Physics of Low dimensional

Materials-1
Nanostructured materials derive their special properties
from having one or more dimensions made small
compared to a length scale critical to the physics of the
process.
Development of electronic properties as a
function of cluster size

Each band has a width that reflects the interaction between atoms, with a
bandgap between the conduction and the valence bands that reflects the
original separation of the bonding and antibonding states.
Electronic DOS and dimensionality

Size effects are most

evident at band edges
(semiconductor NPs).

DOS (dn/dE) as a function

of dimensionality.
3D case is for free particles.
 Quantum Dots: Introduction

Definition:
 Quantum dots (QD) are nanoparticles/structures that exhibit 3
dimensional quantum confinement, which leads to many unique
optical and transport properties.

GaAs Quantum dot containing just 465 atoms.

 Quantum Dot: Introduction

 Quantum dots are usually regarded as semiconductors by definition.

 Similar behavior is observed in some metals. Therefore, in some cases
it may be acceptable to speak about metal quantum dots.
 Typically, quantum dots are composed of groups II-VI, III-V, and IV-
VI materials.
 QDs are bandgap tunable by size which means their optical and
electrical properties can be engineered to meet specific applications.
 Quantum Confinement

Definition:
 Quantum Confinement is the spatial confinement of electron-hole pairs
(excitons) in one or more dimensions within a material.
 1D confinement: Quantum Wells
 2D confinement: Quantum Wire
 3D confinement: Quantum Dot
 Quantum confinement is more prominent in semiconductors because
they have an energy gap in their electronic band structure.
 Metals do not have a bandgap, so quantum size effects are less
prevalent. Quantum confinement is only observed at dimensions below 2
nm.
 Quantum Confinement

 Recall that when atoms are brought together in a bulk material the
number of energy states increases substantially to form nearly continuous
bands of states.

Energy
Energy

N
 Quantum Confinement

 The reduction in the number of atoms in a material results in the

confinement of normally delocalized energy states.
 Electron-hole pairs become spatially confined when the diameter of a
particle approaches the de Broglie wavelength of electrons in the
conduction band.
 As a result the energy difference between energy bands is increased with
decreasing particle size.

Energy

Eg
Eg
 Quantum Confinement

 This is very similar to the famous particle-in-a-box scenario and can be

understood by examining the Heisenberg Uncertainty Principle.
 The Uncertainty Principle states that the more precisely one knows the
position of a particle, the more uncertainty in its momentum (and vice
versa).
 Therefore, the more spatially confined and localized a particle becomes,
the broader the range of its momentum/energy.
 This is manifested as an increase in the average energy of electrons in the
conduction band = increased energy level spacing = larger bandgap
 The bandgap of a spherical quantum dot is increased from its bulk value
by a factor of 1/R2, where R is the particle radius.*

• Based upon single particle solutions of the schrodinger wave equation

• valid for R< the exciton bohr radius.
 Quantum Confinement

 What does this mean?

 Quantum dots are bandgap tunable by size. We can engineer their optical and electrical
properties.
 Smaller QDs have a large bandgap.
 Absorbance and luminescence spectrums are blue shifted with decreasing particle size.

Energy

555 nm 650 nm
 Quantum Dots (QD)

 Nanocrystals (2-10 nm) of

semiconductor compounds
 Small size leads to confinement of
excitons (electron-hole pairs)
 Quantized energy levels and altered
relaxation dynamics Eg

 Examples: CdSe, PbSe, PbTe, InP

 Quantum Dots: Optics

Absorption and emission occur at specific wavelengths, which

are related to QD size

Eg
 Applications of QDs: Light Emitters

 The discovery of quantum dots has led to the development of an

entirely new gamut of materials for the active regions in LEDs and
laser diodes.
 Indirect gap semiconductors that don’t luminesce in their bulk form
such as Si become efficient light emitters at the nanoscale due quantum
confinement effects.
 The study of QDs has advanced our understanding of the emission
mechanisms in conventional LED materials such as InGaN, the active
region of blue LEDs.
 The high radiative-recombination efficiency of epitaxial InGaN is due
to self-assembled, localized, In rich clusters that behave like QDs.
 Quantum Dot Solar Cells

Possible benefits of using quantum dots (QD):

 “Hot carrier” collection: increased voltage due to
reduced thermalization
 Multiple exciton generation: more than one
electron-hole pair per photon absorbed
 Intermediate bands: QDs allow for absorption of light
below the band gap, without sacrificing voltage

MRS Bulletin 2007, 32(3), 236.

