Digitized by the Internet Archive

In 2022 with funding from

Kahle/Austin Foundation

https://archive.org/details/physicstextoookf0000dunc_s8h4
 PHYSICS
A Textbook for Advanced
Level Students
Two-volume version of this book:
Advanced Physics Volume I: Materials and Mechanics
Advanced Physics Volume II: Fields, Waves and Atoms

By the same author:

Electronics and Nuclear Physics

Exploring Physics 1-3
Physics for Today and Tomorrow
Electronics for Today and Tomorrow
Science for Today and Tomorrow (with M. A. Atherton and D. G. Mackean)
Adventures with Electronics
Adventures with Microelectronics
Adventures with Digital Electronics
Adventures with Physics
Success in Electronics
Success in Physics
GCSE Physics
Physics for the Caribbean (with D. Onac)
Basic Skills: Electronics
GCSE Electronics

Cover illustration:

Moiré interference pattern of identical radiating images.

(Paul Brierley)
 PHYSICS
A Textbook for Advanced
Level Students

2nd edition

Tom Duncan Bsc

Formerly Senior Lecturer in Education
University of Liverpool

John Murray
© T. Duncan 1982, 1987

First published 1982 by

John Murray (Publishers) Ltd
50 Albemarle Street, London W1X 4BD

Reprinted 1983, 1984

Second edition 1987
Reprinted 1987, 1988, 1989, 1990 (twice), 1993

All rights reserved

Unauthorized duplication
contravenes applicable laws

Printed in Hong Kong

by Colorcraft Ltd.

British Library Cataloguing in Publication Data

Duncan, Tom
Physics: a textbook for Advanced Level
students — 2nd ed.
1. Physics
530 QC23
ISBN 0-7195-4336-3
Preface

This second edition incorporates recent ‘A’ level sylla- As in the previous edition, the treatment permits con-
bus revisions; it meets the requirements of Boards who siderable flexibility in the order in which the chapters
examine the topics covered, as core material or as are followed. On the assumption that the book will be
options. used selectively, certain topics (e.g. aspects of geometri-
Changes have been made throughout the text, most cal optics, surface tension) which are no longer examined
extensively in chapter 23, Electronics, and three new at this level by most Boards are retained for
chapters have been added. The first, chapter 1, deals those students who require them.
with Materials and their uses (especially in structures It has been ensured for this printing that full account
such as bridges) and provides a practically orientated is taken of the Hong Kong Advanced Level Physics
introduction to chapters 2 and 3 on the structure and 1992/93 syllabus. Questions taken from _ the
mechanical properties of materials. Thesecond new H.K.A.L.E. 1986-1991 appear at ends of chapters and
chapter 10, is concerned with Energy and it uses, a in the banks of objective-type revision questions.
topic of considerable current and future importance, The influence of the Nuffield Advanced Physics
and discusses sources of energy (finite and renewable), Course will be evident and is again acknowledged.
energy conversion, energy consumption and losses, and Physics: A Textbook for Advanced Level Students is
pollution. The third, chapter 24 on Telecommunica- also available in two volumes. Volume I: Materials and
tions, is relevant to this age of information technology Mechanics contains chapters 1 to 12; Volume II: Fields,
and includes sections on the representation and Waves and Atoms contains chapters 13 to 25. These are
transmission of information, radio systems, communica- currently in their third edition.
tion satellites, optical fibre systems and lasers.
Acknowledgements

I am much indebted to Dr J. W Warren and to Mr For permission to use questions from their examina-
J. Dawber, both of whom read the original manuscript tions, grateful acknowledgement is made to the various
and made many helpful comments and suggestions. I examining Boards, indicated by the following abbrevia-
must also thank Professor J. Stringer and Dr J. C. Gib- tions: A. E. B. (Associated Examining Board); C. (Cam-
bings. The former very kindly commented on chapters bridge Local Examination Syndicate); H.K.A.L.E.
2 and 3, and useful discussions were held with the latter (Hong Kong Advanced Level Examinations, by per-
on chapter 12. My sincere thanks are also due to Mr mission of the Hong Kong Examinations Authority);
B. Baker, who constructed and tested the objective-type J.M.B. (Joint Matriculation Board); L. (Univerisity
questions, and to my son-in-law and daughter Drs of London); O.L.E. (Oxford Local Examinations);
B. L. N. and H. M. Kennett, who checked the answers O. and C. (Oxford and Cambridge Schools Examination
to the numerical problems. Thanks are also due to my Board); §S. (Southern Universities Joint Board);
wife for preparing the typescript. W. (Welsh Joint Education Committee).
a:.D,

Photo credits
Dynamics Ltd*; 11.12, L. H. Newman (photo by W. J. C.
The publishers are grateful to the following, who have kindly Murray)*; 11.13, photograph by Oskar Kreisel from Focal
permitted the reproduction of copyright photographs: Encyclopedia of Photography, Focal Press, London and Bos-
Figs 1.7, Autocar/QPL; 1.16, Shell Chemicals UK Ltd.; ton; 12.9, Dr J. C. Gibbings, Department of Mechanical Engi-
2.1a,b, Cambridge Instruments Ltd; 2.1c, The Plessey Co. ple; neering, University of Liverpool, and the Editor, Physics
2.1d, Josiah Wedgwood & Sons Ltd; 2.7, Dr R. T. Southin*; Bulletin - © J. C. Gibbings; 13.3, Aerofilms; 14.37, 25.21,
2.9, Professor Brian Ralph, Institute of Materials, University 25.30a,b, 25.31, United Kingdom Atomic Energy Authority;
College, Cardiff; 2.10, J. W. Martin; 2.16b,c,d,e, 3.13, The 15.23a,b, Leybold-Heraeus; 15.306, National Physical Labora-
Royal Society (Proceedings, 1947, A, 190); 2.19a, Bell & tory*; 15.35c, Walden Precision Apparatus Ltd; 16.2b, The
Hyman Ltd (from An Approach to Modern Physics by E. Royal Institution; 16.145, 23.9, 23.64a,d, Mullard Ltd; 16.16,
Andrade); 2.19b, Dr C. Henderson; 2.20d, 4.18, 14.25, 18.4, Trustees of the Science Museum (London); 16.23a, GEC Tur-
23.27a, 23.45, 23.72a, Unilab Ltd; 3.1, 3.2, Avery-Denison bine Generators Ltd, Stafford; 16.34a, GEC Power Trans-
Ltd, courtesy of Messrs Seamless Tubes, Corby; 3.12, V. A. formers Ltd; 16.34b, Electricity Council; 16.35, Central
Phillips and J. A. Hugo*; 3.14, Joseph V. Laukonis, General Electricity Generating Board; 16.48, IBM; 16.51, Les éditions
Motors Research Laboratories; 3.166, 6.37a,b, 6.76a, Penguin de physique (J. de Physique 1951 Vol. 12, p. 308); 18.3a, A.
Books Ltd (from Revolution in Optics by S. Tolansky); 3.21, M. Lock & Co. Ltd*; 18.86, Griffin & George; 18.10b, 18.155,
British Engine, The Technical Services Division of Royal Insur- 18.20a,b,c, W. Llowarch, Ripple Tank Studies of Wave Motion
ance; 3.23, Vosper Thornycroft (UK) Ltd; 3.24a, A.E.R.E. (Clarendon Press, Oxford) © Oxford University Press 1961;
Harwell; 3.24c, Fulmer Research Institute Ltd; 3.27a,b, Pro- 18.19, Racal-Decca Marine Navigation Ltd; 18.21a,b, Educa-
fessor E. H. Andrews, Queen Mary College; 3.28, Eidenbenz tion Development Center, Inc., Newton, Ma.; 18.34, Racal
and Eglin, used in Science, ed. J. Bronowski (Aldus Books); Marine Radar Ltd; 19.2, Charing Cross and Westminster Medi-
4.20, THORN EMI Instruments Ltd; 4.22, 10.7, Central Elec- cal School; 19.4, 19.17, South Bank Board; 20.1a, 25.7a,b,
tricity Generating Board; 4.24, 9.25, British Crown Copyright/ Philip Harris Ltd; 20.35, D. G. A. Dyson*; 20.12, Bausch
RAE photograph; 4.25, 16.14, 23.9, 23.64a,d, Mullard Ltd; & Lomb Optical Company Ltd*; 20.19, C. B. Daish*; 20.21,
4.30, LLOYD INSTRUMENTS - Educational Division; 5.4, 20.24a, Richards, Sears, Wehr & Zemansky, Modern College
British Steel; 6.88, Palomar Observatory; 6.99, Times News- Physics, © 1962, Addison-Wesley Publishing Co., Inc., Read-
papers Ltd*; 7.16, 13.8a,b, PSSC PHYSICS, 2nd ed., 1965, ing, Massachusetts, p. 695, Fig. 35.21 and p. 687, Fig. 35.16;
D.C. Heath & Company with Education Development Center, 20.53a,b. Open University, from $100 Unit 28, © 1974 The
Inc., Newton, Ma.; 7.24a,b, 7.33a,b, Jim Jardine, Moray Open University Press; 20.57, Barnes Engineering Company*;
House College of Education, Edinburgh; 7.34b, 25.32a,b, 20.67, 24.17, 24.18, TeleFocus: a British Telecom photograph;
Lady Blackett, photographs by Lord Blackett; 8.9, Popperfoto; 21.8, BBC Publications, from Science Extra, Spring 1969;
8.29, C.E.F., Bernard PARIS; 9.1a,b, Special Collections, 21.20, Rolls-Royce Ltd; 22.11, 23.18c, RS Components
University of Washington Libraries, photos by F. B. Farquhar- Limited; 22.17a,b, Nuffield Advanced Physics (1971) Teachers’
son; 10.8a, Pilkington Solarproducts Ltd*; 10.9, Service Scien- Guide, Unit I and Students’ Book, Unit 1, Longman Group
tifique, French Embassy*; 10.11a,b, Daily Telegraph Colour Ltd; 22.22, Kratos Analytical—A division of Spectros; 22.30a,
Library/Julian Wasser; 10.12, Wind Energy Group; 10.16, Professor H. Hill; 22.30b, G. Mdllenstedt and H. Duker;
Hamworthy Engineering Ltd; 10.19, J. Watton, Camborne 23.3b,c, © 1987, TEKTRONIX Inc.; 23.64e,f, Department
School of Mines Geothermal Project; 11.1, Water Authorities of Industry*; 23.80, Thandar Electronics Ltd; 24.20a, 24.21b,
Association; 11.35, Austin Rover Group Ltd; 11.4, Pressure STC PEC; 25. 105925401, (Cy De Re wilson: 925.22, United

vi
Kingdom Atomic Energy Authority, courtesy of the National Figs 15.16, 19.30a,b, Philip Harris Ltd; 18.35, A. M. Lock
Hospital; 25.24, AERE Harwell; A16.1, A. G. Gaydon’. & Co. Ltd; 18.28, Unilab Ltd; 25.12, University of Liverpool.
Page 1, Department of Physics, Royal Holloway and Bedford
New College, University of London; page 141, British Aero- * Every effort has been made to contact these copyright
space; page 247, The MIT Museum; page 365, W. Llowarch, holders, who have given their permission for reproduction in
Ripple Tank Studies of Wave Motion (Clarendon Press, previous editions; we apologize for any breach and will be
Oxford), © Oxford University Press, 1961; page 451, Mullard pleased to rectify this in future impressions.
Ltd.
The copyright holders of the following photographs we have
The following photographs, taken by the author, show equip- been unable to trace; we will be pleased to rectify these omis-
ment or experiments from these sources: sions in future impressions: Figs 2.20b, 3.19.

Vii
Contents

Preface
Acknowledgements Vl

VOLUME I VOLUME II

MATERIALS FIELDS
Materials and their uses 13 Electric fields 249
Structure of materials 14 Capacitors 266
Mechanical properties 15 Magnetic fields 287
Electrical properties 16 Electromagnetic induction 308
Thermal properties 17 Alternating current 341
hk
=
An
WN Optical properties Objective-type revision questions 361
Objective-type revision questions 137
WAVES
MECHANICS 18 Wave motion 367
7 Statics and dynamics 143 19 Sound 390
8 Circular motion and gravitation 165 20 Physical optics 414
9 Mechanical oscillations 185 Objective-type revision questions 449
10 Energy and its uses 198
11 .Fluids at rest 217 ATOMS
12 Fluids in motion 232 21 Kinetic theory: thermodynamics 453
Objective-type revision questions 244 22 Atomic physics 483
23 Electronics . S13
24 Telecommunications 559
25 Nuclear physics 573
Objective-type revision questions, 597

Appendices App 1
Answers I
Index Vil

Vili
 VOLUME |

Materials and Mechanics

Contents

Ss > \ { ms )

Fin
Ant iieloeGgpaiers

vouinn) eaineriseM obit . hill

MATERIALS
1 Materials and their uses 5 4 Electrical properties ad
2 Structure of materials 11 5 Thermal properties 73
3 Mechanical properties 27 6 Optical properties 92

eZ
Film interference on the surface of diamond
 an

eJAIRSTAM
oe a ——
7 rolhreqo'sn Inobetoel = # . eon Wott ban elotateld F
esitweitg trav 2 FE M debianteneta X
ertrqoig inti @ UE a asirouang tenivedald ©
Os 7 = ot a ta a oh

bi fait Te soe! ae-onte nes savorsbroal ellie

ie
1 Materials and their uses

Mankind’s use of materials Polymers

Materials in tension and compression Other materials
Metals and alloys Beams
Timber Bridges
Stone, bricks and concrete Some other structures

Mankind’s use of materials (d) Modern era. The twentieth century has seen the
It has been said that a scientific discovery is incomplete arrival of plastics (p. 5) and composite materials (p.
and immature until the technologist has found a prac- 36) which have opened up a whole range of possibilities.
tical application-for it and improved mankind’s lot. One
of the essential requirements for any technological
Materials in tension and compression
advance is the availability of the right materials. The
importance of this is shown by the use of names such Different materials are used for different jobs, the
as Stone Age, Bronze Age and Iron Age for successive choice depending, among other things, on the properties
cultures in ancient times. of the material. There are good reasons why concrete
is used for constructing large buildings, wood for furni-
(a) Stone Age. In this period, dating from the earliest ture, glass for windows, aluminium for saucepans, plas-
times recorded up to about 2500 Bc, tools and weapons tics for washing-up bowls, cotton and nylon for clothes,
were made of stone. Clay was fired to make pottery, and rubber for elastic bands.
while the weaving of plant and animal fibres provided We shall be concerned in particular with the use of
cloth, fishing nets and baskets. materials in structures such as buildings and bridges,
When agriculture was developed, the settled exis- and in these cases it is their mechanical properties that
tence required for tending crops and animals all the are important. For example, it is essential to know how
year round encouraged the building of permanent they behave under tension and compression. Stretching
houses of wood and stone. Villages and towns’grew up, a material puts it in tension, Fig. 1.1a, while squeezing
requiring roads, drains, bridges and aqueducts. The it puts it in compression, Fig. 1.15. A material which
resulting wealth of some communities led to envy among is strong in tension can be weak in compression, and
others and the need for town dwellers to construct forti- vice versa.

—
fications for protection and to develop weapons tech-
nology. Bar of material

(b) Bronze Age. This era began with the discovery

around 2500 8c in eastern Europe that copper became
harder and tougher when alloyed with tin to make
(a) Pulls cause tension (b) Pushes cause
bronze. The consequent advances in technology were, compression
however, small compared with those of the Iron Age.
Fig. 1.1
(c) Iron Age. Although iron farm implements were
used in China for centuries before, it was not until about The properties of some common types of material
1000 Bc or so that iron came into widespread use in other will now be considered. The mechanical properties are
countries. Iron is one of the commonest metals in the discussed in detail in chapter 3.
earth’s crust. It is extracted from its ore (iron-bearing
rock) by smelting. Metals and alloys
More recently, in the nineteenth century, steel (an
iron alloy) became the dominant material for making (a) Iron and steel. Pure iron is seldom used; it is
tools, utensils, machinery, bridges, ships, weapons, cars usually alloyed with other substances to form steel. Mild
and many other items. steel is iron containing a very small proportion of carbon.

3
4 VOLUME I Materials and Mechanics

It is strong in both tension and compression and, being Tree trunk Plank of wood

cheap, is used in large quantities for mass-produced

goods like cars, cookers and refrigerators. In the con-
struction industry scaffolding, girders, bridges, power
pylons and the reinforcing for concrete are made of it.
Two disadvantages of mild steel are first that it is
a heavy metal and second that it rusts. To counter rust-
ing, an alternative to painting is to coat it with another
metal which resists corrosion, such as tin, forming tin-
plate. Galvanized steel is covered by a thin layer of zinc;
corrugated sheets of this are used as roofing for sheds. (a) (b)
Chromium-plated steel is protected chiefly by a layer Fig. 1.2
of nickel, on top of which a very thin layer of chromium
(a hard, shiny metal) is added by electroplating. New wood contains a great deal of moisture, most
Stainless steel contains large proportions of chromium of which must be removed by seasoning before it is used.
and nickel. It is more expensive and difficult to work This is generally done naturally by stacking the freshly
than mild steel but it is much stronger and harder and cut planks with spacers between, to allow air to circulate
very resistant to corrosion. and dry them out slowly. The operation may take any-
Titanium steel has a very high melting-point and is thing from a few weeks to several years. If a piece of
used to make parts of jet engines, rockets, supersonic wood is not properly seasoned it will gain or lose water
aircraft and space-shuttle nose-cones. unevenly when it is very damp or very dry. Gain of
(b) Aluminium and duralumin. Aluminium is the water produces expansion, loss of water causes shrink-
most widely used metal after iron but is much more ing and the result is warping.
expensive to extract from its ore (bauxite). Its density (a) Plywood. Thin sheets of wood need less time to
is one-third that of iron and the thin, tough layer of season than thick planks. Plywood is made by glueing
oxide which forms on the surface when exposed to the thin, seasoned sheets together with the grains of alter-
air makes it very resistant to atmospheric corrosion. nate sheets at right angles to each other, Fig. 1.34. An
The pure metal tends to be weak and brittle. odd number of sheets is always used (to give, for exam-
Duralumin is made by alloying aluminium with small ple, 3-ply or 5-ply) so that if a crack passes between
amounts of copper, manganese and magnesium. The the grains of one sheet-it meets the next sheet across
tensile strength is then as great as that of mild steel the grain, Fig. 1.3b, and does not spread. For this rea-
and this, combined with its low density, makes it highly son, plywood is stronger than a piece of solid wood
suitable for aircraft bodies. of the same thickness. Since it consists of sheets it is
called a laminate.

Yi{py yyy //
Wp ftp
jiff| /
fi
Timber VINE 1 |
Vay iS 4
7)

There are two main types of wood. Softwoods like pine,

besides being soft, are usually light in both weight and
colour. They are used for general carpentry to make
doors, window frames, floors and roof trusses in house-
building. Hardwoods like oak and teak are stiff (see
p. 27) and strong and are suitable for making furniture. (a) (b)
A tree grows from the centre outwards, a ring of wood
being produced in the trunk each year, Fig. 1.2a. These
annual rings are the grain marks that are seen when
the trunk is cut into long planks, Fig. 1.2b. In hardwoods (b) Blockboard is a sandwich made by filling the
these marks are closer than in softwoods, which explains space between two thin sheets of wood with strips of
why the former are stiffer and stronger. In tension, the solid wood, as in Fig. 1.4. The grain on the outside
strength along the grain is greater than across it because sheets goes the same way. Like plywood, blockboard
wood consists of long tube-like fibres running up and should not warp and can be fixed down firmly (e.g. as
down the tree trunk. Wood is less strong in compression. a worktop) with no fear of it moving.
 MATERIALS AND THEIR USES 5
Thin sheet (c) Cement and mortar. Cement is a cream-coloured
of wood
powder, made by heating a mixture of clay and lime
to a high temperature. If mixed with sand and water,
Solid wood
strip
cement becomes a thick paste called mortar. Mortar
is used to hold bricks or stones together since it becomes
S Thin sheet a hard, stone-like material when it dries.
of wood

Fig. 1.4 (d) Concrete is now used more than any other mater-
ial for building and construction work. It is made by
(c) Chipboard is made from wood particles and resin, mixing cement, sand and gravel (called ‘aggregate’) with
water. A typical mix is 1 part cement, 2 parts sand and
Fig. 1.5a. It can be sawn like wood and, being reason-
4 parts aggregate, but this is varied for different pur-
ably strong though heavier than solid wood, it can be
poses. If allowed to dry in a mould, any shape can be
used as floor boards or in other situations where appear-
obtained and it sets as a hard, white stone.
ance does not matter.
Concrete weathers well and is strong in compression
but weak in tension due to the large number of small
cracks it inevitably contains. As a result, it is brittle
and unsuitable when large tensile strength is required.
In reinforced concrete the strength of concrete in ten-
sion is improved by inserting wires or rods of steel
(a) Wood particles (b) through the wet concrete, Fig. 1.6. As the concrete dries
and resin it sticks to the steel, giving a combination which is strong
Fig. 1.5 in both compression and tension.

