Semiconductor Photonics

THE INTERACTION OF RADIATION AND MATTER: SEMICLASSICAL THEORY PAGE 82
VII. SEMICONDUCTOR P HOTONICS

A. P RELIMINARIES : SEMICONDUCTOR BACKGROUND 29
THE CRYSTAL H AMILTONIAN
For an assembly of atoms the classical energy is the sum of the following:
• the kinetic energy of the nuclei;
• the potential energy of the nuclei in one another's electrostatic field;
• the kinetic energy of the electrons;
• the potential energy of the electrons in the field of the nuclei;
• the potential energy of the electrons in one another's electrostatic field;
• the magnetic energy associated with spin and orbital variables.
Dividing the electrons into core and valence electrons and leaving out magnetic effects
leads to the following expression for the crystal Hamiltonian:
r r r r
r r
H =∑
p 2α
α 2M α
(
+ ∑U R α − R β + ∑
α, β
p 2l
l 2ml
) (
e 2 4 πε
+ ∑V rl − R α + ∑ r r 0
l, α l, m rl − rm
) [ VII-0 ]
r
where α and β label the ions, l and m label the electrons, p is the momentum, M is an
r r
( )
ionic mass, m is the mass of an electron, U R α − Rβ is the interionic potential, and
r r
( )
V rl − R α is the valence-electron-ion potential.
The quantum mechanics of the assembly is treated to a good approximation by taking

the total wavefunction of the system as the product
r r r
Ξ= Ψ r; R Φ R( ) ( ) [ VII-1 ]
29 This discussion draws heavily on B. K. Ridley, Quantum Processes in Semiconductors (3rd edition), Clarendon
Press (1993).
R. Victor Jones, April 4, 2000
r r r r r
( ) ( )
where Φ R = Φ R α , Rβ , Rδ , R γ ,K is the wavefunction of all the ions and
r r r r r r
( ) ( )
r r r r
Ψ r; R = Ψ rk , rl , rm , rn ,K; R α , Rβ , Rδ , R γ ,K is the wavefunction of all the
electrons at the instantaneous ionic positions. The Schrödinger equation is then written
r r r r r r
( ) ( )
Ψ r; R H lattice Φ R +Φ R H elec Ψ ( ) ( ) r; R
[ r r
( )]
r r r r r r r [ VII-2 ]
+ H lattice Ψ r; R Φ R ( ) ( ) ( ) ( ) ( )
− Ψ r; R H lattice Φ R = E Ψ r; R Φ R
where the total Hamiltonian is parsed into two independent components -- viz.
r r r
H lattice =∑
p 2α
α 2M α α, β
(
+ ∑U R α − R β . ) [ VII-3a ]
r
r r
H elec = ∑
p 2l
l 2ml
(
e 2 4 πε
+ ∑V rl − R α + ∑ r r 0 .
l, α l, m rl − rm
) [ VII-3b ]
The essential assumption of the adiabatic approximation is that the bracketed

term is negligible and that the global problem may be treated as two independent
problems -- viz. r r
( )
H lattice Φ R = Elattice Φ R ( )
[ VII-4a ]
r r r r
H elec ( )
Ψ r; R = Eelec Ψ r; R ( ) [ VII-4b ]
As a further refinement, the electron problem must be parsed once more as
H elec = {H elec }static + {H elec }dynamic

where {H elec }static defines the problem of the many electron system interacting with the
static ionic lattice and {H elec }dynamic incorporates the effects of the electron-phonon
interaction.
LATTICE VIBRATIONS
For a brief overview of lattice vibrations see the webpage entitled Self Study Materials
on Solid State Electronics 2 at
http://deas.harvard.edu/courses/ap216/lectures/ls_2/ls2_u7/sse_tut_2/solid2.html
B LOCH ELECTRONS - Solutions of the Schrödinger equation for a single

electron or quasi-particle moving in a periodic lattice
If the electron-electron is averaged out, deviations from the average may be treated
perturbations -- i.e., we make the replacement
∑ err 4− πε
2
r 0
⇒ (H ) + (H
elec 0 )
elec 1 [ VII-5 ]
l, m r l m
where (H )
elec 0 leads to a constant replusive electronic energy component and (H )
elec 1
is a small fluctuating electron-electron interaction. If we neglect these fluctuations, each

electron interacts independently with a screened lattice potential. In this approximation,
the electronic wavefunction can be expressed as
r r v r
( )
Ψ r; R 0 = ∏ ψ n rn ; R 0
n
( ) [ VII-6 ]
with the requirement that the occupation of one-electron states is in accordance with
Pauli exclusion principle. Thus, each the wavefunction for each electrons satisfies a
Schrödinger equation in the form

v r v r v r v r
(H ) n
elec static ( )
ψ n rn ; R 0 = −
2m
( ) ( ) ( )
∇ n ψ n rn ; R 0 + V (r) ψ n rn ; R 0 = ε n ψ n rn; R 0 [ VII-7a ]
h2 2 v
or simplifying the notation
∇ ψ (r) + V ( r) ψ ( r) = ε ψ ( r;)
h2 2 v v v v
− [ VII-7b ]
2m
S CHEMATIC P ERIODIC LATTICE P OTENTIAL
Energy Energy
Position Position
Single ion potential

