2.

1 Dielectric slab waveguide

a Consider the rays 1 and 2 in Figure 2.3. Derive the waveguide condition.
b Consider the two rays 1 and 2 in Figure 2.4. Show that the phase difference when
they meet at C at a distance y above the guide center is
m = k12(a y)cosm m
c Using the waveguide condition show that
y
m = m (y) = m (m + m )
a
Solution
n2
A A n2
B y C ay
1 1

2 2 /2
E 2a z E 2 a
k1 C x y
A k A
n1 1 y
2
2
Guide z
n2 B center x

Figure 2.3 Two arbitrary waves 1
and 2 that are initially in phase must
remain in phase after reflections.
Otherwise the two will interfere
destructively and cancel each other.
Figure 2.4 Interference of waves such as 1
and 2 leads to a standing wave pattern
along the y-direction which propagates
along z.

From the geometry we have the following:

(a y)/AC = cos
and C/AC = cos( 2)
The phase difference between the rays meeting at C is
= kAC kAC = k1AC k1ACcos( 2)
= k1AC[1 cos( 2)] = k1AC[1 + cos(2)]
= k1[(a y)/cos][ 1 + 2cos2 1]
= k1 [(a y)/cos][2cos2]
= 2k1(a y)cos
2 (2a)n1
Given, cos m m = m

 (m + m ) m + m
cos m = =
2 n1 (2a) k1 (2a)
m + m
Then m = 2k1 (a y) cos m m = 2k1 (a y) m
k1 (2a)
y y
m = (1 )(m + m ) m = m (m + m )
a a
y
m = m (y) = m (m + m )
a
2.7 Dielectric slab waveguide Consider a planar dielectric waveguide with a core
thickness 10 m, n1 = 1.4446, n2 = 1.4440. Calculate the V-number, the mode angle m
for m = 0 (use a graphical solution, if necessary), penetration depth, mode field distance
(MFD = 2 + 2), for light wavelengths of 1.0 m and 1.5 m. What is your conclusion?
Compare your MFD calculation with 2wo = 2a(V+1)/V.
Solution
= 1 m, n1 = 1.4446, n2 = 1.4440, a = 5 m. Apply
2a
(n n22 )
1/ 2
V= 2

to obtain V = 1.3079
Solve the waveguide condition
1/ 2
2 n2
2

sin m n

tan ak1 cos m m =
1
= f ( m )
2 cos m
graphically as in Example 2.1.1 to find: c = 88.35 and the mode angle (for m = 0) is o
= 88.85.
Then use
1/ 2
n 2 2
2 n2 1 sin m 1
1 n2
= m =
m
to calculate the penetration depth:
= 1/ = 5.33 m.
MFD = 2a + 2 = 20.65 m
We can also calculate MFD from
MFD = 2a(V+1)/V = 2(5 m)(1.3079+1)/(1.3079) = 17.6 m (Difference
= 15%)
 = 5 m, V = 0.262, single mode. Solve waveguide condition graphically to find that the
mode angle is o = 88.40.
= 1/ = 77.22 m.
MFD = 2a + 2 = 164.4 m.
Compare with MFD = 2a(V+1)/V = 2(5 m)(0.262 + 1)/(0.262) = 48.2 m (Very large
difference)
Notice that the MFD from 2a(V+1)/V gets worse as V decreases. The reason for using
MFD = 2a(V+1)/V, this equation provides a single step calculation of MFD. The
calculation of the penetration depth d requires the calculation of the incidence angle
and .
= 1.5 m, V = 0.872, single mode. Solve waveguide condition graphically that the
mode angle is o = 88.72.
= 1/ = 9.08 m.
MFD = 2a + 2 = 28.15 m.
Compare with MFD = 2a(V+1)/V = 2(5 m)(0.872+1)/(0.872) = 21.5 m (Difference =
24%)
Notice that the MFD from 2a(V+1)/V gets worse as V decreases. The reason for using
MFD = 2a(V+1)/V, this equation provides a single step calculation of MFD. The
calculation of the penetration depth requires the calculation of the incidence angle
and .

2.8 A multimode fiber Consider a multimode fiber with a core diameter of 100 m,
core refractive index of 1.475 and a cladding refractive index of 1.455 both at 850 nm.
Consider operating this fiber at = 850 nm.
a Calculate the V-number for the fiber and estimate the number of modes.
b Calculate the wavelength beyond which the fiber becomes single mode.
c Calculate the numerical aperture.
d Calculate the maximum acceptance angle.
e Calculate the modal dispersion and hence the bit rate distance product given
that rms dispersion 0.29 where is the full spread.

