Professional Documents
Culture Documents
E LECTROMAGNETIC F IELD T HEORY
Bo Thid
Also available
Electromagnetic
Field Theory
Bo Thid
Swedish Institute of Space Physics
and
Department of Astronomy and Space Physics
Uppsala University, Sweden
AB
A
This book was typeset in L TEX 2
on an HP9000/700 series workstation
and printed on an HP LaserJet 5000GN printer.
Contents
Preface
xi
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 . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
1
2
2
5
5
6
8
8
9
10
11
14
14
15
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
Plane Waves . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2.1 Telegraphers equation . . . . . . . . . . . . . . . .
2.2.2 Waves in conductive media . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
23
24
24
24
25
26
28
29
30
ii
C ONTENTS
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 . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
35
35
36
36
38
38
39
.
.
.
.
40
44
44
45
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 . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
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
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
.
.
.
.
.
.
.
.
.
.
.
.
.
.
iv
C ONTENTS
8.3.3
8.3.4
Vavilov-Cerenkov radiation . . . . . . .
Bibliography . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
143
145
146
148
149
151
153
158
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 . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
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.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
163
163
163
163
163
164
164
164
164
164
164
165
165
165
166
159
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 . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Example M.1 Tensors in 3D space .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
167
167
167
167
169
169
169
170
171
Dyadic product .
Vector product .
M.1.4 Vector analysis .
The del operator
.
.
.
.
Example M.6 The four-del operator in Lorentz space .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
178
179
179
179
180
vi
C ONTENTS
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
Coulomb interaction . . . . . . . . . . . . . . . . . . . . . . . 3
Ampre interaction . . . . . . . . . . . . . . . . . . . . . . . . 6
Moving loop in a varying B eld . . . . . . . . . . . . . . . . . 12
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
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
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
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.
1
|x x |
qq
40
qq
qq x x
=
3
40 |x x |
40
1
|x x |
F(x) =
(1.1)
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).
Estat lim
q0
F
q
(1.2)
1.1
E LECTROSTATICS
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
1
|x x |
q
40
q
q xx
=
40 |x x |3
40
1
|x x |
Estat (x) =
(1.3)
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,
1 In
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.
C LASSICAL E LECTRODYNAMICS
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
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
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
Estat (x) =
1
40
(x ) 3
dx
|x x |
Estat (x) =
=0
(1.6)
(1.7)
Estat (x) =
1 In
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.
dl dl
C C
dl (x x )
|x x |3
C C
dl
1
|x x |
(1.8)
0 JJ
4
0 JJ
=
4
F(x) =
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
dl
1
|x x |
0 JJ
4
0 JJ
F(x) =
dl
(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
0 JJ
4
F(x) =
C C
xx
dl dl
|x x |3
(1.11)
C LASSICAL E LECTRODYNAMICS
J
C
dl
xx
dl
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.
dBstat (x)
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:
j(x )
1
d3x
|x x |
xx 3
0
0
j(x )
d x =
3
4 V
4
|x x |
j(x ) 3
0
=
dx
4
V xx
Bstat (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
F(x) = J
dl Bstat (x)
(1.14)
j(x ) 3
dx
xx
=0
Bstat (x) =
(1.15)
1
d3x
|x x |
(1.16)
j(x )
0
4
j(x ) 3
dx =
V xx
1
0
d3x +
[j(x )
4 V
|x x |
Bstat (x) =
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:
j(x )
1
d3x
|x x |
1
d3x
|x x |
V
j(x )
V
d3x
1
d3x
|x x |
V
1
= xk
j(x )
xk |x x |
V
1
= xk j(x )
dS
xk |x x |
S
[j(x )
(1.17)
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
C LASSICAL E LECTRODYNAMICS
stationary currents (no charge accumulation in space). The net result is simply
Bstat (x) = 0
(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) =
(1.19a)
(1.19b)
stat
(1.20a)
(x) = 0 j(x)
(1.20b)
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.
1.3
E LECTRODYNAMICS
(t, x)
+ j(t, x) = 0
t
(1.21)
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).
(t, x )
= 0 j(t, x) + 0 0 E(t, x)
t
1
d3x
|x x |
B(t, x) = 0
(1.22)
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
(t, x )
V
1
d3x =
t
40
|x x |
= E(t, x)
t
1
d3x
|x x |
(t, x )
1
40 t
(1.23)
Later, we will need to consider this formal result further. The result is Maxwells
source equation for the B eld
0 E(t, x)
t
B(t, x) = 0 j(t, x) +
(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
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.
(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 )
(1.26)
E=
(Estat + EEMF ) dl
(1.27)
1.3
11
E LECTRODYNAMICS
E=
EEMF dl
(1.28)
d
dt
E(t, x) dl =
S
d
m (t, x)
dt
B(t, x) dS =
E(t, x) =
dS
B(t, x)
t
(1.29)
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
(1.31)
12
C LASSICAL E LECTRODYNAMICS
dS
B(x)
v
C
dl
B(x)
B dS =
d
dt
dS
E(t, x) =
(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)
(1.34)
1.3
13
E LECTRODYNAMICS
allowing us to rewrite Equation (1.32) on the facing page in the following way:
B
dS
t
d
dt
EEMF dl =
E(t, x) =
B dS
(1.35)
(B v) dS
(EEMF v B) dl =
EEMF dl =
B
dS
t
B
dS
t
E(t, x) =
(B v) dl
(1.36)
(1.37)
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)
(1.39)
(1.40)
(1.41)
Hence, we can conclude that for a stationary observer, the Maxwell equation
E =
B
t
(1.42)
14
C LASSICAL E LECTRODYNAMICS
(t, x)
0
(1.43a)
B
=0
t
B = 0
E
B 0 0
= 0 j(t, x)
t
E+
(1.43b)
(1.43c)
(1.43d)
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.
(1.44a)
H = H[t, x; E, B]
(1.44b)
1.4
15
E LECTROMAGNETIC D UALITY
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
(1.45)
(1.46)
H= B
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.
e
0
B
= 0 jm
t
B = 0 m
E
= 0 je
B 0 0
t
E+
(1.48a)
(1.48b)
(1.48c)
(1.48d)
16
C LASSICAL E LECTRODYNAMICS
( 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 ):
E cB
(1.51a)
cB E
e
(1.51b)
(1.51c)
c
e
cj j
(1.51d)
j cj
(1.51e)
(1.51f)
which is a particular case ( = /2) of the general duality transformation (depicted by the Hodge star operator)
E = E cos + cB sin
(1.52a)
c B = E sin + cB cos
(1.52b)
c = c cos + sin
m
= c sin + cos
e
c j = cj cos + j sin
m
j = cj sin + j cos
(1.52c)
(1.52d)
(1.52e)
(1.52f)
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
17
E LECTROMAGNETIC D UALITY
FARADAY S LAW AS
E XAMPLE 1.1
m (x )
1
d3x
|x x |
"
0
0
xx 3
d x =
m (x )
4 V
4
x |3
|x
1
0
m (x )
d3x
=
4
|x x |
V
Bstat (x) =
(1.53)
"
[cf. Equation (1.5) on page 4 for Estat ] and, if magnetic currents exist, a Biot-Savart
law for electric elds:
V
jm (x )
0
xx 3
dx =
4
|x x |3
jm (x )
0
4
1
d3x
|x x |
(1.54)
"
Estat (x) =
Taking the curl of the latter and using the operator bac-cab rule, Formula (F.62) on
page 165, we nd that
"
jm (x )
1
d3x
|x x |
(1.55)
"
0
4
jm (x ) 3
dx =
V xx
1
0
d3x +
[jm (x )
4 V
|x x |
"
Estat (x) =
Comparing with Equation (1.16) on page 7 for Estat and the evaluation of the integrals
there we obtain
Estat (x) = 0
(1.56)
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,
jm (t, x )(x x ) d3x
= 0 jm (t, x)
B(t, x)
t
0
4 t
m (t, x )
V
1
d3x
|x x |
"
E(t, x) = 0
(1.57)
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
E XAMPLE 1.2
OF EXAMPLE
1.1
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 ) =
"
(1.60)
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
OF EXAMPLE
1.2
"
QED
E ND
(1.59)
1 E
=
c2 t
"
(1.58)
1.4
19
E LECTROMAGNETIC D UALITY
E XAMPLE 1.3
(1.62a)
(1.62b)
=
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
(1.63a)
(1.63b)
(1.63c)
(1.63d)
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
(1.64)
E cos c B sin
(1.65)
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
(1.67)
so that
e
1 e
cos 0 =
cos 0
0
(1.68)
QED
E ND OF
EXAMPLE
E =
1.3
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
T HE
'
&
where E, B
(1.69)
3,
(1.70)
is conserved. I.e.,
E 2 c2 B2 = Const
E B = Const
(1.71a)
(1.71b)
(1.72)
= E E c2 B B + ic(E B) + ic(B E)
= 0 + 0 + ic(E B) ic(E B) = 0
(1.73)
= E E + c2 B B ic(E B) + ic(B E)
= 0 + 0 ic(E B) ic(E B) = 2ic(E B)
(1.74)
E ND
OF EXAMPLE
1.4
1.4
21
B IBLIOGRAPHY
D UALITY EXPRESSED IN
E XAMPLE 1.5
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
(1.75)
while
(( ((
G G =
= ei G ei G = |F|2
(1.76)
G G = e2i G G
(1.77)
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)
EXAMPLE
G
1.5
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.
