Solutions to Problems in Jackson,

Classical Electrodynamics, Third Edition

Homer Reid
May 24, 2002
Chapter 8: Waveguide Derivations
Before starting the problems, I thought it would be useful to run through
my own derivations of some of the formulas from this chapter.
Waveguides and cavities: basic pedagogy
The unifying feature of waveguide and cavity problems is that we single out
one spatial coordinate and announce from the start that the elds will have
sinusoidal dependence on that coordinate. Taking the special coordinate to be z,
this means that all components of all elds have the functional form f(x, y)e
ikz
for some wavevector k.
Assuming harmonic time dependence, we write explicitly
E(x) =
_
E
x
(x, y)i + E
y
(x, y)j + E
z
(x, y)k
_
e
i(kzt)
B(x) =
_
B
x
(x, y)i + B
y
(x, y)j) + B
z
(x, y)k
_
e
i(kzt)
(1)
We have here a total of six functions f(x, y) that we must nd to satisfy
Maxwells equations with the relevant boundary conditions. At rst this would
appear tough since the six elds are all coupled by Maxwells equations, but
after a little algebra we obtain the following simplied situation: The z di-
rection elds E
z
(x, y) and B
z
(x, y) turn out to satisfy (separately) simple one-
dimensional dierential equations, which may be readily solved upon specifying
the boundary conditions for a particular situation. Meanwhile, the remaining
elds (E
x
, E
y
, B
x
, B
y
) can be expressed simply as linear combinations of E
z
and
B
z
and their derivatives, so once we obtain the z elds we have everything. In
what follows well derive the dierential equations satised by E
z
and B
z
and
the equations giving the remaining elds in terms of them.
1
Homer Reids Solutions to Jackson Problems: Chapter 8 2
The dierential equations for E
z
and B
z
The Maxwell curl equations are
E =
B
t
, B =
1
c
2
m
E
t
where c
m
is the speed of light in the medium. We begin by applying the rst
curl equation to our ansatz (1), obtaining

y
E
z
ikE
y
= iB
x
(2)

x
E
z
+ ikE
x
= iB
y
(3)

x
E
y

y
E
x
= iB
z
, (4)
and we pause to solve the rst two of these for E
x
and E
y
:
E
x
=

k
B
y

i
k

x
E
z
, E
y
=

k
B
x

i
k

y
E
z
. (5)
Next we apply the second curl equation to our ansatz, obtaining

y
B
z
ikB
y
=
i
c
2
m
E
x
(6)

x
B
z
+ ikB
x
= i

c
2
m
E
y
(7)

x
B
y

y
B
x
= i

c
2
m
E
z
. (8)
But in (5) we solved for E
x
and E
y
, and if we then plug those solutions into (6)
and (7) we can solve for B
x
and B
y
in terms of B
z
and E
z
:
B
x
=
ikc
2
m

2
k
2
c
2
m
_

x
B
z
+

c
2
m
k

y
E
z
_
(9)
B
y
=
ikc
2
m

2
k
2
c
2
m
_

y
B
z

c
2
m
k

x
E
z
_
. (10)
Finally, with the ansatz (1) the equation B = 0 reads
B
x
x
+
B
y
y
= ikB
z
.
When we plug (9) and (10) into this, the terms involving E
z
elds cancel, and
we obtain an equation involving B
z
alone:
_

2
x
2
+

2
y
2
_
B
z
+
_

2
c
2
m
k
2
_
B
z
= 0
or
_

2
x
2
+

2
y
2
_
B
z
+
2
B
z
= 0 (11)
Homer Reids Solutions to Jackson Problems: Chapter 8 3
where
=
_

2
c
2
k
2
.
If we had carried out this derivation in the reverse order we would have obtained
the same equation for E
z
:
_

2
x
2
+

2
y
2
_
E
z
+ E
z
= 0. (12)
We can think of equations (11) and (12) as eigenvalue equations that have solu-
tions only for certain values of the parameter , which depend on the boundary
conditions.
Armed with equations (11) and (12) and the boundary conditions appropri-
ate to our problem we can now solve for B
z
and E
z
and then use (9) and (10)
to nd the remaining components of the B eld. The remaining components of
the E eld are given by analogous equations:
E
x
=
ikc
2
m

2
k
2
c
2
m
_

x
E
z
+

k

y
B
z
_
(13)
E
y
=
ikc
2
m

2
k
2
c
2
m
_

y
E
z

x
B
z
_
. (14)
Boundary Conditions; TE, TM, TEM Modes
The boundary conditions on the elds at the surfaces of the waveguide or cavity
are that E

and B

be continuous, where denotes the component of the

vector normal to the boundary surface and includes all other components of
the vector. This means that the two eigenvalue equations (11) and (12) must be
solved subject to dierent boundary conditions, which means in general their
eigenvalues will be dierent. If we have a solution of (12) for some value of
(i.e. for some combination of values of and k), then there will be no nonzero
solution of (11) for that value of , and hence we must have B
z
= 0 identically
for the mode at that frequency and wavevector. Since in this case the magnetic
eld has nonzero components only transverse to the direction of propagation,
this is called a transverse magnetic mode. Similarly, if (11) can be solved with
nonzero B
z
at some , then E
z
= 0 and we have a transverse electric mode. A
mode for which both E
z
and B
z
are zero is called a transverse electromagnetic
mode, and can only exist in the region between two conducting surfaces, not
within a single conductor as is possible for TE and TM modes.
Since either E
z
or B
z
is zero, we can simplify some of the equations above
and collect results appropriate to the two cases.
Homer Reids Solutions to Jackson Problems: Chapter 8 4
TM Modes TE Modes
B
z
0