 QDs: Collect Hot Carriers

Band structure of bulk semi- 1. Tune QD absorption (band gap) to

conductors absorbs light match incident light.
having energy > Eg. However, 2. Extract carriers without loss of
photo-generated carriers voltage due to thermalization.
thermalize to band edges.

Conduction
Band

Eg

Valence
Band
QDs: Multiple Exciton Generation

In bulk In QDs:
semiconductors: 1 photon = multiple excitons
1 photon = 1 exciton

Eg Impact ionization

Absorption of one photon of

The thermalization of the
light creates one electron-hole
original electron-hole pair
pair, which then relaxes to the
creates another pair.
band edges.
 QDs: Multiple Exciton Generation

300
Quantum efficiency for exciton
generation: The ratio of excitons
produced to photons absorbed
Quantum Eff (%)

250

>100% means multiple exciton

200 generation

Occurs at photon energies (Ehv)

150
much greater than the band gap
(Eg)
100
1 2 3 4 5
Photon Energy (Ehv/Eg)
 QDs: Intermediate Bands

Conventional band structure Intermediate bands in the band gap

does not absorb light with allow for absorption of low energy
energy < Eg light

Intermediate band
formed by an
Eg array of QDs
 Expected effects for electrons in
nanostructures

 Quantum confinement effect

 Charge discreteness and strong electron-electron Coulomb
interaction effects
 Tunneling effects
 Strong electric field effects
 Ballistic transport effects
 Overview of Quantum Confinement

History: In 1970 Esaki & Tsu proposed fabrication of an

artificial structure, which would consist of alternating
layers of 2 different semiconductors with
Layer Thickness
 1 nm = 10 Å = 10-9 m  SUPERLATTICE
 PHYSICS: The main idea was that introduction of an
artificial periodicity will “fold” the Brillouin Zones into
smaller BZ’s  “mini-zones”.
 The idea was that this would raise the conduction band
minima, which was needed for some device applications.
 Modern growth techniques (starting in the 1980’s), especially MBE &
MOCVD, make fabrication of such structures possible!
 For the same reason, it is also possible to fabricate many other kinds
of artificial structures on the scale of nm
(nanometers)  “Nanostructures”
Superlattices = “2 dimensional” structures
Quantum Wells = “2 dimensional” structures
Quantum Wires = “1 dimensional” structures
Quantum Dots = “0 dimensional” structures!!

 Clearly, it is not only the electronic properties of materials which can

be drastically altered in this way. Also, vibrational properties
(phonons). Here, only electronic properties & only an overview!
 For many years, quantum confinement has been a fast growing field
in both theory & experiment! It is at the forefront of current research!
Quantum Confinement in Nanostructures: Overview
Electrons Confined in 1 Direction:
Quantum Wells (thin films):
 Electrons can easily move in ky
2 Dimensions! kx

nz
1 Dimensional
Electrons Confined in 2 Directions: Quantization!
Quantum Wires: ny
 Electrons can easily move in kx
1 Dimension! 2 Dimensional nz
Quantization!
Electrons Confined in 3 Directions:
Quantum Dots: nz
 Electrons can easily move in 3 Dimensional
nx Quantization!
0 Dimensions! ny
Each further confinement direction changes a continuous k component
to a discrete component characterized by a quantum number n.
 Quantum confinement

 Trap particles and restrict their motion

 Quantum confinement produces new material behavior/phenomena
 “Engineer confinement”- control for specific applications
 Structures

 Quantum dots (0-D) only confined

states, and no freely moving ones
 Nanowires (1-D) particles travel only
along the wire
 Quantum wells (2-D) confines
particles within a thin layer

(Scientific American)
 PHYSICS: Back to the bandstructure:
 Consider the 1st Brillouin Zone for the infinite crystal. The
maximum wavevectors are of the order

km  (/a)
a = lattice constant. The potential V is periodic with period a. In the
almost free e- approximation, the bands are free e- like except near
the Brillouin Zone edge. That is, they are of the form:

