Electrical and Optical Properties of Materials
Electrical and Optical Properties of Materials
E z=0 z
E
E H A H C
B H
Material 1 Material 2
1
5. Optical properties of materials
At the boundary (z = 0), and for any time t, the total E field on each side
of the interface must be equal (and similarly for H). Thus:
EA EB EC
EA + EB = EC and − = (5.4)
Z1 Z1 Z2
Rearranging, we obtain the ratio of the reflected and transmitted amplitudes:
EB Z2 − Z1
reflected = (5.5)
EA Z2 + Z1
EC 2Z2
transmitted= (5.6)
EA Z2 + Z1
For the case where the relative permeability µr = 1, a valid assumption for
most materials at optical frequencies, we can express1 this in terms of the
refractive index n.
EB Z2 − Z1 n1 − n2
reflected = = (5.7)
EA Z2 + Z1 n1 + n2
EC 2Z2 2n1
transmitted = = (5.8)
EA Z2 + Z1 n1 + n2
Note that these represent the field amplitudes, and not power (which goes
as field squared). Similar expressions can be derived for the ratio of the
magnetizing fields H, using the relation that HA,B = EA,B /Z1 and HC =
EC /Z2 . These expressions should look reminiscent of those derived in your
quantum mechanics course for a particle incident on a potential barrier.
1 √ p
Reminder: n = r µr and Z = µ/
2
Electrical and optical properties of materials JJL Morton
B x=0
C
b
c
θb θc
θa
E a y
H
A x
Material 1 Material 2
3
5. Optical properties of materials
which is know as the Law of Reflection. While for the transmitted wave:
n1
n1 sin θa = n2 sin θc and so sin θc = sin θa (5.17)
n2
For nn12 sin θa ≤ 1, this reduces to θc = nn12 θa which is known as Snell’s Law
(for refraction) However, for nn12 sin θa > 1, there is no allowed value of θc ,
and there is no transmission (this shall be useful in total internal reflection
in optical fibres later on).
For the amplitudes, the algebra gets a little bit more cumbersome, but
this is derived in detail in Bleaney & Bleaney (3rd edition, Chapter 8, pp
239-244). Here we will simply quote the results, for the case where µr = 1.
4
Electrical and optical properties of materials JJL Morton
1 b
d
a c
n1 n2 n3
(simplified)
1
~1
a ~1
~c c
n1 n2 n3
The fact that there must be some reflection at any interface means that we
must use interference between two or more beams to reduce it to zero. The
general case is indicated schematically in Figure 5.3 for a coating or bloom of
refractive index n2 on the lens of index n3 . Using the results above, we note
the amplitude reflection and transmission coefficients are:
ni − nj 2ni
rij = and tij = (5.20)
ni + nj ni + nj
We can simplify the problem by assuming reflection is already reasonably
small (as shown in the lower part of Figure 5.3). For zero reflection we need
to ensure that the two components marked a and c are equal in amplitude
and π out of phase so as to destructively interfere. The amplitude of a = r12
and that of c = r23 , so these reflections coefficients will have to be equal for
any chance of complete destructive interference. Thus:
n1 − n2 n2 − n3
r12 = = r23 = (5.21)
n1 + n2 n2 + n3
√
This gives the solution n2 = n1 n3 , so we know what the refractive index n2
of the coating must be. But what about the thickness? Wave c has travelled
an additional length 2L, and so will have picked up an extra phase (think
about the exp(ini k0 z) term for a wave):
φ = 2Ln2 k0 (5.22)
5
5. Optical properties of materials
2Ln2 k0 = π (5.23)
1 π 1 λ0 λ2
L= = = , (5.24)
n2 2k0 n2 4 4
where λ0 = 2π/k0 is the wavelength of the wave, were it to be travelling in
free space and λ2 is the wavelength of the wave when in material 2. Thus,
blooming with a layer of the appropriate thickness will remove all reflection
at a particular wavelength. This is also called a ‘quarter-wave’ coating.