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
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
( B) = 0
j + 0 E
t
t
t
( E) =
(2.2)
( E) = ( E)
(2.3)
(2.4)
and since EEMF = 0, Ohms law, Equation (1.26) on page 10, yields
j = E
(2.5)
E + 0 E = 0
t
t
E 0
(2.6)
E 0
E 1 2 E
=0
t c2 t2
(2.7)
( E) = 0 E + 0 0 ( E) (2.8)
t
t
2.1
25
T HE WAVE E QUATIONS
which, with the use of Equation (F.67) on page 165 and Equation (2.1c) on
page 23 can be rewritten
2
( B)
B = 0
2
B
0 0 2 B
t
t
(2.9)
B 1 2 B
=0
t c2 t2
B 0
(2.10)
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.
(2.11)
and insert this into Equation (2.7) on the facing page. This yields
2
1 2
E0 (x)eit 2 2 E0 (x)eit
t
c t
1
= 2 E 0 (i)E0 (x)eit 2 (i)2 E0 (x)eit
c
1
2
2
= E 0 (i)E 2 (i) E =
c
2
E=0
= 2E + 2 1 + i
c
0
E 0
(2.12)
E+
1+
i
E=0
(2.13)
E+
2
E=0
c2
(2.14)
26
E LECTROMAGNETIC WAVES
E + i0 E = 0
(2.15)
(2.16)
k=
c
=
=
=
0 0 ck k
0
= R0
0 k
(2.17)
where in the last step we introduced the characteristic impedance for vacuum
E XAMPLE 2.1
R0 =
0
376.7
0
WAVE EQUATIONS IN
(2.18)
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
(2.19a)
E = 0 jm
B = 0
B
t
(2.19b)
(2.19c)
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
27
T HE WAVE E QUATIONS
(2.20)
that
1 E
( B) = 0 m B
0 j e + 2
t
t
c t
= 0 m 0 e E +
1 E
E 1 2 E
0 e
c2 t
t c2 t2
"
( E) = 0 jm
"
1/c2 ,
(2.21)
2 E,
m
c2
"
E 1 2 E
2 m e E = 0
0
t c2 t2
(2.22)
2
c2
E+
!0
E+
2
E 2 m e E
0
c2
1 0 m e
e + m /c2
E=0
+i
2
0
0
"
m
c2
(2.23)
E + i0 e +
"
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
1
R2 m e
0
2
245
3"
2
c2
1+i
e + m /c2
R2
0 1 0 m e
2
E+
@8
A97
B
E=0
(2.24)
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
m
c2
"
E=0
(2.25)
EXAMPLE
2.1
28
E LECTROMAGNETIC WAVES
=n
(2.26)
E
n
B
n
B
n
=0
=
(2.27a)
B
t
(2.27b)
=0
(2.27c)
= 0 j(t, x) + 0 0
E
E
= 0 E + 0 0
t
t
(2.27d)
= n 0 + 0 0
E
t
0 = n n
(2.28)
which simplies to the rst-order ordinary differential equation for the normal
component En of the electric eld
dEn
+ En = 0
dt
0
(2.29)
(2.30)
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
(2.31)
(2.32)
2.2
29
P LANE WAVES
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.
=0
2
t c2 t2
(2.33)
=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)
kt)
(2.36)
n= k
c
c
(2.37)
(2.38)
30
E LECTROMAGNETIC WAVES
E
= iB = ik n E
(2.39)
k
1
1
n E = k E = k E = 0 0 n E
(2.40)
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).
(2.41)
where
2
c2
1+i
= k2 1 + i
K 2 = 0 0 2 1 + i
(2.42)
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
K=k
1+i
= + i
0
(2.43)
or
k2 1 + i
= (2 2 ) + 2i
(2.44)
2 = 2 k2
(2.45)
(2.46)
k2
20
2.2
31
P LANE WAVES
Squaring the latter and combining with the former, one obtains the second
order algebraic equation (in 2 )
k 4 2
42 2
0
2 (2 k2 ) =
(2.47)
(2.48a)
D
CDD
) E
) E
=k
1+
CDD
=k
1+
(2.48b)
(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)
= (1 + i)
20
= 0 0 (1 + i)
0
=k i
1i
0
1
2
G
1
2
k(1 + i)
0
2
K = k 1+i
0
20
(2.51)
From this analysis we conclude that when the wave impinges perpendicularly upon the medium, the elds are given, inside this medium, by
)
0
( n E )
)
B = (1 + i)
0
exp i
2
0
t
2
H
E = E0 exp
(2.52a)
(2.52b)
32
E LECTROMAGNETIC WAVES
2
0
(2.53)
(2.54)
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
(2.55)
I
I
2.3
33
B IBLIOGRAPHY
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
1
1
F G Re {F} Re {G} = Re F G = Re F G
2
2
(2.57)
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.
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.
(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)
stat (x) =
(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
Astat (x)
(3.4)
Astat (x) =
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.
A(t, x)
(3.6)
3.3
37
T HE E LECTRODYNAMIC P OTENTIALS
Inserting this expression into the other homogeneous Maxwell equation, Equation (1.30) on page 11, we obtain
E(t, x) =
(3.7)
A(t, x) = 0
t
(3.8)
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
+
(3.11a)
= 0 j(t, x)
(3.11b)
38
E LECTROMAGNETIC P OTENTIALS
A+
1
c2 t
1
A+ 2
c t
A+
(t, x)
0
(3.12a)
= 0 j(t, x)
(3.12b)
1 2
c2 t2
1 2 A
c2 t2
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
1 In
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.
3.3
39
T HE E LECTRODYNAMIC P OTENTIALS
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
=
def
def
(3.14a)
(3.14b)
S
S
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)
+
=
t
t
t
t
t
A = A ( ) = A
E =
(3.16a)
B =
(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.
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
1 2
c2 t2
1 2
c2 t2
(t, x)
0
= 0 j(t, x)
1 2
c2 t2
1 2 A
c2 t2
(3.17a)
(3.17b)
We notice that if we require that the gauge function (t, x) itself be restricted
to full the wave equation
1 2
c2 t2
=0
(3.18)
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.
3.3
41
T HE E LECTRODYNAMIC P OTENTIALS
(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:
2
(t, x) = f (t, x)
(3.19)
def
1
2
f (x) eit d
(3.20a)
f (t, x) eit dt
(3.20b)
def
F 1 [ f (x)] f (t, x) =
exists, and that the same is true for the generic potential component:
1
2
(x) =
(t, x) =
(x) eit d
(t, x) eit dt
(3.21a)
(3.21b)
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)
(3.23)
G(x, x ) + k2G(x, x ) = (x x )
(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
f (x )G(x, x ) d3x
(x) =
(3.25)
(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
(3.27)
eikr
eik|xx |
eikr
eik|xx |
+ C
+ C
C+
r
r
|x x |
|x x |
(3.28)
1
1
+ C
,
|x x |
|x x |
G( x x ) C +
xx 0
(3.29)
The volume integrated Equation (3.24) on the previous page can under this
assumption be approximated by
( x x ) d x
1
d3x + k2 C + + C
|x x |
C+ + C
1
d3x
|x x |
(3.30)
3.3
43
T HE E LECTRODYNAMIC P OTENTIALS
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
1
4
C+ + C =
(3.31)
Insertion of the general solution Equation (3.28) on the facing page into
Equation (3.25) on the preceding page gives
f (x )
eik|xx | 3
d x + C
|x x |
f (x )
(x) = C +
eik|xx | 3
dx
|x x |
(3.32)
|x x |
exp i t + k|xx |
|x x |
d d3x
(3.33)
V
WG
T
f (x )
+ C
f (x )
exp i t k|xx |
V
WG
(t, x) = C
d d3x
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 +
(3.34a)
(3.34b)
(t, x) = C +
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).
44
E LECTROMAGNETIC P OTENTIALS
(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
e
0
(3.37a)
E = 0 jm
B = 0
B
t
B = 0 j e + 0 0
(3.37b)
(3.37c)
E
t
(3.37d)
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
A (t, x) Am
t
1
m (t, x) 2 Am (t, x) + Ae
c t
E = e (t, x)
B=
1
c2
(3.38a)
(3.38b)
45
B IBLIOGRAPHY
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
(3.39b)
Am +
Am +
= 0 je (t, x)
1 m
c2 t
"
1 2 Am
c2 t2
(3.39a)
2 e
"
3.3
(3.39c)
= 0 jm (t, x)
(3.39d)
(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
3.1
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.
46
E LECTROMAGNETIC P OTENTIALS
4
Relativistic
Electrodynamics
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
v
P(t, x, y, z)
P(t , x , y , z )
O
z
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.
4.1
T HE S PECIAL T HEORY
OF
49
R ELATIVITY
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)
(4.3a)
x = (x vt)
(4.3b)
y =y
(4.3c)
z =z
(4.3d)
Taking the difference between the square of (4.3a) and the square of (4.3b) we
nd that
v2
c2
v2
c2
2 2
= c t x2
1
c2 t 2 1
x2 1
v2
c2
c2 t 2 x 2 = 2 c2 t2 2xct + x2 2 x2 + 2xvt v2 t2
(4.4)
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.
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:
(4.6)
(4.7)
This summation process is an example of index contraction and is often referred to as index lowering.
(4.8)
4.1
T HE S PECIAL T HEORY
OF
51
R ELATIVITY
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
`a
g =
as
1
if = = 0
1 if = = i = j = 1, 2, 3
0
if =
(4.10)
ac
hh
ig
(4.11)
deef
(g ) =
1 0
0
0
0 1 0
0
0 0 1 0
0 0
0 1
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)
(4.12)
deef
x0
x1
x2
x3
x0
x1
x2
x3
hh
ig
hh
ig
deef p
deef
1 0
0
0
0 1 0
0
0 0 1 0
0 0
0 1
hh
ig
hh
ig
p
x0
x1
x2
x3
(4.13)
deef
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
52
R ELATIVISTIC E LECTRODYNAMICS
(4.14)
(4.15a)
g = g
g g =
g g =
g = g
g
=
=
(4.15b)
(4.15c)
(4.15d)
= =
1
0
if =
if =
(4.16)
(4.17)
4.1
T HE S PECIAL T HEORY
OF
53
R ELATIVITY
dx 3
+
dt
CDD
E
1
(v x )2 + (vy )2 + (vz )2 dt = c
c2
c
= c 1 2 dt = dt = c d
=c
dx 2
+
dt
dx 1
dt
ds = c
1
1 2
c
dt
v2
dt
c2
(4.18)
where we introduced
d = dt/
(4.19)
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)
(4.21)
(4.22)
(4.23)
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.