Solution
Given n1 = 1.475, n2 = 1.455, 2a = 10010-6 m or a = 50 m and = 0.850 m. The V-
number is,
2(50 m)(1.475 1.455 )
2 2 1 /2

(n )
2a 2 1/ 2
V= n =
2
= 89.47
1 2
(0.850 m)
 Number of modes M,
V 2 89.47 2
M= = 4002
2 2
The fiber becomes monomode when,

(n1 n22 ) < 2.405

2a 2 1/ 2
V=

2 a(n12 n22 ) 2 (50 m)(1.4752 1.4552 )
1/ 2 1/ 2

or > = = 31.6 m
2.405 2.405
For wavelengths longer than 31.6 m, the fiber is a single mode waveguide.
The numerical aperture NA is
NA = (n1 n2 ) = (1.475 1.455 )
2 2 1 /2 2 2 1/ 2
= 0.242

If max is the maximum acceptance angle, then,

NA
max = arcsin = arcsin(0.242/1) = 14
no
Modal dispersion is given by
intermode n1 n2 1.475 1.455
= =
L c 3 108 m s1
= 66.7 ps m-1 or 67.6 ns per km
Given that 0.29, maximum bit-rate is
0.25L 0.25L 0.25
BL = =
total intermode (0.29)(66.7 ns km 1 )
i.e. BL = 13 Mb s-1 km (only an estimate!)
We neglected material dispersion at this wavelength which would further decrease
BL. Material dispersion and modal dispersion must be combined by
total
2
= intermode
2
+ material
2

For example, assuming an LED with a spectral rms deviation of about 20 nm,
and a Dm 200 ps km-1 nm-1 (at about 850 nm)we would find
m = (200 ps km-1 nm-1)(20 nm)(1 km) 4000 ps km-1 or 4 ns km-1,
which is substantially smaller than the intermode dispersion and can be neglected.

2.9 A single mode fiber Consider a fiber with a SiO2-13.5%GeO2 core of diameter
of 8 m and refractive index of 1.468 and a cladding refractive index of 1.464 both
refractive indices at 1300 nm where the fiber is to be operated using a laser source with a
half maximum width of 2 nm.
a Calculate the V-number for the fiber. Is this a single mode fiber?
b Calculate the wavelength below which the fiber becomes multimode.
c Calculate the numerical aperture.
d Calculate the maximum acceptance angle.
e Obtain the material dispersion and waveguide dispersion and hence estimate the
bit rate distance product (BL) of the fiber.
Solution
a Given n1 = 1.475, n2 = 1.455, 2a = 810-6 m or a = 4 m and =1.3 m. The V-
number is,
2(4 m)(1.468 1.464 )
2 2 1/ 2

(n )
2a 2 1/ 2
V= n =
2
= 2.094
1 2
(1.3 m)
b Since V < 2.405, this is a single mode fiber. The fiber becomes multimode when

(
2a 2
n1 n22 ) > 2.405
1/ 2
V=

2 a(n12 n22 ) 2 (4 m)(1.468 2 1.4642 )
1/ 2 1/ 2

or < = =1.13 m
2.405 2.405
For wavelengths shorter than 1.13 m, the fiber is a multi-mode waveguide.
c The numerical aperture NA is
NA = (n1 n2 ) = (1.468 1.464 )
2 2 1 /2 2 2 1/ 2
= 0.108

d If max is the maximum acceptance angle, then,

NA
max = arcsin = arcsin(0.108/1) = 6.2
no

so that the total acceptance angle is 12.4.

e At =1.3 m, from the figure, Dm 7.5 ps km-1 nm-1, Dw 5 ps km-1 nm-1.
1/ 2
= Dm + Dw 1/ 2
L
= |7.55 ps km-1 nm-1|(2 nm) = 15 ps km-1 + 10 ps km-1
= 0.025 ns km-1
Obviously materials dispersion is 15 ps km-1 and waveguide dispersion is 10 ps
-1
km
The maximum bit-rate distance product is then
0.59L 0.59
BL = = 23.6 Gb s-1 km.
1 / 2 0.025 ns km 1
Dispersion coefficient (ps km-1 nm-1 )
20 Dm