What if we want to do the opposite, and have only reflection — i.e. create a
mirror? This can be achieved using multiple layers of very low loss dielectric,
as shown in Figure 5.4. The total reflected wave is the sum of reflections
from each interface in the layered structure. When the refractive index of
the incident material is less than the current one (nj > nj+1 ), the reflection
coefficient rj,j+1 is positive and there no phase shift on reflection. In the
opposite case (nj < nj+1 ), the reflection coefficient rj,j+1 is negative and
reflection is accompanied by an instantaneous π phase shift in the wave. Let’s
count up all the contributions to the reflected wave (assuming n1 > n2 > n0 ):
In other words, all waves are in phase and the assembly acts as a mirror.
Efficiencies can be very high (e.g. 99.8% compared with 98.5% for the best
silvered mirror).
6
Electrical and optical properties of materials JJL Morton
i1 i2 i3 i4 i5
1 ~1 ~1 ~1 ~1
r1 r2
~r2 r3 etc...
~r3 r4
r5
n0 n1 n2 n1 n2
Material 2
C
n2
θc
90-θc
θa
Material 1
n1
Optic fibres or ‘light pipes’ operate on the principle of total internal re-
flection. Above we showed that there is no transmission at an interface (from
region 1 into region 2) when the angle of incidence θ is such that:
n1
sin θ > 1 (5.27)
n2
Take some light entering a light pipe with an angle of incidence θa , as shown
in Figure 5.5. At its front surface, light is accepted into the material provided:
n1 n2
sin θa ≤ , with sin(90 − θc ) = cos θc = sin θa (5.28)
n2 n1
7
5. Optical properties of materials
For most real optic fibers, the surrounding material (with n2 ) is not air,
but rather some cladding to provide mechanical support to the light pipe.
The basic requirements for an optical fibre are:
* good flexibility
8
Electrical and optical properties of materials JJL Morton
2
We have said there is total internal reflection, when in fact there is always decaying
standing wave which penetrates the interface. As long as it is of negligible amplitude by
the time the refractive index changes again, one is alright. However, for a thin layer of
material it is possible for the light to leak out, just as in quantum tunneling through a
barrier
9
5. Optical properties of materials
Figure 5.7: Attenuation versus wavelength for a typical fibre. Also shown
are some available sources and detectors
10
Electrical and optical properties of materials JJL Morton
In our work on electromagnetic waves we have assumed the electric and mag-
netic field are fixed along x and y, respectively, perpendicular to the direction
of propagation z. In this sense we have already been assuming one particular
polarisation (though everything we’ve done could be generalised). The kind
of polarisation we have used is plane polarised, and we could rotate Ex and
Hy around the x − y plane to generate other kinds of plane polarisation.
Ordinary (unpolarised) light has equal intensity of electric field vectors
along both x and y and so there is no net polarisation, and can be thought
of as a combination of two light beams of orthogonal polarisations (see Fig-
ure 5.9).
B E
E
B
11
5. Optical properties of materials
the full amplitude of the incident light (at that polarisation). Thus, as an
alternative, polarisation may be obtained as an increasing fraction of the
transmitted wave, as illustrated in Figure 5.11.
12
Electrical and optical properties of materials JJL Morton
Figure 5.12: The Nicol prism for the production of polarised light
O and extraordinary E rays inside the calcite crystal, with refractive indices
n(O) = 1.66 and n(E) = 1.49. Two pieces of the crystal are glued together
using Canada balsam, which is not birefringent and has n = 1.55). Thus, at
an appropriate range of angles, one of the rays (O say) is totally internally
reflected while the other (E say) is transmitted. Note that the reflected ray
contains O and a component of E which was reflected also at the interface,
so it contains only partial polarisation.
We may take two polarising devices (e.g. two sheets of polaroid) and place
them in series in the path of an initially unpolarised beam of light. If they
are perpendicular then we expect ‘no’ light to get through. However, if we
put some test system between them which has the potential to rotate the
polarisation of light, we will see this in the transmitted light. This kind of
system is referred to as a polariser and analyzer, with the incident light falling
on the polariser, passing through the test system, and then the analyzer. In
the absence of any experimental system, transmission obeys a fairly obvious
law, known as Malus’ Law whereby the transmitted field strength goes as:
A1 = A0 cos θ (5.32)
13
5. Optical properties of materials
where θ is the angle between the polarisers and I0 is the intensity of incident
light (assuming no losses at all from the transmitted polarisation, which is
never really the case).
