Nanomaterials: Properties

Nguyn Tun Anh

Scale and Properties

polycrystalline, nanocrystalline, and am
orphous structures
made up of ordered crystals meeting at
boundaries where there is disorder

Scale and Properties

two structural length scales: the crystals
and the atoms
fraction of disordered material

 the crystal size shrinks until it becomes

of atomic dimensions completely diso
rdered amorphous
Properties of amorphous state and nano
structured would represent limits

Scale Dependence of Properties

basic assumption: small piece has the sa
me values as a large
greatly simplifying the analysis
However, it does not always work

Mechanical Properties
Nanodispersions
high-strength aluminum alloys: 4% copp
er and 96% aluminum
alloy is heated to 550oC, the copper diss
olves.
cooled rapidly, the copper stays dissolve
d

Mechanical properties (cont.)

aged by holding it at 150oC
allow atoms to rearrange by diffusion bu
t too to remain dissolved, the copper pr
ecipitates out as a nanodispersion of a c
ompound with the approximate compos
ition CuAl2

Nanocrystalline

 The boundaries of grains act as obstacle

s to dislocation motion
measured by the force f*
The number N of dislocations

 Breakdown: N=1

Nanolaminates
multilayers, alternating layers, usually of two
different materials.
the layers have to be thin (few atomic)
Data for the tensile strength of copper-nickel
laminates
are plotted as a function of the bilayer period
in the upper part of Figure 7.8.
The properties are remarkable: a nanolamina
te made from two soft metals (copper and nic
kel, for example) with a bilayer period of a fe
w nm can have a strength of several GPa,

THERMAL PROPERTIES
Melting Point
fundamental point of reference
correlates with the bond strength

surface effects should be considered

and Lo is the latent heat of melting, To is the

melting point of the bulk, T is the melting poin
t of the extended system, where surface effect
s are included, and VL is the volume of liquid.

 is the surface tension and A is the in

crement in surface area.

 At equilibrium, the solid core of radius r

has the same chemical potential as the s
urrounding liquid layer of thickness t, w
hich occurs when the differential GT
otal/t = 0.
For a sphere
the upper melting temperature

Melting Point

 nanoparticles in a matrix (e.g. nanocom

posite)
the surface of the nanoparticle is in cont
act with a matrix instead of being expos
ed to the surrounding atmosphere.
according to Youngs theorem

Melting point
assuming that = 90o,
when SM > LM, the melting point of the embedd
ed nanoparticles should be lower than the bulk
when SM > LM, the melting point of the embedd
ed nanoparticles should be higher than the bulk

The later call superheating, and it has been sh

own experimentally.
Due to nanoscale effects, the melting tempera
ture can be either increased or reduced with r
espect to the bulk material.

Thermal Transport
In addition to the melting temperature, many
of the current applications of nanomaterials r
equire knowledge about thermal transport.
In some cases, such as microprocessors and
semiconductor lasers, the goal is to transport
heat away as quickly as possible,
whereas for applications such as thermal barr
iers, the objective is to reduce thermal condu
ction.

 heat is transported by two different mechanism

s: lattice vibration waves (phonons) and free el
ectrons.
In metals, the electron mechanism is dominent
In nonmetals, phonons are the main
In both metals and nonmetals, as the system le
ngth scale is reduced to the nanoscale, there a
re quantum confinement and classical scatteri
ng effects.

 bulk homogeneous solid nonmetal materials, the wa

velengths of phonons are much smaller than the len
gth scale of the microstructure.
However, in nanomaterials the length scale of the m
icrostructure is similar to the wavelength of phonon
s.
In 0-D nanomaterials (e.g. nanoparticles), quantum
confinement occurs in three dimensions.
In 1-D nanomaterials (e.g. nanowires and nanotube
s), confinement is restricted to two dimensions.
In 2-D nanomaterials such as nanofilms and nanoco
atings, quantum confinement takes place in one dim
ension

 the presence of nearby surfaces in 0-D,

1-D, and 2-D nanostructures causes a c
hange in the distribution of the phonon
frequencies as a function of phonon wav
elength as well as the appearance of sur
face phonon modes.
Changes in the velocity with which the v
ariations in the shape of the waves amp
litude propagate, the so-called group ve
locity

 a phonon bottleneck occurs in 0-D nano

structures.
One-dimensional nanomaterials behave
as a phonon waveguide similar to optica
l ones for light.

