Interference 131

C H A P T E R

6 Interference

6.1 INTRODUCTION
In 1678 Christian Huygens propounded the wave theory of light. He suggested that light
propagates in the form of waves. Huygens did not suggest anything about the nature of light
wave; whether it is a transverse wave or longitudinal wave. He had no knowledge about
the speed of light or its wavelength. In 1801 Thomas Young provided the first experimental
evidence for the wave theory of light from the double slit interference experiment and
determined the wavelength of light waves. The interference is not easy to observe in case of
light waves. One cannot observe interference fringes on a screen by placing two light sources
in front of it. Sustained interference would be obtained only when coherent sources of light
are used. Special devices, such as Lloyd’s mirror, Fresnel biprism are designed to create two
coherent sources from a single source of light and to determine accurately the wavelength of
light. The striking colour effects caused by white light on a soap bubble and on a slick of oil
spreading over small puddles of water are familiar to us from our childhood. Less noticed are
the colours exhibited by coatings of oxide on heated metal pans in kitchen and thin layers
of mica, cellophane etc. Thomas Young explained the origin of colours in all these cases
on the basis of interference. The interference of certain colours constructively and others
destructively causes the exhibition of iridescent colours by the thin films in reflected light.
An understanding of the thin film interference phenomenon has led to a number of practical
applications such as non-reflecting coatings, interference filters, and interference mirrors.

6.2 INTERFERENCE
p
Interference is an important consequence of superposition of
waves. Let us consider two (or more) light waves of same frequency
and having a constant phase difference travel in the same region S
1
of a medium simultaneously and cross each other (see Fig.6.1).
Waves having the same frequency and a constant phase difference S2
are known as coherent waves. Waves having fluctuating phase
difference are known as incoherent waves. At the point of crossing, Fig. 6.1
waves overlap or superpose on each other.
According to the principle of superposition, the combined effect of coherent waves at
each point of the region of superposition is obtained by adding algebraically the disturbances
due to individual waves. The resultant intensity at any point in the region of superposition
depends upon the amplitudes and the phase relationships of the component waves. Let us
assume here that the component waves are of the same amplitude, say A.
131
132 A Textbook of Engineering Physics

We focus our attention mainly on two specific types of disposition of the waves (see
Fig.6.2) at the point of observation, P.

time

+ = time

time
(a)

time

+ = time
time
(b)

Fig. 6.2
(i) If the two waves, meeting at P, reach their maxima, zeros and minima at the same
instant of time, then their phase difference is zero (or an integral multiple of 2p).
Such waves will have a crest-to-crest and trough-to-trough correspondence, as
shown in Fig.6.2 (a). Then the waves are said to be in phase. The amplitude of the
resultant wave at the point will then be equal to the sum of the amplitudes of the two
waves, as shown in Fig.6.2 (a).
(a) The amplitude of the resultant wave = A + A = 2A. Hence, the intensity of the
resultant wave is IR ∝ 22 A2 = 22 I. It is obvious that the resultant intensity is greater
than the sum of the intensities (I + I = 2I) due to individual waves. The interference
produced at such points is known as constructive interference. When two waves
are not displaced with respect to each other or when they are displaced through an
integral number of wavelengths, constructive interference takes place. Bright bands
of light are observed at those points. As S1 and S2 are coherent sources, the bright
bands are stationary.
(ii) If one of the waves reaches its maximum at the same time when the other reaches
its minimum, then their phase difference is p radians. Such waves have a crest-to-
trough correspondence, as shown in Fig. 6.2(b). The waves are said to be in opposite
phase or 180° out-of-phase. The amplitude of the resultant wave at the point will
also be equal to the sum of the amplitudes of the two waves, as shown in Fig. 6.2(b).
As the amplitude of one of the waves is negative, the amplitude of the resultant
wave = A – A = 0. Hence, the intensity of the resultant wave is IR ∝ 02 = 0. It is
obvious that the resultant intensity is less than the sum of the intensities (I + I =
2I) due to individual waves. The interference produced at such points is known as
destructive interference. When two waves are displaced with respect to each other
by an odd number of half-wavelengths, destructive interference results. Dark bands
of light are observed at those points. As S1 and S2 are coherent sources, the dark
bands are stationary.
Thus, when two or more coherent waves of light are superposed, the resultant effect is
that at certain points brightness is produced while at other points darkness is produced in the
medium.
Interference 133

The phenomenon of redistribution of light energy due to the superposition of light waves
from two or more coherent sources is known as interference. The stationary bands of alternate
darkness and brightness are known as fringes. The fringe pattern is obtained only when the
interfering waves are coherent.
When incoherent waves overlap on each other, the resultant intensity is a simple
addition of the intensities in the region of overlap. Therefore, the resultant intensity due to
two incoherent waves of equal intensity is I + I = 2I and the region of overlap is uniformly
bright without the formation of fringes. It means that interference does not take place when
incoherent waves superpose on each other.

6.3 CONDITIONS FOR OBSERVING SUSTAINED INTERFERENCE

We may now summarize the conditions that are to be fulfilled in order to observe sustained
interference.
(i) The waves from the two light sources must be of the same frequency.
(ii) The waves from the two light sources must maintain a constant phase difference.
(iii) For obtaining distinct bright and dark fringes, the vector sum of the overlapping
electric field vectors should be zero in the dark regions. The sum will be zero only if
the vectors are anti-parallel and have the same magnitude.
(iv) If the two sets of waves are R
P Q
plane polarized, their planes of
polarization must be the same. 1
Waves polarized in perpendicular
planes cannot produce interference
effects.
2
(v) The path difference between the
overlapping waves must be less S T U
than the coherence length of the Fig. 6.3
waves (see Art.5.20.1). If we
consider two interfering wave trains, having constant phase difference, as in Fig.6.3,
the interference effects occur due to parts QR of wave 1 and ST of wave 2. For the
parts PQ and TU interference will not occur. Therefore, the interference pattern does
not appear distinctly. When the entire wave train PR overlaps on the wave train SU,
interference pattern will be distinct. On the other hand, when the path difference
between the waves 1 and 2 becomes very large, the wave trains arrive at different
times and do not overlap on each other. Therefore, in such cases interference does not
take place. The interference pattern completely vanishes if the path difference is equal
to the coherence length. It is hence required that
D < lcoh (6.1)
(vi) The two coherent sources must lie close to each other in order to discern the fringe
pattern.

6.4 TECHNIQUES OF OBTAINING INTERFERENCE

The phase relation between the waves emitted by two conventional light sources fluctuates
rapidly and therefore they can never be coherent, though they are identical in all respects.
However, two coherent sources are derived from a single source by techniques, which can be
divided into two broad classes.
134 A Textbook of Engineering Physics

1. Division of wave front: One of the methods consists in using a narrow slit as the
source and subsequently, the wave front is divided. For example, in the Young’s
double slit experiment, a wave front emerging from the slit S is divided into two
parts by the double slit S1S2. Fresnel’s biprism, Lloyd’s mirror, etc are the other
examples where the division of wave front method is used.
2. Division of amplitude: In this method, amplitude of the light beam is divided
by partial reflection into two or more beams. Thin films (wedge, Newton’s rings
etc), interferometers such as Michelson’s interferometer etc utilize this method in
producing interference.
6.5 REVIEW OF IMPORTANT CONCEPTS
When studying the topic of interference, we frequently come across the terms optical path and
path difference. We should clearly understand these terms and know how to calculate them.
6.5.1 Geometrical Path
Light travels along a straight line path from a point A to another point B and it is known as the
path of the light. The shortest path between any two points A and B is called the geometrical
path length (GPL). GPL remains the same whether it is measured in a vacuum or in any
medium.
6.5.2 Optical Path
Light travels m times slower in a medium. Therefore, it takes m times more time to cover the
distance AB in the medium than it takes to cover the same distance in a vacuum. This time
delay is accounted for by introducing another distance called optical path length (OPL). It is
defined as
O.P.L. = m × G.P.L.
or D = mL (6.2)
The optical path length, D signifies the number of wavelengths that are accommodated in
a given medium over the corresponding geometrical path length.
6.5.3 Path Difference
Light rays travel along different paths, which may lie in the same medium or in different
media. The difference between optical paths of two rays travelling in different directions is
known as the optical path difference.
6.5.4 Phase Difference
The phase of a wave arriving at a point depends on the optical path length it traversed. We
know that if a wave covers in air a distance of one wavelength, 1l, its phase changes by 2p
radians. Therefore, we compute that if a wave travels a distance L in air, its phase change is
given by
2π L
d= (6.3)
λ
When the wave travels the distance L in a medium, then
2π ∆ 2πµL
d= = (6.4)
λ λ
Comparing equs. (6.3) and (6.4), we find that a light path of geometric length L in a
medium of refractive index m produces the same phase change as a light path of length mL
in a vacuum. Therefore, in the study of optics we always must calculate the optical paths
travelled by light rays.
Interference 135

The path difference between two in-phase waves may be zero or an integral multiple of a
wavelength, l and the path difference between two opposite-phase waves will be l / 2 or an
odd integral multiple of l / 2.
Optical path difference and the consequent phase difference may arise due to two
reasons. One reason is the difference in the optical paths (See Art.5.15) and the other is due to
reflections at optical interfaces.
(a) Phase difference due to optical path difference:
Let us consider two sources of light S1 and S2, as shown in Fig.6.4. Let us assume that the
sources are identical and produce waves of same wavelength and that their vibrations are
in the same phase at S1 and S2. Light from P
these sources travel in air along different
paths, S1P and S2P; and meet at a point P.
The path lengths S1P and S2P are different
and contain different number of waves. S1
The geometric path difference between the
waves at P is (S2P – S1P) and the optical Geometric path difference = S2P – S1P
path difference is m(S2P – S1P). Since the optical path difference =  (S2P – S1P)
paths contain different number of waves, the
optical path difference will be equal either to S2
a few integral number of waves or an integral Fig. 6.4
number of wavelengths plus a fraction of
one wavelength. This optical path difference leads to a phase difference between the waves
meeting at P. It means that though the waves started with the same phase, they arrive at P
with different phases because they travelled along different path lengths. Using equn.(6.4),
the phase difference between the waves at P may be expressed as
2π
d= µ( S2 P − S1P) (6.5)
λ
(b) Phase difference due to reflection at boundaries of optical interfaces:
Light waves may also undergo phase
Air  change
change due to reflection at some point
in their path. If the waves are reflected
at a rarer-to-denser medium boundary, Air
Glass

the reflected waves suffer a phase  change

change of p rad or 180° compared to Glass
the incident waves (See Fig. 6.5 a &
b). It is seen from equn.(6.3) that a
phase change of p rad is equal to a path No  change
change of l/2. Therefore, we must add (a) (b)
(or subtract) l/2 in the calculation of Fig. 6.5
true optical path difference whenever a
reflection occurs at a denser medium.

INTERFERENCE BY DIVISION OF WAVE FRONT

6.6 FRESNEL BIPRISM

Fresnel used a biprism to show interference phenomenon. The biprism consists of two prisms
of very small refracting angles joined base to base. In practice, a thin glass plate is taken and
one of its faces is ground and polished till a prism (Fig.6.6 a) is formed with an obtuse angle
136 A Textbook of Engineering Physics

of about 179° and two 30 S1

side angles of the order A
of 30′. S

When a light ray is 179°  S

incident on an ordinary
S
prism, the ray is bent
B
through an angle S2
called the angle of 30
deviation. As a result, (a) (b) (c)
the ray emerging out Fig. 6.6. Fresnel biprism and formation of virtual sources
of the prism appears to
have emanated from a source S′ located at a small distance above the real source, as shown
in Fig.6.6 (b). We say that the prism produced a virtual image of the source. A biprism, in
the same way, creates two virtual sources S1 and S2, as seen in Fig.6.6 (c). These two virtual
sources are images of the same source S produced by refraction and are hence coherent.
6.6.1 Experimental Arrangement
The biprism is mounted suitably on an optical bench. An optical bench consists of two
horizontal long rods, which are kept strictly parallel to each other and at the same level. The rods
carry uprights on which the optical components are positioned. A monochromatic light source
such as sodium vapour lamp illuminates a vertical slit S. Therefore, the slit S acts as a narrow
linear monochromatic light source. The biprism is placed in such a way that its refracting
edge is P  A
parallel to the S 2 
length of the
1
E
slit S. A single d 
 O
S
cylindrical F
wavefront S2

impinges on a b D
both prisms. D
The top (a) (b)
portion of Fig. 6.7. Fresnel biprism and fringe formation
wavefront
is refracted downward and appears to have emanated from the virtual image S1. The lower
segment, falling on the lower part of the biprism, is refracted upward and appears to have
emanated from the virtual source S2. The virtual sources S1 and S2 are coherent (see Fig.6.7),
and hence the light waves are in a position to interfere in the region beyond the biprism. If a
screen is held there, interference fringes are seen. In order to observe fringes, a micrometer
eyepiece is used.
Theory
Let S1 and S2 be the two virtual images of the source S. Let “d” be the distance between S1
and S2. The fringes are formed on a screen T kept at a distance D from the biprism. The point
O on the screen is equidistant from S1 and S2. Hence, the waves arrive from S1 and S2 arrive
at O simultaneously and the point O is always bright. The point O corresponds to the position
of central bright fringe. On both sides of O, alternate bright and dark fringes, as shown in
Fig.6.7 (b), are produced.
Interference 137

Let P be an arbitrary point on P

screen (Fig.6.8). Let q be the angle
that MP makes with the horizontal
line MO. Let S1N be a normal x
on to the line S2P. The distances S1 E
PS1 and PN are equal. The waves d
2
emitted at the slits, S1 and S2 are d M 
 Axis
O
initially in phase with each other.
The difference in the path lengths of N N
S2 F
these two waves is S2N. We assume
that the experiment is carried out in
air. Therefore, the optical paths are D
identical with geometrical paths. The Screen T
nature of the interference of the two
Fig. 6.8
waves at P depends simply on how
many waves are contained in the length of the path difference S2N. If S2N contains an integral
number of wavelengths, the two waves interfere constructively, producing a maximum in the
intensity of light on the screen at P. If it contains an odd number of half-wavelengths, the
waves interfere destructively and produce a minimum intensity at P.
6.6.2 Optical Path Difference between the Waves at P
Let the point P be at a distance x from O (Fig.6.8). Then
PE = x – d / 2 and PF = x + d / 2.
 2  2 2
d   2  d 
(S2P)2 – (S1P)2 =  D +  x +   −  D +  x −  
  2     2  
\ (S2P)2 – (S1P)2 = 2xd
2 xd
or S2P – S1P =
S2 P + S1P
We can approximate that S2P ≅ S1P ≅ D.
xd
\ Path difference = S2P – S1P = (6.6)
D
We now find out the conditions for observing bright and dark fringes on the screen.
6.6.3 Bright Fringes
Bright fringes occur wherever the waves from S1 and S2 interfere constructively. The first
place this occurs is at O, the axial point. There, the waves from S1 and S2 travel the same
optical path length to O and arrive in phase. The next bright fringe occurs when the wave
from S2 travels one complete wavelength further the wave from S1. In general constructive
interference occurs if S1P and S2P differ by a whole number of wavelengths.
The condition for finding a bright fringe at P is that
S2P – S1P = ml
Using the equation (6.6), it means that
xd
= ml (6.7)
D
138 A Textbook of Engineering Physics

where m is called the order of the fringe. The bright fringe at O, corresponding to m = 0, is
called the zero-order fringe. The first-order bright fringe from the axis corresponds to m = 1
and the second order bright fringe to m = 2 and so on.
6.6.4 Dark Fringes
The first dark fringe occurs when (S2P – S1P) is equal to l / 2. The waves are now in opposite
phase at P. The second dark fringe occurs when (S2P – S1P) equals 3l / 2. The mth dark fringe
occurs when
(S2P – S1P) = (2m + 1) l / 2
The condition for finding a dark fringe is
xd λ
= (2m + 1) (6.8)
D 2
The first-order dark fringe from the axis corresponds to m = 1 and the second order dark
fringe to m = 2 and so on.
6.6.5 Separation between Neighbouring Bright Fringes
The mth order bright fringe occurs when
mλD
xm =
d
th
and the (m + 1) order bright fringe occurs when
(m + 1)λ D
xm+1 =
d
The bright fringe separation, b is given by
λD
b = xm+1 – xm = (6.9)
d
The same result will be obtained for dark fringes. Thus, neighbouring bright and dark
fringes are separated by the same amount everywhere on the screen. The separation b is called
the fringe width.
The width of the dark or bright fringe is given by equ.(6.9).
λD
b=
d
where D(= a + b) is the distance of the sources from the eye-piece.
Example 6.1: A biprism is placed 5 cm from a slit illuminated by sodium light (l = 5890 Å).
The fringe width obtained on the screen is 0.9424 mm. The screen is at a distance of 75 cm
from the biprism. Find the distance between the two coherent sources.
λD
Solution: b=
d
5890 × 10−8 cm (5 + 75)cm
\ 9.424 × 10–2 cm =
d
−8
5890 × 10 × 80
or d= cm = 0.05 cm.
9.424 × 10−2

6.6.6 Determination of Wavelength of Light

The wavelength of the light can be determined using the equ.(6.9). For using the relation, the
values of b, D and d are to be measured. These measurements are done as follows.
Interference 139

Adjustments
A narrow adjustable slit S, the biprism, and a micrometer eyepiece are mounted on the
uprights and are adjusted to be at the same height and in a straight line. The slit is made
vertical and parallel to the refracting edge of the biprism by rotating it in its own plane. It is
illuminated with the light from the monochromatic source. The biprism is moved along the
optical bench till, on looking through it along the axis of the optical bench, two equally bright
vertical slit images are seen. Then the eyepiece is moved till the fringes appear in the focal
plane of the eyepiece.
(i) Determination of fringe width b: When the fringes are observed in the field view
of the eyepiece, the vertical cross-wire is made to coincide with the centre of one of
the bright fringes. The position of the eyepiece is read on the scale. The micrometer
screw of the eyepiece is moved slowly and the number of the bright fringes, N, that
pass across the cross-wire is counted. The position of the cross-wire is again read.
The fringe width is then given by
x − xm
b = m+ N
N
(ii) Determination of ‘d’: A convex lens of short focal length is placed between the
slit and the eyepiece without disturbing their positions. The lens is moved back and
forth near the biprism till a sharp pair of images of the slit is obtained in the field of
view of the eyepiece (Fig. 6.9 a). The distance between the images is measured. Let
it be denoted by d1.

S1 S1
S d1 S
S2 S2 d2

L1 L2
u   u
(a) (b)
Fig. 6.9

If u is the distance of the slit and u that of the eyepiece from the lens (Fig.6.9), then the
magnification is
υ d
= 1 (6.10)
u d
The lens is then moved to a position nearer to the eyepiece, where again a pair of images
of the slit is seen (Fig. 6.96). The distance between the two sharp images is again measured.
Let it be d2. Again magnification is given by
u d
= 2 (6.11)
υ d
Note that the magnification in one position is the reciprocal of the magnification in the
other position.
Multiplying the equations (6.10) and (6.11), we obtain
d1d 2
=1
d2
d = d1d 2 (6.12)
Using the values of b, d and D in the equation (6.9), the wavelength l can be
computed.
140 A Textbook of Engineering Physics

Example 6.2: Fresnel biprism is used to form interference fringes using sodium light of
l = 5890 Å. A convex lens interposed gives two images separated by distances 0.6 mm and
0.27 mm of the coherent sources corresponding to two positions. Calculate the fringe width.
The slit to screen distance is 80 cm.
Solution: Let d1 and d2 be the separation between the images in two positions. Then, the
distance between the coherent sources is
d = d1d 2 = 6 × 10−4 m × 2.7 × 10−4 m = 4.025 × 10–4 m

λD 5890 × 10−10 m × 0.8 m

\ b= = = 1.171 mm.
d 4.025 × 10−4 m

6.6.7 Interference Fringes with White Light

In the biprism experiment if the slit is illuminated by white light, the interference pattern
consists of a central white fringe flanked on both its sides by a few coloured fringes; and
general illumination beyond the fringes. The central white fringe is the zero-order fringe.
With monochromatic light all the bright fringes are of the same colour and it is not
possible to locate the zero-order fringe. Therefore, in order to locate the zero order fringe the
biprism is to be illuminated by white light.
6.6.8 Lateral Displacement of Fringes
The biprism experiment can be used to determine the thickness of a given thin sheet of
transparent material such as glass or mica. If a thin transparent sheet is introduced in the path
of one of the two interfering beams, the fringe system gets displaced towards the beam in
whose path the sheet is introduced. By measuring the amount of displacement, the thickness
of the sheet can be determined.
Suppose S1 and S2 are the virtual X

coherent monochromatic sources. t P

The point O is equidistant from S1
x
and S2, where we obtain the central S1
G
bright fringe. Therefore, the optical
path S1O = S2O. Let a transparent d
O
plate G of thickness t and refractive
index m be introduced in the path of S2
one of the beams (see Fig.6.10). The
optical path lengths S1O and S2O are D
now not equal and the central bright Y
fringe shifts to P from O. The light
Fig. 6.10
waves from S1 to P travel partly in air
and partly in the sheet G; the distance travelled in air is (S1P – t) and that in the sheet is t.
The optical path DS P = (S1P – t) + m t = S1P + (m – 1)t
1
The optical path DS P = S2P
2
The optical path difference at P is DS P – DS P = 0, since in the presence of the thin sheet,
1 1
the optical path lengths S1P and S2P are equal and central zero fringe is obtained at P.
\ ∆ S 1 P = ∆ S2 P
[S1P + (m – 1)t] = S2P
\ S2P – S1P = (m – 1)t
Interference 141

But according to the relation (6.7),

xd
S2P – S1P =
D
where x is the lateral shift of the central fringe due to the introduction of the thin sheet.
xd
\ (m – 1)t =
D
Hence, the thickness of the sheet is
xd
t= (6.13)
D(µ − 1)
Example 6.3: A thin sheet of glass (m = 1.5) of thickness 6 mm is introduced in one of the
paths of a biprism. It shifts the central fringe by 5 bands. Find the wavelength of light.
Solution: The displacement of fringes is given by (m – 1)t = ml. Therefore,
(µ − 1)t (1.5 − 1) × 6 × 10−6 m
l= = = 600 nm.
m 5

INTERFERENCE BY DIVISION OF WAVE AMPLITUDE

6.7 THIN FILM INTERFERENCE

An optical medium is called a thin film when its Reflected rays
thickness is about the order of 1 wavelength of light in
visible region. Thus, a film of thickness in the range 0.5 100%
Inc

%
5%

05
4%
mm to 10 mm may be considered as a thin film. A thin
ide

3.6

0.0
nt

film may be a thin sheet of transparent material such as

ray

glass, mica, an air film enclosed between two transparent 1 2 3

plates or a soap bubble. When light is incident on such a
film, a small part of it gets reflected from the top surface Thin
and a major part is transmitted into the film. Again, a film
small part of the transmitted component is reflected
back into the film by the bottom surface and the rest of
it emerges out of the film. A small portion of the light
thus gets reflected partially several times in succession 92%

within the film (see Fig.6.11).

In transparent thin films, the two bounding surfaces
strongly transmit light and only weakly reflect the Transmitted rays
incident light. Therefore, only the first reflection at the Fig. 6.11
top surface and the first reflection at the bottom surface will be of appreciable strength. For
example, if we consider a glass plate, having a refractive index 1.52, the reflectivity of the top
surface is given by
2
1.52 − 1 
r=   = 0.042 (6.14)
1.52 + 1 
It means that about 4% of the incident light is reflected by the top surface of the glass
plate, while 96% of it is transmitted into the plate. Out of the light reaching the bottom
surface, again 3.8% is reflected and 92% is transmitted out of the plate. Then, again out of the
3.8% of the light 0.15% is reflected at the inner boundary of the top surface and about 3.65%
142 A Textbook of Engineering Physics

is transmitted out into the air. After two reflections, the intensity will become insignificantly
small. At each reflection, the intensity and hence the amplitude of light wave is divided into a
reflected component and a refracted component. The reflected and refracted components travel
along different paths and can be brought to overlap to produce interference. Therefore, the
interference in thin films is called interference by division of amplitude. Newton and Robert
Hooke first observed the thin film interference. However, Thomas Young gave the correct
explanation of the phenomena. A thin film may be uniform or non-uniform in its structure.
However, as long as its thickness lies within the specified limits, interference of light occurs.
6.8 PLANE PARALLEL FILM
A transparent thin film of uniform thickness bounded by two parallel surfaces is known as a
plane parallel thin film.
When light is incident on a parallel thin film, a

s d
ve cte
wa efle
small portion of it gets reflected from the top surface

R
and a major portion is transmitted into the film. Again, 1

a small part of the transmitted component is reflected 2

back into the film by the bottom surface and the rest
of it is transmitted from the lower surface of the film.
Thin films transmit incident light strongly and reflect Thin
film
only weakly. After two reflections, the intensities of
reflected rays drop to a negligible strength. Therefore,
we consider the first two reflected rays only. These S1
two rays are derived from the same incident ray but
N
appear to come from two sources located below the
S2
film. The sources are virtual coherent sources (see
Fig. 6.12
Fig.6.12). The reflected waves 1 and 2 travel along
parallel paths and interfere at infinity. This is a case of two-beam interference.
The condition for maxima and minima can be deduced once we have calculated the
optical path difference between the two rays at the point of their meeting.
6.8.1 Interference Due to Reflected Light
Let us consider a transparent film of uniform A
thickness ‘t’ bounded by two parallel surfaces 1 C E
y
Ra
2

as shown in Fig.6.13. Let the refractive index

y
Ra

of the material be m. The film is surrounded

H
by air on both the sides. Let us consider plane  change
i i
waves from a monochromatic source falling G i Air
on the thin film at an angle of incidence ‘i’. B D
Part of a ray such as AB is reflected along BC,  t
r r
and part of it is transmitted into the film along no  change
BF. The transmitted ray BF makes an angle ‘r’
with the normal to the surface at the point B. F Air

The ray BF is in turn partly reflected back into BF = t / cos r

the film along FD while a major part refracts BG = t.tan r
into the surrounding medium along FK. Part of BH = 2t.tan r. sin i
D = 2t cos r – /2
the reflected ray FD is transmitted at the upper K
surface and travels along DE. Since the film Fig. 6.13
Interference 143

boundaries are parallel, the reflected rays BC and DE will be parallel to each other. The
waves travelling along the paths BC and BFDE are derived from a single incident wave AB.
Therefore they are coherent and can produce interference if they are made to overlap by a
condensing lens or the eye.
(i) Geometrical Path Difference: Let DH be normal to BC. From points H and D
onwards, the rays HC and DE travel equal path. The ray BH travels in air while the ray
BD travels in the film of refractive index m along the path BF and FD. The geometric path
difference between the two rays is
BF + FD – BH.
(ii) Optical Path Difference:
Optical path difference Da = m L
\ Da = m (BF + FD) – 1(BH) (6.15)
In the DBFD, ∠BFG = ∠GFD = ∠r
BF = FD
FG t
BF = =
cos r cos r
2t
\ BF + FD = (6.16)
cos r
Also, BG = GD
\ BD = 2BG
BG = FG tan r = t tan r
\ BD = 2t tan r
In the Dle BHD ∠HBD = (90 – i)
∠BHD = 90°
\ ∠BDH = i
\ BH = BD sin i = 2 t tan r sin i (6.17)
From Snell’s law,
sin i = msin r
2µ t sin 2 r
\ BH = 2t tan r (m sin r) = (6.18)
cos r
Using the equations (6.17) and (6.16) into equ.(6.15), we get
 2t   2µ t sin 2 r 
Da = µ  − 
 cos r   cos r 
2µ t 
= 1 − sin 2 r 
cos r  

2µ t
= cos 2 r
cos r
\ Da = 2m t cos r (6.19)
(iii) Correction on account of phase change at reflection: When a ray is reflected at
the boundary of a rarer to denser medium, a path-change of l/2 occurs for the ray BC (see
Fig.6.13). There is no path difference due to transmission at D. Including the change in path
difference due to reflection in eqn. (6.19), the true path difference is given by
Dt = 2m t cos r – l/2 (6.20)
144 A Textbook of Engineering Physics

6.8.2 Conditions for Maxima (Brightness) and Minima (Darkness)

Maxima occur when the optical path difference D = m l. If the difference in the optical
path between the two rays is equal to an integral number of full waves, then the rays meet
each other in phase. The crests of one wave falls on the crests of the others and the waves
interfere constructively. Thus, when
λ
2m t cos r – = ml (6.21)
2
the reflected rays undergo constructive interference to produce brightness or maxima at the
point of their meeting.
2m t cos r = ml + l/2
or 2m t cos r = (2m + 1)l/2 Condition for Brightness (6.22)
Minima occur when the optical path difference is D = (2m + 1) l/2. If the difference in
the optical path between the two rays is equal to an odd integral number of half-waves, then
the rays meet each other in opposite phase. The crests of one wave falls on the troughs of the
others and the waves interfere destructively. Thus, when
2m t cos r – l/2 = (2m + 1)l/2 (6.23)
the reflected rays undergo destructive interference to produce darkness. Equ.(6.23) may be
rewritten as
2m t cos r = (m + 1)l
The phase relationship of the interfering waves does not change if one full wave is added
to or subtracted from any of the interfering waves. Therefore (m + 1)l can be as well replaced
by ml for simplicity in expression. Thus,
2m t cos r = ml Condition for Darkness (6.24)
6.8.3 Some Important Points
(a) It is seen that the conditions of interference depend on three parameters, namely
mt, l and r. In the case of constant thickness (parallel) film, mt is constant. When
a parallel beam of (see no light) (see no light)
light is incident
Incoming
on such a film, Incoming light
r also remains light
"Reflected
constant. Then light"
the interference
conditions solely
depend on the
wavelength, l. Air
(b) When a parallel Glass

beam of t = m / 2 t = (2m + 1)  / 4
monochromatic
(a) (b)
light is incident
normal to the Fig. 6.14: Interference in film of constant thickness – (a)
film, the whole Interference is destructive and the film appears dark in
film will appear reflected light for thickness satisfying t = m l/2m condition. (b)
uniformly dark or It is constructive and the film appears bright for film thickness
uniformly bright. satisfying the condition t = (2m + 1)l/4m.
The film will appear bright in reflected light, when the film is of l/4m, 3l/4m,
5l/4m, ....... thick and it appears dark when its thickness is l/2m, l/m, 3l/2m, ..........
etc. (Fig.6.14). If the condition of constructive interference is satisfied, the film will
show intense colour corresponding to the colour of the incident light.
Interference 145

(c) A change in the angle of incidence of the rays leads to a change in the path
difference. Consequently, if the inclination of the film with respect to the light beam
is changed gradually, we find that it will appear dark and bright (or bright and dark)
in succession.
(d) If a parallel beam of white light falls on a parallel film, those wavelengths for which
the path difference is ml, will be absent from the reflected light. The other colours
will be reflected. Therefore, the film will appear uniformly coloured with one colour
being absent.
Example 6.4: A soap film of 5 × 10–7m thick is viewed at an angle of 35° to the normal. Find
the wavelengths of light in the visible spectrum, which will be absent from the reflected light.
Given the refractive index of the film = 1.33.
Solution: White light is incident on the film at an angle 30°. Let r be the angle of
sin i
refraction of light into the film. r can be calculated from Snell’s law m = .
sin r
sin 30°
\ sin r = or r = 25.55° and cos r = 0.90.
1.33
The absence of certain wavelengths in reflected light is due to their undergoing destructive
interference. The condition for destructive interference is
2m t cos r = ml
To find out the missing wavelengths, we have to use different m values into the above
equation and find out which of them lie in the visible region 7000 to 4000 Å.
For m = 1, we get λ1 = 2 ×1.33 × 5 × 10–7m × 0.90 = 12 ×10–7m = 12000 Å.
For m = 2, we get λ2 = (2 ×1.33 × 5 × 10–7m × 0.90) ÷ 2 = 6 ×10–7m = 6000 Å.
For m = 3, we get λ3 = (2 ×1.33 × 5 × 10–7m × 0.90) ÷ 3 = 4 ×10–7m = 4000 Å.
For m = 4, we get λ4 = (2 ×1.33 × 5 × 10–7m × 0.90) ÷ 4 = 3 ×10–7m = 3000 Å.
It is clear that the first wavelength lies in the infra red and the last one lies in the UV
region. The middle two wavelengths lie in the visible region. Hence, the absent wavelengths
in the reflected light are 6000 Å and 4000 Å.
6.8.4 Restriction on Thickness of the Film
We know that interference colours are observed only in thin films but not in thick plates such
as windowpanes or glass slabs. This is due to the fact that light waves can interfere only when
both the conditions of temporal and spatial coherence are satisfied. In Fig. 6.14, we have
assumed that a monochromatic wave of infinite length is incident on the film. In reality, the
incident light consists of wave trains of finite length and coherence extends over the length of
each wave train only. Interference can occur only when parts of the same group of wave trains
overlap. Superposition of different wave trains cannot produce interference because they will
be incoherent and do not maintain any constant phase relationship with each other.
Fig. 6.15 shows the real situation. Wave trains 1,2,3 of finite length are incident in
succession on a thin film. Portions of each wave train are reflected by the top and bottom
surfaces of the film. Each wave train is divided into two reflected wave trains (U1, L1, U2,
L2 and U3, L3). In Fig.6.15 (a) the film is thin and the difference in the optical path lengths
of U1 and L1 is small compared to the length of the wave train. Their superposition produces
interference, as U1 and L1 are parts of the same wave train 1 and hence are coherent. In
Fig. 6.15 (b) the film is thicker and the optical path difference between U1 and L1 is large than
the coherence length. Consequently, superposition takes place between parts of different wave
trains, U2 and L1 and U3 and L2. Therefore interference does not take place.
146 A Textbook of Engineering Physics

U1
U1
3
U2
3
U2 L1 2 U3
2 L1
1
U3
L2
1
t L2
L3 t L3

(a) (b)
Fig. 6.15: Role of the thickness of the film—(a) when the film thickness is smaller than coherence
length, superposition of reflected parts of the same wave train occurs leading to inference. (b) In
thick films, different wave trains which are not coherent superpose and interference does
not arise.

It implies that interference occurs only when the optical path difference, D, between the
superposing waves is less than the coherence length.
i.e., D << lcoh (6.25)
\ (2mt cos r – l/2) << lcoh (6.26)
2
λ
But lcoh =
∆λ
\ (2mt cos r – l/2) < l2/Dl
Rearranging the terms, we obtain
 λ 1
λ +
 ∆λ 2 
t< (6.27)
2µ cos r
l/Dl >> 1/2 and for normal incidence cos r = 1.
λ2
\ t< (6.28)
2µ ∆ λ
The above equation indicates that interference in thin film will be observed if the
thickness of the film is less than the coherence length of the incident light waves. Normally,
the coherence length of the light from ordinary sources is of the order of a fraction of a
millimeter. Therefore, interference is seen with the films of thickness of the order of a few
hundred microns only. It is because of this reason that thick films do not exhibit interference.
6.9 VARIABLE THICKNESS (WEDGE-SHAPED) FILM
A wedge is a thin film of varying thickness having a zero thickness at one end and progres-
sively increasing to a particular thickness at the other end. A thin wedge of air film can be
formed by two glass slides resting on each other at one edge and separated by a thin spacer at
the opposite edge.
The arrangement for observing the interference pattern in a wedge shaped air-film is
shown in Fig.6.16. If a parallel beam of monochromatic light illuminates the wedge from
above, the rays reflected from its two bounding surfaces will not be parallel. They appear to
diverge from a point near the film. The path difference between the rays reflected from the
Interference 147

upper and lower surfaces of the air film varies along its length due to variation in thickness.
Therefore, alternate bright and dark fringes are observed on its top surface (Fig.6.17). The
fringes are localized at the top surface of the film.

Microscope

Glass plate S
45° Sodium
Lamp

Microscope
slide
Thin sheet


Fig. 6.16          Fig. 6.17

When the light is incident on the wedge from above, it gets partly reflected from the
glass-to-air boundary at the top of the air film. Part of the light is transmitted through the air
film and gets reflected partly at the air-to-glass boundary, as shown in Fig.6.18. The two rays
BC and FE, thus reflected from the top and bottom of the air film, are coherent as they are
derived from the same ray AB through division of amplitude. The rays are close enough if the
thickness of the film is of the order of a wavelength of light. For small film thickness the rays
interfere producing darkness or brightness depending on the phase difference. The thickness
of the glass plates is large compared with the wavelength of the incident light. Hence, the
observed interference effects are entirely due to the wedge-shaped air film.

Glass-to-air
boundary

λ/2
1

Air-wedge θ
Ray

Ray 1

θ β
θ 1 β 2
O
No π change
Air-to glass Dark Dark Dark
boundary
O
π change

Fig. 6.18           Fig. 6.19

The optical difference between the two rays BC and FE is given by

Δ = 2m t cos r – l/2
where l/2 takes account the gain of half–wave due to the abrupt jump of p radians in the
phase of the wave reflected from the bottom boundary of air – to – glass.
148 A Textbook of Engineering Physics

Maxima occur when the optical path difference D = m l. If the difference in the optical
path between the two rays is equal to an integral number of full waves, then the rays meet
each other in phase. The crests of one wave falls on the crests of the others and the waves
interfere constructively. This needs that
2m t cos r = (2m + 1)l/2
Minima occur when the optical path difference is D = (2m + 1)l/2. If the difference in
the optical path between the two rays is equal to an odd integral number of half-waves, then
the rays meet each other in opposite phase. The crests of one wave fall on the troughs of the
others and the waves interfere destructively. It needs that
2m t cos r = ml.
Referring to Fig.6.19, let us say a dark fringe occurs at A where the relation
2m t cos r = ml
is satisfied. If normal incidence is assumed, cos r = 1 and if the thickness of air film at A is
denoted by t1, then at A
2m t1 = ml (6.29)
The next dark fringe will occur, say, at C where the thickness CL = t2. Then at C
2m t2 = (m + 1)l (6.30)
Subtracting equ. (6.29) from equ. (6.30), we get
2m(t2 – t1) = l (6.31)
But (t2 – t1) = BC
\ 2m(BC) = l
λ
or BC = (6.32)
2µ
From the DleABC, ∠CAB = q and BC = AB tan q
λ
\ (AB) tan q = (6.33)
2µ
AB is the distance between successive dark fringes and it also equals the separation of the
successive bright fringes. It is, therefore, called the fringe width, b. That is AB = b. We may
write equ. (6.33) as
λ
\ b= (6.34)
2µ tan θ
For small values of q, tan q ≈ q.
λ
\ b= (6.35)
2µ θ
According to the relation (6.35), an increase in the angle q makes the fringes move closer.
At an angle q ≈ 1°, the interference pattern vanishes. On the other hand, if q is gradually
decreased, the fringe separation increases and ultimately the fringes disappear since the faces
of the film become parallel at q = 0°.
Example 6.5: Fringes of equal thickness are observed in a thin glass wedge of refractive
index 1.52. The fringe spacing is 0.1 mm, wavelength of light being 5893 Å. Calculate the
wedge angle.
Solution: The fringe width b = l/2mq.
λ 5893 × 10−10 m
\ q = = = 1.938 rad = 0.11°.
2µβ 2 × 1.52 × 10−4 m
Interference 149

6.9.1 Salient Features of the Interference Pattern

(i) Fringe at the apex is dark.
(ii) Fringes are straight and parallel.
(iii) Fringes are equidistant.
(iv) Fringes of equal thickness
(v) Fringes are localized.
(i) Fringe at the apex is dark: At the apex, the two glass slides are in contact with each
other. Therefore, the thickness of the air film at the contact edge is negligible (t ≅ 0). The
optical path difference there becomes
D = 2mt – l/2 = 0 – l/2 = – l/2 (6.36)
It implies that a path difference of l/2 or a phase difference of p occurs between the
reflected waves at the edge. The two waves interfere destructively. Therefore, the fringe at
the apex is always dark (Fig. 6.20).
Dark Bright
y (2)

band band
d Ra
ecte
Refl

Reflected Ray (1)

 
 change

Fig. 6.20: At the contact edge, the reflected rays (1) and (2) are
180° out of phase and produce a dark band.

(ii) Straight and parallel fringes: Each fringe in the pattern is produced by the
interference of rays reflected from sections of the wedge having the same thickness. The locus
of points having the same thickness lies along lines parallel to the contact edge. Therefore, the
fringes are straight. Since the fringes are equidistant, they will be parallel.
(iii) Equidistant fringes: The fringe width b is given by
b ≈ l/2q (6.37)
where l is the wavelength of the incident light and q is the angle of the wedge. As the
quantities l and q are constants, b is constant for a given wedge angle. Therefore, the fringes
are equidistant.
(iv) Fringes of equal thickness: As each bright or dark fringe is a locus of constant film
thickness, the fringes are called fringes of equal thickness.
(v) Localized fringes: The fringes are very close to the top surface of the air wedge and
can be seen with a microscope.
6.9.2 Determination of the Wedge Angle
The wedge angle q can be experimentally determined with the help of a travelling microscope.
Using the microscope the positions of dark fringes at two distant points Q and R are noted
(Fig.6.21). Let the distance OQ be x1 and OR be x2. Let the thickness of the wedge be t1 at Q
and t2 at R.
150 A Textbook of Engineering Physics

t
Q t1
 R t2
O
Dark Dark
x1

x2

Fig. 6.21

The dark fringe at Q is given by

2m t1 = ml (6.38)
But as q is very small, we can write
t1 = x1 tan q ≅ x1q
\ 2m x1 q = ml (6.39)
We can write similarly for the dark fringe at R as
2m x2 q = (m + N)l (6.40)
where N is the number of dark fringes lying between the positions Q and R. Subtracting equ.
(6.39) from equ.(6.40), we get
2m(x2 – x1)q = Nl
Nλ
\ q= (6.41)
2µ( x2 − x1 )
In case of air m = 1 and the above relation reduces to
Nλ
q= (6.42)
2( x2 − x1 )

6.9.3 Determination of the Thickness of the Spacer

The thickness of the spacer used to form the wedge shaped air film between the glass slides
can be determined from the above measurements. If ‘t’ is the thickness of the spacer (foil or
wire) used, we can write
t = l tan q ≅ l q (6.43)
where l is the length of the air wedge.
l Nλ
\ t= (6.44)
2( x2 − x1 )

6.9.4 Number of Dark Fringes in an Air Wedge

λ
The fringe width in air wedge is given by b =
2θ
t
The wedge angle is given by q =
l
λl
\ b= (6.45)
2t
The number of dark fringes observed, N, is related to l, as
l = Nb (6.46)
Interference 151

λNβ
\ b=
2t
2t
\ N= (6.47)
λ
6.10 COLOURS IN THIN FILMS
The colours exhibited in reflection by thin films of oil, mica, soap bubbles and coatings of
oxides on heated metals etc are due to interference of light from an extended source such
as sky. Thomas Young explained the origin of colours in thin films. It may be understood as
follows. The films are usually observed by reflected light.The eye looking at the thin film
receives light waves reflected from the top and bottom surfaces of the film. The reflected
rays are very close to each other and are in a position to interfere. The optical path difference
between the interfering rays is Δ = 2mt cos r – l/2. It is seen that the path difference depends
upon the thickness t of the film, the wavelength l and the angle r, which is related to the
angle of incidence of light on the film. White light consists of a range of wavelengths
and for specific values of t and r, waves of only certain wavelengths (colours) construc-
tively interfere. Therefore, only those colours are present in the reflected light. The other
wavelengths interfere destructively and hence are absent from the reflected light. Hence, the
film at a particular point appears coloured. As the thickness and the angle of incidence vary
from point to point, different colours are intensified at different places. The colours seen are
not isolated colours, as at each place there is a mixture of colours. The composition of colours
is different at different places and contours of impressive hues are observed over the entire
surface of the film.
6.11 NEWTON’S RINGS
Newton’s rings are another example of fringes of
equal thickness. Newton’s rings are formed when a
plano-convex lens L of a large radius of curvature placed
on a sheet of plane glass AB. The combination forms a
thin circular air film of variable thickness in all directions
around the point of contact of the lens and the glass
plate. The locus of all points corresponding to specific
thickness of air film falls on a circle whose centre is at
O. Consequently, interference fringes are observed in Fig. 6.22
the form of a series of concentric rings with their centre at O (Fig. 6.22). Newton originally
observed these concentric circular fringes and hence they are called Newton’s rings.
The experimental arrangement for observing Newton’s rings is shown in Fig. 6.23.
Monochromatic light from an extended source S is rendered parallel by a lens L′. It is
incident on a glass plate inclined at 45° to the horizontal, and is reflected normally down
onto a plano-convex lens placed on a flat glass plate. Part of the light incident on the system
is reflected from the glass-to-air boundary, say from point D (Fig.6.24).The remainder of the
light is transmitted through the air film. It is again reflected from the air-to-glass boundary,
say from point J. The two rays reflected from the top and bottom of the air film are derived
through division of amplitude from the same incident ray CD and are therefore coherent.
The rays 1 and 2 are close to each other and interfere to produce darkness or brightness. The
condition of brightness or darkness depends on the path difference between the two reflected
light rays, which in turn depends on the thickness of the air film at the point of incidence.
152 A Textbook of Engineering Physics

Microscope

C
L′ 2 1

Glass plate s
45°

Sodium
source
Air film Planoconvex P
lens D
L O
O
Place glass
plate A J B
A B AIR FILM

Fig. 6.23           Fig. 6.24

6.11.1 Condition for Bright and Dark Rings

The optical path difference between the rays is given by D = 2mt cos r – l/2. Since m = 1 for
air and cos r = 1 for normal incidence of light,
D = 2t – l/2 (6.48)
Intensity maxima occur when the optical path difference D = ml. If the difference in
the optical path between the two rays is equal to an integral number of full waves, then the
rays meet each other in phase. The crests of one wave falls on the crests of the others and the
waves interfere constructively. Thus, if 2t – l/2 = ml
2t = (2m + 1)l/2 (6.49)
bright fringe is obtained.
Intensity minima occur when the optical path difference is D = (2m + 1)l/2. If the
difference in the optical path between the two rays is equal to an odd integral number of
half-waves, then the rays meet each other in opposite phase. The crests of one wave fall on
the troughs of the other and the waves interfere destructively.
Hence, if 2t – l/2 = (2m + 1)l/2
2t = ml (6.50)
and dark fringe is produced.
6.11.2 Circular Fringes
In Newton’s ring arrangement, a thin air film is enclosed between a plano-convex lens and a
glass plate. The thickness of the air film at the point of contact is zero and gradually increases
as we move outward. The locus of points where the air film has the same thickness then fall
on a circle whose centre is the point of contact. Thus, the thickness of air film is constant at
points on any circle having the point of lens–glass plate contact as the centre. The fringes are
therefore circular.
6.11.3 Radii of Dark Fringes
Let R be the radius of curvature of the lens (Fig. 6.25). Let a dark fringe be located at Q.
Let the thickness of the air film at Q be PQ = t. Let the radius of the circular fringe at Q be
OQ = rm. By the Pythagorus theorem,
PM2 = PN2 + MN2
Interference 153

\ R2 = rm2 + (R – t)2 M

\ rm2 = 2Rt – t2 (6.51)

As R >> t, 2Rt >> t2. R
\ rm2 ≅ 2Rt (6.52) R

The condition for darkness at Q is that

N
2t = ml P
t
\ rm2 ≅ mlR
Q rm O
rm = mλR (6.53) Fig. 6.25
The radii of dark fringes can be found by inserting values 1,2,3, ....... for m.
Thus,
r1 = 1λ R or r1 ∝ 1

r2 = 2λR or r2 ∝ 2

r3 = 3λR or r3 ∝ 3 and so on
It means that the radii of the dark rings are proportional to square root of the natural
numbers.
The above relation also implies that rm ∝ λ
Thus, the radius of the mth dark ring is proportional to square root of wavelength.
Ring Diameter:
Diameter of mth dark ring Dm = 2rm
Dm = 2 2 R t

or Dm = 2 mλ R (6.54)

Example 6.6: In a Newton’s rings experiment, the diameter of 10th dark ring due to
wavelength 6000 Å in air is 0.5 cm. Find the radius of curvature of the lens.
( D / 2)2 (0.5 × 10−2 / 2)2 m2
Solution: Radius of curvature, R = = = 104 cm.
mλ 10 × 6000 × 10−10 m

6.11.4 Spacing between Fringes is not Even

It is seen that the diameter of dark rings is given by
Dm = 2 m λ R
where m = 1,2,3, ....
The diameters of dark rings are proportional to the square root of the natural numbers.
There fore, the diameter of the ring does not increase in the same proportion as the order of
the ring, for example, if m increases as 1, 2, 3, 4, .... the diameters are
D1 = 2 λ R

D2 = 2(1.4) λR

D3 = 2(1.7) λR

D4 = 2(2) λ R and so on
154 A Textbook of Engineering Physics

Therefore, the rings get closer and closer, as m increases. This is why the rings are not
evenly spaced.
6.11.5 Fringes of Equal Thickness
Newton’s rings are formed as result of interference between light waves reflected from the top
and bottom surfaces of a thin air film enclosed between a plano-convex lens and a plane glass
plate. The occurrence of alternate bright and dark rings depends on the optical path difference
arising between the reflected rays. If the light falls normally on the air film the optical path
difference between the waves reflected from the two surfaces of the film is
D = 2t – l/2
It is seen that the path difference between the reflected rays arises due to the variation
in the thickness ‘t’ of the air film. Reflected light will be of minimum intensity for those
thickness for which the path difference is ml and maximum intensity for those thickness
for which the path difference is (2m + 1)l/2. Thus, each maxima and minima is a locus of
constant film thickness. Therefore, the fringes are known as fringes of equal thickness.
6.11.6 Dark Central Spot
The central spot is dark as seen by reflection. Newton’s rings are produced due to superpo-
sition of light rays reflected from the top and bottom surfaces of a thin air film enclosed
between a plano-convex lens and a plane glass plate. The occurrence of brightness or darkness
depends on the optical path difference arising between the reflected rays. The optical path
difference is given by D = 2t – l/2.
At the point of contact ‘O’ of the lens and glass plate
(Fig.6.26), the thickness of air film is negligibly small Dark
compared to a wavelength of light. Reflected Reflected
\ t≅0 ray (1) ray (2)
\ D ≅ l/2
The wave reflected from the lower surface of the air
film suffers a phase change of p while the wave reflected
from the upper surface of the film does not suffer such  change

change. o
Thus, the superposing waves are out of step l/2 which
Fig. 6.26
is equivalent to a phase difference of 180° (or p rad). Thus
the two interfering waves at the centre are opposite in phase and produce a dark spot.
6.11.7 Determination of Wavelength of Light
A plano-convex lens of large radius of curvature (about 100 cm) and a flat glass plate are
cleaned. The lens is kept with its convex face on the glass plate and they are held in position
with the help of a metal ring arrangement. The system is held under a low power travelling
microscope kept before a sodium vapour lamp. It is arranged that the yellow light coming
from the sodium lamp falls on a glass plate held at 45° light beam. The light is turned through
90° and is incident normally on the lens-plate system. The microscope is adjusted till the
circular rings came into focus. The centre of the cross-wire is made to come into focus on the
centre of the dark spot, which is at the centre of the circular ring system. Now, turning the
screw the microscope is moved on the carriage slowly towards one side, say right side Fig.
6.27 (a). As the cross-wires move in the field of view, dark rings are counted. The movement
is stopped when the 22nd dark ring is reached. Then the microscope is moved in the opposite
Interference 155

side and stopped at the 20th or 19th dark ring. The vertical cross-wire is made tangential to the
19th ring and the reading is noted with the help of the scale graduated on the carriage. Thus,
starting from the 19th ring, the tangential positions of the 18th, 17th, 16th, …., 5th dark rings
are noted down. Now, the microscope is moved quickly to the left side of the ring system and
it is stopped at the 5th dark ring. The cross-wire is again made tangential to the 5th dark ring
and its position is noted. The difference between the readings on right and left sides of the 5th
dark ring gives its diameter value. The procedure is repeated till 19th ring is reached and its
reading is noted. From the value of the diameters the squares of the diameters are calculated.
A graph is plotted between Dm2 and between the ring number ‘m’. A straight line would be
obtained, as shown in Fig.6.27 (b).

