Materials: engineering, science, properties, and design (4th edition)

M.F. Ashby, H.R. Shercliff and D. Cebon

Exercises with Solutions for Chapter 15

Exercise E15.1 Why do metals conduct electricity? Why do insulators not conduct electricity, at least when
cold?
Answer. Metals are electron-conductors, meaning that the charge-carriers are the electrons. The electrons
of an atom occupy discrete energy levels that, in a solid, broaden into bands of closely spaced levels. The
electrons fill the bands from the bottom up. Whether the material is a conductor or an insulator depends
on how full the bands are, and whether or not they overlap. Conductors like copper have an unfilled outer
band; there are many very closely spaced levels just above the last full one, and – when accelerated by a
field – electrons can use these levels to move freely through the material. In insulators the outermost band
with electrons in it is full and the nearest empty band is separated from it in energy by a wide band gap. An
electric field, E (volts/m), exerts a force Ee on an electron, where e is the charge it carries. In metals this
force accelerates electrons, causing them to flow through the material and giving conduction. In insulators
the force is insufficient to push electrons across the band gap, so they remain localized on the atom and are
unable to move.

Exercise E15.2 A potential difference of 4 volts is applied across a tungsten wire 200 mm long and 0.1 mm
in diameter. What current flows in the wire? How much power is dissipated in the wire? The electrical
conductivity κ e of tungsten is 1.8 x 107 S/m.

Answer. The current is given by Ohm’s law: V = i R where V is the potential difference, i is the current
and R is the resistance:
ρe L L 1 0.2
R=
A
=
κe A
=
1.8 x 10 7
×
(
π 0.5 × 10 −4 ) 2 = 1.42 Ω
The current in the wire is
V 4
i= = = 2.8 amps
R 1.42

The power dissipated in the wire is P = i 2R = 11.3 Watts

Exercise E15.3 A gold interconnect 1mm long and with a rectangular cross-section of 10 µm × 1 µm has a
potential difference of 1 mV between its ends. What is the current in the interconnect? What is the current
density (amps/m2) in the interconnect? How much power is dissipated in it? The resistivity ρ e of gold is 2.5
x 10-8 Ω.m.

Answer. The current is given by Ohm’s law: V = i R where V is the potential difference, i is the current
and R is the resistance. The resistivity of gold is ρ e = 2.5×10-8 Ω.m, so the resistance R of the
interconnect is
ρ L 1 × 10 −3
R = e = 2.5 × 10 − 8 × = 2.5 Ω
A 10 × 10 − 12

V 1 × 10 −3
The current in the interconnect is: i = = = 4 × 10 − 4 amps
R 2.5

i 4 × 10 −4
The current density in the interconnect is: = = 4 × 107 amps/m 2
A 10 x 10 −12

The power dissipated in the interconnect is P = i 2R = 4 × 10 −7 watts

1
Exercise E15.4 The gold interconnect of the previous question (length 1mm and rectangular cross-section
of 10 µm × 1 µm with a potential difference of 1 mV across its length) carries power for 10 seconds. If no
heat is lost by conduction, radiation or convection, how hot will the interconnect get? The resistivity ρ e of
gold is 2.5×10-8 Ω.m, its specific heat C p is 130 J/kg.C and its density is 19,300 kg/m3.

Answer. The resistance R of the interconnect is 2.5 Ω, the current it carries is 4 x 10-4 amps and the power
dissipated in it is P = i 2 R = 4 × 10 −7 watts. (A watt is a joule per second.) The volume of gold in the
interconnect is 1 × 10-14 m3, corresponding to a mass m = 1.93 × 10-10 kg. The energy Q required to heat
this quantity of gold through a temperature interval ∆T is
Q = m C p ∆T = 1.93 × 10 −10 × 130 × ∆T
Equating this to the electrical power dissipated in 10 second, 4 ×10 −6 joules, gives
∆T = 159oC
The heat dissipated in microcircuit interconnects can be large, requiring forced cooling.

Exercise E15.5 A 0.5mm diameter wire must carry a current i = 10 amps. The system containing the wire
will overheat if the power dissipation, P = i 2R , exceeds 5 watts/m. The table lists the resistivities ρ e of
four possible candidates for the wire. Which ones meet the design requirement?

