PHY-403 :- Electrodynamics – 1

unit-1 - wave propagation

Theory-1.1 :- Propagation through linear media:-

The electromagnetic waves are traveling through empty space, but it is easy to generalize these
results to any linear medium. By change ∈0 to ∈ , and 𝜇0 to 𝜇 .

The speed of propagation becomes,

1 𝑐
𝑣= = 1
√∈𝜇 𝑛

∈𝜇
Here, n = √∈ 𝜇 = ‘Index of refraction’ of material.
0 0

The energy density is,

1 1
U= (∈ 𝐸 2 + 𝐵2) 2
2 𝜇

And the pointing vector is,

1
S= (𝐸 × 𝐵) 3
𝜇

In particular the intensity of the wave is,

1
I= ∈ 𝑣 𝐸2 4
2

As in the case of a wave on a string (waves passes from one such medium to another- air to
water or glass to plastic), we can get a reflected wave and a transmitted wave. The details
depend on the exact nature of of the electrodynamic boundary conditions, so

(i) ∈1 E1 ┴ = ∈2 E2 ┴ , (iii) E1 ║ = E2 ║ ,
1 1
(ii) B1 ┴ = B2 ┴ , (iv)
𝜇1
B1║ =
𝜇2
B2║ , 5

These equations relate the electric and magnetic fields just to the left and just to the right of
the interface between two linear media.

(In the following sections we use them to deduce the laws governing reflection and
refraction of electromagnetic waves.)
Theory-1.2 :- Reflection and Transmission at normal incidence:-

The yz-plane forms the boundary between two linear media. A plane wave of frequency
ω, traveling in the x-direction and polarized in the y-direction, approaches interface from the –x
direction as we can see in the figure below.

F-1: wave propagation at normal incidence.

So the incident wave is,

^ 1 ^
𝐸̃𝐼 (𝑥, t) = 𝐸̃0𝐼 𝑒 i(𝑘𝐼𝑥− ωt)𝑗 , 𝐵̃𝐼 (𝑥, t) = 𝑣 𝐵̃0𝐼 𝑒 i(𝑘𝐼𝑥− ωt)𝑘 1
1

And it gives reflected wave, travels back to left in medium-1,

^ 1 ^
𝐸̃𝑅 (𝑥, t) = 𝐸̃0𝑅 𝑒 i(−𝑘𝐼𝑥− ωt)𝑗 , 𝐵̃𝑅 (𝑥, t) = - 𝐵̃0𝑅 𝑒 i(−𝑘𝐼𝑥− ωt)𝑘 2
𝑣1

Similarly, transmitted wave, travels further right into medium-2

^ 1 ^
𝐸̃𝑇 (𝑥, t) = 𝐸̃0𝑇 𝑒 i(𝑘2𝑥− ωt)𝑗 , 𝐵̃𝑇 (𝑥, t) = 𝑣 𝐵̃0𝑇 𝑒 i(𝑘2𝑥− ωt)𝑘 3
2

Note:- The minus sign in 𝐵̃𝑅 equation, due to fact that the pointing vector must aim in the
direction of propagation.

At x = 0, the combined fields to the left, (𝐸̃𝐼 +𝐸̃𝑅 ) and (𝐵̃𝐼 +𝐵̃𝑅 ), must join the fields
corresponding to the right 𝐸̃𝑇 and 𝐵̃𝑇 , in accordance with the electrodynamic boundary
conditions (of linear media).
 In this case, there are no components perpendicular to the (interface) surface, so
boundary condition (i) and (ii) are trivial. However, (iii) requires,

𝐸̃0𝐼 +𝐸̃0𝑅 = 𝐸̃0𝑇 4

And (iv) gives,

1 1 1 1 1
(𝑣 𝐸̃0𝐼 − 𝑣 𝐸̃0𝑅 ) = (𝑣 𝐸̃0𝑇 )
𝜇1 1 1 𝜇2 2

Or

𝐸̃0𝐼 -𝐸̃0𝑅 = β 𝐸̃0𝑇 5

𝜇1 𝑣1
Where, β = 6
𝜇2 𝑣2

Equations (4) and (5) are easily solved for outgoing amplitudes, so, in terms of the
amplitudes we can right,
1− β 2
𝐸̃0𝑅 =( 1+ β ) 𝐸̃0𝐼 , 𝐸̃0𝑇 = (1+ β) 𝐸̃0𝐼 7

