00 voturi pozitive00 voturi negative

26 vizualizări209 paginiSep 30, 2012

© Attribution Non-Commercial (BY-NC)

PDF, TXT sau citiți online pe Scribd

Attribution Non-Commercial (BY-NC)

26 vizualizări

00 voturi pozitive00 voturi negative

Attribution Non-Commercial (BY-NC)

Sunteți pe pagina 1din 209

Bo Thid

Also available

by Tobia Carozzi, Anders Eriksson, Bengt Lundborg, Bo Thid and Mattias Waldenvik

Bo Thid

Swedish Institute of Space Physics and Department of Astronomy and Space Physics Uppsala University, Sweden

U B C AB U S

A This book was typeset in L TEX 2 on an HP9000/700 series workstation and printed on an HP LaserJet 5000GN printer.

Copyright 1997, 1998, 1999, 2000 and 2001 by Bo Thid Uppsala, Sweden All rights reserved. Electromagnetic Field Theory ISBN X-XXX-XXXXX-X

Contents

Preface 1 Classical Electrodynamics 1.1 Electrostatics . . . . . . . . . . . . . . . . . . . 1.1.1 Coulombs law . . . . . . . . . . . . . . 1.1.2 The electrostatic eld . . . . . . . . . . . 1.2 Magnetostatics . . . . . . . . . . . . . . . . . . 1.2.1 Ampres law . . . . . . . . . . . . . . . 1.2.2 The magnetostatic eld . . . . . . . . . . 1.3 Electrodynamics . . . . . . . . . . . . . . . . . 1.3.1 Equation of continuity for electric charge 1.3.2 Maxwells displacement current . . . . . 1.3.3 Electromotive force . . . . . . . . . . . . 1.3.4 Faradays law of induction . . . . . . . . 1.3.5 Maxwells microscopic equations . . . . 1.3.6 Maxwells macroscopic equations . . . . 1.4 Electromagnetic Duality . . . . . . . . . . . . .

xi 1 1 2 2 5 5 6 8 8 9 10 11 14 14 15

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

Example 1.1 Faradays law as a consequence of conservation of magnetic charge . . . . . . . . . . . . . 16 Example 1.2 Duality of the electromagnetodynamic equations 18 Example 1.3 Diracs symmetrised Maxwell equations for a xed mixing angle . . . . . . . . . . . . . . . . 18 Example 1.4 The complex eld six-vector . . . . . . . . . 20 Example 1.5 Duality expressed in the complex eld six-vector 20

Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 2 Electromagnetic Waves 2.1 The Wave Equations . . . . . . . . . . . . . . . . 2.1.1 The wave equation for E . . . . . . . . . . 2.1.2 The wave equation for B . . . . . . . . . . 2.1.3 The time-independent wave equation for E 2.2 23 24 24 24 25 26 28 29 30

. . . .

. . . .

. . . .

. . . .

. . . .

. . . . . . . .

. . . . . . . .

ii

C ONTENTS

2.3 Observables and Averages . . . . . . . . . . . . . . . . . . . . 32 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 3 Electromagnetic Potentials 3.1 The Electrostatic Scalar Potential . . . . . . . . . . . . . . . . 3.2 The Magnetostatic Vector Potential . . . . . . . . . . . . . . . 3.3 The Electrodynamic Potentials . . . . . . . . . . . . . . . . . 3.3.1 Electrodynamic gauges . . . . . . . . . . . . . . . . . Lorentz equations for the electrodynamic potentials . . Gauge transformations . . . . . . . . . . . . . . . . . 3.3.2 Solution of the Lorentz equations for the electromagnetic potentials . . . . . . . . . . . . . . . . . . . . . The retarded potentials . . . . . . . . . . . . . . . . . Example 3.1 Electromagnetodynamic potentials . . . . . Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Relativistic Electrodynamics 4.1 The Special Theory of Relativity . . . . . . . . . . . . . . . . 4.1.1 The Lorentz transformation . . . . . . . . . . . . . . 4.1.2 Lorentz space . . . . . . . . . . . . . . . . . . . . . . Radius four-vector in contravariant and covariant form Scalar product and norm . . . . . . . . . . . . . . . . Metric tensor . . . . . . . . . . . . . . . . . . . . . . Invariant line element and proper time . . . . . . . . . Four-vector elds . . . . . . . . . . . . . . . . . . . . The Lorentz transformation matrix . . . . . . . . . . . The Lorentz group . . . . . . . . . . . . . . . . . . . 4.1.3 Minkowski space . . . . . . . . . . . . . . . . . . . . 4.2 Covariant Classical Mechanics . . . . . . . . . . . . . . . . . 4.3 Covariant Classical Electrodynamics . . . . . . . . . . . . . . 4.3.1 The four-potential . . . . . . . . . . . . . . . . . . . 4.3.2 The Linard-Wiechert potentials . . . . . . . . . . . . 4.3.3 The electromagnetic eld tensor . . . . . . . . . . . . Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Electromagnetic Fields and Particles 5.1 Charged Particles in an Electromagnetic Field 5.1.1 Covariant equations of motion . . . . Lagrange formalism . . . . . . . . . Hamiltonian formalism . . . . . . . . 35 35 36 36 38 38 39 40 44 44 45 47 47 48 49 50 50 51 52 54 54 54 55 57 59 59 60 62 65 67 67 67 68 70

. . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

iii

5.2

Covariant Field Theory . . . . . . . . . . . . . . . . . . . . . . 74 5.2.1 Lagrange-Hamilton formalism for elds and interactions 74 The electromagnetic eld . . . . . . . . . . . . . . . . . 78

Example 5.1 Field energy difference expressed in the eld tensor . . . . . . . . . . . . . . . . . . . . . . 79

Other elds . . . . . . . . . . . . . . . . . . . . . . . . 82 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 6 Electromagnetic Fields and Matter 6.1 Electric Polarisation and Displacement . . . . . . . . 6.1.1 Electric multipole moments . . . . . . . . . 6.2 Magnetisation and the Magnetising Field . . . . . . . 6.3 Energy and Momentum . . . . . . . . . . . . . . . . 6.3.1 The energy theorem in Maxwells theory . . 6.3.2 The momentum theorem in Maxwells theory Bibliography . . . . . . . . . . . . . . . . . . . . . . . . 85 85 85 88 90 90 91 93 95 97 99 101 104 104 104 106 107 107 108 111 114 114 118 120 121 122 123 125 127 132

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

7 Electromagnetic Fields from Arbitrary Source Distributions 7.1 The Magnetic Field . . . . . . . . . . . . . . . . . . . . . 7.2 The Electric Field . . . . . . . . . . . . . . . . . . . . . . 7.3 The Radiation Fields . . . . . . . . . . . . . . . . . . . . 7.4 Radiated Energy . . . . . . . . . . . . . . . . . . . . . . . 7.4.1 Monochromatic signals . . . . . . . . . . . . . . . 7.4.2 Finite bandwidth signals . . . . . . . . . . . . . . Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

8 Electromagnetic Radiation and Radiating Systems 8.1 Radiation from Extended Sources . . . . . . . . . . . . . . . 8.1.1 Radiation from a one-dimensional current distribution 8.1.2 Radiation from a two-dimensional current distribution 8.2 Multipole Radiation . . . . . . . . . . . . . . . . . . . . . . . 8.2.1 The Hertz potential . . . . . . . . . . . . . . . . . . . 8.2.2 Electric dipole radiation . . . . . . . . . . . . . . . . 8.2.3 Magnetic dipole radiation . . . . . . . . . . . . . . . 8.2.4 Electric quadrupole radiation . . . . . . . . . . . . . . 8.3 Radiation from a Localised Charge in Arbitrary Motion . . . . 8.3.1 The Linard-Wiechert potentials . . . . . . . . . . . . 8.3.2 Radiation from an accelerated point charge . . . . . . The differential operator method . . . . . . . . . . . . Example 8.1 The elds from a uniformly moving charge .

. . . . . . . . . . . . .

iv

C ONTENTS

8.3.3

Example 8.2 The convection potential and the convection force134 Radiation for small velocities . . . . . . . . . . . . . . 137 Bremsstrahlung . . . . . . . . . . . . . . . . . . . . . . 138 Example 8.3 Bremsstrahlung for low speeds and short acceleration times . . . . . . . . . . . . . . . . . 141

Cyclotron and synchrotron radiation . . . Cyclotron radiation . . . . . . . . . . . . Synchrotron radiation . . . . . . . . . . . Radiation in the general case . . . . . . . Virtual photons . . . . . . . . . . . . . . 8.3.5 Radiation from charges moving in matter Vavilov-Cerenkov radiation . . . . . . . Bibliography . . . . . . . . . . . . . . . . . . . . . .

8.3.4

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

143 145 146 148 149 151 153 158 159 159 159 159 159 159 160 160 160 160 160 160 160 160 161 161 161 161 162 162 162 162 162 162 162 163

F Formulae F.1 The Electromagnetic Field . . . . . . . . . . . . . . . . . . . F.1.1 Maxwells equations . . . . . . . . . . . . . . . . . . Constitutive relations . . . . . . . . . . . . . . . . . . F.1.2 Fields and potentials . . . . . . . . . . . . . . . . . . Vector and scalar potentials . . . . . . . . . . . . . . Lorentz gauge condition in vacuum . . . . . . . . . . F.1.3 Force and energy . . . . . . . . . . . . . . . . . . . . Poyntings vector . . . . . . . . . . . . . . . . . . . . Maxwells stress tensor . . . . . . . . . . . . . . . . . F.2 Electromagnetic Radiation . . . . . . . . . . . . . . . . . . . F.2.1 Relationship between the eld vectors in a plane wave F.2.2 The far elds from an extended source distribution . . F.2.3 The far elds from an electric dipole . . . . . . . . . . F.2.4 The far elds from a magnetic dipole . . . . . . . . . F.2.5 The far elds from an electric quadrupole . . . . . . . F.2.6 The elds from a point charge in arbitrary motion . . . F.2.7 The elds from a point charge in uniform motion . . . F.3 Special Relativity . . . . . . . . . . . . . . . . . . . . . . . . F.3.1 Metric tensor . . . . . . . . . . . . . . . . . . . . . . F.3.2 Covariant and contravariant four-vectors . . . . . . . . F.3.3 Lorentz transformation of a four-vector . . . . . . . . F.3.4 Invariant line element . . . . . . . . . . . . . . . . . . F.3.5 Four-velocity . . . . . . . . . . . . . . . . . . . . . . F.3.6 Four-momentum . . . . . . . . . . . . . . . . . . . . F.3.7 Four-current density . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

F.3.8 Four-potential . . . . . . . . . . . F.3.9 Field tensor . . . . . . . . . . . . F.4 Vector Relations . . . . . . . . . . . . . . F.4.1 Spherical polar coordinates . . . . Base vectors . . . . . . . . . . . Directed line element . . . . . . . Solid angle element . . . . . . . . Directed area element . . . . . . Volume element . . . . . . . . . F.4.2 Vector formulae . . . . . . . . . . General vector algebraic identities General vector analytic identities . Special identities . . . . . . . . . Integral relations . . . . . . . . . Bibliography . . . . . . . . . . . . . . . . . . Appendices M Mathematical Methods M.1 Scalars, Vectors and Tensors M.1.1 Vectors . . . . . . . Radius vector . . . . M.1.2 Fields . . . . . . . . Scalar elds . . . . . Vector elds . . . . Tensor elds . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

163 163 163 163 163 164 164 164 164 164 164 165 165 165 166 159 167 167 167 167 169 169 169 170 171

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

M.1.3 Vector algebra . . . . . . . . . . . . . . . . . . . . . . 176 Scalar product . . . . . . . . . . . . . . . . . . . . . . 176 Example M.3 Inner products in complex vector space . . . . 176

Example M.4 Scalar product, norm and metric in Lorentz space . . . . . . . . . . . . . . . . . . . . . . 177 Example M.5 Metric in general relativity . . . . . . . . . . 177

Dyadic product . Vector product . M.1.4 Vector analysis . The del operator

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . . .

. . . . .

. . . . .

vi

C ONTENTS

Example M.7 Gradients of scalar functions of relative distances in 3D . . . . . . . . . . . . . . . . . . . 181

The divergence . . . . . . . . . . . . . . . . . . Example M.8 Divergence in 3D . . . . . . . . . . The Laplacian . . . . . . . . . . . . . . . . . . . Example M.9 The Laplacian and the Dirac delta . . The curl . . . . . . . . . . . . . . . . . . . . . . Example M.10 The curl of a gradient . . . . . . . . Example M.11 The divergence of a curl . . . . . . M.2 Analytical Mechanics . . . . . . . . . . . . . . . . . . . M.2.1 Lagranges equations . . . . . . . . . . . . . . . M.2.2 Hamiltons equations . . . . . . . . . . . . . . . Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

182 182 182 182 183 183 184 185 185 185 186

List of Figures

1.1 1.2 1.3 4.1 4.2 4.3 5.1 7.1 8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9 8.10 8.11 8.12 8.13

Coulomb interaction . . . . . . . . . . . . . . . . . . . . . . . 3 Ampre interaction . . . . . . . . . . . . . . . . . . . . . . . . 6 Moving loop in a varying B eld . . . . . . . . . . . . . . . . . 12 Relative motion of two inertial systems . . . . . . . . . . . . . 48 Rotation in a 2D Euclidean space . . . . . . . . . . . . . . . . . 55 Minkowski diagram . . . . . . . . . . . . . . . . . . . . . . . . 56 Linear one-dimensional mass chain . . . . . . . . . . . . . . . . 75 Radiation in the far zone . . . . . . . . . . . . . . . . . . . . . 103 Linear antenna . . . . . . . . . . . . . . . . . . . . . Electric dipole geometry . . . . . . . . . . . . . . . . Loop antenna . . . . . . . . . . . . . . . . . . . . . . Multipole radiation geometry . . . . . . . . . . . . . . Electric dipole geometry . . . . . . . . . . . . . . . . Radiation from a moving charge in vacuum . . . . . . An accelerated charge in vacuum . . . . . . . . . . . . Angular distribution of radiation during bremsstrahlung Location of radiation during bremsstrahlung . . . . . . Radiation from a charge in circular motion . . . . . . . Synchrotron radiation lobe width . . . . . . . . . . . . The perpendicular eld of a moving charge . . . . . . Vavilov-Cerenkov cone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 109 111 116 119 123 126 139 140 144 146 149 155

vii

To the memory of professor L EV M IKHAILOVICH E RUKHIMOV dear friend, great physicist and a truly remarkable human being.

If you understand, things are such as they are If you do not understand, things are such as they are G ENSHA

Preface

This book is the result of a twenty-ve year long love affair. In 1972, I took my rst advanced course in electrodynamics at the Theoretical Physics department, Uppsala University. Shortly thereafter, I joined the research group there and took on the task of helping my supervisor, professor P ER -O LOF F R MAN , with the preparation of a new version of his lecture notes on Electricity Theory. These two things opened up my eyes for the beauty and intricacy of electrodynamics, already at the classical level, and I fell in love with it. Ever since that time, I have off and on had reason to return to electrodynamics, both in my studies, research and teaching, and the current book is the result of my own teaching of a course in advanced electrodynamics at Uppsala University some twenty odd years after I experienced the rst encounter with this subject. The book is the outgrowth of the lecture notes that I prepared for the four-credit course Electrodynamics that was introduced in the Uppsala University curriculum in 1992, to become the ve-credit course Classical Electrodynamics in 1997. To some extent, parts of these notes were based on lecture notes prepared, in Swedish, by B ENGT L UNDBORG who created, developed and taught the earlier, two-credit course Electromagnetic Radiation at our faculty. Intended primarily as a textbook for physics students at the advanced undergraduate or beginning graduate level, I hope the book may be useful for research workers too. It provides a thorough treatment of the theory of electrodynamics, mainly from a classical eld theoretical point of view, and includes such things as electrostatics and magnetostatics and their unication into electrodynamics, the electromagnetic potentials, gauge transformations, covariant formulation of classical electrodynamics, force, momentum and energy of the electromagnetic eld, radiation and scattering phenomena, electromagnetic waves and their propagation in vacuum and in media, and covariant Lagrangian/Hamiltonian eld theoretical methods for electromagnetic elds, particles and interactions. The aim has been to write a book that can serve both as an advanced text in Classical Electrodynamics and as a preparation for studies in Quantum Electrodynamics and related subjects. In an attempt to encourage participation by other scientists and students in the authoring of this book, and to ensure its quality and scope to make it useful in higher university education anywhere in the world, it was produced within a World-Wide Web (WWW) project. This turned out to be a rather successful

xi

xii

P REFACE

move. By making an electronic version of the book freely down-loadable on the net, I have not only received comments on it from fellow Internet physicists around the world, but know, from WWW hit statistics that at the time of writing this, the book serves as a frequently used Internet resource. This way it is my hope that it will be particularly useful for students and researchers working under nancial or other circumstances that make it difcult to procure a printed copy of the book. I am grateful not only to Per-Olof Frman and Bengt Lundborg for providing the inspiration for my writing this book, but also to C HRISTER WAHLBERG and G RAN FLDT, Uppsala University, and YAKOV I STOMIN, Lebedev Institute, Moscow, for interesting discussions on electrodynamics and relativity in general and on this book in particular. I also wish to thank my former graduate students M ATTIAS WALDENVIK and T OBIA C AROZZI as well as A NDERS E RIKSSON, all at the Swedish Institute of Space Physics, Uppsala Division, who all have participated in the teaching and commented on the material covered in the course and in this book. Thanks are also due to my longterm space physics colleague H ELMUT KOPKA of the Max-Planck-Institut fr Aeronomie, Lindau, Germany, who not only taught me about the practical aspects of the of high-power radio wave transmitters and transmission lines, but A also about the more delicate aspects of typesetting a book in TEX and LTEX. I am particularly indebted to Academician professor V ITALIY L. G INZBURG for his many fascinating and very elucidating lectures, comments and historical footnotes on electromagnetic radiation while cruising on the Volga river during our joint Russian-Swedish summer schools. Finally, 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. I dedicate this book to my son M ATTIAS, my daughter K AROLINA, my high-school physics teacher, S TAFFAN RSBY, and to my fellow members of the C APELLA P EDAGOGICA U PSALIENSIS. Uppsala, Sweden February, 2001 B O T HID

1

Classical Electrodynamics

Classical electrodynamics deals with electric and magnetic elds and interactions caused by macroscopic distributions of electric charges and currents. This means that the concepts of localised electric charges and currents assume the validity of certain mathematical limiting processes in which it is considered possible for the charge and current distributions to be localised in innitesimally small volumes of space. Clearly, this is in contradiction to electromagnetism on a truly microscopic scale, where charges and currents are known to be spatially extended objects. However, the limiting processes used will yield results which are correct on small as well as large macroscopic scales. In this chapter we start with the force interactions in classical electrostatics and classical magnetostatics and introduce the static electric and magnetic elds and nd two uncoupled systems of equations for them. Then we see how the conservation of electric charge and its relation to electric current leads to the dynamic connection between electricity and magnetism and how the two can be unied in one theory, classical electrodynamics, described by one system of coupled dynamic eld equationsthe Maxwell equations. At the end of the chapter we study Diracs symmetrised form of Maxwells equations by introducing (hypothetical) magnetic charges and magnetic currents into the theory. While not identied unambiguously in experiments yet, magnetic charges and currents make the theory much more appealing for instance by allowing for duality transformations in a most natural way.

1.1 Electrostatics

The theory which describes physical phenomena related to the interaction between stationary electric charges or charge distributions in space is called electrostatics.1 For a long time electrostatics was considered an independent phys1 The

C LASSICAL E LECTRODYNAMICS

ical theory of its own, alongside other physical theories such as mechanics and thermodynamics.

It has been found experimentally that in classical electrostatics the interaction between two stationary electrically charged bodies can be described in terms of a mechanical force. Let us consider the simple case described by Figure 1.1 on the facing page. Let F denote the force acting on a electrically charged particle with charge q located at x, due to the presence of a charge q located at x . According to Coulombs law this force is, in vacuum, given by the expression

where in the last step Equation (M.80) on page 181 was used. In SI units, which we shall use throughout, the force F is measured in Newton (N), the electric charges q and q in Coulomb (C) [= Ampre-seconds (As)], and the length |x x | in metres (m). The constant 0 = 107 /(4c2 ) 8.8542 1012 Farad per metre (F/m) is the vacuum permittivity and c 2.9979 108 m/s is the speed of light in vacuum. In CGS units 0 = 1/(4) and the force is measured in dyne, electric charge in statcoulomb, and length in centimetres (cm).

Instead of describing the electrostatic interaction in terms of a force action at a distance, it turns out that it is often more convenient to introduce the concept of a eld and to describe the electrostatic interaction in terms of a static vectorial electric eld Estat dened by the limiting process Estat lim

def

q0

F q

where F is the electrostatic force, as dened in Equation (1.1), from a net electric charge q on the test particle with a small electric net electric charge q.

The whole theory of electrostatics constitutes a group of abstract ideas and general propositions, formulated in the clear and concise language of geometry and algebra, and connected with one another by the rules of strict logic. This whole fully satises the reason of a French physicist and his taste for clarity, simplicity and order. . .

F(x) =

qq qq x x = 3 40 |x x | 40

1 |x x |

qq 40

1 |x x |

(1.1)

(1.2)

1.1

E LECTROSTATICS

q xx q x

O F IGURE 1.1: Coulombs law describes how a static electric charge q, located at a point x relative to the origin O, experiences an electrostatic force from a static electric charge q located at x .

Since the purpose of the limiting process is to assure that the test charge q does not inuence the eld, the expression for Estat does not depend explicitly on q but only on the charge q and the relative radius vector x x . This means that we can say that any net electric charge produces an electric eld in the space that surrounds it, regardless of the existence of a second charge anywhere in this space.1 Using (1.1) and Equation (1.2) on the preceding page, and Formula (F.71) on page 165, we nd that the electrostatic eld Estat at the eld point x (also known as the observation point), due to a eld-producing electric charge q at the source point x , is given by

where in the last step Equation (M.80) on page 181 was used. In the presence of several eld producing discrete electric charges qi , located at the points xi , i = 1, 2, 3, . . . , respectively, in an otherwise empty space,

the preface to the rst edition of the rst volume of his book A Treatise on Electricity and Magnetism, rst published in 1873, James Clerk Maxwell describes this in the following, almost poetic, manner [6]: For instance, Faraday, in his minds eye, saw lines of force traversing all space where the mathematicians saw centres of force attracting at a distance: Faraday saw a medium where they saw nothing but distance: Faraday sought the seat of the phenomena in real actions going on in the medium, they were satised that they had found it in a power of action at a distance impressed on the electric uids.

1 In

Estat (x) =

q q xx = 40 |x x |3 40

1 |x x |

q 40

1 |x x |

(1.3)

C LASSICAL E LECTRODYNAMICS

If the discrete electric charges are small and numerous enough, we introduce the electric charge density located at x and write the total eld as

the assumption of linearity of vacuum1 allows us to superimpose their individual E elds into a total E eld x xi 1 Estat (x) = q (1.4) 40 i x x 3 i

i

1 xx 3 1 1 d x = (x ) (x ) d3x 3 40 V 40 V |x x | |x x | (1.5) (x ) 3 1 = dx 40 V |x x | where we used Formula (F.71) on page 165 and the fact that (x ) does not depend on the unprimed coordinates on which operates. We emphasise that Equation (1.5) above is valid for an arbitrary distribution of electric charges, including discrete charges, in which case can be expressed in terms of one or more Dirac delta distributions. Since, according to formula Equation (M.90) on page 184, [ (x)] 0 for any 3D 3 scalar eld (x), we immediately nd that in electrostatics

V

i.e., that Estat is an irrotational eld. Taking the divergence of the general Estat expression for an arbitrary electric charge distribution, Equation (1.5), and using the representation of the Dirac delta distribution, Equation (M.85) on page 183, we nd that 1 xx 3 (x ) dx 40 V |x x |3 1 1 (x ) d3x = 40 V |x x | 1 1 = (x ) 2 d3x 40 V |x x | 1 = (x )(x x ) d3x 0 V (x) = 0 which is Gausss law in differential form.

1 In

fact, vacuum exhibits a quantum mechanical nonlinearity due to vacuum polarisation effects manifesting themselves in the momentary creation and annihilation of electron-positron pairs, but classically this nonlinearity is negligible.

Estat (x) =

Estat (x) =

1 40

(x ) 3 dx |x x |

=0

Estat (x) =

(1.6)

(1.7)

1.2

M AGNETOSTATICS

1.2 Magnetostatics

While electrostatics deals with static electric charges, magnetostatics deals with stationary electric currents, i.e., electric charges moving with constant speeds, and the interaction between these currents. Let us discuss this theory in some detail.

Experiments on the interaction between two small loops of electric current have shown that they interact via a mechanical force, much the same way that electric charges interact. Let F denote such a force acting on a small loop C carrying a current J located at x, due to the presence of a small loop C carrying a current J located at x . According to Ampres law this force is, in vacuum, given by the expression

C C

Here dl and dl are tangential line elements of the loops C and C , respectively, and, in SI units, 0 = 4 107 1.2566 106 H/m is the vacuum permeability. From the denition of 0 and 0 (in SI units) we observe that 107 1 (F/m) 4 107 (H/m) = 2 (s2 /m2 ) (1.9) 2 4c c which is a useful relation. At rst glance, Equation (1.8) above may appear unsymmetric in terms of the loops and therefore to be a force law which is in contradiction with Newtons third law. However, by applying the vector triple product bac-cab Formula (F.54) on page 164, we can rewrite (1.8) as 0 0 =

C C

(1.10) xx dl dl 3 C C |x x | Recognising the fact that the integrand in the rst integral is an exact differential so that this integral vanishes, we can rewrite the force expression, Equation (1.8) above, in the following symmetric way

F(x) =

0 JJ 4

C C

xx dl dl |x x |3

F(x) =

0 JJ 4 0 JJ 4

dl

1 |x x |

dl

F(x) =

0 JJ 4 0 JJ = 4

C C

dl

dl (x x ) |x x |3

dl dl

1 |x x |

(1.8)

(1.11)

C LASSICAL E LECTRODYNAMICS

J C dl xx x C J x O F IGURE 1.2: Ampres law describes how a small loop C, carrying a static electric current J through its tangential line element dl located at x, experiences a magnetostatic force from a small loop C , carrying a static electric current J through the tangential line element dl located at x . The loops can have arbitrary shapes as long as they are simple and closed. dl

In analogy with the electrostatic case, we may attribute the magnetostatic interaction to a vectorial magnetic eld Bstat . I turns out that Bstat can be dened through dBstat (x)

def

0 J xx dl 4 |x x |3

(1.12)

which expresses the small element dBstat (x) of the static magnetic eld set up at the eld point x by a small line element dl of stationary current J at the source point x . The SI unit for the magnetic eld, sometimes called the magnetic ux density or magnetic induction, is Tesla (T). If we generalise expression (1.12) to an integrated steady state electric current density j(x), we obtain Biot-Savarts law:

Bstat (x) =

xx 3 0 0 j(x ) d x = 3 4 V 4 |x x | j(x ) 3 0 = dx 4 V xx

j(x )

1 d3x |x x |

(1.13)

1.2

M AGNETOSTATICS

where we used Formula (F.71) on page 165, Formula (F.60) on page 165, and the fact that j(x ) does not depend on the unprimed coordinates on which operates. Comparing Equation (1.5) on page 4 with Equation (1.13) on the facing page, we see that there exists a close analogy between the expressions for Estat and Bstat but that they differ in their vectorial characteristics. With this denition of Bstat , Equation (1.8) on page 5 may we written

In order to assess the properties of Bstat , we determine its divergence and curl. Taking the divergence of both sides of Equation (1.13) on the facing page and utilising Formula (F.61) on page 165, we obtain

V

since, according to Formula (F.66) on page 165, ( a) vanishes for any vector eld a(x). Applying the operator bac-cab rule, Formula (F.67) on page 165, the curl of Equation (1.13) on the facing page can be written

In the rst of the two integrals on the right hand side, we use the representation of the Dirac delta function given in Formula (F.72) on page 165, and integrate the second one by parts, by utilising Formula (F.59) on page 165 as follows: 1 d3x |x x | V 1 = xk j(x ) xk |x x | V 1 = xk j(x ) dS xk |x x | S

Then we note that the rst integral in the result, obtained by applying Gausss theorem, vanishes when integrated over a large sphere far away from the localised source j(x ), and that the second integral vanishes because j = 0 for

j(x )

1 d3x |x x |

[j(x )

d3x

j(x )

1 d3x |x x |

0 4

j(x )

Bstat (x) =

0 4

Bstat (x) =

F(x) = J

dl Bstat (x)

(1.14)

0 4

j(x ) 3 dx xx

=0

(1.15)

1 d3x |x x | (1.16)

(1.17)

C LASSICAL E LECTRODYNAMICS

stationary currents (no charge accumulation in space). The net result is simply Bstat (x) = 0 j(x )(x x ) d3x = 0 j(x) (1.18)

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 vector differential equations, namely the equations of classical electrostatics (x) 0 stat E (x) = 0 Estat (x) = and the equations of classical magnetostatics Bstat (x) = 0

stat

(x) = 0 j(x)

Since there is nothing a priori which connects Estat directly with Bstat , we must consider classical electrostatics and classical magnetostatics as two independent theories. However, when we include time-dependence, these theories are unied into one theory, classical electrodynamics. This unication of the theories of electricity and magnetism is motivated by two empirically established facts: 1. Electric charge is a conserved quantity and electric current is a transport of electric charge. This fact manifests itself in the equation of continuity and, as a consequence, in Maxwells displacement current. 2. A change in the magnetic ux through a loop will induce an EMF electric eld in the loop. This is the celebrated Faradays law of induction.

Let j denote the electric current density (measured in A/m2 ). In the simplest case it can be dened as j = v where v is the velocity of the electric charge density . In general, j has to be dened in statistical mechanical terms as

(1.19a) (1.19b)

(1.20a) (1.20b)

1.3

E LECTRODYNAMICS

j(t, x) = qe v f (t, x, v) d3v where f (t, x, v) is the (normalised) distribution function for particle species with electric charge qe . The electric charge conservation law can be formulated in the equation of continuity (t, x) + j(t, x) = 0 t

which states that the time rate of change of electric charge (t, x) is balanced by a divergence in the electric current density j(t, x).

We recall from the derivation of Equation (1.18) on the preceding page that there we used the fact that in magnetostatics j(x) = 0. In the case of nonstationary sources and elds, we must, in accordance with the continuity Equation (1.21), set j(t, x) = (t, x)/t. Doing so, and formally repeating the steps in the derivation of Equation (1.18) on the preceding page, we would obtain the formal result

= 0 j(t, x) + 0 0 E(t, x) t

where, in the last step, we have assumed that a generalisation of Equation (1.5) on page 4 to time-varying elds allows us to make the identication

Later, we will need to consider this formal result further. The result is Maxwells source equation for the B eld B(t, x) = 0 j(t, x) + 0 E(t, x) t (1.24)

where the last term 0 E(t, x)/t is the famous displacement current. This term was introduced, in a stroke of genius, by Maxwell in order to make the

(1.23)

1 40 t

(t, x )

1 1 d3x = t 40 |x x | = E(t, x) t

(t, x )

1 d3x |x x |

B(t, x) = 0

(1.21)

0 4 t

(t, x )

1 d3x |x x | (1.22)

10

C LASSICAL E LECTRODYNAMICS

right hand side of this equation divergence free when j(t, x) is assumed to represent the density of the total electric current, which can be split up in ordinary conduction currents, polarisation currents and magnetisation currents. The displacement current is an extra term which behaves like a current density owing in vacuum. As we shall see later, its existence has far-reaching physical consequences as it predicts the existence of electromagnetic radiation that can carry energy and momentum over very long distances, even in vacuum.