Material ρ e , µΩ.cm
Aluminum 2.7
Copper 1.8
Nickel 9.6
Tungsten 5.7

Answer. If the resistivity of the material of the wire is ρ e , the resistance of 1 meter of wire is

ρe L 1
R= = ρe × = 5.09 × 10 6 ρ e
A (
π 2.5 × 10 )
−4 2

The power dissipated in the wire

P = i 2R = 25 × 5.09 × 10 6 × ρ e = 1.27 × 10 8 ρ e

Equating this to the power limit 5 watt/meter, gives the condition

5
ρe < = 3.93 × 10 − 8 Ω .m = 3.93 µΩ .cm
1.27 × 10 8
Thus aluminum and copper meet the design requirement but nickel and tungsten do not.

Exercise E15.6 Which metallic superconductors have a critical temperature above the boiling point of
helium? Which ceramic superconductors have a critical temperature above the boiling point of liquid
nitrogen?

Answer. Figure 15.3 shows that tantalum, lead, vanadium and niobium have critical temperatures above
the boiling point of helium. Four ceramic superconductors, HgBa2Ca2Cu3O10, Tl2Ba2Ca2Cu3O10,
Bi2Sr2Ca2Cu2O8 and YBa2Cu3O7, have critical temperatures above the boiling point of liquid nitrogen.

2
Exercise E15.7 Superconducting materials remain superconducting if they are held below their critical
temperature Tc and the magnetic field in which they sit remains below the critical field Hc . The critical
field decreases with increasing temperature T as

(
Hc = Hc ,o 1 − (T / Tc )2 )
where Hc ,o is the critical field at 0 K. A superconducting magnet design envisages a superconducting coil
cooled in liquid nitrogen, generating a magnetic field of 30 Tesla. Is it practical to use YBCO (YBa2Cu3O7) as
the superconductor? The critical temperature Tc of YBCO is 93 K and its zero-Kelvin critical field Hc ,o is 140
Tesla. Liquid nitrogen boils at 77 K.

Answer. The critical field depends on temperature as

(
Hc = Hc ,o 1 − (T / Tc )2 )
Inserting data for YBCO we find
( )
Hc = 140 1 − (77 / 93 )2 = 24.1 T
A 30 Tesla magnet is not possible without additional cooling below 77 K.

Exercise E15.8 What is a dielectric? What is meant by polarization? Define the dielectric constant. Why do
some dielectrics absorb microwave radiation more than others?

Answer. The electrons and protons of the atoms of most dielectrics are symmetrically distributed, so the
material carries no net charge or dipole moment. A field E (volts/m) exerts a force, pushing protons in the
direction of the field and electrons in the opposite direction, giving the atom a dipole moment. Two
charges ±q separated by a distance ∆x create a dipole with dipole moment, d , given by
d = q ∆x
The polarization of the material, P , is the volume-average of all the dipole moments it contains:
∑d
P=
Volume
The dielectric constant is a measure of the ease of polarization. The bigger the shift, the bigger the
dielectric constant.
The dielectric loss factor is a measure of the energy dissipated when a dielectric is placed in an oscillating
electric field. Materials with a large loss factor absorb microwave radiation more than those with a small
one.

Exercise E15.9 It is much easier to measure the electrical conductivity of a material than to measure its
thermal conductivity. Use the Weidemann-Franz law to find the thermal conductivities of
(a) an alloy that has an electrical resistivity ρe of 28 μΩ.cm.
(b) tungsten of the kind used for lamp filaments that has an electrical conductivity κ e of 1.8 × 107
S/m (a S/m is 1/Ω.m).