These results are strikingly similar to those describing waves on a string. Indeed, if the
permittivities μ are close to their values in vacuum ( we know that, they are for most media),
then β = (𝑣1 / 𝑣2 ), and we have,
𝑣 −𝑣 2𝑣
𝐸̃0𝑅 =( 𝑣2 + 𝑣1 ) 𝐸̃0𝐼 , 𝐸̃0𝑇 = (𝑣 +2𝑣 ) 𝐸̃0𝐼 8
2 1 2 1

As before, the reflected wave is in phase (right side up) if 𝑣2 > 𝑣1 and out of phase
(upside down) if 𝑣2 < 𝑣1 , so the real amplitudes are related by,

𝑣 −𝑣 2𝑣
𝐸̃0𝑅 =| 𝑣2 + 𝑣1 | 𝐸̃0𝐼 , 𝐸̃0𝑇 = (𝑣 +2𝑣 ) 𝐸̃0𝐼 9
2 1 2 1

or, we can write in terms of the index of refraction,

𝑛 −𝑛 2𝑛
𝐸̃0𝑅 =| 𝑛1 + 𝑛2 | 𝐸̃0𝐼 , 𝐸̃0𝑇 = (𝑛 +1𝑛 ) 𝐸̃0𝐼 10
1 2 1 2

Some fraction of incident energy is being reflected and some fraction being transmitted
(no absorption due to ideal media). So the intensity(average power per unit area) is denoted
by,
1
I = ∈ 𝑣𝐸02
2
If then the ratio of reflected intensity to the incident intensity is,
𝐼 𝐸 𝑛 −𝑛
R (Reflection Coefficient) = 𝐼𝑅 = ( 𝐸0𝑅 )2 = ( 𝑛1 + 𝑛2 ) 2 11
𝐼 0𝐼 1 2

Whereas the ratio of transmitted intensity to the incident intensity is,

∈2 𝑣2 𝐸0𝑇 2 𝑛2 2𝑛1 2
𝐼
T (Refraction Coefficient) = 𝐼𝑇 = ( ) = (𝑛 ) 12
𝐼 ∈1 𝑣1 𝐸0𝐼 𝑛1 1 + 𝑛2

This is the fraction of incident intensity in form of reflection and transmitted, notice that

R+ T = 1 13

As we can see the conservation of energy is fulfilled correct. For example, when light
passes from air (n1= 1) into glass (n2=1.5) calculate R and T co-efficient,

R= 0.04 , T= 0.96 here, R+T = 0.04 + 0.96 = 1

The value of T say, most of intensity is transmitted.

Theory-1.3:- Reflection and transmission at oblique incidence:-

The oblique incidence in which the incoming wave meets the boundary at glancing angle
θI . Of course, normal incidence is really just a special case of oblique incidence, with θ I = 0.

F-2: wave propagation at oblique incidence.

Suppose, a monochromatic plane wave is

 ^ 1 ^
𝐸̃𝐼 (𝑟, t) = 𝐸̃0𝐼 𝑒 i(𝑘𝐼𝑟− ωt)𝑗 , 𝐵̃𝐼 (𝑟, t) = 𝑣 𝐵̃0𝐼 𝑒 i(𝑘𝐼𝑟− ωt)𝑘 1
1

From the left of interface, the reflected wave is

^ 1 ^
𝐸̃𝑅 (𝑟, t) = 𝐸̃0𝑅 𝑒 i(−𝑘𝐼𝑟− ωt)𝑗 , 𝐵̃𝑅 (𝑟, t) = - 𝐵̃0𝑅 𝑒 i(−𝑘𝐼𝑟− ωt)𝑘 2
𝑣1

Similarly, transmitted wave

^ 1 ^
𝐸̃𝑇 (𝑟, t) = 𝐸̃0𝑇 𝑒 i(𝑘2 𝑟− ωt)𝑗 , 𝐵̃𝑇 (𝑟, t) = 𝑣 𝐵̃0𝑇 𝑒 i(𝑘2𝑟− ωt)𝑘 3
2

All three waves have same frequency ω because incident wave incoming from any one source
(flash light, torch etc). So, the wave numbers are related,
𝑣𝑇 𝑛
kI vI = kR vI = kT vT = ω, or kI = kR = kT = 𝑛1 kT, 4
𝑣𝐼 2