2D nanomaterial
Single-layered nanoscale thin films, mos
t results show that the thermal conducti
vity is less than bulk materials

 multilayered films grains, an interface produces a thermal r

esistance.
The interface disturb the regular crystal lattice on which ph
onons propagate.
The interface can separate two crystals of the same materia
l with different orientations, such as a grain boundary, acro
ss which the two regions have a different distribution of ph
onons.
On the other hand, an interface can separate dissimilar mat
erials, such as a multilayer structure, for which the two diffe
rent materials have different
densities and sound velocities.
phonon is scattered, thermal conductivity is reduced.

 For nanoporous materials, the nanosize

effect is determined by the number and
size of the pores.
Due to the porosity, these materials hav
e low permittivity and thermal conductiv
ity

Heat capacity and coefficient of therm

al expansion
The heat capacity and coefficient of ther
mal expansion of nanomaterials have be
en much less studied.
Different trend
Difficult to explain

Electrical properties
Conduction
Large size system
Mechanism: delocalized, electrons can move free
ly in all dimensions
Scattered by phonons, impurities, interafces, ran
dom walk

Nano-size system
Quantum effect: band is replaced by discrete
Classical effect: mean free-path becomes compa
rable, reduce scattering

3-D nanomaterials
The three spatial dimensions are all above
the nanoscale.
Two aforementioned effects can be negle
cted.
Bulk nanocrystalline materials exhibit a hi
gh grain boundary area-to-volume ratio
Electron scattering
Reduce the electrical conductivity.

2-D nanomaterials
Thickness at the nanoscale, quantum confinement will occur
carrier motion is uninterrupted along the plane
the wave functions of electrons are limited to very specific va
lues
only electron wavelengths that are multiple integers of the th
ickness will be allowed.
All other electron wavelengths will be absent.
Reduction in the number of energy states available for cond
uction
The electrons become trapped in potential well of width equ
al to the thickness.
In general, the effects of confinement on the energy state ca
n be written as

 L is the width of the potential well (thickness of

2-D nanomaterial),
and n is the principal quantum number.
assumes an infinite-depth potential well model.
the carriers are free to move along the plane of t
he sheet.
total energy of a carrier has two components
a term related to the confinement dimension
a term associated with the unrestricted motion along
the two other in-plane dimensions.

Unrestricted motion
the z-direction to be the thickness direction
and x and y the in-plane directions in which th
e electrons are delocalized.
Characterized by two wave vectors kx and ky, w
hich are related to the electrons momentum
px = kx and py = ky.
The energy corresponding to these delocalize
d electrons is given by the so-called Fermi ene
rgy,

 where kF= (kx2+ky2)0.5

At the temperature of absolute zero, all t
he conduction electrons are contained
within a circle of radius kF.
The total energy of an electron (due to c
onfinement and unrestricted motion) in
a 2-D nanomaterial with thickness at the
nanoscale can be given by

 Since the electronic states are confined along the na

noscale thickness,
the electron momentum is only relevant along the inplane
directions.
scattering by phonons and impurities occurs mainly i
n-plane, leading to a 2-D electron conduction.
However, for 2-D nanomaterials with nanocrystalline
structure, the large amount of grain boundary area
So, the smaller the grain size, the lower the electrical
conductivity of 2-D nanocrystalline materials.

1-D nanomaterials
quantum confinement occurs in two dim
ensions, whereas unrestricted motion oc
curs only along the long axis of the nano
tube/rod/wire.
For a 1-D nanomaterial, the energy of a
2-D confinement depends on two quant
um numbers, ny and nz, in the form

Total Energy in 1-D

Considering now the electron free motion along the x-direction (long axis),

Because of the confinement, the nanoscale dimensions of 1-D nanomaterial

s act as reflectors, not allowing the electrons to exit the surfaces.
In addition, scattering by impurities and/or phonons becomes restricted to t
he long axis of the tube, despite the fact that boundary scattering is more pr
onounced due to the high surface-to-volume ratio of 1-D nanomaterials.
the transport of electrons along the tube occurs without significant loss of ki
netic energy the transport is ballistic, particularly at low temperatures.
Metallic carbon nanotubes, the conductivity is extremely high, as much as o
ne billion amperes/cm2, in contrast to 1 million amperes/cm2 for copper.
In addition to the effects described here, carbon nanotubes also exhibit a lo
w density of defects and high thermal dissipation, reducing even further the
chances for scattering.