10 SiO2-13.5%GeO2

0 Material and waveguide dispersion

Dw a (m) coefficients in an optical fiber with a
4.0 core SiO2-13.5%GeO2 for a = 2.5 to
3.5
10 3.0 4 m.
2.5

20
1.2 1.3 1.4 1.5 1.6
(m)
Figure 2Q9

2.10 A single mode fiber design According to Question 1.3 (Ch.1), the Sellmeier
dispersion equation provides n vs. for pure SiO2 and SiO2-13.5 mol.%GeO2. The
refractive index increases linearly with addition of GeO2 to SiO2 from 0 to 13.5 mol.%. A
single mode step index fiber for use at 1300 nm is required to have the following
properties: NA = 0.1, core diameter of 9 m and a core of SiO2-13.5% GeO2. What
should be the cladding composition?
Solution
Given
2a = 910-6 m or a = 4.5 m. From Ch 1, Question 1.3, the Sellmeier equation is,
G1 2 G2 2 G3 2
n2 1 = + +
2 21 2 22 2 23
where G1, G2, G3 and 1, 2 and 3 are constants given below where 1, 2, 3 are in m.
Sellmeier constants G1 G2 G3 1 2 3
SiO2-13.5%GeO2 0.711040 0.451885 0.704048 0.0642700 0.129408 9.425478

The fiber is to operate at =1.3 m, thus, using the Sellmeier equation above
with the constants in the table we find
n1 = 1.4682
The V-number is,
 2 a 2 (4.5 m)
V= NA = (0.1) = 2.175
(1.3 m)

NA = (n1 n2 ) (2n1 )
2 2 1 /2 2 1/ 2
Apply

0.1 [2(1.4682) ]
2 1 /2
or
to obtain = 0.002320
n1 n2 1.4681 n2
Apply = , i.e. 0.0232 =
n1 1.4681
Thus, the required cladding refractive index is
n2 = 1.4648
Pure silica has n = 1.4473, SiO2-13.5 mol.%GeO2 has n1 = 1.4682, by linear interpolation
the composition corresponding to n2 = 1.4648 is 11.3 mol.% GeO2. Note, the refractive
index n(x) of SiO2-x mol.%GeO2, assuming a linear relationship, can be written as
x x
n(x) = n(0) 1 + n(13.5)
13.5 13.5
where n(0) = 1.4473; n(13.5) = n1 = 1.4682. Substituting n(x) = 1.4648 gives x = 11.3 .

2.11 Material dispersion If Ng1 is the group refractive index of the core material of a
step fiber, then the propagation time (group delay time) of the fundamental mode is
L LN g1
= =
vg c
Since Ng will depend on the wavelength, show that the material dispersion
coefficient Dm is given approximately by
d d 2n
Dm =
Ld c d2
Using the Sellmeier equation in Question 1.3 in Chapter 1 evaluate material
dispersion at = 1.55 m for pure silica (SiO2) and SiO2-13.5%GeO2 glass.

Solution
From Ch. 1 we know that
dn
Ng n
d
Differentiate with respect to wavelength using the above relationship between
Ng and n.
 L LN g1
= =
vg c
d L dN g1 L dn d 2 n dn L d 2n
= 2 = 2
d c d c d d d c d
d d 2n
Thus, Dm = (1)
Ld c d2
From Ch. 1 we know that the Sellmeier equation is
G1 2 G2 2 G3 2
n2 1 = + + (2)
2 21 2 22 2 23
where G1, G2, G3 and 1, 2 and 3 are constants; given in Table 2Q11 (called Sellmeier
coefficients) that are determined by fitting this expression to the experimental data.
We can substitute Eq. (2) in Eq. (1) to obtain Dm and plot Dm vs. as shown in
Figure 2Q11. The substitution, differentiation and the plot were done on Mathview but
almost any other math-software package can do the same. Thus,
At = 1.55 m, Dm = 14 ps km-1 nm-1
Table 2Q11
The Sellmeier coefficients for SiO2-13.5%GeO2.
The 1, 2, 3 are in m.
G1 G2 G3 1 2 3
SiO2-13.5%GeO2 0.711040 0.451885 0.704048 0.0642700 0.129408 9.425478

Dm (ps km-1 nm-1)

