Professional Documents
Culture Documents
EMFT Book 1300612
EMFT Book 1300612
net/publication/258221118
CITATIONS READS
87 50,607
1 author:
Bo Thidé
Swedish Institute of Space Physics
477 PUBLICATIONS 5,016 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
All content following this page was uploaded by Bo Thidé on 01 June 2014.
Bo Thidé
electromagnetic
field theory
FT
A
second edition
R
D
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 2. of 300
FT
A
R
D
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 3. of 300
FT
A
R
D
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 4. of 300
FT
A
R
D
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 5. of 300
ELECTROMAGNETIC
FIELD THEORY
Second Edition
Bo Thidé
FT
Swedish Institute of Space Physics
Uppsala, Sweden
and
Department of Physics and Astronomy
Uppsala University, Sweden
and
Galilean School of Higher Education
A
University of Padua
Padua, Italy
R
D
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 6. of 300
Also available
FT
www.plasma.uu.se/CED
This book was typeset in LATEX 2" based on TEX 3.1415926 and Web2C 7.5.6
A
Copyright ©1997–2011 by
Bo Thidé
Uppsala, Sweden
All rights reserved.
R
Electromagnetic Field Theory
ISBN 978-0-486-4773-2
D
The cover graphics illustrates the linear momentum radiation pattern of a radio beam endowed with orbital
angular momentum, generated by an array of tri-axial antennas. This graphics illustration was prepared by
J O H A N S J Ö H O L M and K R I S T O F F E R PA L M E R as part of their undergraduate Diploma Thesis work in En-
gineering Physics at Uppsala University 2006–2007.
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 7. of 300
FT
A
R
D
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 8. of 300
FT
A
R
D
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 9. of 300
CONTENTS
Contents ix
FT
1.1.1 Coulomb’s law . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.2 The electrostatic field . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.2 Magnetostatics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.2.1 Ampère’s law . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.2.2 The magnetostatic field . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.3 Electrodynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.3.1 The indestructibility of electric charge . . . . . . . . . . . . . . . . . . . 9
1.3.2 Maxwell’s displacement current . . . . . . . . . . . . . . . . . . . . . . 10
1.3.3 Electromotive force . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.3.4 Faraday’s law of induction . . . . . . .
A
. . . . . . . . . . . . . . . . . . 12
1.3.5 The microscopic Maxwell equations . . . . . . . . . . . . . . . . . . . . 15
1.3.6 Dirac’s symmetrised Maxwell equations . . . . . . . . . . . . . . . . . . 15
1.4 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
1.5 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
R
2 Electromagnetic Fields and Waves 19
2.1 Axiomatic classical electrodynamics . . . . . . . . . . . . . . . . . . . . . . . . 19
2.2 Complex notation and physical observables . . . . . . . . . . . . . . . . . . . . 20
2.2.1 Physical observables and averages . . . . . . . . . . . . . . . . . . . . . 21
2.2.2 Maxwell equations in Majorana representation . . . . . . . . . . . . . . . 22
D
ix
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 10. of 300
x CONTENTS
FT
4.1.3 P symmetry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
4.1.4 T symmetry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
4.2 Continuous symmetries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
4.2.1 General conservation laws . . . . . . . . . . . . . . . . . . . . . . . . . 57
4.2.2 Conservation of electric charge . . . . . . . . . . . . . . . . . . . . . . . 58
4.2.3 Conservation of energy . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
4.2.4 Conservation of linear (translational) momentum . . . . . . . . . . . . . 61
4.2.4.1 Gauge-invariant operator formalism . . . . . . . . . . . . . . . 64
4.2.5 Conservation of angular (rotational) momentum . . . . . . . . . . . . . . 66
A
4.2.5.1 Gauge-invariant operator formalism . . . . . . . . . . . . . . . 69
4.2.6 Electromagnetic duality . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
4.2.7 Electromagnetic virial theorem . . . . . . . . . . . . . . . . . . . . . . . 73
4.3 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
4.4 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
R
5 Fields from Arbitrary Charge and Current Distributions 85
5.1 Fourier component method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
5.2 The retarded electric field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
5.3 The retarded magnetic field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
D
5.4 The total electric and magnetic fields at large distances from the sources . . . . . 93
5.4.1 The far fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
5.5 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
5.6 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
CONTENTS xi
FT
6.5.2.2 The direct method . . . . . . . . . . . . . . . . . . . . . . . . 129
6.5.2.3 Small velocities . . . . . . . . . . . . . . . . . . . . . . . . . . 131
6.5.3 Bremsstrahlung . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
6.5.4 Cyclotron and synchrotron radiation . . . . . . . . . . . . . . . . . . . . 135
6.5.4.1 Cyclotron radiation . . . . . . . . . . . . . . . . . . . . . . . . 137
6.5.4.2 Synchrotron radiation . . . . . . . . . . . . . . . . . . . . . . 138
6.5.4.3 Radiation in the general case . . . . . . . . . . . . . . . . . . . 140
6.5.4.4 Virtual photons . . . . . . . . . . . . . . . . . . . . . . . . . . 141
6.6 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
A
6.7 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
xii CONTENTS
FT
9 Electromagnetic Fields and Matter 191
9.1 Maxwell’s macroscopic theory . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
9.1.1 Polarisation and electric displacement . . . . . . . . . . . . . . . . . . . 192
9.1.2 Magnetisation and the magnetising field . . . . . . . . . . . . . . . . . . 193
9.1.3 Macroscopic Maxwell equations . . . . . . . . . . . . . . . . . . . . . . 195
9.2 Phase velocity, group velocity and dispersion . . . . . . . . . . . . . . . . . . . 196
9.3 Radiation from charges in a material medium . . . . . . . . . . . . . . . . . . . 197
L
9.3.1 Vavilov-Cerenkov radiation . . . . . . . . . . . . . . . . . . . . . . . . . 197
9.4 Electromagnetic waves in a medium . . . . . . . . . . . . . . . . . . . . . . . . 202
A
9.4.1 Constitutive relations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
9.4.2 Electromagnetic waves in a conducting medium . . . . . . . . . . . . . . 205
9.4.2.1 The wave equations for E and B . . . . . . . . . . . . . . . . . 205
9.4.2.2 Plane waves . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
9.4.2.3 Telegrapher’s equation . . . . . . . . . . . . . . . . . . . . . . 207
R
9.5 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
F Formulæ 215
F .1 Vector and tensor fields in 3D Euclidean space . . . . . . . . . . . . . . . . . . . 215
F .1.1 Cylindrical circular coordinates . . . . . . . . . . . . . . . . . . . . . . . 216
D
CONTENTS xiii
FT
F .2 The electromagnetic field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
F .2.1 Microscopic Maxwell-Lorentz equations in Dirac’s symmetrised form . . 225
F .2.1.1 Constitutive relations . . . . . . . . . . . . . . . . . . . . . . . 225
F .2.2 Fields and potentials . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226
F .2.2.1 Vector and scalar potentials . . . . . . . . . . . . . . . . . . . 226
F .2.2.2 The velocity gauge condition in free space . . . . . . . . . . . 226
F .2.2.3 Gauge transformation . . . . . . . . . . . . . . . . . . . . . . 226
F .2.3 Energy and momentum . . . . . . . . . . . . . . . . . . . . . . . . . . . 226
F .2.3.1 Electromagnetic field energy density in free space . . . . . . . 226
A
F .2.3.2 Poynting vector in free space . . . . . . . . . . . . . . . . . . . 226
F .2.3.3 Linear momentum density in free space . . . . . . . . . . . . . 226
F .2.3.4 Linear momentum flux tensor in free space . . . . . . . . . . . 227
F .2.3.5 Angular momentum density around x0 in free space . . . . . . 227
F .2.3.6 Angular momentum flux tensor around x0 in free space . . . . 227
R
F .2.4 Electromagnetic radiation . . . . . . . . . . . . . . . . . . . . . . . . . . 227
F .2.4.1 The far fields from an extended source distribution . . . . . . . 227
F .2.4.2 The far fields from an electric dipole . . . . . . . . . . . . . . . 227
F .2.4.3 The far fields from a magnetic dipole . . . . . . . . . . . . . . 228
F .2.4.4 The far fields from an electric quadrupole . . . . . . . . . . . . 228
D
xiv CONTENTS
FT
M .1.1.1 Position vector . . . . . . . . . . . . . . . . . . . . . . . . . . 235
M .1.2 Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
M .1.2.1 Scalar fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
M .1.2.2 Vector fields . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
M .1.2.3 Coordinate transformations . . . . . . . . . . . . . . . . . . . 237
M .1.2.4 Tensor fields . . . . . . . . . . . . . . . . . . . . . . . . . . . 238
M .2 Vector algebra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
M .2.1 Scalar product . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
A
M .2.2 Vector product . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
M .2.3 Dyadic product . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
M .3 Vector calculus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
M .3.1 The del operator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
M .3.2 The gradient of a scalar field . . . . . . . . . . . . . . . . . . . . . . . . 245
R
M .3.3 The divergence of a vector field . . . . . . . . . . . . . . . . . . . . . . 245
M .3.4 The curl of a vector field . . . . . . . . . . . . . . . . . . . . . . . . . . 245
M .3.5 The Laplacian . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246
M .3.6 Vector and tensor integrals . . . . . . . . . . . . . . . . . . . . . . . . . 246
D
CONTENTS xv
Index 269
FT
A
R
D
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 16. of 300
FT
A
R
D
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 17. of 300
LIST OF FIGURES
FT
6.5 Radiation from a moving charge in vacuum . . . . . . . . . . . . . . . . . . . . . . 122
6.6 An accelerated charge in vacuum . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
6.7 Angular distribution of radiation during bremsstrahlung . . . . . . . . . . . . . . . . 132
6.8 Location of radiation during bremsstrahlung . . . . . . . . . . . . . . . . . . . . . . 134
6.9 Radiation from a charge in circular motion . . . . . . . . . . . . . . . . . . . . . . . 136
6.10 Synchrotron radiation lobe width . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
6.11 The perpendicular electric field of a moving charge . . . . . . . . . . . . . . . . . . 141
6.12 Electron-electron scattering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
6.13 Loop antenna . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
A
7.1 Relative motion of two inertial systems . . . . . . . . . . . . . . . . . . . . . . . . . 156
7.2 Rotation in a 2D Euclidean space . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
7.3 Minkowski diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
xvii
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 18. of 300
FT
A
R
D
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 19. of 300
FT
use in Academia and elsewhere.
The subject matter starts with a description of the properties of electromagnetism when the
charges and currents are located in otherwise free space, i.e., a space that is free of matter and
external fields (e.g., gravitation). A rigorous analysis of the fundamental properties of the elec-
tromagnetic fields and radiation phenomena follows. Only then the influence of matter on the
fields and the pertinent interaction processes is taken into account. In the author’s opinion, this
approach is preferable since it avoids the formal logical inconsistency of introducing, very early
in the derivations, the effect on the electric and mangetic fields when conductors and dielectrics
are present (and vice versa) in an ad hoc manner, before constitutive relations and physical mod-
A
els for the electromagnetic properties of matter, including conductors and dielectrics, have been
derived from first principles. Curved-space effects on electromagnetism are not treated at all.
In addition to the Maxwell-Lorentz equations, which postulate the beaviour of electromag-
netic fields due to electric charges and currents on a microscopic classical scale, chapter chapter 1
also introduces Dirac’s symmetrised equations that incorporate the effects of magnetic charges
R
and currents. In chapter chapter 2, a stronger emphasis than before is put on the axiomatic found-
ation of electrodynamics as provided by the Maxwell-Lorentz equations that are taken as the
postulates of the theory. Chapter chapter 3 on potentials and gauges now provides a more com-
prehensive picture and discusses gauge invariance in a more satisfactory manner than the first
edition did. Chapter chapter 4 is new and deals with symmetries and conserved quantities in a
D
more rigourous, profound and detailed way than in the first edition. For instance, the presentation
of the theory of electromagnetic angular momentum and other observables (constants of motion)
has been substantially expanded and put on a firm basis. Chapter chapter 9 is a complete rewrite
and combines material that was scattered more or less all over the first edition. It also contains
new material on wave propagation in plasma and other media. When, in chapter chapter 9, the
macroscopic Maxwell equations are introduced, the inherent approximations in the derived field
quantities are clearly pointed out. The collection of formulæ in appendix F on page 215 has been
augmented quite substantially. In appendix M on page 233, the treatment of dyadic products and
tensors has been expanded significantly and numerous examples have been added throughout.
xix
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 20. of 300
FT
S O N , and J O H A N L I N D B E R G who during their work on their Diploma theses suggested im-
provements and additions and to H O L G E R T H E N and S TA F F A N Y N G V E for carefully checking
some lengthy calculations and to the numerous undergraduate students, who have been exposed
to various draft versions of this second edtion. In particular, I would like to mention B RU N O
S T R A N D B E R G.
This book is dedicated to my son M AT T I A S , my daughter K A RO L I NA , my four grandsons
M A X , A L B I N , F I L I P and O S K A R , my high-school physics teacher, S TA F F A N R Ö S B Y , and
my fellow members of the C A P E L L A P E DAG O G I C A U P S A L I E N S I S .
A
Padova, Italy BO THIDÉ
February, 2011 www.physics.irfu.se/bt
R
D
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 21. of 300
Of the four known fundamental interactions in nature—gravitational, strong, weak, and electro-
magnetic—the latter has a special standing in the physical sciences. Not only does it, together
with gravitation, permanently make itself known to all of us in our everyday lives. Electro-
dynamics is also by far the most accurate physical theory known, tested on scales running from
sub-nuclear to galactic, and electromagnetic field theory is the prototype of all other field theories.
This book, E L E C T RO M AG N E T I C F I E L D T H E O RY , which tries to give a modern view of
classical electrodynamics, is the result of a more than thirty-five year long love affair. In the
autumn of 1972, I took my first advanced course in electrodynamics at the Department of Theor-
etical Physics, Uppsala University. Soon I joined the research group there and took on the task
of helping the late professor P E R O L O F F RÖ M A N , who was to become my Ph.D. thesis ad-
viser, with the preparation of a new version of his lecture notes on the Theory of Electricity. This
FT
opened my eyes to the beauty and intricacy of electrodynamics and I simply became intrigued by
it. The teaching of a course in Classical Electrodynamics at Uppsala University, some twenty odd
years after I experienced the first encounter with the subject, provided the incentive and impetus
to write this book.
Intended primarily as a textbook for physics and engineering students at the advanced under-
graduate or beginning graduate level, it is hoped that the present book will be useful for research
workers too. It aims at providing a thorough treatment of the theory of electrodynamics, mainly
from a classical field-theoretical point of view. The first chapter is, by and large, a descrip-
tion of how Classical Electrodynamics was established by J A M E S C L E R K M A X W E L L as a
A
fundamental theory of nature. It does so by introducing electrostatics and magnetostatics and
demonstrating how they can be unified into one theory, classical electrodynamics, summarised
in Lorentz’s microscopic formulation of the Maxwell equations. These equations are used as an
axiomatic foundation for the treatment in the remainder of the book, which includes modern for-
mulation of the theory; electromagnetic waves and their propagation; electromagnetic potentials
R
and gauge transformations; analysis of symmetries and conservation laws describing the elec-
tromagnetic counterparts of the classical concepts of force, momentum and energy, plus other
fundamental properties of the electromagnetic field; radiation phenomena; and covariant Lag-
rangian/Hamiltonian field-theoretical methods for electromagnetic fields, particles and interac-
tions. Emphasis has been put on modern electrodynamics concepts while the mathematical tools
D
used, some of them presented in an Appendix, are essentially the same kind of vector and tensor
analysis methods that are used in intermediate level textbooks on electromagnetics but perhaps a
bit more advanced and far-reaching.
The aim has been to write a book that can serve both as an advanced text in Classical Elec-
trodynamics and as a preparation for studies in Quantum Electrodynamics and Field Theory, as
well as more applied subjects such as Plasma Physics, Astrophysics, Condensed Matter Physics,
Optics, Antenna Engineering, and Wireless Communications.
The current version of the book is a major revision of an earlier version, which in turn was an
outgrowth of the lecture notes that the author prepared for the four-credit course Electrodynam-
xxi
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 22. of 300
ics that was introduced in the Uppsala University curriculum in 1992, to become the five-credit
course Classical Electrodynamics in 1997. To some extent, parts of those notes were based on
lecture notes prepared, in Swedish, by my friend and Theoretical Physics colleague B E N G T
L U N D B O R G , who created, developed and taught an earlier, two-credit course called Electro-
magnetic Radiation at our faculty. Thanks are due not only to Bengt Lundborg for providing
the inspiration to write this book, but also to professor C H R I S T E R WA H L B E R G , and professor
G Ö R A N F Ä L D T , both at the Department of Physics and Astronomy, Uppsala University, for
insightful suggestions, to professor J O H N L E A R N E D , Department of Physics and Astronomy,
University of Hawaii, for decisive encouragement at the early stage of this book project, to pro-
fessor G E R A R D U S T ’ H O O F T , for recommending this book on his web page ‘How to become
a good theoretical physicist’, and professor C E C I L I A J A R L S KO G , Lund Unversity, for pointing
out a couple of errors and ambiguities.
I am particularly indebted to the late professor V I TA L I Y L A Z A R E V I C H G I N Z B U R G , for
FT
his many fascinating and very elucidating lectures, comments and historical notes on plasma
physics, electromagnetic radiation and cosmic electrodynamics while cruising up and down the
Volga and Oka rivers in Russia at the ship-borne Russian-Swedish summer schools that were
organised jointly by late professor L E V M I K A H I L OV I C H E RU K H I M OV and the author during
the 1990’s, and for numerous deep discussions over the years.
Helpful comments and suggestions for improvement from former PhD students T O B I A C A -
RO Z Z I , R O G E R K A R L S S O N , and M AT T I A S WA L D E N V I K , as well as A N D E R S E R I K S S O N
at the Swedish Institute of Space Physics in Uppsala and who have all taught Uppsala students
on the material covered in this book, are gratefully acknowledged. Thanks are also due to the late
A
H E L M U T K O P K A , for more than twenty-five years a close friend and space physics colleague
working at the Max-Planck-Institut für Aeronomie, Lindau, Germany, who not only taught me
the practical aspects of the use of high-power electromagnetic radiation for studying space, but
also some of the delicate aspects of typesetting in TEX and LATEX.
In an attempt to encourage the involvement of other scientists and students in the making of
R
this book, thereby trying to ensure its quality and scope to make it useful in higher university
education anywhere in the world, it was produced as a World-Wide Web (WWW) project. This
turned out to be a rather successful move. By making an electronic version of the book freely
downloadable on the Internet, comments have been received from fellow physicists around the
world. To judge from WWW ‘hit’ statistics, it seems that the book serves as a frequently used
D
Internet resource. This way it is hoped that it will be particularly useful for students and research-
ers working under financial or other circumstances that make it difficult to procure a printed copy
of the book. I would like to thank all students and Internet users who have downloaded and
commented on the book during its life on the World-Wide Web.
FOUNDATIONS OF CLASSICAL
ELECTRODYNAMICS
The classical theory of electromagnetism deals with electric and magnetic fields
and their interaction with each other and with charges and currents. This theory
is classical in the sense that it assumes the validity of certain mathematical limit-
ing processes in which it is considered possible for the charge and current distri-
FT
butions to be localised in infinitesimally small volumes of space.1 Clearly, this 1Accepting the mere existence of
is in contradistinction to electromagnetism on an atomistic scale, where charges an electrically charged particle re-
quires some careful thinking. In his
and currents have to be described in a nonlocal quantum formalism. However, excellent book Classical Charged
the limiting processes used in the classical domain, which, crudely speaking, as- Particles, F R I T Z R O H R L I C H
writes
sume that an elementary charge has a continuous distribution of charge density,
‘To what extent does it
will yield results that agree perfectly with experiments on non-atomistic scales, make sense to talk about an
however small or large these scales may be.2 electron, say, in classical
terms? These and similar
It took the genius of J A M E S C L E R K M A X W E L L to consistently unify, in questions clearly indicate
the mid-1800’s, the theory of Electricity and the then distinctively different the- that ignoring philosophy
A
in physics means not
ory Magnetism into a single super-theory, Electromagnetism or Classical Elec- understanding physics.
trodynamics (CED), and also to realise that optics is a sub-field of this super- For there is no theoretical
physics without some
theory. Early in the 20th century, H E N D R I K A N T O O N L O R E N T Z took the philosophy; not admitting
electrodynamics theory further to the microscopic scale and also paved the way this fact would be self-
deception.’
for the Special Theory of Relativity, formulated in its full extent by A L B E RT
R
E I N S T E I N in 1905. In the 1930’s PAU L A D R I E N M AU R I C E D I R AC expan- 2Electrodynamics has been tested
experimentally over a larger range
ded electrodynamics to a more symmetric form, including magnetic as well as of spatial scales than any other
electric charges. With his relativistic quantum mechanics and field quantisation existing physical theory.
concepts, Dirac had already in the 1920’s laid the foundation for Quantum Elec-
trodynamics (QED ), the relativistic quantum theory for electromagnetic fields
D
1
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 24. of 300
FT
rents into the theory in a symmetric way. In practical work, such as in antenna
engineering, magnetic currents have proved to be a very useful concept. We
shall make some use of this symmetrised theory of electricity and magnetism.
1.1 Electrostatics
A
3The physicist, mathematician and The theory that describes physical phenomena related to the interaction between
philosopher P I E R R E M AU R I C E
M A R I E D U H E M (1861–1916)
stationary electric charges or charge distributions in a finite space with stationary
once wrote: boundaries is called electrostatics. For a long time, electrostatics, under the
‘The whole theory of name Electricity, was considered an independent physical theory of its own,
electrostatics constitutes
a group of abstract ideas alongside other physical theories such as Magnetism, Mechanics, Optics, and
R
and general propositions, Thermodynamics.3
formulated in the clear
and concise language of
geometry and algebra,
and connected with one
another by the rules of 1.1.1 Coulomb’s law
strict logic. This whole
D
1.1. Electrostatics j3
mathematics as
qq 0 x x0 qq 0
FT qq 0 0
es 1 1
F .x/ D D r D r
4"0 jx x0 j3 4"0 jx x0 j 4"0 jx x0 j
(1.1)
where, in the last step, formula (F.114) on page 222 was used. In SI units, which
we shall use throughout, the electrostatic force5 F es is measured in Newton (N), 5 Massive particles also interact
the electric charges q and q 0 in Coulomb (C), i.e. Ampere-seconds (As), and gravitationally but with a force that
is typically 10 36 times weaker;
the length jx x0 j in metres (m). The constant "0 D 107 =.4c 2 / Farad per we shall not take gravitational
metre (Fm 1 ) is the permittivity of free space and c ms 1 is the speed of light interactions into acocunt.
A
in vacuum.6 In CGS units, "0 D 1=.4/ and the force is measured in dyne, 6 The notation c for speed stems
electric charge in statcoulomb, and length in centimetres. from the Latin word ‘celeritas’
which means ‘swiftness’. This
notation seems to have been intro-
duced by W I L H E L M E D UA R D
1.1.2 The electrostatic field W E B E R (1804–1891), and
R U D O L F K O H L R AU S C H (1809–
R
1858) and c is therefore sometimes
Instead of describing the electrostatic interaction in terms of a ‘force action at a
referred to as Weber’s constant.
distance’, it turns out that for many purposes it is useful to introduce the concept In all his works from 1907 and
of a field. Thus we describe the electrostatic interaction in terms of a static onward, A L B E RT E I N S T E I N
(1879–1955) used c to denote the
vectorial electric field Estat defined by the limiting process speed of light in free space.
F es .x/
D
def
Estat .x/ lim (1.2)
q!0 q
where F es is the electrostatic force, as defined in equation (1.1) above, from a net
electric charge q 0 on the test particle with a small net electric charge q.7 In the SI 7 If we picture the test charge as an
system of units, electric fields are therefore measured in NC 1 or, equivalently, electrically charged particle, the
charge of such a particle cannot
in Vm 1 . Since the purpose of the limiting process is to ascertain that the test tend smoothly to 0 simply because
charge q does not distort the field set up by q 0 , the expression for Estat does the lowest allowable amount of
charge is that of an individual
not depend explicitly on q but only on the charge q 0 and the relative position quark, namely 1=3 jej where e is
vector x x0 . This means that we can say that any net electric charge produces the elementary charge
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 26. of 300
an electric field in the space that surrounds it, regardless of the existence of a
8 In the preface to the first edition second charge anywhere in this space.8 However, in order to experimentally
of the first volume of his book detect a charge, a second (test) charge that senses the presence of the first one,
A Treatise on Electricity and
Magnetism, first published in 1873, must be introduced and done so in an adiabatic way.
Maxwell describes this in the Using equations (1.1) and (1.2) on the previous page, and formula (F.114) on
following almost poetic manner:
page 222, we find that the electrostatic field Estat at the observation point (also
‘For instance, Faraday, in
his mind’s eye, saw lines of known as the field point) x, due to a field-producing electric charge q 0 at the
force traversing all space source point x0 , is given by
where the mathematicians
saw centres of force q 0 x x0 q0 q0
stat 1 0 1
attracting at a distance: E .x/ D D r D r
Faraday saw a medium 4"0 jx x0 j3 4"0 jx x0 j 4"0 jx x0 j
where they saw nothing
(1.3)
but distance: Faraday
sought the seat of the
In the presence of several field producing discrete electric charges qi0 , located
FT
phenomena in real actions
going on in the medium, at the points x0i , i D 1; 2; 3; : : : , respectively, in otherwise empty space, the
they were satisfied that they
had found it in a power assumption of linearity of vacuum9 allows us to superimpose their individual
of action at a distance electrostatic fields into a total electrostatic field
impressed on the electric
fluids.’ 1 X 0 x x0i
Estat .x/ D qi ˇ (1.4)
4"0 ˇx x0 ˇ3
ˇ
i i
9 In fact, a vacuum exhibits a If the discrete electric charges are small and numerous enough, we can, in
quantum mechanical non-linearity a continuum limit, assume that the total charge q 0 from an extended volume
due to vacuum polarisation
to be built up by local infinitesimal elemental charges dq 0 , each producing an
A
effects, manifesting themselves
in the momentary creation and elemental electric field
annihilation of electron-positron
pairs, but classically this non- 1 x x0
linearity is negligible. dEstat .x/ D dq 0 (1.5)
4"0 jx x0 j3
1 x x0
Z Z
stat
E .x/ D dE .x/ D stat
d3x 0 .x0 /
4"0 V 0 jx x0 j3
0
1 1 1 3 0 .x /
Z Z
3 0 0
D d x .x /r D r d x (1.7)
4"0 V 0 jx x0 j 4"0 V0 jx x0 j
where we used formula (F.114) on page 222 and the fact that .x0 / does not
depend on the unprimed (field point) coordinates on which r operates.
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 27. of 300
1.1. Electrostatics j5
FT
We emphasise that under the assumption of linear superposition, equation
(1.7) on the facing page is valid for an arbitrary distribution of electric charges,
including discrete charges, in which case is expressed in terms of Dirac delta
distributions:10 10 Since, by definition, the integral
X
.x0 / D qi0 ı.x0 x0i /
Z
(1.8) d3x 0 ı.x0 x0i /
i V0
Z
as illustrated in figure 1.2. Inserting this expression into expression (1.7) on the d3x 0 ı.x 0 xi0 /
V0
facing page we recover expression (1.4) on the preceding page, as we should. ı.y 0
yi0 /ı.z 0 zi0 / D 1
According to Helmholtz’s theorem, discussed in subsection M.3.7, any well- is dimensionless, and x has the SI
A
behaved vector field is completely known once we know its divergence and curl dimension m, the 3D Dirac delta
distribution ı.x0 x0i / must have
at all points x in 3D space.11 Taking the divergence of the general Estat ex-
the SI dimension m 3 .
pression for an arbitrary electric charge distribution, equation (1.7) on the facing
page, and applying formula (F.126) on page 224 [see also equation (M.75) on 11 H E R M A N N L U DW I G
page 248], we obtain F E R D I NA N D VO N H E L M -
H O LT Z (1821–1894) was a
R
1 .x0 / .x/
Z
physicist, physician and philo-
stat
r E .x/ D r r d3x 0 0j
D (1.9a) sopher who contributed to wide
4"0 V 0 jx x "0
areas of science, ranging from
which is the differential form of Gauss’s law of electrostatics. Since, according electrodynamics to ophthalmology.
to formula (F.100) on page 222, r r ˛.x/ 0 for any R3 scalar field ˛.x/,
we immediately find that in electrostatics
D
1 .x0 /
Z
r Estat .x/ D r r d3x 0 D0 (1.9b)
4"0 V0 jx x0 j
i.e. that Estat is a purely irrotational field.
To summarise, electrostatics can be described in terms of two vector partial
differential equations
.x/
r Estat .x/ D (1.10a)
"0
stat
r E .x/ D 0 (1.10b)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 28. of 300
1.2 Magnetostatics
Whereas electrostatics deals with static electric charges (electric charges that do
not move), and the interaction between these charges, magnetostatics deals with
static electric currents (electric charges moving with constant speeds), and the
interaction between these currents. Here we shall discuss the theory of magneto-
statics in some detail.
FT
Experiments on the force interaction between two small loops that carry static
electric currents I and I 0 (i.e. the currents I and I 0 do not vary in time) have
shown that the loops interact via a mechanical force, much the same way that
static electric charges interact. Let F ms .x/ denote the magnetostatic force on
a loop C , with tangential line vector element dl, located at x and carrying a
current I in the direction of dl, due to the presence of a loop C 0 , with tangential
line element dl0 , located at x0 and carrying a current I 0 in the direction of dl0
A
12 ANDRÉ-MARIE AMPÈRE
(1775–1836) was a French math- in otherwise empty space. This spatial configuration is illustrated in graphical
ematician and physicist who,
only a few days after he learned form in figure 1.3 on the facing page.
about the findings by the Danish According to Ampère’s law the magnetostatic force in question is given by
physicist and chemist H A N S
C H R I S T I A N Ø R S T E D (1777– the expression12
1851) regarding the magnetic
R
0 II 0 x x0
I I
effects of electric currents, presen-
ted a paper to the Académie des F ms .x/ D dl dl0
Sciences in Paris, postulating the 4 C C0 jx x0 j3
(1.11)
force law that now bears his name. 0 II 0
1
I I
0
D dl dl r
4 C C0 jx x0 j
space. From the definition of "0 and 0 (in SI units) we observe that
107 1 7 1 1 2 2
"0 0 D (Fm ) 4 10 (Hm ) D (s m ) (1.12)
4c 2 c2
1.2. Magnetostatics j7
FT
we can rewrite (1.11) as
0 II 0
1
I I
ms 0
F .x/ D dl ˝ dl r
4 C0 C jx x0 j
0 (1.13)
0 II x x0
I I
dl dl0 ˝
4 C C0 jx x0 j3
Since the integrand in the first integral over C is an exact differential, this in-
tegral vanishes and we can rewrite the force expression, formula (1.11) on the
preceding page, in the following symmetric way
A
0 II 0 x x0
I I
ms
F .x/ D dl dl0 ˝ (1.14)
4 C C0 jx x0 j3
which clearly exhibits the expected interchange symmetry between loops C and
C 0.
R
magnetic field dBstat .x/ at the field point x due to a line current element di 0 .x0 / D
I 0 dl0 .x0 / D d3x 0 j.x0 / of static current I 0 with electric current density j, meas-
ured in Am 2 in SI units, directed along the local line element dl0 of the loop at
x0 , is
def dF ms .x/ 0 0 0 x x0
dBstat .x/ lim D di .x /
I !0 I 4 jx x0 j3
(1.15)
0 3 0 x x0
D d x j.x0 /
4 jx x0 j3
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 30. of 300
FT
4 V 0 jx x0 j3
(1.16)
0 1
Z
D d3x 0 j.x0 / r
4 V 0 jx x0 j
0
0 j.x /
Z
D r d3x 0
4 V 0 jx x0 j
where we used formula (F.114) on page 222, formula (F.87) on page 221, and
the fact that j.x0 / does not depend on the unprimed coordinates on which r
operates. Comparing equation (1.7) on page 4 with equation (1.16) above, we
see that there exists an analogy between the expressions for Estat and Bstat but
A
that they differ in their vectorial characteristics. With this definition of Bstat ,
equation (1.11) on page 6 may be written
I I
F ms .x/ D I dl Bstat .x/ D di Bstat .x/ (1.17)
C C
R
In order to assess the properties of Bstat , we determine its divergence and
curl. Taking the divergence of both sides of equation (1.16) above and utilising
formula (F.99) on page 222, we obtain
0 j.x0 /
Z
r Bstat .x/ D r r d3x 0 D0 (1.18)
4 V0 jx x0 j
D
since, according to formula (F.99) on page 222, r .r a/ vanishes for any
vector field a.x/.
With the use of formula (F.128) on page 224, the curl of equation (1.16)
above can be written
0 j.x0 /
Z
stat
r B .x/ D r r d3x 0
4 V0 jx x0 j
(1.19)
0 1
Z
3 0 0 0 0
D 0 j.x/ d x Œr j.x /r
4 V 0 jx x0 j
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 31. of 300
1.3. Electrodynamics j9
assuming that j.x0 / falls off sufficiently fast at large distances. For the stationary
currents of magnetostatics, r j D 0 since there cannot be any charge accumu-
lation in space. Hence, the last integral vanishes and we can conclude that
1.3 Electrodynamics
As we saw in the previous sections, the laws of electrostatics and magnetostatics
can be summarised in two pairs of time-independent, uncoupled partial differen-
FT
tial equations, namely the equations of classical electrostatics
.x/
r Estat .x/ D (1.21a)
"0
r Estat .x/ D 0 (1.21b)
FT
1.3.2 Maxwell’s displacement current
We recall from the derivation of equation (1.20) on the previous page that there
we used the fact that in magnetostatics r j.x/ D 0. In the case of non-
stationary sources and fields, we must, in accordance with the continuity equa-
tion (1.23) above, set r j.t; x/ D @.t; x/=@t. Doing so, and formally re-
peating the steps in the derivation of equation (1.20) on the previous page, we
would obtain the result
0
0 3 0 @.t; x / 0 1
Z
r B.t; x/ D 0 j.t; x/ C dx r (1.25)
A
4 V 0 @t jx x0 j
If we assume that equation (1.7) on page 4 can be generalised to time-varying
14 Later, we will need to consider fields, we can make the identification14
this generalisation and formal
1 @ 1
Z
identification further.
d3x 0 .t; x0 /r 0
jx x0 j
R
4"0 @t V 0
@ 1 1
Z
3 0 0
D d x .t; x /r (1.26)
@t 4"0 V 0 jx x0 j
.t; x0 /
@ 1 @
Z
D r d3x 0 0j
D E.t; x/
@t 4"0 V 0 jx x @t
D
where "0 @E.t; x/=@t is the famous displacement current. This, at the time, un-
observed current was introduced by Maxwell, in a stroke of genius, in order to
make also the right-hand side of this equation divergence-free when j.t; x/ is
assumed to represent the density of the total electric current. This total current
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 33. of 300
1.3. Electrodynamics j 11
FT
and chemical processes can give rise to electric currents. Under certain physical
conditions, and for certain materials, one can assume that a linear relationship
exists between the electric current density j and E. This approximation is called
Ohm’s law:15 15In semiconductors this approx-
imation is in general applicable
only for a limited range of E. This
j.t; x/ D E.t; x/ (1.28) property is used in semiconductor
diodes for rectifying alternating
currents.
where is the electric conductivity measured in Siemens per metre (Sm 1 ).
We can view Ohm’s law equation (1.28) as the first term in a Taylor ex-
A
pansion of a general law jŒE.t; x/. This general law incorporates non-linear
effects such as frequency mixing and frequency conversion. Examples of media
that are highly non-linear are semiconductors and plasma. We draw the atten-
tion to the fact that even in cases when the linear relation between E and j is
a good approximation, we still have to use Ohm’s law with care. The conduct-
R
ivity is, in general, time-dependent (temporal dispersive media) but then it is
often the case that equation (1.28) above is valid for each individual temporal
Fourier (spectral) component of the field. In some media, such as magnetised
plasma and certain material, the conductivity is different in different directions.
For such electromagnetically anisotropic media (spatial dispersive media) the
D
field Eemf , Ohm’s law, equation (1.28) on the previous page, takes the form
16 The term ‘electromotive force’ where dl is a tangential line element of the closed loop C .16
is something of a misnomer since
E represents a voltage, i.e. its SI
dimension is V.
1.3.4 Faraday’s law of induction
In subsection 1.1.2 we derived the differential equations for the electrostatic
FT
field. Specifically, on page 5 we derived equation (1.9b) stating that r Estat D 0
and hence that Estat is a conservative field (it can be expressed as a gradient of a
scalar field). This implies that the closed line integral of Estat in equation (1.30)
above vanishes and that this equation becomes
I
E D dl Eemf (1.31)
C
ferential equation
@
r E.t; x/ D B.t; x/ (1.33)
@t
that is valid for arbitrary variations in the fields and constitutes the Maxwell
equation which explicitly connects electricity with magnetism.
Any change of the magnetic flux ˆm will induce an EMF. Let us therefore
consider the case, illustrated in figure 1.4 on the facing page, when the ‘loop’ is
moved in such a way that it encircles a magnetic field which varies during the
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 35. of 300
1.3. Electrodynamics j 13
C
dl
operator formula
d
dt
D
@
@t
C
dx
dt
r
FT
movement. The total time derivative is evaluated according to the well-known
(1.34)
This follows immediately from the multivariate chain rule for the differentiation
of an arbitrary differentiable function f .t; x.t//. Here, dx=dt describes a chosen
path in space. We shall choose the flow path which means that dx=dt D v and
A
d @
D C v r (1.35)
dt @t
where, in a continuum picture, v is the fluid velocity. For this particular choice,
the convective derivative dx=dt is usually referred to as the material derivative
R
and is denoted Dx=Dt .
Applying the rule (1.35) to Faraday’s law, equation (1.32) on the preceding
page, we obtain
@B
Z Z Z
d
E.t/ D d2x nO B D d2x nO d2x nO .v r /B (1.36)
dt S S @t S
D
Furthermore, taking the divergence of equation (1.33) on the facing page, we see
that
@ @
r B.t; x/ D r B.t; x/ D r Œr E.t; x/ D 0 (1.37)
@t @t
where in the last step formula (F.99) on page 222 was used. Since this is true for
all times t , we conclude that
r B.t; x/ D 0 (1.38)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 36. of 300
also for time-varying fields; this is in fact one of the Maxwell equations. Using
this result and formula (F.89) on page 221, we find that
r .B v / D .v r /B (1.39)
FT
@B
I Z I
emf 2
E.t / D dl E D d x nO dl .B v / (1.41)
C S @t C
@B
I Z
dl .Eemf v B/ D d2x nO (1.42)
C S @t
where Eemf is the field induced in the ‘loop’, i.e. in the moving system. The
application of Stokes’ theorem ‘in reverse’ on equation (1.42) above yields
A
@B
r .Eemf v B/ D (1.43)
@t
An observer in a fixed frame of reference measures the electric field
E D Eemf v B (1.44)
R
and an observer in the moving frame of reference measures the following Lorentz
force on a charge q
Eemf D E C v B (1.46)
@B
r ED (1.47)
@t
is indeed valid even if the ‘loop’ is moving.
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 37. of 300
1.3. Electrodynamics j 15
FT
r B D 0 j (1.48d) in Heaviside-Lorentz units (one of
c 2 @t several natural units)
r E D
In these equations D .t; x/ represents the total, possibly both time and space r B D 0
dependent, electric charge density, with contributions from free as well as in- 1 @B
r EC D0
duced (polarisation) charges. Likewise, j D j.t; x/ represents the total, possibly c @t
both time and space dependent, electric current density, with contributions from 1 @E 1
r B D j
c @t c
conduction currents (motion of free charges) as well as all atomistic (polarisation
and magnetisation) currents. As they stand, the equations therefore incorporate and in Planck units (another set of
natural units)
the classical interaction between all electric charges and currents, free or bound,
A
in the system and are called Maxwell’s microscopic equations. They were first r E D 4
r B D 0
formulated by Lorentz and therefore another name frequently used for them is
@B
the Maxwell-Lorentz equations and the name we shall use. Together with the r EC
@t
D0
appropriate constitutive relations that relate and j to the fields, and the initial @E
r B D 4j
and boundary conditions pertinent to the physical situation at hand, they form a @t
R
system of well-posed partial differential equations that completely determine E
18 JULIAN SEYMOUR
and B. S C H W I N G E R (1918–1994)
once put it:
‘. . . there are strong theor-
etical reasons to believe
scalar and two coupled vectorial partial differential equations (SI units):
e
r E D (1.49a)
"0
r B D 0 m (1.49b)
@B
r EC D 0 j m (1.49c)
@t
1 @E
r B D 0 j e (1.49d)
c 2 @t
We shall call these equations Dirac’s symmetrised Maxwell equations or the
electromagnetodynamic equations.
Taking the divergence of (1.49c), we find that
FT
@
r .r E/ D .r B/ 0 r j m 0 (1.50)
@t
where we used the fact that the divergence of a curl always vanishes. Using
(1.49b) to rewrite this relation, we obtain the equation of continuity for magnetic
charge
@m
C r jm D 0 (1.51)
@t
which has the same form as that for the electric charges (electric monopoles) and
A
currents, equation (1.23) on page 10.
1.4 Examples
R
E X A M P L E 1.1 BFaraday’s law derived from the assumed conservation of magnetic charge
ists and is indestructible in the same way that electric charge exists and is indestructible.
In other words, we postulate that there exists an equation of continuity for magnetic charges:
@m .t; x/
C r j m .t; x/ D 0 (1.52)
@t
Use this postulate and Dirac’s symmetrised form of Maxwell’s equations to derive Faraday’s
law.
The assumption of the existence of magnetic charges suggests a Coulomb-like law for mag-
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 39. of 300
1.4. Examples j 17
netic fields:
x0
0 0 1
Z Z
x
Bstat .x/ D d3x 0 m .x0 / D d3x 0 m .x0 /r
4 V0 jx x0 j3 4 V0 jx x0 j
(1.53)
0 m .x0 /
Z
D r d3x 0
4 V 0 jx x0 j
[cf. equation (1.7) on page 4 for Estat ] and, if magnetic currents exist, a Biot-Savart-like
law for electric fields [cf. equation (1.16) on page 8 for Bstat ]:
x0
0 0 1
Z Z
stat 3 0 m 0 x 3 0 m 0
E .x/ D d x j .x / D d x j .x / r
4 V0 jx 0
xj 3 4 V0 jx x0 j
0 m 0
3 0 j .x /
Z
D r dx
4 V0 jx x0 j
(1.54)
r Estat .x/ D
FT
Taking the curl of the latter and using formula (F.128) on page 224
0 j m .x/
0
4
Z
V0
d3x 0 Œr 0 j m .x0 /r 0
assuming that j m falls off sufficiently fast at large distances. Stationarity means that
r j m D 0 so the last integral vanishes and we can conclude that
jx
1
x0 j
(1.55)
(1.56)
A
It is intriguing to note that if we assume that formula (1.54) above is valid also for time-
varying magnetic currents, then, with the use of the representation of the Dirac delta func-
tion, equation (F.116) on page 222, the equation of continuity for magnetic charge, equation
(1.51) on the preceding page, and the assumption of the generalisation of equation (1.53)
above to time-dependent magnetic charge distributions, we obtain, at least formally,
R
@
r E.t; x/ D 0 j m .t; x/ B.t; x/ (1.57)
@t
[cf. equation (1.25) on page 10] which we recognise as equation (1.49c) on the facing page.
A transformation of this electromagnetodynamic result by rotating into the ‘electric realm’
of charge space, thereby letting j m tend to zero, yields the electrodynamic equation (1.48c)
on page 15, i.e. the Faraday law in the ordinary Maxwell equations. This process would also
D
1.5 Bibliography
[1] T. W. BARRETT AND D. M. G RIMES, Advanced Electromagnetism. Founda-
tions, Theory and Applications, World Scientific Publishing Co., Singapore, 1995,
ISBN 981-02-2095-2.
[4] E. H ALLÉN, Electromagnetic Theory, Chapman & Hall, Ltd., London, 1962.
[5] K. H UANG, Fundamental Forces of Nature. The Story of Gauge Fields, World Sci-
entific Publishing Co. Pte. Ltd, New Jersey, London, Singapore, Beijing, Shanghai,
FT
Hong Kong, Taipei, and Chennai, 2007, ISBN 13-978-981-250-654-4 (pbk).
[6] J. D. JACKSON, Classical Electrodynamics, third ed., John Wiley & Sons, Inc.,
New York, NY . . . , 1999, ISBN 0-471-30932-X.
[7] L. D. L ANDAU AND E. M. L IFSHITZ, The Classical Theory of Fields, fourth re-
vised English ed., vol. 2 of Course of Theoretical Physics, Pergamon Press, Ltd.,
Oxford . . . , 1975, ISBN 0-08-025072-6.
[8] F. E. L OW, Classical Field Theory, John Wiley & Sons, Inc., New York, NY . . . ,
1997, ISBN 0-471-59551-9.
A
[9] J. C. M AXWELL, A Treatise on Electricity and Magnetism, third ed., vol. 1, Dover
Publications, Inc., New York, NY, 1954, ISBN 0-486-60636-8.
[10] J. C. M AXWELL, A Treatise on Electricity and Magnetism, third ed., vol. 2, Dover
Publications, Inc., New York, NY, 1954, ISBN 0-486-60637-8.
[13] F. ROHRLICH, Classical Charged Particles, third ed., World Scientific Publishing
D
Co. Pte. Ltd., New Jersey, London, Singapore, . . . , 2007, ISBN 0-201-48300-9.
[15] J. VANDERLINDE, Classical Electromagnetic Theory, John Wiley & Sons, Inc., New
York, Chichester, Brisbane, Toronto, and Singapore, 1993, ISBN 0-471-57269-1.
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 41. of 300
As a first step in the study of the dynamical properties of the classical elec-
tromagnetic field, we shall in this chapter, as an alternative to the first-order
Maxwell-Lorentz equations, derive a set of second-order differential equations
for the fields E and B. It turns out that these second-order equations are wave
FT
equations for the field vectors E and B, indicating that electromagnetic vector
wave modes are very natural and common manifestations of classical electro-
dynamics.1 1 In 1864, in a lecture at the
But before deriving these alternatives to the Maxwell-Lorentz equations, we Royal Society of London, J A M E S
C L E R K M A X W E L L (1831–1879)
shall discuss the mathematical techniques of making use of complex variables to himself said:
represent physical observables in order to simplify the mathematical treatment. ‘We have strong reason to
In this chapter we will also describe how to make use of the single spectral conclude that light itself
— including radiant heat
component (Fourier component) technique, which simplifies the algebra, at the and other radiation, if any
same time as it clarifies the physical content. — is an electromagnetic
disturbance in the form of
A
waves propagated through
the electro-magnetic field
according to electro-
In chapter 1 we described the historical route which led to the formulation of the
R
microscopic Maxwell equations. From now on we shall consider these equations
as postulates, i.e. as the axiomatic foundation of classical electrodynamics.2 As 2F R I T Z R O H R L I C H writes in
such, these equations postulate, in scalar and vector differential equation form, Classical Charged Particles that
the behaviour in time t 2 R and in space x 2 R3 of the relation between the ‘A physical theory, in
the narrow sense of the
electric and magnetic fields E.t; x/ 2 R3 and B.t; x/ 2 R3 , respectively, and word, is a logical structure
the charge density .t; x/ 2 R and current density j.t; x/ 2 R3 [cf. equations
D
19
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 42. of 300
and are not to be viewed as equations that only describe how the fields E and B
are generated by and j, but also the other way around.