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
a (x ) = g a (x ) = (a0 (x ), a(x ))
(4.24)
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)
hh
ig
p
deef
0 0
0 0
0
0
1 0
0
0
0 1
(4.26)
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
(4.27)
(4.28)
1 + 2
1 + 1 2
(4.29)
4.1
T HE S PECIAL T HEORY
OF
55
R ELATIVITY
X0
X
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.
$ #
$ #
(4.30a)
X 1 = x1
(4.30b)
(4.30c)
(4.30d)
X =x
X =x
dS = ids
(4.30e)
56
R ELATIVISTIC E LECTRODYNAMICS
x0
= ct
x0
x0 = x 1
x1
ct
O=O
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.
(4.31)
dS 2 = (dX 0 )2 + (dx1 )2
(4.32)
(4.33a)
(4.33b)
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.
4.2
57
(4.34a)
x = x cosh ct sinh
(4.34b)
(4.35a)
cosh =
(4.35b)
tanh =
(4.35c)
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.
v2
c2
v
1
v2
c2
u fd
dx
= (c, v) =
d
u =
= (u0 , u)
(4.37)
which, when multiplied with the scalar invariant m0 yields the four-momentum
v2
c2
u fd
m0 c
m0 v
1
v2
c2
dx
= m0 (c, v) =
d
p = m0
= (p0 , p)
(4.38)
58
R ELATIVISTIC E LECTRODYNAMICS
(4.39)
where
m = m0 =
m0
(4.40)
1 v2
c
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
m0 c2
cp0 = m0 c2 =
v2
c2
= mc2
(4.41)
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)
(4.42)
1 v2
c
v2
1 2
c
(4.43)
2 2
(m0 c2 )2
= (m0 c )
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
(4.44)
4.3
59
dx
= 0 u = 0 (c, v) = (c, v)
d
(4.45)
where we introduced
= 0
(4.46)
= = =
1 2
c2 t2
(4.47)
Since it has the characteristics of a four-scalar, the dAlembert operator is invariant and, hence, the homogeneous wave equation is Lorentz covariant.
,A
c
A =
(4.48)
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:
2
A = 0 j
(4.49)
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)
60
R ELATIVISTIC E LECTRODYNAMICS
and the Lorentz gauge condition, Equation (3.13) on page 38, can be written
A = 0
(4.51)
(4.52)
If only one dimension Lorentz contracts (for instance, due to relative motion along the x direction), a 3D spatial volume transforms according to
1 2 = dV0
1
dV = d3x = dV0 = dV0
v2
c2
(4.53)
(4.54)
i.e., the charge in a given volume is conserved. We can therefore conclude that
the elementary charge is a universal constant.
,A
c
=
v=0
1
q
,0
40 c |x x |0
(4.55)
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 )
(4.56)
R R = (c(t t ), x x ) (c(t t ), (x x )) = c2 (t t )2 x x
(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 )
(4.58)
Inserting this into Equation (4.57) on the facing page, we see that
R R = 0
(4.59)
R = ( x x , x x )
(4.60)
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
v
1
v2
c2
= (c, 0)
(4.61)
u fd
(4.62)
v=0
Q
and
(R )0 = ( x x , x x )0 = ( x x
, (x x )0 )
= (c, 0) ( x x
, (x x )0 ) = c x x
0
u R = u R
(4.63)
0
q
u
40 cu R
(4.64)
u R = (c, v) x x , (x x ) = c x x v (x x )
(4.65)
62
R ELATIVISTIC E LECTRODYNAMICS
def
= v
(4.66)
and introducing
v (x x )
x x (x x )
c
def
s xx
(4.67)
we can write
u R = cs
(4.68)
and
1 v
,
cs c2 s
(4.69)
u
=
cu R
1 v
,
cs c2 s
q
40
,A
c
(4.70)
A (x ) =
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) =
(4.71a)
(4.71b)
i jk ai b j xk
(4.72)
i, j = k
(4.73)
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:
A
B=
(4.76a)
A
t
In component form, this can be written
E =
(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:
Bi j =
,A
c
(4.78)
A =
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
deef
hh
ig
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
(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)
hh
ig
(4.82)
deef
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
(4.83)
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
(4.84)
or, equivalently,
E = 0 c2 =
(4.85)
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)
4.3
65
B IBLIOGRAPHY
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
t
B = 0
(4.89)
(4.90)
(4.91)
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
67
68
E LECTROMAGNETIC F IELDS
AND
PARTICLES
Lagrange formalism
Let us no introduce a function L(4) which fulls the variational principle
1
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
m0
u u + qu A d
2
0
A
m0 (u u )
u + q A u + u x
2 u
x
R
m0 u u + q A u + u A x
0
1
L(4) (x , u ) d =
(5.4)
d = 0
5.1
C HARGED PARTICLES
IN AN
69
E LECTROMAGNETIC F IELD
dx
d
(5.5)
which means that we can write the variation of u as a total derivative with
respect to :
=
d
x
d
dx
d
u =
(5.6)
Inserting this into the rst two terms in the last integral in Equation (5.4) on
the facing page, we obtain
1
L(4) (x , u ) d
d
d
m0 u
x + qA
x + qu A x d
d
d
(5.7)
Partial integration in the two rst terms in the right hand member of (5.7)
gives
1
L(4) (x , u ) d
(5.8)
du
dA
m0
x q
x + qu A x d
d
d
where the integrated parts do not contribute since the variations at the endpoints vanish. A change of irrelevant summation index from to in the 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
=
0
du
dA
m0
q
+ qu A x d
d
d
(5.9)
(5.10)
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
L(4) (x , u ) d
du
m0
+ qu A A
d
(5.11)
x 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:
du
= qu A A
d
m0
(5.12)
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)
5.1
C HARGED PARTICLES
IN AN
71
E LECTROMAGNETIC F IELD
L(4)
u
(5.15)
(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)
(5.18)
Furthermore, from the denition (5.15) of the canonically conjugate fourmomentum p , we see that
L(4)
=
u
u
1
m0 u u + qu A (x ) = m0 u + qA
2
p =
(5.19)
72
E LECTROMAGNETIC F IELDS
AND
PARTICLES
(5.20)
m0 c2
2
(5.21)
u =
(5.22)
p p 2qA p + q2 A A
(5.23)
p qA
p qA
1
1
p qA
p qA
m0
m0
m0
2
1
=
2m0
1
=
2m0
H(4) =
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
(5.24)
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
du
= qu A A = qu F
d
m0
(5.25)
5.1
C HARGED PARTICLES
IN AN
73
E LECTROMAGNETIC F IELD
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
x
iw
v
(pqA)2
y
(p0 )2
1
2
q2
m0 c2
=
(p0 )2 (p)2 qp0 + 2q(p A) + 2 2 q2 (A)2
2
2m0
c
c
(5.27)
(5.28)
q
p0 =
c
(p qA)2 + m2 c2
0
(5.29)
H cp0 = q + c
(p qA)2 + m2 c2
0
(5.30)
(5.31)
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
(p qA)2 + m2 c2
0
(5.32)
(5.33)
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
2
= mv q( A v) mc = q + qA v m0 c
v2
1 2
c
(5.34)
5.2
75
i1
m
k
a
i
m
k
a
i+1
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
L(i , i , t) dt = 0
(5.35)
(5.36)
L=
i=1
L = a
(5.37)
Here,
i+1 i
1 m 2
ka
2 a i
a
G
(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
(5.39a)
linear mass density
(5.39b)
Youngs modulus
(5.39c)
(5.39d)
we obtain
(5.40)
, , , t =
t x
2
t
(5.41)
where
dx
L=
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.
L dt
(5.42)
t
x
t
x
G F
G F
dx dt
=0
,
t x
dx dt
5.2
77
The last integral can be integrated by parts. This results in the expression
G F
fd
G F
fd
dx dt = 0
(5.43)
where the variation is arbitrary (and the endpoints xed). This means that the
integrand itself must vanish. If we introduce the functional derivative
G F
fd
G F
fd
(5.44)
=0
(5.45)
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
d4x
=0
(5.47)
fd
d3x dt
L dt =
d4x
where the variation is arbitrary and the endpoints are xed. This means that
the integrand itself must vanish:
fd
=0
(5.48)
78
E LECTROMAGNETIC F IELDS
AND
PARTICLES
xi
fd
G F
fd
xi
(5.49)
=0
(5.50)
G F
(x ) = (t, x) =
,
t xi
(5.52)
;t =
i
x
t
, ,
If, as usual, we differentiate this expression and identify terms, we obtain the
following Hamilton density equations
(5.53a)
(5.53b)
The Hamilton density functions are in many ways similar to the ordinary
Hamilton functions and lead to similar results.
