14

CHAPTER

Conductors and
Resistors
The electrons in the outermost orbitals of the atoms of a solid determine its
electrical properties. In this chapter, the free electron model is introduced,
starting with a brief description of the wide range of resistivity of materials. The
justification for Ohm’s law is given, in terms of the free electron motion and the
collision processes in crystals. Materials for conductive and resistive functions
are discussed. The phenomenon of superconductivity and the recent discovery of
oxide superconductors are discussed at the end.

Units

SI units
Quantity Other units
Unit Symbol
Resistivity r ohm metre ohm m microohm-inch,
ohm-cm
Temperature coefficient per kelvin K–1 per °F
of resistance a
Conductivity s per ohm per ohm–1 m–1 mho/cm
metre
de Broglie wavelength l metre m Å
Wave number k per metre m–1 –
123

Kinetic energy E
Fermi energy EF joule J erg, eV
Drift velocity v metre per second m s–1 –
Field gradient e volt per metre V m–1 volts/mil
Current density Je ampere per m2 A m–2 amp/cm2

Constants
Planck’s constant h = 6.626 ´ 10–34 J s
Rest mass of electron m0 = 9.109 ´ 10–31 kg
Charge of electron e = 1.602 ´ 10–19 C
332
 The Resistivity Range 333

14.1 The Resistivity Range

Electrical resistivity (or conductivity) is probably the most remarkable of all

physical properties, in that it varies over 25 orders of magnitude. To get a feel
for this wide range, Table 14.1 lists the electrical resistivity, at room temperature,
of a number of materials which are important from the engineering point of
view. The materials fall into three broad categories.

TABLE 14.1
The Resistivity of Materials (ohm m)

10–9 10–7 10–5 10–3 10–1 101 103

Ag
Cu Al Ni Sb Bi Ge Ge Si
Au Fe Graphite (doped) (pure) (pure)
Metals Semiconductors

105 107 109 1011 1013 1015 1017

Window Bakelite Porcelain Lucite PVC SiO2
glass Diamond Mica (pure)
Rubber, Nylon
Polyethylene
Insulators

Conductors are metals and alloys. Gold, silver and copper are among the
best conductors of electricity. Therefore, their electrical resistivities are the
lowest, as shown in Table 14.1. They are followed by aluminium whose
resistivity is 60% higher than that of copper. Transition metals such as iron and
nickel are not as good conductors as the above. Still poorer conductors are the
semimetals of the fifth column, e.g., antimony and bismuth. Graphite, with one
of its bonding electrons resonating between the (sp2) bonds, also fall in this
category of semimetals. The electrical resistivity of conductors ranges from 10–9
to 10–3 ohm m. The electrical conductivity, being the reciprocal of resistivity,
ranges from 109 ohm–l m–l to 103 ohm–1 m–1.
When the resistivity is in the range 10–3–103 ohm m, we have the second
category of materials known as semiconductors. They form the base for a
number of solid state devices. Here, the resistivity is a strong function of small
concentrations of impurities. Doped germanium, with an impurity content of a
few tens per million, can have a resistivity about two orders of magnitude lower
than that of pure germanium, see Table 14.1. Pure silicon has a higher resistivity
than pure germanium.
The third category of materials are insulators. Common electrical insulating
materials such as polyethylene, bakelite, lucite, mica, PVC, rubber and porcelain
fall in this category. The resistivity range for this category extends from 104 to
beyond 1017 ohm m. Here, a difference in resistivity of some twelve orders of
334 Conductors and Resistors

magnitude is noticeable between a silica glass and soda-lime-silica (window

glass). This striking difference is a result of the ionic conductivity of window
glass. The relatively loosely-bound sodium and calcium cations in soda-lime
glass diffuse and conduct much more readily, as compared to the tightly-bound
immobile silicon cations in pure silica. Ionic conduction and ionic diffusivity are
closely related phenomena, recall Sec. 8.6.

14.2 The Free Electron Theory

The conducting properties of a solid are not a function of the total number of
electrons in the solid, as only the outermost electrons of the atoms can take part
in conduction. In the free electron model, the outermost electrons of an atom are
not bound to that atom, but are free to move through the whole solid. These
electrons have been variously called the free electron cloud, the free electron gas
or the Fermi gas.
In the free electron theory, the basic assumption is that the potential field
due to the ion cores is uniform throughout the solid. The free electrons have the
same potential energy everywhere in the solid. Due to the electrostatic attraction
between a free electron and the ion cores, this potential energy will be a finite
negative value. As we are interested only in energy differences, we can assume
this constant potential to be zero. Then the only energy that we have to consider
is the kinetic energy. This kinetic energy is substantially lower than that of the
bound electrons in an isolated atom, as the field of motion for the free electron
is considerably enlarged in the solid as compared to the field around an isolated
atom. This effect is explained below.
Electrons have both particle-like and wave-like characteristics. The de
Broglie wavelength of an electron l is related to its momentum mv as
h
l= (14.1)
mv
where h is Planck’s constant, m is the mass of the free electron and v is its
velocity. The wavelength is inversely related to the magnitude of the wave
number vector k:
2Q
k= (14.2)
M
As the velocity of the free electrons is much smaller than that of light, we can
ignore relativistic effects and use the classical relation for kinetic energy E:

1
E= mv 2 (14.3)
2
Substituting Eqs. (14.1) and (14.2) into Eq. (14.3), we obtain

h2 k 2
E= (14.4)
8Q 2 m
 The Free Electron Theory 335

The kinetic energy E increases as the square of the wave number. This parabolic
relationship between E and k is shown in Fig. 14.1. As l is inversely related to
k, the electron with the largest de Broglie wavelength will have the lowest
kinetic energy. With the enlarged field of motion in the solid, the electrons can
have larger wavelengths and hence lower kinetic energies.

