Conformal Hamiltonian Dynamics of General Relativity

A.B. Arbuzova , B.M. Barbashova, R.G. Nazmitdinova,b , V.N. Pervushina,∗,

A. Borowiecc, K.N. Pichugind, A.F. Zakharove
a Bogoliubov Laboratory of Theoretical Physics, Joint Institute for Nuclear Research, 141980
Dubna, Russia
b Department de Fı́sica, Universitat de les Illes Balears, E-07122 Palma de Mallorca, Spain
c Institute of Theoretical Physics, University of Wroclaw, Pl. Maxa Borna 9, 50-204

Wroclaw, Poland
d Kirensky Institute of Physics, 660036 Krasnoyarsk, Russia
e Institute of Theoretical and Experimental Physics, B. Cheremushkinskaya str. 25, 117259

Moscow, Russia
arXiv:1007.0293v1 [gr-qc] 2 Jul 2010

Abstract
The General Relativity formulated with the aid of the spin connection coeffi-
cients is considered in the finite space geometry of similarity with the Dirac
scalar dilaton. We show that the redshift evolution of the General Relativity
describes the vacuum creation of the matter in the empty Universe at the elec-
troweak epoch and the dilaton vacuum energy plays a role of the dark energy.
Keywords: General Relativity, Cosmology, dilaton gravity
PACS: 95.30.Sf 98.80.-k 98.80.Es

Submitted to Phys. Lett. B 14 April 2010; accepted 23 June 2002.

The distance-redshift dependence in the data of the type Ia supernovae [1]

is a topical problem in the standard cosmology (SC). As it is known, the SNeIa
distances are greater than the ones predicted by the SC based on the matter
dominance idea [2]. There are numerous attempts to resolve this problem with
a various degree of success (see for review [3]). One of the popular approaches is
the Λ-Cold-Dark-Matter model [4]. It provides, however, the present-day slow
inflation density that is less by factor of 10−57 than the fast primordial inflation
density proposed to include the Planck epoch.
Approaches to the General Relativity (GR) with conformal symmetry pro-
vide a natural relation to the SC [5]. The Dirac version [6] of the geometry
of similarity [7] is an efficient way to include the conformal symmetry into the
GR. In fact, the latter approach allows to explain the SNeIa data without the
inflation [8]. In the present paper, the Dirac formulation of the GR in the ge-
ometry of similarity is adapted to the diffeo-invariant Hamiltonian approach by
means of the spin connection coefficients in a finite space-time, developed in
[9]. In this way we study a possibility to choose variables and their initial data
that are compatible with the observational data associated with the dark energy

∗ Corresponding author
Email address: [email protected] (V.N. Pervushin)

Preprint submitted to Elsevier July 5, 2010

content. We find integrals of motion of the metric and matter fields in terms of
the variables distinguished by the conformal initial data.
Within the Dirac approach the Einstein-Hilbert action takes the form
R
WHilbert = − d4 x| − e| 16 R(e)
e=e−D e e
 
R 4 |−ee|e−2D −D µν −D


=− d x 6 R(ee) − e ∂µ | − e
e|e
g ∂ ν e . (1)

2 3
Hereafter, we use the units MPlanck 8π = 1. The interval is defined via diffeo-
µ
invariant linear forms ω(α) = e(λ)µ dx with the tetrad coefficients

ds2 = gµν dxµ dxν = ω(α) (d) ⊗ ω(β) (d)η(α)(β) ; η(α)(β) = Diag(1,−1,−1,−1). (2)

The geometry of similarity [6, 7] means the identification of measured physical

quantities F (n) , where (n) is the conformal weight, with their ratios in dilaton
units e−nD
Fe (n) = enD F (n) , e 2 = e2D ds2 .
ds (3)
We define the measurable space-time coordinates in the GR as the scale-invariant
quantities in the framework of the Dirac-ADM 4=1+3 foliation [10, 11]

e2 = ω
ds e(0) ⊗ ω
e(0) − ω
e(b) ⊗ ω
e(b) , (4)

e(α) = ee(α)µ dxµ are

where the linear forms ω

e(0) = e−2D N dx0 ,

ω (5)
j 0
ω
e(b) = ω(b) + e(b)j N dx , (6)
i
ω (b) = e(b)i dx . (7)

