82 Electromagnetic Waves & Antennas – S. J.

Orfanidis

         

4 E
H
=
1
η−1
1
−η−1
E+
E−
,
E+
E−
=
1
2
1
1
η
−η
E
H
(4.1.4)

Two useful quantities in interface problems are the wave impedance at z:

Reflection and Transmission E(z)
Z(z)= (wave impedance) (4.1.5)
H(z)

and the reflection coefficient at position z:

E− (z)
Γ(z)= (reflection coefficient) (4.1.6)
E+ (z)

Using Eq. (4.1.3), we have:

4.1 Propagation Matrices 1 E

E− (E − ηH) −η Z−η
Γ= = 2 = H =
In this chapter, we consider uniform planes waves incident normally on material inter- E+ 1 E Z+η
(E + ηH) +η
faces. Using the boundary conditions for the fields, we will relate the forward-backward 2 H
fields on one side of the interface to those on the other side, expressing the relationship
in terms of a 2×2 matching matrix. Similarly, using Eq. (4.1.1) we find:
If there are several interfaces, we will propagate our forward-backward fields from E−
one interface to the next with the help of a 2×2 propagation matrix. The combination of 1+
E E+ + E− E+ 1+Γ
a matching and a propagation matrix relating the fields across different interfaces will Z= = =η =η
H 1 E− 1−Γ
be referred to as a transfer or transition matrix. (E+ − E− ) 1−
η E+
We begin by discussing propagation matrices. Consider an electric field that is lin-
early polarized in the x-direction and propagating along the z-direction in a lossless Thus, we have the relationships:
(homogeneous and isotropic) dielectric. Setting E(z)= x̂ Ex (z)= x̂ E(z) and H(z)=
ŷ Hy (z)= ŷ H(z), we have from Eq. (2.2.6): 1 + Γ(z) Z(z)−η
Z(z)= η
Γ(z)= (4.1.7)
1 − Γ(z) Z(z)+η
−jkz jkz
E(z) = E0+ e + E0− e = E+ (z)+E− (z)
(4.1.1) Using Eq. (4.1.2), we find:
1
 1

H(z) = E0+ e−jkz − E0− ejkz = E+ (z)−E− (z)
η η E− (z) E0− ejkz
Γ(z)= = = Γ(0)e2jkz
where the corresponding forward and backward electric fields at position z are: E+ (z) E0+ e−jkz

E+ (z)= E0+ e−jkz where Γ(0)= E0− /E0+ is the reflection coefficient at z = 0. Thus,
(4.1.2)
E− (z)= E0− ejkz
Γ(z)= Γ(0)e2jkz (propagation of Γ) (4.1.8)
We can also express the fields E± (z) in terms of E(z), H(z). Adding and subtracting
the two equations (4.1.1), we find: Applying (4.1.7) at z and z = 0, we have:

1
 Z(z)−η Z(0)−η 2jkz

E+ (z)= E(z)+ηH(z) = Γ(z)= Γ(0)e2jkz = e
2 Z(z)+η Z(0)+η
(4.1.3)
1

E− (z)= E(z)−ηH(z) This may be solved for Z(z) in terms of Z(0), giving after some algebra:
2
Eqs.(4.1.1) and (4.1.3) can also be written in the convenient matrix forms: Z(0)−jη tan kz
Z(z)= η (propagation of Z) (4.1.9)
η − jZ(0)tan kz

81
4.1. Propagation Matrices 83 84 Electromagnetic Waves & Antennas – S. J. Orfanidis

The reason for introducing so many field quantities is that the three quantities which gives after some algebra:
{E+ (z), E− (z), Γ(z)} have simple propagation properties, whereas {E(z), H(z), Z(z)}     
do not. On the other hand, {E(z), H(z), Z(z)} match simply across interfaces, whereas E1 cos kl jη sin kl E2
= (propagation matrix) (4.1.13)
{E+ (z), E− (z), Γ(z)} do not. H1 jη−1 sin kl cos kl H2
Eqs. (4.1.1) and (4.1.2) relate the field quantities at location z to the quantities at
Writing η = η0 /n, where n is the refractive index of the propagation medium,
z = 0. In matching problems, it proves more convenient to be able to relate these
Eq. (4.1.13) can written in following form, which is useful in analyzing multilayer struc-
quantities at two arbitrary locations.
tures and is common in the thin-film literature [181,183,187,198]:
Fig. 4.1.1 depicts the quantities {E(z), H(z), E+ (z), E− (z), Z(z), Γ(z)} at the two
    
locations z1 and z2 separated by a distance l = z2 − z1 . Using Eq. (4.1.2), we have for E1 cos δ jn−1 η0 sin δ E2
= (propagation matrix) (4.1.14)
the forward field at these two positions: H1 jnη0−1 sin δ cos δ H2
E2+ = E0+ e−jkz2 , E1+ = E0+ e−jkz1 = E0+ e−jk(z2 −l) = ejkl E2+ where δ is the propagation phase constant, δ = kl = k0 nl = 2π(nl)/λ0 , and nl the
optical length. Eqs. (4.1.13) and (4.1.5), imply for the propagation of the wave impedance:

E2
cos kl + jη sin kl
E1 E2 cos kl + jηH2 sin kl H2
Z1 = = −
=η
H1 jE2 η sin kl + H2 cos kl
1 E2
η cos kl + j sin kl
H2

which gives:

Z2 cos kl + jη sin kl
Z1 = η (impedance propagation) (4.1.15)
η cos kl + jZ2 sin kl
Fig. 4.1.1 Field quantities propagated between two positions in space.
It can also be written in the form:
−jkl
And similarly, E1− = e E2− . Thus, Z2 + jη tan kl
Z1 = η (impedance propagation) (4.1.16)
E1+ = e jkl
E2+ , −jkl
E1− = e E2− (4.1.10) η + jZ2 tan kl

and in matrix form: A useful way of expressing Z1 is in terms of the reflection coefficient Γ2 . Using (4.1.7)
     and (4.1.12), we have:
E1+ ejkl 0 E2+
= (propagation matrix) (4.1.11)
E1− 0 e−jkl E2− 1 + Γ1 1 + Γ2 e−2jkl
Z1 = η =η or,
1 − Γ1 1 − Γ2 e−2jkl
We will refer to this as the propagation matrix for the forward and backward fields.
It follows that the reflection coefficients will be related by:
1 + Γ2 e−2jkl
Z1 = η (4.1.17)
E1− E2− e−jkl 1 − Γ2 e−2jkl
Γ1 = = = Γ2 e−2jkl , or,
E1+ E2+ ejkl We mention finally two special propagation cases: the half-wavelength and the quarter-
wavelength cases. When the propagation distance is l = λ/2, or any integral multiple
Γ1 = Γ2 e−2jkl (reflection coefficient propagation) (4.1.12) thereof, the wave impedance and reflection coefficient remain unchanged. Indeed, we
Using the matrix relationships (4.1.4) and (4.1.11), we may also express the total have in this case kl = 2πl/λ = 2π/2 = π and 2kl = 2π. It follows from Eq. (4.1.12)
electric and magnetic fields E1 , H1 at position z1 in terms of E2 , H2 at position z2 : that Γ1 = Γ2 and hence Z1 = Z2 .
         If on the other hand l = λ/4, or any odd integral multiple thereof, then kl = 2π/4 =
E1 1 E1+ 1 1 1 ejkl 0 E2+ π/2 and 2kl = π. The reflection coefficient changes sign and the wave impedance
= −1 =−1 −1 −1 −jkl
H1 η −η E1− η −η 0 e E2− inverts:
    
1 1 1 ejkl 0 1 η E2 1 + Γ1 1 − Γ2 1 η2
=
η−1
−η −1
0 e−jkl 1 −η H2 Γ1 = Γ2 e−2jkl = Γ2 e−jπ = −Γ2 ⇒ Z1 = η =η =η =
2 1 − Γ1 1 + Γ2 Z2 /η Z2
4.2. Matching Matrices 85 86 Electromagnetic Waves & Antennas – S. J. Orfanidis

Thus, we have in the two cases:     