"Free carrier" energy
Bound state energy
L
+a +a +a + a +a + a +

The lattice potential

v far from the surface of the crystal has the property that, for all
v v v
lattice vectors R [l ] = l 1 a1 + l 2 a 2 + l 3 a 3
v
(
v v
V ( r) = V r + R [ l ] ) [ VII-8 ]
Bloch's (Floquet's) theorem : The function

v v
v −1 v
(
ψ vk ( r) = (V) 2 uvk( r) exp i k ⋅ r ) [ VII-9 ]
v v
where urk ( r) is a function with the same spatial periodicity as V ( r) that is an
eigenfunction of Equation [ VII-7 ].
Proof of theorem:
First rewrite Equation [ VII-7b ] as
H ( r) ψ( r) = ε ψ( r)
v v v
[ VII-7b' ]
Shifting the origin by a lattice vector, we again rewrite Equation [ VII-7b ] as
(
v v
) (v v
) v v
(
H r + R [ l ] ψ r + R [ l] = ε ψ r + R [ l ] ) [ VII-7b" ]
In light of periodicity expressed in Equation [ VII-8 ], it follows that
(
v v
) v v
(
H ( r) ψ r + R [ l ] = ε ψ r + R [ l ]
v
) [ VII-10 ]
Thus, any linear combination of the possible eigenfunctions

(
v v
)
ψ r + R [ l ] i s a valid eigenstate of the energy ε ! In particular, let us
choose the combination

v v
( ) ( )
v v v
ψ vk( r) = ∑ ψ r + R [ l ] exp − i k ⋅ R [ l ]
[l ]
) ∑ ψ ( vr + R [ ] ) exp[− i k ⋅ ( vr + R [ ] )]
v v v v v [ VII-11 ]
(
= exp i k ⋅ r l l
[l ]
v
where k is, for the moment, taken to be an arbitrary complex vector. Since the
v
summation includes all possible lattice vectors it must be a periodic function of r
v
with the period of the lattice and may be identified with uvk( r) of Equation
v
[ VII-9 ]. 30 If we impose cyclical boundary conditions on ψ vk( r) , we see that, of
v
necessity, k is a real vector! QED
In light of Bloch's theorem, Equation [ VII-7b ] may be rewrite in the form
v v v  h2 v v 2 v 
H v
k ( r) uk( r) = −
 2m
( 
v
)
∇ + i k + V ( r)  uvk( r) = ε vk u vk( r) .
v
[ VII-12 ]
( )
v v
Since the reciprocal lattice vectors of crystal are defined31 so that exp i R [ l ] ⋅ G [ m ] = 1 ,
any function written in the form
( )
v v
∑C[ m] exp i G [ m ] ⋅ r
[ m]
v v
( ) [ ( )]
v v v v
30 We keep for the record uvk( r) ⇒ ∑ ψ r + R [ l ] exp − i k ⋅ r + R [ l ]
[ l]
v v v v
31 The reciprocal lattice vectors: G[ m ] = m1 b1 + m2 b2 + m3 b3
where v v v v v v
v a 2 × a3 v a3 × a1 v a1 × a2
b1 = 2π v v v ; b2 = 2π v v v ; b3 = 2π v v v
a1 ⋅ a2 × a 3 a 1 ⋅ a2 × a 3 a1 ⋅ a 2 × a 3
v
has the spatial periodicity of the lattice. Conversely, we are allowed to expand V ( r)
v
and uvk (r ) in Fourier series -- viz.
( )
v v v
V ( r) = ∑V[ m ] exp i G [ m ] ⋅ r [ VII-13a ]
[m ]
∑ u[ ] ( k) exp ( i G [ )
v v v v
uvk( r) = m′ m ′] ⋅r [ VII-14a ]
[ m ′]
∫∫∫ d vr V ( vr) exp(− i G ⋅ vr)

v
where V[ m ] = V-1 [ m ′]
[ VII-13b ]
( k) = V ∫∫∫ d vr u ( vr) exp (− i G ⋅ vr)

v -1
v
and u[ m ] v
[m ]
[ VII-14b ]
k
Substituting these expressions into Equation [ VII-12 ] and equating various Fourier
components, we obtain the following infinite set of algebraic equations:
(
h2 v
)
v v v v
( ) ( )
u [m ] k + ∑V[ m ′] u[ m ] −[ m ′] k = ε vk u[ m ] k ( )
2
G[m ] + k [ VII-15 ]
2m [ m ′]
v
which can, in principle, be solved for ε v
k ()
and u[ m ] k . If we were to turn off the
lattice potential off to effectuate the e m p t y l a t t i c e approximation, the
eigenfunctions and eigenvalues should, obviously, be those of a free electron and
given by
v v
v v
ψ vk( r) ⇒ψ 0vk( r) = (V) 2 exp i k ⋅ r
−1
( ) [ VII-16 ]
v2
h2 k
and ε v
k
⇒ ε vk0 = [ VII-17 ]
2m