The combined fields in medium-(1), (𝐸̃𝐼 +𝐸̃𝑅 ) and (𝐵̃𝐼 +𝐵̃𝑅 ), must join the fields corresponding to
the right 𝐸̃𝑇 and 𝐵̃𝑇 in medium(2). So all the boundary condition have general form,

( ) 𝑒 i(𝑘𝐼𝑟− ωt) = ( ) 𝑒 i(𝑘𝑅𝑟− ωt) = ( ) 𝑒 i(𝑘𝑇 𝑟− ωt) at 𝑥=0 5

It noticeable to that, the y-, z-, t-dependence is confined to the exponents. Because the
boundary conditions must hold at all points on the plane, and for all time, these exponents
factors must be equal. Otherwise, a slight change in y, would destroy the equality. As for the
spatial terms,

kI ∙ r= kR ∙ r= kT ∙ r at 𝑥=0 6

[That means,

(𝑘𝐼 )𝑦 y + (𝑘𝐼 )𝑧 z = (𝑘𝑅 )𝑦 y + (𝑘𝑅 )𝑧 z = (𝑘𝑇 )𝑦 y + (𝑘𝑇 )𝑧 z for all y and all z (a)

Equation holds only, if components are separately equal, for y = 0, we get

(𝑘𝐼 )𝑧 z = (𝑘𝑅 )𝑧 z = (𝑘𝑇 )𝑧 z (b)

And for z = 0,

(𝑘𝐼 )𝑦 y = (𝑘𝑅 )𝑦 y = (𝑘𝑇 )𝑦 y (c) ]

Here kI lies on xy-plane ((𝑘𝐼 )𝑧 = 0), so according to equation-(b), so too will kR and kT.

Conclusion,
 (A) The incident, reflected and transmitted wave vector form a plane called the plane of
incidence, which also include normal to the surface. (here, x-axis)
𝑘𝐼 sin θI = 𝑘𝑅 sin θR = 𝑘𝑇 sin θT 7
(B) The angle of incidence is equal to the angle of reflection, θI = θR 8
(C) “Snell’s law”.
sin θ𝑇 𝑛
= 𝑛1 9
sin θ𝐼 2
(A),(B) and (C) are fundamental laws of geometrical optics. We have yet to invoke any
specific boundary condition, so it is wel taken care of exponential factors of equation-5.
The boundary conditions we have,
(i) ∈1 (𝐸̃0𝐼 +𝐸̃0𝑅 )x = ∈2 (𝐸̃0𝑇 )x
(ii) (𝐵̃0𝐼 +𝐵̃0𝑅 )x = (𝐵̃0𝑇 )x
(iii) (𝐸̃0𝐼 +𝐸̃0𝑅 )y, z = ∈2 (𝐸̃0𝑇 )y, z
1 1
(iv) (𝐵̃0𝐼 +𝐵̃0𝑅 )y, z = (𝐵̃0𝑇 )y, z (X)
𝜇1 𝜇2
Where, 𝐵̃0 = (k x 𝐸̃0 )/ 𝑣 (the last two represent pairs of equations, one for y-component and
one for z-component.)
Suppose, the polarization of the incident wave is parallel to the plane of incidence.then
equation (X-(i)) reads,
∈1 (−𝐸̃0𝐼 sin 𝜃𝐼 +𝐸̃0𝑅 sin 𝜃𝑅 ) = ∈2 (−𝐸̃0𝑇 sin 𝜃𝑇 ) 10

(X-(ii)) gives nothing, but (X-(iii)) and (X-(iv)) becomes,

𝐸̃0𝐼 cos 𝜃𝐼 +𝐸̃0𝑅 cos 𝜃𝑅 = 𝐸̃0𝑇 cos 𝜃𝑇 11

1 1
(𝐸̃0𝐼 − 𝐸̃0𝑅 )= 𝐸̃0𝑇 12
𝜇1 𝑣1 𝜇2 𝑣2

Given the laws of reflection and refraction (eq. (7), (8), (10) & (12)) both reduce to,

𝐸̃0𝐼 − 𝐸̃0𝑅 = β 𝐸̃0𝑇 13

𝜇 𝑣 𝜇 𝑛
Where, β= 𝜇1𝑣1 = 𝜇1𝑛2 14
2 2 2 1

Similarly eq. (11) becomes,

𝐸̃0𝐼 + 𝐸̃0𝑅 = α 𝐸̃0𝑇 15

cos 𝜃
Where, α = cos 𝜃𝑇 16
𝐼

Solving eq. (13) & (15) for the reflection and refraction amplitudes, we obtain,
𝛼−𝛽 2
𝐸̃0𝑅 = (𝛼+𝛽) 𝐸̃0𝐼 , 𝐸̃0𝑇 = (𝛼+𝛽) 𝐸̃0𝐼 17
These are known as Fresnel’s equations, for the case of polarization in the plane of incidence.
Notice that the transmitted wave is always in plane with the incident one: the reflected wave is
either in phase (“right side up”), if α>β, or 180⁰ out of phase (“upside down”), if α<β.

