Plane Waves Reflection From Media Interface: Dr. Rakhesh Singh Kshetrimayum Dr. Rakhesh Singh Kshetrimayum

6.
Plane waves reflection from media

interface
Dr. Rakhesh Singh Kshetrimayum
1 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

6.1 Introduction
Plane waves reflection
from a media interface
Normal Oblique
incidence Perfect incidence
conductor TE
Good TM Effect on
conductor Brewster angle polarization
Lossy conducting
medium
Total internal reflection
Lossless medium
Fig. 6.1 Plane waves reflection from media interface
6.1 Introduction
Till now, we have studied plane waves in various medium
Let us try to explore how plane waves will behave at a media
interface
In practical scenarios of wireless and mobile
communications,
radio wave will reflect from
walls &
other obstacles on its path
When a radio wave reflects from a surface,
the strength of the reflected waves is less than that of the
incident wave

6.1 Introduction
The ratio of the two (reflected wave/incident wave) is
known as the ‘reflection coefficient’ of the surface
This ratio depends on the
conductivity (σ),
permittivity (ε) and
permeability (µ)
of the material that forms the reflective surface
as well as material properties of the
air
from which the radio wave is incident

6.1 Introduction
Some part of the wave will be transmitted through the
material
How much of the incident wave has been transmitted
through the material is
also dependent on the material parameters mentioned above
It is given by another ratio known as ‘transmission
coefficient’
It is the ratio of the transmitted wave divided by the incident
wave

6.1 Introduction
In plane wave reflection from media interface,
what we will be doing is
basically writing down the
electric and
magnetic field expressions
in all the regions of interest and
apply the boundary conditions
to get the values of the
transmission and
reflection coefficients

6.1 Introduction
One of the possible applications of such an exercise is in
wireless communication,
where we have to find the
transmission and
reflection coefficients of
multipaths
Electromagnetic waves are often
reflected or
scattered or
diffracted
at one or more obstacles before arriving at the receiver

6.2 Plane wave reflection from media
interface at normal incidence
For smooth surfaces,
EM waves are reflected;
For rough surfaces,
EM waves are scattered;
For edges of surfaces,
EM waves are diffracted
Fig. 6.2 depicts multipath for a mobile receiver in the car from
direct,
reflected and
diffracted waves
of the signal sent from the base station

6.1 Introduction
Fig. 6.2 Illustration of reflected and diffracted wave in mobile

communications (this chapter’s study will be very helpful in
understanding wireless and mobile communications)
We will consider the case of normal incidence,
when the incident wave propagation vector is along the normal
to the interface between two media
6.2.1 Lossy conducting medium
We will assume plane waves with electric field vector
oriented along the x-axis and
propagating along the positive z-axis without loss of
generality
For z<0 (we will refer this region as region I and it is
assumed to be a lossy medium)

Let us do a more generalized analysis by assuming this region
I as a lossy medium
It could be a lossless medium like free space
The incident electric and magnetic fields can be expressed as
r
ˆ o e−γ1z
Ei = xE
r 1
H i = y Eo e −γ1z
ˆ
η1

where η1 is the medium 1 wave impedance and
Eo is the arbitrary amplitude of the incident electric field
The expression for intrinsic wave impedance and propagation
constant
jωµ1 jωµ1 jωµ1 jσ
η1 = = = ; γ 1 = α1 + j β1 = jω µ1ε1 1 − 1
γ1 jωµ1 ( jωε1 + σ 1 ) jωε1 + σ 1 ωε1

rr r
EEii r Et r
γi γt
r 1
r 2
Hi Ht
rr
r E
Err
γr1 r
Hr
ε1 , µ1 , σ 1 ε 2 , µ2 , σ 2
Fig. 6.3 A plane EM wave is incident from region I or medium 1 for z<0
Convention:
a circle with a dot in the center an arrow pointing
perpendicularly out of the page and
a circle with a cross an arrow pointing perpendicularly into
the page)
Notation for fields:
subscript i incident, r reflected and ttransmitted

Note that the incident wave from region I will be partially
reflected and transmitted at the media interface between two
regions
For reflected wave, z<0, wave direction is in the –z axis and
the reflected electric field is expressed as
r
Er = xˆΓEo e + γ 1z
where the Γ is the reflection coefficient

The reflection coefficient is defined as the ratio of amplitude
of the reflected electric field divided by amplitude of the
incident electric field as follows:
Er
Γ=
Ei
The reflected magnetic field can be obtained from the

reflected electric field using the Maxwell’s curl equations as

r r
∇ × Er = − jωµ1 H r
xˆ yˆ zˆ
r r
r ∇ × Er j∇ × E r j ∂ ∂ ∂
⇒ Hr = − = =
jωµ1 ωµ1 ωµ1 ∂x ∂y ∂z
ΓEo e + γ1z 0 0
j ∂ +γ1z  j γ1
= yˆ 
ωµ1  ∂z
Γ E(
o e 

= ˆ
y)ωµ1
γ 1Γ E(
o e + γ1z
= − ˆ
y )
jωµ1
ΓEo e + γ1z ( )
r Γ
∴ H r = − yˆ Eo e + γ1z
η1