We reiterate that in these equations .t; x/ and j.t; x/ are the total charge
and current densities, respectively. Hence, these equations are considered mi-
croscopic in the sense that all charges and currents, including the intrinsic ones
in matter, such as bound charges in atoms and molecules as well as magnetisa-
tion currents in magnetic material, are included, but macroscopic in the sense
that quantum effects are neglected. Despite the fact that the charge and cur-
rent densities may not only be considered as the sources of the fields, but may
equally well be considered being generated by the fields, we shall follow the
convention and refer to them as the source terms of the microscopic Maxwell
equations. Analogously, the two equations where they appear will be referred to
as the Maxwell-Lorentz source equations.
FT
I we allow for magnetic charge and current densities m and j m , respect-
ively, in addition to electric charge and current densities e and j e j,
we will have to replace the Maxwell-Lorentz equations by Dirac’s symmetrised
equations
e
r E D (2.2a)
"0
m
r B D 0 m D (2.2b)
c 2 "0
@B
A
r EC D 0 j m (2.2c)
@t
1 @E
r B D 0 j e (2.2d)
c 2 @t
and consider them to be the postulates of electromagnetodynamics.
R
mathematical variables that represent the fields, the charge and current densities,
and other physical quantities be analytically continued into the complex domain.
3 A physical observable is some- However, when we use such a complex notation we must be very careful how
thing that can, one way or another, to interpret the results derived within this notation. This is because every phys-
be ultimately reduced to an input
to the human sensory system. In ical observable is, by definition, real-valued.3 Consequently, the mathematical
other words, physical observables expression for the observable under consideration must also be real-valued to be
quantify (our perception of) the
physical reality and as such they physically meaningful.
must, of course, be described by If a physical scalar variable, or a component of a physical vector or tensor,
real-valued quantities.
is represented mathematically by the complex-valued number , i.e. if 2 C,
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 43. of 300
FT
imaginary part corresponds to a
physical observable.
1
i!t i!t
C a0 .x/ ei!t
˚
Re fag D Re a0 .x/ e D Œa0 .x/ e (2.4a)
2
and
1
i!t i!t
C b0 .x/ ei!t
˚
Re fbg D Re b0 .x/ e D Œb0 .x/ e (2.4b)
2
1 1
D Œa0 .x/ e i!t C a0 .x/ ei!t ı Œb0 .x/ e i!t C b0 .x/ ei!t
2 2
1
D a0 ı b0 C a0 ı b0 C a0 ı b0 e 2i!t C a0 ı b0 e2i!t
4
1 1 ˚ (2.5)
D Re fa0 ı b0 g C Re a0 ı b0 e 2i!t
2 2
1 ˚ 1 ˚
i!t
ı b0 ei!t C Re a0 e i!t ı b0 e i!t
D Re a0 e
2 2
1 1
D Re fa.t; x/ ı b .t; x/g C Re fa.t; x/ ı b.t; x/g
2 2
In physics, we are often forced to measure the temporal average (cycle aver-
FT
age) of a physical observable. We use the notation h i t for such an average and
find that the average of the product of the two physical quantities represented by
a and b can be expressed as
1 1
ha ı bi t hRe fag ı Re fbgi t D Re fa ı b g D Re fa ı bg
2 2 (2.6)
1 1
D Re fa0 ı b0 g D Re fa0 ı b0 g
2 2
This is because the oscillating function exp. 2i!t/ in equation (2.5) above van-
ishes when averaged in time over a complete cycle 2=! (or over infinitely many
A
cycles), and, therefore, ha.t; x/ ı b.t; x/i t gives no contribution.
r G D (2.8a)
"0
i @G
r G D ic0 j (2.8b)
c @t
In regions where D 0 and j D 0 these equations reduce to
r G D 0 (2.9a)
i @G
r G D (2.9b)
c @t
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 45. of 300
yD
p i}r (2.10)
can be rewritten
yG D 0
p (2.11a)
@G
i} D ciy
pG (2.11b)
@t
The first equation is the transversality condition p D }k ? G where we anti-
cipate the usual quantal relation between the field momentum p and the wave
vector k,5 whereas the second equation describes the dynamics. 5 The scalar quantity } D h=.2/
Using formula (F.105) on page 222, we can rewrite equation (2.11b) above is the reduced Planck constant
FT
where the Planck constant proper
as h 6:62606957 10 34 Js.
@G
i} y/G
D c.S p (2.12)
@t
where
0 1 0 1 0 1
0 0 0 0 0 i 0 i 0
S D @0 0 O C O
iA 1 @ 0 0 0A 2 @ i
x x C 0 0A x O3 (2.13)
B C B C B C
0 i 0 i 0 0 0 0 0
FT
ors. If we apply the curl vector operator (r ) to both sides of the two Maxwell-
Lorentz vector equations (2.1c) and (2.1d) on page 19, assuming that the fields
vary in such a regular way that temporal and spatial differentiation commute, we
obtain the second-order differential equations
@B @
r .r E/ C r D r .r E/ C .r B/ D 0 (2.16a)
@t @t
1 @
r .r B/ .r E/ D 0 r j (2.16b)
c 2 @t
A
As they stand, these second-order partial differential equations still appear to be
coupled. However, by using the Maxwell-Lorentz equations (2.1) on page 19
once again we can formally decouple them into
1 @2 E @j
C r .r E/ D 0 (2.17a)
c 2 @t 2 @t
R
2
1 @ B
C r .r B/ D 0 r j (2.17b)
c 2 @t 2
If we use the operator triple product ‘bac-cab’ formula (F.96) on page 222, which
gives
r 2E
D
r .r E/ D r .r E/ (2.18)
when applied to E and similarly to B, Gauss’s law equation (2.1a) on page 19,
and then rearrange the terms, we obtain the two inhomogeneous vector wave
equations
1 @2 E r @j
r 2 E D 2 E D 0 (2.19a)
c 2 @t 2 "0 @t
1 @2 B
r 2 B D 2 B D 0 r j (2.19b)
c 2 @t 2
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 47. of 300
FT
1 @2 B
2 B D 2 2 r 2B D 0 (2.20b)
c @t
These equations describe how the fields that were generated in the source region,
propagate as vector waves through free space. Once these waves impinge upon
another region that can sustain charges and/or currents for a long enough time,
such as a receiving antenna or other electromagnetic sensors, the fields inter-
act with the charges and the currents in this second region in accordance with
equations (2.19) on the facing page.
A
2.3.1 The time-independent wave equations for E and B
Often one can assume that the temporal dependency of E and B and of the
sources and j is well-behaved enough that it can be represented by the sum
of a finite number N of temporal spectral components (temporal Fourier com-
R
ponents), or, in other words, in the form of a temporal Fourier series. In such
situations it is sufficient to study the properties of one arbitrary member of this
set of spectral components f!n W n D 1; 2; 3; : : : ; N g, i.e.
i!n t i!n t
˚
E.t; x/ D En .x/ cos.!n t / D En .x/Re e En .x/e (2.21a)
D
i!n t i!n t
˚
B.t; x/ D Bn .x/ cos.!n t / D Bn .x/Re e Bn .x/e (2.21b)
FT
and similarly for B. Solving this equation, multiplying the solution obtained
by exp. i!n t / and summing over all N such Fourier (spectral) components
with frequencies !n ; n D 1; 2; 3; : : : ; N present in the sources, and hence in the
fields, the complete solution of the original wave equation is obtained. This is a
consequence of the superposition principle which is valid as long as nonlinear
effects can be neglected.
In the limit of very many frequency components, the Fourier sum goes over
into a Fourier integral . To illustrate this generic case, let us introduce the Fourier
transform of E.t; x/
A
Z 1
1
E! .x/ D dt E.t; x/ ei!t (2.24a)
2 1
and the corresponding inverse Fourier transform
Z 1
E.t; x/ D d! E! .x/ e i!t (2.24b)
R
1
3
where the amplitude E! .x/ 2 C is a continuous function of (angular) fre-
quency ! 2 R and of x 2 R3 .
We see that the Fourier transform of @E.t; x/=@t becomes
Z 1
1 @E.t; x/ i!t
dt e
D
2 1 @t
Z 1
1 1 1
D E.t; x/ ei!t 1 i! dt E.t; x/ ei!t (2.25)
2 „ ƒ‚ … 2 1
D0 since E!0; t!˙1
D i! E! .x/
and that, consequently,
1
@2 E.t; x/ i!t
1
Z
dt e D ! 2 E! .x/ (2.26)
2 1 @t 2
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 49. of 300
Fourier transforming equation (2.19a) on page 24, and using formulæ (2.25)
and (2.26) on the facing page, we thus obtain
!2 r !
r 2 E! C 2
E! C i!0 j! D (2.27)
c "0
FT
need for the formal forward and inverse Fourier transform technique.
What was said above in general terms about temporal spectral components is
true also for spatial spectral components (spatial Fourier components) only that
we must use a three-dimensional Fourier representation
Z 1
1
Ek .t / D d3x E.t; x/ e ik x (2.28a)
.2/3 1
Z 1
E.t; x/ D d3k Ek .t/ eik x (2.28b)
1
A
Since we always assume that the real part shall be taken (if necessary), we
can pick any pair of the spatial amplitudes in equations (2.21a) and (2.21b) on
page 25, denote the members of this pair by E0 and B0 , respectively, and then
represent them as the Fourier modes
Now, since
@ i!t i!t
e D i!e (2.30a)
@t
and
@ ikj xj
r eik x D x
Oi e D iOxi kj ıij eikj xj D iOxi ki eikj xj D ikeik x (2.30b)
@xi
we see that for each spectral component in equations (2.29) above, temporal
and spatial differential operators turn into algebraic operations according to the
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 50. of 300
following scheme:
@
7! i! (2.31a)
@t
r 7! ik (2.31b)
r 7! ik (2.31c)
r 7! ik (2.31d)
We note that
r E D ik E D ik Ek (2.32a)
r E D ik E D ik E? (2.32b)
r B D ik B D ik Bk (2.32c)
FT
r B D ik B D ik B? (2.32d)
r D ik (2.32e)
r j D ik j D ik jk (2.32f)
r j D ik j D ik j? (2.32g)
Hence, with respect to the wave vector k, the r operator projects out the
spatially longitudinal component, and the r operator projects out the spatially
transverse component of the field vector in question. Put in another way,
A
r E? D 0 (2.33a)
r Ek D 0 (2.33b)
and so on for the other observables. This can be viewed as a instance in k space
of the Helmholtz’s theorem, discussed in subsection M.3.7 on page 249, saying
R
that if E falls off suitably rapidly at infinity, it can be written as the sum of a
rotational part Erotat and an irrotational part Eirrot :
where
r Brotat D 0 (2.38a)
irrot
r B D0 (2.38b)
B? D Brotat (2.39)
irrot
Bk D B (2.40)
FT
As we see from equations (2.31) on the preceding page, the Fourier transform
of a function of time t is a function of angular frequency !, and the Fourier
transform of a function of the position vector x is a function of the wave vector
k. One says that ! is a reciprocal space to t and that k spans a space that is
reciprocal to x. In the reciprocal ! and k space the Maxwell-Lorentz equations
are
ik Ek D (2.41a)
"0
ik Bk D 0 (2.41b)
A
ik E? i!B D 0 (2.41c)
!
ik B? C i 2 E D 0 j (2.41d)
c
Applying the Helmholtz decomposition, the Maxwell-Lorentz equations be-
R
come
r Eirrot D (2.42a)
"0
r Birrot D0 (2.42b)
@B
r Erotat C D0 (2.42c)
D
@t
1 @E
r Brotat D 0 j (2.42d)
c 2 @t
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 52. of 300
2.4 Examples
FT
Since ‘length’ is a scalar quantity that is invariant under rotations, we find that
E2 c 2 B 2 D Const (2.44a)
E B D Const (2.44b)
2. The inner (scalar) product defined as G scalar multiplied with the complex conjugate of
itself
is also an invariant scalar quantity. As we shall see in chapter chapter 4, this quantity is
proportional to the electromagnetic field energy density.
A
3. As with any 3D vector, the cross product of G with itself vanishes:
G G D .E C icB/ .E C icB/
DEE c 2 B B C ic.E B/ C ic.B E/ (2.46)
D 0 C 0 C ic.E B/ ic.E B/ D 0
R
4. The cross product of G with the complex conjugate of itself does, however, not vanish.
Instead it is
G G D .E C icB/ .E icB/
2 (2.47)
DEECc BB ic.E B/ C ic.B E/
D0C0 ic.E B/ ic.E B/ D 2ic.E B/
D
which is proportional to the electromagnetic energy flux (the so called Poynting vector
or the electromagnetic linear momentum density), to be introduced in chapter chapter 4.
5. The dyadic product of G with itself is
2.4. Examples j 31
7. The dyadic product of G with its own complex conjugate from the right is
FT
ematical idealisation. In reality,
magnetic field vectors are restricted to a two-dimensional plane that is perpendicular to the plane waves appear as building
propagation direction. Let us choose this plane to be the x1 x2 plane and the propagation blocks of wave packets, i.e. super-
positions of a (possibly infinite)
vector (wave vector) k to be along the x3 axis: k D k x O 3 . A generic temporal Fourier number of individual plane waves
mode of the electric field vector E with (angular) frequency ! is therefore described by the with different properties (frequen-
real-valued expression cies, directions,. . . ). E.g. a radio
beam from a transmitting antenna
E.t; x/ D E1 cos.!t O 1 C E2 cos.!t
kx3 C ı1 / x O2
kx3 C ı2 / x (2.53) is a superposition (Fourier sum or
integral) of many plane waves with
where the amplitudes Ei and phases ıi , can take any value. In complex notation we can slightly different angles relative to
write this as a fixed, given axis or a plane.
!t / !t /
E.t; x/ D E1 eiı1 ei.kx3 O 1 C E2 eiı2 ei.kx3
x O2
x
A
D E1 eiı1 x O 2 ei.kx3 !t /
O 1 C E2 eiı2 x
(2.54)
O 2 ei.kx3 !tCı1 /
O 1 C E2 eiı0 x
D E1 x
where ı0 D ı2 ı1 . When this phase difference ı0 vanishes, the electric field oscillates
along a line directed at an angle arctan .E2 =E1 / relative to the x1 axis. This is called linear
wave polarisation. When ı0 ¤ 0 the wave is in general in a state of elliptical wave polar-
R
isation. Later, in subsection 4.2.5, we will show that wave polarisation is a manifestation of
the fact that the electromagnetic field carries angular momentum.
For the special cases ı0 D ˙=2 and E1 D E2 D E0 the wave can, in complex notation,
be described as
As discussed in example M.1 on page 252, this shows that the field vector E rotates around
the x3 axis as it propagates along this axis. This rotation is called circular wave polarisation.
For ı1 D 0 in equation (2.55), the linear superpostion 1=2.EC C E / represents a wave
that is linearly polarised along xO 1 and for ı1 D =2 the superposition 1=2.EC E /
O 2.
represents a wave that is linearly polarised along x
The helical base vectors
1
hO ˙ D p xO 1 ˙ iOx2 / (2.56)
2
which are fixed unit vectors, allow us to write equation (2.55) above
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 54. of 300
FT
Taking the real and imaginary parts of this equation, assuming that E; B 2 R3 , we recover
the wave equations (2.17) on page 24, as expected.
2.5 Bibliography
[16] J. D. JACKSON, Classical Electrodynamics, third ed., John Wiley & Sons, Inc.,
A
New York, NY . . . , 1999, ISBN 0-471-30932-X.
ELECTROMAGNETIC POTENTIALS
AND GAUGES
As described in chapter 1 the concepts of electric and magnetic fields were in-
troduced such that they are intimately related to the mechanical forces between
(static) charges and currents given by Coulomb’s law and Ampère’s law, respect-
ively. Just as in mechanics, it turns out that in electrodynamics it is often more
FT
convenient to express the theory in terms of potentials rather then in terms of
the electric and magnetic fields (Coulomb and Ampère forces) themselves. This
is particularly true for problems related to radiation and relativity. As we shall
see in chapter 7, the potentials play a central rôle in the formulation of relativist-
ically covariant electromagnetism. And at the quantum level, electrodynamics
is almost exclusively formulated in terms of potentials rather than electric and
magnetic fields.
In this chapter we introduce and study the properties of such potentials and
find that they exhibit some remarkable properties that elucidate the fundamental
A
aspects of electromagnetism, lead naturally to the special theory of relativity,
and pave the way for gauge field theories.
R
3.1 The electrostatic scalar potential
As we saw in equation (1.9b) on page 5, the time-independent electric (electro-
static) field Estat .x/ is irrotational. According to formula (F.100) on page 222 we
may therefore express it in terms of the gradient of a scalar field. If we denote
D
Taking the divergence of this and using equation (1.9a) on page 5, we obtain
Poisson’s equation
.x/
r 2 ˆstat .x/ D r Estat .x/ D (3.2)
"0
33
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 56. of 300
1 .x0 /
Z
ˆstat .x/ D d3x 0 (3.3)
4"0 V 0 jx x0 j
where the integration is taken over all source points x0 at which the charge dens-
ity .x0 / is non-zero. The scalar function ˆstat .x/ is called the electrostatic scalar
potential .
FT
Let us consider the equations of magnetostatics, equations (1.22) on page 9. Ac-
cording to formula (F.99) on page 222 any vector field a has the property that
r .r a/ 0 and in the derivation of equation (1.18) on page 8 in magneto-
statics we found that r Bstat .x/ D 0. We therefore realise that we can always
write
where Astat .x/ is called the magnetostatic vector potential . In the magnetostatic
A
case, we may start from Biot-Savart’s law as expressed by equation (1.16) on
page 8. Identifying this expression with equation (3.4) above allows us to define
the static vector potential as
0 j.x0 /
Z
stat
A .x/ D d3x 0 (3.5)
4 V 0 jx x0 j
R
From equations (3.1) and (3.4) on pages 33–34 we conclude that if we trans-
form equations (3.3) and (3.5) in the following way
0
ˆstat .x/ 7! ˆstat .x/ D ˆstat .x/ C ˛.x/ (3.6a)
stat 0
Astat .x/ 7! A .x/ D Astat .x/ C a.x/ (3.6b)
D
r ˛.x/ D 0 (3.7a)
r a.x/ D 0 (3.7b)
scalar field. In other words, the fields are unaffected by the transformation (3.6)
if
˛.x/ D Const (3.8a)
a.x/ D r ˇ.x/ (3.8b)
where ˇ is an arbitrary, at least twice continuously differentiable function of x.
FT
the pertinent fields E.t; x/ and B.t; x/, as described by the Maxwell-Lorentz
equations (2.1) on page 19. In other words, let us study the electrodynamic
potentials ˆ.t; x/ and A.t; x/.
According to the non-source Maxwell-Lorentz equation (2.1b), the magnetic
field B.t; x/ is divergence-free also in electrodynamics (if magnetic charges are
not included). Because of this divergence-free nature of the time- and space-
dependent magnetic field, we can express it as the curl of an electromagnetic
vector potential :
B.t; x/ D r A.t; x/ (3.9)
A
Inserting this expression into the other non-source Maxwell-Lorentz equation
(2.1c) on page 19, we obtain
@ @
r E.t; x/ D Œr A.t; x/ D r A.t; x/ (3.10)
@t @t
R
or, rearranging the terms,
@
r E.t; x/ C A.t; x/ D 0 (3.11)
@t
As before we utilise the vanishing curl of a vector expression to write this
vector expression as the gradient of a scalar function. If, in analogy with the
D
and B.t; x/ from formula (3.9) on the previous page. Hence, it is a matter of
convention (or taste) whether we want to express the laws of electrodynamics in
terms of the potentials ˆ.t; x/ and A.t; x/, or in terms of the fields E.t; x/ and
B.t; x/. However, there is an important difference between the two approaches:
in classical electrodynamics the only directly observable quantities are the fields
themselves (and quantities derived from them) and not the potentials. On the
other hand, the treatment becomes significantly simpler if we use the potentials
in our calculations and then, at the final stage, use equation (3.9) on the preceding
page and equation (3.13) on the previous page to calculate the fields or physical
quantities expressed in the fields. This is the strategy we shall follow.
FT
3.4 Gauge conditions
Inserting (3.13) and (3.9) on the preceding page into Maxwell’s equations (2.1)
on page 19, we obtain, after some simple algebra and the use of equation (1.12)
on page 6 and formula (F.96) on page 222, the equations
@ .t; x/
r 2ˆ C .r A/ D (3.14a)
@t "0
1 @2 A 1 @ˆ
r 2 A C r .r A/ D 0 j.t; x/ r (3.14b)
A
c 2 @t 2 c 2 @t
Subtracting .1=c 2 /@2 ˆ=@t 2 from both sides of the first equation and rearranging,
the above two equations turn into the following general inhomogeneous wave
equations
.t; x/ @ 1 @ˆ
R
2 ˆ D C r A C 2 (3.15a)
"0 @t c @t
2 1 @ˆ
A D 0 j.t; x/ r r A C 2 (3.15b)
c @t
FT
the coupled inhomogeneous wave equations (3.15) on page 36 simplify to the
condition (3.16). It had been dis-
following set of uncoupled inhomogeneous wave equations: covered by the Danish physicist
L U DV I G VA L E N T I N L O R E N Z
.t; x/ (1829-1891) already in 1867. This
2 ˆ D (3.17a) fact has sometimes been over-
"0 looked in the literature and the
2 A D 0 j.t; x/ (3.17b) condition was earlier referred to as
the Lorentz gauge condition. Prior
to that, B E R N H A R D R I E M A N N ,
Each of these four scalar equations is an inhomogeneous wave equation of the had discussed this condition in a
following form: lecture in 1858.
2 ‰.t; x/ D s.t; x/
A
(3.18)
exists, and that the same is true for the generic potential component ‰:
Z 1
‰.t; x/ D d! ‰! .x/ e i!t (3.20a)
1
Z 1
1
‰! .x/ D dt ‰.t; x/ ei!t (3.20b)
2 1
Inserting the Fourier representations (3.19) and (3.20) into equation (3.18) above,
and using the vacuum dispersion relation for electromagnetic waves relating the
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 60. of 300
angular frequency !, the speed of light c, and the wave number k D .2/=
where is the vacuum wavelength ,
! D ck (3.21)
the generic 3D inhomogeneous wave equation (3.18) on the previous page, turns
into
FT
front is formed by a linear superposition of the individual weighted wavelets
from each of the point sources on the old wave front. The solution of (3.22) can
therefore be expressed as a weighted sum of solutions of an equation where the
source term has been replaced by a single point source
and the solution of equation (3.22) above which corresponds to the frequency !
is given by the weighted superposition
A
Z
‰! .x/ D d3x 0 s! .x0 /G.x; x0 / (3.24)
V0
(plus boundary conditions) where s! .x0 / is the wavelet amplitude at the source
point x0 . The function G.x; x0 / is called the Green function or the propagator.
Due to translational invariance in space, G.x; x0 / D G.x x0 /. Furthermore,
R
in equation (3.23) above, the Dirac generalised function ı.x x0 /, which repres-
ents the point source, depends only on x x0 and there is no angular dependence
in the equation. Hence, the solution can only be dependent on r D jx x0 j and
not on the direction of x x0 . If we interpret r as the radial coordinate in a
spherically polar coordinate system, and recall the expression for the Laplace
D
d2
.rG/ C k 2 .rG/ D rı.r/ (3.25)
dr 2
Away from r D jx x0 j D 0, i.e. away from the source point x0 , this equation
takes the form
d2
.rG/ C k 2 .rG/ D 0 (3.26)
dr 2
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 61. of 300
the volume integrated equation (3.23) on the preceding page can be approxim-
FT
ated by
1
Z
3 0 2
.CC C C / dx r
V0 jx x0 j
(3.29)
1
Z Z
2 3 0 3 0 ˇ 0ˇ
ˇ ˇ
C k .CC C C / dx D d x ı. x x /
V0 jx x0 j V0
In virtue of the fact that the volume element d3x 0 in spherical polar coordinates
is proportional to r 2 D jx x0 j2 [see formula (F.19) on page 218], the second
integral vanishes when jx x0 j ! 0. Furthermore, from equation (F.116)
on page 222, we find that the integrand in the first integral can be written as
A
4ı.jx x0 j/ and, hence, that the two constants C ˙ must fulfil the condition
1
CC C C D (3.30)
4
Now that we have determined the relation between CC and C , we insert the
R
general solution equation (3.27) above into equation (3.24) on the facing page
and obtain the general solution in the ! domain:
0 0
eikjx x j e ikjx x j
Z Z
3 0 0
‰! .x/ D CC d x s! .x / CC d3x 0 s! .x0 / (3.31)
V0 jx x0 j V0 jx x0 j
D
In order to find the solution in the t domain, we take the inverse Fourier trans-
form of this by inserting the above expression for ‰! .x/ into equation (3.20) on
page 37:
h 0
i
Z Z 1 exp i! t kjx! x j
‰.t; x/ D CC d3x 0 d! s! .x0 /
V0 1 jx x0 j
h 0
i (3.32)
Z Z 1 exp i! t C kjx! x j
CC d3x 0 d! s! .x0 /
V0 1 jx x0 j
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 62. of 300
0 0
If we introduce the retarded time tret and the advanced time tadv in the following
way [using the fact that k=! D 1=c in free space, according to formula (3.21)
on page 38]:
k ˇx x0 .tret
ˇ 0 ˇ
ˇ ˇ
ˇx x0 .t 0 /ˇ
ˇ
0 0
ˇ 0
ˇ / ret
tret D tret .t; ˇx x ˇ/ D t Dt (3.33a)
ˇ ! 0 0 ˇ ˇ c 0 0 ˇ
0 0 k ˇx x .tadv /ˇ ˇx x .t /ˇ
.t; ˇx x0 ˇ/ D t C adv
ˇ ˇ
tadv D tadv DtC (3.33b)
! c
and use equation (3.19) on page 37, we obtain
0 0 0
f .tret ; x0 / 3 0 f .tadv ; x /
Z Z
‰.t; x/ D CC d3x 0 C C d x (3.34)
V0 jx x0 j V0 jx x0 j
This is a solution to the generic inhomogeneous wave equation for the potential
components equation (3.18) on page 37. We note that the solution at time t at the
FT
field point x is dependent on the behaviour at other times t 0 of the source at x0
and that both retarded and advanced t 0 are mathematically acceptable solutions.
However, if we assume that causality requires that the potential at .t; x/ is set
0
up by the source at an earlier time, i.e. at .tret ; x0 /, we must in equation (3.34)
above set C D 0 and therefore, according to equation (3.30) on the preceding
2 In fact, inspired by ideas put page, CC D 1=.4/.2
forward by PAU L A D R I E N From the above discussion about the solution of the inhomogeneous wave
M AU R I C E D I R AC (1902–
1984), J O H N A R C H I BA L D equations in the Lorenz-Lorentz gauge we conclude that, if we discard the ad-
W H E E L E R (1911–2008) and vanced potentials, the electrodynamic potentials in free space can be written
A
RICHARD PHILLIPS FEYNMAN
(1918–1988) derived, in 1945, 0 0 0
1 3 0 tret ; x .tret /
Z
a consistent electrodynamics ˆ.t; x/ D dx 0 (3.35a)
based on both the retarded and the 4"0 V 0 jx.t/ x0 .tret /j
0 0 0
advanced potentials.
1 3 0 j tret ; x .tret /
Z
A.t; x/ D d x 0 (3.35b)
4"0 c 2 V 0 jx.t/ x0 .tret /j
R
These retarded potentials were obtained as solutions to the Lorenz-Lorentz in-
homogeneous wave equations (3.17). The expressions (3.35) are therefore valid
3 In 1897, T U L L I O L E V I - C I V I TA only in the Lorenz-Lorentz gauge.3 scalar first-order, coupled, partial differen-
(1873–1941) showed that the tial equations. In other gauges (other choices of r A) the expressions for the
retarded Lorenz-Lorentz potentials
solves equations (3.14). potentials are analytically different but will, of course, yield the very same phys-
D
ical fields E and B as the expressions (3.35) do. As they stand, we shall use
expressions (3.35) quite frequently in the sequel.
1 @2 A 1 @ˆ
r 2A D 0 j.t; x/ C r (3.36b)
c 2 @t 2 c 2 @t
The first of these two is the time-dependent Poisson’s equation which, in analogy
with equation (3.3) on page 34, has the solution
1 .t; x0 .t/
Z
ˆ.t; x/ D d3x 0 (3.37)
4"0 V 0 jx.t/ x0 .t/j
We see that in the scalar potential expression, the charge density source is evalu-
ated at time t . Hence, the scalar potential does not exhibit any retardation, which
means that the effect of the charge shows up in the potential instantaneously, as
if the propagation speed were infinite. This is not in disagreement with the laws
of nature since in classical physics potentials are not physical observables.
FT
Since in the Coulomb gauge the scalar potential ˆ does not suffer any re-
tardation (or advancement) but the fields E and B themselves must be physical
and therefore must exhibit retardation effects, all retardation must occur in the
vector potential A, i.e. the solution of the inhomogeneous wave equation (3.36b)
above. As we see, the last term in the RHS of this equation contains the scalar
potential ˆ, which, according to equation (3.37) above, in turn depends on the
charge density . The continuity equation (1.23) on page 10 provides a relation
between and j on which we can apply Helmholtz decomposition and find that
r j rotat D 0 (3.39a)
irrot
r j D0 (3.39b)
R
[cf. equations (2.36) on page 28]. Then the equation of continuity becomes
@ @
C r j irrot D "0 r 2 ˆ C r j irrot
@t @t (3.40)
@ˆ irrot
Dr "0 r Cj D0
D
@t
@ˆ
"0 r D j irrot (3.42)
@t
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 64. of 300
The inhomogeneous wave equation (3.36b) on the previous page thus becomes
1 @2 A 1 @ˆ
r 2A D 0 j C r D 0 j C 0 j irrot D 0 j rotat (3.43)
c 2 @t 2 c 2 @t
which shows that in Coulomb gauge the source of the vector potential A is the
rotational (transverse) component of the current, j rotat . The irrotational (longit-
udinal) component of the current j irrot does not contribute to the vector potential.
The retarded (particular) solution in Coulomb gauge of equations (3.14) on page
36 is therefore [cf. equations (3.35) on page 40]:
0
1 3 0 t; x .t/
Z
ˆ.t; x/ D dx (3.44a)
4"0 V 0 jx.t/ x0 .t/j
0
; x0 .tret
0
0 j rotat tret /
Z
d3x 0
FT
A.t; x/ D 0 0 (3.44b)
4 V 0 jx.t/ x .tret /j
The Coulomb gauge is also called the transverse gauge or the radiation gauge.
˛ @ˆ c2
A
r A C D 0; ˛D (3.45)
c 2 @t v2
we obtain the Lorenz-Lorentz gauge condition in the limit v D c, i.e. ˛ D 1,
and the Coulomb gauge condition in the limit v ! 1, i.e. ˛ D 0, respectively,
where v is the propagation speed of the scalar potential. Hence, the velocity
4 gauge is a generalisation of both these gauges.4 Inserting equation (3.45) above
R
The value ˛ D 1, corres-
ponding to an imaginary speed into the coupled inhomogeneous wave equations (3.15) on page 36 they become
v D ic, gives the Kirchhoff gauge,
introduced already in 1857 by
G U S TAV R O B E RT K I R C H H O F F
.t; x/ 1 ˛ @ @ˆ
(1824–1884). 2 ˆ D (3.46a)
"0 c2
@t @t
1 ˛ @ˆ
D
2 A D 0 j.t; x/ C r (3.46b)
c2 @t
and still get the same physical result. In fact, the way the electromagnetic scalar
potential ˆ.t; x/ and the vector potential A.t; x/ are related to the physical ob-
servables gives leeway for similar manipulation of them also in electrodynamics.
In analogy with equations (3.6) on page 34 we introduce
By inserting these transformed potentials into equation (3.13) on page 35 for the
electric field E.t; x/ and into equation (3.9) on page 35 for the magnetic field
B.t; x/, we see that the fields will be unaffected provided
@a.t; x/
r ˛.t; x/ C D0 (3.48a)
FT
@t
r a.t; x/ D 0 (3.48b)
@.t; x/
ˆ.t; x/ 7! ˆ0 .t; x/ D ˆ.t; x/ (3.50a)
@t
R
A.t; x/ 7! A0 .t; x/ D A.t; x/ C r .t; x/ (3.50b)
and insert the transformed potentials into equation (3.13) on page 35 for the
electric field and into equation (3.9) on page 35 for the magnetic field, we obtain
the transformed fields
D
@A0
0 0 @ @A @.r /
E D rˆ D rˆ C r
@t @t @t @t
(3.51a)
@A @.r / @.r / @A
D rˆ C D rˆ
@t @t @t @t
B0 D r A0 D r A C r .r / D r A (3.51b)
where, once again equation (F.100) on page 222 was used. This explicit calcu-
lation clearly demonstrates that the fields E and B are unaffected by the gauge
transformation (3.50). The function .t; x/ is called the gauge function.
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 66. of 300
A transformation of the potentials ˆ and A which leaves the fields, and hence
Maxwell’s equations, invariant is called a gauge transformation. Any physical
law that does not change under a gauge transformation is said to be gauge in-
variant. It is only those quantities (expressions) that are gauge invariant that are
observable and therefore have experimental significance. Trivially, the electro-
magnetic fields and the Maxwell-Lorentz equations themselves are gauge invari-
5 A very important extension is ant and electrodynamics is therefore a gauge theory and as such the prototype
the Yang-Mills theory, introduced for all gauge theories.5
in 1954. This theory has had
a profound impact on modern As just shown, the potentials ˆ.t; x/ and A.t; x/ calculated from equations
physics. (3.14) on page 36, with an arbitrary choice of r A, can be gauge transformed
according to (3.50) on the preceding page. If, in particular, we choose r A
according to the Lorenz-Lorentz condition, equation (3.16) on page 37, and ap-
ply the gauge transformation (3.50) on the resulting Lorenz-Lorentz potential
FT
equations (3.17) on page 37, these equations will be transformed into
1 @2 ˆ 1 @2
2 @ 2 .t; x/
r ˆ C r D (3.52a)
c 2 @t 2 @t c 2 @t 2 "0
1 @2 A 1 @2
2 2
r A r r D 0 j.t; x/ (3.52b)
c 2 @t 2 c 2 @t 2
We notice that if we require that the gauge function .t; x/ itself be restricted to
fulfil the wave equation
1 @2
r 2 D 0
A
(3.53)
c 2 @t 2
these transformed Lorenz-Lorentz equations will keep their original form. The
set of potentials which have been gauge transformed according to equation (3.50)
on the previous page with a gauge function .t; x/ restricted to fulfil equation
(3.53) above, or, in other words, those gauge transformed potentials for which
R
the equations (3.17) on page 37 are invariant, comprise the Lorenz-Lorentz
gauge.
The Weyl gauge, also known as the temporal gauge or Hamilton gauge,
defined by ˆ0 D 0.
FT
A
R
D
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 68. of 300
3.6 Examples
1 1
FT
D
jx x0 j j.x x0 / .x0 x0 /j
1 3 3
1 X 1 X X
D C
jx x0 j nŠ
nD1 i1 D1 in D1
1
@n jx x0 j
Œ .xi01 x0i / Œ .xi0n x0in /
@xi1 @xin 1 (3.56)
1
1 X X . 1/n
D C
jx x0 j n1 Šn2 Šn3 Š
nD1 n1 Cn2 Cn3 Dn
ni 0
A
1
@n jx x0 j
n n n .x10 x01 /n1 .x20 x02 /n2 .x30 x03 /n3
@x1 1 @x2 2 @x3 3
Inserting this into the integral in formula (3.55) above, we obtain the expansion
1
"
1 1 . 1/n
Z
d3x 0 .x0 / C
X X
ˆstat .x/ D
4"0 jx x0 j V 0 n1Šn2 Šn3 Š
R
nD1 n1 Cn2 Cn3 Dn
ni 0 (3.57)
1
@n jx Z #
x0 j
n n n d3x 0 .x10 x01 /n1 .x20 x02 /n2 .x30 x03 /n3 .x0 /
@x1 1 @x2 2 @x3 3 V0
Clearly, the first integral in this expansion is nothing but the static net charge
D
Z
qD d3x 0 .x0 / (3.58)
V0
If we introduce the electrostatic dipole moment vector
Z
d.x0 / D d3x 0 .x0 x0 / .x0 / (3.59)
V0
3.6. Examples j 47
1
@ jx x0 j xi x0i
D (3.61)
@xi jx x0 j3
and that
1
@2 jx x0 j 3.xi x0i /.xj x0j / jx x0 j2 ıij
D (3.62)
@xi @xj jx x0 j5
then we can write the first three terms of the expansion of equation (3.55) on the facing page
as
x x0
d
jx x0 j
" ‚ …„ ƒ
1 q 1 xi x0i
ˆstat .x/ D C 2
di
4"0 jx x0 j jx x0 j jx x0 j
#
1 3 xi x0i xj x0j 1
C Q
3 ij 2 jx
ıij C : : : (3.63)
jx x0 j x0 j jx x0 j 2
FT
where Einstein’s summation convention over i and j is implied. We see that at large dis-
tances from a localised charge distribution, the electrostatic potential can, to the lowest
order, be approximated by the (Coulomb) potential from a single point charge q located at
the moment point x0 . We also see that
Z Z Z
d.x0 / D d3x 0 .x0 x0 / .x0 / D d3x 0 x0 .x0 / x0 d3x 0 .x0 /
V0 V0 V0
Z
D d3x 0 x0 .x0 / x0 q (3.64)
V0
from which we conclude that if q ¤ 0, it is always possible to choose the point x0 such that
A
d D 0. We also see that if q D 0, then d is independent of the choice of moment point x0 .
Furthermore, one can show that
1 3.xi x0i /.xj x0j / jx x0 j2 ıij
˛ ıij D0 (3.65)
2 jx x0 j5
where ˛ is an arbitrary constant. Choosing it to be
R
1
Z
ˇ2
d3x 0 ˇx0 x0 ˇ .x0 /
ˇ
˛D (3.66)
3 V0
we can transform Qij into
1 ˇˇ 0
Z
0
ˇ2
Qij D Qij ˛ıij D d3x 0 .xi0 x0i /.xj0 x0j / x x0 ˇ ıij .x0 /
V0 3
D
(3.67)
or
1 ˇˇ 0
Z
ˇ2
Q0 D Q 13 ˛ D d3x 0 .x0 x0 / ˝.x0 x0 / 13 x x0 ˇ .x0 / (3.68)
V0 3
where 13 D x O i is the unit tensor. It follows immediately that Qi0 i D 0 (Einstein summa-
O ix
tion), i.e. that Q0 is traceless. Rotating the coordinate system, it is possible to diagonalise
the tensors Q and Q0 . For any spherical symmetric distribution of charge, all components
of Q0 vanish if the moment point x0 is chosen as the symmetry centre of the distribution.
If the charge distribution .x/ is made up of discrete point charges qn with coordinates xn ,
the definitions above of q; d; Q and Q0 become
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 70. of 300
X
qD qn (3.69a)
n
X
dD qn .xn x0 / (3.69b)
n
X
QD qn .xn x0 / ˝.xn x0 / (3.69c)
n
1
Q0 D x0 j2
X
qn .xn x0 / ˝.xn x0 / 13 jxn (3.69d)
n
3
FT
e
r E D (3.70a)
"0
r B D 0 m (3.70b)
@B
r EC D 0 j m (3.70c)
@t
1 @E
r B D 0 j e (3.70d)
c 2 @t
In this theory, one derives the inhomogeneous wave equations for the usual ‘electric’ scalar
and vector potentials .ˆe ; Ae / and their ‘magnetic’ counterparts .ˆm ; Am / by assuming
A
that the potentials are related to the fields in the following symmetrised form:
@ e
ED r ˆe .t; x/ A .t; x/ r Am (3.71a)
@t
1 1 @ m
BD r ˆm .t; x/ A .t; x/ C r Ae (3.71b)
c2 c 2 @t
R
In the absence of magnetic charges, or, equivalently for ˆm 0 and Am 0, these for-
mulæ reduce to the usual Maxwell theory, formulæ (3.9) and (3.13) on page 35 respectively,
as they should.
Inserting the symmetrised expressions (3.71) above into equations (3.70) above, one obtains
[cf., equations (3.14) on page 36]
@ e .t; x/
D
r 2 ˆe C r Ae D
(3.72a)
@t "0
2 m @ m
m .t; x/
r ˆ C r A D (3.72b)
@t "0
1 @2 Ae e
1 @ˆ
r 2 Ae C r r Ae C 2 D 0 j e .t; x/ (3.72c)
c 2 @t 2 c @t
1 @2 Am 1 @ˆm
2 m m
r A C r r A C 2 D 0 j m .t; x/ (3.72d)
c 2 @t 2 c @t
By choosing the conditions on the divergence of the vector potentials as the generalised
Lorenz-Lorentz condition [cf. equation (3.16) on page 37]
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 71. of 300
3.6. Examples j 49
1 @ e
r Ae C ˆ D0 (3.73a)
c2 @t
1 @ m
r Am C 2 ˆ D0 (3.73b)
c @t
these coupled wave equations simplify to
e .t; x/
2 ˆe D (3.74a)
"0
2 Ae D 0 j e .t; x/ (3.74b)
m .t; x/
2 ˆm D (3.74c)
"0
2 Am D 0 j m .t; x/ (3.74d)
exhibiting, once again, the striking properties of Dirac’s symmetrised Maxwell theory.
FT
BLongitudinal and transverse components in gauge transformations
If we represent the vector potential A.t; x/ in the reciprocal k space as described at the
end of subsection 2.3.1 on page 25, the gauge transformation equation (3.50b) on page 43
becomes
Ak .t/ 7! A0k .t/ D Ak .t/
we can separate it into its longitudinal and transverse components
ikk .t/
(3.76a)
E X A M P L E 3.3
A
(3.76b)
A Helmholtz decomposition [see formula (M.83) on page 250]
A D Arotat C Airrot (3.77)
shows that
rotat
A0 D Arotat (3.78a)
R
0 irrot irrot
A DA C r (3.78b)
Hence, a law (expression) that depends on A only through its transverse/rotational com-
ponent A? D Arotat is gauge invariant, whereas a law that depends on the longitud-
inal/irrotational component Ak D Airrot is in general not gauge invariant and, if so, it
does not represent a physical observable.
D
With the caveat about plane waves mentioned near formulæ (2.36) on page 28, the following
then applies for the electric field and the magnetic field E D E? C Ek and B D B? C Bk ,
respectively:
@A?
E? D (3.79a)
@t
@Ak
Ek D rˆ (3.79b)
@t
B? D r A? (3.79c)
Bk D 0 (3.79d)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 72. of 300
In terms of rotational and irrotational parts, the electric and magnetic fields are given by the
sums E D Erotat C Eirrot and B D Brotat C Birrot , respectively, where the individual terms
are
@Arotat
Erotat D (3.80a)
@t
@Airrot
Eirrot D rˆ (3.80b)
@t
Brotat D r Arotat (3.80c)
Birrot D 0 (3.80d)
FT
rather than the real part of our mathematical variables in order to turn them into quantities
that represent physical observables. In non-relativistic quantum mechanics, the physical
observable probability density is D j j2 , where the wave function 2 C solves the
Schrödinger equation
@
i} y
DH (3.81)
@t
y is the Hamilton operator.
and H
The non-relativistic Hamiltonian for a classical particle with charge q in an electromagnetic
field, described by the scalar potential ˆ and vector potential A, is
A
1
H D .p qA/2 C qˆ (3.82)
2m
where p is the linear momentum. The corresponding quantal Hamilton operator is obtained
y D i}r , referred
from the correspondence principle, viz., by replacing p by the operator p
to as minimal coupling. This gives the Schrödinger equation
@ 1
qA/2
R
i} D . i}r C qˆ (3.83)
@t 2m
The idea is to perform a gauge transformation from the potentials ˆ.t; x/ and A.t; x/ to
new potentials
@.t; x/
ˆ 7! ˆ0 .t; x/ D ˆ.t; x/ C (3.84a)
@t
D
0
A 7! A .t; x/ D A.t; x/ r .t; x/ (3.84b)
and then find a .t; x/, expressed in the gauge function .t; x/, so that the transformed
Schrödinger equation can be written
@.ei / 1
i} D . i}r qA/2 ei C qˆei (3.85)
@t 2m
Under the gauge transformation equation (3.84) above, the Schrödinger equation (3.83)
transforms into
@ 0 1 @
i} D Œ i}r qA C .qr /2 0
C qˆ 0
Cq 0
(3.86)
@t 2m @t
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 73. of 300
3.6. Examples j 51
Now, setting
0
.t; x/ D ei.t;x/ .t; x/ (3.87)
we see that
2 0
i}r qA C .qr /
D Œ i}r qA C .qr / Œ i}r qA C .qr / ei
D Œ i}r qA C .qr /
h i
i}ei .r / i}ei .r i/ qAei C .qr /eı
D Œ i}r qA C .qr / ei Œ i}r i}.r i/ qA C .qr / (3.88)
h i
D i}.r i/ei i}ei r qAei C .qr /ei
FT
Œ i}r i}.r i/ qA C .qr /
i
D e Œ i}r i}.r i/ qA C .qr /2
D ei Œ i}r qA C }.r / C .qr /2
Clearly, the gauge transformed Hamilton operator is unchanged iff }.r / D q.r /, or,
equivalently, iff .t; x/ D q.t; x/=}. This has as a consequence that
@ 0 @ 0 @ 0 @ 0 @ @
i} q D i} C} D i} .ei / C } ei
@t @t @t @t @t @t (3.89)
2 @ i i @ @ i i @
A
Di } e C i}e C} e D e i}
@t @t @t @t
Inserting this into the transformed Schrödinger equation (3.86) on the facing page, we re-
cover the untransformed Schrödinger equation (3.83).
We conclude that under a gauge transformation of the potentials ˆ and A and with minimal
coupling as in equation (3.83) on the preceding page, the wave function changes from
R
to ei where the phase angle is real-valued. Hence, even if the wavefunction is not
invariant, the quantum physical observable j j2 is unaffected by a gauge transformation of
the classical potentials ˆ and A that appear in the Hamilton operator. The fact that .t; x/
is coordinate dependent, means that we are dealing with a local gauge transformation.
2
r /=.2m/ qA j j =m do not change under a gauge transformation of ˆ and A,
the charge density D q j j2 and therefore also the charge V 0 d3x 0 are conserved. In
R
For the gauge transformation given by formulæ (3.84) on the facing page, W O L F G A N G
PAU L I introduced the notation gauge transformation of the second kind whereas he called
a wavefunction phase change, expression (3.87) above, a gauge transformation of the first
kind .
3.7 Bibliography
[19] L. D. FADEEV AND A. A. S LAVNOV, Gauge Fields: Introduction to Quantum
Theory, No. 50 in Frontiers in Physics: A Lecture Note and Reprint Series. Ben-
jamin/Cummings Publishing Company, Inc., Reading, MA . . . , 1980, ISBN 0-8053-
9016-2.