Cross wire

4 R

m
2
D
22nd 5th 5th 22nd
dark ring dark ring
Direction of motion
Stop measurement Start measurement
LHS Jump from RHS RHS
5th to LHS 5th ring No. of ring m

(a) (b)

Fig. 6.27

We have
Dm2 = 4mlR (6.55)
For the (m + p)th ring,
D2m + p = 4(m + p)lR (6.56)
D m+p – D2m = 4plR
2

Dm2 + p − Dm2
l= (6.57)
4 pR
The slope of the straight line (Fig.6.27 b) gives the value of 4lR. Thus,
Slope
l= (6.58)
4R
The radius of curvature R of the lens may be determined using a spherometer and l is
computed with the help of the above equation.
Example 6.7: In a Newton’s rings experiment the diameter of the 15th ring was found to be
0.59 cm and that of the 5th ring was 0.336 cm. If the radius of the plano-convex lens is 100
cm, calculate the wavelength of light used.
Dm2 + p − Dm2 D 2 − D52 (5.9 − 3.36)2 × 10−6 m 2
Solution: l = = 15 = = 5880 Å
4 pR 4 × 10 × R 4 × 10 × lm
Example 6.8: In a Newton’s rings experiment the diameter of the 4th and 12th dark rings are
0.400cm and 0.700 cm respectively. Determine the diameter of 20th dark ring.
156 A Textbook of Engineering Physics

Dm2 + p − Dm2 2
D12 − D42
Solution: l= =
4 pR 4×8× R
Considering 20th and 4th dark rings,
2
D20 − D42
l= .
4 × 16 × R
Dividing these two equations, we get

2
D20 − D42 = (2
2 D12 − D42 )
2 2
or D20 = 2 D12 − D42 = 2(0.700)2 – (0.400)2cm2 = (0.98 – 0.16)cm2 = 0.82 cm2.
\ D20 = 0.906 cm.
6.11.8 Refractive Index of a Liquid
The liquid whose refractive index is to be determined is filled in the gap between the lens and
plane glass plate. Now the liquid film substitutes the air film. The condition for interference
may then be written as
2mt cos r = ml  Darkness
where m is the refractive index of the liquid. For normal incidence the equation becomes
2mt = ml
th
The diameter of m dark ring is given by
4m λ R
[ Dm2 ]L = (6.59)
µ
Similarly, the diameter of the (m + p)th ring is given by
4(m + p)λR
[ Dm2 + p ]L = (6.60)
µ
Subtracting equ.(6.59) from equ.(6.60), we get

 Dm2 + p  −  Dm2  = 4 p λ R (6.61)

 L  L µ
But we know that
( Dm2 + p ) air − ( Dm2 ) air = 4plR (6.62)

(D ) − (D )
2
m+ p
air
2
m
air
\ m= (6.63)
(D ) − (D )
2
m+ p
liq
2
m
liq

Example 6.9: In a Newton’s Rings experiment, the diameter of the 15th ring was found to
be 0.59 cm and that of the 5th ring was 0.336 cm. If the radius of the plano-convex lens is
100 cm, calculate the wavelength of light used. What happens to ring diameter if air film is
replaced with liquid of refractive index 1.5?
Dm2 + p − Dm2 2
D15 − D52 (0.592 − 0.3362 ) cm 2
Solution: l = = = = 5902 Å.
4 pR 4 × 10 × R 4 × 10 × 100 cm
The diameters of the Newton’s rings will be reduced when the air film is replaced with a
liquid film.
Interference 157

2
2 ( D15 ) air 0.592 cm 2
For example, ( D15 )liq = = = 0.232 cm2
µ 1.5

\ D15 = 0.232 cm 2 = 0.4817 cm

Example 6.10: In a Newton’s rings experiment the diameter of 10th ring changes from 1.40 to
1.27 cm when a drop of liquid is introduced between the lens and the glass plate. Calculate
the refractive index of the liquid.
(D )
2
m
air
=
(1.40 cm )2
Solution: m= = 1.215.
(D )
2
m
liq.
(1.27 cm) 2

6.11.9 Newton’s Rings by Transmitted Light

Newton’s rings are observed also when we view the
light transmitted through the lens-plate combination.
The light rays passing through the system gets partly
reflected at the top and bottom surfaces of the air film
enclosed between the lens and glass plate (Fig.6.28).
The optical path difference between two rays reflected
from the top of the air film and the bottom of the air
film is given by
D = 2 mt cos r
Since m = 1 for air and cos r = 1 for normal Transmitted light
incidence of light, Fig. 6.28
D = 2t
Intensity maxima occur when the optical path difference D = m l. Thus, if
2t = ml (6.64)
bright fringe is obtained.
Intensity minima occur when the optical path difference is D = (2m + 1)l/2. Hence, if
2t = (2m + 1)l/2 (6.65)
dark fringe is produced.
rm2
Taking the value of t = , the radii of the bright and dark rings can be calculated. We
2R
find that for bright rings
rm = mλ R (6.66)
(2m + 1)λ R
and for dark rings rm = (6.67)
2
Thus, the ring system due to transmitted light is just opposite to that we observe in the
reflected light. Wherever we get bright rings in the reflected light, we observe dark rings in
the transmitted light and vice versa. At the center, we find bright spot in the transmitted light.
Thus, the ring system in transmitted light is complementary to that seen in reflected light.
However, the rings in transmitted light are much poorer in contrast.
158 A Textbook of Engineering Physics

6.12 APPLICATIONS OF INTERFERENCE

The applications of interference phenomenon are wide and varied. Interference is used for
making precision measurements. For example, the wavelength of light can be measured with
accuracy up to eight significant digits. Interference is used for measuring small displacements.
The refractive indices of liquids and gases are measured using interference. We study here
two important applications, which utilize the phenomenon of thin film interference.
6.12.1 Testing of Flatness of Surfaces

AC F
optical E
flat B M
D -change
work piece
contact
edge (a) Flat (b) Concave (c) Convex

Fig. 6.29
In modern technology thin film interference is widely used. One of the applications is
testing of flatness of surfaces. Machine components retain surface irregularities left after
machining. The extent of suitability of the component for a particular application depends
on the irregularities which act as sources of stress leading to fatigue cracks. The surfaces of
components which are going to be subjected to high stress and load reversals are therefore
required to have a smooth surface finish. The smoothness of a surface can be quickly inspected
visually by keeping an optical flat on the component at an angle and illuminating it with a
monochromatic light (Fig.6.29). The air wedge formed between the component and optical
flat produces straight and equidistant fringes if the component surface is smooth. If the fringes
are curved towards the contact edge, the surface is concave and if the fringes curve away, it is
convex (Fig.6.29).
Testing of a lens surface: One of the
important uses of Newton’s rings is in the
testing of the optical components manufac-
tured for use in telescopes and other
instruments. The grinding of a lens surface is
tested by keeping it on a master. A master is
an optical flat which is a cylindrical disc made
of fused quartz. The two faces of the optical Fig. 6.30: Lens inspection using Newton’s
flat are perfectly parallel to each other. The rings (a) Circular ring pattern indicates the
departure from the flatness of each face is less high quality of grinding. (b) Distorted pattern
indicates irregularities.
than a light wavelength. If a lens is ground
perfectly, a circular fringe pattern is observed. Otherwise variations are observed which give
an indication of how the lens must be ground and polished to remove the imperfections. High
quality lenses are ground with a precision of less than a light wavelength.
6.12.2 Thickness of a Thin Film Coating
Dielectric and metallic thin films are often coated on optical components, solar cells etc.
One of the methods of determination of thickness of such thin films is based on multiple
beam interference. A partially coated substrate is used for the determination. The surfaces
Interference 159

of the substrate and the thin film on it are coated with a transparent metallic film of uniform
thickness. A glass plate is also coated on one of its surfaces with the transparent metallic film.
When the substrate and the glass plate are placed in contact and examined under monochro-
matic light, the reflected light shows a fringe system, as shown in Fig.6.31. A shift occurs in
the fringes as we pass from the region occupied by thin film to the region where thin film is
absent. The amount of displacement of one set of the fringes with respect to the second set of
fringes is given by
s = 2t or t = s/2
where t is the thickness of the thin film. By measuring ‘s’, t can be calculated.

(a) (b)

Fig. 6.31: Determination of the thickness of a thin coating.

6.12.3 Anti-Reflecting Coatings

Optical instruments such as telescopes and cameras use multicomponent glass lenses. When
light is incident on the lens, part of the incident light is reflected away and that much amount
of light is lost and wasted. When more surfaces are there, the number of reflections will be
large and the quality of the image produced by a device will be poor. In case of solar cells,
which operate on sunlight (daylight), the electrical energy produced will be less because of
the loss of part of light energy due to reflection, at the cell surface. It is found that coating the
surface with a thin transparent film of suitable refractive index can reduce such loss of energy
due to reflections at surface. Such coatings are called antireflection coatings. Thus,
“Antireflection (AR) coatings are thin transparent coatings of optical thickness of one–
quarter wavelength given on a surface in order to suppress reflections from the surface”.
Alexander Smakula discovered in 1935 that the reflections from a surface can be reduced
by coating the surface with a thin transparent dielectric film.
A thin film can act as an AR coating if it meets the following two conditions:
(i) Phase condition: The waves reflected from the top and bottom surfaces of the thin
film are in opposite phase such that their
overlapping leads to destructive interference, 1
and π change 2
(ii) Amplitude condition: The waves have
equal amplitudes. Air
The above conditions enable us determine B D
respectively (a) the required thickness of the film and
C Film
(b) the refractive index of the material to be used for
forming the film. t
π change
(i) Phase condition and minimum thickness
of the film: Let the thickness of the film be t and Glass
the refractive of the film-material be mf.The phase Fig. 6.32
160 A Textbook of Engineering Physics

condition requires that the waves (ray 1 and ray 2) reflected from the top and bottom surfaces
of thin film be 180° out of phase. It requires that the optical path difference between the two
rays must equal one half-wave or an odd number of half-waves. Referring to Fig.6.32, the
optical path difference between ray 1 and ray 2 is
D = 2mf t cos r – l/2 – l/2
the first l/2 corresponds to the p change at the top surface of the film (air-to-film boundary)
and the second l/2 to the p change that occurs at the film-to glass boundary because mf < mg.
If we assume normal incidence of light, cos r = 1 and the above equation reduces to
D = 2mf t – l = 2mf t
We wrote the above equality remembering that an addition of a full wave or subtraction
of a full wave from a train of waves does not affect the original phase relation. The ray 1
and ray 2 interfere destructively if the optical path difference satisfies the condition that
D = (2m + 1)l/2.
Thus, it requires that 2mf t = (2m + 1)l/2
For the film to be transparent, its thickness should be a minimum, which happens when
m = 0.
2mf tmin = l/2
λ
\ tmin = (mf < mg) (6.68)
4µ f
It means that the optical thickness of the AR coating should be of one-quarter wavelength.
Such quarter-wavelength coatings suppress the reflections and cause the light to pass into the
transmitted component.
(ii) Amplitude condition: The amplitude condition requires that the amplitudes of
reflected rays, ray 1 and ray 2 are equal. That is,
E1 = E2 (6.69)
It requires that 2 2
 µ f − µa   µg − µ f 
  =   (6.70)
 µ f + µ a   µ g + µ f 
where ma, mf , and mg are the refractive indices of air, thin film and glass substrate respectively.
As ma = 1, the above expression may be rewritten as
2 2
 µ f −1  µg − µ f 
  =  
 µ f + 1   µ g + µ f 
Expanding the above equation, we get
µ 2f − 2µ f + 1 µ 2g − 2µ g µ f + µ 2f
=
µ 2f + 2µ f + 1 µ 2g + 2µ g µ f + µ 2f

4µ3f µ g + 4µ f µ g = 4µ3f + 4µ f µ2g

Dividing by 4mf and rearranging the terms
µ 2f − µ g µ f + µ 2g − µ g = 0

(
µ 2f = µ g 1 + µ 2f − µ g )
Interference 161

\ µ 2f ≅ mg (as mf ≈ mg)

\ mf = µg (6.71)

It implies that the refractive index of thin film should be less than that of the substrate and
possibly nearer to its square root.
In case of glass, if we take mg = 1.5, mf = µ g = 1.22.
The materials which have refractive index nearer to this value are magnesium fluoride,
MgF2 (m = 1.38) and cryolite, 3NaF.AlF3 (m = 1.36). Apart from the refractive index, the
material should possess some more additional properties. The film should adhere well, should
be durable, scratch proof and insoluble in ordinary solvents. MgF2 and cryolite satisfy these
requirements. However, among the two, magnesium fluoride is cheaper and is hence widely
used as AR coating.
It may be noted that the condition (6.68) is satisfied only at one particular wavelength. The
wavelength normally chosen is 5500 Å for which the eye is most sensitive. This wavelength
is located in the yellow-green portion of the spectrum. Consequently, the reflection of red and
violet light will be larger when white light is incident on the component such as a camera
lens. Hence, the component shows purple hue in reflected light.
Example 6.11: A glass microscope lens (μ = 1.5) is coated with magnesium fluoride
(μf = 1.38) film to increase the transmission of normally incident light λ = 5800 Å. What
minimum film thickness should be deposited on the lens?
Solution: λ 5800 × 10−10 m
tmin = = = 1051 Å
4µ f 4 × 1.38
Example 6.12: Can a thin film of water (mf = 1.33) formed on a glass window pane (mf =
1.52) act as a non-reflecting film? If so, how thick should be the water film?
Solution: A film of refractive index mf can act as a non-reflecting film on a substrate
having refractive index m, if mf = µ .

Here, µ = 1.52 = 1.233.

As the refractive index of water is 1.33, it is nearer to water film µ can act as a
non-reflecting film on glass.
The minimum thickness of the film is given by
λ
tmin = .
4µ f
As human eye is more sensitive to green, it may be assumed that l = 5500 Å.
5500 × 10−10 m
\ tmin = = 1034 Å.
4 × 1.33
Multilayer AR coatings:
A single layer AR coating is effective only at one particular wavelength. A much wider
coverage across the spectrum is possible with multiple coatings, called multilayers. In
practice three layer coatings are widely used and are highly effective over most of the visible
spectrum. The central layer is half-wave (l / 2) thick and is of high refractive index materials
162 A Textbook of Engineering Physics

such as zirconium dioxide (ZrO2, m = 2.1). The outside layer is of magnesium fluoride having
l / 4 thickness and the layer adjacent to the substrate is again a l /4 thick coating of cesium
fluoride (CeF3, m =1.63) or aluminium oxide (Al2O3, m = 1.76). Some of the antireflection
coatings use up to 100 layers of alternating high and low refractive index materials.

1
g g g
a
2 3
a 1 2 a
d

Fig. 6.33

6.12.4 Interference Filters

An interference filter is an optical system that will transmit a very narrow range of wavelengths
and thus provides a monochromatic beam of light. Interference filters are fabricated earlier as
follows. A thin metallic film, usually of aluminium or silver, is deposited on a glass substrate
by vacuum deposition technique. Then a thin layer of cryolite is deposited over this. The
structure is again covered by another metallic film. Another plate is placed over it to protect the
thin film structure. By varying the thickness of the dielectric film, any particular wavelength
can be filtered out. However, the filtered light will have a narrow spectrum centered on the
chosen wavelength. By increasing the reflectivity of the surfaces, the transmitted spectrum can
be made narrower. But it is not possible to increase the thickness of metallic films indefinitely,
as they start absorbing the light.
Semireflective layers

max

HW Tmax

Glass

Transparent Dielectric
(a) (b)

Fig. 6.34

In modern versions metallic films are not used; instead dielectric films are used. In an
all dielectric interference filter, layers of dielectric materials of appropriate refractive indices
are deposited. To obtain an interference filter, a l/4 thick film of titanium oxide is deposited
and then over it a film of dielectric material with lower refractive index, such as magnesium
fluoride is deposited. On this, again a l/4 thick film of titanium oxide is deposited. In this way
alternately high and low refractive index materials are deposited to obtain an interference filter.
With multiple coatings, it is possible to fabricate filters, which are capable of transmitting a
Quantum Mechanics 553

C H A P T E R

20 Quantum Mechanics

20.1 INTRODUCTION
In 1925, de Broglie introduced the concept of matter waves and the idea of wave-particle
duality. He suggested that the wave-particle duality observed in case of light should be
extended to microparticles also. The combination of the idea of quantization with the idea of
wave-particle duality proved to be very fruitful for the development of quantum mechanics. The
whole apparatus of quantum mechanics was built in 1925-26. In 1925 Heisenberg suggested
that any reference to conceptual pictures which are not amenable to direct experimental
verification, should be discarded. He formulated matrix mechanics which is set in terms of the
observable quantities alone. In 1926, Schrödinger developed wave mechanics. His theory is
based on explicit use of a mental picture of matter wave which replaced the classical picture
of point particle. He developed the well-known differential equation for a wave function. The
problem of calculating the energy levels of a bound microparticle was reduced by Schrödinger
to the problem of finding eigenvalues. Heisenberg’s theory came to be known as matrix
mechanics while that of Schrödinger as wave mechanics. In 1926, Max Born proposed the
probability interpretation of the wave function. In 1927 the wave behaviour of microparticles
was confirmed by experiments on electron diffraction conducted simultaneously in several
different laboratories. In 1927, Heisenberg introduced the uncertainty principle. Through this
principle Heisenberg showed how the concepts of coordinate, momentum, energy etc should
be applied to microparticles. The uncertainty principle marked the final break of quantum
mechanics from classical determinism and established quantum mechanics as a statistical
theory. In quantum mechanics, the waves and particles are not classes of objects. They are
distinct modes of behaviour shared by all atomic particles. Every microparticle can behave
like a particle and like a wave too. In 1930, P.A.M.Dirac proposed a general formalism which
is a unifying concept of the matrix mechanics and wave mechanics.
The new laws applicable for atoms and subatomic particles constitute quantum
mechanics. The laws of conservation of momentum, angular momentum and energy are
still valid but we are not in a position to obtain on their basis a detailed description of the
motion of the subatomic particles. New ideas such as quantization of physical quantities and
allowed values of physical quantities are required to be incorporated into the theory. Planck’s
hypothesis of energy quanta, Einstein’s ideas on photons, de Broglie’s visualization of the
wave properties of micro-particles and the Heisenberg’s uncertainty principle provided the
basis for the development of quantum mechanics.
554 A Textbook of Engineering Physics

20.2 DE BROGLIE HYPOTHESIS

In 1924, Louis de Broglie extended the wave–particle dualism of light to the material particles.
He reasoned out that nature exhibits a great amount of symmetry. Therefore, if a light wave
can act as a wave sometimes and as a particle at other times, then particles such as electrons
should also act as waves at times. This is known as de Broglie hypothesis.
According to de Broglie hypothesis any moving particle is associated with a wave.
The waves associated with particles are known as de Broglie waves or matter waves. The
wavelength l of matter waves associated with a particle moving with velocity u is inversely
proportional to the magnitude of the momentum of the particle. Thus,
h
l= (20.1)
mυ
De Broglie deduced the connection between the particle and wave properties as follows.
De Broglie wavelength of Matter waves
As a photon travels with the velocity c, we can express its momentum as
E hν h
p= = =
c c λ
Thus, the wavelength and momentum p of a photon are related to each other through the
expression
h
l= (20.2)
p
De Broglie proposed that the relation (20.2) between the momentum and the wavelength
of a photon is a universal one and must be applicable to photons and material particles as
well.
The quantities n and l are wave properties and the quantities E and p are particle
properties. They are tied to each other through the relations
E = hv and p = h /l
which demonstrate that the wave and particle natures of a photon are intimately tied up to
each other. These equations reflect the wave-particle dualism of light.
Now let us consider a moving particle. A particle of mass ‘m’ moving with a velocity υ
carries a momentum p = mυ and it must be associated with a wave of wavelength
h h
l= = (20.3)
p mυ
The waves associated with moving particles are called matter waves or de Broglie
waves.
The relation l = h /mυ is known as de Broglie equation and the wavelength l is called
the de Broglie wavelength.
From equ.(20.3), we may draw the following conclusions.
1. l → ∞ when the velocity of the particle is zero. It means that matter waves are
detectable only for moving particles.
2. Lighter the particle, smaller the value of mass m and hence the longer is the wavelength
of the matter wave associated with it. Therefore, wave behaviour of micro-particles will
be significant whereas waves associated with macro-bodies can never be detected.
Quantum Mechanics 555

3. The smaller the velocity of the micro-particle, the longer is the wavelength of the matter
wave associated with it.
If a photon is considered to be a particle, then the corresponding electromagnetic wave
is the de Broglie wave for the photon. Similarly, atomic particles also can be viewed as
associated with matter waves, which do not have any similarity to any known waves. It is
later understood that the waves associated with particles are not real three dimensional waves
in the way sound waves are, but are probability waves related to the probabilities of finding
the particles in various places and with various properties.
De-Broglie wavelength associated with an accelerated charged particle
If a charged particle, say an electron is accelerated by a potential difference of V volts,
then its kinetic energy is given by K.E. = eV.
1
Or mυ2 = eV
2
2eV
\ u=
m
Then the electron wavelength is given by
h h m
l= = ⋅ .
mυ m 2eV
h
\ l= (20.4)
2emV

De Broglie Wavelength expressed in terms of K.E.

1 m 2 υ2 p 2
If a particle has kinetic energy K.E., then K.E. = mυ2 = =
2 2m 2m
or p= 2m ( K .E.)
h
\ l= (20.5)
2m ( K.E.)

De Broglie wavelength associated with particles in thermal equilibrium

If particles are in thermal equilibrium at temperature T, then their kinetic energy is given by
3
K.E. = kT
2
h h
\ l= = (20.6)
2m ( K.E.) 3mkT

20.3 DE BROGLIE’S JUSTIFICATION OF BOHR’S POSTULATE

Bohr’s theory of atomic structure was successful in explaining a large body of experimental
observations concerning atomic behaviour but the three adhoc postulates underlying his
theory remained without a valid theoretical justification for a long time. In support of his
hypothesis of matter waves, de Broglie demonstrated that it could provide an explanation for
the postulate regarding quantization of angular momentum of electron in Bohr’s model of
atom.
556 A Textbook of Engineering Physics

One of the postulates that Bohr used in formulating a model of atom is that the angular
momentum L of the electron revolving in a stationary orbit is quantized. Thus,
L = nħ (20.7)
The above postulate of Bohr follows directly from the concept of matter waves.
As the electron travels round in one of its circular orbits, the associated matter waves
propagate along the circumference again and again. A wave must meet itself after going round
one full circumference. If it did not meet, the wave would be out of phase with itself after
going round one orbit. After a large number of orbits, all possible phases would be obtained
and the wave would be annihilated by destructive interference. It implies that the wave should
produce a standing wave profile in the orbit to preclude the electron energy from radiating
away.
If a stretched string is fastened at both ends and is made to vibrate, standing waves are
formed provided the length of the string is an integral number of half–wavelengths of the
disturbance. If the string is formed into a circular loop, the condition for standing waves is
that the circumference of the loop should be an integral number of whole wavelengths of the
disturbance. Thus, if r is the radius of the circular loop,
2p r = nl    n = (1,2,3,....) (20.8)
We may regard the stationary electron orbits in an atom to be analogous to the circular
loop of string. We conclude that stationary electron wave pattern can form in the orbit if only
an integral number of electron wavelengths fit into the orbit, as shown in Fig. 20.1 (b).

Nucleus

n=3
n=6

(a) (b)
Fig. 20.1
The above equation (20.8) can be applied for electron waves, taking λ as de Broglie
wavelength of electron waves.
The de Broglie wavelength of electron wave is given by
h
l=
mυ
where υ is the speed of the electron in the orbit. Using the de Broglie wavelength into
equ.(20.8), we obtain
nh
2pr =
mυ
nh
\ mur = (20.9)
2π
But the quantity ‘mυr’ is the angular momentum, L, of electron in the orbit of radius r.
Thus, L = mυr
Quantum Mechanics 557

It, therefore, follows that

L = nħ
which is precisely Bohr’s postulate. De Broglie thus demonstrated that the quantization of
angular momentum is a direct consequence of wave nature of electron.
Let us calculate the wavelength of the electron in the first orbit of hydrogen atom. The
electron speed υ in the orbit is given by
e
u=
4πε o mr
h 4πε o r
\ l= (20.10)
e m
Taking r = 5.3 ×10–11 m, we get l = 3.3 Å. The circumference of the orbit is 2πr = 3.3 Å.
Hence, the first orbit of the electron in a hydrogen atom corresponds to one complete
electron wave joined on itself. The de Broglie hypothesis thus offered a new meaning to the
quantum number ‘n’. n is the number of de Broglie wavelengths that fit into the circumference
of Bohr allowed orbits.
We may now picture the electron in the atom in two ways: either as a particle moving in
an orbit with a certain quantized value of mυr, or as a standing de Broglie wave occupying a
certain region around the nucleus.
20.4 DE BROGLIE WAVES ARE INSIGNIFICANT IN CASE OF
MACRO-BODIES
According to de Broglie hypothesis a moving body is associated with matter waves and the
wavelength of the waves is given by
h
l=
mυ
where u is the velocity with which the body moves.
As the mass m of the body increases, the wavelength tends to be insignificant. Therefore,
the wavelength associated with macroscopic bodies become insignificant in comparison to the
size of the bodies themselves even at very low velocities. Because of the smaller magnitude
of Planck’s constant h, the wavelength l will be significant only in case of micro-particles.
For example, if we consider a cricket ball of mass 500 gm flying with a velocity of
50 km/hr, its wavelength comes to
6.62 × 10−34 J .s
l= = 10 −34 m = 10 −24 Å.
0.5kg × 13.9m / s
It is easy to see that this wavelength is insignificant in comparison to the size of the ball.
On the other hand, if we consider the case of an electron, having energy 100 eV, the de
Broglie wavelength of the electron is given by
h
l=
2meV
6.62 × 10−34 J .s
= = 1.33 Å.
2 × 9.11 × 10−31 kg × 1.602 × 10−19 C × 100V

The size of an electron is about 10–5Å, which is far smaller than the wavelength of 1.33Å.
It means that the electron behaves more as a wave than a particle under the circumstances.
558 A Textbook of Engineering Physics

20.5 PROPERTIES OF MATTER WAVES

1. Matter waves are produced by the motion of the particles and are independent of the
charge. Therefore, they are neither electromagnetic nor acoustic waves but are new kind
of waves.
2. They can travel through vacuum and do not require any material medium for their
propagation.
3. The smaller the velocity of the particle, the longer is the wavelength of the matter wave
associated with it.
4. The lighter the particle, the longer is the wavelength of the matter wave associated with
it.
5. The velocity of matter waves depends on the velocity of the material particle and is not
a constant quantity.
6. The velocity of matter waves is greater than the velocity of light.
7. They exhibit diffraction phenomenon as any other waves.

20.6 DAVISSON–GERMER EXPERIMENT

Waves exhibit diffraction. If the de Broglie hypothesis is valid, then the matter waves should
exhibit diffraction effects. Diffraction is observed when the wavelength is comparable to the
size of the object causing diffraction. The wavelength of 100 eV electrons is of the order of 1
Å and the interatomic spacing in a crystal is of the order of 2 to 3 Å. Therefore, we expect that
the wave behaviour of a micro-particle such as an electron becomes noticeable when a beam
of particles interact with crystals. In 1927, Davisson and Germer observed the diffraction of
an electron beam incident on a nickel crystal. The experiment provided a convincing proof of
the wave nature of matter.
Apparatus
The experimental arrangement of Davisson and Germer is shown in Fig. 20.2. The apparatus
consisted of an electron gun, which produced collimated beam of electrons. An anode, A,
connected to a variable voltage source accelerated the electrons. The energy of the electrons
can be computed from the accelerating potential. These electrons were scattered by a nickel
crystal located at C. The crystal can be rotated on the axis. The number of electrons scattered
by the crystal in different directions was measured with the help of a detector D, which can be
moved on a scale.
F G

A
D
D V = 54V

Circular
scale

Nickel crystal =50°

Fig. 20.2 Fig. 20.3

Quantum Mechanics 559

The experimental arrangement of Davisson and Germer is shown in Fig. 20.2. An electron
beam is generated from a hot tungsten filament F and an anode A connected to a variable
voltage source accelerated the electrons. The energy of the electrons can be computed from
the accelerating potential, V applied between the filament F and the anode A. The electrons
emerge through an opening in the anode and fall normally on the surface of a nickel crystal,
C. These electrons are scattered by a nickel crystal located at C. The crystal can be rotated on
the axis. The detector D measured the number of electrons scattered by the crystal in different
directions. The detector could be moved on a graduated semicircular scale. Thus, the intensity
of the scattered electron beam was determined as a function of the scattering angle, f.
Investigations
During their experiments Davisson and Germer moved the detector on the circular scale to
various positions and the current was measured. The detector current is a measure of the
intensity of the diffracted beam. A polar graph was then plotted between the detector current
and the angle between the incident beam and the diffracted beam. Such polar curves were
obtained for electrons accelerated through different voltages. It was found that a hump
appears in the polar curve when 44 eV electrons were incident on the crystal. The hump grew
in size as the accelerating voltage is increased and became most pronounced at 54 volts. The
polar curve corresponding to 54 V is shown in Fig. 20.3. It is found that for the accelerating
voltage of 54 volts, the electrons are scattered more pronouncedly at an angle of 50° with
the direction of the incident beam. The maximum is an indication that electrons are being
diffracted. Incident beam
Reflected
Analysis 50° beam
It may be interpreted that the rows of atoms at the
25° 25°
surface of the nickel crystal act like rulings of a
natural diffraction grating and the de Broglie waves
associated with the electrons underwent diffraction
when they were incident on the crystal. The hump 65°
produced at 50° in Fig. 20.3 then corresponds
to the first order diffraction maxima. Braggs’ d
law, applicable for X-ray diffraction by crystals,
would be valid for electron wave diffraction also.
Fig. 20.4 shows atomic planes and the incident
and scattered beams. The interplanar spacing is
Fig. 20.4
obtained from X-ray analysis to be d = 0.91 Å.
From the Fig. 20.4, it is seen that the glancing angle q = 65°. Applying Braggs’ equation,
l = 2d sin q = 2 × 0.91 Å × sin 65° = 1.65 Å.
The wavelength of the electron wave is thus determined to be 1.65 Ǻ. The wavelength of
electron wave can be computed from the accelerating potential V using de Broglie equation.
h
l=
2meV
6.63 × 10−34 Js
= = 1.66Å.
 2 × 9.1 × 10−31 kg × 1.602 × 10−19 C × 54V 
 
It is seen that the values obtained experimentally using Braggs’ equation and de Broglie
equation agreed well. Therefore, Davisson-Germer experiment gave conclusive evidence that
electrons exhibit diffraction property.
560 A Textbook of Engineering Physics

20.7 G.P. THOMSONS’S EXPERIMENT

The de Broglie hypothesis was further
T
confirmed in 1927 by the experiments
conducted independently by G.P.Thomson in
C
England and by Kikuchi in Japan.
Thomson’s experimental arrangement is F P
shown in Fig. 20.5. Electrons are produced
from a heated filament F and accelerated A B G
through a high positive potential given to the
Fig. 20.5
anode A. The whole apparatus is kept highly
evacuated. The electron beam passes through a fine hole in a metal block B and falls on
a gold foil of thickness 0.1 mm. The electrons passing through the foil are received on a
photographic plate P.
Metals are polycrystalline in which the grains are oriented completely at random.
Therefore, some grains have always the right inclination θ towards the incident beam in order
to produce a Bragg reflection. Owing to the random orientation, the reflections from a given
set of lattice planes at the glancing angle occur in every azimuth about the incident beam.
Consequently, the reflected beams form a cone of semi-vertical angles 2θ. Each set of lattice
planes with its particular spacing d in the grain produces its own cone of diffracted rays. A
concentric rings pattern is produced on the photographic plate when it intercepts the coaxial
cones of diffracted rays.
The diffraction pattern produced by the electron beam was strikingly similar to the x-ray
diffractions obtained from powder samples. Thus, the experiments of G.P. Thomson and
Kikuchi provided irrefutable proof to the existence of de Broglie waves.
In 1937, C.J.Daavisson and G.P.Thomson were jointly awarded the Noble Prize in physics
for their experimental discovery of electron diffraction.
In 1929, soon after the discovery of wave properties of electrons, the German physicist
Otto Stern and his coworkers detected diffraction phenomena with neutral atomic and
molecular beams.
20.8 VELOCITY OF DE BROGLIE WAVES
Any harmonic wave is characterized by a precise wavelength l and constant amplitude. It is
non-localized and has no beginning and end. It means that such a wave extends over a very
large volume of space.
20.8.1 Phase Velocity
If we consider a harmonic wave, the wave has a single wavelength and a single frequency.
The velocity of propagation of the wave is given by
υp = nl
Using, v = w/2p and l = 2p/k into the above equation, we get
ω 2π ω
up = ⋅ = (20.11)
2π k k
υp is called the phase velocity. The velocity with which the plane of equal phase travels
through a medium is known as the phase velocity. It thus represents the velocity of propagation
of the wave front.
As E = hv and p = h/l, we get
E h E
up = ⋅ = (20.12)
h p p
Quantum Mechanics 561

(i) When the atomic particle velocity is non-relativistic, the total energy E = mc2 and
momentum p = mu.
Therefore, the phase velocity of the de Broglie wave associated with the particle is
E mc 2 c 2
up = = = (20.13)
p mυ υ
As u < c, the phase velocity of the de Broglie wave associated with the atomic particle
is always greater than c.
(ii) When the atomic particle velocity is relativistic, the total energy E = mo2c 4 + p 2c 2 ,
where m­o is the rest mass of the particle.
Therefore, the phase velocity of the de Broglie wave associated with the particle is
1/2
E  mo2c 4 + p 2c 2 
up = = 
p  p2 
1/2 1/2
 m 2c 2   m 2c 2 λ 2 
= c  o 2 + 1 = c  o 2 + 1 (20.14)
 p   h 
m 2c 2 λ 2
As the term o 2 is always a positive quantity, the phase velocity of the de Broglie wave
h
associated with the atomic particle is always greater than c.
According to the theory of relativity, it is not possible that the velocity of the particle
wave be greater than or equal to the velocity of light. Hence, a harmonic wave of wavelength
l cannot represent a moving atomic particle. Thus, de Broglie waves cannot be harmonic
waves.
20.9 WAVE PACKET – REPRESENTS A MICROPARTICLE
We have so far assumed that a particle may be represented by a monochromatic de Broglie
wave. However, a wave spreads over a large
region of space and cannot represent a highly
localized particle. Schrödinger postulated that a
wave packet rather than a single harmonic wave
represents a particle. A wave packet consists
of a group of harmonic waves. Each wave has
slightly different wavelength. The superpo-
sition of a very large number of harmonic
waves differing infinitesimally in frequency
will produce a single wave packet (see Fig. 20.6
c). The waves interfere constructively over only
a small region of space and cancel each other
everywhere except in that small region. The Fig. 20.6: Formation of a Wave packet. (a)
two waves of slightly different frequencies
position of the particle would then be approxi-
produce constructive interference. (b) three
mately determined by the position of the wave waves produce interference maxima of
packet. larger size separated by larger distance.
The velocity with which the wave packet (c) A large number of waves-having slightly
propagates is called the group velocity ug. different frequencies produces only one
maxima and it is called a wave packet.
562 A Textbook of Engineering Physics

The individual waves forming the wave packet propagate at a velocity known as the phase
velocity up.
20.9.1 Group Velocity
When a number of plane waves of slightly different wavelengths travel in the same direction,
they form wave groups or wave packets. The velocity with which the wave group advances
in the medium is known as the group velocity υg. Each component wave has its own phase
velocity, υp = nl. The wave packet has amplitude that is large in a small region and very small
outside it. The amplitude of the wave packet varies with x and t. Such a variation of amplitude
is called the modulation of the wave. The
velocity of propagation of the modulation is
known as the group velocity, υg.
Here, we should note that wave packets
are only theoretical artifices to aid our
visualization of various phenomena in the
micro-world.
Expression for the Group Velocity Fig. 20.7 Beats are formed when two waves of
We derive now an expression for group slightly different frequencies combine
velocity considering a group of waves (a) the individual waves (b) the resultant wave.
consisting of two components of equal
amplitude and slightly differing angular velocities ω1 and ω 2 .
Let the waves in Fig. 20.7 (a) be represented by the equations
y1 = A sin (w1t – k1x)
y2 = A sin (w2t – k2x)
The superposition of these two waves is given by
y1 + y2 = A sin (w1t – k1x) + A sin (w2t – k2x)
 α+β   α − β 
Using the trigonometric relation sin α + sin β = 2sin   sin   , we write the above
 2   2 
equation as
 ( ω + ω 2 ) ( k1 + k2 )   ( ω − ω 2 ) ( k1 − k2 ) 
y1 + y2 = 2 A sin  1 t− x  cos  1 t− x
 2 2   2 2 
 ∆ω t ∆kx 
= 2A sin ( ωt – kx ) cos  –  (20.15)
 2 2 
where w = (w1+ w2)/2, k = (k1+ k2)/2, ∆w = w1 – w2 and ∆k = k1 – k2. Equ.(20.15) represents
the resultant wave which is seen to have the following two parts.
(i) A wave of angular frequency w and propagation constant k , moving with a velocity
ω
up = = νλ and
k
(ii) A second wave of angular frequency Dw/2 and propagation constant Dk/2, moving
∆ω
with a velocity ug = .
∆k
When ∆w and ∆k are very small, we can write the above equation as
dω
ug = (20.16)
dk
Quantum Mechanics 563
2π d ν dν
or ug = = −λ 2
2π d (1 / λ ) dλ

20.9.2 Relation between Phase Velocity and Group Velocity

The velocity of the individual component wave of the wave packet is given by
up = nl
Using, v = w / 2p and l = 2p / k into the above equation, we get
ω 2π ω
up= ⋅ = (20.17)
2π k k
∴ w = kup
The group velocity is given by the relation (20.16) as
dω d dυp
ug=
dk dk
= ( )
kυp = υp + k
dk
2π
But k= .
λ
2π
Therefore, dk = − 2 d λ
λ
k λ
and = − .
dk dλ
dυp
\ ug = υ p − λ (20.18)
dλ
Group velocity will be the same as phase velocity if the entire constituent waves travel
with the same velocity. It means that in a nondispersive medium, υg = υp. However, the waves
of different wavelengths travel in a medium with different velocities. Therefore, the group
velocity is in general less than the phase velocity.
20.9.3 The Velocity of a Particle Equals the Group Velocity of the
Associated Matter Waves
A particle moving with a velocity u is supposed to consist of a group of de Broglie waves.
The group velocity of a wave packet is given by
dω
ug =
dk
 d ω   dE   dp 
which we can write as ug =  (20.19)
 dE   dp   dk 

ω h dω 1
As E = hν = h ⋅= ⋅ ω = ω , =
2π 2π dE 
h 1 k h dp
and p= = h⋅ = h⋅ = ⋅ k = k , =
λ λ 2π 2π dk
 d ω   dE   dp  1  dE   dE 
\ ug =  = =  (20.20)
 dE   dp   dk    dp   dp 

1 ( mυ )
2
1 p2
For a particle, E = mυ 2 = = .
2 2 m 2m
564 A Textbook of Engineering Physics
dE p
\ ug = = = υ. (20.21)
dp m
Thus, the de Broglie wave group associated with an atomic particle travels with the same
velocity as that of the particle itself.
20.9.4 Relation Between the Group Velocity and Particle Velocity (in a
Non-dispersive Medium)
A particle moving with a velocity u is supposed to consist of a group of de Broglie waves. For
an atomic particle of rest mass mo moving with a velocity u, the total energy and momentum
are given by
mo c 2 mo υ
E = mc 2 = and p = mυ = respectively.
2 2
1− υ / c 1 − υ2 / c 2
The frequency of the associated de Broglie wave is
E mo c 2 2π mo c 2
n = = and w = 2πν = .
h h 1 − υ2 / c 2 h 1 − υ2 / c 2
2πmo
Therefore, dw = υ⋅ dυ . (20.22)
( )
3/2
h 1 − υ2 / c 2
The wavelength of the de Broglie wave is
( )
2 2 1/2
h h 1− υ / c 2π 2π mo υ
l = = and k = =
( )
1/2
p mo υ λ h 1 − υ2 / c 2

2πmo 
( ) υ
( ) 
2 2 −1/2 −3/2
∴ dk =  1 − υ / c dυ + υ⋅ 1 − υ2 / c 2 d υ
h c2 
2πmo d υ
or dk = (20.23)
( )
3/2
h 1 − υ2 / c 2
Dividing eq. (20.22) by (20.23), we get
dω
ug = =υ (20.24)
dk
Thus, the de Broglie wave group associated with an atomic particle travels with the same
velocity as that of the particle itself in a non-dispersive medium.
20.10 APPLICATIONS OF DE BROGLIE WAVES
We discuss here some of the applications of de Broglie waves.
1. Possible energy states of a microparticle trapped in a box
Let us consider a microparticle trapped in a one-dimensional box of length L. According to de
Broglie hypothesis, the particle is associated with a wave having a wavelength l. The particle
cannot move beyond the walls of the box. Hence, the amplitude of the de Broglie wave drops
to zero at the walls. It implies that the de Broglie wave of the particle forms a standing wave
pattern with nodes at the walls. The formation of standing wave pattern requires that the
distance L must be an integral multiple of half-wavelength. Thus,
Quantum Mechanics 565

λ 2L
L= n or l = (20.25)
2 n
where n = 1,2,3,......
The possible values of linear momentum are given by
h h
p= =n (20.26)
λ 2L
The possible values of the kinetic energy of the microparticle are given by
p 2 n2h2
K.E. = = (20.27)
2m 8mL2
The above expression indicates that a microparticle trapped in a box can take only certain
discrete energy states. Secondly, the particle cannot have zero energy and the minimum
h2
kinetic energy that it can take is when n = 1.
8mL2

2. Neutron diffraction
Experiments have showed that neutrons exhibit diffraction. The diffraction of neutrons is
used to study atomic structures of solids containing hydrogen atoms. Normally, x-ray and
electron diffraction analysis are used to determine the structures of solids. In solids containing
hydrogen atoms, the scattering of x-rays and electrons by hydrogen is not sufficient.
Therefore, it is difficult to locate the position of hydrogen atoms using these techniques. In
contrast, neutrons interact largely with atomic nuclei, notably hydrogen nuclei. Therefore,
the scattering of neutrons by hydrogen-containing atoms in solids reveals the presence and
location of hydrogen atoms in the lattice.
Matter waves are less penetrating than x-rays and so are useful in studying surface
features of materials.
3. Electron microscope
A microscope is an optical instrument used to magnify small objects in order to study their
structural details. It consists of two high power lenses called the objective and the eyepiece.
The objective forms a real image of the object kept in front of it and the image is viewed
by the eye through the eyepiece. The magnification of the object by the microscope is given
by the product of the magnifications of the objective and eyepiece. It may appear at the first
instance that one may get an image magnified to any desired extent by increasing the magnifi-
cations of the objective and eyepiece. It does not happen so in practice. The light reflected
from each point of the object has to pass through the objective, which is a circular aperture.
Consequently, circular diffraction is produced by the objective corresponding to each point.
The higher the objective power, the smaller the aperture and the larger is the diffraction
pattern. When the image is viewed further through the eyepiece of higher power, the points
appear as blurred patches overlapping on each other and the details are not discernible. Thus,
the diffraction effects restrict the ultimate useful magnification achievable by a microscope.
The maximum useful magnification of an optical microscope is about 1000×. Any further
magnification does not show more details though a larger image is obtained. The diffraction
effects depend on the wavelength of light. The useful magnification can be increased by
making use of shorter wavelength radiation. Thus, UV radiation can give higher magnifi-
cation of around 2000×.
566 A Textbook of Engineering Physics

Very large magnification is obtained by exploiting the wave character of electrons. The de
Broglie wavelengths of electrons are extremely small, of the order of 0.001 to 1 Å as against
the light wavelength of 5000 Å. As a result, magnifications
of the order of 106 × can be easily attained using electron
waves. The Transmission Electron Microscope (TEM)
was the first type of Electron Microscope to be developed
and is built exactly on the model of Light Transmission
Microscope except that a focused beam of electrons is
used instead of light to “see through” the specimen. It was
developed by Max Knoll and Ernst Ruska in Germany in
1931.
Construction: A schematic diagram of the electron
microscope is shown in Fig. 20.8. It is essentially a very
large modified cathode ray tube. It consists of an electron
gun at one end of the tube, a number of magnetic lenses
in the path of the electron beam and a fluorescent screen
at the other end of the tube. Each magnetic lens is a
solenoid encased in soft iron and has a soft iron pole piece
to concentrate the magnetic field lines. The focal lengths
are of the order of a few millimeters and can be varied by
varying current through the solenoid. The tube is mounted
vertically as illustrated in Fig. 20.8. Fig. 20.8: Transmission
Electron Microscope (TEM)
Working: Electrons are emitted by a hot cathode and are
accelerated to high velocities with the help of an anode held
at about 50 to 100 kV. The electrons pass through a magnetic lens that acts as a condenser
lens and is formed into a parallel beam. The electron beam then passes through the specimen
to be viewed. The specimen is prepared in the form of a very thin slice of thickness of about
100 to 1000 Å so that electrons are not scattered and blur the image. Different number of
electrons passes through different portions of the specimen depending on its structure. After
passing through the specimen, the electron beam goes through a second lens, which acts as
the objective lens and forms an intermediate image of the object. After that the beam passes
through a third lens that acts as the eyepiece and forms the final image of the object on a
fluorescent screen. The fluorescent screen converts the image into an optical image. The
magnified image of the object is viewed through a side window. The resolving power of an
electron microscope is of the order of 10 to 100 Å and the magnification of the order of 106 to
107 is easily attainable.
In view of high magnification power the electron microscope proved to be a very
valuable tool for studying the microstructures of a variety of materials and also for studying
micro-organisms. The structural details of virus, proteins etc could be understood only with
the help of an electron microscope.
4. Scanning electron microscope (SEM)
An important variation is the scanning electron microscope. A schematic diagram of the
scanning electron microscope is shown in Fig. 20.9. In this, a beam of electrons is generated
in the electron gun, located at the top of the column. This beam is attracted through the anode,
condensed by a condenser lens, and focused as a very fine point on the sample by the objective
Quantum Mechanics 567

lens. The scan coils are energized Electron gun

(by varying the voltage produced
by the scan generator) and create
a magnetic field which deflects the
beam back and forth in a controlled Condenser Lens Aperture
pattern. The varying voltage is also Scan generator
applied to the coils around the neck Scan coils
of the Cathode-ray tube (CRT) Objective Lens CRT display/
—
e camera
which produces a pattern of light Sample
deflected back and forth on the Electron Amplifier
surface of the CRT. The pattern of collector High vacuum pump

deflection of the electron beam is

Fig. 20.9: Scanning electron microscope (SEM)
the same as the pattern of deflection
of the spot of light on the CRT. The electron beam hits the sample, producing secondary
electrons from the sample. These electrons are collected by a collecting anode that is held
at a positive potential with respect to the specimen. The current in the electron collector is
converted to a voltage, and amplified. The amplified voltage is applied to the grid of the
CRT and causes the intensity of the spot of light to change. The image consists of thousands
of spots of varying intensity on the face of a CRT that correspond to the topography of the
sample.
Advantages: Generally, the TEM resolution is about an order of magnitude greater than
the SEM resolution. However, the advantages of SEM are as follows. (i) As the process of
forming image involves surface processes, bulk samples can be used. (ii) It provides a much
greater depth of view, and so can produce images that are a good representation of the 3D
structure of the sample.
Example 20.1: An electron beam is accelerated from rest through a potential difference of
200 V.
(i) Calculate the associated wavelength.
(ii) This beam is passed through a diffraction grating of spacing 3 Å. At what angle of
deviation from the incident direction will be the first maximum observed?
Solution: The wavelength of the waves associated with the electron beam is given by
h
l=
2meV

6.63 × 10−34 Js
=
2 × 9.11 × 10−31 kg × 1.602 × 10−19 C × 200V
= 0.86 Å.
The diffraction is governed by the equation 2d sin q = ml
For first order m = 1 and 2d sin q = l .
 λ  0.86 × 10−10 m 
\ q = sin −1   = sin −1   = 8.31°.
 2d   2 × 3 × 10−10 m 
568 A Textbook of Engineering Physics

Example 20.2. An enclosure filled with helium is heated to 400K. A beam of He-atoms
emerges out of the enclosure. Calculate the de Broglie wavelength corresponding to He
atoms. Mass of He atom is 6.7×10–27kg.
h
Solution. De Broglie wavelength l =
2mkT
6.63 × 10−34 Js
=
2 × 6.7 × 10 −27 kg × 1.376 × 10−21 J / deg × 400
= 0.769 Å
Example 20.3: Find the de Broglie wavelength of
(i) an electron accelerated through a potential difference of 182 volts, and
(ii) a 1 kg object moving with a speed 1 m/s. Comparing the results explain why the wave
nature of matter is not more apparent in daily observations.
Solution:
h 6.626 × 10−34 J .s
(i) le = =
2 emV ( )(
2 1.602 × 10−19 C 9.11 × 10−31 kg 182V )
6.626 × 10−34 J .s. 2
−10 kg .m / s
  = −24
= 0.91 × 10 = 9.1 × 10−11 m = 0.91Å
7.29 × 10 kg .m / s kg.m / s

h 6.626 × 10−34 J .s kg .m 2 / s
(ii) λ m = = = 6.6 ×10−34 = 6.6 × 10−34 m .
Mυ 1kg × 1m / s kg .m / s
It is seen from the above that the wavelength of the accelerated electron is about 105
times larger then its own size (≈10-15 m) and is therefore significant. On the other hand, the
wavelength associated with the macroscopic object is negligibly small and is thus not apparent
in its interactions with other objects.
20.11 HEISENBERG UNCERTAINTY PRINCIPLE
The wave nature of atomic particles leads to some inevitable consequences. Classically, the
state of a particle can be defined by specifying its position and momentum at any given time t.
If a body is moving along x-direction with a velocity u, its position is given by x = u t and its
momentum by p = mu. From this,
p
x= t (20.28)
m
At each instant, the position and momentum can be measured to a very high accuracy.
When an atomic particle is conceptualized as a de Broglie wave packet such a precision
becomes restricted.
Schrödinger postulated that a moving microparticle is equivalent to a wave packet.
A wave packet spreads over a region of space. Therefore, it is difficult to locate the exact
position of the microparticle. Although the particle is somewhere within the wave packet, it
is impossible to know where exactly the particle is at a given instant. If the linear spread of
the wave packet is Δx, the particle would be located somewhere within the region Δx. The
probability of finding the particle is a maximum at the centre of the wave packet and falls off
to zero at its ends. Therefore, there is an uncertainty Δx in the position of the particle. As a
Quantum Mechanics 569

result, the momentum of the particle at that instant cannot be determined precisely. It means
that the location and momentum of a microparticle cannot be simultaneously determined
with certainty. Any attempt to determine these variables will lead to uncertainties in each of
the variables.
In 1927 Heisenberg showed that the product of uncertainty Dx in the x-coordinate of a
quantum particle and the uncertainty Dpx in the x-component of the momentum would always
be of the order of Planck’s constant h. Thus,
Dx ⋅ Dpx ≈ h

or more precisely Dx ⋅ Dpx ≥ (20.29)
2
This is known as Heisenberg’s uncertainty principle for position and momentum, which
may be stated as follows:
“It is not possible to know simultaneously and with exactness both the position and the
momentum of a microparticle”.
The Uncertainty Principle implies a built-in, unavoidable limit to the accuracy with which
we can make measurements. Classically, it is thought that the precision of any measurement
was limited only by the accuracy of the instruments the experimenter used. Heisenberg
showed that whatever may be the accuracy of the instruments used, quantum mechanics limits
the precision when two properties are measured at the same time. These are not just any two
properties but pairs of measurable quantities whose product has dimensions of energy × time.
Such quantities are called conjugate quantities in quantum mechanics, and have a special
relation to each other. Position–linear momentum, energy-time, time-frequency and angular
momentum-angular displacement are conjugate pairs of variables.
The uncertainty principle asserts that it is physically impossible to know simultaneously
the exact position (Dx = 0) and exact momentum (Dpx = 0) of a micro-particle. According to
it, the more precisely we know the position of the particle, the less precise is our information
about its momentum. To localize a wave packet, we have to add more wavelengths to form
the wave packet. More wavelengths mean larger ∆l and more uncertainty in momentum (note
that Dp ∝ Dl). Conversely, in order to have more precise value of momentum, the wave packet
should contain less number of waves. Less number of waves produces a longer wave packet.
Thus, the momentum of a particle cannot be precisely specified without our loss of knowledge
of the position of the particle at that time. Similarly, a particle cannot be precisely localized in
a particular direction without our loss of knowledge of momentum in that particular direction.
We can at best specify that certain momentum of the particle is more probable than the other
or that the particle is more likely to be here than there. We cannot use classical notions like
coordinates and momentum to describe the motion of quantum particles. Thus, the uncertainty
principle implies that we can never define the path of an atomic particle with the absolute
precision indicated in classical mechanics. Therefore, concepts such as velocity, position, and
acceleration are of limited use in quantum world.
Relations similar to (20.29) hold good for other components of position and linear
momentum. Thus,

Dy ⋅ Dpy ≥ (20.29a)
2

Dz ⋅ Dpz ≥ (20.29b)
2
570 A Textbook of Engineering Physics

20.11.1 Energy - Momentum Uncertainty

The uncertainty relation for the simultaneous measurement of energy E and time t is expressed
as

DE ⋅ Dt ≥ (20.30)
2
The physical significance of the energy-time uncertainty relation is different from that of
the position –momentum uncertainty. If ∆E is the maximum uncertainty in the determination
of the energy of a particle, then the minimum time interval for which the particle remains in
that state is given by
/2
Dt =
∆E
And, if a particle remains in a particular energy state for a maximum time Dt, then the
minimum uncertainty in the particle energy is given by
/2
DE =
∆t
Derivation: We can obtain the result (20.30) as follows. Let us consider a microparticle of
mass m moving with a velocity u. Its kinetic energy will be
1
E = mυ2
2
1 
If the uncertainty in the energy is DE, then ∆E = ∆  mυ2  = mυ ∆υ = υ ∆p
2 
∆x ∆x
As the velocity υ = , the uncertainty in energy may be written as ∆E = ∆p
∆t ∆t
Thus, ∆E ⋅ Dt = ∆x ⋅ Dp