mech
inter
tot
eld
(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
inter
= j A
(5.56)
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
(5.57)
1
2
0u
1
F F
40
u + j A +
(5.58)
E XAMPLE 5.1
(5.59)
"
2445
E y /c
Bz
0
Bx
E z /c
By
B x
0
@@ B8
2445
F =
0
E x /c
E y /c
E z /c
@@ 8B
F =
0
E x /c
E y /c
E z /c
(5.60)
(5.61)
where denotes the row number and the column number. Then, Einstein summation
80
E LECTROMAGNETIC F IELDS
AND
PARTICLES
(5.62)
E 2 /c2 + 0 + B2 + B2
x
z
y
2
E y /c2 + B2 + 0 + B2
z
x
2
E z /c2 + B2 + B2 + 0
y
x
2
2
= 2E 2 /c2 2E y /c2 2E z /c2 + 2B2 + 2B2 + 2B2
x
x
y
z
B2
0 E 2
0
(5.63)
"
QED
E ND
OF EXAMPLE
1 2
B2
E
0 c 2 0
"
1
1
F F =
40
2
5.1
inter
EM
eld
= j A +
1
F F
40
(5.64)
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
(5.65)
5.2
81
Furthermore,
EM
1
=
F F
( A )
40
( A )
1
=
( A A )( A A )
40
( A )
(5.66)
A A + A A
A A A A
20
( A )
9
A A A A
40
( A )
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 )
(5.68)
so that
EM
1
1
= A A = F
( A )
0
0
(5.69)
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
1
= j F = 0
( A )
0
EM
(5.70)
82
E LECTROMAGNETIC F IELDS
AND
PARTICLES
or
F = 0 j
(5.71)
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:
1
m2 2
2
(5.72)
m2 ) = 0
(5.73)
em|x|
|x|
(5.74)
which describes the Yukawa meson eld for a scalar meson with mass m. With
=
1
c2 t
(5.75)
(5.76)
inter
EM
eld
= j A +
1
F F + m2 A A
40
(5.77)
5.2
83
B IBLIOGRAPHY
(5.78)
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.
q=
(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
(6.3)
3 (x x0 )i (x x0 ) j 1
i j + . . .
2 |x x0 | |x x0 | 2
1
Qi j
+
|x x0 |3
(6.4)
stat (x) =
P(x )
xx 3
1
d x =
3
|x x |
40
V
1
P(x )
d3x
|x x |
V
P(x )
1
d3x
|x x |
1
40
1
=
40
p (x) =
(6.5)
Using the expression Equation (M.83) on page 182 and applying the divergence theorem, we can rewrite this expression for the potential as follows:
(6.6)
P(x )
P(x ) 3
d3x
dx
|x x |
V
V |x x |
P(x ) n 2
P(x ) 3
d x
dx
|x x |
|x x |
S
V
1
40
1
=
40
p (x) =
6.1
E LECTRIC P OLARISATION
AND
87
D ISPLACEMENT
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
p =
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
(6.9)
we obtain
(0 E + P) =
D = true (x)
(6.10)
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)
88
E LECTROMAGNETIC F IELDS
AND
M ATTER
(6.13)
where, approximately,
= 0 (1 + )
(6.14)
m=
(x x0 ) j(x ) d3x
(6.15)
6.2
M AGNETISATION
AND THE
89
M AGNETISING F IELD
volume, M. Via the denition of the vector potential one can show that the
magnetisation current and the magnetisation is simply related:
jM =
(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)
B = 0 jtrue +
(6.18)
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)
(6.21)
where, approximately,
= 0 (1 + m )
(6.22)
90
E LECTROMAGNETIC F IELDS
AND
M ATTER
E =
(6.23a)
(6.23b)
B
t
(6.23c)
H = j(t, x) +
D
t
(6.23d)
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.
(6.24)
Integration over the entire volume V and using Gausss theorem (the divergence theorem), we obtain
1
(H B + E D) d3x =
2
j E d3x +
(E H) nd2x
(6.25)
(6.26)
j2 3
dx
j E d3x =
j EEMF d3x
(6.27)
6.3
E NERGY
AND
91
M OMENTUM
x
iw
Field energy
v y
x
iw v
Joule heat
1
(E D + H B) d3x +
2
(E H) nd2x
x
w
x
iw
j2 3
dx +
Radiated power
j EEMF d3x =
(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 .
(D B) + D
t
t
= E( D) B ( H)
(D B) D ( E) + H( B)
t
H
E + j B = ( D)E +
(6.32)
y iw v
x
=0
(D B)
t
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
1
D
E
E
D
+
2
xi
xi
x j
1
E i D j E D i j
2
(6.33)
[E( D) D ( E)]i =
and
1
B
H
H
B
+
2
xi
xi
x j
1
Hi B j B H i j
2
(6.34)
[H( B) B ( H)]i =
respectively.
Using these two expressions in the ith component of Equation (6.32) on
the preceding page and re-shufing terms, we get
+ H
B
H
B
xi
xi
1
1
E i D j E D i j + Hi B j H B i j
2
2
(D B)i
t
D
E
D
xi
xi
x j
1
2
(E + j B)i
(6.35)
Introducing the electric volume force Fev via its ith component
D
E
B
H
D
+ H
B
xi
xi
xi
xi
1
2
H
(Fev )i = (E + j B)i
(6.36)
(6.37)
= ( T)i
(D B) =
t
x j
i
Fev +
(6.38)
(6.39)
B = m 0 H = H
(6.40)
6.3
93
B IBLIOGRAPHY
T i j
m S
= Fev + 2
x j
c t
(6.41)
i
where S is the Poynting vector dened in Equation (6.29) on page 91. Integration over the entire volume V yields
x
iw
m 3
Sd x =
c2
Field momentum
d
dt
T n d2x
x
iw
x
iw
(6.42)
Fev d3x
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
S d3x =
d mech d eld
p
+ p
=
dt
dt
1 d
c2 dt
T n d2x
or
(E + v B) d3x +
T n d2x
(6.43)
(6.44)
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.
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
1
2
(x) =
(t, x) =
(x) eit d
(7.1a)
(t, x) eit dt
(7.1b)
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
95
96
E LECTROMAGNETIC F IELDS
AND
M ATTER
pairs
(t, x) =
1
2
and
1
2
(7.2b)
j (x) eit d
j (x) =
(7.2a)
(t, x) eit dt
j(t, x) =
(x) =
(x) eit d
(7.3a)
j(t, x) eit dt
(7.3b)
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
1
2
(x) eit d
(t, x) eit dt =
(x) =
(7.4a)
1
40
(x )
(t, x) =
eik|xx | 3
dx
|x x |
(7.4b)
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:
1
2
A (x) =
A (x) eit d
A(t, x) eit dt =
(7.5a)
0
4
j (x )
A(t, x) =
eik|xx | 3
dx
|x x |
(7.5b)
(7.6)
in order that all physical quantities be real. Similar transform pairs and requirements of real-valuedness exist for the elds themselves.
1 In
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].
7.1
97
T HE M AGNETIC F IELD
(7.7a)
i0 t
(7.7b)
(t, x) = 0 (x)ei0 t
(7.7c)
i0 t
(7.7d)
j(t, x) = j0 (x)e
A(t, x) = A0 (x)e
A(t, x)
(7.8)
(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
A (x) =
B (x) =
j (x )
eik|xx | 3
dx
|x x |
(7.10)
98
E LECTROMAGNETIC F IELDS
FROM
j (x )eik|xx | (x x ) 3
dx
|x x |3
V
(ik)j(x )eik|xx | (x x ) 3
+
dx
V
|x x |2
0
4
(7.11)
j (x )
0
4
0
=
4
B (x) =
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):
v y
x
w
Induction eld
d3x
(7.12)
ret , x ) (x x )
j(t
d3x
2
|x x |
x
w
Radiation eld
|x x |
(i)j (x )ei(k|xx |t) d (x x )
|x x |
j(tret , x ) (x x ) 3
0
dx +
3
4c
|x x |
V
d3x
0
4
1
c
j (x )ei(k|xx |t) d (x x )
0
4
B (x) eit d
B(t, x) =
v
where
j
t
ret , x ) def
j(t
(7.13)
t=tret
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 ()!
7.2
99
T HE E LECTRIC F IELD
eik|xx | 3
eik|xx | 3
i0
j (x )
dx +
dx
4 V
|x x |
|x x |
V
(7.14)
(x )eik|xx | (x x ) 3
dx
3
V
|x x |
(x )(x x ) j (x ) eik|xx | 3
ik
dx
c
|x x |
|x x |
V
(x )
1
40
1
=
40
(7.15)
j (x )
(7.16)
Doing so in the last term of Equation (7.14), and also using the fact that
k = /c, we can rewrite this Equation as
x
iw
1
c
(x )eik|xx | (x x ) 3
dx
|x x |3
[ j (x )](x x )
eik|xx | 3
ikj (x )
dx
|x x |
|x x |
(7.17)
y
1
40
E (x) =
(7.18)
j (x )](x x )
eik|xx | 3
ikj (x )
dx
|x x |
|x x |
eik|xx | 3
jm xl xl
ik jl (x ) xl
dx
xm |x x |
|x x |
[
I =
100
E LECTROMAGNETIC F IELDS
FROM
But, since
xl xl ik|xx |
e
|x x |2
xl xl ik|xx |
+ jm
e
xm |x x |2
jm
xm
(7.19)
xl xl ik|xx |
e
|x x |2
jm
xm
we can rewrite I as
(7.20)
xm
xl xl
eik|xx | 3
dx
xl eik|xx | + ikj
|x x |
|x x |2
xl xl
jm
xl eik|xx | d3x
|x x |2
xm
jm
I =
|x x |3
ik|xx |
eik|xx |
+ j
|x x |
d3x
(7.21)
dx
j (x x ) (x x )
ik
eik|xx |
2
j (x x ) (x x )eik|xx |
+
2
4
|x x |
|x x |
I =
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:
1
40
1
=
40
E (x) =
ik
c
1
c
1
c
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 )
7.3
101
T HE R ADIATION F IELDS
Intermediate eld
x
w
1
40 c2
Intermediate eld
[ ret , x ) (x x )] (x x ) 3
j(t
dx
|x x |3
Radiation eld
(7.23)
[j(tret , x ) (x x )] (x x ) 3
dx
|x x |4
x
iw
v
v
+
1
40 c
[j(tret , x ) (x x )](x x ) 3
dx
|x x |4
x
w
1
40 c
(tret , x )(x x ) 3
dx
|x x |3
1
40
x
iw
E (x) eit d
E(t, x) =
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.
