Professional Documents
Culture Documents
Electromagnetic Sources and Electromagnetic Fields 1St Edition Gaobiao Xiao Online Ebook Texxtbook Full Chapter PDF
Electromagnetic Sources and Electromagnetic Fields 1St Edition Gaobiao Xiao Online Ebook Texxtbook Full Chapter PDF
https://ebookmeta.com/product/natural-electromagnetic-fields-in-
pure-and-applied-geophysics-springer-geophysics-kalyan-kumar-roy/
https://ebookmeta.com/product/electromagnetic-field-theory-1st-
edition-amit-sarin/
https://ebookmeta.com/product/electromagnetic-vortices-wave-
phenomena-and-engineering-applications-1st-edition/
https://ebookmeta.com/product/wave-fields-in-real-media-wave-
propagation-in-anisotropic-anelastic-porous-and-electromagnetic-
media-4th-edition-jose-m-carcione/
Advanced Electromagnetic Wave Propagation Methods 1st
Edition Gonzalez
https://ebookmeta.com/product/advanced-electromagnetic-wave-
propagation-methods-1st-edition-gonzalez/
https://ebookmeta.com/product/electromagnetic-waves-and-lasers-
second-edition-kimura-wayne-d/
https://ebookmeta.com/product/principles-and-technologies-for-
electromagnetic-energy-based-therapies-1st-edition/
https://ebookmeta.com/product/principles-and-techniques-of-
electromagnetic-compatibility-3rd-edition-christos-christopoulos/
https://ebookmeta.com/product/electromagnetic-metasurfaces-
theory-and-applications-wiley-ieee-1st-edition-karim-achouri/
Modern Antenna
Gaobiao Xiao
Electromagnetic
Sources and
Electromagnetic
Fields
Modern Antenna
Editors-in-Chief
Junping Geng, Department of Electronic Engineering, Shanghai Jiao Tong
University, Shanghai, China
Jiadong Xu, School of Electronics and Information, Northwestern Polytechnical
University, Xi’an, Shaanxi, China
Series Editors
Yijun Feng, School of Electronic Science and Engineering, Nanjing University,
Nanjing, Jiangsu, China
Xiaoxing Yin, School of Information Science and Engineering, Southeast
University, Nanjing, Jiangsu, China
Gaobiao Xiao, Electronic Engineering Department, Shanghai Jiao Tong University,
Shanghai, China
Anxue Zhang, Institute of Electromagnetic and Information Technology, Xi’an
Jiaotong University, Xi’an, Shaanxi, China
Zengrui Li, Communication University of China, Beijing, China
Kaixue Ma , School of Microelectronics, Tianjin University, Tianjin, China
Xiuping Li, School of Electronic Engineering, Beijing University of Posts and
Telecommunications, Beijing, China
Yanhui Liu, School of Electronic Science and Engineering, University of
Electronic Science and Technology of China, Chengdu, Sichuan, China
Shiwei Dong, National Key Laboratory of Science and Technology on Space
Microwave, China Academy of Space Technology (Xi’an), Xi’an, Shaanxi, China
Mingchun Tang, College of Microelectronics and Communication Engineering,
Chongqing University, Chongqing, China
Qi Wu, School of Electronics and Information Engineering, Beihang University,
Beijing, China
The modern antenna book series mainly covers the related antenna theories and
technologies proposed and studied in recent years to solve the bottleneck problems
faced by antennas, including binary coded antenna optimization method, artificial
surface plasmon antenna, complex mirror current equivalent principle and low profile
antenna, generalized pattern product principle and generalized antenna array, cross
dielectric transmission antenna, metamaterial antenna, as well as new antenna tech-
nology and development. This series not only presents the important progress of
modern antenna technology from different aspects, but also describes new theoret-
ical methods, which can be used in modern and future wireless communication, radar
detection, internet of things, wireless sensor networks and other systems.
The purpose of the modern antenna book series is to introduce new antenna
concepts, new antenna theories, new antenna technologies and methods in recent
years to antenna researchers and engineers for their study and reference. Each book
in this series is thematic. It gives a comprehensive overview of the research methods
and applications of a certain type of antenna, and specifically expounds the latest
research progress and design methods. As a collection, the series provides valuable
resources to a wide audience in academia, the engineering research community,
industry and anyone else who are looking to expand their knowledge of antenna
methods. In addition, modern antenna series is also open. More antenna researchers
are welcome to publish their new research results in this series.
Gaobiao Xiao
Electromagnetic Sources
and Electromagnetic Fields
Gaobiao Xiao
Department of Electronic Engineering
Shanghai Jiao Tong University
Shanghai, China
© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature
Singapore Pte Ltd. 2024
This work is subject to copyright. All rights are solely and exclusively licensed 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, expressed 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.
This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd.
The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721,
Singapore
In the first semester of 2018, I told one of my master’s students to calculate the Q
factors of a small bowtie antenna associated with its characteristic mode currents.
Several days later, he came back and asked me which formulation to use because there
are several of them in the literature. Furthermore, the results obtained using different
formulations may be different. I was a little bit surprised because I had thought this
was a solved problem. After I had checked those available formulations, I realized
that the key issue is about how to interpret the energies involved in electromagnetic
radiations and mutual couplings, which is a century old problem that absolutely needs
to be revisited. Personally, I think that for such a commonly encountered issue, if
there are many formulations that are not satisfactorily consistent with each other,
then we must have missed something in it.
On the other hand, I have spent more than fifteen years in the research of compu-
tational electromagnetics, analysis and synthesis of antenna arrays, and electromag-
netic inverse source problems. I have deep interest in the performance of the numer-
ical methods involved in these problems, especially their stability behaviors. In some
situations, we may still have no uncontroversial and insightful explanation to the root
of the instability that may possibly exist and ruin our numerical methods.
I believe that all these issues are certainly associated with each other. They are
related to the relationships of the electromagnetic sources and the electromagnetic
fields in the electromagnetic environment. I decided to revisit these issues and made
my mind to try my best to find a consistent solution to them before my retirement.
However, I fully understand that it is a very challenging task, so in this book I mainly
focus on investigating the issues in the vacuum. It is obvious that if the interactions
between the electromagnetic sources and the electromagnetic fields cannot be clearly
interpreted in free space, it will become more complicated when various effects of
media are taken into account.
Basically, the methods of calculating the Q factor of an antenna can be divided into
two categories. One is to evaluate the ratio of the stored energy of the antenna versus
the dissipated power. The result can be considered as a kind of unloaded Q factor
that mainly depends on the antenna itself. The other is to evaluate the corresponding
parameters at the exciting port of the antenna. It may be considered as a kind of loaded
v
vi Preface
Q factor as it usually contains some kind of external information, such as the port
structure and the matching network. The first kind of Q factor may be more general
because it can be applied for cases with no specified ports, such as small antennas
with characteristic mode currents, corner reflectors, or resonators. Moreover, if the
stored energy and dissipated power can be accurately evaluated, the Q factor of
the antenna at the exciting port can also be calculated using a projection from the
antenna structure onto the port. Therefore, estimations of the electromagnetic energy,
the radiation power, and the loss in the antenna structure are the basic concerns.
Conventionally, it is considered that the total electromagnetic energy of an antenna
consists of a radiative electromagnetic energy and a reactive electromagnetic energy.
Looking from the antenna, the radiative energy may be treated as a kind of energy
dissipation, while the reactive energy is related to the stored energy. However, it is
difficult to accurately separate and evaluate the two parts of the energy because there
are no explicit expressions for them.
In frequency domain analysis, it is assumed that the antenna begins radiating from
the infinite past, so the radiative electromagnetic energy fills the whole space, leading
to an unbounded radiative electromagnetic energy and an unbounded total electro-
magnetic energy. Pulse radiator in free space is a suitable example to use for deriving
the energy separation formulae because all the energies are finite and their perfor-
mances with respect to the source can be examined rigorously. By analogy with the
electromagnetic energy concepts in the classical charged particle theory, and using the
relationships derived from the Maxwell equations, the total electromagnetic energy
of a pulse radiator can be divided into three parts, each part is expressed with an inte-
gral with its integrand consisting of a source-potential or a field-potential term. The
nonzero periods of the three parts of the electromagnetic energy can be determined
from their explicit expressions. The first part of the energy disappears immediately
after the source has disappeared and is termed as the Coulomb-velocity energy. The
second part also disappears but not immediately. It remains nonzero for a short while
after the source has disappeared. It is called the macroscopic Schott energy in this
book because its behavior is similar to the Schott energy in the charged particle
theory. The third part of the energy becomes constant after the macroscopic Schott
energy has vanished. It is the radiative electromagnetic energy which keeps propa-
gating in free space till it encounters other sources. From their temporal evolution
property, we may reasonably take the Coulomb-velocity energy and the macroscopic
Schott energy as the reactive energy.
A closely related issue is the electromagnetic mutual coupling, which plays a very
important role in many systems. Efficient and accurate analysis of the electromagnetic
mutual couplings is still a challenging issue. The theory for electromagnetic energy
separation can be extended for handling multiple radiators. We can aggregate all
radiators together and treat them as a single larger radiator, similar to an antenna
array. The mutual electromagnetic coupling energies can be separated and defined
in the same way for a single radiator. The key issue involved in the electromagnetic
mutual couplings is the same as that in the electromagnetic radiation problems.
It is not the aim of this book to build up a bridge between the classical Maxwell
theory and the classical charged particle theory. I simply borrowed the concept of
Preface vii
the Schott energy from the classic charged particle theory and proposed the concept
of the macroscopic Schott energy by noticing that they have similar performances.
The two Schott energies are both full-time derivatives. Making use of the Liénard-
Wiechert potentials and some approximations, the Schott energy of a moving charge
in the charged particle theory can be derived from the macroscopic Schott energy.
Although the energy separation is obtained with time-varying pulse radiators, the
formulae for time harmonic waves are also available. The results in time domain
and frequency domain are completely in consistent because they are respectively
derived from the time domain Maxwell equations and the frequency domain Maxwell
equations directly. The electromagnetic fields in frequency domain and time domain
can be converted with Fourier transform.
The theory is verified with the Hertzian dipole both in frequency domain and in
time domain. The reactive electric energy and the reactive magnetic energy are found
to be exactly in agreement respectively with the energy stored in the capacitor and
the inductor in the equivalent circuit model of the Hertzian dipole proposed by Chu.
The real radiative power and the pseudo-electromagnetic power associated with the
Hertzian dipole can be explicitly separated.
For a source in free space, its electromagnetic fields can be calculated with inte-
grations over the source region using some powerful numerical tools, so are the
corresponding electromagnetic energies and powers. However, in many practical
applications such as antenna synthesis and current source reconstructions, it may
require more insightful information and interpretation of the electromagnetic radi-
ation process than those the numerical solutions can provide. Analytical and semi-
analytical methods are still very important because they can help to illustrate more
clearly the characteristic relationship between the electromagnetic fields and the
electromagnetic sources and provide bases for predicting the stability and accuracy
of the numerical methods.
