p-n Junction and Transport in

Semiconductors
Lecture 8, OEN-630

Scope
•Mobility
•Current in Semiconductors
•p-n Junction
•Basic Characteristics of p-n Junction
•Manufacturing

CMR
Band Diagram in
V(x), PE (x)

Applied Field V(x)

Electron Energy PE (x) = – eV

Ec
EF

Ev E c − eV
E F − eV

E v − eV
A

B
n -Type Semiconductor

Energy band diagram of an n -type semiconductor connected to a

voltage supply of V volts. The whole energy diagram tilts because
the electron now has an electrostatic potential energy as well
© 1999 S.O. Kasap, Optoelectronics (Prentice Hall)
 Drift Velocity
In electric field appears additional velocity component:
 Mobility
Definition of mobility through the life time:

mn vn = FΔt = −eEτ C
eτ C
vn = − E = −μn E
mn
eτ C
vp = − E = −μ p E
mp
 Mobility and Scattering
e
The mobility through
μn ≡ τ C
mn
the mean free time
(average time between e
two collisions (C)): μp ≡ τC
mp

1 1 1 1
The overall
w= = + +K = ∑
probability of τC τ C ,lattice τ C ,imp i τ C ,i
collision is a sum of
the probabilities of 1 1 1
every single collision = +
mechanism: μ μL μI
 Mobility
Mobility of electrons and holes in Silicon at room temperature:
Electric Conductivity and Resistance

The electric current which flows in an intrinsic semiconductor

consists of both electron and hole currents
Electric Conductivity and Resistance
The conductivity of a semiconductor can be modeled in terms of the band
theory of solids. The band model of a semiconductor suggests that at
ordinary temperatures there is a finite possibility that electrons can reach
the conduction band and contribute to electrical conduction
Electric Conductivity and Resistance
r r r
In external electric field every
electron experiences a force from F = −eE = −∇V (r )
the field:
dEC dEi
Bottom of the conduction F =− =
band corresponds to the dx dx
potential energy

dV ( x )
Electric field
E≡−
dx
Electrostatic potential Ei = −eV ( x )
 Drift Current
The transport of carriers under the influence of an applied electric field
produces a current called the drift current:

n
J n = = ∑ (− evi ) = −envn = enμ n E
In
A i =1
J n = enμ n E
J p = epμ p E
J = Jn + J p
Total current is the sum of electron and hole current
components
Electric Conductivity and Resistance

Total current is determined through the geometry of the conductor

Conductivity and Resistivity
The total drift current is the sum of electron and hole
current components:

J = J n + J p = e(nμ n + pμ p )E = σE
σ = e(nμ n + pμ p )
1 1
ρ≡ =
σ e(nμ n + pμ p )
Temperature as an Energy

Average kinetic energy of the electrons:

1 3
KE = mn vth = kT
2

2 2
 Thermal Velocity
1 3
KE = mn vth = kT ,
2

2 2

~ 105 ms −1 (T = 300 K )
3kT
v~ = v =
2

m
L 4 ×10 m (2500mi )
6
t~~ ≈ = 40 sec Estimated cost-to-cost travel
v 105 ms −1 time of electrons
Direct and Indirect Semiconductors

E-k Diagrams

CMR
 Potential Energy in Crystals
PE(r)
PE of the electron around an
r isolated atom

When N atoms are arranged to form

the crystal then there is an overlap
of individual electron PE functions.
V(x)
a a
0
PE of the electron, V(x), inside
the crystal is periodic with a
period a.
x
x=0 a 2a 3a x=L

Surface Crystal Surface

The electron potential energy (PE), V(x), inside the crystal is periodic with the same
periodicity as that of the crystal, a. Far away outside the crystal, by choice, V = 0 (the
electron is free and PE = 0).
© 1999 S.O. Kasap, Optoelectronics (Prentice Hall)
 Direct Bandgap Semiconductors
The E-k Diagram The Energy Band
Ek Diagram

CB
Conduction
Empty ψk e-
Band (CB) e-
Ec Ec
Eg hυ hυ

Valence
Ev Ev
Band (VB) h+ Occupied ψk h+

VB

k
–š /a š /a
The E-k diagram of a direct bandgap semiconductor such as GaAs. The E-k
curve consists of many discrete points with each point corresponding to a
possible state, wavefunction ψk(x), that is allowed to exist in the crystal.
The points are so close that we normally draw the E-k relationship as a
continuous curve. In the energy range Ev to Ec there are no points (ψk(x)
solutions).
© 1999 S.O. Kasap, Optoelectronics (Prentice Hall)
 Indirect Bandgap Semiconductors

E E E

CB
Indirect Bandgap, Eg
Ec CB
Direct Bandgap Eg Photon CB Ec Er Ec
Ev kcb Phonon
Ev Ev
VB
VB kvb VB
–k k –k k –k k
(a) GaAs (b) Si (c) Si with a recombination center

(a) In GaAs the minimum of the CB is directly above the maximum of the VB. GaAs is
therefore a direct bandgap semiconductor. (b) In Si, the minimum of the CB is displaced from
the maximum of the VB and Si is an indirect bandgap semiconductor. (c) Recombination of
an electron and a hole in Si involves a recombination center .
© 1999 S.O. Kasap, Optoelectronics (Prentice Hall)
 p-n Junction
Fabrication

(a) A bare n-type Si wafer.

An oxidized Si wafer by dry or
wet oxidation.
Application of resist.
Resist exposure through the
mask.
 p-n Junction
Fabrication

a. The wafer after the

development.
The wafer after SiO2
removal.
The final result after a
complete lithography
process.
A p-n junction is formed in
the diffusion or implantation
process.
The wafer after metalization.
A p-n junction after the
compete process.
 Thermal Equilibrium Conditions
Band Diagram

(a) Uniformly doped p-type and n-type semiconductors before the junction is formed. (b) The electric
field in the depletion region and the energy band diagram of a p-n junction in thermal equilibrium.

CMR
 Equilibrium Condition
p-type Semiconductor

J p = J p (drift ) + J p (diffusion )
dp
= eμ p pE − eD p 1 dEi
dx E= electric field
e dx
⎛ 1 dEi ⎞ dp
= eμ p p⎜ ⎟ − kTμ p =0 Dp =
kT
μ p Einstein relation
⎝ e dx ⎠ dx e

There is no electric current at steady-state condition

CMR
 Equilibrium Fermi Level

( Ei − F ) / kT dp p ⎛ dEi dF ⎞
p = ni e = ⎜ − ⎟
dx kT ⎝ dx dx ⎠
dF
J p = μp p =0
dx
dF
= 0, F = const
dx
Fermi energy is constant at steady-state condition

CMR
 p n

B- As+
h+
(a)
e–
P-n Junction M
Metallurgical Junction
M
Eo E (x)
Neutral p-region Neutral n-region –Wp 0 Wn
x
(e)
(b)
–Eo

V(x)
M Vo
Space charge region
Wp Wn (f)
log(n), log(p)
ppo
x
nno
PE(x)
ni (c)
eVo

pno Hole PE(x)

npo
x
x=0
ρnet x (g)
M

eNd Electron PE(x)

–Wp –eVo
x (d)
Wn

-eNa

Properties of the pn junction.