102
E LECTROMAGNETIC F IELDS
FROM
1
40 c2
where
Erad (t, x) =
j
t
ret , x ) def
j(t
0
4c
Brad (t, x) =
ret , x ) (x x )
j(t
d3x (7.24a)
2
|x x |
[ ret , x ) (x x )] (x x ) 3
j(t
dx
|x x |3
(7.24b)
(7.25)
t=tret
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
Brad (x) =
(7.26a)
(7.26b)
k x x k (x x ) k (x x0 ) k (x x0 )
k |x x0 | k (x x0 )
(7.27)
7.3
103
T HE R ADIATION F IELDS
S
dS = nd2x
xx
x x0
x x0
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
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)
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
1
1
Re E B =
Re E eit (B eit )
20
20
1
1
=
Re E B eit eit =
Re E B
20
20
(7.30)
S = EH =
Using the far-eld approximations (7.29a) and (7.29b) and the fact that 1/c =
(j k)e
ik(x x0 ) 3
1
1
S =
R0
322 |x x0 |2
x x0
|x x0 |
(7.31)
(j k)eik(x x0 ) d3x
dP
1
=
R0
d 322
(7.32)
7.4
105
R ADIATED E NERGY
over the whole bandwidth. The total energy transmitted through a unit area is
the time integral of the Poynting vector:
(E H) dt
S(t) dt =
(7.33)
(E H )ei(+ )t dt
If we carry out the temporal integration rst and use the fact that
ei(+ )t dt = 2( + )
(7.34)
(E H ) d
(E H ) d +
(E H ) d
(E H ) d
(E B + E B ) d
(7.35)
(E H ) d
(E H ) d +
2
0
2
=
0
= 2
= 2
(E H ) d
= 2
S(t) dt = 2
(E B + E B ) 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
j k ik|xx | 3
dx
e
|x x |
k ndS d
0
0
1
4
U=
(7.36)
Inserting the approximations (7.27) and (7.28) into Equation (7.36) and
also introducing
U=
0
U d
(7.37)
and recalling the denition (2.18) on page 26 for the vacuum resistance R 0 we
obtain
(j k)e
dx
ik(x x0 ) 3
1
dU
d R0
d
4
(7.38)
106
E LECTROMAGNETIC F IELDS
FROM
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.
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.
j0 (x ) = I0 (x1 )(x2 )
sin[k(L/2 x3 )]
x3
sin(kL/2)
(8.1)
8.1
R ADIATION
FROM
109
E XTENDED S OURCES
x3
L
2
j (x )
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
L/2
2
= 4I0
k2 sin2
eikx0 cos
sin2 (kL/2)
sin[k(L/2 x3 )]
I0
k sin eikx3 cos eikx0 cos dx3
sin(kL/2)
L/2
L/2
2
= I0
j0 k)eik(x x0 ) d3x
(8.2)
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
110
E LECTROMAGNETIC R ADIATION
R ADIATING S YSTEMS
AND
is
kL0
12
sin d
(8.3)
2
R0 I0
lim P(L) =
1
4
2
P(L) = R0 I0
(8.4)
197
P(L) P(L)
= 1 2 = R0
2
6
Ieff
I0
2
Rrad (L) =
(8.5)
cos2
cos
d
sin
1
4
2
P(/2) = R0 I0
(8.6)
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
cos2
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 .
8.1
R ADIATION
111
E XTENDED S OURCES
FROM
x3 = z = z
k
x2
x1
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).
Brad =
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
= sin x1 + cos x2
and
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
z = x3 = cos sin
(8.10)
(8.11)
(8.12)
k x = ka sin cos( )
(8.13)
(8.14)
and
= I0 ak
eikx j k d3x = a
+ I0 ak cos
(8.15)
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
113
E XTENDED S OURCES
trigonometric identities, the rst integral in the RHS of (8.15) can be rewritten
2
= cos
=
1 1
+ cos 2
2 2
2
1
cos
eika sin cos d
2
0
2
1
eika sin cos cos 2 d
+ cos
2
0
= cos
(8.16)
2
1
sin
eika sin cos d
2
0
2
1
sin
eika sin cos cos 2 d
2
0
(8.17)
in
ei cos cos n d =
in
2
Jn () =
ei cos cos n d
(8.18)
(8.20)
114
E LECTROMAGNETIC R ADIATION
AND
R ADIATING S YSTEMS
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:
i0 e(ikrt )
I +I
4r
Brad (t, x) = Re
Brad (x) =
0
sin(kr t ) I + I
4r
I0 ak0
=
sin(kr t ) cos [J0 (ka sin ) J2 (ka sin)]
4r
(8.22)
From this expression for the radiated B eld, we can obtain the radiated E eld
with the help of Maxwells equations.
(8.23)
= jtrue
t
(8.24a)
(8.24b)
8.2
115
M ULTIPOLE R ADIATION
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 =
(8.26)
This equation is of the same type as Equation (3.19) on page 41, and has
therefore the retarded solution
1
40
e (t, x) =
(tret , x ) 3
dx
|x x |
(8.27)
e (x) =
(x )eik|xx | 3
dx
|x x |
(8.28)
(8.29)
we see that we can calculate the magnetic and electric elds, respectively, as
follows
B=
1 C
c2 t
(8.30a)
E=
(8.30b)
116
E LECTROMAGNETIC R ADIATION
AND
R ADIATING S YSTEMS
xx
x
x x0
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 )|
(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
8.2
117
M ULTIPOLE R ADIATION
Pn (cos ) =
(8.32)
m=n
x x0 = ( x x0 , , )
(8.33a)
x x0 = (|x x0 | , , )
(8.33b)
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
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
e =
x0 ) Pm (cos
n
)e
im
(8.34)
dx
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 |
(8.35)
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:
2n n!
k x x0
(2n + 1)!
jn (k x x0 )
(8.37)
Inserting these expansions into Equation (8.34), we obtain the multipole expansion of the Fourier component of the Hertz vector
e
e (n)
(8.38a)
n=0
118
E LECTROMAGNETIC R ADIATION
AND
R ADIATING S YSTEMS
where
V
1 eik|xx0 | 2n n!
40 |x x0 | (2n)!
e (n) = (i)n
V (x ) (k x x0 )n Pn (cos ) d3x
(8.38b)
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.
e (0) =
(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 |
def
e e (0) r =
def
e e (0) = 0
(8.40a)
(8.40b)
(8.40c)
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
1
40
1
eik|xx0 |
ik
p sin
|x x0 |
|x x0 |
(0)
C = C, =
(8.41)
8.2
119
M ULTIPOLE R ADIATION
k
x3
Brad
Erad
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
0
1
eik|xx0 |
ik
p sin
4 |x x0 |
|x x0 |
1
ik
x x0
1
2
cos
E =
2
40
|x x0 |
|x x0 |
|x x0 |
ik|xx0 |
1
ik
e
k2 sin
p
+
|x x0 |
|x x0 |2 |x x0 |
(8.42a)
(8.42b)
B = i
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 =
120
E LECTROMAGNETIC R ADIATION
R ADIATING S YSTEMS
AND
These elds constitute the electric dipole radiation, also known as E1 radiation.
e (1) = i
(8.44)
(8.45)
and introducing
i = xi x0,i
(8.46a)
i = xi x0,i
(8.46b)
1
{[(x x0 ) (x x0 )] (x )} j = i , j i + ,i j
2
1
+ i , j i ,i j
2
(8.47)
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
(8.48)
j
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)
8.2
121
M ULTIPOLE R ADIATION
allow us to express in j as
= i
(8.50)
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
(8.51)
m =
(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
(8.53)
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
0 eik|xx0 |
(m k) k
4 |x x0 |
k eik|xx0 |
Erad (x) =
m k
40 c |x x0 |
Brad (x) =
(8.54a)
(8.54b)
which are the elds of the magnetic dipole radiation (M1 radiation).
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 ) =
(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) =
(8.56a)
(8.56b)
Q
(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
R ADIATION
FROM A
L OCALISED C HARGE
x(t)
IN
123
A RBITRARY M OTION
8.3
dr
v(t )
xx
dS
x (t )
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 .
V (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
124
E LECTROMAGNETIC R ADIATION
AND
R ADIATING S YSTEMS
or
(tret , x ) d3x =
(8.59)
dq
(8.60)
1 (xx )v
c|xx |
(8.61)
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)
1
40
A(t, x) =
0
4
(t, x) =
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) =
(8.63a)
(8.63b)
where
(x x ) v
c
xx v
= xx 1
|x x | c
v
xx
= (x x )
|x x | c
(8.64a)
(8.64b)
s = xx
(8.64c)
8.3
R ADIATION
FROM A
L OCALISED C HARGE
IN
125
A RBITRARY M OTION
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
q
40
1 v
,
s cs
A (x ) =
= (, cA)
(8.65)
A(t, x)
(8.66a)
A(t, x)
E(t, x) = (t, x)
t
(8.66b)
(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
a(t ) = v(t ) =
(8.68a)
dv d2 x
=
dt
dt 2
(8.68b)
126
E LECTROMAGNETIC R ADIATION
AND
R ADIATING S YSTEMS
?
|x x |
v
c
q
x (t )
x0 (t)
v(t )
0
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).
