the CMB anisotropy
pattern with tracers of the large scale structure (LSS) in the local universe.
In our analysis 6 we improve previous studies in two directions. On one side we
applied a new type of spherical wavelets, the so-called needlets,7 to extract cross-
correlation signal beetwen WMAP3 CMB sky maps with the radio galaxy NVSS
catalogue. Needlets have the great advantage to show a double localization: both in
multipole space and in pixel space due to the analitic properties of the shape of the
window function. This makes our analysis almost insensitive to the cuts present in
the maps, increasing our ability in the detection ofISW effect. On the other side we
treated dark energy as a generic fluid characterized by three physical parameters: its
overall density ODE, its equation of state w = p / p, and the sound speed c; = !5p /!5 p.
This parameterization has the advantage of being model independent, allowing one
to encompass a rather broad set of fundamental models, and of giving a more
realistic description of the dark energy fluid.
The first step in our work was to build the two maps at the proper resolution,
useful for our purpose, paying particular attention to remove noise and systematic
errors; secondly we extract the cross-correlation coefficients from maps. To make a
comparison with the theoretical prevision, we numerically integrated the Boltzmann
equation for photon brightness coupled to the other relevant equations, suitably
modifing the CMBFast code.

2. Results
In the left hand panel of the figure, we show the cross-correlation signal in wavelet
space extracted from the WMAP3 and NVSS data, superimposed to the predicted
cross-correlation for some dark energy models. The excess signal peaks at value
7 < j < 11, corresponding to angular scales between 2° and 10°, as expected from
theoretical studies. We checked that the observed signal was not produced by casual
alignment of sources in the NVSS catalogue with the CMB pattern at decoupling
(z c:::: 1100) and the errors, calculated through the Monte Carlo procedure, are close
to the analytical estimates, that confirm the consistence of our approach.
The first conclusion we can draw from our analysis is that the evidence for
non zero dark energy density is rather robust: we find 0.32 :<::: ODE :<::: 0.78 at 95%
confidence level. A null value of ODE is excluded at more than 4a, independently
of c;. When we model the dark energy as a cosmological constant (i.e. we assume
the value w = -1 for its equation of state), the bounds on its density shrinks to
0.41:<::: ODE :<::: 0.79 at 95% confidence level.
Quite interestingly, we find that although the case for a non zero dark energy
1676

0,6 VAl

Fig. 1. Cross-correlation signal in necdlet space extract from WMAP3 CMB data and NVSS radio
galaxy survey (left hand panel). Note that a simple Cold Dark Matter model does not reproduce
the signal (dotted line). in the right hand panel, constraints at 68%, 95% and 99% confidence level
in the rtDE-W plane for scalar field model (speed of sonnd c; 1).

contribution to the total density is compelling, the constraints on w do depend

on the assumed clustering properties of the dark energy component, namely its
sound speed c;. Phantom models, and also the ordinary cosmological constant case
w= perform worse when a quintessence behaviour c; 1 is assumed. However,
we emphasize that, for values of nDE rv 0.7, models with w -1 are a good fit to
the data, as it is evident from the right hand panel in figure.
Clearly, the observation of ISW is proving quite promising as a tool to answer
the questions arising from the mysterious nature of dark energy. While the CMB
data have reached a great degree of accuracy on the angular scales that are more
relevant for the detection of ISW, deeper red shift surveys and better catalogues can,
in the future, improve the tracing of tho localmattor distribution, thus allowing to
reduce tho orrors on the cross-correlation determination.

References
1. Riess, A. G., et al., ApJ 607, 665 (2004).
2. Springel, V., C. S. Frenk, & S. D. M. White, Natur 440, 1137 (2006).
3. Spergel, D. N., et al., arXiv:astro-ph/0603449 (2006).
4. Caldwell, R. R., R. Dave, & P. J. Steinhardt, PhRvL 80,1582 (19!:l8).
5. Saehs, R. K. & A. M. Wolfe, ApJ 147, 73 (1967).
6. Pietrobon, D., A. Balbi & D. Marinucd, PhRvD 74, 043524 (2006)
7. Baldi, P., G. Kerkyacharian, D. Marinucd, D. Picard, arXiv:math.ST /0606599.
 THE MATTER POWER SPECTRUM AS A TOOL TO
DISCRIMINATE DARK MATTER-DARK ENERGY
INTERACTIONS

GERMAN OLIVARES' and DIEGO PAVONt

Departmento de Fisica,
Universidad Aut6noma de Barcelona,
08193 Bellaterra (Barcelona),
Spain
• german. [email protected]
t [email protected]

FERNANDO ATRIO-BARANDELA
Department of Physics and Astronomy,
University of Pennsylvania,
Philadelphia, Pennsylvania 19104,
USA
[email protected]

The coincidence problem of late cosmic acceleration gets significantly alleviated when
a suitable interaction between matter and dark energy, either of phantom type or not,
enters the picture. We show that a class of models featuring this interaction fares rather
well when contrasted with the anisotropies of the CMBR and the matter power spectrum.
The latter test is very sensible to the interaction and may be used to discriminate between
different models.

Keywords: Cosmology, Cosmic microwave background, Matter power spectrum

1. Introduction

In order to account for the current stage of cosmic accelerated expansion many
dark energy candidates have been invoked up to now. By far, the simplest one is
the cosmological constant. However, while models based on it seem to fare rather
well when confronted with the observational data they face two severe difficulties on
the theoretical side: (i) its measured value is about 120 orders of magnitude lower
than the one predicted by QFT, and (ii) the "coincidence problem", namely, why
are the densities of matter and dark energy of precisely the same order today. This
is why dynamic dark energy models, not based on the zero-point energy of vacuum,
have been proposed in the recent years.
There is a particular class of models in which the dark energy interacts with
the dark matter (assumed non-relativistic) so that neither of them conserve sep-
arately.l Here, we consider a specific model within this class with the interesting
feature that it evades the first difficulty (the dark energy no longer relies on the
quantum vacuum) and considerable alleviates the second one. 2 In this model the
balance equations for the cold dark matter and dark energy are Pc + 3H Pc = Q,
and Px + 3H(1 + wx)p" = -Q, respectively, with Q = 3Hc 2 (pc + Px), where c2 is
a dimensionless, constant parameter that measures the strength of the interaction.

1677
1678

Thus, the ratio, Pel Px, between the aforesaid energy densities is found to evolve
from a constant, but unstable value, at early times to a lower, constant and stable
value at late times 2 -see also Fig. 2 of. 3
The target of short Communication is to use the cosmic microwave background
radiation (CMBR) and large scale structure to constrain this model (for full ac-
counts, see Refs. 3 and,4 respectively). As we shall see, the latter can be used as
a tool to discriminate between interacting and non-interacting models as well as
to discriminate between different interactions. The SNIa data practically do not
constrain the model as the c 2 shows large degeneracy -see Fig. 3 of. 3

2. Constraints from the CMBR

Upon exploring the whole parameter space and using the the first year data provided
by the WMAP satellite we found that the parameters of the model fall into the
following ranges: 3 Ox = 0.43 ± 0.12, Ob = 0.08 ± 0.01, ns = 0.98 ± 0.02, Ho =
56 ± 4km/s/Mpc. As for the equation of state parameter of the dark energy only
an upper bound can be set, Wx ::::; -0.86, while the preferred value of c 2 is 5 X 10- 3
but it still exhibits degeneration. The Bayesian Information Criteria5 allows to show
that this extra parameter fits the data better than models with no interaction. After
combining the WMAP and the SNIa data of Riess et al., Ox goes up to 0.68 and c2
increases to 6.3 x 10- 3 .

3. Constraints from the matter power spectrum

For the sake of comparison, we assume that in interacting and non-interacting mod-
els density perturbations have the same amplitude when they come within the hori-
zon. For non-interacting models, this prescription leads to the Harrison-Zeldovich
power spectrum, P(k) ~ k- n with ns = 1 on large scales. During the matter epoch,
if density perturbations and the background energy density evolve as be ~ aP / 2
and Pe ~ a-a, respectively, we have that P(k) ~ k-3+2p/(a-2). During the matter-
dominated era, p c:::' 2(0: - 2) + 0.6c 2 and the slope of the scale-invariant spectrum
is ns = 1 with a very weak dependence on the interaction.
The slope of the matter power spectrum on scales k ::;:. keq is determined by the
growth rate of subhorizon sized matter perturbations during radiation domination.
If a mode that crosses the horizon before matter radiation equality grows as be ~
q 2
T / during the radiation dominated era, then the amplitude of the power spectrum

today would be P(k) = P(k eq )(k eq lk)-3+q. For cold dark matter models, dark
matter perturbations experience only logarithm growth, so models with less growth
will have less power at small scales as do, for instance, mixed dark matter models 7
which contain a significant fraction of massive neutrinos. Figs. 2(a) and 2(b) of Ref.4
depicts the power spectrum for different interacting models and mixed dark matter
models, respectively. With increasing c 2 or mv the matter power spectrum exhibits
larger oscillations as a consequence to the increased ratio of baryons to dark matter.
Figs. 2(c) and 2(d), of the same reference, show that the slope of P(k) decreases
 1679

with increasing c 2 and mv. In both cases the behavior is rather similar. Therefore,
observations of large scale structure that constrain the neutrino mass also serve to
set constraints on the strength of the dark matter-dark energy interaction during the
radiation-dominated era and discriminate interacting from non-interacting models.
Further, there is an important difference between interacting and mixed models.
In the former the maximum of P(k) shifts to the left for increasing c2 , in the
latter the maximum does not shifts by increasing the neutrino mass. This can be
understood as follows. At larger c2 , the dark matter density becomes smaller at
any given redshift and the matter-radiation equality is delayed. This does not occur
with massive neutrinos where the matter-radiation equality takes place always at
the same redshift.
Since the interaction affects the slope of P(k), we resorted to the 2dFRGS data 8
to constrain c2 . We used a Monte Carlo Markov chain to run the CMBFAST code,
adapted to solve the interacting model. We ran the chain for 105 models, that were
sufficient to reach convergence. Fig. 4 of Ref. 4 depicts the joint confidence contours
at the 68%, 95%, and 99% level for pairs of parameters after marginalizing over the
rest. The leT confidence levels and upper limits for the model parameters resulted
to be: c2 ::; 3 X 10- 3 , Dch2 = 0.1 ± 0.02, Ho = 83:+:~okm/s/Mpc. The data are
very insensitive to Wx and baryon fraction. It is only fair to say that quintessence
non-interacting models are also compatible with the 2dFGRS data at leT level.

4. Concluding remarks
In summary, (i) the interacting dark matter-dark energy model of Ref. 2 significantly
alleviates the coincidence problem and is consistent with the observational data,
c2 < 10- 2 at 99% confidence level. (ii) keq is sensitive to the c2 value and decreases
with increasing c2 . (iii) The slope of the matter power spectrum is also affected by
the interaction -the stronger the interaction, the more negative the slope. This may
be of help to determine c 2 •

References
1. L. Amendola, Phys. Rev. D 62, 043511 (2000); L. Amendola and D. Tocchini-
Valentini, Phys. Rev. D 64, 043509 (2001); S. del Campo, R. Herrera, G. Olivares,
and D. Pavon Phys. Rev. D 74, 023501 (2006).
2. L.P. Chimento, A.S. Jakubi, D. Pavon and W. Zimdahl, Phys. Rev. D 67, 083513
(2003).
3. G. Olivares, F. Atrio-Barandela and D. Pavon, Phys. Rev. D 71, 063523 (2005).
4. G. Olivares, F. Atrio-Barandela and D. Pavon, Phys. Rev. D 74, 043521 (2006).
5. G. Schwarz, Ann. Statistics 5, 461 (1978); A.R. Liddle, Mon. Not. R. Astron. Soc.
351, L49 (2004).
6. A.G. Riess et ai., Astrophys. J. 607, 665 (2004).
7. M. Davies, F.J. Summers, and d. Schlegel, Nature (London) 359, 393 (1992).
8. S. Cole et al., Mon. Not. R. Astron. Soc. 362, 505 (2005).
 THE BRAIN EXPERIMENT

G. POLENTA, M. CALVO, L. CONVERSI, P. de BERNARDIS, C. GIORDANO, A.

IACOANGELI, M. MAIELLO, C. MARINI-BETTOLO, S. MASI, F. NATI, L. NATI, S.
PETERZEN, F. PIACENTINI, R. SORDINI and M. VENEZIANI
Dip. di Fisica, Univ. di Roma "La Sapienza"
piazzale A. Moro 5, 00185 Roma, Italy
gianluca. [email protected]

J. BARTLETT, E. BREELLE, C. DUFOUR, S. GALLI, K. GANGA, A. GHRIBI, Y.

GIRAUD-HERAUD, L. GUGLIELMI, J.C. HAMILTON, J. KAPLAN and M. PlAT
APC, Universite Paris 7, France

M. GERVASI, G. SIRONI, S. SPINELLI, A. TARTARI and M. ZANNONI

Dipartimento di Fisica, Univ. di Milano "Bicocca", Italy

B. MAFFEI, L. PICCIRILLO and G. PISANO

School of Physics and Astronomy, Univ. of Manchester, UK

P. ADE, A. ORLANDO and G. SAVINI

Department of Physics and Astronomy, Univ. of Cardiff, UK

J. BROSSARD, M. GIARD and J. LANDE

CESR - Toulouse, France

L. BERGE, L. DUMOULIN, A. JUILLARD and S. MARNIEROS

CSNSM - Orsay, France

F. PAJOT
lAP - Orsay, Prance

C. ROSSET
LAL - Orsay, Prance

The rotational component of the CMB polarization, the so-called B-modes, is one of
the major topic for next generation CMB experiments. This signal traces the effect
on the CMB due to primordial gravitational waves produced during the inflationary
epoch, probing the physics of the very early universe at GUT energy scales. This is a
challenge, being the expected amplitude of B-mode polarization ~ O.lj1.K. In this paper
we describe the BRAIN experiment, a bolometric interferometer which combines high-
sensitivity bolometric detectors with the excellent control of systematic effects proper
of interferometers. Being a ground based experiment, we identified Dome Charlie in
Antarctica as the best site for such measurements. In order to validate the goodness of
the site, as well as some of the implemented technical solutions, we built a pathfinder
experiment which has been successfully operated during last Antarctic summer, and we
report about preliminary results obtained.

1680
 1681

1. Introduction

The three defining characteristics of the Cosmic Microwave Background (CMB)

are the spectrum, the spatial anisotropy and polarization. The spectrum has been
measured by COBE-FIRAS and is consistent with a pure blackbody at 2.725K.1
Despite the recent new measurements of anisotropy and polarization,2-13 important
questions remain. Some of these, like "Does the observed CMB emanate directly
from the de-coupling era, or has it instead scattered from free electrons in an in-
tervening re-ionization of the universe? What causes the universal expansion and
is it accelerating? What is the nature of the dark matter and dark energy?" can
be answered by niche experiments, not covered by large space programs. One of
these is the detection of CMB polarization, and in special mode the rotational com-
ponent of the polarization field, known as the B-modes of CMB polarization. 14 ,15
This detection is very difficult to obtain and very important. Symmetry properties
of perturbations insure that B-modes cannot be produced by scalar perturbations,
the ones responsible for CMB anisotropy.16 The detection of B-modes opens the ex-
traordinary possibility of a direct investigation of tensor components (gravitational
waves) generated during the Inflation Phase, immediately after the Big Bang. The
amplitude of the B-modes component due to gravity waves, which is of the order of
0.1,uK,16,17 is proportional to the square of the inflationary energy scale, which can
thus be measured in this way. Primordial magnetic fields would also produce tensor
perturbations, but their level is negligible. For scale invariant tensor perturbations
the B-mode power spectrum peaks roughly at multipole f! ~ 100. Gravitational
lensing by the intervening matter distribution distorts the polarization pattern in
the sky mixing E and B modes. 18 ,19 However this effect peaks at small angular
scales (f! > 1000) and is negligible at large angular scales, where gravity waves are
important. The level of the B-modes signal is very low, about 100 times below the
CMB anisotropy. We want to be able to measure the Stokes parameters I, U and
Q of the CMB at multipoles between 40 and 250, with a precision of about O.l,uK
rms.

2. The BRAIN experiment

The polarized signal is expected to be extremely weak and this requires a design
oriented to minimize all possible sources of instrumental random and systematic ef-
fects, and exceptional atmospheric transmission and stability, characteristics which
can be obtained only in very high altitude, dry places on the earth, as Dome Charlie
in Antarctica. 2o ,21 BRAIN (Background RAdiation INterferometer) is a bolometric
interferometer designed to take advantage of both high sensitivity of bolometric
detectors and low level of systematic effects proper of interferometers.
Reflection off a tilted metal surface normally changes the Stokes parameters of
the incident radiation, so it is necessary to use the simplest front-end optics. The
same is applicable for refraction: any lenses or mirrors used to define the beam of the
polarimeter in the sky are then sources of spurious effects. Cross polarization and
1682

common mode signals are to be minimized because of their ability of mimicking true
sky signals: the result is that normal telescopes cannot be used. In BRAIN we use
circular corrugated feed horns for the front-end, and we combine the EM signals at
the output of the antennas, in front of the detector, in order to achieve the necessary
angular resolution. In this way we can use detectors which are not sensitive to the
phase of the EM field, like the cryogenic bolometers which are optimal in terms of
instantaneous CMB sensitivity.
The optimal operating frequency is 150 GHz. This is determined by high atmo-
spheric transmittance, good bolometer sensitivity and low Galactic contamination.
In order to measure the interesting range of multipoles we need a compact inter-
ferometer, with longest baselines of the order of 1m. Shorter baselines (which we
can obtain by combining signals from antennas closer to each other) will measure
the lower multipoles (largest angular scales). In order to reduce drifts and instru-
mental instabilities, we decided to use modulation and synchronous demodulation
in order to extract the small polarized signals from the dominant common mode
one. The signal from each antenna is modulated at a different frequency, in or-
der to unambiguously retrieve it in the combined signal. Modulation is obtained
by phase shifting the EM signal in a waveguide at the output of the antenna. All
phase-modulated signals are summed and squared in a non linear detector (the
bolometer), in total analogy with RF mixers. A schematic of the working principle
of BRAIN is shown in fig. 1. 22 ,23

Front Horns

Beam Combiner

Phase Shiftar

Seam Combiner

Back Homs

Feddlng Horns

Fig. 1. Working principle of the BRAIN interferometer. 22 ,23

A further modulation of the signal will be obtained by slowly rotating the in-
strument. The cross polarization of the system will be below 0.001. This means that
 1683

we can accept modulated common-mode signals 1000 times larger than the signal to
be measured. The CMB anisotropy signal, which is about 100 times larger than the
B-modes polarization, will produce a negligible spurious contamination. Other EM
signals present (like atmospheric and instrumental emission) are not modulated and
do not produce detected signals. Using conservative parameters for the bolometers
noise (N ETc M B = 250fJ,K Vs during the polar night on the Antarctic plateau) the
system has a noise of about 660f-lK Vs for each baseline. With 14 antennas we get
91 +91 independent baselines, which are summed in order to synthesize 6 window
functions (slightly overlapping) in the multipoles range of interest. We obtain a
noise of about 170f-lK Vs for each window. A 12 months integration produces a final
sensitivity of O.lf-lK for each window, which is the expected order of magnitude for
the primordial B-modes of the CMB.

3. The BRAIN pathfinder

In order to validate the use of a cryocooler based system as well as to characterize the
goodness of the Dome C site we built a pathfinder experiment. Of course, a good site
for astrophysical observation should have a very high transparency in the selected
wavelengths. However, for CMB B-mode observation we have to consider carefully
the polarization properties of the atmosphere as well. As described in Hanany and
Rosencranz 2003,24 the Earth magnetic field produces a Zeeman splitting of the
emission lines of spin 1 Oxygen molecules. The relative contribution of the three lines
corresponding to m = 0, ±1 depends on the angle between the line of sight and the
direction of the magnetic field. In particular, the lines associated to m = ±1 are fully
circularly polarized along the line of sight parallel to the magnetic field. Circular
polarization can be converted into linear polarization in real-world polarimeters.
Since the two lines are completely circular polarized, this effect is of particular
concern. This effect has not been measured yet, but the expected level should be of
the order of 10- 9 and 10- 6 of the total intensity for linear and circular polarization
respectively.

3.1. Instrument description

The instrument is based on a dry cryostat cooled by a Sumitomo Pulse Tube (PT)
cooler. All the parts of the cooler have been thermally insulated to allow operation
of the PT in the Antarctic environment. The first stage of the PT has a temperature
of rv 30K and acts as a thermal shield to reduce the radiative load on the second
stage, which has limit temperature of rv 2K and is used to cool down the back-to-
back horns collecting the incoming radiation and as a starting point for the fridge.
This is a double stage 4He/ 3 He one, able to cool the detector down to rv 0.3K.
Two 45° degree off-axis mirrors collect the radiation from the sky into the cryo-
stat through 8.5mm thick UHMW polypropylene windows, placed on opposite sides
of the cryostat. Three stages of optical filtering at 30, 4 and 0.3K reduce the load
on the detectors and define the spectral response which is centred at 1430Hz with
1684

~ 30% bandwidth.
Pointing of the telescope is allowed by both azimuth rotation of the whole in-
strument and elevation movement of the mirrors. The position is read by three
absolute encoders which have been calibrated using a sun sensor (azimuth) and
moon observations (elevation).
The detectors are silicon nitride micro mesh spider web bolometers, consisting
of a photolithographed mesh which provides high absorption efficiency, low heat
capacity and cosmic-ray cross section, and a neutron transmutation doped (NTD)
germanium thermistor which provides high sensitivity and stability.
In order to measure the circular polarization of the atmosphere, we put a polar-
izer and a Quarter-Wave Plate (QWP) in front of one of the window. The QWP is
a retarder tuned to shift by 90 0 the two orthogonal directions of the electric field of
the incoming radiation at 1450Hz, thus converting circular into linear polarization
(and vice versa). Using the Stokes description for electromagnetic waves, it can be
shown that the expected signal on the detector in the ideal case is:

S ="21 [
I + Q 1 + cos4wt
2 + Usin4wt
--
2
+ V sin2wt ] (1)

Hence, rotating the QWP with a frequency w the linear and circular polarization
signals will be modulated at 4w and 2w respectively, allowing to recover all the four
Stokes parameters I, Q, U and V, and making the instrument a circular polarization
polarimeter.

3.2. Measured instrument performance and preliminary results

from the 2006/2007 Antarctic campaign
The BRAIN pathfinder has been installed in Dome Charlie, Antarctica, 300m far
from the French-Italian Concordia Station, at the beginning of December 2006,
within the framework of the XXII Italian Antarctic expedition.
The whole instrument, was mounted on the roof of a sea container modified to
host the housekeeping of the experiment (fig. 2). The container itself is placed on
top of 5m high snow berm, reducing the accumulation of the snow. An external
shield prevents sun light from going into the telescope, and an inner shield avoids
ground spillover (fig. 2).
The cryostat ran successfully throughout the whole campaign (25 days). The
temperatures of operation for both stages of the PT and for the fridge were respec-
tively 29,2.1 and 0.349K. In particular, the temperature of the fridge was quite
stable among many different cycles of ~ 2 days each. To our knowledge, this is the
first successful operation of a PT cooler in Antarctica.
We characterized the detectors by means of standard V - I curves. The re-
sistance of the two bolometers was 1.9 and 3.6MO, corresponding to a ~ 5pW
background and a ~ 25K equivalent temperature of the atmosphere at the time of
the calibration.
 1685

Fig. 2. Left. Schematic of BRAIN pathfinder. Right. Installation of BRAIN in Dome C.

We divided the observation time in three parts: sky dips to characterize the
atmosphere in both temperature and polarization (4 hours); observation of the
moon for calibration (1 hour); observation of a selected region of the Galactic plane
(10 days).
Typical scans on the moon are shown in fig. 3. The signal is obviously visible
with very high signal to noise ratio. In particular, the white noise level is 2 and
3rnK / vYiZ for the two detectors, which is of course too high for sensitive ClVIB
observations but is good enough for the goal of the pathfinder. Better bolometers,
optimized for the measured radiative background, are being prepared.

Moon scons

1000

2002 2004 2006 2008 2010

tir0e (5)

2002 2004 2006 7008 2010 0.01 0.10 l.OO 10.00 100.00
time (5)

Fig. 3. Left. Typical scans with moon crossing for the two detectors. Right:. Signal pIns noise
power spectra for both bolometers during observation of the moon.

In fig. 4 we show a typical sky dip. Using a secant law model for atmospheric
emission, we estimate an optical depth T < 0.05 at 95%. However this is a prelimi-
nary result. A full analysis of all the collected data is in progress.
1686

sky dip
~-~-~~'--r--,-,--~~---c
-1.6x1O'

-1.8xl0'

-20x1O'
;:J

~ -22xl0'

~ -24xl0'
bJJ

lfJ -2.6xl04

-30xl04t
2)05 )710 2215 2220 )225 o 20 40 60
time Z (deg)

Fig. 4. Left. Typical sky dip in time ordered data. Right. Secant law fit for the sky dip.

4. DIBO: a bolometric interferometer demonstrator

DIBO is a demonstrator for bolometric interferometry at 90GHz. It is made with
classical waveguide technology using devices such as corrugated horns, guide twists,
phase shifters and hybrid junctions (see fig. 5). It has been designed to combine the
signals coming from two horns before loosing the phase information on the bolome-
ter. This system is now been fully characterised on a VNA and with a 4K bolometer
and has already validated the bolometric interferometry detection method. System-
atic effects are now being analysed in details.

5. Conclusions
In this paper we presented the BRAIN experiment, a bolometric interferometer
devoted to measure the B-modes of the Cosmic Microwave Background. We built a
pathfinder experiment which has been installed and operated in Dome C during the
last austral summer to test the atmospheric emission and the automatic operations
of a cryogenic system. Preliminary results suggest that the quality of the site is
as good as expected and the cryogenic system based on a pulse-tube cryocooler
performed well. Full analysis of the collected data is ongoing. Meanwhile we started
testing the prototype for the bolometric interferometer.

6. Acknowledgements
BRAIN is supported by PNRA and MIUR in Italy, and by IPEV in France. The
BRAIN team is especially thankful to all the logistic staff in Dome C for their
support during the 2006/2007 Antarctic campaign.
 1687

Fig. 5. Top. Schematics of the bolometer interferometric demonstrator. Each horn is sensitive to
one polarisation. The phase information is preserved by nsing two controlled phase shifters. The
hybrid junction combines the two signal components. Bottom. Photo of DIBO.

References
1. J. Mather et aI, Astrophysical Journal, 420, 439, 1994
2. S. Masi et aI, Astronomy fI:J Astrophysics, 458, 687, 2006
3. W.C. Jones et aI, Astrophysical J01trnal, 647, 823, 2006
4. F. Piacentini et aI, Astrophysical Journal, 647, 833, 2006
5. T.E. Montroy et aI, Astrophysical Journal, 647, 813, 2006
6. G. Hinshaw et aI, Astrophysical Journal accepted, in press, 2007
7. L. Page et aI, Astrophysical Journal accepted, in press, 2007
8. C.L. Kuo et aI, submitted to ApJ, astro-ph/0611198, 2006
9. A.C.S. Readhead et aI, Astrophysical Journal, 609, 498, 2004
10. A.C.S. Readhead et aI, Science, 306, 836, 2004
11. J.M. Kovac et aI, Nature, 420, 772, 2002
12. E. Leitch et aI, Astrophysical Journal, 624, 10, 2005
13. J.H.P. Wu et aI, submitted to ApJ, astro-ph/0611392, 2006
14. M. Kamionkowski, A. Kosowsky & A. Stebbins, Phys. Rev. D, 55, 7:368, 1997
1688

15. M. Zaldarriaga & U. Seljak, Phys. Rev. D, 58, 023003, 1998.

16. W. Hu, Phys. Rev. D, 65, 023003, 2002.
17. M.S. Turner & M. White, Phys. Rev. D, 53, 6822, 1996
18. Kaplinghat et aI, Phys. Rev. Lett. 91, 241301, 2003.
19. K. Smith et aI, Phys. Rev. D, 70, 043002, 2004
20. J. Storeyet aI, 2005, EAS Publication series, 14, 7-12, 2005
21. G. Dall' Oglio & L. Valenziano L, PASA, 16, 167, 1999
22. S. Masi et aI, EAS Publication series, 14, 87-92, 2005
23. G. Polenta et aI, New Astronomy Reviews, 51, 256, 2007
24. S. Hanany and P. Rosencranz, New Astronomy Review, 47, 1159, 2003
Observational Gravitational
Lensing
This page intentionally left blank
 MICROLENSING WITH THE RADIOASTRON SPACE
TELESCOPE*

ALEXANDER F. ZAKHAROV
National Astronomical Observatories of Chinese Academy of Sciences, 20A Datun Road,
Chao yang District, Beijing, 100012, China;
Institute of Theoretical and Experimental Physics, 117259, Moscow, Russia;
Bogoliubov Laboratory for Theoretical Physics, JINR, 141980 Dubna, Russia;
Center of Advanced Mathematics and Physics, National University of Science and Technology,
Rawalpindi, Pakistan
[email protected]

It is well-known that gravitational lensing is a powerful tool in the investigation of

the distribution of matter, including that of dark matter (DM). Typical angular dis-
tances between images and typical time scales depend on the gravitational lens masses.
For the of microlensing, angular distances between images or typical astrometric shifts
are about 10- 5 - 10- 6 as. Such an angular resolution will be reached with the space-
ground VLBI interferometer, Radioastron. The basic targets for microlensing searches
should be b point like images of gravitationally lensed systems and bright point-like ra-
dio sources at cosmological distances. In this case, an analysis of their variability and
a solid determination of micro lensing could lead to an estimation of their cosmological
mass density.
According to the schedule of the Russian Space Agency the space radio telescope
Radioastron will be launched in 2007. a This space based, 10-meter radio telescope
will be used for space - ground VLBI measurements. Observations will be made
in four wavelength bands, corresponding to A = 1.35 cm, A = 6.2 cm, A = 18 cm,
A = 92 cm. The fringe sizes (in micro-arc seconds) for the apogee of the above
mentioned orbit and for all Radioastron bands are given in Table 1. Thus, as it was
noted earlier, there are non-negligible chances to observe mirages (shadows) around
the black hole at the Galactic Center and in nearby AGNs in the radio-band (or
in the mm-band) using Radioastron (or Millimetron) facilities. Since a shadow size
should be about 50 /-las for the black hole in the Galactic Center and analyzing the
shadow size and shape one could evaluate the spin and charge of the black hole
since fringe sizes for shortest wave lengths are smaller than shadow diameters. 1

Table 1. The fringe sizes (in micro-arc seconds)

for the standard apogee Brnax (3.5 X 10 5 km)
Brnax(km)\>..(cm) 92 18 6.2 1.35
3.5 x 105 540 106 37 8

Gravitational micro lensing was discussed in details in a number of papers. 2 For

cosmological locations of gravitational lenses and stellar masses, typical angles be-

* This research has been partially supported by the National Natural Science Foundation of China
(NNSFC) (Grant # 10233050) and National Basic Research Program of China (2006CB806300).
ahttp://www.asc.rssi.ru/radioastron/.

1691
1692

tween images are about rv 10- 6 arcsec,3 or more precisely

BE = -RE :::::;
Ds
2.36 x 10- 6 h65 - 1/2{fff
--
M8
arcsec, (1)

where RE is the Einstein- Chwolson radius, Ds is an angular diameter distance

Ho .
between a source and an observer, h65 = ( /( ))' Ho IS the Hubble con-
65 km c· Mpc
stant. One could point out that gravitational lensed systems are the most perspec-
tive objects to search for microlensing. Astrometric microlensing (or displacements
of images or their splitting) could be detected in the gravitational lens system such
as B1600+434 in the case if a proper motion of source, lens and an observer are
generated mostly by a superluminal motion of knots in jet. In the case if there is
microlensing of core in the B1600+434 system for example, then astrometric mi-
crolensing in the system could be about should be about 20 - 40 ILaS 4 and the
Radioastron interferometer will have enough sensitivity to detect such an astromet-
ric displacement.
In principle microlensing for distant sources could be the only tool to evaluate
rlL from microlensing event rate. 5 To solve this problem with the Radioastron in-
terferometer one should analyze variabilities of compact sources with a core size
;S 40JLas and with high enough flux densities about ,2: 20 m,ly at 6 cm wavelength
and about ,2: 100 m,ly at 1.35 cm wavelength. To fit the most reliable model for
variabilities of the sources such as scintillations, microlensing etc. A fraction of the
sources in the list of extragalactic targets for VSOP and Radioastron about 13%
- 14 %. 6 Therefore, one could say that astrometric micro lensing (or direct image
resolution with Radioastron interferometer) is the crucial test to confirm (or rule
out) microlens hypothesis for gravitationally lensed systems and for point like dis-
tant objects at cosmological distances. Thus, just after the Radioastron launch it
will be the first chance to detect microlensing by a direct way. So, the main goal
of the paper to attract an attention to such a challenging possibility because, pre-
flight time is very short now and perspective targets should be analyzed carefully
by observational and theoretical ways in advance. A number of point like bright
sources at cosmological distances and gravitational lensed systems with point like
components demonstrating photometrical microlensing signatures is not very high
and the sources should be analyzed by the careful way to search for candidates
where microlens model is preferable in comparison with alternative explanations of
variabilities.

References
1. A.F. Zakharov, A.A. Nucita, F. De Paolis, G. Ingrosso, New Astronomy 10, 479
(2005); A.F. Zakharov, A.A. Nucita, F. De Paolis, G. Ingrosso, Retro gravitational
lensing for Sgr A * with Radioastron, in Pmc. of the 16th SIGRA V Conference on Gen-
eral Relativity and Gravitational Physics, eds. G. Vilasi, G. Esposito, G. Lambiase,
G. Marmo, G. Scarpetta, (AlP Conference Proceedings, 2005) 751, p. 227; A.F. Za-
kharov, A.A. Nucita, F. De Paolis, G. lngrosso, Observational Features of Black Holes,
 1693

in Proc. of the XXVII Workshop on the Fundamental Problems of High Energy and
Field Theory, ed. V.A. Petrov (Institute for High Energy Physics, Protvino, 2005)
p. 21; gr-qc/0507118; A.F. Zakharov, A.A. Nucita, F. De Paolis, G. Ingrosso, Mea-
suring parameters of supermassive black holes, in Proc. of XXXXth Rencontres de
Moriond "Very High Energy Phenomena in the Universe", eds. J. Tran Thanh Van
and J. Dumarchez, (The GIOI Publishers, 2005) p. 223; A.F. Zakharov, A.A. Nucita,
F. De Paolis, G. Ingrosso, Shadows (Mirages) Around Black Holes and Retro Gravi-
tational Lensing, in Proc. of the 22nd Texas Symposium on Relativistic Astrophysics
at Stanford University, SLAC-R-752, eds. P. Chen, E. Bloom, G. Madejski, V. Pet-
rosian, SLAC-R-752, eConf:C041213, http://www.slac.stanford.edu/econf/C041213,
paper 1226 (2005); A.F. Zakharov, F. De Paolis, G. Ingrosso, A.A. Nucita, Astron. fj
Astrophys. 442, 795 (2005); A.F. Zakharov, A.A. Nucita, F. De Paolis, G. Ingrosso,
Shadow Shapes around the Black Hole in the Galactic Centre, in Proc. of" Dark M at-
ter in Astro- and Particle Physics" (DARK 2004), eds. H.V. Klapdor-Kleingrothaus
and D. Arnowitt, (Springer, Heidelberg, Germany, 2005), p. 77; A.F. Zakharov, F. De
Paolis, G. Ingrosso, A.A. Nucita, Measuring the black hole parameters from space, in
Gravity, Astrophysics,. and Strings '05, Proc. of the 3rd Advanced Workshop, eds. P. P.
Fiziev and M. D. Todorov, St. Kliment Ohridski University Press, Sofia, 2006, p. 290.
2. A.F. Zakharov, Gravitational Lensing and Microlensing, (Janus-K, Moscow, 1997);
A.F. Zakharov, M.V. Sazhin, Physics-Uspekhi 41, 945 (1998); E. Kerins, MACHOs
and the clouds of uncertainty, in Cosmological Physics with Gravitational Lensing,
Proceedings of the XXXVth Rencontres de Moriond, eds. J. Tran Than Van, Y. Mel-
lier, M. Moniez, (EDP Sciences, 2001), p. 43; K. Griest, Baryonic Dark Matter and
Machos, in "Dark Matter in Astro- and Particle Physics", Proc. of the International
Conference DARK-2002, eds. H.V. Klapdor-Kleingrothaus, R.D. Villier (Springer-
Verlag Heidelberg, 2002), p. 62; A.F. Zakharov, Gravitational Microlensing and Dark
Matter Problem: Results and Perspectives, Publ. Astron. Obs. Belgrade 75, 27; astro-
ph/0212009; A.F. Zakharov, Gravitational microlensing and dark matter problem in
our Galaxy: 10 years later, in Proc. of the Eleven Lomonosov Conference on Elemen-
tary Particle Physics, ed. A.I. Studenikin (World Scientific, Singapore, 2005) p. 106;
astro-ph/0403619; A.F. Zakharov, Gravitational microlensing: results and perspectives
in brief, Letters to Physics of Particles and Nuclei (accepted), astro-ph/0610857.
3. J. Wambsganss, Gravitational Microlensing, Dissertation der Fakultat fur Physik der
Ludwig-Maximilians-Universitat, (Preprint MPA-550, 1990); J. Wambsganss, Cosmo-
logical Microlensing, in Micro lensing 2000: A new Era of Microlensing Astrophysics,
eds. J.W. Menzies and P.D. Sackett, (ASP Con£. Series, 2001) 239, p. 351.
4. M. Treyer, J. Wambsganss, Astron. fj Astrophys. 416, 19 (2004).
5. A.F. Zakharov, L. C. Popovic, P. Jovanovic, 2004, Astron. fj Astrophys. 420, 881
(2004); A.F. Zakharov, L. C. Popovic, P. Jovanovic, Contribution of microlensing to
X-ray variability of distant QSOs, in Gravitational Lensing Impact on Cosmology,
Proc. of the IAU Symposium, eds. Y. Mellier and G. Meylan, 225, (Cambridge, UK,
Cambridge University Press, 2005) p. 363; A.F. Zakharov, L. C. PopoviC, P. JovanoviC,
Searches of Stellar Mass Dark Matter from Analysis of Variabilities of High-Redshifted
QSOs, in XXXIX Rencontres de Moriond "Cosmology: Exploring the Universe", eds.
J. Dumarchez and J. Tran Thanh Van, (The GIOI Publishers, 2005) p. 41; astro-
ph/0406417.
6. A.F. Zakharov, Astran. Reports 50, 79 (2006); A.F. Zakharov, Intern. J. Mod. Phys.
D (accepted).
 MICROLENSING TOWARDS M31

SEBASTIANO CALCHI NOVATI

(on behalf of the POINT-AGAPE collaboration)
Dipartimento di Fisica "E. R. Caianiello", UniversitiL di Salerno,
Via S. Allende, 1-84081 Baronissi (SA), Italy
[email protected]·it

We report the results of a search for gravitational microlensing toward M31 to reveal
galactic dark matter in the form of MACHOs (Massive Astrophysical Compact Halo
Objects) in the halos of the Milky Way and M31. A high-threshold analysis of 3 years
of data yields 6 bright, short-duration microlensing events, which are confronted to a
simulation of the observations and the analysis. The observed signal is much larger than
expected from self lensing alone and we conclude, at the 95% confidence level, that at
least 20% of the halo mass in the direction of M31 must be in the form of MACHOs
if their average mass lies in the range 0.5-1 M8. This lower bound drops to 8% for
MACHOs with masses ~ 0.01 M 8 .

1. Introduction
Gravitational microlensing, as first noted by,l is a powerful tool for the detection
of massive astrophysical compact halo objects (MACHOs), a possible component
of dark matter halos. The main results towards the Magellanic Clouds have been
reported by the MACH0 2 and the EROS 3 groups: the former claims for a detection
of a MACHO signal, it results an halo mass fraction in form of MACHOs of ~ 0.5
M8 of about f ~ 20%, whereas the latter, in the same mass range, put a rather
strong upper limit, f < 7%. The possible contribution of the LMC dark matter
halo, besides that of the LMC self lensing, has also been discussed. 4
As an alternative line of sight, the Andromeda galaxy (M31), being both nearby
and similar to the MW, is a suitable target for such a search. 5 It allows the MW
halo to be explored along a different line of sight. It has its own halo that can be
studied globally, and its high inclination is expected to give a strong gradient in the
spatial distribution of microlensing events belonging to the M31 dark matter halo.
Several collaborations have undertaken searches for micro lensing toward M31:
AGAPE,6 SLOTT-AGAPE,7,8 WeCAPP,9 POINT-AGAPE,lO MEGAY Here we
report the results of a search for short duration and bright microlensing events
towards M31 carried out by the POINT-AGAPE collaboration. 12

2. Evidence for a MACHO contribution to Galactic Halos

To search for micro lensing events, we use the "pixel-lensing" technique,13 in which
one monitors the flux variations of unresolved sources of each pixel element of the
image. Looking, in particular, for a MACHO signal we must deal with two main
backgrounds: (i) variable stars masquerading as microlensing events and (ii) self-
lensing events (for which both the lens and the source are part of the luminous
components of M31 or MW).

1694
 1695

Three distinct steps can be identified during the analysis: (i) the selection
pipeline; (ii) the Monte Carlo simulation; (iii) the detection efficiency analysis.
First, a catalogue of flux variations is established. Second, a serie of cuts is ap-
plied where we remove flux variations: with too small signal-to-noise ratio; showing
more than one bump; that badly fit the expected Paczynski shape; that are not
sufficiently sampled along the bump. This way we select about 10 4 flux variations.
Eventually, we look for microlensing only among short duration (t 1 / 2 < 25 days)
and bright (f::lR < 21) events.
As a result of the selection pipeline, out of the initial rv 10 5 set of detected
flux variations, 6 short-duration and bright microlensing-like flux variations are
selected. Because of t.heir characteristics (spat.ial and duration dist.ribut.ions), even
if we cannot absolutely rule out non-micro lensing sources of stellar variability, all
scenarios that would invoke variability to explain our candidate list are extremely
constrained, indeed contrived: we assume all 6 of them to be genuine microlensing
events.
The aim of the Monte Carlo is to predict the number and the characteristics of
the expected signal in the experiment, trying to mimic the observational conditions

1.0~_ _ _~

O.OL-~~~~~~~~~~~~~~~~~_~~~
10- 5 10- 4

Fig. 1. Most probable value, upper and lower 95% CL limit for the halo fraction as a function of
the MACHO mass.
1696

and the selection process. Together with the overall observational setup, we have
to build an astrophysical model of the processes involved: the surface brightness
distribution of the target galaxy, the luminosity function for the sources, the mass
function and a model for the density distribution for the lenses, the velocity dis-
tribution for both lenses and sources. Particular attention must be payed to the
evaluation of the self-lensing contribution. To this purpose, we vary the parame-
teres of our model, in particular the }VI/ L stellar ratio that drives the values of
the total mass to be associated with the stellar lens component. In a further step,
microlensing events selected within the Monte Carlo are simulated on the real data
set so to evaluate the detection efficiency for the selection pipeline.
We conclude with a prediction for about 1 expected self-lensing event and about
7 MACHO lensing events for a full halo of 0.5 M8 dark objects. Taking into account
the spatial distribution of the events, self lensing is expected to be more concentrated
in the innermost M31 region, whereas at least 1 event (N2) is detected in the outer
region, we conclude that, in the mass range 0.1-1 M 8 , MACHOs should contribute
for a sizeable mass fraction, f > (10 - 20)%, to the galactic halos (Fig. 1).
This result has been challenged by the MEGA collaboration who analysed the
same data set. 14 They claim that the observed microlensing events can all be ex-
plained with the expected self-lensing signal and put, in the same mass range around
0.5 M 8 , an upper- limit on the halo fraction of f < 30%. The debate about the dark
matter nature of galactic halos is still open.

References
1. B. Paczynski 1986, ApJ 304, 1
2. C. Alcock, R. AUsman, D. Alves, et al. 2000, ApJ 542, 281
3. P. Tisserand, L. Le Guillou, C. Afonso, et al. 2004, astro-ph/0607207
4. S. Calchi Novati, F. De Luca, P. Jetzer & C. Scarpetta 2006, A&A 459, 407
5. P. Baillon, A. Bouquet, Y. Giraud-Heraud & J. Kaplan 1993, A&A 277, 1
6. R. Ansari, M. Auriere, P. Baillon, et al. 1999, A&A 344, L49
7. S. Calchi Novati, G. lovane, A. Marino, et al. 2002, A&A 381, 848
8. S. Calchi Novati, P. Jetzer, G. Scarpetta, et al. 2003, A&A 405, 851
9. A. Riffeser, J. Fliri, R. Bender, S. Seitz & C. Gossl 2003, ApJ 599, L17
10. S. Paulin-Henriksson, P. Baillon, A. Bouquet, et al. 2003, A&A 405, 15
11. J. de Jong, K. Kuijken, A. Crotts, et al. 2004, A&A 417, 461
12. S. Calchi Novati, S. Paulin-Henriksson, J. An, et al. 2005, A&A 443, 911
13. A. Gould 1996, ApJ 470, 201
14. J. de Jong, L. Widrow, P. Cseresnjes, et al. 2006, A&A 446, 855
 DOES THE LMC HALO CONTRIBUTE SIGNIFICANTLY TO THE
MACHO EVENTS?

GAETANO SCARPETTA
Dipariimenio di Fisica UE. R. Caianiello ", Univer'siiri di Saler'rw,
Via S. Allende, 1-84081 Baronissi (SA), Italy
[email protected]

Our main purpose is to determine the lens population to which the events found by the
MACHO collaboration towards the LMC may belong. We show that only a fraction of
the events have characteristics that match those expected for lenses belonging to the
MACHO population of the Galactic halo. This suggests that this component cannot
explain all the candidates. Accordingly, we challenge the view that the dark matter halo
fraction of both the Galaxy and the LMC halos are equal, and indeed we show that, for
a MACHO mass in the range 0.1-0.3 MG), the LMC halo fraction can be significantly
larger than the Milky Way's. In this perspective, our main conclusion is that up to about
half of the observed events could be attributed to the LMC MACHO dark matter halo.

1. Introduction
Gravitational galactic microlensing is an efficient tool for the study of the dark mat-
ter component of the halos in form of MACHOs l (Massive Astrophysical Compact
Halo Objects). The line of sight towards the Magellanic Clouds has been explored
but the results obtained up to now are highly debated. The MACHO collabora-
tion, 2 according to the detection of 13-17 microlensing events, claims for a signif-
icant (Galactic) halo fraction (f rv 20%) in form of rv 0.5 MC') MACHO objects,
while the EROS collaboration,3 in the same mass range, put a rather strong upper-
limit, f < 7% at 95% CL. Overall, the picture remains unclear, in particular the
problem of the nature and location of the lens population of the observed events.
Following the original suggestion of Sahu,4 many authors tried to explain them as
belonging to some known luminous component of the LMC ("self lensing"). How-
ever these conclusions have been challenged: 5- 7 indeed it has been argued that the
microlensing events towards the LMC, as observed by the MACHO collaboration,
do not belong all to the same lens population.
The question we address here is whether and to what extent events due to the
lens population residing in the LMC dark matter halo can contribute to the observed
rate. l3

2. The LMC MACHO contribution to micro lensing events

As a main tool of investigation we use the differential rate of microlensing events
with respect to the Einstein time TE,9 to make predictions of the timescale, the
number and spatial distribution of the expected events, which we compare to the
observed quantities. To evaluate the expected quantities we take into account the
detection efficiency. 10
We model the LMC with a luminous part, a bar and a disc, plus a stellar and

1697
1698

a dark matter halo. ll •12 For both the LMC and the Galactic halo we assume the
"standard" isothermal profile with a core radius.
With respect to the rv 10 observed events, we find a very small contribution of
self-lensing events, 1-2, an even smaller one from the stellar halo, and a large one of
MACHOs, strongly dependent from the value of the MACHO mass. For a full halo
of 0.2 (0.5) M8 we find rv 80 (rv 60) expected events, for the Galactic halo, and rv 8
(rv 5) for the LMC one. This of course points towards a significant halo fraction in
form of MACHOs.
Two strong indications on the nature of the lens population come from the
analysis of the timescale and spatial distribution. For the former, we find that, for
a given MACHO mass, Galactic lenses give rise to shorter events than those due to
LMC MACHO lenses. However, a stastical analysis carried out with a Kolmokorov-
Smirnov test allows us to show that the latter component is in better agreement with

0.6
0.4
0.2

0.2 0.4 0.6 0.8 1

m (solar mass)
u
::E 1
4-<.....:l
0.8
0.6
0.4
0.2

0.2 0.4 0.6 0.8 1

m (solar mass)
Fig. 1. Galactic and LMC dark matter halo mass fraction in form of MACHOs, median value
with 68% CL errors, as a function of the MACHO mass. Values found after marginalisation of the
distribution function P (fMW, hMC) (see text).
 1699

the observed timescales. In particular it results that the preferred value is for '" 0.2
M(') LMC MACHOs. As for the spatial distribution, the inclination of the LMC
disc with respect to the line of sight gives rise to an expected spatial asymmetry
for LMC halo lenses, as first proposed and discussed by Gould. 8 We use the non-
parametric Pearson test, that allows us to evaluate the probability to accept the null
hypothesis that the probability scheme matches the measured distribution. We find
a probability of 46% and 26% for a MACHO LMC and MW halo lens population,
respectively. This outcomes make the lack of a predominant lens population very
clear, and show that the usual implicit assumption of equal halo fractions for both
the MW and the LMC dark matter halos should be reconsidered. We carry out
a maximum likelihood analysis in the fMW, fLMC space (the halo fraction of the
Galaxy and of the LMC, respectively), keeping the mass fixed as a parameter. It is
then possible to compute the probability distribution function for the halo fractions
given the observed events, P (fMw, hMC)'
The main outcome of this analysis is that in a significant range of the MACHO
mass, the LMC and MW dark matter halo fractions are not expected to be equal
(Fig. 1). In particular, from 0.1 up to 0.3 M(,), both a high value for hMC and a low
one for fMW are expected. For our preferred value for the MACHO mass we find
hMC '" 80% so that up to about 6 events might be attributed to this population.
For higher values of the mass, the LMC MACHO halo fraction turns out to be
almost degenerate though compatible with zero, and, for 0.5 M(,), we recover the
result 2 of a Galactic MACHO halo fraction of about 20%.

References
1. B. Paczynski 1986, ApJ 304, 1
2. C. Alcock, R. AUsman, D. Alves, et al. 2000, ApJ 542, 281
3. P. Tisserand, L. Le Guillou, C. Afonso, et al. 2004, arXiv:astro-ph/0607207
4. K. Sahu 1994, PASP 106, 492
5. Ph. Jetzer, L. Mancini & G. Scarpetta 2002, A&A 293, 129
6. L. Mancini, S. Calchi Novati, Ph. Jetzer & G. Scarpetta 2004, A&A 427, 61
7. A. Green & K. Jedamzik 2002, A&A 395, 31
8. A. Gould 1993, ApJ 404, 451
9. K. Griest 1991, ApJ 366, 412
10. D. Bennett 2005, ApJ 633, 906
11. van der Marel et al. 2002, AJ 124 2639
12. D. Alves 2004, ApJ 601, L151
13. S. Calchi Novati, F. De Luca, P. Jetzer & G. Scarpetta 2006, A&A 459, 407
 A NEW ANALYSIS OF THE MEGA M31 MICROLENSING
EVENTS

A.A. NUCITA, G. INGROSSO, F. DE PAOLIS, and F. STRAFELLA

Department of Physics and INFN, Sezione di Leece,
University of Leece, Via Arnesano, 1-73100 Leece, Italy

S. CALCHI NOVATI and G. SCARPETTA

Dipartimento di Fisica "E. R. Caianiello", Universita di Salerno, 1-84081 Baronissi (SA) and
INFN, Sezione di Napoli, Italy

Ph. JETZER
Institute for Theoretical Physics, University of Zurich, Winterthurerstmsse 190, CH-8057
Zurich, Switzerland

We discuss the results of the MEGA micro lensing campaign towards M31. Our analysis
is based on analytically evaluating the microlensing rate, taking the observational effi-
ciency into account as given by the MEGA collaboration. We find that only for extreme
models of the M31 luminous components it is possible to reconcile the total observed
MEGA events with the expected self-lensing contribution. However, the expected spatial
distribution of self-lensing events is more concentrated with respect to the distribution of
observed events. We thus find difficult to explain all observed events as due to self-lensing
alone. On the other hand, the small number of events detected up to date does not yet
allow to infer firm conclusions on the halo dark matter fraction in the form of MACHOs.

The two Collaborations MEGA and POINT-AGAPE analyzed independently

the same pixel-lensing observations towards M31 by the INT Telescope reaching al-
most opposite conclusions about the existence of MACHOs in MW and M31 halos.
In particular, the analysis of the MEGA collaboration,l although not conclusive,
is in agreement with a no MACHO hypothesis since the observed event rate is
consistent with the rate predicted for self-lensing alone. On the other hand, the
POINT-AGAPE result 2 is that the observed signal is much larger than that ex-
pected by self-lensing, and at least 20% of the halo mass in direction of M31 must
be in form of MACHOs. It is noticing that this result is compatible with that of the
MACHO collaboration, who has put a lower limit on the halo fraction in form of
MACHOs of cv 20% for objects in the mass range 0.5 -1 MG' Of course, the conclu-
sion about the halo fraction in form of MACHOs depends critically on the prediction
of the expected signal due to known luminous populations, this being dominated
by the "self-lensing" signal where both source and lens belong to same star popula-
tion residing in the M31 galaxy. In this respect, the modelling of the M31 luminous
components is a crucial aspect to be dealt with in order to get meaningful results.
We have first considered these aspects in a Monte Carlo simulation program 4 which
we have used to investigate the nature and location of the microlensing candidates
events towards M31 as reported in a first paper by the MEGA collaboration.
Here, our aim is to further explore these issues taking into account the latest

1700
 1701

MEGA results. Thus, we consider the question whether the expected self-lensing
signal due to stars belonging either to the bulge or the disc of M31 is able, as
claimed by the MEGA collaboration, to fully explain their results. Hence, we report
about the expectation on the event number by using

• an analytical evaluation of the micro lensing rate

• the observational efficiency as given by the MEGA collaboration.

In doing this, we started from the available observations in order to have a good
model for the three-dimensional structure of the Andromeda galaxy. In fact, the
visible mass distributions for the M31 bulge and disk are derived by fitting the ob-
served brightness profiles 3 and by further assuming realistic mass-to-light ratios for
bulge and disk stellar populations. 6 Note, however, that these distributions depend
on the chosen mass-to-light ratios so that also extreme models (i.e. massive bulge
and disk) can be obtained. Moreover, consideration of the M31 rotation curve data
allows us to derive the distribution of the dark matter in the M31 halo. Once the
mass density distributions (for the bulge, disk and halo components) are obtained,
one can evaluate the expected number of pixel-lensing events towards different line
of sight to M31. 5
We find that it is hard to explain all detected events as being due to self-lensing
alone. On the other hand, the small number of events does not yet allow to draw
firm conclusions on the halo dark matter fraction in form of MACHOs.
In particular, our main result is that only for extreme models of the M31 lumi-
nous components it is possible to reconcile the total observed MEGA events with
the expected self-lensing contribution. Nevertheless, the expected spatial distribu-
tion of self-lensing events is more concentrated and hardly in agreement with the
observed distribution.
To be more quantitative, by using the Kolmogorov-Smirnov test, we find a K-
S probability ~ 0.51 for self+dark lensing and ~ 0.18 for self-lensing only, thus
implying that a dark matter contribution to microlensing seems to be favored.

References
1. J. T. A. de Jong et al., Astronomy and Astrophysics 446, 855, (2006).
2. S. Calchi Novati et al., Astronomy and Astrophysics 443, 911, (2005).
3. S. M. Kent, Astrophysical Journal 97, 1614, (1989).
4. G. Ingrosso et al., Astronomy and Astrophysics, 445, 375, (2006).
5. G. Ingrosso et al., Astronomy and Astrophysics, in press, (2006).
6. Riffeser A. et al., Astrophysical Journal Supplement Series 163, 225, (2006).
 ON THE LENS NATURE IN MICROLENSING SEARCHES

F. DE PAOLIS, G. INGROSSO, A. A. NUCITA

Department of Physics and 1NFN, Sezione di Leece,
University of Leece, Via Arnesano, 1-73100 Leece, Italy
francesco. [email protected]

Some recent results on the attempts for direct searches for the lens object in selected
microlensing events are presented.

It is well known that in microlensing observations it is not possible to determine

unambiguously the lens parameters (mass, distance and transverse velocity) since
only two quantities are measured from observations (the event duration and its
amplification). Due to this parameter degeneracy, only statistical information on
the lens can be inferred from observations, and that is also true for parallax events.
It would be of course of the greatest importance to determine unambiguously
the lens parameters since it is still open in most cases the issue of the nature of
the lens objects. In fact, only in two cases (i.e. for two microlensing events) it has
been possible to directly image the lens object. HST has been pointed towards the
direction of all the 18 events towards the Magellanic Clouds and only in one case
(towards the microlensing event MACHO LMC-5) the lens has been observed! and
has been classified as an M-dwarf with mass 0.097 ± 0.0l6.iY18 at distance of about
550 pc from Earth (see als0 2 were it is discussed that LMC-5 is likely a jerk-parallax
event). In the second case, the Spitzer telescope has conducted a program to follow-
up in the mid-infrared the MACHO Large Magellanic Cloud microlensing fields. A
significant flux excess towards the MACHO LMC-20 has been discovered and this
observation, combined with high resolution near-infrared Magellan/PANIC data has
allowed to classify the lens as an early M dwarf in the thick disk of the Milky Way,
at a distance of about 2 kpc. 3
We would like to emphasize that only direct observations towards microlensing
events may allow discovering the true lens nature. To this aim we would like to
briefly report about such an attempt recently made towards the event MACHO-96-
BULGE-5, the longest event in the MACHO database. Due to its extremely long
duration (969 days) this is a parallax event and from its light curve it has been
possible to infer (statistically) the lens mass ]vI = 6!1° M8 and distance between
about 0.5 and 2 kpc. 4 The lens object in this microlensing event is therefore thought
to be most likely a black hole, since HST observations have practically ruled out

1702
 1703

the possibility of the lens being a main sequence star. If the lens is a black hole,
then it would accrete by the surrounding interstellar medium thus emitting X-rays.
To this aim, it as been recently conducted a very deep XMM observation for more
than 100 ks towards the MACHO-96-BLG-5 direction. The detection of the source
would have been the first observation of an isolated stellar-mass black hole.
Because of the long exposure, the XMM observation has been affected by flares
due to soft protons. After a usual screening of the row data a detailed analysis
has been performed in order to identify any possible X-ray signature towards the
MACHO-96-BLG-5 direction.
Unfortunately, the observation has shown no clear detection of an X-ray excess
towards the target. Only an upper limit of 9.1 x 10- 15 erg cm- 2 S-l to the X-ray
flux can be put, constraining the black hole accretion parameters (for more details
see 6 ).
Before closing this communication, we would like to emphasize once more the
importance of making any effort by the micro lensing community in performing
deep observations towards any microlensing or pixel-lensing event aiming to image
the lens object. To this aim, it would be perhaps useful to build up an up-dated
database - a specialized kind ov Virtual Observatory - available to the community,
with detailed information about the performed, planned and suggested observations
in any band of the electromagnetic spectrum towards any microlensing target.

References
1. C. Alcock et al., Nature 414, 617 (2001).
2. A. Gould, D. P. Bennett and D. R. Alves, Astrophysical Journal 614, 404 (2004).
3. N. Kallivayalil, preprint astro-ph/0609619 (2006).
4. D. P. Bennett et al., Astrophysical Journal 579,639 (2002).
5. A. A. Nucita et al., Proceedings of the Meeting Neutron Stars and Black Holes in Star
Clusters, 26th meeting of the IAU, Prague, Czech Republic, JD06, N. 15 (2006).
6. A. A. Nucita et al., preprint astro-ph/0608692, in press on Astrophysical Journal
(2006).
This page intentionally left blank
Theoretical Gravitational
Lensing
This page intentionally left blank
 GRAVITATIONAL LENSING BY BRANEWORLD BLACK HOLES

RICHARD WHISKER
Institute for Particle Physics Phenomenology, Department of Physics,
University of Durham, South Road, Durham, DHl 3£E, UK
r. s. [email protected]

Black holes in the braneworld scenario can be different from the classical black hole
solutions of general relativity, so it is important to investigate these solutions and the
observational signatures they could give rise to. In this contribution I discuss gravita-
tional lensing in the strong deflection limit for some candidate braneworld black hole
metrics.

1. Introduction
The Randall-Sundrum braneworld scenario 1 ,2 provides an interesting framework
within which to explore the possibility that there exist extra dimensions. We focus
on the second Randall-Sundrum model, 2 consisting of a single brane in an infinite
extra dimension. An elegant description of gravity in this model was provided by
Shiromizu, Maeda and Sasaki,3 who projected the 5D Einstein equations onto the
brane to obtain the effective 4D field equations:
(1)
Here, A4 is a cosmological constant and TJ1,v is the usual energy-momentum tensor
of matter on the brane. SJ1,v(T2) consists of squares of T,.LV and thus is a local, high
energy correction term. EJ1,v consists of the projection of the bulk Weyl tensor onto
the brane, and is non-local from the brane point of view.
The general static, spherically symmetric metric on the brane can be written as:
(2)
The vacuum brane field equations following from Eq. (1) are GJ1,V = -EJ1,v, In order
to close this system of equations and obtain a braneworld black hole (BBH) solution,
an assumption must be made about EJ1,v or gJ1,v' The assumption A2 = 1/ B2 leads
to the tidal Reissner-Nordstrom (RN) solution of Dadhich et al.: 4
-1
1 1
2
ds = - (
1- -
r
+ 2rq ) dt
2
+ (1- -
r
+ r2q )
dr
222
+r dDj[, (3)

where r is in units of 2GM. Unlike the standard RN solution, the 'tidal charge'
parameter q can take both positive and negative values. Another solution is the
so-called 'U = 0' solution of Ref. 5:
d 2 __ (r - rh)2 d 2 (r + 1/4)4 d 2 (7' + 1/4)4 dD2
S - (r + 1/4) 2 t + r4 r + r2 j[ . (4)

This has a turning point in the 'area function' C(r) at r = 1/4 (i.e. a wormhole
region) and the horizon at r = rh is singular, showing that BBHs could be radically
different from the standard Schwarzschild black hole! For a more detailed discussion
of BBHs, see Ref. 6.

1707
1708

2. Gravitational Lensing
The lensing setup is shown in Fig. 1. As the impact parameter u of a light ray
s I

Dos L

Fig. 1. Gravitational lensing diagram.

decreases, the deflection angle a increases. At some point, a exceeds 27f and the
photon performs a complete loop around the black hole. As ro approaches the radius
of the photon sphere r p , with corresponding impact parameter uP' the deflection
angle diverges. Bozza 7 has shown that this divergence is logarithmic for all black
hole metrics of the form of Eq. (2). Hence the deflection angle can be expanded
close to the divergence in the form

0'(8) = -aln (8D o t/up - 1) + b + O(u - up), (5)

where the strong deflection limit (SDL) coefficients a and b depend on the metric
functions evaluated at rp. The SDL coefficients for the two BBH metrics given in
Eqs. (3) and (4) were calculated in Ref. 8 and are shown in Fig. 2.

2.5

1.5
1.5
a

0.5
0.5
q
rh -0.2 -0.1 0.1 0.2
0.1 0.15 0.2 0.3 0.35 0.4
-0.5
-0.5 6 6

(a) (b)
Fig. 2. SDL coefficients for two BBH metrics. The vertical axis corresponds to the Schwarzschild
black hole in both cases. (a) The U = 0 metric. (b) The tidal-RN metric.
 1709

2.1. Observables
We focus on the simplest situation, where only the outermost 'relativistic image'
(formed by light rays that wind around the black hole once) is resolved as a single
image, with the remaining images (formed by light rays winding around the black
hole mulitple times) packed together at the angular radius of the photon sphere,
800 = Up/DoL. We define the observable 8 as the separation between the outermost
image and the others, and f as the flux ratio between the outermost image and
all the others. With further simplifying assumptions, it can be shown that these
observables are directly related to the SDL coefficients: 7 ,8
8 = 800 e(b-27r)/a, (6)
27r a
f = e / . (7)
Taking as a concrete example the black hole at the centre of our galaxy, with mass
M = 2.8xl0 6 M 8 and Dol = 8.5kpc, the observables 800 ,8 and f are calculated for
the two BBH metrics, as well as the Schwarz schild metric, in Table 1.

Table 1. Estimates for the lensing observables for the central black hole of our galaxy. 8= and s
are defined in Sec. 2.1 and given in J1 arc seconds, and 1m = 2.51og1 is 1 converted to magnitudes.

Sch. U = 0 metric Tidal RN metric

T'h=O.1 T'h=0.2 T'h=0.3 T'h=O.4 q=-0.2 q=-O.1 q=O.1 q=0.2

8= 16.87 13.24 15.65 18.11 20.62 18.85 17.92 15.64 14.07

s 0.0211 0.0303 0.0235 0.0192 0.0164 0.0150 0.0173 0.0286 0.0502
1m 6.82 6.13 6.61 7.01 7.32 7.26 7.08 6.44 5.70

3. Discussion
Unfortunately, the resolutions required for this type of observation are beyond
reach of current observational facilities. Nevertheless, Table 1 clearly shows that
BBHs could have significantly different observational signatures than the standard
Schwarz schild black hole. This encourages the investigation of more realistically ob-
servable situations, and an interesting possibility in this direction is the study of
the accretion disc emission from black holes.

References
1. L. Randall and R. Sundrum, Phys. Rev. Lett. 83, 3370 (1999).
2. L. Randall and R. Sundrum, Phys. Rev. Lett. 83,4690 (1999).
3. T. Shiromizu, K. 1. Maeda and M. Sasaki, Phys. Rev. D 62, 024012 (2000).
4. N. Dadhich et al., Phys. Lett. B 487, 1 (2000).
5. R. Gregory, R. Whisker, K. Beckwith and C. Done, leAP 0410, 013 (2004).
6. R. Whisker, Braneworld Black Holes, PhD Thesis, University of Durham (2006).
7. V. Bozza, Phys. Rev. D 66, 103001 (2002).
8. R. Whisker, Phys. Rev. D 71, 064004 (2005).
 GRAVITATIONAL LENSING OF STARS SURROUNDING
SUPERMASSIVE BLACK HOLES

V. BOZZA and L. MANCINI

Dipartimento di Fisica "E.R. Caianiello", Universitri di Salerno, Baronissi (SA), Italy
Istituto Nazionale di Fisica Nucleare, Sezione di Napoli, Italy

Many indications suggest that both the Milky Way and M31 host a super massive black
holes in their central regions. General Relativity tells us that these black holes, acting
as gravitational lenses, are able in principle to bend the light rays of the source stars
moving in their neighborhood. As a consequence of this fact, a secondary image and
two infinite series of relativistic images will be generated. In the case of our Galaxy, the
central supermassive black hole corresponds to the radio source Sgr A * and, thanks to
the observations in the near infrared band, it has been possible to determine the orbits
of several stars moving in the neighborhood of the Galactic center. In the framework
of Schwarzschild black hole, it is now possible to calculate the expected light curves for
the secondary and for the main relativistic images for these stars. In this way, it is easy
to predict the best times to observe the secondary images, which occur when the stars
approach the minimum distance from the black hole. Concerning M31, we draw some
perspectives for observations.

Keywords: black hole physics - gravitational lensing

1. Supermassive black holes

In the last decade the search for supermassive black holes has been remarkably
successful. In fact, dozens of excellent candidates have been identified, especially in
the nuclei of spiral galaxies. The presence of a supermassive black hole in the center
of our Milky Way is suggested by many clues related to the motions of stars and
gas in the vicinity of the radio source Sgr A *. Indeed, two independent groups have
completely reconstructed the orbits of individual stars around this object thanks to
high resolution NIR observations. 1,2
Also for M31 there are strong evidences for the existence of a supermassive
black hole in its center. Using telescopes on the ground and in space, astronomers
discovered two large clumps of stars that appear to form a double nucleus in M31.
Recently, new spectroscopic observations by Hubble Space Telescope reveal that the
two bright clumps that form this double nucleus are actually the brightest regions
of a disk of stars that completely encircle a supermassive black hole. 3 By measuring
the speeds at which these stars orbit the center of the galaxy, it has been possible to
deduce that the M31 central black hole is 10 times as massive as the central black
hole in the Milky Way.

2. Gravitational Lensing around Sgr A *

The system formed by the stars and the black hole at the Galactic center provides a
unique lab for gravitational lensing research, in that we have an evident advantage:
we precisely know the source location in space as a function of time and thus the
geometry of the lensing configuration. Since we already observe the direct image, at
any time we can predict where to look for a secondary lensing image and what its
1710
 1711

apparent magnitude should be. Thanks to our knowledge of the orbits of these stars
and the light curves, we can easily predict the best time to observe these secondary
images. Finally, since the distances between the sources and the lens are known, the
information extraction from gravitational lensing observations is much easier and
unambiguous.
Once we know the absolute magnitude and the position of a star around Sgr
A *, we can give the position and the apparent magnitude of its four most relevant
images. Following this path, we have developed a complete analysis of several stars
as potential sources for gravitational lensing by the central black hole. We found
very interesting results, especially for the star S14.

2.1. Star S14

This star represents the most interesting lensing candidate, since its orbit is almost
edge-on (i = 97.3 deg), and its eccentricity brings it very close to the black hole.
We refer the reader to the works of Bozza & Mancini 4 ,5 for the analysis of other
stars.
The apparent magnitudes in the K-band of its gravitational lensing images
change very slowly during the whole orbital period, save for a sharp peak corre-
sponding to the periapse epoch. This is easily understood as a consequence of the
fact that the magnification is inversely proportional to the square of the distance
between the lens and the source. The best alignment and anti-alignment times are
very close to the periapse epoch and are responsible for two subpeaks: first we have
the anti-alignment peak, then we have the alignment peak.
What makes S14 particularly interesting is the high brightness attained by the
secondary image. At the best alignment time the secondary image has K=23. This
is because we are close to a weak field gravitational lensing situation. The duration
of the main peak is about one month and is determined by the velocity of S14 at
the peri apse epoch.
The next brightness peak of the secondary image of S14 will occur in 2038, ac-
cording to the current estimates of the orbital parameters, and the maximal angular
distance from the apparent event horizon is about 0.125 mas. 5 The rapid motion of
this image would help to distinguish it from background sources of similar bright-
ness, while its distortion should also be an unambiguous sign of its gravitational
lensing nature with respect to fast moving stars around Sgr A *.
As regards the angular resolution in the K-band, the VLT units can be combined
to perform interferometry observations with an equivalent baseline of 200 m and
a maximal angular resolution of 2.2 mas. Some space missions performing nulling
interferometry (TPF, DARWIN) should be launched in the near future. According
to the mission designs, some spacecraft should fly in formation at distances of the
order of tens of meters. A futuristic development of such idea might lead to much
higher resolutions. The baseline needed for 0.1 mas resolution is of the order of
several kilometers. High precision formation flying may be achieved by laser ranging
1712

and microthrusters in the wake of what is being studied for LISA, in which the
distance between the spacecraft is 5 million km. Given the present situation and
all the technical advances that will soon be exploited in scientific researches, 0.1
mas resolution in the K-band and the detection of secondary gravitational lensing
images around Sgr A * seem just the next step after the generation of telescopes
currently under study.

