Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Lecture Notes

Download as pdf or txt
Download as pdf or txt
You are on page 1of 77

Topic 1: Introduction

•Optical coefficients
•Complex dielectric constant

•Complex refractive index

•Introduction to optical materials

Optical coefficients
propagation through
the medium

incident light transmitted light

reflected light

Figure 1.1: Optical coefficients

• Reflectivity = reflected / incident power


• Transmissivity = transmitted / incident power
• T + R =1 if medium is transparent

1
refraction
Propagation
absorption and luminescence Figure 1.2:
Propagation of
light through a
scattering medium

• Velocity v= c/n, n is the refractive index


• I(z) = I0 exp(-az), a is the absorption coefficient
• T = (1-R1) exp(-aL) (1-R2)
• Luminescence : re-emission at lower frequency
• scattering: elastic- change of direction
inelastic - change of direction and frequency

Luminescence
• Luminescence comes
excited states relaxation
out at lower frequency
than absorption due to
emission
absorption

internal relaxation
• The energy shift
between absorption
ground state and luminescence is

Figure 1.3: luminescence mechanisms called the Stokes shift.

2
Complex optical coefficients
Complex relative dielectric constant: εr = ε1 + iε 2
Complex refractive index: n = n + iκ
Absorption coefficient: α = 4πκ / λ
εr = n2 ; ε1 = n 2 − κ 2 ; ε 2 = 2nκ
1/2
1 " 1/2 %
n= 2
( 2
$ε1 + ε1 + ε 2
2#
) '
&
1/2
1 " 1/2 %
κ= 2
(2
$ −ε1 + ε1 + ε 2
2#
) '
&
2
n −1 (n −1) 2 + κ 2
Reflectivity: R = =
n +1 (n +1) 2 + κ 2

Insulators / semiconductors
UV visible infrared
1.0 • Infrared absorption due
(a)
Al2O3
to phonons
Transmission

• ultraviolet/visible
absorption due to bound
0.0 electrons
0.8 (b) • position of fundamental
absorption edge depends
CdSe on the size of band gap
• transparency region in
0.0 between
0.1 1 10
Wavelength (microns)
Figure 1.4 : transmission spectra of
(a) sapphire (Al2O3) (b) CdSe

3
Metals
infrared visible UV

1.0
Figure 1.5:
Reflectivity

0.8 Reflectivity spectrum


silver
0.6 of silver
0.4
0.2
0.0
10 1.0 0.1
Wavelength (µm)
• Free electrons in the metal absorb

Þ High reflectivity up to “plasma frequency” in the UV

Organic materials
UV visible
1.0
Absorption (a.u.)

UV/blue band
0.8 Figure 1.6 : Absorption
0.6 spectrum of polyfluorene
0.4 polyfluorene
(F8)
0.2
0.0
300 400 500 600 700
Wavelength (nm)

• Strong absorption in UV/visible spectral region due to


electronic transitions
• Stokes-shifted emission across the visible spectral region

4
Doped insulators
1.0
sapphire
0.8 Figure 1.7:
Transmission

ruby Transmission spectra


0.6 of sapphire (Al2O3)
yellow/green and ruby (Al2O3: Cr3+)
0.4 band
0.2 blue band

0.0
200 400 600 800 1000
Wavelength (nm)
• Sapphire is transparent for visible wavelengths
• Cr3+ ions doped into sapphire absorb in the blue and yellow/green
spectral regions, hence red colour

Lifting of degeneracies
atom free atom
atom in
in B field
crystal
degenerate
magnetic
levels
crystal Zeeman
field effect
effect

Figure 1.8 : Lifting of degeneracies in the solid state


• Crystals have directions Þ lower symmetry than atoms or molecules
• new effects due to lower symmetry e.g. lifting of degeneracies

5
E solid free Band formation
atom

Figure 1.9:
Band formation in
the solid state

Interatomic separation

• Free atoms have sharp absorption and emission lines


• Solids have broad absorption and emission bands
• Two types of bands: electronic and vibronic

6
Topic 2: Dipole oscillators

• Introduction

• Lorentz oscillator model

• Kramers–Kronig relationships

• Dispersion

• Optical anisotropy & chirality

Electron oscillators

Figure 2.1: classical model of an atom.


Electrons are bound to the nucleus by springs
which determine the natural frequencies

Bound electrons (insulators, intrinsic semiconductors)


• restoring force for small displacements F = – k x
Þ natural resonant frequency w0 = (k/m)1/2
• w0 lies in the near infrared / visible / UV spectral regions

Free electrons (metals, doped insulators)


• k = 0 Þ w0 = 0

7
Dipole oscillators Figure 2.2

p(t) t p(t) = – e x(t)


x(t) P=Np
t
D = e0 E + P = ere0 E

x
Light emission
+ • natural oscillations at w0
p 2p cause radiation at w0
t=0 t=w t= w
0 0 • “Hertzian dipoles”

Refractive index Absorption


• light º AC electric field at w (¹w0) • light frequency = w0
• forced oscillations at w • resonance phenomenon
• atoms lag on the driving field • atoms absorb energy from
and re-radiate in phase the light beam

Vibrational oscillators
+
Figure 2.3
polar molecule
e w0

Molecular physics
IR absorption at resonant frequency of the vibrational modes

Ionic crystals
IR absorption at frequency of optically-active lattice vibrations
i.e. the transverse optic (TO) phonons

8
Lorentz oscillators
Dielectric constant Refractive index
30 6 Figure 2.4
e1

4 Lorentz
10

n
oscillator
2
-10 with
w0 w0 w0 = 1014 rad/s
40 4 g = 5 1012 s-1
est = 12.1
e2

g k
20 2 e¥ = 10
0 0
60 100 140 60 100 140
w (1012 rad/s)
• absorption at w0
2
Ne 1 • g = 1/t = damping rate
e r (w ) = 1 + c +
(
e 0 m0 w 0 - w 2 - igw
2
) • FWHM = g

Example: atomic absorption line


a Lorentzian lineshape
1700 m-1 γ2
α (ω ) = α 0
4Δω 2 + γ 2
0
n 4γΔω
n(ω ) = n0 − Δn
3.95 ´ 10-5 4Δω 2 + γ 2
n0 » 1
Δω = ω − ω 0

-200 0 200
Dn (MHz)

Figure 2.5
• sodium gas, N = 1´1017 m–3 (low density Þ n0 » 1)
• D2 line at 589.0 nm, hyperfine component with FWHM = 100 MHz

