Photodetectors

Photodetectors
Principle of the p-n junction Photodiode

SiO2 Vr Iph R Vout

 Schematic diagram of a reverse Electrode
biased p-n junction photodiode
hv > Eg h+ e–
 Photocurrent is depend on number
of EHP and drift velocity. n
p+
 The electrode do not inject carriers E
but allow excess carriers in the AR Electrode
sample to leave and become coating
W
collected by the battery. r Depletion
net
region

 Net space charge across the diode eN d

x
in the depletion region. Nd and Na
are the donor and acceptor
concentrations in the p and n sides.
–eN a

E (x )
 The field in the depletion region. x

E max
 Vr
Principle of pn junction photodiode
(a) SiO 2 R Vout
• (a) Reversed biased p+n junction Electrode p+ Iph

photodiode.
h> Eg h+ e–
• Annular electrode to allow photon n
to enter the device. E
Antireflection
• Antireflection coating (Si3N4) to coating
Electrode
W Depletion region
reduce the reflection. rnet
(b)
• The p+-side thickness < 1 μm.
eNd
x
• (b) Net space charge distribution,
within SCL.
• (c) The E field across depletion –eNa
region. E (x)
(c)
x

6
E max
Photodetectors
Principle of the p-n junction Photodiode

 Operation of a p-i-n photodiode.

(b) Energy band diagram under reverse bias.

(a) Cross-section view of a p-i-n photodiode.

(c) Carrier absorption characteristics.

Photodetectors
Principle of the p-n junction Photodiode

 A generic photodiode.
Photodetectors
Principle of the p-n junction Photodiode

 Variation of photon flux with distance.

A physical diagram showing the depletion region.

 A plot of the the flux as a function of distance.

 There is a loss due to Fresnel reflection at the surface,
followed by the decaying exponential loss due to absorption.
 The photon penetration depth x0 is defined as the depth at
which the photon flux is reduced to e-1 of its surface value.
Photodetectors
RAMO’s Theorem and External Photocurrent

 An EHP is photogenerated at x = l. The electron and the hole drift in opposite directions with
drift velocities vh and ve.
 The electron arrives at time telectron = (L-l )/ve and the hole arrives at time thole = l/vh.
V
Iphoto(t)
e vh  e vh e ve 
  
0 L  L L 
iphoto(t)
Semiconductor
Area = Charge = e
e- E telectron
h+
vhole velectron thole

l L-l t
0 l L 0 evh /L eve /L
x i (t)
h+ e–
telectron ielectron(t)

thole thole photocurrent

ihole(t)
t t t
Photodetectors
RAMO’s Theorem and External Photocurrent

 As the electron and hole drift, each generates ielectron(t) and ihole(t).
 The total photocurrent is the sum of hole and electron photocurrents each
lasting a duration th and te respectively.
L- l l
te t   and t h t   Transit time
ve vh
V dx
Work done  e  E dx  V  ie t  dt E ve 
L dt

e ve e vh
ie  t   ; t  te ih  t   ; t  th Photocurrent
L L

Qcollected   ie t  dt   ih t  dt  e
te th The collected charge is not
0 0 2e but just “one electron”.

If a charge q is being drifted with a velocity vd(t) by a field between two biased
electrodes separated by L, the motion of q generates an external current
given by
e v d t 
i (t )  ; t  t transit Ramo’s Theorem
L
Photodetectors
Absorption Coefficient and Photodiode Materials

 Absorbed Photon create Electron-Hole Pair.

1.24
 g [ m]  Cut-off wavelength
E g [eV ] vs. Energy bandgap

 Incident photons become absorbed as they travel in the

semiconductor and light intensity decays exponentially
with distance into the semiconductor.