0 k
Fig. 14.1 The parabolic relationship between the kinetic energy E of a free
electron and its wave number k.

Even though the variation in the kinetic energy E is shown to be continuous

in Fig. 14.1, these are in fact very closely spaced discrete energy levels, as the
quantum restrictions arising from the Pauli exclusion principle apply to the free
electrons as well. Let us consider a solid, in which the electron motion is
unidirectional. Let L be the length of this axis of motion. The longest
wavelength that is permissible for the free electron is equal to twice this length
as shown in Fig. 14.2, as the amplitude of the wave should be zero at both ends

l = (2/3)L l = 2L
l=L
Amplitude

0 L
x
Fig. 14.2 The de Broglie wavelengths of the first few electrons moving along x.
336 Conductors and Resistors

of this length. The next smaller wavelength that is permissible is equal to L, the
third one is equal to (2/3)L, and so on. The corresponding values of the
wavenumber k are p /L, 2p /L, ¼, as obtained from Eq. (14.2).
In a solid that is electrically neutral, a free electron having a certain speed in
one direction will always be associated with another electron having the same
speed but moving in the opposite direction. (If this were not so, there will be a net
flow of electrons in one direction, even in the absence of an externally applied
field.) In other words, the wave number k takes on both positive and negative
values. Thus, for the first two states, we have k equal to ± p /L and
± 2p /L. In general, k = ± np /L, where the quantum number n takes on successively
increasing integer values. Substituting k = ±np /L in Eq. (14.4), we obtain
h 2 n2
E= (14.5)
8mL2
In a three-dimensional solid, the free electron can move in any direction in
space. So, three quantum numbers nx, ny and nz corresponding to the three
coordinate axes are used to define each quantum state. nx, ny and nz take on
successively-increasing integer values. (nx2 + ny2 + nz2) is substituted in place of
n2 used for the unidirectional motion. Equation (14.5) can be rewritten for three-
dimensional motion as
È nx2  n y2  nz2 Ø
E = h2 É Ù (14.6)
Ê 8mL2 Ú
Each distinct combination of nx , ny and nz corresponds to a quantum state, where
two electrons of opposite spins can reside. Several combinations of nx , ny and nz
can result in the same value of E. Quantum states with the same energy are said
to be degenerate. The successive energy levels calculated from Eq. (14.6) are so
close to one another that E can be considered to be varying continuously as
illustrated for unidirectional motion in Fig. 14.1.

Example 14.1 Calculate the energy difference between the nx = ny = nz = 1

level and the next higher energy level for free electrons in a solid cube of
10 mm ´ 10 mm ´ 10 mm.

Solution
L = 10 mm = 10–2 m
From Eq. (14.6),
(6.626 – 10 34 )2 (12  12  12 )
E=
8 – 9.109 – 10 31 – (10 2 )2
= 1.81 ´ 10–33 J
There are many equal energy quantum states above this energy level, with values
of nx, ny and nz as (1,1,2), (1,2,1), (2,1,1), ¼ . For these states,
 The Free Electron Theory 337

(6.626 – 10 34 )2 (12  12  22 )

E=
8 – 9.109 – 10 31 – (10 2 )2
= 3.62 ´ 10–33 J
The energy difference between the first and the next higher energy levels is
extremely small, only 1.81 ´ 10–33 J, so that we are justified in assuming E to be
continuously varying with k. Compare this energy difference with the value of
thermal energy at temperature as low as 1 K, 1.38 ´ 10–23 J, which is ten orders
of magnitude larger.

The wave forms for the free electrons starting from the lowest energy level
are sketched in Fig. 14.2 for unidirectional motion. The probability of finding an
electron at any point along the length is proportional to the square of the
amplitude of the wave at that point. This means that, for the free electron with
l = 2L, the probability of finding it at the midpoint of the length is a maximum.
This, however, is not an acceptable result, as the probability of finding the free
electron must be the same anywhere within the solid. To overcome this
difficulty, the waves are considered to be travelling waves with a constant
velocity, so that the time averaged probability of finding an electron is constant
throughout the solid.
In keeping with the minimum energy criterion and the Pauli exclusion
principle, the free electrons occupy successive quantum states of increasing
kinetic energy. The energy corresponding to the highest filled level at 0 K is
called the Fermi energy EF. At 0 K, the free electrons occupy all the levels up to
the Fermi level, leaving all those above it empty. At temperatures above 0 K, due
to thermal excitation, there is a finite probability of some of the electrons from
below the Fermi level moving to levels above E F. This probability is given by
the Fermi–Dirac statistics, which takes into account the quantum restrictions due
to the Pauli exclusion principle. The probability of occupation P(E) of an energy
level E by an electron is given by
1
P(E) = (14.7)
1  exp [( E  EF )/ kT ]

If EF is independent of temperature, P(E) varies with E as shown in

Fig. 14.3. At 0 K, P(E) remains constant at unity with increasing E up to E F,
where it falls abruptly to zero. At T1 > 0 K, some of the electrons just below the
Fermi level are thermally excited to higher levels just above EF. So, P(E) is less
than one just below the Fermi level and is greater than zero just above the Fermi
level. At a still higher temperature, more electrons leave the lower energy levels
and occupy higher levels. The cross-over point, where P(E) is 0.5 for different
temperatures in Fig. 14.3, occurs at the same value of the energy level which is
EF. This is true, provided that EF is independent of temperature, an assumption
valid for most ordinary temperatures. Under such conditions, the Fermi level can
be defined as that level which has a 50% probability of occupation by an
electron at any temperature.
338 Conductors and Resistors