1D E MPTY LATTICE MODEL

extended zone scheme reduced zone scheme
With gradual restoration of the lattice potential, the wave function is gradually
transformed from a plane wave, Equation [ VII-16 ], to a Bloch function, Equation
v
[ VII-9 ]. In general, the value of k is well defined since it does not change from its
v
original value and the plane wave merely becomes modulated by the function uvk (r ) .
Clearly, it is also possible to write the Bloch wave function in the form
v
ψ vk( r) = (V) 2
−1
{ v
(
v
)} [ (
v v v
) ]v
uvk( r) exp i G [ m ] ⋅ r exp i k − G [ m ] ⋅ r [ VII-18 ]
v
v
( v
)
and since both uvk (r ) and exp i G[ m] ⋅ r are spatially periodic, so is their product.
Thus, the wave function canvbe vconsidered as obeying the Bloch theorem with any
empty lattice wave vector k − G [ m] . Accordingly, the theorem allows for two
alternative modes of classifying solutions -- i.e. a solution may be specified with the
v
original value of k (extended zone scheme) or with that value "reduced" by a reciprocal
lattice vector plus a band index (reduced zone scheme).
0 v
In the nearly free electron approximation we use the ψ vk( r) functions as a basis
v
set and take the periodic potential V ( r) as a small static perturbation. Using Equations
[ A-6c ] and [ A-7b ] of this set of lecture notes (at the end of the Section II, Review of
Basic Quantum Mechanics: Dynamic Behavior of Quantum Systems) we obtain
ε V ε
v0 v0
ε 1
≈ ε v0 +∑ ε v0 k′ k
ε −ε
v [ VII-19 ]
k k k′ v
0
v
0
k′ k
ε ε
2
v0 V v0
ε ≅ ε + ∑ ε −ε
v
k
v0
k
k′
0
v
0
v
k
[ VII-20 ]
k′ k
In light of Equation [ VII-13a ] we may write
∫∫∫ d vr exp (− i k ′ ⋅ vr) V ( vr) exp ( i k ⋅ vr)

v v
ε v0
k′
V ε v0
k
= (V )
−1
= (V) ∫∫∫ d r exp [− i ( k ′ − k) ⋅ r] ∑V exp ( i G ⋅ r) [ VII-21 ]

v−1 v v v v v
[m ] [m ]
[m ]
= V δ (k + G − k ′)
v v v
[m ] [m ]
so that Equations [ VII-19 ] and [ VII-20 ] become
V
ε ≈ ε v0 +∑ ε v0 v
[m ]
ε −ε
v
k k k+ G [ m ] 0
v v
0
v
[ VII-22 ]
[m ] k+ G [ m ] k

2
V[ m ]
ε ≈ε +∑ ε v0
−ε
v [ VII-23 ]
k k 0
v v
0
v
[ ] m k+ G [ m ] k
Clearly, the strongest departure from free electron behavior occurs when the
denominator vanishes -- i.e. when
h2  v v 2 v 2
ε v0 v −ε =
v0 G + k − k 
2m  
k+ G [m ] k [m ]
. [ VII-24 ]
h2  v 2 v v = 0
= G + 2G ⋅ k
2m  
[ m ] [ m ]
Equation[ VII-24 ] is the famous Bragg scattering condition or the condition

which defines the boundaries of the Brillouin zones. At the zone edge for a particular
pair of degenerate free electron states the appropriate pair of equation from the set in
Equation [ VII-19 ] may be approximated
v v
[ ] ( )
ε vk0 −ε vk u[ 0] k + V[ m ] u[ m ] k = 0 ( )
[ ]
v v
( ) ( )
[ VII-25a ]
ε vk0+ Gv [m ] −ε vk u[ m ] k + V[∗m ] u[ 0] k = 0
or
1  v0 v
± (ε ) +4 V

ε ε + ε vk0 −ε
2 2
v = v0 v v0
[m ] . [ VII-25b ]
k
2  k+ G [m ] k+ G [ m ] k


1D N EARLY F REE ELECTRON MODEL

extended zone scheme reduced zone scheme

These considerations then provide the context for the examination of real band
structures.
Diamond Structure Brillouin Zone and Nomincature

For more on band structures go to:

http://www.deas.harvard.edu/courses/ap216/images/bandgap_engineering/bandgap_engineering.html
For more on pn junction physics go to:

http://www.deas.harvard.edu/courses/ap216/images/pn_junction/pn_junction.html

B . OPTICAL P ROPERTIES OF B ULK (3-D) S EMICONDUCTOR 32

F REE -C ARRIER THEORY
For most of our considerations a parabolic-band model provides a reasonable
picture of optical properties. In this model we assume that near a band edge the single
particle energy of the electron measured with respect to the edge is given by
v 2
h2 k
ε v = =
h2 k 2
[ VII-26a ]
k
2m∗ 2m∗
where the e f f e c t i v e m a s s m∗ is a measure of the inverse curvature of the band.33