3. Gravitational Lensing around M31

Our horne galaxy, the Milky Way, is part of a collection of about three dozen galaxies
known as the Local Group. These galaxies move through space as a single unit,
bound together by their mutual gravitational pull. The largest member of the Local
Group is the Andromeda Galaxy, M31.
Recent HST spectroscopy of the nucleus of M31 has identified three distinct
stellar populations near the center of M31, denoted as PI, P2 and P3. In particular,
the core of M31 is composed of a ring of old red stars (P2) and a disk of hot
young blue (A-type) stars (P3), yielding convincing evidence for the presence of a
central supermassive black hole in M31. Indeed, the young disk is trapped within
the gravitational field of a 1.4 x 10 8 Mo black hole. 3
From the point of view of lensing, the estimate of the probability distribution
that a single event is due to a particular source class, as a function of the dis-
tance form the black hole, as well as the number of events, as a function of the
secondary image threshold magnitude, tell us that few gravitational lensing events
are expected, essentially due to stars located in the bulge. For this purpose, we
need images with a long time-exposure in order to achieve an apparent magnitude
better than 26. The required resolution is of 0.012 arcsec. Another important fact
is that the duration of the events is very long. This means that, since we are in a
quasi-static situation, a VLT interferometer measure is sufficient to say if there are
some gravitational lensing effects by the M31 central supermassive black hole.

References
1. Eisenhauer F., Genzel R., Alexander T. et al., Astroph. J. 628, 246 (2005)
2. Ghez A.M., Salim S., Hornstein S. et al., Astroph. J. 620, 744 (2005)
3. Bender R., Kormendy J., Bower G., Astroph. J. 631, 280 (2005)
4. Bozza V. & Mancini L., Astroph. J. 611, 1045 (2004)
5. Bozza V. & Mancini 1., Astroph. J. 627, 790 (2005)
 KERR BLACK HOLE GRAVITATIONAL LENSING IN THE
STRONG DEFLECTION LIMIT: AN ANALYTICAL APPROACH

FABIANA De LUCA
Institut Jor theoretical Physics, University oj Zurich,
Zurich, Switzerland
* E-mail: [email protected]

The caustics, critical curves and Kerr black hole shadow structure are derived from the
lens equations solved by means of a perturbative approach and assuming large deflec-
tions. Starting from the Schwarzschild lens we will add terms up to the second order
in the lens spin. We will focus on the degeneracy, which arises at this order of approx-
imation, between the lens angular momentum and the height of the observer on the
equatorial plane that cannot be disentangled, no matter which observable is measured.
Some applications of the results will be described.

Keywords: Gravitational lensing.

1. Introduction

As predicted by General Relativity, photons passing near a black hole suffer devi-
ations from their original trajectory. If the minimum distance between photon and
black hole is much larger than the gravitational radius, a weak field approximation
of the metric tensor is sufficient to describe the light deflection. Two images of the
original source are then detected by the observer. On the other hand, photons pass-
ing very close to the black hole may suffer very large deviations without falling into
the black hole. These photons may perform one or more loops around the black
hole before reemerging in the observer direction, thus generating two infinite sets
of so-called relativistic images very close to the black hole shadow. It can be easily
guessed that these relativistic images represent a unique probe to gain information
on the very strong gravitational fields surrounding the black holes. Through their
study it would be possible to learn the properties of black holes and get new insight
on General Relativity in a strong field regime. The features of relativistic images
will thus represent a possible challenge arena for alternative theories of gravitation.

2. Kerr Geodesics
We consider 3 ,4 an observer at Boyer-Lindquist coordinates 2 (DOL, 1J o, CPo), with 1J
and cp respectively the colatitude and the azimuth, and introduce the new vari-
able f1.o = cos(1J o ) so that af1.o is the lens spin projection along the line of sight
observer-lens. All the distances will be measured in units of the Schwarzschild ra-
dius, 2G!VI / c2 , where M is the lens mass.
Lightlike geodesics are given in terms of radial and angular integrals 1 which must
be evaluated along the photon trajectories and the signs must be chosen so that
all the contributions sum up. The distance of closest approach between photon and
lens depends on the incoming trajectory, so a continous family of allowed closest

1713
1714

approaches, Xm , can be defined. In particular all the distances parametrized as

follows
3 2 ~ 4 2 2)
Xm = 2" - V3a~ y 1 - IL~ - ga (1 + ILa

_ 4 r.>a3~(5 + 6IL;h/l- IL~ + O(a 4 ), (1)

27v3
with ~ in the range [-1,1], are allowed closest approaches.

3. Results
3.1. Lens shadow
The shadow of a lens is defined as the locus of points in the observer sky where pho-
tons with minimum closest approach would show up while no gravitational lensing
images can appear inside this locus. We define angular coordinates on the observer
sky (8 1 , 82 ) centered on the black hole position and such that the spin axis of the
black hole is projected on the 82 -axis. Up to second order in a, the lens shadow can
be parametrized as follows

3V3 + ay~
D OL 81 ,m = --2-~
~ [5 -
1 - IL~(l + ~ 2) + a 2 3V3 2 2 2)]
2ILa - 2~ (1 - ILa ,

As to zero order 81 ,m ex -~ and 82 ,m ex y'1=12, one can identify ~ as the cosine

of the position angle in the (8 1 , 82 ) plane as taken from the opposite of the 81 -axis.
One can show that the previous equations satisfy the ellipse equation
2
(81,m -
8)2
0 + 82,m = 1 (2)
Ai A3
with the origin shifted rightward by 80 = 2aF
OL
and serniaxes given by
A 1 -- D- 1 -2- - v'3
OL (3v'3
2
a ) an d A 2 -_ D- 1 -2- - v'3Ma
OL (3v'3 a 2) . 2

The lens shadow is the first available observable but one cannot disentangle the lens
spin a from the inclination ILa but in the case one knows to a good accuracy DOL
so extracting a from a measure of A1 solely. However a affects A1 only at second
order and very high accuracy measures are needed in order to appreciate such a
small correction.

3.2. Caustics and critical curves

Considering corrections up to second order in a, one can show that, with respect
to the non rotating lens, caustics suffer a shift from the optical axis (first order
effect) and are not point-like anymore but have an extended, atroid-shaped structure
 1715

(second order effect). Introducing a set of observer-oriented coordinates (x, ¢) e,

centered at the black hole with the polar axis perpendicular to the optical axis, the
azimuth q; taken from the direction opposite to the observer and x being the radial
distance, one shows that the caustics are symmetric on the plane ee,
¢), they are
astroid-shaped and their semi-amplitude is

(3)

where k is directly connected to the number of loops performed by the photon

around the lens. Notice that, once again, the lens spin and the inclination fLo are
combined in the usual product aJI - fL; and for polar observers (fLo = I) the
caustic turns out to be point-like again. This analysis also showed that relativistic
images appear very near to the lens shadow with a separation that decreases as the
number of loops performed by the photon around the lens increases. This is due to
the shape of the critical curves; the higher is k and the more the critical curve tends
to the lens shadow. The aforementioned separation is of the order of fLas or less,
thus requiring a great sensitivity to detect relativistic images. Moreover, the image
magnification is exponentially decreasing with the number of loops, so the nearer
the image is to the lens shadow and the fainter it is. Neverthless, direct observation
of relativistic images seem to be possible.

4. Applications
Many X-ray sources have been detected around Sgr A * with a luminosity compara-
ble or even higher than that of the supermassive black hole. These sources are the
so called low mass X - ray binaries (LMXB). For such sources, the signal-to-noise
ratio on a pixel collecting radiation emitted by a relativistic image is

~= O.9x10 C~:s) (lO~~m) C~~u)

6
-1 -1 -1, (4)

where D LS is the distance lens-source, Rs is the linear extension of the LMXB

emitting region ans wp is the detection accuracy. If the latter is of the order of fLas,

of D LS and R" *
value that should be reached by the next realistic projects, and for reasonable values
should be high enough to allow direct detection of relativistic
images and get new insight on General Relativity in the strong deflection limit.
For more details about the topics discussed here, we refer to. 3 ,4

References
1. B. Carter, Phys. Rev. 174, 1559 (1968).
2. R.H. Boyer and R.W. Lindquist, Jour. of Math. Phys. 8, 265 (1967).
3. V. Bozza, F. De Luca, G. Scarpetta, M. Sereno, Phys. Rev. D 72, 083003 (2005).
4. V. Bozza, F. De Luca, G. Scarpetta gr-qc/0604093.
 ON GRAVITATIONAL LENSING BY A KERR BLACK HOLE

MAURO SERENO
Institut Jur Theoretische Physik, Universitiit Zurich,
Winterthurerstmsse 190, CH-80S7 Zurich, Switzerland
[email protected]

FABIANA DE LUCA
Institut Jur Theoretische Physik, Universitiit Zurich,
Winterthurerstmsse 190, CH-80S7 Zurich, Switzerland
and
Dipartimento di Fisica 'E.R. Caianiello', UniversitiL di Salerno,
via Allende, 1-84081 Baronissi (SA), Italy.
[email protected]

The gravitational deviation suffered by photons passing near massive compact

bodies provided one of the first observational tests of general relativity and is still
considered as an excellent probe for gravity theories. Black hole lensing has been
emerging as a pretty promising tool for gravitational investigations in both weak and
strong fields. On the observational side, interest in this topic is mainly motivated by
the super-massive black hole supposed to be hosted in the radio source Sgr A * in the
Galactic center. Planned high-resolution observations at the astrometric resolution
of the microarcsecond (/-Las) should allow in the next future a clean detection of
higher order effects in gravitational lensing. It is now well understood that a photon
passing near a black hole can suffer either a strong or a weak deflection. The latter
occurs when the minimum distance is much larger than the gravitational radius.
The former occurs when photons wind around the black hole making one or more
loops and producing images very near to the shadow.
Whereas lensing by either Schwarzschild or generic spherically symmetric black
holes has been extensively investigated, a full analytical description of lensing by a
Kerr Black hole is still missing. 1 An intrinsic angular momentum breaks the spheri-
cal symmetry heavily affecting the gravitational field. The separation of variables of
the Hamilton-Jacobi equation is an unique method to address the light propagation
near a rotating body. Either numerical studies or analytical investigations can be
used. 1 Studies in the weak deflection limit usually consider only the first correction
due to the angular momentum, the equations of motion for a light ray being eval-
uated up to and including second order corrections in m/b e alb, where m is the
black hole mass, a the angular momentum and b is the impact parameter. We took
a step forward and performed an analytical treatment of gravitational lensing by a
Kerr black hole in the weak deflection limit up to the next order. We considered
the standard gravitational lensing scenario in which both source and observer lie
in the asymptotically flat region of the spacetime and took care of expressing the
results in terms of the invariants of the light ray, avoiding ambiguities connected to
coordinate-dependent quantities. The crucial point in such an investigation is that
light like geodesics ca be expanded as a Taylor series up to and including third-order

1716
 1717

terms in m/b and a/b.

The lensing quantities were solved for with a standard perturbation technique.
This method allowed us to consider corrections proportional to a 2rsch and ar§ch.
We showed as pure spin terms ex: a 2 , a 3 do not contribute to the observable lensing
quantities, in particular to the deflection angle. Up to the first order correction in
the spin, the Kerr lens is equivalent to a displaced Schwarzschild deflector. This
is a very general property of spinning lenses. To the next order, this degeneracy
is broken and some particular features show up. The two perturbed images are
no more aligned with a fixed position. The degeneracy between the absolute value
of the spin and its inclination on the line of sight is also broken. All observable
quantities at the first order correction in the spin are functions of a sin 13 0 , with 13 0
the complement to the inclination angle of the spin axis, but terms proportional to
a cos 13 0 appear at the next order in the angular coordinates of the images in the
plane of the sky. However, the angular displacement of the images from the center
is still a function of a sin 130 •
Positions and magnifications of individual images are computed with a pertur-
bative analysis. At this order, the degeneracy with the translated Schwarzschild
lens is broken. The shape of the critical curve is still a circle displaced along the
equatorial direction and the caustic is still point-like. The finite size of the caustic
should show up at the next order due to terms ex: a 2r§ch.
It could be of interest to draw some comparison with the case of the strong
deflection limit. Such a limit has been treated considering small values of the angular
momentum and including corrections proportional to a 2 . That is two orders beyond
the Schwarzschild lens. This was enough to obtain finite shaped caustics. In our
study of the weak deflection limit, we made no assumptions on the absolute value
of the spin and still considered two orders beyond the spherically symmetric lens
but we did non get the caustic structure. This is only an apparent discrepancy, as
we have to remind that the minimum distance in the strong deflection limit is of
order of the gravitational radius. In fact the finite size of the caustic springs from
terms proportional to (a 2r§ch) / r;;'in. Since in the strong deflection limit rmin '"
rSch, we see as these terms are included in an analysis up to the second order in
a.
If the supermassive black hole at the Galactic center has a significant angu-
lar momentum, some features of Kerr lensing could be detected by future space
astrometric mission with a planned resolution of the microarcsecond.

Acknowledgments
M.S. is supported by the Swiss National Science Foundation and by the Tomalla
Foundation. F.D.L.'s work was performed under the auspices of the EU, which has
provided financial support to the 'Dottorato di Ricerca Internazionale in Fisica
1718

della Gravitazione ed Astrofisica' of the Salerno University, through 'Fondo Sociale

Europeo, Misura IlI.4'

References
1. M. Sereno and F. De Luca, Phys. rev. D, in press; arXiv:astro-ph/0609435 (2006)
 TESTING THEORIES OF GRAVITY WITH BLACK HOLE
LENSING

CHARLES R. KEETON
Department of Physics €3 Astronomy, Rutgers University
136 Prelinghuysen Road, Piscataway, NJ 08837 USA
[email protected]

A.O.PETTERS
Departments of Mathemathics and Physics, Duke University
Science Drive, Durham, NC 27708 USA
[email protected]

The gravitational deflection of light provided one of the first observational confirmations
of general relativity. Now we are considering how gravitational lensing can provide novel
tests of Einstein's theory, and intriguing alternatives. We have developed a comprehen-
sive analytical framework for lensing by black holes, and made concrete predictions that
are testable with existing or planned instruments. Two examples: (1) In parametrized
post-Newtonian models, there are universal relations among lensing observables. Ob-
served violations of these relations would falsify all PPN models in one fell swoop. (2) In
braneworld gravity, there could be many primordial black holes in our Solar System that
would produce interference fringes in the energy spectra of gamma ray bursts, which
could be detected with the GLAST satellite starting in 2007.

1. Introduction
In conventional cosmology, we comprehend the universe by extrapolating Einstein's
theory of general relativity from the Solar System (where it is well tested) 1 to
many different length scales and gravity regimes. General relativity succeeds in
explaining many observations of galaxies and cosmology, but only if we postulate
exotic dark matter and dark energy.2-4 Some have questioned the supposition that
we know so much about gravity but so little about the substance of the universe.
Various modifications to Einstein's theory have been proposed, either to allow simple
generalizations, or to create a theory that incorporates ideas from the search for a
quantum theory of gravity, or specifically to build a cosmology that does not require
dark matter and/or dark energy.l,5,6
Black holes provide extreme environments in which gravity is strong and dif-
ferences between theories are amplified. Strong gravity effects have been inferred
from the motions of gas around black holes,7,8 but such observations are indirect
and complicated by non-gravitational physics. The gravitational deflection of light
by black holes represents a cleaner phenomenon. While lensing by black holes has
not yet been observed, we predict that it can be detected if one knows what to look
for. Our recent work 9- 11 reveals exciting new opportunities for testing general rela-
tivity and related gravity theories, and for determining whether space has a fourth
dimension as postulated in some alternative theories, using observations that are
accessible to current or planned instruments.

1719
1720

2. Parametrized Post-Newtonian (PPN) Theories of Gravity

We are building a new mathematical theory of lensing by black holes in general
gravity theories. Briefly, we analyze how light propagates through the spacetime
around a black hole, and compute the observable properties of the lensed images
that are created. (Predicting observables distinguishes our work from many previous
studies.) We have completed 9 ,10 a rigorous and comprehensive study of lensing by
static, spherical black holes in the parametrized post-Newtonian (PPN) family of
gravity theories, representing generalizations of general relativity. One important
result is a precise, analytic description of the relation between the spacetime ge-
ometry around the black hole and lensing observables. The second surprising result
is the discovery of new fundamental relations among lensing observables. These re-
lations tell us which combinations of observables reveal strong gravity effects, and
which ones test the entire PPN model framework (see ref. [10] for details).
One use of the formal theory is to identify astrophysical systems in which it
will be possible to measure strong gravity effects. We have found that lensing tests
of gravity theories are within the reach of current technology. For example, timing
measurements of pulsars or X-ray binaries could be used to observe lensing by
supermassive black holes in our own galaxy or nearby galaxies. 9 ,10 Current work will
allow us to study the brightness fluctuations when hotspots in accretion disks around
spinning black holes are lensed. These observational feasibility studies will help us
design realistic observational programs for existing and planned instruments. They
play a crucial role in connecting mathematical formalism with real astrophysical
systems and astronomical observations, which is one of the hallmarks of our research.

3. Braneworld Gravity
One alternative gravity theory that deserves special mention is braneworld gravity.
In a model developed by Randall and Sundrum, 6 familiar 4-dimensional spacetime is
actually a submanifold (a brane) in a 5-dimensional universe. The model is inspired
by concepts from string theory, and designed to explain why gravity is so much
weaker than the other forces. We have discovered a new way to test the Randall-
Sundrum braneworld model with gravitational lensing.
The test involves small black holes thought to have been created in the early
universe. In general relativity, black holes smaller than about 10- 19 M(o) '" 10 14 g
would have evaporated by now due to Hawking radiation. 12 In the Randall-Sundrum
model, by contrast, the evaporation rate is lower,13 so black holes with masses as
small as 1000 grams may be able to survive to the present. 14 We have found that
such primordial braneworld black holes could be detected via gravitational lensing. 11
Specifically, a primordial braneworld black hole would bend passing gamma rays to
create an interference pattern, in a phenomenon we call "attolensing" (see Fig. 1).
The interference pattern would be detectable in the energy spectra of gamma ray
bursts observed with the Gamma-ray Large Area Space Telescope (GLAST), sched-
uled to be launched in 2007. 15 ,16 It is exciting that gravitational lensing provides
 1721

'&1 (a)
t>D
2
-
0
' --"
'0
"---
Z 1
'0
t>D
.3 0

w
t>D
2
-
0
' --"
'0
"---
Z 1
'0
t>D
.3 0

1.5 2 2.5 3
log E [MeV]

Fig. 1. Sample energy spectrum of a gamma ray burst attolensed by a braneworld black hole
with mass 10- 18 lvIr:). (a) The dashed lines shows the intrinsic energy spectrum, while the dotted
line shows the attolensed spectrum. (b) The histogram shows a sample GLAST "observation,"
assuming 100 counts with energies between 30 MeV and 1 GeV.

a real opportunity to make measurements in the next few years that will test the
speculative but intriguing braneworld hypothesis.

References
1. C. M. Will, Astrophys. Sp. Sci., 283:543 (2003)
2. A. G. Riess, et al., Astron. J., 116:1009 (1998)
3. S. Perlmutter, et al., Astrophys. J., 517:565 (1999)
4. D. N. Spergel, et al., astro-ph/0603449 (2006)
5. S. M. Carroll, V. Duvvuri, M. Trodden, & M. S. Turner, Phys. Rev. D70:043528 (2004)
6. L. Randall & R. Sundrum, Phys. Rev. Lett., 83:4690 (1999)
7. K. Nandra, et al., Astrophys. J., 477:602 (1997)
8. R. Shafee, et al., Astrophys. J. Lett., 636:L113 (2006)
9. C. R. Keeton & A. O. Petters, Phys. Rev. D72:104006 (2005)
10. C. R. Keeton & A. O. Petters, Phys. Rev. D73:044024 (2006)
11. C. R. Keeton & A. O. Petters, Phys. Rev. D73:104032 (2006)
12. D. N. Page, Phys. Rev. D13:198 (1976), Phys. Rev. D14:3260 (1976)
13. R. Guedens, D. Clancy, & A. R. Liddle, Phys. Rev. D66:043513 (2002), Phys. Rev.
D66:083509 (2002)
14. A. S. Majumdar, Phys. Rev. Lett., 90:031303 (2003); A. S. Majumdar & N. Mukherjee,
Intl. J. Mod. Phys. D14:1095 (2005)
15. http://glast.gsfc.nasa.gov
16. N. Omodei, et al., astro-ph/0603762 (2006)
GRAVITATIONAL LENSING BY HIGHER DIMENSIONAL BLACK
HOLES

ARCHAN S. MAJUMDAR and NUPUR MUKHERJEE

S. N. Bose National Centre for Basic Sciences,
Block JD, Sector III, Salt Lake, Kolkata 700098, India
[email protected]

We consider black hole solutions inspired by higher dimensional string and braneworld
models. The deflection of light in these geometries is modified compared to that in the
Schwarzschild metric. Expressions for the strong-field gravitational lensing observables
are derived in terms of the metric coefficients. Numerical estimates for these observables
are obtained for the galactic centre black hole for various candidate geometries.
The bending of light due to the gravitational potential of a massive object is
one of the first predictions of the general theory of relativity, and gravitational
lensing 1 in the weak field limit is one of the most widely used tool in astrophysics
and cosmology. On the other hand, strong field gravitational lensing, though limited
in observational utility because of presently inadequate instruments, remains our
ultimate scope for exploring the physics of strong gravitational fields, such as the
modifications of the standard general relativistic field theory entailed in higher
dimensional string theoretic and braneworld models.
Strong gravitational lensing is endowed with richer phenomenological features
like relativistic images 2 and retrolensing,3 and the general technique for analysing
strong gravitational lensing for spherically symmetric metrics was developed by
Bozza. 4 Analysis of gravitational lensing for various black hole solutions derived
from higher dimensional metrics such as the string-inspired dilaton black hole,5
braneworld black holes,6 and the Myers-Perry 5-dimensional black hole 3 ,7 have been
undertaken. The latter solution represents primordial braneworld black holes that
could grow in size due to accretion of radiation, and consequently survive till much
later stages in the evolution of the universe. s Recently there have been further
developments in the formalism used for studying strong lensing. 9
In this work we consider the supermassive black hole at the centre of our galaxy
and present the values of strong lensing observables for a braneworld black hole with
tidal charge,lO and also for a string theoretic dilaton black hole in an accelerating
universeY A comparitive estimate of the lensing observables is also provided for
the Schwarzschild and the Reissner-Nordstrom geometries.
The general formalism for strong gravitational lensing as developed by Bozza 4
considers a spherically symmetric metric
ds 2 = -A(r)dt 2 + B(r)dx 2 + C(r) (d0 2 ) (1)
The largest positive solution of the equation (C'(r))j(C(r)) (A'(r))j(A(r)) is
defined to be the photon sphere rph. A photon emanating from a distant source
and having an impact parameter u will approach near the black hole at a minimum
distance ro before emerging in a different direction. The deflection angle of the
photon can be expressed in terms in terms of the distance of closest approach.

1722
 1723

Decrease of the impact parameter to a minimum value Urn corresponding to the

distance of closest approach ro = rph results in the divergence of the deflection
angle, i.e., the photon is captured by the black hole. The deflection angle can be
written as a function of B = u/ Dd (where B is the angular separation of the image
from the lens, and Dd is the distance between the lense and the observer) as
BDd )
a( B) = -0; log ( Urn - 1 + -b (2)

where the expressions for 0; and b are obtained in terms of the metric coefficients
by extracting out the divergent part of the deflection angle. 4
In strong lensing there exist n relativistic images given by the number of times
a light ray loops around the black hole. The positions of these are obtained as
solutions of the lense equation 2 given by

Ds [tanB + tan(a - B) 1
tan6 = tanB - Dds (3)
with 6 representing the angular position of the source. When n ----+ 00 an observable
Boo can be defined 4 representing the asymptotic position approached by a set of im-
ages. The minimum impact parameter can then be obtained as Urn = DdBoo. In the
simplest situation where only the outermost image B1 is resolved as a single image,
while all the remaining ones are packed together at Boo, two lensing observables can
00
be defined as S = B1 - Boo and R = (tLd/[ I: tLn], with tLn the magnification of the
n=2
n-th image. In terms of the parameters 0; and b, these observables are given by4
S- B - CX)
i/
a - 271)a . ~ - e 27r / a , ,"-'- (4)
These two observables can be used to reconstruct the deflection angle.
We now consider a particular braneworld black hole with negative tidal charge
Q having the line element lO

ds 2 = - [ 1 - -2M Q]
+ -r2 dt 2 + [ 1 - -2M Q] -1 dr 2 + r 2dD- 2
+ -r2 (5)
r r
The photon sphere and the minimum impact parameter are given by
(3 + y'9 - 32Q) (3 + y'9 - 32Q)2
(6)
rph = 4 ; Urn = 4v'2J3 _ 8Q + y'9 - 32Q
in units of distance 2M, and the expressions for the lensing observables can be
derived in terms of the metric coefficients. 4,6
We next consider the dilaton-de Sitter black hole metric given byll

ds 2 = -[1- ~ -
r
~
r2(1- )H2]dt2+[1 -
r r r
~
- r2(1 - )H 2 2 + r2(1- )dD-2(7)
r
~ ]i/ ~
where H is the Hubble parameter measured in Schwrzschild radius units, and ~ =
Q;~:o (¢Yo corresponds to the minimum of the dilaton potential l l ). The photon
1724

sphere and the minimum impact parameter are given respectively by12
3+~+7]
rph = 4
(3 + ~ + 7]))3 + 7] - 3~
Urn = -'==7=======7=~~~~~~==~~~==~~======~ (8)
J15(3 + ~ + 7]) - 54 - H2(3 + ~ + 7])3 + 4~H2(3 + ~ + 7])2
where 7] = J9 -
10~ + ~2. The coefficients a and b in the deflection angle, and the
lensing observables can be derived in terms of the metric coefficients. 12

Table 1. Estimates for the strong field lensing observables for the black hole at the center of
our galaxy with mass M = 4.3 x 10 6 lvl0 for different spacetime geometries with Rs = 2M as
Schwarzschild radius. We have taken Q = O.lM, CPo = 0.1, H ~ 1O-26 sec, and Tm = 2.5logR.
Observables Schwarzschild Brane tidal charge Dilaton-de sitter Reissner- Nordstrom
(Joo (f.L arc sec) 27.505 29.09 27.42 27.47
S (f.L arc sec) 0.034 0.024 0.0338 0.038
Tm 6.81 7.02 6.8184 6.81
um/Rs 2.6 2.75 2.592 2.597

In Table 1 we provide the numerical estimates of the observables Boo, S, and r m ,

corresponding to the braneworld black hole with tidal charge and also the dilaton-de
Sitter black hole. These observables are computed for the galactic centre black hole
(taking its mass 4.3 x 10 6 MO and its distance from us 8kpc). For comparison we
also furnish the estimates of these observables for the Schwarzschild as well as the
Reissner-Nordstrom metrics. In order to detect the existence of relativistic images a
high order sensitive instrument with very long baseline interferometry is required. If
future experiments can attain O.01fLarcsec resolution, and if such relativistic images
are detected, then strong gravitational lensing may be able to distinguish between
the predictions of different black hole models and probe deeply the character of
gravity in the strong field regime.

References
1. P. Schneider, J. Ehlers and E. E. Falco, Gravitational Lenses, (Springer, 1992).
2. K. S. Virbhadra and G. F. R. Ellis, Phys. Rev. D62, 084003 (2000); K. S. Virbhadra
and G. F. R. Ellis, Phys. Rev. D65, 103004 (2002);
3. D. E. Holtz and J. A. Wheeler, Astrophys. J. 587, 330 (2002); E. F. Eiroa and D. F.
Torres, Phys. Rev. D69, 063004 (2004).
4. V. Bozza, Phys. Rev. D66, 103001 (2002).
5. A. Bhadra, Phys. Rev. D67, 103009 (2003).
6. R. Whisker, Phys. Rev. D71, 064004 (2005); A. S. Majumdar and N. Mukherjee, Int.
J. Mod. Phys. D14, 7 (2005).
7. V. Frolov and D. Stojkovic, Phys. Rev. D68, 064011 (2003).
8. A. S. Majumdar, Phys. Rev. Lett. 90, 031303 (2003).
9. C. R. Keeton and A. O. Petters, Phys. Rev. D73, 044024 (2006).
10. N. Dadhich et al., Phys. Lett. B487, 1 (2000).
11. C. J. Gao and S. N. Zhang, Phys.Rev. D70, 124019 (2004).
12. N. Mukherjee and A. S. Majumdar, eprint astro-ph/0605224.
 IRON Ko. LINE PROFILES AND SHADOW SHAPES AS
EVIDENCES OF GRAVITATIONAL LENSING IN A STRONG
GRAVITATIONAL FIELD NEAR BHs*

ALEXANDER F. ZAKHAROV
National Astronomical Observatories of Chinese Academy of Sciences, 20A Datun Road,
Chaoyang District, Beijing, 100012, China;
Institute of Theoretical and Experimental Physics, 117259, Moscow, Russia;
Bogoliubov Laboratory for Theoretical Physics, JINR, 141980 Dubna, Russia;
Center of Advanced Mathematics and Physics, National University of Science and Technology,
Rawalpindi, Pakistan
zakharov@itep. ru

FRANCESCO DE PAOLIS, ACHILLE A. NUCITA, GABRIELE INGROSSO

Department of Physics, University of Leece and INFN, Section of Leece, Via Arnesano, 1-73100
Leece, Italy

Based on ray-shooting method we simulate Fe Ka. line profiles discovered in microquasars

and Seyfert galaxies. Using narrow disk model for the emission line we describe transfor-
mations of characteristic two-peak profiles into one-peak profiles. In 2007 radio telescope
Radioastron will be launched at a high elliptical orbit and it will be used as a space dish
for VLBI observations with space-ground base. This interferometer will have extraordi-
nary angular resolution (10- 6 as). We discuss a procedure to measure masses, charges
and spins analyzing these forms of mirages (shadows). Actually, that is a manifestation
of gravitational lens effect in the strong gravitational field near black hole horizon.
Comprehensive reviews summarize the detailed discussion of theoretical aspects
of possible scenarios for generation of broad iron lines in AGNs 1 (an influence of
micro lensing on Fe Kcx line shapes and spectra was discussed 2 ). An ray shooting
approach was described earlier. 3 The approach was used in particular to simulate
spectral line shapes and it was based on results of qualitative analysis. 4 There is a
detailed of the analysis in the framework of the simple model for radiating annuli. 5
Recently it was proposed to use VLBI technique to observe mirages around mas-
sive black holes and in particular, towards the black hole at Galactic Center. 6 The
boundaries of the shadows are black hole mirages. We use the length parameter
OM .
7"g = = 6 X 1011 cm for the black hole at Sgr A* and analytIcal approach to
-2-
c
calculate shadow sizes, as it was explained in the text. By taking the distance of
Sgr A* to be DGC = 8 kpc, the length 7"g corresponds to angular size r-v 5 f.1as.
Since the minimum arc size for the considered mirages are about at least 4 7" g, the
standard Radioastron fringe size of about 8 I.WS is comparable with the required
precision. In principle, it is possible to evaluate the black hole charge Q by ob-
serving the shadow size. The mirage size difference between the extreme charged
black hole and Schwarzschild black hole case is about 30% (the mirage diameter for
Schwarzschild black hole is about 10.4 and for the extreme charged black hole the

* This research has been partially supported by the National Natural Science Foundation of China
(NNSFC) (Grant # 10233050) and National Basic Research Program of China (2006CB806300).

1725
1726

diameter is equal to 8 or (in black hole mass units)) and typical angular sizes are
about rv 52 IWS for the Schwarzschild and rv 40 /-Las for the Reissner-Nordstrom
black hole cases, respectively.6 Therefore, for Sgr A * a charged black hole could be
distinguished by Schwarzschild black hole with Radioastron, at least if its charge is
close to the maximal value. For stellar mass black holes we need a much higher an-
gular resolution to distinguish charged and uncharged black holes since the typical
shadow (mirage) angular sizes are about 2 x 1O-5/-Las, even for galactic black holes.
In principle, measuring the mirage shapes one could evaluate the black hole mass,
inclination angle (e.g. the angle between the black hole spin axis and line of sight)
and spin if a distance toward the black hole is known. For example, for the black
hole at the Galactic Center the mirage size is rv 52 /-Las for the Schwarzschild case.
In the case of a Kerr black hole, 6 the mirage is deformed depending on the black
hole spin a and on the angle of the line of sight, but its size is almost the same.
In the case of a Reissner-Nordstrom black hole its charge changes the size of the
shadows up to 30 % for the extreme charge case. Therefore, the charge of the black
hole can measured by observing the shadow size, if the other black hole parameters
are known with a sufficient precision. In general, one could say that a measure of
the mirage shape (in size) allows to evaluate all the black hole "hairs".
Few years ago the possibility of observing images of distant sources around black
holes in the X-ray band was discussed 7 by using X-ray interferometer. Indeed, the
aim of the MAXIM project is to realize a space based X-ray interferometer capable
of observing with angular resolution as small as 0.1 /-Las.
In spite of the difficulties of measuring the shapes of shadow images, to look at
black hole "faces" is an attractive challenge since mirages outline the "faces" and
correspond to fully general relativistic description of the region nearby the black hole
horizon without any assumption about a specific model for astrophysical processes
around black holes (of course we assume that there are sources illuminating black
hole surroundings). There is no doubt that the rapid growth of observational facil-
ities will give a chance to measure the mirage shapes using not only Radioastron
facilities but also other instruments and spectral bands, like the X-ray interfer-
ometer MAXIM, the Radioastron mission or other space based interferometers in
millimeter and sub-millimeter bands.

References
1. A.C. Fabian, K. Iwasawa, K.S. Reynolds, A.J. Young, Publ. Astron. Soc. Pacific 112,
1145 (2000).
2. L. Popovic, P. Jovanovic, E. Mediavilla et al. Astrophys. J. 637620 (2006).
3. A.F. Zakharov, Mon. Not. Roy. Astron. Soc. 269, 283 (1994); A.F Zakharov, On the
Hot Spot near a Kerr Black Hole, in Proc. of the 17th Texas Symposium on Relativistic
Astrophysics, eds. H. Bohringer, G.E. Morfill, J.E. Trumper, (Ann. NY Academy of
Sciences, 1995), 759, 550; A.F. Zakharov, S.V. Repin, Astron. Rep. 43, 705 (1999).
A.F. Zakharov, S.V. Repin, The shape of the Ka-line as a possible indication of the
black hole existence, in Proc. of the XXIV International Workshop on High Energy
Physics and Field Theory "Fundamental Problems of High Energy Physics and Field
 1727

Theory", ed. V.A. Petrov, (Institute for High Energy Physics, Protvino, Russia, 2001)
p. 99; A.F. Zakharov, S.V. Repin,Astron. Rep. 46, 360 (2002); A.F. Zakharov, S.V. Re-
pin, The shape of the Ka line of iron as the evidence of the black hole existence, in
Proc. of the Eleven Workshop on General Relativity and Gravitation in Japan, eds.
J. Koga, T. Nakamura, K. Maeda, K. Tomita (Waseda University, Tokyo, 2002), p. 68;
A.F. Zakharov, S.V. Repin, The shape of iron Ka line as the evidence for the black hole
existence in Seyfert Galaxies, 2002c, in Proc. of the XXXVIIth Rencontres de Moriond
"The Gamma-ray Universe", eds. A. Goldwurm, D.N. Neumann, J. Tran Thanh Van
(The GIOI Publishers, 2002) p. 203; A.F. Zakharov, S.V. Repin, Astron. 8 Astrophys.
406, 7 (2003); A.F. Zakharov, Publ. of the Astron. Observatory of Belgrade, 76, 147
(2003); A.F. Zakharov, The iron Ka line as a tool for analysis of black hole param-
eters, in The Physics of Ionized Gases, eds. L. Hadzievski, T. Gvozdanov, N. Bibic,
(AlP Conf. Proc. 2004), 740, p. 398; A.F. Zakharov, Intern. J. Mod. Phys. A 20,
2321 (2005); A.F. Zakharov, Massive Black Holes: Theory vs. Observations, in Pmc.
of the Helmholtz International School and Workshop "Hot Points in Astrophysics and
Cosmology", eds. V.B. Belyaev, D. Blaschke, (JINR, Dubna, Russia, 2005), p. 332;
A.F. Zakharov, N.S. Kardashev, V.N. Lukash, S.V. Repin, Mon. Not. Roy. Astron.
Soc. 342, 1325 (2003).
4. A.F. Zakharov, Sov. Phys. - J. Exper. Theor. Phys. 64, 1 (1986); A.F. Zakharov, Sov.
Phys. - J. Exper. Theor. Phys. 68, 217 (1989).
5. A.F. Zakharov, S.V. Repin, Mem. S. A. It. Suppl. 7, 60 (2005); A.F. Zakharov,
S.V. Repin, New Astron. 11, 405 (2006); A.F. Zakharov, Phys. Atom. Nucl., 70,
159 (2007).
6. A.F. Zakharov, A.A. Nucita, F. De Paolis, G. Ingrosso, New Astronomy 10, 479
(2005); A.F. Zakharov, A.A. Nucita, F. De Paolis, G. Ingrosso, Retro gravitational
lensing for Sgr A * with Radioastron, in Proc. of the 16th SIGRA V Conference on Gen-
eral Relativity and Gravitational Physics, eds. G. Vilasi, G. Esposito, G. Lambiase,
G. Marmo, G. Scarpetta, (AlP Conference Proceedings, 2005) 751, p. 227; A.F. Za-
kharov, A.A. Nucita, F. De Paolis, G. Ingrosso, Observational Features of Black Holes,
in Proc. of the XXVII Workshop on the Fundamental Problems of High Energy and
Field Theory, ed. V.A. Petrov (Institute for High Energy Physics, Protvino, 2005)
p. 21; gr-qc/0507118; A.F. Zakharov, A.A. Nucita, F. De Paolis, G. Ingrosso, Mea-
suring parameters of supermassive black holes, in Proc. of XXXXth Rencontres de
Moriond "Very High Energy Phenomena in the Universe", eds. J. Tran Thanh Van
and J. Dumarchez, (The GIOI Publishers, 2005) p. 223; A.F. Zakharov, A.A. Nucita,
F. De Paolis, G. Ingrosso, Shadows (Mirages) Around Black Holes and Retro Gravi-
tational Lensing, in Proc. of the 22nd Texas Symposium on Relativistic Astrophysics
at Stanford University, SLAC-R-752, eds. P. Chen, E. Bloom, G. Madejski, V. Pet-
rosian, SLAC-R-752, eConf:C041213, http://www.slac.stanford.edu/econf/C041213,
paper 1226 (2005); A.F. Zakharov, F. De Paolis, G. Ingrosso, A.A. Nucita, Astmn. 8
Astrophys. 442, 795 (2005); A.F. Zakharov, A.A. Nucita, F. De Paolis, G. Ingrosso,
Shadow Shapes around the Black Hole in the Galactic Centre, in Pmc. of "Dark Mat-
ter in Astro- and Particle Physics" (DARK 2004), eds. H.V. Klapdor-Kleingrothaus
and D. Arnowitt, (Springer, Heidelberg, Germany, 2005), p. 77; A.F. Zakharov, F. De
Paolis, G. Ingrosso, A.A. Nucita, Measuring the black hole parameters from space,
in Gravity, Astrophysics, and Strings'05, Proc. of the 3rd Advanced Workshop, eds.
P.P. Fiziev and M.D. Todorov, (St. Kliment Ohridski University Press, Sofia, 2006)
p.290.
7. W. Cash, A. Shipley, S. Osterman, M. Joy, Nature, 407, 160 (2000); N. White, Nature
407, 146 (2000).
 LENSING EFFECTS ON GRAVITATIONAL WAVES IN A
CLUMPY UNIVERSE

CHUL-MOON YOO*, KEN-ICHI NAKAOt and HIROSHI KOZAKI+

Department of Mathematics and Physics, Graduate School of Science, Osaka City University,
Osaka 558-8585, Japan
* [email protected]
t [email protected]
t [email protected]

RYUICHI TAKAHASHI
Division of Theoretical Astronomy, National Astronomical Observatory of Japan, Mitaka, Tokyo
181-8588, Japan
[email protected]

The distance-redshift relation via gravitational waves in the clumpy universe is sim-
ulated numerically by taking account of the effects of gravitational lensing. It is assumed
that all of the matter in the universe takes the form of randomly distributed point
masses, each of which has the identical mass ML. Calculations are carried out in the
two extreme cases: .\ » GML/C 2 and .\ « GML/C 2 , where .\ denotes the wavelength
of gravitational waves. In the former, the distance-redshift relation for the fully homo-
geneous and isotropic universe is reproduced with small distance dispersion, whereas, in
the latter, the distance dispersion is larger. This result suggests that we might obtain the
information about the typical mass of lens objects through the distance-redshift relation
gleaned through observation of gravitational waves of various wavelengths.

1. Introduction
Owing to recent developments of observational techniques, it is expected that de-
tection of gravitational waves will be achieved in the near future. In preparation
for future breakthroughs, we have to discuss the information provided by gravita-
tional waves data. Here, we focus on inhomogeneities of our universe that can be
investigated with gravitational wave observations.
Many cosmological observations suggest that our universe is globally homoge-
neous and isotropic. However, our universe is locally inhomogeneous. It is difficult to
directly observe the inhomogeneities with optical observations because most of the
matter in our universe is not luminous. One useful means of obtaining information
about aspects of inhomogeneities is to examine gravitational lensing.
There are some candidates for the dark matter such as weakly interacting mas-
sive particles (WIMPs) and massive compact halo objects(MACHOs). In this pa-
per, we show that it is possible to extract information about the properties of the
macroscopic compact objects from the observational data of gravitational waves by
analyzing the gravitational lensing effects due to these compact objects. For this
purpose, we consider an idealized model of the inhomogeneous universe which has
the following properties. First, this model is a globally Friedmann-Lemaitre uni-
verse. Second, all of the matter takes the form of point masses, each of which has
the identical mass lvh. Finally, the point masses are uniformly distributed.

1728
 1729

One of the main targets of gravitational-wave astronomy is the binary of com-

pact objects. Then, we need wave optics for analyzing the gravitational lensing of
gravitational waves since the wavelength of gravitational waves may be comparable
to or longer than the Schwarzschild radii of lens objects. When we consider gravita-
tionallensing effects on the gravitational waves with G M L / c2 ,:S A, we have to take
account of the wave effect.5 In fact, remarkable differences between the extreme
cases A» GML /C 2 and A« GML /C 2 have already been reported. s
In this paper, we focus on the relation between the distance from the observer to
the source of the gravitational waves and its redshift. We assume that the redshift of
each source is independently given by the observation of the electromagnetic coun-
terpart 3 or the waveform of the gravitational waves. 4 The distance is determined
using the information contained in the amplitudes of the gravitational waves from
the so-called standard sirens. 1 Although there are several similar analyses via the
optical observation of Type Ia supernovae,2 gravitational lensing effects on gravita-
tional waves can add new information about the properties of the lensing compact
objects to that obtained by the optical observations by virtue of the wave effect; we
can gain an understanding of the typical mass and number density of the lensing
compact objects.

2. The clumpy universe model and the calculation method

In order that the clumpy universe is the same as a homogeneous and isotropic
universe from a global perspective, the point masses must be distributed uniformly.
In this paper we use the so-called "the on-average Friedmann universe" ,67
In the case of A» GML/C 2 , the amplification factor of the single lens case is
almost equal to unity because of the diffraction effects. 5 Thus we assume that a wave
can be treated as a spherical wave whose center is at the source during all stages
of the propagation. In this approximation, we can calculate the total amplification
due to lensing, multiplying the amplification factor of the single lens case. 9
In the case of A « G M L / c2 , we adopt the geometrical optics approximation. The
observed waveform which is given by superposition of waves from each image have
received the amplification and the phase shift. In this paper, we assume stationary
configurations and the phase shift is not an observational variable. Therefore, we
can observe only the amplitude of the superposed waves, and we cannot identify
the signal as superposed one. In order to calculate the amplification factor, we use
the multiple lens-plane method. 6

3. Results
In the case of the long wavelength A » GML/C 2 , the magnification due to a single
lens is always very small, but is not negligible. The total magnification in the mul-
tiple lens system is determined by the cumulation of all of the magnification effects
due to each lens system and is not small. We found that the relation between the
average distance and redshift depends on the ratio AC2 /GML and it approaches the
1730

distance-redshift relation of the global universe, as AC2 /GML increases. Further,

we found that the distance dispersion also depends on the ratio AC2 /GML and it
approaches zero as AC2 /GM L increases. The distance-redshift relation in the global
universe is found even in the clumpy universe if the wavelength of gravitational
waves is much longer than the Schwarzschild radius of the lens objects.
In the case of the short wavelength A « G ML / c2 , the geometrical optics approx-
imation is valid. We found that the distance dispersion is larger than that in the case
of the long wavelength. The total magnification comes from the focusing and inter-
ference effects on the ray bundles which emanate from the source to the observer.
In this case, diffraction effects are negligible and thus the large magnifications due
to the focusing effects on the ray bundles are possible. Moreover, interference effects
enhance the magnifications or cause demagnifications.

4. Conclusion
The lensing effects cause the distance dispersion in the distance-redshift relation
obtained by the data of the gravitational waves. Making use of the properties of this
distance dispersion, we can probe the inhomogeneities of our universe. The distance
dispersion is sensitive to the ratio GML /AC 2 and therefore our result suggest that
we might obtain the information about the typical mass of clumps which cause
gravitational lensing effects.

References
1. Holz, D. E. and Hughes, S. A. 2005 Astrophys. J., 629, 15
2. Holz, D. E. and Linder, E. V. 2005 Astrophys. J., 631, 678
3.Kocsis, B., Frei, Z., Haiman, Z. and Menou, K. 2006 Astrophys. J., 637, 27
4.Markovic, D. 1993 Phys. Rev. D, 48, 4738
5.Nakamura, T. T. 1998 Phys. Rev. Lett, 80, 1138
6.Schneider, P., Elers, J. and Falco, E. E. 1992 Gravitational Lenses (New York
Springer)
7. Seitz, S., Schneider, P. and Ehlers J. 1994 Class. Quant. Grav., 11, 2345
8. Takahashi, R. and Nakamura, T. 2003 Astrophys. J., 595, 1039
9. C. M. Yoo, K. Nakao, H. Kozaki and R. Takahashi, arXiv:astro-phj0604123.
 QSO LENSING

M. MIRANDA, P. JETZER and A.V. MACCIO

Institute for Theoretical Physics, University of Zurich, CH-805'"/ Zurich, Switzerland
E-mail: [email protected]

Gravitational lenses on arcsec~nds scales provide a tool to probe the mass distribution
in the lensing galaxies at redshift z 0.5-1.0. Image positions can be fitted using simple
smooth galaxy mass models, but observed fluxes are more difficult to match. We analyze
the effects of substructures in galaxy halos to explain such anomalies in the cusp and fold
lensing configurations. In the first case we use detailed numerical simulations combined
with a Monte Carlo approach to compare predictions from the ACDM small scale mass
function with observed flux ratios. We extended our analysis down to a mass of;::o 10 5 M0
for the subhalos. In the second case we try to constrain, with semi analytical approach,
the mass and the position of a substructure by considering its effects on the flux of
the images: we add to a smooth lens model, which reproduces well the positions of the
images but not the anomalous fluxes, one or two substructures (in the mass range ~ 10 6 -
10 8 M0) described as singular isothermal spheres. In a work in progress we take into
account the effect of substructures along the line of the sight.

Keywords: Gravitational lensing

1. Introduction
Theory predicts that the magnifications of the close triplet of images should sum to
zero (cusp case) and the magnifications of the close pair should be equal (fold case),
(Keeton 2005). The standard lens models, although reproduce in general the relative
positions of the images quite accurately, often have difficulties explaining the relative
fluxes of multiply-imaged sources (Mao et a1. 2004). The discrepancy between the
predicted and observed flux ratios is referred to as "anomalous flux ratio problem"
(Mao & Schneider 1998). Several possible explanations have been considered in the
literature (extinction, absorption, stellar microlensing, dark matter substructures)
the most plausible being that the lensing potential of real galaxies are not fully
described by the simple lens models used to compute the lens characteristics. The
most often invoked solution is to consider additional small-scale structures (Dalal
& Kochaneck, Metcalf 2005 and references therein) which, if located nearby the
images, can modify significantly the observed flux ratio between different images,
in particular the so-called cusp or fold relations. Here we summarize methods and
results of our recent works.

1.1. Cusp Relation

By using detailed numerical simulations combined with a Monte Carlo approach, we
compare predictions from the ACDM small scale mass function with observed flux
ratios in the cusp case. First studies claimed that substructures with;:::;! 10 7 -10 9 M0
can explain the cusp relation violation (Bradac et al 2003, Dalal & Kochanek 2002,
Chiba 2002, Keeton 2003). A recent analysis points in the opposite direction (Amara

1731
1732

et al2006 -DM only simulations) A recent work (Maccio et al. 2006) studies the im-
pact of the substructures on the cusp relation using hydro simulations: substructure
with ~ 10 7 Mo are unable to reproduce the observed number of cusp violations.
We extended our analysis down to a mass of ~ 105 Mo for the subhalos by consid-
ering the effect of substructures randomly distributed in a small area around the
images of a simulated lens system and we found the same result. We are able to per-
form simulations with subhalos masses between ~ 10 5 -10 7 Mo (Macci6 & Miranda
2005 ). Also including the effects of extra-halos, like other galaxies surrounding the
primary lens, does not change the results a. This indicates that there is no direct
evidence for substructures (in form of dark dwarf galaxies) from multiple imaged
QSOs.

1.2. Fold configuration

Also we study in detail the anomalous flux ratio which is observed in several four-
image lens systems, where the source lies close to a fold caustic. In particular, we
analyze the two lens systems PGl115+080 and B1555+375, for which there are
not yet satisfactory models which explain the observed anomalous flux ratios (re-
cently Chiba et al. 2005 put limits on masses of substructures inside the halo of
PG1115+080 lens galaxy by analyzing mid-infrared observation, in which the close
pair flux ratio is 0.93). We add to a smooth lens model, which reproduces well
the positions of the images but not the anomalous fluxes, one or two substructures
described as singular isothermal spheres. Knowing fluxes (flux ratios) may help us
to constrain position and mass of a subhalo inside the halo of the main lens: con-
vergence and shear of the whole lens system depend on the mass and position of
the lens components. So, if we know these quantities for the main lens we can get
constraints on the subhalo mass (Einstein radius) and its position. Improving the
fit with this semianalytical approach we find a real good agreement between ob-
servation and theory. For the PGl115+080 system we find two solutions (modeling
the group which the main lens is part of with a SIS or an external shear) with one
substructure (~ 106 Mo for the first solution and 108 Mo for the second one). For
the B1555+375 system we find a solution with 2 substructures (in the mass range
~ 10 6 - 10 8 Mo) with which we can reproduce both positions and radio fluxes of
the images (Miranda & Jetzer 2006). Anyway, since the positions and magnifica-
tions of the images are only known within a certain accuracy, we determined the
corresponding 10- and 20- range for the values of the mass of the substructure and
its position in order to study possible degeneracies. On the other hand stars and/or
very low mass subhalos should be responsible for microlensing effects, that have
still not been accurately detected in the last decade. Nevertheless both effects could
be present. Further observations with instruments with higher resolution will cast
light on these anomalous flux ratios.

aSince we are interested in the flux anomalies, we consider only cases in which we do not have
image splitting.
 1733

1.3. Work in progress

We consider a theoretical cusp configuration as obtained from the lensmodel pack-
age (Keeton 200l)b using a SIE to model the main lens galaxy (with M = r v
5 X 1011 MC')). By using the Sheth & Tormen (1999) mass function profile, we
determine the number of substructures with mass rv 10 5 - 10 9 MC') inside a 1 Mpc 3
volume and model them as SIS, so we can easily (and analytically!) calculate their
contribution to the image deflection and to the magnification factor. The SIS model
has one free parameter directly related to the mass: the velocity dispersion. We take
a typical source (zs = 2) and a main lens (Zd = 1.3) redshift. We diVIde this path
in different bins (depending on the redshift, in front of and behind the main lens).
Choosing an area around the images, we can calculate the volume inside which we
distribute our substructures and their effective number. In one run the light passes
thought the space filled with substructures from the source to the observer: each
image is deflected and amplified. Finally, the sum of the magnifications of the close
triplet of images is calculated c : the so-called R-cusp values seem to match the ob-
servations if we consider sub halos along the line of the sight. For some simulated
configurations, we try to fit the data (positions and magnification factors) with the
lensmodel package, using a simple smooth lens model. We do not find any good
models. We thus conclude that substructures are needed to explain the data.

2. Acknowledgements
Marco Miranda is supported by the Swiss National Science Foundation.

References
1. Amara, A., Metcalf, R B., Cox, T. J., & Ostriker, J. P. 2006, MNRAS, 367, 1367
2. Bradac, M. et al. 2004, A&A, 423, 797
3. Chiba, M. et al. 2005, ApJ, 627, 53
4. Dalal, N. & Kochanek, C. S. 2002, ApJ, 572, 25
5. Keeton, C. R, Gaudi, B. S., & Petters, A. O. 2005, ApJ, 635, 35
6. Keeton, C. 2001, astro-ph/0112350
7. Mao, S., Jing, Y., Ostriker, J.P., Weller, J. 2004, ApJL, 604, 5L
8. Mao, S. & Schneider,P. 1998 MNRAS 295, 587
9. Metcalf, R B. 2005, ApJ, 629, 673
10. Moore B. et al. 1999, ApJ, 524, L19
11. Miranda, M. Jetzer, Ph. 2005 submitted
12. Maccio, A.V., Miranda, M. 2006, MNRAS, 368, 599
13. Maccio, A. V., Moore, B., Stadel, J., & Diemand, J. 2006, MNRAS, 366, 1529
14. Sheth, R K., & Tormen, G. 1999, MNRAS, 308, 119

bThe software is public available via web site: http://cfa-www.harvard.edu/castles

cFor subhalo mass bigger then 10 5 MG, with an area A=2 x2 arcsec 2 around each image, with
Zs = 2.0, and Zd = 0.3, we find 4000 substructures.
 JLENSES AND XFGLENSES

FRANCISCO FRUTOS-ALFARO* and Hugo Solis-Sanchez

School of Physics and Space Research Center,
University of Costa Rica, San Jose, Costa Rica
[email protected]

Astrophysical phenomena like Gravitational Lenses can be easily visualized with Java.
Gravitational Lensing is an effect produced by deflection of light rays passing by stars,
galaxies or cluster of galaxies. It could play an important role in cosmology, because it is
possible to estimate the age of the Universe. JLenses is the Java version of XFGLenses, a
program intended for visualizing and modelling Gravitational Lenses. Among the appli-
cations of the programs are animations, time delay contour lines, caustic visualizations,
and modelling. The OpenGL library for Java is needed to run it. JLenses runs on all
computer systems (XFGLense runs only under Unix and Linux). There is a Website to
download them.

Keywords: Gravitational Lenses

1. Theory of Gravitational Lensing

1.1. The Gravitational Lens Equation
The image positions of a source object behind the gravitational lens can be found
by means of the scaled ray tracing or gravitational lens equation: 1

Yl]=(1-a-.,cos2¢ -,sin2¢ ).[Xl]_[O'l(X'P)], (1)

[ Y2 -,sm2¢ 1-a+,cos2¢ X2 0'2(X,P)
where the variables are defined as in Schneider 1 et al. The set of model parameters
are represented by the variable p.

1.2. Modelling Gravitational Lenses

The function to be optimized has the following form (parametric modelling):2

(2)

where N is the number of images, Wa and wp are weight factors for magnification
and position, Yi are the source positions, Rij are the observed magnification ratios,
and J i are the Jacobians at the image positions.
When this function is optimized, we have found the optimized model parameter
P for the gravitational lens system. Now, we have to calculate the cosmological

'Current address: Theoretical Astrophysics, University of Tiibingen, Auf der Morgenstelle 10C,
72076 Tiibingen, Germany, [email protected]

1734
 1735

distances for a given cosmological model. After that, it is possible to find the mass
of the gravitational lens, the shape of the source, etc.

2. JLenses and XFGLenses

The first version of XFGLenses appeared in 1998. 3 This version was mainly in-
tended for visualizing gravitational lenses. The new version is intended not only for
visualizing, but also for modelling gravitational lenses. The new programs JLenses
and XFGLenses can be downloaded from the following Website:
http://www.tat.physik.uni-tuebingen.de/-frutto/

2.1. Gravitational Lens Models

The implemented models in JLenses and XFGLenses can be found in our Website.
We are planning to implement new model as well as combination of them.

2.2. Solution of Gravitational Lens Equation

To solve Equation (1), we have implemented two algorithms:

• Kayser-Schramm method 5
• Brent method 4
For the sake of visualization (extended sources), it is used the Kayser-Schramm
method. The Brent method was implemented to find the roots for a point source.

2.3. Optimizing the Gravitational Lens Model

In our programs, the function to be optimized is Eq. (2), but the magnification
term is not implemented yet. We will include it soon. At the moment, we have
implemented two procedures to optimize this function for a given model:

• Evolution
• Simplex6
Evolution is a random walk procedure used by R. Kayser. We are planning to imple-
ment other methods. To find the cosmological distances, one can use the ANGSIZ
program. 7 In our program, this is not implemented yet.

3. Applications
At the moment, the programs can be used to do the following applications:

• Animations
• Caustic visualization (Ray Plot)
• Contours lines
• Optimization (not finished yet)
1736

In Fig. 1, some of the above mentioned features can be seen. We are working to
improving the program. New subroutines will be implemented to visualize:

~ Critical and Caustic Curves

~ Light Curve
~ Shape of the source object (Inversion)

4. Conclusions
Visualization is useful in science to understand complex systems in nature.
JLenses and XFGLenses are useful and versatile toolkit to visualize and model
gravitational lenses. For more information about our software visit our Website or
send us an email. Comments are welcome.

Fig. 1. Lens Visualization with JLenses and XGFLenses.

References
1. P. Schneider, J. Ehlers and E. E. Falco, Gravitational Lenses (Springer, 1992).
2. R. Kayser, Ap. J. 357, 309 (1990).
3. F. Frutos-Alfaro, Am. J. Phys. 69, 218 (2001).
4. J. J. More and M. Y. Cosnard, ACM Trans. on Math. Software 6, 240 (1980).
5. R. Kayser and T. Schramm, A f3 A 191, 39 (1988).
6. J. A. Neider and R. Mead, Computer Journal 7, 308 (1965).
7. P. Helbig, R. Kayser and T. Schramm, A f3 A 318, 680 (1997).
 WAVE FRONTS IN GENERAL RELATIVITY THEORY

FRANK GRAVE
FRANCISCO FRUTOS-ALFARO'
THOMAS MULLER
DARIA ADIS
Theoretical Astrophysics, University of Tiibingen,
Auf der Morgenstelle 10C, 72076 Tiibingen, Germany,
[email protected]

JWFront is a program written in Java to visualize, in two or three dimensions, the

wavefronts passing by an object, i.e. a black hole or gravitational lens. The light cones
can also be displayed. The caustics for a given metric can be visualized in two or three
dimensions. This program has a graphical user interface to control input and output. The
implemented metrics so far are: Minkowski, Schwarzschild, and Kerr. We will describe
how the program works and present its applications.

1. Wavefronts and Caustics in General Relativity

The mathematical theory of wavefronts and caustics in general relativity were devel-
oped by Friedrich and Stewart 1 (see also Stewart 2 ). They used Arnold's singularity
theory 3 to tackle this problem. More recently Hasse 4 et al., Low,5 and Ehlers and
Newman 6 have revived this topic. Petters 7 (see also Petters 8 et al.), and Frittelli and
Petters 9 developed this theory for gravitational lenses. More recently, Grave 10 made
movies in MPG format of the gravitational collapse and wavefronts in presence of
Schwarz schild and Kerr fields. To have a perspective on this topic, the interested
reader is referred to Perlick's paper. l l

1.1. Wavefrants
A wavefront is generated by a bundle of light rays orthogonal to a spacelike 2-
surface in a four-dimensional Lorentzian manifold. 4 For the sake of visualization,
the wavefront is defined as the surface A generated by all points of the null geodesic
bundle at a given time ti:

A(t;) = b(t i ) !,(ti) is a null geodesic with ,(to) = (to, Xo, Yo, zo), ti ~ to}.
To get the wavefronts, the geodesic equation is solved for a bundle of light rays
starting from a common point.

1.2. Caustics
The caustic of a wavefront is defined as the set of all points where the wavefront
fails to be an (immersed) submanifold. 4 Roughly speaking, a caustic appears if the

'permanent address: School of Physics and Space Research Center, University of Costa Rica, San
Jose, Costa Rica [email protected]

1737
1738

Fig. 1. A 3d visualization of the wavefront for the Kerr metric. The structure of the caustics can
be seen.

wavefront crosses itself.

1.3. Light Cones

The light cone is defined as the surface generated by all points (x, y, t) that fulfil
the geodesic equation with the null geodesic condition, where the axes are x, y, and
t (t as the z axis). Light cones can be used to visualize caustics.

2. JWEront
JWFront, our program interface, was written in Java and uses LightWeight Java
OpenGL Library (lwjgl), which native drivers support Windows, Linux and Mac
computers. With this graphical user interface, the user gets, after entering the initial
 1739

data, his or her visualizations practically in real time depending on the speed of the
computer. The program can be downloaded from the following Website:
http://www.tat.physik.uni-tuebingen.de/-boot/programs.html

3. Applications
Among the applications of the program are:

• wavefront animations in 2d and 3d

• light cone structure visualizations

The visualization of a 3d wavefront for the Kerr metric is shown in Fig. 1. In this
picture, one sees that the deformation of the wavefront occurs not only in 2d but in
3d as well. For Schwarzschild, such distortion is also possible. To see more examples
visit our website:
http://www.tat.physik.uni-tuebingen.de/-boot/jwfront-doc/

4. Conclusions
JWFront is a very useful interactive program to visualize wavefronts and the light
cone structure for the most important metrics in general relativity and astrophysics.
The applications presented here show that it helps understand the light propagation
in strong fields, such as in Schwarzschild and Kerr spacetimes. The caustics produced
by the chosen metric can be easily visualized in two or three dimensions, and also
with the light cone.

References
1. H. Friedrich and J. M. Stewart, Prac. R. Soc. London Serie A 385, 345 (1983).
2. J. M. Stewart, Advanced General Relativity (Cambridge, 1993).
3. V. I. Arnold, Singularities of Caustics and Wavefronts, (Kluwer, 1990).
4. W. Hasse, M. Kriele and V. Perlick, Class. Quantum Grav. 13, 1161 (1996).
5. R. Low, Jour. Math. Phys. 36, 3332 (1998).
6. J. Ehlers and E. Newman, Jour. Math. Phys. 41, 3344 (2000).
7. A. O. Petters, Jour. Math. Phys. 34, 3555 (1993).
8. A. O. Petters, H. Levine and J. Wambsganss, Singularity Theory and Gravitational
Lensing (Birkhiiuser, 2001).
9. S. Frittelli and A. O. Petters, Jour. Math. Phys. 43, 5578 (2002).
10. F. Grave, Visualizations of Gravitational Collapse and Wavefrants in General Rela-
tivity (Master Thesis in german), Eberhard-Karls-Universitiit Tiibingen (2004).
11. V. Perlick, Living Rev. in Rei. 7, (# 9, 2004).
http://re1ativity.livingreviews.org/Articles/lrr-2004-9/
This page intentionally left blank
Galaxies and the Large-Scale
Structure
This page intentionally left blank
 SPHERICAL VOIDS IN A NEWTON-FRIEDMANN UNIVERSE

R. TRIAY
Centre de Physique Theorique *
CNRS Luminy Case 907, 13288 Marseille Cedex 9, France
* E-mail: [email protected]
www.cpt.univ-mrs·fr

H. H. FLICHE
LMMrt, Fac. des Sciences et Techniques de St Jerome
avo Normandie-Niemen, 13397 Marseille Cedex 20, Prance
E-mail: [email protected]

The understanding of voids formation, which is at the ongm of the foam like pat-
terns in the distribution of galaxies within scale up to 100 Mpc, has become an im-
portant challenge for the large scale formation theory.23 Such a structure has been
observed since three decades and confirmed by recent surveys.4,5,7,14-16,25,27 Investi-
gations has been performed - on their statistical properties by improving identification
techniques,24 by exploring their formation process in a A CDM model through N-body
simulations,3,1l,12,19,22,26,29 by probing their origins;21 - on the kinematics of giant
voids 17 and the dynamics by testing models of void formation. 1,2,10 Herein, we investi-
gate the effect of the cosmological constant A on the evolution of a spherical void through
an exact solution of Euler-(modified) Poisson equations system (EPES).9 Let us remind
that Friedmann-Lemaitre models, which provide us with a suitable description of the
universe at large scales (thanks to their stability with respect to linear perturbations 18 ),
can be described within a Newtonian approach by means of EPES solutions for whom
kinematics satisfy Hubble (cosmological) law. The void consists of three distinct media:
a material shell (S) with null thickness and negligible tension-stress, an empty inside and
a uniform dust distribution outside which expands according to Friedmann equation. We
use a covariant formulation of EPES6,28 for deriving the evolution with time of S acting
as boundaries condition for the inside and outside media.
As a result, S expands with a huge initial burst that freezes asymptotically up to
matching Hubble flow. The related perturbation on redshift of sources located on S does
not exceed f. z ~ 10- 3 . In the Friedmann comoving frame, its magnification increases
nonlinearly with 0 0 and A. These effects interpret respectively by the gravitational
attraction from the outer parts and repulsion (of vacuum) from the inner parts of S,
with a sensitiveness on 0 at primordial epochs and on A later on by preserving the
expansion rate from an earlier decreasing. This dependence of the expansion velocity
v = yHr on A is shown on Fig. 1 through the corrective factor y to Hubble expansion,
where rand H stand respectively for radius of S and Hubble parameter at time t. It
is characterised by a protuberance at redshift z ~ 1.7, the larger the A the higher the
bump. It is due to the existence of a minimum value of Hubble parameter H which
is reached during the cosmological expansion (also referenced as a loitering period). It
characterises spatially closed Friedmann models that expands for ever, they offer the
property of sweeping out the void region, what interprets as a stability criterion.

Keywords: Cosmology: Theory, Cosmological Constant, Voids, Large Scale Structures.

* Unite Mixte de Recherche (UMR 6207) du CNRS, et des universites Aix-Marseille I, Aix-Marseille
II et du Sud Toulon-Var. Laboratoire affilie a la FRUMAM (FR 2291).
tUPRES EA 2596

1743
1744

1.3

Q~ 0.3 Q~

Q~ 1.1

I
-4 ·2 0
In(a)

Fig. l. The corrective factor y to Hubble expansion. It results from a void that initially expands
with Hubble flow at expansion parameter aj = 0.003, with 0 0 = 0.3. and 0.\ = .A = 0, 0.7, l.4.

References
1. N. Benhamidouche, B. Torresani, R. Triay Mon. Not. R. Astron. Soc. 302,807 (1999)
2. K. Bolejko, Mon. Not. R. Astron. Soc. 370, 924 (2006)
3. J.M. Colberg, in Outskirts of Galaxy Clusters: intense life in the s·uburbs, Proc. IAU
ColI. 195, 51 (2004) A. Diaferio, ed.
4. C. Conroy, A.L. Coil, M. White, J.A. Newman, R. Van, M.C. Cooper, B.F. Gerke, M.
Davis, D.C. Koo Astrophys. 1. 635, 990 (2005)
5. D.J. Croton, M. Colless, E. Gaztaaga, C.M. Baugh, P. Norberg, I. K. Baldry, J.
Bland-Hawthorn, T. Bridges, R. Cannon, S. Cole, C. Collins, W. Couch, G. Dalton,
R. De Propris, S. P. Driver, G. Efstathiou, R. S. Ellis, C. S. Frenk, K. Glazebrook, C.
Jackson, O. Lahav, 1. Lewis, S. Lumsden, S. Maddox, D. Madgwick, J. A. Peacock,
B. A. Peterson, W. Sutherland, K. Taylor, The 2dFGRS team, Mon. Not. R. Astron.
Soc. 352, 828 (2004)
6. C. Duval, H.P. Kiinzle, Rep. Math. Phys. 13, 351 (1978)
7. PJ. Einasto, in Proc. of Ninth Marcel Grossmann Meeting on General Relativity, p.
291 . Edits. V.G. Gurzadyan, R.T. Jantzen, Ser. Edit. R. Ruffini. World Scientific
(2002)
8. H.H. Fliche, in Evaluation des parametres cosmologiques a I'aide des propriete optiques
des quasars. Fluctuations des modeles de Friedmann-Lemaitre, these d'etat - Univ.
de Provence (1981)
9. H.H. Fliche, R. Triay, arXiv:gr-qc/0607090, submitted to Gravit. Gen. Relat.
10. y. Friedmann , T. Piran, Astrophys. 1. 548, 1 (2001)
11. D.M. Goldberg, M.S. Vogeley Astrophys. 1. 605, 1 (2004)
12. S. Gottli.iber, E.L. Lokas, A. Klypin, Y. Hoffman, Mon. Not. R. Astron. Soc. 344, 715
(2003)
13. V. Icke, Mon. Not. R. Astron. Soc. 206, 1 (1984)
14. J. Jaaniste, M. Einasto, J. Einasto, Astroph. Space Sci. 290, 187 (2004)
15. M. Joevee, J. Einasto, in The Large Scale Structure of the Universe. IAU Symp. 79,
241. Eds. M.S. Longair & J. Einasto (1978)
16. R.P. Kirshner, A. Oemler, P.L. Shechter, S.A. Shechtman Astrophys. 1. 248, L57
 1745

(1981)
17. H. Mathis, J. Silk, L.M. Griffiths, M. Kunz, Astron. Astrophys. 382, 389 (2002)
18. E.M. Lifchitz, LM. Khalatnikov, Usp. Fiz. Nauk. 80, 391 (1963); Adv. Phys. 12, 185
(1963)
19. H. Mathis, J. Silk, L.M. Griffiths, M. Kunz, Mon. Not. R. Astron. Soc. 350, 287
(2004)
20. W.H. Mc Crea, Astron. J. 60, 271 (1955)
21. L.M. Ord, M. Kunz, H. Mathis, J. Silk, Pub. Astron. Soc. Austr. 22, 2,166 (2005)
22. N.D. Padilla, L. Ceccarelli, D.G. Lambas, Mon. Not. R. Astron. Soc. 363, 977 (2005)
23. P. J. E. Peebles Astrophys. J. 557, 495 (2001)
24. S. G. Patiri, J. E. Betancort-Rijo, F. Prada, A. Klypin, S. Gottlober, Mon. Not. R.
Astron. Soc./369, 335 (2006)
25. RR Rojas, M.S. Vogeley, F. Hoyle, J. Brinkmann, Astrophys. J. 617, 50 (2004)
26. S.F. Shandarin, J.V. Sheth, V. Sahni, Mon. Not. R. Astron. Soc. 353, 162 (2004)
27. RM. Soneira, P.J.E. Peebles, Astron. J. 83, 845 (1978)
28. J.-M. Souriau, (French) C. R Acad. Sci. Paris Ser. A , p. 751 (1970); (French) C.
R Acad. Sci. Paris Ser. A , p.1086 (1970); "Milieux conti nus de dimension 1, 2 ou 3
: statique et dynamique" in Actes du 13eme Congnis Fmn~ais de Mecanique, AUM,
Poitiers-Futuroscope, p.41 (1997)
29. R van de Weygaert, R Sheth, and E. Platen, in Outskirts of Galaxy Clusters . intense
life in the suburbs, Proc. lAU Coli. 195, 1 (2004) A. Diaferio, ed.
This page intentionally left blank
Dark Energy and Universe
Acceleration
This page intentionally left blank
 DARK ENERGY AND UNIVERSE ACCELERATION OF
NONLINEAR SUIPERSYMMETRIC GENERAL RELATIVITY *

KAZUNARI SHIMA and MOTOMU TSDUDA

Laboratory of Physics, Saitama Institute of Technology,
Fukaya, Saitama 369-0293, Japan
E-mail: [email protected]@sit.ac.jp

The cosmological constant of nonlinear supersymmetric general relativity(NLSUSYGR)

may be the origin of everything and give new insights into the origin of mass and the
mysterious relations between the cosmology and the (low energy) particle physics.