Generally, semi-analytical solutions for radiators can be obtained with spherical
harmonic mode expansion method. Conventionally, the three sets of vector basis
functions, namely the two solenoidal basis function sets Mnm , Nnm , and the longitu-
dinal basis function set Lnm , are adopted for spherical harmonic mode expansions, as
in Stratton’s book and in Collin’s book. However, it should not be taken for granted
that it is optimal to use them for expanding general vectors that may neither satisfy
the Helmholtz equations nor the vector wave equations. In this book, we choose three
sets of vector basis functions different from the conventional ones. The electromag-
netic fields and their sources are expanded with the same set of spherical vector basis
functions in a similar procedure. Explicit expressions for the electromagnetic fields,
potentials, energies, and the related Green’s functions are derived for the spherical
modes. In particular, the time domain Green’s function for the spherical harmonics
is derived for the first time, with which the radiation process of the spherical modes
can be illustrated more clearly.
An equivalent nonuniform transmission line (NTL) model is developed for intu-
itively characterizing the total radiation process. The introduction of the cutoff radius
and the cutoff mode degree provides a simple reference for determining the number
of degrees of freedom (NDF) of the fields associated with the sources in a bounded
viii Preface
region. In particular, we use the concept of the effective NDF of a source and that
of the electromagnetic field radiated by the source in antenna synthesis and inverse
problems. We emphasize that the two effective NDFs are dependent on each other
and should be matched. When we discuss the effective NDF of a field, we may have
to associate it with a source region and specify the size and shape of the source
region, as well as the distance between the field and the source region. On the other
hand, when we discuss the effective NDF of a source, we may have to associate it
with the electromagnetic fields outside the source region and also have to specify the
distance between them. The connection between the two effective NDFs provides
us a useful criterion for determining some of the main parameters of the radiation
pattern that can be realized with sources in a certain region; or for determining to
what extent the sources can be reconstructed from the electromagnetic fields outside
the source region. Based on the theory, an efficient hybrid method for synthesizing
antenna arrays with complex footprints is proposed and demonstrated with numerical
examples. Effective algorithms are also developed for reconstructing the radiating
part of the current sources.
This book focuses on the fundamental principles and semi-analytical methods
involved in the non-relativistic electromagnetic radiation problems. It is intended for
researchers, engineers, and graduate students who are interested in the topics of the
energy transfer in electromagnetic radiation, synthesis and measurement of antenna
arrays, and applications of electromagnetic inverse source problems. Basically, we
have to resort to some kind of numerical methods for analyzing the electromagnetic
radiation of a general radiator. Various numerical techniques have been developed
by the computational electromagnetics society, like the methods based on integral
equations or differential equations. We will discuss the numerical techniques for
evaluating the fields and energies in electromagnetic radiations in our next book,
especially the formulations based on surface integral equations in frequency domain
and in time domain.
Acknowledgments The author would like to acknowledge the assistance from his Ph.D. students
Mengxia Hu, Ting Zang, Guomin Liu, Xiaocheng Wang, and Rui Liu in preparing the manuscript
of the book.
Contents
ix
x Contents
xiii
Abbreviations
AR Axial ratio
CBF Characteristic basis function
CMBF Characteristic mode basis function
CP Circular polarization
EFIE Electric field integral equation
FDFD Finite difference frequency domain
FDTD Finite difference time domain
GA Genetic algorithm
LP Linear polarization
MB-RWG Multi-branch RWG
MoM Method of moment
MOT Marching-on in time
NDF Number of degrees of freedom
NTL Nonuniform transmission line
PEC Perfectly electrically conducting
PSO Particle swarm optimization
RWG Rao–Wilton–Glisson
SA Simulated annealing
SBF Synthetic basis function
SIE Surface integral equation
SNR Signal to noise ratio
SVD Singular value decomposition
SWG Schaubert–Wilton–Glisson
TE mode Transverse electric mode
TEM mode Transverse electromagnetic mode
TM mode Transverse magnetic mode
VIE Volume integral equation
VSWR Voltage standing wave ratio
WGN White Gaussian noise
xv
Chapter 1
Basic Concepts in Electromagnetic
Radiation
Abstract This chapter will not discuss the incompleteness of the classical macro-
scopic electromagnetic theory. Readers who are interested in this topic may find
comprehensive discussions in the literature, such as the books by Barrett T W
and Rohrlich F. On the contrary, we confine our discussions within the frame of
the Maxwell’s theory and try to give an intuitive interpretation to the relationship
between the electromagnetic fields and the electromagnetic sources in free space. The
electromagnetic fields are solutions to the Maxwell equations. They are generally
characterized by some conserved dynamic quantities, of which the most commonly
known ones are the electromagnetic energy, the electromagnetic power, the electro-
magnetic linear momentum, and the electromagnetic angular momentum. We will
give a brief introduction to the fundamental notations, concepts, and principles in
the classical electromagnetic theory. Detailed and systematic descriptions about the
classical electromagnetic theory can be found in classical books.
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024 1
G. Xiao, Electromagnetic Sources and Electromagnetic Fields, Modern Antenna,
https://doi.org/10.1007/978-981-99-9449-6_1
2 1 Basic Concepts in Electromagnetic Radiation
For this purpose, we will give a brief introduction to the fundamental nota-
tions, concepts, and principles in the classical electromagnetic theory in this chapter.
Detailed and systematic descriptions about the classical electromagnetic theory can
be found in classical books [3–6].
Electromagnetic fields are vector fields. Generally, their properties can be charac-
terized through their divergences and curls. For an electric field E(r, t) [V/m] at
position r in vacuum and at time t, its divergence and curl are respectively expressed
by [5–7]
∂
∇ × E(r, t) = − B(r, t) (1.2)
∂t
where ε0 [F/m] is the permittivity in vacuum, and B(r, t) [Wb/m2 ] is the magnetic
flux density. It can be regarded that the charge density ρ(r, t) [C/m3 ] is the source to
induce the divergence of the electric field multiplied by ε0 , while the time derivative
of B(r, t) is the source to the curl of the electric field. We will simply refer the former
as a divergence source and the latter as a curl source.
For a magnetic field H(r, t) [A/m] in vacuum, its divergence and curl are expressed
by
μ0 ∇ · H(r, t) = 0 (1.3)
∂D(r, t)
∇ × H(r, t) = + J(r, t) (1.4)
∂t
where μ0 [H/m] is the permeability in vacuum, and D(r, t) [C/m2 ] is the electric
2
/ the current density J(r, t) [A/m ] and the
flux density. The total current, including
displacement current density ∂D(r, t) ∂t, is the curl source of the magnetic field.
There is no divergence source for the magnetic field as there are no magnetic charges
in nature.
The flux densities and the field intensities are subject to the constitutive
relationships. In vacuum, the relationships are
A curl source is conservative because its divergence is always zero. From Eq. 1.4
we can show that
( )
∂D(r, t)
∇· + J(r, t) = 0. (1.7)
∂t
Substituting Eqs. 1.1 and 1.5 into Eq. 1.7, we obtain the current continuity law,
∂ρ(r, t)
∇ · J(r, t) + = 0. (1.8)
∂t
Equations 1.7 and 1.8 state that, although the total current is always conservative,
the current density J(r, t) alone is generally not conservative unless the charge density
is static.
Equations 1.1–1.4 are the Maxwell equations for the classical electromagnetic
theory in vacuum. The Maxwell equations describe the relationships between the
electromagnetic fields and their sources, i.e., the field intensities E(r, t) and H(r, t),
the flux densities D(r, t) and B(r, t), and the sources ρ(r, t) and J(r, t). Substi-
tuting the constitutive relationships into the Maxwell equations, we get the equa-
tions consisting of only the electric field E(r, t) and the magnetic field H(r, t). Two
equations are for their divergences and two equations for their curls. However, as
E(r, t) and H(r, t) are coupled in the equations, it is not an easy task to obtain their
solutions even in free space.
A scalar electric potential φ(r, t) and a vector magnetic potential A(r, t) are
introduced to simplify the solving of the Maxwell equations. The electromagnetic
fields can be expressed with their potentials as
1
H(r, t) = ∇ × A(r, t) (1.9)
μ0
∂A(r, t)
E(r, t) = −∇φ(r, t) − . (1.10)
∂t
The potentials are not unique. Introducing the Gauge transformation
⎧
⎨ φ (r, t) → φ(r, t) − ∂ /(r, t)
1
∂t , (1.11)
⎩
A1 (r, t) → A(r, t) + ∇/(r, t)
we can check that the electromagnetic fields are Gauge-invariant under Eq. 1.11.
However, it is important to note that the potentials can be uniquely determined in the
whole space if we put the reference zero points of the potentials at the infinity, and
make them subject to the Lorentz Gauge,
4 1 Basic Concepts in Electromagnetic Radiation
∂
∇ · A(r, t) + μ0 ε0 φ(r, t) = 0. (1.12)
∂t
In this way, the potentials can be handled like the fields. We will show in later
chapters that the electromagnetic energies can be separated explicitly with terms
including the potentials.
Under the Lorentz Gauge, φ(r, t) and A(r, t) satisfy
∂ 2 φ(r, t) ρ(r, t)
∇ 2 φ(r, t) − μ0 ε0 =− (1.13)
∂t 2 ε0
∂ 2 A(r, t)
∇ 2 A(r, t) − μ0 ε0 = −μ0 J(r, t). (1.14)
∂t 2
Consider a simple linear system. The sources ρ(r, t) and J(r, t) are assumed to
be some known distributions in a bounded region Vs in the free space. The potentials
are derived to be
{ {
1 ρ(r1 , t1 ) 1
φ(r, t) = dr1 = G t (r, r1 ; t) ∗ ρ(r1 , t)dr1 (1.15)
ε0 4π R ε0
Vs Vs
{ {
J(r1 , t1 )
A(r, t) = μ0 dr1 = μ0 G t (r, r1 ; t) ∗ J(r1 , t)dr1 (1.16)
4π R
Vs Vs
/
where “*” means temporal convolution. t1 = t − R c is the retarded time, and
R = |r − r1 | is the distance between the field point r and the source point r1 . The
retarded time accounts for the propagation delay of the fields from the source point
r1 to the observation point r. In this book, if there is no risk of causing confusion,
we generally use r to indicate the position for both of the fields and the sources.
In places where we have to distinguish them, then we will use r for the fields and
potentials, and r1 for the sources.
The potentials expressed in Eqs. 1.15 and 1.16 are called retarded potentials: they
are related to sources some time earlier than the present time t. In three-dimensional
space, the Green’s function in time domain has the expression of [8]
( )
1 R
G (r, r1 ; t) =
t
δ t− (1.17)
4π R c
where c is the light velocity in vacuum. As shown in Eqs. 1.15 and 1.16, the potentials
include the contributions from sources for t1 ≤ t.