E  (k)2/(2mo)
So, the energy at the Brillouin Zone edge has the form:

Em  (km)2/(2mo)
or

Em  ()2/(2moa2)
 PHYSICS
 SUPERLATTICES  Alternating layers of material. Periodic, with
periodicity L (layer thickness). Let kz = wavevector perpendicular
to the layers.
 In a superlattice, the potential V has a new periodicity in the z
direction with periodicity L >> a
 In the z direction, the Brillouin Zone is much smaller than that for an
infinite crystal. The maximum wavevectors are of the order:
ks  (/L)
 At the BZ edge in the z direction, the energy has the form:
Es  ()2/(2moL2) + E2(k)
E2(k) = the 2 dimensional energy for k in the x,y plane.
Note that: ()2/(2moL2) << ()2/(2moa2)
 Primary Qualitative Effects of Quantum Confinement

 Consider electrons confined along 1 direction (say, z) to a layer of

width L:
Energies
 The energy bands are quantized (instead of continuous) in kz &
shifted upward. So kz is quantized:
kz = kn = [(n)/L], n = 1, 2, 3
 So, in the effective mass approximation (m*), the bottom of the
conduction band is quantized (like a particle in a 1 d box) & shifted:
En = (n)2/(2m*L2)
 Energies are quantized! Also, the wavefunctions are 2
dimensional Bloch functions (traveling waves) for k in the x,y
plane & standing waves in the z direction.
Quantum Confinement Terminology

Quantum Well  QW
= A single layer of material A (layer thickness L), sandwiched between 2
macroscopically large layers of material B. Usually, the bandgaps satisfy:

EgA < EgB

Multiple Quantum Well  MQW
= Alternating layers of materials A (thickness L) & B (thickness L). In this case:
L >> L
So, the e- & e+ in one A layer are independent of those in other A layers.

Superlattice  SL
= Alternating layers of materials A & B with similar layer thicknesses.
Brief Elementary Quantum Mechanics & Solid State
Physics Review

 Quantum Mechanics of a Free Electron:

 The energies are continuous: E = (k)2/(2mo) (1d, 2d, or 3d)
 The wavefunctions are traveling waves:
ψk(x) = A eikx (1d)
ψk(r) = A eikr (2d or 3d)
 Solid State Physics: Quantum Mechanics of an Electron in a Periodic
Potential in an infinite crystal :
 The energy bands are (approximately) continuous: E= Enk
 At the bottom of the conduction band or the top of the valence band,
in the effective mass approximation, the bands can be written:
Enk  (k)2/(2m*)
 The wavefunctions are Bloch Functions = traveling waves:
Ψnk(r) = eikr unk(r); unk(r) = unk(r+R)
Some Basic Physics

 Density of states (DoS)

dN dN dk
DoS  
dE dk dE
k space vol
N (k ) 
vol per state
4 3 k 3
in 3D: 
(2 ) 3 V

Structure Degree of dN
Confinement dE
Bulk Material 0D E
Quantum Well 1D 1
Quantum Wire 2D 1/ E
Quantum Dot 3D d(E)
 QM Review: The 1d (infinite) Potential Well (“particle in a
box”) In all QM texts!!

 We want to solve the Schrödinger Equation for:

x < 0, V  ; 0 < x < L, V = 0; x > L, V 
 -[2/(2mo)](d2 ψ/dx2) = Eψ
 Boundary Conditions:
ψ = 0 at x = 0 & x = L (V  there)
 Energies:
En = (n)2/(2moL2), n = 1,2,3
Wavefunctions:
ψn(x) = (2/L)½sin(nx/L) (a standing wave!)
Qualitative Effects of Quantum Confinement:
Energies are quantized & ψ changes from a traveling
wave to a standing wave.
 In 3Dimensions…

 For the 3D infinite potential well:R

 ( x, y, z ) ~ sin( nLxx ) sin( mLyy ) sin( qLzz ), n, m, q  integer

2 2
Energy levels  8nmLh 2  8mmLh 2  8qmLh 2
2 2 2 2

x y z

Real Quantum Structures aren’t this simple!!