Keywords: General relativity; Nonlinear supersymmetry; neutrino mass; Dark energy.

1. Nonlinear Supersymmetric General Relativity(NLSUSYGR)

NLSUSY GR is obtained by extending the Einstein general relativity (EGR) prin-
ciple to new (called SCM from composite viewpoints 1 ) spacetime just inspired by
NLSUSY, where tangent spacetime is specified not only by the ordinary SO(3,1)
Minkowski coordinates x a but also by the SL(2,C) Grassmann coordinates 7j; turn-
ing subsequently to the dynamical Nambu-Goldstone (NG) fermion 7j;( called superon
hereafter) of NLSUSY Volkov-Akulov (VA) mode1. 2 NLSUSY GR (N = 1 SGM)
action 1 is written as the following geometrical form of empty SG M spacetime (of
everything) :

c4
L(w) = --Iwl{n(w) - A}, (1)
167fG
where Iwl detw aJL = det{e aJL + t aJL (7j;)} and taJL('l/;) = ~: (1jry a8JL7j; - 8 JL {;,a7j;).
=
waJL(x) and new) are the (composite) unified vierbein and the corresponding
scalar curvature defining SGM spacetime. eaJL and t aJL (7j;) are the ordinary vier-
bein and the mimic analogue for SL(2C), whose dynamical parallel alignments
for the spacetime orientations of SG M spacetime are encoded in w a JL (x). G is
the Newton gravitational constant and A is a (small) cosmological term. Remark-
ablly "', which is the arbitrary constant with the dimension (length)2 in NLSUSY
VA model, is now related to G and A by ",2 = (~:~)-l in the NLSUSY GR
(SGM) scenario. (N-extended) NLSUSY GR action (1) is invariant at least un-
der [new NLSUSY] 0 [local GL(4,R)] 0 [local Lorentz] 0 [local spinor translation]0
[global SO(N)] 0 [local U(1)N].3

2. Cosmological Implications of NLSUSYGR

(1) is unstable due to the NLSUSY structure of asymptotic tangent space and
decays spontaneously (called Big Decay of SGM spacetime) to ordinary Riemann

-Talk given by K. Shima

1749
1750

spacetime described by the Einstein-Hilbert action coupled with massless superon

(matter) 1/; of NLSUSY VA model:

c4
L(e, 1/;) = 161fC1el{R(e) - A + T(e, 1/;)}, (2)

where T( e, 1/;) is the kinetic term and the gravitational interaction of superon. The
second and the third terms reduce to (N-extended) NLSUSY VA action in asymp-
totic Riemann-fiat (eal"(x) ---+ bal") spacetime with ",2 = (~;~)-1, which is different
from SCM-fiat (wal"(x) ---+ bal") spacetime for (1).
As some NL theories can be recasted into equivalent linear (L) renormalizable
theories, NLSUSY is also the case. To investigate the mass (gap) generation mech-
anism in SCM scenario it is useful to go to asymptotic fiat spacetime. We have
shown 4 that the four dimensional(D=4) (more realistic) N = 2 NLSUSY VA model
in terms of two superons (Majorana NC fermion fields) 1/;i (i = 1,2), which cor-
responds to the N = 2 SCM action (2) in asymptotic Riemann-fiat spacetime, is
(algebraically) equivalent to the spontaneouly broken N = 2 LSUSY model in terms
of the massless supermultiplet with component fields (va, Ai, r/J, pI) (1 = 1,2,3) for
a U(l) gauge field, two Majorana spinor fields of ~ of SU(2), a complex scalar field
and three auxiliary real scalar fields of Q of SU(2). The N = 2 NLSUSY VA action,
LVA = - 2~21wl = - 2~2 {I + t a + ~(ta atbb - tabtb a) + ... } with Iwl = det Wab =
a
det(bg +tab) and tab = -iK,2?jJi'"'(a[h1/;i, is recasted into the equivalent spontaneouly
broken N = 2 LSUSY action L a = [Ja'l-'A,[JaA,* 'I-'
- lp2
4 ab
+ i).i
2)U :7l Ai + l(pI)2 _ leI pI
2 '" <" ,
where Pab = [JaVb - [JbVa, the e are arbitrary real parameters for the induced global
SO(3) (SU(2)) rotation satisfying (e)2 = 1 and La means the (almost) free action.
Note that the Fayet-Iliopoulos (FI) P terms indicating spontaneous SUSY break-
ing with the vacuum expectation value < pI >= ~ are generated automatically.
And all component fields of the LSUSY supermultiplet can be constructed as the
composites of superons 1/;i (called SUSY invariant relations), in such a way as the
familiar LSUSY transformations defined on the elementary supermultiplet are re-
produced on the abovementiond composite supermultiplet in terms of the NLSUSY
. 1· - . . .
transformations b(1/" = ;;zC - iK,(1 ,",(a1/,J [Ja 1/;' on the constituent superons 1/,'. We
investigate the cosmological implications of SCM scenario described by (1) and the
subsequent Big Decay to Riemann spacetime with massless superon (2) igniting the
Big Bang. 5 The variation of (2) with respect to e a I" gives the equation of motion
for e a I" depicted formally as follows

(3)

where Tl"v (e, 1/;) represents the stress-energy-momentum of superon including the
gravitational interactions. Note that - ~;~ can be interpreted as the constant
negative energy density of empty spacetime, i.e. the dark energy density -PD re-
sponsible for the observed present accelleration of the universe. While, on asymp-
totic fiat spacetime, the low energy theorem of the particle physics gives the fol-
 1751

j
lowing superon-vacuum coupling < W a(q)!J kll i3 !O >= iJ ~:~(rIl)ai3ojkeiqX +"',
where Jkll = iJ ~:~,IlWk + ... is the conserved supercurrent of NLSUSY VA ac-

tion 6 and J ~:~ is the coupling constant gsv of superon with the vacuum. Further
as shown before the right hand side of (3) for N = 2 flat case is essentially N = 2
NLSUSY VA action. And it is equivalent to the broken LSUSY action La which
contains the SUSY breaking mass (gap) M su Sy for the (composite) fields of the
(massless) LSUSY supermultiplet

Msusy 2 rv< F
I
>rv V~
S;C~
I
rv
I
gsv~ . (4)
Suppose that among the LSUSY supermultiplet the stable and the lightest particle
retains the mass of the order of the spontaneous SUSY breaking Msusy. And if
the neutrino is such a particle Ai(x), i.e. mv 2 rv J
~:~e, then SGM predicts the
observed value of the (dark) energy density of the universe and naturally explains
the mysterious coincidence between mv and pobs rv PD, provided rv 0(1), e
pobs rv (1O-12GeV)4 rv mu 4 . (5)
For actually gennerating mass in the LSUSY supermultiplet the familiar Yukawa
interaction term should appear in the equivalent LSUSY theory La. Recently we
have found in D=2 for simplicity that the familiar N = 2 LSUSY invariant mass
terms Lm = - ~m (), i Ai - 2AF + Eab ¢Fab) as well as the Yukawa interaction terms
Lf = f(A),i Ai _Eij¢),i'5Aj +A2 F_¢2 F-E ab A¢Fab) vanish identically under SUSY
invariant relations in the linearization of N = 2 NLSUSY. Therefore we can add
these terms to the linearized free action La without violating the equivalence and the
invariance as well. The consequent equivalent Lagrangian Lo + L f generates mass
at the tree level and reproduces (4) and (5), 5 where besides the nutrinos( fermions)
Ai one neutral scalar field obtains the mass of the same magnitude. 7
Now we summarize the results as follows. The cosmological constant of SGM space-
time may be the origin of everything, i.e. mass (and the equivalence principle),
dark matter, dark energy and the present accelerated expansion of the universe.
NLSUSYGR may suggest a certain composite (SGM) structure of nature beyond
the SM(GUT), a la G-L theory v.s. BeS theory for the superconductivity.

References
1. K. Shima, Phys. Lett. B501 (2001) 237. K. Shima, Eur. Phys. J. C7 (1999) 341.
2. For a review on Land NL SUSY, J. Wess and J. Bagger, Supersymmetry and Super-
gravity (Princeton University Press, Princeton, New Jersey, 1992).
3. K. Shima and M. Tsuda, Phys. Lett. B507 (2001) 260. See the talk by M. Tsuda.
4. K. Shima, Y. Tanii and M. Tsuda, Phys. Lett. B546 (2002) 162.
5. K. Shima and M. Tsuda, PaS HEP2005 (2006) OIl.
6. K. Shima, Phys. Rev. D20 (1979) 574.
7. K. Shima and M. Tsuda, hep-th/0611072. To appear in Physics Letters B.
 TESTING THE DARK-ENERGY-DOMINATED COSMOLOGY BY
THE SOLAR-SYSTEM EXPERIMENTS

YURII V. DUMIN
Theoretical Department, IZMIRAN, Russian Academy of Sciences, Troitsk, 142190 Russia
[email protected]

According to the recent astronomical data, the most part of energy in the Uni-
verse is in the 'dark' form, which is effectively described by A-term in Einstein
equations. All arguments in favor of the dark energy were obtained so far from the
observational data related to very large (intergalactic) scales. Is it possible to find a
manifestation of the dark energy at much less scales (e.g. inside the Solar system)?
In general, such effects can be expected from the solution of the equations of
General Relativity (GR) for a point-like mass M in the A-dominated (de Sitter) Uni-
verse, which was obtained by Kottler 1 very long time ago. The presence of A-term
should change, particularly, the standard relativistic shift of Mercury's perihelion.
This was the idea by Cardona & Tejeiro,2 who proposed using the measure of the
uncertainty in our knowledge of Mercury's perihelion shift to impose the upper
bound on A. The result obtained was not so good as other cosmological estimates
but, surprisingly, the accuracy was worse by only 1-;.-2 orders of magnitude. A more
skeptical viewpoint on the same subject was presented recently by Iorio. 3
Since accuracy of the above method is insufficient, it was proposed in our pre-
vious papers 4 ,5 to utilize the data of radial (rather than angular) measurements of
the Moon to reveal the anomalous increase in its distance from the Earth produced
by the A-term, which looks formally as 'local' Hubble expansion. Why is it nec-
essary to reexamine the problem of local Hubble expansion just in the context of
'dark-energy'-dominated cosmological models'?
Hubble dynamics at small scales is studied for a long time, starting from the
pioneering work by McVittie. 6 Although the results by various authors were quite
contradictory (e.g. review by Bonnor 7 and references therein), the most popular
point of view was that the Hubble expansion manifests itself only at the sufficiently
large distances (from a few Mpc) and is absent at the less scales. s There were a
few arguments in favor of such conclusion, such as the so-called Einstein-Straus
theorem,9 a quasi-Newtonian treatment of Hubble effect in a small volume as a
tidal-like action by distant matter (e.g. the recent work by Dominguez & Gaite lO
and references therein), and the Einstein-Infeld-Hoffmann (EIH) surface integral
method, which was applied to the problem of local Hubble expansion by Anderson. 11
Unfortunately, as is shown in Ref. 12, all these approaches become inapplicable when
the Universe evolution is governed by A-term, uniformly distributed in space.
A frequent experimental argument against the Hubble expansion within So-
lar system is based on the available constraint on time variation in the gravi-
tational constant derived from the lunar dynamics, which is now as strong as
CIG = (4±9)xlO- 13 yr- 1 (Ref. 13). Unfortunately, the equivalence between the

1752
 1753

Table 1. Rates of secular increase in the mean Earth-Moon distance.

Method Immediate measurement by Independent estimate from the

the lunar laser ranging Earth's tidal deceleration
Effects involved (1) geophysical tides (1) geophysical tides
(2) local Hubble expansion
Numerical value 3.8±O.1 cmyr- 1 1.6±O.2 cmyr- 1

effect of variable G and the cosmological expansion, stated by some authors, is

based solely on the Newtonian arguments. A more accurate treatment of this prob-
lem in the GR framework 14 shows that manifestation of A-term in some components
of the metric tensor really looks like the influence of variable G if we assume that
G = Go + Gt, where G = -cJ
A/3 ; but such interpretation is not self-consistent:
the A-dependence of a few other components is not expressible in terms of the vari-
able coefficient of gravitational coupling. Therefore, the available limits on G/ G ,
in general, cannot be reinterpreted as a constraint on local cosmological dynamics.
Since all the commonly-used arguments against the local Hubble expansion fail
in the case of dark energy, it becomes reasonable to seek for the corresponding
effect; and the most sensitive tool seems to be the lunar laser ranging (LLR) .15,16
For example, if we assume that planetary systems experience Hubble expansion with
the same rate as everywhere in the Universe (60-;-70 kms- 1 Mpc 1), then average
radius of the lunar orbit R should increase by ~ 50 cm for the period of 20 years.
On the other hand, the accuracy of LLR during the last 20 years was not worse than
2-;-3 cm; so the perspective of revealing the local Hubble effect looks very good.
The main problem is to exclude the effect of geophysical tides, which also con-
tributes to the secular increase in the Earth-Moon distance as it = k: TE , where
TE is the Earth's diurnal period, and k = 1.81 X 10 5 cms- 1 (e.g. Ref. 5). So, if TE
is known from independent astrometric measurements of the Earth's rotation de-
celeration with respect to distant objects, then the above relation can be used to
exclude the geophysical tides and, thereby, to reveal a probable Hubble expansion.
The telescopic data, accumulated from the middle of the 17th century, were pro-
cessed by a few researches; and one of the most detailed compilations was presented
recently in Ref. 17. Of course, the value of secular trend derived from the quite short
time series can suffer from considerable periodic and quasi-periodic variations in T E .
So, the main aim of our statistical analysis, described in more detail in Ref. 12, was
to estimate as carefully as possible the 'mimic' effect of such influences. The result
can be written as TE = (8.77±l.04)x10- 6 syr- 1 . (This value is appreciably less
than in our previous work,5 where it was taken from the older literature.)
The entire analysis of LLR vs. the astrometric data is summarized in Table 1.
The excessive rate of increase in the lunar orbit, 2.2±O.3 cm yr-1, can be attributed
just to the local Hubble expansion with rate H~loc) = 56±8 kms- 1 Mpc- 1 .
Next, it is reasonable to assume that the local Hubble expansion is formed only
by the uniformly-distributed dark energy, while the irregularly-distributed (aggre-
gated) forms of matter begin to affect the Hubble flow at the larger distances,
1754

thereby increasing its rate up to the standard intergalactic value. If the Universe
is spatially flat and filled with the A-term and a dust-like ('cold') matter, with
densities PAO and PD~ respectively, then 18

Ho = !87rG
-3- J PAD + PD~ . (1)

So, if Ho is formed locally only by PAD, while globally by both these terms, PAO and
PD~ (or, in terms of the relative densities, ~ho= PAO I Per and nDO = PD~ I Per), then
H(loe)
_0_ =
[ n ] -1/2.
1+ ~ (2)
Ho nAD
At nAD = 0.75 and n DO = 0.25, we get Hoi H6 ) ~ 1.15. Therefore, Ho =
1oe

65±9 kms- 1 Mpc- 1 , which is in reasonable agreement both with the well-known
WMAP result, 71±3.5 kms- 1 Mpc-l, and with the recent Hubble diagram for type
Ia supernovae,19 whose interpretation requires a slightly reduced value of Ho.
Therefore, the presence of local Hubble expansion, caused by the A-term, gives
us a reasonable explanation of the anomalous increase in the lunar orbit, consistent
with the 'large-scale' astronomical data. Thereby, this is one more argument in favor
of the dark energy. Besides, if the local Hubble expansion really exists, it should
result in profound consequences not only for cosmological evolution but also for the
dynamics of planetary systems and other 'small-scale' astronomical phenomena.

References
1. F. Kottler, Ann. Phys. 56, p. 401 (1918).
2. J. F. Cardona and J. M. Tejeiro, Astrophys. J. 493, p. 52 (1998).
3. L. Iorio, Int. J. Mod. Phys. D 15, p. 473 (2006).
4. Yu. V. Durnin, Geophys. Res. Abstr. 3, p. 1965 (2001).
5. Yu. V. Durnin, Adv. Space Res. 31, p. 2461 (2003).
6. G. C. McVittie, MNRAS 93, p. 325 (1933).
7. W. B. Bonnor, Gen. Rei. Grav. 32, p. 1005 (2000).
8. C. W. Misner, K. S. Thorne and J. A. Wheeler, Gravitation (W.H. Freeman & Co.,
San Francisco, 1973).
9. A. Einstein and E. G. Straus, Rev. Mod. Phys. 17, p. 120 (1945).
10. A. Dominguez and J. Gaite, Europhys. Lett. 55, p. 458 (2001).
11. J. L. Anderson, Phys. Rev. Lett. 75, p. 3602 (1995).
12. Yu. V. Durnin, astro-ph/0507381.
13. J. G. Williams, S. G. Turyshev and D. H. Boggs, Phys. Rev. Lett. 93, p. 261101 (2004).
14. Yu. V. Durnin, Phys. Rev. Lett. 98, p. 059001 (2007).
15. J. O. Dickey, et al., Science 265, p. 482 (1994).
16. K. Nordtvedt, Class. Quant. Grav. 16, p. A101 (1999).
17. N. S. Sidorenkov, Physics of the Earth's Rotation Instabilities (Nauka-Fizmatlit,
Moscow, 2002, in Russian).
18. L. D. Landau and E. M. Lifshitz, The Classical Theory of Fields (Pergamon, Oxford,
1975).
19. B. Reindl, G. A. Tammann, A. Sandage and A. Saha, Astrophys. J. 624, p. 532 (2005).
 A DARKLESS SPACE TIME

A. TARTAGLIA * and M. CAPONE

Dipartimento di Fisica, Politecnico di Torino, and INFN, Turin section
Torino, 1-10129, Italy
* E-mail: [email protected]@polito.it
www.polito.it/relgmv

Instead of introducing an ad hoc dark energy in order to explain the accelerated ex-
pansion of the universe, we try a different approach treating space time as a physical
continuum endowed of specific properties. A Lagrangian modeled on the one valid for
simple dissipative phenomena in fluids is build and used for empty space time. The inter-
nal "viscosity" is shown to correspond to a four-vector field. Using the known symmetry
of the universe, assuming the vector field to be divergence-less and solving the Euler-
Lagrange equation we directly obtain inflation and a phase of accelerated expansion of
space time. The vector field is shown to correspond to the displacement vector field
induced by a point defect in a four-dimensional continuum. The presence of ordinary
matter does not spoil the obtained general behaviour.

Keywords: cosmology, defects, accelerated expansion.

1. Introduction
Since the moment when growing evidence supported the idea that the universe is
undergoing an accelerated expansion 1 Einstein's "big blunder" ,2 the cosmological
constant, has seen a big revival and has been enriched with new and sophisticated
theories. One has been led to think that the universe is filled up with some" dark
energy" exerting a negative pressure, responsible for the acceleration. Here we try a
different explanation taking advantage of analogies with other branches of physics.
The symmetry of the whole universe in four dimensions implies the presence of
a singular event: a center of symmetry. Our approach assumes that the singularity
is not due to the content (mass-energy) of the spacetime, but is built in the very
spacetime. The next step consists in interpreting and treating the singularity just
as a defect in a continuous medium is in the classical elasticity theory.3 A point
defect in an otherwise homogeneous and isotropic medium induces a strained state,
characterizable by means of a radial vector field: the "rate of stretching" in the
radial position. The theory of defects is a well established one and was developed
more than one century ago together with the elasticity theory; the idea of extending
it to more than three dimensions is not new,4-8 but it has never been considered
more than a curiosity.
We start from the correspondence between the bidimensional phase space of a
Robertson Walker universe and the phase space of a point particle moving across a
viscous medium. From this analogy, we are able to write down and then generalize
an appropriate Lagrangian. What we obtain in the end is a spacetime displaying
inflationary expansion in the neighbourhood of the center of symmetry (i.e. the Big
Bang), then a deceleration-acceleration-deceleration sequence.
Introducing matter in the form of an ordinary perfect fluid we find that in the

1755
1756

negligible pressure era (today) the presence of matter has no influence on the global
solution. In the radiation dominated era the general behaviour is preserved when
the matter-energy content is lower than a critical value. The theory contains one
free parameter, which may be fixed in such a way that the present value of the
Hubble constant as well as the age of the universe are reproduced. Doing so, we see
that the currently estimated content of matter-energy in the cosmos is well below
the critical value, and the position and duration of the accelerated expansion are
consistent with the data from observation.

2. The appropriate Lagrangian

As already mentioned in the introduction, we started from considering that the state
of a Robertson Walker (RW) universe is described by the only scale factor a with
an evolution parameter T (cosmic time). The phase space displays da/dT versus a.
A simple transliteration (from a to x) converts the evolution of the universe into
the interaction of a point particle with an isotropic fluid. The interesting feature of
this analogy is that we know how to write the equations governing the motion of a
point particle of mass m in a viscous fluid. Although dissipative, the problem can
be treated in a relativistic covariant form by means of the Lagrangian

S = -m lB e'7c<p,G x {3 ds , (1)

where TJex{3 is an element of the Minkowski metric tensor, ,ex

is a component of
a vector containing the information on the viscosity of the fluid and x{3 are the
components of the position four-vector of the particle. Exploiting this analogy we
construct the action integral for space-time in the form:

S = 1112 e-g"v,",v RdO + "" r ll2

pdO , (2)
Il, Jn l

Now dO is the invariant four-volume element, R is the scalar curvature, , is our

four-vector field, K is a coupling constant, and p is the pressure of the ideal fluid rep-
resenting the matter content of the universe. Assuming the universe has a Robertson
Walker symmetry, the action integral depends only on the scale parameter a and
its cosmic time derivatives:

(3)

From the above action integral it is possible to obtain the "dynamics" of the
universe or, to say better, the four-dimensional equilibrium state that we read as
an evolution in time.

3. The results
The vector field that we have introduced has a direct interpretation in a four-
dimensional theory of elastic continua. It corresponds to the flux density of strain
 1757

induced by a pointlike defect. As such it must be divergence-less except at the

location of the defect. This condition implies:

(4)
Introducing the said constraint and solving the field equations we obtain an inter-
esting result which is shown in Fig. 1. As it can be seen, we are able to reproduce

I:
I
I

\!:c:. . . -- - -- ---

Fig. 1. Behaviour of the expansion rate of the universe in the presence of ordinary matter: the
solid line shows the dependence of a on the cosmic time in the case of a spacetime with matter in
subcritical conditions with 'ljJ = 0.5. For the sake of comparison we also show the empty spacetime
case ('ljJ = 0, dashed line) and a case of supercritical matter density ('ljJ = 1, thick grey line).
We note that the three curves start with an accelerated expansion phase. For later times this is
converted into deceleration, but for'ljJ = 0.5 and'ljJ = 0 an effective "re-heating" occurs.

a decelerated-accelerated-decelerated evolution with a model containing just one

parameter. The details of this theory may be found in .9

References
1. S.M. Carroll in Carnegie Observatories Astrophysics series, Vol. 2 : Measuring and
Modelling the Universe, W.L. Freedman Ed., Cambridge Univ. Press, (Cambridge
2003).
2. G. Gamow, My World Line (Viking Press, New York, 1970).
3. J.D. Eshelby, The continuum theory of lattice defects, in Solid State Physics, F.D. Seitz
and D. Turnball Eds. (Academic Press Inc., New York, 1956).
4. H. Kleinert Gauge Fields in Condensed Matter, Vol II: Stresses and Defects(World
Scientific, Singapore, 1989).
5. M.O. Katanaev and LV. Volovich , Ann. Phys. 216, 1 (1992).
6. C. Malyshev, Ann. Phys. 286, 249 (2000).
7. R.A. Puntingam and H.H. Soleng, Class. Q. Grav. 14, 1129 (1997).
8. T. Padmanabhan, Int. Jour. of Mod. Phys. D 13, 2293 (2004).
9. A. Tartaglia, M. Capone, gr-qc/0601033 (2006).
 ON INTRINSIC INVARIANCE IN GURZADYAN-XUE
COSMOLOGICAL MODELS

H. KHACHATRIAN*,t, G.V. VERESHCHAGINU and G. YEGORIAN*,t

* Yerevan Physics Institute, Alikhanyan brs. st. 2, 315036 Yerevan, Armenia
t ICRANet, P.le della Repubblica 10, 65100 Pescara, Italy and
ICRA, Dip. Fisica, Univ. "Sapienza", P.le A. Mora 5, 00185 Rome, Italy
[email protected]

Analysis of cosmological solutions of GX models shows interesting features, in particular

for simple models with pressureless fluid the presence of separatrix in the phase space of
solutions which divides them into two classes: Friedmannian-like with initial singularity
and non-Friedmannian solutions. Invariants were found that reveal intrinsic symmetry
in GX models.

The formula for dark energyl

(1)
where c, G and a are speed of light, gravitational constant, and scale factor of
the Universe, derived by Gurzadyan and Xue predicts the value of dark energy
density in agreement with present observations. 2 This formula leads to a possibility
of variation of physical constants such as speed of light, gravitational constant,
cosmological term. Classification of models based on the GX scaling (1) is given
in. 3 Analysis of the cosmological equations for GX models is performed in. 4
It was shown that each GX model is characterized by the critical value of the
density parameter Dm :::::; 2/3 which represents a separatrix in the space of solutions
of cosmological equations. 5 To illustrate the presence of separatrix we provide at
fig.1 phase portraits for expansion stage with zero spatial curvature. Numeration
of models is as follows: I - variable dark energy density, II - variable speed of light,
constant cosmological term, III - variable gravitational constant, IV - variable speed
of light, constant dark energy density (details see e.g. in 4 ).
Phase space variables are
2 2
a = -arctg(a), m = -arctg(/L), (2)
7r 7r
where /-L is the matter density,
The separatrix divides the space of solutions into two classes: Friedmannian-
like with initial singularity and non-Friedmannian solutions starting with vanishing
density.5 The value of separatrix is given by
2 1
Dm = '3 (1 _ k/7r 2 ) . (3)
The upper left corner in each diagram at fig.1 corresponds to the Friedmann
cosmological singularity with /-L ----+ 00 and a ----+ O. In the lower right corner /L ----+ 0
and a ----+ 00. The separatrix in each diagram is shown by a thick curve. Above the
separatrix solutions start with a classical singularity and end up with infinite scale
factor and zero density (or constant density for models II and III) but the solutions

1758
 1759

m I m IV
~~~-=~-===~~==~~~~~1

u a
o 0
m m
~~~~==========~~~~1

u
o
II III
Fig. 1. Phase portraits for models I-IV. Directions of phase trajectories are shown by arrows. See
detailed explanations in the text.

below the separatrix start at a positive scale factor and zero density and tend to
the same limit.
The presence of separatrix in GX models is a crucial difference with the Fried-
mann solutions within General Relativity. It is shown that such symmetries are due
to existence of invariant in GX models;6 the value of separatrix can be obtained
from expression of vanishing invariant.
The continuty equation for GX models reads 3

(4)
1760

where /-lex == pex / c2 is dark energy mass density. Invariant was found for this
equation6 which has the form
3
IW C 7ra
~ - 4 = canst == bm · (5)

The value bm = 0 corresponds exactly to the above condition (3).

Apart from the demonstrated symmetry, present in all GX models, which is
linked with the form of (1) it was shown that GX models agree well with super-
novae and radio galaxies data. 7 Perturbation dynamics analysis 8 also shows some
interesting features of GX models, in particular that density fluctuations grow also
during curvature- or vacuum-dominated stages, following matter-dominated stage
of expansion, unlike perturbations within Friedmannian models.

References
l. V.G. Gurzadyan, S.-S. Xue in: "From Integrable Models to Gauge Theories; volume
in honor of Sergei Matinyan", ed. V.G. Gurzadyan, A.G. Sedrakian, p.l77, World Sci-
entific, 2002; Mod. Phys. Lett. A18 (2003) 561 [astro-ph/0105245], see also [astro-
ph/0510459].
2. S.G. Djorgovski, V.G. Gurzadyan, Nucl.Phys. B (in press), astro-ph/0610204.
3. G.V. Vereshchagin, Mod. Phys. Lett. A2l (2006) 729.
4. G.V. Vereshchagin and G. Yegorian, Class. Quantum Grav. 23 (2006) 5049.
5. G.V. Vereshchagin and G. Yegorian, Phys. Lett. B636 (2006) 150.
6. H.G. Khachatryan, astro-ph/0608649.
7. G.V. Vereshchagin and G. Yegorian, astro-ph/0604566.
8. G.V. Vereshchagin and G. Yegorian, astro-ph/0610197; see also this volume.
 PHANTOM DARK ENERGY AND ITS COSMOLOGICAL
CONSEQUENCES

MARIUSZ P. D~BROWSKI

Institute of Physics, University of Szczecin,

Wielkopolska 15, 70-},51 Szczecin, Poland
[email protected]

I discuss the dark energy characterized by the violation of the null energy condition
(e + p ::: 0), dubbed phantom. Amazingly, it is admitted by the current astronomical
data from supernovae. We discuss both classical and quantum cosmological models with
phantom as a source of matter and present the phenomenon called phantom duality.

1. Introduction

Until a couple of years ago, the standard cosmological source of dark energy was
considered to be a slightly negative pressure matter (-g < p < 0) with perhaps time-
evolving equation of state (quintessence), but not exceeding a "mysterious" barrier
p = -g = -A, which corresponded to the cosmological constant. Only the strong
energy condition of Hawking and Penrose (g + 3p > 0, g + p > 0) was presumably
violated, and the evolution of the universe in future contained two options: an
asymptotic emptiness or a Big-Crunch. However, a deeper analysis of the data
from supernovae, cosmic microwave background (WMAP) and large-scale structure
(SDSS)l shows that the dark energy may also be the matter whose pressure is
less than minus the energy density and so violates the null energy condition, and
consequently, all the energy conditions. This matter is dubbed phantom,2 and it
leads to qualitatively new types of the evolution of the universe as a whole. I will
discuss these opportunities both in classical 3 and quantum 4 cosmological context.

2. Classical phantom cosmologies - Big-Rip and phantom duality

Phantom is dark energy of a strongly negative pressure which in the easiest case
may be simulated by a scalar field ¢ of negative kinetic energy with the Lagrangian

(1)

where l = -1 for phantom, l = + 1 for standard scalar field, and V (¢) is the
potential. In terms of the energy-momentum tensor for a perfect fluid this gives (g
- the energy density, p - the pressure): g = (l/2)¢2 + V(¢) , p = (l/2)¢2 - V(¢) ,
and it surely violates the null energy condition g + p = l¢2 > 0, if l = -1. There
are many other examples where phantom matter appears naturally. For example,
this is the case in Brans-Dicke theory in the Einstein frame (provided the Brans-
Dicke parameter w < -3/2), in superstring cosmology, in brane cosmology, and in
viscous cosmology. Due to the energy conditions violation, it makes a failure of the
standard cosmic censorship conjectures, black hole thermodynamics, positive mass
theorems and other renowned theorems of general relativity. It also leads to classical

1761
1762

and quantum instabilities,5 which encourages some researchers to disregard it as a

serious candidate for the dark energy. However, as it was already mentioned, its main
motivation does not come from the theoretical considerations. On the contrary, it
comes from the observational data. Of course its stability is a problem, and there
are suggestions how to avoid that problem, too. 6
The most striking result which refers to phantom is that its energy density (}
grows proportional to the scale factor a(t), i.e.,
for w < -1. (2)
Then, unlike in a more intuitive standard matter case, where the growth of the
energy density corresponds to the decrease of the scale factor, here, the growth of
the energy density accompanies the expansion of the Universe. This really gives a
new scenario for the future evolution of the universe, which has not been considered
so far in cosmology. There is a future singularity, which due to its peculiar properties
is called Big-Rip. On the approach to a Big-Rip everything in the universe is pulled
apart in a reverse order - first clusters, then galaxies, solar systems, atoms, nuclei
etc. 2
One of the standard cosmological scenarios in cosmology is the evolution from
a Big-Bang to a Big-Crunch. In phantom cosmology it is possible to start with a
Big-Rip reach the minimum and terminate at another Big-Rip. This is an example
of the phantom duality3 - a new symmetry of the field equations which allows to
map a large scale factor (cf. (2)) onto a small one and vice versa due to a change
a(t) f-7 l/a(t) or w + 1 f-7 -(w + 1) . (3)
Similar symmetry was already discovered in the context of superstring cosmology
under the name of the scale factor duality 7 and further extended onto the brane
cosmology8 ("phantom triality").
Admission of phantom with p < - (} enlarges possible set of cosmological solu-
tions. The most desirable are the solutions which begin with a Big-Bang and termi-
nate at a Big-Rip. This is because they preserve all standard Hot-Big-Bang scenario
results and agree with the observational data, which suggests that there was a turn-
ing point of the evolution of the universe (the standard matter stopped dominating
against phantom) just at the redshift z = 0.46. 1 However, as shown explicitly,3
other interesting phantom cosmologies, due to its strongly repulsive contribution
to the dynamics appear. Among them there are: Einstein Static Universe with two
monotonic solutions - one of them is monotonic towards a Big-Bang and another is
monotonic towards a Big-Rip; a monotonic solution which transits between the two
Einstein Static Universes (both ways); non-singular oscillating solutions etc.

3. Quantum phantom cosmologies and the large-scale quantum

effects
Due to the new types of classical trajectories it is also interesting to ask about
possible quantum cosmological implications of phantom dark energy. Despite the
 1763

large size of the universe it is advisable to think of quantum effects in the region of
Big-Rip singularity since according to (2) the energy density may reach the Planck
scale of 10 19 GeV there. In fact, it was already shown 9 that there is a dispersion of
the wave packets in configuration space (with classical time coordinate eliminated)
at Big-Bang, Big-Crunch and at the maximum of expansion point in recollapsing
models. The question was studied whether such quantum effects may also appear
at a Big-Rip and at the minimum point of expansion. 4 It emerged that the answer
is positive for Big-Rip, but negative so far for the minimum point of expansion.
The latter might be due to the simple form of the phantom potentials taken into
account, and the question of such quantum effects is still open. Besides, it seems
reasonable to ask if spreading of the wave packets in configuration space signaling
large-scale quantum effects may appear at the turning point of the expansion - the
one for z = 0.46 at which standard matter loses domination against the phantom.
This is the matter for future work.

4. Discussion
Apart from quintessence (-f} < P < 0), global acceleration of the universe gives
strong observational motivation for a non-standard phantom (p < -f}) type of
matter as a candidate for dark energy. Phantom dark energy may dominate the
evolution of the universe and additionally lead it to a Big-Rip singularity - a state
in which the whole matter pulled apart, though very dense. A Big-Rip singularity
may be dual to a Big-Bang/Big-Crunch singularity. Finally, quantum effects may
smear out Big-Rip singularity due to large-scale quantum effects.

Acknowledgments
This research has partially been supported by the Polish Ministry of Science and
Education grant No 1P03B 043 29 (years 2005-07).

References
1. J.L. Tonry et al., Astroph. 1. 594, 1 (2003); M. Tegmark et al., Phys. Rev. D69,
103501 (2004).
2. RR Caldwell, Phys. Lett. B 545, 23 (2002); RR Caldwell, M. Kamionkowski, and
N.N. Weinberg, Phys. Rev. Lett. 91, 071301 (2003).
3. M.P. Di}browski, T. Stachowiak and M. Szydlowski, Phys. Rev. D68, 103519 (2003);
M.P. Di}browski and T. Stachowiak, Ann. Phys. (N. Y) 321, 771 (2006).
4. M.P. Di}browski, C. Kiefer and B. Sandhi:ifer, Phys. Rev. D74, 044022 (2006).
5. S.M. Carroll, M. Hofman, and M. Trodden, Phys. Rev. D 68, 023509 (2003); S.D.H.
Hsu, A. Jenkins, and M.B. Wise, Phys. Lett. B 597, 270 (2004); J.M. Cline, S. Jeon,
and G.D. Moore, Phys. Rev. D70, 043543 (2004).
6. V.A. Rubakov, hep-th/0604153; P. Creminelli et al., hep-th/0606090.
7. K.A. Meissner and G. Veneziano, Mod. Phys. Lett. A 6, 1721 (1991).
8. J.E. Lidsey, Phys. Rev. D 70, 041302 (2004).
9. C. Kiefer, Phys. Rev. D38, 1761 (1988); C. Kiefer and H.D. Zeh, Phys. Rev. D51,
4145 (1995).
 THE GENERALIZED SECOND LAW IN DARK ENERGY
DOMINATED UNIVERSES

GERMAN IZQUIERDO' and DIEGO PAVONt

Department of Physics,
Autonomous University of Barcelona,
08193 Bellaterra (Barcelona),
Spain
• german. [email protected]
t [email protected]

The generalized second law of gravitational thermodynamics is examined in scenarios

where the dark energy dominates the cosmic expansion. For quintessence and phantom
fields this law is fulfilled but it may fail when the dark energy is in the form of a Chaplygin
gas. However, if a black hole is allowed in the picture, the law can be violated if the field
is of phantom type.

Keywords: Theoretical cosmology, black holes, thermodynamics

1. Introduction
Nowadays, the overwhelming observational evidence suggests that the Universe is
undergoing an accelerated expansion driven by some form of energy (dubbed "dark
energy") that violates the strong energy condition, p+3p > O. What is more, models
in which this field also violates the dominant energy condition (DEC), p + p > 0
-dark energy of "phantom" type- appears marginally favored by the data. This short
Communication briefly summarizes our research about the validity of generalized
second law (GSL) of gravitational thermodynamics in spatially flat, dark energy
dominated, FLRW universes. 1 ,2 This law asserts that the entropy of matter and
fields within the event horizon plus the entropy of the horizon is a never decreasing
quantity. Event horizons, ofradius RH = a(t) It"
dt' /a(t'), are unavoidable features
of ever-accelerating cosmologies and are widely assumed to possess thermodynamic
properties as temperature and entropy, though this has been shown in a rigurous
manner for the de Sitter horizon only.3

2. The GSL in accelerated universes

\Ve begin by considering a phantom dominated universe with equation of state
parameter w = constant < -1. Its horizon radius RH = n/[(l + n)H], with n =
-2/[3(1 + w)], decreases with expansion whence its associated temperature, TH =
n/[(l + n)27TRHl augments with time while its entropy, SH = 7TR~, diminishes.
However, the entropy of the phantom, assumed in thermal equilibrium with the
horizon, calculated through Gibbs' equation 4 exactly compensates the decrease of
the horizon entropy, S = -SH, whereby the GSL is satisfied. (It is noteworthy that,
in agreement with previous authors,5 the entropy of the phantom fluid is negative.
However, strange as it may be, this is not forbidden by any thermodynamical law). A
parallel study for non-phantom dark energy dominated cosmologies, with -1 < w =

1764
 1765

constant < -1/3 (e.g., "quintessence"), yields identical result only that, now, Sis
positive and decreasing, and SH augments.
As an example of a phantom-dominated expansion with variable w we consider
the model of Sami and Toporensky6 which features a scalar field with negative
kinetic energy and potential V = Va qP, where 0 < a :::; 4. A similar analysis shows
that S+ SH ~ 0, the equality sign holding just for t ----7 00, i.e., when RH vanishes.
Again, S is negative and increasing and such that ISI/SH ~ l.
The Chaplygin gas,7 of equation of state p = -A/ p, corresponds to a non-
phantom dark energy field with w = -Aa6 /(Aa 6 + B). Here A and Bare positive-
definite constants. In this case, the radius of the horizon increases with expansion
up to the de Sitter value, H- 1 , for t ----7 00, as it should -see, Fig. 3 of ReU-, and
the GSL is fulfilled for a ~ [2.509 B / AP/6 but it may be violated at earlier times.
We next consider the impact of a small black hole within the event horizon in
phantom dominated universes on the validity of the GSL. By "small" we mean that
the black hole mass is much lower than the phantom energy inside the horizon,
i.e., M / E¢ « 1, so that neither the scale factor nor the event horizon radius gets
significantly modified. As is well known, Schwarzschild black holes, immersed in
a phantom environment, are bound to lose mass by accreting phantom energy at
a rhythm if = -167fM 2¢2 -see Ref.8 Therefore, the black hole entropy, SBH =
47f M2, will necessarily decrease. Thus, for scenarios with w = constant < -1 it
follows that S + SBH + SH < 0, i.e., the GSL is violated.
It remains to be seen whether it will be also violated when w varies with time.
To this end we consider again the model of Sami and Toporensky.6 In this scenario,
SBH = -87fooH 2 M3 x-I, with x == 47fq} / a, and the GSL is satisfied provided the
black hole mass does not exceed the critical value, 2

which decreases with time at fixed a much slowly than M. This implies that, re-
gardless the initial mass of the black hole, sooner or later we will have M > Mer
while the condition M / E¢ « 1 still holds -see Fig. 1 in Ref. 2 As a consequence the
GSL will be violated. At some point further ahead the said condition will no longer
be met and our analysis will break down. From this point on we can say nothing
about the validity of the GSL.

3. Discussion
The assumption of thermal equilibrium between the dark energy and the event
horizon may seem artificial. However, this condition must be fulfilled for the entropy
concept to be meaningful. In other words, the entropy is an exclusive property of
1766

equilibrium systems whence the entropy of two systems cannot be meaningfully

added unless they are into equilibrium with one another.4
In view of the failure of the GSL in phantom dominated scenarios when black
holes are present different reactions may arise: (i) Some phantom energy fields
might be physical but not those considered in this Communication. Indeed, several
predictions lending support to phantom fields may have come from an erroneous
interpretation of the observational data. 9 (ii) The GSL was initially formulated for
systems complying with the DEC, so there is no reason why it ought to be satisfied
by systems that violate it. (iii) Strictly speaking, a general proof of the GSL even for
systems complying with the DEC is still lacking, 10 therefore we should not wonder
at its failure in some particular cases.
It is for the reader to decide which of these alternatives, if any, is more to his/her
liking.
Yet, one may argue that it is unclear that black holes retain their thermodynamic
properties (entropy and temperature) in presence of a field that does not comply
with the DEC. In such an instance, one may think, that there is no room for the black
hole entropy in the expression for the GSL. However, the latter is often formulated
by replacing SBH by the black hole area. Again, this variant of the GSL will fail in
the two cases of above. To sum up, if eventually phantom energy is shown to be a
physical reality, it will pose a serious threat to the generalized second law.

References
1. G. Izquierdo and D. Pavon, Phys. Lett. B 633,420 (2006).
2. G. Izquierdo and D. Pavon, Phys. Lett. B 639, 1 (2006).
3. G. Gibbons and S.W. Hawking, Phys. Rev. D 15,2738 (1977).
4. H. Callen, Thermodynamics (J. Wiley, N.Y., 1960).
5. M.D. Pollock and T.P. Singh, Class. Quantum Gmv. 6, 901 (1989); 1. Brevik et al.,
Phys. Rev. D 70, 043520 (2004); J.A.S. Lima and J.S. Aleaniz, Phys. Lett. B 600, 191
(2004).
6. M. Sami and A. Toporensky, Mod. Phys. Lett. A 19, 20 (2004).
7. A. Kamenshchik, U. Moschella and V. Pasquier, Phys. Lett. B 511, 265 (2001).
8. E. Babichev, V. Dokuchev and Yu. Eroshenko, Phys. Rev. Lett. 93, 021102 (2003).
9. S. Das, P.S. Corasaniti, and J. Khoury, Phys. Rev. D 73,083509 (2006).
10. R.M. Wald, Quantum Field Theory in Curved Spacetime and Black Hole Thermody-
namics (University of Chicago Press, Chicago, 1994).
 ACCELERATED EXPANSION BY NON-MINIMALLY COUPLED
SCALAR FIELDS

ROGER BIELI
Max Planck Institute for Gravitational Physics
Am Muhlenberg 1, 14476 Golm, Germany

In a class of spatially homogeneous cosmologies including those of Bianchi type I-VIII

mathematical results are presented which show that a scalar field non-minimally cou-
pled to the scalar curvature of spacetime can dynamically yield a positive cosmological
constant without the potential being required to include one. More precisely, it is shown
that in an exponential potential any positive coupling constant leads eventually to late-
time de Sitter expansion and isotropization corresponding to a positive cosmological
constant and that this behaviour is independent of the steepness of the potential. This
is in marked contrast to the minimally coupled case where power-law inflation occurs at
most, provided the potential is sufficiently shallow.

Keywords: Dark energy; quintessence; non-minimal coupling.

1. Introduction
An ongoing major effort m theoretical cosmology is to satisfactorily explain the
observed late-time acceleration of the universe. For an explanation to be considered
satisfactory the underlying model together with its parameters is commonly required
to come with good motivation from particle physics theories. This is because, up to
now, all observational data are fully consistent with the ACDM concordance model
having a positive cosmological constant driving the expansion. Every other model
that is capable ofreproducing this simple behaviour, e.g. a scalar field in a potential
with a positive lower bound,l can presently not be ruled out experimentally and
cosmologists are therefore endeavoring to find models that naturally originate from
fundamental physics. However, it seems quite intricate to construct a viable particle
physics theory that yields a positive cosmological constant - or, more generally, a
non-vanishing lower bound on the potential of some effective field - comparable
to the observed value. It is the aim of this proceedings contribution to give an
example of a cosmological model that is well-motivated from a quantum gravity
point of view and includes a scalar field that drives de Sitter expansion at late
times. The potential itself does not contain a cosmological constant, rather, one is
established dynamically by a non-minimal coupling of the field to the spacetime
scalar curvature. This mechanism has previously been considered heuristically by
Tsujikawa2 in an inflationary context for the purpose of sustaining inflation in steep
potentials. Here, mathematical results from Ref. 3 are presented that generalize
these findings in particular to spatially homogeneous models of Bianchi type 1-VIn
and apply them to obtain late-time exponential acceleration.

1767
1768

2. The model
A direct coupling of a real scalar field ej; to the spacetime scalar curvature R can be
realized by an additional term _~Rej;2 /2 in the action

where * is the Hodge dual, V the scalar field potential and Sm the action for all
other forms of matter which are supposed to satisfy the dominant and strong energy
conditions. Such a term arises regularly in quantum field theory in curved spacetimes
on renormalizability grounds. 4 Because the coupling constant ~ is dimensionless it
persists in the classical limit and it is thus sensible to include the coupling term
already in the effective action. If the dimension of spacetime is n + 1 with n 2:: 3,
the prominent case of conformal coupling corresponds to ~ = (n - 1) /4n.
What remains to be specified is the potential V for the scalar field. It will be
taken exponential, V(ej;) = Aexp(-f\:ej;) > 0, with f\: > 0 being referred to as the
steepness of the potential. The reason for taking simple exponentials is that, on one
hand, they are easily motivated as originating from Kaluza-Klein type dimensional
reductions or conformal rescalings and on the other hand, the late-time asymptotics
of the respective models is sufficiently well understood in the minimally coupled case
due to the work of Kitada and Maeda. 5 Specifically, it is known that all initially ex-
panding solutions of Bianchi type I-VIII approach a power-law inflationary solution
if 0 < f\: < V2 and that the field ej; diverges to infinity.

3. Late-time asymptotics
Suppose that there is an n-dimensional Lie group G acting simply transitively on
the surfaces of spatial homogeneity such that the spacetime Ai can be written as
A1 = G x 1 where 1 = [to, oo[ shall be an interval of comoving time not bounded
from above. Assume that G allows only for left invariant Riemannian metrics of
non-positive scalar curvature. For n = 3 this restricts to groups of Bianchi type 1-
VIII. Let a metric g, an energy-momentum tensor T and a scalar field ej; be given as
families, E C 2(1, T~g), p E C 1 (1), j E C 1 (1, TPg), S E C 1 (1, T~g) and ej; E C 2 (1)
of tensors on the Lie algebra 9 of G by

9= 7f*, - dt 0 dt, T = 7f* S - 7f* j 0 dt - dt 0 7f* j + p dt 0 dt

such that the Einstein-scalar field-matter equations according to the action above

(1 - ~ej;2) ( Rc - ~R9) = (1- 2~)Vej; 0 Vej; + (2~ - ~) IVej;I~g

- 2~ej;V2ej; + 2~(ej;Dej;)g - V(ej;)g +T
Dej;-~Rej;-V'(ej;)=O

divT = 0
 1769

are satisfied. Re is the Ricci tensor, R the scalar curvature, \7 the covariant deriva-
tive, D the covariant wave operator and I . I~ the induced fibre metric, all with
respect to g. ]f and t are the canonical projections from M onto its factors and ]f* is
the pullback of families of covariant tensors on 9 back to M. The Hubble parameter
H is given by H = tr "y 12n.
If for ~ > 0, initially 1 - ~¢2(tO) > 0 and H(to) > -[log(I - ~¢2)]"(to)/(n - 1)
hold, at late times the field ¢ approaches

with its first two derivatives vanishing, the Hubble parameter H monotonically
decreases towards
2 V( ¢co)
H co .-
.-
n(n - 1) 1 - ~¢~

and the spatial curvature, the shear as well as matter density p, flux Ijl, and pressure
tr S decay exponentially. In fact, the deceleration parameter q converges to -1
and exponential acceleration and isotropization occurs asymptotically mimicking a
positive cosmological constant of magnitude
A = Aexp(-li¢co)
dyn 1- ~¢~ .
Note that this is true irrespective of the steepness Ii of the potential and for any
arbitrarily small positive coupling constant ~, in whose limit ~ ---> 0 the expansion
Adyn = (I/~) exp( -.lil ~)(C + O( ~)) holds for a constant C> o.
The results stated above can be obtained by applying a cosmic no-hair theorem 3
in a conformally related frame, in which the field is minimally coupled to gravity but
directly coupled to the ordinary matter instead. It might be interesting that scalar
fields non-minimally coupled to matter were, apart from being conformally related
to certain scalar-tensor theories of gravity, studied per se as they can yield scaling
dark energy6 or hide a putative fifth force from laboratory experiments. 7 Another
consequence of the convergence of the field ¢ to a finite value ¢co worth noting is
that, given the hypotheses above, the quantity 1 - ~¢2 stays bounded away from
zero for all times, a topic relevant for the boundedness of the effective gravitational
constant. 8

References
1. A. D. Rendall, Class. Quantum Grav. 21 (2004) 2445-2454
2. S. Tsujikawa, Phys. Rev. D 62 (2000) 043512
3. R. Bieli, Class. Quantum Grav. 23 (2006) 5983-5995
4. C. G. Callan, Jr., S. Coleman and R. Jackiw, Ann. Phys. 59 (1970) 42-73
5. Y. Kitada and K. Maeda, Class. Quantum Grav. 10 (1993) 703-734
6. L. Amendola, Phys. Rev. D 60 (1999) 043501
7. J. Khoury, A. Weltman, Phys. Rev. D 69 (2004) 044026
8. A. A. Starobinsky, Pis'ma Astron. Zh. 7 (1981) 67-72
 VACUUM ENERGY GENERATING MECHANISMS IN COSMIC
EXPANSION WITH NATURAL UV CUTOFF

ACHIM KEMPF*
Department of Applied Mathematics, University of Waterloo,
Waterloo, Ontario, N2L 3Cl, Canada

In expanding space-times with a natural ultraviolet cutoff, new comoving modes are
continually being generated for each field and, a priori, each mode should come with
gravitating ground state energy. Once generated, the ground state energy of a mode
becomes diluted as its proper wavelength increases. Given a model for quantum gravity,
the balance of the two effects can be nontrivial depending, for example, on the effective
dispersion relations close to the scale of the natural UV cutoff.

As is well known, general relativity and quantum theory together indicate that the
concept of distance loses operational meaning at the Planck scale: test particles
aimed to resolve a region of Planckian size would possess sufficient momentum
uncertainty to significantly randomly curve and thereby disturb the very region
in space that the test particles were meant to resolve. The existence, then, of a
shortest length in nature should have cosmological implications. This is because
in an expanding universe the independent modes are comoving modes, i.e., modes
whose wavelength expands with the universe. If there exists a finite minimum length,
(say for observers that are at rest in comoving coordinates) then new comoving
modes are continually being created. More precisely, new modes are continually
entering the regime of physical wavelengths that are large enough so that the usual
framework of QFT in curved spacetime holds. A priori, each new mode comes with
its own ground state energy. This energy should of course contribute to the overall
vacuum energy and it should gravitate. At the same time, each co moving mode's
vacuum energy decreases as the mode's physical wavelength increases.
The balance of vacuum energy creation and dilution through expansion should
naively be expected to be zero. Thus, roughly speaking, if the universe had its
quantum origin at zero energy (in the sense of how much gravitates), which is not a
fine-tuned value, then this would explain why the universe would stay at that value.
Notice that, in this scenario, the overall vacuum energy is essentially zero while the
ground state energy of field modes is real and, in particular, it is gravitating. Let
us recall that condensed matter physics indeed provides reasons to expect that the
ground state energy of field modes does gravitate:
The ground state energy of a quantum harmonic oscillator can be changed not
only by changing the parameters of the Hamiltonian (as would be the case for co-
moving modes during expansion). It can also be lowered by coupling the oscillator

*This research has been partially supported by CFI, OrT, PREA and the Canada Research Chairs
Program of NSERC

1770
 1771

to degrees of freedom which decrease the amplitude of its zero-point fluctuations.

This happens, e.g., when these extra degrees of freedom effectively increase the oscil-
lator's mass. An example would be the vibrational oscillator of a diatomic molecule
that captures a neutron in one of its nuclei. Similarly, the ground state energy of
field modes of the electromagnetic field can be lowered by suppressing the zero point
fluctuations of some of its electromagnetic field modes. As Casimir! showed, this
happens, for example, when the electromagnetic field couples to conducting charge
carriers which are confined to two parallel plates. In this case the zero point fluctua-
tions of certain modes in the direction orthogonal to the plates become suppressed.
Let us now consider a Gedanken experiment. Consider two parallel plates con-
sisting of high temperature superconducting material (a ceramic such as YBCO).
Assume that the plates are separated by vacuum and freely floating in space with
no relative velocity. Let their temperature be initially just above Tc and now let us
lower their temperature to just below T e , which is possible with negligible energy
budget. The plates started as insulators but now the Casimir force between the
newly superconducting plates sets in and the plates therefore accelerate towards
another. Thereby the two plates gain kinetic energy, which certainly gravitates.
Interestingly, this energy had initially been locked up in the form of ground state
energy of electromagnetic modes. Thus the ground state energy of field modes should
also be assumed to gravitate.
Returning to the question of the balance of the expansion-induced vacuum en-
ergy creation and vacuum energy dilution, it is clear that the two may not offset
another completely, depending on the model of quantum gravity. Intuitively, some-
one who keeps buying land by taking on loans should not be getting any richer,
since the continual gain of land is offset by continually growing debt. Realistically,
however, the values of land and money evolve nonproportionally, and therefore that
person generally accumulates a net gain or net loss. Similarly, the predicted energy
balance in cosmic expansion may be nontrivial, for example due to nontrivial effec-
tive dispersion relations close to the UV cutoff scale, depending on the model for
quantum gravity that one uses. It is tempting to speculate about a possible relation
to the origin of the inflaton potential or to late-time acceleration.
In principle, any candidate quantum gravity theory,4 such as loop quantum
gravity or string theory should show exactly how quantum field theory behaves when
approaching the scale of a natural ultraviolet cutoff. For a simple model for physics
approaching the cutoff scale we recall that any theory can generally be formulated
entirely in terms of the (scalar, vector, tensor or operator- ) valued functions over the
set of spacetime points. As first proposed in,3 fields in space-time could be functions
over a continuous manifold while, crucially, the class of fields that can occur may
be naturally restricted such that if a field is sampled only at discrete points then
its amplitudes can already be reconstructed at all points in the manifold - if one
chooses sampling points that are spaced densely enough. The maximum average
sample spacing which allows one to reconstruct the continuous field from discretely
taken samples defines the scale of the natural UV cutoff. The mathematics of classes
1772

of functions which can be reconstructed from discrete samples is well-known, namely

as sampling theory, in the information theory community, where it plays a central
role in Shannon's theory of continuous information, see. 5
The descrip~ion of a minimum length in nature as a sampling theoretical max-
imum information density can indeed be implemented6 in quantum field theory
and it thereby yields explicit candidate models of quantum field theory close to
the Planck scale. The calculations 7 were done for certain FRW background space-
times and the implications for the predictions within standard inflation have been
studied in detail. 8 Of interest to us here is that, as expected, comoving modes
are continually being created. The modes' wave equations are still of the form
1]?/1 + m(TJ)I]?! + n(TJ)¢ = 0 but each mode now possesses a starting time, TJo, which
is when the comoving mode's physical wave length is the length of the UV cutoff.
Mode generation happens by continually adding the new modes' af and a1 operators
to the fields and thereby also to the Hamiltonian. Schematically:

( I]?~II
k k
- - f»k

Here, a(TJ) is the scale factor. As expected, the dilution of each mode's vacuum
energy as its proper wavelength increases is countered by the continual production
of vacuum energy as modes enter the region covered by the integral. A detailed
study of the balance of vacuum energy creation and dilution in this model of a UV
cutoff is in progress. Notice that a sampling theoretic natural UV cutoff can be fully
covariant. 9

References
1. H.B.G. Casimir, D. Polder, Phys. Rev. 73, 360 (1948)
2. A. Kempf, Proc. 10th Marcel Grossmann Meeting, Ed. R. Ruffini, 2271, World Scien-
tific (2005)
3. A. Kempf, Phys. Rev. Lett. 85, 2873 (2000), hep-th/9905114
4. E. Witten, Phys. Today 49,24 (1996), F. Markopoulou, L. Smolin, Phys. Rev. D70,
124029 (2004), gr-qc/0311059
5. J.J. Benedetto, P.J.S.G. Ferreira, Modern Sampling Theory, Birkiiuser (2001)
6. A. Kempf, J. Math. Phys. 35, 4483 (1994), hep-th/9311147, A. Kempf, G. Mangano,
R. B. Mann, Phys. Rev. D52 H08 (1995), hep-th/9412167, A. Kempf, G. Mangano,
Phys. Rev. D55, 7909 (1997), hep-th/9612084
7. A. Kempf, Phys. Rev. D63, 083514 (2001), astro-ph/0009209, A. Kempf, J. C.
Niemeyer, Phys. Rev. D64 103501 (2001), astro-ph/0103225
8. R. Easther, B. R. Greene, W. H. Kinney, G. Shiu, 2001, Phys. Rev. D64, 103502,
hep-th/0104102, R. Easther, B. R. Greene, W. H. Kinney, G. Shiu, 2002, Phys. Rev.
D66, 023518, hep-th/0204129, A. Kempf, L. Lorenz, Phys. Rev. D74, 103517 (2006),
gr-qc/0609123
9. A. Kempf, Phys. Rev. Lett. 92, 221301 (2004), gr-qc/0310035
 NEW KINEMATICAL CONSTRAINTS
ON COSMIC ACCELERATION

DAVID RAPETTI, STEVE W. ALLEN, MUSTAFA A. AMIN and ROGER D. BLANDFORD

Kavli Institute for Particle Astrophysics and Cosmology,
382 Via Pueblo Mall, Stanford, CA, 94305-4060, USA [email protected]

We present and employ a new kinematical approach to 'dark energy' studies. We con-
struct models in terms of the dimensionless second and third derivatives of the scale
factor a(t) with respect to cosmic time t, namely the present-day value of the decel-
eration parameter qO and the cosmic jerk parameter, j(t). An elegant feature of this
parameterization is that all ACDM models have j(t) = 1 (constant), which facilitates
simple tests for departures from the ACDM paradigm. Applying our model to redshift-
independent distance measurements, from type Ia supernovae and X-ray cluster gas mass
fraction measurements, we obtain clear statistical evidence for a late time transition from
a decelerating to an accelerating phase. For a flat model with constant jerk, j(t) = j, we
measure qo = -O.Sl ± 0.14 and j = 2.16:+::g:~~, results that are consistent with ACDM
at about the 1cr confidence level. In comparison to dynamical analyses, the kinemati-
cal approach uses a different model set and employs a minimum of prior information,
being independent of any particular gravity theory. The results obtained with this new
approach therefore provide important additional information and we argue that both
kinematical and dynamical techniques should be employed in future dark energy studies,
where possible.

1. Introduction
Late-time acceleration of the Universe is now an observed fact. 2 ,4,9,11 Most current
analyses of cosmological data assume General Relativity and the Friedmann equa-
tions and employ the mean matter density of the Universe, Om, and the dark energy
equation of state, W, as model parameters. Other dynamical analyses employ modi-
fied Friedmann equations for a particular gravity model. However, a purely kinemat-
ical approach is also possible that does not assume any particular gravity theory.
Kinematical models provide important, complementary information when seeking
to understand the origin of the observed late-time accelerated expansion. 1,6-8,11-14
In Rapetti et al. (2007)10 we develop an improved method for studying the kine-
matical history of the Universe. Instead of using parameterizations constructed in
terms of the deceleration parameter q(z), we introduce a new kinematical frame-
work using the cosmic jerk,5 the dimensionless third derivative of the scale factor
with respect to cosmic time. We apply our method to the 'gold' sample of type
Ia supernovae (SNIa) measurements,l1 the SNIa data from the first year of the
Supernova Legacy Survey (SNLS) project,4 and the X-ray galaxy cluster distance
measurements of Allen et al. (2007).3

2. A new kinematical framework

We rewrite the defining equation for the jerk parameter, j(a) = (a 2H2)" j2H 2, in
a more convenient form 5 a2V"(a) - 2j(a)V(a) = 0 where prime denotes derivative
with respect to a and V(a) is defined as V(a) = _a 2H2 j2H'5. We specify the two
constants of integration required by this differential equation in terms of the present
1773
1774

6 -0.2
-0.4
-0.6
4
-O.S
-1
;: -1.2
-1.4
1 - - - --
-1.6

o -1.S
-2
-1
-2.2
-1.5 -1 -0.5 0.1 0.2 0.4 0.5
qo

Fig. 1. The left panel shows the 68.3 and 95.4 per cent confidence limits in the (qO,j) plane for
the kinematical model with a constant jerk, j, obtained using all three data sets: both SNla data
sets 4 • Jl and the X-ray clusters data'" The right panel shows the results in the standard (Om,w)
plane obtained using the same three data sets and assuming HST, BBNS and b priors. (Note that
the kinematical analysis does not use the HST, BBNS and b priors). The dashed lines show the
expectation for a cosmological constant model in both formalisms (j = 1, w = -1, respectively).

Hubble parameter Ho and the present deceleration parameter go, V(l) = -1/2 and
V ' (l) = go where aCto) = 1 at the present time to. Here the first condition comes
from H(l) = Ho and the second from V ' (l) = -(Hb/ Ho) - 1 = go. Allowing a
constant deviation from ACDM (j = 1), i.e. a constant j model (for more compli-
cated j(a) models see Rapetti et al. (2007)10), we solve the jerk differential equation
analytically obtaining

v (a) = - V; [( p ~ U ) aP + (p ~ U ) a -P ] (1)

where p == (1/2))(1 + 8j) and U == 2(go + 1/4). The Hubble parameter, H(a),
obtained from (1) is used to calculate the angular diameter (d A ) and luminos-
ity (d L ) distances for a fiat Friedmann-Robertson-Walker-Lemaitre (FRWL) metric
dA(a) = a 2 dda) = c/Hoa Ja11/(a2E(a)da, where c is the speed of light and
E(a) = H(a)/ Ho. These theoretical distances dda) and dA(a) are then used to fit
the data (for details about the data analysis see Rapetti et al. (2007)10).

3. Results
Combining all three data sets, we obtain tight constraints on go = -0.81 ± 0.14 and
j = 2.16:+:g:~;. Our result represents the first measurement of the jerk parameter
from cosmological data. Our dynamical analysis of the same three data sets gives
w = -1.15:+:g:i~ and Om = 0.306:+:g:g!~. Figure 1 shows the constraints for both
the kinematical (go,j; left panel) and dynamical (Om,w; right panel) models, using
all three data sets combined. In both cases, the dashed lines indicate the expected
range of results for ACDM models (i.e. a cosmological constant). It is important to
recognise that the results from the kinematical and dynamical analyses constrain
 1775

different sets of departures from ACDM. We are using two simple, but very different
parameterizations based on different underlying assumptions.

4. Conclusions
We have developed a new kinematical approach 10 to study the expansion of the
history of the Universe. Our technique uses the parameter space defined by the
current value of the cosmic deceleration parameter qo and the jerk parameter j.
The use ofthis (qO, j) parameter space provides a natural framework for kinematical
studies. In particular, it provides a simple prescription for searching for departures
from ACDM, since the complete set of ACDM models are characterized by j = l.
Using type Ia supernovae and X-ray galaxy clusters data and assuming geometric
flatness, we measure qo = -0.82 ± 0.14 and j = 2.16!g:~~ (Figure 1). Note that this
represents the first measurement of the cosmic jerk parameter, j. We suggest that
future studies should endeavour to use both kinematical and dynamical approaches
where possible, in order to extract the most information from the data. The com-
bination of techniques may be especially helpful in to distinguish between an origin
for cosmic acceleration that lies with dark energy (i.e. a new energy component to
the Universe) from modifications to General Relativity.

Acknowledgments
The computational analysis was carried out using the KIPAC XOC computer clus-
ter at SLAC. SWA acknowledges support from the National Aeronautics and Space
Administration through Chandra Award Number DD5-6031X. RDB acknowledges
support from National Science Foundation grant AST05-07732. This work was sup-
ported in part by the U.S. Department of Energy under contract number DE-AC02-
76SF00515.

References
1. U. Alam, V. Sahni, T. D. Saini, A. A. Starobinsky, MNRAS 344 1057 (2003).
2. S. W. Allen, R. W. Schmidt, H. Ebeling, A. C. Fabian, L. van Speybroeck, MNRAS
353 457 (2004).
3. S. W. Allen et al., in preparation (2007).
4. P. Astier et al., Astron.Astrophys. 447 31 (2006).
5. R. D. Blandford, M. A. Amin, E. A. Baltz, K. Mandel, P. J. Marshall, ASP Conf.
Ser. 339 27 (2004).
6. R. R. Caldwell, Kamionkowski M., JCAP 0409009 (2004).
7. T. Chiba, T. Nakamura, Pron Theor. Phys. 1001077 (1998).
8. O. Elgaroy, T. Multamaki, JCAP 0609002 (2006).
9. D. Rapetti, S. W. Allen, J. Weller, MNRAS 360 555 (2005).
10. D. Rapetti, S. W. Allen, M. A. Amin, R. D. Blandford, MNRAS 375 1510 (2007).
11. A. Riess et al., ApJ 607665 (2004).
12. V. Sahni, T. D. Saini, A. A. Starobinsky, A. Ujjaini, JETP Lett. 77201 (2003).
13. C. Shapiro, M. S. Turner, ApJ 649 563 (2006).
14. M. Visser, Class. Quant. Grav. 21 2603 (2004).
 DARK ENERGY AND DECAYING DARK MATTER*

GRANT J. MATHEWS
University of Notre Dame, Center for Astrophysics, Notre Dame, IN 46556 USA
[email protected]

NGUYEN Q. LAN
Dept. of Physics, Hanoi University of Education, Hanoi Vietnam

JAMES R. WILSON
Lawrence Livermore National Laboratory, Livermore, CA 94550, USA

We discuss a cosmology in which cold dark-matter particles decay into relativistic parti-
cles. Such decays lead to a bulk viscosity in the cosmic fluid. This bulk viscosity enters
the cosmic energy equation as an effective negative pressure. We investigate whether
this negative pressure is of sufficient magnitude to account fo the observed cosmic ac-
celeration. We have shown that a single decaying species in a A = 0, flat, dark-matter
dominated cosmology can not reproduce the observed magnitude-redshift relation from
Type La supernovae. However, late decaying particles may be able to account for a sig-
nificant fraction of the apparent cosmic acceleration.