Therefore, the Poynting vector for reflected wave of region I
is
r r r* 2 2 zˆ
S r = Er × H r == − Γ Eo e +2α z *
η1
shows power is traveling in the -z axis for the reflected wave
For transmitted wave, z>0, wave is propagating in lossy
medium 2 (we will refer this region as region II),

r
E t = xˆτE o e −γ 2 z
r E
H t = yˆ τ o e − γ 2 z
η2
where η2 is the wave impedance of lossy medium 2 and τ is

the transmission coefficient
intrinsic wave impedance and propagation constant are

jωµ 2 jωµ 2 jωµ 2 jσ

η2 = = = ; γ 2 = α 2 + j β 2 = jω µ 2ε 2 1 − 2
γ2 jωµ 2 ( jωε 2 + σ 2 ) jωε 2 + σ 2 ωε 2
transmission coefficient is defined as the ratio of amplitude of

the transmitted electric field divided by amplitude of the
incident electric field as follows:
Et
τ=
Ei

Our intention here is to find the transmission and reflection
coefficients
Let us rewrite the fields in the two regions: region I (z<0)
and region II (z>0) (see Table 6.1) and
Apply the boundary conditions to obtain the two unknown
coefficients

Table 6.1 Fields in the two lossy regions (normal incidence)
Region I (lossy medium 1) Region II (lossy medium 2)
r r 1 r r Eo −γ 2 z
ˆ o e−γ1z
Ei = xE H i = yˆ Eo e −γ1z E t = xˆτE o e − γ 2 z H t = ˆ
y τ e
η1 η
2
r r Γ
Er = xˆΓEo e + γ 1z H r = − yˆ Eo e + γ1z
η1

This is basically boundary value problem with boundary
conditions at z=0
Note that total electric and magnetic fields (both incident
and reflected) are tangential to the interface at z=0
Similarly, the transmitted electric and magnetic field are also
tangential to the interface at z=0
Also note that there are no surface current density at the
interface
Hence, the tangential components of electric and magnetic
fields must be continuous at the interface z=0

r r r
Ei + Er = Et ⇒ (1 + Γ ) E0 = τ E0 ⇒ 1 + Γ = τ
z =0
r r r  1− Γ   τ  1− Γ τ
Hi + H r = Ht ⇒ E0   = E0 ⇒ =
z =0
 η1   η2  η1 η2
Therefore,
1− Γ τ 1− Γ 1+ Γ η 2 − η1
= ⇒ = ⇒ (1 − Γ )η 2 = η1 (1 + Γ ) ⇒ Γ =
η1 η2 η1 η2 η 2 + η1
Hence,
η 2 − η1 2η 2
τ = 1+ Γ = 1+ =
η 2 + η1 η1 + η 2

6.2.2 Lossless medium
If the regions are lossless dielectric,
then, σ = 0 and µ and ε are real quantities
The propagation constant for this case is purely imaginary
and can be written as
γ = jβ =jω√(µε)
The wave impedance of the dielectric is
µr
η= jωµ = jωµ = ωµ
= µ
= η0
εr
γ jβ ω µε ε

For a lossless medium,
η is real, so,
both Γ and τ are real
Electric and magnetic fields are in phase with each other
The wavelength in the dielectric is
2π 2π
λ = =
β ω µε
And the phase velocity is
ω 1 c
v p = = =
β µε µrε r
where c is the speed of light in free space

Table 6.2 Fields in the two lossless regions (normal incidence)
Region I (lossless medium 1) Region II (lossless medium 2)
r r 1
ˆ o e− j β1z
Ei = xE H i = yˆ Eo e − j β1 z r E
η1 H t = yˆτ o e − j β 2 z
η2
r r Γ
Er = xˆΓEo e + j β1z H r = − yˆ Eo e + j β1z r
η1 Et = xˆτ Eo e − j β 2 z

Applying the boundary conditions like before,
we can get the expression for
reflection and
transmission coefficients,
we will get the same expression as before
η 2 − η1 2η 2
Γ= ; τ=
η 2 + η1 η1 + η2

Poynting vector in region I
r1 r 1 r1 * r r r r *  E0* j β1z E0* − jβ1z 

(
S = Etot × H tot ) ( ) (
= Ei + Er × H i + H r ) = (E e 0
− j β1 z
xˆ + ΓE0e + j β1 z
)
xˆ ×  e yˆ − Γ e yˆ 
 η1 η1 
 E0 2 Γ E0 2 e −2 j β1z Γ E0 2 e +2 j β1z Γ 2 E0 2   E0 2 Γ E
2

= − + −  zˆ =  (1 − Γ2 ) +
0
e+2 jβ1z − e −2 j β1z( )  zˆ
 η1 η1 η1 η1 
  η1 η1 
 E0 2 Γ E
2