 In Superlattices & Quantum Wells, the potential barrier is obviously
not infinite!
 In Quantum Dots, there is usually ~ spherical confinement, not
rectangular.
 The simple problem only considers a single electron. But, in real
structures, there are many electrons & also holes!
 Also, there is often an effective mass mismatch at the boundaries.
That is the boundary conditions we’ve used are too simple!
 Quantization in a Thin Crystal
E

An energy band with continuous k

is quantized into N discrete points kn
in a thin film with N atomic layers.

Electron
Scattering

EVacuum
Inverse
Photoemission
EFermi
Photoemission

0 /d k /a d
= zone
boundary
n = 2d / n
N atomic layers with the spacing a = d/n
N quantized states with kn ≈ n  /d ( n = 1,…,N ) kn = 2 / n = n  /d
 Quantization in Thin Graphite Films

1 layer =
graphene

2 layers

EVacuum

EFermi 3 layers
Photoemission

0 /d k /a

4 layers
 layers
N atomic layers with spacing a = d/n : =
graphite
 N quantized states with kn ≈ N  /d
Quantum Well States
in Thin Films

becoming
continuous for
N

discrete for
small N

Paggel et al.
Science 283, 1709 (1999)
 Counting Quantum Well States
n
 (eV)
hAg/Fe(100) (N, n')
4 3 (2, 1)
13 N
15 (3, 1)
11.5 0.3 (7, 2)
14 (12, 3) Periodic Fermi level crossing
14 (13, 3)
13 of quantum well states with
Photoemission Intensity (arb. units)

13
12 increasing thickness
14 11
2 (a) Quantum Well States for Ag/Fe(100)

Line Width (eV)

14 10
16

Binding Energy (eV)

9
13
0.2 0 n
8 1 2 3 4 5
16
7
16 6
16 6 1
1 7
5
14
4 8
16 0.1 2
3
16
10 2 (b) Number of monolayers N
1
Work Function (eV)

2 1 0 100 4.4 200 300

Binding Energy (eV) Temperature (K)

4.3
 The Important Electrons in a Metal

Energy  EFermi
Energy Spread  3.5 kBT

Transport (conductivity, magnetoresistance, screening length, ...)

Width of the Fermi function:
FWHM  3.5 kBT

Phase transitions (superconductivity, magnetism, ...)

Superconducting gap:
Eg  3.5 kBTc (Tc= critical temperature)
 Energy Bands of Ferromagnets

Calculation Photoemission data

4
Ni

Relative to EF
2

Energy
0

[eV]
-2
0.7 0.9
1.1k || along [011] [Å-1 ]
-4

-6

States near the Fermi level cause the energy

-8
splitting between majority and minority spin
-10
 K X
bands in a ferromagnet (red and green).
 39
Some Unusual Physical Properties of Nanomaterials

1. Reduced Melting Point -- Nanomaterials may have a significantly lower melting point or
phase transition temperature and appreciably reduced lattice constants (spacing between
atoms is reduced), due to a huge fraction of surface atoms in the total amount of atoms.
2. Ultra Hard -- Mechanical properties of nanomaterials may reach the theoretical strength,
which are one or two orders of magnitude higher than that of single crystals in the bulk form.
The enhancement in mechanical strength is simply due to the reduced probability of defects.
3. Optical properties of nanomaterials can be significantly different from bulk crystals.
--- Semiconductor Blue Shift in adsorption and emission due to an increased band gap.
Quantum Size Effects,
Particle in a box.
--- Metallic Nanoparticles Color Changes in spectra due to Surface Plasmons
Resonances
Lorentz Oscillator Model.
4. Electrical conductivity decreases with a reduced dimension due to increased surface
scattering.
Electrical conductivity increases due to the better ordering and ballistic transport.
5. Magnetic properties of nanostructured materials are distinctly different from that of bulk
materials. Ferromagnetism disappears and transfers to superparamagnetism in the
nanometer scale due to the huge surface energy.
6. Self-purification is an intrinsic thermodynamic property of nanostructures and nanomaterials
due to enhanced diffusion of impurities/defects/dislocations to the nearby surface.
7. Increased perfection enhances chemical stability.
Most are tunable with size!
 2D Nanostructures: Superhydrophobic surfaces