E+ 1 1 ρ E+
 = (matching matrix) (4.2.4)
λ E− τ ρ 1 E−
l= ⇒ Z1 = Z2 , Γ1 = Γ2
2
(4.1.18) where {ρ, τ} and {ρ , τ } are the elementary reflection and transmission coefficients
λ η2
l= ⇒ Z1 = , Γ1 = −Γ2 from the left and from the right of the interface, defined in terms of η, η as follows:
4 Z2
η − η 2η
ρ= , τ= (4.2.5)
4.2 Matching Matrices η + η η + η

Next, we discuss the matching conditions across dielectric interfaces. We consider a η − η 2η

ρ = , τ = (4.2.6)
planar interface (taken to be the xy-plane at some location z) separating two dielec- η + η η + η
tric/conducting media with (possibly complex-valued) characteristic impedances η, η ,
Writing η = η0 /n and η = η0 /n , we have in terms of the refractive indices:
as shown in Fig. 4.2.1.†

n − n 2n
ρ= , τ=
n + n n + n
(4.2.7)
n − n
  2n
ρ =  , τ =
n +n n + n

These are also called the Fresnel coefficients. We note various useful relationships:

τ = 1 + ρ, ρ = −ρ, τ = 1 + ρ = 1 − ρ, ττ = 1 − ρ2 (4.2.8)

In summary, the total electric and magnetic fields E, H match simply across the
Fig. 4.2.1 Fields across an interface. interface, whereas the forward/backward fields E± are related by the matching matrices
of Eqs. (4.2.3) and (4.2.4). An immediate consequence of Eq. (4.2.1) is that the wave
Because the normally incident fields are tangential to the interface plane, the bound- impedance is continuous across the interface:
ary conditions require that the total electric and magnetic fields be continuous across
E E
the two sides of the interface: Z= =  = Z
H H
E = E On the other hand, the corresponding reflection coefficients Γ = E− /E+ and Γ =
(continuity across interface) (4.2.1)
H = H 
E− 
/E+ match in a more complicated way. Using Eq. (4.1.7) and the continuity of the
wave impedance, we have:
In terms of the forward and backward electric fields, Eq. (4.2.1) reads:
1+Γ 1 + Γ

E+ + E− = E+ 
+ E− η = Z = Z = η
1−Γ 1 − Γ
1  1  
 (4.2.2)
which can be solved to get:
E+ − E− =  E+ − E−
η η
ρ + Γ ρ + Γ
Eq. (4.2.2) may be written in a matrix form relating the fields E± on the left of the Γ= and Γ =

interface to the fields E± on the right: 1 + ρΓ 1 + ρ Γ
     The same relationship follows also from Eq. (4.2.3):

E+ 1 1 ρ E+
=  (matching matrix) (4.2.3)
E− τ ρ 1 E− 
1 
E−

(ρE+ + E− ) ρ+
E− 
E+ ρ + Γ
and inversely: Γ= = τ =  =
E+ 1   E 1 + ρΓ
† The arrows in this figure indicate the directions of propagation, not the direction of the fields—the field (E + ρE− ) 1+ρ −
τ + E+
vectors are perpendicular to the propagation directions and parallel to the interface plane.
4.2. Matching Matrices 87 88 Electromagnetic Waves & Antennas – S. J. Orfanidis

To summarize, we have the matching conditions for Z and Γ: 4.3 Reflected and Transmitted Power

ρ + Γ ρ + Γ For waves propagating in the z-direction, the time-averaged Poynting vector has only a
Z = Z
Γ=
Γ = (4.2.9)
1 + ρΓ 1 + ρ Γ z-component:
1   1
Two special cases, illustrated in Fig. 4.2.1, are when there is only an incident wave P = Re x̂ E × ŷ H∗ = ẑ Re(EH∗ )
2 2

on the interface from the left, so that E− = 0, and when the incident wave is only from
A direct consequence of the continuity equations (4.2.1) is that the Poynting vector
the right, so that E+ = 0. In the first case, we have Γ = E−  
/E+ = 0, which implies
    
is conserved across the interface. Indeed, we have:
Z = η (1 + Γ )/(1 − Γ )= η . The matching conditions give then:
1 1
ρ + Γ P=
∗
Re(EH∗ )= Re(E H )= P (4.3.1)
Z = Z = η , Γ= =ρ 2 2
1 + ρΓ
In particular, consider the case of a wave incident from a lossless dielectric η onto a
The matching matrix (4.2.3) implies in this case:
lossy dielectric η . Then, the conservation equation (4.3.1) reads in terms of the forward
       
E+ 1 1 ρ 
E+ 1 E+ and backward fields (assuming E− = 0):
= = 
E− τ ρ 1 0 τ ρE+
1    1   2
 P= |E+ |2 − |E− |2 = Re |E+ | = P
Expressing the reflected and transmitted fields E− , E+ in terms of the incident field E+ , 2η 2η
we have:
The left hand-side is the difference of the incident and the reflected power and rep-
resents the amount of power transmitted into the lossy dielectric per unit area. We saw
E− = ρE+
 (left-incident fields) (4.2.10) in Sec. 2.6 that this power is completely dissipated into heat inside the lossy dielectric
E+ = τE+
(assuming it is infinite to the right.) Using Eqs. (4.2.10), we find:
This justifies the terms reflection and transmission coefficients for ρ and τ. In the
1   1 
right-incident case, the condition E+ = 0 implies for Eq. (4.2.4): P= |E+ |2 1 − |ρ|2 )= Re |E+ |2 |τ|2 (4.3.2)
       2η 2η
  
E+ 1 1 ρ 0 1 ρ E−
 = = This equality requires that:
E− τ ρ 1 E− τ E−
1 1
These can be rewritten in the form: (1 − |ρ|2 )= Re  |τ|2 (4.3.3)
η η

E+ 
= ρ E− This can be proved using the definitions (4.2.5). Indeed, we have:
(right-incident fields) (4.2.11)
E− = τ E−
  
η 1−ρ η 1 − |ρ|2 1 − |ρ|2
  = ⇒ Re = =
which relates the reflected and transmitted fields E+
to the incident field, E− In this E− . η 1+ρ η |1 + ρ|2 |τ|2
case Γ = E− /E+ = ∞ and the third of Eqs. (4.2.9) gives Γ = E−  
/E+ = 1/ρ , which is 

consistent with Eq. (4.2.11). which is equivalent to Eq. (4.3.3), if η is lossless (i.e., real.) Defining the incident, re-
 flected, and transmitted powers by
When there are incident fields both from both sides, that is, E+ , E− , we may invoke
the linearity of Maxwell’s equations and add the two right-hand sides of Eqs. (4.2.10)
1

and (4.2.11) to obtain the outgoing fields E+ , E− in terms of the incident ones: Pin = |E+ |2
2η

E+ = τE+ + ρ E−
 1 1
(4.2.12) Pref = |E− |2 = |E+ |2 |ρ|2 = Pin |ρ|2
E− = ρE+ + τ E−
 2η 2η
 1   2  1  η
This gives the scattering matrix relating the outgoing fields to the incoming ones: Ptr = Re |E+ | = Re |E+ |2 |τ|2 = Pin Re  |τ|2
2η 2η η
    

E+ τ ρ E+
= (scattering matrix) (4.2.13) Then, Eq. (4.3.2) reads Ptr = Pin − Pref . The power reflection and transmission
E− ρ τ 
E−
coefficients, also known as the reflectance and transmittance, give the percentage of the
Using the relationships Eq. (4.2.8), it is easily verified that Eq. (4.2.13) is equivalent incident power that gets reflected and transmitted:
to the matching matrix equations (4.2.3) and (4.2.4).
4.3. Reflected and Transmitted Power 89 90 Electromagnetic Waves & Antennas – S. J. Orfanidis

Solution: For a good conductor, we have ω0 /σ  1. It follows from Eq. (2.8.4) that Rs /η0 =
Pref Ptr η  n  2 ω0 /2σ  1. From Eq. (2.8.2), the conductor’s characteristic impedance is ηc = Rs (1 +
= |ρ|2 , = 1 − |ρ|2 = Re  |τ|2 = Re |τ| (4.3.4)
Pin Pin η n j). Thus, the quantity ηc /η0 = (1 + j)Rs /η0 is also small. The reflection and transmission
If both dielectrics are lossless, then ρ, τ are real-valued. In this case, if there are coefficients ρ, τ can be expressed to first-order in the quantity ηc /η0 as follows:
incident waves from both sides of the interface, it is straightforward to show that the
2η c 2η c 2η c
net power moving towards the z-direction is the same at either side of the interface: τ=  , ρ = τ − 1  −1 +
ηc + η0 η0 η0
1   1   2  2

P= |E+ |2 − |E− |2 = |E+ | − |E− | = P (4.3.5) Similarly, the power transmission coefficient can be approximated as
2η 2η
This follows from the matrix identity satisfied by the matching matrix of Eq. (4.2.3): 2 Re(ηc ) 4R s
1 − |ρ|2 = 1 − |τ − 1|2 = 1 − 1 − |τ|2 + 2 Re(τ) 2 Re(τ)= 2 =
      η0 η0
1 1 ρ 1 0 1 ρ η 1 0
=  (4.3.6)
τ2 ρ 1 0 −1 ρ 1 η 0 −1 where we neglected |τ|2 as it is second order in ηc /η0 . For copper at 1 GHz, we have
ω0 /2σ = 2.19×10−5 , which gives Rs = η0 ω0 /2σ = 377×2.19×10−5 = 0.0082 Ω. It
If ρ, τ are real, then we have with the help of this identity and Eq. (4.2.3):
follows that 1 − |ρ|2 = 4R2 /η0 = 8.76×10−5 .
  
