Springer Theses

Recognizing Outstanding Ph.D. Research

Amin Kianinejad

Metamaterial
Surface
Plasmon-Based
Transmission Lines
and Antennas
Springer Theses

Recognizing Outstanding Ph.D. Research

Aims and Scope

The series “Springer Theses” brings together a selection of the very best Ph.D.
theses from around the world and across the physical sciences. Nominated and
endorsed by two recognized specialists, each published volume has been selected
for its scientific excellence and the high impact of its contents for the pertinent field
of research. For greater accessibility to non-specialists, the published versions
include an extended introduction, as well as a foreword by the student’s supervisor
explaining the special relevance of the work for the field. As a whole, the series will
provide a valuable resource both for newcomers to the research fields described,
and for other scientists seeking detailed background information on special
questions. Finally, it provides an accredited documentation of the valuable
contributions made by today’s younger generation of scientists.

Theses are accepted into the series by invited nomination only

and must fulfill all of the following criteria
• They must be written in good English.
• The topic should fall within the confines of Chemistry, Physics, Earth Sciences,
Engineering and related interdisciplinary fields such as Materials, Nanoscience,
Chemical Engineering, Complex Systems and Biophysics.
• The work reported in the thesis must represent a significant scientific advance.
• If the thesis includes previously published material, permission to reproduce this
must be gained from the respective copyright holder.
• They must have been examined and passed during the 12 months prior to
nomination.
• Each thesis should include a foreword by the supervisor outlining the signifi-
cance of its content.
• The theses should have a clearly defined structure including an introduction
accessible to scientists not expert in that particular field.

More information about this series at http://www.springer.com/series/8790

Amin Kianinejad

Metamaterial Surface
Plasmon-Based Transmission
Lines and Antennas
Doctoral Thesis accepted by
the National University of Singapore, Singapore

123
Author Supervisor
Dr. Amin Kianinejad Prof. Zhi Ning Chen
Department of Electrical and Computer Department of Electrical and Computer
Engineering Engineering
National University of Singapore National University of Singapore
Singapore Singapore
Singapore Singapore

ISSN 2190-5053 ISSN 2190-5061 (electronic)

Springer Theses
ISBN 978-981-10-8374-7 ISBN 978-981-10-8375-4 (eBook)
https://doi.org/10.1007/978-981-10-8375-4
Library of Congress Control Number: 2018931493

© Springer Nature Singapore Pte Ltd. 2018

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt from
the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, express or implied, with respect to the material contained herein or
for any errors or omissions that may have been made. The publisher remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.

Printed on acid-free paper

This Springer imprint is published by Springer Nature

The registered company is Springer Nature Singapore Pte Ltd.
The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721,
Singapore
Supervisor’s Foreword

Advanced electronic circuits and systems are composed of the guiding wave-based
electronic devices and components with double metal configurations. Recently, the
spoof surface plasmon (SSP) modes have been proposed as a novel concept for the
design of electronic circuits. With the high field confinement, the SSP-based
structures do not suffer from the compactness limitations of conventional circuits so
that they are capable of playing the new and crucial role of an alternative for the
future generation of electronic circuits and electromagnetic systems. Despite the
vast number of research effort devoted to the study of the SSP modes, a deep
theoretical study and a reliable and repeatable modeling for the SSP structures are
strongly demanded for engineering applications. Here, we tackle the quest by
elaborating the basic requirements for the implementation of the SSP modes in
microwave circuits to enable the SSP-based circuit component designs.
For the SSP-based transmission systems, the equivalent circuit models for the
electromagnetic structures are important for the analysis and the design of their
characteristics. This thesis first proposes the equivalent circuit models for the
accurate characterization of the SSP structures, serving as an insightful guideline to
design the SSP-based circuits.
To efficiently excite the SSP modes, it is necessary to couple the SSP structures
to the guiding mode-based microwave sources and the conventional transmission
lines (conventional TLs) through a transition. Here, a new type of efficient transi-
tions is developed for the connection of the SSP structures to the conventional TLs
in order to form slow-wave transmission lines (SW-TLs). The simulation and
experiments have verified that the proposed SW-TL achieves as low as half of the
ohmic loss of the conventional counterparts. Moreover, the low cross talk between
the proposed SW-TLs is numerically and experimentally substantiated to be up to
one order of magnitude lower than that between the conventional microstrip TLs.
For the antenna feeding systems, the presence of the ground plane impedes the
proper operation of the antenna and distorts their desired radiation performance.
This thesis tackles this challenge and proposes a new method based on the
SSP-based structure to remove the unwanted effect of the ground plane in the
antenna systems. As an example, a cylindrical dielectric resonator antenna excited

v
vi Supervisor’s Foreword

to operate at the otherwise unavailable TE dominant modes. The simulation and

measurement have evidenced the unique features of prohibited TE01d mode such as
the lower thickness dependency of the resonant frequencies, ultra-compactness, and
horizontally polarized omnidirectional radiation pattern.
In addition to non-radiative transmission lines, the SSP structures have a great
potential to form highly efficient antennas. This thesis applies these modes to solve
one of the persistent challenges in traveling wave antennas. These antennas are
highly demanded for frequency scanning imaging systems, although they are suf-
fering from the inconsistent performance over their wideband frequency response.
Here, a single-layered leaky-wave antenna (SL-LWA) using a meander SSP struc-
ture is proposed and proved to provide a consistent gain variation less than 2.5 dB of
scanning beams within the 80% fractional frequency bandwidth.
In summary, the concept of the SSP has opened a new field for guided and
radiative systems in electronic systems. The work is at the early stage of research
and development. With deeper and deeper study, we can expect the more and more
applications of SSP-based structures in electronic systems in future.

Singapore Prof. Zhi Ning Chen

January 2018
Parts of the thesis have been published in the following journal articles:

1. A. Kianinejad, Z. N. Chen, and C.-W. Qiu, “Design and modeling of spoof

surface plasmon modes-based microwave slow-wave transmission line,” IEEE
Trans. Microw. Theory Tech., vol. 63, no. 6, pp. 1817–1825, Jun. 2015.
2. A. Kianinejad, Z. N. Chen, L. Zhang, W. Liu, and C. W. Qiu, “Spoof
plasmon-based Slow-wave excitation of dielectric resonator antennas,” IEEE
Trans. Antennas Propag., vol. 64, no. 6, pp. 2094–2099, Jun. 2016.
3. A. Kianinejad, Z. N. Chen, and C. W. Qiu, “Low-loss spoof surface plasmon
slow-wave transmission lines with compact transition and high isolation,” IEEE
Trans. Microw. Theory Tech., vol. 64, no. 10, pp. 3078–3086, Oct. 2016.
4. A. Kianinejad, Z. N. Chen, and C. W. Qiu, “A single-layered
spoof-plasmon-mode leaky wave antenna with consistent gain,” IEEE Trans.
Antennas Propagat., vol. 65, no. 2, pp. 681–687, Feb. 2017.
5. A. Kianinejad, Z. N. Chen, and C.-W. Qiu, “Modeling and implementation of
meander spoof plasmon modes for design of highly tunable transmission lines,”
to be submitted.

vii
Acknowledgements

Firstly, I would like to express my sincere gratitude to my advisors Prof. Zhi Ning
Chen and Prof. Cheng-Wei Qiu for their continuous support of my Ph.D. study and
related research, for their patience, motivation, and immense knowledge. Their
guidance helped me in all the time of research and writing of this thesis.
Besides my advisor, I would like to thank the rest of my thesis advisory com-
mittee: Prof. Tat Soon Yeo, Prof. Xudong Chen, and Prof. Minghui Hong, for their
insightful comments and encouragement, and for the hard question which incented
me to widen my research from various perspectives.
Special thanks to my family. Words cannot express how grateful I am to my
mother, Nasrin, my father, Abodlrahim, and my brothers, Iman and Mohammad, for
all of the sacrifices that they’ve made on my behalf. Their prayer for me was what
sustained me thus far. I would also like to thank all of my friends in the MMIC
Modelling and Packaging Laboratory, specially Wei Liu, Srien Sithara, Andong
Huang, Yuanyan Su, and Ashraf Adam Salih who supported me in writing and
incented me to strive toward my goal. I would also like to thank my friends outside
NUS, specially Sajjad Seifozzakerini, Hossein Dehghani Tafti, Mohammad Danesh,
and Rasool Maghareh, who made the journey more rewarding and enjoyable.

ix
Declaration

I, Amin Kianinejad, declare that the thesis submitted is my own original work.
I declare that the thesis contains research reported in co-authored work that has been
published, accepted for publication, or submitted for publication. I declare that the
thesis does not contain material which has been accepted, or submitted, for any
other degree or diploma at a university or other institution of higher learning.

January 2017

xi
Contents

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Physical Concept of Spoof Surface Plasmon Modes . . . . . . . . . . . 2
1.2 Surface Wave-Based Single Line Transmission Lines . . . . . . . . . . 3
1.3 SSP Based Microwave Components . . . . . . . . . . . . . . . . . . . . . . 3
1.4 Motivation and Organization of This Thesis . . . . . . . . . . . . . . . . . 4
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2 Spoof Surface Plasmon Modes Modeling Using Circuit
Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.1 Field Confinement of SSP Modes . . . . . . . . . . . . . . . . . . . . . . . . 12
2.2 Principles of Circuit Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.3 Equivalent Circuit Models for SSP Cells . . . . . . . . . . . . . . . . . . . 16
2.3.1 U-Shaped SSP Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.3.2 Symmetric SSP Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.3.3 Meander SSP Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.4 SSP-Based Circuit Design Using Circuit Models . . . . . . . . . . . . . 23
2.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3 SSP-Based Transmission Lines (TLs) . . . . . . . . . . . . . . . .. . . . . . . . 29
3.1 Conversion of SSP Modes to Guided Waves . . . . . . . .. . . . . . . . 29
3.1.1 Polarization Matching . . . . . . . . . . . . . . . . . . .. . . . . . . . 29
3.1.2 Momentum Matching . . . . . . . . . . . . . . . . . . . .. . . . . . . . 30
3.1.3 Impedance Matching . . . . . . . . . . . . . . . . . . . .. . . . . . . . 31
3.2 Transition Design from SSPs to Conventional TLs . . . .. . . . . . . . 33
3.2.1 Transition from U-Shaped Cells to Microstrip TLs . . . . . . 33
3.3 Loss in SW-TLs . . . . . . . . . . . . . . . . . . . . . . . . . . . .... . . . . . . 42
3.3.1 The Leaky Loss . . . . . . . . . . . . . . . . . . . . . . .... . . . . . . 42
3.3.2 Ohmic Loss . . . . . . . . . . . . . . . . . . . . . . . . . .... . . . . . . 42

xiii
xiv Contents

3.4 Mutual Coupling Between SW-TLs . . . . . . . . . . . . . . . . . . . . . . . 43

3.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4 Spoof Surface Plasmon Excitation of Dielectric Resonator
Antennas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
4.1 Resonance Modes of Cylindrical DRAs . . . . . . . . . . . . . . . . . . . . 47
4.2 Feeding Configuration Design . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
4.3 Excitation of the TE Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4.4 The Radiation Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
4.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
5 Spoof Surface Plasmon-Based Leaky-Wave Antenna (LWA) . . . . . . 57
5.1 Radiation Mechanism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
5.1.1 Single-layered Leaky-Wave Antenna . . . . . . . . . . . . . . . . . 58
5.1.2 SL-LWA Without 2nd Converter . . . . . . . . . . . . . . . . . . . 60
5.1.3 Effect of Dielectric and Metal . . . . . . . . . . . . . . . . . . . . . . 61
5.1.4 Comparison with Other LW Structures Similar
to SL-LWA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
5.2 Design Procedure and Optimization . . . . . . . . . . . . . . . . . . . . . . . 65
5.3 Experimental Verification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
5.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
6 Future Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
6.1 Spoof Surface Plasmon Modes in Antenna Design . . . . . . . . . . . . 74
6.2 Spoof Surface Plasmon-Based Circuit Design . . . . . . . . . . . . . . . . 74
Appendix A: Excitation of TE-Dominant Higher Order Modes
in DRAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Appendix B: Farfield Radiation Pattern Results of SL-LWA . . . . . . . . . 81
Abbreviations

CPW Coplanar waveguide

DRA Dielectric resonator antenna
EM Electromagnetic
LWA Leaky-wave antenna
MC Mutual coupling
MMIC Monolithic microwave integrated circuit
PA Power amplifier
PEC Perfect electric conductor
SL-LWA Single-layered leaky-wave antenna
SMA Subminiature version A
SSP Spoof surface plasmon
SW-TL Slow-wave transmission line
THz Terahertz
TL Transmission line
TSA Tapered slot antenna

xv
Symbols

b Wave number (rad/m)

k Wavelength (m)
k0 Wave number of light (rad/m)
x Angular frequency (rad/s)
c0 Speed of light (m/s)
e Permittivity (F/m)
l Permeability (H/m)
E Electric field (V/m)
H Magnetic field (A/m)
f Frequency (1/s)
V Electric potential (V)
I Electric current (A)
Zc Bloch impedance (Ω)
L Inductance (H)
C Capacitance (F)
R Resistance (Ω)
r Conductivity (S/m)
G Conductance (S)
hmax Maximum radiation direction (degree)

xvii
List of Figures

Fig. 2.1 A symmetric SSP cell. a A period cell of the symmetric

SSP structure. b Electric fields distribution near the cell.
b Amplitude and c vector field plots. d The dispersion
curves and e field distribution comparison between the SSP
cells and microstrip lines with and without the dielectric
substrate. Inset The microstrip line along with the coordinate
system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 12
Fig. 2.2 Dispersion study of SSP cells. a Dispersion curves
of the symmetric SSP cells for varying groove widths and
depths. b The effect of cell widths on the dispersion curves.
c The effect of the period length on the cut-off frequency
and momentum of the SSP modes. . . . . . . . . . . . . . . . . . . . . . .. 14
Fig. 2.3 Field distribution of SSP cells. a Comparison of the field
confinement near the symmetric SSP cells with various groove
depths. Inset An SSP cell along with the coordinate system.
b The field confinement near SSP cells for various period
lengths at 10 GHz. c The comparison of the maximum field
confinement near SSP cells with various period lengths.
The operating frequency for each case is as follows: p = 6 mm
at f = 11 GHz; p = 4 mm at f = 13 GHz; and p = 2 mm
at f = 16 GHz . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 15
Fig. 2.4 The circuit model for U-shaped SSP cells. a A unit cell
of the SSP structure along with the equivalent circuit model.
b The dimensions of the SSP structure. c The equivalent areas
of inductors. d The equivalent areas of capacitors. e The
dispersion curves in four background media. Inset The Bloch
impedance calculated for the SSP unit cells on Rogers
4003 1.52 mm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 17

xix
xx List of Figures

Fig. 2.5 The circuit model for symmetric SSP cells. a The equivalent
circuit model. b Comparison between dispersion curve results
from circuit model and eigen mode solver of CST Microwave
Studio. c The characteristic impedance of the symmetric
SSP cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 20
Fig. 2.6 The circuit model for meander SSP cells. a One unit cell
of meander SSP structure. b The equivalent circuit model.
c The effective area for the equivalent capacitors. d The
effective area of the current paths for the equivalent
inductors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 21
Fig. 2.7 Meander SSP cells: dispersion curves and characteristic
impedance. The comparison between the dispersion results
of the SSP modes in meander cells calculated from equivalent
circuit model and full wave analysis for various a cell widths
and b cell period lengths. c The characteristic impedance
of the meander SSP cell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 23
Fig. 2.8 Loading design for symmetric SW-TLs. a Current distribution
at the SW-TL. b Loaded SW-TL with two resistors. c Block
diagram representation of a loaded SW-TL. d The circuit
model for a loaded SSP cell . . . . . . . . . . . . . . . . . . . . . . . . . . .. 24
Fig. 2.9 Reflection results from loaded SW-TLs. a Circuit model
and b full wave simulation results. The loading resistors
for the four sets are tabulated in Table 2.4 . . . . . . . . . . . . . . . .. 26
Fig. 3.1 Polarization transformation of mode converter. a The proposed
SSP structure connected to the mode converter to construct
the SW–TL. b The normal and c the azimuthal components
of the electric field at six perpendicular planes to the mode
converter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 30
Fig. 3.2 Dispersion conversion. Inset The separate parts
of the mode converter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 31
Fig. 3.3 Impedance matching of mode converter. a The proposed
mode converter for the meander SW-TL. b The effective area
for the equivalent capacitors. c Matching of the impedance
through the mode converter . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 32
Fig. 3.4 Mode converter design for U-shaped SSP cells. a The SSP
structure connected to the microstrip line via matching
transitions to build the SW-TL. b The specified parameters
for the strip connector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 33
Fig. 3.5 Mode converter design for U-shaped SSP cells. a The ground
regulator, b Simulated |S21| and c |S11| for five parameter sets
(Table 3.2) for the entire frequency range. . . . . . . . . . . . . . . . .. 35
Fig. 3.6 Effect of ground connector. a Surface current on the SW–TL
at 7 GHz, b Comparison of the S-parameter results (simulation)
of SW-TL with and without port connection . . . . . . . . . . . . . . .. 36
List of Figures xxi

Fig. 3.7 Mode converter design for U-shaped SSP cells. a Substitution
of curved lines of the ground regulator by straight lines.
b S-parameter results of SW–TL with curved ground regulator,
straight ground regulator and SW–TL with parameters given
in Table 3.3. c S-parameter results for SW–TL with single, two,
three and four unit cells strip connectors . . . . . . . . . . . . . . . . . .. 37
Fig. 3.8 Experimental evaluation of U-shaped SW-TL. a The fabricated
prototype. b Comparison between the experimental
and simulation S-parameters results . . . . . . . . . . . . . . . . . . . . . .. 38
Fig. 3.9 Mode converter design for symmetric SSP cells.
a The proposed compact design. b The conventional
mode converter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 39
Fig. 3.10 S-parameter results of symmetric SW-TL. a Simulated |S21|
for five parameter sets given in Table 3.4 as well as the SW-TL
with the conventional mode converter. b Experimental
evaluation of the design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 40
Fig. 3.11 Meander SW-TL. a The schematic SW-TL. Inset the
dimensions of one SSP unit cell. b The simulated transmission
and reflection from the designed SW-TL. . . . . . . . . . . . . . . . . .. 41
Fig. 3.12 Loss in symmetric SW-TLs. a Comparison between |S21|
results for symmetric SW-TLs with different lengths.
b Simulation and measured insertion loss
of symmetric SW-TL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 43
Fig. 3.13 Mutual coupling between SW-TLs. a The fabricated prototype
of the two closely located and loaded SW-TLs. Comparison
of the MC between SW-TLs and microstrip lines for various
b overlapping part lengths and c distance between
two SW-TLs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 44
Fig. 4.1 Resonance frequencies fr of cylindrical DRAs. Comparison
between fr of TE01d and HEM11d for various antenna
thicknesses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 48
Fig. 4.2 Field distribution of a cylindrical DRA. a The schematic
of an isolated DRA and the electric field vector distributions
of the fundamental mode (TE01d). b The schematic of a DRA
on a ground plane and the electric field vector distributions
of the nominal fundamental mode (HEM11d). . . . . . . . . . . . . . .. 48
Fig. 4.3 The proposed meander SW-TLs. a The schematic
of the design. Inset The dimensions of one SSP unit cell.
b The magnetic field distribution at two cutting planes . . . . . . .. 50
Fig. 4.4 A DRA fed by SW-TL with four orientations. Arrows represent
the magnetic field vectors. a The DRA’s axial direction
is parallel to the SW-TL. b The DRA is placed on top
of the SW-TL and its axial direction is parallel to the grooves
of the SSP cells. c The DRA is placed on top of the SW-TL
xxii List of Figures

and its axial direction is perpendicular to the SW-TL.