For making furniture, shelves, etc., it is sold with

_ Reinforcing
a thin, more attractive sheet (a veneer) already on one ~ steel rod
or both surfaces, Fig. 1.55. Veneers of white plastic,
of plastic with wood grain or colour effects, or of wood ~~—Concrete
itself are used.
Fig. 1.6
_ Stone, bricks and concrete
A variety of materials is used to construct houses, build- In prestressed concrete even greater tensile strength
ings, bridges, roads and dams. Cost, climate and availa- is obtained, as explained in chapter 3 (p. 35). In light-
bility are often factors that have to be considered when weight concrete cinders are used as the aggregate.
deciding which to use.
(a) Stone. Deposits of stone are found in many parts Polymers
of the world. They occur as granite which is hard and
The properties of some of the commoner man-made
long-lasting, marble which is hard and attractive but
polymers are outlined here; (a) to (f) are thermoplastics,
does not last as long, and sandstone which is soft, easy
(g) and (h) are thermosets (see chapter 2, pp. 24-5,
to work and fairly long-lasting.
where the molecular structure of polymers is discussed).
Stone is strong in compression but weak in tension.
The mechanical properties of rubber, a natural polymer,
In cities where there is atmospheric pollution, stone
are considered in detail in chapter 3 (p. 41).
buildings need cleaning periodically if they are to retain
their appearance. (a) Polythene is tough (i.e. not brittle) but flexible,
and resistant to water and most solvents. It can be rolled
(b) Bricks are a cheap alternative to stone and have
into thin sheets and moulded into complicated shapes.
a convenient size. They are made by mixing clay with
It is a very good electrical insulator.
water and are then moulded into shape before being
baked in ovens at a high temperature. The colour and (b) Polystyrene is more brittle than some plastics but
hardness of the brick produced depends on the clay used its stiffness is taken advantage of for making small con-
and the baking temperature. Bricks, like stone, are tainers and toys. Expanded polystyrene is a solid foam
weak in tension and strong in compression. containing a large number of air bubbles. Its very low
6 VOLUME I Materials and Mechanics

density and ease of moulding to almost any shape make Other materials
it a good packaging material. It is also a good heat insula-
(a) Fibre-reinforced materials. The properties of
tor.
composite materials using fibres of glass or carbon in
(c) PVC (polyvinyl chloride) is strong, tough, flexible a plastic resin (GRP and CFRP) will be considered in
and waterproof, which makes it suitable for protective chapter 3 (pp. 36-9).
sheeting and floor coverings. Being a good electrical
(b) Laminated glass, often called ‘bullet-proof’ glass,
insulator it is used to cover electric cables.
is even stronger than toughened (prestressed) glass (p.
(d) Perspex is stiff and transparent but not so hard 35). It is made by fixing together layers of toughened
and brittle as glass for which it is sometimes used as glass with a transparent adhesive. More layers give
a substitute. It is easily cut and drilled, but scratches. greater strength. It is used for aircraft as well as car
windscreens. If a crack gets through one layer of glass,
(e) PTFE (polytetrafluorethylene) or Teflon has a
it gets ‘blunted’ on meeting a layer of adhesive and is
much higher melting-point than most plastics and also
unable to penetrate the next layer, Fig. 1.7a. So when
has ‘non-stick’ properties. These make it useful for coat-
struck, it cracks but does not break into lots of small
ing the insides of saucepans and other cooking con-
fragments; Fig. 1.7b shows the effect.
tainers and for making bearings that do not need
lubricating. (c) Ceramics are made by mixing clay, fine sand and
water into a paste which is shaped as required and fired
(f) Nylon fibres are used to make strong ropes and
at a high temperature. Very fine clay (e.g. china clay
hard-wearing fabrics. Being water-resistant, clothing
or kaolin) when fired at a sufficiently high temperature
made from it dries quickly and requires no ironing.
forms porcelain. Ordinary clay is unable to withstand
However, nylon shirts and blouses often feel damp to
such high temperatures and is used to make earthenware
wear in hot weather because they do not absorb sweat.
or pottery. Porcelain is usually ‘glazed’ by adding a layer
Terylene is another fibre used in the textile industry.
of glass.
(g) Bakelite is hard and brittle, but is much streng-
Crack ‘blunted’
thened by the addition of, for example, sawdust. Its
cheapness, low density and resistance to corrosion have Glass
made it popular for electrical fittings.
Glass
(h) Formica and melamine are two other thermosets
with hard, smooth surfaces which make them suitable
veneers for table tops and other surfaces. (a)
 MATERIALS AND THEIR USES 7
Beams (b) Trussed beams. A simple beam can be streng-
Beams or ‘girders’ form parts of larger structures such thened if a truss is joined to it as in Fig. 1.10. If, for
as bridges. example, the structure is a bridge, the weight of a car
on it makes CBD bend down. AB moves down too,
(a) Simple beams. You can see what happens when but AC and AD hold it back so CBD does not bend
a beam is loaded by drawing lines on a piece of foam so much. Loading the trussed beam therefore tends to
rubber, Fig. 1.8a, and then pushing down on its top stretch AB and puts it under tension. A beam in tension
surface. The lines at the top become shorter while those is called a tie.
at the bottom become longer. The length of the central
line is unchanged, Fig. 1.8b. Therefore, when a beam Struts (in compression)
is loaded and bends,

the top is in compression (squeezed),

the bottom is in tension (stretched), and ~ Tie (in tension
the centre, called the neutral layer, is neither squeezed
nor eae ae

‘lg
Compression —
wy
Neutral
layer
On the other hand, AC and AD are under compres-
sion and are called struts. They are put in this state
by AB pushing down on them at A and by the bridge
Foam rubber Tension supports pushing up on them at C and D. The latter
forces arise because the truss transfers the load to the
(a)
supports.
Fig. 1.8
Bridges
In a solid beam, most of the material in the central
region (the neutral layer) is not needed. It is wasted’ In its simplest form a bridge consists of a beam, called
material whose weight simply acts as an extra load that the bridge-deck, supported at the ends, Fig. 1.11.
has to be supported. If this material is removed, the
Compression Bridge-deck
much used I-beam is obtained, Fig. 1.9, which is as
strong as a solid beam but much lighter. The top and
bottom flanges withstand the compression and tension A =" Se eee”
forces produced when the beam is loaded. Other com- Tension
mon types of girder are L- and T-shaped.

Fig. 1.11
As we have seen, when a beam is loaded it bends
and, since bending involves compression and tension,
bridge-decks must be made of materials which can with-
stand both compressive and tensile stresses. They must
I-beam L-beam T-beam Circular also be fire- and water-resistant. Other important factors
tube
are cost and the amount of maintenance required.
Fig. 1.9 Early bridges were built of stone, then came steel,
and today reinforced, prestressed and lightweight con-
Tubes use the same idea, the removal of unstressed crete are most common. Where steel and concrete or
material giving similar advantages. Circular tubes are stone are used, the design is often such that steel bars
most common, being equally strong in all directions at are in parts under tension, while the stone or concrete
right angles to the surface. is arranged to experience compression. Cables (of steel)
8 VOLUME I Materials and Mechanics

are only used for parts in tension. If steel is to be under or from below as in Fig. 1.13b. The arch may be of
compression it is usually in the form of I- or T-shaped reinforced concrete or of steel girders.
girders. There are many different types of bridge. A load on the bridge-deck causes slight compression
of the arch, whether it is above or below the deck. The
(a) Beam and pier. As well as having supports at both material of the arch should therefore be strong in com-
ends, this type has one or more pillars or piers in the pression. Stone or brick bridges often have arches below
middle, Fig. 1.12. The piers stop the bridge-deck from
the bridge-deck, Fig. 1.13c.
bending too much and make the bridge much stronger.

Bridge-deck (c) Suspension. Most of the world’s largest bridges

use this construction, Fig. 1.14. They are in effect beams
(0) Oe OO eee)
supported by steel cables, all of which are in tension.
Pier in The main cables hang from tall towers (pylons) at each
compression end, which must be built on rocks that can withstand
the large downward forces exerted by the towers.

Fig. 1.12
A load on the bridge puts the piers under compression
but, being made of stone, brick or concrete, they can
withstand this.
(b) Arch. In an arch bridge the bridge-deck is sup-
ported by an arch either from above, as in Fig. 1.13a,
Arch in compression
Fig. 1.14

The Forth road bridge in Scotland, Fig. 1.16, is of

this type.
(d) Cantilever. A cantilever is a beam which is sup-
ported only at one end, Fig. 1.15. The top is in tension
while the bottom is in compression.

Tension Load

yi —]

Arch :
compression Cantilever Compression

(b) Fig. 1.15

Stone arch in
compression The Forth railway bridge, built in 1890, part of which
is visible in the background of Fig. 1.16, uses the canti-
lever principle.
(e) Girder. Larger bridges made from steel girders
are of several types. They are designed so that there
is no material in the neutral layer. The top of the bridge
is under compression and the bottom is in tension. Fig.
1.17a shows a truss girder bridge and Fig. 1.17b a lattice
girder type. Fig. 1.17c shows how a modern box girder
bridge is built from steel boxes.
 MATERIALS AND THEIR USES 9

NV WEP:

_z Compression <a Some other structures

(a) Roof truss of a house. The walls of a house with

a tiled or slated roof would be pushed outwards, Fig.
1.18a, if the roof were not supported by a truss and
tie beam, Fig. 1.18b. The walls are in compression, the
a ye é ee
“Tension Zo tie beam is in tension.

/
Tiled roof _y Truss

le ee /
Tie beam

(a) (b)
(c) Fig. 1.18
Fig. 1.17
10 VOLUME I Materials and Mechanics

(b) Balconies. The floor of a flat and its adjoining 3. (a) What are (i) the advantages, (ii) the disadvantages, of
balcony are shown in Fig. 1.19. The prestressed concrete wood as a material for building a bridge?
beam supports both. The steel reinforcing bar is at the (b) Why do reinforced concrete bridge-decks have steel rods
bottom of the beam in the floor of the flat, because at the bottom, Fig. 1.21?
it is in tension (like a bridge-deck). However, the rein-
forcing steel bar is in the top of the beam forming the Concrete
balcony because it is a cantilever and is in tension at
the top.

Floor of flat Balcony

Fig. 1.21

4. (a) Why does a dome not fall down?

(b) What property of stone did some cathedral builders
Steel rod Steel rod
employ when they used flying buttresses to prevent the roof
pushing the top of the walls outwards?

Fig. 1.19 5. The crane in Fig. 1.22 is mounted on a wall and supports
a load.
(a) Is AB atie ora strut?
QUESTIONS (b) Is BCa tie ora strut?
1. (a) A metal is softened or annealed if it is heated to a dull
red heat and cooled slowly. If after heating it is plunged into
ee
cold water instead, it is hardened or tempered and becomes
brittle. Which of the following articles have been (i) annealed,
(ii) tempered: a paper clip, a needle, a knife blade, a pin,
a file?
(b) Why does wood split more easily along the grain than
across it?
(c) List some of the advantages of concrete over stone.
(d) Why are fabrics used for clothing often combinations
of plastics like Terylene and natural fibres such as cotton?
(e) Why do cracks not spread so readily through laminated
glass?
Fig. 1.22
2. (a) Which of AB, AC and AD in Fig. 1.10 (p. 7) could
be replaced by a cable?
(b) A beam can also be strengthened by a truss underneath
it, as in Fig. 1.20. Which of AB, AC and AD are in (i) tension,

aN all
6. In the girder bridge shown in Fig. 1.23 which girders are
i.e. ties, (di) compression, i.e. struts? (a) ties, (b) struts?

Fig. 1.20 Fig. 1.23

2 Structure of materials

Materials science Types of solid

Atoms, molecules and Brownian motion Crystal structures
The Avogadro constant: mole Bubble raft
Size of a molecule X-ray crystallography
Periodic Table Microwave analogue
Interatomic bonds Polymers
States of matter
a a a

Materials science development from the electron microscope and ‘scans’

a surface with electrons in the way that a television
Advances in technology depend increasingly on the
screen is scanned. It gives higher magnifications and
development of better materials. This is especially true
much greater depth of focus than optical microscopes
of those industries engaged in aircraft production, space
using reflected light. It is useful for examining the sur-
projects, telecommunications, computer manufacture
faces of semiconductors, the hairlike fibres and
and nuclear power engineering. Structural materials are
‘whiskers’ that are so important in the manufacture of
required to be stronger, stiffer and lighter than existing
the new generation of composite materials, man-made
ones. In some cases they may have to withstand high
fibres, and corroded and fractured surfaces. A view of
temperatures or exposure to intense radioactivity.
the end of a torn wire (X75) is shown in Fig. 2.1b and
Materials with very precise electrical, magnetic, ther-
of lead—tin telluride crystals (x 30) in Fig. 2.1c.
mal, optical or chemical properties are also demanded.
Materials science is a rapidly advancing subject with
A great deal has been known for many years about
exciting prospects for the future. Its importance lies in
materials that are useful in everyday life and industry.
the help it can give with the selection of materials for
For example, the metallurgist has long appreciated that
particular applications, with the design of new materials
alloys can be made by adding one metal to another,
and with the improvement of existing ones. The strength
and that heating, cooling or hammering metals changes
of even a tea cup has been improved by research into
their mechanical behaviour. Materials technology is a
ceramics, as Fig. 2.1d shows.
long-established subject. The comparatively new sub-
ject of materials science is concerned with the study of
materials as a whole and not just with their physical,
Atoms, molecules and Brownian motion
chemical or engineering properties. As well as asking
how materials behave, the materials scientist also wants The modern atomic theory was proposed in 1803 by
to know why they behave as they do. Why is steel strong, John Dalton, an English schoolmaster. He thought of
glass brittle and rubber extensible? To begin to find atoms as tiny, indivisible particles, all the atoms of a
answers to such questions has required the drawing given element being exactly alike and different from
together of ideas from physics, chemistry, metallurgy those of other elements in behaviour and mass. By mak-
and other disciplines. ing simple assumptions he explained the gravimetric
The deeper understanding of materials which we now (i.e. by weight) laws of chemical combination but failed
have has come from realizing that the properties of to account satisfactorily for the volume relationships
matter in bulk depend largely on the way the atoms which exist between combining gases. This required the
are arranged when they are close together. Progress introduction in 1811 by the Italian scientist, Amedeo
has been possible because of the invention of instru- Avogadro, of the molecule as the smallest particle of
ments for ‘seeing’ finer and finer details. The electron an element or compound capable of existing indepen-
microscope, which_uses beams of electrons instead of dently and consisting of two or more atoms, not neces-
beams of light as in the optical microscope, reveals struc- sarily identical. Thus, whilst we could only have atoms
ture just above the atomic level. The field ion micro- of elements, molecules of both elements and compounds
scope and X-ray apparatus allow investigation at that were possible.
level. At the end of the nineteenth century some scientists
The scanning electron microscope, Fig. 2.la, is a felt that evidence, more direct than that provided by
11
12 VOLUME I Materials
and Mechanics

Fig. 2.1

the chemist, was needed to justify the basic assumption many kinds of small particles suspended in both liquids
that atoms and molecules exist. In 1827 the Scottish and gases, is called Brownian motion. It is now con-
botanist, Robert Brown, discovered that fine pollen sidered to be due to the unequal bombardment of the
grains suspended in water were in a state of constant suspended particles by the molecules of the surrounding
movement, describing small, irregular paths but never medium.
stopping. The effect, which has been observed with Very small particles are essential. If the particle is
 STRUCTURE OF MATERIALS 13
very large compared with the size of a molecule, the atom of carbon (more precisely, on the carbon-12 iso-
impacts, occurring on every side and irregularly, will tope '2C). On the carbon scale the atomic mass of car-
cancel out and there will be no average resultant force bon-12 is taken as exactly 12, making that of hydrogen
on the particle. However, if the particle is small enough 1.008 and of oxygen 16.00. Nowadays atomic masses
to suffer impacts with only a few hundred molecules are found very accurately using a mass spectrometer.
at any instant, the chances of these cancelling out are It follows from the definition of atomic mass that any
proportionately less. It is then likely that for a short number of atoms of carbon will have, near enough, 12
time most of the impacts will be in one direction; shortly times the mass of the same number of atoms of hydro-
afterwards the direction will have changed. The pheno- gen. Therefore any mass of hydrogen, say 1 g, will con-
menon can be observed in smoke in a small glass cell tain the same number of atoms as 12g of carbon. In
which is illuminated strongly from one side and viewed general, the atomic mass of any element expressed in
from above with a low-power microscope, Fig. 2.2. How grams, contains the same number of atoms as 12g of
would the random motion be affected by (i) cooling carbon. This number is thus, by definition, a constant.
the air to a low temperature, (ii) using smaller smoke It is called the Avogadro constant and is denoted by
particles? L. Its accepted experimental value is 6.02 x 10”.
The number of molecules in the molecular mass in
grams of a substance is also (because of the way molecu-
lar masses are defined) the same for all substances and
equal to the Avogadro constant. There are, therefore,
6.02 x 10” molecules in 2 g of hydrogen (molecular mass
2) and in 18g of water (molecular mass 18). In fact,
icroscope
the Avogadro constant is useful when dealing with other
particles besides atoms and molecules and a quantity
which contains 6.02 x 10” particles is called, especially
by chemists, a mole. We can thus have a mole of atoms,
Glass rod Window a mole of molecules, a mole of ions, a mole of electrons,
(as lens)
etc.—all contain 6.02 x 10” particles. We must always
a Glass cell
have a mole of some kind of particle and so
L =6.02 x 10” particles per mole
Smoke
It should be noted that the mole (abbreviation mol)
> Fig. 2.2
is based on the gram and not the kilogram, which makes
The effect, on its own, does not offer conclusive proof it an anomaly in the SI system of units. Sometimes,
for molecules but it clearly reveals that on the micro- however, it is expressed in terms of the number of parti-
scopic scale there is great activity in matter which mac- cles per kilogram-mole and its value then is 6.02 x 10”.
roscopically (on a large scale) appears to be at rest. The Avogadro constant has been measured in various
The theory of the motion was worked out by Einstein ways. In an early method alpha particles emitted by
and found to correspond closely with observation. His a radioactive source were counted by allowing those
basic assumption was that the suspended particles have within a small known angle to strike a fluorescent screen.
the same mean kinetic energy as the molecules of the Each particle produced one scintillation on the screen
fluid and so behave just like very large molecules. Their and if it is assumed that one particle is emitted by each
motions should therefore be similar to those of the fluid radioactive atom an approximate value for L can be
molecules. obtained (see question 6, p. 26). Other methods give
more reliable results—one involves X-ray crystallo-
graphy.
The Avogadro constant: mole
Atomic and molecular masses (previously called atomic Size of a molecule
and molecular weights) give the masses of atoms and
molecules compared with the mass of another kind of (a) Monolayer experiments. An experimental deter-
atom. Originally the hydrogen atom was taken as the mination of the size of a molecule was made by Lord
standard, with atomic mass 1, since it has the smallest Rayleigh in 1899. He used the fact that certain organic
mass. In 1960 it was agreed internationally, for various substances, such as olive oil, spread out over a clean
reasons, to base atomic and molecular masses on the water surface to form very thin films.
14 VOLUME I Materials and Mechanics

Lycopodium powder

Waxed tray
Card overbrimming
with water
Loop of wire

Rubber wedge
Transparent for levelling
4mm scale
(a)
Fig. 2.3 (b)