Here N is the Dirac lapse function, N j are the shift vector components, and
(3)
gij | ≡
e(b)i are the triads corresponding to the unit spatial metric determinant |e
|e(b)j e(b)i | = 1.
(3)
The Dirac dilaton D = −(1/6) log |gij | = hDi + D, is taken in the Lich-
nerowicz gauge [12]. The Dirac lapse function N = N0 (x0 )N (τ, x) is split on
the global factor N0−1 = hN −1 i which determines all time intervals used in the
observational cosmology: the redshift interval dτ = N0 dx0 [13], the conformal
one dη = dτ e−2hDi , and the world interval
R dt3= e
−hDi
dη = dτ e−3hDi . In this
−1
case the dilaton zeroth mode hDi = V0 V0
d xD (defined in the finite diffeo-
invariant volume) coincides with the logarithm of the redshift of spectral line
energy Em
hDi = log(1 + z) = log (Em (η0 − η)/Em (η0 )) , (8)
where η0 is the present-day conformal time interval, and η0 −η = r/c is the SNeIa
distance. In accord with the new Poincaré group classification, the ”redshift”
(8) is treated as one of the matter components, on the equal footing with the
matter.
The key point of our approach is to express the GR action directly in terms
of the redshift factor. The action can be represented as a sum of the dilaton

2
and the graviton terms:
" #
2
R
(∂0 hDi)
0 −2hDi
WHilbert = dx − + N0 e Lg , (9)
N0
" 2
#
R v(ab) R (3)
Lg = e2hDi d3 xN −(vD )2 + − e−4D . (10)
6 6

Here,
4
R(3) = R(3) (e) − e7D/2 △e−D/2 , (11)
3
is the curvature, where R(3) (e) is expressed via the spin-connection coefficients
1h j i
±
ω(ab) (∂(c) ) = e(a) ∂(c) ej(b) ± ei(b) ∂(c) ei(a) , (12)
2

and △ = ∂i [ei(a) ej(a) ∂j ] is the Laplace operator.

The dependence of the linear forms

ω (b) (d) = e(b)i dxi = dX(b) − X(c) ei(c) de(b)i (13)

R
on the tangent space coordinates X(b) ≡ dxi e(b)i = xi e(b)i by means of the
spin connection coefficients can be obtained by virtue of the Leibniz rule AdB =
d(AB)−(AB)d log(A) (in particular d[xi ]eTbi = d[xi eTbi ]−xi d[eTbi ]). The difference
between this approach to gravitation waves and the accepted one [14, 15] is that
the symmetry with respect to diffeomorphisms is imposed on spin connection
coefficients.
The linear graviton form (12) can be expressed via two photon-like polar-
(α)
ization vectors ε(a) (k). By virtue of the condition

X (α) (α) k(a) k(b)

ε(a) (k)ε(b) (k) = δ(a)(b) − , (14)
α=1,2 k(2)

one obtains
X eikX
+ + −
ω(ab) (∂(c) ) = √ k(c) [εR R
(ab) (k)gk (η) + ε(ab) (−k)gk (η)], (15)
2
2ω k
k 6=0

where εR ε (1) (k),

(ab) (k) = diag[1, −1, 0] in the orthogonal basis of spatial vectors [~
~ε (2) (k), k]. Here, ±
√ g are the holomorphic variables of the single degree of
2
freedom, ωk = k is the graviton energy normalized (like a photon in QED)
on the units of a volume and time
√
± 8π
gk = g±.
1/2 k
(16)
MPlanck V0

The triad velocities

h i
1 +
v(ab) = N ω(a)(b) (∂0 − N l ∂l ) + ∂(a) N(b)
⊥ ⊥
+ ∂(a) N(b) (17)

3
 + ⊥
depend on the symmetric forms ω(ab) , and the shift vector components ∂(b) N(b) =
0 are treated as the non-dynamical potentials. This means that the anti-
−
symmetric forms ω(ab) are not dynamically independent variables but are deter-
mined by a matter distribution.
Following Dirac [10, 16] one can define such a coordinate system, where the
covariant velocity vD of the local volume element and the momentum

