0% found this document useful (0 votes)
2 views

Chapter 6

The document discusses non-equilibrium statistics related to recombination-generation in semiconductors, covering concepts such as steady-state and transient responses, derivations of relevant formulas, and various recombination mechanisms including direct, indirect, and Auger recombination. It also addresses carrier transport, dynamics of trap populations, and the impact of surface states on recombination rates. Suggested readings are provided for further understanding of the topics discussed.

Uploaded by

林于仁
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
2 views

Chapter 6

The document discusses non-equilibrium statistics related to recombination-generation in semiconductors, covering concepts such as steady-state and transient responses, derivations of relevant formulas, and various recombination mechanisms including direct, indirect, and Auger recombination. It also addresses carrier transport, dynamics of trap populations, and the impact of surface states on recombination rates. Suggested readings are provided for further understanding of the topics discussed.

Uploaded by

林于仁
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 40

6.

Non-Equilibrium Statistics:
Recombination-Generation
• Non‐equilibrium systems
• Recombination generation events
• Steady‐state and transient response
• Derivation of R‐G formula
• Derivation of SRH formula
• Application of SRH formula for special
cases
• Direct and Auger recombination
Suggested readings: Pierret Ch. 5, Neamen Ch. 6
EE 5530 C. F. Huang 1
Current Flow Through Semiconductors

I = GV
= qnvA

n: carrier concentration determined by Quantum Mechanics +


Equilibrium Statistical Mechanics
⇒ Encapsulated into concepts of effective masses and occupation
factors (discussed)

v: carrier velocity determined by Transport with scattering,


non‐equilibrium Statistical Mechanics
⇒ Encapsulated into drift‐diffusion equation with
recombination‐generation (to be discussed)

EE 5530 C. F. Huang 2
Non‐equilibrium Systems
Ch. 7 Ch. 6

Carrier transportation Carrier generation/recombination


Moving in real space Moving in energy space

EE 5530 C. F. Huang 3
Direct Band‐to‐band Recombination

GaAs, InP, InSb (in 3D systems)


Lasers LEDs etc.

EE 5530 C. F. Huang 4
Direct Excitonic Recombination

CNT, Organic electronics, InP, (in 1D systems)


Transistors, Lasers, Solar cells, etc.

EE 5530 C. F. Huang 5
Indirect Recombination
(Trap‐assisted, SRH)

Ge, Si, ….(Indirect bandgap)


Transistors, Solar cells, etc

EE 5530 C. F. Huang 6
Auger Recombination

InP, GaAs, …(high current density)


Lasers, etc.

EE 5530 C. F. Huang 7
Impact Ionization – A Generation
Mechanism

Si, Ge, InP (high electric field)


Lasers, Transistors, Microwave dioides etc.

EE 5530 C. F. Huang 8
Indirect vs. Direct Bandgap

The top & bottom of bands do not align at same wavevector k


for indirect bandgap materials

EE 5530 C. F. Huang 9
Photon Energy and Wavevector

EV + E photon =
EC
Conservation of energy

kV + k photon =
 kC
Conservation of momentum

2π 2π 2π 2π
k photon
= = =
<<
λ in µ m 1.23/ E photon in eV a 5 × 10−4 µ m
a : lattice constant
Photon has large energy for excitation through bandgap, but its
wavevector is negligible compared to size of BZ

EE 5530 C. F. Huang 10
Phonon Energy and Wavevector

EV + E phonon =EC

kV + k phonon =
 kC

2π 2π 2π 2π
k phonon = = ≈ =
λ υ sound / E phonon a 5 ×10−4 µ m

Phonon has large wavevector comparable to BZ,


but negligible energy compared to bandgap

EE 5530 C. F. Huang 11
Localized Traps and Wavevector

phonon
photon photon
Et

2π 2π
ktrap ≈ ≈
a 5 × 10−4 µ m

Trap provides a temporary state and the wave vector


which facilitates indirect transition

EE 5530 C. F. Huang 12
Detailed Balance: Simple Explanation

• 9 in & 9 out of USA


• All numbers are people/unit time.
• The rates of exchange of people (particles) between every
pair of countries (energy levels) is balanced

EE 5530 C. F. Huang 13
Steady‐state (∂/∂t=0)

• 12 in & 12 out of USA


• The rates of exchange of people (particles) between every
pair of countries (energy levels) is NOT balanced