It is assumed that all the source distributions are known in the expressions. We have
to note that this is a strong condition that is hard to satisfy in general situations even
if we restrict the problem in vacuum. The following issues may be more commonly
encountered. The first one is to find the electromagnetic fields from the sources with
an initial source distribution at time t = t0 . The second one is that only a very small
1.1 Electromagnetic Fields in Free Space 5
part of the sources is well controlled and can be treated as a known excitation. In
these cases, the source distributions at time t are affected by the fields generated
by the sources at time t ' with t0 ≤ t ' < t. It is required to determine the source
distributions at t ' in the period of t0 ≤ t ' < t before we can calculate the fields at t.
In this book, we simply assume that the sources ρ(r, t) and J(r, t) are the resultant
distributions that have already taken into account of these interactions and need not
to consider the change of distribution caused by the surrounding electromagnetic
fields once again.
In some situations, the Hertz vectors can be used for solving the Maxwell equa-
tions. Although the electric Hertz vector |e and the magnetic Hertz vector |m can
be used together, the Maxwell equation can be solved with the single electric Hertz
vector |e . In free space, the general potentials are related to |e by [8]
1
φ(r, t) = − ∇ · |e (r, t)
ε0
∂
A(r, t) = μ0 |e (r, t).
∂t
In the meantime, the current density and the charge density are related to an
electric polarization vector P with
ρ(r, t) = −∇ · P(r, t)
∂
J(r, t) = P(r, t).
∂t
Substituting the relationships into Eq. 1.14, we obtain the equation for the electric
Hertz vector
∂2
∇ 2 |e (r, t) − μ0 ε0 |e (r, t) = −P(r, t). (1.18)
∂t 2
By solving the electric Hertz vector through Eq. 1.18, we obtain the scalar potential
φ(r, t), the vector potential A(r, t), and the electromagnetic fields generated by the
source P(r, t).
In the static situation, all derivatives with respect to time t in the formulae are zeros,
including those in the Maxwell equations, the current continuity law, the relationships
between the fields and the potentials, and the Lorentz Gauge. Therefore, the electric
fields only have divergence sources, while the magnetic fields only have curl sources.
The static electric fields and the static magnetic fields are decoupled.
6 1 Basic Concepts in Electromagnetic Radiation
Since the sources are time independent, the convolution in Eqs. 1.15 and 1.16
simply change to an integration of the time domain Green’s function with respect to
the time t, resulting the Green’s function for the static fields,
{∞
1
G 0 (r, r1 ) = G t (r, r1 ; t)dt = . (1.19)
4π R
−∞
In a spherical
( / ) coordinate system, the Green’s function can be expanded with power
series of r1 r [9]
1 E ( r 1 )n
∞
1 1
= / = Pn (cos α) (1.20)
4π R 4π r 2 + r12 − 2rr1 cos α 4πr n=0 r
where cos α = r̂ · r̂1 , and α is the angle between the unit vector r̂ and r̂1 . Pn (cos α) is
the n-th order Legendre polynomial. Therefore, the scalar potential can be expressed
as the sum of the contributions from the multipoles of the charge source,
∞
E ∞ {
1 E 1
φ(r) = φn (r) = ρ(r1 )r1n Pn (cos α)dr1 . (1.21)
n=0
4π ε0 n=0
r n+1
Vs
Basically, the potential φn (r) decreases in the order of r −(n+1) . The first three
lowest Legendre polynomials and scalar potentials are respectively
P0 (x) = 1
P1 (x) = x
P2 (x) = 0.5(3x 2 − 1).
{
1 Q0
φ0 (r) = ρ(r1 )dr1 =
4π ε0 r 4π ε0 r
Vs
{
1 1 ( )
φ1 (r) = ρ(r1 )r1 dr1 · r̂ = p · r̂
4π ε0 r 2 4π ε0 r 2
Vs
{ ( )
1 3 1 1
φ2 (r) = (r1 · r̂) − r1 · r1 ρ(r1 )dr1 =
2
Q quad .
4π ε0 r 3 2 2 4π ε0 r 3
Vs
The total charge Q 0 is the monopole of the source, p is the electric dipole moment
of the source, and Q quad is the electric quadrupole moment of the source at r̂ direction.
The potentials from the higher-order poles decrease faster with the increase of the
distance to the source.
1.1 Electromagnetic Fields in Free Space 7
The static electric field can be derived from the scalar potential as
{
1
E(r) = −∇φ(r) = − ∇G 0 (r, r1 )ρ(r1 )dr1 . (1.22)
ε0
Vs
In the region far away from the sources, the asymptotical behavior of the electric
field is found to be
{
1 Q0
E(r) ≈ ρ(r1 )dr1 r̂ ≈ r̂. (1.23)
4π ε0 r 2 4π ε0 r 2
Vs
An electric dipole is shown in Fig. 1.1a, where two charges are symmetrically
placed along the z-axis with a small distance of l. When the charges are static and
q1 = −q2 = q, they compose a static electric dipole. The first three components of
the scalar potential of the dipole are found to be
{
1
φ0 (r) = ρ(r1 )dr1 = 0
4π ε0 r
Vs
1 ( ) 1 ( )
φ1 (r) = ql ẑ · r̂ = p · r̂
4π ε0 r 2 4π ε0 r 2
φ2 (r) = 0.
According to Eq. 1.21, the amplitude of the scalar potential of the n-th multipole
is in proportional to l n . Therefore, the electric potential mainly consists of the contri-
bution of the electric dipole moment p = ql ẑ. The static electric field in the spherical
coordinate system is then,
ql | |
E = −∇φ(r) ≈ −∇φ1 (r) = 2 cos θ r̂ + sin θ θ̂ . (1.24)
4π ε0 r 3
The electric flux line is shown in Fig. 1.1b. At every point on an electric flux line,
the electric field is in the tangential direction of the flux line.
The vector potential of a static current source can be handled in a similar way.
Substituting Eq. 1.20 into Eq. 1.16 gives the expansion for the vector potential,
⎡ ⎤
∞
E ∞ {
μ0 E ⎣ 1
A(r, t) = An (r, t) = n+1
J(r1 )r1n Pn (cos α)dr1 ⎦. (1.25)
n=0
4π n=0
r
Vs
The potential component An (r) also decays in the order of r −(n+1) . The first two
terms are
{
μ0
A0 (r, t) = J(r1 )dr1
4πr
Vs
{
μ0 ( )
A1 (r, t) = 2
J(r1 ) r1 · r̂ dr1
4πr
Vs
We usually do not directly use Eq. 1.25 to define the multipoles of the current
source like in handling the charge sources. Generally, multipole expansion for the
vector potential is carried out with the spherical harmonic expansion of the Green’s
function in spherical coordinate system. We will provide a detailed discussion on
this issue in Chap. 4. However, Eq. 1.25 can also provide some useful information
about the behavior of a static current source.
A small loop current is shown in Fig. 1.2a. For a static loop current I , it can be
derived that
{ {2π
μ0 μ0 bI
A0 (r, t) = J(r1 )dr1 = ϕˆ1 dϕ1 = 0
4πr 4πr
Vs 0
{
μ0 ( ) μ0 I π b2 μ0
A1 (r, t) = J(r1 ) r1 · r̂ dr1 = ẑ × r̂ = m × r̂
4πr 2 4πr 2 4πr 2
Vs
{
μ0
A2 (r, t) = J(r1 )r12 P2 (cos α)dr1 = 0.
4πr 3
Vs
The main contribution comes from the second term, so it can be considered as
a magnetic dipole with magnetic dipole moment of m = m ẑ = I π b2 ẑ. The static
magnetic flux density is the curl of the vector potential,
μ0 m | |
B = ∇ × A(r) ≈ ∇ × A1 (r) = 3
2 cos θ r̂ + sin θ θ̂ . (1.26)
4πr
1.1 Electromagnetic Fields in Free Space 9
Fig. 1.2 Small loop current. a Coordinate system. b Magnetic field flux lines. c Magnetic dipole
and its flux lines
The magnetic flux lines of the loop current are illustrated in Fig. 1.2b. For compar-
ison, the magnetic flux lines of a magnetic dipole consisting of two virtual magnetic
charges are illustrated in Fig. 1.2c. Obviously, in the interior region of the loop, the
magnetic field distribution is different from that of the magnetic dipole. However, in
the exterior region of the loop, the magnetic field distribution is very close to that of a
magnetic dipole. Therefore, we may simply use a very small current loop to emulate
a magnetic dipole.
For time harmonic fields with time dependence of exp( j ωt), the Green’s function in
frequency domain is the Fourier Transform of the Green’s function in time domain,
{∞ ( )
1 R − jωt 1 − jk0 R
G(r, r1 ) = δ t− e dt = e (1.27)
4π R c 4π R
−∞
√
where k0 = ω/ c = ω μ0 ε0 is the wavenumber in free space. The potentials are
then expressed by
{
1
φ(r, k0 ) = G(r, r1 )ρ(r1 )dr1 (1.28)
ε0
Vs
{
A(r, k0 ) = μ0 G(r, r1 )J(r1 )dr1 . (1.29)
Vs
In the near region that satisfies k0 R << 1, we may take k0 R ≈ 0 in the phase
terms of G(r, r1 ) in Eqs. 1.28 and 1.29. The potentials are obtained to be
10 1 Basic Concepts in Electromagnetic Radiation
{
1 ρ(r1 )
φ(r) ≈ dr1 (1.30)
ε0 4π R
Vs
{
J(r1 )
A(r) ≈ μ0 dr1 . (1.31)
4π R
Vs
They are the same as the potentials for the static sources. The electric field and
the magnetic field can be approximately expressed by
{
1 ρ(r1 )
E(r) ≈ â R dr1 (1.32)
ε0 4π R 2
Vs
{
J(r1 ) × â R
H(r) ≈ dr1 (1.33)
4π R 2
Vs
where â R is the unit vector of (r − r1 ). As expected, the near fields behave similarly
to the static fields. ( )
At region far ) | we have G(r, r1 ) ∼ g0 exp
| away from( the sources, / jk0 r̂ · r1 and
∇G(r, r1 ) ∼ − jk0 g0 exp jk0 r̂ · r1 r̂ , where g0 = exp(− jk0 r ) 4πr . Then we
can derive
{
( )
E(r) ≈ − jωμ0 g0 I − r̂r̂ · e jk0 r̂·r1 J(r1 )dr1 (1.34)
Vs
1
H(r) ≈ r̂ × E(r) (1.35)
η0
/ /
where η0 = μ0 ε0 [/] is the intrinsic impedance in vacuum. Equations 1.34 and
1.35 show that the far fields are nonuniform spherical transverse electromagnetic
(TEM) fields. The fields have concentric spherical equi-phase surfaces, but their
amplitudes are not uniform on the equi-phase surfaces.
The directivity of the fields is determined by the transverse vector current moment,
which is defined by
{
( )
P J (k) = I − r̂r̂ · e jk·r1 J(r1 )dr1 (1.36)
Vs
Performing curl operation on Eq. 1.37 and substituting Eqs. 1.2 and 1.3 into it
yields
Therefore, the magnetic field in vacuum is determined by the curl of the current
density and is independent on the divergence of the current density. Borrowing the
solution for the vector potential A(r) in vacuum, we can write that
{
H(r) = G(r, r1 )∇1 × J(r1 )dr1 (1.39)
Vs
where ∇1 means performing the operator with respect to r1 . The electric field outside
the source region can be obtained using Eq. 1.37, which is
1
E(r) = ∇ × H(r). (1.40)
j ωε0
When the two charges in Fig. 1.1a vary periodically in time with the angular frequency
ω, they compose the Hertzian dipole with a moment of ql cos ωt, as shown in Fig. 1.3.