2  
PD = 2vD = (∂0 − N l ∂l )D + ∂l N l /3 = 0 (18)
N
are zero. As a result, the dilaton deviation D can be treated as a static po-
tential. The dilaton contribution to the curvature (11) with matter sources
yield the Schwarzschild solution of classical equations △[exp{−7D/2}N ] = 0
and △ exp{−D/2} = 0. The solutions are exp{−7D/2}N = 1 + rg /(4r) and
exp{−D/2} = 1 − rg /(4r) in the isotropic coordinates of the Einstein interval
ds, where rg is the gravitation radius of a matter source. These solutions dou-
ble the angle of the photon beam deflection by the Sun field, exactly as the
Einstein’s metric determinant. Note that the GR theory provides also the New-
tonian limit in our variables (see details in [9]). Furthermore, in empty space
without a matter source (rg = 0), the mean field approximation (N = 1, D = 0,
N l = 0) becomes exact.
−
If there are no matter sources one can impose the condition ω(a)(b) = 0,
+
since the kinetic term (17) depends only on ω(ab) components. In this case the
curvature (11) takes the bilinear form
+ +
R(3) (e) = ω(ab) (∂(c) )ω(ab) (∂(c) ). (19)

The variation of the Hilbert action with respect to the lapse function leads
to the energy constraint [17]

(∂τ hDi)2 = ρcr ΩhDi + e−2hDi Hg /V0 , (20)

where the dilaton integral of motion ρcr ΩhDi is added, ρcr = H02 MPl
2
3/(8π) is
the critical density, and
 
R e−4D R(3)
Hg = e2hDi d3 xN 3p2(ab) + (21)
6

is the graviton Hamiltonian, p(ab) = v(ab) /3 is a canonical momentum (see

Eq.(17)).
Straightforward calculations define a set of evolution equations for the La-
grangian Lg (10) and the Hamiltonian Hg (21)

∂hDi Hg = 2Lg , (22)

−2hDi
∂hDi Tg = 2e Lg , (23)
∂hDi Lg = 2Hg − 2e−2hDi Tg , (24)
q
where Tg = H2g − L2g .
Note, the GR equations in terms of the spin-connection coefficients (22)-
(24) coincide with the evolution equations for the parameters of squeezing rb

4
and rotation θb [18]

∂hDi rb = cos 2θb , (25)

ωso − ∂hDi θb = coth 2rb sin 2θb (26)

of the Bogoliubov transformations A+ = B + cosh reiθ + B − sinh reiθ for a

squeezed oscillator (SO) ∂hDi A± = ±iωso A± + A∓ . Indeed, Eqs.(25),(26) estab-
lish similar relations for the expectation values of various combinations of the
operators A± with respect to the Bogoliubov vacuum B − | >= 0 (see details in
[17])

cosh 2rb − 1
Nb ≡< |A+ A− | > = −1
≡ ωso : Hb :, (27)
2
i sinh 2rb sin 2θb
< A− A− − A+ A+ > = −1
≡ ωso Tb , (28)
4 2
1 sinh 2rb cos 2θb
< A+ A+ + A− A− > = −1
≡ ωso Lb , (29)
4 2
On the other hand, Eqs. (10), (15), (19), and (21) show up that the graviton
action (9) has a bilinear oscillator-like form
X ωk + −
Hg = Hk , Hk = [g g + gk− g−k
+
],
2 k −k
k
X ωk + +
Lg = Lk , Lk = [g g + gk− g−k
−
], (30)
2 k −k
k
X iωk + +
Tg = T k, Tk= [g g − gk− g−k
−
],
2 k −k
k

where √
√ √
gk± = [g k ωk ∓ ipk / ωk ] / 2 (31)
are the classical variables in the holomorphic representation [15]. The form
(31) suggests itself to replace the variables gk± by creation and annihilation
graviton operators. Evidently, in this case we have to postulate the existence
of a stable vacuum |0i. As a consequence, it is reasonable to suppose that
the classical graviton Hamiltonian (see Eqs.(30)) is the quantum Hamiltonian
averaged over coherent states [19]. One may speculate that such procedure
reflects a transformation of a genuine quantum Hamiltonian (describing the
initial dynamics of the Universe) to the classical Hamiltonian, associated with
present-day dynamics.
Having the correspondence between two sets of equations (22)-(24) for the
GR and (27)-(29) for the SO, we are led to the ansatz that the SO is the quantum
version of our graviton Hamiltonian (see also [14]). This is a central point of
our construction. As a result, the normal ordering of the graviton Hamiltonian
yields
ωc
Hg = Hb =: Hb : + , Lg = Lb , Tg = Tb , (32)
2
where ωc = ωso e2hDi [17]. The normal ordering creates the Casimir–type vac-
uum energy ωc = 0.09235/(2rh) [20], where rh is the radius of the sphere defined
by the Hubble parameter.