Introducing the notion of a deficiency of electrons in the valence band -- i.e. holes --
as positive charge carriers, the parabolic-band model is re-interpreted as, respectively,
the quasiparticle energies of an electron in the conduction band and a hole in the valence
band
v 2
h2 k
εe , vk = 2m = h2mk
2 2
[ VII-26b ]
e e
v 2
h2 k
ε v
h, k
= =
h2 k 2
[ VII-26c ]
2mh 2mh
32 This section draws heavily on Weng W. Chow, Stephan W. Koch and Murray. Sargent III, Semiconductor-
Laser Physics, Springer-Verlag (1994)
33 That is, the effective mass is given by
1 ∂2 v
1
= ε
m∗ h2 ∂k 2 k Band edge
and, thus, in the c o n d u c t i o n b a n d t h e e f f e c t i v e m a s s o f e l e c t r o n s i s p o s i t i v e a n d i n the
valence band it is negative -- i.e. mc > 0 and mv < 0 .
where me = mc > 0 and mh > 0 accounts for aggregate dynamics of all the electron in
the valence minus the single empty state.34 Of course, the single particle picture
discussed thus far, does not include the critically important effects of the inter-electronic
Coulomb interactions and, in particular, carrier-carrier scattering. In what is
usually called free-carrier theory, it is assumed that carrier-carrier scattering
causes a rapid "thermalization" (relaxation times less than 0.1 picoseconds) of excited
conduction band electrons (and valence band holes) and, consequently, it is assumed
that carriers within a band are in quasiequilibrium with energies distributed
according to a Fermi-Dirac distribution (see illustration on the next page).35,, 36
Accordingly, the carrier density for a given band (i.e. α= e for the conduction band
and α= h for the valence band) is determined by the condition
−1
Nα =
1
∑
V states
fα vk =
1
∑ exp β

V states   (ε v
αk )
− µ α  +1
 
[ VII-27 ]
where V is the volume of the sample and µα is the carrier quasichemical

potential (or imref ≡ quasi-Fermi energy).37 In the discussion of Equation
34 The hole energy is the energy of the filled valence minus the energy of the filled with a vacant electronic state so
that an increase in hole momentum leads to an increase in hole energy. Since mh must of necessity include the
many-body Coulomb interactions among the valence electrons.
35 See a Derivation of the Fermi-Dirac distribution function at http://ece-www.colorado.edu/~bart/book/fermi-
dirac_derivation.htm
36 See the applets Fermi Level vs. Carrier Concentration and Doping of Donor and Acceptor Impurities
(at http://www.acsu.buffalo.edu/~wie/applet/fermi/fermi.html)
and Fermi Level, Fermi Function and Electron Occupancy of Localized Energy States
(at http://www.acsu.buffalo.edu/~wie/applet/fermi/functionAndStates/functionAndStates.html).
37 As are the electron and hole energies ε v
αk, the imrefs µα are measured with respect to the appropriate band
edge -- viz.
µ e = (EF )C − EC and µ h = EV − ( EF )V
Thus, if µα < 0 the band does contain enough carriers to populate any state with a probability greater than one-half.
v
[ IV-5 ] of this set of lectures, it was argued that the 3-D density of states in k space is
given by
V 2
ρ (k) dk = k dk [ VII-28 ]
π2
which translates (within a given band) into a density of states per unit energy per unit
volume of
(ε )
3
1  2mα  2
ρ
1
= ρ( k)
dk
= ε
dε α vk 2π 2  h2 
v v
αk αk [ VII-29]
V
Hence, for a 3D semiconductor, Equation [ VII-27 ] becomes
∞ ∞
∫ ∫
3
ρ (ε α ) f (ε α ) dε α =
1  2mα  2
ε { exp [ β(ε α − µ α )] } dε
−1
Nα = [ VII-30 ]
2π 2  h2  α α
0 0