[20] M. G UIDRY, Gauge Field Theories: An Introduction with Applications, John Wiley
& Sons, Inc., New York, NY . . . , 1991, ISBN 0-471-63117-5.
[21] J. D. JACKSON, Classical Electrodynamics, third ed., John Wiley & Sons, Inc.,
New York, NY . . . , 1999, ISBN 0-471-30932-X.
[22] H.-D. NATURE, The Physical Basis of The Direction of Time, fourth ed., Springer-
Verlag, Cambridge . . . , 1920, ISBN 3-540-42081-9.
FT
[23] W. K. H. PANOFSKY AND M. P HILLIPS, Classical Electricity and Magnetism,
second ed., Addison-Wesley Publishing Company, Inc., Reading, MA . . . , 1962,
ISBN 0-201-05702-6.
FUNDAMENTAL PROPERTIES OF
THE ELECTROMAGNETIC FIELD
FT
physics. This includes both discrete and continuous geometric symmetries (con-
jugation, reflection, reversion, translation, rotation) and intrinsic symmetries
(duality, reciprocity). Intimately related to symmetries are conserved quantit-
ies (constants of motion) of which our primary interest will be the electromag-
netic energy, linear momentum, centre of energy, and angular momentum. These
ten conserved quantities1 (one scalar, two vectors and one pseudovector ), can 1The concomitant symmetries
carry information over large distances and are all more or less straightforwardly comprise the ten-parameter
Poincaré group P .10/.
related to their counterparts in classical mechanics (indeed all classical field the-
ories). But we will also consider other conserved quantities where the relation
A
to classical mechanics is less straightforward.
In order to derive convenient and useful expressions for the physical ob-
servables that we want to study, we will take the microscopic Maxwell-Lorentz
equations (2.1) on page 19 as our axiomatic starting point.
R
53
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 76. of 300
/q (4.1a)
and that
dx
j D v mech D
FT
(4.1b)
dt
The transformation properties then follow directly from the Maxwell-Lorentz
equations (2.1) on page 19 as shown in the following three subsections.
4.1.2 C symmetry
A charge conjugation q 7! q 0 D q results in the following changes:
7! 0 D (4.2a)
A
0 0
dx dx dx
j 7! j 0 D 0 D D D j (4.2b)
dt 0 dt 0 dt
7 r0 D r
r ! (4.2c)
@ @ @
7! 0 D (4.2d)
@t @t @t
R
When we apply them to the Maxwell-Lorentz equation (2.1a) we see that
0 . /
r 0 E0 .t 0 ; x0 / D D D D r E.t; x/ (4.3a)
"0 "0 "0
Since r 0 D r , and "0 is unaffected by the changes described by equations (4.2),
D
@B0 .t 0 ; x0 / @B.t; x/
r 0 E0 .t 0 ; x0 / D D r Œ E.t; x/ D (4.3b)
@t 0 @t
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 77. of 300
E0 .t 0 ; x0 / D E.t; x/ (4.4a)
0 0 0
B .t ; x / D B.t; x/ (4.4b)
4.1.3 P symmetry
A 3D spatial inversion x 7! x0 D
in the following changes:
7! 0 D
dx
j 7! j 0 D 0
dt 0
7 r0 D r
r !
D
d. x/
dt
D
0 FT
x, i.e. a mirror-like reflection in 3D, results
dx
dt
D j
(4.5a)
(4.5b)
(4.5c)
A
@ @ @
7! 0 D (4.5d)
@t @t @t
When applied to the Maxwell-Lorentz equations (2.1a) and (2.1c) on page 19,
we see that
R
0
r 0 E0 .t 0 ; x0 / D r E0 .t 0 ; x/ D D D r Œ E.t; x/ (4.6a)
"0 "0
@B0 .t 0 ; x0 / @B.t; x/
r 0 E0 .t 0 ; x0 / D D r Œ E.t; x/ D (4.6b)
@t 0 @t
D
4.1.4 T symmetry
A time reversal t 7! t 0 D t results in the following changes:
7! 0 D (4.8a)
0
j 7! j D j (4.8b)
0
r 7! r D r (4.8c)
@ @ @
7! 0 D (4.8d)
@t @t @t
When we insert these changes into the Maxwell-Lorentz equations (2.1a) and
(2.1c) on page 19, we see that
0
FT
r 0 E0 .t 0 ; x0 / D r E0 . t; x/ D D D r E.t; x/ (4.9a)
"0 "0
@B0 .t 0 ; x0 / @B0 . t; x0 / @B0 . t; x0 /
r 0 E0 .t 0 ; x0 / D D D
@t 0 @. t/ @t
(4.9b)
@B.t; x/
D r E.t; x/ D
@t
implying that under time reversal
E0 .t 0 ; x0 / D E.t; x/ (4.10a)
0 0 0
B .t ; x / D B.t; x/ (4.10b)
A
Hence, E is even and B odd under the T symmetry. The Universe as a whole
is asymmetric under time reversal. On quantum scales this is manifested by
the uncertainty principle and on classical scales by the arrow of time which is
related to the increase of thermodynamic entropy in a closed system.
R
The CPT theorem states that the combined CPT symmetry must hold for all
physical phenomena. No violation of this law has been observed to date. How-
ever, in 1964 it was experimentally discovered that the combined CP symmetry
2This discovery led to the Nobel is violated in neutral kaon decays.2
Prize in Physics 1980.
D
FT
surface S of the volume V at the rate %v per unit time and unit area. Clearly, an
inward flow (antiparallel to n)O will increase the amount of entity in V . We also
have to allow for an increase due to the production of the entity inside V . This
increase is quantified by the source density s. Recalling that the normal vector
nO points outward, the following balance equation must then hold:
Z I Z
d
d3x % D d2x nO %v C d3x s (4.11)
dt V
„ ƒ‚ … „ S ƒ‚ … „Vƒ‚ …
Total change within V Flow into V Production inside V
A
With the help of the divergence theorem, identity (F.121b) on page 223, this
balance equation can be written
Z @%
d3x C r .%v / s D 0 (4.12)
V @t
R
where %v is the flux density of the entity under consideration. Since this balance
equation must hold for any volume V , the integrand must vanish and we obtain
the continuity equation
@%
C r .%v / D s (4.13)
@t
D
This inhomogeneous, linear partial differential equation expresses the local bal-
ance between the explicit temporal change of the density of the entity, the flow
of it across the surface of a control volume, and the local production of the en-
tity within the control volume. In the absence of such a production, i.e. if s D 0,
equation shows that the amount of entity V d3x % in V is only changed if the
R
r .%v / D %r v C v r % (4.16)
FT
we can rewrite the equation of continuity as
d%
C %r .v C r a/ D s (4.17)
dt
where, again, a is an arbitrary, differentiable pseudovector field with dimension
m2 s 1 , e.g. the moment of velocity with respect to x0 , i.e. .x x0 / v .
total electric charge within the control volume V at time t , q D V d3x .t; x/,
R
changes if and only if a charge density flux, or, equivalently, an electric current
density (amount of current per unit area), j passes across the surface enclosing
the control volume V at time t . Hence, it postulates that electric charge can 5 That electricity is indestructible
neither be created nor destroyed. Consequently, the Maxwell-Lorentz equations was postulated already 1747 by
B E N JA M I N F R A N K L I N (1706–
postulate that electric charge is indestructible.5 1770), printer, scientist, inventor,
This is the first conservation law that we have derived from the microscopic philosopher, statesman, and one of
the founding fathers of the United
Maxwell equations (2.1) on page 19 and it can be shown to be a manifestation States of America.
of the gauge symmetry of electrodynamics. This result is, of course, consistent
a posteriori with the fact that these axiomatic laws were formulated under the a
priori assumption that the continuity equation (1.23) on page 10 was valid. As
shown in example 3.4 on page 50, the symmetry that is associated with charge
FT
conservation is the gauge symmetry.
FT
density and is equivalent to the time rate of change of the mechanical kinetic
energy density (Jm 3 ) of the current carrying particles
@umech
D j E D v mech E (4.27)
@t
Equation (4.24) on the previous page can therefore be written
@umech @ufield
C C r S D 0 (4.28)
@t @t
This is the energy density balance equation in local (differential) form.
A
Expressing the Lorentz power, V d3x j E, as the time rate of change of the
R
mechanical energy:
dU mech 3 @ mech
Z Z
D dx u .t; x/ D d3x j E (4.29)
dt V @t V
R
and introducing the electromagnetic field energy
U field D U e C U m (4.30)
(4.31)
2 V
is the electric field energy and
"0
Z
U m .t/ D d3x c 2 B B (4.32)
2 V
is the magnetic field energy, we can write the integral version of the balance
equation (4.28) above as
dU mech dU field
I
C C d2x nO S D 0 (4.33)
dt dt S
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 83. of 300
This is the energy theorem in Maxwell’s theory, also known as Poynting’s the-
orem.
Allowing for an EMF and assuming that Ohm’s law
j D .E C Eemf / (4.34)
FT
which, when inserted into equation (4.33) on the facing page and use is made of
equation (4.29) on the preceding page, yields
dU field
Z I Z
jj
d3x j Eemf D C d2x nO S C d3x (4.37)
V „ dt …
ƒ‚ S V
„ ƒ‚ … „ ƒ‚ … „ ƒ‚ …
Supplied electric power Field power Radiated power Joule heat
This shows how the supplied power (left-hand side, LHS) is expelled in the form
of a time rate change of electric and magnetic field energy, radiated electromag-
netic power, and Joule heat power (Ohmic losses) in the system (right-hand side,
A
RHS).
The conservation of energy is a manifestation of the temporal translational
invariance of the Maxwell-Lorentz equations.
@.E B/ @E @B @B @E
D BCE DE B
@t @t @t @t @t
D jB
2 1 ‚…„ƒ
D E .r E/ c B .r B/ C Bj (4.38)
"0
A combination of the identities (F.79) and (F.86) on page 221 yields
Using the Maxwell-Lorentz equations (2.1a) and (2.1b) on page 19, and the
identity (F.102) on page 222 allows us to write
E .r E/ D r 1
2 .E E/13 E˝E C E (4.40a)
"0
B .r B/ D r 1
2 .B B/13 B˝B C 0 (4.40b)
Let us introduce the electromagnetic linear momentum flux tensor, also known
as (the negative of) the Maxwell stress tensor or the electromagnetic linear mo-
mentum current density,
FT
Tij D ufield ıij "0 Ei Ej "0 c 2 Bi Bj (4.42)
Tij D Tj i (4.43a)
field
Tr.T/ D Ti i D u (4.43b)
A
det.T/ D ufield .cgfield /2 .ufield /2
(4.43c)
c 2 , we see that this relation we can therefore write equation (4.38) on the previous page as
forebodes the relativistic relation
E D mc 2 , where E is energy and @gfield
m is mass. C f C r T D 0 (4.46)
@t
where
f D E C j B (4.47)
FT
for linear momentum
dpmech dpfield
I
C C d2x nO T D 0 (4.50)
dt dt S
where
Z Z
3
p mech
.t / D d xgmech
D d3x %m v mech (4.51a)
V V
and Z Z
pfield .t / D d3x gfield D d3x "0 .E B/ (4.51b)
V V
A
or
Z Z I
d
d3x f C d3x "0 .E B/ C d2x nO T D 0 (4.52)
dt V
„Vƒ‚ … „ ƒ‚ … „S ƒ‚ …
Force on the matter EM field momentum Linear momentum flow
R
This is the linear momentum theorem in Maxwell’s theory which shows that not
only the mechanical particles (charges) but also the electromagnetic field itself
carries linear momentum (translational momentum) and can thus be assumed to
be particle- or fluid-like.
If we assume that we have a single localised charge q, such that the charge
density is given in terms of a Dirac distribution as in equation (1.8) on page 5,
D
with the summation running over one particle only, the evaluation of the first
integral in equation (4.52) above shows that the force on this charge is
Z Z
F D d x f D d3x .E C j B/ D q.E C v mech B/
3
(4.53)
V V
which is the Lorentz force; see also equation (1.45) on page 14. Note that equa-
tion (4.53) above follows directly from a conservation law, and therefore is a
consequence of a symmetry of the Maxwell-Lorentz postulates equations (2.1).
Hence, the Lorentz force does not have to be separately postulated.
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 86. of 300
we see that we can write the field momentum in the form of the following mani-
festly gauge-invariant expression:
Z Z
p .t / D "0 d x E .r A / C "0 d3x Erotat .r Arotat /
field 3 irrot rotat
V V
FT
(4.55)
Let us first evaluate the first integral in the RHS of this expression. Apply-
ing identities (F.79) and (F.86) on page 221, with a 7! Eirrot and b 7! Arotat ,
recalling from the Helmholtz decomposed Maxwell-Lorentz equations (2.42) on
page 29 that r Eirrot D ="0 and that r Eirrot D 0, the first integrand becomes
rotat
Eirrot .r Arotat / D A Arotat .r ˝ Eirrot / (4.56)
"0
C r .Eirrot Arotat / r Eirrot ˝ Arotat (4.57)
A
Performing the integration and using identities (F.121) on page 223, we find that
the first integral in expression (4.55) above can be expressed as the sum of four
integrals:
1
Z Z Z
3 3
irrot
d x E .r A / D rotat
d x A rotat
d3x Arotat .r ˝ Eirrot /
"0 V
R
V V
I I
C d2x n.E O irrot Arotat / d2x nO .Eirrot ˝ Arotat / (4.58)
S S
The second integral in the RHS of this equation can be integrated by parts by
using identity (F.86) on page 221 with a 7! Arotat and b 7! Eirrot . The result is
D
Z I Z
3 2
rotat irrot
d x A .r ˝ E / D d x n.E O irrot rotat
A / d3x .r Arotat /Eirrot
V S V
(4.59)
Inserting this, using the fact that, by definition, r Arotat D 0, we find that the
first integral in equation (4.55) is
Z Z I
"0 d3x Eirrot .r Arotat / D d3x Arotat "0 d2x nO Eirrot ˝ Arotat
V V S
(4.60)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 87. of 300
We evaluate the second integral in equation (4.55) exactly analogously and find
that
Z Z
"0 d3x Erotat .r Arotat / D "0 d3x Arotat .r Erotat /
V
IV (4.61)
"0 d2x nO Erotat ˝ Arotat
S 9 Strictly speaking, in the expres-
Putting all of this together, we find that the electromagnetic linear (translational) sion for Erotat and Arotat , we have
neglected the respective surface
momentum is given by the exact,9 manifestly gauge-invariant formula integrals that must be retained if
Z Z V R3 .
pfield .t / D d3x Arotat "0 d3x Arotat .r Erotat /
V V
I (4.62)
2
"0 d x nO E ˝ Arotat
S
FT
If we restrict ourselves to consider a single temporal Fourier component of
the rotational (‘transverse’) component of E in a region where D 0, and use
the results in example 3.3 on page 49, we find that in complex representation
@A rotat
Erotat D D i!Arotat (4.63)
@t
which allows us to replace Arotat by iErotat =!, yielding, after applying identity
(F.93) on page 222 and equation (2.6) on page 22,
"0
Z
A
3 rotat
˝ field ˛ rotat
p t
D Re i ˝
d x .r E / .E /
2! V
(4.64)
"0
I
2 rotat
C Re i d x nO E ˝ .E /
2! S
If the tensor (dyadic) E ˝ .Erotat / is regular and falls off sufficiently rapidly at
R
large distances (or if nO E D 0), we can discard the surface integral term and find
that the cycle averaged linear momentum carried by the rotational components
of the fields, Erotat and Brotat B D r Arotat , is
"0
Z
3 rotat
˝ field ˛ rotat
p t
D Re i ˝
d x .r E / .E / (4.65)
2! V
D
yD
p i}r (4.67)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 88. of 300
we see that we can write the expression for the linear momentum of the electro-
magnetic field in terms of this operator as
3 Z
"0 X
pfield t D d3x .Eirotat / py Eirotat
˝ ˛
(4.68)
2}! Vi D1
FT
t
i D1 V
pfield t D h‰i j p
˝ ˛
y j‰i i (4.71)
That is, we can represent the cycle (temporal) averaged linear momentum carried
by a monochromatic electromagnetic field as a sum of diagonal quantal matrix
elements (expectation values) where the rotational (‘transverse’) component of
the (scaled) electric field vector behaves as a kind of vector wavefunction.
A
The conservation of linear (translational) momentum is a manifestation of
the spatial translational invariance of the Maxwell-Lorentz equations.
dJ mech
ND (4.73)
dt
Starting from the definition of the mechanical linear momentum density
mech
g , formula (4.48) on page 63, we define the mechanical angular momentum
density about a fixed moment point x0 as
FT
Analogously, we define the electromagnetic moment of momentum density or
electromagnetic angular momentum density about a moment point x0 as the
pseudovector
where gfield is given by equation (4.45) on page 62. Partial differentiation with
respect to time yields
@hfield .t; x; x0 / @
A
.x x0 / gfield
D
@t @t
dx @gfield
D gfield C .x x0 / (4.76)
dt @t
@gfield .t; x/
D .x x0 /
@t
R
where we used the fact that dx=dt is the energy density velocity v field and, hence,
is parallel to gfield ; see equation (4.26) on page 60. We now use equation (4.46)
on page 62 to immediately find that
@hfield .t; x; x0 /
C .x x0 / f .t; x/ C .x x0 / r T.t; x/ D 0 (4.77)
D
@t
Identifying
@hmech .t; x; x0 /
D .x x0 / f .t; x/ (4.78)
@t
as the Lorentz torque density and introducing the electromagnetic angular mo-
mentum flux tensor
dJ mech .x0 /
FT
dJ field .x0 /
I
C C d2x nO K.x0 / D 0 (4.81)
dt dt S
C d2x nO K.x0 / D 0
S
„ ƒ‚ …
Angular momentum flow
This angular momentum theorem is the angular analogue of the linear mo-
mentum theorem, equation (4.52) on page 63. It shows that the electromagnetic
field, like any physical field, can carry angular momentum, also known as rota-
tional momentum.
For a single localised charge q, i.e. for a charge density given by equation
(1.8) on page 5 with summation over one particle only, the evaluation of the first
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 91. of 300
integral in equation (4.83) on the facing page shows that the mechanical torque
on this localised charge q is
Z Z
3
N.x0 / D d x .x x0 / f D d3x .x x0 / .E C j B/
V V (4.84)
D .x x0 / q.E C v mech B/ D .x x0 / F
where F is the Lorentz force given by expression (4.53) on page 63. The phys-
ical observable N.x0 / is known as the Lorentz torque.
FT
A is regular enough and falls off sufficiently fast at large distances [cf. example
3.3 on page 49], we can write
Z
J field
.t; x0 / D "0 d3x .x x0 / ŒE.t; x/ B.t; x//
V
Z (4.85)
3 rotat
D "0 d x .x x0 / E.t; x/ Œ.r A .t; x/
V
Employing the vector identity (F.92) on page 221, we can rewrite this as
A
Z
J field
.t; x0 / D "0 d3x .x x0 / Œ.r ˝ Arotat / E
V
Z (4.86)
"0 d3x .x x0 / ŒE .r ˝ Arotat /
V
which consists of one intrinsic term (a term that is not dependent on the choice
of moment point x0 ):
Z
† field .t / D "0 d3x E.t; x/ Arotat .t; x/ (4.88)
V
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 92. of 300
where use was made of the Maxwell-Lorentz equation (2.1a) on page 19.
We note that the first term in equation (4.89) above can be written
Z Z
3 rotat
d x .x x0 / A .t; x/ D d3x x Arotat .t; x/ (4.90)
V V
FT
Z
x0 d3x Arotat .t; x/ (4.91)
V
Hence, the total field angular momentum J field is the sum of two manifestly
gauge invariant components, one intrinsic († field ) and one extrinsic (Lfield ):
in perfect analogy with the total mechanical angular momentum; cf. equation
(4.82a) on page 68.
If there is no net electric charge density in the integration volume, the first
integral in the RHS of equation (4.89) above vanishes, and if E ˝.x x0 /Arotat
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 93. of 300
is regular and falls off sufficiently rapidly with jxj, the contribution from the
surface integral in equation (4.89) on the facing page can be neglected if we
make the enclosing volume V large enough. Furthermore, for a single temporal
Fourier component and in complex notation, we obtain the expressions
"0
Z
d3x .E E/
˝ field ˛
† t
D i (4.95a)
2! V
"0
Z
d3x Ei Œ.x x0 / r Ei
˝ field ˛
L .x0 / t D i
2! V
"0
Z
D i d3x Ei .x r /Ei
2! V (4.95b)
"0
Z
C i x0 d3x Ei r Ei
2! V
D Lfield .0/ t x0 pfield t
FT
˝ ˛ ˝ ˛
˝ ˛
where pfield t is the cycle averaged EM field linear momentum given by expres-
˝ ˛
sion (4.66) on page 65. We see that if pfield t ¤ 0 it is always possible to choose
˝ ˛ ˝ ˛
the moment point x0 such that Lfield .x0 / t D 0. But if pfield t D 0, i.e. if the
cycled average of the linear momentum of the monochromatic field in question
˝ ˛
vanishes, then Lfield t is independent of x0 .
Recalling that in quantum mechanics the spin angular momentum operator
is
A
bj k D
† Oi
i}ij k x (4.96)
which, with the help of the matrix vector expression (M.26) on page 240 can be
written
bD
† }S (4.97)
R
and that the quantal orbital angular momentum operator is
yD
L i}x r D Oi
i}ij k xj @k x (4.98)
d3x Ej †
˝ field ˛
† t
D bj k Ek (4.99a)
2}! V
"0
Z
d3x Ei Ly Ei
˝ field ˛
L .0/ t D (4.99b)
2}! V
or, with the use of the vector ‰ introduced in equation (4.69) on page 66,
Z
d3x ‰i †
˝ field ˛
† t
D b ‰i D h‰i j †
b j‰i i (4.100a)
V
Z
L .0/ t D d3x ‰i L y ‰i D h‰i j L
y j‰i i
˝ field ˛
(4.100b)
V
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 94. of 300
Hence, under the assumptions made above, we can interpret † field as the
electromagnetic spin angular momentum and Lfield as the electromagnetic orbital
angular momentum.
The conservation of angular (rotational) momentum is a manifestation of the
rotational invariance of the Maxwell-Lorentz equations.
E 7! cB (4.101a)
cB 7! E
FT
(4.101b)
e m
c 7! (4.101c)
m e
7! c (4.101d)
cj e 7! j m (4.101e)
m e
j 7! cj (4.101f)
This transformation leaves the symmetrised Maxwell equations, and hence the
physics they describe (often referred to as electromagnetodynamics), invariant.
D
Since E and j e are true (polar) vectors, B a pseudovector (axial vector), and e
13 Recall that the Taylor expansion a (true) scalar, we conclude that the magnetic charge density m as well as the
of cos contains only even angle , which behaves as a mixing angle in a two-dimensional charge space,
powers of and therefor is an
ordinary (true) scalar, whereas must be pseudoscalars13 and the magnetic current density j m a pseudovector.
the expansion of sin contains The invariance of Dirac’s symmetrised Maxwell equations under the duality
only odd powers and therefore is a
pseudoscalar. transformation (4.102) means that the amount of magnetic monopole density m
is irrelevant for the physics as long as the ratio m =e D tan is kept constant.
So whether we assume that charged particles are only electrically charged or
also have an amount of magnetic charge with a given, fixed ratio between the
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 95. of 300
4.3. Examples j 73
FT
@ .x x0 / gmech @ .x x0 / gfield
C r .x x0 / T D ufield
C
@t @t
(4.103)
This is the electromagnetic virial theorem, analogous to the virial theorem of
Clausius in mechanics. The quantity .x x0 / gfield is the electromagnetic
virial density. When integrated over space and time averaged, this theorem is a
statement of the partitioning of energy in electrodynamics and finds use in, e.g.
plasma physics.
A
4.3 Examples
C HARGE CONJUGATION
C W ˆ 7! ˆ (4.104a)
C W A 7! A (4.104b)
P W ˆ 7! ˆ (4.105a)
P W A 7! A (4.105b)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 96. of 300
T IME REVERSAL
T W ˆ 7! ˆ (4.106a)
T W A 7! A (4.106b)
FT
is conserved.
To show that the total energy U of the closed system is conserved, i.e. is constant in time,
is to show that the derivative of U with respect to time t vanishes. Direct differentiation
yields
N
@E @B
Z
dU dvi
C "0 d3x E C c2B
X
D mi vi (4.108)
dt dt V @t @t
i D1
From the Lorentz force equation (4.53) on page 63 and Newton’s second law we find that
dvi
mi D qi E.t; xi / C vi B.t; xi /
A
dt
which means that
N N N N
X dvi X X X
mi vi D qi vi E.t; xi / C qi vi vi B.t; xi / D qi vi E.t; xi /
dt
iD1 i D1 i D1
„ ƒ‚ … iD1
0
N Z
dU
d3x j E.t; x/
X
D qi vi E.t; xi /
dt V
iD1 (4.110)
1
Z
d3x E .r B/ B .r E/
C
0 V
Since the current is carried by discrete charged particles i D 1; 2; : : : ; N , the current density
can be represented as
N
X
jD qi v i ı x xi .t/ (4.111)
i D1
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 97. of 300
4.3. Examples j 75
and therefore the two first terms in equation (4.110) on the facing page cancel. If we use
also formula (F.83) on page 221 we see that equation (4.110) can be written
1
Z Z I
dU
D d3x r .E B/ D d3x r S D d2x nO S (4.112)
dt 0 V V S
where in the last step the divergence theorem, formula (F.121b) on page 223, was used.
This result shows that the rate at which the total energy is lost in a volume equals the
amount of energy flux that flows outward across a closed surface enclosing this volume.
Of course, this is nothing but what is stated in the energy theorem (Poynting’s theorem),
formula (4.33) on page 60. Now, if the surface lies entirely outside the boundaries of the
system under study, and this system is closed, no energy flux passes through the surface and
hence
dU
D0
dt
showing that the total energy U D U mech C U field of a closed system is indeed a conserved
FT
quantity. QED
Put in another way: If we observe a change in the radiated energy from a given system, we
can deduce that there has been a similar change, but with opposite sign, of the mechanical
energy of the system. This can be useful to know if we want to study the dynamics of a
remote system, e.g. in the Universe.
End of example 4.2C
BConservation of the total linear momentum in a closed system E X A M P L E 4.3
Show, by explicit calculation, that the total linear momentum p D pmech C pfield of a closed
electromechanical system comprising N non-relativistic particles of mass mi , speed vi and
A
charge qi , and pertinent electromagnetic fields E and B,
N Z
d3x .E B/
X
pD mi vi C "0 (4.113)
i D1 V
is conserved.
To show that the total linear momentum p of the closed system is conserved, i.e. is con-
R
stant in time, is to show that the derivative of p with respect to time t vanishes. Direct
differentiation yields
N @E @B
Z
dp dvi
C "0 d3x
X
D mi BCE (4.114)
dt dt V @t @t
iD1
D
From the Lorentz force equation (4.53) on page 63 and Newton’s second law we find that
dvi qi
D E.t; xi / C vi B.t; xi /
dt mi
Substitution of this expression and equations (4.109) on the preceding page into equation
(4.114) yields, after some rearrangement of terms,
N Z
dp
d3x j B.t; x/
X
D qi E.t; xi / C vi B.t; xi /
dt V
iD1 (4.115)
Z
"0 d3x E .r E/ C c 2 B .r B/
V
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 98. of 300
By first applying formula (F.93) on page 222 and then formula (F.86) on page 221 on the
last term, we find that
N Z
dp
d3x j B.t; x/
X
D qi E.t; xi / C vi B.t; xi /
dt V
i D1
Z
(4.116)
"0 d3x .E ˝ r E C c 2 B ˝ r B/
V
Z
"0 d3x 21 r .E E C c 2 B B/ r .E ˝ E C c 2 B ˝ B/
V
We can now use the Maxwell-Lorentz equations to make the substitutions "0 r E D and
r B D 0 to obtain
N Z
dp
d3x E C j B.t; x/
X
D qi E.t; xi / C vi B.t; xi /
dt V
i D1 (4.117)
Z
FT
3 2 2
"0 d x r 12 .E E C c B B/13 .E ˝ E C c B ˝ B/
V
Since the current is carried by discrete charged particles the current density j can be rep-
resented as formula (4.111) and the charge density as
N
X
D qi ı.x xi / (4.118)
iD1
the two first terms in equation (4.117) above cancel and, after applying the divergence
formula (F.121b) on page 223, we are left with
I I
dp
D "0 d2x nO 21 .E E C c 2 B B/13 .E ˝ E C c 2 B ˝ B/ D "0 d2x nO T
dt
A
S S
(4.119)
where the tensor T is the electromagnetic linear momentum current density (the negative of
the Maxwell stress tensor); cf. equation (4.41) on page 62.
This result shows that the rate at which the total linear momentum is lost in a volume equals
the amount of linear momentum flux that flows outward across a closed surface enclosing
R
this volume. If the surface lies entirely outside the boundaries of the system under study,
and this system is closed, no linear momentum flux passes through the surface and hence
dp
D0
dt
showing that indeed the total linear momentum p D pmech C pfield of a closed system is a
conserved quantity. QED
D
If our system under study can be considered to be a closed electromechanical system and
we observe a change in the electromagnetic linear momentum in this system, we can de-
duce that there has been a similar change, but with opposite sign, of the mechanical linear
moment of the system. This allows us to determine mechanical properties of a system by
analysing the radiation (radio, light, . . . ) from it.
In the quantum picture the linear momentum of a photon is }k where k D jkj D 2= is
the wavenumber and the radiated wavelength. The shift in is the translational Doppler
shift. Since ! D ck for a photon in free space, we experience the change of electromagnetic
linear momentum as an associated frequency shift, a redshift if it is to the long wavelength
(low frequency) side and a blueshift if it is to the short wavelength (high frequency) side.
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 99. of 300
4.3. Examples j 77
This vocabulary is used regardless of whether the radiation falls in the optical range or not.
.x O k .xl
x0 / .r T/ D klm x x0l /.r T/m (4.120)
In tensor notation
O i @i x
r T D x Oj x
O k Tj k D ıij x
O k @i Tj k D x
O k @ i Ti k x
O j @i Tij (4.121)
FT
where, in the last step, we made a summation (dummy) index replacement k 7! j so that
we recognise this as identity (F.72) on page 221 for A D T.
By definition
O m D ıj m @i Tij D @i Ti m
.r T/m .Oxj @i Tij /m D .Oxj @i Tij / x (4.122)
fields E and B,
N Z
d3x .x
X
J .t; x0 / D .xi x0 / mi vi C "0 x0 / .E B/ (4.124)
i D1 V
is conserved.
To show that the total angular momentum J of the closed system is conserved, i.e. is con-
stant in time, is to show that the derivative of J with respect to time t vanishes.
For simplicity we put the origin of our coordinate system at the moment point x0 and then
follow the same procedure as for the proof of the constancy of linear momentum in example
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 100
. of 300
From the Lorentz force equation (4.53) on page 63 and Newton’s second law we find that
dvi qi
D E.t; xi / C vi B.t; xi /
dt mi
FT
Substitution of this expression and equations (4.109) on page 74 into equation (4.125)
yields, after some rearrangement of terms,
N
dJ X
D qi xi E.t; xi / C xi vi B.t; xi /
dt
iD1 (4.126)
Z Z
d3x x .j B/ "0 d3x x E .r E/ C c 2 B .r B/
V V
By first applying formula (F.93) on page 222, and then formula (F.86) on page 221 on the
last term, we find that
A
N
dJ X
D xi qi E.t; xi / C vi B.t; xi /
dt
i D1
Z Z
(4.127)
d3x x .j B/ "0 d3x x .E ˝ r E C c 2 B ˝ r B/
V V
Z
"0 d x x 2 r .E E C c 2 B B/ r .E ˝ E C c 2 B ˝ B/
3
R
1
V
We then use the Maxwell-Lorentz equations to make the substitutions "0 r E D and
r B D 0 and obtain
N Z
dJ
d3x x E C .j B/
X
D xi qi E.t; xi / C vi B.t; xi /
dt
D
V
i D1 (4.128)
Z
"0 d3x x r 12 .E E C c 2 B B/13 .E ˝ E C c 2 B ˝ B/
V
The electric current is carried by discrete charged particles so that the current density j is
given by formula (4.111) on page 74 and the charge density by formula (4.118) on page 76.
Hence, the two first terms in equation (4.128) above cancel. After applying the divergence
theorem, formula (F.121b) on page 223, on the remaining third term, we are left with
I
dJ
D "0 d2x nO K (4.129)
dt S
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 101
. of 300
4.3. Examples j 79
where the tensor K is the electromagnetic angular momentum current density; cf. equation
(4.79) on page 67.
This result shows that the rate at which the total angular momentum is lost (gained) in a
volume is equal to the amount of angular momentum flux that flows outward (inward) across
a closed surface enclosing this volume. If the surface lies entirely outside the boundaries
of the system under study, and this system is closed, no angular momentum flux passes
through the surface and hence
dJ
D0
dt
Hence, we have shown that the total angular momentum J D J mech C J field of a closed
system is conserved. QED
If we observe a change in the electromagnetic angular momentum in a given system, we
can deduce that there has been a similar change, but with opposite sign, of the mechanical
FT
angular moment of the system.
and
1
hO ˙ D p xO 1 ˙ iOx2 / (4.132)
2
R
As before, we use the convention that hO C represents left-hand circular polarisation and hO
right-hand circular polarisation. Noting that
.hO ˙ / hO ˙ D ˙iOz (4.133)
we see that
D
When we insert this into equation (4.95a), we find that the cycle averaged spin angular
momentum of a circularly polarised wave is
˝ field ˛
"0 U
D E Z
† field D ˙ d3x E 2 zO D ˙ t
zO (4.135)
t 2! V !
where U field is the field energy. Considering the fact that the wave-particle duality shows
that the wave can be considered to be a gas with N photons so that the kinetic energy of the
field is U field D N }!. This means that the spin of the wave is
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 102
. of 300
˝ field ˛
D E U t N }!
† field D ˙ zO D ˙ zO D ˙N } zO (4.136)
t ! !
Hence, each photon of a right-hand or a left-hand circular polarised wave carries a spin
angular momentum of } or }, respectively.
FT
@ @
yz D i} x
L y (4.137c)
@y @x
as is well known from Quantum Mechanics.
In cylindrical coordinates .; '; z/ the components are
@
@ z @
y
Lx D i} sin ' z C cos ' (4.138a)
@ @z @'
@
@ z @
yy D i} cos ' z
L sin ' (4.138b)
@ @z @'
@
yz D i}
L (4.138c)
@'
A
and in spherical coordinates .r; '; / they are
@ @
yx D i} sin '
L C cos ' cot (4.139a)
@ @'
@ @
yy D i} cos '
L sin ' cot (4.139b)
@ @'
R
@
yz D i}
L (4.139c)
@'
For an electric field E that depends on the azimuthal angle ' in such a way that
E D E0 .t; x/ˆ.'/ (4.140)
D
we find that
@E .t; x/ @ˆ.'/
yz E.t; x/ D 0
L i} ˆ.'/ E0 .t; x/ (4.141)
@' @'
If E0 .t; x/ is rotationally symmetric around the the z axis, so that E0 D E0 .t; ; z/ in
cylindrical coordinates, E0 D E0 .t; r; / in spherical (polar) coordinates and equivalently
in other coordinates, the first term in the RHS vanishes.
If the azimuthal part is expressed in a Fourier series
1
X
ˆ.'/ D cm eim' (4.142)
mD 1
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 103
. of 300
4.3. Examples j 81
we see that
1
X 1
X
yz E.t; x/ D
L cm m}E0 eim' D cm m}Em (4.143)
mD 1 mD 1
yz Em D m}Em
L (4.144)
which means that for a rotationally symmetric beam with an azimuthal phase dependence
given by exp.im'/ one deduces, according to formula (4.95b) on page 71, that the z com-
ponent of the orbital angular momentum of each photon in this beam is m}.
E X A M P L E 4.8
FT
BDuality of the electromagnetodynamic equations
Show that the symmetric, electromagnetodynamic Maxwell-Lorentz equations (2.2) on
page 20 (Dirac’s symmetrised Maxwell equations) are invariant under the duality trans-
formation (4.102).
Explicit application of the transformation yields
e
r ?E D r .E cos C cB sin / D cos C c0 m sin
"0
?e (4.145a)
1 1
D e cos C m sin D
"0 c "0
1 e
A
?
r B D r . E sin C B cos / D sin C 0 m cos
c c"0 (4.145b)
D 0 ce sin C m cos D 0 ?m
@?B
@ 1
r ?E C D r .E cos C cB sin / C E sin C B cos
@t @t c
@B 1 @E
D 0 j m cos cos C c0 j e sin C sin
R
@t c @t (4.145c)
1 @E @B
sin C cos D 0 j m cos C c0 j e sin
c @t @t
D 0 . cj e sin C j m cos / D 0 ?j m
1 @?E 1 1 @
r ?B D r . E sin C B cos / .E cos C cB sin /
c 2 @t c c 2 @t
D
1 1 @B 1 @E
D 0 j m sin C cos C 0 j e cos C 2 cos
c c @t c @t
1 @E 1 @B
2 @t
cos sin
c c @t
1 m
D 0 j sin C j e cos D 0 ?j e
c
(4.145d)
QED
?
G D ?E C ic ?B D E cos C cB sin iE sin C icB cos
(4.146)
i i
D E.cos i sin / C icB.cos i sin / D .E C icB/e D Ge
i
ˇ ˇ2
?
G ?G D ˇ?G ˇ D Ge i i
G ei D jGj2
Ge D Ge (4.147)
whereas
?
G ?G D G Ge 2i
FT
(4.148)
If this transformation were local, i.e. if D .t; x/, we see that spatial and temporal differ-
entiation of ?G would lead to
@?G @ @G @ @G
D i e i G C e i D i ?G C e i
(4.149a)
@t @t @t @t @t
r ?G D i.r / e i G C e i r G D i.r / ?G C e i
r G (4.149b)
? i i ? i
r GD i.r / e GCe r G D i.r / G C e r G (4.149c)
A
However, if we require that the physics be unaffected by a duality transformation, the free-
space Maxwell-Lorentz equations (2.9) on page 22 must hold also for ?G. I.e.
r ?G D 0 (4.150a)
i @?G
r ?G D (4.150b)
R
c @t
Then, using formulæ (4.149), as well as equations (2.9), we find that this would mean that
.r / ?G D 0 (4.151a)
i @ ?
.r / ?G D G (4.151b)
D
c @t
@
D0 (4.152a)
@t
r D 0 (4.152b)
4.3. Examples j 83
For instance, as can be derived directly from the Maxwell-Lorentz equations, the following
conservation law holds in free space (vacuum):
@
C r X D 0 (4.153)
@t
where the pseudoscalar
def
E .r E/ C c 2 B .r B/ (4.154)
FT
is the chirality density, and the pseudovector
def @E @B
X E C c2B (4.155)
@t @t
is the chirality flow.
If we use the notation j cond for the conduction current associated with the actual motion
of electric charges, both free and bound, i.e. j cond D j free C j bound , and j disp for the
displacement current "0 @.E/=@t, the Maxwell-Lorentz equation (2.1d) on page 19 can be
A
written
1
r B D j cond C j disp D j tot (4.156)
0
Differentiating this with respect to time t and using the Maxwell equation (2.1c) on page 19
and formula (F.104) on page 222, we obtain the following local conservation law for the
total current
R
@j tot 1
C r 13 .r E/ D 0 (4.157)
@t 0
which is a rather obfuscated way of writing the wave equation for the electric field vector
E. Normally it is written as in equation (2.19a) on page 24.
We note that the global (i.e. volume integrated) version of this wave-equation-turned-
D
conservation-law formula can, with the help of the rank two tensor
W D 13 .r E/ (4.158)
be written
1
Z I
d
d3x j tot C d2x nO W D 0 (4.159)
dt V 0 S
and hence as a conservation law for the integrated curl of the magnetic field:
Z I
d
d3x .r B/ C d2x nO W D 0 (4.160)
dt V S
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 106
. of 300
4.4 Bibliography
[25] J. D. JACKSON, Classical Electrodynamics, third ed., John Wiley & Sons, Inc.,
New York, NY . . . , 1999, ISBN 0-471-30932-X.
FT
[26] F. M ELIA, Electrodynamics, Chicago Lectures in Physics. Cambridge University
Press, Cambridge and London, 1991, ISBN 0-226-51957-0.
A
R
D
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 107
. of 300
The electric and magnetic fields E and B generated by prescribed charge and
current sources and j can — at least in principle — be obtained by directly
FT
solving the Maxwell-Lorentz differential equations given at the beginning of
chapter 2, or the wave equations given later in the same chapter. However, it is
often technically easier and physically more lucid to calculate the fields from the
electromagnetic potentials ˆ and A that we introduced in chapter 3. We saw in
that chapter that these potentials can, in a suitable gauge, be readily obtained in
the form of a volume integral over the spatial distribution of the source elements,
divided by the distance between the actual source element and the observer.
In this chapter we will use electromagnetic potentials to derive exact, closed-
form, analytic expressions for the electric and magnetic fields generated by pre-
A
scribed but completely arbitrary charge and current sources at rest, distributed
arbitrarily within a volume of finite extent in otherwise free space. As we shall
find, both the electric and magnetic field vectors are actually a sum of several
vector composants, each characterized by its particular vectorial property and
fall-off behaviour with respect to distance from the source elements. This means
R
that different field vector composants have different magnitudes, directions and
phases in different zones. These zones are customarily — and self-explanatorily
— referred to as the near zone, the intermediate zone, and the far zone, respect-
ively.
Those composants of the field vectors that have the slowest fall-off with dis-
D
tance from the source and therefore dominate in the far zone, will be referred
to as the far fields. Because of their importance, a special analysis of these
far fields is given at the end of the chapter. Surprising as it may seem, it will
be shown in chapter 6 that certain physical observables receive their far-zone
contributions not from the dominant far fields but from a combination of far
fields and sub-dominant near-zone and intermediate-zone fields. For complete-
ness, we therefore include a derivation of approximate expressions for all field
composants, dominant as well as sub-dominant, valid at large distances from an
arbitrary source.
85
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 108
. of 300
FT
well-behaved, non-erratic physical variables which are neither strictly monoton-
ically increasing nor strictly monotonically decreasing with time. For charge and
current densities that vary in time, we can therefore, without loss of generality,
work with individual temporal Fourier components ! .x/ and j! .x/, respect-
ively. Strictly speaking, the existence of a signal represented by a single Fourier
component assumes a monochromatic source (i.e. a source containing only one
single frequency component), which requires that this source must have existed
for an infinitely long time. However, by taking the proper limits, we may still
use this approach even for sources and fields of finite temporal duration.
A
This is the method we shall utilise in this chapter in order to derive the elec-
tric and magnetic fields in vacuum from arbitrary given charge densities .t; x/
and current densities j.t; x/, defined by the temporal Fourier transform pairs
Z 1
i!t
.t; x/ D d! ! .x/ e (5.1a)
R
1
1
1
Z
! .x/ D dt .t; x/ ei!t (5.1b)
2 1
and
D
Z 1
i!t
j.t; x/ D d! j! .x/ e (5.2a)
1
1
1
Z
j! .x/ D dt j.t; x/ ei!t (5.2b)
2 1
be written
Z 1
ˆ.t; x/ D d! ˆ! .x/ e i!t (5.3a)
1
1
1 1
Z Z
d3x 0 ! .x0 /G.ˇx x0 ˇ/
ˇ ˇ
ˆ! .x/ D dt ˆ.t; x/ ei!t D (5.3b)
2 1 "0 V 0
where, in the last step, we made use of the explicit expression for the temporal
Fourier transform of the generic potential component ‰! .x/, equation (3.31) on
page 39, and introduced the Green function
x0 j
ˇ 0ˇ
ˇ eikjx
G.ˇx x /D (5.4)
4 jx x0 j
Similarly, we must require that the following Fourier transform pair for the
vector potential exists:
FT
Z 1
A.t; x/ D d! A! .x/ e i!t (5.5a)
1
Z 1
1 1
Z
d3x 0 j! .x0 /G.ˇx x0 ˇ/ (5.5b)
ˇ ˇ
A! .x/ D dt A.t; x/ ei!t D 2
2 1 "0 c V 0
Analogous transform pairs must exist for the fields themselves.
In the limit that the sources can be considered monochromatic, containing
one single frequency !0 only, we can safely assume that ! D 0 ı.! !0 /,
j! D j0 ı.! !0 / etc.. Our Fourier integrals then become trivial and we obtain
the simpler expressions
A
i!0 t
.t; x/ D 0 .x/e (5.6a)
i!0 t
j.t; x/ D j0 .x/e (5.6b)
i!0 t
ˆ.t; x/ D ˆ0 .x/e (5.6c)
i!0 t
A.t; x/ D A0 .x/e (5.6d)
R
where the real-valuedness of all these quantities is implied. As discussed above,
all formulæ derived for a general temporal Fourier representation of the source
(general distribution of frequencies in the source) are valid in these limiting
cases. In this context, we can therefore, without any essential loss of stringency,
D
FT
D dx
4"0 V 0 jx x0 j3
0
! .x0 /eikjx x j .x x0 /
Z
ik
d3x 0
4"0 V 0 jx x0 j2
0
eikjx x j
Z
ik
C d3x 0 j! .x0 /
4"0 c V 0 jx x0 j
Taking the inverse Fourier transform of this expression, the following expression
for the retarded electric field is obtained:
0
1 .tret ; x/ x x0
Z
E.t; x/ D d3x 0
A
4"0 V 0 jx x0 j2 jx x0 j
0
1 P ret
.t ; x0 / x x0
Z
C d3x 0 (5.9)
4"0 c V 0 jx x0 j jx x0 j
1 P 0 0
3 0 j.tret ; x /
Z
d x
4"0 c 2 V 0 jx x0 j
R
Letting dq 0 D d3x 0 .tret
0
; x0 / and di 0 D I 0 dl0 D d3x 0 j.tret
0
; x0 /, the corresponding
formula in infinitesimal differential form becomes
x x0 1 0 x x0
1 1 P0 1
dE.t; x/ D dq 0 C d qP d i (5.10)
4"0 jx x0 j3 c jx x0 j2 c 2 jx x0 j
D
which in the static limit reduces to the infinitesimal Coulomb law, formula (1.5)
on page 4.
We shall now further expand equation (5.9) above. To this end we first note
that the Fourier transform of the continuity equation (4.20) on page 58
Doing so in the last term of equation (5.8) on the facing page, and also using the
fact that k D !=c, we can rewrite this equation as
0
! .x0 /eikjx x j .x x0 /
Z
1
E! .x/ D d3x 0
4"0 V 0 jx x0 j3
0 0 ikjx x0 j
1 3 0 Œr j! .x /.x x0 /
Z
0 e
dx 0
ikj! .x /
c V 0 jx x j jx x0 j
„ ƒ‚ …
K!
(5.13)
The last vector-valued integral can be further rewritten in the following way
(where l and m are summation indices and Einstein’s summation convention is
FT
assumed):
0
Œr 0 j! .x0 /.x x0 / eikjx x j
Z
K! D d3x 0 ikj! .x0 /
V0 jx x0 j jx x0 j
0 0 (5.14)
eikjx x j
@j!;l xm xm
Z
3 0 0
D dx Om
ikj!m .x / x
V0 @xl0 jx x0 j jx x0 j
But, since
0 0
@ xm xm @j!;l xm xm
A
x0 j 0
j!;l eikjx D eikjx x j
@xl0 jx x0 j2 @xl0jx x0 j2
0 (5.15)
@ xm xm x0 j
C j!;l 0 eikjx
@xl jx x0 j2
we can rewrite K! as
R
" 0
#
0
eikjx x j
@ xm xm
Z
3 0 x0 j
K! D d x j!;l 0 O m eikjx
x C ikj!