I ( x )  I 0  e - x Absorption coefficient
 Absorption Coefficient
• Absorption
coefficient α is a
material property.
• Most of the photon
absorption (63%)
occurs over a
distance 1/α (it is
called penetration
depth δ)

13
Photodetectors
Absorption Coefficient and Photodiode Materials

 Absorption
The indirect-gap materials are shown with a broken line.
Photon energy (eV)
54 3 2 1 0.9 0.8 0.7
8
1 10

7 Ge
1 10 In0.7Ga0.3As0.64P0.36
Absorption Coefficient (m-1)

Si In0.53Ga0.47As
6
1 10
GaAs
InP
5
1 10
a-Si:H

4
1 10

3
1 10
0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8
Wavelength (mm)
 Absorption Coefficient
• Direct bandgap semiconductors
(GaAs, InAs, InP, GaSb, InGaAs,
GaAsSb), the photon absorption does
not require assistant from lattice
vibrations. The photon is absorbed
and the electron is excited directly
from the VB to CB without a change
in its k-vector (crystal momentum
ħk), since photon momentum is very
small.
kCB - kVB  photon momentum  0
 Absorption coefficient α for direct bandgap semiconductors
rise sharply with decreasing wavelength from λg (GaAs and
InP). 15
 Absorption Coefficient
• Indirect bandgap
semiconductors (Si and Ge), the
photon absorption requires
assistant from lattice vibrations
(phonon). If K is wave vector
of lattice wave, then ħK
represents the momentum
associated with lattice vibration
 ħK is a phonon momentum.
kCB - kVB  phonon momentum  K
 Thus the probability of photon absorption is not as high as in a
direct transition and the λg is not as sharp as for direct bandgap
semiconductors.
16
 Photodetectors
Absorption Coefficient and Photodiode Materials

Photon absorption in Photon absorption in

a direct bandgap semiconductor. an indirect bandgap semiconductor

E E

CB

EC Indirect Bandgap
CB
Direct Bandgap Eg Photon Eg
EC
EV Photon
EV
VB
VB Phonons
–k k –k k
Photodetectors
Quantum Efficiency and Responsivity

 External Quantum Efficiency

Number of EHP geberated and collected I ph e

 
Number of incidnet photons P0 h

 Responsivity

Photocurrent (A) I ph
R 
Incident Optical Power (W) P0

e e
R   Spectral Responsivity
h hc
Photodetectors

 Responsivity vs. wavelength for a typical Si photodiode.

1
0.9
Responsivity (A/W)

0.8 Ideal Photodiode

0.7 QE = 100% (  = 1)

0.6
0.5
0.4 g
0.3 Si Photodiode
0.2
0.1
0
0 200 400 600 800 1000 1200
Wavelength (nm)
 The pin Photodiode SiO 2
Electrode Electrode
• The pn junction photodiode has two p+

drawbacks:
– Depletion layer capacitance is not i-Si n+
sufficiently small to allow (a)

photodetection at high modulation

frequencies (RC time constant rnet
limitation). eNd
– Narrow SCL (at most a few microns) 
(b)
long wavelengths incident photons are x
absorbed outside SCL  low QE
• The pin photodiode can significantly -eNa
reduce these problems. E(x)
 Intrinsic layer has less doping and wider region (5 – 50 μm).
x
(c)

Eo
20
W
Photodetectors
The pin Photodiode

 Reverse-biased p-i-n photodiode  pin energy-band diagram

 pin photodiode circuit

Photodetectors
The pin Photodiode

 Schematic diagram of pin photodiode

In contrast to pn junction
SiO2 E(x) built-in-field is uniform
Electrode Electrode
p+
x

E0
i-Si n+
W

rnet h > Eg E
eNd –
h+ e
x
Iph
R Vout
–eNa
Vr

 Small depletion layer capacitance gives high modulation frequencies.

 High Quantum efficiency.
Photodetectors
The pin Photodiode

 A reverse biased pin photodiode is illuminated with a short wavelength

photon that is absorbed very near the surface.
 The photogenerated electron has to diffuse to the depletion region
where it is swept into the i- layer and drifted across.

p+ i-Si

Diffusion
h > Eg e– E

Drift
h+

l W

Vr
Photodetectors
The pin Photodiode

p-i-n diode
(a) The structure;

(b) equilibrium energy band diagram;

(c) energy band diagram under reverse bias.

Photodetectors
The pin Photodiode

 The responsivity of pin photodiodes

Photodetectors
Photoconductive Detectors and Photoconductive gain

 Quantum efficiency versus wavelength for various photodetectors.