Answer. The Wiedemann-Franz law Figure 15.6 and equation 15.18) relates thermal conductivity λ to the
resistivity ρ e
1000
λ≈ = 10 −5 S / m ( λ in W/m.K, κ e in S/m in ρe in μΩ.cm, ). Using this equation gives:
ρe
(a) the alloy has a predicted thermal conductivity of 36 W/m.K.
(b) the tungsten has a predicted thermal conductivity of 180 W/m.K

3
Exercise E15.10 The metal zinc has free electron concentration per unit volume, nv of 1.3×1029 / m3 and an
electron mobility µe of 8×10-4 m2/V.s. The charge carried by an electron, e , is 1.6×10-19 Coulomb. Based
on this information, what is the electrical conductivity of zinc? Handbooks list the measured resistivity of
zinc as 5.9 μΩ.cm. Is this consistent with your calculation? (Watch the units.)
Answer. The electrical conductivity κ e depends on the number, charge and mobility of charge carriers
(equation 15.20 of the text) as: κ e = nv e µ e

Inserting the data from the question gives: κ e =1.66×107 S/m.

This translates into a resistivity of 6.0 μΩ.cm, and thus is within 2% of the Handbook value, 5.9 μΩ.cm.

Exercise E15.11 The resistivity of pure silver is ρ e = 1.59 µΩ .cm . Each silver atom provides one free
electron. The atomic volume of silver is 1.71×10-29 m3. What is the electron mobility in silver?
Answer. If each atom contributes one electron the electron concentration in copper is
1
nv = = 5.85 × 10 28 per m 3
Atomic volume
The resistivity ρ e is related to this by equation 15.20 of the text:
1 1
ρe = = where e = 1.6 × 10 −19 Coulombs is the charge on an electron.
κ e nv e µe
Converting the resistivity from µΩ .cm to Ω .m by multiplying by 10-8 gives the electron mobility in copper
1
as: µ e = = 0.0067 m2/V.s
nv e ρ e

Exercise E15.12 Estimate the drift velocity of free electrons when a potential difference of 3 volts is applied
between the ends of a copper wire 100 mm long. The electron mobility in copper is µe = 0.0032 m2/V.s

Answer. The potential difference of 3 volts across 100mm of wire creates a field E = 3 / 0.1 = 30 V/m. The
drift velocity is: vd = µ e E = 0.0032× 30 = 0.096 m/s

Exercise E15.13 The resistivity of brass with 60 atom% copper and 40 atom% of zinc is ρ e = 6.8 µΩ .cm .
Each copper atom contributes one free electron. Each zinc atom contributes two. The atomic volumes of
copper and zinc are almost the same (assume them to be equal) and the electron concentration in copper is
nv = 8.45×1028/m3 . What is the electron mobility µe in brass? How does it compare with that of pure
copper ( µe = 0.0032 m2/V.s)?

Answer. If each copper atom contributes one free electron and each zinc atom contributes two, the
average number of free electrons per atom is (0.6 + 2×0.4) = 1.4, so the electron concentration in the brass
is
nv = 1.4 × 8.45 × 10 28 = 1.18 × 10 29
The resistivity ρ e is related to this electron concentration (equation 15.20 of the text) by
1 1
ρe = = where e = 1.6 × 10 −19 Coulombs is the charge on an electron.
κe nv e µe
Solving for electron mobility µe and converting the resistivity into Ω.m gives, for brass,
1
µe = = 0.0008 m2/V.s
nv e ρ e
This is about one quarter of the mobility of electrons in pure copper, suggesting that the mean free path of
electrons in brass is much less than that in copper.

4
Exercise E15.14 Graphene is a semi-conductor with remarkable electrical properties. It is reported that the
free-electron mobility µe in graphene is 1.5 m2/V.s and that the effective free electron concentration per
unit volume, nv , is 5×1025/m3. If these values are accepted, what would you expect the resistivity of
graphene to be?

Answer. The conductivity of an electronic conductor is κ e = nv e µe where e = 1.6 × 10 −19 Coulombs is the
charge on an electron. Inserting the data in the question gives
κ e = 5 × 10 25 × 1.6 × 10 −19 × 1.5 = 1.2 × 107 S / m
1
Thus the resistivity is ρ e = = 8.3 × 10 − 8 Ω .m = 8.3 µΩ .cm . This is about the same as a metal.
κe

Exercise E15.15 A power line is to carry 5 kA at 11 kV using pylons 400 m apart. The dip d , in a wire of
weight ml per unit length strung between pylons L apart at a tension T is given by d = L2 ml / 8T . The
maximum tensile stress allowed is 0.8 of the yield stress, σ y . If the maximum allowable dip is 6 m, which
of the materials in the table could be used?