But ∆x ⋅ Dp ≥
2

Therefore, ∆E ⋅ Dt ≥
2
The above relations are to be supplemented by the following uncertainty relation

∆Mx ⋅ Dϕx ≥ (20.31)
2
where DMx is the uncertainty in the projection of the angular momentum on the x-axis and
DFx is the uncertainty in the angular coordinates of the microparticle.
By analogy with (20.28a) and (20.28b), we may write down relations for other projections
of momentum and angular momentum as follows:
∆Mx ⋅ Dϕy ≥ ħ/2 and ∆Mz ⋅ Dϕz ≥ ħ/2 (20.31a)
In general if q and p denote two canonically conjugate variables, the uncertainty relation
is given by
Dq ⋅ Dp ≥ ħ/2 (20.32)
The above relations do not mean that the uncertainty principle creates certain obstacles
to the understanding of the atomic phenomena; it only reflects certain peculiarities of the
objective properties of a quantum particle.
Quantum Mechanics 571

20.12 ELEMENTARY PROOF OF UNCERTAINTY PRINCIPLE USING DE

BROGLIE WAVE CONCEPT
A wave packet produced by a superposition of large number of harmonic waves is shown
in Fig. 20.10. Since a wave packet is not an infinite harmonic wave, it has a range of wave
numbers Dk instead of one definite wave number.
Dk is thus the uncertainty in wave number. Further,
the position of the particle cannot be given with
certainty. It will lie somewhere between the two
consecutive nodes (see Fig. 20.7b and Fig. 20.10).
Thus, the uncertainty in the position of the particle
is equal to the distance between two consecutive
nodes. Referring to equ. (20.15), the condition for
formation of a node is
 ∆ωt ∆kx  Fig. 20.10
cos  −  =0
 2 2 
∆ω ∆k π 3π 5π π
i.e., t− x = , , , .... , ( 2n + 1)
2 2 2 2 2 2
Thus, if x1 and x2 are the positions of two consecutive nodes, then
∆ω ∆k π
t− x1 = ( 2n + 1)
2 2 2
∆ω ∆k π
and t− x2 = ( 2n + 3)
2 2 2
Subtracting the above upper equation from the lower one, we find that
∆k ∆k

2
( x2 − x1 ) = p or
2
∆x = π
where Dx = (x2 – x1) is the uncertainty in the position of the particle. Thus,
2π
Dx = .
∆k
2π 2πp
Since k= = ,
λ h
2π
Dk = ∆p
h
2π 2πh h
Therefore, Dx = = =
∆k 2π ∆p ∆p
or Dx ⋅ Dp = h
When we consider a group consisting of very large number of harmonic waves of continu-
ously varying frequencies, the product of the uncertainties comes to
1
Dx ⋅ Dp ≥ 
2

20.13 IMPLICATION OF UNCERTAINTY PRINCIPLE

The uncertainty principle expresses a fundamental limitation in nature that also limits the
precision of our measurements. According to classical mechanics the position and momentum
572 A Textbook of Engineering Physics

of a macro-particle can be determined exactly. But the uncertainty principle asserts that it
is physically impossible to know simultaneously the exact position (Dx = 0) and exact
momentum (Dpx = 0) of a microparticle. According to it, the more precisely we know the
posistion of the particle, the less precise is our information about its momentum. Thus, the
momentum of a particle cannot be precisely specified without our loss of knowledge of the
position of the particle at that time. Similarly, a particle cannot be precisely localized in a
particular direction without our loss of knowledge of momentum in that particular direction.
We can at best specify that certain momentum of the particle is more probable than the other
or that the particle is more likely to be here than there. It means that our classical notions like
coordinates and momentum derived from ordinary macroscopic experiences are inadequate to
describe the atomic world. The uncertainty principle points out that in the microscopic world,
(1) the dynamical variables of a particle are combined in sets of simultaneously determined
quantities which are known as complete sets of quantities;
(2) the coordinate and momentum components of a particle etc are pairs of concepts which
are interrelated and fall in different complete sets of quantities. They cannot be defined
simultaneously in a precise way.
Thus, the uncertainty principle implies that we can never define the path of an atomic
particle with the absolute precision indicated in classical mechanics. Therefore, concepts such
as velocity, position, and acceleration are of limited use in quantum world. To describe the
quantum particle the concept of energy becomes important since it is related to the state of the
system rather than to its path.
20.14 UNCERTAINTY PRINCIPLE IS NOT SIGNIFICANT IN CASE OF
MACRO-BODIES
The Heisenberg Principle is of no practical importance for heavy bodies where the de Broglie
wavelength is negligibly small.
For example, let us take the case of a cricket ball in flight. The indeterminacy in the
position of the ball is, say, 1 mm. We can determine the indeterminacy of velocity of the ball
from uncertainty principle.
Dx ⋅ Dp ≈ h
\ Dx ⋅ mDu ≈ h
h 6.62 × 10−34 J .s
Du ≈ = ≅ 10 −30 m / s.
m∆x 0.5kg × 10 −3 m
The above inaccuracy is negligible and not detectable. It implies that the uncertainties are
of no importance in case of macro bodies; and the position and velocity of a macro body can
be simultaneously determined with a high degree of accuracy. As a result, macroscopic body
follows a well defined trajectory.
In contrast if we take the example of an electron orbiting in a hydrogen atom, the
inaccuracy in its position is ± 1Å. The uncertainty in its speed is
h 6.62 × 10−34 J .s
Du = = ≈ 2 × 105 m / s
m∆x 9.11 × 10 −31 kg × 2 × 10−10 m
which is of the same order as the velocity of the electron in the orbit. It means that it is not
possible to determine the velocity and the position of a microparticle with certainty and as
such we cannot talk of a specific trajectory. Instead we have to be content knowing only the
probable values.
Quantum Mechanics 573

20.15 THOUGHT EXPERIMENTS

It is not possible to verify Heisenberg uncertainty principle in the laboratory. Therefore, we
illustrate it with the help of two thought experiments.
1. Treating the electron as a wave
We assume that an electron has wave
character. Suppose that we want to determine
the y-coordinate of an electron moving
along the x-axis. We place a slit of width ‘d’
perpendicular to the direction of motion of
the electron (Fig. 20.11). The precision of
the position of electron, Δy, in y-direction is
limited by the size of the slit. That is, Dy ≈ d. Fig. 20.11
If the slit is narrow enough, it causes a
change in the motion of the electron after going through the slit and brings out the wave
character of electron as evidenced from the diffraction pattern observed on the screen. The
uncertainty in the electron momentum parallel to y-axis depends on the diffraction angle q.
According to the theory of diffraction at a single slit, the angle q is given by
sin q = l/d (20.33)
The uncertainty in the momentum of the electron parallel to y-axis is given by
hλ h h
Dpy = p sin θ = = ≈
λ d d ∆y
∴ ∆y ∆py ≈ h (20.34)
If we wish to determine the exact position of the electron along the y-axis, we have to
use a very narrow slit. However, a very narrow slit produces a wider diffraction pattern,
which leads to a larger uncertainty in our knowledge of the Y-component of the momentum.
Conversely, if we attempt to reduce the uncertainty in our knowledge of Y-component of
momentum, the diffraction pattern should be very Y
narrow. Therefore, we have to use a very wide slit,
which in turn results in a large uncertainty in the
y-coordinate of the electron. Thus, our efforts to
simultaneously reduce the uncertainties Dy and Dp d
will get frustrated.
2. Treating the Electron as a Particle
We now consider that electron is a particle and 2p
attempt to measure its position. For the sake of
convenience, we assume that the electron is at rest Scattered
2

P P
and use a microscope to locate it. We cannot use light 
an optical microscope for this purpose. The act of Incident
observing an object requires that light from a source light
x
be reflected by the object and enter a recording Electron
instrument such as eye. An object reflects a wave x
when the size of the object is sufficiently larger than
Fig. 20.12: Measurement of position
the wavelength of the wave. In the present case, if we and momentum of an electron by
use light waves, they would pass on without getting means of a g-microscope.
reflected by the electron, since the size of the electron
574 A Textbook of Engineering Physics

is about 105 times smaller than the wavelength of light. Therefore, we use a g-ray microscope
to detect and locate an electron.
Let a free electron be directly beneath the center of the g-ray microscope’s lens. The
circular lens forms a cone of angle 2a from the electron. The electron is illuminated from
the left by g-rays. The microscope can resolve objects to a size of Δx. Δx is given by the
expression
λ
 ) 53 . 02 ( Δx=
2sin α
To be observed by the microscope, the g-ray must be scattered into any angle within the
cone of angle 2α. A g-photon carries a very large momentum. When the g-photon strikes the
electron, part of the momentum and energy are transferred to the electron due to Compton
scattering. Consequently, as the scattered photon enters the microscope, the electron has
already moved away in a certain direction (Fig. 20.12). The total momentum p is related to
h
the wavelength by the formula p = .
λ
In the extreme case of diffraction of the gamma ray to the right edge of the lens, the
total momentum in the x direction would be the sum of the electron’s momentum p′x in the X
direction and the gamma ray’s momentum in the x-direction:
 h sin α 
px′ +  (20.36)
 λ′ 
where λ´ is the wavelength of the deflected gamma ray. In the other extreme, the observed
gamma ray recoils backward, just hitting the left edge of the lens. In this case, the total
momentum in the X-direction is:
 h sin α 
px′′ −  (20.37)
 λ′′ 
The final X-momentum in each case must equal the initial X-momentum, since momentum
is conserved. Therefore, the final X-momenta are equal to each other:
 h sin α   h sin α 
px′ +  = px′′ − 
 λ′   λ′′ 
If α is small, then the wavelengths are approximately the same, λ′ ≈ λ′′ ≈ λ. And we have
2h sin α
p″x – p′x =
λ
2h sin α
or Dpx = (20.38)
λ
Using eq.(20.35) into the above equation, we obtain
h
Dpx =
∆x
\ Dx ⋅ Dpx = h
20.16 APPLICATIONS OF UNCERTAINTY PRINCIPLE
We deal here with three simple examples to illustrate the application of uncertainty principle.
(a) Bohr’s Orbit and Energy
Let us consider the electron in a hydrogen atom. We cannot know at any instant the position
of the electron in its orbit. It might be on the left or right of the nucleus, as sketched in
Quantum Mechanics 575

Fig. 20.13. The electron position has an uncertainty ±r. We cannot know likewise whether the
electron is moving upward or downward .The uncertainty in its velocity therefore ±u. Taking
Dx = r ≈ 0.5 × 10–10 m, the uncertainty in the electron speed is
h
Du =
2πm∆x

6.62 × 10 −34 J .s
=
2 × 3.124 × 9.11 × 10−31 kg × 0.5 × 10−10 m
≅ 2 × 106 m/s
The velocity ‘u’, of an electron in an atom
is of the order of 1.0 × 106 m/s and is of the
same order as the uncertainty Du. Therefore,
we conclude that the uncertainty in momentum
is of the same order as the momentum. That ∆υ = –υ ∆υ = + υ
is Dp ≈ p. It means that sharp position and Nucleus
momentum do not exist simultaneously for the
electron in an atom. Hence it is not possible to
ascribe any specific trajectory to an electron
in an atom. It can only be said that atomic
electrons traverse the whole of the space about
the nucleus, but however, they move most ∆x = –r ∆x = r

of the time at a distance corresponding to a Fig. 20.13: The uncertainties in the position
permitted Bohr radius. and velocity of electron in an atom.
Now let us calculate the energy of the
Bohr’s first orbit. The total energy of the electron in the first orbit is given by
E = K.E. + P.E.
1  e2  p2  e2
=  mυ 2  − =  − (20.39)
2  4πε o r  2m  4πε o r
where p is the momentum of the electron.
h h
As Dp ≈ p, we can write p ≈ = . (20.40)
2π∆x 2πr

h2 e2
\ E= − (20.41)
8π 2 mr 2 4πε o r
εo h2
But r is given by r = .
π me2
me4
\ E= − (20.42)
8ε o2 h 2
The above expression (20.42) is the same as that is given by Bohr theory.
(b) Particle in a Box:
Let us consider a particle confined to a box of length l. The uncertainty Dx in the position is l.
Dx ⋅ Dp ≅ ħ
 
∴ Dp = = (20.43)
∆x l
576 A Textbook of Engineering Physics

p 2 ( ∆p ) (  / l )2
2
2
Energy is given by E= ≈ = = (20.44)
2m 2m 2m 2ml 2
This result agrees with the result obtained from Schrödinger equation. Refer to § 20.95.
(c) Electrons cannot be present in the nucleus:
The radiation emitted by radioactive nuclei consists of a, b and g-rays, out of which b-rays
are identified to be electrons. We apply uncertainty principle to find whether electrons are
coming out of the nucleus. The radius of the nucleus is of the order of 10–14 m. Therefore,
if electrons were to be in the nucleus, the maximum uncertainty Dx in the position of the
electron is equal to the diameter of the nucleus. Thus,
∆x = 2 × 10–14 m.
The minimum uncertainty in its momentum is then given by
 1.04 × 10−34 J .s
∆p = = = 5.2 × 10–21 kg-m/s.
∆x 2 × 10−14 m
The minimum uncertainty in momentum can be taken as the momentum of the electron.
Thus,
p = 5.2 × 10–21 kg-m/s.
The minimum energy of the electron in the nucleus is then given by
Emin = pmin c = (5.2 × 10–21 kg-m/s)(3 × 108 m/s) = 1.56 × 10–12 J = 9.7 MeV.
It implies that if an electron exists within the nucleus, it must have a minimum energy
of about 10 MeV. But the experimental measurements showed that the maximum kinetic
energies of b-particles were of the order of 4 MeV only. Hence electrons are not present in the
nucleus. It is subsequently established that emission of b-particles occurs due to transforma-
tions in the nucleus. The transformation of a neutron into a proton produces an electron.
Example 20.4: Uncertainty in time of an excited atom is about 10–8s. What are the
uncertainties in energy and in frequency of the radiation?
h
Solution: DE Dt ≈
2π

1.054 × 10−34 J .s 1.054 × 10−26

\ DE = = eV = 6.58 × 10−8 eV .
10−8 s 1.602 × 10 −19
∆E 1.054 × 10−26 J
Dn = = = 15.9 MHz.
h 6.626 × 10−34 J .s
Example 20.5. An electron is confined to a potential well of width 10 nm. Calculate the
minimum uncertainty in its velocity.
h h
Solution. ∆x ∆p ≈ or ∆x ∆p ≈
2π 2π
h
\ ∆u =
2π m ⋅ ∆x

6.63 × 10−34 Js
\ ∆u = = 12.1 km/s.
2 × 3.143 × 9.11 × 10−31 kg × 10 × 10 −9 m
Quantum Mechanics 577

Example 20.6: If the kinetic energy of an electron known to be about 1 eV, must be measured
to within 0.0001 eV, what accuracy can its position be measured simultaneously?
p2 2 p ∆p m
Solution: E=   \ ∆E =   \ ∆p = ∆E
2m 2m p
h
∆x∆p =
2π
h h p h 2mE h E
\ ∆x = = ⋅ = =
2π ⋅ ∆p 2π m∆E 2π m∆E π ∆E 2m

6.63 × 10−34 Js 1.602 × 10−19 J

\ ∆x = ⋅
3.143 × 0.0001 × 1.602 × 10−19 J 2 × 9.11 × 10−31 kg

= 1.95 mm.
Example 20.7: An electron and a 150 gm base ball are traveling at a velocity of 220 m/s,
measured to an accuracy of 0.005 %. Calculate and compare uncertainty in position of each.
0.065
Solution: The uncertainty in the velocity is ∆u = u × 0.065% = ( 220m / s ) × = 0.143 m/s.
100
(i) The uncertainty in the position of electron is
 1.05 × 10 −34 J .s
Dxe = = = 0.4 mm.
2 m∆υ 2 × 9.11 × 10−31 kg × 0.143 m / s
(ii) The uncertainty in the position of baseball is
 1.05 × 10 −34 J .s
DxB = = = 2.5 ×10–33m.
2 M ∆υ 2 × 0.15 kg × 0.143 m / s

20.17 WAVE FUNCTION AND PROBABILITY INTERPRETATION

Waves represent the propagation of a disturbance in a medium. We are familiar with light
waves, sound waves, and water waves. These waves are characterized by some quantity that
varies with position and time. Light waves consist of variations of electric and magnetic fields
in space, and sound waves consist of pressure variations. We cannot specify in a similar way
what is actually varying in de Broglie waves. Since microparticles exhibit wave properties, it
is assumed that a quantity y represents a de Broglie wave. This quantity y is called a wave
function. y describes the wave as a function of position and time. However, it has no direct
physical significance, as it is not an observable quantity. In general, y is a complex-valued
function. According to Heisenberg uncertainty principle, we can only know the probable
value in a measurement. The probability cannot be negative. Hence y cannot be a measure
of the presence of the particle at the location (x, y, z). But it is certain that it is in someway an
index of the presence of the particle at around (x, y, z, t).
Probability Interpretation of Wave Function given by Max Born
A probability interpretation of the wave function was given by Max Born in 1926. He
suggested that the square of the magnitude of the wave function |y|2 evaluated in a particular
region represents the probability of finding the particle in that region. In other words,
578 A Textbook of Engineering Physics

Probability, P, of finding the particle in an infinitesimal volume dV (= dx dy dz) is

proportional to |y(x, y, z)|2 dx dy dz at time t.
or P ∝ |y(x,y,z)|2 dV (20.45)
2
|y| is called the probability density and y is the probability amplitude.
Since the particle is certainly somewhere in the space, the probability P = 1 and the
integral of |y|2dV over the entire space must be equal to unity. That is
+∞
2
∫ ψ dV = 1 (20.46)
−∞
The wave function y is in general a complex function. But the probability must be real.
Therefore to make probability a real quantity, y is to be multiplied by its complex conjugate
y*. +∞
∫ ψψ * dV =1
−∞
Thus, y has no physical significance but |y|2 gives the probability of finding the atomic
particle in a particular region.
20.17.1 Normalization Condition
If at all the particle exists, it must be found somewhere in the universe. Since we are sure that
the particle must be somewhere in space, the sum of the probabilities over all values of x,y,z
must be unity. If Y(x, y, z, t) is multiplied by a constant C such that YN (x,y,z,t) = C Y(x,y,z,t),
where YN (x, y, z, t) satisfies the relation
∞ ∞
2 2
∫ Ψ N ( x, y, z , t ) dxdydz = C Ψ ( x, y,z,t ) dxdydz = 1
2
∫ (20.47)
−∞ −∞

then YN (x,y,z,t) is said to be normalized wave function and C the normalization constant.
This condition (20.47) is known as the normalization condition. From equ.(20.47), we have
1
|C|2 = ∞ (20.48)
2
∫ Ψ ( x,y,z ,t ) dxdydz
−∞
|yN(x, y, z,t)|2 dxdydzis called the probability density.
Whenever wave functions are normalised, |y|2dV equals the probability that a particle
will be found in an elemental volume dV.
Thus,
Probability P = |y(x, y, z)|2dV (20.49)
20.17.2 Well-behaved Wave Functions - Conditions to be satisfied by
y-function
An acceptable wave function y must be normalized and fulfill the following requirements:
(i) y function must be finite: The wave function must be finite everywhere. Even if x →
∞ or – ∞, y → ∞ or – ∞, z → ∞ or – ∞ , the wave function should not tend to infinity.
It must remain finite for all values of x, y, z. If y is infinite, it would imply an infinitely
large probability of finding the particle at that point. This would violate the uncertainty
principle.
Quantum Mechanics 579

(ii) y function must be single–valued: Any physical

quantity can have only one value at a point. For
this reason, the function related to a physical
quantity cannot have more than one value at that
point. If it has more than one value at a point,
it means that there is more than one value of
probability of finding the particle at that point
(Fig. 20.14).
(iii) y function must be continuous: y function
should be continuous across any boundary. Since
y is related to a physical quantity, it cannot have
a discontinuity at any point. Therefore, the wave
∂ψ ∂ψ
function y and its space derivatives ,
∂x ∂y
∂ψ
and should be continuous across any
∂z Fig. 20.14
boundary. Since y is related to a real particle, it
cannot have a discontinuity at any boundary where potential changes.
Wave functions satisfying the above mathematical conditions are called well-behaved
wave functions.
20.18 SCHRÖDINGER WAVE EQUATION
In 1926, Erwin Schrödinger, an Austrian physicist, reasoned that the de Broglie waves
associated with electrons would resemble the classical waves of light and developed wave
equation that describes the behaviour of matter waves. This equation, which has been named
after him, defines the wave properties of electrons and also predicts particle-like behaviour.
Time dependent Schrödinger Wave Equation
Since the concept of de Broglie waves is not a result of previous physical theories, it is not
possible to derive a wave equation for the wave associated with a particle. However, a wave
equation can be developed. Let us assume that a particle of mass ‘m’ is in motion along the
x-direction. Let the wave function y be the dependent variable of the de Broglie wave which
is a function of the coordinates x and t. Analogous to the classical wave, we may expect that,
y will be a function of (x – υt). As υ = ω / k, the wave function may be written as a function
of (kx – ωt). Using the relation p = ħk and E = ħω, we can write
 px − Et 
y= f (20.50)
  
The more general wave would be a sum of a sine and cosine waves. Taking help of
Euler’s identity, we write the above equation in an exponential form as follows:
 In 
y = A exp  ( px − Et ) (20.51)
 
We assume that the energy and momentum of the particle are constant. Differentiating the
above equation with respect to x, we get
580 A Textbook of Engineering Physics

∂ψ ip  In 
= A exp  ( px − Et )
∂x   
∂ψ ip
\ = ψ
∂x 
Rearranging the terms in the above equation, we get
 ∂ψ ∂ψ
py = = −i (20.52)
i ∂x ∂x
Differentiating the equ.(20.51) with respect to t gives
∂ψ iE  In  iE
= − A exp  ( px − Et ) = − ψ
∂t     
Rearranging the terms in the above equation, we get
 ∂ψ ∂ψ
Ey = − = i (20.53)
i ∂t ∂t
The partial derivatives with respect to x and t are connected by means of the relation
between the energy and momentum. The classical expression for the kinetic energy in terms
of the momentum is
m υ 2 ( m υ) 2 p 2
Ek = = = (20.54)
2 2m 2m
The total energy and momentum are related by the expression
p2
+V = E (20.55)
2m
where V is the potential energy of the particle. Multiplying the equ.(20.55) with ψ, we obtain
p2
ψ + V ψ = Ey
2m
Using the relations (20.52) and (20.53) into the above equation, we get
2 ∂2ψ ∂ψ
− + V ψ = i
2 m ∂x 2 ∂t
The above equation may be rewritten as
∂Ψ  2 ∂2 
i = − 2
+V  Ψ (20.56)
∂t  2 m ∂x 
The above equation is known as the time-dependent Schrödinger wave equation.
When extended to the three-dimensional case, we find that
2  ∂2ψ ∂2ψ ∂2ψ  ∂ψ
−  2
+ 2 + 2  + V ψ = i
2 m  ∂x ∂y ∂z  ∂t

2  ∂2 ∂2 ∂2  ∂ψ
−  2
+ 2
+ 2
ψ + V ψ = i
2 m  ∂x ∂y ∂z  ∂t

2 2 ∂ψ
or − ∇ ψ + V ψ = i
2m ∂t
Quantum Mechanics 581

∂2 ∂2 ∂2
where ∇2 = 2
+ 2+ 2.
∂x ∂y ∂z
Therefore, the Schrödinger equation for three dimensional motion may be written as
∂  2 2 
i Ψ ( r , t ) =  − ∇ + V ( r ) Ψ ( r , t ) (20.57)
∂t  2m 
Time independent Schrödinger Wave Equation
Knowing the form of V, equn.(20.56) can be solved for the wave function y. In a number
of cases the potential energy V of a particle does not depend on time; it varies with the
position of the particle only and the field is said to be stationary. In the stationary problems
Schrödinger equation can be simplified by separating out time and position-dependent parts.
Accordingly, we can write the wave function as a product of x, y(x) and a function of t, j(t).
We, therefore, write that
y(x,t) = y(x)j(t) (20.58)
Equation (20.56) may be written as
2 ∂2 ∂
− 2
(ψ , φ) + V ψφ = i (ψ , φ)
2 m ∂x ∂t
2 ∂2 ∂
− φ (ψ , φ) + V ψφ = i (ψ , φ)
2 m ∂x 2 ∂t
Dividing the above equation with yf, we get
2 1 d 2ψ 1 dφ
− + V = i (20.59)
2m ψ dx 2 φ dt
If we assume that the potential energy V is a function of x only, the entire left hand side
of equ.(20.59) is a function of x only while the right hand side is a function of t only. Since x
and t are independent variables, both the function of x and t must be equal to a constant. The
constant that each side must equal is called the separation constant E. Thus,
2 1 d 2ψ
− +V = E (20.60)
2m ψ dx 2
1 dφ
and i =E
φ dt
Eq.(20.60) may be rewritten as
2 d 2ψ
− + V ψ = Ey (20.61)
2m dx 2
The above equation involves only the space coordinates and is called the time-independent
Schrödinger wave equation.
Eq. (20.61) may be rewritten as
d 2 ψ 2m
+ ( E − V )ψ = 0
dx 2  2
d 2 ψ 8π 2 m
\ + 2 ( E − V )ψ = 0 (20.61a)
dx 2 h
582 A Textbook of Engineering Physics

In three dimensions, equation (20.61) may be written as

2 2
− ∇ ψ +Vψ = 0
2m

20.18.1 Allowed Wave Functions and Energies

The time-independent wave equation (20.61) is the pertinent equation for studying properties
of atomic systems in stationary conditions. Wave mechanical methods of solving the problem
of particle motion are essentially based on ψ functions. Appropriate wave equation is
formulated by incorporating the particle mass ‘m’ and potential energy function V for the
region in which the particle is located. The next step consists of solving the differential
equation for solutions, namely for the ψ functions which will satisfy the differential equation.
By solving the Schrödinger equation, we obtain the possible set of ψ functions. In case of
bound particles the acceptable solutions for the differential equation are possible only for
certain specified values of energy. These energy values will be the only possible results of
precise measurements of the total energy of the particle. These discrete values of energy E1,
E2,…..En are called eigenvalues or allowed values of the energy of the particle. The solutions
ψ1, ψ2, ……ψn corresponding to the eigen energy values En are called the eigen functions.
The quantization of energy thus appears as a natural element of the wave equation.
In Schrödinger’s theory, the quantization of energy values follows directly from the
mathematical formulation of the wave and particle nature of the electron. The existence of the
stationary states has not been assumed and no assumptions have been made about orbits. The
new theory yields all the results of the Bohr’s theory without having any of the inconsistent
hypotheses of the earlier theory. The new theory also accounts for the experimental information
for which the Bohr’s theory failed to account, such as the probability of an electron changing
from one state to another.
We now apply the Schrödinger wave equation to different cases of particles confined to
move in different enclosures and obtain information regarding the energy values it can take
and its probability density etc.

20.19 THE FREE PARTICLE

A particle is said to be a free particle when it is moving in space without being subjected to
any external force in any region of space, and its potential energy is constant [V = constant]
everywhere. We study here the one dimensional motion of a particle under no forces.
Let us assume that the particle is moving in space along the positive x-direction. As
the particle is not acted upon by a force, the potential energy of the particle is constant.
For convenience, we take the constant potential to be zero. Then, the time-independent
Schrödinger wave equation for the free particle is written as
d 2 ψ 8π 2 m
+ 2 Eψ = 0 (20.62)
dx 2 h
As the particle is moving freely with zero potential energy, its total energy will be kinetic
energy which is given by
p2
E= x (20.63)
2m
where px is the momentum of the particle along the x-direction.
We may rewrite the equation (20.62) as
Quantum Mechanics 583

d 2ψ
2
+ k 2ψ = 0 (20.64)
dx
8π 2 mE
where = k2 (20.65)
h2
(a) Wave function
The general solution of the equation (20.65) is of the form
y(x) = Aeikx +Be–ikx (20.66)
where A and B are constants of integration. Eq.(20.66) gives the time-independent part of the
wave function. The complete wave function is given by
y(x, t) = y(x)e–iwt
= (Aeikx + Be–ikx) e–iwt
= Ae–i(wt–kx) + Be–i(wt+kx) (20.67)
Equ.(20.67) represents a continuous plane simple harmonic wave. The first term represents
the wave traveling along the positive x-direction while the second term represents the wave
traveling along the negative x-direction. The particle traveling in the positive x-direction is
represented by
y(x,t) = Ae–i(wt–kx) (20.68)
(b) Energy:
It is seen from eq. (20.65) that k has the following value.
8π 2 mE 2mE
k= 2
= (20.69)
h 2
Therefore, the particle energy is given by
2k 2
E= (20.70)
2m
There are no boundary conditions to be applied to the particle motion and hence there are
no restrictions placed on k. It follows from eq.(20.70) that the particle is permitted to have
any value of energy. In other words, the energy is not quantized. It means that a freely moving
particle possesses a continuous energy spectrum as shown in Fig. 20.15.

Fig. 20.15 Fig. 20.16

The k-vector describes the wave properties of the particle. It may be seen from equ.
(20.70) that E ∝ k2. The plot of E as a function of k gives a parabola, as illustrated in Fig.
20.16.
584 A Textbook of Engineering Physics

(c) Position of the Particle

The probability of finding the particle between x and x + dx is given by
P dx = y*(x,t)y(x,t)dx (20.71)
or 2
P dx = A dx (20.72)
The probability density P for the position of the particle with the definite value of
momentum is constant over the x-axis. It means that the particle is not located at a definite
point but all positions are equally probable. This conclusion is in agreement with the
Heisenberg uncertainty principle. As the momentum is well defined in this case, it is difficult
to assign a position to the particle. The uncertainty in position will be infinity which means
that the particle position is indeterminate.

20.20 POTENTIAL ENERGY STEP

A particle moving in a region of constant
potential energy region may suddenly encounter
a region of higher potential energy, which is
also constant. Thus, the potential energy step is
a square potential step as shown in Fig. 20.17.
We will not discuss in detail the solutions but
only give the outline of the salient steps and
conclusions. Fig. 20.17
(a) Potential energy step, E > Vo:
Let E be the fixed total energy of the particle and V be the value of the constant potential
energy.
It is seen from Fig. 20.17 that V = 0 for x < 0
and V = Vo for x > 0
(i) In the region x < 0 the time independent Schrödinger equation is given by
d 2 ψ 8π 2 m
+ 2 Eψ = 0
dx 2 h
d 2ψ
or + ko2 ψ = 0 (20.73)
dx 2
8π 2 mE
where = k2o.
h2
(ii) In the region x > 0 the time independent Schrödinger equation is given by
d 2 ψ 8π 2 m
+ 2 ( E − Vo )ψ = 0
dx 2 h
d 2ψ
or + k 2ψ = 0 (20.74)
dx 2
8π 2 m
where k2 = ( E − Vo ) .
h2
The solutions of equations (20.73) and (20.74) are
y(x) = Aeikox+Be–ikox for x < 0 (20.75 a)
Quantum Mechanics 585

y(x) = Ceikx +De–ikx for x > 0 (20.75 b)

where A, B, C and D are constants of integration. The Schrödinger equation changes discontin-
uously at x = 0, as there is a discontinuity in V at x = 0. Relationships among the four
coefficients A, B, C and D may be found by applying the condition that y(x) and its derivative
dψ
must be continuous at the boundary. Thus,
dx
when x = 0, y(x) = A + B and y(x) = C + D
\ A+B = C+D (20.76 a)
dψ
At x = 0, = ko(A – B)
dx
dψ
and = k(C – D)
dx
\ ko (A – B) = k (C – D)
 k
or (A – B) =   (C − D ) (20.76 b)
 ko 
Solving the equations (20.76 a) and (20.76 b), we get
k − k   2k 
B=  o  A+  D (20.77 a)
 ko + k   ko + k 
 2 ko   ko − k 
and C=  A− D (20.77 b)
 ko + k   ko + k 
Substituting the value of B in equ.(20.75 a), we get
 k − k   2k   −iko x
y(x) = Aeiko x +  o  A+  D e
  ko + k   ko + k  
  k − k  −iko x   2k  −ik x 
y(x) = A eiko x +  o  e  + D  e
o

  ko + k     ko + k  

or y(x) = Ay1 + Dy2 (20.78 a)

Substituting the value for C in equn.(20.75 b), we get
 2ko  ikx    k − k  ikx 
y(x) =   e  A + e −ik x −  o e D
  ko + k     ko + k  
or y(x) = Ay1 + Dy2 (20.78 b)
The equations (20.78 a) and (20.78 b) are the same and conditions of continuity at x = 0
are satisfied by both y1 and y2.
The complete solution of equn.(20.73) is obtained when the equn.(20.75 a) is multiplied
by the time dependent term e–iwt. Thus,
y(x, t) = Ae (
− i ωt − ko x )
+ Be (
− i ωt + ko x )
(20.79)
The first term in the above equation represents a wave moving along the positive
x-direction while the second term represents a wave moving along the negative x-direction.
586 A Textbook of Engineering Physics

Therefore, equ.(20.79) describes the superposition of a wave of intensity |A|2 moving along
the positive x-direction (from –∞ to zero) and a wave of intensity |B|2 moving along the
negative x-direction. Therefore, when the particles are incident on the step from left on the
potential energy step, |A|2 gives the intensity of the incident wave and |B|2 gives that of the
2
B
refelcted wave. The ratio 2
gives us the reflected fraction of the incident wave intensity.
A

(b) Potential energy step, E < Vo:

Whenever E becomes less than V, a different solution results.
y(x) = Aekx + Be − kx (20.80 a)
If E were less than Vo, then the solution for yo(for x < 0) would still be given by equn.
(20.75 a), but the solution for y (for x > 0) becomes
y1(x) = Cek1x + De − k1x (20.80 b)

8π 2 m
where k1 = (Vo − E ) .
h2
We set, C = 0 to keep y1(x) from becoming infinite as x → ∞, and apply the boundary
conditions on y1(x) at x = 0. The resulting solution is shown in Fig. 20.18 (b).
This illustrates an important difference
between classical and quantum mechanics.
Classically, the particle can never be found in the
region x > 0, since its total energy is not sufficient
to overcome the potential energy step. However,
quantum mechanics allows the wave function,
and therefore the particle, to penetrate into the
classically forbidden region. The particle can
never be observed in the forbidden region, but a
particle can pass through a classically forbidden Fig. 20.18
region and emerge into an allowed region where it can be observed.
Penetration into the forbidden region is associated with the wave nature of the particle.
The penetration distance is consistent with the uncertainty in defining the location of the
particle.
20.21 RECTANGULAR POTENTIAL BARRIER
According to classical ideas, a particle striking
a hard wall has no chance of leaking through
it. But the behaviour of a quantum particle is
different owing to the wave nature associated
with it. We know that when an electromagnetic
wave strikes at the interface of two media, it is
partly reflected and partly transmitted through Fig. 20.19
the interface and enters the second medium. In
a similar way the de Broglie wave also has a possibility of getting partly reflected from the
boundary of the potential well and partly penetrating through the barrier. The penetration of a
barrier by a quantum particle is called tunneling.
Quantum Mechanics 587

Fig. 20.19 represents a potential barrier of height Vo and thickness L. The region around
the barrier can be divided into three regions as shown in Fig. 20.20. The potential energy is
zero in regions I and III, i.e., for x < 0 and x > L and has a constant value Vo in the region II,
i.e., for 0 < x < L. Let us consider a particle of total energy E approaching the barrier from the
left. In regions I and III the total energy of the particle is kinetic energy. In region II, the total
energy is partly kinetic and partly potential energy. From the view-point of classical physics,
the electron would be reflected from the barrier because its energy E is less than V. For the
particle to overcome the potential barrier, it must have energy equal to or greater than V.
Quantum mechanics leads to an entirely new result.
If E < V, according to quantum mechanics (20.20), there is a finite chance for the electron
to leak to the other side of the barrier. We say that the electron tunneled through the potential
barrier and hence in quantum mechanics, the phenomenon is called tunneling.
We write down the Schrödinger wave equation for the electron wave in the three regions
and solve them. The Schrödinger wave equation for a particle constrained to move along
x-axis is given by
d 2 ψ 8π 2 m
+ 2 ( E − V )ψ = 0 (20.81)
dx 2 h
In the region I, V = 0 and the equation (20.81) takes the form
d 2 ψ1 8π 2 mE
+ ψ1 = 0
dx 2 h2
d 2 ψ1
or + ko2 ψ1 = 0 (20.82)
dx 2
\ In the region II, E < Vo, V = Vo and the equation (20.81) takes the form
d 2ψ 8π 2 m
2 2 + 2 ( E − Vo )ψ 2 = 0
dx h
d 2ψ2 8π 2 m
or − (Vo − E )ψ 2 = 0 (20.83)
dx 2 h2
d 2ψ 2
or 2
− k 2ψ 2 = 0 (20.84)
dx

8π 2 m
where k= (Vo − E ) .
h2
In the region III, V = 0. Therefore,
d 2ψ 3
+ ko2 ψ 3 = 0
dx 2
The solutions of equations (20.82), (20.83), and (20.84) are
y1 = Aeiko x + Be −iko x (20.85 a)
kx −k x
y2 = Ce + De (20.85 b)
iko x − iko x
y3 = Le + Me (20.85 c)
In the region III, no particle comes from right and hence M = 0.
588 A Textbook of Engineering Physics

\ y3 = Leiko x (20.85 d)
In eq. (20.85 a), Aeikoxrepresents the de Broglie wave traveling along the x-direction
in the region I, with amplitude A and Beikox represents the wave reflected along the negative
x-direction with amplitude B. Since the probability that the particle is present in a region of
the space is proportional to the square of the de Broglie wave amplitude, the ratio
2
B
R= 2
(20.86)
A
is the coefficient of reflection of the particle from the barrier.
The wave function Y2 is not zero inside the barrier (the region forbidden by classical
mechanics), but decreases exponentially. De–kx represents the exponentially decreasing wave
in the barrier. The square of the amplitude of this wave defines the probability of penetration
of the particle into the region II. The ratio is the coefficient of penetration of the interface.
Cekx is the reflected wave within the barrier. Since such a wave is nonexistent, C should be
equal to zero.
2
D
T= 2
(20.87)
A
In eq.(20.85 d) Leikox represents the transmitted wave moving along the x-direction in the
region III.
The form of the wave function in the
region (I), (II) and (III) is also shown in the
Fig. 20.20. The wave function Y1
corresponds to the free electron with
momentum p = 2mE . Since Y3 is not
equal to zero at x = L, there is a finite
probability of finding the electron in the
region III. That means the particle that is
initially to the left of the barrier has some Fig. 20.20
probability of being found to the right of the barrier. The wave function Y3 represents the
wave transmitted through the barrier and the free particle on the right side of the barrier. The
particle has the same momentum as the incident particle but has smaller amplitude. Thus, it is
possible for a particle to penetrate through the potential barrier even if its kinetic energy is
less than the height of the potential barrier.
This probability is proportional to the square of the modulus of the wave function Y2.
 2x 
2m (Vo − E ) 
2
P ∝ |y2|2 = C exp  − (20.88)
  
This probability is indicative of the fact that the particle is able to penetrate the potential
barrier of finite width L. Such a penetration is called the tunnel effect. The probability that
the particle gets through the barrier is called the transmission coefficient. It is defined as
Probability density of the transmitted wave
T=
Probability density of the incident wave
It is shown that the transmission coefficient is given approximately by
16 E  E   −2 L 8 π 2m(V − E ) / h
T = Ge −2 kL =  1 −   e (20.89)
 V  V 
Quantum Mechanics 589

where G is a constant close to unity. Equ.(20.89) shows that the probability of particle
penetration through a potential barrier depends on the height, V and width, L of the barrier.
The probability decreases rapidly with increasing barrier width, L.
Tunneling is significant in many areas of physics.
20.21.1 Application of Tunnelling
1. The tunnel diode is a semiconductor diode. The current in this device is largely due to
tunneling of electrons through a potential barrier. The rate of tunneling or current can
be controlled over a wide range by varying the height of the barrier, which is done by
varying the applied voltage. Potential
2. a-decay: The quantum possibility of energy

penetrating potential barriers is the basis

of the explanation of the a-decay of
radioactive nuclei. In a-decay, an unstable
nucleus disintegrates into a lighter
nucleus and an a-particle. For example, a
uranium nucleus 238U undergoes a-decay x
and forms thorium nucleus 234Th.
In 1928 George Gamow explained
a-decay of unstable nuclei on the basis
of quantum tunneling. The forces in the
nucleus set up a potential barrier of height
of the order of 30 MeV against a-particle
emission. Classically, the a-particle
would be trapped unless its energy
exceeds 30 MeV. The a-particles have Fig. 20.21: Alpha particle having less energy
than the height of the potential barrier
energies in the range of 4 to 9 MeV only. tunnels through the barrier
Therefore, it is impossible for a a-particle
to jump the barrier. According to quantum mechanics, the a-particle tunnels through the
potential barrier (Fig. 20.21). The experimental studies of Geiger-Nuttal on a- decay
confirmed the predictions of Gamow’s theory.
3. Scanning tunnelling electron microscope (STM): The instrument was invented in
the early 1979 by Gerd Binnig and Heinrich Rohrer, who were awarded the 1986
Nobel prize in physics for their work. The scanning tunnelling electron microscope
uses electron tunneling to produce images of surfaces down to the scale of individual
atoms. If two conducting samples are brought in close proximity, with a small but finite
distance between them, electrons from one sample flow into the other if the distance is
of the order of the spread of the electronic wave into space. One says that the electrons
“tunnel” through the barrier into the adjacent sample. For electrons, the barrier width
which may be overcome via a tunneling process is of the order of nm, i.e. of the order
of several atomic spacings. The probability of an electron to get through the tunneling
barrier decreases exponentially with the barrier width, i.e. the so called tunneling
current is extremely sensitive measure of the distance between two conducting samples.
The STM makes use of this sensitivity.
Working: The schematic diagram of a scanning tunneling electron microscope is shown
in Fig. 20.22 (a).In the scanning tunneling microscope the sample is scanned by a very fine
metallic tip. The tip is mechanically connected to the scanner, an XYZ positioning device. The
590 A Textbook of Engineering Physics

Control voltages for piezotube

Piezoelectric tube
with electrodes
Tunneling Distance control
current amplifier and scanning unit

Tip

Sample

Tunneling
voltage
Data processing
and display

(a)
z piezo
o
ez
pi
y

x piezo

(b) (c)

Fig. 20.22: (a) The schematic diagram of STM (b) The tunneling tip is scanned over the specimen,
producing an image of the tunneling current (c) An image of a silicon crystal surface produced by
a STM.

sharp metal needle is brought close to the surface to be imaged. The distance is of the order of
a few angstroms. A bias voltage is applied between the sample and the tip. When the needle
is at a positive potential with respect to the surface, electrons can tunnel through the across
the gap and set up a small “tunneling current” in the needle. This feeble tunneling current
is amplified and measured. With the help of the tunneling current the feedback electronics
keeps the distance between tip and sample constant. The sensitivity of the STM is so large
that electronic corrugation of surface atoms and the electron distribution around them can be
detected. Fig. 20.22 (b) shows the tip of the sharp needle and Fig. 20.22 (c) shows the image
of silicon crystal surface.
Example 20.8. A stream of electrons, each of energy E = 3eV, is incident on a potential barrier
of height V = 4eV. The width of the barrier is 20 Å. Calculate the percentage transmission of
the beam through this barrier.
Solution. The probability of transmission through a potential barrier is given by
Quantum Mechanics 591

16 E  E   −2 L 8 π 2m(V − E ) / h
T = Ge –2kl =  1 −   e
 V  V 
2L 2 × 20 × 10−10 m
8π 2 m (V − E ) = 8 × (3.143) × 9.11 × 10 −31 kg ( 4 − 3) × 1.602 × 10 −19 J
2
−34
h 6.63 × 10 Js
= 20.49
16 E  E   −20.49
T= 
\ 1 −   e
 V  V 
16 × 3eV  3eV  −20.49
= 1 −  e
4eV  4eV 
= 3 × e–20.49 = 3.8×10–9
Therefore, the percentage of transmission = 3.8×10–7%.
 
20.22 INFINITE POTENTIAL WELL
A potential well is a potential energy function V(x) V
V=0
that has a minimum (see Fig. 20.23). A potential well
is the opposite of a potential barrier; it is a potential-  0
energy function with a minimum. If a particle is left
in the well and the total energy of the particle is less
=0 =0
than the height of the potential well, we say that the
particle is trapped in the well. In classical mechanics a
particle trapped in a potential well can vibrate back and
forth with periodic motion but cannot leave the well.
In quantum mechanics, such a trapped state is called a
bound state. 0 x L
Let us consider a particle confined to the region 0 Fig.20.23
< x < L. It can move freely within the region 0 < x < L
but subject to strong forces at x = 0 and x = L. Therefore, it can never cross to the right to the
region x > L or to the left of 0. It means that V = 0 in the region 0 < x < L and rises to infinity
(V = ∞) at x = 0 and x = L. This situation is called a one-dimensional potential box.
For a particle trapped in a one-dimensional potential box, V = 0 and the Schrodinger
equation (20.61a) takes the following form
d 2 ψ 8π 2 mE
+ ψ =0 (20.90)
dx 2 h2
8π 2 mE
Putting = k2, we rewrite the above equation as
h2
d 2ψ
2
+ k 2ψ = 0 (20.91)
dx
Because the particle can move back and forth freely between x = 0 and x = L, the solution
of the equation (20.91) is of the form
y(x) = Aeikx + Be −ikx (20.92)
592 A Textbook of Engineering Physics

which contains motion in both directions. We can evaluate the constants A and B with the
help of boundary conditions. The boundary conditions are as follows. The particle cannot
jump over the walls and therefore the function does not exist outside the box. The particle is
located within the box and therefore it exists within the box. It means that
y(x) = 0 at x = 0
y(x) = 0 at x = L
\ yx=o = A+B = 0
\ B = –A
Using this result into equ.(20.92), we get
y(x) = A(eikx − e −ikx )
or y(x) = 2iA sin kx
Again y(x) = 0 at x = L
\ y x=L = 2iA sin kL = 0
The factor 2iA cannot be zero, because we would not then have the wave function. It
means that
sin kL = 0
or kL = np (20.93)
As k = 2p/n, l = 2L/n
where n (=1,2,3, ....) is an integer.
The above conclusion implies that the wave equation has solutions only when the particle
wavelength is restricted to discrete values such that only a whole number of half-wavelengths
are formed over the length L of the box. It means that particle waves form standing wave
pattern within the potential box. We write (20.93) as
k = np / L
(i) The possible values of the momentum of the particle are then given by
nπ nh
p = k = = (20.94)
L 2L
(ii) The possible values of the energy are
p 2 n2h2
E= = (20.95)
2m 8mL2
The above equation indicates that a particle confined in a certain region can have only
certain values of energy. Other energy values are not allowed. In other words, energy
quantization is a consequence of restricting a microparticle to a certain region.
The allowed wave functions which are the solutions of the Schrodinger equation are
given by
yn = 2iA sin kx = C sin kx
As k = np / L, we write the above as
 nπx 
yn = C sin  (20.96)
 L 
Applying normalization condition, we can determine the value of the constant C in the
above equation. The normalisation condition is that
Quantum Mechanics 593

∫ ψ n ψ n dx
*
=1
0
L
nπ x
\ C 2 ∫ sin 2 dx = 1
0
L
2 L
C  2nπ x 

2 ∫ 1 − cos L 
dx = 1
0
L
 2n π x 
C 2 sin
L 
 x − 2nπ  = 1
2  
 L 0

L
\ C2 =1 Fig. 20.24
2
2
i.e., C2 =
L
or C= 2 L (20.97)
Using the value of C into eq.(20.96), we obtain the wave function as
2  nπx 
yn = sin  (20.98)
L  L 
Energy Levels:
The allowed energy states are given by
h2 2
En = n
8mL2
where n is the quantum number given by n = 1,2,3, ....
The energy states are obtained by inserting the value of n in the above equation. Thus
h2
First energy level E1 =
8mL2
h2
Second energy level E2 =
2mL2
9h 2
Third energy level E3 = and so on.
8mL2
The energy levels are shown in Fig. 20.24.
It is important to note that the particle cannot have zero energy. The lowest energy
allowed for the particle is E1 = h2/8 mL2. It cannot possess energy less than this in the one–
dimensional potential box.
The energy value E1 is called zero–point energy. The zero point energy is a consequence
of the uncertainty principle. If the energy of the particle is zero, its momentum also would be
zero, and the uncertainty principle requires that the wavelength l be infinite. In such case the
particle cannot be confined to the box. Therefore, the particle must have a certain minimum
amount of kinetic energy.
594 A Textbook of Engineering Physics

Comparison with Classical Results:

1. Classically, a particle in a box could have any positive energy. Quantum theory shows
that the particle is restricted to take only certain discrete values.
2. In classical theory, a particle in a box can have energy of zero value. The quantum
theory shows that a particle in a box cannot have zero energy. The minimum energy
possessed by the particle is E1 = h2/8 mL2.
Probability of locating the particle over an interval x and x + dx
The probability P(x)dx of finding the particle over a infinitesimal distance dx at position x in
the potential well is given by
2 2  nπx 
P(x)dx = ∫ ψ n dx = sin 2  dx (20.99)
L  L 
The probability of finding the microparticle between the positions x1 and x2 is
x
2 2 2  nπx 
L x∫
P= sin  dx
 L 
1

x2
1  2nπx 
= ∫
Lx
1 − cos
L 
 dx
1

x
1 x 1  2nπx  2
= [ x ]x2 − sin (20.100)
L 1 2π  L  x1

The probability density is given by

2 2  nπx 
P(x) = sin  dx (20.101)
L  L 

Fig. 20.25
nπx π 3π 5π L 3L 5 L
The probability density is a maximum when = , , ,..... or x = , , ,..
L 2 2 2 2n 2n 2n
The variation of the probability densities with x for n = 1, n = 2, and n = 3 are shown in
Fig. 20.25.
Quantum Mechanics 595

It is easy to see that in the state y1 the probability is seen to be largest at x = L / 2 ,i.e.,
in the middle of the box (Fig. 20.25a) and it decreases towards the walls. It means that the
particle will stay more in the centre of the box and avoids the region near walls. Similarly, in
the state y2 the most probable positions are at x = L/4 and x = 3L/4. Therefore, the particle
will be found either in the right–half or in the left–half of the box but never in the middle. In
classical theory, all positions within the box are equally probable for the particle. Classically,
the particle passes back and forth between the walls and it has an equal probability of being
found any where between x = 0 and x = L. The classical probability function is a constant
with a value of 1/L. The quantum results show peaks of magnitude 2/L and valleys where
the probability is very small. The number of peaks is equal to the quantum number n. As n
increases, the number of peaks increase and when n is very large, the distribution approaches
the classical distribution.
Dependence of quantization on the width of the box:
Equn.(20.95) suggests that the quantization of energy is dependent on the width of the
potential box, L and on the quantum number n. Let us now study the effect of L on the
quantization of particle energy. If we designate two adjacent energy levels by En+1 and En, the
separation between these levels is given by
∆E = En+1 –En
h2 
= (n + 1) 2 − n 2 
8mL2 
h2
= (2n + 1) (20.102)
8mL2
As 2n > 1, 2n+1 ≈ 2n
h2
\ ∆E = n (20.103)
4mL2
Case (i): Let us take the case of an electron moving in a box of side L = 1 cm = 10–2 m.
From equ.(20.102) we get
ΔE @ n × 10–15 eV
The thermal energy possessed by the electron is
kT @ 10–3eV at T =1°K
\ ΔE << kT
It means that the electron can move
form a lower energy level to an upper energy
due to its own thermal energy and without E << KT
the need of any energy input from external
agency. As the levels are nearly continuous, En+1
E >> KT
the electron behaves as a classical particle En
without revealing the quantum aspects. 0 L
0 L >> 1mm L L  10Å
Case (ii) Let us next consider the case of
an electron moving in a box of side L = 10 Å. Fig. 20.26
Then
ΔE @ 0.75 (n) eV.
\ ΔE >> kT.
754 A Textbook of Engineering Physics

C H A P T E R

26 Crystal Structures

26.1 INTRODUCTION
A solid consists of atoms or clusters of atoms arranged in close proximity. The physical
structure of a solid and its properties are closely related to the scheme of arrangement of
atoms within the solid. In amorphous solids the arrangement of atoms is random while in
crystals there is a regular arrangement of atoms. Simple geometrical concepts of a lattice
and unit cell are used to describe the atomic arrangement in crystals. The advantage of such
a description is that all possible crystal structures are represented by a limited number of
basic unit cell geometries. Majority of metals are found to belong to three simple structures.
The crystal structures are analyzed using x-ray diffraction technique invented by Max von
Laue and extensively employed by Bragg and Bragg. The study of crystal geometry helps
us understand the diverse behaviour of solids in their mechanical, metallurgical, electrical,
magnetic and optical properties.
26.2 CLASSIFICATION OF SOLIDS
Solids are classified into the following three categories basing on the atomic arrangement
within the solid.
(i) single crystals
(ii) polycrystalline solids and
(iii) amorphous solids.
(i) Single crystals: Single crystals are polyhedrons that have a distinctive shape for
each material and are bounded by smooth shiny faces and straight edges. When a crystal
is broken, it cleaves along certain preferred directions. The same substance may crystallize
under different conditions of crystal growth to form different geometrical shapes but the
angles between the faces are always constant for different shapes. This is known as law
of constancy of angles. X-ray diffraction studies have shown that the atoms in crystalline
solids are arranged in a regular periodic pattern in three dimensions, as shown in Fig. 26.1(a).
The arrangement of atoms in specific relation to each other is called order. In crystals the
order exists in the immediate neighbourhood of a given atom as well as over large distances
corresponding to several layers of atoms. Therefore, crystals possess both short-range order
and long-range order.
In single crystals, since the periodic arrangement of atoms differs in the three directions,
the physical properties vary with direction and therefore, they are called anisotropic
substances.
Crystal Structures 755

Quartz, alum, diamond and rock salt are examples of solids that occur as large size single
crystals.
(ii) Polycrystalline solids: Polycrystalline solids consist of fine grains, having a size of
103 to 104 Å, separated by well-defined boundaries but oriented in different directions (Fig.
26.1b). Each such grain is a single crystal of an irregular shape. Since the grains are oriented
randomly, a polycrystalline material is isotropic and the physical properties do not vary with
direction. Majority of the natural solids have polycrystalline structure. Metals are examples of
polycrystalline solids.