A simple procedure for performing the experiment If copper atoms are spherical, would their radius be
is to obtain a drop of olive oil by dipping the end of larger or smaller than this even if they were packed
a loop of thin wire, mounted on a card, into olive oil, tightly? Why? A more accurate way of calculating the
quickly withdrawing it and then estimating the diameter size of a copper atom is indicated in questions 10 to
of the drop by holding it against a} mm scale and viewing 13 on p. 26.
the drop and scale through a lens, Fig. 2.3a. If the drop A word of caution is necessary regarding atomic
is then transferred to the centre of a waxed tray over- dimensions. Nowadays atoms and molecules are no
brimming with water, the surface of which has been longer pictured as having hard, definite surfaces like
previously cleaned by drawing two waxed booms across a ball and there is, therefore, little point in trying to
it and then lightly dusted with lycopodium powder, Fig. give their diameters too exact values; most are within
2.3b, it spreads out into a circular film pushing the the range 0.1 to 0.5nm. Also, although we shall usually
powder before it. Assuming the drop is spherical, the treat atoms and molecules as spheres, it is necessary
thickness of the film can be calculated if its diameter on occasion to consider them as having other shapes.
is measured. It is found to be about 2 x 10-° metre,
i.e. 2 nanometres (2 nm).
Oil-film experiments do not necessarily prove that Periodic Table
matter is particulate but from them we can infer that With the passage of time the early nineteenth-century
if molecules exist and if the film is one molecule thick, picture of an indivisible atom came to be doubted in
i.e. amonolayer, then in the case of olive oil one dimen- the light of fresh information. During the 1860s chemical
sion of its molecule is 2nm. knowledge increased sufficiently for it to be clear that
(b) Predictions from kinetic theory of gases. Informa- there were elements with similar chemical properties.
tion about the molecular world can sometimes be Moreover, atomic masses were being established with
obtained from observations of the behaviour of matter greater certainty and attempts were made to relate
in bulk, i.e. from macroscopic observations. Thus with properties and atomic masses.
the help of the kinetic theory of gases, expressions can It was found that if the elements were arranged in
be derived relating such properties as rate of diffusion order of increasing atomic masses then, at certain
with the size of the gas molecules involved. repeating intervals, elements occurred with similar
chemical properties. Sometimes it was necessary to
(c) Using the Avogadro constant. Consider copper place an element of larger atomic mass before one of
which has atomic mass 64 and density 9.0gcm~*. One slightly smaller atomic mass to preserve the pattern.
mole of copper atoms, therefore, has mass 64g and The first eighteen elements of this arrangement, called
volume 64/9cm*; it contains 6.0 x 107? atoms. The the Periodic Table, are shown in Table 2.1. The third
volume available to each atom is 64/(9 x 6 x 1073) cm3 and eleventh (3 + 8) elements are the alkali metals lith-
and the radius r of a sphere having this volume is given ium and sodium; the ninth and seventeenth (9 + 8) are
by the halogens fluorine and chlorine—here the repeating
4 64 interval is eight. The serial number of an element in
37” -Ox6x108 the table is called its atomic number.
The Periodic Table suggests that the atoms of the
r= 0114 X10-7
em =0.14 x 10 nr =0: 14m elements may not be simple entities but are somehow
 STRUCTURE OF "MATERIALS 15
Table 2.1
Group Group Group Group Group Group
iL 2 3 4 3) 6 7 0
1 2
Hydrogen Helium
3 4 5 6 7 8 y) 10
Lithium Beryllium Boron Carbon Nitrogen Oxygen Fluorine Neon
11 12 13 14 5 16 1 yf 18
Sodium Magnesium Aluminium Silicon Phosphorus Sulphur i
Chlorine Argon

related. There must be similarities between the atoms types. The type formed in any case depends on the outer
of similar elements and it would seem that the similarity electrons in the electron clouds of the atoms involved.
might be due to the way they are built up.
(a) Ionic bond. This is formed between the atoms
We now believe that atoms are composed of three
of elements at opposite sides of the Periodic Table, for
types of particle—protons, neutrons and electrons.
example between sodium (Group 1) and chlorine
(Many other subatomic particles, such as positrons,
(Group 7) when they are brought together to form com-
mesons and antiprotons, are known but most are short-
mon salt. A sodium atom has a loosely held outer elec-
lived and are not primary components.) Protons and
tron which is readily accepted by a chlorine atom. The
neutrons are packed together into a very small nucleus
sodium atom becomes a positive ion, i.e. an atom defi-
which is surrounded by a cloud of electrons, the dia-
cient of an electron, and the chlorine atom becomes
meter of the atom as a whole being at least 10000 times
a negative ion, i.e. an atom with a surplus electron.
greater than that of the nucleus. The comparative
The two ions are then bonded by the electrostatic attrac-
masses and charges of the three basic particles are given
tion between their unlike charges.
in Table 2.2. The nucleus is positively charged and the
A sodium ion attracts all neighbouring chloride ions
electron cloud negatively charged but the number of
in other pairs of bonded ions and vice versa. Each ion
protons equals the number of electrons so that the atom
becomes surrounded by ions of opposite sign and the
is electrically neutral.
resulting structure depends among other things on the
The number of protons in the nucleus of an atom
relative sizes of the two kinds of ion.
has been found to be the same as its atomic.number
The ionic bond is strong. Ionic compounds are usually
' which therefore means that each element in the Periodic
solid at room temperature and have high melting-points.
Table has one more proton and one more electron in
They are good electrical insulators in the solid state since
its atom than the previous element. Hydrogen, the first
the electrons are nearly all firmly bound to particular
element, has one proton and one electron. Helium, the
ions and few are available for conduction.
second element, has two protons and two electrons.
Lithium, with atomic number three, has three protons (b) Covalent bond. In ionic bonding electron transfer
and three electrons. Neutrons are present in all nuclei occurs from one atom to another. In covalent bonding
except that of hydrogen. electron sharing occurs between two or more atoms.
Thus the atoms of carbon can form covalent bonds with
Table 2.2 other carbon atoms. Each carbon atom has four outer
electrons, Fig. 2.4a, and all can be shared with four
other carbon atoms to make four bonds, Fig. 2.4b, each
consisting of two interlocking electron clouds.

e = electronic charge
-

Carbon atom
Interatomic bonds
e Outer electron
Materials consist of atoms held together by the attractive
forces they exert on each other. These forces are electri- (a) (b)
cal in nature and create interatomic bonds of various Fig. 2.4
16 VOLUME I Materials and Mechanics

Covalent bonds are also strong and many covalent States of matter
compounds have similar mechanical properties to ionic
The existence of three states or phases of matter is due
compounds. However, unlike the latter, they do not
to a struggle between interatomic (intermolecular)
conduct electricity when molten.
forces and the motion which atoms (molecules) have
(c) Metallic bond. Metal atoms have one or two outer because of their internal energy (see p. 76).
electrons that are in general loosely held and are readily
lost. In a metal we picture many free electrons drifting (a) Solids. In thé above discussion of the four types
around randomly, not attached to any particular atom of interatomic force only attractions were considered
as they are in covalent bonding. All atoms share all
but there must also be interatomic repulsions, otherwise
the free electrons. The atoms thus exist as positive ions
matter would collapse. Evidence, both theoretical and
in a ‘sea’ of free electrons, Fig. 2.5; the strong attraction
experimental, suggests that at distances greater than one
between the ions and electrons constitutes the metallic
atomic diameter the attractive force exceeds the repul-
bond.
sive one, whilst for small distances, i.e. less than one
atomic diameter, the reverse is true. In Fig. 2.6a the
Electron sea
dotted graphs show how the short-range attractive force
and the very short-range repulsive force between two
Positive metal ion atoms vary with the separation of the atoms; the total
(nucleus plus inner or resultant force is shown by the continuous graph.
electrons) It can be seen that for one value of the separation,
ro, the resultant interatomic force is zero. This is the
e Electron
situation that normally exists in a solid, but if the atoms
Fig. 2.5
come closer together—for example, when the solid is
compressed—they repel each other; they attract when
The nature of the metallic bond has a profound they are pulled farther apart. We have only considered
influence on the various properties of metals, as we shall two atoms whereas in a solid each atom has interactions
see later. with many of its close neighbours. The conclusion about
the existence of an equilibrium separation, however,
(d) Van der Waals bond. Van der Waals forces are
will still hold good.
very weak and are present in all atoms and molecules.
In an ionic bond the short-range attractive part of
They arise because, although the centres of negative
the interatomic force arises from the attraction between
and positive charges in an atom coincide over a period
positive and negative ions which pulls them together
of time, they do not coincide at any instant—for reasons
until their electron clouds start to overlap, thus creating
too advanced to be considered here. There is a little
a very short-range repulsive force. The attractive and
more of the electron cloud on one side of the nucleus
repulsive forces in the other types of bond also arise
than the other. A weak electric ‘dipole’ is produced
from the electric charges in atoms.
giving rise to an attractive force between opposite ends
Now consider the motion of the atoms, the other con-
of such dipoles in neighbouring atoms.
testant. In a solid the atoms vibrate about their equili-
The condensation and solidification at low tempera-
brium positions, alternately attracting and repelling one
tures of oxygen, hydrogen and other gases is caused
another, but the interatomic forces have the upper hand.
by van der Waals forces binding their molecules
The atoms are more or less locked in position and so
together. (In the molecules of such gases the atoms are
solids have shape and appreciable stiffness.
held together covalently.) Van der Waals forces are also
The corresponding potential energy—separation curve
important when considering polymers (p. 25).
for two atoms (or molecules) is shown in Fig. 2.6b. At
Two further points: first, sometimes more than one the equilibrium separation ry when the resultant force
of the previous four types of bonding is involved in a is zero, the p.e. must have its minimum value Ep. This
given case; second, information about the strength of is so because any attempt to change the separation
interatomic bonds in solids is obtained from heat of sub- involves overcoming an opposing force—an attractive
limation measurements in which solid is converted one if the separation increases and a repulsive one if
directly to vapour and all atoms separated from their it decreases. Ep is called the bonding energy; it is the
neighbours (see question 8, p. 26) and for liquids latent energy needed to pull the atoms apart so that their p.e.
heat of vaporization measurements provide the informa- increases to zero. They are then quite free from one
tion (see p. 81). another’s influence.
 SERUGTRURE;/ OF) MATERIALS, 17
> +
lop)
\ Resultant force ©
i=
o
Repulsion i)
fe
oO
is)
Separation a 0 Separation
Force of atoms of atoms

Eo

(a)
Fig. 2.6 (b)

Bonding energy will be considered later in connection Types of solid

with latent heat (p. 81) and surface energy (p. 230).
There are two main types of solid.
(b) Liquids. As the temperature is increased the
atoms have larger amplitudes of vibration and even- (a) Crystalline. Most solids, including all metals and
tually they are able partly to overcome the interatomic many minerals, are crystalline. In substances such as
forces of their immediate neighbours. For short spells sugar the crystal form is evident but less so in the case
they are within range of the forces exerted by other of metals, although large crystals of zinc are often visible
atoms not quite so near. There is less order and the on a freshly galvanized iron surface.
solid melts. The atoms or molecules of a liquid are not The crystalline structure of a metal can be revealed
much farther apart than in a solid but they have greater by polishing the surface, treating it with an etching
speeds, due to the increased temperature, and move chemical, sometimes a dilute acid, and then viewing
randomly in the liquid while continuing to vibrate. The it under an optical microscope. The metal is seen to
difference between solids and liquids involves-a differ- consist of a mass of tiny crystals, called ‘grains’, at var-
ence of structure rather than a difference of distance ious angles to one another; it is said to be polycrystalline.
between atoms or molecules. Grain sizes are generally small, often about 0.25mm
Although the forces between the molecules in a liquid across. Fig. 2.7 shows crystal grains in a cross-section
do not enable it to have a definite shape, they must of an aluminium-—copper casting. The grains show up
still exist otherwise the liquid would not hold together on the surface after etching because ‘steps’ are formed
or exhibit surface tension (i.e. behave as if it had a on each grain due to the rate of chemical action differing
skin on its surface) and viscosity, nor would it have latent with different grain orientations. Light is then reflected
heat of vaporization. in various directions by the different grains so that some
appear light and others dark, Fig. 2.8.
(c) Gases. Ina gas or vapour the atoms and molecules Grain boundaries are revealed at the atomic level by
move randomly with high speeds through all the space the field ion microscope, Fig. 2.9, which uses beams
available and are now comparatively far apart. On aver- of helium ions to ‘illuminate’ the object rather than elec-
age their spacing at s.t.p. is about 10 molecular dia- trons or light. The field ion micrograph in Fig. 2.10
meters and their mean free path (i.e. the distance shows the tip of an iridium needle, viewed from the
travelled between collisions) is roughly 300 molecular point and magnified 2500000 times. Each bright spot
diameters. Molecular interaction only occurs for those represents an iridium atom and the abrupt pattern
brief spells when molecules collide and large repulsive change is clearly visible at the grain boundary.
forces operate between them. The essence of the structure of a crystal, whether it
be a large single crystal or a tiny grain in a polycrystalline
Conditions in gases and solids are, by comparison, specimen, is that the arrangement of atoms, ions or
simpler than those in liquids and in general their beha- molecules repeats itself regularly many times, i.e. there
viour is better understood. is a long-range order.
18 VOLUME I Materials and Mechanics

1Oem

Light

Steps etched
oncrystal
grain
Grain boundary

Fig. 2.8

Fig. 2.9
 STRUCTURE OF MATERIALS 19

Fig. 2.10

(b) Amorphous or glassy. Here the particles are atoms. This does not encourage close-packing since the
assembled in a more disordered way and only show number of atoms immediately surrounding each atom
_ order over short distances; there is no long-range order. is limited to the number of covalent bonds it forms.
The structure of an amorphous solid has been likened Would you describe the ionic bond as directional or
to that of an instantaneous photograph of a liquid. It non-directional?
is much more difficult to unravel but is the subject of Some common crystal structures will now be des-
considerable research. The many types of glass are the cribed.
commonest of the amorphous solids; we can think of
(a) Face-centred cubic (FCC) packing is shown in Fig.
them as having a structure of groups of atoms (e.g. of
2.1la; there is a particle at the centre of each of the
silicon and oxygen) that would have been crystalline
six faces of the cube in addition to the eight at the
had it not been distorted.
corners. Copper and aluminium have this structure. The
sodium chloride crystal can be regarded as two interpe-
netrating FCC structures, one of sodium ions and the
Crystal structures
other of chloride ions, Fig. 2.11b; each sodium ion is
The structure adopted by a crystalline solid depends surrounded by six chloride ions and vice versa.
on various factors including the kind of bond(s) formed
(b) Hexagonal close-packing (HCP) is represented in
and the size and shape of the particles involved. For
Fig. 2.12; it is built up from layers of hexagons. Zinc
example, in metals where all the positive ions attract
and magnesium form HCP crystals.
all electrons (p. 16), the bonding pulls equally in all
directions, i.e. is non-directional, and every ion tends These two structures give the closest possible packing
to surround itself by as many other ions as is geometri- and account for 60% of all metals. They are not very
cally possible. A close-packed structure results. On the dissimilar if we consider how they can be assembled
other hand, in covalent solids the bonding is directional, from successive layers. Fig. 2.13 shows a layer A of
i.e. every shared electron is localized between only two hexagonal close-packed spheres in which each sphere
20 VOLUME I Materials and Mechanics

2, re

Chloride ion Sodium ion

(a) (b)

Fig. 2.11 FCC structure

das
Cae
Fig. 2.14 BCC structure

ih (a)
Fig. 2.13 Fig. 2.15 Tetrahedral structure

touches six others—this is the closest packing possible four corners, Fig. 2.15a. This more open arrangement
for spheres. A second hexagonal close-packed layer B is found in carbon (as diamond), silicon and germanium
can be placed on top and the packing of these two layers —all substances which form covalent bonds. The hard-
will be closest when the spheres of B sit in the hollows ness of diamond is partly due to the fact that its atoms
formed by three neighbouring spheres of A. A third are not in layers and so cannot slide over each other
hexagonal close-packed layer can be placed on top of as they can in graphite, the other crystalline form of
B in two ways. If it rests in the hollows of B so that carbon. Graphite forms layers of six-membered rings
its spheres are directly above the spheres in A, then of carbon atoms that are about two-and-a-half times
an HCP crystal results, Fig. 2.13 (bottom), and the layer farther apart than are the carbon atoms in the layers,
stacking is ABAB. However, if the third layer rests in Fig. 2.15b. The forces between the layers are weak,
other hollows in B, its spheres can be above hollows thus explaining why graphite flakes easily and is soft
in A and the structure is FCC with layer stacking and suitable for use in pencils and as a lubricant.
ABCABC. Graphite and diamond provide a good example of the
importance of structure in determining properties.
(c) Body-centred cubic (BCC) packing has a particle
at the centre of the cube and one at each corner, Fig.
Two further points: first, there is in every crystal struc-
2.14. Alkali metals have this less closely packed struc-
ture a typical cell, called the unit cell, which is repeated
ture.
over and over again—Figs. 2.1la, 2.12 and 2.14 are
(d) Tetrahedral structures have a particle at the examples of unit cells; second, the structures described
centre of a regular tetrahedron and one at each of the are those of perfect crystals. In practice there are imper-
 STRUCTURE: OFAMATERIALS 21

fections in crystals and these are important in determin- which gives bubbles of about 2mm diameter, Fig. 2.16a.
ing the properties of a material, as we shall see later. If the dish is placed on an overhead projector a magni-
fied image of the raft can be viewed. What pulls the
Bubble raft thes together and what keeps them from getting too

Soap bubbles pack together in an orderly manner and A perfect, hexagonally close-packed array is shown
provide a good representation, in two dimensions, of in Fig. 2.16b. Grain boundaries are readily seen in Fig.
how atoms are packed in a crystal. 2.16c, ‘vacancies’ in Fig. 2.16d and a bubble of a differ-
A bubble raft is made by attaching a glass jet (about ent size in Fig. 2.16e. What might (d) and (e) represent
1mm bore) or a 25 gauge hypodermic needle on a 1 cm? in a real crystal and what effect do they have on the
syringe barrel, to the gas tap, via a length of rubber structure? Gcrevrclie
tubing and a screw clip. The jet is held below the surface
of a ‘soap’ solution (1 Teepol, 8 glycerol and 32 water
is satisfactory) in a shallow glass dish, at a constant depth

= From gas tap ‘Soap solution

Barrel of
Tubing
1cm3 syringe
=
Hypodermic needle (25 gauge)
Shallow glass dish ‘Soap bubbles

OO OOOO
ROOMS OLE
Ones
ere wetome

CPx)
ee
eae
One
(b)
22 VOLUME I Materials and Mechanics

X-ray crystallography
Before the discovery which enabled the arrangement
of atoms in a crystal to be determined experimentally, Single
crystallographers had simply assumed that the regular plane of
external shapes of crystals were due to the atoms being atoms
arranged in regular, repeating patterns. Strong
In 1912 three German physicists, Max von Laue, W. beam
Friedrich and P. Knipping, found that a beam of X-rays,
on passing through a crystal, formed a pattern of spots
(a)
on a photographic plate (see Fig. 2.19b). Shortly after-
wards W. L. (later Sir Lawrence) Bragg and his father Incident beam Reflected beam
Sir William Bragg showed how the pattern could be (all waves in step)
used to reveal the positions of the atoms in a crystal. wa

Together they proceeded to unravel the atomic struc-

tures of many substances and started the science of X-
ray crystallography. In recent years the structures of
many complex organic molecules, including some like .~ Layers of
DNA (deoxyribonucleic acid) that play a vital part in 4 atom
the life process, have been discovered by this technique.
The analysis of crystal structures by X-rays depends
on the fact that X-rays, like light, have a wave-like
nature and when they fall on a crystal they are scattered Fig. 2.17
by the atoms so that in some directions the scattered
beams reinforce each other while in others they cancel
each other. X-rays are used because their wavelengths
are of the same order as the atomic spacings in crystals—
about 10000 times less than those of light.
The simplest way of regarding what occurs when X-
rays fall on a crystal is to consider the crystal as made
up of regularly spaced layers of atoms, each of which
produces a weak ‘reflected’ beam such that the angle
a
of incidence equals the angle of reflection, as for the Fig. 2.18
reflection of light by a mirror. In Fig. 2.17a reflection
by a single layer of atoms is shown: most of the beam photograph. Fig. 2.19a is due to a polycrystalline sample
passes through. Fig. 2.17b shows a beam of X-rays fall- of gold and Fig. 2.196 to a single crystal of the same
ing on aset of parallel layers of atoms. If all the reflected material.
waves are to combine to produce a strong reflected beam
(and give an intense spot on a photograph) then they
Microwave analogue
must all emerge in step. For this to happen the path
difference between successive layers must be a whole Microwaves are very short radio waves with wavelengths
number of wavelengths of the X-rays—as they are in extending from 1cm or so to about 1m and are used
this case. Otherwise crests of one wave may coincide for radar and satellite communication. A large-scale
with troughs from another and the two tend to cancel demonstration with microwaves shows that regularly
out. spaced polystyrene tiles or layers of spheres give strong
The atoms in a crystal can be considered as arranged reflections at certain angles, depending on the tile or
in several different sets of parallel planes, from all of layer separation, and that this can be explained as being
which strong reflections may be obtained to give a pat- due to interference between waves reflected from suc-
tern of spots characteristic of the particular structure. cessive layers—as it can with X-rays and layers of atoms.
Two other possible sets of planes are shown for the The apparatus is shown in Fig. 2.20a; wax lens A
array in Fig. 2.18. In a polycrystalline sample many produces a parallel beam from the 3cm microwave
planes are involved at once and thousands of spots are transmitter and B focuses it on the detector. First it
produced resulting in circles or circular arcs on the should be shown, with the transmitter and detector in
 STRUGCTURE-OF MATERIALS 23

Fig. 2.19 ia)

Microwave ae ee Amplifier and :

transmitter loudspeaker :
Alignment ‘A /
rod To 1:5V battery
and variable resistor
(a)

From
transmitter (b)

Plan view

(c)

Fig. 2.20 (d)