= (1− Γ ) −
2 0
2 j sin ( 2 β1 z )  zˆ
 η1 η1 

Poynting vector in region II
2 *
r2 r r *  τ E0 − jβ 2 z  2
τ E0
S = Et × H t = x τ E0 e
ˆ (
− j β2 z
× e )  y=z
ˆ ˆ
 η2  η2
Q1 − Γ 2 = 1 −
(η 2 − η1 ) =
4η η
2 1
⇒
1 − Γ 2
=
4η2
2 2 2
( 2 1) ( 2 1)
η + η η + η η 1 ( 2 1)
η + η
τ=
2η 2
⇒τ 2 =
4η 2
2
⇒
1− Γ τ r 2
2
= ∴ S = zˆ
(
21 − Γ 2
) E0
2
2
η 2 + η1 ( 2 1)
η + η η1 η2 η1

Power conservation
Compare the time average power flow in the two regions
For z <0, the time average power flow through 1m2 cross
section is
1 r1 1 21
1−− Γ2
S 1
avg
2
( 2
)
= Re S • zˆ = E0
η1
And for z >0, the time average power flow through 1m2
cross section is
1 r2 1 2 1− Γ
2
S 2
avg
2
(
= Re S • zˆ = E0
2
) η1

Hence,
1 2
Savg = Savg
So real power is conserved
6.2.3 Good conductor:
If the region II (z >0) is a good (but not perfect) conductor
and
region I is a lossless medium like free space,
the propagation constant for medium 2 is

ωµ2σ 2 1
γ 2 = α 2 + j β 2 = (1 + j ) = (1 + j )
2 δ s2
where δ is the skin depth for the medium 2,
2
s
which is a good conductor

Similarly, the intrinsic impedance of the conducting medium
2 (see also example 6.2) simplifies to
2
jωµ 2 jωµ 2 j 2 (ωµ 2 ) j 2ωµ 2
η2 = = = =
γ2 ωµ2σ 1 + j ωµ 2σ 1+ j σ
(1 + j )
2
j (1 − j ) 2ωµ 2 ( j + 1) 2ωµ 2 ( j + 1) 2ωµ 2 ωµ 2
= = = = ( j + 1)
(1 + j )(1 − j ) σ (1 − j 2 ) σ 2 σ 2σ

Now, the wave impedance of medium 2 is complex, with a
phase angle of 45o so the electric
and magnetic field will be 45o out of phase and Γ and τ will
be complex.
Now, let us write down the field expressions in region I and
II for lossless dielectric (region I) and good conductor
(region II) interface

Table 6.4 Fields in the two regions (normal incidence)
Region I (lossless medium) Region II (good conductor)
r r 1 r
ˆ o e− jβ1z
Ei = xE H i = yˆ Eo e − j β1z Et = xˆτ Eo e − γ 2 z
η1
r r Γ r
Er = xˆΓEo e + j β1z H r = − yˆ Eo e + jβ1z
E
H t = yˆτ o e − γ 2 z
η1 η2

Applying the boundary conditions like before,
we can get the expression for
reflection and
transmission coefficients,
we will get the same expression as before
η 2 − η1 2η 2
Γ= ; τ=
η 2 + η1 η1 + η2

Note that we have to use the previous expressions for
intrinsic wave impedances for lossless medium and good
conductor
µ ωµ2
η1 = ; η2 = (1 + j )
ε 2σ 2
Power conservation
let us find out the Poynting vector in the two regions
Noting that the field in region I comprise of incident and
reflected wave,
we can write the Poynting vector for z <0

r1 r 1 r1 * r r r r *  E0* j β1z * E0 − j β1 z 
*
S = Etot × H tot ( ) ( ) (
= Ei + Er × H i + H r ) = (E e
0
− j β1 z
xˆ + ΓE0 e + j β1 z
)
xˆ ×  e yˆ − Γ e yˆ 
 η1 η1 
2
E0
=
η1
(1 − Γ e* −2 j β1 z
+ Γe +2 j β1z − Γ
2
) zˆ
For z >0, only transmitted fields exist and we can write
down the Poynting vector as
2 2
r2 r r *  τ * *
E *  τ E
S = Et × H t = xˆ (τ E0 e −γ z ) ×  * 0 e −γ z  yˆ = zˆ 0
e −2α z
 η2  η 2*

At z=0,
2
r1 E0
S =
η1
(
1 − Γ* + Γ − Γ
2
) zˆ
r 2 E0 2 τ 2 2
4 E0 η2 E0
2
S =
η2*
ˆ
z =
(η1 + η2 )
2
ˆ
z =
η1
1 − Γ (
*
+ Γ − Γ
2
)zˆ
Complex power is conserved

Then the time average power flow through a 1m2 cross
section can be calculated as follows
Since

−Γ*e −2 j β1z + Γe +2 j β1z
is purely an imaginary number, therefore,

2
1 r 1 1 − Γ
1
S avg ( )
= Re S 1 • zˆ = E0
2 2
2
η1
2
r2 1 − Γ
2
S avg
1
( 1
= Re S • zˆ = E0
2 2
)
2
η2
e − 2αz
which shows power balance at z=0 only

Note that the Poynting vector for z>0,
this power density in the lossy conductor decays
exponentially according to the attenuation factor
6.2.4 Perfect conductor:
Now assume that the region II (z >0) contains a perfect
conductor and region I is a lossless medium
Let us rewrite the expressions for propagation constant and
wave impedance of a good conductor and deduce the
expressions for perfect conductor