(a) (b) (c)

Fig. 26.1. Atomic arrangement in (a) Single crystal (b) Polycrystalline solid and
(c) Amorphous substance
Types of crystals
The atoms are bound to each other by electrostatic forces. Depending on the nature of
the bond between the neighbouring atoms, the crystals can be classified as metallic crystals,
covalent crystals, and ionic crystals. In ionic crystals, the electron is transferred from the
cation to anion and the atoms exist in the crystal as ions and hence the name ionic crystal.
The interatomic forces in such crystals are non-directional. These forces are directional in
covalent crystals where electron is shared by the neighbouring atoms. In metallic crystals,
the electrons are very loosely bound to the corresponding nuclei.
(iii) Amorphous solids: Solid materials possessing only short-range order (Fig. 26.1c)
are known as amorphous solids. These materials maintain a fixed volume and shape
and resemble solids in their external features, but internally they do not have the ordered
crystalline arrangement. The atomic arrangement is random in these materials. They are in
fact considered as supercooled liquids having a very high viscosity. Glass, rubber and many
polymers are amorphous solids. The amorphous materials are generally referred to as ‘glass’.
Note that the word ‘glass’ here does not refer to the transparent glass used for windowpanes.
The amorphous materials are called glass because they have features similar to glass. The
physical properties of amorphous solids are not dependent on the direction of measurement
and therefore they are isotropic substances.
Though the word ‘solid’ has been traditionally used for crystalline, polycrystalline and
amorphous materials, nowadays the word ‘solid’ is meant to convey only crystalline materials.
26.2.1 Phase
When the crystals are grown there are two important requirements to be satisfied. (i) The atoms
must occupy minimum space, and (ii) the bond angle requirements must be satisfied. For
metallic crystals, the second condition does not exist and therefore, the atomic arrangement in
metallic crystals is usually much simpler than the atomic arrangement in nonmetallic crystals.
756 A Textbook of Engineering Physics

Since the atomic arrangement everywhere in the crystal is the same, the crystal growth occurs
at a fixed temperature. The atoms in the crystal arrange and disperse like soldiers in the army,
each soldier going through identical movement. A given atomic arrangement in the material
is called a phase. Thus, a given chemical like water may exist in vapour, water or ice phase.
If the atomic arrangement in the crystal changes, we say that a new crystal has formed and a
phase change has occurred.
26.2.2 Periodicity
The periodicity of atomic arrangement can be described as
r = ax+by+cz
where x, y, z are unit vectors in x, y, z directions respectively. Note that periodicity is the only
condition for crystallinity.

26.3 SPACE LATTICE

A regular and periodic arrangement of atoms is the most important feature of crystals. The
actual arrangement of atoms is called the structure. Suppose the atoms or clusters of atoms in
a crystal are represented by points that t1 t1 t1
correspond to their mean positions.
Fig. 26.2 a. A sequence of periodically spaced points
Then we obtain a regular distribution
of points in space. These points are
called lattice points or lattice sites. Although lattice points represent atom locations, the
lattice points and atom centres need not be coincident. The three-dimensional network of
regularly arranged points is known
as a space lattice. A point is a
dimensionless and shapeless entity; t2
therefore, a lattice is merely an
imaginary geometrical framework.
A space lattice is defined t2
as an array of points in three
dimensions in which every point
has surroundings identical to Fig. 26.2 b. A planar array of points
that of every other point. A space
lattice can be generated by successive translations of an initial point. A very simple operation
of repetition consists of repeating the unit without change after translating it a distance ‘t’.
Repeated application of a translation of given length and direction, t1, to an initial point
generates a sequence of periodically spaced points or a row of points.
The repeated application of some other translation which is not along the same line (i.e.
non-collinear), t2, to the above row generates a planar array of points.
A third translation not in the same plane (i.e.non-coplanar), t3, applied to a two-dimensional
lattice generates a three-dimensional array of points called a space lattice or simply lattice.
Thus, lattice points satisfy the condition of periodicity. Hence, the crystal structure, i.e.,
the arrangement of atoms in the crystal can be understood with reference to a lattice. It is also
observed that a given point in the lattice (Fig. 26.2c) is surrounded by a specific number of
lattice points located at equal distances. The arrangement of points around a given lattice point
is called its environment. The environment continues to change as we traverse in a particular
direction from one lattice point to the next lattice point, but at the next lattice point it is
Crystal Structures 757

exactly the same as that at the initial Lattice Planes

point. Thus, the same environment
would be found after equal intervals
of distance in the same direction. The
environment would be different in
different directions. The space lattice Lattice Lines
is the skeleton upon which crystal
structure is built by placing atoms on Lattice Points
or near the lattice points.
The study of crystal structure
becomes simpler when it is
Unit Cell
represented by a space lattice.
Fig. 26.2 c. Three dimensional lattice or space lattice
26.4 CRYSTAL STRUCTURE
A space lattice is a mathematical abstraction. The crystal structure is formed only when a
group of atoms is identically attached to each lattice point, as shown in Fig. 26.3. The
group of atoms
that is associated
with every lattice
point is called a
basis. The basis
must be identical + =
in composition,
arrangement and
orientation such that
the crystal appears
Fig. 26.3
exactly the same at
one point as it does at other equivalent point.
A crystal structure is thus the result of two quantities; namely a lattice and a basis. Thus,
Lattice + basis → crystal structure
In the simplest crystals such as copper, silver, gold, iron and the alkali metals, the basis is a
single atom. Often the basis contains a few atoms.

26.5 UNIT CELL

Because of
the inherent a b
periodicity, the c
space lattice can
be represented by   d c
a unit cell. The unit b
a
cell is the smallest  e
geometrical f
unit, which when
repeated in space (a) (b)
indefinitely, Fig. 26.4. (a) A primitive unit cell. (b) Primitive and non-primitive unit cells
generates the space in a two-dimensional lattice
758 A Textbook of Engineering Physics

lattice. Hence, it is the smallest volume that carries a full description of the entire lattice.
The lines drawn parallel to the lines of intersection of any three faces of the unit cell which
do not lie in the same plane are called crystallographic axes. The three translational vectors
a, b, and c lie along the crystallographic axes. The intercepts a, b, and c (Fig. 26.4 a) define
the dimensions of the unit cell and are known as primitives. The angle g represents the angle
between a and b axes, the angle a represents the angle between b and c axes and the angle b
represents the angle between the c and a axes. The axial lengths a, b, c and the three inter-axial
angles a, b, g are known as the basic lattice parameters. The volume of the unit cell is
(a × b) · c . If this volume in space contains only one lattice point, we call it a primitive unit
cell. It is not necessary that every unit cell should be a primitive cell (Fig. 26.4 b). Depending
on the requirement and symmetry of the lattice we can choose larger cell containing more
than one lattice point as a unit cell. If there are two or more lattice points per unit cell, then
it is called a non-primitive unit cell. Most of the unit cells of various crystal lattices contain
two or more lattice points and are non-primitive cells.
Example 26.1. The unit cell of copper is a cube. The side of the cube is 3.6 Å. If the unit
cell in copper solid are lined up side by side, how many unit cells will be there along 10 mm
length of the solid?
Solution. The number of unit cells in a length l of solid is given by
Length of the solid l 10−2 m
N = = = = 28 × 106
side of the unit cell a 3.6 × 10−10 m

26.6 BRAVAIS LATTICES

Bravais introduced the concept of space lattice in the study of crystal structures. A
three-dimensional space lattice is generated by repeated translations of three non-coplanar
vectors. One would expect that many lattices can be generated in three dimensions with
different primitive and non-primitive cells. However, Bravais showed that there are only
fourteen different ways of arranging identical points in three-dimensional space, satisfying
the condition of periodicity, so that they are in every way equivalent in their surroundings.
These fourteen types of arrangements are called the space lattices or Bravais lattices. There
are seven primitive cells and seven non-primitive cells. With the 14 types of lattices and on
the basis of primitive cell, crystals are grouped into 7 systems. They are cubic, tetragonal,
orthorhombic, monoclinic, triclinic, hexagonal and rhombhohedral (trigonal). The seven types
of crystals and the corresponding Bravais lattices with their features are described below.
1. Cubic system
In cubic crystals, the crystal axes are perpendicular to one another. Thus, a = b = g =
90°. The length of the primitives (edges of unit cell) is the same along the three axes. Thus,
a = b = c. Cubic lattice has three possible types of arrangements, simple (SC) or cubic-P,
body centered (BCC) or cubic-I and face
centered (FCC) or cubic-F. a
(i) Simple cubic lattice: It has
lattice points at all 8 corners of a
the unit cell. a

Cubic—P Cubic—I Cubic—F

Fig. 26.5 (A): The three Bravais lattices of a cubic
system
Crystal Structures 759

(ii) Body-centered cubic lattice: It has lattice points at all 8 corners of the unit cell and
one lattice point at the centre of the body.
(iii) Face-centered cubic lattice: It has
lattice points at all 8 corners of the unit
cell and one lattice point each at the c
centre of six faces of the cube.
a
2. Tetragonal system
In tetragonal crystals, the crystal axes are a
perpendicular to one another. Thus, a = b = g =
90°. The lengths of the edges of unit cell along Tetragonal—P Tetragonal—I
two axes are the same but the edge along the third
Fig. 26.5 (B): The two Bravais lattices of
axis is different. That is, a = b ≠ c. Tetragonal a tetragonal system
lattice has two possible types of arrangements,
simple (P) and body centered (I).
(i) Simple tetragonal lattice: It has lattice points at all 8 corners of the unit cell.
(ii) Body-centered tetragonal lattice: It has lattice points at all 8 corners of the unit
cell and one
lattice point at
the centre of the c
body.
3. Orthorhombic b

system a

In orthorhombic Orthorhombic—P Orthorhombic—C Orthorhombic—I Orthorhombic—F

crystals, the crystal axes Fig. 26.5 (C): The four Bravais lattices of a orthorhombic system
are perpendicular to one
another. Thus, a = b = g = 90°. But the lengths of the edges of unit cell along the three axes
are different. That is, a ≠ b ≠ c. Orthorhombic lattice has four possible types of arrangements,
simple (P), base centered (C), body centered (I) and face centered (F).
(i) Simple orthorhombic lattice: It has lattice points at all 8 corners of the unit cell.
(ii) Base-centered orthorhombic lattice: It has lattice points at all 8 corners of the unit
cell and two lattice points, one each at the base and top face of the body.
(iii) Body-centered orthorhombic lattice: It has lattice points at all 8 corners of the unit
cell and one lattice point at the centre of the body.
(iv) Face-centered orthorhombic lattice: It has lattice points at all 8 corners of the
unit cell and six lattice points, one each at the
centre of six faces of the unit cell.
4. Monoclinic system c

In monoclinic crystals, out of the three crystal axes, β

β
two are not perpendicular to each other but the third is b
perpendicular to both of them. Thus, a = g = 90° ≠ b.
a
The lengths of the edges of unit cell along the three axes
are different. That is, a ≠ b ≠ c. Monoclinic lattice has Monoclinic—P Monoclinic—C
two possible types of arrangements, simple (P) and base
Fig. 26.5 (D): The two Bravais
centered (C).
lattices of a monoclinic system
760 A Textbook of Engineering Physics

(i) Simple monoclinic lattice: It has lattice points at all 8 corners of the unit cell.
(ii) Base-centered monoclinic lattice: It has lattice points at all 8 corners of the unit
cell and two lattice points, one each at the base and top face of the unit cell.
5. Triclinic system
In triclinic crystals, none of the three crystal axes is perpendicular to any of the other.
Thus, a ≠ b ≠ g. The lengths of the edges of unit cell along the three axes are different. That
is, a ≠ b ≠ c. Triclinic lattice has only one possible type of arrangement, namely, simple (P).

c a
 
 b 
a
 a
a  b
a
Triclinic Trigonal—R

Fig. 26.5(E): The Bravais Fig. 26.5 (F): The Bravais Hexagonal—P
lattice of a triclinic system lattice of a trigonal system
Fig. 26.5 (G): The
Bravais lattice of a
hexagonal system

(i) Simple triclinic lattice: It has lattice points at all 8 corners of the unit cell.
6. Trigonal or Rhombhohedral system
In trigonal crystals, the angle between each pair of crystal axes is the same but is not
equal to 90°. Thus, a = b = g ≠ 90° < 120°. The length of the edges of unit cell is the same
along all the three axes. That is, a = b = c. Trigonal lattice has only one possible type of
arrangement, namely, simple (R).
(i) Simple trigonal lattice: It has lattice points at all 8 corners of the unit cell.
7. Hexagonal system
In hexagonal crystals, two of the crystal axes are 120° apart, while the third axis is
perpendicular to both of them. Thus, a = b = 90° and g = 120°. The length of the edges of
unit cell is the same along all the axes that are 120° apart but the edge along the third axis
is different. That is, a = b ≠ c. Hexagonal lattice has only one possible type of arrangement,
namely, simple (P).
(ii) Simple hexagonal lattice: It has lattice points at all 12 corners of the hexagonal unit
cell and two lattice points, one each at the base and top face of the hexagonal prism.
The seven crystal systems and the parameters of the corresponding unit cells are
summarized in Table-1.
Crystal Structures 761

Table 1: The seven crystal systems

Sr. Crystal System Type and number Unit cell Angle between axes
No. of Bravais lattices axes
1. Triclinic P,1 a≠b≠c a ≠ b ≠ g ≠ 90°
2. Monoclinic P a≠b≠c a = g = 90° ≠ b

  2
C 
3. Orthorhombic P a≠b≠c a = b = g = 90°

C


  4
I
F

4. Tetragonal P  a=b≠c a = b = g = 90°

  2
I 
5. Cubic P a=b=c a = b = g = 90°

I  3
F 
6. Hexagonal P, I a=b≠c a = b = 90°, g = 120°
7. Trigonal R, I a=b=c a = b = g ≠ 90° < 120°
(Rhombohedral)

26.7 SYMMETRIES IN CRYSTALS

Some similarities of atomic arrangements may occur in crystals as a result of periodic
arrangement. This is called symmetry of atomic arrangement. Indeed, the crystal can be
characterized by the symmetry it possesses. Crystals possess different external symmetries,
which are described by certain mental operations. A symmetry operation is one that takes
the crystal into a configuration identical to the initial configuration. The crystal is said to
possess a symmetry element corresponding to an operation, if after performing the particular
operation the crystal goes into a position indistinguishable from the initial position. The most
important elements of symmetry are a center, an axis and a plane.
1. A crystal is said to have a center of symmetry, if a point exists within the crystal
such that any line drawn through it will have similar situation at both of its ends.
This means that if there is an atom at some distance at one end, there is also an atom
at the other end at the same distance. It is also called the center of inversion.
2. A crystal is said to possess an axis of symmetry if when the crystal is rotated about
the axis, the atomic arrangement looks exactly the same more than once during
one complete revolution. If the atomic arrangement looks the same twice in one
revolution, the axis is said to be a diad axis. Similarly, the axis may be a triad axis,
tetrad axis, and hexad axis. Note that only these types of axes are possible. The
other types of axes are not consistent with the condition of periodicity. The diad
axis is represented by 2, triad axis by 3, tetrad axis by 4, and the hexad axis by 6,
corresponding to 180°, 120°, 90°, 60°, rotations respectively.
3. A crystal is said to have a plane of symmetry when it can be divided by an
imaginary plane into two parts, such that one is the exact mirror image of the other.
762 A Textbook of Engineering Physics

The mirror plane is represented by the letter m. Usually, the mirror plane is parallel
to rotation axis or perpendicular to it.
In addition to the above point operations, there exist the following four hybrid operations.
Hybrid operations
(i) Rotoreflection: This is the combination of an n-fold rotation followed by a reflection
in a plane perpendicular to the rotation axis.
(ii) Rotoinversion: This is the combination of an n-fold rotation followed by an
inversion.
(iii) Screw translation: In this the n-fold rotation axis is coupled with the translation
parallel to rotation axis.
(iv) Glide reflection: In this a mirror plane is coupled with a translation parallel to the
reflecting plane.
Each type of symmetry is called a symmetry element. The crystal may have more than
one symmetry element. All the symmetry elements possessed by a crystal, grouped together,
are called a symmetry group or a point group. Because of the restriction of the condition
of periodicity, the number of possible point groups is not large. It was shown that all possible
combinations of symmetry elements lead to 230 space groups divided into 32 symmetry
classes. These are grouped into the seven crystal systems.
26.7.1 Symmetry Elements in a Cubic Crystal
Basing on the internal structure, only certain elements of external symmetry can occur in a
crystal. We now take the example of a cubic crystal and illustrate the elements of symmetry it
exhibits. R
(i) Let us consider the cube
→r

shown in Fig. 26.6. If the N

body center point is I, and →

r
body diagonals are drawn →r I P Q
through it, each diagonal
connects identical lattice
→r

M
points located at equal
S
distances and in opposite
Fig. 26.6 Fig. 26.7
directions from I, The point
I acts as a point mirror which generates the second lattice point at an equal distance in
opposite direction. Therefore, the point I is
C K
the center of symmetry or inversion point.
Thus, for a cubic crystal inversion point I is
located at the body center.
(ii) Let us consider a normal MN through
mid-points of the pair of opposite
parallel faces of a cube. If the cube is
rotated through 90° rotation, it goes into
an indistinguishable configuration. In
one complete rotation of 360°, the cube D L
becomes indistinguishable four times. Fig. 26.8       Fig. 26.9
Therefore, the face-normal is a 4-fold axis
of symmetry. There are three such 4-fold axes of symmetry, MN, PQ and RS, one
normal to each of the three pairs of parallel faces (see Fig. 26.7).
Crystal Structures 763

Now consider the axis CD passing through the body diagonal of the cube (Fig. 26.8).
If the cube is rotated around the body diagonal through 120°, it comes into a congruent
position. In one full rotation, the cube becomes indistinguishable three times. Therefore, the
body diagonal is a 3-fold axis of R
symmetry. A cube has four body S
diagonals and therefore possesses
four 3-fold axes of symmetry.
Next, consider a normal
KL at mid points of parallel Q
edges of Fig. 26.9. If the cube is P

rotated through 180°, it becomes Fig. 26.10

congruent. In one rotation of 360°, the cube assumes indistinguishable configuration twice.
Therefore, KL is a 2-fold axis of symmetry. As there are 12 edges in a cube, the number of
2-fold axes of symmetry are six. N
The axes of symmetry in a cube
are thus M
2 - fold axes (diad axes) 6
3 - fold axes (triad axes) 4 K

4 - fold axes (tetrad axes) 3 L

------
13
There are in total 13 axes of
symmetry in a cube.
(iii) Let us consider a plane such
Fig. 26.11
as PQRS (Fig. 26.10) in the
middle of the cube and parallel to one pair of the faces. If it is a plane mirror, and
one half of the crystal is cut and removed, the plane PQRS forms the image of that
half of the crystal in it. That means, if we reflect one half of the crystal in PQRS,
the image will coincide with the other half. Therefore, PQRS is called a plane of
symmetry or a mirror plane. There are
three such planes of symmetry parallel to
the faces of the cube.
Further, consider the diagonal plane KLMN
(Fig. 26.11) in the cube. If KLMN is imagined to Simple cubic Body-centered Face-centered
be a mirror, it is readily seen that the prism behind cubic cubic
it is the reflection of the prism in contact with it in Fig. 26.12. The three space lattices of a
its front. Thus it is also a mirror plane. There are cubic system
six such diagonal mirror planes in a cube.
The total elements of symmetry of a cubic crystal are 23 comprising of one center of
inversion, 13 axes of rotation and 9 mirror planes.

26.8 CALCULATION OF PARAMETERS OF A CUBIC LATTICE

A unit cell is characterized by a set of characteristics, namely, volume, effective number of
atoms, coordination number, atomic packing fraction, and density. These parameters can be
computed with relative ease in case of cubic unit cells. There are three varieties of the cubic
764 A Textbook of Engineering Physics

lattice as shown in Fig. 26.12. They are simple cubic (SC), body centered cubic (BCC), and
face centered cubic (FCC) unit cells.
26.8.1 Simple cubic (SC) cell
A unit cell is said to be primitive when the
cell has lattice points only at its corners. The
primitive cubic unit cell is also called a simple
cubic (SC) unit cell. Fig. 26.13 shows a SC
cell.
(i) Unit cell volume, V: In case of cubic
cell all the edges of the cube are of equal (a) (b)
length,’a’. Therefore, the volume is given by Fig. 26.13. Simple cubic cell (a) atomic site
V = a3 (26.1) model (b) hard sphere model
(ii) Effective number of atoms per unit cell, Z: A unit cell is a part of an infinite
scheme, and is not an isolated entity. Therefore, several adjacent cells share each lattice point.
As a result, the basis attached to a lattice site contributes only a fraction of its mass and
volume to one unit cell. Let only one atom, having a radius R, be attached to one lattice point.
The effective number of atoms per unit cell is given by
Z Z
Z = ZB + F + C (26.2)
2 8
where ZB = Number of body centered atoms,
ZF = Number of face centered atoms and
ZC = Number of
corner atoms
In the a
three-dimensional array,
each corner atom is linked
to eight surrounding cells,
as shown in Fig. 26.14 a
(a). Hence, in effect, the A R R B
a
atom contributes 1/8th of
its content to a unit cell, (a) (b)
as shown in Fig. 26.14 (b). Fig. 26.14. (a) An atom contributes one eighth to a unit cell
SC cell being a primitive when it is located at the corner of the cell. (b) Isolated SC cell
cell does not contain lattice
points within the body volume or in the centre of faces. Therefore, the total contribution to Z
comes from the corner atoms of the unit cell and it is given by
Z Z 8
Z = Z B + F + C = 0 + 0 + = 0 + 0 + 1 atom / cell
2 8 8
\ Z = 1 atom / cell (26.3)
(iii) Coordination Number, CN: The coordination number of an atom in a crystal is
the number of nearest neighbour atoms. It signifies the tightness of packing of atoms in the
crystal.
Around an atom in a SC cell, there would be six equally spaced nearest neighbour atoms
each at a distance ‘a’ from that atom, as shown in Fig. 26.15. Four atoms lie in the plane
Crystal Structures 765

of the atom while one is vertically above it and one

vertically below. Therefore, the coordination number 3

CN = 6.
(iv) Atomic radius, R: The relationship between 6
the apparent size of the atom and the edge of the unit 1 2
A
cell can be determined where one atom is attached to
5
a lattice site. The specific direction along which atoms
are in contact is identified and by applying simple
geometry, the relation between the atomic size and the 4
unit cell edge can be computed.
In a SC cell the atoms would be in contact along
the edges of the cube, as seen from Fig. 26.14 (b). If Fig. 26.15. Determination of nearest
‘a’ is the edge of the cubic cell and R is the radius of neighbours. There are six close
the atom. neighbours for any selected atom
in a SC cell.
a = 2R  or  R = a / 2 (26.4)
(v) Packing Fraction, APF: The fraction of space occupied by atoms in a unit cell is
known as atomic packing fraction. It is defined as the ratio of volume of effective number of
atoms in the unit cell to the total volume of the unit cell. Thus,
(Number of atoms/unit cell) (Volume of each atom)
APF = (26.5)
Volume of the unit ceell
Zυ
\ APF =
V
In case of SC cell, Z = 1
Volume of unit cell, V = a3 = (2R)3 = 8R3.
4 3
Volume of spherical atom, u = πR
3
4 3
1 × πR
3 π
\ APF = 3
= = 0.52 (26.6)
8R 6
(vi) Percentage void space: The void space in the unit cell is the vacant space left
unutilised in the cell. It is often expressed as percentage.
% void space = (1 – APF) × 100
= (1 – 0.52) × 100
= 48% (26.7)
(vii) Density of SC crystal, r: As a unit cell possesses all the structural properties of a
bulk crystal, the density of a unit cell must be the same as that of the bulk crystal. Thus,
Mass ZW
Density, r = = (26.8)
Volume V
where W is the mass of each atom, which is given by
M
W= (26.9)
NA
M is the molecular weight of the material and NA is the Avogadro number.
766 A Textbook of Engineering Physics

Therefore, the density of a cubic crystal is given by

ZM ZM
r= = (26.10)
N AV N A a3
In case of SC crystal Z = 1.
M
Hence, r= (26.11)
N A a3

26.9 BODY CENTRED CUBIC (BCC) CELL

The body centered cubic cell has a lattice point within the cell in addition to the eight corner
lattice points, as shown in Fig. 26.16 (a).
(i) Unit cell volume, V: In case of cubic cell all the edges of the cube are of equal length
‘a’. Therefore, the volume is given by
V = a3
(ii) Effective number of atoms per
H
unit cell, Z:
E
A BCC cell has eight lattice points F
at the eight corners of the cube and one
a
lattice point at centre within the cell,
as shown in Fig. 26.16 (a). There are no A C
points in the face. An atom at the lattice a a
point within the cell belongs completely B
to the cell and the atoms at the corners of (a) (b)
the cell contribute 1/8th each, as shown in Fig. 26.16. (a) BCC cell (b) Isolated BCC unit cell
Fig. 26.16 (b). The effective number of
atoms per BCC cell is then
Z Z
Z = ZB + F + C
2 8
8
Z = 1+0+
8
\ Z = 2 atoms/cell (26.12)
(iii) Coordination Number, CN:
In the BCC cell, atoms occupying the corner lattice points
do not touch each other. However, each corner atom is in
contact with the atom at the body center (Fig. 26.17). As there
are eight unit cells around each corner of the cell, the atom
located at a corner would be simultaneously touched by the
eight body centered atoms of the eight surrounding cells. Thus,
the coordination number
CN = 8
(iv) Atomic radius, R: Fig. 26.17. A given atom in
In a BCC cell atoms would be in contact along the body a BCC cell is surrounded
diagonal. From the Fig. 26.18, by 8 closed neighbours
Crystal Structures 767

(AG)2 = (AC)2 + (CG)2 = (AB)2 + (BC)2 + (CG)2 (26.13)

AG = 4R and  AB = BC = CG = a H
\ (4R)2 = 3(a)2
E R G
\ 4R = a 3 F

a 3 2R 3a a
\ R = (26.14)
4 R

(v) Atomic packing fraction, APF: A C

4R
For BCC cell, Z = 2 and a = B
3 2a
4 3
Volume of the spherical atom u = πR 3 a = 4R
3
Fig. 26.18. Determination of the
64 R3 relation between atomic radius
Volume of the unit cell V = a3 =
3 3 and lattice edge in a BCC cell
4
2 × πR3
Z 3 π 3
APF = υ = = = 0.68 (26.15)
V 64 3 8
R
3 3
(vi) Percentage void space:
Percentage void space = (1 – APF) × 100
= (1 – 0.68) × 100 = 32% (26.16)
(vii) Density of BCC crystal, r:
In case of BCC crystal Z = 2. Hence
2M
r= (26.17)
N A a3
Example 26.2. Molybdenum has a BCC structure. Its density is 10.2 × 103 kg/m3 and its
atomic weight is 95.94. Determine the radius of molybdenum atom.
ZM
Solution: Density of a cubic unit cell is given by r =
N A a3
ZM 2 × 95.94 kg / k.mol
\ a3 = = = 31.25 × 10–30m3
ρ NA 10.2 × 10 kg / m3 × 6.02 × 1026 / k.mol
3

\ a = 3.15 Å.
a 3 3.15 Å × 1.732
The atomic radius in BCC structure is given by R = = = 1.364 Å.
4 4
Example 26.3. Sodium crystallizes in a cubic lattice. The edge of the unit cell is 4.3Å. The
density of sodium is 963 kg/m3 and its atomic weight is 23. What type of unit cell does sodium
form?
Solution:
ZM
Density of a cubic unit cell is given by r =
N A a3
768 A Textbook of Engineering Physics

ρ N A a3
The effective number atoms per unit cell is given by Z =
M

963kg/m3 ×6.02×1026 /k.mol ×(4.3×10−10 m)3 46.09

\ Z= = = 2 atoms/unit cell
23kg/k.mol 23
\ Sodium forms BCC structure.

26.10 FACE CENTRED CUBIC (FCC) CELL

The face centred cubic cell is a non-primitive cell having six lattice points at the centres of its
six faces and eight atoms at the eight corners of the cube, as shown in Fig. 26.19.
(i) Unit cell volume, V: In case of cubic cell all the edges of the cube are of equal length
‘a’. Therefore, the volume is given by
V = a3
(ii) Effective number of atoms per unit cell, Z:
In a three dimensional array, a unit cell is surrounded by other unit cells, while each
corner of the cell is shared by eight adjacent cells, and each face is shared by two adjacent
cells, as shown in Fig. 26.20. Therefore, each face-centred atom contributes half of its mass
and volume to one cell, while each corner atom contributes 1/8th, as shown in Fig. 26.21.

H
G
E
a

C
A a
a
Face-centered
B
cubic Fig. 26.20. An atom contributes
Fig. 26.19. Face half of its volume to the unit cell Fig. 26.21. Determination of affe-
Centered Cubic cell when it is located in the face ctive number of atoms per unit cell.

There are 6 faces and 8 corners in a cubic cell. Thus, the number of atoms effectively contrib-
uting to the FCC cell is
Z Z
Z = Z B + F + C
2 8
6 8
Z = 0+ + (26.18)
2 8
\ Z = 4 atoms/cell.
(iii) Coordination Number, CN:
In a FCC cell, each corner atom is in contact with the face-centered atom. It would
therefore be in contact with 4 atoms in the xy plane, 4 atoms in the yz-plane and 4 atoms in
Crystal Structures 769

Z
YX-Plane ZX-Plane yz-plane
Y
7 8 X 7 8
4
12 A
XY-Plane
2 1 A
6 5

(a) (b)

Fig. 26.22. Determination of nearest neighbours in a FCC cell

the zx-plane, as shown in Fig. 26.22. Therefore,
Coordination number = 4 + 4 + 4 = 12.
(iv) Atomic radius, R:
In a FCC cell, atoms are in contact along the face diagonal of the cube, as shown in Fig.
26.23. It is seen from the figure that
AF2 = AB2 + BF2 (26.19) E
R
F

But AF = 4R and AB = a = BF
\ (4R)2 = 2 a2 R 2 a
\ 4R = a 2
a 2 a R
\ R = = (26.20)
4 2 2
A B
4R = 2 a
(v) Atomic packing fraction, APF: Fig. 26.23. Determination of
Zυ 4υ atomic radius in a FCC cell
APF = = 3
V a
3 3
a3 =  2 2 R  = 16 2 R
4
4 × πR3
3 π
APF =
\ 3
= = 0.74 (26.21)
16 2 R 3 2
(vi) Percentage void space:
Percentage void space = (1 – APF) × 100 = (1 – 0.74) × 100 = 26 % (26.22)
(vii) Density of FCC crystal, r:
The density of a cubic crystal is given by
ZM ZM
r = =
N AV N A a3
In case of FCC crystal Z = 4.
4M
Hence r = (26.23)
N A a3
Example 26.4. Lead exhibits FCC structure. Each side of the unit cell is of 4.95 Å. Calculate
the radius of a lead atom.
770 A Textbook of Engineering Physics

a 4.95 Å
Solution: The radius of the lead atom is given by R = = = 1.75 Å.
2 2 2(1.414)

26.11 HCP STRUCTURE N

Hexagonal close packed (HCP) structure is one of the most
common metallic structures. About 25 metals exhibit this
P R
structure. Metals do not crystallize into the simple hexagonal
crystal structure. The atoms attain a lower energy and a more
stable condition only by forming the HCP structure. c
Q
(i) Effective number of atoms per unit cell
The isolated HCP unit cell is shown in Fig. 26.24. Each 
corner atom of the hexagonal face is shared by six unit cells.  b
Consequently it contributes 1/6th of its volume and mass to
a M

one unit cell. There are six corner atoms in the base. Further,
there is an atom at the center of the each hexagonal face, Fig. 26.24
which is shared by two adjacent unit cells. Three atoms forming a triangle in the middle layer
are within the body of the cell and cannot be shared by adjacent cells. The total contribution
to the effective number of atoms in the cell is thus

 base   1 atom   corners   bases   1 atoms   atoms 

Z =  2 ×  × 6  + 2 ×  + 3 
 cell   6 corner   base   cell   2 base   cell 
or Z = 6 atoms/unit cell.
(ii) Atomic radius, R: The atoms are in contact along the edges of the hexagon as seen
from Fig. 26.24. Therefore,
2R = a or R = a/2
(iii) Coordination number, CN: Each atom in the structure is positioned in valleys
formed by three adjacent atoms of the top layer and by three adjacent atoms in the bottom
layer; and is surrounded by six neighbour atoms. Thus, twelve atoms are in contact with the
given atom under consideration. Therefore,
C
CN = 12 F E

(iv) Volume of the unit cell,V: C

The volume of the unit cell may be G D a h
determined by computing the area of the 60° 60°
base of the unit cell and then by multiplying A a B A
a
B
it by the cell height. The area of the base
(a) (b)
of the unit cell is the area of the hexagon
ABDEFG, shown in Fig. 26.25 (a). It is Fig. 26.25
equal to the sum of the areas of the six equal-sized triangles (Fig. 26.25 b).
Area of the hexagon ABDEFG = 6 (area of D ABC) = 6(1/2a)(a sin 60°)
2 3 3 3 2
= 3a2 sin 60° = 3a   = a
 2  2

\ Volume of the unit cell = (Area of the hexagon) × (height of the unit cell)
Crystal Structures 771

3 3 2
=  a  c
 2 

3 3 2
or V = a c (21.24)
2
Calculation of c/a ratio:
Referring to the HCP unit cell shown in Fig. 26.26, it is seen N
that a = b and the angle q = 120°. The axis is normal to the plane
containing a and b. Therefore a = b = 90°. Let P, Q and R be
the centers of the adjacent atoms in the middle plane and N and
M be the centers of the adjacent atoms in the plane immediately
c
above and below the plane PQR, as shown in Fig. 26.26. Now let P R
us join M to N and M and N to P, Q and R. It forms two tetrahe- S T L
drons MPQR and NPQR with PQR as the common base. The line Q 90°
MN passes through the point T, which is the intersection of the
three medians of the triangle PQR. RS is one of the medians and
therefore,
TR = 2 TS M

Further, MN = c and PQ = QR = a Fig. 26.26

As the median of an equilateral triangle is also perpendicular to the opposite side,

∠QSR = 90°.
2
2 a 3
\ RS = a −   = a
2 2
2 2 2 3 a
Also, RT = 2TS = (TS + TR ) = RS = ⋅ a =
3 3 3 2 3
a2 2
And NT = NR 2 − RT 2 = a2 − = a ( NR = NQ = NP = a)
3 3
2 8
c = 2NT = 2 a = a
3 3
c 8
\ = (26.25)
a 3
3 3 2 3 3 2 8
\ The volume of the hexagonal cell, V = a c = a ⋅ a = 3 2 a3 (26.26)
2 2 3
(v) Atomic Packing Fraction (APF):
4
6 × πr 3
Zυ 3 π
APF = = 3
= = 0.74
V 3 2 ( 2r ) 3 2
(vi) Void space
Percentage void space = (1 – APF) × 100 = (1 – 0.74) × 100 = 26%
772 A Textbook of Engineering Physics

(vii) Density, r:
The theoretical density of HCP cell is given by
ZM 6M 2M
r= = 3
= (26.27)
N AV N A (3 2a ) N A a3
The characteristics of the three types of cubic unit cells and the HCP cell are summarized
in Table-2.
Table - 2
Sr.No. Characteristics Unit Cell
SC BCC FCC HCP

1. Unit cell volume, V a3 a3 a3 3 2 a3

2. Atoms per unit cell, Z 1 2 4 6
3. Atomic radius, r a/2 a 3/4 a/2 2 a/2
4. Coordination number, CN 6 8 12 12
5. Atomic packing fraction, APF p/6 π 3 /8 π/3 2 π/3 2
= 0.52 = 0.68 = 0.74 = 0.74
6. Void space 48% 32% 26% 26%
M 2M 4M 2M
7. Density, r
N Aa 3 N Aa 3 N Aa 3 N Aa 3

It is seen from the above table that the SC cell is loosely packed and the FCC and HCP
cells are the close packed cells among the four types of unit cells.

26.12 ATOM POSITIONS IN CUBIC UNIT CELLS

Let a rectangular right handed coordinate system be attached to the cubic lattice, as shown in
Fig. 26.27. The position of a lattice site is described by three coordinates (x, y, z) which are
expressed as
x = pa, y = qb, z = rc (26.28) z
where a, b and c are lattice constants in x, y and z (0, 0, 1)
(0, 1, 1)
directions respectively and p, q and r are integers. If
the lattice constants a, b and c are taken as unit axial (1, 0, 1)
lengths, a = 1, b = 1, and c = 1, and the co-ordinates of
the lattice site will be (p, q, r). These are the indices of 1, 1, 1
(0, 1, 0)
the lattice site and are written with commas separating (0, 0, 0) 2 2 2 y
0 1 ,1,0
the numbers and enclosed in parenthesis. 2

As an example, the indices of atomic sites in the (1, 0, 0) (1, 1, 0)

x
BCC cell are shown in Fig. 26.27. The atom positions
for the eight corner atoms are (0,0,0), (1,0,0), (0,1,0), Fig. 26.27
(0,0,1), (1,1,1), (1,1,0), (1,0,1), and (0,1,1). The body cnetre atom in the cell has the
1 1 1
coordinates  , ,  . By convention, the set of coordinates (0,0,0) stands for the locations
2 2 2
of all eight corners of cubic unit cell. Therefore, it is sufficient if the atom positions (0,0,0)
1 1 1
and  , ,  are specified for the BCC cell.
2 2 2
Crystal Structures 773

The minimum number of coordinate sets necessary to specify the locations of all atoms in
a unit cell are equal to the value of Z, the effective number atoms/cell.

26.13 INDICES OF CRYSTALLOGRAPHIC DIRECTION

Many crystals exhibit properties that are dependent on the direction in which they are measured
and on the direction of the external stimuli, such as an electric field, magnetic field or mechanical
stress, acting on the crystal. Therefore, it is essential that we know reference directions in
crystals. To describe a direction in a crystal lattice, a straight line passing through the origin
is chosen. The co-ordinates of the first lattice point lying on this line are utilized to denote
the direction of the line. Thus, the indices of direction are the vector components of direction
resolved along z z
each of the axes
R
and reduced to the [0 0 1]
smallest integers. [1
The vector 1
0]
components are
multiples of lattice [1 0 0] O P y
O y
[0 1 0]
constants. If the 1/2
S
[1 [1
direction passes Q 1/2 10
] 1
0]
through the origin, x x
then the indices of Fig. 26.28. Indices of direction Fig. 26.29. Indices of parallel directions
the lattice will also
be the indices of direction.
The indices of direction are calculated by the following procedure.
(i) First, find the coordinates of the lattice site nearest to the origin in a given direction.
(ii) The coordinates are then divided by appropriate unit translations.
(iii) If fractions are obtained, each of the fractions is multiplied by smallest common
divisor.
(iv) The integers obtained are the indices of direction written in square brackets, as
[p q r].
Thus, the indices of direction in a crystal are the set of smallest integers, which have the
same ratios as the components of a vector in the desired direction, referred to the axes.
As an example let us consider a cubic cell shown in Fig. 26.28. Let the point ‘O’ be the
origin. The point P is at the position (0, b, 0). ‘b’ represents one unit distance. Therefore, the
indices of the point P are (0,1,0). The direction OP is given as [0 1 0]. On the other hand, the
direction of PO is specified as 0 1 0  . A bar on the number indicates a negative direction.
Similarly, the directions OQ and OR are indicated as [1 0 0] and [0 0 1] respectively.
The direction of OS may be determined as follows. The position indices of the point S are
1 1 
 , , 0  . Then, using the instruction (iii), the direction of the line OS is obtained as
2 2 

 1 1 
 2 × 2 , 2 × 2 , 0  = [110].
 
774 A Textbook of Engineering Physics

All parallel directions have the same direction indices (Fig. 26.29) and the parallel
directions are equivalent. <pqr> represents the family of directions [0 1 0], [0 0 1], [1 0 0],
[0 1 0] , [0 0 1 ] , [ 1 0 0] . All of them are grouped as <1 0 0>.

26.14 LATTICE PLANES AND MILLER INDICES

A crystal lattice may be regarded as an aggregate of a set of parallel, equally spaced planes
passing through the lattice points. The planes are called lattice planes and the perpendicular
distance between adjacent planes is called interplanar spacing. A given space lattice may
have infinite sets of lattice planes, each having its characteristic interplanar spacing. In a
crystal the geometrical location of a plane is not important. All the planes that are parallel to
each other play a similar role.
The position and orientation of a lattice plane Z

in a crystal are determined by three smallest whole

numbers which have the same ratios with one
C
another as the reciprocals of the intercepts of the
plane on the three crystal axes. These numbers are
known as Miller indices of that plane. B
Y
We find the Miller indices of a given plane as O
follows. We choose a set of coordinate axes parallel
to unit cell edges at a convenient point. Let OX, OY X A
and OZ be positive directions of the axes along the
Fig. 26.30. A lattice plane intercepting
three edges of the unit cell, as shown in Fig. 26.30. the three crystallographic axes.
We consider a plane ABC oriented with respect to the
coordinate axes and which intercepts the x-, y-, and z-coordinates at distances OA, OB and
OC respectively. It will be seen that Miller indices for all the planes parallel to the plane ABC
and containing similar type of atomic array are the same. The intercepts are not measured in
metrical distances of cm or Å, but are measured in terms of respective unit lengths assigned
to each edge of the cell, regardless the actual dimension of the edge. Thus, the intercepts OA,
OB, and OC made by a plane such as ABC are expressed as multiples of the axial lengths.
For the plane of interest, we determine the intercepts x, y, and z on the coordinate axes.
• We express the intercepts in terms of the base vectors of the unit cell, pa, qb, and rc;
p, q and r are not necessarily integers but they do have rational ratios.
• We form the reciprocals 1/p, 1/q, and 1/r.
• If the reciprocals are fractions, each fraction is multiplied by the least common
denominator, D, and reduce them to the smallest set of integers. Thus, we obtain
D/p = h, D/q = k and D/r = l respectively.
• We put the integers in parenthesis (hkl). These are the Miller indices of the given
plane.
All the planes, having the same structure of atoms, which are parallel to each other and
therefore have the same orientation with respect to the three edges of the unit cell belong to
the same family of planes and are represented by the same Miller indices. Consider a plane, of
a family of similar planes, lying within the unit cell and nearest to the origin. This plane cuts
off intercepts equal to (1/h)a, (1/k)b,(1/l)c. This implies that the (hkl) plane divides “a” into
h parts, “b” into k parts and “c” into l parts. Other members of the family cut off intercepts
2a/h, 2b/k, 2c/l,……..n a/h, n b/k, n c/l so that the ratio between the intercepts is the same.
Crystal Structures 775

Some important features of Miller indices

1. Parallel planes spaced equally have the same Miller indices. Thus, all the planes in a
set parallel to a particular plane (hkl) are denoted by (hkl).
2. A plane parallel to one coordinate axis has Miller index 0 for that direction.
3. A plane passing through origin is denoted by Miller indices of a parallel plane having
non-zero intercepts.
4. When the intercept of a plane is negative on an axis, a bar is placed on the
corresponding Miller index.
26.14.1 Miller Indices of Principal Planes in a Cubic Cell
Any plane in the crystal containing some atoms is a crystal plane. However, the effect of the
planes with small atomic density on the physical properties of the crystal is negligible. Hence,
the planes with appreciable atomic density are called principal planes.
Fig. 26.31 z z
z
depicts the principal
crystallographic (110) (111)
planes of cubic (100)
crystal structure. y
y
One of the faces of y

the cube is shown x x x

shaded in Fig. 26.31 (a) (b) (c)

(a). The shaded Fig. 26.31. The three principal Miller index planes in a cubic lattice.
plane intercepts the
axes x, y, z at 1, ∞, ∞ respectively. The reciprocals of the intercepts are 1,0,0. The Miller
indices for this plane are (100). The plane is called one-zero-zero plane. The plane in Fig.
26.31 (b) has the intercepts at 1, 1, ∞. Therefore, the Miller indices are (110) and the plane is
called one-one-zero plane. Finally, the plane in Fig. 26.31 (c) has intercepts at 1,1,1 and hence
the Miller indices are (111). This plane is called one-one-one plane. These three planes are
known as the principal planes of cubic crystal system.
26.14.2 Sketching a Lattice Plane (hkl)
The intercepts made by a plane can be found if the Miller indices of the plane are known.
● Let the Miller indices of a plane be (hkl).
● A unit cell (for e.g. a cube) is drawn with three coordinate axes.
● From the Miller indices, the z
1 1 1
reciprocals are obtained → , ,
h k l
● The intercepts are marked on the
coordinate axes.
1
● The three non-planar points are joined 1 2 y
and the plane is obtained. 3

Example: Draw a plane (321)

Step (i): A cube having unit axial lengths
(a = 1) along the three coordinate axes is drawn x
as shown in Fig. 26.32. Fig. 26.32. (321) plane
776 A Textbook of Engineering Physics

1 1
Step (ii): The reciprocals of Miller indices are p = ,q= and r = 1.
3 2
1 1
Step (iii): The intercepts, , and 1, are marked within the cube on x, y and z axes
3 2
respectively.
Step (iv): A plane is drawn through the points and is shaded.

26.15 INTERPLANAR SPACING IN A CUBIC LATTICE

The distance ‘d’ between successive lattice
Z
planes is known as the interplanar distance. Y
The interplanar distance ‘d’ involves the axial
lengths of the unit cell and the Miller indices of
the planes. We shall derive here an expression Normal to
B
for d in the case of a cubic system only. We C Plane ABC
know that the three axes of a cubic crystal are
mutually perpendicular. c/l
b/k d N

Let ABC be one of the family of parallel γ β

lattice planes in the crystal (Fig. 26.33). Let α A
X
its Miller indices be (hkl) and the intercepts O a/h
on the crystallographic axes be OA = a/h, Fig. 26.33. Determination of interplanar
OB = b/k and OC = c/l. The next plane of the set distance in a cubic crystal
parallel to ABC passes through the origin of the
coordinates O (which is not shown in the Fig. 26.33). Therefore, ON, the length of the normal
from the origin to the plane is equal to ‘d’. Let a, b, and g be the angles ON makes with the
three crystallographic axes respectively. Then the direction cosines of ON are
ON d
cos a = =
OA a / h
ON d
cos b = =
OB b/k
ON d
cos g = = (26.29)
OC c/l
The sum of the squares of the direction cosines of a line equals unity. Thus,
cos2 a + cos2 b + cos2 g = 1 (26.30)
Substituting the values of cos a, cos b, and cos g in the above equation, we get
d2 d2 d2
2 2 + 2 2 + 2 2 = 1
a /h b /k c /l

 h2 k 2 l 2 
d2  2 + 2 + 2  = 1
 a b c 
−1/ 2
 h2 k 2 l 2 
\ d =  2 + 2 + 2 (26.31)
 a b c 
Crystal Structures 777

In case of cubic system, a = b = c. The above equation then reduces to

a
dhkl = (26.32)
(h 2 + k 2 + l 2 )

Example 26.5. Determine lattice constant for FCC lead crystal of radius 1.746 Å. Also find
the spacing of (i) (111) planes, (ii) (200) planes and (iii) (220) planes.
Solution: The lattice constant is related to the atomic radius in FCC structure through
a = 2 2 r = 2 × 1.414 × 1.746 Å = 4.938 Å.
a
The interplanar spacing is given by d =
h2 + k 2 + l 2
4.938
(i) d111 = Å = 2.851 Å
12 + 12 + 12
4.938
(ii) d200 = Å = 2.469 Å
22 + 02 + 02
4.938
(iii) d220 = Å = 1.746 Å
22 + 22 + 02
Example 26.6. The density of copper is 8980 kg/m3 and unit cell dimension is 3.61Å. Atomic
wt. of copper is 63.54. Determine crystal structure. Calculate atomic radius and interplanar
spacing of (1 1 0) plane.
ρ N A a3
Solution: (i) The effective number atoms per unit cell is given by Z =
M
8980 kg/m3 × 6.02 × 1026 / k.mol × (3.61 × 10−10 m)3 254.332
\ Z= = = 4 atoms/unit cell.
63.54 kg/k.mol 63.54
\ Copper exhibits FCC structure.
(ii) The lattice constant is related to the atomic radius in FCC structure through a = 2 2 r
a 3.61 Å
r= = = 1.276 Å
2 2 2 × 1.4142
a
(iii) The interplanar spacing is given by d =
h + k2 + l2
2

3.61 Å
\ d= = 2.553 Å.
1 + 12 + 02
2

Example 26.7. A crystal with primitives 1.2 Å, 1.8 Å and 2 Å has a plane (231) which cuts an
intercept 1.2 Å along x-axis. Calculate the intercepts along y- and z-axes.
1.2Å
Solution: Intercept on X-axis, pa = 1.2 Å \p= = 1.
1.2Å
D 2 D 2
L.C.M, D = hp = 2 × 1 = 2.   \ q= = and r = = .
k 3 l 1
778 A Textbook of Engineering Physics

2
\ Y-intercept, qb = × 1.8Å = 1.2 Å.
3
and Z-intercept, rc = 2 × 2Å = 4 Å.