The amplitudes of the transmitted and reflected waves depend on the angle of incidence,
because α is a function of𝜃𝐼 .

2
√1− (𝑛1 sin 𝜃𝐼 )
√1−sin 𝜃𝑇 𝑛 2
α= = 18
cos 𝜃𝐼 cos 𝜃𝐼

In the case of normal incidence (θI = 0⁰), α = 1, and we recover amplitudes of normal incidence.
At grazing incidence (θI = 90⁰), α diverges, and the wave is totally reflected (a fact that is
painfully familiar to anyone who has driven at night on a wet road). Interestingly, there is an
intermediate angle, θB (called Brewster’s angle) at which the reflected wave is completely
extinguished. According to Fresnel’s equation, this occurs when α = β, from eq.-18 we get,

1−𝛽 2
Sin2 θI = 19
(𝑛1 ⁄𝑛2 )2 −𝛽 2

For the typical case μ1 ≈ μ2 , so that β = 𝑛2 ⁄𝑛1 , sin2 θB = 𝛽 2 /(1+𝛽 2 ), and hence

𝑛2
Tan θB = ( ) 20
𝑛1

In the figure we can see a plot of the transmitted and reflected amplitudes as functions of θ I,
for light incidence on glass (n2 = 1.5) from air (n1 = 1). ( on the graph, a negative number
indicates that the wave is 180⁰ out of phase with the incident beam- the amplitude itself is the
absolute value).

The power per unit area striking the interface is given by S ∙ 𝑛̂, where ∙ 𝑛̂ is the normal unit
vector, in the case ∙ 𝑥̂ .thus the incident intensity is,
1 2
II = 𝜖1 𝑣1 𝐸0𝐼 cos 𝜃𝐼 21
2

While the reflected and transmitted intensities are,

1 2 1 2
IR = 2 𝜖1 𝑣1 𝐸0𝑅 cos 𝜃𝑅 , IT = 2 𝜖2 𝑣2 𝐸0𝑇 cos 𝜃𝑇 22

The reflection and the transmission coefficients for waves polarized parallel to the plane of
incidence are,
 𝐼𝑅 𝐸 2 𝛼−𝛽 2
R= = ( 𝐸0𝑅 ) = (𝛼 + 𝛽) 23
𝐼𝐼 0𝐼

𝐼𝑇 𝜖2 𝑣2 𝐸 2 cos 𝜃 2 2
T= = ( 𝐸0𝑇 ) 𝑇
= αβ (𝛼 + 𝛽) 24
𝐼𝐼 𝜖1 𝑣1 0𝐼 cos 𝜃𝐼

R is the fraction of the incidence energy that is reflected – naturally it goes to zero
at Brewster’s angle; T is the fraction transmitted – it goes to 1 at θB. Note that R +T = 1.

Theory-1.4:- Total internal reflection.

Consider radiation is incident from a medium of higher refractive index on to the surface of a
medium of lower refractive index, i.e. n1 > n2, from Snell’s law,
𝑛1
Sin θT = Sin θI 1
𝑛2

What happens if θI is gradually increased starting from zero? Obviously θT will also increase
until it attains the value π/2. Let us represent the value of θI, when, this happens, by the symbol
θC, from eq.-1,
𝑛2
Sin θC = 2
𝑛1

The angle θC is called the critical angle, since, when θI = θC there is only reflected wave, there
being no transmitted wave. What will happen if θC increased further, i.e. when θI > θC ?

Let us express θT in terms of θI and θC.

𝑛1 sin θ𝐼
Cos θT = √1 − 𝑆𝑖𝑛2 𝜃𝑇 but, Sin θT = Sin θI =
𝑛2 sin θ𝐶

Therefore,

𝑆𝑖𝑛2 𝜃
Cos θT = √1 − 𝑆𝑖𝑛2 𝜃 𝐼 3
𝐶

We see that the value of cos θT decreases as θI is increased: it becomes 0 at θI = θC and for value
θI of greater than θC, cos θT becomes an imaginary number.