ωµ 2σ 2 1
γ 2 = α 2 + j β 2 = (1 + j ) = (1 + j )
2 δ s2
ωµ2
η2 = (1 + j )
2σ 2
Perfect conductor implies σ2→∝, and

correspondingly, α2→∝, δs2→0, η2→0
Note that 2η2 η −η
τ= ,Γ = 2 1
η1 + η2 η2 + η1
Therefore, τ→0 and Γ→-1

Table 6.5 Fields in the two regions (normal incidence)
Region I (lossless medium) Region II (perfect conductor)
r r 1
ˆ o e− j β1z
Ei = xE H i = yˆ Eo e − j β1 z r r
η1 Et ≅ 0 Ht ≅ 0
r r Γ
Er = xˆΓEo e + j β1z H r = − yˆ Eo e + jβ1z
η1

The field for z >0 thus decay infinitely fast and
are identically zero in the perfect conductor as we have seen
in example 6.2 as well
The perfect conductor can be thought of as ‘shorting out’ the
incident electric field
Now let us write down the field expressions in region I and II
for lossless dielectric and perfect conductor interface (see
Table 6.5)

Power conservation
In order to see the power conservation,
let us find out the Poynting vector in the two regions
Noting that the field in region I comprise of
incident and
reflected wave,
we can write the Poynting vector for z <0 (Γ=-1)
r1 r r
( )
ˆ 0 e − j β1z − e j β1z = − xˆ 2 jE0 sin β1 z
Etot = Ei + Er = xE

r1 r E E
( )
H tot = H i + H r = yˆ 0 e − j β1 z + e j β1z = yˆ 2 0 cos β1 z
η1 η1
Therefore, the Poynting vector for z<0 is

2
r1 r1 * 4 E0
1
S = Etot × H tot( ) = − zj
ˆ
η1
sin β1 z cos β1 z
which has a zero real part and

thus indicates that no real power is delivered to the perfect conductor
Fields in region II is anyway zero and hence power is also zero

interface at oblique incidence
We will consider the problem of a plane wave
obliquely incident on a plane interface
between two lossy conducting regions
We will first consider two particular cases of this problem as
follows:
the electric field is in the xz plane (parallel polarization)
the electric field is in normal to the xz plane (perpendicular
polarization)

Any arbitrary incident plane wave can be expressed
as a linear combination of these two principal polarizations
The plane of incidence is that plane containing
the normal vector to the interface and
the direction of propagation vector of the incident wave

For Fig. 6.4, this is the xz plane
For perpendicular polarization (TE),
electric field is perpendicular to the plane of incidence
For parallel polarization (TM),
electric field is parallel to the plane of incidence

x
r r
r r Et , H t
Er , H r
θr θt z
θi
r r
Ei , H i n̂
ε , µ ,σ
1 1 1
ε 2 , µ2 , σ 2
Region I Region II
Fig. 6.5 Oblique incidence of plane EM wave at a media interface

6.3.1 Perpendicular polarization (TE):
In this case, electric field vector is perpendicular to the xz
plane,
Hence, it will have component along the y-axis
Since the electric field is transversal to the plane of incidence
They are also known transverse electric (TE) waves

x̂ x̂
r
ri γr
γ 1
γ 1 sin θ r
γ 1 sin θ i 1
n̂ θi ẑ n̂ θr ẑ
γ 1 cosθ i γ 1 cos θ r
x̂
r
γ 2t
γ 2 sin θ t
n̂ θt ẑ
γ 2 cos θt
Fig. 6.6 Wave propagation vector for (a) incident (b) reflected and (c) transmitted
EM waves at oblique incidence
Electromagnetic Field Theory by R. S. Kshetrimayum
52 3/25/2014
Let us assume that the incident wave propagates in the first
quadrant of xz plane without loss of generality and
r
γ (incident propagation vector) makes an angle θi with
1
i
the normal (see Fig. 6.6 (a))

r r
γ 1i • z ' = (γ 1 cos θi zˆ + γ 1 sin θi xˆ ) • ( zzˆ + xxˆ ) = γ 1 cos θi z + γ 1 sin θi x = γ 1 ( z cos θi + x sin θi )
r
Ei = E0 e 1 ( i
− γ z cosθ + x sin θi )
yˆ
r
r r r ∇ × Ei
Q ∇ × Ei = − jωµ1 H i ⇒ H i =
− jωµ1

xˆ yˆ zˆ
E0  ∂e 1 ( i   ∂e −γ1 ( z cosθi + x sin θi )  
− γ z cosθ + x sin θi )
1 ∂ ∂ ∂
= =  − xˆ  +  zˆ  
− jωµ1 ∂x ∂y ∂z − jωµ1  ∂z  
  ∂x 
 