If an electric eld E(t, x) is applied to a conducting medium, a current density j(t, x) will be produced in this medium. There exist also hydrodynamical and chemical processes which can create currents. Under certain physical conditions, and for certain materials, one can sometimes assume a linear relationship between the electric current density j and E, called Ohms law: j(t, x) = E(t, x) (1.25)

where is the electric conductivity (S/m). In the most general cases, for instance in an anisotropic conductor, is a tensor. We can view Ohms law, Equation (1.25) above, as the rst term in a Taylor expansion of the law j[E(t, x)]. This general law incorporates non-linear effects such as frequency mixing. Examples of media which are highly non-linear are semiconductors and plasma. We draw the attention to the fact that even in cases when the linear relation between E and j is a good approximation, we still have to use Ohms law with care. The conductivity is, in general, time-dependent (temporal dispersive media) but then it is often the case that Equation (1.25) is valid for each individual Fourier component of the eld. If the current is caused by an applied electric eld E(t, x), this electric eld will exert work on the charges in the medium and, unless the medium is superconducting, there will be some energy loss. The rate at which this energy is expended is j E per unit volume. If E is irrotational (conservative), j will decay away with time. Stationary currents therefore require that an electric eld which corresponds to an electromotive force (EMF) is present. In the presence of such a eld EEMF , Ohms law, Equation (1.25) above, takes the form j = (Estat + EEMF ) The electromotive force is dened as (1.26)

E=

(Estat + EEMF ) dl

(1.27)

1.3

E LECTRODYNAMICS

11

In Subsection 1.1.2 we derived the differential equations for the electrostatic eld. In particular, on page 4 we derived Equation (1.6) which states that Estat (x) = 0 and thus that Estat is a conservative eld (it can be expressed as a gradient of a scalar eld). This implies that the closed line integral of Estat in Equation (1.27) on the preceding page vanishes and that this equation becomes

It has been established experimentally that a nonconservative EMF eld is produced in a closed circuit C if the magnetic ux through this circuit varies with time. This is formulated in Faradays law which, in Maxwells generalised form, reads E(t, x) = E(t, x) dl =

S

where m is the magnetic ux and S is the surface encircled by C which can be interpreted as a generic stationary loop and not necessarily as a conducting circuit. Application of Stokes theorem on this integral equation, transforms it into the differential equation E(t, x) = B(t, x) t (1.30)

which is valid for arbitrary variations in the elds and constitutes the Maxwell equation which explicitly connects electricity with magnetism. Any change of the magnetic ux m will induce an EMF. Let us therefore consider the case, illustrated if Figure 1.3.4 on the following page, that the loop is moved in such a way that it links a magnetic eld which varies during the movement. The convective derivative is evaluated according to the wellknown operator formula d = +v dt t

E=

EEMF dl

(1.28)

d m (t, x) dt

S

d dt

B(t, x) dS =

dS

B(t, x) t

(1.29)

(1.31)

12

C LASSICAL E LECTRODYNAMICS

dS B(x)

v C dl

B(x)

F IGURE 1.3: A loop C which moves with velocity v in a spatially varying magnetic eld B(x) will sense a varying magnetic ux during the motion.

which follows immediately from the rules of differentiation of an arbitrary differentiable function f (t, x(t)). Applying this rule to Faradays law, Equation (1.29) on the previous page, we obtain E(t, x) = d dt B dS = dS B t (v )B dS (1.32)

During spatial differentiation v is to be considered as constant, and Equation (1.15) on page 7 holds also for time-varying elds: B(t, x) = 0 (1.33)

(it is one of Maxwells equations) so that, according to Formula (F.62) on page 165, (B v) = (v )B (1.34)

1.3

E LECTRODYNAMICS

13

allowing us to rewrite Equation (1.32) on the facing page in the following way: E(t, x) = EEMF dl = d dt B dS

With Stokes theorem applied to the last integral, we nally get E(t, x) = EEMF dl = B dS t (B v) dl (1.36)

where EEMF is the eld which is induced in the loop, i.e., in the moving system. The use of Stokes theorem backwards on Equation (1.37) above yields (EEMF v B) = B t (1.38)

In the xed system, an observer measures the electric eld E = EEMF v B (1.39)

Hence, a moving observer measures the following Lorentz force on a charge q qEEMF = qE + q(v B) (1.40)

corresponding to an effective electric eld in the loop (moving observer) EEMF = E + (v B) (1.41)

Hence, we can conclude that for a stationary observer, the Maxwell equation E = B t (1.42)

B dS t

B dS t

(1.35)

(B v) dS

(1.37)

14

C LASSICAL E LECTRODYNAMICS

We are now able to collect the results from the above considerations and formulate the equations of classical electrodynamics valid for arbitrary variations in time and space of the coupled electric and magnetic elds E(t, x) and B(t, x). The equations are E = E+ (t, x) 0 (1.43a) (1.43b) (1.43c) (1.43d)

B =0 t B = 0 E B 0 0 = 0 j(t, x) t

In these equations (t, x) represents the total, possibly both time and space dependent, electric charge, i.e., free as well as induced (polarisation) charges, and j(t, x) represents the total, possibly both time and space dependent, electric current, i.e., conduction currents (motion of free charges) as well as all atomistic (polarisation, magnetisation) currents. As they stand, the equations therefore incorporate the classical interaction between all electric charges and currents in the system and are called Maxwells microscopic equations. Another name often used for them is the Maxwell-Lorentz equations. Together with the appropriate constitutive relations, which relate and j to the elds, and the initial and boundary conditions pertinent to the physical situation at hand, they form a system of well-posed partial differential equations which completely determine E and B.

The microscopic eld equations (1.43) provide a correct classical picture for arbitrary eld and source distributions, including both microscopic and macroscopic scales. However, for macroscopic substances it is sometimes convenient to introduce new derived elds which represent the electric and magnetic elds in which, in an average sense, the material properties of the substances are already included. These elds are the electric displacement D and the magnetising eld H. In the most general case, these derived elds are complicated nonlocal, nonlinear functionals of the primary elds E and B: D = D[t, x; E, B] H = H[t, x; E, B] (1.44a) (1.44b)

1.4

E LECTROMAGNETIC D UALITY

15

Under certain conditions, for instance for very low eld strengths, we may assume that the response of a substance to the elds is linear so that D = E H= B

1

(1.45) (1.46)

i.e., that the derived elds are linearly proportional to the primary elds and that the electric displacement (magnetising eld) is only dependent on the electric (magnetic) eld. The eld equations expressed in terms of the derived eld quantities D and H are D = (t, x) B E+ =0 t B = 0 D H = j(t, x) t (1.47a) (1.47b) (1.47c) (1.47d)

and are called Maxwells macroscopic equations. We will study them in more detail in Chapter 6.

If we look more closely at the microscopic Maxwell equations (1.48), we see that they exhibit a certain, albeit not a complete, symmetry. Let us further make the ad hoc assumption that there exist magnetic monopoles represented by a magnetic charge density, denoted m = m (t, x), and a magnetic current density, denoted jm = jm (t, x). With these new quantities included in the theory, and with the elecric charge density denoted e and the electric current density denoted je , the Maxwell equations can be written E = E+ e 0 (1.48a) (1.48b) (1.48c) (1.48d)

B = 0 jm t B = 0 m E = 0 je B 0 0 t

16

C LASSICAL E LECTRODYNAMICS

We shall call these equations Diracs symmetrised Maxwell equations or the electromagnetodynamic equations Taking the divergence of (1.48b), we nd that ( E) = ( B) 0 jm 0 t (1.49)

where we used the fact that, according to Formula (M.94) on page 184, the divergence of a curl always vanishes. Using (1.48c) to rewrite this relation, we obtain the equation of continuity for magnetic monopoles m + jm = 0 t (1.50)

which has the same form as that for the electric monopoles (electric charges) and currents, Equation (1.21) on page 9. We notice that the new Equations (1.48) on the preceding page exhibit the following symmetry (recall that 0 0 = 1/c2 ): cB E

e m m e

E cB

c cj j

m e

c

m

j cj

which is a particular case ( = /2) of the general duality transformation (depicted by the Hodge star operator) E = E cos + cB sin c = c cos + sin c j = cj cos + j sin

m e m

c B = E sin + cB cos

e e m m e m

= c sin + cos

e e m

j = cj sin + j cos

which leaves the symmetrised Maxwell equations, and hence the physics they describe (often referred to as electromagnetodynamics), invariant. Since E and je are (true or polar) vectors, B a pseudovector (axial vector), e a (true) scalar, then m and , which behaves as a mixing angle in a two-dimensional charge space, must be pseudoscalars and jm a pseudovector.

1.4

E LECTROMAGNETIC D UALITY

17

Postulate 1.1 (Indestructibility of magnetic charge). Magnetic charge exists 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. Use this postulate and Diracs symmetrised form of Maxwells equations to derive Faradays law. The assumption of existence of magnetic charges suggests that there exists a Coulomb law for magnetic elds:

V

[cf. Equation (1.5) on page 4 for Estat ] and, if magnetic currents exist, a Biot-Savart law for electric elds:

V V

Taking the curl of the latter and using the operator bac-cab rule, Formula (F.62) on page 165, we nd that

Comparing with Equation (1.16) on page 7 for Estat and the evaluation of the integrals there we obtain Estat (x) = 0

V

We assume that Formula (1.54) is valid also for time-varying magnetic currents. Then, with the use of the representation of the Dirac delta function, Equation (M.85) on page 183, the equation of continuity for magnetic charge, Equation (1.50) on the facing page, and the assumption of the generalisation of Equation (1.53) above to timedependent magnetic charge distributions, we obtain, formally,

= 0 jm (t, x)

B(t, x) t

(1.57)

"

E(t, x) = 0

0 4 t

m (t, x )

1 d3x |x x |

"

"

0 4

jm (x )

"

Estat (x) =

0 4

jm (x ) 3 dx = V xx 1 0 d3x + [jm (x ) 4 V |x x |

1 d3x |x x | (1.55)

"

Estat (x) =

0 4

jm (x )

0 xx 3 dx = 4 |x x |3

jm (x )

1 d3x |x x | (1.54)

"

"

FARADAY S LAW AS

E XAMPLE 1.1

Bstat (x) =

0 0 xx 3 d x = m (x ) 4 V 4 x |3 |x 1 0 m (x ) d3x = 4 |x x | V

m (x )

1 d3x |x x |

(1.53)

(1.56)

18

C LASSICAL E LECTRODYNAMICS

[cf. Equation (1.22) on page 9] which we recognise as Equation (1.48b) on page 15. A transformation of this electromagnetodynamic result by rotating into the electric realm of charge space, thereby letting jm tend to zero, yields the electrodynamic Equation (1.48b) on page 15, i.e., the Faraday law in the ordinary Maxwell equations. By postulating the indestructibility of an hypothetical magnetic charge, we have thereby been able to replace Faradays experimental results on electromotive forces and induction in loops as a foundation for the Maxwell equations by a more appealing one.

E ND

OF EXAMPLE

1.1

E XAMPLE 1.2

Show that the symmetric, electromagnetodynamic form of Maxwells equations (Diracs symmetrised Maxwell equations), Equations (1.48) on page 15, are invariant under the duality transformation (1.52). Explicit application of the transformation yields e cos + c0 m sin 0 1 e 1 = e cos + m sin = 0 c 0 B 1 = (E cos + cB sin ) + E+ E sin + B cos t t c B 1 E = 0 jm cos cos + c0 je sin + sin t c t 1 E B sin + cos = 0 jm cos + c0 je sin c t t = 0 (cje sin + jm cos ) = 0 jm 1 e B = ( E sin + B cos ) = sin + 0 m cos c c0 = 0 ce sin + m cos = 0 m E= (E cos + cB sin ) = B 1 E = c2 t

1 1 ( E sin + B cos ) 2 (E cos + cB sin ) c c t 1 m 1 B 1 E = 0 j sin + cos + 0 je cos + 2 cos c c t c t (1.61) 1 E 1 B 2 cos sin c t c t 1 m = 0 j sin + je cos = 0 je c

E ND

OF EXAMPLE

1.2

"

"

"

(1.58)

(1.59)

(1.60)

QED

1.4

E LECTROMAGNETIC D UALITY

19

Show that for a xed mixing angle such that m = ce tan jm = cje tan the symmetrised Maxwell equations reduce to the usual Maxwell equations. Explicit application of the xed mixing angle conditions on the duality transformation (1.52) on page 16 yields 1 1 e = e cos + m sin = e cos + ce tan sin c c 1 1 e e 2 e 2 ( cos + sin ) = = cos cos m = ce sin + ce tan cos = ce sin + ce sin = 0 1 e 1 e 2 (j cos + je sin2 ) = j je = je cos + je tan sin = cos cos jm = cje sin + cje tan cos = cje sin + cje sin = 0 (1.62a) (1.62b)

Hence, a xed mixing angle, or, equivalently, a xed ratio between the electric and magnetic charges/currents, hides the magnetic monopole inuence ( m and jm ) on the dynamic equations. We notice that the inverse of the transformation given by Equation (1.52) on page 16 yields E = E cos c B sin This means that E = E cos c B sin (1.65) (1.64)

Furthermore, from the expressions for the transformed charges and currents above, we nd that e 1 e E= = (1.66) 0 cos 0 and B = 0 m = 0 so that E = e 1 e cos 0 = cos 0 0 (1.68) (1.67)

E ND OF

EXAMPLE

1.3

E XAMPLE 1.3

QED

20

C LASSICAL E LECTRODYNAMICS

The invariance of Diracs symmetrised Maxwell equations under the similarity transformation means that the amount of magnetic monopole density m is irrelevant for the physics as long as the ratio m /e = tan is kept constant. So whether we assume that the particles are only electrically charged or have also a magnetic charge with a given, xed ratio between the two types of charges is a matter of convention, as long as we assume that this fraction is the same for all particles. Such particles are referred to as dyons. By varying the mixing angle we can change the fraction of magnetic monopoles at will without changing the laws of electrodynamics. For = 0 we recover the usual Maxwell electrodynamics as we know it.

E XAMPLE 1.4

3,

1. The inner product of G with itself G G = (E + icB) (E + icB) = E 2 c2 B2 + 2icE B is conserved. I.e., E 2 c2 B2 = Const E B = Const as we shall see later. 2. The inner product of G with the complex conjugate of itself G G = (E + icB) (E icB) = E 2 + c2 B2 is proportional to the electromagnetic eld energy. 3. As with any vector, the cross product of G itself vanishes: G G = (E + icB) (E + icB) = E E c2 B B + ic(E B) + ic(B E) = 0 + 0 + ic(E B) ic(E B) = 0 (1.73) (1.72) (1.71a) (1.71b) (1.70)

4. The cross product of G with the complex conjugate of itself G G = (E + icB) (E icB) = E E + c2 B B ic(E B) + ic(B E) = 0 + 0 ic(E B) ic(E B) = 2ic(E B)

E ND

'

&

where E, B

T HE

and hence G

(1.74)

OF EXAMPLE

1.4

1.4

B IBLIOGRAPHY

21

Expressed in the complex eld vector, introduced in Example 1.4 on the facing page, the duality transformation Equations (1.52) on page 16 become G = E + ic B = E cos + cB sin iE sin + icB cos from which it is easy to see that = E(cos i sin ) + icB(cos i sin ) = ei (E + icB) = ei G (1.75)

while

G G = e2i G G

Furthermore, assuming that = (t, x), we see that the spatial and temporal differentiation of G leads to G t G = i(t )ei G + ei t G (1.78a) t G G = iei G + ei G (1.78b) G G = iei G + ei G (1.78c) G

1.5

which means that t G transforms as G itself if is time-independent, and that and G transform as G itself if is space-independent.

E ND OF

EXAMPLE

Bibliography

[1] R. B ECKER, Electromagnetic Fields and Interactions, Dover Publications, Inc., New York, NY, 1982, ISBN 0-486-64290-9. [2] W. G REINER, Classical Electrodynamics, Springer-Verlag, New York, Berlin, Heidelberg, 1996, ISBN 0-387-94799-X. [3] E. H ALLN, Electromagnetic Theory, Chapman & Hall, Ltd., London, 1962. [4] J. D. JACKSON, Classical Electrodynamics, third ed., John Wiley & Sons, Inc., New York, NY . . . , 1999, ISBN 0-471-30932-X. [5] 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. [6] 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.

(( ((

D UALITY EXPRESSED IN

E XAMPLE 1.5

G G =

= ei G ei G = |F|2

(1.76)

(1.77)

22

C LASSICAL E LECTRODYNAMICS

[7] D. B. M ELROSE AND R. C. M C P HEDRAN, Electromagnetic Processes in Dispersive Media, Cambridge University Press, Cambridge . . . , 1991, ISBN 0-52141025-8. [8] 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. [9] J. S CHWINGER, A magnetic model of matter, Science, 165 (1969), pp. 757761. [10] J. S CHWINGER , L. L. D E R AAD , J R ., K. A. M ILTON , AND W. T SAI, Classical Electrodynamics, Perseus Books, Reading, MA, 1998, ISBN 0-7382-0056-5. [11] J. A. S TRATTON, Electromagnetic Theory, McGraw-Hill Book Company, Inc., New York, NY and London, 1953, ISBN 07-062150-0. [12] J. VANDERLINDE, Classical Electromagnetic Theory, John Wiley & Sons, Inc., New York, Chichester, Brisbane, Toronto, and Singapore, 1993, ISBN 0-47157269-1.

2

Electromagnetic Waves

In this chapter we investigate the dynamical properties of the electromagnetic eld by deriving an set of equations which are alternatives to the Maxwell equations. It turns out that these alternative equations are wave equations, indicating that electromagnetic waves are natural and common manifestations of electrodynamics. Maxwells microscopic equations (1.43) on page 14, which are usually written in the following form (t, x) 0 B E = t B = 0 E = B = 0 j(t, x) + 0 0 (2.1a) (2.1b) (2.1c) E t (2.1d)

can be viewed as an axiomatic basis for classical electrodynamics. In particular, these equations are well suited for calculating the electric and magnetic elds E and B from given, prescribed charge distributions (t, x) and current distributions j(t, x) of arbitrary time- and space-dependent form. However, as is well known from the theory of differential equations, these four rst order, coupled partial differential vector equations can be rewritten as two un-coupled, second order partial equations, one for E and one for B. We shall derive these second order equations which, as we shall see are wave equations, and then discuss the implications of them. We shall also show how the B wave eld can be easily calculated from the solution of the E wave equation.

23

24

E LECTROMAGNETIC WAVES

We restrict ourselves to derive the wave equations for the electric eld vector E and the magnetic eld vector B in a volume with no net charge, = 0, and no electromotive force EEMF = 0.

In order to derive the wave equation for E we take the curl of (2.1b) and using (2.1d), to obtain

According to the operator triple product bac-cab rule Equation (F.67) on page 165 ( E) = ( E) E = 0 and since EEMF = 0, Ohms law, Equation (1.26) on page 10, yields j = E we nd that Equation (2.2) can be rewritten (2.5)

2

2

E 0

E 1 2 E =0 t c2 t2

The wave equation for B is derived in much the same way as the wave equation for E. Take the curl of (2.1d) and use Ohms law j = E to obtain ( B) = 0 j + 0 0 ( E) = 0 E + 0 0 ( E) (2.8) t t

E 0

E + 0 E = 0 t t

( E) =

( B) = 0 j + 0 E t t t

(2.2)

(2.3)

(2.6)

(2.7)

2.1

T HE WAVE E QUATIONS

25

which, with the use of Equation (F.67) on page 165 and Equation (2.1c) on page 23 can be rewritten ( B)

2

B = 0

2 B 0 0 2 B t t

(2.9)

Using the fact that, according to (2.1c), ing, we can rewrite this equation as

2

B 0

B 1 2 B =0 t c2 t2

This is the wave equation for the magnetic eld. We notice that it is of exactly the same form as the wave equation for the electric eld, Equation (2.7) on the preceding page.

We now look for a solution to any of the wave equations in the form of a timeharmonic wave. As is clear from the above, it sufces to consider only the E eld, since the results for the B eld follow trivially. We therefore make the following Fourier component Ansatz E = E0 (x)eit and insert this into Equation (2.7) on the facing page. This yields

2

(2.11)

E 0

(2.12)

Introducing the relaxation time = 0 / of the medium in question we can rewrite this equation as

2

E+

2 E=0 c2

E+

2 c2

1+

i E=0

(2.13)

(2.14)

26

E LECTROMAGNETIC WAVES

which is a time-independent wave equation for E, representing weakly damped propagating waves. In the short limit we have instead

2

E + i0 E = 0

(2.15)

which is a time-independent diffusion equation for E. For most metals 1014 s, which means that the diffusion picture is good for all frequencies lower than optical frequencies. Hence, in metallic conductors, the propagation term 2 E/c2 t2 is negligible even for VHF, UHF, and SHF signals. Alternatively, we may say that the displacement current 0 E/t is negligible relative to the conduction current j = E. If we introduce the vacuum wave number (2.16) k= c we can write, using the fact that c = 1/ 0 0 according to Equation (1.9) on page 5, 1 1 = = = 0 0 ck k 0 = R0 0 k (2.17)

where in the last step we introduced the characteristic impedance for vacuum R0 = 0 376.7 0 (2.18)

E XAMPLE 2.1

WAVE EQUATIONS IN

ELECTROMAGNETODYNAMICS

Derive the wave equation for the E eld described by the electromagnetodynamic equations (Diracs symmetrised Maxwell equations) [cf. Equations (1.48) on page 15] E = e 0 B t (2.19a) (2.19b) (2.19c)

E = 0 jm B = 0

m

E (2.19d) 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. B = 0 j e + 0 0 Taking the curl of (2.19b) and using (2.19d), and assuming, for symmetry reasons, that there exists a linear relation between the magnetic current density jm and the magnetic

2.1

T HE WAVE E QUATIONS

27

eld B (the analogue of Ohms law for electric currents, je = e E) jm = m B one nds, noting that 0 0 = 1/c2 , that (2.20)

Using the vector operator identity ( E) = ( E) and the fact that E = 0 for a vanishing net electric charge, we can rewrite the wave equation as

This is the homogeneous electromagnetodynamic wave equation for E we were after. Compared to the ordinary electrodynamic wave equation for E, Equation (2.7) on page 24, 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 component. Then our wave equation becomes

Realising that, according to Formula (2.18) on the facing page, 0 /0 is the square of the vacuum radiation resistance R0 , and rearranging a bit, we obtain the timeindependent wave equation in Diracs symmetrised electrodynamics

0 1 0 m e 2

R2

From this equation we conclude that the existence of magnetic charges (magnetic monopoles), 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 vanishes, the imaginary term does not necessarily vanish and the wave might therefore experience damping (or growth) according as m is positive (or negative) in a perfect electric isolator. Finally, we note that in the particular case that = R0 m e , the wave equation becomes a (time-independent) diffusion equation

E ND OF

EXAMPLE

"

E + i0 e +

m c2

E=0

@8 A97 B

245 3"

E+

2 c2

R2 m e 0 2

1+i

e + m /c2

"

!0

E+

2 c2

"

E + i0 e +

m c2

"

E 0 e +

m c2

E 1 2 E 2 m e E = 0 0 t c2 t2

E+

2 E 2 m e E 0 c2

1 0 m e e + m /c2 E=0 +i 2 0 0

"

= 0 m 0 e E +

1 E E 1 2 E 0 e c2 t t c2 t2

2 E,

(2.21)

(2.22)

(2.23)

E=0

(2.24)

(2.25)

2.1

"

( E) = 0 jm

1 E ( B) = 0 m B 0 j e + 2 t t c t

28

E LECTROMAGNETIC WAVES

Consider now the case where all elds depend only on the distance to a given plane with unit normal n. Then the del operator becomes =n (2.26)

and Maxwells equations attain the form E E n B n B n n =0 = =0 = 0 j(t, x) + 0 0 E E = 0 E + 0 0 t t B t (2.27a) (2.27b) (2.27c) (2.27d)

which simplies to the rst-order ordinary differential equation for the normal component En of the electric eld dEn + En = 0 dt 0 with the solution En = En0 et/0 = En0 et/ (2.30) (2.29)

This, together with (2.27a), 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. Scalar multiplying (2.27b) by n, we similarly nd that

or n B =0 t

0 = n n

= n

B t

0 = n n

= n 0 + 0 0

E t

(2.28)

(2.31)

(2.32)

2.2

P LANE WAVES

29

From this, and (2.27c), we conclude that the only longitudinal component of B must be constant in both time and space. In other words, the only non-static solution must consist of transverse components.

In analogy with Equation (2.7) on page 24, we can easily derive the equation E 1 2 E 2 E 0 =0 2 t c2 t2 (2.33)

This equation, which describes the propagation of plane waves in a conducting medium, is called the telegraphers equation. If the medium is an insulator so that = 0, then the equation takes the form of the one-dimensional wave equation 2 E 1 2 E =0 2 c2 t2 (2.34)

As is well known, each component of this equation has a solution which can be written Ei = f ( ct) + g( + ct), i = 1, 2, 3 (2.35)

where f and g are arbitrary (non-pathological) functions of their respective arguments. This general solution represents perturbations which propagate along , where the f perturbation propagates in the positive direction and the g perturbation propagates in the negative direction. If we assume that our electromagnetic elds E and B are time-harmonic, i.e., that they can each be represented by a Fourier component proportional to exp{it}, the solution of Equation (2.34) becomes E = E0 ei(tk) = E0 ei(

kt)

(2.36)

30

E LECTROMAGNETIC WAVES

Let us consider the lower sign in front of k in the exponent in (2.36). This corresponds to a wave which propagates in the direction of increasing . Inserting this solution into Equation (2.27b) on page 28, gives n E = iB = ik n E k 1 1 n E = k E = k E = 0 0 n E c (2.39)

Hence, to each transverse component of E, there exists an associated magnetic eld given by Equation (2.40) above. If E and/or B has a direction in space which is constant in time, we have a plane polarised wave (or linearly polarised wave).

Assuming that our medium has a nite conductivity , and making the timeharmonic wave Ansatz in Equation (2.33) on the preceding page, we nd that the time-independent telegraphers equation can be written 2 E 2 E + 0 0 2 E + i0 E = 2 + K 2 E = 0 2 where 0 2 c2 0 0 (2.41)

where, in the last step, Equation (2.16) on page 26 was used to introduce the wave number k. Taking the square root of this expression, we obtain

or 2 = 2 k2 = k2 20

k2 1 + i

K=k

1+i

= + i 0

= (2 2 ) + 2i

K 2 = 0 0 2 1 + i

1+i

= k2 1 + i

(2.42)

(2.43)

(2.44)

(2.45) (2.46)

2.2

P LANE WAVES

31

Squaring the latter and combining with the former, one obtains the second order algebraic equation (in 2 ) 2 (2 k2 ) = k 4 2 42 2 0 (2.47)

=k

=k

As a consequence, the solution of the time-independent telegraphers equation, Equation (2.41) on the facing page, can be written E = E0 e ei(t) (2.49)

With the aid of Equation (2.40) on the preceding page we can calculate the associated magnetic eld, and nd that it is given by B= 1 1 1 K k E = ( k E)( + i) = ( k E) |A| ei (2.50)

where we have, in the last step, rewritten + i in the amplitude-phase form |A| exp{i}. 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. In the case that 0 , we can approximate K as follows:

From this analysis we conclude that when the wave impinges perpendicularly upon the medium, the elds are given, inside this medium, by

B = (1 + i)

0 ( n E ) 2

H

)

)

E = E0 exp

0 exp i 2

= 0 0 (1 + i)

= (1 + i) 20

0 2

0 t 2

G

K = k 1+i 0

1 2

1+

G

2

D CDD ) E ) E

CDD D

1+

+1 (2.48a) 1

(2.48b)

0 =k i 1i 0

1 2

k(1 + i)

20

(2.51)

(2.52a) (2.52b)

32

E LECTROMAGNETIC WAVES

In the above we have used complex notation quite extensively. This is for mathematical convenience only. For instance, in this notation differentiations are almost trivial to perform. However, every physical measurable quantity is always real valued. I.e., Ephysical = Re {Emathematical }. It is particularly important to remember this when one works with products of physical quantities. For instance, if we have two physical vectors F and G which both are time-harmonic, i.e., can be represented by Fourier components proportional to exp{it}, then we must make the following interpretation

Furthermore, letting denotes complex conjugate, we can express the real part of the complex vector F as 1 Re {F} = Re F0 (x) eit = [F0 (x) eit + F (x) eit ] 0 2

and similarly for G. Hence, the physically acceptable interpretation of the scalar product of two complex vectors, representing physical observables, is F(t, x) G(t, x) = Re F0 (x) eit Re G0 (x) eit 1 1 = [F0 (x) eit + F (x) eit ] [G0 (x) eit + G (x) eit ] 0 0 2 2 1 = F0 G + F G0 + F0 G0 e2it + F G e2it 0 0 0 0 4 (2.56) 1 = Re F0 G + F0 G0 e2it 0 2 1 = Re F0 eit G eit + F0 G0 e2it 0 2 1 = Re F(t, x) G (t, x) + F0 G0 e2it 2

(2.54)

(2.55)

2.3

B IBLIOGRAPHY

33

Often in physics, we measure temporal averages ( ) of our physical observables. If so, we see that the average of the product of the two physical quantities represented by F and G can be expressed as

Bibliography

[1] J. D. JACKSON, Classical Electrodynamics, third ed., John Wiley & Sons, Inc., New York, NY . . . , 1999, ISBN 0-471-30932-X. [2] 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.

1 1 F G Re {F} Re {G} = Re F G = Re F G 2 2

(2.57)

34

E LECTROMAGNETIC WAVES

3

Electromagnetic Potentials

Instead of expressing the laws of electrodynamics in terms of electric and magnetic elds, it turns out that it is often more convenient to express the theory in terms of potentials. In this chapter we will introduce and study the properties of such potentials.

As we saw in Equation (1.6) on page 4, the electrostatic eld Estat (x) is irrotational. Hence, it may be expressed in terms of the gradient of a scalar eld. If we denote this scalar eld by stat (x), we get Estat (x) = stat (x) (3.1)

Taking the divergence of this and using Equation (1.7) on page 4, we obtain Poissons equation

2 stat

(x) 0

(3.2)

A comparison with the denition of Estat , namely Equation (1.5) on page 4, shows that this equation has the solution stat (x) = 1 40 (x ) 3 d x + |x x | (3.3)

where the integration is taken over all source points x at which the charge density (x ) is non-zero and is an arbitrary quantity which has a vanishing gradient. An example of such a quantity is a scalar constant. The scalar function stat (x) in Equation (3.3) above is called the electrostatic scalar potential.

35

36

E LECTROMAGNETIC P OTENTIALS

Consider the equations of magnetostatics (1.20) on page 8. From Equation (M.94) on page 184 we know that any 3D vector a has the property that ( a) 0 and in the derivation of Equation (1.15) on page 7 in magnetostatics we found that Bstat (x) = 0. We therefore realise that we can always write Bstat (x) = Astat (x) (3.4)

where Astat (x) is called the magnetostatic vector potential. We saw above that the electrostatic potential (as any scalar potential) is not unique: we may, without changing the physics, add to it a quantity whose spatial gradient vanishes. A similar arbitrariness is true also for the magnetostatic vector potential. In the magnetostatic case, we may start from Biot-Savarts law as expressed by Equation (1.13) on page 6. Identifying this expression with Equation (3.4) above allows us to dene the static vector potential as Astat (x) = 0 4 j(x ) 3 d x + a(x) |x x | (3.5)

where a(x) is an arbitrary vector eld whose curl vanishes. From Equation (M.90) on page 184 we know that such a vector can always be written as the gradient of a scalar eld.

Let us now generalise the static analysis above to the electrodynamic case, i.e., the case with temporal and spatial dependent sources (t, x) and j(t, x), and corresponding elds E(t, x) and B(t, x), as described by Maxwells equations (1.43) on page 14. In other words, let us study the electrodynamic potentials (t, x) and A(t, x). From Equation (1.43c) on page 14 we note that also in electrodynamics the homogeneous equation B(t, x) = 0 remains valid. Because of this divergence-free nature of the time- and space-dependent magnetic eld, we can express it as the curl of an electromagnetic vector potential: B(t, x) = A(t, x) (3.6)

3.3

T HE E LECTRODYNAMIC P OTENTIALS

37

Inserting this expression into the other homogeneous Maxwell equation, Equation (1.30) on page 11, we obtain E(t, x) = [ A(t, x)] = A(t, x) t t (3.7)

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 electrostatic case, we introduce the electromagnetic scalar potential function (t, x), Equation (3.8) above becomes equivalent to E(t, x) + A(t, x) = (t, x) t (3.9)

This means that in electrodynamics, E(t, x) can be calculated from the formula E(t, x) = (t, x) A(t, x) t (3.10)

and B(t, x) from Equation (3.6) on the facing page. Hence, it is a matter of taste whether we want to express the laws of electrodynamics in terms of the potentials (t, x) and A(t, x), or in terms of the elds E(t, x) and B(t, x). However, there exists an important difference between the two approaches: in classical electrodynamics the only directly observable quantities are the elds themselves (and quantities derived from them) and not the potentials. On the other hand, the treatment becomes signicantly simpler if we use the potentials in our calculations and then, at the nal stage, use Equation (3.6) on the preceding page and Equation (3.10) to calculate the elds or physical quantities expressed in the elds. Inserting (3.10) and (3.6) on the preceding page into Maxwells equations (1.43) on page 14 we obtain, after some simple algebra and the use of Equation (1.9) on page 5, the general inhomogeneous wave equations (t, x) ( A) = t 0 1 2 A 1 2 A 2 2 A+ 2 c t c t

2

38

E LECTROMAGNETIC P OTENTIALS

2

These two second order, coupled, partial differential equations, representing in all four scalar equations (one for and one each for the three components Ai , i = 1, 2, 3 of A) are completely equivalent to the formulation of electrodynamics in terms of Maxwells equations, which represent eight scalar rstorder, coupled, partial differential equations.