1   1
∗ ∗ 1 0 E+ This represents only a small power loss of 8.76×10−3 percent and the sheet acts as very
P= |E+ |2 − |E− |2 = E ,E
2η 2η + − 0 −1 E− good mirror at microwave frequencies.
     On the other hand, at optical frequencies, e.g., f = 600 THz corresponding to green
1
 ∗  ∗ 1 1 ρ∗ 1 0 1 ρ 
E+
= E , E− light with λ = 500 nm, the exact equations (2.6.5) yield the value for the character-
2η + ττ∗ ∗
ρ 1 0 −1 ρ 1 
E− istic impedance of the sheet ηc = 6.3924 + 6.3888i Ω and the reflection coefficient
   ρ = −0.9661 + 0.0328i. The corresponding power loss is 1 − |ρ|2 = 0.065, or 6.5 percent.
1 η
 ∗  ∗ 1 0 
E+ 1   2  2

= E , E− = |E+ | − |E− | = P Thus, metallic mirrors are fairly lossy at optical frequencies. 

2η η + 0 −1 
E− 2η
Example 4.3.3: A uniform plane wave of frequency f is normally incident from air onto a thick
Example 4.3.1: Glasses have a refractive index of the order of n = 1.5 and dielectric constant
conductor with conductivity σ , and  = 0 , µ = µ0 . Determine the reflected and trans-
 = n2 0 = 2.250 . Calculate the percentages of reflected and transmitted powers for
mitted electric and magnetic fields to first-order in ηc /η0 and in the limit of a perfect
visible light incident on a planar glass interface from air.
conductor (ηc = 0).
Solution: The characteristic impedance of glass will be η = η0 /n. Therefore, the reflection and
Solution: Using the approximations for ρ and τ of the previous example and Eq. (4.2.10), we
transmission coefficients can be expressed directly in terms of n, as follows:
have for the reflected, transmitted, and total electric fields at the interface:

η − η0 n−1 − 1 1−n 2  
ρ= = −1 = , τ=1+ρ= 2η c
η + η0 n +1 1+n 1+n E− = ρE+ = −1 + E+
η0
For n = 1.5, we find ρ = −0.2 and τ = 0.8. It follows that the power reflection and 2ηc

transmission coefficients will be E+ = τE+ = E+
η0
2η c 
|ρ|2 = 0.04, 1 − |ρ|2 = 0.96 E = E+ + E − = E+ = E + = E
η0
That is, 4% of the incident power is reflected and 96% transmitted. 

For a perfect conductor, we have σ → ∞ and ηc /η0 → 0. The corresponding total tangen-
Example 4.3.2: A uniform plane wave of frequency f is normally incident from air onto a thick tial electric field becomes zero E = E = 0, and ρ = −1, τ = 0. For the magnetic fields, we
conducting sheet with conductivity σ , and  = 0 , µ = µ0 . Show that the proportion need to develop similar first-order approximations. The incident magnetic field intensity
of power transmitted into the conductor (and then dissipated into heat) is given approxi- is H+ = E+ /η0 . The reflected field becomes to first order:
mately by  
1 1 2ηc
H− = − E− = − ρE+ = −ρH+ = 1− H+
Ptr 4R s 8ω0 η0 η0 η0
= =
Pin η0 σ
Similarly, the transmitted field is
Calculate this quantity for f = 1 GHz and copper σ = 5.8×107 Siemens/m.
4.4. Single Dielectric Slab 91 92 Electromagnetic Waves & Antennas – S. J. Orfanidis

Let ρ1 , ρ2 be the elementary reflection coefficients from the left sides of the two
  interfaces, and let τ1 , τ2 be the corresponding transmission coefficients:
 1  1 η0 η 0 2η c 2η0 ηc
H+ = E+ = τE+ = τH+ = H+ = H+  2 1 − H+
ηc ηc ηc ηc ηc + η0 ηc + η 0 η0 η1 − ηa ηb − η1
ρ1 = , ρ2 = , τ1 = 1 + ρ1 , τ2 = 1 + ρ2 (4.4.1)
η1 + ηa ηb + η1
The total tangential field at the interface will be:
To determine the reflection coefficient Γ1 into medium ηa , we apply Eq. (4.2.9) to
 
ηc relate Γ1 to the reflection coefficient Γ1 at the right-side of the first interface. Then, we

H = H+ + H − = 2 1 − H+ = H+ = H propagate to the left of the second interface with Eq. (4.1.12) to get:
η0

In the perfect conductor limit, we find H = H = 2H+ . As we saw in Sec. 2.6, the fields just ρ1 + Γ1 ρ1 + Γ2 e−2jk1 l1
Γ1 =  = (4.4.2)

inside the conductor, E+ 
, H+ , will attenuate while they propagate. Assuming the interface 1 + ρ1 Γ1 1 + ρ1 Γ2 e−2jk1 l1
is at z = 0, we have:
At the second interface, we apply Eq. (4.2.9) again to relate Γ2 to Γ2 . Because there
  −αz −jβz   −αz −jβz
are no backward-moving waves in medium ηb , we have Γ2 = 0. Thus,
E+ (z)= E+ e e , H+ (z)= H+ e e
ρ2 + Γ2
where α = β = (1 − j)/δ, and δ is the skin depth δ = ωµσ/2. We saw in Sec. 2.6 that Γ2 = = ρ2
1 + ρ2 Γ2
the effective surface current is equal in magnitude to the magnetic field at z = 0, that is,

Js = H + . Because of the boundary condition H = H = H+ 
, we obtain the result Js = H, We finally find for Γ1 :
or vectorially, Js = H × ẑ = n̂ × H, where n̂ = −ẑ is the outward normal to the conductor.
This result provides a justification of the boundary condition Js = n̂ × H at an interface ρ1 + ρ2 e−2jk1 l1
Γ1 = (4.4.3)
with a perfect conductor. 
 1 + ρ1 ρ2 e−2jk1 l1

This expression can be thought of as function of frequency. Assuming a lossless

4.4 Single Dielectric Slab medium η1 , we have 2k1 l1 = ω(2l1 /c1 )= ωT, where T = 2l1 /c1 = 2(n1 l1 )/c0 is the
two-way travel time delay through medium η1 . Thus, we can write:
Multiple interface problems can be handled in a straightforward way with the help of
the matching and propagation matrices. For example, Fig. 4.4.1 shows a two-interface ρ1 + ρ2 e−jωT
Γ1 (ω)= (4.4.4)
problem with a dielectric slab η1 separating the semi-infinite media ηa and ηb . 1 + ρ1 ρ2 e−jωT

This can also be expressed as a z-transform. Denoting the two-way travel time delay
in the z-domain by z−1 = e−jωT = e−2jk1 l1 , we may rewrite Eq. (4.4.4) as the first-order
digital filter transfer function:

ρ1 + ρ2 z−1
Γ1 (z)= (4.4.5)
1 + ρ1 ρ2 z−1

An alternative way to derive Eq. (4.4.3) is working with wave impedances, which
are continuous across interfaces. The wave impedance at interface-2 is Z2 = Z2 , but
Z2 = ηb because there is no backward wave in medium ηb . Thus, Z2 = ηb . Using the
propagation equation for impedances, we find:

Z2 + jη1 tan k1 l1 ηb + jη1 tan k1 l1

Z1 = Z1 = η1 = η1
η1 + jZ2 tan k1 l1 η1 + jηb tan k1 l1
Fig. 4.4.1 Single dielectric slab.
Inserting this into Γ1 = (Z1 − ηa )/(Z1 + ηa ) gives Eq. (4.4.3). Working with wave
Let l1 be the width of the slab, k1 = ω/c1 the propagation wavenumber, and λ1 = impedances is always more convenient if the interfaces are positioned at half- or quarter-
2π/k1 the corresponding wavelength within the slab. We have λ1 = λ0 /n1 , where λ0 is wavelength spacings.
the free-space wavelength and n1 the refractive index of the slab. We assume the incident If we wish to determine the overall transmission response into medium ηb , that is,
field is from the left medium ηa , and thus, in medium ηb there is only a forward wave. the quantity T = E2 + /E1+ , then we must work with the matrix formulation. Starting at
4.4. Single Dielectric Slab 93 94 Electromagnetic Waves & Antennas – S. J. Orfanidis

the left interface and successively applying the matching and propagation matrices, we For lossless media, energy conservation states that the energy flux into medium η1
obtain: must equal the energy flux out of it. It is equivalent to the following relationship between
         Γ and T, which can proved using Eq. (4.4.6):
E1+ 1 1 ρ1 E1 + 1 1 ρ1 ejk1 l1 0 E2+
= =
E1− τ1 ρ1 1 E1 − τ1 ρ1 1 0 e−jk1 l1 E2− 1   1
1 − |Γ1 |2 = |T|2 (4.4.10)
      ηa ηb
1 1 ρ1 ejk1 l1 0 1 1 ρ2 E2 +
= Thus, if we call |Γ1 |2 the reflectance of the slab, representing the fraction of the
τ1 ρ1 1 0 e−jk1 l1 τ2 ρ2 1 0
incident power that gets reflected back into medium ηa , then the quantity
where we set E2 − = 0 by assumption. Multiplying the matrix factors out, we obtain:
ηa nb
1 − |Γ1 |2 = |T|2 = |T|2 (4.4.11)
ejk1 l1   ηb na
E1+ = 1 + ρ1 ρ2 e−2jk1 l1 E2 +
τ1 τ2 will be the transmittance of the slab, representing the fraction of the incident power that
ejk1 l1   gets transmitted through into the right medium ηb . The presence of the factors ηa , ηb
E1− = ρ1 + ρ2 e−2jk1 l1 E2 + can be can be understood as follows:
τ1 τ2
1
These may be solved for the reflection and transmission responses: |E |2
Ptransmitted 2ηb 2+ ηa
= = |T|2
E1− ρ1 + ρ2 e−2jk1 l1 Pincident 1 2 ηb
Γ1 = = |E1+ |
2ηa
E1+ 1 + ρ1 ρ2 e−2jk1 l1
(4.4.6)
E τ1 τ2 e−jk1 l1
T = 2+ = 4.5 Reflectionless Slab
E1+ 1 + ρ1 ρ2 e−2jk1 l1

The transmission response has an overall delay factor of e−jk1 l1 = e−jωT/2 , repre- The zeros of the transfer function (4.4.5) correspond to a reflectionless interface. Such
senting the one-way travel time delay through medium η1 . zeros can be realized exactly only in two special cases, that is, for slabs that have either
For convenience, we summarize the match-and-propagate equations relating the field half-wavelength or quarter-wavelength thickness. It is evident from Eq. (4.4.5) that a
quantities at the left of interface-1 to those at the left of interface-2. The forward and zero will occur if ρ1 + ρ2 z−1 = 0, which gives the condition:
backward electric fields are related by the transfer matrix: ρ2
      z = e2jk1 l1 = − (4.5.1)
jk1 l1 ρ1
E1+ 1 1 ρ1 e 0 E2+
=
E1− τ1 ρ1 1 0 e−jk1 l1 E2− Because the right-hand side is real-valued and the left-hand side has unit magnitude,
     (4.4.7) this condition can be satisfied only in the following two cases:
E 1+ 1 ejk1 l1 ρ1 e−jk1 l1 E2+
= z = e2jk1 l1 = 1, ρ2 = −ρ1 , (half-wavelength thickness)
E1− τ1 ρ1 ejk1 l1 e−jk1 l1 E2−
z=e 2jk1 l1
= −1, ρ2 = ρ1 , (quarter-wavelength thickness)
The reflection responses are related by Eq. (4.4.2):
The first case requires that 2k1 l1 be an integral multiple of 2π, that is, 2k1 l1 = 2mπ,
ρ1 + Γ2 e−2jk1 l1
Γ1 = (4.4.8) where m is an integer. This gives the half-wavelength condition l1 = mλ1 /2, where λ1
1 + ρ1 Γ2 e−2jk1 l1
is the wavelength in medium-1. In addition, the condition ρ2 = −ρ1 requires that:
The total electric and magnetic fields at the two interfaces are continuous across the
ηb − η1 ηa − η1
interfaces and are related by Eq. (4.1.13): = ρ2 = −ρ1 =
ηa = ηb
ηb + η1 ηa + η1
    
E1 cos k1 l1 jη1 sin k1 l1 E2 that is, the media to the left and right of the slab must be the same. The second pos-
= (4.4.9)
H1 jη1−1 sin k1 l1 cos k1 l1 H2 sibility requires e2jk1 l1 = −1, or that 2k1 l1 be an odd multiple of π, that is, 2k1 l1 =
Eqs. (4.4.7)–(4.4.9) are valid in general, regardless of what is to the right of the second (2m + 1)π, which translates into the quarter-wavelength condition l1 = (2m + 1)λ1 /4.
interface. There could be a semi-infinite uniform medium or any combination of multiple Furthermore, the condition ρ2 = ρ1 requires:
slabs. These equations were simplified in the single-slab case because we assumed that ηb − η1 η1 − ηa
there was a uniform medium to the right and that there were no backward-moving waves. = ρ2 = ρ1 =
η21 = ηa ηb
ηb + η1 η1 + ηa
4.5. Reflectionless Slab 95 96 Electromagnetic Waves & Antennas – S. J. Orfanidis

To summarize, a reflectionless slab, Γ1 = 0, can be realized only in the two cases:

λ1 4ρ21
half-wave: l1 = m , η1 arbitrary, ηa = ηb |Γ1 |2max =
2 (4.5.2) (1 + ρ21 )2
λ1 √
quarter-wave: l1 = (2m + 1) , η1 = ηa ηb , ηa , ηb arbitrary
4 Fig. 4.5.1 shows the magnitude responses for the three values of the reflection co-
efficient: |ρ1 | = 0.9, 0.7, and 0.5. The closer ρ1 is to unity, the narrower are the reflec-
An equivalent way of stating these conditions is to say that the optical length of
tionless notches.
the slab must be a half or quarter of the free-space wavelength λ0 . Indeed, if n1 is the
refractive index of the slab, then its optical length is n1 l1 , and in the half-wavelength
case we have n1 l1 = n1 mλ1 /2 = mλ0 /2, where we used λ1 = λ0 /n1 . Similarly, we have
n1 l1 = (2m + 1)λ0 /4 in the quarter-wavelength case. In terms of the refractive indices,
Eq. (4.5.2) reads:

λ0
half-wave: n1 l1 = m , n1 arbitrary, na = nb
2 (4.5.3)
λ0 √
quarter-wave: n1 l1 = (2m + 1) , n1 = na nb , na , nb arbitrary
4