15 ps km-1 nm-1

Wavelength (m) 1.55 m

Figure 2Q11
2.12 Waveguide dispersion Waveguide dispersion arises as a result of the
dependence of the propagation constant on the V-number which depends on the
wavelength. It is present even when the refractive index is constant; no material
dispersion. Let us suppose that n1 and n2 are wavelength (or k) independent. Suppose that
is the propagation constant of mode lm and k = 2/ where is the free space
wavelength. Then the normalized propagation constant b and propagation constant are
related by (see Example 2.3.4)
= n2k[1 + b] (1)
The group velocity is defined and given by
d dk
vg = =c
d d
Show that the propagation time, or the group delay time, of the mode is
L Ln2 Ln2 d( kb)
= = + (2)
vg c c dk
Given the definition of V,
V = ka[n1 n2 ] kan 2 (2)
2 2 1 /2 1/ 2
(3)
and
d (Vb )
dV
=
d
dV
[bkan2 (2 )1 / 2 ]= an2 (2 )1/ 2
d
dV
(bk ) (4)

Show that
d Ln d 2 (Vb)
= 2 V (5)
d c dV 2
and that the waveguide dispersion coefficient is
d n d 2 (Vb)
Dw = = 2 V (6)
Ld c dV 2
Figure 2Q12 shows the dependence of V[d2(Vb)/dV2] on the V-number. In the
range 1.5 < V < 2.4,
d 2 (Vb) 1.984
V
dV 2 V2
Show that,
n2 1.984 (n n ) 1.984
Dw = 1 2 (7)
c V 2
c V2
which simplifies to
1.984
Dw (8)
c (2a) 2 2 n2
 Equation (2) should really have Ng2 instead of n2 in which case Eq. (8) would be
1.984 Ng 2
Dw 2 (9)
c (2a) 2 n2
2

Consider a fiber with a core of diameter of 8 m and refractive index of 1.468 and
a cladding refractive index of 1.464 both refractive indices at 1300 nm. Suppose that a
1.3 m laser diode with a spectral linewidth of 2 nm is used to provide the input light
pulses. Estimate the waveguide dispersion per kilometer of fiber using Eqs. (6) and (9).

1.5

V[d2(Vb)/dV2]

0.5

0
0 1 2 3
V - number
[d2(Vb)/dV2] vs. V-number for a step index fiber.
Figure 2Q12

Solution
Waveguide dispersion arises as a result of the dependence of the propagation constant on
the V-number which depends on the wavelength. It is present even when the refractive
index is constant; no material dispersion. Let us suppose that n1 and n2 are wavelength (or
k) independent. Suppose that is the propagation constant of mode lm and k =
2/ where is the free space wavelength. Then the normalized propagation constant b
is defined as,
( / k)2 n22
b= (1)
n12 n22
Show that for small normalized index difference = (n1 n2)/n1, Eq. (1)
approximates to
( / k) n2
b= (2)
n1 n2
which gives as,
= n2k[1 + b] (3)
The group velocity is defined and given by
d dk
vg = =c
d d
Thus, the propagation time of the mode is
L L d Ln 2 Ln2 d(kb)
= = = + (4)
vg c dk c c dk

where we assumed constant (does not depend on the wavelength). Given the
definition of V,
V = ka[n12 n22 ]1 / 2 = ka[(n1 + n2 )(n1 n2 )]1/ 2
n n2 1/ 2
ka[(n1 + n2 )n1 1 ] (5)
n1
ka[2n2 n1 ]1/ 2 kan2 (2)1/ 2
From Eq. (5),
d(Vb)
dV
=
d
dV
[bkan2 (2)1 / 2 ]= an2 (2)1/ 2
d
dV
(bk)

This means that depends on V as,

Ln Ln d(Vb)
= 2+ 2 (6)
c c dV
Dispersion, that is, spread in due to a spread can be found by
differentiating Eq. (6) to obtain,
d Ln2 dV d d(Vb) Ln2 V d 2 (Vb)
= =
d c d dV dV c dV 2
(7)
Ln d 2 (Vb)
= 2 V
c dV 2
The waveguide dispersion coefficient is defined as
d n d 2 (Vb)
Dw = = 2 V (8)
Ld c dV 2
Figure 2Q5 shows the dependence of V[d2(Vb)/dV2] on the V-number.
In the range 2 < V < 2.4,
d 2 (Vb) 1.984
V
dV 2 V2
so that Eq. (8) becomes,
 n2 1.984 (n n ) 1.984
Dw = 1 2 (9)
c V 2
c V2
We can simplify this further by using
1/ 2
n 1.984 1.984 n2
Dw 2
c V 2
c 2 an (2 )
1 / 2
2

1.984
Dw (10)
c(2a) 2 2n2
Equation (6) should really have Ng2 instead of n2 in which case Eq. (10) would be
1.984 Ng 2
Dw 2 (11)
c (2a) 2 n2
2

Consider a fiber with a core of diameter of 8 m and refractive index of 1.468 and
a cladding refractive index of 1.464 both refractive indices at 1300 nm. Suppose that a1.3
m laser diode with a spectral linewidth of 2 nm is used to provide the input light pulses.
Estimate the waveguide dispersion per kilometer of fiber using Eqs. (8) and (11).