(8.69a)
d2
(x x (t )) = (t )
v
dt 2
(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
L OCALISED C HARGE
FROM A
IN
127
A RBITRARY M OTION
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.
xx
|x x | t
x x (t ) =
x x (t ) =
(x x ) v
|x x |
(8.71)
Furthermore, by applying the operator (/t)x to Equation (8.70) on the preceding page we nd that
|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
t
(8.72)
x
t
t
|x x |
|x x |
=
s
|x x | (x x ) v/c
=
x
(8.73)
t
t
=
x
|x x |
s
(8.74)
x
128
E LECTROMAGNETIC R ADIATION
AND
R ADIATING S YSTEMS
( )t t = ( )t
(8.75)
xx
cs
(8.76)
+ ( )t =
xx
cs
( )t = ( )t t
+ ( )t
(8.77)
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
( )t =
(8.78a)
(8.78b)
x
Utilising these relations in the calculation of the E eld from the LinardWiechert potentials, Equations (8.63) on page 124, we obtain
A
t
q
(x x ) |x x | v/c
=
2
40 s
|x x |
E(t, x) =
s
t
(x x ) |x x | v/c
cs
|x x | v
2
c
(8.79)
Starting from expression (8.64a) on page 124 for the retarded relative distance
8.3
R ADIATION
FROM A
L OCALISED C HARGE
IN
129
A RBITRARY M OTION
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 |
x x (t )
v(t )
v (x x )
t
t
1
x x (t )
t
c
(x x ) v
c
xx
s
t
(8.80)
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
1
v2
c2
|x x | v
(x x )
v
c
x
iw
Radiation eld
d
xx
+ 2
c
(8.81)
y
|x x | v
c
x
iw
(x x )
q
40 s3
E(t, x) =
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)
q
v2
(x x0 ) 1 2
3
40 s
c
+ (x x )
(x x0 ) v
2
c
E(t, x) =
(8.83)
130
E LECTROMAGNETIC R ADIATION
AND
R ADIATING S YSTEMS
xx
cs
t
A
xx
q
xx
v
=
40 c2 s2 |x x |
c |x x |
t x
A ( ) t A = ( )t A
(8.84)
B(t, x) =
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 |
(8.85)
so that
( )t
A
t
E(t, x)
xx
c |x x
xx
=
c |x x
B(t, x) =
(8.86)
The radiation part of the electric eld is obtained from the acceleration
eld in Formula (8.81) on the previous page as
lim E(t, x)
|xx |
q
|x x | v
(x x ) (x x )
v
2 s3
40 c
c
q
=
(x x ) [(x x0 ) v]
40 c2 s3
=
Erad (t, x) =
(8.87)
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)
(x x ) v
(x x ) v
+
2 xx
c
c
s (t , x) = x x
(8.89)
8.3
R ADIATION
FROM A
L OCALISED C HARGE
IN
131
A RBITRARY M OTION
(x x ) v
c
(x x ) v
c
|x x |2 v2
|x x |2 v2 2
cos2 +
sin
c2
c2
|x x |2 v2
|x x |2 v2
=
(cos2 + sin2 ) =
c2
c2
(8.90)
we nd that
(x x ) v
|x x |2 v2
2
c
c
(8.91)
(x x ) v
c
Furthermore, from Equation (8.82) on page 129, we obtain the following identity:
(x x ) v = (x x0 ) v
(8.92)
(x x0 ) v
|x x |2 v2
=
c2
c
(8.93)
(x x ) v
c
Inserting the above into expression (8.89) on the preceding page for s2 , this
expression becomes
(x x0 ) v
c
(x x0 (t)) v
= (x x0 (t))2
c
(x x0 ) v
c
(8.94)
|x x0 |2
|x x | v
c
= (x x )
(x x ) v |x x |2 v2
(x x0 ) v
2 xx
+
c
c2
c
s = xx
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):
v2
1 2
c
v2 2
sin 0
c2
(8.95a)
(8.95b)
(x x0 ) v
+
c
(8.95c)
|x x0 |
= |x x0 |
(x x0 ) v
c
|x x0 |2
s(t, x) =
(x x0 ) v
c
vv
(x x0 )
c2
(8.96)
v2
c2
= 2 (x x0 ) 1
v2
c2
|x x0 |2 1
s2 =
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.
T HE
V
WG
E XAMPLE 8.1
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
8.3
R ADIATION
FROM A
L OCALISED C HARGE
133
A RBITRARY M OTION
IN
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.
6e
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
f7
v
v v
vv v
(x x0 ) 2
(x x0 )
c c
c
c
c
h
f 7
e
" !
q
v v
v2
=
(x x0 ) +
(x x0 ) (x x0 ) 2
40 s3
c c
c
v v
(x x0 )
c c
(8.100)
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.
"
From Equation (8.100) we conclude that E is directed along the vector from the simul-
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
0 q
4s3
v2
c2
"
B(t, x) =
v (x x0 ) =
1
vE
c2
(8.101)
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
E XAMPLE 8.2
T HE
OF EXAMPLE
8.1
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),
1 2
c2 t2
"
(t, x) =
(t, x) = f (t, x)
(8.102)
"
v2
c2
2 2 2
+
+
= f (x)
x2 x2 x2
1
2
3
(8.103)
1 =
2 = x 2
3 = x 3
x1
(8.104a)
1 v2 /c2
(8.104b)
(8.104c)
def
(/1 , /2 , /3 ),
8.3
R ADIATION
FROM A
L OCALISED C HARGE
IN
135
A RBITRARY M OTION
2
()
1 v2 /c2 1 , 2 , 3 ) f ()
(8.105)
in this space. This equation has the well-known Coulomb potential solution
f ( ) 3
d
((
1
4
(8.106)
((
() =
1
4
f (x ) 3
dx
s
(8.107)
1
2
[(x2 x2 ) + (x3 x3 ) ]
(8.108)
"
v2
s = (x1 x1 ) + 1 2
c
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) =
(8.109a)
(8.109b)
q
40 s
qv
A(t, x) =
40 c2 s
(t, x) =
(8.110a)
(8.110b)
(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-
136
E LECTROMAGNETIC R ADIATION
R ADIATING S YSTEMS
AND
6e
f7
=q
v
E
c2
f7
F = q E+v
(8.112)
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
(8.113)
(8.114)
where
v2
c2
"
= 1
(8.115)
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
v2
c2
q
40 s
"
= 1
(8.116)
1 v2 /c2
x1 x 1
+ (x2 x2 )2 + (x3 x3 )2
(8.117)
OF EXAMPLE
8.2
8.3
R ADIATION
FROM A
L OCALISED C HARGE
137
A RBITRARY M OTION
IN
s = xx
(8.118)
|x x | v
xx ,
c
(8.119)
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],
(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
(8.121)
(8.122)
x x = x x0
(8.123a)
(8.123b)
v
0 q 2 ( )2
xx
sin2
2 c |x x |2
|x x |
16
(8.124)
q 2 v2
0 q 2 ( )2
v
=
6c
60 c3
(8.125)
where is the angle between v and x x0 . The total radiated power (integrated
over a closed spherical surface) becomes
P=
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
(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
sin2
0 q 2 v 2
162 c |x x |2 1 v cos
c
S=
xx
|x x |
(8.128)
It is interesting to note that the magnitudes of the electric and magnetic elds
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
139
A RBITRARY M OTION
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 :
t
t
dU rad
dU rad
d =
dt
dt
(8.130)
(8.131)
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
sin2
0 q 2 v 2
dU rad ()
d =
dt
162 c 1 v cos
c
(8.132)
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
140
E LECTROMAGNETIC R ADIATION
AND
R ADIATING S YSTEMS
dS
dr
d
q
x2 vdt x1
x x2 + c dt
((
((
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
dS
x x2
d =
(8.133)
is
d3x = dS dr = x x2
d dr
(8.134)
Here, dr denotes the differential distance between the two spheres and can be
evaluated in the following way
dt =
(8.135)
cs
dt
x x2
x
iw
v cos
x x2
v
x x2
x x2
v dt x x2
x x2
dr = x x2 + c dt
8.3
R ADIATION
L OCALISED C HARGE
FROM A
IN
141
A RBITRARY M OTION
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
1
(8.137)
1 v2
c
dU rad 0 q 2 v2
=
dt
6c
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.
B REMSSTRAHLUNG FOR LOW SPEEDS AND SHORT ACCELERATION TIMES
E XAMPLE 8.3
v(t ) = v (t t0 )
(8.138)
v dt
((
s xx
(8.139)
((
(8.141)
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 |
(8.142)
142
E LECTROMAGNETIC R ADIATION
AND
R ADIATING S YSTEMS
((
E
c
((
B=
(8.143)
Fourier transforming expression (8.142) on the preceding page for E is trivial, yielding
E =
q sin
v eit0
2
0 c |x x |
(8.144)
82
= 0 c
1
0
EB dt d2x =
1
0 c
B
0
"
dU rad
dt =
dt
U rad =
dS dt
E 2 dt d2x
(8.145)
E 2 dt d2x
According to Parsevals identity [cf. Equation (7.35) on page 105] the following equality holds:
E 2 dt = 4
0
|E |2 d
(8.146)
|E |2 d2x d
(8.147)
"
rad
U d = 40 c
q
163 0 c3
q2
30 c
(v)2
d
0
v
c
sin2 sin d d
(8.148)
0
2
"
sin2 2
d x d
|x x |2
d
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
8.3
R ADIATION
FROM A
L OCALISED C HARGE
IN
143
A RBITRARY M OTION
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
"
v
c
d
1 2
137 3
2
e2
40 hc 3
v
c
"
dN =
(8.150)
(8.151)
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
8.3
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
(8.152a)
x (t ) = a[ x1 cos (t ) + x2 sin (t )]
(8.152b)
v = |v| = a0
0
v = | | =
v
a2
0
(8.152c)
(8.152d)
(8.152e)
(8.152f)
x x = x x ( x2 sin + x3 cos )
(8.153)
144
E LECTROMAGNETIC R ADIATION
AND
R ADIATING S YSTEMS
x2
(t, x)
xx
x
v
q
(t , x )
v
(t )
x1
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
(8.154a)
(8.154b)
where in the last step we simply used the denition of a scalar product and the
1
1
xx
(E B) =
|E|2
0
c0
|x x |
(8.155)
(8.156)
8.3
R ADIATION
FROM A
L OCALISED C HARGE
145
A RBITRARY M OTION
IN
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
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
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
1
(8.158)
1 v2
c
dU rad 0 q 2 v2
=
dt
6c
2. v/c
Cyclotron radiation
For a non-relativistic speed v
dU rad (, ) 0 q 2 v2
(1 sin2 sin2 )
=
2c
dt
16
(8.159)
(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.