As they stand, Equations (3.11) on the previous page and Equations (3.12) look complicated and may seem to be of limited use. However, if we write Equation (3.6) on page 36 in the form A(t, x) = B(t, x) we can consider this as a specication of A. But we know from Helmholtz theorem that in order to determine the (spatial) behaviour of A completely, we must also specify A. Since this divergence does not enter the derivation above, we are free to choose A in whatever way we like and still obtain the same physical results!

With a judicious choice of A, the inhomogeneous wave equations can be simplied considerably. To this end, Lorentz introduced the so called Lorentz gauge condition1 A+ 1 =0 c2 t (3.13)

because this condition simplies the system of coupled equations Equation (3.12) into the following set of uncoupled partial differential equations

fact, the Dutch physicist Hendrik Antoon Lorentz, who in 1903 demonstrated the covariance of Maxwells equations, was not the original discoverer of this condition. It had been discovered by the Danish physicist Ludwig Lorenz already in 1867.

1 In

A+

A+

1 c2 t 1 A+ 2 c t

(t, x) 0

(3.12a) (3.12b)

= 0 j(t, x)

3.3

T HE E LECTRODYNAMIC P OTENTIALS

39

where 2 is the dAlembert operator discussed in Example M.6 on page 180. We shall call (3.14) the Lorentz potential equations for the electrodynamic potentials.

Gauge transformations

We saw in Section 3.1 on page 35 and in Section 3.2 on page 36 that in electrostatics and magnetostatics we have a certain mathematical degree of freedom, up to terms of vanishing gradients and curls, to pick suitable forms for the potentials and still get the same physical result. In fact, the way the electromagnetic scalar potential (t, x) and the vector potential A(t, x) are related to the physically observables gives leeway for similar manipulation of them also in electrodynamics. If we transform (t, x) and A(t, x) simultaneously into new ones (t, x) and A (t, x) according to the mapping scheme (t, x) t A(t, x) A (t, x) = A(t, x) (t, x) (t, x) (t, x) = (t, x) + (3.15a) (3.15b)

where (t, x) is an arbitrary, differentiable scalar function called the gauge function, and insert the transformed potentials into Equation (3.10) on page 37 for the electric eld and into Equation (3.6) on page 36 for the magnetic eld, we obtain the transformed elds E = B = A ( ) A ( ) A = + = t t t t t A = A ( ) = A (3.16a) (3.16b)

where, once again Equation (M.90) on page 184 was used. Comparing these expressions with (3.10) and (3.6) we see that the elds are unaffected by the gauge transformation (3.15). A transformation of the potentials and A which leaves the elds, and hence Maxwells equations, invariant is called a gauge transformation. A physical law which does not change under a gauge transformation is said to be gauge invariant. By denition, the elds themselves are, of course, gauge invariant.

def

def

1 2 c2 t2 1 2 c2 t2

1 2 (t, x) 2 = c2 t2 0 2A 1 A = 2 2 2 A = 0 j(t, x) c t =

S S

(3.14a) (3.14b)

40

E LECTROMAGNETIC P OTENTIALS

The potentials (t, x) and A(t, x) calculated from (3.11) on page 37, with an arbitrary choice of A, can be further gauge transformed according to (3.15) on the preceding page. If, in particular, we choose A according to the Lorentz condition, Equation (3.13) on page 38, and apply the gauge transformation (3.15) on the resulting Lorentz potential equations (3.14) on the preceding page, these equations will be transformed into

We notice that if we require that the gauge function (t, x) itself be restricted to full the wave equation 1 2 c2 t2

2

=0

these transformed Lorentz equations will keep their original form. The set of potentials which have been gauge transformed according to Equation (3.15) on the previous page with a gauge function (t, x) which is restricted to full Equation (3.18), i.e., those gauge transformed potentials for which the Lorentz equations (3.14) are invariant, comprises the Lorentz gauge. Other useful gauges are The radiation gauge, also known as the transverse gauge, dened by A = 0. The Coulomb gauge, dened by = 0, A = 0. The temporal gauge, also known as the Hamilton gauge, dened by = 0. The axial gauge, dened by A3 = 0. The process of choosing a particular gauge condition is referred to as gauge xing.

As we see, the Lorentz equations (3.14) on the previous page for (t, x) and A(t, x) represent a set of uncoupled equations involving four scalar equations

= 0 j(t, x)

1 2 c2 t2 1 2 A c2 t2

1 2 c2 t2 1 2 c2 t2

(t, x) 0

(3.17a) (3.17b)

(3.18)

3.3

T HE E LECTRODYNAMIC P OTENTIALS

41

(one equation for and one equation for each of the three components of A). Each of these four scalar equations is an inhomogeneous wave equation of the following generic form:

where is a shorthand for either or one of the vector components of A, and f is the pertinent generic source component. We assume that our sources are well-behaved enough in time t so that the Fourier transform pair for the generic source function F 1 [ f (x)] f (t, x) = F [ f (t, x)] f (x) =

def def

exists, and that the same is true for the generic potential component: (t, x) = (x) =

Inserting the Fourier representations (3.20a) and (3.21a) into Equation (3.19), and using the vacuum dispersion relation for electromagnetic waves = ck (3.22)

the generic 3D inhomogeneous wave equation, Equation (3.19) above, turns into

2

which is a 3D inhomogeneous time-independent wave equation, often called the 3D inhomogeneous Helmholtz equation. As postulated by Huygens principle, each point on a wave front acts as a point source for spherical waves which form a new wave from a superposition of the individual waves from each of the point sources on the old wave front. The solution of (3.23) can therefore be expressed as a superposition of solutions of an equation where the source term has been replaced by a point source:

2

G(x, x ) + k2G(x, x ) = (x x )

1 2

(x) eit d

(t, x) eit dt

1 2

(t, x) = f (t, x)

(3.19)

(3.20a) (3.20b)

(3.21a) (3.21b)

(3.23)

(3.24)

42

E LECTROMAGNETIC P OTENTIALS

and the solution of Equation (3.23) on the preceding page which corresponds to the frequency is given by the superposition (x) = f (x )G(x, x ) d3x (3.25)

The function G(x, x ) is called the Green function or the propagator. In Equation (3.24) on the previous page, the Dirac generalised function (x x ), which represents the point source, depends only on x x and there is no angular dependence in the equation. Hence, the solution can only be dependent on r = |x x | and not on the direction of x x . If we interpret r as the radial coordinate in a spherically polar coordinate system, and recall the expression for the Laplace operator in such a coordinate system, Equation (3.24) on the preceding page becomes d2 (rG) + k2 (rG) = r(r) dr2 (3.26)

Away from r = |x x | = 0, i.e., away from the source point x , this equation takes the form d2 (rG) + k2 (rG) = 0 dr2 with the well-known general solution G = C+ eikr eik|xx | eikr eik|xx | + C + C C+ r r |x x | |x x | (3.28) (3.27)

where C are constants. In order to evaluate the constants C , we insert the general solution, Equation (3.28) above, into Equation (3.24) on the preceding page and integrate over a small volume around r = |x x | = 0. Since G( x x ) C + 1 1 + C , |x x | |x x | xx 0

The volume integrated Equation (3.24) on the previous page can under this assumption be approximated by

( x x ) d x

C+ + C

1 d3x + k2 C + + C |x x |

(3.29)

1 d3x |x x |

(3.30)

3.3

T HE E LECTRODYNAMIC P OTENTIALS

43

In virtue of the fact that the volume element d3x in spherical polar coordinates is proportional to |x x |2 , the second integral vanishes when |x x | 0. Furthermore, from Equation (M.85) on page 183, we nd that the integrand in the rst integral can be written as 4(|x x |) and, hence, that C+ + C = 1 4 (3.31)

Insertion of the general solution Equation (3.28) on the facing page into Equation (3.25) on the preceding page gives (x) = C + f (x ) eik|xx | 3 d x + C |x x | f (x ) eik|xx | 3 dx |x x | (3.32)

The Fourier transform to ordinary t domain of this is obtained by inserting the above expression for (x) into Equation (3.21a) on page 41: (t, x) = C

+

f (x )

+ C

f (x )

|x x |

If we introduce the retarded time tret and the advanced time tadv in the following way [using the fact that in vacuum k/ = 1/c, according to Equation (3.22) on page 41]: k |x x | |x x | = t c k |x x | |x x | = t+ tadv = tadv (t, x x ) = t + c tret = tret (t, x x ) = t and use Equation (3.20a) on page 41, we obtain (t, x) = C + f (tret , x ) 3 d x + C |x x | f (tadv , x ) 3 dx |x x | (3.35)

This is a solution to the generic inhomogeneous wave equation for the potential components Equation (3.19) on page 41. We note that the solution at time t at the eld point x is dependent on the behaviour at other times t of the source at x and that both retarded and advanced t are mathematically acceptable solutions. However, if we assume that causality requires that the potential at (t, x) is set up by the source at an earlier time, i.e., at (tret , x ), we must in Equation (3.35) set C = 0 and therefore, according to Equation (3.31) above, C + = 1/(4).

V WG

exp i t + k|xx |

|x x |

V WG

exp i t k|xx |

(3.34a) (3.34b)

44

E LECTROMAGNETIC P OTENTIALS

From the above discussion on the solution of the inhomogeneous wave equation we conclude that, under the assumption of causality, the electrodynamic potentials in vacuum can be written (tret , x ) 3 1 dx 40 |x x | 0 j(tret , x ) 3 A(t, x) = dx 4 |x x | (t, x) = (3.36a) (3.36b)

Since these retarded potentials were obtained as solutions to the Lorentz equations (3.14) on page 39 they are valid in the Lorentz gauge but may be gauge transformed according to the scheme described in subsection 3.3.1 on page 39. As they stand, we shall use them frequently in the following.

E XAMPLE 3.1

E LECTROMAGNETODYNAMIC POTENTIALS

In Diracs symmetrised form of electrodynamics (electromagnetodynamics), Maxwells equations are replaced by [see also Equations (1.48) on page 15]: E = e 0 B t E t (3.37a) (3.37b) (3.37c) (3.37d)

E = 0 jm B = 0

m

B = 0 j e + 0 0

In this theory, one derives the inhomogeneous wave equations for the usual electric scalar and vector potentials (e , Ae ) and their magnetic counterparts (m , Am ) by assuming that the potentials are related to the elds in the following symmetrised form: E = e (t, x) B= 1 c2 e A (t, x) Am t 1 m (t, x) 2 Am (t, x) + Ae c t (3.38a) (3.38b)

In the absence of magnetic charges, or, equivalenty for m 0 and Am 0, these formulae reduce to the usual Maxwell theory Formula (3.10) on page 37 and Formula (3.6) on page 36, respectively, as they should. Inserting the symmetrised expressions (3.38) above into Equations (3.37), one obtains [cf., Equations (3.11) on page 37]

3.3

B IBLIOGRAPHY

45

By choosing the conditions on the vector potentials according to Lorentz prescripton [cf., Equation (3.13) on page 38] 1 e =0 c2 t 1 A m + 2 m = 0 c t these coupled wave equations simplify to Ae + (3.40) (3.41)

e (t, x) 1 2 e 2 e = (3.42a) c2 t2 0 1 2 m m (t, x) 2 m = (3.42b) c2 t2 0 1 2 Ae 2 Ae = 0 je (t, x) (3.42c) c2 t2 1 2 Am 2 Am = 0 jm (t, x) (3.42d) c2 t2 exhibiting once again, the striking properties of Diracs symmetrised Maxwell theory.

E ND OF

EXAMPLE

Bibliography

[1] 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. Benjamin/Cummings Publishing Company, Inc., Reading, MA . . . , 1980, ISBN 08053-9016-2. [2] M. G UIDRY, Gauge Field Theories: An Introduction with Applications, John Wiley & Sons, Inc., New York, NY . . . , 1991, ISBN 0-471-63117-5. [3] J. D. JACKSON, Classical Electrodynamics, third ed., John Wiley & Sons, Inc., New York, NY . . . , 1999, ISBN 0-471-30932-X.

"

1 2 Am c2 t2

Am +

Am +

1 m c2 t

"

e (t, x) Ae = t 0 m (t, x) 2 m Am = + t 0 1 2 Ae 1 e 2 Ae + Ae + 2 c2 t2 c t

$ $

2 e

3.1

46

E LECTROMAGNETIC P OTENTIALS

[4] 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.

4

Relativistic Electrodynamics

We saw in Chapter 3 how the derivation of the electrodynamic potentials led, in a most natural way, to the introduction of a characteristic, nite speed of propagation in vacuum that equals the speed of light c = 1/ 0 0 and which can be considered as a constant of nature. To take this nite speed of propagation of information into account, and to ensure that our laws of physics be independent of any specic coordinate frame, requires a treatment of electrodynamics in a relativistically covariant (coordinate independent) form. This is the object of this chapter.

An inertial system, or inertial reference frame, is a system of reference, or rigid coordinate system, in which the law of inertia (Galileos law, Newtons rst 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 relativity1 describes how physical processes are interrelated when observed in different inertial systems in uniform, rectilinear motion relative to each other and is based on two postulates:

1 The Special Theory of Relativity, by the American physicist and philosopher David Bohm, opens with the following paragraph [4]:

The theory of relativity is not merely a scientic development of great importance in its own right. It is even more signicant as the rst stage of a radical change in our basic concepts, which began in physics, and which is spreading into other elds of science, and indeed, even into a great deal of thinking outside of science. For as is well known, the modern trend is away from the notion of sure absolute truth, (i.e., one which holds independently of all conditions, contexts, degrees, and types of approximation etc..) and toward the idea that a given concept has signicance only in relation to suitable broader forms of reference, within which that concept can be given its full meaning.

47

48

R ELATIVISTIC E LECTRODYNAMICS

vt y y v P(t, x, y, z) P(t , x , y , z ) O z x z O x

F IGURE 4.1: Two inertial systems and in relative motion with velocity v along the x = x axis. At time t = t = 0 the origin O of coincided with the origin O of . At time t, the inertial system has been translated a distance vt along the x axis in . An event represented by P(t, x, y, z) in is represented by P(t , x , y , z ) in .

Postulate 4.1 (Relativity principle; Poincar, 1905). All laws of physics (except the laws of gravitation) are independent of the uniform translational motion of the system on which they operate. Postulate 4.2 (Einstein, 1905). The velocity of light in empty space is independent of the motion of the source that emits the light. A consequence of the rst postulate is that all geometrical objects (vectors, tensors) in an equation describing a physical process must transform in a covariant manner, i.e., in the same way.

Let us consider two three-dimensional inertial systems and in vacuum which are in rectilinear motion relative to each other in such a way that moves with constant velocity v along the x axis of the system. The times and the spatial coordinates as measured in the two systems are t and (x, y, z), and t and (x , y , z ), respectively. At time t = t = 0 the origins O and O and the x and x axes of the two inertial systems coincide and at a later time t they have the relative location as depicted in Figure 4.1. For convenience, let us introduce the two quantities v = (4.1) c 1 = (4.2) 1 2

4.1

T HE S PECIAL T HEORY

OF

R ELATIVITY

49

where v = |v|. In the following, we shall make frequent use of these shorthand notations. As shown by Einstein, the two postulates of special relativity require that the spatial coordinates and times as measured by an observer in and , respectively, are connected by the following transformation: ct = (ct x) x = (x vt) y =y z =z (4.3a) (4.3b) (4.3c) (4.3d)

Taking the difference between the square of (4.3a) and the square of (4.3b) we nd that c2 t 2 x 2 = 2 c2 t2 2xct + x2 2 x2 + 2xvt v2 t2 1

From Equations (4.3) we see that the y and z coordinates are unaffected by the translational motion of the inertial system along the x axis of system . Using this fact, we nd that we can generalise the result in Equation (4.4) above to c2 t 2 x 2 y2 z2 = c2 t 2 x 2 y 2 z 2 (4.5)

which means that if a light wave is transmitted from the coinciding origins O and O at time t = t = 0 it will arrive at an observer at (x, y, z) at time t in and an observer at (x , y , z ) at time t in in such a way that both observers conclude that the speed (spatial distance divided by time) of light in vacuum is c. Hence, the speed of light in and is the same. A linear coordinate transformation which has this property is called a (homogeneous) Lorentz transformation.

Let us introduce an ordered quadruple of real numbers, enumerated with the help of upper indices = 0, 1, 2, 3, where the zeroth component is ct (c is the

v2 c2 2 2 = c t x2

c2 t 2 1

v2 c2

x2 1

v2 c2

R

(4.4)

Downloaded from http://www.plasma.uu.se/CED/Book

50

R ELATIVISTIC E LECTRODYNAMICS

speed of light and t is time), and the remaining components are the components of the ordinary 3 radius vector x dened in Equation (M.1) on page 168: x = (x0 , x1 , x2 , x3 ) = (ct, x, y, z) (ct, x)

We want to interpret this quadruple x as (the component form of) a radius four-vector in a real, linear, four-dimensional vector space.1 We require that this four-dimensional space to be a Riemannian space, i.e., a space where a distance and a scalar product are dened. In this space we therefore dene a metric tensor, also known as the fundamental tensor, which we denote by g .

The radius four-vector x = (x0 , x1 , x2 , x3 ) = (ct, x), as dened in Equation (4.6) above, is, by denition, the prototype of a contravariant vector (or, more accurately, a vector in contravariant component form). To every such vector there exists a dual vector. The vector dual to x is the covariant vector x , obtained as (the upper index in x is summed over and is therefore a dummy index and may be replaced by another dummy index ): x = g x (4.7)

This summation process is an example of index contraction and is often referred to as index lowering.

The scalar product of x with itself in a Riemann space is dened as g x x = x x (4.8)

This scalar product acts as an invariant distance, or norm, in this space. If we want the Lorentz transformation invariance, described by Equation (4.5) on the preceding page, to be the manifestation of the conservation

1 The British mathematician and philosopher Alfred North Whitehead writes in his book The

Concept of Nature [13]: I regret that it has been necessary for me in this lecture to administer a large dose of four-dimensional geometry. I do not apologise, because I am really not responsible for the fact that nature in its most fundamental aspect is fourdimensional. Things are what they are. . .

(4.6)

4.1

T HE S PECIAL T HEORY

OF

R ELATIVITY

51

of the norm in a 4D Riemann space, then the explicit expression for the scalar product of x with itself in this space must be x x = c2 t 2 x 2 y2 z2 (4.9)

We notice that our space will have an indenite norm which means that we deal with a non-Euclidean space. We call the four-dimensional space (or spacetime) with this property Lorentz space and denote it 4 . A corresponding real, linear 4D space with a positive denite norm which is conserved during ordinary rotations is a Euclidean vector space. We denote such a space 4 .

Metric tensor

i.e., a matrix with a main diagonal that has the sign sequence, or signature, {+, , , }, the index lowering operation in our chosen at 4D space becomes nearly trivial: x = g x = (ct, x) Using matrix algebra, this can be written x0 x1 x2 x3 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 x0 x1 x2 x3 x0 x1 x2 x3 (4.12)

Hence, if the metric tensor is dened according to expression (4.10) above the covariant radius four-vector x is obtained from the contravariant radius fourvector x simply by changing the sign of the last three components. These

deef

deef p

deef

hh ig

hh ig

hh ig

deef

(g ) =

hh ig

`a

g =

1 if = = 0 1 if = = i = j = 1, 2, 3 0 if =

as

ac

(4.10)

(4.11)

hh ig p

(4.13)

deef

52

R ELATIVISTIC E LECTRODYNAMICS

components are referred to as the space components; the zeroth 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 x x = (ct, x) (ct, x) = c2 t2 x2 y2 z2 (4.14)

which indeed is the desired Lorentz transformation invariance as required by Equation (4.9) on the preceding page. Without changing the physics, one can alternatively choose a signature {, +, +, +}. The latter has the advantage that the transition from 3D to 4D becomes smooth, while it will introduce some annoying minus signs in the theory. In current physics literature, the signature {+, , , } seems to be the most commonly used one. The 4 metric tensor Equation (4.10) on the previous page has a number of interesting properties: rstly, we see that this tensor has a trace Tr g = 2 whereas in 4 , as in any vector space with denite norm, the trace equals the space dimensionality. Secondly, we nd, after trivial algebra, that the following relations between the contravariant, covariant and mixed forms of the metric tensor hold: g = g g = g g g = g g =

(4.15a) (4.15b)

g g

= =

(4.15c) (4.15d)

Here we have introduced the 4D version of the Kronecker delta , a mixed four-tensor of rank 2 which fulls

The differential distance ds between the two points x and x + dx in 4 can be calculated from the Riemannian metric, given by the quadratic differential form ds2 = g dx dx = dx dx = (dx0 )2 (dx1 )2 (dx2 )2 (dx3 )2

where the metric tensor is as in Equation (4.10) on the preceding page. As we see, this form is indenite as expected for a non-Euclidean space. The square

= =

1 0

if = if =

(4.16)

(4.17)

4.1

T HE S PECIAL T HEORY

OF

R ELATIVITY

53

2

Since d measures the time when no spatial changes are present, it is called the proper time. Expressing Equation (4.5) on page 49 in terms of the differential interval ds and comparing with Equation (4.17) on the facing page, we nd that ds2 = c2 dt2 dx2 dy2 dz2 (4.20)

is invariant during a Lorentz transformation. Conversely, we may say that every coordinate transformation which preserves this differential interval is a Lorentz transformation. If in some inertial system dx2 + dy2 + dz2 < c2 dt2 ds is a time-like interval, but if dx2 + dy2 + dz2 > c2 dt2 ds is a space-like interval, whereas dx2 + dy2 + dz2 = c2 dt2 (4.23) (4.22) (4.21)

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 is invariant under a Lorentz transformation.

=c

1 (v x )2 + (vy )2 + (vz )2 dt = c c2 c = c 1 2 dt = dt = c d

ds = c

dx 2 + dt

dx 3 + dt

CDD E

dt v2 dt c2 (4.18)

54

R ELATIVISTIC E LECTRODYNAMICS

Four-vector elds

Any quantity which relative to any coordinate system has a quadruple of real numbers and transforms in the same way as the radius four-vector x does, is called a four-vector. In analogy with the notation for the radius four-vector we introduce the notation a = (a0 , a) for a general contravariant four-vector eld in 4 and nd that the lowering of index rule, Equation (M.32) on page 174, for such an arbitrary four-vector yields the dual covariant four-vector eld

The scalar product between this four-vector eld and another one b (x ) is g a (x )b (x ) = (a0 , a) (b0 , b) = a0 b0 a b (4.25)

which is a scalar eld, i.e., an invariant scalar quantity (x ) which depends on time and space, as described by x = (ct, x, y, z).

Introducing the transformation matrix 0 0 0 0 0 0 1 0 0 0 0 1

the linear Lorentz transformation (4.3) on page 49, i.e., the coordinate transformation x x = x (x0 , x1 , x2 , x3 ), from one inertial system to another inertial system , can be written x = x The inverse transform then takes the form x = (1 ) x (4.28) (4.27)

It is easy to show, by means of direct algebra, that two successive Lorentz transformations of the type in Equation (4.28) above, and dened by the speed parameters 1 and 2 , respectively, correspond to a single transformation with speed parameter = 1 + 2 1 + 1 2 (4.29)

deef

hh ig

a (x ) = g a (x ) = (a0 (x ), a(x ))

(4.24)

(4.26)

4.1

T HE S PECIAL T HEORY

OF

R ELATIVITY

55

X0 X

0

x1 x1 F IGURE 4.2: Minkowski space can be considered an ordinary Euclidean space where a Lorentz transformation from (x1 , X 0 = ict) to (x 1 , X 0 = ict ) corresponds to an ordinary rotation through an angle . This rotation 2 2 leaves the Euclidean distance x1 + X 0 = x2 c2 t2 invariant.

This means that the nonempty set of Lorentz transformations constitutes a closed algebraic structure with a binary operation which is associative. 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 transformations constitutes a mathematical group. However tempting, we shall not make any further use of group theory.

Specifying a point x = (x0 , x1 , x2 , x3 ) in 4D space-time is a way of saying that something takes place at a certain time t = x0 /c and at a certain place (x, y, z) = (x1 , x2 , x3 ). Such a point is therefore called an event. The trajectory for an event as a function of time and space is called a world line. For instance, the world line for a light ray which propagates in vacuum is the trajectory x0 = x1 . Introducing X 0 = ix0 = ict X 1 = x1 X =x X =x

3 2 2 3

dS = ids

$ #

$ #

56

R ELATIVISTIC E LECTRODYNAMICS

x0 = ct

w x0 x0 = x 1

P O=O P ct

x1

x1 = x

F IGURE 4.3: Minkowski diagram depicting geometrically the transformation (4.34) from the unprimed system to the primed system. Here w denotes the world line for an event and the line x0 = x1 x = ct the world line for a light ray in vacuum. Note that the event P is simultaneous with all points on the x1 axis (t = 0), including the origin O while the event P , which is also simultaneous with all points on the x axis, including O = O, to an observer at rest in the primed system, is not simultaneous with O in the unprimed system but occurs there at time |P P | /c.

where i = 1, we see that Equation (4.17) on page 52 transforms into dS 2 = (dX 0 )2 + (dX 1 )2 + (dX 2 )2 + (dX 3 )2 (4.31) i.e., into a 4D differential form which is positive denite just as is ordinary 3D Euclidean space 3 . We shall call the 4D Euclidean space constructed in this way the Minkowski space 4 .1 As before, it sufces to consider the simplied case where the relative motion between and is along the x axes. Then dS 2 = (dX 0 )2 + (dx1 )2

and we consider X 0 and x1 as orthogonal axes in an Euclidean space. As in all Euclidean spaces, every interval is invariant under a rotation of the X 0 x1 plane through an angle into X 0 x 1 : X 0 = x1 sin + X 0 cos x 1 = x1 cos + X 0 sin

1 The

fact that our Riemannian space can be transformed in this way into an Euclidean one means that it is, strictly speaking, a pseudo-Riemannian space.

(4.32)

(4.33a) (4.33b)

4.2

57

See Figure 4.2 on page 55. If we introduce the angle = i, often called the rapidity or the Lorentz boost parameter, and transform back to the original space and time variables by using Equation (4.30) on page 55 backwards, we obtain ct = x sinh + ct cosh x = x cosh ct sinh (4.34a) (4.34b)

which are identical to the transformation equations (4.3) on page 49 if we let sinh = cosh = tanh = (4.35a) (4.35b) (4.35c)

It is therefore possible to envisage the Lorentz transformation as an ordinary rotation in the 4D Euclidean space 4 This rotation i 4 corresponds to a coordinate change in 4 as depicted in Figure 4.3 on the facing page. Equation (4.29) on page 54 for successive Lorentz transformation then corresponds to the tanh addition formula tanh 1 + tanh 2 (4.36) tanh(1 + 2 ) = 1 + tanh 1 tanh 2 The use of ict and 4 , which leads to the interpretation of the Lorentz transformation as an ordinary rotation, may, at best, be illustrative, but is not very physical. Besides, if we leave the at 4 space and enter the curved space of general relativity, the ict trick will turn out to be an impasse. Let us therefore immediately return to 4 where all components are real valued.

The invariance of the differential distance ds in 4 , and the associated differential proper time d [see Equation (4.18) on page 53] allows us to dene the four-velocity

v2 c2

which, when multiplied with the scalar invariant m0 yields the four-momentum

v2 c2

v2 c2

u fd

p = m0

dx = m0 (c, v) = d

m0 c

m0 v

v2 c2

u fd

u =

dx = (c, v) = d

= (u0 , u)

= (p0 , p)

(4.37)

(4.38)

58

R ELATIVISTIC E LECTRODYNAMICS

We can interpret this such that the Lorentz covariance implies that the masslike term in the ordinary 3D linear momentum is not invariant. A better way to look at this is that p = mv = m0 v is the covariantly correct expression for the kinetic three-momentum. Multiplying the zeroth (time) component of the four-momentum p with the scalar invariant c, we obtain

Since this component has the dimension of energy and is the result of a covariant description of the motion of a particle with its kinetic momentum described by the spatial components of the four-momentum, Equation (4.38) on the previous page, we interpret cp0 as the total energy E. Hence, cp = (cp0 , cp) = (E, cp) Scalar multiplying this four-vector with itself, we obtain cp cp = c2 g p p = c2 [(p0 )2 (p1 )2 (p2 )2 (p3 )2 ] = (E, cp) (E, cp) = E 2 c2 p2 (4.42)

Since this is an invariant, this equation holds in any inertial frame, particularly in the frame where p = 0 and there we have E = m 0 c2 This is probably the most famous formula in physics history. (4.44)

1 v2 c

(m0 c2 )2

cp0 = m0 c2 =

m = m0 =

m0 1 v2 c

2

(4.40)

m0 c2 1

v2 c2

= mc2

(4.41)

v2 1 2 c

(4.43)

= (m0 c )

2 2

4.3

59

In the rest inertial system the charge density is 0 . The four-vector (in contravariant component form) j = 0 dx = 0 u = 0 (c, v) = (c, v) d (4.45)

is called the four-current. The contravariant form of the four-del operator = /x is dened in Equation (M.69) on page 180 and its covariant counterpart = /x in Equation (M.70) on page 180, respectively. As is shown in Example M.6 on page 180, the dAlembert operator is the scalar product of the four-del with itself:

2

= = =

1 2 c2 t2

Since it has the characteristics of a four-scalar, the dAlembert operator is invariant and, hence, the homogeneous wave equation is Lorentz covariant.

If we introduce the four-potential A = ,A c

where is the scalar potential and A the vector potential, dened in Section 3.3 on page 36, we can write the inhomogeneous wave equations (Lorentz potential equations), Equations (3.14) on page 39, in the following compact (and covariant) way:

A = 0 j

With the help of the above, we can formulate our electrodynamic equations covariantly. For instance, the covariant form of the equation of continuity, Equation (1.21) on page 9 is j = 0 (4.50)

S S

(4.47)

(4.48)

(4.49)

60

R ELATIVISTIC E LECTRODYNAMICS

and the Lorentz gauge condition, Equation (3.13) on page 38, can be written A = 0 The gauge transformations (3.15) on page 39 in covariant form are A A = A + (x ) (4.52) (4.51)

If only one dimension Lorentz contracts (for instance, due to relative motion along the x direction), a 3D spatial volume transforms according to 1 dV = d3x = dV0 = dV0

then from Equation (4.46) on the preceding page we see that dV = 0 dV0 (4.54)

i.e., the charge in a given volume is conserved. We can therefore conclude that the elementary charge is a universal constant.