OPTICAL MATRIX ELEMENTS

Since optical interactions in a semiconductor are essential distributed, it is probably
more precise and appropriate in this case to write the interaction Hamiltonian as 38
q v v e v v
[ ( )]
v v v
[eˆ ⋅ P ] exp i ω rad t − k rad ⋅ r + K
e A0
H I =− A ⋅P = A ⋅P ⇒ [ VII-31 ]
m m 2m
Thus, the critical optical matrix element 39 between a state in the valence band and one
in the conduction band is
ε v
vk
H I ε v
c k′
=
e A0
exp [ i ω rad t] M v vk, c vk ′ [ VII-32a ]
2m
where the so called transition matrix element is given by
∫ d vr u∗ ( vr) exp[− i k ⋅ vr] {eˆ ⋅ Pv exp[− i k ⋅ vr]} u

v v v v
( r) exp [ ]
v
M v vk, c vk ′ ≡ v
vk rad
v
c k′
i k′⋅ r [ VII-32b ]
Using the expansion for un kv in Equation [ VII-12a ] the matrix element becomes
v v
Mv kv, c kv ′ = ∑ ∑ uv∗[ m ](k) uc [ m ′] ( k′ )
[ m ] [ m ′]
[ VII-33 ]
[ ( ) ] [( ) ]
v v v v v v
∫∫∫
v v v
× d r exp −i k + k rad + G [ m] ⋅ r {eˆ ⋅ P } exp i k ′ + G [m ′] ⋅ r
38 In the next set of lectures, The Interaction of Radiation and Matter: Quantum Theory, we develop the
following expression for the nonrelativistic Hamiltonian of a single charged particle:
r
[ ]
1 r 2
H = qϕ+ P − qA
r 2m
where P is the canonical conjugate momentum of the charged particle.
39 The Fermi Golden Rule and first-order perturbation theory tells us that the radiation-induced transition
rate is given by
∑ 2πh f H int i δ(ε f − ε i ) .

f
v v
However, P ⇒ − i h ∇ so that
{ ( v v
M v vk, c vk ′ = ∑ ∑ h eˆ ⋅ k ′ + G [ m ′]
[ m ] [ m ′]
)} u∗ ( k) u ( k ′)
v [m ]
v
c [ m ′]
v
[ VII-34 ]
∫∫∫ [ ( ) ]
v v v v v v v
× d r exp − i k + k rad − k ′ + G [ m ] − G [ m ′] ⋅ r
v v v v
v
With the parsing r = R[ l] + ρ and using the defining relationship exp i R[ l] ⋅ G[ m ] =1, ( )
the volume integral may be expressed
∫∫∫ d vr {vr} exp [ i ( k ′ − k − k + G − G ) ⋅ vr]

v v v v v
rad [ m ′] [m ]
= ∫∫∫ d r {ρ } exp [ i ( k ′ − k − k + G − G ) ⋅ ρ ]
v v v v v v v v
rad [ m ′] [m ] [ VII-35 ]
unit cell
× ∑ exp [ i ( k ′ − k − k ) ⋅ R ]
v v v v
rad [l ]
[ l]
∑ exp [ i ( k ′ − k − k ) ⋅ R [ ] ] ⇒ 0 unless
v v v v
The summation rad l
[ l]
v v v v
k ′ = k + k rad ≈ k . [ VII-36 ]
-- i.e. essentially only vertical transitions are allowed! Therefore,
V 
{ ( v v v
)}
v
M v vk, c vk ′ ⇒δ vk vk ′  sample  ∑ ∑ h eˆ ⋅ k + G [ m ′] u∗v [ m ] k u c [ m ′] k
 Vunit cell  [ m ] [ m ′]
( ) ( )
.[ VII-37a ]
∫∫∫ [( ) ]
v v v v
× d r exp i G [ m ′] − G [m ] ⋅ ρ
unit cell
For a given polarization of the applied field, the transition matrix can be written

M v vk, c vk ′ = {polarization factor} M . [ VII-37b ]

2
where the polarization factors are of order one and typical values of M are given in
the following table:40
Material system 2 M m
2
(in eV )
Ga As 28.8
Alx Ga1-x As (x < 0.3) 29.83 + 2.85 x
Inx Ga1-x As 28.8 - 6.6 x
In P 19.7
In1-x Gax Asy P1-y (x = 0.47 y) 19.7 + 5.6 y
Of course, the radiation must also satisfy the energy conservation condition
∆ ε = h ω = εg = εg +
h2 k2 h2 k 2 h2 k 2
v
k
v
k + + [ VII-38a ]
2me 2mh 2mr
1 1 1
where = + . [ VII-38b ]
mr me m h
"FREE - CARRIER THEORY " OF OPTICAL P ROCESSESS

The bottom line is that in the free-carrier theory of optical interactions, the
effective Hamiltonian for the carriers -- i.e. "free" particle kinetic energy plus
v
electromagnetic interaction -- is separable into a series of k -dependent terms.41 Thus,
40 From Diode Lasers and Photonic Integrated Circuits by Larry A. Coldren and Scott W. Corzine, Wiley (1995)
41 In particular, it can be shown -- see Weng W. Chow, Stephan W. Koch and Murray. Sargent III,
Semiconductor-Laser Physics, Springer-Verlag (1994) -- that
vv
if we neglect k k ′ correlations in the treatment of optical properties, we need only
consider the following manifold of states:
Equation [ VII-37a ] tells us that the absorption or emission of a single

v photon connect
v
only the two states in the manifold which can be identified as 0e 0 h : k and 1e 1h : k
where the first number in the ket specifies electron occupancy of a state with
v v
momentum h k and the second hole occupancy of a state with momentum − h k .
  h2 k 2  †v v  h 2 k 2  †v v 
H eff = ∑
k 
v