9
Multiple resonances
Figure 2.6
Refractive index

vibrational electronic
bands resonant frequencies:
transitions phonons IR
valence electrons vis/UV
core electrons X-ray
1
0
Ne 2 fj
Absorption

e r (w ) = 1 + å
(
e 0 m0 j w 0 j - w 2 - ig jw
2
)
0
IR visible UV X-ray • fj = oscillator strength
• classical theory Þ fj = 1
1011 1013 1015 1017 • quantum theory Þ fj determined
Frequency (Hz) by transition probability

Example: SiO2 glass


SiO2 glass (a) Figure 2.7
Extinction coefficient refractive index

3
real and imaginary
2 part of the refractive
n

index for SiO2 glass


1

1 (b)
Transparency region
10-2 between
phonon frequencies in IR
k

10-4 and band gap in UV

1012 1013 1014 1015 1016 1017


Frequency (Hz)

10
Optical fibre losses

http://nobelprize.org/nobel_prizes/physics/laureates/2009/phyadv09.pdf

Kramers–Kronig relationships
2 w ¢k (w ¢)
¥
n(w ) - 1 = Pò dw ¢
p w ¢2 - w 2
0

¥ w ¢ [ n(w ¢) - 1]
2
2
pw ò0 w ¢2 - w 2
k (w ) = - P dw ¢

Figure 2.9
• Refractive index (n) and absorption
Refractive index

4
(k) are the real and imaginary parts
of the same function: ñ = n+ik
• Absorption and refraction are 2
n at 10 µm
related to each other by the
vs band gap
Kramers–Kronig relationships
0
0 1 2
Band gap wavelength (µm)

11
UV transmission of glass

(at 546 nm) (at 310 nm)


Add UV absorbers:
• reduces UV transmission
• increases refractive index at
visible frequencies

Dispersion Figure 2.10


glass dispersion
Wavelength (nm)
1000 600 400 300 200
IR visible ultraviolet
• Group velocity
Refractive index, n

1.55 dispersion (GVD)


µ d2n/dl2
d2n • GVD determines
1.50 =0
dl 2 pulse spreading
in optical fibres
• GVD = 0 near
1.45 SiO2 glass 1300 nm

0.2 0.4 0.6 0.8 1.0 1.2 1.4


Frequency (1015 Hz)

12
Dispersive prisms

white red Figure 2.11


Dispersion of light by a prism

blue

• NORMAL dispersion: n increase with frequency


• ANOMALOUS dispersion: occurs near resonance lines

For Pink Floyd fans …

red red

blue
blue

anomalous dispersion ? normal dispersion

13
Double refraction (Birefringence)

Double refraction in calcite

optic axis Figure 2.12


Birefringent crystals
such as calcite
109°
e-ray (Iceland Spar)
separate orthogonal
6.2° light polarizations
o-ray
unpolarized into the
light 71° o(rdinary) and
e(xtraordinary) rays

Birefringence (optical anisotropy) : refractive index depends on


the direction of the polarization vector relative to the crystal axes

14
Optical anisotropy
optic axis optic axis
z z
propagation propagation
direction direction
q q
e e
y y

(a) o-ray (b) e-ray

Figure 2.13
æ Dx ö æ e11 0 0 öæ E x ö
Birefringence caused by ç ÷ ç ÷ç ÷
difference of dielectric constants ç D y ÷ = e 0 ç 0 e 22 0 ÷ç E y ÷
(and hence refractive index) çD ÷ ç 0 0 e 33 ÷øçè E z ÷ø
è zø è
along the different crystal axes.

Polarizing beams splitters


Figure 2.14
air gap Glan–Foucault
unpolarized q prism
light
o-ray
q e-ray

optic axis

Critical angle: sin qc = 1/n


Calcite: no = 1.658, qc = 37.1°; ne = 1.486, qc = 42.3°
Hence choose 37.1° ≤ q ≤ 42.3°

15
Wave plates Figure 2.15

ax tic
(a) (b)

op
is
q
o-ray e-ray
s
axi
ic
output q
opt

input
input
polarization
d

2p Half wave plate: Df = p


Df = ( no - ne ) d Quarter wave plate: Df = p / 2
l

Induced birefringence
• Isotropic materials are non-birefringent
• Induce birefringence Dn with strain or electric field
• Hence photo-elastic and electro-optic effects
• Kerr effect (quadratic electro-optic effect) observed in
all materials, including liquids and glass:
Δn = l K E2 ; K = Kerr constant
• Hence Kerr cells (see Fig. 11.8)
• Contrast with linear electro-optic effect (Pockels effect)
observed only in anisotropic crystals (See Fig 11.6)

16
Chirality
• Optical properties different for left or right circularly polarized
light due to chirality (helicity) of molecules or crystal structure
• Circular dichroism: different absorption for left or right
circular light
• Optical activity: different refractive index for left or right
circular light.
• Optical activity causes rotation of linear light: Examples:
dextrose, laevulose (fructose) [latin dexter, laevus]

pd
q= ( n - nL )
l R
amino acid

Magneto-optics
• Induce chirality in non-chiral materials with a magnetic field
• magnetic circular dichrosim in absorbing materials
• Faraday effect in transparent materials: rotation of linear
polarization by magnetic field
q=VBd; V = Verdet coefficient

input output Figure 2.16


The Faraday effect
d

17
Appendix: Local field corrections
Figure 2.8
P local field ¹ applied field in dense medium
- - - - q- - -

Lorentz correction:
++ + elocal = e + P/3e0
++++ in cubic crystal

e
Clausius Mossotti relationship
e r - 1 Nc a
=
er + 2 3

18
Topic 3: Interband absorption

• Interband transitions: direct and indirect


• Direct gap materials

• Optical orientation

• Indirect gap materials

• Photodetectors & solar cells

Interband absorption

Energy • Photon excites electron


upper band from filled valence to
empty conduction band
Ef
• Fundamental
absorption edge at Eg
Eg !w
• Process creates an
Ei electron–hole pair
lower band
Figure 3.1

19
Direct and Indirect absorption
(a) Direct band gap: Figure 3.2 (b) Indirect band gap
C.B. minimum at k = 0 C.B. minimum at k ¹ 0
E conduction band E

Eg q Eg
!w !w

valence band
k k
0 0
• kphoton = 2p/l ~ 107 m–1 negligible compared to B.Z. size p/a ~ 1011 m–1
• Transitions appear as vertical lines on E – k diagrams
• Phonon needed to conserve momentum for indirect gap materials
• Indirect absorption 2nd order process, therefore low probability

Atomic physics of semiconductors


p antibonding Figure 3.3

p s antibonding conduction band


Eg
s p bonding valence band

s bonding

ATOM MOLECULE CRYSTAL

• Four valence electrons per atom: Group IV (C, Si, Ge),


• III-V compounds (GaAs, InAs, InSb, GaN ….)
• II-VI compounds (ZnS, ZnSe, CdSe, HgTe, …. )
• V.B. ® C.B. is p ® s, hence allowed transition