Let us now solve the Lorentz equation (the inhomogeneous wave equation) (3.14) on page 39 in vacuum for the case of a well-localised charge q at a source point dened by the radius four-vector x = (x 0 = ct , x 1 , x 2 , x 3 ). The eld point (observation point) is denoted by the radius four-vector x = (x0 = ct, x1 , x2 , x3 ). In the rest system we know that the solution is simply

0

v=0

where |x x |0 is the usual distance from the source point to the eld point, evaluated in the rest system (signied by the index 0). Let us introduce the relative radius four-vector between the source point and the eld point: R = x x = (c(t t ), x x ) Scalar multiplying this relative four-vector with itself, we obtain R R = (c(t t ), x x ) (c(t t ), (x x )) = c2 (t t )2 x x (4.56)

,A c

1 q ,0 40 c |x x |0

1 2 = dV0

v2 c2

(4.53)

(4.55)

(4.57)

4.3

61

We know that in vacuum the signal (eld) from the charge q at x propagates to x with the speed of light c so that x x = c(t t )

Inserting this into Equation (4.57) on the facing page, we see that R R = 0 or that Equation (4.56) on the preceding page can be written R = ( x x , x x ) (4.59)

Now we want to nd the correspondence to the rest system solution, Equation (4.55) on the facing page, in an arbitrary inertial system. We note from Equation (4.37) on page 57 that in the rest system

v2 c2

and (R )0 = ( x x , x x )0 = ( x x

As all scalar products, u R is invariant, which means that we can evaluate it in any inertial system and it will have the same value in all other inertial systems. If we evaluate it in the rest system the result is:

= (c, 0) ( x x q u 40 cu R

, (x x )0 ) = c x x 0

subject to the condition R R = 0 has the proper transformation properties (proper tensor form) and reduces, in the rest system, to the solution Equation (4.55) on the facing page. It is therefore the correct solution, valid in any inertial system. According to Equation (4.37) on page 57 and Equation (4.60)

u R = (c, v) x x , (x x ) = c x x v (x x )

u R = u R

= (u )0 (R )0

u fd

v2 c2

(4.58)

(4.60)

= (c, 0)

v=0

(4.61)

, (x x )0 )

(4.62)

(4.63)

0

(4.65)

62

R ELATIVISTIC E LECTRODYNAMICS

Generalising expression (4.1) on page 48 to vector form: = v and introducing s xx we can write u R = cs and (4.68)

def def

v c v (x x ) x x (x x ) c

(4.66)

where in the last step the denition of the four-potential, Equation (4.48) on page 59, 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 velocity v, the scalar and vector potentials are given by the expressions 1 q 1 q = 40 s 40 |x x | (x x ) q v v q = A(t, x) = 2 s 2 |x x | (x x ) 40 c 40 c (t, x) = These potentials are called the Linard-Wiechert potentials. (4.71a) (4.71b)

Consider a vectorial (cross) product c between two ordinary vectors a and b: c = ab = i jk ai b j xk = (a2 b3 a3 b2 ) x1 + (a3 b1 a1 b3 ) x2 + (a1 b2 a2 b1 ) x3

We notice that the kth component of the vector c can be represented as ck = ai b j a j bi = ci j = c ji , i, j = k (4.73)

A (x ) =

q 40

1 v , cs c2 s

u = cu R

1 v , cs c2 s

,A c

(4.67)

(4.69)

(4.70)

(4.72)

4.3

63

In other words, the pseudovector c = a b can be considered as an antisymmetric tensor of rank two. The same is true for the curl operator . For instance, the Maxwell equation B (4.74) E = t can in this tensor notation be written E j Ei Bi j = (4.75) xi x j t We know from Chapter 3 that the elds can be derived from the electromagnetic potentials in the following way: B= A t In component form, this can be written E = Bi j = A (4.76a) (4.76b)

A j Ai = i A j j Ai (4.77a) xi x j Ai Ei = i = i t Ai (4.77b) x t 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 elds in a tensor form where the components are functions of the covariant form of the four-potential, Equation (4.48) on page 59:

Inspection of (4.78) and Equation (4.77) makes it natural to dene the fourtensor A A F = = A A (4.79) x x This anti-symmetric (skew-symmetric), four-tensor of rank 2 is called the electromagnetic eld tensor. In matrix representation, the contravariant eld tensor can be written F = 0 E x /c Ey /c Ez /c E x /c 0 Bz By Ey /c Bz 0 B x Ez /c By Bx 0

hh ig

deef

A =

,A c

(4.78)

(4.80)

64

R ELATIVISTIC E LECTRODYNAMICS

The covariant eld tensor is obtained from the contravariant eld tensor in the usual manner by index contraction (index lowering): F = g g F = A A (4.81)

It is perhaps interesting to note that the eld tensor is a sort of four-dimensional curl of the four-potential vector A . The matrix representation for the covariant eld tensor is 0 E x /c Ey /c Ez /c E x /c 0 Bz By Ey /c Bz 0 B x Ez /c By Bx 0

is immediately observed by explicitly setting = 0 in this covariant equation and using the matrix representation Formula (4.80) on the previous page for the covariant component form of the electromagnetic eld tensor F , to obtain

which is the Maxwell source equation for the electric eld, Equation (1.43a) on page 14. For = 1, Equation (4.84) above yields F 01 F 11 F 21 F 31 Bz By 1 E x +0 + = 0 j1 = 0 v x (4.86) + 1 + 2 + 3 = 2 0 x x x x c t y z or, using 0 0 = 1/c2 , By Bz E x 0 0 = 0 j x z y t (4.87)

1 F 00 F 10 F 20 F 30 + + + = 0+ 0 1 2 3 x x x x c 1 = E = 0 j0 = 0 c c

p

E x Ey Ez + + x y z

deef

F =

hh ig

(4.82)

(4.84)

4.3

B IBLIOGRAPHY

65

and similarly for = 2, 3. In summary, in three-vector form, we can write the result as B 0 0 E = 0 j(t, x) t (4.88)

which is the Maxwell source equation for the magnetic eld, Equation (1.43d) on page 14. The two Maxwell eld equations E = B = 0 correspond to (no summation!) F + F + F = 0 (4.91) B t (4.89) (4.90)

Hence, Equation (4.83) on the preceding page and Equation (4.91) constitute Maxwells equations in four-dimensional formalism.

Bibliography

[1] J. A HARONI, The Special Theory of Relativity, second, revised ed., Dover Publications, Inc., New York, 1985, ISBN 0-486-64870-2. [2] A. O. BARUT, Electrodynamics and Classical Theory of Fields and Particles, Dover Publications, Inc., New York, NY, 1980, ISBN 0-486-64038-8. [3] R. B ECKER, Electromagnetic Fields and Interactions, Dover Publications, Inc., New York, NY, 1982, ISBN 0-486-64290-9. [4] D. B OHM, The Special Theory of Relativity, Routledge, New York, NY, 1996, ISBN 0-415-14809-X. [5] W. T. G RANDY, Introduction to Electrodynamics and Radiation, Academic Press, New York and London, 1970, ISBN 0-12-295250-2. [6] 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. [7] F. E. L OW, Classical Field Theory, John Wiley & Sons, Inc., New York, NY . . . , 1997, ISBN 0-471-59551-9.

66

R ELATIVISTIC E LECTRODYNAMICS

[8] C. M LLER, The Theory of Relativity, second ed., Oxford University Press, Glasgow . . . , 1972. [9] H. M UIRHEAD, The Special Theory of Relativity, The Macmillan Press Ltd., London, Beccles and Colchester, 1973, ISBN 333-12845-1. [10] 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. [11] J. J. S AKURAI, Advanced Quantum Mechanics, Addison-Wesley Publishing Company, Inc., Reading, MA . . . , 1967, ISBN 0-201-06710-2. [12] B. S PAIN, Tensor Calculus, third ed., Oliver and Boyd, Ltd., Edinburgh and London, 1965, ISBN 05-001331-9. [13] A. N. W HITEHEAD, Concept of Nature, Cambridge University Press, Cambridge . . . , 1920, ISBN 0-521-09245-0.

5

Electromagnetic Fields and Particles

In previous chapters, we calculated the electromagnetic elds and potentials from arbitrary, but prescribed distributions of charges and currents. In this chapter we study the general problem of interaction between electric and magnetic elds and electrically charged particles. The analysis is based on Lagrangian and Hamiltonian methods, is fully covariant, and yields results which are relativistically correct.

We rst establish a relativistically correct theory describing the motion of charged particles in prescribed electric and magnetic elds. From these equations we may then calculate the charged particle dynamics in the most general case.

We will show that for our problem we can derive the correct equations of motion by using in 4D 4 a function with similiar properties as a Lagrange function in 3D and then apply a variational principle. We will also show that we can nd nd a Hamiltonian-type function in 4D and solve the corresponding Hamilton-type equations to obtain the correct covariant formulation of classical electrodynamics.

67

68

E LECTROMAGNETIC F IELDS

AND

PARTICLES

Lagrange formalism

Let us no introduce a function L(4) which fulls the variational principle

1 0

L(4) (x , u ) d = 0

(5.1)

where d is the proper time dened via Equation (4.18) on page 53, and the endpoints are xed. We shall show that L(4) acts as a kind of generalisation to the common 3D Lagrangian. We require that L(4) fulls the following conditions: 1. The Lagrange function must be invariant. This implies that L(4) must be a scalar. 2. The Lagrange function must yield linear equations of motion. This implies that L(4) must not contain higher than the second power of the fourvelocity u . According to Formula (M.96) on page 185 the ordinary 3D Lagrangian is the difference between the kinetic and potential energies. A free particle has only kinetic energy. If the particle mass is m0 then in 3D the kinetic energy is m0 v2 /2. This suggests that in 4D the Lagrangian for a free particle should be 1 free L(4) = m0 u u 2 (5.2)

For an interaction with the electromagnetic eld we can introduce the interaction with the help of the four-potential given by Equation (4.78) on page 63 in the following way 1 L(4) = m0 u u + qu A (x ) 2 (5.3)

We call this the four-Lagrangian and shall now show how this function, together with the variation principle, Formula (5.1), yields covariant results which are physically correct. The variation principle (5.1) with the 4D Lagrangian (5.3) inserted, leads to

1 0 1 0

R

m0 u u + q A u + u A x

d = 0

L(4) (x , u ) d =

m0 u u + qu A d 2 0 A m0 (u u ) u + q A u + u x 2 u x

(5.4)

5.1

C HARGED PARTICLES

IN AN

E LECTROMAGNETIC F IELD

69

which means that we can write the variation of u as a total derivative with respect to : =

Inserting this into the rst two terms in the last integral in Equation (5.4) on the facing page, we obtain

1 0

L(4) (x , u ) d

Partial integration in the two rst terms in the right hand member of (5.7) gives

1 0

L(4) (x , u ) d

1 0

where the integrated parts do not contribute since the variations at the endpoints vanish. A change of irrelevant summation index from to in the rst two terms of the right hand member of (5.8) yields, after moving the ensuing common factor x outside the partenthesis, the following expression:

1 0

L(4) (x , u ) d

1 0

Applying well-known rules of differentiation and the expression (4.37) for the four-velocity, we can express dA /d as follows: dA A dx = = A u d x d (5.10)

du dA m0 q + qu A x d d d

du dA m0 x q x + qu A x d d d

d d m0 u x + qA x + qu A x d d d

u =

dx d

d x d

(5.6)

(5.7)

(5.8)

(5.9)

70

E LECTROMAGNETIC F IELDS

AND

PARTICLES

By inserting this expression (5.10) into the second term in right-hand member of Equation (5.9) above, and noting the common factor qu of the resulting term and the last term, we obtain the nal variational principle expression

1 0

L(4) (x , u ) d

Since, according to the variational principle, this expression shall vanish and x is arbitrary between the xed end points 0 and 1 , the expression inside in the integrand in the right hand member of Equation (5.11) must vanish. In other words, we have found an equation of motion for a charged particle in a prescribed electromagnetic eld: m0

With the help of Equation (4.79) on page 63 we can express this equation in terms of the electromagnetic eld tensor in the following way: m0 du = qu F d (5.13)

This is the sought-for covariant equation of motion for a particle in an electromagnetic eld. It is often referred to as the Minkowski equation. As the reader can easily verify, the spatial part of this 4-vector equation is the covariant (relativistically correct) expression for the Newton-Lorentz force equation.

Hamiltonian formalism

The usual Hamilton equations for a 3D space are given by Equation (M.101) on page 185 in Appendix M. These six rst-order partial differential equations are H dqi = pi dt H dpi = qi dt (5.14a) (5.14b)

where H(pi , qi , t) = pi qi L(qi , qi , t) is the ordinary 3D Hamiltonian, qi is a generalised coordinate and pi is its canonically conjugate momentum.

du = qu A A d

du m0 + qu A A d

(5.11) x d

(5.12)

5.1

C HARGED PARTICLES

IN AN

E LECTROMAGNETIC F IELD

71

We seek a similar set of equations in 4D space. To this end we introduce a canonically conjugate four-momentum p in an analogous way as the ordinary 3D conjugate momentum: p = L(4) u (5.15)

and utilise the four-velocity u , as given by Equation (4.37) on page 57, to dene the four-Hamiltonian H(4) = p u L(4) (5.16)

With the help of these, the radius four-vector x , considered as the generalised four-coordinate, and the invariant line element ds, dened in Equation (4.18) on page 53, we introduce the following eight partial differential equations: H(4) dx = p d dp H(4) = x d (5.17a) (5.17b)

which form the four-dimensional Hamilton equations. Our strategy now is to use Equation (5.15) on the preceding page and Equations (5.17) to derive an explicit algebraic expression for the canonically conjugate momentum four-vector. According to Equation (4.42) on page 58, c times a four-momentum has a zeroth (time) component which we can identify with the total energy. Hence we require that the component p0 of the conjugate fourmomentum vector dened according to Equation (5.15) on the facing page be identical to the ordinary 3D Hamiltonian H divided by c and hence that this cp0 solves the Hamilton equations, Equations (5.14) on the preceding page. This later consistency check is left as an exercise to the reader. Using the denition of H(4) , Equation (5.16), and the expression for L(4) , Equation (5.3) on page 68, we obtain 1 H(4) = p u L(4) = p u m0 u u qu A (x ) 2 (5.18)

Furthermore, from the denition (5.15) of the canonically conjugate fourmomentum p , we see that

p =

L(4) = u u

1 m0 u u + qu A (x ) = m0 u + qA 2

(5.19)

72

E LECTROMAGNETIC F IELDS

AND

PARTICLES

Since the four-velocity scalar-multiplied by itself is u u = c2 , we clearly see from Equation (5.20) that H(4) is indeed a scalar invariant, whose value is simply H(4) = m0 c2 2 (5.21)

p qA

That this four-Hamiltonian yields the correct covariant equation of motion can be seen by inserting it into the four-dimensional Hamiltons equations (5.17) and using the relation (5.22): H(4) q A = (p qA ) x m0 x A q = m0 u m0 x A = qu x du dp A = = m0 q u d d x

where in the last step Equation (5.19) on the previous page was used. Rearranging terms, and using Equation (4.80) on page 63, we obtain m0 du = qu A A = qu F d

p p 2qA p + q2 A A

p qA

H(4) =

m0 2 1 = 2m0 1 = 2m0

1 1 p qA p qA m0 m0

u =

1 p qA m0

(5.22)

(5.23)

(5.24)

(5.25)

5.1

C HARGED PARTICLES

IN AN

E LECTROMAGNETIC F IELD

73

which is identical to the covariant equation of motion Equation (5.13) on page 70. We can then safely conclude that the Hamiltonian in question is correct. Recalling expression (4.48) on page 59 and representing the canonically conjugate four-momentum as p = (p0 , p), we obtain the following scalar products: p p = (p0 )2 (p)2 1 A p = p0 (p A) c 1 A A = 2 2 (A)2 c (5.26a) (5.26b) (5.26c)

Inserting these explicit expressions into Equation (5.23) on the preceding page, and using the fact that for H(4) is equal to the scalar value m0 c2 /2, as derived in Equation (5.21) on page 71, we obtain the equation 1 2 q2 m0 c2 = (p0 )2 (p)2 qp0 + 2q(p A) + 2 2 q2 (A)2 2 2m0 c c which is the second order algebraic equation in p0 : (5.27)

(pqA)2

(p qA)2 + m2 c2 0

Since the fourth component (time component) p0 of a four-momentum vector p multiplied by c represents the energy [cf. Equation (4.42) on page 58], the positive solution in Equation (5.29) must be identied with the ordinary Hamilton function H divided by c. Consequently, H cp0 = q + c

(p qA)2 + m2 c2 0

is the ordinary 3D Hamilton function for a charged particle moving in scalar and vector potentials associated with prescribed electric and magnetic elds. The ordinary Lagrange and Hamilton functions L and H are related to each other by the 3D transformation [cf. the 4D transformation (5.16) between L (4) and H(4) ] L = pvH (5.31)

y

x iw

v

(p0 )2

(5.28)

(5.29)

(5.30)

74

E LECTROMAGNETIC F IELDS

AND

PARTICLES

Using the explicit expressions (Equation (5.30) above) and (Equation (5.31)), we obtain the explicit expression for the ordinary 3D Lagrange function L = p v q c

where the quantity mv is the usual kinetic momentum, we can rewrite this expression for the ordinary Lagrangian as follows: L = qA v + mv2 q c m2 v2 + m2 c2 0

What we have obtained is the relativstically correct (covariant) expression for the Lagrangian describing the motion of a charged particle in scalar and vector potentials associated with prescribed electric and magnetic elds.

So far, we have considered two classes of problems. Either we have calculated the elds from given, prescribed distributions of charges and currents, or we have derived the equations of motion for charged particles in given, prescribed elds. Let us now put the elds and the particles on an equal footing and present a theoretical description which treats the elds, the particles, and their interactions in a unied way. This involves transition to a eld picture with an innite number of degrees of freedom. We shall rst consider a simple mechanical problem whose solution is well known. Then, drawing inferences from this model problem, we apply a similar view on the electromagnetic problem.

Consider N identical mass points, 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 constants k. At equilibrium the mass

= mv q( A v) mc = q + qA v m0 c

(p qA)2 + m2 c2 0

(5.32)

(5.33)

v2 1 2 c

(5.34)

5.2

75

i1 m m m

i+1 m m x

k a

k a

k a

k a

F IGURE 5.1: A one-dimensional chain consisting of N discrete, identical mass points m, connected to their neighbours with identical, ideal springs with spring constants k. The equilibrium distance between the neighbouring mass points is a and i1 (t), i (t), i+1 (t) are the instantaneous deviations, along the x axis, of positions of the (i 1)th, ith, and (i + 1)th mass point, respectively.

points are at rest, distributed evenly with a distance a to their two nearest neighbours. After perturbation, the motion of mass point i will be a one-dimensional oscillatory motion along x. Let us denote the deviation for mass point i from its equilibrium position by i (t) x. The solution to this mechanical problem can be obtained if we can nd a Lagrangian (Lagrange function) L which satises the variational equation L(i , i , t) dt = 0 (5.35)

According to Equation (M.96) on page 185, the Lagrangian is L = T V where T denotes the kinetic energy and V the potential energy of a classical mechanical system with conservative forces. In our case the Lagrangian is

Let us write the Lagrangian, as given by Equation (5.36) above, in the following way:

i=1

Here,

G

i+1 i 1 m 2 ka 2 a i a

L = a

L=

1 N m 2 k(i+1 i )2 i 2 i=1

(5.36)

(5.37)

(5.38)

76

E LECTROMAGNETIC F IELDS

AND

PARTICLES

is the so called linear Lagrange density. If we now let N and, at the same time, let the springs become innitesimally short according to the following scheme: a dx m dm = a dx ka Y i+1 i a x we obtain (5.39a) linear mass density Youngs modulus (5.39b) (5.39c) (5.39d)

where

Notice how we made a transition from a discrete description, in which the mass points were identied by a discrete integer variable i = 1, 2, . . . , N, to a continuous description, where the innitesimal mass points were instead identied by a continuous real parameter x, namely their position along x. A consequence of this transition is that the number of degrees of freedom for the system went from the nite number N to innity! Another consequence is that has now become dependent also on the partial derivative with respect to x of the eld coordinate . But, as we shall see, the transition is well worth the price because it allows us to treat all elds, be it classical scalar or vectorial elds, or wave functions, spinors and other elds that appear in quantum physics, on an equal footing. Under the assumption of time independence and xed endpoints, the variation principle (5.35) on page 74 yields:

L dt

=0

G F

+ + t x t x

G F

, t x

dx dt (5.42) dx dt

1 , , , t = t x 2 t

L=

dx

(5.40)

(5.41)

5.2

77

The last integral can be integrated by parts. This results in the expression

where the variation is arbitrary (and the endpoints xed). This means that the integrand itself must vanish. If we introduce the functional derivative

which is the one-dimensional Euler-Lagrange equation. Inserting the linear mass point chain Lagrangian density, Equation (5.41) on the preceding page, into Equation (5.45), we obtain the equation of motion for our one-dimensional linear mechanical structure. It is: 2 2 2 2 2 =0 (5.46) 2Y 2 = t x Y t2 x i.e., the one-dimensional wave equation for compression waves which propa gate with phase speed v = Y/ along the linear structure. A generalisation of the above 1D results to a three-dimensional continuum is straightforward. For this 3D case we get the variational principle

=0

where the variation is arbitrary and the endpoints are xed. This means that the integrand itself must vanish:

fd

=0

fd

L dt =

G F

fd

G F

=0

fd

= x

G F

G F

fd

fd

dx dt = 0

(5.43)

(5.44)

(5.45)

(5.48)

78

E LECTROMAGNETIC F IELDS

AND

PARTICLES

This constitutes the four-dimensional Euler-Lagrange equations. Introducing the three-dimensional functional derivative

In analogy with particle mechanics (nite number of degrees of freedom), we may introduce the canonically conjugate momentum density (x ) = (t, x) =

If, as usual, we differentiate this expression and identify terms, we obtain the following Hamilton density equations = t = t

The Hamilton density functions are in many ways similar to the ordinary Hamilton functions and lead to similar results.

Above, when we described the mechanical eld, we used a scalar eld (t, x). If we want to describe the electromagnetic eld in terms of a Lagrange density and Euler-Lagrange equations, it comes natural to express in terms of the four-potential A (x ). The entire system of particles and elds consists of a mechanical part, a eld part and an interaction part. We therefore assume that the total Lagrange density tot for this system can be expressed as

tot

mech

inter

eld

, ,

;t = i x t

G F

G F

fd

=0

xi

p

, , t xi

fd

= xi

(5.49)

(5.50)

(5.51)

(5.52)

(5.53a) (5.53b)

(5.54)

5.2

79

where the mechanical part has to do with the particle motion (kinetic energy). It is given by L(4) /V where L(4) is given by Equation (5.2) on page 68 and V is the volume. Expressed in the rest mass density 0 , the mechanical Lagrange density can be written

mech

1 2

0u

(5.55)

The inter part which describes the interaction between the charged particles and the external electromagnetic eld. A convenient expression for this interaction Lagrange density is

For the eld part eld we choose the difference between magnetic and electric energy density (in analogy with the difference between kinetic and potential energy in a mechanical eld). Using the eld tensor, we express this eld Lagrange density as

eld

1 F F 40 1 F F 40

tot

1 2

0u

u + j A +

Show, by explicit calculation, that 1 B2 1 F F = 0 E 2 40 2 0 i.e., the difference between the magnetic and electric eld energy densities. From Formula (4.80) on page 63 we recall that F = 0 E x /c E y /c E z /c E x /c E y /c E z /c 0 Bz By Bz 0 B x By Bx 0 E x /c 0 Bz By E y /c Bz 0 Bx E z /c By B x 0 (5.60) (5.59)

where denotes the row number and the column number. Then, Einstein summation

@@ 8B

@@ B8

"

2445

2445

inter

= j A

(5.56)

(5.57)

(5.58)

E XAMPLE 5.1

(5.61)

80

E LECTROMAGNETIC F IELDS

AND

PARTICLES

2 2 = 0 E 2 /c2 E y /c2 E z /c2 x 2 E y /c2 + B2 + 0 + B2 z x

E 2 /c2 + 0 + B2 + B2 x z y

(5.62)

2 E z /c2 + B2 + B2 + 0 y x

E ND

OF EXAMPLE

5.1

Using tot in the 3D Euler-Lagrange equations, Equation (5.48) on page 77 (with replaced by A ), we can derive the dynamics for the whole system. For instance, the electromagnetic part of the Lagrangian density

EM

inserted into the Euler-Lagrange equations, expression (5.48) on page 77, yields two of Maxwells equations. To see this, we note from Equation (5.64) and the results in Example 5.1 that

EM

= j

inter

eld

= j A +

1 F F 40

(5.64)

(5.65)

"

"

1 1 F F = 40 2

1 2 B2 E 0 c 2 0

1 2

B2 0 E 2 0

(5.63)

QED

5.2

81

Furthermore,

But A A = A A + A A ( A ) ( A ) ( A ) A + A g g A = A ( A ) ( A ) = A A + g g A A (5.67) ( A ) ( A ) = A A + A A ( A ) ( A )

= 2 A Similarly, A A = 2 A ( A ) so that EM 1 1 = A A = F ( A ) 0 0

This means that the Euler-Lagrange equations, expression (5.48) on page 77, for the Lagrangian density EM and with A as the eld quantity become

EM

EM 1 = j F = 0 ( A ) 0

1 A A A A 20 ( A )

9

1 A A A A 40 ( A ) A A + A A

(5.66) (5.68) (5.69) (5.70)

EM 1 = F F ( A ) 40 ( A ) 1 = ( A A )( A A ) 40 ( A )

82

E LECTROMAGNETIC F IELDS

AND

PARTICLES

or F = 0 j (5.71)

which, according to Equation (4.83) on page 64, is the covariant version of Maxwells source equations.

Other elds

In general, the dynamic equations for most any elds, and not only electromagnetic ones, can be derived from a Lagrangian density together with a variational principle (the Euler-Lagrange equations). Both linear and non-linear elds are studied with this technique. As a simple example, consider a real, scalar eld which has the following Lagrange density: =

Insertion into the 1D Euler-Lagrange equation, Equation (5.45) on page 77, yields the dynamic equation (

m2 ) = 0 em|x| |x|

which describes the Yukawa meson eld for a scalar meson with mass m. With = 1 c2 t (5.75)

which is positive denite. Another Lagrangian density which has attracted quite some interest is the Proca Lagrangian

EM

inter

eld

= j A +

1 2 2 c + ( )2 + m2 2 2

R

1 F F + m2 A A 40

1 m2 2 2

(5.72)

(5.73)

(5.76)

(5.77)

5.2

B IBLIOGRAPHY

83

This equation describes an electromagnetic eld with a mass, or, in other words, massive photons. If massive photons would exist, large-scale magnetic elds, including those of the earth and galactic spiral arms, would be signicantly modied to yield measurable discrepances from their usual form. Space experiments of this kind onboard satellites have led to stringent 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.

Bibliography

[1] A. O. BARUT, Electrodynamics and Classical Theory of Fields and Particles, Dover Publications, Inc., New York, NY, 1980, ISBN 0-486-64038-8. [2] V. L. G INZBURG, Applications of Electrodynamics in Theoretical Physics and Astrophysics, Revised third ed., Gordon and Breach Science Publishers, New York, London, Paris, Montreux, Tokyo and Melbourne, 1989, ISBN 288124-719-9. [3] H. G OLDSTEIN, Classical Mechanics, second ed., Addison-Wesley Publishing Company, Inc., Reading, MA . . . , 1981, ISBN 0-201-02918-9. [4] W. T. G RANDY, Introduction to Electrodynamics and Radiation, Academic Press, New York and London, 1970, ISBN 0-12-295250-2. [5] 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. [6] 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. [7] J. J. S AKURAI, Advanced Quantum Mechanics, Addison-Wesley Publishing Company, Inc., Reading, MA . . . , 1967, ISBN 0-201-06710-2. [8] D. E. S OPER, Classical Field Theory, John Wiley & Sons, Inc., New York, London, Sydney and Toronto, 1976, ISBN 0-471-81368-0.

84

E LECTROMAGNETIC F IELDS

AND

PARTICLES

6

Electromagnetic Fields and Matter

The microscopic Maxwell equations (1.43) derived in Chapter 1 are valid on all scales where a classical description is good. However, when macroscopic matter is present, it is sometimes convenient to use the corresponding macroscopic Maxwell equations (in a statistical sense) in which auxiliary, derived elds are introduced in order to incorporate effects of macroscopic matter when this is immersed fully or partially in an electromagnetic eld.

In certain cases, for instance in engneering applications, it may be convenient to separate the inuence of an external electric eld on free charges and on neutral matter in bulk. This view, which, as we shall see, has certain limitations, leads to the introduction of (di)electric polarisation and (di)electric displacement.

The electrostatic properties of a spatial volume containing electric charges and located near a point x0 can be characterized in terms of the total charge or electric monopole moment q=

V

(x ) d3x

(6.1)

where the is the charge density introduced in Equation (1.7) on page 4, the electric dipole moment vector p(x0 ) = (x x0 )(x ) d3x (6.2)

85

86

E LECTROMAGNETIC F IELDS

AND

M ATTER

with components pi , i = 1, 2, 3, the electric quadrupole moment tensor Q(x0 ) = (x x0 )(x x0 )(x ) d3x (6.3)

with components Qi j , i, j = 1, 2, 3, and higher order electric moments. In particular, the electrostatic potential Equation (3.3) on page 35 from a charge distribution located near x0 can be Taylor expanded in the following way:

where Einsteins summation convention over i and j is implied. As can be seen from this expression, only the rst few terms are important if the eld point (observation point) is far away from x0 . For a normal medium, the major contributions to the electrostatic interactions come from the net charge and the lowest order electric multipole moments induced by the polarisation due to an applied electric eld. Particularly important is the dipole moment. Let P denote the electric dipole moment density (electric dipole moment per unit volume; unit: C/m2 ), also known as the electric polarisation, in some medium. In analogy with the second term in the expansion Equation (6.4) on page 86, the electric potential from this volume distribution P(x ) of electric dipole moments p at the source point x can be written

Using the expression Equation (M.83) on page 182 and applying the divergence theorem, we can rewrite this expression for the potential as follows:

p (x) =

1 40 1 = 40

p (x) =

1 40 1 = 40

P(x )

1 Qi j + |x x0 |3

stat (x) =

q 1 1 (x x0 )i + p 2 i |x x | 40 |x x0 | |x x0 | 0

3 (x x0 )i (x x0 ) j 1 i j + . . . 2 |x x0 | |x x0 | 2

(6.4)

1 d3x |x x | (6.5)

(6.6)

6.1

E LECTRIC P OLARISATION

AND

D ISPLACEMENT

87

where the rst 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 P n at the surface. Doing so, we nd that the contribution from the electric dipole moments to the potential is given by p = 1 40

V

P(x ) 3 dx |x x |

(6.7)

Comparing this expression with expression Equation (3.3) on page 35 for the electrostatic potential from a static charge distribution , we see that P(x) has the characteristics of a charge density and that, to the lowest order, the effective charge density becomes (x) P(x), in which the second term is a polarisation term. The version of Equation (1.7) on page 4 where true and polarisation charges are separated thus becomes E = (x) P(x) 0 (6.8)

Rewriting this equation, and at the same time introducing the electric displacement vector (C/m2 ) D = 0 E + P we obtain (0 E + P) = D = true (x) (6.10) (6.9)

where true is the true charge density in the medium. This is one of Maxwells equations and is valid also for time varying elds. By introducing the notation pol = P for the polarised charge density in the medium, and total = true + pol for the total charge density, we can write down the following alternative version of Maxwells equation (6.23a) on page 90 E = total (x) 0 (6.11)

Often, for low enough eld strengths |E|, the linear and isotropic relationship between P and E P = 0 E (6.12)

is a good approximation. The quantity is the electric susceptibility which is material dependent. For electromagnetically anisotropic media such as a

88

E LECTROMAGNETIC F IELDS

AND

M ATTER

magnetised plasma or a birefringent crystal, the susceptibility is a tensor. In general, the relationship is not of a simple linear form as in Equation (6.12) on the previous page but non-linear terms are important. In such a situation the principle of superposition is no longer valid and non-linear effects such as frequency conversion and mixing can be expected. Inserting the approximation (6.12) into Equation (6.9) on the preceding page, we can write the latter D = E where, approximately, = 0 (1 + ) (6.14) (6.13)

An analysis of the properties of stationary magnetic media and the associated currents shows that three such types of currents exist: 1. In analogy with true charges for the electric case, we may have true currents jtrue , i.e., a physical transport of true charges. 2. In analogy with electric polarisation P there may be a form of charge transport associated with the changes of the polarisation with time. We call such currents induced by an external eld polarisation currents. We identify them with P/t. 3. There may also be intrinsic currents of a microscopic, often atomic, nature that are inaccessible to direct observation, but which may produce net effects at discontinuities and boundaries. We shall call such currents magnetisation currents and denote them jM . No magnetic monopoles have been observed yet. So there is no correspondence in the magnetic case to the electric monopole moment (6.1). The lowest order magnetic moment, corresponding to the electric dipole moment (6.2), is the magnetic dipole moment m= 1 2 (x x0 ) j(x ) d3x (6.15)

For a distribution of magnetic dipole moments in a volume, we may describe this volume in terms of the magnetisation, or magnetic dipole moment per unit

6.2

M AGNETISATION

AND THE

M AGNETISING F IELD

89

volume, M. Via the denition of the vector potential one can show that the magnetisation current and the magnetisation is simply related: jM = M (6.16)

In a stationary medium we therefore have a total current which is (approximately) the sum of the three currents enumerated above: jtotal = jtrue + P + M t (6.17)

Moving the term M to the left hand side and introducing the magnetising eld (magnetic eld intensity, Ampre-turn density) as H= B M 0 (6.19)

and using the denition for D, Equation (6.9) on page 87, we can write this incorrect equation in the following form H = jtrue + E P D = jtrue + 0 t t t (6.20)

As we see, in this simplistic view, we would pick up a term 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 overcome this inconsistency he was forced to add his famous displacement current term which precisely compensates for the last term in the right hand side. In Chapter 1, we discussed an alternative way, based on the postulate of conservation of electric charge, to introduce the displacement current. We may, in analogy with the electric case, introduce a magnetic susceptibility for the medium. Denoting it m , we can write H= B (6.21)

90

E LECTROMAGNETIC F IELDS

AND

M ATTER

Maxwells equations expressed in terms of the derived eld quantities D and H are D = (t, x) B = 0 E = B t D t (6.23a) (6.23b) (6.23c) (6.23d)

H = j(t, x) +

and are called Maxwells macroscopic equations. These equations are convenient to use in certain simple cases. Together with the boundary conditions and the constitutive relations, they describe uniquely (but only approximately!) the properties of the electric and magnetic elds in matter.