ε g + ∗
2m ec 
 a k a k +  ∗ 
 2mhv 
b − k b− k − µ[v a †v b †v + µ∗
k k − k k k − k ]
v a v b v E( z, t )

{ } { }
where a vk, akv and bvk , bkv are, respectively, electron and hole {creation, destruction} operators.
† †
µvk is
the dipole matrix element between vertical states in the valence and conduction bands.
E NERGY D IAGRAM FOR B ULK L ASERS
Electron Energy
vv
Since no k k ′ correlations are involved in the free carrier theory, the density operator of
the complete system may be expressed as a product of component density operators
v
ρ= ∏ v
ρ( k ) [ VII-39 ]
k
so that the Schrödinger equation of motion for the density operator is also separable -- viz.
[ ]
v v v
ρ˙ (k) = − H eff (k), ρ( k) .
i
[ VII-39 ]
h
C. INJECTION LASER THEORY
(For pictorials on laser configurations go to:

http://www.deas.harvard.edu/courses/ap216/images/injection_lasers/laser_structures/laser_structures.html)
A fairly satisfactory model of lasing (and other optical processes) in semiconductors

may be obtain by adaptation of the two-level, semiclassical discussed earlier. In light
of the discussion in the previous section, we adapt the two-level theory by making the
identifications
v v v v v v v
( ) ( )
ρ aa k ⇒ρ1 1 1 1 k = 1e 1h : k ρ 1e 1h : k = 3:k ρ 3:k = ρ 3 3 k ( )
v v v v v v v
( ) ( )
ρ bb k ⇒ρ 00 00 k = 0e 0h : k ρ 0e 0h : k = 0:k ρ 0:k = ρ 00 k () [ VII--40 ]
v v v v v v v
( ) ( )
ρ ab k ⇒ρ1 1 0 0 k = 1e 1h : k ρ 0e 0h : k = 3:k ρ 0:k ==ρ 3 0 k ()
Given the free carrier effective Hamiltonian discussed above and the equation of
motion expressed in Equation [ VII-31 ], we can write
v v v v
( )
ρ˙ 30 k = ( )
3:k ρ˙ k 0:k
[ ] ()
v v v v v v v
=−
i
h
( ) ( )
3:k H eff k 3:k − 0:k H eff k 0:k ρ 30 k
[ VII--41a ]
[ () ( )]
v v v v v
+
h
i
( )
3:k H eff k 0:k ρ 33 k − ρ 00 k
[ () ]
v v v v v v v v
( )
ρ˙ 33 k = ( )
3:k ρ˙ k 3:k =
i
h
ρ 30 k 0:k H eff k 3:k ( )
[ VII--41b ]
[ () ]
v v v v
−
h
i
ρ 03 k 3:k H eff k ( ) 0:k

[ () ]
v v v v v v v v
( )
ρ˙ 00 k = 0:k ρ˙ k( ) 0:k = −
i
h
ρ 30 k 0:k H eff k 3:k ( )
[ VII--41c ]
[ () ]
v v v v
+
i
h
ρ 03 k 3:k H eff k ( ) 0:k
which tells us, once again, that ρ ab ⇒ ρ 30 is driven by ρ aa − ρ bb ⇒ ρ 33 − ρ 00 and vice