The angle formed by a tangent to a flat surface of a drop of water at the

point of contact (contact angle) is given in terms of the interfacial
energies of the system by the Young equation:

 AB   AC
cos c 
 BC

γAB= air/surface interfacial tension

γAC= water/surface interfacial tension

γBC= air/water interfacial tension

cos c  1 Water/surface repulsion (large interfacial tension)

Water drop

Si Nanowires Coated Si surface Coated

(planar) nanostructured
surface (rough)
Roughening on the nanoscale can greatly increase
hydrophobicity.
 42

Optical Properties

 The reduction of materials' dimension has

pronounced effects on the optical properties.
 The size dependence can be generally classified
into two groups.
 One is due to the increased energy level spacing as
the system becomes more confined, and the other is
related to surface plasmon resonance.
43
 Plasmonics

• The long wavelength of light (≈ m) creates a problem for

extending optoelectronics into the nanometer regime.

• A possible way out is the conversion of light into plasmons.

• They have much shorter wavelengths than light and are

able to propagate electronic signals.
 What is a Plasmon ?

A plasmon is a density wave in an electron gas. It is analogous to a sound

wave, which is a density wave in a gas consisting of molecules.

Plasmons exist mainly in metals, where electrons are weakly bound to the
atoms and free to roam. The free electron gas model provides a good
approximation (also known as jellium model).

The electrons in a metal can wobble like a piece of jelly, pulled back by the
attraction of the positive metal ions that they leave behind.

In contrast to the single electron wave function that we encountered

already, a plasmon is a collective wave where billions of electrons oscil-late
in sync.
 The Plasmon Resonance
Right at the plasmon frequency p the electron gas has a resonance,
it oscillates violently. This resonance frequency increases with the
electron density n , since the electric restoring force is proportional to
the displaced charge (analogous to the force constant of a spring).
Similar to an oscillating spring one obtains the proportionality: p  n
The plasmon resonance can be 1
observed in electron energy loss
spectroscopy (EELS). Electrons with
2
and energy of 2 keV are re-flected 3
4 5
from an Al surface and lose energy
by exciting 1, 2, 3,… plasmons. The
larger peaks at multiples of 15.3 eV
are from bulk plasmons, the smaller
peaks at multiples of 10.3 eV from
surface plasmons.
 Why are Metals Shiny ?
An electric field cannot exist inside a metal, because metal electrons react to it by creating
an opposing screening field. An example is the image charge, which exactly cancels the
field of any external charge. This is also true for an electromagnetic wave, where
electrons respond to the changing external field and screen it at any given time. As a result,
the electromagnetic wave cannot enter a metal and gets reflected back out.
However, at high frequency (= high photon energy) there comes a point when the external
field oscillates too fast for the electrons to follow. Beyond this frequency a metal loses its
reflectivity.
The corresponding energy is the plasmon energy Ep = ħp (typically 10-30 eV,
deep into the ultraviolet).
The reflectivity of aluminum cuts off at its plasmon energy Data (dashed) are
compared to the electron gas model (full).

Ep
 Plasmons and Energy-Saving
Window Coatings
The reflectivity cutoff at the plasmon energy can be used for energy-
saving window coatings which transmit visible sunlight but reflect
thermal radiation back into a heated room.
To get a reflectivity cutoff in the infrared one needs a smaller electron
density than in a metal. A highly-doped semiconductor is just right, such
as indium-tin-oxide (ITO). We encountered this material already as
transparent front electrode for solar cells and LCD screens.

An ITO film transmits visible light and

reflects thermal infrared radiation,
keeping the heat inside a building.
R = Reflectivity
T = Transmission
 Low-Dimensional Plasmons in
Nanostructures

We have showed how single electron waves become quantized by

confinement in a nanostructure. Likewise, collective electron waves (=
plasmons) are affected by the boundary conditions in a thin film, a
nano-rod, or a nano-particle.

