Monday, March 11, 2013

em01 Fields ~E and ~B are in phase, reaching their maximum and minimum values at the

http://www.phy.iitb.ac.in/~dkg/PH-102/emw.pdf

Fields
~E and ~B are in phase, reaching their maximum and minimum values at the

same time. The electric field oscillates in the x-z plane and the magnetic field oscillates

in the y-z plane

ElectromagneticWaves :



In the absence of any source of charge or current, Maxwell’s equations in free

space are as follows :

~

~E = 0 (1)

~

~B = 0 (2)

~

×
~E =

∂~B

∂t

(3)

~

×
~B = ǫ0μ0

∂~E

∂t

(4)

The last two equations couple the electric and the magnetic fields. If
~B is time

dependent,
~∇ × ~E is non-zero. This implies that ~E is a function of position. Further,

if
∂~B /∂t itself changes with time, so does ~∇ × ~E . In such a case ~E also

varies with time since the
~operator cannot cause time variation. Thus, in general,

a time varying magnetic field gives rise to an electric field which varies both

in space and time. It will be seen that these coupled fields propagate in space.

We will first examine whether the equations lead to transverse waves. For simplicity,

assume that the electric field has only x-component and the magnetic field

only y-component. Note that we are only making an assumption regarding their

directions – the fields could still depend on all the space coordinates
x, y, z, in

addition to time
t.

Gauss’s law gives

~

~E =

∂E
x

∂x

+

∂E
y

∂y

+

∂E
z

∂z

= 0

Since only
Ex 6= 0, this implies

∂E
x

∂x

= 0

Thus
Ex is independent of x coordinate and can be written as Ex(y, z, t). A similar

analysis shows that
By is independent of y coordinate and can be written explicitly

as
By(x, z, t).

Consider now the time dependent equations eqns. (3) and (4). The curl equation

for
~B gives, taking z-component

∂B
y

∂x

∂B
x

∂y

=

∂E
z

∂t

= 0

1

Since
Bx = 0, this gives

∂B
y

∂x

= 0

showing that
By is independent of x and hence depends only on z and t. In a

similar manner we can show that
Ex also depends only on z and t. Thus the fields

~E

and
~B do not vary in the plane containing them. Their only variation takes

place along the z-axis which is perpendicular to both
~E and ~B. The direction of

propagation is thus
zdirection.

Z

X

Y

E

B B

E


To see that propagation is really a wave disturbance, take y-component of Eqn.

(3) and x-component of Eqn. (4)

∂E
x

∂z

=

∂B
y

∂t

(5)


∂B
y

∂z

=
μ0ǫ0

∂E
x

∂t

(6)

To get the wave equation for
Ex, take the derivative of eqn. (5) with respect to z

and substitute in eqn. (6) and interchange the space and time derivatives,

2Ex

∂z
2 = μ0ǫ02By

∂z∂t

=


∂t



∂B
y

∂z

!


=
μ0ǫ0

2Ex

∂t
2

Similarly, we can show,We get

2By

∂z
2 = μ0ǫ0

2By

∂t
2

2

Each of the above equations represents a wave disturbance propagating in the zdirection

with a speed

c
=

1

μ0ǫ0

On substituting numerical values, the speed of electromagnetic waves in vacuum

is
3 × 108 m/sec.

Consider plane harmonic waves of angular frequency
ω and wavlength λ = 2π/k.

We can express the waves as

E
x = E0 sin(kz ωt)

B
y = B0 sin(kz ωt)

The amplitudes
E0 an B0 are not independent as they must satisfy eqns. (5) and

(6) :

∂E
x

∂z

=
E0k cos(kz ωt)

∂B
y

∂t

=
B0ω cos(kz ωt)

Using Eqn. (5) we get

E
0k = B0ω

The ratio of the electric field amplitude to the magnetic field amplitude is given

by

E
0

B
0

=

ω

k

=
c

Fields
~E and ~B are in phase, reaching their maximum and minimum values at the

same time. The electric field oscillates in the x-z plane and the magnetic field oscillates

in the y-z plane. This corresponds to a
polarized wave. Conventionally,

the plane in which the electric field oscillates is defined as the plane of polarization.

In this case it is x-z plane. The figure shows a harmonic plane wave

propagating in the z-direction. Note that
~E , ~B and the direction of propagation ˆk

form a right handed triad.

3

z



B

E


y

x


ct


Example :


The electric field of a plane electromagnetic wave in vacuum is
Ey = 0.5 cos[2π×

10
8(t x/c)] V/m, Ex = Ez = 0. Determine the state of polarization and the

direction of propagation of the wave. Determine the magnetic field.