The reflectionless matching condition can also be derived by working with wave
impedances. For half-wavelength spacing, we have from Eq. (4.1.18) Z1 = Z2 = ηb . The
condition Γ1 = 0 requires Z1 = ηa , thus, matching occurs if ηa = ηb . Similarly, for the
quarter-wavelength case, we have Z1 = η21 /Z2 = η21 /ηb = ηa .
Fig. 4.5.1 Reflection responses |Γ(ω)|2 . (a) |ρ1 | = 0.9, (b) |ρ1 | = 0.7, (c) |ρ1 | = 0.5.
We emphasize that the reflectionless response Γ1 = 0 is obtained only at certain slab
widths (half- or quarter-wavelength), or equivalently, at certain operating frequencies.
It is evident from these figures that for the same value of ρ1 , the half- and quarter-
These operating frequencies correspond to ωT = 2mπ, or, ωT = (2m + 1)π, that is,
wavelength cases have the same notch widths. A standard measure for the width is the
ω = 2mπ/T = mω0 , or, ω = (2m + 1)ω0 /2, where we defined ω0 = 2π/T.
3-dB width, which for thge half-wavelength case is twice the 3-dB frequency ω3 , that
The dependence on l1 or ω can be seen from Eq. (4.4.5). For the half-wavelength
is, ∆ω = 2ω3 , as shown in Fig. 4.5.1 for the case |ρ1 | = 0.5. The frequency ω3 is
case, we substitute ρ2 = −ρ1 and for the quarter-wavelength case, ρ2 = ρ1 . Then, the
determined by the 3-dB half-power condition:
reflection transfer functions become:
1
ρ1 (1 − z−1 ) |Γ1 (ω3 )|2 = |Γ1 |2max
Γ1 (z) = , (half-wave) 2
1 − ρ21 z−1
(4.5.4) or, equivalently:
ρ1 (1 + z−1 )
Γ1 (z) = , (quarter-wave)
1 + ρ21 z−1 2ρ21 (1 − cos ω3 T) 1 4ρ21
=
where z = e2jk1 l1 = ejωT . The magnitude-square responses then take the form: 1− 2ρ21 cos ω3 T + ρ41 2 (1 + ρ21 )2

Solving for the quantity cos ω3 T = cos(∆ωT/2), we find:

 
2ρ21 1 − cos(2k1 l1 ) 2ρ21 (1 − cos ωT)
|Γ1 |2 = = , (half-wave)  ∆ωT  2ρ21  ∆ωT  1 − ρ21
1 − 2ρ21 cos(2k1 l1 )+ρ41 1 − 2ρ21 cos ωT + ρ41 cos =
tan = (4.5.6)
(4.5.5) 2 1+ ρ41 4 1 + ρ21
 
2ρ21 1 + cos(2k1 l1 ) 2ρ21 (1 + cos ωT)
|Γ1 |2 = = , (quarter-wave) If ρ21 is very near unity, then 1 − ρ21 and ∆ω become small, and we may use the
1 + 2ρ21 cos(2k1 l1 )+ρ41 1 + 2ρ21 cos ωT + ρ41
approximation tan x  x to get:
These expressions are periodic in l1 with period λ1 /2, and periodic in ω with period
∆ωT 1 − ρ21 1 − ρ21
ω0 = 2π/T. In DSP language, the slab acts as a digital filter with sampling frequency  
4 1+ ρ21 2
ω0 . The maximum reflectivity occurs at z = −1 and z = 1 for the half- and quarter-
wavelength cases. The maximum squared responses are in either case: which gives the approximation:
4.5. Reflectionless Slab 97 98 Electromagnetic Waves & Antennas – S. J. Orfanidis

Example 4.5.1: Determine the reflection coefficients of half- and quarter-wave slabs that do not
∆ωT = 2(1 − ρ21 ) (4.5.7) necessarily satisfy the impedance conditions of Eq. (4.5.2).

This is a standard approximation for digital filters relating the 3-dB width of a pole Solution: The reflection response is given in general by Eq. (4.4.6). For the half-wavelength case,
peak to the radius of the pole [52]. For any desired value of the bandwidth ∆ω, Eq. (4.5.6) we have e2jk1 l1 = 1 and we obtain:
or (4.5.7) may be thought of as a design condition that determines ρ1 .
η1 − ηa ηb − η1
Fig. 4.5.2 shows the corresponding transmittances 1 − |Γ1 (ω)|2 of the slabs. The +
ρ1 + ρ 2 η1 + η a ηb + η 1 ηb − ηa na − nb
transmission response acts as a periodic bandpass filter. This is the simplest exam- Γ1 = = η1 − ηa ηb − η1 = ηb + ηa = na + nb
1 + ρ 1 ρ2 1+
ple of a so-called Fabry-Perot interference filter or Fabry-Perot resonator. Such filters η1 + ηa ηb + η1
find application in the spectroscopic analysis of materials. We discuss them further in
Chap. 5. This is the same as if the slab were absent. For this reason, half-wavelength slabs are
sometimes referred to as absentee layers. Similarly, in the quarter-wavelength case, we
have e2jk1 l1 = −1 and find:

ρ1 − ρ 2 η21 − ηa ηb na nb − n21
Γ1 = = =
1 − ρ 1 ρ2 η21 + ηa ηb na nb + n21

The slab becomes reflectionless if the conditions (4.5.2) are satisfied. 



Example 4.5.2: Antireflection Coating. Determine the refractive index of a quarter-wave antire-
flection coating on a glass substrate with index 1.5.

Solution: From Eq. (4.5.3), we have with na = 1 and nb = 1.5:

√ √
n1 = na nb = 1.5 = 1.22

Fig. 4.5.2 Transmittance of half- and quarter-wavelength dielectric slab. The closest refractive index that can be obtained is that of cryolite (Na3 AlF6 ) with n1 =
1.35 and magnesium fluoride (MgF2 ) with n1 = 1.38. Magnesium fluoride is usually pre-
Using Eq. (4.5.5), we may express the frequency response of the half-wavelength ferred because of its durability. Such a slab will have a reflection coefficient as given by
the previous example:
transmittance filter in the following equivalent forms:

(1 − ρ21 )2 1 ρ1 − ρ2 η2 − η a η b na nb − n21 1.5 − 1.382

1 − |Γ1 (ω)|2 = = (4.5.8) Γ1 = = 12 = 2 = = −0.118
1 − ρ1 ρ2 η1 + η a ηb na nb + n 1 1.5 + 1.382
1 − 2ρ21 cos ωT + ρ41 1 + F sin2 (ωT/2)
where the F is called the finesse in the Fabry-Perot context and is defined by: with reflectance |Γ|2 = 0.014, or 1.4 percent. This is to be compared to the 4 percent
reflectance of uncoated glass that we determined in Example 4.3.1.
2ρ21
F= Fig. 4.5.3 shows the reflectance |Γ(λ)|2 as a function of the free-space wavelength λ. The
(1 − ρ21 )2
reflectance remains less than one or two percent in the two cases, over almost the entire
visible spectrum.
The finesse is a measure of the peak width, with larger values of F corresponding
to narrower peaks. The connection of F to the 3-dB width (4.5.6) is easily found to be: The slabs were designed to have quarter-wavelength thickness at λ0 = 550 nm, that is, the
optical length was n1 l1 = λ0 /4, resulting in l1 = 112.71 nm and 99.64 nm in the two cases
 ∆ωT  1 − ρ21 1 of n1 = 1.22 and n1 = 1.38. Such extremely thin dielectric films are fabricated by means
tan = = (4.5.9) of a thermal evaporation process [181,183].
4 1 + ρ21 2+F
Quarter-wavelength slabs may be used to design anti-reflection coatings for lenses, The MATLAB code used to generate this example was as follows:

so that all incident light on a lens gets through. Half-wavelength slabs, which require that
n = [1, 1.22, 1.50]; L = 1/4; refractive indices and optical length
the medium be the same on either side of the slab, may be used in designing radar domes
lambda = linspace(400,700,101) / 550; visible spectrum wavelengths
(radomes) protecting microwave antennas, so that the radiated signal from the antenna Gamma1 = multidiel(n, L, lambda); reflection response of slab
goes through the radome wall without getting reflected back towards the antenna.
4.5. Reflectionless Slab 99 100 Electromagnetic Waves & Antennas – S. J. Orfanidis

Antireflection Coating on Glass where we used ρ1 = (1 − n)/(1 + n). This explains why glass windows do not exhibit a
5
frequency-selective behavior as predicted by Eq. (4.5.5). For n = 1.5, we find 1 − |Γ1 |2 =
nglass = 1.50 0.9231, that is, 92.31% of the incident light is transmitted through the plate.
4

| Γ1 (λ)|2 (percent)
The same expressions for the average reflectance and transmittance can be obtained by
n1 = 1.22
3 n1 = 1.38 summing incoherently all the multiple reflections within the slab, that is, summing the
uncoated glass multiple reflections of power instead of field amplitudes. The timing diagram for such
multiple reflections is shown in Fig. 4.6.1.
2
Indeed, if we denote by pr = ρ21 and pt = 1 − pr = 1 − ρ21 , the power reflection and trans-
1 mission coefficients, then the first reflection of power will be pr . The power transmitted
through the left interface will be pt and through the second interface p2t (assuming the
0
same medium to the right.) The reflected power at the second interface will be pt pr and
400 450 500 550 600 650 700
will come back and transmit through the left interface giving p2t pr .
λ (nm)
Similarly, after a second round trip, the reflected power will be pt2 p3r , while the transmitted
Fig. 4.5.3 Reflectance over the visible spectrum. power to the right of the second interface will be pt2 p2r , and so on. Summing up all the
reflected powers to the left and those transmitted to the right, we find:

The syntax and use of the function multidiel is discussed in Sec. 5.1. The dependence pt2 pr 2pr
|Γ1 |2 = pr + p2t pr + p2t p3r + p2t p5r + · · · = pr + =
of Γ on λ comes through the quantity k1 l1 = 2π(n1 l1 )/λ. Since n1 l1 = λ0 /4, we have 1 − p2r 1 + pr
k1 l1 = 0.5πλ0 /λ. 

p2t 1 − pr
1 − |Γ1 |2 = pt2 + pt2 p2r + p2t p4r + · · · = =
Example 4.5.3: Thick Glasses. Interference phenomena, such as those arising from the mul- 1 − p2r 1 + pr
tiple reflections within a slab, are not observed if the slabs are “thick” (compared to the
where we used pt = 1 − pr . These are equivalent to Eqs. (4.5.10). 

wavelength.) For example, typical glass windows seem perfectly transparent.
If one had a glass plate of thickness, say, of l = 1.5 mm and index n = 1.5, it would have Example 4.5.4: Radomes. A radome protecting a microwave transmitter has  = 40 and is
optical length nl = 1.5×1.5 = 2.25 mm = 225×104 nm. At an operating wavelength designed as a half-wavelength reflectionless slab at the operating frequency of 10 GHz.
of λ0 = 450 nm, the glass plate would act as a half-wave transparent slab with nl = Determine its thickness.
104 (λ0 /2), that is, 104 half-wavelengths long.
Next, suppose that the operating frequency is 1% off its nominal value of 10 GHz. Calculate
Such plate would be very difficult to construct as it would require that l be built with the percentage of reflected power back towards the transmitting antenna.
an accuracy of a few percent of λ0 /2. For example, assuming n(∆l)= 0.01(λ0 /2), the
Determine the operating bandwidth as that frequency interval about the 10 GHz operating
plate should be constructed with an accuracy of one part in a million: ∆l/l = n∆l/(nl)=
frequency within which the reflected power remains at least 30 dB below the incident
0.01/104 = 10−6 . (That is why thin films are constructed by a carefully controlled evapo-
power.
ration process.)
More realistically, a typical glass plate can be constructed with an accuracy of one part in a Solution: The free-space wavelength is λ0 = c0 /f0 = 30 GHz cm/10 GHz = 3 cm. The refractive
thousand, ∆l/l = 10−3 , which would mean that within the manufacturing uncertainty ∆l, index of the slab is n = 2 and the wavelength inside it, λ1 = λ0 /n = 3/2 = 1.5 cm. Thus,
there would still be ten half-wavelengths, n∆λ = 10−3 (nl)= 10(λ0 /2). the slab thickness will be the half-wavelength l1 = λ1 /2 = 0.75 cm, or any other integral
multiple of this.
The overall power reflection response will be obtained by averaging |Γ1 |2 over several λ0 /2
cycles, such as the above ten. Because of periodicity, the average of |Γ1 |2 over several cycles Assume now that the operating frequency is ω = ω0 + δω, where ω0 = 2πf0 = 2π/T.
is the same as the average over one cycle, that is, Denoting δ = δω/ω0 , we can write ω = ω0 (1 + δ). The numerical value of δ is very
small, δ = 1% = 0.01. Therefore, we can do a first-order calculation in δ. The reflection
1 ω0 coefficient ρ1 and reflection response Γ are:
|Γ1 |2 = |Γ1 (ω)|2 dω
ω0 0
η − η0 0 .5 − 1 1 ρ1 (1 − z−1 ) ρ1 (1 − e−jωT )
ρ1 = = =− , Γ1 (ω)= 2 −1 =
where ω0 = 2π/T and T is the two-way travel-time delay. Using either of the two expres- η + η0 0 .5 + 1 3 1 − ρ1 z 1 − ρ21 e−jωT
sions in Eq. (4.5.5), this integral can be done exactly resulting in the average reflectance
and transmittance: where we used η = η0 /n = η0 /2. Noting that ωT = ω0 T(1 + δ)= 2π(1 + δ), we can
expand the delay exponential to first-order in δ:
2ρ21 1 − ρ21 2n
|Γ1 |2 = , 1 − |Γ1 |2 = = (4.5.10)
1+ ρ21 1 + ρ21 n2 + 1 z−1 = e−jωT = e−2πj(1+δ) = e−2πj e−2πjδ = e−2πjδ  1 − 2πjδ
4.5. Reflectionless Slab 101 102 Electromagnetic Waves & Antennas – S. J. Orfanidis

Thus, the reflection response becomes to first-order in δ: 4.6 Time-Domain Reflection Response
 
ρ1 1 − (1 − 2πjδ) ρ1 2πjδ ρ1 2πjδ We conclude our discussion of the single slab by trying to understand its behavior in
Γ1  = 
1 − ρ21 (1 − 2πjδ) 1 − ρ21 + ρ21 2πjδ 1 − ρ21 the time domain. The z-domain reflection transfer function of Eq. (4.4.5) incorporates
where we replaced the denominator by its zeroth-order approximation because the numer- the effect of all multiple reflections that are set up within the slab as the wave bounces
ator is already first-order in δ. It follows that the power reflection response will be: back and forth at the left and right interfaces. Expanding Eq. (4.4.5) in a partial fraction
expansion and then in power series in z−1 gives:
ρ21 (2πδ)2
|Γ1 |2 = ρ1 + ρ2 z−1 1 1 (1 − ρ21 )
∞
(1 − ρ21 )2
Γ1 (z)= −
= − = ρ1 + (1 − ρ21 )(−ρ1 )n−1 ρn2 z−n
1 + ρ1 ρ2 z 1 ρ1 ρ1 1 + ρ1 ρ2 z−1 n=1
Evaluating this expression for δ = 0.01 and ρ1 = −1/3, we find |Γ|2 = 0.00049, or
0.049 percent of the incident power gets reflected. Next, we find the frequency about Using the reflection coefficient from the right of the first interface, ρ1 = −ρ1 , and the
ω0 at which the reflected power is A = 30 dB below the incident power. Writing again, transmission coefficients τ1 = 1 + ρ1 and τ1 = 1 + ρ1 = 1 − ρ1 , we have τ1 τ1 = 1 − ρ21 .
ω = ω0 + δω = ω0 (1 + δ) and assuming δ is small, we have the condition: Then, the above power series can be written as a function of frequency in the form:
ρ21 (2πδ)2 Prefl 1 − ρ21 ∞ ∞
|Γ1 |2 = = = 10−A/10 ⇒ δ= 10−A/20 Γ1 (ω)= ρ1 + τ1 τ1 (ρ1 )n−1 ρn2 z−n = ρ1 + τ1 τ1 (ρ1 )n−1 ρn2 e−jωnT
(1 − ρ21 )2 Pinc 2π|ρ1 |
n=1 n=1
Evaluating this expression, we find δ = 0.0134, or δω = 0.0134ω0 . The bandwidth will
where we set z−1 = e−jωT . It follows that the time-domain reflection impulse response,
be twice that, ∆ω = 2δω = 0.0268ω0 , or in Hz, ∆f = 0.0268f0 = 268 MHz. 

that is, the inverse Fourier transform of Γ1 (ω), will be the sum of discrete impulses:
Example 4.5.5: Because of manufacturing imperfections, suppose that the actual constructed
∞
thickness of the above radome is 1% off the desired half-wavelength thickness. Determine
the percentage of reflected power in this case.
Γ1 (t)= ρ1 δ(t)+ τ1 τ1 (ρ1 )n−1 ρn2 δ(t − nT) (4.6.1)
n=1
Solution: This is essentially the same as the previous example. Indeed, the quantity θ = ωT = This is the response of the slab to a forward-moving impulse striking the left inter-
2k1 l1 = 2ωl1 /c1 can change either because of ω or because of l1 . A simultaneous in- face at t = 0, that is, the response to the input E1+ (t)= δ(t). The first term ρ1 δ(t) is the
finitesimal change (about the nominal value θ0 = ω0 T = 2π) will give: impulse immediately reflected at t = 0 with the reflection coefficient ρ1 . The remaining
δθ δω δl1 terms represent the multiple reflections within the slab. Fig. 4.6.1 is a timing diagram
δθ = 2(δω)l1 /c1 + 2ω0 (δl1 )/c1 ⇒ δ= = + that traces the reflected and transmitted impulses at the first and second interfaces.
θ0 ω0 l1

In the previous example, we varied ω while keeping l1 constant. Here, we vary l1 , while
keeping ω constant, so that δ = δl1 /l1 . Thus, we have δθ = θ0 δ = 2πδ. The correspond-
ing delay factor becomes approximately z−1 = e−jθ = e−j(2π+δθ) = 1 − jδθ = 1 − 2πjδ.
The resulting expression for the power reflection response is identical to the above and its
numerical value is the same if δ = 0.01. 