The corresponding electric polarization vector is.
P(r, t) = ql cos ωtδ(r)ẑ.
The electric current J(r, t) can be expressed in terms of the polarization vector
P(r, t) by
∂P(r, t)
J(r, t) = = −ωql sin ωtδ(r)ẑ (1.41)
∂t
from which the vector potential and the scalar potential of the Hertzian dipole are
derived to be
ωμ0 ql ( )
A(r, t) = − sin(ωt − k0 r ) cos θ r̂ − sin θ θ̂ (1.42)
4πr
| |
ω2 μ0 ql 1 1
ϕ(r, t) = cos θ 2 2 cos(ωt − k0 r ) − sin(ωt − k0 r ) . (1.43)
4π k0 r k0 r
k02 ql
E f ar (r, t) ≈ − cos(ωt − k0 r ) sin θ θ̂ (1.46)
4π ε0 r
1 k02 ql 1
H f ar (r, t) ≈ − cos(ωt − k0 r ) sin θ ϕ̂ = r̂ × E f ar (r, t). (1.47)
η0 4π ε0 r η0
The far fields are spherical TEM fields. The magnetic field and the electric field
are in-phase.
The two varying charges can be equivalently treated as a very short uniform linear
current with length l locating symmetrically at the origin. The phasor of the current
can be found from Eq. 1.41 as
where I = − jωq. We can obtain the fields by directly solving the vector potential
from the short current, which only has z-component,
μ0 I le− jk0 r
A(r, t) = ẑ. (1.49)
4πr
1.1 Electromagnetic Fields in Free Space 13
The components of the electric field and the magnetic field in the spherical
coordinate system are respectively
⎧ ( )
⎪ jk0 η0 I l 1 1
⎪
⎪
⎪ Eθ = sin θ 1 + − 2 2 e− jk0 r
⎪
⎨ 4πr jk0 r k0 r
( )
η0 I l 1 (1.50)
⎪
⎪ Er = cos θ 1 + e− jk0 r
⎪
⎪ 2πr 2 jk 0r
⎪
⎩
Eϕ = 0
⎧
⎪
⎨ Hr = Hθ = 0
( )
jk0 I l 1 . (1.51)
⎪
⎩ Hφ = sin θ 1 + e− jk0 r
4πr jk0 r
According to their asymptotic behaviors, the fields of the Hertzian dipole can be
divided into three parts: a far field, a middle field, and a near field. The far field is
the part that decays most slowly,
⎧
⎪ jk0 η0 I l
⎪ f ar
⎨ Eθ = sin θ e− jk0 r
4πr
. (1.52)
⎪
⎪ jk0 I l 1 f ar
⎩ Hϕf ar = sin θ e− jk0 r = Eθ
4πr η0
For a Hertzian dipole, its near field mainly consists of electric field. It forms a kind
of electric field sheath in the region near the source.
The middle field is the part that decays slower than the near field but faster than
the far field,
⎧ η0 I l
⎪
⎪ E θmid = sin θ e− jk0 r
⎪
⎪ 4πr 2
⎪
⎨
η0 I l
Ermid = cos θ e− jk0 r . (1.54)
⎪
⎪ 2πr 2
⎪
⎪
⎪
⎩ H mid = I l sin θ e− jk0 r
ϕ
4πr 2
14 1 Basic Concepts in Electromagnetic Radiation
It is clear that the amplitudes of the far field, the middle field, and the near field
decay respectively with the speed of (k0 r )−1 , (k0 r )−2 , and (k0 r )−3 , as shown in
Fig. 1.4. The near field components dominate in the near region, the far field compo-
nents dominate in the far region. In the middle region, the distance to the source is
of the order of a wavelength. No component dominates in this region.
Hertzian dipole is an infinitesimal dipole. It can be used as a fundamental unit
block of a wire current or other complicated current sources. We can decompose the
source of a radiator into many Hertzian dipoles, and then calculate the fields of the
radiator approximately by adding up the fields of all those Hertzian dipoles.
The fields of the magnetic dipole are dual to that of the Hertzian dipole. The near
fields are mainly magnetic fields instead of electric fields.
The densities of the electric energy and the magnetic energy in vacuum are
respectively defined as
1 1
we (r, t) = D(r, t) · E(r, t) = ε0 E 2 (r, t) (1.55)
2 2
1 1
wm (r, t) = B(r, t) · H(r, t) = μ0 H 2 (r, t) (1.56)
2 2
The energy density is a local variable. Recalling that the dimension of E(r, t) is
[V/m], and that of ε0 is [F/m]. Intuitively, we may regard E(r, t) as the voltage per
unit length at the point r, and ε0 as the capacitance of the cube with unit edge length
at the point r or simply as the capacitance density at r in vacuum. The dimension of
we (r, t) is deduced to be [(F V2 )/m3 ]. It can be considered that we (r, t) is the electric
energy stored in the capacitor with capacitance of ε0 . Similarly, μ0 can be considered
1.2 Electromagnetic Energy and Power 15
as the inductance density at the point r in vacuum, and wm (r, t) is the magnetic energy
stored in the inductor with inductance of μ0 . The total electric energy and the total
magnetic energy are obtained by integrating their energy densities over the whole
three-dimensional space,
{ {
1
We (t) = we (r, t)dr = D(r, t) · E(r, t)dr (1.57)
2
V∞ V∞
{ {
1
Wm (t) = wm (r, t)dr = B(r, t) · H(r, t)dr. (1.58)
2
V∞ V∞
1 1 1 ∂A 1
D · E = ρφ − D · − ∇ · (φD) (1.59)
2 2 2 ∂t 2
where the vector identity ∇ · (φD) = ∇φ · D + φ∇ · D is used. Substituting Eq. 1.59
into Eq. 1.57, the total electric energy can be expressed alternatively by [10]
{ { {
1 1 ∂A 1
We (t) = ρφdr − D· dr− φD · r̂d S (1.60)
2 2 ∂t 2
Vs V∞ S∞
where V∞ represents the whole three-dimensional space and S∞ represents the spher-
ical surface with (radius) approaching
( / ) infinitely large. According
( / ) to the asymptotical
behavior of lim D · r̂ ∼ O 1 r 2 and lim φ ∼ O 1 r , the surface integral at
r →∞ r →∞
the right-hand side of Eq. 1.60 approaches zero at S∞ . Therefore,
{ { ( )
1 1 ∂A
We (t) = ρφdr + − D· dr. (1.61)
2 2 ∂t
Vs V∞
The total electric energy is composed of two parts. The first one is stored in the
source region Vs . We denote it as the Coulomb energy,
{
1
Wρ (t) = ρ(r, t)φ(r, t)dr. (1.62)
2
Vs
For a charge distribution ρ(r, t) with finite amplitude in a bounded region, the
Coulomb energy Wρ (t) is always finite.
The second part of We (t) is a volume integral in terms of the field and the vector
potential, as shown in the second term of the right-hand side of Eq. 1.61. It spreads
over the whole three-dimensional space. /
For static fields, the second part is zero because ∂A ∂t = 0. We have
16 1 Basic Concepts in Electromagnetic Radiation
{ {
1 1
We = D(r) · E(r)dr = ρ(r)φ(r)dr = Wρ . (1.63)
2 2
V∞ Vs
Similarly, substituting Eqs. 1.4 and 1.9 into B · H and making use of the vector
identity of ∇ · (H × A) = A · ∇ × H − H · ∇ × A, we obtain
1 1 1 ∂D 1
B·H = J·A+ · A − ∇ · (H × A). (1.64)
2 2 2 ∂t 2
Substituting Eq. 1.64 into Eq. 1.58, the total magnetic energy can be expressed
alternatively by [10–12]
{ { {
1 1 ∂D 1
Wm (t) = J · Adr + · Adr − H × A · r̂d S. (1.65)
2 2 ∂t 2
Vs V∞ S∞
From the asymptotic behaviors of the ( /time) varying magnetic field and vector
potential, we have lim (H × A) · r̂ ∼ O 1 r 2 . The surface integral in Eq. 1.65 at
r →∞
S∞ is usually a bounded but nonzero value. This means that there is a part of energy
leaking to or coming back from the infinity.
The total magnetic energy consists of three parts. Denote the first part at the
right-hand side of Eq. 1.65 as
{
1
W J (t) = J(r, t) · A(r, t)dr (1.66)
2
Vs
which is stored in the source region. If we assume that the charge density ρ(r, t)
moves with velocity v(r, t) at (r, t), then we have J(r, t) = v(r, t)ρ(r, t). By analogy
with the energies in the charged particle theory, we call W J (t) the velocity energy. For
a current source distribution with finite amplitude in a bounded region, the velocity
energy is also bounded.
As can be seen from Eq. 1.65, the second part of the magnetic energy occupies
the whole space, and the third part is related to the energy flowing to or coming back
from the infinity. Their properties will be discussed in Chap. 6.
For
/ the static fields, the second part of the magnetic energy is zero because
∂D ∂t = 0. The third part is also zero because lim (H × A) · r̂ decays in the
( / ) ( / ) r →∞
order of 1 r 3 instead of 1 r 2 . We can check that
{ {
1 1
Wm = B(r) · H(r)dr = J(r) · A(r)dr = W J . (1.67)
2 2
V∞ V∞
Equations 1.63 and 1.67 show that the electric energy and the magnetic energy
associated with a static source distribution in a bounded region are finite.
1.2 Electromagnetic Energy and Power 17
Now consider static point sources. Enclose the electric dipole in Fig. 1.1a with a
small sphere with radius of a. The center of the sphere is at the origin of the coordinate
system. The total static electric energy outside the small sphere is calculated to be
(ql)2
We = . (1.68)
12π ε0 a 3
The electric energy outside the small region is bounded. However, we can check
by letting a → 0 in Eq. 1.68 that the electric energy in the interior region of the
sphere is not bounded. This infinitely large electric energy is nonphysical. It is caused
by the point charge model.
The magnetic field of the small static loop current shown in Fig. 1.2a can be
evaluated similarly. It can also be checked that the magnetic energy in the region
outside the small sphere containing the loop is bounded, and the total magnetic energy
inside the small sphere is not bounded. However, in the case of a loop current, the
infinite magnetic energy is not caused by the small size of the loop but caused by
the line current model. The static magnetic energy of any static line current is not
bounded.
