Optical Properties of Materials

Overview
• The study of the optical properties of materials is
a huge field and we will only be able to touch on
some of the most basic parts
• So we will consider the essential properties such
as absorption/reflection/transmission and
refraction
• Then we will look at other phenomena like
luminescence and fluorescence
• Finally we will mention applications, in particular
optical fibres and lasers
 Nature of light
• Light is an electromagnetic wave:

 with a velocity given by c = 1/(00) = 3 x 108 m/s

• In view of this, it is not surprising that the electric
field component of the wave should interact with
electrons electrostatically
http://www.astronomynotes.com/light/emanim.gif
• Many of the electronic properties of materials,
information on the bonding, material composition
etc. was discovered using spectroscopy, the
study of absorbed or emitted radiation
 evidence for energy levels in atoms
 evidence for energy bands and band-gaps
 photoelectric effect
 General description of absorption
• Because of conservation of energy, we can say
that I0 = IT + IA + IR
 Io is the intensity (W/m2) of incident light and subscripts refer
to transmitted, absorbed or reflected
• Alternatively T + A + R = 1 where T, A, and R are
fractions of the amount of incident light
 T = IT/I0, etc.
• So materials are broadly classed as
 transparent:relatively little absorption
and reflection
 translucent:light scattered within
the material (see right)
 opaque:relatively little transmission

http://www.tekano.pwp.blueyonder.co.uk/tekano/translucent.jpg
• If the material is not perfectly transparent, the
intensity decreases exponentially with distance
• Consider a small thickness of material, x
• The fall of intensity in x is I so I = -.x.I
 where  is the absorption coefficient (dimensions are m -1)
• In the limit of x  0, we get
dI
  I
dx
• The solution of which is I = I0 exp(–x)
• Taking “ln” of both sides, we have:
 I 
x   ln 
 I(he
which is known as Lambert’s Law 0  also has a unit of light
intensity named for him)
• Thus, if we can plot -ln(I) against x, we should
find  from the gradient
• Depending on the material and the wavelength,
light can be absorbed by
 nuclei – all materials
 electrons – metals and small band-gap materials
 ATOMIC ABSORPTION
• How the solid absorbs the radiation depends on
what it is!
• Solids which bond ionically, show high
absorption because ions of opposite charge
move in opposite directions
 in the same electric field
 hence we get effectively twice the interaction between the
light and the atoms
• Generally, we would expect absorption mainly in
the infrared
 because these frequencies match the thermal vibrations of
the atoms
• If we think of our atom-on-springs model, there is
a single resonance peak:

absorption

f
f0

• But things are more complex when the atoms are

connected – phonons
 recall transverse and longitudinal optical phonons
 Electronic absorption
• Absorption or emission due to excitation or
relaxation of the electrons in the atoms

http://www.nhn.ou.edu/~kieran/reuhome/vizqm/figs/hydrogen.gif
 3.0 3.0
kondisi awal
Kontrol kondisi awal Fotoreaktor PD
8 jam iradiasi 8 jam iradiasi
2.5 16 jam iradiasi 2.5 16 jam iradiasi
24 jam iradiasi 24 jam iradiasi

A bsorbansi (a.u)
A bsorbansi(a.u)

2.0 32 jam iradiasi 2.0 32 jam iradiasi

40 jam iradiasi 40 jam iradiasi
1.5 48 jam iradiasi 1.5 48 jam iradiasi

1.0 1.0

0.5 0.5

0.0 0.0
400 500 600 700 800 400 500 600 700 800

Panjang gelom bang (nm ) P anjang gelom bang (nm )

100
kontrol reaktor Fotoreaktor:
fotoreaktor
80
96,54% MB terdegradasi selama 48
jam iradiasi (reaksi fotokatalisis)
C /C 0 M B (% )