© 1999 S.O. Kasap, Optoelectronics (Prentice Hall)

 Abrupt p-n Junction

(a) A p-n junction with abrupt doping

changes at the metallurgical junction.
Energy band diagram of an abrupt
junction at thermal equilibrium.
Space charge distribution.
Rectangular approximation of the
space charge distribution.
 Abrupt p-n Junction
Neutral regions

d 2Ψ ρ
= − = − (N D − N A + p − n )
dE e
≡− Poisson’s equation
dx 2
dx εs εs
kT ⎛ N A ⎞
Ψ p ≡ − (Ei − F ) | x≤ − x p = −
1
ln⎜⎜ ⎟⎟ p-type neutral region:
e e ⎝ ni ⎠ Assume p>>n, ND=0
⎛ ⎞
Ψn ≡ −
1
(Ei − F ) |x≥ xn = − kT ln⎜⎜ N D ⎟⎟ n-type neutral region
e e ⎝ ni ⎠ Assume n>>p, NA=0

kT ⎛ N A N D ⎞
Vbi = Ψn − Ψ p = ln⎜⎜ 2
⎟⎟ Built-in potential
e ⎝ ni ⎠

CMR
Built-in Potentials Depend on Doping
Concentration

Built-in potentials on the p-

side and n-side of abrupt
junctions in Si and GaAs as
a function of impurity
concentration.

CMR
Abrupt p-n Junction
Space charge region

Electric field from Poisson’s equation

d 2 Ψ eN A
= , for − xp ≤ x < 0
dx 2 εs

dΨ eN A (x + x p )
E (x ) = − =− , for − xp ≤ x < 0
dx εs

d 2Ψ eN D
= − , for 0 < x ≤ x n
dx 2
εs
dΨ eN A (x − xn )
E ( x ) = − Em + = , for 0 < x ≤ xn
dx εs
eN D xn eN A x p
Em | x =0 = =
εs εs
CMR
 Abrupt p-n Junction
Space charge region

N A x p = N D xn Neutrality condition

NB = ND for p+-n junction

eN A x 2p eN D xn2 1
Vbi = + = EmW
2ε s 2ε s 2
⎛ x⎞
E (x ) = (− W + x ) = − Em ⎜1 − ⎟
eN B
εs ⎝ W⎠
⎛ x2 ⎞ V x⎛ x⎞
Ψ ( x ) = − ∫ Edx = Em ⎜⎜ x −
x
⎟⎟ = bi ⎜ 2 − ⎟
0
⎝ 2W ⎠ Ψ (0 )=0 W ⎝ W ⎠

(a) One-sided abrupt junction (with NA >> ND) in thermal equilibrium.

(b) Space charge distribution.
(c) Electric-field distribution.
(d) Potential distribution with distance, where Vbi is the built-in potential.
 Biased p-n Junction
Log (carrier concentration)
Eo – E
Neutral p-region Neutral n-region
Minute increase
ppo
Hole PE(x)
nno M eVo
Excess holes e(Vo–V)
pn(0)
Excess electrons np(0) Hole
diffusion
Electron
diffusion pno
npo SCL x
x
x′
W
Wo
V

(a) (b)

Forward biased pn junction and the injection of minority carriers (a) Carrier
concentration profiles across the device under forward bias. (b). The hole
potential energy with and without an applied bias. W is the width of the SCL
with forward bias
© 1999 S.O. Kasap, Optoelectronics (Prentice Hall)
 I-V Characteristics
Basic assumptions

•Space charge region has abrupt boundaries .

•Outside the boundaries the material is neutral.
•Carrier densities at the boundaries are related to the
potential difference across the junction.
•No generation and/or recombination in depletion regions.

Low-injection condition: pn 0 << nn 0

n p 0 << p p 0

CMR
 Eo (a) Eo–E (b)
p n
M
Ec Ec
e(Vo–V)
eVo Ec
Ec EFn
EFp EFp eV
EFn Ev
Ev
Ev
Ev
p n p n

SCL I
V

Eo+E (c) Eo+E (d)

Ec
Ec
e(Vo+Vr) e(Vo+Vr)
Thermal
EFp EFp generation
Ec Ec
Ev EFn Ev EFn

Ev Ev
p n p n

I = Very Small
Vr Vr

Energy band diagrams for a pn junction under (a) open circuit, (b) forward
bias and (c) reverse bias conditions. (d) Thermal generation of electron hole
pairs in the depletion region results in a small reverse current.
© 1999 S.O. Kasap, Optoelectronics (Prentice Hall)
Biased p-n Junction
Total depletion layer thickness

2ε s ⎛ N A + ND ⎞
W= ⎜⎜ ⎟⎟Vbi
e ⎝ N AND ⎠
W = xn + x p

2ε s (Vbi − V )
W=
e NB
Schematic representation of depletion layer width and
energy band diagrams of a
p-n junction under various biasing conditions.
a) Thermal-equilbrium condition. (b) Forward-bias
condition. (c) Reverse-bias condition
Biased p-n Junction
Built-in potential

kT N A N D
Vbi = ln
e ni2
kT p p 0 nn 0 kT nn 0
= ln 2
= ln
e ni e n p0
eVbi
nn 0 = n p 0 e kT

eVbi
p p 0 = pn 0 e kT

CMR
 Biased p-n Junction
At low-injection condition:
e (Vbi −V ) eVbi
n ≈ nn 0 = n p e kT
= n p 0e kT

eV
⎛ eV ⎞
n p = n p 0 e , or
kT
n p − n p0 = n p 0 ⎜⎜ e − 1⎟⎟
kT For injected electrons
⎝ ⎠ at the boundary x=-xp
eV
⎛ eV ⎞
pn = pn 0 e , or
kT
pn − pn 0 = pn 0 ⎜⎜ e − 1⎟⎟
kT For injected holes
⎝ ⎠ at the boundary x=xn

CMR
 Biased p-n Junction
At the condition of no generation and no electric fields in depletion region:

d 2 p pn − pn 0
− =0
dx 2
D pτ p
( x − xn )
⎛ eV
⎞ −
pn − pn 0 = pn 0 ⎜⎜ e − 1⎟⎟e
kT Lp Concentration of injected holes
⎝ ⎠
(x + x p )
⎛ eV
⎞ Concentration of injected electrons
n p − n p0 = n p 0 ⎜⎜ e − 1⎟⎟e
kT Lp

⎝ ⎠

CMR
Ideal Diode Equation
Currents at the boundaries:

eD p pn 0 ⎛ eV ⎞
J p ( xn ) = −eD p
dpn
| xn = ⎜ e − 1⎟,
kT
dx L p ⎜⎝ ⎟
⎠
eDn n p 0 ⎛ eV ⎞
J n (− x p ) = eDn
dn p
|− x p = ⎜ e − 1⎟
kT
dx Ln ⎜⎝ ⎟
⎠
Total current:

⎛ eV ⎞
J = J p ( xn ) + J n (− x p ) = J s ⎜⎜ e kT − 1⎟⎟,
⎝ ⎠
eD p pn 0 eDn n p 0
Js ≡ +
Lp Ln
CMR
 Total Current of p-n Junction

J
p-region SCL n-region

J = Jelec + Jhole
Total current
Majority carrier diffusion The total current
and drift current anywhere in the device is
Jhole constant. Just outside the
Jelec Minority carrier diffusion depletion region it is due
current to the diffusion of
minority carriers.
x
–Wp Wn

© 1999 S.O. Kasap, Optoelectronics (Prentice Hall)

Log (carrier concentration)
p-side n-side
SCL
ppo nno
C
nM pM
Electrons Holes
np(0) pn(0)
npo A B D pno
Wp Wn
x
M

Forward biased pn junction and the injection of

carriers and their recombination in the SCL.
© 1999 S.O. Kasap, Optoelectronics (Prentice Hall)
I-V Characteristics
of p-n Junction

Cartesian plot

Semilog plot

CMR
I-V Characteristic of p-n Junction

Current-voltage characteristic of p-n junction

CMR
Current-Voltage Characteristics

Current-voltage
characteristics of a typical
silicon p-n junction.