146
E LECTROMAGNETIC R ADIATION
AND
R ADIATING S YSTEMS
x2
(t, x)
xx
v
q
(t , x )
v
(t )
0
x1
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
(8.162)
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
2
v
v
(8.163)
8.3
R ADIATION
L OCALISED C HARGE
FROM A
IN
147
A RBITRARY M OTION
cos 0 =
sin 0 =
(8.164a)
v2
c2
(8.164b)
1,
1 v2
c
v2
c2
1,
(8.165)
v2 1
=
c2
(8.166)
Hence, synchrotron radiation from ultra-relativistic charges is characterized by a radiation lobe width which is approximately
(8.167)
This angular interval is swept by the charge during the time interval
t =
(8.168)
(8.169)
in the direction toward the observer who therefore measures a pulse width of
length
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
y iw v
x
t = t
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
(x x )
dU rad () 0 q 2 |x x |
(x x )
=
dt
162 cs5
(8.172)
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
(8.173)
8.3
R ADIATION
L OCALISED C HARGE
FROM A
IN
149
A RBITRARY M OTION
vt
q
v = vx
|x x0 |
B
E z
Virtual photons
According to Formula (8.100) on page 133 and Figure 8.12,
1
v2
c2
(x x0 ) x3
q
40 s3
E = Ez =
(8.174)
E =
(8.175)
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
q
42
0 bv
E (t) eit dt =
b
K1
v
1
2
b
v
E, =
(8.176)
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
E, 0,
0 bv
b
v
(8.177a)
(8.177b)
150
E LECTROMAGNETIC R ADIATION
AND
R ADIATING S YSTEMS
bmax
bmin
2
E d3x = 0
U = 0
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
v/
bmin
db
d
b
q2
2
2 0 v
bmin
bmax
U d = 40 v
E,
U=
d 2b db
(8.179)
(8.180)
bmin
N d
(8.181)
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
c
2
ln
p1 p1
N d
(8.182)
2 , we see that
E1 and E1 = m0 c
E1 cp1 cp1
m0 c2 E 1 E 1
2
ln
N d
(8.183)
8.3
R ADIATION
FROM A
L OCALISED C HARGE
IN
151
A RBITRARY M OTION
(8.184)
B = (t, x)H = m 0 H
(8.185)
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)
(8.188)
0 m 0
m
(8.189)
152
E LECTROMAGNETIC R ADIATION
AND
R ADIATING S YSTEMS
i = 1, 2, 3
(8.190)
(8.191)
k k k = k =
v
v
v v
(8.192)
(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)
(8.195)
v =
8.3
R ADIATION
FROM A
L OCALISED C HARGE
IN
153
A RBITRARY M OTION
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 =
(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
2
p
2
(8.197)
N q2
0 m
(8.198)
n2 () = 1
where
2 =
p
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
(8.199a)
= 0
(8.199b)
c
1
=
<c
n
0
(8.200)
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
q
1 q
1
=
40 |x x | n (xx )v
40 s
c
1
qv
1 qv
=
2
(xx )v
40 c |x x | n
40 c2 s
c
(8.201b)
A(t, x) =
(8.201a)
(t, x) =
where now
(x x ) v
c
s = xx n
(8.202)
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+
(8.203a)
(8.203b)
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)
8.3
R ADIATION
L OCALISED C HARGE
IN
155
A RBITRARY M OTION
FROM A
x(t)
x (t )
For v/c 1/n, the retarded distance s, and therefore the denominators in
Equations (8.201) on the facing page vanish when
nv
v
= xx
cos c = x x
c
c
n(x x )
(8.205)
c
nv
(8.206)
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)
c
nv
(8.208)
156
E LECTROMAGNETIC R ADIATION
AND
R ADIATING S YSTEMS
with speed c/n in the form of a shock front, called Vavilov-Cerenkov radiation.1
(8.209)
(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]
d
x
kx cos
exp i
v
dx
(8.212)
q 2 n2
=
163 0 c3
(j k)eikx d3x
1
40 nc
rad
U d
sin d
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
8.3
R ADIATION
L OCALISED C HARGE
FROM A
IN
157
A RBITRARY M OTION
where is the angle between the direction of motion, x1 , and the direction to
Q
R
rad
U d =
(8.213)
Q
sin2
1 + nv
c
1 + nv
c
X
v
2
1
1
rad
U () d
(8.214)
V G
F
oU
F
iU
q 2 n2 sin2 c
22 0 c3
rad
U () sin d = cos = = 2
rad
U = 2
sin2 x
dx
2
x
(8.215)
c2
cX
1 2 2
n
n v
V G
F
oU
c2
n2 v2
d 1
F
iU
1 + nv
c
X
v
2
1 + nv
c
sin c
sin2
leading to the nal approximate result for the total energy loss in the frequency
interval (, + d)
1
c2
n2 v2
q 2X
rad
U d =
20 c2
(8.216)
c2
n2 ()v2
rad
U d
q 2
=
2X
40 c2
(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)
B = 0
(F.2)
E =
(F.3)
B
t
H = j+ D
t
(F.4)
Constitutive relations
D = E
B
H=
(F.5)
j = E
(F.7)
P = 0 E
(F.8)
(F.6)
E =
(F.9)
A
t
(F.10)
159
160
F ORMULAE
1
=0
c2 t
(F.11)
(F.12)
(F.13)
kE
c
(F.14)
40 c |x|
V
Brad (x) =
(F.15)
(F.16)
40 |x|
Brad (x) =
(F.17)
(F.18)
F.3
161
S PECIAL R ELATIVITY
40 c |x|
Brad (x) =
(F.19)
(F.20)
80 |x|
(F.21)
(F.22)
Brad (x) =
Q
v
c
x x0 = (x x ) |x x |
t
t
=
x
s = x x (x x )
+ (x x )
(x x0 ) v
c2
q
v2
(x x0 ) 1 2
40 s3
c
E(t, x)
B(t, x) = (x x )
c|x x |
E(t, x) =
(F.23)
(F.24)
(F.25)
v
c
(F.26)
|x x |
s
(F.27)
(x x0 ) v
c
(F.28)
(F.29)
|x x0 |2
s=
q
v2
(x x0 ) 1 2
40 s3
c
v E(t, x)
B(t, x) =
c2
E(t, x) =
(F.30)
162
F ORMULAE
hh
ig
(F.31)
deef
g =
1 0
0
0
0 1 0
0
0 0 1 0
0 0
0 1
(F.32)
(F.33)
hh
ig
p
0 0
0 0
=
0
0
1 0
0
0
0 1
1
=
1 2
v
=
c
(F.34)
(F.35)
(F.36)
deef
dt
= c d
(F.37)
F.3.5 Four-velocity
u =
dx
= (c, v)
d
(F.38)
F.3.6 Four-momentum
E
,p
c
p = m0 cu =
(F.39)
F.4
163
V ECTOR R ELATIONS
(F.40)
F.3.8 Four-potential
,A
c
(F.41)
A =
hh
ig
p
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
(F.42)
xi
i=1
def
def
xi
xi
xi
(F.43)
,
,
x1 x2 x3
= i =
, ,
x y z
(F.44)
= sin x1 + cos x2
(F.45a)
(F.45b)
(F.45c)
164
F ORMULAE
(F.46a)
(F.46b)
x3 = cos sin
(F.46c)
dx x = dl = dr r + r d + r sin d
(F.47)
(F.48)
(F.49)
Volume element
d3x = dV = drdS = r2 dr d
(F.50)
(F.51)
a (b c) = (a b) c
(F.53)
a b = b a =
i jk a j bk xi
(F.52)
a (b c) = b(a c) c(a b)
(F.54)
(F.56)
a (b c) + b (c a) + c (a b) = 0
(F.55)
(a b) (c d) = (a b d)c (a b c)d
(F.57)
F.4
165
V ECTOR R ELATIONS
(F.58)
(a) = a + a
(F.59)
(a b) = b ( a) a ( b)
(F.61)
(a) = a a
(F.60)
(a b) = a( b) b( a) + (b )a (a )b
(a b) = a ( b) + b ( a) + (b )a + (a )b
2
(F.62)
(F.63)
(F.64)
=0
(F.65)
( a) = 0
(F.66)
( a) = ( a)
(F.67)
Special identities
In the following k is an arbitrary constant vector while x, as before, is the
radius vector.
x = 3
(F.68)
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)
(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
dS a
dS
(F.77)
(F.78)
( a) d3x =
( ) d3x =
V
V
( a) d3x =
dS a
(F.79)
If S (C) is an open surface bounded by the contour C(S ), whose line element is dl, then
dl =
S
a dl =
dS
dS ( a)
(F.80)
(F.81)
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.