V0 @xl jx x0 j 2 jx x0 j
0
@ xm xm
Z
ikjx x0 j
C d3x 0 0 j!;l Ome
x
0 @x jx x0 j2
D
V
" l 0
#
x0 eikjx x j
Z
3 0 0 0 x ikjx x0 j 0
D d x j! .x / r e C ikj! .x /
V0 jx x0 j2 jx x0 j
x0
Z
3 0 0 0 x ikjx x0 j
C d x r j! .x / e
V0 jx x0 j 2
(5.16)
where, according to identity (F.121e) on page 223, the last term vanishes if the
dyadic inside the big parentheses is regular and tends to zero at large distances.
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 112
. of 300
Further evaluation of the derivative in the first term makes it possible to write
x0 j
eikjx
Z
d3x 0 j! .x0 / .x x0 / .x x0 /
K! D 2
V0 jx x0 j4
x0 j
eikjx
Z
3 0 0 0 0
C ik d x j! .x / .x x / .x x/
V0 jx x0 j3
(5.17)
x0 j
eikjx
Z
3 0 0
C d x j! .x /
V0 jx x0 j2
0
eikjx x j
Z
3 0 0
ik d x j! .x /
V0 jx x0 j
FT
Using the triple product ‘bac-cab’ formula (F.53) on page 220 backwards, and
inserting the resulting expression for K! into equation (5.13) on the previous
page, we arrive at the following final expression for the temporal Fourier trans-
form of the total E field:
A
0
1 eikjx x j
Z
E! .x/ D r d3x 0 ! .x0 /
4"0 V0 jx x0 j
Z ikjx x j 0
i! 3 0 0 e
C d x j ! .x /
4"0 c 2 V 0 jx x0 j
R
0
! .x0 /eikjx x j .x x0 /
Z
1
D d3x 0
4"0 V 0 jx x0 j3
0 (5.18)
1 Œj! .x0 /eikjx x j .x x0 /.x x0 /
Z
C d3x 0
c V0 jx x0 j4
0
1 Œj! .x0 /eikjx x j .x x0 / .x x0 /
Z
D
C d3x 0
c V0 jx x0 j4
0 ikjx x0 j
.x x0 / .x x0 /
3 0 Œj! .x /e
Z
ik
dx
c V0 jx x0 j3
Taking the inverse Fourier transform of equation (5.18) above, once again
using the vacuum relation ! D kc, we find, at last, the expression in time domain
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 113
. of 300
FT
„ ƒ‚ …
Intermediate field
1
Z P ret
Œj.t 0
; x0 / .x x0 / .x x0 /
C d3x 0
4"0 c 2 V0 jx x0 j3
„ ƒ‚ …
Far field
where, as before,
P ret
0 def @j
j.t ; x0 / (5.20)
@t 0
tDtret
A
Here, the first term represents the retarded Coulomb field and the last term rep-
resents the far field which dominates at very large distances. The other two
terms represent the intermediate field which contributes significantly only to the
fields themselves in the near zone and must be properly taken into account there.
R
equation (5.5) and equation (5.5b) on page 87, and formula (3.9) on page 35:
Using the Fourier transformed version of this equation and equation (5.5b) on
page 87, we obtain
0
1 eikjx x j
Z
B! .x/ D r A! .x/ D r d3x 0 j! .x0 / (5.22)
4"0 c 2 V0 jx x0 j
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 114
. of 300
Utilising formula (F.87) on page 221 and recalling that j! .x0 / does not depend
on x, we can rewrite this as
" 0
!#
1 eikjx x j
Z
3 0 0
B! .x/ D d x j! .x / r
4"0 c 2 V 0 jx x0 j
x x0
Z
1 3 0 0 0
D 2
d x j! .x / 3
eikjx x j
4"0 c V0 jx x j 0
0
1
Z
x x ikjx x0 j
C d3x 0 j! .x0 / ik e (5.23)
V0 jx x0 j jx x0 j
0 ikjx x0 j
.x x0 /
Z
1 3 0 j! .x /e
D d x
4"0 c 2 V 0 jx x0 j3
0 ikjx x0 j
.x x0 /
3 0 . ik/j! .x /e
Z
C dx
FT
V0 jx x0 j2
From this expression for the magnetic field in the frequency (!) domain, we
finally obtain the total magnetic field in the temporal (t ) domain by taking the
inverse Fourier transform (using the identity ik D i!=c):
Z 1
B.t; x/ D d! B! .x/ e i!t
1
hR i
1 0 i.!t kjx x0 j/
1
Z
1 d! j ! .x /e .x x0 /
3 0
D d x
4"0 c 2 V 0 jx x0 j3
A
hR i
1 0 i.!t kjx x0 j/
1
Z
1 d! . i!/j ! .x /e .x x0 /
3 0
C dx
c V0 jx x0 j2
(5.24)
R
Comparing with equations (3.33) on page 40, we can identify the exponents
0
of the exponentials in the integrands as i!tret and find that the total retarded
1 Equation (5.9) and equation magnetic field can be written as the sum of two terms1
(5.25) seem to have been first
0 0
1 3 0 j.tret ; x / .x x0 /
Z
introduced by Panofsky and
Phillips. Later they were given by B.t; x/ D d x
4"0 c 2 V 0 jx x0 j3
D
where
P ret
0 def @j
j.t ; x0 / (5.26)
@t 0
tDtret
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 115
. of 300
5.4. The total electric and magnetic fields at large distances from the sources j 93
The first term, the retarded induction field that dominates near the current source
but falls off rapidly with distance from it, is the electrodynamic version of the
Biot-Savart law in electrostatics, formula (1.16) on page 8. The second term, the
far field , dominates at large distances. The spatial derivatives (r ) gave rise to
a time derivative (P) so this term represents the part of the magnetic field that is
generated by the time rate of change of the current, i.e. accelerated charges, at
the retarded time.
In infinitesimal differential form, formula (5.25) on the preceding page be-
comes
x x0 x x0
1 0 0 0 1 P0 0
dB.t; x/ D di .t ; x / C d i .t; x /
4"0 c 2 jx x0 j3 c jx x0 j2
(5.27)
FT
where
di 0 .t 0 ; x0 / D d3x 0 j.tret
0
; x0 / (5.28a)
and
def @.di/ @j
diP 0 .t 0 ; x0 / 3 0
Ddx (5.28b)
@t 0
tDtret @t 0
tDtret
FT
V0
O
such a limited spatial extent that sup jx0 x0 j inf jx x0 j, and the integration
surface S.x0 /, centred on x0 and with an outward pointing normal unit vector
nO D x\ x0 , has a large enough radius jx x0 j sup jx0 x0 j.
The exact wave vector of fields generated at x0 reaching an observer at x can
be written
A
O 0 / k x[ x x0
k.x0 / D k k.x x0 D k (5.29)
jx x0 j
expressing the fact that its magnitude k D !=c is constant but its direction is
along x x0 and thus is dependent on the location of the source element at x0 in
R
V 0 . Now,
ˇx x0 ˇ ˇ.x x0 / .x0 x0 /ˇ
ˇ ˇ ˇ ˇ
q ˇ2
D jx x0 j2 2.x x0 / .x0 x0 / C ˇx0 x0 ˇ
ˇ
12
x0 / .x0 jx0 x0 j2
D
2.x x0 /
D jx x0 j 1 C
jx x0 j2 x0 j2
jx
x0 x0 1 jx0 x0 j2 1 2nO .x0 x0 / 2
jx x0 j 1 C nO C C C : : :
jx x0 j 2 jx x0 j2 8 jx x0 j
0 0 2
jx x0 j 1 jx x0 j 2
D jx x0 j 1 C cos ‚ C sin ‚ C : : :
jx x0 j 2 jx x0 j2
(5.30)
where we made a binomial expansion. We can, for the geometry just described,
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 117
. of 300
5.4. The total electric and magnetic fields at large distances from the sources j 95
x0 ˇ k jx k.x0 / .x0
ˇ ˇ
k ˇx x0 j x0 / (5.31)
x0 ˇ jx
ˇ ˇ
ˇx x0 j (5.32)
FT
distances from the bounded source volume:
1 eikjx x0 j
Z
ik.x0 / .x0 x0 / O 0/
E! .x/ d3x 0 ! .x0 /e k.x
4"0 jx x0 j2 V0
1 eikjx x0 j
Z
ik.x0 / .x0 x0 / O 0 / k.x
O 0/
C d3x 0 Œj! .x0 /e k.x
4"0 c jx x0 j2 V0
1 eikjx x0 j
Z
ik.x0 / .x0 x0 / O 0 / k.x
O 0/
C d3x 0 Œj! .x0 /e k.x
4"0 c jx x0 j2 V0
ik eikjx x0 j
Z
A
ik.x0 / .x0 x0 / O 0 / k.x
O 0/
d3x 0 Œj! .x0 /e k.x
4"0 c jx x0 j V0
(5.33a)
1 eikjx x0 j
Z
ik.x0 / .x0 x0 / O 0/
B! .x/ d3x 0 j! .x0 /e k.x
4"0 c 2 jx x0 j2 V0
(5.33b)
ik eikjx x0 j
R
Z
3 0 0 ik.x0 / .x0 x0 / O /0
d x j! .x /e k.x
4"0 c 2 jx x0 j V0
At a field (observation) point x located sufficiently far away from the source
volume V 0 such that inf jx x0 j sup jx0 x0 j, and jx x0 j sup jx0 x0 j,
we can assume that the direction of all wave vectors from the sources in V 0 are
D
O 0/ D x x0 O 0/ D x x0
k.x 0
D x[x0 k.x D x\x0 nO (5.34)
jx xj jx x0 j
where nO is the constant unit vector normal to the surface S.x0 / of a large sphere
centred on x0 and passing through the (fixed) field point x (see figure 5.1 on the
preceding page). Then formulæ (5.33) on page 95 can be further approximated
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 118
. of 300
as
1 eikjx x0 j
Z
O .x0 x0 /
E! .x/ d3x 0 ! .x0 /e ik n
nO
4"0 jx x0 j2 V0
1 eikjx x0 j
Z
O .x0 x0 /
C d3x 0 Œj! .x0 /e ik n
O nO
n
4"0 c jx x0 j2 V0
(5.35a)
1 eikjx x0 j
Z
3 0 0 O .x0 x0 /
ik n
C d x Œj! .x /e O nO
n
4"0 c jx x0 j2 V 0
ik eikjx x0 j
Z
0
d3x 0 Œj! .x0 /e ik nO .x x0 / n O nO
4"0 c jx x0 j V 0
1 eikjx x0 j
Z
0
B! .x/ 2 2
d3x 0 j! .x0 /e ik nO .x x0 / nO
4"0 c jx x0 j V 0
(5.35b)
FT
ik eikjx x0 j
Z
3 0 0 O .x0 x0 /
ik n
d x j! .x /e nO
4"0 c 2 jx x0 j V 0
which, after reordering of scalar and vector products, using the fact that nO is a
constant unit vector, can be written
1 eikjx x0 j
Z
3 0 0 ik nO .x0 x0 /
E! .x/ d x ! .x /e nO
4"0 jx x0 j2 V0
eikjx x0 j
Z
1 3 0 0 O .x0 x0 /
ik n
C d x j! .x /e nO nO
4"0 c jx x0 j2 V0
eikjx x0 j
A
Z
1 3 0 0 O .x0 x0 /
ik n
C d x j ! .x /e O
n nO
4"0 c jx x0 j2 V0
ik eikjx x0 j
Z
0
d3x 0 j! .x0 /e ik nO .x x0 / nO nO
4"0 c jx x0 j V0
(5.36a)
R
eikjx x0 j
1
Z
0
B! .x/ d3x 0 j! .x0 /e ik nO .x x0 / nO
4"0 c jx x0 j2
2
V 0
(5.36b)
ik eikjx x0 j
Z
3 0 0 O .x0 x0 /
ik n
d x j! .x /e nO
4"0 c 2 jx x0 j V0
or
D
1 eikjx x0 j
E! .x/ Q! .x0 / nO
4"0 jx x0 j2
1 eikjx x0 j
C I ! .x0 / nO nO
4"0 c jx x0 j2
(5.37a)
1 eikjx x0 j
C O nO
.I ! .x0 / n/
4"0 c jx x0 j2
ik eikjx x0 j
O nO
.I ! .x0 / n/
4"0 c jx x0 j
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 119
. of 300
5.4. The total electric and magnetic fields at large distances from the sources j 97
1 eikjx x0 j
B! .x/ I ! .x0 / nO
4"0 c jx x0 j2
2
(5.37b)
ik eikjx x0 j
I ! .x0 / nO
4"0 c 2 jx x0 j
where
Z
def 0
Q! .x0 / d3x 0 ! .x0 / e ik nO .x x0 /
V0
Z (5.38a)
O x0 0
De ik n
d3x 0 ! .x0 / e ik nO x
V0
and
Z
def 0
I ! .x0 / d3x 0 j! .x0 / e ik nO .x x0 /
FT
V0
Z (5.38b)
O x0 0
De ik n
d3x 0 j! .x0 / e ik nO x
V0
Inverse Fourier transforming these expressions we find that the total retarded
E and B fields very far away from a source are, to a good approximation, given
by
1
Z
E.t; x/ d3x 0 .t 0 ; x0 /nO
4"0 jx x0 j2 V 0
1
Z
A
C d3x 0 Œj.t 0 ; x0 / nO nO
4"0 c jx x0 j2 V 0
(5.39a)
1
Z
3 0 0 0
C d x Œj.t ; x / O
n O
n
4"0 c jx x0 j2 V 0
1
Z
C P 0 ; x0 / n
d3x 0 Œj.t O nO
4"0 c 2 jx x0 j V 0
R
and
1
Z
B.t; x/ d3x 0 j.t 0 ; x0 / nO
4"0 c 2 jx x0 j2 V 0
(5.39b)
1
Z
3 0 P 0 0
d x j.t ; x / nO
4"0 c 3 jx x0 j V 0
D
k jx k.x0 / .x0 x0 /
x0 j jx x0 j jx0 x0 j cos 00
t 0 .x0 / t C Dt C
! ! c c
jx0 x0 j
jx x0 j
Dt 1 cos 00 (5.40)
c jx x0 j
O 0 / and x0
where 00 is the angle between k.x x0 .
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 120
. of 300
1
Z P ret
Œj.t 0
; x0 / .x x0 / .x x0 /
far
E .t; x/ D d3x 0 (5.41a)
4"0 c 2 V0 jx x0 j 3
1
Z P ret
j.t 0
; x0 / .x x0 /
far
B .t; x/ D d3x 0 (5.41b)
4"0 c 3 V0 jx x0 j2
In the frequency (temporal Fourier) domain, these far fields are represented ex-
actly by
Efar
! .x/
B!far .x/
D
D
ik
ik
ik
1
4"0 c
1
4"0 c
1
Z
Z
FT
V0
V0
d3x 0
d3x 0
d3x 0
Œj! .x0 /eikjx
e
jx
ikjx x0 j
x0 j
j! .x0 /eikjx
x0 j
jx
x0 j
.x
x0 j3
O 0 / k.x
Œj! .x0 / k.x
.x
O 0/
x0 / .x
x0 /
x0 /
(5.42a)
A
4"0 c 2 V0 jx x0 j 2 (5.42b)
ikjx x0 j
1
Z
e O 0/
D ik d3x 0 j! .x0 / k.x
4"0 c 2 V0 jx x0 j
respectively.
R
Within the approximation (5.31), the expressions (5.42) above for the far
fields can be simplified to
1 eikjx x0 j
Efar
! .x/ ik
4"0 c jx x0 j
(5.43a)
D
Z
0 0
d3x 0 Œj! .x0 /e ik.x / .x x0 / O 0 / k.x
k.x O 0/
V0
1 eikjx x0 j
B!far .x/ ik 2
4"0 c jx x0 j
Z (5.43b)
0 0
d3x 0 Œj! .x0 /e ik.x / .x x0 / O /
k.x 0
V0
Assuming that also the paraxial approximation (5.34) is applicable, the ap-
proximate expressions (5.43) above for the far fields can be further simplified to
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 121
. of 300
5.4. The total electric and magnetic fields at large distances from the sources j 99
1 eikjx x0 j
Z
3 0 0 ik .x0 x0 /
Efar
! .x/ ik d x Œj! .x /e nO nO
4"0 c jx x0 j V0
(5.44a)
ikjx x0 j Z
1 e ik .x0 x0 /
B!far .x/ ik 2
d3x 0 Œj! .x0 /e nO (5.44b)
4"0 c jx x0 j V0
We see that at very large distances r D jx x0 j the fields fall off as 1=rare, are
O and mutually orthogonal.
purely transverse (perpendicular to n)
FT
A
R
D
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 122
. of 300
5.5 Examples
.t 0 ; x0 / x0 x
‚.t 0 ; x0 ; x/ D
4"0 jx x0 j 2x0 j
jx
x x0 x x0
1 0 0
C j.t ; x /
0
4"0 c jx x j 2 jx x j jx x0 j
0
(5.45)
x x0 x0
1 0 0 x
C j.t ; x /
FT
0 2 jx x 0j jx x0 j
4"0 c jx x j
1
0 x0
C P 0 ; x0 / x x
j.t
x
4"0 c 2 jx x0 j jx x0 j jx x0 j
where
x0 ˇ
ˇ ˇ
ˇx
t 0 D tret
0
Dt (5.46)
c
one can, for non-moving sources, express the infinitesimal differential fields as
1 1 0 0
‚.t 0 ; x0 ; x/ D .t 0 0
; x / C j.t ; x / O
r 0
rO 0
4"0 r 02 c
(5.49)
1 1 0 0 0 1P 0 0 0
C j.t ; x / O
r C j.t ; x / O
r rO 0
4"0 cr 0 r 0 c
where
r0
t 0 D tret
0
Dt (5.50)
c
5.6 Bibliography
[27] M. A. H EALD AND J. B. M ARION, Classical Electromagnetic Radiation, third ed.,
Saunders College Publishing, Fort Worth, . . . , 1980, ISBN 0-03-097277-9.
[29] J. D. JACKSON, Classical Electrodynamics, third ed., John Wiley & Sons, Inc.,
New York, NY . . . , 1999, ISBN 0-471-30932-X.
FT
[31] L. D. L ANDAU AND E. M. L IFSHITZ, The Classical Theory of Fields, fourth revised
English ed., vol. 2 of Course of Theoretical Physics, Pergamon Press, Ltd., Oxford . . . ,
1975, ISBN 0-08-025072-6.
FT
A
R
D
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 125
. of 300
In chapter 3 we were able to derive general expressions for the scalar and vector
potentials from which we later (in chapter 5) derived exact analytic expressions
of general validity for the total electric and magnetic fields generated by com-
pletely arbitrary given distributions of charge and current sources that are loc-
FT
ated within a certain region in space. The only limitation in the calculation of
the fields was that the advanced potentials were discarded on — admittedly not
totally convincing — physical grounds.
In chapter 4 we showed that the electromagnetic energy, linear momentum,
and angular momentum are all conserved quantities and in this chapter we will
show explicitly that these quantities can be radiated all the way to infinity and
therefore be used for remote studies of physical processes in Nature, even on
cosmic scales as in astronomy, and for wireless communications over long dis-
tances. Radiation processes are irreversible in that the radiated quantity in ques-
A
tion, produced by changes in the dynamical state of the sources (the charge and
current densities), does not return to the radiator but is lost from it forever.1 1This is referred to as time arrow
Other charges and currents that are located in free space, e.g. the electrons in of radiation.
an antenna, possibly very far away from the sources, are affected by the radi-
ated quantity and, as a result, change their dynamical states. These changes can
R
be measured and the embedded information decoded. This is precisely what
makes it possible for our eyes to observe light, and even more so our telescope
to observe and analyse optical and radio signals, from extremely distant stars
and galaxies. This consequence of Maxwell’s equations, with the displacement
current included, was verified experimentally by H E I N R I C H R U D O L F H E RT Z
D
about twenty years after Maxwell had made his theoretical predictions. Hertz’s
experimental and theoretical studies, followed by the demonstration of the prac-
tical use of radio communication by G U G L I E L M O M A R C O N I paved the way
for radio and TV broadcasting, radar, wireless communications, radio astronomy
and a host of other applications and technologies.
Thus, one can, use the results in chapter 5 to calculate the flux of energy,
linear momentum and angular momentum, as well as other electromagnetic ob-
servables generated by arbitrary charge and current distributions in sources at
rest (no bulk motion) at any time at any point in space. However, in practice it
103
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 126
. of 300
is often difficult to evaluate the source integrals, at least analytically, unless the
charge and current densities have a simple distribution in space. In the general
case, one has to resort to approximations. We shall consider both these situations
in this chapter.
FT
field is wholly located within two concentric spheres of radii r0 D ct0 and r0 C
r D ct0 C ct, respectively. The total linear momentum carried by this
electromagnetic pulse (signal) is the volume integral of the linear momentum
density gfield D S=c 2 D "0 E B [cf. equation (4.45) on page 62]. In spherical
polar coordinates with the origin chosen at x0 this volume integral becomes
Z Z Z 2 Z r0 Cr
2 field
pfield
D drd r g D "0 d sin d' dr r 2 .E B/
V 0 0 r0
(6.1)
In chapter 5 we derived the two expressions (5.44) on page 99 for the E
A
and B fields, respectively. These expressions show that at large distances r 0 D
jx0 x0 j from the source, the leading order contributions to these fields are
purely transverse, mutually orthogonal, and fall off as 1=r. As a result, at
large distances r, the dominating component of the linear momentum density
gfield D Sfar =c 2 D "0 Efar Bfar , which is purely radial and falls off as 1=r 2 .
R
The total linear momentum pfield is obtained when gfield is integrated over
a large spherical shell (centred on the source) of width dr D cdt , where dt
is the short duration of the signal, and for which the directed area element is
d2x nO D r 2 d rO D r 2 sin d d' rO [cf. formula (F.18) on page 217]. But the
total integrated power, as given by the surface integral S d2x nO S in equation
H
D
(4.33) on page 60, tends to a constant at infinity, showing that energy U field
and electromagnetic linear momentum pfield is carried all the way to infinity
and is irreversibly lost there. This is the physical foundation of the well-known
fact that pfield and U field can be transmitted over extremely long distances. The
force action on charges in one region in space can therefore cause a force ac-
tion on charges in a another region in space; see equation (4.50) on page 63.
Today’s wireless communication technology, be it classical or quantal, is based
almost exclusively on the utilisation of this translational degree of freedom of
the charges (currents) and the fields.
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 127
. of 300
We also note from figure 5.1 on page 94 that kO and nO are nearly parallel. Hence,
FT
we can approximate
kO d2x nO d2x
kO nO d kO nO d (6.3)
jx x0 j2 jx x0 j2
Using the far-field approximations (??) for the fields and the fact that 1=c D
p
"0 0 , and also introducing the characteristic impedance of vacuum
r
def 0
R0 376:7 (6.4)
"0
A
we obtain
ˇZ ˇ2
˝ far ˛ 1 1 ˇ d3x 0 .j! k/e
ˇ ik .x0 x0 / ˇ
ˇ
S tD
32 2
R0 2
jx x0 j ˇ V 0 ˇ nO (6.5)
Consequently, the amount of power per unit solid angle d that flows across an
R
infinitesimal surface element r 2 d D jx x0 j2 d of a large spherical shell
with its origin at x0 and enclosing all sources, is
ˇZ ˇ2
dP 1 ˇ 3 0 ik .x0 x0 / ˇ
ˇ
D R 0
ˇ d x .j ! k/e (6.6)
d 32 2 ˇ 0
V
ˇ
D
This formula is valid far away from the sources and shows that the radiated
power is given by an expression which is resistance (R0 ) times the square of
the supplied current (the integrated current density j! ), as expected. We note
that the emitted power is independent of distance r D jx x0 j and is therefore
carried all the way to infinity. The possibility to transmit electromagnetic power
over large distances, even in empty space, is the physical foundation for the
extremely important wireless communications technology. Besides determining
the strength of the radiated power, the integral in formula (6.6) also determines
its angular distribution.
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 128
. of 300
If we carry out the temporal integration first and use the fact that
Z 1
FT
0
dt e i.!C! /t D 2ı.! C ! 0 / (6.8)
1
over the area S of a large sphere which encloses the source volume V 0 , is
r 1
ˇZ ˇ2
1 0 j! k ikjx
I Z
x0 j ˇ O
d2x nO d! ˇˇ d3x 0
ˇ ˇ
U D e ˇ k (6.10)
4 "0 S 0 V0 jx x0 j
Inserting the approximations (5.31) and (6.3) into equation (6.10) above, in-
troducing the spectral energy density U! via the definition
Z 1
def
U d!U! (6.11)
0
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 129
. of 300
FT
Not only electromagnetic linear momentum (Poynting vector) can be radiated
from a source and transmitted over very long distances, but the same is also true
for electromagnetic angular momentum J field . Then torque action (the time rate
of change of J field ) in one region causes torque action on charges. The use of this
rotational degree of freedom of the fields has only recently been put to practical
use even if it has been known for more than a century.
After straightforward calculations, based on the results obtained in chapter 5,
one finds that the complete cycle averaged far-zone expression for a frequency
component ! of the electromagnetic angular momentum density generated by
A
arbitrary charge and current sources can be approximated by
nO Re .cq C In /IP
˚
˝ field ˛ 1
h .x0 / t D
32 2 "0 c 3 c jx x0 j2
nO Re f.cq C In /I g
C (6.13)
R
jx x0 j3
where, in complex notation,
Z
I.t 0 / d3x 0 j.t 0 ; x0 / (6.14)
V0
and
D
Z
P 0/
I.t P 0 ; x0 /
d3x 0 j.t (6.15)
V0
We see that at very large distances r, the angular momentum density hfield
falls off as 1=r 2 , i.e. it has precisely the same behaviour in the far zone as the
linear momentum density and can therefore also transfer information wirelessly
over large distances. The only difference is that while the direction of the linear
momentum (Poynting vector) becomes purely radial at infinity, the angular mo-
mentum becomes perpendicular to the linear momentum, i.e. purely transverse,
there.
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 130
. of 300
FT
assumed to be located in a finite volume around a point x0 . At a field (observing)
point x at large distance jx x0 j from the source region, one can then, to a
good approximation, approximate the retarded potential by the Taylor expansion
(Einstein’s summation convention over i and j is implied); cf. example 3.1 on
page 46
"
0
1 q.tret / 1 0 x x0
ˆ.t; x/ D C 2
d.tret ; x0 /
4"0 jx x0 j jx x0 j jx x0 j
#
1 0
3 x
i x0i xj x0j 1
A
C Qij .tret ; x0 / ıij C : : : (6.16)
jx x0 j3 2 jx x0 j jx x0 j 2
where
Z
0
q.tret /D d3x 0 .tret
0
; x0 / (6.17a)
V0
R
is the total charge or electric monopole moment,
Z
0
d.tret ; x0 / D d3x 0 .x0 x0 / .tret
0
; x0 / (6.17b)
V0
with components di , i D 1; 2; 3 is the electric dipole moment vector, and
Z
0
Q.tret ; x0 / D d3x 0 .x0 x0 / ˝.x0 x0 / .tret 0
; x0 / (6.17c)
D
V0
0
Hence the transformation between t and tret is a trivial. In the subsequent ana-
0 0
lysis in this subsection we shall use t to denote this approximate tret .
For a normal medium, the major contributions to the electrostatic interactions
come from the net charge and the lowest order electric multipole moments in-
duced by the polarisation due to an applied electric field. Particularly important
is the dipole moment. Let P denote the electric dipole moment density (elec-
tric dipole moment per unit volume; unit: C m 2 ), also known as the electric
polarisation, in some medium. In analogy with the second term in the expan-
sion equation (6.16) on the facing page, the electric potential from this volume
distribution P .t; x0 / of electric dipole moments d at the source point x0 can be
written
1 x x0
Z
ˆd .t; x/ D d3x 0 P .t 0 ; x0 /
4"0 V 0 jx x0 j3
FT
1 1
Z
3 0 0 0
D d x P .t ; x / r (6.20)
4"0 V 0 jx x0 j
1 1
Z
D d3x 0 P .t 0 ; x0 / r 0
4"0 V 0 jx x0 j
Using expression (M.155a) on page 260 and applying the divergence theorem,
we can rewrite this expression for the potential as follows:
P .t 0 ; x0 / 0 0 0
Z Z
1 3 0 r P .t ; x /
ˆd .t; x/ D d3x 0 r 0 d x
4"0 V 0 jx x0 j V0 jx x0 j
(6.21)
P .t 0 ; x0 / 0 0 0
I
1 3 0 r P .t ; x /
Z
A
D d2x 0 nO 0 d x
4"0 S 0 jx x0 j V0 jx x0 j
where the first term, which describes the effects of the induced, non-cancelling
dipole moment on the surface of the volume, can be neglected, unless there is
a discontinuity in nO P at the surface. Doing so, we find that the contribution
from the electric dipole moments to the potential is given by
R
1 r 0 P .t 0 ; x0 /
Z
ˆd .t; x/ D d3x 0 (6.22)
4"0 V 0 jx x0 j
Comparing this expression with expression equation (3.35a) on page 40 for the
potential from a charge distribution .t; x/, we see that r P .t; x/ has the
characteristics of a charge density and that, to the lowest order, the effective
D
charge density becomes .t; x/ r P .t; x/, in which the second term is a
polarisation term that we call pol .t; x/.
If we adopt the same idea for the ‘true’ charge density due to free charges and
introduce a vector field .t; x/, analogous to P .t; x/, but such that
def
true r (6.24a)
which means that the associated ‘polarisation current’ now is the true current:
@
D j true (6.24b)
@t
As a consequence, the equation of continuity for ‘true’ charges and currents [cf.
expression (1.23) on page 10] is satisfied:
@true .t; x/ @ @
C r j true .t; x/ D r C r D0 (6.25)
@t @t @t
FT
The vector is called the polarisation vector because, formally, it treats also the
‘true’ (free) charges as polarisation charges. Since in the microscopic Maxwell-
Lorentz equation (2.1a) on page 19, the charge density must include all charges,
we can write this equation
true C pol r r P
r E D D D (6.26)
"0 "0 "0
i.e. in a form where all the charges are considered to be polarisation charges.
We now introduce a further potential …e with the following property
A
r …e D ˆ (6.27a)
e
1 @…
DA (6.27b)
c 2 @t
where ˆ and A are the electromagnetic scalar and vector potentials, respectively.
R
As we see, …e acts as a ‘super-potential ’ in the sense that it is a potential from
which we can obtain other potentials. It is called the Hertz vector or polarisation
potential . Requiring that the scalar and vector potentials ˆ and A, respectively,
satisfy their inhomogeneous wave equations, equations (3.15) on page 36, one
finds, using (6.24) and (6.27), that the Hertz vector must satisfy the inhomogen-
D
1 @2 e
2 …e D … r 2 …e D (6.28)
c 2 @t 2 "0
This equation is of the same type as equation (3.18) on page 37, and has
therefore the retarded solution
0
1 ; x0 /
Z
.tret
e
… .t; x/ D d3x 0 (6.29)
4"0 V 0 jx x0 j
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 133
. of 300
FT
4"0 V0 jx x0 j
Assume that the source region is a limited volume around some central point
x0 far away from the field (observation) point x illustrated in figure 6.1. Under
these assumptions, we can expand the Hertz vector, expression (6.30) above,
0
due to the presence of non-vanishing .tret ; x0 / in the vicinity of x0 , in a formal
series. For this purpose we recall from potential theory that
0 0
eikjx x j eikj.x x0 / .x x0 /j
jx x0 j j.x x0 / .x0 x0 /j
1 (6.31)
A
X
.2n C 1/Pn .cos ‚/jn .k ˇx0 x0 ˇ/h.1/ .k jx
ˇ ˇ
D ik n x0 j/
nD0
where (see figure 6.1)
0
eikjx x j
is a Green function or propagator
jx x0 j
R
‚ is the angle between x0 x0 and x x0
Pn .cos ‚/ is the Legendre polynomial of order n
jn .k ˇx0 x0 ˇ/ is the spherical Bessel function of the first kind of order n
ˇ ˇ
h.1/
n .k jx x0 j/ is the spherical Hankel function of the first kind of order n
D
where Pnm is an associated Legendre polynomial of the first kind , related to the
spherical harmonic Ynm as
s
m 2n C 1 .n m/Š m
Yn .; '/ D P .cos / eim'
4 .n C m/Š n
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 134
. of 300
C D r …e (6.34)
we see that we can calculate the electric and magnetic fields, respectively, as
follows
EDr C (6.35a)
FT
1 @C
BD (6.35b)
c 2 @t
Clearly, the first formula is valid only if r E D 0, i.e. if we are outside the
region where the sources are located. Since we are mainly interested in the
fields in the far zone, a long distance away from the source region, this is no
essential limitation.
Inserting equation (6.31) on the previous page, together with formula (6.32)
on the preceding page, into equation (6.30) on the previous page, we can in a
formally exact way expand the Fourier component of the Hertz vector as
A
1 n
ik X X
…e! D .2n C 1/. 1/m h.1/n .k jx x0 j/ Pnm .cos / eim'
4"0 nD0 mD n
Z
0
d3x 0 ! .x0 / jn .k ˇx0 x0 ˇ/ Pn m .cos 0 / e im'
ˇ ˇ
V0
R
(6.36)
k ˇx0
ˇ ˇ
x0 ˇ 1 k jx x0 j (6.37)
eikjx x0 j
h.1/
n .k jx x0 j/ . i/nC1 (6.38)
k jx x0 j
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 135
. of 300
and replace jn with the first term in its power series expansion:
2n nŠ ˇn
jn .k ˇx0 k ˇx 0
ˇ ˇ ˇ
x 0 ˇ/ x0 ˇ (6.39)
.2n C 1/Š
Inserting these expansions into equation (6.36) on the facing page, we obtain the
multipole expansion of the Fourier component of the Hertz vector
1
X
…e! …e!.n/ (6.40a)
nD0
where
1 eikjx x0 j 2n nŠ
…e!.n/ D . i/n
4"0 jx x0 j .2n/Š
FT
Z
d3x 0 ! .x0 / .k ˇx0 x0 ˇ/n Pn .cos ‚/ (6.40b)
ˇ ˇ
V0
d rO
x2
'
x1
def
FT
[cf. equation (6.17b) on page 108]. If a spherical coordinate system is chosen
with its polar axis along d! as in figure 6.2, the components of …e!.0/ are
def
…er …e!.0/ rO D
…e …e!.0/ ™O D
1 eikjx x0 j
4"0 jx x0 j
d! cos
1 eikjx x0 j
d! sin
(6.43a)
(6.43b)
A
4"0 jx x0 j
def
…e' …e!.0/ ®
O D0 (6.43c)
Evaluating formula (6.34) on page 112 for the help vector C, with the spher-
ically polar components (6.43) of …e!.0/ inserted, we obtain
R
ikjx x0 j
.0/ 1 1 e
C! D C!;' ® O D ik O
d! sin ® (6.44)
4"0 jx x0 j jx x0 j
Applying this to equations (6.35) on page 112, we obtain directly the Fourier
components of the fields
"
D
1 1 ik x x0
E! D 2 2
cos
4"0 jx x0 j jx x0 j jx x0 j
# (6.45a)
ikjx x0 j
1 ik e
C k sin ™O
2
d!
jx x0 j2 jx x0 j jx x0 j
ikjx x0 j
!0 1 e
B! D i ik O
d! sin ® (6.45b)
4 jx x0 j jx x0 j
Keeping only those parts of the fields which dominate at large distances (the
radiation fields) and recalling that the wave vector k D k.x x0 /= jx x0 j
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 137
. of 300
where k D !=c, we can now write down the Fourier components of the radiation
parts of the magnetic and electric fields from the dipole:
1 eikjx x0 j 1 eikjx x0 j
Efar
! D d! k 2 sin ™O D Œ.d! k/ k
4"0 jx x0 j 4"0 jx x0 j
(6.46a)
!0 eikjx x0 j !0 eikjx x0 j
B!far D O D
d! k sin ® .d! k/ (6.46b)
4 jx x0 j 4 jx x0 j
These fields constitute the electric dipole radiation, also known as E1 radiation.
FT
The next term in the expression (6.40b) on page 113 for the expansion of the
Fourier transform of the Hertz vector is for n D 1:
eikjx x0 j
Z
…e!.1/ D i d3x 0 k ˇx0 x0 ˇ ! .x0 / cos ‚
ˇ ˇ
4"0 jx x0 j V 0
(6.47)
1 eikjx x0 j
Z
3 0 0 0
D ik d x Œ.x x0 / .x x0 / ! .x /
4"0 jx x0 j2 V 0
i D xi x0;i (6.49a)
0i D xi0 x0;i (6.49b)
R
the j th component of the integrand in …e! .1/ can be broken up into
1
x0 / .x0 x0 / ! .x0 /gj D i !;j 0i C !;i j0
fŒ.x
2 (6.50)
1
C i !;j 0i !;i j0
2
D
i.e. as the sum of two parts, the first being symmetric and the second antisym-
metric in the indices i; j . We note that the antisymmetric part can be written
as
1 1
i !;j 0i !;i j0 D Œ!;j .i 0i / j0 .i !;i /
2 2
1
D Œ! . 0 / 0 . ! /j (6.51)
2
1˚
.x x0 / Œ! .x0 x0 / j
D
2
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 138
. of 300
The utilisation of equations (6.24) on page 110, and the fact that we are
considering a single Fourier component,
i!t
.t; x/ D ! e (6.52)
allow us to express ! in j! as
j!
! D i (6.53)
!
Hence, we can write the antisymmetric part of the integral in formula (6.47) on
the preceding page as
1
Z
.x x0 / d3x 0 ! .x0 / .x0 x0 /
2 V 0
1
Z
FT
D i .x x0 / d3x 0 j! .x0 / .x0 x0 / (6.54)
2! V 0
1
D i .x x0 / m!
!
where we introduced the Fourier transform of the magnetic dipole moment
1
Z
m! D d3x 0 .x0 x0 / j! .x0 / (6.55)
2 V0
The final result is that the antisymmetric, magnetic dipole, part of …e!.1/ can
be written
A
.1/ k eikjx x0 j
…e,! antisym D .x x0 / m! (6.56)
4"0 ! jx x0 j2
In analogy with the electric dipole case, we insert this expression into equation
(6.34) on page 112 to evaluate C, with which equations (6.35) on page 112 then
R
gives the B and E fields. Discarding, as before, all terms belonging to the near
fields and transition fields and keeping only the terms that dominate at large
distances, we obtain
k eikjx x0 j
Efar
! .x/ D m! k (6.57a)
4"0 c jx x0 j
D
0 eikjx x0 j
B!far .x/ D .m! k/ k (6.57b)
4 jx x0 j
which are the fields of the magnetic dipole radiation (M1 radiation).
Again we use this expression in equation (6.34) on page 112 to calculate the
fields via equations (6.35) on page 112. Tedious, but fairly straightforward al-
gebra (which we will not present here), yields the resulting fields. The compon-
ents of the fields that dominate in the far field zone (wave zone) are given by
i0 ! eikjx x0 j
B!far .x/ D .k Q! / k (6.59a)
8 jx x0 j
i eikjx x0 j
FT
Efar
! .x/ D Œ.k Q! / k k (6.59b)
8"0 jx x0 j
This type of radiation is called electric quadrupole radiation or E2 radiation.
axis, and the azimuthal angle ' is calculated relative to the direction of positive
x axis toward the positive y axis in a right-handed sense. In this polar coordinate
system the observation (field) point is located at x D .r; ; '/ with r D jx x0 j.
The origin is chosen at x0 so that x x0 D x D r D r rO and the wave vector
k D k kO D k rO .
The Fourier amplitude of the far-zone electric field generated by the antenna
current density
i!t 0
j.t 0 ; x0 / D j! .x0 /e (6.60)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 140
. of 300
is
1 eikjx x0 j
Z
0
Efar
! .x/ D i rO d3x 0 e ik .x x0 / j! .x0 / k.x0 /
4"0 c jx x0 j V0
ikjx x0 j Z
(6.61)
1 e 3 0 ik .x0 x0 / 0 0
B!far .x/ D i dx e j! .x / k.x /
4"0 c 2 jx x0 j V 0
In the paraxial approximation described by equation (5.34) on page 95 one as-
sumes that the distant observer views all parts of the antenna under one and the
same fixed polar angle , i.e. that all wave vectors has the same constant direc-
tion as that coming from the midpoint x0 D 0. In this approximation
i eikr
Z
far 3 0 ik x0 0 (6.62)
E! D rO k dx e j! .x /
4"0 c r V0
FT
where j! .x0 / k D k j! .x0 / was used. In the chosen geometry, the 1D
antenna current flows along zO :
Let us apply equation (6.6) on page 105 to calculate the radiated EM power
from a one-dimensional, time-varying current. Such a current can be set up
by feeding the EMF of a generator (e.g. a transmitter) onto a stationary, linear,
straight, thin, conducting wire across a very short gap at its centre. Due to the
applied EMF, the charges in this thin wire of finite length L are set into linear
motion to produce a time-varying antenna current which is the source of the
EM radiation. Linear antennas of this type are called dipole antennas. For sim-
A
plicity, we assume that the conductor resistance and the energy loss due to the
electromagnetic radiation are negligible.
Choosing our coordinate system such that the x3 axis is along the antenna
axis, the antenna current density can be represented, in complex notation, by
O 3 (measured in Am 2) where J.t 0 ; x30 / is the
j.t 0 ; x0 / D ı.x10 /ı.x20 /J.t 0 ; x30 / x
current (measured in A) along the antenna wire. Since we can assume that the
R
antenna wire is infinitely thin, the antenna current must vanish at the endpoints
L=2 and L=2. At the midpoint, where the antenna is fed across a very short gap
in the conducting wire, the antenna current is, of course, equal to the supplied
current.
For each Fourier frequency component !0 , the antenna current J.t 0 ; x30 / can
D
be written as I.x30 / exp. i!0 t 0 / so that the antenna current density can be written
as j.t 0 ; x0 / D j0 .x0 / exp. i!0 t 0 / [cf. equations (5.6) on page 87] where
and where the spatially varying Fourier amplitude I.x30 / of the antenna current
fulfils the time-independent wave equation (Helmholtz equation)
d2 I
C k 2 I.x30 / D 0 ; I. L=2/ D I.L=2/ D 0 ; I.0/ D I0 (6.64)
dx302
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 141
. of 300
FT
sinŒk.L=2 ˇx30 ˇ/
ˇ ˇ
0
I.x3 / D I0 (6.65)
sin.kL=2/
where I0 is the amplitude of the antenna current (measured in A), assumed to
be constant and supplied, via a non-radiating transmission line by the gener-
ator/transmitter at the antenna feed point (in our case the midpoint of the antenna
wire) and 1= sin.kL=2/ is a normalisation factor. The antenna current forms a
standing wave as indicated in figure 6.3.2 2 This rather accurate model of the
When L is much smaller than the wavelength , we can approximate the antenna current was introduced
in 1987 by H E N RY C A B O U R N
A
current distribution formula (6.65) by the first term in its Taylor expansion: P O C K L I N G T O N (1870–1952).
Hence, in the most general case of a straight, infinitely thin antenna of fi-
nite, arbitrary length L directed along the x30 axis, the Fourier amplitude of the
R
antenna current density is
ˇ 0ˇ
ˇx ˇ/
0 0 0 sinŒk.L=2 3
j0 .x / D I0 ı.x1 /ı.x2 / O3
x (6.67)
sin.kL=2/
For a half-wave dipole antenna (L D =2), the antenna current density is
(6.68)
In the case of a travelling wave antenna, in which one end of the antenna is
connected to ground via a resistance so that the current at this end does not
vanish, the Fourier amplitude of the antenna current density is
0 kO
j! .x /
x2
'
x1
FT L
2
In order to evaluate formula (6.6) on page 105 with the explicit monochro-
matic current (6.67) inserted, we use a spherical polar coordinate system as in
A
figure 6.4 to evaluate the source integral
ˇZ ˇ2
3 0 ik .x0 x0 / ˇ
ˇ ˇ
I D ˇ d x j0 k e
ˇ
V0
ˇ
ˇZ L ˇ2
sinŒk.L=2 ˇx30 ˇ/
ˇ ˇ
R
ˇ 2 ˇ
0 ikx30 cos ikx0 cos 0 ˇ
D ˇ dx3 I0 k sin e e
ˇ
ˇ
ˇ L2 sin.kL=2/ ˇ
ˇ Z L ˇ2
2 2
k sin ˇ ˇ2 ˇ 2
ˇ
D I02 2 L ˇeikx0 cos 0 ˇ ˇ2 dx30 sin.kL=2 kx30 / cos.kx30 cos /ˇ
ˇ ˇ ˇ ˇ
sin .k 2 / ˇ 0 ˇ
cosŒ.kL=2/ cos cos.kL=2/ 2
D
D 4I02
sin sin.kL=2/
(6.71)
Inserting this expression and d D 2 sin d into formula (6.6) on page 105
and integrating over , we find that the total radiated power from the antenna is
2
1 cosŒ.kL=2/ cos cos.kL=2/
Z
P .L/ D R0 I02 d sin (6.72)
4 0 sin sin.kL=2/
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 143
. of 300
FT
The integral in (6.75) can always be evaluated numerically. But, it can in fact
also be evaluated analytically as follows:
cos2 2 cos cos2 2 u
Z Z 1
d D Œcos 7! u D du D
0 sin 1 1 u2
1 C cos.u/
cos2 u D
2 2
Z 1
1 1 C cos.u/
D du
2 1 .1 C u/.1 u/
1 1 1 C cos.u/ 1 1 1 C cos.u/
A
Z Z
D du C du
4 1 .1 C u/ 4 1 .1 u/
1 1 1 C cos.u/ vi
Z h
D du D 1 C u 7!
2 1 .1 C u/
Z 2
1 1 cos v 1
D dv D Œ
C ln 2 Ci.2/
R
2 0 v 2
1:22
(6.76)
where in the last step the Euler-Mascheroni constant
D 0:5772 : : : and the co-
sine integral Ci.x/ were introduced. Inserting this into the expression equation
D
1111
0000
creases, with velocity c D c.x
x0 /= jx x0 j outward from the
0000
1111
centre. The source charge element
moves with an arbitrary velocity v 0
and gives rise to a source ‘leakage’
111111111
000000000
out of the volume dV 0 D d3x 0 . dr 0
x.t /
00000
11111
v 0 .t 0 /
000000000
111111111
x
00000
11111
x0 dS0
000000000
111111111
dV 0
0 0
x .t /
0000
1111
dq 0
FT 0000
1111 c
and consider a source region with such a limited spatial extent that the charges
and currents are well localised. Specifically, we consider a charge q 0 , for in-
stance an electron, which, classically, can be thought of as a localised, unstruc-
D
tured and rigid ‘charge distribution’ with a small, finite radius. The part of this
‘charge distribution’ dq 0 which we are considering is located in dV 0 D d3x 0
in the sphere in figure 6.5. Since we assume that the electron (or any other
similar electric charge) moves with a velocity v 0 whose direction is arbitrary
and whose magnitude can even be comparable to the speed of light, we can-
not say that the charge and current to be used in (6.77) is V 0 d3x 0 .tret 0
; x0 / and
R
R 3 0 0 0
V 0 d x v .tret ; x /, respectively, because in the finite time interval during which
the observed signal is generated, part of the charge distribution will ‘leak’ out of
the volume element d3x 0 .