Let us now calculate the amplitude of the reflected electric vector when θI > θC. we can write,

𝑆𝑖𝑛2 𝜃
Cos θT = √1 − 𝑆𝑖𝑛2 𝜃 𝐼 = iQ
𝐶
 𝑆𝑖𝑛2 𝜃
i.e. Q = √𝑆𝑖𝑛2 𝜃 𝐼 − 1 4
𝐶

in the case when E is polarized perpendicular to the plane of incidence ,

𝑛2 𝑛2
𝐸0𝑅 cos 𝜃𝐼 − cos 𝜃𝑇 cos 𝜃𝐼 − 𝑖𝑄
𝑛1 𝑛1
= 𝑛2 = 𝑛2
𝐸0𝐼 cos 𝜃𝐼 + cos 𝜃𝑇 cos 𝜃𝐼 + 𝑖𝑄
𝑛1 𝑛1

Therefore,

𝐸 2
| 𝐸0𝑅 | = 1, i.e. |𝐸0𝑅 | = |𝐸0𝐼 |
0𝐼

Similarly when E is polarized parallel to the plane of incidence,

𝐸 2
| 𝐸0𝑅 | = 1, i.e. |𝐸0𝑅 | = |𝐸0𝐼 |
0𝐼

Thus the wave is totally reflected. This phenomenon is known as total internal reflection.

It must be noted that there is a change of phase on reflection; hence, if the incidence wave is
polarized in a plane intermediate between the plane of incidence and the plane normal to it,
the two components, will not be in phase after reflection and the wave will be elliptically
polarized.

We conclude: when the angle of incidence is greater than critical angle there is no refracted
wave and all the energy is reflected. That this is so can be shown by computing the average rate
of energy flow across the boundary.

1 1 𝑘𝑇 ×𝐸𝑇∗
The rate of energy flow = <N> ∙ ê𝑛 = 2 Re(𝐸𝑇 × 𝐻𝑇∗ ) ∙ ê𝑛 = 2 Re(𝐸𝑇 × ) ∙ ê𝑛
𝜔𝜇

1
= Re [(𝐸𝑇 × 𝐸𝑇∗ )𝑘𝑇 − (𝐸𝑇 × 𝑘𝑇 )𝐸𝑇∗ ] ∙ ê𝑛
2𝜔𝜇

1
= Re [(𝐸𝑇 × 𝐸𝑇∗ )𝑘𝑇 ] ∙ ê𝑛 ( ∵ 𝐸𝑇 ⊥ 𝑘𝑇 )
2𝜔𝜇

1 1
= Re |𝐸0𝑇 |2 𝑘𝑇 cos 𝜃𝑇 = Re iQ 𝑘𝑇 |𝐸0𝑇 |2 5
2𝜔𝜇 2𝜔𝜇

Since this expression is purely imaginary, <N> ∙ ê𝑛 = 0 which justifies our conclusion.

We must not hasten to conclude at this stage if n1 > n2 and the wave is incident an angle
greater than the critical angle, there will be no field on the other side of the boundary. Even
though there is no energy flow across the surface, the field does exist on the other side of the
surface as is shown below.
Consider the equation for the refracted wave,

ET = E0T exp {𝑖(𝑘𝑇 ∙ 𝑟 − 𝜔𝑡)} 6

For the co-ordinate axes from the normal incidence case, this expression becomes,

ET = E0T exp {𝑖(𝑘𝑇 ∙ 𝑥 cos 𝜃𝑇 + 𝑘𝑇 ∙ 𝑦 sin 𝜃𝑇 − 𝜔𝑡)}

= E0T exp {(−𝑘𝑇 ∙ 𝑥𝑄 exp{ 𝑖(𝑘𝑇 ∙ 𝑦 sin 𝜃𝑇 − 𝜔𝑡}) } 7

This shows that the field does exist on the far side of the surface in the medium 2; but is rapidly
attenuated as below figure.

F-3: Skin depth

How far does the wave penetrate the medium? – the penetrating distance or skin depth is
given by,
−1/2
1 1 𝑠𝑖𝑛2 𝜃
X=𝑘 = (𝑠𝑖𝑛2 𝜃 𝐼 − 1) 8
𝑇 𝑄 𝑘 𝑇 𝐶

Consider for example a wave going from glass to air. The refractive index of glass is 1.5; hence
the critical angle for glass is 𝜃𝐶 = sin-1 2/3, i.e. ∼42⁰.