Solution :


Comparing with the standard form for a harmonic wave

ω
= 2π × 108 rad/s

k
= 2π × 108/c

so that
λ = c/108 = 3 m. the direction of propagation is x-direction. Since

the electric field oscillates in x-y plane, this is the plane of polarization. Since

~B

must be perpendicular to both the electric field direction and the direction of

propagation,
~B has only z-component with an amplitude B0 = E0/c 1.66 ×

10
9 T. Thus

B
z = 1.66 × 109 cos[2π × 108(t x/c)] T

Exercise :


The magnetic field of a plane electromagnetic wave is given by

B
y = Bz = 108 sin[

2
π

3

x
2π × 108t] T

Determine the electric field and the plane of polarization. (Ans. Strength of

electric field is
32 V/m)

Plane, Circular and Elliptic Polarization :


4

We have shown that
~E and ~B lies in a plane perpendicular to the direction of propagation,

viz., the plane of polarization. This does not imply that the direction of

these fields are constant in time. If it so happens that the successive directions of

~E

(and hence
~B) remains parallel, the electric vectors at different points in space

along the direction of propagation at a given time will lie in a plane. (Equivalently,

the directions of electric vectors at a given point in space at different times will be

parallel). Such a situation is called a
plane polarized wave.

Z

X

y

X

y

Z


E− parallel to x−direction

E parallel to y−direction

X

y

Z


q


E inclined at an angle

q to x−axis

A plane polarized wave propagating in z-direction can be described by an electric

field given by

~E

=
0 sin ωtˆn = E0 sin ωt cos θˆi + E0 sin ωt sin θˆj

We know that because of linearity of the wave equation, any linear combination

of two solutions of the wave equation is also a solution of the wave equation.

This allows us to construc new states of polarization of the electromagnetic wave

5

from the plane polarized wave. Suppose we have two plane polarized waves of

equal amplitude one with the electric vector parallel to the x-direction and the

other parallel to y-direction, the two waves having a phase difference of
π/2. The

resultant, which is also a solution of the wave equation, has

E
x = E0 sin ωt

E
y = E0 cos ωt

The tip of the resultant electric vector
~E describes a circle of radius E0, which

is independent of
t. The state of polarization is known as circular polarization.

Looking along the direction of propagation if the radius vector is moving clockwise,

the polarization is called
right circularly polarized and if anticlokwise it is

called left circularly polarized.

x

y

x

y


w

t=0

w

t=p /2

w

t

w

t

w

t=0

w

t=p /2

Right Circularly polarized Left Circularly polarized


In the same way if the two plane polarized solutions have a phase difference of
π/2

but have different amplitudes
a and b, they produce what is known as elliptically

polarized light. In this case, we have

E
x = a sin ωt

E
y = b cos ωt

so that the equation to the trajectory is given by

E
2

x


a
2 +

E
2

y


b
2 = 1

6

x

y

x

y


w

t=0

w

t=p /2

w

t

w

t

w

t=0

w

t=p /2

Right Elliptically polarized Left Elliptically polarized

b

a

a

b


Wave Equation in Three Dimensions :


We can obtain the wave equation in three dimensions by using eqns. (1) to (4).

On taking the curl of both sides of eqn. (3), we get

~

×
(~∇ × ~E ) =


∂t

(
~∇ × ~B)

Using the operator identity

~

×
(~∇ × ~E ) = ~(~∇ ・ ~E ) ~2~E = ~2~E

wherein we have used
~∇ ・ ~E = 0, and substituting eqn. (4) we get

~

2

~E = μ0ǫ0

2~E

∂t
2

A three dimensional harmonic wave has the form
sin(~k ~r ωt) or cos(~k ~r ωt) Instead of using the trigonometric form, it is convenient to use the complex

exponential form

f
(~r, t) = exp(i~k ~r ωt)

and later take the real or imaginary part of the function as the case may be. The

derivative of
f(~r, t) is given as follows :


∂x

f
(~r, t) =


∂x

exp(
ikxx + ikyy + ikzz iωt) = ikxf(~r, t)

7

Since

= ˆı


∂x

+ ˆ


∂y

+ ˆ
k


∂z

we have,

f(~r, t) = i~kf(~r, t)

In a similar way,


∂t

f
(~r, t) = iωf(~r, t)

Thus for our purpose, the differential operators
and ∂/∂t may be equivalently

replaced by


∂t
→ −

~

i~k

Using these, the Maxwell’s equations in free space become

~k
~E = 0 (7)

~k
~B = 0 (8)

~k
× ~B = μ0ǫ0~E (9)

~k
× ~E = B (10)

We can see that
~E , ~B and ~k form a mutually orthogonal triad. The electric field

and the magnetic field are perpendicular to each other and they are both perpendicular

to the direction of propagation.