60
Kontrol reaktor:
40 27,71% MB terdegradasi selama 48
jam iradiasi (reaksi fotolisis)
20

0
0 8 16 24 32 40 48
W aktu iradiasi (jam )
 Molecular materials
• Materials such as organic (carbon containing)
solids or water consist of molecules which are
relatively weakly connected to other molecules
• Hence, the absorption spectrum is dominated by
absorptions due to the molecules themselves
• e.g. water molecule:

http://www.sbu.ac.uk/water/images/molecul5.jpg
• The spectrum of liquid water

http://www.sbu.ac.uk/water/images/watopt.jpg
 TiO 2/K A
1.00

Transm itansi
0.99

0.98
KA
1.00
Transm itansi

0.99

0.98
TiO 2
1.00
Transm itansi

0.99

0.98
4000 3500 3000 2500 2000 1500 1000 500
-1
B ilangan G elom bang (cm )
• Since the bonds have different “spring
constants”, the frequencies of the modes are
different
 when the incident illumination is of a wavelength that excites
one of these modes, the illumination is preferentially
absorbed
• This technique allows us to measure
concentrations of different gas species in, for
example, the atmosphere
 by fitting spectra of known gases to the measured
atmospheric spectra, we can figure out the quantities of
each of the gases
 Optical properties of metals
• Recall that the energy diagram of a metal looks like:

empty T = 0K
levels
EF

full
 EF levels
is the energy below which, at 0K, all electron states are full
and above which they are empty
 this is the Fermi Energy
• For T > 0, EF is the energy at which half of the
available energy states are occupied
• Semiconductors also have a Fermi level
 for an intrinsic material EF is in the middle of the bandgap
 nearer Ec for n-type; nearer Ev for p-type
• This structure for metals means that almost any
frequency of light can be absorbed
• Since there is a very high concentration of
electrons, practically all the light is absorbed
within about 0.1µm of the surface
• Metal films thinner than this will transmit light
 e.g. gold coatings on space suit helmets
• Penetration depths for some materials are:
 water: 32 cm
 glass: 29 cm
 graphite: 0.6 µm
 gold: 0.15µm
• So what happens to the excited atoms in the
surface layers of metal atoms?
 they relax again, emitting a photon
• The energy lost by the descending electron is the
same as the one originally incident
• So the metal reflects the light very well – about
95% for most metals
 metals are both opaque and reflective
 the remaining energy is usually lost as heat
• In terms of electrostatics, the field of the radiation
causes the free electrons to move and a moving
charge emits electromagnetic radiation
 hence the wave is re-emitted = reflected
• The metal appears “silvery” since it acts as a
perfect mirror
• OK then, why are gold and copper not silvery?
 because the band structure of a real metal is not always as
simple as we have assumed
 there can be some empty levels below EF and the energy re-
emitted from these absorptions is not in the visible spectrum
• Metals are more transparent to very high energy
radiation (x- & - rays) when the inertia of the
electrons themselves is the limiting factor
• Reflection spectra for gold and aluminum are:

aluminum
spectrum is
relatively flat

gold reflects lots of

red wavelengths

blue red

http://www.thermo.com/eThermo/CMA/Images/Various/109Image_12275.gif
 Electronic absorption in non-metals
• Dielectrics and semiconductors behave essentially
the same way, the only difference being in the size
of the bandgap
• We know that photons with energies greater than E g
will be absorbed by giving their energy to electron-
hole pairs

EC

EG

EV

 which may or may not re-emit light when they relax

hole
• Hence, the absorption coefficients of various
semiconductors look like: Photon energy (eV)
5 4 3 2 1 0.7
0.9 0.8
108

Ge
.

107 In0.7Ga0.3As0.64P0.36

In0.53Ga0.47As
Si
106
( -1 )

GaAs
InP
m

105
a-Si:H

104

103
0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8
Wavelength (m)

Fig. 9.19: Absorption coefficient () vs. wavelength () for

various semiconductors (Data selectively collected and combined
from various sources.)
From Principles of Electronic Materials and Devices, Second Edition, S.O. Kasap (© McGraw-Hill, 2002)
http://Materials.Usask.Ca
• Semiconductors can appear “metallic” if visible
photons are all reflected (like Ge) but those with
smaller Eg, such as CdS look coloured
 yellow for CdS which absorbs 540nm and above
• The above picture is good for pure materials but
impurities can add extra absorption features
EC
hf1 phonon