The simple coincidence that dark matter and dark energy currently contribute
comparable mass energy toward the closure of the universe begs the question as to
whether they could be different manifestations of the same physical phenomenon
(unified dark-matter). The possibility of particular interest in this work 1 is that of
a bulk viscosity within the cosmic fluid. Such a term resists the cosmic expansion
and therefore acts as a negative pressure. Indeed, it has been shown 2 ,3 that for the
right viscosity coefficient, an accelerating cosmology can be achieved without the
need for a cosmological constant.
We have proposed 1 a simple mechanism for the formation of such bulk viscosity
by the decay of a dark matter particle into relativistic products. Such decays heat
the cosmic fluid and cause it to fall out of pressure and temperature equilibrium and
can therefore be represented by a bulk viscosity. We have computed the magnitude-
redshift relation for Type Ia supernovae in this cosmology and have shown that
a single decay does not reproduce these data unless decays are delayed, e.g. by a
cascade of particle decays, or a late decaying particle.
The physical origin of bulk viscosity in a system can be traced to deviations from
local thermodynamic equilibrium. The bulk viscosity, therefore, is a measure of the
pressure required to restore equilibrium to a compressed or expanding system. 4 - 6 It
is natural for such a term to exist in the cosmologically expanding universe anytime

·Work supported in part by the US Department of Energy under Nuclear Theory grant DE-FG02-
95ER40934. N.Q.L. also supported in part by NSF grant PHY 02-16783 for the Joint Institute for
Nuclear Astrophysics (JINA). Work at Lawrence Livermore National Laboratory performed under
the auspices of the U.S. Department of Energy under under contract W-7405-ENG-48 and NSF
grant PHY-9401636

1776
 1777

the fluid is out of equilibrium. For the cosmology proposed here, the attainment of
equilibrium as the universe expands is delayed by the gradual decay of one or more
species to another which occurs over ~ 10 10 yrs. This leads to nontrivial dependence
of pressure on density as the universe expands, and therefore a bulk viscosity.
The existence of such dissipation leads 1 to a modification of the perfect-fluid
energy-momentum tenor. The effect of bulk viscosity is to replace the fluid pressure
with an effective pressure given by, Peff = p-(3(aja). Thus, for large (it is possible
for the negative pressure term to dominate and an accelerating cosmology to ensue.
Although such a pressure term is absent from the Friedmann equation, the bulk
viscosity does appear in the conservation condition T"'~v = O. The total density for
the Friedmann equation will then include not only terms from heavy and light dark
matter, but a dissipated energy density in bulk viscosity.
Bulk viscosity can be thought of 4 ,7 as a relaxation phenomenon. It derives from
the fact that the fluid requires time to restore its equilibrium pressure from a depar-
ture which occurs during expansion. The viscosity coefficient ( depends upon the
difference between the pressure j5 of a fluid being compressed or expanded and the
pressure P of a constant volume system in equilibrium. Of the several formulations 4
the basic non-equilibrium method 8 is most consistent with Pej j as defined above.
Following the derivation in [7], we have inferred 1 the following ansatz for the bulk
viscosity of the cosmic fluid due to particle decay,

( _ [1
-PhTe -
PI
---
+ p,] 2
, (1)
P
where the square of the term in brackets comes from inserting our derivation into
the linearized relativistic transport equation of Ref. [9], and Te is an effective decay
rate defined in Ref. [1].
To avoid observational constraints the decay products must have very little en-
ergy in photons or charged particles. Neutrinos are thus a natural candidate for
such a background which might arise from decaying heavy sterile neutrinos 10 or
SUSY dark matter. For supersymmetric dark matter candidates, the initially pro-
duced dark matter relic might be a superWIMP to produce the correct relic den-
sity. Later, this superWIMP decays to a lighter stable dark matter particle, or the
light supersymmetric particle itself might be unstable to decay. One might imagine
an R-parity violating decay in which a particle might decay by coupling to right-
handed neutrinos which then decay to normal neutrinos. Another possibility could
be gauge-mediated supersymmetry breaking involving the decay of a supersymmet-
ric sneutrino into a gravitino plus a light neutrino.
We have compared various cosmological models with some of the recent com-
bined data from the High-Z Supernova Search Team and the Supernova Cosmology
Projecty,12 We found 1 that, although the bulk viscosity indeed provides a nega-
tive pressure, a flat A = 0 cosmology with bulk viscosity from decay of a single
dark-matter species does not do better than a ACDM in reproducing the supernova
distance-red shift relation. In fact it is much worse than the usual ACDM cosmol-
1778

ogy and is even worse than a pure matter dominated cosmology. The reason for
this is that, although the bulk viscosity is substantial, it scales with the decaying
dark matter which falls off faster with time than a- 3 because of the decay. An
accelerating cosmology requires a nearly constant value of Ptot with time.
We have shown,l however, that if the emergence of the bulk viscosity can be
delayed by a cascading decay process, then some of the acceleration required by
observations of Type Ia supernovae at high redshift can be explained. However if
we consider a late-decaying dark matter particle then an excellent fit can be obtained
to the SNla luminosity-redshift relation. This can be achieved in this context if the
dark matter particle becomes unstable only at late times due to a time dependent
mass crossing or a cosmic phase transition.
Obviously, one must decide whether the dilemma of a cosmological constant is
less plausible than the dilemma of a bulk viscosity produced by delayed decaying
dark matter. Here, we merely argue that the possibility exists. Nevertheless, having
established that a possible paradigm exists, are examining the possible influence of
this scenario on the CMB and the growth of large scale structure. We have employed
a model similar to that of Ref. [13] that includes the decay of heavy neutrinos. Our
analysis indicates that the cosmic acceleration in a universe with dark matter decay
may be enhanced by its effect on large scale structuret.

References
1. J. R. Wilson, G. J. Mathews, and G. M. Filler, Phys. Rev.D75 043521 (2007), astro-
ph/0609687
2. R. F. Sawyer, Phys. Rev. D74, 043527 (2006), astro-ph/0501525.
3. J. C. Fabris, S.V.B. Goncalves, and R. de Sa Ribeiro, astro-ph/0503362.
4. H. Okumura and F. Yonezawa, Physics A, 321, 207 (2003).
5. P. Ilg and H. C. Ottinger, PRD, 61, 023510 (1999).
6. C. Xinzhong and E. A. Spiegel, Mon. Not. T. Astron. Soc., 323, 865 (2001).
7. S. Weinberg, Gravitation and Cosmology: Principles and Applications of the General
Theory of Relativity, (John Wiley & Sons, Inc. New York) (1972); Astrophys. J., 168,
175 (1971).
8. W. G. Hoover, A. J. Ladd, R. B. Hickman, B. L. Holian, Phys. Rev. A21, 1756 (1980);
W. G. Hoover, D. J. Evans, R. B. Hickman, A. J. Ladd, W. T. Ashurst, B. Moran,
Phys. Rev. A22, 1690 (1980)
9. L. H. Thomas, Quart. J. Math. (Oxford), 1, 239 (1930).
10. K. Abazajian, G. M. Fuller, and M. Patel, Phys. Rev., D64, 023501 (2001).
11. P. M. Garnavich, et al., Astrophys. J., 509, 74 (1998); S. Perlmutter, et al. Nature,
391, 51 (1998).
12. A. G. Riess, et al., Astrophys. J., 607, 665 (2004).
13. J. M. Centrella and J. R. Wilson Astrophys. J Supp., 54, 229 (1984).
 GRAVITATIONAL INSTANTON SOLUTION
TO COSMOLOGICAL TERM

SHE-SHENG XUE
ICRANet, Piazzale della Repubblica 10-12, 65122 Pescara, Italy
[email protected]

We study the gravitational instanton solution to the problem of cosmological term.

Finite temperature and action. The attractive nature of gravity that cannot be
screened is the essential reason for many inevitable instabilities of classical gravi-
tating systems, such as the gravitational collapse and Jean instability. One might
also worry about the instability of classical solutions of the Einstein equation against
quantum fluctuations. To explore the structure of quantum (thermal) fluctuations
about these classical solutions of the vacuum Einstein equation, one starts with the
Euclidean functional integral for quantum gravity,

Z = J V[gab(x)]e-SE(g)+gauge-fixingterms. (1)

In the canonical ensemble at temperature T = 1/ (J, the metrics gab in (1) are strictly
periodic in the Euclidean time T with period (J: gab( T, x) = gab( T+ (J, x). The saddle-
point evaluation of the functional integral Z starts by constructing the saddle-point
of the action, namely, the classical solutions g~b of the vacuum Einstein equation
and R = O. Expansion about these saddle points is performed by writing

(2)

where ¢ab are small quantum (thermal) fluctuation fields about the saddle-point
metricg~b' colloquially termed a gravitational instanton. Up to the lowest order,
i.e., the quadratic terms of quantum fields ¢ab, Eq. (1) becomes

where SE(gS) is the action of gravitational instantons and N(3 is the number of
period (J.
In addition to the trivial periodic solution: flat space g~b (x) = Oab (trivial in-
stanton), there is another periodic solution: g~b (x) is the Euclidean section of the
Schwarzschid solution (non-trivial instanton), 1

(4)

in Boyer-Lindquist coordinates (t,r,e,¢), where M is not a fixed mass parameter,

but rather determined by inverse temperature {J,2

{J = 87rGM. (5)

1779
1780

This gravitational instanton (4) is a special classical field configuration. The classical
Euclidean action of such an instanton is non-zero,2

(6)
For N-instanton sectors with different topological numbers, the finite action is
NSE(gS), where the Coulomb interaction between instantons is neglected.
Instability and vacuum transition. The quantum (thermal) fluctuations ¢ab about
the Schwarzschid instanton gS are unstable 2,3 This instanton is therefore not a strict
minimum, rather an unstable saddle point of the action. In Ref.,2 this instability
is interpreted as a decay of fiat space at finite temperature T, and the decay pro-
cess proceeds by quantum (thermal) fiuctuations spontaneously nucleating quantum
black holes (QBHs) of radius R = (47fT)-1 and mass M = (S7fGT)-l. The decay
rate per unit of volume V is

r
_
=
4
d N
(g)1/2dTd3x C::'
0.S7
73(11/3)
212/45 ( m~
647f 3
) [ S ]
exp -SE(g ) . (7)

The canonical partition function Z(N) (gS) of N-instanton sectors [see (3)J is pro-
portional to the transition amplitude from the vacuum of zero-instanton (Oab fiat
space) to the vacuum of N-instanton gS at the finite temperature T,

Z(N)(gS) = (Ng~b' OIOab, Ohio ex: e-NSE(gS) = e-~(3MN. (S)

(Oab, OIOab, 0h=o
Summing over contributions of all instanton sectors (S), the total vacuum transition
amplitude has the form

exp [i~llTV] , (9)

and the instanton number N = rv llT.

Cosmological term Suppose that the space-time symmetry of Universe at temper-
ature T is described by the Euclidean Robertson-Walker metric with zero curvature
k = 0 and the scale factor a(T): g{J,(x) = a(T)c5ab , which is strictly periodic in Eu-
clidean time T with period (3. The scale factor a(T) is a time-independent constant
at a fixed temperature T. Thus the fiat Robertson-~Walker metrics g{J, = a(T)c5ab ,
is a trivial instanton, which has zero action SE(gR) = O. As discussed, the non-
trivial N instantons (4) has the finite action N S E (gs) at temperature T. The fiat
Robertson-Walker metrics g;;b = a( T )c5ab at temperature T is not stable against
quantum (thermal) fiuctuations, and is bound to spontaneously decay by thermal
nucleating QBHs.
We define the functional integral for the time-evolution of a canonical ensem-
ble as a time-ordering product of canonical partition functions (8) for canonical
ensembles at different temperature T(Ti) = (3-1(Ti),
N
Z = TIl Z[{3(Ti)J := Z[{3(TO)]' Z[{3(TdJ' Z[{3(T2)J··· Z[{3(TN)J. (10)
i=O
 1781

Then, the functional integral (10) can be written as,

II II J'D[gab(X, T')]e-NSdgSl-SE(gl+gauge-fixing,
N (3( r;)
Z = (11)
i=O r'=O

where the Euclidean action SE(g) and finite instanton action NSdg S ) are respec-
tively given by the sum over their counterparts at different temperature T(Ti). The
finite N-instanton action NSE(gS) in (11) becomes,

NSE(gs) = ~;3(Ti)V(Ti) [~Ti~M(T;)f(Ti)] . (12)

The factor 2.:~0 ;3h)Vh) gives the four-dimensional volume in the function inte-
gral. Because the mass parameter A1(Ti) and decay rate f(Ti) in Eq. (12) are time-
dependent, we should integrate M(Ti)f(Ti)/2 over the time-interval ~Ti = Ti - TO
when thermal nucleating process occurs,

(13)

As results, the finite N-instanton action NSE(gS) (12) is given by

)1
PA(TO, T ==
r

ro
11 I
dT -M(T )f(T ).
2
I
(14)

Actually, PA (TO, T) is the energy-density of N QBHs nucleated in the Euclidean time

interval ~ = T - TO. After the Wick rotation from the Euclidean spacetime back to
Minkowski one, the finite action (14) gives the cosmological term,

PA(tO, t) == it
to
dt' ~M(t')r(t').
2
(15)

The cosmological constant is then given by,

(16)
Such cosmological term has two features: (i) its geometric origin from gravitational
instantons; (ii) its energetic origin from the energy-gain of nucleating QBHs at finite
temperature due to thermal energy of relativistic particles. The above formulations
and discussions can be straightforwardly extended to the case that matter fields are
present. The details of this study can be found in Ref.4

References
1. G.W. Gibbons and M.J. Perry, Proc. R. Soc. London A358 (1978) 467.
2. D.J. Gross, M.J. Perry and L.G. Yaffe, Phys. Rev. D. 25 (1982) 330.
3. G. Preparata, S. Rovelli, and S.-S. Xue, Phys. Lett., B427 (1998) 254 and General
Relativity and Gravitation, Vol 32, No.9, (2000).
4. S.-S. Xue, Journal of the Korean Physical Society, Vol. 49, NO.2, August 2006, pp. 759
and hep-th/0608220.
 AN ALTERNATIVE SOURCE FOR DARK ENERGY

M. 1. WANAS
Astronomy Department, Faculty of Science, Cairo University, Giza, Egypt
[email protected]

In the present work, an alternative interpretation of the source of accelerated expansion of

the Universe is suggested. A probable candidate is the interaction between the quantum
spin of a moving particle and the torsion of space-time, produced by the background
gravitational field of the Universe. This interaction has been suggested by the author in
a previous work, with some experimental and observational evidences for its existence.
It has been shown that this interaction gives rise to a repulsive force. The accelerated
expansion of the Universe may give a further evidence on the existence of this interaction
on the cosmological scale.
Recently, observation of red-shift of type I a supernova [1] indicate that our Uni-
verse is in an accelerating expansion phase. The four known fundamental interac-
tions, as we understand them today, can not account for the accelerating expansion
of the universe. Weak and strong interactions can be easily ruled out, since both
are very short range and cannot play any role in the large scale behavior of our
Universe. Also, electromagnetism cannot play any role, in this respect, since nei-
ther the universe nor its constituents are electrically charged, while monopoles are
not present in our Universe. The only interaction which is playing the main role,
in the structure and evolution of the universe, is gravity. But gravity, as far as we
know, is attractive and not repulsive. Consequently, it can not give any satisfactory
interpretation of the accelerating phase of the Universe. So, what is real cause of
the accelerating expansion of the universe? Is the existing physics sufficient to deal
with this problem? Do we need new physics?
Theories of gravity consider the interaction between matter (energy) and the
gravitational field. The General Theory of Relativity (GR), the most successful
gravity theory so far, considers gravity as a space-time property. Its field equations
shows how matter produces gravity. i.e. how matter affects space-time structure.
Its equation of motion shows how space-time structure (gravity) affects trajectories
of test particles (matter). In the context of GR, this interaction is between space-
time structure (curvature) and one of properties of matter (mass or energy). Mass
(energy) of a material object is collection of the masses (energies) of its elementary
constituents. These constituents have other properties beside mass, e.g. spin, charge,
... etc. GR does not answer questions neither about the interaction between space-
time structure and such other properties of matter, nor whether such interactions
exists or not.
If we consider geometries with curvatures and torsion, in particular the Pa-
rameterized Absolute Parallelism(PAP)-geometry [2], it has been shown that the

1782
 1783

parameterized path equation of this geometry can be written as [3].

2 Ct Ct
d xf1 +{ JL } dx dx{3 = -bA". f1 dx dx{3 (1)
ds 2 a(3 ds ds Ct{3. ds ds

where {:{3} is the Christoffel symbol, Af1Ct{3 is the torsion of the space-time and b
is a dimensionless parameter. The term on the R.H.S. of this equation has been
suggested to represent an interaction between the spin of the moving particle and
the torsion of space-time. In the presence of this interaction, the total potential felt
by a spinning particle, in the linearization regime of (1), is found to be [3]

<I>s = (1 - b)<I>N (2)

where <I> N is the Newtonian potential. The second term with the (- )sign on the
R.H.S. of (2)gives rise to a repulsive force. This force is needed to interpret the
results of the observation of SN type Ia [1].
For some experimental and observational reasons, the parameter b is suggested
to the form

b= -a,
n
2
(3)

where n is a natural number is, a is the fine structure constant and, is a dimen-
sionless parameter characterizes the source of the background field. Equation (2)
has been used to interpret the discrepancy of in the results of the COW-experiment
[4] considering the gravitational field of the Earth as weak, static field and taking
n = 1 (for neutrons)and , = 1 for the Earth. The same equation has been used
to construct a temporal model of SN1987A [5]. Also, it has been used to study the
effect of spin-torsion interaction on the Chandrasekhar-limit [6].
Many authors have tried to interpret the accelerating expansion of the universe
using different assumptions. Most of them consider the cosmological term in the
equations of General Relativity, as a probable candidate (for a review of alternative
interpretations see [7]). Others consider special types of matter to be the source of
dark energy [8]!
In conclusion we draw the following comments:
1. Torsion of space-time should be taken into account in order to complete our
understanding of physical phenomena, especially on large scale.
2. Spin-torsion interaction, giving rise to a repulsive force, can be used to interpret
the accelerating expansion of the Universe, and may prevent the existence of space-
time singularities.
3. Torsion may give a reasonable geometric origin for the cosmological term.
4. A field theory, taking into account the torsion of space-time is needed now in
order to find the type of dependence of the parameter , on the energetic contents
of any system under consideration.
An extended version of the present work will be published elsewhere.
1784

Acknowledgement
The author would like to thank the authorities of the lCTP Trieste, Italy for giving
him a grant to participate in this conference.

References
[1] Tonry, J.L., Schmdit, B.P. et al. (2003) Astrophys. J. 594, 1.
[2] Wanas, M.L (2000) Turk. J. Phys. 24, 473; gr-qc/0010099
[3] Wanas, M.L (1998) Astrophys. Space Sci., 258, 237; gr-qc/9904019
[4] Wanas, M.L, Melek, M. and Kahil, M.E. (2000) Gravit. Cosmol.,6, 319.
[5] Wan as , M.I., Melek, M. and Kahil, M.E. (2002) Proc. MG IX Part B, p.ll00 ;
gr-qc/0306086
[6] Wanas, M.l. (2003)Gravit. Cosmol.,9, 109.
[7] Mannheim, P.D. (2006) Prog.Part.Nucl.Phys. 56 340; gr-qc/0505266
[8] Babourova, O.V. and Frolov, B. (2003) Class.Quant.Grav. 201423 ;
gr-qc/0209077
 AN AWESOME HYPOTHESIS FOR DARK ENERGY: THE
ABNORMALLY WEIGHTING ENERGY

ANDRE FUZFA 1 ,2*

JEAN-MICHEL ALIMI2
1 GAMASCO, Facultes Universitaires Notre-Dame de la Paix, Namur, Belgium
2 Laboratory Universe and Theories, CNRS UMR 8102,
Observatoire de Paris-Meudon and Universite Paris VII, France

We introduce the Abnormally Weighting Energy (AWE) hypothesis in which dark en-
ergy (DE) is presented as a consequence of the violation of the weak equivalence principle
(WEP) at cosmological scales by SOme dark sector. Indeed, this implies a violation of
the strong equivalence principle (SEP) for ordinary matter and consequent cosmic ac-
celeration in the observable frame as well as variation of the gravitational constant. The
consequent DE mechanism build upon the AWE hypothesis (i) does not require a viola-
tion of the strong energy condition p < -pc 2 /3, (ii) assumes rather non-negligible direct
couplings to the gravitational scalar field (iii) offers a natural convergence mechanism
toward general relativity (iv) accounts fairly for supernovae data from various couplings
and equations of state of the dark sector as well as density parameters very close to the
ones of the concordance model AC D 1\;1. Finally (v), this AWE mechanism typically ends
up with an Einstein-de Sitter expansion regime once the attractor is reached.

The AWE hypothesis can be seen as a generalization of many models aimed to

address the DE problem. Let us begin by the heart of the concordance model: the
controversial cosmological constant. The theory of gravitation behind the concor-
dance model is the usual Einstein-Hilbert action of general relativity (GR):

SEH = _1_
2",,*
JV-g.d4x {R* + 2A} + Sm [1/Jm' g~vl ' (1)

where g~v is the Einstein metric, A the cosmological constant, ""* = 87rG* is the
gravitational coupling constant, R* is the curvature scalar and 1/Jm are the matter
fields. The cosmological constant is well-known to lead to intricate fine-tuning and
coincidence problems whose possible way-out is to assume a cosmological mecha-
nism for DE which would rule the non-trivial gravitational vacuum. Quintessence
is problably the most popular of these mechanisms, and is based on a real massive
scalar field rolling down some self-interaction potential. During a slow-roll phase of
this massive scalar field (when its kinetic energy is much less than its self-interacting
one), negative pressures are achieved, producing the desired cosmic acceleration if
they violate the strong energy condition (SEC): p < -p/3. More precisely, the
quintessence mechanism is implemented by the following theory:

SQuint = 2~* J 4
V-g*d x {R* - 2g~Val"rpavrp + V(rp)} + Sm [1/Jm,g~vl ' (2)

with rp the quintessence scalar field and V (rp) its self-interaction potential. The
quintessence mechanism relies strongly on an appropriate choice of this potential

*F.N.R.S. Postdoctoral Researcher

1785
1786

whose shapes can be quite sophisticated in order to reproduce Hubble diagram

data. The cosmological constant A in (1) corresponds in (2) to a quintessence scalar
field frozen (O"i.PCY'i.P : : : : 0) in a non-vanishing energy state V(i.p) > O. However,
quintessence is a very discrete form of DE whose interactions with matter are purely
gravitational. This can be extended to non-minimal couplings in order to make
this field running physical constants and account for a violation of the equivalence
principle (EP). This constitutes the famous tensor-scalar theories of gravitation

STS = 2~* J 4
y' -g*d x {R* - 2g~V O"i.pOvi.p + V( i.p)} + Sm ['ljIm, A;" (i.p )g~v l, (3)

where Am (i.p) denotes the coupling function to the Einstein metric (quintessence
corresponds to Am (i.p) = 1). However, if we consider that the 70% of missing en-
ergy is the whole contribution of such a non-minimally coupled scalar field, then
it might be difficult to match the present tests of GR without assuming the non-
minimal couplings to be extremely weak. Even worst, as soon as these couplings are
non-vanishing, the likely endless domination of a non-minimally coupled scalar field
with quintessence potential might also inescapably lead to a disastrous violation of
the EP in the future.

The AWE Hypothesis 1 consists of assuming that DE does not couple to gravitation
in the same way as usual matter (baryons, photons, ... ). This implies a violation
of the weak equivalence principle (WEP), mostly on cosmological scales, where DE
dominates the energy content of the Universe. This anomalous weight also implies
that the related gravitational binding energy of DE does not generate the same
amount of gravity than the one of usual matter, yielding a violation of the strong
equivalence principle (SEP). This results in a running gravitational coupling con-
stant on cosmological scales whose dynamics will produce cosmic acceleration. The
corresponding action can be written down:

S = 2~* J 4
y'-g*d x {R* - 2g~Vo"i.pOvi.p} + Sm ['ljIm, A;"(i.p)g~vl
(4)
with Sawe is the action for the AWE sector with fields 'ljIawe and Sm is the usual mat-
ter sector with matter fields 'ljIm; Aawe (i.p) and Am (i.p) being the coupling functions
to the metric g;v for the AWE and matter sectors respectively. Usual tensor-scalar
theories (3) correspond to a universal coupling Aawe(i.p) = Am (i.p). However, the
action (4) is written in the so-called "Einstein frame" of the physical degrees of
freedom while cosmology is expressed in the observable frame 9"v = A;" (i.p)g;v.
This subtelty will allow to produce cosmic acceleration without violation of SEC
through the acceleration of i.p in a = Am(i.p)a*. The AWE mechanism is based on
a competition between the non-minimal couplings of the two material sectors to
rule the EP. This competition is similar to the dynamics of the so-called chameleon
fields 3 between the self-interaction term and the non-minimal coupling. The compe-
tition ends up with the stabilization of i.p on a attractor, leaving the corresponding
 1787

gravitational theory asymptotically similar to GR. During some time before the
stabilization, both transient cosmic acceleration and variation of G appear allowing
to explain Hubble diagram of type Ia supernovae. This mechanism has been imple-
mented in the case of an AWE Born-Infeld gauge interaction 1 but also works with
a pressureless fluid. 2 Figure 1 illustrates the behavior of the scale factor in different
AWE cosmologies fitting the Hubble diagram of type Ia supernovae. The AWE hy-
pothesis exhibits key features for DE (see the abstract) and we are confident that
it will surely constitute a seducing alternative to the cosmological constant.

0.9

0.8

0.7

~ 0.6
<1l
~ 0.5
(ij
c7l 0.4

0.3
- AWE Born-Infeld Interaction (without G(z»
- _. AWE Born-Infeld Interaction (with G(z»
0.1 AWE Dark Matter
ACDM
2 4 6 8 10 12 14
Time (Gyr)

Fig. 1. Evolutions of the scale factor in some AWE models and comparison with the concordance
model ACDM (straight line is the AWE Born-Infeld (BI) gauge interaction 1 without a variation
of G ; the dashed one is the same model accounting for this variation ((Dm,o, DBI,O) ::::; (0.3,0.7»;
the bold dotted line is an AWE Dark Matter (DM) model 2 for which (Dm,o, DDM,O) ::::; (0.04,0.26)
and dotted line is a (Dm,o,DA,O)::::; (0.3,0.7) ACDM model)

References
1. A. Fiizfa, J.-M. Alimi, Phys. Rev. Lett. 97, 061301 (2006);
A. Fiizfa, J.-M. Alimi, Phys.Rev. D73 (2006) 023520 A. Fiizfa, J.-M. Alimi, Proceed-
ings SF2A 2006, astro-ph/0609099
2. A. Fiizfa, J.-M. Alimi, "Toward a Unified Description of Dark Energy and Dark Matter
from the A WE Hypothesis", in preparation.
3. J. Khoury, A. Weltman, Phys. Rev. Lett. 93, 172204 (2004);
J. Khoury, A. Weltman, Phys. Rev. D69, 044026 (2004);
P. Brax, C. van de Bruck, A.C. Davis, J. Khoury, A. Weltman, Phys. Rev. D 70,
123518 (2004)
 PERTURBATIONS OF A COSMOLOGICAL CONSTANT
DOMINATED UNIVERSE

MATyAs VASUTH* and VIKTOR CZINNER*t

* KFKI Research Institute for Particle and Nuclear Physics
Budapest 114, P.D.Box 49, H-1525 Hungary
E-mail: [email protected]
t Yukawa Institute for Theoretical Physics, Kyoto University, Kyoto 606-8502, Japan
E-mail: [email protected]

One of the proposed form of dark energy is the cosmological constant. We give the
complete Coo solution governing linear perturbations of a spatially flat Friedmann-
Robertson-Walker model with a pressureless ideal fluid source and non-vanishing cosmo-
logical constant. Using the potential which belongs to the growing mode of the relative
energy density fluctuations, we separate the Sachs-Wolfe and Integrated Sachs-Wolfe
terms. It is found that the part of the Sachs-Wolfe term which becomes time dependent
in the presence of a cosmological constant underwent a reversal approximately two billion
years ago. Using the time dependence of the perturbation potentials we give the analytic
expression for the argument of the ISW integral for an arbitrary initial power spectrum
of the fluctuations.

1. Introduction
The accelerating expansion of the universe has been confirmed by recent cosmo-
logical measurements, i.e. the Wilkinson Microvawe Anisotropy Probe,l type Ia
supernovae observations and other sky surveys.2,3 This kinematic property of the
universe can be represented by the introduction of dark energy. At present, all exist-
ing observational data are in agreement with the concordance model. In this model
the universe is flat, homogeneous and isotropic with nonzero cosmological constant
A which is the simplest possibility of dark energy.
There are several effects which contribute to the fluctuations of the cosmic mi-
crowave background radiation (CMBR). The dominant part of primary anisotropies
comes from the Sachs-Wolfe effect.4 In a cosmological constant (or curvature) dom-
inated universe the gravitational potentials are time varying which give rise to the
appearance of the integrated Sachs-Wolfe (ISW) effect. In a linear theory for a flat,
[20 = 1 universe the potentials are constants and this appearance does not occur.
Here we consider an analytic framework for the description of the SW and ISW
effects and their contributions to the fluctuations of the CMBR in the presence
of a cosmological constant. Following the treatment of the perturbations due to
White 5 all Coo solutions of the linear field equations in a dust filled universe can
be obtained. 6 The metric is a perturbed FRW model in the conformal form, gab =
a 2 (TJ)(1Jab + h ab ), with the Minkowski metric 1Jab, the metric perturbation hab and
the scale factor a = a(1J). The metric and density perturbations for a flat universe
and pressure free source are

3B,a{3
h a{3 = 1Ja{3B - 4/3 2C2 coth(Ct)I(t) ,
2 aD

1788
 1789

where B = B(x a ) is a space function, C and eM are constants and the time-
dependent amplitude I(t) is an elliptic integra1. 6 In Eq. (1) we have presented only
those terms that are proportional to the relatively growing mode of the density
contrast.

2. The integrated Sachs-Wolfe effect

The contribution to the temperature fluctuations of the CMBR due to the gravita-
tional perturbations along the path of the photon is the Sachs-Wolfe effect which
can be computed as 4

JT = ~ (lR-TiE ( 8h a(3 eae(3 _ 2 8h oa ea) dw .

(2)
T 2k ~ ~ .
where 7]R and 7]E are the time of reception and emission, respectively, and w is the
affine length along the null geodesic. The different contributions sum up to
JT [ 3 ] TiR
T = - 7/3 2C2 B,a ea !::"(t) - B!::"sw(t) - 21/3 ao C
lTiR
B!::"ISW(t)dw . (3)
2 ao TiE TiE
4
The B.aea term has a dipole character. The amplitude !::"sw decays exponentially.
The terms in the square brackets represent the SW effect, while the last term in
Eq. (3) is the ISW term with the transfer function
A 2cosh2(Ct) + 3 () 21/ 3
L.l.ISW = It - , (4)
3 sinh2 (Ct) sinh1/3( Ct) cosh( Ct)
whose contribution becomes dominant only in late times. 6
The complete analytic solution of the perturbed field equations can also be
obtained by using the covariant linear perturbation formalism of Bardeen. 7,8 With
the use of the gauge invariant quantities the CR momenta of the anisotropy power
spectrum can be easily obtained. The general perturbations of a spatially flat FRW
universe is written as
goo = -(1 + 2A), gOa = -aBa, ga(3 = a 2 [(1 + 2HL ) 'Ya(3 + 2HTa (3] (5)
with the perturbation quantities A, H L ) Boo and H TCi (3. These functions can be rep-
resented by the amplitudes of the scalar, vector and tensor eigenmodes of the Lapla-
cian operator and the Einstein equations decouple into a set of ordinary differential
equations. For the spatially flat model these eigenmodes are simply plane waves. In
the covariant formalism three gauge invariant quantities can be constructed, \jJ and
<I> from the metric perturbations and V from the matter velocity.7 Using the New-
tonian gauge condition we have the following expressions for the relatively growing
mode 9
1 cosh(Ct)
\jJ = - 4/3 5/3 I(t)Bk, <I> = -\jJ ) (6)
2 sinh (Ct)
13
V - ~ [ I(t) _ sinh / (Ct)] Bk (7)
- 4Cao 21/ 3 sinh / 3(Ct)
4 cosh(Ct) )
1790

where Bk is the amplitude of the eigenmode of the Laplacian operator corresponding

to the kth wave number for the function B.
With the solution W the anisotropy power spectrum of the ISW effect can be
calculated. The early ISW effect describes the variation in the perturbation po-
tentials occurring in the time interval between the matter-radiation equality and
recombination. The late ISW effect, corresponding to the variations caused by the
presence of A, can be approximated by the following formula lO

CISW ~ 2V (r[(€+ 1)/2])2 roo Wf2('11·,k, k)dk , (8)

R r[(€ + 2)/2] io
where the prime denotes derivative with respect to the conformal time 71. Inserting
the solution for W into Eq. (8) the analytic form of the argument of the integral is

Wf2 = 25r2/3C2a2 2
1/3 __1_ (2 x 3 + 5) I(x)
]2 IW(O, kW , (9)
9 0 [JX(l + x3) 3x3
where W(O, k) = -3Bk/10 and x = sinh 2 / 3 (Ct).
To calculate the CfsW momenta a general power law initial spectrum is chosen
P(k) = k3 Iw(O, kW = Bkn-l, where B is constant. It is usual to set n = 1 for the
scale invariant Harrison-Zeldovich spectrum. Although the argument of the integral
in Eq. (8) have been expressed analytically, the integration over k has to be done
numerically. After performing the calculations the obtained power spectrum of the
late ISW effect is in very good agreement with earlier numerical calculations. 9

3. Concluding remarks
We have calculated the anisotropy power spectrum of the late ISW effect in a
cosmological constant dominated universe with the use of the analytic expression
of the argument of the ISW integral. The results are in agreement with earlier
calculations in the case of the scale invariant Harrison-Zeldovich spectrum. Our goal
is that in the case of an arbitrary initial power spectrum the anisotropy spectra can
be more easily calculated with the help of the obtained analytic formulae.
This work was supported by OTKA grants no. TS044665, T046939 and F049429.

References
1. D. N. Spergel et al., Astrophys J. Suppl. 148, 175 (2003).
2. J. L. Tonryet al., Astrophys J. Suppl. 594, 1 (2003).
3. M. Tegmark et al., Phys. Rev. D69, 103501 (2004).
4. R. K. Sachs and A. M. Wolfe, Astrophys. J. 147, 73 (1967).
5. P. C. White, J. Math. Phys. 14, 831 (1973).
6. Z. Perjes, M. Vasuth, V. Czinner, and D. Eriksson, Astron. Astrophys. 431,415 (2005).
7. J. M. Bardeen, Phys. Rev. D22, 1882 (1980).
8. V. Czinner, M. Vasuth, A. Lukacs, and Z. Perjes, Int. J. Mod. Phys. A20, 5671 (2005).
9. V. Czinner, M. Vasuth, and A. Lukacs, Int. J. Mod. Phys. A20, 7233 (2005).
10. W. Hu, Ph.D. Thesis (1995).
 ON THE GURZADYAN-XUE COSMOLOGICAL MODELS AND
THE DYNAMICS OF DENSITY PERTURBATIONS

G. YEGORIAN
ICRANet, P.le della Repubblica 10, 165100 Pescara, Italy and
ICRA, Dip. Fisica, Univ. La Sapienza, P.le A. Mora 5, IO0185 Rome, Italy
and
Yerevan Physics Institute, 375036 Yerevan, Armenia

The Gurzadyan-Xue formula for dark energy predicts the value of the density parameter
in agreement with observations. We explore cosmological consequences of this formula,
discussing solutions of cosmological equations and behavior of density perturbations.

Present dark energy density can be attributed to vacuum fluctuations, and after
taking only relevant modes, as was shown by Gurzadyan and Xue,l dark energy
density is given by

'if nc 1c4 1 'if

(1)
pex = "8 L~ a 2 = "8 G a 2 '

where II, is the Planck constant, the Planck length is Lp = (nG)! C 3 / 2 and c is the
speed oflight, G is the gravitational constant and a is the scale factor of the U nivese.
Following the original idea by Zeldovich,3 the vacuum energy (1) corresponds to the
cosmological term Aex = 87r~~GX = 'if2 (~) 2.
This formula does not contain free parameters, and being written for the density
parameter gives the impressive result, nex = 3.29, which is very close to the
value inferred from observations, in contrast with the standard approach leading
to a difference of 123 orders of magnitude. This small difference can be further
eliminated 2 by introducing a phenomenological parameter.
In General Relativity all the quantities on both sides of (1) are constants, except
for the scale factor, which is a function of the cosmic time. Therefore one naturally
comes to the possibility of varying physical constants. 4
Assuming the validity of the Einstein equations also when physical constants
vary in time, the cosmological equations 5 read

(2)

(3)

where p is energy density, p is the pressure, k is the sign of spatial curvature, H is

the Hubble parameter; the dot denotes time derivative. From these two equation (or
from energy-momentum conservation (8~PTflV);v = 0) follows continuity equation

p+3H(p+p) = -PA + (p+ PA) ( 4~c - 6)

G . (4)

1791
1792

In a multi-component cosmological model, based on Eqs. (2)-(4) with energy

densities Pi of each component particular solution can be characterized completely
by the present density parameter which has the same form as in the Friedmann
model
87fGo
Di = 3H2 2 Pia,
oC o
where the index "0" refers to the values today.
'We found that in the phase space of solutions for simple models with a pres-
sureless fluid there is a separatrix which divides the space of solutions into two
clases: Friedmannian-like solutions with initial singularity and non-Friedmann-like
solutions which begin with nonzero scale factor and vanishing matter density.6 In
spite of the fact that the equations for our models look very different, the value of
separatrix for all models is the same and depends only on k (the sign of the spatial
curvature)

When k = 0, we have a simple result 12m = ~. The presence of the same

separatrix in all models may be a result of some symmetry in the GX models. 7
The simple models with pressureless fluid are generalized to include radiation
field,5 assuming that the dynamics of radiation is not influenced by variation of
constants, i.e., taking the anzatz for radiation density and pressure Pr = PrO (..£.)
ao
-4,

p = '?f.In this multi-component model the separatrix is absent. Here all solutions
(except for the case where explicit coupling of matter and radiation due to variation
of constants is absent) belong to the non-Friedmann-like class, i.e., solutions start
with vanishing density and nonzero scale factor in a radiation-dominated epoch.
We also considered density perturbations in GX models in the Newtonian ap-
proximation.
Density perturbations obey the following equationS

6+J(2H+~) - ~~V'20-47fGp6=0,
where 0 is dimensionless density contrast, fl is the unperturbed density, and v;
= g~.
Here again, in contrast with the corresponding equation in Friedmann models, there
is additional term due to the speed of light variation. It is interesting that in the
absence of cosmic expansion this term is already sufficient to change the character
of instability from an exponential one, as is the case in the original Jeans treatment,
to a polynomial one.
The Jeans length can be introduced in a standard way by using the Fourier
decomposition, A] = Vs /iii, but since both the speed of sound and background
density behavior is different from the Friedmann case, we expect different charac-
teristic scales of structure formation in GX models.
 1793

In contrast with the Friedmann case, in GX models fluctuations continue to

grow after the matter-dominated epoch. As for the matter-dominated stage, in
some GX models perturbations grow exactly as in the Friedmann models, and in
some models they grow slowly, but still with polynomial dependence on the scale
factor (see details in Ref. 8).

References
1. V.G. Gurzadyan, S.-S. Xue in: "From Integrable Models to Gauge Theories; volume
in honor of Sergei Matinyan", ed. V.G. Gurzadyan, A.G. Sedrakian, p.177, World Sci-
entific, 2002; Mod. Phys. Lett. A18 (2003) 561 [astro-ph/0l05245]' see also [astro-
ph /0510459].
2. S.G. Djorgovski, V.G. Gurzadyan, Nue!. Phys. B, in press [astro-ph/0610204].
3. Va. B. Zeldovieh JETP Lett. 6 (1967) 883; Sov. Phys. - Uspekhi 95 (1968) 209.
4. G.V. Vereshehagin, Mod. Phys. Lett. A21 (2006) 729.
5. G.V. Vereshehagin and G. Yegorian, Class. Quantum Grav. 23 (2006) 5049.
6. G.V. Vereshehagin and G. Yegorian, Phys. Lett. B636 (2006) 150; see also this volume.
7. H.G. Khaehatryan, Mod. Phys. Lett. A22, No.4 (2007).
8. G.V. Vereshehagin and G. Yegorian, astro-ph/0610l97.
 SCALAR-TENSOR DARK ENERGY MODELS

R. GANNOUJI, D. POLARSKI, A. RANQUET

Lab. de Physique Theorique et Astroparticules, CNRS
Universite Montpellier II, 34095 Montpellier Cedex 05, Prance

A. A. STAROBINSKY
Landau Institute for Theoretical Physics, Moscow, 119334, Russia

We present here some recent results concerning scalar-tensor Dark Energy models. These
models are very interesting in many respects: they allow for a consistent phantom phase,
the growth of matter perturbations is modified. Using a systematic expansion of the
theory at low redshifts, we relate the possibility to have phantom like DE to solar system
constraints.

1. Introduction
The late-time accelerated expansion of the universe is a major challenge for cos-
mology. The component producing this acceleration accounts for about two thirds
of the total energy density. While this has gradually become a building block of
our present understanding, the nature of Dark Energy (DE) still remains myste-
rious. 1- 3 The simplest solution is a cosmological constant A. A major contender
is Quintessence, a minimally coupled scalar field (with canonical kinetic term). We
will consider scalar-tensor (ST) DE models, a more elaborate alternative involving a
new physical degree of freedom, the scalar partner ¢ of the graviton responsible for
a modification of gravity.4-6 It is not clear yet whether some modification of gravity
is required or even preferred in order to explain the bulk of data. The increasing
accuracy of the data, should allow to severely constrain the various viable models.
ST DE models allow for phantom DE, WDE < -1, moreover the equation for the
growth of matter perturbations is modified. 4 We will review here results concerning
their low z behaviour, in particular how the DE equation of state is related to solar
system constraints. 9

2. Scalar-tensor DE models
We consider the microscopic Lagrangian density in the Jordan frame

L= ~ (F(P) R - Z(p) g ILV OM POv P ) - U(p) + Lm(gMV) . (1)

We define what we mean by the energy density PDE and the pressure PDE by writing
the gravitational equations in the following Einsteinian form:
3Fo H2 = Pm + PDE (2)
- 2Fo if = Pm + PDE + PDE (3)
1
This can be seen as the Einsteinian form, with constant Go = G N (to) = F O- , of the
gravitational equations of ST gravity. With these definitions, the usual conservation

1794
 1795

equation applies, and the equation of state parameter WDE == PDE

PDE
plays its usual
role. Using (2,3), One gets WDE(Z) from the observations through
h 2 + 1D (1 + Z)2
2
Hz dh _
( ) 3 dz 3 k,O
(4)
WDEZ = h2-Dm,0 (1+z)3-Dk,O (l+z)2 ,
if we allow for a nonzero spatial curvature and Dm == 2Fo' 3#
Looking at the equations above, everything looks the same as in GR, ST gravity
is hidden in the definitions of PDE, PDE, and the various D's. The condition for DE
to be of the phantom type, WDE < -1, reads
dh 2
dz < 3 Dm,o (1 + z? + 2 Dk,o (1 + z) . (5)
in the presence of spatial curvature.1,4,7 As first emphasized,4 the weak energy
condition for DE can be violated in scalar-tensor gravity (see also S ).

3. General low z expansion of the theory

We investigate now the low z behaviour of the model and the possibility to have
phantom boundary crossing in a recent epoch. For each solution H(z), <I>(z), the
basic microscopic functions F (<I» and U ( <I» can be expressed as functions of z and
expanded into Taylor series in z:
F(z) 2
- - = 1 + Fl z + F2 z + ... > 0 , (6)
Fo
U(z)
3F H'6 == Du,ou = Du,o + Ul Z + U2 z2 + .... (7)
o
From (6,7), all other expansions can be derived, in particular:
WDE(Z) =WO+Wl Z+W2 z2+ ... , (8)
G
H o-1 -0
elf
= go + gl Z + g2 Z2 +..... ( )
9
elf
A viable ST gravity model must be very close to General Relativity, viz.
_ 6(DDE,0 - Du,o - Fd _ b,2 4 10 4 (10)
WBD,O - F2 - F2 > X ,
1 1

with b,2 == 6 (DDE,O - Du,o - Fd. Therefore, we must have IFII « 1 and b,2 ~
6(DDE,0 - Du,o) > O. Moreover, for positive U, b,2 < 6DDE,0 < 5

IFII < (_5_)

WBD,O
1/2 ;s 10- 2 . (11)

It can be shown that the condition IFII « 1 is sufficient to ensure here that solar
system constraints are satified. 9
We now specialize to the case IFII « 1 yet assuming that other Fi are not as
small. Then all expansions simplify considerably and we have in particular,
2F2 + 6(DDE 0 - Du 0) (12)
1 +wo c::: "
3DDE,0
1796

From (12), the necessary condition to have phantom DE today reads

(dd<l>2F)
2
(13)
0

Hence F2 < 0 is necessary for phantom DE, because DDE,O -Du,o > 0 from,t,2 > O.
In addition significant phantom DE requires IF21 rv 1. If IFII rv 1F21 « 1, the present
phantomness is very small.
It is actually possible to invert all expansions and to obtain all coefficients in
function of the post-Newtonian parameters ,,(, f3 and go. The following results are
finally obtained
,,(-1
FI = go ----'-----"7' (14)
"( - 1 - 4(p - 1)
2 f3 - 1
(15)
F2 = -2 go h- 1 - 4(f3 - 1)]2
1 2 "( -1
DDE,O - Du,o = -'6 go h- 1 - 4(p - 1)]2 (16)

1 2 4(f3-1)+,,(-1
(17)
1 + WDE,O = -'3 go DDE,O h - 1 - 4(f3 - 1)]2
The best present bounds are "(PN - 1 = (2.1 ± 2.3) . 10- 5 , f3PN - 1 (0 ± 1) .
10- 4 , cGeff .o = (-0.2±0.5)·10- I3 y-I. Though possible in principle, 10 the interesting
eff ,0
possibility to test phantomness in the solar system is very hard while its amount
depends critically on the small quantity g6.
In this respect cosmological data are
certainly better suited, a conclusion reminiscent of that reached in6 concerning the
viability of ST DE models with vanishing potential.

References
1. V. Sahni and A. A. Starobinsky, Int. J. Mod. Phys. D 9, 373 (2000).
2. T. Padmanabhan, Phys. Rep.380, 235 (2003).
3. V. Sahni, astro-ph/0502032 (2005); E. J. Copeland, M. Sami and S. Tsujikawa, Int.
J . Mod. Phys. D 15, 1753 (2006).
4. B. Boisseau, G. Esposito-Farese, D. Polarski and A. A. Starobinsky, Phys. Rev. Lett.
85, 2236 (2000).
5. Y. Fujii, Phys. Rev. D62, 044011 (2000); N. Bartolo and M. Pietroni, Phys. Rev. D
61 023518 (2000); F. Perrotta, C. Baccigalupi and S. Matarrese, Phys. Rev. D 61,
023507 (2000).
6. G. Esposito-Farese and D. Polarski, Phys. Rev. D 63, 063504 (2001).
7. D. Polarski and A. Ranquet, Phys. Lett. B 627, 1 (2005).
8. D. Torres, Phys. Rev. D 66, 043522 (2002).
9. R. Gannouji, D. Polarski, A. Ranquet, A. A. Starobinsky, JeAP 0609, 016 (2006).
10. J. Martin, C. Schimd and J.-P. Uzan, Phys. Rev. Lett. 96, 061303 (2006).
 RECONSTRUCTION OF DARK ENERGY USING SUPERNOVA
AND OTHER DATASETS

UJJAINI ALAM1, VARUN SAHNe and ALEXEI A. STAROBINSKy3

1 International Centre for Theoretical Physics, Strada Costiera 11, 34100 Trieste, Italy
2 Inter- University Centre for Astronomy and Astrophysics, Pune 411 007, India
3 Landau Institute for Theoretical Physics, 119334 Moscow, Russia

We explore different methods of extracting information about dark energy from the
presently available data. Supernova data is of primary importance to the study of dark
energy, and over the last decade, many different reconstruction methods have been ap-
plied to this data. In this talk, we summarize some methods of model independent re-
construction of dark energy from supernova data. We also look at the sensitivity of other
complementary probes, such as the cosmic microwave background (CMB) and baryon
acoustic oscillations (BAO) in large scale structure, to the nature of dark energy.

Dark Energy is one of the most tantalizing problems of current cosmology. A host
of cosmological observations- supernova Type la, CMB and large scale structure-
suggest that about two-third of the energy density of the present universe is made
up of this unknown, negative-pressure component. The simplest candidate for DE
is the cosmological constant A. Several other phenomenological explanations have
also been suggested, such as quintessence and phantom models, the Chaplygin gas,
Braneworld models, scalar-tensor theories etc. 1
It is not efficacious to examine this multitude of DE models individually against
the data. We employ instead a quasi-model-independent approach where a suffi-
ciently general ansatz for dark energy is tested against the data, and confidence
levels on derived cosmological parameters are then used to rule out/accept different
cosmological models of DE. For this purpose we use an ansatz for the DE density
which has been found to give accurate results for a variety of cosmological models.
h 2(z) = H 2(z)/ Hg = Oom(1 + z)3 + Ao + Ad1 + z) + A2(1 + z)2 .
The strongest evidence in favour of the accelerating universe (and therefore DE)
comes from SNe Type Ia data. We examine two recent datasets- the Gold+HST
sample from the HZT team 2 and SuperNova Legacy Survey (SNLS) data 3 - to see
what constraints follow from them on the possible evolution of DE. We also look at
the constraints available from data complementary to SNe. We consider jointly the
parameter R characterizing acoustic peaks in the angular power spectrum of CMB 4
and the SDSS result on the baryon acoustic oscillation peak (BAO). 5
Figure 1 shows the reconstructed equation of state for all three datasets. 6 In these
results, we have marginalized over the matter density using OOm = 0.28 ± 0.03, the
currently accepted value from SDSS. 5 Two points are to be noted-

• Firstly, there is a slight discrepancy in the results from the two SNe
datasets, with Gold+HST preferring an evolving dark energy model with
w(z = 0) ~ - 1 and w(z = 1) ::: 0 while SNLS prefers ACDM. However,
this discrepancy is quite small (within 2a both datasets agree with ACDM)

1797
1798

20 0.2 04 06 08 1 12 14 16

Fig. 1. Best-fit and 2a confidence levels are shown for the reconstructed equation of state of dark
energy w(z) for the Gold+HST dataset (left panel), the SNLS dataset (middle panel) and the
BAO+CMB dataset (right panel) with the dashed line showing ACDM.

and can probably be attributed to the fact that the two datasets use dif-
ferent techniques for calibrating their SNe (MLCS2k2 for Gold+HST and
SALT for SNLS). With better data and uniform calibration, this discrep-
ancy should disappear.
• Secondly, unlike the SNe data, the results from BAO+CMB depend quite
strongly on the value of matter density. Choosing even a slightly different
value (say DOm = 0.25 ± 0.03, within the 10" limits of the currently accepted
value) would give quite different results for w(z). Thus, for this data, it is
important to know the value of Dom model-independently and accurately.

Given that the current data is not sensitive enough to obtain strong constraints
on the nature of dark energy, we next take a look at the prospects from future data.
A number of future SNe probes have been suggested- DUNE, SNAP, JED!. Using
these, we can resolve the issue whether DE is the cosmological constant, or if it is
evolving. We simulate data according to SNAP specifications using a ACDM fiducial
model. We use this dataset to reconstruct the Hubble parameter and equation of
state using a smoothing technique 7

1nddz, ~)S = 1n ddz)9+N(z) L, [lnddz;) -111 dd z i)9] exp [-ln2 (~: :i) / 2~2] ,

where N(z) is a normalization factor. The results are shown in fig 2. We see that
this technique can reconstruct cosmological parameters quite well with future data.
We also define new diagnostics for dark energy- the Statefinder pair {T, s}-
which could break the degeneracy between dark energy models,S
T ="(i /(aH 3 ), s = (T - 1)/[3(q - 1/2)] .
Figure 3 shows the reconstructed Statefinder pair for a simulated ACDM universe.
We see that the Statefinder pair may be used in future to discriminate the cosmo-
logical constant from other dark energy models.
In conclusion, we reconstruct dark energy parameters from current data and find
that ACDM is consistent with all observations, but evolving DE models are not yet
 1799

co
cO
co I
0 N
I
~ '5:
I
,-
co

co
0.5 1.5 ,- 0.5 1.5

Z Z

Fig. 2. The reconstructed Hubble parameter and equation of state of dark energy at 20" confidence
level for ACDM using the smoothing method. The dashed line shows ACDM.

-- .. 00

QUINTESSENCE

~ w=,_O.5\~.CI"'4
,.a",2
1'1=-0 ~:t~"

IC
O
~ K=l ',~
",=6.,
CHAPLYGIN GAS

Fig. 3. The statefinder pair averaged over redshift {T, s} with 30" confidence levels for a ACDM
universe simulated using SNAP specifications.

ruled out, and better data is required to reach a firm conclusion on the true nature
of dark energy. With an eye to future observations, we define new diagnostics for
dark energy such as the Statefinder pair, and new reconstruction techniques such
as smoothing of data.

References
1.V. Sahni and A. A. Starobinsky, 2000, Int. J. Mod. Phys. D 9, 373.
2.A G. Riess, et al., 2006, astro-ph/0611572.
3.P. Astier et al., 2005, Astron. Astroph. 447, 3l.
4.D. N. Spergel et ai., 2006, astro-ph/0603449.
5.D. J. Eisenstein, et ai., 2005, Astroph. J. 633, 560.
6.U. Alam, V. Sahni and A. A. Starobinsky, 2007, JCAP 0702, OIL
7.A. Shafieloo, U. Alam, V. Sahni and A. A. Starobinsky, 2006,
Mon. Not. Roy. Ast. Soc. 366, 108l.
8. V. Sahni, T. D. Saini, A. A. Starobinsky and U. Alam, 2003, JETP Lett. 77 201;
U. Alam, V. Sahni, T. D. Saini and A. A. Starobinsky, 2003, Mon. Not. Roy. Ast. Soc.
3441057.
 F(R) DARK ENERGY:
FROM THE TIME OF RECOMBINATION TILL PRESENT DAY

VIKTOR GUROVICH, VLADIMIR FOLOMEEV

Institute of Physics of NAS KR, 265 a, Chui str., Bishkek, 720071, Kyrgyz Republic
[email protected]

IYA TOKAREVA
Physics Department, Technion, Technion-city, Haifa 32000, Israel
[email protected]

The geometric model of dark energy within the framework of f(R) ex: Rn theory is
considered. The value of the parameter n corresponding in the best way to observations
from the epoch of recombination (z c::; 1000) till present day (z c::; 0) is found.

1. Introduction
The discovery of accelerated expansion of the Universe 1 has forced looking for mod-
els for explanation of this acceleration. In this Letter we consider the geometric
model of dark energy (DE) within the framework of modified Einstein theory with
additional term feR). This sort of additional terms was widely used in the models
of the early Universe, including the models of inflation. In this work we will use the
term feR) IX Rn with n > O. The main purpose is searching for a such value of n
which corresponds as much as possible to the given set of observational data from
z::::; 1000 till present time (z ::::; 0). At the same time, we try to construct the model
which does not contradict to the scenario of the large-scale structure formation.

2. Basic equation
As it was mentioned above, we start with the following gravitational Lagrangian:

where a is a positive constant, and the flat FRW metric

The corresponding 0-0 component of gravitational equations will be:

(1) I] (yl)2-n
f3 [ n(n-l)y"y+ ;n (y')2+ n (4-3n)YaY = a4 - 3n (y-Dm a ). (1)

1800
 1801

Here j3 = (_3)n- 1 a, y == (iLa?, Om == (Pm/p*)a 3 , Pm is the a-dependent cold dark

matter (CDM) energy density, P* = 3H'!;/87fG is the critical density. A general
approach to investigation of the last equation is given in Ref.2 Let us examine the
Eq. (1) in this context. In Eq. (1) there are two parameters, j3 and n, determined
by the observational data. These parameters could be chosen according to different
requirements. 3,4 Here we will choose parameter n from requirement of closeness of
evolution of our model to the classical solution for the flat FRW Universe with cold
matter in the past. 5 This fact allows this scenario to be close to the scenario of the
large scale structure formation. This requirement can be realized at condition that
the classical Friedmann solution

(2)
is a particular solution of Eq.(l). It is easy to see from Eq.(l), the last condition is
equivalent to the choice of n to be satisfying the equation

n - 1 = 2n(4 - 3n) with roots nl = 1.295, n2 = -0.129. (3)

The first of the roots leads to the type of models of papers,3,4 while the second root
corresponds the models with correction of the form ex ,4 Rln21 investigated in Ref.6
As it was shown in Ref.,5 a choice of n = nl approaches the model to the set of the
observational data in the best way.

3. Conclusions
As a set of observational data, the analysis of SNe and CMB data from Ref. 7 is used.
The cited paper presents two conceptions of the analysis of the observational data:
the first of them is the best fit to the data which uses only the hydro dynamical
describing of DE and does not impose restrictions on the values of Om and ho,
while the second conception follows the priority of the concordance ACDM model,
so authors of Ree put Om = 0.27 ± 0.04 and ho = 0.71 ± 0.06. We will give the
comparison of our results with both of them.
As it was found in Ref.,7 the best fit values are: Om = 0.385, ho = 60. In the
model DE evolves in time strongly enough. For given Om and ho we compared
the results of the f(R)-model with n = 1.2955 (bn = 0.0002) and j3 = 0.273 for
the "geometric equation of state" parameter w = (y - ay')/3(y - Om a) and the
deceleration parameter q with the results of Ree In this f(R)-model the age of
the Universe is 14.9 Hyr, the deceleration parameter is qo = -0.91 at present and
the transition to acceleration occurs at z = 0.38. Similarly to results of Ref.,7 the
WDE < -1 at lower redshifts (WDEO = -1.53), however, the evolution of equation
of state of "geometric DE" is more weak contrary to the results of Ref. 7
However, if strong priors have been imposed on Om and ho (i.e. the ACDM
model priors: Om = 0.27 ± 0.04 and ho = 0.71 ± 0.06), the evolution of DE is
extremely weak and in good agreement with the ACDM model. The best fit in the
case is Om = 0.29 and there is a good enough coincidence of our model and their
1802

Gold+CMB
Gold+CMB

o 0"

-1

-1

o o
z z

Fig. 1. The comparison of results of feR) ex: R n model with n 1.296 and (3 = 0.467 (the thick
solid line) to results of analysis of SNe+CMB data with Om 0.29 done in Ref. 7 The evolution
of the deceleration parameter with redshift is shown in a right panel and variation of equation of
state of DE is shown in a left panel. The best fit of SNe+CMB data in such case is represented
by the thin solid line, the 10' and 20' confidence levels are represented by the light and dark grey
contours, respectively, and ACDM is represented by the dashed line.

analysis for parameters of the model n = 1.296 (On 0.001) and (3 = 0.467 (see
Fig. 1). The deceleration parameter qo = -0.683 at present, and the deceleration
was changed by the acceleration at z 0.51 (qO = -0.63±0.12 and z = 0.57±0.07
in Ref.7). The age of the Universe in this case is 13.6 Hyr.
Thus, the feR) ex: (3Rn model with parameters (3 and n chosen according to the
principles mentioned in Introduction describes the evolution of the Universe quite
corresponding to the SNe+CMB data.

References
1. A.Riess et al., Astron. J. 116, 1009 (1998), astro-ph/9805201;
M.Tegmark et al., Astroph. J. 606, 702 (2004), astro-ph/0310725.
2. V.Ts. Gurovich and A. A. Starobinsky, Sov. Phys. JETP 50,844 (1979).
3. S. Capozziello et. al., Int. J. Mod. Phys. D12, 1969 (2003).
4. V. Folomeev, V. Gurovich and H. Kleinert, astro-ph/0501209.
5. V. Gurovich and I. Tokareva, astro-ph/0509071;
V. Folomeev, V. Gurovich and I. Tokareva, Grav.f3Cosm. 46-47, 163 (2006).
6. S. M. Carroll et. al., Phys. Rev. D 70, 043528 (2004).
7. U. Alam, V. Sahni, A. A. Starobinsky, JCAP 0406, 008 (2004), astro-ph/0403687.
 PROBING DYNAMICAL DARK ENERGY WITH
PRESS-SCHECHTER MASS FUNCTIONS

MORGAN LE DELLIOU
CFTC, Lisbon University,
Lisbon, Portugal
delliou@cii·fc.ul.pt

This project proposes to discriminate in the wealth of models for dark energy using
the formation of non-linear dark matter structures. In particular,it focuses on structures
traced by the mass function of dark matter haloes.

Keywords: Semi-analytic modeling, Dark matter, Galaxy clusters, Dark energy theory

1. Introduction

Among the various models proposed for the explanation of the accelerated expansion
in terms of dark energy (DE), distinctions can be made between static (cosmological
constant) or dynamical (e.g. quintessence), coupled or uncoupled to dark matter
(DM), clustering or unclustering or even unified (Chaplygin Gas) DE. This wealth of
models calls for discriminating schemes. The goal of this work is to propose a unified
analysis extending previous studies (see 1e Delliou 2006,1 Manera & Mota 2006 2
and 3 references therein) of DE impact on dark matter haloes on mass functions, for
confrontation with other DE assessments.

Table 1. Choice of scalar field's potentials

Explored modelsr;" rrlOdels under scrutiny

Model Potential V Origin Model Potential V

R.P.
A 4 +a
_Q_-
QQ
Historical potential; Albrecht & At (("Q - M,)2
global SUSY Skordis 2000 4 +M2 ) e-AKQ

SUGRA
SUSY+extra
dim.=SUGRA
Sahni & Wang At (cosh (')'"Q)
2000,0 -1)"
superpot.
Ferreira & Joyce extra dim. Dodelson et ai, Ate-AKQ (1 +
1998 cOlllpactification 2000 6 asin(v"Q))
Steinhardt et al. A 4+a

1999
RP.xF.J. 3" e-
AKQ

Voe.3(a l)4>X

Barreiro et al. At (e-,,"Q double exponential

Bertolarni et ((COSh U)~)) Q~'
2000 +e-,6KQ) al. 2004 7
+ (cosh (2;";,')) ;;:!~)

2. Models and Mass functions

We model a cosmic fluid with baryons, radiation and, either uncoupled (and
coupled) DM with a(n) (un)clustering scalar field DE (quintessence) Q, (non-)
minimally coupled to DM, or a Chaplygin Gas (GCG) - DMjDE unified component

1803
1804

Mass Functions at z=0.5 and 1

for variou~ QCDM mudels (and L=O for Reference)

-2
-

~
:>:
LCDM
~:E
-4
RP6
RP]I
A
"-'
'ob
SUGRA6
.Q SUGRAll
-6
PJ
Steinhardtet al

-8

-2

-~

~ -4
:>: LCDM
<:'
~ RP6

~ -6 RPII
SUGRA6
"-'
~ SUGRAll
FJ
-s Steinhardtcl al

-10
1O() 10' !O2 10' 10' 5
10
I
M (in lO" h- MJ

Le Delliou 2006 1

Barreiro et al. model from Manera & Mota 2006: 2 A = non dust., large 0cDMo' B non Clust.,
small OcDMo' C = Clust., large OcDMo ' D = Clust_, small OcDllfO'

-----JA]
... ' B

[== ~CDM
',c

N
A
;;:
Z

Manera & Mota 2006 2

Fig. 1. Cumulative mass functions for different models

 1805

- defined either from P ex _p-a or from a scalar field mimicking it. We restrict
to fiat backgrounds, a linear coupling and model clustering through energy conser-
vation. The models are defined by their potential (Table 1). We already studied
homogeneous minimal quintessences (left side, upper part 1 ) and a coupled, cluster-
ing quintessence (left side, lower part 2 ). All potentials shown will have clustering
and interaction.
We use the top hat spherical collapse to model non-linear structure formation
as a Friedmann sphere with higher, varying curvature. We extract the linearly ex-
trapolated overdensity as a function of non-linear collapse scale factor bco(a c). This
is combined in a Press-Schechter scheme to get the mass function of large scale
structures. The results obtained so far are presented in Fig. 1.

3. Conclusions
Extending previous evidence of DE models impact on DM mass functions, our re-
sults permit the confrontation of several homogeneous models and the examination
of clustering and interacting quintessence. This have shown that more insights can
be drawn from confrontation of several homogeneous models 1 and that strong effects
on mass function evolution proceed from clustering and interacting quintessence. 2
Indeed, the spread of ",10% at 10 14 h- 1 M0 between mass functions and the hierar-
chy between models on the lower (z = 1) panel of the left part of Fig. 1 shows that
the method should be most discriminant on clusters scales and that the impact of
wQ dominates other effects. Moreover, its right part entails that, contrary to ho-
mogeneous models, DE clustering increases DM clustering while coupling decreases
it. This motivates our extended study of DE models with mass functions. Some
pending questions remain: our use of Birkhoff's theorem with spherical symmetry in
cosmology may require some mass function corrections; geometric effects are argued
to induce a degeneracy in angular mass functions, 8 not taking the bias-geometry
dependence 9 into account. We are extending our results to other models (Table 1;
Chaplygin gas), including clustering and interacting DE. Further developments are
also planned.

Acknowledgements
See acknowledgements in. 1 Current work involves J.P. Mimoso, U. Lisboa, D.F.
Mota, U. Heidelberg, C. van de Bruck, U. Sheffield and O. Bertolami, 1ST Lisboa.

References
1. M. Le Delliou, leAP 0601, 021 (2006) [astro-ph/0506200]
2. M. Manera and D. F. Mota, MNRAS 371, 1373 (2006) [astro-ph/0504519]
3. R. Mainini and S. A. Bonometto, Phys. Rev. D 74, 043504 (2006) [astro-ph/0605621]
4. A. Albrecht and C. Skordis, Phys. Rev. Lett. 84, 2076 (2000) [astro-ph/9908085]
5. V. Sahni and L. Wang, Phys. Rev. D 62, 103517 (2000) [astro-ph/9910097]
6. S. Dodelson, M. Kaplinghat and E. Stewart, Phys. Rev. Lett 85, 5276 (2000) [astro-
ph/0002360]
1806

7. O. Bertolami, A. A. Sen, S. Sen and P. T. Silva, MNRAS 353, 329 (2004) [astro-
ph/0402387]
8. P. Solevi, R. Mainini and S. A. Bonomento, ApJ (2004) [astro-ph/0412054]
9. N. Kaiser, ApJ 284, L9 (1984)
 BROKEN SCALE INVARIANCE AND QUINTESSENCE
(a quarter of a century ago)

GIOVANNI VENTURI
Dipartimento di Fisica, Universita degli Studi di Bologna, and INFN, via Irnerio 46, 40126
Bologna, Italy. [email protected]

The cosmological consequences of a simple scalar field model for the generation of New-
ton's constant through the spontaneous breaking of scale invariance in a curved space-
time are again presented and discussed. Such a model leads to a consistent description
wherein the introduction of matter introduces a small perturbation on a de Sitter uni-
verse and a time dependence of the gravitational coupling.

A quarter of a century ago we suggested a simple scalar field model for the gen-
eration of the gravitational coupling (Newton's constant) through the spontaneous
breaking of scale invariance in a curved space time. 1 In this brief note we wish to re-
visit the model which consisted of the following globally scale invariant Lagrangian
density for a scalar filed (J'

L 1 gfl. v 0 (J'O ,\ 4
+ '22 R + L m
= __
2 fl. v
(J' - -(J'
4
-(J' (1)

where " ,\ are dimensionless positive constants (,\ > 0 for stability), Lm is the
matter Lagrangian density (does not contain (J') and R the curvature scalar. In
flat space one can have a non-zero vacuum expectation value for (J' (\(J')o =F 0), in
our case we assume the vacuum is a scalar particle condensate and examine the
cosmological consequences. We consider a Robertson-Walker line element

(2)

and matter behaving like an isentropic perfect fluid having energy-momentum tensor

(3)

where p(t), pet) and uQ. are the energy density, pressure, and velocity four-vector,
respectively. The Einstein equations obtained from Eq. (1) are

p=-3~(p+p) (4)
a

a2
-2 + -2
k
= --
p
+- - -
1 &2 a& 1 2
2- - + -,\(J' (5)
a a 3''')'0'2 6, (J'2 a (J' 12,

. 3) -_ (p - 3p) a.
-d ((J'(J'a 3
(6)
dt 6, + 1
where the dot denotes differentiation with respect to the time t.

1807
1808

In the absence of matter (p = p = 0) one obtains a time independent "vacuum"

solution given by

(7)

where Ro is a constant and corresponds to a fiat space (k = 0) of constant curvature

R o/12 (de Sitter),
a = ao(t) = ao(O)exp(Hot) (8)

Ho -
_ (~ (Jo2)1/2 _
-
(Ro)1/2 (9)
12, 12
Further, from the weak-field limit,2 Newton's coupling constant G is given by:
1 26, + 1
--=,(J - - (10)
87rG o 0 8, + 1

One now treats the introduction of matter as a small perturbation and writes

(J = (Jo[1 + X(t)], (11)

a = ao(t) + oa(t), (12)

where X and Oa are assumed small. We keep k = 0 and assume an equation of state
p = wp(t) where w is a positive or zero constant.
To lowest order Eqs. (4-6) become
p cio
- = -3(w + 1)- = -3(w + I)Ho, (13)
p ao

2Ho . p 2
-(oa - Hooa) = - - 2 - 2HoX + 2HoX, (14)
ao 3,(Jo

.. 3 .H __ 1 _ (1 - 3w)p
X+ X 0 - 6, + 1 (J5 ' (15)

and their solutions are

pet) = p(0)e- 3 (1+ w )Ho t, (16)

(1 - 3w) p(O) (e- 3 (1+W)H ot

1
X = --
_
- e 3Hot + -w- ) (17)
6, + 1 (J5 9H6W 1+w w+1

oa = (1 - 3w) p(O)ao(t) [(1- e-3Hot(1+W)) (4 + 3w) _ ~ (1 _ e3Hot ) + WHot]

6,+19(J5H6w 3(I+w)2 3 w+1
+ p(O)aoCt) (1 _e-3Hot(1+W)) (18)
18'"1'(1 + w)H{;(J6 '
where we have imposed the boundary conditions XeO) = xCO) = oa(O) = o.
 1809

As a consequence of the above, the gravitational constant G acquires a time

dependence given by
G
-~-2X=-
. (1 - 3w) 2p(t)
(_1+e+3wHot) , (19)
G 6, + 1 3How(J"6
from which:
(1 - 3w) 2p(t) ( 1 3
we+ (1+W)H t)] o
G(t) = Goexp - _ - - e+3wHot
[ 6, + 1 9H;5w(J"6 1 + W
+- ---- (20)
W +1

On setting W = 0 (dust), taking p(to) = 2 x 10- 30 g/cm3 , to ~ 2·7 X 10 17 sec

and Ho c::=3;0
one has

(21)

g
and at present I It=to < 2 X 10- 18 sec- 1 . Further, again for W = 0 and, small
(actually from solar system measurements, « 1), it is sufficient to just consider
the last term in Eq. (18) obtaining:

HT(t O) = ~I = ao(to) (1 + ~a(to) _ oao(to)) c::= Ho(l + 0.1) (22)

a t=to ao( to) ao (to) ao(to)
where oa(to) ~ 0.2a(to). Finally one obtains for the deceleration parameter

qo(to) ~ - ~~ I ~ -0.7 (23)

a t=to
We thus see that, at the present time (just as a quarter of a century ago!), it is
consistent to introduce matter as a perturbation on a de Sitter universe.

References
1. F. Cooper and G. Venturi, Phys. Rev. D 24 (1981) 3338.
2. G. Turchetti and G. Venturi, Nuovo Cim. A 66 (1981) 221.
This page intentionally left blank
Topology of the Universe
This page intentionally left blank
 AN AXISYMMETRIC OBJECT-BASED SEARCH FOR A FLAT
COMPACT DIMENSION*

GRANT J. MATHEWS and DYLAN MENZIESt

Center for Astrophysics, University of Notre Dame, Department of Physics, Notre Dame, IN
46556 USA
[email protected]

A new method is presented to search for a hypertorus symmetry axis by the alignment of
distant objects. This method offers greater sensitivity than previously proposed object-
based methods that rely on accurate true distances. When applied to the catalog of
objects with z > 1, we deduce a lower limit to the compact dimension size D > 0.9 of the
distance to the present cosmic horizon. This is consistent with independent constraints
from the recent analysis of the WMAP microwave background data.