Example 4.5.6: Because of weather conditions, suppose that the characteristic impedance of
the medium outside the above radome is 1% off the impedance inside. Calculate the per-
centage of reflected power in this case.

Solution: Suppose that the outside impedance changes to ηb = η0 + δη. The wave impedance
at the outer interface will be Z2 = ηb = η0 + δη. Because the slab length is still a half-
wavelength, the wave impedance at the inner interface will be Z1 = Z2 = η0 + δη. It
follows that the reflection response will be:

Z1 − η 0 η0 + δη − η0 δη δη
Γ1 = = = 
Z1 + η 0 η0 + δη + η0 2η0 + δη 2η 0
Fig. 4.6.1 Multiple reflections building up the reflection and transmission responses.
where we replaced the denominator by its zeroth-order approximation in δη. Evaluating
at δη/η0 = 1% = 0.01, we find Γ1 = 0.005, which leads to a reflected power of |Γ1 |2 = The input pulse δ(t) gets transmitted to the inside of the left interface and picks up
2.5×10−5 , or, 0.0025 percent. 
 a transmission coefficient factor τ1 . In T/2 seconds this pulse strikes the right interface
4.6. Time-Domain Reflection Response 103 104 Electromagnetic Waves & Antennas – S. J. Orfanidis

and causes a reflected wave whose amplitude is changed by the reflection coefficient ρ2 For a causal waveform E1+ (t), the summation over n will be finite, such that at each
into τ1 ρ2 . time t ≥ 0 only the terms that have t − nT ≥ 0 will be present. In a similar fashion, we
Thus, the pulse τ1 ρ2 δ(t − T/2) gets reflected backwards and will arrive at the left find for the overall transmitted response into medium ηb :
interface T/2 seconds later, that is, at time t = T. A proportion τ1 of it will be transmit-
−∞ ∞
ted through to the left, and a proportion ρ1 will be re-reflected towards the right. Thus,
E2 + (t)= T(t )E1+ (t − t )dt = τ1 τ2 (ρ1 )n ρn2 E1+ (t − nT − T/2) (4.6.5)
at time t = T, the reflected pulse into the left medium will be τ1 τ1 ρ2 δ(t − T), and the −∞ n=0
re- reflected pulse τ1 ρ1 ρ2 δ(t − T).
The re-reflected pulse will travel forward to the right interface, arriving there at time We will use similar techniques later on to determine the transient responses of trans-
t = 3T/2 getting reflected backwards picking up a factor ρ2 . This will arrive at the left mission lines.
at time t = 2T. The part transmitted to the left will be now τ1 τ1 ρ1 ρ22 δ(t − 2T), and
the part re-reflected to the right τ1 ρ1 2 ρ22 δ(t − 2T). And so on, after the nth round trip,
the pulse transmitted to the left will be τ1 τ1 (ρ1 )n−1 ρn
4.7 Two Dielectric Slabs
2 δ(t − nT). The sum of all the
reflected pulses will be Γ1 (t) of Eq. (4.6.1).
Next, we consider more than two interfaces. As we mentioned in the previous section,
In a similar way, we can derive the overall transmission response to the right. It is
Eqs. (4.4.7)–(4.4.9) are general and can be applied to all successive interfaces. Fig. 4.7.1
seen in the figure that the transmitted pulse at time t = nT+(T/2) will be τ1 τ2 (ρ1 )n ρn 2.
shows three interfaces separating four media. The overall reflection response can be
Thus, the overall transmission impulse response will be:
calculated by successive application of Eq. (4.4.8):
∞
T(t)= τ1 τ2 (ρ1 )n ρn2 δ(t − nT − T/2) ρ1 + Γ2 e−2jk1 l1 ρ2 + Γ3 e−2jk2 l2
Γ1 = , Γ2 =
n=0 1 + ρ1 Γ2 e−2jk1 l1 1 + ρ2 Γ3 e−2jk2 l2

It follows that its Fourier transform will be:

∞
T(ω)= τ1 τ2 (ρ1 )n ρn2 e−jnωT e−jωT/2
n=0

which sums up to Eq. (4.4.6):

τ1 τ2 e−jωT/2 τ1 τ2 e−jωT/2
T(ω)= = (4.6.2)
1 − ρ1 ρ2 e−jωT 1 + ρ1 ρ2 e−jωT
For an incident field E1+ (t) with arbitrary time dependence, the overall reflection
response of the slab is obtained by convolving the impulse response Γ1 (t) with E1+ (t).
This follows from the linear superposition of the reflection responses of all the frequency
components of E1+ (t), that is,
∞ ∞
dω dω Fig. 4.7.1 Two dielectric slabs.
E1− (t)= Γ1 (ω)E1+ (ω)ejωt , where E1+ (t)= E1+ (ω)ejωt
−∞ 2π −∞ 2π
If there is no backward-moving wave in the right-most medium, then Γ3 = 0, which
Then, the convolution theorem of Fourier transforms implies that:
implies Γ3 = ρ3 . Substituting Γ2 into Γ1 and denoting z1 = e2jk1 l1 , z2 = e2jk2 l2 , we
∞ −∞ eventually find:
dω
E1− (t)= Γ1 (ω)E1+ (ω)ejωt = Γ1 (t )E1+ (t − t )dt (4.6.3)
−∞ 2π −∞
ρ1 + ρ2 z1−1 + ρ1 ρ2 ρ3 z2−1 + ρ3 z1−1 z2−1
Inserting (4.6.1), we find that the reflected wave arises from the multiple reflections Γ1 = (4.7.1)
1 + ρ1 ρ2 z1−1 + ρ2 ρ3 z2−1 + ρ1 ρ3 z1−1 z2−1
of E1+ (t) as it travels and bounces back and forth between the two interfaces:
The reflection response Γ1 can alternatively be determined from the knowledge of
∞
the wave impedance Z1 = E1 /H1 at interface-1:
E1− (t)= ρ1 E1+ (t)+ τ1 τ1 (ρ1 )n−1 ρn2 E1+ (t − nT) (4.6.4)
n=1 Z1 − ηa
Γ1 =
Z1 + ηa
4.7. Two Dielectric Slabs 105 106 Electromagnetic Waves & Antennas – S. J. Orfanidis