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 145
. of 300
FT
dq 0 D .tret
0
; x0 / d3x 0 .tret
0
; x0 / dx
c jx x0 j
(6.79)
.x x0 / v 0
0
D .tret ; x0 / 1 d3x 0
c jx x0 j
or
0 dq 0
.tret ; x0 / d3x 0 D .x x0 / v 0
(6.80)
1 cjx x0 j
suming that the integrands do not change sign in the integration volume, use the
mean value theorem of calculus to evaluate these expressions to become
1 1 q0 1
Z
0
ˆ.t; x/ D 0 0 dq D (6.83a)
4"0 jx x0 j .x x / v 4"0 s
c
1 v0 q0 v0
Z
0
A.t; x/ D 0 0 dq D
4"0 c 2 jx x0 j .x x / v 4"0 c 2 s
c (6.83b)
v0
D 2ˆ
c
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 146
. of 300
where
s D s.t 0 ; x/ D ˇx
ˇ
ˇ
D ˇx
FT
x0 .t 0 /ˇ
0 0 ˇ
x .t / 1
ˇ
ˇ
Œx
x
x0 .t 0 / v 0 .t 0 /
x
c
.t / v 0 .t 0 /
0 0
jx x0 .t 0 /j
0 0
x x .t /
0 0
v .t /
c
(6.84a)
(6.84b)
A
D Œx x0 .t 0 / 0 0
(6.84c)
jx x .t /j c
3 These results were derived is the retarded relative distance. The potentials (6.83) are the Liénard-Wiechert
independently by A L F R E D - potentials.3 In section 7.3.2 on page 167 we shall derive them in a more elegant
M A R I E L I É NA R D (1869–1958)
R
in 1898 and E M I L J O H A N N and general way by using a relativistically covariant formalism.
W I E C H E RT (1861–1928) in It should be noted that in the complicated derivation presented above, the
1900. When v 0 k .x x0 / and
v ! c, the potentials become observer is in a coordinate system that has an ‘absolute’ meaning and the velocity
singular. This was first pointed v 0 is that of the localised charge q 0 , whereas, as we shall see later in the covariant
out by A R N O L D J O H A N N E S
WILHELM SOMMERFELD derivation, two reference frames of equal standing are moving relative to each
(1868–1951) in 1904. other with v 0 .
D
x0 D x0 .t 0 / (6.86)
(in the interest of simplifying our notation, we drop the subscript ‘ret’ on t 0 from
now on). This means that we know the trajectory of the charge q 0 , i.e. x0 , for all
times up to the time t 0 at which a signal was emitted in order to precisely arrive
at the field point x at time t . Because of the finite speed of propagation of the
fields, the trajectory at times later than t 0 cannot be known at time t .
The retarded velocity and acceleration at time t 0 are given by
FT
dx0
v 0 .t 0 / D (6.87a)
dt 0
dv 0 d2 x0
a0 .t 0 / DD 0 D 0 2 (6.87b)
dt dt
As for the charge coordinate x0 itself, we have in general no knowledge of the
velocity and acceleration at times later than t 0 , and definitely not at the time of
observation t ! If we choose the field point x as fixed, the application of (6.87)
to the relative vector x x0 yields
A
d
Œx x0 .t 0 / D v 0 .t 0 / (6.88a)
dt 0
d2
Œx x0 .t 0 / D a0 .t 0 / (6.88b)
dt 0 2
The retarded time t 0 can, at least in principle, be calculated from the implicit
R
relation
jx x0 .t 0 /j
t 0 D t 0 .t; x/ D t (6.89)
c
and we shall see later how this relation can be taken into account in the calcula-
D
tions.
According to formulæ (6.85) on the facing page, the electric and magnetic
fields are determined via differentiation of the retarded potentials at the observa-
tion time t and at the observation point x. In these formulæ the unprimed r , i.e.
the spatial derivative differentiation operator r D x O i @=@xi means that we dif-
ferentiate with respect to the coordinates x D .x1 ; x2 ; x3 / while keeping t fixed,
and the unprimed time derivative operator @=@t means that we differentiate with
respect to t while keeping x fixed. But the Liénard-Wiechert potentials ˆ and A,
equations (6.83) on page 123, are expressed in the charge velocity v 0 .t 0 / given
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 148
. of 300
by equation (6.87a) on page 125 and the retarded relative distance s.t 0 ; x/ given
by equation (6.84) on page 124. This means that the expressions for the po-
tentials ˆ and A contain terms that are expressed explicitly in t 0 , which in turn
is expressed implicitly in t via equation (6.89) on the previous page. Despite
this complication, it is possible, as we shall see below, to determine the electric
and magnetic fields and associated quantities at the time of observation t . To
this end, we need to investigate meticulously the action of differentiation on the
potentials.
FT
and t , respectively. With this convention, we find that
0
@ ˇx x0 .t 0 /ˇp D p x x
ˇ ˇ @ 0 0
2 p
x x .t /
@t 0 x jx x0 j @t 0 x
0 0 0 (6.90)
.x x / v .t /
D p
jx x0 j2 p
Furthermore, by applying the operator .@=@t/x to equation (6.89) on the preced-
ing page we find that
0
jx x0 .t 0 .t; x//j
@t @
D1
@t x @t c
A
x
jx x0 j
0
@ @t
D1 0
(6.91)
@t x c @t x
.x x0 / v 0 .t 0 / @t 0
D1C
c jx x0 j @t x
This is an algebraic equation in .@t 0 =@t/x that we can solve to obtain
R
0
@t jx x0 j jx x0 j
D 0 0 0 0
D (6.92)
@t x jx x j .x x / v .t /=c s
where s D s.t 0 ; x/ is the retarded relative distance given by equation (6.84) on
page 124. Making use of equation (6.92) above, we obtain the following useful
operator identity
D
0
jx x0 j @
@ @t @
D D (6.93)
@t x @t x @t 0 x s @t 0 x
Likewise, by applying .r / t to equation (6.89) on the previous page we ob-
tain
jx x0 .t 0 .t; x//j x x0
0
r .x x0 / t
.r t / t D r D 0
c t c jx x j
(6.94)
x x0 .x x0 / v 0 .t 0 / 0
D C .r t / t
c jx x0 j c jx x0 j
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 149
. of 300
FT
4"0 s 2 jx x0 j c cs @t 0 x
and, from equation (6.83b) on page 123 and equation (6.93) on the preceding
page,
@ 0 q 0 v 0 .t 0 /
@A @A
D
@t @t x @t 4 s x
0 (6.97b)
q ˇx x ˇ sa .t / ˇx x0 ˇ v 0 .t 0 / @s
ˇ 0
ˇ 0 0
ˇ ˇ
D
4"0 c 2 s 3 @t 0 x
Utilising these relations in the calculation of the E field from the Liénard-Wiechert
potentials, equations (6.83) on page 123, we obtain
A
@
E.t; x/ D r ˆ.t; x/ A.t; x/
@t
"
q0 Œx x0 .t 0 / jx x0 .t 0 /j v 0 .t 0 /=c
D
4"0 s 2 .t 0 ; x/ jx x0 .t 0 /j
(6.98)
Œx x0 .t 0 / jx x0 .t 0 /j v 0 .t 0 /=c @s.t 0 ; x/
R
cs.t 0 ; x/ @t 0 x
#
0 0 0 0
jx x .t /j a .t /
c2
Starting from expression (6.84a) on page 124 for the retarded relative distance
s.t 0 ; x/, we see that we can evaluate .@s=@t 0 /x in the following way
D
ˇ .x x0 / v 0 .t 0 /
@s @ ˇˇ 0ˇ
D x x
@t 0 x @t 0 c x
@ ˇ ˇ 0 0 ˇ
ˇ
D x x .t /
@t 0 x
0 0 (6.99)
1 @Œx x0 .t 0 /
0 0 0 0 @v .t /
v .t / C Œx x .t /
c @t 0 x @t 0 x
0 0 0 0 02 0 0 0 0 0
.x x .t // v .t / v .t / .x x .t // a .t /
D C
jx x0 .t 0 /j c c
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 150
. of 300
where equation (6.90) on page 126 and equations (6.87) on page 125, respect-
ively, were used. Hence, the electric field generated by an arbitrarily moving
localised charge at x0 .t 0 / is given by the expression
q0 x0 .t 0 /j v 0 .t 0 / v 02 .t 0 /
jx
E.t; x/ D Œx x0 .t 0 / 1
4"0 s 3 .t 0 ; x/ c c2
„ ƒ‚ …
Velocity field (tends to the Coulomb field when v ! 0)
( )
q0 x x0 .t 0 / x0 .t 0 /j v 0 .t 0 /
0 0 jx 0 0
C Œx x .t / a .t /
4"0 s 3 .t 0 ; x/ c2 c
„ ƒ‚ …
Acceleration (radiation) field
(6.100)
The first part of the field, the velocity field , tends to the ordinary Coulomb field
FT
when v 0 ! 0 and does not contribute to the radiation. The second part of the
field, the acceleration field , is radiated into the far zone and is therefore also
called the radiation field .
From figure 6.6 on page 124 we see that the position the charged particle
would have had if at t 0 all external forces would have been switched off so that
the trajectory from then on would have been a straight line in the direction of
the tangent at x0 .t 0 / is x0 .t/, the virtual simultaneous coordinate. During the
arbitrary motion, we interpret x x0 .t/ as the coordinate of the field point x
relative to the virtual simultaneous coordinate x0 .t/. Since the time it takes for
A
a signal to propagate (in the assumed free space) from x0 .t 0 / to x is jx x0 j =c,
this relative vector is given by
jx x0 .t 0 /j v 0 .t 0 /
x x0 .t/ D x x0 .t 0 / (6.101)
c
R
This allows us to rewrite equation (6.100) above in the following way
"
q0 v 02 .t 0 /
E.t; x/ D x x 0 .t/ 1
4"0 s 3 c2
# (6.102)
0 0
x x 0 .t/ a .t /
C x x0 .t 0 /
D
c2
x0
x @A
B.t; x/ D r A.t; x/ r A t D r A t 0
cs @t 0 x
q0 x0 x x0
x @A
D 2 2
v0
4"0 c s jx x0 j c jx x0 j @t x
(6.103)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 151
. of 300
where we made use of equation (6.83) on page 123 and formula (6.93) on
page 126. But, according to (6.97a),
x x0 q0 x x0
v0
0
r ˆ t
D 2 2
(6.104)
c jx x j 4"0 c s jx x0 j
so that
x x0
@A
B.t; x/ D 0
rˆ t
c jx x j @t x
(6.105)
x x0 .t 0 /
D E.t; x/
c jx x0 .t 0 /j
The electric far field is obtained from the acceleration field in formula (6.100)
on the facing page as
FT
q0 jx x0 j v 0
0 0 0
Efar .t; x/ D .x x / .x x / a
4"0 c 2 s 3 c
0
(6.106)
q 0 0 0 0
D Œx x .t / fŒx x 0 .t/ a .t /g
4"0 c 2 s 3
where in the last step we again used formula (6.101) on the preceding page.
Combining this formula and formula (6.105), the radiation part of the magnetic
field can be written
x x0 .t 0 /
Bfar .t; x/ D Efar .t; x/ (6.107)
c jx x0 .t 0 /j
A
6.5.2.2 The direct method
An alternative to the differential operator transformation technique just described
is to try to express all quantities in the potentials directly in t and x. An example
R
of such a quantity is the retarded relative distance s.t 0 ; x/. According to equation
(6.84) on page 124, the square of this retarded relative distance can be written
ˇ2 ˇ Œx x0 .t 0 / v 0 .t 0 /
s 2 .t 0 ; x/ D ˇx x0 .t 0 /ˇ x0 .t 0 /ˇ
ˇ ˇ
2 ˇx
c
(6.108)
x0 .t 0 / v 0 .t 0 / 2
Œx
C
D
c
If we use identity (F.37) on page 220 we find that
.x x0 / v 0 2 jx x0 j2 v 02
2
x0 / v 0
.x
D 2
(6.109)
c c c
Furthermore, from equation (6.101) on the preceding page, we obtain the iden-
tity
Œx x0 .t 0 / v 0 D Œx x0 .t/ v 0 (6.110)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 152
. of 300
which, when inserted into equation (6.109) on the preceding page, yields the
relation
.x x0 / v 0 2 jx x0 j2 v 02 .x x0 / v 0 2
D (6.111)
c c2 c
Inserting the above into expression (6.108) on the previous page for s 2 , this
expression becomes
ˇ .x x0 / v 0 jx x0 j2 v 02 .x x0 / v 0 2
ˇ2
s 2 D ˇx x 0 ˇ 2 ˇx x0 ˇ
ˇ ˇ
C
c c2 c
0 0 2
0 2
jx x j v .x x0 / v
D .x x0 /
c c
0 2
.x x0 / v
D .x x0 /2
FT
c
Œx x0 .t/ v 0 .t 0 / 2
jx x0 .t /j2
c
(6.112)
where in the penultimate step we used equation (6.101) on page 128.
What we have just demonstrated is that if the particle velocity at time t can
be calculated or projected from its value at the retarded time t 0 , the retarded
distance s in the Liénard-Wiechert potentials (6.83) can be expressed in terms
of the virtual simultaneous coordinate x0 .t/, viz., the point at which the particle
A
will have arrived at time t , i.e. when we obtain the first knowledge of its existence
at the source point x0 at the retarded time t 0 , and in the field coordinate x D
x.t /, where we make our observations. We have, in other words, shown that all
quantities in the definition of s, and hence s itself, can, when the motion of the
charge is somehow known, be expressed in terms of the time t alone. I.e. in this
R
special case we are able to express the retarded relative distance as s D s.t; x/
and we do not have to involve the retarded time t 0 or any transformed differential
operators in our calculations.
Taking the square root of both sides of equation (6.112) above, we obtain
the following alternative final expressions for the retarded relative distance s in
D
If we know what velocity the particle will have at time t , expression (6.113) on
the facing page for s will not be dependent on t 0 .
Using equation (6.113c) on the preceding page and standard vector analytic
formulae, we obtain
" #
v 02 .x x0 / v 0 2
2 2
r s D r jx x0 j 1 C
c2 c
v 02 v0 ˝ v0
D 2 .x x0 / 1 C .x x / (6.114)
0
c2 c2
v0
0
v
D 2 .x x0 / C .x x0 /
c c
We shall use this result in example 6.4 on page 150 for a uniform, unaccelerated
FT
motion of the charge.
.x x0 / v 0
x0 ˇ x0 ˇ ; v0 c
ˇ ˇ ˇ ˇ
s D ˇx ˇx (6.115)
c
A
and formula (6.101) on page 128
jx x0 j v 0
x x0 D .x x0 / x x0 ; v0 c (6.116)
c
so that the radiation field equation (6.106) on page 129 can be approximated by
R
q0
Efar .t; x/ D .x x0 / Œ.x x0 / a0 ; v 0 c (6.117)
4"0 c 2 jx x0 j3
from which we obtain, with the use of formula (6.105) on page 129, the magnetic
field
D
q0
Bfar .t; x/ D Œa0 .x x0 /; v0 c (6.118)
4"0 c 3 jx x0 j2
It is interesting to note the close correspondence that exists between the non-
relativistic fields (6.117) and (6.118) and the electric dipole field equations (6.46)
on page 115 if we introduce the electric dipole moment for a localised charge
[cf. formula (6.42) on page 113]
d D q 0 x0 .t 0 / (6.119)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 154
. of 300
FT
and at the same time make the transitions
q 0 a0 D dR 7! ! 2 d! (6.120a)
0
x x Dx x0 (6.120b)
The energy flux in the far zone is described by the Poynting vector as a
function of Efar and Bfar . We use the close correspondence with the dipole case
to find that it becomes
0 q 0 2 .a0 /2 x x0
Sfar D sin2 (6.121)
16 2 c jx x0 j2 jx x0 j
A
where is the angle between a0 and x x0 . The total radiated power (integrated
over a closed spherical surface) becomes
0 q 0 2 .a0 /2 q 0 2 a02
P D D (6.122)
6c 6"0 c 3
R
which is the Larmor formula for radiated power from an accelerated charge.
Note that here we are treating a charge with v 0 c but otherwise totally un-
specified motion while we compare with formulæ derived for a stationary oscil-
lating dipole. The electric and magnetic fields, equation (6.117) on the previous
page and equation (6.118) on the preceding page, respectively, and the expres-
D
sions for the Poynting flux and power derived from them, are here instantaneous
values, dependent on the instantaneous position of the charge at x0 .t 0 /. The an-
gular distribution is that which is ‘frozen’ to the point from which the energy is
radiated.
6.5.3 Bremsstrahlung
An important special case of radiation is when the velocity v 0 and the accelera-
tion a0 are collinear (parallel or anti-parallel) so that v 0 a0 D 0. This condition
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 155
. of 300
FT
the energy flux (Poynting vector) far away from the source therefore becomes
0 q 0 2 a02 sin2 x x0
Sfar D (6.125)
x0 j2 1 v0
6 jx x0 j
16 2 c jx cos
c
It is interesting to note that the magnitudes of the electric and magnetic fields are
the same whether v 0 and a0 are parallel or anti-parallel.
We must be careful when we compute the energy (S integrated over time).
The Poynting vector is related to the time t when it is measured and to a fixed
surface in space. The radiated power into a solid angle element d, measured
A
relative to the particle’s retarded position, is given by the formula
dU rad dU rad s
0
d D d D S .x x0 /s d (6.128)
dt dt jx x0 j
Inserting equation (6.125) above for S into (6.128), we obtain the explicit
expression for the energy loss due to radiation evaluated at the retarded time
FT
The angular factors of this expression, for three different particle speeds, are
plotted in figure 6.7 on page 132.
Comparing expression (6.126) on the preceding page with expression (6.129)
on the previous page, we see that they differ by a factor 1 v 0 cos =c that comes
from the extra factor s= jx x0 j introduced in (6.128). Let us explain this in
geometrical terms.
A
During the interval .t 0 ; t 0 C dt 0 / and within the solid angle element d the
particle radiates an energy ŒdU rad ./=dt 0 dt 0 d. As shown in figure 6.8 this
energy is at time t located between two spheres, one outer with its origin at x01 .t 0 /
and radius c.t t 0 /, and one inner with its origin at x02 .t 0 Cdt 0 / D x01 .t 0 /Cv 0 dt 0
and radius cŒt .t 0 C dt 0 / D c.t t 0 dt 0 /.
R
From Figure 6.8 we see that the volume element subtending the solid angle
element
d2x
d D ˇ ˇ2 (6.130)
x02 ˇ
D
ˇx
is
ˇ2
d3x D d2x dr D ˇx x02 ˇ d dr
ˇ
(6.131)
Here, dr denotes the differential distance between the two spheres and can be
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 157
. of 300
x x02
x02 ˇ C c dt 0 x02 ˇ ˇ v 0 dt 0
ˇ ˇ ˇ ˇ
dr D ˇx ˇx
ˇx x 0 ˇ
ˇ
„ ƒ‚2 …
v 0 cos (6.132)
!
x x02 cs
D c ˇ v0 dt 0 D ˇ ˇ dt 0
x02 ˇ x02 ˇ
ˇ
ˇx ˇx
where formula (6.84) on page 124 was used in the last step. Hence, the volume
element under consideration is
s
d3x D d2x dr D ˇ ˇ d2x cdt 0
FT
(6.133)
ˇx x02 ˇ
We see that the energy that is radiated per unit solid angle during the time interval
.t 0 ; t 0 C dt 0 / is located in a volume element whose size is dependent. This
explains the difference between expression (6.126) on page 133 and expression
(6.129) on page 133.
Let the radiated energy, integrated over , be denoted UQ rad . After tedious,
but relatively straightforward integration of formula (6.129) on page 133, one
obtains
A
3
dUQ rad 0 q 0 2 a02 2 q 0 2 a02 v 02
1
D 3 D 1 (6.134)
dt 0 6c 3 4"0 c 3 c2
02
1 vc 2
If we know v 0 .t 0 /, we can integrate this expression over t 0 and obtain the total
R
energy radiated during the acceleration or deceleration of the particle. This way
we obtain a classical picture of bremsstrahlung (braking radiation, free-free ra-
diation). Often, an atomistic treatment is required for obtaining an acceptable
result.
D
x3
0 0
FT
With the charged particle orbiting in the x1 x2 plane as in figure 6.9, an orbit
radius b, and an angular frequency !0 , we obtain
'.t 0 / D !0 t 0
0
O 2 sin '.t /
x .t / D bŒOx1 cos '.t / C x 0
(6.135a)
(6.135b)
A
0 0 0 0 0 0
P .t / D b!0 Œ x
v .t / D x O 1 sin '.t / C x
O 2 cos '.t / (6.135c)
0
ˇ 0ˇ
v D ˇv ˇ D b!0 (6.135d)
R 0 .t 0 / D b!02 ŒOx1 cos '.t 0 / C x
a0 D vP 0 .t 0 / D x O 2 sin '.t 0 / (6.135e)
v 0 D ˇa0 ˇ D b! 2
ˇ ˇ
0 (6.135f)
R
Because of the rotational symmetry we can, without loss of generality, rotate
our coordinate system around the x3 axis so the relative vector x x0 from the
source point to an arbitrary field point always lies in the x2 x3 plane, i.e.
x0 D ˇx x0 ˇ .Ox2 sin ˛ C x
ˇ ˇ
x O 3 cos ˛/ (6.136)
D
where ˛ is the angle between x x0 and the normal to the plane of the particle
orbit (see Figure 6.9). From the above expressions we obtain
where in the last step we simply used the definition of a scalar product and the
fact that the angle between a0 and x x0 is .
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 159
. of 300
The energy flux is given by the Poynting vector, which, with the help of
formula (6.105) on page 129, can be written
1 1 x x0
SD .E B/ D jEj2 (6.138)
0 c0 jx x0 j
Inserting this into equation (6.128) on page 133, we obtain
dU rad .˛; '/ jx x0 j s
D jEj2 (6.139)
dt 0 c0
where the retarded distance s is given by expression (6.84) on page 124. With
the radiation part of the electric field, expression (6.106) on page 129, inserted,
and using (6.137a) and (6.137b) on the preceding page, one finds, after some
algebra, that
2
v0 v 02
rad 0 2 02 1 sin ˛ cos ' 1 sin2 ˛ sin2 '
FT
dU .˛; '/ 0 q a c c2
D 5
dt 0 16 2 c 0
1 v sin ˛ cos '
c
(6.140)
The angles and ' vary in time during the rotation, so that refers to a moving
coordinate system. But we can parametrise the solid angle d in the angle ' and
the angle ˛ so that d D sin ˛ d˛ d'. Integration of equation (6.140) above over
this d gives, after some cumbersome algebra, the angular integrated expression
dUQ rad 0 q 0 2 a02 1
0
D 2 (6.141)
dt 6c
02
1 vc 2
A
In equation (6.140) above, two limits are particularly interesting:
1. v 0 =c 1 which corresponds to cyclotron radiation.
2. v 0 =c . 1 which corresponds to synchrotron radiation.
R
6.5.4.1 Cyclotron radiation
For a non-relativistic speed v 0 c, equation (6.140) above reduces to
dU rad .˛; '/ 0 q 0 2 a02
D .1 sin2 ˛ sin2 '/ (6.142)
dt 0 16 2 c
D
FT
termine the characteristics of the particle motion both in particle accelerators
and in astrophysical objects where a direct measurement of particle velocities
are impossible.
In the orbit plane (˛ D =2), equation (6.140) on the previous page gives
2
v0 v 02
rad
dU .=2; '/ 0 2 02 1
0 q a c
cos ' 1 c2
sin2 '
D 5 (6.146)
dt 0 16 2 c 0
1 v cos '
c
This vanishes for angles '0 such that
v0
A
cos '0 D (6.147a)
c
s
v 02
sin '0 D 1 (6.147b)
c2
Hence, the angle '0 is a measure of the synchrotron radiation lobe width ;
see figure 6.10 on the facing page. For ultra-relativistic particles, defined by
R
s
1 v 02
Ds 1; 1 1; (6.148)
02 c2
v
1
c2
D
FT
x3
This angular interval is swept by the charge during the time interval
t 0 D (6.151)
!0
during which the particle moves a length interval
l 0 D v 0 t 0 D v 0 (6.152)
A
!0
in the direction toward the observer. The observer therefore measures a com-
pressed pulse width of length
l 0 v 0 t 0 v0
t D t 0 D t 0 D 1 t 0
c c c
R
v0 v0
0
v
v 0
1 1 c
1 C c 1
D 1 1 D
c !0 c
!0 v0
!0
1C
„ ƒ‚c… (6.153)
2
D
v 02
1 1 1
1 D 3
c 2 2
!0 2
!0
„ ƒ‚ …
1=
2
Typically, the spectral width of a pulse of length t is ! . 1=t . In the ultra-
relativistic synchrotron case one can therefore expect frequency components up
to
1
!max D 2
3 !0 (6.154)
t
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 162
. of 300
1. All N radiating particles are spatially much closer to each other than a typical
wavelength. Then the relative phase differences of the individual electric and
magnetic fields radiated are negligible and the total radiated fields from all
individual particles will add up to become N times that from one particle.
This means that the power radiated from the N particles will be N 2 higher
than for a single charged particle. This is called coherent radiation.
2. The charged particles are perfectly evenly distributed in the orbit. In this case
FT
the phases of the radiation fields cause a complete cancellation of the fields
themselves. No radiation escapes.
3. The charged particles are somewhat unevenly distributed in the orbit. This
happens for an open ring current, carried initially by evenly distributed charged
particles, which is subject to thermal fluctuations. From statistical mechanics
we know that this happens forpall open systems and that the particle densities
exhibit fluctuations of order N . This means that out of the N particles,
p
N will exhibit deviation from perfect randomness — and thereby perfect
A
radiation field cancellation
p — and give rise to net radiation fields which are
proportional
p to N . As a result, the radiated power will be proportional to
. N /2 D N , and we speak about incoherent radiation. Examples of this can
be found both in earthly laboratories and under cosmic conditions.
R
6.5.4.3 Radiation in the general case
We recall that the general expression for the radiation E field from a moving
charge concentration is given by expression (6.106) on page 129. This expres-
sion in equation (6.139) on page 137 yields the general formula
0 q 0 2 jx x0 j
2
jx x0 j v 0
D
jx x0 j
O3
E? x
FT
v ? a0 , sin D 1, which corresponds to cyclotron radiation or synchrotron
0
radiation.
Sheet: 164
. of 300
p1 p01
FT
fields, the total field energy can be written
Z Z bmax Z 1
UQ D "0 d3x E? 2
D "0 db 2b 2
dt v 0 E? (6.161)
V bmin 1
where the volume integration is over the plane perpendicular to v 0 . With the use
of Parseval’s identity for Fourier transforms, formula (6.9) on page 106, we can
rewrite this as
Z 1 Z bmax Z 1
Q
U D Q
d! U! D 4"0 v 0
db 2b 2
d! E!;?
A
0 bmin 0
Z 1 Z v0
=! (6.162)
q 02 db
d!
2 2 "0 v 0 1 bmin b
from which we conclude that
q 02
0
v
R
UQ ! 2 0
ln (6.163)
2 "0 v bmin !
where an explicit value of bmin can be calculated in quantum theory only.
It is intriguing to quantise the energy into photons. We then find that
2˛ c
d!
N! d! ln (6.164)
D
bmin ! !
6.6 Examples
FT
BLinear momentum emitted by a non-moving, oscillating Hertzian dipole in free space
Let us consider an infinitesimally small electric dipole d (also known as a Hertzian dipole)
with a moment point x0 , oscillating along the x3 D z axis at single the angular frequency !.
We describe it in complex notation as d.t; x0 / D d!;z .x0 / exp. i!t/Oz. The Fourier amp-
litudes of the fields generated by this dipole are given by expressions (6.45) on page 114.
Inverse Fourier transforming these expressions to the time domain, and using a spherical
coordinate system .r; ; '/ where r D jx x0 j, the physically observable fields are readily
found to be
E X A M P L E 6.1
A
!0 1 k
B.t; x/ D d! sin sin.kr !t/ cos.kr !t/ ® O (6.167a)
4 r2 r
!
1 1 k k2
E.t; x/ D d! sin cos.kr !t/ C 2 sin.kr !t/ cos.kr !t/ ™O
4"0 r3 r r
1 1 k
C d! cos cos.kr !t/ C sin.kr !t/ rO (6.167b)
R
2"0 r3 r2
Letting the subscript . n/ denote the quantity in question falls off as r n, we can
Applying formula (4.45) on page 62 for the electromagnetic linear momentum density to
the fields from an electric dipole, equations (6.177) on page 145, we obtain
!0 2 1
gfield .t; x/ D "0 E.t; x/ B.t; x/ D d sin cos sin.kr !t/ cos.kr !t/
8 2 !
D
r5
k2
k
C 4 sin2 .kr !t/ cos2 .kr !t/ sin.kr !t/ cos.kr !t/ ™O
r r3
!0 2 2 1
C d sin sin.kr !t/ cos.kr !t/
16 2 ! r5
k k2
C 4 sin2 .kr !t/ cos2 .kr !t/
2 3 sin.kr !t/ cos.kr !t/
r r
k3
C 2 cos2 .kr !t/ rO
r
(6.168)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 166
. of 300
Using the well-known double-angle trigonometric relations, this can be put in the form
field !0 2 1 k
g .t; x/ D d sin cos 5 sinŒ2.kr !t/ 2 4 cosŒ2.kr !t/
16 2 ! r r
k2
sinŒ2.kr !t/ ™O
r3
!0 2 2 1 k
C d sin 5 sinŒ2.kr !t/ 2 4 cosŒ2.kr !t/
32 2 ! r r
k2 k3
2 3 sinŒ2.kr !t/ C 2 1 C cosŒ2.kr !t/ rO
r r
(6.169)
We see that the linear momentum density gfield , and hence the Poynting vector S D gfield c 2
O and therefore
[recall identity (4.45) on page 62], has a perpendicular component (along ™)
performs, in general, a spiralling motion.
FT
Consequently, at finite distances from the source the linear momentum has a perpendicular
component. It is only at infinity that it is strictly radial (along rO ).
Applying formula (4.75) on page 67 for the electromagnetic angular momentum density
around the momentum point x0 , i.e.
and using equation (6.179) on page 146, we find that for a pure electric dipole
field !0 2 1
h D d sin cos 4 sin.kr !t/ cos.kr !t/
8 2 ! r
A
k2
k
C 3 sin2 .kr !t/ cos2 .kr !t/ sin.kr !t/ cos.kr !t/ ®O
r r2
(6.171)
or
!0 2 1
hfield D 2
d! sin cos sinŒ2.kr !t/
16 r4
R
k2
k
2 3 cosŒ2.kr !t/ sinŒ2.kr !t/ ®O (6.172)
r r2
The total electromagnetic linear momentum is (cf. formula (4.51b) on page 63)
Z
pfield D d3x 0 gfield .t 0 ; x0 / (6.173)
D
V0
and the total electromagnetic angular momentum is (cf. formula (4.94) on page 70)
Z
J field D d3x 0 hfield .t 0 ; x0 / (6.174)
V0
In order to get a total net J field , it is convenient to superimpose several individual dipoles of
(possibly) different strengths and relative phases. Perhaps the most common configuration
yielding a total net J field is two orthogonal co-located dipoles with =2 phase shift between
them.
We note that in the far zone the linear and angular momentum densities tend to
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 167
. of 300
!0 2 k 3
gfield,far .t; x/ d sin2 cos2 .kr !t/Or
16 2 ! r 2
!0 2 k 3
d! 2 sin2 1 C cosŒ2.kr !t/ rO
D 2
(6.175)
32 r
and
!0 2 k 2
hfield,far .t; x/ d sin cos sin.kr !t/ cos.kr O
!t/®
8 2 ! r 2
!0 2 k 2
D d O
sin cos sinŒ2.kr !t/® (6.176)
16 2 ! r 2
respectively. I.e. to leading order, both the linear momentum density gfield and the angular
momentum density hfield fall off as 1=r 2 far away from the source region. This means
that when they are integrated over a spherical surface / r 2 located at a large distance from
the source [cf. the last term in the LHS of formula (4.33) on page 60], there can be a net flux
so that the integrated momenta do not fall off with distance and can therefore be transported
FT
all the way to infinity.
BLinear and angular momentum densities from an electric dipole in vacuum E X A M P L E 6.2
The Fourier amplitudes of the fields generated by an infinitesimal electric dipole (also
known as a Hertzian dipole), oscillating along the x3 D z axis at the angular frequency
!, d.t; x0 / D d! .x0 / exp. i!t/, are given by expressions (6.45) on page 114. Inverse
Fourier transforming these expressions to the time domain, and using a spherical coordinate
system .r; ; '/ where r D jx x0 j, the physically observable fields are readily found to
be
A
!0 1 k
B.t; x/ D d! sin sin.kr !t/ cos.kr !t/ O
® (6.177a)
4 r2 r
!
1 1 k k2
E.t; x/ D d! sin cos.kr !t/ C 2 sin.kr !t/ cos.kr !t/ ™O
4"0 r3 r r
1 1 k
C d! cos cos.kr !t/ C sin.kr !t/ rO (6.177b)
R
2"0 r3 r2
Letting the subscript . n/ denote the quantity in question falls off as r n, we can
Applying formula (4.45) on page 62 for the electromagnetic linear momentum density to
the fields from an electric dipole, equations (6.177) above, we obtain
!0 2 1
gfield .t; x/ D "0 E.t; x/ B.t; x/ D d sin cos sin.kr !t/ cos.kr !t/
8 2 !
D
r5
k2
k
C 4 sin2 .kr !t/ cos2 .kr !t/ sin.kr !t/ cos.kr !t/ ™O
r r3
!0 2 2 1
C d sin sin.kr !t/ cos.kr !t/
16 2 ! r5
k k2
C 4 sin2 .kr !t/ cos2 .kr !t/
2 3 sin.kr !t/ cos.kr !t/
r r
k3
C 2 cos2 .kr !t/ rO
r
(6.178)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 168
. of 300
Using the well-known double-angle trigonometric relations, this can be put in the form
field !0 2 1 k
g .t; x/ D d sin cos 5 sinŒ2.kr !t/ 2 4 cosŒ2.kr !t/
16 2 ! r r
k2
sinŒ2.kr !t/ ™O
r3
!0 2 2 1 k
C d sin 5 sinŒ2.kr !t/ 2 4 cosŒ2.kr !t/
32 2 ! r r
k2 k3
2 3 sinŒ2.kr !t/ C 2 1 C cosŒ2.kr !t/ rO
r r
(6.179)
We see that the linear momentum density gfield , and hence the Poynting vector S D gfield c 2
O and therefore
[recall identity (4.45) on page 62], has a perpendicular component (along ™)
performs, in general, a spiralling motion.
FT
Consequently, at finite distances from the source the linear momentum has a perpendicular
component. It is only at infinity that it is strictly radial (along rO ).
Applying formula (4.75) on page 67 for the electromagnetic angular momentum density
around the momentum point x0 , i.e.
and using equation (6.179) above, we find that for a pure electric dipole
field !0 2 1
h D d sin cos 4 sin.kr !t/ cos.kr !t/
8 2 ! r
A
k2
k
C 3 sin2 .kr !t/ cos2 .kr !t/ sin.kr !t/ cos.kr !t/ ®O
r r2
(6.181)
or
!0 2 1
hfield D 2
d! sin cos sinŒ2.kr !t/
16 r4
R
k2
k
2 3 cosŒ2.kr !t/ sinŒ2.kr !t/ ®O (6.182)
r r2
The total electromagnetic linear momentum is (cf. formula (4.51b) on page 63)
Z
pfield D d3x 0 gfield .t 0 ; x0 / (6.183)
D
V0
and the total electromagnetic angular momentum is (cf. formula (4.94) on page 70)
Z
J field D d3x 0 hfield .t 0 ; x0 / (6.184)
V0
In order to get a total net J field , it is convenient to superimpose several individual dipoles of
(possibly) different strengths and relative phases. Perhaps the most common configuration
yielding a total net J field is two orthogonal co-located dipoles with =2 phase shift between
them.
We note that in the far zone the linear and angular momentum densities tend to
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 169
. of 300
kO
x2
zO 0
j! .x0 / '
x0
FT
x1 O0
®
'0
¡O 0
!0 2 k 3
gfield,far .t; x/ d sin2 cos2 .kr !t/Or
16 2 ! r 2
!0 2 k 3
d! 2 sin2 1 C cosŒ2.kr !t/ rO
D 2
(6.185)
32 r
A
and
!0 2 k 2
hfield,far .t; x/ d sin cos sin.kr !t/ cos.kr O
!t/®
8 2 ! r 2
!0 2 k 2
D d O
sin cos sinŒ2.kr !t/® (6.186)
16 2 ! r 2
R
respectively. I.e. to leading order, both the linear momentum density gfield and the angular
momentum density hfield fall off as 1=r 2 far away from the source region. This means
that when they are integrated over a spherical surface / r 2 located at a large distance from
the source [cf. the last term in the LHS of formula (4.33) on page 60], there can be a net flux
so that the integrated momenta do not fall off with distance and can therefore be transported
all the way to infinity.
D
i0 eikjxj
Z
ik x0
far
B! D d3x 0 e j! k (6.187)
4 jxj V0
In our case the generator produces a single frequency ! and we feed the antenna across a
small gap where the loop crosses the positive x1 axis. The circumference of the loop is
chosen to be exactly one wavelength D 2c=!. This means that the antenna current
oscillates in the form of a sinusoidal standing current wave around the circular loop with a
Fourier amplitude
j! D I0 cos ' 0 ı.0 a/ı.z 0 /®
O0 (6.188)
For the spherical coordinate system of the field point, we recall from subsection F.1.2 on
page 217 that the following relations between the base vectors hold:
rO D sin cos ' x
O 1 C sin sin ' x
O 2 C cos x
O3
™O D cos cos ' x
O 1 C cos sin ' x
O2 O3
sin x
O D
® O 1 C cos ' x
sin ' x O2
FT
and
O 1 D sin cos ' rO C cos cos ' ™O
x O
sin ' ®
O 2 D sin sin ' rO C cos sin ' ™O C cos ' ®
x O
O 3 D cos rO
x sin ™O
With the use of the above transformations and trigonometric identities, we obtain for the
cylindrical coordinate system which describes the source:
¡O 0 D cos ' 0 x
O 1 C sin ' 0 x
O2
(6.189)
D sin cos.' 0 '/Or C cos cos.' 0 '/™O C sin.' 0 O
'/®
A
O0 D
® sin ' 0 x
O 1 C cos ' 0 x
O2
(6.190)
D sin sin.' 0 '/Or cos sin.' 0 '/™O C cos.' 0 O
'/®
zO 0 D x
O 3 D cos rO sin ™O (6.191)
Utilising the periodicity of the integrands over the integration interval Œ0; 2, introducing
the auxiliary integration variable ' 00 D ' 0 ', and utilising standard trigonometric identit-
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 171
. of 300
FT
Z 2
ika sin cos ' 00
e sin ' 00 cos.' 00 C '/ d' 00
0
1
Z 2
ika sin cos ' 00
D sin ' e d' 00 (6.196)
2 0
1
Z 2
ika sin cos ' 00
sin ' e cos 2' 00 d' 00
2 0
Jn . / D . 1/n Jn ./
ı n i cos ' ı n 2 (6.197)
A
Z Z
i cos '
Jn . / D e cos n' d' D e cos n' d'
0 2 0
V0
(6.199)
D I0 ak cos ' ŒJ0 .ka sin / J2 .ka sin / ™O
O
C I0 ak cos sin ' ŒJ0 .ka sin / C J2 .ka sin / ®
To obtain the desired physical magnetic field in the radiation (far) zone we must Fourier
transform back to t space and take the real part:
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 172
. of 300
!t 0 /
( )
i0 e.ikr
far
B .t; x/ D Re I ™O C I' ®
O
4 r
0
D sin.kr !t 0 / I ™O C I' ®O
4 r (6.201)
I0 ak0
D sin.kr !t 0 / cos ' ŒJ0 .ka sin / J2 .ka sin / ™O
4r
O
C cos sin ' ŒJ0 .ka sin / C J2 .ka sin / ®
From this expression for the radiated B field, we can obtain the radiated E field with the
help of Maxwell’s equations.
FT
E X A M P L E 6.4 BThe fields from a uniformly moving charge
4This problem was first solved by In the special case of uniform motion,4 the localised charge moves in a field-free, isolated
O L I V E R H E AV I S I D E in 1888. space and we know that it will not be affected by any external forces. It will therefore move
uniformly in a straight line with the constant velocity v 0 . This gives us the possibility to
extrapolate its position at the observation time, x0 .t/, from its position at the retarded time,
x0 .t 0 /. Since the particle is not accelerated, vP 0 0, the virtual simultaneous coordinate
x0 will be identical to the actual simultaneous coordinate of the particle at time t , i.e.
x0 .t/ D x0 .t/. As depicted in figure 6.6 on page 124, the angle between x x0 and v 0 is
0 while then angle between x x0 and v 0 is 0 .
In the case of uniform velocity v 0 , i.e. a velocity that does not change with time, any
A
physical observable f .t; x/ has the same value at time t and position x as it has at time
t C dt and position x C v 0 dt . Hence,
Taylor expanding f .t C dt; x C v 0 dt ), keeping only linear terms in the infinitesimally small
R
dt , we obtain
@f
f .t C dt; x C v 0 dt/ D f .t; x/ C dt C v 0 rf dt C O .dt/2
(6.203)
@t
@f
D
D v 0 rf (6.204)
@t
Since f is an arbitrary physical observable, the time and space derivatives must be related
in the following way when they operate on any physical observable dependent on x.t/ [cf.
equation (1.35) on page 13]:
@
D v0 r (6.205)
@t
Hence, the E and B fields can be obtained from formulæ (6.85) on page 124, with the
potentials given by equations (6.83) on page 123 as follows:
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 173
. of 300
@A 1 @v 0 v 0 @
ED r D r D r
@t c 2 @t c 2 @t
v0 v0 v0 ˝ v0
D r C r D 1 r (6.206a)
c c c2
0
v ˝ v0
D 13 r
c2
0
v v0 v0
B Dr ADr 2
D r 2 D r
c c c2
v0 v0 v0 v0 v0 ˝ v0
D 2 r r D 2 13 r (6.206b)
c c c c c2
v0
D 2 E
c
Here 13 D x O i is the unit dyad and we used the fact that v 0 v 0 0. What remains is
O ix
FT
just to express r in quantities evaluated at t and x.
From equation (6.83a) on page 123 and equation (6.114) on page 131 we find that
q0 q0
1
r D r D r s2
4"0 s 8"0 s 3
(6.207)
q0 v0
0
v
D .x x 0 / C .x x0 /
4"0 s 3 c c
When this expression for r is inserted into equation (6.206a) above, the following result
v0 ˝ v0 q0
0
v ˝ v0
E.t; x/ D 13 r D 1 3 r s2
c2 8"0 s 3 c2
A
(
q0 v0
0
v
D .x x 0 / C .x x 0 /
4"0 s 3 c c
0 0 )
v ˝ v0
0
0 0
v v v v
.x x0 / .x x0 /
c c c2 c c
(6.208)
R
"
q0 v0 v0 v 02
D 3
.x x 0 / C .x x 0 / .x x0 / 2
4"0 s c c c
#
v0 v0
.x x0 /
c c
!
q0 v 02
D .x x0 / 1
D
4"0 s 3 c2
obtains. Of course, the same result also follows from equation (6.102) on page 128 with
vP 0 0 inserted.
From equation (6.208) above we conclude that E is directed along the vector from the
simultaneous coordinate x0 .t/ to the field (observation) coordinate x.t/. In a similar way,
the magnetic field can be calculated and one finds that
!
0 q 0 v 02 1 0
B.t; x/ D 1 v 0 .x x0 / D v E (6.209)
4s 3 c2 c2
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 174
. of 300
From these explicit formulae for the E and B fields and formula (6.113b) on page 130 for
s, we can discern the following cases:
1. v 0 ! 0 ) E goes over into the Coulomb field ECoulomb
2. v 0 ! 0 ) B goes over into the Biot-Savart field
3. v 0 ! c ) E becomes dependent on 0
4. v 0 ! c; sin 0 0 ) E ! .1 v 02 =c 2 /ECoulomb
5. v0 ! c; sin 0 1 ) E ! .1 v 02 =c 2 / 1=2 ECoulomb
FT
which means that
vP 0 .t 0 / D v 0 ı.t 0
v 0 .t0 / D
1
t0 /
dt 0 vP 0
x0
(6.210)
(6.211)
(6.212)
A
x x0 x (6.213)
Because of the Dirac ı behaviour in time, Fourier transforming expression (6.214) above
D
Z 1 dUQ rad
Z 1 I
B
UQ rad D dt 0 D dt 0
d2 0 0
x O
n E
1 dt 0 1 S0 0
Z 1 Z 1
1 1
I I
2 0 0 2 0
D dx dt EB D dx dt 0 E 2 (6.217)
0 S 0 1 0 c S 0 1
I Z 1
2 0
D "0 c dx dt 0 E 2
S0 1
According to Parseval’s identity [cf. equation (6.9) on page 106] the following equality
holds:
Z 1 Z 1
dt 0 E 2 D 4 d! jE! j2 (6.218)
1 0
which means that the radiated energy in the frequency interval .!; ! C d!/ is
I
Q rad 2 0 2
U! d! D 4"0 c d x jE! j d! (6.219)
FT
S0
For our infinite spectrum, equation (6.216) on the preceding page, we obtain
q 0 2 .v 0 /2 sin2 0
I
UQ !rad d! D d2x 0 d!
16 3 "0 c 3 S0 jx x0 j2
q 0 2 .v 0 /2 2 0 0
Z Z
D d' d sin 0 sin2 0 d! (6.220)
16 3 "0 c 3 0 0
q02
0 2
v d!
D
3"0 c c 2
We see that the energy spectrum UQ !rad is independent of frequency !. This means that if we
A
would integrate it over all frequencies ! 2 Œ0; 1/, a divergent integral would result.
In reality, all spectra have finite widths, with an upper cutoff limit set by the quantum
condition
1 1 02
}!max D m.v 0 C v 0 /2 mv (6.221)
2 2
R
which expresses that the highest possible frequency !max in the spectrum is that for which
all kinetic energy difference has gone into one single field quantum (photon) with energy
}!max . If we adopt the picture that the total energy is quantised in terms of N! photons
radiated during the process, we find that
UQ !rad d!
D dN! (6.222)
D
}!
or, for an electron where q 0 D
jej, where e is the elementary charge,
e2 2 v 0 2 d! 1 2 v 0 2 d!
dN! D (6.223)
4"0 }c 3 c ! 137 3 c !
where we used the value of the fine structure constant ˛ D e 2 =.4"0 }c/ 1=137.
Even if the number of photons becomes infinite when ! ! 0, these photons have negligible
energies so that the total radiated energy is still finite.
6.7 Bibliography
[34] H. A LFVÉN AND N. H ERLOFSON, Cosmic radiation and radio stars, Physical Review,
78 (1950), p. 616.