hf2

EV
• Impurity levels divide up the bandgap to allow
transitions with energies less than Eg
• Recombination can be either radiative (photon) or
non-radiative (phonon) depending on the
transition probabilities
• Practical p-n diodes usually contain a small
amount of impurity to help recombination
because Si has a relatively low recombination
“efficiency”
 for the same reason that Si is inefficient at generating light
 Refraction in non-metals
• One of the most important optical properties of
non-metallic materials is refraction
• This refers to the bending of a light beam as it
passes from one material into another
 e.g. from air to glass
• We define the index of refraction to be
n = c/v
 where c is the speed of light in a vacuum and v is the speed
of light in the material (which is in general wavelength-
dependent)
• A familiar example is the prism where the
different amounts of bending separates out the
wavelengths
• Refraction is also vital for other applications,
such as:
 optical fibres – keeps the light in
 semiconductor laser – keeps the light in the amplifying cavity
of the laser
• Given that
1 1
v and c 
  0 0
 where µ and µ0 (= µrµ0) are the permeability of the material
and free space, respectively (a magnetic property)
 and  and 0 (= r0) are the permittivity of the material and
free space, respectively (an electrostatic property)
• We find that n = √(µrr) (≈ √r for many materials)
• Since light is an electromagnetic wave, the
connection with both the dielectric permittivity ()
and the magnetic permeability (µ) is not
surprising
• The index of refraction is therefore a
consequence of electrical polarization, especially
electronic polarization

– +

• Hence, the radiation loses energy to the electrons

• Since E = hv/, and  doesn’t change, the velocity
must be smaller in the material than in free space
 since we lose E to the atoms, v must also decrease
• Electronic polarization tends to be easier for
larger atoms so n is higher in those materials
 e.g. glass: n ~ 1.5
 lead crystal: n ~ 2.1 (which makes glasses and chandeliers
more sparkly!)
• n can be anisotropic for crystals which have non-
cubic lattices
 Reflection in non-metals
• Reflection occurs at the interface between two
materials and is therefore related to index of
refraction
• Reflectivity, R = IR/I0, where the I’s are intensities
• Assuming the light is normally incident to the
interface:
n  n 2
R   2 1 
n2  n1 
n1 n2
 where n1 and n2 are the indices for the two materials
• Optical lenses are frequently coated with
antireflection layers such as MgF2 which work by
reducing the overall reflectivity
 some lenses have multiple coatings for different wavelengths
 Spectra
• So we have seen that reflection and absorption
are dependent on wavelength
 and transmission is what’s left over!
• Thus the three components for a green glass are:

Callister Fig. 21.8

 Colours
• Small differences in composition can lead to large
differences in appearance
• For example, high-purity single-crystal Al2O3 is
colourless
 sapphire
• If we add only 0.5 - 2.0% of Cr2O3 we find that the
material looks red
 ruby
• The Cr substitutes for the Al and introduces
impurity levels in the bandgap of the sapphire
• These levels give strong absorptions at:
 400nm (green) and 600nm (blue)
 leaving only red to be transmitted
• The spectra for ruby and sapphire look like:

• A similar technique is used to colour glasses or

pottery glaze by adding impurities into the molten
state:
 Cu2+: blue-green, Cr3+: green
 Co2+: blue-violet, Mn2+: yellow
http://www.valleydesign.com/images/sapp.jpg
http://home.achilles.net/~jtalbot/glossary/photopumping.gif
 Translucency
• Even after the light has entered the material, it
might yet be reflected out again due to scattering
inside the material
• Even the transmitted light can lose information
by being scattered internally
 so a beam of light will spread out or an image will become
blurred
• In extreme cases, the material could become
opaque due to excessive internal scattering
• Scattering can come from obvious causes:
 grain boundaries in poly-crystalline materials
 fine pores in ceramics
 different phases of materials
• In highly pure materials, scattering still occurs
and an important contribution comes from
Rayleigh scattering
• This is due to small, random differences in
refractive index from place to place
• In amorphous materials such as glass this is
typically due to density or compositional
differences in the random structure
• In crystals, lattice defects, thermal motion of
atoms etc. also give rise to Rayleigh scattering
• Rayleigh scattering also causes the sky to be
blue. The reason for this is the wavelength-
dependence of Rayleigh scattering
 scattering goes as -4
 so since red ~ 2blue blue light is scattered ~16 times more
than red light
• This mechanism is of great technological
importance because it governs losses in optical
fibres for communication
• But before we get onto fibres, we will mention a
couple more basic effects
 A dielectric particle smaller than wavelength