In frequency domain, the time averaged electric energy density and the time averaged
magnetic energy density in vacuum are obtained by replacing the fields and sources
with their corresponding phasors,
{ }
1 ∗ 1
we,av (r) = Re E(r) · D (r) = ε0 E 2 (r) (1.69)
4 4
{ }
1 ∗ 1
wm,av (r) = Re B(r) · H (r) = μ0 H 2 (r). (1.70)
4 4
The time-averaged Coulomb energy and the time-averaged velocity energy are
respectively their volume integrations,
⎧ ⎫
⎨{ 1 ⎬
Wρ,av = Re ρ ∗ (r)φ(r)dr (1.71)
⎩ 4 ⎭
Vs
⎧ ⎫
⎨{ 1 ⎬
W J,av = Re J∗ (r) · A(r)dr . (1.72)
⎩ 4 ⎭
Vs
18 1 Basic Concepts in Electromagnetic Radiation
We can derive the time-averaged electric energy and the time-averaged magnetic
energy using Eqs. 1.61 and 1.65. A better way is to derive them directly from the
Maxwell equations in frequency domain so that we can correctly judge which phasors
should use their conjugates. The results are
⎧ ⎫
⎨{ ( jω ) ⎬
We,av = Wρ,av + Re − D∗ · A dr (1.73)
⎩ 4 ⎭
V∞
⎧ ⎫
⎨{ ( j ω ) ⎬
Wm,av = W J,av + Re − D∗ · A dr . (1.74)
⎩ 4 ⎭
V∞
{{ {
The fact that Re S∞ jω(H∗ × A) · r̂d S = 0 has been used to derive Eq. 1.74.
The rigorous proof can be found in [13].
Time harmonic sources are considered to exist over the entire time axis, i.e., in the
time interval of (−∞ < t < ∞). Their radiation fields expand in the whole space
and are not bounded. It is straightforward to check that lim (D∗ · A) ∼ O(1/r 2 ).
r →∞
Therefore, the volume integrals in the right-hand side of Eqs. 1.73 and 1.74 are
usually not bounded.
The Poynting Theorem describes the relationship between the work rate done by the
source, the total electromagnetic energy in region Va that contains the source, and
the total electromagnetic power flux crossing the boundary Sa of the region,
{ { ( ) {
∂ 1 1
− J · Edr = D · E + B · H dr + E × H · n̂d S. (1.75)
∂t 2 2
Vs Va Sa
where dFe = (ρE + ρv × B)dr is the local Lorentz force. Note that we have inserted
a null term v · (ρv × B)dr in Eq. 1.76. By analogy with Newton mechanics, Eq. 1.76
can be interpreted as the work rate done by the source Jdr. Its integration, the
left-hand side of Eq. 1.75, represents the total work rate done by the source.
The Poynting vector is defined as S(r, t) = E(r, t) × H(r, t). Denote the total
work done by the source as Wext (t). Equation 1.75 can be rewritten as
1.3 Electromagnetic Momentum 19
{
∂ ∂
Wext (t) = [We (t) + Wm (t)] + S(r, t) · n̂d S. (1.77)
∂t ∂t
Sa
Based on energy conservation principle, the second term in the right-hand side of
Eq. 1.77 must be the total electromagnetic power crossing the surface Sa to the outside
region, so the Poynting vector can be naturally interpreted as the electromagnetic
power flux density.
For time harmonic fields, Poynting Theorem can be written in terms of the phasors,
{ { ( ) {
1 ∗ 1 ∗ 1 ∗ 1
− J · Edr =2 j ω B · H − E · D dr+ E × H∗ · n̂d S. (1.78)
2 4 4 2
Vs Va Sa
The imaginary part of the relationship in the Poynting Theorem Eq. 1.78 is about
the energy conservation, which states that the imaginary part of the work rate done
by the source equals the increasing rate of the total electromagnetic energy.
When a particle with charge q moves in the electromagnetic field E(r, t) and B(r, t)
with velocity v(r, t), the electromagnetic force on the charge is given by
which is usually called the Lorentz force. For a charge density ρ(r, t) and a current
density J(r, t) in region Vs , the total electromagnetic force from the electromagnetic
field is
{
Fem (t) = (ρE + J × B)dr (1.81)
Vs
where the current density can be expressed with the charge density and the local
velocity v(r, t) by J(r, t) = v(r, t)ρ(r, t). The electromagnetic force density is
defined as
If the electromagnetic fields are generated by the source pair ρ(r, t) and J(r, t)
themselves, the electromagnetic force is a kind of self-force. It is the reaction force
acting on the sources by the electromagnetic fields they have generated. Assume that
there is an external force to balance this reaction force, that is,
d
fext (r, t) = pmec (t) = −fem (r, t)
dt
where pmec (t) is the mechanical momentum density corresponding to the external
force. Making use of the Maxwell equations and Eq. 1.82, we obtain
d
fext = pmec (t) = −ρE − J × B
dt
∂
= (D × B) + [E∇ · D − D × (∇ × E) + H∇ · B − B × (∇ × H)]
∂t | |
∂ 1
= (D × B) + ∇ · (D · E + B · H)I − DE − BH . (1.83)
∂t 2
1
T= (D · E + B · H)I − DE − BH. (1.85)
2
Therefore, Eq. 1.83 can be rewritten as
1.3 Electromagnetic Momentum 21
∂
fext (r, t) = −fem (r, t) = −ρE − J × B = pem + ∇ · T. (1.86)
∂t
Integrating Eq. 1.86 over the region Va ⊃ Vs enclosed by the surface Sa gives
{ {
∂
Fext (t) = − (ρE + J × B)dr = Pem + T · n̂d S (1.87)
∂t
Vs Sa
∂ ∂
Fext = Pext = Pem (1.89)
∂t ∂t
which states that the external force converts the mechanical momentum to the same
amount of electromagnetic momentum.
The total Lorentz force is related to the total electromagnetic momentum by
{
d
Fem (t) = fem (t)dr = − Pem (t). (1.90)
dt
Vs
Besides the electromagnetic energy and the electromagnetic momentum, the elec-
tromagnetic field also carries an electromagnetic angular momentum. The canonical
angular momentum is defined by
{ {
Lem = r × pem dr = ε0 r × (E × B)dr. (1.91)
Va Va
The angular momentum is conserved. It can be divided into two parts, an orbital
angular momentum (OAM) Lam and a spin angular momentum (SAM) Sam ,
In free space, Lam and Sam are conserved. At the angular frequency ω, with the
democracy formalism (dual-symmetry formalism) [14], the SAM density can be
explicitly expressed by
1 { }
sam = Im ε0 E∗ × E + μ0 H∗ × H . (1.93)
4ω
The OAM density is defined by the cross product of the radius vector r and the
orbital linear momentum density po , i.e.,
lam = r × po (1.94)
where po is defined by
1 { }
po = Im ε0 E∗ · (∇)E + μ0 H∗ · (∇)H . (1.95)
4ω
In addition, there is a spin linear momentum with density of
1
ps = ∇ × sam (1.96)
2
In the definitions, we have used the following notation for the sake of brevity,
A · (∇)B = A x ∇ Bx + A y ∇ B y + A z ∇ Bz .
The total linear momentum is the sum of the orbital linear momentum and the
spin linear momentum,
pem = po + ps . (1.97)
The electromagnetic fields are solutions to the Maxwell equations. They are
generally characterized by some conserved dynamic quantities, of which the most
commonly known ones are the electromagnetic energy, the electromagnetic power,
the electromagnetic linear momentum, and the electromagnetic angular momentum
[14, 15]. Because the total linear momentum is linearly proportional to the Poynting
vector, it is natural to separate the Poynting vector into two parts, an orbital energy
flux density that is related to the orbital linear momentum and a spin energy flux
density that is related to the spin linear momentum. However, the separation of
electromagnetic energy flux is not unique, as will be discussed in Chap. 6.
References 23
References
For a resonant circuit, the port voltage or the port current reaches its maximum at
the resonance frequency f 0 and will decrease when the frequency shifts away from
f 0 . A 3dB bandwidth / f is introduced to represent the frequency range that the port
voltage or current drops to 0.707 of the peak value [1–4].
Q factor of a resonant circuit is often defined in two ways. The first definition is
the ratio of the central resonance frequency f 0 to the 3dB frequency bandwidth,
f0 ω0
Q F BW = = (2.1)
/f /ω
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024 25
G. Xiao, Electromagnetic Sources and Electromagnetic Fields, Modern Antenna,
https://doi.org/10.1007/978-981-99-9449-6_2
26 2 Q Factor of a Radiator
The second definition for Q factor is based on the energies in the resonator at the
resonance frequency,
2π E sto
QE = (2.2)
E dis
where E sto is the energy stored in the resonator and E dis is the energy dissipated per
cycle.
A Lorentz type resonator may be modelled with a RLC series circuit or a RLC
shunt circuit, as shown in Fig. 2.1. For sinusoidal resonance with voltage/source
√
U (ω), it can be checked that the Q factor at the resonance frequency ω0 = 1 LC
can be expressed by
ω0 L 1
Q F BW = Q E = = . (2.3)
R ω0 C R
For the Lorentz type resonator, the current I (ω) flowing in the resistor may be
regarded as the response to the voltage source U (ω) imposing on the input port. The
system response function is the input admittance,
|I (ω)| ωC
|h(ω)| = =/ ( )2 (2.4)
|U (ω)|
R 2 ω2 C 2 + ω2 LC − 1
√ /√
It is easy to check that ωU ω L = ω0 = 1 LC. The fractional bandwidth is
found to be
ωU − ω L R 1 1
F BW = = = = . (2.5)
ω0 ω0 L Q F BW QE
2.1 Q Factor for Circuits 27
In this case, the Q factor is a clear index for the fractional bandwidth.
Substituting Eq. 2.5 into Eq. 2.4 gives the normalized system function as
1
|H (ω)| = / ( )2 . (2.6)
ω ω0
1+ ω0
− ω
Q 2F BW
The typical relationship between the bandwidth and the Q factor for a Lorentz
type resonator is plotted in Fig. 2.2a. The line shapes are approximately symmetrical
with respect to the center frequency.
The Fano resonance was first discovered by Fano U in 1961 [3]. It has become
a distinctive characteristic of interacting quantum system, in which the Fano reso-
nance is usually caused by the interference between a broad background state and
a narrow discrete state. Unlike the Lorentz resonance, the Fano resonance basically
has an asymmetric line shape with a narrow bandwidth. Fano resonances also exist
in classical optics, as well as in metamaterials, plasmonic nanostructures, and many
classical oscillator systems.
Light scattering by high permittivity homogeneous spheres or cylinders can also
exhibit Fano resonances. Mie theory can be used to theoretically analyze the Fano
resonance that occurs in the scattering of homogeneous spheres or cylinders, in which
the Mie coefficients an can be expressed generically as a Fano function [5]:
| / |2
1 (ω − ω0 ) | + q
|an | =
2
| / | (2.7)
1 + q 2 1 + (ω − ω0 ) | 2
where q is the Fano parameter. It affects the line shape of the Fano resonance. | is a
positive number that represents the resonance width. It can be much small compared
with the resonance frequency, so the Q factor of the Fano resonator can become very
high. The Fano resonance expression of the Mie scattering coefficient |an |2 in Eq. 2.7
28 2 Q Factor of a Radiator
is accurate near the resonance frequency. ω0 is the resonance frequency related to the
radius of the sphere. The two half power frequencies ωU and ω L can be determined
according to Eq. 2.7, and the Q factor can be calculated with Eq. 2.1. The typical
normalized line shapes for the Fano resonances are shown in Fig. 2.2b.