We shall use Maxwells macroscopic equations in the following considerations on the energy and momentum of the electromagnetic eld and its interaction with matter.

Scalar multiplying (6.23c) by H, (6.23d) by E and subtracting, we obtain H ( E) E ( H) = (E H) B D 1 = H EjE = (H B + E D) j E t t 2 t (6.24)

Integration over the entire volume V and using Gausss theorem (the divergence theorem), we obtain t 1 (H B + E D) d3x = 2 j E d3x + (E H) nd2x (6.25)

But, according to Ohms law in the presence of an electromotive force eld, Equation (1.26) on page 10: j = (E + EEMF ) which means that

V

j E d3x =

j2 3 dx

j EEMF d3x

(6.26)

(6.27)

6.3

E NERGY

AND

M OMENTUM

91

Inserting this into Equation (6.25) on the preceding page j EEMF d3x = j2 3 dx + t 1 (E D + H B) d3x + 2

Field energy

Joule heat

Radiated power

(6.28) which is the energy theorem in Maxwells theory also known as Poyntings theorem. It is convenient to introduce the following quantities: 1 E D d3x 2 V 1 H B d3x Um = 2 V S = EH Ue = (6.29) (6.30) (6.31)

where U e is the electric eld energy, U m is the magnetic eld energy, both measured in J, and S is the Poynting vector (power ux), measured in W/m2 .

Let us now investigate the momentum balance (force actions) in the case that a eld interacts with matter in a non-relativistic way. For this purpose we consider the force density given by the Lorentz force per unit volume E+jB. Using Maxwells equations (6.23) and symmetrising, we obtain

y iw v x

E + j B = ( D)E +

D B t D = E( D) + ( H) B B t = E( D) B ( H) B (D B) + D t t = E( D) B ( H) (D B) D ( E) + H( B) t H

=0

(6.32)

x w

v y

x iw

x iw v

x iw

(E H) nd2x

92

E LECTROMAGNETIC F IELDS

AND

M ATTER

One veries easily that the ith vector components of the two terms in square brackets in the right hand member of (6.32) can be expressed as

and

respectively. Using these two expressions in the ith component of Equation (6.32) on the preceding page and re-shufing terms, we get

Introducing the electric volume force Fev via its ith component

and the Maxwell stress tensor T with components 1 1 T i j = E i D j E D i j + Hi B j H B i j 2 2 we nally obtain the force equation Fev + T i j = ( T)i (D B) = t x j i

If we introduce the relative electric permittivity and the relative magnetic permeability m as D = 0 E = E B = m 0 H = H (6.39) (6.40)

H

(Fev )i = (E + j B)i

1 2

D E B H D + H B xi xi xi xi

x j

1 1 E i D j E D i j + Hi B j H B i j 2 2

(E + j B)i

1 2

D E D xi xi

+ H

B H B xi xi

(D B)i t

(6.35)

(6.36)

(6.37)

(6.38)

[H( B) B ( H)]i =

1 B H H B + 2 xi xi x j

1 Hi B j B H i j 2 (6.34)

[E( D) D ( E)]i =

1 D E E D + 2 xi xi x j

1 E i D j E D i j 2 (6.33)

6.3

B IBLIOGRAPHY

93

where S is the Poynting vector dened in Equation (6.29) on page 91. Integration over the entire volume V yields Fev d3x + d dt m 3 Sd x = c2 T n d2x (6.42)

Field momentum

Maxwell stress

which expresses the balance between the force on the matter, the rate of change of the electromagnetic eld momentum and the Maxwell stress. This equation is called the momentum theorem in Maxwells theory. In vacuum (6.42) becomes (E + v B) d3x + 1 d c2 dt S d3x = T n d2x (6.43)

or

Bibliography

[1] E. H ALLN, Electromagnetic Theory, Chapman & Hall, Ltd., London, 1962. [2] J. D. JACKSON, Classical Electrodynamics, third ed., John Wiley & Sons, Inc., New York, NY . . . , 1999, ISBN 0-471-30932-X. [3] 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. [4] J. A. S TRATTON, Electromagnetic Theory, McGraw-Hill Book Company, Inc., New York, NY and London, 1953, ISBN 07-062150-0.

d mech d eld p + p = dt dt

T n d2x

x iw

x iw

x iw

(6.44)

94

E LECTROMAGNETIC F IELDS

AND

M ATTER

7

Electromagnetic Fields from Arbitrary Source Distributions

While, in principle, the electric and magnetic elds can be calculated from the Maxwell equations in Chapter 1, or even from the wave equations in Chapter 2, it is often physically more lucid to calculate them from the electromagnetic potentials derived in Chapter 3. In this chapter we will derive the electric and magnetic elds from the potentials. We recall that in order to nd the solution (3.35) for the generic inhomogeneous wave equation (3.19) on page 41 we presupposed the existence of a Fourier transform pair (3.20a) on page 41 for the generic source term (t, x) = (x) =

That such transform pairs exists is true for most physical variables which are neither strictly monotonically increasing nor strictly monotonically decreasing with time. For charge and current densities varying in time we can therefore, without loss of generality, work with individual Fourier components w (x) and jw (x), respectively. Strictly speaking, the existence of a single Fourier component assumes a monochromatic source (i.e., a source containing only one single frequency component), which in turn requires that the electric and magnetic elds exist for innitely long times. However, by taking the proper limits, we may still use this approach even for sources and elds of nite duration. This is the method we shall utilise in this chapter in order to derive the electric and magnetic elds in vacuum from arbitrary given charge densities (t, x) and current densities j(t, x), dened by the temporal Fourier transform

1 2

(x) eit d

(7.1a) (7.1b)

(t, x) eit dt

95

96

E LECTROMAGNETIC F IELDS

AND

M ATTER

under the assumption that only retarded potentials produce physically acceptable solutions.1 The temporal Fourier transform pair for the retarded vector potential can then be written (t, x) = (x) =

where in the last step, we made use of the explicit expression for the temporal Fourier transform of the generic potential component (x), Equation (3.32) on page 43. Similarly, the following Fourier transform pair for the vector potential must exist: A(t, x) = A (x) =

in order that all physical quantities be real. Similar transform pairs and requirements of real-valuedness exist for the elds themselves.

fact, John A. Wheeler and Richard P. Feynman derived in 1945 a fully self-consistent electrodynamics using both the retarded and the advanced potentials [6]; See also [1].

1 In

1 2

A (x) eit d

A(t, x) eit dt =

0 4

1 2

1 2

1 2

(x) eit d

(7.2a) (7.2b)

(t, x) eit dt

j (x) eit d

(7.3a) (7.3b)

j(t, x) eit dt

(x) eit d

(t, x) eit dt =

(7.6)

7.1

T HE M AGNETIC F IELD

97

In the limit that the sources can be considered monochromatic containing only one single frequency 0 , we have the much simpler expressions (t, x) = 0 (x)ei0 t j(t, x) = j0 (x)e

i0 t

i0 t

where again the real-valuedness of all these quantities is implied. As discussed above, we can safely assume that all formulae derived for a general temporal Fourier representation of the source (general distribution of frequencies in the source) are valid for these simple limiting cases. We note that in this context, we can make the formal identication = 0 ( 0 ), j = j0 ( 0 ) etc., and that we therefore, without any loss of stringence, let 0 mean the same as the Fourier amplitude and so on.

Let us now compute the magnetic eld from the vector potential, dened by Equation (7.5a) and Equation (7.5b) on the preceding page, and Formula (3.6) on page 36: B(t, x) = A(t, x) (7.8)

The calculations are much simplied if we work in space and, at the nal stage, Fourier transform back to ordinary t space. We are working in the Lorentz gauge and note that in space the Lorentz condition, Equation (3.13) on page 38, takes the form k A i = 0 c (7.9)

which provides a relation between (the Fourier transforms of) the vector and scalar potentials. Using the Fourier transformed version of Equation (7.8) and Equation (7.5b) on the preceding page, we obtain B (x) = A (x) = 0 4 j (x ) eik|xx | 3 dx |x x | (7.10)

98

E LECTROMAGNETIC F IELDS

FROM

From this expression for the magnetic eld in the frequency () domain, we obtain the total magnetic eld in the temporal (t) domain by taking the inverse Fourier transform (using the identity ik = i/c): B(t, x) = =

3

Induction eld

Radiation eld

where

The rst term, the induction eld, dominates near the current source but falls off rapidly with distance from it, is the electrodynamic version of the BiotSavart law in electrostatics, Formula (1.13) on page 6. The second term, the radiation eld or the far eld, dominates at large distances and represents energy that is transported out to innity. Note how the spatial derivatives ( ) gave rise to a time derivative ()!

j t

t=tret

x w

v y

x w

0 4

|x x | j(tret , x ) (x x ) 3 0 dx + 3 4c |x x |

1 c

0 4

d3x

0 4

d3x

B (x) =

0 4 0 = 4

(7.11)

(7.12)

(7.13)

7.2

T HE E LECTRIC F IELD

99

In order to calculate the electric eld, we use the temporally Fourier transformed version of Formula (3.10) on page 37, inserting Equations (7.4b) and (7.5b) as the explicit expressions for the Fourier transforms of and A: E (x) = (x) + iA (x)

Using the Fourier transform of the continuity Equation (1.21) on page 9 j (x ) i (x ) = 0 we see that we can express in terms of j as follows (x ) = i j (x ) (7.16) (7.15)

Doing so in the last term of Equation (7.14), and also using the fact that k = /c, we can rewrite this Equation as

The last vector-valued integral can be further rewritten in the following way:

I =

x iw

y

1 c

E (x) =

1 40

1 40 1 = 40

(7.17)

(7.18)

100

E LECTROMAGNETIC F IELDS

FROM

But, since

we can rewrite I as

where, according to Gausss theorem, the last term vanishes if j is assumed to be limited and tends to zero at large distances. Further evaluation of the derivative in the rst term makes it possible to write

Using the triple product bac-cab Formula (F.54) on page 164 backwards, and inserting the resulting expression for I into Equation (7.17) on the preceding page, we arrive at the following nal expression for the Fourier transform of the total E eld: E (x) = 1 40 1 = 40 + + i0 eik|xx | 3 eik|xx | 3 dx + dx j (x ) 4 V |x x | |x x | V (x )eik|xx | (x x ) 3 dx V |x x |3 [j (x )eik|xx | (x x )](x x ) 3 (7.22) dx |x x |4 V [j (x )eik|xx | (x x )] (x x ) 3 dx V |x x |4 [j (x )eik|xx | (x x )] (x x ) 3 dx V |x x |3 (x )

ik c

1 c

1 c

ik

j (x x ) (x x ) |x x |3

ik|xx |

eik|xx | + j |x x |

dx

I =

eik|xx | 2 j (x x ) (x x )eik|xx | + 2 4 |x x | |x x |

xm

I =

jm

xm

xm

jm

xl xl ik|xx | e |x x |2

xl xl ik|xx | e |x x |2 xl xl ik|xx | + jm e xm |x x |2 jm xm

(7.19)

7.3

T HE R ADIATION F IELDS

101

Taking the inverse Fourier transform of Equation (7.22) on the preceding page, once again using the vacuum relation = kc, we nd, at last, the expression in time domain for the total electric eld: E(t, x) = =

Intermediate eld

Intermediate eld

V

Radiation eld

Here, the rst term represents the retarded Coulomb eld and the last term represents the radiation eld which carries energy over very large distances. The other two terms represent an intermediate eld which contributes only in the near zone and must be taken into account there. With this we have achieved our goal of nding closed-form analytic expressions for the electric and magnetic elds when the sources of the elds are completely arbitrary, prescribed distributions of charges and currents. The only assumption made is that the advanced potentials have been discarded; recall the discussion following Equation (3.35) on page 43 in Chapter 3.

In this section we study electromagnetic radiation, i.e., the part of the electric and magnetic elds elds, calculated above, which are capable of carrying energy and momentum over large distances. We shall therefore make the assumption that the observer is located in the far zone, i.e., very far away from the source region(s). The elds which are dominating in this zone are by denition the radiation elds. From Equation (7.12) on page 98 and Equation (7.23), which give the total

x w

1 40 c2

[ ret , x ) (x x )] (x x ) 3 j(t dx |x x |3

x iw

1 40 c

[j(tret , x ) (x x )] (x x ) 3 dx |x x |4

x w

1 40 c

x iw

1 40

(7.23)

102

E LECTROMAGNETIC F IELDS

FROM

electric and magnetic elds, we obtain Brad (t, x) = Erad (t, x) = = where

Instead of studying the elds in the time domain, we can often make a spectrum analysis into the frequency domain and study each Fourier component separately. A superposition of all these components and a transformation back to the time domain will then yield the complete solution. The Fourier representation of the radiation elds Equation (7.24a) above and Equation (7.24b) were included in Equation (7.11) on page 98 and Equation (7.22) on page 100, respectively and are explicitly given by 1 rad B (t, x) eit dt 2 j (x ) (x x ) ik|xx | 3 k0 dx = i e 4 V |x x |2 0 j (x ) k ik|xx | 3 e = i dx 4 V |x x | 1 rad Erad (x) = E (t, x) eit dt 2 [j (x ) (x x )] (x x ) ik|xx | 3 k dx = i e 40 c V |x x |3 1 [j (x ) k] (x x ) ik|xx | 3 = i dx e 40 c V |x x |2

where we used the fact that k = k k = k(x x )/ |x x|. If the source is located inside a volume V near x0 and has such a limited spatial extent that max |x x0 | |x x |, and the integration surface S , centred on x0 , has a large enough radius |x x0 | max |x x0 |, we see from Figure 7.1 on the next page that we can approximate k x x k (x x ) k (x x0 ) k (x x0 ) k |x x0 | k (x x0 )

Brad (x) =

j t

t=tret

1 40 c2

[ ret , x ) (x x )] (x x ) 3 j(t dx |x x |3

0 4c

(7.25)

(7.26a)

(7.26b)

(7.27)

7.3

T HE R ADIATION F IELDS

103

S dS = nd2x k xx x

x x0

x x0

x x0 V

O F IGURE 7.1: Relation between the surface normal and the k vector for radiation generated at source points x near the point x0 in the source volume V. At distances much larger than the extent of V, the unit vector n, normal to the surface S which has its centre at x0 , and the unit vector k of the radiation k vector from x are nearly coincident.

Recalling from Formula (F.48) and Formula (F.49) on page 164 that dS = |x x0 |2 d = |x x0 |2 sin d d and noting from Figure 7.1 that k and n are nearly parallel, we see that we can approximate. k dS k n = dS d 2 |x x0 | |x x0 |2 (7.28)

Both these approximations will be used in the following. Within approximation (7.27) the expressions (7.26a) and (7.26b) for the radiation elds can be approximated as 0 ik|xx0 | j (x ) k ik(x x0 ) 3 e e dx 4 V |x x | (7.29a) 0 eik|xx0 | ik(x x0 ) 3 dx i [j (x ) k] e 4 |x x0 | V [j (x ) k] (x x ) ik(x x0 ) 3 1 Erad (x) i eik|xx0 | dx e 40 c |x x |2 V 1 eik|xx0 | (x x0 ) [j (x ) k] eik(x x0 ) d3x i 40 c |x x0 | |x x0 | V (7.29b) Brad (x) i

104

E LECTROMAGNETIC F IELDS

FROM

I.e., if max |x x0 | |x x |, then the elds can be approximated as spherical waves multiplied by dimensional and angular factors, with integrals over points in the source volume only.

Let us consider the energy that is carried in the radiation elds Brad , Equation (7.26a), and Erad , Equation (7.26b) on page 102. We have to treat signals with limited lifetime and hence nite frequency bandwidth differently from monochromatic signals.

If the source is strictly monochromatic, we can obtain the temporal average of the radiated power P directly, simply by averaging over one period so that

Using the far-eld approximations (7.29a) and (7.29b) and the fact that 1/c = 0 0 and R0 = 0 /0 according to the denition (2.18) on page 26, we obtain dx

A signal with nite pulse width in time (t) domain has a certain spread in frequency () domain. To calculate the total radiated energy we need to integrate

dP 1 = R0 d 322

(j k)eik(x x0 ) d3x

1 1 S = R0 322 |x x0 |2

(j k)e

ik(x x0 ) 3

x x0 |x x0 |

S = EH =

(7.30)

(7.31)

(7.32)

7.4

R ADIATED E NERGY

105

over the whole bandwidth. The total energy transmitted through a unit area is the time integral of the Poynting vector:

S(t) dt = =

(E H) dt d

If we carry out the temporal integration rst and use the fact that

ei(+ )t dt = 2( + )

S(t) dt = 2

(E H ) d

where the last step follows from the real-valuedness of E and B . We insert the Fourier transforms of the eld components which dominate at large distances, i.e., the radiation elds (7.26a) and (7.26b). The result, after integration over the area S of a large sphere which encloses the source, is U=

Inserting the approximations (7.27) and (7.28) into Equation (7.36) and also introducing U=

0

U d

and recalling the denition (2.18) on page 26 for the vacuum resistance R 0 we obtain dx

V

1 dU d R0 d 4

(j k)e

ik(x x0 ) 3

1 4

0 0

j k ik|xx | 3 dx e |x x |

2 0 2 = 0

0 0

(E B + E B ) d (E B + E B ) d

k ndS d

= 2

(E H ) d +

(E H ) d

= 2

(E H ) d

(E H ) d

= 2

(E H ) d +

(E H )ei(+ )t dt

(7.33)

(7.34)

(E H ) d

(7.35)

(7.36)

(7.37)

(7.38)

106

E LECTROMAGNETIC F IELDS

FROM

which, at large distances, is a good approximation to the energy that is radiated per unit solid angle d in a frequency band d. It is important to notice that Formula (7.38) includes only source coordinates. This means that the amount of energy that is being radiated is independent on the distance to the source (as long as it is large).

Bibliography

[1] F. H OYLE , S IR AND J. V. NARLIKAR, Lectures on Cosmology and Action at a Distance Electrodynamics, World Scientic Publishing Co. Pte. Ltd, Singapore, New Jersey, London and Hong Kong, 1996, ISBN 9810-02-2573-3(pbk). [2] J. D. JACKSON, Classical Electrodynamics, third ed., John Wiley & Sons, Inc., New York, NY . . . , 1999, ISBN 0-471-30932-X. [3] 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. [4] 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. [5] J. A. S TRATTON, Electromagnetic Theory, McGraw-Hill Book Company, Inc., New York, NY and London, 1953, ISBN 07-062150-0. [6] J. A. W HEELER AND R. P. F EYNMAN, Interaction with the absorber as a mechanism for radiation, Reviews of Modern Physics, 17 (1945), pp. 157.

8

Electromagnetic Radiation and Radiating Systems

In Chapter 3 we were able to derive general expressions for the scalar and vector potentials from which we then, in Chapter 7, calculated the total electric and magnetic elds from arbitrary distributions of charge and current sources. The only limitation in the calculation of the elds was that the advanced potentials were discarded. Thus, one can, at least in principle, calculate the radiated elds, Poynting ux and energy for an arbitrary current density Fourier component and then add these Fourier components together to construct the complete electromagnetic eld at any time at any point in space. However, in practice, it is often difcult to evaluate the source integrals unless the current has a simple distribution in space. In the general case, one has to resort to approximations. We shall consider both these situations.

Certain radiation systems have a geometry which is one-dimensional, symmetric or in any other way simple enough that a direct calculation of the radiated elds and energy is possible. This is for instance the case when the current ows in one direction in space only and is limited in extent. An example of this is a linear antenna.

107

108

E LECTROMAGNETIC R ADIATION

AND

R ADIATING S YSTEMS

L 2

L 2

F IGURE 8.1: A linear antenna used for transmission. The current in the feeder and the antenna wire is set up by the EMF of the generator (the transmitter). At the ends of the wire, the current is reected back with a 180 phase shift to produce a antenna current in the form of a standing wave.

Let us apply Equation (7.32) on page 104 to calculate the power from a linear, transmitting antenna, fed across a small gap at its centre with a monochromatic source. The antenna is a straight, thin conductor of length L which carries a one-dimensional time-varying current so that it produces electromagnetic radiation. We assume that the conductor resistance and the energy loss due to the electromagnetic radiation are negligible. Since we can assume that the antenna wire is innitely thin, the current must vanish at the end points L/2 and L/2. The charges in this thin wire are set in motion due to the EMF of the generator (transmitter) to produce an antenna current which is the source of the EM radiation. The current is reected at the ends of the antenna wire and undergoes there a phase shift of radians. The combined effect of this is that the antenna current forms a standing wave as indicated in Figure 8.1 For a Fourier component 0 the standing wave current density can be written as j(t , x ) = j0 (x ) exp{i0 t } [cf. Equations (7.7) on page 97] where j0 (x ) = I0 (x1 )(x2 ) sin[k(L/2 x3 )] x3 sin(kL/2)

(8.1)

8.1

R ADIATION

FROM

E XTENDED S OURCES

109

r x3

L 2

j (x )

k x2

x1 L 2 F IGURE 8.2: We choose a spherical polar coordinate system (r = |x| , , ) and orientate it so that the linear antenna axis (and thus the antenna current density j ) is along the polar axis with the feed point at the origin.

In order to evaluate Formula (7.32) on page 104 with the explicit monochromatic current (8.1) inserted, we use a spherical polar coordinate system as in Figure 8.2 to evaluate the source integral

Inserting this expression and d = 2 sin d into Formula (7.32) on page 104 and integrating over , we nd that the total radiated power from the antenna

2 = 4I0

(8.2)

2 = I0

L/2

L/2

j0 k)eik(x x0 ) d3x

110

E LECTROMAGNETIC R ADIATION

AND

R ADIATING S YSTEMS

is

kL0

is called the radiation resistance. For the technologically important case of a half-wave antenna, i.e., for L = /2 or kL = , Formula (8.3) reduces to

The integral in (8.6) can always be evaluated numerically. But, it can in fact also be evaluated analytically as follows:

0

where in the last step the Euler-Mascheroni constant = 0.5772 . . . and the cosine integral Ci(x) were introduced. Inserting this into the expression Equation (8.6) above we obtain the value Rrad (/2) 73 .

cos2

1 cos2 u cos 2 d = [cos u] = du = 2 sin 1 1 u 1 + cos(u) cos2 u = 2 2 1 1 + cos(u) 1 du = 2 1 (1 + u)(1 u) (8.7) 1 1 1 + cos(u) 1 1 1 + cos(u) = du + du 4 1 (1 + u) 4 1 (1 u) 1 1 + cos(u) 1 v = du = 1 + u 2 1 (1 + u) 1 2 1 cos v 1 = dv = [ + ln 2 Ci(2)] 2 0 v 2 1.22

2 P(/2) = R0 I0

1 4

cos2

cos d sin

Rrad (L) =

2 2 R0 I0

197

2 P(L) = R0 I0

1 4

sin d

(8.3)

(8.4)

(8.5)

(8.6)

8.1

R ADIATION

FROM

E XTENDED S OURCES

111

r k

x3 = z = z

z x1 x

x2

F IGURE 8.3: For the loop antenna the spherical coordinate system (r, , ) describes the eld point (the radiation eld) and the cylindrical coordinate system ( , , z ) describes the source (the antenna current).

As an example of a two-dimensional current distribution we consider a circular loop antenna and calculate radiated elds from such an antenna. We choose the Cartesian coordinate system x1 x2 x3 with its origin at the centre of the loop as in the Figure 8.3 According to Equation (7.29a) on page 103 in the formula collection the Fourier component of the radiation part of the magnetic eld generated by an extended, monochromatic current source is Brad = i0 eik|x| 4 |x| eikx j k d3x (8.8)

In our case the antenna current oscillates in the form of a sinusoidal standing current wave around the circular loop, i.e. j = I0 cos ( a)(z ) (8.9)

For the spherical coordinate system of the eld point, we recall from subsection F.4.1 on page 163 that the following relations between the base vectors

112

E LECTROMAGNETIC R ADIATION

AND

R ADIATING S YSTEMS

hold: r = sin cos x1 + sin sin x2 + cos x3 = cos cos x1 + cos sin x2 sin x3

x1 = sin cos + cos cos sin r x2 = sin sin + cos sin + cos r x3 = cos sin r With the use of the above transformations and trigonometric identities, we obtain for the cylindrical coordinate system which describes the source: = cos x1 + sin x2 = sin x1 + cos x2 = sin cos( ) + cos cos( ) + sin( ) r (8.10) (8.11) (8.12)

This choice of coordinate systems means that k = k and x = a so that r k x = ka sin cos( ) and k = k[cos( ) + cos sin( )] (8.14) (8.13)

With these expressions inserted and d3x = d d dz , the source integral becomes eikx j k d3x = a

2 2 0

= I0 ak

2 0

+ I0 ak cos

Utilising the periodicity of the integrands over the integration interval [0, 2], introducing the auxiliary integration variable = , and utilising standard

8.1

R ADIATION

FROM

E XTENDED S OURCES

113

trigonometric identities, the rst integral in the RHS of (8.15) can be rewritten

2 0

2

2 0

2 1 sin eika sin cos d 2 0 2 1 sin eika sin cos cos 2 d 2 0

2 0

ika sin cos

2 0

Putting everything together, we nd that eikx j k d3x = I + I = I0 ak cos [J0 (ka sin ) J2 (ka sin )] + I0 ak cos sin [J0 (ka sin ) + J2 (ka sin )]

in

ei cos cos n d =

in 2

2 0

= cos

= cos

0 2

eika sin cos cos2 d + a vanishing integral eika sin cos 1 1 + cos 2 2 2 d (8.16)

(8.17)

ei cos cos n d

(8.18)

(8.20)

114

E LECTROMAGNETIC R ADIATION

AND

R ADIATING S YSTEMS

so that, in spherical coordinates where |x| = r, i0 eikr I +I (8.21) 4r To obtain the desired physical magnetic eld in the radiation (far) zone we must Fourier transform back to t space and take the real part and evaluate it at the retarded time:

(8.22) From this expression for the radiated B eld, we can obtain the radiated E eld with the help of Maxwells equations.

In the general case, and when we are interested in evaluating the radiation far from the source volume, we can introduce an approximation which leads to a multipole expansion where individual terms can be evaluated analytically. We shall use Hertz method to obtain this expansion.

Let us consider the equation of continuity, which, according to expression (1.21) on page 9, can be written (t, x) + j(t, x) = 0 t (8.23)

In Section 6.1.1 we introduced the electric polarisation P such that P = pol , the polarisation charge density. If we introduce a vector eld (t, x) such that = true = jtrue t (8.24a) (8.24b)

Brad (t, x) = Re

i0 e(ikrt ) I +I 4r

Brad (x) =

8.2

M ULTIPOLE R ADIATION

115

and compare with Equation (8.23) on the facing page, we see that (t, x) satises this equation of continuity. Furthermore, if we compare with the electric polarisation [cf. Equation (6.9) on page 87], we see that the quantity is related to the true charges in the same way as P is related to polarised charge. The quantity is referred to as the polarisation vector since, formally, it treats also the true (free) charges as polarisation charges. We introduce a further potential e with the following property e = 1 e =A c2 t (8.25a) (8.25b)

where and A are the electromagnetic scalar and vector potentials, respectively. 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 and, as can be seen from (8.24) and (8.25), it satises the inhomogeneous wave equation

2

e =

1 2 e c2 t2

e =

This equation is of the same type as Equation (3.19) on page 41, and has therefore the retarded solution e (t, x) = 1 40 (tret , x ) 3 dx |x x | (8.27)

we see that we can calculate the magnetic and electric elds, respectively, as follows B= E= 1 C c2 t C (8.30a) (8.30b)

(8.26)

116

E LECTROMAGNETIC R ADIATION

AND

R ADIATING S YSTEMS

xx x x0 x x x0 V

O F IGURE 8.4: Geometry of a typical multipole radiation problem where the eld point x is located some distance away from the nite source volume V centered around x0 . If k |x x0 | 1 k |x x0 |, then the radiation at x is well approximated by a few terms in the multipole expansion.

Clearly, the last equation is valid only outside the source volume, where E = 0. Since we are mainly interested in the elds in the far zone, a long distance from the source region, this is no essential limitation. Assume that the source region is a limited volume around some central point x0 far away from the eld (observation) point x illustrated in Figure 8.4. Under these assumptions, we can expand the Hertz vector, expression (8.27) on the preceding page, due to the presence of non-vanishing (tret , x ) in the vicinity of x0 , in a formal series. For this purpose we recall from potential theory that eik|(xx0 )(x x0 )| eik|xx | |x x | |(x x0 ) (x x0 )|

n=0

(8.31)

x0 )h(1) (k |x x0 |) n

where eik|xx | is a Green function |x x | is the angle between x x0 and x x0 (see Figure 8.4 on page 116) Pn (cos ) is the Legendre polynomial of order n jn (k x x0 ) is the spherical Bessel function of the rst kind of order n

8.2

M ULTIPOLE R ADIATION

117

According to the addition theorem for Legendre polynomials, we can write Pn (cos ) =

m=n

(8.32)

where Pm is an associated Legendre polynomial and, in spherical polar coorn dinates, x x0 = ( x x0 , , ) x x0 = (|x x0 | , , )

Inserting Equation (8.31) on the preceding page, together with Equation (8.32), into Equation (8.28) on page 115, we can in a formally exact way expand the Fourier component of the Hertz vector as e = ik n (2n + 1)(1)mh(1) (k |x x0 |) Pm (cos ) eim n n 40 n=0 m=n

V

(x ) jn (k x

x0 ) Pm (cos n

We notice that there is no dependence on x x0 inside the integral; the integrand is only dependent on the relative source vector x x0 . We are interested in the case where the eld point is many wavelengths away from the well-localised sources, i.e., when the following inequalities k x x0

k |x x0 |

hold. Then we may to a good approximation replace h(1) with the rst term in n its asymptotic expansion: h(1) (k |x x0 |) (i)n+1 n eik|xx0 | k |x x0 | (8.36)

and replace jn with the rst term in its power series expansion: jn (k x x0 )

Inserting these expansions into Equation (8.34), we obtain the multipole expansion of the Fourier component of the Hertz vector e

n=0

e (n)

2n n! k x x0 (2n + 1)!

(8.33a) (8.33b)

(8.34)

)e

im

dx

(8.35)

(8.37)

(8.38a)

118

E LECTROMAGNETIC R ADIATION

AND

R ADIATING S YSTEMS

where

This expression is approximately correct only if certain care is exercised; if many e (n) terms are needed for an accurate result, the expansions of the spherical Hankel and Bessel functions used above may not be consistent and must be replaced by more accurate expressions. Taking the inverse Fourier transform of e will yield the Hertz vector in time domain, which inserted into Equation (8.29) on page 115 will yield C. The resulting expression can then in turn be inserted into Equation (8.30) on page 115 in order to obtain the radiation elds. For a linear source distribution along the polar axis, = in expression (8.38b) above, and Pn (cos ) gives the angular distribution of the radiation. In the general case, however, the angular distribution must be computed with the help of Formula (8.32) on the preceding page. Let us now study the lowest order contributions to the expansion of Hertz vector.