d The DRA is placed on top of the SW-TL and its axial
direction is parallel to the SW-TL . . . . . . . . . . . . . . . . . . . . . . .. 50
Fig. 4.5 A DRA fed by meander SW-TL. a The design configuration.
Inset: the simulated electric field vector of the fundamental
mode TE01d and the feeding end. b A DRA fed by microstrip
transmission line. c The S-parameter results for various DRA
locations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 51
Fig. 4.6 Excitation of a cylindrical DRA. a Comparison
of S-parameter results among five cases: simulation
results for a DRA fed with the conventional microstrip
TL and with the SW-TL (T = 4 mm and T = 1.5 mm),
experimental results for a DRA fed with the SW-TL
(T = 4 mm). b Photographs of the fabricated SW-TL
feeding configuration and the DRA . . . . . . . . . . . . . . . . . . . . . .. 52
Fig. 4.7 Far field radiation patterns of DRA (realised gain in dBi).
Comparison among three cases: Simulation results for the
conventional microstrip transmission line feeding technique
and for SW-TL feeding method, and the experimental
results for SW-TL feeding method a1 Eu in x-y plane.
a2 Eh in x-y plane. b1 Eu in x-z plane. b2 Eh in x-z plane.
c1 Eu in y-z plane. c2 Eh in y-z plane . . . . . . . . . . . . . . . . . . . .. 53
Fig. 5.1 Meander SW-TL. a The schematic of the design. Inset:
the meander SSP cell. b Scattering parameter results
from the meander SW-TL . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 58
Fig. 5.2 Radiation from the SW-TL. a Dispersion curves for the
dominant mode and the first space harmonic. b Normalized
co-component (u) radiation patterns of the SL-LWA in the y-z
plane at frequencies. c 3D radiation pattern at 17 GHz . . . . . . .. 59
Fig. 5.3 Designed SL-SW-TL. a SL-LWA without the second
converter. b The magnetic field distribution near
the SL-LWA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 61
Fig. 5.4 Effects of the 2nd mode converter. a The reflection spectrum,
b the maximum radiation direction and the gain . . . . . . . . . . . .. 61
Fig. 5.5 Dielectric substrate effect of SL-SW-TL. a The scattering
parameter results. b The maximum radiation angle
and the gain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 62
Fig. 5.6 Efficiency of SL-LWA. a Total efficiency with and without
the dielectric substrate. b The Radiation efficiency for three
cases: SL-LWA (1) with lossy metal and dielectric,
(2) with lossless dielectric and (3) with PEC. . . . . . . . . . . . . . .. 62
Fig. 5.7 Effect of the loss on SL-LWA. The normalized decay
rate of the SL-LWA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 63
List of Figures xxiii

Fig. 5.8 A simple strip antenna without grooves. a The schematic.

b The reflection results for varying geometrical parameters . . .. 64
Fig. 5.9 Effect of the period length on SL-LWA. a The reflection and
b the maximum radiation angle with the varying
period lengths . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 66
Fig. 5.10 Realized gain of SL-LWA. The maximum realized gain
in the y-z plane with varying a d and w. b Nm and lt.
c The broadside realized gain . . . . . . . . . . . . . . . . . . . . . . . . . .. 67
Fig. 5.11 Experimental evaluation of SL-LWA. a The fabricated
prototype. b The measured reflection spectrum compared
with the simulation results . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 68
Fig. 5.12 Normalized radiation pattern of SL-LWA. Comparison
between the measured and simulated results for the
co-component (u) as well as cross-component (h)
in the y-z plane at a 10.5, b 17, c 20 and d 24 GHz. . . . . . . . .. 69
Fig. 5.13 Maximum realized gain in y-z plane . . . . . . . . . . . . . . . . . . . . .. 69
Fig. A.1 TE-dominant higher order modes excited by SW-TL. The axial
component of the magnetic field for EH11d (a1), EH21d (a2) and
EH31d (a3). The electric field vectors for EH11d (b1), EH21d (b2)
and EH31d (b3). The far field radiation pattern of the
co-polarized component of the electric fields at the x-y-plane
for EH11d (c1), EH21d (c2) and EH31d (c3). The resonance
frequency of each mode for the isolated DRA (value in
parentheses) as well as the DRA fed by SW-TL
(value out of parentheses) are indicated in (a1–a3)
in unit of GHz . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 78
Fig. B.1 Normalized radiation pattern of SL-LWA.
Comparison between the measured and simulated results
for the co-component (u) as well as cross-component (h)
in the y-z plane at a 10.5, b 11, c 12, and d 13 GHz . . . . . . . .. 81
Fig. B.2 Normalized radiation pattern of SL-LWA. Comparison
between the measured and simulated results
for the co-component (u) as well as cross-component (h)
in the y-z plane at a 14, b 15, c 16, d 17, e 18, f 19,
g 20 and h 21 GHz . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 82
Fig. B.3 Normalized radiation pattern of SL-LWA. Comparison
between the measured and simulated results for the
co-component (u) as well as cross-component (h)
in the y-z plane at a 22, b 23, c 24, and d 24.5 GHz . . . . . . . .. 83
List of Tables

Table 2.1 Circuit element parameters of the U-shaped SSP cells

in media with different dielectric coefficients . . . . . . . . . . . . . .. 18
Table 2.2 Equivalent circuit elements for six symmetric SSP cells
with varying cell widths . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 21
Table 2.3 Circuit element parameters for meander SSP cells . . . . . . . . . .. 23
Table 2.4 Four sets of resistors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 26
Table 3.1 Relation between parameters of strip connector
in Fig. 3.4b . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 34
Table 3.2 Five parameter sets for mode converter . . . . . . . . . . . . . . . . . .. 36
Table 3.3 Parameters of U-shaped SW–TL after optimization
(in mm) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 37
Table 3.4 Five parameter sets for mode converter . . . . . . . . . . . . . . . . . .. 39

xxv
Chapter 1
Introduction

Electronic circuits are compositions of electronic components such as diodes,

resistors, capacitors, inductors, transistors, and sensors, which are connected by
transmission lines. At very low frequencies with the wavelengths larger than the
circuit dimensions, transmission lines are as simple as pairs of wires; however, by
increasing the frequency TLs become more complex and their configurations, sizes
and designs change.
Compactness is a necessity for all electronic circuits. To meet the compactness
requirement of electronic systems and condense all the required electrical func-
tionalities in a limited area, multilayer circuits are developed composing of several
layers, where each layer is designed to perform a specific task. Layers are separated
by ground planes and to transfer the signals between layers, they are connected
through via holes. However, further miniaturization of multilayer integrated circuits
is limited by the general performance of systems and the mutual effects between the
layers and embedded components and specially the transmission lines.
Consequently, the compactness still remains the formidable challenge of the elec-
tronic systems.
Transmission lines and other components in electronic systems operate based on
the guiding waves and with their double-metal configurations they are always
attached to the inevitable ground plane. The compactness of the multilayer circuits
is mainly limited by the ground plane, where reducing the distance between the
layers drastically increases the mutual effect between the ground and the adjacent
components and consequently degrades their performance.
Recently, the Spoof Surface Plasmon (SSP) modes have been introduced as
highly localized surface waves ranging from sub-gigahertz [1, 2] to terahertz
(THz) regime [3, 4]. Benefiting from a simple configuration while presenting a very
high field confinement, these ground-free modes have shown promises in realiza-
tion of various circuit functions. Consequently, the SSPs do not suffer from the
compactness limitations of traditional circuits and are capable of being a new
platform for the guiding wave-based components in the future generations of the

© Springer Nature Singapore Pte Ltd. 2018 1

A. Kianinejad, Metamaterial Surface Plasmon-Based Transmission Lines
and Antennas, Springer Theses, https://doi.org/10.1007/978-981-10-8375-4_1
2 1 Introduction

electronic circuits and electromagnetic based systems. This chapter addresses the
background of the spoof surface plasmon modes and recent developments of the
SSP structures. The realized circuit functionalities using SSP modes will be dis-
cussed in the following sections.

1.1 Physical Concept of Spoof Surface Plasmon Modes

Surface plasmons (SPs) are the highly localized modes at the surface of two
materials (e.g., metal-dielectric interface) of opposite signs in the real parts of the
electric permittivities. These waves propagate along the interface of a conductor and
decay exponentially in transverse direction [5]. SPs reduce the effective wavelength
significantly and therefore present high spatial confinement [6]. Ebbesen et al.
reported extraordinary transmission in optical wavelengths, through subwavelength
apertures owing to the coupling of light with surface plasmons (SPs) [7].
SPs have found potential applications in ultrafast optical switching [8],
biomedical sensing (fiber sensors) [9], near-field microscopy [10–12], optoelec-
tronics [13, 14], magneto-optic data storage [5], photovoltaics [15], and nanopho-
tonics [13], optical signal processing [16, 17] etc.
SPs originally exist at optical frequencies [5]. However, the surface wave modes
at microwave frequencies were reported in couples of works [18–22]. A.P. Hibbins
et al. reported wavelength-dependent reflectivity with the same trend as the surface
plasmon polaritons in metallic sinusoidal diffraction gratings at microwave fre-
quencies [23]. These modes were also observed through various designs such as a
single aperture surrounded by concentric grooves [24], annular apertures [25], a
dual pitch metal grating [26], deep metallic gratings [27] and metallic grating with a
subwavelength aperture [28]. Enhanced transmission due to the excitation of sur-
face plasmon modes was also reported at THz in several studies [29, 30]. Pendry
et al. analytically demonstrated these SP-like modes at microwave regime in a
perforated conducting surface [1]. The structure is composed of the metal layer with
two dimensional periodic arrangement of square holes with the size of (a
a) and
the spacing (d), where a and d are much smaller that the wavelength. The authors in
[1] analytically showed that the dispersion behavior of the modes near the perfo-
rated metallic layer is similar to the real surface plasmons:

1 64a4 x4
kjj2 c20 ¼ x2 þ ; ð1:1Þ
x2p  x p4 d 4
2

where, k jj is the transverse wavenumber of the plasmon waves at the metal layer. At
lower frequencies, the surface wavenumber can be approximated by the light line.
As the frequency x approaches xp, the wavenumber approaches infinity. xp is the
plasmon frequency and is:
1.1 Physical Concept of Spoof Surface Plasmon Modes 3

pffiffiffiffiffiffiffiffiffi
xp ¼ pc0 =a eh lh : ð1:2Þ

c0 is the speed of light in the free space and eh and lh are the hole’s permittivity
and permeability, respectively. A design composed of a perforated metal layer was
experimentally verified to support SP-like modes by Hibbins et al. [31]. Later on,
the term “spoof surface plasmon” or SSP has been used to describe the SP-like
modes at microwave and terahertz frequencies [32–39].
Shen et al. applied their planar U-shaped structure to transfer SSP modes [40].
The structure was fabricated on a flexible or ultrathin dielectric film. They showed
that the field propagates through their design even when the SSP unit cells form a
curve path. They also realized a simple power splitter using the symmetrical SSP
structures. The authors in [40] took the advantage of a flexible dielectric film to
demonstrate the flexibility of their designs.

1.2 Surface Wave-Based Single Line Transmission Lines

The “single wire” transmission line (TL) was first introduced by Goubau [41] and
composed of a dielectric coated wire connected to two coaxial connectors that form
a groundless TL. The structure transmit the electromagnetic energy by stimulating
the surface wave modes of a single wire [22]. In other works, planar transitions
were developed for the connection of the groundless Goubau line to co-planar
waveguides (CPWs) at THz bands [42, 43]. The transition by Akalin et al. in [43] is
composed of tapered strips and flared ground planes and with the transmission of
−5 dB at THz. The coupling between the Goubau line and resonating structures
(ring resonators) has been studied [44]. Akalin and Padila examined the 90°
bending of a Goubau line and designed a filter by adding periodic corrugations to a
Goubau line [45]. Xu et al. designed the feeding transition for the Goubau loads and
power divider/combiner at millimeter wave range [46]. Another group proposed a
coplanar transition to a planar Goubau line at millimeter wave bands [47]. Laurette
et al. added corrugations to the planar Goubau line to form a slow-wave trans-
mission line at the THz regime [48]. They showed that adding corrugations
increases both the propagation effective index and transmission from the Goubau
line. The Goubau line has also been applied to form a low-loss leaky-wave antenna
by adding dipole antennas to the transmission line [49].

1.3 SSP Based Microwave Components

Spoof Surface Plasmons (SSPs) have been applied to form waveguides in several
works. Navarro-Cía et al. designed a waveguide by their proposed split ring
resonator-based SSP structure [39]. Another related study took advantage of an
4 1 Introduction

L-shaped metallic element array that was horizontally attached to a metal surface to
construct a transmission line [50]. Navarro-Cía et al. designed a slow-wave trans-
mission line and splitter by coaxial spoof plasmons [51]. They applied a monopole
antenna as the source. Pandey et al. employed 3D printing technology to fabricate a
THz plasmonic waveguide [52].
In another study, the multi-channel composite SSPs were excited by a CPW line
[53]. Pan et al. applied electrically resonant metamaterials to control the rejection of
an SSP-based TL [54]. The SSP modes have inspired the design of many micro-
wave components, including couplers and frequency splitters [55, 56] and wide-
band SSP filters [57–59]. The usage of the SPP modes has been further enhanced by
the broadband amplification technique in [60]. SSPs have been applied to design
multiband transmission lines [61]. SSP mode-based TLs have also found their way
into on-chip CMOS designs [62].
Conversion of the SSP modes to radiating modes has been studied [63–70]. In
[65], an SSP-based waveguide feeds an array of microstrip patch antennas. Another
study applied the SSP-based waveguides to control the radiation from dielectric
slabs [66]. Wu et al. studied the leaky-wave radiating form the SSP modes [67].

1.4 Motivation and Organization of This Thesis

Recent development of highly efficient interfaces for the connection of the SSPs to
the conventional microwave TLs was the major step toward the development of the
SSP-based circuits. Since then, much effort has been devoted to realize the various
electronic components based on the SSP modes. However, it is still necessary to
further theoretically explore the mechanism and modelling of the SSPs at micro-
wave bands in order to realize all the required functionalities and electrical com-
ponents to enable the all-SSP-based circuits and benefit from the SSP modes to
break the compactness limitation of the current microwave systems. This thesis
aims to address this issue by providing the fundamental requirements for the
implementation of the SSP modes in electromagnetic engineering in order to realize
the novel and effective plasmonic-inspired microwave devices.
In the second chapter, the spoof surface plasmon modes are studied and the first
equivalent circuit elements for the SSP structures are proposed. The modeling
method is the key to better understanding of the SSP modes as well as to facilitate
their design procedure and increases their compatibilities with other microwave
components. We apply the proposed method to derive the equivalent circuit models
for U-shaped, symmetric and meander SSP cells. The models are applied to cal-
culate the dispersion curve and the Bloch impedance of the periodic arrangement of
the cells. The results are evaluated using the full wave simulation from commercial
software for different cells. Design and optimization of the EM structures using
full-wave analysis software is time and memory hungry. We show that the proposed
1.4 Motivation and Organization of This Thesis 5

circuit models are effective tools in the circuit design using the SSP structures. We
applied the circuit model to design a broadband loading for the single line
SSP-based transmission lines while the results are in good agreement with the
full-wave analysis.
In the third chapter, we develop efficient transitions from SSP structures to
conventional transmission lines. The proposed transitions smoothly convert the SSP
modes to the guiding waves through the transformation of the polarization,
momentum and impedance. While gradually converting the quasi-TEM modes in
the conventional TLs to the TM polarized SSP modes, the momentum of the EM
waves slowly increases via the designed transition to ensure the maximum coupling
to the SSP cells. The proposed transitions provide gradual impedance matching for
the meander SSP-cells with inductive characteristic impedance. With high trans-
mission and low reflections, the periodic SSP structure along with the compact
transitions form an efficient slow-wave transmission lines (SW-TLs). We experi-
mentally characterize the important features of the SW-TLs including the leaky
loss, ohmic loss and mutual coupling. The results show much lower loss and cross
talk for these designs in comparison with the conventional TLs. Besides, we model
the ohmic loss using cell geometries and applied the model to tune the loss in the
SW-TLs. Our further exploration on the field distribution of the SSP modes helps to
design the SW-TLs with minimized mutual coupling as highly isolated transmission
lines appealing for compact circuit designs.
In the fourth chapter, we apply the designed groundless SW-TL for antenna
feeding applications. We study the effect of adding a ground plane to a cylindrical
dielectric resonator antenna (DRA) on resonance modes. We show that the
unavoidable ground plane in the conventional feeding methods suppresses the
tangential electric fields near the ground, prohibits the excitation of the TE modes
and consequently changes the fundamental modes of the DRA from TE modes to
hybrid TM modes. TE modes in cylindrical DRA possess various advantageous
features including the lower resonance frequency, less antenna thickness depen-
dency and horizontally polarized omnidirectional radiation pattern. We design a
meander slow-wave transmission line to feed a cylindrical DRA and numerically
and experimentally evaluate the excitation of the TE dominant modes as well as the
hybrid TE higher order modes with all the merits. We show that using the SW-TLs
enables the effective excitation of the planar DRAs.
The fifth chapter presents a leaky-wave antenna based on the spoof surface
plasmon modes. The conventional LWAs are suffering from the low radiation
performance at their higher frequency range. The SSP-based LWA solve this
challenge by providing the consistent radiation conversion through the whole fre-
quency range. We proposed a straight forward dispersion-based design method to
develop a new generation of LWAs with a single layer and simple configuration
operating based on the SSP modes. The design is numerically and experimentally
evaluated and demonstrates the flat realized gain for backward, forward and
broadside radiation.
6 1 Introduction

References

1. J.B. Pendry, L. Martin-Moreno, F.J. Garcia-Vidal, Mimicking surface plasmons with

structured surfaces. Science 305(5685), 847–848 (2004)
2. F.F. Qin, J.J. Xiao, Z.Z. Liu, Q. Zhang, Multiple fano-like transmission mediated by
multimode interferences in spoof surface plasmon cavity-waveguide coupling system. IEEE
Trans. Microw. Theory Tech. 64(4), 1186–1194 (2016)
3. J.R. Middendorf, J.S. Cetnar, J. Owsley, E.R. Brown, High fill-factor substrate-based
wire-grid polarizers with high extinction ratios. IEEE Trans. Terahertz Sci. Technol. 4(3),
376–382 (2014)
4. M. Aghadjani, P. Mazumder, THz polarizer controller based on cylindrical spoof surface
plasmon polariton (C-SSPP). IEEE Trans. Terahertz Sci. Technol. 5(4), 556–563 (2015)
5. W.L. Barnes, A. Dereux, T.W. Ebbesen, Surface plasmon subwavelength optics. Nature 424
(6950), 824–830 (2003)
6. J.M. Pitarke, V.M. Silkin, E.V. Chulkov, P.M. Echenique, Theory of surface plasmons and
surface-plasmon polaritons. Rep. Prog. Phys. 70(1), 1 (2007)
7. T.W. Ebbesen, H.J. Lezec, H.F. Ghaemi, T. Thio, P.A. Wolff, Extraordinary optical
transmission through sub-wavelength hole arrays. Nature 391(6668), 667–669 (1998)
8. K.F. MacDonald, Z.L. Sámson, M.I. Stockman, N.I. Zheludev, Ultrafast active plasmonics.
Nat. Photonics 3(1), 55–58 (2009)
9. W.-H. Tsai, Y.-C. Tsao, H.-Y. Lin, B.-C. Sheu, Cross-point analysis for a multimode fiber
sensor based on surface plasmon resonance. Opt. Lett. 30(17), 2209–2211 (2005)
10. N. Fang, H. Lee, C. Sun, X. Zhang, Sub–diffraction-limited optical imaging with a silver
superlens. Science 308(5721), 534–537 (2005)
11. S. Kawata, Y. Inouye, P. Verma, Plasmonics for near-field nano-imaging and superlensing.
Nat. Photonics 3(7), 388–394 (2009)
12. M. Ozaki, J. Kato, S. Kawata, Surface-plasmon holography with white-light illumination.
Science 332(6026), 218–220 (2011)
13. E. Ozbay, Plasmonics: merging photonics and electronics at nanoscale dimensions. Science
311(5758), 189–193 (2006)
14. W. Cai, W. Shin, S. Fan, M.L. Brongersma, Elements for plasmonic nanocircuits with
three-dimensional slot waveguides. Adv. Mater. 22(45), 5120–5124 (2010)
15. A. Polman, H.A. Atwater, Photonic design principles for ultrahigh-efficiency photovoltaics.
Nat. Mater. 11(3), 174–177 (2012)
16. J.A. Conway, S. Sahni, T. Szkopek, Plasmonic interconnects versus conventional intercon-
nects: a comparison of latency, crosstalk and energy costs. Opt. Express 15(8), 4474–4484
(2007)
17. D.K. Gramotnev, S.I. Bozhevolnyi, Plasmonics beyond the diffraction limit. Nat. Photonics 4
(2), 83–91 (2010)
18. A. Sommerfeld, Ueber die fortpflanzung elektrodynamischer wellen längs eines drahtes. Ann.
Phys. 303(2), 233–290 (1899)
19. G. Goubau, Surface waves and their application to transmission lines. J. Appl. Phys. 21(11),
1119–1128 (1950)
20. A.F. Harvey, Periodic and guiding structures at microwave frequencies. IRE Trans. Microw.
Theory Tech. 8(1), 30–61 (1960)
21. R. Ulrich, M. Tacke, Submillimeter waveguiding on periodic metal structure. Appl. Phys.
Lett. 22(5), 251–253 (1973)
22. G. Goubau, On the excitation of surface waves. Proc. IRE 40(7), 865–868 (1952)
23. A.P. Hibbins, J.R. Sambles, C.R. Lawrence, Grating-coupled surface plasmons at microwave
frequencies. J. Appl. Phys. 86(4), 1791–1795 (1999)
References 7