Q E can be extended for energy storage devices other than resonators and oscilla-
tors. In a circuit, a capacitor is a storage device for electric energy, and an inductor
is a storage device for magnetic energy. We can define a Q factor for each of them.
Instead of using Q E directly, a popular definition for energy storage devices is slightly
modified as
ωE sto,max (ω)
Q E M (ω) = (2.8)
Pdis (ω)
where E sto,max (ω) is the maximum storage energy in the capacitor or the inductor
evaluated at ω, and Pdis (ω) is the time averaged dissipated power per cycle.
A capacitor circuit model is shown in Fig. 2.3. If the loss mainly comes from the
leakage current in the capacitor, it can be modeled with a capacitance C in parallel
with a conductance G C , as shown in Fig. 2.3a. When the capacitor is connected with
a sinusoidal voltage source U (ω), the energy stored in the capacitor is 0.25CU 2 , and
the power dissipated in one period is 0.5G C U 2 . We can explicitly define its Q factor
as
ωC
QC = . (2.9)
2G C
If the loss mainly comes from the leads, then it can be modeled with a capacitance
C in series with a resistance RC , as shown in Fig. 2.3b. When the capacitor is
/ with a sinusoidal current source I (ω), the energy stored in the capacitor is
connected
0.25I 2 ω2 C, and the power dissipated in one period is 0.5RC I 2 . We can explicitly
define its Q factor as
1
QC = . (2.10)
2ωC RC
The Q factor for an inductor can be defined in the same way. When the inductor is
modelled as an inductance L in parallel with a resistance R L , as shown in Fig. 2.4a,
its Q factor can be defined as
1
QL = . (2.11)
2ωLG L
ωL
QL = . (2.12)
2R L
1
Z in (ω) = Rin (ω) + j X in (ω) ≈ R0 + j ωL 0 + (2.14)
j ωC0
where the
/ √parameters R0 , L 0 , and C0 are not dependent on the frequency. Note that
ω0 = 1 L 0 C0 . Taking the derivative of Z in (ω) with respect to ω at ω0 , we have
By expanding Z in (ω) in a power series of ω and equating Eq. 2.14 to its first order
of ω at ω0 , we obtain
| ' '
|
R0 = Rin (ω0 ), L 0 = 0.5 X in (ω0 ) − j Rin (ω0 ) .
30 2 Q Factor of a Radiator
/ |
We have used the notation of X in'
(ω0 ) = d X in (ω) dω|ω=ω0 in the expression.
The Q factor can then be defined in a unified formula as
| ' |
ω0 L 0 ω0 | Rin '
(ω0 ) + j X in (ω0 )|
QZ = = . (2.16)
R0 2Rin (ω0 )
We can verify that the five definitions for the Q factors in Eq. 2.3 and Eqs. 2.9–
2.12 can all be derived from Eq. 2.16. In most situations, we may take R ' (ω0 ) ≈ 0.
If X in (ω0 ) ≈ 0, the circuit may be resonant at ω0 , or it is simply a resistive device.
Otherwise, it is a capacitive device for X in (ω0 ) < 0 and an inductive device for
X in (ω0 ) > 0.
Z a (ω) = Ra + j X a . (2.17)
The input resistance Ra accounts for the effect of the electromagnetic radiation
and the loss in the antenna structure. The fields of the short dipole are similar to that
of the Hertzian dipole. As has discussed in the previous section, it may have very
large electric field near the antenna, and hence may store a large amount of electric
energy in the antenna region. Intuitively, we can expect that the input impedance is
capacitive, and must have a very large negative reactance X a .
We have two natural approaches to evaluate the Q factor [7–9]. The first one is to
use the input impedance Eq. 2.17 and obtain the Q factor at ω0 with
Fig. 2.5 Dipole antenna. a Tuning at the feeding port. b Equivalent circuit
2.2 Q Factor for Antennas 31
| |
ω0 | Ra' (ω0 ) + j X a' (ω0 )|
QZ = . (2.18)
2Ra (ω0 )
Since it requires the derivative of the input reactance with respect to ω, it is not
suitable for evaluating the Q factor at a single frequency. In practical applications,
some kind of tuning components may be used to cancel the input reactance [10]. In
the calculation of the Q factor, we have to take into account the effect of the tuning
devices, which may be a series inductor or a parallel capacitor.
The second choice is to evaluate the Q factor in terms of the stored electromagnetic
energy and the radiation power. In some situations, we do not care about the excitation
conditions. Some radiators simply have no excitation ports. In these cases, it is
more convenient and reasonable to calculate the Q factor in terms of their stored
energy versus their radiation power. For time harmonic radiators, the time-averaged
Coulomb-velocity energy is the stored energy. If we take the radiation power as the
loss of the radiator, we obtain
( )
ω Wρ + W J
Q unload = . (2.19)
Prad
1 1 1 1
= + + (2.20)
Q load Q unload Qc Qd
where
( )
ω Wρ + W J
Qc =
Pc
( )
ω Wρ + W J
Qd = .
Pd
For a general antenna, we can still use the procedure to obtain its Q factor. However,
the role of the Q factor is not as significant as that in an electrically small antenna.
On the one hand, the relationship between the Q factor and the bandwidth is not so
simple and straightforward. On the other hand, the definition of the bandwidth of an
antenna itself is diverse.
Further discussions on the Q factors of antennas will be given in Chap. 6, together
with several numerical examples.
Another random document with
no related content on Scribd:
Sable noticed it. He had his own irritation, growing more focused each
day. To begin with, they would lose the Laidlaw case and it was all, Sable
thought, due to that false start which Gage had made. He had rather
decisively taken the matter out of Gage’s hands towards the end but the
thing had been lost already—or he preferred to think so. Sable could bear to
lose cases but not a case which involved so much money. It frightened off
the right sort of clients.
When Gage was a cub lawyer, arguing cases with flaring energy in the
local courts, Sable had picked him out as a bright young man. He had kept
his eye on him and his progress, with sheer admiration for the practical
genius with which he picked up important clients and gained and held their
confidence. He edged in on politics after a little—and in Mr. Sable’s own
party. Then King and Sable had made a proposition to young Flandon—that
he join them, bringing his clients, of course, and coming in, not as an
ordinary apprentice lawyer but as the colleague of Mr. Sable. It was an
amazing offer to be made to so young a man. Gage accepted it. Two years
later King, rather elderly now and ready to retire, went to Congress and the
firm name changed from King and Sable to Sable and Flandon. Flandon
made good. He made important alliances for Mr. Sable, he played the
political end for Mr. King, he made money for himself.
These things were not to be passed lightly over and Mr. Sable had them
all docketed somewhere in his mind. He was fond of Gage too, in his own
restrained way. But Sable was fifty-eight. He had seen many a brilliant start
end in disaster, many a man with ability fail. He knew most of the signs of
failure in men. He knew further exactly what steps Gage should take to
achieve eminence. They were broad and fair before him. Instead it was
increasingly clear that Gage was not keeping his mind on his work—that he
was letting his nerves get the better of his judgment. For some reason or
other he was making a fool of himself. When a man made a fool of himself,
there were, in Sable’s experience, one of three things back of it—a woman,
liquor or speculation. He was watching Gage to see which of these things it
might be in his case.
All this talk which Flandon was always getting off about women now—
thought the senior partner—that was camouflage. He felt fairly convinced
that Gage must be playing the fool with some woman. Irregular and
disappointing, with a lovely, fine looking, distinguished wife like Mrs.
Flandon. Rotten streak in Flandon probably. Sable chose the woman
solution rather definitely. Gage drank when he could get it of course. And
he nearly always had a supply on hand. But he used his head about it pretty
well. It didn’t seem like liquor trouble. As for speculation—surely he
wouldn’t play the fool there. There was plenty of money coming to Gage,
and he always could get more.
It must be a woman. Probably Flandon was trying to keep it from his
wife and that was what was on his nerves. Some little—Sable characterized
Gage’s visionary lady impolitely. He thought on, his mind lighting, for no
apparent reason, on Freda. And there it stopped. Queer, Flandon’s bringing
that girl into the office. Bright enough but no experience. Unlike him too,
considering his usual impatience with inexpert assistance. He wondered—
So while the Brownley girls gossiped in ugly, furtive, rather lustful
conversations and Ted Smillie told his little discovery on occasion as being
an instance of what those “smooth touch-me-not girls were usually up to”—
while Mr. Sable, his mouth tight in repression and his eyes keen, watched
and noted Freda. Freda went on her serene way. She was serene and she was
happy. At times her happiness seemed to shut her completely off from every
one—even in her thoughts from her father. She never tired of exploring her
memory for the sound of Gregory’s voice, the touch of his hands, the
mystery of love. More and more as the days went by she hugged her secret
to herself. She could not have shared a vestige of it. Its exquisite privacy
was part of its quality. She had the vaguest notions of what might be
waiting her as Gregory’s wife. Certainly she might have a baby—normally
that probably would happen to her in the next nine months. Gregory was
poor. They’d have to work. And there might be hard things. She thought
once or twice that it might be an ugly sort of proposition if she did not have
the particular feeling she did for Gregory. But there it was. It wasn’t a
matter of the mind—nor of physiology either. She didn’t believe it was
physiology which made her deliciously faint and weak as she read
Gregory’s strange letters—letters so frequent, so irregular, so curiously
timed and written—on the back of a menu, on a scrap of envelope, on a
dozen sheets of hotel paper. Each message, beating, alive, forcing its
entrance. This was the love that according to Margaret was the undoing of
her sex. She knew she would go anywhere Gregory wanted her to go, to be
with him. That she knew her life with him would have its independence
completely in so far as her own love allowed it, did not make it less clear to
her that even if the independence had been less, if she had found him a man
of convention she would none the less—but would she?
She was immensely interested in possibly having a baby, and anxious to
know about it. She wanted to tell Gregory. She wrote him letters in which
she spent the deepest of her thought. She said things in her letters which
would have astounded her if she had read them over. But she never did read
them after she had written them. It would have seemed almost like cheating
to read them as if for criticism.
But to-day she had not had a letter from Gregory and several unpleasant
things broke in upon her absorbed happiness. She missed his letter which
she usually went home at noon to get. In the afternoon as she sat at her desk
working and trying to feel that she could fill up the time until she went
home that night to see if there was a letter, Bob and Allison Brownley came
in with another young girl. They were as resplendent as usual and Freda
judged that they were collecting for some fashionable charity, from their
intrusion with pencils and notebooks. She had seen women invade these
offices almost every day for some such reason but it was her first encounter
with Bob since that night on which she had left her house. To her horror she
found herself flushing, and hoping that Barbara would not notice her and
that thought enraged her so that she raised her head and looked full at the
girls coming towards Mr. Flandon’s office, evidently referred to her.