If the light is incident internally from the glass at an angle greater than 42 say 45, there will be
total internal reflection, and

1 1 1 9 −1/2 𝜆
X=𝑘 = (2 × 4 − 1) = 2𝜋 2√2 =0.45𝜆
𝑇 𝑄 𝑘 𝑇

Thus the field becomes negligible beyond distances of the order of a few wavelengths.

Theory-1.5:- Reflection from the surface of a metal.

We shall now show how the arguments of the preceding section can be extended to a boundary
surface of a conducting medium. Since the case of oblique incidence is more involved, we shall
confine to the simple case of normal incidence only.

We have incidence, reflected and transmitted waves,

 𝑘𝐼 ×𝐸𝐼
EI = E0I exp{𝑖(𝑘𝐼 𝑟 − 𝜔𝑡)}; HI = 1.1
𝜔𝜇1

𝑘𝐼 ×𝐸𝑅
ER = E0R exp{𝑖(𝑘𝐼 𝑟 − 𝜔𝑡)}; HR = - 1.2
𝜔𝜇1

𝑘𝐼 ×𝐸𝐼
ET = E0T exp{𝑖(𝑘𝑇 𝑟 − 𝜔𝑡)}; HT = 1.3
𝜔𝜇2

Since the medium-2 is a conducting medium, the propagation vector kT is given by,
𝑖𝜎
𝑘𝑇2 = 𝜖2 𝜇2 𝜔2 [1 + ] (Remember, 𝑘𝐼2 = 𝜖1 𝜇1 𝜔2) 2
𝜖2 𝜔

The boundary condition require that,

E0I – E0R = E0T 3.1

kI (E0I + E0R) = kT E0T 3.2

Since kT is complex, E0R and E0T cannot both be real and, hence, one should expect phase shift
other than 0 or π in the reflected and transmitted waves.

Solving eq.-3.1 and 3.2, we have,

𝑘𝑇 −𝑘𝐼
E0R = E0I 4
𝑘𝑇 +𝑘𝐼

2𝑘𝐼
E0T = 𝑘 E0I 5
𝑇 +𝑘𝐼

Substituting the values of kI and kT, we have,

𝑖𝜎 1/2
√𝜖2 𝜇2 𝜔2 (1+ ) − ω (𝜖1 𝜇1 )1/2
𝜖2 𝜔
E0R =
𝑖𝜎 1/2
E0I 6.1
√𝜖2 𝜇2 𝜔2 (1+ ) + ω (𝜖1 𝜇1 )1/2
𝜖2 𝜔

2ω (𝜖1 𝜇1 )1/2
E0T =
𝑖𝜎 1/2
E0I 6.1
√𝜖2 𝜇2 𝜔2 (1+ ) + ω (𝜖1 𝜇1 )1/2
𝜖2 𝜔

Let us now examine the case of a perfect conductor for which σ=∞. Then

E0R = E0I and E0T = 0

Hence, the reflection is complete.
 𝜎
If the conductor is not a perfect conductor but a very good conductor, 𝜖 ≫ 1.
2𝜔

The approximation adopted in the preceding theory gives,

𝜔𝜎𝜇1 1+𝑖
kT = α + iβ = (1 + i) √ = A
2 𝛿

Where we have made use of the definition of skin depth – 𝛿,

Substituting (A) into eq.-6.1,

1+𝑖 1 𝑖
− ω (𝜖1 𝜇1 )1/2 { − ω (𝜖1 𝜇1 )1/2 }+
𝛿 𝛿 𝛿
E0R = 1+𝑖 E0I = 1 𝑖 E0I B
+ ω (𝜖1 𝜇1 )1/2 { + ω (𝜖1 𝜇1 )1/2 }+
𝛿 𝛿 𝛿

Therefore the reflection coefficient R given by,

2
1/2
|𝐸0𝑅 |2 {1− ω (𝜖1 𝜇1 ) 𝛿} +1
R= |𝐸0𝐼 |2
= 1/2 2
{1+ ω (𝜖1 𝜇1 ) 𝛿} +1

𝜎
Because, ≫ 1, ω (𝜖1 𝜇1 )1/2 𝛿 ≪ 1 , and, hence
𝜖2 𝜔

R ≅ 1 - 2 ω (𝜖1 𝜇1 )1/2 𝛿 7

2ω 𝜖1 𝜇1
= 1- 2 √ 8
𝜎𝜇2

If the magnetic permeabilities are assumed to be equal,

2ω 𝜖1
R =! – 2 √ 9
𝜎

The measure of energy transmitted into the conducting medium is obtained by calculating the
transmission coefficient T which gives the ratio of the energy transmitted to the energy incident