( )
1/ 2
2(4 m) 1.4682 1.4642
V=
2a

( 2
n1 n1 )
2 1/ 2
=
(1.3 m)
= 2.094

and = (n1 n2)/n1 = (1.468-1.464)/1.468 = 0.00273.

From the graph, Vd2(Vb)/dV2 = 0.45,
n2 d 2 (Vb ) (1.464)(2.73 10 3 )
Dw = V = 9 (0.45)
c dV
2
(3 10 m s )(1300 10 m)
8 -1

Dw - 4.610-6 s m-2 or -4.6 ps km-1 nm-1

Using Eq. (10)
1.984 1.984(1300 10 9 m)
Dw =
c(2 a) 2 2n2 (3 108 m s-1 )[2 4 10 6 m]2 2(1.464)]
Dw - 4.610-6 s m-2 or -4.6 ps km-1 nm-1
For 1/2 = 2 nm we have,
1/2 = |Dw|L 1/2 = (4.6 ps km-1 nm-1)(2 nm) = 9.2 ps/km

2.13 Profile dispersion Total dispersion in a single mode step index fiber is primarily
due to material dispersion and waveguide dispersion. However, there is an additional
dispersion mechanism called profile dispersion that arises from the propagation
constant of the fundamental mode also depending on the refractive index difference .
Consider a light source with a range of wavelengths coupled into a step index fiber.
We can view this as a change in the input wavelength . Suppose that n1, n 2, hence
depends on the wavelength . The propagation time, or the group delay time, g per unit
length is
1 1 d
g = = (1)
vg c dk
g g n1 g V g
= + + (2)
n1 V
Total dispersion = Materials dispersion (due to n1/)
+ Waveguide dispersion (due to V/)
+ Profile dispersion (due to /)
where the last term is due to depending on ; although small this is not zero. Even the
above statement in Eq. (2) is over simplified but nonetheless provides an insight into the
problem. The total intramode (chromatic) dispersion coefficient Dch is then given by
Dch = Dm + Dw + Dp (3)
where Dm, Dw, Dp are material, waveguide and profile dispersion coefficients
respectively. The waveguide dispersion is given by Eq. (8) in Question 2.6 and the
profile dispersion coefficient is (very) approximately,
N g1 d 2 (Vb) d
Dp V (4)
c dV 2 d
where b is the normalized propagation constant and Vd2(Vb)/dV2 vs. V is shown in Figure
2Q12. The term Vd2(Vb)/dV2 1.984/V2.
Consider a fiber with a core of diameter of 8 m. The refractive and group indices
of the core and cladding at = 1.55 m are n1 = 1.4504, n 2 = 1.4450, Ng1 = 1.4676, Ng 2
= 1.4625, and d/d = 161 m-1. Estimate the waveguide and profile dispersion per km of
fiber per nm of input light linewidth at this wavelength.
Solution
Total dispersion in a single mode step index fiber is primarily due to material dispersion
and waveguide dispersion. However, there is an additional dispersion mechanism called
profile dispersion that arises from the propagation constant of the fundamental mode
also depending on the refractive index difference . Consider a light source with a range
of wavelengths coupled into a step index fiber. We can view this as a change in the
input wavelength . Suppose that n1, n 2, hence depends on the wavelength . The
propagation time, or the group delay time, g per unit length is
1 1 d
g = = (1)
v g c dk
g g n1 g V g
= + + (2)
n1 V
 Total dispersion = Materials dispersion (due to n1/)
+ Waveguide dispersion (due to V/)
+ Profile dispersion (due to /)
where the last term is due depending on ; although small this is not zero. Even the
above statement in Eq. (2) is over simplified but nonetheless provides an sight into the
problem. The total intramode (chromatic) dispersion coefficient Dch is then given by
Dch = Dm + Dw + Dp (3)
where Dm, Dw, Dp are material, waveguide and profile dispersion coefficients
respectively. The waveguide dispersion is given by Eq. (8) in Question 2.6 and the
profile dispersion coefficient away is (very) approximately,
N g1 d 2 (Vb) d
Dp V (4)
c dV 2 d
where b is the normalized propagation constant and Vd2(Vb)/dV2 vs. V is shown in Figure
2Q6. The term Vd2(Vb)/dV2 1.984/V2.
Consider a fiber with a core of diameter of 8 m. The refractive and group indexes of the
core and cladding at = 1.55 m are n1 = 1.4504, n 2 = 1.4450, Ng1 = 1.4676, Ng 2 =
1.4625. d/d = 161 m-1.
2 (4 m)(1.4504 1.4450 )
2 2 1 /2
2a
V=