1. Introduction
General relativity makes no direct statement about the universal topology, only
its local curvature. Indeed, a compact topology seems much more natural than an
infinite three-space. Moreover, observations of the cosmic microwave background
(CMB)l and the accumulation extragalactic object surveys2 have spurred interest
in detecting evidence of a compact topology by looking for correlations in the posi-
tions of distant objects,3-5 or patterns in the CMB. 6 ,7 The strongest limits on this
possibility appear to come from recent analyses of the CMB. 8 ,9
Recently, the motivation for a search for a hypertorus cosmology has been
boosted by possible evidence of a suppression 10 in the quadrupole and octupole
moments of the CMB along a common axis. This suppression appears significant
despite the increased cosmic variance at large scales. l l While other explanations of
this suppression have been put forward,12-14 there remains the possibility that it is
due to a sub-horizon scale compact dimension. IN this work, we have considered 15
a compact dimension in flat space as produced by toroidal compactification, or as
an approximation to slightly curved space. 16 For this case, we show that it is possi-
ble to construct a specialized object-based test with much greater sensitivity than
previous object-based tests. This provides an independent approach to the CMB
methods.

2. Axisymmetric test
The test we have developed 15 is specific to cylindrical compactification. This ap-
proach can be modified for the other flat-space compactifications that arise from

*Work supported in part by the US Department of Energy under grant DE-FG02-95ER40934

and a University of Notre Dame Center for Applied Mathematics (CAM) summer fellowship. This
research has made use of the NASA/IPAC Extragalactic Database (NED) which is operated by the
Jet Propulsion Laboratory, California Institute of Technology, under contract with the National
Aeronautics and Space Administration.
tPresent Address; Monohan University, UK

1813
1814

variations of the hypertorus, and even for curved space, but then the upper limit
for the detected compact dimension is not as useful. The test is based upon the
observation that pairs of images (each corresponding to two copies of an object in
the covering space) are each aligned with the center. This observation suggests that
we look for an axis direction with an unusually high number of such alignments.
Specifically, we must look for image matches on opposite sides of the sky for
each possible axis of the compact dimension. The alignment check is made more
efficient by maintaining a list of objects ordered by angle, ¢, about the search axis.
Alignments are then found by running through the list only once. After shifting
the axis, the list is reordered slightly using a fast sorting algorithm. Due to the
finite angular resolution of the search D.a, each object is given an allowed range of
e e
¢ values [¢ - D.¢, ¢ + D.¢] where D.¢ = D.a/ sin and is the angle of the object
away from the axis. When the ¢ ranges of two objects on opposite sides of the axis
overlap, a candidate match has been found.
To reduce background noise from chance alignments, we performed further tests
once an alignment was found. Quasars are not expected to live longer than 10 9
years,4 so images that differ in look-back time by more than this are discarded. The
look-back time is determined from the redshift. We assume a flat ACDM models
with 0.17 ::; Om ::; 0.42. The look-back filter also has the effect of greatly reducing
the angular error caused by the peculiar velocity of the object to a value which is well
below other errors. We also found that it is necessary to exclude false peaks which
result from tight clusters of correlated objects. These exist due to the duplication of
objects in the database and the high concentrations of objects found in pencil-beam
surveys. These are excluded by organizing pairs into groups that are separated by
a minimum angle of 0.5 0 • This angle was chosen because it is sufficient to remove
the clustering effect, while giving a very small chance of reducing the numbers of
genuine matching pairs.
Once a pair of images has passed all tests, the compact dimension size is es-
e
timated as D = d 1 cos e1 + d 2 cos 2 . The alignment is used to increment the bin
corresponding to D in a histogram corresponding to Om. As a test of the method
and to calibrate the bin sizes we added simulated aligned objects to the database.
The D bin and Om spacing are made just large enough so that a simulated set of
aligned images, of look-back difference less than 1 Gyr, with maximum expected
noise added and maximum misalignment between the Om values, can be fully de-
tected in one bin.
The test was applied to a catalog of 40000 objects with z > 1.0 (mostly quasars)
obtained from the NASA/IPAC Extragalactic Database t (NED). This data base
was corrected for the aberration due to Earth's motion as well as the aberration of
the observed redshifts. The search space was divided evenly between 20 1.4 GHz
Linux workstations, and ran for 5 days. There was no clear evidence of a peak. Figure
1 shows the maximum number of alignments found in each D bin for Om = 0.3, once

thttp://nedwww.ipac.caltech.edu/
 1815

all the searches are combined. The contours are for 10", 20", and 30" significance.
We have deduced 15 that the survey is complete enough to make detection possible
over a wide range of D: A completeness of 0.02 rules out a compact dimension up
to 0.9 TJo, using the 30" criterion.
The approach described here 15 is more sensitive than previous object-based tests
for nearly flat compact cosmology, because it does not rely on accurate true dis-
tance determination, and searches in a focused way for a particular symmetry.
Although there is uncertainty over the degree of completeness among the observa-
tions, the results presented here suggest that a hypertorus cosmology is ruled out
for D < 0.9 TJo. Selection could be improved by matching magnitude and spectral
profiles, using greater an enlarged catalog. Deeper observations going beyond the
current:::::: 0.5 TJo limit will also allow us to extend the range of D tested.

References
1. Bennett C L et al 2003, Astrophys.J.Suppl., 148, 1.
2. Adelman-McCarthy, J., Agueros, M.A., Allam, S.S., et al. 2006, ApJS, 162, 38.
3. Lehoucq R et al 2000, Astron.Astrophys. 363 (2000) 1, astro-ph/0005515.
4. Roukema B F 1996, M.N.R.A.S. 283, 1147, astro-ph/9603052.
5. Roukema B F 2002, astro-ph/0201092.
6. Cornish N J, Spergel D N & Starkman G 1998, Phys. Rev. D., 57, 5982.
7. Levin J 2001, gr-qc/0108043.
8. Cornish N J et al. 2004, Phys. Rev. Lett. 92, 201302.
9. Key J S, Cornish N J, Spergel D N & Starkman G 2006, astro-ph/0604616.
10. Tegmark M, de Oliveira-Costa A, Hamilton A 2003, Phys. Rev. D, 68, 123523, astro-
ph/0302496.
11. de Oliveira-Costa A et al 2003, Phys. Rev. D, 69, 063516.
12. Piao Y-S et al 2004, Phys. Rev. D, 69, 103520, hep-th/0310206
13. Moroi T, Takahashi T 2004, Phys. Rev. Lett. 92, 091301, astro-ph/0308208.
14. Contaldi C R et al 2003, JCAP 0307, 002, astro-ph/0303636.
15. Menzies D and Mathews G J 2005 J. Cosmol. Astropart. Phys. 10, 008.
16. Mota B et al. 2004, Class.Quant.Grav. 21, 3361-3368.
17. Aurich, R., Lustig, S. and Steiner, F. Class.Quant.Grav. 22 (2005) 2061-2083.
18. Luminet et al. 2003, Nature 425, 593.
19. Roukema et al. 2004, A&A 423, 821-831.
20. Aurich et al., astro-ph/0412569.
21. Weeks et al. 2004, MNRAS 352, 258-262.
22. Levin J 2004, astro-ph/0403036.
23. Lineweaver C H 1996, Moriond CMB Proceedings, astro-ph/9609034.
24. D. Menzies and G. J. Mathews 2005, ApJ, 624, 7.
25. Carroll S M et al 1992, ARA&A 30 499-542.
26. Gorski K M, Hivon E, Wandelt B D 1999, Analysis Issues for Large CMB Data Sets,
in Proceedings of the MPA/ESO Cosmology Conference, astro-ph 9812350.
 TOPOLOGICAL GRAVITATION ON GRAPH MANIFOLDS

N.V. MITSKIEVICH, V.N. EFREMOV, AND A.M. HERNANDEZ MAGDALENO

GUGEl, Universidad de Guadalajara, Guadalajara, Jalisco, Mexico
Apdo. Postal 1-2011 G.P. 44100, Guadalajara, Jalisco, Mexico
[email protected]

A model of topological field theory is presented in which the vacuum coupling constants
are topological invariants of the four-dimensional spacetime. Thus the coupling constants
are theoretically computable, and they indicate the topological structure of our universe.
We construct an Abelian BF-type model in analogy with the ordinary four-
dimensional topological field theory I and with the low-energy effective U (1 Y -theory
ofSeiberg-Witten (SW)2 , beginning with a U(1Y-bundle E over a four-dimensional
topological space X with a non-empty boundary ax, E being a direct sum of
linear bundles LI EEl ... EEl L r . Let us define locally connection I-forms A a (a =
1, ... , r) on E with values in the algebra L of the group U(I), and 2-forms Ba
with values in the dual algebra. Due to these analogies it is natural to write the
action as S = J Fa II Ba - ~AabBa II Bb + ~8abFa II Fb where Fa = dAa; Aab and
8 ab are non-degenerate symmetric matrices called those of the coupling constants
and theta angles matrices, respectively. Our action admits symmetry under dual
conjllgation similar to the electro-magnetic one (EM duality) of the SW theory,
namely Fa -+ FDa = AabFb; Ba -+ BD = A ab B b; Aab -+ A ab ; 8 ab -+ 8 ab =
AacA bd 8 cd , AabAbc = 6~. These duality transformations carry S into its equivalent,
SD, and they involve the strong-weak coupling duality.2 We call Aab and Aab strong
and weak coupling constants matrices (in this sense), respectively.
Then we generalize Dirac's quantization conditions: the flux through non-trivial
2-cycles 2::1 must be f"EJ Fa = 27rm I , m'} E Z. Then from dynamical equations
of our BF system dB a = 0, Fa = AabBb together with the gauge symmetries
it follows that the moduli space of this BF system is H2(X, Z) EEl H2(x, ax, Z)
meaning that [2~Fal E H2(X, Z) and [2~Bal E H2(X, ax, Z). Thus there are
no local degrees of freedom and like in the case of the low energy efFective SW
action, our model describes the moduli space of vacua. The spacetime topology
is non-trivial since we model the spacetime by the graph manifold. 3 Each tree
graph corresponds to a unique four-dimensional space X with a boundary con-
taining lens spaces and Z-homology spheres. The latter ones are results of splic-
ing of Seifert fibred homology (Sfh) spheres. The most important construction
element is Sfh-sphere 2::(gJ == 2::(al' a2, a3) having three special orbits and be-
ing a three-dimensional manifold which is an intersection of the Brieskorn sur-
face ZI a, + Z2 a2 + Z3 Q3 = 0 (Zi E rei) and a sphere S5. Here aI, a2 and a3 are
mutually prime integers (Seifert invariants). To the end of constructing our cos-
mological model we need a specific family of Sfh-spheres to which we give the
following definition: 3 We take a succession of Sfh-spheres calling it the primary one:
{2::(q2n-I,P2n,P2n+I)ln = 0, ... ,4}. Here Pi is the i-th prime number in the natu-
ral series and qi = PI ... Pi. Then we define the "derivative" of a Sfh-sphere 2::(g)

1816
 1817

~ (510510,19,23) _---HIe+-_---+<H-l-_-+...,I--__---1__+-_____-~

~ (2310, 13, 17) __-HIe+-_---H~-_---H~----<_~

~ (30, 7,11) ...-HIe+-_-HM+-_---1itl

~ (2,3,5) _._---1__+-_____- __)

~(1,1,2)

Fig. 1. The graph associated with a four-dimensional manifold interpreted as a Euclidean region
of spacetime X with boundaries ax = - U;=l L(pi' qi) U M. This graph describes a universe with
five low energy interactions related to the first nine prime numbers as (1,2), (3,5), (7,11), (13,17),
(19,23) determining Seifert's invariants of Sfh-spheres being glued together. Vertices on horizontal
lines represent (from left to right) successive "derivatives" of the primary Sfh-spheres shown on
the diagonal line. At the four-dimensional level splicing gives the plumbing operation.

as another Sfh-sphere ~(1)(gJ := ~(a1,a2o.3,a + 1) == ~(ail),a~l),a~1)).The spatial

sections M of our universe model we construct gluing together (by splicing) the
Sfh-spheres in agreement with tree-type graphs like that we give in Fig. l.
It is interesting to compute the intersection matrices for these manifolds. But
let us first recall that due to the Poincare-Lefschetz duality Wz ; H2(x, ax, Z) ®
H2(X, Z) -+ Z the integer intersection form can be determined as a cup product
wz(b,.f) = (b U f, [X, aX]) for any b E H2(x, ax, Z) and f E H2(X, Z). This is
a generalization of the intersection form in the de Rham representation, Jx b 1\ f.
It is well known that, if one takes some bases bi and r
of groups H2(X, ax, Z)
and H2(X,Z) ('i = 1,,,:,1', l' = rankH2(X,Z)) which are mutually dual in the
sense that wz(b i , f j ) = <S;, the integer intersection matrix wz(b i , bj ) will be inverse
to the rational one WiQ(fi, fj). These intersection matrices represent basic topo-
logical invariants of any graph manifold. We consider solutions of the dynamical
equations dB" = 0, Fa = Aab Bb as mutually dual bases of the respective coho-
mology groups, so that fa = [2171' Fa] and ba = [2~Bal. The second equation yields
WiQ(f'\fb) = Aab, wz(ba,bb) = Aab. Thus the coupling constants matrices Aab and
Aab should be identified with the rational and integer intersection matrices, respec-
tively. Now we return to the graph which describes spacetime of our model. Each
free vertex of this graph (i.e., each Sfll-sphere) corresponds to an element ba of the
basis and to an element fa of the dual basis. Then the intersection matrix elements
correspond to the respective Sfll-spheres forming the graph. We associate with any
Sfll-sphere a certain interaction (inclusion of one more Sfll-sphere into our universe
1818

model results in switching on one new interaction). We associate with the Sfu-
spheres in the extreme right-hand "column" low-energy (LE) interactions (strong,
electromagnetic, weak, gravitational and cosmological). Elimination of these Sfu-
spheres yields a graph describing the earlier stage of the cosmological evolution
and a family of higher energy interactions, thus our model involves an interactions
unification scheme. Each Sfu-sphere has three edges corresponding to three special
orbits, so one may glue it into the graph in three different manners which yields 3 15
different universes. A hypothesis that the "real universe" is found in a mixed state
suggests that the physical meaning belongs to the average intersection matrix:
10- 4 0 0 0 10- 3 0 0 0 0 0 0 0 0 0 0
0 10- 20 10- 13 0 0 0 0 0 0 0 0 0 0 0 0
0 10- 13 10- 6 10- 7 0 0 0 0 0 0 0 0 0 0 0
0 0 10- 7 10- 4 10- 4 0 10- 6 0 0 0 0 0 0 0 0
10- 3 0 0 10- 4 10- 2 10- 2 0 0 0 0 0 0 0 0 0
0 0 0 0 10- 2 10- 1 0 0 0 0 0 0 0 0 0
0 0 0 10- 6 0 0 10- 5 10- 12 0 0 10- 10 0 0 0 0
0 0 0 0 0 0 10- 12 1O- 11 10- 23 0 0 0 0 0 0
0 0 0 0 0 0 0 10- 23 10- 21 10- 44 0 0 0 0 0
0 0 0 0 0 0 0 0 10- 44 10- 68 0 0 0 0 0
0 0 0 0 0 0 10- 10 0 0 0 10- 9 10- 17 0 0 0
0 0 0 0 0 0 0 0 0 0 10- 17 10- 16 10- 33
0 0
0 0 0 0 0 0 0 0 0 0 0 10- 33 10- 31 10- 650
0 0 0 0 0 0 0 0 0 0 0 0 10-65 10- 61 10- 130
0 0 0 0 0 0 0 0 0 0 0 0 0 10-130 10- 197

In this matrix, the boldface non-diagonal elements represent the dimensionless cou-
pling constants of LE interactions. These elements really reproduce the hierarchy
experimentally known for coupling constants of the fundamental interactions. 3 It is
remarkable that in our model the coupling constants of which A ab is built, are basic
topological invariants of the four-dimensional spacetime X. In fact, we theoretically
reproduced the vacuum coupling constants hierarchy existing in the real universe
if the spacetime is a graph manifold. In our scheme the five LE interactions are
related to the first nine prime numbers. To obtain any new interaction, one has
to attach a new pair of prime numbers to the preceding set. With the next pair
(29,31), the same algorithm yields a new coupling constant of the order of magni-
tude (};6 ;::::; 10- 361 • Thus our model answers the question: How many fundamental
interactions may exist in the universe? To the infinite succession of prime numbers
should correspond infinite number of interactions. We simply cannot detect too
weak interactions beginning with (};6, and all subsequent are even much weaker.

References
1. G. Thompson, New Results in Toplogical Field Theory and Abelian Gauge Theory,
hep-th/9511038 (1995).
2. P. Argyres and K. Narayan, String Webs from Field Theory, hep-th/0101114 (2001).
3. V.N. Efremov, N.V. Mitskievich, A.M. Hernandez Magdaleno and R. Serrano Bautista,
Class. Quantum Grav. 22, 3725 (2005); see references therein.
 SUPERNOVAE CONSTRAINTS ON COSMOLOGICAL DENSITY
PARAMETERS AND COSMIC TOPOLOGY*
MARCELOJ.REBOUQAS
Centro Brasileiro de Pesquisas Fisicas
Rua Dr. Xavier Sigaud 150, 22290-180 Rio de Janeiro - RJ, Brazil

We illustrate the constraints that a possible detection of a non-trivial spatial topology

may place on the cosmological density parameters by considering the ACDM model
Poincare dodecahedal space (PDS) topology as a circles-in-the-sky detectable topology.
To this end we reanalyze the type Ia supernovae constraints on the density parameter
plane Ok - OA and show that a circles-in-the-sky detectable PDS topology gives rise to
important constraints on this parameters plane.

1. Introduction

Let us begin by stating our basic context. In the light of the current observations,
we assume that the current matter content of the Universe is well approximated
by a dust of density Pm, (baryonic plus dark mater) plus a cosmological constant
A (p = -PA). In this ACDM context, we additionally assume that the Universe
is modelled by a space-time manifold M4 = IR X M3 with a locally (spatially)
homogeneous and isotropic metric

(1)

where, depending on the spatial curvature k, the geometry of the 3-space M3 is

either Euclidean (k = 0), spherical (k = 1), or hyperbolic (k = -1).
The metric (1) only expresses the principle of spatial homogeneity and isotropy
along with the existence of a cosmic time t, it does not specify the underlying
space-time manifold M4 nor the spatial section M, which are often taken to be
the (simply-connected) spaces: Euclidean lE 3, spherical §3, or hyperbolic space !H[3.
This has led to a common misconception that the Gaussian curvature k of M3 is
all one needs to determine the topology of the spatial sections M 3 . However, it is a
mathematical result that the great majority of constant curvature 3-spaces, M 3 , are
multiply-connected quotient manifolds of the form IR 3/r, §3/r, and JH[3/r, where
r is a fixed-point free group of isometries of the corresponding covering space (see
Ref. 1 for details).
In the general relativity (GR) framework the Friedmann written in the form
Ok = Om + OA -1 makes apparent that the chief point in the search for the spatial
curvature (and the associated geometry) is to constrain the total density Otot =
Om + OA from observations. This amounts to determining regions in the parametric
planes Om - OA (or Om - Ok), which consistently account for the observations, and
from which one expects to deduce the geometry of the Universe.

'This research has been partially supported by CNPq.

1819
1820

Now given that the spatial geometry can be probed but its knowledge does does
not determine the topology of the 3-space M 3 , the question arises as to whether
the spatial topology is an observable that can be be used to set constraints on
the density parameters nm and nk • An important observational consequence of a
observable nontrivial topology 2 of M3 is the existence of the circles-in-the-sky,3 i.e.,
pairs of correlated circles with the same fluctuation of temperature distribution oT
will be imprinted on the CMBR anisotropy sky maps.3,4 Hence, to observationally
probe a putative a nontrivial topology of M 3 , one ought to examine the full-sky
CMBR maps in order to extract the pairs of correlated circles and determine the
spatial topology.
Here we briefly illustrate the constraints that a possible detection of a non-trivial
spatial topology may place on the cosmological density parameters in the ACDM,
i.e., by assuming a Poincare dodecahedral (PDS) topology as a circles-in-the-sky
observable spatial topology we reanalyze the type la supernovae constraints on the
density parameter plane nk - nm that arise from gold set of 157 SNe la, as compiled
by Riess et al.,5 and show that a circles-in-the-sky detection of the PDS topology
gives rise to important additional constraints on parameters of this plane.

2. Constraints and Concluding Remarks

Using the three-year data the WMAP team 6 reports six different values for the total
density ntot ranging from a very nearly flat ntot = l.003~g:gg to positively curved
n tot = l.037~g:g~~ depending on the combination of data set used to resolve the
geometrical degeneracy. The Poincare dodecahedral space (PDS), D = §3 / J*, fits
both this latter density and the suppression of power of the low multi poles observed
by the WMAP team. a Attempts to find antipodal or nearly-antipodal circles-in-
the-sky in the WMAP data have failed. g ,lo On the other hand, hints of matching
circles l l in first year lLC WMAP maps have be found, but although a second
group has confirmed the circles,10 they have also shown that the circle detection
lies below the false positive threshold. 1o Even if one adopts the result that pairs
of antipodal (or nearly antipodal) circles of radius f' :2: 5° are undetectable in the
current CMBR maps the question arises as to whether the circles are not there or
are simply hidden or destroyed by various sources of contamination or even due to
the angular resolution of the current CMBR maps.13 The answer to these questions
requires great care, among other things, because the level of contamination depends
on both the choice of the cosmological models (parameters) and on the topology. 14
Results so far remain non-conclusive, i.e., one group finds their negative outcome to
be robust for globally homogeneous topologies, including the dodecahedral space, in
spite of contamination, 10 while another group finds the contamination strong enough
to hide the possible correlated circles in the current CMBR maps.12,13 Thus, it is

aOther spherical topologies, notably 0 = §3 /0*, where 0* is the binary octahedral group, also
explain these data analysis. 8
 1821

conceivable that the correlated circles may have been overlooked in the CMBR sky
maps searches.

/
/
/

/
/

Fig. 1. A schematic illustration of two antipodal matching circles in the LSS.

In the Poincare dodecahedral space V, which we shall assume here, the pairs of
matching circles are necessarily antipodal as shown in Fig. 1. Clearly the distance
between the centers of each pair of the correlated circles is twice the injectivity
radius of the smallest sphere inscribable V. A straightforward use of trigonometric
relations for the right-angled spherical triangle shown in Fig. 1 yields

Xl = ~=VlOJ
d jl+ZISS dx = tan- 1 [ t anrinj ] (2)
ss ao 1 JDmx3+Dkx2-(Dm+Dk)+1 COSO'

where d lss is the radius of the L88, x = l+z is an integration variable, Dk = 1-Dtot ,
rinj is a topological invariant that is equal to 'if /10 for V, the distance Xl ss is
measured in units of the curvature radius, ao = a(to) = (HoJI1 - Dtotl )-1, and
15
Zlss = 1089. Equation (2) makes apparent that Xl ss depends on the cosmological
scenario, which we have taken to be ACDM.
Equation (2) give the relation between the angular radius 0 and the parameters
of the ACDM model, and thus can be used to set constraints on these parameters.
For a detailed analysis of topological constraints in the context of other models and
data sets, including braneworld inspired models, we refer the readers to Refs. 16
and Refs. 17.
To illustrate the role of the cosmic topology in constraining the density parame-
ter in the context of ACDM model, we consider the V spatial topology, and assume
the angular radius 0 = 50° and uncertainty 60 ':'::'. 6°. Figure 2 shows the results
of our joint 8Ne Ia gold sample 5 plus cosmic topology analysis. There we display
the confidence regions in the parametric plane Dk - Dm and also the regions from
the conventional analysis. The comparison between these regions makes clear that
the effect of the V topology is to reduce considerably the area corresponding to the
confidence intervals in the parametric plane as well as to break degeneracies aris-
1822

ACDM model (SNe la + Topology)

-0.01

~.~....
-0,02

-0.03 ,:.~,
-0.04

c:,"'" -0.05

-0,06

-0.07

-0.08

-0.09

-O·d.~'5~~O.2~=;:'O.~25~~O.3~~O~.35~~O.~4~O~.45~~O.~5~Oc';'.55~~O.·6
"m
Fig. 2. Confidence regions (1-0', 2-0' and 3-0') in the plane Om - Ok ACDM model obtained with
the SNe Ia gold sample assuming a D space topology with., = 50° ± 6°. Also shown are the
contours obtained assuming no topological data (dash-dotted lines) and the ones corresponding to
topology only (dotted lines).

ing from the current SNe Ia measurements. The best-fit parameters for this joint
analysis are Om = 0.32 and Ok = -0.022.

Acknowledgments
The author is grateful to J.S. Alcaniz, M.C. Bento, O. Bertolami, M. Makler, B.
Mota, N.M.C. Santos and P.T. Silva for valuable discussions.

References
1. M. Lachieze-Rey and J.-P. Luminet, Phys. Rep. 254, 135 (1995); J. Levin, Phys. Rep.
365, 251 (2002); M.J. Rebou<;as and G.!. Gomero, Braz. J. Phys. 34, 1358 (2004);
M.J. Rebou<;as, astro-ph/0504365.
2. G.!. Gomero, M.J. Rebou<;as and R. Tavakol, Class. Quantum Grav. 18, 4461 (2001);
G.!. Gomero, M.J. Rebou<;as and R. Tavakol, Class. Quantum Grav. 18, L145 (2001);
G.!. Gomero, M.J. Rebou<;as and R. Tavakol, Int. J. Mod. Phys. A17, 4261 (2002);
J.R. Weeks, R. Lehoucq and J.-P. Uzan, Class. Quantum Grav. 20, 1529 (2003); J.R.
Weeks, Mod. Phys. Lett. A18, 2099 (2003); B. Mota, M.J. Rebou<;as and R. Tavakol,
Class. Quantum Grav. 20, 4837 (2003); B. Mota, G.!. Gomero, M.J. Rebou<;as and
R. Tavakol, Class. Quantum Grav. 21, 3361 (2004); B. Mota, M.J. Rebou<;as and R.
Tavakol, Int. J. Mod. Phys. A20, 2415 (2005).
3. N.J. Cornish, D. Spergel and G. Starkman, Class. Quantum Grav. 15, 2657 (1998).
4. M.O. Calvao, G.!. Gomero, B. Mota and M.J. Rebou<;as, Class. Quantum Grav. 22,
1991 (2005).
5. A.G. Riess et al., Astrophys. J. 607, 665 (2004).
6. D.N. Spergel et al., astro-ph/0603449.
7. J.-P. Luminet, J. Weeks, A. Riazuelo, R. Lehoucq and J.-P. Uzan, Nature 425, 593
(2003).
8. R. Aurich, S. Lustig and F. Steiner, Class. Quantum Grav. 22, 2061 (2005); R. Aurich,
S. Lustig and F. Steiner, Class. Quantum Grav. 22, 3443 (2005).
 1823

9. N.J. Cornish, D.N. Spergel, G.D. Starkman and E. Komatsu, Phys. Rev. Lett. 92,
201302 (2004).
10. J. Key, N. Cornish, D. Spergel and G. Starkman, astro-phj0604616.
11. B.F. Roukema et al., Astron. Astrophys. 423, 821 (2004).
12. R. Aurich, S. Lustig and F. Steiner, Mon. Not. Roy. Astron. Soc. 369, 240 (2006).
13. H. Then, Mon. Not. R. Astron. Soc. 373, 139 (2006).
14. J. Weeks, Class. Quantum Grav. 23, 6971 (2006).
15. D.N. Spergel et al., Astrophys. J. Suppi. 148, 175 (2003).
16. M.J. Rebou~as, J.8. Alcaniz, B. Mota and M. Makler, Astron. Astrophys. 452, 803
(2006); M.J. Rebou~as and J. S. Alcaniz, Eraz. J. Phys 35,1062 (2005); M.J. Rebou~as
and J.S. Alcaniz, Mon. Not. Roy. Astron. Soc. 369, 1693 (2006); M.J. Rebou~as, astro-
phj0605214; M. Makler, B. Mota and M. J. Rebou~as, astro-phj0507116.
17. M.C. Bento, O. Bertolami, M.J. Rebou~as and P.T. Silva, Phys. Rev. D73, 043504
(2006); M.C. Bento, O. Bertolami, M.J. Rebou~as and N.M.C. Santos, Phys. Rev.
D73, 103521 (2006); M.J. Rebou~as, astro-ph/0702428.
 SUPERNOVAE CONSTRAINTS ON DGP MODEL AND COSMIC
TOPOLOGY*

MARCELO J. REBOUQAS
Centro Brasileiro de Pesquisas Fisicas
Rua Dr. Xavier Sigaud 150, 22290-180 Rio de Janeiro - RJ, Brazil

We study the constraints that the detection of a non-trivial spatial topology may place
on the parameters of braneworld models by considering the Dvali-Gabadadze-Porrati
(DGP) and the globally homogeneous Poincare dodecahedral spatial (PDS) topology as
a circles-in-the-sky observable topology. To this end we reanalyze the type Ia supernovae
constraints on the parameters of the DGP model and show that PDS topology gives rise
to strong and complementary constraints on the parameters of the DGP model.

1. Introduction
In the standard cosmology, the Universe is described by a space-time manifold
M4 = lR x lvh endowed with a locally (spatially) homogeneous and isotropic metric

(1)

where, depending on the spatial curvature k, the geometry of the 3-space lv13 is
either Euclidean (k = 0), spherical (k = 1), or hyperbolic (k = -1). The spatial
section lv13 is usually taken to be one of the simply-connected spaces: Euclidean
lR 3 , spherical §3, or hyperbolic lHI 3 . However, given that the connectedness of the
spatial sections A13 has not been determined by cosmological observations, and
since geometry does not fix the topology, our 3--dimensional space may be OIle
of the possible multiply connected quotient manifolds of the form lR 3 If, §3/r, and
lHI 3 If, where f is a fixed-point free group of isometrics of the corresponding covering
space. Thus, for example, for the Euclidean geometry (k = 0) besides lR 3 there arc
6 classes of topologically distinct compact orient able spaces lvh.
The immediate observational consequence of a detectable nontrivial topologyl
of Jvh is the existence of the circles-in-the-sky,2 i.e., pairs of matching circles will be
imprinted on the CMBR anisotropy sky maps.2 Hence, to observationally probe a
putative nontrivial topology of A13, one should examine the full-sky CMBR maps in
order to extract the pairs of correlated circles and determine the spatial topology.
In the context of the 5D braneworld models the universe is described by a 5-
dimensional metrical orbifold (bulk) 0 5 that is mirror symmetric (1':2) across the
4D brane (manifold) M4. Thus, the bulk can be decomposed as 0 5 = M4 X El =
lR X Jv13 X lEI, where El is a 1':2 symmetric Euclidean space, and where M4 is
endowed with a Robertson-Walker metric (1), which is recovered when w = 0 for
the extra non-compact spatial dimension. In this way, the multiplicity of possible
inequivalent topologies of our 3 -dimensional space, and the physical consequences

'This research has been partially supported by CNPq.

1824
 1825

of a non-trivial detectable topology of M3 as the circles-in-the-sky are brought on

the braneworld scenario.
Here we briefly study the constraints that a detection of a spatial topology may
place on the parameters of a simple braneworld modified-gravity model that ac-
counts for the accelerated expansion of the universe via infrared modifications to
general relativity, namely the Dvali-Gabadadze-Porrati (DGP) model,4 as general-
ized to cosmology by Deffayet. 5 To this end we reanalyze the type Ia supernovae
constraints on the parameters of the DGP model and show that PDS topology gives
rise to strong and complementary constraints on the parameters of the DGP model.

2. Constraints and Concluding Remarks

Using the first-year data the WMAP team reported a total density value 6 D tot =
l.02 ± 0.02, while the three-year WMAP article 7 reports six different values for
the Dtot ranging from a very nearly flat Dtot = l.003:::~:~g to positively curved
Dtot = 1.037:::~:~~~ depending on the combination of data set used to resolve the
geometrical degeneracy.
The Poincare dodecahedral space (PDS), V = §3 / J*, explains both the sup-
pression of power of the low multipoles and this observed total density. We note,
however, that other topologies as 0 = §3/0* also remain viable. 9 Attempts to find
antipodal or nearly-antipodal circles-in-the-sky in the WMAP data have failed. 1o
There is, however, claim of hints of matching circles l l in ILC WMAP maps, which
a second group has confirmed 12 but have also shown that the circle detection lies
below the false positive threshold. 12 On the other hand, even if one embraces the
result that pairs of antipodal (or nearly antipodal) circles of radius I 2 5° are unde-
tectable in the current CMBR maps,12 the question arises as to whether the circles
are not there or are merely hidden by various sources of contamination (Doppler,
integrated Sachs-Wolfe, e.g.), or even due to the angular resolution of the current
CMBR maps, as suggested in Ref. 14. The answer to these questions requires great
care, among other things, because the level of contamination depends on both the
choice of the cosmological models (parameters) and on the topology.15 Results so
far remain non-conclusive, i.e., one group finds their negative outcome to be robust
for globally homogeneous topologies, including the dodecahedral space, in spite of
contamination,12 while another group finds the contamination strong enough to
hide the possible correlated circles in the current CMBR maps.13,14
In V space the pairs of matching circles are necessarily antipodal as shown in
Fig. l. Clearly the distance between the centers of each pair of the correlated circles
is twice the injectivity radius of the smallest sphere inscribable V. A straightforward
use of trigonometric relations for the right-angled spherical triangle shown in Fig. 1
yields

x = d zss = Jlf2J /l+z

ISS

Ho dx = tan- 1 [tan Tin j ] , (2)

Zss ao J1 H(x) cosO'
where d zss is the radius of the LSS, x = 1 + z is an integration variable, H is the
1826

Fig. 1. A schematic illustration of two antipodal matching circles in the L88.

Hubble parameter, nk1 - ntot , rinj is a topological invariant (equals to w/10

=
for V), the distance is measured in units of the curvature radius, ao = a(to) =
Xl ss
(HovI1- ntotl )-1, and Zlss = 1089. 6
Equation (2) makes apparent that Xl ss depends on the cosmological scenario.
For the DGP model one has

(3)

where r c is a length scale beyond which gravity starts to leak out into the bulk.
Equations (2) and (3) give the relation between the angular radius a and the pa-
rameters of the DGP model, and thus can be used to set constraints on these
parameters.
To illustrate the role of the cosmic topology in constraining the DGP parameter
we consider the V spatial topology, and assume the angular radius a = 50° and
uncertainty 800 c::: 6°. Figure 2 shows the results of our joint SNe la plus cosmic
topology analysis, where the gold sample of 157 SNe la, as compiled by Riess et
al.,18 was used. There we display the confidence regions in the parametric plane
n k - nm and also the regions from the conventional analysis with no such a topology
assumption. The comparison between these regions makes clear that the effect of
the V topology is to reduce considerably the area corresponding to the confidence
intervals in the parametric plane as well as to break degeneracies arising from the
current SNe la measurements. The best-fit parameters for this joint analysis are
nm = 0.232 and n k = -0.018.
For a detailed analysis of topological constraints in the context of the DGP
and other models, including braneworld inspired models, see Refs. 16 and Refs. 17.
Finally, we note that in Refs. 19 constraints are placed on the DGP models using
supernova and other data but with no such a topological constraint.
 1827

DGP model (SNe ta + Topology)

-0.01

-0.02

-0.03

-0.04

a-'" -0.05

-0.06

-0.07

-0.08

-0.09

-O'"b .1L--O~.1C-5~--:O~.2----'-'O.2-5-~O~.3-~O.3~5-~0.4-----10A5

Fig. 2. Confidence contours (68.3%, 95.4% and 99.7%) in the Om - Ok plane for DGP model
obtained with the SNe Ia gold sample assuming a D space topology with 'Y = 50 0 ± 6°. Also
shown are the contours obtained assuming no topological data (dash-dotted lines) and the ones
corresponding to topology only (dotted lines).

Acknowledgments
Valuable discussions with M.C. Bento, O. Bertolami and N.M.C. Santos are grate-
fully acknowledged. I am grateful to A.F.F. Teixeira for indicating misprints and
omissions.

References
1. G.I. Gomero, M.J. Rebouc;as and R. Tavakol, Class. Quantum Grav. 18,4461 (2001);
G.I. Gomero, M.J. Rebouc;as and R. Tavakol, Class. Quantum Grav. 18, L145 (2001);
G.I. Gomero, M.J. Rebouc;as and R. Tavakol, Int. J. Mod. Phys. A17, 4261 (2002);
J.R. Weeks, R. Lehoucq and J.-P. Uzan, Class. Quantum Grav. 20, 1529 (2003); J.R.
Weeks, Mod. Phys. Lett. A18, 2099 (2003); G.I. Gomero and M.J. Rebouc;as, Phys.
Lett. A31l, 319 (2003); B. Mota, M.J. Rebouc;as and R. Tavakol, Class. Quantum
Grav. 20, 4837 (2003); B. Mota, G.I. Gomero, M.J. Rebouc;as and R. Tavakol, Class.
Quantum Grav. 21, 3361 (2004); B. Mota, M.J. Rebouc;as and R. Tavakol, Int. J.
Mod. Phys. A20, 2415 (2005).
2. N.J. Cornish, D. Spergel and G. Starkman, Class. Quantum Grav. 15, 2657 (1998).
See also M.O. Calvao, G.!. Gomero, B. Mota and M.J. Rebouc;as, Class. Quantum
Grav. 22, 1991 (2005).
3. M. Lachieze-Rey and J.-P. Luminet, Phys. Rep. 254, 135 (1995); J. Levin, Phys. Rep.
365, 251 (2002); M.J. Rebouc;as and G.I. Gomero, Braz. J. Phys. 34, 1358 (2004);
M.J. Rebouc;as, astro-ph/0504365.
4. G.R. Dvali, G. Gabadadze G and M. Porrati, Phys. Lett. B485, 208 (2000).
5. C. Deffayet, Phys. Lett. B502, 199 (2001).
6. D.N. Spergel et al., Astrophys. J. Suppl. 148, 175 (2003).
7. D.N. Spergel et al., astro-ph/0603449.
8. J.-P. Luminet, J. Weeks, A. Riazuelo, R. Lehoucq and J.-P. Uzan, Nature 425, 593
(2003).
9. R. Aurich, S. Lustig and F. Steiner, Class. Quantum Grav. 22, 2061 (2005); R. Aurich,
S. Lustig and F. Steiner, Class. Quantum Grav. 22, 3443 (2005).
1828

10. N.J. Cornish, D.N. Spergel, G.D. Starkman and E. Komatsu, Phys. Rev. Lett. 92,
201302 (2004).
11. B.F. Roukema et al., Astron. Astrophys. 423, 821 (2004).
12. J. Key, N. Cornish, D. Spergel and G. Starkman, astro-ph/0604616.
13. R. Aurich, S. Lustig and F. Steiner, Mon. Not. Roy. Astron. Soc. 369, 240 (2006).
14. H. Then, Mon. Not. R. Astron. Soc. 373, 139 (2006).
15. J. Weeks, Class. Quantum Grav. 23, 6971 (2006).
16. M.C. Bento, O. Bertolami, M.J. Rebou<;as and P.T. Silva, Phys. Rev. D73, 043504
(2006); M.C. Bento, O. Bertolami, M.J. Rebouc,:as and N.M.C. Santos, Phys. Rev.
D73, 103521 (2006);
17. M.J. Rebouc,:as, J.S. Alcaniz, B. Mota and M. Makler, Astron. Astrophys. 452, "803
(2006); M.J. Rebouc,:as and J. S. Alcaniz, Eraz. J. Phys. 35, 1062 (2005); M.J. Re-
bouc,:as and J.S. Alcaniz, Mon. Not. Roy. Astron. Soc. 369, 1693 (2006); M.J. Re-
bouc,:as, astro-ph/0605214;
18. A.G. Riess et al., Astrophys. J. 607, 665 (2004).
19. S. Rydbeck, M. Fairbairn and A. Goobar, astro-ph/0701495; R. Maartens and E.
Majerotto, Phys. Rev. D74, 023004 (2006); M. Fairbairn and A. Goobar, Phys. Lett.
B642, 432435 (2006).
Inhomogeneous Cosmology
This page intentionally left blank
 REINTERPRETING DARK ENERGY THROUGH
BACKREACTION: THE MINIMALLY COUPLED MORPHON
FIELD

JULIEN LARENA
Laboratoire de l'Univers et de ses theoies (LUTH),
CNRS UMR 8102, Observatoire de Paris and Universite Paris 1 Denis Diderot
julien.larena@obspm·fr

THOMAS BUCHERT
Arnold Sommerfeld Center for Theoretical Physics ASC,
Ludwig-M aximilians- Universitiit, Theresienstrafle 31,
80333 Miinchen, Germany
[email protected]

JEAN-MICHEL ALIMI
Laboratoire de l'Univers et de ses theoies (LUTH),
CNRS UMR 8102, Observatoire de Paris and Universite Paris 1 Denis Diderot
jean-michel. [email protected]

In the context of averaged cosmologies, the effective equations can be written in the form
of "regional" Friedmannian equations with additional sources arising from the so-called
backreaction of inhomogeneities. We propose a mean field description of this backreaction
in terms of a regionally homogeneous scalar field: this provides a physical motivation
to the phenomenological scalar fields generically called quintessence fields. We explicitly
reconstruct the potential of the scalar field for a one-parameter family of scaling solutions
to the backreaction problem, showing that it entails most of the standard scalar fields
including e.g. standard and phantom quintessence scenarii.

1. Introduction

Regionally averaged relativistic cosmologies are a way to tackle the so-called coinci-
dence problem of standard Friedmannian cosmologies: why is the expansion acceler-
ating approximately at the same time when the Universe becomes structured, that
is when the density contrast in the matter field is no longer small on a wide range
of scales? The main issue is to try and link the dynamics of the Universe on large
scales to its structuration on smaller scales. 1 It consists in building, from a fully
non-homogeneous Universe, cosmological models that are homogeneous, thanks to
a spatial averaging procedure. It results in equations for a volume scale factor that
not only include an averaged matter source term, but also additional terms that
arise from the coarse-grained inhomogeneities. These additional terms are commonly
named backreaction. In this contribution, we review a correspondence, proposed and
discussed in a recent paper2 between regionally averaged cosmologies and Friedman-
nian scalar field cosmologies, the scalar field being interpreted in this context as a
mean field description of the inhomogeneous Universe, that can play the role of a
field responsible for the dark energy phenomenon.

1831
1832

2. Regionally averaged cosmologies: the backreaction context

In this paper, since we are interested in the late time behavior of the cosmological
model, we restrict the analysis to a Universe filled with an irrotational fluid of dust
matter with density p(t,Xi). Foliating the spacetime by flow-orthogonal hypersur-
faces with the 3-metric gij (that has no a priori symmetry), the line element reads
ds 2 = -dt2 + gijdXidX j . The large scale homogeneous model is then built byav-
eraging the scalar parts of the general relativistic equations on a spatial domain V
with a spatial averager applied to any scalar function Yet, Xi):

(Y(t,Xi))'O = ~ r Y(t,Xi)Jd X
V'O i'O
3
, (1)

where V'O is the volume of the domain V and J = J

det (gij ). Then, one can define a
volume scale factor a'O = (V'O jV'O ,) 1/3 a, and one can build an effective homogeneous
cosmological model on the domain V. The system of equations for that model
characterizes the properties of the Universe on large scales corresponding to the
scale of the domain V. It preserves the main feature of the standard FRW Universe:
the properties of the Universe on large scales can be deduced from a single scale
factor; but, due to the non-commutativity of the spatial averaging and the time
derivatives, this scale factor now obeys dynamical equations given by (2) that differ
from the FRW equations for a dust fluid because of additional source terms, Q'O =
2 ((e - (e)'O)2)'O /3 - 2 ((52)'0' called kinematical backreaction, that are essentially
e
the spatial variances on V of the local expansion rate and of the rate of shear (5:

(2)

Moreover, the averaged 3-curvature (R)'O is no longer a constant curvature term

because it explicitly couples to these variances (see the third equation of system
(2), that is simply an integrability condition and is therefore not an independent
equation). One immediately infers from the second equation of (2) that an effective
acceleration is possible iff Q'O > 41fG (p)'O.

3. Correspondence with scalar field cosmologies

It has been shown 2 that the system describing averaged cosmologies can be cast into
the form of a Friedmannian dust cosmology in the presence of an additional stan-
dard minimally coupled scalar field <P'O with a self-interaction potential U'O(<p'O),
this scalar field representing the backreaction and averaged 3-curvature effects. The

aThis volume scale factor has a simple physical interpretation, since the averaged dust density
r/,
evolves as (p)p ex: a due to our foliation that is locally comoving with the fluid.
 1833

potential UD (<1'>v) is identical, up to a constant factor, with the averaged 3-curvature

and the kinetic energy of the scalar field is a linear combination of the backreac-
tion and averaged curvature. Thanks to that correspondence, one can infer the
dynamical properties of the backreaction and averaged curvature by using all the
constraints that are known on quintessence fields in the context of Friedmannian
cosmologies. For example, if one knows the dark energy equation of state in a par-
ticular quintessence model, one can find the evolution of the backreaction that leads
to such an effective equation of state. For example particular solutions have been
studied: 2 considering backreaction and averaged curvature that are power laws of
the effective volume scale factor, the potential Uv(<I>v) has been reconstructed, and
it corresponds to known quintessence models. 3 Apart from the quasi-Friedmannian
case in which the averaged curvature and the backreaction decouple, the ratio be-
tween backreaction and averaged curvature (IV = Qv/ (R)v) is constant, implying
a strong coupling between them. Then, the properties of the corresponding scalar
field essentially depends on this ratio, and it has been shown that varying this
ratio can lead to a scale-dependent cosmological constant (for 'v = -1/3), and
to almost all types of dark energy scalar fields such as standard quintessence (for
'v E] - 1/3, OD and phantom fields (for 'v E] - 1, -1/3[). So, in this class of so-
lutions, the measure of the dark energy equation of state today can fix the ratio
between backreaction and averaged 3-curvature that is necessary today to explain
dark energy. Finally, it is interesting to note that in the phase plane of these scaling
solutions, the Einstein-de Sitter scenario is a saddle point for the dynamics, so that
a small amount of initial backreaction and/or averaged curvature naturally pushes
the system far from it and may lead to a late-time accelerating phase. The open
question is whether this mechanism can produce a quantitatively sufficient amount
today, a subject of ongoing work. 4

References
1. G. F. R. Ellis, Geneml Relativity and Garvitation, Dordrecht: Reidel (1984); T.
Buchert, Gen. Rei. Gmv. 32, 105 (2000); T. Buchert, Gen. Rei. Gmv. 33, 1381
(2001), T. Buchert, Class. Quant. Gmv. 23, 817 (2006); E. W. Kolb, S. Matarrese
and A. Riotto, New 1. Phys. 8, 322 (2006); T. Buchert, Class. Quantum Gmv. 22,
L113 (2005); Y. Nambu and M. Tanimoto, gr-qc/OS070S7 (2005); A. Paranjape and
T.P. Singh, astro-ph/060S19S (2006); S. Riisiinen, Class. Quantum Gmv. 23, 1823
(2006); S. Riisiinen, J.C.A.P. 0611, 003 (2006); G. F. R. Ellis and T. Buchert, Phys.
Lett. A 347, 38 (2005); M.-N. Celerier, astro-ph/0612222, contribution to this volume
2. T. Buchert, J. Larena and J.-M. Alimi, Class. Quantum Gmv. 23, 6379-640 (2006)
3. V. Sahni and A. A. Starobinskii, Int. J. Mod. Phys. D 9, 373 (2000); V. Sahni, T. D.,
A. A. Starobinskii and U. Alam, JETP Lett. 77, 201 (2003); L. A. Urena-Lopez and
T. Matos, Phys. Rev. D 62, 081302 (2000)
4. T. Buchert, gr-qc/0612166
 INITIAL CONDITIONS FOR PRIMORDIAL
BLACK HOLE FORMATION

ILIA MUSCO
Astronomy Unit, Queen Mary University of London, Mile End Road, London E1 4NS, England
i. [email protected]

ALEXANDER G. POLNAREV
Astronomy Unit, Queen Mary University of London, Mile End Road, London E1 4NS, England
a. g. [email protected]

We study Primordial Black Hole (PBH) formation in the early universe. Working within
spherical symmetry, we specify an initial configuration in terms of a curvature profile,
which represents initial conditions for the large amplitude metric perturbations away
from the homogeneous Friedmann Robertson Walker model, which are required for PBH
formation. Using an asymptotic quasi-homogeneous solution, we relate the curvature
profile with the density and velocity perturbations, which at an early enough time, when
the length scale of the configuration is much larger than the cosmological horizon, can
be treated as small perturbations of the background values. This solution enable us to
consider in a self-consistent way the formation of PBHs in a wide variety of cosmolog-
ical situations. We use two parametric description of the curvature profiles and follow
numerically the evolution of the initial configurations.

1. Introduction
At the end of 60' it has been realised that black holes are possibly formed in the
very early universe due to the collapse of non linear cosmological perturbations. 1-3
Numerical calculations 4- 8 have shown that the necessary requirement to form a
PBR is that the amplitude of metric perturbation should be larger than a certain
threshold value which depends on the shape of the initial curvature profile. 4 To
calculate numerically the formation of PBRs we need to specify the initial conditions
in a self consistent way, based on the quasi homogeneous solution of the Einstein
equations. 9 As a result we express the energy density and the velocity profiles in
terms of the curvature profile K(r), parametrised in terms of two independent
parameters; one describes the strength of the curvature perturbation, the other
describes its sharpness. Finally we present the results of our numerical simulations.

2. Initial conditions in terms of the curvature profile

In the asymptotic quasi homogeneous solution the spatial metric is arbitrary inho-
mogeneous and anisotropic, while the density of the matter tends to become homo-
geneous as time t -> 0. 10 The quasi homogeneous solution is valid if the length-scale
of the perturbation is sufficiently larger than the cosmological horizon. The curva-
ture perturbation does not depend on time in this regime and pressure gradients
are negligible. 4 ,9
The initial perturbation is characterised by a small parameter E == (RH / Ro)2,
where RH is the length scale of the cosmological horizon and Ro is the initial length
scale of the perturbation.
1834
 1835

The initial profiles of the energy density e and the velocity U of the perturbation
are related to K(r) as follows
(1)

(j = ~ [<p(t) - 1] K(r)r6 , (2)

where eb and Ub are the background values described by the Friedmann equation,
and ro is the comoving radius of the initial perturbation. The function of time
<p is the solution of a differential equation written in the first order of the quasi
homogeneous solution for an arbitrary mixture of ideal fluids.u

3. Parametrisation of curvature profiles

We consider two versions of a two parametric description of K(r) given by:

K (r) = (1 + ;;2 )
a exp ( - ;;2 ), (3)

and

(r-~.)')
if r < .6.*
(4)
K(r)={(: (_(r-.6.*)2)
+ 2.6. 2 exp 2.6. 2 if r > .6.*,

where the parameters (3 and .6.* specify amplitude of the gravitational field, while
a and .6. specify the pressure gradients within the initial perturbation.

3. Disconnected region
3. Disconnected region
0.75

0.75
<2.j)
<::Q. <l 0.5
«&
0.5

1. No Black Hole formation

0.25
0.25 1. No Black Hole

formation

0 0
0 0.25 0.5 0.75 0 0.25 0.5 0.75

ex 6,

Fig. 1. These plots shows which values of (a,;3) and (.6.,.6.*) lead to formation of PBHs,
or to an initial perturbation already disconnected from the rest of the Universe.
1836

4. Results
Our results confirm and generalise the conclusion of. 4 In Figure 1 we summarise
the results of our numerical calculations. The left plot refers to the parametrisation
given by (3), while the right hand plot refers to the parametrisation given by (4).
In the plane of the two parameters [(0:,,6) or (,0,.,,0,.*)] one can see three different
regions: the first one corresponds to the dispersion of perturbations into acustic
waves, when the gravitational field is not strong enough to form PBHs. The second
one corresponds to the formation of PBHs and the third one corresponds to causally
disconnected closed universes (not PBHs). One can see that the formation of PBHs
is determined not only by the amplitude of the perturbation but also by the shape
of its profile.

Acknowledgements: In the course of this work, we have benefited from helpful

discussions with many colleagues including John Miller, Bernard Carr, Pavel Ivanov,
Luciano Rezzolla and Carlo Baccigalupi.

References
1. Zel'dovich Ya.B. & Novikov I.D. 1966 Astron.Zh. 43 758 [Sov.Astron. 10 602 (1967)J
2. Hawking S.W. 1971 MNRAS 152 75
3. Carr B.J. 1975 Astrophys.J. 201 1
4. Nadezhin D.K., Novikov I.D. & Polnarev A. G. 1978 Astron.Zh. 55 216 [Sov.Astron.
22(2) 129 (1978)J
5. Bicknell G.V. & Henriksen R. N. 1979 Astrophys.J. 232 670
6. Niemeyer J.C. & Jedamzik K. 1999 Phys.Rev.D 59 124013
7. Shibata M. & Sasaki M. 1999 Phys.Rev.D 60084002
8. Musco I., Miller J.C., Rezzolla L. 2005 Class. Quantum Gmv., 22, 1405
9. Lifshitz E.M. & Khalatnikov I.M. 1963 Usp. Fiz. Nauk. 80, 391 [Sov. Phys. Usp. 6,
496 (1964)]
10. Landau L.D. & Lifshitz E.M. 1975, The Classical Theory of Fields Vol. 2 (Butterwort
Heinemann)
11. Polnarev A.G & Musco I. 2007 Class. Quantum Gmv., 24, 1405
 IS THE APPARENT ACCELERATION OF THE UNIVERSE
EXPANSION DRIVEN BY A DARK ENERGY-LIKE COMPONENT
OR BY INHOMOGENEITIES?

CELERIER, MARIE-NOELLE
Laboratoire de l'Univers et de ses THeories (LUTH), Observatoire de Paris-Meudon,
5 place Jules Janssen, Meudon, 92195 cedex, France
E-mail: [email protected]

Since its decovery during the late 90's, the dimming of distant SN Ia apparent lumi-
nosity has been mostly ascribed to the influence of a mysterious dark energy component.
Based upon the cosmological "principle" hypothesis, this interpretation has given rise
to the "concordance" model, developed in the context of a Friedmann-Lemaltre cos-
mology. However, a caveat of this reasoning is that the cosmological "principle" derives
from a philosophical Copernican assumption and has never been tested. Furthermore, a
weakness of its conclusion, i. e., the existence of a negative-pressure fluid or a cosmo-
logical constant, is that it would have profound implications for the current theories of
physics. This is why we have proposed a more conservative explanation, ascribing the
departure of the observed universe from an Einstein-de Sitter model to the influence of
inhomogeneities. This idea has been independently developed by other authors and fur-
ther enlarged to the reproduction of different cosmological data. We review here the main
proposals which has been made along these lines of though and present some prospects
for future developments.

Keywords: Cosmology; Large Scale Inhomogeneities; Dark Energy

1. Introduction
Since its decovery during the late 90's, the dimming of distant SN Ia apparent
luminosity has been mostly ascribed to the influence of a mysterious dark energy
component 1 which is supposed to drive an acceleration of the Universe expansion.
Based upon the cosmological "principle" hypothesis and coupled to an analysis of
the other cosmological data, this interpretation has given rise to the "concordance"
or ACDM model, developed in the context of a Friedmann-Lemaitre cosmology. A
drawback to this reasoning is that the cosmological "principle" merely derives from
a philosophical Copernican assumption which still deserve to be tested. 2 ,3
The dark energy component, a cosmological constant or a negative pressure
fluid, represents, in the "concordance" model, about 73% of the Universe energy
density. However, a cosmological constant is usually interpreted as a vacuum energy
of which particle physics cannot explain such an amplitude and a negative pressure
fluid remains a mysterious phenomenum. This is known as the cosmological constant
problem.
Another feature of the supernova data analysed in a Friedmannian framework
is to yield a late-time acceleration of the expansion rate, about the epoch when
structure formation enters the nonlinear regime. This would imply that we live
at a time when matter density energy and dark energy are of the same order of
magnitude. This is known as the coincidence problem.
General relativity having only been tested up to scales of order a planetary
system, a second type of explanation has been proposed which implies a modification
1837
1838

of this theory at larger distance scales. 4

Our current purpose is to review the works dedicated to a more simple and
natural proposal, which makes only use of known physics and phenomena. Since it
appears that the onset of the apparent acceleration and the beginning of structure
formation in the Universe are concomitant, the idea that the SN Ia observations
could be reproduced by the effect of inhomogeneities have been put forward. This
interpretation has been first proposed independently by a few authors,2,5-7 shortly
after the release of the data. Then, after a period of relative disaffection, it has
experienced a reniewed interest, specially during the last two years.
Another natural explanation is the possible effect of an actual geometrical cos-
mological constant. Its value such as it appears in the ACDM model has been indeed
predicted in the framework of scale relativity, as A = 1.36 1O-56cm -2, long before
the release of the supernova results. 8 ,g However, we shall not develop this approach
here since it is beyond the scope of this contribution.

2. Presentation and discussion of the methods

It is not obvious that one can keep using FLRW models to interpret the high preci-
sion observational data collected in a regime where most of the mass is clumped into
or is forming structures. Three main physical effects might actually be missing: 10
(i) the overall (average) dynamics of such a universe could be significantly different
from the FLRW one. This effect is usually taken into account by calculating so-
called backreaction terms added to effective equations for the dynamical evolution
of the physical quantities under study, (ii) light propagation in a clumpy universe
might be different from that in a homogeneous one and the luminosity distance-
redshift (LDR) relation could hence be affected, (iii) the fact that we have only one
observer could influence the results, since we might live in an under or over-dense
region which should induce significant corrections in the data interpretation.
What is the most subtle point of this issue is that what is observed in the
supernova data is not an accelerated universe expansion (this is only an artefact due
to the a priori assumption that our universe, even in the local region of structure
formation, can be represented by a FLRW model) but only a dimming of the SN 1a
luminosity taken into account by the infered LDR relation.
However, many authors have focussed their attention on effect (i) and backreac-
tion has been studied with two types of methods: (a) in the linear regime of small
amplitude fluctuations, perturbative expansions which were subsequently averaged
out, (b) for the more general problem of the dynamics induced by all scale inhomo-
geneities including the nonlinear regime, spatial averaging of nonperturbed models.
The local dynamics has also been analysed using different exact solutions of the
general relativity equations.
Effects (ii) and (iii) have been mostly studied using toy models constructed with
such a kind of exact nonaveraged inhomogeneous solutions.
 1839

2.1. Perturbative analysis

The perturbative analysis is a method employed when the deviations from homo-
geneity and isotropy are assumed to be "small" (see Mukhanov et alY for a review).
One considers two spacetimes, the physical, perturbed spacetime and a fictious
background spacetime described by a FLRW model. A one-to-one correspondence
between points in the background and points in the physical spacetime carries the
coordinates labelling the points in the background over into the physical spacetime
and defines a choice of gauge. The perturbation in some quantity is the difference
betwen the values it possesses at a point in the physical spacetime and at the corre-
sponding point in the background spacetime. Since we are interested in considering
the influence of inhomogeneities, the perturbed quantity is here the density.
To study the influence of such fluctuations on the expansion rate of the Uni-
verse, one identifies a local physical variable which describes this expansion rate,
calculates the backreaction of the cosmological perturbations on this variable to a
given order, and then spatially averages the result. Note that it is of the upmost
importance to avoid the deficient procedure consisting of calculating an "observ-
able" from the spatially averaged metric, which, in general, does not give the same
result as calculating the spatial average of the observable. 12 ,13 As we shall stress
in Sec. 2.2, it is also crucial to define the hypersurfaces on which the averaging is
performed by a clear physical prescription.
The cleanest available calculation of the effect of density fluctuations on the aver-
aged expansion rate of a matter-dominated universe up to second order in the metric
variables has been performed by Kolb et aI,14 for an application to superhorizon
scale perturbations, in the adiabatic case. Owing to the perturbation development
employed, this method is only consistent for the study of the effect of very large
scale fluctuations, with small amplitude, e. g., in the linear regime of structure for-
mation. 14- 16 It is also sensitive to the order at which the development is performed.
It has been demonstrated by Notari 17 that, while at early times the contribution of
subhorizon inhomogeneous gravitational fields is perturbatively sub dominant , the
series is likely to diverge around the time of structure formation. There one must
employ a nonperturbative method.
To deal with this issue either spatial averaging of a nonperturbed solution or
exact solutions of Einstein's equations are usually used. A new formalism for the
study of nonlinear perturbations in cosmology, based on a covariant and fully non-
perturbative approach, has recently been proposed 18 but it has not yet been applied
to our purpose.

2.2. Spatial averaging of nonperturbed models

Spatial averaging aims at obtaining the impact of a given inhomogeneity profile upon
the assumed large scale FLRW background, in terms of backreaction terms added
to the Friedmannian evolution equations of scalar quantities such as the expansion
rate, the energy density, the isotropic pressure. These backreaction terms can be
1840

assimilated to a black energy component if they exhibit the right properties.

In general relativity, spatial averaging is very much involved 19 since the equa-
tions which determine the metric tensor and the quantities calculated from it are
highly nonlinear. However, when modelling the Universe, the usual method is to
use continuous functions representing, e. g., energy-density, pressure, or other kine-
matical scalars of the velocity field, implicitly assuming that they represent volume
averages of the corresponding fine-scale quantities. But we know that our local Uni-
verse is highly inhomogeneous from the scales of planetary systems up to currently
unknown distances. 2o Anyhow, the scale, i. e., the size of the volume, over which
the averagings are performed are never explicitly defined, while the results of this
process obviously depend on it.
Moreover, a volume average is a well-defined quantity for scalars only. For vec-
tors, and all the more for tensors, it leads usually to noncovariant quantities. An-
other drawback is that a gauge problem arises when relating the "true" and the
averaged metric. Scalar quantities only are invariant under coordinate transforma-
tions, not tensors. One must also be aware that, in a generic spacetime, there are
no preferred time-slice one could average over and the results depend on the choice
of the hypersurfaces on which this average is performed. 21
But the main issue is the noncommutating property of the two operations: av-
eraging the metric and calculating the Einstein tensor. In other words, the Einstein
tensor calculated from an averaged metric and energy-momentum tensor is not
equal to the Einstein tensor first calculated from the fine-scale metric and energy-
momentum tensor, then averaged. However, in the standard cosmological approach,
the Universe is modelled by adopting exactly the wrong method: take a metric which
is assumed already averaged, calculate the corresponding Einstein tensor and equate
it to an already averaged energy-momentum tensor.
The question raised by the method consisting of determining the parameters
of an a priori assumed FLRW model from observational data is the "fitting prob-
lem" .22 The departure of the "real" Universe from the averaged one is known as
the "backreaction" effect. The study of backreaction can be implemented by two
methods: either one tries to obtain directly the equations satisfied by the averaged
quantities, with minimum assumptions as regards an underlying background, or one
takes as a background a FLRW model and analyses the effect of linear perturbations
applied on it. In this section, we consider the first approach. The second one has
already been studied in Sec. 2.1.
This issue has been de It with by Buchert and collaborators for simplified inhomo-
geneous cosmological models with an irrotational perfect fluid as the gravitational
source. First, Buchert and Ehlers 23 have proposed a simple averaging procedure
which have been used by Palle 24 to deal with the cosmological constant problem.
Then, Buchert 25 ,26 has developed another procedure aimed at constructing an "ef-
fective dynamics" of spatial portions of the Universe from which observable average
characteristics can be infered like the Hubble constant, the effective 3-Ricci scalar
curvature and the mean density (and isotropic pressure) of a spatial domain.
 1841

Relations between average scalar sources (energy density, pressure) and an aver-
age scalar geometric quantity, the expansion rate, have been derived for an adapted
foliation of spacetime. They involve domain dependent backreaction terms that have
been splitted into a "kinematical" backreaction comparing the variance of the ex-
pansion rate to the shear, and a "dynamical" backreaction, i. e., pressure forces.
These equations show that the averaged shear fluctuations tend to increase the
expansion rate as do the averaged energy source terms (provided the energy condi-
tion holds), while the averaged expansion fluctuations have an opposite effect and
therefore work to a stabilization of structures. The dynamical backreaction can do
both.
However, an averaging procedure is not complete unless one also averages the ge-
ometrical inhomogeneities. Since geometrical fields are tensorial variables for which
possible strategies of averaging are not straightforward, Buchert and Carfora 27 have
suggested a Lagrangian smoothing of these variables as opposed to their Eulerian
averaging on spatial domains. Curvature fluctuations turn out to be crucial and
may even outperform the effect of kinematical fluctuations. The authors define ef-
fective cosmological parameters that would be assigned to the smoothed cosmologi-
cal spacetime. These parameters are "dressed" after smoothing out the geometrical
fluctuations. Relations between the "dressed" and "bare" parameters are derived.
The former provide the framework for interpreting observations with a "Friedmann
bias", i. e., as if the observer was living in a Friedmannian universe. The latter
represent the actual inhomogeneous cosmological model, spatially averaged.
In subsequent articles,28,29 the same authors have identified two effects that
quantify the difference between "bare" and "dressed" parameters: the "curvature
backreaction" and the "volume effect". To summarize we can say that, in the
smoothed model, the averaged scalar curvature is dressed both by the volume effect
and the curvature backreaction effect. The volume effect is expected when compar-
ing two regions of distinct volumes, but with the same matter content, in a constant
curvature space. The backreaction term encodes the deviation of the averaged scalar
curvature from a constant curvature model, e. g., a FLRW space section.
However, the interpretation of cosmological parameters remains far from trivial.
One must also average on the observer's light cone in which case the above effects
interact with the time evolution of the model. Moreover, smoothing will have to be
performed in a dynamical setting. Apostolopoulos et apo have uncovered another
subtelty of inhomogeneous cosmology: the volume increase at a given point results
from averaging over various directions a possible anisotropic acceleration parameter
and might thus not be the most appropriate for an expansion characterization.

2.3. Use of exact solutions of Einstein's equations

The method consisting in using exact solutions to modelize the inhomogeneities
observed in the Universe is the most straightforward and devoided of theoretical
pitfalls. It is adapted to represent either strong or weak inhomogeneities.
1842

For mathematical simplification and also to account for the local quasi-isotropy
of the CMB as measured on our wordline, most of the retained models exhibit
spatial spherical symmetry. Therefore, some authors have claimed that unphysical
properties of this symmetry, and specially of the Lemaitre- Tolman-Bondi (LTB)
class,31,32 might prevent these models to be used in this framework, specially when
the observer is assumed to be located at the centre. However, in the literature, the
observer has been put either at the center of the mode1 33 ,34 or offcenter. 35 Moreover,
a spherically symmetric model can represent an anisotropic inhomogeneous universe
averaged over angular scales, which is not physically worse than a uniformly aver-
aged universe such as in the FLRW picture.
Other spherically symmetrical models has been used in this framework, i. e.,
peculiar classes of Stephani solutions. One feature of the Stephani models that has
been the subject of much debate is their matter content. 19 ,36,37 Individual fluid
elements can behave in a rather exotic manner, e. g., exhibiting negative pressure.
But this is also the case of dark energy.
A voiding spherical symmetry, Moffat 38 has considered a peculiar class of models
of the Szafron family. Other exact solutions have consisted of FLRW patch( es)
embedded in a FLRW background with different energy densities. 7,39,40
Depending on the authors, these exact solutions have been used either to barely
fit the data as they are obtained by observation or to provide the inhomogenous
models on which an averaging procedure "a la Buchert" is implemented. 41

3. Studied physical quantities

3.1. The deceleration parameter
When reasoning in the framework of a Friedmannian cosmology, the dimming of
the supernovae is associated with an acceleration of the Universe expansion. This
is why a number of authors have focussed on the issue of either demonstrating or
ruling out an effect of the inhomogeneities on the expansion rate.
Some have tried to derive 42 ,43 or rule out 30 ,44 no-go theorems. However, when
spatially averaged, a physical quantity associated with the expansion rate behaves
quite differently.44-46 It is thus difficult to yield general rules from such theorems.
Some have stressed that the definition of a deceleration parameter in an inho-
mogeneous framework is tricky. Hirata and Seljak42 have proposed four different
definitions of such a parameter. Apostolopoulos et al. 30 have shown that an ob-
server located away from the centre of a spherically symmetric configuration can
measure acceleration or deceleration along radial or perpendicular directions de-
pending on local under or over-density. This demonstrates that, even locally, the
effect of inhomogeneities on the dynamics of the Universe is not trivial.
Ishibashi and Wald 47 have also argued that an averaged quantity representing
the scale factor or the deceleration parameter may accelerate without there being
any observable consequence.
Another unexpected effect has been put forward by Tomita7,39,40 who has con-
 1843

sidered a cosmological model composed of a low-density inner homogeneous region

connected to an outer homogeneous region of higher-density. Both regions deceler-
ate, but, since the void expands faster than the outer region, an apparent accelera-
tion is experienced by the observer located inside the void.
We can therefore conclude that the computation of some local quantity (gen-
erally the deceleration parameter), eventually subsequently averaged, and behav-
ing the same way as in FLRW models with dark energy can lead to spurious re-
sults 13 ,48,49 and must therefore be avoided.

3.2. The Effective Stress Energy Tensor (SET)

Martineau and Brandenberger 50 have tried to estimate the effect of a backreaction
of Super-Hubble modes, by computing it in terms of the second order Effective
SET of cosmological perturbation theory. This has been criticized by Ishibashi and
Wald 47 who have argued that a large SET implies the contribution of higher order
terms in the perturbative scheme, therefore impairing the claimed results.