24 VOLUME I Materials and Mechanics

line, that microwaves can largely penetrate a polysty- spheres—see question 9, p. 26) is responsible for the
rene ceiling tile but not a metal sheet. strong signals when, twice in each revolution, it bisects
the angle between the beam and the detector.
(a) One tile. A single tile held vertically on the turn-
table reflects (partially) the beam at any angle. Through
what angle from the straight-through position must the
detector be turned when the glancing angle (i.e. the Polymers
angle between the tile and the incident beam) is 6 (see
Polymers are materials with giant molecules, each con-
p. 93)?
taining anything from 1000 to 100000 atoms, and are
(b) Two tiles. When a second tile is brought up behind usually carbon (organic) compounds. An example of
and parallel to the first (already positioned for the ref- a natural polymer is cellulose whose long, tough fibres
lected beam to be detected), the intensity of the reflec- give strength and stiffness to the roots, stems and leaves
tion rises and falls as the extra path to the second tile of plants and trees. Rubber, wool, proteins, resins and
varies between an even and an odd number of half- silk are others. Man-made polymers include plastics
wavelengths. Interference is occurring between the such as polythene, Perspex and polystyrene, fibres like
waves reflected from each tile when they come together. nylon and Terylene, synthetic rubbers and the epoxy
Will this occur for all glancing angles? resins which are well known for their strong bonding
properties and toughness.
(c) Ten tiles at 3cm centre-to-centre spacing, Fig.
The unravelling of the intricacies of nature’s polymers
2.20b. If the detector is swung round as the array of
required X-ray apparatus, the electron microscope and
tiles revolves on the turntable, a strong reflection is
other instruments. Their molecules were found to con-
obtained only when the detector makes an angle of 60°
sist of a large number of repeating units, called
with the straight-through position. The tiles then bisect
monomers, arranged in a long flexible chain. Thus every
the angle between the detector and the straight-through
molecule of cellulose comprises a long chain of from
position, 1.e. they make an angle of 30° with the straight-
a few hundred to several thousand glucose sugar
through position, and give a glancing angle of 30°, Fig.
(C,H,0¢,) molecules.
2.20c.
Artificial polymers are made by a chemical reaction
(d) Polystyrene ball ‘crystal’. This is made from seven known as polymerization, in which large numbers of
hexagonal close-packed layers of 5 cm diameter spheres small molecules join together to form a large one.
glued together to give a face-centred cubic (FCC) struc- Polythene or polyethylene (to give it its full name) is
ture (see Appendix 6). As the crystal rotates on the made by polymerizing ethylene (C,H,), a gas obtained
turntable, Fig. 2.20d, two pairs of strong reflections are when petroleum is ‘cracked’. In one process the ethylene
obtained per revolution when the detector is at 44° from molecules, heated to 100-300°C under a pressure
the straight-through position (@=22°), the time several thousand times greater than atmospheric, link
between each pair being greater than that between the with one another to give the long chain molecules of
two signals in each. At 50° (@ = 25°) there are two single polythene, Fig. 2.21.
strong reflections per revolution and similarly at 74° If the chains run parallel to each other, like wires
(0= 37°). in a cable, the structure shows a certain amount of order
The first pair of 44° reflections is produced when two and is said to be ‘crystalline’, Fig. 2.22a. This contrasts
easily identifiable sets of vertical ‘hexagonal’ layers (i.e. with the disorder of tangled chains in an ‘amorphous’
with the packing of spheres in each layer the closest structure, Fig. 2.22b. Many polymers have both crystal-
possible) bisect in turn the angle between the straight- line and amorphous regions, Fig. 2.22c. If crystallinity
through position and the detector; the second pair arises predominates an X-ray photograph shows sharp spots
half a revolution later when the ‘backs’ of the two sets (but the pattern is never as sharp as for wholly crystalline
of the same layers are in the bisecting position. At 50°, materials) and the polymer is fairly strong and rigid.
a set of ‘square’ layers (i.e. with less closely packed A polymer with a largely amorphous structure is soft

> + OO + @ + & —— ——OCR090%Y3}OYO and so on

Ethylene Polythene
molecule © Carbon atom molecule
(C5H,)
¢ Hydrogen atom
Fig. 2.21
 SHERUGLURESOLUMA TERIALTS 25

Long chain
molecule

WySOK
oe aces

CLES
Fig. 2.22

and flexible and gives diffuse rings in an X-ray photo- hold the chains together and these materials can be sof-
graph. The proportion of crystalline to amorphous tened by heating and if necessary remoulded. On cool-
regions in a polymer depends on its chemical composi- ing they recover their original properties and retain any
tion, molecular arrangement and how it has been pro- new shape. This treatment can be repeated almost inde-
cessed. The intermolecular forces between chains are finitely so long as temperatures are below those causing
of the weak van der Waals type, but in crystalline struc- decomposition, i.e. breakdown of the covalent bonds
tures the chains_are close together over comparatively that hold together the atoms in the long chain.
large distances and so the total effect of these forces The possibility of using man-made polymers in the
is to produce a stiff material. future as load-bearing structural materials for houses,
When an amorphous (glassy) polymer is stretched the buildings, cars, boats and aircraft depends largely on
chains become less coiled and tangled. They line up how far their strength and stiffness can be increased.
giving a more ordered (‘crystalline’) structure, making As we have seen, two methods are used to do this at
the material stiffer and able to take strains of 100% present. One is by having ‘crystallized’ long chains—a
or more. Rubber behaves similarly when stretched (p. physical feature—and the other requires cross-links to
41) and the polymer is said to change from the glassy be formed between chains—a chemical feature. Current
to the rubbery state if stressed. research is directed towards producing molecular chains
Crystallization is one of two principles that have been which are themselves stiff (most existing man-made
used to produce strong, stiff polymers (e.g. polythene, polymer chains are inherently flexible) by polymerizing
nylon); the other is the formation of strong covalent monomers which have a ring-shaped structure. It is then
bonds between chains—a process known as cross-link- hoped to achieve the desired strength and stiffness by
ing. In vulcanizing raw rubber, i.e. heating it with a crystallizing and cross-linking those chains.
_ controlled amount of sulphur, a certain number of sul-
phur atoms form cross-links between adjacent rubber
molecules to give a more solid material than raw rubber
which is too soft for use, Fig. 2.23. As more cross-links
are added to rubber it stiffens and ultimately becomes
the hard material called ebonite.
QUESTIONS
1. Experiment shows that 3g of carbon combine with 8g of
oxygen to form 11g of carbon dioxide. If 1 atom of carbon
reacts with 2 atoms of oxygen to give 1 molecule of carbon
Cross-links
dioxide (i.e. C + O, = CO,)
(a) compare the mass of an oxygen atom with that of a car-
bon atom. (Hint: start by supposing that 1g of C contains x
atoms.)
Fig. 2.23 (b) what is the atomic mass of oxygen on the carbon-12
Polymers such as ebonite and bakelite (the first plastic scale?
(c) what mass of oxygen contains the same number of atoms
to be made) with many strong cross-links do not soften
as 12g of C?
with increased temperature but set once and for all after
their initial moulding. They are called thermosetting 2. (a) If the atomic mass of nitrogen is 14, what mass of nitro-
plastics or thermosets and remain comparatively strong gen contains the same number of atoms as 12 g of carbon?
until excessive heating leads to breakdown of the cross- (b) What mass of chlorine contains the same number of
links and chemical decomposition. By contrast, in ther- atoms as 32 g of oxygen? (Atomic masses of chlorine and oxy-
moplastic polymers only the weak van der Waals forces gen are 35.5 and 16 respectively.)
26 VOLUME I Materials and Mechanics

3. Taking the value of the Avogadro constant as 6.0 x 107 spheres are touched by (i) A, (ii) B? In which arrangement
how many atoms are there in (a) 14g of iron (at. mass 56), is the packing closest?
(b) 81g of aluminium (at. mass 27), (c) 6.0g of carbon (at. (b) Fig. 2.24c is a pyramid of spheres in which the second
mass 12)? and successive layers are formed by placing balls in the hollows
of the layer below it. How are the balls packed in (7) the sloping
4. What is the mass of (a) one atom of magnesium (at. mass sides of the pyramid, (ii) the horizontal layers?
24), (b) three atoms of uranium (at. mass 238), (c) one molecule
of water (mol. mass 18)? Take the value of the Avogadro
10. One face of the.unit cell of an FCC crystal is shown in
constant as 6.0 x 10°. Fig. 2.25, atoms being represented by spheres. Ifr is the atomic
radius in cm, calculate (a) the length of a side of the unit cell,
5. (a) A mole is the name given to the quantity of substance
(b) the volume of a unit cell, (c) the number of unit cells in
which contains a certain number of particles. What is this
1cm’.
number?
(b) What is the mass of 1 mole of hydrogen molecules?
(c) If the density of hydrogen at s.t.p. is 9.0 x 10> gcm~?
Side of
what volume does 1 mole of hydrogen molecules occupy at unit cell
s.t.p.?
(d) How many molecules are there in 1cm* of hydrogen
ats.t.p.?
Fig. 2.25
6. By counting scintillations it is found that 1.00 mg of polo-
nium in decaying completely emits approximately 2.90 x 10!8
alpha particles. If one particle is emitted by each atom and 11. In a crystal built up from a large number of similar unit
the atomic mass of polonium is 210, what is the Avogadro cells the atoms at the corners and on the faces of individual
constant? cells are shared with neighbouring cells. In an FCC unit cell
(a) how many corner atoms are there?
7. Estimate (a) the mass and (b) the diameter of a water mole- (b) how many neighbouring cells share each corner atom?
cule (assumed spherical) if water has molecular mass 18 and (c) what is the effective number of corner atoms per cell?
the Avogadro constant is 6.0 x 10” per mole. (d) how many face atoms are there?
(e) how many neighbouring cells share each face atom?
8. (a) Suggest an approximate but reasonable value for the (f) what is the effective number of face atoms per cell?
heat of sublimation of copper (in Jg~!) from the following (g) what is the total effective number of atoms per cell?
data. (h) what is the total effective number of atoms in 1 cm? of
unit cells (use your answer from 10(c))?
Specific latent heat of fusion = 2.0 x 10°Jg™!
Specific latent heat of vaporization = 4.8 x 10° J g7!
12. X-ray diffraction shows that copper, atomic mass 64 and
What additional information would enable a better estimate density 9.0 gcm~?, has an FCC structure. If the Avogadro con-
to be made? stant is 6.0 x 10% per mole, how many atoms are there in
(b) If the Avogadro constant is 6.0 x 107° per mole and the 1.0m? of copper?
atomic mass of copper is 64, what is the heat of sublimation
of copper in J/atom? Why is a knowledge of this quantity
useful? 13. Using your answer to 11(h) and 12, calculate the atomic
radius of copper.
9. (a) ‘Square’ and ‘hexagonal’ methods of packing spheres
are shown in Figs. 2.24a and b respectively. How many other 14. For crystalline sodium chloride, draw the unit cell which
is repeated throughout the lattice. Label precisely the two kinds
of particle at the lattice sites. What are the forces maintaining
them in their relative positions?
GOCS@e) Calculate the distance between adjacent particles in crystal-
C0868 line sodium chloride, given that its formula weight is 58.5 and
Square packing Hexagonal packing Pyramid
its density is 2.16 gcm~> (2.16 x 10°kgm~*). (Avogadro con-
stant = 6.03 x 1073 mole7!)
(a) (b) (c) Discuss the effect of a small stress on a crystalline lattice.
Fig. 2.24 (CG. PhysaSe:)
3 Mechanical properties
Stress and strain Fatigue and creep
The Young modulus Composite materials
Stretching experiments Strain energy
Deformation and dislocations Rubber
Strengthening metals Elastic moduli
Cracks and fracture

Stress and strain all ‘give’ in some degree although the deformation may
often be very small. Ductility or workability relates to
The mechanical properties of a material are concerned the ability of the material to be hammered, pressed,
with its behaviour under the action of external forces—a bent, rolled, cut or stretched into useful shapes. A tough
matter of importance to engineers when selecting a material is one which is not brittle, i.e. it does not crack
material for a particular job. Four important mechanical readily. Steel has all four properties, putty has none of
properties are strength, stiffness, ductility and tough- them. Glass is strong and stiff but not tough or ductile.
ness. Which properties would you ascribe to rubber, nylon
Strength deals with how great an applied force a and diamond?
material can withstand before breaking. Stiffness tells Information about mechanical properties may be
us about the opposition a material sets up to being dis- obtained by observing the behaviour of a wire or strip
torted by having its shape or size, or both, changed. of material when it is stretched. The stretching of short
A stiff material is not very flexible. There is no such rods or ‘test-pieces’ is done using a machine like that
thing as a perfectly stiff or rigid (unyielding) material; in Figs. 3.1 and 3.2.

Fig. 3.2
28 VOLUME I Materials and Mechanics

The extension produced in a sample of material the stress doubles the strain. Over this range the mater-
depends on (i) the nature of the material, (ii) the stretch- ial suffers elastic deformation, i.e. it returns to its origi-
ing force, (iii) the cross-section area of the sample and nal length when the stress is removed and none of the
(iv) its original length. What effect would you expect extension remains.
(iii) and (iv) to have? To enable fair comparisons to
(b) Plastic deformation. As the stress is increased the
be made between samples having different sizes the
graph becomes non-linear but the deformation remains
terms ‘stress’ and ‘strain’ are used when referring to
elastic until at a certain stress corresponding to point
the deforming force and the deformation it produces.
P, and called the yield point, permanent or plastic defor-
Stress (a: pronounced ‘sigma’) is the force (in N) act-
mation starts. Henceforth the material behaves rather
ing on unit cross-section area (1m7’). For a force F and
like Plasticine and retains some of its extension if the
area A (Fig. 3.3) we can write stress is removed. Recovery is no longer complete and
orce F on reducing the stress at A, for example, the specimen
stress = orc=— recovers along AO’ where AO’ is almost parallel to
area A
OE. OO’ is the permanent plastic extension produced.
The unit of stress is the pascal (Pa) which equals one If the stress is reapplied, the curve O’AD is followed.
newton per square metre (Nm~°). At D the specimen develops one or more ‘waists’ and
Strain(e: pronounced ‘epsilon’) is the extension of unit ductile fracture occurs at one of them. The stress at D
length (1m). If e=extension and /=original length is the greatest the material can bear and is called the
(Fig. 3.3) then breaking stress or ultimate tensile strength; it is a useful
extension e measure of the strength of a material.
strain = ——————— or € The specimen appears to ‘give’ at P, and over the
original length l
plastic region a given stress increase produces a greater
Strain is a ratio and has no unit.
increase of strain than previously. None the less it still
opposes deformation and any increase of strain requires
Force exerted by increased stress. Beyond P the material is said to work-
support harden or strain-harden.

Sample
Stress

A
” (Force exerted by
external agent)
Fig. 3.3
Strain
Fig. 3.4
A stress which causes an increase of length puts the
sample in tension, and so we talk about a tensile stress
and a tensile strain.
Few ductile materials behave elastically for strains
The shape of the stress-strain graph for the stretching as great as }% (i.e. for an extension of $cm in a 100cm
of a sample (e.g. a wire) depends not only on the mater-
long wire) but they may bear large plastic strains, up
to 50%, before fracture. The breaking stress for steels
ial but also on its previous treatment and method of
may occur at stresses half as great again as that at the
manufacture. For a ductile material, i.e. a metal, it has
the general form shown by OEPAD in Fig. 3.4. There yield point, while for very ductile metals with low yield
are two main parts. points it may be several times greater than the stress
at the yield point. It is desirable that metals used in
(a) Elastic deformation. The first part of the graph engineering structures should carry loads which only
from O to E is a straight line through the origin showing deform them elastically. On the other hand the fabrica-
that strain is directly proportional to stress, i.e. doubling tion of metals into objects of various shapes requires
 ME GHANICAL PROPERTIES. 29
them to withstand considerable plastic deformation area A and original length / causes an extension e we
before fracture, i.e. be very ductile. The dominant posi- can write
tion of metals in modern technology arises from their
strength and ductility.
A brittle material such as a glass may give a curve tensilestrain e/l Ae
like OB in Fig. 3.4 and fractures almost immediately
Like stress, E is expressed in pascals (Pa) since strain
after the elastic stage; little or no plastic deformation
is a ratio. Suppose a load of 1.5kg attached to the end
occurs and the glass is non-ductile.
of a wire 3.0m long of diameter 0.46mm stretches it
by 2.0mm then F = ma = mg = 1.5 X 9.8N
(g=9.8ms~’)

Fl
The Young modulus
Ae
The stress-strain curve for the stretching of metals and
some other materials (e.g. glasses), over almost all the - (1.5 x 9.8N)(3.0m)
elastic region, is a straight line through the origin. That (a X 0.237 x 107° m?)(2.0 x 1073m)
is, tensile strain is directly proportional to tensile stress
during elastic deformation. This statement is known as * 1.5 x 9.8 x 3.0 Nm
Hooke’s law and in more elementary work it is often ar X 0.232 x 10-6 x 2.0 x 1073 m2m
stated in the form: extension varies as the load. In
mathematical terms it can be written =1.3 x 10!! Pa

tensile strain « tensile stress Approximate values of E for some common materials
are given in Table 3.1.
tensile stress
or aia constant
tensile strain Table 3.1
This constant is called the Young modulus and is Material The Young modulus
denoted by E. Its value is given by the slope of the E/10! Pa
straight part of the stress-strain graph and depends on steel 21
the nature of the material and not on the dimensions copper 13
glasses 7
of the sample. If a material has large E, it resists elastic
polythene about 0.5
deformation strongly and a large stress is required to rubber about 0.005
produce a small strain. F is thus a measure of the opposi-
tion of a material to change of length strains such as
Glasses are surprisingly stiff (and strong). The high
occur when a wire or rod is stretched elastically, i.e.
elasticity of rubber (i.e. its ability to regain its original
it measures elastic stiffness.
shape after a very large deformation) is not to be con-
The Young modulus is of great importance in engi-
fused with its low elastic modulus. For steel a large stress
neering. In the early days of iron railway bridge con-
gives a small strain while the same stress applied to rub-
struction, engineers relied heavily on ‘rule of thumb’
ber will give a very much larger strain. This means that,
methods. It required a series of disasters like that of by our scientific definition, steel has a greater modulus
the Tay Bridge in Scotland in 1879 and a reputed col- of elasticity than rubber.
lapse rate of twenty-five bridges per year at about the
same time in the U.S.A. before it was accepted that
reliable strength calculations were necessary for safety
and the economical use of materials. The value of E
Stretching experiments
is one of the pieces of information which must be known
to calculate accurately the deformation (deflections) 1. Copper. Using an arrangement like that in Fig.
that will occur in a loaded structure and its parts. As 3.5 the extensions produced in a 2 metre length of cop-
we have seen in chapter 1, when a beam bends one per wire (SWG 32) are found as it is loaded to breaking
surface is compressed and the other stretched, so that with 100 gram slotted weights.
E is involved; ‘beam theory’ is one of the foundation A load-extension graph is plotted to see if Hooke’s
stones of engineering. law is obeyed and to find the percentage strain copper
If a stretching force F acting on a wire of cross-section can withstand before its elastic limit is exceeded.
30 VOLUME I Materials and Mechanics

Gummed paper on
wire to measure
extension

Slotted weight
Fig. 3.5 hanger

The breaking stress of copper can be determined and, 5. Glass. The breaking stress of glass may be found
if wire of another gauge is used (say SWG 26), whether by hanging weights from a glass thread which has been
the breaking stress depends on wire thickness. The per- freshly drawn from a length of 3 mm diameter soda glass
centage plastic strain borne by copper before it breaks rod. If parts of the rod are left at the top and bottom,
may also be found. the thread can be supported by clamping one end and
a hook made at the other end for the weights.
2. Steel. If J is repeated for 2 metres of steel wire
(SWG 44) the same kind of information can be obtained.
6. The Young modulus for a wire. Using the appara-
3. Rubber. A load—extension graph can be plotted
tus of Fig. 3.6a or b the extensions of a wire can be
for a strip of rubber (5 cm long and 2mm wide cut from
measured with greater accuracy.
a rubber band), suspended vertically and loaded with
In Fig. 3.6a the right-hand wire is under test and car-
100 g slotted weights and then unloaded. Information
ries a vernier scale (see Appendix 3) which, when the
may be obtained about adherence to Hooke’s law, and
right-hand wire is loaded, moves over a millimetre scale
also the number of times its original length that rubber
attached to the left-hand wire and enables the extension
can be extended. The breaking stress of rubber should
to be measured. The alternative and more accurate
be found.
arrangement in Fig. 3.65 is known as Searle’s apparatus.
4. Polythene. A strip 15cm long and 1cm wide, cut In this case the micrometer screw (see Appendix 3) is
cleanly from a piece of polythene, can be investigated adjusted, after the addition of a load to the right-hand
asin 3. wire, so that the bubble of the spirit level is centralized.