− γ 1 ( z cosθi + x sin θi )
0 E0 e 0
E0  ∂e −γ1 ( z cosθi + x sinθi )   ∂e −γ1 ( z cosθi + x sin θi )   E0γ 1 −γ1 ( z cosθi + x sin θi )
=  xˆ  −  zˆ   = e {− cos θi xˆ + zˆ sin θi }
jωµ1  ∂z  
  ∂x 
  jωµ1
E −γ z cosθ + x sin θi )
= 0 e 1( i ( − xˆ cos θi + zˆ sin θi )
η1

Let us assume that the reflected wave propagates in the
second quadrant of xz plane and
rr
γ1 (reflected propagation vector) makes an angle θr with
the normal (see Fig. 6.6 (b))
r r
γ 1r • z ' = ( −γ 1 cos θ r zˆ + γ 1 sin θ r xˆ ) • ( zzˆ + xxˆ ) = −γ 1 cos θ r z + γ 1 sin θ r x = γ 1 ( − z cos θ r + x sin θ r )
r
Er = E0 ΓTE e 1 (
−γ − z cosθ r + x sin θ r )
yˆ

r r
Note that γ 1r and γ 1i will have the same magnitude
since both the waves are still in the same region I,
only their direction changes
Since the Poynting vector must be negative like the previous
case of normal incidence,
r E
H r = 0 ΓTE e 1 (
− γ − z cosθ r + x sin θ r )
( xˆ cos θ r + zˆ sin θ r )
η1
You could also use the Maxwell’s curl equation below to find
r
this r
Hr =
∇ × Er
− jωµ1
The transmitted fields will have similar expression with the
incident fields except
that now the θi should be replaced by θt (angle that transmitted
propagation vector makes with the normal),
γ1 should be replaced by γ2 (wave is in region II now) and
multiplication by (transmission coefficient)
The transmitted fields are r
ˆ 0τ TE e−γ 2 ( z cosθt + x sinθt )
Et = yE
r
r ∇ × Et E0τ TE −γ 2 ( z cosθt + x sin θt )
Ht = = e ( − xˆ cos θt + zˆ sin θt )
− jωµ2 η2

Table 6.5 Fields in two regions (oblique incidence:
perpendicular polarization)

r
Ei = E0e 1 ( i
−γ z cosθ + x sin θi )
yˆ r
r
ˆ 0τ TE e−γ 2 ( z cosθt + x sinθt )
Et = yE
E −γ z cosθ + x sin θi )
H i = 0 e 1( i ( − xˆ cos θi + zˆ sin θi )
η1
r Eτ
H t = 0 TE e 2 ( t
− γ z cosθ + x sin θt )
r ( − xˆ cos θt + zˆ sin θt )
Er = E0 ΓTE e 1 (
− γ − z cosθr + x sin θ r )
yˆ η2
r EΓ
H r = 0 TE e 1 ( i
− γ z cosθ + x sin θ i )
( xˆ cos θ r + zˆ sin θ r )
η1
Equating the tangential components of electric field
(electric field has only Ey component and it is tangential at the
interface z=0) and
magnetic field
(magnetic field has two components: Hx and Hz and only Hx is
tangential at the interface z=0)
at z=0 gives
e −γ1x sin θi + ΓTE e −γ1x sin θr = τ TE e −γ 2 x sin θt
−1 ΓTE τ TE
cos θ i e −γ1x sin θi + cos θ r e −γ1x sin θr = − cos θ t e −γ 2 x sin θt
η1 η1 η2

If Ex and Hy are to be continuous at the interface z = 0 for all
x,
then, this x variation must be the same on both sides of the
equations (also known as phase matching condition)
γ 1 sin θ i = γ 1 sin θ r = γ 2 sin θ t
⇒ θi = θ r ; γ 1 sin θi = γ 2 sin θt

The first is Snell’s law of reflection
which states that the angle of incidence equals the angle of
reflection
The second result is the Snell’s law of refraction
(refraction is the change in direction of a wave due to change in
velocity from one medium to another medium)
Also note that refractive index of a medium is defined as
c µ r ε r µ 0ε 0
n= = = µr ε r
vp µ 0ε 0

hence, for a lossless dielectric media,
sin θi γ 2 β 2 µ 2ε 2 v1 ε n
= = = = = 2 = 2
sin θt γ 1 β1 µ1ε1 v2 ε1 n1
Now we can simplify above two equations by applying Snell’s

two laws as follows
1 + ΓTE = τ TE
cos θi cos θr τ TE
− + ΓTE =− cos θt
η1 η1 η2

The above two equations has two unknowns τTE and ГTE and
it can be solved easily as follows
 cos θi cos θ r  η2
τ TE =  − ΓTE  1 + ΓTE = τ TE
 η1 η1  cos θt
Therefore,
 η cos θi η cos θr   η2 cos θr   η2 cos θi 

1 + ΓTE =  2 − ΓTE 2  ⇒ ΓTE 1 + = − 1
η
 1 cos θ t η1 cos θ t   η1 cos θt η
  1 cos θ t 
 η cos θt + η2 cos θr   η2 cos θi −η1 cos θt  η2 cos θi −η1 cos θt
⇒ ΓTE  1 =  ⇒ ΓTE =
 η1 cos θt   η1 cos θt  η1 cos θt + η2 cos θr