20
GaAs band structure
4
Figure 3.4
2 GaAs
Energy (eV)

Eg !w • Direct gap at 1.5 eV


0
• Very important
-2 optoelectronic material
-4
• Strong absorption for
-6 !w > Eg
L L G D X
Wave vector k

fcc lattice Brillouin zone


kz
Symmetry points

G 000
X 100
010 L
001 G
K 110 X ky
L 111 W
K
kx

Fig D.5

21
Four-band model
E Figure 3.5

Electron (e) • Simplified band


band structure first proposed
by Kane (1957)
Eg • valid near k = 0
 ‚

0 k  Heavy hole transition


Heavy holes (hh)
-D ‚ Light hole transition
Light holes (lh)
• Split-off hole
Split-off holes (so) transitions also possible

InAs band edge absorption


1.0
InAs Figure 3.6
a2 (1012 m-2)

0.8 room temperature


InAs is a direct gap
0.6 III–V semiconductor
with Eg = 0.35 eV
0.4
!w < Eg : a = 0
0.2
!w > Eg :
0.0 a µ (!w – Eg)1/2
0.3 0.4 0.5 0.6
Energy (eV)

22
Magneto absorption in germanium
Figure 3.7
45
germanium
Transmission (%)

40 300 K • direct gap at 0.80 eV


35
B = 3.6 T • B-Field quantizes motion
30
in direction perpendicular
25 B = 0 to B
20
• Landau level absorption
15
when
0.78 0.80 0.82 0.84 0.86 !w = Eg + (n+½) !wc
Energy (eV) wc = eB / me
(cyclotron frequency)

Spin injection (optical orientation)


E |M12|2 µ MJ
conduction band
Þ hh transitions three
Eg J = 1/2 times stronger than lh
E1 selection rule
s+ s+ s- s- Dm = ±1 for s±
hh lh lh hh
0 J = 3/2
-D so so J = 1/2
valence band Figure 3.8
MJ
-3/2 -1/2 +1/2 +3/2
50% spin
N (+1/2) - N (-1/2)
P= polarization for
N (+1/2) + N (-1/2) s± excitation

23
Direct versus indirect absorption
Figure 3.9
Absorption coefficient (m-1)

106
105 • Direct absorption is
much stronger than
104 indirect absorption
103 GaAs
102 silicon •Silicon has indirect gap
at 1.1 eV
1.0 1.2 1.4 1.6 1.8 2.0
Energy (eV) • GaAs has direct gap at
1.4 eV

Germanium band structure


4

2 Eg = 0.66eV Figure 3.10


Energy (eV)

direct gap
0
• Indirect gap
-2 at 0.66 eV

-4 • Direct gap
at 0.80 eV
-6
L L G D X
Wave vector k

24
Germanium band edge absorption
10
80 (a) 291 K (b) 300 K germanium
8
a1/2 (m-1/2)

a (105 m-1)
60 phonon a ~ (!w - Egdir)1/2
6
absorption

emission
phonon

40 4
20 K Eg
20 2
0 0
0.6 0.7 0.8 0.9 0.6 0.7 0.8 0.9
Energy (eV) Energy (eV)
Figure 3.11
at 300K :
• indirect absorption: a µ (!w – Egind !Wphonon )2 Egind = 0.66 eV
• direct absorption: a µ (!w – Egdir )1/2 Egdir = 0.80 eV

Silicon absorption
Absorption coefficient (10 m )
-1

E2
8

2.5
• Indirect band gap
2.0 at 1.1 eV
E1
1.5
• Critical points
1.0 (van Hove
Eg silicon
0.5 singularities) at E1
300 K
(3.2 eV) and E2
0 (4.3 eV)
0 2 4 6 8 10
Energy (eV)
Figure 3.12

25
Silicon band structure
•density of states g(E)
4
µ 2 g(k) (dE/dk)–1
E1 Eg
0
Energy (eV)

E2 • Critical points (van


Hove singularities)
-4
whenever dE/dk = 0
-8 • occurs when
conduction and
-12 valence bands are
L L G D X parallel to each other:
Wave vector k Parallel band effect

Figure 3.13

Absorption spectroscopy Figures 3.14-15


white-light
white-light
source
source
scanning collimation lenses
monochromator sample
reflection in cryostat
detector
collection lenses
sample
entrance slit
transmission
detector spectrograph
valve silicon
diode array : computer
detector
vacuum pump

26
p-i-n diodes (Appendix E)
E p i n p i n
E

Eg Eg
Figs E.1-2 | eV0 |
li
li
z z
V0 = 0 Reverse bias V0 applied

• Electric field E = (Vbi – V0) / li li

• Vbi = built-in voltage » Eg / e p i n


V0
• V0 is negative in reverse bias z

Photodetectors
Figure 3.16
• Embed absorbing region within
!w p-n junction to form p-i-n diode
Ipc • Apply reverse bias V0
(i.e. V0 negative )
p
e • Electric field e = (Vbi – V0) / li
V0 i li
+ - • Vbi = built-in voltage » Eg / e
n • Photons absorbed if !w > Eg
• Creates electron-hole pairs in
i-region
Examples:
• photodiodes • Carriers swept out by e field
• solar cells (V0 = 0) into external circuit to generate
photocurrent Ipc

27
Solar cells (photovoltaics) Figure 3.17

!w pi n

Ipc
R
+ –
V = Ipc R
Ipc
maximum
ISC power • Small gap: large ISC , small VOC
• Large gap: small ISC , large VOC
• Single junction: max efficiency ~30%
V • Larger for multi junction
VOC

Solar cell efficiency

28
Topic 4: Excitons

• Introduction

• Wannier excitons

• Excitonic nonlinearities

• Frenkel excitons

Excitons Figure 4.1

a h
e

Free (Wannier) Tightly-bound (Frenkel)


radius >> a radius ~ a
small binding energy large binding energy
moves freely through crystal localized on one lattice site

29
Figure 4.2

Free exciton absorption

• Hydrogenic series of
n=1 lines satisfying :
a !w = Eg – RX / n2

n=2 • enhanced absorption for


!w > Eg
(!w – Eg)1/2
• only observed when
Photon energy T £ ( RX / kB)
Eg

Excitons in bulk GaAs Figs 4.3-4

exciton
n=1 n=2

1.2 1.2 n=3


294 K
a (106 m-1)
a (106 m-1)

0.8 RX
0.8 21 K
0.4 Eg
0.4 a µ (!w - 1.425) 1/2
0
1.514 1.520
0
1.42 1.46 1.50 1.54
Photon energy (eV)
Photon energy (eV)
• ultra pure sample
• standard purity sample
• T = 1.2 K
• T = 21 K – 294 K
• RX = 4.2 meV

30
Figure 4.5
Field ionization in GaAs
F ~ 2 RX / e aX
V0 = +1.00 V, E » 5 ´ 105 V / m
E F aX
!w
p
Photocurrent

V0 i E li
n ionized if e > F
T=5K GaAs parameters :
RX ~ 4.2 meV
V0= +1.44 V aX ~ 13 nm
E»0 li = 1 µm (typical)
0 F ~ 6 ´ 105 V / m
1.50 1.51 1.52 E ~ 1.5 ´ 106 V / m
Photon energy (eV) for V0 = 0 !