versa. There is, however, an important new element to be considered in this problem to
wit, as we have illustrated above, we are, in
v fact, actuallyv dealing with a four-level
system which includes of the states 1e 0h : k and 0e 1h : k as well as the pseudo "a"
and "b" states. Fortunately, inclusion of these states does not unduly complicate the
analysis since, as discussed above, electron-electron scattering induces a rapid relaxation or
"thermalization" of the probability of finding a given state occupied. After thermalization,
the probability of finding a state with a particular momentum value in a given band is
provided by a Fermi-Dirac distribution referenced to a quasi-Fermi energy or imref
appropriate to that band. To obtain an expression for ρ aa − ρ bb we note that42
[ ( ) ] = ρ ( k) +ρ ( k) +ρ ( k) +ρ ( k) = 1
v v v v v
tr ρ k 00 11 22 33 [ VII--42 ]
so that
( ) { ( ) } + { ρ ( k) +ρ ( k) } − 1
v v v v v v
( )
ρ aa − ρ bb = ρ 33 k − ρ 00 k = ρ 33 k +ρ 22 k ( ) 33 11
v v [ VII--43 ]
( ) ( )
= n e k + nh k − 1
v v
where ne (k ) is the probability43 of finding a electron with momentum h k independent of
v v
whether or not there is a hole with momentum − h k and nh (k) is the corresponding
r
42 That is, there must be either 0 or 1 electrons and 0 or 1 holes in the given k state.
43 In the second quantization representation of the problem, the n’s are expectation values of the appropriate
number operators.
probability for a hole. Adapting Equations [ VI-13a-c ] from earlier in this set of lecture
notes, 44 we may write
[ ( ) ( ) ]
v v v v v v v
( ) ( )
n˙e k = Λ e vk − γ nr ne k − Γvk ne k nh k − n˙e k () ( ) ( ) c-c
− i h−1 Vab z ,t; k ρ ab k + c.c [ VII--44a ]
n˙ ( k) = Λ − γ n ( k) − Γ n (k) n ( k) − n˙ ( k) − [ i h V ( z , t; k) ρ ( k) + c.c]
v v v v v v v −1
v v
h hk nr h k e h h ab ab [ VII--44b ]
c-c
ρ˙ ( k) = − ( i ω + γ ) ρ ( k) + i h V ( z ,t; k) [ n ( k) + n ( k) − 1]
v v v v
−1
v
ab ab ab ab ab e h [ VII--44c ]
v
where Λ α vk ( α= e or h ) represents the pumping rate due to carrier injection and n˙α k c-c
v v v
( )
( )
the effect of carrier-carrier scattering. γ nr nl k and Γvk ne k nh k are, respectively, ( ) ( )
phenomenological representations of nonradiative decay and radiative “bimolecular”
recombination (spontaneous emission), respectively. The probability difference (gain
factor)
v v v v v   1  1 
( ) () ( )
∆ k ≡ ne k + n h k − 1 = ne k nh k 1− 
n k
( ) ( )
v − 1 
n
v − 1 
k ( ) ( ) [ VII--45 ]
  e  h 
is the critical factor in the analysis of stimulated processes -- i.e. for inversion
[ ]
v v v v
∆ (k) > 0 . This gain factor may be written as ∆(k) = ne( k) − 1 − nh( k) -- i.e. the
population inversion is proportional to the probability difference of an electron in
r
corresponding k -states in the conduction and valence bands. Alternatively and more
usefully, we see that it varies directly with the product of
44 That is [
ρ˙ a a = λ a − γ a ρ a a − i h−1 V ab ρ a b + c.c. ]
+ [i h + c.c.]
−1
ρ˙ b b = λ b − γ b ρ b b V ab ρ a b
( )
ρ˙ a b = − i ω ab + γ a b ρ ab + i h−1 V ab ρ a a − ρ b b [ ]
v v
the spontaneous emission factor ne (k ) nh(k)
and the so called absorption factor  1  1 .
1−  v − 1  v − 1
n e k ( )
  nh k  ( )
v
In the spirit of the free carrier model and in the unsaturated limit the nl (k ) 's are given
by the quasi equilibrium Fermi-Dirac distributions -- viz.
−1
(ε )
v
( )
nα k ⇒ fα vk = exp  β
 
v
αk
− µ α  +1
  [ VII--46 ]
Thus, the absorption factor is given by
 1  1 
1−  v − 1 
( )
v − 1 = 1− exp β
( )
 (ε v
ek
+ ε h vk − µ e − µ h 
)
n e k   nh k  [ VII--47 ]
= 1− exp β [ (ε − ∆µ) ]v
k
and the spontaneous emission factor is given by

−1 −1
(ε ) [( )] ( )
v v
( ) ( )
ne k nh k =  exp −β
 
v
hk
− µ h  exp β ε vk − ∆ µ +1  expβ ε h vk − µ h  +1 [ VII--48 ]
     
where ε and ∆ µ ≡ µ e + µ h = (EF ) C − ( EF )V − ε g .

h2 k 2
v
k
= 45
2mr
45 For plotting purposes it is useful to write the complete gain factor as
−1 −1
v    x      r  
( )
∆ k = exp  β ε g
 
+
r
1+ r 1+ r
− xF + xh  +1  − exp  β ε g −
     1+ r
x+
r
1+ r
+ xh   +1 
 
where r ≡ me mh , x = hω vk ε g , xF = ∆ EF ε g
, and xh = µ h ε g
.
Therefore, for population inversion ∆ µ− ε = (EF ) C − ( EF )V − (ε g + ε ) > 0 which is

a very stringent condition!
Adapting Equation [ VI-26a ] from earlier in this set of lecture notes,46 we may see that
the small signal gain in a semiconductor, may be expressed in the form

t v v
( ) ( ) {the density of paired statesat k} L(ω )
−1
Gain ω vk ∝ M v vk, c vk ∆ k v
k
−ω n ; γ a b [ VII--49 ]
in bulk material ⇒ the density of paired states varies as the square root of the energy.
Therefore, we may draw the following gain curves:
0.14
µ h / ε g = 0.1
µ h / ε g = 0.05
Gain (arbitrary units)
0.12 0.05

0.1 0.04
0.08
0.03
0.06
0.02
0.04
0.02 0.01
ω -k / ε g
0.98 1.02 1.04 1.06 1.08 1.1 1.12 0.98 1.02 1.04 1.06 1.08 1.1 1.12
Gain in a bulk semiconductor with ∆ EF / ε g = 1.00, 1.02, 1.04, 1.06, 1.08, and 1.10
℘2  1  ω 
46 That is αn =
1
2
ωn
hε o  γ a b 
( )
L ω a b −ω n ; γ a b N −  n 
 2Q n 
GAIN V ARIATION WITH CARRIER CONCENTRATION
increasing
injection
current
relative gain
β ( hω − Egap )
QUANTUM CONFINEMENT
0 z < Lc 2
In an idealized quantum well V( z) = 
∞ z > Lc 2
which yields a single electron energy

2
r 2
r h
( ) h n π
2 2 2 k ⊥
Εn k ⊥ = 2 +
2 mz Lc 2 m⊥
and density of paired states which is a constant or a staircase.