(n2
1 n
1 )
2 1/ 2
=
(1.55 m)
= 2.03

and = (n1 n2)/n1 = (1.4504-1.4450)/1.4504 = 0.00372

From the graph in Figure 2Q12, Vd2(Vb)/dV2 = 0.5,
Profile dispersion:
Ng1 d 2 (Vb) d
(0.5)(161 m 1 )
1.4676
Dp = V =
c dV d
2
3 108 m s 1
Dp = 3.8 10-7 s m-1 m-1 or 0.38 ps km-1 nm-1
Waveguide dispersion:
1.984 1.984(1500 10 9 m)
Dw =
c(2 a) 2 2n2 (3 108 m s-1 )[2 4 10 6 m]2 2(1.4450)]
Dw 5.6 ps km-1 nm-1
Profile dispersion is more than 10 times smaller than waveguide dispersion.

3.5 AlGaAs LED emitter An AlGaAs LED emitter for use in a local optical fiber
network has the output spectrum shown in Figure 3Q5. It is designed for peak emission at
820 nm at 25C.
a What is the linewidth between half power points at temperatures 40C, 25C
and 85C? What is the empirical relationship between and T given three temperatures
and how does this compare with (h) 2.5kBT 3kBT? B B

b Why does the peak emission wavelength increase with temperature?

c Why does the peak intensity decrease with temperature?
d What is the bandgap of AlGaAs in this LED?
e The bandgap, Eg, of the ternary alloys AlxGa1-xAs follows the empirical
expression,
Eg(eV) = 1.424 + 1.266x + 0.266x2.
What is the composition of the AlGaAs in this LED?
f When the forward current is 40 mA, the voltage across the LED is 1.5V and the
optical power that is coupled into a multimode fiber through a lens is 25 W. What is the
efficiency?
Relative spectral output power

40C
1
25C

The output spectrum from GaAlAs LED. Values

85C normalized to peak emission at 25C.

0
740 800 840 880 900
Wavelength (nm)

Figure 3Q5
Solution
a We note that the emitted wavelength is related to the photon energy Eph by
= c/ = hc/Eph.
If we differentiate with respect to photon energy Eph we get
d hc
= 2
dE ph E ph

We can represent small changes or intervals (or ) by differentials, e.g. /Eph

|d/dEph|, then
hc
2 E ph
E ph
We are given the energy width of the output spectrum, Eph = (h) 3kBT. B

Then, using the latter and substituting for Eph in terms of we find,
 3kB T 3k T
2 or 2 B
hc hc
Temperature 40 C 25 C 85 C Comment
Parameter
peak (nm) 804 820 837
(nm) (Measured) 30 40 48
(nm) (Calculated); Eph = 2.5 kT 26.2 34.8 43.6
(nm) (Calculated); Eph = 3 kT 31.4 41.7 52.3 Very close
70000 B
60000 B
50000
B
40000
/(2)
30000

20000 Best line forced through zero is

10000 /(2) = 1956T ; R2 = 0.9932374
0
0 100 200 300 400
Temperature (K)

The theory predicts that /2 vs T should be a straight line because,

Eph
2
hc
mkB T
so that =
2 hc
where Eph = mkBT and m is a numerical constant that represents the ratio (h)/(kBT)
B B

and is determined from the slope of the /2 vs. T plot.