Spirit level

Vernier scale

Micrometer

Fixed load
Fig. 3.6
 MECHANICAL PROPERTIES 31
The extension is then found from the scale readings.
By having two wires of the same material suspended
from the same support, errors are eliminated if there 8 S
& S)
is a change of temperature or if the support yields, since
2 , 2
both wires will be affected equally. € Separation is
2 of atoms S / Separation
Initially both wires should have loads that keep them = © of atoms
taut and free from kinks. Readings are then taken as © ©
£ £
the load on the right-hand wire is increased by equal
steps, without exceeding the elastic limit. The strain
should therefore not be more than 0.1%, i.e. the wire (a) (b)
should not be stretched much beyond 1/1000th of its Fig. 3.8
original length. The length / of the wire to the top of
the vernier or micrometer is measured with a metre
rule and the diameter (27) found at various points along The Young modulus E, the measure of elastic stiffness
its length with a micrometer screw gauge. (i.e. resistance to elastic deformation), is high for mater-
From a graph of ‘load’ against ‘extension’, an average ials with strong interatomic bonds. Covalent and ionic
value of ‘load’/‘extension’ in kgmm™! is given by solids, and to a lesser extent metals, are in this category.
PQ/OQ, Fig. 3.7. The Young modulus can then be cal- Diamond (pure carbon), the hardest known natural sub-
culated from stance, has a large number of very strong covalent bonds
Fl per unit volume and a very high value of E.
As well as determining the stiffness of a material,
Ae
the Young modulus also governs ultimately, in theory,
where F/e is expressed in Nm“, i.e. F/e=PQ x g/ its strength since this too depends on the forces between
(OQ x 1073), A(= a7’) in m? and /in m. atoms. However, we shall see later (pp. 32-5) that there
are other factors which prevent solids displaying their
theoretical strengths.
(b) Plastic strain. The ability to undergo plastic strain
(and be ductile) is a property of crystalline materials.
The yielding which occurs could therefore be attributed
to the slipping of layers of atoms (or ions) over one
Load/kg
another. With close-packed layers like those in Fig.
3.9a, the atoms would have to be moved farther apart,
Fig. 3.9b. This would be resisted by the interatomic
Extension/mm bonds, many of which will have to be broken simultan-
Fig. 3.7 eously.

CED BE
(a) (b)
Fig. 3.9
Deformation and dislocations
The deformation behaviour of materials can be
Calculations based on the known strength of bonds
explained at the atomic level. show that the stresses needed to produce slip in this
(a) Elastic strain. This is due to the stretching of the way are many times greater than those which actually
interatomic bonds that hold atoms together. The atoms cause plastic strain. The problem is, therefore, not so
are pulled apart very slightly; each is displaced a tiny much to explain the strength of metals as their weakness.
distance from its equilibrium position and the material This led to a search for defects in crystal structures and
lengthens. Hooke’s law is a result of the fact that the in 1934 G. I. Taylor of Cambridge University proposed
‘interatomic force-separation’ graph, Fig. 3.8a, is a the dislocation as one such defect.
straight line for atomic separations close to the equili- The idea is that occasionally, due perhaps to growth
brium separation 7, Fig. 3.8). faults during crystallization, there is an incomplete plane
32 VOLUME I Materials and Mechanics

Fig. 3.10

of atoms (or ions) in the crystal lattice, for example The first and most direct evidence for the existence
AB in Fig. 3.10a. We shall now see how the movement of dislocations was obtained in 1956 by J. W. Menter,
of the dislocation produces the same effect as a plane also of Cambridge University, using an electron micro-
of atoms slipping over other planes, but much more scope. An electron micrograph (an electron microscope
easily. photograph) is shown in Fig. 3.12 of an aluminium—
If a stress is applied as shown by the arrows, atom copper alloy in which the planes of atoms are spaced
B, whose bonds have already been weakened by the about 0.20 nm apart. A dislocation can be seen; the extra
distortion of the structure, moves a small distance to plane of atoms ends in the white circle and distorts the
the right and forms a bond with atom C, Fig. 3.10b. arrangement of the surrounding planes.
Plane of atoms DE is now incomplete. Then D flicks Dislocations can be obtained in a bubble raft (p. 21)
over and joins with F leaving GH as the incomplete and made to move if the raft is squeezed between two
plane, Fig. 3.10c. The result is just the same as if half- glass slides dipping into the ‘soap’ solution. There is
plane AB had slipped over the planes below BDG to one to be seen in Fig. 3.13.
the surface of the crystal. This process would have
involved breaking a great many bonds at the same time.
Instead, the dislocation, by moving a single line at a Strengthening metals
time, has broken many fewer bonds and required a much
smaller stress to do it. No atom has in fact moved more Pure metals produced commercially are generally too
than a small fraction of the atomic spacing. Plastic defor- weak or soft to be of much mechanical use—a rod of
mation by this mechanism is clearly only possible in pure copper the thickness of a pencil is easily bent by
the well-ordered structure of a crystalline material. hand. Their weakness can be attributed to the fact that
The passage of a dislocation in a crystal is like the they contain a moderate number of dislocations which
movement of a ruck in a carpet. A greater force is can move about easily in the orderly crystal structure,
needed to drag one carpet over another by pulling one thus allowing deformation under relatively small
end of it, than to make a ruck in the carpet and kick stresses. The traditional methods of making metals
it along, Fig. 3.11. Calculations confirm that the stresses stronger and stiffer all involve obstructing dislocation
to make dislocations move in metals are in good agree- movement by ‘barriers’, i.e. pockets of disorder in the
ment with their measured plastic flow stresses. lattice. Three barriers will be considered.
(a) ‘Foreign’ atoms. In an alloy such as steel ‘foreign’
atoms (e.g. carbon) are introduced into the lattice of
iron, disturbing its perfection and opposing dislocation
motion. This makes for greater strength and stiffness.
(b) Other dislocations. One problem with the disloca-
tion model is that when the dislocations have slipped
out of the crystal, as in Fig. 3.10c, the crystal is then
perfect and should have its theoretical strength. In
general this is not observed and it would seem that
further dislocations are generated whilst slip is occur-
ring. This view can be justified in a more advanced treat-
ment but here it is sufficient to say that as the metal
Fig. 3.11 is submitted to further stress, dislocations are created,
 MECHANICAL PROPERTIES 33

Re me Sti amie Niacin

BE
AO ARE, EW ET le ilps
VM AB Mg i, RAL EF Loteh Milage

i eee
on eee a
Bene Ant
Te: Reale t
BO EE COMES sinh
} INS gsc A ites Sao: pag
ine ke ee ee exes
OS MEY taygh eign § SPhe4 bE PME LEI 8G
9 OREM AE BE, Seat 5 PE Meee
atesSAS

A PEM. BELL SOBling

OE MES Sli Sg ia gt lly A to: Satna
hi Bh

le ee ee ee ee
Sic pal
Ae BY Stet EP Ath NEE Rah te ae etn
AS i let ts 6 Mp GM AP ep
EMA SEI GE A Nat ale GO By NySet: te 8
a ee a

OM eo Mealy sauabedeeecmantentneuet :
SBE CHC AIG, POG ces /
GPER CTW Ma Ry hte Males

Fig. 3.12

Fig. 3.13

move, meet and thereby obstruct each other’s progress. (c) Grain boundaries. In practice most metal samples
A ‘traffic jam’ of dislocations builds up. Such stress are polycrystalline, i.e. consist of many small crystals
occurs when a metal is hammered, stretched, bent, etc. or grains at different angles to each other (see Fig. 2.7,
and work-hardens. p. 18). The boundary between two grains is imperfect
On the other hand, in the process of annealing, in and can act as an obstacle to dislocation movement.
which a metal is heated and cooled slowly so that it In general, the smaller the grains the more difficult it
is softer and easier to work, dislocations can disappear. isto deform the metal. Why?
This may be due either to dislocations reaching the sur-
face of the metal and forming a ‘step’ there, or to differ- An obvious way of strengthening metals would be
ent types of dislocation meeting, cancelling each other to eliminate dislocations altogether and produce in
out to give a perfect lattice. effect perfect crystals. So far this has only been possible
34 VOLUME Materials and Mechanics

for tiny, hairlike single-crystal specimens called ‘bending’ it, as in Fig. 3.15—with the notch on the far
‘whiskers’ that are only a few micrometres thick and side of the rod. Why?
are seldom more than a few millimetres long. Their Round a scratch, notch or crack there is a concen-
strength, however, approaches the theoretical value and tration of stress which in general is greater the smaller
they can withstand elastic strains of 4 or 5% (compared the radius of curvature of the tip of the crack (i.e. the
with 3% or less for most common engineering mater- sharper the crack). Such stress concentrations may be
ials). Unfortunately, perhaps due to surface oxidation, seen by viewing a lamp through two ‘crossed’ Polaroid
dislocations soon develop and the ‘whisker’ weakens. squares (i.e. one*square is rotated to cut off most of
At present they are the subject of much research. Fig. the light coming from the other to give a dark field
3.14 shows ‘whiskers’ of pure iron magnified four times, of view) having a strip of polythene between them, Fig.
each one virtually a perfect crystal, and having about 3.16a. Pulling the strip causes colours to appear and
fifty times the tensile strength of the same thickness is an indication that the polythene is under stress. (The
of ordinary soft iron. phenomenon is called photoelasticity and is used to study
stresses in plastic models of engineering structures. Fig.
3.16b shows these round a triangular hole in a plastic
block under pressure.) When the strip is cut halfway

Projector Crossed Polaroids

Strip of stretched polythene

(a) -

Fig. 3.14

Cracks and fracture

Cracks, both external and internal and however small,
play an important part in the fracture of a material and
prevent it displaying its theoretical strength. Different
types of fracture usually occur in brittle and ductile
materials.
(a) Brittle fracture. This happens after little or no
plastic deformation (i.e. during elastic deformation) by
the very rapid propagation of a crack. It takes place,
for example, when a glass rod is cut by making a small
but sharp notch on it with a glass knife or file and then

Notch in glass
Opposite thumbs

Fig. 3.15 Fig. 3.16

 MECHANICAL PROPERTIES 35

across and again pulled, colours are seen at the tip of tern of the air jets used for cooling the prestressed glass
the cut showing the stress is high there. If it is sufficiently of a car windscreen can be seen through polarizing spec-
high, interatomic bonds are broken, the cut spreads and tacles or sometimes by sunlight that has been partially
breaks a few more bonds at the new tip. Eventually polarized by reflection from the car bonnet. Prestressing
complete fracture occurs. glass in this way can increase its toughness 3000 times.
Even tiny surface scratches, and they seem to arise
inevitably on all materials, can lead to fracture. For (b) Ductile fracture. In this case fracture follows
example, a freshly-drawn 3-metre-long glass fibre can appreciable plastic deformation, by slow crack propaga-
be bent into an arc but it fractures if ‘scratched’ at A, tion. After thinning uniformly along its length during
Fig. 3.17, by gently stroking a few times with another the plastic stage, the specimen develops a ‘waist’ or
glass fibre. It is less likely to break if scratched at B. ‘neck’ in which cavities form. These join up into a crack
Why? and this travels out to the surface of the specimen, Fig.
3.19a, to give ‘cup and cone’ shaped, dull fracture sur-
faces, like those in Fig. 3.19b for an aluminium rod.
It is possible that the internal cavities are formed during
the later stages of plastic strain when stress concen-
Glass fibre trations arise in regions having a large number of inter-
locking dislocations.

{ {
Clam
“ Glass rod |
Ductile
Fig. 3.17 material

In brittle materials like concrete and glass, cracks | | E

spread more readily when the specimen is stretched or
bent, i.e. is in tension. Crack propagation is much more
difficult if such materials are used in compression (i.e.
squeezed) so that any cracks close up, Fig. 3.18. Thus
prestressed concrete contains steel rods that aré in ten-
sion because they were stretched whilst the concrete
was poured on them and set. As well as providing extra
tensile strength these keep the concrete in compression
even if the whole prestressed structure is in tension.

Crack ;
Compression

Fig. 3.19
Brittle material Crack

Fig. 3.18 Cracks propagate with much greater difficulty in

metals due to the action of dislocations. These can move
to places of high stress such as the tip of a crack, thus
Glass can also be prestressed and its surface com- reducing the effective stress by causing it to be shared
pressed, making it more resistant to crack propagation. among a greater number of interatomic bonds, Fig. 3.20.
In thermal toughening, jets of air are used to cool the The crack tip is thereby deformed plastically, blunted
hot glass and cause the outside to harden and contract by the dislocation and further cracking possibly stopped.
whilst the inside is still soft. Later the inside contracts, For the same reason surface scratches on metals have
pulls on the now reluctant-to-yield surface which is thus practically no effect. In brittle materials dislocation
compressed. As we saw previously stresses in transpar- movement is impossible (why?) and high local stresses
ent materials are revealed by polarized light. The pat- can build up at cracks under an applied force.
36 VOLUME I Materials and Mechanics

High stress at tip of | Crack Tip of crack blunted

crack may break bond | Crack by dislocation .’.stress
reduced and bond
restored

Dislocation Dislocation moves

towards crack
Fig. 3.20

Fatigue and creep surface (as on right of photograph), until it was halfway
across the shaft which then broke suddenly. Stress con-
These are two other important aspects of the mechanical
centrations may be due to bad design (e.g. rapid changes
behaviour of metals that are conveniently considered
of diameter), bad workmanship (e.g. a tool mark) and
here.
are common at holes—the Comet aircraft disasters of
(a) Fatigue. This may cause fracture, often with little 1954 were caused by fatigue failure started round small
or no warning, and happens when a metal is subjected rivet holes.
to a large number of cycles of varying stress, even if For many ferrous metals there is a safe stress variation
the maximum value of the stress could be applied stea- below which failure will not occur even for an infinite
dily with complete safety. It is estimated that about 90% number of cycles. With other materials ‘limited-life’
of all metal failures are due to fatigue; it occurs in air- design only is possible. As yet no fully comprehensive
craft parts, in engine connecting rods, axles, etc. theory of fatigue exists.
‘ A typical fatigue fracture in a steel shaft is shown
(b) Creep. In general this occurs at high temperature
in Fig. 3.21; starting as a fine crack, probably at a point
and results in the metal continuing to deform as time
of high stress, it has spread slowly, producing a smooth
passes, even under constant stress. The effect is thought
to be associated with dislocation motion due to the
vibratory motion of atoms, and in the creep-resistant
alloys that have been developed this motion is restricted.
Such alloys are used, for example, to make the turbine
blades of jet engines where high stress at high tempera-
ture has to be withstood without change of dimensions.
Some low melting-point metals can creep at room
temperature, thus unsupported lead pipes gradually sag
and the lead sheeting on church roofs has to be replaced
periodically.

Composite materials
Composites are produced by combining materials so
that the combination has the most desirable features
of the components. The idea is not new. Wattle and
daub (interlaced twigs and mud) have been used to build
homes for a long time; straw and clay are the ingredients
of bricks; Eskimos freeze moss into ice to give a less
brittle material for igloo construction; reinforced con-
crete contains steel rods or steel mesh. In all cases the
composite has better mechanical properties than any
Fig. 3.21 of its components.
 MECHANICAL PROPERTIES 37
The production of composites is man’s attempt to acting as a barrier to the crack and deflecting it harm-
copy nature. Wood is a composite of cellulose fibres lessly along the interface it forms with the fibre, Fig.
cemented together with lignin. Bone is another compos- 3.22. A plastic resin or a ductile metal makes a suitable
ite material. Many modern technological applications matrix. Fibre-reinforced composites are strong to
require materials that are strong and stiff but light and stresses applied along the fibres.
heat-resistant. The development of composites to meet
these requirements is at present a major concern of
materials scientists throughout the world and offers Crack
exciting possibilities to engineers in the future.
The highest strength-to-weight and stiffness-to-weight
Fibre Matrix Crack deflected
ratios are possessed not by metals but by materials such Interface along interface
as glass, carbon and boron whose atoms are linked by
Fig. 3.22
many strong covalent bonds. (The strength of covalent
bonds and their number per unit volume in these cova-
lent solids accounts for the high strength and stiffness;
the directional nature of the bond explains the non- Fibreglass was the first of the successful modern com-
closed-packed structure and consequently small density, posites. It consists of high-strength glass fibres in a plas-
see p. 19.) Unfortunately these materials are brittle, tic resin (called glass-reinforced plastic—GRP) and is
partly because their structures make dislocation motion used for making boat hulls, storage tanks, pipes, car
difficult under an applied stress and partly because they components. Fig. 3.23 shows H.M.S. Bicester, one of
usually have small surface scratches that develop into the Hunt-class Mine Countermeasures Vessels, believed
cracks. to be the largest ships built of GRP afloat. The main
In modern composite materials the desirable proper- reason for using GRP for these vessels is that it is non-
ties of covalent solids are exploited by incorporating magnetic (unlike steel). Its other advantages are that
them as fibres in a weaker, yielding material called the it does not corrode, rot, warp, shrink or split, and is
matrix. Freshly drawn fibres are fairly scratch-free and resistant to marine borers. Also, it is fire-resistant and
are therefore strong. The matrix has three functions: needs three times less maintenance than steel or wooden
first, it has to bond with and hold the fibres together hulls. Lightweight lift-jet engines for vertical take-off
so that the applied load is transmitted to them; second, also use GRP extensively for low-temperature parts and
it must protect the surface of the fibres from scratches; produce a thrust sixteen times their own weight. The
third, if cracks do appear it should prevent thenr from best engines in commercial service today produce a
spreading from one fibre to another—it can do this by thrust less than five times their weight.
38 VOLUME I Materials and Mechanics

Carbon-fibre-reinforced plastics (CFRP) are similar to the aircraft industry. They are not subject to fatigue
but carbon fibres (about 6 x 10° per cm? of cross-section) failure or to high-stress concentrations round holes and
of greater strength and stiffness replace glass fibres. cracks as metals are. They do not, however, flow plasti-
These are stronger and stiffer than steel and much lighter cally like metals but are elastic until failure, when exten-
(see question 7, p. 43). At present they are costly to sive damage may occur. Good design is therefore
produce; nevertheless CFRP are particularly attractive necessary to avoid overloading. Their resistance to cor-

Fibre Resin

Fig. 3.24
 MECHANICAL PROPERTIES 39
rosion is also poor and prohibits certain applications. If Hooke’s law is obeyed, OA is a straight line (as shown)
Fig. 3.24a shows a bundle of carbon fibres; Fig. 3.24b and OAB is a triangle.
shows a section of a piece of CFRP. Therefore the strain energy in the wire for the whole
Much research is taking place on composites. The extension e and final stretching force F, is given by
interface between a fibre and its matrix holds the key
to the formation of a successful composite; at present strain energy = area AOAB
a great deal is not understood about the properties of = AB X OB
the interface. If the fibres are not covered uniformly
Fe
with the matrix, small holes form at the interface. In bole

service these cause high-stress concentrations and pre- The expression 3Fe gives the strain energy in joules if
mature failure. They may also allow liquids and gases Fis in newtons and e in metres.
to penetrate the composite and attack the fibres. The If / is the original length of the wire and A its cross-
effect of coating the fibre with some other material to section area, then volume of wire = Al.
protect its surface from scratches, corrosion, etc., before
Strain energy per unit volume = }Fe/(Al)
combining it with the matrix is being studied. Fig. 3.24c
is a scanning electron microscope photograph of a tung-
sten-coated carbon fibre (x 3000).
The fibre-reinforcement of metals and concrete is also
being investigated. But F/A = stress and e// = strain.
‘. Strain energy per unit volume = 3(stress X strain)
This is the area under the stress-strain graph.