CMR
 I I = Io[exp(eV/ηkBT) − 1]

mA
Reverse I-V characteristics of a
pn junction (the positive and
V negative current axes have
Shockley equation different scales)

nA
Space charge layer
generation.

© 1999 S.O. Kasap, Optoelectronics (Prentice Hall)

 Minority Carrier (a)
Concentration
Eo+E (b)
Neutral p-region Neutral n-region Hole PE(x)
Thermally M
generated e(Vo+Vr)
EHP eVo

Holes
Electrons pno
npo
x x
Wo Wo
W Diffusion W(V = –Vr)
Drift

V
r
Reverse biased pn junction. (a) Minority carrier profiles and the origin of the
reverse current. (b) Hole PE across the junction under reverse bias
© 1999 S.O. Kasap, Optoelectronics (Prentice Hall)
Reverse diode current (A) at V = −5 V

10 - 4 Ge Photodiode
323 K Reverse diode current in a Ge pn
10 - 6
junction as a function of temperature in
0.63 eV
a ln( I rev ) vs. 1/ T plot. Above 238 K, Irev
10 - 8 is controlled by n i 2 and below 238 K it
is controlled by n i . The vertical axis is
10 - 1 0 a logarithmic scale with actual current
0.33 eV values. (From D. Scansen and S.O.
238 K
10 - 1 2 Kasap, Cnd. J. Physics. 70 , 1070-1075,
1992.)
10 - 1 4

10 - 1 6
0.002 0.004 0.006 0.008
1/Temperature (1/K)

© 1999 S.O. Kasap, Optoelectronics (Prentice Hall)

 Depletion Capacitance
a – increase of p-n junction width
due to the incremental bias voltage dV

b – equal incremental charges dQ

on both n- and p-sides

c – increase in the electric field dE

caused by the incremental charge dQ

(a) p-n junction with an arbitrary impurity profile under reverse bias. (b) Change in space charge distribution due
to change in applied bias. (c) Corresponding change in electric-field distribution.
 p-n Junction Capacitance
dQ
C= General definition of capacitance.
dV

dV = dE ⋅ W Change in the applied voltage.

ρ dQ Electric field in p-n junction is defined by

dE = dx = Poisson’s equation.
εs εs Here ρ is charge density in the depletion layer

dQ dQ εs eε s N B
C= = = =
2(Vbi − V )
Capacitance of abrupt
dV W dQ W p-n junction is dependent

εs
on reverse-bias voltage

CMR
Varactor
Doping profile in n-region

m
⎛ x⎞
N D ∝ B⎜⎜ ⎟⎟
⎝ x0 ⎠
Hyper abrupt p-n junction

C ∝ V −2
Frequency of varactor-L circuit

1
ω= ∝V
LC
CMR
Recombination
 Recombination
R = βn0 p0 direct recombination
Gth = Rth = β n0 p0 generation − recombination ballance
R = βnn pn = β (n0 + Δn )( p0 + Δp )

Δn = nn − n0 = Δp = pn − p0 , with light

dpn
= G − R = GL + Gth − Rth = 0, steady state
dt
 Recombination

U = R − Gth = GL = β (nn + p0 + Δp )Δp ≈ β nn Δp

p n − p0 1
U = GL ≈ βnn Δp = τp = , life time
τp βn0
dpn p n − p0
= Gth − R = −U = −
dt τp

⎛ t ⎞
pn (t ) = p0 + τ p GL exp⎜ − ⎟,
⎜ τ ⎟
⎝ p⎠
Generation by Light

⎛ t ⎞
pn (t ) = p0 + τ pGL exp⎜ − ⎟,
⎜ τ ⎟
⎝ p⎠
Boundary condition:

pn − p0
GL = , t =0
τp
pn = p0 + τ pGL
Surface Recombination

Real atomic configurations on solid surfaces

Surface Recombination

Real atomic configurations on solid surfaces with ad-atoms

 Surface Recombination

Surface structure, defects, and oxidation-reduction at the

atomic scale can be studied using SPM techniques. (a) An
ultrahigh vacuum scanning tunneling microscope (STM)
image shows iron atoms at an iron oxide (hematite) surface
(1). (b) An STM image of the same type of hematite surface
as in part (a) taken in air. Note that the unit cell in parts (a)
and (b) are virtually identical. The cells are marked, and
the edge length in both views is 0.5 nm. (c) An STM image
of a TiO2 surface (2) shows areas of ordered surface
structure (box A), steps (box B), and both point (box C), see
also inset) and line (box D) defects (the scale bar represents
10 nm). (d) An atomic force microscopy (AFM) image
taken in aqueous solution shows a step with kink sites on a
calcite (CaCO3) surface (3). (e) An AFM image of gypsum
(4) shows one kink site along the step (arrow). (f) In the
STM image of protoporphyrin molecules on a graphite
surface, the bright spots are Fe atoms complexed within
these molecules in a 1:4 mixture of Fe-containing to Fe-free
molecules (The scale bar represents 5 nm.)
∂n ⎡ J (x )A J n (x + dx)A ⎤
Adx = ⎢ n − ⎥ + (Gn − Rn )Adx
∂t ⎣ −e −e ⎦

∂J n
J n ( x + dx) = J n (x ) + dx + ...
∂x
∂p 1 ∂J p
∂n 1 ∂J n
= + (Gn − Rn ), electron CE =− + (G p − R p ), hole CE
∂t e ∂x ∂t e ∂x

Continuity Equation
 Continuity Equation (2)

dn dp
J n = eμn nE + eDn . J p = eμ p pE − eDp .
dx dx

np − np0 pn − pn 0
Rn = , Rp = , net recombination rate
τn τp

∂n p ∂E ∂n p ∂ 2np np − np0
= n p μn + μn E + Dn + Gn − ,
∂t ∂x ∂x ∂x 2
τn

∂pn ∂E ∂pn ∂ 2 pn pn − pn 0
= pn μ p + μpE + Dp + G − ,
∂t ∂x ∂x ∂x τp
2 p
 Home work
OEN-630 L8:

1. Problems 3.3
2. Problem 3.4 a, b
Ref. [1], Chapter 3

Deadline: September 26

CMR
 References
The following references include recommended
advanced reading for the subject:

1. S. O. Kasap, Optoelectronics and Photonics, Prentice Hall,

2001, Ch. 3
2. S. O. Kasap, Principles of Electronic Materials and
Devices, McGraw Hill, 2006, Chapters 1, 2
3. P. Yu, M. Cardona. Fundamentals of Semiconductors.
Springer, 2001
4. C. Kittel, Introduction to Solid State Physics, Benjamin
Inc., NY 1986

Website: http://vigyan.nsu.edu/~cmr/vgavrilenko.htm

CMR
6. Light emitting devices
6.1 The light emitting diode
6.1.1 Introduction
A light emitting diode consist of a p-n diode which is designed so that radiative
recombination dominates. Homojunction p-n diodes, heterojunction p-i-n diodes where the
intrinsic layer has a smaller bandgap (this structure is also referred to as a double-hetero-
structure) and p-n diodes with a quantum well in the middle are all used for LEDs. We will
only consider the p-n diode with a quantum well because the analytical analysis is more
straight forward and also since this structure is used often in LEDs and even more frequently
in laser diodes.
6.1.2 Rate equations
The LED rate equations are derived from the continuity equations as applied to the p-n
diode:

∂n 1 ∂Jn
= -R+G [6.1.1]
∂t q ∂x
where G is the generation rate per unit volume and R is the recombination rate per unit
volume. This equation is now simplified by integrating in the direction perpendicular to the
plane of the junction. We separate the integral in two parts: one for the quantum well, one
for the rest of the structure.