Figs 4.6 - 7

Nonlinear excitonic absorption


GaAs, 1.2 K
10
8 low density
a (106 m-1)

6 ´ 1022 m-3
(a) Low density 6 2 ´ 1023 m-3
Separation » diameter
4
2
0
1.512 1.516 1.520
(b) High density Photon energy (eV)
Separation » diameter
NMott = [ (4/3)paX3 ]–1
~ 1.1 ´ 1023 m-3 in GaAs

31
Figs 4.8 - 9
Frenkel excitons
Alkali halides Organic crystals
3 LiF C16H10 (pyrene)
300 K
a (108 m-1)

Absorption
2 NaCl
Eg
Eg
1
300 K
0
8 10 12 14 3.20 3.30 3.40
Energy (eV) Energy (eV)

Eg(eV) RX (eV)
Also: Rare gas crystals
NaCl 8.8 0.9
(solid Ne, Ar, Kr, Xe)
LiF 13.7 0.9

32
Topic 5: Luminescence

• Introduction
• Photoluminescence

• Electroluminescence

• LEDs and lasers


• Cathodoluminescence

Luminescence
• Luminescence spontaneous emission in solids

• Fluorescence fast luminescence


electric-dipole allowed, tR ~ ns
• Phosphorescence slow luminescence
electric-dipole forbidden, tR ~ µs – ms

• Electroluminescence electrical excitation


• Photoluminescence optical excitation
• Cathodoluminescence cathode ray (e–beam) excitation

33
Radiative quantum efficiency Fig. 5.1

inject
electrons
relaxation
EXCITED hR = radiative quantum
STATE efficiency
= radiative transition rate
tNR tR !w total transition rate

GROUND 1
hR =
STATE 1 + t R / t NR
inject holes

• Radiative transition rate determined by Einstein A-coefficient


• tR = A–1
• tNR determined by phonon population, number of traps etc

Direct gap materials


GaN
conduction band
T = 4 K, Eg = 3.50 eV
E
Luminescence
intensity

Absorption

electrons
Eg !w
holes

k=0 k 3.40 3.50 3.60


valence band Energy (eV)

• Strong emission at the band gap


• most III-V and II-VI semiconductors Figs 5.2 – 3
• linewidth ³ kBT

34
Indirect gap materials
conduction band
E
electrons Figure 5.4
phonon

!w Eg
holes
valence band
k
k=0
• Low emission probability (2nd order process)
• Long radiative lifetime Þ low radiative quantum efficiency
• diamond, silicon, germanium, AlAs

Photoluminescence
E E
electrons
conduction
band
Eg
hnL hn Figure 5.5
0
k
valence holes
k=0 band Density of states
• Excite using laser with photon energy > Eg
• electrons and holes relax to the bottom of their bands
• thermal distributions formed according to statistical mechanics
• emission from Eg to top of carrier distributions

35
Classical (Boltzmann)Statistics
103 Figure 5.6
PL intensity (a.u.)

GaAs
T = 100 K
102
1.50 1.52 kBT = 8.6 meV
kBT Eg = 1.501 eV
Eg
Fermi’s golden rule
1.49 1.50 1.51 1.52 1.53 Rate µ |M|2 r(hn)
Energy (eV)
• Boltzmann statistics: f(E) µ exp(–E/kBT) (occupancy factors)
• I(E) µ Density of states ´ fe(E) fh(E)
• PL rises sharply at Eg, then decays exponentially. Linewidth ~ kBT

Degeneracy Figs 5.7 – 8

E Ga0.47In0.53As, TL = 10 K
Photoluminescence (a.u.)

electrons
6nJ/pulse
E Fc

Eg hn 24 ps
250 ps 180 K
EFv
Eg 55 K
holes
Density of states 0.80 0.90 1.00
Energy (eV)

• Degeneracy observed at high density and low temperatures


• Emission from Eg to (Eg + EFc +EFv )

36
Photoluminescence spectroscopy
sample in cryostat Photoluminescence (PL) spectroscopy
• fixed frequency laser, measure spectrum
laser by scanning spectrometer

mirror PL excitation spectroscopy (PLE)


PL collection • detect at peak emission,
lenses vary laser frequency
entrance slit • effectively measures absorption

spectrometer Time-resolved PL spectroscopy


• short pulse laser + fast detector
: • measure lifetimes, relaxation processes

detector computer

Electroluminescence
holes hn » Eg
p-type • Epitaxial growth of high
epitaxial
n-type layers purity light-emitting layers on
electrons substrate crystal
• MBE, MOCVD, LPE ….
substrate

current
• Forward-biassed p-n junction
p • Electrons and holes recombine at the
V0 hn junction
n
• photon energy ~ Eg
Figure 5.10

37
Lattice matching
substrate materials
SiC sapphire GaAs
InP
hexagonal cubic direct band gap
6 AlN indirect band gap
Band gap (eV)

4 AlP
GaN GaP AlAs
visible blue
2 spectrum red
GaAs InP fibre
InN InAs optics
0
3 4 5 6
Figure 5.11
Lattice constant (Å)

Figs 5.12 – 3
Junction electroluminescence
depletion
region GaAs, Eg = 1.42 eV
p n
Electroluminescence

Eg electrons
EFv EFc 1 mA
holes 293 K

(a) V0 = 0
1.2 1.4 1.6 1.8
p n Energy (eV)
eV0 hn = Eg • Emission at Eg
• Operating voltage ~ Eg / e
(b) V0 » +Eg/e • Spectral width ~ kBT

38
Diode lasers Figs 5.14 – 16

l
current
metal contact
hn oxide
R1 R2 p - AlGaAs
i - GaAs
n- GaAs
n - AlGaAs
Output

substrate
power

light
Iin output
gth metal contact
Gain gn

Iin • Mirrors formed by “facets” (i.e. edges) of chip


Ith
• Emission wavelength ~ hc / Eg
• Linewidth determined by cavity modes

Figure 5.17
Cathodoluminescence
VACUUM • Focussed spot size ! 100nm
electron
beam • Primary electrons generate
Primary electrons secondary electrons in
few keV excitation volume
• Secondary electrons generate
back-scattered
electrons e–h pairs, hence luminescence

penetration excitation
depth (Re) volume
! 1 µm

CRYSTAL cathodoluminescence

• Commercial use in cathode ray tubes


• Research tool for investigating nanostructures:
Observe luminescence from electron microscope

39
40
Topic 6: Quantum confinement
• Dimensionality
• Quantum wells
ØEnergy levels
ØOptical transitions
ØQuantum confined Stark effect

• Quantum dots
• Carbon nanostructures

Fig. 6.1
Dimensionality
Dimensionality /
Confinement
z
3 /0 bulk
y
x
Density of states

quantum
2/1 well
quantum
1/2 wire
quantum
0/3 dot
Eg Energy

41
Fig. 6.2
Semiconductor quantum wells
AlGaAs GaAs GaAs quantum wells
crystal

substrate
GaAs growth
direction AlGaAs
substrate
z
d b d
C.B. e- C.B.