26.16 ATOMIC PACKING

The unit cells are geometric models connecting points which are purported to be the atomic
centres. Atoms are space-filling entities. Therefore, the crystal structure may be viewed,
alternatively, as resulting from the packing of hard spheres in three dimensions. The most
efficient packing of spheres is called closest packing, or simply close packing.
Let us consider the
packing of hard spheres of
identical size. They can be
close packed in a row by
arranging them in contact
with each other. A two
dimensional layer of spheres
can be built in two ways as
shown in Fig. 26.34. Fig.
26.34 (a) shows square
packing and triangular (a) (b)
packing of two rows. A Fig. 26.34
comparison of the two types
suggests that the spheres fit more compactly in triangular packing rather than in square
packing. When more rows are added to build the layer, a close packing such as the one shown
in Fig. 26.34 (b) is obtained.
In a layer of close packed spheres, each sphere is in contact with six neighbours. The lines
joining the centres of the six spheres form a hexagon. Therefore, such a packing is commonly
called hexagonal packing. Hexagonal packing is the most compact packing possible within
a layer. Such a layer is known as a hexagonal close packed layer. Three dimensional close
packed structures
are built by
A A A A A
keeping hexagonal C C C C
close packed layers A
B
A
B
A
B
A
B
A
C B
B
B
B
B
B
B
B
on top of one C
B
C
B
C
B
C
B C-Valley BC BC BC BC
another in a regular A
C
A
C
A
C
A
C
A
B B B B
sequence. A
B
A
B
A
B
A
B
A
C C C C

In practice, C
B
C
B
C
B
C
B
B

identical spheres A A A A A
can be packed in
different ways (a)
B-Valley
(b)
to produce three
Fig. 26.35
dimensional
structures. Different kinds of close packing are theoretically possible but all are combinations
of two different basic structures, namely HCP structure and CCP structure. They are built
starting with a hexagonal packed layer.
Crystal Structures 779

26.16.1 HCP Structure

Fig. 26.35 (a) shows the top view of a hexagonal close packed layer of identical spheres. Let
us designate the spheres as A and hence the layer as A-layer.
Successive
layers of spheres
are added on top of
the A-layer to form
three dimensional
structures. As spheres B B B
have curved surfaces, A A
a regular array of C C C
valleys are formed
between the spheres B B B B
in the A layer. Each
valley is formed where
three spheres are in
contact with each Fig. 26.36
other. Each sphere in
the layer is surrounded by six neighbour spheres and six valleys (Fig. 26.35 a). The valleys
may be divided into two sets of differing shapes and are labeled B and C. Thus, there are three
B-type valleys and three type C-valleys. If a sphere is placed in a B valley, none of the three
adjacent C-valleys can be filled with a sphere. Similarly, if a sphere is placed in a C valley,
none of the three adjacent B-valleys can be filled with a sphere. It means that the B and C
valleys are so close that they cannot be simultaneously occupied. Therefore, when spheres are
placed on top of the A-layer, all of them must roll down into one kind of valley, either B or C.
Thus, the second layer
spheres occupy only
one kind of valley A
leaving the other B B B
kind vacant. It means A A
B B
A A
B
A A A A A A A
that there is only one B
A A A A A A
way of arranging the A A A
second layer. Fig. A
26.36 shows the close (a)
packing of identical
spheres in two layers.
The centres of the Sequence of Hexagonal
spheres in the upper layers prism
layer lie above the (b) (c)

B-valleys. Let us Fig. 26.37

designate this layer as B-layer.
Again in the B-layer, each sphere is surrounded by two sets of valleys marked CCC and
AAA. Now there are two options. The spheres can be positioned in A valleys or C valleys to
form the third layer. If spheres are placed in A valleys, the third layer becomes equivalent to
the first layer at the bottom. It means that it is again an A-layer. Similarly, the fourth layer may
be arranged such that its spheres sit in B valleys of third layer, forming a B-layer. Spheres
may be arranged to form the fifth layer as layer, sixth layer as B-layer and so on. This kind of
780 A Textbook of Engineering Physics

stacking of spheres is essentially a repeating two-layer-close-packing. The packing is known

as AB AB AB ……layer sequence. In this structure, the C valleys are left vacant from the
bottom layer to the topmost layer. Fig. 26.37 (a) shows the schematic of AB AB AB….layer
sequence.
If the centres of six spheres surrounding a particular sphere in the first layer spheres and
those of the third layer spheres are joined, a right prism having a regular hexagon for its base
is obtained (Fig. 26.37 b). Therefore, the unit cell of this structure is a hexagon. In view of
this, the AB AB AB….layer packing is called hexagonal close packing and the resulting three
dimensional structure is known as hexagonal close packed structure or HCP structure.
26.16.2 CCP Structure
If the spheres of
the third layer are
positioned in C A
valleys instead of A A A A
A valleys of second A A A
layer, a different A
B B B
A
C
B B
sequence of layers A A A
B
A A A
is produced. The A A A A B

third layer built on A A A

C valleys is denoted (a) A
as C-layer. In the
C-layer, Each sphere Unit cell
is surrounded by
AAA and BBB Sequence of
layers
valleys. If the spheres (b) (c)
of fourth layer are Fig. 26.38
stacked on A-valleys,
it becomes a repetition of the first layer. Similarly, the fifth layer may be arranged as a
repetition of B-layer, the sixth that of C-layer , the seventh that of A-layer and so on. The
structure is symbolically represented as ABC ABC ABC……..layer sequence. Fig. 26.38
shows the sequence of layers.
The fundamental block of ABC ABC ABC...structure is a cube (Fig. 26.38 a). Therefore,
the packing is known as cubic close packed structure or CCP structure. The atoms are
located at the corners of the cube and at the centres of the faces of the cube. The CCP structure
is in fact FCC structure only.
Atoms which are attracted to each
other by nondirectional forces are likely O
to form close packed structures. Thus,
the noble gases crystallize in CCP
structure, with the exception of helium T
which crystallizes in HCP structure.
Metallic bond is nondirectional and
non saturable. Therefore, metals also Fig. 26.39. Formation of tetrahedral and octahedral
voids in a three dimensional close packed structure
crystallize in close packed structures.
Thus, Cd, Co, Mg, Zn, Ti etc crystallize in HCP structure whereas Ag, Al, Cu, Ni, Pt etc
crystallize in CCP structure.
Crystal Structures 781

26.17 VOIDS
All the available space in a crystal is not filled with atoms. Even in the close packed structures,
it is seen that about 26% of the volume is left vacant. These vacant spaces between the atoms
in the crystal are called voids. They are also known as interstices. Two kinds of voids form
in close packed structures, such as FCC close packed structures. They are tetrahedral voids
and octahedral voids (see Fig. 26.39). A tetrahedral void is formed by a sphere fitting into
the valley formed between three adjacent spheres of a close packed layer, as shown in Fig.
26.40 (a). If the centres of the four spheres are joined, a regular tetrahedron is produced. The
vacant site enclosed by the tetrahedron is known as the tetrahedral void. If the fourth sphere
belongs to the upper layer and is on top of three spheres, an upright tetrahedral void forms.
On the other hand, if the fourth sphere belongs to the bottom layer an inverted tetrahedral void
forms. Thus, in a three-dimensional structure two tetrahedral voids form for every sphere.

(a) Schematic (b) Cutaway view (c) Schematic (d) Cutaway view

Fig. 26.40. Tetrahedral Void (a) Schematic (b) cutaway view; Octahedral void (c) Schematic
(d) cutaway view

An octahedral void is produced when a void formed by three spheres in one layer comes
on the top of the void formed by three spheres in the top layer, as depicted in Fig. 26.40 (c).
When the centres of the six spheres are joined, an octahedron is formed. The space enclosed
is an octahedral void. An octahedral void is surrounded by six spheres. There is no sphere
in the central valley. Octahedral voids are bigger in size and smaller in number compared to
tetrahedral voids. In a three-dimensional close packed structure, one octahedral void forms
per one sphere.
26.17.1 Sizes of the Spheres that Fit into Voids
(i) The tetrahedron in Fig. 26.40 (b) may be used to determine the radius of the sphere that
can fit into the tetrahedral void. A cube ABCDEFGH is formed using the tetrahedron CAEH
(Fig. 26.41a). The H G
point C is the apex 2a 2a
of the tetrahedron D G
and is one of the D a 3a
corners of the cube. C O
The edges AE, a r
AH and EH of the
E R+
A E
tetrahedron are the 2a a R
J
face diagonals of A a B
the faces ABEF,
(a) (b)
ADFH, and EHGF
of the cube. The Fig. 26.41
body diagonal AEGD of the cube is shown separately in Fig. 26.41 (b).
From the Fig. 26.41 (b) it is seen that the Dles AJO and AEG are similar.
782 A Textbook of Engineering Physics
AO AG R+r 3a
\ = \ =
AJ AE R 2a

r 3  3− 2
\ 1 + = \ r =   R = 0.225 R (26.33)
R 2  2 
(ii) For finding the size of the sphere that can fit tightly into an octahedral void, let us
consider the octahedron. In Fig. 26.41(d). B
The plane ABCD passing through four
spheres surrounding the central valley is
shown in Fig. 26.42.
The triangles ABC and ADC are right a a
isosceles triangles. If each side of the
square is a, in the Dle ABC
AC 2a
= A C
BC a 2a

2 R + 2r
\ = 2
2R

or  R + r = 2R
D
r = ( 2 − 1) R
\ or r = 0.414 Fig. 26.42

R (26.34)

26.18 IONIC SOLIDS

Ionic solids are made up of ions of different sizes. The cations are ordinarily smaller than
anions as cations have given up their valence electrons to the anions. The anions that are
relatively larger in size than the cations, form close packed structures and the cations are
accommodated in the voids. The packing depends upon the ionic radii of cations, rC and
anions, rA. Each cation prefers to have as many nearest neighbours as possible. So does each
anion. The coordination number is therefore determined by cation-anion radius ratio rC/rA.
This is called radius ratio effect. This determines the efficiency of packing and forms the
basis of many ceramic (inorganic materials) structures.
Some of the simple ionic solids are those in which there are equal number of cations and
anions. These are often referred to as AX compounds where A denotes the cation and X the
anion. The most common AX crystal structure is the sodium chloride type.
NaCl CRYSTAL STRUCTURE
The NaCl crystal is an ionic crystal. During the formation of the crystal, Na atom loses its outer
electron and becomes a positive ion. On the other hand, the chlorine ion acquires the electron
lost by the Na atom and becomes a negative ion. When the two ions approach each other, at a
certain distance the forces of attraction are balanced by forces of repulsion and a stable system
is produced. Na+ and Cl– ions arranged alternately in a cubic pattern in space so that the
electrostatic attraction between the nearest neighbours is maximum. The resultant structure has
Crystal Structures 783

a coordination number + a
Cl
6. Six anions surround –
Cl
each cation and each +
Na
anion is surrounded
by six cations. Thus,
the cations occupy
the octahedral voids
formed by the anions
(Fig. 26.43 a). Since
the ratio of anions
to cations is 1:1 and
there is one octahedral (a) (b)
void per anion, all Fig. 26.43: NaCl crystal
octahedral voids in the
structure are occupied.
NaCl is an example of face centered cubic (fcc) Bravais lattice. Each cell has 8 corners
and 8 cells meet at each corner. Thus, an ion at a corner of the cell is shared by 8 cells, i.e.
only 1/8 ion belongs to any one cell. Similarly, an ion at the center of a face of the cell is
shared by 2 cells, i.e. only 1/2 ion belongs to any one cell. Since a cell has 8 corners and,
6 faces, it has [(8×1/8) + (6 × 1/2)] = 4 ions of one kind, and similarly 4 ions of other kind.
Thus, there are 4 Na+ – Cl– ion pairs (molecules) per unit cell. The position coordinates of
Na+ and Cl– are as follows.
1 1 1 1 1 1
Na 000 0 0 0
2 2 2 2 2 2
1 1 1 1 1 1
Cl 00 0 0 00
2 2 2 2 2 2
Fig. 26.43 (b) shows a unit cell of NaCl lattice. NaCl lattice can be regarded to be made
up of two fcc sub-lattices, one of Na+ ions having origin at (0,0,0) and the other of Cl– ions
having its origin midway along the cube edge.
Each Na+ ion has 6 Cl– ions as nearest neighbours and similarly, each Cl– ion has 6 Na+
ions. Hence, the coordination number of NaCl is 6, the same as that for, simple cubic lattice.

26.19 DIAMOND CUBIC STRUCTURE

In a diamond crystal the carbon atoms are linked by directional covalent bonds. Each carbon
atoms forms covalent bonds with four other carbon atoms that occupy four corners of a cube
in a tetrahedral structure
(Fig. 26.44 a). The length 0 1
2 0
each bond is 1.53 Å and 3 1
the angle between the 4 4
bonds is 109.5°. The 1 1
0
entire diamond lattice 2 2

is constructed of such 1 3
4 4
tetrahedral units.
1
Diamond exhibits 0 2 0

both cubic and hexagonal (a) (b)

type structure. The Fig. 26.44: Diamond crystal
784 A Textbook of Engineering Physics

diamond cubic (dc) structure is more common. The space lattice of diamond is face centered
(fcc) with a basis of two carbon atoms associated with each lattice point. Fig. 26.44 (b) shows
the position of atoms in the cubic cell of the diamond structure projected on a cubic face. The
fractions denote the height above the base in units of a cube edge. The points at 0 and 1/2
are on the fcc lattice, those at 1/4 and 3/4 are on a similar lattice displaced along the body
diagonal by one fourth of its length. Thus, the diamond lattice is composed of two interpen-
etrating fcc sublattices, one of which is shifted relative to the other by one fourth of a body
diagonal.
In the diamond lattice, each atom has four nearest neighbours with which it forms
covalent bonds. Thus, the coordination number of diamond crystal is 4. The number of atoms
per unit cell is 8.
Silicon, germanium and gray tin crystallize in the diamond structure.
26.20 ZnS STRUCTURE
The structure of zinc sulphide is identical to
the diamond cubic structure. It consists of two
interpenetrating FCC sub lattices which are Zn
occupied by two different elements and are S
displaced from each other by one quarter of the
body diagonal. Zinc sulphide structure results
when Zn atoms are placed on one FCC lattice and
S atoms on the other FCC lattice, as shown in Fig.
26.45. The conventional pattern of this structure is
a cube. There are four molecules per conventional
cell. For each atom there are four equidistant Fig. 26.45: Zinc sulphide crystal
atoms of other kind arranged at the vertices of a regular tetrahedron. Some of the important
compounds which possess this structure are semiconductors such as InSb, GaAs, and CdS.
26.21 POLYMORPHISM AND ALLOTROPY
The phenomenon by which a single substance crystallizes in two or more different forms
under different conditions is called polymorphism and the different crystalline forms are said
to be polymorphous with each other. Polymorphism is found among pure elements as well as
among chemical compounds. Polymorphism of an element is frequently called allotropy.
An element that exists in more than one crystalline form is said to be allotropic and the
forms are referred to as allotropes and the phenomenon is called allotropy. The prevailing
crystal structure depends on the temperature and the external pressure. Pure iron has a BCC
crystal structure at room temperature, which changes to FCC structure at 912°C. Another
example is carbon. Graphite is the stable allotrope at ambient conditions, whereas diamond is
formed at extremely high pressures.
26.22 GRAPHITE STRUCTURE
The crystalline allotropes of carbon are diamond and graphite. The atoms of carbon can form
two types of space lattices, namely diamond and graphite structures. Diamond structure is
discussed in Art. 26.19.
Graphite structure
The second type of packing of carbon atoms occurs in graphite. In graphite each atom is
bound to four neighbours but the forces and the directions of the bonds differ (Fig. 26.46).
Crystal Structures 785

An atom has a strong bond with three

other atoms that lie together with it in
the same layer at angles of 120° and a
weaker bond with a fourth atom lying
in the adjacent layer. This last bond
makes an angle of 90° with the plane
of the layer. The unit cells are right
hexagonal prisms with edges 1.42 Å

3.35Å
and 3.35 Å in length. The distance
between the layers is 2.36 times more
than between the atoms lying in one
layer and therefore the bonds between
layers are weak.
1.42Å
In each layer, each carbon atom
is covalently bonded to three others Fig. 26.46: Graphite
involving sp2 hybrid orbital instead of four as in diamond. Thus, all atoms in a single plane are
linked to give flat hexagons. The hexagons are held together in sheet-like structure, parallel
to one another. The C-C covalent bond distance is 1.42 Å. The distance between the sheets of
layers, however is comparatively larger being about 3.35 Å. This rules out the possibility of
covalent bonding between the layers. Thus graphite has a layer structure.
The difference in structure of the diamond and graphite lattices is the cause of the striking
differences in their physical properties. Diamond is one of the hardest substances in nature;
it can cut glass, as well as any hard rock like granite. Graphite is very soft; it writes on the
paper because its layers slide easily with respect to each other. The diamond is an insulator as
it has no free electrons; but graphite is a conductor. Diamond is transparent whereas graphite
is opaque. The density of diamond is 3500 kg/m3; that of graphite is 2100 kg/m3.
26.23 CRYSTAL STRUCTURE ANALYSIS
Much of our knowledge of the internal structure of crystals has been obtained from x-ray
diffraction experiments. Diffraction of waves occurs when the waves are scattered by a
periodic array of scattering centres separated by distances of the order of a wavelength. A plane
transmission grating is a familiar device used to study diffraction of light. It is made by ruling
thousands of parallel lines on a piece of glass. The lines are opaque and the spaces between
the lines form openings for light to pass through. A grating normally used in laboratories has
15000 lines per inch. The slit width or grating period in the grating comes to about 17,000 Å,
which is about three times the wavelength of light. When light passes through these narrow
slits, the waves spread in all directions. The diffracted waves interfere constructively in a
few specific directions and destructively in all other directions, producing a spatial pattern
of alternate regions of brightness and darkness. The dimension of atoms and the interatomic
spacing in a crystal are of the order of 2 to 5 Å which is of the order of the wavelength of
X-rays. In view of this Max von Laue, the German physicist predicted in 1912 that a crystal
acts as a natural three dimensional grating for x-rays, where the regular periodic lines of
atoms serve the role of parallel ruled lines. His associates, W.Friedrich and P.Knipping duly
tested and confirmed the prediction. Laue was awarded the Noble prize in Physics in 1914.
26.23.1 Laue Method
X-rays produced by an x-ray tube are defined into a narrow beam by a set of lead screens S1
and S2 having pin holes at their centres. A thin crystal C is mounted in the path of the X-ray
786 A Textbook of Engineering Physics

beam and a Photographic

photographic plate
fim is
positioned C
beyond it, as S1 S2
Undiffracted
shown in Fig. beam
26.47 (a).
As the
X-ray beam
penetrates the
crystal C, some (a) (b)

of the rays Fig. 26.47

are scattered
by atoms from their initial direction. The scattered x-rays emerge from the crystal in specific
directions as highly narrow beams and they are intercepted by the photographic film. On developing
the exposed film, a pattern of bright spots corresponding to maximum intensity are observed
(Fig. 26.47 b). They are more commonly referred to as Laue spots. The pattern of Laue spots
is uniquely characteristic of the crystal C. The central bright spot on the film corresponds to
the main beam. A hole is often cut in the film so that the central spot is not recorded. In the
experiment conducted by Friedrich and Knipping, the wavelength of X-rays was determined
from the measurement of angular positions of Laue spots and from the knowledge of the
separation of atoms in the crystal. This method has been subsequently used to determine the
crystal structures using X-rays of known wavelength. However, the analysis of Laue pattern is
mathematically complicated. At present the Laue method is only of historical interest.

26.24 BRAGGS’ LAW

In 1912 W.H.Bragg and W.L.Bragg discovered that X-rays can be regularly reflected by the
cleavage planes of the crystals. The cleavage planes are successive atomic planes in the crystal
(Fig. 26.48).
They are
also known d
as Braggs’
planes. Thus,
a crystal may
d
be regarded
as consisting
of a stack
of several
(a) (b)
parallel
planes of Fig. 26.48. Two different sets of Bragg planes
atoms. Each plane in a given set has the same distribution of atoms. When an X-ray beam is
incident on the crystal, each atom in the crystal scatters a portion of the incident beam. The
scattered waves travel in all directions. It is convenient to consider the net scattering by atoms
in terms of the diffraction (scattering) by the crystal planes. Each family of planes gives rise
to scattering but only a certain family scatters constructively in a given direction. W.H.Bragg
and W.L Bragg derived a relation between the wavelength of X-rays and the angular positions
of the scattered beam and the separation of atomic planes in the crystal.
Crystal Structures 787

Braggs made the following simplifications regarding the diffraction of X-rays from a
crystal:
1. Any crystal may be viewed as a regular stack of atomic planes;
2. The atomic planes act like semitransparent mirrors for X-rays;
3. X-rays reflected from the successive atomic planes would interfere constructively or
destructively depending upon the path difference between the rays;
4. Whenever the path difference between the rays is an integer multiple of wavelength
(λ), the rays interfere constructively and produce a bright spot in that direction.
Thus, the complex phenomenon of diffraction of X-rays by the atoms was converted
into the problem of reflection of X-rays by the parallel atomic planes. Hence, the words
‘diffraction’ and ‘reflection’ are mutually interchangeable in Braggs’ treatment. Based on these
considerations, Braggs derived a simple mathematical relationship that serves as a condition
for the Bragg reflection to occur. This condition is known as Braggs’ law.
A
A C
1 F
1
D C
2 D F
H G 2
   B 
M  N

E B d  
G H
P Q
E d sin 
R S
(a) (b)

Fig. 26.49. Determination of path difference between X-rays diffracted by consecutive lattice planes

Let us consider a set of parallel atomic planes with interplanar spacing d. We represent
the row of atoms as a single horizontal line, as shown in Fig. 26.49. Let a parallel beam of
monochromatic X-rays of wavelength l, represented by the parallel lines AB and DE, be
incident on these planes at a glancing angle q (Fig. 26.49 a). The scattered beam emerges
along BC and EF. The contributions of two adjacent atoms in the same plane are considered
in the Fig. 26.49 (a).
The rays BC and EF are coherent and reinforce each other, if they are in phase. It requires
that the path lengths BK and EL are equal. They will be equal when q = q′. This is the
condition of reflection.
We consider next the contributions of two adjacent atoms to the reflection from successive
planes MN and PQ are considered (see Fig. 26.49 b). The path difference Δ between the
reflected rays BC and EF is
Δ = GE + EH (26.35)
∠ABG = q + a = 90° (26.36)
BE is normal to the plane MN. Therefore,
∠MBE = a + ∠GBE = 90° (26.37)
From equ. (26.36) and (26.37) we get
∠GBE = q
788 A Textbook of Engineering Physics

Similarly, ∠EBH = q
In the Δle BGE, BE = d
\ GE = d sin q
Similarly, in the Δle EBH, EH = d sin q
\ The path difference Δ = GE + EH = d sin q + d sin q = 2d sin q (26.38)
The rays BC and EF will constructively interfere only when Δ = ml, where m = 1,2,3…
\ The condition for reinforcement of scattered waves is
2d sin q = ml (26.39)
The above equation is called Bragg’s equation or Braggs’ law.
The meaning of the above equation is that the reinforcement of reflected waves will
take place only at certain values of q, corresponding to specific values of l and d. In these
directions the resultant wave has the maximum amplitude and produces bright spots on the
photographic plate placed in their path. On the other hand, at other angles, the scattered waves
may not be in phase with each other and hence their amplitude will be zero, leading to dark
spots in those directions.
Order of reflection:
The reflection angle q is given by
−1  mλ 
q = sin   (26.40)
 2d 
For a fixed wavelength, l, the angle q depends dhkl, the interplanar distance.
When the lattice planes of indices (hkl) give rise to x-ray reflection, the path difference
between the rays reflected from successive planes is one wavelength. For example, if the
reflection from (100) planes of the lattice occurs, say, at an angle q1, then the path difference
between the reflections will be l and the order of reflection m = 1. When the path difference
between the reflected rays is 2l, the reflection will occur at an angle q2 and the Bragg
equation is satisfied with m = 2. Thus, as q is increased gradually, a number of positions
will be found at which the reflections are intense. These positions correspond to m = 1,2,3…
values. The diffraction lines appearing for m = 1, 2 and 3 are called first, second, third order
diffraction lines respectively. Thus, m denotes the order of reflection. The highest possible
order is determined by the condition that sin q cannot exceed unity.
If the value of q is determined experimentally, and knowing the wavelength l, the
interplanar spacing ‘d’ can be determined with the help of Bragg law.
Example 26.8. X-rays of unknown wavelength give first order Bragg reflection at glancing
angle 20° with (212) planes of copper having FCC structure. Find the wavelength of X-rays,
if the lattice constant for copper is 3.615 Å.
a
Solution: The interplanar spacing is given by d =
h + k2 + l2
2

3.615Å
\ d= = 1.205 Å.
2 + 12 + 22
2
Crystal Structures 789

According to Braggs’ law

2d sin q = ml.
2d sin θ 2 × 1.205 Å × sin 20°
\ l= = = 0.824 Å.
m 1

26.25 BRAGGS’ SPECTROMETER

W.H.Bragg and W.L.Bragg devised an X-ray spectrometer in which a crystal is used as a
reflection grating. It is used to measure glancing angle q.
Construction: Braggs’ spectrometer is very much similar in construction to the optical
spectrometer. The schematic diagram of the Braggs’ spectrometer is shown in Fig. 26.50.
X-rays from an X-ray tube are collimated into a narrow beam by two slits S1 and S2 cut in lead
plates. The beam is then allowed to C
be incident at a glancing angle on S1 S2

the face of the crystal D, which is A

mounted in wax on the turntable of  D
the spectrometer. The turntable is
V S3
capable of rotation about a vertical
axis passing through its center and
the position of the crystal can be
read from the circular scale, C.
T.
Most of the incident beam passes H.
straight through the crystal. Some

E
of the X-rays are however scattered
by the regularly arranged atoms
in different crystal planes. The Fig. 26.50. The schematic diagram of the Bragg’s
scattered X-rays can be regarded spectrometer
as having been reflected from the
crystal planes rich in atoms. The reflected X-ray beam enters an ionization chamber carried by
the spectrometer arm, which is capable of rotation about the same axis as the turntable. The
turntable and the arm are so linked together that when the turntable rotates through an angle
q, the arm turns through an angle 2q. In this way, the X-ray beam is always reflected into the
ionization chamber whatever its incidence
angle is at the crystal face. The ionization A1
current produced by the reflected beam is A2
measured by a sensitive electrometer E or
Ionization current

A3
recorded on a photographic plate.
Working: Initially, the single crystal
under investigation is mounted on the
turntable such that the glancing angle q =
0° and the ionization chamber is adjusted to
receive X-rays. Then the crystal is moved n= 1 2 3
in small angles and the corresponding Incident angle
deflection obtained in the electrometer is
noted down. A graph is plotted for glancing Fig. 26.51. A typical diffraction spectrum
angle versus intensity of diffracted X-rays. produced by a monochromatic X-rays scattered
from a crystal.
790 A Textbook of Engineering Physics

X-ray spectrum: Fig. 26.51 shows a typical diffraction spectrum produced by a

monochromatic X-rays scattered from a crystal. The diffraction spectrum is a graph plotted
between the intensity of ionization and the glancing angle. It is seen from the graph that the
intensity of ionization current increases abruptly for certain values of glancing angle. The
peaks A1, A2, A3 etc. in intensity occur whenever the Braggs’ equation is satisfied. If q1, q2
and q3 are the angles of incidence where intense peaks occur, then they correspond to the
different orders of reflection with n = 1, 2 and 3 respectively for a given wavelength. Using
Braggs’ law
2d sin q1 = l, 2d sin q2 = 2l and 2d sin q3 = 3l
That is, 2d sin q1 : 2d sin q2 : 2d sin q3 = l : 2l : 3l
\ sin q1 : sin q2 : sin q3 = 1 : 2 : 3 (26.41)
The above ratio indicates that the peaks correspond to first, second and third order
reflections respectively.
It may be seen from Fig. 26.51 that
• the intensity of the reflected beam decreases as the order of the spectrum increases.
• The intensity of the reflected beam never falls to zero. It reaches down only to a
minimum value. This is due to the presence of continuous spectrum.
At a given time more than one set of planes give rise to reflections. Usually, the reflections
are produced due to principal planes in the crystals.
26.25.1 Determination of Lattice Constant:
Bragg’s law may be used to find the interplanar spacing, if the wavelength of X-rays is
known. If we substitute the value of d in terms of the Miller indices of the planes in the
relation connecting the interplanar spacing, d and the lattice constant, a, we can determine the
lattice constant of the crystal. Thus, in case of a cubic crystal, the relation between d and a is
given by
a
dhkl = (26.42)
h + k2 + l2
2

Knowing the value of d, the lattice constant a can be calculated from the above equation.
26.25.2 Determination of Crystal Structure
The Braggs’ method of determining crystal structure consists in finding out the interplanar
spacing, dhkl of the various sets of planes rich in atoms. From the knowledge of dhkl and from
the relative intensities of diffracted beams of different orders, the crystal structure can be
deduced. A
For example, let us A
consider a crystal of a simple B d1 d1 D
cubic system. In the unit cell 45°
d3 C
the atoms are located at the E L
d2
L
d1
corners of the cell. It may be d2
P
d3
seen that three sets of planes F H
d2
K
are rich in atom. E K
G
(i) Referring to Fig. 26.52,
Fig. 26.52
we see that the set of planes
ABFE, CDHG, ADHE, BCGF, ABCD and EFGH are all alike. Let the d2 distance between
the consecutive planes be d1. These planes are known as (100) planes.
Crystal Structures 791

(ii) The second set of planes consists of parallel planes like ADGF, inclined at an angle
of 45° to the planes mentioned in (i) Let be the spacing between the second set of planes, as
shown in Fig. 26.52 by CP. Note that a plane parallel to ADGF contains the line BC. Then,
from the DCDP, we find that
d2 1
= sin 45° =
d1 2

d2 1
or = (26.43)
d1 2
These planes are known as (110) planes.
(iii) The third set of planes consists of planes like AFH. Let us draw EK perpendicular to
FH and join AK to complete the triangle AEK. The perpendicular EL denoted by d3 represents
the distance between the plane AFH and a parallel plane passing through E. In DAEK,
AE = d1, EK = d2 and AK = d12 + d 22 . Further,
d3 d2
sin A = and also sin A =
d1 d12 + d 22

d1d 2 d1d 2 d1
\ d3 = = = (26.44)
d12 + d 22 2d 22 + d 22 3

These planes are known as the (111) planes. The interplanar distances are in the ratio
d1 : d2 : d3 = 1 : 1 / 2 : 1 / 3 (26.45)
Using a KCl crystal, Bragg found the values of q for reflections from the faces ABCD,
ADGF and AHF respectively as
q1 = 5°22′, q2 = 7°36′ and q3 = 9°25′.
Hence for the first order spectrum,
d1 : d2 : d3 = 1/sin 5°22′ : 1/sin 7°36′ : 1/sin 9°25′
= 1/0.0936 : 1/0.1323 : 1/0.1636
= 1/1 : 1/1.410 : 1/1.748
These ratios, within the limits of experimental error, are
d1 : d2 : d3 = 1 : 1 / 2 : 1 / 3
Hence it is concluded that KCl crystal has a simple cubic structure. On the other hand,
NaCl crystal which is very much similar to KCl is found to have the ratio of interplanar
distances as
d1 : d2 : d3 = 1 : 1 / 2 : 2 / 3
which agree with the theoretical values for a face centred cubic system. The observed dissimi-
larity between the conclusions on these similar crystals can be resolved once we determine
the relative intensities of the diffracted beams. From such a study, it has been confirmed that
both the KCL and NaCl crystals have face centred cubic crystal structure.
792 A Textbook of Engineering Physics

Example 26.9. A beam of X-rays λ = 0.842 Å is incident on a crystal at a glancing angle of

8°35′ when the first order Braggs’ diffraction occurs. Calculate the glancing angle for 3rd
order diffraction.
Solution: According to Braggs’ law, 2d sin q = ml.
mλ 1 × 0.842 Å
d= = = 2.826 Å
2 sin θ 2 × sin 8°35′
−1  mλ  −1  3 × 0.842 
q = sin   = sin   = 26.55°.
 2d   2 × 2.826 
Example 26.10. X-rays of wavelength 1.5 Å are incident on Na Cl crystal having a grating
spacing of 2.8 Å. What is the highest order that the crystal can diffract?
Solution: According to Braggs’ law, 2d sin q = ml. m will be maximum if sin q = 1
2d sin θ 2 × 2.8Å × 1
\ m= = = 3.73.
λ 1.5Å
It means that the highest order of diffraction is 3.
26.26 POWDER CRYSTAL METHOD
There are many materials for which it is impossible to obtain single crystals of required
size. For such materials powder photography is highly suitable. One from of the powder
photography is known as Debye-Scherrer method invented by P.Debye and Scherrer.
Fig. 26.53 shows the experimental arrangement. It consists of a cylindrical camera called
Debye-Scherrer camera, whose width is smaller as compared to its diameter. The material
Film
Specimen Exit for
X-rays 2

X-ra y beam
Film
Specimen

Collimator Point where

incident beam 2 = 0°
enters (2 = 180°)

(a) (b)
Fig. 26.53

under investigation is crushed into fine grain powder and compressed into a thin rod or packed
into a small capillary tube. A strip of photographic film wrapped in opaque paper is mounted
round the inside of the cylindrical drum of the camera. The specimen is positioned vertically
at the centre of the drum. A narrow beam of monochromatic x-rays enter and leave the drum
through the apertures on the opposite sides of the drum. Each crystallite has the same system
of atomic planes. Some of the crystallites are bound to lie with their planes at glancing angle
q to the incident ray such that Braggs’ condition is satisfied. Each such crystal produces a spot
on the photographic plate. As the specimen contains a large number of crystallites oriented
in all directions, almost all the possible values of q and d are available. Also, for a particular
838 A Textbook of Engineering Physics

be represented as the product of two functions: a free particle wave function and a periodic
function u (x) that has the same period as the lattice. Thus
y(x) = u(x)eikx (29.5)
with u(x) = u(x + a) (29.6)
The wave functions of the above type are known as Bloch functions which change
periodically with increasing x.
29.3.3 Energy Bands
We substitute the above wave functions into the Schrodinger equation and solve it in the usual
way. When we apply the periodic boundary condition, we get the following expression.
maV b sin α a
2 0 ⋅ + cos α a = cos k a (29.7)
 a
2mE
where a=

Equ. (29.7) provides the allowed solutions to the Schrodinger equation. As the relation
involves trigonometric functions, only certain values of a are possible. The right hand side
of equation (29.7)
is cosine function Allowed
and can take values band
Energy E

only between –1 Forbidden

Gap
and +1. Therefore,
the left-hand side Allowed
of the equation is band
restricted to vary Forbidden
between those two Gap
limits. Hence, only Allowed
band
certain values of –3 –2 – 0  2 3
a are allowed. It a a a a a a

means that energy k

Fig. 29.5. Electron energy, E versus wave number, k plot for a solid
E is restricted to lie
within certain ranges.
This concept is best understood by drawing the plot of energy E as a function of the wave
number, k. The plot is shown in Fig. 29.5. The parabolic relation between E and k obtained in
case of free electron is interrupted at certain values of k, as shown by the broken curve.
Fig. 29.5 shows discontinuities in E. The discontinuities occur at ka = ± np i.e., at
π 2π 3π
k = ± ,± ,± , .... (29.8)
a a a
The origin of the allowed energy bands and forbidden gaps is clear from Fig. 29.5.
29.4 ENERGY BAND STRUCTURE OF A SOLID
A crystal (i.e., solid) consists of an enormous number of atoms arranged in a regular periodic
structure. The extent of energy level splitting in the solid depends on the nearness of atoms
in it. Let us assume that N identical atoms form the crystal. The energy levels of the isolated
atoms are shown in Fig. 29.6(c). All the N atoms have identical sets of energy levels. The
electrons fill the energy levels in each atom independently. Fig. 29.6 (b) shows an atom sitting
Band Theory of Solids 839

at the origin of the coordinate system. Now let us imagine that other atoms approach this
atom along the three axial directions and halt at the distance a0, which is the lattice constant
of the crystal. As the atoms approach, a continuously increasing interaction occurs between
the atoms. Each of the energy levels splits into many distinct levels and form energy bands,
as shown in Fig. 29.6 (b). Fig. 29.6 (a) depicts the effect of slicing of Fig. 29.6 (b) at a0 and
it represents the energy band structure of the crystal. It is seen that corresponding to each
allowed energy level of an isolated atom, there forms an allowed energy band; and that the
allowed energy bands are separated by forbidden bands of energy.
E E E
N Energy levels
2 N States 3s
Conduction Band No Electrons
Actual spacing
3 N Energy levels
6 N States
Valence Band 2p
6 N Electrons
N Energy levels
2 N States 2s
2 N Electrons
N Energy levels
1s
2 N States
2 N Electrons
a0
Distance
Interatomic distance
(a) (b) (c)
Fig. 29.6. Energy level splitting in a solid as a function of interatomic distance (a) Energy band
structure of the solid corresponding to the actual spacing of atoms in the solid. (b) Energy level
splitting as a function of distance. (c) Discrete energy levels in an isolated atom.

The degree of splitting of energy levels depends on their depth in the atom. The electrons
in outer shells screen the core electrons belonging to inner shells. Consequently, the energy
levels of inner shell electrons are split to a lesser degree. They form narrow core bands. They
are always completely filled and do not participate in electrical conduction. In contrast the
energy levels of valence electrons are split more and form wider bands.
In general, N interacting atoms cause an energy level to split into (2l + 1) N levels. Thus,
s-level (l = 0) splits into N levels whereas the p-level, consisting of three sublevels px, py
and pz, splits into 3N levels. Thus, in a solid each level of an individual atom splits into
(2l + 1) N number of levels where N is the number of atoms in the system. Consequently, the
maximum electron capacity of an s-band is 2N electrons whereas the capacity of a p-band is
6N electrons.
While occupying an energy band, electrons start from the lowest energy level in the band
and fill the levels one after the other in the ascending order of energy. When 2N electrons
occupy the N levels available in the band, the band is said to be completely filled. In case
of non-availability of 2N electrons, the energy band gets partially filled. When there are no
electrons to occupy the levels, the energy band remains vacant.
The width of an allowed or forbidden energy band is generally of the order of a few
electron-volts. As N is very large, the energy separation between successive energy levels
in an allowed band is very small and is of the order of 10–27eV. At room temperature, the
kinetic energy of the electrons of the order of kT(≈ 0.026 eV) which is very large compared
to the energy level separation in an allowed band. Consequently, electrons can easily move
into higher vacant levels within the allowed energy band either due to thermal energy or due
to a small externally applied electric field. On the other hand, electrons cannot jump across a
840 A Textbook of Engineering Physics

forbidden band under normal thermal energy possessed by them or due to an applied electric
field. High temperatures are required to cause inter-band electron transitions.
29.5 ELECTRICAL CONDUCTION FROM THE VIEW POINT OF BAND
THEORY
In a solid, the allowed values of electron energy are distributed into bands (Fig. 29.6), each
band consisting of a sequence of closely spaced discrete energy levels arranged in a manner
akin to the rungs of a ladder. The electrons are distributed in the energy levels according to
the Pauli exclusion principle. The motion of an electron corresponds to its transition from
a lower energy level to an upper vacant energy level. This implies that the following two
conditions are to be fulfilled for electrical conduction to take place in a solid:
(i) There should be free electrons available in the solid.
(ii) Vacant energy levels should be available immediately above the levels occupied by
free electrons.
If a band has vacant energy levels but is devoid of electrons, there would be no carriers to
move through the vacant levels when energy is supplied to the solid from a source such as a
battery. Hence current does not flow through the solid.
On the other hand, if all the energy levels within a band were completely occupied by
electrons, there would be no energy level to which an electron can jump. Therefore, even
though the energy is supplied to the solid from a source such as a battery, the electrons cannot
acquire energy and electrical conduction cannot occur in the solid.
If an energy band is partially filled, then the electrons will have vacant upper energy
levels into which they can jump. On acquiring energy from the electric field applied across the
solid, the electrons move into successive upper energy levels and cause electrical conduction.
Thus, partially filled energy band is required for electrical conduction in a solid.
29.6 ENERGY BAND DIAGRAM
Conduction
An energy band diagram is a graphic represen- band
tation of the energy levels associated with top energy
band and the next lower energy band in a solid.
The energy band diagram shows two bands with a Electron energy Eg
gap in-between (see Fig. 29.7). The upper band is
called the conduction band and the lower energy
band is called the valence band. These two bands Valence
band
are separated by a forbidden gap. This energy gap is
more popularly called band gap and is denoted by
Distance
the symbol Eg. The conduction band corresponds to
Fig. 29.7. Energy band diagram
the energy values of free electrons that have broken
their valence bonds, and hence have become free to move in the crystal. The bottom of the
conduction band represents the smallest energy that the electron must possess to become free.
Only the free electrons can move in the crystal under the influence of the externally applied
electric field. Hence, these electrons are called conduction electrons and the energies of such
electrons constitute the conduction band. The band showing the energy values of valence
electrons that are engaged in covalent bonding is called the valence band.
29.7 CLASSIFICATION OF SOLIDS
The concept of energy bands helps us in understanding the division of solids into three
groups. The nature of the energy bands determines whether the solid is an electrical conductor
Band Theory of Solids 841

or insulator. According to the band theory, the electrical conductivity a solid is characterized
by the energy gap Eg separating the outermost energy bands namely, the valence band and the
conduction band. The ability of electrical conduction is decided by the order of magnitude of
the energy gap Eg.
E E

Empty levels
Empty levels
Overlapping
N/2 Region
Filled levels
Filled levels

(a) (b)

Fig. 29.8. Energy band formation in a conductor (a) Half filled conduction band.
(b) Empty upper band overlaps on a totally filled lower band.

In some solids, an upper vacant band overlaps E

the valence band or the valence band itself is
half-filled, as shown in Fig. 29.8. It means that
electrons in the valence band have easy access to
levels in the upper vacant band. For this reason, Conduction band
empty
very large numbers of electrons are available for
conduction, even at extremely low temperatures.
When electric field is impressed across the solid, Eg  2 eV
electrons readily jump into upper unoccupied
energy levels of the vacant band and current flows
Valence band
in a large measure in the solid. Therefore, these filled
solids exhibit good electrical conductivity and are
called conductors.
In some solids the band gap is narrow and Distance
of the order of 2 eV or less, as shown in Fig. Fig. 29.9. Energy band structure of a
29.9. Acquisition of small amounts of energy semiconductor
from the vibrations of atom can raise electrons
from the valence band to the conduction band. E
The conduction band is then partially filled. If a
potential is applied across the material, it causes
the electrons in the conduction band to move to Conduction band
empty
upper levels. As a result, current flows in a modest
measure in the solid. Such solids are called
semiconductors.
Eg  3 eV
Some solids (Fig. 29.10) have band gaps that
are very wide (Eg > 3 eV). It would require the
acquisition of very large amounts of energy to
Valence band
cause an electron to jump from the valence band filled
to the conduction band. Very few electrons can get
this large amount of energy to jump from valence
Distance
band to conduction band at ambient temperature.
Hence, there are very few electrons in the Fig. 29.10. Energy band structure of an
insulator
conduction band. When a voltage is applied across
842 A Textbook of Engineering Physics

the solid, negligible current flows and the solid exhibits very low electrical conductivity.
These solids are called insulators.
29.8 ENERGY BAND DIAGRAMS FOR SOME TYPICAL SOLIDS
29.8.1 Lithium
Let us consider the element lithium E E E
belonging to Group I in the periodic Empty levels N Energy levels
2s
table. The electron configuration of Filled levels
N/2
N Electrons
lithium atom is 1s2 2s1. The 1s shell is
closed and there is only one electron
at the 2s level. In solid lithium, 1s and N Energy levels 1s
2s bands form corresponding to the 2 N Electrons
1s and 2s levels, as illustrated in Fig.

Spacing
Actual
29.11.
Both 1s and 2s bands have N Distance a0 r
levels each. The 1s band is completely Fig. 29.11. Energy level splitting and energy band
filled as 2N electrons occupy N configuration in lithium solid showing half-filled 2s band.
energy levels whereas the 2s band is
half-filled because the N available electrons fill N/2 lower levels in the band leaving the upper
N/2 levels vacant. In general, the solids of Group-I elements form such half-filled energy
bands at the top and therefore belong to the group of conductors.
29.8.2 Beryllium
Let us next consider the case of alkaline E E E
earth elements of Group-II. The first Empty levels 3 N Energy levels 2p

element in this group is beryllium. Its N Energy levels 2s

Filled levels
electron configuration is 1s2 2s2. From 2N Electrons

the electron configuration, it is expected

Spacing
Actual

that beryllium solid would be an insulator.

However, it is known to be a conductor.
The reason is that the upper vacant 2p band N Energy levels 1s
2N Electrons
overlaps the lower completely filled 2s
a0
band leading to the formation of a partially Distance
Fig. 29.12. Energy level splitting and energy
filled hybrid band, as shown in Fig. 29.12.
band configuration in beryllium solid showing
In general, solids of Group-II elements overlapping of completely filled 2s band and
exhibit such partially filled bands, and vacant 2p band.
therefore belong to the conductors group.
29.8.3 Energy Band Diagrams for Silicon and Diamond
Silicon belongs to Group IV elements in the periodic table. The electron configuration of
silicon atom is 1s2 2s2 2p6 3s2 3p2. It is seen that the inner K and L shells are closed and the
corresponding bands would be completely filled. In the outer subshells 3s and 3p, 3s-subshell
is closed. The 3p sub-shell is partially filled. Hence it is expected to behave as a good
conductor. But because of formation of a hybrid band, which later branches out, the Si solid
behaves as a semiconductor.
In the crystal formation process, when the atoms are very far apart, as at position ‘d’ in
Fig. 29.13, the electrostatic interaction among them is negligible. Consequently, the electronic
Band Theory of Solids 843

energy levels of the crystal will be 2N Levels 4N Levels

4N States 8N States
the same as those of isolated atoms.

Valence Energy Conduction

No Electrons 4N Electrons
As the separation between atoms E

band
E 3N Levels
6N States
decreases, the 3s and 3p levels Empty Levels 2N Electrons
split and two bands are formed, as 3p

gap
shown at position ‘c’ in Fig. 29.13. Eg Eg
3s
The band corresponding to 3s

band
Filled Levels 2N Levels N - Levels
level has N energy levels and the 4N Electrons 2N States
2N Electrons
bad corresponding to 3p level has

Spacing
Actual
3N levels. 2N electrons occupy N
levels in 3s-band and 2N electrons 2s
1s
occupy N levels in 3p-band. It may 0
a a0 b c d

now be noted that there is an energy Distance Interatomic Distance

gap between the two bands. The Fig. 29.13. Energy level splitting and band formation in
energy gap is seen to decrease with crystals of Group IV A elements
the decrease in atomic spacing. At
position ‘b’ in Fig. 29.13 the two bands merge and form a composite band. The 3N upper
levels merge with N lower levels, giving rise to a total of 4N levels. These levels have to be
occupied by the 4N electrons available in total, and so the lowermost 2N levels are filled.
When the atomic distance in our imaginary crystal is further reduced, the interaction among
the atoms becomes very strong. Beyond the lattice spacing ‘b’ in Fig. 29.13, we find that the
composite band branches out and once again two bands are formed, separated by a forbidden
gap, Eg. The significant point is that the 4N energy levels are equally divided between the
two branches. There is an equal distribution of levels, 2N in each, in the two bands. The
4N electrons available in total at 3s and 3p levels, now occupy the lower energy band and
leave the upper band vacant. The lower band constitutes the valence band and the upper band
the conduction band. This is the situation at the actual spacing ‘ao’ in the silicon crystal.
At position ao the two bands are not widely separated from each other. The value of Eg at
0 K is 1.12 eV. At normal temperatures, a significant number of electrons will be thermally
excited from valence band to conduction band. The electrons excited to conduction band
respond to the external voltage and produce a modest flow of current. Thus, Si behaves as a
semiconductor.
Diamond
It is evident from Fig. 29.13 that the energy gap between the two branches goes on increasing
with decreasing atomic distance. At the interatomic distance corresponding to line at ‘a’ in Fig.
29.13, the distance between the two bands becomes considerably large. In case of diamond
the two bands are separated by 5.47 eV. Even at high temperatures, the thermal energy would
be insufficient to excite enough number of electrons to the conduction band. Because of the
non-availability of electrons in the conduction band electrical conduction cannot take place in
the material and hence diamond behaves as insulator.

29.9 ENERGY BAND STRUCTURE OF A CONDUCTOR

Conductors are characterized by a single energy band, namely conduction band which contains
more energy levels than free electrons. At the temperature 0K, electrons occupy the lower
energy levels in the conduction band up to a certain energy level called the Fermi level EF.
844 A Textbook of Engineering Physics

29.9.1 Fermi-Dirac Distribution Function

We are next interested in knowing how electrons are distributed among the various energy
levels in the conduction band at a given temperature. We cannot apply Maxwell-Boltzmann
distribution to electrons because (i) they obey exclusion principle and (ii) they are indistin-
guishable particles. The statistical distribution function applicable to quantum particles is the
Fermi–Dirac distribution function.
The probability that an electron occupies an energy level E at thermal equilibrium is
given by
1
f(E) =
1 + exp[ E − EF ) / kT ]
where EF is known as Fermi level. In general EF may or may not correspond to an energy
level but it provides a reference with which other energies can be compared.The function f(E)
is known as Fermi factor.
The above equation is known as Fermi-Dirac equation or Fermi-Dirac distribution
function. Note that the probability of the electron to occupy the energy level E increases with
temperature. We first discuss about the distribution function and the related topics with reference
to conductors. We shall find later that these concepts are equally applicable to other cases.
Example 29.1. Evaluate the Fermi function for energy kT above the Fermi energy.
1
Solution. The Fermi function is given by f(E) = ( E − EF )/ kT
1+ e
1 1 1 1
If (E – EF) = kT, then f (E) = ( E − EF )/ kT
= 1
= = = 0.269.
1+ e 1+ e 1 + 2 .78 3 .78

Example 29.2. In a solid, consider the energy level lying 0.01 eV below Fermi level. What is
the probability of this level not being occupied by an electron?
Solution. (EF – E) = [EF – (EF – 0.01)] = 0.01 eV and kT = 0.026 eV at T = 300 K
The probability of an energy level E not being occupied by an electron is given by
[1 – f(E)].
1 1 1 1
[1 – f(E)] = 1 − ( E − EF )/ kT
= ( EF − E )/ kT
= 0.01eV / 0.026eV
=
1+ e 1+ e 1+ e 1 + e0.385
1
= = 0.405
1 + 1.47

29.9.2 Fermi Level

The occupancy of the energy levels by electrons in conductors is described by the Fermi-Dirac
distribution function.
1
f(E) = (29.9)
1 + exp[( E − EF ) / kT ]
We distinguish two situations - one at absolute zero and the other at higher temperatures.
Case 1: T = 0 K
Fig. 29.14 (a) depicts the conduction band of a conductor at 0K. At absolute zero,
electrons occupy energy levels in pairs starting from the bottom of the band up to an upper
level designated as EF, leaving the upper levels vacant. Fermi level can be, therefore, defined
as the uppermost filled energy level in a conductor at 0K. Correspondingly, Fermi energy
Band Theory of Solids 845

is defined as maximum energy that a free electron can have in a conductor at 0K. To use an
analogy, the electron distribution in the conduction band can be likened to water at rest in a
container. The Fermi level corresponds to the top surface of water. The Fermi function at 0K
is shown in Fig. 29.14 (b).
Let us now apply equ. (29.9) to the solid taking the value of T as 0 K.
(i) For energy levels E lying below EF, E < EF, (E – EF) is a negative quantity.
1 1 1
\ f(E) = − ( E − EF )/ 0
= −∞
= =1
1+ e 1+ e 1+ 0
f(E) = 1 indicates that all the energy levels lying below the level EF are occupied.
(ii) For energy levels located above EF, E > EF, (E – EF) is a positive quantity.
1 1 1 1
\ f(E) = ( E − EF )/ 0
= ∞
= = =0
1+ e 1+ e 1 + ∞ ∞

The result ƒ(E) = 0 implies that all the levels above EF are vacant at T = 0K.
(iii) For E = EF, the quantity (E – EF) = 0.
1
\ f(E) = = indeterminate
1 + e0 / 0
E
E
The above result
implies that the occupancy
of Fermi level at 0K ranges
EF N/2
from zero to one.
EF
Case 2: T > 0K
On heating the
conductor, electrons 0
are excited to higher Distance f (E) 1
(a) (b)
energy levels. In general,
EF >> kT. Therefore, for Fig. 29.14
most of the electrons lying
deep in the conduction band, the thermal energy is not sufficient to cause a transition to an
upper unoccupied level. At normal temperatures, only those electrons occupying the energy
levels near the Fermi level can be thermally excited. These levels make up a narrow band of
width kT directly adjacent to the Fermi level. Therefore, upon heating the solid, electrons
having energy a little below EF,
jump into levels with energy E E
somewhat above EF and a new
energy distribution of electrons is kT kT
obtained. EF
kT EF
kT
Thus, as a result of thermal
excitation, the probability of
finding electrons in the levels
0 0.5 1
immediately below EF will Distance f(E)
decrease. On the same hand, the (a) (b)
probability of finding electrons in Fig. 29.15
846 A Textbook of Engineering Physics

the levels immediately above EF increases. This fact is reflected in the graph (Fig. 29.15 b) as
a blurring of the step plot.
At T > 0K, if we consider an electron at Fermi level, then E = EF.
1 1 1
\ f(E) = 0 / kT
= =
1+ e 1+1 2
This implies that the probability of occupancy of Fermi
level at any temperature above 0K is 0.5 or 50%. Now we can
define Fermi level as the energy level, which has a probability
of occupancy of 0.5. An operational definition of Fermi T=0K

Energy of the level E

C
energy can be given now. Fermi energy is the average energy 300 K

possessed by electrons participating in conduction in metals at 1000 K

temperatures above 0K. 1500 K

29.9.3 Effect of Temperature on Fermi Function

The Fermi-Dirac distribution curves for different tempera-
tures are shown in Fig. 29.16. At T = 0 K, there is an abrupt 0 0.5 1
jump in the value of ƒ(E) from 1 to zero at EF. At T > 0K the f(E)
change is gradual. The higher the temperature, more gradual is Probability of filling the level
the change. Fig. 29.16
It is seen from the curves for different temperatures in Fig.
29.16 that they all pass through a crossover point C, at which the probability of occupancy is
0.5. This is due to the fact that ƒ(E) has a value of 0.5 for any temperature greater than 0K.
We may deduce from the curves that Fermi energy EF is the average energy possessed
by electrons that participate in conduction process in a conductor at temperatures above
absolute zero.
Example 29.3. The Fermi level for potassium is 2.1 eV. Calculate the velocity of the electrons
at the Fermi level.
1 2
Solution: EF = mυF
2
2 EF 2 × 2.1 × 1.602 × 10−19 C ⋅ V
\ υ2F = = = 0.74 × 1012 m2/s2
m 9.10 × 10−31 kg
uF = 8.6 × 105 m/s.
Example 29.4. The Fermi level of silver is 5.5 eV. Calculate the fraction of free electrons at
room temperature located up to a width of kT otn either side of the Fermi level.
Solution. The number of electrons that occupy levels above EF at a temperature T is
proportional to kT. Therefore, the fraction of electrons that occupies levels higher than EF is
given by
kT 0.026 eV
= = 0.0047
EF 5.5 eV
Similarly, the fraction of electrons that are deprived of partners ≅ 0.0047
\ The fraction of free electrons that is located up to a width kT on either side of EF
= 2 × 0.0047 = 0.0094 ≈ 0.01
Example 29.5. At what temperature we can expect a 10% probability that electrons in silver
have an energy which is 1% above the Fermi energy? The Fermi energy of silver is 5.5 eV.
Band Theory of Solids 847

Solution. Given that the electron energy is E = EF + 1%EF.