Electron Energy

0.8

µ h / ε g = 0.05
0.6
0.4
0.2
ω -k / ε g
0.98 1.02 1.04 1.06 1.08 1.1 1.12
Gain in a 2D semiconductor with ∆ EF / ε g = 1.06, 1.08, and 1.10
increasing
injection
current
relative gain
β ( hω − Egap )
2

Semiconductor Photonics

Uploaded by

Copyright:

Available Formats

Semiconductor Photonics

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Semiconductor Photonics

Uploaded by

Copyright:

Available Formats

THE INTERACTION OF RADIATION AND MATTER: SEMICLASSICAL THEORY PAGE 82

VII. SEMICONDUCTOR P HOTONICS

The quantum mechanics of the assembly is treated to a good approximation by taking

The essential assumption of the adiabatic approximation is that the bracketed

As a further refinement, the electron problem must be parsed once more as

H elec = {H elec }static + {H elec }dynamic

R. Victor Jones, April 4, 2000

B LOCH ELECTRONS - Solutions of the Schrödinger equation for a single

is a small fluctuating electron-electron interaction. If we neglect these fluctuations, each

R. Victor Jones, April 4, 2000

or simplifying the notation

S CHEMATIC P ERIODIC LATTICE P OTENTIAL

Single ion potential

Bound state energy

R. Victor Jones, April 4, 2000

The lattice potential

Bloch's (Floquet's) theorem : The function

Shifting the origin by a lattice vector, we again rewrite Equation [ VII-7b ] as

In light of periodicity expressed in Equation [ VII-8 ], it follows that

Thus, any linear combination of the possible eigenfunctions

choose the combination

R. Victor Jones, April 4, 2000

In light of Bloch's theorem, Equation [ VII-7b ] may be rewrite in the form

∫∫∫ d vr V ( vr) exp(− i G ⋅ vr)

( k) = V ∫∫∫ d vr u ( vr) exp (− i G ⋅ vr)

R. Victor Jones, April 4, 2000

1D E MPTY LATTICE MODEL

In light of Equation [ VII-13a ] we may write

∫∫∫ d vr exp (− i k ′ ⋅ vr) V ( vr) exp ( i k ⋅ vr)

= (V) ∫∫∫ d r exp [− i ( k ′ − k) ⋅ r] ∑V exp ( i G ⋅ r) [ VII-21 ]

so that Equations [ VII-19 ] and [ VII-20 ] become

R. Victor Jones, April 4, 2000

Equation[ VII-24 ] is the famous Bragg scattering condition or the condition

R. Victor Jones, April 4, 2000

1D N EARLY F REE ELECTRON MODEL

R. Victor Jones, April 4, 2000

Diamond Structure Brillouin Zone and Nomincature

R. Victor Jones, April 4, 2000

For more on band structures go to:

For more on pn junction physics go to:

R. Victor Jones, April 4, 2000

B . OPTICAL P ROPERTIES OF B ULK (3-D) S EMICONDUCTOR 32

where the e f f e c t i v e m a s s m∗ is a measure of the inverse curvature of the band.33

where V is the volume of the sample and µα is the carrier quasichemical

Hence, for a 3D semiconductor, Equation [ VII-27 ] becomes

R. Victor Jones, April 4, 2000

OPTICAL MATRIX ELEMENTS

where the so called transition matrix element is given by

∫ d vr u∗ ( vr) exp[− i k ⋅ vr] {eˆ ⋅ Pv exp[− i k ⋅ vr]} u

∑ 2πh f H int i δ(ε f − ε i ) .

∫∫∫ d vr {vr} exp [ i ( k ′ − k − k + G − G ) ⋅ vr]

-- i.e. essentially only vertical transitions are allowed! Therefore,

R. Victor Jones, April 4, 2000

M v vk, c vk ′ = {polarization factor} M . [ VII-37b ]

"FREE - CARRIER THEORY " OF OPTICAL P ROCESSESS

Equation [ VII-37a ] tells us that the absorption or emission of a single

E NERGY D IAGRAM FOR B ULK L ASERS

C. INJECTION LASER THEORY

(For pictorials on laser configurations go to:

A fairly satisfactory model of lasing (and other optical processes) in semiconductors