Material Yield stress, σ y MPa Density. ρ kg/m3

Aluminum 125 2700
Copper 120 8900
Steel 380 7800

Answer. The mass per unit length of cable is ml = ρ g A . The tension T = σ A so the required strength
given by

L2 ρ gA L2 ρ g σ L2 ρ g
d= or, on inverting σ = or =
8T 8d σ y 8 dσ y

Evaluating this for the materials of the table gives the results listed below.
Aluminum Copper Steel
Density , ρ (kg/m3 ) 2700 8900 7800
Strength, σ y (MPa) 125 120 380
σ /σy 0.71 2.43 0.67

Thus aluminum or steel meet the criterion that σ / σ y ≤ 0.8 , but not copper.

Exercise E15.16 A material is required for a transmission line that gives the lowest full-life cost over a
twenty-year period. The total cost is the sum of the material cost and the cost of the power dissipated in
Joule heating. The cost of electricity CE is 6×10-3 $/MJ. Material prices are listed in the Table below.
Derive an expression for the total cost per meter of cable in terms of the cross-sectional area A (which is a
free parameter), the material and electrical costs and the material parameters. Show that the minimum
cost occurs when the two contributions to the cost are equal. Hence derive a performance index for the
material and decide on the best of the materials in the Table.

Material Electrical resistivity Density Price

ρe , Ωm ρ kg/m3 C m , $/kg
Aluminum 1.7×10-8 2700 1.6
Copper 1.5×10 -8 8900 5.2
Steel 55×10 -8 7800 0.5

5
Answer. If the material cost is C m /kg the cost/ meter is C m ρ A . The resistance/meter is R = ρe / A . The
power dissipation = i 2 ρ e / A . The cost of the dissipated power is C = C e i 2 ρ e t / A where t is the time
(20 years). Thus the total cost
Ce i 2 ρe t
Ctot = C m ρ A +
A
Minimizing Ctot with respect to A gives the optimum value

(
C~tot = 2 C m ρ ρ e C e i 2t )1 / 2
A suitable performance index is thus C m ρ ρ e . This group has the value 7.3×10-5 for aluminum, 6.9x10-4 for
copper and 2.2×10-3 for steel. Aluminum offers the lowest cost C~ .tot

Exercise E15.17 In the discussion of conductors, a 50-50 mix of copper and steel strands was suggested for
transmission cables. Using the values in the table for resistivity and strength, calculate the effective values
for the cable, assuming a rule of mixtures for both strength and resistivity. Plot this on a copy of the
σ y − ρ e chart for copper alloys of Figure 15.10 to explore its performance.

Material Strength, σ y (MPa) Resistivity, ρe ( µΩ.cm )

High strength steel, cold drawn 1700 22
High conductivity copper, cold drawn 300 1.7

Answer. Using a rule of mixtures, the strength of

the cable is 1000 MPa and its resistivity is 11.9 The cable
µΩ.cm. The result is plotted on the chart. The
composite cable is stronger than all but the
strongest copper alloys – the copper-beryllium
alloys (very expensive) – and has a resistivity
comparable with that of brass.

Exercise E15.18 Roughly 50% of all cork that is harvested in Portugal ends up as cork dust, a worthless by-
product. As a materials expert, you are approached by an entrepreneur who has the idea of making useful
products out of cork dust by compacting it and heating it, using microwave heating. The loss factor
L = ε r tanδ of cork is 0.21. The entrepreneur reckons he needs a power density P of at least 2 kW per m3
for the process to work. If the maximum field E is limited to 102 V/m, what frequency f of microwaves
will be needed?

Answer. When a dielectric material is placed in a cyclic electric field of amplitude E and frequency f ,
power P is dissipated and the field is correspondingly attenuated. The power dissipated per unit volume, in
W/m3, is

P ≈ f E 2 ε tanδ = f E 2 ε o ε r tanδ
where ε r is the dielectric constant of the material and tan δ is its loss tangent (equation (15.17) of the
text). This power appears as heat and is generated uniformly through the volume of the material.
6
The higher the frequency or the field strength and the greater the loss factor ε r tanδ the greater is the
heating and energy loss. Inserting the data into this equation and setting the permittivity of free space ε o
= 8.85 × 10−12 F/m, we find that the frequency required to meet the design requirements is 1.1× 1011/s.