η2 cos θi − η1 cos θt η1 cos θt + η2 cos θr + η2 cos θi − η1 cos θt 2η2 cos θi
∴τ TE = 1 + ΓTE = 1 + = =
η1 cos θt + η2 cos θr η1 cos θt + η2 cos θ r η1 cos θt + η2 cos θr
Hence, the reflection and transmission coefficients for

oblique incidence (Fresnel coefficients) for perpendicular
polarization are given as follows:
η2 cos θi − η1 cos θt 2η2 cos θi
ΓTE = τ TE =
η2 cos θi + η1 cos θt η2 cos θi + η1 cos θt
For normal incidence, it is a particular case and put
θi = θ r = θt = 0

6.3.2 Parallel Polarization (TM):
In this case, electric field vector lies in the xz plane
Since the magnetic field is transversal to the plane of
incidence such waves are also called as transverse magnetic
(TM) waves
So let us start with H vector which will have only y
component,
r E
H i = yˆ 0 e − γ 1 ( z cos θ i + x sin θ i )
η1

r
r r r ∇ × Hi
Q ∇ × H = jωε1 E ⇒ Ei =
jωε1
xˆ yˆ zˆ
1 ∂ ∂ ∂ E0   ∂e−γ1 ( z cosθi + x sin θi )   ∂e−γ1 ( z cosθi + x sin θi )  
= = −  xˆ  +  zˆ  
jωε1 ∂x ∂y ∂z jωε1η1   ∂z  
  ∂x 
 
E0 − γ 1 ( z cosθi + x sin θi )
0 e 0
η1
E0γ 1 − γ 1 ( z cosθi + x sin θi )
= e {cos θi xˆ − zˆ sin θi } = E0e−γ ( z cosθ + x sinθ ) ( cos θi xˆ − zˆ sin θi )
1 i i
jωε1η1

Similar to the previous case of parallel polarization, we can
write the reflected magnetic field vector as
r EΓ
ˆ −γ 1 ( − z cosθr + x sin θr )
H r = − yˆ 0 TM ye
η1
The transmitted fields will have similar expression with the

incident fields except that
now the θi should be replaced by θt (angle that transmitted
propagation vector makes with the normal),
γ1 should be replaced by γ2 (wave is in region II) and
multiplied by a factor (transmission coefficient for TM case)

r − γ ( z cosθt + x sin θt )
Et = E0τ TM e 2 ( xˆ cos θt − zˆ sin θt )
r τ E −γ z cosθ + x sinθt )
H t = yˆ TM 0 e 2 ( t
η2
Table 6.6 Fields in two regions (oblique incidence: parallel

polarization) {next page}

r
Ei = E0e 1 ( i
− γ z cosθ + x sin θi ) r
( cos θi xˆ − zˆ sin θi ) − γ ( z cosθt + x sin θt )
Et = E0τ TM e 2 ( xˆ cos θt − zˆ sin θt )
r E
H i = yˆ 0 e − γ 1 ( z cos θ i + x sin θ i ) r τ E −γ z cosθ + x sin θt )
η1 H t = yˆ TM 0 e 2 ( t
η2
r
Er = E0ΓTM e 1 (
− γ − z cosθ r + x sin θ r )
{cos θr xˆ + zˆ sin θi }
r EΓ
ˆ −γ 1 ( − z cosθr + x sin θr )
H r = − yˆ 0 TM ye
η1

Equating the tangential components of magnetic field
(magnetic field has only Hy component and it is tangential at the
interface z=0) and
electric field
(electric field has two components: Ex and Ez and only Ex is
tangential at the interface z=0)
at z=0 gives
e −γ1x sin θi cos θi + e −γ1x sin θr ΓTM cos θ r = τ TM e −γ 2 x sin θt cos θ t
e −γ1x sin θi ΓTM e−γ1x sin θr τ TM e−γ x sinθ

2 t
− =
η1 η1 η2

Therefore,
γ 1 sin θ i = γ 1 sin θ r = γ 2 sin θ t
⇒ θi = θ r ; γ 1 sin θi = γ 2 sin θt
which is Snell’s law of reflection and refraction and this

implies that
cos θi + ΓTM cos θ r = τ TM cos θt
1 τ TM
(1 − ΓTM ) =
η1 η2

The above two equations has two unknowns τTM and ГTM and
it can be easily solved as follows
cos θi + ΓTM cos θ r η2
τ TM =
cos θt
τ TM = (1 − ΓTM )
η1
Therefore,
cos θi + ΓTM cos θr η2 η cos θi ΓTM cos θ r
= (1 − ΓTM ) ⇒ 2 (1 − ΓTM ) = +
cos θt η1 η1 cos θt cos θt
 η cos θ r   cos θi η2   −η2 cos θt −η1 cos θ r   η1 cos θi −η2 cos θt 
⇒ ΓTM  − 2 −  = −  ⇒ Γ TM  = 
 η1 cos θt   cos θt η1   η1 cos θt   η1 cos θt 
η cos θt −η1 cos θi η2 cos θt −η1 cos θi
⇒ ΓTM = 2 =
η2 cos θt + η1 cos θ r η2 cos θt + η1 cos θi