T2 > T1 T1 > T 0

1
T0 = 0 K
P(E)

0 EF
E
Fig. 14.3 The Fermi–Dirac distribution of free electrons at different
temperatures.

14.3 Conduction by Free Electrons

We have already noted that the wave number k takes both positive and negative
values. For every electron moving with a certain speed in a direction, there is
another electron moving with the same speed in the opposite direction. This equal
and opposite velocity distribution in a neutral solid can be biased by an externally
applied electric field to yield a net velocity in one direction. With this biasing, the
solid conducts electricity. Under an applied field, the E–k relationship of Fig. 14.1
gets modified to the distribution shown in Fig. 14.4. The negatively charged
Field
– +

EF

0
k
Fig. 14.4 Electrons moving towards the positive end of the applied field acquire
extra velocity, while those moving in the opposite direction lose some velocity.

electrons are accelerated towards the positive end of the field. The velocity of the
fastest electron moving in the direction of the positive end has a larger magnitude
than that of the fastest electron moving towards the negative end of the field. Such
redistribution is possible, only when empty electron states are available
immediately above the Fermi level. This availability is a basic characteristic of
conductors, as opposed to semiconductors and insulators.
 Conduction by Free Electrons 339

The force experienced by an electron of charge e in an applied field of

gradient e can be equated to the force as defined in the classical law:
e e = ma (14.8)
where m is the mass of the electron and a is the acceleration due to the applied
field. The electrons that are accelerated towards the positive end of the field do
not continue to increase their velocity indefinitely. They collide with obstacles
on their way. Depending on the time interval between two successive collisions,
the electrons acquire an average increment of velocity called drift velocity, all of
which they lose during a collision, as illustrated in Fig. 14.5. The drift velocity
is the extra velocity that electrons acquire over and above their normal velocity
in the absence of a field.

Collisions
Velocity

vd

t Time

Fig. 14.5 The extra velocity acquired by an electron due to an applied field is
lost on collision with an impurity, imperfection or phonon.

If the average collision time is t and vd is the drift velocity acquired by the
electrons, Eq. (14.8) can be rewritten as
m (vd /t) = e e (14.9)
or
F eU
vd = (14.10)
m
The flux Je due to the flow of electrons is called the current density:
ne2UF
Je = nevd = (14.11)
m
where n is the number of free electrons of charge e. This is in the form of
Ohm’s law. As conductivity s is by definition the flux per unit potential
gradient, we have
ne2U
s= (14.12)
m
The electrical resistivity r is the reciprocal of conductivity.
340 Conductors and Resistors

Example 14.2 Calculate the conductivity of copper at 300 K. The collision

time t for electron scattering is 2 ´ 10–14 s at this temperature.

Solution The number of free electrons per m3 of copper is given by

n = 6.023 ´ 1023 ´ 8960/0.06354 = 8.50 ´ 1028 m–3
Electronic charge e = 1.602 ´ 10–19 C
Mass of a free electron ~ rest mass
= 9.1 ´ 10–31 kg
Using Eq. (14.12),
s = 8.50 ´ 1028 ´ 1.6022 ´ 10–38 ´ 2 ´ 10–14/(9.1 ´ 10–31)
= 4.8 ´ 107 ohm–1 m–1

The mean free path l of an electron is the mean distance it travels between
successive collisions. For an ideal crystal with no impurities and imperfections,
the mean free path at 0 K is infinite. That is, there are no collisions and the
electrical conductivity is ideally infinite.
Solute atoms provide effective scattering centres for the electrons. Their
introduction into the crystal results in collisions, decreasing the mean free path
and the conductivity. Similarly, other point imperfections, dislocations and grain
boundaries also increase the scattering and decrease the conductivity. Among
these scattering centres, impurities dissolved in the crystal lattice as solutes are
more effective than the others.
At temperatures above 0 K, the atoms vibrate randomly about their mean
positions. These vibrations can be considered as elastic waves in the crystal and
are called phonons. Their random nature destroys the ideal periodicity of a
crystal and interferes with the electron motion. Consequently, the mean free path
and the conductivity decrease with increasing temperature. At low temperatures,
the mean free path and the collision time are proportional to the cube of the
reciprocal of temperature in kelvin. At higher temperatures, the vibrational
frequency tends to become constant. Here, the mean free path and the collision
time are proportional to the reciprocal of temperature.
The electrical resistivity of pure copper with a low density of imperfections
is shown as a function of temperature in Fig. 14.6. The variation is cubic at low
temperatures and is linear at higher temperatures. If nickel is introduced in the
copper lattice as a solute, the resistivity of the alloy is higher at all temperatures
than that of pure copper. The increase in resistivity is proportional to the amount
of nickel added. The resistivities of Cu-2% Ni and Cu-3% Ni are shown in
Fig. 14.6. The resistivities of the Cu–Ni alloys at 0 K have a finite residual
value, in contrast to pure copper, where the resistivity is tending to zero at 0 K.
This residual resistivity is attributable to the scattering centres provided by the
solute atoms which are present even at 0 K, where there is no phonon scattering.
The additivity rule of Mattheissen describes this behaviour:
r = rT + rr (14.13)
 Conductor and Resistor Materials 341

Cu-3% Ni

Cu-2% Ni
4
Resistivity, 10–8 ohm m

2
Pure Cu

0 100 200 300

Temperature, K

Fig. 14.6 The electrical resistivity of pure copper and Cu –Ni alloys as a function
of temperature.

where rT is the thermal part of the resistivity and rr is the residual resistivity
due to solute atoms and other imperfections in the crystal.