24. M.J. Lockyear, A.P. Hibbins, J.R. Sambles, C.R. Lawrence, Enhanced microwave
transmission through a single subwavelength aperture surrounded by concentric grooves.
J. Opt. Pure Appl. Opt. 7(2), S152 (2005)
25. H. Caglayan, I. Bulu, E. Ozbay, Extraordinary grating-coupled microwave transmission
through a subwavelength annular aperture. Opt. Express 13(5), 1666–1671 (2005)
26. A.P. Hibbins, J.R. Sambles, C.R. Lawrence, Excitation of remarkably nondispersive surface
plasmons on a nondiffracting, dual-pitch metal grating. Appl. Phys. Lett. 80(13), 2410–2412
(2002)
27. H.E. Went, J.R. Sambles, Resonantly coupled surface plasmon polaritons in the grooves of
very deep highly blazed zero-order metallic gratings at microwave frequencies. Appl. Phys.
Lett. 79(5), 575–577 (2001)
28. S.S. Akarca-Biyikli, I. Bulu, E. Ozbay, Enhanced transmission of microwave radiation in
one-dimensional metallic gratings with subwavelength aperture. J. Appl. Phys. 85,
1098–1100 (2004)
29. A.K. Sarychev, V.A. Podolskiy, A.M. Dykhne, V.M. Shalaev, Resonance transmittance
through a metal film with subwavelength holes. IEEE J. Quantum Electron. 38(7), 956–963
(2002)
30. J.G. Rivas, C. Schotsch, P.H. Bolivar, H. Kurz, Enhanced transmission of THz radiation
through subwavelength holes. Phys. Rev. B 68(20), 201306 (2003)
31. A.P. Hibbins, B.R. Evans, J.R. Sambles, Experimental verification of designer surface
plasmons. Science 308(5722), 670–672 (2005)
32. S.A. Maier, S.R. Andrews, L. Martin-Moreno, F.J. Garcia-Vidal, Terahertz surface
plasmon-polariton propagation and focusing on periodically corrugated metal wires. Phys.
Rev. Lett. 97(17), 176805 (2006)
33. C.R. Williams, S.R. Andrews, S.A. Maier, A.I. Fernández-Domínguez, L. Martín-Moreno,
F.J. García-Vidal, Highly confined guiding of terahertz surface plasmon polaritons on
structured metal surfaces. Nat. Photonics 2(3), 175–179 (2008)
34. N. Yu et al., Designer spoof surface plasmon structures collimate terahertz laser beams. Nat.
Mater. 9(9), 730–735 (2010)
35. D. Martin-Cano, M.L. Nesterov, A.I. Fernandez-Dominguez, F.J. Garcia-Vidal,
L. Martin-Moreno, E. Moreno, Domino plasmons for subwavelength terahertz circuitry.
Opt. Express 18(2), 754–764 (2010)
36. Y.J. Zhou, Q. Jiang, T.J. Cui, Bidirectional bending splitter of designer surface plasmons.
Appl. Phys. Lett. 99(11), 111904 (2011)
37. Q. Gan, Y. Gao, K. Wagner, D. Vezenov, Y.J. Ding, F.J. Bartoli, Experimental verification of
the rainbow trapping effect in adiabatic plasmonic gratings. Proc. Natl. Acad. Sci. 108(13),
5169–5173 (2011)
38. G. Kumar, S. Pandey, A. Cui, A. Nahata, Planar plasmonic terahertz waveguides based on
periodically corrugated metal films. New J. Phys. 13(3), 033024 (2011)
39. M. Navarro-Cía, M. Beruete, S. Agrafiotis, F. Falcone, M. Sorolla, S.A. Maier, Broadband
spoof plasmons and subwavelength electromagnetic energy confinement on ultrathin
metafilms. Opt. Express 17(20), 18184–18195 (2009)
40. X. Shen, T.J. Cui, D. Martin-Cano, F.J. Garcia-Vidal, Conformal surface plasmons
propagating on ultrathin and flexible films. Proc. Natl. Acad. Sci. 110(1), 40–45 (2013)
41. G. Goubau, Open wire lines. IRE Trans. Microw. Theory Tech. 4(4), 197–200 (1956)
42. A. Treizebré, T. Akalin, B. Bocquet, Planar excitation of Goubau transmission lines for THz
bioMEMS. IEEE Microwave Wirel. Compon. Lett. 15(12), 886–888 (2005)
43. T. Akalin, A. Treizebré, B. Bocquet, Single-wire transmission lines at terahertz frequencies.
IEEE Trans. Microw. Theory Tech. 54(6), 2762–2767 (2006)
44. T. Akalin, E. Peytavit, J.F. Lampin, THz long range plasmonic waveguide in membrane
topology, in 33rd International Conference on Infrared, Millimeter and Terahertz Waves,
2008. IRMMW-THz 2008 (2008), pp. 1–2
8 1 Introduction

45. T. Akalin, W. Padilla, Plasmonic waveguides and metamaterial components at terahertz

frequencies, in 2009 Asia Pacific Microwave Conference, Singapore (2009), pp. 2444–2446
46. Y. Xu, C. Nerguizian, R.G. Bosisio, Wideband planar Goubau line integrated circuit
components at millimetre waves. IET Microwaves Antennas Propag. 5(8), 882–885 (2011)
47. J. Emond et al., A low-loss planar goubau line and a coplanar-PGL transition on
high-resistivity silicon substrate in the 57–64 GHz band. Microw. Opt. Technol. Lett. 54
(1), 164–168 (2012)
48. S. Laurette, A. Treizebre, B. Bocquet, Corrugated Goubau lines to slow down and confine
THz waves. IEEE Trans. Terahertz Sci. Technol. 2(3), 340–344 (2012)
49. D. Sanchez-Escuderos, M. Ferrando-Bataller, J.I. Herranz, M. Cabedo-Fabres, Periodic
leaky-wave antenna on planar Goubau line at millimeter-wave frequencies. IEEE Antennas
Wirel. Propag. Lett. 12, 1006–1009 (2013)
50. D. Martin-Cano, O. Quevedo-Teruel, E. Moreno, L. Martin-Moreno, F.J. Garcia-Vidal,
Waveguided spoof surface plasmons with deep-subwavelength lateral confinement. Opt. Lett.
36(23), 4635–4637 (2011)
51. M. Navarro-Cía, M. Beruete, M. Sorolla, S.A. Maier, Enhancing the dual-band guiding
capabilities of coaxial spoof plasmons via use of transmission line concepts. Plasmonics 6(2),
295–299 (2011)
52. S. Pandey, B. Gupta, A. Nahata, Terahertz plasmonic waveguides created via 3D printing.
Opt. Express 21(21), 24422–24430 (2013)
53. L. Liu et al., Multi-channel composite spoof surface plasmon polaritons propagating along
periodically corrugated metallic thin films. J. Appl. Phys. 116(1), 013501 (2014)
54. B.C. Pan, Z. Liao, J. Zhao, T.J. Cui, Controlling rejections of spoof surface plasmon
polaritons using metamaterial particles. Opt. Express 22(11), 13940–13950 (2014)
55. X. Gao et al., Ultrathin dual-band surface plasmonic polariton waveguide and frequency
splitter in microwave frequencies. Appl. Phys. Lett. 102, 151912 (2013)
56. X. Liu, Y. Feng, K. Chen, B. Zhu, J. Zhao, T. Jiang, Planar surface plasmonic waveguide
devices based on symmetric corrugated thin film structures. Opt. Express 22(17),
20107–20116 (2014)
57. X. Gao, L. Zhou, Z. Liao, H.F. Ma, T.J. Cui, An ultra-wideband surface plasmonic filter in
microwave frequency. Appl. Phys. Lett. 104(19), 191603 (2014)
58. J.Y. Yin, J. Ren, H.C. Zhang, Q. Zhang, T.J. Cui, Capacitive-coupled series spoof surface
plasmon polaritons. Sci. Rep. 6, 24605 (2016)
59. B. Xu, Z. Li, L. Liu, J. Xu, C. Chen, C. Gu, Bandwidth tunable microstrip band-stop filters
based on localized spoof surface plasmons. JOSA B 33(7), 1388–1391 (2016)
60. H.C. Zhang, S. Liu, X. Shen, L.H. Chen, L. Li, T. J. Cui, Broadband amplification of spoof
surface plasmon polaritons at microwave frequencies. Laser Photonics Rev. 9(1), 83–90
(2014)
61. X. Liu, Y. Feng, B. Zhu, J. Zhao, T. Jiang, High-order modes of spoof surface plasmonic
wave transmission on thin metal film structure. Opt. Express 21(25), 31155–31165 (2013)
62. Y. Liang et al., On-chip sub-terahertz surface plasmon polariton transmission lines with mode
converter in CMOS. Sci. Rep. 6, 30063 (2016)
63. H. Shi, X. Wei, Z. Zhao, X. Dong, Y. Lu, C. Du, A new surface wave antenna-based spoof
surface plasmon mechanism. Microw. Opt. Technol. Lett. 52(10), 2179–2183 (2010)
64. J.J. Xu, H.C. Zhang, Q. Zhang, T.J. Cui, Efficient conversion of surface-plasmon-like modes
to spatial radiated modes. Appl. Phys. Lett. 106(2), 021102 (2015)
65. Huan Yi, Qu Shi-Wei, Xue Bai, Antenna array excited by spoof planar plasmonic waveguide.
IEEE Antennas Wirel. Propag. Lett. 13, 1227–1230 (2014)
66. O. Quevedo-Teruel, Controlled radiation from dielectric slabs over spoof surface plasmon
waveguides. Prog. Electromagn. Res. 140, 169–179 (2013)
References 9

67. J.J. Wu et al., High-directivity radiation based on the leaky mode of spoof surface plasmon
polaritons. IET Microwaves Antennas Propag. 8(13), 1075–1079 (2014)
68. S.-H. Kim et al., Subwavelength localization and toroidal dipole moment of spoof surface
plasmon polaritons. Phys. Rev. B 91(3), 035116 (2015)
69. B. Xu et al., Tunable band-notched coplanar waveguide based on localized spoof surface
plasmons. Opt. Lett. 40(20), 4683–4686 (2015)
70. J.Y. Yin, H.C. Zhang, Y. Fan, T.J. Cui, Direct radiations of surface plasmon polariton waves
by gradient groove depth and flaring metal structure. IEEE Antennas Wirel. Propag. Lett. 15,
865–868 (2016)
Chapter 2
Spoof Surface Plasmon Modes Modeling
Using Circuit Elements

This chapter studies the spoof surface plasmon modes. One of the important fea-
tures of the SSP modes is their confinement, which is directly related to their
moment. In the first section of this chapter, the effects of different geometrical
parameters on the confinements of the modes are discussed. Moreover, a method is
discussed to increase the momentum of the SSP modes in order to maximize their
field confinement and consequently minimize the mutual effect between SSP-based
circuits.
Establishing an equivalent circuit model is the key to characterize and thereby
design of microwave structures. In the second section of this chapter, we propose a
model that offers a general understanding of SSP-based transmission lines, and
facilitate further advanced engineering designs and optimizations with desired
performance. By modeling one unit cell of a periodic structure using an equivalent
circuit model, the network parameters such as scattering and transmission matrices
are calculated. These parameters, in return, are helpful in calculating the dispersion
curve and Bloch impedance of the periodic structures.
Here, the U-shaped, symmetric and meander SSP cells are modeled and the
equivalent circuit elements are calculated using electrostatic approximation and
based on the geometrical features of the cells. The model is applied to calculate the
dispersion curve and Bloch impedance of the SSP cells and verified by full-wave
numerical solutions using time domain and eigen mode solvers in commercial
software CST Microwave Studio. In addition, the proposed models are imple-
mented in design of SSP-based circuits and the method is validated with full wave
analysis results.

© Springer Nature Singapore Pte Ltd. 2018 11

A. Kianinejad, Metamaterial Surface Plasmon-Based Transmission Lines
and Antennas, Springer Theses, https://doi.org/10.1007/978-981-10-8375-4_2
12 2 Spoof Surface Plasmon Modes Modeling Using Circuit Elements

2.1 Field Confinement of SSP Modes

Figure 2.1a depicts a symmetric SSP unit cell. The cell is composed of two grooves
with the depth of d and groove width of g. The period is p, the total cell width
is w and the thickness is t. Figure 2.1b, c plot the electric field distribution near
the symmetric SSP cells in vacuum with the following geometrical parameters:
p = 4 mm, w = 2p, d = 3.6 mm, g = 0.6p and t = 0.02 mm. As it can be seen, the
amplitude of the electric field reaches its maximum at the open end of the grooves,
while it is zero at its closed end. The electric field vector distribution is also plotted
in Fig. 2.1c.
The dispersion curves of the SSP cells and microstrip line are compared in
Fig. 2.1d for two cases of with and without the dielectric substrate. The substrate is
Rogers 4003 with the thickness of 1.5 mm. To achieve the 50 Ω input impedance,
the strip width of the microstrip line for the two cases of with and without the
substrate are set as 3.4 and 7 mm, respectively. As seen, with the substrate, the
wavenumber of the microstrip line at lower frequencies is slightly higher than the
SSP cells. This shows the strong effect of dielectric on the momentum in microstrip
lines. The majority of EM energy is distributed between the ground and the central
strip, where is filled with the substrate. However, by increasing the frequency, the
wavenumber of the slow-wave SSP modes increases drastically, which leads to
higher momentum and field confinement in comparison with the microstrip TL
modes.

Fig. 2.1 A symmetric SSP cell. a A period cell of the symmetric SSP structure. b Electric fields
distribution near the cell. b Amplitude and c vector field plots. d The dispersion curves and e field
distribution comparison between the SSP cells and microstrip lines with and without the dielectric
substrate. Inset The microstrip line along with the coordinate system. Figure adopted and
reproduced with permission from Ref. [71]
2.1 Field Confinement of SSP Modes 13

Figure 2.1e compares the normalized electric field distribution (|E|/|Emax|) of the
SSP cells and microstrip TLs in the x direction at z = −0.5 mm and f = 10 GHz,
where |E| is calculated by calculating an integral of the absolute value of the electric
field over one period:

Zp=2 rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 
 2
jEðx; zÞj ¼ jEx ðx; y; zÞj2 þ Ey ðx; y; zÞ þ jEz ðx; y; zÞj2 dy ðv/mÞ; ð2:1Þ
p=2

where, z = 0 and x = w/2. The coordinate system for the microstrip TL and SSP cell
are shown in Fig. 2.1e inset and Fig. 2.3a inset, respectively. As seen in Fig. 2.1e,
due to the larger wavenumber of the SSP modes at 10 GHz, the decay rates for
these modes are higher than the microstrip TLs. Moreover, adding a substrate
increases the wavenumber, which results in higher decay rate. The dispersion curve
and the spatial field decay of the SSP cells can be changed by changing their
geometrical parameters, which are discussed as follows.
Figure 2.2a plots the dispersion curves of SSP cells without dielectric substrate
for varying groove depths d, and groove widths, g. The other parameters are set as
p = 4 mm and w = 2p. As seen, d mainly affects the dispersion characteristics; while
the curves are almost independent of b. Increasing the groove depth increases the
momentum while it decreases the cut-off frequency. Figure 2.2b presents the dis-
persion curves for varying widths, w and a constant groove depth, d. The geo-
metrical parameters of the cells considered in this figure are as: p = 4 mm, g = 0.6p,
and d = 3.6 mm. As it can be seen, changing w does not change the wavenumber
significantly.
Figure 2.3a compares the field distribution of the SSP cells with different groove
depths at 12 GHz. This figure shows the normalized electric field, as described by
Eq. (2.1), over the z axis at x = 3.5 mm. For d = 0.9p, the decay rate of the fields is as
much as 12 dB at 0.2k distance from the cell, 4 dB higher than that for the cell with
d = 0.7p. In summary, increasing the groove depth enhances the field confinement.
The phase change of the wave in one period is Δu = bp, where b is the
wavenumber of the SSP modes. The frequency at which Δu = p is the cut-off
frequency of the symmetric SSP cells shown in Fig. 2.1a. In the following, the
effect of the period length on the wavenumber, cut-off frequency and field distri-
bution are studied. Figure 2.2c shows the dispersion curves of SSP modes for
varying period lengths: p = 2, p = 4 and p = 6 (all in mm). The other parameters are
set as: g = 0.6p, w = 8 mm and d = 0.45w. The abscissa is wavenumber without
normalization to the period length, due to variation of the period length in this plot.
As seen, the wavenumber is almost independent of the period length. The cut-off
frequency, however, increases by decreasing p. In other words, for an SSP cell with
smaller period length (p1 < p2), the phase change of the dominant spoof plasmon
mode in one period is smaller (Δu1 = bp1 < Δu2 = bp2) and consequently, the phase
change limit for its periodic structure changes to a higher frequency, which cor-
responds to a higher cut-off limit.
14 2 Spoof Surface Plasmon Modes Modeling Using Circuit Elements

Fig. 2.2 Dispersion study of

SSP cells. a Dispersion curves
of the symmetric SSP cells for
varying groove widths and
depths. b The effect of cell
widths on the dispersion
curves. c The effect of the
period length on the cut-off
frequency and momentum of
the SSP modes.
Figure adopted and
reproduced with permission
from Ref. [1]

Figure 2.3b depicts the field distribution at 10 GHz over the z axis at x = 3.5 mm
for the cells with different period lengths. The wavenumber at 10 GHz for the three
cases is identical (see Fig. 2.2c). Therefore, the field confinement is similar for
these cases and their electric fields decay at the same rate. Nevertheless, the cells
with shorter period can exhibit higher field confinement near their cut off
frequencies. To compare the maximum field confinement of the cells, the field
distributions of three cases are plotted at frequencies near their cut off in Fig. 2.3c:
p = 6 mm at 11 GHz, p = 4 mm at 13 GHz and p = 2 mm at 16 GHz. The abscissa
is normalized to the operating wavelength for each curve to make a fair comparison
between the results at different frequencies. As seen, the SSP cell with p = 2 mm
shows much higher field confinement in comparison with the other two cases,
which is expectable from the higher maximum momentum of SSP modes for the
shorter SSP cells (see Fig. 2.2c).
2.1 Field Confinement of SSP Modes 15

Fig. 2.3 Field distribution

of SSP cells. a Comparison
of the field confinement near
the symmetric SSP cells with
various groove depths. Inset
An SSP cell along with the
coordinate system. b The field
confinement near SSP cells
for various period lengths at
10 GHz. c The comparison
of the maximum field
confinement near SSP cells
with various period lengths.
The operating frequency
for each case is as follows:
p = 6 mm at f = 11 GHz;
p = 4 mm at f = 13 GHz;
and p = 2 mm at f = 16 GHz.
Figure adopted and
reproduced with permission
from Ref. [1]

In summary, the confinement of SSP modes is adjustable and strongly depends

on the geometries of the cells. This is an important feature of these structures that
enables highly controllable SSP-based components. Reliable circuit models for SSP
structures are required for their implementation in microwave circuit design. The
circuit models of various SSP structures are discussed as follows.

2.2 Principles of Circuit Modeling

The SSP cells with their highly confined EM energy could form an ideal trans-
mission line (TL) when they are arranged in periodic configurations. All trans-
mission lines require a circuit model to make them compatible with other circuit
components. In the conventional two-conductor TLs, the voltage is defined as the
16 2 Spoof Surface Plasmon Modes Modeling Using Circuit Elements

voltage difference between the core conductor and the ground plane. The modeling
of the single layer SSP structures has been remained a challenge due to lack of a
ground plane around them. Here, we propose a definition for the voltage at SSP
cells slightly different from the conventional cases. In a single conductor TL there is
no ground plane; however, we still can consider a ground plane very far from the
structure and the voltages can be defined as the voltage difference between the TL
and infinity. The current in the circuit model is defined as the surface current on the
conductor, similar to the definition of the current in a two-conductor TL.
For periodic structures in general, the following equations holds for input (Vn, In)
and output (Vn+1, In+1), i.e., voltages and currents of the n-th unit cell:

Vn þ 1 ¼ Vn ecp ðVÞ
ð2:2Þ
In þ 1 ¼ In ecp ðAÞ;

where p is the period length. The complex propagation constant is c = a + ib, where
b and a are the propagation and decay constants, respectively. The transmission
parameters of a unit cell are defined in a matrix as follows [2]:
    
Vn A B Vn þ 1
¼ : ð2:3Þ
In C D In þ 1

For a symmetric structure (A = D) and the complex propagation constant c is

calculated as:

cos hðc pÞ ¼ A: ð2:4Þ

Another important parameter to be extracted from the circuit model is the Bloch
impedance. The Bloch impedance is the characteristic impedance of the TL peri-
odically constructed by the SSP unit cells. This parameter ensures the compatibility
of the SW–TL with other transmission lines and is calculated as:

Vn þ 1 B
Zc ¼ ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffi : ð2:5Þ
In þ 1 A 1
2

The transmission parameters can be calculated from the measured scattering

parameters [2, 3].

