EECS 117

Lecture 20: Plane Waves

Prof. Niknejad
University of California, Berkeley

University of California, Berkeley

EECS 117 Lecture 20 p. 1/2

Maxwells Eq. in Source Free Regions

In a source free region = 0 and J = 0
D=0
B=0

B
H
E=
=
t
t
D
E
=
H=
t
t
Assume that E and H are uniform in the x-y plane so

=
0
and
x
y = 0

For this case the E simplifies

University of California, Berkeley

EECS 117 Lecture 20 p. 2/2

Curl E for Plane Uniform Fields

Writing out the curl of E in rectangular coordinates

x
y

z

E= 0
0 z
Ex Ey Ez
Ey
Hx
( E)x =
=
z
dt
Hy
Ex
=
( E)y =
z
dt
Hz
( E)z = 0 =
dt

University of California, Berkeley

EECS 117 Lecture 20 p. 3/2

Curl of H for Plane Uniform Fields

Similarly, writing out the curl H in rectangular
coordinates
E
H=
t
Hy
Ex
=

z
t
Ey
Hx
=
z
t
Ez
0=
t
Time variation in the
z direction is zero. Thus the fields
are entirely transverse to the direction of propagation.
We call such fields TEM waves
University of California, Berkeley

EECS 117 Lecture 20 p. 4/2

Polarized TEM Fields

For simplicity assume Ey = 0. We say the field polarized
in the x
-direction. This implies that Hx = 0 and Hy 6= 0
Hy
Ex
=
z
t
Hy
Ex

=
z
t
2 Hy
2 Ex
=
2
z
zt
2 Hy
2 Ex
= 2
zt
t

University of California, Berkeley

EECS 117 Lecture 20 p. 5/2

One Dimensional Wave Eq.

We finally have it, a one-dimensional wave equation
2 Ex
2 Ex
= 2
2
z
t

Notice similarity between this equation and the wave

equation we derived for voltages and currents along a
transmission line
As before, the wave velocity is v =

The general solution to this equation is

z
z
Ex (z, t) = f1 (t ) + f2 (t + )
v
v

University of California, Berkeley

EECS 117 Lecture 20 p. 6/2

Wave Solution
Lets review why this is the general solution
1 1
Ex
= f1 + f2
z
v
v

Ex
= f1 + f2
t

2E
2 Ex

1
1
x

= f1 + f2
= 2 f1 + 2 f2
2
2
t
z
v
v
A point on the wavefront is defined by (t z/v) = c
where c is a constant. The velocity of this point is
therefore v
1 z
1
=0
v t

z
=v
t
University of California, Berkeley

EECS 117 Lecture 20 p. 7/2

Wave Velocity
We have thus shown that the velocity of this wave
moves is
1
v=c=

In free-space, c 3 108 m/s, the measured speed of

light
In a medium with relative permittivity r and relative
permeability r , the speed moves with effective velocity
c
v=
r r

This fact alone convinced Maxwell that light is an EM

wave
University of California, Berkeley

EECS 117 Lecture 20 p. 8/2

Sinusoidal Plane Waves

For time-harmonic fields, the equations simplify
dEx
= jHy
dz
dHy
= jEx
dz
This gives a one-dimensional Helmholtz equation
d2 Ex
2
=

Ex
2
d z

University of California, Berkeley

EECS 117 Lecture 20 p. 9/2

Solution of Helmholtz Eq.

The solution is now a simple exponential
Ex = C1 ejkz + C2 ejkz

The wave number is given by k = =

We can recover a traveling wave solution

jt
Ex (z, t) = Ex e


Ex (z, t) = C1 ej(tkz) + C2 ej(t+kz)

Ex (z, t) = C1 cos (t kz) + C2 cos (t + kz)

The wave has spatial variation =

University of California, Berkeley

2
k

2v

v
f

EECS 117 Lecture 20 p. 10/2

Magnetic Field of Plane Wave

We have the following relation

1 dEx
1 
Hy =
jkC1 ejkz + C2 jkejkz
=
j dz
j

k 
C1 ejkz C2 jkejkz
Hy =

By definition, k =
r 


Hy =
C1 ejkz C2 ejkz

The ratio Ex+ and Hy+ has units of impedance and is

p
given by the constant = /. is known as the
impedance of free space
University of California, Berkeley

EECS 117 Lecture 20 p. 11/2

Plane Waves
Plane waves are the simplest wave solution of
Maxwells Eq. They seem to be a gross
oversimplification but they nicely approximate real
waves that are distant from their source

source of radiation

wavefront at distant points

University of California, Berkeley

EECS 117 Lecture 20 p. 12/2

Wave Equation in 3D
We can derive the wave equation directly in a
coordinate free manner using vector analysis
H
( H)
E =
=
t
t

Substitution from Maxwells eq.