Example 14.3 The resistivity of pure copper at room temperature is

1.8 ´ 10–8 ohm m. The resistivity of Cu-4% Ni alloy at room temperature is
7.0 ´ 10–8 ohm m. Estimate the resistivity due to impurity scattering by 1% of
nickel in the copper lattice.

Solution Taking the resistivity of copper at 0 K to be negligible,

1.8 ´ 1.0–8 ohm m is the thermal part of resistivity of copper as well as of the
alloy between 0 K and room temperature. The impurity scattering by 1% of
nickel is then equal to (7.0 – 1.8) ´ 10 –8/4 = 1.3 ´ 10 –8 ohm m.

14.4 Conductor and Resistor Materials

The resistivity, the temperature coefficient of resistance, the density and the
tensile strength of typical conductor and resistor materials are listed in
Table 14.2.
342 Conductors and Resistors

TABLE 14.2
Properties of Typical Conductors and Resistors at Room Temperature

Resistivity, Temperature Density, Tensile

Material 10–8 ohm m coefficient a , 103 kg m–3 strength*,
K–1 MN m–2
Silver 1.5 0.0040 10.49 125
Copper 1.7 0.0043 8.96 210
Gold 2.2 0.0035 19.32 138
Aluminium 2.8 0.0042 2.70 60
Tungsten wire 5.5 0.0045 19.3 2800
Molybdenum wire 4.9 0.0050 10.2 700
Platinum wire 10.9 0.0037 21.45 350
Tantalum wire 15.5 0.0032 16.6 490
Nichrome wire 108 0.0001 8.41 1000
Manganin 48 0.00002 8.2 420
Kanthal wire 135 0.00003 7.2 800
*
The tensile strength values given here are approximate, as they depend on the prior
thermal and mechanical history of the metal.

For use as conductors in applications such as transmission lines and

distribution lines, low I 2R loss is the primary consideration and the choice would
be from amongst the best conductors, keeping in view the cost, fabricability and
mechanical strength. Copper and aluminium are the most likely choices. For
long distance transmission lines, aluminium is chosen. As a large cross section
would reduce the I 2R loss, thick cables are preferred. If the elastic modulus of
the aluminium cables is improved by reinforcement with steel as in ACSR
(aluminium conductor steel reinforced) cables, the distance between successive
poles along the transmission line can be substantially increased. More expensive
copper is used for distribution lines, busbars and other energy conversion
applications. OFHC (oxygen-free high conductivity) copper is often specified.
Among the common solutes in copper, Fe, P and As are the most harmful in
impairing the electrical conductivity.
For electrical contacts used in switches, brushes and relays, the material
must possess high electrical conductivity, high thermal conductivity, high
melting point and good oxidation resistance. High thermal conductivity helps to
dissipate the heat effectively. High melting point is desirable so that any
accidental overheating does not fuse together the contact points. Good oxidation
resistance is necessary to keep the contact clean and free of insulating oxides.
Copper and silver largely satisfy the above requirements. For low cost, copper is
commonly used. For critical contacts such as those used in aircrafts, silver is
preferred. The low mechanical strength of pure silver is increased by the
dispersion of fine particles of CdO. Dislocations moving in silver have to bend
 Conductor and Resistor Materials 343

around the dispersed CdO particles and a fine dispersion increases the strength.
CdO improves the wear resistance of silver. It also decomposes at the melting
point of silver, thereby absorbing much of the heat generated by arcing and
minimizing the loss of expensive silver by evaporation.
For resistor applications, the primary requirements are uniform resistivity
(achieved in a homogeneous alloy), stable resistance (achieved by avoiding
metallurgical changes such as ageing and relaxation of residual stresses), small
temperature coefficient of resistance (a ) and low thermoelectric potential with
respect to copper. A small a minimizes the error in measurements due to
variations in the ambient temperature. a is defined as
1 dR
a= (14.14)
R dT
where R is the resistance of the alloy at temperature T. For manganin alloy
(87% Cu and 13% Mn), a is only 20 ´ 10–6 K–1 as against 4000 ´ 10–6 K–l for
pure copper. Constantan (60% Cu and 40% Ni) is another such alloy. These
alloys have also good resistance to atmospheric corrosion, another desirable
property in a resistor.
A low thermoelectric potential with respect to copper, to which the resistor
is commonly connected, reduces errors due to temperature differences between
junctions. For high precision, dissimilar junctions should be maintained at the
same temperature so that no thermoelectric potential develops.
Ballast resistors are used to maintain constant current in some industrial
circuits. If the flow of current increases, the temperature increases and the
resistance of the ballast increases. This in turn decreases the current in the circuit
towards the initial value. An iron–nickel alloy with 71% Fe and 29% Ni with
excellent oxidation resistance and a high a of 4500 ´ 10–6 K–l is used for this
application.
For heating elements, the primary requirements are high melting point, high
electrical resistance, good oxidation resistance, good creep strength, low elastic
modulus and low thermal expansion. The last two requirements help in reducing
thermal fatigue due to repeated heating and cooling. The heating elements
should be designed in a way as to allow unhindered expansion and contraction,
for example, in the form of a coil of wire. Nichrome (80% Ni and 20% Cr) and
Kanthal (69% Fe, 23% Cr, 6% Al and 2% Co) are used for heating elements up
to 1300°C. SiC and MoSi2 can be used at higher temperatures up to 1700°C.
Graphite, by virtue of its high sublimation temperature and good fabrication
properties, is also widely used up to 1800°C. Molybdenum and tantalum need
protective atmospheres at high temperatures, as their oxidation resistance is poor.
By virtue of its very high melting point (3410°C), tungsten is used for filaments
in incandescent lamps. Its creep resistance at white heat (above 1500°C) can be
improved by dispersion hardening with thoria (ThO2).
Resistance thermometers should have a high temperature coefficient of
resistance for good sensitivity. A pure metal is obviously the choice for this
application. Platinum, obtainable in very pure form, is used.
344 Conductors and Resistors

14.5 Superconducting Materials

14.5.1 The Superconducting Phenomenon

Referring to Fig. 14.7, the resistivity of pure silver decreases gradually to a low
but measurable value of ~10–11 ohm m at 0 K. Silver is not a superconductor.