Generation of Electromagnetic Waves :


We have looked for solutions to Maxwell’s equations in free space which does

not have any charge or current source. In the presence of sources, the solutions

become complicated. If
ρ = constant, i.e. if ~ J = 0, we only have a steady electric

field. If
ρ varies uniformly with time, we have steady currents which gives

us both a steady electric field as well as a magnetic field. Clearly, time varying

electric and magnetic fields may be generated if the current varies with time, i.e.,

if the charges accelerate. Hertz confirmed the existence of electromagnetic waves

in 1888 using these principles. A schematic diagram of Hertz’s set up is shown in

the figure.

8

Source

Primary

Plate

Plate

Secondary

Gap


The radiation will be appreciable only if the amplitude of oscillation of charge is

comparable to the wavelength of radiation that it emits. This rules out mechanical

vibration, for assuming a vibrational frequency of 1000 cycles per second,

the wavelength work out to be 300 km. Hertz, therefore, made the oscillating

charges vibrate with a very high frequency. The apparatus consists of two brass

plates connected to the terminals of a secondary of a transformer. The primary

consists of an LC oscillator circuit, which establishes charge oscillations at a frequency

of
ω = 1/LC. As the primary circuit oscillates, oscillations are set up

in the secondary circuit. As a result, rapidly varying alternating potential difference

is developed across the gap and electromagnetic waves are generated. Hertz

was able to produce waves having wavelength of 6m. It was soon realized that

irrespective of their wavelength, all electromagnetic waves travel through empty

space with the same speed, viz., the speed of light.

9

Depending on their wavelength range, electromagnetic waves are given different

names. The figure shows the electromagnetic spectrum. What is known as visible

light is the narrow band of wavelength from 400 nm (blue) to 700 nm (red). To

its either side are the infrared from 700 nm to 0.3 mm and the ultraviolet from 30

nm to 400 nm. Microwaves have longer wavelength than the infrared (0.3 mm to

300 mm) and the radio waves have wavelengths longer than 300 mm. The television

broadcast takes place in a small range at the end of the microwave spectrum.

Those with wavelengths shorter than ultraviolet are generally called
rays. Prominent

among them are x-rays with wavelengths 0.03 nm t0 30 nm and
γ-rays with

wavelengths shorter than 0.03 nm.

Poynting Vector :


Electromagnetic waves, like any other wave, can transport energy. The power

through a unit area in a direction normal to the area is given by
Poynting vector,

given by

~S

=

1

μ
0

~E

×
~B

As
~E , ~B and ~k form a right handed triad, the direction of ~S is along the direction

of propagation. In SI units
~S ismeasured in watt/m2.

The magnitude of
~Sfor the electromagnetic wave travelling in vacuum is given

10

by

S
=

EB

μ
0

=

E
2

0

where we have used the relationship between
E and B in free space. For harmonic

waves, we have

S
=

E
2

0


0

sin
2(kx ωt)

The average power transmitted per unit area, defined as the
intensity is given by

substituting the value 1/2 for the average of the square of sine or cosine function

I
=

E
2

0


2
0

Example :


Earth receives 1300 watts per squar meter of solar energy. assuming the energy

to be in the form of plane electromagnetic waves, compute the magnitude of the

electric and magnetic vectors in the sunlight.

Solution :


From the expression for the average Poynting vector

E
2

0


2
0

= 1300

which gives
E0 = 989 V/m. The corresponding rms value is obtained by dividing

by
2, Erms = 700 V/m. The magnetic field strength is Brms = Erms/c =

2
.33 × 106 T.

Exercise :


A 40 watt lamp radiates all its energy isotropically. Compute the electric field at

a distance of 2m from the lamp. (Ans. 30 V peak)

Radiation Pressure :


We have seen that electric field, as well as magnetic field, store energy. The

energy density for the electric field was seen to be
(1/2)ǫ0E2 and that for the

magnetic field was found to be
(1/2)B2/2μ0. For the electromagnetic waves,

where
E/B = c, the total energy density is

u
=

1

2

ǫ
0E2 +

B
2

2
μ0

=
ǫ0E2

where we have used
c2 = 10ǫ0.

In addition to carrying energy, electromagnetic waves carry momentum as well.

11

The relationship between energy (
U) and momentum (p) is given by relatistic

relation for a massless photons as
p = U/c. Since the energy density of the

electromagnetic waves is given by
ǫ0E2, the momentum density, i.e. momentum

per unit volume is

|
p |=

ǫ
0E2

c

=
ǫ0 | ~E × ~B |

Since the direction of momentum must be along the direction of propagation of

the wave, the above can be converted to a vector equation

~p
= ǫ0~E × ~B

If an electromagnetic wave strikes a surface, it will thus exert a pressure. Consider

the case of a beam falling normally on a surface of area
A which absorbs the wave.