5
 Figure 1: The creation of the Uni-
verse distribution [Nk = Nb ] (27) ver-
sus dimensionless time η and energies
0.5 ≤ ωk at the initial data Nk (η =
0) = 0 and the Hubble parameter
H(η) = 1/(1 + 2η) = (1 + z)2 .

The solution of Eqs. (22)-(24) is shown at Fig. 1. In accordance with this

solution, at the tremendous redshift 1 + z = ehDi = a−1 , z → ∞, a = 0,
Eq.(20) is reduced to the zeroth mode dilaton integral of motion ΩhDi which
corresponds to the z-dependence of the Hubble parameter H(z) = H0 (1 + z)2 .
At this moment, the Universe was empty, and all particle densities had the zero
initial data. The same dilaton vacuum regime H(z) = H0 (1 + z)2 is compatible
with the SNeIa data [1] in the geometry of similarity (3) [8].
The next step is the creation of gravitons induced by the direct dilaton
interaction. A hypothetic observer being at the first instance at rI = 1/HI
in the primordial volume VI = 4πrI3 /3 observes the vacuum creation of these
particles with the primordial density
H02 8
ΩgI = ωc · 2 · (1 + zI ) (33)
MPl.
defined by the Casimir energy. The question which remains to answer is how to
define zI ?
In order to estimate the instance of creation (1 + zI ), one can add the Hamil-
tonian of the Standard Model (SM): Hg → H = Hg + HSM ,q – when in the limit
(1+zI ) → ∞ and a → 0 all particles become nearly massless k2 + a2 M02 → ωk .
In this case, the same mechanism of intensive particle creation works also for
any scalar fields including four Higgs bosons [21]
ΩI Higgs = 4ΩgI . (34)
The decays of the Higgs sector including longitudinal vector W and Z bosons
approximately preserve this partial energy density for the decay products. These
products are Cosmic Microwave Background (CMB) photons and nν neutrino.
Therefore, one obtains
(1 + nν )ΩCMB ≈ 4ΩgI . (35)
In our model there is the coincidence of two epochs:
• the creation of SM bosons in the Universe in electroweak epoch
1 + zW = [MW /H0 ]1/3 = 0.37 · 1015 , (36)

6
 when the horizon H(zW ) = (1 + zW )2 H0 = (1 + zW )2 1.5 · 10−42 GeV
contains only a single W boson;
• and the CMB origin time

1 + zCMB = [λCMB H0 ]−1/2 = [10−29 · 2.35/1.5]1/2 = 0.39 · 1015 , (37)

when the horizon contains only a single CMB photon with mean wave
length λCMB that is approximately equal to the inverse temperature λ−1
CMB =
TCMB = 2.35 · 10−13 GeV.
In the same epoch zI ≈ zW ≈ zCMB , if the primordial graviton density (33)
coincides with the CMB density normalized to a single degree of freedom (as
it was supposed in [14]). The coincidence of the Planck epoch zI with the first
two ones solves cosmological problems with the aid of the geometry of similarity
(3), without the inflation (see also [8]).
While adding the SM sector to the theory in order to preserve the conformal
symmetry, we should exclude the unique dimensional parameter from the SM
Lagrangian, i.e. the Higgs term with a negative squared mass. However, fol-
lowing Kirzhnits [22], we can include the vacuum expectation of the Higgs field
(its zeroth harmonic) hφi. The latter appears as a certain external initial data
or a condensate. In our construction we can choose it in the most simple form:
hφi = Const = hφiI = 246 GeV which could be consider as the initial condition
at the beginning of the Universe. The fact, that the Higgs vacuum expectation
is equal to its present day value, allows us to preserve the status of the SM
as the proper quantum field theory during the whole Universe evolution. The
standard vacuum stability conditions

< 0|0 > |φ=hφi = 1, < 0|0 >′ |φ=hφi = 0 (38)

yield the following constraints on the Coleman–Weinberg effective potential of

the Higgs field:
′
Veff (hφi) = 0, Veff (hφi) = 0. (39)