Silver 20 Tin
10
r, 10–11 ohm m

r, 10–11 ohm m
10
5

0 10 20 0 2 4 6 8 10
T, K T, K
Fig. 14.7 The electrical resistivity r of (a) pure silver, and (b) tin, as a function
of temperature near 0 K.

The resistivity of tin, on the other hand, falls abruptly at about 4 K to a value
below the measurable limit of 10–20 ohm m. The temperature at which this
abrupt fall occurs is called the superconducting transition temperature Tc. Tin is
in the normal state above Tc and in the superconducting state below Tc. The
resistivity in the superconducting state is zero for all practical purposes. A
current induced in a superconducting ring persists for years with negligible
decay.
The first application that was conceived of the superconducting effect was to
produce a large permanent magnetic field by inducing a perpetual current in a
superconductor. However, at temperatures below Tc, as the magnetic field
strength reaches a critical value Hc, the superconductivity disappears. This is
illustrated in Fig. 14.8. At T = Tc, Hc = 0. Hc increases as the temperature
decreases below Tc. A similar graph relates the critical current density Jc that a
superconductor can carry at temperatures below Tc. At T = Tc, Jc = 0. Jc
increases as the temperature decreases below Tc in a manner similar to Hc in
Fig. 14.8. In other words, the maximum current that a superconductor carries at
a given temperature below Tc is limited by the magnetic field it produces at the
surface of the superconductor.
In the superconducting state, the flux lines of a magnetic field are ejected
out of the superconductor, as illustrated in Fig. 14.9. This effect is known as the
Meissner effect. A superconductor is a perfect diamagnetic material, with the
magnetic susceptibility c equal to –1.
 Superconducting Materials 345

0.09

0.06
m0 Hc, Wb m–2

0.03

0 2 4 6 Tc 8
T, K
Fig. 14.8 The critical magnetic field Hc as a function of temperature for the
superconducting-to-normal transition. Hc = 0 at Tc.

T > Tc T < Tc
(a) (b)
Fig. 14.9 The flux lines of an applied magnetic field are ejected out of the solid
on crossing to the superconducting state.

The superconducting effect has been explained in the Bardeen–Cooper–

Schreiffer (BCS) theory as a three-way interaction between two electrons and a
phonon (a phonon is the quantum of energy in an elastic wave, analogous to a
photon, which is the quantum of energy in an electromagnetic wave). The
phonon scattering due to lattice vibrations felt by electron 1 is exactly nullified
by electron 2, such that the electron pair moves through the lattice without
feeling any scattering effect from lattice vibrations. This attractive interaction
energy is more than the repulsive energy between the two electrons; yet it is
small enough in magnitude to be disrupted by the thermal energy available at
T > T c.
346 Conductors and Resistors

14.5.2 Type I and Type II Superconductors

Type I or the ideal superconductor when placed in a magnetic field repels the
flux lines totally, till the magnetic field attains the critical value Hc. The
magnetization M is equal to –H up to Tc, where it drops to zero, as shown in
Fig. 14.10a. Type II or hard superconductors are those in which the ideal
behaviour is seen up to a lower critical field Hc1, beyond which the magnetiz-
ation gradually changes and attains zero at an upper critical field designated Hc2,
see Fig. 14.10b. The Meissner effect is incomplete in the region between Hc1 and
Hc2; this region is known as the vortex region. The normal behaviour is observed
only beyond Hc2. The magnetic flux lines gradually penetrate the solid, as the
field is increased beyond Hc1 and the penetration is complete only at Hc2.

Type I Type II

Super-
–M

–M
Normal

conducting

Normal
Super-
conducting
Vortex

Hc Hc1 Hc Hc2
H H
(a) (b)
Fig. 14.10 The magnetization M versus the critical magnetic field Hc for
(a) Type I and (b) Type II superconductors.

Type II superconductors are of great practical interest because of the high

current densities that they can carry. The type II state is determined by the
microstructural condition of the material. Heavily cold-worked and recovery
annealed material have cell walls of high dislocation density and this
microstructure effectively pins the magnetic flux lines and makes their
penetration difficult. Grain boundaries also exert a pinning action; an extremely
fine grain size is effective in increasing Hc2. Similarly, the dispersion of very
fine precipitates in the matrix with the interparticle spacing of about 300 Å
results in optimum flux pinning. The critical current density Jc also increases, as
Hc2 increases. The very high current densities obtainable in a Nb-40% Ti alloy at
4.2 K at a magnetic field strength of 0.9Hc2 are listed below as a function of the
microstructural condition.