0-D nanomaterials
Confined along the three directions Lx, Ly,
and Lz.
Total energy can be given by
In this fashion, all the energy states are disc
reet and no electron delocalization occurs.
Metallic systems can behave as insulators d
ue to the formation of an energy band gap,
which is not allowed in the bulk form.

In practical point of view

These materials need to be coupled to external circuits by electrodes.
For 2-D and 3-D nanomaterials, ohmic contacts are possible. However, for 0D and 1-D nanomaterials, the contact resistances between nanomaterials and
the connecting leads are usually high.
Thus one mechanism of providing conduction is through electron tunneling.
This is a quantum mechanical effect in which an electron can penetrate a pot
ential barrier higher than the kinetic energy of the electron.
Think of a configuration in which two metals are separated by a thin insulator
For an electron to tunnel from one metal to the other across the insulator, on
e of the metals must have unoccupied energy states.
A simple way of achieving this is to apply a voltage V across the circuit to raise
the Fermi energy (the energy of the highest occupied quantum state) of one o
f the metals.
In this fashion, electrons can tunnel from the metal with the highest Fermi en
ergy to the metal with the lowest Fermi energy, producing a current I along th
e circuit.

 The current I = V/R, where R is the resistance.

However, in this case, the resistance is primarily due to el
ectron tunneling.
As an example, arrays of gold nanoparticles have been el
ectrically coupled by connecting the nanoparticles to eac
h other by organic molecules. The nanoparticles act as th
e metal electrodes, whereas the organic molecules play t
he role of a thin insulator.
Under these conditions, the conductance C, defined as C
= I/V, was shown to increase due to electron tunneling co
mpared to the case in which the nanoparticles were not c
onnected by the organic molecules.

Application electron tunneling

The phenomenon of can also be used to develop field-effect transistors (FET) made fro
m quantum dots
two electrodes, a source and a drain, are coupled to a quantum dot and connected thro
ugh a circuit (see Figure 7.20). In addition, a gate voltage is provided to the quantum do
t to control its resistance and ultimately the current I passing between the lead and the
drain.
Due to the discrete nature of the electrical charge, electrons tunnel from the source to t
he quantum dot and then to the drain, one at the time.
Therefore the junction, which acts as a capacitor, suffers a raise in voltage V = e/C (e is
the elementary charge) when a single electron is added.
If the change in voltage is large enough, another electron can be prevented from tunneli
ng. This effect is called a Coulomb blockade.
electrons will not tunnel until a discreet voltage is reached.
To promote electron tunneling, the temperature has to be low enough so that the energ
y (e2/C) necessary to charge the junction with one electron exceeds the thermal energy
kT.
As the capacitance decreases with the size of the particle, a nanoparticle allows the Co
ulomb blockade to be observable at higher temperatures.

dielectric behavior
The large number of grain boundaries in nanocrystalline materials is expected
to increase the dielectric constant.
Nanocrystalline TiO2, a higher dielectric constant was found, compared with
coarse-grained samples.
This is due to the fact that under an applied electric field, the positive and ne
gative charges that are segregated at the interfaces will lead to some form of
polarization.
Since for nanocrystalline TiO2 the volume fraction of grain boundaries is muc
h larger than in coarse-grained TiO2, the polarization mechanisms will be mu
ch more important, leading to a higher dielectric constant
A similar effect was found for polymermatrix nanocomposites reinforced with
nano TiC fillers.
An increase of the dielectric constant was observed for higher loading levels o
f TiC, especially when the TiC content was near the percolation threshold. As
in the case of nanocrystalline TiO2, the increase of the dielectric constant is d
ue to interface polarization.