It results in a zero contribution of the Higgs field vacuum expectation into

the Universe energy density. In other words, the SM mechanism of a mass
generation can be completely repeated. However, the origin of the observed
conformal symmetry breaking is not a dimensional parameter of the theory but a
certain non-trivial (and very simple at the same moment) set of the initial data.
In particular, one obtains that the Higgs boson mass is determined from the
′′ 2
equation Veff (hφi) = MH . Note that in our construction the Universe evolution
is provided by the dilaton, without making use of any special potential and/or
any inflaton field. In this case we have no reason to spoil the renormalizablity
of the SM by introducing the non-minimal interaction between the Higgs boson
and the gravity [23].
In summary, following the ideas of the conformal symmetry [6, 7], we for-
mulated the GR in terms of the spin-connection coefficients. The cosmological
evolution of the metrics is induced by the dilaton, without the inflation hy-
pothesis and the Λ-term. In the suggested model, the Planck epoch coincides
with the thermalization and the electroweak ones. In this case the CMB power

7
spectrum can be explained by two gamma processes of SM bosons [24], avoid-
ing dynamical dilaton deviations with negative energy by means of the Dirac
constraint (18). We have provided a few arguments in favour that the exact
evolution of the GR as a theory of spontaneous conformal symmetries breaking
is related to the equations for the quantum squeezed oscillator. We found that
the dilaton evolution yields the vacuum creation of matter.

Acknowledgments
The authors thank D. Blaschke, K. Bronnikov, D.V. Gal’tsov, A.V. Efre-
mov, N.K. Plakida, and V.B. Priezzhev for useful discussions. V.N.P. thanks
Yu.G. Ignatev and N.I. Kolosnitsyn for the discussion of experimental conse-
quences of the General Relativity.

References

[1] A. G. Riess et al., Astron. J. 116 (1998) 1009; S. Perlmutter et al., Astro-
phys. J. 517 (1999) 565; P. Astier et al., Astronomy and Astrophysics 447
(2006) 31.
[2] A. Einstein and W. de-Sitter, Proc. Nat. Acad. of Scien. 18 (1932) 213.
[3] A. D. Linde, Lect. Notes Phys. 738 (2008) 1.
[4] M. Giovannini, Int. Jour. Mod. Phys. D14 (2005) 363.
[5] D. Grumiller, W. Kummer, and D. V. Vassilevich, Phys. Rep. 369 (2002)
327.
[6] P. A. M. Dirac, Proc. Roy. Soc. A333 (1973) 403.
[7] H. Weyl, Sitzungsber. d. Berl. Akad., 465 (1918).
[8] D. Behnke et al., Phys. Lett. B530 (2002) 20; A.F. Zakharov and V.N. Per-
vushin, arXiv:1006.4745 [gr-qc].
[9] B. M. Barbashov et al., Phys. Lett. B633 (2006) 458; Int. Jour. Mod. Phys.
A21 (2006) 5957; Int. J. Geom. Meth. Mod. Phys. 4 (2007) 171.
[10] P. A. M. Dirac, Proc. Roy. Soc. A246 (1958) 333; Phys. Rev.114 (1959)
924.
[11] R. Arnowitt, S. Deser, and C. W. Misner, The dynamics of general rel-
ativity, in L. Witten, Gravitation: An Introduction to Current Research
(Wiley, New York, 1962) pp.227-265.
[12] A. Lichnerowicz, Journ. Math. Pures and Appl. B37 (1944) 23.
[13] C. Misner, Phys. Rev. 186 (1969) 1319.
[14] L. P. Grishchuk, Sov. Phys. Usp. 20 (1977) 319.
[15] V.N. Pervushin and V.I. Smirichinski, J. Phys. A32 (1999) 6191.
[16] L. D. Faddeev and V. N. Popov, Sov. Phys. Usp. 16 (1974) 777.

8
[17] A. F. Zakharov, V. A. Zinchuk, and V. N. Pervushin, Phys. Part. Nucl. 37
(2006) 104.
[18] L. Parker, Phys. Rev. 183 (1969) 1057.
[19] J. P. Blaizot and G. Ripka, Quantum Theory of Finite Systems (The MIT
Press, London, 1986).
[20] J. Schwinger, L. DeRaad, and K. A. Milton, Ann. Phys. 115 (1979) 1.
[21] V.N. Pervushin, Acta Phys. Slov. 53 (2003) 237; D. B. Blaschke et al.,
Phys. Atom. Nucl. 67 (2004) 1050.
[22] D. A. Kirzhnits, JETP Lett. 15 (1972) 529; A. D. Linde, JETP Lett. 19
(1974) 183.
[23] F. L. Bezrukov and M. Shaposhnikov, Phys. Lett. B659 (2008) 703.
[24] A. B Arbuzov et al., Physics of Atomic Nuclei 72 (2009) 744.