Strain energy
Energy has to be supplied to stretch a wire. If the stretch-
ing force is provided by hanging weights, there is a loss
of potential energy, some of which is stored in the
stretched wire as strain energy. Provided the elastic limit
is not exceeded this energy can usually be recovered
completely (rubber is a notable exception as we shall
see in the next section). If it is exceeded, the part of
the energy used to cause crystal slip (i.e. plastic strain)
is retained by the wire.
We need to determine the strain energy stored in a
wire stretched by a known amount. Consider a material
with a force—extension graph like that in Fig. 3.25. (This
{ Extension
is of the same form as its stress-strain graph.)
Suppose the wire is already extended by e, and then
suffers a further extension Se, which is so infinitesimally Fig. 3.25
small that the shaded area is near enough a rectangle.
If F, is the average but nearly constant value of the
stretching force during the extension 5e,, then the
energy transferred to the wire, i.e. the work done 6W,,
If Hooke’s law is not obeyed so that OA is not a
is given by
straight line, the value of the strain energy in the wire
dW, = force x distance for extension e is still area OAB. If the wire suffers
= F, x 5e;
plastic deformation, e.g. an extension OD, the total
strain energy stored is again the area under the force—
=‘area of shaded strip extension graph, i.e. area OCD, and can be calculated
by counting the squares on the graph paper, knowing
Total work done during whole extension e
the ‘energy value’ of one square.
= area OAB
The term specific energy is also used and is the energy
i.e. Strain energy stored in wire = area OAB stored in the wire per unit mass.
40 VOLUME I Materials
and Mechanics

H CH; H H!/H CH; H H|H CH;H H

Sesniiae Oe | ea ee
! He Hale :
lsoprene monomer

Fig. 3.26 (C,H)

Fig. 3.28
 MECHANICAL PROPERTIES 41
Rubber stiffer at large extensions. It has been shown (p. 39)
that the area enclosed by OABC and the strain axis
The two most striking mechanical properties of rubber represents the energy supplied to cause stretching; simi-
are (a) its range of elasticity is great—some rubbers larly the area under CDEO represents the energy given
can be stretched to more than ten times their original up by the rubber during contraction. The shaded area
length (i.e. 1000% strain) before the elastic limit is is called a hysteresis loop; it is a measure of the energy
reached and (b) its value of the Young modulus is about ‘lost’ as heat during one expansion-contraction cycle.
10* times smaller than most solids and increases as the The changes in temperature can be felt by placing a
temperature rises, an effect not shown by any other wide rubber band on the lips; if it is stretched quickly
material. the temperature rises due to the transfer of mechanical
Rubber is a polymer consisting of up to 10* isoprene energy to the rubber. When released under control the
molecules (C;Hg) joined end-to-end into a long chain temperature falls. (The temperature change on free con-
of carbon atoms, Fig. 3.26. The enormous extensibility traction is small.)
and low value of E cannot be due to the stretching of
the strong covalent bonds between atoms in the carbon
chain. ;
If a sample of stretched rubber is ‘photographed’ by
a beam of high-energy electrons, sharp spots are
obtained, Fig. 3.27a, like those produced by X-rays and
Stress
a crystal. This suggests there is some order among the
molecules in such a sample. Fig. 3.275 is a similar photo-
graph of unstretched rubber. A plausible explanation
of the behaviour of rubber might be that its long-chain Strain
molecules are intertwined and jumbled up like cooked Fig. 3.29
spaghetti. Fig. 3.28 is a model of one rubber molecule.
Rubber with a hysteresis loop of small area is said
A stretching force would tend to make the chains uncoil
to have resilience. This is an important property where
and straighten out into more or less orderly lines along-
the rubber undergoes continual compression and relaxa-
side each other. When the force is removed they coil
tion as does each part of a car tyre when it touches
up again. There is also some cross-bonding between
the road and rotates on. If the rubber used in tyres
chains, achieved during manufacture by vulcanizing raw
does not have high resilience there is appreciable loss
_tubber (see p. 25); this cross-linking as well as causing of energy resulting in increased petrol consumption or
stiffening also greatly increases the reversible strain pos-
lower maximum speed. Should the heat build-up be
sible by anchoring together the long molecules. Fully
large the tyre may disintegrate.
extended rubber is strong because the bonds are then When rubber is stretched and released there may be
stretched directly. a small permanent set as shown by the dotted line EF
The rise in value of E with temperature can be attri- in Fig. 3.29.
buted to the greater disorder among the chains when
the material is heated; their resistance to alignment by
a stretching force therefore increases.
If a stress-strain curve is plotted for the loading and Elastic moduli
unloading of a piece of rubber the two parts do not
All deformations of a body whether stretches, compres-
coincide, Fig. 3.29. OABC is for stretching and CDEO
sions, bends or twists can be regarded as consisting of
for contracting. The strain for a given stress is greater
one or more of three basic types of strain. For many
when unloading than when loading. The unloading
materials experiment shows that provided the elastic
strain can be considered to ‘lag behind’ the loading
strain; the effect is called elastic hysteresis. It occurs with
limit is not exceeded
other substances, noticeably with polythene and glasses stress
and to a small extent with metals. =aconstant
strain
We note that Hooke’s law is not obeyed. This is typi-
cal of non-crystalline polymers and contrasts with the This is a more general statement of Hooke’s law. The
usually linear behaviour of crystalline materials, e.g. constant is called an elastic modulus of the material for
metals which obey the law during elastic strain. It is the type of strain under consideration. There are three
also evident that rubber stretches easily at first but is moduli, one for each kind of strain.
42 VOLUME 1 Materials and Mechanics

(a) The Young modulus (E). This has already been

considered (p. 29) and is concerned with change of
length strains. It is defined by
tensile stress
tensile strain

where stress is force per unit area (F/A) and strain is

change of length per unit length (e//).
Fig. 3.32
(b) Shear modulus (G). In this case the strain involves
a change of shape without change of volume. Thus if is a change of volume without a change of shape. The
a tangential force F is applied along the top surface bulk stress is 5p, i.e. increase in force per unit area,
of area A of a rectangular block of material fixed to and the bulk strain is 8V/V, i.e. change of volume/
the bench, the block suffers a change of shape and is original volume; the bulk modulus K is defined by
deformed so that the front and rear faces become paral-
_ bulkstress _ —8p
lelograms, Fig. 3.30. The shear stress is F/A and angle
a is taken as a measure of the strain produced. (The ~ bulkstrain 8V/V
force F on the bottom surface of the block is exerted
by the bench.) The shear modulus is defined as Spare5
5V
shear stress F
The negative sign is introduced to make K positive since
shear strain Aa SV, being a decrease, is negative.
Note. 5 (pronounced ‘delta’) is the Greek letter d
and when used as a prefix to the symbol for a quantity
it indicates a change in that quantity is being considered.
Solids have all three moduli, liquids and gases only
K. All moduli have the same units—pascals (Pa).

Fig. 3.30

When a wire is twisted, a small square on the surface QUESTIONS

becomes a rhombus, Fig. 3.31, and is an example of 1. (a) Why are stresses and strains rather than forces and
a shear strain. G can be found from experiments on extensions generally considered when describing the deforma-
the twisting of wires. If a spiral spring is stretched, the tion behaviour of solids?
wire itself is not extended but is twisted, i.e. sheared. (6b) A length of copper of square cross-section measuring
The extension thus depends on the shear modulus of 1.0mm by 1.0mm is stretched by a tension of 40N. What is
the material as well as on the dimensions of the spring. the tensile stress in Pa?
(c) If the breaking stress of steel is 1.0 x 10° Pa will a wire
Wire
of this material of cross-section area 4.0 x 10-*cm? break
when a 10 kg mass is hung from it? (g = 10ms~?)
(d) A strip of rubber 6cm long is stretched until it is 9cm
long. What is the tensile strain in the rubber as (i) a ratio,
(ii) a percentage?
(e) A wire originally 2m long suffers a 0.1% strain. What
is its stretched length?

Fig. 3.31 7 2. (a) The Young modulus for steel is greater than that for
brass. Which would stretch more easily? Which is stiffer?
(b) How does a deformed body behave when the deforming
force is removed if the strain is (i) elastic, (ii) plastic?
(c) Bulk modulus (K). If a body of volume V is sub- (c) A brass wire 2.5m long of cross-section area
jected to an increase of external pressure 8p which 1.0x 10-7cm? is stretched 1.0mm by a load of 0.40 kg.
changes its volume by 8V, Fig. 3.32, the deformation Calculate the Young modulus for brass. (Take g = 10ms~?.)
 MECHANICAL PROPERTIES 43

Fig. 3.33
aoe: pee
Stress

Strain

(a) (b) (c) (d)

What percentage strain does the wire suffer? Use the value (c) In structures like aircraft the designer is concerned with
of E to calculate the force required to produce a 4.0% strain obtaining as much strength and stiffness as possible for a given
in the same wire. Is your answer for the force reliable? If it weight. Using the values in the table derive measures of ‘stiff-
isn’t, would it be greater or less than your answer? Explain. ness-to-weight’ and ‘strength-to-weight’ for the four materials
listed.
3. Stress-strain curves for four different materials are shown
in Fig. 3.33. Describe what you would feel if you pulled a Material The Young Ultimate tensile Relative
specimen of each. modulus strength density
E/10" Pa
4. (a) A 0.50kg mass is hung from the end of a wire 1.5m aluminium
long of diameter 0.30 mm. If the Young modulus for the mater- carbon fibres
ial of the wire is 1.0 x 10!!Pa, calculate the extension pro- glasses
duced. (Take g= 10ms~*.) steel
(b) Two wires, one of steel and one of phosphor bronze,
each 1.5 metres long and of diameter 0.20cm, are joined end 8. Fig. 3.34 shows the force—-extension graph for a length of
to end to form a composite wire of length 3.0 metres. What steel wire A. The wire obeys Hooke’s law over the range of
tension in this wire will produce a total extension of 0.064 cm? extensions considered.
(The Young modulus for steel = 2.0 x 10!! Pa and for phosphor (a) Ifthe wire has a diameter of 0.40 mm and its unstretched
bronze = 1.2 x 10" Pa.) (L. part qn.) length is 2.0 m, calculate the Young modulus of the steel.
(b) A second wire B is made of the same steel. It has the
5. Write a short essay on ‘strong’ materials. Your essay might same unstretched length as A, but twice the diameter. Draw
include discussion of such topics as: the need for strong mater- accurately on a copy of Fig. 3.34 the force—-extension graph
- ials, why some materials are weak, how the weaknesses can for this wire. Label your graph B.
be avoided, composite materials such as plywood, reinforced (c) A student discovers that for a given force within the
concrete and fibre-glass, and the new technology of composite range considered above, the elastic energy stored by wire A
materials. These are only suggestions: you might write about is four times greater than that stored by wire B. Use the two
a few of them, or, if you prefer, about all of them, and you graphs, or some other theoretical argument, to explain this
may of course discuss other related issues if you wish. fact.
(O. and C. Nuffield) (d) If the breaking stress of the steel is 5.0 x 108N m~, cal-
culate the maximum force which can be applied longitudinally
6. Define stress, strain, the Young modulus. to wire A. (H.K.A.L.E., 1984)
Describe in detail how the Young modulus for a steel wire
may be determined by experiment.
A vertical steel wire 350cm long, diameter 0.100cm, has
a load of 8.50 kg applied at its lower end. Find (a) the extension,
(b) the energy stored in the wire. (Take the Young modulus
for steel as 2.00 x 10!!Pa and g= 9.81 ms *.) (LE)
fop)

7. Approximate values for the Young modulus, ultimate ten-

sile strength and relative density of some materials are given
Force/N
in the table. <a
(a) Explain the meaning and significance of each of these
three terms.
(b) The tensile strength of glass equals that of aluminium. 0 0.1 OA We O74) “OS OG Wie WOES OW) ake
Why isn’t glass used for the same kind of constructional applica- Extension/mm
tions as aluminium? Fig. 3.34
4 Electrical properties
Conduction in solids Electrical strain gauge
Current and charge Effects of temperature on resistance
Drift velocity of electrons Electromotive force
Potential difference Internal resistance
Resistance Kirchoff’s laws
Current-p.d. relationships Power and heating effect
Types of resistor Wheatstone bridge
Resistor networks Potentiometer
Using ammeters and voltmeters Electrolysis
Shunts, multipliers and multimeters Electric cells
Rough calibration of a voltmeter Thermoelectric effect
Resistivity

Conduction in solids lattice but are not free to undergo translational motion.
As a result the electrons lose kinetic energy and slow
Materials exhibit a very wide range of electrical conduc-
down whilst the ions gain vibrational energy. The net
tivities. The best conductors (silver and copper) are over
effect is to transfer chemical energy from the battery,
1073 times better than the worst conductors, i.e. the best
via the electrons, to internal energy (see p. 76) of the
insulators (e.g. polythene). Between these extreme
ions. This shows itself on the macroscopic scale as a
cases is the now important group of semiconductors
temperature rise in the conductor and subsequently
(e.g. germanium and silicon).
energy may pass from the conductor to the surroundings
The first requirement for conduction is a supply of
as heat. The electrons are again accelerated and the
charge carriers that can wander freely through the
process is repeated.
material. In most solid conductors, notably metals, we
The overall acceleration of the electrons is zero on
believe that the carriers are loosely held outer electrons.
account of their collisions: They acquire a constant aver-
With copper, for example, every atom has one ‘free’
age drift velocity in the direction from negative to posit-
electron (the one that helps to form the metallic bond,
ive of the battery and it is this resultant drift of charge
p. 16) which is not attached to any particular atom and
that is believed to constitute an electric current. An
sO can participate in conduction. On the other hand
analogous situation may arise when a ball rolls down
if all electrons are required to form the bonds (covalent
a long flight of steps. The acceleration caused by the
or ionic) that bind the atoms of the material together
earth’s gravitational field when the ball drops can be
then the material will be an insulator. In semiconductors
cancelled by the force it experiences on ‘colliding’ with
only a small proportion of the outer electrons are ‘free’
the steps. The ball may roll down the stairs with zero
to move.
average acceleration, i.e. at constant average speed.
The ‘free’ electrons in a solid conductor are in a state
The ‘free’ electron theory is able to account in a
of rapid motion, moving to and fro within the crystal
general way for many of the facts of conduction and
lattice at speeds calculated to be about 1/1000 of the
although in more advanced work it has been extended
speed of light. This motion is normally completely
by the ‘band’ theory it will be adequate for our present
haphazard (like that of gas molecules) and as many elec-
purposes.
trons with a given speed move in one direction as in
the opposite direction with the same speed. There is,
therefore, no net flow of charge and so no current.
Current and charge
If a battery is connected across the ends of the conduc-
tor, an electric field is created in the conductor which In metals, current is the movement of negative charge,
causes the electrons to accelerate and gain kinetic i.e. electrons; in gases and electrolytes both positive
energy. Collisions occur between the accelerating elec- and negative charges may be involved. Under the action
trons and atoms (really positive ions) of the conductor of a battery, charges of opposite sign move in opposite
that are vibrating about their mean position in the crystal directions and so a convention for current direction has
44
 EGE CERICALZP ROPER TUBS, , 145

to be chosen. As far as most external effects are con- n free electrons per unit volume
cerned, positive charge moving in one direction is the
same as negative charge moving in the opposite direc-
tion. By agreement all current is assumed to be due
Conductor
to the motion of positive charges and when current
arrows are marked on circuits they are directed from
Fig. 4.1
the positive to the negative of the supply. If the charge
carriers are negative they move in the opposite direction
to that of the arrow.
The basic electrical unit is the unit of current—the
Suppose that a battery across the ends of the conduc-
ampere (abbreviated to A); it is defined in terms of the
magnetic effect of a current. The unit of electric charge, tor causes the charge Q to pass through length / in time
the coulomb (C), is defined in terms of the ampere. t with average drift velocity v. The resulting steady cur-
rent / is given by
One coulomb is the quantity of electric charge carried
past a given point in a circuit when a steady current of
1 ampere flows for 1 second.
_ nAle
If 2 amperes flow for 1 second, 2 x 1 coulombs
(ampere-seconds) pass; if 2 amperes flow for 3 seconds ae,
then 2 x 3 coulombs pass. In general if a steady current But v = //t, therefore ¢ = //v.
I (in amperes) flows for time f (in seconds) then the
quantity Q (in coulombs) of charge that passes is given
by
Q=It
The flow of charge in a conductor is often compared
with the flow of water in a pipe. The flow of water in To obtain a value for v consider a current of 1.0
litres per second say, corresponds to the flow of charge ampere in SWG28 copper wire of cross-section area
in coulombs per second, i.e. amperes. 1.1 x 107’ square metre. If we assume that each copper
The charge on an electron, i.e. the electronic charge, atom contributes one ‘free’ electron, it can be shown
is 1.60 x 10-!°C and is much too small as a practical (see question 2a, p. 69) that n ~ 10” electrons per cubic
‘unit. In 1C there are therefore 1/(1.60 x 107%), i.e. metre. Then, since e = 1.6 x 10~'? coulomb (the charge
6.24 x 108 electronic charges. A current of 1 A is thus on an electron),
equivalent to a drift of 6.24 x 10'8 electrons past each
point in a conductor every second. =<
ik
Smaller units of current are the milliampere (107° A), nAe
abbreviated to mA, and the microampere (107° A),
abbreviated to pA.
7 (1.0A)
~ (1029 m=) x (1.1 x 1077 m2) x (1.6 x 10-8C)
v 1.0 A
Drift velocity of electrons ~ 102° 1.1 x 10-7X 1.6107 m 3m2C
On the basis of the ‘free’ electron theory an expression Cs!
= 6 x 10-4 — (1A =1Cs"!)
can be derived for the drift velocity of electrons in a C
current and an estimate made of its value. The results
=6x104*ms !
are surprising.
Consider a conductor of length / and cross-section =(0.6mms~!
area A having n ‘free’ electrons per unit volume each
This is a remarkably small velocity and means that
carrying a charge e, Fig. 4.1.
it takes electrons about half an hour to drift 1m when
Volume of conductor All a current of 1 A flows in this wire. The tiny drift velocity
Number of ‘free’ electrons =nAl of electrons contrasts with their random speeds due to
Total charge Q of ‘free’ electrons = nAle their vibrational motion (about 1/1000 of the speed of
46 VOLUME I Materials and Mechanics

light) and is not to be confused with the speed at which The potential difference between two points in a circuit
the electric field causing their drift motion travels along is the amount of electrical energy changed to other forms
a conductor. This is very great and is nearly equal to of energy when unit charge passes from one point to
the speed of light, i.e. 3 x 10®ms~! (see Appendix 7). the other.
Current therefore starts to flow almost simultaneously
The unit of potential difference is the volt (V) and
at all points in a circuit.
equals the p.d. between two points in a circuit in which
The same expression for drift velocity holds for charge
1 joule of electrical energy is converted when 1 coulomb
carriers other than electrons (in fact, it holds for the
passes from one point to the other. If 2 joules are con-
transport of other things as well as electric charge). In
verted per coulomb then the p.d. is 2 volts. If the passage
an electrolyte, conduction is due to ions and using the
of 3 coulombs is accompanied by the conversion of 9
arrangement of Fig. 4.2 information can be obtained
joules of energy, the p.d. is 9/3 joules per coulomb,
about their motion. On applying an electric field (from
i.e. 3 volts.
a 250 volt d.c. supply), the purple stains from the per-
It therefore follows that if the p.d. between two points
manganate crystals travel very slowly towards the posit-
A and B (we more commonly talk about the p.d. across
ive of the supply, and if we make the not unreasonable
AB) is 5 volts then when 4 coulombs pass from A to
assumption that the stain travels with the charge carriers
B, the electrical energy changed will be 5 joules per
then ions too would appear to have tiny drift velocities
coulomb, i.e. 5 X 4 joules. In general if a charge of Q
of a similar value to those calculated for electrons.
(in coulombs) flows in a part of a circuit across which
there is a p.d. of V (in volts) then the energy change
Strip of filter paper soaked in bench W (in joules) is given by
ammonium hydroxide solution
Crystals of potassium permanganate
W=OQV
Crocodile
If Q is in the form of a steady current J (in amperes)
clip
flowing for time ¢ (in seconds) then Q = Jt and
W=ItV
Microscope slide To HT.+
Although it is always the p.d. between two points
Fig. 4.2 which is important in electric circuits there are some
occasions when it is helpful to consider what is called
the potential at a point. This involves selecting a conven-
Despite the slow movement of the carriers in conduc- ient point in the circuit and saying it has zero potential.
tors and their very small charge, large currents are pos- The potentials of all other points are then stated with
sible. Why? reference to it, i.e. the potential at any point is then
Note. The term current density J is often used and the p.d. between the point and the point of zero poten-
is defined as the current per unit cross-section area at tial. In practice one part of a piece of electrical equip-
right angles to the direction of flow. It is given by ment (e.g. a power supply) is often connected to earth;
the earth and all points in the circuit joined to it are
I then taken as having zero potential.
J=—=nev
A If positive charge moves (i.e. conventional current
flows) from a point A to a point B then A is regarded
as being at a higher potential than B. Negative charge
flow is therefore from a lower to a higher potential,
i.e. from B to A. We can look upon p.d. as a kind
Potential difference of electrical ‘pressure’ that drives conventional current
from a point at a higher potential to one at a lower
In an electric circuit electrical energy is converted into
potential.
other forms of energy. A lamp converts electrical energy
into heat and light, and an electric motor converts elec-
trical energy into mechanical energy. Such energy con-
versions, produced by suitable devices, are a useful
Resistance
feature of electric circuits and form the basis of the defi-
nition of the term potential difference (p.d.)—an idea When the same p.d. is applied across different conduc-
that helps us to make sense of circuits. tors different currents flow. Some conductors offer more
 ELECTRICAL FROPERITIEg 47
opposition or resistance to the passage of current than in one direction only and is a non-ohmic or non-linear
others. conductor. This one-way property makes it useful as
The resistance R of a conductor is defined as the ratio a rectifier for changing alternating current (a.c.) to direct
of the potential difference V across it to the current I current (d.c.).
flowing through it. That is,
(c) Filament lamps. The I-V graph of, for example,
a torch bulb, Fig. 4.3c, bends over as V and J increase,
indicating that a given change of V causes a smaller
change in J at larger values of V. That is, the resistance
The unit of resistance is the ohm (symbol ©, the (V/1) of the tungsten wire filament increases as the cur-
Greek letter omega) and is the resistance of a conductor rent raises its temperature and makes it white-hot. In
in which the current is 1 ampere when a p.d. of 1 volt general, the resistance of metals and alloys increases
is applied across it. Larger units are the kilohm with temperature rise.
(10° ohm), symbol kQ, and the megohm (10° ohm), sym-
bol MQ. The ratio V/J is a sensible measure of the (d) Thermistors. These are made of semiconductors
resistance of a conductor since the smaller / is for a and the [-V graph of the commonest type bends
given V, the greater must be opposition of the conduc- upwards, Fig. 4.3d, i.e. their resistance decreases
tor, that is, the greater is R. sharply as their temperature rises.
The resistance of a metal can be regarded as arising
from the interaction which occurs between the crystal
lattice of the metal and the ‘free’ electrons as they drift Hf
through it under an applied p.d. This interaction is due
mainly to collisions between electrons and the vibrating
ions of the metal but collisions between defects in the
crystal lattice (e.g. impurity atoms and dislocations) also
play a part, especially at very low temperatures.
The conductance of a specimen is the reciprocal of
its resistance and is measured in siemens (S).
V
(b) 0