She expected some embarrassment in Barbara and instead met a glance
of insolence and surprise. She looked at Allie but Allie looked away and
left it to Barbara.
“Can I take your message?” asked Freda with a little hauteur.
“We prefer to see Mr. Flandon personally,” said Barbara, and went by. It
was in Freda’s mind to stop them but Barbara was swift. Freda could hear
Mr. Flandon’s voice greeting her and judged it was too late to do anything.
She sat down at her desk frowningly and was further surprised when the
door opened very shortly and the girls went out. They, especially Barbara,
had heads unpleasantly held, angrily tilted. The buzzer sounded for Freda.
She found her employer sitting at his desk looking as angry as his
departing guests.
“Sit down a moment, Miss Thorstad, will you?”
She did as he told her. It was evident that he had something important
and difficult to say. She watched him. He looked nervous, tired too, she
thought.
“That young lady made some unpleasant remarks about you and I asked
her to leave the office,” he said.
“Oh—I’m sorry,” answered Freda. “She’s been abominable, Mr.
Flandon. But it’s too bad you should have been involved.”
“Don’t let that bother you,” said Gage grimly; “it’s of no consequence.
But I wonder if you ought to let her be quite so broadcast in her remarks. It
could be stopped.”
“It doesn’t matter—truly it doesn’t. Let her say what she pleases. If any
one wants to know the truth of the matter I always can tell it, you see.”
“Would you think it infernal impudence if I asked you what the truth
was?”
She hesitated and then laughed a little.
“You know the funny thing is that I had almost completely forgotten the
whole business. It seemed important at the time but it was really trivial.
Except for the fact that it opened up other things to me. Of course I’ll tell
you, if you want to know.”
She did tell him in outline, stressing the fact of the misunderstanding all
around, on the whole, dealing rather gently with Barbara, now that anger
had gone out of her.
“I had made rather a fool of myself you see,” she finished.
He looked at her as if waiting for her to go on.
“That’s all.”
“I see. She—well—.” He let that pass. “Now ordinarily it is easy to say
that gossip and slander don’t make any difference to a high minded person.
I think you are high minded. I do feel however that she has made this
incident a basis for a kind of slander that is dangerous. Her accusations
against you are, from what I hear, absolutely libelous. It wouldn’t take ten
minutes to shut her mouth if I could talk to her. But I want you to fully
refute her specific attacks.”
“I know. I imagine she might say almost anything.”
“Well, then, you have never stayed at the Roadside Inn, have you?”
To his amazement the face of the girl in front of him changed. She had
been calm and half smiling. Now astonishment, consciousness, and
something like panic showed in her eyes, her suddenly taut body.
“Does she say that? How did she know?” There was a little moan of
dismay in Freda’s answer.
Gage’s face grew stern. He sat looking at the girl across from him,
whose eyes were closed as if in pain.
“To lay her hands on that,” said Freda, under her breath.
“I don’t understand you,” said Gage rather curtly.
She lifted her face.
“It hurts to have any one know that—but for her to know it most of all.”
“Such things are usually public knowledge sooner or later, my dear
young lady. Clandestine—”
“Don’t say that,” cried Freda, her voice rising, “don’t use that word.”
And then as if some gate had been opened her words poured out. “Can’t
you understand something being too beautiful to be anything except secret?
It was something I couldn’t have let even those who love me know about.
And to have her ugly devastating hands on it! It soils it. I feel her finger
marks all over me. It was mine and she’s stolen it.”
Her head went down on her arms on the desk in front of her. Gage
watched her with curiosity, embarrassment and pity. To his mind this love
affair was a shady business but she didn’t see it so. That was evident. Her
abandonment touched a chord of sympathy in him. He knew how she was
being rent by pain.
“My dear girl,” he told her, more gently, “I’m afraid you’ve been very
unwise.”
“No—not unwise.” She raised her head and smiled unsteadily. “I’ve
been quite wise. It’s just bad luck—that’s all.”
“Could you tell me about it?”
She got up and walked to the window, evidently trying to compose
herself. “It’s nothing that matters to any one but me. And I suppose you are
thinking things that, even if they don’t matter, had better be set straight. For
perhaps you think they matter. There’s nothing that I’ve done that I
shouldn’t have done. I was there at that Inn—with—with my husband. It
was just that we wanted—he even more than I at first until I learned why—
to keep that little bit of life for ourselves. We didn’t want people to know—
we didn’t want to share with any one except each other. I know you won’t
understand but there’s nothing to condemn except that we had our own way
of—caring.”
“But I do understand,” answered Gage, “and I’m glad you told me. I do
most entirely understand. Because I’ve felt that way. Is your husband here?”
“He’s gone,” said Freda, “but he’ll come back. You see I married
Gregory Macmillan.”
A memory of that slim, gaunt young poet came to Gage. Yes, this was
how he would do it. And how perfect they were—how beautiful it all was.
“Mr. Flandon,” said Freda, “let them say what they please about me. Let
them talk—they don’t know about Gregory—or do they?”
“No—they don’t.”
“Then don’t tell them, will you? Don’t tell any one. I don’t care what
they say now if they don’t lay their hands on the truth. I can’t bear to have
the truth in their mouths. Please—what do I care what any one says? I don’t
know any one. I never see those people. He will be back and we’ll go away
and they’ll forget me.”
She was very beautiful as she pleaded with him, eyes fresh from their
tears, her face full of resolution.
“It’s all right, my dear,” said Gage, “no one shall know. You are right.
Keep your memories to yourself. What they say doesn’t matter.”
He was standing by her at the window now, looking down at her with a
tenderness that was unmistakable. It was unfortunate that at that moment
Mr. Sable entered without notice.
II
There was an argument that night. Sable had forced it. He had said that
Gage had to “cut it out in his own office.”
Gage had asked him what he meant by cutting it out and his partner said
that he definitely meant getting that girl out of the office at once.
“And my advice to you is to keep away from her after she is out.”
The upshot was that Gage had refused. He had simply said that there was
no reason why he should turn out a useful employee simply because any
one disliked her or thought evilly of her. Miss Thorstad was extremely
useful to him and there was nothing further to say. At which Sable had
snorted in disdain.
But, seeing Gage’s stubbornness he had possibly guessed at what might
be the depth of it and grown milder.
“It’s a difficult business for me, Gage,” he said, “but I’ve got to go
through with it. She must leave the office. We can’t afford scandal.”
“Suppose I won’t discharge her?”
“I’m not supposing any such nonsense. You aren’t going to act that way
unless you’re crazy.”
“But if I did?”
Sable looked at him.
“It means a smash probably. Don’t let’s talk foolishness. You know
you’ve got too much tied up in this business to let it go. You couldn’t afford
to say you smashed up your business for a woman. That’s not the way
things are done. I can’t insist on your giving up the girl but I can ask you to
remove the scandal from an office in which not alone your name is
involved.”
“Such rotten minds,” thought Gage, almost without anger. He was
feeling curiously clear and light and deft. He had felt that way ever since he
had found how Freda felt. Something had been strengthened in his own
philosophy by her simple refusal to share her secret with every one. She put
other things higher than the opinion of gossip. So must he.
They let the thing ride for a few days. Gage thought of nothing else and
found himself dreaming a great deal when he should have been working,
according to Sable. He also found that Helen was becoming almost anti-
pathetic to him. She was to make the seconding speech for one of the
candidates at Chicago and was busy with its preparation. There were
conferences constantly, and she had allowed a picture of herself with her
children to be syndicated. Gage found it before him everywhere and it
enraged him. He felt it on his raw mind as an advertisement of the result of
their love, as a dragging into publicity of the last bond between them.
“I feel like the husband of a moving picture actress,” he told her,
viciously, one day.
She said what she had never meant to say. She was tired and full of
worrying and important matters. Gage and his brooding seemed childish
and morbid. And she had her own secret grievance.
“From what I hear of your escapades at the Roadside Inn you act like the
husband of one,” she retorted.
She had not meant to say that. But when the gossip about Freda had
reached her there had come an ugly coupling in her mind of that gossip and
Gage’s interest in the girl. During that very week-end Gage had been absent
from the city—on political business—he had said vaguely. Yet she had tried
to control her suspicions, convince herself that there was no cause for
investigation or accusation. This flare of hers was unexpected and
unguarded—dangerous too.
A shudder of misery shot through both of them at their own coarseness.
But they were launched. And it was clear to Gage that in some way or other
not only Sable but Helen had thought him involved with Freda. It did not
make him particularly angry. He rather courted the injustice of the suspicion
because it justified him in his own position. This was where this business of
Helen’s had landed them then. Alienated, loveless, suspicious—this was the
natural outcome of the whole thing. Minds running on sex all the time—that
was what happened to these women—yet without delicacy, without reserve.
So she thought he was like that, did she? She was thinking that sort of
viciousness while he’d been trying to protect her even from himself. What
was the use of it all?
“I don’t know what you hear of my escapades as you call them,” he
answered. “Possibly you might inform me?”
She was sick with shame at her own impulse but perhaps it had been at
the bottom of her mind corroding it more than she knew.
“I didn’t mean to say that, Gage.”
“You must have meant something.”
He was insistent, brutal. He would have the truth out of her. He wanted
the inside of her mind, to torture himself with it if he could. He wanted it
over with.
“Not to-night, Gage. I’m tired. Let’s talk over some of these things when
we are both fresh. I—I apologize.”
She moved towards the door of the living-room on her way upstairs. But
Gage caught her hand. He stood looking down at her and as she met his
eyes she saw that his face was almost strange. His eyes looked queer. They
were brutal, excited, strange glints. His mouth seemed to hang loose and
heavy.
“Not to-night, Gage,” she repeated. In her voice was a droop of
weariness that was unmistakable.
“Why not to-night? Because you want to save yourself fresh for your
public to-morrow? You don’t want to be bothered with a husband and his
annoyances?”
“Not to-night because you aren’t in the right mood.”
He still held her hand.
“But suppose I want to go into it to-night. There’ll be no better time.
Day after to-morrow my wife goes to the National Convention to dazzle the
American public. Suppose she sets her house in order first. Every good
politician does that, Helen.”
“There’s a devil in you, Gage, isn’t there?”
“A hundred, and every one bred by you. Tell me, what you were
referring to as my escapades? Tell me.”
He shook her a little. She felt a hairpin loosened and the indignity
suddenly made her furious.
“Let me go.”
“I will not let you go. I want you to tell me.”
“I’ll tell you,” she said bitterly, her words coming as if anger pushed
them out. “Heaven knows I’ve tried to conceal it even from myself. But
your viciousness shows you’ve got a rotten conscience. When you took that
Thorstad girl into your office I wondered why—and then after I told you
she’d been seen at that place with a man, your silly defence of her might
have told me what was the situation. You talk of her—all the time—all the
time. You were away that week-end. Where were you if you weren’t with
her?”
He let her go then. She had said it. It was said, as he had wanted it said.
He felt triumphant. And he would give her no satisfaction. He would hurt
her—and hurt her.
She went on in a tumbled burst of words.