Choosing n = 0 in expression (8.38b) above, we obtain e (0) = eik|xx0 | 40 |x x0 | (x ) d3x = 1 eik|xx0 | p 40 |x x0 | (8.39)

where p = V (x ) d3x is the Fourier component of the electric dipole moment p(t, x0 ) = V (x x0 )(t, x ) d3x [cf. Equation (6.2) on page 85 which describes the static dipole moment]. If a spherical coordinate system is chosen with its polar axis along p as in Figure 8.5 on the next page, the components of e (0) are 1 eik|xx0 | p cos 40 |x x0 | 1 eik|xx0 | def e e (0) = p sin 40 |x x0 | e e (0) r = r

def def

e e (0) = 0

Evaluating Formula (8.29) on page 115 for the help vector C, with the spherically polar components (8.40) of e (0) inserted, we obtain

(0) C = C, =

1 40

1 eik|xx0 | ik p sin |x x0 | |x x0 |

e (n) = (i)n

1 eik|xx0 | 2n n! 40 |x x0 | (2n)!

(8.41)

8.2

M ULTIPOLE R ADIATION

119

k x3 x Brad Erad

p r x2

x1 F IGURE 8.5: If a spherical polar coordinate system (r, , ) is chosesn such that the electric dipole moment p (and thus its Fourier transform p ) is located at the origin and orientated along the polar axis, the calculations are simplied.

Applying this to Equation (8.30) on page 115, we obtain directly the Fourier components of the elds

Keeping only those parts of the elds which dominate at large distances (the radiation elds) and recalling that the wave vector k = k(x x0 )/ |x x0 | where k = /c, we can now write down the Fourier components of the radiation parts of the magnetic and electric elds from the dipole: 0 eik|xx0 | 0 eik|xx0 | pw k sin = (p k) (8.43a) 4 |x x0 | 4 |x x0 | 1 eik|xx0 | 1 eik|xx0 | p k2 sin = [(p k) k] Erad = 40 |x x0 | 40 |x x0 | (8.43b) Brad =

B = i

(8.42a)

(8.42b)

120

E LECTROMAGNETIC R ADIATION

AND

R ADIATING S YSTEMS

These elds constitute the electric dipole radiation, also known as E1 radiation.

The next term in the expression (8.38b) on page 118 for the expansion of the Fourier transform of the Hertz vector is for n = 1: e (1) = i eik|xx0 | k x x0 (x ) cos d3x 40 |x x0 | V 1 eik|xx0 | = ik [(x x0 ) (x x0 )] (x ) d3x 40 |x x0 |2 V

Here, the term [(x x0 ) (x x0 )] (x ) can be rewritten [(x x0 ) (x x0 )] (x ) = (xi x0,i )(xi x0,i ) (x ) and introducing i = xi x0,i (8.46a) (8.46b) (8.45)

i = xi x0,i the jth component of the integrand in e (1) can be broken up into 1 {[(x x0 ) (x x0 )] (x )} j = i , j i + ,i j 2 1 + i , j i ,i j 2

i.e., as the sum of two parts, the rst being symmetric and the second antisymmetric in the indices i, j. We note that the antisymmetric part can be written as 1 1 i , j i ,i j = [, j (i i ) j (i ,i )] 2 2 1 = [ ( ) ( )] j 2 1 = (x x0 ) [ (x x0 )] 2

The utilisation of Equations (8.24) on page 114, and the fact that we are considering a single Fourier component, (t, x) = eit (8.49)

(8.44)

(8.47)

(8.48)

j

8.2

M ULTIPOLE R ADIATION

121

Hence, we can write the antisymmetric part of the integral in Formula (8.44) on the facing page as 1 (x x0 ) (x ) (x x0 ) d3x 2 V 1 = i (x x0 ) j (x ) (x x0 ) d3x 2 V 1 = i (x x0 ) m

where we introduced the Fourier transform of the magnetic dipole moment m = 1 2 (x x0 ) j (x ) d3x (8.52)

The nal result is that the antisymmetric, magnetic dipole, part of e (1) can be written e,antisym

(1)

k eik|xx0 | (x x0 ) m 40 |x x0 |2

In analogy with the electric dipole case, we insert this expression into Equation (8.29) on page 115 to evaluate C, with which Equations (8.30) on page 115 then gives the B and E elds. Discarding, as before, all terms belonging to the near elds and transition elds and keeping only the terms that dominate at large distances, we obtain Brad (x) = 0 eik|xx0 | (m k) k 4 |x x0 | k eik|xx0 | Erad (x) = m k 40 c |x x0 | (8.54a) (8.54b)

which are the elds of the magnetic dipole radiation (M1 radiation).

The symmetric part of the n = 1 contribution in the Equation (8.38b) on page 118 for the expansion of the Hertz vector can be expressed in terms

e,sym (1)

(8.51)

(8.53)

122

E LECTROMAGNETIC R ADIATION

AND

R ADIATING S YSTEMS

of the electric quadrupole tensor, which is dened in accordance with Equation (6.3) on page 86: Q(t, x0 ) = (x x0 )(x x0 )(t, x ) d3x (8.55)

Again we use this expression in Equation (8.29) on page 115 to calculate the elds via Equations (8.30) on page 115. Tedious, but fairly straightforward algebra (which we will not present here), yields the resulting elds. The radiation components of the elds in the far eld zone (wave zone) are given by i0 eik|xx0 | k Q k 8 |x x0 | i eik|xx0 | Erad (x) = k Q k k 80 |x x0 | Brad (x) =

The derivation of the radiation elds for the case of the source moving relative to the observer is considerably more complicated than the stationary cases studied above. In order to handle this non-stationary situation, we use the retarded potentials (3.36) on page 44 in Chapter 3 (tret , x ) 3 1 dx 40 V |x x | j(tret , x ) 3 0 dx A(t, x) = 4 V |x x | (t, x) = (8.57a) (8.57b)

and consider a source region with such a limited spatial extent that the charges and currents are well localised. Specically, we consider a charge q , for instance an electron, which, classically, can be thought of as a localised, unstructured and rigid charge distribution with a small, nite radius. The part of this charge distribution dq which we are considering is located in dV = d3x in the sphere in Figure 8.6 on the facing page. Since we assume that the electron (or any other other similar electric charge) is moving with a velocity v whose direction is arbitrary and whose magnitude can be almost comparable to the speed of light, we cannot say that the charge and current to be used in (8.57) is

Q

(8.56a) (8.56b)

8.3

R ADIATION

FROM A

L OCALISED C HARGE

IN

A RBITRARY M OTION

123

x(t)

dr dV q

F IGURE 8.6: Signals which are observed at the eld point x at time t were generated at source points x (t ) on a sphere, centred on x and expanding, as time increases, with the velocity c outward from the centre. The source charge element moves with an arbitrary velocity v and gives rise to a source leakage out of the source volume dV = d3x .

) d3x and V v(tret , x ) d3x , respectively, because in the nite time interval during which the observed signal is generated, part of the charge distribution will leak out of the volume element d3x .

The charge distribution in Figure 8.6 on page 123 which contributes to the eld at x(t) is located at x (t ) on a sphere with radius r = |x x | = c(t t ). The radius interval of this sphere from which radiation is received at the eld point x during the time interval (t , t + dt ) is (r , r + dr ) and the net amount of charge in this radial interval is dq = (tret , x ) dS dr (tret , x ) (x x ) v dS dt |x x | (8.58)

where the last term represents the amount of source leakage due to the fact that the charge distribution moves with velocity v(t ). Since dt = dr /c and

V (tret , x

v(t ) xx x (t )

dS

124

E LECTROMAGNETIC R ADIATION

AND

R ADIATING S YSTEMS

dS dr = d3x we can rewrite this expression for the net charge as dq = (tret , x ) d3x (tret , x ) (x x ) v 3 dx c |x x | (x x ) v = (tret , x ) 1 d3x c |x x | dq 1 (xx )v c|xx |

(8.59)

or (tret , x ) d3x =

This is the expression to be used in the Formulae (8.57) on page 122 for the retarded potentials. The result is (recall that j = v) (t, x) = A(t, x) = 1 40 dq |x x | (xx )v c v dq |x x | (xx )v c (8.62a) (8.62b)

For a sufciently small and well localised charge distribution we can, assuming that the integrands do not change sign in the integration volume, use the mean value theorem and the fact that V dq = q to evaluate these expressions to become 1 q 1 q = 40 |x x | (xx )v 40 s c v v v q q A(t, x) = = = (t, x) 40 c2 |x x | (xx )v 40 c2 s c2 c (t, x) = where s = xx (x x ) v c xx v = xx 1 |x x | c v xx = (x x ) |x x | c

0 4

(8.60)

(8.63a) (8.63b)

8.3

R ADIATION

FROM A

L OCALISED C HARGE

IN

A RBITRARY M OTION

125

is the retarded relative distance. The potentials (8.63) are precisely the LinardWiechert potentials which we derived in Section 4.3.2 on page 60 by using a covariant formalism. It is important to realise that in the complicated derivation presented here, the observer is in a coordinate system which has an absolute meaning and the velocity v is that of the particle, whereas in the covariant derivation two frames of equal standing were moving relative to each other with v. Expressed in the four-potential, Equation (4.48) on page 59, the Linard-Wiechert potentials become

The Linard-Wiechert potentials are applicable to all problems where a spatially localised charge emits electromagnetic radiation, and we shall now study such emission problems. The electric and magnetic elds are calculated from the potentials in the usual way: B(t, x) = A(t, x) (8.66a) (8.66b)

Consider a localised charge q and assume that its trajectory is known experimentally as a function of retarded time x = x (t ) (8.67)

(in the interest of simplifying our notation, we drop the subscript ret on t from now on). This means that we know the trajectory of the charge q , i.e., x , for all times up to the time t at which a signal was emitted in order to precisely arrive at the eld point x at time t. Because of the nite speed of propagation of the elds, the trajectory at times later than t is not (yet) known. The retarded velocity and acceleration at time t are given by v(t ) = dx dt dv d2 x = dt dt 2 (8.68a) (8.68b)

a(t ) = v(t ) =

As for the charge coordinate x itself, we have in general no knowledge of the velocity and acceleration at times later than t , in particular not at the time of

A (x ) =

q 40

1 v , s cs

= (, cA)

(8.65)

126

E LECTROMAGNETIC R ADIATION

AND

R ADIATING S YSTEMS

? |x x | v c

q x (t )

x0 (t) 0

v(t )

xx

x x0 x(t)

F IGURE 8.7: Signals which are observed at the eld point x at time t were generated at the source point x (t ). After time t the particle, which moves with nonuniform velocity, has followed a yet unknown trajectory. Extrapolating tangentially the trajectory from x (t ), based on the velocity v(t ), denes the virtual simultaneous coordinate x0 (t).

observation t. If we choose the eld point x as xed, application of (8.68) to the relative vector x x yields d (x x (t )) = v(t ) dt d2 (x x (t )) = (t ) v dt 2 (8.69a) (8.69b)

The retarded time t can, at least in principle, be calculated from the implicit relation t = t (t, x) = t |x x (t )| c (8.70)

and we shall see later how this relation can be taken into account in the calculations. According to Formulae (8.66) on the preceding page the electric and magnetic elds are determined via differentiation of the retarded potentials at the observation time t and at the observation point x. In these formulae the unprimed , i.e., the spatial derivative differentiation operator = xi /xi means that we differentiate with respect to the coordinates x = (x1 , x2 , x3 ) while keeping t xed, and the unprimed time derivative operator /t means that we differentiate with respect to t while keeping x xed. But the Linard-Wiechert

8.3

R ADIATION

FROM A

L OCALISED C HARGE

IN

A RBITRARY M OTION

127

potentials and A, Equations (8.63) on page 124, are expressed in the charge velocity v(t ) given by Equation (8.68a) on page 125 and the retarded relative distance s(t , x) given by Equation (8.64) on page 124. This means that the expressions for the potentials and A contain terms which are expressed explicitly in t , which in turn is expressed implicitly in t via Equation (8.70) on the facing page. Despite this complication it is possible, as we shall see below, to determine the electric and magnetic elds and associated quantities at the time of observation t. To this end, we need to investigate carefully the action of differentiation on the potentials.

We introduce the convention that a differential operator embraced by parentheses with an index x or t means that the operator in question is applied at constant x and t, respectively. With this convention, we nd that

Furthermore, by applying the operator (/t)x to Equation (8.70) on the preceding page we nd that

This is an algebraic equation in (t /t)x which we can solve to obtain t t = |x x | |x x | = s |x x | (x x ) v/c (8.73)

where s = s(t , x) is the retarded relative distance given by Equation (8.64) on page 124. Making use of Equation (8.73) above, we obtain the following useful operator identity

t t

|x x | s

t t

|x x (t (t, x))| c x x t |x x | = 1 t x c t (x x ) v t = 1+ c |x x | t x = 1 t

Q

x

x x (t ) =

x x (t ) =

xx |x x | t

(x x ) v |x x |

(8.71)

(8.72)

(8.74)

128

E LECTROMAGNETIC R ADIATION

AND

R ADIATING S YSTEMS

(8.75)

which gives the following operator relation when ( )t is acting on an arbitrary function of t and x:

With the help of the rules (8.77) and (8.74) we are now able to replace t by t in the operations which we need to perform. We nd, for instance, that 1 q 40 s xx v x x s q = 40 s2 |x x | c cs t x 0 q v A A = t t x t 4 s s q = x x s x x v v 2 s3 40 c t

Utilising these relations in the calculation of the E eld from the LinardWiechert potentials, Equations (8.63) on page 124, we obtain E(t, x) = A t q (x x ) |x x | v/c = 2 40 s |x x |

Starting from expression (8.64a) on page 124 for the retarded relative distance

(x x ) |x x | v/c cs

( )t =

s t

( )t = ( )t t

+ ( )t =

xx cs

+ ( )t

x

(8.77)

(8.78a)

(8.78b)

(8.79) |x x | v 2 c

8.3

R ADIATION

FROM A

L OCALISED C HARGE

IN

A RBITRARY M OTION

129

s(t , x), we see that we can evaluate (s/t )x in the following way

(x x ) v v2 (x x ) v + c c |x x |

where Equation (8.71) on page 127 and Equations (8.68) on page 125, respectively, were used. Hence, the electric eld generated by an arbitrarily moving charged particle at x (t ) is given by the expression

Radiation eld

The rst part of the eld, the velocity eld, tends to the ordinary Coulomb eld when v 0 and does not contribute to the radiation. The second part of the eld, the acceleration eld, is radiated into the far zone and is therefore also called the radiation eld. From Figure 8.7 on page 126 we see that the position the charged particle would have had if at t 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 x (t ) is x0 (t), the virtual simultaneous coordinate. During the arbitrary motion, we interpret x x0 as the coordinate of the eld point x relative to the virtual simultaneous coordinate x0 (t). Since the time it takes from a signal to propagate (in the assumed vacuum) from x (t ) to x is |x x | /c, this relative vector is given by x x0 = (x x ) |x x | v c (8.82)

E(t, x) =

q v2 (x x0 ) 1 2 3 40 s c

+ (x x )

(x x0 ) v 2 c

y

x iw

xx + 2 c

|x x | v (x x ) v c

x iw

v d

E(t, x) =

q 40 s3

(x x )

|x x | v c

v2 c2

1 x x (t ) t c

x x (t ) v(t ) v (x x ) t t

xx

s t

(x x ) v c

(8.80)

(8.81)

(8.83)

130

E LECTROMAGNETIC R ADIATION

AND

R ADIATING S YSTEMS

where we made use of Equation (8.63) on page 124 and Formula (8.74) on page 127. But, according to (8.78a), q xx xx ( )t = v c |x x | 40 c2 s2 |x x | so that B(t, x) = xx c |x x xx = c |x x ( )t E(t, x) A t (8.85)

The radiation part of the electric eld is obtained from the acceleration eld in Formula (8.81) on the previous page as Erad (t, x) = =

|xx |

lim E(t, x)

where in the last step we again used Formula (8.82) on the preceding page. Using this formula and Formula (8.86) above, the radiation part of the magnetic eld can be written Brad (t, x) = xx Erad (t, x) c |x x | (8.88)

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 of such a quantity is the retarded relative distance s(t , x). According to Equation (8.64) on page 124, the square of this retarded relative can be written

s (t , x) = x x

(x x ) v (x x ) v + 2 xx c c

q |x x | v (x x ) (x x ) v 2 s3 40 c c q = (x x ) [(x x0 ) v] 40 c2 s3

B(t, x) =

xx cs t A xx q xx v = 40 c2 s2 |x x | c |x x | t x A ( ) t A = ( )t A

A

x

(8.84)

(8.86)

(8.87)

(8.89)

8.3

R ADIATION

FROM A

L OCALISED C HARGE

IN

A RBITRARY M OTION

131

we nd that

Furthermore, from Equation (8.82) on page 129, we obtain the following identity: (x x ) v = (x x0 ) v which, when inserted into Equation (8.91), yields the relation (8.92)

Inserting the above into expression (8.89) on the preceding page for s2 , this expression becomes

where in the penultimate step we used Equation (8.82) on page 129. What we have just demonstrated is that, in the case the particle velocity at time t can be calculated or projected, the retarded distance s in the LinardWiechert potentials (8.63) can be expressed in terms of the virtual simultaneous coordinate x0 (t), viz., the point at which the particle will have arrived at

|x x0 |2

(x x0 ) v c

(x x0 (t)) v = (x x0 (t))2 c

= (x x )

|x x | v c

(x x0 ) v c

s = xx

(x x ) v |x x |2 v2 (x x0 ) v 2 xx + c c2 c

(x x ) v c

(x x0 ) v |x x |2 v2 = c2 c

(x x ) v c

(x x ) v |x x |2 v2 2 c c

(x x ) v c

(x x ) v c

(8.90)

(8.91)

(8.93)

(8.94)

132

E LECTROMAGNETIC R ADIATION

AND

R ADIATING S YSTEMS

time t, i.e., when we obtain the rst knowledge of its existence at the source point x at the retarded time t , and in the eld coordinate 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 special case we are able to express the retarded relative distance as s = s(t, x) and we do not have to involve the retarded time t or transformed differential operators in our calculations. Taking the square root of both sides of Equation (8.94) on the previous page, we obtain the following alternative nal expressions for the retarded relative distance s in terms of the charges virtual simultaneous coordinate x 0 (t):

v v = 2 (x x0 ) + (x x0 ) c c which we shall use in the following example of a uniformly, unaccelerated motion of the charge.

E XAMPLE 8.1 T HE

FIELDS FROM A UNIFORMLY MOVING CHARGE

In the special case of uniform motion, the localised charge moves in a eld-free, isolated 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. This gives us the possibility to extrapolate its position at the observation time, x (t), from its posi tion at the retarded time, x (t ). Since the particle is not accelerated, v 0, the virtual simultaneous coordinate x0 will be identical to the actual simultaneous coordinate of the particle at time t, i.e., x0 (t) = x (t). As depicted in Figure 8.7 on page 126, the angle between x x0 and v is 0 while then angle between x x and v is . We note that in the case of uniform velocity v, time and space derivatives are closely

V WG

= 2 (x x0 ) 1

v2 c2

vv (x x0 ) c2

s2 =

|x x0 |2 1

v2 c2

(x x0 ) v c

|x x0 |

= |x x0 |

1

2

v2 2 sin 0 c2

v2 1 2 c

(x x0 ) v + c

s(t, x) =

|x x0 |2

(x x0 ) v c

(8.96)

8.3

R ADIATION

FROM A

L OCALISED C HARGE

IN

A RBITRARY M OTION

133

related in the following way when they operate on functions of x(t) : v (8.97) t Hence, the E and B elds can be obtained from Formulae (8.66) on page 125, with the potentials given by Equations (8.63) on page 124 as follows: 1 v v A = 2 = 2 E = t c t c t v v vv (8.98a) = 1 2 = + c c c vv = 2 1 c v v v B = A = 2 = 2 = 2 c c c v v v v vv = 2 = 2 2 1 (8.98b) c c c c c v = 2 E c Here 1 = xi xi is the unit dyad and we used the fact that v v 0. What remains is just to express in quantities evaluated at t and x. From Equation (8.63a) on page 124 and Equation (8.96) on the preceding page we nd that q q 1 = = s2 40 s 80 s3 (8.99) v v q (x x0 ) (x x0 ) + = 40 s3 c c When this expression for is inserted into Equation (8.98a), the following result q vv vv 1 s2 E(t, x) = 2 1 = 3 c2 c 80 s v v q (x x0 ) + (x x0 ) = 3 40 s c c

q v v v2 = (x x0 ) + (x x0 ) (x x0 ) 2 40 s3 c c c

From Equation (8.100) we conclude that E is directed along the vector from the simul-

q v2 (x x0 ) 1 2 3 40 s c follows. Of course, the same result also follows from Equation (8.83) on page 129 with v 0 inserted.

"

v v (x x0 ) c c

h f 7

v v v vv v (x x0 ) 2 (x x0 ) c c c c c

f7

e " !

6e

(8.100)

134

E LECTROMAGNETIC R ADIATION

AND

R ADIATING S YSTEMS

taneous coordinate x0 (t) to the eld (observation) coordinate x(t). In a similar way, the magnetic eld can be calculated and one nds that

From these explicit formulae for the E and B elds we can discern the following cases: 1. v 0 E goes over into the Coulomb eld ECoulomb 2. v 0 B goes over into the Biot-Savart eld 3. v c E becomes dependent on 0 4. v c, sin 0 0 E (1 v2 /c2 )ECoulomb 5. v c, sin 0 1 E (1 v2 /c2 )1/2 ECoulomb

E ND

OF EXAMPLE

E XAMPLE 8.2

T HE

Let us consider in more detail the treatment of the radiation from a uniformly moving rigid charge distribution. If we return to the original denition of the potentials and the inhomogeneous wave equation, Formula (3.19) on page 41, for a generic potential component (t, x) and a generic source component f (t, x),

we nd that under the assumption that v = v x1 , this equation can be written 1 v2 c2 2 2 2 + + = f (x) x2 x2 x2 1 2 3 (8.103)

2 = x 2 3 = x 3

1 =

x1 1 v2 /c2

"

(t, x) =

1 2 c2 t2

"

B(t, x) =

0 q 4s3

v2 c2

v (x x0 ) =

1 vE c2

(8.101)

8.1

"

(t, x) = f (t, x)

(8.102)

def

(/1 , /2 , /3 ),

8.3

R ADIATION

FROM A

L OCALISED C HARGE

IN

A RBITRARY M OTION

135

2 ()

= f( 1 4

in this space. This equation has the well-known Coulomb potential solution () = f ( ) 3 d f (x ) 3 dx s (8.106)

After inverse transformation back to the original coordinates, this becomes (x) = 1 4 (8.107)

Applying this to the explicit scalar and vector potential components, realising that for a rigid charge distribution moving with velocity v the current is given by j = v, we obtain 1 (x ) 3 dx 40 V s 1 v v(x ) 3 d x = 2 (t, x) A(t, x) = 40 c2 V s c (t, x) = For a localised charge where d3x = q , these expressions reduce to q 40 s qv A(t, x) = 40 c2 s (t, x) = (8.109a) (8.109b)

which we recognise as the Linard-Wiechert potentials; cf. Equations (8.63) on page 124. We notice, however, that the derivation here, based on a mathematical technique which in fact is a Lorentz transformation, is of more general validity than the one leading to Equations (8.63) on page 124. Let us now consider the action of the elds produced from a moving, rigid charge distribution represented by q moving with velocity v, on a charged particle q, also moving with velocity v. This force is given by the Lorentz force F = q(E + v B) (8.111)

With the help of Equation (8.101) on the facing page and Equations (8.109) above, and the fact that t = v [cf.. Formula (8.97) on page 133], we can rewrite expres-

"

((

((

1 v2 /c2 1 , 2 , 3 ) f ()

(8.105)

[(x2 x2 ) + (x3 x3 ) ]

1 2

(8.108)

(8.110a) (8.110b)

136

E LECTROMAGNETIC R ADIATION

AND

R ADIATING S YSTEMS

Applying the bac-cab rule, Formula (F.54) on page 164, on the last term yields v v v v v2 = c c c c c2 which means that we can write F = q where = 1 v2 c2 (8.115) (8.114)

The scalar function is called the convection potential or the Heaviside potential. When the rigid charge distribution is well localised so that we can use the potentials (8.110) the convection potential becomes = 1 v2 c2 q 40 s (8.116)

The convection potential from a point charge is constant on attened ellipsoids of revolution, dened through Equation (8.108) on the previous page as

= 2 (x1 x1 )2 + (x2 x2 )2 + (x3 x3 )2 = Const These Heaviside ellipsoids are equipotential surfaces, and since the force is proportional to the gradient of , which means that it is perpendicular to the ellipsoid surface, the force between two charges is in general not directed along the line which connects the charges. A consequence of this is that a system consisting of two comoving charges connected with a rigid bar, will experience a torque. This is the idea behind the Trouton-Noble experiment, aimed at measuring the absolute speed of the earth or the galaxy. The negative outcome of this experiment is explained by the special theory of relativity which postulates that mechanical laws follow the same rules as electromagnetic laws, so that a compensating torque appears due to mechanical stresses within the charge-bar system.

E ND

OF EXAMPLE

x1 x 1

1 v2 /c2

+ (x2 x2 )2 + (x3 x3 )2

6e

=q

f7

f7

F = q E+v

v E c2

v v v v c c c c

(8.112)

"

"

(8.113)

(8.117)

8.2

8.3

R ADIATION

FROM A

L OCALISED C HARGE

IN

A RBITRARY M OTION

137

If the charge moves at such low speeds that v/c simplies to s = xx (x x ) v xx , c |x x | v xx , c 1, Formula (8.64) on page 124

so that the radiation eld Equation (8.87) on page 130 can be approximated by Erad (t, x) = q 40 c2 |x x |3 (x x ) [(x x ) v], v c (8.120)

from which we obtain, with the use of Formula (8.86) on page 130, the magnetic eld Brad (t, x) = q [ (x x )], v 4c3 |x x |2 v c (8.121)

It is interesting to note the close correspondence which exists between the nonrelativistic elds (8.120) and (8.121) and the electric dipole eld Equations (8.43) on page 119 if we introduce p = q x (t ) and at the same time make the transitions q v = p 2 p (8.123a) (8.123b) (8.122)

x x = x x0

The power ux in the far zone is described by the Poynting vector as a function of Erad and Brad . We use the close correspondence with the dipole case to nd that it becomes S= v 0 q 2 ( )2 xx sin2 2 c |x x |2 |x x | 16 (8.124)

where is the angle between v and x x0 . The total radiated power (integrated over a closed spherical surface) becomes P= q 2 v2 0 q 2 ( )2 v = 6c 60 c3 (8.125)

(8.118)

138

E LECTROMAGNETIC R ADIATION

AND

R ADIATING S YSTEMS

which is the Larmor formula for radiated power from an accelerated charge. Note that here we are treating a charge with v c but otherwise totally unspecied motion while we compare with formulae derived for a stationary oscillating dipole. The electric and magnetic elds, Equation (8.120) on the previous page and Equation (8.121) on the preceding page, respectively, and the expressions for the Poynting ux and power derived from them, are here instantaneous values, dependent on the instantaneous position of the charge at x (t ). The angular distribution is that which is frozen to the point from which the energy is radiated.

8.3.3 Bremsstrahlung

An important special case of radiation is when the velocity v and the acceler ation v are collinear (parallel or anti-parallel) so that v v = 0. This condition (for an arbitrary magnitude of v) inserted into expression (8.87) on page 130 for the radiation eld, yields Erad (t, x) = q (x x ) [(x x ) v], 40 c2 s3 v v (8.126)

from which we obtain, with the use of Formula (8.86) on page 130, the magnetic eld Brad (t, x) = q |x x | [ (x x )], v 40 c3 s3 v v (8.127)

The difference between this case and the previous case of v c is that the approximate expression (8.118) on the previous page for s is no longer valid; we must instead use the correct expression (8.64) on page 124. The angular distribution of the power ux (Poynting vector) therefore becomes S= sin2 0 q 2 v 2 162 c |x x |2 1 v cos c

6

xx |x x |

(8.128)

It is interesting to note that the magnitudes of the electric and magnetic elds are the same whether v and v 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 xed surface in space. The radiated power into a solid angle element d, measured relative to the particles retarded position, is given by the formula dU rad () 0 q 2 v 2 sin2 d = S (x x ) x x d = dt 162 c 1 v cos c

d (8.129)

8.3

R ADIATION

FROM A

L OCALISED C HARGE

IN

A RBITRARY M OTION

139

v = 0.5c

v = 0.25c v=0 v

F IGURE 8.8: Polar diagram of the energy loss angular distribution factor sin2 /(1 v cos /c)5 during bremsstrahlung for particle speeds v = 0, v = 0.25c, and v = 0.5c.

On the other hand, the radiation loss due to radiation from the charge at retarded time t :

Inserting Equation (8.128) on the facing page for S into (8.131), we obtain the explicit expression for the energy loss due to radiation evaluated at the retarded time

The angular factors of this expression, for three different particle speeds, are plotted in Figure 8.8. Comparing expression (8.129) on the facing page with expression (8.132) above, we see that they differ by a factor 1 v cos /c which comes from the extra factor s/ |x x | introduced in (8.131). Let us explain this in geometrical terms. During the interval (t , t + dt ) and within the solid angle element d the particle radiates an energy [dU rad ()/dt ] dt d. As shown in 8.9 this energy

dU rad dU rad d = dt dt

t t

d

x

(8.130)

(8.132)

140

E LECTROMAGNETIC R ADIATION

AND

R ADIATING S YSTEMS

dS dr x d q x2 vdt x1 x x2 + c dt

F IGURE 8.9: Location of radiation between two spheres as the charge moves with velocity v from x1 to x2 during the time interval (t , t + dt ). The observation point (eld point) is at the xed location x.

is at time t located between two spheres, one outer with its origin in x1 (t ) and one inner with its origin in x1 (t + dt ) = x1 (t ) + v dt and radius c[t (t + dt )] = c(t t dt ). From Figure 8.9 we see that the volume element subtending the solid angle element x x2

is d3x = dS dr = x x2

Here, dr denotes the differential distance between the two spheres and can be evaluated in the following way

v cos

x x2 v x x2

dt =

cs dt x x2

x iw

dr = x x2 + c dt

d =

dS

2

d dr

x x2 v dt x x2 x x2

((

((

(8.133)

(8.134)

(8.135)

8.3

R ADIATION

FROM A

L OCALISED C HARGE

IN

A RBITRARY M OTION

141

where Formula (8.64) on page 124 was used in the last step. Hence, the volume element under consideration is s d3x = dS dr = dS c dt (8.136) x x2 We see that the energy which is radiated per unit solid angle during the time interval (t , t + dt ) is located in a volume element whose size is dependent. This explains the difference between expression (8.129) on page 138 and expression (8.132) on page 139. Let the radiated energy, integrated over , be denoted U rad . After tedious, but relatively straightforward integration of Formula (8.132) on page 139, one obtains dU rad 0 q 2 v2 = dt 6c 1

3

If we know v(t ), we can integrate this expression over t and obtain the total energy radiated during the acceleration or deceleration of the particle. This way we obtain a classical picture of bremsstrahlung (braking radiation). Often, an atomistic treatment is required for an acceptable result.