⌠ ∂n ⌠ ∂n J JSHR Jbb Jideal NP - Ni12 1

 +  = q - q - q - q - Σ (NkPk - Nik2)Bk - (N + P + 2N )
⌡ ∂t ⌡ ∂t k i1 τnr
qw p-n
[6.1.2]
where k refers to the quantum number in the well. If we ignore the carriers everywhere
except in the quantum well and assume that only the first quantum level is populated with
electrons/holes and that the density of electrons equals the density of holes, we obtain:

∂N J N S
= - B1N2 - + [6.1.3]
∂t q 2τnr τab
where the last term is added to include reabsorption of photons. The rate equation for the
photon density including loss of photons due to emission (as described with the photon
lifetime τph) and absorption (as described with the photon absorption time τab) equals:

∂S S S
= B1N2 - - [6.1.4]
∂t τph τab
The corresponding voltage across the diode equals:
Principles of Electronic Devices 6.1 © Bart J. Van Zeghbroeck 1996
 Egqw
Va = + Vt ln [(eN/Nc - 1) (eN/Nv* - 1)] [6.1.5]
q
Where the modified effective hole density of states in the quantum well, Nv*, accounts for
the occupation of multiple hole levels as described in section 4.4.3.d. The optical output
power is given by the number of photons which leave the semiconductor per unit time,
multiplied with the photon energy:

S Θ 2
Pout = hν A (1-R) c [6.1.6]
τph 4
where A is the active area of the device, R is the reflectivity at the surface and Θc is the
critical angle for total internal reflection1

n1 - n2 n1
R=( )2 and Θc = sin-1( ) [6.1.7]
n1 + n2 n2
The reflectivity and critical angle for a GaAs Air interface are 30 % and 16º respectively.
6.1.3. DC solution to the rate equations
The time independent solution in the absence of reabsorption, as indicated with the subscript
0, is obtained from:

J0 N0
0= - B N02 - [6.1.8]
q 2τnr

S0
0 = B N02 - [6.1.9]
τph
where B is the bimolecular recombination constant. Solving these equations yields:

1 16 τnr2 B J0
N0 = [ 1+ -1] [6.1.10]
4Bτnr q
for small currents this reduces to: (J << q/16τnr2B)

2τnr J0
N0 = [6.1.11]
q
which indicates that SHR recombination dominates, whereas for large currents one finds: (J
>> q/16τnr2B)

1See Appendix A.7 for the derivation of the reflectivity at dielectric interfaces.

Principles of Electronic Devices 6.2 © Bart J. Van Zeghbroeck 1996

 J0
N0 = [6.1.12]
qB
The DC optical output power is:

Θc2
P0 = hν B N02 A(1-R) [6.1.13]
4
This expression explains the poor efficiency of an LED. Even if no non-radiative
recombination occurs in the active region of the LED, most photons are confined to the
semiconductor because of the small critical angle. Typically only a few percent of the
photons generated escape the semiconductor. This problem is most severe for planar surface
emitting LEDs. Better efficiencies have been obtained for edge emitting, "superluminescent"
LEDs (where stimulated emission provides a larger fraction of photons which can escape the
semiconductor) and LEDs with curved surfaces.
6.1.4 AC solution to the rate equations
Assume that all variables can be written as a sum of a time independent term and a time
dependent term (note that n(t) is still a density per unit area):

N = N0 + n1(t) J = J0 + j1(t)

S = S0 + s1(t) P = P0 + p1(t)

Va = Va + va(t) [6.1.14]
The rate equations for the time dependent terms the given by:

∂n1 n (t) j (t)

= -B 2N0 n1(t) - B n12(t) - 1 + 1 [6.1.15]
∂t 2τ0 q

∂s1 s
= B 2N0 n1(t) + B n12(t) - 1 [6.1.16]
∂t τph
Assuming the AC current of the form j1 = j10ejωt and ignoring the higher order terms we can
obtain a harmonic solution of the form:

n1 = n10 ejωt s1 = s10 ejωt p1 = p10 ejωt [6.1.17]

yielding:

1 B 2N0 j10 τph τeff

s10 = [6.1.18]
q (1 + jω τph) (1 + jω τeff)
where τeff depends on N0 as:

Principles of Electronic Devices 6.3 © Bart J. Van Zeghbroeck 1996

 1
τeff = [6.1.19]
1
2B N0 +
2τ0
and the AC responsivity is:

p10 hν πΘ 2 B 2N0 τeff

= (1-R) c [6.1.20]
j10 q 4π (1 + jω τph) (1 + jω τeff)
at ω = 0 this also yields the differential quantum efficiency (D.Q.E)

p q (1-R)Θc2B N0 τeff (1-R)Θc2 B N0 τ0

D.Q.E. = 10 = = [6.1.21]
j10 hν 2 (4 B N0 τ0 + 1)
6.1.5 Equivalent circuit of an LED
The equivalent circuit of an LED consist of the p-n diode current source parallel to the diode
capacitance and in series with a linear series resistance R. The capacitance, C, is obtained
from

1 N/Nc N/Nv 1 N/Nv N/Nc

[ e (e - 1) + e (e - 1)]
1 dV a dV a 1 Nc Nv
= = = Vt [6.1.22]
C dQ q dN q N/N N/N
(e c - 1) (e v - 1)
or

qN0
C= [6.1.23]
mVt
with

N0 eN/Nc N0 eN/Nv
m= + [6.1.24]
Nc (eN/Nc - 1) Nv (eN/Nv - 1)
N (N +N )
for N0 << Nc and/or Nv m = 2 while for N0 >> Nc and/or Nv m = 0 c v
NcNv

Principles of Electronic Devices 6.4 © Bart J. Van Zeghbroeck 1996

6.2 The laser diode
6.2.1 Emission, Absorption and modal gain
In order to find the modal gain one first has to calculate the photon absorption spectrum as
well as the spontaneous photon emission spectrum from the quasi-Fermi levels. Energy and
momentum conservation requires that

Eph = En - Ep [6.2.1]
with

/h2 kn2
En = Ec + E1n + [6.2.2]
2 mn*

/h2 kp2
Ep = Ev - E1p - [6.2.3]
2 mp*
and

kn = kp = k [6.2.4]
if we assume that the photon momentum is negligible2. These equations can be reduced to

/h2 k2 1 1 1
Eph = Egqw1 + , with * = * + * [6.2.5]
2 mr* mr mn mp
where Egqw1 is the energy between the lowest electron energy in the conduction band and
the lowest hole energy in the valence band. En and Ep then become:

mr*
En = Ec + E1n + (Eph - Egqw1) [6.2.6]
mn*

mr*
Ep = Ev - E1p - (Eph - Egqw1) [6.2.7]
mp*
The emission and absorption spectra (β(Eph) and α(Eph)) are obtained from:

β(Eph) = βmax Fn (En) [1-Fp(Ep)] [6.2.8]

α(Eph) = αmax [1 - Fn(En)] Fp(Ep) [6.2.9]

Stimulated emissions occurs if an incoming photon triggers the emission of another photon.
The spectrum equals that for spontaneous emission, minus that for absorption since both are
competing processes:

g(Eph) = β(Eph) - α(Eph) = gmax [Fn(En) - Fp(Ep)] [6.2.10]

2This assumption causes an error of a few percent.