5.5 eV
\ E – EF = 1%EF = = 0.055 eV
100
1
Also f(E) =
1 + e( E − EF )/ kT
E − EF 1
Putting = x, f(E) =
kT 1 + ex
1
As f(E) = 10% = 0.1, we get = 0.1 or x = 2.197
1 + ex
E − EF
\ = 2.197
kT
E − EF 0.055 eV
or T= = = 290 K.
2.197 × k 2.197 × 8.61 × 10−5 eV / K
Example 29.6. Find the temperature at which there is 1% probability that a state with energy
2 eV is occupied. Given that Fermi energy is 1.5 eV.
Solution. The probability of an energy state E being occupied by an electron is given by
1
f(E) = ( E − EF )/ kT
1+ e
E – EF = 2 eV – 1.5 eV = 0.5 eV and f(E) = 1%
1 1 0.99
\ = 0 . 5 / kT
or e0.5/kT = = 99
100 1+ e 0.01
Taking logarithm on both the sides, we get
0.5 eV
= 2.303 log 99
kT
0.5 eV 0.5 eV
or = T= = 1262 K.
kT 2.303 log 99 × 8.61 × 10−5 eV
Example 29.7. Show that the probability of finding an electron of energy ΔE above the Fermi
level is same as probability of not finding an electron at energy ΔE below the Fermi level.
OR
Show that the probability that a state ΔE above the Fermi level EF is filled equals the
probability that a state ΔE below EF is empty.
Solution. Let us consider an energy level E2 that is above the Fermi level by an amount
of energy ΔE. The probability that the energy level E2 is occupied is given by
1
f(E2) = f(EF + DE) =
1 + exp[( E2 − EF ) / kT
1
=
1 + exp[( EF + ∆E − EF ) / kT ]
1
\ f(E2) = (1)
1 + exp (∆E / kT )
Next, let us consider the energy level E1 that is below the Fermi level by energy ΔE.
[1 – ƒ(E1)] gives the probability that the level E1 is not occupied.
848 A Textbook of Engineering Physics

[1 – f(E1)] = [1 – f(EF – DE)]

1
= 1−
1 + exp[( EF − ∆E − EF ) / kT ]
1
= 1−
1 + exp[(− ∆E ) / kT ]
exp [(− ∆E ) / kT ]
=
1 + exp[(−∆ E ) / kT ]
1
or [1 – f(E1)] = (2)
1 + exp[(∆E ) / kT ]
The R.H.S of equation (1) and (2) are the same.
\ ƒ(E2) = [1 – ƒ(E1)]
It means that the probability of an energy level [EF + ΔE] (ΔE above the Fermi level)
being occupied is the same as the probability of an energy level [EF – ΔE] (ΔE below EF),
being vacant.
Example 29.8. Show that the occupancy probabilities of two states whose energies are
equally spaced above and below the Fermi energy add up to one.
Solution. Let us consider two energy levels E2 and E1, which are equally spaced above
and below the Fermi energy EF.
Let E2 = EF + ΔE  and
E1 = EF – ΔE
The probability of occupancy of the level E2 is given by
1
F(E2) = F(EF + DE) =
1 + exp[( E2 − EF ) / kT ]
1
=
1 + exp[( EF + ∆E − EF ) / kT ]
1
\ F(E2) =
1 + exp (∆ E / kT )
The probability of occupancy of the level E1 is given by
1
f(E1) = f(EF – DE) =
1 + exp [( EF − ∆E − EF ) / kT ]
1
or f(E1) =
1 + exp[−∆E / kT ]
1 1
\ f(E1) + f(E2) = +
1 + exp(∆E / kT ) 1 + exp(−∆E / kT )
1 exp(∆E / kT )
= +
1 + exp(∆E / kT ) 1 + exp(∆E / kT )
1 + exp(∆E / kT )
=
1 + exp(∆E / kT )
\ f(E1) + f(E2) = 1
Band Theory of Solids 849

Thus, the occupancy probabilities of two states whose energies are equally spaced above
and below the Fermi energy add up to one.

29.10 ENERGY BAND STRUCTURE OF AN INSULATOR

Insulators are characterized by two
energy bands – conduction band
and valence band, separated by a Conduction Band
large energy gap. At 0K all valence Empty
EC
electrons are engaged in covalent
bonds, the valence band is full. The EF Eg
absence of mobile charge carriers EF
keeps the conduction band vacant. EV
The situation is same even at Valence Band
higher temperatures (300K), as the full
valence band and conduction bands
are separated by a large gap (> 3 0 1
Distance
eV) and it is not possible to excite f(E)
electrons from valence band to (a) (b)

conduction band by thermal energy Fig. 29.17

(kT ≅ 0.026 eV). Consequently, insulators do not allow flow of current even at temperatures
higher than room temperature.
The concept of Fermi level can be extended to insulators also. As the energy levels in
valence band are filled, f(E) is equal to unity throughout the valence band. As there are no
electrons in the conduction band, f(E) is equal to zero throughout the conduction band. Since
the Fermi function is symmetrical about EF, the Fermi level may be expected to be situated in
the middle of the energy gap. Even though there are no energy levels and no electrons in the
band gap, the meaning of Fermi level remains the same. It is a reference energy position. The
energy band diagram for an insulator is shown in Fig. 29.17 along with probability function.
29.11 ENERGY BAND STRUCTURE OF A SEMICONDUCTOR
A semiconductor is characterized by two energy bands – conduction band and valence band
separated by a smaller energy gap. At normal temperatures, a significant number of electrons
are thermally excited from valence E
band to conduction band. An equal E

number of vacancies are produced

in the valence band. These vacancies
are treated as particles having a mass Area  Electrons
equal to that of electron and carry EC
positive charge. They are called EF Eg EF Area  Holes
holes. EV
The Fermi-Dirac distribution
function is applicable to a semicon-
ductor. Fig. 29.18 depicts the
probability function plotted alongside
0 0.5 1
the band diagram for a semicon- f(E)
Distance
ductor. Because the probability of (a) (b)
electron occupancy of the conduction Fig. 29.18
850 A Textbook of Engineering Physics

band increases at temperatures greater than 0K, the probability function is blurred and tapers
off towards higher energy in the conduction band. Similarly, the probability of hole occupancy
of the valence band increases, and the probability curve is blurred near the top edge of the
valence band. The extent of blurring of blurring of probability curve in both the bands is
equal indicating that the concentration of electrons in the conduction band and that of holes
in the valence band are equal. Secondly, the probability function f(E) rapidly approaches zero
value with increasing E. It suggests that the electrons in the conduction band are clustered
very close to the bottom edge of the band. In a similar way, the holes are grouped very close
to the top edge of the valence band. Therefore, it may be approximated that electrons are
located right at the bottom edge of the conduction band whereas the holes are at the top edge
of the valence band.
The Fermi level represents the average energy of charge carriers participating in
conduction. Both electrons and holes participate in conduction in semiconductor and they lie
in two different bands separated by a forbidden gap. Therefore, it is expected that the Fermi
level lies in the middle of the forbidden gap. If the Fermi level is located elsewhere, it would
mean that the number of electrons in the conduction band would be different from the number
of holes in the valence band. It, in turn, would imply that the material does not exhibit overall
neutrality which is not at all true.

29.12 EFFECTIVE MASS

We generally assume that
the mass of an electron in a 1
>0 Energy
Energy E

solid is the same as the mass m*

A band
of a free electron. However, Energy
1
experimentally measured values 1 B m*
<0 gap
0
indicate that in some solids the m*
First
electron mass is larger while C 1

1 energy
m* m
for others it is slightly smaller – 0 +
band
than the free electron mass. a a
The experimentally determined k

electron mass is called the Fig. 29.19. The effective mass of electron in a solid depends
on electron’s location in the band.
effective mass m*. The cause
for the deviation of the effective mass from the free electron mass is due to the interactions
between the drifting electrons and the atoms in a solid. It has been found that the effective
mass is inversely proportional to the curvature of an allowed energy band. It means that the
effective mass depends on the location of an electron in the allowed energy band (Fig. 29.19).
Considering the electron as a wave packet having a group velocity, ug, an expression for the
effective mass is derived as
2
m* = (29.10)
d 2 E / dk 2
(i) Near the bottom of the band, the form of the E – k curve does not differ much from
the curve for free electron. Therefore, in these regions m* ≈ m.
(ii) At the point of inflection B, the derivative d2E/dk2 = 0. Therefore, in these regions
m* ≈ ∞. It means that an external field cannot exert any action on the motion of the
electron in the region.
Band Theory of Solids 851

(iii) Near the top of the allowed band, C the derivative 2E/dk2 < 0. Therefore, the
effective mass m* of the electrons occupying levels near the top of the band is
negative.
The concept of effective mass provides a satisfactory description of the charge carriers
in crystals. In crystals such as alkali metals, which have partially filled energy band, the
conduction takes place mainly through electrons. However, in crystals for which the energy
band is nearly full, the negative charge and negative mass vacancies may be considered as
positive charge and positive mass particles called holes. It explains the origin of positive Hall
coefficient in certain metals such as zinc.

QUESTIONS
1. Define energy level and energy band. Explain with proper diagrams, how on the basis of band
theory, solids are classified as conductors, insulators and semiconductors.
(C.S.V.T.U., 2005, 2007, 2009)
2. Describe the formation of energy bands in a crystalline solid.
Define valence band, conduction band and forbidden gap in the energy band structure.
Hence classify solids into conductors, semiconductors and insulators. (Bombay Univ.)
3. Explain formation of energy bands in solids on the basis of band theory of solids.
(R.T.M.N.U., 2007)
4. Explain the ‘Kronig-Penny’ model of solids and show that it leads to energy band structure of
solids. (RGPV, 2010)
5. (a) Discuss with suitable mathematical expressions, the motion of an electron in a periodic potential.
(b) Explain how the above theory leads to the concept of band structure of solids.
(c) What is effective mass of electron ? (JNTU, 2010)
6. Explain quantitatively band theory of solids. Explain energy band diagram and distinguish
metal, semiconductor and insulator on the basis of above theory. (RGPV, 2008)
7. Explain how solids are classified on the basis of energy band gap. (Calicut Univ., 2005)
8. Describe in short the formation of energy bands in solids and hence explain how it helps to
classify the materials into conductors, semiconductors and insulators (with an example in each).
(C.S.V.T.U., 2008)
9. Explain how the materials are classified into conductors, semiconductors and insulators with the
help of energy band diagrams. (G.T.U., 2009)
10. Explain the formation of energy bands in solids and briefly explain how solids are classified on
the basis of energy band gap. (Calicut Univ., 2006)
11. According to band theory, a completely filled or empty band is not associated with electrical
conduction. Only partially filled band is responsible for electrical conduction. Explain, why?
(R.T.M.N.U., 2007)
12. How does the band theory differ from the free electron model in explaining the properties of
metals?
13. Explain energy band diagram of silicon showing a graph of variation of potential energy with
distance. Explain semi-conducting nature of silicon. With similar band structure why is diamond
insulator?
14. Draw a graph showing variation of electron energy in germanium crystals as a function of
interatomic distance and explain why it shows semiconducting behaviour?
852 A Textbook of Engineering Physics

15. Draw a graph showing variation of electron energy levels of germanium as a function of its
interatomic distance. Explain from it why germanium is an insulator at 0°K and semiconductor
at 7°K. (R.T.M.N.U., 2007)
16. What is Fermi level and Fermi energy?
17. Explain Fermi energy function. How does it vary with temperature? (RGPV, 2008)
18. Write and explain Fermi function. Explain with the help of a diagram how it varies with change
of temperature.
19. Write down the Fermi-Dirac equation for the probability of occupation of an energy level E by
an electron. Show that the probability of its occupancy by an electron is zero if E > EF and unity
if E < EF at temperature 0K.
20. Define Fermi distribution function. Show that at all temperatures (T > 0K) probability of
occupancy of Fermi level is 50%.
21. What is Fermi function? Draw a graph showing its variation with energy at different tempera-
tures and discuss it.
22. Write down Fermi distribution function f(E). Show graphically and analytically that f(E) as
 1
function of E always passes through a point  EF ,  at different temperatures.
 2
23. Why is that only the electrons near the Fermi level contribute to electrical conductivity?
24. Explain the concept of hole. (R.T.M.N.U., 2006)
25. What is meant by effective mass of electron?
26. Explain the concept of negative mass on the basis of band theory.
27. Write down an expression for the probability of occupancy of a particular energy state of an
electron in an intrinsic semiconductor. Represent it graphically at 0° K and at room temperature.
28. Explain in brief the concept of Fermi level. Show diagrammatically the Fermi level in metals,
intrinsic semiconductors and insulators at 0° K and at higher temperature. (R.T.M.N.U., 2006)
29. How are the band structures of insulators and semiconductors similar? How are they different?
(R.T.M.N.U., 2006)

PROBLEMS
1. In a solid, consider the energy level lying 0.01 eV above Fermi level. What is the probability of
this level being occupied by an electron at 200 K? [Ans: 0.359]
2. In a solid, consider the energy level lying 0.01 eV below Fermi level. What is the probability of
this level being occupied by an electron at 300 K? [Ans: 0.595]
3. In a solid, consider the energy level lying 0.01 eV above Fermi level. What is the probability of
this level being occupied by an electron at 300 K? [Ans: 0.405]
4. In a solid, consider the energy level lying 0.01 eV above Fermi level. What is the probability of
this level being not occupied by an electron at 300 K? [Ans: 0.595]
Semiconductors 853

C H A P T E R

30 Semiconductors

30.1 INTRODUCTION
Semiconductors are materials having electrical conductivity considerably greater than that of
an insulator but significantly lower than that of a conductor. Of all the elements in the periodic
table, eleven elements are semiconductors. Germanium and silicon are the most widely used
semiconductors in device manufacturing applications. They are known as elemental semicon-
ductors. Besides these, there are certain compound semiconductors such as Gallium
Arsenide, Indium Phosphide etc which are formed from the combinations of the elements of
groups III and V or groups II and VI. The unique and interesting feature of semiconductors
is that they are bipolar and two charge carriers, namely electrons and holes, transport current
in these materials. The electrical conductivity of a pure semiconductor, known as intrinsic
conductivity, is significantly low and is drastically influenced by temperature. As such pure
semiconductors cannot be used in device fabrication. Through the technique of doping, the
conductivity of a semiconductor can be increased in magnitude to a desired value and can
be made independent of temperature in a certain temperature interval. Doped semiconductors
are known as extrinsic semiconductors. The remarkable feature of extrinsic semiconductors
is that current is transported in them by two different charge carriers, electrons and holes;
and through two different processes, drift and diffusion. Extrinsic semiconductors are widely
used in fabrication of solid-state devices. An understanding of the mechanism of conduction
in intrinsic and extrinsic semiconductors helps us understand the working of solid-state devices.
30.2 CRYSTAL STRUCTURE
Silicon and germanium are elemental semiconductors. They belong to Group IVA in the
periodic table. Silicon atom has 14 electrons and germanium atom 32 electrons. Each of them
has four valence electrons which are distributed among the outermost s and p orbitals. In case
of silicon atom, there are four energy levels, 3s, 3px, 3py, and 3pz, out of which the lower
two levels are filled by four valence electrons. In an actual piece of silicon crystal, the 3s and
3p orbitals of the atom combine to yield 3sp3 hybrid molecular orbitals. The 3sp3 orbitals
form four covalent bonds of equal angular separation, leading to a tetrahedral arrangement
of atoms in space (Fig. 30.1 a). Atoms occupy the corners of a regular tetrahedron and each
atom in turn bonds with four neighbours and so on. The resulting crystal structure is known as
diamond cubic (DC) crystal structure, shown in Fig. 30.1 b.
854 A Textbook of Engineering Physics

Si

a/2

109.5°
109.5° Si
Si

Si 109.5°
°
9.5
10

Si a

(a) (b)
Fig. 30.1

30.3 INTRINSIC SEMICONDUCTOR

Chemically pure semiconductors are known as intrinsic semiconductors. A semiconductor is
considered to be pure when there is less than one impurity atom in a billion host atoms.
A two-dimensional representation of silicon crystal is shown in Fig. 30.2a. Each silicon
atom forms covalent bonds with four surrounding atoms. The shaded circles in Fig. 30.2a,
represent the cores of the silicon atoms. The four valence electrons are shown by the small
black dots surrounding each hatched circle. The probability of valence electrons being in any
place between the bonding atoms is indicated by the dashed curves. In terms of energy band
diagram, a conduction band and a valence band separated by a smaller energy gap characterize
a semiconductor (Fig. 30.2b).
In a real crystal, the concentration of atoms N is given by
N ρ
N= A (30.1)
M
where NA is the Avogadro number, r the density and M the atomic weight of the material.
Using the data for silicon into equ. (30.1), we obtain

N=
( 6.02×10 26
)(
atoms/k.mol 2330 kg/m 3 )
28.09 kg / k.mol
or N = 5 × 1028 atoms/m3 (30.2)
The valence and conduction bands of silicon crystal contain 2N energy levels each. Therefore,
the number of energy levels in each band is 1029 levels/m3. In other words, there are 2×1029 states/
m3. The number of valence electrons available in the silicon crystal is 4N = 2×1029 electrons/m3.
These electrons occupy the valence band and leave the conduction band vacant.
(a) At 0K an Intrinsic Semiconductor Behaves as a Perfect Insulator
At 0K and temperatures close to 0K, all valence electrons are locked in covalent bonds
(Fig. 30.2 a) and spend most of the time between neighbouring atoms. Since all the valence
electrons are engaged in covalent bonds, the bonds are complete. The energy available at 0K
is not sufficient to break the covalent bonds.
Therefore, there are no free electrons within the material at absolute zero. Consequently,
the semiconductor at 0K cannot conduct electricity and acts as a perfect insulator.
Semiconductors 855

In terms of energy T = 0K
E
band diagram, the situation
is as follows. There are 2N Si Si Si
Conduction Band
29 3
29
= 10 energy levels/m in 3 N = 10 /m Levels

the valence band and 2N E

Vacant
29
= 10 energy levels/m in 3
C

Si E E
the conduction band. The Si Si F g

total number of valence EV

band electrons available in Valence Band

29 3
2N = 2 × 10 /m
the crystal is 4N=2×1029 Si Si Si Valence Electrons
occupy N levels
3
electrons/m . At 0K
these 2×1029 electrons/ Distance
m3 completely occupy (a) (b)
29
the 10 energy levels/m 3 Fig. 30.2: Silicon Crystal at 0K
available in the valence
band. There are no electrons left to go into the conduction band (Fig. 30.2b). At 0K, electrons
in the valence band do not possess enough energy to jump into the conduction band. As free
electrons do not exist in the conduction band, an externally applied electric field cannot cause
flow of current through the crystal. Hence, the intrinsic semiconductor behaves as a perfect
insulator at 0K.
(b) Mechanism of Conduction in an Intrinsic Semiconductor
At temperatures above absolute zero, the finite thermal energy causes each atom in the crystal
to vibrate about its mean position. When the vibrations become violent, some of the electrons
acquire sufficient energy and break away from covalent bonds (Fig. 30.3a). Whenever a
covalent bond is ruptured by thermal energy, a valence electron becomes free. The higher the
temperature, the more covalent bonds are broken. The electrons liberated from bonds move
randomly in the void spaces between the atoms in the crystal. If an electric field is applied,
these free electrons cause electrical conduction.
From the T > 0°K
energy band E
point of view,
it means that Si Si Si
some of the Broken Free
electrons in Covalent Electron
C E
Bond
the valence
band convert Si Si Si F E

part of their
E
thermal energy Vacancy V

into potential
energy. Those Si Si Si
electrons which
acquire energy Distance
equal to or (a) (b)
in excess of
Fig. 30.3: Silicon Crystal at T > 0K
the band gap
energy Eg are excited to the conduction band (Fig. 30.3b). Thus, the band gap energy Eg is the
minimum amount of energy required to excite an electron from valence band to conduction
band. Eg is characteristic of the material. The number of electrons excited to the conduction
856 A Textbook of Engineering Physics

band depends on the amount of thermal energy received by the crystal. For example, the
concentration of broken bonds in a silicon crystal at 300K is 1.5×1016 /m3. Thus, there are
1.5×1016 electrons/m3 in the conduction band at 300K. In fact, the conduction band can
accommodate 2×1029 electrons/m3 and hence it is partially filled.
When an electron from the valence band jumps to the conduction band, an empty state
(quantum vacancy) arises in the valence band. In silicon crystal at 300K, 1.5×1016 vacant
states/m3 appear in the valence band, which are very small in number compared to the number
of electrons remained in the band. Thus, now both the bands are partially filled. The electrons
in the conduction band and the electrons in the valence band can be excited to upper vacant
levels within the respective bands. Therefore, if an electric field is applied, these electrons
can move into higher vacant levels and current flows in the crystal at ordinary temperatures.
The motion of valence electrons in the valence band is customarily described in terms of a
fictitious particle called hole which is bequeathed with a positive charge +e and a mass mh
equal to that of an electron.
In pure semiconductors all available charge carriers, electrons and holes, arise due to
thermally ruptured bonds and these thermally generated electron-hole pairs cause electrical
conduction. Thermal generation is an intrinsic process. Therefore, we may define an intrinsic
semiconductor as follows.
Definition: An intrinsic semiconductor is a semiconductor crystal in which electrical
conduction arises due to thermally excited electrons and holes.
Significance of Band Gap Eg
The band gap energy Eg is the minimum amount of energy required for breaking a covalent
bond. It is the minimum amount of energy required to excite an electron from valence band
to conduction band. Also, it is the minimum amount of energy required to convert a bound
electron into a free electron. The energy required to break a covalent bond in a germanium
crystal is about 0.72 eV at 300K and that in silicon is 1.12 eV.

30.4 CORRELATION BETWEEN CRYSTAL LATTICE AND ENERGY

BAND DESCRIPTIONS
The correlation between the crystal lattice description and the energy band description of a
semiconductor can be
summed up as follows: Electron
kinetic
The statement that Electron energy
an electron is in the energy Electrons
valence band means that
the electron is partici- Electron
E C

pating in a covalent band. potential energy

The excitation of the Hole potential
electron from the valence energy
band to the conduction E V

band implies rupturing Hole

Holes
of a covalent bond. energy Hole
Accordingly, the band gap kinetic
Eg may be regarded as energy
representing the strength of
the covalent bond. When it Distance
Fig. 30.4
Semiconductors 857

is said that the electron is in the conduction band, it means that the electron is wandering
haphazardly in the interstitial spaces of the crystal. The electrons in the conduction band are
equally attracted to all nuclei and hence they experience a constant potential in the crystal.
The magnitude of the potential energy may be taken as Eg. It is the energy corresponding to
the lowest energy level in the conduction band.
Out of the thermal energy E supplied to a valence electron, an amount Eg is translated
into the potential energy and the electron reaches the bottom edge EC, of the conduction band
(Fig. 30.4). Any additional energy goes into the form of kinetic energy. Thus,
E = Eg + 12 mυ2 (30.3)
Consequently, the electrons in the conduction band are characterized by their velocity
only and act as free electrons. It is seen from the relation (30.3) that an electron with an
energy Eg just reaches the bottom edge of the conduction band and will be at rest since u = 0.
With more energy, it is positioned higher and higher above the conduction band edge EC. The
situation is very much akin to the billiard balls on a billiard table. The balls are confined to
the table but can have wide ranging velocities.
The situation is similar for holes. The top of the valence band, EV represents the potential
energy of the hole (Fig. 30.4). The energies below EV indicate the increasing kinetic energy of
a hole. The directions of increasing energy for electrons and holes are opposite to each other
because they have charges of opposite sign.
30.5 HOLES
In a silicon crystal, when a covalent bond is broken and an electron is set free, there arises a
quantum vacancy at the site of the broken bond. The removal of a negative charge from an
otherwise neutral atmosphere accords an effective positive charge +e to the resulting vacancy.
Obviously, the actual seat of the positive charge is the nucleus of the silicon atom which
is deprived of its valence electron. The positively charged vacancy, say, created at the site
A can attract an electron from the adjacent bonds. An electron from site, say, B jumps into
the vacancy at A and bridges the bond but leaves a vacancy at B and so on. In this way, a
new type of electron motion is created in the valence band. In quantum mechanical terms,
the electron at a lower energy level makes an upward transition to a vacant level. As the
electron moved up, the vacancy moves down. Note that the vacancy and the electron move in
opposite directions in the valence band. As both motions are related, the hopping of a large
number of valence electrons can be equivalently described in terms of normal motion of a
small number of holes. The concept of hole helps in differentiating the two different kinds of
electron motion, namely the motion of high energy electrons in the conduction band and the
motion of low energy electrons in the valence band.
The concept of holes simplifies the understanding of electrical conduction in semicon-
ductors. It is seen as follows:
The instantaneous current produced by one electron moving at a speed uk is given by
i = −euk
The current set up by the motion of many electrons in a valence band is therefore
I = −e∑ υi
i
For a completely filled band, the resultant current is zero. That is,
I = −e∑ υi = 0 (30.4)
i
858 A Textbook of Engineering Physics

The above equation may be rewritten as

 
I = −e  ∑ υi + υk  = 0
 i≠k 
\ −e∑ υi = − e ( −υk ) = euk (30.5)
i≠k

\ ∑ υi = −uk (30.6)
i≠k

The above relation (30.6) implies that if the kth electron is absent in the valence band,
then the sum of the velocities of the remaining electrons is −uk. The relation (30.5) shows that
these electrons will set up a current equal to +euk. It means that the current set up by all the
electrons in the valence band having one vacancy is equivalent to the current resulting from
the motion of a single particle with a positive charge +e that occupies the vacancy.
If a vacancy is to be given the status of a particle, a mass is also to be ascribed to it. We
can arrive at the mass from the band theory. In solids, electrons experience forces due to
atoms and other electrons. The motion of electron in a solid subjected to an electric field may
be written in the form
F = me*a
where me* is the effective mass Energy
of the electron. The effective
mass of an electron heavily
depends on its location in the
energy band, as discussed in
Art. 29.12 of Chapter 29. "Light" electrons Region 4
Electrons near the bottom of
the conduction band have an Forbidden Region 3
effective mass which is nearly
identical to the mass of a free
electron. Electrons near the top Negative "mass" Region 2

of valence band have negative

effective mass. The concept of Classical
negative effective mass is not behaviour Region 1
easy to grasp. The removal of
an electron with a negative
effective mass is identical to k
creating in its place a particle –k
1 0 +k1

of positive effective mass. Fig. 30.5

*
Thus, a hole is assigned an effective mass, mh .
A hole in the sea of valence electrons in the valence band is analogous to a bubble in
a liquid. A bubble has shape, size and also a velocity. Its behaviour is the same as that of a
particle. In reality, it is the absence of liquid. One can in principle describe the motion of a
bubble by describing the motion of the surrounding liquid; but it will be a tedious task. In
a similar way, the motion of the valence electrons is difficult to be described. On the other
hand, the description of their motion in terms of holes is a far simpler task.
Semiconductors 859

Direct evidence for the existence of holes furnished by the results of Hall effect measure-
ments justifies the use of parlance of holes.

30.6 GENERATION AND RECOMBINATION

In semiconductors a single event of covalent bond breaking leads to the generation of two
charge carriers, an electron in the conduction band and a hole in the valence band (Fig.
30.3). The electron and hole are produced simultaneously as a pair and the process is called
electron-hole pair generation. The process may be represented as
Rupture
Covalent bond + Thermal energy  → (Electron + Hole) pair
In the process of generation, a covalent bond is broken and a bound electron is transformed
into a free electron. Thermal energy is one of the agents which causes pair generation. Another
agent is optical illumination. At any temperature T, the number of electrons generated would
be equal to the number of holes produced. If n denotes the concentration of electrons in the
conduction band and p is the concentration of holes in the valence band, then
n=p (30.7)
After generation the charge carriers move independently. The electrons move in the conduction
band and the holes move in the valence band. Their motion is at random in the respective
bands, as long as external electric field is not applied.
It is likely that the electron in conduction band may lose its energy due to collision with
other particles in the lattice and fall into the valence band (Fig. 30.6). When a free electron
falls into valence band, it merges with a hole. This process is called recombination. When a
recombination event occurs, the free electron enters a ruptured covalent bond and re-bridges it.
Electron + Hole → Covalent bond + Energy↑
Therefore, recombination means that
a free electron transforms into a valence Conduction
electron and that a ruptured covalent bond band
is re-bridged. In the process the electron-
hole pair disappears and energy is released. Electron Generation Recombination
E
The released energy is mainly in the form of energy g

thermal energy.
At a steady temperature a dynamic Valence
equilibrium exists which balances the two band
processes of electron-hole pair generation and
electron-hole recombination (Fig. 30.6). Distance

Just as thermal energy generates electron- Fig. 30.6: Generation and recombination of
electron-hole pairs
hole pairs, light radiation can produce
electron-hole pairs in a semiconductor. When a semiconductor material is irradiated with
optical radiation, electron-hole pairs are generated if the frequency n of the radiation satisfies
the condition
hv ≥ Eg (30.8)
When the radiation is switched off the excess electron-hole pairs recombine and the
material returns to thermal equilibrium. Optically generated electron-hole pairs are responsible
for the working of LDRs, photodiodes, etc optical detectors.
860 A Textbook of Engineering Physics

30.7 INTRINSIC CONDUCTIVITY

A single event of bond breaking in a pure semiconductor leads to generation of an electron-
hole pair. At any temperature T, the number of electrons generated will be equal to the
number of holes generated per unit volume. As the two charge carrier concentrations are
equal, they are denoted by a common symbol ni, which is called intrinsic density or intrinsic
concentration. Thus,
n = p = ni (30.9)
In a semiconductor under thermal equilibrium condition, free electrons move in the
conduction band and holes in the valence band, which are in a state of random motion. At a
temperature T, they possess an average kinetic energy given by
1 2 3 Total conventional
mυth = kT current
I=I +I
e h

2 2
where uth is the mean thermal velocity. When Electron flow
a potential difference is applied across the Ie

Electric field E
solid, the equilibrium condition is disturbed.
The electric field accelerates the electrons Hole flow
I
and holes but their motion is hindered due h

to interactions with the lattice vibrations. In I

the steady state condition there arises a net V
movement of electrons in a direction opposite + –
to that of the electric field and movement Fig. 30.7: Components of drift current due
of holes in the direction of the electric field. to electron and hole motion in an intrinsic
This net movement of electrons and holes is semiconductor
called drift, and the corresponding mean velocity is known as drift velocity, ud. The drift
motion is superposed on the random thermal motion of the charge carriers. The drift motion
is directional and causes drift current flow, which is more often called conduction current.
The drift velocity is given by
ud = mE (30.10)
Since the electrons move in relatively less populated conduction band the properties such
as mobility, conductivity etc of electrons are larger compared to those of holes as the latter
move in nearly full valence band. We designate the drift velocity of electron with ude and
that of hole with udh. Similarly, we denote the mobility of electron and hole with me and mh
respectively. Then, the current density due to electrons is given by
Je = neude = nemeE (30.11)
and the current density due to holes is
Jh = peudh = pemhE (30.12)
Comparing the above expressions with Ohm’s law J = sE, we obtain the expressions for
electronic and hole conductivities as follows.
se = neme (30.13)
and sh = pemh (30.14)
Let us now consider a sample of semiconductor across which a potential difference V is
applied. The potential difference V establishes an electric field E in the semiconductor. It
causes a current Ie due to electrons drifting in the conduction band and a current Ih due to holes
drifting in the valence band (see Fig. 30.7). The total current through the semiconductor is
I = Ie + Ih (30.15)
Semiconductors 861

Ie Ih
\ Current density J = + where A is cross-sectional area of the crystal.
A A
or J = Je + Jh
= (neme + pemh)E (30.16)
Therefore, the intrinsic conductivity is given by
s = (neme + pemh) (30.17)
As n = p = ni in an intrinsic semiconductor, equ. (30.17) may be rewritten as
s = eni (me + mh) (30.18)
The equation (30.18) does not explain the temperature dependence of electrical
conductivity in semiconductors. In general the variation of mobility with temperature is too
small and the large variation in electrical conductivity in semiconductors is linked to the
variation of electron concentration with temperature. Thus, as a first approximation, taking e,
me and mh as constants, we find that
s(T) ∝ ni (T) (30.19)
The expression for the intrinsic carrier concentration is derived in the band theory. Here,
we take the final results from the band theory and use them to obtain an expression for n(T),
the variation of intrinsic carrier concentration with temperature.

30.8 CARRIER CONCENTRATIONS

With an increase in temperature covalent bonds are broken in an intrinsic semiconductor and
electron-hole pairs are generated. We expect that a large number of electrons can be found in
the conduction band and similarly, a large number of holes in the valence band. As electrons
and holes are charged particles, they are together are called charge carriers. Carrier
concentration is the number of electrons in the conduction band per unit volume (n) and the
number of holes in the valence band per unit volume (p) of the material. Carrier concentration
is also known as the density of charge carriers. We would like to calculate the electron
concentration, n, in the conduction band and the hole concentration, p, in the valence band.
30.8.1 Calculation of Electron Density
Let dn be the number of electrons whose energy lies in the energy interval E and E + dE in
the conduction band. Then,
dn = Z(E)f(E)dE (30.20)
where Z(E) dE is the density of states in the energy interval E and E + dE and f(E) probability
that a state of energy is occupied by an electron.
The electron density in the conduction band is given by integrating the above equation
between the limits EC and ∞. EC is the energy corresponding to the bottom edge of the
conduction band and ∞ the energy corresponding to the top edge of the conduction band. As
the probability of electrons occupying upper levels of conduction band f(E) readily approaches
zero for higher energies, the upper limit, namely the top of conduction band is taken as ∞.
Thus,
∞
n= ∫ Z ( E ) f ( E ) dE (30.21)
EC

The density of states in the conduction band is given by

862 A Textbook of Engineering Physics

4π
( )
3/2 1/2
Z(E)dE = 3
2me* E dE for E > EC. (30.22)
h
The bottom edge of the conduction band EC corresponds to the potential energy of an
electron at rest. Therefore, (E – EC) will be the kinetic energy of the conduction electron at
higher energy levels. Hence, equ. (30.22) is to be modified as follows.
4π
( )
3/2
Z(E)dE = 3 2me* ( E − EC ) dE
1/2
(30.23)
h
The probability of an electron occupying an energy level is given by
1
f(E) =
1 + exp ( E − EF ) / kT 
When the number of particles is very small compared to the available energy levels, the
probability of an energy state being occupied by more than one electron is very small. Such
a situation is valid when (E – EF) >> 3kT. Under this circumstance, the number of available
states in the conduction band is far larger than the number of electrons in the band. Then,
Fermi-Dirac function can be approximated to Boltzmann function,
f(E) = exp[– (E – EF)/kT]. (30.24)
Using the equations (30.23) and (30.24) into (30.21), we obtain
∞
4π
( ) ∫ (E − E
3/2
\ n = 2me* C )1/2e−( E − E F )/ kT dE (30.25)
h3 EC
∞
4π
( )
3/2
e(
EF − EC )/ kT 1/2 −( E − EC )/ kT
or n =
h3
2me* ∫ ( E − EC ) e dE (30.26)
EC
The integral in eq. (30.26) is of the standard form which has a solution of the following
form. ∞
π
∫ x e dx = 2a a
1/2 − ax

0
where a = 1/kT and x = (E – EC).
 π 
4π
( )
3/2
e(
EF − EC )/ kT
\ n= 2me*  ( kT )3/2  (30.27)
h3  2 
Rearranging the terms, we get
3/2
 2πme*kT  −( EC − EF )/ kT
n = 2 2  e (30.28)
 h 
The above equation is the expression for the electron concentration in the conduction
band of an intrinsic semiconductor.
3/2
 2πme*kT 
Designating NC = 2  2  (30.29)
 h 
in the above equation, we obtain
n = NCe – (EC – EF)/kT (30.30)
NC is a temperature dependent material constant known as the effective density of states
in the conduction band. In silicon at 300 K, NC = 2.8 × 1025/m3.
Semiconductors 863

The importance of the relation (30.30) is that it relates the equilibrium electron
concentration to a single variable, namely the Fermi level EF. Therefore, electron concentration
is specified, if EF is specified.
30.8.2 Calculation of Hole Density
Let dp be the number of holes whose energy lies in the energy interval E and E + dE in the
valence band. Then,
dp = Z(E)[1 – f(E)]dE (30.31)
where Z(E)dE is the density of states in the energy interval E and E + dE and [1 – f(E)] the
probability that a state of energy is vacant and not occupied by an electron. If f(E) is the
probability for occupancy of an energy state at E by an electron, then the probability that the
energy state is vacant is given by [1 – f(E)]. Since a hole represents a vacant state in valence
band, the probability for occupancy of a state at E by a hole is equal to the probability of
absence of electron at that state.
We can write
1 1
≈e ( F )
− E − E / kT
[1 – f(E)] = 1 − ( E − EF )/ kT
=
( EF − E )/ kT
(30.32)
1+ e 1+ e
The density of states in the valence band is given by
4π
( )
3/2 1/2
Z(E)dE = 3 2mh* E dE (30.33)
h
The top edge of the valence band EV corresponds to the potential energy of a hole at rest.
Therefore, (EV – E) will be the kinetic energy of the hole at lower energy levels. Hence, equ.
(30.23) is to be modified as follows.
4π
( )
3/2
Z(E)dE = 3 2mh* ( EV − E )1/2 dE (30.34)
h
The number of holes in the energy interval E and E + dE is
4π
( )
3/2
dp = 3 2mh* ( EV − E )1/2 e−( EF − E )/ kT dE (30.35)
h
In order to calculate the number of holes in the valence band equ. (30.35) is to be
integrated between the limits –∞ and EV. The hole density in the valence band is therefore
given by
EV
4π
( ) ∫ (E
3/2
− E ) e ( F ) dE
1/2 − E − E / kT
p = 2mh* V (30.36)
h3 −∞
EV
4π
( )
3/2 −( E − E )/ kT 1/2 −( EV − E )/ kT
= 3 2mh*
h
e F V
∫ ( EV − E ) e dE (30.37)
−∞
The integral in equ. (30.37) is of the standard form which has a solution of the following
form.
∞
π
∫x
1/2 − ax
e dx =
0
2a a
where a = 1/kT and x = (EV – E).
 π 
4π
( )
3/2 −( E − E )/ kT
\ p = 2mh* e F V
 ( kT )3/2  (30.38)
h3  2 
864 A Textbook of Engineering Physics

Rearranging the terms we get

3/2
 2π mh* kT 
e ( F V)
− E − E / kT
or p = 2 2  (30.39)
 h 
The above equation is the expression for the hole concentration in the valence band of
an intrinsic semiconductor.
3/2
 2π mh* kT 
Denoting NV = 2  2  (30.40)
 h 
\ p = NV e–(EF – EV)/kT (30.41)
NV is called the effective density of states in the valence band. For silicon at 300 K, NV =
1025/m3. It is seen that NC/NV = 2.8. To a first approximation, NC is taken to be equal to NV.
The electron and hole concentrations in an intrinsic semiconductor are shown in Fig.
30.8.
Density Carrier
Ferm-Dirac of states concentrations
E
distribution

Conduction Band n0 = ni

EC EC
gc = (E) f (E) gc (E)
EF EF +
[1 – f(E)] gV(E)
EV EV
gc = (E)
p 0 = ni
Valence band
f(E) 0
1 1/2
Fig. 30.8

30.9 INTRINSIC CARRIER CONCENTRATION

A single event of bond breaking in a pure semiconductor leads to generation of an electron-
hole pair. At any temperature T, the number of electrons generated will be equal to the number
of holes generated. As the two charge carrier concentrations are equal, they are denoted by a
common symbol ni, which is called intrinsic density or intrinsic concentration. Thus,
n = p = ni
We can write that
n2i = np (30.42)
= (NCe–(EC – EF)/kT)(NVe–(EF – EV)/kT)
= (NCNV)–(EC – EV)/kT (30.43)
The term (EC – EV) stands for the difference in energy between the top level of valence
band and the bottom level of conduction band. Thus, it represents the separation between the
valence and conduction bands, which is the band gap Eg.
(EC – EV) = Eg (30.44)

\ n2i = (NC NV)e–Eg/kT

Semiconductors 865

Substituting the values of NC and NV into the above equation, we obtain

3
 2π kT 
( )
3/2 − E / kT
= 4  2  me*mh* e g

 h 
3/2
\
 2π kT 
ni = 2  2 
 h 
(m m )
*
e
* 3/4 − Eg /2 kT
h e (30.45)

This is the expression for intrinsic carrier concentration.

30.9.1 Variation of Intrinsic Carrier Concentration with Temperature
Equ. (30.45) may be written as
3/2

 2π k 
ni = 2  2 
 h 
(m m )
*
e
* 3/4 3/2 − Eg /2 kT
h T e (30.46)

The above relation shows that the free charge carrier concentration varies with
temperature.
Equ. (30.46) may be approximated to
ni = 1021.7T3/2 × 10–2500 Eg/T (30.46a)
The following important points may be inferred from the relation (30.46):
● The intrinsic concentration is independent of Fermi level.
● The intrinsic concentration has an exponential dependence on the band gap value Eg.
● It strongly depends on the temperature.
● The factor 2 in the exponent indicates that two charge carriers are produced for one
covalent bond broken.
The experimental value of ni in silicon at room temperature is 1.5 × 1016 carriers/m3 and
in germanium 2.5 × 1019 carriers/m3.
Example 30.1: The forbidden gap in pure silicon is 1.1 eV. Compare the number of conduction
electrons at temeperatures 37°C and 27°C.
Solution: Let n1 be the number of conduction electrons at 27°C and n2 at 37°C. Using equn.
(30.46 a), we get
n1 = 1021.7 (300 K)3/2 (10–2500 × 1.1eV/300 K)
and n2 = 1021.7 (300 K)3/2 (10–2500 × 1.1eV/310 K)

n2
=
(310K ) × 10−8.87 = 2.96
3/2

n1 (300K )3/2 × 10−9.2

Thus the number of electrons in the conduction band increased nearly three-fold as the
temperature of the material increased from 27°C to 37°C.
Example. 30.2: Compute the concentration of intrinsic charge carriers in a germanium
crystal at 300 K. Given that Eg = 0.72 eV and assume me* = me.
3/2
 2π me*kT   Eg 
Solution: Intrinsic charge carrier concentration, ni = 2   exp  −
 h
2
  2kT 
3/2
3/2  
 2π me*k   2 × 3.143 × 9.11 × 10 −31 kg × 1.38 × 10−23 J/K 
2  = 2

( )

2 2
 h  6.626 × 10−34 J.s
 
866 A Textbook of Engineering Physics

= 4.83 × 1021 , T 3/2 = 3003/2 = 5196

 Eg   0.72 eV 
and exp  − = exp  −

 2kT   2 × 8.61 × 10 eV/K × 300K 
−5

= exp (-13.846) = 9.7 × 10-7

\ ni = 4.83 × 1021 × 5196 × 9.7 × 10-7 = 3.4 × 1019/m3.

30.10 THE FRACTION OF ELECTRONS IN THE CONDUCTION BAND

The Fermi-Dirac probability function gives the fractional occupancy of the energy states.
1
Thus, f ( E ) = gives the probability that an electron occupies the
1 + exp ( E − EF ) / kT 
energy state E. The probability that an electron occupies the energy state EC may then be
found from
1
f(EC) = (30.47)
1 + exp ( EC − EF ) / kT 
n
But, according to the definition of probability, f ( EC ) =
N
where n is the number of electrons excited to conduction band levels and n is the total number
of electrons available in the valence band initially.
n 1
\ = (30.48)
N 1 + exp ( EC − EF ) / kT 
As (EC – EF) = Eg/2, we write
n 1
=
N 1 + exp  Eg / 2kT 

Since Eg > 2kT, the factor unity may be neglected in comparison to the exponential term.
n 1 − E /2 kT
\ = =e g (30.49)
N exp  Eg /2kT 
The above equation gives the fraction of electrons in the conduction band.
Example 30.3. What is the probability of an electron being thermally promoted to the
conduction band in diamond at 27° C, if the bandgap is 5.6 eV wide?
Solution. The probability that an electron being thermally promoted to the conduction
band is given by
1 1
f(EC) = =
(
1 + exp Eg /2 kT )  5.6eV
1 + exp 

eV 
 2 ( 0.026) 
1
= = 1.7 × 1047
1 + e107.69
Example 30.4. Estimate the fraction of electrons in the conduction band at 300°K of
(i) Germanium (Eg = 0.72 eV) (ii) Silicon (Eg = 1.1 eV) and (iii) Diamond (Eg = 5.6 eV)
What is the significance of these results?
Semiconductors 867

Solution.
0.72eV
− Eg /2 kT
(i) Germanium: f(EC) = e = e 2×0.026eV = e −13.85 = 9.66 × 10–7
1.1eV
− Eg /2 kT
(ii) Silicon: f(EC) = e = e 2×0.026eV = 6.5 × 10−10
5.6eV
− Eg /2 kT
(iii) Diamond: f(EC) = e = e 5×0.026eV = e −107.7 = 1.7 × 10−47
The above results show that the larger the band gap the smaller the electrons that can go
into the conduction band, at a given temperature.
Example 30.5: Assuming that the number of electrons near the top of the valence band
available for thermal excitation is 5 × 1025/m3 and the intrinsic carrier density is 2.5 × 1019/
m3, calculate the energy gap of germanium at room temperature.
n  Eg 
Solution: Fraction of electrons in conduction band = exp  − 
N  2kT 
 n
\ Eg = −2kT ln  
 N

 2.5 × 1019 /m3 

(
= −2 8.61 × 10−5 eV/K ( 300 K ) ln  ) 25 3 
 5 × 10 /m 
= 0.052 × 14.509 eV = 0.75 eV.
Example 30.6. Estimate the fraction of electrons in conduction band at room temperature in
Ge with Eg = 0.72 eV and in diamond with Eg = 5.6 eV.
n  Eg 
Solution. Fraction of electrons in conduction band = exp  − 
N  2 kT 
(i) Fraction of electrons in conduction band in Ge,
n  0.72 eV 
= exp  − = exp (-13.846) = 9.7 × 10-7.
N  2 × 0.026 eV 
(ii) Fraction of electrons in conduction band in diamod,
n  5.6 eV 
= exp  −
N  2 × 0.026 eV 
= exp (-107.692) = 1.7 × 10-47.
30.11 FERMI LEVEL IN INTRINSIC SEMICONDUCTOR
In a pure semiconductor, the electrons in the conduction band cluster very close to the
bottom edge of the band, and we assume that electrons are located right at the bottom edge of
the conduction band, as shown in Fig. 30.9. Similarly, we assume that the holes are at the top
edge of the valence band. The electron concentration in the conduction band is given by
n = NC e–(EC–EF)/kT
The hole concentration in the valence band is given by
p = NV e–(EF–EV)/kT
In an intrinsic semiconductor, the electron and hole concentrations are equal. Thus, n = p
868 A Textbook of Engineering Physics

NC e–(EC–EF)/kT = NV e–(EF–EV)/kT (30.50)

Taking logarithm on both sides, we get
( EC − EF ) N ( E − EV )
− = ln V − F
kT NC kT

NV
–EC + EF = kT ln − EF + EV
NC

NV
2EF = ( EC + EV ) + kT ln
NC
E + EV 1 N
\ EF = C + kT ln V (30.51)
2 2 NC
3/2 3/2
 2π me*kT   2π mh* kT 
But NC = 2  2  and NV = 2  2 
 h   h 
3/2
NV  m* 
\ =  h* 
NC  me  E Conduction Band

N  3  m* 
\ ln  V  = ln  *h  EC
 NC  2  me 
Electron
Energy

EF
EC + EV 3  m* 
\ EF = + kT ln  *h  (30.52)
2 4  me  EV

We can also write the above equation as

Valence band
E + EV 3  m* 
EF = C − kT ln  *e  (30.53)
2 4  mh  Distance
If the effective mass of a free electron is Fig. 30.9: Fermi level in intrinsic
assumed to be equal to the effective mass of a semiconductor
hole, i.e.,
mh* = me*
  m*
 = 0ln  *h
m
  e
E + EV
\ EF = C (30.54)
2
To make the meaning of the above equation more explicit, we write
E − EV
EF = C + EV
2
Eg
EF = + EV
2
If we denote the top of the valence band EV as zero level, EV = 0.
Semiconductors 869

Eg
\ EF = (30.55)
2
The above result shows that in an intrinsic semiconductor the Fermi level lies in the
middle of the forbidden gap. We made the following assumptions in obtaining the relation
(30.55).
● It was assumed that the electrons undergo transitions from the top edge level, EV,
of the valence band to the bottom edge level, EC, of the conduction band. In reality,
transitions are possible between the other levels also. However, the above result does
not appreciably differ if other transitions are also taken into account.
● It was assumed that the effective mass of electrons in the conduction band, me*, is
exactly equal to the effective mass of the holes in the valence band, mh*. In practice,
the effective masses differ from each other. However, the difference does not alter the
above result significantly.
An important point to be noted here is that the Fermi level is not an allowed energy
level in semiconductors. It only serves as a reference energy with reference to which we
specify the energies of electrons and holes in a semiconductor.

30.11.1 Variation of Fermi Level E

with Temperature in an Intrinsic

Semiconductor Conduction Band
With an increase in temperature, the Fermi level
gets displaced upward to the bottom edge of the EC
conduction band if mh* > me* or downward to
*
the top edge of the valence band if mh* < me*, as m*h
> me
indicated in Fig. 30.10. mh = me
EF
In most of the materials, the shift of Fermi
level on account of mh* ≠ me* is insignificant. The mh* <
Fermi level in an intrinsic semiconductor may me*
be considered as independent of temperature EV
and as staying in the middle of the band gap.

Example 30.7: Determine the position of Fermi

level in silicon semiconductor at 300 K. Given Valence band

that the band gap is 1.12 eV, and me* = 0.12 m

and mh* = 0.28m. T
200 K 400 K 600 K
Eg 3kT  m*h  Fig. 30.10: Variation of Fermi level with
Solution: EF = + ln  * 
2 4 m temperature in an intrinsic semiconductor
 e 

1.12 eV 3 × 8.61 × 10−5 eV/K × 300 K  0.28 m 

= + ln  
2 4  0.12 m 

= 0.56 eV + (0.0194) ln 2.333 eV

= 0.56 eV + (0.0194) (0.8473) eV
870 A Textbook of Engineering Physics

= 0.56 eV + 0.016 eV = 0.576 eV.

The Fermi level is 0.016 eV above the centre of the forbidden gap. In other words, it is at
0.576 eV from the top of the valence band.
Example 30.8: If the effective mass of an electron is equal to twice the effective mass of hole,
determine the position of the Fermi level in an intrinsic semiconductor from the center of
forbidden gap at room temperature.
Eg 3  m* 
Solution: EF = + kT ln  h* 
2 4  me 
Eg 3 Eg 3
− kT ln 2 = − ( 0.026 eV )( 0.69)
2 4 2 4
Eg
or EF = − 0.0135 eV
2
The Fermi level is below the centre of the forbidden gap by 0.014 eV.

30.12 VARIATION OF INTRINSIC CONDUCTIVITY WITH TEMPERATURE

Incorporating the expression for ni from equ. (30.43) into the equation (30.18), we get

NC NV e ( µ e + µ h ) e
– Eg / 2kT
s =

− Eg / 2kT
or s = σ o e (30.56)
where so is a constant.
The relation (30.56) gives the temperature
dependence of conductivity of an intrinsic semicon- In i
ductor, which is dominated by the exponential term
e–Eg/2kT.
Taking logarithm on both sides of eq. (30.56),
we get
Eg 1
ln si = ln σ o −
(30.57) T
2 kT
Fig. 30.11: Variation of intrinsic
conductivity with temperature
Fig. 30.11 shows a plot of ln si versus 1/T. It
shows that the conductivity increases with temperature.
Example 30.9: Find the resistivity of intrinsic germanium at 300°K. Given that the intrinsic
density carriers is 2.5 × 1019/m3.
Solution: si = eni(me + mh)
= 1.602 × 10−19 C × 2.5 × 1019 × ( 0.39 + 0.19) m 2 V −1s −1
= 2.32 mho/m
1
\ ri = = 0.43ohm − m.
ρi
Semiconductors 871

30.13 DETERMINATION OF BAND GAP

Equ. (30.56) may be rewritten in terms of resistivity of the intrinsic semiconductor as
follows:
ri = roeEg/2kT (30.58)
Ri A
or = roeEg/2kT
L
\ Ri = CeEg/2kT (30.59)
Taking logarithms on both sides, we obtain
Eg
ln Ri = ln C + (30.60)
2kT
A plot of ln Ri versus 1/T gives a straight
line, as shown in Fig. 30.12. The slope
of the straight line gives the value of Eg.
In practice we measure the resistance
of intrinsic semiconductor at different
temperatures with the help of four-point
probe technique and draw a plot between
ln Ri and 103/T. The slope of the straight
line thus obtained gives the value of Eg/
(2k). Hence,

 dy 
Eg = m(2k) =   2k (30.61)
 dx 
Fig. 30.12: Variation of resistance with
 dy 
where m  =  is the slope of the temperature in intrinsic semiconductor
 dx 
straight line.
Example 30.10: The resistivity of an intrinsic semiconductor is 4.5 W.m at 20°C and 2.0 W.m
at 32°C. Find the energy gap.
Solution:
 dy   log ρ2 − log ρ1 
Eg = 2k   = 2k  
 dx   (1 / T1 − 1 / T2 ) 
 log 2 − log 4.5 
= 2 × 8.61 × 10 −5 eV/K  K
 (1 / 293) − (1 / 305) 

 0.6532 − 0.3010 
= 2 × 8.61 × 10 −5 eV 
 0.134 × 10−3 
= 2 × 8.61 × 10–5 eV × 2.63 × 103 = 0.45 eV
Example 30.11. The resistivity of intrinsic silicon is 2.3 × 103 W-m at 300K. Calculate its
resistivity at 100°C. Assume Eg = 1 eV and k = 1.38 × 10–23 J/K.
872 A Textbook of Engineering Physics

Solution. The resistivity of an intrinsic semiconductor is given by ri = roeEg/2kT.