Microstructure Jc , A m–2
Recrystallized 105
Cold worked and recovery annealed 107
Cold worked and precipitation hardened 108
 Superconducting Materials 347

14.5.3 Potential Applications

Superconducting materials are already in use for producing very strong magnetic
fields of about 50 Tesla, which is much larger than the field obtainable from an
electromagnet. Such high magnetic fields are required in MHD power
generators. At high magnetic field strengths, a conventional copper solenoid
consumes about 3 MW, whereas a superconducting magnet consumes about
10 kW. In addition, the copper solenoid will require about 2000 l/min of water
circulation to avoid burning down due to Joule heating. Magnetic energy can be
stored in large superconductors and drawn as required, to counter voltage
fluctuations during peak loading.
Superconductors can be used to perform logic and storage functions in
computers. A Josephson junction consists of a thin layer of insulating material
between two superconducting solids. The unique current-voltage characteristics
associated with the Josephson junction are suitable for memory elements.
Switching times of the order of 10 ps (l0–11 s) have been measured.
Arising from the Meissner effect, a superconducting material can be
suspended in air against the repulsive force from a permanent magnet. This
magnetic levitation effect can be used in transportation. As there is no I 2R loss
in a superconductor, power can be transmitted through superconducting cables
without loss. With the liquid N2 environment, the economics of such
transmission has become more favourable for adoption.

14.5.4 New Developments

Till the year 1986, the highest known transition temperature Tc was 23 K in the
Nb3Ge alloy. In 1986, Bednorz and Mueller reported a significant increase in Tc
to 34 K in a La–Ba–Cu–O ceramic material. This Nobel prize-winning discovery
was soon followed by further big increases in the transition temperature. The
oxide with the nominal formula YBa2Cu3O7–x has a transition temperature of
~90 K.* This transition temperature is 13 K above the boiling point of liquid N2
(77 K). Compare this with the boiling point of liquid He (4 K) and that of liquid
H2 (23 K), which is a safety hazard. That a superconductor can function in
liquid nitrogen is itself a remarkable achievement. A liquid nitrogen environment
is far easier and cheaper to obtain than a liquid helium medium. Further, it takes
about 25 times more energy to cool from 77 K to 4 K than from room
temperature to 77 K.
The oxide YBa2Cu3O7–x is prepared by heating compacted powder mixtures
of Y2O3, BaCO3 and CuO in the right proportion to temperatures between 900
and 1100°C. BaCO3 decomposes at this temperature to BaO and CO2. This is
often followed by another annealing treatment at 800°C in an atmosphere of
oxygen. The heat treatment conditions such as the partial pressure of oxygen in
the atmosphere are critical for obtaining a high Tc. The crystal structure of the
powder product obtained is related to the cubic perovskite structure as illustrated
in Fig. 14.11. Three body-centered cubic unit cells are stacked one above
another. The atom distributions in the unit cells are as follows:
*
New compounds such as Tl(Bi)-Ba(Sr)-Ca-Cu-O are now known with a reproducible Tc
of about 125 K.
348 Conductors and Resistors

Copper
Barium

Yttrium

Oxygen

Vacant oxygen site

Fig. 14.11 The crystal structure of yttrium–barium–copper oxide.

Cu at the body corners : 8 ´ 1/8 ´ 3 = 3

Ba at two body centres : 1´2=2
Y at one body centre : 1´1=1
O at mid-points of edges : 12 ´ 1/4 ´ 3 = 9
Some oxygen positions are vacant and their number is between 6.5 and 7.
The superconducting properties appear to be a sensitive function of the oxygen
content and, therefore, of the partial pressure of oxygen during heat treatment.
The engineering aspects of this oxide superconductor still remain elusive. It
is reactive, brittle, unable to support any significant stress and cannot be easily
formed or joined. Also, the superconducting properties deteriorate during heating
for forming purposes or even in a humid room. Explosive forming and isotactic
pressing are two forming processes that appear to be promising. In explosive
forming, the oxide powder is placed between two copper plates in a container,
which is filled with an explosive. On detonating, shock waves are generated and
compressive pressures up to 50 000 atm are exerted on the composite sandwich-
like arrangement, which gets compacted into one solid mass. The temperature
attained in the chamber is less than 100°C. The very high compressive stress
plastically deforms the oxide, which should help to increase its critical-current
carrying capacity. Once encapsulated, the outer copper layer provides for easy
external connection and also ensures that there is no deterioration of the
superconducting properties due to environmental effects.
 Summary/Problems 349

SUMMARY

1. Electrical resistivity of materials covers a very wide range, over 25 orders

of magnitude.
2. In a solid where the potential field is uniform, the total energy of the free
electrons can be taken to be their kinetic energy. This energy varies as the
square of the wave number k.
3. The Fermi energy level is defined as that level which has a 50%
probability of occupation by an electron. At 0 K, the Fermi level is the
highest filled electron energy level.
4. At temperatures above 0 K, due to thermal excitation, the distribution of
electrons near the Fermi level changes as given by the Fermi–Dirac
statistics.
5. The availability of empty electron states immediately above the Fermi level
is a basic requirement for electrical conduction.
6. Ohm’s law arises, as conducting electrons acquire a drift velocity under an
applied field, due to collisions with impurities, imperfections and phonons.
7. For high conductivity and large temperature coefficient of resistance, a,
pure metals are chosen. For heating purposes by I 2R dissipation and for
low temperature coefficient of resistance, alloys are more suitable.
8. The transition temperature Tc, the critical magnetic fields Hc1 and Hc2, and
the critical current density Jc are the main properties of interest of
superconductors. Among these, Jc is structure sensitive and can be
maximized by proper control of the microstructure.
9. The recent discovery of high Tc oxide superconductors is an important
advance in scientific research, with far reaching technological implications.