Principles of Electronic Devices 6.5 © Bart J. Van Zeghbroeck 1996

The normalized gain spectrum is shown in figure 6.1 for different values of the carrier
density. The two staircase curves indicate the maximum possible gain and the maximum
possible absorption in the quantum well.

2.5

1.5

1
Ise/Ise,max1

0.5

-0.5

-1

-1.5

-2

-2.5
1.4 1.45 1.5 1.55 1.6 1.65 1.7 1.75 1.8 1.85 1.9
Photon Energy [eV]

Fig.6.1. Normalized gain versus photon energy of a 10nm GaAs quantum well for a carrier
density of 1012 (lower curve), 3 x 1012 ,5 x 1012 ,7 x 1012 and 9 x 1012 (upper
curve) cm-2.

The theoretical gain curve of figure 6.1 exhibits a sharp discontinuity at Eph = Egqw1. The
gain can also be expressed as a function of the carrier density when assuming that only one
electron and one hole level are occupied:

1 - e-N/Nc
g(Eph) = gmax [
Eph - Eqw1 mr*
1 + e-N/Nv(exp{ } - 1)
kT mn*

e-P/Nv
- ] [6.2.12]
Eqw1 - Eph mr*
exp{ }(1 - e-P/Nv) + e-P/Nv
kT mp*
The peak value at Eph = Eqw1, assuming quasi-neutrality (N = P) is then:

gpeak = g (Eqw1) = gmax (1 - e-N/Nc - e-N/Nv ) [6.2.13]

The maximum gain can be obtained from the absorption of light in bulk material since the
wavefunction of a free electron in bulk material is the same as the wavefunction in an infinite
stack of infinitely deep quantum wells, provided the barriers are infinitely thin and placed at
the nodes of the bulk wavefunction. This means that for such a set of quantum wells the
absorption would be the same as in bulk provided that the density of states is also the same.

Principles of Electronic Devices 6.6 © Bart J. Van Zeghbroeck 1996

This is the case for Eph = Eqw1 so that the maximum gain per unit length is given by:

K h2 1
gmax = K Eqw1-E1 = [6.2.14]
2 2 mr* Lx
where Lx is the width of the quantum well. This expression shows that the total gain of a
single quantum well due to a single quantized level is independent of the width3. The
corresponding value for GaAs quantum wells is 0.006 or 0.6%.
Experimental gain curves do not show the discontinuity at Eph = Eqw1 due to inter-carrier
scattering which limits the lifetime of carriers in a specific state. The line width of a single set
of electron and hole level widens as a function of the scattering time which disturbs the
phase of the atomic oscillator. Therefore, an approximation to the actual gain curve can be
obtained by convoluting [6.2.10] with a Lorenzian line shape function:

∆ν
g(Eph) = ⌠ gmax [Fn(En) - Fp(Ep)] dν [6.2.15]
⌡ ∆ν
2π[(ν - Eph/h)2 + ( 2 )2]
1 1
with ∆ν = , where τ is the carrier collision time in the quantum well. The original and
π τ
convoluted gain curve are shown in Fig. 6.2.

2.5
2
1.5
1
Ise/Ise,max1

0.5
0
-0.5
-1
-1.5
-2
-2.5
1.4 1.6 1.8 2 2.2 2.4
Photon Energy [eV]

Fig.6.2. Original and convoluted gain spectrum of a 10 nm GaAs quantum well with a

3There is a weak dependence of m* on the width of the well.

Principles of Electronic Devices 6.7 © Bart J. Van Zeghbroeck 1996

 carrier density of 3 x 1012 cm-2 and a collision time of 0.09 ps.

For lasers with long cavities such as edge emitter lasers, one finds that the longitudinal
modes are closely spaced so that lasing will occur at or close to the peak of the gain
spectrum. It is therefore of interest to find an expression for the peak gain as a function of
the carrier density4. A numeric solution is shown in Fig.6.3 where the peak gain is
normalized to the maximum value of the first quantized energy level. Initially, the gain peak
is linear with carrier concentration but saturates because of the constant density of states,
until the gain peak associated with the second quantized level takes over. Since the peak
gain will be relevant for lasing we will consider it more closely. As a first order
approximation we will set the peak gain g(N) equal to:

g(N) = (N - Ntr) [6.2.16]

where is the differential gain coefficient. This approximation is only valid close to N = Ntr,
and even more so for quantum well lasers as opposed to double-hetero-structure lasers. An
approximate value for the differential gain coefficient of a quantum well can be calculated
from [6.2.13] yielding:

e-Ntr/Nc e-Ntr/Nv
= gmax [ + ] [6.2.17]
Nc Nv

4Experimental values for the gain versus current density can be found in: G. Hunziker, W. Knop and C.
Harder, "Gain Measurements on One, Two and Three Strained GaInP Quantum Well Laser Diodes", IEEE
Trans. Quantum Electr., Vol. 30, p 2235-2238, 1994.

Principles of Electronic Devices 6.8 © Bart J. Van Zeghbroeck 1996

 6000

5000

4000
Peak gain [cm-1]

3000

2000

1000

0
0 2 4 6 8 10 12 14 16 18 20
Carrier density [1E12 cm-2]

Fig.6.3. Calculated gain versus carrier density for a 10 nm GaAs quantum well (solid line)
compared to equation [6.2.13]

From Fig. 6.3 one finds that the material becomes "transparent" when the gain equals zero
or:

g(Ntr) = 0 = gmax [Fn(En)(1 - Fp(Ep))-(1 - Fn(En))Fp(Ep)] [6.2.18]

which can be solved yielding:

Eph = En - Ep = Efn - Efp = qVa [6.2.19]

The transparency current density is defined as the minimal current density for which the
material becomes transparent for any photon energy larger than or equal to Egqw1. This
Egqw1
means that the transparency condition is fulfilled for Va = . The corresponding carrier
q
density is referred to as Ntr, the transparency carrier density. The transparency carrier
density can be obtained from by setting gmax = 0, yielding

Ntr = - Nc ln(1 - eNtr/Nv) [6.2.20]

This expression can be solved by iteration for Nv > Nc. The solution is shown in Figure 6.4.

Principles of Electronic Devices 6.9 © Bart J. Van Zeghbroeck 1996

 2.5

Ntr/Nc 1.5

0.5

0
0 5 10 15 20
Nv/Nc

Fig.6.4. Normalized transparency carrier density versus the ratio of the effective density of
states in the valence and conduction band.