EgAlGaAs EgGaAs EgGaAs EgAlGaAs


V.B. V.B.
h+
Single quantum well MQW or superlattice
growth • Molecular beam epitaxy (MBE)
methods • Metal-organic chemical vapour deposition (MOCVD)

Fig. 6.3
Infinite quantum well
10
E (h 2/8m*d 2)

n=3
• kn = np/d
5
n=2 • En = (!kn)2/2m* = (!2p2/2m*d2) n2
• yn = (2/d)1/2 sin (knz +np/2)
n=1
0
z
-d/2 0 d/2

• symmetry about z = 0 Þ wave functions have definite parity


• yn has (n–1) nodes
• En depends on m*, hence heavy and light holes split

42
Figs 6.4–5
Finite quantum well
E y = 0.85 (13.2-x2)½ / x
V0
E2
n=2 8 y = tan(x)
E1 n=1
0 z 4
–d 0 d

y
2 2
0 x
• Wave functions tunnel into the 0 2 4
barrier Example : GaAs/AlGaAs
• wave function still identified by V0 = 0.3 eV, d = 10 nm
parity and number of nodes mw*= 0.067me, mb*= 0.092me
• Confinement energy reduced
E1 = 31.5 meV
compared to infinite well c.f. infinite well: E1 = 57 meV
• graphical solution to find En

Figs 6.6–7

Optical transitions
quantum well
conduction band

!w
n=1 n=2

z valence band

• Light polarized in x,y plane for normal incidence


• Parity selection rule: Dn = even number
• Infinite well selection rule: Dn = 0

43
Figs 6.8–9
2-D absorption

Absorption coefficient
conduction
band E
3-D

Eg
!w n=3
0 n=2
2-D
kxy n=1
0
valence band 0 5 10
(!w-Eg) in units of (h 2/8d 2µ)
• Absorption µ density of states
• Density of states constant in 2-D: g2D(E) = m / p!2
• Thresholds whenever !w exceeds (Eg + Een+ Ehn)
• Band edge shifts to (Eg + Ee1+ Ehh1)

Figs 6.10–11
GaAs quantum wells
GaAs/AlAs MQW, d = 7.6 nm GaAs/AlGaAs MQW
d = 10 nm
n=1 n=2 hh lh
Absorption (au)

1.0 10 bulk 4
hh hh
a (105 m-1)

hh lh n=3 n=2
lh lh
0.5 5 2
T=6K n =1 300 K
0.0 0 0
1.6 1.8 2.0 2.2 1.4 1.5 1.6
Photon energy (eV) Photon Energy (eV)
• Excitonic effects enhanced in quantum wells: strong at room temp
• Pure 2-D: RX2D = 4 ´ RX3D
• Typical GaAs quantum well: RX ~ 10 meV ~ 2.5 ´ RX (bulk GaAs)
• Splitting of heavy and light hole transitions

44
Spin injection in quantum wells
Selection rules
E s+: Dm = +1
conduction band s-: Dm = -1
J = 1/2
Eg

s+ s+ s- s-
0
lh lh J = 3/2
hh hh
-D J = 1/2
so so
valence band

-3/2 -1/2 +1/2 +3/2


MJ

Figure 6.12

Quantum confinement Stark effect (1)


Figure 6.13
(a) Ez = 0 (b) Ez = 107 V/m

1800
Energy (meV)

1600 e1
1400 1455.0 1447.9
1438.5
1462.4

–7.4 hh1 9.4


0
–200
–10 0 10 –10 0 10
Position (nm) Position (nm)
• 10 nm GaAs / Al0.3Ga0.7As quantum well in an electric field
• Eg (GaAs) = 1420 meV

45
The quantum confined Stark effect (2)
GaAs MQW, d = 9.0 nm, 300 K
!w pi n MQW
lh1®e1
hh1®e1
hh2®e2
V0 Ez (a) 0 V

Photocurrent (arb. units)


1.5´106 V/m
photocurrent

• Red shift of excitons


• Excitons stable to high (b) -10 V
hh1®e2 1.1´107 V/m
fields (c.f. Fig 4.5)
• Parity selection rule hh2®e1
broken 1.4 1.5 1.6 1.7
• used to make modulators Energy (eV) Figs 6.14–15

Figure 6.16

Emission spectrum
Zn0.8Cd0.2Se/ZnSe quantum well
Eg = 2.55eV (10K)
d = 2.5 nm
Eg = 2.45eV (300K)
10 K
PL intensity

300 K

• Emission energy shifted from


Eg to (Eg + Ee1 + Ehh1)
0 • Tune l by changing d
2.4 2.5 2.6 2.7 • Brighter than bulk due to
Energy (eV) improved electron-hole overlap
• Used in laser diodes and LEDs

46
Figure 6.17

Intersubband transitions
n-type quantum well
• Need z polarized light
hn n=2 • Parity selection rule:
electrons Dn = odd number
n=1

• Transition energy ~ 0.1 eV (~ 10 µm, infrared)


• Absorption used for infrared detectors
• Emission used for infrared lasers (Quantum cascade lasers)

Figure 6.18
Quantum dots

Bulk (3-D)
Density of states

Quantum well (2-D)

Quantum wire (1-D)

z Quantum dot (1-D)

Eg Energy

47
Quantum dots
Spherical dot
Cuboid dot d

! 2 2 2$
 2π 2 # n x n y n z &
R
E= + +
2m* #" d x2 d y2 d z2 &%
! 2 2$
 2 # Cnl π &
E=
Density of states

2m* #" R2 &%

3-D C10 = 1
C11 = 1.43
C12 = 1.83
0 5 10 C20 = 2
(E-Eg) in units of (h 2/8d 2m*) !