2.3 Equivalent Circuit Models for SSP Cells

In this section, the proposed definitions for the voltage and current are applied to
derive the equivalent circuit model of the SSP modes. The circuit elements are
calculated based on the geometries of the cells. Here, three different types of planar
SSP cells, which have been applied in many works, are modeled.
2.3 Equivalent Circuit Models for SSP Cells 17

2.3.1 U-Shaped SSP Cells

Figure 2.4a depicts a unit cell of the SSP structure proposed in [4]. Each current
path of one unit cell is modeled as an inductor while the electric fields near the cell
indicate the presence of the capacitors in the model. The equivalent circuit model is
presented in Fig. 2.4a. The dimensions of the unit cell are specified in Fig. 2.4b.
The equivalent inductors and capacitors are calculated in static electromagnetics.
The inductance of the inductors is [5]:

l0 li 2li 1
Li ¼ ln þ ðHÞ; ð2:6Þ
2p wi þ t 2

Fig. 2.4 The circuit model for U-shaped SSP cells. a A unit cell of the SSP structure along with
the equivalent circuit model. b The dimensions of the SSP structure. c The equivalent areas of
inductors. d The equivalent areas of capacitors. e The dispersion curves in four background media.
Inset The Bloch impedance calculated for the SSP unit cells on Rogers 4003 1.52 mm.
Figure adopted and reproduced with permission from Ref. [3]
18 2 Spoof Surface Plasmon Modes Modeling Using Circuit Elements

where wi, t and li are the width, thickness and length of the metallic path for the
current, respectively. It is obvious that decreasing the width or increasing the length
of the current path increases the equivalent inductance. As the thickness is much
smaller than the width, changing the thickness hardly affects any equivalent
inductor. L1, L2 and L3 as shown in Fig. 2.4a are calculated using Eq. (2.6).
Figure 2.4c specifies the geometries of the inductors of the unit cell. In this figure,
the length of L3 (l3) is equal to half the width of the SSP unit cell, i.e., w as shown in
Fig. 2.4b.
There are two types of equivalent capacitors in the proposed model. One consists
of a pair of metallic plates (C1). This capacitor is calculated by electrostatic software
such as CST EM Studio. It is worthy to mention that C1 includes parts of the
neighboring cells as depicted in Fig. 2.4d. Simply, the capacitance increases as
increasing the area of the plate and decreasing the distance between two plates. The
distance between two plates (pe) is equal to the period length of the structure (p).
The other equivalent capacitors correspond to the connection of the proposed
structure to the ground (C2 and C3) and are calculated based on the proposed
voltage definition. These capacitors also model the leaky currents from the TL to
the infinite ground, similar to the leaky current in double-conductor TLs. Each
single metallic plate in electrostatics has a capacitance due to the electric fields from
the structure that end at infinity. As no ground plane exists around the SSP
structure, the accurate calculation of these elements is of great importance. For each
capacitor, the capacitance can be calculated by EM software and measuring the
capacitance of a single separate as well as a pair of metallic plates with the same
geometries, as indicated in Fig. 2.4d.
Table 2.1 presents the calculated circuit element values for the SSP with four
background media of different dielectric constants. The parameters of the cells are
w = p = 5 mm, g = 0.4p, d = 0.8p, and t = 0.02 mm. All the inductances are
identical because any change in the permittivity value does not affect the inductors.
However, the capacitance linearly changes with the dielectric constant. Therefore,
the capacitance for each medium is equal to the capacitance of the reference
medium with er = 1 multiplied by the relative permittivity of that medium.

Table 2.1 Circuit element parameters of the U-shaped SSP cells in media with different dielectric
coefficients
Case 1: er = 1 Case 2: er = 2 Case 3: er = 3 Case 4: er = 4
L1 (nH) 0.75 0.75 0.75 0.75
L2 (nH) 0.2 0.2 0.2 0.2
L3 (nH) 0.82 0.82 0.82 0.82
C1 (fF) 15 30 45 60
C2 (fF) 94 188 282 376
C3 (fF) 63 126 189 252
The dimensions are: w = p = 5 mm, d = 0.8p, g = 0.4p, and t = 0.02 mm (see Fig. 2.4b)
2.3 Equivalent Circuit Models for SSP Cells 19

For a better understanding of the capacitance of single plates, one can consider
simpler cases of circular metallic discs with the same areas as the rectangular
C2 and C3. For a metallic circular disc of the radius rc and the area of A, the
capacitance is [6]:
pffiffiffiffiffiffiffiffi
C ¼ 8er ¼ 8ec A=p ðFÞ: ð2:7Þ

So, the capacitance is proportional to the square root of the area of the metallic
plate. For Case 1, C2 and C3 from Eq. (2.7) are calculated to be 98 fF and 63 fF that
are very close to the simulated values by numerical electrostatic calculation in
Table 2.1.
The dispersion curves from the equivalent circuit model solved by Advanced
Design System, indicated as E. C., and the eigenmode solver of CST Microwave
Studio, indicated as CST, are plotted in Fig. 2.4e. The results indicate that the
cut-off frequency is higher for the media with lower permittivity. Excellent
agreement between the two methods demonstrates the accuracy of the proposed
equivalent circuit model. The difference between the results caused by two main
reasons: firstly, the equivalent elements are calculated in electrostatics and general
definitions of the voltage and current have been applied, which are not accurate at
high frequencies. Secondly, the approximation of the surface currents and charge
densities are slightly different from the real values at the periodic U-shaped
structure. For example, a constant voltage over the area of a capacitor is considered,
where this assumption is not very accurate at higher frequencies. In order to
improve the model, one can divide the unit cells into even smaller pieces in order to
account for the non-uniform voltage and current distribution. However, the model
will be much more complicated and therefore, not desirable for further engineering
design purposes. The current simple model provides a simple and viable tool for
designing SSP structures for specific frequency ranges.
Another adjustable parameter in the equivalent circuit model is the Bloch
impedance calculated by Eq. (2.5). By changing the parameters to achieve the 50-Ω
impedance, the structure will be more compatible with a conventional transmission
line while it is surely quite straightforward to tune the structure to achieve other
impedances. Here, the Bloch impedance of SSP modes is studied. The simulation
has been done for the SSP cells printed on a piece of Rogers 4003 slab (1.52 mm)
with an effective dielectric constant of er = 2 used in our experiment. The effective
permittivity for this Rogers substrate is calculated by CST Microwave Studio. For
this purpose, the dispersion curves of the structure for two cases are compared: the
SSP structure on Rogers 4003 (as a partially filled medium) and the SSP structure in
the fully filled medium with the effective permittivity. Our calculation is plotted in
Fig. 2.4e inset and shows an almost constant Bloch impedance of 42 X over a wide
frequency range for the SSP structure for the medium with er = 2 as a function of
frequency. These results indicate that SSP modes can be matched to any 50-X
transmission line.
20 2 Spoof Surface Plasmon Modes Modeling Using Circuit Elements

2.3.2 Symmetric SSP Cells

The symmetric SSP cells are double-side SSP structures and have been studied in
many works [7–12]. Here, we model these structures using circuit elements.
Figure 2.5a depicts the proposed model, where C1 is the capacitance between two
metal plates at the open end of each groove, and C2 and C3 are the capacitances
between the metal and infinity. All the capacitors are calculated using CST
EM-Studio. L1, L2 and L3 are the inductances caused by the current flowing the
metal surface and are calculated using Eq. (2.6).
Here, six SSP cells with the following geometrical parameters are modeled:
p = 4 mm, g = 0.6p and d = 0.45w, where w changes from 6 to 16 mm. The
geometrical parameters of the cell are shown in Fig. 2.1a. The cell is on a piece of
FR4 substrate with a thickness of 0.5 mm. The effective permittivity for this sub-
strate is calculated to be 1.7 [3] and the calculated circuit elements are tabulated in
Table 2.2.
To verify the model, the dispersion curve is calculated for the periodic
arrangement of the cells and the results are validated by the eigen mode solution
using CST Microwave Studio in Fig. 2.5b. To confirm the compatibility of the SSP
modes with the other transmission lines, the Bloch impedance is calculated for the
symmetric SSP cell in Table 2.2 with w = 8 mm and depicted in Fig. 2.5c, where a
constant 50-Ω impedance over a wide frequency range is observed.

Fig. 2.5 The circuit model for symmetric SSP cells. a The equivalent circuit model.
b Comparison between dispersion curve results from circuit model and eigen mode solver of
CST Microwave Studio. c The characteristic impedance of the symmetric SSP cells.
Figure adopted and reproduced with permission from Ref. [1]
2.3 Equivalent Circuit Models for SSP Cells 21

Table 2.2 Equivalent circuit w (mm) 6 8 10 12 14 16

elements for six symmetric
C1 (fF) 12 14 18 23 28 32
SSP cells with varying cell
widths C2 (fF) 50 80 100 120 130 140
C3 (fF) 40 70 80 90 100 110
L1 (nH) 1.2 1 1 0.9 0.8 0.76
L2 (nH) 0.04 0.02 0.004 0 0 0
L3 (nH) 0.54 0.8 1.1 1.5 1.8 2.2

2.3.3 Meander SSP Cells

The meander SSP cell is composed of two reversely attached U-shaped structures.
This configuration supports double grooved SSPs that promotes better connection
to the symmetric microwave transmission lines in comparison with single-sided
SSP cells. At the same time, it offer a lower profile in comparison with other double
sided grooved SSP cells such as the symmetric cells. Each groove of the meander
cell acts like a cavity; the incoming electromagnetic waves couple with these
cavities and slow down their propagation resulting in highly confined SSP modes
[13]. A single layered meander structure as depicted in Fig. 2.6a supports the SSP

Fig. 2.6 The circuit model for meander SSP cells. a One unit cell of meander SSP structure.
b The equivalent circuit model. c The effective area for the equivalent capacitors. d The effective
area of the current paths for the equivalent inductors
22 2 Spoof Surface Plasmon Modes Modeling Using Circuit Elements

modes and acts as low-profile SSP cells with great potential for performing as a
transmission line in microwave circuits.
Similar to the U-shaped and symmetric SSP cells, the meander cells require a
reliable and repeatable circuit model to increase their compatibilities with other
circuit components. Figure 2.6a depicts the meander SSP cell with the following
geometrical parameters: the period length of p, the width of l, the groove width of g,
the groove depth of d and the thickness of t. Figure 2.6b presents the circuit model,
where the capacitor C1 models the electric fields inside the groove and C2 and C3
initiate the connection of the cell to the ground plane. All the capacitors are being
modeled using electrostatic calculation in the CST EM Studios and Fig. 2.6c
depicts the corresponding metal area for each capacitor in the model.
The equivalent inductors L1 and L2 model the surface current at the SSP cells
and are being approximated by the effective current paths according to Fig. 2.6d,
where w1 is (w − d)/2 and l2 is d/3. The inductors are being calculated using
Eq. (2.6). Our calculation shows that the model is accurate for meander SSP cells
with 0.5w < p < 6w and 0.7w < d.
To validate the equivalent circuit, the SSP cells of various widths and period
lengths are modelled and the dispersion curves are compared with the results from a
full wave analysis. For this purpose, the periodic structure is solved using the
transient solver of CST Microwave Studios and the dispersion curves are calculated
by calculating the Fourier transform of the electric fields at different frequencies.
Figure 2.7a compares the results for various cell widths (l). The other parameters of
the SSP cells are as follows: p = 3 mm, d = 0.85l and g = 0.25p. Figure 2.7b depicts
the results for different period lengths (p). The other parameters are as follows:
l = 2.3 mm, d = 0.85l, and g = 0.25p. The comparison validates the model. The
values for the equivalent circuit elements of the SSP cells studied in Fig. 2.7a, b are
proposed in Table 2.3.
According to Fig. 2.7a, b, the phase difference in one period of the meander SSP
structure is not limited to p radian. Consequently, In comparison with U-shaped [3]
and symmetric SSPs [1, 14], the meander SSP cells do have no sharp cut off
frequency at bp = p.
The proposed circuit model is helpful to find the characteristic impedance of the
periodic arrangement of the cells [2]. Figure 2.7c plots the real and imaginary parts
of impedance for an SSP cell of the following parameters: p = 4 mm, l = 5 mm,
d = 0.85l and g = 0.25p. Similar to the U-shaped [3] and symmetric SSP cells [1],
the real part is almost constant over the frequency range from 2 to 7 GHz. However,
in contrast to the zero imaginary part of the impedance in the two SSP cells
mentioned above, Xc for the meander cells is positive exhibiting inductive char-
acteristics and increases from 0 to 7 Ω over the frequency bandwidth of 2–7 GHz.
2.4 SSP-Based Circuit Design Using Circuit Models 23

Fig. 2.7 Meander SSP cells:

dispersion curves and
characteristic impedance. The
comparison between the
dispersion results of the SSP
modes in meander cells
calculated from equivalent
circuit model and full wave
analysis for various a cell
widths and b cell period
lengths. c The characteristic
impedance of the meander
SSP cell

Table 2.3 Circuit element w (mm) 1 2.3 5 2.3 2.3 2.3

parameters for meander SSP
p (mm) 3 3 3 2 6 14
cells
C1 (fF) 4 9 23 10 9 11
C2 (fF) 19 29 43 24 41 63
C3 (fF) 11 17 26 14 24 37
L1 (nH) 0.5 0.4 0.3 0.2 1 2.9
L2 (nH) 0.05 0.2 0.7 0.3 0.1 0.03

2.4 SSP-Based Circuit Design Using Circuit Models

The proposed circuit models for the SSP structures are effective tools enabling their
implementation into microwave circuits. In this section, we apply the equivalent
circuit model to design broadband loading for SSP-based slow wave transmission
24 2 Spoof Surface Plasmon Modes Modeling Using Circuit Elements

lines (SW-TLs). The loaded TLs are applied for the mutual coupling characteri-
zation of SW-TLs in Sect. 3.4.
The SSP structures are capable of transferring the electromagnetic energy and
composing the low-profile and groundless slow-wave transmission lines (SW-TLs),
which are studied in the next chapter. To load a conventional double-line trans-
mission line, one can put a resistor between the lines. However, this method is not
applicable for the single-line SW-TLs. Here, we propose a method to solve this
dilemma by delivering the power of the SSP modes to the broadband loads. The
current is redirected to the loading resistors that are placed in the current path.
Figure 2.8a depicts the current distribution on the SW-TL. As seen, the current
amplitude is maximum between the grooves. gap is made at the maximum current
location, while a resistor is placed on top of the gap.
Figure 2.8b presents a truncated SW-TL with two loaded cells at its end. The
loading resistors are R1 and R2. The loaded SSP cell is modelled by adding a
resistor in series with L1 in Fig. 2.5a, without changing the other elements. The
modified model is presented in Fig. 2.8c.
The full-wave analysis of the loaded SW-TLs for optimization of R1 and R2 is
time and memory consuming; however, the implementation of the circuit model is a
fast and efficient alternative for this purpose. Figure 2.8d presents the schematic of
the loaded SW-TL. This configuration is composed of 10 SSP cells connected to two
loaded cells by R1 and R2. The transmission matrix of each block is depicted inside it

Fig. 2.8 Loading design for symmetric SW-TLs. a Current distribution at the SW-TL. b Loaded
SW-TL with two resistors. c Block diagram representation of a loaded SW-TL. d The circuit
model for a loaded SSP cell. Figure adopted and reproduced with permission from Ref. [1]
2.4 SSP-Based Circuit Design Using Circuit Models 25

and can be calculated from the circuit model. The input and output ports are tagged
in the figure. The transmission matrix of the whole structure is as follows:

½Tt  ¼ ½T      ½T  ½T2   ½T1 : ð2:8Þ

[T], [T1] and [T2] are the transmission matrices of SSP cells with no loading,
loaded with R1 and loaded with R2, respectively. The reflection from the input port
or S11 is calculated from the Zin as:

Zin  Z0
S11 ¼ ; ð2:9Þ
Zin þ Z0

where Z0 = 50 Ω. Zin is the input impedance of the two port network in Fig. 2.8d
when the port 2 (output) is open corresponding to the definition of Z11. Z11 relates to
the transmission matrix as [2]:

Z11 ¼ A/C, ð2:10Þ

 
A B
where ½Tt  ¼ .
C D
Here, the loading resistors are chosen for broadband matching of the SW-TL
composed of symmetric SSP cells with p = 4 mm, d = 0.45w and b = 0.6p, and
w = 8 mm. The equivalent circuit model is derived in Table 2.2. The substrate is
Rogers 4003 with a thickness of 1.5 mm. To study the loading termination of the
SW-TL and exclude the energy loss by the substrate and metal, all materials are
chosen as lossless. The capacitors calculated in the circuit model linearly depend on
the permittivity of the filling medium, while the inductors do not [3]. Consequently,
all the circuit elements calculated in Table 2.2 can be used here; however, the
capacitors are multiplied by the ratio of the effective permittivity of the new
medium (2 for Rogers 4003 1.5 mm) to that of the original medium (1.7 for FR4
0.5 mm). The method for calculation of the effective permittivity is discussed in
Sect. 2.3.1. Figure 2.9a plots the reflection spectrum for four sets of R1 and R2,
given in Table 2.4. As seen, set 4 gives the lowest reflection among them and
results in a broadband matching for the frequency range of 5.5–8.5 GHz. These
results confirm the possibility of broadband termination of a single-line SW-TL by
the resistive loads.
To evaluate the method, full wave analysis has been performed for the loaded
SW-TLs with the resistors given in Table 2.4 and the results are presented in
Fig. 2.9b. The agreement between the results in Fig. 2.9a, b endorses the accuracy
of the application of SSP circuit models in the microwave circuit design.
26 2 Spoof Surface Plasmon Modes Modeling Using Circuit Elements

Fig. 2.9 Reflection results

from loaded SW-TLs.
a Circuit model and b full
wave simulation results. The
loading resistors for the four
sets are tabulated in
Table 2.4. Figure adopted and
reproduced with permission
from Ref. [1]

Table 2.4 Four sets of Set 1 Set 2 Set 3 Set 4

resistors
R1 (in Ω) 300 70 300 50
R2 (in Ω) 100 50 25 25

2.5 Conclusion

We have studied the field confinement of the spoof plasmon modes of symmetric
SSP cells. The effect of the various geometrical parameters on the maximum field
confinement is discussed. It has been concluded that increasing the groove depth
increases the momentum of SSP modes. Moreover, it has been shown that
decreasing the period length increases the cut off frequency and consequently,
increases the maximum field confinement of the SSP modes.
We have proposed the method to derive the equivalent circuit model for single
metal SSP structures. The voltage for groundless SSP cells has been defined as the
voltage difference between the metal and infinity. This definition has been applied
to establish the connection of the SSP cells with the ground. The equivalent circuit
elements have been calculated using electrostatic approximation and accordingly,
the circuit models have been derived for the U-shaped, symmetric and meander SSP
cells. We have applied the equivalent model for calculation of the dispersion curves
and characteristic impedance of the periodic composition of the cells. The results
have been validated using commercial software.
The equivalent model is advantageous in the circuit design using the SSP
structures. Here, we have applied the model to design the broadband loading for the
SSP-based slow wave transmission lines. The design method has been validated
2.5 Conclusion 27

using full-wave solution by CST Microwave Studio. With high field confinement
and the constant input impedance of *50 X for a wide range of frequencies, along
with the reliable and repeatable circuit model, the SSP structures are able to provide
a plausible option for inexpensive and small transmission lines.

References

1. A. Kianinejad, Z.N. Chen, C.W. Qiu, Low-loss spoof surface plasmon slow-wave
transmission lines with compact transition and high isolation. IEEE Trans. Microw. Theory
Tech. 64(10), 3078–3086 (2016)
2. D.M. Pozar, Microwave Engineering, 3rd edn. (Wiley, New York, 2004)
3. A. Kianinejad, Z.N. Chen, C.-W. Qiu, Design and modeling of spoof surface plasmon
modes-based microwave slow-wave transmission line. IEEE Trans. Microw. Theory Tech. 63
(9), 1817–1825 (2015)
4. X. Shen, T.J. Cui, D. Martin-Cano, F.J. Garcia-Vidal, Conformal surface plasmons
propagating on ultrathin and flexible films. Proc. Natl. Acad. Sci. 110(1), 40–45 (2013)
5. Z. Piatek, B. Baron, T. Szczegielniak, D. Kusiak, A. Pasierbek, Self inductance of long
conductor of rectangular cross section. Przeglad Elektrotechniczny Electr. Rev. 88(8),
323–326 (2012)
6. J.D. Jackson, Classical Electrodynamics, vol. 3 (New York: Wiley, 1975), p. 128, problem 3.3
7. H.F. Ma, X. Shen, Q. Cheng, W.X. Jiang, T.J. Cui, Broadband and high-efficiency conversion
from guided waves to spoof surface plasmon polaritons. Laser Photonics Rev. 8(1), 146–151
(2013)
8. G.S. Kong, H.F. Ma, B.G. Cai, T.J. Cui, Continuous leaky-wave scanning using periodically
modulated spoof plasmonic waveguide. Sci. Rep. 6, 29600 (2016)
9. B.C. Pan, Z. Liao, J. Zhao, T.J. Cui, Controlling rejections of spoof surface plasmon
polaritons using metamaterial particles. Opt. Express 22(11), 13940–13950 (2014)
10. Y.J. Zhou, B.J. Yang, Planar spoof plasmonic ultra-wideband filter based on low-loss and
compact terahertz waveguide corrugated with dumbbell grooves. Appl. Opt. 54(14),
4529–4533 (2015)
11. B. Xu, Z. Li, L. Liu, J. Xu, C. Chen, C. Gu, Bandwidth tunable microstrip band-stop filters
based on localized spoof surface plasmons. JOSA B 33(7), 1388–1391 (2016)
12. J.Y. Yin, J. Ren, H.C. Zhang, Q. Zhang, T.J. Cui, Capacitive-coupled series spoof surface
plasmon polaritons. Sci. Rep. 6, 24605 (2016)
13. A. Kianinejad, Z.N. Chen, C.-W. Qiu, Full modeling, loss reduction and mutual coupling
control of spoof surface plasmon based meander slow wave transmission lines. IEEE Trans.
Microw. Theory Tech. (Submitted)
14. A. Kianinejad, Z.N. Chen, C.-W. Qiu, Design and modeling of low-loss symmetric
slow-wave transmission lines, in 2015 Asia-Pacific Microwave Conference (APMC), vol.
2 (2015), pp. 1–3
Chapter 3
SSP-Based Transmission Lines (TLs)

With high field confinement and a constant input impedance of *50 X for a wide
range of frequencies in addition to their single layered, low profile and flexible
configurations, the SSP structures are able to provide a plausible platform for cost
effective and compact microwave circuits. It is necessary to design an appropriate
transition to couple the SSP modes to the conventional transmission lines and
realize the slow-wave transmission lines (SW–TL).