To include multiple hole levels one simply replaces Nv by Nv* as described in section
4.4.3.d.
6.2.2 Principle of operation of a laser diode
A laser diode consists of a cavity, defined as the region between two mirrors with reflectivity
R1 and R2, and a gain medium, in our case a quantum well. The optical mode originates in
spontaneous emission which is confined to the cavity by the waveguide. This optical mode is
amplified by the gain medium and partially reflected by the mirrors. The modal gain depends
on the gain of the medium, multiplied with the overlap between the gain medium and the
optical mode which we call the confinement factor, Γ, or:

modal gain = g(N)Γ [6.2.21]

This confinement factor will be calculated in section 6.2.5. Lasing occurs when for light
traveling round trip through the cavity the optical gain equals the losses. For a laser with
modal gain g(N)Γ and wave guide loss α this condition implies:

R1R2 e2(g(N)Γ-α)L = 1 [6.2.22]

where L is the length of the cavity. The distributed loss of the mirrors is therefore:

1 1
mirror loss = ln [6.2.23]
L R1R2
6.2.3 Longitudinal modes in the laser cavity.

Principles of Electronic Devices 6.10 © Bart J. Van Zeghbroeck 1996

Longitudinal modes in the laser cavity correspond to standing waves between the mirrors. If
we assume total reflection at the mirrors this wave contains N/2 periods where N is an
integer. For a given wave length λ and a corresponding effective index, neff, this yields:

2 neff L
N= [6.2.24]
λ
Because of dispersion in the waveguide, a second order model should also include the
wavelength dependence of the effective index. Ignoring dispersion we find the difference in
wavelength between two adjacent longitudinal modes from:

2 L neff
N= [6.2.25]
λ1

2 L neff
N+1= [6.2.26]
λ2

1 1 λ12
∆λ = 2 L neff ( - ) ≅ [6.2.27]
N N+1 2 L neff
Longer cavities therefore have closer spaced longitudinal modes. An edge emitting (long)
cavity with length of 300 µm, neff = 3.3, and λ = 0.8 µm has a wavelength spacing ∆λ of
0.32 nm while a surface emitting (short) cavity of 3 µm has a wavelength spacing of only 32
nm. These wavelength differences can be converted to energy differences using:

∆λ
∆E = - Eph [6.2.28]
λ
so that 0.32 nm corresponds to -6.2 meV and 32 nm to 620 meV. A typical width of the
optical gain spectrum is 60 meV, so that an edge emitter biased below threshold can easily
contain 10 longitudinal modes, while for a surface emitter the cavity must be carefully
designed so that the longitudinal mode overlaps with the gain spectrum.
A more detailed analysis of a Fabry-Perot etalon is described in A.7.3, providing the
reflectivity, absorption and transmission as a function of photon energy
6.2.4 Waveguide modes5
The optical modes in the waveguide determine the effective index used to calculate the
longitudinal modes as well as the confinement factor which affects the modal gain. Starting
from Maxwell's equations in the absence of sources:

5A detailed description of modes in dielectric waveguides can be found in Marcuse, "dielectric waveguides",
2nd ed.

Principles of Electronic Devices 6.11 © Bart J. Van Zeghbroeck 1996

 ∂
∇×H = ε0 n2(x,y,z) [6.2.29]
∂t

∂H
∇ × = - µ0 [6.2.30]
∂t
and assuming a propagating wave in the z-direction and no variation in the y-direction we
obtain the following one-dimensional reduced wave equation for a time harmonic field, =
jωt
x e , of a TM mode:

∂2 x
+ (n2(x) k2 - β2) x = 0 [6.2.31]
∂x2
ω ω
with the propagation constant given by β = neff, and k = , this equation becomes:
c c

d2 x
2(x) - n 2)
ω2
+ (n eff c2 x = 0 [6.2.32]
dx2
this equation is very similar to the Schrödinger equation. In fact previous solutions for
quantum wells can be used to solve Maxwell's equation by setting the potential V(x) equal to
ω2 2m*
-n2(x) and replacing 2 by / 2 . The energy eigenvalues, E , can then be interpreted as
c h
minus the effective indices of the modes: -n2eff, . One particular waveguide of interest is a
slab waveguide consisting of a piece of high refractive index material, n1, with thickness d,
between two infinitely wide cladding layers consisting of lower refractive index material, n2.
From Appendix A.1.3. one finds that only one mode exists for:

c2 π [6.2.33]
V0 = -n22 + n12 E10 = 2 ( )2
ω d

c 1 λ
or d ≤ π = [6.2.34]
ω 2
n1 -n2 2 2 n12 - n22
For λ = 0.8 µm, n1 = 3.5 and n2 = 3.3 one finds d ≤ 0.34 µm.
6.2.5 The confinement factor
The confinement factor is defined as the ratio of the modal gain to the gain in the active
medium at the wavelength of interest:

∞
⌠ 2
⌡ g(x) | x| dx
modal gain -∞
Γ= = [6.2.35]
g ∞
⌠ 2
⌡ | x| dx
-∞

Principles of Electronic Devices 6.12 © Bart J. Van Zeghbroeck 1996

for a quantum well with width Lx, the confinement factor reduces to

Lx/2
⌠ 2
⌡ | x| dx
-Lx/2
Γ= ≅ 0.02...0.04 for a typical GaAs single quantum well laser [6.2.36]
∞
⌠ 2
⌡ | x| dx
-∞
6.2.6 The rate equations for a laser diode.
Rate equations for each longitudinal mode, λ, with photon density Sλ and carrier density Nλ
which couple into this mode are:

∂Nλ Jλ N N N ∂S ∂x
= - Bλ Nλ2 - λ + Σ k - Σ k - λ , λ = 1, 2, ..., λmax [6.2.37]
∂t q 2τ0 τ τ ∂x ∂t
k kλ k λk

∂Sλ S ∂S ∂x
= βλ BλNλ2 - λ + λ , λ = 1, 2, ..., λmax [6.2.38]
∂t τphλ ∂x ∂t
Rather than using this set of differential equations for all waveguide modes, we will only
consider one mode with photon density S, whose photon energy is closest to the gain peak.
The intensity of this mode will grow faster than all others and eventually dominate. This
simplification avoids the problem of finding the parameters and coefficients for every single
mode. On the other hand it does not enable to calculate the emission spectrum of the laser
diode. For a single longitudinal mode the rate equations reduce to:

dN J N
= - BN2 - - v Γ (N - Ntr) S [6.2.39]
dt q 2τ0 gr

dS S
= β BN2 - + vgr Γ (N - Ntr) S [6.2.40]
dt τph

1
P1 = vgr S W ln [6.2.41]
R1

a) DC solution to the rate equations

The time independent rate equations, ignoring spontaneous emission are:

J N
0 = 0 - B N02 - 0 = vgr Γ (N0 - Ntr)S0 [6.2.42]
q 2τ0

Principles of Electronic Devices 6.13 © Bart J. Van Zeghbroeck 1996

 S
0 = - 0 + vgr Γ (N0 - Ntr)S0 [6.2.43]
τph
where the photon life time is given by:

1 1 ∂x ∂S 1 1
= = vgr (α + ln ) [6.2.44]
τph S ∂t ∂x L R1R2
from which we can solve the carrier concentration while lasing:

1
N0 = Ntr + [6.2.45]
τph vgr Γ
which is independent of the photon density6. The threshold current density is obtained when
S0 = 0

N
J0 | = Jth = q (B N02 + 0 ) [6.2.46]
2 τ 0
| (S0 = 0)
The photon density above lasing threshold, and power emitted through mirror R1, are given
by:

J -J 1
S0 = 0 th [6.2.47]
q vgr Γ (N0-Ntr)
and the power emitted through mirror 1 is:

1
P10 = hν S0 W vgr ln [6.2.48]
R1
The differential efficiency of the laser diode is:

1
ln
dP0 hν R1
D.E. = = [6.2.49]
dI0 q 1
ln + αL
R1R2
and the quantum efficiency is:

1
ln
q dP0 R1
η = = [6.2.50]
hν dI0 1
ln + αL
R1R2
Efficient lasers are therefore obtained by reducing the waveguide losses, increasing the
reflectivity of the back mirror, decreasing the reflectivity of the front mirror and decreasing

6a more rigorous analysis including gain saturation reveals that the carrier concentration does increase with
increasing current, even above lasing. However this effect tends to be small in most laser diodes.