EE 5530 C. F. Huang 14
Detailed Balance, Transient,
Steady‐state

9 in 9 out (∂/∂t≠0) 12 in 12 out


Population conserved Population stabilized

EE 5530 C. F. Huang 15
Detailed Balance, Transient,
Steady‐state

Detailed Balance

Transient

Steady-state

EE 5530 C. F. Huang 16
Indirect Recombination
(Trap‐assisted, SRH)

Ge, Si, ….(Indirect bandgap)


Transistors, Solar cells, etc

EE 5530 C. F. Huang 17
Physical View of Carrier
Capture/Recombination
N=
T nT + pT
N T : total trap number
nT : electron captured trap number
pT : empty trap number

EE 5530 C. F. Huang 18
Carrier Capture Coefficients

1 ∗ 2 3
m υth = kT
2 2
d (# of electrons)
=−n × υth × σ n × (# of empty trap)
dt
dn
⇒ =−n × υth × σ n × pT
dt
≡ − cn pT n
where cn ≡ σ nυth capture coefficient
EE 5530 C. F. Huang 19
Capture Cross‐section

σ n = π r0 2

EE 5530 C. F. Huang 20
SRH Recombination

(1)+(3): one electron reduced from conduction‐band


& one‐hole reduced from valence‐band

(2)+(4): one hole created in valence band


& one electron created in conduction band

EE 5530 C. F. Huang 21
Changes in Electron and Hole
Densities

electron capture electron emission

hole capture hole emission

∂n
−cn n pT + en nT (1 − f c )
=
∂t 1,2
where (1 − f c ) accounts for the empty conduction band states,
en is the emission coefficient
∂p
−c p pnT + e p pT fυ
=
∂t 3,4
assuming f C ≅ 0 and fV ≅ 1 for non-degenerate semiconductors
EE 5530 C. F. Huang 22
Detailed Balance in Equilibrium

∂n ∂p
0
−cn n0 pT 0 + en nT 0 =
= 0
−c p p0 nT 0 + e p pT 0 =
=
∂t 1,2 ∂t 3,4
n0 pT 0 c p p0 nT 0
=en cn ≡ cn n1 ep = ≡ c p p1
nT 0 pT 0
0=
− cn ( n0 pT 0 − nT 0n1 ) 0=
−c p ( p0 nT 0 − pT 0 p1 )

n p pn
Note that n1 p1 = 0 T 0 × 0 T 0 =n0 p0 =ni2
nT 0 pT 0

EE 5530 C. F. Huang 23
Expressions for n1 and p1

Assuming ET is donor-like

NT
nT 0 N=
= T fD
1 β ( ET − EF )
1+ e
gD
n0 pT 0 N T (1 − f D )
=n1 = n0
nT 0 NT ( f D )
β ( EF − Ei ) 1 β ( ET − EF )
n1 = ni e e
gD p1n1 = ni2
1 β ( ET − Ei ) β (E −E ) p1 = n i2 n1
ni e ≡ ni e T ,eff i
gD β ( Ei − ET ) β ( Ei − ET , eff )
= n=g
i D e ni e
n1 i , if ET ,eff
n= Ei

EE 5530 C. F. Huang 24
Dynamics of Trap Population

(1) ∂nT ∂n ∂p (2)


=
− +
∂t ∂t 1,2 ∂t 3,4
= cn n pT − en nT − c p p nT + e p pT
= cn ( npT − nT n1 ) − c p ( p nT − pT p1 )

EE 5530 C. F. Huang 25
Steady‐state Trap Population

∂nT
=0= cn ( npT − nT n1 ) − c p ( pnT − pT p1 )
∂t
cn N T n + c p N T p1
nT =
cn ( n + n1 ) + c p ( p + p1 )
remember pT + nT =
NT

EE 5530 C. F. Huang 26
Net Rate of Recombination‐Generation

dp
rp = − = c p ( p nT − pT p1 )
dt
np − ni2
=
 1   1 
  ( n + n1 ) +   ( p + p1 )
c N
 p T c N
 n T
τp τn

rp= rn ≡ R
R = 0 in equilibrium.
Trap level is described by n1 and p1. Usually the mid-level traps dominate.
27