Two primary methods for rotating the polarisation of light use i) optically
active chemicals or ii) magnetic fields (known as the Faraday or Kerr effect
depending on whether transmission or reflection is used). We shall not be
studying these, but rather looking at transmission through a series of n an-
alyzers, each rotated by a small angle δθ with respect to its predecessor (as
shown in Figure 5.13). Following Eq. 5.32,
An = A0 cosn δθ (5.34)
In = I0 cos2n δθ (5.35)
Using a Taylor expansion for cos (assuming δθ is small)3 :
2n
δθ2
In ≈ I0 1 − , (5.36)
2
In ≈ I0 1 − nδθ2
(5.37)
If the total angle between the first and the last analyzer is φ = nδθ, then:
φ2
In = I0 1 − (5.38)
n
The polarisation of the light coming out of the final analyzer is determined
by the angle of the final analyzer, and hence a total polarisation rotation
of φ has been accomplished, but at some cost in intensity. Plugging some
numbers in, for φ = π/2 and n = 10, 75% of the intensity is transmitted,
while for n = 40, the transmission is 94%. So, provided the ‘twist’ angle is
small enough, the plane of polarisation can be ‘guided’ round without too
much loss.
3
Taylor expansion: cos θ ≈ 1 − θ2 /2
4
Taylor expansion: (1 + x)m ≈ 1 + xm
14
Electrical and optical properties of materials JJL Morton
δθ φ
light
Figure 5.13: Slowly rotating analyzers can rotate the polarisation of light
Liquid crystals are a particular group of materials which show a ‘double’ melt-
ing point: at the lower melting point the solid ‘melts’ but remains ordered
in at least one dimension until some higher melting temperature is reached
and the material becomes truly liquid. Between these two temperatures the
material is called a liquid crystal. There are three types of liquid crystal:
nematic, cholesteric and smectic, though only the first is widely used for its
electrical properties. In the nematic class, the rod-shaped (or long ellipsoidal)
molecules tend to line up parallel to each other locally — the extent of align-
ment may be enhanced by the application of relatively small electric fields.
In terms of the various polarisation mechanisms described in Part 2 of this
course, liquid crystals behave a bit like large permanent dipole fluid particles
at low temperature (although the molecules are not permanently polarised).
Nematic liquid crystals have the further property of aligning with respect to
a true solid surface with which they are in contact (e.g. Figure 5.14) with
either perpendicular (homeotropic) or parallel (homogeneous) ordering.
15
5. Optical properties of materials
Although the required field 150 kV/m may seem high, this can be achieved
with a relatively small voltage (1.5 V) across the 10 µm gap. The response
time of the realignment is < 100 ms, which is adequate for most display
purposes. This time dependence arises from Debye relaxation.
16
Electrical and optical properties of materials JJL Morton
the liquid crystal is held between two plates and then twisted by 90◦ , in the
absence of any electric field, polarised light will follow the molecular axes and
undergo a 90◦ rotation. If an electric field is now applied so as to alter the
nematic alignment, the polarisation no longer rotates (see Figure 5.16). Suit-
ably arranged polariser/analyser combinations can thus distinguish between
the two cases, and light can controllably be allowed to transmit through the
device, or vice versa. A typical arrangement of the actual components is
shown in Figure 5.17, while Figure 5.18 shows how this might be constructed
in practice. The reflection mode is typically used in watches or pocket cal-
culators, allowing them to work well around sources of bright ambient light
(e.g. outside).
A.1 Acknowledgements
These notes are based extensively on those of Mike J Goringe, with additional
input from Peter Wilshaw and Brendon Lovett. Please don’t hesitate to email
me at john.morton@materials.ox.ac.uk to point out any errors/corrections or
for any other feedback.
17
5. Optical properties of materials
18