3.3. The LDR relation

The LDR relation is the only direct product of the supernova data, obtained without
any a priori cosmological assumption. This relation is therefore the best observable
to be fitted and this has been the aim of a number of workS. 2,5,6,10,30,31,33,34,37,51-53

4. Main physical results

It has been proposed that energy density perturbations of wave-length larger than
the Hubble radius, generated during inflation, might produce upon cosmic parame-
ters effects that could mimic an accelerated expansion of the Universe. 14 ,15,50,54 But
it has subsequently been shown by a number of authors,42,43,55-57 among whom some
of the proponents themselves,58,59 that this could not be the case.
Working in the framework of a peculiar toy model, Biswas et al. IO have com-
puted a minimal overall (average) effect which amounts to a correction in apparent
magnitudes at all redshifts of order L:::,.m "-' 0.15.
This effect is to be added to the estimation made by Buchert and Carfora29 of
the volume effect alone in a naive swiss-cheese model. Its magnitude of 67% reduces
the necessary dark energy in the concordance model roughly from 70% to 50%.60
A mismatch of similar magnitude has been reported by Hellaby,61 using volume
matching 22 in LTB models of clusters and voids.
Studies of the constraints which apply to a dust model of universe if one wants
to explain dark energy by the backreaction effect of inhomogeneities have shown
that a negative average spatial curvature is mandatory to compensate a strong
backreaction increasing the expansion rate of the model. 21 ,44,45,56,60,62,63 This result
is actually consistent with the fact that most of the exact inhomogeneous models
which reproduce best the observed LDR relation are those which exhibit negative
1844

spatial curvature near the observer, or, equivalently, where the observer is located
in an underdense region. 10 ,34,39,52,53,64 The assumption that we might be located
within an underdense region seems to be consistent with observations leading to the
identification of a Local Void and of its suggested expansion. 4o ,65
The matching of the observed LDR relation has been successfully performed
by a number of proposed models: homogeneous void models,39 LTB models with
a centered observer,34,51,52 LTB models with an outcentre observer,1O,64 Stephani
models. 5,33,37,53 Most of these models reproduce the ACDM LDR relation up to a
redshift of the order unity. This feature, which has been considered as a ruling out
drawback by some authors,31,32 must be viewed as a nice way out of the coincidence
problem. The physical explanation is that the appearance of small scale inhomo-
geneities corresponds to the onset of structure formation at the time (around z rv 1)
where apparent "acceleration" begins to be observed. 41
However, it has been shown 2 ,66 that the problem of fitting a given LDR relation
with a peculiar inhomogeneous (LTB) model is completly degenerate. Therefore,
the model parameters must be constraint by other cosmological data. This has
been done by some authors: the model proposed by Biswas et al.lO reproduces
matter abundance, the first acoustic peak in the CMB power spectrum and the
baryon oscillations; Alnes et al. 34 obtain the matter density measured at low z and
the two first CMB peaks; Godlowski et al.,33 the three first CMB peaks; Bolejko 31
uses observed cosmological properties to constraint his models. The main challenge
remains to fit the whole set of available cosmological data with a given model. Since
very few exact solutions to Einstein's equations can be of use in a cosmological
framework, only oversimple toy models have been studied up to now to deal with
this issue. However, a project to begin implementing this is currently underway.67

5. Conclusion and prospects

One can find in the literature inhomogeneous cosmological toy models able to solve
both the cosmological constant and coincidence problems, i. e., to mimic an "ac-
celerated expansion" with no need for dark energy up to the epoch when structure
formation enters the nonlinear regime, i. e., around z rv l.
It has been shown that inhomogeneities likely to solve these problems must be of
the subhorizon and strong type, which cannot be studied with perturbation meth-
ods. Averaging and smoothing procedures have been proposed which can provide
some insight into the issue for very simple cases but which much be used with care
since they are not devoided of pitfalls and incompleteness.
Exact solutions of Einstein's equations have the nice property of being able to
modelize both strong and weak inhomogeneities. What these models must reproduce
is not an accelerated expansion, which is an artefact of the homogeneous assumption,
but the observed dimming of the SN Ia luminosity, i. e., the LDR relation. Some
classes have been nicely fitted to this relation at low redshifts. However, the proposed
toy models do not pretend to be fair representations of our neighbouring patch of
 1845

universe. Actually, some of them exhibit controversed properties issued from their
spherical symmetry (LTB models) or their matter content (Stephani models). But
since very few exact solutions to Einstein's equations can be of use in a cosmological
framework, these simple models are usefull to obtain some insight into the issue.
Even if some of them have been shown to match other cosmological observational
constraints, the main challenge remains to fit the whole set of available cosmological
data. A project to begin implementing this issue is currently underway.67 A solution,
if any, would be to use non pathological exact inhomogeneous solutions, reproducing
the nearby Universe, coupled to or asymptotically matching nearly homogeneous
ones valid up to last scattering.

Acknowledgements
I wish to thank Brandon Carter and Julien Larena for a usefull discussion about
averaging procedures. I also want to thank the organizers of the 11th Marcel Gross-
mann Meeting and specially Andrzej Krasinski to have invited me to present this
contribution.

References
1. P. J. E. Peebles & B. Ratra, Rev. Mod. Phys. 75, 559 (2003).
2. M. N. Celerier, A & A 353, 63 (2000).
3. M. N. Celerier, in Proceedings of the 22nd Texas Symposium on Relativistic Astro-
physics, Stanford University, 2004, P. Chen, et ai, eds, eConf C041213, 1403 (2005).
4. S. M. Carroll, V. Duvvuri, M. Trodden & M. Turner, Phys. Rev. D 70,043528 (2004);
S. Capozziello, "Dark Energy and Dark Matter as Curvature Effects", this issue (2006).
5. M. P. Dabrowski & M. A. Hendry, ApJ 498, 67 (1998).
6. J. F. Pascual-Sanchez, Mod. Phys. Lett. A 14, 1539 (1999).
7. K. Tomita, ApJ 529, 38 (2000).
8. L. Nottale, Fractal Space- Time and Microphysics: Towards a Theory of Scale Relativ-
ity, World Scientific, Singapore (1993).
9. L. Nottale, Chaos, Solitons & Fractals 7, 877 (1996).
10. T. Biswas, R. Mansouri & A. Notari, arXiv: astro-ph/0606703.
11. V. F. Mukhanov, H. A. Feldman & R. H. Brandenberger, Phys. Rep. 215, 203 (1992).
12. W. Unruh, "Cosmological long wavelength perturbations", arXiv: astro-ph/9802323.
13. G. Geshnizjani & R. Brandenberger, Phys. Rev. D 66, 123507 (2002).
14. E. W. Kolb, S. Matarrese, A. Notari & A. Riotto, Phys. Rev. D 71,023524 (2005).
15. E. Barausse, S. Matarrese & A. Riotto, Phys. Rev. D 71,063537 (2005).
16. C. Bonvin, R. Durrer & M. A. Gasparini, Phys. Rev. D 73,023523 (2006).
17. A. Notari, arXiv: astro-ph/0503715 (2005).
18. D. Langlois & F. Vernizzi, Phys. Rev. Lett. 95, 091303 (2005); Phys. Rev. D 72,
103501 (2005); JCAP 02, 014 (2006).
19. A. Krasinski, Inhomogeneous Cosmological Models, Cambridge University Press
(1997).
20. see, e.g., F. Sylos Labini, M. Montuori & L. Pietronero, Phys. Rep. 293, 61 (1998).
21. S. Riisiinen, JCAP 0411, 010 (2004).
22. G. F. R. Ellis & W. Stoeger, Class. Quant. Grav. 4, 1697 (1997).
23. T. Buchert & J. Ehlers, A & A 320, 1 (1997).
1846

24. D. Palle, Nuovo Cim. 117 B, 687 (2002).

25. T. Buchert, Gen. ReI. Grav. 32, 105 (2000).
26. T. Buchert, Gen. ReI. Grav. 33, 1381 (2001).
27. T. Buchert & M. Carfora, Class. Quant. Grav. 19, 6109 (2002).
28. T. Buchert & M. Carfora, Phys. Rev. Lett. 90, 031101 (2003).
29. T. Buchert & M. Carfora, in Proceedings of the 12th JGRG Conference, Tokyo, 2002,
M. Shibata, et ai, eds, (2003) p. 157.
30. P. S. Apostolopoulos, N. Brouzakis, N. Tetradis & E. Tzavara, JCAP 06, 009 (2006).
31. K. Bolejko, arXiv: astro-phj0512103 (2005).
32. R. A. Vanderveld, E. E. Flanagan & 1. Wasserman, Phys. Rev. D 74,023506 (2006).
33. W. Godlowski, J. Stelmach & M. Szydlowski, Class. Quant. Grav. 21, 3953 (2004).
34. H. Alnes, M. Amarzguioui & O. Gron, Phys. Rev. D 73, 083519 (2006).
35. J. W. Moffat, JCAP 0510, 012 (2005).
36. D. Lorenz-Petzold, Astrophys. Astron. 7, 155 (1986).
37. R. K. Barrett & C. A. Clarkson, Class. Quant. Grav. 17, 5047 (2000).
38. J. W. Moffat, arXiv: astro-ph/0606124 (2006).
39. K. Tomita, MNRAS 326, 287 (2001).
40. K. Tomita, ApJ 584, 540 (2003).
41. S. Rasanen, arXiv: astro-phj0605632 (2006).
42. C. M. Hirata & U. Seljak, Phys. Rev. D 72083501 (2005).
43. E. E. Flanagan, Phys. Rev. D 71, 103521 (2005).
44. J. W. Moffat, JCAP 0605, 001 (2006).
45. J. W. Moffat, to be published in Proceedings of the Albert Einstein Century Interna-
tional Conference, UNESCO, Paris, July 2005; arXiv: astro-phj0603777.
46. T. Kai, H. Kozaki, K. Nakao, et ai, arXiv: gr-qcj0605120 (2006).
47. A. Ishibashi & R. M. Wald, Class. Quant. Grav. 23, 235 (2006).
48. S. Rasanen, JCAP 0402, 003 (2004).
49. G. Bene, V. Czinner & M. Vasuth, Mod. Phys. Lett. A 21, 1117 (2006).
50. P. Martineau & R. Brandenberger, arXiv: astro-phj0510523 (2005).
51. H. Iguchi, T. Nakamura & K. Nakao, Prog. Theor. Phys. 108, 809 (2002).
52. D. Garfinkle, Class. Quant. Grav. 23,4811 (2006).
53. J. Stelmach & 1. Jakacka, Class. Quant. Grav. 18, 2643 (2001).
54. E. W. Kolb, S. Matarrese, A. Notari & A. Riotto, arXiv: hep-thj0503117 (2005).
55. G. Geshnizjani, D. J. H. Chung & N. Afshordi, Phys. Rev. D 72, 023517 (2005).
56. T. Buchert, Class. Quant. Grav. 22, L113 (2005).
57. S. Riisanen, Class. Quant. Grav. 23, 1823 (2006).
58. E. W. Kolb, S. Matarrese, A. Notari & A. Riotto, Mod. Phys. Lett. A 20, 2705 (2005).
59. E. W. Kolb, S. Matarrese & A. Riotto, arXiv: astro-phj0506534 (2005).
60. T. Buchert, Class. Quant. Grav. 23, 817 (2006).
61. C. Hellaby, Gen. ReI. Grav. 20, 1203 (1988).
62. T. Buchert, J. Larena & J. M. Alimi, Class. Quant. Grav. 23, 6379 (2006).
63. A. Paranjape & T. P. Singh, Class. Quant. Grav. 23, 1 (2006).
64. R. Mansouri, arXiv: astro-phj0512605 (2005); astro-phj0601699 (2006).
65. 1. Iwata, K. Ohta, K. Nakanishi, et ai, in Nearby Large-Scale Structures and the Zone
of Avoidance, ASP Conference Series, Vol. CS-329, A. P. Fairall & P. A. Woudt, eds,
(2004).
66. N. Mustapha, B. A. C. C. Bassett, C. Hellaby & G. F. R. Ellis, Class. Quant. Grav.
15, 2363 (1998).
67. Lu & C. Hellaby, in preparation (2006).
 EVOLUTION OF A VOID AND AN ADJACENT GALAXY
SUPERCLUSTER IN THE QUASISPHERICAL SZEKERES MODEL

KRZYSZTOF BOLEJKO
Nicolaus Copernicus Astronomical Center, Polish Academy of Sciences,
ul. Bartycka 18, 00-716 Warsaw, Poland
[email protected]

This paper investigates the evolution of a void and an adjacent galaxy supercluster.
For this purpose the quasispherical Szekeres model is employed. The Szekeres model
is an exact solution of the Einstein field equations. In this way investigations of the
evolution of the cosmic structures presented here can be freed from such assumptions
as small amplitude of the density contrast. Studying the evolution of a void and an
adjacent supercluster results not only in better understanding the evolution of cosmic
structures but also in acquainting us with the Szekeres model. The main results include
the conclusion that small voids surrounded by large overdensities evolve slower than
large, isolated voids do. On the other hand, large voids enhance the evolution of adjacent
superclusters which evolve much faster than isolated galaxy superclusters.

Keywords: Cosmology; Structure formation; Szekeres model

1. Introduction

The structures which can be observed in the local Universe include small voids
among compact clusters, super clusters and large voids surrounded by large walls
or long filaments. The present day density contrast of overdense regions is larger
than 11 and inside voids it descends to _1. 2 To describe these structures, an exact
solution of the Einstein field equations must be employed. However, among the
known exact solutions none is flexible enough to describe such complicated structure
as our Universe. Therefore, the analysis of this paper will focus on smaller scales.
The structures on small scales, up to Mpc can be described by the Szekeres model
which is an exact solution of the Einstein field equations.
The structure of this paper is as follows: Sec. 2 presents the Szekeres model; Sec.
3 presents the evolution of pairs void-supercluster in the quasispherical Szekeres
model; Sec. 4 presents the role of expansion in the process of structure formation.

2. The Szekeres model

For our purpose it is convenient to use a coordinate system different from that in
which Szekeres 3 originally found his solution. The metric is of the following form: 4

([>1 - ([> fY dr2 _ ([>2 (dp2 + dq2) (1)

(c:-k) E2
where I == a/or, c: = ±1, 0 and k = k(r) ~ c: is an arbitrary function of r.
The function E is given by:
1 2 2 P Q ()
E(r,p,q) = 2S(P +q) - SP- sq+C, 2

1847
1848

where the functions S = S(r), P = P(r), Q Q(r), and C = C(r) satisfy the
relation:
p2 Q2 S
C = 2S + 2S + "iE, E = 0,±1, (3)
but are otherwise arbitrary.
The quasispherical Szekeres model is the case of E = 1. As wa.-; shown in Ref. 5,
surface of t = const and r = const has a topology of a sphere. However, as S, P
and Q are now functions of r, the spheres are not concentric. For the spheres to
be concentric, the functions S, P and Q must be constant. Such conditions entail
spherical symmetry, with which the Szekeres model becomes the the Lemaitre-
Tolman modelY·lo
Substituting the metric (1) in the Einstein equations, and assuming the energy
momentum tensor for a dust, the Einstein equations reduce to the following two:

1 .2 2M(r) 1 2
-<I>
c2
(t r) = - - - k(r)
'<I>(t,r)
+3
-A<I> (t r).
' .
(4)

4 G ( )_ M'(r) -3M(r)E'(r,p,q)/E(r,p,q)
(5)
7r c2P t, r,p, q - <I>2(t, r)[<I>'(t, r) - <I>(t, r)E'(r,p, q)/ E(r,p, q)]'

Equation (4) can be integrated:

<P _

JJ2A~(r)
o -
d<I>
k(r) + iA<I>2
= c [t - tB(T)]. (6)

As can be seen the Szekeres model is specified by 6 functions:

M(r),k(r),tB(T),S(r),Q(r),P(r). However, by a choice of the coordinates, the
number of independent functions can be reduced to .5.
The equations of motion TOi(3;(3 = ° are reduced to the continuity equation:

p+ p8 = 0, (7)
where 8 is the scalar of expansion and is equal to:

_ 3 <I:>(t, r) <I:>'(t, r) - <I:>(t, r)<I>'(t, r)/<I>(t, T)

8( t,r,p,q ) - -(-)
<I> t, T
+ <I>'(t, r) )
- <I>(t, r E'(r,p, q)/ E(r,p, q)
. (8)
In the expanding Universe 8 is positive so the density decreases. The structures
which exist in the Universe, emerged either due to slower expansion of the space
(formation of overdense regions) or due to faster expansion (formation of underdense
regions). In the Friedmann limit R ~ m, where a is the scale factor and 8 ~ 3Ho.
The Szekeres model is known to have no symmetry.6 It is of great flexibility and
wide application in cosmology7 and in astrophysics,4,5 and still it can be used as a
model of many astronomical phenomena. In this paper it will be employed to study
the evolution of cosmic structures in different environments.
 1849

2.1. Density contrast

To compare the evolution of different models the change in their density contrast
will be considered. Two different types of density contrast indicators are taken into
account. The first one is the usual density contrast, 6 = pi Pb - 1, the second one is
the spatially invariant density contrast: 8

SIK
r I
h e¥.f3 ap ap
= IE ----;7 axe¥. ax f3
IK dV, (9)

where I E JR., and K E JR.\ {O}. This family of the density contrast indicators can be
considered as local or global depending on the size of ~. Such a quantity not only
describes the change of density but also the change of gradients and the volume of
a perturbed region. So this density indicator describes the evolution of the whole
region in a more sophisticated way than the 6. Here only the case I = 2, K = 1/2
will be considered.

2.2. Model set-up

To specify the model 5 functions of the radial coordinate need to be known. Let us
define the radial coordinate as a value of <I> at the initial instant to = 0.5 My after
the big bang, i.e., r := <I>(r, to).
Two of these functions will be t8(r) and Nf(r). Let us write the mass function
in the following form:

M(r) = Mo(r) + 6M(r), (10)

where AIo is the mass distribution as in the homogeneous universe, and 6Nf is a mass
correction, which can be either positive or negative. The 6NI is defined similarly as
in the spherically symmetric case:

6M(r) = 47r2
C
c
1" 0
drR(r, to)2 R'(r, to)c5j5(r) , (11)

where 6j5(r) is an arbitrary function chosen to specify the 6M. Although c5j5(r) is not
the initial function of density fluctuations (since an initial density fluctuation is a
function of all coordinates) it gives some estimation on the initial density fluctuation
of the monopole density component.
The next three functions are P(r), Q(r), S(r). All functions defining the model
are presented in Table 1. The numerical algorithm used to solve the Szekeres model's
equations is presented in detail in Ref. 11.
The chosen background model is the homogeneous Friedmann model with the
density:
3H2
Pb = Dm x Per = 0.24 x CO , (12)
87r
1850

where the Hubble constant is Ho = 74 km S-l Mpc- 1 . The cosmological constant,

A, corresponds to OA = 0.76, where OA = (1/3)(c 2 A/HJ).

3. Models of a void and an adjacent galaxy supercluster

In this section the evolution a void with an adjourning galaxy supercluster is in-
vestigated. Although within the Szekeres model more than two structures can be
described, such investigations of less complex cases are useful because they enable
us .to draw some general conclusions without going into too much detail, which could
easily obscure the larger picture.

3.1. Models with p' = 0 = S', Q' =F 0

As mentioned above, if P' = 0 = 5' = Q' the Szekeres model becomes the Lemaitre-
Tolman model. Hence, the class of models considered in this subsection is the sim-
plest generalisation of the spherically symmetric models.
The double structure of a void and adjourning supercluster can be described in
the Szekeres model in two different ways. The first alternative is when 8M < 0, the
second when 8M > O. Both these possibilities are examined here.

3.1.1. Model specification

The exact form of the functions used to define models 1 and 2 is presented in Table
1. The density distributions of models 1 and 2 are presented in Fig. 1. As can be seen
the model with 8M < 0 has the void in the center, and the adjacent supercluster
has an elongated shape. It is the opposite in model 2. The overdense region at the
origin is more compact than in modell, and the adjacent element is the void.

Table 1. The exact form of the functions used to define models 1-4.

Model tB 5p S p Q
1 0 -5A x exp[-(C/8)2] 1 0 b In(l + C) x exp( -3AC)
2 0 1.14A x exp[-(€/9)2] 1 0 cln(l + 0.2C) x exp( -3AC)
3 0 -5A x exp[-(C/8)2] _Co. 4 0.55Co.4 0.33Co.4
4 0 1.14A x exp[-(C/9)2] _CO. 9 0.551'°·8 0.33Co.8

Note: C = r /Kpc, A = 10- 3 , b = -0.6, c = 1.45,.

3.1.2. Evolution
In this section we compare the evolution of the density contrast, 8(t, r, B, 4;), and
the 5 2,1/2(t,2:) density indicator for models 1, 2, with the corresponding mod-
els of a single void and the models of a single supercluster obtained within the
Lemaitre-Tolman model. The Lemaitre-Tolman model is considered because within
 1851

density
density

3.5 4
3 3.5
2.5
2 2.5
1.5
1.5
0.5
0 0.5
0

(a) (b)
Fig.1. The present-day density distribution, p/Pb. Fig. l(a) presents model 1 (8M < 0). Fig.
l(b) presents model 2 (8M> 0).

this model one can descrihe a single spherically symmetric structure. Such a com-
parison can demonstrate how the evolution of a structure changes if there is another
structure in its close neighbourhood.
Fig. 2 presents the evolution of the density contrast of model 1 in comparison
with the corresponding models obtained within the Lemaitre-Tolman model. The
Lemaitre-Tolman model was specified by assuming the same condition as the ones in
the Szekeres model at the initial instant. The local density contrast, 8, is compared
at the point of the maximal and minimal density value. Fig. 2(a) presents the
evolution of the density contrast inside the void. As can be seen the bahaviour
of the density contrast in both models is similar. This due to the conditions of
regularity at the origin (for a detailed description of the regularity conditions at
the origin see Ref. 4). These conditions imply that the origin in the quasispherical
Szekeres and Lemaitre-Tolman model behaves like a Friedmann models Fig. 2(b)
presents the evolution of the density contrast at the very center of the overdense
region of the model 1 and the corresponding Lemaitre-Tolman model. The growth
of density contrast in the Szekeres model is much faster than in the corresponding
Lemaitre-Tolman model. The results of this comparison indicate that within the
perturbed region of mass below the background mass (8M < 0) the evolution of
underdensities does not change but the evolution of the overdense regions situated
at the edge of the underdense regions is much faster than the similar evolution of
isolated structures.
The evolution of the density contrast of model 2 (8M> 0) is presented in Fig.
3, the evolution of the density contrast at the point of minimal density is depicted
in Fig. 3(a), and the evolution at the origin is depicted in Fig. 3(b). Similarly as
in modell, the evolution at the origin in the Szekeres model and in the Lemaitre-
Tolman model are very much alike. The evolution of the void, however, is slower
1852

within the Szekeres model than it is in the Lemaitre-Tolman model. This implies
that single, isolated voids evolve much faster than the ones which are in the neigh-
borhood of large overdensities where the mass of the perturbed region is above the
background mass (rSM > 0).

3.5
·0.2
sz
1;; w
l"
~ ·0.4 C
2.5
8 8 2
·f
w
·0.6 ·fw 1.5
"0 "0

LT
·0.8 sz
LT 0.5
-~------

·1 0
10 12 14 16 0 4 10 12 14 16
9
time [My] time [10 y]
(a) (b)
Fig. 2. The evolution of the density contrast inside the void (a), and inside the supercluster (b)
for model 1 (b1>1 < 0). The curve SZ presents the evolution within the Szekeres model; curve LT
presents the evolution within the Lemaitre-Tolman model.

3.5
·0.2
sz
~
w LT
C ·0.4 ~ 2.5
8 8
""
'g?
w
"0
·0.6 1.5

·0.8
- - - -_ _ sz 0.5
O"---.~-~-~_~~_~_~---...J

o 2 4 10 12 14 16 o 4 6 10 12 14 16
9 9
time [10 y] time [10 y]
(a) (b)
Fig. 3. The evolution of the density contrast inside the void (a), and inside the supercluster (b)
for model 2 (bM > 0). The curve SZ presents the evolution within the Szekeres model; curve LT
presents the evolution within the Lemaitre-Tolman model.

3.2. Models with P' =f. 0 =f. 8', Q' =f. 0

In this section models of non-constant P, Q and S are investigated. The evolution
of these models is compared with the evolution of models which were considered in
Sec. 3.l.
 1853

3.2.1. Models specification

The exact form of the functions used to define models 3 and 4 is presented in
Table 1. Fig. 4 presents the comparison of the present day density distribution in
models 1 and 3 in colour coded diagrams. It presents the vertical cross-sections of
the considered structures. Fig. 4(a) presents the vertical cross-section through the
surface of ¢ 1T /2 and Fig. 4(b) presents the cross section through the surface
of ¢ ~ 1T / 6. The comprehensive study of the vertical and horizontal cross--sections
of similar models was presented in Ref. 12. Fig. 5 also presents the vertical cross-
sections of models 2 and 4. As can be seen, both structures appear to be similar
but, in comparison with modell, in model 3 the additional component is moved
down and right. Model 4 on the other hand presents the structure moved down and
right in comparison with model 2.

3.5

3
20 20

2.5
10 10
2
0'
a.
~
N
0 I
N
0
1.5
-10 -10

-20 -20
0.5

-30 ,-_~~~~_-,-_ ......._ - l -30 0

-30 -20 -10 0 10 20 30 -30 -20 -10 0 10 20 30
Y[Mpc] R[Mpc]

(a) (b)

Fig. 4. The present--day colour coded density distribution, pi Pb, of models with 8M < O. (a)
presents the model with pi S' O. (b) presents the model with pi of 0 of S'.

3.2.2. Evolution
The evolutions of the density contrast inside the voids and super clusters of models
3 and 1 are very similar which needn't be suprising as model 3 has the same 8(r)
as model 1. Also, the evolutions of the corresponding density contrasts of model 4
and 2 are similar. The functions S, P, Q were chosen so they reproduce the same
shape of current structures and the same density contrast inside them - that is
why the evolution of a local density contrast is comparable for models 1 and 3,
and for models 2 and 4. However, it is not clear whether or not the evolution of
S2,1/2 is comparable too. When the functions S, P, Q are not constant, the axis of
a density dipole changes. Also, the volume of the perturbed region as well as the
density gradients can be different. So it may be interesting to compare the evolution
1854

30 ~--T--- __--__ --~--~--~

30 3,5

20 3
20

2,5
10 10

'0' 2
C-
~ O
N 1.5
-10 -10

-20 -20 0,5

-30 '---~--'-----'-----' -30 '---~--'-----'----""" o

-30 -20 -10 0 10 20 30 -30 -20 -10 a 10 20 30
Y[Mpc] R[Mpc]

(a) (b)
Fig, 5, The present-day colour coded density distribution, pi Pb, of models with 8M > 0, (a)
model with P' = 8' O. (b) model with p' 0 =I- 8'.

of the whole perturbed under dense and overdense regions of models 1, 2, 3, and 4.
Fig. 6 presents the comparison of evolution of 5 2 ,1/2 for models 1-4. The primed
letters denote models of 51 =I- 0 =I- pi, QI =I- O. As can be seen the evolution of 5 2 ,1/2
for all these models is also comparable. These results imply that the evolution in
the quasispherical Szekeres model does not depend on the position of the dipole
component. As long as the shape and density contrast of the analysed models are
similar, such models evolve in a very similar way.

1,4 3

1,2 2,5

2
0,8
~ 1,5
(flN
0,6

0.4

0.2 0.5

0 0
0 2 4 6 8 10 12 14 16 0 2 4 6 8 10 12 14 16
9 time [109 y]
time [10 y]
(a) (b)
Fig. 6. Comparison of 8 2 ,1/2 for models with oM < 0 (a) and with 8M < 0 (b). C corresponds
to a "supercluster" an overdense region, and V corresponds to a "void" underdense region.
Primes denote models with P' =I- 0 =I- 8'. Since the value of 8IK depends on units, the results
presented here were normalised so they are now of order of unity.
 1855

4. The role of expansion

The faster or slower evolution rate of the previously presented models is reflected
by their current expansion rate. As has been shown above, the evolution does not
depend on a relative position of the dipole component (evolution of models 1 and 3
is similar). Thus, let us focus on model 1 and model 2 only.
Fig. 7 presents the ratio, 8sz/8 o, of the expansion parameter in the considered
Szekeres models to the expansion parameter in the homogeneous background. As can
be seen, modell, with 8M < 0, has a larger amplitude ofthis ratio, and the evolution
of a supercluster in this model is much faster than in the corresponding Lemaitre-
Tolman model. On the other hand, model 2 (8M> 0) has smaller amplitude of
the 8sz/8 o ratio and within model 2 the evolution of the density contrast inside
the void was much slower than in the Lemaitre-Tolman model. So clearly the rate
of the evolution is connected with the rate of the expansion. This conclusion is
also supported by the continuity equation [Eq. (7)]. Higher mass in the perturbed
region slows down the expansion rate this is a condition hindering the evolution
of cosmic voids. On the other hand, if the mass of perturbed region is below the
background mass, such region expands much faster than the background, leading
to the formation of the large underdense regions. Such large voids enhance the
formation of large elongated overdensities (walls) formed at the edges of voids.

1.2

30 30

20 20

10 10
g: g:
~ 0 0.9
~ 0
>- >-
-10 -10
0.8

-20 -20
OJ
-30 -30
-40 1--.............._ .......--"_"'---'-_ _...1
-40 1--__--'~-'-~-----...1 0.8
-40 -30 -20 -10 0 10 20 30 40 -40 -30 -20 -10 0 10 20 30 40
X [Mpc] X [Mpc]

(a) (b)
Fig. 7. The 8sz/80 ratio. (a) presents the ratio of model 1. (b) presents the ratio of model 2.

5. Conclusions
The galaxy redshift surveys show that the Universe is patchy with various structures.
These structures include small voids among compact superclusters and large voids
surrounded by large walls or long filaments.
The evolution of these cosmic structures in different environments in the quasi-
1856

spherical Szekeres model was investigated. The Szekeres model is the most complex
spatially inhomogeneous exact solutions of the Einstein field equations, and it is
of great use in cosmology. Since it is an exact solution of Einstein's equations, it
enables us to investigate the evolution of cosmic structures without such approxi-
mations as linearity and small amplitude of density contrast. Moreover, the Szekeres
model is flexible enough to describe more than one structure.
Having investigated various models with two or three structures within one
frame it may be concluded that the evolution of the cosmic structures depends
on the environment. In perturbed a region whose mass is below the background
mass the amplitude of the expansion's fluctuations is large and as can be seen from
the continuity equation [Eq. (7)], such conditions enhance the evolution of cosmic
structures.
The analyses presented in this paper indicate that small voids among large
overdense regions do not evolve as fast as the large voids do. This is because the ex-
pansion of the space is faster inside large voids than inside smaller voids. Moreover,
this higher expansion rate inside the large voids leads to the formation of large and
elongated structures such as walls and filaments which emerge at the edges of these
large voids.

Acknowledgments
I would like to thank Andrzej Krasinski and Charles Hellaby for their valuable
comments and discusions concerning the Szekeres model.

References
1. E. Branchini, et al., Mon. Not. R. Astron. Soc. 308, 1 (1999).
2. F. Hoyle and M. S. Vogeley, Astrophys. J. 607, 751 (2004).
3. P. Szekeres, Commun. Math. Phys. 41, 55 (1975).
4. C. Hellaby and A. Krasinski, Phys.Rev. D 66, 084011 (2002).
5. P. Szekeres, Phys. Rev. D 12, 2941 (1975).
6. W. B. Bonnor, A. H. Sulaiman, and N. Tomimura, Gen. Relativ. Gravit. 8,549 (1977).
7. W. B. Bonnor and N. Tomimura, Mon. Not. R. Astron. Soc. 175, 85 (1976).
8. F. Mena and R. Tavakol, Class. Quant. Grav. 16, 435 (1999).
9. G. Lemaitre, Ann. Soc. Sci. Bruxelles A53, 51 (1933); reprinted in Gen. Relativ.
Gravit. 29, 641 (1997).
10. R. C. Tolman, Proc. Nat. Acad. Sci. USA 20, 169 (1934); reprinted in Gen. Relativ.
Gravit. 29, 935 (1997).
11. K. Bolejko K., in 13th Young Scientists' Conference on Astronomy and Space Physic,
eds. A. Golovin, G. Ivashchenko and A. Simon, (Kiev University Press, 2006); astro-
ph/0607130.
12. K. Bolejko K., Phys. Rev. D 73, 123508 (2006).
 COVARIANT DESCRIPTION OF THE INHOMOGENEOUS
MIXMASTER CHAOS

R. BENINI12t and G. MONTANI 23 :J:

1 Dipartimento
di Fisica - UniversitiI di Bologna and INFN
Sezione di Bologna, via Irnerio 46, 40126 Bologna, Italy
2 ICRA -International Center for Relativistic Astrophysics c/o Dipartimento di Fisica (Gg)
UniversitiI di Roma "La Sapienza", Piazza A.Moro 5 00185 Roma, Italy
3 ENEA C.R. Fmscati (U. T.S. Fusione),
via Enrico Fermi 45, 00044 Fmscati, Rome, Italy
t riccardo. [email protected]
'+ [email protected]

We outline the covariant nature of the chaos characterizing the generic cosmological
solution near the initial singularity. Our analysis is based on a "gauge" independent
ADM-reduction of the dynamics to the physical degrees of freedom, and shows that the
dynamics is isomorphic point by point in space to a billiard on a Lobachevsky plane. The
Jacobi metric associated to the geodesic flow is constructed and a non-zero Lyapunov
exponent is explicitly calculated. The chaos covariance emerges from the independence
of the form of the lapse function and the shift vector.
In recent years, the chaoticity of the homogeneous Mixmaster model has been
widely studied in the literature (see l - 4) in view of understanding the features of its
covariant nature. Two convincing arguments, appeared in,1,4 support the idea that
the Mixmaster chaos (described by the invariant measure introduced in 3 ,5) remains
valid in any system of coordinates.
The main issue of the present work is to show that the property of space-time co-
variance can be extended to the inhomogeneous Mixmaster model.
A generic cosmological solution is represented by a gravitational field having avail-
able all its degrees of freedom and, therefore, allowing to specify a generic Cauchy
problem. In the Arnowitt-Deser-Misner (ADM) formalism, the metric tensor corre-
sponding to such a generic model takes the form
(1)
where Nand Na denote (respectively) the lapse function and the shift-vector, --Ya(3
the 3-metric tensor of the spatial hyper-surfaces r;3 for which t = canst, being 6

--Yex(3 = e qa uad
£ oaod!Cl b"
b cUaY U(3Y
c a, b, c, d, 0, f3 = 1,2,3; (2)
qa = qa(x, t) and yb = yb(x, t) are six scalar functions, and 0b = Obex) a 80(3)
matrix. By the metric tensor (2), the action for the gravitational field is

8 = r
j"E(3) x Di
dtd 3 x (PaOtqa + IIdOtyd - N H - N ex Hex) , (3)

(4)

(5)

1857
1858

in (4) and (5) Pa and IId are the conjugate momenta to the variables qa and yb re-
spectively, and the (3) R is the Ricci 3-scalar which plays the role of a potential term.
We use the Hamiltonian constraints H = Ha = 0 to reduce the dynamics to the
physical degrees of freedom; the super-momentum constraints can be diagonalized
and explicitly solved by choosing the function ya as special coordinates:

(6)

(7)

The potential term appearing in the super-Hamiltonian behaves as an infinite po-

tential wall as the determinant of the 3-metric goes to zero, and we can model it as
follows

+oo if < 0,
X
8(x) = (8)
a
{0 if x> o.

where the Qa's are called anisotropy parameters. By (8) the Universe dynamics
evolves independently in each space point and the point-Universe can move within
a dynamically-closed domain rQ only6 (see figure (1)). Since in rQ the potential U
asymptotically vanishes, near the singularity we have oPal aT) = 0; then the term
2Pa( 0- 1)~OTJO~ in (7) behaves as an exact time-derivative that can be ruled out of
the variational principle.
The ADM reduction is completed by introducing the so-called Misner-Chitre like
(T,~, 8) variables,1 in terms of which the anisotropy parameters Q a become T in-
dependent: When expressed in terms of such variables the super-Hamiltonian con-
straint can be solved in the domain r Q according to the ADM prescription:

_-----------i,"

Fig. 1. rQ(~,B)
 1859

1 ~
Ql = 3" - 3~ (cosB + V3sinB)

1 ~
Q2 = 3" - 3~ (cosB - V3sinB) (9)

1 2~
Q3 = 3" + 3~ cos B

(10)

(11 )

By virtue of the asymptotic limit (8) and the Hamilton equations associated with
(11) it follows that E is a constant of motion, i.e. dE/dry = OE/Ory = 0 '*
E = E(ya);

furthermore EOTJT behaves as an exact time derivative.

This dynamical scheme allows us to construct the Jacobi metric corresponding to
the dynamical flow, and the line element reads

ds 2 = E 2 (ya) ( de
(t;2 - 1)
+ (e - 1)dB 2
) . (12)

Here the space coordinates behaves like external parameters since the evolution is
spatially uncorrelated. The Ricci scalar takes the value R = -2/ E2, so that (12)
describes a two-dimensional Lobachevsky space; the role of the potential wall (8)
consists of cutting a closed domain r Q on such a negative curved surface.
In this scheme the Lyapunov exponent associated to the geodesic deviation along
the direction orthonormal to the geodesic 4-velocity can be evaluated, and it results
to be equal to

(13)

Hence the chaotic behavior of the inhomogeneous Mixmaster model can be described
in a generic gauge (i. e. without assigning the form of the lapse function and of the
shift vector) as soon as Misner-Chitre like variables are adopted. The value (13) is
a positive definite function of the space point, making this calculation extendible
point by point to the whole space.

References
1. G. Imponente and G. Montani, Phys. Rev. D 63 (2001) 103501
2. G. Montani, Class. Quant. Grav. 18 (2001) 193
3. A. A. Kirillov and G. Montani, Phys. Rev. D 56 (1997) 6225.
4. N. J. Cornish and J. J. Levin, Phys. Rev. Lett. 78, 998 (1997)
5. D.F. Chernoff and J.D. Barrow, Phys. Rev. Lett, 50, 134,(1983).
6. R. Benini, G. Montani and R. Benini, Phys. Rev. D 70 (2004) 103527
 THE MASS AND GEOMETRY OF THE COSMOS

CHARLES HELLABY
Department of Mathematics and Applied Mathematics, University of Cape Town
[email protected]

HUI-CHING LU
Department of Mathematics and Applied Mathematics, University of Cape Town
[email protected]

As the cosmological data generated by redshift surveys becomes increasingly accurate,

the proper reduction and interpretation of the high redshift data will require knowledge
of the cosmic geometry that is traveled through by the light rays we observe. It will no
longer be necessary to assume homogeneity, the data will make it possible to quantify
the level of homogeneity on different scales. The ultimate application of Einsteins field
equations is to determine the relation between matter and geometry in the real universe.
A set of observations of the redshifts, angular diameters, and apparent luminosities of
galaxies, as well as their number counts, combined with knowledge of their true diameters,
luminosities and masses, plus the cosmic equation of state, can be turned into metric
information. Though much theoretical development has been done, the proposed methods
have never been implemented, so the 2 key issues of choosing appropriate numerical
methods and handling real observational data have not been addressed. A preliminary
numerical reduction scheme has been written and tested. The methods and difficulties
encountered will be discussed. The locus where the past null cone crosses the apparent
horizon has significant theoretical, numerical and observational properties. It allows us
to determine a characteristic mass of the cosmos that is quite model independent.

1. Obtaining the Metric of the Cosmos from Observations

1.1. Introduction
Large-scale automated redshift surveys, such as SDSS, 2dF, 6dF, etc, are opening
up new possibilities for measuring the content and dynamics of the universe.
Given sufficiently accurate and complete galaxy data on our past null cone
(PNC), viz redshifts, apparent luminosities and/or angular diameters, and number
counts in redshift space, and given the source properties, viz absolute luminosities
and/or true diameters, and source masses, plus an assumption about the cosmic
equation of state, we can determine the metric of our cosmos - the geometry of
the spacetime we live in.
The primary measures of cosmic distance are the diamter distance
D true diameter
dD = R= £"
u angular diameter
and the luminosity distance

dL = /!j dlO =
absolute luminosity
apparent luminosity
x 10 pc

of sources. The density in redshift space is

mn = mass per source x number density in z space

1860
 1861

where the number density per steradian per unit redshift interval, n, is dimension-
less. Note that the observables are 6, €, n, and that each one is multiplied or divided
by a source property D, L, m. Since galaxy properties change with time, these source
evolution functions must be known, for example from a theory of galaxy evolution
supported by independent observations.

1.2. Observer's Past Null Cone

time
Us, now

Our
past
null Galaxy
cone worldlines
Our
worldline
Several things affect a source's appearance; if far away objects look different, is
it because of
• Cosmic evolution (EFEs): the equation of state determines the scale factor evo-
lution which determines redshift down the null cone, p(p) ~ S(t) ~ z(t);
• Source evolution: the same type of object looks different long ago (at large z)
than recently (small z), L = L(t), D = D(t), m = m(t) ~ L = L(z), D =
D(z), m = m(z);. or Inhomogeneity: spatial variation makes source properties
different far away (large z) and nearby (small z), L = L(r), D = D(r), m =
m(r) ~ p = p(t, r) ~ p(z)?

1.3. How Homogeneous?

The conventional wisdom is that the Copernican principle ensures homogeneity
on very large scales, but this scale is still not well defined, and the homogeneous
Robertson-Walker (RW) model is in fact assumed. This assumption has been essen-
tial up to now, and has served us well. But now we're getting all this data, we won't
have to assume it much longer, and we should rather try to prove (or disprove)
homogeneity. Indeed we should quantify it - how close to homogeneity are we on
a variety of scales?
To do this, we must analyse observational data without assuming homogeneity,
and here lies a significant problem because so much of the current analysis does
assume it. There's a danger of a circular argument. Isotropy about the obsever
1862

can be established with no assumptions about the correct cosmological model. Ra-
dial variation due to (possible) inhomogeneity is hard to separate from the other
cosmological effects.
The two key papers that considered the problem of obtaining geometry from
cosmological observations are "Observations in Cosmology" by Kristian and Sachs 1
and "Ideal Observational Cosmology" by Ellis, Nel, Maartens, Stoeger and Whit-
man. 2 Multicolour observations as a way to check source evolution were considered
in. 3 For a list of references and some details of earlier work, see. 4 The algorithm we
shall be using was given by Mustapha, Hellaby and Ellis.5

1.4. Lemaitre-Tolman model

To get the investigation going, we start with simplest case - the simplest inhomoge-
neous cosmology. The Lemaitre-Tolman model (LT) model is spherically symmetric,
having only radial variation, and it evolves with time. The metric is 6 - 9

where and '=8/8r.

Here R = R( t, r) is the areal radius, which equals the diameter distance, and E =
E(r) ~ -1/2 is an arbitrary function that represents the local geometry. It has a
dust equation of state, Tab = puau b , and the coordinates are comoving with the
matter, u a = of. From the Einstein field equations we obtain the density and the
differential equation (DE) for the evolution,

2M'
,",p(t,r) = R2R' '
·2 2M AR2
R =-+2E+--
R 3
where .=8/8t
where A is the cosmological constant, and the second arbitrary function M = M(r)
gives the gravitational mass interior to the comoving shell at T. We see that E(r)
also represents local energy per unit mass of worldlines. The solution depends on
E, and for the hyperbolic case E > 0 it is
3/2
.h ) ( 2E ) (t-tB)
(sm Tl- TJ = M '

where tB = tB(r) is the third arbitrary function that gives the local time of big
bang. The parabolic and elliptic solutions are similar.

1.5. Integrating Down the P NC

The observer's past null cone is null and radial, i.e. it obeys

dt R' A,
dr vI
--====-
+ 2E = t .
 1863

We write the solution, the PNC locus, as t = i(r), and use a "hat" to denote
evaluation on the past null cone. Thus the diameter and luminosity distances are
d D = R = R(£(r), r), d L = d D (1 + z)2 and the redshift is

(1 + z) = r
jPNC vI R'+ 2E dr .

The number density in z-space is

From the above we derive DEs for extracting the LT model from the data, i.e. DEs
for the 3 arbitrary functions M(z), E(z), tB(Z):lO

dr=",
dz '+' ,
d¢=
dz ¢
{_1_
(1 + z) +
~+~}
dR
dz
'

+ 2E dR (1 _ 21;:1) ¢
dM
dz
41fmny'1
¢
where vI + 2E = dz
2¢
+ ~
2 dR
dz
The right hand sides of the DEs all contain the unknown functions and the obser-
vational data. In addition, there is an algebraic equation for tB(Z).

t observer's worldline

galaxy worldlines

r r+dr
........r
z+dz

1.6. Four Integration Regions

In practice, there are regions that need different treatments, illustrated in the
schematic R vs z plot below. These regions are:
(a) Origin: There is no actual data at the origin, and very little near it, but the DEs
1864

need initial values at the origin. However, the origin of an LT model is RW-like, so
we use the first few data points to extrapolate back to the origin.
(b) Inner region: The DEs are integrated numerically.
(c) Maximum: Near the maximum in R(z) the DEs go singular, so we do a series
expansion about R max , and join it to the end of the numerical run.
(d) Outer region: The DEs are again integrated numerically.
Each region must be matched properly to the next.
X 10-3
5~----~------~------~-------,------~------,

4.5
Serie

3.5

Near origin - LTis RW-like - extrapolate back from nearby data

o Ori in - no data
o 0.5 1.5 2.5 3

1.7. The Maximum in R(z)

At the maximum we have ~~ = 0 and it is evident that the above DEs become
singular. Actually this problem occurs even if the observations are exactly RW, and
is a generic feature of any integration method. It would probably be worse in non
spherical symmetry. We do a series expansion about R max , which occurs at Zm, and
use subscript m to indicate evaluation on the PNC at Zm:

+ R36z3 + ...
R = Rm - R 2 6z 2
mn = (mn)m + (mnh6z + (mnh6z 2 + ...
M = Mm + M 1 6z + lvh6z 2 + ...
¢ = ¢m + ¢16z + ¢2 6z2 + ...
viI + 2E = W = Wm + W 1 6z + W 2 6z 2 + ...
where 6z = (z - zm), and insert these into the DEs, which, to zero & first order,
give
 1865

We find that all the 1> coefficients are determined, but the M and W coefficients
all depend on either lvh (or ~Vm), which is not determined. However, the numerical
integration has already given us Ma and (dM / dZ)a at some point Za before Zm, i.e.

Ma = Mm + lVh(za ~ zm)
+ M2(Za ~ zm)2 + ...

so we can solve for M 1 , W m , etc.

1.8. Results
To test a numerical procedure, you must use data for which you know the correct
answer. The above numerical procedure was run for a range of fake data, includ-
ing open, nearly-flat, and closed homogeneous models and a variety of mildly and
strongly inhomogeneous models. The graphs below show the accuracy of the results
for a strongly inhomogeneous model, by comparing the numberical output func-
tions with original functions. They show M vs z, W = VI + 2E vs z, and tB vs Z
respectively.
In each case, agreement of the output functions (M, E & t B) with those of
original model was good to excellent. Not surprisingly, the function with the largest
discrepancy was W = VI + 2E . In some cases there was a noticeable discrepancy
near the origin in t B, indicating an imperfect estimate of the origin parameters.
1866

06

05

04
,.
03

0.2

0.1

0
0

13

Nume~cs
1-25
COITe0

12

1"15

;:
11 :· ...............L-'·.'

1.05

09~Lr_·__'·_.. _··_·~______~____~______ - L_ _ _ _ _ _ ~ ~ ____

o 0.5 1.5 2.5

01

-01

Correct
-0.2
.,." •• NUfTIencs
-0.3
- - - Now
-04

-0.5

-06

-07

-OB

-OJ '·.n

o 0.5 1.5 25
 1867

1.9. Conclusions - Metric of Cosmos

We have successfully implemented the MHE algorithm, and demonstrated its via-
bility. It was tested with fake data, for many homogeeous & inhomogeneous models.
The accuracy was quite good, with E having the highest error, as might be expected.
The reduction of actual data is underway - this is for practice only, as much
larger quantities of much more accurate data is needed. Many improvements & ex-
tensions to the method will also be needed. But we are gaining experience with
reducing cosmological data to metric information, in anticipation of a flood of cos-
mological data.
Knowing the nearby metric will assist in analysing distant observations in more
than just a statistical sense. Eventually we'll be able to quantify the degree of
homogeneity, and won't have to assume homogeneity.

2. Measuring a Characteristic Mass of the Cosmos

We next point out a significant characterisation of the cosmos on giga-parsec scales
that will become measurable with the next generation of surveys. 11 The insight here
is in putting together long-known results, not in the maths itself. As is well known,
for objects of a given physical size, there is a minimum in their apparent size, i.e.
there is a maximum in the areal radius of our past null cone (PNC).

2.1. Mass, Radius and Apparent Horizon

The gravitational mass M within comoving radius r, used in calculating light bend-
ing, orbital velocities, etc, is not the same as the integrated density on a constant t
section, i.e. the sum of the rest masses of galaxies, gas clouds, dark matter concen-
trations, etc,
MI
M =

The derivative of R down the PNC,

J
pdV =
J dr-.
Vl+2E

d
-R=Rt+R
A AI I]
[ •
= [( -R +1 ) R I]
dr PNC VI + 2E PNC

in terms of physically measureable quantities is

-dR = -k =[. RI ( VI + 2E - R. ) ] ,
dz Zl RI(1 + z) PNC

For there to be a maximum in R(z), dRjdz must go through zero as R inceases to

the past. The apparent horizon (AH) is the locus of all such points - for all PNCs
in the observer's history:
AR3
2M=R--
3
Thus the maximum in dD is where our PNC crosses the AH.
1868

2.2. Significance of the Maximum in R

This maximum in d D = R is (a) a distinctive feature of an observational plot, (b)
a characteristic of the cosmos we inhabit and the time we observe it, and (c) it
determines the gravitational mass contained within that radius,
~3

2Mm = Rm - ARm
3 where A = 3Ha2 0 A·

There are two roots: the smaller, at Rm , is the AH; the larger, at Rm , is the de
Sitter horizon. Any given worldline (at constant Mm) either encounters both or nei-
ther. This is illustrated below for a family of incoming light rays in a homogeneous
model, with HO = 70, Om = 0.3, OA = 0.7.

Areal radius of past null cone vs t

160

140

120

100

~
~ 80

60

40

20

0
0 3
Rn (Gpc)

The R-z Curves for families of RW models with Om varying, with OA varying,
and with Ha varying can be seen in. l l The curves are most sensitive to variations
in OA & Ha near Zm, the maximum in R, and have close to maximum sensitivity
to Om variations there too. In addition, variations in OA, H a, and Om move the
maximum in very different directions.

2.3. Conclusions - Mass of the Cosmos

• The next generation of surveys should determine Rm and Zrn quite well.
• A determination of Rm and A will fix the cosmic mass Mm on giga-parsec scales,
even in inhomogeneous universe, i.e. independently of whether homogeneity is as-
sumed or not. This is not true for any other point on our PNC.
• For the RW model, determination of Rm & Zm puts constraints on Om, OA, Ha.
 1869

• This result provides an amendment to the MHE theorem - the total mass from
mn and the mass deduced from Rm must mesh correctly here - i.e. the observations
n(z), €(z) & 5(z), plus the evolution functions m(z), L(z) & D(z) plus the A value
must satisfy the AH equation.
• Key features of R-z curve _. the location of the maximum & the initial slope --
Rm, Zm & Ho - define a natural "best fit" or "average" RW model. This definition
is based on geometry, but uses observations.
• A determination of Rm and Zm is likely to be the first observational detection of
a relativistic horizon.

References
1. J. Kristian and RK. Sachs (1966) Astrophys. J. 143, 379-99, "Observations in Cos-
mology" .
2. G.F.R Ellis, S.D. Nel, R Maartens, W.R Stoeger, & A.P. Whitman (1985) Phys.
Reports 124, 315-417, "Ideal Observational Cosmology".
3. C. Hellaby, (2001) Astron. Astrophys. 372,357-363, "Multicolour Observations, Inho-
mogeneity and Evolution" .
4. C. Hellaby (2003) Mathematics of Gravitation II, Warsaw, September 2003,
http://www.impan.gov.pl/Gravi tation/ConfProc/, "Finding the Metric of the Cos-
mos".
5. N. Mustapha, C. Hellaby and G.F.R Ellis (1997) Mon. Not. Roy. Astron. Soc. 292,
817-30, "Large Scale Inhomogeneity Versus Source Evolution: Can We Distinguish
Them Observationally?".
6. G. Lemaitre (1933) Ann. Soc. Sci. Bruxelles A53 51-85; reprinted in Gen. Rei. Grav.
29, 641-80 (1997).
7. R.C. Tolman (1934) Froc. Nat. Acad. Sci. 20 169-76, reprinted in Gen. Rei. Grav.29,
935-43 (1997).
8. C. Hellaby & K. Lake, Astrophys. J. 282, 1 (1984); and errata in Astrophys. J. 294,
702 (1985)
9. C. Hellaby, Class. Q. Grav. 4, 635-50 (1987)
10. H.-C. Lu, C. Hellaby, in preparation (2005).
11. C. Hellaby, (2006) Mon. Not. Roy. Astron. Soc. 370, 239-244, "The Mass of the
Cosmos" .
This page intentionally left blank
N onsingular Cosmology -
Inflation
This page intentionally left blank
 EMERGENT UNIVERSE WITH BULK VISCOSITY

S MUKHERJEE
Dept of Physics, North Bengal University, Darjeeling 734 430, India
[email protected]

B C PAUL
Dept of Physics, North Bengal University, Darjeeling 734 430, India
[email protected]

N K DADHICH
Inter-University Centre for Astronomy and Astrophysics, Pune 411007, India
[email protected]

S D MAHARAJ
Astrophysics and Cosmology Research Unit, School of Mathematical Sciences
University of KwaZulu-Natal Durban 4000, South Africa
[email protected]

A BEESHAM*
Department of Mathematical Sciences, University of Zululand, P Bag X1001,
Kwa-Dlangezwa 3886, South Africa
[email protected]

The relationship of emergent universe models to cosmological models with bulk viscosity
is studied. It is found that the two sets of models have close resemblance.

1. Introduction
In a recent paper, 1 we have presented a general framework for an emergent universe
scenario in which one seeks a universe which is eternal and large enough so that
space-time may be treated classically. There is no time-like singularity. In these
models, the universe in the infinite past is in an almost static state but it eventually
evolves into an inflationary stage. These ideas are in conformity with the Lemaitre-
Eddington concepts forwarded in the early days of modern cosmology, although the
details and the context are different now. An emergent universe model, if developed
in a consistent way is capable of solving the well known conceptual problems of
the big-bang model. A model of an ever-existing universe, which eventually enters
at some stage into the standard big bang epoch, and which is consistent with fea-
tures precisely known to us, will be of considerable interest. The purpose of this
paper is to elucidate the relationship between the emergent universe scenario and a
cosmological model with bulk viscosity.
In seeking a model of the emergent universe, in addition to the above, we assume
the following features for the universe:

* Presenter

1873
1874

(1) The universe is isotropic and homogeneous at large scales.

(2) It is spatially fiat, as indicated by WMAP results. 2
(3) The matter, or, in general, the source of gravity has to be described by quantum
field theory.
(4) The universe may contain exotic matter so that energy conditions may be vio-
lated.
(5) The universe is accelerating as suggested by supernova observations. 3

2. Emergent Universe
There are several solutions in the literature describing an emergent universe,4 and
we attempted to classify and study these models in a general way. A simple approach
was to look for an equation of state which could possibly lead to such models. To
this end, the following equation of state was proposed 1
p(p) = Ap - BJP (1)
where A and B are constants. The energy density p may have different components,
each satisfying its own equation of state. Thus, the choice of Eq. (1) may be looked
upon as a mathematical tool for generating solutions for an emergent universe. For a
fiat universe and the Robertson-Walker metric, the following solution was obtained
2

at
() = (
3K(A+1) (
a
2
+ -- v'3
e Bt
))3(A+l)
2 (2)
2 V3B
where K and a are two constants of integration.

3. Bulk Viscous Models

In order to include the effects of bulk viscosity, consider a fiat universe and the
Robertson-Walker metric. The Friedmann equation is
0,2
p=3 2 , (3)
a
and the energy conservation equation is

p + 3(p + Peff)~ = 0, (4)

a
where Pef f is the total pressure given by
a
Peff = P - 3TJ-, TJ > O. (5)
a
Here P is the perfect fiuid pressure given by

p=b-1)p, (6)
where r = constant and TJ is the coefficient of bulk viscosity, usually taken to be a
power law function of the energy density p. Using Eq. (3), Eq. (5) becomes

Peff = b- l)p - TJV3P. (7)

 1875

For comparison with the emergent models, consider only the case 1] = constant. 5
From Eqs. (1) and (6), it is clear that the two equations are the same if we take
A = (r - 1) and B = V31].

4. Conclusion
We have demonstrated a close resemblance between the emergent universe models
and cosmological models with bulk viscosity. This comparison may facilitate the
construction of models to describe the later evolution of the universe.

References
1. S. Mukherjee, B. C. Paul, N. K. Dadhich, S. D. Maharaj and A. Beesham, Class.
Quantum Gmv. 23, 6927 (2006).
2. H. V. Peiris, E. Komatsu, L. Verde, D. N. Spergel, C. L. Bennett, M. Halpern, G.
Hinshaw, N. Jarosik, A. Kogut, M. Limon, S. S. Meyer, L. Page, G. S. Tucker, E.
Wollack, and E. L. Wright, Astrophys. J. (Suppl.) 148,213 (2003).
3. S. Perlmutter et aI, Astrophys. J. 517, 565 (1999)
4. G. F. R. Ellis and R. Maartens, Class. Quantum Gmv. 21, 223 (2004); G. F. R. Ellis,
J. Murugan and C. G. Tsagas, Class. Quantum Gmv. 21, 233; S. Mukherjee, B. C.
Paul, S. D. Maharaj and A. Beesham, gr-qc/0506103; D. J. Mulryne, R. Tavakol, J.
E. Lidsey and G. F. R. Ellis, Phys. Rev. D 71, 123512 (2005).
5. J. D. Barrow, Nucl. Phys. B 310, 743 (2988); C. Murphy, Phys. Rev. D 35, 1146
(1973); R. Treciokas and C. F. R. Ellis, Commun. Math. Phys. 23, 1 (1971).
 BULK VISCOSITY IMPACT ON THE SCENARIO
OF WARM INFLATION

JOSE PEDRO MIMOSO* and ANA NUNESt

Departmento de Fisica e Centro de Fisica Teorica e Computacional
Universidade de Lisboa,
Avenida Professor Gama Pinto 2, P-1649-003 Lisboa,
Portugal
* jpmimoso@cii·fc.ul.pt
t [email protected]

DIEGO PAv6Nt
Departmento de Fisica,
Universidad Autonoma de Barcelona,
08193 Bellaterm (Barcelona),
Spain
t [email protected]

The decay of the inflaton into radiation and particles during the slow-roll suggests that
these may interact with each other and that the latter may also decay into subproducts
before inflation is completed. As a consequence, the fluid is no longer perfect and a non-
negligible bulk viscosity necessarily sets in. We write the corresponding equations as an
autonomous system and study the asymptotic behavior, the conditions for the existence
of scaling solutions, and show that the late time effect of fluid dissipation alleviates the
depletion of matter and increases the duration of inflation.

Keywords: Theoretical cosmology, early universe, inflation

1. Introduction
In the warm inflation scenario the inflaton decays during the slow-roll into particles
and radiation at such a rate, r, that at the end of the accelerated expansion the
temperature of the cosmic fluid is high enough for the conventional hot big-bang
scenario to proceed right away.l Thus, the reheating mechanism -typical of cool
inflation 2 - can be dispensed with altogether. Aside from this, warm inflation has
other significant advantages: (i) the slow-roll condition ¢2 « V(¢) can be fulfilled
for steeper potentials, (ii) the density perturbations generated by thermal fluctua-
tions may be larger than those of quantum origin 3 , and (iii) it may provide a very
useful mechanism for baryogenesis. 4
To simplify the dynamics of warm inflation many works treated the particles
created in the decay of the inflaton purely as radiation, thereby ignoring the ex-
istence of particles with mass in the decay fluid. In a recent paper,5 we took into
account the presence of particles with mass as part of the decay products, and pro-
vided a hydrodynamical description of the mixture of massless and non-massless
particles by an overall fluid with equation of state p = (r -l)p, where the adiabatic
index I lies in the interval [1,2] (for pure radiation I = 4/3), and endowed with
a dissipative bulk pressure, II. The latter originates quite naturally via: (i) inter-
particle interactions, 6 and (ii) the decay of particles within the fluid. 7 With regard

1876
 1877

to (i), it should be noted that II arises spontaneously in mixtures of different par-

ticles species, or of the same species but with different energies -a typical instance
in laboratory physics is the Maxwell-Boltzmann gas. 8 One may think of II as the
internal "friction" that sets in as a consequence of the diverse cooling rates in the
expanding mixture, something to be expected in the matter fluid originated by the
decay of the inflaton. As for (ii), it is well known that the decay of particles within
a fluid can be formally described by a bulk dissipative pressure -see, e.g.,7,g. In
this connection, it has been proposed that the inflaton may first decay into a heavy
boson which subsequently decays in two light fermions. 10
In this short Communication we limit ourselves to briefly summarize the salient
findings of Ref., 5 paying special attention to the conditions for the existence of
scaling solutions -i.e., solutions for which the energy densities of the inflaton and
the fluid (matter plus radiation) keep a constant ratio.

2. Warm inflation with n = -3C,H and r = constant

The autonomous system of equations can be written as,
x' = x[Q - 3(1 + fX)] - W(¢)y2 , y' = [Q + W(¢)x]y,
x' =xQ, ¢'=V6x (W(¢) = J372(o¢V/V) , X=II/(3H 2), f=-r/(3(»,

where x 2 = J} /(6H 2), y2 = V(¢)/(3H 2), Q = (3/2)[2x 2 + ,(1- x 2 - y2) + X], and
a prime stands for derivative with respect to In a (see Ref. 5 for details).
Fixed points are found at finite ¢, namely, x* = y* = 0, ¢ = ¢*' X* = -, (which
correspond to matter dominated de Sitter solutions), and x* = 0, ,(I - y;) = -X*
(de Sitter scaling solutions), and also at ¢ --+ 00 but these do not show scaling
solutions.

3. Warm inflation with n = -3C, p O: H{3 and r = r(<t» H8

In this more general case the autonomous system reads,

x' = x[Q - 3(1 + 3)]- W(¢)y2 , y' = [Q + W(¢)x]y,

r' = [V6(o¢f /f)x + Q(l - o)]r, ¢' = V6x,

where X = _3 CX (1_x 2_ y2)CX and Q = (3/2)[2x2+,(1-x2-y2)+xL with 6" > 0, < °

Q < 1, and (3 > 0. The results for the scaling solutions can found in the Table III of.5

The fixed points at finite ¢ constitute: (i) a line of fixed points with x = y = r = °
(these are matter dominated and the solutions are unstable), and (ii) another line
x = 0, ¢ = ¢o, (1 - y2)1-CX = 3a (/! for any r i= o. In this second case, when V(¢)
has a minimum, there is either a stable node, 0 < W'(¢o) < 9(1 + r)2/(4V6y;),
1878

or a stable sink, W'(¢o > 9(1 + r)2j(4y6y;) > 0 -here y; = 1 - (30{!r)1/(1-a).

A sufficiently flat effective potential can thus be achieved with the help of viscous
effects, whence "assisted" inflation l l can be dispensed with. As for fixed points
at ¢ -----> 00, scaling solutions exist when W(¢CXJ) = -J3/2,\ < 0 provided that
W(¢) = y6 (8¢f jf)j(l - 8).

4. Concluding remarks
Our main findings can be summarized as follows:

(1) Relevant asymptotic regimes arise in connection with maxima and minima of
V( ¢), at finite ¢, and with the asymptotic behavior at ¢ -----> 00.
(2) In the latter case, the existence of scaling solutions depends on f( ¢). A necessary
and sufficient condition for scaling solutions is that f(¢) ex H.
(3) Scaling solutions exist for f =I 0 regardless the steepness of V(¢).
(4) When II =I 0 the inflationary region takes a larger portion of the phase space
than otherwise -compare Figs. 1 and 2 of Ref. 5 Thus, the presence of a dissi-
pative pressure -which arises on very general grounds- facilitates inflation and
lends support to the warm inflationary proposal.

The next step should be to study the scalar and tensorial perturbations stem-
ming from this model.

References
1. A. Berera, Phys. Rev. Lett, 75,3218 (1995); A. Berera and L.Z. Fang, Phys. Rev. Lett.
74,1912 (1995).
2, A.R Liddle and D. Lyth, Cosmological Inflation and Large Scale Structure (CUP,
Cambridge, 2000); A. Linde, hep-thj0503203.
3. A. Berera, Nucl. Phys. B 585,666 (2000); A.N. Taylor and A. Berera, Phys. Rev. D
62, 083517 (2000).
4. RH. Brandenberger and M. Yamaguchi, Phys. Rev. D 68, 023505 (2003).
5, J.P. Mimoso, A. Nunes and D. Pavon, Phys. Rev. D 73, 023502 (2006).
6. L. Landau and E.M. Lifshitz Mecanique des Fluids (MIR, Moscow, 1971); K. Huang,
Statistical Mchanics (J. Wiley, N.Y., 1987).
7. Ya. B. Zel'dovich, JETP Lett. 12, 307 (1970); J.D. Barrow, Nucl. Phys. B 310, 743
(1988).
8. S. Harris, An Introduction to the Theory of Boltzann Equation (Holt,Reinhart and
Winston, N.Y., 1971).
9. W. Zimdahl and D. Pavon, Phys. Lett. A 176, 57 (1993); W. Zimdahl, Mon. Not. R.
Astron. Soc. 280, 1239 (1996); W. Zimdahl, Phys. Rev. D 53, 5483 (1996).
10. A. Berera and RO. Ramos, Phys. Lett. B 567, 294 (2003).
11. A.A. Coley and RJ. van den Hoogen, Phys. Rev. D 62, 023517 (2000); L.P. Chimento,
A.S. Jakubi, D. Pavon and N. Zuccala, Phys. Rev. D 65, 083510 (2002).
Quantum Cosmology and
Quantum Effects in
the Early Universe
This page intentionally left blank
 SCALAR FIELD PHASE DYNAMICS IN PREHEATING

T. CHARTERSt, A. NUNES+ and J. P. MIMOSO§

t Departamento de Mecanica/Sec<;ao de Matematica
Instituto Superior de Engenharia de Lisboa
Rua Conselheiro Emidio Navarro, 1, P-194g-014 Lisbon, Portugal
Centro de Fisica Te6rica e Computacional da Universidade de Lisboa
Avenida Professor Gama Pinto 2, P-1649-003 Lisbon, Portugal
§ + Departamento de Fisica, Faculdade de Ciencias da Universidade de Lisboa
Centro de Fisica Te6rica e Computacional da Universidade de Lisboa
Avenida Professor Gama Pinto 2, P-1649-003 Lisbon, Portugal
t [email protected], [email protected], § [email protected]

We study the model of a massive inflaton field ¢ coupled to another scalar filed X with
interaction term g2¢2X 2 for the first stage of preheating. We obtain the the behavior of
the phase in terms of the iteration of a simple family of circle maps. When expansion
of the universe is taken into account the qualitative behavior of the phase and growth
number evolution is reminiscent of the behavior found in the case without expansion.

1. Introduction

The reheating mechanism was proposed as a period, immediately after inflation,

during which the inflaton field ¢ oscillates coherently about its ground state and
swiftly transfers its energy into ultra-relativistic matter and radiation, here modelled
by another scalar field x. Consider the potential V(¢) = 1/2m~¢2 and interaction
potentiaP-4 Vint(¢, X) = g2¢2X 2. The evolution of the flat FRW universe is given
by
2
3H = !~l (~¢2 + V(¢) + ~X2 + g2¢2X2) , (1)

where H = R/ Rand R is the FRW scalar factor. The equations of motion in a

FRW universe for a homogeneous scalar field ¢ coupled to the k-mode of the X field
are given by
¢ + 3H¢ + (m~ + g2X%) ¢ = 0, (2)
Xk + 3HXk + W~(t)xk = 0, (3)
where w~(t) = k 2/ R2+g2¢2. The number density nk(t) of particles with momentum
k can be evaluated as the energy of that mode divided by the energy of each particle
nk(t) = (IXkI 2 + w2(t)lxkI 2) /(2Wk(t)) - 1/2.
In this communication we extend the formalism Of1 to give a full description
of the dynamics of the phase of the field modes Xk, which in turn determines the
evolution of the growth factor ILk. For a complete reference see Charters et al. 6

2. Phase dynamics in Minkowski space-time

Typical values of the parameters 1,5 are g2 .:::; 10- 6 , m = 1O- 6 m pl, A = oompl, where
0< a < 1, and thus b::; 00 2 X 106 . In the present case we have, Vb» 1, and it is

1881
1882

-0.1

-0.2

_0.3LL-~.L-~~---:-'-~c:-:'-:~LL--'----'J

31.5 32.0 32.5 33.0 33.5 34.0 34.5

l12
b
(a) (b)

Fig. 1. (a) Bifurcation diagram of the family of circle maps (7) for Vb E [0, 1r]. (b) The asymptotic
value of /-Lo as a function of b for Vb E [101r, llrr] computed analytically (full line) and numerically
(dotted line). Also shown (in grey) are all the values of /-L~, j = 100,101, ... ,200.

possible to construct an approximate global solution

which is valid except in the neighbourhood of tj = jJr, j = 0,1, .... a{ and f3~ are
constants and w~(t) = ak + bsin 2(t) with ak = k 2/m~ and b = g2 A2 /m~, and A
is the constant amplitude of the field cp, and the time variable is now t --+ m¢t.
The parameters (a{, f3{) for consecutive j are determined by the behaviour of the
solution of (3) in Minkowski space-time for t close to tj. In terms of the phase
v~ = arg f3~ + e~ of the field Xk when t = tj, one getsl

(5)

where e(b, f£) = fa"" w(s)ds, PK = exp( -Jrf£2 /2), 1'£2 = ak/Vb, 'f'K arg(f((l +
iK,2)/2)) + 1'£2/2(1 + In(2/K,2)), and 1'£ = k(jAgmq,' Since nk = lf3kI 2, the growth
index J1t defined by n{+l = n{ exp(2JrJ1~), is given by,l in terms of the phase
v~ = arg f3~ + e{ of the field Xk when t = tj,

(6)

The properties of the family (5) is best understood by looking at the behavior (see
Figure 1) of the family Pb,o(v) parametrised by Vb,6

f1 y2 sin v - cos v
Pb,O(V) = 2v b + arctan y2 .' (7)
2cosv - smv

The map has a two dynamical regimes, a strongly attractive fixed point, and random
oscillations around a mean value (see Figure 1).
 1883

3.0 a :.- • 0 ••••

•• .• 0.2 0
o·il
i~
-"0

... • o
.000 0 • •
00.

2.0 . tl
tl
0.1 "\I
a
a
A
0
• •
eo
o·
..
o·

;-

.
A
a
il
il tl
•
•
::l. 0.0
bl • • - ~Map(av~
ODE(avg)
.~Map
A A IA
1.0 OA
o• o (avg)!
ODE (avgll
-0.1
.. • 10a Map
• ••
A A 0. ODE _

.-
• • Map , -0.2
a
a
•
• 0 *0•• OO~
- ..
0.00 10 15 20 25 30 10 15 20 25 30
j j
(a) (b)

Fig. 2. For bo = 5 X 103 and K:O = 0.1 and with initial conditions n~ = 1/2 and v~ = 0 (a) The
phase (b) The growth factor. For the same initial conditions and parameter values, the iteration
of equations are plotted as full circles, and the numerical integration of equations and are plotted
as open circles. This were carried out until Jb(tj) R; 1.