Current-p.d. relationships if if
Using one of the ammeter—voltmeter circuits described
later (Fig. 4.15), the p.d. V across a component can
be varied and the corresponding current J measured.
A graph of J against V shows the relationship between
these two quantities and is called the characteristic of
the component. It summarizes pictorially how the com-
0 V 0 V
ponent behaves.
(a) Metals and alloys. These give I-V graphs which (c) (d)
are straight lines through the origin, Fig. 4.3a, and are
called linear or ohmic conductors. For them J « V and
it follows that V/J =a constant (equal to the reciprocal
of the slope of the graph, i.e. by OB/AB). They obey
Ohm’s law, which states that the resistance of a metallic
Types of resistor
conductor does not change with p.d., provided the tem-
perature is constant. Conductors especially constructed to have resistance are
called resistors, denoted by +1; they are required
(b) Semiconductor diodes. The typical I-V graph in
for many purposes in electric circuits. Several types
Fig. 4.3b shows that current passes when the p.d. is
exist.
applied in one direction but is almost zero when its acts
in the opposite direction. A diode thus has a small resis- (a) Carbon composition resistors. These are made
tance if the p.d. is applied one way round but a very from mixtures of carbon black (a conductor), clay and
large resistance when the p.d. is reversed. It conducts resin binder (non-conductors) which are pressed and
48 VOLUME I Materials and Mechanics

moulded into rods by heating. The resistivity of the mix- Resin coating Metal end

ture depends on the proportion of carbon. The stability

of such resistors is poor and their values are usually
only accurate to within +10% but they are cheap, small
and good enough for many jobs. Three sizes are avail-
able with power ratings of 3, 1 and 2 watts. Carbon film on Spiral groove
Values are shown by colour markings as in Fig. 4.4. ceramic rod
The tolerance colours are gold +5%, silver +10%, no Fig. 4.5
colour +20%.
film is protected by an epoxy resin coating. The stability
2nd figure and hence accuracy of this type of resistor is commonly
1st figure Number of noughts
+2% and the power rating § to } watt.
Tolerance (accuracy)
(c) Wire-wound resistors. High-accuracy, high-stabi-
lity resistors are always wire-wound, as are those
required to have a large power rating (i.e. over 2 watts).
Red Violet Orange Silver They use the fact that the resistance of a wire increases
2 i 000 410% with its length. Manganin (copper, manganese, nickel)
27 000Q (27kQ) wire is used for high-precision standard resistors because
of its low temperature coefficient of resistance (p. 56);
constantan or Eureka (copper-nickel) wire is used for
several purposes and Nichrome (nickel-chromium) wire
Black for commercial resistors.
Brown Adjustable known resistances, called resistance
Red boxes, are used for electrical measurements in the labor-
Orange |} Colours atory. They consist of a number of constantan coils
Yellow |§ of
which can be connected in series by switches or plugs
Green spectrum
to give the required value, Figs 4.6 and 4.7. It is espe-
Blue
cially important with resistance boxes not to exceed the
Violet
Grey
maximum safe currents since overheating may change
OC
- White
OMANONFWNH
the resistance value or even burn out the coils. (The
power limit is about 1 watt per coil and so a 1 ohm
Fig. 4.4

Brass plug Brass block

This colour code is now being replaced by a code HOF ioe! w™‘\
with simpler markings:
pe

5 / 10 Sf 2 50
Resistance coils
Fig. 4.6 Plug type (reading =65Q)
68kQ | 100kQ 6.8MQ
Mio Rotary switch

Tolerances are indicated by adding a letter; F=+1%,

G=+2%, J=+5%, K=+10%, M=+20%. Thus
SK6K = 5.6kQ +10%.
(b) Carbon film resistors. Ceramic rods are heated to
about 1000 °C in methane vapour which decomposes and Terminal
SS
deposits a uniform film of carbon on the rod. The resis- x1 x10
tance of the film depends on its thickness and can be
increased by cutting a spiral groove in it, Fig. 4.5. The Fig. 4.7 Switch type (reading = 27Q)
 ELECTRICAL PROPERTIES 49
coil should not carry more than 1 ampere and a 100 by the sliding contact; if it is ‘logarithmic’ it is proportio-
ohm coil not more than 0.1 ampere—from power = nal to the log of the distance, and at the end of the
I’ R, see p. 60). track a small movement of the sliding contact causes
a larger increase of resistance than at the start.
(d) Variable resistors. Those used in electronic cir-
Larger current versions as used in many electrical
cuits, often as sound volume or other controls and some-
experiments consist of constantan wire wound on a
times called potentiometers, consist of an incomplete
straight ceramic tube with the sliding contact carried
circular track of carbon composition or wire-wound
on a metal bar above the tube.
card, with fixed connections to each end and a rotating
There are two ways of using a variable resistor. It
arm contact which can slide over the track. Figs 4.8a
may be used as a rheostat for controlling the current
and b show the outside and inside respectively of a wire-
in a circuit, when only one end connection and the slid-
wound potentiometer. If the track is ‘linear’, the resis-
ing contact are required, Fig. 4.9. It can also act as
tance tapped off is proportional to the distance moved
a potential divider for controlling the p.d. applied to
a device, all three connections then being used. In Fig.
4.10 any fraction of the total p.d. from the battery can
be tapped off by varying the sliding contact.

Rheostat

Fig. 4.9

Potential
divider

Fig. 4.10

Resistor networks
A network of resistors like that in Fig. 4.11 has a com-
bined or equivalent resistance which can be found exper-
imentally from the ratio of the voltmeter reading to the
ammeter reading. Its value may also be calculated.

Fig. 4.11

(a) Resistors in series. Resistors are in series if the

Fig. 4.8 same current passes through each in turn. In Fig. 4.12
50 VOLUME I Materials and Mechanics

Peery OV
Se
ee
Rik eRe;
eel 1 1

Fig. 4.12 The single resistance R which would have the same resis-
tance as the whole network can be calculated.
For the special case of two equal resistors in parallel
we have R, = R, and
if the total p.d. across all three resistors is V and the
current is J, the combined resistance R is given by

The electrical energy changed per coulomb in passing

through all the resistors equals the sum of that changed In general for n equal resistances R, in parallel, the
in each resistor. Therefore if V,;, V, and V; are the p.d.s
combined resistance is R,/n.
across R,, R, and R; respectively then The combined resistance of the network in Fig. 4.11
V=V,+V,+ V3 is 6; do you agree?

By the definition of resistance, R; = V,/I, R, = V2/I and

R, = V3/1, therefore V, = IR,, V, = IR), V; = IR; and

since V = JR we have Using ammeters and voltmeters

[R= Ry EIR IR: Most ammeters and voltmeters are basically galvano-
meters (i.e. current detectors capable of measuring cur-
R=R,+R,+R;
rents of the order of milliamperes or microamperes)
(b) Resistors in parallel. Here alternative routes are of the moving-coil type which have been modified by
provided to the current, which splits, and we would connecting suitable resistors in parallel or in series with
expect the combined resistance to be less than the smal- them as described in the next sections. Moving-coil
lest individual resistance. In Fig. 4.13 if J is the total instruments are accutate, sensitive and reasonably
cheap and robust.
Connecting an ammeter or voltmeter should cause
the minimum disturbance to the ‘current or p.d. it has
to measure. The current in a circuit is measured by
breaking the circuit and inserting an ammeter in series
so that the current passes through the meter. The resis-
tance of an ammeter must therefore be smal/ compared
with the resistance of the rest of the circuit. Otherwise,
inserting the ammeter changes the current to be mea-
Fig. 4.13 sured. The perfect ammeter would have zero resistance,
the p.d. across it would be zero and no energy would
be absorbed by it.
current through the network and /,, J, and J, are the The p.d. between two points A and B in a circuit
currents in the separate branches then since current is is most readily found by connecting a voltmeter across
not used up in a circuit the points, i.e. in parallel with AB. The resistance of
the voltmeter must be /arge compared to the resistance
Il=[,+1,+1,
of AB, otherwise the current drawn from the main cir-
In a parallel circuit the p.d. across each parallel branch cuit by the voltmeter (which is required to make it oper-
is the same. If this is V then by the definition of resistance ate) becomes an appreciable fraction of the main current
R, = V/I,, Ry = V/b, R3 = V/1,and if R is the combined and the p.d. across AB changes. A voltmeter can be
resistance then R = V/J. Therefore = V/R, 1, =V/R,, treated as a resistor which automatically records the
1, = V/R,and I, = V/R;. Hence p.d. between its terminals. The perfect voltmeter would
 BRE CTRICAL (PROPERTIES 1
have infinite resistance, take no current and absorb no full-scale deflection (f.s.d.) when 100A (0.0001 A)
energy. passes through it. If we wish to convert it to an ammeter
It is instructive to set up the circuit in Fig. 4.14 for reading 0-1 A this can be done by connecting a resistor
measuring the p.d. tapped off by the potential divider (perhaps a misnomer) of very low value in parallel with
between X and Y, using first a high-resistance voltmeter it. Such a resistor is called a shunt and it must be chosen
(e.g. one with a 1002A movement) for @ and then so that only 0.0001 A passes through the meter and the
a low-resistance voltmeter (e.g. one with a 10 mA move- rest of the 1A, namely 0.9999 A, passes through the
ment). The reading in the second case is much lower. shunt, Fig. 4.16. A full-scale deflection of the meter
What will it be if both voltmeters are connected across will then indicate a current of 1 A.
XY?

BA )
gag
10) O000IA 7 | Coil of meter

gate eee

1A ‘ | Ammeter
terminal
0-9999A
Fig. 4.14
Fig. 4.16
The most straightforward method of measuring resis-
tance uses an ammeter and a voltmeter as in Fig. 4.15a.
To obtain the value S of the shunt, we use the fact
The voltmeter records the p.d. across R but the ammeter
that the meter and the shunt are in parallel. Therefore
gives the sum of the currents in R and in the voltmeter.
If the voltmeter has a much higher resistance than R, p.d. across meter = p.d. across shunt
the current through it will be small by comparison and
Applying Ohm’s law to both meter and shunt
the error in calculating R can be neglected. However,
if the resistance of the voltmeter is not sufficiently high, 0.0001 x 1000 = 0.9999 x $ (from V = JR)
perhaps because R is very high, the voltmeter should
be connected across both R and the ammeter as in Fig.
_ 0.0001 x 1000
4.15b. The ammeter now gives the true current in R. ~ §,9999
The voltmeter indicates the p.d. across R and the =0.10
ammeter together, but the resistance of the latter is
usually negligible compared with that of R and so the The combined resistance of the meter and the shunt
p.d. across it will be so small as to make the error in in parallel will now be very small (less than 0.1) and
calculating R negligible. the current ina circuit will be virtually undisturbed when
the ammeter is inserted.
(b) Conversion of a microammeter into a voltmeter.
To convert the same moving-coil meter of resistance

SCAee
1000© and f.s.d. 100 wA to a voltmeter reading 0-1V,
a resistor of high value must be connected in series with
the meter. The resistor is called a multiplier and it must
be chosen so that when a p.d. of 1 V is applied across
the meter and resistor in series, only 0.0001A flows
through the meter and a full-scale deflection results,
Fig. 4.15
Fig. 4.17.
To obtain the value M of the multiplier, we apply
Ohm’s law when there is an F.S.D. of 0.0001 A. Hence
Shunts, multipliers and multimeters
p.d. across multiplier and meter in series =
(a) Conversion of a microammeter into an ammeter. 0.0001 (M + 1000)
Consider a moving-coil meter which has a resistance
(due largely to the coil) of 10000 and which gives a But the meter is to give an f.s.d. when the p.d. across
52 VOLUME I Materials and Mechanics

In electronic circuits resistances of 1MQ or higher

Pant 4 peers are encountered and electronic voltmeters which have

|gt tt) nd

very high resistances have to be used.

| :
Coil of meter
| M 10002-T_voltmeter terminal (c) Multimeters. A multi-range instrument or multi-
meter is a moving-coil galvanometer adapted to measure
/ 0:0001A |\ current, p.d. and resistance. There is a tapped shunt S
across the meter and a tapped multiplier M in series
with it, Fig. 4.19. A rotary switch allows the various
ranges to be chosen.

it and the multiplier in series is 1 V. Therefore

Gb eS ahs
0.0001 (M + 1000) = 1
1 Zap 00 EP 200% 65
M + 1000
= —— = 10000 | 5000 Ohms 50 ¢ |
0.0001 Coil of meter

M = 9000 0
In Fig. 4.18 a microammeter (20-0-100 wA) with its
matching shunts and multipliers is shown.
Voltmeters are often graded according to their ‘resis-
tance per volt’ at f.s.d. For the above voltmeter, 1 V
applied across its terminals produces a full-scale deflec-
tion, i.e. a current of 100A, and so the resistance of
the meter (coil + multiplier) must be 100002 (since R
= V/I = 1/0.0001 = 10000). The ‘resistance per volt’
of the meter is thus 100000/V. To be used as a volt-
meter with an f.s.d. of 10V it would need to have a
total resistance of 100000 Q, i.e. a multiplier of 99 000 0
to limit the full-scale current to 100 tA—but its resis- One other position of the switch is marked ‘ohms’
tance for every volt of deflection is still 100009. A and puts a dry cell B (usually 1.5 volts) and a rheostat
100. 0/V voltmeter has a resistance of 100 9 for an f.s.d. R in series with the meter. To measure resistance the
of 1V and draws a full-scale current of 10mA (J = terminals are short-circuited and R adjusted until the
V/R = wo = 0.01A = 10mA). Hence the higher the pointer gives a full-scale deflection, i.e. is on the zero
‘resistance per volt’ of a voltmeter the smaller is the of the ohms scale. The unknown resistance then replaces
current it draws and the less will it disturb the circuit the short circuit across the terminals. The current falls
to which it is connected. A good voltmeter should have and the pointer indicates the value in ohms. Fig. 4.20
a resistance of at least 10000/V. shows a multimeter.

Fig. 4.18
 BEUROTRIGALPPROPRERIIES 63

measured. Assuming that the amounts of electrical

energy and mechanical energy needed to cause a 1°C
rise in the drum are the same, the p.d. can be calculated.
(a) Mechanical experiment. The apparatus is shown
in Fig. 4.21a. The drum can be rotated about a horizon-
tal axis by a handle and has a nylon cord wound round
it five or six times. A heavy mass m (e.g. 8kg) is hung
from one end of the cord and a fixed rubber band
attached to the other end. The initial temperature of
the drum is noted. When the handle is turned the mass
Cut Out should lift slightly from the floor, the rubber band be
xt Mxt00, Mxt0k METEREIGHT 7 7 just slack and the drum rotate steadily. The number
: « AVOME Ak.6 ie
of turns n to give a temperature rise of 6 (about 10°C)
in time ¢ is noted, as is the diameter d (in metres) of
the drum.
The mass m is supported by a frictional force F
between the cord and the drum. If g is the acceleration
due to gravity (9.8 ms) then F= mg. During n revolu-
tions, F acts over a total distance of zdn where 7d is
the circumference of the drum. The mechanical energy
supplied by the person turning the handle is measured
by the work, and work = force X distance in direction
of force. Here the work = F X mdn = mg X trdn. There-
Fig. 4.20
fore the mechanical energy W to produce a temperature
rise of 1°C in the drum is

Rough calibration of a voltmeter _ mgndn

The calibration, which is done in two parts, is only

eed
approximate, but uses the definition of the volt as a (b) Electrical experiment. The copper drum is
joule per coulomb. First, the temperature rise produced allowed to cool to room temperature and is connected
in a solid copper drum by a known force acting through into the electric circuit of Fig. 4.215 with the nylon cord
a known distance is found and from it the amount of wound round it as before. The voltmeter to be calibrated
mechanical energy required to cause a 1 °C rise is calcu- is connected across the built-in heating coil in the drum
lated. Second, the drum is heated electrically for a cer-
tain time by a known current and the p.d. to be Thermometer

Rubber band

Nylon
Thermometer cord

Copper dru 3 § Heating

5 Coil
Nylon cord

Slack
adjustment
slide

Fig. 4.21 (a)

54 VOLUME I Materials and Mechanics

and a steady current (0.7A is suitable for some makes Knowing the resistivity of a material the resistance
of apparatus) passed so that the temperature rises by of any specimen of that material may be calculated.
about the same as in (a) (i.e. about 10°C) in roughly For example, if the cross-section area of the live rail
the same time ¢. This ensures the heat losses in each of an electric railway is 50 cm? and the resistivity of steel
experiment are similar and can be neglected. The elec- is 1.0 10-7Qm then neglecting the effect of joints,
trical energy supplied to produce a temperature rise of the resistance per kilometre of rail R follows—
1°C is W, the same as the mechanical energy required pl
and Is given by
we
(1.0 x 10-70m) x (10° m)
a (50 x 10-4 m2)
where / is the ammeter reading in amperes, ¢ the time es 10e a
in seconds for a temperature rise of 0, and V the = ( 50 x 10-4 m g
unknown p.d. across the coil. Hence
= Opel UaeoO
ItV _mgrdn
A
chniad The resistivities at 20 °C of various materials are given
in Table 4.1; their experimental determination is briefly
_mgndn 6,
described on p. 62.
ait atora The conductivity (a) of a material is the reciprocal
V can thus be calculated in joules per coulomb (volts) of its resistivity (p), i.e. o=1/p, and has unit ohm7!
and compared with the reading on the voltmeter. In metre7!(Q7!m7!).
this experiment the same amount of internal energy (not Silver is the best conductor, i.e. has the lowest resisti-
heat as is often stated, see p. 76) is produced in the vity, and is followed closely by copper which, being
drum, first from mechanical energy and then from elec- much less expensive, is used for electrical connecting
trical energy. wire. Although the resistivity of aluminium is nearly
twice that of copper, its density is only about one-third
of copper’s. The ratio of current-carrying-capacity to
weight of aluminium is therefore greater than that of
Resistivity copper. This accounts for its use in the overhead power
The resistance of a conductor depends on its size as cables of the Grid System where aluminium strands are
well as on the material of which it is made. To make wrapped round a core of steel wires (54 aluminium
fair comparisons of the abilities of different materials strands to 7 steel wires for example, Fig. 4.22). The
to conduct, the resistance of specimens of the same size cable then has the strength it requires for suspension
must be considered. in long spans between pylons.
Experiment shows that the resistance R of a uniform The resistivity of a pure metal is increased by small
conductor of a given material is directly proportional amounts of ‘impurity’ and alloys have resistivities appre-
to its length / and inversely proportional to its cross- ciably greater than those of any of their constituents.
section area. A. Hence On the other hand, the addition of tiny traces of ‘impuri-
ties’ to pure semiconductors (a process known as
‘doping’ the semiconductor) reduces their resistivity.
‘Impurity’ atoms in a crystal lattice act as ‘defects’ and
restrict the movement of charge carriers. When a semi-
or = —
conductor is ‘doped’ this is more than offset by the pro-
duction of extra ‘free’ charges.
where p is a constant (for fixed temperature and other
physical conditions), called the resistivity of the material
Electrical strain gauge
of the conductor.
Hence since p= AR// we can say that the resistivity One device which engineers employ to obtain informa-
of a material is numerically the resistance of a sample tion about the size and distribution of strains in struc-
of unit length and unit cross-section area, at a certain tures such as buildings, bridges and aircraft is the
temperature. The unit of p is ohm metre (1m) since electrical strain gauge. It converts mechanical strain into
those of AR//are metre? X ohm/metre, i.e. ohm metre. a resistance change in itself by using the fact that the
a
 ELECTRIGAL, PROPBRTIES,; 55
Table 4.1

Material Resistivity
Qm
CONDUCTORS
Silver 1.6 x 10-8 Contacts on small switches
ietais Copper EES 10° Connecting wires
Aluminium 2IX107° Power cables
Tungsten Spore Lamp filaments
Manganin ade 10m” High-precision standard resistors
Alloys Constantan or Eureka ADS XO Resistance boxes, variable resistors
Nichrome (Ni-Cr) 110 =< 105° Heating elements
Carbon 3000 x 1078 Radio resistors
SEMICONDUCTORS
Germanium 0.6 Transistors
Silicon 2300 Transistors
INSULATORS
Glass. 1019-194

Polystyrene 1045

resistance of a wire depends on its length and cross-

section area.
One type of gauge consists of a very fine wire (of
an alloy containing mostly nickel, iron and chromium)
cemented to a piece of thin paper as in Fig. 4.23. In
use it is securely attached with a very strong adhesive
to the component under test so that it experiences the
same strain as the component. If, for example, an
increase of length strain occurs, the gauge wire gets
longer and thinner and on both counts its resistance
increases. Thick leads connect the gauge to a resistance-
measuring circuit (e.g. a Wheatstone bridge, p. 60) and
previous calibration of the gauge enables the strain to
be measured directly. What is the advantage of using
a parallel-wire arrangement for the strain gauge?
Strain gauges are used to check theoretical calcula-
tions on new aircraft designs. Fig. 4.24 shows a model
of a slender-wing aircraft with weights and strain gauges
attached at various places. The weights simulate the
aerodynamic loading when the aircraft is in flight.
Fig. 4.22