EE 5530 C. F. Huang
Case 1: Low‐level Injection in p‐type

p0 >> ∆n, p0 >> n0


np − ni2
R=
τ p (n + n1 ) + τ n ( p + p1 )
(n0 + ∆n)( p0 + ∆n) − ni2
=
τ p (n0 + ∆n + n1 ) + τ n ( p0 + ∆p + p1 )
∆n( n0 + p0 ) + ∆n 2
=
τ p ( n0 + ∆n + n1 ) + τ n ( p0 + ∆p + p1 )
∆n( p0 ) ∆n
= = = cn NT ∆n
τ n ( p0 ) τ n
τ n : minority carrier lifetime under low-level injection
∆n = ∆p to maintain charge neutrality

EE 5530 C. F. Huang 28
Case 2: High‐level Injection

∆n >> p0 , p0 >> n0
e.g. organic solar cells,
np − ni2 thyristors
R=
τ p (n + n1 ) + τ n ( p + p1 )
( n0 + ∆n )( p0 + ∆p ) − ni2
=
τ p (n0 + ∆n + n1 ) + τ n ( p0 + ∆p + p1 )
∆n( n0 + p0 ) + ∆n 2
=
τ p ( n0 + ∆n + n1 ) + τ n ( p0 + ∆n + p1 )
∆n 2 ∆n
=
(τ n + τ p ) ∆n (τ n + τ p )

∆n = ∆p to maintain charge neutrality

EE 5530 C. F. Huang 29
High/Low Level Injection …

∆n
Rhigh = ∆n >> p0
(τ n + τ p )

∆n
Rlow = ∆n << p0
τp

which one is larger and why?

EE 5530 C. F. Huang 30
Case 3: Generation in Depletion
Region
p << p1 , n << n1

np − ni2
R=
τ p ( n + n1 ) + τ n ( p + p1 )
−ni2
=
τ p (n1 ) + τ n ( p1 )

β ( E −E )
=p1 ni e= i if ET ,eff
n= Ei
i T , eff

ni 2
and =
n1 = ni
p1
−ni
then R =
τ p +τn

EE 5530 C. F. Huang 31
Band‐to‐band Recombination

=R B(np − ni2 )

Direct recombination at low‐level injection

( ∆n = ∆p ) << p0
2 ∆n
=R B ( n0 + ∆n )( p0 + ∆p ) − n  ≈ Bp0 =
i × ∆n
τ direct
Direct generation in depletion region
n, p ~ 0
R B( np − ni2 ) ≈ − Bni2
=

EE 5530 C. F. Huang 32
Auger Recombination
Involving two electrons and one hole

R cn n( np − ni2 ) + c p p( np − ni2 )
=

cn , c p ~ 10−29 cm 6 / sec

Auger recombination at low‐level injection

( ∆n = ∆p ) << ( p0 = N A in extrinsic temp. range)

∆n 1
c p N A2 ∆n
R ≈= where
= τ auger
τ auger c p N A2

EE 5530 C. F. Huang 33
Effective Carrier Lifetime

R = RSRH + Rdirect + RAuger


 1 1 1 
∆n 
= + + 
τ τ τ
 SRH direct Auger 

∆n(cn NT + BN D + cn ,auger N D2 )
=
τ eff = (cn NT + BN D + cn ,auger N D2 ) −1 (assuming low level injection)

EE 5530 C. F. Huang 34
Effective Carrier Lifetime with all
Processes

EE 5530 C. F. Huang 35
Atomic configuration of Surface States

Dangling bonds

“God made the bulk; the surface was invented by the devil”
Wolfgang Pauli, Nobel Prize in Physics, 1945

EE 5530 C. F. Huang 36
Surface States

EE 5530 C. F. Huang 37
Multiple Levels of Surface States

Physically apart

EE 5530 C. F. Huang 38
Surface Recombination Rate

For single level traps in the bulk


np − ni2
Rbulk = NT DIT
1 1
( n + n1 ) + ( p + p1 )
c p NT cn N T
( np − ni2 ) N T
=
1 1
( n + n1 ) + ( p + p1 )
cp cn
For single level interface traps at E
( ns ps − ni2 ) DT ( E )dE
R( E ) =
1 1
( ns + n1s ) + ( ps + p1s )
c ps cns
Where DT is number of traps per area per unit energy

EE 5530 C. F. Huang 39
Surface Recombination Velocity
∆E around midgap, most important
It can be shown that by
using the conclusion of
low level injection
EF

R≈ ∫c
EF '
ps ∆ps DIT ( E )dE

= c ps DIT ( EF − EF' )∆ps


= c ps DIT ∆E ∆ps
≡ sg ∆ps
where sg is defined as surface velocity

EE 5530 C. F. Huang 40

You might also like