Exercise E15.19 Derive the expression (equation (15.14))

1
Max energy density = ε r ε o E b2
2
for the maximum limiting electrical energy density that can be stored in a capacitor with plate area A
separated by a dielectric of thickness t . Here ε r is the dielectric constant of the dielectric, ε o is the
permittivity of free space and E b is the field at which the dielectric breaks down.

1 2
Answer. The energy in a dielectric film of thickness t and area A is C V . The capacitance is
2
A
C = εr εo
t
and the electric field is E = V / t . Thus the energy stored in the dielectric is
1 1 A
C V 2 = ε r ε o (E t )2
2 2 t
The volume of the dielectric is A t and the maximum permissible field is the breakdown field Eb .

Thus the upper limiting energy density is

1 CV 2 1
= ε r ε o E b2
2 At 2

Exercise E15.20 You are asked to suggest a dielectric material for a capacitor with the highest possible
energy density. What material would you suggest? Use Figure 15.7 to find out.

Answer. The axes of Figure 15.7 are dielectric constant ε r and breakdown potential Eb . The dielectrics
capable of storing the most electrical energy are those with the greatest values of the product ε r E b2
(equation 15.14 of the text) – this product appears as a grid of diagonal contours on Figure 15.7. The
materials with the highest value of this product are ferroelectrics lead zirconium titanate (PZT) and lead
lanthanum zirconium titanate.

Exercise E15.21 The electron mobility µ e in undoped germanium at 300 K is 0.36 m2/V.s and that of holes
µ h is 0.19 m2/V.s. The carrier density nv is 2.3×1019 /m3. What is the conductivity of germanium at 300 K?
What is its resistivity?

Answer. The conductivity is

κ e = nv e (µ e + µ h ) = 2.3 × 10 19 × 1.6 × 10 −19 × ( 0.36 + 0.19 ) = 2.0 S / m
where e = 1.6 × 10−19 coulombs is the charge on an electron. The conductivity of germanium at 300 K is 2
S/m. The resistivity is the reciprocal of the conductivity, so the resistivity of germanium is 0.5 Ω .m or 0.5x103
µΩ .cm .

7
Exercise E15.22 The atomic volume of silicon is 2.0×10-29 m3 and the free electron density nv at 300K is
1.5×1016 /m3. What fraction of silicon atoms have provided a conduction electron?

1
Answer. There are natoms = = 5 × 10 28 silicon atoms per cubic meter. The fraction of
Atomic volume
silicon atoms providing a free electron is then

1.5 × 10 16
= 3 × 10 −13
5 × 10 28
This compares with a fraction of 1 for conducting electrons in copper and silver.

Exercise E15.23 You are asked to design a heat-sensing switch to turn the lights off in the garage when no
one is moving around in it. What principle would you base it on? What materials would you choose for the
sensor?

Answer. The principle is that of pyroelectricity – polarization caused by change of temperature. Some
materials have a permanent dipole moment because their positive and negative ions balance electrically but
are slightly out of line with each other. A thin disk of one of these, appropriately cut, has a dipole moment
of its own. The dipole moment per unit volume of the material is the spontaneous polarization Ps . This net
dipole moment exists in the absence of an applied electric field and is equivalent to a layer of bound charge
on each surface. Nearby free charges such as electrons or ions are attracted to the surfaces, neutralizing the
charge. If the temperature of the sample is constant, then so is Ps and no current flows through a circuit
linking the faces of the disk. An increase in temperature, however, causes expansion and that changes the
net dipole moment and the polarization. Redistribution of free charges to compensate for the change in
bound charge induces a pyroelectric current in the circuit. Cooling rather than heating reverses the sign of
the current. The pyroelectric current only flows while the temperature is changing – it is this effect that can
be used for motion sensors.

Pyroelectric materials include minerals such as tourmaline, ceramics such as barium titanate, and polymers
such as polyvinylidene fluoride.