FT
[38] J. D. JACKSON, Classical Electrodynamics, third ed., John Wiley & Sons, Inc.,
New York, NY . . . , 1999, ISBN 0-471-30932-X.
[39] H.-D. NATURE, The Physical Basis of The Direction of Time, fourth ed., Springer-
Verlag, Cambridge . . . , 1920, ISBN 3-540-42081-9.
RELATIVISTIC ELECTRODYNAMICS
We saw in chapter chapter 3 how the introduction of electrodynamic potentials 1 The Special Theory of Relativity,
led, in a most natural way, to the existence of a characteristic, finite speed of by the physicist and philosopher
D AV I D J O S E P H B O H M (1917–
propagation of electromagnetic fields (and related quantities) in free space (va- 1992), opens with the following
p
cuum) that equals the speed of light c D 1= "0 0 and which can be considered paragraph:
a constant of Nature. To take this finite speed of propagation of information into ‘The theory of relativity
is not merely a scientific
account, and to ensure that our laws of physics be independent of any specific development of great
FT
coordinate frame, requires a treatment of electrodynamics in a relativistically importance in its own right.
It is even more significant
covariant (coordinate independent) form. This is the objective of this chapter.1 as the first stage of a
The technique we shall use to study relativity is the mathematical apparatus radical change in our basic
concepts, which began
developed for non-Euclidean spaces of arbitrary dimensions, here specialised to in physics, and which is
four dimensions. It turns out that this theory of Riemannian spaces, derived for spreading into other fields
of science, and indeed,
more or less purely mathematical reasons only, is ideal for a formal description even into a great deal of
of relativistic physics. For the simple case of the special theory of relativity, thinking outside of science.
For as is well known, the
the mathematics is quite simple, whereas for the general theory of relativity it modern trend is away
becomes more complicated. from the notion of sure
A
‘absolute’ truth, (i.e. one
which holds independently
of all conditions, contexts,
degrees, and types of
approximation etc.) and
7.1 The special theory of relativity toward the idea that a given
concept has significance
only in relation to suitable
R
broader forms of reference,
An inertial system, or inertial reference frame, is a system of reference, or rigid within which that concept
coordinate system, in which the law of inertia (Galileo’s law, Newton’s first can be given its full
meaning.’
law) holds. In other words, an inertial system is a system in which free bodies
move uniformly and do not experience any acceleration. The special theory of
relativity describes how physical processes are interrelated when observed in
D
155
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 178
. of 300
FT
tensors) in an equation describing a physical process must transform in a covari-
ant manner, i.e. in the same way.
z0 D z (7.2d)
Taking the difference between the square of (7.2a) and the square of (7.2b) we
find that
c 2 t 02 x 02 D
2 c 2 t 2 2xcˇt C x 2 ˇ 2 x 2 C 2xvt v 2 t 2
v2 v2
1 2 2 2
D c t 1 x 1
v2 c2 c2 (7.3)
1 2
c
D c2t 2 x2
From equations (7.2) on the preceding page we see that the y and z coordinates
are unaffected by the translational motion of the inertial system †0 along the x
FT
axis of system †. Using this fact, we find that we can generalise the result in
equation (7.3) above to
c2t 2 x2 y2 z 2 D c 2 t 02 x 02 y 02 z 02 (7.4)
which means that if a light wave is transmitted from the coinciding origins O and
O 0 at time t D t 0 D 0 it will arrive at an observer at .x; y; z/ at time t in † and an
observer at .x 0 ; y 0 ; z 0 / at time t 0 in †0 in such a way that both observers conclude
that the speed (spatial distance divided by time) of light in vacuum is c. Hence,
A
the speed of light in † and †0 is the same. A linear coordinate transformation
which has this property is called a (homogeneous) Lorentz transformation.
x D g x (7.6)
where the upper index in x is summed over and is therefore a dummy index
and may be replaced by another dummy index , say. This summation process
is an example of index contraction and is often referred to as index lowering.
FT
7.1.2.2 Scalar product and norm
The scalar product of x with itself in a Riemannian space is defined as
g x x D x x (7.7)
0 0 0 1
i.e. a matrix with a main diagonal that has the sign sequence, or signature,
fC; ; ; g. As we see, the index lowering operation in our flat 4D space
L4 is nearly trivial.
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 181
. of 300
Hence, if the metric tensor is defined according to expression (7.8) on the facing
page, the covariant position four-vector x is obtained from the contravariant
position four-vector x simply by changing the sign of the last three compon-
ents. These components are referred to as the space components; the zeroth
FT
component is referred to as the time component.
As we see, for this particular choice of metric, the scalar product of x with
itself becomes
space dimensionality. Secondly, we find, after trivial algebra, that the following
relations between the contravariant, covariant and mixed forms of the metric
tensor hold:
Here we have introduced the 4D version of the Kronecker delta ı , a mixed
four-tensor of rank 2 that fulfils
(
1 if D
ı D ı
D (7.14)
0 if ¤
0 0 0 1
FT
7.1.2.4 Invariant line element and proper time
The differential distance ds between the two points x and x C dx in L4 can
be calculated from the Riemannian metric, given by the quadratic differential
form
where the metric tensor is as in equation (7.8) on page 158. As we see, this
.dx 3 /2 (7.16)
A
form is indefinite as expected for a non-Euclidean space. The square root of this
expression is the invariant line element
q 2 !2 !2 !2 3
1 4 dx 1 dx 2 dx 3
ds D c dt 1 C C 5
c2 dt dt dt
R
s
r (7.17)
1 v2
.vx /2 C .vy /2 C .vz /2 D c dt
D c dt 1 1
c2 c2
dt
q
D c dt 1 ˇ 2 D c D c d
D
where we introduced
d D dt= (7.18)
Since d measures the time when no spatial changes are present, i.e. by a clock
that is fixed relative the given frame of reference, it is called the proper time.
As equation (7.18) above shows, the proper time of a moving object is always
less than the corresponding interval in the rest system. One may say that moving
clocks go slower than those at rest.
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 183
. of 300
ds 2 D c 2 dt 2 dx 2 dy 2 dz 2 (7.19)
dx 2 C dy 2 C dz 2 < c 2 dt 2 (7.20)
FT
dx 2 C dy 2 C dz 2 > c 2 dt 2
dx 2 C dy 2 C dz 2 D c 2 dt 2
(7.21)
(7.22)
is a light-like interval ; we may also say that in this case we are on the light cone.
A vector which has a light-like interval is called a null vector. The time-like,
space-like or light-like aspects of an interval ds are invariant under a Lorentz
A
transformation. I.e. it is not possible to change a time-like interval into a space-
like one or vice versa via a Lorentz transformation.
The scalar product between this four-vector field and another one b .x / is
which is a scalar field , i.e. an invariant scalar quantity ˛.x / which depends on
time and space, as described by x D .ct; x; y; z/.
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 184
. of 300
the linear Lorentz transformation (7.2) on page 156, i.e. the coordinate trans-
formation x ! x 0 D x 0 .x 0 ; x 1 ; x 2 ; x 3 /, from one inertial system † to
another inertial system †0 in the standard configuration, can be written
FT
x 0 D ƒ x (7.26)
This means that the nonempty set of Lorentz transformations constitutes a closed
algebraic structure with a binary operation (multiplication) that is associative.
R
Furthermore, one can show that this set possesses at least one identity element
and at least one inverse element. In other words, this set of Lorentz transform-
ations constitutes a mathematical group. However tempting, we shall not make
any further use of group theory.
D
x1
Introducing
X 0 D ix 0 D ict
FT
(7.28a)
1 1
X Dx (7.28b)
2 2
X Dx (7.28c)
3 3
X Dx (7.28d)
dS D ids (7.28e)
p
where i D 1, we see that equation (7.16) on page 160 transforms into
dS 2 D .dX 0 /2 C .dX 1 /2 C .dX 2 /2 C .dX 3 /2 (7.29)
A
i.e. into a 4D differential form that is positive definite just as is ordinary 3D
Euclidean space R3 . We shall call the 4D Euclidean space constructed in this
way the Minkowski space M4 .3 3 The fact that our Riemannian
As before, it suffices to consider the simplified case where the relative motion space can be transformed in this
way into a Euclidean one means
between † and †0 is along the x axes. Then that it is, strictly speaking, a
pseudo-Riemannian space.
R
dS 2 D .dX 0 /2 C .dX 1 /2 D .dX 0 /2 C .dx 1 /2 (7.30)
and we consider the X 0 and X 1 D x 1 axes as orthogonal axes in a Euclidean
space. As in all Euclidean spaces, every interval is invariant under a rotation of
the X 0 x 1 plane through an angle into X 00 x 01 :
X 00 D x 1 sin C X 0 cos (7.31a)
D
01 1 0
x D x cos C X sin (7.31b)
See figure 7.2.
If we introduce the angle ' D i , often called the rapidity or the Lorentz
boost parameter, and transform back to the original space and time variables by
using equation (7.28) above backwards, we obtain
ct 0 D x sinh ' C ct cosh ' (7.32a)
0
x D x cosh ' ct sinh ' (7.32b)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 186
. of 300
we let
FT
which are identical to the original transformation equations (7.2) on page 156 if
sinh ' D
ˇ
cosh ' D
tanh ' D ˇ
(7.33a)
(7.33b)
(7.33c)
A
It is therefore possible to envisage the Lorentz transformation as an ‘ordinary’
rotation in the 4D Euclidean space M4 . Such a rotation in M4 corresponds to a
coordinate change in L4 as depicted in figure 7.3. Equation (7.27) on page 162
for successive Lorentz transformation then corresponds to the tanh addition for-
mula
R
tanh '1 C tanh '2
tanh.'1 C '2 / D (7.34)
1 C tanh '1 tanh '2
The use of ict and M4 , which leads to the interpretation of the Lorentz trans-
formation as an ‘ordinary’ rotation, may, at best, be illustrative, but is not very
physical. Besides, if we leave the flat L4 space and enter the curved space of
D
general relativity, the ‘ict ’ trick will turn out to be an impasse. Let us therefore
immediately return to L4 where all components are real valued.
four-velocity
0 1
dx
B C
B c v C D .u0 ; u/
C
u D D
.c; v / D B s ;s (7.35)
d B
2 2
C
@ v v A
1 1
c2 c2
which, when multiplied with the scalar invariant m0 yields the four-momentum
0 1
dx
B C
B m0 c m0 v C
p D m0 D m0
.c; v / D B s ; s C D .p 0 ; p/ (7.36)
d B C
@ v2 v2 A
1 1
c2 c2
FT
From this we see that we can write
p D mv (7.37)
where
m0
m D
m0 D s (7.38)
2
v
1
c2
We can interpret this such that the Lorentz covariance implies that the mass-like
term in the ordinary 3D linear momentum is not invariant. A better way to look
A
at this is that p D mv D
m0 v is the covariantly correct expression for the
kinetic three-momentum.
Multiplying the zeroth (time) component of the four-momentum p by the
scalar invariant c, we obtain
m0 c 2
cp 0 D
m0 c 2 D s D mc 2 (7.39)
R
v2
1
c2
Since this component has the dimension of energy and is the result of a Lorentzco-
variant description of the motion of a particle with its kinetic momentum de-
scribed by the spatial components of the four-momentum, equation (7.36) above,
D
Since this is an invariant, this equation holds in any inertial frame, particularly
in the frame where p D 0 and there we have
E D m0 c 2 (7.42)
FT
dx
j D 0 D 0 u D 0
.c; v / D .c; v / (7.43)
d
with
D 0 (7.44)
is the four-current.
The contravariant form of the four-del operator @ D @=@x is defined in
equation (M.52) on page 244 and its covariant counterpart @ D @=@x in equa-
tion (M.53) on page 244, respectively. As is shown in example M.9 on page 260,
A
the d’Alembert operator is the scalar product of the four-del with itself:
1 @2
2 D @ @ D @ @ D r2 (7.45)
c 2 @t 2
Since it has the characteristics of a four-scalar, the d’Alembert operator is in-
R
variant and, hence, the homogeneous wave equation 2 f .t; x/ D 0 is Lorentz
covariant.
2 A D 0 j (7.47)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 189
. of 300
With the help of the above, we can formulate our electrodynamic equations
covariantly. For instance, the covariant form of the equation of continuity, equa-
tion (4.20) on page 58 is
@ j D 0 (7.48)
and the Lorenz-Lorentz gauge condition, equation (3.16) on page 37, can be
written
@ A D 0 (7.49)
The Lorenz-Lorentz gauge is sometimes called the covariant gauge. The gauge
transformations (3.50) on page 43 in covariant form are
FT
If only one dimension Lorentz contracts (for instance, due to relative motion
along the x direction), a 3D spatial volume element transforms according to
s
1 v2
q
dV D d3x D dV0 D dV0 1 ˇ 2 D dV0 1 (7.51)
c2
where dV0 denotes the volume element as measured in the rest system, then from
equation (7.44) on the facing page we see that
0
ˆ q 1
.A /0 D ;A D ;0 (7.53)
c vD0 4"0 c jx x0 j0
where jx x0 j0 is the usual distance from the source point to the field point,
evaluated in the rest system (signified by the index ‘0’).
Let us introduce the relative position four-vector between the source point
and the field point:
R D x x 0 D c.t t 0 /; x x0
(7.54)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 190
. of 300
We know that in vacuum the signal (field) from the charge q 0 at x 0 propag-
ates to x with the speed of light c so that
ˇx x0 ˇ D c.t t 0 /
ˇ ˇ
(7.56)
R R D 0 (7.57)
FTR D .ˇ x x 0 ˇ ; x x 0 /
ˇ ˇ
(7.58)
Now we want to find the correspondence to the rest system solution, equation
(7.53) on the preceding page, in an arbitrary inertial system. We note from
equation (7.35) on page 165 that in the rest system
0 1
B C
c v
.u /0 D B
B C
B s ; s C
C D .c; 0/ (7.59)
2 2 A
v v
A
@
1 1
c2 c 2 vD0
and
u R D u R 0 D .u /0 .R /0
(7.61)
D .c; 0/ .ˇx x0 ˇ ; .x x0 /0 / D c ˇx x0 ˇ
ˇ ˇ ˇ ˇ
D
0 0
According to equation (7.35) on page 165 and equation (7.58) on the facing
page
u R D
.c; v / ˇx x0 ˇ ; .x x0 / D
c ˇx x0 ˇ v .x x0 /
ˇ ˇ ˇ ˇ
(7.63)
Generalising expression (7.1a) on page 156 to vector form:
def v
ˇ D ˇ vO (7.64)
c
and introducing
def ˇ v .x x0 /
x0 ˇ x0 ˇ x0 /
ˇ ˇ ˇ
s ˇx ˇx ˇ .x (7.65)
c
we can write
and
u
u R D
cs
cu R
D
1 v
;
A .x / D
q0
1 v
4"0 cs c s
; 2
D
ˆ
c
;A
FT
cs c 2 s
(7.66)
(7.67)
(7.68)
A
where in the last step the definition of the four-potential, equation (7.46) on
page 166, was used. Writing the solution in the ordinary 3D way, we conclude
that for a very localised charge volume, moving relative an observer with a ve-
locity v , the scalar and vector potentials are given by the expressions
q0 1 q0 1
R
ˆ.t; x/ D D (7.69a)
4"0 s 4"0 jx x0 j ˇ .x x0 /
q0 v q0 v
A.t; x/ D 2
D (7.69b)
4"0 c s 2
4"0 c jx x j ˇ .x x0 /
0
Ok
c D a b D ij k ai bj x
D .a2 b3 a3 b2 /Ox1 C .a3 b1 a1 b3 /Ox2 C .a1 b2 a2 b1 /Ox3
(7.70)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 192
. of 300
FT
tromagnetic potentials in the following way:
B Dr A (7.74a)
@A
ED rˆ (7.74b)
@t
In component form, this can be written
@Aj @Ai
Bij D i
D @i Aj @j Ai (7.75a)
@x @x j
@ˆ @Ai
Ei D D @i ˆ @ t Ai (7.75b)
@x i @t
A
From this, we notice the clear difference between the axial vector (pseudovector)
B and the polar vector (‘ordinary vector’) E.
Our goal is to express the electric and magnetic fields in a tensor form where
the components are functions of the covariant form of the four-potential, equa-
tion (7.46) on page 166:
R
ˆ
A D ;A (7.76)
c
Inspection of (7.76) and equation (7.75) above makes it natural to define the
four-tensor
@A @A
F D D @ A @ A
D
(7.77)
@x @x
This anti-symmetric (skew-symmetric), four-tensor of rank 2 is called the elec-
tromagnetic field tensor or the Faraday tensor. In matrix representation, the
contravariant field tensor can be written
0 1
0 Ex =c Ey =c Ez =c
BE =c 0 Bz By C
B x
.F / D B (7.78)
C
@Ey =c Bz 0 Bx A
C
Ez =c By Bx 0
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 193
. of 300
We note that the field tensor is a sort of four-dimensional curl of the four-
potential vector A .
The covariant field tensor is obtained from the contravariant field tensor in
the usual manner by index lowering
Ez =c By Bx 0
Comparing formula (7.80) with formula (7.78) on the preceding page we see
FT
that the covariant field tensor is obtained from the contravariant one by a trans-
formation E ! E.
That the two Maxwell source equations can be written
@ F D 0 j (7.81)
For D 1 [the second column in equation (7.78) on the facing page], equa-
tion (7.81) above yields
@F 01 @F 11 @F 21 @F 31 1 @Ex @Bz @By
C C C D C0C
@x 0 @x 1 @x 2 @x 3 c 2 @t @y @z (7.84)
D 0 j 1 D 0 vx
This result can be rewritten as
@Bz @By @Ex
"0 0 D 0 jx (7.85)
@y @z @t
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 194
. of 300
or, equivalently, as
@Ex
.r B/x D 0 jx C "0 0 (7.86)
@t
and similarly for D 2; 3. In summary, we can write the result in three-vector
form as
@E
r B D 0 j.t; x/ C "0 0 (7.87)
@t
which we recognise as the Maxwell source equation for the magnetic field, equa-
tion (2.1d) on page 19.
With the help of the fully antisymmetric pseudotensor of rank 4
8
<1 if ; ; ; is an even permutation of 0,1,2,3
ˆ
FT
ˆ
D 0 if at least two of ; ; ; are equal (7.88)
ˆ
ˆ
: 1 if ; ; ; is an odd permutation of 0,1,2,3
? 1
F D F D ?
F (7.89)
2
A
with the further property
? ?
F D F
(7.90)
Bz Ey =c Ex =c 0
i.e. the dual field tensor is obtained from the ordinary field tensor by the duality
D
@ ?F D 0 (7.94)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 195
. of 300
sometimes referred to as the Jacobi identity. Hence, equation (7.81) on page 171
and equation (7.95) above constitute Maxwell’s equations in four-dimensional
formalism.
It is interesting to note that equation (7.81) on page 171 and
@ ?F D 0 jm (7.96)
7.4 Bibliography
FT
[43] J. A HARONI, The Special Theory of Relativity, second, revised ed., Dover Publica-
tions, Inc., New York, 1985, ISBN 0-486-64870-2.
[44] A. O. BARUT, Electrodynamics and Classical Theory of Fields and Particles, Dover
Publications, Inc., New York, NY, 1980, ISBN 0-486-64038-8.
[47] L. D. L ANDAU AND E. M. L IFSHITZ, The Classical Theory of Fields, fourth re-
vised English ed., vol. 2 of Course of Theoretical Physics, Pergamon Press, Ltd.,
Oxford . . . , 1975, ISBN 0-08-025072-6.
R
[48] F. E. L OW, Classical Field Theory, John Wiley & Sons, Inc., New York, NY . . . ,
1997, ISBN 0-471-59551-9.
[49] H. M UIRHEAD, The Special Theory of Relativity, The Macmillan Press Ltd., London,
Beccles and Colchester, 1973, ISBN 333-12845-1.
D
[50] C. M ØLLER, The Theory of Relativity, second ed., Oxford University Press, Glas-
gow . . . , 1972.
[53] B. S PAIN, Tensor Calculus, third ed., Oliver and Boyd, Ltd., Edinburgh and London,
1965, ISBN 05-001331-9.
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 196
. of 300
FT
A
R
D
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 197
. of 300
FT
general problem of interaction between electric and magnetic fields and electric-
ally charged particles. The analysis is based on Lagrangian and Hamiltonian
methods, is fully covariant, and yields results which are relativistically correct.
electrodynamics.
175
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 198
. of 300
where d is the proper time defined via equation (7.17) on page 160, and L4
acts as a kind of generalisation to the common 3D Lagrangian. As a result, the
variational principle
Z 1
ıS4 D ı L4 .x ; u / d D 0 (8.2)
0
FT
1
Lfree
4 D m0 u u (8.3)
2
Again drawing inferences from analytical mechanics in 3D, we introduce a gen-
eralised interaction between the particles and the electromagnetic field with the
help of the four-potential given by equation (7.76) on page 170 in the following
way
1
L4 D m0 u u C qu A .x / (8.4)
2
A
We call this the four-Lagrangian and shall now show how this function, together
with the variation principle, formula (8.2) above, yields Lorentz covariant results
which are physically correct.
The variation principle (8.2) with the 4D Lagrangian (8.4) inserted, leads to
Z 1
m0
R
ıS4 D ı u u C qu A .x / d
2
Z 1 0
m0 @.u u /
@A
D ıu C q A ıu C u ıx d (8.5)
0 2 @u @x
Z 1
m0 u ıu C q A ıu C u @ A ıx d D 0
D
D
0
Inserting this into the first two terms in the last integral in equation (8.5) on the
preceding page, we obtain
Z 1
d d
ıS4 D m0 u .ıx / C qA .ıx / C qu @ A ıx d (8.8)
0 d d
Partial integration in the two first terms in the right hand member of (8.8) gives
Z 1
du dA
ıS4 D m0 ıx q ıx C qu @ A ıx d (8.9)
0 d d
where the integrated parts do not contribute since the variations at the endpoints
vanish. A change of irrelevant summation index from to in the first two terms
of the right hand member of (8.9) yields, after moving the ensuing common
factor ıx outside the parenthesis, the following expression:
FT
Z 1
du dA
ıS4 D m0 q C qu @ A ıx d
(8.10)
0 d d
Applying well-known rules of differentiation and the expression (7.35) for
the four-velocity, we can express dA =d as follows:
dA @A dx
D D @ A u (8.11)
d @x d
By inserting this expression (8.11) into the second term in right-hand member of
equation (8.10) above, and noting the common factor qu of the resulting term
and the last term, we obtain the final variational principle expression
A
Z 1
du
C qu @ A @ A ıx d
ıS4 D m0 (8.12)
0 d
Since, according to the variational principle, this expression shall vanish and ıx
is arbitrary between the fixed end points 0 and 1 , the expression inside in
R
the integrand in the right hand member of equation (8.12) above must vanish.
In other words, we have found an equation of motion for a charged particle in a
prescribed electromagnetic field:
du
D qu @ A @ A
m0 (8.13)
d
D
With the help of formula (7.79) on page 171 for the covariant component form
of the field tensor, we can express this equation in terms of the electromagnetic
field tensor in the following way:
du
m0 D qu F (8.14)
d
This is the sought-for covariant equation of motion for a particle in an electro-
magnetic field. It is often referred to as the Minkowski equation. As the reader
may easily verify, the spatial part of this 4-vector equation is the covariant (re-
lativistically correct) expression for the Newton-Lorentz force equation.
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 200
. of 300
@H dqi
D (8.15a)
@pi dt
@H dpi
D (8.15b)
@qi dt
FT
canonically conjugate four-momentum p in an analogous way as the ordinary
3D conjugate momentum
@L4
p D (8.16)
@u
H4 D p u L4 (8.17)
A
With the help of these, the position four-vector x , considered as the generalised
four-coordinate, and the invariant line element ds, defined in equation (7.17) on
page 160, we introduce the following eight partial differential equations:
R
@H4 dx
D (8.18a)
@p d
@H4 dp
D (8.18b)
@x d
which form the four-dimensional Hamilton equations.
D
Our strategy now is to use equation (8.16) above and equations (8.18) above
to derive an explicit algebraic expression for the canonically conjugate mo-
mentum four-vector. According to equation (7.40) on page 165, c times a four-
momentum has a zeroth (time) component which we can identify with the total
energy. Hence we require that the component p 0 of the conjugate four-momentum
vector defined according to equation (8.16) above be identical to the ordinary 3D
Hamiltonian H divided by c and hence that this cp 0 solves the Hamilton equa-
1Recall that in 3D, the Hamilto- tions, equations (8.15) above.1 This latter consistency check is left as an exercise
nian equals the total energy. to the reader.
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 201
. of 300
Using the definition of H4 , equation (8.17) on the facing page, and the ex-
pression for L4 , equation (8.4) on page 176, we obtain
1
H4 D p u L4 D p u m0 u u qu A .x / (8.19)
2
Furthermore, from the definition (8.16) of the canonically conjugate four-mo-
mentum p , we see that
@L4 @ 1
p D D m0 u u C qu A .x / D m0 u C qA (8.20)
@u @u 2
Inserting this into (8.19), we obtain
1 1
H4 D m0 u u C qA u m0 u u qu A .x / D m0 u u (8.21)
2 2
FT
Since the four-velocity scalar-multiplied by itself is u u D c 2 , we clearly
see from equation (8.21) above that H4 is indeed a scalar invariant, whose value
is simply
m0 c 2
H4 D (8.22)
2
However, at the same time (8.20) provides the algebraic relationship
1
u D .p qA / (8.23)
m0
A
and if this is used in (8.21) to eliminate u , one gets
m0 1 1
H4 D .p qA / p qA
2 m0 m0
1
.p qA / p qA
D (8.24)
R
2m0
1
p p 2qA p C q 2 A A
D
2m0
That this four-Hamiltonian yields the correct covariant equation of motion can
be seen by inserting it into the four-dimensional Hamilton equations (8.18) and
D
where in the last step equation (8.20) on the previous page was used. Rearran-
ging terms, and using equation (7.79) on page 171, we obtain
du
D qu @ A @ A D qu F
m0 (8.26)
d
which is identical to the covariant equation of motion equation (8.14) on page 177.
We can therefore safely conclude that the Hamiltonian in question yields correct
results.
Recalling expression (7.46) on page 166 for the four-potential, and repres-
enting the canonically conjugate four-momentum as p D .p 0 ; p/, we obtain
the following scalar products:
p p D .p 0 /2 .p/2 (8.27a)
FT
1
A p D ˆp 0 .p A/ (8.27b)
c
1
A A D 2 ˆ2
.A/2 (8.27c)
c
Inserting these explicit expressions into equation (8.24) on the previous page,
and using the fact that H4 is equal to the scalar value m0 c 2 =2, as derived in
equation (8.22) on the preceding page, we obtain the equation
m0 c 2 q2 2
1 0 2 2 2 0 2 2
D .p / .p/ qˆp C 2q.p A/ C 2 ˆ q .A/
2 2m0 c c
A
(8.28)
2q q2
.p 0 /2 ˆp 0 .p/2 2qp A C q 2 .A/2 C 2 ˆ2 m20 c 2 D 0 (8.29)
c … c
R
„ ƒ‚
.p qA/2
is the ordinary 3D Hamilton function for a charged particle moving in scalar and
vector potentials associated with prescribed electric and magnetic fields.
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 203
. of 300
The ordinary Lagrange and Hamilton functions L and H are related to each
other by the 3D transformation [cf. the 4D transformation (8.17) between L4
and H4 ]
L D pv H (8.32)
Using the the explicit expressions given by equation (8.31) and equation (8.32),
we obtain the explicit expression for the ordinary 3D Lagrange function
q
L D pv qˆ c .p qA/2 C m20 c 2 (8.33)
FT
p qA D s D mv (8.34)
v2
1
c2
where the quantity mv is the usual kinetic momentum, we can rewrite this ex-
pression for the ordinary Lagrangian as follows:
q
L D qA v C mv 2 qˆ c m2 v 2 C m20 c 2
(8.35)
s
2 2 2 v2
A
D mv q.ˆ A v/ mc D qˆ C qA v m0 c 1
c2
FT
each with mass m and connected to its neighbour along a one-dimensional straight
line, which we choose to be the x axis, by identical ideal springs with spring con-
stants k (Hooke’s law). At equilibrium the mass points are at rest, distributed
evenly with a distance a to their two nearest neighbours so that the equilibrium
coordinate for the i th particle is xi D iax O . After perturbation, the motion of
mass point i will be a one-dimensional oscillatory motion along x O . Let us denote
the deviation for mass point i from its equilibrium position by i .t/Ox.
As is well known, the solution to this mechanical problem can be obtained if
we can find a Lagrangian (Lagrange function) L which satisfies the variational
A
equation
Z
ı L.i ; Pi ; t/ dt D 0 (8.36)
Let us write the Lagrangian, as given by equation (8.37) above, in the fol-
lowing way:
N
X
LD a Li (8.38)
i D1
where
" 2 #
1 m P2 i C1 i
Li D ka (8.39)
2 a i a
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 205
. of 300
FT
where
" 2 #
@ 2
@ @ 1 @
L ; ; ;t D Y (8.42)
@t @x 2 @t @x
Notice how we made a transition from a discrete description, in which the mass
points were identified by a discrete integer variable i D 1; 2; : : : ; N , to a con-
tinuous description, where the infinitesimal mass points were instead identified
by a continuous real parameter x, namely their position along x O.
A consequence of this transition is that the number of degrees of freedom for
A
the system went from the finite number N to infinity! Another consequence is
that L has now become dependent also on the partial derivative with respect to
x of the ‘field coordinate’ . But, as we shall see, the transition is well worth the
cost because it allows us to treat all fields, be it classical scalar or vectorial fields,
or wave functions, spinors and other fields that appear in quantum physics, on
an equal footing.
R
Under the assumption of time independence and fixed endpoints, the vari-
ation principle (8.36) on the preceding page yields:
@ @
Z ZZ
ı L dt D ı L ; ; dx dt
@t @x
2 3
D
L L L
@ @ @ @ @
ZZ
D 4 ı C ı C ı 5 dx dt D 0
@ @ @ @t @ @ @x
@t @x
(8.43)
As before, the last integral can be integrated by parts. This results in the expres-
sion
2 0 1 0 13
@L @ @L @ @ L
ZZ
4 @ A @ A5 ı dx dt D 0 (8.44)
@ @t @ @ @x @ @
@t @x
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 206
. of 300
where the variation is arbitrary (and the endpoints fixed). This means that the
integrand itself must vanish. If we introduce the functional derivative
0 1
ıL @L @ @ @L A
D (8.45)
ı @ @x @ @
@x
FT
the preceding page, into equation (8.46) above, we obtain the equation of motion
for our one-dimensional linear mechanical structure. It is:
@2 @2
Y D0 (8.47a)
@t 2 @x 2
or !
1 @2 @2
D0 (8.47b)
v'2 @t 2 @x 2
i.e. the one-dimensionalp wave equation for compression waves which propagate
with phase speed v' D Y = along the linear structure.
A
A generalisation of the above 1D results to a three-dimensional continuum is
straightforward. For this 3D case we get the variational principle
@
Z ZZ Z
ı L dt D ı L d3x dt D ı L ; d4x
@x
2 0 13
(8.48)
R
@L @ @ L
ZZ
D 4 @ A5 ı d4x D 0
@x @ @
@
@x
where the variation ı is arbitrary and the endpoints are fixed. This means that
the integrand itself must vanish:
D
0 1
@L @ @ @L A @L @L
D @ D0 (8.49)
@x @ @
@ @ @.@ /
@x
FT
If, as usual, we differentiate this expression and identify terms, we obtain the
following Hamilton density equations
@H @
D (8.54a)
@ @t
ıH @
D (8.54b)
ı @t
The Hamilton density functions are in many ways similar to the ordinary Hamilton
functions for a system of a finite number of particles and lead to similar results.
However, they describe the dynamics of a continuous system of infinitely many
A
degrees of freedom.
The L interaction part describes the interaction between the charged particles
and the external electromagnetic field. A convenient expression for this interac-
tion Lagrange density is
L interaction D j A (8.57)
For the field part L field we choose the difference between magnetic and elec-
tric energy density (in analogy with the difference between kinetic and potential
energy in a mechanical field). With the help of the field tensor, we express this
field Lagrange density as
1
L field D F F (8.58)
40
so that the total Lagrangian density can be written
L tot D
FT
1
2
%0 u u C j A C
L EM D L interaction C L field D j A C
1
40
F F (8.60)
A
inserted into the Euler-Lagrange equations, expression (8.49) on page 184, yields
two of Maxwell’s equations. To see this, we note from equation (8.60) above and
the results in Example 8.1 that
@L EM
D j (8.61)
R
@A
Furthermore,
@L EM
1 @
@ D @ F F
@.@ A / 40 @.@ A /
D
i
1 @ h
D @ .@ A @ A /.@ A @ A /
40 @.@ A /
(
1 @
D @ @ A @ A @ A @ A (8.62)
40 @.@ A /
)
@ A @ A C @ A @ A
1 @
D @ @ A @ A
@ A @ A
20 @.@ A /
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 209
. of 300
But
@ @ @
@ A @ A D @ A @ A C @ A @ A
@.@ A / @.@ A / @.@ A /
@ @
D @ A @ A C @ A ˛ @˛ ˇ Aˇ
@.@ A / @.@ A /
@ @ (8.63)
D @ A @ A C ˛ ˇ @ A @˛ Aˇ
@.@ A / @.@ A /
@ @
D @ A @ A C @˛ Aˇ @˛ Aˇ
@.@ A / @.@ A /
D 2@ A
Similarly,
FT
@
@ A @ A D 2@ A (8.64)
@.@ A /
so that
@L EM
1 1
@ D @ .@ A @ A / D @ F (8.65)
@.@ A / 0 0
This means that the Euler-Lagrange equations, expression (8.49) on page 184,
for the Lagrangian density L EM and with A as the field quantity become
A
@L EM @L EM
1
@ D j @ F D 0 (8.66)
@A @.@ A / 0
or
@ F D 0 j (8.67)
R
which, according to equation (7.81) on page 171, is a Lorentz covariant formu-
lation of Maxwell’s source equations.
!
1 1 B2
F F D "0 E 2
(8.68)
40 2 0
i.e. the difference between the magnetic and electric field energy densities.
From formula (7.78) on page 170 we recall that
0 1
0 Ex =c Ey =c Ez =c
BEx =c 0 Bz By C
F DB @Ey =c
C (8.69)
Bz 0 Bx A
Ez =c By Bx 0
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 210
. of 300
where denotes the row number and the column number. Then, Einstein summation and
direct substitution yields
FT
(8.71)
Ex2 =c 2 C 0 C Bz2 C By2
Ey2 =c 2 C Bz2 C 0 C Bx2
Ez2 =c 2 C By2 C Bx2 C 0
D 2Ex2 =c 2 2Ey2 =c 2 2Ez2 =c 2 C 2Bx2 C 2By2 C 2Bz2
D 2E 2 =c 2 C 2B 2 D 2.B 2 E 2 =c 2 /
where, in the last step, the identity "0 0 D 1=c 2 was used. QED
ied with this technique. As a simple example, consider a real, scalar field
which has the following Lagrange density:
1
L D @ @ m2 2
(8.73)
2
Insertion into the 1D Euler-Lagrange equation, equation (8.46) on page 184,
yields the dynamic equation
.2 m2 / D 0 (8.74)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 211
. of 300
FT
Proca Lagrangian
mc 2
1
L EM D L interaction C L field D j A C F F C A A
40 }
(8.78)
mc 2
@ F A D 0 j (8.79)
}
A
Clearly, this equation describes an electromagnetic field with mass m, or, in
other words, massive photons. If massive photons do exist, large-scale magnetic
fields, including those of the earth and galactic spiral arms, should be signific-
antly modified from what they are to yield measurable discrepancies from their
usual form. Space experiments of this kind on board satellites have led to strin-
R
gent upper bounds on the photon mass. If the photon really has a mass, it will
have an impact on electrodynamics as well as on cosmology and astrophysics.
8.3 Bibliography
D
[54] A. O. BARUT, Electrodynamics and Classical Theory of Fields and Particles, Dover
Publications, Inc., New York, NY, 1980, ISBN 0-486-64038-8.
[58] L. D. L ANDAU AND E. M. L IFSHITZ, The Classical Theory of Fields, fourth revised
English ed., vol. 2 of Course of Theoretical Physics, Pergamon Press, Ltd., Oxford . . . ,
1975, ISBN 0-08-025072-6.
[61] D. E. S OPER, Classical Field Theory, John Wiley & Sons, Inc., New York, London,
Sydney and Toronto, 1976, ISBN 0-471-81368-0.
FT
A
R
D
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 213
. of 300
FT
a correct physical picture for arbitrary field and source distributions, on mac-
roscopic and, under certain assumptions, microscopic scales. A more complete
and accurate theory, valid also when quantum effects are significant, is provided
by quantum electrodynamics. QED gives a consistent description of how elec-
tromagnetic fields are quantised into photons and describes their intrinsic and
extrinsic properties. However, this theory is beyond the scope of the current
book.
In a material medium, be it in a solid, fluid or gaseous state or a combination
thereof, it is sometimes convenient to replace the Maxwell-Lorentz equations
A
(2.1) on page 19 by the corresponding macroscopic Maxwell equations in which
auxiliary, derived fields are introduced. These auxiliary fields, viz., the electric
displacement vector D (measured in C m 2 ) and the magnetising field H (meas-
ured in A m 1 ), incorporate intrinsic electromagnetic properties of macroscopic 1 In the excellent textbook Intro-
matter, or properties that appear when the medium is immersed fully or partially duction to Electrodynamics, Third
R
Edtition, the author D AV I D J .
in an electromagnetic field.1 Consequently, they represent, respectively, elec- G R I F F I T H S writes
tric and magnetic field quantities in which, in an average, approximate sense, ‘Many authors call H,
the material properties of the substances are already included. In the most gen- not B, the “magnetic
field”. Then they have
eral case, these derived fields are complicated, possibly non-local and nonlinear, to invent a new word for
functions of the primary fields E and B : B: the “flux density,” or
magnetic “induction” (an
D
191
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 214
. of 300
D "E (9.2)
1
H B (9.3)
i.e. that the electric displacement vector D.t; x/ is only linearly dependent on
the electric field E.t; x/, and the magnetising field H.t; x/ is only linearly de-
pendent on the magnetic field B.t; x/. In this chapter we derive these linearised
forms, and then consider a simple, explicit linear model for a medium from
FT
which we derive the expression for the dielectric permittivity ".t; x/, the mag-
netic susceptibility .t; x/, and the refractive index or index of refraction n.t; x/
of this medium. Using this simple model, we study certain interesting aspects of
the propagation of electromagnetic particles and waves in the medium.
2
and at the same time introducing the electric displacement vector (C m )
This approximation is often valid for regular media if the field strength jEj is
low enough. Here the variations in time and space of the the material dependent
electric susceptibility, e , are usually on much slower and longer scales than for
E itself.2 Inserting the approximation (9.7) into equation (9.5) on the preceding 2 The fact that the relation between
page, we can write the latter the dipole moment per unit volume
P and the applied electric field E
is local in time and space is yet
D.t; x/ D ".t; x/E.t; x/ (9.8) another approximation assumed in
macroscopic Maxwell theory.
where, approximately,
e .t; x/ D
".t; x/
"0
FT
".t; x/ D "0 Œ1 C e .t; x/ D "0 e .t; x/
D 1 C e .t; x/
(9.9)
(9.10)
will have to be replaced by a tensor. This would still describe a linear relation-
A
ship between E and P but one where the linear proportionality factor, or, as we
shall call it, the dispersive property of the medium, is dependent on the direction
in space.
In general, however, the relationship is not of a simple linear form as in
equation (9.7) above but non-linear terms are important. In such a situation
R
the principle of superposition is no longer valid and non-linear effects such as
frequency conversion and mixing can be expected.3 3 The nonlinearity of semicon-
ductor diodes is used, e.g. in radio
receivers to convert high radio
9.1.2 Magnetisation and the magnetising field frequencies into lower ones, or
into the audible spectrum. These
techniques are called heterodyning
D
An analysis of the properties of magnetic media and the associated currents and demodulation, respectively.
Another example of the nonlinear
shows that three such types of currents exist:
response of a medium is the Kerr
effect.
1. In analogy with true charges for the electric case, we may have true currents
j true , i.e. a physical transport of true (free) charges.
FT
Analogously to the electric case, one may, for a distribution of magnetic dipole
moments in a volume, describe this volume in terms of its magnetisation, or
magnetic dipole moment per unit volume, M. Via the definition of the vector
potential A one can show that the magnetisation current and the magnetisation
is simply related:
jM D r M (9.12)
Ampère-turn density) as
B
HD M (9.14)
0
and using the definition for D, equation (9.5) on page 192, we find that
LHS D r H
@P @D @E
RHS D j true C D j true C "0
@t @t @t
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 217
. of 300
Hence, in this simplistic view, we would pick up a term "0 @E=@t which makes
the equation inconsistent: the divergence of the left hand side vanishes while the
divergence of the right hand side does not! Maxwell realised this and to over-
come this inconsistency he was forced to add his famous displacement current
term which precisely compensates for the last term the RHS expression.4 In 4 This term, which ensures that
chapter 1, we discussed an alternative way, based on the postulate of conserva- electric charge is conserved also
in non-stationary problems, is the
tion of electric charge, to introduce the displacement current. one that makes it possible to turn
We may, in analogy with the electric case, introduce a magnetic susceptibil- the Maxwell equations into wave
equations (see chapter 2) and,
ity for the medium. Denoting it m , we can write hence, the term that, in a way, is
the basis for radio communications
1
H.t; x/ D .t; x/B.t; x/ (9.15) and other engineering applications
of the theory.
where, approximately,
FT
.t; x/ D 0 Œ1 C m .t; x/ D 0 m .t; x/ (9.16)
and
.t; x/
m .t; x/ D D 1 C m .t; x/ (9.17)
0
is the relative permeability. In the case of anisotropy, m will be a tensor, but it
is still only a linear approximation.5 5 This is the case for the Hall
effect which produces a potential
difference across an electric
9.1.3 Macroscopic Maxwell equations conduction current channel,
A
perpendicular to this current,
in the presence of an external
Field equations, expressed in terms of the derived, and therefore in principle magnetic field that is likewise
perpendicular to the current. This
superfluous, field quantities D and H are obtained from the Maxwell-Lorentz
effect was discovered 1879 by the
microscopic equations (2.1) on page 19, by replacing the E and B in the two US physicist E DW I N H E R B E RT
source equations by using the approximate relations formula (9.8) on page 193 H A L L (1855–1938).
R
and formula (9.15), respectively:
r D D true (9.18a)
@B
r ED (9.18b)
@t
D
r B D 0 (9.18c)
@D
r H D j true C (9.18d)
@t
This set of differential equations, originally derived by Maxwell himself, are
called Maxwell’s macroscopic equations. Together with the boundary condi-
tions and the constitutive relations, they describe uniquely (but only approxim-
ately) the properties of the electric and magnetic fields in matter and are conveni-
ent to use in certain simple cases, particularly in engineering applications. How-
ever, the structure of these equations rely on certain linear approximations and
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 218
. of 300
there are many situations where they are not useful or even applicable. There-
fore, these equations, which are the original Maxwell equations (albeit expressed
in their modern vector form as introduced by O L I V E R H E AV I S I D E ), should be
6 It should be recalled that Max- used with some care.6
well formulated these macroscopic
equations before it was known that
matter has an atomistic structure
and that there exist electrically
charged particles such as elec-
trons and protons, which possess 9.2 Phase velocity, group velocity and dispersion
a quantum mechanical prop-
erty called spin that gives rise to
magnetism! If we introduce the phase velocity in the medium as
1 1 c
v' D p Dp Dp (9.19)
" e "0 m 0 e m
FT
p
where, according to equation (1.12) on page 6, c D 1= "0 0 is the speed of
light, i.e. the phase speed of electromagnetic waves, in vacuum. Associated with
the phase speed of a medium for a wave of a given frequency ! we have a wave
vector, defined as
def ! v'
k k kO D k vO ' D (9.20)
v' v'
D je m j1=2
refractive index is larger than unity (e.g. glass and water at optical frequencies),
Such negative refractive index in others, it can be smaller than unity (e.g. plasma and metals at radio and optical
materials, have quite remarkable
electromagnetic properties. frequencies).
It is important to notice that depending on the electric and magnetic proper-
ties of a medium, and, hence, on the value of the refractive index n, the phase
speed in the medium can be smaller or larger than the speed of light:
c !
v' D D (9.23)
n k
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 219
. of 300
where, in the last step, we used equation (9.20) on the facing page.
If the medium has a refractive index which, as is usually the case, dependent
on frequency !, we say that the medium is dispersive. Because in this case also
k.!/ and !.k/, so that the group velocity
@!
vg D (9.24)
@k
has a unique value for each frequency component, and is different from v' .
Except in regions of anomalous dispersion, vg is always smaller than c. In a gas
of free charges, such as a plasma, the refractive index is given by the expression
!p2
n2 .!/ D 1 (9.25)
!2
where
FT
X N q 2
!p2 D (9.26)
"0 m
is the square of the plasma frequency !p . Here m and N denote the mass and
number density, respectively, of charged particle species . In an inhomogen-
eous plasma, N D N .x/ so that the refractive index and also the phase and
group velocities are space dependent. As can be easily seen, for each given fre-
quency, the phase and group velocities in a plasma are different from each other.
If the frequency ! is such that it coincides with !p at some point in the medium,
then at that point v' ! 1 while vg ! 0 and the wave Fourier component at !
A
is reflected there.
L
9.3.1 Vavilov-Cerenkov radiation
As we saw in example 6.4 on page 150, a charge in uniform, rectilinear motion
in vacuum does not give rise to any radiation; see in particular equation (6.206a)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 220
. of 300
c 1
v' D Dp <c (9.30)
n "0
Hence, in this particular medium, the speed of propagation of (the phase planes
FT
of) electromagnetic waves is less than the speed of light in vacuum, which we
know is an absolute limit for the motion of anything, including particles. A
medium of this kind has the interesting property that particles, entering into
the medium at high speeds jv 0 j, which, of course, are below the phase speed
in vacuum, can experience that the particle speeds are higher than the phase
L
speed in the medium. This is the basis for the Vavilov-Cerenkov radiation, more
commonly known in the western literature as Cherenkov radiation, that we shall
now study.
If we recall the general derivation, in the vacuum case, of the retarded (and
advanced) potentials in chapter 3 and the Liénard-Wiechert potentials, equations
A
(6.83) on page 123, we realise that we obtain the latter in the medium by a simple
formal replacement c 7! c=n in the expression (6.84) on page 124 for s. Hence,
the Liénard-Wiechert potentials in a medium characterized by a refractive index
n, are
R
1 q0 1 q0
ˆ.t; x/ D ˇ ˇ D (9.31a)
4"0 ˇˇjx x0 / v 0 ˇ 4"0 s
x0 j n .x c ˇ
1 q0v 0 1 q0v 0
A.t; x/ D ˇ ˇ D (9.31b)
4"0 c 2 ˇˇjx x0 j n .x x0 / v 0 ˇ 4"0 c 2 s
c ˇ
D
where now
x0 / v 0 ˇˇ
ˇ ˇ
.x
x0 ˇ
ˇˇ ˇ
s D ˇˇˇx n (9.32)
c ˇ
The need for the absolute value of the expression for s is obvious in the case
when v 0 =c 1=n because then the second term can be larger than the first term;
if v 0 =c 1=n we recover the well-known vacuum case but with modified phase
speed. We also note that the retarded and advanced times in the medium are [cf.