Principles of Electronic Devices 6.14 © Bart J. Van Zeghbroeck 1996

the length of the cavity. Decreasing the reflectivity of the mirror also increases the threshold
current and is therefore less desirable. Decreasing the cavity length at first decreases the
threshold current but then rapidly increases the threshold current.
b) AC solution to the rate equations
Assuming a time-harmonic solution and ignoring higher order terms (as we did for the LED)
the rate equations become:

j n
jω n1 = 1 - 1 - Γ (N0-Ntr) s1 vgr - Γ n1 S0 vgr [6.2.51]
q τeff

s
jω s1 = Γ (N0-Ntr) s1 vgr - 1 + Γ n1 S0 vgr [6.2.52]
τph
where τeff is the same as for an LED and given by equation [6.1.19]. Using Γ (N0-Ntr) vgr
1
= these equations can be solved yielding:
τph

1 Γ S0 vgr
j1 = jω q n1 + q n1 ( + Γ S0 vgr) + q n1 [6.2.53]
τeff jω τph
replacing n1 by relating it to the small signal voltage v1

mVtn1
v1 = [6.2.54]
N0
The equation for the small signal current i1 can be written as

1 1
i1 = (jω C + + )v [6.2.55]
R jω L 1
q N0 A N0 eN/Nc N0 eN/Nv
with C = , and m = + , where A is the area of the
mVt
Nc (eN/Nc - 1) Nv (eN/Nv - 1)
laser diode.

1 1
=C( + Γ S0 vgr) [6.2.56]
R τeff
and

1 τph
L= [6.2.57]
C Γ S0 vgr
c) Small signal equivalent circuit
Adding parasitic elements and the circuit described by the equation [6.2.48] we obtain the
following equivalent circuit, where LB is a series inductance, primarily due to the bond wire,
Rs is the series resistance in the device and Cp is the parallel capacitance due to the laser
contact and bonding pad.

Principles of Electronic Devices 6.15 © Bart J. Van Zeghbroeck 1996

 LB Rs

R C L
Cp

Rd

Fig.6.5 Small signal equivalent circuit of a laser diode

The resistor, Rd, in series with the inductor, L, is due to gain saturation7 and can be obtained
by adding a gain saturation term to equation [6.2.16]. The optical output power is
proportional to the current through inductor L, i1L, which is given by:

q A s1 1 1
i1L = = q A s1 vgr (α + ln ) [6.2.58]
τph L R1R2
and the corresponding power emitted from mirror R1

1
p1 = s1 hν vgr W ln [6.2.59]
R1
Ignoring the parasitic elements and the gain saturation resistance, Rd, one finds the ac
responsivity p1/i1 as:

1
ln
p1 hν R1 1
= [6.2.60]
i1 q 1 L
(αL + ln ) 1 + jω + (jω)2LC
R1R2 R
from which we find the relaxation frequency of the laser:

1 Γ S0 vgr Γ P0
ω0 = = = [6.2.61]
LC τph 1
τph hν W ln
R1
or the relaxation frequency is proportional to the square root of the DC output power. The
amplitude at the relaxation frequency relative to that at zero frequency equals:

7for a more detailed equivalent circuit including gain saturation see: Ch. S. Harder et al. High-speed
GaAs/AlGaAs optoelectronic devices for computer applications, IBM J. Res. Develop., Vol 34, No. 4, July
1990, p. 568-584.

Principles of Electronic Devices 6.16 © Bart J. Van Zeghbroeck 1996

 p1
|
j1
| ω = ω0 R 1
= = [6.2.62]
L ω0 1
p1 + τph ω0
| ω0 τeff
j1
|ω=0
6.2.7 Threshold current of multi-quantum well laser
Comparing threshold currents of laser diodes with identical dimensions and material
parameters but with a different number of quantum wells, m, one finds that the threshold
currents are not simple multiples of that of a single quantum well laser.
Let us assume that the modal gain, g, is linearly proportional to the carrier concentration in
the wells and that the carriers are equally distributed between the m wells. For m quantum
wells the modal gain can be expressed as:

g = m (N - Ntr) = ∆N m [6.2.63]
where is the differential gain coefficient and Ntr is the transparency carrier density. Since
the total modal gain is independent of the number of quantum wells we can express the
carrier density as a function modal gain at lasing8.

g ∆N
N= + N0 = [6.2.64]
m m
The radiative recombination current at threshold is then

∆N ∆N2
Jtr = q B1 m (Ntr + )2 = q B1 (Ntr2 m + 2Ntr ∆N + ) [6.2.65]
m m
This means that the threshold current density is a constant plus a component which is
proportional to the number of quantum wells. The last term can be ignored for m>>1 and
∆N<<Ntr.
6.2.8 Large signal switching of a laser diode
Because of the non-linear terms in the rate equations the large signal switching of a laser
diode exhibits some peculiar characteristics. The response to a current step is shown in the
figure below. The carrier density initially increases linearly with time while the photon
density remains very small since stimulated emission only kicks in for N > N0.

8We assume here that we are comparing identical lasers which only differ by the number of quantum wells.

Principles of Electronic Devices 6.17 © Bart J. Van Zeghbroeck 1996

 8

P [mW] , N/Ntr
4

0
0 0.5 1 1.5 2 2.5 3
t [ns]

Fig.6.6 Optical power and normalized carrier concentration versus time when applying a
step current at t = 0 from I = 0.95 Ith to I = 1.3 Ith.

Both the carrier density and the photon density oscillate around their final value. The
oscillation peaks are spaced by roughly 2π/ω0, where ω0 is the small signal relaxation
frequency at the final current. The photon and carrier densities are out of phase as carriers
are converted into photons due to stimulated emission, while photons are converted back
into electron-hole pairs due to absorption. High speed operation is obtained by biasing close
to the threshold current and driving the laser well above the threshold. In addition one can
use the non-linear behavior to generate short optical pulses. By applying a current pulse
which is long enough to initiate the first peak in the oscillation, but short enough to avoid
the second peak, one obtains an optical pulse which is significantly shorter that the applied
current pulse. This method is referred to as gain switching or current spiking.

Principles of Electronic Devices 6.18 © Bart J. Van Zeghbroeck 1996

Intro. to Fibre Optic Communications Systems OE & Recvrs - 51

Introduction to photodetectors

• Photodetector: A device which converts optical

power into an electrical signal
• Desirable properties include:
– high responsivity/sensitivity to light
– fast response time, i.e. large bandwidth
– low noise, i.e. minimally degrades the SNR
– insensitive to temperature
• Numerous types of detectors, some of which are:
– photomultiplier tubes
– pyroelectric detectors
Ö reverse biased diode photodetectors
(PIN, Avalanche (APD), MSM)
– phototransistors
• The semiconductor photodetectors are commonly
used because of their low cost, high reliability and
high performance (Note: Fibre optic communication
systems at 1.3 and 1.55 µm use InGaAs pin
devices almost exclusively. For wavelengths < 1
µm, Si photodetectors are generally used.)
Intro. to Fibre Optic Communications Systems OE & Recvrs - 52

Photodetectors cont’d:

Basic Concepts: Absorption and photocurrent

Absorption: Photons with an energy greater than the bandgap
energy of the semiconductor can be absorbed to create an ehp.