\ The resistivity of the intrinsic semiconductor at 300 K is given by r1 = roeEg/2k(300 K)
and the resistivity of the semiconductor at 373 K is given by r2 = roeEg/2k(373K).
 1eV 
exp  −5 
ρ2  2 × 8.61 × 10 eV × 373K 
\ =
ρ1  1eV 
exp  −5 
 2 × 8.61 × 10 eV × 300K 

 1eV 
exp 
\ r2 = ρ1  2 × 8.61 × 10 eV × 373K 
−5

 1eV 
exp 
 2 × 8.61 × 10 eV × 300K 
−5

 1eV 
exp 
= 2.3 × 103  2 × 8.61 × 10 eV × 373K  Ω.m
−5

 1eV 
exp 
 2 × 8.61 × 10−5 eV × 300K 
= 52 W.m.

30.14 LIMITATIONS OF INTRINSIC SEMICONDUCTOR

Intrinsic semiconductors are not useful for device manufacture because of low conductivity
and the strong dependence of conductivity on temperature. If we take a crystal of pure silicon
or germanium and connect it in the circuit, we will find that the current in the circuit will
gradually increase as the temperature of the crystal is increased. We would also expect the
current to increase if the voltage is increased and it does. But increasing the voltage increases
the current only proportionally, obeying Ohm’s law, while increasing the temperature increases
the current at an exponential rate. Thus the temperature over which we have no control exerts
more influence upon the current than the voltage, which we customarily do control.
We summarize the limitations as follows:
● Conductivity is low. Germanium has a conductivity of 1.67 S/m, which is nearly 107
times smaller than that of copper.
● Conductivity is a function of temperature and increases exponentially as the temperature
increases.
● Conductivity cannot be controlled from outside.
30.15 EXTRINSIC SEMICONDUCTORS
A judicious introduction of impurity atoms in an otherwise perfect semiconductor crystal
produces useful modifications of its electrical conductivity. It makes the current more voltage
dependent than temperature dependent. An intentional introduction of controlled amount of
impurity into an intrinsic semiconductor is called doping. The impurity added is called a
dopant. A semiconductor doped with impurity atoms is called an extrinsic semiconductor.
The impurity-produced electrons are not temperature-dependent but are voltage-dependent
and they will be under our control.
One of the important methods of doping is to add precisely determined quantities of
impurity to the melt from which the semiconductor crystal is grown. This way, a crystal
Semiconductors 873

having a constant impurity density is obtained. Typical doping levels range from 1020 to 1027
impurity atoms/m3. Pentavalent elements from Group V or trivalent elements from Group
III are used as dopants. The atoms belonging to these two groups are nearly of the same
size as silicon or germanium atoms and easily substitute themselves in place of some of the
host atoms in the semiconductor crystal. Thus, they are substitutional impurities and do not
cause any distortion in the original crystal structure. Depending on the two different types
of doping, two types of extrinsic semiconductors are possible. They are n-type and p-type
semiconductors.
Common Dopant Elements for Silicon and Germanium
n-type p-type
Phosphorous Aluminium
Arsenic Boron
Antimony Gallium
Indium
Advantages of Extrinsic Semiconductors
• Conductivity is high.
• Conductivity can be tailored to the desired value through the control of doping
concentration.
• Conductivity is not a function of temperature.
30.16 n -TYPE SEMICONDUCTOR
An n-type semiconductor is produced when a pure semiconductor is doped with a pentavalent
impurity such as phosphorous. A phosphorous atom has five valence electrons. Out of the
five electrons, only four participate in bonding with four host silicon atoms while the fifth
electron remains loosely bound. The host silicon lattice is a dielectric medium having a
dielectric constant of 12. As a result, the Coulomb force between the phosphorous nucleus
and the fifth electron is smaller than that it would be in free space. Therefore, the ionization
energy of the fifth electron is very small. It is Intrinsic
Donated
E electron
found to be 0.045 eV. The ionization energy is so electron
small that the thermal energy can easily liberate
the fifth electron from the nucleus. It means that Conduction Band
the energy levels corresponding to phosphorous
atoms are nearer to the bottom edge of the E C

conduction band. At normal temperatures, the

fifth electron becomes free to move about in the ED
+ +
crystal and acts as a charge carrier. That is, the Phosphorus atom Phosphorus ion
electron jumps into the conduction band leaving
behind the positive phosphorous ion that is
fixed in the crystal lattice. As the phosphorous Donor Energy level
atom is donating an electron for the purpose of
E
electrical conduction, it is called a donor atom. V

Valence band
Energy Band Diagram
The energy band diagram of n-type semicon-
ductor is shown in Fig. 30.13. If the donor atom
density is low, the donor atoms are distantly Intrinsic hole

spaced from one another, approximately, by 100 Fig. 30.13: Energy band diagram of n-type
semiconductor
874 A Textbook of Engineering Physics

atom spacings. In such a situation, the donor atoms cannot interact with each other and their
energy levels are discrete levels, ED. They are called donor levels and represent the ground
state of the fifth electron of impurity atom. As even small amount of thermal energy can
readily liberate the fifth electron from the atom and send it into the conduction band, the
donor levels are expected to be located very near to the bottom edge of the conduction band.
30.16.1 Temperature Variation of Carrier Concentration
The general dependence of electron concentration on temperature is shown in Fig. 30.14. At
0 K, the donor atoms are not ionized which means that all the donor electrons are bound to
the donor atoms. The conduction band is empty, while the valence band is full. The material
behaves essentially as an insulator.
At slightly elevated temperatures, the donor atoms are ionized and the donor electrons
go into the conduction band (Fig. 30.14, ionization region) by getting energy from lattice
vibrations. In this process holes are not produced in the valence band. At about 100K,
the donor levels are all ionized. Once all the donor atoms are ionized, further increase in
temperature does not produce electrons and the curve levels off. The plateau region is called
the depletion region. In the depletion region, the electron concentration in the conduction
band is nearly identical to the concentration of the dopant atoms.
If ND is the concentration of donor atoms, then
nn ≅ ND (30.62)
where nn is the electron concentration in the conduction band of n-type material.
In n Intrinsic
region
Extrinsic Extrinsic
complete incomplete
ionization ionization

Depletion Ionization
region region

III II I
1/T
Ti Td
Conduction Band Conduction Band Conduction Band

EC

ED + + + + + + + +

EV

Valence band Valence band Valence band

Fig. 30.14. Variation of electron concentration in an n-type semiconductor as

a function of the inverse of temperature.

As temperature increases further, some electrons from the valence band are also excited
into the conduction band (Fig. 30.14, depletion region). The conduction band, therefore,
Semiconductors 875

contains electrons that have come through two different processes-namely (i) donor atom
ionization and (ii) intrinsic process. The intrinsic process produces holes in the valence band.
At high enough temperatures the intrinsic behaviour takes over since large number of
electrons in the valence band gets thermally excited to conduction band and their number far
exceeds the number of donor electrons (Fig. 30.14, intrinsic region). Therefore, in intrinsic
region
nn = ni (30.63)
In an n type semiconductor at moderate temperatures, there are some 1010 electrons
and 1010 holes which have been generated due to heat-ruptured bonds in addition to the
1016 free electrons received from the donors. Donor electrons are in the majority and their
number is essentially constant at all usual temperatures; this is because they are not part of
covalent bonds and only 0.05 eV is required to free them. Thus, by doping we produce a less
temperature-sensitive semiconductor with an abundance of electrons.
In n-type material the electrons outnumber the holes and constitute the majority
carriers (in region II). Holes are minority carriers. The number of carriers is independent of
temperature in the depletion region. The current in this type of crystal is mainly due to the
negatively charged electrons and hence the material is called n-type semiconductor.
30.16.2 Carrier Concentration in n-type Semiconductor at Low
Temperatures: (In the Ionization Region)
The energy band diagram of a n-type semiconductor is shown in Fig. 30.13. Let ND be the
concentration of donors in the material. At 0K, the donor atoms are not ionized and are at the
level ED which is very near to EC. When the temperature is raised above 0°K, the donor atoms
get ionized and free electrons appear in the conduction band. With increase in temperature
more and more donor atoms get ionized and the electron concentration in the conduction band
increases. Electrons require an energy ED, for their transition to the conduction band from the
donor levels. Therefore, we may assume that the electron concentration, n, in the conduction
band is
n = N D+
or n = N D − N D0
where N D+ is the number of donor atoms that are ionized and N D0 is the number of atoms left
unionized at the energy level ED .
( )
The concentration of ionized donors N D+ = N D − N D0 = ND[1 −f(ED)] =
1+ e
ND
−( ED − EF )/ kT
ND
∴ n=
1+ e ( D F )
− E − E / kT

From the operational definition of Fermi level it is expected that the Fermi level in n-type
semiconductor lies a few kT above ED. Therefore, the above equation may be simplified as
n = N e( D F )
E − E / kT
D (30.64)
But the electron concentration, n, in the conduction band is given by
n = N e ( C F)
− E − E / kT
C (30.65)

N D e( D F ) = NC e ( C F )
E − E / kT − E − E / kT
∴
Taking logarithm and rearranging the terms we get
876 A Textbook of Engineering Physics

 ED − EF   EC − EF  NC
  +   = ln (30.66)
kT kT ND
NC
(ED + EC) − 2EF = ( kT ) ln
ND
ED + EC  kT  N
or EF = −   ln C
2  2 ND

ED + EC  kT  N
or EF = +   ln D
2  2 NC
ED + EC  kT  ND
or EF = +   ln (30.67)
 
( )
3/2
2 2 2 2π m*e kT /h 2
It is seen from equ. (30.67) that at T = 0K,
E + EC
EF = D (30.68)
2
That is, the equilibrium Fermi level lies midway between the bottom of the conduction
band and donor levels. Now,
 
 EF − EC   ED + EC  1  ND EC 
exp  = exp +   ln −
 kT   2kT  2
2(2π m*
kT /h 2 3/2
)kT 
 e 
 
E − EC  1  ND
= exp  D +   ln 
 2kT
( )
 2 3/2 
 2 2π m*e kT /h 2 
 
 E − EC +  ND  1 
= exp  D ln  2 ln x = ln x 
2 3/2 

2kT *
2 2π me kT /h (  )  

 
 ED − EC    ND   exp ( a + b ) = 
= exp    ⋅ exp  ln 3/2   
  2kT   2 2π m*e kT /h 2 ( )  exp ( a ) + exp (b )

 
 ED − EC    ND 
= exp  2kT   ⋅ [exp (ln x) = x]
2 3/2 
  
 ( *
2 2π me kT /h )


3/2  
 E − EC   2π me*kT   ED − EC    ND 
∴ n = NC exp  F = 2   exp  ⋅ 
 kT   h
2
  2kT    (*
2 2π me kT /h 2 3/2 
)
 

3/4
1  2π me*kT   E − EC  
or n = (2N D ) 2
  exp  D  (30.69)
 2kT  
2
 h 
Semiconductors 877

Thus, the electron concentration in the conduction band of an n-type semiconductor is

proportional to the square root of the donor concentration at moderately low temperatures.
Electrons are the majority carriers in an n-type semiconductor and the above expression
therefore gives the majority carrier concentration.
30.17 p-TYPE SEMICONDUCTOR
A p-type semiconductor is produced when a pure semiconductor is doped with a trivalent
impurity such as boron. Boron atom has three valence electrons. Therefore, it falls short of
one electron for completing the four covalent bonds with its neighbours. When an electron
from a neighbouring atom acquires energy and jumps into the vacancy to form the fourth
bond, it leaves behind a hole. The boron atom having acquired an additional electron becomes
a negative ion. The hole can move freely in the valence band whereas the impurity ion is fixed
in position by the covalent bonds. As the boron atom accepted an electron from the valence
band, it is called an acceptor atom. The acceptor impurity atoms produce holes without the
simultaneous generation of the electrons in the conduction band.
Energy Band Diagram
The energy band diagram of p-type semiconductor is shown in Fig. 30.15. If the acceptor
atom density is low, the acceptor atoms are distantly spaced from one another. As such, the
acceptor atoms cannot interact with each other and their energy levels are discrete levels,
EA. They are called acceptor levels and Intrinsic
represent the ground state of the hole. As electron
even small amount of thermal energy can E
make an electron in the valence band jump
Conduction Band
into the acceptor level, the acceptor levels
are expected to be located very near to the
E
top edge of the valence band. They are at C

Acceptor energy
about 0.01 eV above the valence band. A level
hole may be said to have moved from the Boron ion
E
acceptor atom to the valence band. with accepted Boron
g

electron atom
At 0 K the acceptor levels are vacant
and the valence band is full. The conduction E – –
band is also vacant. The material behaves
A

essentially as an insulator. E V

At slightly elevated temperature,

electrons from the valence band jump into Valence band

the acceptor levels and holes are generated in

the valence band (Fig. 30.15). In this process Intrinsic Hole left by the
holes are generated without the simultaneous hole electron accepted
by boron atom
generation of electrons. At normal tempera-
tures the acceptor levels are saturated and a Fig. 30.15. Energy band diagram for a p-type
few electrons are excited to the conduction semiconductor
band also. At about 100K, the acceptor atoms are all ionized. Once all the acceptor atoms
are ionized, further increase in temperature does not produce holes and we say the acceptor
levels are saturated. The region is called the saturation region. In the saturation region, the
hole concentration in the valence band is nearly identical to the concentration of the acceptor
atoms.
878 A Textbook of Engineering Physics

If NA is the concentration of acceptor atoms, then

pp ≅ NA (30.70)
where pp is the hole concentration in the valence band of p-type material.
As temperature is increased further, some electrons from the valence band are excited
into the conduction band. The valence band now contains holes that have been generated by
two different processes-namely, (i) acceptor atom ionization and (ii) intrinsic process. The
intrinsic process causes electrons to appear in the conduction band.
At high enough temperatures, a large number of electron-hole pairs are generated and
the number of holes generated thermally far exceeds the number of holes due to acceptor
impurity. The material behaves as an intrinsic semiconductor. In the intrinsic region
pp = ni (30.71)
In p-type material the holes outnumber the electrons and constitute the majority carriers.
Electrons are minority carriers. The number of majority carriers is independent of temperature
in the depletion region. The current in this type of crystal is mainly due to the positively
charged holes and hence the material is called p-type semiconductor.
30.17.1 Carrier Concentration in p-type Semiconductor at Low
Temperatures : (In the Ionization Region)
The energy band diagram of a p-type semiconductor is shown in Fig. 30.15. Let NA be the
concentration of acceptors in the material. At 0K, the acceptor atoms are not ionized and
are at the level EA which is very near to EV. When the temperature is raised above 0°K, the
acceptor atoms get ionized and holes appear in the valence band. With increase in temperature
more and more acceptor atoms get ionized and the hole concentration in the valence band
increases. Since transition of electrons to the acceptor levels from the valence band requires
an energy EA, they can go to acceptor levels and ionize acceptor atoms. Therefore, we may
assume that the hole concentration, p, in the conduction band is
p = N A–
–
where N A is the number of acceptor atoms that are ionized.
 E − EA 
The concentration of ionized acceptors N A– = NA f(EA) = N A exp  F
 kT 
 E − EA 
∴ p = N A exp  F (30.72)
 kT 
But the hole concentration, p, in the valence band is given by
 E − EF 
p = NV e − ( EF − EV )/ kT = NV exp  V (30.73)
 kT 
 E − EA   E − EF 
∴ N A exp  F  = NV exp  V
 kT   kT 
Taking logarithm and rearranging the terms we get
 E − E A   EV − EF  N
 F  −  = ln V
 kT   kT  NA
NV
− (EV + EA) + 2EF = ( kT ) ln
NA
EV + E A  kT  N
or EF = +   ln V
2  2  NA
Semiconductors 879

EV + E A  kT  N
or EF = −   ln A
2  2 NV
EV + E A  kT  NA
or EF = −   ln (30.74)
 2
( )
3/2
2 2 2π m*h kT /h 2
It is seen from equ. (30.74) that at T = 0K,
E + EA
EF = V (30.75)
2
That is, the equilibrium Fermi level lies midway between the bottom of the valence band
and acceptor levels. Now,
 
exp  V
 E − EF 
= exp  EV − EV + E A +  1  ln NA 
    3/2 
 kT 

kT 2kT  2 *
(
2 2πmh kT /h 2
 )
 
E − EA  1  NA
= exp  V +   ln 
3/2 

2kT  2
2 2πm*h kT /h 2 ( ) 

 
 E − EA NA   1 
= exp  V + ln  2 ln x = ln x 
2 3/2 

2kT
( *
2 2πmh kT /h  )  

 
 EV − E A    NA   exp ( a + b ) = 
= exp    ⋅ exp  ln 3/2   
  2kT   2 2 π m*h kT /h 2 ( )  exp ( a ) + exp (b )

 
 EV − E A    NA 
= exp  ⋅    exp (ln x ) = x 
 2kT    ( )
3/2
2 2π m*h kT /h 2 


3/2  
 E − EA   2π mh* kT   EV − E A    NA 
∴ p = NV exp  V  = 2   exp  ⋅  2 3/2 
 2kT    ( )
2
 kT   h  *
2 2 π mh kT /h 

3/4
1  2π mh* kT   E − E A  
or p = (2N A ) 2
  exp  V  (30.76)
 2kT  
2
 h 

Thus, the hole concentration in the valence band of a p-type semiconductor is proportional
to the square root of the acceptor concentration at moderately low temperatures. Holes are the
majority carriers in a p-type semiconductor. Therefore, the above expression represents the
majority carrier concentration in a p-type semiconductor.
880 A Textbook of Engineering Physics

30.18 BAND DIAGRAMS OF EXTRINSIC SEMICONDUCTORS

AT 0K AND 300K
The energy band diagrams of extrinsic semiconductors at 0K and 300K showing the position
of Fermi level in each case is shown in Fig. 30.16 and Fig. 30.17 respectively.
T = OK T = 300K
E E
Conduction Band Conduction Band

EC
EFn ED + + + + +
ED EFn

EFi Eg/2 EFi Eg/2

EV EV

Valence band Valence band

Distance Distance
(a) (b)
Fig. 30.16 Energy band diagram of an n-type semicondcutor (a) at 0K (b) at 300K
T = 0K T = 300K
E E
Conduction Band Conduction Band

EC EC

EFi Eg/2 EFi Eg/2

EFP
EA EA – – – – –
EFp
EV EV

Valence band Valence band

Distance Distance
(a) (b)
Fig. 30.17 Energy band diagram of an p-type semicondcutor (a) at 0K (b) at 300 K

30.19 EXTRINSIC CONDUCTIVITY

The temperature range in which a doped semiconductor used is the extrinsic region where
all the impurity atoms are ionized (Region II, Fig. 30.14). This region is of practical interest
since the carrier concentration is essentially independent of temperature in this region and
any desired conductivity can be achieved by controlling the amount of impurities added from
Semiconductors 881

outside. Note that whenever extrinsic semiconductor is discussed, we invariably imply its
characteristics in the extrinsic region. The conductivity of n-type semiconductor is given by
sn = nneme + pnemh
As pn << nn, the second term is negligible and it is the electrons that contribute to the
conductivity. Therefore,
sn = nneme
As nn = ND, we can write the above relation as
sn = NDeme (30.77)
Similarly, the conductivity of p-type semiconductor is given by
sp = ppemh
or sp = NAemh (30.78)
A general dependence of conductivity on temperature in extrinsic semiconductors is
shown in Fig. 30.18.
–Eg
Slope =
2k

Extrinsic

(Ec – Ed)
Slope =
2k
In 

Intrinsic

1/T
Low Temp.
Fig. 30.18

Example 30.12. In a doped semiconductor, there are 4.52 × 1024 holes and 1.25×1014electrons
per cubic metre. What will be the carrier density in undoped specimen? Electron and hole
mobilities are 0.38 m2/V.s and 0.18 m2/V.s respectively. Calculate the conductivity of intrinsic
and the doped semiconductors.
Solution. Intrinsic carrier density, ni = p p n p

∴ ni = (4.52 × 10 24
)(
/ m3 1.25 × 1014 / m3 )
= (5.65 ×1038 / m6)1/2 = 2.38 × 1019 /m3
Conductivity of the intrinsic semiconductor
si = eni (me + mh)
m2
(
= 1.602 × 10−19 C 2.38 × 1019 / m3 ( 0.38 + 0.18) )( ) V.s
C
= 2.14 = 2.14 S/m
m.V.s
Conductivity of the doped semiconductor
sp = eppmh
882 A Textbook of Engineering Physics

m2
(
= 1.602 × 10−19 C 4.52 × 1024 / m3 (0.18) )( ) V.s
C
= 1.30 × 105 = 130 kS/m
m.V.s
C A
Unit: 1 =1 = 1W−1 m−1 = 1 S /m
m.V.s mV
Example 30.13: Silicon has a conductivity of only 5 × 10-4 W-1m-1 in its pure form. An
engineer wanted it to have conductivity of 200 W-1m-1 and doped it with aluminium to
produce p-type semiconductor. Calculate the impurity concentration. Assume mh=0.05 m2/V.s.
Solution. s = pemh = NAemh.
σ 200 Ω −1m –1
\ NA = =
eµ h 1.602 × 10−19 C × 0.05 m 2 /V.s
= 2.5 × 1022 atoms/m3.

30.20 LAW OF MASS ACTION

In case of intrinsic semiconductors the product of n and p is a constant for a certain semicon-
ductor at a certain temperature and is given by
− E / kT
np = ni2 = NC NV e g
There is no condition in the expression that restricts it to intrinsic semiconductors because
Eg does not change with impurity concentration and NC and NV are constants. This product is
therefore a constant equally valid for intrinsic as well as for extrinsic semiconductors.
The electron and hole concentrations in extrinsic semiconductors may be given by
expressions similar to equ. (30.30) and (30.41). Denoting the electron concentration in n-type
semiconductor by nn and the hole concentration by pn, we can write
n = N e ( C F)
− E − E / kT
n C (30.79)
pn = NV e ( F V )
− E − E / kT
and (30.80)
− Eg / kT
∴ nn pn = NC NV e (30.81)
∴ nnpn = ni2 (30.82)
Similarly, if we denote the hole concentration in a p-type semiconductor by pp and
electron concentration by np, we can express
pp = NV e ( F V )
− E − E / kT
(30.83)

np = NC e ( C F )
− E − E / kT
and (30.84)
− Eg / kT
∴ ppnp = NC NV e (30.85)
∴ ppnp = ni2 (30.86)
The relations (30.82) and (30.86) show that the product of majority and minority carrier
concentrations in an extrinsic semiconductor at a particular temperature is a constant and is
equal to the square of intrinsic carrier concentration at that temperature.
Semiconductors 883

The law of mass action is very important relation because it in conjunction with charge
neutrality condition enables us calculate minority carrier concentration. The law suggests that
the addition of impurities to an intrinsic semiconductor increases the concentration of one
type of carrier, which consequently becomes majority carrier and simultaneously decreases
the concentration of the other carrier, which as a result becomes minority carrier. The minority
carriers decrease in number below the intrinsic value because the majority carriers increase
the rate of recombinations. The law of mass action states that the product of majority and
minority carriers remains constant in an extrinsic semiconductor and it is independent of the
amount of donor and acceptor impurity concentrations. Note that when the doping is heavy,
the minority carrier concentration will be low and if doping is lighter, the minority carrier
concentration will be larger.

30.21 CHARGE NEUTRALITY CONDITION

An extrinsic semiconductor is an electrically neutral body in its equilibrium condition. In
n-type semiconductor, the total number of electrons in the conduction band must be equal
to the sum of electrons originated from the donor atoms and electrons excited from the
valence band. Electrons coming from donor levels leave behind positive donor ions while
electrons excited from the valence band leave behind holes. These processes have not created
any additional charges, and the equality between positive and negative charges remain
undisturbed. The charge neutrality condition applied to the n-type semiconductor implies
that the total negative charge of mobile electrons is equal to the total positive charge created
in the crystal. It means that
nn = ND + pn
where ND is the donor impurity concentration and all the donor atoms are assumed to have
got ionized.
But nn >> pn
\ nn = ND (30.87)
The above relation indicates that the majority carrier concentration, nn, in an n-type
semiconductor is equal to the donor impurity concentration, ND.
The charge neutrality for a p-type semiconductor requires that
pp = NA + np
As pp >> np
pp = NA (30.88)
30.22 CALCULATION OF MINORITY CARRIER CONCENTRATION
In the case of an n-type semiconductor, the majority carrier concentration nn is given by
nn = ND
where ND is the donor impurity concentration and all the donor atoms are assumed to have
got ionized. The above relation indicates that the majority carrier concentration, nn, in an
n-type semiconductor is equal to the donor impurity concentration, ND. The minority carrier
concentration, pn, is then given by eq. (30.82), as
n2
pn = i
nn
ni2
or pn = (30.89)
ND
884 A Textbook of Engineering Physics

Similarly, in case of a p-type semiconductor, the minority carrier concentration, np is

given by
n2
np = i
pp

ni2
or np = (30.90)
NA
Example 30.14. A sample of intrinsic germanium at room temperature has a carrier
concentration of 2.4 × 1019 / m3. It is doped with antimony at a rate of one antimony atom per
million atoms of germanium. If the concentration of the germanium atoms is 4 × 1028 /m3,
determine the hole concentration.
4 × 1028 atoms/m3
Solution. ND = = 4 ×1022 donors/m3
106 atoms/m3
nn = ND = 4 ×1022 electrons/m3

∴
n2
pn= i =
(2.4 × 10 carriers/m )
19 3 2

nn (4 × 10 electrons/m )
22 3

or pn = 1.4 ×1016 holes/m3.

Example 30.15: A sample of intrinsic silicon at room temperature has a carrier concentration
of 1.5 × 1016/m3. A donor impurity is added to the extent of 1 donor atom per 108 atoms of
silicon. If the concentration of silicon atoms is 5 × 1028atoms/m3, determine the resistivity of
the material. (Given me=0.135 m2/V.s and mh=0.048 m2/V.s)
Solution:
N 5 × 1028 /m3
Donor atom density is given as ND = = = 5×1020 /m3
108 108
Free electron concentration is given by nn = ND = 5 × 1020 /m3
ni2 (1.5 × 1016 /m3 )2
Hole concentration is given by pn = = = 4.5×1011/m3.
nn 5 × 1020 /m3
1 1
Resistivity of the material r = =
σ e ( nn µ e + pn µ h )
1
= = 0.092 Wm.
(1.602 × 10 −19
)(
C 5 × 10 × 0.135m 2 /V.s + 4.5 × 1011 × 0.048 m2 /V.s
20
)
Example 30.16: A sample of intrinsic germanium at room temperature has a carrier
concentration of 2.4 × 1019/m3. It is doped with antimony at a rate of one antimony atom
per million atoms of germanium. If the concentration of the germanium atoms is 4 × 1028/m3,
determine the hole concentration.
Solution. N 4 × 1028 /m3
Donor atom density is given as ND = 6 = = 4×1022 /m3
10 106
Free electron concentration is given by nn= ND = 4 × 1022 /m3
Semiconductors 885

ni2 (2.4 × 1019 /m3 )2

Hole concentration is given by pn = = = 1.4×1016/m3.
nn 4 × 1022 /m3

30.23 FERMI LEVEL IN EXTRINSIC SEMICONDUCTORS

In intrinsic semiconductors, the Fermi level, EF lies in the middle of the band gap. However,
the situation is different with extrinsic semiconductors. In an n-type semiconductor, the Fermi
level lies in the upper half of the gap, as the majority carriers reside in the conduction band
and their average energy is more than EFn. In a p-type semiconductor, the Fermi level lies
in the lower half of the gap, as the majority carriers reside in the valence band and their
average energy is less than EFp. The carrier concentrations in extrinsic semiconductors vary
with temperature and impurity concentration. It means that the probability of occupancy of
respective bands varies and consequently the position of Fermi level changes with temperature
and impurity concentration.
(a) Variation of Fermi Level with Temperature in an n-type
Semiconductor
In the n-type semiconductor at low temperatures, some donor atoms are ionized and provide
electrons to the conduction band while others remain neutral. As electrons in the conduction
band are only due to the transitions from the donor levels, the Fermi level must lie between
the impurity donor levels and the bottom of the conduction band. When T = 0K, EFn lies
midway between the donor levels and the bottom of the conduction band. It is thus,
E + ED
EFn = C at T = 0 K (30.91)
2
As the temperature increases the donor levels gradually get depleted and the Fermi level
moves downward. At the temperature of complete depletion of donor levels, Td, the Fermi
level coincides with the donor level ED. Thus
EFn = ED at T = Td (30.92)
As the temperature grows further above Td, the Fermi level shifts downward in an
approximately linear fashion. At a temperature Ti, the intrinsic process contributes to electron
concentration significantly. At higher T T
temperatures, the n-type semiconductor E
d i

loses its extrinsic character and behaves as Conduction Band

an intrinsic semiconductor. In the intrinsic E
E
region, the electron concentration in
C
Fn
E
D
conduction band increases exponentially Td T
i
E
E = 2
g

and the Fermi level approaches the intrinsic

Fi

value. Thus, E
V

Eg Valence band
EFn = EFi = at T ≥ Ti (30.93)
2 T[K]
The variation of Fermi level EFn in an 100 200 300 400

n-type semiconductor with temperature is Fig. 30.19: Qualitative dependence of Fermi

level on temperature in an n-type semiconductor
illustrated in Fig. 30.19.
886 A Textbook of Engineering Physics

Variation of Fermi Level with

Temperature in a p-type
Semiconductor
In case of p-type semiconductor, in the low
temperature region, holes in the valence band
are only due to the transitions of electrons
from the valence band to the acceptor levels.
As the valence band is the source of electrons
and the acceptor levels are the recipients for
them, the Fermi level must lie between the
top of the valence band and the impurity
acceptor levels. When T = 0, Fermi level lies
midway between the acceptor levels and the Fig. 30.20: Qualitative dependence of Fermi
top of the valence band. Thus, level on temperature in a p-type semiconductor

EV + E A
EFP = (30.94)
2
As the temperature increases the acceptor levels gradually get filled and the Fermi level
moves upward. At the temperature of saturation Ts, the Fermi level coincides with the acceptor
level EA. Thus,
EFP = EA at T = Ts (30.95)
As the temperature grows above Ts, the Fermi level shifts upward in an approximately
linear fashion.
At a temperature Ti intrinsic behaviour sets in. At higher temperatures, the p-type
semiconductor loses its extrinsic character and behaves as an intrinsic semiconductor. In the
intrinsic region, the hole concentration in the valence band increases exponentially and the
Fermi level approaches the intrinsic value. Thus
Eg
EFP = Ei = at T = Ti (30.96)
2
The variation of Fermi level EFn in an p-type semiconductor with temperature is illustrated
in Fig. 30.20.
30.24 VARIATION OF FERMI LEVEL WITH IMPURITY CONCENTRATION
n-type Semiconductor
The addition of donor impurity to an intrinsic semiconductor leads to the formation of discrete
donor levels below the bottom edge of the conduction band. At low impurity concentra-
tions, the impurity atoms are distantly spaced from one another, approximately, by 100 atom
spacings. Therefore, they do not interact with each other. With an increase in the impurity
concentration, the separation between impurity atoms decreases and they tend to interact. As a
result, the donor levels undergo splitting and form an energy band below the conduction band,
as shown in Fig. 30.21.
The larger the doping concentration, the broader is the impurity band; and at one stage
the impurity band overlaps on the conduction band. Then the upper vacant levels in the
conduction band are accessible to the donor electrons. The broadening of donor levels into a
band is a accompanied by a decrease in the width of the forbidden gap and also by the upward
displacement of Fermi level. The Fermi level shifts closer and closer to the conduction band
Semiconductors 887

with increasing impurity concentration and finally moves into the conduction band when the
donor band overlaps on the conduction band.
E
Conduction Band Conduction Band Conduction Band

EC EF
EC EC
EF
EF
ED ED

Eg
Eg
Eg
EV EV EV

Valence band Valence band Valence band

(a) (b) (c)

Fig. 30.21: Energy band diagrams of an n-type semiconductor at three different levels
of doping; (a) low level doping; (b) medium doping; (c) heavy doping.

p-type Semiconductor
In p-type semiconductor, the acceptor levels broaden and form into a band with increasing
impurity concentration. The acceptor band ultimately overlaps on the valence band. The Fermi
level moves down closer to the valence band and finally at very high impurity concentration it
will shift into the valence band.

30.25 DRIFT AND DIFFUSION CURRENTS

Under the condition of thermal equilibrium, the electrons and holes are uniformly distributed
in the crystal and in the absence of an external stimulus their average velocity is zero and
no current flows through the crystal. This is equally true for an intrinsic or an extrinsic
semiconductor.
The thermal equilibrium may be disturbed by an external agent and the chaotic motion
of charge carriers acquire a directional movement leading to a flow of current in the material.
Electric field and concentration gradients are examples of such disturbing agents.
1. Drift Current: When an electric field E is applied across a semiconductor, the charge
carriers acquire a directional motion over and above their thermal motion and produce drift
current.
The electrons drifting in the conduction band produce a current component Je given by
Je (drift) = nemeE
The holes drifting in the valence band cause a current component Jh given by
Jh (drift) = pemhE
Therefore, the total drift current density is,
Jdr = Je (drift) + Jh (drift)
888 A Textbook of Engineering Physics

Drift current occurs only when external electric field is present across the solid. Although
electrons and holes move in opposite directions, the direction of conventional current flow
due to both the carriers is in the same direction.
\ Jdr = e(nme + pmh)E (30.97)
2. Diffusion Current: In case of semicon- p-type
ductors, current can also flow without the
application of an external electric field. If a
spatial variation of carrier density is created
in the semiconductor, current flows in it. If we
consider an arbitrary surface in the volume of the
solid and if there are more charge carriers on its
one side than on the other side, we say there is
a concentration gradient. This concentration n(0)
gradient causes a directional movement of charge
carriers, which continues until all the carriers
are evenly distributed throughout the material.
Any movement of charge carriers constitutes 0 x (or t)
an electric current, and this type of movement Fig. 30.22
produces a current component known as
diffusion current.
Concentration gradient may be produced in an extrinsic semiconductor by applying heat
or light locally at one region. Suppose an external agent such as light or heat acts momentarily
at one end of a p-type semiconductor, as shown in Fig. 30.22. The external agent generates
additional electron-hole pairs leading to a sudden increase in the concentration of charge
carriers at that end. In the rest of the volume, the concentration of carriers is at equilibrium
value. The difference in the concentration of charge carriers initiates the carriers to diffuse
from the region of higher concentration to the region of lower concentration in order to restore
the equilibrium condition. As the carriers are charged particles, their migration produces a
current flow, which is the diffusion current. The diffusion current strength is proportional to
the concentration gradient, i.e. the rate of change of carrier concentration per unit length. In
case of electrons moving left to right (see Fig. 30.22), current flows from right to left in the
negative x-direction.
The current component due to electron diffusion is given by
dn
Je (diff) = eDe (30.98)
dx
The current component due to hole diffusion is given by
dp
Jh (diff) = −eDh (30.99)
dx
De and Dh are diffusion coefficients for electrons and holes respectively.
Drift and diffusion currents coexist in semiconductors. The total current density due to
drift and diffusion of electrons may be written as
 dn 
Je = e  nµe E + De  (30.100)
 dx 
Semiconductors 891

30.28 HALL EFFECT

If a metal or a semiconductor carrying a current I is placed in a transverse magnetic field B,
a potential difference VH is produced in a direction normal to both the magnetic field and
current directions. This is known as Hall effect. This effect was discovered by E.H. Hall in
1879 and showed that it is negatively charged particles that carry current in metals.
30.28.1 Importance of Hall Effect
The importance of Hall effect in the field of semiconductors is that it helps to determine
(i) the type of semiconductor,
(ii) the sign of majority charge carriers,
(iii) the majority charge carrier concentration,
(iv) the mobility of majority charge carriers, and
(v) the mean drift velocity of majority charge carriers.
30.28.2 Experimental Arrangement
The experimental set up for the measurement of Hall voltage and determination of Hall
coefficient is shown in Fig. 30.23 (a). A thin rectangular semiconductor wafer is mounted on
an insulating strip and two pairs of electrical contacts are provided on opposite sides of the
wafer. One pair of contacts is connected to a constant current source. And the other pair is
connected to a sensitive voltmeter. This arrangement is mounted in between two pole pieces
of an electromagnet such that the magnetic field acts perpendicular to the lateral faces of the
semiconductor wafer.
30.28.3 Hall Voltage
Let us assume that the semiconductor is a p-type semiconductor. Let a potential difference
V be applied across its ends. A current of strength I flows through it along the x-direction
(Fig. 30.23 b). Holes are the majority charge carriers in the p-type semiconductor. The current
through the wafer is given by
I = peAud (30.111)
where p is the hole concentration
A is the area of cross-section of the end face of semiconductor wafer,

+
–

V
VH + –
– VH
+
B
S F'
B EH
I t
E
N Ex
H w
+ + + + ++ ++ ++ ++
Semiconductor
wafer p-type
F
semiconductor
Electro V
+ –
magnet
(a) (b)
Fig. 30.23: (a) Basic experimental arrangement to study Hall effect (b) Generation of Hall voltage.
892 A Textbook of Engineering Physics

e is the electrical charge associated with a hole, and

ud is the average drift velocity of holes.
I
The current density Jx = = peυd (30.112)
A
Any plane perpendicular to the current
flow direction is an equipotential surface. F´
Therefore, the potential difference between p-type
the front and rear faces F and F’ is zero (see I
F
I
Fig. 30.24).
L

E
Now, if a magnetic field B is applied
H

F
normal to the wafer surface and hence to + + + + + + + + + +
the direction of current flow in it, then a
transverse potential difference is produced Fig. 30.24: Top view of the wafer—The
between faces F and F ′. It is known as Hall directions of magnetic force and
Voltage VH. Hall field in p-type semiconductor
Before the application of magnetic
field, holes move parallel to faces F and F ′. On application of magnetic field B, the holes
experience a sideways deflection due to the magnetic force FL, which is given by
FL = eBud (30.113)
Holes are deflected toward the front face F and pile up there. Due to this, a corresponding
equivalent negative charge is left on the rear face F′. These opposite charges produce a
transverse electric field, EH, whose direction is from the front to the rear face. Due to the
action of EH, holes experience an electric force in addition to the Lorentz force. When the
force FE due to this transverse electric field balances the magnetic force FL, equilibrium
condition is attained and the holes once again flow along x-direction parallel to the faces F
and F’.
In the equilibrium condition
FE = FL
\ eEH = eudB
V
If ‘w’ is the width of the semiconductor wafer, EH = H
w
 VH 
∴   = Bud (30.114)
w
Jx
From equ. (30.112), we have ud =
pe
Therefore, we can write equ. (30.114) as
VH BJ x
∴ = (30.115)
w pe
wBJ x wBI
VH = =
pe peA
If ‘t’ is the thickness of the semiconductor plate, A = w t.
BI
\ VH = (30.116)
pet
Semiconductors 893

30.28.4 Hall Coefficient

Hall coefficient, RH is defined as Hall field per unit current density per unit magnetic
induction. Thus,
E V /w
RH = H = H
JxB JxB

Using equ. (30.114), we get

BJ x
RH =
peJ x B
1
or RH = (30.117)
pe
Hall voltage, VH can now be written as
BI
VH = RH (30.118)
t

VH t
\ RH = (30.119)
BI
The above equation is derived assuming that the p-semiconductor contains only holes.
When the two types of charge carriers are taken into account, it is shown that the Hall
coefficient is given by

RH =
(
pµ 2h − nµ e2 ) (30.120)
e ( pµ h + nµ e )
2

From the above equation, we see that the Hall coefficient and Hall voltage are smaller for
intrinsic materials than for extrinsic materials.
30.28.5 Drift Velocity
According to the equilibrium condition, namely FE = FL, we have
V 
e  H  = eBud
 w
VH
ud = (30.121)
Bw
As it is possible to experimentally measure the Hall voltage, magnetic field B and width
of the wafer w, the mean drift velocity of the charge carriers can be obtained from the above
equation.
30.28.6 Carrier Concentration
From the measurement of Hall voltage and current through the wafer and knowing the strength
of magnetic induction and thickness of the wafer, RH can computed using equ. (30.119). Once
RH is known, the carrier concentration can be determined with the help of equ. (30.108),
which may be rewritten as follows:
1
p=
RH e
In case of conductors and n-type semiconductors, the expression for Hall coefficient is
894 A Textbook of Engineering Physics

1
RH = − (30.122)
ne
where n represents the concentration of electrons in the conductor or n-type semiconductor.
1
\ n= − (30.123)
RH e

30.28.7 Doping Level

It is possible to estimate the doping concentration, NA, from the value of p using the relation
pp @ NA.
30.28.8 Carrier Sign and Type of Semiconductor
With the direction of the magnetic field and the current as depicted in Fig. 30.23 (b), the sign
of the Hall voltage is positive. For an n-type semiconductor, the Hall voltage will be negative
when the directions of B and I are kept same. Therefore, knowing the sign of Hall voltage, the
type of the semiconductor and hence the sign of the majority carrier can be determined.
30.28.9 Hall Mobility
Mobility is defined as the drift velocity acquired in unit electric field. We know that the
current density J is given by the following expressions.
J = peud and J = sE
\ peud = sE
υd σ
or =
E pe
\ mh = s RH (30.124)
30.28.10 Hall Angle
The net electric field E in the semiconductor is a vector sum of Ex and EH. It acts at an angle
qH to the x-axis, where qH is called the Hall angle. From the Fig. 30.23 (b), we have
E
tan qH = H (30.125)
Ex
V BJ x
But EH = H =
w pe
J
Also Ex = x = ρJ x
σ
BJ x
Thus, tan qH =
peρJ x
B
\ tan qH =
peρ
EH B
= (30.126)
Ex peρ
1 1
Using = RH and = σ , equ. (30.125) can be rewritten as
pe ρ
tan qH = sRHB
or tan qH = mhB (30.127)
Semiconductors 895

30.28.11 Factors Affecting Hall Voltage

Hall voltage is given the following expression.
BI
VH =
pet
It is seen from the above equation that Hall voltage is directly proportional to the magnetic
field strength and the current passing through the wafer. Therefore, the Hall voltage will be
larger, the larger the magnetic field or the current. Further, the Hall voltage is larger, the
smaller is the carrier concentration, and the thinner is the wafer. It is easy to select thin wafers
of the material, so making t small. The fact that the concentration of carriers must be small
implies that large Hall voltage is obtained with poor conductors, i.e. those with relatively
few free electrons or holes per unit volume. Thus, the best voltage is obtained with semicon-
ductors like silicon and germanium that have been lightly doped with impurities to make them
conducting slightly.
For the semiconductors the number of charge carriers per unit volume is about 1023/m3
while in case of metals it is about 1028/m3. Therefore, the Hall voltage is about 105 greater in
semiconductors than in metals.
30.28.12 Variation of Hall Coefficient with Temperature
In metals it was found that the Hall coefficient does not depend on temperature. Hence, it
may be concluded that the concentration of free electrons does not vary with temperature in
metals.
In semiconductors, the Hall coefficient decreases sharply with increase in temperature. It
indicates that the concentration of charge carriers in semiconductors increases with increasing
temperature.
30.28.13 Applications
Hall effect is widely used in various fields for a variety of applications. In almost all cases a
Hall effect sensor is employed. A Hall effect sensor is a transducer that produces its output
voltage in response to changes in magnetic fields.
1. Determination of semiconductor type: The Hall coefficient is negative for a p-type
semiconductor and positive for a p-type semiconductor. Therefore, the sign of the Hall
coefficient can be used to determine whether a given semiconductor is n- or p-type.
2. Determination of carrier concentration: By measuring the Hall coefficient, the carrier
concentration in a semiconductor can be determined making use of the relations
1 1
n= or p = .
RH e RH e
3. Determination of carrier mobility: By measuring the Hall coefficient and conductivity
of the semiconductor, the carrier mobility can be determined using the relation
mh = s|RH|
4. Measurement of magnetic fields: Hall voltage is proportional to the magnetic field
intensity, for a given current through the sample. Therefore, one of the important
applications of Hall effect consists in measuring magnetic fields. Knowing the
parameters of the Hall probe, and applied current, we can determine the intensity of the
magnetic field. Hall probes can be used for static as well as high-frequency magnetic
fields. Hall probes measure variable magnetic fields up to a frequency of 1012 Hz.
896 A Textbook of Engineering Physics

5. Measurement of power in an electromagnetic wave: In an electromagnetic wave in

free space, the electric and magnetic fields are at right angles. Now, if a semiconductor
is kept parallel to E, it will produce a current I in the semiconductor. Since the semicon-
ductor is simultaneously subjected to a transverse magnetic field, Hall voltage is
produced across the sample. The Hall voltage is proportional to the product EH which
represents power of the wave. Thus, Hall effect can be used to determine the flow of
power of an electromagnetic wave.
6. Miscellaneous applications: Hall sensors are used for proximity switching, positioning,
speed detection, and current sensing applications.
Current flowing through a conductor produces a magnetic field that varies with current,
and a Hall sensor can be used to measure the current without interrupting the circuit. They are
especially used in measuring extremely heavy currents, where conventional ammeters cannot
be used.
A Hall sensor is combined with circuitry to act as a digital switch. Such switches are used
in consumer equipment; for example in computer printers to detect missing paper and open
covers.
Example 30.18: An n-type germnanium sample has a donor density of 1021/m3 . It is arranged
in a Hall experiment having magnetic field of 0.5 T and the current density is 500 A/m2. Find
the Hall voltage if the sample is 3 mm wide.
BI BJA BJ wt
Solution: Hall voltage VH = = =
net net net
As n = ND
BJw
∴ VH =
ND e
(0.5 T ) (500A/m 2 )(3 × 10−3 m)
=
(10 21
)(
/m3 1.602 × 10−19 C )
= 4.7×10−3 V = 4.7 mV
TAm 2 N A.m 2 N .m J
Units: 1 = 1. . = 1. =1 =1V
C A.m C C C

Example 30.19. A copper strip 2.0cm wide and 1.0mm thick is placed in a magnetic field with
B=1.5 wb/m2. If a current of 200 A is set up in the strip, calculate Hall voltage that appears
across the strip. Assume RH= 6 × 10-7m3/C.
IB
Solution. Hall voltage VH = RH
t
200 A × 1.5 wb/m 2
= 6 × 10−7 m3 /C ⋅
10−3 m
= 0.18V
Example 30.20. An electric field of 100 V/m is applied to a sample of n-type semiconductor
whose Hall coefficient is -0.0125 m3/C. Determine the current density in the sample, assuming
me= 0.6 m2 / V.s.
Semiconductors 897

µe E
Solution. Current density J=
RH

=
( 0.36 m 2
)
/V.s (100 V/m )
–0.0125 m3 /C
= - 2880 C/s.m2 = - 2880 A/m2.
Example 30.21. In a Hall coefficient experiment, a current of 0.25 A is sent through a metal
strip having thickness 0.2 mm and width 5 mm. The Hall voltage is found to be 0.15 mV when
a magnetic field of 2000 gauss is used.
(a) What is the carrier concentration?
(b) What is the drift velocity of the carriers?
Solution. (a) The carrier concentration,
IB
n=
VH e t

=
(0.25 A )(0.2 T )
(0.15 × 10 V)(1.602 × 10
–3 −19
)(
C (0.2 × 10 –3m )
= 1.04×1025 carriers/m3.
(b) Drift velocity of the carriers,
V
ud = H
wB
0.15 × 10−3 V
=
5 × 10−3 m × 0.2 T
= 0.15 V/m.T = 0.15 m/s.

QUESTIONS
1. Show that the intrinsic concentration ni for a semiconductor is given by
− Eg /2kT
ni = ( N C NV )1/2 e
where the symbols have their usual meaning.
2. Write the expressions for electron and hole concentrations in an intrinsic semiconductors and
hence derive the expression:
EC + EV 3  m* 
EF = + kT ln  *h 
2 4  me 
for Fermi level in the intrinsic semiconductor. Assume the symbols to have their usual meanings.
3. Draw a neat sketch of a band diagram of intrinsic semiconductor at room temperature and show
that the Fermi level in an intrinsic semiconductor lies in the middle of the energy gap.
(R.T.M.N.U., 2007)
4. Derive an expression for Fermi energy in intrinsic semiconductor. What is the effect of
temperature on Fermi level in an intrinsic semiconductor?
5. Describe a method of determining the band gap of a semiconductor. How does electrical
conductivity vary with temperature for an intrinsic semiconductor? (Anna Univ., 2004)
Superconductivity 1031

C H A P T E R

35 Superconductivity

35.1 INTRODUCTION
Superconductivity is a state of matter exhibited usually at very low temperatures where the
resistivity of the material drops to zero. The superconducting state is a state in which quantum
mechanics operates on a macroscopic scale of the order of many atomic distances rather than
the usual atomic and subatomic scale. The superconducting state is influenced by temperature,
current and magnetic field. There exist critical values for these three parameters, above which
the material passes into normal state. Besides being of immense theoretical interest, the
phenomenon has many possible practical applications. Superconducting magnets are designed
for use in the fields of medicine and particle physics. It is being vigorously tried for use in
transportation and transmission of power. Until recent times the main hurdle in the way of
their use is the requirement of extremely low temperatures of the order of 20 K and less.
Recent discoveries of high temperature ceramic superconductors raised the hopes of using
superconductors in making more efficient and smaller electrical and electronic devices at
normal temperatures. Extensive research is being carried out all over the world to improve the
properties of ceramic superconductors to make them suitable for various practical applications.

35.2 SUPERCONDUCTIVITY
Phenomenon
Metals are good conductors of electricity as they contain a tremendous number of free
electrons. The low resistance offered by them to the flow of current is attributed to the
scattering of the free electrons by vibrating ions of the lattice. When temperature increases,
the amplitude of the lattice vibrations increase and cause more scattering of electrons leading
to more resistance. Even at 0 K, metals offer finite resistance, called residual resistance,
which is attributed to the scattering of electrons by impurities and crystal defects present in
the material. The variation of resistivity of normal metals with temperature is shown in Fig.
35.1, which indicates the existence of residual resistance, rr. H.K.Onnes was verifying the
behaviour of metals at very low temperatures and in 1911, he discovered that the electrical
resistance of highly purified mercury dropped abruptly to zero at a temperature of 4.15 K (see
Fig. 35.2). The sudden drop in resistivity was quite unexpected and Onnes recognized it to
be an entirely new phenomenon. Onnes also found that the transition was reversible. When
heated above the transition temperature 4.15 K, mercury regained its resistivity. Onnes named
the phenomenon as superconductivity. Subsequently, superconductivity was discovered in
lead, zinc, aluminium, and other metals as well as in a number of alloys. Superconductivity
1032 A Textbook of Engineering Physics

was strictly a low temperature phenomenon till 1980’s, when certain ceramic materials were
found to exhibit superconductivity at higher temperatures of about 120 K.

0.125

Resistance (ohms)
Electrical Resistivity

Superconductor

0.025
–5
10 
Pr Normal metal

4.1 4.2 4.3

0 TC
Temperature (K) Temperature (K)
Fig. 35.1                 Fig. 35.2
Superconductors
Superconductivity is the phenomenon in which electrical resistance of materials
suddenly disappears below a certain temperature. The materials that exhibit superconductivity
and which are in the superconducting state are called superconductors.
Transition temperature
The temperature at which a normal material abruptly changes into a superconductor is
called transition temperature, TC. It is also known as the critical temperature.
35.3 MATERIALS (LOW TC MATERIALS)
Since the discovery of superconductivity in the metal mercury, many metals have been found
to exhibit superconductivity at liquid helium temperatures. About 40 elements from the
H He

Li Be B C N O F Ne
0.03
Na Mg Al Si P S Cl Ar
1.19 6.7 4.6-6.1

K Ca Se Ti V Cr Mn Fe Co Ni Cu Zn Ga Ge As Se Br Kr
0.39 5.3 0.9 1.09 5.4 0.5 6.9
6.9

Rb Sr Y Zr Nb Mo Tc Ru Rh Pb Ag Cd In Sn Sb Te I Xe
3.7:
0.5;2.7 0.55 9.2 0.92 7.8 0.5 325 µ 0.55 3.4 5.3 3.6 4.5

Ca Ba La Hf Ta W Re Os Ir Pt Au Hg Ti Pb Bi Po At Rn
1.8: 4.15 2.39 3.9
1.5 1.8;5.1
5.1 4.8;5.9 4.4 0.01 1.7 0.65 0.14 3.95 1.45 7.2 7.2;8.5
Fr Ra Ac

Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
1.7 0.1-0.7

Th Pa U Np Pu Am Cm Bk Cf Es Fm Md No Lw
1.37 1.3 0.2

Fig. 35.3
Superconductivity 1033

periodic table are found to exhibit superconductivity (Fig. 35.3). Apart from these, thousands
of alloys display superconductivity.
Good conductors of electric current such as silver, copper and gold are not supercon-
ductors. In fact, the superconductors are relatively poor conductors at room temperatures.
Non-transition metals such as Nb, Mo and Zn exhibit superconductivity. Semiconductors
like Si, Ge, Se and Te transform to a metallic phase when subjected to high pressure and then
become superconducting at low temperatures. The ferromagnetic materials like Fe, Co, Ni
do not show superconductivity. Table-1 lists some of the superconducting materials and their
transition temperatures TC.
Table 1: Transition Temperatures and Critical Field of Some Superconducting Materials
Material Transition Temperature (K) Critical Field at 0K (Wb/m2)
Type I. Superconductors
Tungsten (W) 0.015 0.0001
Titanium (Ti) 0.390 0.0100
Cadmium (Cd) 0.560 0.0030
Zinc (Zn) 0.850 0.0054
Molybdenum (Mo) 0.920 0.0095
Indium (In) 3.408 0.0281
Tin (Sn) 3.722 0.0305
Mercury (Hg) 4.153 0.0411
Vanadium (V) 5.380 0.1420
Lead (Pb) 7.193 0.0803
Niobium (Nb) 9.460 0.1980
Type II. Superconductors
Nb Ti 10.0 15.7
Nb NN 15.7 1.5
B3Al 18.7 32.4
Nb3Ge 23.2 38.4
V3G 14.8 2.1
V3Si 16.9 2.35
La – Ba – Cu – O 40 –
Y Ba2 Cu3O7 92 –
Bi2Sr2Ca2Cu3O10+ d 105 –
Th2Ca Ba2 Cu2O10+ d 125 –

The following points may be noted from Table-1.