3. Dynamics in an expanding universe

In the first stage of preheating, that ends when nx(t) ~ m~1>(t)/g, where 1>(t)
is the varying amplitude of the inflaton field ¢, equations (1), (2) decouple from
(3), and the evolution of the inflaton field and of the scale factor R(t) is given byl
¢(t) = 1>(t) sint, 1>(t) = mpz/(3(71-j2 + t)), R(t) = (2t/7r)2/3. Equation (3) can be
reduced to the form of an harmonic oscillator the change of variable X k = R 3 / 2 Xk,
with frequency given by w 2 = k 2/(m~R(t)2) + g2¢(t)2 /m~ + 6/m~ and where
6/m~ « 1. The preheating period ends when g1>(t)/m¢ '::::' 1, and, during preheating,
the rate of variation of those parameters and the oscillations of the inflaton field
are much slower than the oscillations of the Xk modes. The changes in occupation
numbers nk will occur at t = j7r with exponential growth rate given by (6), provided
we take /'1,j = k/(R(tj) Jgm¢1>(tj)) and Jbj = g1>(tj)/m¢ (see Figure (2)).

4. Conclusions
We use the theory of developed by Kofman et al. l to show that the phase dynamics
of the modes Xk in Minkowski space-time is given by the properties of a simple family
of circle maps. In the case of an expanding universe the qualitative behaviour of
the phase and growth number evolution is reminiscent of the behavior found in the
case without expansion.

References
1. L. Kofman, A. D. Linde and A. A. Starobinsky, Phys. Rev. D 56 (1997) 3258.
2. A. D. Dolgov and A. D. Linde, Phys. Lett. B 116 (1982) 329.
3. L. F. Abbott, E. Farhi and M. B. Wise, Phys. Lett. B 117 (1982) 29.
4. J. H. Traschen and R. H. Brandenberger, Phys. Rev. D 42 (1990) 2491.
5. A. D. Linde Phys. Lett. 108B, (1982) 389.
6. T. Charters, A. Nunes and J. P. Mimoso, Phys. Rev. D 71 (2005) 083515.
 BRANCH WAVE FUNCTIONS FOR QUASI-CLASSICAL
HOMOGENEOUS UNIVERSES

DAVID CRAIG
Le Moyne College
1419 Salt Springs Road, Syracuse, New York 13214
[email protected]

Quantum mechanics normally requires that a history of a system be measured before

the probability of that history can be meaningfully discussed. In quantum cosmology,
external measurements have no meaning and other strategies must be sought. "Deco-
herent histories" formulations of quantum theory take the hint from ordinary quantum
mechanics that measurement serves to destroy interference between alternative histories
and formalizes this observation into a scheme for making internally consistent quantum
predictions for closed systems even in the absence of anything that resembles a clas-
sical measurement situation. I discuss the construction of such a quantum theory for
the example of the Friedmann-Robertson-Walker cosmology. In particular, coarse grain-
ings suitable to characterize quasiclassical behavior are described, and the branch wave
functions for quasiclassical FRW cosmologies are exhibited.

Whatever the correct theory of quantum gravity turns out to be, it is not unrea-
sonable to suppose that it will contain elements of a recognizable union of existing
notions of particle physics, general relativity, and quantum mechanics. In order to
provide a satisfactory account of quantum gravitational phenomena, however, the
theory must be capable of making predictions concerning the nature of the very
early universe, when it is likely that no subsystems were approximately classical,
and concerning the universe as a whole, of which there are by definition no exter-
nal observers. These circumstances undermine the usual predictive framework of
Copenhagen quantum theory, which relies on some variant of the notion of mea-
surement in order to assign probabilities to quantum mechanical alternatives. In
the absence of measurements, the theory simply cannot make predictions!
Any sensible quantum theory of gravity, therefore, must be formulated within
a framework for quantum prediction which does not rely on notions of approxi-
mate classicality or external measurements. One such framework is the generalized
quantum mechanics of closed systems,l a generalization of consistent histories for-
mulations of quantum theory.2,3 The essential ingredients of the quantum mechanics
of closed systems are (i) a set n = {h} of histories which characterize in the finest
detail available the possible histories of the system; and (ii) a decoherence functional
d(h, hi) on pairs of histories in n which both measures the quantum interference
between histories, and determines the probabilities of each history in those "consis-
tent" or "decoherent" sets for which the interference among the histories, as mea-
sured by the decoherence functional, vanishes. In such decoherent sets, therefore,
d(h, hi) = p(h)6hhl. The decoherence functional for ordinary quantum mechanics is
(1)
Here I'ljJh) 1S the "branch wave function" corresponding to the history h, and is

1884
 1885

simply the initial state lib) propagated according to the physical properties of h -
in Copenhagen quantum theory it would simply be the state of a system that had
been observed to have followed that particular history.
As a model for how this program might be implemented in a realistic quan-
tum gravitational theory, a decoherence functional has been given for recollapsing,
homogeneous cosmological models: 5

(2)
The ingredients in this construction are as follows. The product • is a positive
Rieffel-type inner product between cosmological branch wave functions Wh. The Wh
are constructed from the initial cosmological state W as

(3)
Here the q's are minisuperspace coordinates, 0 is a conserved Klein-Gordon-type
product, and Ch(q", q') is the "class operator" or "restricted propagator" for the
history h. In our example model it is constructed schematically as

(4)

where oq is a gauge-fixed measure, S is a suitable action, and the integral is over all
paths in minisuperspace connecting q' to q" which have the physical characteristics
consistent with the properties specified by the history h.
(Though we do not expect this particular path integral representation to pro-
duce a realistic quantum theory of gravitation, it is a model in which the predictive
structure of the theory may be fruitfully explored. Within the context of a genuine
quantum theory of gravity such as loop quantum gravity, we would expect that
the details of the products employed in the construction of the decoherence func-
tional and the specific definition of the restricted propagators would be replaced by
constructions more appropriate to that particular theory. The overall flavor of the
construction of the decoherence functional, however, we would expect to be similar.)
Within this framework, one may prove 5 that the probability that the universe
"begins" in a minisuperspace region tJ. i and "ends" in tJ. j is given by
(5)
where d~ is the volume element on the initial hypersurface and J ij is a conserved
current appropriately constructed from the branch wave functions. The traditional
"J . d~" heuristic of quantum cosmology may therefore be derived in this setting.
As a specific example consider universes which evolve in an approximately clas-
sical fashion throughout most of their history. The histories appropriate to such
evolutions, and therefore the paths integrated over in Eq. (4), are those which re-
main close to a classical trajectory. The corresponding branch wave functions for
FRW universes with scalar matter may be calculated to one loop.6 The result of this
calculation of a branch wave function for a particular WKB inital state concentrated
around a family of classical trajectories is illustrated in Fig. 1.
1886

Fig. 1. The branch wave function for a recollapsing Friedmann-Robertson-Walker universe with
scalar matter which evolves in an approximately classical fashion throughout its history. The axis
into the page is essentially the log of the scale factor and the horizontal axis the value of the
scalar field. For WKB initial conditions favoring the classical trajectories shown, the branch state
remains concentrated around those classical trajectories throughout its evolution. (An arbitrary
scaling has been applied in order to emphasize certain features.)

We have offered a concrete predictive framework within which more realistic

quantum theories of gravitation, mutatis mutandis, may be formulated. We argue
strenuously that any theory of quantum gravity must be formulated within a frame-
work of this kind in order not to be devoid of predictive content! The fundamental
questions of quantum theory do not go away simply because the systems to which
it is applied are in themselves distant from familiar laboratory settings. Indeed, the
difficulties become, if anything, considerably more accute. We hope we have offered
a specific formalism within which the predictions of quantum gravitational theories
may be addressed in an unambiguous fashion.

References
1. J.B. Hartle, in Gravitation and Quantizations, B. Julia and J. Zinn-Justin, eds. (North
Holland, 1995).
2. R. Griffiths, Consistent Quantum Theory (Cambridge University Press, 2002), and
references to earlier work therein.
3. R. Omnes, Interpretation of Quantum Mechanics (Princeton University Press, 1994),
and references to earlier work therein.
4. M. Gell-Mann and J.B. Hartle, in Complexity, Entropy, and the Physics of Informa-
tion, SFI Studies in the Sciences of Complexity, Volume VIII, W. Zurek, ed. (Addison-
Wesley, 1990), or in Proceedings of the 3rd International Symposium on the Founda-
tions of Quantum Mechanics in the Light of New Technology, S. Kobayashi, H. Ezawa,
M. Murayama, and S. Nomura, eds. (Physical Society of Japan, 1990).
5. D.A. Craig and J.B. Hartle, Phys. Rev. D69, 123525-123547 (2004).
6. D.A. Craig, Branch Wave Functions for Quasi-Classical Priedmann-Robertson- Walker
Cosmologies, in preparation.
 QUANTUM PHANTOM COSMOLOGY

BARBARA SAND HOFER

Institute for Theoretical Physics,
Cologne University, 50937 Cologne, Germany
[email protected]

We apply the formalism of quantum cosmology to models containing a phantom field.

Three models are discussed explicitly: a toy model, a model with an exponential phan-
tom potential, and a model with phantom field accompanied by a negative cosmological
constant. In all these cases we calculate the classical trajectories in configuration space
and give solutions to the Wheeler-DeWitt equation in quantum cosmology. In the cases
of the toy model and the model with exponential potential we are able to solve the
Wheeler-DeWitt equation exactly. For comparison, we also give the corresponding solu-
tions for an ordinary scalar field. We discuss in particular the behaviour of wave packets
in minisuperspace. For the phantom field these packets disperse in the region that cor-
responds to the Big Rip singUlarity. This thus constitutes a genuine quantum region at
large scales, described by a regular solution of the Wheeler-DeWitt equation.

1. Introduction

It is a striking fact that our universe is currently accelerating. A major open problem
.is to provide a fundamental type of matter which may be responsible for this. Such
a matter component is called 'Dark Energy' and in the simplest case modelled
by a cosmological constant with equation of state parameter w = -1. But on
observational grounds alone, a very negative equation of state, W < -1, cannot be
excluded. Such a dark energy component was dubbed phantom. In the simplest case
it is modelled by a scalar field with negative kinetic energy, in the following refered
to as phantom field. Such phantom models of the universe admit a new type of
singularity called a Big Rip singularity. At the Big Rip, energy density and pressure
diverge as the scale factor a(t) goes to infinity at a finite time.
Even though this type of cosmology has stirred intensive research, a full quantum
cosmological treatment is still lacking. So far, quantum effects were only considered
in certain phases of the evolution. But we know that quantum theory is a universally
valid theory. Therefore, the universe as a whole has to be described by quantum
theory, too.
Quantum cosmology must be based on a theory of quantum gravity. Our analysis
rests on the Wheeler-DeWitt equation of quantum geometrodynamics. Indepen-
dently of the correct theory of quantum gravity, this framework should yield an
adequate description at least on the energy scales below the Planck scale (if not on
all scales).
In the following we will give a brief outline of our studies and results on such a
quantum cosmological analysis of phantom cosmologies. For further particulars and
references we direct the reader to the published version. 1

1887
1888

2. Classical models
We studied flat Friedmann cosmologies with phantom matter content. Phantom
matter is mimicked by a homogeneous scalar field ¢ with reversed sign of the kinetic
energy term. Apart from a toy model, two models were studied which differ by the
choice of potential for the phantom field:

1. Simple model with Big Rip

A= 0,

2. Big Rip to Big Rip model

V(¢) = Vocosh2 (f), A < o.

Here, A denotes the cosmological constant. The classical equations of motion were
solved and the classical trajectories given in configuration space. Whereas the first
model evolves from an initial value of the scale factor into the Big Rip singularity,
the second model contains a turning point in the evolution of the scale factor. It
starts and ends with a Big Rip singularity.
For comparability the same models were also studied with a conventional scalar
field replacing the phantom field, thus containing a Big Bang singularity.

3. Quantum models
Using minisuperspace quantisation, one is led to the Wheeler-DeWitt equation
(WDW),

1i2 [j2 1i2 8 2 )

( "2 8a2 + "2 8¢2 + e6a V(¢) 1l1(a, ¢) = 0 .

In contrast to conventional matter content, the presence of phantom matter destroys

the hyperbolic nature of the WDW equation. We adopted a pragmatic approach to
solve this equation. As boundary condition we required that out of the solutions, one
should be able to form a wave packet which is peaked around the classical trajectory
at least throughout some region in configuration space. Our solution thus represents
a semi-classical state in this region of configuration space. The question is whether
and where this semi-classical approximation breaks down.

4. Results
For the model with exponential potential of the phantom field, we were able to
solve the WDW equation exactly. The second model was solved in the vicinity of
the singularity using an approximation of the potential there. We found that in
both models, the semi-classical approximation breaks down in the vicinity of the
classical Big Rip singularity. The wave packets disperse in that region 1. The Big
 1889

Rip singularity is thus 'smeared out' by quantum effects at large scale factor. Once
the wave packets disperse, no approximate time parameter can be defined and the
classical evolution terminates in a singularity-free way. Thus we found q'uant'um
effects for large scale factors.
For the reference models containing a standard scalar field we found that, maybe
contradicting naive expectations, the wave packet does not spread at the Big Bang.
Rather the wave packet vanishes there due to the implementation of boundary
conditions 1. In this way, the Big Bang singularity is removed from the quantum
theory.

Fig. 1. The amplitude of the wave packet for an exponential potential solution of the WDW
equation for the phantom (left) and scalar (right) field model. The wave packet for the phantom
field model is seen to spread near the classical singularity. For the scalar field model \jI --> 0 at
the origin. In each sector corresponding to one copy of the (a, cf;) plane, the same wave packet
propagates. The coordinates u and v are complicated functions of the scale factor and cf;.

Acknowledgements
The author thanks Claus Kiefer and Mariusz Df}browski for collaboration. She also
thanks the Friedrich-Ebert-Stiftung for financial support.

References
1. M. Di}browski, C. Kiefer and B. Sandhofer, Phys. Rev. D 74, 044022 (2006).
 GENERIC EVOLUTIONARY QUANTUM UNIVERSE

MARCO VALERIO BATTISTI

ICRA - International Center for Relativistic Astrophysics
Physics Department (G9) , University of Rome "La Sapienza"
P.le A. Moro 5, 00185 Rome, Italy
[email protected]

GIOVANNI MONTANI
ICRA - International Center for Relativistic Astrophysics
Physics Department (G9), University of Rome "La Sapienza"
P.le A. Moro 5, 00185 Rome, Italy
ENEA C.R. Frascati (U. T.S. Fusione), Via Enrico Fermi 45, 00044 Frascati, Rome, Italy
[email protected]

We consider a Schrodinger quantum dynamics for the gravitational field associated to a

generic cosmological model and then we solve the corresponding eigenvalue problem. We
show that, from a phenomenological point of view, an Evolutionary Quantum Cosmology
overlaps the Wheeler-DeWitt approach.

1. Introduction
The canonical approach in Quantum Gravity is characterized by the so-called frozen
formalism, i.e. the absence of a time evolution for the wave functiona1. 1 ,2 It has
been proposed 3 ,4 that such feature disappears as soon as the impossibility of a
physical slicing without frame fixing is recognized for a quantum spacetime. In this
work 5 we start with a Schrodinger dynamics for the gravitational field using Planck
mass particle as a "clock" for the system; and we will analyze the meaning of the
corresponding spectrum (we deal with a new energy density) in the framework of a
generic inhomogeneous Universe.

2. Evolutionary Quantum Gravity

In this section we briefly analyze the implication of a Schrodinger formulation of
the quantum dynamics for the gravitational field. We require the theory to evolve
along the spacetime slicing so that \II = \II (t, {h ij } ); so the quantum evolution is
governed by a smeared Schrodinger equation

iOt \II = H\II == ~3 d3 x (N if) \II (1)

t

being H the super-Hamiltonian operator and N the lapse function. If we now take
the right expansion for the wave functional, the Schrodinger dynamics is reduced
to an eigenvalues problem of the form

HX = EX, HiX = 0, (2)

which outlines the appearance of a non zero super-Hamiltonian eigenvalue.

1890
 1891

Is not difficult to show that the classical limit (in the sense of WKB approx-
imation) of the above model is characterized by the appearance of a new matter
contribution, which admits the following energy density:
_ E(Xi)
P = Too = - 2Vh' h = dethij . (3)

The explicit form of (3) is that of a dust fluid co-moving with the slicing 3-
hypersurfaces, i.e. we deal with TfJ,v = pnfJ,nv.

3. The Generic Quantum Universe and its Spectrum

We now apply the Schrodinger approach to a Quantum Universe that has to be
described by a generic inhomogeneous model,6 which has a dynamics summarized,
asymptotically to the Big-Bang, by the following variational principle 7

(4)

where adopting Misner-like variables R, fJ± 8 (R is the scale factor and fJ± describes
the anisotropies), the super-Hamiltonian has the structure

i) [p~ 1 (2 2 )] 3 p~ R3 3
H (x = t£ -Ii + R3 p+ + p- + 87r R3 - 4t£l;n V(fJ±) + R (Pur + Ppg), (5)

where t£ = 87rl~. We have added to the dynamics of the system an ultrarelativistic

energy density (Pur = JL2 / R 4 ), a perfect gas contribution (ppg = (J2 / R 5 ) and a
scalar field cp (a free inflaton field).
Performing the canonical quantization of this model we obtain the following
eigenvalue problem (2), with the right normal ordering:

1 1 (2 2 )] 3 2 R3 3 }
{ t£ [ORfjDR - R3 0++0_ - 87rR3 0¢ - 4t£ lrn V(fJ±)+R (Pur+Ppg) X=EX·
(6)
The appropriate boundary condition for this problem are: i) X(R = O,fJ±,cp) < 00
that relies on the idea that the quantum Universe is singularity-free and ii) X(R ----+
00, fJ±, cp) = °
that ensures a physical behavior at "large" scale factor.
In order to study the previous eigenvalue problem we expand the wave func-
tion as X(R, fJ±, cp) = J
BK(R)FK(R, fJ±, cp)dK, and then performing an adiabatic
approximation (loRFI « IORBI) we obtain the following reduced problems:

t£ d~ (~:~) + (t£ ~: + R3 (Pur + Ppg) - E) e = 0, (7)

-(8~ + 0: + -38 o~)F + 4t£~6l2in V(jh)F = K 2(R)F.

7rt£
(8)

The function F is a plane wave as soon as we neglect the potential term in (8)
for same R* « 1. The solution of (7) is a series in R multiplied by a Gaussian
1892

function peaked around R = El~/167r. Since we required the wave function to decay
at large scale factor R we have to terminate the series and obtain the spectrum of
the super-Hamiltonian:
a2
n ,
E • = l~(n + 'Y _ 1/2)' (9)
so that the ground state n = 0 eigenvalue, for 'Y < 1/2, is negative; therefore is
associated via (3) to a positive dust energy density.

4. Phenomenology of the Dust Fluid

In order to analyze the cosmological implication of this new matter contribution,
we have to impose a cut-off length in our model, requiring that the Planck length
lp is the minimal physical length accessible by an observer (l ~ lp). So, from the
thermodynamical relation for the perfect gas, we obtain a constraint on the Ppg and
then on the super-Hamiltonian eigenvalue:
3 _ V 3 lp 3
l - - = --- > l Ppg S; O(lll~), (10)
- N 2p,\2
pg
- P

where l is the length per particle and ,\ the corresponding thermal length (,\ = lp).
Therefore we get a 2 S; O(lp) and so IEol S; (lllp): the spectrum is limited by below.
The contribution of our dust fluid to the actual critical parameter is
Odust rv Pdust rv 0 (10- 60 ) . (11)
PToday

Such a parameter is much less then unity and so no phenomenology can came out
(today) from our dust fluid. In other words an Evolutionary Quantum Cosmology
overlaps the Wheeler-DeWitt approach. Finally we face the question of the classical
limit of the spectrum in the sense of large occupation numbers n ----+ 00. As we can
see from (9) the eigenvalue approaches zero as lin. Therefore for very large n, our
quantum dynamics would overlap the Wheeler-DeWitt approach.

References
1. B.S.DeWitt, Phys. Rev., (1967), 160, 1113.
2. K.V.Kuchar, in Quantum Gravity II, a second Oxford symposium, (1981), eds C. J.
Isham et al., Clarendom Press., Oxford.
3. G.Montani, Nucl. Phys. E., (2002), 634, 370.
4. S.Mercuri and G.Montani, Int. J. Mod. Phys., (2004), D13, 165.
5. M.V.Battisti and G.Montani, Phys. Lett. B637, (2006), 203-209.
6. V.A.Belinski, I.M.Khalatnikov and E.M.Lifshitz, Adv. Phys., (1982), 31, 639.
7. R.Benini and G.Montani, Phys. Rev. D, (2004), 70, 103527.
8. C.Misner, Phys. Rev. Lett., (1969), 22, 1071.
 QUANTUM COSMOLOGY FROM THREE DIFFERENT
PERSPECTIVES

GIAMPIERO ESPOSITO
INFN, Sezione di Napoli, Complesso Universitario di Monte S. Angelo, Via Cintia, Edificio N',
80126 Napoli, Italy
and
Universita degli Studi di Napoli Federico II, Dipartimento di Scienze Fisiche, Complesso
Universitario di Monte S. Angelo, Via Cintia, Edificio N', 80126 Napoli, Italy
giampiero. [email protected]

Our review is devoted to three promising research lines in quantum cosmology and the
physics of the early universe. The nonperturbative renormalization programme is mak-
ing encouraging progress that we here assess from the point of view of cosmological
applications: Lagrangian and Hamiltonian form of pure gravity with variable G and
A; power-law inflation for pure gravity; an accelerating universe without dark energy. In
perturbative quantum cosmology, on the other hand, diffeomorphism-invariant boundary
conditions lead naturally to a singularity-free one-loop wave function of the Universe.
Last, but not least, in the braneworld picture one discovers the novel concept of cos-
mological wave function of the bulk space-time. Its impact on quantum cosmology and
singularity avoidance is still, to a large extent, unexplored.
Keywords: quantum cosmology; renormalization group; braneworld

In this brief review we focus on three (among the many) peculiar aspects of
modern quantum cosmology, with the hope of leading quickly the reader towards
open research problems. No completeness can be achieved in such a short presenta-
tion, and we therefore apologize in advance with the colleagues who might not find
a proper acknowledgment of their work.
The familiar formulation of quantum cosmology via functional integrals relies
upon the pioneering work of Misner,1 Hartle and Hawking. 2 The main idea of the
functional-integral approach is to build in-out amplitudes following Feynman: the
amplitude to go from a metric g1 and a (matter) field configuration CP1 on a spacelike
surface 51 to a metric g2 and a (matter) field configuration CP2 on a spacelike surface
52 is (formally) expressed as the functional integral of the exponential of i times
the action, supplemented by gauge-fixing and ghost terms,3 taken over all metrics
and (matter) fields matching the given boundary data on 51 and 52. In order to
obtain a well-defined prescription, in-out amplitudes are sometimes written first
as Euclidean functional integrals, but severe technical problems occur: integration
measure over all four-geometries with their topologies and unboundedness from
below of the Euclidean action among the many.4
In a cosmological setting, one therefore arrives at the Hartle-Hawking quan-
tum state. 2 According to these authors, the quantum state of the Universe 5 can be
expressed by an Euclidean functional integral over compact four-geometries match-
ing the boundary data on the surface 52, while the three-surface 51 shrinks to
a point (hence the name "no boundary proposal"). One can therefore derive, in
principle, all we know about cosmology from a choice of boundary conditions,6 in-

1893
1894

cluding formation of structure, 7 coupling to matter fields, 8 inflationary solutions, 9

supersymmetric models. lO ,11
Recent progress relies instead on a completely different approach: one builds a
scale-dependent effective action f(k) for quantum Einstein gravity, which is ruled
by the renormalization-group (hereafter RG) equation. If r( k) equals the classical
Einstein-Hilbert action at the ultraviolet cut-off scale /)', one uses the RG equation
to evaluate r(k) \/k < /)', and then sends k -+ 0 and /), -+ 00. The continuum limit
as /), -+ 00 should exist after renormalizing finitely many parameters in the action,
and is taken at a non-Gaussian fixed point of the RG-flow. 12 Over the years, strong
evidence has been obtained in favour of the new ultraviolet fixed point, regardless of
the truncation used. 13 The plot in Ref. 13 shows part of theory space of the Einstein-
Hilbert truncation with its RG flow. The arrows therein point in the direction of
decreasing values of k .14 The flow is dominated by a non-Gaussian fixed point in
the first quadrant and a trivial one at the origin.14
After investigating the RG-improved equations for self-interacting scalar fields
coupled to gravity in a FLRW Universe,15 we have improved the action principle
itself, building the Lagrangian and Hamiltonian formalism with variable G and A
treated as dynamical variables. 16 The latter point is substantially innovative, since
all other investigations in the literature treated G and A as external parameters at
the very best, but not as dynamical variables with an Euler-Lagrange equation for
G.
Unlike models where only the Einstein equations are RG-improved, our frame-
work allows for a non-trivial dynamics of the scale factor even in the absence of
coupling to a matter field. Indeed, if in the pure-gravity case we look for power-law
solutions of the Euler-Lagrange equations of the type 16 ,17
t2 ~2
a(t) = At"', G(t) = 9* e' A(t) = A*{i' (1)

we find for example, in a spatially flat FLRW Universe, that A is undetermined,

while

(2)

Our modified Lagrangian 16 allows therefore for power-law inflation in pure-gravity

models, unlike all previous models in the literature. 13 .18 ,19
The derivation of an infrared fixed point is not on a footing as firm as the
evidence in favour of an ultraviolet fixed point. 13 Nevertheless, on assuming its ex-
istence, we have linearized the RG-flow and, after evaluating the critical exponents,
we have found how the infrared fixed point would be approached. 2o We have also ob-
tained a smooth transition between FLRW cosmology and the observed accelerated
expansion of the universe. 2o
Perturbative quantum cosmology studies instead the first quantum corrections
to the underlying classical dynamics. In particular, one-loop effects can be evaluated
after imposing gauge-invariant boundary conditions, according to the recipe for im-
 1895

posing gauge-invariant boundary conditions in quantum field theory.21 On denoting

bY?Tij a projector acting on the gauge fields zpj, by pO:(zp) and'l/J{3 the gauge-fixing
functionals and ghost fields, respectively, such boundary conditions read as

(3)

[pO:(zp )]aM = 0, (4)

['l/Jf3]aM = o. (5)
For pure gravity, one-loop quantum cosmology in the limit of small three-
geometry22 describes a vanishing probability of reaching the singularity at the
origin (of the Euclidean four-ball) only with diffeomorphism-invariant boundary
conditions,23,24 which are a particular case of the previous scheme. All other sets of
boundary conditions lead instead to a divergent one-loop wave function. 22 ,24,25
We stress that we do not require a vanishing one-loop wave function. We rather
find it, on the Euclidean four-ball, as a consequence of diffeomorphism-invariant
boundary conditions. Peculiar cancellations occur on the Euclidean four-ball, and
the spectral (also called gener alized) (-function remains regular at the origin,23,24
despite the lack of strong ellipticity of the boundary-value problem. 21
In the braneworld picture, branes are timelike surfaces with metric go:{3 embed-
ded into bulk space-time with metric GAB. The action functional can be taken to
be the sum of a four-dimensional (brane) and five-dimensional (bulk) contribution,
i.e. 26
(6)

In general, there exist vector fields RB, RI/ on the space of histories such that

(7)
with Lie brackets given by

(8)

The components of the vector fields RB and RI/ generate five-dimensional and four-
dimensional diffeomorphisms, respectively, while the bulk and brane ghost operators
read 26 (with FA and Xlt the bulk and brane gauge-fixing functionals, respectively)

(9)

J It1 /-= R I/X It -X,i

- It Ri
1/'
(10)

On denoting by SA and TB the bulk ghost fields, the cosmological wave function of
the bulk space-time can be written as 26

(11)
1896

with p,( G AB, S, T) a suitable measure functional, while

- 1 A B A B
S5 = S5[GAB] + 2F WABF + SAQ BT . (12)

The braneworld effective action r can (in principle) be obtained from the for-
mula26

e ir = J( )
v ga(3, p, IJ e iS4 <pBulk,
0 /,
(13)

where v(ga(3, p, IJ) is a suitable measure functional over brane metrics and brane
ghost fields, while 26

(14)
Recent developments in this respect can be found in Refs. 26,27, where the authors
lay the foundations for a sistematic application of the background-field method to
the braneworld picture.
In our opinion, it is of crucial importance to work at least on the following
unsettled issues:

(i) Can one prove in a rigorous way that an infrared fixed point occurs III the
nonperturbative approach?

(ii) Can the spectral cancellations found in 23 ,24 survive the choice of curved back-
grounds with boundary?

(iii) Is braneworld quantum cosmology one-loop singularity free?

Hopefully, the years to come will shed some light on these open problems, and
we also hope that quantum cosmology will make a closer contact with the rich world
of observational cosmology.

Acknowledgments
The work of C. Esposito has been partially supported by PRIN SINTESI. Collabo-
ration with I. Avramidi, A. Bonanno, C. Fucci, A.Yu. Kamenshchik, K. Kirsten, C.
Rubano, P. Scudellaro has been very helpful in the course of working on the topics
this short review is devoted to.

References
1. C.W. Misner, Rev. Mod. Phys. 29, 497 (1957).
2. J.B. Hartle and S.W. Hawking, Phys. Rev. D28, 2960 (1983).
3. B.S. DeWitt, The Global Approach to Quantum Field Theory, Int. Ser. Mon. Phys.
114 (Clarendon Press, 2003).
4. S.W. Hawking, in General Relativity, an Einstein Centenary Survey, eds. S.W. Hawk-
ing and W. Israel (Cambridge University Press, 1979).
5. S.W. Hawking, Nucl. Phys. B239, 257 (1984).
 1897

6. S.W. Hawking, Pont. Acad. Sci. Scr. Varia 48, 563 (1982).
7. J.J. Halliwell and S.W. Hawking, Phys. Rev. D31, 1777.
8. P.D. D'Eath and J.J. Halliwell, Phys. Rev. D35, 1100 (1987).
9. G. Esposito and G. Platania, Class. Quantum Grav. 5, 937 (1988).
10. P.D. D'Eath, Supersymmetric Quantum Cosmology (Cambridge University Press,
1996).
11. P.V. Moniz, Int. 1. Mod. Phys. A11, 4321 (1996); Gen. ReI. Grav. 38, 577 (2006).
12. A. Bonanno and M. Reuter, Phys. Rev. D65, 043508 (2002).
13. O. Lauscher and M. Reuter, hep-th/0511260.
14. http)/www.na.infn.it/Theor /gruppoIV /giespo/COT5july06.ppt
15. A. Bonanno, G. Esposito and C. Rubano, Gen. ReI. Grav. 35, 1899 (2003).
16. A. Bonanno, G. Esposito and C. Rubano, Class. Quantum Grav. 21, 5005 (2004).
17. A. Bonanno, G. Esposito and C. Rubano, Int. J. Mod. Phys. A20, 2358 (2005).
18. A.O. Barvinsky, A.Yu. Kamenshchik and I.P. Karmazin, Phys. Rev. D48, 3677 (1993).
19. F. Bauer, gr-qc/0512007.
20. A. Bonanno, G. Esposito, C. Rubano and P. Scudellaro, Class. Quantum Grav. 23,
3103 (2006).
21. I.G. Avramidi and G. Esposito, Commun. Math. Phys. 200,495 (1999).
22. K. Schleich, Phys. Rev. D32, 1889 (1985).
23. G. Esposito, G. Fucci, A.Yu. Kamenshchik and K. Kirsten, Class. Quantum Grav. 22,
957 (2005).
24. G. Esposito, G. Fucci, A.Yu. Kamenshchik and K. Kirsten, JHEP 05 09:063 (2005).
25. G. Esposito, A.Yu. Kamenshchik and G. Pollifrone, Euclidean Quantum Gravity on
Manifolds with Boundary, Fundamental Theories of Physics, Vol. 85 (Kluwer, 1997).
26. A.O. Barvinsky, hep-th/0504205; Phys. Rev. D74, 084033 (2006).
27. A.O. Barvinsky and D.V. Nesterov, Phys. Rev. D73, 066012 (2006).
 ON THE THERMAL BOUNDARY CONDITION OF THE WAVE
FUNCTION OF THE UNIVERSE*

MARIAM BOUHMADI-LOPEZ 1 ,2t and PAULO VARGAS MONIZ 2 ,q

1 Centro Multidisciplinar de Astrofisica - CENTRA, Departamento de Fisica,
Instituto Superior Tecnico, Av. Rovisco Pais 1, 1096 Lisboa, Portugal
2 Departamento de Fisica, Universidade da Beira Interior,
Rua Marques d'Avila e Bolama, 6200 Covilhii, Portugal

We broaden the domain of application of the recently proposed thermal boundary con-
dition of the wave function of the Universe, which has been suggested as the basis of a
dynamical selection principle on the landscape of string solutions.

1. Introduction
The existence of a multiverse of vast solutions l - 3 to string theory constitutes cur-
rently an important challenge: How to select a Universe or a class from the multi-
verse that will bear significant similarities to ours?
The framework of quantum cosmology4-6 provides a methodology to establish
a probability distribution for the dynamical parameters of the Universe. In this
context, Brustein and de Alwis proposed in, 7 using FRW quantum cosmology, a
dynamical selection principle§ on the landscape of string solutions. 3
We prove that the thermal boundary condition applied in 7 corresponds to the
particular physical situation where the amount of radiation is very large. We then
provide a broader and improved analysis of the generalised thermal boundary con-
dition that is independent of such restrictive limit; 10 i.e. we consider an arbitrary
amount of radiation consistent with the tunnelling of a closed radiation-filled Uni-
verse with a positive cosmological constant.

2. The generalised thermal boundary condition

The thermal boundary condition for the wave function of the Universe states that
the Universe emerges from the string era in a thermally excited state above the
Hartle-Hawking (HH) vacuum. 7 Furthermore, the primordial thermal bath is ef-
fectively described by a radiation fluid whose energy density p "depends" on the
cosmological constant ),
- v
K -~­
),2' (1)

In the previous expressions k and v are parameters that quantify the amount of
radiation, G is the gravitational constant and a the scale factor. Therefore, the

* This research work was supported by the grants

POCI/FP /63916/2005,FEDER-POCI/P /FIS/57547 /2004 and Acc;6es Integradas (CRUP-CSIC)
Luso-Espanholas E-138/04.
t mariam. [email protected]!.pt
[email protected]
§See also Refs. 8 ,g

1898
 1899

transition amplitude of a closed radiation-filled FRW Universe ll ,12 to tunnel from

the first Lorentzian region (a < a_, see Fig. I-a) to the larger Lorentzian region
(a+ < a, see also Fig. I-a) within a WKB approximation reads 7 ,lO

A = exp (E2/) , J-~.!.

- 2~ G v g , (2)
where E = ±I and

9 = *\1"1 + m [E(an) - (1 - m)K(an)] , an = J ~mm' 1 m = \11 - 4io.., (3)

with K(m) and E(m) as complete elliptic integrals of the first and second kind,
respectively. Consequently, the thermal boundary condition implies a switch in the
usual features of the HH4 (E = 1 choice in Eq. (2)) and the tunnelling (E = -1 choice
in Eq. (2))5,6 wave functions; The HH wave function, once the thermal boundary
condition of 7 is assumed, favours a non-vanishing cosmological constant; A '::::' 8.33v,
larger than the one preferred by the tunnelling wave function; A'::::' 4v, (see Fig. I-b).
It turns out that the thermal effect considered in 7 corresponds to a large amount
of radiation where K is close to Kmax (see Fig. I-a); i.e. the turning points a_ and
a+ are very close or equivalently the height of the barrier separating both Lorentzian
regions is very small.
Next we consider a generalised thermal boundary condition for the wave func-
tion lO where we will assume instead an arbitrary amount of radiation, consistent
with a tunnelling of the Universe; i.e. K < Kmax (see Fig. I-a). Consequently,
Eqs. (1)-(3) are replaced by consistent and more general relations where the amount
of radiation as measured by K depends also on A and reads
~ 4VA- 2
K = (1 + 4v A-1)2 . (4 )
Within this broader range, the relevant features is that the transition amplitude as
a function of v / A will be unlike the one deduced in. 7 Indeed, this is the case as is
shown in Figs. I-b and I-c.

0.04
_ 0.06
0.06 0.03
KmaL 0.05 )---------------7'""""<"
0.04 g
9 0.04 0.02
VII 0.03
0.02 0.02 0.01
k 0.Q1
0
-0.01 o 0.05 0.1 0.15 0.2 0.25 o 0.1 0.2 0.3 0.4
[//). [//).

I-a I-b I-c

Fig. 1. Figure I-a corresponds to the potential barrier (Vo - K) separating the two Lorentzian
regions of a closed radiation-filled FRW Universe. Kmax corresponds to the maximum amount
of radiation consistent with the tunnelling of the Universe. We will refer to this situation as a
large amount of radiation. Figures l-b and l-c corresponds to 9 defined in Eq_ (3) for the thermal
boundary condition and the generalised thermal boundary condition, respectively.
1900

Regarding the HH wave function, it now favours a vanishing cosmological con-

stant (v/).. ----+ (0) and K rv 1/(4v). This physical case is represented schematically
by a star in Fig. I-c. In this manner, the role of the HH wave function and subse-
quent transition amplitude is returned to its "original" implication, with the ther-
mal boundary condition being implemented in a fully consistent manner and not
restricted to a narrow (perhaps not fully valid) limit.
Concerning the tunnelling wave function, it favours two possible physical situ-
ations depicted by a circle and a square in Fig. I-c. On the one hand, the "circle"
option corresponds to a large cosmological constant (v /).. ----+ 0) and a small amount
of radiation as measured by K (K)" ----+ 0). On the other hand, the "square" option
implies no tunnelling, that is, 4v /).. ----+ 1 or equivalently 4K)" ----+ 1; i.e. both turning
points coincide. In order to select one of these two possibilities for the tunnelling
wave function, we employed the DeWitt's argument,13.14 since there is a curvature
singularity at small scale factors. It turns out that the preferred value of the cosmo-
logical constant in this case is a large one. Moreover, this condition implies a small
amount of radiation (as measured by the parameter K) allowing consequently the
tunnelling of the Universe.

3. Conclusions
We prove that the thermal boundary condition applied in 7 corresponds to the partic-
ular physical situation where the amount of radiation is very large. We then provide
a broader and improved analysis of the generalised thermal boundary condition that
is independent of such restrictive limit. 10
Acknowledgments
MBL acknowledges the support of a Marcel Grossmann fellowship to attend the
meeting. MBL also acknowledges the support of CENTRA-1ST BPD (Portugal) as
well as the fellowship FCT/BPD/26542/2006.
References
1. R. Bousso and J. Polchinski, JHEP 0006, 006 (2000).
2. M. R. Douglas, JHEP 0305, 046 (2003).
3. L. Susskind, arXiv:hep-th/0302219.
4. J. B. Hartle and S. W. Hawking, Phys. Rev. D 28, 2960 (1983).
5. A. Vilenkin, Phys. Rev. D 27, 2848 (1983).
6. A. D. Linde, Sov. Phys. JETP 60, 211 (1984) [Zh. Eksp. Teor. Fiz. 87, 369 (1984)].
7. R. Brustein and S. P. de Alwis, Phys. Rev. D 73, 046009 (2006).
8. S. Sarangi and S. H. Tye, arXiv:hep-th/0505104.
9. R. Holman and L. Mersini-Houghton, Phys. Rev. D 74, 043511 (2006).
10. For more details see: M. Bouhmadi-L6pez and P. Vargas Moniz, arXiv:hep-th/0612149.
11. A. Vilenkin, arXiv:gr-qc/9812027.
12. M. Bouhmadi-L6pez, L. J. Garay and P. F. GonzaJez-Diaz, Phys. Rev. D 66, 083504
(2002).
13. B. DeWitt, Phys. Rev. 160, 1113 (1967).
14. M. Bouhmadi-L6pez and P. Vargas Moniz, Phys. Rev. D 71, 063521 (2005);
M. Bouhmadi-Lopez and P. Vargas Moniz, AlP Conf. Proc. 736, 188 (2005).
 DARK ENERGY FROM QUANTUM WAVE FUNCTION
COLLAPSE OF DARK MATTER

ARCHAN S. MAJUMDAR
S. N. Bose National Centre for Basic Sciences,
Block JD, Sector III, Salt Lake, Kolkata 700098, India
[email protected]

D. HOME
Dept of Physics, Bose Institute, A.P. C. Road, Kolkata 700009, India
[email protected]

Dynamical wave function collapse models entail the continuous liberation of a specified
rate of energy arising from the interaction of a fluctuating scalar field with the matter
wave function. We consider the wave function collapse process for the constituents of
dark matter in our universe. Beginning from the era of electroweak phase transition in
the early universe the rate of the associated energy liberation is integrated to yield the
requisite magnitude of dark energy during the era of galaxy formation. The equation of
state for the liberated energy approaches w ---7 -1 at late times, providing a mechanism
to generate the present acceleration of the universe.
Since the advent of quantum mechanics and relativity over a century ago, there
have been profound advancements of our understanding of the microcosm as well
as the macrocosm. Yet certain deep mysteries have emerged in the study of par-
ticular aspects like the measurement problem l of the former, and the mechanism
for the presently accelerating universe 2 of the latter arena. Resolution of these two
fundamental challenges encountered by modern physics calls for introspection of
any possible domain of overlap between the solutions offered separately for either
of them. This work investigates a possible connection between these two problems.
The linear and unitary evolution enshrined in the basic equations of quantum
theory give rise to the quantum entanglement of elementary particles having po-
tentially widespread applications. However, it is difficult to realize quantum en-
tanglement over macroscopic scales. Several non-standard approaches 3 have been
suggested to explain the emergence of classicality and the related measurement
problem l of quantum theory. Among these schemes are the environment induced
decoherence approach, the consistent histories approach, the Bohmian ontological
interpretation, and the dynamical models of wave function collapse. 4 - 7
In dynamical collapse models the unitary Schrodinger evolution is modified by
stochastic nonlinear terms that affect the dynamics on time scales relevant to typical
macrophysical situations. The emergence of classicality and the ocurrence of single
outcomes in measurements is achieved by the interaction of the fluctuating modes
of a scalar field with the relevant particle wave functions. The scheme ensures that
the behaviour of microsystems stays in conformity with quantum mechanics and
there is no conflict with experiments. An associated liberation of energy occurs at
a rate proportional to the mass 5 of the collapsing wave packet, which for a system

1901
1902

with mass Ai is given by

dE 3fL2M
~~-=(3 (1)
dt 4m6a;T*
where mo is the proton mass, and a* = 10- 5 cm and T* = 10 16 S are the parameters
of the collapse models. 4 - 7 These values are chosen such that any superpositions
of the wave functions of microparticles are kept intact over the relevant distance
and time scales, consistent with all laboratory experiments. 6 Recent relativistic
generalizations of collapse models 7 satisfy the conservation of energy exchanged
between the scalar field and the collapsing matter.
In the cosmological arena one or more scalar fields have beem invoked to imple-
ment the inflationary paradigm in the early universe, and also to drive the presently
accelerating phase of the universe. 8 However, these models involve the construction
of "ad-hoc" potentials for the scalar field, and also fail to explain why the scalar field
energy density should start dominating during a narrow window of time beyond the
era of galaxy formation.
In the present analysis we consider a uniform distribution of mass in the comov-
ing volume R 3 , where the rate of energy liberation can be written as fft(pR 3 ) = (3,
with P as the density of the energy gained through wave function collapse. The fluc-
tuating scalar field ¢ driving collapse loses energy at the rate p¢ and conservation of
energy7 between the scalar field and the collapsing matter dictates that p¢R 3 = -(3.
The effect of the energy liberation due to the dynamical collapse of baryonic wave
functions 9 is subdominant to that for dark matter, and in the present analysis we
focus on the collapse of the dark matter. Replacing P by Pm (the energy density of
dark matter), one obtains
. 3 R (3
(2)
Pm + PmIi = R3
The integration of Eq.(2) over a time period (say, from ti to tend) gives the total
energy density of dark matter at the time tend. We set ti = tEW, assuming the
creation of dark matter along with baryonic matter at the time of the electroweak
phase transition in the early universe. The rate of energy liberation in Eq.(2) is not
valid beyond the time when most of the dark matter constituents have decoupled
from the background RW expansion by gravitational collapse. On setting tend =
tgalaxy (where tgalaxy is the epoch of galaxy formation), and integrating Eq.(2) one
obtains the total matter density at tgalaxy. Substituting the standard values of the
cosmological parameters, one gets Pm ~ 0.5pc + 0(10-22)pc, where Pc is the critical
energy density at tgalaxy. The energy liberated in dynamical collapse is thus unable
to alter the matter density.
The total energy liberated by the scalar field ¢ in driving the dynamical collapse
process is obtained by integrating the equation p¢ = -(3/ R3 during the time from
ti = tEW to tend = tgalaxy. The magnitude of the energy liberated by the field ¢ till
the era of galaxy formation is thus given by
_jtga1u,"y. 2(3tEQ (tgalaXY) 2( tEQ ) 1/2
(P*)galaxy = (-p¢dt) c::::' R3 -t- ~ (3)
tEW galaxy EQ EH
 1903

Substituting the standard values of tgalaxy = 10 16 8, tEQ = 10 11 8, and tEW =

10- 10 8, one obtains (P*)galaxy ~ (Pm)galaxy Therefore, the excess energy P* forms a
significant part ('" 50%) of the total energy density at the time of galaxy formation.
This energy liberated by the scalar field does not, of course, add to the matter
energy, but nonetheless does contribute to the expansion of the universe.
If the RW expansion takes place in the standard adiabatic manner, the total
energy density PT of all the constituents and the pressure P must satisfy the re-
It
lation (PT R 3 ) = -pit(R 3 ). Around the time of galaxy formation in the matter
dominated era, PT c::: Pm + P* (assuming any remnant energy residing in the scalar
field to be negligible compared to Pm and P*) and Pm ~ O. Hence, one obtains,
2/3
P = P. = - P* - 3R2 R (4 )
The second term in the Eq.(4) falls off as lit, and w == (pip) approaches the value
of -1 asymptotically, i.e., the equation of state for the "dark" energy at the time of
galaxy formation (tgalaxy) turns out to be P = P* = -P* - [(/3tgalaxy)/(R~alaxy)l c:::
-P* - O(10- 17 )p*. Hence, the liberated dark energy P* can generate the accelerated
expansion of the universe beyond the era of galaxy formation.
To summarize, the scheme presented here combines some essential features of
two hitherto unconnected solutions offered respectively for the quantum measure-
ment problem 1 and the dark energy problem. 2 Unlike in other models of dark energy,
construction of complicated potentials for the scalar fieldS is avoided, and the req-
uisite magnitude of dark energy with equation of state (w = -1) emerges beyond
the era of galaxy formation. The mathematical structure of the collapse models has
striking similarities 10 with that of certain other decoherence schemes. More detailed
calculations are needed to generalise this approach and develop it further taking into
account a variety of constraints from rapidly emerging astrophysical data. s

References
1. See, for example, A. J. Leggett, Science, 307, 871 (2005).
2. S. Perlmutter et al., Nature 391, 51 (1998).
3. D. Home, Conceptual Foundations of Quantum Physics, (Plenum, New York, 1997).
4. A. Bassi and G. C. Ghirardi, Phys. Rep. 379, 257 (2003).
5. L. Diosi, Phys. Rev. A40, 1165 (1989); ibid.42, 5086 (1990); G. C. Ghirardi, R. Grassi
and A. Rimini, Phys. Rev. A42, 1057 (1990).
6. P. Pearle and E. Squires, Phys. Rev. Lett. 73, 1 (1994); P. Pearle and J. Ring, Found.
Phys. 34, 1467 (2004).
7. P. Pearle, Phys. Rev. A59, 80 (1999); A. Bassi and G. C. Ghirardi, Phys. Rev. A65,
042114 (2002); P. Pearle, eprint quant-ph/0506177.
8. P. J. E. Peebles and B. Ratra, Rev. Mod. Phys. 75, 559 (2003).
9. D. Home and A. S. Majumdar, Phys. Lett. A220, 17 (1996).
10. L. Diosi, N. Gisin, J. Halliwell and 1. C. Percival, Phys. Rev. lett. 74, 203 (1995); J.
B. Hartle, Phys. Rev. A69, 042111 (2004).
 COSMOLOGICAL PERTURBATIONS IN QUANTUM
COSMOLOGICAL BACKGROUNDS

NELSON PINTO-NETO*
lCRA - Centro Brasileiro de Pesquisas Fisieas - CBPF,
rua Xavier Sigaud, 150, Urea, CEP22290-180, Rio de Janeiro, Brazil
[email protected]

We discuss the general properties of cosmological perturbations of quantum mechanical

origin in quantum cosmological models with a bonce. We obtain a suiable nonsingular
model were scalar and tensor perturbations have an almost scale invariant spectrum
which is naturally led to the standard cosmological model evolution before nucleosyn-
thesis.

In the theory of linear cosmological perturbations, simple evolution equations

for the perturbations have been obtained. l Lagrangians and hamiltonians describing
the dynamics of scalar, vector, and tensor perturbations coming from the Einstein-
Hilbert lagrangian have been greatly simplified in different cosmological scenarios
under the assumption that the background metric satisfies Einstein classical field
equations, and after taking out space and time total derivatives. l Once these simple
lagrangians and hamiltonians are obtained, the quantization of linear cosmological
perturbations becomes easy, with a quite simple interpretation: they can be seen
as quantum fields which behave essentially as scalar fields with a time dependent
effective mass. In this framework, one can assume an initial vacuum state for the
perturbations, yielding primordial perturbation spectra which can be compared with
observations. In the cosmological inflationary scenario,2 the resulting spectrum for
scalar perturbations is in good agreement with the data. 3
Some of the primordial cosmological backgrounds proposed in the literature
are quantum cosmological models which share the attractive properties of exhibit-
ing neither singularities nor horizons,4-6 leading the universe evolution through a
bouncing phase due to quantum effects, and a contracting phase before the bounce.
They constitute an example of a bouncing model, a possibility that has attracted
the attention of many authors,7 without the presence of any phantom field. These
features of the background introduce a new picture for the evolution of cosmological
perturbations: vacuum initial conditions may now be imposed when the Universe
was very big and almost fiat, and effects due to the contracting and bouncing phases,
which are not present in the standard background cosmological model, may change
the subsequent evolution of perturbations in the expanding phase. Hence, it is quite
important to study the evolution of perturbations in these quantum backgrounds to
confront them with the data. Howeever, one must face the problem of simplifying
the complicate hamiltonian of cosmological perturbations 8 without using the the
background classical Einstein's equations as long as the background models above
do not satisfy them.

*This research has been partially supported by CNPq of Brazil, CAPES (Brazil) and COFECUB
(France) for financial support.

1904
 1905

Recently, we have managed to put the hamiltonian of tensor and scalar perturba-
tions into a very simple form through the implementation of canonical transforma-
tions and redefinitions of the lapse function only, without recurring to any classical
equations of motion. 9 ,lD These simple lagrangians and hamiltonians can then be
used in situations where the background metric is also quantized, hence providing
a substantial simplification over the direct approach originally developed by Halli-
well and Hawking. Its consequences were explored in Ref. 1l •l2 This contribution is
summary of these results.
After performing Legendre and canonical transformations, redefining the lapse
function N with terms which do not alter the equations of motion up to first order,
all this without ever using the background equations of motion, the Hamiltonian up
to second order for scalar perturbations (the case of tensor perturbations is simpler
and similar, see9 ) is simplified to (see Ref.lD for details)

(1)

where

The quantities N, r/J, and A¢ play the role of Lagrange multipliers of the constraints
H6 D
) + H6 2 ) "'" 0, 7r'ljJ "'" 0, and 7r ¢ "'" 0, respectively. The momenta Pa , 7r ¢' 7r'ljJ, and
PT are conjugate respectively to the scale factor a, the Bardeen potentials r/J,?/J,
and the fluid degree of freedom T, this last variable playing the role of time. The
variable v is the usual Mukhanov-Sasaki variable.
In the Dirac quantization procedure, the first class constraints must annihilate
the wave functional x[N,a,r/J(xi),?/J(x),v(xi),T], yielding OX/or/J = 0, OX/o'ljJ = 0,
and H X = 0. The first two equations impose that the wave functional does not
depend on r/J and ?/J: as mentioned above, Nand r/J are, respectively, the homoge-
neous and inhomogeneous parts of the total lapse function, which are just lagrange
multipliers of constraints, and ?/J has been substituted by V(Xi), the unique degree
of freedom of scalar perturbations, as expected.
As PT appears linearly in H, and making the gauge choice N = a 3w , one can
interpret the T variable as a time parameter. Hence, the equation H X = assumes °
the Schrodinger form

~~x = ~ {a(3W-l)/2~ [a(3W-l)/2~] } X

aT 4 aa aa

(3)

where we have chosen the factor ordering in a in order to yield a covariant

Schrodinger equation under field redefinitions, and V and I have been absorbed
in redefinitions of the fields.
1906

If one makes the ansatz

x[a, v, T] = X(O)(a, T)X(2)[V, T] (4)

where X(O) (a, T) satisfies the equation,

o
ZoTX(o)(a,T) = 4"1 { a (3w - 1)/2 oa
0 [a (3w - 1)/2 0 ] } X(o)(a,T),
oa (5)

then we obtain for X(2)(a, v, T) the equation

o
ZoTX(2)(a,v,T) = -
a(3w-1)
2
J 3 52
d x 5v2X(2)(a,v,T)

+
wa(3w+1)
2
J 3 .
d XV"V,iX(2)(a,v,T) (6)

Solutions of the zeroth order equation (5) for a perfect fluid p = wp are known. 4 ,6
They are

8To ] 1/4 [ -4Toa 3(1-w) ]

X(O)(a, T) = [ 7r (T2 + TJ) exp 9(T2 + TJ)(1 _ W)2

exp {-i [9(T24:~;~lw~ w)2 + ~ arctan(~) - ~]}. (7)

with Bohmian trajectory13

T )2j3(l'-Wl
a(T) = ao [ 1 + ( To (8)

Note that this solution has no singularities and tends to the classical solution when
T ---> ±oo, and we are in the gauge N = a 3w .
This a(T) can be viewed as a given function of time in the second equation (5).
Going to conformal time d7] = a3w - 1dT, and performing the unitary transformation

(9)
the Schrodinger functional equation for the perturbations is transformed to

.OX(2)[V,7]]
Z 07] =
J 3
d x
( 1 52 W
-'2 5v 2 + "'2 V,i V'
i a" 2)
- 2a v X(2)[V, 7]]. (10)

The corresponding time evolution equation for the operator v in the Heisenberg
picture is given by
a"
v" - WV,i ,Z. - -av = 0, (11)

where a prime means derivative with respect to conformal time. In terms of the
normal modes Vk, the above equation reads

"
Vk + ( wk 2 - a")
--;; Vk = o. (12)
 1907

These equations have the same form as the equations for scalar perturbations
obtained in Ref. I (for one single fluid, the pump function Zll I z obtained in l is exactly
equal to a" la obtained here, if we make use of the background equations). The
difference is that the function a( T]) is no longer a classical solution of the background
equations but a quantum Bohmian trajectory of the quantized background, which
may lead to different power spectra.
Applying the solutions (8) to these equations and calculating the spectrum by a
matching procedure or numerically, one obtains (see Ref.12) for the power spectrum
Pip == 2k 3 1<1>12 lrr2 ex kns -1 the result
12w
ns = 1 + 1 + 3w· (13)

which is the same for gravitational waves (see RefY for details). Note that in the
limit w ---* 0 (dust) we obtain a scale invariant spectrum for both tensor and scalar
perturbations.
Numerically, taking into account the spectral index constraint as well as the
CMB normalization measure yields an equation of state that should be less than
w ;S 8 X 10- 4 , implying ns - 1 ~ 0 (10- 4 ), and that the characteristic size of the
Universe at the bounce is La ~ 10 3 lp), a region where one expects that the Wheeler-
De Witt equation should be valid without being spoiled by string or loop quantum
gravity effects. We have also obtained a consistency relation between the tensor to
scalar ratio TIS and the scalar spectral index as TIS ~ 4 x 10- 2y'n s -1, leading
to potentially measurable differences with inflationary predictions.
Concluding, using quantum cosmology and the Bohm interpretation, we have
obtained a simple bouncing model whose scalar and tensor perturbations can be
made arbitrary close to scale invariance. Perturbations begin in a vacuum state
during the contraction epoch, when the Universe was very large, dominated by dust,
and almost flat, and are subsequently evolved in a fully quantum way through its all
history up to the present expansion phase. Hence, we have presented a nonsingular
model without horizons, where perturbations of quantum mechanical origin can be
described all along and generate structures in the Universe.

References
1. V. F. Mukhanov, H. A. Feldman, and R. H. Brandenberger, Phys. Rep. 215, 203
(1992) .
2. V. Mukhanov and G. Chibisov, JETP Lett. 33, 532 (1981); A. H. Guth, Phys. Rev. D
23, 347 (1981); A. Linde, Phys. Lett. B 108,389 (1982); A. Albrecht and P. J. Stein-
hardt, Phys. Rev. Let. 48, 1220 (1982).
3. D. N. Spergel et al., Ap. J. Suppl. 148, 175 (2003).
4. J. Acacio de Barros, N. Pinto-Neto, and M. A. Sagioro-Leal, Phys. Lett. A 241, 229
(1998).
5. R. Colistete Jr., J. C. Fabris, and N. Pinto-Neto, Phys. Rev. D 62, 083507 (2000).
6. F.G. Alvarenga, J.C. Fabris, N.A. Lemos and G.A. Monerat, Gen.Rel.Grav. 34, 651
(2002).
1908

7. G. Murphy, Phys. Rev. D8, 4231 (1973); A. A. Starobinsky, Sov. Astron. Lett. 4, 82
(1978); M. Novello and J. M. Salim, Phys. Rev. D20, 377 (1979); V. Melnikov and
S. Orlov, Phys. Lett A70, 263 (1979); P. Peter and N. Pinto-Neto, Phys. Rev. D 65,
023513 (2002); P. Peter and N. Pinto-Neto, Phys. Rev. D 66, 063509 (2002); V.A. De
Lorenci, R. Klippert, M. Novello and J.M. Salim, Phys. Rev. D 65, 063501 (2002); J.
C. Fabris, R. G. Furtado, P. Peter and N. Pinto-Neto Phys. Rev. D 67, 124003 (2003).
8. J. J. Halliwell and S. W. Hawking, Phys. Rev. D 31, 1777 (1985).
9. P. Peter, E. Pinho, and N. Pinto-Neto, JCAP 07, 014 (2005).
10. E. J. C. Pinho, and N. Pinto-Neto, hep-th 0610192.
11. P. Peter, E. J. C. Pinho, and N. Pinto-Neto, Phys. Rev. D75 023516 (2007).
12. E. J. C. Pinho, P. Peter, and N. Pinto-Neto, hep-th 0610192.
13. See e.g. P. Holland, The Quantum Theory of Motion, Cambridge University Press
(Cambridge, UK, 1993) and references therein.
 CLASSICAL AND QUANTUM ASPECTS OF THE
INHOMOGENEOUS MIXMASTER CHAOTICITY

R. BENINI12t and G. MONTANI 23+

1 Dipartimento di Fisica - Universitii di Bologna and INFN
Sezione di Bologna, via Irnerio 46, 40126 Bologna, Italy
2 ICRA -International Center for Relativistic Astrophysics c/o Dipartimento di Fisica (Gg)
Universitii di Roma "La Sapienza"', Piazza A.NIoTa 500185 Roma, Italy
3 ENEA C.R. Frascati (U. T.S. Fusione), Via Enrico Fermi 45, 00044 Prascati, Roma, Italy
t riccardo. [email protected]
[email protected]

We refine Misner's analysis of the classical and quantum Mixmaster in the fully inhomo-
geneous picture; we both connect the quantum behavior to the ensemble representation,
both describe the precise effect of the boundary conditions on the structure of the quan-
tum states.
Near the cosmological singularity, the dynamics of a generic inhomogeneous cos-
mological model is reduced by an ADM procedure to the evolution of the physical
degrees of freedom, i,e. the anisotropies of the Universe. In fact asymptotically to
the Big-Bang the space points dynamically decouple 1 because the spatial gradients
of the dynamical variables become of higher order 2 and we can model the inhomo-
geneous Mixmaster via the reduced action:

1= f 3
d xdT (Pu 8 T u + PV 8 T V - E) , E = vJp~ + p~, (1)

The dynamics of such a model is equivalent to (the one of) a billiard-ball on a

Lobachevsky plane; this can be shown by the use of the Jacobi metric. 1 The man-
ifold described turns out to have a constant negative curvature, where the Ricci
scalar is given by R = -2/£2: the complex dynamics of the generic inhomogeneous
model results in a collection of decoupled dynamical systems, one for each point of
the space, and all of them equivalent to a billiard problem on a Lobachevsky plane.
We want to investigate the relation existing between the classical and the semiclas-
sical dynamics, and from this analysis we will derive the correct operator ordering
to be used when quantizing the system.
Let's write down the Hamilton-Jacobi equation for the system

(2)

This can be explicitly solved by separation of constants

(3)

The semiclassical analysis can be developed furthermore, and the stationary con-
tinuity equation for the distribution function can be worked in order to obtain
informations about the statistical properties of the model; as soon as we restrict

1909
1910

the dynamics to the configuration space, we get the following equation for the dis-
tribution function w

aw(u,v;k)
---'--::----'--+
~E)2
- -1
aw(u,v;k)
+
E2_2k 2v 2 w(u,v;k)
=0 (4)
au kv av kv 2 JE2 - (kv)2
This can be solved, and the exact distribution function can be obtained as soon as
we eliminate by integration the constant k

-
w(u v) =
J~ g(u+vJS~2-I) dk (5)
, _~ vy'E2 - k 2v 2

It is worth nothing how in the case g = canst, the micro canonical Liouville measure
wmc(u,v) = v~ is recovered.
We expect that the distribution function w( u, v) is re-obtained as soon as the quan-
tum dynamics is investigated to the first order in n. This can be easily done as a
WKB approximation to the quantum dynamics is constructed; as soon as we retain
only the lower order in n,
we obtain that: i) the phase S( u, v) coincides with the
Hamilton-Jacobi function, and ii) the probability density function r( u, v) obeys the
following equation:

2 2
ar(u,v)
k --'::--'----'-+ J(E)2
- - k2ar(u,v) + a(E22 - k v ) 2- 2E2 ruv
( . )_
-0. (6)
au v av v y'E2 - k v '

that coincides with (4) for a particular choice for the operator-ordering only, i. e.

v~2 P~2 -+ -IL

~2 - a (V 2 - a ) . (7)
v av av
With this result, the problem of the full quantization of the system can be taken
in consideration. The main problem is the presence of the root square function
in the definition of the Hamiltonian (1), but well grounded motivations exist 3 to
assume that the real Hamiltonian and the squared one have same eigenfunctions
and squared eigenvalues. This way the solution of the eigenvalue equation can be
obtained

w(u, v) = L an K s - 1 / 2 (2mrv) sin(2mru) (8)

n>O

The spectrum is obtained as Dirichelet boundary conditions on the domain r Q (see

Fig.I) are taken into account. The condition on the vertical lines can be imposed
exactly, but the one on the semicircle cannot be solved exactly; so we approximate
it as in Fig.2 with a straight line v = I/7T (it preserves the domain area /1 = 7T).
All these imply that s = 1/2 + t, with t E ~, and that the spectrum assumes the
following form

(9)
 1911

Fig. 1. The real domain rQ Fig. 2. The approximate domain

The values of the real parameter t have to be numerically evaluated by solving the
equation Kit(2n) = 0 for every natural n. This condition implies a discrete but
quite complicated shape for the spectrum.
Asymptotic expansions for high occupation numbers can be derived for different
regions of the parameters (t, n)4

References
1. R. Benini, G. Montani, Phys. Rev. D 70, 103527 (2004)
2. A. A. Kirillov, Zh. Eksp. Theor·. Fiz. 103,721-729 (1993)
3. R. Puzio, Class. Quantum Grav. 11, 609-620 (1994)
4. R. Benini, G. Montani, Class. Quantum Grav. in publication
 COSMOLOGICAL DYNAMICS WITH VACUUM POLARIZATION

A. V. TOPORENSKyt and P. V. TRETYAKOV+

Sternberg Astronomical Institute, Universitetsky prospekt, 13, Moscow 119992, Russia
t [email protected]
[email protected]

Theories of modified gravity may lead to rather unusual cosmological dynamics, including
existence of effective phantom branches. We study the conditions of de Sitter stability
with respect to vacuum polarization in modified gravity theories and show that de Sitter
solution is unstable on phantom branches.

Investigations of quantum effects in a strong gravitational field and their applica-

tions in cosmology have a long story. Since the beginning of 70-th many interesting
results modifying the standard Friedmann cosmology due to vacuum polarization
and particle production have been obtained. In particular, a vacuum de Sitter solu-
tion and an inflationary regime driven solely by vacuum polarization without any
matter 1 was described even earlier than a common scalar field inflationary sce-
nario. A detailed analysis of cosmology with vacuum polarization (we will consider
only this effect in the present communication) have been done many years ago. 2
After some period of stagnation, this problem begins to attract a considerable at-
tention last several years, mainly due to development of modified gravity models.
The form of extra terms in cosmological equations of motion caused by vacuum
polarization docs not depend on a particular theory of gravity, however, peculiar-
ities of a background metric in these theories could result in some new dynamical
regimes. Recently modifications caused by the vacuum polarization have been stud-
ied for regimes with soft future singularities. Such regimes, being impossible in the
standard cosmology, are rather typical in some modern cosmological scenarios,3 in
particular, they are present in induced gravity brane models. 4 Quantum correc-
tions change the dynamics significantly, leading to a softer singularity or even to
non-singular solutions. 5-7
Another interesting problem is stability of classical solutions with respect to
quantum corrections. As quantum terms contain higher time derivatives in com-
parison with corresponding classical equation of motion, some classical solutions
may become unstable. Recently this instability is described for certain regimes in
induced gravity brane cosmology.7,8
It is well known that the vacuum polarization in a flat FRW background leads
to the following vacuum expectation value for the energy density:2
4 " . 2 . 2
pq =< Too >= k2H + k3(2HH + 6HH - H ), (1)
where H is the Hubble parameter, k2, k3 depend upon the spin weight of the different
fields contributing to the vacuum polarization.
If the classical Friedmann equation has the form of algebraic dependence of the
Hubble parameter upon the energy density of the Universe H = H(p), substituting
p ---+ p + pq we get a differential equation which governs the cosmological evolution

1912
 1913

with quantum corrections.

We consider a general situation when matter density in the corresponding effec-
tive Friedmann equation is some algebraic function of the Hubble parameter H, so
the evolution equation hass the form:
p + () = F(H), (2)
where () is the cosmological constant which is necessary for existence of de Sitter
solution in a classical theory.
In the de Sitter point the matter density is diluted, so the equation becomes

() = F(H).
Now we add the quantum corrections (1). The k2 -term can be incorporated into
the algebraic functions

The new de Sitter point is given by

() = F(H).
The resulting second order differential equation for H can be written in the form
of a system of two first order equations
iI= C,
. c2 1 _
C = -3CH + 2H + 2Hk3 (F - a) = f(H, C).
There is an equilibrium point of this system (iI,6) = (0,0), which corresponds to
the de Sitter solution. In order to investigate stability of this equilibrium point we
need a linearized system:
. _ D1 D1
C - (DC loC + (DH )oH,
iI =C.
The eigenvalues of this linearized system have the form:

We can easily see that the real part of the first eigenvalue is negative if and only
if(gfr)o < O. The real part of the second eigenvalue is always negative (because
(¥C)o = -3H is negative in an expanding universe).
Let us now evaluate (gfr
)0:
8f -1 1 8F
(oH)o = 2H 2 k 3 [-() + F] + 2Hk 3 oH·
In the de Sitter stable point -() + F = 0, and we have finally
of 1 of
(oH)o = 2Hk3 (oH)·
1914

As a result, if Z;-;' > 0 (a normal branch), the de Sitter solution is stable if k3 < 0
(as in the standard cosmology). In the opposite case (a phantom branch) we need
k3 > 0 for stability.
We can conclude, that if we have k3 < 0 in our Universe (which is needed for
stability of the Minkowski space), all de Sitter points on phantom branches are
unstable.
Note, that we did not specify any particular modified gravity theory. All we need
is a function H(p). Of course, this analysis can not be applied to a situation when
equation which relates Hand p is a differential one (the most important example
of this case is a scalar field with non-zero potential playing the role of matter).

References
1. A. Starobinsky, Phys. Lett. B91, 99 (1980)
2. M.Fischetti, J.Hartle and B. Hu, Phys. Rev. D20, 1757 (1979).
3. J. Barrow, Class. Quant. Grav., 21, L79 (2004)
4. Yu. Stanov and V. Sahni, Class. Quant. Grav., L101 (2002).
5. S. Nojiri and S. Odintsov, Phys. Lett. B595, 1 (2004)
6. S. Nojiri, S. Odintsov and S. Tsujikawa, Phys. Rev. D71, 063004 (2005)
7. P.Tretyakov, A.Toporensky, Yu.Shtanov and V.Sahni, Class. Quant. Grav. 23, 3259
(2006).
8. A. Toporensky and P. Tretyakov, Grav. Cosmo!., 12, 55 (2006).
 SOME COSMOLOGICAL CONSEQUENCES OF
LOOP QUANTUM GRAVITY

DAVID J. MULRYNE
DAMTP, Centre for Mathematical Sciences,
University of Cambridge, Wilberforce Road, Cambridge, CBS OWA, UK.
d. [email protected]

REZA TAVAKOL
Astronomy Unit, School of Mathematical Sciences,
Queen Mary, University of London, Mile End Road, London, E1 4NS, U.K.
r. [email protected]

Among the most fundamental questions about the universe are those relating to its origin
and early evolution. Attempts at answering these questions within the classical relativis-
tic framework have failed so far due to the presence of initial singularities, where the laws
of physics break down. Recent results from Loop Quantum Cosmology, however, provide
a framework where these questions can be addressed. Here we briefly discusses this frame-
work and a number of recent results concerning early universe cosmology, focusing on the
connections between the semi-classical effects of Loop Quantum Cosmology, singularity
avoidance and inflation.

1. Introduction
Recent years have witnessed tremendous advances in cosmology from both the ob-
servational and theoretical perspectives. The high-precision data obtained by Cos-
mic Microwave Background (CMB) and high redshift surveys provide strong evi-
dence for a nearly spatially-fiat universe with a primordial spectrum of adiabatic,
gaussian and nearly scale-·invariant density perturbations. l These features are very
successfully described by the so called 'standard model' of cosmology with an early
infiationary phase. Despite these successes, however, a number of fundamental ques-
tions remain. These include (i) the absence of a fully successful model of infiation
that can be situated within a fundamental theory of quantum gravity and (ii) the
fact that the standard model based on Einstein general relativity leads to a singular
initial state for the universe where the classical theory breaks down.
It has long been expected that these questions can be resolved once a complete
theory of quantum gravity is known. At present there are two main contenders for
such a theory: string/M theory and Loop Quantum Gravity (LQG). Here we shall be
concerned with the latter and in particular with Loop Quantum Cosmology (LQC)
(i.e. applications of LQG to cosmological settings).
Within the framework of LQC, the evolution of the universe can be divided into
three distinct phases. An initial high energy/high curvature quantum phase, which
is described in terms of quantum difference equations; an intermediate semi-classical
phase, where the spacetime can be well approximated by the classical one, but in
which the dynamics of the early universe is modified by non-perturbative quantiza-
tion effects, and finally a classical phase, where the usual continuous cosmological
equations are recovered.
1915
1916

In this short review we briefly discuss some recent developments in LQC con-
cerning the questions outlined above. We focus on the intermediate semi-classical
regimes where the classical evolution equations are modified by LQG effects. In par-
ticular, we concentrate on homogeneous and isotropic settings and study some of
the effects of the quantization of the inverse volume in LQC. We shall outline three
major phenomenological consequences of these effect, and briefly discuss recent work
which employ them to construct novel early universe scenarios. For a discussion of
other settings, as well as the discrete quantum regime and a more thorough review
of recent advances, the reader is referred to the comprehensive review by Bojowald 2
and references therein. For further noteworthy recent work on quantization issues
and additional semi-classical effects the reader is pointed to Ref [3].