Plasmons in metal nanoparticles are often called Mie-resonances, after

Gustav Mie who calculated them hundred years ago. Their resonance
energy and color depend strongly on their size, similar to the color
change induced in semiconductor nanoparticles by confinement of
single electrons. In both cases, smaller particles have higher resonance
energy (blue shift).
 50

Nanotechnology in Roman Times: The Lycurgus Cup

Plasmons of gold nanoparticles in glass reflect green, transmit red.
 Quantum Numbers of Plasmons

Like any other particle or wave in a (crystalline) solid, a plasmon has the
energy E and the momentum p as quantum numbers, or the circular
frequency  = E/ħ and the wavevector k = p/ħ . One can use the same E(k)
plots as for single electrons.

Photon

Bulk Plasmon

Surface Plasmon
 Coupling of Light and Plasmons

To combine optoelectronics with plasmonics one has to convert light (photons)

into plasmons. This is not as simple as it sounds.

Bulk plasmons are longitudinal oscillations (parallel to the propa-gation

direction), while photons are transverse (perpendicular to the propagation). They
don’t match.

Surface plasmons are transverse, but they are mismatched to photons in their
momentum. The two E(k) curves never cross. It is possible to provide the
necessary momentum by a grating, which transmits the wavevector k = 2/d
(d = line spacing) .
 Attenuated Total Reflection

Another method to couple photons and surface plasmons uses attenuated

total reflection at a metal-coated glass surface. The exponentially
damped (evanescent) light wave escaping from the glass can be matched
to a surface plasmon (or thin film plasmon) in the metal coating. This
technique is surface sensitive and can be used for bio-sensors.

Gold
film
 Plasmons and the Dielectric
Constant 
The dielectric constant is a complex number:  = 1 + i 2

The real part 1 describes refraction of light,

The imaginary part 2 describes absorption .

The bulk plasmon occurs at an energy Ep where 1 = 0,

the surface plasmon occurs at an energy Es where 1 = -1 .
(More precisely: Im[1/] and Im[1/(+1)] have maxima.)

1
2 Typical behavior of the dielectric
constant versus energy E for a solid
1 with an optical transition at E=E0 . A
0 metal has E0=0 .
E
-1 Ep
0 E0 s E
55
 56

Mesoscopic oscillation of
Bandgap (s & p band)
charge around a positive
+ Envelope (curvature of wave
lattice
function)
 57

Theoretical Tools

 Metals:
– Lorentz Oscillator Theory of Materials
– Plasmons and Plasmonics
 Semiconductors:
– Band Theory for Crystals
– Band Transport at Nanoscales:
Molecular Metals and Semiconductors
– Microscopic Description of Optical
Properties
58
 Optical Properties: Surface Plasmon Resonance
59

•Surface plasmon resonance is the coherent excitation of all

the "free" electrons within the conduction band, leading to an
in-phase oscillation.

•When the size of a metal nanocrystal is smaller than the

wavelength of incident radiation, a surface plasmon resonance
is generated.

• The figure shows schematically how a surface plasmon

oscillation is created. The electric field of an incoming light
induces a polarization of the free electrons relative to the
cationic lattice.

•The net charge difference occurs at the nanoparticle

boundaries (the surface), which in turn acts as a restoring
force. In this manner a dipolar oscillation of electrons is
created with a certain frequency.
 60

Surface plasmon resonance

 In this manner a dipolar oscillation of electrons is created with a

certain frequency. The energy of the surface plasmon
resonance depends on both the free electron density and the
dielectric medium surrounding the nanoparticle.
 The resonance frequency increases with decreasing particle size
if the size of the particles is smaller than the wavelength of the
particles.
 The width of the resonance varies with the characteristic time
before electron scattering. For larger nanoparticle, the
resonance sharpens as the scattering length increases. Noble
metals have the resonance frequency in the visible light range.
 Mie was the first to explain the red color of gold nanoparticle
colloidal in 1908 by solving Maxwell's equation for an
electromagnetic light wave interacting with small metallic
spheres.
 Bulk plasmon: quantization of collection valence
electron density oscillation at frequency ne 2
p 
 0m
ħp is typically ~ 10-20 eV

Surface plasmon
p
sp 
2

localized at the surface

and its amplitude
decays with the depth