The fields E1 , H1 are obtained by successively applying Eq. (4.4.9): The frequency dependence of Eq. (4.7.1) arises through the factors z1 , z2 , which can
     be written in the forms: z1 = ejωT1 and z2 = ejωT2 , where T1 = 2l1 /c1 and T2 = 2l2 /c2
E1 cos k1 l1 jη1 sin k1 l1 E2
= are the two-way travel time delays through the two slabs.
H1 jη1−1 sin k1 l1 cos k1 l1 H2
A case of particular interest arises when the slabs are designed to have the equal
   
cos k1 l1 jη1 sin k1 l1 cos k2 l2 jη2 sin k2 l2 E3 travel-time delays so that T1 = T2 ≡ T. Then, defining a common variable z = z1 =
= z2 = ejωT , we can write the reflection response as a second-order digital filter transfer
jη1−1 sin k1 l1 cos k1 l1 jη2−1 sin k2 l2 cos k2 l2 H3
function:
But at interface-3, E3 = E3 = E3 + and H3 = Z3−1 E3 = η− 1 
b E3+ , because Z3 = ηb .
Therefore, we can obtain the fields E1 , H1 by the matrix multiplication: ρ1 + ρ2 (1 + ρ1 ρ3 )z−1 + ρ3 z−2
Γ1 (z)= (4.7.2)
      1 + ρ2 (ρ1 + ρ3 )z−1 + ρ1 ρ3 z−2
E1 cos k1 l1 jη1 sin k1 l1 cos k2 l2 jη2 sin k2 l2 1
= E3 + In the next chapter, we discuss further the properties of such higher-order reflection
H1 jη1−1 sin k1 l1 cos k1 l1 jη2−1 sin k2 l2 cos k2 l2 η−
b
1
transfer functions arising from multilayer dielectric slabs.
Because Z1 is the ratio of E1 and H1 , the factor E3 + cancels out and can be set equal
to unity.
4.8 Problems
Example 4.7.1: Determine Γ1 if both slabs are quarter-wavelength slabs. Repeat if both slabs
are half-wavelength and when one is half- and the other quarter-wavelength. 4.1 Fill in the details of the equivalence between Eq. (4.2.2) and (4.2.3), that is,


E+ + E − = E + 
+ E−     
Solution: Because l1 = λ1 /4 and l2 = λ2 /4, we have 2k1 l1 = 2k2 l2 = π, and it follows that E+ 1 1 ρ 
E+
z1 = z2 = −1. Then, Eq. (4.7.1) becomes: 1  1  

= 
E + − E − =  E + − E− E− τ ρ 1 E−
η η
ρ1 − ρ2 − ρ1 ρ2 ρ3 + ρ3
Γ1 =
1 − ρ1 ρ2 − ρ2 ρ3 + ρ1 ρ3 4.2 Fill in the details of the equivalences stated in Eq. (4.2.9), that is,

A simpler approach is to work with wave impedances. Using Z3 = ηb , we have: ρ + Γ ρ + Γ

Z = Z
Γ=
Γ =
1 + ρΓ 1 + ρ Γ
η21 η2 η2 η2
Z1 = = 2 1 = 12 Z3 = 21 ηb Show that if there is no left-incident field from the right, then Γ = ρ, and if there is no
Z2 η2 /Z3 η2 η2
right-incident field from the left, then, Γ = 1/ρ . Explain the asymmetry of the two cases.
Inserting this into Γ1 = (Z1 − ηa )/(Z1 + ηa ), we obtain: 4.3 Let ρ, τ be the reflection and transmission coefficients from the left side of an interface and
let ρ , τ be those from the right, as defined in Eq. (4.2.5). One of the two media may be
η21 ηb − η22 ηa
Γ1 = lossy, and therefore, its characteristic impedance and hence ρ, τ may be complex-valued.
η21 ηb + η22 ηa Show and interpret the relationships:

The two expressions for Γ1 are equivalent. The input impedance Z1 can also be obtained η 2
1 − |ρ|2 = Re |τ| = Re(τ∗ τ )
by matrix multiplication. Because k1 l1 = k2 l2 = π/2, we have cos k1 l1 = 0 and sin k1 l1 = 1 η
and the propagation matrices for E1 , H1 take the simplified form:
4.4 Show that the reflection and transmission responses of the single dielectric slab of Fig. 4.4.1
       
E1 0 jη1 0 jη2 1  −η1 η− 2
1

are given by Eq. (4.4.6), that is,
= E3 + = E3+
H1 jη−
1
1
0 jη−
2
1
0 ηb−1 −η2 η−
1
1 −1
η b
ρ1 + ρ2 e−2jk1 l1 E2 + τ1 τ2 e−jk1 l1
Γ= , T= =
The ratio E1 /H1 gives the same answer for Z1 as above. When both slabs are half-wavelength, 1 + ρ1 ρ2 e−2jk1 l1 E1+ 1 + ρ1 ρ2 e−2jk1 l1
the impedances propagate unchanged: Z1 = Z2 = Z3 , but Z3 = ηb .
Moreover, using these expressions show and interpret the relationship:
If η1 is half- and η2 quarter-wavelength, then, Z1 = Z2 = η22 /Z3 = η22 /ηb . And, if the
quarter-wavelength is first and the half-wavelength second, Z1 = η21 /Z2 = η21 /Z3 = η21 /ηb . 1   1
1 − |Γ|2 = |T|2
The corresponding reflection coefficient Γ1 is in the three cases: ηa ηb

ηb − η a η22 − ηa ηb η21 − ηa ηb 4.5 A 1-GHz plane wave is incident normally onto a thick copper plate (σ = 5.8×107 S/m.) Can
Γ1 = , Γ1 = , Γ1 = the plate be considered to be a good conductor at this frequency? Calculate the percentage
ηb + η a η22 + ηa ηb η21 + ηa ηb
of the incident power that enters the plate. Calculate the attenuation coefficient within the
These expressions can also be derived by Eq. (4.7.1), or by the matrix method. 
 conductor and express it in units of dB/m. What is the penetration depth in mm?
4.8. Problems 107 108 Electromagnetic Waves & Antennas – S. J. Orfanidis

4.6 With the help of Fig. 4.5.1, argue that the 3-dB width ∆ω is related to the 3-dB frequency 4.11 An alternative way of representing the propagation relationship Eq. (4.1.12) is in terms of the
ω3 by ∆ω = 2ω3 and ∆ω = ω0 − 2ω3 , in the cases of half- and quarter-wavelength slabs. hyperbolic w-plane variable defined in terms of the reflection coefficient Γ, or equivalently,
Then, show that ω3 and ∆ω are given by: the wave impedance Z as follows:

2ρ21 ∆ωT 1 − ρ21 Γ = e−2w
Z = η coth(w) (4.8.1)
cos ω3 T = ± , tan =
1+ ρ41 4 1 + ρ21
−2w1 −2w2
Show the equivalence of these expressions. Writing Γ1 = e and Γ2 = e , show that
4.7 A fiberglass ( = 40 ) radome protecting a microwave antenna is designed as a half-wavelength Eq. (4.1.12) becomes equivalent to:
reflectionless slab at the operating frequency of 12 GHz.
w1 = w2 + jkl (propagation in w-domain) (4.8.2)
a. Determine three possible thicknesses (in cm) for this radome.
This form is essentially the mathematical (as opposed to graphical) version of the Smith
b. Determine the 15-dB and 30-dB bandwidths in GHz about the 12 GHz operating fre-
chart and is particularly useful for numerical computations using MATLAB.
quency , defined as the widths over which the reflected power is 15 or 30 dB below the
incident power.

4.8 A 5 GHz wave is normally incident from air onto a dielectric slab of thickness of 1 cm and
refractive index of 1.5, as shown below. The medium to the right of the slab has an index of
2.25.

a. Write an analytical expression of the reflectance |Γ(f )|2 as a function of frequency

and sketch it versus f over the interval 0 ≤ f ≤ 15 GHz. What is the value of the
reflectance at 5 GHz?
b. Next, the 1-cm slab is moved to the left by a distance of 3 cm, creating an air-gap
between it and the rightmost dielectric. Repeat all the questions of part (a).
c. Repeat part (a), if the slab thickness is 2 cm.

4.9 Consider a two-layer dielectric structure as shown in Fig. 4.7.1, and let na , n1 , n2 , nb be the
refractive indices of the four media. Consider the four cases: (a) both layers are quarter-
wave, (b) both layers are half-wave, (c) layer-1 is quarter- and layer-2 half-wave, and (d) layer-1
is half- and layer-2 quarter-wave. Show that the reflection coefficient at interface-1 is given
by the following expressions in the four cases:

na n22 − nb n21 na − n b na nb − n21 na nb − n22

Γ1 = , Γ1 = , Γ1 = , Γ1 =
na n22 + nb n21 na + n b na nb + n21 na nb + n22

4.10 Consider the lossless two-slab structure of Fig. 4.7.1. Write down all the transfer matrices
relating the fields Ei± , i = 1, 2, 3 at the left sides of the three interfaces. Then, show the
energy conservation equations:

1   1   1   1
|E1+ |2 − |E1− |2 = |E2+ |2 − |E2− |2 = |E3+ |2 − |E3− |2 = |E |2
ηa η1 η2 η b 1+