2ω 𝜖1
T= 1-R = 2 √ 10
𝜎

In order to have an idea of the magnitude of the fraction of the energy transmitted in a good
conductor, let calculate T for copper with the following data;

𝜎for copper= 6 x 107 (Ohm m)-1 , 𝜈 = 1010 sec-1 = 2.722 x 10-4 Htz
 2 ×2π ×1010 × 8.85 × 10−12
T=2√ ≅ 3 x 10-4
6 × 107

Which is extremely small for direct measurements to be made.

The total energy in the medium-1is,

E = E I + ER here, EI = - ER = 𝑒̂ n E0I 𝑒 𝑖(𝑘𝐼 𝑥− 𝜔𝑡)

Note that the amplitudes in both the waves are taken to be equal. Justification for this is found
in the fact that for metallic surface the reflection coefficient is almost unity. Therefore

E = 𝑒̂ n E0I 𝑒 −𝑖𝜔𝑡 {𝑒 𝑖𝑘𝐼𝑥 − 𝑒 −𝑖𝑘𝐼𝑥 }

On taking the real parts,

E = 2 𝑒̂ n E0I sin 𝜔𝑡 sin 𝑘𝐼 𝑥

Therefore, the field in medium-1 is represented by a standing wave. The field will behave in
below figure.

F-4: standing wave

Theory- 1.6:- the frequency dependence of ϵ, μ and σ.

We have seen that the propagation of electromagnetic waves through matter is governed by
three properties of the materials. Which we took to be constants: the permittivity ϵ, the
permeability μ, and the conductivity σ. Actually, each of these parameters depends to some
extent on the frequency of the waves you are considering. Indeed, if ϵ and μ were truly
constant, then the speed of waves in a nonconductor.
1 𝑐
𝜈= and the index of refraction, n =
√∈𝜇 𝜈

would likewise be constant. And yet, it is well known from optics that n is a function of 𝜔. Thus
a prism bends blue light more sharply than red and spreads white light out into a rainbow of
colors. This phenomenon is called dispersion. By extension, whenever the speed of waves
varies with frequency, the supporting medium is called dispersive. (note that:- the transverse
waves on a stretched string- will not serve to illustrate dispersion, for the propagation speed
√𝑇/𝜇 is independent of frequency. Dispersion does occur in water waves, however, and in
many other contexts.

F-5: wave dispersion & wave envelope

Because waves of different frequency travel at different speeds in a dispersive medium, a wave
form that incorporates a range of frequencies will change shape as it progress. A sharpely
peaked wave typically flattens out as it goes alone, and whereas each sinusoidal component
travels at the ordinary waves (or phase) velocity,
𝜔
𝜈=𝑘

The packet as a whole (the “envelope”) travels at the so-called group velocity.
𝑑𝜔
𝜈𝑔 = 𝑑𝑘

You can demonstrate this by dropping a rock into the nearest pond and watching the waves
that form: while the disturbance as a whole spreads out in a circle., moving at speed 𝜈𝑔 , the
ripples that go to make it up will be seem to travel twice as fast (𝜈 =2𝜈𝑔 for these waves). They
appear at the back end of the packet, growing as they move forward to the center, then
shrinking again and fading away at the front (see in figure-1.6.2). we shall not concern ourselves
with these matters- we’ll stick to monochromatic waves. For which the problem does not arise.
But please note that the energy carried by a wave packet in a dispersive medium ordinarily
travels at the group velocity, not the phase velocity. Don’t be too alarmed, if in some
circumstances 𝜈 comes out greater than c.
It is not unreasonable that ϵ, μ and σ should be functions of𝜔. After all, they describe the
response of the material to electric and magnetic fields, and this response can depend on the
driving frequency, much as the response of a swing depends on the frequency with which you
push it. In practice, μ is very close to 𝜇0 , for most materials, and its variation with is
insignificant. This leaves ϵ and σ. It describes, a model for the behavior of electrons inside
matter, from which we can deduce the functions ϵ(𝜔) and σ(𝜔).