The force exerted on the surface is equal to the rate of change of momentum of the

wave. The momentum change per unit time is given by the momentum contained

within a volume
cA. The pressure, obtained by dividing the force by A is thus

given by

P
= cp = 0EB = ǫ0E2

which is exactly equal to the energy density
u.

If on the other hand, the surface reflects the wave, the pressure would be twice the

above value.

The above is true for waves at normal incidence. If the radiation is diffuse, i.e., if it

strikes the wall from all directions, it essentially consists of plane waves travelling

in all directions. If the radiation is isotropic, the intensity of the wave is the

same in all directions. The contribution to the pressure comes from those waves

which are travelling in a direction which has a component along the normal to the

surface. Thus on an average a third of the radiation is responsible for pressure.

The pressure for an absorbing surface is
u/3 while that for a reflecting surface is

2
u/3.

The existence of radiaton pressure can be verified experimentally. The curvature

of a comet’s tail is attributed to the radiation pressure exerted on the comet by

solar radiation.

Exercise :


Assuming that the earth absorbs all the radiation that it receives from the sun,

calculate he radiation pressure exerted on the earth by solar radiation. (Ans.

Assuming diffuse radiation
1.33 × 106 N/m2)

Wave Propagation in Matter :


12

Inside matter but in region where there are no sources, the relevant equations are

∇ ・
~D = 0

∇ ・
~B = 0

∇ ×
~E =

∂B

∂t

∇ ×
~H =

∂D

∂t

which results in a travelling wave with speed
v = 1/ǫμ where ~B = μ~H and

~D

=
ǫ~H .

Waves in a conducting Medium :


If the medium is conducting, we need to include the effect due to conduction

current. The two curl equations become

∇ ×
~E =

∂~B

∂t

=
μ

∂~H

∂t

∇ ×
~H =

∂~D

∂t

+
~ J = ǫ

∂~E

∂t

+
σ~E

where we have used Ohm’s law as another constitutive relation. Thus, we have,

∇ ×
(∇ × ~E) = μ


∂t

(
∇ × ~H )

Using the expansion for
∇ × (∇ × ~E) = (∇ ・ ~E ) − ∇2E, we get

1

ǫ
(∇ ・ ~D) − ∇2~E = μǫ∂2~E

∂t
2 μσ

∂E

∂t

It may be noted that though
∇ ・ ~D term equals ρf , free charges, if they exist in a

conductor soon depletes. This may be seen from the equation of continuity. Using

∂ρ

∂t

=
−∇ ・ ~ J

and using Ohm’s law one can see that if at time
t = 0 there exists some free charge

ρ
0

f

, the charge drops exponentially to 1/e th of its value after a time ǫ/σ which is

very small for conductors. Thus the wave equation that we have is

2~E = μǫ∂2~E

∂t
2 + μσ

∂E

∂t

13

and a similar equation for
~H , i.e.

2~H = μǫ∂2~H

∂t
2 + μσ

∂t

∂t

In practice, most generators produce voltages and currents which vary sinusoidally.

Corresponding electric and magnetic fields also vary simusoidally with time. This

implies
∂E/∂t −→ −iωE and 2E/∂t2 −→ −ωE. The wave equation becomes

2~E = μǫω2~E + iμσωE γ2E

where

γ
2 = μǫω2 + iμσω

is a complex constant. We may write
γ = α + , squaring and equating it to the

expression for
γ2, we can see that α and β have the same sign. When we take the

square root of
γ2 and write it as γ we assume that α and β are positive. One can

explicitly show that

α
= ω

vuuut


μǫ
2



s


1 +

σ
2

ω
2ǫ2 1




β
= ω

vuuut


μǫ
2



s


1 +

σ
2

ω
2ǫ2 + 1




We have seen that in Maxwell’s equation,
σ~E is conduction current density while

the term
iωǫE is the displacement current density. This suggest that the value of

the ratio
σ/ωǫis a good measure to divide whether a material is a dielectric or a

metal. For good conductors this ratio is much larger than 1 over the entire radio

frequency spectrum. For instance, at a frequency as high as 30, 000 MHz, Cu has

σ/ωǫ
108 while for the same frequency this ratio is 0.0002 for mica.

For good conductors, we take
σ/ωǫ1. We have, in this limit

α
= β =

r


ωσμ

2

The velocity of the wave in the conductor is

v
=

ω

β

=

s


2
ω

μσ

14

The wave is greatly attenuated inside a conductor.
Depth of penetration or skin

depth

δ is defined as the distance at which an eletromagnetic wave would have

attenuated to 1/e of its value on the surface.

δ
=

1

α

=

s


2

ωμσ

For conductors like copper this distance is typically less than a fraction of a millimeter.

15

No comments:

Post a Comment