η2 η  η cos θt −η1 cos θi  η2  η2 cos θt + η1 cos θi −η2 cos θt + η1 cos θi 
∴τ TM = (1 − ΓTM ) = 2 1 − 2 =  
η1 η1  η2 cos θt + η1 cos θi  η1  η2 cos θt + η1 cos θi 
η2  2η1 cos θi  2η2 cos θi
=  =
η1  η2 cos θt + η1 cosθi  η2 cosθt + η1 cosθi
Hence, the reflection and transmission coefficients for oblique

incidence (Fresnel coefficients) for parallel polarization are given as
follows:
η2 cos θt − η1 cos θi 2η2 cos θi
∴ΓTM = τ TM =
η2 cos θt + η1 cos θi η2 cos θt + η1 cos θi
Note that cosine terms multiplication in the above equations is

different from the previous expression for TE waves

6.3.3 Brewster angle:
For Γ=0, the angle of incidence is known as Brewster angle
(θB)
This means at this angle of incidence of the EM wave from
region I,
there will be no reflection from region II,
all the EM waves will be absorbed

For parallel polarization,
θi = θ BTM
η2 cos θt − η1 cos θi
ΓTM = =0
η2 cos θt + η1 cos θi
⇒ η2 cos θt = η1 cos θ BTM
Also we have from 2nd Snell’s law
γ 1 sin θ BTM = γ 2 sin θ t

Using the above two equations, we can get (see textbook)

 η12 
 η2 − 1 
 2 
sin 2 θ BTM = 2 2 
 η1 − γ 1 
 η 22 γ 22 
For lossless media, we have,

µ ε2
2 2 2
1
−1
η12 µ γ µ µ2ε 2 µ ε 2 γ 12 µ ε1
2 ε1µ 2
Q 2 = 2 = 1
2
=
2
; 2 =
1
∴ sin 2 θ BTM
1 1
=
η2 γ 1 µ µ1ε1µ2 ε1µ2 γ 2 ε 2 µ2
2 µ ε 2 µ ε1
1
− 1
ε1µ2 ε 2 µ2

For magnetic media, ε1 = ε 2

2 TM
sin θ B is of the form 1/0 and Brewster angle does not
exist
But, for dielectric media, it exist and can be calculated as
follows
ε2
−1
ε1 ε2 ε n
⇒ sin θ BTM = = ⇒ tan θ BTM = 2 = 2
ε 2 ε1 ε 2 + ε1 ε1 n1
−
ε1 ε 2

For perpendicular polarization,
Similarly, Brewster’s angle doesn’t exist for perpendicular
polarization at oblique incidence for a non-magnetic or
dielectric medium (see textbook for derivation)
For magnetic medium
2
µ2 µ1 − ( µ2 ) µ2 µ2 µ2 n2
∴ sin 2 θBTE = 2 2
= ⇒ sin θBTE = ; tan θBTE = =
( µ1 ) − ( µ2 ) µ1 + µ2 µ1 + µ2 µ1 n1

Brewster angle exist for oblique incidence (perpendicular
polarization) for interface between two magnetic media
An incident wave consisting of both polarizations will have
reflected wave for only one polarization for Brewster angle
Brewster angle is also known as Polarizing angle (used in
optics and lasers)
In other words, an incident wave which is composed of both
TE and TM waves will have reflected TE (TM) waves only for
dielectric (magnetic) media interface
Thus, a circularly polarized incident wave at polarizing angle
will become a linearly polarized wave

6.3.4 Total internal reflection:
Used in optical fibers
Light strikes an angle of incidence greater than critical angle,
all of the light energy will be reflected back, and
none of them will be absorbed
This critical angle is related to the refractive index of the two
media and is given by  n2 
θ c = sin −1  
n
 1
where are the refractive index of medium 2 and 1
respectively

In fact, this can be observed for both the polarizations we
have discussed above
From Snell’s Second law, we have,
γ 2 sin θt
γ 1 sin θi = γ 2 sin θt ⇒ sin θi =
γ1
If we consider non-magnetic materials (dielectrics) of , in
that case,
ε 2 sin θt
⇒ sin θi =
ε1

Hence, for
π
θi = θ c ,θt =
2
(it means the transmitted wave travels along the interface, for
our case, it is x-axis) which implies that
ε2  ε2 
⇒ sin θ c = ⇒ θ c = sin 
−1

ε1 ε
 1

We can write the above relation as
 ε2   µ 0 µ r 2 ε 0ε r 2   µ r 2ε r 2  −1  n2 
θ c = sin −1   = sin −1
  = sin −1
  = sin  
ε
 1  µ0 µ r1ε 0ε r1   µ r1ε r1   n1 
Note that ε1 > ε 2 to have a real value of critical angle

Besides, we have assumed that for a dielectric, µ r1 = µ r 2 = 1

What will happen when the angle of incidence is greater than
the critical angle?
For all angles greater than the critical angle,
using Snell’s second law, and
noting that waves are moving from a denser to a rarer dielectric
medium,
µ r1 = µ r 2 = 1 ε1 > ε 2
µ1ε1 sin θi
Qθi > θ c ∴ sin θt = > 1 ⇒ cos θt = ± 1 − sin 2 θt
µ 2ε 2
The value of cos θt should be imaginary