3.1 Conversion of SSP Modes to Guided Waves

In order to obtain highly efficient connection of the SSP structures to the con-
ventional TLs, two transmission structures should be matched in terms of impe-
dance, polarization, and momentum. This section studies the conversion of the SSP
modes to the guiding waves.

3.1.1 Polarization Matching

Figure 3.1a presents the U-shaped SSP structure connected to a microstrip line
through the mode converter. The design procedure is presented in Sect. 3.2.1. The
microstrip line supports the Quasi-TEM modes, while the SSP modes are TM
polarized. Figure 3.1b, c show the normal (|Ez|) and azimuthal (√(|Ex|2 + |Ey|2))
magnitudes of electric field at six planes perpendicular to the mode converter,
respectively. These results clearly demonstrate the gradual transformation from
Quasi-TEM modes to TM modes via the proposed mode converter (at 7 GHz). It
can be observed from Fig. 3.1 that the normal component of Quasi-TEM waves at

© Springer Nature Singapore Pte Ltd. 2018 29

A. Kianinejad, Metamaterial Surface Plasmon-Based Transmission Lines
and Antennas, Springer Theses, https://doi.org/10.1007/978-981-10-8375-4_3
30 3 SSP-Based Transmission Lines (TLs)

Fig. 3.1 Polarization transformation of mode converter. a The proposed SSP structure connected
to the mode converter to construct the SW–TL. b The normal and c the azimuthal components of
the electric field at six perpendicular planes to the mode converter. Figure adopted and reproduced
with permission from Ref. [1]

the microstrip line (as shown in Fig. 3.1b-Plane 1) is almost zero, while the major
part of the electric field is the azimuthal component (Fig. 3.1c-Plane 1). Figure 3.1b
depicts a gradual increase in the normal component of the electric field via the mode
converter. This component becomes the major electric field at the SSP structure as
shown in both Fig. 3.1b, c-Plane 6 [1].

3.1.2 Momentum Matching

The mode converter acts as a momentum converter while it transforms the polar-
ization. The microstrip line supports the waves with the light-line dispersion
3.1 Conversion of SSP Modes to Guided Waves 31

Fig. 3.2 Dispersion conversion. Inset The separate parts of the mode converter. Figure adopted
and reproduced with permission from Ref. [1]

relation and the wavenumber of k0; however, the dispersion curve for the SSP
structure (Fig. 2.4e) indicates much larger wavenumber for the SSP modes.
Momentum conversion is achieved through gradually increasing the wavenumber
as the wave propagates through the mode converter. Figure 3.2 presents the dis-
persion curve at various points along the mode converter. These curves are obtained
by considering a periodic configuration of each section of the mode converter
separately. These separate sections are specified in Fig. 3.2 inset. This figure
clarifies the gradual increase of the momentum of the wave propagating through the
mode converter.

3.1.3 Impedance Matching

The impedance is another important parameter that should be smoothly matched via
the mode converter. In Fig. 2.7c we showed that the characteristic impedance of the
meander SSP cells is complex and the meander cells exhibit an inductive behavior.
The schematic of the mode converter for connection of meander SSP cells to the
CPW line is depicted in Fig. 3.3a. The mode converter is composed of unit cells
with the same period length but with the gradually increasing width and groove
depth. This design is discussed with more details in Sect. 3.2.3. This configuration
provides smooth polarization, momentum and impedance matching.
To study the impedance transformation, the equivalent circuit models of the cells
in the mode converter are needed. The model is similar to the other cells except for
32 3 SSP-Based Transmission Lines (TLs)

Fig. 3.3 Impedance matching of mode converter. a The proposed mode converter for the meander
SW-TL. b The effective area for the equivalent capacitors. c Matching of the impedance through
the mode converter

the effect of the ground plane in the mode converter, which is taken into account by
C2g and C3g as the capacitance between the ground plane of the CPW line and the
SSP cell. Figure 3.3b identifies the approximated surface area and the distance
between them for these two capacitors. To include them into the model in Fig. 2.6b,
the capacitances C2 and C3 in the model change to C20 and C30 , respectively as
follows:

C20 ¼ C2 þ C2g ;
ð3:1Þ
C30 ¼ C3 þ C3g :

By applying the modified circuit model for the mode converter, the impedance
matching is examined. The imaginary part of the impedance Xc for four cells tagged
in Fig. 3.3a as m1 to m4 is plotted in Fig. 3.3c. As the wave propagates through the
mode converter, Xc gradually increases to match the inductive meander SW-TL
from zero to positive.
3.2 Transition Design from SSPs to Conventional TLs 33

3.2 Transition Design from SSPs to Conventional TLs

In Chap. 2, we studied the modeling of U-shaped, symmetric and meander SSP

cells. Each SSP cell has its own unique features which make it appealing for
specific application. In this section, the design procedures to construct transmission
lines using different SSP cells are discussed.

3.2.1 Transition from U-Shaped Cells to Microstrip TLs

Here, the mode converter is designed in order to connect U-shaped SSP unit cells to
a 50-X microstrip line on a piece of 1.52 mm Rogers 4003 substrate to construct the
SW–TL for the reflection lower than −12 dB.

3.2.1.1 Structure

The mode converter is composed of two parts: the strip connector and the ground
regulator as shown in Fig. 3.4a. These two parts are designed based on the
parameters of the microstrip line and the SSP unit cell. The dimensions of the SSP
cell are w = p = 5 mm, h = 0.8p g = 0.4p (Fig. 2.4b). To achieve the 50-Ω
impedance for the design on Rogers 4003 1.52 mm, the microstrip line width is set
as ws = 3.2 mm. Figure 2.4e shows that the cutoff frequency (fc) for the SSP
structure on a piece of 1.52 mm Rogers 4003 substrate with the effective epsilon of
2 is fc = 9.1 GHz. This frequency can be changed by scaling the SSP dimensions.
Here, we design the mode converter to efficiently operate at the bandwidth of
5–8 GHz. All the parameters are presented with respect to the period length (d).

Fig. 3.4 Mode converter design for U-shaped SSP cells. a The SSP structure connected to the
microstrip line via matching transitions to build the SW-TL. b The specified parameters for the
strip connector. Figure adopted and reproduced with permission from Ref. [1]
34 3 SSP-Based Transmission Lines (TLs)

Consequently, the working frequency range for both the mode converter and the
SSP structure can be scaled by changing the period length of the SSP unit cells.
Figure 3.4b shows the strip connector, which includes three SSP unit cells. The
effect of the number of unit cells on the performance of the converter will be
discussed later. The parameters of the strip connectors are specified in Fig. 3.4b and
are related to each other based on the formulas given in Table 3.1. These relations
can be applied to design the transitions for SSPs working at other frequency ranges.
The width of the strip decreases from ws to w1 = ws/x and then increases to
w2 = w. These changes make two major effects on the fields: first, it helps the
transformation of the polarization (Fig. 3.1b from Planes 1 to 4); secondly, the
increase in the strip width from w1 to w2 compensates for the increase at the
distance between the ground and the strip where the ground is gradually removed.
This keeps the impedance of the line almost the same for the rest of the microstrip
line guiding modes before their complete transformation to the SSP modes and
therefore, reduces their reflection.
S1 and S2 are the required lengths for the gradual width change and chosen
according to the equations in Table 3.1, where d1, d2, d3, g1, g2, g3, l1, l2 and l3 are
the dimensions of the three separate parts of the strip connector as shown in
Fig. 3.1d inset and chosen to achieve a gradual momentum increase as it is depicted
in Fig. 3.1d and the optimized relations for the best performance of these param-
eters are given in Table 3.1. Referring to Table 3.1, the crucial parameters to design
the strip connector include: Δd, Δg, ΔS and x.
The ground regulator regulates the discontinuity of the ground plane and is
located where the ground is removed. On both the sides of the strip, the edges of the
ground regulator have the functionality of y = A(erx − 1) as shown in Fig. 3.5a,
where both r and A are the design parameters. One can select the initial values so
that the edges of the ground regulator follow the change of the strip connector
properly. For simplicity, we chose A1 and A2 to be p/500(0.01 mm) and p/5(1 mm)
for the two sides, respectively and further selectivity can be achieved by tuning r for
the smallest reflection and highest transmission. The ground regulator truncates at
the half length of the strip connector:

G1 ¼ 0:5ðS2 þ g1 þ l1 þ g2 þ l2 þ g3 þ l3 Þ ð3:2Þ

Consequently, the parameters of the ground regulator are found to be r1 and r2.

Table 3.1 Relation between d1 = d − 2  Δd g1 = g l1 = l = p − g

parameters of strip connector d2 = d − 1  Δd g2 = a + 2  Δg l2 = l − 1  Δl
in Fig. 3.4b d3 = d g3 = a + 1  Δg l3 = l − 2  Δl
Δl = Δg
S1 = 1.5  p − ΔS w1 = ws/x
S2 = p − ΔS w2 = w
3.2 Transition Design from SSPs to Conventional TLs 35

Fig. 3.5 Mode converter

design for U-shaped SSP
cells. a The ground regulator,
b Simulated |S21| and c |S11|
for five parameter sets
(Table 3.2) for the entire
frequency range.
Figure adopted and
reproduced with permission
from Ref. [1]

3.2.1.2 Tuning and Optimization

The parameters Δd, Δg, ΔS, x, r1 and r2 are tuned for high transmission and low
reflection results. Δd, Δg and ΔS are small fractions of p. Here, we chose
Δd = Δg = ΔS = p/16 (or 0.25 mm). The parameter x determines the strip width
change and the initial value of 3 can be used for designing the transition to any SSP
structure. The parameters r1 and r2 determine the shape of the ground regulator and
their initial values are chosen as following: r1 = 10.5/p (or 2.1 mm−1) and
r2 = 0.75/p (or 0.15 mm−1). Figure 3.5b, c study the effect of these geometrical
parameters on the scattering results for five parameter sets listed in Table 3.2. As it
can be seen, the SW-TL with Set 5 achieves the best transmission performance
among the five sets.
Figure 3.6a presents the absolute values of the surface current on the metallic
surfaces of the structure at 7 GHz. As seen, gradually removing the ground at the
connection of the SSP structure and the microstrip TL (the matching part on the
ground plane) prevents the discontinuity of the surface currents and consequently,
reduces the possible reflection caused by this abrupt change.
36 3 SSP-Based Transmission Lines (TLs)

Table 3.2 Five parameter Set 1 Set 2 Set 3 Set 4 Set 5

sets for mode converter
Δd p/20 p/20 p/20 0 p/20
Δg p/20 p/20 0 p/20 p/20
ΔS p/20 p/20 0 p/20 p/20
x 3 1 3 3 3
r1 5/p 10.5/p 10.5/p 10.5/p 10.5/p
r2 0.5/p 0.75/p 0.75/p 0.75/p 0.75/p

Fig. 3.6 Effect of ground connector. a Surface current on the SW–TL at 7 GHz, b Comparison of
the S-parameter results (simulation) of SW-TL with and without port connection. Figure adopted
and reproduced with permission from Ref. [1]

The both ports of the designed SW-TL have a common ground connected by a
metallic connection as shown in Fig. 3.6a. Figure 3.6a shows that the current
density on this metallic bar is at least 25 dB smaller than the maximum current
density at the SW-TL. In order to ensure that this metallic connection does not have
any significant effect on the performance of the groundless transmission line, the
design is compared with the same SW-TL but without any ground connection in
Fig. 3.6b. The results indicate that this connection does not change the transmission
significantly.
The strip connector consists of straight lines. We anticipate better results by
substituting the curve lines of ground regulator by the straight lines as it is shown in
Fig. 3.7a. G2 = p/2.8 = 1.8 mm is tuned for good transmission. For better perfor-
mance, “Interpolated Quasi Newton” optimization method (in CST Microwave
3.2 Transition Design from SSPs to Conventional TLs 37

Fig. 3.7 Mode converter

design for U-shaped SSP
cells. a Substitution of curved
lines of the ground regulator
by straight lines.
b S-parameter results of SW–
TL with curved ground
regulator, straight ground
regulator and SW–TL with
parameters given in
Table 3.3. c S-parameter
results for SW–TL with
single, two, three and four
unit cells strip connectors.
Figure adopted and
reproduced with permission
from Ref. [1]

Studio) is performed on the whole parameters. Figure 3.7b compares the results
from the both SW-TLs with curved and straight ground regulators as well as the
results from final optimization. The parameters after final optimization are pre-
sented in Table 3.3.
The ideal transition should be composed of an infinite number of gradually
changing SSP elements; however, we should find the most effective transition with
as few elements as possible. The strip connector in Fig. 3.4a is composed of three
SSP elements. The most important role of these elements is to gradually increase
the momentum and Fig. 3.2 indicates that three elements provides smooth

Table 3.3 Parameters of d1 3.5 g1 2 l1 3.2

U-shaped SW–TL after
d2 3.4 g2 2.5 l2 2.8
optimization (in mm)
d3 3.8 g3 2.3 l3 2.7
w1 0.5 S1 7 G1 10.1
w2 5 S2 4.7
38 3 SSP-Based Transmission Lines (TLs)

Fig. 3.8 Experimental

evaluation of U-shaped
SW-TL. a The fabricated
prototype. b Comparison
between the experimental and
simulation S-parameters
results. Figure adopted and
reproduced with permission
from Ref. [1]

momentum matching and consequently, increasing the number of matching ele-

ments does not improve the transmission and reflection. Figure 3.7c compares the
S-parameter results for four types of mode converters: single-element, two-element,
three-element, and four-element strip connectors. The results show that the strip
connector with three elements has better performance than single-element and
two-element types. However, increasing the number of elements to four does not
change the transmission significantly. Consequently, due to its shorter length,
three-element mode converter is chosen.
The S-parameters were measured by a vector network analyzer. Figure 3.8a
shows the fabricated SW-TL connected to the vector network analyzer and Fig. 3.8b
presents the results. |S21| for the SW–TL is calibrated based on a 50-X microstrip line
on a similar substrate material, in order to cancel out the connection loss and
dielectric loss as well as to study the role of the mode converter and the SSP structure
on the transmission. The difference between the measured and simulated |S11| is
attributed to the effect of Sub Miniature version A (SMA) connectors, which is not
considered in the simulation. The experiment validates the simulation results and
confirms that the designed SW-TL operates as an efficient transmission line.

3.2.1.3 Transition from the Symmetric Cells to CPW Lines

This subsection presents the design procedure for the symmetric SSP-based
SW-TLs. Fig. 3.9a depicts the proposed compact mode converter for connection to
3.2 Transition Design from SSPs to Conventional TLs 39

Fig. 3.9 Mode converter design for symmetric SSP cells. a The proposed compact design. b The
conventional mode converter. Figure adopted and reproduced with permission from Ref. [2]

CPW lines. The mode converter is composed of two parts: the strip connector and
the ground regulator. The former connects the strip to the symmetric SSP cells,
while the latter regulates the ground removal. The ground plane flared by an
exponential function (y = erx − 1 where x and y are in millimeter) as indicated in
Fig. 3.9a. The geometrical parameters for optimization are w1, w2, l1, l2, l3 and r.
Here, the mode converter is designed for the symmetric SSP Cell with the
following geometrical parameters: p = 4 mm, d = 0.45w and g = 0.6p, w = 8 mm
(see Fig. 2.1a). The substrate is FR4 with a thickness of 0.5 mm. The strip width
(wc) and the gap between the ground and the strip (gc) of the CPW line are set to be
2.3 and 0.3 mm, respectively, to achieve the 50-Ω characteristic impedance.
Five sets of geometrical parameters for the mode converter are chosen as tab-
ulated in Table 3.4. To study the mode converter performance, other possible losses
are factored out of the simulation; the substrate is considered lossless and the metal

Table 3.4 Five parameter sets for mode converter

Set 1 Set 2 Set 3 Set 4 Set 5
w1 wc/1.1 wc/1.1 wc/1.5 wc/1.1 wc/1.1
w2 wc/2.5 wc/2.5 wc/4.6 wc/4.6 wc/4.6
l1 0.8  p 0.8  p 0.8  p 0.5  p 0.8  p
l2 p p p 0.7  p P
l3 0.47  p −0.75  p 0.47  p 0.47  p 0.47  p
r 0.84/p 1.56/p 1.56/p 1.56/p 1.56/p
40 3 SSP-Based Transmission Lines (TLs)

is modelled as perfect electric conductor (PEC). The whole SW-TL with the total
length of 11 cm is simulated using transient solver of CST Microwave Studio. The
parameters are given in terms of the strip width of CPW (wc) and the period length
of the SSP cells and consequently, can be scaled to other SSP cells. Figure 3.10a
plots the scattering parameters, where the highest transmission (higher than
−1.5 dB) for a wide frequency range of 5–9 GHz is achieved for set 5. The total
length of the designed transition is 12.4 mm or 0.2k, where k is the wavelength in
the free space at 5 GHz.
In [3] a transition has been proposed for the connection of the symmetric SSP
cells to CPW lines. While the transition is proven to be highly efficient in con-
version of guiding modes to SSP modes, it requires to be sufficiently long to
achieve its expected performance. As a benchmark, a conventional transition
similar to that in [3] but with the same total length as our compact mode converter is
simulated as presented in Fig. 3.9b. The conventional transition is composed of
three SSP cells and its total length is 12 mm. The scattering parameter result for this
mode converter is tagged as ‘Conv.’ in Fig. 3.10a. Due to the short length of the
conventional transition, the conversion is not efficient especially at lower fre-
quencies, while the proposed compact mode converter presents an acceptable
performance for the wide frequency range.
The designed SW-TL is fabricated on a piece of FR4 substrate with the thickness
of 0.5 mm. To experimentally evaluate the performance of the mode converter, the
measured scattering parameter results for the SW-TL are presented in Fig. 3.10b.

Fig. 3.10 S-parameter results

of symmetric SW-TL.
a Simulated |S21| for five
parameter sets given in
Table 3.4 as well as the
SW-TL with the conventional
mode converter.
b Experimental evaluation of
the design. Figure adopted
and reproduced with
permission from Ref. [2]
3.2 Transition Design from SSPs to Conventional TLs 41

A good agreement between the results confirms the efficient conversion of the
guided modes of CPW line to the surface waves of the SW-TL.

3.2.1.4 Transition from Meander Cells to CPWs

This subsection discusses the design of meander SW-TLs. A groundless trans-

mission line based on meander spoof surface plasmons (SSPs) is constructed as
shown in Fig. 3.11a. The parameters of the cell (Fig. 3.11a) are set as d = 4 mm,
w = d + 2 mm, g = 2.4 mm, t = 0.02 mm and p = 8 mm. This SSP cell is chosen
to form an SW-TL with a cutoff frequency of 5 GHz.
Here, the SW-TL is fed by a coplanar waveguide (CPW) as shown in Fig. 3.11a
on a piece of Rogers 4003 with the thickness of 1.5 mm. The strip width (wc) is
2.3 mm and the gap size between the strip and the ground (gc) is chosen to be
0.2 mm to achieve a 50-Ω characteristic impedance. The SW-TL is connected to a
CPW line via a mode converter.
In order to realize a gradual matching of momentum between the light line
(for CPW) and SSP modes (for SW-TL), the mode converter is composed of ten
similar unit cells with gradually increasing width (width w = wc + Sl  y :
Sl = (w − wc)/10p) and groove depths (linearly increasing from 0 to d) [3]. The
ground of the CPW is also flared by an exponential function (x = A(ery − 1)),
where r = 44 and A = 0.001. The high conversion efficiency between the SSP
structures and the CPW is verified by the high transmission and low reflection
depicted in Fig. 3.11b.