Colloidal quantum dots Figure 6.20

(a) CdSe (b)


Absorption (arb. units)

10 K 6 nm CdTe
D 300 K
C 5 nm
B
A 4 nm

1.5 2.0 2.5 3.0 3.5 1.6 1.8 2.0 2.2


Energy (eV) Energy (eV)

• Found in semiconductor doped glass (Colour glass filters &


stained glass)
• Available commercially

48
Self-organized epitaxial dots Figs 6.19, 21

2-D harmonic oscillator levels


m
-2 -1 0 1 2

Energy in unit of !w0


3 ‘d’ shell
InAs 200 nm
2 ‘p’ shell
quantum dot
1 ‘s’ shell
V(r)
0 r
0
GaAs 10 nm
z
ℏ$% $&' $ dz
!≈ + & + 1 ℏ./
2)∗ +' $ dz ≪ a
a

InAs quantum dots Figure 6.22


Quantum confined Stark effect
(a) (b) 2.6V
Photocurrent (pA)
PL Intensity (arb. units)

T = 10 K
Bias voltage

Spatially 10 pA
200 nm resolved PL
aperture Single QD

PC
PL Intensity

Far field PL PL
(arb. units)

200 µm ~107 QDs


spot
T = 10 K 0.6V

1.25 1.30 1.35 1.40 1.250 1.255 1.260


Energy (eV) Energy (eV)

49
Figure 8.20
Carbon nanostructures
graphene nanotube C60 Bucky ball *

* Buckminster fullerene named after


architect R. Buckmintser Fuller

Fig. 8.21
Graphene band structure

Ideal 2-D material

(a) (b) M
4 p* G
K
0 EF

–4 s p E
Energy (eV)

(c)
–8 p*

–12 !w
ky
–16 s
kx
–20 p
G K M G

Taken from Machon, Phys. Rev B 66, 155410 (2002)

50
Graphene transmission Figure 8.22

100 100

Transmission (%)
(a)
Transmission (%)

99 (b)

98 95
97
graphene, single layer
96 90
95
400 500 600 700 0 1 2 3 4 5
Wavelength (nm) Number of layers

After Nair et al. Science, 320, 1308 (2008).

Nanotube chirality Quantum wire (1-D)


D.O.S.

armchair
(5,5) Eg Energy
chiral
q (7,3)
zig
zag
a2
(9,0)
a1 c
Figure 8.23

chiral vector c = n1 a1 + n2 a2
Metallic if n1−n2 = 3m ; otherwise semiconductor

51
Nanotube transitions
E E

semiconducting conduction band


nanotube kz emission
valence band

density of states

E E
conduction band
metallic kz EF
nanotube
valence band

density of states Figure 8.24

Semiconducting Nanotube PL spectra


s m SWNT ensemble
E22 E11
3 1.0 (7,5)
Normalized PL intensity (au)
Energy gap (eV)

(8,3)
2 (9,1)

0.0 single
1 1.0
Es11
SWNTs
(9,1)
0 (8,3)
0 1 2 3
Tube diameter (nm) (7,5)
0.0
900 1000 1100 1200
Figure 8.24-5 Wavelength (nm)

52
Carbon bucky ball: C60

PL (a.u.)

Absorption
(a) (b)

3 S2(u)
Energy (eV)

2 1.2 ns T1(g) 1.6 1.8 2.0 2.2


S1(g)
4 Energy (eV)

Absorption
(104 cm–1)
1 1.8 µs 3
2 S1 S2 C60
0 S0(g)
singlets triplets 1
0
2 3 4 5
Energy (eV)

Figure 8.27

53
54
Topic 7: Free electrons
• Free carrier reflectivity

• Metals

• Doped semiconductors

• Plasmons (bulk & surface)

• Negative refraction

Fig 7.1
Free electron reflectivity
1.0 • Key parameter:
Reflectivity

Plasma frequency
0.5 wp = (Ne2/e0m0)1/2
• Lightly-damped system
er(w) = 1 - wp2/w2
0.0
0 1 2 • R = 100% below wp
w / wp • transmitting above wp
Free electron effects observed in: Þ ultraviolet transparency
• metals of metals
• doped semicondcutors

55
The Drude model Fig 7.2

!wp
Aluminium
1.0
!wp = 15.8 eV,
0.8
Reflectivity

experimental data s = 3.6 ´ 107 W–1 m–1


0.6
0.4 g=0
t = 8.0 fs
wp2
0.2 e r (w ) = 1 -
0.0
w 2 + igw
0 5 10 15 20 1/ 2
æ Ne 2 ö
Energy (eV) w p = çç ÷
÷
è e 0 m0 ø

• t = 1/g = momentum scattering time = m0s/Ne2, s = DC conductivity


• Reflectivity < 100% below wp
• Discrepancies caused by interband absorption transitions

Interband transitions in Aluminium


Transition energy ~ 1.5 eV
• Transitions from filled
16 states below EF to
EF empty states above EF
Energy (eV)

12
• Very high density of
8 states for parallel bands

• “parallel band effect”


4
at the critical points
0 • Hence reflectivity dip
G X W L G K X
at 1.5 eV
Wave vector k
Fig 7.3

56
Fig. 7.4, 7.6
Transition metals
E 1.0
4s band 0.8 copper

Reflectivity
optical 0.6
transitions
EF 0.4 !wp

visible
0.2
3d band 0.0
0 2 4 6 8 10 12 14
Density of states Energy (eV)
• Inner d orbitals form narrow bands below the Fermi energy
• threshold for optical transitions to outer half-full s band gives a
sharp absorption edge
• hence characteristic colours of metals, eg copper, gold

Copper band structure Fig 7.5

Electronic configuration: 3d10 4s1

minimum transition energy Integrated density of states


~ 2.2 eV (560 nm)
0 2 4 6 8 10
2 2
Energy (eV)

0 EF 0
Energy (eV)

-2 3d -2
-4 bands -4
-6 copper -6
-8 3d and 4s bands -8
W L G X WK 0 2 4 6
Wave vector k Density of states (states eV-1)

57
Fig 7.7
Doped semiconductors
n-type InSb at 300 K
100
4.0 ´ 1024
Reflectivity (%)

80 2.8 ´ 1024
1.2 ´ 1024
60
6.2 ´ 1023
40 3.5 ´ 1023
20 Plasma frequency
0 wp = (Ne2/e0erm*)1/2
10 20 30
Wavelength (µm)

• Control plasma frequency by varying the doping density


• Plasma edge in the infrared

Fig 7.8
Free carrier absorption
E
!k 2 scattering
E= process
2me*
Ne 2
!w a free carrier =
EF m * e 0 nctw 2

• Transitions from filled states below EF to empty ones above EF


• a µ l2 / t
• Scattering process required to conserve momentum
• Continuous infrared absorption in doped semiconductors