It can be shown that (see textbook for derivation)
2 2
r r r* E0τ − j 2 β2 z cosθt E0τ − j 2 β2 z cosθt
∴ S = E × H = xˆ e sin θt + zˆ e cos* θt
η2 η2
For z>0, since cos θ t is an imaginary number, cos* θt is

also purely imaginary
Hence, the second term is imaginary, therefore, no real
power is flowing along z-axis in region II
The first term which has direction along x-axis is real and is
exponentially decaying with z

Hence,
When a wave is incident at an angle of incidence greater than
or equal to the critical angle from region I,
the wave will be total internally reflected in region I
Surface waves exist in the region II,
which is propagating along x-axis (direction of power flow),
and
attenuating very fast along z-axis
People have used this for communications using submarines

6.3.5 Effects on polarization:
Since any arbitrary wave incident obliquely from region I can
be decomposed into linear combinations of TE and TM
waves and
we also know that the expressions for reflection and
transmission coefficients for parallel and perpendicular
polarizations are different
η2 cos θi − η1 cos θt η2 cos θt − η1 cos θi
ΓTE = ; ΓTM =
η2 cos θi + η1 cos θt η2 cos θt + η1 cos θi

So, any circularly polarized incident wave will
become elliptically polarized after reflection or transmission
Similarly, a linearly polarized wave (note that linear
polarization could be expressed as linear combination of
RHCP and LHCP) may be
rotated after reflection

For instance,
If the region II is a perfect conductor, η2=0, Г=-1 for both
TE and TM cases,
this means a RHCP wave after reflection will become LHCP
wave and vice versa
This is why circular polarization is not used for indoor

communication systems since we have many reflections inside a
room

Plane waves reflection from a media interface
6.4 Summary
Normal incidence Oblique incidence
TE
Perfect ΓTE =
η2 cos θi − η1 cos θt
η2 cos θi + η1 cos θt
Lossy conducting conductor 2η2 cos θi
Effect on
τ TE
medium polarization
=
η2 cos θi + η1 cos θt
η −η 2η2 τ→0 and Γ→-1
Γ= 2 1; τ =
η 2 + η1 η1 + η2 TM
Brewster CPEP
Lossless η cos θt − η1 cos θi
angle
1
Savg 2
= Savg ∴ΓTM = 2 Total internal
medium η2 cos θt + η1 cos θi
1 r 1 2 1− Γ
2
2η2 cos θi
reflection
1
Savg =
2
( )
Re S 1 • zˆ = E0
2 η1
τ TM =
η2 cos θt + η1 cos θi tan θ BTM =
ε 2 n2
=
ε1 n1
Good conductor TE µ 2 n2  ε2 
θ c = sin −1  
tan θ = = ε 
B
µ1 n1  1
r+ 2 µ1ε1 sin θi
2 1− Γ > 1 ⇒ cos θt = ± 1 − sin 2 θt
S 2
avg
1
( )1
= Re S • zˆ = E0
2 2 η2
e −2αz Qθ i > θ c ∴ sin θt =
µ 2ε 2
Fig. 6.7 Plane waves reflection from media interface in a nutshell


Plane Waves Reflection From Media Interface: Dr. Rakhesh Singh Kshetrimayum Dr. Rakhesh Singh Kshetrimayum

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Plane Waves Reflection From Media Interface: Dr. Rakhesh Singh Kshetrimayum Dr. Rakhesh Singh Kshetrimayum

Uploaded by

Copyright:

Available Formats

6.

Plane waves reflection from media

1 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

3 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

4 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

5 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

6 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

7 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

8 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

Fig. 6.2 Illustration of reflected and diffracted wave in mobile

10 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

11 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

12 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

14 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

where the Γ is the reflection coefficient

15 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

The reflected magnetic field can be obtained from the

16 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

17 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

18 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

where η2 is the wave impedance of lossy medium 2 and τ is

19 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

jωµ 2 jωµ 2 jωµ 2 jσ

transmission coefficient is defined as the ratio of amplitude of

20 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

21 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

Region I (lossy medium 1) Region II (lossy medium 2)

22 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

23 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

24 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

25 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

26 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

Region I (lossless medium 1) Region II (lossless medium 2)

27 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

28 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

r1 r 1 r1 * r r r r *  E0* j β1z E0* − jβ1z 

29 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

30 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

31 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

32 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

which is a good conductor

33 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

34 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

Region I (lossless medium) Region II (good conductor)

35 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

36 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

37 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

38 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

Complex power is conserved

39 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

is purely an imaginary number, therefore,

which shows power balance at z=0 only

40 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

41 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

Perfect conductor implies σ2→∝, and

Therefore, τ→0 and Γ→-1

42 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

Region I (lossless medium) Region II (perfect conductor)

43 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

44 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

45 Electromagnetic Field Theory by R. S. Kshetrimayum 3/25/2014

Therefore, the Poynting vector for z<0 is

which has a zero real part and