“I don’t blame the girl, though she’s a little fool. But I won’t stand
having her let in for that sort of thing.”
“Why not?” asked Gage, lighting a cigarette. “Isn’t it a perfectly proper
thing for a modern woman to choose her lovers where she will?”
Helen felt herself grow dizzy, not at his question but at the admission it
made. She drew herself up and Gage wondered at her beauty with a hot
surge of desire even while he wanted to torture her more. It was such a
relief to have found a weapon.
“Come,” he went on, “we won’t discuss that young lady. There’s not a
thing in the world against her. If you have been bending your ear to the
ground and heard a lot of rotten gossip I’m not responsible. If the people
who talk about her had half her quality—”
“I warn you, Gage, you’re going to pieces,” interrupted Helen. “I can’t
stop you if you’re determined to ruin yourself. But you’ve acted like a
pettish child for months about the fact that I wanted to do some work you
didn’t approve of, apparently you’ve run off and got mixed up with this girl,
you’ve been drinking far too much—you had whisky before breakfast this
morning—it’s beginning to tell on you.”
“I miss you, Helen,” said Gage with a kind of sinister sarcasm.
She shivered.
“I’m going upstairs.”
“We’re not through.”
“Yes, we are.”
“Aren’t you going to divorce me—or would that hurt your career?”
“You’re not yourself, Gage,” said Helen. She had regained a loose hold
on herself. “I’d sooner not talk to you any more to-night.”
He flattened the end of his lighted cigarette and pulled the chain of the
table light.
“Then we’ll talk upstairs.”
“Not to-night.”
“Yes, we will, Helen. I’m lonely for you.” He came to where she stood.
“Come along, my dear.”
There was not a tone in his voice that Helen could recognize. A kind of
ugly caress—she shuddered.
He put his arm around her shoulders.
“Gage—you mustn’t touch me like this.”
He laughed at her.
“It’s quite the new way, as I understand it, my dear, isn’t it? Nature—
openness—no false modesties, no false sentiments. After all we are married
—or to be more modern, we’re openly living together. The pictures in the
paper prove it. There’s no use being silly. You’ve had your way a lot lately
—now how about mine?”
He pulled her close to him and pushing back her head sought her lips
roughly, as if he were dying of thirst and cared little what healthy or
unhealthy drink he had found.
III
“You know,” said Cele Nesbitt to Freda, “I think Mr. Flandon acts kind
of queer, don’t you?”
“He’s tired, probably,” she told Cele.
“Doesn’t look tired. He seems so excited. I thought he and old Sable
must be having a row. I went into Sable’s office with some papers to-day
and there they were glowering at each other and mum as oysters all the time
I was in the room. They don’t stop talking business when I’m around.”
“Well, don’t worry about them,” answered Freda, “Mr. Flandon is the
kindest person I know and there’s something wrong with people who can’t
agree with him.”
“Hate him, don’t you?” Cele teased her. “Isn’t it a pity he’s married. And
such a stunning wife and children. Did you see her picture on Sunday? She
ought to be in the movies instead of politics with that hair.”
Except for Margaret Freda saw only one other person at very close
range. That was Gage’s stenographer, Cecilla Nesbitt, commonly known as
Cele. Cele was a joyous soul who had taken a liking to Freda and shortly
invited her to come home for dinner. Freda had gone and been made happy
and intimate at once. There were all the traces of the cottage that the
Nesbitts had before they moved to St. Pierre—old rattan rocking chairs and
scroll topped beds. Over everything, invading everything was the Church.
There was a little holy water font inside the door, there were pictures and
holy cards framed and unframed everywhere, crucifixes over the beds, holy
pictures in the bureau frames and rosaries on the bed posts. To Freda in her
sparsely religious home, God had been a matter of church on Sunday and
not much more than that except a Bible for reference and a general
astronomical warder at the enormity of God’s achievements. This difference
—this delightful easy intimacy with God was all fascinating. This was the
comfort of religion, religion by your bedside and at your table. She
expanded under it. There was a plenitude of Nesbitts, sleeping rather thickly
in the four bedrooms—two brothers, young men of twenty or thereabouts—
there was Cele after them and then two younger girls of ten and thirteen and
stepping rapidly downward the twins of nine, Mrs. Nesbitt having finished
her family with a climax, especially as the twins were boys and made up for
being altar boys on Sunday by being far from holy on all other occasions.
Still their serving of Mass endowed them in the eyes of Mrs. Nesbitt with
peculiar virtues. She had a gently conciliatory Irish way towards her sons
rather different from her tone to her daughters. Freda contrasted it with
some amusement with the cold classicism of Margaret’s attitude. To Mrs.
Nesbitt they were obviously slightly inferior in the sight of God and man,
being female, to be cherished indeed, frail perhaps, and yet not made in the
exact image of the Creator.
They were headed for the Nesbitt flat. Freda had no letter from Gregory,
had had none for two days and her heart felt as if it were thickening and
sinking. She would not let it be so. She set to work to make herself
interested. She would not mope. It was not in her to mope. But she did not
know where Gregory was, for his last letter had said he was waiting advice
from the bureau—one of his talks having been cancelled—and that he
didn’t know where he would go now. It did not make her worried or
nervous but she had been drugging her emotions with his letters and the
sudden deprivation hurt her cruelly. So she was going home with Cele to
forget it.
They got on the street car and hung from their straps with the
nonchalance of working girls who have no hopes or wishes that men will
give up their seats to them, their attitude strangely different from that of
some of the women, obviously middle class housewives, who
commandeered seats with searching, disapproving, nagging eyes. Freda
loved this time of day—the sense of being with people all going to their
places of living, fraught with mystery and possibility. Her spirits rose. She
was not thinking sadly of Gregory. She thought of how her intimate thought
and knowledge of him reached out, over her unfamiliarity with these others,
touching him wherever he was, in some place unknown to her. The thought
put new vigor into her loneliness.
It was an oppressively hot evening for June. They climbed the three
flights to the Nesbitt flat with diminishing energy and Cele sank on one of
the living-room chairs in exhaustion as she went in.
“Hot as hell,” she breathed. “Let’s sit down a minute before we wash,
Freda.”
Freda took off her hat and brushed her hair back with her hand.
“Pretty hot all right. Bad weather for dispositions.”
“My idea of this kind of weather is that it’s preparation for the
hereafter.”
Mrs. Nesbitt opened the door to the kitchen and hot heavy smells from
the cooking food came through to the girls. But Mrs. Nesbitt herself,
mopping great hanging drops of sweat from her forehead, was serene
enough. She shook hands with Freda with vast smiling cordiality.
“You’re as cool looking as the dawn,” she said to her. “Are you tired,
dear?”
“Not a bit.”
“There’s a little droop to your eyes, dear. I thought maybe it was bad
news now.”
Freda had a sudden impulse to confidence, a leap of the mind towards it.
But she drew back.
“No—not bad news at all.”
“Your mother and father’s well?”
“My mother is coming to see me for a few days, I think. She’s going to
Chicago for the Convention for the clubs and she’ll come back this way to
see me.”
“Now, isn’t that the blessing for you,” said Mrs. Nesbitt rejoicingly.
The family streamed in, the boys from their work and the twins from
school. Last came Mr. Nesbitt, his tin lunch pail in his hand, his feet
dragging with weariness. They talked of the heat, all of them, making it
even more oppressive than it was by their inability to escape the thought of
it. And Mrs. Nesbitt who knew nothing of salads and iced tea, or such hot
weather reliefs stirred the flour for her gravy and set the steaming pot roast
before her husband. They ate heavily. Freda tried to keep her mind on what
she was doing. She talked to the boys and let Mrs. Nesbitt press more food
on her unwilling appetite. It was very unwilling. She did not want to eat.
She wanted to sit down and close her eyes and forget food and heat and
everything else—except Gregory.
Vaguely she was aware of Mr. Nesbitt talking.
“It was in the paper and no more stir made of it than if a stray dog was
run over by an automobile—shot down they were, martyrs to Ireland.” His
voice was oratorical, funereal, heavy with resentment.
“Who?” asked Freda.
“Fine young Irishmen with the grace of God in their hearts shot down by
the hired wastrels of the Tyrants. Gentlemen and patriots.”
“What an outrage it is,” she answered.
He burst into invective at her sympathy, rolling his mighty syllabled
words in denunciation, and his family sat around and listened in agreement
yet in amusement.
“Come now, pop, you’ll be going back, if you get as hot under your shirt
as all that,” said Mike.
“It’s too hot for excitement, pa,” Mrs. Nesbitt contributed equably. “Pass
him the mustard, do you, Cele.”
“I’ll show you a true account of it in The Irish News,” said Mr. Nesbitt,
to Freda, ignoring his family.
He wiped his mouth noisily and abandoned the table, coming back to
press into Freda’s hands his Irish News, a little out of fold with much
handling.
“The city papers tell you nothing but lies,” he said, “read this.”
To please him, Freda read. She read the account of the shooting of three
young men poets and patriots, whose names struck her as familiar. And then
she read:
“These young martyrs were part of the group who banded together for
restoration of the Gaelic tongue to Ireland. They with Seumas, McDermitt
and Gregory Macmillan now on tour in this country—”
She read it again. It gave her a sense of wonder to come on his name
here, his name so secretly dear, in this cold print. And then came more than
that. This was Gregory—her Gregory who might have been killed too if he
had been there—who might be killed when he returned to Ireland. She
didn’t know where he was. Perhaps—perhaps he had heard of this and gone
back. Perhaps he had forgotten, forgotten about her—about them. This was
so big—
She had to take her thought away from the presence of all these people.
She wanted to con it over—she must get away. Suddenly she stood up and
the heat and distaste for food—the accurate sight of a piece of brown
stringy meat, embedded in lifeless gravy, sickened her. She pressed her
hand before her eyes and swayed a little.
Mrs. Nesbitt jumped up with Cele.
“She’s sick—poor dear. The heat now has quite overcome her.”
They helped her into the least hot of the little bedrooms and she found
herself very sick—nauseated—chilled even while she was conscious of the
heat that oppressed while it did not warm her. The family was all astir. Mr.
Nesbitt underwent censure for having bothered her. But when Freda,
apologetic and recovered, went home on Mike’s arm, getting the first breath
of air which came as a relief to the hot city, Mrs. Nesbitt came into the
room where Cele hung half out of the window trying to catch the breeze.
“Sick she was, poor thing.”
“Rotten heat got her. She’s not used to working, either, I think. She felt a
lot better. Her stomach got upset too.”
Mrs. Nesbitt pressed her lips together.
“It was a funny way she was taken. If she was a married woman I should
have said the cause was not the heat.”
“Huh?” said Cele, pulling herself in. “What’s that you mean?”
“I mean nothing,” said Mrs. Nesbitt. “Nothing at all. Only I would have
you always be sure to make sure your friends are good girls, my darlin’.
Mind ye, I say nothing against the young lady. But she’s a pretty and
dangerous face and she’s away from her home where by rights should every
girl be.”
CHAPTER XV
THE CONVENTION