Fig. 3.11 Meander SW-TL.

a The schematic SW-TL.
Inset the dimensions of one
SSP unit cell. b The simulated
transmission and reflection
from the designed SW-TL.
Figure adopted and
reproduced with permission
from Ref. [3]
42 3 SSP-Based Transmission Lines (TLs)

3.3 Loss in SW-TLs

Transmission lines are the building blocks of the microwave circuits and their loss
significantly affects the performance of the whole systems. Consequently, it is a
matter of great importance to consider this crucial aspect in design of the TLs and
optimize the transmission lines for the minimum loss values. This section studies
the various types of loss in SW-TLs.
The loss in the SW-TLs includes the conversion loss (Losscon by the mode
converter) and the transmission loss (Losst by the SSP cells):

Loss ¼ Losscon þ Losst ð3:3Þ

The transmission loss is composed of the leaky loss (LossLeaky) and the ohmic loss
(Losso):

Losst ¼ LossLeaky þ Losso ð3:4Þ

3.3.1 The Leaky Loss

Radiation from the transmission line causes the leaky loss and is proportional to the
length of the SW-TL. To study this type of loss, the transmission results of the
SW-TLs with lossless materials and with different lengths are compared. With their
lossless dielectric and metal, the ohmic loss of these TLs is zero, the conversion loss
is constant and the difference between the transmission results is the leaky loss.
Figure 3.12a depicts the |S21| for the SW-TLs with the length of k, 6k and 16k and
composed of symmetric SSP cells with the following parameters: p = 4 mm,
d = 0.45w and b = 0.6p, w = 8 mm (see Fig. 2.1a). k is the wavelength in the free
space at 5 GHz. The substrate is FR4 with a thickness of 0.5 mm. The constant
transmission results for various lengths in this figure indicate that the leaky loss is
negligible for the SW-TL, which is expected of highly confined SSP modes.

3.3.2 Ohmic Loss

To examine the ohmic loss caused by lossy dielectric in symmetric SW-TLs, two
SW-TLs with different lengths of L1 = 12 mm and L2 = 160 mm are fabricated on
a piece of FR4 with a thickness of 0.5 mm and loss tangent of 0.013 with sym-
metric SSP cells of the following geometrical parameters: p = 4 mm, d = 0.45w
and g = 0.6p, w = 8 mm (see Fig. 2.1a). The scattering parameters are measured
using a vector network analyzer. The measured |S21| includes connection loss
(caused by SMA connectors), conversion loss, leaky loss and ohmic loss. The leaky
3.3 Loss in SW-TLs 43

Fig. 3.12 Loss in symmetric

SW-TLs. a Comparison
between |S21| results for
symmetric SW-TLs with
different lengths. b Simulation
and measured insertion loss
of symmetric SW-TL.
Figure adopted and
reproduced with permission
from Ref. [2]

loss in the SW-TL is negligible. To exclude the connection loss and conversion loss
from the results, the results are offset by the |S21| of the shorter SW-TL to indicate
the ohmic loss of an SW-TL with the length of L2 − L1 = 148 mm. Figure 3.12b
depicts the ohmic loss per 10k, where k is the free space wavelength at 5 GHz. As a
bench mark, the ohmic loss of a CPW line on the same substrate is plotted in
Fig. 3.12b. The measured ohmic loss for the SW-TL confirms the simulation
prediction and shows much lower loss in comparison with the CPW line for the
wide frequency range of 4–7.5 GHz.

3.4 Mutual Coupling Between SW-TLs

This section studies the mutual coupling (MC) between two closely located sym-
metric SW-TLs with loading termination according to Fig. 3.13a. The loading
design method was discussed in Sect. 2.4. The SW-TLs are composed of symmetric
SSP cells with the following geometrical parameters: p = 4 mm, d = 0.45w and
g = 0.6p, w = 8 mm (see Fig. 2.1a). The overlapping part length is Lmc and the
distance between the two SW-TLs is dmc. As a benchmark, the mutual coupling
between loaded microstrip transmission lines on the same substrate is calculated.
The strip width of the microstrip TL is 3.4 mm for a 50-Ω intrinsic impedance.
Figure 3.13b depicts the mutual coupling at 8.5 GHz for various overlapping
part lengths, Lmc. The abscissa is normalized to the free space wavelength at the
44 3 SSP-Based Transmission Lines (TLs)

Fig. 3.13 Mutual coupling

between SW-TLs. a The
fabricated prototype of the
two closely located and
loaded SW-TLs. Comparison
of the MC between SW-TLs
and microstrip lines for
various b overlapping part
lengths and c distance
between two SW-TLs.
Figure adopted and
reproduced with permission
from Ref. [2]

operating frequency. The distance between the two SW-TLs is dmc = 0.3k. The MC
increases by increasing Lmc for the both cases; however, the MC between the
SW-TLs is around 2 dB better than that between the microstrip TLs. This is
expected from the high field confinement of the SSP modes.
Increasing the field confinement decreases the interference between the EM
fields of the closely located TLs and therefore, decreases the MC between them.
According to the discussion given in Sect. 2.1, the maximum field confinement of
the SSP modes increases by decreasing their period lengths. Here, the effect of the
period length on the MC between the SW-TLs is studied. The period length of the
SSP cell is changed to half its original value. The geometrical parameters of the cell
are as follows: p = 2 mm, w = 8 mm, d = 0.45w and g = 0.6p. The MC between
3.4 Mutual Coupling Between SW-TLs 45

two SW-TLs for varying distances between two SW-TLs at 10 GHz is presented in
Fig. 3.13c. The overlapping part length is k and the MC is around 10 dB lower than
the microstrip TLs.
To evaluate the results, the closely located SW-TLs are fabricated for three dmc
values of 4 mm, 7 mm, and 10 mm. The fabricated prototype is displayed in
Fig. 3.13a. The measured MC is compared with the simulation in Fig. 3.13c and
validates the simulation prediction. The lower MC between the SSP cells with
shorter period length is in accordance with their higher maximum field confinement
plotted in Fig. 2.3c. The SW-TL with much lower crosstalk in comparison with the
conventional TLs is a perfect choice for compact integrated microwave circuits.

3.5 Conclusion

We have designed efficient and compact transitions from different SSP structures to
conventional transmission lines. The proposed transitions act as a mode converters
providing smooth conversion of momentum, polarization and impedance for
maximized coupling of the modes and minimized reflection. Through the mode
converters, the quasi TEM guided waves gradually transform to the TM polarized
SSP modes. The momentum of the guided waves increases smoothly to match the
high momentum SSP modes. For the meander SSP cells with inductive behavior,
the proposed mode converter gradually introduces the positive imaginary part to the
characteristic impedance to provide the required impedance matching for efficient
connection to 50-Ω TLs.
We have proposed and experimentally evaluated various slow wave transmis-
sion lines with much better performance in comparison with the conventional TL.
The simulation has showed that the leaky loss in the proposed SSP SW-TL is
negligible. The proposed SW-TL has been experimentally evaluated as a wideband
and low-loss TL showing up to half of the ohmic loss of conventional TLs. The MC
between the SW-TLs has been numerically as well as experimentally proved to be
up to 10 dB lower than the microstrip transmission lines. The proposed low loss
and highly isolated SW-TL along with its simplified and applicable models hold the
promises for implementation in future microwave circuits.

References

1. A. Kianinejad, Z.N. Chen, C.-W. Qiu, Design and modeling of spoof surface plasmon
modes-based microwave slow-wave transmission line. IEEE Trans. Microw. Theory Tech. 63
(9), 1817–1825 (2015)
2. A. Kianinejad, Z.N. Chen, C.W. Qiu, Low-loss spoof surface plasmon slow-wave transmission
lines with compact transition and high isolation. IEEE Trans. Microw. Theory Tech. 64(10),
3078–3086 (2016)
46 3 SSP-Based Transmission Lines (TLs)

3. H.F. Ma, X. Shen, Q. Cheng, W.X. Jiang, T.J. Cui, Broadband and high-efficiency conversion
from guided waves to spoof surface plasmon polaritons. Laser Photonics Rev. 8(1), 146–151
(2013)
4. A. Kianinejad, Z.N. Chen, L. Zhang, W. Liu, C.W. Qiu, Spoof plasmon-based slow-wave
excitation of dielectric resonator antennas. IEEE Trans. Antennas Propag. 64(6), 2094–2099
(2016)
Chapter 4
Spoof Surface Plasmon Excitation
of Dielectric Resonator Antennas

Dielectric resonators have been of great interest for many years [1–3]. These
resonators were first proposed to function as radiators in 1983 and named as
dielectric resonator antennas (DRAs) [4]. DRAs are usually considered to possess
striking advantages owing to their much lower loss than their conductive coun-
terparts. This loss only results from the imperfect dielectric materials, and thus
facilitates highly efficient operation especially at microwave, millimeter-wave and
terahertz bands [5].
This chapter designs a spoof surface plasmon feeding structure for the DRAs to
excite their TE dominated modes [6]. These modes are experimentally and
numerically demonstrated and their unique features and merits including the
reduced thickness-dependency of the resonance frequency and the horizontally
polarized omnidirectional radiation are thus verified. Our proposed method opens a
vista to find new potential applications of DRAs in antenna designs.

4.1 Resonance Modes of Cylindrical DRAs

The two fundamental modes of a cylindrical DRA are HEM11d and TE01d modes.
Figure 4.1 compares the resonance frequencies fr of TE01d and HEM11d modes for
various antenna thicknesses (T). The diameter and the relative permittivity are
25 mm and 25, respectively. TE01d mode also possesses lower eigen frequency than
HEM11d mode. Moreover, TE01d mode is less dependent on the thickness variation
due to its radial resonance nature [7]. This lower thickness-dependency enables
realization of planar and low frequency DRAs that overcomes the current limita-
tions in antenna engineering and monolithic microwave integrated circuit (MMIC)
designs [8, 9].

© Springer Nature Singapore Pte Ltd. 2018 47

A. Kianinejad, Metamaterial Surface Plasmon-Based Transmission Lines
and Antennas, Springer Theses, https://doi.org/10.1007/978-981-10-8375-4_4
48 4 Spoof Surface Plasmon Excitation of Dielectric Resonator Antennas

Fig. 4.1 Resonance 15

frequencies fr of cylindrical TE
01

DRAs. Comparison between HEM

10 11

f r (GHz)
fr of TE01d and HEM11d for
various antenna thicknesses.
Figure adopted and 5
reproduced with permission
from Ref. [6]
0
1 2 3 4 5 6
T (mm)

It has been verified that the horizontally polarized transmitting and receiving
antennas with omnidirectional coverage for indoor communication systems result in
a 10-dB higher power gain in comparison with their vertically polarized counter-
parts [10, 11]. Most horizontally polarized omnidirectional antennas are quite
sophisticated and are realized by combining different types of antennas [10, 12–14].
Fortunately, the fundamental mode TE01d of a DRA with its simplified feeding
design offers a promising alternative to achieve the desired horizontal polarization
with omnidirectional radiation.
The electric fields of TE01d mode circulate azimuthally. The field vectors are
tangential to the circular plane of the cylinder as seen in Fig. 4.2a, due to the high
permittivity dielectric boundary conditions. However, with the introduction of a
PEC as the ground plane in many traditional feeding methods, such as the coaxial
probe feeding [4], direct microstrip feedline [15], aperture-coupling [16], and
co-planar feeding [17], the fundamental mode TE01d cannot be effectively excited
owing to the absence of tangential electric fields. Instead, the lowest order mode
observed in these systems is the HEM11d mode. In the HEM11d mode, the tangential
components of the electric fields almost disappear at the lower end of the DRA and
are dominated by the vertical component Ez as seen in Fig. 4.2b.

Fig. 4.2 Field distribution of

a cylindrical DRA. a The
schematic of an isolated DRA
and the electric field vector
distributions of the
fundamental mode (TE01d).
b The schematic of a DRA on
a ground plane and the
electric field vector
distributions of the nominal
fundamental mode (HEM11d).
Figure adopted and
reproduced with permission
from Ref. [6]
4.1 Resonance Modes of Cylindrical DRAs 49

The excitation of the higher order modes in DRAs has also been of interest
[18, 19]. Applications of the dominant modes for polarization diversity have been
studied in [20]. In [21], HEM12d mode is excited simultaneously with the con-
ventional modes to achieve a broad-side radiation pattern.
Ground based feeding configurations have been proposed to excite TE01d mode
in cylindrical DRAs for filtering [22, 23] and radiation [24, 25] purposes. The
proposed feeding configuration in [24] achieves an omnidirectional radiation pat-
tern for the horizontal components at 3.9 GHz, while the ground plane in this
method affects the resonance frequency and the field distribution of the excited
mode. Eigen mode solution shows that the same DRA without the feeding network
resonates at 3 GHz at TE01d mode. In other words, the resonance frequency in the
presence of the feeding configuration is shifted by 30%. Moreover, the far-field
beam is titled upward and therefore, the main radiation direction is positioned 45°
to the horizontal plane. Consequently, effective excitation of TE01d mode has
always remained a challenge in DRA feedings. In this chapter, a groundless slow
wave transmission line is applied to feed a DRA in order to eliminate the ground
effects in the field distribution and farfield pattern, as well as effectively excite
TE01d mode.

4.2 Feeding Configuration Design

To excite the fundamental mode TE01d, a groundless transmission line based on

SSPs in meander cells, similar to Fig. 3.11a is designed. The SW-TL composed of
meander unit cells connected to two CPW lines through the mode converter. The
mode converter is composed of 10 unit cells with gradually increasing groove depth
(d) and the width (l). The ground plane flares with an exponential function. The
designed procedure is detailed in Sect. 3.2.3. The cutoff frequency of the SW-TL
can be readily tuned by adjusting the dimensions of the unit cells. An SSP cell with
the period length of p = 8 mm, metal thickness of t = 0.02 mm, groove depth of
d = 9 mm, groove width of g = 2.4 mm and cell width of l = d + 2 mm (Fig. 4.3a
inset) is chosen to form an SW-TL with a cutoff frequency of 5 GHz for the DRA
fundamental mode excitation.
Here, the SW-TL is fed by a coplanar waveguide (CPW) as shown in Fig. 4.3a
on a piece of Rogers 4003 1.5 mm. To have a 50-Ω characteristic impedance, the
strip width (wc) is 2.3 mm and the gap size between the strip and the ground (gc) is
0.2 mm (see Fig. 3.11a).
A DRA resonating at TE01d mode is dominated by Hz [7] and acts similar to a
magnetic dipole in an axial direction. The magnetic vector fields within one unit cell
of the SW-TL are depicted in Fig. 4.3b, where the maximum field is located inside
the two cavities of the meander cell and is perpendicular to the unit cell. In order to
50 4 Spoof Surface Plasmon Excitation of Dielectric Resonator Antennas

Fig. 4.3 The proposed meander SW-TLs. a The schematic of the design. Inset The dimensions of
one SSP unit cell. b The magnetic field distribution at two cutting planes. Figure adopted and
reproduced with permission from Ref. [6]

Fig. 4.4 A DRA fed by SW-TL with four orientations. Arrows represent the magnetic field
vectors. a The DRA’s axial direction is parallel to the SW-TL. b The DRA is placed on top of the
SW-TL and its axial direction is parallel to the grooves of the SSP cells. c The DRA is placed on
top of the SW-TL and its axial direction is perpendicular to the SW-TL. d The DRA is placed on
top of the SW-TL and its axial direction is parallel to the SW-TL. Figure adopted and reproduced
with permission from Ref. [6]

efficiently excite a DRA with the SW-TL, the magnetic fields of both TE01d mode
and the SSP mode should be parallel to each other within an optimized area.
Figure 4.4a–d show four different orientations to feed a cylindrical DRA by the
proposed SW-TL. A side view of each configuration is also presented for better
understanding. Only in Fig. 4.4c the magnetic fields of both structures are in the
same direction. This orientation is chosen for excitation of the dominant mode.
4.3 Excitation of the TE Modes 51

4.3 Excitation of the TE Modes

The designed SW-TL is loaded with a cylindrical DRA according to the configu-
ration in Fig. 4.4c and is schematically shown in Fig. 4.5a. A cylindrical DRA with
the fundamental TE01d mode resonating at 3 GHz is chosen, where the thickness is
T = 4 mm. The same feeding configuration can be applied to excite the higher order
TE-dominant modes, where the details are presented in Appendix A. The electric
field distribution of the fundamental mode with the feeding structure aside, as
shown in Fig. 4.5a inset, indicates the slight difference from the ideal one plotted in
Fig. 4.2a inset, which is an effect of the presence of the SW-TL near the DRA. For
an optimized excitation, the overlapping area of electric fields between the SSP
mode and the target mode should be maximized, which can be achieved by finely
tuning x1 and x2. According to the scattering parameter results in Fig. 4.5c, x1 and
x2 are set to be 3.9 and −1.5 mm, respectively to achieve the lowest reflection at the
resonance frequency.
To study the effect of the feeding configuration on the resonance frequency, the
resonance frequency of an isolated DRA is calculated. The eigen mode solution

Fig. 4.5 A DRA fed by meander SW-TL. a The design configuration. Inset: the simulated electric
field vector of the fundamental mode TE01d and the feeding end. b A DRA fed by microstrip
transmission line. c The S-parameter results for various DRA locations. Figure adopted and
reproduced with permission from Ref. [6]
52 4 Spoof Surface Plasmon Excitation of Dielectric Resonator Antennas

Fig. 4.6 Excitation of a

cylindrical DRA.
a Comparison of S-parameter
results among five cases:
simulation results for a DRA
fed with the conventional
microstrip TL and with the
SW-TL (T = 4 mm and
T = 1.5 mm), experimental
results for a DRA fed with the
SW-TL (T = 4 mm).
b Photographs of the
fabricated SW-TL feeding
configuration and the DRA.
Figure adopted and
reproduced with permission
from Ref. [6]

shows the resonance frequency of 3 GHz for TE01d mode for the same DRA on the
same substrate without presence of the SW-TL. This corresponds to the results in
Fig. 4.5c and confirms that the feeding effect on the resonance frequency is neg-
ligible. The reflection spectrum is also measured experimentally, plotted in
Fig. 4.6a and good agreement with the simulation is observed. The simulation
results indicate 2.8% impedance bandwidth, while the measured bandwidth is 7%.
The difference between the results may be due to dielectric constant tolerance of the
DR material (MCT-25) [26] and change of the substrate thickness during the
fabrication process. Two photographs of the fabricated prototype are presented in
Fig. 4.6b.
As a benchmark, the same DRA is fed by a traditional microstrip (Mic.) TL as
seen in Fig. 4.5b. The lowest order mode of a cylindrical DRA excited by
microstrip TL changes into HEM11d mode and its resonance frequency shifts to
4 GHz. According to the reflection spectra in Fig. 4.6a, it is verified that TE01d
mode shows less thickness-dependence. When the DRA thickness decreases from 4
to 1.5 mm, only a 33% frequency shift is obtained for TE01d mode, while the shift
amount is as large as 60% for HEM11d mode.
4.4 The Radiation Performance 53

4.4 The Radiation Performance

To explore its radiation properties, both the simulated and measured Eh and Eu
components of the radiated far fields (realised gain in dBi) of TE01d mode are
plotted in Fig. 4.7 in the x-y, x-z and y-z planes for the cases fed with the microstrip
transmission line (simulation) and SW-TL. As seen, the fundamental mode pro-
duces horizontally polarized omnidirectional radiation patterns. According to
Fig. 4.2a, the electric field vectors of TE01d mode circulate along the / direction,
which mainly contribute to the far field in the / direction. As expected, the mag-
nitudes of E/ components (co-polarized component) of the radiated field, depicted
in Figs. 4.7a1, b1 and c1 are much higher than the counterparts of Eh components
(cross-polarized component), plotted in Fig. 4.7a2, b2 and c2 (except for the h = 0°
and 180°, where there is a null in the radiation pattern) in all the corresponding
cutting planes.
The far-field radiation patterns of HEM11d have both the vertical and horizontal
components, owing to the presence of both the Ez and E/ components for the fields
inside the DRA at this mode. Presence of the ground plane blocks the lower space

Fig. 4.7 Far field radiation patterns of DRA (realised gain in dBi). Comparison among three
cases: Simulation results for the conventional microstrip transmission line feeding technique and
for SW-TL feeding method, and the experimental results for SW-TL feeding method a1 Eu in x-
y plane. a2 Eh in x-y plane. b1 Eu in x-z plane. b2 Eh in x-z plane. c1 Eu in y-z plane. c2 Eh in y-
z plane. Figure adopted and reproduced with permission from Ref. [6]
54 4 Spoof Surface Plasmon Excitation of Dielectric Resonator Antennas

and breaks the symmetry of the radiation pattern. Consequently, the HEM11d cannot
radiate towards the -z direction and the directivity decreases drastically for h > 90◦.
However, TE01d mode covers both the upper and lower space equally. The maxi-
mum realized gain of 2.5 dBi is achieved in the x-y plane. The experimental results
agree well with the simulations in all the three planes and indicate the efficient
excitation of the fundamental TE01d mode.
The DR material is MCT-25 with a loss tangent of 0.0015 [26] and the substrate
is Rogers 4003 1.524 mm. The simulation results for the total efficiency for the
DRA with loss but the lossless substrate and perfect PEC is 94%. When including
substrate loss and copper loss, the total antenna efficiency is 87%.