2
t
E
z
+
2
E
z
= 0, E
n

S
= 0
E
x
=
ikc
2
m

2
k
2
c
2
m

x
E
z
E
y
=
ikc
2
m

2
k
2
c
2
m

y
E
z
B
x
=
i

2
k
2
c
2
m

y
E
z
B
y
=
i

2
k
2
c
2
m

x
E
z
E
z
0

2
t
B
z
+
2
B
z
= 0,
B
n
n

S
= 0
E
x
=
ic
2
m

2
k
2
c
2
m

y
B
z
E
y
=
ic
2
m

2
k
2
c
2
m

x
B
z
B
x
=
ikc
2
m

2
k
2
c
2
m

x
B
z
B
y
=
ikc
2
m

2
k
2
c
2
m

y
B
z
(A factor of e
i(kzt)
is understood in all of these expressions.)
For TM modes, the boundary condition is E

= 0, and E
z
is always perpen-
dicular to the boundary surfaces, so the boundary condition for the eigenvalue
equation is E
z
= 0. For the case of TE modes, the boundary condition is B

= 0.
Suppose one boundary surface is the yz plane. The normal to this plane is the
x direction, so B
x
must vanish at this surface; but we just saw that in the TE
case B
x

x
B
z
, i.e. the derivative of B
z
normal to the boundary surface must
vanish. This is general: the boundary condition for the eigenvalue equation in
the TM case is B
z
/n = 0.
Power ow; Energy Loss
The ow of power down a waveguide is described by the z component of the
Poynting vector S = EH =
1

EB. Using the boxed expressions above, for

the two types of modes we obtain
S
TM
z
=
1

(E
x
B
y
E
y
B
x
)
=
kc
2
m
(
2
k
2
c
2
m
)
2
_
(
x
E
z
)
2
+ (
y
E
z
)
2

e
2i(kzt)
or, in the time average,
=
kc
2
m
2(
2
k
2
c
2
m
)
2
_
(
x
E
z
)
2
+ (
y
E
z
)
2

=
k
2
4
_
(
x
E
z
)
2
+ (
y
E
z
)
2

. (15)
Homer Reids Solutions to Jackson Problems: Chapter 8 5
Similarly, for TE modes we obtain
S
TE
z
=
k
2
4
_
(
x
B
z
)
2
+ (
y
B
z
)
2

. (16)
To address the issue of dissipation in the boundaries, we solve Maxwells
equations within the boundary surfaces. The two curl equations are
E = iB (17)
B = J iE
= ( i) E
E (18)
since in most cases. (For example, for a copper waveguide with air
or vacuum interior we have we have 6 10
7

1
m
1
, while 9
10
12

1
m
1
( in rad/sec), so the approximation is good up to frequencies
10
19
rad/sec.)
Now we assume that the elds are only changing signicantly in the direc-
tion normal to the boundary surface (i.e., as we go deeper and deeper into the
boundary surface the elds die out rapidly, whereas as we move along parallel
to the boundary surface the elds dont change much) and keep only the normal
derivative in the curl equations. If measures the depth of penetration into the
surface, the curl equations become

E

= iB

B

= E
Dierentiating the rst of these, taking the cross product with of both sides,
and substituting in the second equation yields


2
E

2
= iE
or, using the bac-cab rule,

2
E

2


2
E

2
= iE.
Evidently the component of the LHS vanishes here, so E

= 0; the electric
eld within the conducting boundary has no component normal to the surface.
For the remaining components we obtain

2
E

2
+ iE

= 0
Homer Reids Solutions to Jackson Problems: Chapter 8 6
with solution
E

= e

i
E
0
= e
(1+i)

E
0
where =
_
2/ is the skin depth and E
0
is the eld just at the surface of
the boundary. To keep the solution from blowing up as we penetrate into the
conductor we take the negative sign in the exponent. From (17) we then obtain
B =
i 1

( E
0
)e
(1+i)

.
Evaluating this at the surface yields the modied boundary condition on the
elds in the cavity or waveguide:
B
0
=
i 1

( E
0
). (19)
From this equation we can work out the power loss per unit length in the
cavity or waveguide. The power dissipated in a volume dV is
_
(J E) dV =

_
E
2
dV. We integrate over the volume occupied by the boundary surfaces in
a length dz :
dP = dz
__ _

0
E
2
0
e
2(1+i)

d dl
_
e
2i(kzt)
= dz
_

2(1 + i)
_
E
2
0
dl
_
e
2i(kzt)
or, taking the time average,
dP
dz
=

4

2
_
E
2
0
dl (20)
where the line integral is over the cross section of the surface boundary at a
xed value of z.