Photodetectors
The pin Photodiode

 Junction capacitance of pin

 0 r A  Small capacitance: High modulation frequency

Cdep 
W  RCdep time constant is  50 psec.

 Electric field of biased pin  Response time

W
E  E0 
Vr Vr
 t drift 
W W vd
v d  d E

 The speed of pin photodiodes are invariably limited by the transit time of
photogenerated carriers across the i-Si layer.
 For i-Si layer of width 10 m, the drift time is about is about 0.1 nsec.
Photodetectors
The pin Photodiode

 Drift velocity vs. electric field for holes and electrons in Silicon.

105
Drift velocity (m sec-1)

Electron

104
Hole

103

102
104 105 106 107
Electric field (V m-1)
Example
Bandgap and photodetection
(a) Determine the maximum value of the energy gap which a semiconductor, used as a
photoconductor, can have if it is to be sensitive to yellow light (600 nm).
(b) A photodetector whose area is 510-2 cm2 is irradiated with yellow light whose
intensity is 20 mW cm-2. Assuming that each photon generates one electron-hole
pair, calculate the number of pairs generated per second.

Solution

(a) Given,  = 600 nm, we need Eph = h = Eg so that,

Eg = hc/ = (6.62610-34 J s)(3108 m s-1)/(60010-9 m) = 2.07 eV

(b) Area = 510-2 cm2 and Ilight = 2010-3 W/cm2.

The received power is
P = Area Ilight = (510-2 cm2)(2010-3 W/cm2) = 10-3 W
Nph = number of photons arriving per second = P/Eph
= (10-3 W)/(2.0591.6021810-19 J/eV)
= 2.97871015 photons s-1 = 2.97871015 EHP s-1.
Example
Bandgap and Photodetection
(c) From the known energy gap of the semiconductor GaAs (Eg = 1.42 eV), calculate the
primary wavelength of photons emitted from this crystal as a result of electron-hole
recombination. Is this wavelength in the visible?
(d) Will a silicon photodetector be sensitive to the radiation from a GaAs laser? Why?

Solution
(c) For GaAs, Eg = 1.42 eV and the corresponding wavelength is
 = hc/ Eg = (6.62610-34 J s)(3108 m s-1)/(1.42 eV1.610-19 J/eV)
= 873 nm (invisible IR)
The wavelength of emitted radiation due to EHP recombination is 873 nm.

(d) For Si, Eg = 1.1 eV and the corresponding cut-off wavelength is,
g = hc/ Eg = (6.62610-34 J s)(3108 m s-1)/(1.1 eV1.610-19 J/eV)
= 1120 nm
Since the 873 nm wavelength is shorter than the cut-off wavelength of 1120 nm, the
Si photodetector can detect the 873 nm radiation (Put differently, the photon energy
corresponding to 873 nm, 1.42 eV, is larger than the Eg, 1.1 eV, of Si which mean
that the Si photodetector can indeed detect the 873 nm radiation)
Example
Absorption coefficient
(a) If d is the thickness of a photodetector material, Io is the intensity of the incoming
radiation, the number of photons absorbed per unit volume of sample is

I 0 1 - exp(-  d )
n ph 
d h

Solution

(a) If I0 is the intensity of incoming radiation (energy flowing per unit area per
second), I0 exp(- d ) is the transmitted intensity through the specimen with
thickness d and thus I0 exp(- d ) is the “absorbed” intensity
Example

(b) What is the thickness of a Ge and In0.53Ga0.47As crystal layer that is needed for
absorbing 90% of the incident radiation at 1.5 m?
For Ge,   5.2  105 m-1 at 1.5 m incident radiation.
For In0.53Ga0.47As,   7.5  105 m-1 at 1.5 m incident radiation.

(b) For Ge,   5.2  105 m-1 at 1.5 m incident radiation.

 1 - exp(-  d )  0.9
 1 
1 1  1 
d  ln   ln    4.428  10 -6
m  4.428m
  1 - 0.9  5.2  10  1 - 0.9 
5

For In0.53Ga0.47As,   7.5  105 m-1 at 1.5 m incident radiation.