Incident wave Through wave

Scattered waves

Fig. 9.21: Rayleigh scattering involves the polarization of a small

dielectric particle or a region that is much smaller than the light
wavelength. The field forces dipole oscillations in the particle (by
polarizing it) which leads to the emission of EM waves in "many"
directions so that a portion of the light energy is directed away from the
incident beam.
From Principles of Electronic Materials and Devices, Second Edition, S.O. Kasap (© McGraw-Hill, 2002)
http://Materials.Usask.Ca
 Dispersion
• Dispersion is a general name given to things
which vary with wavelength
• For example, the wavelength-dependence of the
index of refraction is termed the dispersion of the
index
• Another important case arises because the speed
of the wave depends on its wavelength
• If a pulse of white light is transmitted through a
material, different wavelengths arrive at the other
end at different times
 this is also called dispersion
 Luminescence
• Luminescence is the general term which
describes the re-emission of previously absorbed
radiative energy
• Common types are photo- , electro-, and cathodo-
luminescence, depending on whether the original
incident radiation was
 light of a different wavelength – e.g. fluorescent light
 electric field – e.g. LED
 electrons – e.g. electron gun in a cathode ray tube (CRT)
• There is also chemo-luminescence due to
chemical reactions which make the glowing rings
seen at fairgrounds!
• Luminescence is further divided into
phosphorescence and fluorescence
• Fluorescence and phosphorescence are
distinguished by the electron transitions
requiring no change or a change of spin,
respectively
 hence fluorescence is a faster process because no change
of spin is required, around 10-5 – 10-6s
 phosphorescence takes about 10-4 – 101s
• Thus the energy diagram might be like:
E2 flip phosp.
fluor. E3
incident
phosp.
E1 flip
• If the energy levels are actually a range of
energies, then:
phonon emission
~10-12s per hop

fluorescence, ~10-5s

• So the light emitted by fluorescence is of longer

wavelength than the incident light
 since the energy is smaller
 and phosphorescent light is typically longer wavelength than
fluorescent light
• In fluorescent lights, the plasma generates UV
light, and a fluorescent coating on the walls of
the tube converts this to visible light
 these lights have a visible flicker because (60Hz)-1 > 10-5s
• Rather confusingly, materials that do this are
generally called phosphors
• To obtain a white light, a mixture of phosphors
must be used, each fluorescing at a different
wavelength
• TV tubes usually use materials doped with
different elements to give the colours:
 ZnS doped with Cu+ gives green
 ZnS:Ag gives blue
 YVO4:Eu gives red
 Optical fibres
• Fibre-optic technology has revolutionised
telecommunications owing to the speed of data
transmission:
 equivalent to >3 hrs of TV per second
 24,000 simultaneous phone calls
 0.1kg of fibre carries same information as
30,000kg of copper cable
• Owing to attenuation in the cable, transmission is
usually digital and the system requires several
sections: optical optical

encoder conversion repeater detection decoder

to optical
http://www.ngflscotland.gov.uk/connected/connected5/images/fibreoptic.jpg
• Obviously, the loss in the cable is important
because is determines the maximum
uninterrupted length of the fibre
• We know that losses depend on the wavelength
of the light and the purity of the material
 recall the penetration depth for glass was ~30cm
• In 1970, 1km of fibre attenuated 850nm light by a
factor of 100
• By 1979, 1km of fibre attenuated 1.2µm light by a
factor of only 1.2
 this light is infrared
• Now, over 10km of optical fibre silica glass, the
loss is the same as 25mm of ordinary window
glass!
• For such high-purity materials, Rayleigh scattering is
the dominant loss mechanism:
10
.
5 water
-1 OH-absorption peaks

1.0
0.5 1550 nm

Rayleigh Lattice
scattering absorption
km
A

dB
1310 nm
en

on
ua

0.1
tt

ti

0.05
0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Wavelength (µm)
Fig. 9.22: Illustration of a typical attenuation vs. wavelength characteristics
of a silica based optical fiber. There are two communications channels at
1310 nm and 1550 nm.
From Principles of Electronic Materials and Devices, Second Edition, S.O. Kasap (© McGraw-Hill, 2002)
http://Materials.Usask.Ca
• The Rayleigh scattering results from minute local
density variations which are present in the liquid
glass due to Brownian motion and become frozen
into the solid
• The really clever part about optical fibres is that
the light is guided around bends in the fibre
• This is achieved by total internal reflection at the
boundary of the fibre
• Thus, the cross section of the fibre is designed
as follows

http://www.datacottage.com/nch/images/fibreconstruct.gif
• The light is transmitted in the core and total
internal reflection is made possible by the
difference in the index of refraction between the
cladding and the core
• A simple approach is the “step-index” design:

n
• The main problem with this design is that
different light rays follow slightly different
trajectories
• So different light rays from an input pulse will
take slightly different paths and will therefore
reach the output at different times
• Hence the input pulse is found to broaden during
transmission:
signal signal

t t
in out
• This limits the data rate of digital communication
• Such broadening is largely eliminated by using a
“graded-index” design:

• This is achieved by doping the silica with B2O3 or

GeO2 parabolically as shown above
• Now, waves which travel in the outer regions, do
so in a lower refractive index material
 and their velocity is higher (v = c/n)
• Therefore, they travel both further and faster
 as a result, they arrive at the output at almost the same time
as the waves with shorter trajectories
• Anything that might cause scattering in the core
must be minimised
 Cu, Fe, V are all reduced to parts per billion
 H2O and OH concentrations also need to be very low
• Variations in the diameter of the fibre also cause
scattering
 this variation is now <1µm over a length of 1km
• To avoid dispersion of different wavelengths,
lasers are used as the light sources
 many data channels are possible using wavelength division
multiplexing (WDM)
• A convenient fact is that compound
semiconductor lasers can emit IR light close to
the 1.55µm wavelength where the fibre absorbs
least
• Referring back to the system diagram, it would be
advantageous to integrate the encoder and
transmitter
 so the circuits and the light emitter can be integrated
• This is why there is so much interest in getting
light out of porous silicon or Si compounds
 where thin strands of material exhibit quantum-mechanical
effects which adjust the Si band structure to facilitate
efficient light emission
http://ghuth.com/Porous%20silicon.jpg

http://porous.silicon.online.fr/images/poreux.jpg
 Lasers
• LASER stands for Light
Amplification by the Stimulated
Emission of Radiation
• The key word here is “stimulated”
• All of the light emission we have mentioned so far
is spontaneous
 it happened just due to randomly occurring “natural” effects
• Stimulated emission refers to electron transitions
that are “encouraged” by the presence of other
photons
• Einstein showed that an incident photon with E ≥
Eg was equally likely to cause stimulated emission
of light as to be absorbed
http://www.007sdomain.com/gf_laser.jpg
 equally likely
as

• The emitted light has the same energy and phase

as the incident light (= coherent)
• Under normal circumstances, there are few
excited electrons and many in the ground-state,
 so we get predominantly absorption
• If we could arrange for more excited than non-
excited electrons, then we would get mostly
stimulated emission
• Since we get more photons out than we put in,
this is optical amplification
 hence lAser
 this system was first used to amplify microwaves for
communications (maser)
• Such a condition is called a population inversion
• This stimulated emission is what gives the laser
its coherent output
 which is what makes it useful for holography, for example
• Clearly, random spontaneous emission “wastes”
electron transitions by giving incoherent output
 so we minimise them by using transitions for which the
spontaneous emissions are of low probability
 so-called metastable states
• The energy levels of a laser material therefore look
like:

• Ruby is a common laser material, which we saw

was Al2O3 (sapphire) with Cr3+ impurities

http://kottan-labs.bgsu.edu/teaching/workshop2001/chapter4a_files/image022.gif
• So all we need to make a laser is to achieve
 (i) a population inversion
 (ii) enough photons to stimulate emission
• The first is achieved by filling the metastable
states with electrons generated by light from a
xenon flash lamp
• The second condition is achieved by confining
the photons to travel back and forth along the rod
of ruby using mirrored ends
 next slide
• The ruby laser has an output at 694.3 nm
http://www.repairfaq.org/sam/laserop.gif
• In order to keep the coherent emission, we must
ensure that the light which completes the round
trip between the mirrors returns in phase with
itself
• Hence the distance between the mirrors should
obey 2L = N
 where N is an integer,  is the laser wavelength and L is the
cavity length
• Semiconductor lasers work in just the same way
except that they achieve the population inversion
electrically
 by using a carefully designed band structure
• Some laser characteristics are given in the
following table:

Callister
 Summary
• We have looked at how the electronic structure of
atoms and their bonding leads to varying optical
behaviours in materials
• In particular, properties such as absorption and
emission are closely related to the electrons
• Applications of this knowledge include
 anti-reflective coatings for lenses
 fibre-optic communications
 lasers
 Closing remarks
• this first half of ENG2000 is an introduction to a
subject area that is very subtle, and the course
covers a huge range of subjects
• As you gain more experience, the pieces of the
jigsaw will fit better and better
• So, if all the connections etc are not crystal clear
right now, have patience!
• For me, the success of the course is how often
you say “oh yes, we saw that in ENG2000” !
THE END