58
Fig 7.9

Intervalence band absorption


E

EF  ‚ ƒ
k
D
hh band

SO band lh band

• Infrared absorption in p-type semiconductors

Impurity transitions Fig. 7.10

• Impurity levels: En = –(m*/m0)(1/er2) RH / n2

n conduction band conduction band n


2 donor 2
levels
1 1

n-type
material

valence band valence band

(a) low temperature: (b) high temperature:


infrared absorption band tail below Eg

59
Donor absorption in n-type silicon
Phosphorous doped Silicon
1.0
Absorption (103 m-1)

2p0
2p N = 1.2 ´ 10 20 m–3
3p
T = 4.2 K
4p
3p0 5p me* ~ 0.5 m0
4p0 er = 12
0 R* ~ 45 meV
32 34 36 38 40 42 44
Photon energy (meV)
m* 1
• simple model: hn = R* æç1 - ö÷ , R* = e
1
RH
è n2 ø m0 e r 2
• complications due to effective mass anisotropy Fig. 7.11

Plasmons

k
• Quantized plasma waves: i.e. longitudinal
oscillations of the electron plasma at wp

Fig. 7.12

60
Observation of bulk plasmons
sample Doped semiconductors: !wp ~ 10 meV
!wp Measure by Raman scattering:
!wout = !win ± n!wp
Ein Eout

Raman scattering rate


!wp plasmon plasmon
emission absorption

Metals: !wp ~ 3–20 eV


observe by electron energy -400 -200 0 200 400
loss spectroscopy : Energy shift (cm-1)
Eout = Ein − n!wp n-type GaAs, T = 300 K
Ein ~ keV N = 1.8 ´ 10 23 m–3
Figs 7.12-13 !wp = 19 meV (150 cm–1)

Surface plasmons
z
dielectric
dielectric: ed
z=0 x +++ --- +++ ---
k
metal: em
metal

z • Surface waves at boundary between


metal and insulator
Ezd (z) • Plasmonics: propagate energy as
surface plasmons in nano-circuits
0 Ez • Cannot couple directly to free-space
photons due to wave vector
Ezm (z) mismatch: hence need for couplers
(e.g. gratings)
Figs 7.14 - 15

61
Surface plasmon polaritons
w bulk w surface
(a) (b)

w = ck w = ckx /Öεd
wp wp
wsp

0 0
0 1 ck /wp 2 0 1 ckx /wp 2

wp
wsp = Hence wsp = wp/Ö2 if ed = 1 (i.e. air)
1+ed
Fig. 7.16

Localized surface plasmons


Gold nanoparticles 6-7nm diameter
1.2
Absorption (a.u.)

0.8 Courtesy
MesoGold® M. Sugden
• pure gold 0.4 T. Richardson
nanoparticles
suspended in 0.0
300 500 700 900
water Wavelength (nm) Fig. 7.17
• Colloidal gold is not gold coloured !
• Localized plasma oscillations at surface of metal nanoparticles
• wsp ~ wp/Ö(ed+2) i.e. wp/Ö3 for air-metal boundary, but varies
somewhat with size and shape

62
Negative refraction Figs 7.18 - 19
µr
n=1 n<0
qi qr
II I
ñ = ik n>0
k>0 k=0 perfect lens
er
n<0 ñ = ik
k=0 k>0
III IV

• Need both er and µr to be negative


i.e. region III
• Not possible in natural materials:
hence need metamaterial metamaterial

63
64
Topic 8: Phonons
(Chapter 10)
• Infrared active phonons
• Reststrahlen

• Inelastic light scattering


Ø Raman scattering

• Phonon lifetimes

Figs 10.1–2
Lattice absorption
• Resonant absorption when
w, k W, q photon frequency and wave-
vector match the phonon W, q
optical branch • Only possible for optical
phonons of ionic or polar
Frequency

photon

(partly ionic) crystals


• Phonon must be transverse:
hence TO phonons involved
acoustic branch • Resonance occurs in the
0 p infrared ~ 10 – 100 µm
Wave vector (q)
a

65
TO phonon resonance
e = e0eiqz
Figure 10.3
x+ z
x-

• TO phonon modes of ionic crystals: positive and negative ions


move in opposite transverse directions
• This generates transverse electric field waves
• Light resonates with these modes when the wave vectors and
frequencies match

Reststrahlen Figure 10.4

• 100% reflectivity between nTO and nLO


• Lyddane-Sachs-Teller relationship: nLO / nTO = (est / e¥ )1/2

n TO 2 Example:
e r (n ) = e ¥ + (e st - e ¥ ) 2
n TO - n 2 nTO = 10 THz est = 12.1
nLO = 11 THz e¥ = 10

1.0
40
Reflectivity

(a) (b)
20
er

0
-20
0.0
6 8 10 12 14 6 8 10 12 14
Frequency (THz) Frequency (THz)

66
Figure 10.5
Infrared reflectivity
1.0
GaAs :
nTO = 273.3 cm–1
Reflectivity

nLO = 297.3 cm–1


InAs GaAs
InAs :
nTO = 218.9 cm–1
0.0 nLO = 243.3 cm–1
200 240 280
1 cm–1 = 2.998 ´ 1010 Hz
Wave number (cm-1)

• Reststrahlen band observed between nTO and nLO


• reflectivity less than 100% due to damping
• damping caused by anharmonic phonon decay. Typical lifetime ~ 10 ps

Polaritons Figs 10.6 – 7


• Polariton = coupled TO phonon polarization wave & photon wave
• anticrossing of photon and TO phonon modes when w ~ WTO
• modifies the dispersion of both photons and TO phonons
20 52
w = v¥ q w = vst q
(THz)

!WLO
Energy (meV)

upper branch 48
WLO / 2p
10 44
w / 2p

WTO / 2p lower branch 40

0 36
0.0 1.0 2.0 0 1 2 3
Wave vector q (106 m-1) Wave vector (10 m-1)
6

Calculated: WTO / 2p = 10 THz, Raman scattering data on GaP


est = 12.1, e¥ = 10

67
Fig. 10.8
Polarons
• free electron in a polar solid
produces a local lattice distortion
through the electron-phonon
e- coupling aep
• modifies the effective mass to :
m** = m*(1 – aep/6)–1

• Polaron effects are small in III-V s (e.g. ~ 1% in GaAs) but increase as


the crystal becomes more ionic.
• Self-trapping leads to hopping conduction instead of band conduction
• Organic conductors show strong polaron effects

Fig. 10.9
Inelastic light scattering
• Conservation laws:
w1 = w2 W ; k1 = k2 Q
w2 , k2 • Stokes scattering:
w1 , k1 phonon emission (+ sign)
• Anti-Stokes scattering:
W,q phonon absorption (– sign)
• Ianti-Stokes / IStokes = exp(–!W/kBT)

• Raman scattering: scattering from optical phonons


• Brillouin scattering: scattering from acoustic phonons
• Qmax = 2k ~ 106 m–1 for back-scattering geometry
\ can only probe small wave vector phonons
• Scattering can also be observed from plasmons, magnons etc

68
Fig. 10.10

Raman scattering arrangement

laser beam computer

sample
in cryostat
reflected entrance slit
laser
scattered
scanning
double
:
spectrometer
light
CCD array
detector

collection
lenses

Fig. 10.11
Raman scattering data
TO LO

• Raman scattering from TO


GaAs and LO phonons in various
Raman signal (a.u.)