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 221
. of 300
FT
equation (3.33) on page 40]
0 0 k jx x0 j jx x0 j n
x0 ˇ/ D t
ˇ ˇ
tret D tret .t; ˇx Dt (9.33a)
! c
0
0 0 k jx x j jx x0 j n
x 0 ˇ/ D t C
ˇ ˇ
tadv D tadv .t; ˇx DtC (9.33b)
! c
so that the usual time interval t t 0 between the time measured at the point of
observation and the retarded time in a medium becomes
jx x0 j n
t t0 D (9.34)
c
A
For v 0 =c 1=n, the retarded distance s, and therefore the denominators in
equations (9.31) on the facing page, vanish when
v0 ˇ nv 0
x0 / x0 ˇ x0 ˇ
ˇ ˇ ˇ
n.x D ˇx cos c D ˇx (9.35)
c c
or, equivalently, when
R
c
cos c D (9.36)
nv 0
In the direction defined by this angle c , the potentials become singular. Dur-
ing the time interval t t 0 given by expression (9.34) above, the field exists within
a sphere of radius jx x0 j around the particle while the particle moves a distance
D
l 0 D .t t 0 /v 0 (9.37)
FT
radiation. This was indeed the
correct interpretation, but the paper
2
was rejected. The paper was then This Fourier component can be used in the formulæ derived for a linear current
sent to Physical Review and was,
after some controversy with the in subsection 6.4.1 on page 117 if only we make the replacements
American editors, who claimed
the results to be wrong, eventually "0 7! " D n2 "0 (9.41a)
published in 1937. In the same
n!
year, I G O R ’ E V G E N ’ E V I C H k 7! (9.41b)
TA M M (1895–1975) and I LYA c
M I K H A I L OV I C H F R A N K (1908–
1990) published the theory for the In this manner, using j! from equation (9.40) above, the resulting Fourier trans-
effect (‘the singing electron’). L
forms of the Vavilov-Cerenkov magnetic and electric radiation fields can be cal-
In fact, predictions of a similar
culated from the expressions (5.23) on page 92 and (5.18) on page 90, respect-
A
effect had been made as early as
1888 by O L I V E R H E AV I S I D E ively.
(1850–1925), and by A R N O L D
JOHANNES WILHELM SOM-
The total energy content is then obtained from equation (6.9) on page 106
M E R F E L D (1868–1951) in his (integrated over a closed sphere at large distances). For a Fourier component
1904 paper ‘Radiating body mov-
ing with velocity of light’. On 8
one obtains [cf. equation (6.12) on page 107]
May, 1937, Sommerfeld sent a
R
ˇZ ˇ2
letter to Tamm via Austria, saying rad 1 ˇ 3 0 ık x0 ˇ
ˇ
that he was surprised that his old
U! d ˇ d x .j! k/e ˇ d
4"0 nc V 0 ˇ
1904 ideas were now becoming ˇ2 (9.42)
q 0 2 n! 2 ˇˇ 1
ˇZ h !
interesting. Tamm, Frank and
i
0 0ˇ 2
ˇ
L
Cerenkov received the Nobel Prize D exp ix k cos dx ˇ sin d
16 3 "0 c 3 ˇ 1 v0
in 1958 ‘for the discovery and
L
the interpretation of the Cerenkov where is the angle between the direction of motion, x O 01 , and the direction to the
D
effect’ [V I TA L I Y L A Z A R E V I C H
G I N Z B U R G (1916–2009), private observer, k.O The integral in (9.42) is singular of a ‘Dirac delta type’. If we limit
communication]. The Vavilov- the spatial extent of the motion of the particle to the closed interval Œ X; X on
L
Cerenkov cone is similar in nature
to the Mach cone in acoustics. the x 0 axis we can evaluate the integral to obtain
h 0
i
q 02
n! 2
sin 2
sin2 1 nvc cos X! v 0
U!rad d D 2 d (9.43)
4 3 "0 c 3 0
1 nv cos !0
c v
FT
another variable substitution we can therefore approximate
h 0
i
Z 1 sin2 1 C nvc X! v 0
c2
cX
Z 1
sin2 x
2
sin c i2 d 1 dx
n2 v 02 !n x2
h 0
1
1 C nvc v!0 1
c2
cX
D 1
!n n2 v 02
(9.45)
A
leading to the final approximate result for the total energy loss in the frequency
interval .!; ! C d!/
q02X c2
UQ !rad d! D 1 ! d! (9.46)
2"0 c 2 n2 v 02
R
As mentioned earlier, the refractive index is usually frequency dependent.
Realising this, we find that the radiation energy per frequency unit and per unit
length is
D
UQ !rad d! q02! c2
D 1 d! (9.47)
2X 4"0 c 2 n2 .!/v 02
This result was derived under the assumption that v 0 =c > 1=n.!/, i.e. under
the condition that the expression inside the parentheses in the right hand side is
positive. For all media it is true that n.!/ ! 1 when ! ! 1, so there exist
L
always a highest frequency for which we can obtain Vavilov-Cerenkov radiation
from a fast charge in a medium. Our derivation above for a fixed value of n is
valid for each individual Fourier component.
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 224
. of 300
1 @2 E r @j
r 2E D 0 (9.48a)
c 2 @t 2 "0 @t
1 @2 B
r 2 B D 0 r j (9.48b)
c 2 @t 2
where the charge density and the current density j were viewed as the sources
of the wave fields. As we recall, these wave equations were derived from the
Maxwell-Lorentz equations (2.1) on page 19, taken as an axiomatic foundation,
FT
or postulates, of electromagnetic theory. As such, these equations just state what
relations exist between (the second order derivatives of) the fields, i.e. essentially
dynamic generalisations of the Coulomb and Ampère forces, and the dynamics
of the charges (charge and current densities) in the region under study.
Even if the and j terms in the Maxwell-Lorentz equations are often referred
to as the source terms, they can equally well be viewed as terms that describe
the impact on matter in a particular region upon which an electromagnetic wave,
produced in another region with its own charges and currents, impinges. In order
to do so, one needs to find the constitutive relations that describe how charge
and current densities are induced by the impinging fields. Then one can solve
A
the wave equations (9.48) above. In general, this is a formidable task, and one
must often resort to numerical methods.
Let us, for simplicity, assume that the linear relations, as given by formula
(9.8) on page 193 and formula (9.15) on page 195, hold, and that there is also
a linear relation between the electric field E and the current density, known as
R
Ohm’s law:
where is the conductivity of the medium. Let us make the further assumption
that " D ".x/, D .x/, and .x/ are not explicitly dependent on time and are
D
@E @2 E
r 2E .x/ .x/ ".x/.x/ D .r E/ r ln .x/
@t @t 2 (9.50a)
r Œr ln ".x/ E C r
"
2
@H @ H
r 2 H .x/ .x/ ".x/.x/ 2 D .r H/ r ln ".x/
@t @t (9.50b)
r Œr ln .x/ H C r ln ".x/ ..x/E/ Œr .x/ E
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 225
. of 300
For the case D 0 (no magnetisation) and D Const in the medium, equa-
tions (9.50) on the facing page simplify to
@E @2 E
r 2E 0 .x/ ".x/0 2 D r Œr ln ".x/ E (9.51a)
@t @t
@B @2 B
r 2B 0 .x/ ".x/0 2 D .r B/ r ln ".x/
@t @t (9.51b)
C0 r ln ".x/ .E/ 0 Œr .x/ E
Making the further assumption that the medium is not conductive, i.e. that
D 0, the uncoupled wave equations
".x/ @2 E
r 2E D r fŒr ln ".x/ Eg (9.52a)
"0 c 2 @t 2
FT
".x/ @2 B
r 2B D .r B/ Œr ln ".x/ (9.52b)
"0 c 2 @t 2
are obtained.
where m and q are the mass and charge of the electron, respectively, the ef-
fective collision frequency representing the frictional dissipative force from the
surrounding medium, and E the effective applied electric field sensed by the
electron. For a Fourier component of the electric field E D E0 exp . i!t/, the
equation of motion becomes
q2
! 2 qx.t/ i!qx.t/ D E (9.54)
m
If the electron is at equilibrium x D 0 when E D 0, then its dipole moment is
d.t / D qx.t /. Inserting this in equation (9.54) above, we obtain
q2
dD E (9.55)
m.! 2 C i!/
This is the the lowest order contribution to the dipole moment of the medium
FT
from each electron under the influence of the assumed electric field. If Nd .x/
electrons per unit volume can be assumed to give rise to the electric polarisation
P , this becomes
Nd q 2
P D Nd d D E (9.56)
m.! 2 C i!/
Using this in formula (9.5) on page 192, one finds that
is the refractive index. At points in the medium where the wave frequency !
equals this plasma frequency and the collision frequency vanishes, the refract-
ive index n D 0, and the wave is totally reflected. In the ionised outer part of the
atmosphere called the ionosphere this happens for radio waves of frequencies
up to about 5–10 MHz dependent on the solar radiation which causes most of
the ionisation. This is the basis for over-the-horizon radio communications and
is also the reason why low-frequency radio signals from space do not reach the
surface of the Earth.
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 227
. of 300
FT
@E 1 @2 E
r 2E 0 D0 (9.61)
@t c 2 @t 2
@B 1 @2 B
r 2B 0 D0 (9.62)
@t c 2 @t 2
which are the homogeneous vector wave equations for E and B in a conducting
medium without EMF.
We notice that for the simple propagation media considered here, the wave
equation for the magnetic field B has exactly the same mathematical form as
the wave equation for the electric field E, equation (9.61) above. Therefore,
A
in this case it suffices to consider only the E field, since the results for the B
field follow trivially. For EM waves propagating in more complicated media,
containing, e.g. inhomogeneities, the wave equations for E and for B do not
have the same mathematical form.
Following the spectral component prescription leading to equation (2.23) on
R
page 26, we obtain, in the special case under consideration, the following time-
independent wave equation
!2
r 2 E0 C 2 1 C ı E0 D 0 (9.63)
c "0 !
D
Multiplying by e ı!t and introducing the relaxation time D "0 = of the me-
dium in question, we see that the differential equation for each spectral compon-
ent can be written
!2 ı
r 2 E.t; x/ C 2 1 C E.t; x/ D 0 (9.64)
c !
In the limit of long (low conductivity ), (9.64) tends to
!2
r 2E C ED0 (9.65)
c2
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 228
. of 300
r 2 E C i!0 E D 0 (9.66)
FT
c
p
we can write, using the fact that c D 1= "0 0 according to equation (1.12) on
page 6,
r
1 1 0
D D D D R0 (9.68)
! "0 ! "0 ck k "0 k
where in the last step we used the characteristic impedance of vacuum defined
according to formula (6.4) on page 105.
@ @t
@B
nO D0 (9.70c)
@
@B @E @E
nO D 0 j.t; x/ C "0 0 D 0 E C "0 0 (9.70d)
@ @t @t
O we find that
Scalar multiplying (9.70d) by n,
@B @
0 D nO nO D nO 0 C "0 0 E (9.71)
@ @t
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 229
. of 300
which simplifies to the first-order ordinary differential equation for the normal
component En of the electric field
dEn
C En D 0 (9.72)
dt "0
with the solution
t ="0 t=
En D En0 e D En0 e (9.73)
This, together with (9.70a), shows that the longitudinal component of E, i.e. the
component which is perpendicular to the plane surface is independent of and
has a time dependence which exhibits an exponential decay, with a decrement
given by the relaxation time in the medium.
FT
O we similarly find that
Scalar multiplying (9.70b) by n,
@E @B
0 D nO nO D nO (9.74)
@ @t
or
@B
nO D0 (9.75)
@t
From this, and (9.70c), we conclude that the only longitudinal component of B
must be constant in both time and space. In other words, the only non-static
A
solution must consist of transverse components.
so that D 0, then the equation takes the form of the one-dimensional wave
equation
@2 E 1 @2 E
D0 (9.77)
@ 2 c 2 @t 2
As is well known, each component of this equation has a solution which can be
written
FT
By introducing the wave vector
! !
k D k nO D nO D kO (9.81)
c c
this solution can be written as
!t/
E D E0 ei.k x (9.82)
Let us consider the lower sign in front of k in the exponent in (9.80). This
corresponds to a wave which propagates in the direction of increasing . Insert-
ing this solution into equation (9.70b) on page 206, gives
A
@E
nO D ı!B D ik nO E (9.83)
@
or, solving for B,
k 1 1 p
R
BD nO E D k E D kO E D "0 0 nO E (9.84)
! ! c
Hence, to each transverse component of E, there exists an associated magnetic
field given by equation (9.84) above. If E and/or B has a direction in space
which is constant in time, we have a plane wave.
Allowing now for a finite conductivity in our medium, and making the
D
spectral component Ansatz in equation (9.76) on the previous page, we find that
the time-independent telegrapher’s equation can be written
@2 E 2 @2 E
C "0 0 ! E C ı 0 !E D C K 2E D 0 (9.85)
@ 2 @ 2
where
!2
2 2 2
K D "0 0 ! 1Cı D 2 1Cı Dk 1Cı (9.86)
"0 ! c "0 ! "0 !
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 231
. of 300
where, in the last step, equation (9.67) on page 206 was used to introduce the
wave number k. Taking the square root of this expression, we obtain
r
K Dk 1Cı D ˛ C ıˇ (9.87)
"0 !
ˇ2 D ˛2 k2 (9.89)
2
k
FT
˛ˇ D (9.90)
2"0 !
Squaring the latter and combining with the former, one obtains the second order
algebraic equation (in ˛ 2 )
k4 2
˛ 2 .˛ 2 k2/ D (9.91)
4"20 ! 2
ˇ i.˛ !t/
E D E0 e e (9.93)
D
With the aid of equation (9.84) on the facing page we can calculate the associated
magnetic field, and find that it is given by
1 O 1 1
BD K k E D .kO E/.˛ C ıˇ/ D .kO E/ jAj eı
(9.94)
! ! !
where we have, in the last step, rewritten ˛ C ıˇ in the amplitude-phase form
jAj exp.ı
/. From the above, we immediately see that E, and consequently also
B, is damped, and that E and B in the wave are out of phase.
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 232
. of 300
In this limit we find that when the wave impinges perpendicularly upon the me-
dium, the fields are given, inside the medium, by
r r
0 0 ! 0 !
E D E0 exp exp ı !t (9.96a)
2 2
r
0
B0 D .1 C i/ .nO E0 / (9.96b)
FT
2!
@2 E @E @2 E
" D0 (9.98)
@ 2 @t @t 2
R
describing (1D) wave propagation in a material medium.
In chapter 2 we concluded that the existence of a finite conductivity, mani-
festing itself in a collisional interaction between the charge carriers, causes the
waves to decay exponentially with time and space. Let us therefore assume that
D
@2 E @2 E
" D0 (9.99)
@ 2 @t 2
!t/
E D E0 ei.k x (9.100)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 233
. of 300
where now k is the wave vector in the medium given by equation (9.20) on
page 196. With these definitions, the vacuum formula for the associated mag-
netic field, equation (9.84) on page 208,
p O 1 O 1
BD " k E D kED kE (9.101)
v' !
FT
e
r E D (9.102a)
"0
@B
r ED 0 j m (9.102b)
@t
r B D 0 m (9.102c)
@E
r B D "0 0 C 0 j e (9.102d)
@t
under the assumption of vanishing net electric and magnetic charge densities and in the
absence of electromotive and magnetomotive forces. Interpret this equation physically.
A
Assume, for symmetry reasons, that there exists a linear relation between the magnetic
current density j m and the magnetic field B (the magnetic dual of Ohm’s law for electric
currents, j e D e E)
jm D m B (9.103)
Taking the curl of (2.2c) and using (2.2d), one finds, noting that "0 0 D 1=c 2 , that
R
@
r .r E/ D 0 r j m .r B/
@t
@ 1 @E
D 0 m r B 0 j e C 2 (9.104)
@t c @t
1 @2 E
1 @E @E
D 0 m 0 e E C 2 0 e
D
c @t @t c 2 @t 2
Using the vector operator identity r .r E/ D r .r E/ r 2 E, and the fact that
r E D 0 for a vanishing net electric charge, we can rewrite the wave equation as
m @E 1 @2 E
r 2 E 0 e C 2 20 m e E D 0 (9.105)
c @t c 2 @t 2
This is the homogeneous electromagnetodynamic wave equation for E that we were after.
Compared to the ordinary electrodynamic wave equation for E, equation (9.61) on
page 205, we see that we pick up extra terms. In order to understand what these extra
terms mean physically, we analyse the time-independent wave equation for a single Fourier
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 234
. of 300
m !2
r 2 E C i!0 e C 2 E C 2 E 20 m e E
c c
2
" # (9.106)
e C m =c 2
! 1 0 m e
D r2E C 2 1 C i ED0
c ! 2 "0 "0 !
Realising that, according to formula (6.4) on page 105, 0 ="0 is the square of the vacuum
radiation resistance R0 , and rearranging a bit, we obtain the time-independent wave equa-
tion in Dirac’s symmetrised electrodynamics
0 1
!
!2 R02 m e e C m =c 2
r2E C
B C
B1 C i
c2
1
!2
R02 AE D 0;
C
(9.107)
@
"0 ! 1 m e
FT
!2
p
! ¤ R0 m e
From this equation we conclude that the existence of magnetic charges (magnetic mono-
poles), and non-vanishing electric and magnetic conductivities would lead to a shift in the
effective wave number of the wave. Furthermore, even if the electric conductivity e van-
ishes, the imaginary term does not necessarily vanish and the wave therefore experiences
damping or growth according as m is positive or negative, respectively. This would hap-
pen in a hypothetical medium which is a perfect insulator for electric currents but which
can carry magnetic currents.
p
A
def
Finally, we note that in the particular case ! D R0 m e !m , the time-independent
wave equation equation (9.106) above becomes a time-independent diffusion equation
m
r 2 E C i!0 e C 2 E D 0 (9.108)
c
R
which in time domain corresponds to the time-dependent diffusion equation
@E
Dr 2 E D0 (9.109)
@t
1
DD (9.110)
m
0 e C c2
Hence, electromagnetic waves with this particular frequency do not propagate. This means
that if magnetic charges (monopoles) exist in a region in the Universe, electromagnetic
waves propagating through this region would, in this simplistic model, exhibit a lower cutoff
at ! D !m . This would in fact impose a lower limit on the mass of the photon, the quantum
of the electromagnetic field that we shall come across later.
9.5 Bibliography
[62] E. H ALLÉN, Electromagnetic Theory, Chapman & Hall, Ltd., London, 1962.
[63] J. D. JACKSON, Classical Electrodynamics, third ed., John Wiley & Sons, Inc.,
New York, NY . . . , 1999, ISBN 0-471-30932-X.
FT
A
R
D
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 236
. of 300
FT
A
R
D
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 237
. of 300
FORMULÆ
FT
Let x be the position vector (radius vector, coordinate vector) from the origin
of the Euclidean space R3 coordinate system to the coordinate point .x1 ; x2 ; x3 /
in the same system and let jxj denote the magnitude (‘length’) of x. Let further
˛.x/; ˇ.x/; : : :, be arbitrary scalar fields, a.x/; b.x/; : : : , arbitrary vector fields,
and A.x/; B.x/; : : : , arbitrary rank two tensor fields in this space. Let denote
complex conjugate and denote Hermitian conjugate (transposition and, where
A
applicable, complex conjugation).
The differential vector operator r is in Cartesian coordinates given by
3
X @ def @ def
r Oi
x x Oi @ (F.1)
@xi @xi
iD1
R
O i , i D 1; 2; 3 is the i th unit vector and x
where x O1 x
O, x
O 2 y,
O and x
O 3 zO . In
component (tensor) notation r can be written
@ @ @ @ @ @
ri D @i D ; ; D ; ; (F.2)
@x1 @x2 @x3 @x @y @z
D
3
X @ def @ def
r0 Oi
x xO i 0 @0 (F.3)
@xi0 @xi
iD1
or
@ @ @ @ @ @
ri0 D @0i D 0 ; 0 ; D ; 0; 0 (F.4)
@x1 @x2 @x30 0
@x @y @z
215
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 238
. of 300
216 j 6. FORMULÆ
O 1 C sin x
¡O D cos x O2 (F.5a)
O D
® O 1 C cos x
sin x O2 (F.5b)
zO D x
O3 (F.5c)
O 1 D cos ¡O
x O
sin ® (F.6a)
FT
O 2 D sin ¡O C cos ®
x O (F.6b)
O 3 D zO
x (F.6c)
O D d ¡O C d' ®
dl D dx x O C dz zO (F.7)
T HE GRADIENT
@˛ 1 @˛ @˛
r ˛ D ¡O C ™O C zO (F.10)
@ @' @z
T HE DIVERGENCE
T HE CURL
1 @az @a'
r a D ¡O
@' @z
@a @az
C® O (F.12)
@z @
1 @.a' / @a
C zO
@ @'
T HE L APLACIAN
1 @2 ˛ @2 ˛
1 @ @˛
r 2˛ D C 2 2C 2 (F.13)
@ @ @' @z
F .1.2
F .1.2.1
Spherical polar coordinates
Base vectors
C ARTESIAN TO SPHERICAL POLAR
O D dr rO C r d ™O C r sin d' ®
dl D dx x O (F.16)
D
dS D r 2 d rO C r sin dr d' ™O C r dr d ®
O (F.18)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 240
. of 300
218 j 6. FORMULÆ
dV D d3x D dr r 2 d (F.19)
T HE GRADIENT
@˛ 1 @˛ 1 @˛
r ˛ D rO C ™O O
C® (F.20)
@r r @ r sin @'
FT
T HE DIVERGENCE
T HE CURL
1 @.a' sin / @a
r a D rO
r sin @ @'
1 1 @a @.ra '/
A
O r
C™ (F.22)
r sin @' @r
1 @.ra / @ar
C® O
r @r @
T HE L APLACIAN
R
@2 ˛
2 1 @ 2 @˛ 1 @ @˛ 1
r ˛D 2 r C 2 sin C (F.23)
r @r @r r sin @ @ r 2 sin2 @' 2
O ix
13 D x Oi (F.24)
(
0 if i ¤ j
ıij D (F.26)
1 if i D j
FT
8
<1 if i; j; k is an even permutation of 1,2,3
ˆ
ˆ
ij k D 0 if at least two of i; j; k are equal (F.28)
ˆ
ˆ
: 1 if i; j; k is an odd permutation of 1,2,3
S1 D @0 0 iA D i.1j k / (F.33a)
B C
0 i 0
0 1
0 0 i
S2 D @ 0 0 0A D i.2j k / (F.33b)
B C
i 0 0
0 1
0 i 0
S3 D @ i 0 0A D i.3j k / (F.33c)
B C
0 0 0
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 242
. of 300
220 j 6. FORMULÆ
FT
Tr.a ˝ b/ D a b (F.43)
a ˝.b C c/ D a ˝ b C a ˝ c (F.44)
.a C b/ ˝ c D a ˝ c C b ˝ c (F.45)
c a ˝ b D .c a/b (F.46)
a ˝ b c D a.b c/ (F.47)
.a b/ D ia S ˝ b D ia ˝ S b (F.48)
c a ˝ b D .c a/ ˝ b (F.49)
A
a ˝ b c D a ˝.b c/ (F.50)
a ˝ b c ˝ d D .b c/a ˝ d (F.51)
a .b c/ D .a b/ c (F.52)
a .b c/ D b ˝ a c c ˝ a b D b.a c/ c.a b/ (F.53)
.a b/ c D b ˝ a c a ˝ b c D b.a c/ a.b c/ (F.54)
R
a .b c/ C b .c a/ C c .a b/ D 0 (F.55)
.a b/ .c d/ D a Œb .c d/ D .a c/.b d/ .a d/.b c/ (F.56)
.a b/ .c d/ D .a b d/c .a b c/d (F.57)
D
13 a D a 13 D a (F.58)
13 a D a 13 (F.59)
0 1
0 a3 a2
.13 a/ D .a 13 / D @ a3 0 a1 A D iS a (F.60)
B C
a2 a1 0
a .13 b/ D a b (F.61)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 243
. of 300
a .13 b/ D a b (F.62)
13 .a b/ D b ˝ a a˝b (F.63)
O i @i ˛
r˛ D x (F.64)
r a D @i ai (F.65)
a r D ai @i (F.66)
O i @j ak
r a D ij k x (F.67)
O i aj @k
a r D ij k x (F.68)
r ˝a D x
Oi ˝x
O j @i aj (F.69)
FT
a˝r D x
Oi ˝x
O j ai @j (F.70)
Tr.r ˝ a/ D r a (F.71)
O j @i Aij
r A D x (F.72)
a ˝ r b D a@i bi (F.73)
2
r r˛ D r ˛ (F.74)
r ˝ r a D r .r a/ D x
O i @i @j aj (F.75)
a r ˝ r D .a r /r D x
O i aj @j @i (F.76)
A
r .˛ˇ/ D ˛r ˇ C ˇr ˛ (F.77)
r ˝.˛a/ D .r ˛/ ˝ a C ˛r ˝ a (F.78)
r .a b/ D a .r b/ C b .r a/ C a r ˝ b C b r ˝ a (F.79)
r .a b/ D .r a/ b .r b/ a (F.80)
R
r .˛a/ D a r ˛ C ˛r a (F.81)
2
.r ˛/ r ˇ D r .˛r ˇ/ ˛r ˇ (F.82)
r .a b/ D b .r a/ a .r b/ (F.83)
.r ˛/ .r a/ D r .a r ˛/ (F.84)
.r a/ .r b/ D b Œr .r a/ r Œ.r a/ b (F.85)
D
222 j 6. FORMULÆ
FT
In the following S is the matrix vector defined in formula (F.32) and k is an
arbitrary constant vector.
.r a/ D
r .13 ˛/ D r ˛
r .13 a/ D r a
r .13 a/ D r a
ir S ˝ a D
r x D 3
ir ˝ S a
(F.102)
(F.103)
(F.104)
(F.105)
(F.106)
A
r xD0 (F.107)
r ˝ x D 13 (F.108)
r .x k/ D k (F.109)
r .x a/ D a C x.r a/ C .x r / a (F.110)
R
x
r jxj D (F.111)
jxj
x x0
r jx x0 j D D r 0 jx x0 j
0
(F.112)
jx x j
1 x
r D (F.113)
D
jxj jxj3
0
1 x x 0 1
r D D r (F.114)
jx x0 j jx x0 j3 jx x0 j
x 1
r 3
D r2 D 4ı.x/ (F.115)
jxj jxj
x x0
1
r 0 3
D r2 D 4ı.x x0 / (F.116)
jx x j jx x0 j
k 1 kx
r D k r D (F.117)
jxj jxj jxj3
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 245
. of 300
x kx
r k D r if jxj ¤ 0 (F.118)
jxj3 jxj3
k 1
r2 D kr 2 D 4kı.x/ (F.119)
jxj jxj
r .k a/ D k.r a/ C k .r a/ r .k a/ (F.120)
FT
Z I I
3
d x r˛ D dS ˛ d2x n˛
O (F.121a)
V S S
Z I I
d3x r a D dS a d2x nO a (F.121b)
V
Z IS IS
d3x r a D dS a d2x nO a (F.121c)
V S S
Z I I
d3x r ˝ a D dS ˝ a d2x nO ˝ a (F.121d)
V S S
Z I I
d3x r A D dS A d2x nO A (F.121e)
A
V S S
Z I I
d3x r A D dS A d2x nO A (F.121f)
V S S
If S.C / is an open surface bounded by the contour C.S/, whose line element
is dl, then
R
I Z
dl ˛ D dS r ˛ (F.122)
I C ZS
dl a D dS r a (F.123)
C S
D
224 j 6. FORMULÆ
A.x0 /
1
Z Z
3 0 3 0 0 2
r r dx D d x A.x /r D 4A.x/ (F.126)
V0 jx x0 j V0 jx x0 j
where A D ˛ or A D a.
a.x0 /
1
Z Z
3 0 3 0 0 0 0
r ˝r dx D d x Œr a.x /r
V0 jx x0 j V0 jx x0 j
0 ˝.x (F.127)
x0 /
a.x /
I
2 0 0
C d x nO
FT
S0 jx x0 j3
a.x0 / 0
3 0 a.x /
Z Z
r r d3x 0 D r ˝ r d x
V0 jx x0 j V0 jx x0 j
a.x0 /
Z
r r d3x 0
V0 jx x0 j
1
Z
D 4a.x/ d3x 0 Œr 0 a.x0 /r 0
V0 jx x0 j
a.x0 / ˝.x x0 /
I
C d2x 0 nO 0
A
S0 jx x0 j3
(F.128)
H ELMHOLTZ DECOMPOSITION
Any regular, differentiable vector field u that falls off sufficiently fast asymptot-
R
ically can be decomposed into two components, one irrotational and one rota-
tional, such that
u irrot
.x/ D r d3x 0 (F.129b)
V0 4 jx x0 j
r 0 u.x0 /
Z
urotat .x/ D r d3x 0 (F.129c)
V0 4 jx x0 j
If v is a vector field with the same properties as u, or u itself, then
Z
d3x uirrot .x/ v rotat .x/ D 0 (F.130a)
ZV
d3x uirrot .x/ v irrot .x/ D 0 (F.130b)
V
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 247
. of 300
r 0 uirrot .x0 /
Z Z Z
d3x uirrot .x/ v irrot .x/ D d3x Œr v irrot .x/ d3x 0
V V V0 4 jx x0 j
(F.130c)
r 0 v rotat .x0 /
Z Z Z
d3x urotat .x/ v rotat .x/ D d3x Œr urotat .x/ d3x 0
V V V0 4 jx x0 j
(F.130d)
r 0 v rotat .x0 /
Z Z Z
3 3
d xu rotat
.x/ v rotat
.x/ D d x Œr u rotat
.x/ d3x 0
V V V0 4 jx x0 j
(F.130e)
r 0 v rotat .x0 /
Z Z Z
3 3
d xu irrot
.x/ v rotat
.x/ D d x Œr u irrot
.x/ d3x 0
V V V0 4 jx x0 j
(F.130f)
F .2
F .2.1
The electromagnetic field
e
FT
A
r E D (F.131)
"0
r B D 0 m (F.132)
@B
r EC D 0 j m (F.133)
@t
1 @E
R
r B D 0 j e (F.134)
c 2 @t
1 1
cDp .D 299 792 458 ms / (F.135)
"0 0
r
0
D R0 .D 119:9169832 376:7 / (F.136)
"0
D "E (F.137)
1
H B (F.138)
j E (F.139)
P "0 E (F.140)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 248
. of 300
226 j 6. FORMULÆ
@A
ED rˆ (F.141a)
@t
B Dr A (F.141b)
8
<˛ D 1 ) Lorenz-Lorentz gauge
ˆ
c2
ˆ
˛ @ˆ
FT
r A C D 0; ˛D ; ˛ D 0 ) Coulomb gauge
c 2 @t v2 ˆ
ˆ
:˛ D 1 ) Kirchhoff gauge
(F.142)
@.t; x/
ˆ.t; x/ 7! ˆ0 .t; x/ D ˆ.t; x/ (F.143a)
@t
A
A.t; x/ 7! A0 .t; x/ D A.t; x/ C r .t; x/ (F.143b)
1
ufield D "0 .E E C c 2 B B/ (F.144)
2
D
1
SD EB (F.145)
0
g D "0 E B (F.146)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 249
. of 300
T D ufield 13 "0 E ˝ E C c 2 B ˝ B
(F.147)
field 2
Tij D u ıij "0 Ei Ej "0 c Bi Bj (F.148)
h D .x x0 / g D .x x0 / "0 E B (F.149)
F .2.4
F .2.4.1
Efar
! .x/
Electromagnetic radiation
K.x0 / D .x
k eikjx x0 j
Z
d3x 0
FT
x0 / T
ik .x0 x0 /
kO kO
(F.150)
A
i j! e (F.151)
4"0 c jx x0 j V0
ikjx x0 j
k e
i O nO
.I ! .x0 / n/ (F.152)
4"0 c jx x0 j
k eikjx x0 j
Z
0
B!far .x/ d3x 0 j! e ik .x x0 / kO
i (F.153)
4"0 c 2 jx x0 j V0
R
ikjx x0 j
k e
i I ! .x0 / nO (F.154)
4"0 c 2 jx x0 j
Z
0
I ! .x0 / D d3x 0 j! .x0 / e ik nO .x x0 / (F.155)
V0
D
k 2 eikjx x0 j O kO
Efar
! .x/ D .d! .x0 / k/ (F.156)
4"0 jx x0 j
k 2 eikjx x0 j
B!far .x/ D d! .x0 / kO (F.157)
4"0 c jx x0 j
Z
d.t; x0 / D d3x 0 .x0 x0 / .tret
0
; x0 / (F.158)
V0
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 250
. of 300
228 j 6. FORMULÆ
k 2 eikjx x0 j
Efar
! .x/ D m! .x0 / kO (F.159)
4"0 c jx x0 j
k 2 eikjx x0 j
B!far .x/ D m! .x0 / kO kO
2
(F.160)
4"0 c jx x0 j
1
Z
m.t; x0 / D d3x 0 .x0 x0 / j.tret
0
; x0 / (F.161)
2 V0
k 3 eikjx x0 j O
FT
Efar k Q! .x0 / kO kO
! .x/ D i (F.162)
8"0 jx x0 j
k 3 eikjx x0 j O
B!far .x/ D i k Q! .x0 / kO
(F.163)
8"0 c jx x0 j
Z
Q.t; x0 / D d3x 0 .x0 x0 / ˝.x0 x0 / .tret0
; x0 / (F.164)
V0
1O
A
BD kE (F.165)
c
v 02 x0 / a0
R
q 0 .x
E.t; x/ D .x x0 / 1 C .x x/
4"0 s 3 c2 c2
(F.166)
1 x x0
B.t; x/ D E.t; x/ (F.167)
c jx x0 j
dx0 .t 0 /
D
v 0 .t 0 / D (F.168)
dt 0
dv 0 .t 0 /
a0 .t 0 / D (F.169)
dt 0
v0
s D ˇx x0 ˇ .x x0 /
ˇ ˇ
(F.170)
c
0
ˇ 0
ˇ v0
x x0 D .x x / ˇx x ˇ (F.171)
0 c
@t jx x0 j
D (F.172)
@t x s
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 251
. of 300
F .3 Special relativity
0 0 0 1
x 0 D ƒ x
0
B
ˇ
B
ƒ D B
@ 0
Dp
ˇD
1
0
0
1 ˇ2
v
ˇ
FT 0
0
1
0
0
1
0C
0A
1
C
C
(F.174)
(F.175)
(F.176)
(F.177)
A
c
def
x D .x 0 ; x 2 ; x 2 ; x 3 / D .ct; x; y; z/ .ct; x/ (F.178)
D
C OVARIANT REPRESENTATION
x D .x0 ; x2 ; x2 ; x3 / D .x 0 ; x 2 ; x 2 ; x 3 /
def
(F.179)
D .ct; x; y; z/ .ct; x/
a .x / D a .x / (F.180)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 252
. of 300
230 j 6. FORMULÆ
1 @2
2 1 @ 1 @
D @ @ D @ @ D ; r ;r D 2 2 r2 (F.182)
FT
c @t c @t c @t
F .3.3.5 Four-velocity
dx
u D D
.c; v / (F.184)
A
d
F .3.3.6 Four-momentum
E
p D m0 u D ;p (F.185)
c
R
F .3.3.8 Four-potential
ˆ
A D ;A (F.187)
c
F D @ A @ A (F.188)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 253
. of 300
0 1
0 Ex =c Ey =c Ez =c
BE =c 0 Bz By C
B x
F D B
(F.189)
C
@Ey =c Bz 0 Bx A
C
Ez =c By Bx 0
F .4 Bibliography
[66] G. B. A RFKEN AND H. J. W EBER, Mathematical Methods for Physicists, fourth,
FT
international ed., Academic Press, Inc., San Diego, CA . . . , 1995, ISBN 0-12-059816-
7.
FT
A
R
D
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 255
. of 300
MATHEMATICAL METHODS
Physics is the academic discipline that systematically studies and describes the
physical world, postulates new fundamental laws of Nature or generalises exist-
ing ones that govern Nature’s behaviour under various conditions, and — per-
haps most important of all — makes new predictions about Nature, based on
these new postulates. Then these predictions are put to systematic tests in inde-
pendent, carefully designed and performed, repeatable experiments that produce
FT
objective emprical data. Merely describing Nature and explaining physical ex-
periments in terms of already existing laws is not physics in the true sense of the
word.1 Had this non-creative, static view been adopted by all physicists since 1 As S T E V E N W E I N B E R G puts it
the days of Newton, we would still be doing essentially Newtonian physics. in the Preface To Volume I of The
Quantum Theory of Fields:
Even if such a scientific giant as M I C H A E L F A R A DAY , who had very little
‘. . . after all, our purpose
mathematical training, was able to make truly remarkable contributions to phys- in theoretical physics is
ics (and chemistry) using practically no formal mathematics whatsoever, it is for not just to describe the
world as we find it, but to
us mere mortals most convenient to use the shorthand language of mathematics, explain — in terms of a few
together with the formal methods of logic (inter alia propositional calculus), fundamental principles —
A
why the world is the way it
in physics. After all, mathematics was once introduced by us human beings is.’
to make it easier to quantitatively and systematically describe, understand and
predict the physical world around us. Examples of this from ancient times are
arithmetics and geometry. A less archaic example is differential calculus, needed
by S I R I S A AC N E W T O N to formulate, in a compact and unambiguous manner,
R
the physical laws that bear his name. Another more modern example is the delta
‘function’ introduced by PAU L A D R I E N M AU R I C E D I R AC . But the oppos-
ite is also very common: the expansion and generalisation of mathematics has
more than once provided excellent tools for creating new physical ideas and to
better analyse observational data. Examples of the latter include non-Euclidean
D
233
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 256
. of 300
FT
which suggested the
new hypothesis. And
thirdly it must predict
new phenomena or new
relationships between
different phenomena, which
were not known or not
clearly understood at the
M .1 Scalars, vectors and tensors
time when it was invented.’
Every physical observable can be represented by a mathematical object. We
have chosen to describe the observables in classical electrodynamics in terms of
scalars, pseudoscalars, vectors, pseudovectors, tensors and pseudotensors, all of
A
which obey certain canonical rules of transformation under a change of coordin-
ate systems and are completely defined by these rules. Despite certain advant-
ages (and some shortcomings), differential forms will not be exploited to any
significant degree in our mathematical description of physical observables.
A scalar, which may or may not be constant in time and/or space, describes
R
the scaling of a physical quantity. A vector describes some kind of physical
3 The Latin word ‘vector’ means motion along a curve in space due to vection.3 A tensor describes the local
‘carrier’. motion or deformation of a surface or a volume due to some form of tension
and is therefore a relation between a set of vectors. However, generalisations to
more abstract notions of these quantities have proved useful and are therefore
D
commonplace. The difference between a scalar, vector and tensor and a pseudo-
scalar, pseudovector and a pseudotensor is that the latter behave differently un-
der those coordinate transformations that cannot be reduced to pure rotations.
For computational convenience, it is often useful to allow mathematical ob-
jects representing physical observables to be complex valued, i.e. to let them be
analytically continued into (a domain of) the complex plane. However, since
by definition our physical world is real, care must be exercised when compar-
ing mathematical results with physical observables, i.e. real-valued numbers ob-
tained from physical measurements.
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 257
. of 300
M .1.1 Vectors
Mathematically, a vector can be represented in a number of different ways. One
suitable representation in a vector space of dimension N is in terms of an ordered
N -tuple of real or complex4 numbers .a1 ; a2 ; : : : ; aN / of the components along 4 It is often very convenient to use
N orthogonal coordinate axes that span the vector space under consideration. complex notation in physics. This
notation can simplify the math-
Note, however, that there are many ordered N -tuples of numbers that do not ematical treatment considerably.
FT
comprise a vector, i.e. do not have the necessary vector transformation proper- But since all physical observables
are real, we must in the final step
ties! of our mathematical analysis of a
physical problem always ensure
that the results to be compared
M .1.1.1 Position vector with experimental values are real-
valued. In classical physics this
The most basic vector, and the prototype against which all other vectors are is achieved by taking the real (or
imaginary) part of the mathem-
benchmarked, is the position vector (radius vector, coordinate vector) which is atical result, whereas in quantum
the vector from the origin of the chosen coordinate system to the actual point physics one takes the absolute
value.
of interest. Its N -tuple representation simply enumerates the coordinates of the
position of this point. In this sense, the vector from the origin to a point is
A
synonymous with the coordinates of the point itself.
In the 3D Euclidean space R3 , we have N D 3 and the position vector x can
be represented by the triplet .x1 ; x2 ; x3 / of its coordinates xi 2 R, i D 1; 2; 3.
The coordinates xi are scalar quantities which describe the position along the
unit base vectors x O i which span R3 . Therefore one convenient representation of
R
the position vector in R3 is5 5 We introduce the symbol
def
In Cartesian coordinates
.x1 ; x2 ; x3 / D .x; y; z/ (M.3)
This component notation is particularly useful in a 4-dimensional Rieman-
nian space where we can represent the (one and the same) position vector either
in its contravariant component form, (superscript index form) as the quartet
def
x .x 0 ; x 1 ; x 2 ; x 3 / (M.4)
or its covariant component form (subscript index form)
def
x .x0 ; x1 ; x2 ; x3 / (M.5)
The contravariant and covariant forms represent the same vector but the numer-
ical values of the components of the two may have different numerical values.
FT
The relation between them is determined by the metric tensor (also known as
the fundamental tensor) whose actual form is dictated by the properties of the
vector space in question. The dual representation of vectors in contravariant and
covariant forms is most convenient when we work in a vector space with an in-
definite metric. An example of 4D Riemannian space is Lorentz space L4 which
is a frequently employed to formulate the special theory of relativity.
M .1.2 Fields
A
A field is a physical entity that depends on one or more continuous parameters.
Such a parameter can be viewed as a ‘continuous index’ that enumerates the
infinitely many ‘coordinates’ of the field. In particular, in a field that depends on
the usual position vector x of R3 , each point in this space can be considered as
one degree of freedom so that a field is a representation of a physical entity with
R
an infinite number of degrees of freedom.
FT
Re fcg D cR D cR cO R 2 R3
Im fcg D cI D cI cO I 2 R 3
a .x / D .a0 .x /; a1 .x /; a2 .x /; a3 .x // (M.12)
@x 0
dx 0 D dx (M.14)
@x
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 260
. of 300
FT
† with coordinates x to a system †0 with coordinates x 0 , it transforms to the
new system according to the rule
@x 0
y 0 D y (M.16)
@x
i.e. in the same way as the differential coordinate element dx transforms ac-
cording to equation (M.14) on the previous page.
The analogous requirement for a covariant four-vector is that it transforms,
during the change from † to †0 , according to the rule
@x
A
y0 D y (M.17)
@x 0
i.e. in the same way as the differential operator @=@x transforms according to
equation (M.15) above.
R
M .1.2.4 Tensor fields
We denote an arbitrary tensor field in R3 by A.x/. This tensor field can be
represented in a number of ways, for instance in the following matrix represent-
6 When a mathematical object ation:6
representing a physical observable 0 1
A .x/ A12 .x/ A13 .x/
D
Strictly speaking, the tensor field described here is a tensor of rank two.
A particularly simple rank two tensor in R3 is the 3D Kronecker delta tensor
ıij , with the following properties:
(
0 if i ¤ j
ıij D (M.19)
1 if i D j
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 261
. of 300
Another common and useful tensor is the fully antisymmetric tensor of rank
three, also known as the Levi-Civita tensor
8
<1 if i; j; k is an even permutation of 1,2,3
ˆ
ˆ
ij k D 0 if at least two of i; j; k are equal (M.21)
ˆ
ˆ
: 1 if i; j; k is an odd permutation of 1,2,3
and
ij k D
ij k i lm D ıj l ıkm
FT
ij k D j ki D kij
ikj
ıj m ıkl
(M.22)
(M.23)
(M.24)
A
In fact, tensors may have any rank n. In this picture, a scalar is considered
to be a tensor of rank n D 0 and a vector to be a tensor of rank n D 1. Con-
sequently, the notation where a vector (tensor) is represented in its component
form is called the tensor notation. A tensor of rank n D 2 may be represented by
a two-dimensional array or matrix, and a tensor of rank n D 3 may be represen-
R
ted as a vector of tensors of rank n D 2. Assuming that one of the indices of the
Levi-Civita tensor ij k , e.g. the first index i D 1; 2; 3, denotes the component of
such a vector of tensors, these components have the matrix representations (the
second and third indices, j; k D 1; 2; 3, are the matrix indices)
0 1 0 1
111 112 113 0 0 0
D
B
.ij k /i D1 D @121 122 123 A D @0 0 1A D iS1 (M.25a)
C B C
131 132 133 0 1 0
0 1 0 1
211 212 213 0 0 1
B
.ij k /i D2 D @221 222 223 A D @0 0 0 A D iS2 (M.25b)
C B C
231 232 233 1 0 0
0 1 0 1
311 312 313 0 1 0
B
.ij k /i D3 D @321 322 323 A D @ 1 0 0A D iS3 (M.25c)
C B C
331 332 333 0 0 0
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 262
. of 300
FT
coordinates. This is a very strict constraint.
In 4D, we have three forms of four-tensor fields of rank n. We speak of
a contravariant four-tensor field , denoted A1 2 :::n .x /,
a covariant four-tensor field , denoted A1 2 :::n .x /,
:::
a mixed four-tensor field , denoted A1kC12 :::kn .x /.
The 4D metric tensor (fundamental tensor) mentioned above is a particularly
important four-tensor of rank two. In covariant component form we shall denote
A
it g . This metric tensor determines the relation between an arbitrary contrav-
ariant four-vector a and its covariant counterpart a according to the following
rule:
def
a .x / g a .x / (M.29)
This rule is often called lowering of index. The raising of index analogue of the
R
index lowering rule is:
def
a .x / g a .x / (M.30)
More generally, the following lowering and raising rules hold for arbitrary
rank n mixed tensor fields:
D
gk k A1kC1
2 :::k 1 k 1 2 :::k 1
kC2 :::n .x / D Ak kC1 :::n .x / (M.31)
g k k A1k2kC1
:::k 1 1 2 :::k 1 k
:::n .x / D AkC1 kC2 :::n .x / (M.32)
Successive lowering and raising of more than one index is achieved by a repeated
application of this rule. For example, a dual application of the lowering operation
on a rank two tensor in its contravariant form yields
A D g g A (M.33)
i.e. the same rank two tensor in its covariant form.
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 263
. of 300
M .2 Vector algebra
O i bj x
a b D ai x Oj D x
Oi x
O j ai bj D ıij ai bj D ai bi (M.34)
FT
and similarly for b. This allows us to write
a b D ab cos (M.36)
a b D g a b D a b D g a b (M.37)
where we made use of the index lowering and raising rules (M.29) and (M.30).
A
The result is a four-scalar, i.e. an invariant which is independent of in which 4D
coordinate system it is measured.
The quadratic differential form
a b D ab sin eO (M.40)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 264
. of 300
where is the angle between a and b and eO is a unit vector perpendicular to the
plane spanned by a and b.