1  1 
d ln    3.07  10 -6
m  3.07 m
7.5  10  1 - 0.9 
5
Example
InGaAs pin Photodiodes
Consider a commercial InGaAs pin photodiode whose responsivity is shown in fig.
Its dark current is 5 nA.
(a) What optical power at a wavelength of 1.55 m would give a photocurrent that
is twice the dark current? What is the QE of the photodetector at 1.55 m?
(b) What would be the photocurrent if the incident power in a was at 1.3 m?
What is the QE at 1.3 m operation?

Responsivity (A/W)
1
0.8
0.6
 The responsivity of an InGaAs
0.4 pin photodiode
0.2
0
800 1000 1200 1400 1600 1800
Wavelength (nm)
Solution
(a) At  = 1.55´10-6 m, from the responsivity vs. wavelength curve we
have R  0.87 A/W. From the definition of responsivity,
Photocurrent ( A) I ph
R 
Incident Optical Power (W ) P0
I ph2 I dark 2  5  10-9 ( A)
we have P0     11.5 nW
R R 0.87 A / W )
From the definitions

of quantum efficiency  and responsivity,
e e
R  
h hc
hcR (6.62  10-34 J  sec)(3  108 m / s )(0.87 A / W )
   0.70 (70 %)
e -19
(1.6  10 coul )(1.55  10 m ) -6

Note the following dimensional identities: A = C s-1 and W = J s-1 so that A W-1 = C J-1.
Thus, responsivity in terms of photocurrent per unit incident optical power is also charge
collected per unit incident energy.
Solution

(b) At  = 1.310-6 m, from the responsivity vs. wavelength curve, R = 0.82 A/W.

Since Po is the same and 11.5 nW as in (a),

I ph  R  P0  (0.82 A / W )(1.15 nW )  9.43 nA

The QE at  = 1.3 m is

hcR (6.62  10-34 J  sec)(3  108 m / s )(0.82 A / W )

   0.78 (78 %)
e -19
(1.6  10 coul )(1.3  10 m ) -6
Photodetectors
Avalanche Photodiode (APD)

 Impact ionization processes

Iphoto R
Electrode SiO2 resulting avalanche multiplication

h > Eg E
e – h+
n+ p  p+ h+
E e–
rnet Electrode

n+ p 
Avalanche region
e-
x Ec
Ev
E(x)
h+

x
Absorption
region
 Impact of an energetic electron's kinetic
Avalanche region energy excites VB electron to the CV.
Photodetectors
Avalanche Photodiode (APD)

 Schematic diagram of typical Si APD.

Electrode
Antireflection coating SiO2

n+ Guard ring n n+
p p n

 Avalanche breakdown 

p+ p+
Substrate Substrate
Electrode Electrode

Si APD structure without a More practical Si APD

guard ring
 Breakdown voltage around periphery is higher and avalanche is
confined more to illuminated region (n+p junction).
Photodetectors
Heterojunction Photodiode
Separate Absorption and Multiplication (SAM) APD

InGaAs-InP heterostructure Separate Absorption and Multiplication APD

Ip h

Electrode R Vout
InP InP Vr InGaAs

E e–
h

h+
E
P+ N n n+
P and N refer to p- and
E (x) Avalanche Absorption
n-type wider-bandgap
region region
semiconductor.
x
Photodetectors
Heterojunction Photodiode
Separate Absorption and Multiplication (SAM) APD

Ec E

(a) Energy band diagram for a SAM

InP
e–
heterojunction APD where there is
a valence band step DEv from
Ev InGaAs to InP that slows hole
DEv Ec entry into the InP layer.
InGaAs
h+
Ev

InP
Ev InGaAsP grading layer (b) An interposing grading layer
(InGaAsP) with an intermediate
InGaAs bandgap breaks DEv and makes it
h+ Ev easier for the hole to pass to the InP
layer.
Photogenerated electron concentration
exp( -  x) at time t = 0

v
de

x
A B
W

h > E E
g

e–
h+

i R
ph

V
r