2. Effective field equations in loop quantum gravity

We consider a homogeneous and isotropic universe sourced by a scalar field ¢ with
potential V (¢). The semi-classical regime is characterized by the regime where
the scale factor lies in the range ai < a < a*. Below the scale ai, the discrete
nature of spacetime becomes important, whereas the standard classical cosmology
is recovered above a*. The dynamics in semi-classical LQC is generated by the
Hamiltonian constraint equation, which classically takes the form
(1)
where H¢ = ~a-3p~ + a 3V(¢) is the matter Hamiltonian for a scalar field. In LQG
there is a well defined prescription for the quantization of inverse quantities such
as the inverse volume a- 3 which appears in the scalar field Hamiltonian. 4 This
procedure, however, introduces a number of ambiguities, the most important of
which is the parameter j which is related to a* via a* == V'Tjj3l p l h ~ 0.274). In
this way j sets an effective scale below which quantum effects become important. j
takes half-integer values, but is otherwise unconstrained by theory, though smaller
values are considered to be more natural.
In the semi-classical regime the consequence of the quantization for the inverse
volume a- 3 in Eq. (1) is to replace it by a- 3D j (a), which is a continuous approxi-
mation to the quantized inverse volume. 4 This turns Eq. (1) into the semi·-classical
Hamiltonian constraint. The function Dj(a) is rather complicated and given by

Dj(a) = D(q) = (:7) 6 q3/2 {7 [(q + 1)11/4 _\q _1\11/4]

llq [(q + 1)7/4 _ sgn(q _ l)\q _ 1\7/4]} 6 (2)

where q = a2 ja;. Its main features, however, can easily be gleaned from its plot
which is given in Fig. 1. As can be seen, Dj(a) has the following important features:
(a) it approaches unity as a - ; 00, i.e, the inverse volume a- 3Dj(a) approximates
the classical expression a -3 for a » a* (b) it peaks around a* and importantly (c)
for a« a*, Dj(a) ex a 15 - ; 0 as a - ; O.
 1917

1.5r-----~----~----__,

D(q)

0.5

%~--~--~-----2~----~3
q

Fig. 1. Small-a behaviour of the function D(q).

The cosmological evolution equations in the semi-classical regime can then be

obtained using the semi-classical Hamiltonian constraint together with Hamiltonian
equations of motion. 5 The resulting modified Friedmann, Raychaudhuri and Klein-
Gordon equations are then respectively given by:

81[-[2
H2 = ~
3
(i2 + (1;) ) _ _ ,
_'P_
2D j
~T k
a2
(3)

if = _ 47rl~IJy2 (1- ~ +~
dlnD j )
(4)
Dj a a
6 dIn 2 '

¢= [-3+ dlnDj]H¢_DV" ,
dIna J ,<p
(5)

where the modifications to dynamics are completely parametrized by the function

D j (a). It is worth noting that dIn D j / d In a ranges between 15 for a « a* and 0 for

3. Some important cosmological consequences and scenarios

Studying the above semi-classical equations together with the properties of the
function D j (a), a number of LQC effects with consequences of potential importance
for the early can readily be seen. These include:

Super-inflation. An expanding universe undergoes a phase of super-inflationary

expansion, characterized by if > O. As can be seen from Eq. (4), this occurs when
the universe is sufficiently small for dIn D / d In a > 6 to be satisfied.
This effect was initially shown to occur in Ref. [5], and the unperturbed dynam-
ics further studied in Ref. [6], where it was shown that very large j values would be
required for this phase of inflation to solve the cosmological problems (Horizon, flat-
ness etc.) which initially motivated the introduction of inflation. Another important
question is whether the perturbations produced during this phase can have a near
scale-invariant spectrum to account for structure formation in the universe. Initial
1918

studies were optimistic,7 but more recent work, using more standard methods, are
less s08,9 (though a steep negative potential might allow a scale invariant spectrum
to be produced 8). So far, all the studies have necessarily neglected perturbations
in the background spacetime. Very recently, however, a framework has been devel-
oped which could allow full perturbed equations, including background effects, to
be studied.lO Utilizing this framework should allow more thorough understanding
of the consequences of the super-inflationary phase to be obtained.

Anti-friction. In an expanding universe the usual frictional term in the Klein-

Gordon equation (5), which acts to slow down the field, changes sign when
din D / d In a > 3 to become an anti-friction term. The importance of this effect,
originally noted in Ref. [5], lies in its ability to accelerate a scalar field up its po-
tential during the semi-classical phase. Then, once the semi-classical phase ends,
the field can roll back down, and if the potential is of the correct form, a phase of
standard slow-roll inflation can begin. Thus the anti-frictional effect in principle
provides a mechanism to set the initial conditions for inflation, even if the field
is initially situated at the minimum of its potential. It has been suggested that
this could lead to observational consequences on the CMBY The anti-frictional
effects and their sensitivity to quantization ambiguities have also been studied. 12
This study seems to indicate that in general the increase in kinetic energy during
the anti--frictional phase could compromise the validity of the semi-classical approx-
imation before the field has moved sufficiently far up its potential for 60 e-folds of
subsequent slow-roll inflation to occur. The general effect, however, was found to
be generic and robust to quantization ambiguities, though relatively large j values
are required. 12

Bounce. An important observation has been that LQC effects can induce a bounce
in a collapsing universe. 13 ,14 This can most easily be seen by rewriting the semi-
classical equations of LQC in terms of an effective perfect fluid with effective energy
density and pressure, and a variable equation of state thus: 14
2 87rl~1 k
H = -3-Peff - a 2 (6)
{Jeff = -3H (1 + w(a)) Peff (7)
where
1 ¢2
Peff == "2 D + V (8)

_
w(a) = Peff/Peff = -1 + . 2¢2 ( 1- -
1 din D)
-- . (9)
cjJ2 + 2DV 6 dIna
Now when dIn D / d In a > 6, the effective fluid behaves like a 'phantom' fluid with
(w < -1) violating the null energy condition independently of the form of the po-
tential, and thus removing an important obstacle to singularity avoidance. 14 During
a collapse, the universe's energy density starts to decrease once this condition has
 1919

been met. In a positively-curved universe, this allows the increasing magnitude of

the negative curvature term in Eq. (6) to cancel the positive energy density term
and for H to go to zero. At this point the contraction ceases and a bounce oc-
curs. A bounce is also possible for a flat universe if the scalar field's potential is
negative-definite. 14,15
The LQC induced bounce has arguably been the most dramatic of the three
effects discussed above, and allows the development of bouncing or cyclic scenar-
ios. 14 ,15 In particular, two closely connected and highly novel scenarios have been
constructed. These follow from the observation that in a positively-curved universe
with a massless scalar field, a bounce would be followed by an infinite set of iden-
tical oscillations. 14 These oscillations are associated with a center fixed point of
the LQC-modified dynamics,17 which is absent in the normal relativistic models.
Moreover, during these oscillations the scalar field moves monotonically in one di-
rection. If a potential is introduced into this model, and the field is initially at the
minimum of its potential, the behaviour is at first similar to that of the massless
case. Since the field moves monotonically, however, it gradually becomes more im-
portant. Once the field moves sufficiently far for the weak energy condition to be
violated, the oscillations end, the field turns around and, if the potential has the
correct form, slow-roll inflation occurs. This could provide a mechanism to set the
initial conditions for inflation, and only small j values are required. This behaviour
also allows a stable version of emergent universe scenario 16 to be developed. 17

References
1. D. N. Spergel et al., arXiv:astro-ph/0603449.
2. M. Bojowald, Living Rev. ReI. 8, 11 (2005).
3. A. Ashtekar, T. Pawlowski and P. Singh, Phys. Rev. Lett. 96, 141301 (2006); Phys.
Rev. D 73, 124038 (2006); Phys. Rev. D 74, 084003 (2006).
4. M. Bojowald, Phys. Rev. D 64, 084018 (2001).
5. M. Bojowald, Phys. Rev. Lett. 89, 261301 (2002).
6. M. Bojowlad and K. Vandersloot, Phys. Rev. D 67, 124023 (2003).
7. G. M. Hossain, Class. Quant. Grav. 22, 2511 (2005).
8. D. J. Mulryne and N. J. Nunes, Phys. Rev. D 74, 083507 (2006).
9. G. Calcagni and M. Cortes, Class. Quant. Grav. 24, 829 (2007).
10. M. Bojowald, H. H. Hernandez, M. Kagan, P. Singh and A. Skirzewski, Phys. Rev.
D 74,123512 (2006).
11. S. Tsujiawa, P. Singh and R. Maartens, Class. Quant. Grav. 21, 5767 (2004).
12. M. Bojowald, J. E. Lidsey, D. J. Mulryne, P. Singh and R. Tavakol, Phys. Rev. D
70,043530 (2004).
13. P. Singh and A. Toporensky, Phys. Rev. D 69, 104008 (2004).
14. J. E. Lidsey, D. J. Mulryne, N. J. Nunes and R. Tavakol, Phys. Rev. D 70, 063521
(2004); D. J. Mulryne, N. J. Nunes, J. E. Lidsey and R. Tavakol, Int. J. Mod. Phys.
A 20, 2347 (2005).
15. S. Tsujikawa, P. Singh and R. Maartens, Class. Quant. Grav. 21, 5767 (2004).
16. G. F. R. Ellis, R. Maartens, Class. Quant. Grav. 21, 223 (2004).
17. D. J. Mulryne, R. Tavakol, J. E. Lidsey and G. F. R. Ellis, Phys. Rev. D 71, 123512
(2005).
 SEMICLASSICAL SUPERSYMMETRIC QUANTUM GRAVITY

CLAUS KIEFER and TOBIAS LUCK

Institut fur Theoretische Physik, Universitiit zu Koln,
Zulpicher Str. 77, 50937 Koln, Germany
[email protected]
[email protected]

PAULO VARGAS MONIZ

Departamento de Fisica, Universidade da Beira Interior,
Rua Marques d'Avila e Bolama, 6200-307 Covilhfi, Portugal
[email protected]

We develop a semiclassical approximation scheme for the constraint equations of super-

symmetric canonical quantum gravity. This is achieved by a Born-Oppenheimer type of
expansion, in analogy to the case of the usual Wheeler-DeWitt equation. We recover at
consecutive orders the Hamilton-Jacobi equation, the functional Schrodinger equation,
and quantum gravitational correction terms to this Schrodinger equation. In particular,
our work has the following implications: (i) the Hamilton-Jacobi equation and therefore
the background spacetime must involve the gravitino, (ii) a (many fingered) local time
parameter has to be present on SuperRiem ~ (the space of all possible tetrad and grav-
itino fields), (iii) quantum supersymmetric gravitational corrections affect the evolution
of the very early universe.

1. Introduction

The essential feature of supersymmetric quantum cosmology (SQC) is that it sub-

scribes to the idea that treating both quantum gravity and supersymmetry (SUSY)
effects as dominant would give an improved description of the early universe. Such
an approach has profound consequences for the wave function of the universe: the
quantum state can be written as an expansion in linearly independent fermionic sec-
tors, each associated with a specific bosonic functional (of the same type as those
satisfying the Wheeler-DeWitt equation). Besides the pertinent question of how
to interpret such quantum states, investigating whether conserved currents and a
positive probability density can be obtained in this setting must be performed by
taking into account the enlarged structure of the wave function. Moreover, any cos-
mological evolution determined within SQC must eventually be consistent with a
mechanism for SUSY breaking. Nevertheless, in spite of all the progress achieved so
far, further efforts are required to find new states determining a consistent dynam-
ical path from a supersymmetric quantum cosmological to a classical cosmological
stage. 1
It is in the context of the above description that the purpose of the present work
can be seen: investigating the semiclassical approximation of supersymmetric canon-
ical quantum gravity. The viability of such a scheme is crucial for the framework of
quantum gravity in the presence of supersymmetry. In canonical quantum gravity,
a formal Born-Oppenheimer type of approximation scheme has been successfully
applied to the Wheeler-De Witt equation. 2 This has led in particular to the deriva-

1920
 1921

tion of quantum gravitational corrections terms which modify the limit of quantum
theory on a fixed background spacetime. 3- 5 Here we extend this formalism to the
supersymmetric case. We restrict ourselves to the theory of N = 1 supergravity in
four spacetime dimensions. Compared with the case ofthe bosonic Wheeler~DeWitt
equation, this leads to equations of the same type, bearing similarities as well as
various important differences arising from the presence of fermions (via SUSY).
The framework and results presented herewith may constitute an efficient means to
study the influence of SUSY in the physics of the very early universe.
For details of the calculations and the formalism as well as for further references
we refer to our previous article,6 on which the present contribution is based.

2. The semiclassical approximation

Here we sketch how the semiclassical approximation scheme can be applied to SQC.
We start with the algebra of quantum constraints 1 ,7 and take HAA' \[I = O. From
this constraint, the supersymmetric analogue of the Wheeler-DeWitt equation can
be derived. It reads
(HJ. + HT)\[I = 0. (1)
The part HJ. is obtained by the decomposition HAN = -nAA,HJ. + eAA,H i , where
nAA' and e~N are the spinorial versions of the timelike normal vector and the
triad, respectively; Hi and HJ. denote, respectively, the gravitational momentum
and Hamiltonian constraints. The matter part HT is assumed to be the Hamiltonian
density of a minimally coupled scalar field <P. For the wave functional we use an
ansatz of the form

(2)

where e and 'l/J denote the tetrad and the gravitino, respectively, and expand 5 into
a power series with respect to the gravitational constant G,
00

5[e, 'l/J, <p] = L5 n [e, 'I/J, <p]Gn - 1

. (3)
n=O
By means of this procedure, we then investigate the expansion of (1) in powers of
G.

3. Results and interpretations

The lowest order occuring in the expansion is G~2. In this order the independence
of 50 on the matter field <P is obtained, that is, 50 == 50 [e,I/J].
To handle the higher orders we introduce a supersymmetric extension of the
DeWitt metric, which we call 'Super-DeWitt metric'. This procedure is only essen-
tial for the treatment of the corrections to the Schrodinger equation. For the lower
orders it is useful as it allows us to formulate the results in a concise and short
manner.
1922

This metric is given on the space Super Riem ~, which is the direct sum of the
tetrad and the gravitino space. It contains vectors of the form (eAA'i, 1/;f) == qa·
The metric Q reads in block form:

(4)

The block B is pure bosonic and is essentially equivalent to the DeWitt metric on
Riem ~. The sector :F is purely fermionic, and the off-diagonal blocks Sl and S2
are mixed.
We continue with the semiclassical expansion scheme and find at the order C- 1
the Hamilton-Jacobi equation of the theory. Using the Super-DeWitt metric, it
assumes the form
1 oSo oSo
-Qab-- + A(So) - V = 0 (5)
2 oqa Oqb
with an operator A and a potential term V. 6 To discuss the solutions of the
Hamilton-Jacobi equation (5) we first remark that any solution So can be decom-
posed according to So[e, 1/;] = Bo[e] + Fo[e, 1/;]. Also the Hamilton-Jacobi equation
can be splitted with respect to Bo and Fo.6 A particular aspect distinguishing (5)
from its bosonic analogue 3 .4 is the presence of the gravitino, on which So generically
will depend. Moreover, we indirectly prove that So must depend on the gravitino.
It is known from the full theory that a pure bosonic solution, W[e], to the full set
of constraints cannot exist. 8 In fact, this argument can be extended in a straight-
forward way to each term in the semiclassical expansion. 6 The consequence is that
the Hamilton-Jacobi equation - and therefore the 'background spacetime' retrieved
from So - must necessarily involve the gravitino. For the pure bosonic case (that
is, for pure general relativity), a solution of the Hamilton-Jacobi equation yields
a classical spacetime which can serve as the appropriate background for higher or-
ders. 2 ,9,10 This is due to the fact that such a solution is equivalent to the field
equations originating from the Einstein-Hilbert action. This constitutes DeWitt's
interpretation: 9 every solution So describes a family of solutions to the classical
field equations. For every three-geometry there is one member of this family with a
spacelike hypersurface being equal to this three-geometry. But, as mentioned, the
semiclassical approximation of supersymmetric quantum gravity deviates notably
from this scenario: it contains as an additional ingredient the gravitino.
At the next order, Co, the Tomanaga-Schwinger (or local Schrodinger) equation
is found; it reads

.~r.
tn~ab~~
oSo oX
uqa uqb
+ ·~A X _=
21L
.~OX
tn~ =
uT
'1.JID
I L.lX . (6)

The wave functional X is defined by X = W[e, 1/;] exp(iS1 [e, 1/;, <f>]/n) , with a WKB-
like prefactor W.
Integrating (6) over space yields the functional Schrodinger equation. Further-
more, this equation defines the time functional T via the functional derivative. As in
 1923

canonical quantum gravity, this functional gives a 'many-fingered' time parameter

on Super Riem I;.
Proceeding with the semiclassical expansion scheme leads at the order G 1 to
corrections to the Schrodinger equation. The terms obtained by the expansion can
be handled by decomposing the coefficient S2[e, 1/;, <p] in a part containing the matter
field and a pure gravitational part, S2 [e, 1/;, <p] = a2 [e ,Ij!] + 7] [e, 1/;, <p]. Since the pure
gravitational phase a2 can be fixed, and we are mainly interested in the matter part,
we can restrict our attention to e == xexp (i7]Gjn), which contains the functional X
and the not yet determined part 7] of S2. The corrected local Schrodinger equation
up to the order of G 1 then reads

(7)

If the potential term V vanished, this would be the mathematical expression with
which we would have to work. In our case of non-vanishing V it makes sense to
decompose the correction terms into parts normal and tangential to hypersurfaces
So = constant 3 in Super Riem I;. In other words, we consider a trajectory of a
classical spacetime in configuration space and split it into a part in the direction of
the evolution and a part transverse to it. To perform this decomposition we use the
Super-DeWitt metric, which enables us to define what 'normal' means. Applying
this idea leads to the final result: the normal part of the corrected local Schrodinger
equation is given by

in
6e
157
= wne +
1..
c:
4VX
[(HT)2 _ i~V (6V
157
- (AV)) (HT - inA)

+z
'n 15 (HT - inA) _ 2(AW) ('1Jffi _ 'IlA)
157 W I L 1.. Z. +
3AHT _ 2A2]
in
e' (8)

where V is a slightly modified potential term. On a formal level, (8) is almost

exactly the result that has been obtained from the expansion of the Wheeler-DeWitt
equation. 3 ,4 However, there is a difference: the definition of the time functional is
different due to the involvement of the gravitino. But it can be seen that a vanishing
gravitino would yield exactly the same time functional as in the pure bosonic case.
In addition, this would lead to a vanishing operator A.6 Therefore, the 'bosonic
limit' of super symmetric quantum gravity yields up to the first order of correction
terms bosonic canonical quantum gravity. This is a strong argument for the overall
consistency of the supersymmetric theory.
It can be shown by a simple estimate 6 that the corrections in (8) become impor-
tant if we deal with a universe with a small scale parameter, that is, if we consider
the very early universe.
1924

Let us add that a violation of unitarity due to the purely imaginary terms cannot
be immediately concluded. This would require an inner product that we have not
defined here. 2 Furthermore, it is important that Eq. (8) is not independent of the
chosen factor ordering.

4. Outlook
We successfully applied the semiclassical approximation scheme to SQC. The results
could be the starting point for further research. In particular, the application of (8)
in the context of quantum cosmology and structure formation should be the topic of
future investigations. There are also questions arising from the introduction of the
Super-DeWitt metric, that is, the structure of SuperRiem I: has to be analyzed as
it has been done for Riem ~.9 One could also think of an extension of the work of
Gerlach lO with the aim to derive the complete set of equations of motion of N = 1
supergravity.

Acknowledgements
This work was supported by DAAD-GRICES/2004/2005 - D/03/40416, CRUP-AI-
A-21/2004 as well as the grants POCTI(FEDER) P - FIS - 57547/2004 and CERN
P - FIS - 49529/2003. P.V.M. is supported by the grant FCT (FEDER) SFRH -
BSAB 396/2003. He also wishes to thank Queen Mary College, London, for kind
hospitality during his sabbatical and expresses his warmest thanks to the Institut
fiir Theoretische Physik, Universitiit zu Kaln, for kind hospitality during his visit.
C.K. and T.L. acknowledge kind hospitality at the Universidade da Beira Interior,
Covilha.

References
1. P. V. Moniz, Int. J. Mod. Phys. A 11, 4321 (1996).
2. C. Kiefer, Quantum Gravity (Clarendon Press, Oxford, 2004).
3. C. Kiefer and T. P. Singh, Phys. Rev. D 44,1067 (1991).
4. C. Kiefer, in Canonical gravity: from classical to quantum, edited by J. Ehlers and H.
Friedrich (Springer, Berlin, 1994).
5. A. O. Barvinsky and C. Kiefer, Nucl. Phys. B 526, 509 (1998).
6. C. Kiefer, T. Liick, and P. Moniz, Phys. Rev. D 72, 045006 (2005).
7. P. D. D'Eath, Supersymmetric quantum cosmology (Cambridge University Press, Cam-
bridge, 1996); P. D. D'Eath, Phys. Rev. D 29, 2199 (1984).
8. S. M. CaroU, D. Z. Friedmann, M. E. Ortiz, and D. N. Page, Nucl. Phys B 423, 661
(1994).
9. B. S. DeWitt, Phys. Rev. 160, 1113 (1967); in Relativity, edited by M. Carmeli, S. 1.
Fickler, and L. Witten (Plenum Press, New York, 1970).
10. U. H. Gerlach, Phys. Rev. 177, 1929 (1969).
 MULTIGRAVITY AND SPACE TIME FOAM

REMO GARATTINI

\Ve consider a multigravity approach to spacetime foam. As an application we give indi-

cations on the computation of the cosmological constant, considered as an eigenvalue of
a Sturm-Liouville problem.

Universitii degli Studi di Bergamo, Facolta di Ingegneria,

Viale Marconi 5, 24044 Dalmine (Bergamo) ITALY.
INFN - sezione di Milano, Via Celoria 16, Milan, Italy
remo. [email protected]

Despite of its recent interest, the term "Multigravity" is not completely new.
Indeed, in the early seventies, some pioneering works appeared under the name
"strong gravity" or "f-g theory" 1 as a tentative to describe a sector of hadronic
physics where a massive spin-2 field (the f-meson with Planck mass !vI j cv 1 GeV)
plays a dominant role. Multigravity coincides with "strong gravity" or "f-g theory"
,vhen the number of gravitational fields involved is exactly 2 ("bigravity"). In this
paper, we would like to use the Multigravity idea as a model of space-time foam 2 to
compute the cosmological constant. Such a computation has been done looking at
the foam as a large N composition of Schwarzschild wormholes. 3 Nevertheless, the
Multigravity framework seems to be more appropriate for such a computation. We
recall that there exists a very large discrepancy between the recent estimates on the
cosmological constant, which approximately are of the order of 10- 47 Ge V 4 , while
a crude estimate of the Zero Point Energy (ZPE) of some field of mass m with a
cutoff at the Planck scale gives E ZPE ;::::: 10 71 GeV4 with a difference of about 118
orders.4 Let us see how to use multigravity, to approach such a problem. To this
purpose we begin with the following action involving N massless gravitons without
matter fields 5

So =
N
L
1=1
S [gil = L
N

1.=1
16~Gi Jd4xV-gi [R (gi) - Ai] , (1)

where Ai and G i are the cosmological constant and the related Newton constant
corresponding ith universe, respectively. Generally speaking, the total action should
be of the form
N

Stot = L S [gil + )..,Sint (gl, g2, ... , 9 N ) . (2)

;=1

When).., -+ O,the N world are non-interacting. This will be the context we are going
to examine. In this way, the action So describes a gas of gravitons. For N = 1,
introduce a time-like unit vector ull. such that u . u = -1, then it is possible to show
that

1 JV [gij] 1//* [YiJ] J2:, d3 xHI// [YiJ] ~ (1// IJ2:, d3 xA2:, 11//) A
(3)
V J V [YiJ ]1//* [gij] 1// [gij] V (1///1//;

1925
1926

where we have integrated over a hypersurface ~, divided by its volume and function-
ally integrated over quantum fluctuation with the help of some trial wave function-
also Note that Eq.(3) can be derived starting with the Wheeler-De Witt equation
(WDW)6 which represents invariance under time reparametrization. Eq.(3) repre-
sents the Sturm-Liouville problem associated with the cosmological constant. The
related boundary conditions are dictated by the choice of the trial wavefunction-
als which, in our case are of the Gaussian type. Different types of wavefunctionals
correspond to different boundary conditions. If we fix our attention on TT tensor,
then the one loop contribution assumes the following relevant expression

mt(r)
Pi (c) = - - -
2567r 2 c
[1 + In
2
(J..L
- -)
mT (r)
+ 21n 2 - 1] ,
-
2
(4)

i = 1,2, where we have used the zeta function regularization method to compute the
energy densities Pi and where we have introduced the additional mass parameter
J..L in order to restore the correct dimension for the regularized quantities. Such an
arbitrary mass scale emerges unavoidably in any regularization scheme. mT (r) play
the role of an effective mass. The energy density is renormalized via the absorption
of the divergent part (in the limit c ---> 0) into the re-definition of the bare classical
constant A ---> Ao + Adiv and the removal of the dependence on the arbitrary mass
scale J..L can be done by the use of the renormalization group-like equation 7

(5)

where p~h (J..L, r) is the renormalized energy density. Solving it we find that the
renormalized constant Ao should be treated as a running one in the sense that it
varies provided that the scale J..L is changing. We can now discuss the multigravity
gas. For each gravitational field introduce the following variables (N, Ni )(k) and
choose the gauge Ni(k) = 0, (k = 1 ... N w ). Define the following domain D~k)

{ WI [[(2~) Gijkl7rij7rkl - t! R] (k) WCk) [gg)] = - ~f~;) A~k)W(k) [92~)] 1}, (6)
Nw
and assume the following assumption: :=I a covering of ~ s.t. ~ = U ~k with ~k n
k=l
~j = 0 for k =1= j. Then the Eq.(3) turns into

_1_ J V [gij(k)] W*(k) [(k)]

gij
JEk d3 xAEk
~ (k) W(k) [ gij(k)]
(7)
V,
(k)
J V [(k)]
gij
* [(k)]
W(k) gij W(k) [gij(k)]

Each ~k has topology S2 x Rl. Therefore, the whole physical space ~ is a collection of
N w copies of the same world,s if the further assumption of having all the Newton's
constants equal is made. Thus, the final evaluation of the "global" cosmological
 1927

where A~: is the eigenvalue obtained evaluating Eq.(7) on each I: k . The computa-
tion of the max is taken on each disjoint I: k . Note that in any case, the maximum
of A~: corresponds to the minimum of the energy density computed on the related
hypersurface. It is interesting also to note that the whole procedure can be applied
even in case of a massive graviton with a term of the form 9 ,10
2
Sm = mg Jd 4 x
8""
R [hijhij] , (9)

which is a particular sub-case of the Pauli-Fierz termY

References
1. C.J. Isham, A. Salam, and J. Strathdee, Phys. Rev. D 3,867 (1971). A. Salam, and
J. Strathdee, Phys. Rev. D 16, 2668 (1977). C.J. Isham and D. Storey, Phys. Rev. D
18, 1047 (1978).
2. J.A. Wheeler, Ann. Phys. 2 (1957) 604; J.A. Wheeler, Geometrodynamics. Academic
Press, New York, 1962.
3. R. Garattini, Int. J. Mod. Phys. D 4, 635 (2002); gr-qc/0003090.
4. For a pioneering review on this problem see S. Weinberg, Rev. Mod. Phys. 61, 1
(1989). For more recent and detailed reviews see V. Sahni and A. Starobinsky, Int.
J. Mod. Phys. D 9, 373 (2000), astro-ph/9904398; N. Straumann, The history of the
cosmological constant problem gr-qc/0208027; T.Padmanabhan, Phys.Rept. 380, 235
(2003), hep-th/0212290.
5. T. Damour and 1.1. Kogan, Phys. Rev. D 66 104024 (2002).
6. B. S. DeWitt, Phys. Rev. 160, 1113 (1967).
7. J.Perez-Mercader and S.D. Odintsov, Int. J. Mod. Phys. D 1, 401 (1992).1.0. Chered-
nikov, Acta Physica Slovaca, 52, (2002), 221. 1.0. Cherednikov, Acta Phys. Polan. B
35, 1607 (2004). M. Bordag, U. Mohideen and V.M. Mostepanenko, Phys. Rep. 353,
1 (2001). Inclusion of non-perturbative effects, namely beyond one-loop, in de Sitter
Quantum Gravity have been discussed in S. Falkenberg and S. D. Odintsov, Int. J.
Mod. Phys. A 13, 607 (1998); hep-th 9612019.
8. R. Garattini, TSPU Vestnik 44 N7, 72 (2004); gr-qc/0409016 .
9. R. Garattini, J. Phys. Conf. Ser. 33215 (2006); gr-qc/0510062. R. Garattini, J. Phys.
A 39 6393, (2006 ); gr-qc/0510061.
10. V.A. Rubakov, Lorentz- Violating Gmviton Masses: getting around ghosts, low strong
coupling scale and VDVZ discontinuity. hep-th/0407104.
11. M. Fierz and W. Pauli, Proc. Roy. Soc. Land. A 173, 211 (1939).
 BOUNDARY CONDITIONS AND PREDICTIONS OF QUANTUM
COSMOLOGY *

DON N. PAGE
Theoretical Physics Institute, University of Alberta, Edmonton, Alberta, Canada T6G 2G7,
[email protected]

A complete model of the universe needs at least three parts: (1) a complete set of
physical variables and dynamical laws for them, (2) the correct solution of the dynamical
laws, and (3) the connection with conscious experience. In quantum cosmology, item (1)
is often called a 'theory of everything,' and item (2) is the quantum state of the cosmos.
Hartle and Hawking have made the 'no-boundary' proposal, that the wavefunction of the
universe is given by a path integral over all compact Euclidean 4-dimensional geometries
and matter fields that have the 3-dimensional argument of the wavefunction on their one
and only boundary. This proposal has had several partial successes, mainly when one
takes the zero-loop approximation of summing over a small number of complex extrema
of the action. However, it has also been severely challenged by an argument by Susskind.

1. Introduction
A complete model of the universe needs at least three parts:

(1) A complete set of physical variables (e.g., the arguments of the wavefunction)
and dynamical laws (e.g., the Schrodinger equation for the wavefunction, the
algebra of operators in the Hilbert space, or the action for a path integral.)
Roughly speaking, these dynamical laws tell how things change with time. Typ-
ically they have the form of differential equations.
(2) The correct solution of the dynamical laws (e.g., the wavefunction of the uni-
verse). This picks out the actual quantum state of the cosmos from the set
of states that would obey the dynamical laws. Typically a specification of the
actual state would involve initial and/or other boundary conditions for the dy-
namicallaws.
(3) The connection with conscious experience (e.g., the laws of psycho-physical
experience) These might be of the form that tells what conscious experience
occurs for a possible quantum state for the universe, and to what degree each
such experience occurs (i.e., the measure for each set of conscious experiences 1 ).

Item 1 alone is called by physicists a TOE or 'theory of everything,' but it

is not complete by itself. Even Items 1 and 2 alone are not complete, since by
themselves they do not logically determine what, if any, conscious experiences occur
in a universe.

*This research has been partially supported by Natural Sciences and Engineering Research Council
of Canada.

1928
 1929

2. The Hartle-Hawking Proposal for the Quantum State

Here I shall focus on Item 2, the quantum state of the cosmos, and in particular focus
on a proposal by Hawking 2 and by Hartle and Hawking 3 for this quantum state.
They have proposed that the quantum state of the universe, described in canonical
quantum gravity by what we now call the Hartle-Hawking wavefunction, is given
by a path integral over compact four-dimensional Euclidean geometries and matter
fields that each have no boundary other than the three-dimensional geometry and
matter field configuration that is the argument of the wavefunction.
In particular, the wavefunction for a three-geometry given by a three-metric
gij(X k ), and for a matter field configuration schematically denoted by ¢A(x k ), where
the three-metric and the matter field configuration are functions of the three spatial
coordinates xk (with lower-case Latin letters ranging over the three values {I, 2, 3}),
is given by the wavefunction

(1)

where the path integral is over all compact Euclidean four-dimensional geometries
that have the three-dimensional configuration [gij (xk), ¢A (xk) 1 on their one and
only boundary. Here a four-geometry are given by a four-metric gl"v(x Ct ), and four-
dimensional matter field histories are schematically denoted by ¢o(x Ct ), both func-
tions of the four Euclidean spacetime coordinates x Ct (with lower-case Greek letters
ranging over the four values {O, 1, 2, 3}).

3. Incompleteness of the Hartle-Hawking Proposal

The Hartle-Hawking 'one-boundary' proposal is incomplete in various ways. For ex-
ample, in quantum general relativity, using the Einstein-Hilbert-matter action, the
path integral is ultraviolet divergent and nonrenonnalizable. 4 This nonrenormaliz-
ability also occurs for quantum supergravity.5 String/M theory gives the hope of
being a finite theory of quantum gravity (at least for each term of a perturbation
series, though the series itself is apparently only an asymptotic series that is not
convergent.) However, in string/M theory it is not clear what the class of paths
should be in the path integral that would be analogous to the path integral over
compact four-dimensional Euclidean geometries without extra boundaries that the
Hartle-Hawking proposal gives when general relativity is quantized.
Another way in which the Hartle-Hawking 'one-boundary' proposal is incomplete
is that conformal modes make the Einstein-Hilbert action unbounded below, so the
path integral seems infinite even without the ultraviolet divergence. 6 If the analogue
of histories in string/M theory that can be well approximated by low-curvature
geometries have actions that are similar to their general-relativistic approximations,
then the string/M theory action would also be unbounded below and apparently
exhibit the same infrared divergences as the Einstein-Hilbert action for general
relativity. There might be a uniquely preferred way to get a finite answer by a
suitable restriction of the path integral, but it is not yet clear what that might be.
1930

A third technical problem with the Hartle-Hawking path integral is that one
is supposed to sum over all four-dimensional geometries, but the sum over topolo-
gies is not computable, since there is no algorithm for deciding whether two four-
dimensional manifolds have the same topology. This might conceivably be a problem
that it more amenable in string/M theory, since it seems to allow generalizations
of manifolds, such as orbifolds, and the generalizations may be easier to sum over
than the topologies of manifolds.
A fourth problem that is likely to plague any proposal for the quantum state of
the cosmos is that even if the path integral could be uniquely defined in a computable
way, it would in practice be very difficult to compute. Thus one might be able to
deduce only certain approximate features of the universe from such a path integral.
One can avoid many of the problems of the Hartle-Hawking path-integral, and
achieve some partial successes, by taking a 'zero-loop' approximation. 7

4. Partial Successes of the Hartle-Hawking Proposal

Despite the difficulties of precisely defining and evaluating the Hartle-Hawking 'one-
boundary' proposal for the quantum state of the universe, it has had a certain
amount of partial successes in calculating certain approximate predictions for highly
simplified toy models:

(1) Lorentzian-signature spacetime can emerge in a WKB limit of an analytic con-

tinuation. 2,3
(2) The universe can inflate to large size. 2
(3) Models can predict near-critical energy density.2,s
(4) Models can predict low anisotropies. 9
(5) Inhomogeneities start in ground states and so can fit cosmic microwave back-
ground data. 10
(6) Entropy starts low and grows with timeJl-13

5. Susskind's Objection to the Hartle-Hawking Proposal

Leonard Susskind 14-17 has argued that the cosmological constant or quintessence
or dark energy that is the source of the present observations of the cosmic accelera-
tion 1S ,19 would give a large Euclidean 4-hemisphere as an extremum of the Hartle-
Hawking path integral that would apparently swamp the extremum from rapid early
inflation. Therefore, to very high probability, the present universe should be very
nearly empty de Sitter spacetime, which is certainly not what we observe.
This argument is a variant of Vilenkin's old objection 20 that the no-boundary
proposal favors a small amount of inflation, whereas the tunneling wavefunction
favors a large amount. Other papers have also attacked the Hartle-Hawking wave-
function. 21 - 23 However, Susskind was the first to impress upon me the challenge to
the Hartle-Hawking no-boundary proposal from the recent cosmic acceleration.
Of course, it may be pointed out that most of de Sitter spacetime would not
have observers and so would not be observed at all, so just the fact that such
 1931

an unobserved universe dominates the path integral is not necessarily contrary to

what we do observe. To make observations, we are restricted to the parts of the
universe which have observers. One should not just take the bare probabilities for
various configurations (such as empty de Sitter spacetime in comparison with a
spacetime that might arise from a period of rapid early inflation). Rather, one
should consider conditional probabilities of what observers would see, conditional
upon their existence. 1.24 ,25
However, the bare probability of an empty de Sitter spacetime forming by a
large 4-hemisphere extremum of the Hartle-Hawking path integral dominates so
strongly over that of a spacetime with an early period of rapid inflation that even
when one includes the factor of the tiny conditional probability for an observer
to appear by a vacuum fluctuation in empty de Sitter, the joint probability for
that fluctuation in de Sitter dominates over the probability to form an inflationary
universe and thereafter observers by the usual evolutionary means. Therefore, the
argument goes, almost all observers will be formed by fluctuations in nearly empty
de Sitter, rather than by the processes that we think occurred in our apparently
inflationary universe.
The problem then is that almost all of these fluctuation-observers will not see
any significant ordered structures around them, such as the ordered large-scale
universe we observe. Thus our actual observations would be highly atypical in this
no-boundary wavefunction, counting as strong observational evidence against this
theory (if the calculation of these probabilities has indeed been done correctly). As
Dyson, Kleban, and Susskind put it in a more general challenge to theories with
a cosmological constant,15 "The danger is that there are too many possibilities
which are anthropically acceptable, but not like our universe." See 26 ,27 for further
descriptions of this general problem.
The general nature of this objection was forcefully expressed by Eddington 75
years ago: 28 "The cTude assertion would be that (unless we admit something which
is not chance in the architecture of the universe) it is practically certain that at any
assigned date the universe will be almost in the state of maximum disorganization.
The amended assertion is that (unless we admit something which is not chance in
the architecture of the universe) it is practically certain that a universe containing
mathematical physicists will at any assigned date be in the state of maximum
disorganization which is not inconsistent with the existence of such creatures. I
think it is quite clear that neither the original nor the amended version applies. We
are thus driven to admit anti-chance; and apparently the best thing we can do with
it is to sweep it up into a heap at the beginning of time."
In Eddington's language, Susskind's challenge is that the Hartle-Hawking no-
boundary proposal seems to lead to pure chance (the high-entropy nearly-empty
de Sitter spacetime), whereas to meet the challenge, we need to show instead that
somehow in the very early universe (near, if not at, the "beginning of time") it
actually leads to anti-chance, something far from a maximal entropy state.
For further details of Susskind's challenge, see my recent account. 29
1932

References
1. D. N. Page, quant-ph/9506010; Int. J. Mod. Phys. D5, 583 (1996), gr-qc/9507024; in
Consciousness: New Philosophical Perspectives, eds. Q. Smith and A. Jokic (Oxford,
Oxford University Press, 2003), pp. 468-506, quant-ph/0108039 in Universe or Mul-
tiverse?, ed. B. J. Carr (Cambridge University Press, Cambridge, 2007), pp. 401-419;
hep-th/0610101.
2. S. W. Hawking, in Astrophysical Cosmology: Proceedings of the Study Week on Cos-
mology and Fundamental Physics, September 28 - October 2, 1981, eds. H.A. Bruck,
C.V. Coyne, and M.S. Longair (Pontificiae Academiae Scientiarum Scripta Varia, Vat-
ican), pp. 563-574; Nucl. Phys. B239, 257 (1984).
3. S. W. Hawking and J. B. Hartle, Phys. Rev. D28, 2960 (1983).
4. M. H. Gorot! and A. Sagnotti, Nucl. Phys. B266, 709 (1986).
5. S. Deser, hep-th/9905017.
6. G. W. Gibbons, S. W. Hawking, and M. J. Perry, Nucl. Phys. B138, 141 (1978).
7. D. N. Page, in The Future of Theoretical Physics and Cosmology: Celebrating Stephen
Hawking's 60th Birthday, eds. G. W. Gibbons, E. P. S. Shellard, and S. J. Rankin
(Cambridge University Press, Cambridge, 2003), pp. 621-648, hep-th/0610121.
8. S. W. Hawking and D. N. Page, Nucl. Phys. B264, 185 (1986).
9. S. W. Hawking and J. C. Luttrell, Phys. Lett. B143, 83 (1984).
10. J. J. Halliwell and S. W. Hawking, Phys. Rev. D31, 1777 (1985).
11. S. W. Hawking, Phys. Rev. D32, 2489 (1985).
12. D. N. Page, Phys. Rev. D32, 2496 (1985).
13. S. W. Hawking, R. Laflamme, and G. W. Lyons, Phys. Rev. D47, 5342 (1993).
14. L. Susskind, private communication (2002).
15. L. Dyson, M. Kleban, and L. Susskind, J. High Energy Phys. 0210, 011 (2002), hep-
th/0208013.
16. N. Goheer, M. Kleban, and L. Susskind, J. High Energy Phys. 0307, 056 (2003),
hep-th/0212209.
17. L. Susskind, in Universe or Multiverse?, edited by B. J. Carr (Cambridge University
Press, Cambridge, 2007), pp. 241-260, hep-th/0302219.
18. S. Perlmutter et al. Astrophys. J. 483, 565 (1997), astro-ph/9608192; Nature 391, 51
(1998), astro-ph/9712212; Astrophys. J. 517, .565 (1999), astro-ph/9812133.
19. A. G. Riess et al. Astronom. J. 116, 1009 (1998), astro-ph/9805201; Astrophys. J.
560, 49 (2001), astro-ph/0104455; Astrophys. J. 607, 665 (2004), astro-ph/0402512.
20. A. Vilenkin, Nucl. Phys. B252, 141 (1985).
21. R. Bousso and A. Chamblin, Phys. Rev. D59, 063504 (1999), hep-th/9805167.
22. S. Gratton and N. Turok, Phys. Rev. D63, 123514 (2001), hep-th/0008235.
23. R. Bousso and J. Polchinski, J. High Energy Phys. 0006, 006 (2000), hep-th/0004134;
Sci. Am. 291, 60 (2004).
24. A. Vilenkin, Phys. Rev. Lett. 74,846-849 (1995), gr-qc/9406010.
25. S. W. Hawking and T. Hertog, Phys. Rev. D73, 123527 (2006), hep-th/0602091.
26. D. N. Page, hep-th/0610079; hep-th/0611158; hep-th/0612137.
27. R. Bousso and B. Freivogel, hep-th/0610132.
28. A. S. Eddington, Nature 127, 3203 (1931); reprinted in The Book of the Cosmos:
Imagining the Universe from Heraclitus to Hawking, ed. D. R. Danielson (Perseus,
Cambridge, Mass., 2000), p. 406.
29. D. N. Page, J. Cosmol. Astropart. Phys. (in press), hep-th/0610199.
 QUANTUM COSMOLOGY WITH NONTRIVIAL TOPOLOGY*

HELLO V. FAGUNDES
Instituto de Fisica TeoTica (IFT-UNESPj,
Sao Paulo, SP 01405-900, Bmzil.

TEOFILO VARGAS
Instituto de Ciencias Exatas, UniveTsidade Fedeml de Itajub6.
Itajubd, MG 37500-903, Bmzil.

Quantum creation of a universe with a nontrivial spatial topology is considered. Using

the Euclidean functional integral prescription, we calculate the wave function of such
a universe with cosmological constant and without matter. The minisuperspace path
integral is calculated in the semiclassical approximat.ion, and it is shown that in order to
include the nontrivial topologies in the path integral approach to quantum cosmology,
it seems necessary to generalize the sum over compact and smooth 4-manifolds to sum
over finite-volume compact 4-orbifolds.

1. Introduction

Whether our universe has a finite or infinite spatial extension is still an open ques-
tion related to that of the global topology of the universe. The Einstein field equa-
tions and the demand for homogeneity and isotropy deal only with local geometrical
properties, leaving the global topology of the universe undetermined. 1 In the stan-
dard cosmological framework, the universe is described by a Friedmann-Lamaitre-
Robertson-Walker (FLRW) solution, and the spatial sections are usually assumed to
be simply connected: hyperbolic, Euclidean spaces with infinite volumes and spheri-
cal one with finite volume. But each of these geometries can support many nontrivial
topologies with finite volumes, without altering the dynamics or the curvature.
Recently many works have appeared studying the possibility that the universe
may possess a compact manifold with nontrivial topology as spatial sections, and
also different methods have been proposed to detect or observationally constrain
these space topologies. Then a natural question appears: where did all kinds of
possible cosmic topologies corne from? Quantum cosmology aims at getting some
insight into this problem by treating the whole universe quantum mechanically. In
fact, it was argued by Gibbond and others2 that the global topology of the present
universe would be a relic of its quantum era, since the global topology would not
have changed under evolution after the Planck era.
In the pioneering works on the quantum creation of a closed universe, in both the
"tunneling from nothing" 3 and "no-boundary" 4 proposals, the solution of Einstein's
equations is a Riemannian 4-sphere S4 instanton joined across an equator to a
Lorentzian de Sitter space S3 at its minimal radius. A subsequent works on the
birth of a closed universe have been limited to the description of an S4 instanton

*This research has been supported by FAPESP and FAPEMIG

1933
1934

tunneling into a simply connected Lorentzian de Sitter spacetime R 161 S3. Here we
briefly report the case of a closed universe with s3/r spatial topology.5

2. Orbifold instanton and quantum cosmology

The notion of orbifold was first introduced by Satake,6 who used for it the term V-
manifold, and was rediscovered by Thurston,7 where the term orbifold was coined.
An orbifold is quotient space obtained by identify points in a manifold under discrete
symmetry group generated by translations, rotations and reflections. Let M be a
manifold and r be a dicrete group with an action r x M -> M. We say that r
acts freely, if for all x E M, IX = x implies I = 1, then the quotient space is
another manifold; but if some elements in r have fixed points, points for which
IX = x for any I E r with Ii-I, then the M/r is an orbifold 0. Here we will
consider an orbifold with boundary, which is defined in analogy to a manifold with
boundary: ao is obtained by taking orbifold charts (U, r, 7/)) where U is a connected
open subset of the positive half-space R+. = {(xl, x 2 , ... , xn) E R n : xn ~ O} and
lau au~ for each IE r.
Thurston's proposition leads to a classification of orbifolds into two types: If Mis
a smooth manifold with boundary and f a group that acts properly discontinuously
on M, then the quotient space M/f is an orbifold with boundary. An orbifold is
called good or global if it arises as a global quotient of a manifold by a properly
discontinouos group action. Otherwise the orbifold is called bad. If f is a discrete
subgroup of the isometries of either a spherical, a hyperbolic, or an Euclidean n-
manifold, then the quotient spaces sn If, H n If, and R n If will be respectively
called a spherical, a hyperbolic and a Euclidean good orbifold. A Riemannian metric
on an orbifold is obtained as in the manifold case, namely these quantities being
locally defined via coordinate charts, and then patched together through a partition
of the unity. The Levi-Civita connection, the curvature tensor and integration on
an orbifold are obtained in similar way. The main difference in the orbifold case is
that the structure involved must be invariant under the local group actions.
In Euclidean approach to quantum gravity on orbifolds, the basic quantity of
interest is the transition probability K(L,f, hf; L,i, hi,) from an initial three-orbifold
L,i to a final three-orbifold L, f. Neglecting a phase factor due to the nontrivial
topology of the orbifold, this propagator can be formally written as

K(L,f,hf;L,i,h i ,) = L
(O,g)
J
D(gfLV)exp[-SE(O,g)] , (1)

where the sum includes any 4-dimensional compact orbifold °

with metric g,
°
SE[O, g] is the Einstein-Hilbert action with cosmological constant on with disjoint
pair of boundaries ao = L,i UL, f. But we are interested in explaining the origin of
the universe, which we assume to be described by a de Sitter model during the infla-
tionary era and later by a FRW model. Since these models are based on manifolds
the final and initial 3-orbifolds should be manifolds, then the singular points are in
 1935

the interior of O. The important point is that these quotients of M by r will induce
a nontrivial topology on the boundary ao. The induced metric and the curvature
will have singularities at the fixed points of the orbifold. However, as observed by
Schleich and Witt,8 the character of the curvature singularity at these points is
dimension dependent; and in dimensions greater than two the integral leaves out
only a set of zero measure, so the Einstein-Hilbert functional is finite.
Now, Following the Hartle and Hawking's (HH) proposa1 4 of no-boundary
boundary condition, we will calculate the probability of creation of a closed universe
with 53/r spatial topology in the semiclassical approximation. In order to obtain
such a universe, it is necessary to begin with a more general instanton, namely 5 4 /r.
As shown by Fagundes at all. ,9 this instanton is an orbifold with two cone-like points,
which are the poles of 54 and are fixed points lO with respect to the nontrivial el-
ements of r. On this Euclidean closed de Sitter universe with 0(4) invariance, the
solution to the field equation will be almost a 4-sphere: aCT) : : : : 1IHcos(HNT).
The full spacetime solution is obtained attaching smoothly the lower half of this
orbifold across an equator to a Lorentzian de Sitter space.
Using the co moving volume of the space section given by V(5 3 If) = 27f 2 /Irl,
where Irl is the order of the group, the unnormalized probability of creation of this
multiply connected universe in semiclassical approximation for a < H- 1 is

IWHHI 2 = aexp (3Ir~;H2 [1- (1- H2a 2)3/2J) , (2)

and its analytical continuation for a > H- 1 is given by

2 [27f] [27f(H2 a2 - 1)3/2 7f] (3)
IWHHI = !3exp 31f/GH2 cos 31f/GH2 -"4.
Thus, we considered the probability of quantum creation of a closed universe with
nontrivial topology, and we see that in HH approach the probability of creation is
maximum for minimum order of r.

References
1. M. Lachieze-Rey and J.-P. Luminet, Phys. Rep. 254 (1995) 135; J. Levin, Phys. Rep.
254 (2002) 135.
2. L. Z. Fang and H. J. Mo, in Observational Cosmology, Proc. 12th Symp. lAU (Reidel,
Dordrecht, 1987); G. W. Gibbons and J.B. Hartle, Phys. Rev. D42 (1990) 2458.
3. D. Atkatz and H. Pagels, Phys. Rev. D25 (1982) 2065; A. Vilenkin, Phys. Lett. 1I7B
(1982) 25; A. D. Linde, Lett. Nuovo Cimento 39 (1984) 401; V. A. Rub akov , Phys.
Lett. 148B (1982) 280.
4. J. B. Hartle and S. W. Hawking, Phys. Rev. D28 (1983) 2960.
5. H. V. Fagundes and T. Vargas gr-qc/0611048
6. I. Satake, Math. Soc. Japan 9 (1957) 464.
7. W. Thurston, The Geometry and Topology of 3-Manifolds (1978).
8. K. Schleich and D. M. Witt, Phys. Rev. D60 (1999) 064013.
9. S. S. e Costa and H. V. Fagundes, Gen. Relat. Grav. 33 (2001) 1489.
10. P. Scott, Bull. London Math. Soc. 15 (1983) 401.
NON-SINGULAR SOLUTIONS IN LOOP QUANTUM COSMOLOGY

G.V. VERESHCHAGIN
ICRANet, P.le della Repubblica 10, 65100 Pescara, Italy
ICRA and Dip. Fisica, Univ. "Sapienza", P.le A. Mora 5,00185 Rome, Italy
[email protected]

We discuss non-singular solutions with scalar field in Loop Quantum Cosmology with
a number of effective potentials: massless, massive scalar field, nonlinear self-interacting
potential, cyclic, bicyclic potentials. General analysis with phase portrait technique is
presented. Such features of solutions as inflation and cyclic behavior are considered.

Loop quantum cosmology (LQC), at least within homogeneous scheme, does

not contain singularities. From the mathematical point of view this is because its
dynamical equations are no longer differential equations, like e.g. in the case of
Wheeler-deWitt theory, or classical General Relativity, but a difference equation.
From the physical point of view this reflects basic outcomes of the underlying theory,
Loop Quantum Gravity, namely that the spacetime is no longer a continuum at
the quantum level, but has a descrete structure. In this way surface and volume,
becoming operators in the quantum theory, do not possess zero eigenvalues.
In semi-classical approximation quantum difference equations reduce to differ-
ential equations, the spacetime may be represented as a continuum but with some
different structure unlike the classical spacetime. It is natural then, that at the
semi-classical level LQC leads to modifications of Friedmann equation. There are
two kinds of modifications found at present. From the one hand, there are so called
inverse volume operator corrections coming from the fact that the volume does not
vanish when the scale factor vanishes. These modifications appear as an anti-friction
term in the Klein-Gordon equation and lead to quite rich dynamics. From the other
hand quantum effects lead to appearance of quadratic on energy density term in
the Friedmann equation. This term, being negative definite, excludes occurence of
singularities. It turns out that at the bounce kinetic energy is dominating over the
potential one, which allows the scalar field to climb the potential wall and gener-
ate suitable initial conditions for inflation right after the bounce. Clearly for small
density (with respect to the Planck density) and large scale factor (with respect to
the value at the bounce) one recovers Friedmann cosmological equations.
Dynamics of solutions of semi-classical cosmological equations is in a remarkable
agreement with fully quantum dynamics. Being inspired by this fact we performed
qualitative analysis of cosmological equations with quantum modifications. Exten-
sive study of semi-classical equations with inverse volume operator corrections was
carried out in a number of papers. 1 Here we report briefly our main results of
the analysis of modified Friedmann equations contanining quadratic energy density
term. There are convincing indications that corrections of this type are the most
important at the bounce. 3
We considered single scalar field ¢ with energy density p and pressure p

1936
 1937

.
~

Fig. 1. Phase portrait for massive scalar field. Dashed curves represent GR case and solid curves
show LQC case. Dots correspond to points where initial conditions for numerical solutions were
given. The direction of phase trajectories is from the boundary to the origin.

1 ·2
P= 2¢ - V(¢),
where dot denotes time derivative, taking several forms of the effective potential
V (¢) like quadratic and fourth order potential, cyclic, bicyclic and negative poten-
tials, see 3 for details.
For the above definitions Friedmann equations with zero spatial curvature with
quantum modifications read

X (1- -p)
( -a0,)2 = -P - ,
3 PeTit

p + 3~ (p + p) = 0,
a
with x = 87fG, where G is Newton's constant,

V3
PeTit = 167f 2 ,3 Ppl,

where Planck's density is Ppl = 1/ (fiG 2 ) , , is Immirzi parameter, a is the scale

factor.
To illustrate performed analysis we provide at fig. 1 phase portrait for massive
scalar field at expansion stage (a > 0) on the plane (;p, ¢). Scale factor or its
1938

derivative, being required for the full analysis, depend uniquely on scalar field and
its derivative and may be found from the cosmological equations.
It is clear from the phase portrait that solutions deviate from their classical
couterparts only close to the boundary shown by thick cicrle which represents the
condition P = Perit· At the boundary bounce takes place so that expanding part of
solutions follows the corresponding contracting part. Phase portrait for contraction
stage may be obtained simply by reflecting the one shown at fig. 1 with respect to
the vertical axis and by inversion of directions of phase trajectories. The boundary
is common for both portraits.
There are attractors of solutions at expansion stage like in classical General
Relativity found for the first time in 4 reflecting occurence of inflation for most
solutions (assuming equal probability to have bounce with arbitrary ¢ and ¢).
We have considered also negative potentials and found cycling behavior of so-
lutions which is generic. For most solutions inflation is present making them asym-
metric and expanding after averaging over cycles.
The presence of negative definite quadratic on the energy density term in the
cosmological equations lead in general to occurance of turn-arounds. 3 In this way
problems of cyclic models like the absence of natural turn-around may be overcomed
within semi-classical LQC.
Thus, both inflationary and cyclic models may be naturally incorporated in
LQC.
Acknowledgement. I thank Parampreet Singh and Kevin Vandersloot for
many fruitful discussions.

References
1. M. Bojowald, G. Date, and K. Vandersloot, Class. Quantum Grav. 21 (2004) 1253;
P. Singh and A. Toporensky, Phys. Rev. D69 (2004) 104008; G.V. Vereshchagin, 1.
Cosmol. Astropart. Phys. 07 (2004) 013; G. Date, Phys. Rev. D71 (2005) 127502; M.
Bojowald, Phys. Rev. Lett. 95 (2005) 061301; M. Bojowald, R. Goswami, R. Maartens,
and P. Singh, Phys. Rev. Lett. 95 (2005) 091302; G. Date, Phys. Rev. D71 (2005)
127502; G. Date and G. M. Hossain, Phys. Rev. Lett. 94 (2005) 011302; A. Ashtekar
and M. Bojowald, Class. Quantum Grav. 23 (2006) 391; R. Goswami, P. S. Joshi, and
P. Singh, Phys. Rev. Lett. 96 (2006) 031302.
2. P. Singh and K. Vandersloot, Phys. Rev. D72 (2005) 084004.
3. P. Singh, K. Vandersloot and G.V. Vereshchagin, Phys. Rev. D74 (2006) 043510.
4. V. A. Belinsky, L. P. Grishchuk, 1. M. Khalatnikov, and Y. B. Zeldovich, Phys. Lett.
B155 (1985) 232.
 MINIMAL ENERGY AND FACTOR ORDERING
IN QUANTUM COSMOLOGY

F. HINTERLEITNER and R. STEIGL*

Department of Theoretical Physics and Astrophysics, Masaryk University,
Kotllifskd 2, 61137 Bmo, Czech Republic
* [email protected]
* www.muni.cz

This contribution investigates the influence of factor ordering in the hamiltonian con-
straint and the energy-momentum of the matter field on the wave function of homogenous
and isotropic cosmological models.

Keywords: Quantum cosmology; Friedmann models; Hamiltonian constraint; Critical

energy.

1. Introduction

Although this contribution is written in the framework of continuous standard quan-

tum cosmology, the motivation comes from loop quantum cosmology, where the
problem of the cosmological singularity is solved thanks to the discreteness of ge-
ometrical quantities, like the scale factor of the universe 1 or the volume. 2 As a
by-product of the considerations of Friedmann models with a scalar field as source
of gravity it was observed that there is a minimal or critical thrershold energy (of
the order of the Planck energy) of the source field for a dynamical wave function of
the model universe to arise. 3
The question we were mainly interested in is whether the critical energy is tied to
the discrete features of loop quantum cosmology or to some special factor ordering in
the Hamiltonian constraint operator. Furthermore, an advantage of the continuous
treatment of the Friedmann models is the possibility to handle the cases of positive,
zero, and negative spatial curvature in the same way (the relatively trivial spatially
flat case will be omitted here, 8ee4 ). For the problem of symmetry reduction before
quantization, see e.g. 5 The ingredients of our paper are the standard continuous
quantization of Friedmann models, a massless scalar field and the factor ordering
in the Hamiltonian constraint operator left open by introducing free factor ordering
parameters.

2. The Classical Model

We start with the usual Friedmann metric with time-dependent scale factor a(t),
which represents together with the minimally coupled homogenous field ¢(t) the two
degrees of freedom of the model. Their conjugate momenta are given by Pa = -12aa
and P¢ = 2a 3 Jy, respectively. The hamiltonian of the system is

'" P; 6c- 1 P~
H = - - - - - k a + - -3 (1)
24c a '" 4c a

1939
1940

with k = 0, 1, or -1 being the spatial curvature and", the gravitational constant.

Classical solutions may be given in terms of hypergeometric functions. 4

3. Quantization
Quantization is carried out by promoting the variables of the system to operators,
Pa = -i118/8a and P¢ = -i118/8¢, and writing down the Wheeler-DeWitt wave
equation H'ljJ = 0 with the aid of (1). The operator ambiguity in the first term is
expressed in a form containing three factor ordering parameters i, j, k,
Pa2 -iA _jA -k
(2)
----+ a Paa Paa
a
with i + j + k = 1 (i is a parameter, i = A). The Wheeler-DeWitt equation
1 ",11 2 1 8 1 8 1 1 112 1 8 2 C - )
----;------,--
( 24 c at 8a aJ 8a a k - - - - - - 6-ka 'ljJ(a,¢) = 0 (3)
4 c a3 8¢2 '"
can be solved by a product ansatz 'ljJ(a, ¢) = A(a)<I>(¢) with the ¢-part having the
eigensolutions <I> ( ¢) = e±p¢¢.
After a suitable transformation of variables the a-part assumes the form of a
Bessel equation (open case) or a modified Bessel equation (closed case) (for details,
see 4 ). In both cases the solution can be written in the form

where lp is the Planck length and Bv is a Bessel or modified Bessel function with
index v. The index, which determines the form of the wave function - oscillating
(dynamical) for imaginary and non-oscillating (non-dynamical) for real v - is a
combination of the field energy-momentum P¢ and the factor ordering parameter j.
In the supercritical case P¢ > Jfi(j + 1) we write v = iE.
We can also introduce an indefinite inner product of wave functions

("(.0/, cp ) = J
da
a
(0(.* 8cp
0/ 8¢
_ 8'ljJ* )
cp 8¢ . (5)
In the closed case this defines a finite norm, in the open case wave functions can be
normalized to delta-functions.
In the closed model the only normalizable supercritical functions Bv are modified
Hankel functions K iE , which have a reasonable semiclassical probability interpreta-
tion for large E, see. 4 The subcritical wave functions Kv decrease rapidly for growing
a, their behaviour at a = 0 (diverging, finite, or zero) depends on factor ordering.
For the open model we obtain in the supercritical case a superposition of the
Bessel functions J iE and J- iE and in the subcritical case a superposition of J v and
J- v . Again, only the former ones have a suitable classical probability interpretation
of the norm in the sense of (5) in the limit of large E. With a classical probability
interpretation we mean comparison with the classical probability density to find the
model universe at a certain value of a, given by 1/ iL, the inverse of the velocity of
change of the scale factor, written as function of a.
 1941

4. Conclusion, Examples
In both of the considered cases (and in the flat case, too) the wave functions can
be classified as "supercritical" = dynamical or "subcritical" = nondynamical. The
latter ones certainly do not represent a macroscopical universe, but they may be
classified w. r. to a further criterion: According to their behaviour at the classical
singularity a = 0 they mayor may not be normalizable in the sense of the above
inner product, again in dependence of Pcp and the factor ordering. It is unclear to
us whether subcritical but normalizable wave functions have a physical meaning.
Special factor orderings found in the literature are for example 'i = 2, j = -1,
k = 0, introduced as d'Alembert factor ordering,6 or i = k = 0, j = l.7 In the
former one v is imaginary for all values of Pcp, so there are no nondynamical wave
functions. This might be an argument in favour of this factor ordering. In the sec-
ond example the critical p¢ is equal to y7Z76 and all subcritical wave functions are
normalizable. For a nonvanishing cosmological constant the wave equation becomes
more complicated, the main qualitative features of the solutions, however, remain
unchanged. As a conclusion we have shown that the phenomenon of a "critical
energy-momentum" is of general nature, not restricted to a certain loop quantum
cosmological model. Further, not only a possible potential of cjJ, which was not con-
sidered here, plays an important role in the issue of normalizability and behaviour
at the singularity, but also the kinetic energy and factor ordering. Further work on
the interplay between field energy-momentum and factor ordering can be found, for
example, in the review. 8

Acknowledgements
Supported by the Ministry of Education of the Czech Republic, project MSM
0021622409 and Grant Agency of the Czech Republic, grant GACR 205/03/H144.

References
1. M. Bojowald, Isotropic Loop Quantum Cosmology, Clas. Quantum Grav. 19 2717-42
(2002).
2. M. Bojowald, Loop Quantum Cosmology, Liv. Rev. ReI. 8, 11 (2005), gr-qc/0601085.
A. Ashtekar, Proceedings of the Albert Einstein Century Conference, Paris, 2005, AlP
Conference Proceedings, Vol. 861, 3-13 (2006).
3. M. Bojowald and F. Hinterleitner, Loop quantum cosmology with matter, Phys. Rev.
D 66 104003 (2002), gr-qc/0207038.
4. R. Steigl and F. Hinterleitner, Factor ordering in standard quantum cosmology, Class.
Quantum Grav. 23 3879-93 (2006), gr-qc/0511149.
5. M.A.H. McCallum and A.H. Taub, Variational Principles and Spatially-Homogenous
Universes, Including Rotation. Commun. Math. Phys. 25 173-89 (1972).
6. N. Kontoleon and D. L. Wiltshire, Operator ordering and consistency of the wave
function of the Universe, Phys. Rev. D 59 063513 (1999).
7. A. Vilenkin, Phys. Rev. D 33 3560 (1986) and Phys. Rev. D 37 888 (1988).
8. D. Coule, Quantum Cosmological Models, Class. Quantum Grav. 22 R125 (2005). gr-
qc/0412026.
 ON THE FALSE VACUUM BUBBLE NUCLEATION

BUM-HOON LEE*'t,§, CHUL H. LEE:j:,~, WONWOO LEEt,11 and

CHANYONG PARKt,u
* Department of Physics, Sogang University, Seoul 121-742, Korea
t Center for Quantum Spacetime, Sogang University, Seoul 121-742, Korea
:j: Department of Physics, and BK21 Division of Advanced Research and
Education in Physics, Hanyang University, Seoul 133-791, Korea
§ E-mail: [email protected]
~ E-mail: [email protected]
II E-mail: [email protected]
** E-mail: [email protected]

We study how a false vacuum bubble can be nucleated within the true vacuum back-
ground via the true-to-false vacuum phase transition in curved spacetime. The mecha-
nism is not known in curved spacetime with arbitrary vacuum energy in the pure Einstein
theory of gravity. In this work we consider a semiclassical Euclidean bubble in the Ein-
stein theory of gravity with a nonminimaliy coupled scalar field.

1. Brief Summary of the Talk

What did the spacetime look like in the very early universe? Probably there was
a dynamical spacetime foam structure which was introduced by John A. Wheeler,
indicating that quantum fluctuations come into play at the Planck scale, changing
topology and metric 1. But such a phenomenon is difficult to describe in the theory
of gravity. On the other hand, there were investigations on the mechanism of creat-
ing the inflationary universe in the laboratory 2. However there is no regular method
for the nucleation of small regions of false vacuum. In this paper we show that such
complicated vacuum structure (spacetime structure) or false vacuum bubble can oc-
cur by the vacuum-to-vacuum phase transition in the semiclassical approximation.
In this paper, we summarize the results following the work presented in Ref. 3.
We have considered the true-to-false vacuum phase transitions in the five particular
cases; (Case 1) from de Sitter space to de Sitter space, (Case 2) from flat space to
de Sitter space, (Case 3) from anti-de Sitter space to de Sitter space, (Case 4) from
anti-de Sitter space to flat space and (Case 5) from anti-de Sitter space to anti-de
Sitter space. We have obtained numerical solutions. In case 1, we have obtained the
false vacuum smaller than the true vacuum horizon. The case 3 can be interesting
from the aspect of the so-called string landscape 4, which has included both anti-de
Sitter and de Sitter minima. Our solution represent how the false vacuum bubble,
corresponding to the de Sitter spacetime, can be nucleated within the true vacuum
background, corresponding to the anti-de Sitter spacetime.

1942
 1943

30

20

p('l) 15

10

10 15 30 35
'l 'l
(a) (b)

Figure 1. (a) The false vacuum bubble profiles for several values of E and t; in case 3. Here t;
is taken to be positive. The three curves are (i) solid curve: E = 0.01 and t; e:= 0.328, (ii) dotted
curve: E = 0.02 and t; e:= 0.414, (iii) dashed curve: E = 0.03 and t; e:= 0.508. (b) The evolution of
p(i]) in case 3. The solid curve is the solution of p with E = 0.01. In the region inside the bubble,
p = A sin X, and outside the bubble, p = A2 sinh j\~ •

We obtained the exponent B and the radius of the false vacuum bubble approx-
imately by employing Coleman's thin-wall approximation. The vacuum-to-vacuum
phase transition rate per unit time per unit volume is given by 5
(1)
where the coefficient B is given by

B~ '~;2 [dF { (l- '";lUF )'I' -l}

_(1 - Ur4b2~/'i;)2 { (1 _3(1/'i;p2Ur
-
)
4b2~/'i;)
3/2 _ I}] + 2 2p3 (SO _ ~~).
Jr
p2
(2)
The radius of the false vacuum bubble is given by
-2 H± VH2 -ED
p = E ' (3)
2
where E = [1 + 2(t;!? + (t;)4] + 8b2~/'i;3 A§Ur (8b 1UT - ~;), H = ~1 [(2 - 8b2~/'i;)
(~ - 4b2~Ur) + !;(UFi
UT
) (16b4~/'i; - 8b 2 - Soi")], D = ~~ [~(64b4~ + 2S3'C) -
256b 6 e/'i; - 8b
2e;oCK], and Po =
3So/[UF - Ur ] is the bubble size without gravity.
One can analyze the spacetime of inside and outside bubble using the metric
junction condition, which needs to be further studied.
In summary, we conclude that the false vacuum bubble can be nucleated within
the true vacuum background due to the term, -~REiI>, in the Einstein theory of
gravity with a nonminimally coupled scalar field and expect the phenomenon can
be possible in many other theories of gravity with similar terms.
1944

Acknowledgements
This work was supported by the Science Research Center Program of the Korea
Science and Engineering Foundation through the Center for Quantum Spacetime
(CQUeST) of Sogang University with grant number Rll - 2005- 021, Korea Re-
search Foundation grant number COOll1, and supported by the Sogang University
Research Grants in 2004.

References
1. J. A. Wheeler, Ann. Phys. 2, 604 (1957).
2. E. Farhi and A. H. Guth, Phys. Lett. B183, 149 (1987); E. Farhi, A. H. Guth, and J.
Guven, Nue!. Phys. B339, 417 (1990).
3. W. Lee, B.-H. Lee, C. H. Lee, and C. Park, Phys. Rev. D74, 123520 (2006).
4. L. Susskind, arXiv:hep-th/0302219.
5. S. Coleman, Phys. Rev. D15, 2929 (1977); ibid. D16, 1248(E) (1977); S. Coleman and
F. De Lueeia, Phys. Rev. D2I, 3305 (1980).
This page intentionally left blank