III-V semiconductors using


InP a Nd:YAG laser at 1.06 µm

AlSb • 1cm–1 = 3 ´ 1010 Hz


= 0.124 meV
300 K GaP

200 300 400


Energy shift (cm-1)

69
Figs 10.12 – 13
Phonon lifetimes
“Klemens channel” (see Klemens 1966)
W2 , q2
WTO optical

Frequency (W)
W1 , q1 branch
WTO
W3 , q3
2 acoustic
W1 , q1 branch
0
p 0 p
W3 , q3 a Wave vector (q) a
W2 , q2

• 3 phonon processes • TO phonon can decay into two


caused by anharmonicity acoustic phonons
in the crystal • Lifetime typically ~ 10 ps.

70
Topic 9: Nonlinear optics
(Chapter 11)
• Definitions

• Resonant vs non-resonant nonlinearities

• Second-order nonlinearities
Øfrequency doubling

Øelectro-optics

• third-order nonlinearities

Nonlinear susceptibility
D = e0 E + P º e0 er E
P = e0 c(1) E + e0 c(2) E2 + e0 c(3) E3 + ….
Ä er = 1 + c(1) + c(2) E + c(3) E2 + ….

• c(n) = nth order nonlinear susceptibility


• c(2) = 0 in materials with inversion symmetry (centrosymmetric)
ie: all isotropic materials (gases, liquids, glasses)
+ some crystals (eg NaCl)
• optical properties depend on the light intensity
• needs large electric fields (ie high power lasers)

71
Nonresonant nonlinear response
P P Figure 11.1
(a) P(t) (b) P(t)
t e t
e
sine wave
output
t t

e (t) e (t) distorted output


small amplitude contains terms
large amplitude
sine wave input at 2w, 3w etc
sine wave input

• laser is tuned to the transparency region of crystal


• far from resonance with any atomic transitions

Resonant nonlinear response


N2
2
A
un I(z) I – dI
1
N1
Figs. 11.2-3 dz
z
• laser is tuned to resonance with an atomic transition
• stimulated emission reduces net absorption rate at high intensity
• equivalent to “Pauli blocking” of upper state
• saturable absorption: a(I) = a0 / (1 + I / Is) ; Is = saturation intensity
• a(I) » a0 – a2 I at small I ; a2 = a0 / Is
• Der µ I µ E2 \third-order nonlinearity

72
Second order nonlinear effects
Pi(2) = e0 Sjk c(2)ijk Ej Ek c(2) = 0 in centrosymmetric crystals

w2 -w2
(a) w1+w2 (b) w1-w2

w1 w1 Fig. 11.4

Process Input Ouptut


Frequency mixing w 1, w 2 w1 + w2
Frequency doubling w, w 2w
Down conversion w1 w 2, w 1 - w 2

Phase matching
c(2) q
w 2w
kw k2w

• Phase matching condition: 2kw = k2w


(conservation of momentum)
Þ 2´ n(w) w / c = n(2w) 2w / c
Þ n(w) = n(2w)
• not usually possible due to dispersion.
• use a birefringent crystal with optic axis at angle q :

1 sin 2 q cos 2 q
= +
no (w ) 2 ne ( 2w ) 2 no ( 2w ) 2

73
Frequency doubling / tripling etc
nonlinear
• use a 2nd order nonlinear crystal crystal
266 nm
2w 4w
nonlinear
crystal
Nd:YAG laser w 2w 532 nm
1064 nm

2w 3w
355 nm
w
Figure 11.5 nonlinear
crystal

Electro-optic modulators
y
crossed
45 z polarizer
y' E-O crystal
E x
polarizer output
x'

V
input
• Linear electro-optic (Pockels) effect:
Field-induced birefringence: Dn µ rij E
• rij = electro-optic coefficient
¹ 0 for non centro-symmetric crystal
• Consider as c(2) process with DC field
• Use to make intensity modulators Figure 11.6

74
Third-order nonlinear processes
w1
w
w2 w1+w2+w3 w w
-w
w3
(a) Four-wave mixing (c) Optical Kerr effect

w w
w 3w -w ws
w ws

(b) Frequency tripling (d) Raman effect

Figure 11.7

Kerr cells
crossed crossed
Kerr polarizer Kerr polarizer
medium output medium
polarizer output
polarizer

input E V input

E
laser pulse
• Quadratic electro-optic effect:
Field-induced birefringence: Dn µ c(3) E 2
• Occurs even in isotropic materials (eg glass, liquids)
• DC Kerr effect: rotate polarization with DC field
• Optical Kerr effect: induce birefringence with light pulse
Figure 11.8

75
Nonlinear refractive index
Der = c(3) E2
n(I) = n0 + n2 I
Dn = n2 I
Þ n2 = c(3) / n02 c e0

• Third order nonlinear effects found in all materials


including isotropic media e.g. glasses, liquids, gases
• self-phase modulation: Df = 2 p Dn L / l = 2 p n2 I L / l

• soliton: balance dispersion


and self-phase modulation
• propagation of high intensity
short pulses along optical fibres
Figure 11.9

Semiconductor nonlinearities
E Figs. 11.10-11 40
GaInAs/InP MQW
a (103 cm-1)

30
20 exciton
Eg 10 300 K
k
0
0.7 0.9 1.1 E (eV)
• strong Pauli blocking when 8 6700
a(I) = + 2000
1 + I / 290
!w ~ Eg
a (103 cm-1)

6
Þ saturable absorption 4
• third order nonlinearity 2
• excitonic enhancements E = 0.77 eV
0
• see also Fig. 4.7 10 102 103 104 105
I (W cm-2)

76
Fig. 11.12
Saturable absorption in quantum dots
Absorbance (units of 10–3)

15

Peak absorbance
10–2
10 3.5 µeV
10–3
5

0 10–4
1.26545 1.26550 1.26555 0.1 1 10 100 1000
Energy (eV) Power (nW)

N2
2
• Tune laser to QD exciton un
• Behaves as saturable two-level atom 1
N1

77

You might also like