The three points plotted in the figure seems to follow this behavior. The best line
forced through zero has a slope that indicates m = 2.8.
b The bandgap decreases with temperature
c There are two factors to consider. (i) Spectral intensity means intensity per unit
wavelength, that is, dI/d. The integration of the spectral curve gives the total intensity,
the total number of photons emitted per unit area per unit time. As the spectrum broadens
with temperature we would naturally expect the peak to decrease with temperature. (ii)
Higher the temperature, the stronger are the lattice vibrations (there are more phonons).
Indirect or radiationless transitions, those that do not emit photons, require phonons
(lattice vibrations) which encourage indirect transitions. Thus increasing the temperature
increases indirect transitions at the expense of direct transitions and the light intensity
decreases. If (ii) was totally absent then the areas under the curves for all the three
spectra would be identical.
d Use the peak emission wavelength to find Eg as follows:
At -40 C (233 K) , peak = 804 nm.
At 25 C (298 K), peak = 820 nm.
At 85 C (358 K), peak = 837 nm.
We first note that we need the required bandgap Eg at the wavelength of interest.
The photon energy at peak emission is hc/peak = Eg + kBT. Then,
B

ch kT
Eg = B
epeak e
and at peak = 820109 m, taking T = 25 + 273K,
(3 108 )(6.626 1034 )
Eg = 0.0257 eV = 1.4863 eV
(1.6 1019 )(820 109 )
e The bandgap Eg of the ternary alloys AlxGa1-xAs follows the empirical expression,
Eg(eV) = 1.424+1.266x+0.266x2.
Eg(eV) = 1.4863 = 1.424 + 1.266x + 0.266x2.
Solving for x we find, x = 0.05.
f From the definition of efficiency ,
Output optical power
=
Input electrical power
Po 25 10 6 W
= = = 0.000417 = 0.0417%
IV (40 10 3 A )(1.5 V)
3.6 III-V compound semiconductors in optoelectronics Figure 3Q6 represents the
bandgap Eg and the lattice parameter a in the quarternary III-V alloy system. A line
joining two points represents the changes in Eg and a with composition in a ternary alloy
composed of the compounds at the ends of that line. For example, starting at GaAs point,
Eg = 1.42 eV and a = 0.565 nm, and Eg decreases and a increases as GaAs is alloyed with
InAs and we move along the line joining GaAs to InAs. Eventually at InAs, Eg = 0.35 eV
and a = 0.606 nm. Point X in Figure 3Q6 is composed of InAs and GaAs and it is the
ternary alloy InxGa1-xAs. It has Eg = 0.7 eV and a = 0.587 nm which is the same a as that
for InP. InxGa1xAs at X is therefore lattice matched to InP and can hence be grown on an
InP substrate without creating defects at the interface.
Further, InxGa1-xAs at X can be alloyed with InP to obtain a quarternary alloy
InxGa1-xAsyP1-y whose properties lie on the line joining X and InP and therefore all have
the same lattice parameter as InP but different bandgap. Layers of InxGa1-xAsyP1-y with
composition between X and InP can be grown epitaxially on an InP substrate by various
techniques such as liquid phase epitaxy (LPE) or molecular beam expitaxy (MBE) .
The shaded area between the solid lines represents the possible values of Eg and a
for the quarternary III-V alloy system in which the bandgap is direct and hence suitable
for direct recombination.
The compositions of the quarternary alloy lattice matched to InP follow the line
from X to InP.
a Given that the InxGa1-xAs at X is In0.535Ga0.465As show that quarternary alloys
InxGa1-xAsyP1-y are lattice matched to InP when y = 2.15x.
b The bandgap energy Eg, in eV for InxGa1-xAsyP1-y lattice matched to InP is given
by the empirical relation,
Eg (eV) = 1.35 0.72y + 0.12 y2
Find the composition of the quarternary alloy suitable for an emitter operating at
1.55 m.

Eg (eV)
2.6 Quaternary alloys
with indirect bandgap Direct bandgap
2.4 GaP Indirect bandgap
2.2
2
Quaternary alloys
1.8
with direct bandgap
1.6
InP
1.4 GaAs
1.2
1 In1-xGa xAs
0.8
In0.535 Ga 0.465As
0.6 X InAs
0.4
0.2
0.54 0.55 0.56 0.57 0.58 0.59 0.6 0.61 0.62
Lattice constant, a (nm)

Bandgap energy E g and lattice constant a for various III-V alloys of GaP, GaAs,
InP and InAs. A line represents a ternary alloy formed with compounds from the
end points of the line. Solid lines are for direct bandgap alloys whereas dashed
lines for indirect bandgap alloys. Regions between lines represent quaternary
alloys. The line from X to InP represents quaternary alloys In1-xGaxAs1-yPy made
from In0.535Ga0.465As and InP which are lattice matched to InP.
Figure 3Q6
Solution
a The IIIV quaternary alloy is made of (In0.535Ga0.465As) alloyed with InP.
Suppose that z fraction of In0.535Ga0.465As is alloyed with (1 z) fraction of InP.