D
E
H=
=
t
t
2E
E = 2
t

University of California, Berkeley

EECS 117 Lecture 20 p. 13/2

Wave Eq. in 3D (cont)

Using the identity E = 2 E + ( E)
Since E = 0 in charge free regions
2E

2 E = 2
t

In Phasor form we have k 2 = 2

2 E = k 2 E

Now its trivial to get a 1-D version of this equation

2E

x
2
Ex = 2
t

University of California, Berkeley

2 Ex
2 Ex
= 2
2
x
t

EECS 117 Lecture 20 p. 14/2

Penetration of Waves into Conductors

Inside a good conductor J = E
In the time-harmonic case, this implies the lack of free
charges
E
H=J+
= ( + j) E
t

Since H 0, we have
( + j) E 0

Which in turn implies that = 0

For a good conductor the conductive currents
completely outweighs the displacement current, e.g.

University of California, Berkeley

EECS 117 Lecture 20 p. 15/2

Conductive vs. Displacement Current

To see this, consider a good conductor with 107 S/m
up to very high mm-wave frequencies f 100GHz
The displacement current is still only

1011 1011 1

This is seven orders of magnitude smaller than the

conductive current
For all practical purposes, therefore, we drop the
displacement current in the volume of good conductors

University of California, Berkeley

EECS 117 Lecture 20 p. 16/2

Wave Equation inside Conductors

Inside of good conductors, therefore, we have
H = E
E = ( E) 2 E = 2 E = j H
2 E = jE

One can immediately conclude that J satisfies the same

equation
2 J = jJ
Applying the same logic to H, we have

H = ( H)2 H = 2 H = (j+) E
2 H = jH
University of California, Berkeley

EECS 117 Lecture 20 p. 17/2

Plane Waves in Conductors

Lets solve the 1D Helmholtz equation once again for
the conductor
d2 Ez
2
=
jE
=

Ez
z
2
d x

We define 2 = j so that
1 + j
=

Or more simply, = (1 + j) f =

1+j

The quantity = f1 has units of meters and is an

important number
University of California, Berkeley

EECS 117 Lecture 20 p. 18/2

Solution for Fields

The general solution for the plane wave is given by
Ez = C1 e x + C2 e x

Since Ez must remain bounded, C2 0

Ez = E0 e x = Ex ex/ e|jx/
| {z } {z }
mag

phase

Similarly the solution for the magnetic field and current

follow the same form
Hy = H0 ex/ ejx/
Jz = J0 ex/ ejx/
University of California, Berkeley

EECS 117 Lecture 20 p. 19/2

Penetration Depth
The wave decays exponentially into the conductor. For
this reason, is called the penetration depth, or more
commonly, the skin depth. The fields drop to 1/e of their
values after traveling one skin depth into the conductor.
After several skin depths, the fields are essentially zero
You may also say that the wave exists only on the skin
of the conductor
For a good conductor at f = 1GHz
1
=
106 m
f

As the frequency is increased, 0, or the fields

completely vanish in the volume of the conductor
University of California, Berkeley

EECS 117 Lecture 20 p. 20/2

Total Current in Conductor

Why do fields decay in the volume of conductors?
The induced fields cancel the incoming fields. As
, the fields decay to zero inside the conductor.
The total surface current flowing in the conductor
volume is given by
Z
Z
Jsz =
Jz dx =
J0 e(1+j)x/ dx
0

Jsz

J0
=
1+j

At the surface of the conductor, Ez0 =

University of California, Berkeley

J0

EECS 117 Lecture 20 p. 21/2

Internal Impedance of Conductors

Thus we can define a surface impedance
Ez0
1+j
Zs =
=
Jsz

Zs = Rs + jLi

The real part of the impedance is a resistance

r
1
f
Rs =
=

The imaginary part is inductive

Li = Rs

So the phase of this impedance is always /4

University of California, Berkeley

EECS 117 Lecture 20 p. 22/2

Interpretation of Surface Impedance

The resistance term is equivalent to the resistance of a
conductor of thickness
The inductance of the surface impedance represents
the internal inductance for a large plane conductor
Note that as , Li 0. The fields disappear from
the volume of the conductor and the internal impedance
is zero
We commonly apply this surface impedance to
conductors of finite width or even coaxial lines. Its
usually a pretty good approximation to make as long as
the conductor width and thickness is much larger than

University of California, Berkeley

EECS 117 Lecture 20 p. 23/2