PROBLEMS

14.1 Calculate the kinetic energy and the de Broglie wavelength of an electron
that falls through a potential of (i) 500, (ii) 5000, and (iii) 50000 V.
Answer: (i) 500 eV, 0.55 Å, (ii) 5000 eV, 0.17 Å, and (iii) 50000 eV,
0.055 Å
14.2 Derive the kinetic energy of free electrons as a function of their wave
number.
14.3 What should be the energy of the quantum state, which has 19 for the sum
of the squares of the quantum numbers nx, ny and nz ? What is the
degeneracy of this state? Assume the solid to be 10 mm ´ 10 mm ´ 10 mm.
Answer: 1.15 ´ 10–32 J; 24.
350 Problems

14.4 Show that the probability of occupancy of energy level E by an electron

is (i) negligible for E  EF, (ii) 0.5 for E = EF, and (iii) ~1 for
E  EF .
14.5 The Fermi level for potassium is 2.1 eV. Calculate the velocity of the
electrons at the Fermi level.
Answer: 8.6 ´ 105 m s–l.
14.6 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 on either side
of EF.
Answer: 0.01.
14.7 The resistance of a sample of copper wire of 0.1 mm diameter and 0.2 m
long is 0.439 ohm. Calculate its resistivity and conductivity.
Answer: 1.72 ´ 10–8 ohm m and 5.81 ´ 107 ohm–1 m–1.
14.8 The resistivity of silver at room temperature is 1.6 ´ 10–8 ohm m.
Calculate the collision time for electron scattering.
Answer: 3.8 ´ 10–14 s.
14.9 The resistivity of aluminium at room temperature is 2.62 ´ 10–8 ohm m.
Calculate the incremental velocity acquired by the free electrons in a
field gradient of 100 V m–l.
Answer: 0.39 m s–1.
14.10 Calculate the per cent increase in resistivity of (i) copper and
(ii) nichrome on heating from 300 K to 1000 K.
Answer: (i) 300%, (ii) 7%.
14.11 If the ambient temperature varies by ±10°C about the mean value of
20°C, what is the maximum possible error between two measurements of
resistance made at different times of (i) a copper resistor, and (ii) a
manganin resistor?
Answer: (i) 8.6%, (ii) 0.04%
14.12 Resistivity changes of 1 in 104 can be detected by means of a bridge.
What is the smallest change in temperature that you can measure with a
platinum thermometer at –200°C. At this temperature, a = 0.03 K–1.
Answer: 0.0033°C.
14.13 Explain why nichrome and not copper is used as a heating element.
14.14 A wire whose diameter is 2 mm must carry a 20 A current in an
application. The maximum power dissipation along the wire is 5 W m–l.
Calculate the minimum permissible conductivity of the wire for this
application.
Answer: 2.55 ´ 107 ohm–1 m–l.
14.15 Estimate the mean free path of free electrons in pure copper at 4 K. The
collision time for phonon scattering at 4 K is 10–9 s. The Fermi energy
level for copper is 7 eV.
Answer: 1.57 mm.
 Multiple Choice Questions 351

14.16 Explain why aluminium used in long distance transmission lines cannot
be strengthened by solid solution.

MULTIPLE CHOICE QUESTIONS

1. The SI units of electrical conductivity are

A. kg–1 m–3 s3 A2 B. kg m3 s–3 A–2 C. ohm m D. ohm–l m–3
2. The correct order of increasing resistivity of the following is
A. nickel, doped silicon, sodium silicate, pure silica
B. doped silicon, pure silica, nickel, sodium silicate
C. pure silica, doped silicon, sodium silicate, nickel
D. nickel, pure silica, doped silicon, sodium silicate
3. Volt in SI units is
A. W A B. N m2 s–1 A–1 C. kg m2 s–3 A–1 D. ohm A–1
4. The degeneracy of the quantum states with (nx2 + ny2 + nz2) = 6 is
A. 8 B. 12 C. 24 D. 48
5. The energy level difference between two successive levels for the lowest
energy free electrons is in order of magnitude
A. 1 eV B. 10–33 J C. 10–23 J D. 10–19 J
6. In a metal of 1 cm ´ 1 cm ´ 1 cm, the lowest energy free electron has a
wavelength of
A. 0.5 cm B. 1 cm
C. 2 cm D. need Fermi energy to compute
7. The probability of finding a free electron in Ag at 300 K at an energy level
1.01EF is given by (EF for Ag = 5.5 eV)
A. 0.5 B. 0.11 C. 0.005 D. 0
8. If the Fermi energy of silver is 5.5 eV, the wave number of the fastest
electron at 0 K has the magnitude (in m–l)
A. 0.85 ´ 1010 B. 7.54 ´ 1010 C. 1.20 ´ 1010 D. 0.19 ´ 1010
9. The Fermi level for Cu is 7 eV. The maximum velocity of free electrons at
0 K is
A. 1570 km s–1 B. 1110 km s–1 C. 860 km s–1 D. 0 km s–1
10. The acceleration (m s–2) of a free electron in an electric field of
100 V m–1 is
A. 1.76 ´ 103 B. 1.1 ´ 1013 C. 1.76 ´ 1013 D. 1.1 ´ 1032
352 Multiple Choice Questions