4.5 Conclusion

A novel slow-wave feeding method based on SSPs has been proposed to eliminate
the effects of the ground plane on the excitation of operating modes of DRAs and
therefore, excite their fundamental horizontally polarized modes. It has been shown
that the conventional ground-based feeding methods are not capable of exciting the
real fundamental mode of DRAs. The method has been applied to feed a cylindrical
DRA and the fundamental TE01d mode is excited with better performance in
comparison with the previous works. Both the resonance frequency and far-field
radiation patterns have been measured and the experiment matches well with the
simulation.
The conventional methods for the generation of omnidirectional radiation pat-
terns with horizontal polarization include complex arrangement of different
antennas. Here, the horizontally polarized omnidirectional radiation patterns have
been verified experimentally and numerically for a simple DRA fed by SW-TLs. In
addition, the SW-TL feeding method enables excitation of planar DRAs. Being
successfully applied to feed cylindrical DRAs, the proposed groundless technique
could be a promising solution to minimize undesired effects of the ground plane in
antenna systems.

References

1. T. Nobis, E.M. Kaidashev, A. Rahm, M. Lorenz, M. Grundmann, Whispering gallery modes

in nanosized dielectric resonators with hexagonal cross section. Phys. Rev. Lett. 93(10),
103903 (2004)
2. S. Chang, A.G. Mann, A.N. Luiten, D.G. Blair, Measurements of radiation pressure effect in
cryogenic sapphire dielectric resonators. Phys. Rev. Lett. 79(11), 2141 (1997)
3. A.A. Savchenkov, V.S. Ilchenko, A.B. Matsko, L. Maleki, Kilohertz optical resonances in
dielectric crystal cavities. Phys. Rev. A 70(5), 051804 (2004)
4. S.A. Long, M.W. McAllister, L.C. Shen, The resonant cylindrical dielectric cavity antenna.
IEEE Trans. Antennas Propag. 31, 406–412 (1983)
References 55

5. Q. Lai, G. Almpanis, C. Fumeaux, H. Benedickter, R. Vahldieck, Comparison of the radiation

efficiency for the dielectric resonator antenna and the microstrip antenna at Ka band. IEEE
Trans. Antennas Propag. 56(11), 3589–3592 (2008)
6. A. Kianinejad, Z.N. Chen, L. Zhang, W. Liu, C.W. Qiu, Spoof plasmon-based slow-wave
excitation of dielectric resonator antennas. IEEE Trans. Antennas Propag. 64(6), 2094–2099
(2016)
7. A. Petosa, Dielectric Resonator Antenna Handbook (Artech House, 2007)
8. K.W. Leung, K.Y. Chow, K.M. Luk, E.K.N. Yung, Low-profile circular disk DR antenna of
very high permittivity excited by a microstripline. Electron. Lett. 33(12), 1004–1005 (1997)
9. A. Petosa, A. Ittipiboon, Dielectric resonator antennas: a historical review and the current state
of the art. Antennas Propag. Mag. IEEE 52(5), 91–116 (2010)
10. C. Soras, M. Karaboikis, G. Tsachtsiris, V. Makios, Analysis and design of an inverted-F
antenna printed on a PCMCIA card for the 2.4 GHz ISM band. Antennas Propag. Mag. IEEE
44(1), 37–44 (2002)
11. D. Chizhik, J. Ling, R.A. Valenzuela, The effect of electric field polarization on indoor
propagation, in IEEE 1998 International Conference in Universal Personal Communications,
1998, vol. 1, pp. 459–462
12. X. Qing, Z.N. Chen, Metamaterial-based wideband horizontally polarized omnidirectional
5-GHz WLAN antenna array, in Antennas and Propagation (EuCAP), 2014 8th European
Conference on, 2014, pp. 605–608
13. R.M. Honda, R.R. Johnson, Horizontally-polarized omni-directional antenna. Google
Patents, 2011
14. K. Wei, Z. Zhang, Z. Feng, M.F. Iskander, A MNG-TL loop antenna array with horizontally
polarized omnidirectional patterns. IEEE Trans. Antennas Propag. 60(6), 2702–2710 (2012)
15. R.A. Kranenburg, S.A. Long, Microstrip transmission line excitation of dielectric resonator
antennas. Electron. Lett. 24(18), 1156–1157 (1988)
16. K.W. Leung, K.M. Luk, E.K.N. Yung, Spherical cap dielectric resonator antenna using
aperture coupling. Electron. Lett. 30(17), 1366–1367 (1994)
17. R.A. Kranenburg, S.A. Long, J.T. Williams, Coplanar waveguide excitation of dielectric
resonator antennas. IEEE Trans. Antennas Propag. 39, 119–122 (1991)
18. D. Guha, A. Banerjee, C. Kumar, Y.M.M. Antar, New technique to excite higher-order
radiating mode in a cylindrical dielectric resonator antenna. IEEE Antennas Wirel. Propag.
Lett. 13, 15–18 (2014)
19. D. Guha, P. Gupta, C. Kumar, Dualband cylindrical dielectric resonator antenna employing
and modes excited by new composite aperture. IEEE Trans. Antennas Propag. 63(1), 433–438
(2015)
20. L. Zou, C. Fumeaux, A cross-shaped dielectric resonator antenna for multifunction and
polarization diversity applications. IEEE Antennas Wirel. Propag. Lett. 10, 742–745 (2011)
21. D. Guha, A. Banerjee, C. Kumar, Y.M.M. Antar, Higher order mode excitation for high-gain
broadside radiation from cylindrical dielectric resonator antennas. IEEE Trans. Antennas
Propag. 60(1), 71–77 (2012)
22. S.B. Cohn, Microwave bandpass filters containing high-Q dielectric resonators. IEEE Trans.
Microw. Theory Tech. 16(4), 218–227 (1968)
23. E.H. Lim, K.W. Leung, Use of the dielectric resonator antenna as a filter element. IEEE Trans.
Antennas Propag. 56(1), 5–10 (2008)
24. L. Zou, C. Fumeaux, Horizontally polarized omnidirectional dielectric resonator antenna, in
Microwave Conference Proceedings (APMC), 2011 Asia-Pacific, 2011, pp. 849–852
25. L. Zou, D. Abbott, C. Fumeaux, Omnidirectional cylindrical dielectric resonator antenna with
dual polarization. IEEE Antennas Wirel. Propag. Lett. 11, 515–518 (2012)
26. MCT-25*. [Online]. Available: http://www.trans-techinc.com/products_detail.asp?ID=116&
Name=MCT-25*
Chapter 5
Spoof Surface Plasmon-Based
Leaky-Wave Antenna (LWA)

Radiation from travelling wave structures was first proposed by Hansen [1]. With
their non-resonant nature, travelling wave antennas offer unique features such as
high gain, wide operation bandwidth, and frequency scanning beam for applications
in millimeter wave and microwave systems [2–4].
Usually, leaky-wave antennas are terminated by a broadband load to absorb the
non-radiated power and prevent it from reflecting back to the antenna and dis-
turbing desired radiation patterns. This termination reduces the antenna efficiency,
especially at higher frequencies where non-radiated power increases. H. V. Nguyen
et al. proposed a power recycling feedback system to solve this issue for maxi-
mizing the antenna efficiency [5]. However, this additional power recycling unit
increases the total size as well as the total ohmic loss of the antenna.
Among LWAs, the planar versions have been of more interest due to their
low-profile configurations [6–10]. Recently, the planar LWAs designed by com-
posite right/left handed (CRLH) structures have been proposed to realize a wide
bandwidth of the consistent broadside gain (4.2%) [11]. This design as well as
many other CRLH material based leaky-wave antennas require an arrangement of
metal patches and via holes with two metallic layers.
Recently, spoof surface plasmon modes have been of interest in antenna engi-
neering [12–20]. In [14], an array of microstrip patch antennas are fed by an
SP-based waveguide. In [18], additional gaps are implemented between the cells of
an SP-based transmission line to convert the slow waves to fast radiating waves,
and design an SP-based leaky wave antenna with fixed radiation patterns.
In this chapter, the spoof surface plasmon modes are implemented to design a
single-layered leaky-wave antenna (SL-LWA). With a simple and single layer
configuration, the proposed design offers all the advantageous features of the
conventional leaky-wave antennas such as frequency scanning beam, forward,
broadside and backward radiations as well as broadband operation for broadside
radiation. With very low non-radiative power at the end of the antenna, the pro-
posed leaky-wave antenna does not require any loading termination. In the next

© Springer Nature Singapore Pte Ltd. 2018 57

A. Kianinejad, Metamaterial Surface Plasmon-Based Transmission Lines
and Antennas, Springer Theses, https://doi.org/10.1007/978-981-10-8375-4_5
58 5 Spoof Surface Plasmon-Based Leaky-Wave Antenna (LWA)

section, the radiation from the proposed single-layered leaky-wave antenna is

studied. In Sect. 5.2, a simple design procedure for the SL-LWA is proposed. In
Sect. 5.3, the radiation performance of the antenna is experimentally evaluated.

5.1 Radiation Mechanism

The operation principle of the periodic leaky-wave antennas is based on generation

of a radiating space harmonic [2]. Here, the operating leaky-wave mode of the
SSP-based structures is employed to design a new type of leaky-wave antennas.

5.1.1 Single-layered Leaky-Wave Antenna

The meander-line slow wave transmission line (SW-TL) is composed of periodic

meander SSP cells connected to CPW lines via two converters and is depicted in
Fig. 5.1a. The meander SSP cell is presented in Fig. 5.1a inset with the geometrical
parameters as follows: the period length is p, groove depth and groove width are
d and g, respectively and the total width is l. The design procedure of the meander

Fig. 5.1 Meander SW-TL. a The schematic of the design. Inset: the meander SSP cell.
b Scattering parameter results from the meander SW-TL. Figure adopted and reproduced with
permission from Ref. [19]
5.1 Radiation Mechanism 59

SW-TL is detailed in Sect. 3.2.3. The simulation results for the scattering param-
eters from the designed meander-line SW-TL are presented in Fig. 5.1b. The
substrate is Rogers 4003 with the thickness of 1.5 mm and the loss tangent of
0.0027 and the geometrical parameters of the cell are p = 12 mm, d = 2 mm,
g = p/2 and l = 5 mm. The cut off frequency for this cell is 9 GHz, as seen in
Fig. 5.1b. Below 9 GHz, the meander-line SW-TL operates in the transmission line
mode with transmission above −0.5 dB and reflection below −15 dB. For the
frequencies range of 9–10.4 GHz transmission is suppressed to below −30 dB with
high reflection, while above 10.4 GHz, reflection decreases with negligible
transmission.
Figure 5.2a depicts the dispersion curves for the two modes in the meander-line
SW-TL. These results are calculated by taking the Fourier transform of the waves
on the SW-TL at various frequencies. The space harmonic corresponds to the Bloch
mode with n = −1 and the wavenumber (b−1):

b
1 ¼ b0
2p=p; ð5:1Þ

where b0 is the wavenumber of the dominant mode. The red dashed lines in
Fig. 5.2a indicate the fast wave limit with lower wavenumber than the free space k0.
For the frequency range of 10.4–24.5 GHz, the first space harmonic lies in the
radiation range, where the reflection and the transmission are very low, according to

Fig. 5.2 Radiation from the SW-TL. a Dispersion curves for the dominant mode and the first
space harmonic. b Normalized co-component (u) radiation patterns of the SL-LWA in the y-z
plane at frequencies. c 3D radiation pattern at 17 GHz. Figure adopted and reproduced with
permission from Ref. [19]
60 5 Spoof Surface Plasmon-Based Leaky-Wave Antenna (LWA)

the results in Fig. 5.1b. Therefore, the energy leaks from the transmission line and
the SW-TL acts as a single-layered LWA (SL-LWA).
The main beam radiation direction hmax is calculated from the radiating space
harmonic dispersion curve as below:
8

>
> b
1
>
< Arcsin
k0
hmax ¼
 ðdegreeÞ ð5:2Þ
>
> b
1
> 
: 180
Arcsin
k0

The first angle corresponds to radiation in the upper space (z > 0) while the
second angle corresponds to that in the lower space (z < 0).
Figure 5.2b depicts the normalized radiation patterns of the SL-LWA for the
co-component (u) in the y-z plane. At 10.5 GHz the main beam is titled toward
hmax = −120°. As the frequency increases from 10.5 to 17 GHz, the radiation
direction changes from backward to broadside radiation. This is in accordance with
the results in Fig. 5.2a, where the momentum of the radiating space harmonics
increases from a negative value at 10.5 GHz to zero at 17 GHz. Figure 5.2c plots
the three-dimensional radiation pattern at 17 GHz. By increasing the frequency
from 17 to 24.5 GHz, the momentum increases and the radiation beam moves from
broadside to forward direction (hmax = 30°). These results are in accordance with
Eq. (5.2).

5.1.2 SL-LWA Without 2nd Converter

The scattering parameter results in Fig. 5.1b shows that the jS21 j of the SL-LWA
within the radiation range from f = 10.4 to 24.5 GHz is lower than −12 dB. In other
words, majority of the energy over the entire bandwidth radiates and the transmitted
energy to Port 2 is almost zero. This is an advantage of the proposed design in
comparison with other types of LWAs, where the transmitted power increases by
increasing the frequency and the radiation efficiency of the antenna decreases
significantly at higher frequencies.
The SL-LWA is composed of two converters as seem in Fig. 5.1a. To study the
effect of the 2nd converter on the radiation, this part is removed (see Fig. 5.3a), and
for the two cases of with and without the 2nd converter, the reflection spectrum and
radiation performance are compared in Fig. 5.4a, b, respectively. Removing the 2nd
converter does not affect the reflected power, radiation angle, and maximum gain.
In other words, the majority of the electromagnetic energy has been radiated before
reaching the 2nd converter in Fig. 5.1b while this part of antenna does not have a
significant role in the radiation performance. Besides that, without any loading
termination at the end of the SL-LWA, the performance of the antenna remains
unchanged.
5.1 Radiation Mechanism 61

Fig. 5.3 Designed SL-SW-TL. a SL-LWA without the second converter. b The magnetic field
distribution near the SL-LWA. Figure adopted and reproduced with permission from Ref. [19]

Fig. 5.4 Effects of the 2nd

mode converter. a The
reflection spectrum, b the
maximum radiation direction
and the gain. Figure adopted
and reproduced with
permission from Ref. [19]

5.1.3 Effect of Dielectric and Metal

To study the effect of the dielectric substrate on the radiation performance, the
antenna without the substrate is simulated and the scattering parameter results are
presented in Fig. 5.5a. Removing the substrate changes the impedance bandwidth
to 12.7–35 GHz and increases the bandwidth from 80 to 94%. The second con-
verter of the SL-LWA without dielectric is removed, similar to the configuration in
Fig. 5.3a. Figure 5.5b shows the maximum radiation angle and the maximum gain
of the antenna in the y-z plane and indicates that the SL-LWA without the sup-
porting dielectric still operates as a leaky-wave antenna.
Figure 5.6a shows the total efficiency of the SL-LWA with and without the
substrate. For the SL-LWA with substrate, three cases are studied with different lt
62 5 Spoof Surface Plasmon-Based Leaky-Wave Antenna (LWA)

Fig. 5.5 Dielectric substrate

effect of SL-SW-TL. a The
scattering parameter results.
b The maximum radiation
angle and the gain.
Figure adopted and
reproduced with permission
from Ref. [19]

Fig. 5.6 Efficiency of

SL-LWA. a Total efficiency
with and without the dielectric
substrate. b The Radiation
efficiency for three cases:
SL-LWA (1) with lossy metal
and dielectric, (2) with
lossless dielectric and (3) with
PEC. Figure adopted and
reproduced with permission
from Ref. [19]
5.1 Radiation Mechanism 63

(see Fig. 5.3a). For the SL-LWA without substrate, the dielectric loss is zero and
the antenna efficiency is about 95% for the entire frequency range. This figure
indicates that at 23.6 GHz, the efficiency decreases to 90%. At this frequency, the
main beam points at hmax = 0° and the radiation is broadside as seen in Fig. 5.5b.
The slight reduction of the antenna efficiency is due to the slight increase of the
reflection at the broadside frequency, according to Fig. 5.5a. Although, the antenna
efficiency is above 90% over the entire bandwidth showing the excellent radiation
performance of the SL-LWA without the substrate.
The efficiency of the SL-LWA with a 1.5-mm thick Rogers 4003 dielectric
substrate layer reduces to 82%. The antenna efficiency reduces due to the additional
losses introduced by the substrate. Moreover, according to the results in Fig. 5.6a,
changing lt does not affect the total antenna efficiency significantly. Although, this
parameter affects the antenna gain and is studied in Sect. 5.2. According to the
results in this figure, the antenna efficiency with the dielectric substrate is changing
over the frequencies; however, it is above 80% for the entire bandwidth.
To further study the effect of the omhic losses caused by dielectric and metal on
the antenna performance, the radiation efficiency of the SL-LWA for three cases are
studied in Fig. 5.6b:
(1) the SL-LWA with lossy metal and on a lossy substrate layer tagged as “Lossy”;
(2) the SL-LWA with lossy metal and on a lossless substrate layer tagged as
“Lossless substrate”;
(3) the SL-LWA with PEC as metal and on a lossy substrate layer tagged as
“PEC”.
These results demonstrate that the effect of the metal loss on the radiation
efficiency is less than that by the dielectric loss. In addition, the both losses reduce
the efficiency by about 1 dB over the entire bandwidth from 10.4 to 24 GHz.
Figure 5.7a shows the normalized attenuation constant of the fields of the
SL-LWA for two cases of the lossy and lossless antennas. These results indicate
that presence of the loss cause negligible effect on the field decay rate and radiation
operation.

Fig. 5.7 Effect of the loss on 0.16

SL-LWA. The normalized
decay rate of the SL-LWA. 0.14 Lossy
Figure adopted and Lossless
reproduced with permission 0.12
from Ref. [19]
0.1

0.08

0.06
10 12 14 16 18 20 22 24
Frequency (GHz)
64 5 Spoof Surface Plasmon-Based Leaky-Wave Antenna (LWA)

5.1.4 Comparison with Other LW Structures Similar

to SL-LWA

The SL-LWA is composed of the SSP cells as well as the flaring ground. The
periodic cells generate the space harmonics and the radiating EM energy. The
flaring ground converts the CPW modes to the SSP modes and does not have any
significant effect on the radiation. Figure 5.3b shows the magnetic field distribution
of the SL-LWA at 17 GHz and demonstrates that the majority of the electromag-
netic energy is distributed near the meander strip, while the field distribution near
the flaring ground is negligible.
To further study the effect of the grooves on the antenna operation, a structure
similar to Fig. 5.3a without grooves is simulated, as shown in Fig. 5.8a.
Figure 5.8b shows the reflection results for varying geometrical parameters. These
results indicate that the flaring ground along with the central conducting strip
without the grooves could not generate the wideband radiation for any of the
parameter sets and therefore, this structure could not operate as a wideband antenna.

Fig. 5.8 A simple strip antenna without grooves. a The schematic. b The reflection results for
varying geometrical parameters. Figure adopted and reproduced with permission from Ref. [19]
5.1 Radiation Mechanism 65

The flaring ground of the SL-LWA may also resemble a tapered slot antenna
(TSA) [21, 22]. While this part of the SL-LWA does not radiate, the flared slot in
the TSAs radiates the electromagnetic energy and generates a fixed endfire beam.
This pattern is different from the frequency scanning radiating beam of the
SL-LWA. Their different radiation patterns indicate the major difference between
the radiation mechanisms of these two types of antennas.