Calculate the bremsstrahlung when a charged particle, moving at a non-relativistic speed, is accelerated or decelerated during an innitely short time interval. We approximate the velocity change at time t = t0 by a delta function: v(t ) = v (t t0 ) which means that v =

From the general expression (8.86) on page 130 we conclude that E B and that it sufces to consider E Erad . According to the bremsstrahlung expression for Erad , Equation (8.126) on page 138, E= q sin v (t t0 ) 40 c2 |x x |

((

((

((

s xx

1 v2 c

(8.137)

((

E XAMPLE 8.3

(8.138)

(8.139)

(8.140)

(8.142)

142

E LECTROMAGNETIC R ADIATION

AND

R ADIATING S YSTEMS

Fourier transforming expression (8.142) on the preceding page for E is trivial, yielding E = 82 q sin v eit0 2 0 c |x x | (8.144)

We note that the magnitude of this Fourier component is independent of . This is a consequence of the innitely short impulsive step (t t0 ) in the time domain which produces an innite spectrum in the frequency domain. The total radiation energy is given by the expression

1 0

EB dt d2x = E 2 dt d2x

1 0 c

= 0 c

According to Parsevals identity [cf. Equation (7.35) on page 105] the following equality holds:

E 2 dt = 4

0

|E |2 d

For our innite spectrum, Equation (8.144), we obtain q 2 (v)2 rad U d = 163 0 c3 = = q 163 0 c3

2

(v)2

2 0

sin2 2 d x d |x x |2

d

0 2

rad We see that the energy spectrum U is independent of frequency . This means that if we integrate it over all frequencies [0, ], a divergent integral would result. In reality, all spectra have nite widths, with an upper cutoff limit set by the quantum

"

q2 30 c

v c

d 2

"

rad U d = 40 c

|E |2 d2x d

sin2 sin d d

"

U rad =

dU rad dt = dt

((

((

(8.143)

B 0

dS dt

E 2 dt d2x

(8.145)

(8.146)

(8.147)

(8.148)

8.3

R ADIATION

FROM A

L OCALISED C HARGE

IN

A RBITRARY M OTION

143

condition 1 (8.149) h = m(v)2 2 which expresses that the highest possible frequency in the spectrum is that for which all kinetic energy difference has gone into one single eld quantum (photon) with energy h. If we adopt the picture that the total energy is quantised in terms of N photons radiated during the process, we nd that rad U d = dN h or, for an electron where q = |e|, where e is the elementary charge, (8.150)

where we used the value of the ne structure constant e2 /(40 hc) 1/137.

Even if the number of photons becomes innite when 0, these photons have negligible energies so that the total radiated energy is still nite.

E ND OF

EXAMPLE

Formula (8.86) and Formula (8.87) on page 130 for the magnetic eld and the radiation part of the electric eld are general, valid for any kind of motion of the localised charge. A very important special case is circular motion, i.e., the case v v. With the charged particle orbiting in the x1 x2 plane as in Figure 8.10 on the following page, an orbit radius a, and an angular frequency 0 , we obtain (t ) = 0 t x (t ) = a[ x1 cos (t ) + x2 sin (t )] v(t ) = x (t ) = a0 [ x1 sin (t ) + x2 cos (t )] v(t ) = x (t ) = a2 [ x1 cos (t ) + x2 sin (t )] 0 v = | | = v a2 0 v = |v| = a0 (8.152a) (8.152b) (8.152c) (8.152d) (8.152e) (8.152f)

Because of the rotational symmetry we can, without loss of generality, rotate our coordinate system around the x3 axis so the relative vector x x from the source point to an arbitrary eld point always lies in the x2 x3 plane, i.e., x x = x x ( x2 sin + x3 cos )

"

"

dN =

2 e2 40 hc 3

v c

d 1 2 137 3

v c

(8.151)

8.3

(8.153)

144

E LECTROMAGNETIC R ADIATION

AND

R ADIATING S YSTEMS

x2 (t, x) x v q a 0 v (t ) x1 (t , x ) xx

x3 F IGURE 8.10: Coordinate system for the radiation from a charged particle at x (t ) in circular motion with velocity v(t ) along the tangent and constant acceleration v(t ) toward the origin. The x1 x2 axes are chosen so that the relative eld point vector x x makes an angle with the x3 axis which is normal to the plane of the orbital motion. The radius of the orbit is a.

where is the angle between x x and the normal to the plane of the particle orbit (see Figure 8.10). From the above expressions we obtain (x x ) v = x x v sin cos (x x ) v = x x v sin sin = x x v cos

where in the last step we simply used the denition of a scalar product and the fact that the angle between v and x x is . The power ux is given by the Poynting vector, which, with the help of Formula (8.86) on page 130, can be written S= 1 1 xx (E B) = |E|2 0 c0 |x x | (8.155)

Inserting this into Equation (8.131) on page 139, we obtain dU rad (, ) |x x | s 2 = |E| dt c0 (8.156)

(8.154a) (8.154b)

8.3

R ADIATION

FROM A

L OCALISED C HARGE

IN

A RBITRARY M OTION

145

where the retarded distance s is given by expression (8.64) on page 124. With the radiation part of the electric eld, expression (8.87) on page 130, inserted, and using (8.154a) and (8.154b) on the preceding page, one nds, after some algebra, that

2 2 v v dU rad (, ) 0 q 2 v2 1 c sin cos 1 c2 sin sin = (8.157) 5 dt 162 c 1 v sin cos c

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 (xed) angle so that d = sin d d. Integration of Equation (8.157) over this d gives, after some cumbersome algebra, the angular integrated expression dU rad 0 q 2 v2 = dt 6c 1

2

In Equation (8.157) above, two limits are particularly interesting: 1. v/c 1 which corresponds to cyclotron radiation. 1 which corresponds to synchrotron radiation.

Cyclotron radiation

For a non-relativistic speed v c, Equation (8.157) reduces to (8.159) dU rad (, ) 0 q 2 v2 (1 sin2 sin2 ) = 2c dt 16 But, according to Equation (8.154b) on the preceding page sin2 sin2 = cos2 (8.160)

where is dened in Figure 8.10 on the facing page. This means that we can write 0 q 2 v 2 dU rad () 0 q 2 v2 = (1 cos2 ) = sin2 dt 162 c 162 c (8.161)

Consequently, a xed observer near the orbit plane will observe cyclotron radiation twice per revolution in the form of two equally broad pulses of radiation with alternating polarisation.

2. v/c

1 v2 c

(8.158)

146

E LECTROMAGNETIC R ADIATION

AND

R ADIATING S YSTEMS

(t, x) xx

x2

v q a 0 v (t ) x1 (t , x )

x3 F IGURE 8.11: When the observation point is in the plane of the particle orbit, i.e., = /2 the lobe width is given by .

Synchrotron radiation

When the particle is relativistic, v c, the denominator in Equation (8.157) on the previous page becomes very small if sin cos 1, which denes the forward direction of the particle motion ( /2, 0). Equation (8.157) on the preceding page then becomes dU rad (/2, 0) 0 q 2 v2 1 = 2c dt 16 1 v c

which means that an observer near the orbit plane sees a very strong pulse followed, half an orbit period later, by a much weaker pulse. The two cases represented by Equation (8.161) on the previous page and Equation (8.162) are very important results since they can be used to determine 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 ( = /2), Equation (8.157) on the previous page gives

2 v v dU rad (/2, ) 0 q 2 v2 1 c cos 1 c2 sin = 5 dt 162 c 1 v cos c

(8.162)

(8.163)

8.3

R ADIATION

FROM A

L OCALISED C HARGE

IN

A RBITRARY M OTION

147

Hence, the angle 0 is a measure of the synchrotron radiation lobe width ; see Figure 8.11 on the facing page. For ultra-relativistic particles, dened by

1 v2 c

Hence, synchrotron radiation from ultra-relativistic charges is characterized by a radiation lobe width which is approximately 1 (8.167)

This angular interval is swept by the charge during the time interval t = 0 (8.168)

in the direction toward the observer who therefore measures a pulse width of length

y iw v x

t = t

l vt v v v 1 = t = 1 t = 1 1 c c c c 0 c 0 v v 2 1 c 1+ c 1 v 1 1 1 = 1 2 = 3 v 0 c 20 2 0 1+ c 1/2 2

y iw v x

sin 0 =

1,

v2 c2

1,

(8.165)

148

E LECTROMAGNETIC R ADIATION

AND

R ADIATING S YSTEMS

(8.170) As a general rule, 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 = 23 0 (8.171) t A spectral analysis of the radiation pulse will exhibit Fourier components n 0 from n = 1 up to n 23 . When N electrons are contributing to the radiation, we can discern between three situations: max 1. All electrons are very close to each other so that the individual phase differences are negligible. The power will be multiplied by N 2 relative to a single electron and we talk about coherent radiation. 2. The electrons are perfectly evenly distributed in the orbit. This is the case, for instance, for electrons in a circular current in a conductor. In this case the radiation elds cancel completely and no far elds are generated. 3. The electrons are unevenly distributed in the orbit. This happens for an open ring current which is subject to uctuations of order N as for all open systems. As a result we get incoherent radiation. Examples of this can be found both in earthly laboratories and under cosmic conditions.

We recall that the general expression for the radiation E eld from a moving charge concentration is given by expression (8.87) on page 130. This expression in Equation (8.156) on page 144 yields the general formula

Integration over the solid angle gives the totally radiated power as

2 dU rad 0 q 2 v2 1 v2 sin c = 3 dt 6c v2 1 c2

2

where is the angle between v and v. In the limit v v, sin = 0, which corresponds to bremsstrahlung. For v v, sin = 1, which corresponds to cyclotron radiation or synchrotron radiation.

(x x )

|x x | v c

(8.172)

(8.173)

8.3

R ADIATION

FROM A

L OCALISED C HARGE

IN

A RBITRARY M OTION

149

vt q b 0 |x x0 | B E z

F IGURE 8.12: The perpendicular eld of a charge q moving with velocity v = v x is E z.

v = vx

Virtual photons

According to Formula (8.100) on page 133 and Figure 8.12,

Utilising expression (8.95a) on page 132 and simple geometrical relations, we can rewrite this as E = b q 40 2 (vt)2 + b2 /2

3/2

This represents a contracted eld, approaching the eld of a plane wave. The passage of this eld pulse corresponds to a frequency distribution of the eld energy. Fourier transforming, we obtain

Here, K1 is the Kelvin function (Bessel function of the second kind with imaginary argument) which behaves in such a way for small and large arguments that E, q 42

0 bv

b v

E, 0,

42

0 bv

E, =

1 2

E (t) eit dt =

E = Ez =

q 40 s3

v2 c2

(x x0 ) x3

(8.174)

(8.175)

b K1 v

b v

(8.176)

(8.177a) (8.177b)

150

E LECTROMAGNETIC R ADIATION

AND

R ADIATING S YSTEMS

showing that the pulse length is of the order b/(v). Due to the equipartition of the eld energy into the electric and magnetic elds, the total eld energy can be written U = 0

2 E d3x = 0 bmax 2 E vdt 2b db

(8.178)

where the volume integration is over the plane perpendicular to v. With the use of Parsevals identity for Fourier transforms, Formula (7.35) on page 105, we can rewrite this as U=

0

U d = 40 v

v/ bmin

where an explicit value of bmin can be calculated in quantum theory only. As in the case of bremsstrahlung, it is intriguing to quantise the energy into photons [cf. Equation (8.150) on page 143]. Then we nd that

where = e2 /(40 hc) 1/137 is the ne structure constant. Let us consider the interaction of two electrons, 1 and 2. The result of this interaction is that they change their linear momenta from p1 to p1 and p2 to p2 , respectively. Heisenbergs uncertainty principle gives bmin h/ p1 p1 so that the number of photons exchanged in the process is of the order

Since this change in momentum corresponds to a change in energy h = E 1 2 , we see that E1 and E1 = m0 c N d 2 ln E1 cp1 cp1 m0 c2 E 1 E 1

N d

c 2 ln p1 p1 h

(8.182)

(8.183)

N d

2 c ln bmin

q2 v ln 2 v 2 0 bmin

q2 2 2 0 v

db d b

bmin

E,

bmax

bmin

d 2b db (8.179)

(8.180)

(8.181)

8.3

R ADIATION

FROM A

L OCALISED C HARGE

IN

A RBITRARY M OTION

151

When electromagnetic radiation is propagating through matter, new phenomena may appear which are (at least classically) not present in vacuum. As mentioned earlier, one can under certain simplifying assumptions include, to some extent, the inuence from matter on the electromagnetic elds by introducing new, derived eld quantities D and H according to D = (t, x)E = 0 E B = (t, x)H = m 0 H (8.184) (8.185)

Expressed in terms of these derived eld quantities, the Maxwell equations, often called macroscopic Maxwell equations, take the form D = (t, x) E+ B = 0 t B = 0 H D = j(t, x) t (8.186a) (8.186b) (8.186c) (8.186d)

Assuming for simplicity that the electric permittivity and the magnetic permeability , and hence the relative permittivity and the relative permeability m all have xed values, independent on time and space, for each type of material we consider, we can derive the general telegraphers equation [cf. Equation (2.33) on page 29] E 2 E 2 E 2 = 0 2 t t (8.187)

describing (1D) wave propagation in a material medium. In Chapter 2 we concluded that the existence of a nite conductivity, manifesting itself in a collisional interaction between the charge carriers, causes the waves to decay exponentially with time and space. Let us therefore assume that in our medium = 0 so that the wave equation simplies to 2 E 2 E 2 = 0 2 t If we introduce the phase velocity in the medium as 1 c 1 = v = = 0 m 0 m (8.189) (8.188)

152

E LECTROMAGNETIC R ADIATION

AND

R ADIATING S YSTEMS

where, according to Equation (1.9) on page 5, c = 1/ 0 0 is the speed of light, i.e., the phase speed of electromagnetic waves in vacuum, then the general solution to each component of Equation (8.188) on the preceding page Ei = f ( v t) + g( + v t), c def = m = c n v i = 1, 2, 3 (8.190)

The ratio of the phase speed in vacuum and in the medium (8.191)

is called the refractive index of the medium. In general n is a function of both time and space as are the quantities , , , and m themselves. If, in addition, the medium is anisotropic or birefringent, all these quantities are rank-two tensor elds. Under our simplifying assumptions, in each medium we consider n = Const for each frequency component of the elds. Associated with the phase speed of a medium for a wave of a given frequency we have a wave vector, dened as k k k = k = v

def

v v v

(8.192)

As in the vacuum case discussed in Chapter 2, assuming that E is time-harmonic, i.e., can be represented by a Fourier component proportional to exp{it}, the solution of Equation (8.188) can be written E = E0 ei(kxt) (8.193)

where now k is the wave vector in the medium given by Equation (8.192) above. With these denitions, the vacuum formula for the associated magnetic eld, Equation (2.40) on page 30, 1 1 kE = kE B = k E = v (8.194)

is valid also in a material medium (assuming, as mentioned, that n has a xed constant scalar value). A consequence of a = 1 is that the electric eld will, in general, have a longitudinal component. It is important to notice that depending on the electric and magnetic properties 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: v = c = n k (8.195)

8.3

R ADIATION

FROM A

L OCALISED C HARGE

IN

A RBITRARY M OTION

153

where, in the last step, we used Equation (8.192) on the preceding 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 = k (8.196)

has a unique value for each frequency component, and is different from v . Except in regions of anomalous dispersion, v is always smaller than c. In a gas of free charges, such as a plasma, the refractive index is given by the expression n2 () = 1 where 2 = p

2 p 2

(8.197)

N q2 0 m

(8.198)

is the plasma frequency. Here m and N denote the mass and number density, respectively, of charged particle species . In an inhomogeneous plasma, N = 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 frequency, 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 while vg 0 and the wave Fourier component at is reected there.

Vavilov-Cerenkov radiation

As we saw in Subsection 8.1, a charge in uniform, rectilinear motion in vacuum does not give rise to any radiation; see in particular Equation (8.98a) on page 133. Let us now consider a charge in uniform, rectilinear motion in a medium with electric properties which are different from those of a (classical) vacuum. Specically, consider a medium where = Const > 0 = 0 This implies that in this medium the phase speed is v = c 1 = <c n 0 (8.200) (8.199a) (8.199b)

154

E LECTROMAGNETIC R ADIATION

AND

R ADIATING S YSTEMS

Hence, in this particular medium, the speed of propagation of (the phase planes 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 |v|, which, of course, are below the phase speed in vacuum, can experience that the particle speeds are higher than the phase speed in the medium. This is the basis for the Vavilov-Cerenkov 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 Linard-Wiechert potentials, Equations (8.63) on page 124, we realise that we obtain the latter in the medium by a simple formal replacement c c/n in the expression (8.64) on page 124 for s. Hence, the Linard-Wiechert potentials in a medium characterized by a refractive index n, are

where now

The need for the absolute value of the expression for s is obvious in the case when v/c 1/n because then the second term can be larger than the rst term; if v/c 1/n we recover the well-known vacuum case but with modied phase speed. We also note that the retarded and advanced times in the medium are [cf. Equation (3.34) on page 43] k |x x | |x x | n = t c k |x x | |x x | n tadv = tadv (t, x x ) = t + = t+ c tret = tret (t, x x ) = t

so that the usual time interval t t between the time measured at the point of observation and the retarded time in a medium becomes tt = |x x | n c (8.204)

s = xx n

(x x ) v c

A(t, x) =

1 qv 1 qv = 2 (xx )v 40 c |x x | n 40 c2 s c

(t, x) =

q 1 q 1 = 40 |x x | n (xx )v 40 s c

(8.201a)

(8.201b)

(8.202)

(8.203a) (8.203b)

8.3

R ADIATION

FROM A

L OCALISED C HARGE

IN

A RBITRARY M OTION

155

x(t)

c x (t )

F IGURE 8.13: Instantaneous picture of the expanding eld spheres from a point charge moving with constant speed v/c > 1/n in a medium where n > 1. This generates a Vavilov-Cerenkov shock wave in the form of a cone.

For v/c 1/n, the retarded distance s, and therefore the denominators in Equations (8.201) on the facing page vanish when n(x x ) nv v = xx cos c = x x c c

In the direction dened by this angle c , the potentials become singular. During the time interval t t given by expression (8.204) on the preceding page, the eld exists within a sphere of radius |x x | around the particle while the particle moves a distance l = v(t t ) (8.207)

along the direction of v. In the direction c where the potentials are singular, all eld spheres are tangent to a straight cone with its apex at the instantaneous position of the particle and with the apex half angle c dened according to sin c = cos c = c nv (8.208)

(8.205)

156

E LECTROMAGNETIC R ADIATION

AND

R ADIATING S YSTEMS

This cone of potential singularities and eld sphere circumferences propagates with speed c/n in the form of a shock front, called Vavilov-Cerenkov radiation.1 The Vavilov-Cerenkov cone is similar in nature to the Mach cone in acoustics. In order to make some quantitative estimates of this radiation, we note that we can describe the motion of each charged particle q as a current density: j = q v (x vt ) = q v (x vt )(y )(z ) x1 (8.209)

which has the trivial Fourier transform q ix /v j = e (y )(z ) x1 (8.210) 2 This Fourier component can be used in the formulae derived for a linear current in Subsection 8.1.1 if only we make the replacements 0 = n 2 0 (8.211a) n k (8.211b) c In this manner, using j from Equation (8.210), the resulting Fourier trans forms of the Vavilov-Cerenkov magnetic and electric radiation elds can be calculated from the expressions (7.11) and (7.22) on page 100, respectively. The total energy content is then obtained from Equation (7.35) on page 105 (integrated over a closed sphere at large distances). For a Fourier component one obtains [cf. Equation (7.38) on page 105]

dx

1 The rst systematic exploration of this radiation was made by P. A. Cerenkov in 1934, who was then a post-graduate student in S. I. Vavilovs research group at the Lebedev Institute in Moscow. Vavilov wrote a manuscript with the experimental ndings, put Cerenkov as the author, and submitted it to Nature. In the manuscript, Vavilov explained the results in terms of radioactive particles creating Compton electrons which gave rise to the radiation (which was the correct interpretation), but the paper was rejected. The paper was then sent to Physical Review and was, after some controversy with the American editors who claimed the results to be wrong, eventually published in 1937. In the same year, I. E. Tamm and I. M. Frank published the theory for the effect (the singing electron). In fact, predictions of a similar effect had been made as early as 1888 by Heaviside, and by Sommerfeld in his 1904 paper Radiating body moving with velocity of light. On May 8, 1937, Sommerfeld sent a letter to Tamm via Austria, saying that he was surprised that his old 1904 ideas were now becoming interesting. Tamm, Frank and Cerenkov received the Nobel Prize in 1958 for the discovery and the interpretation of the Cerenkov effect [V. L. Ginzburg, private communication]. The rst observation of this type of radiation was reported by Marie Curie in 1910, but she never pursued the exploration of it [7].

q 2 n2 = 163 0 c3

x kx cos exp i v

rad U d

1 40 nc

(j k)eikx d3x

d

2

(8.212) sin d

2

8.3

R ADIATION

FROM A

L OCALISED C HARGE

IN

A RBITRARY M OTION

157

where is the angle between the direction of motion, x1 , and the direction to the observer, k. The integral in (8.212) is singular of a Dirac delta type. If we limit the spatial extent of the motion of the particle to the closed interval [X, X] on the x axis we can evaluate the integral to obtain

which has a maximum in the direction c as expected. The magnitude of this maximum grows and its width narrows as X . The integration of (8.213) over therefore picks up the main contributions from c . Consequently, we can set sin2 sin2 c and the result of the integration is

The integrand in (8.214) is strongly peaked near = c/(nv), or, equivalently, near cos c = c/(nv). This means that the integrand function is practically zero outside the integration interval [1, 1]. Hence, one can extend the integration interval to (, ) without introducing too much an error. Via yet another variable substitution we can therefore approximate sin c

2

leading to the nal approximate result for the total energy loss in the frequency interval (, + d)

As mentioned earlier, the refractive index is usually frequency dependent. Realising this, we nd that the radiation energy per frequency unit and per unit length is

rad U d q 2 = 2X 40 c2

c2 n2 ()v2

q 2X rad U d = 20 c2

c2 n2 v2

cX 1 2 2 n n v

c2

V G

1 + nv c

F oU

F iU

sin2

1 + nv c

X v 2

d 1

V G

F iU

1 + nv c

c2 n2 v2

F oU

q 2 n2 sin2 c 22 0 c3

sin2

1 + nv c

X v 2

cX n

rad U = 2

R

Q

rad U d =

X v 2

(8.213)

Q

1 1

rad U () d

(8.214)

sin2 x dx 2 x

(8.215)

(8.216)

(8.217)

158

E LECTROMAGNETIC R ADIATION

AND

R ADIATING S YSTEMS

This result was derived under the assumption that v/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 , so there exist al ways a highest frequency for which we can obtain Vavilov-Cerenkov radiation from a fast charge in a medium. Our derivation above for a xed value of n is valid for each individual Fourier component.

Bibliography

[1] R. B ECKER, Electromagnetic Fields and Interactions, Dover Publications, Inc., New York, NY, 1982, ISBN 0-486-64290-9. [2] M. B ORN AND E. W OLF, Principles of Optics. Electromagnetic Theory of Propagation, Interference and Diffraction of Light, sixth ed., Pergamon Press, Oxford,. . . , 1980, ISBN 0-08-026481-6. [3] V. L. G INZBURG, Applications of Electrodynamics in Theoretical Physics and Astrophysics, Revised third ed., Gordon and Breach Science Publishers, New York, London, Paris, Montreux, Tokyo and Melbourne, 1989, ISBN 288124-719-9. [4] J. D. JACKSON, Classical Electrodynamics, third ed., John Wiley & Sons, Inc., New York, NY . . . , 1999, ISBN 0-471-30932-X. [5] J. B. M ARION AND M. A. H EALD, Classical Electromagnetic Radiation, second ed., Academic Press, Inc. (London) Ltd., Orlando, . . . , 1980, ISBN 0-12472257-1. [6] 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. [7] J. S CHWINGER , L. L. D E R AAD , J R ., K. A. M ILTON , AND W. T SAI, Classical Electrodynamics, Perseus Books, Reading, MA, 1998, ISBN 0-7382-0056-5. [8] J. A. S TRATTON, Electromagnetic Theory, McGraw-Hill Book Company, Inc., New York, NY and London, 1953, ISBN 07-062150-0. [9] J. VANDERLINDE, Classical Electromagnetic Theory, John Wiley & Sons, Inc., New York, Chichester, Brisbane, Toronto, and Singapore, 1993, ISBN 0-47157269-1.

F

Formulae

F.1 The Electromagnetic Field

F.1.1 Maxwells equations

D = B t H = j+ D t E = B = 0 (F.1) (F.2) (F.3) (F.4)

Constitutive relations

D = E B H= (F.5) (F.6)

j = E P = 0 E

(F.7) (F.8)

Vector and scalar potentials

B= A (F.9) A t (F.10)

E =

159

160

F ORMULAE

A+ 1 =0 c2 t (F.11)

Poyntings vector

S = EH (F.12)

1 T i j = Ei D j + Hi B j i j (Ek Dk + Hk Bk ) 2 (F.13)

F.2.1 Relationship between the eld vectors in a plane wave

B= kE c (F.14)

i0 eik|x| d3 x eikx j k 4 |x| V i eik|x| Erad (x) = d3 x eikx j k x 40 c |x| V Brad (x) = (F.15) (F.16)

0 eik|x| p k 4 |x| 1 eik|x| (p k) k Erad (x) = 40 |x| Brad (x) = (F.17) (F.18)

F.3

S PECIAL R ELATIVITY

161

Brad (x) = 0 eik|x| (m k) k 4 |x| k eik|x| Erad (x) = m k 40 c |x| (F.19) (F.20)

s = x x (x x ) t t |x x | s

v c

x x0 = (x x ) |x x | =

v c

s=

|x x0 |2

(x x0 ) v c

E(t, x) =

q v2 (x x0 ) 1 2 40 s3 c v E(t, x) B(t, x) = c2

E(t, x) =

Q

Brad (x) =

(F.21) (F.22)

+ (x x )

(x x0 ) v c2

(F.23) (F.24)

(F.28) (F.29)

(F.30)

162

F ORMULAE

F.3.1 Metric tensor

1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1

v = g v (F.32)

ds = c dt = c d (F.37)

F.3.5 Four-velocity

u = dx = (c, v) d (F.38)

F.3.6 Four-momentum

p = m0 cu =

E ,p c

0 0 0 0 = 0 0 1 0 0 0 0 1 1 = 1 2 v = c

hh ig

x = x

deef

deef

g =

hh ig

(F.31)

(F.33)

(F.34)

(F.35) (F.36)

(F.39)

F.4

V ECTOR R ELATIONS

163

j = 0 u (F.40)

F.3.8 Four-potential

Let x be the radius vector (coordinate vector) from the origin to the point (x1 , x2 , x3 ) (x, y, z) and let |x| denote the magnitude (length) of x. Let further (x), (x), . . . be arbitrary scalar elds and a(x), b(x), c(x), d(x), . . . arbitrary vector elds. The differential vector operator is in Cartesian coordinates given by xi

i=1 3

def def xi xi xi

where xi , i = 1, 2, 3 is the ith unit vector and x1 x, x2 y, and x3 z. In component (tensor) notation can be written

i

Base vectors

r = sin cos x1 + sin sin x2 + cos x3 = cos cos x1 + cos sin x2 sin x3 (F.45a) (F.45b) (F.45c)

= sin x1 + cos x2

= i =

, , x1 x2 x3

, , x y z

deef

F = A A =

0 E x /c Ey /c Ez /c E x /c 0 Bz By Ey /c Bz 0 B x Ez /c By Bx 0

hh ig

A =

,A c

(F.41)

(F.42)

(F.43)

(F.44)

164

F ORMULAE

x1 = sin cos + cos cos sin r x2 = sin sin + cos sin + cos r x3 = cos sin r

dx x = dl = dr r + r d + r sin d (F.47)

d = sin d d (F.48)

d2x n = dS = dS r = r2 d r (F.49)

Volume element

d3x = dV = drdS = r2 dr d (F.50)

General vector algebraic identities

a b = b a = i j ai b j = ab cos i jk a j bk xi

a b = b a =

a (b c) + b (c a) + c (a b) = 0

a (b c) = b(a c) c(a b)

a (b c) = (a b) c

F.4

V ECTOR R ELATIONS

165

() = + (a) = a + a (a) = a a (F.58) (F.59) (F.60) (F.61) (F.62) (F.63) (F.64) (F.65) (F.66)

2

(a b) = b ( a) a ( b)

(a b) = a ( b) + b ( a) + (b )a + (a )b =

2

(a b) = a( b) b( a) + (b )a (a )b

( a) = ( a)

( a) = 0

=0

(F.67)

Special identities

In the following k is an arbitrary constant vector while x, as before, is the radius vector. x = 0 x |x| = |x| 1 x = 3 |x| |x| x 1 = 2 = 4(x) 3 |x| |x| kx k 1 = k = 3 |x| |x| |x| x kx = if |x| = 0 k |x|3 |x|3 k 1 2 =k 2 = 4k(x) |x| |x| (k a) = k( a) + k ( a) (k a) x = 3 (F.68) (F.69) (F.70) (F.71) (F.72) (F.73) (F.74) (F.75) (F.76)

Integral relations

Let V(S ) be the volume bounded by the closed surface S (V). Denote the 3dimensional volume element by d3x( dV) and the surface element, directed

166

F ORMULAE

along the outward pointing surface normal unit vector n, by dS( d2x n). Then ( a) d3x = dS a (F.77) (F.78) (F.79)

If S (C) is an open surface bounded by the contour C(S ), whose line element is dl, then dl =

C

Bibliography

[1] G. B. A RFKEN AND H. J. W EBER, Mathematical Methods for Physicists, fourth, international ed., Academic Press, Inc., San Diego, CA . . . , 1995, ISBN 0-12059816-7. [2] P. M. M ORSE AND H. F ESHBACH, Methods of Theoretical Physics, Part I. McGraw-Hill Book Company, Inc., New York, NY . . . , 1953, ISBN 07-043316-8. [3] 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.

a dl =

dS dS ( a)

( a) d3x =

( ) d3x =

dS

S

dS a

(F.80) (F.81)

M

Mathematical Methods

M.1 Scalars, Vectors and Tensors

Every physical observable can be described by a geometric object. We will describe the observables in classical electrodynamics mathematically in terms of scalars, pseudoscalars, vectors, pseudovectors, tensors or pseudotensors and will not exploit differential forms to any signicant degree. A scalar describes a scalar quantity which may or may not be constant in time and/or space. A vector describes some kind of physical motion due to vection and a tensor describes the motion or deformation due to some form of tension. However, generalisations to more abstract notions of these quantities are commonplace. The difference between a scalar, vector and tensor and a pseudoscalar, pseudovector and a pseudotensor is that the latter behave differently under such coordinate transformations which cannot be reduced to pure rotations. Throughout we adopt the convention that Latin indices i, j, k, l, . . . run over the range 1, 2, 3 to denote vector or tensor components in the real Euclidean three-dimensional (3D) conguration space 3 , and Greek indices , , , , . . . , which are used in four-dimensional (4D) space, run over the range 0, 1, 2, 3.

M.1.1 Vectors

Radius vector

Any vector can be represented mathematically in several different ways. One suitable representation is in terms of an ordered N-tuple, or row vector, of the coordinates xN where N is the dimensionality of the space under consideration. The most basic vector is the radius vector which is the vector from the origin to the point of interest. Its N-tuple representation simply enumerates the coordinates which describe this point. In this sense, the radius vector from the origin to a point is synonymous with the coordinates of the point itself.

167

168

M ATHEMATICAL M ETHODS

In the 3D space 3 , we have N = 3 and the radius vector can be represented by the triplet (x1 , x2 , x3 ) of coordinates xi , i = 1, 2, 3. The coordinates xi are scalar quantities which describe the position along the unit base vectors x i which span 3 . Therefore a representation of the radius vector in 3 is x = x i xi x i xi

def i=1 3

where we have introduced Einsteins summation convention (E) which states that a repeated index in a term implies summation over the range of the index in question. Whenever possible and convenient we shall in the following always assume E and suppress explicit summation in our formulae. Typographically, we represent a 3D vector by a boldface letter or symbol in a Roman font. Alternatively, we may describe the radius vector in component notation as follows: xi (x1 , x2 , x3 ) (x, y, z)

def

This component notation is particularly useful in 4D space where we can represent the radius vector either in its contravariant component form x (x0 , x1 , x2 , x3 ) or its covariant component form x (x0 , x1 , x2 , x3 )

def def

The relation between the covariant and contravariant forms is determined by the metric tensor (also known as the fundamental tensor) whose actual form is dictated by the physics. The dual representation of vectors in contravariant and covariant forms is most convenient when we work in a non-Euclidean vector space with an indenite metric. An example is Lorentz space 4 which is a 4D Riemannian space. 4 is often utilised to formulate the special theory of relativity. We note that for a change of coordinates x x = x (x0 , x1 , x2 , x3 ), due to a transformation from a system to another system , the differential radius vector dx transforms as dx = x dx x

which follows trivially from the rules of differentiation of x considered as functions of four variables x .

(M.1)

(M.2)

(M.3)

(M.4)

(M.5)

M.1

S CALARS , V ECTORS

AND

T ENSORS

169

M.1.2 Fields

A eld is a physical entity which depends on one or more continuous parameters. Such a parameter can be viewed as a continuous index which enumerates the coordinates of the eld. In particular, in a eld which depends on the usual radius vector x of 3 , each point in this space can be considered as one degree of freedom so that a eld is a representation of a physical entity which has an innite number of degrees of freedom.