11. The Fermi level EF depends on the length L of a linear solid as

A. l/L2 B. l/L3
C. l/L D. is independent of L
12. At 0 K, the probability of finding an electron at energy level E is unity,
when
A. E  EF B. E < EF C. E > EF D. E  EF
13. The classical equation for the kinetic energy can be used for the free
electrons, because, as compared to the velocity of light, their velocity is
A. more B. slightly less C. much less D. zero
14. The probability of occupation by a free electron of the energy level 3.5 eV
at room temperature in Cu is (EF for Cu = 7 eV)
A. 0 B. 0.5 C. 1 D. exp (–134.6)
15. The collision time for electron scattering in pure Ag (EF = 5.5 eV) at
300 K is 10–14 s. The mean free path of electrons is
A. 19.4 mm B. 1.39 ´ 10–8 m C. 3.3 ´ 10–8 m D. none of these
16. If the collision time for electron scattering in copper is 10–9 s at 4 K, the
conductivity of copper in units of ohm–1 m–1 is
A. 2.4 ´ 1012 B. 9.6 ´ 107 C. 4.8 ´ 107 D. 1.04 ´ 10–8
17. The number of degenerate quantum states corresponding to 9 for the sum
of the squares of the quantum numbers nx, ny and nz in the free electron
model is
A. 32 B. 6 C. 30 D. 24
18. The probability of occupation of an energy level E, when E – EF = kT, is
given by
A. 0.73 B. 0.63 C. 0.27 D. 0.5
19. The resistivity of pure copper at 300 K is 2 ´ 10–8 ohm m and that of
a copper-3% nickel alloy at 300 K is 5 ´ 10–8 ohm m. The impurity
scattering effect for 1 per cent of nickel in copper is
A. 5 ´ 10–8 ohm m B. 3 ´ 10–8 ohm m
–8
C. 10 ohm m D. 7 ´ 10–8 ohm m
20. The error in measurement of resistance per kelvin with manganin
(temperature coefficient of resistance of 20 ´ 10–6 K–1) is
A. 0.04% B. 0.08% C. 0.00002 D. nil
21. The unit of electrical conductivity is
A. ohm m B. mho–1 m–1 C. m–3 C–2 s–1 kg–1 D. ohm–1 m–1
22. To increase the mechanical strength of an Al conductor, we can use
A. solute strengthening B. cold working
C. doping D. steel reinforcement
 Multiple Choice Questions 353

23. High conductivity aluminium should not have

A. steel rod reinforcement
B. solute atoms such as Cu, Ag and Au
C. high dislocation density
D. dissolved impurities
24. The electrical resistivity of very pure silver near 0 K is
A. not measurable B. below 10–20 ohm m
C. very low but measurable D. very high
25. Tick the materials with the approximately correct superconducting
transition temperature given in brackets
A. Sn (4 K) B. Nb3Ge (23 K)
C. Y–Ba–Cu oxide (90 K) D. Y–Ba–Cu oxide (300 K)
26. The magnetization of a superconductor is
A. 0 B. –B C. –1 D. –H
27. The atoms that are located at the body centres of the cubic unit cells of the
YBa2Cu3O7-x superconductor are
A. Y B. Ba C. Cu D. O
28. The first measured Tc in a ceramic superconductor by Bednorz and
Mueller was
A. 4 K B. 23 K C. 34 K D. 90 K
29. The following microstructural features can improve Jc in a superconductor
A. dislocation tangles B. grain boundaries
C. fine precipitates D. low test temperature
30. Explosive forming of a ceramic superconductor causes
A. Jc to increase B. Tc to increase
C. plastic deformation D. brittle fracture of the ceramic
31. Cold working of oxide superconductors is possible with
A. high hydrostatic pressure B. moderate tensile stress
C. large tensile force D. none of these
32. Switching times with a Josephson junction are in order of magnitude
A. 10–2 ns B. 0.1 ms C. 10–15 s D. 1 ms
33. The room temperature electrical resistivity (ohm-m) of the new oxide
superconductors lies around
A. 10–9 B. 10–5 C. 107 D. 1018
34. The critical current density Jc that a superconductor can carry depends on
A. T with respect to Tc B. dislocation density
C. precipitate distribution D. none of these
354 Sources for Experimental Data/Suggestions for Further Reading

Answers

1. A 2. A 3. C 4. C 5. B
6. C 7. B 8. C 9. A 10. C
11. D 12. A, B 13. C 14. C 15. B
16. A 17. C 18. C 19. C 20. C
21. D 22. D 23. B, C, D 24. C 25. A, B, C
26. D 27. A, B 28. C 29. A, B, C 30. A, C
31. A 32. A 33. B 34. A, B, C

Sources for Experimental Data

G.W.A. Dummer, Materials for Conductive and Resistive Functions, Hayden

Book Co., New York (1971).
Metals Handbook, 10th ed., Vol. 3, Special Purpose Materials, ASM
International, Materials Park, Ohio (1990), pp. 804–1024.

Suggestions for Further Reading

C. Kittel, Introduction to Solid State Physics, Wiley, New York (1976), Chaps. 6
and 12.
L.E. Murr, A.W. Hare and N.G. Eror, Introducing the Metal Matrix High
Temperature Superconductor, in Advanced Materials and Processes, ASM
International, Materials Park, Ohio (1987), Vol. 145(10), pp. 36–44.
15
CHAPTER

Semiconductors

A number of solid state devices have come into use in the last several decades,
using semiconductor crystals. The field of electronics has been revolutionized
since the discovery of the transistor in 1948. The fragile vacuum tubes have now
been largely replaced by solid state diodes and triodes. This technological
progress has been in a large measure due to improved techniques in the
preparation, purification and characterization of the raw materials needed to
make the solid state devices.
In this chapter, we shall consider briefly the energy band model for solids in
general and for pure and doped semiconductors in particular. We shall also deal
with common semiconductor materials and their fabrication. The last section is
devoted to a description of some common semiconductor devices.

Units

Quantity SI units Other units

Unit Symbol
de Broglie wavelength l metre m Å
Wave number k per metre m–1 –
Energy gap Eg joule J eV
Contact potential eV0 joule J eV
Concentration of conduction per cubic metre m–3 –
electrons or holes, ne or nh
Mobility me or mh metre squared per m2 V–1 s–1 –
volt per second
Conductivity s per ohm per metre ohm–1 m–1 –

Constants
Boltzmann’s constant k = 1.380 ´ 10–23 J K–1
Electronic charge e = 1.602 ´ 10–19 C
Electron rest mass m0 = 9.109 ´ 10–31 kg
355