5.2 Design Procedure and Optimization

To change the operation frequency range of the SL-LWA, the dispersion curve of
the radiating space harmonic is being tuned and shifted to the desired frequency
ranges. By increasing and decreasing the periodicity, the dispersion line of this
harmonic in Fig. 5.2b moves right and left, respectively. Figure 5.2b indicates that
the SL-LWA operating at the dominant mode can be approximated by a medium
with an effective permittivity of eeff = 2.2. Our simulation for SL-LWAs shows that
above the cut off frequency and within the leaky-wave frequency range, the
effective permittivity of the dominant mode only depends on the substrate per-
mittivity and its thickness and does not change significantly by changing the
geometrical parameters of the SSP cells. Therefore, according to Eq. (5.2), the
space harmonic follows the following dispersion relation:

pffiffiffiffiffiffi 2p
b
1 ¼ eeff k0
: ð5:3Þ
p

For the meander cell studied in Fig. 5.2b, p = 12 mm and the broadside radia-
tion (b−1 ’ 0) occurs at f = 17 GHz. According to Eq. (5.3), to set the broadside
radiation at 14.5 and 20 GHz, the periodicities of p = 14 mm and p = 10 mm are
required. Figure 5.9a, b studies the effects of the periodicity (p) on the operating
bandwidth and radiation angle, respectively. As p increases, the operating fre-
quency range shifts to lower edge while the bandwidth is kept as 80%.
Consequently, the radiation angle curve shifts to lower frequencies while the
scanning range remains constant. The broadside radiation frequency also corre-
sponds to f = 14.5 and 20 GHz, as were calculated from Eq. (5.3).
Figure 5.10a shows the maximum gain of the u-component in the y-z plane for
the three sets of parameters, where for d = 1.9 mm and w = 5.1 mm the most
consistent as well as highest gain over the frequency are achieved.
The other parameters to be tuned are the length of the converter (lm = Nm p) and
lt, as tagged in Fig. 5.3a. Figure 5.9b plots the maximum gain of the u-component
in the y-z plane for varying lt and Nm. Among all the parameters studied in this
66 5 Spoof Surface Plasmon-Based Leaky-Wave Antenna (LWA)

Fig. 5.9 Effect of the period

length on SL-LWA. a The
reflection and b the maximum
radiation angle with the
varying period lengths.
Figure adopted and
reproduced with permission
from Ref. [19]

figure, with the selection of Nm = 10 and lt = 80 mm chosen for prototyping the

most consistent gain over the frequency range is achieved.
Figure 5.9c shows the broadside realized gain over the frequency range. This
figure indicates that the operation band for less than 1-dB gain variation is
16.5–17.2 GHz or 4.2% of fractional bandwidth. Benefiting from a single layer and
planar configuration, the proposed leaky-wave antenna offers its broadband oper-
ation with a consistent broadside gain comparable with previous designs.

5.3 Experimental Verification

The designed single-layered leaky-wave antenna was fabricated, as shown in

Fig. 5.11a, and experimentally validated. The reflection from the antenna was
measured using a vector network analyzer and compared with the simulation results
in Fig. 5.11b and both the results agree well and demonstrate wide impedance
bandwidth from 10.4 to 24.5 GHz. The measured jS11 j is slightly higher than the
simulated values, probably caused by presence of the SMA connector that was not
included in the simulation.
5.3 Experimental Verification 67

Fig. 5.10 Realized gain of

SL-LWA. The maximum
realized gain in the y-z plane
with varying a d and w. b Nm
and lt. c The broadside
realized gain. Figure adopted
and reproduced with
permission from Ref. [19]

The radiation patterns were measured in an anechoic chamber. As the operating

frequency range of the power amplifier (PA) in the measurement set up was limited,
the measurement was performed for two frequency ranges of 10.5–18 GHz and
18–24.5 GHz separately with two different PAs. For the higher frequency range,
two low noise amplifiers were employed before the transmit antenna as well as after
the antenna under the test to compensate for higher loss at higher frequencies and
keep the measured results above the noise floor of the network analyzer.
68 5 Spoof Surface Plasmon-Based Leaky-Wave Antenna (LWA)

Fig. 5.11 Experimental evaluation of SL-LWA. a The fabricated prototype. b The measured
reflection spectrum compared with the simulation results. Figure adopted and reproduced with
permission from Ref. [19]

The normalized radiation patterns in the y-z planes are compared with the
simulation results in Fig. 5.12. More results are presented in Appendix B. The
cross-polarized h-component is at least 12 dB lower than the co-polarized u-
component at all the frequencies and all the angles. The experiment agrees the
simulation and demonstrates the frequency scanning radiation of the SL-LWA.
According to Fig. 5.2a, the momentum of the first space harmonic at 17 GHz is
zero, corresponding to the broadside radiation, as seen in Fig. 5.12b. Below
17 GHz, b−1 is negative corresponding to the backward radiation as seen in
Fig. 5.12a. As the frequency increases, b−1 increases and the main beam tilts to the
forward direction, as seen in Fig. 5.12c, d.
The gain in the maximum radiation direction at frequencies was measured and
compared with the simulation in Fig. 5.13, where a good agreement between the
results was observed. The gain variation across the entire bandwidth is limited to
2.5 dB. The constant gain is an important advantage of the proposed antenna over
the conventional LWAs with dropping gain at the broadside and/or forward radi-
ation. At 17 GHz, where the radiation is broadside, the gain is as high as the rest of
the bandwidth and no gain reduction is observed.
5.4 Conclusion 69

Fig. 5.12 Normalized radiation pattern of SL-LWA. Comparison between the measured and
simulated results for the co-component (u) as well as cross-component (h) in the y-z plane at
a 10.5, b 17, c 20 and d 24 GHz. Figure adopted and reproduced with permission from Ref. [19]

Fig. 5.13 Maximum realized 15

gain in y-z plane.
Figure adopted and
reproduced with permission
10
from Ref. [19]
Gmax (dBi)

Exp
Sim
0
10 15 20 25
Frequency (GHz)

5.4 Conclusion

A single-layered leaky-wave antenna based on meander SSP cells has been

proposed and experimentally verified. The antenna has achieved a wide operating
bandwidth of 80% with high and consistent total efficiency, gain, and wideband
broadside radiation over 4.2% of fractional bandwidth. The reflection bandwidth
70 5 Spoof Surface Plasmon-Based Leaky-Wave Antenna (LWA)

has been up to 95%. With the unique simple design free from ground planes and via
holes, the SL-LWAs open a vista for application of leaky-wave antennas in inte-
grated microwave circuits.

References

1. W.W. Hansen, Radiating electromagnetic wave guide. US2402622 A, 1946

2. D.R. Jackson, C. Caloz, T. Itoh, Leaky-wave antennas. Proc. IEEE 100(7), 2194–2206 (2012)
3. J.L. Gomez-Tornero, A.T. Martinez, D.C. Rebenaque, M. Gugliemi, A. Alvarez-Melcon,
Design of tapered leaky-wave antennas in hybrid waveguide-planar technology for millimeter
waveband applications. IEEE Trans. Antennas Propag. 53(8), 2563–2577 (2005)
4. F. Xu, K. Wu, X. Zhang, Periodic leaky-wave antenna for millimeter wave applications based
on substrate integrated waveguide. IEEE Trans. Antennas Propag. 58(2), 340–347 (2010)
5. H.V. Nguyen, A. Parsa, C. Caloz, Power-recycling feedback system for maximization of
leaky-wave antennas’ radiation efficiency. IEEE Trans. Microw. Theory Tech. 58(7),
1641–1650 (2010)
6. Y.-D. Lin, J.-W. Sheen, Mode distinction and radiation-efficiency analysis of planar
leaky-wave line source. IEEE Trans. Microw. Theory Tech. 45(10), 1672–1680 (1997)
7. Y. Dong, T. Itoh, Substrate integrated composite right-/left-handed leaky-wave structure for
polarization-flexible antenna application. IEEE Trans. Antennas Propag. 60(2), 760–771
(2012)
8. L. Liu, C. Caloz, T. Itoh, Dominant mode leaky-wave antenna with backfire-to-endfire
scanning capability. Electron. Lett. 38(23), 1414–1416 (2002)
9. A. Alu, F. Bilotti, N. Engheta, L. Vegni, Subwavelength planar leaky-wave components with
metamaterial bilayers. IEEE Trans. Antennas Propag. 55(3), 882–891 (2007)
10. S.K. Podilchak, A.P. Freundorfer, Y. Antar, Planar leaky-wave antenna designs offering
conical-sector beam scanning and broadside radiation using surface-wave launchers. IEEE
Antennas Wirel. Propag. Lett. 7, 155–158 (2008)
11. N. Nasimuddin, Z.N. Chen, X. Qing, Substrate integrated metamaterial-based leaky-wave
antenna with improved boresight radiation bandwidth. IEEE Trans. Antennas Propag. 61(7),
3451–3457 (2013)
12. H. Shi, X. Wei, Z. Zhao, X. Dong, Y. Lu, C. Du, A new surface wave antenna-based spoof
surface plasmon mechanism. Microw. Opt. Technol. Lett. 52(10), 2179–2183 (2010)
13. O. Quevedo-Teruel, Controlled radiation from dielectric slabs over spoof surface plasmon
waveguides. Prog. Electromagn. Res. 140, 169–179 (2013)
14. H. Yi, S.-W. Qu, X. Bai, Antenna array excited by spoof planar plasmonic waveguide. IEEE
Antennas Wirel. Propag. Lett. 13, 1227–1230 (2014)
15. S.-H. Kim et al., Subwavelength localization and toroidal dipole moment of spoof surface
plasmon polaritons. Phys. Rev. B 91(3) (2015)
16. B. Xu et al., Tunable band-notched coplanar waveguide based on localized spoof surface
plasmons. Opt. Lett. 40(20), 4683–4686 (2015)
17. A. Kianinejad, Z.N. Chen, L. Zhang, W. Liu, C.W. Qiu, Spoof plasmon-based slow-wave
excitation of dielectric resonator antennas. IEEE Trans. Antennas Propag. 64(6), 2094–2099
(2016)
18. A. Kianinejad, Z.N. Chen, C.-W. Qiu, in Spoof Surface Plasmon-Based Leaky Wave
Antennas. 2016 Asia-Pacific Microwave Conference (APMC), New Delhi (2016), pp. 1–3
19. A. Kianinejad, Z.N. Chen, C.W. Qiu, A single-layered spoof-plasmon-mode leaky wave
antenna with consistent gain. IEEE Trans. Antennas Propag. 65(2), 681–687 (2017)
References 71

20. A. Kianinejad, Z.N. Chen, C.W. Qiu, in Highly Radiative Symmetric Plasmonic Leaky Wave
Antenna. 2017 International Workshop on Antenna Technology: Small Antennas, Innovative
Structures, and Applications (iWAT) (2017), pp. 92–95
21. R. Janaswamy, D.H. Schaubert, Analysis of the tapered slot antenna. IEEE Trans. Antennas
Propag. 35(9), 1058–1065 (1987)
22. K.S. Yngvesson, T.L. Korzeniowski, Y.-S. Kim, E.L. Kollberg, J.F. Johansson, The tapered
slot antenna-a new integrated element for millimeter-wave applications. IEEE Trans. Microw.
Theory Tech. 37(2), 365–374 (1989)
Chapter 6
Future Work

Spoof surface plasmon modes with their high field confinement and simple and
planar configuration are the possible alternatives for the guiding waves in future
microwave circuits. In this research, we have established the basic requirements for
implementation of the spoof surface plasmon modes in circuit design. The
repeatable and reliable SSP mode circuit modeling has been proposed to enable the
design of the SSP-based circuit components. The designed slow wave transmission
lines with their advantage including low mutual effect, small profile, low loss and
flexible configuration have been developed as the ideal choices for compact
microwave integrated circuits.
In this thesis, we have demonstrated the implementation of the SSP modes in the
design of antennas and antenna feeds. With their unique groundless configuration,
the SSP-based transmission lines provide the real isolated boundary condition for
the antenna feeding and consequently, are capable of excitation of the otherwise
unavailable modes of the dielectric resonator antennas.
In addition, operating within their stop band frequency range, the SSP cells form
leaky-wave antennas with high radiation conversion and consistent performance. In
this research, we have experimentally evaluated the broadband SSP-based
single-layer LWA with extra advantage of flat gain characteristic for both maxi-
mum gain and broadside radiation.
In this chapter we propose two possible future research directions for the spoof
surface plasmon modes in the microwave engineering. Section 6.1 addresses the
potential implementation of the SSP modes in antenna engineering. Section 6.2
introduces the possible SSP application in microwave engineering and circuit
design.

© Springer Nature Singapore Pte Ltd. 2018 73

A. Kianinejad, Metamaterial Surface Plasmon-Based Transmission Lines
and Antennas, Springer Theses, https://doi.org/10.1007/978-981-10-8375-4_6
74 6 Future Work

6.1 Spoof Surface Plasmon Modes in Antenna Design

The SSP modes have shown the great potential to be implemented in antenna
design in order to eliminate the ground effect on the antenna at RF and microwave
bands. We have realized the plasmonic feeds for cylindrical DRAs to excite their
otherwise unavailable modes. The work can be extended to develop the feed for
different types of DRAs including spherical, semi-spherical, rectangular as well as
other shapes to study the operation modes of different DRAs in their isolated
conditions. It is expected that the DRAs resonating at their hidden modes would
exhibit better characteristics in terms of their resonance frequencies, sensitivity to
the size, radiation pattern etc.
In addition to the DRAs, the SSP modes can be implemented to feed metallic
patch antennas in order to realize their hidden aspects. With the isolated boundary
condition provided by the groundless plasmonic TLs, the patch antennas will be
excited without any additional antenna image in order to exhibit their unique
performance in the isolated condition. Eventually, the plasmonic feeding for dif-
ferent metallic patch antennas will be studied.
The wavelengths of the microwave plasmon modes are much lower than the
guided waves. This feature is advantageous in design of resonant antennas, which
require being at least quarter wavelength long. The SSP modes can be hybridized in
the design of resonant antennas to possibly overcome the current limitation on the
antenna size.

6.2 Spoof Surface Plasmon-Based Circuit Design

Our goal is to take advantage of the SSP modes in design of microwave circuits.
Three main parts of this research direction are as follows: (1) miniaturization of the
SSP cells, (2) design of the new SSP-based TLs and (3) the design of the SSP-based
microwave components and circuits.
The size limitation has always been a challenge in implementation of the SSP
structures in microwave circuits. The SSP modes are the interaction between the
electromagnetic waves at different frequencies with the metallic cavities in the
structures. The cavity sizes at lower microwave frequencies are quite large and
consequently, the miniaturization of the SSP cells while keeping all their features
has been one of the challenges in this area.
The designed low profile SSP cells should be applied to form transmission lines
by designing the required transitions from conventional TLs to SSP cells. The
transitions should be compact with high conversion efficiency and low reflection.
The proposed SSP-based transmission lines will be evaluated and their important
characteristics including ohmic loss, conversion loss, radiation loss and mutual
6.2 Spoof Surface Plasmon-Based Circuit Design 75

coupling will be compared with other TLs. The goal is to design efficient TLs with
as small profile as possible to be served as the building blocks of the future
SSP-based circuits.
To develop a full SSP-based circuit, it is necessary to design all the microwave
components using SSP cells. The goal is to design and verify high pass, low pass
and band pass filters using SSP cells. Next, SSP-based power dividers, couplers and
transformers will be designed. In addition to linear component design, it is required
to realize nonlinear circuit functions using SSP cells. Effective transitions from
different SSP-based TLs to amplifiers and mixers will be designed. The oscillators,
modulators will be developed using SSP modes and finally simple transmitter/
receiver systems based on SSP cells will be implemented. The final goal is to design
all the proposed SSP-based components and sub-systems for Monolithic
Microwave Integrated Circuit (MMIC) design purposes. By performing this
research work, the first SSP-based transmitter/receiver microwave circuit will be
delivered.
Appendix A
Excitation of TE-Dominant Higher Order
Modes in DRAs

Here, the results of the excitation of the TE-dominant higher order modes of a
cylindrical DRA are presented. EH11d, EH21d and EH31d modes are hybrid modes
with dominant magnetic fields in the axial direction. Besides the fundamental TE
mode, the proposed SW-TL feeding method in Chap. 4 is capable of exciting the
TE-dominant higher order modes. With the same DRA, the resonance frequencies
are 4.5, 5.3 and 6.3 GHz for EH11d, EH21d and EH31d modes, respectively, cal-
culated for an isolated DRA on the same substrate (Rogers 4003 1.5 mm) by the
Eigen mode solver of CST microwave studio. These values are displayed in the
parentheses in Fig. A.1a1–a3.
The feeding configuration in Chap. 4 as shown in Fig. 4.5a is applied to excite
the EH11d mode. However, this design is no longer suitable for exciting the EH21d
and EH31d modes due to its lower cutoff frequency. Therefore, the meander SSP
cells as seen in Fig. 4.3a are then modified and the length of the groove of the
meander slot, d is reduced from 9 to 6 mm for a higher cutoff frequency of 6.8 GHz.
Similar to the fundamental mode, the magnetic vector field distributions of these
higher order modes are in the axial direction of the DRA as shown in Fig. A.1a1–a3
and therefore the same feeding configuration is appropriate for these higher order
modes. Different from the centralized field distribution of TE01d mode, the higher
order modes are more spread on the sides, as it can be seen in Fig. A.1. By taking
advantage of the field distributions, the feeding distance x1 and x2 should be finely
tuned for an optimized excitation. For the EH11d mode, x1 and x2 are set to be 5.5
and −1.5 mm, respectively, while they are tuned to be 2.1 and −2 mm for both the
EH21d and EH31d modes. The resonance frequencies of the DRA fed by the SW-TL
at these modes are also mentioned in Fig. A.1a1–a3 (the values out of parentheses).
It should be noted that there is a slight frequency difference for each mode between
the isolated DRA and the case with a feeding configuration aside, as indicated by
the values in Fig. A.1a1–a3 for the TE-dominant higher order modes.

© Springer Nature Singapore Pte Ltd. 2018 77

A. Kianinejad, Metamaterial Surface Plasmon-Based Transmission Lines
and Antennas, Springer Theses, https://doi.org/10.1007/978-981-10-8375-4
78 Appendix A: Excitation of TE-Dominant Higher Order Modes in DRAs

Fig. A.1 TE-dominant higher order modes excited by SW-TL. The axial component of the
magnetic field for EH11d (a1), EH21d (a2) and EH31d (a3). The electric field vectors for EH11d (b1),
EH21d (b2) and EH31d (b3). The far field radiation pattern of the co-polarized component of the
electric fields at the x-y-plane for EH11d (c1), EH21d (c2) and EH31d (c3). The resonance frequency
of each mode for the isolated DRA (value in parentheses) as well as the DRA fed by SW-TL (value
out of parentheses) are indicated in (a1–a3) in unit of GHz

The axial component of magnetic (Hz) and the electric field (E) distributions of
the EH11d mode in the x-y plane are plotted in Fig. A.1a1 and b1, respectively.
According to Fig. A.1a1, the EH11d resonates in the azimuth direction while there is
no resonance in the axial direction, which acts like two parallel magnetic dipoles
with p radian phase difference. The electric field vectors are plotted in Fig. A.1b1
and aligned in the concentric circles around each dipole. Fig. A.1c1 depicts
the co-polarized radiation patterns in the x-y plane and presents two main lobes at
u = 60° and 240°, corresponding to the direction of the magnetic dipoles in
Fig. A.1a1. The field distributions of EH21d mode are also presented in Fig. A.1a2
and b2. In comparison with EH11d mode, the Hz component as shown in Fig. A.1a2
Appendix A: Excitation of TE-Dominant Higher Order Modes in DRAs 79

forms four resonating dipoles along with the electric field vectors, plotted in
Fig. A.1b2, swirling around them. The far-field co-polarized radiation patterns in the
x-y plane, plotted in Fig. A.1c2, also demonstrate four main lobes corresponding to
the four magnetic dipoles. As the mode order increases, for the EH31d, the number
of dipoles in the field distribution and lobes in the far-field patterns increases
accordingly as shown in Fig. A.1a3, b3 and c3. With the total efficiency of higher
than 90% and directional horizontal radiation pattern, the TE-dominant modes
could find application in cellular mobile communication systems.
Appendix B
Farfield Radiation Pattern Results
of SL-LWA

See Figs. B.1, B.2 and B.3.

Fig. B.1 Normalized radiation pattern of SL-LWA. Comparison between the measured and
simulated results for the co-component (u) as well as cross-component (h) in the y-z plane at
a 10.5, b 11, c 12, and d 13 GHz. Figure adopted and reproduced with permission from Ref. [1]

© Springer Nature Singapore Pte Ltd. 2018 81

A. Kianinejad, Metamaterial Surface Plasmon-Based Transmission Lines
and Antennas, Springer Theses, https://doi.org/10.1007/978-981-10-8375-4
82 Appendix B: Farfield Radiation Pattern Results of SL-LWA

Fig. B.2 Normalized radiation pattern of SL-LWA. Comparison between the measured and
simulated results for the co-component (u) as well as cross-component (h) in the y-z plane at a 14,
b 15, c 16, d 17, e 18, f 19, g 20 and h 21 GHz. Figure adopted and reproduced with permission
from Ref. [1]
Appendix B: Farfield Radiation Pattern Results of SL-LWA 83

Fig. B.3 Normalized radiation pattern of SL-LWA. Comparison between the measured and
simulated results for the co-component (u) as well as cross-component (h) in the y-z plane at a 22,
b 23, c 24, and d 24.5 GHz. Figure adopted and reproduced with permission from Ref. [1]

Reference

[1] A. Kianinejad, Z.N. Chen, C.W. Qiu, A single-layered spoof-plasmon-mode leaky wave
antenna with consistent gain, IEEE Trans. Antennas Propag. 65(2), 681–687 (2017).