A spatial reversal of the coordinate system, .x10 ; x20 ; x30 / D . x1 ; x2 ; x3 /,
known as a parity transformation, changes sign of the components of the vectors
a and b so that in the new coordinate system a0 D a and b0 D b, which is to
say that the direction of an ordinary vector is not dependent on the choice of the
directions of the coordinate axes. On the other hand, as is seen from equation
(M.39) on the previous page, the cross product vector c does not change sign.
Therefore a (or b) is an example of a ‘true’ vector, or polar vector, whereas c is
an example of an pseudovector or axial vector.
A prototype for a pseudovector is the angular momentum vector L D x p
and hence the attribute ‘axial’. Pseudovectors transform as ordinary vectors un-
der translations and proper rotations, but reverse their sign relative to ordinary
FT
vectors for any coordinate change involving reflection. Tensors (of any rank)
that transform analogously to pseudovectors are called pseudotensors. Scalars
are tensors of rank zero, and zero-rank pseudotensors are therefore also called
pseudoscalars, an example being the pseudoscalar x O i .Oxj x O k/ D x
O i .ij k x
O i /.
This triple product is a representation of the ij k component of the rank three
Levi-Civita pseudotensor ij k .
A D a ˝ b D a1 x
O 1 ˝ b1 x O 1 ˝ b2 x
O 1 C a1 x O 3 ˝ b3 x
O 2 C C a3 x O3 (M.41a)
D
O i ˝ bj x
D ai x Oi ˝x
O j D ai bj x O i ai ˝ bj x
Oj D x Oj (M.41b)
0 10 1
a1 b1 a1 b2 a1 b3 O1
x
D x O1 x O2 x O 3 @a2 b1 a2 b2 a2 b3 A @x O 2A (M.41c)
B CB C
a3 b1 a3 b2 a3 b3 O3
x
In matrix representation
0 1 0 1
a1 a1 b1 a1 b2 a1 b3
.A/ D .a ˝ b/ D @a2 A b1 b2 b3 D @a2 b1 a2 b2 a2 b3 A (M.42)
B C B C
a3 a3 b1 a3 b2 a3 b3
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 265
. of 300
which we identify with expression (M.18) on page 238, viz. a tensor in matrix
representation. Hence, a dyadic of two vectors is intimately related to a rank two
tensor, emphasising its vectorial characteristics.
Scalar multiplication from the right or from the left of the dyad A D a ˝ b
by a vector c, produces other vectors according to the scheme
def
A c D a ˝ b c a.b c/ D aj bi ci x
Oj (M.43a)
def
c A D c a ˝ b .c a/b D ai bj ci x
Oj (M.43b)
respectively. These two vectors, proportional to a and b, respectively, are in
general not identical to each other. In the first case, c is known as the postfactor,
in the second case as the prefactor.
O j , then
Specifically, if c D x
FT
O j D abj D ai bj x
Ax Oi (M.44a)
O j A D aj b D aj bi x
x Oi (M.44b)
which means that
Oi Ax
x O j D ai bj D Aij (M.44c)
The vector product can be represented in matrix form as follows:
0 1 0 1
c1 a2 b3 a3 b2
.c/ D @c2 A D .a b/ D @a3 b1 a1 b3 A D ia S ˝ b D ia ˝ S b
B C B C
A
c3 a1 b2 a2 b1
(M.45)
where S ˝ b is the dyadic product of the matrix vector S, given by formula
(M.26) on page 240, and the vector b, and a ˝ S is the dyadic product of the
vector a and the matrix vector S.
R
Vector multiplication from the right and from the left of the dyad A by a
vector is another dyad according to the scheme
def
A c D a ˝ b c a ˝.b c/ D j kl ai bk cl x
O ix
Oj (M.46a)
def
c A D c a ˝ b .c a/ ˝ b D j kl al bi ck x
Oj x
O i D ikl al bj ck x
O ix
Oj
D
(M.46b)
respectively. In general, the two new dyads thus created are not identical to each
other.
Specifically, if A D 13 D x Oi ˝x
O i , i.e. the unit dyad or the second-rank unit
tensor, then
13 c D O1˝x
c3 x O1˝x
O 2 C c2 x O3
O2˝x
C c3 x O1 O2˝x
c1 x O3 (M.47)
O1˝x
c2 x O3˝x
O 3 C c1 x O 2 D c 13
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 266
. of 300
M .3 Vector calculus
where x
The del operator
FT
In R3 the del operator is a differential vector operator, denoted in Gibbs’ nota-
tion by the boldface nabla symbol r and defined as10
O i ri x
r Dx Oi
def @ def @ def
@xi
@x
@ (M.50)
O i is the ith unit vector in a Cartesian coordinate system. Since the oper-
A
ator in itself has vectorial properties, we denote it with a boldface nabla (r ).
In ‘component’ (tensor) notation the del operator can be written
@ @ @
@i D ; ; (M.51)
@x1 @x2 @x3
In 4D, the contravariant component representation of the four-del operator is
R
defined by
@ @ @ @
@ D ; ; ; (M.52)
@x0 @x1 @x2 @x3
whereas the covariant four-del operator is
@ @ @ @
D
@ D ; ; ; (M.53)
@x 0 @x 1 @x 2 @x 3
We can use this four-del operator to express the transformation properties
(M.16) and (M.17) on page 238 as
y 0 D @ x 0 y
(M.54)
and
y0 D @0 x y
(M.55)
respectively.
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 267
. of 300
@˛./
r ˛.x/ D r ˛./ D —O O
D —r˛./ (M.57)
@
and, therefore, r D —r. O From this we see that the boldface notation for the
gradient (r ) is very handy as it elucidates its 3D vectorial property and separates
it from the nabla operator (r) which has a scalar property.
FT
In 4D, the four-gradient of a four-scalar is a covariant vector, formed as a
derivative of a four-scalar field ˛.x /, with the following component form:
@˛.x /
@ ˛.x / D (M.58a)
@x
with the contravariant form
@˛.x /
@ ˛.x / D (M.58b)
@x
A
M .3.3 The divergence of a vector field
We define the 3D divergence of a vector field a in R3 as
@a .x /
@ a .x / D (M.60)
@x
D
where use was made of the Levi-Civita tensor, introduced in equation (M.21) on
page 239. If a is an ordinary vector (polar vector), then r a is a pseudovector
(axial vector) and vice versa.
Similarly to formula (M.45) on page 243, we can write the matrix represent-
ation of the curl in R3 as
0 1
@2 a3 @3 a2
.r a/ D @@3 a1 @1 a3 A D ir S ˝ a D ir ˝ S a (M.62)
B C
@1 a2 @2 a1
FT
A .x / D @ a .x / @ a .x / D
3
def def X @2 def @2 def 2
r r D r2 @i (M.64)
iD1
@xi2 @xi2
R
The symbol r 2 is sometimes read del squared . If, for a scalar field ˛.x/,
r 2 ˛ < 0 at some point in 3D space, ˛ has a concentration at that point.
Numerous vector algebra and vector calculus formulæ are given in appendix F
on page 215. Those which are not found there can often be easily derived by us-
ing the component forms of the vectors and tensors, together with the Kronecker
D
and Levi-Civita tensors and their generalisations to higher ranks and higher di-
mensions.
FT
V0 V0
We can now integrate the RHS by part by invoking formula (F.77) once more,
but this time with r 0 instead of r . The result is
0 0 0
˛.x0 /
3 0 ˛.x / 3 0 r ˛.x /
Z Z Z
3 0 0
r dx D dx dx r (M.67)
V0 jx x0 j V0 jx x0 j V0 jx x0 j
Formula (F.121a) on page 223 enables us to replace the last volume integral with
a surface integral, yielding the final result
˛.x0 / 0 0
3 0 r ˛.x / ˛.x0 /
Z Z I
d3x 0 d2x 0 nO
A
r 0
D d x 0
(M.68)
V0 jx x j V0 jx x j S0 jx x0 j
An analogous approach for the divergence of a volume integral of a regular
vector field a.x0 / divided by jx x0 j yields, with the use of formula (F.81) on
page 221 and the results in example M.10 on page 260,
R
a.x0 /
1
Z Z
r d3x 0 D d 3 0
x a.x 0
/ r
V0 jx x0 j V0 jx x0 j
(M.69)
1
Z
D d3x 0 a.x0 / r 0
V0 jx x0 j
We integrate this by part, again employing identity (F.81) on page 221 and for-
D
The above results can be summarised in the general partial integration for-
mula
0 0 0
3 0 A.x / 3 0 r ı A.x / A.x0 /
Z Z I
2 0 0
rı dx D d x d x O
n ı (M.72)
V0 jx x0 j V0 jx x0 j S0 jx x0 j
where ı is either (i) nothing (juxtaposition) and A D ˛, or (ii) ı D or ı D
and A D a. In the surface integrals in the formulæ above, the surface ele-
ment d2x 0 D dS0 nO 0 is proportional to r 2 D jx x0 j2 [cf. formula (F.18) on
page 217]. Hence, if A.x0 / falls off faster than 1=r, this surface integral vanishes
when we let the radius r of the sphere S 0 , over which the surface integral is to be
evaluated, tend to infinity. If A.x0 / falls off exactly as 1=r for large r, the surface
integral tends to a constant, which, in the special cases ı D ; A.x0 / D a ? nO 0
and ı D ; A.x0 / D a k nO 0 is zero. If A.x0 / falls off slower than 1=r at in-
FT
finity, the surface integral is singular and, consequently, the above formulæ are
inapplicable. When the surface integral vanishes the following simple and very
useful formula obtains:
A.x0 / 0 0
3 0 r ı A.x /
Z Z
rı d3x 0 D d x (M.73)
V0 jx x0 j V0 jx x0 j
˛.x0 /
Z
r r d3x 0 D 4 ˛.x/ (M.75)
V0 jx x0 j
The curl of the curl of an integral with an integrand of the second kind is
0 0
3 0 a.x / 3 0 a.x /
Z Z
r r dx D r ˝ r d x
V0 jx x0 j V0 jx x0 j
0 (M.76)
3 0 a.x /
Z
r r dx
V0 jx x0 j
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 271
. of 300
where we used the identity (F.96) on page 222. The second integral in the RHS
is the vector form of formula (M.75) on the facing page :
a.x0 /
Z
r r d3x 0 D 4a.x/ (M.77)
V0 jx x0 j
Using the fact that r operates on unprimed coordinates whereas a depends only
on x0 , we can rewrite the first integral in the RHS as
Z
1 1
Z
d3x 0 a.x0 / r ˝ r D d 3 0
x a.x0
/ r 0
˝ r 0
V0 jx x0 j V0 jx x0 j
(M.78)
FT
r 0 .1= jx x0 j/ to integrate the RHS by parts as follows:
1
Z
d3x 0 a.x0 / r 0 ˝ r 0
V0 jx x0 j
Z
1 1
Z
D d3x 0 Œr 0 a.x0 /r 0 C d3 0
x r 0
a.x 0
/ ˝ r 0
V0 jx x0 j 0 jx x0 j
IV
1 1
Z
3 0 0 0 0 2 0 0 0 0
D d x Œr a.x /r O
C d x n a.x / r ˝
V0 jx x0 j S0 jx x0 j
(M.79)
A
where, in the last step, we used the divergence theorem for tensors/dyadics, for-
mula (F.121e) on page 223. Putting it all together, we finally obtain the integral
identity
a.x0 /
1
Z Z
r r d3x 0 D 4a.x/ d 3 0
x Œr 0
a.x0
/r 0
R
V0 jx x0 j V0 jx x0 j
0 ˝.x 0
a.x / x/
I
C d2x 0 nO 0 3
(M.80)
S 0 jx x j 0
If u.x/ is regular and falls off rapidly enough with distance r D jx x0 j (typic-
ally faster than 1=r when r ! 1), we can use formula (M.73) on page 248 to
rewrite this as
where
r 0 u.x0 /
Z
˛.x/ D d3x 0 (M.82b)
V0 4 jx x0 j
r 0 u.x0 /
Z
a.x/ D d3x 0 (M.82c)
V0 4 jx x0 j
FT
rotational (also called rotation-less or lamellar), and, according to example M.14
on page 262, r .r a/ D 0, i.e. r a is rotational (also called divergence-
less or solenoidal ), u can always be decomposed into one irrotational and one
rotational component
where
r 0 u.x0 /
Z
uirrot .x/ D r ˛.x/ D r d3x 0 (M.83b)
V0 4 jx x0 j
A
r 0 u.x0 /
Z
urotat .x/ D r a.x/ D r d3x 0 (M.83c)
V0 4 jx x0 j
r u D r uirrot D r . r ˛/ D r 2˛ (M.85a)
rotat
r uDr u D r .r a/ (M.85b)
from which we see that a vector field that is well-behaved at large distances is
completely and uniquely determined if we know its divergence and curl at all
points x in 3D space (and at any given fixed time t if the vector field is time
dependent). This is the (first) Helmholtz’s theorem, also called the fundamental
theorem of vector calculus.
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 273
. of 300
M .4 Analytical mechanics
FT
mechanical system, If we use the action
Z t2
SD dt L.qi ; qP i ; t/ (M.88)
t1
ıS D 0 (M.89)
@L0 @L d @˛
D C (M.94b)
@qi @qi dt @q
Or, in other words,
d @L0 @L0 d @L @L
D (M.95)
dt @qP i @qi dt @qP i @qi
where
@L0 @L @˛ @˛
pi0 D D C D pi C (M.96a)
@qP i @qP i @qi @qi
and
@L0 @L
qi0 D D D qi (M.96b)
@pPi @pP
FT
M .4.2 Hamilton’s equations
From L, the Hamiltonian (Hamilton function) H can be defined via the Le-
gendre transformation
H.pi ; qi ; t/ D pi qP i L.qi ; qP i ; t/ (M.97)
After differentiating the left and right hand sides of this definition and setting
them equal we obtain
@H @H @H @L @L @L
dpi C dqi C dt D qP i dpi C pi dqP i dqi dqP i dt
@pi @qi @t @qi @qP i @t
A
(M.98)
According to the definition of pi , equation (M.91) on the previous page, the
second and fourth terms on the right hand side cancel. Furthermore, noting that
according to equation (M.92) on the preceding page the third term on the right
hand side of equation (M.98) above is equal to pPi dqi and identifying terms, we
R
obtain the Hamilton equations:
@H dqi
D qP i D (M.99a)
@pi dt
@H dpi
D pPi D (M.99b)
@qi dt
D
M .5 Examples
p
c D jcjei D 2C.cos C i sin / ; C 2 R (M.100)
and, without lack of generality, choose as base vectors the two complex unit base vectors,
expressed in the two orthogonal base vectors x1 ; x2 2 R3
1
hO ˙ D p .Ox1 ˙ iOx2 / (M.101)
2
that fulfil the conditions
O C hO D hO hO D 1
h (M.102)
C
O C hO D h
h O hO D 0 (M.103)
C
FT
In order to interpret this expression correctly in physical terms, we must take the real part
O 1 sin x
Re fc˙ g D C.cos x O 2/ (M.105)
If D !t, as is be the case when measures the angle of rotation with angular frequency
!, then
Re fc˙ g D C cos.!t/ xO 1 sin.!t/ xO2 (M.106)
We see that the physically meaningful real part describes a rotation in positive or negative
O ˙ are called helical base vectors.
sense, depending on the choice of sign. and therefore h
T O
n D TnO (M.107)
Using (M.107) and Newton’s second law, we find that the matter of mass m, which at a
given instant is located in V obeys the equation of motion
TnO d2x cos 1 TxO 1 d2x cos 2 TxO 2 d2x cos 3 TxO 3 d2x C Fext D ma (M.108)
where Fext is the external force and a is the acceleration of the volume element. In other
words
m Fext
TnO D n1 TxO 1 C n2 TxO 2 C n3 TxO 3 C 2 a (M.109)
dx m
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 276
. of 300
nO
d2x
x2
x1 FT V
A
Since both a and Fext =m remain finite whereas m=d2x ! 0 as V ! 0, one finds that in
this limit
3
X
TnO D ni TxO i ni TxO i (M.110)
i D1
From the above derivation it is clear that equation (M.110) above is valid not only in equi-
R
librium but also when the matter in V is in motion.
Introducing the notation
Tij D TxO i (M.111)
j
for the j th component of the vector TxO i , we can write equation (M.110) above in compon-
D
Using equation (M.112) above, we find that the component of the vector TnO in the direction
O is
of an arbitrary unit vector m
O D Tn
TnO m O
O m
3 3 3
!
X X X (M.113)
D O mj D
Tnj ni Tij mj ni Tij mj D nO T m
O
j D1 j D1 iD1
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 277
. of 300
Hence, the j th component of the vector TxO i , here denoted Tij , can be interpreted as the
ij th component of a tensor T. Note that TnO m
O is independent of the particular coordinate
system used in the derivation.
We shall now show how one can use the momentum law (force equation) to derive the
equation of motion for an arbitrary element of mass in the body. To this end we consider a
part V of the body. If the external force density (force per unit volume) is denoted by f and
the velocity for a mass element dm is denoted by v , we obtain
Z Z Z
d
v dm D f d3x C TnO d2x (M.114)
dt V V S
The j th component of this equation can be written
Z Z Z Z Z
d
vj dm D fj d3x C Tnj
O d2
x D fj d3
x C ni Tij d2x (M.115)
V dt V S V S
where, in the last step, equation (M.112) on the facing page was used. Setting dm D d3x
FT
and using the divergence theorem on the last term, we can rewrite the result as
@Tij 3
Z Z Z
d
vj d3x D fj d3x C dx (M.116)
V dt V V @xi
Since this formula is valid for any arbitrary volume, we must require that
d @Tij
vj fj D0 (M.117)
dt @xi
or, equivalently
@vj @Tij
C v r vj fj D0 (M.118)
@t @xi
A
Note that @vj =@t is the rate of change with time of the velocity component vj at a fixed
point x D .x1 ; x1 ; x3 /.
The 4D Lorentz space L4 has a simple metric which can be described by the metric tensor
R
8
<1
ˆ if D D 0
g D D 1 if D D i D j D 1; 2; 3 (M.119)
ˆ
0 if ¤
:
0 1
1 0 0 0
B0 1 0 0C
. / D B
@0
C (M.120)
0 1 0A
0 0 0 1
i.e. a matrix with a main diagonal that has the sign sequence, or signature, fC; ; ; g.
Alternatively, one can define the metric tensor in L4 as
8
< 1 if D D 0
ˆ
D 1 if D D i D j D 1; 2; 3 (M.121)
ˆ
0 if ¤
:
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 278
. of 300
0 1
1 0 0 0
B 0 1 0 0C
. / D @
B C (M.122)
0 0 1 0A
0 0 0 1
i.e. a matrix with signature f ; C; C; Cg. Of course, the physics is unaffected by the choice
of metric tensor.
def
a .a0 ; a1 ; a2 ; a3 / D .a0 ; a/ (M.123)
FT
According to the index lowering rule, equation (M.29) on page 240, we obtain the covariant
version of this vector as
D0W
def
a .a0 ; a1 ; a2 ; a3 / D a
a0 D 1 a 0 C 0 a 1 C 0 a 2 C 0 a 3 D a 0
(M.124)
(M.125)
A
0 1 2 3 1
D1W a1 D 0 a 1a C 0a C 0a D a (M.126)
0 1 2 3 2
D2W a2 D 0 a C 0 a 1a C 0a D a (M.127)
0 1 2 3 3
D3W a3 D 0 a C 0 a C 0 a 1a D a (M.128)
or
R
a D .a0 ; a1 ; a2 ; a3 / D .a0 ; a1 ; a2 ; a3 / D .a0 ; a/ (M.129)
x D .x 0 ; x 1 ; x 2 ; x 3 / D .ct; x; y; z/ D .ct; x/
(M.130)
x D .x0 ; x1 ; x2 ; x3 / D .ct; x 1 ; x 2 ; x 3 / D .ct; x/
a D .a0 ; a1 ; a2 ; a3 / D . a0 ; a1 ; a2 ; a3 / D . a0 ; a/ (M.131)
def
c2 c c D .cR C icI / .cR C icI / D c2R c2I C 2icR cI
(M.132)
def
D cR2 cI2 C 2icR cI c 2 2 C
q q
cD cR2 cI2 C 2icR cI D cR2 cI2 C 2icR cI cos 2 C (M.133)
FT
where is the (real-valued) angle between cR and cI .
Using this in equation (M.10) on page 237, we see that we can define the complex unit
vector as being
c cR cI
cO D D q cO R C i q cO I
c
cR2 cI2 C 2icR cI cos cR2 cI2 C 2icR cI cos
q q
cR cR2 cI2 2icR cI cos cI cR2 cI2 2icR cI cos
D cO R C i cO I 2 C3
A
s s
4cR2 cI2 sin2 4cR2 cI2 sin2
.cR2 C cI2 / 1 .cR2 C cI2 / 1
.cR2 C cI2 /2 .cR2 C cI2 /2
(M.134)
On the other hand, the definition of the scalar product in terms of the inner product of a
R
complex vector with its own complex conjugate yields
def
jcj2 c c D .cR C icI / .cR C icI / D c2R C c2I D cR2 C cI2 D jcj2 (M.135)
c cR cI
cO D D q cO R C i q cO I
jcj
cR2 C cI2 cR2 C cI2
q q (M.136)
cR cR2 C cI2 cI cR2 C cI2
D cO R C i cO I 2 C3
cR2 C cI2 cR2 C cI2
FT
End of example M.5C
EXAMPLE M .6 BThe vector triple product
The vector triple product is the vector product of a vector a with a vector product b c and
can, with the help of formula (M.24) on page 239, be evaluated as
O i aj .b c/k D ij k x
a .b c/ D ij k x O i aj .lmn x
O l bm cn /k
O i aj kmn bm cn D ij k kmn x
D ij k x O i aj bm cn
O i aj bm cn D .ıi m ıj n
D kij kmn x ıi n ıj m /Oxi aj bm cn (M.140)
O i aj bm cn
D ıi m ıj n x ıi n ıj m x Oi
O i aj bm cn D aj cj bi x Oi
aj bj ci x
D .a c/b .a b/c D b.a c/ c.a b/ b ˝ a c c˝ab
A
which is formula (F.53) on page 220. This is sometimes called Lagrange’s formula, but is
more often referred to as the bac-cab rule.
End of example M.6C
EXAMPLE M .7 BMatrix representation of the vector product in R3
Prove that the matrix representation of the vector product c D a b is given by formula
R
(M.45) on page 243.
According to formula (M.43) on page 243, the scalar multiplication of a dyadic product of
two vectors (in our case S and b) from the left by a vector (in our case a) is interpreted as
a S ˝ b D .a S/b where
a S D ai Si (M.141)
and the components Si are given by formula (M.27) on page 240. Hence
D
0 1 0 1 0 1
0 0 0 0 0 i 0 i 0
a S D a1 @0 0 iA C a2 @ 0 0 0A C a3 @ i 0 0A
0 i 0 i 0 0 0 0 0
20 1 0 1 0 13
0 0 0 0 0 a2 0 a3 0
D i 4@0 0 a1 A C @ 0 0 0 A C @a3 0 0A5 (M.142)
0 a1 0 a2 0 0 0 0 0
0 1
0 a3 a2
D i @ a3 0 a1 A
a2 a1 0
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 281
. of 300
In other words,
0 1
a2 b3 a3 b2
i.a S/b D ia S ˝ b D @a3 b1 a1 b3 A D .a b/ (M.144)
a1 b 2 a2 b1
FT
S b D Si bi D @0 0 iA b1 C @ 0 0 0A b2 C @ i 0 0 A b3
0 i 0 i 0 0 0 0 0
0 1 (M.145)
0 b3 b2
D i @ b3 0 b1 A
b2 b1 0
and
0 1 0 1
0 b3 b2 a2 b 3 a3 b2
a.S b/ D i a1 a2 a3 @ b3 0 b1 A D i @a3 b1 a1 b3 A (M.146)
b2 b1 0 a1 b 2 a2 b1
A
from which follows
0 1
a2 b3 a3 b2
ia.S b/ D ia ˝ S b D @a3 b1 a1 b3 A D .a b/ (M.147)
a1 b2 a2 b1
R
Hence, .a b/ D ia S ˝ b D ia ˝ S b QED
The gradient of the scalar product of two vector fields a and b can be calculated in the
following way:
O j bk x
r .a b/ D .Oxi @i /.aj x O k/
O j / .bk x
D Œ.Oxi @i /.aj x O k / C .aj x
O j / Œ.Oxi @i /.bk x
O k / (M.148)
O k / a D .r ˝ a/ b C .r ˝ b/ a
O j / b C .Oxi @i bk x
D .Oxi @i aj x
FT
End of example M.9C
EXAMPLE M .11 BDivergence and curl of a vector field divided by the relative distance
D
This demonstrates how the primed divergence and curl, defined in terms of the primed del
operator in equation (M.152) above, work.
End of example M.11C
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 283
. of 300
where ı.x x0 / is the 3D Dirac delta ‘function’. This formula follows directly from the
fact that
x x0 x x0
Z I
1
Z
d3x 0 r r D d3 0
x r D d2 0 0
x nO
V0 jx x0 j V0 jx x0 j3 S0 jx x0 j3
(M.157)
FT
End of example M.12C
Using the definition of the R3 curl, equation (M.61) on page 245, and the gradient, equation
(M.56) on page 245, we see that
O i @j Œr ˛.x/k D ij k x
r Œr ˛.x/ D ij k x O i @j @k ˛.x/ (M.158)
r Œr ˛.x/ 0 (M.160)
Using the definition for the Levi-Civita symbol, defined by equation (M.21) on page 239,
we find that, due to the assumed well-behavedness of a.x/,
@ @
@i ij k @j ak .x/ D a
@xi ij k @xj k
!
@2 @2
D a1 .x/
@x2 @x3 @x3 @x2
!
@2 @2 (M.163)
C a2 .x/
FT
@x3 @x1 @x1 @x3
!
@2 @2
C a3 .x/
@x1 @x2 @x2 @x1
0
i.e. that
for any arbitrary, well-behaved R3 vector field a.x/. The 3D curl is therefore solenoidal
(has a vanishing divergence).
A
In 4D, the four-divergence of the four-curl is not zero, for
@ A D @ @ a .x / 2 a .x / ¤ 0 (M.165)
O i @j .r a/k D ij k x
r .r a/ D ij k x O i @j .lmn x
O l @m an /k
O i @j kmn @m an D ij k kmn x
D ij k x O i @j @m an
O i @j @m an D .ıi m ıj n
D kij kmn x ıi n ıj m /Oxi @j @m an
(M.166)
O i @j @m an
D ıi m ıj n x O i @j @m an
ıi n ıj m x
O i @j @i aj
Dx O i @j @j ai D x
x O i @i @j aj O i @j2 ai
x
D r .r a/ r2a r ˝ r a r ra
Derive the scalar and vector products of the rotational and irrotational components of two
vector fields u.x/ and v.x) that have been Helmholtz decomposed into u D urotat C uirrot
and v D v rotat C v irrot , respectively.
Let us, in addition to expressions (M.82), introduce the definitions
r 0 v.x0 /
Z
ˇ.x/ D d3x 0 (M.167a)
V0 4 jx x0 j
r 0 v .x0 /
Z
b.x/ D d3x 0 (M.167b)
V0 4 jx x0 j
in order to simplify our notations and calculations.
S CALAR PRODUCTS
FT
The scalar product of the two irrotational components uirrot and v irrot is given by the formula
where we can rewrite the RHS by using formula (F.82) on page 221 so that
If we insert the expression (M.82b) on page 250 for ˛.x/ and expression (M.167a) for ˇ.x/
and then integrate over V , the first term in the RHS can, with help ˇ of theˇ divergence theorem,
be written as a surface integral where, at large distances r D ˇx x0 ˇ, the integrand tends
to zero as 1=r 3 , ensuring that this integral vanishes. Then, also using the identity (M.77)
on page 249 and equations (M.85) on page 250, we obtain the following non-local scalar
A
product expression:
Œr 0 uirrot .x0 /Œr v irrot .x/
Z Z Z
d3x uirrot .x/ v irrot .x/ D d3x d3x 0
V V V0 4 jx x0 j
Z Z 0 irrot .x0 /
(M.170)
3 irrot 3 0 r u
D d x Œr v .x/ dx
V V0 4 jx x0 j
R
This is identity (F.130c) on page 225.
We now evaluate the scalar product of the two rotational (solenoidal) components urotat and
v rotat
As the last term in the RHS will vanish when we integrate over V (divergence theorem), we
focus our attention on the first term. According to formula (F.96) on page 222
and, according to equation (M.82c) on page 250 and formula (F.81) on page 221, we obtain
r 0 v .x0 /
1
Z Z
r a D r d3x 0 D d3 0
x Œr 0
v.x0
/ r (M.174)
V0 4 jx x0 j V0 4 jx x0 j
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 286
. of 300
With the help of formula (F.114) on page 222, formula (M.155) on page 260, and formula
(F.99) on page 222, we can rewrite this as
0
r v.x0 /
0
r v .x0 /
Z I
r a D d3x 0 r D d2 0 0
x O
n (M.175)
V0 4 jx x0 j S0 4 jx x0 j
which tends to zero for a surface at large distances. Hence, we have found that
Integration over V and the use of the divergence theorem, where the resulting surface in-
tegral vanishes, and using equations (M.85) on page 250 gives the result
Œr urotat .x/ Œr 0 v rotat .x0 /
Z Z Z
d3x urotat .x/ v rotat .x/ D d3x d3x 0
V V V0 4 jx x0 j
(M.177)
r 0 v rotat .x0 /
Z Z
D d3x Œr urotat .x/ d3x 0
V V0 4 jx x0 j
FT
This is identity (F.130d) on page 225.
The scalar product of uirrot and v rotat becomes
With the use of a standard vector analytic identity (F.84) on page 221, we can rewrite this
as
V ECTOR PRODUCTS
According to formulæ (M.83) on page 250, the vector product of the two irrotational com-
ponents uirrot and v irrot is
which can be evaluated in a similar manner as the scalar product. Using formula (F.88) on
page 221 we immediately see that the result is
Integration over V yields, with the use of formula (F.121c) on page 223, the result
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 287
. of 300
Z I
d3x uirrot .x/ v irrot .x/ D d2x nO ˛.x/r ˇ.x/ (M.185)
V S
where the RHS goes to zero asymptotically, resulting in the non-local parallelity condition
Z
d3x uirrot .x/ v irrot .x/ D 0 (M.186)
V
[see formulæ (M.83)], we use the vector identity (F.79) on page 221 that gives (with c
replacing a)
FT
Next, we use the identity (F.86) on page 221 to replace c r ˝ b and find that we can write
Finally we make the change c 7! r a with the result that we obtain the identity
Recalling that a is given by formula (M.82c) on page 250, we see that the expression within
A
square brackets in the first term of the RHS of equation (M.190) is
r 0 u.x0 /
Z
r .r a/ D r r d3x 0
V0 4 jx x0 j
1
Z
D r u.x/ d3x 0 fr 0 Œr 0 u.x0 /gr 0 (M.191)
V0 „ ƒ‚ … 4 jx x0 j
R
0
D r u.x/
where we used formula (M.80) on page 249, not including the vanishing surface integral
(last term in the RHS of that formula). Consequently,
D b .r u/ b r ˝.r a/ (M.192)
C r Œ.r a/ b C r Œ.r a/ ˝ b
Integrating the RHS of this expression over V , using the divergence theorem and formula
(F.121a) on page 223, and assuming that the ensuing surface integrals vanish, we therefore
obtain
Z Z Z
d3x urotat .x/ v rotat .x/ D d3x b .r u/ d3x b r ˝.r a/ (M.193)
V V V
Integration by parts of the second term in the RHS of this expression over V , making use
of the identity (F.86) on page 221 with a D b and b D r a, gives the result
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 288
. of 300
Z Z Z
d3x b r .r a/ D d3x .r b/.r a/ d3x r Œb ˝.r a/
V V V
Z I (M.194)
D d3x .r b/.r a/ d2x nO b ˝.r a/
V S
where we used formula (F.121e) on page 223. We have thereby shown that
Z Z Z
d3x urotat .x/ v rotat .x/ D d3x b .r u/ C d3x .r b/.r a/
V
IV V
(M.195)
2
d x nO b ˝.r a/
S
When we insert b as given by expression (M.167b) on page 263, the surface integral van-
ishes and we get
r 0 v .x0 /
Z Z Z
d3x urotat .x/ v rotat .x/ D d3x d3x 0 .r u/
V V V0 4 jx x0 j
(M.196)
r 0 v.x0 /
Z Z
FT
C d3x r d3x 0 .r a/
V V0 4 jx x0 j
However, for the case at hand with sufficiently rapid fall-off of the integrands in the surface
integrals, we can, according to formula (M.73) on page 248 with A D r 0 v .x0 /, rewrite
the expression within the large parentheses in the last term as
0
‚ …„ ƒ
0 0 0
r0 v .x0 / 3 0 r Œr v .x /
Z Z
r d3x 0 D dx D0 (M.197)
V0 jx x0 j V0 jx x0 j
This means that the second term itself vanishes and we are left with the final result
Œr u.x/ Œr 0 v.x0 /
Z Z Z
d3x urotat .x/ v rotat .x/ D d3x d3x 0
A
V V V0 4 jx x0 j
(M.198)
r 0 v .x0 /
Z Z
D d3x Œr u.x/ d3x 0
V V0 4 jx x0 j
This is identity (F.130e) on page 225.
The vector product of an irrotational and a rotational component is
R
uirrot .x/ v rotat .x/ D Œr ˛.x/ Œr b.x/ (M.199)
Using identity (F.79), identity (F.89) with a D r ˛, and identity (F.100) on page 222,
together with identity (F.74) on page 221, this can be written
uirrot .x/ v rotat .x/ D b r ˝ r ˛ br 2 ˛ r Œ.r ˛/ b C r Œ.r ˛/ ˝ b (M.200)
When we insert the expression (M.83b) on page 250 and integrate over V , employing iden-
D
tity (F.121a) and identity (F.121b) on page 223 and neglecting the resulting surface integrals,
we get
r 0 u.x0 /
Z
uirrot .x/ v rotat .x/ D b r ˝ uirrot .x/ br 2 d3x 0 (M.201)
V0 4 jx x0 j
With the help of the generic formula (F.126) on page 224 we can simplify the last integral
and obtain
Z Z Z
d3x uirrot .x/ v rotat .x/ D d3x b.x/Œr uirrot .x/ d3x b.x/ r ˝ uirrot .x/
V V V
(M.202)
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 289
. of 300
After integrating the last integral on the RHS by parts, making use of formula (F.86) on
page 221 and discarding the surface integral which results after applying the divergence
theorem, we can write
Z Z Z
d3x uirrot .x/ v rotat .x/ D d3x b.x/Œr uirrot .x/ C d3x Œr b.x/uirrot .x/
V V V
(M.203)
To obtain the final expression, we use formula (M.167b) on page 263 and recall that
r b D 0. The result is
Œr uirrot .x/Œr 0 v rotat .x0 /
Z Z Z
d3x uirrot .x/ v rotat .x/ D d3x d3x 0
V V V0 4 jx x0 j
(M.204)
r 0 v rotat .x0 /
Z Z
D d3x Œr uirrot .x/ d3x 0
V V0 4 jx x0 j
This is identity (F.130f) on page 225.
FT
End of example M.16C
M .6 Bibliography
[69] M. A BRAMOWITZ AND I. A. S TEGUN, Handbook of Mathematical Functions, Dover
Publications, Inc., New York, 1972, Tenth Printing, with corrections.
A
[70] G. B. A RFKEN AND H. J. W EBER, Mathematical Methods for Physicists, fourth,
international ed., Academic Press, Inc., San Diego, CA . . . , 1995, ISBN 0-12-059816-
7.
[71] R. A. D EAN, Elements of Abstract Algebra, John Wiley & Sons, Inc.,
New York, NY . . . , 1967, ISBN 0-471-20452-8.
R
[72] A. A. E VETT, Permutation symbol approach to elementary vector analysis, American
Journal of Physics, 34 (1965), pp. 503–507.
[73] A. M ESSIAH, Quantum Mechanics, vol. II, North-Holland Publishing Co., Amster-
dam, 1970, Sixth printing.
[75] B. S PAIN, Tensor Calculus, third ed., Oliver and Boyd, Ltd., Edinburgh and London,
1965, ISBN 05-001331-9.
FT
A
R
D
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 291
. of 300
INDEX
L
Cerenkov, Pavel Alekseevich, 200 CGS units, 3, 15
Ørsted, Hans Christian, 6 characteristic impedance of vacuum, 105, 206
CPT theorem, 56 charge conjugation, 53
charge density flux, 59
acceleration field, 128 charge space, 72
adiabatic, 4 Cherenkov radiation, 198
advanced time, 40 chiral media, 191
Ampère’s law, 6 chirality density, 83
Ampère, André-Marie, 6 chirality flow, 83
Ampère-turn density, 194 circular wave polarisation, 31
angular frequency, 25, 38 Classical Electrodynamics, 1, 9
angular momentum, 68 closed algebraic structure, 162
angular momentum commutation rule, 240 coherent radiation, 140
FT
angular momentum theorem, 68 collision frequency, 204
anisotropic, 196 collisional interaction, 210
anisotropic medium, 193 complete ˛-Lorenz gauge, 42
anomalous dispersion, 197 complex conjugate, 215
antecedent, 242 complex notation, 20, 235
antenna, 117 complex vector field, 237
antenna current, 118 complex-field six-vector, 22
antenna current density, 117
component notation, 235
antenna feed point, 119
concentration, 246
antisymmetric tensor, 170
conductivity, 202
arrow of time, 56
conductivity tensor, 11
A
associated Legendre polynomial of the first kind,
consequent, 242
111
conservation law, 56, 59
associative, 162
conservation law for angular momentum, 68
axial gauge, 45
conservation law for linear momentum, 63
axial vector, 170, 242
conservation law for the total current, 83
axiomatic foundation of classical electrodynam-
conservative field, 12
ics, 19
R
conservative forces, 182
azimuthal phase, 81
conserved quantities, 53, 56
bac-cab rule, 258 constants of motion, 53, 56
Barbieri, Cesare, xx constitutive relations, 15, 202
Bessel functions, 149 continuity equation, 57
binomial expansion, 94 continuous symmetries, 56
D
269
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 292
. of 300
270 INDEX
FT
Curie, Pierre, 15 electric displacement current, 17
curl, 245 electric displacement vector, 191, 192
cutoff, 153 electric field, 3, 192
cycle average, 22 electric field energy, 60
cyclotron radiation, 137, 141 electric monopole moment, 108
electric permittivity, 210
d’Alembert operator, 25, 36, 166, 260 electric polarisation, 109
de Coulomb, Charles-Augustin, 2 electric quadrupole moment tensor, 108
definiendum, 235 electric quadrupole radiation, 117
definiens, 235 electric quadrupole tensor, 117
del operator, 244 electric susceptibility, 193
A
del squared, 246 Electricity, 1, 2
demodulation, 193 electrodynamic potentials, 35
dielectric permittivity, 192 electromagnetic angular momentum current dens-
differential distance, 160 ity, 68, 79
differential vector operator, 244 electromagnetic angular momentum density, 67,
diffusion coefficient, 212 144, 146
R
dipole antennas, 118 electromagnetic angular momentum flux tensor, 67
Dirac delta, 261 electromagnetic energy current density, 59
Dirac’s symmetrised Maxwell equations, 16 electromagnetic energy flux, 59
Dirac, Paul Adrien Maurice, 1, 40, 233 electromagnetic field density vector, 100
direct product, 242 electromagnetic field energy, 60
dispersive, 197 electromagnetic field energy density, 30, 59
D
INDEX 271
FT
EMF, 11 four-vector, 161, 237
energy density balance equation, 60 four-velocity, 165
energy density velocity, 60 Fourier amplitude, 25
energy theorem in Maxwell’s theory, 61 Fourier integral, 26
equation of continuity, 167 Fourier transform, 26, 37
equation of continuity for electric charge, 10 Fröman, Per Olof, xxi
equation of continuity for magnetic charge, 16 Frank, Ilya Mikhailovich, 200
equations of classical electrostatics, 9 Franklin, Benjamin, 59
equations of classical magnetostatics, 9 free-free radiation, 135
Eriksson, Anders, xxii frequency conversion, 11
Eriksson, Marcus, xx frequency mixing, 11
A
Erukhimov, Lev Mikahilovich, xxii fully antisymmetric tensor, 239
Euclidean space, 163 functional derivative, 184
Euclidean vector space, 159 fundamental tensor, 157, 236, 240
Euler’s first law, 63 fundamental theorem of vector calculus, 250
Euler, Leonhard, 66
Euler-Lagrange equation, 184 Galilei, Galileo, 233
R
Euler-Lagrange equations, 184, 251 Galileo’s law, 155
Euler-Mascheroni constant, 121 gauge field, 185
event, 162 gauge fixing, 45
extrinsic, 70 gauge function, 43
gauge invariant, 44
Fäldt, Göran, xxii gauge symmetry, 59
D
272 INDEX
FT
Hamilton operator, 50, 244 Jarlskog, Cecilia, xxii
Hamilton, William Rowen, 244 Jefimenko equations, 92
Hamiltonian, 50, 252 Joule heat power, 61
Heaviside, Oliver, 59, 150, 196, 200
Heaviside-Larmor-Rainich transformation, 72 Karlsson, Roger, xxii
Heaviside-Lorentz units, 15 Kelvin function, 141
helical base vectors, 31, 253 Kelvin, Lord, 59
Helmholtz decomposition, 250 Kerr effect, 193
Helmholtz equation, 118 kinetic energy, 182, 251
Helmholtz’s theorem, 5, 28, 36, 250 kinetic momentum, 181
help vector, 112 Kirchhoff gauge, 42
Kirchhoff, Gustav Robert, 42
A
Hermitian conjugate, 215
Hertz vector, 110 Kohlrausch, Rudolf, 3
Hertz’s method, 108 Kopka, Helmut, xxii
Hertz, Heinrich Rudolf, 103 Kronecker delta tensor, 238
Hertzian dipole, 143, 145 Kronecker product, 242
heterodyning, 193
Hodge star operator, 72 Lagrange density, 183
R
homogeneous vector wave equations, 205 Lagrange function, 182, 251
Hooke’s law, 182 Lagrange’s formula, 258
Huygens’s principle, 38 Lagrangian, 182, 251
lamellar, 250
identity element, 162 Laplace operator, 246
in a medium, 199 Laplacian, 246
D
INDEX 273
FT
Lorentz force, 14, 63, 69 matrix representation, 238
Lorentz force density, 62 Maxwell stress tensor, 62, 76
Lorentz power, 60 Maxwell’s displacement current, 9
Lorentz power density, 60 Maxwell’s macroscopic equations, 195
Lorentz space, 159, 236 Maxwell’s microscopic equations, 15
Lorentz torque, 69 Maxwell, James Clerk, xxi, 1, 19
Lorentz torque density, 67 Maxwell-Lorentz equations, 15
Lorentz transformation, 157 Maxwell-Lorentz source equations, 20
Lorentz, Hendrik Antoon, 1, 37 mechanical angular momentum, 66
Lorenz, Ludvig Valentin, 37 mechanical angular momentum density, 67
Lorenz-Lorentz gauge, 44 mechanical energy, 60
A
Lorenz-Lorentz gauge condition, 37, 167 mechanical kinetic energy density, 60
lowering of index, 240 mechanical Lagrange density, 185
Lundborg, Bengt, xxii mechanical linear momentum, 66
mechanical linear momentum density, 63
M1 radiation, 116 mechanical moment of momentum, 66
Møller scattering, 143 mechanical orbital angular momentum, 67
R
Mach cone, 200 mechanical spin angular momentum, 67
magnetic charge density, 15, 72 mechanical torque, 67
magnetic current density, 15, 72 Mechanics, 2
magnetic dipole moment, 116, 194 metamaterials, 196
magnetic dipole moment per unit volume, 194 metric, 236, 241
magnetic dipole radiation, 116 metric tensor, 157, 236, 240
D
274 INDEX
FT
null vector, 161 Poynting’s theorem, 61
Poynting, John Henry, 59
observation point, 2, 4 prefactor, 243
Ohm’s law, 11, 202 probability density, 50
Ohmic losses, 61 Proca Lagrangian, 189
one-dimensional wave equation, 207 propagation vector, 31
Optics, 2 propagator, 38, 111
orbital angular momentum operator, 71, 80 proper time, 160
orthogonality condition, 264 propositional calculus, 233
outer product, 242 pseudo-Riemannian space, 163
pseudoscalar, 234
Palmer, Kristoffer, vi, xx
A
pseudoscalars, 242
parallelity condition, 265 pseudotensor, 234
paraxial approximation, 95 pseudotensors, 242
parity transformation, 54, 242 pseudovector, 55, 170, 234, 242
Parseval’s identity, 142, 153
Pauli, Wolfgang, 51 QCD, 1
Peierls, Sir Rudolf, 234 QED, 1, 191
R
permeability of free space, 6 quadratic differential form, 160, 241
permittivity of free space, 3 Quantum Chromodynamics, 1
phase velocity, 196 Quantum Electrodynamics, 1
photon, 153, 212 quantum electrodynamics, 40, 191
photons, 191 quantum mechanical non-linearity, 4
physical observable, 20, 157
D
INDEX 275
FT
Riemann-Silberstein vector, 22 superposition principle, 26
Riemannian metric, 160 symmetries, 53
Riemannian space, 157, 236 synchrotron radiation, 137, 141
right-hand circular polarisation, 32, 79 synchrotron radiation lobe width, 138
Rohrlich, Fritz, 1, 19
rotation-less, 250 t’Hooft, Gerardus, xxii
rotational, 9, 250 Tamburini, Fabrizio, xx
rotational degree of freedom, 107 Tamm, Igor’ Evgen’evich, 200
rotational invariance, 72 telegrapher’s equation, 207, 210
rotational momentum, 68 temporal average, 22
temporal dispersive media, 11
A
Salam, Abdus, 1 temporal Fourier components, 25
scalar, 234, 245 temporal Fourier series, 25
scalar field, 161, 236 temporal gauge, 45
scalar product, 241 temporal spectral components, 25
Schrödinger equation, 50 temporal translational invariance, 61
Schwinger, Julian Seymour, 1, 15 tensor, 234
R
self-force effects, 203 tensor field, 238
shock front, 200 tensor notation, 239
SI units, 3, 15 tensor product, 242
signature, 158, 255 Then, Holger, xx
simultaneous coordinate, 150 thermodynamic entropy, 56
Sjöholm, Johan, vi, xx Thermodynamics, 2
D
276 INDEX
FT
tryad, 244
uncertainty principle, 56
uncoupled inhomogeneous wave equations, 37
unit dyad, 243
unit tensor, 243
universal constant, 167
unmagnetised plasma, 203
vacuum polarisation, 4
vacuum wave number, 206
vacuum wavelength, 38
A
variational principle, 251
Vavilov, Sergey Ivanovich, 200
L
Vavilov-Cerenkov cone, 200
L
Vavilov-Cerenkov radiation, 198, 200
vector, 234
vector product, 241
R
vector wave equations, 24
vector waves, 25
velocity field, 128
velocity gauge condition, 42
virtual simultaneous coordinate, 128
von Helmholtz, Hermann Ludwig Ferdinand, 5
D
FT
A
R
D
Draft version released 12th June 2013 at 18:10 CET—Downloaded from http://www.plasma.uu.se/CED/Book
Sheet: 300
. of 300
FT
A
R
ISBN 978-0-486-4773-2