Magnetic Properties
In general, for any ferromagnetic material, the total energy can be writt
en as the sum of various terms, in the form

where Eexc is the exchange energy, Eani is the anisotropic energy, Edem is
the demagnetization energy, and Eapp is the energy associated with an ap
plied magnetic field.
For macroscopic magnetic materials, a magnetostrictive energy must al
so be included, but for nanoscale materials this energy can be neglect
ed.
This magnetization energy Etotal can then be related to a magnetic field
according to the expression

where M is the magnetization vector and H is the applied magnetic field

 The first term is due to the quantum mechanical interaction

between atomic magnetic moments and represents the tend
ency for the magnetization vectors to align in one direction.
In other words, if the magnetic moment is sufficiently large, t
he resulting magnetic field can drive a nearest neighbor to a
lign in the same direction, provided the exchange energy is
greater than the thermal energy.
The second term represents the anisotropy energy that resu
lts from the spins tendency to align parallel to specific cryst
allographic axes, called easy axes. Thus, a soft magnetic
material will exhibit low anisotropy energy, whereas a har
d magnetic material shows high anisotropy energy.

 Though both the exchange energy and the anisotropy ener

gy try to order the spins in a parallel configuration, the thir
d term, namely the demagnetization energy, which is relat
ed to the magnetic dipole character of spins, leads to the f
ormation of magnetic domains. Thus, for macroscopic ferr
omagnetic materials, all the magnetic moments are aligne
d in magnetic domains, although the magnetization vector
s of different domains are not parallel to each other. Each
domain is magnetized to saturation, with the moments typi
cally aligned in an easy direction. Depending on the ratio
of anisotropy to demagnetization energy, we can expect o
pen (ratio <1) or closure domain (ratio >1) structures.

 Finally, the energy associated with an applied magnetic f

ield, called Zeeman energy and represented by the last t
erm, results from the tendency of spins to align with a m
agnetic field.
Initially, as the magnetic field increases, the magnetizati
on of the material increases.
However, at some point, a saturation point, called satura
tion magnetization, is reached, above which an increase
in magnetic field does not produce an increase in magn
etization.
The saturation magnetization is material and temperatur
e dependent.

 For nanocrystalline ferromagnetic materials, an import

ant consideration is the interaction among exchange e
nergy, anisotropic energy, and demagnetization energ
y.
For very small particles or grain sizes, the exchange for
ces are dominant due to strong coupling, causing all th
e spins in neighboring grains to align, superseding in th
is way the anisotropic and demagnetizing forces.
Therefore there is a critical grain size, below which the
material will be single domain. For spherical grains, the
critical diameter is given by

 where B = 4(AK1)1/2 is the wall energy of the material; A is an exch

ange constant, also known as exchange stiffness, which is a functio
n of the material and temperature; K1 is the anisotropic constant;
0 is the permittivity of free space; and Ms is the saturation magne
tization.
Thus if the particle or grain size is below the critical diameter, the
material is single domain.
For example, the critical diameter for Co is around 70 nm, whereas
for Fe it is 15 nm.
If the particle or grain size becomes significantly smaller (typically a
few nm) than the critical diameter, the magnetization is likely to be
come unstable and loss of magnetization occurs due to thermal flu
ctuations.
These materials are called superparamagnetic.

 Other magnetic properties are also strongly affected by scal

e.
To address this point, first recall the magnetic response of a
bulk ferromagnetic material to an applied magnetic field
As previously discussed, the hysteresis shown in Figure 7.24
is associated with the fact that, on removal of the magnetic fi
eld, the magnetic domains do not revert to their original conf
iguration. In other words, there is a remnant magnetization.
On the other hand, the coercive field is the applied magnetic
field that needs to be applied in the direction opposite the in
itial magnetic field, to bring the magnetization back to zero.
By reducing the particle size or grain size to the nanoscale, t
he magnetization curve can be altered.

 In general, the coercive field of a ferromagnetic m

aterial increases with decreasing particle size or g
rain size, reaching a maximum within a range aro
und the critical diameter.
If the particle size or grain size is further decrease
d below this range, the coercive field will be drast
ically reduced until the magnetization becomes u
nstable due to the superparamagnetic behavior.
Within this regime, the hysteresis can be complet
ely removed at any temperature.