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
x = x i xi x i xi
def
(M.1)
i=1
xi (x1 , x2 , x3 ) (x, y, z)
(M.2)
x (x0 , x1 , x2 , x3 )
(M.3)
x (x0 , x1 , x2 , x3 )
(M.4)
x
dx
x
dx =
(M.5)
M.1
S CALARS , V ECTORS
AND
169
T ENSORS
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
3
by
def
(M.6)
(x0 , x1 , x2 , x3 ) (x )
(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
(M.8)
Vector elds
as follows:
a(x) = xi ai (x)
(M.9)
(M.10)
170
M ATHEMATICAL M ETHODS
(M.11)
(M.12)
x
y
x
(M.13)
x
y
x
(M.14)
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:
def
Ai j (xk )
fd
A(x) =
(M.15)
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.
S CALARS , V ECTORS
AND
0 if i = j
1 if i = j
i j =
171
T ENSORS
M.1
(M.16)
g
p
fd
(i j ) =
1 0 0
0 1 0
0 0 1
(M.17)
`a
i jk
1
0
1
(M.18)
ac
= jl km jm kl
(M.19)
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
E XAMPLE M.1
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
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
3
T n = ni T xi n i T x i
(M.23)
i=1
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.1
S CALARS , V ECTORS
AND
173
T ENSORS
T nj = (T n) j = ni T i j ni T i j
(M.25)
i=1
Using Equation (M.25), we nd that the component of the vector T n in the direction
j=1
ni T i j
j=1
i=1
= T nj m j =
T nm = T n m
(M.26)
m j ni T i j m j = n T m
Hence, the jth component of the vector T xi , here denoted T i j , can be interpreted as the
f d3x +
v dm =
V
T n d2x
(M.27)
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
d
v j d3x =
dt
f j d3x +
T i j 3
dx
xi
(M.29)
Since this formula is valid for any arbitrary volume, we must require that
T i j
d
vj fj
=0
dt
xi
(M.30)
or, equivalently
T i j
v j
+ v v j f j
=0
t
xi
(M.31)
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
M.1
174
M ATHEMATICAL M ETHODS
a (x ) g a (x )
(M.32)
This rule is often called lowering of index. The raising of index analogue of
the index lowering rule is:
def
a (x ) g a (x )
(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
(M.34)
2 ...k1
...k1
gk k A1k+1 ...n (x ) = A1 2k+2 ...nk (x )
k+1
k
(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.
C ONTRAVARIANT AND COVARIANT VECTORS IN
E XAMPLE M.2
FLAT
L ORENTZ SPACE
M.1
S CALARS , V ECTORS
AND
175
T ENSORS
metric tensor
1
if = = 0
1 if = = i = j = 1, 2, 3
0
if =
s
qr
g =
(M.37)
rt
(M.38)
@@ 8B
(g ) =
1 0
0 1
0 0
0 0
2445
i.e., a matrix with a main diagonal that has the sign sequence, or signature, {+, , , }
or
s
qr
g =
1 if = = 0
1
if = = i = j = 1, 2, 3
0
if =
(M.39)
rt
0
0
0
1
(M.40)
@@
AB8
2445
(g ) =
1 0
0 1
0 0
0 0
a (a0 , a1 , a2 , a3 ) = (a0 , a)
(M.41)
According to the index lowering rule, Equation (M.32) on the facing page, we obtain
the covariant version of this vector as
def
a (a0 , a1 , a2 , a3 ) = g a
(M.42)
a 0 = 1 a 0 + 0 a 1 + 0 a 2 + 0 a 3 = a0
0
(M.43)
1
(M.44)
a2 = 0 a0 + 0 a1 1 a2 + 0 a3 = a2
(M.45)
a1 = 0 a 1 a + 0 a + 0 a = a
0
a3 = 0 a + 0 a + 0 a + 1 a = a
(M.46)
or
a = (a0 , a1 , a2 , a3 ) = (a0 , a1 , a2 , a3 ) = (a0 , a)
(M.47)
176
M ATHEMATICAL M ETHODS
(M.48)
E ND OF
EXAMPLE
M.2
a b = x i ai x j b j = x i x j ai b j = i j ai b j = ai bi
(M.49)
(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 .
I NNER PRODUCTS IN COMPLEX VECTOR SPACE
'
def
'
&
def
A aR + iaI = aR aR + iaI aI A A
&
E XAMPLE M.3
6 ),
expressed in terms of
(M.52)
A=
a2 a2 + 2iaR aI
R
I
'
'
def
A2 A A = a2 a2 + 2iaR aI A2
R
I
(M.53)
(M.54)
Using this in Equation (M.52), we see that we can interpret this so that the complex
M.1
S CALARS , V ECTORS
AND
177
T ENSORS
unit vector is
a2 a2 2iaR aI
R
I
a2 + a 2
R
I
a2 a2 + 2iaR aI
R
I
aI
aR + i
aR
aI
aR + i
aI
a2 a2 2iaR aI
R
I
a2 + a 2
R
I
aI
'
a2 a2 + 2iaR aI
R
I
aR
A
A= =
A
(M.55)
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
def
|A|2 A A = a2 + a2 = |A|2
R
I
(M.56)
a2 + a 2
R
I
a2 + a 2
R
I
a2 + a 2
R
I
aR + i
aI
aI
(M.57)
a2 + a 2
R
I
aR
a2 + a 2
R
I
aI
aR + i
a2 + a 2
R
I
aI
'
aR
A =
A=
|A|
E ND OF
METRIC IN
EXAMPLE
M.3
L ORENTZ SPACE
E XAMPLE M.4
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
a b = (a0 , a) (b0 , b) = a0 b0 a b
(M.58)
(M.59)
The
(M.60)
E ND OF
EXAMPLE
M.4
178
E XAMPLE M.5
M ATHEMATICAL M ETHODS
M ETRIC IN
GENERAL RELATIVITY
0
0
r2
0
0
0
0
r2 sin2
(M.61)
@@ B8
2445
(g ) =
e
0
0
0
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
M.5
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
def
def
def
def
A c ab c a(b c)
(M.63a)
c A c ab (c a)b
(M.63b)
fd p
fd R
x3
x2
x1
x2
x3
ab = x1
a1 b1 a1 b2 a1 b3
a1 b2 a2 b2 a2 b3
a1 b3 a3 b2 a3 b3
(M.64)
which means that we can represent the tensor A(x) in matrix form as
a1 b1 a1 b2 a1 b3
a1 b2 a2 b2 a2 b3
a1 b3 a3 b2 a3 b3
fd
Ai j (xk ) =
(M.65)
which we identify with expression (M.15) on page 170, viz. a tensor in matrix
notation.
M.1
S CALARS , V ECTORS
AND
179
T ENSORS
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 =
i jk a j bk xi
(M.66)
def
xi
def
xi
(M.67)
where xi is the ith unit vector in a Cartesian coordinate system. Since the
i =
(M.68)
180
M ATHEMATICAL M ETHODS
,
,
,
x0 x1 x2 x3
(M.70)
(M.71)
y = x y
(M.72)
and
respectively.
L ORENTZ SPACE
1
,
c t
In
"
1
, =
c t
"
1
,
c t
(M.74)
"
(M.73)
"
T HE
E XAMPLE M.6
(M.75)
1 2
c2 t2
E ND OF
EXAMPLE
M.6
M.1
S CALARS , V ECTORS
AND
181
T ENSORS
With the help of the del operator we can dene the gradient, divergence
and curl of a tensor (in the generalised sense).
The gradient
3
(x), is an
The gradient of an
a(x):
vector eld
(M.76)
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
E XAMPLE M.7
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:
xx
=
|x x |3
"
1
|x x |
|x x |
(M.79)
1
|x x |
(M.80)
"
#
and
|x x |
xx
|x x |
=
= xi
xi
|x x |
xi
|x x | = xi
E ND OF
EXAMPLE
M.7
182
M ATHEMATICAL M ETHODS
The divergence
3
as
ai (x)
= (x)
xi
(M.81)
a (x ) = a (x ) =
(M.82)
D IVERGENCE IN 3D
"
a(x )
|x x |
a(x )
+ a(x )
|x x |
1
|x x |
(M.83)
"
For an arbitrary
&
E XAMPLE M.8
a (x )
x
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
M.8
The Laplacian
The 3D Laplace operator or Laplacian can be described as the divergence of
the gradient operator:
2
==
2
2
xi x j
= i j i j = 2 = 2 2
i
xi
x j
xi i=1 xi
(M.84)
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.
M.1
S CALARS , V ECTORS
AND
183
T ENSORS
"
1
x |
|x
is
1
x |
|x
"
&
E XAMPLE M.9
= 4(x x )
(M.85)
EXAMPLE
M.9
The curl
a(x) =
i jk xi j ak (x) = i jk xi
ak (x)
= b(x)
x j
vector eld
(M.86)
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 )
(M.87)
E XAMPLE M.10
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)
(M.88)
2
2
x2 x3 x3 x2
(x) x1
2
2
x3 x1 x1 x3
+
+
(x) xi
x j xk
2
2
x1 x2 x2 x1
(x) x2
"
i jk
"
"
i jk xi j k (x)
(M.89)
(x) x3
184
M ATHEMATICAL M ETHODS
We thus nd that
[ (x)] 0
(M.90)
&
(M.91)
E XAMPLE M.11
T HE
EXAMPLE
&
3.
M.10
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),
a1 (x)
2
2
x3 x1 x1 x3
ak
x j
2
2
x2 x3 x3 x2
+
+
i jk
2
2
x1 x2 x2 x1
(M.93)
a2 (x)
"
xi
"
i jk j ak (x)
"
a3 (x)
i.e., that
[ a(x)] 0
(M.94)
&
G = a (x )
a (x ) = 0
(M.95)
E ND OF
EXAMPLE
M.11
M.2
185
A NALYTICAL M ECHANICS
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].
L(qi , qi , t) = L qi ,
(M.96)
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
L
qi
L
=0
qi
(M.97)
L
qi
(M.98)
qi
(M.99)
(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
186
M ATHEMATICAL M ETHODS
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
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
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
190
M ATHEMATICAL M ETHODS
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