Scalar elds

def

(x0 , x1 , x2 , x3 ) (x )

def

(M.7)

which indicates that the four-scalar depends on all four coordinates spanning this space. Since a four-scalar has the same value at a given point regardless of coordinate system, it is also called an invariant. Analogous to the transformation rule, Equation (M.5) on the facing page, for the differential dx , the transformation rule for the differential operator /x under a transformation x x becomes x = x x x which, again, follows trivially from the rules of differentiation. (M.8)

Vector elds

a(x) = xi ai (x) In component notation this same vector can be represented as ai (x) = (a1 (x), a2 (x), a3 (x)) = ai (x j )

as follows: (M.9)

(M.10)

This eld describes how the scalar quantity varies continuously in 3D space. In 4D, a four-scalar eld is denoted

by (M.6)

3

170

M ATHEMATICAL M ETHODS

In 4D, an arbitrary four-vector eld in contravariant component form can be represented as a (x ) = (a0 (x ), a1 (x ), a2 (x ), a3 (x )) or, in covariant component form, as a (x ) = (a0 (x ), a1 (x ), a2 (x ), a3 (x )) (M.12) (M.11)

where x is the radius four-vector. Again, the relation between a and a is determined by the metric of the physical 4D system under consideration. Whether an arbitrary N-tuple fulls the requirement of being an (N-dimensional) contravariant vector or not, depends on its transformation properties during a change of coordinates. For instance, in 4D an assemblage y = (y0 , y1 , y2 , y3 ) constitutes a contravariant four-vector (or the contravariant components of a four-vector) if and only if, during a transformation from a system with coordinates x to a system with coordinates x , it transforms to the new system according to the rule y= x y x (M.13)

i.e., in the same way as the differential coordinate element dx transforms according to Equation (M.5) on page 168. The analogous requirement for a covariant four-vector is that it transforms, during the change from to , according to the rule y = x y x (M.14)

i.e., in the same way as the differential operator /x transforms according to Equation (M.8) on the preceding page.

Tensor elds

We denote an arbitrary tensor eld in 3 by A(x). This tensor eld can be represented in a number of ways, for instance in the following matrix form: A(x) = Ai j (xk )

where, in the last member, we have again used the more compact component notation. Strictly speaking, the tensor eld described here is a tensor of rank two.

A11 (x) A12 (x) A13 (x) A21 (x) A22 (x) A23 (x) A31 (x) A32 (x) A33 (x)

def

(M.15)

fd

M.1

S CALARS , V ECTORS

AND

T ENSORS

171

Another common and useful tensor is the fully antisymmetric tensor of rank 3, also known as the Levi-Civita tensor 1 0 1 if i, j, k is an even permutation of 1,2,3 if at least two of i, j, k are equal if i, j, k is an odd permutation of 1,2,3

i jk

i jk ilm

In fact, tensors may have any rank n. In this picture a scalar is considered to be a tensor of rank n = 0 and a vector a tensor of rank n = 1. Consequently, the notation where a vector (tensor) is represented in its component form is called the tensor notation. A tensor of rank n = 2 may be represented by a twodimensional array or matrix whereas higher rank tensors are best represented in their component forms (tensor notation).

T ENSORS IN 3D SPACE

Consider a tetrahedron-like volume element V of a solid, uid, or gaseous body, whose atomistic structure is irrelevant for the present analysis; gure M.1 on the next page indicates how this volume may look like. Let dS = d2x n be the directed surface ele ment of this volume element and let the vector T n d2x be the force that matter, lying on the side of d2x toward which the unit normal vector n points, acts on matter which lies on the opposite side of d2x. This force concept is meaningful only if the forces are short-range enough that they can be assumed to act only in the surface proper. According to Newtons third law, this surface force fulls T n = T n (M.20)

Using (M.20) and Newtons second law, we nd that the matter of mass m, which at a

`a

= jl km jm kl

fd

(i j ) =

1 0 0 0 1 0 0 0 1

ac

i j =

0 if i = j 1 if i = j

(M.16)

(M.17)

(M.18)

(M.19)

E XAMPLE M.1

172

M ATHEMATICAL M ETHODS

x3

d2x x2 V

x1 F IGURE M.1: Terahedron-like volume element V containing matter. given instant is located in V obeys the equation of motion T n d2x cos 1 T x1 d2x cos 2 T x2 d2x cos 3 T x3 d2x + Fext = ma (M.21)

where Fext is the external force and a is the acceleration of the volume element. In other words m Fext T n = n 1 T x1 + n 2 T x 2 + n 3 T x 3 + 2 a (M.22) dx m Since both a and Fext /m remain nite whereas m/d2x 0 as V 0, one nds that in this limit T n = ni T xi n i T x i

i=1 3

From the above derivation it is clear that Equation (M.23) is valid not only in equilibrium but also when the matter in V is in motion. Introducing the notation T i j = T xi (M.24)

for the jth component of the vector T xi , we can write Equation (M.23) above in com

"

(M.23)

M.1

S CALARS , V ECTORS

AND

T ENSORS

173

i=1 3

(M.25)

Using Equation (M.25), we nd that the component of the vector T n in the direction of an arbitrary unit vector m is

j=1

j=1

i=1

Hence, the jth component of the vector T xi , here denoted T i j , can be interpreted as the i jth component of a tensor T. Note that T nm 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 d dt v dm =

V V

f d3x +

S

T n d2x

The jth component of this equation can be written d v j dm = dt f j d3x + T nj d2x = f j d3x + ni T i j d2x (M.28)

where, in the last step, Equation (M.25) above was used. Setting dm = d 3x and using the divergence theorem on the last term, we can rewrite the result as

V

d v j d3x = dt

f j d3x +

T i j 3 dx xi

Since this formula is valid for any arbitrary volume, we must require that T i j d vj fj =0 dt xi T i j v j + v v j f j =0 t xi (M.30)

Note that v j /t is the rate of change with time of the velocity component v j at a xed point x = (x1 , x1 , x3 ).

E ND OF

EXAMPLE

= T nj m j =

T nm = T n m

ni T i j

m j ni T i j m j = n T m

(M.26)

(M.27)

(M.29)

M.1

174

M ATHEMATICAL M ETHODS

In 4D, we have three forms of four-tensor elds of rank n. We speak of a contravariant four-tensor eld, denoted A1 2 ...n (x ), a covariant four-tensor eld, denoted A1 2 ...n (x ), a mixed four-tensor eld, denoted A1 2...nk (x ). k+1 The 4D metric tensor (fundamental tensor) mentioned above is a particularly important four-tensor of rank 2. In covariant component form we shall denote it g . This metric tensor determines the relation between an arbitrary contravariant four-vector a and its covariant counterpart a according to the following rule: a (x ) g a (x )

def ...

(M.32)

This rule is often called lowering of index. The raising of index analogue of the index lowering rule is: a (x ) g a (x )

def

(M.33)

More generally, the following lowering and raising rules hold for arbitrary rank n mixed tensor elds:

...k1 k gk k A1 2k+2 ...n (x ) = A1 2 ...k1 (x ) k k+1 ...n k+1 2 ...k1 ...k1 gk k A1k+1 ...n (x ) = A1 2k+2 ...nk (x ) k+1 k

(M.34) (M.35)

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 2 tensor in contravariant form yields A = g g A (M.36)

i.e., the same rank 2 tensor in covariant form. This operation is also known as a tensor contraction.

E XAMPLE M.2

4

FLAT

L ORENTZ SPACE

M.1

S CALARS , V ECTORS

AND

T ENSORS

175

metric tensor

i.e., a matrix with a main diagonal that has the sign sequence, or signature, {+, , , } or

Consider an arbitrary contravariant four-vector a in this space. In component form it can be written: a (a0 , a1 , a2 , a3 ) = (a0 , a)

def

According to the index lowering rule, Equation (M.32) on the facing page, we obtain the covariant version of this vector as a (a0 , a1 , a2 , a3 ) = g a In the {+, , , } metric we obtain =0: =1: =2: =3: or a = (a0 , a1 , a2 , a3 ) = (a0 , a1 , a2 , a3 ) = (a0 , a) Analogously, using the {+, , , } metric we obtain (M.47)

0 def

a3 = 0 a + 0 a + 0 a + 1 a = a

a2 = 0 a0 + 0 a1 1 a2 + 0 a3 = a2

0 1 2 3 3

a1 = 0 a 1 a + 0 a + 0 a = a

a 0 = 1 a 0 + 0 a 1 + 0 a 2 + 0 a 3 = a0

1 2 3

@@ AB8

2445

(g ) =

qr

g =

1 if = = 0 1 if = = i = j = 1, 2, 3 0 if =

@@ 8B

2445

(g ) =

qr rt rt s

g =

1 if = = 0 1 if = = i = j = 1, 2, 3 0 if =

(M.37)

(M.38)

(M.39)

(M.40)

(M.41)

(M.42)

(M.43)

1

176

M ATHEMATICAL M ETHODS

E ND OF

(M.48)

EXAMPLE

M.2

Scalar product

The scalar product (dot product, inner product) of two arbitrary 3D vectors a and b in ordinary 3 space is the scalar number a b = x i ai x j b j = x i x j ai b j = i j ai b j = ai bi

where we used the fact that the scalar product xi x j is a representation of the Kronecker delta i j dened in Equation (M.16) on page 171. In Russian literature, the scalar product is often denoted (ab). In 4D space we dene the scalar product of two arbitrary four-vectors a and b in the following way a b = g a b = a b = g a b (M.50)

where we made use of the index lowering and raising rules (M.32) and (M.33). The result is a four-scalar, i.e., an invariant which is independent of in which inertial system it is measured. The quadratic differential form ds2 = g dx dx = dx dx (M.51)

i.e., the scalar product of the differential radius four-vector with itself, is an invariant called the metric. It is also the square of the line element ds which is the distance between neighbouring points with coordinates x and x + dx .

E XAMPLE M.3

6 ),

Using this in Equation (M.52), we see that we can interpret this so that the complex

'

A=

a2 a2 + 2iaR aI R I

'

A2 A A = a2 a2 + 2iaR aI A2 R I

def

def

'

A aR + iaI = aR aR + iaI aI A A

def

def

&

A 3D complex vector A is a vector in 3 (or, if we like, in two real vectors aR and aI in 3 in the following way

'

&

(M.49)

(M.53)

(M.54)

M.1

S CALARS , V ECTORS

AND

T ENSORS

177

a2 + a 2 R I

a2 + a 2 R I

On the other hand, the denition of the scalar product in terms of the inner product of complex vector with its own complex conjugate yields |A|2 A A = a2 + a2 = |A|2 R I with the help of which we can dene the unit vector as A = A= |A| = aR a2 + a 2 R I aR + i aI a2 + a 2 R I aI (M.57)

def

a2 + a 2 R I

a2 + a 2 R I

METRIC IN

L ORENTZ SPACE

In 4 the metric tensor attains a simple form [see Equation (4.10) on page 51 for an example] and, hence, the scalar product in Equation (M.50) on the facing page can be evaluated almost trivially and becomes

4. 0

E ND OF

4

The

'

aR + i

aR

a2 + a 2 R I

aI

a2 + a 2 R I

aI

E ND OF

EXAMPLE

EXAMPLE

'

aR + i

aR

a2 a2 2iaR aI R I

aI

a2 a2 2iaR aI R I

aI

(M.55)

(M.56)

M.3

E XAMPLE M.4

(M.58)

M.4

178

M ATHEMATICAL M ETHODS

E XAMPLE M.5

In the general theory of relativity, several important problems are treated in a 4D spherical polar coordinate system. Then the radius four-vector can be given as x = (ct, r, , ) and the metric tensor is e 0 0 0 0 e 0 0 0 0 r2 0 0 0 0 r2 sin2

where = (ct, r, , ) and = (ct, r, , ). In such a space, the metric takes the form ds2 = c2 e (dt)2 e (dr)2 r2 (d)2 r2 sin2 (d)2 (M.62)

In general relativity the metric tensor is not given a priori but is determined by the Einstein equations.

E ND OF

EXAMPLE

Dyadic product

The dyadic product eld A(x) a(x)b(x) with two juxtaposed vector elds a(x) and b(x) is the outer product of a and b. Operating on this dyad from the right and from the left with an inner product of an vector c one obtains A c ab c a(b c) c A c ab (c a)b

def def def def

i.e., new vectors, proportional to a and b, respectively. In mathematics, a dyadic product is often called tensor product and is frequently denoted a b. In matrix notation the outer product of a and b is written ab = x1 x2 x3

which means that we can represent the tensor A(x) in matrix form as Ai j (xk ) =

which we identify with expression (M.15) on page 170, viz. a tensor in matrix notation.

a1 b1 a1 b2 a1 b3 a1 b2 a2 b2 a2 b3 a1 b3 a3 b2 a3 b3

fd p

a1 b1 a1 b2 a1 b3 a1 b2 a2 b2 a2 b3 a1 b3 a3 b2 a3 b3

x1 x2 x3

@@ B8

fd R

fd

2445

M ETRIC IN

GENERAL RELATIVITY

(g ) =

(M.61)

M.5

(M.63a) (M.63b)

(M.64)

(M.65)

M.1

S CALARS , V ECTORS

AND

T ENSORS

179

Vector product

The vector product or cross product of two arbitrary 3D vectors a and b in ordinary 3 space is the vector c = ab =

Here i jk is the Levi-Civita tensor dened in Equation (M.18) on page 171. Sometimes the vector product of a and b is denoted a b or, particularly in the Russian literature, [ab]. A spatial reversal of the coordinate system (x1 , x2 , x3 ) = (x1 , x2 , x3 ) changes sign of the components of the vectors a and b so that in the new coordinate system a = a and b = b, which is to say that the direction of an ordinary vector is not dependent on the choice of directions of the coordinate axes. On the other hand, as is seen from Equation (M.66) above, 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 axial vector, or pseudovector. A prototype for a pseudovector is the angular momentum vector L = r p and hence the attribute axial. Pseudovectors transform as ordinary vectors under translations and proper rotations, but reverse their sign relative to ordinary vectors for any coordinate change involving reection. Tensors (of any rank) which 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 xi ( x j xk ). This triple product is a representation of the i jk component of the Levi-Civita tensor i jk which is a rank three pseudotensor.

The del operator

In 3 the del operator is a differential vector operator, denoted in Gibbs notation by and dened as

def

where xi is the ith unit vector in a Cartesian coordinate system. Since the operator in itself has vectorial properties, we denote it with a boldface nabla. In component notation we can write

i =

i jk a j bk xi

(M.66)

xi

def xi

(M.67)

, , x1 x2 x3

(M.68)

180

M ATHEMATICAL M ETHODS

We can use this four-del operator to express the transformation properties (M.13) and (M.14) on page 170 as y = x y and

y = x y respectively.

E XAMPLE M.6

T HE

4

L ORENTZ SPACE

2

1 2 c2 t2

"

"

1 , = c t

1 , c t

"

"

In

, , , x0 x1 x2 x3

E ND OF

, , , x0 x1 x2 x3

(M.69)

(M.70)

(M.71)

(M.72)

(M.74)

EXAMPLE

M.6

M.1

S CALARS , V ECTORS

AND

T ENSORS

181

With the help of the del operator we can dene the gradient, divergence and curl of a tensor (in the generalised sense).

The gradient

From this we see that the boldface notation for the nabla and del operators is very handy as it elucidates the 3D vectorial property of the gradient. In 4D, the four-gradient is a covariant vector, formed as a derivative of a four-scalar eld (x ), with the following component form: (x ) = (x ) x (M.77)

RELATIVE DISTANCES IN

3D

Very often electrodynamic quantities are dependent on the relative distance in 3 between two vectors x and x , i.e., on |x x |. In analogy with Equation (M.67) on page 179, we can dene the primed del operator in the following way: = xi = xi (M.78)

Using this, the unprimed version, Equation (M.67) on page 179, and elementary rules of differentiation, we obtain the following two very useful results: |x x | = xi |x x | xx |x x | = = xi xi |x x | xi

(M.79)

and

"

"

1 |x x |

xx = |x x |3

|x x |

1 |x x |

(M.80)

E ND OF

EXAMPLE

&

(x), is an

vector eld

(M.76)

E XAMPLE M.7

M.7

182

M ATHEMATICAL M ETHODS

The divergence

which, as indicated by the notation (x), is a scalar eld in 3 . We may think of the divergence as a scalar product between a vectorial operator and a vector. As is the case for any scalar product, the result of a divergence operation is a scalar. Again we see that the boldface notation for the 3D del operator is very convenient. The four-divergence of a four-vector a is the following four-scalar: a (x ) = a (x ) = a (x ) x

E XAMPLE M.8

D IVERGENCE IN 3D

3

which demonstrates how the primed divergence, dened in terms of the primed del operator in Equation (M.78) on the previous page, works.

E ND OF

EXAMPLE

The Laplacian

The 3D Laplace operator or Laplacian can be described as the divergence of the gradient operator:

2

==

3 2 2 xi x j = i j i j = 2 = 2 2 i xi x j xi i=1 xi

The symbol 2 is sometimes read del squared. If, for a scalar eld (x), 2 < 0 at some point in 3D space, it is a sign of concentration of at that point.

"

"

a(x ) |x x |

&

For an arbitrary

vector eld a(x ), the following relation holds: = a(x ) + a(x ) |x x | 1 |x x | (M.83)

as (M.81)

ai (x) = (x) xi

(M.82)

M.8

(M.84)

M.1

S CALARS , V ECTORS

AND

T ENSORS

183

E ND OF

EXAMPLE

The curl

a(x) =

i jk xi j ak (x) = i jk xi

ak (x) = b(x) x j

where use was made of the Levi-Civita tensor, introduced in Equation (M.18) on page 171. The covariant 4D generalisation of the curl of a four-vector eld a (x ) is the antisymmetric four-tensor eld G (x ) = a (x ) a (x ) = G (x ) A vector with vanishing curl is said to be irrotational. (M.87)

T HE CURL OF

A GRADIENT

Using the denition of the 3 curl, Equation (M.86) above, and the gradient, Equation (M.76) on page 181, we see that [ (x)] = i jk xi j k (x) i jk xi j k (x) (x) xi x j xk (x) x1

i jk

"

2 2 x1 x2 x2 x1

"

2 2 x3 x1 x1 x3

"

2 2 x2 x3 x3 x2

(x) x2 (x) x3

In 3 the curl of a vector eld a(x), denoted a(x), is another b(x) which can be dened in the following way:

"

&

"

1 x | |x

&

3

E XAMPLE M.9

is = 4(x x ) (M.85)

1 x | |x

M.9

vector eld

(M.86)

E XAMPLE M.10

(M.88)

(M.89)

184

M ATHEMATICAL M ETHODS

We thus nd that [ (x)] 0 In 4D we note that for any well-behaved four-scalar eld (x ) ( )(x ) 0 Hence, a gradient is always irrotational.

E ND OF

EXAMPLE

(M.90)

3

M.10

E XAMPLE M.11

T HE

DIVERGENCE OF A CURL

With the use of the denitions of the divergence (M.81) and the curl, Equation (M.86) on the previous page, we nd that [ a(x)] = i [ a(x)]i =

i jk i j ak (x)

(M.92)

Using the denition for the Levi-Civita symbol, dened by Equation (M.18) on page 171, we nd that, due to the assumed well-behavedness of a(x), i

i jk j ak (x)

= =

xi

i jk

ak x j a1 (x)

&

a (x ) = 0

"

2 2 x1 x2 x2 x1

"

2 2 x3 x1 x1 x3

"

2 2 x2 x3 x3 x2

a2 (x) a3 (x)

(M.93)

(M.94)

(M.95)

E ND OF

EXAMPLE

M.11

&

&

(M.91)

3.

M.2

A NALYTICAL M ECHANICS

185

Numerous vector algebra and vector analysis formulae are given in Chapter F. Those which are not found there can often be easily derived by using the component forms of the vectors and tensors, together with the Kronecker and Levi-Civita tensors and their generalisations to higher ranks. A short but very useful reference in this respect is the article by A. Evett [3].

M.2.1 Lagranges equations

As is well known from elementary analytical mechanics, the Lagrange function or Lagrangian L is given by

where qi is the generalised coordinate, T the kinetic energy and V the potential energy of a mechanical system, The Lagrangian satises the Lagrange equations

To the generalised coordinate qi one denes a canonically conjugate momentum pi according to pi = L qi (M.98)

From L, the Hamiltonian (Hamilton function) H can be dened via the Legendre transformation H(pi , qi , t) = pi qi L(qi , qi , t) (M.100)

After differentiating the left and right hand sides of this denition and setting them equal we obtain H H H L L L dpi + dqi + dt = qi dpi + pi dqi dqi dqi dt (M.101) pi qi t qi qi t

L qi

L =0 qi

L(qi , qi , t) = L qi ,

dqi ,t = T V dt

(M.96)

(M.97)

186

M ATHEMATICAL M ETHODS

According to the denition of pi , Equation (M.98) on the preceding page, the second and fourth terms on the right hand side cancel. Furthermore, noting that according to Equation (M.99) on the previous page the third term on the right hand side of Equation (M.101) on the preceding page is equal to pi dqi and identifying terms, we obtain the Hamilton equations: H dqi = qi = pi dt H dpi = pi = qi dt (M.102a) (M.102b)

Bibliography

[1] G. B. A RFKEN AND H. J. W EBER, Mathematical Methods for Physicists, fourth, international ed., Academic Press, Inc., San Diego, CA . . . , 1995, ISBN 0-12059816-7. [2] R. A. D EAN, Elements of Abstract Algebra, John Wiley & Sons, Inc., New York, NY . . . , 1967, ISBN 0-471-20452-8. [3] A. A. E VETT, Permutation symbol approach to elementary vector analysis, American Journal of Physics, 34 (1965), pp. 503507. [4] P. M. M ORSE AND H. F ESHBACH, Methods of Theoretical Physics, Part I. McGraw-Hill Book Company, Inc., New York, NY . . . , 1953, ISBN 07-043316-8. [5] B. S PAIN, Tensor Calculus, third ed., Oliver and Boyd, Ltd., Edinburgh and London, 1965, ISBN 05-001331-9. [6] W. E. T HIRRING, Classical Mathematical Physics, Springer-Verlag, New York, Vienna, 1997, ISBN 0-387-94843-0.

Index

acceleration eld, 129 advanced time, 43 Ampres law, 5 Ampre-turn density, 89 anisotropic, 152 anomalous dispersion, 153 antisymmetric tensor, 63 associated Legendre polynomial, 117 associative, 55 axial gauge, 40 axial vector, 63, 179 Bessel functions, 113 Biot-Savarts law, 6 birefringent, 152 braking radiation, 141 bremsstrahlung, 141, 148 canonically conjugate four-momentum, 71 canonically conjugate momentum, 70, 185 canonically conjugate momentum density, 78 characteristic impedance, 26 classical electrodynamics, 8 closed algebraic structure, 55 coherent radiation, 148 collisional interaction, 151 complex eld six-vector, 20 complex notation, 32 complex vector, 176 component notation, 168 concentration, 182 conservative eld, 11 conservative forces, 75 constitutive relations, 14 contravariant component form, 50, 168 contravariant eld tensor, 63 contravariant four-tensor eld, 174 contravariant four-vector, 170 contravariant four-vector eld, 54 contravariant vector, 50 convection potential, 136 convective derivative, 11 cosine integral, 110 Coulomb gauge, 40 Coulombs law, 2 covariant, 48 covariant component form, 168 covariant eld tensor, 64 covariant four-tensor eld, 174 covariant four-vector, 170 covariant four-vector eld, 54 covariant vector, 50 cross product, 179 curl, 183 cutoff, 142 cyclotron radiation, 145, 148 dAlembert operator, 39, 59, 180 del operator, 179 del squared, 182 differential distance, 52 differential vector operator, 179 Dirac delta, 183 Diracs symmetrised Maxwell equations, 16 dispersive, 153 displacement current, 9 divergence, 182 dot product, 176 dual vector, 50 duality transformation, 16 dummy index, 50 dyadic product, 178 dyons, 20 E1 radiation, 120 E2 radiation, 122 Einstein equations, 178 Einsteins summation convention, 168 electric charge conservation law, 9 electric charge density, 4

187

188

M ATHEMATICAL M ETHODS

electric conductivity, 10 eld Lagrange density, 79 electric current density, 6 eld point, 3 electric dipole moment, 118 eld quantum, 143 electric dipole moment vector, 85 ne structure constant, 143, 150 electric dipole radiation, 120 four-current, 59 electric displacement, 14 four-del operator, 180 electric displacement vector, 87 four-dimensional Hamilton equations, electric eld, 2 71 electric eld energy, 91 four-dimensional vector space, 50 electric monopole moment, 85 four-divergence, 182 electric permittivity, 151 four-gradient, 181 electric polarisation, 86 four-Hamiltonian, 71 electric quadrupole moment tensor, 86 four-Lagrangian, 68 electric quadrupole radiation, 122 four-momentum, 57 electric quadrupole tensor, 122 four-potential, 59 electric susceptibility, 87 four-scalar, 169 electric volume force, 92 four-tensor elds, 174 electrodynamic potentials, 36 four-vector, 54, 170 electromagnetic eld tensor, 63 four-velocity, 57 electromagnetic scalar potential, 37 Fourier component, 25 electromagnetic vector potential, 36 Fourier transform, 41 electromagnetodynamic equations, 16 functional derivative, 77 electromagnetodynamics, 16 fundamental tensor, 50, 168, 174 electromotive force (EMF), 10 Galileos law, 47 electrostatic scalar potential, 35 gauge xing, 40 electrostatics, 1 gauge function, 39 energy theorem in Maxwells theory, gauge invariant, 39 91 gauge transformation, 39 equation of continuity, 9, 59 equation of continuity for magnetic monopoles, Gausss law, 4 general inhomogeneous wave equations, 16 37 equations of classical electrostatics, 8 generalised coordinate, 70, 185 equations of classical magnetostatics, generalised four-coordinate, 71 8 Gibbs notation, 179 Euclidean space, 56 gradient, 181 Euclidean vector space, 51 Green function, 42, 116 Euler-Lagrange equation, 77 group theory, 55 Euler-Lagrange equations, 78 group velocity, 153 Euler-Mascheroni constant, 110 event, 55 Hamilton density, 78 Hamilton density equations, 78 far eld, 98 far zone, 101 Hamilton equations, 70, 186 Faradays law, 11 Hamilton function, 185 eld, 169 Hamilton gauge, 40

189

Hamiltonian, 185 Heaviside potential, 136 Helmholtz theorem, 38 help vector, 115 Hertz method, 114 Hertz vector, 115 Hodge star operator, 16 homogeneous wave equation, 24 Huygens principle, 41 identity element, 55 in a medium, 154 incoherent radiation, 148 indenite norm, 51 index contraction, 50 index lowering, 50 induction eld, 98 inertial reference frame, 47 inertial system, 47 inhomogeneous Helmholtz equation, 41 inhomogeneous time-independent wave equation, 41 inhomogeneous wave equation, 41 inner product, 176 instantaneous, 138 interaction Lagrange density, 79 intermediate eld, 101 invariant, 169 invariant line element, 53 inverse element, 55 irrotational, 4, 183 Kelvin function, 149 kinetic energy, 75, 185 kinetic momentum, 74 Kronecker delta, 171 Lagrange density, 76 Lagrange equations, 185 Lagrange function, 75, 185 Lagrangian, 75, 185 Laplace operator, 182 Laplacian, 182 Larmor formula for radiated power, 138 law of inertia, 47

Legendre polynomial, 116 Legendre transformation, 185 Levi-Civita tensor, 171 Linard-Wiechert potentials, 62, 125, 135 light cone, 53 light-like interval, 53 line element, 176 linear mass density, 76 linearly polarised wave, 30 longitudinal component, 28 Lorentz boost parameter, 57 Lorentz force, 13, 91, 135 Lorentz gauge, 40 Lorentz gauge condition, 38, 60 Lorentz inhomogeneous wave equations, 39 Lorentz potential equations, 39 Lorentz space, 51, 168 Lorentz transformation, 49, 135 lowering of index, 174 M1 radiation, 121 Mach cone, 156 macroscopic Maxwell equations, 151 magnetic charge density, 15 magnetic current density, 15 magnetic dipole moment, 88, 121 magnetic dipole radiation, 121 magnetic eld, 6 magnetic eld energy, 91 magnetic eld intensity, 89 magnetic ux, 11 magnetic ux density, 6 magnetic induction, 6 magnetic monopoles, 15 magnetic permeability, 151 magnetic susceptibility, 89 magnetisation, 88 magnetisation currents, 88 magnetising eld, 14, 89 magnetostatic vector potential, 36 magnetostatics, 5 massive photons, 83 mathematical group, 55

190

M ATHEMATICAL M ETHODS

matrix form, 170 Maxwell stress tensor, 92 Maxwells macroscopic equations, 15, 90 Maxwells microscopic equations, 14 Maxwell-Lorentz equations, 14 mechanical Lagrange density, 79 metric, 168, 176 metric tensor, 50, 168, 174 Minkowski equation, 70 Minkowski space, 56 mixed four-tensor eld, 174 mixing angle, 16 momentum theorem in Maxwells theory, 93 monochromatic, 95 multipole expansion, 114, 117 near zone, 101 Newtons rst law, 47 Newton-Lorentz force equation, 70 non-Euclidean space, 51 non-linear effects, 10 norm, 50, 177 null vector, 53 observation point, 3 Ohms law, 10 one-dimensional wave equation, 29 outer product, 178 Parsevals identity, 105, 142, 150 phase velocity, 151 photon, 143 physical measurable, 32 plane polarised wave, 30 plasma, 153 plasma frequency, 153 Poisson equation, 135 Poissons equation, 35 polar vector, 63, 179 polarisation charges, 87 polarisation currents, 88 polarisation potential, 115 polarisation vector, 115

positive denite, 56 positive denite norm, 51 potential energy, 75, 185 potential theory, 116 power ux, 91 Poynting vector, 91 Poyntings theorem, 91 Proca Lagrangian, 82 propagator, 42 proper time, 53 pseudo-Riemannian space, 56 pseudoscalar, 167 pseudoscalars, 179 pseudotensor, 167 pseudotensors, 179 pseudovector, 63, 167, 179 quadratic differential form, 52, 176 quantum mechanical nonlinearity, 4 radiation eld, 98, 101, 129 radiation elds, 101 radiation gauge, 40 radiation resistance, 110 radius four-vector, 50 radius vector, 167 raising of index, 174 rank, 170 rapidity, 57 refractive index, 152 relative electric permittivity, 92 relative magnetic permeability, 92 relative permeability, 151 relative permittivity, 151 Relativity principle, 48 relaxation time, 25 rest mass density, 79 retarded Coulomb eld, 101 retarded potentials, 44 retarded relative distance, 125 retarded time, 43 Riemannian metric, 52 Riemannian space, 50, 168 row vector, 167 scalar, 167, 182

191

scalar eld, 54, 169 scalar product, 176 shock front, 156 signature, 51, 175 simultaneous coordinate, 132 skew-symmetric, 63 skin depth, 32 source point, 3 space components, 52 space-like interval, 53 space-time, 51 special theory of relativity, 47 spherical Bessel function of the rst kind, 116 spherical Hankel function of the rst kind, 116 spherical waves, 104 standing wave, 108 super-potential, 115 synchrotron radiation, 145, 148 synchrotron radiation lobe width, 147 telegraphers equation, 29, 151 temporal dispersive media, 10 temporal gauge, 40 tensor, 167 tensor contraction, 174 tensor eld, 170 tensor notation, 171 tensor product, 178 three-dimensional functional derivative, 78 time component, 52 time-harmonic wave, 25 time-independent diffusion equation, 26 time-independent telegraphers equation, 30 time-independent wave equation, 26 time-like interval, 53 total charge, 85 transverse components, 29 transverse gauge, 40 vacuum permeability, 5 vacuum permittivity, 2

vacuum polarisation effects, 4 vacuum wave number, 26 Vavilov-Cerenkov radiation, 154, 156 vector, 167 vector product, 179 velocity eld, 129 virtual simultaneous coordinate, 126, 129 wave equations, 23 wave vector, 29, 152 world line, 55 Youngs modulus, 76 Yukawa meson eld, 82

## Mult mai mult decât documente.

Descoperiți tot ce are Scribd de oferit, inclusiv cărți și cărți audio de la editori majori.

Anulați oricând.