z = 0 for InP and z = 1 for In0.535Ga0.465As

The formula for the quaternary alloy is In1xGaxAsyP1y.
Thus, z(In0.535Ga0.465As) + (1z)InP = In1xGaxAsyP1y
Use x = x to represent point X or (In0.535Ga0.465As), that is, x = 0.535 for the In content.
We can now balance the contents for each element.
For P 1z=1y z=y
For Ga (1 x)z = x y = 2.15x
For In xz + (1 z) = 1 x z = x/(1 x')
For As z=y y = 2.15x
b Consider a lattice matched InGaAsP to InP. Given = 1.55 m.
Eg = 1.24/ = 0.80 eV
Thus, Eg (eV) = 1.35 0.72y + 0.12 y2 = 0.80 eV
0.12y2 0.72y + 0.55 = 0
y = 0.90, x = 0.42
In0.58Ga0.42As0.9P0.1
3.7 External conversion efficiency The external power or conversion efficiency ext
is defined as
Optical power output P
ext = = o
Electrical power input IV
One of the major factors reducing the external power efficiency is the loss of
photons in extracting the emitted photons which suffer reabsorption in the pn junction
materials, absorption outside the semiconductors and various reflections at interfaces.
The total light output power from a particular AlGaAs red LED is 2.5 mW when
the current is 50 mA and the voltage is 1.6 V. Calculate its external conversion
efficiency.
Solution
Po 2.5 10 3 W
ext = = 3 = 0.03125 = 3.125 %
IV (50 10 A)(1.6 V)
3.9 SLEDs and ELEDs Experiments carried out on an AlGaAs SLED (surface
emitting LED) and an ELED (edge emitting LED) give the light output power vs. current
data in Table 3Q9.
a Show that the output light power vs. current characteristics are not linear.
b By plotting the optical power output (Po) vs current (I) data on a log-log plot
show that Po In . Find n for each LED.
Table 3Q9
Light output power vs. DC current for surface and edge emitting LEDs.
SLED I (mA) 25 50 75 100 150 200 250 300
SLED Light output 1.04 2.07 3.1 4.06 5.8 7.6 9.0 10.2
power (mW) Po
ELED I (mA) 25 50 75 100 150 200 250 300
ELED Light output 0.46 0.88 1.28 1.66 2.32 2.87 3.39 3.84
power (mW) Po

Solution
Light Output Power (mW)
20
n = 0.92
SLED
10
ELED






n = 0.86

1

0.1
10 100 1000
Current (mA)

SLED
Po = (0.00556)I0.925
Correlation coefficient = 0.9979
ELED
Po = (0.0306)I0.857
Correlation coefficient = 0.9971

3.11 Internal quantum efficiency The internal efficiency int gauges what fraction
of electron hole recombinations in the forward biased pn junction are radiative and
therefore lead to photon emission. Nonradiative transitions are those in which an electron
and a hole recombine through a recombination center such as a crystal defect or an
impurity and emit phonons (lattice vibrations). By definition,
Rate of radiative recombination
int = (1)
Total rate of recombination (radiative and nonradiative)
1

or int = 1 r 1 (2)
+
r nr
where r is the mean lifetime of a minority carrier before it recombines radiatively and nr
is the mean lifetime before it recombines via a recombination center without emitting a
photon. The total current I is determined by the total rate of recombinations whereas the
number of photons emitted per second (ph) is determined by the rate of radiative
recombinations.
Photons emitted per second ph Pop(int) / h
int = = = (3)
Total carriers lost per second I / e I /e
where Pop(int) is the optical power generated internally (not yet extracted).
For a particular AlGaAs LED emitting at 850 nm it is found that r = 50 ns and nr
= 100 ns. What is the internal optical power generated at a current of 100 mA?
Solution
1 1
r 50 ns
Consider int = 1 1 = 1 1 = 0.667 = 66.7 %
+ +
r nr 50 ns 100 ns

Pop(int) / h
From int =
I /e
I Ihc (100 10 3 )(6.626 1034 )(3 108 )
Pop(int) = int ( )h = int = (0.667) 19 9
e e (1.6 10 )(850 10 )
= 0.097 W or 97 mW