Iinc

I( z ) = Iinc exp[−α s (λ ) z ]
Intensity [W/m2]

Distance into photodetector [m]

Photon energy (eV)

5 4 3 2 1 0.8 0.7
0.9
1×108

Ge In0.7Ga0.3As0.64P0.36
1×107

In0.53Ga0.47As
Si
1×106
GaAs
α (m-1)
InP
1×105
a-Si:H

1×104

1×103
0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8
Wavelength (µm)
Absorption coefficient (α) vs. wavelength (λ) for various semiconductors
(Data selectively collected and combined from various sources.)
Intro. to Fibre Optic Communications Systems OE & Recvrs - 53

Photodetectors cont’d:
Basic Concepts: Absorption and photocurrent

Optical power absorbed by semiconductor of width, w:

P( w) = P0 (1 − exp[−α s (λ ) w])

P0=Iinc*Area
The primary photocurrent, IP, that flows from the photodetector is
the result the absorbed optical power.
e
IP = P0 (1 − exp[−α s (λ ) w])(1 − R f )
hν

Takes reflections at surface

into account … this term is
frequently ignored.

The quantum efficiency, η, is defined as

electron generation rate I p / e Pabs

η= = = = 1 − exp[−α s (λ ) w](1 − R f )
photon incidence rate P0 / hv P0

It is also possible to define the primary photocurrent, IP, as being

directly proportional to the incident optical power P0.
I P = R P0
where R is the responsivity in [A/W].

η e ηλ
R = ≈
hν 1.24
Intro. to Fibre Optic Communications Systems OE & Recvrs - 54

Photodetectors cont’d:

Basic Concepts: Time response

Time response of photodetector (photodiode) and its associated
circuitry depend mainly on 3 things:
1) carrier transit time through depletion region
2) diffusion time of carriers outside generated outside of depletion
region
3) RC time constant of diode and circuitry.

Fig. 6-10 Schematic representation of a reverse biased pin

diode.

Transit time: fundamental limit to response time of photodiode.

Transit time is the time required for the generated carriers to
travel across the depletion/active region (drift).

w
td =
vd
Intro. to Fibre Optic Communications Systems OE & Recvrs - 55

Photodetectors cont’d:

Basic Concepts: Time response

The bias field in the depletion region is usually large enough for
carrier to reach their saturated velocities…electron and holes
velocities are typically different. (Example Si: electrons - 8.4 × 106
cm/s, holes 4.4 × 106 cm/s.)

Diffusion: these processes are very slow compared to the drift of

carriers in the depletion region (~ 1ns to drift 1 µm).
The electric field outside of the depletion region is quite small so
that the generated carriers do not drift. The contribution of
diffusion to the time response can be minimized by increasing the
depletion region width. Can generally be ignored if the diffusion
time is much larger than the width of the optical pulse.

Capacitance effects: due to parasitic capacitance of the diode

packaging and the capacitance of the diode junction.
Junction capacitance: C j =
εs A
w

Generally, Rload >> Rd, such that the RC time constant associated
with the circuit is: trc = CjRload.
Rd

Iph Cj
Rload
Intro. to Fibre Optic Communications Systems OE & Recvrs - 56

Photodetectors cont’d:

Basic Concepts: Time response

Fig. 6-11 Photodiode response time to an optical input pulse

showing the 10%-90% rise time and fall time. Note that the rise
and fall times are not necessarily equal. Usually just specify PD
by a risetime, tr.
Input w >> 1/α
Pulse Small Cj

w >> 1/α w < 1/α

Large Cj Small Cj

w↑ ⇒ td↑
Tradeoff between speed and
w↑ ⇒ η↑ efficiency.

w↓ ⇒ td ↓ Reduce area of device so that capacitance doesn’t

w↓ ⇒ trc↑ become excessive.
Intro. to Fibre Optic Communications Systems OE & Recvrs - 57

Photodetectors cont’d:
PIN photodetector

Fig. 6-1 Schematic representation of a pin photodiode circuit with

an applied reverse bias, and photocurrent Ip. Width of intrinsic
region on order of 3 -5 µm.
Iph

Vbias
No Popt

Popt > 0

Fig. 6-2 Simplified energy-band I-V characteristic for a diode.

diagram for a pin photodiode. If Dark current exists when
a photon has an energy > Eg there is no light. Silicon has
then an electron-hole pair will be low dark current, Ge high. III-
generated. Vs between.
Intro. to Fibre Optic Communications Systems OE & Recvrs - 58

Photodetectors cont’d:

PIN photodetector

Fig. 6-4 Comparison of

the responsivity and
quantum efficiency
as a function of
wavelength for PIN
diodes fabricated
from different
materials.

pin photodiode
electron generation rate I p / e Pabs
η= = = = 1 − exp[−α s (λ ) w](1 − R f )
photon incidence rate P0 / hv P0
η e ηλ
I P = R P0 R = ≈
hν 1.24
Intro. to Fibre Optic Communications Systems OE & Recvrs - 59

Photodetectors cont’d:
Incident photons
Avalanche photodiodes

Fig. 6-5 Reach-through

avalanche photodiode
structure and the electric
fields in the depletion and
multiplication regions.
Electrons and holes have
different ionization rates.
Photocurrent has a gain
or multiplication factor, M.
Substrate

Carrier multiplication:
IM
M=
Ip

APD

I P = R P0 I M = RAPD P0

ηe
RAPD = M =R M
hν

Fig. 6-7 Typical room temperature

current gains of a Si reach-through
avalanche photodiode.
Intro. to Fibre Optic Communications Systems OE & Recvrs - 60

Photodetectors cont’d:
Avalanche photodiodes

Figure 6-6 Carrier

ionization rates for
various semiconductor
materials.
Intro. to Fibre Optic Communications Systems OE & Recvrs - 61

Photodetectors cont’d:
Metal-semiconductor-metal (MSM) photodetectors (not in
book)

Typical structure: Side, top and symbol views for an MSM

photodetector. The device is bipolar and is identified by the
characteristic interdigitated fingers. Simplest MSM requires
fingers, absorption layer and a substrate. Speeds up tp 300 GHz
have been recorded!

MSM Arrays. 25x25 µm MSMs attached to a common electrical

bus. The common bus can be used to accept signals from
several different optical sources.
Intro. to Fibre Optic Communications Systems OE & Recvrs - 62

Photodetectors cont’d:

Photodetector Noise and Receivers

Notes: Rs << RL → Ignore Rs

Ra >> RL → Ignore Ra

APD Only

Photodetector Avalanche Gain Load/Bias Resistor Preamplifier

Quantum Noise Excess noise due to Thermal Noise Thermal Noise

Dark Current Noise random nature of Noise in Transistors
avalanche process

Noise Mechanisms
Shot Noise - due to the quantum nature of light (photons) and
current (electrons/holes) → granular quality to what appears to be
a continuous signal.
quantum noise(light)
dark current noise(thermal generation of electrons)