● Among the pure metals, niobium has the highest transition temperature (TC = 9.46
K) and tungsten the lowest (TC = 0.015 K).
● Metal alloys have higher transition temperatures. Nb3Ge alloy has the highest
transition temperature of 23.2 K.
● The superconducting binary compounds and alloys may be formed from elements
where both of them are superconducting or one of them is superconducting or
neither of them is superconducting. For example, CuS is a superconductor where
neither copper nor sulphur is a superconductor.
1034 A Textbook of Engineering Physics

● Crystalline structure is not a necessary condition for super conductivity. Amorphous

and polycrystalline samples also display superconductivity.

35.4 PROPERTIES OF SUPERCONDUCTORS

Superconductors exhibit many unusual and interesting properties.
35.4.1 Zero Electrical Resistance
A super conductor is characterized by zero electrical resistivity. It is not fundamentally
possible to test experimentally whether the resistance is zero. A method devised by Onnes
consists of measuring the decrease of the current in a closed ring of superconducting wire.
The superconducting ring is kept in a magnetic field and it is cooled to below the critical
temperature so that it goes into the superconducting state. When the external magnetic field is
switched off, a current is induced in the ring. If the ring had a finite resistance, R, the current
circulating in the ring would decrease according to the equation
I(t) = I(0)e–Rt/L (35.1)
where L is the inductance of the ring. The decay current is monitored by a change in the
magnetic flux through a test coil held close to the superconducting ring. Any change in
the magnetic flux of the superconducting ring will induce an emf in the test coil. Careful
measurements established that the resistivity of superconductors could be taken as zero.
35.4.2 Persistent Current T > TC T < TC, Hext = 0
Once a current is started in a closed Hext
loop of superconducting material, it
will continue to keep flowing, of its
own accord, around the loop as long
as the loop is held below the critical
temperature (see Fig. 35.4 b). Such a
steady current, which flows without
diminishing in strength, is called a
persistent current. The persistent (a) (b)
current does not need external power Fig. 35.4
to maintain it because there do not
exist I2R losses. Calculations show that once the current flow is initiated, it persists for more
than 105 years. Persistent current is one of the most important properties of a superconductor.
Superconductor coils with persistent current flowing through them produce magnetic fields
and can therefore act as magnets. Such a superconducting magnet does not require power
supply to maintain its magnetic field.
35.4.3 Critical Temperature
When a superconducting material is cooled below a certain temperature, it goes into the
superconducting state from normal state. The temperature at which a material in normal state
goes into superconducting state is known as the critical temperature, TC (see Fig. 35.1).
Different materials have different critical temperatures. The transition is reversible. When
the temperature of the material is increased above the critical temperature, it passes into the
normal state. The transition is a thermodynamic phase transition. Just as the order in the
arrangement of atoms increases at the transition of a material from liquid to solid state, a
rearrangement of conduction electrons takes place leading to an increase in the order at the
transition from normal to superconducting state. The superconducting transition is sharp for
Superconductivity 1035

a chemically pure and structurally perfect specimen while the transition range is broad for
specimens which are structurally imperfect or which contain impurities.
35.4.4 Critical Magnetic Field
Superconducting state depends on the strength of the magnetic field in which the material
is placed. Superconductivity vanishes if a sufficiently strong magnetic field is applied. The
minimum magnetic field, which is necessary to regain the normal resistivity, is called the
critical magnetic field, HC. When the applied magnetic field exceeds the critical value HC,
the superconducting state is destroyed and the material goes into normal state.
The value of HC varies with temperature. Fig.
HC(O)
35.5 shows the dependence of HC on temperature Not
in a typical superconductor. At temperatures below HC1 superconductive
TC, in the absence of magnetic field, the material is
in superconducting state. When a magnetic field is HC H
C2
applied and as its strength reaches the critical value
HC, the superconductivity in the material disappears.
Superconductive
At any temperature T < TC the material remains
superconducting until a corresponding critical
magnetic field is applied. When the magnetic field TC1 TC2 TC
exceeds the critical value, the material goes into Temperature
normal state. The critical field required to destroy the
superconducting state decreases progressively with Fig. 35.5 : Schematic representation
increasing temperature. The dependence of critical of the critical magnetic field as a
function of temperature.
field on temperature is governed by the following
relation.
  T 2 
HC(T) = H C (0) 1 −    (35.2)
  TC  
 

where HC(0) is the critical field at 0 K.

Example 35.1. The transition temperature for Pb is 7.2 K. However, at 5 K it loses the
superconducting property if subjected to magnetic field of 3.3 × 104 A/m. Find the maximum
value of H which will allow the metal to retain its superconductivity at 0 K.
H C (T ) 3.3 × 104 A/m
Solution. HC(0) = = = 6.37 × 104A/m.
1 − (T 2 / TC2 ) 1 − (25 / 51.28)
Example 35.2. The critical field of niobium is 1 × 105A/m at 8 K and 2 × 105A/m at 0 K.
Calculate the transition temperature of the element.
T 8K
Solution. TC = 1
= 1
= 11.3 K
 H C (T )  2 5
 1 × 10 A/m  2
1 −  1 − 
 H C (0)  5
 2 × 10 A/m 
Example 35.3. The transition temperature for lead is 7.26 K. The maximum critical field for
the material is 8 × 105A/m. Lead has to be used as a superconductor subjected to a magnetic
field of 4 × 104A/m. What precaution will have to be taken?
1036 A Textbook of Engineering Physics

1 1
 H (T )  2  4 × 104 A/m  2
Solution. T = TC 1 − C  = 7.26 K 1 − 5  = 7.08 K.
 H C (0)   8 × 10 A/m 
Therefore, the temperature of the metal should be held below 7.08 K.
35.4.5 Critical Current Density
The critical magnetic field required to destroy superconductivity need not be necessarily
applied externally. An electric current flowing through the superconducting material itself
may produce magnetic field of requisite strength. Thus, if a superconducting ring carries a
current I, it gives rise to its own magnetic field. As the current increases to a critical value,
IC, the associated magnetic field increases to HC and the superconductivity disappears. The
maximum current density at which the superconductivity disappears is called the critical
current density, JC. For any value of J < JC, the current can sustain itself whereas for values
J > JC, the current cannot sustain itself. This effect was observed in 1916 by Silsbee and is
known as Silsbee effect.
A superconducting ring of radius R ceases to be a superconductor when the current is
IC = 2p R HC (35.3)
Thus, the existence of a critical current JC (T = 0K. H = 0)
sets a definite limit to the size of the current
that can flow through a superconducting coil
without disturbing its superconducting state. The
maximum current that a superconductor can carry
decreases as the temperature is raised and falls to
zero at the transition temperature of the material. H
Since the critical current falls with temperature, HC (T = 0K. J = 0)

the critical magnetic field will also decrease as

the transition temperature is approached. The T TC (H = 0.J = 0)
variation of critical current density JC and critical Fig. 35.6
magnetic field HC with temperature is shown in
Fig. 35.6.
Example 35.4. The critical magnetic field at 5 K is 2 × 103 A/m in a superconductor ring of
radius 0.02 m. Find the value of critical current.
Solution. IC = 2p R HC = 2 × 3.143 × 0.02m × 2 × 103 A/m = 251.4 A
Example 35.5. Calculate the critical current for a wire of lead having a diameter of 1 mm at
4.2 K. The critical temperature for lead is 7.18 K and HC(0) = 6.5 × 104A/m.
  T 2    4.2 K 2 
4
Solution. HC(T) = H C (0) 1 −    = 6.5 × 10 A / m 1 −  4
  = 4.28 × 10 A/m
  TC     7 . 18 K  
 
The critical current IC = 2prHC = pdHC = 1 × 3.14 × 10–3m × 4.28 × 104 A/m = 134.5 A.
35.4.6 Perfect Diamagnetism – Meissner Effect
In 1933 W.Hans Meissner and Robert Ochsenfeld found that when superconductors are
cooled below their critical temperature in the presence of a magnetic field, the magnetic
flux is expelled from the interior of the specimen and the superconductor becomes a perfect
diamagnetic. This phenomenon is known as Meissner effect (see Fig. 35.7). Meissner and
Superconductivity 1037

Ochsenfeld found that as the temperature of the specimen is lowered to TC, the magnetic
flux is suddenly and completely expelled from it. The flux expulsion continues for T < TC.
The effect is reversible. When the temperature is raised from below TC, the flux suddenly
penetrates the specimen at T = TC and the material returns to the normal state.
The magnetic induction inside the specimen is given by
B = m0(H + M) = m0(1 + c)H
where H is the magnetic field applied externally and M is the magnetization produced within
the specimen.
At T < TC, B = 0 and therefore m0(H + M) = 0
It follows that M = –H
The susceptibility of the material is
M
c= =–1 (35.4)
H
The specimen is therefore
diamagnetic and the state in which
magnetization cancels the external
magnetic field completely is referred
to as perfect diamagnetism.
The Meissner effect contradicts
the fundamental principles of electro-
magnetism. The condition of perfect
diamagnetism cannot be explained
from the simple definition that
superconductivity is a state of zero Normal Superconducting
resistivity. Meissner effect shows that T > TC T < TC
or H > HC or H < HC
dB
in the superconductor not only =0 Fig. 35.7
dt
but also B = 0. Thus, two mutually independent properties, namely zero resistivity and perfect
diamagnetism are the essential properties that characterize the superconducting state.
Applications of Meissner effect
● The Meissner effect is the standard test used to conclusively prove whether a
particular material is a superconductor or not.
● A material in superconducting state is a perfect dia-magnet and hence strongly repels
external magnets. A smaller magnet repelled by a bigger superconductor hovers in
air. This is known as levitation effect. In a similar way, a small chip of supercon-
ducting material hangs on to a bigger magnet and this effect is known as suspension
effect. The levitation effect is utilized in the operation of Maglev trains.
35.4.7 London Penetration Depth
When a magnetic field is applied to a superconductor, the applied field does not suddenly
drop to zero at the surface. Instead the field decays exponentially according to the formula
H(x) = H(0)e–x/l (35.5)
where H(0) is the field applied at the surface at x = 0 and x is the distance from the surface. The
length l is called the London penetration depth. It may be defined as the effective depth to
1038 A Textbook of Engineering Physics

which a magnetic field penetrates a superconductor.

The penetration depth l ranges from 300 to about
5000 Å depending on the material. It is independent
of frequency of the magnetic field but it strongly H
depends on temperature.
The temperature dependence of l is given by 
the relation
λ(0)
l(T) = (35.6)
  T 4 
1 −   
  TC  
 
where l(T) and l(0) are the penetration depths at T
and 0 K. Fig. 35.8
It follows from equ. (35.6) that l increases with
the increase in temperature and at the critical temperature, it becomes infinite. At T = TC
the material goes into the normal state and hence the magnetic field penetrates the whole
specimen.
Example 35.6. Calculate the penetration depth of lead at 5.2 K if the London penetration
depth at 0 K is 37 nm. The critical temperature of lead is 7.193 K.
−1/ 2 −1/ 2
  T 4    5.2 K 4 
Solution. l(T) = λ(0) 1 −    = 37 nm 1 −    = 43.4 nm.
  TC     7.193K  
 
35.4.8 Flux Quantization
In 1957 A.A.Abrikosov predicted the existence of magnetic flux quanta. According to him,
a closed superconducting loop can enclose magnetic flux only in integral multiples of a
fundamental quantum of flux. The magnetic flux enclosed by a superconducting ring is given
by
h
j= n = nϕ0   n = 1, 2, 3,....
2e
h
where j0 = (35.7)
2e
is the flux quantum and is called a fluxon. Its value is
j0 = 2.07 × 10–15 weber
The quantization of magnetic flux has been confirmed experimentally in 1961 by Deaver
and Fairbank.
35.4.9 Entropy
Entropy S

Entropy is a measure of the disorder tate

m al S
r
of a system. Generally, entropy No Tc
decreases with decreasing temperature. Superconducting state
In superconducting materials, the
entropy decreases linearly up to critical
temperature and more remarkably below T
Fig. 35.9
the critical temperature (see Fig. 35.9).
It indicates that a superconducting state is more ordered than the normal state.
Superconductivity 1039

35.4.10 Heat Capacity

The transition of a metal from its normal state to superconducting state does not involve a
change of crystallographic structure. It is only a thermodynamic phase transition where the
specific heat changes discontinuously at the transition temperature TC. The specific heat of a
normally conducting metal is composed of a lattice part and an electronic part. The electronic
contribution varies smoothly at low temperatures. In superconductors it is found that the
electronic part decreases exponentially, as shown in Fig. 35.10, and it is given by
Cu = e–b TC/T (35.8)
where b is a constant.
4
ctor
Specific heat capacity

3 0.585
ndu

Sn
erco

r 0.580
to
uc
Sup

2 d –M
–

Con 0.575
log TC
0.570
1
0.565
0 0.560
0 0.5 1.0 Tc 1.5 2.0 2.05 2.06 2.07 2.08 2.09 2.10
Temperature (K) log M

Fig. 35.10                 Fig. 35.11

35.4.11 Isotope Effect

In 1950, C.A.Reynolds and E.Maxwell found that the critical temperature decreases with
increasing isotopic mass M. The variation is given by the relation
TC ∝ M–a
or MaTC = constant (35.9)
1
where a is a constant and is approximately equal to . The phenomenon of decrease of
2
critical temperature with increasing atomic mass is called isotope effect. Fig. 35.11 shows the
experimental results for tin. Since the mean square of amplitude of atomic vibrations is
proportional to M at low temperatures, the equ. (35.9) suggests that the lattice vibrations
are involved in causing superconductivity.
∂ (ln TC )
The value of a can be expressed as a= − (35.10)
∂ (ln M )
Change in the isotopic mass does not change the electronic structure. Dependence
of the transition temperature on the isotopic mass suggests involvement of electron-lattice
interaction.
Example 35.7. In a superconducting material isotopic mass is 199.5 amu and critical
temperature is 5K. Calculate isotopic mass at 5.1 K.
1
Solution. TC ∝ . Therefore, T1 M1 = T2 M 2
M
1040 A Textbook of Engineering Physics

2 2
T   5K 
or M2 = M1  1  = 199.5amu   = 191.68 amu.
 T2   5.1K 

35.5 OTHER EXTERNAL FACTORS THAT AFFECT

SUPERCONDUCTIVITY
1. Stress: The transition temperature can be changed by application of stress. Usually,
stress which increases the dimensions increases the transition temperature. Highly
sensitive methods are required to detect the change.
2. Impurities: Addition of chemical impurities modifies nearly all the superconduc-
tivity properties and in particular the magnetic properties.
3. Size: If the size of the specimen is reduced below about 100 mm, its superconducting
properties are modified in many respects. The magnetic permeability of the small
specimen is no longer zero and varies with temperature while the critical magnetic
field becomes greater than that of bulk material.
4. Frequency: The zero resistance of superconductors is modified at very high
frequencies of alternating current. Up to 10 MHz the resistance is still zero but for
frequencies of the order of GHz, it is found that considerable resistance is observed
even below TC.
35.6 TYPE-I AND TYPE-II SUPERCONDUCTORS
Superconductors are divided into two categories depending on the way in which the transition
from superconducting to normal state proceeds when the externally applied magnetic field
exceeds HC.
Type-I Superconductors:
In type I supercon-
Type I
ductors, the transition Normal
Magnetization–µ0M

Sup.cond. superconductor
from superconducting state state
ρ
state to normal state in
the presence of magnetic
field occurs sharply at the
critical value of HC, as
shown in Fig. 35.12 (a). HC H O HC
Type I superconductors External field H
are perfectly diamagnetic (a) (b)
below HC and completely Fig. 35.12
expel the magnetic field
from the interior of the superconducting phase. Up to the critical field strength, the magneti-
zation of the material grows in proportion to the external field and then abruptly drops to
zero at the transition to the normal state, as shown in Fig. 35.12 (b). The magnetic field can
penetrate only the surface layer and current can flow only in this layer. Consequently, type I
superconductors are poor carriers of electrical current.
Aluminium, lead and indium are examples of Type I superconductors. The critical field
is relatively low for type I superconductors. They would generate magnetic fields of about 100
to 2000 G only. Hence, they are not of much use in production of high magnetic fields. Type I
superconductors are also called soft superconductors. We summarize here the characteristics
of Type I superconductors.
Superconductivity 1041

Characteristics of Type-I superconductors

1. They are perfectly diamagnetic and exhibit complete Meissner effect.
2. They have only one critical field. At the critical field the magnetization drops to zero.
3. The maximum critical field for type I superconductor is of the order of 0.1 Wb/m2.
4. The transition at HC is reversible. Below HC the material behaves as a supercon-
ductor, and above HC it behaves as a normal conductor.
Disadvantages
Type I superconductors cannot carry large currents and hence are not of much use in
producing high magnetic fields.
Type-II Superconductors
Type II Type II

Magnetization-0M
superconductivity superconductor
was discovered p
by Schubnikov
in 1930s and
was explained
by Abrikosov HC1 HC2
H
O HC1 HC2
in 1957. Type II External field H
superconductors Complete
Mixed State
are characterized Diamagnetism
Normal
State
by two critical
fields HC1 and Fig.35.13
HC2. The transition from superconducting state to normal state occurs gradually as the
magnetic field is increased from HC1 to HC2, as shown in Fig. 35.13. The magnetization
of the material grows in proportion to the external field up to the lower critical field HC1.
The external magnetic flux is expelled from the interior of the material till then. At HC1 the
magnetic field lines begin penetrating the material.
As the magnetic field increases further, the magnetic flux through the material increases.
At the upper critical field HC2, the magnetization vanishes completely and the external field
has completely penetrated and destroyed the superconductivity. In the region between HC1
and HC2, the material is in a magnetically mixed state but electrically it is a superconductor.
HC2 can be as high as 20 to 50 Wb/m2 and the retention of superconductivity in such high
magnetic fields make type II materials very useful in applications of creating very high
magnetic fields.
Transition metals and alloys
consisting of niobium, silicon and Hext
vanadium exhibit type II superconduc-
tivity. Ceramic superconductors also
belong to this category.
A distinguishing feature of type II
superconductors is that super currents
arising in an external magnetic field
can flow not only over the surface of a
conductor but also in the bulk. Above
the lower critical field HC1, it becomes
Fig. 35.14
energetically more favourable to admit a
1042 A Textbook of Engineering Physics

single flux quantum rather than have the superconductor exclude HC1. The superconductor
passes into a mixed state where the bulk of the material is superconducting but is threaded
by very thin filaments of normal material (see Fig. 35.14). The thin filaments of normal
material serve as the paths along which the magnetic field penetrates. In the center of the
filament superconductivity is absent. These normal regions are surrounded by vortices of
super currents. A flux line together with its current vortex is called a fluxoid. At HC1, fluxoids
appear in the material and increase in number as the magnetic field is increased. An increase
in the magnetic field will not cause an increase of the flux in each vortex line but will cause
an increase in the number of fluxoids threading the superconductor. At HC2 the fluxoids fill
the entire specimen and superconductivity disappears.
Type II superconductors can carry larger currents when the magnetic field lies between
HC1 and HC2. Type II superconductors are called hard superconductors. We summarize here
the characteristics of Type II superconductors.
Characteristics of Type-II superconductors
1. They have two critical magnetic fields, HC1 and HC2.
2. The material is perfect diamagnetic below the lower critical field, HC1. Meissner
effect is complete in this region. Above the upper critical field, HC2, magnetic flux
enters the specimen.
3. Above HC1 they do not show complete Meissner effect and therefore do not behave
as perfect diamagnetic materials.
4. They exist in an intermediate state in between the critical fields, HC1 and HC2. The
intermediate state is a mixture of the normal and superconducting states, magneti-
cally but electrically the material is a superconductor.
5. At HC2 the magnetization vanishes and the specimen returns to normal conducting
state.
6. The upper critical field is very high and is of the order of 30 Wb/m2.
Applications
They are used in applications of generating very high magnetic fields.
COMPARISON BETWEEN TYPE I AND TYPE II SUPERCONDUCTORS
Sl.No. Type-I Superconductors Type-II Superconductors
1. They exhibit complete Meissner effect They do not exhibit complete Meissner effect
2. They show perfect diamagnetic behaviour They do not show perfect diamagnetic behaviour
3. They have only one critical magnetic field, They have two critical magnetic fields, lower
HC critical magnetic field, HC1 and upper critical
magnetic field, HC2
4. There is no mixed state or intermediate state Mixed state or intermediate state is present in
in case of these materials these materials
5. The material loses magnetization abruptly The material loses magnetization gradually
6. Highest value for HC is about 0.1 Wb/m2 Upper critical field is of the order of 30 Wb/m2
7. They are known as soft superconductors They are known as hard superconductors
8. Lead, tin, mercury are examples Nb-Sn, Nb-Ti, Nb-Zr, Va-Ga are examples

35.7 BCS THEORY

In 1957, the American physicists J. Bardeen, L.N. Cooper and J.R. Schrieffer developed the
quantum theory of superconductivity, which came to be known as BCS theory. Starting from
Superconductivity 1045

  2eV 
\ I = I C sin ϕ0 + 2π t   (35.15)
  h 
The current given by eq. (35.15) represents an alternating current of frequency
2eV
n= (35.16)
h
Equ. (35.16) shows that a photon of frequency n is emitted or absorbed when a Cooper
pair crosses the junction. Thus, when a dc voltage is applied across a Josephson junction, an
ac current is produced by the junction. This is known as the ac Josephson effect. At V = 1 m
V, ac current of frequency 483.6 MHz is produced.
Example 35.8. A Josephson junction with a voltage difference of 650 mV radiates electromag-
netic radiation. Calculate its frequency.
2eV 2(1.602 × 10−19 C )(650 × 10−6 V )
Solution. n = Hz = = 3 × 1011 Hz.
h 6.626 × 10−34 J
Oxygen
35.9 HIGH SUPERCONDUCTORS Copper

Superconductors are divided into low TC and

high TC superconductors based on their transition
Ba
temperature. Broadly, materials having TC below 24
K are regarded as low TC superconductors and those Oxygen
having TC above 27 K are regarded as high TC supercon- Vacancy
ductors. However, in practice, materials for which liquid Y
nitrogen cooling can cause transition to superconducting
state may be regarded as high TC superconductors while
those that require liquid helium coolant are considered Ba
as low TC superconductors. The maximum transition
temperature that could be achieved before 1980’s was
23.2 K in Nb3Ge, which is a metallic alloy. Therefore,
it was hoped that metallic alloy systems could be made
Fig. 35.18
to have higher transition temperatures but such systems
were not discovered. Therefore, the focus has shifted to ceramic oxides, which are insulating
materials at normal temperatures. In 1986 Bednortz and Muller discovered superconductivity
in ceramic materials. They found that the mixed metallic oxide of lanthanum-barium-copper
(La1Ba2Cu3O7) exhibited superconductivity at about 30 K. The superconductivity of the
oxide was linked with the deficiency of oxygen ions in the oxide compound. When this
deficiency of oxygen was carefully controlled, by keeping the samples in oxygen atmosphere,
it was found that the material would exhibit superconductivity in the temperature range
30 to 40 K. In 1987 Chu and coworkers replaced lanthanum with yittrium and prepared
YBa2Cu3O7 with transition temperature of about 95 K. This was a major breakthrough, as
this can be maintained in the superconducting state with far less expensive liquid nitrogen
coolant (nitrogen exists in liquid state below 77 K) and marks the beginning of preparation of
high TC superconductors. This oxide also contains a deficiency of oxygen with the chemical
formula YBa2Cu3O7– d where d indicates the deficiency of oxygen and is in the range 0 to 0.1.
YBa2Cu3O7 is the most extensively studied high TC superconductor and exhibits a defective
perovskite structure with three perovskite cubic unit cells stacked on top of each other
1046 A Textbook of Engineering Physics

(Fig. 35.18). For an ideal stack of three perovskite cubic unit cells, the YBa2Cu3Ox compound
should have the composition YBa2Cu3O9, in which x would be equal to 9. However, analysis
shows that x ranges from 6.65 to 6.90 for this material to exhibit superconductivity. At x =
6.90, its TC is highest (≈ 90 K) and at x = 6.65, superconductivity disappears. Thus, oxygen
vacancies are found to play a key role in the superconducting behaviour of ceramic oxides.
If the cell contains one atom of rare earth metal, two barium atoms, three copper atoms and
seven oxygen atoms, then such compounds are called 1-2-3 superconductors. These high TC
copper oxide superconductors belong to type II and their upper critical field is very high (of
the order of 150 to 200 tesla). Apart from YBa2Cu3O7–d compounds, the compounds based on
bismuth like Bi-Sr-Cu-O or Ba-Ca-Cu-O systems are also superconducting. For example, the
compound Bi2CaSr2Cu2O8+x exhibits superconductivity and has the transition temperature 85
K and Bi2Ca2Sr2Cu3O10+x has the transition temperature 110 K. In 1993, a still higher of 133
K was achieved in mercury based copper oxide HgBa2Ca2Cu3O1+ x.
The above discussed mixed oxide materials may be arranged into four major families.
(i) LBCO (mixed oxide of lanthanum, barium and copper)
(ii) YBCO (mixed oxides of yttrium, barium and copper)
(iii) BSCCO (mixed oxides of bismuth, strontium, calcium and copper)
(iv) TBCCO (mixed oxides of thallium, barium, calcium and copper)
Continuous search is going on to discover materials that may exhibit superconducting state
around room temperature. Interesting practical applications of superconductors are visualized
in many fields, but they could not be used on large scale because of the requirement of liquid
helium, which is highly expensive. Therefore, they remained as a laboratory curiosity for a
long time. The discovery of high temperature materials opened up possibilities of putting
superconductors to large-scale use. They require only liquid nitrogen, which is easily available
and also cheaper.
Properties of high TC superconductors
Some of the properties of high TC superconductors are as follows:
1. The high TC superconductors are brittle in nature.
2. The properties of the normal state of these materials are highly anisotropic.
3. The Hall coefficient is positive indicating that the charge carriers are holes.
4. Their behaviour cannot be explained by BCS theory.
5. The isotope effect is almost absent in these materials.
6. The magnetic properties of these materials are highly anisotropic.
7. The effect of pressure is different on different materials. For example the application
of pressure increases the critical temperature of LBCO compounds but decreases the
critical temperature of YBCO compounds.

35.10 APPLICATIONS
Utilization of superconductivity in practical applications is severely limited by the very low
temperatures required to maintain the superconducting material in the superconducting state.
Till 1986 the highest critical temperature known was about 27 K. Only using liquid helium
as the coolant, which is very costly, one can attain such low temperatures. In the last two
decades certain high TC ceramic materials have been discovered. These materials are brittle,
difficult to be drawn into wires and cannot carry large currents. Vigorous research is going on
around the world to overcome the drawbacks of high materials and to gainfully utilize them
in different applications. We discuss here some of the interesting applications.
Superconductivity 1047

1. The most obvious application of superconductors is in power transmission. If the

national grid were made of superconductors rather than aluminium, then the savings
would be enormous - there would be no need to transform the electricity to a higher
voltage (this lowers the current, which reduces energy loss to heat) and then back
down again.
2. Superconducting coils in transformers and electrical machines generate much
stronger magnetic fields than magnetic circuits employing ferromagnetic materials
produce. The normal eddy current losses and hysteresis losses will not be present
in superconducting devices and hence the size of motors and generators will be
drastically reduced. Thus, superconductors are likely to revolutionize the whole
range of rotating electrical machines, making them smaller, lighter and highly
efficient. For example, a superconducting generator about half the size of a copper
wire generator is about 99% efficient; typical generators are around 50% efficient.
3. High magnetic fields are required in many areas of research and diagnostic
equipments in medicine. The electromagnets are cumbersome being very big,
demand large electrical power and require continuous cooling. Superconducting
solenoids produce very strong magnetic fields. They are small in size and are less
cumbersome. They do not need either large power supplies or the means of removing
heat. The only power required is to bring the solenoid into the superconducting
state and maintain it in that state. The low power requirement and simple cooling
technique leads to a large saving in cost. The development of superconductors has
improved the field of MRI, as the superconducting magnet can be smaller and more
efficient than an equivalent conventional magnet.
4. Type II superconductors can be used as very fast electronic switches due to the way
in which a magnetic field can penetrate into the superconductor - this has allowed
researchers to build a 4-bit computer microchip operating at about 500 times the
speed of current processors, where heat output is currently a major problem with
typical speeds approaching the 1GHz mark.
5. Cryotrons: The application of a magnetic field greater than its critical magnetic field
changes the superconducting state of a superconducting material to normal state and
removal of the field brings the material back from normal state to the supercon-
ducting state. This fact is used in developing cryotron switches.
Fig. 35.19 (a) shows 2
Gate Wire, A
a schematic diagram of a 1

simple cryotron. It consists

of a superconducting material
(control) coil, B, wound around control coil, B
another superconducting (gate)
wire, A. For example, the (a) (b)
control coil is of niobium (TC Fig. 35.19
= 9.3K) or lead (TC = 7.2K)
whereas the gate wire is of tantalum. The value of [HC]A of material A is less than
the value of [HC]B of material B. Initially, let the temperature of the device be below
the transition temperature of the two materials A and B. In such a condition the wire
B does not offer resistance to the flow of current through it as it is in the supercon-
ducting state. Now, at the operating temperature, the magnetic field produced by the
1048 A Textbook of Engineering Physics

coil B may exceed the critical field of the core wire A. It makes the material to go
into normal state because the critical field of A is less than that of B. However, B
will not go to normal state at the critical field of A because [HC]B > [HC]A. Hence,
the current in the central wire A can be controlled by the current in the coil B. Thus,
whenever the current passing through the coil B exceeds the critical current value,
the wire A will become a normal conductor exhibiting a finite resistance. This closes
the gate for the flow of current through the core wire, A. Removal of the current
reopens the gate. Thus the system acts as a fast acting relay or switching element
and is highly suitable as fast acting memory element in computers.
The speed of the cryotron switch is dependent on its time constant t = L/R, where
L is the inductance of the control coil B and R is the resistance of the gate coil A
in the normal state. Wire wound cryotrons have t of the order of 10–3s. To reduce
the value of t, R is to be increased and L is to be decreased as far as possible. This
objective is achieved by depositing two crosses strips on a substrate and separating
them by a thin dielectric layer (Fig. 35.19 b). Strip 1 acts as the gate and is usually
made of tin (TC = 3.7K) and strip 2 acts as the control element usually made of lead
(TC = 7.2 K). Varying the current the strip 2 enables switching the strip 1 from the
superconducting state to the normal state and vice versa. Thus opening and closing
of the circuit is achieved. Through an appropriate design, the switching can be made
faster as the value of t can be decreased up to 10–7s.
6. MagLev Trains: The most spectacular application would be the so-called ‘MagLev’
trains. The coaches of the train do not slide over steel rails but float on a four inch air
cushion above the track using superconducting magnets; this eliminates friction and
energy loss as heat, allowing the train to reach high speeds of the order of 500 km/
hr. Such magnetic levitation trains would make train travel much faster, smoother,
and more efficient due to the lack of friction between the tracks and train.
Operation
A typical plan of Maglev train is shown Carriage
in Fig. 35.20. The train has supercon-
Wheel
ducting magnets built into the base of its Wheel
carriages. An aluminium guideway is laid
on the ground and carries electric current.
The repulsion between the two powerful
magnetic fields, namely the field produced
by the superconductor magnet and the Superconducting
Aluminium
field produced by the electric current in magnet

the aluminium guideway causes magnetic Fig. 35.20

levitation of the train. A levitation of about
10 to 15 cm is achieved. The train is fitted with retractable wheels, which act in a way similar
to the wheels of an airplane. Once the train is levitated in air, the wheels are retracted into the
body and the train glides forward on the air cushion. When the train is to be halted, the wheels
are drawn out and the train descends slowly onto the guideway and runs forward till it stops.
7. SQUIDS
A superconducting quantum interference device (SQUID) is a device used to measure
extremely weak magnetic flux. Thus, it is basically a sensitive magnetometer. The heart of
a SQUID is a superconducting ring, which contains one or more Josephson junctions. There
Superconductivity 1049

are two main types of SQUID: DC SQUID and RF (or AC) SQUID. The DC SQUID was
invented in 1964 by Robert Jaklevic, John Lambe, Arnold Silver, and James Mercereau. The
RF SQUID was invented in 1965 by J.E.Zimmernan and Arnold Silver at Ford. An RF SQUID
is made of one Josephson junction and a dc SQUID consists of two Josephson junctions in
parallel and relies on the interference of the currents from each junction. We study here
the dc SQUID. DC SQUIDs are more difficult and expensive to produce, but DC SQUID
magnetometers are much more sensitive.
Fabrication
SQUIDs are usually fabricated from lead or pure niobium. The lead is taken in the form
of an alloy with 10% gold or indium. A thin niobium layer deposited on to it acts as the base
electrode of the SQUID and the tunnel barrier is oxidized onto this niobium surface. The
top electrode is a layer of lead alloy deposited on top of the other two, forming a sandwich
arrangement. The entire device is then cooled to within a few degrees of absolute zero with
liquid helium.
The schematic of a two-junction SQUID [direct current (DC) SQUID] is shown in Fig.
35.21 (a). It consists of two Joseph junctions arranged in parallel so that electrons tunneling
through the junctions demonstrate quantum interference.
Working
Tunnel barrier
Voltage Superconductor P I 1

C B D

Current Current b
Magnetic I2
Josephson
Field
Junction Q Superconductor
(a) (b)

Fig. 35.21

A dc supercurrent is applied to the SQUID (Fig.35.21 b). This current, known as bias
current, enters the device through the arm C. It is divided along two paths a and b and again
merge into one and leaves through the arm D. P and Q are the Josephson junctions and the
insulating layers at P and Q are of different thickness. I1 and I2 are the currents tunneling
through the junctions P and Q respectively. In a superconductor, a single wave function
describes all the Cooper pairs. The wave function experiences a phase shift at the junctions
P and Q. Let the phase difference between points C and D taken on a path through junction
P be da and the phase difference between points C and D taken on a path through junction
Q be db. In the absence of magnetic field these two phases are equal. That is, db – da = 0.
When a magnetic field B is applied perpendicular to the loop, the flux passes through the
loop, and changes the quantum mechanical phase difference across each of the two junctions.
The wave functions at the two Josephson junctions interfere with each other. In other words,
the supercurrents flowing along the paths a (PD) and b (QD) interfere. Hence, the device
is named SQUID. The interference closely resembles the optical interference observed with
Young’s double slit. In the case of light, the phase difference between light waves is due to
the difference in optical path lengths. In case of supercurrent interference, the waves are the
de Broglie waves of Cooper pairs, and the phase difference is caused by the applied magnetic
1050 A Textbook of Engineering Physics

field. According to Josephson’s theory, the phase difference between the reunited currents is
directly proportional to the magnetic flux, F, through the ring. It can be shown that the total
current through two parallel Josephson junctions is given by
eΦ
IT = 2( I 0 sin δ0 ) cos (35.17)
c
The above relation indicates that a progressive increase or decrease of the magnetic flux,
causes the current to oscillate between a maximum and a minimum value. Maxima in the
current occur whenever the magnetic flux increases by one flux quantum. Thus, the period of
these oscillations is one flux quantum F0.
h
F0 = = 2.06 × 10–15 webers
2e
Fig. 35.22 shows the variation of the current through a pair of Josephson junctions as a
function of the magnetic flux applied.

Josephson current (arb. units)

In practice, instead of the current, we measure
the voltage across the SQUID, which also oscillates
with the changing magnetic field. Thus, the SQUID
is a flux-to-voltage transducer, converting a tiny
change in magnetic flux into voltage. The flux F is
related to the magnetic field B through the relation
F = BA
where A is the area of the ring. With the help of –5 –4 –3 –2 –1 0 1 2 3 4 5
this relation, flux measurements are converted into
–5
Magnetic field B (10 Tesla)
magnetic field measurements. Fig. 35.22
SQUID is a very sensitive magnetometer, which can measure very weak magnetic fields of
the order of 10–13 Wb/m2. The sensitivity of a SQUID to magnetic fields can be enhanced by
using a flux transformer (Fig. 35.23). A flux transformer consists of a loop of superconducting
material, which is coupled MAGNETOMETER
to the SQUID. An external MAGNETIC FIELD
magnetic field produces a SQUID
persistent supercurrent in the INPUT
loop and this current induces COIL
a flux in the SQUID. As the
loop encloses a much larger
area than can a SQUID, the
sensitivity of the device gets PICKUP
LOOP CURRENT
enhanced.
Applications of Squids Fig. 35.23

SQUIDs are used to measure very small magnetic fields. Since the current is sensitive to
very small changes in the magnetic field, the SQUID acts as a very sensitive magnetometer.
As ordinary magnetometers, SQUIDs are capable of measuring magnetic fluctuations of the
order of 10–13 T. Because of their extreme sensitivity, SQUIDs find applications in many
fields, engineering, medicine and many other fields. For example, geologists use them for
measuring rock magnetism and continental drift. Physical processes, such as muscular or
neural activity, in humans (and other animals) create magnetic fields as small as a thousand
billionth of a tesla (as a comparison, a fridge magnet generates about a tenth of a tesla).
Human heart generates magnetic fields of about 10–14 wb/m2 and the human brain generates
Superconductivity 1051

magnetic fields of about 10–14 Wb/m2. SQUIDs can detect these feeble fields and an array
of SQUIDs is used in magnetoencephalography (MEG), the process of brain imaging. The
SQUIDs are also used in nondestructive testing. In testing for corrosion of aluminum sheets
riveted together in aircraft, the SQUID measures the influence of the aircraft skin on an
applied oscillating magnetic field; a change in electrical conductivity reveals the defects.
Example 35.9. For a certain metal the critical magnetic field is 5 × 103 A/m at 6K and
2 × 104 A/m at 0K. Determine its transition temperature.
T 6K
Solution. TC = 1
= 1
= 6.93 K.
 H C (T )  2 3
 5 × 10 A/m  2
1 −  1 − 
 H C (0)  4
 2 × 10 A/m 

QUESTIONS
1.The dc resistance of a superconductor is practically zero. What about its ac resistance?
2.How can you change a superconductor from type I to type II?
3.What is the importance of isotope effect in superconductivity?
4.What are Cooper pairs?
5.What is coherence length?
6.How is Josephson tunneling different from single particle tunneling?
7.A superconducting wire and a copper wire are connected in parallel. Does the copper wire carry
current when a potential difference is applied?
8. Give an account of the phenomenon of superconductivity.
9. What is the significance of critical temperature, critical magnetic field and critical current density
for superconductors?
10. Explain the Silsbee effect.
11. What is Meissner effect? (M.G.Univ., 2005)
12. Explain Meissner effect and isotope effect. (Calicut Univ., 2007)
13. Compare the dependence of resistance on temperature of a superconductor with that of a normal
conductor.
14. Define superconductivity. Explain the effect of isotopes on superconductors. (M.G.Univ., 2005)
15. Discuss the relation between isotopic mass and transition temperature.
16. What is flux quantization?
17. What do you mean by “perfect diamagnetism” of a superconductor?
18. Describe how Cooper pairs are formed and explain the salient features of superconductivity.
(V.T.U., 2007)
19. Give a short account of high temperature superconductivity.
20. Explain the term high temperature superconductivity. Give the various applications of supercon-
ductors. (M.G.Univ., 2005)
21. Explain high temperature superconductivity. Explain its advantage. (M.G.Univ., 2005)
22. Distinguish between dc and ac Josephson effects. (Calicut Univ., 2007)
23. What is superconductivity? Explain Meissner effect. Describe type-I and type-II supercon-
ductors. (C.S.V.T.U.,2005, 2009)
24. What are type I and type II superconductors? Explain B.C.S. theory with key note of Cooper
pairs. (M.G.Univ., 2006)
25. Explain the BCS theory with key note of Cooper pairs. Distinguish between type I and II
superconductors. (Calicut Univ., 2005)
26. Write short notes on BCS theory of superconductivity.
(Calicut Univ., 2005, 2007), (C.S.V.T.U.,2006, 2008)
27. What is superconductivity? Describe type I and type II superconductors. (V.T.U., 2008)
1058 A Textbook of Engineering Physics

Metallic glasses are also increasingly used in inductive components such as reactors,
inverter transformers, chokes for magnetic switches, tape recorder heads, magnetic shields
etc. Applications in the areas of magnetic sensors and transducers are under development.
The chief advantage of metallic glasses is that they are fabricated using inexpensive
metals and they can be fabricated with relative ease in the form of thin tapes.
(i) Metallic glasses possess good tensile strength and are superior to common steels.
They are used as reinforcing elements in concrete, plastic and rubber.
(ii) Metallic glasses possess high ductility, good corrosion resistance and hence used in
makings springs for different applications.
(iii) One of the most important applications of metallic glasses is as inductors in
transformers, recording devices, magnetic shielding, motors and others.
(iv) A transformer having a core made from metallic glass is found to have eight times
less core loss and requires about twenty times less current for excitation, compared
to an equal sized commercial transformer having a crystalline silicon steel core.
(v) Magnetic recording heads made of metallic glasses are superior in overall
performance to those made of ferrites and permalloys, because of their high flux
density and high wear resistance.
(vi) Metallic glasses offer opportunities for increasing the efficiency of magnets, motors,
and transformers used in energy-conversion devices. They would reduce costly
losses from power-distribution systems and corrosion damage.

36.3 LIQUID CRYSTALS

Liquid crystals are substances that exhibit a phase of matter that has properties between
those of a conventional liquid and those of a crystal. Liquid crystals have long rod-shaped
organic molecules of about 25 Angstroms in length. In the liquid crystal state, the molecules
can move about and, therefore, lack positional order. However, the molecules strongly
interact and are arranged and
oriented in a crystal-like way.
Thus, they possess long-range Increase Increase
orientational order. The temperature temperature
common axis along which the
molecules tend to orient is Crystal Liquid Crystal Liquid
called the director. Fig. 36.6
The tendency of the
liquid crystal molecules to point along the director leads to a condition known as anisotropy.
This term means that the properties of a material depend on the direction in which they
are measured. The anisotropic nature of liquid crystals is responsible for the unique optical
properties exploited in a variety of applications.
Liquid crystal phases can be divided into two classes: thermotropic and lyotropic liquid
crystals. Thermotropic liquid crystals pass into the liquid crystal phase as temperature is
increased. The liquid crystal state exists within some temperature range, Tm < T < TC, where
Tm is temperature of melting from solid state into liquid crystal state, and TC is clearing
temperature, when the liquid crystal transforms into an isotropic liquid. In the solid state, the
centers of gravity of molecules posses long-range positional order, and, also, the molecules
orientation points in the same direction providing the long-range orientational order. When
Modern Engineering Materials 1059

solid melts into a liquid crystal at Tm, the positional order is lost although some orientational
order of the molecular long axes remains. At still higher temperature TC, liquid crystal melts
into an isotropic liquid with no positional and orientational order.
An example of a
compound displaying
thermotropic liquid C4H9 N=CH O CH3

crystal behavior is
methoxybenzilidene Fig. 36.7. Methoxybenzilidene Butylanaline (“MBBA”)
butylanaline (MBBA),
which is a nematic liquid crystal between 21°C and 48°C.
Lyotropic liquid crystals exhibit phase transitions as a function of concentration of the
liquid crystal in a solvent (typically water) as well as temperature.
36.3.1 Liquid Crystal Phases
Liquid crystal phases are broadly divided into three types, depending upon the positional
order and orientational order of the molecules in the phase. They are
(i) Nematic phase
(ii) Smectic phase and
(iii) Cholesteric phase.
(i) Nematic phase: The nematic liquid crystal phase is
characterized by molecules that have no positional order but
tend to point in the same direction (along the director). In the
following diagram, notice that the molecules point vertically n
but are arranged with no particular order.
In the nematic phase, the molecules flow and their center of
mass positions are randomly distributed as in a liquid, but they
all point in the same direction within a small domain. Nematics
have fluidity similar to that of ordinary (isotropic) liquids but
they can be easily aligned by an external magnetic or electric Fig. 36.8. Schematic of
field. An aligned nematic has the optical properties of a uniaxial molecular ordering in a
crystal and this makes them extremely useful in liquid crystal nematic phase
displays.
The characteristic properties of nematic liquid crystals may be summed up as follows:
● In nematic phase the molecules have no positional order; the location of their centres
of gravity is irregular as in a liquid.
● The molecules have long range orientational order; the axes of the molecules are
parallel to each other.
● The nematic liquid crystal consists of smaller regions each having its molecules
aligned parallel to a unique axis, called director. This axis varies in direction in
different regions of the crystal. Therefore, nematic liquid crystal appears turbid.
● The ordering of the molecules varies with temperature.
● The molecular orientation can be controlled with applied electric and magnetic
fields.
● When an external field orients the molecules in the same direction, the liquid crystal
appears transparent.
1060 A Textbook of Engineering Physics

● Special treatment of the surfaces supporting nematic liquid crystal causes alignment
of the molecules in a desired way: either parallel to the surface or perpendicular to
the surface.
● If the molecules are aligned parallel to the surface, the alignment is said to be
planar. If the molecules are aligned perpendicular to the surface, the alignment is
said to be homeotropic.
(ii) Smectic Phase: In the smectic state, the molecules maintain the general orientational
order of nematics, but also tend to align themselves in layers or planes. Motion is restricted to
within these planes, and separate planes are observed to flow past each other. The increased
order means that the smectic state is more “solid-like” than the nematic.

(a) Smectic - A (b) Smectic - C

Fig. 36.9. Schematic of molecular alignment in the smectic phases. (a) The smectic A phase has
molecules organized into layers. (b) In the smectic C phase, the molecules are tilted inside the
layers

Many compounds are observed to form more than one type of smectic phase. About 12 of
these variations have been identified. Three of the more often observed variations are known
as smectic-A, smectic-B and smectic-C.
The characteristic properties of smectic liquid crystals may be summed up as follows:
● These phases occur at lower temperatures than nematic phase.
● Smectic liquid crystals have a layer structure. Molecules align themselves parallel
in each layer. The layers can slide over one another
causing fluidity to the liquid crystal.
● In the smectic-A phase, the director is perpendicular to A B
the smectic plane, and there is no particular positional
order in the layer.
● Similarly, the smectic-B molecules orient perpendicular
to the smectic plane, but the molecules are arranged into Pitch

a network of hexagons within the layer. B A

● In the smectic-C phase, molecules are arranged as in
the smectic-A phase, but the director is at a constant tilt x

angle measured normally to the smectic plane.

● The value of the angle varies with temperature and can
be altered by an external electric field. A B
(iii) Cholesteric phase: This phase is called the cholesteric Cholesteric
phase because it was first observed for cholesterol derivatives. Fig. 36.10. Schematic of
Only chiral molecules (i.e., those that lack inversion symmetry) molecular alignment in the
can give rise to this phase. Therefore, the phase is also called cholesteric phase
Modern Engineering Materials 1061

chiral nematic phase. The phase exhibits a twisting of the molecules perpendicular to the
director, with the molecular axis parallel to the director. The finite twist angle between adjacent
molecules is due to their asymmetric packing, which results in longer-range chiral order.
When the molecules align in layers, this causes the director orientation to rotate slightly
between the layers, eventually bringing the molecules back into the original orientation.
Thus, the distance it takes for the director to rotate one full turn in the helix is known as the
pitch which is an important characteristic of the cholesteric phase. Note that the structure
of the chiral nematic phase repeats itself every half-pitch, since in this phase directors at 0°
and ±180° are equivalent. The pitch changes when the temperature is altered or when other
molecules are added to the host. A byproduct of the helical structure of the chiral nematic
phase is its ability to selectively reflect light of wavelengths equal to the pitch length, so that
a color will be reflected when the pitch is equal to the corresponding wavelength of light
in the visible spectrum. Mixtures of various types of cholesteric liquid crystals are often
used to create sensors with a wide variety of responses to temperature change. Such sensors
are used as thermometers often in the form of heat sensitive films to detect flaws in circuit
board connections, fluid flow patterns, condition of batteries, the presence of radiation, or in
novelties such as “mood” rings.
The characteristic properties of cholesteric liquid crystals may be summed up as follows:
● The phase exhibits twisting of molecules in successive layers, the molecular axis
being parallel to the director.
● In going from one layer to the next layer, the director turns by a certain angle and it
describes a helix with a pitch of about 0.2 to 20 μm.
● Though the chiral pitch refers to the distance it takes for the director to rotate one full
(360°) turn, the structure of the chiral nematic phase repeats itself every half-pitch.
● The periodicity of the structure leads to Braggs reflection of light at a wavelength
equal to the pitch divided by the refractive index of the cholesteric liquid crystal.
● The pitch varies with temperature. It also varies when other molecules are added to
the liquid crystal.
● Since the pitch varies with temperature, the wavelength of the reflected light and the
colour of the liquid crystal vary with temperature.
● Cholesteric liquid crystals exhibit high optical activity and cause the plane of
polarization of light to turn through angles of the order of 6000° to 7000°/mm.
36.3.2 Electric and Magnetic Field Effects
The response of liquid crystal molecules to an electric field is the major characteristic utilized
in industrial applications. The ability of the director to align along an external field is caused
by the electric nature of the molecules. When an external electric field is applied to the liquid
crystal, the dipole molecules tend to orient themselves along the direction of the field.
36.3.3 Liquid Crystal Displays (LCDs)
The most common application of liquid crystal technology is liquid crystal displays (LCDs).
They come in two different formats – (i) segment displays and (ii) matrix displays. A liquid
crystal display (LCD) is a thin, flat cell filled with liquid crystal material and arrayed in front
of a light source. It is often utilized in battery-powered electronic devices because it uses very
small amounts of electric power.
A segment LCD consists of two thin glass plates having transparent electrically
conducting coatings of indium tin oxide on their inner surfaces. These two plates will also
serve as the electrodes. The conducting coating on one of the glass plates is etched in the
1062 A Textbook of Engineering Physics

form of a digit or character, as illustrated in Fig. 36.11 (a). The surfaces of the electrodes
that are in contact with the liquid crystal material are treated so as to align the liquid crystal
molecules in a direction that is parallel to the glass and also to each other. In case of a twisted
nematic LCD, the surfaces are treated in mutually perpendicular directions. When the cell
is filled with liquid crystal material, the molecular layers spiral a quarter of a turn about the
twist axis normal to the glass plates. The molecules are aligned vertically on one electrode
and gradually they are turned until they are horizontal on the other electrode. This results
in a twisted nematic device. The cell is then sandwiched between two Polaroid sheets held
in crossed configuration, as shown in Fig. 36.11 (b). The rear Polaroid sheet is backed with
a reflecting film. With no actual liquid crystal between the polarizing filters, light passing
through the first filter would be blocked by the second (crossed) polarizer and appears dark.
On the other hand, a cell filled with liquid crystal rotates the plane of polarization of the
incident light and hence light is not blocked. As a result the device appears grey.

Polarizer

a
f b Glass plate A
g
Liquid Crystal Seal
e c
d Glass plate B

1 2 3 4 5 6 7
Glass plate A Glass plate B Polarizer
(a) (b)

Fig. 36.11

When a voltage is applied across Polarizer

the cell, the liquid crystal molecules
turn into the direction of the electric
field. If the applied voltage is large
enough, the liquid crystal molecules
in the center of the layer are almost
completely untwisted and the
polarization of the incident light is not
rotated as it passes through the liquid
crystal layer. The light will then be
mainly polarized perpendicular to the
second filter, and thus gets blocked
Light transmitted Light not transmitted
(Fig. 36.12 b). As a result, the areas
(a) Voltage OFF (b) Voltage ON
where electrodes are present in the
form of digits or character, appears Fig. 36.12
dark (see Fig. 36.13).
36.3.4 Flat Panel Liquid Crystal Displays
Small liquid crystal displays such as used in calculators and other devices have direct driven
segments. Voltage can be applied across one segment without interfering with other segments