Fine wire

Paper

Thick leads

Fig. 4.23 Fig. 4.24 ¥

56 VOLUME I Materials and Mechanics

Effects of temperature on resistance This approximate method is quicker but in this example
introduces an error of 0.3 in 34.5, i.e. about 1%, which
(a) Temperature coefficient. The resistance of a mater-
is acceptable for many purposes.
ial varies with temperature and the variation can be
The experimental determination of a is outlined on
expressed by its temperature coefficient of resistance a.
p. 62. Metals and alloys have positive temperature coef-
If a material has resistance Ry at 0°C and its resistance
ficients (they are p.t.c. materials), i.e. their resistance
increases by 5R due to a temperature rise $6 then a
increases with temperature rise. The values for pure
for the material is defined by the equation
metals are of the order of 4 x 1073 per °C or roughly
1/273 per °C, the same as the cubic expansivity of a
gas. In a tungsten-filament electric lamp the current
raises the temperature of the filament to over 2730°C
In words, a is the fractional increase in the resistance when lit. The ‘hot’ resistance of the filament is, there-
at 0°C (i.e. 8R/Ry) per unit rise of temperature. The fore, more than ten times the ‘cold’ resistance. Why
unit of a is °C“! since 8R and Ry have the same units doesn’t a fuse blow every time lights are switched on?
(ohms) and 8R/R, is thus a ratio. For copper Alloys have much lower temperature coefficients of
a~ 4x 10-3°C™!, which means that a copper wire hav- resistance than pure metals; that for manganin is about
ing a resistance of 1 ohm at 0°C increases in resistance 2x 107-5 per °C and a small temperature change has
by 4 x 10-7 ohm for every 1 °C temperature rise. little effect on its resistance.
Experiment shows that the value of a varies with the Graphite, semiconductors and most non-metals have
temperature at which 66 occurs but, to a good approxi- negative temperature coefficients (they are n.t.c. mater-
mation, for metals and alloys we can generally assume ials), i.e. their resistance decreases with temperature
it is constant in the range 0 to 100°C. Thus if a specimen rise.
has resistances R, and Ry at temperatures 6 and 0°C
respectively then replacing 5R by R,— Ry and 80 by (b) Superconductors. When certain metals (e.g. tin,
0 in the expression for a, we obtain lead) and alloys are cooled to near —273°C with liquid
helium an abrupt decrease of resistance occurs. Below
Ro— Ro
aFt_——— a definite temperature, different for each material, the
Ro resistance vanishes and a current once started seems
Rearranging gives Ry — Ry = Road to flow for ever. Such materials are called superconduc-
tors and their use in electrical power engineering and
and R,= Ro(1 + a) electronics is being explored.
Recently superconductors have been made from the
When using the equation where accuracy is important oxides of copper, barium and yttrium, which need only
Ro should be the resistance at 0°C. A calculation shows be cooled with the plentiful and cheaper liquid nitrogen
the procedure when Rj is not known. (at =210°@).
Suppose a copper coil has a resistance of 30.0 at 20°C
and its resistance at 60°C is required. Taking a for cop- (c) Thermistors (derived from thermal resistors).
per as 4.0 x 10-3 °C~! we have These are devices whose resistance varies quite mar-
kedly with temperature. Depending on their composi-
Ry a Ro te 20a)
tion they can have either n.t.c. or p.t.c. characteristics.
and Réo = R,(1 SF 60a)
The n.t.c. type consists of a mixture of oxides of iron,
nickel and cobalt with small amounts of other substances
and is used in electronic circuits to compensate for resis-
Dividing, — = ——_— tance increase in other components when the tempera-
Rx 1+ 20a
ture rises and also as a thermometer for temperature
_ 30(1 + 60 x 4.0 x 1073) measurement. The p.t.c. type, which is based on barium
60 = 34.50
(1 + 20 x 4.0 x 10-3) titanate, can show a resistance increase of 50 to 200
times for a temperature rise of a few degrees. It is useful
If the calculation had not been based on the resistance as a temperature-controlled switch. Why? Fig. 4.25
at 0°C and we had taken the original resistance (i.e. shows a selection of thermistors.
Ry) as Ro then using Roy = Ro(1+a0) where @ =
(60 — 20) °C = 40°C and Ry = 300, we get (d) Electrons, resistance and temperature. The ‘free’
electron theory can account qualitatively for the varia-
Rey = 30(1 + 4.0 x 1073 x 40) = 34.80 tion of resistance with temperature of different mater-
 EPDECEBRICALNPROPERTIBS 57

Fig. 4.25
(a) disc n.t.c. type;
(b) plate n.t.c. type;
(c) p.t.c. type;
(d) rod n.t.c. type;
(e) rod voltage-
dependent resistor;
(f) bead-in-glass type

ials. The increased average separation of the ions in ple, to heat, but it is replenished at the same rate at
a metal which accompanies a temperature rise (see p. which it is transferred. The electric and magnetic fields
84) causes local distortion of the crystal lattice. As a thus act as a temporary storage reservoir of electrical
result there is increased interaction between the lattice energy in the transfer of chemical energy to heat. A
and the ‘free’ electrons when they drift under an applied battery or generator is said to produce an electromotive
p.d. The average drift speed is reduced and the resis- force (e.m.f.), defined in terms of energy transfer:
tance thus increases. In semiconductors this is more than
The electromotive force of a source (a battery, genera-
compensated when greater vibration of the atoms
tor, etc.) is the energy (chemical, mechanical, etc.) con-
breaks bonds, ‘freeing’ more electrons (an insignificant — verted into electrical energy when unit charge passes
effect in metals) and thereby produces a marked de- through it.
crease of resistance with temperature rise. Heavily
‘doped’ (see Volume II, chapter 23) semiconductors can The unit of e.m.f., like the unit of p.d., is the volt
‘acquire metallic properties, a temperature rise increas- and equals the e.m.f. of a source which changes 1 joule
ing their resistance. of chemical, mechanical or other form of energy into
electrical energy when 1 coulomb passes through it. A
car battery with an e.m.f. of 12 volts supplies 12 joules
Electromotive force per coulomb passing through it; a power station genera-
Batteries and generators are able to maintain one termi- tor with an e.m.f. of 25 000 volts is a much greater source
nal positive (i.e. deficient in electrons) and the other of energy and supplies 25000 joules per coulomb—2
negative (i.e. with an excess of electrons). If we consider coulombs would receive 50000 joules and so on. In
the motion of positive charges, then a battery, for exam- general, if a charge Q (in coulombs) passes through
ple, moves positive charges from a place of low potential a source of e.m.f. E (in volts), the electrical energy
(the negative terminal) through the battery to a place supplied by the source W (in joules) is
of high potential (the positive terminal). The action may
W=QE
be compared with that of a pump causing water to move
from a point of low gravitational potential to one of It should be noted that although e.m.f. and p.d. have
high potential. the same unit, they deal with different aspects of an
A battery or generator therefore does work on charges electric circuit. Whilst e.m.f. applies to a source supply-
and so energy must be changed within it. (Work is a ing electrical energy, p.d. refers to the conversion of
measure of energy transfer, see p. 76.) In a battery electrical energy in a circuit. The term e.m.f. is mislead-
chemical energy is transferred into electrical energy ing to some extent, since it measures energy per unit
which we consider to be stored in the electric and mag- charge and not force. It is true, however, that the source
netic fields produced. When current flows in an external of e.m.f. is responsible for moving charges round the
circuit this stored electrical energy is changed, for exam- circuit.
58 VOLUME I Materials and Mechanics

A voltmeter measures p.d. and one connected across In symbols E = V+oyv

the terminals of an electrical supply such as a battery emit, useful ‘lost’
volts volts
records what is called the terminal p.d. of the battery.
If the battery is not connected to an external circuit where v is the p.d. across the internal resistance of the
and the voltmeter has a very high resistance then the cell, a quantity which cannot be measured directly but
current through the battery will be negligible. We can obtained only by subtracting V from E. From the equa-
regard the voltmeter as measuring the number of joules tion E=V+v we see that the sum of the p.d.s across
of electrical energy the battery supplies per coulomb, all the resistance in a circuit (external and internal) equals
i.e. its e.m.f. A working but less basic definition of the e.m.f.
e.m.f. is to say that it equals the terminal p.d. of a Since V = JR and v = /r we can rewrite the previous
battery or generator on open circuit, i.e. when not main- equation
taining current. E=IR+Ir
E=I(R+r)
Internal resistance Suppose a high-resistance voltmeter reads 1.5 V when
connected across a dry battery on open circuit, and 1.2 V
A high-resistance voltmeter connected across a cell on when the same battery is supplying a current of 0.30A
open circuit records its e.m.f. (very nearly), Fig. 4.26a. through a lamp of resistance R. What is (a) the e.m-f.
Let this be E. If the cell is now connected to an external of the battery, (b) the internal resistance of the battery
circuit in the form of a resistor R and maintains a steady and (c) the value of R?
current J in the circuit, the voltmeter reading falls; let Using symbols with their previous meanings:
it be V, Fig. 4.26b. V is the terminal p.d. of the cell
(but not on open circuit) and it is also the p.d. across (a) Since the terminal p.d. on open circuit equals the
R (assuming the connecting leads have zero resistance). e.m.f., we have EF=1.5V.
Since V is less than E, then not all the energy supplied
per coulomb by the cell (i.e. F) is changed in the external (b) E = V+v = V+Ir where V = 1.2V and] =
circuit to other forms of energy (often heat). What has 0.30 A. Therefore
happened to the ‘lost’ energy per coulomb?
Ir=E-V an d e=O
7 030 1.00
=].

(c) From
V= IR, .
Vale?
=—=— =400
L030
R
Fig. 4.26 (b) The internal resistance of an electrical supply depends
on several factors and is seldom constant as is often
assumed in calculations. However, it is sometimes useful
The deficiency is due to the cell itself having some
resistance. A certain amount of electrical energy per to know its rough value and estimates can be made by
coulomb is wasted in getting through the cell and so taking p.d. and current measurements and proceeding
less is available for the external circuit. The resistance as in the above example. Sources such as low-voltage
of a cell is called its internal or source resistance (r). supply units and car batteries from which large currents
Taking stock of the energy changes in the complete cir- are required must have very low internal resistances.
cuit including the cell, we can say, assuming conserva- On the other hand if a 5000 V, E.H.T. power supply
tion of energy: does not have an internal resistance of the order of
megohms, to limit the current it supplies, it will be dan-
energy supplied energy changed energy wasted gerous.
per coulomb by = per coulomb by + per coulomb on The effect of internal resistance can be seen when
cell external circuit internal __resis- a bus or car starts with the lights on. Suppose the starter
tance of battery motor requires a current of 100A from the battery of
Or, from the definitions of e.m.f. and p.d., e.m.f. 12 V and internal resistance 0.042 to start the
engine. How many volts are ‘lost’? What is the terminal
e.m.f. = p.d. across R + p.d. across r p.d. of the battery with the starter motor working? Why
 ELECTRICAL: PROPERTIES’ 59

do the lights dim if they are designed to operate on (i) e.m.f.s are positive if we pass from the positive
a 12 V supply? terminal of the supply round the rest of the circuit or
The effect of a load (i.e. an external circuit drawing loop to the negative terminal, and negative if the oppo-
current) on a source is to reduce its terminal p.d. The site happens, and
equation V = E — Ir shows that if either J or r increases, (ii) IR products are positive when there is a drop
V decreases. Also note that the maximum current that of potential and negative when there is a rise, then in
can be supplied by a source occurs when R = 0 and is Fig. 4.27b we can say for a clockwise journey round
E/r; it is called the ‘short-circuit’ current and would ABCDA, and with the current directions shown,
damage most sources.
SE E, 2 E,
The terminal p.d. of a battery on open circuit as mea-
sured by even a very-high-resistance voltmeter is not IR =1,R, +1,R,—-—1LhR;
quite equal to the e.m.f. because the voltmeter must
EB, roe E, a LR, + LR = LR;
take some current, however small, to give a reading.
A small part of the e.m.f. is, therefore, ‘lost’ in driving The law is a statement of the conservation of energy
current through the internal resistance of the battery. using electrical quantities.
A potentiometer is used to measure e.m.f. to a very
high accuracy (p. 62).

Kirchoff’s laws
The statements that have been made about steady cur-
rents and p.d.s in series and parallel circuits are summar-
ized by Kirchoff’s two laws.
(a) First law. This states that at a junction in a circuit, E,
the current arriving equals the current leaving. If currents I, R, Ro

arriving are given, say, a positive sign and those leaving B

a negative sign, then the law may be stated in symbols
A Cc
as
D
~I1=0
- Y (pronounced ‘sigma’) stands for ‘the algebraic sum
h Rs z
of’—in this case the currents J. For example in Fig.
(b)
4.27a, at junction A,
Fig. 4.27
current arriving = J
currentleaving =/,+1,+ J, Power and heating effect
f= resi, or — LL 0 Current flow is accompanied by the conversion of elec-
Ke. x1=0 trical energy into other forms of energy and it is often
necessary to know the rate at which a device brings about
The law is a statement of our belief that when we are
this conversion.
dealing with steady currents, charge is conserved and
flows in a circuit without being destroyed or accumulat- The power of a device is the rate at which it converts
ing at any point. energy from one form into another.
If the p.d. across a device is V and the current through
(b) Second law. This states that round any closed cir- it is J, the electrical energy W converted by it in time
cuit or loop the algebraic sum of the e.m.f. E is equal tis (from the definition of p.d., p. 46)
to the algebraic sum of the products of current I and
W=ItV
resistance R. In symbols
The power P of the device will be
2E=XIR
If we adopt the sign conventions that, in going round
a circuit or loop,
60 VOLUME I Materials and Mechanics

The unit of power is the watt (W) and equals an energy sten iodide with the tungsten vapour and remains as
conversion rate of 1 joule per second, i.e. 1 W=1Js7!. vapour when the lamp is working, thereby preventing
In the expression P= JV, P will be in watts if J is in blackening.
amperes and V in volts. A larger unit is the kilowatt
(b) Fuses. When current flows in a wire its tempera-
(kW) which equals 1000 watts.
ture rises until the rate of loss of heat to the surroundings
If all the electrical energy is converted into heat by
equals the rate at which heat is produced. If this temper-
the device it is called a ‘passive’ resistor and the rate
ature exceeds the melting-point of the material of the
of production of heat will also be JV. If its resistance
wire, the wire melts. A fuse is a short length of wire,
is R, Fig. 4.28, then since R = V/I we have
often tinned copper, selected to melt when the current
P=IV through it exceeds a certain value. It thereby protects
a circuit from excessive currents.
V yY It can be shown (see question 21, p. 71) that:
Sela
R R (i) the temperature reached by a given wire depends
only on the current through it and is independent of
=]. IR=IJAR
its length (provided it is not so short for heat loss from
the ends where it is supported to matter); and
(ii) the current required to reach the melting-point
of the wire increases as the radius of the wire increases.
Fuses which melt at progressively higher tempera-
/ R ~Passive tures can thus be made from the same material by using
>, resistor
P=IV=I‘R wires of increasing radius.
Fig. 4.28 (c) The kilowatt-hour (kWh). For commercial pur-
poses the kilowatt-hour is a more convenient unit of
electrical energy than the joule.
There are thus three alternative expressions for power
but the last two are only true when all the electrical The kilowatt-hour is the quantity of energy converted
energy is changed to heat. The first, P=JV, gives the to other forms of energy by a device of power 1 kilowatt
rate of production of all forms of energy. For example in 1 hour.
if the current in an electric motor is 5 A when the applied
The energy converted by a device in kilowatt-hours
p.d. is 10V then 50W of electric power is supplied to
is thus calculated by multiplying the power of the device
it. However, it may only produce 40W of mechanical
in kilowatts by the time in hours for which it is used.
power, the other 10W being the rate of production of
Hence a 3kW electric radiator working for 4 hours uses
heat by the motor windings due to their resistance.
12kWh of electrical energy—often called 12 ‘units’.
(a) Heating elements and lamp filaments. The expres- How many joules are there in 1 kWh?
sion P = V’/R shows that for a fixed supply p.d. of V,
(d) Maximum power theorem. It can be shown that
the rate of heat production by a resistor increases as
the power delivered to a load of resistance R is a maxi-
R decreases. Now R = pl/A, therefore P = V? A/pl and
so where a high rate of heat production at constant p.d. mum when R equals the source resistance r. The maxi-
mum power P is given by
is required, as in an electric fire on the mains, the heating
element should have a large cross-section area A, a small P=[R= E°R/(R +r) = E°R/4R? = E2/4R
resistivity p and a short length /. It must also be able
to withstand high temperatures without oxidizing in air The power wasted in the source is E*/4r, i.e. the same
(and becoming brittle). Nichrome is the material which as that in R since r=R. The efficiency of the power-
transfer process is thus 50%.
best satisfies all these requirements.
Electric lamp filaments have to operate at even higher
temperatures if they are to emit light. In this case, tung-
Wheatstone bridge
sten, which has a very high melting-point (3400 °C), is
used either in a vacuum or more often in an inert gas (a) Theory. The Wheatstone bridge circuit enables
(nitrogen or argon). The gas reduces evaporation of resistance to be measured more accurately than by the
the tungsten (why?) and prevents the vapour condensing ammeter—voltmeter method (p. 51). It involves making
on the inside of the bulb and blackening it. In modern adjustments until a galvanometer is undeflected and so,
projector lamps there is a little iodine which forms tung- being a ‘null’ method, it does not depend on the accuracy
 BLECTRICAL PROPRERILES 61
of an instrument. Other known resistors are, however, The position of D is adjusted until there is no deflec-
required. tion on G, then
Four resistors P, Q, R, S are joined as in Fig. 4.29a.
If P is the unknown resistor, Q must be known as must Ie BYR ie resistance of AD
the values of R and S or their ratio. A sensitive galvan- ee resistance of DC
ometer G and a cell (dry or Leclanché) are connected
Since the wire is uniform, resistance will be proportional
as shown. One or more of Q, R and S are adjusted
to length and therefore
until there is no deflection on G. The bridge is then
said to be balanced and it can be shown that

Four practical points should be noted. Resistor Q

should be chosen to give a balance point near to the
from which P can be found. The proof of this expression
centre of the wire, say between 30 and 70cm, for three
follows.
reasons. First, any errors in reading the balance lengths
At balance, no current flows through G, therefore AD and DC are then small in comparison with their
the p.d. across BD is zero and so values. Second, where the resistors and metre wire are
p.d. across AB = p.d. across AD screwed or soldered to the copper strips there are ‘con-
nection’ resistances which may, if desired, be deter-
Also, current through P = current through Q = J, and, mined experimentally and expressed in millimetres of
current through R = current through S = J,. Therefore bridge wire. However, when this is not done the ‘end
I, x P = L x R corrections’, as they are called, will have least effect
if neither AD nor DC is small. Third, the bridge is more
Similarly, p.d. across BC = p.d. across DC. Therefore sensitive near the middle since the unbalanced current
is larger per mm change of position.
In finding the balance point the cell key should be
Dividing, FieB closed before the jockey makes contact with the wire.
O. 3S This is necessary because, due to an effect known as
‘self-induction’, the currents in the circuit take a short
It can be shown that the same condition holds if the
time to grow to their steady values. During this time
cell and G are interchanged.
a momentary deflection of the galvanometer might be
(b) Metre bridge. This is the simplest practical form obtained even when the bridge is balanced for steady
of the Wheatstone bridge, Fig. 4.29b. The resistors R currents.
and § consist of a wire AC of uniform cross-section A high resistor should be joined in series with the
and 1m long, made of an alloy such as constantan, with galvanometer to protect it from damage whilst the
a resistance of several ohms. The ratio of R to Sis altered balance point is being found. In the final adjustment it
by changing the position on the wire of the movable is shorted out and maximum sensitivity of the galvano-
contact or ‘jockey’ D. The other arm of the bridge con- meter obtained, Fig. 4.29c.
tains the unknown resistor P and a known resistor Q. Having obtained a balance point, P and Q should
Thick copper strips of low resistance connect the various be interchanged and a second pair of values for AD
parts. Figs 4.29a and b have identical lettering to show and DC obtained and the means taken. This helps to
their similarity. compensate for errors arising from non-uniformity of

Thick copper strip

ee
Protective resistor

Fig. 4.29 (a) (b)

62 VOLUME I Materials and Mechanics

the wire, from the wrong positioning of the millimetre Potentiometer

scale in relation to the wire and from ‘end corrections’
(a) Theory. A potentiometer is an arrangement
provided the balance point is near the centre of the
which measures p.d. accurately. It can be adapted to
wire.
measure current and resistance.
Wheatstone bridge methods are unreliable for finding
In its simplest form it consists of a length of resistance
resistances of less than 1 ohm, due to ‘connection’ resis-
wire AB of uniform cross-section area, lying alongside
tance errors becoming appreciable, and are insensitive
a millimetre scale, and through which a steady current
for resistances greater than 1 megohm unless a highly
is maintained by a cell, called the driver cell, Fig. 4.32.
sensitive galvanometer is used. A modern dial-operated
(This is usually an accumulator because it gives a steady
form of Wheatstone bridge for the use of maintenance
current for a long time.) As a result there is a p.d.
engineers is shown in Fig. 4.30.
between any two points on the wire which is propor-
tional to their distance apart. Part of the p.d. across
AB is tapped off and used to counter-balance the p.d.
to be measured. If the p.d. across AB is 2 volts and
if the wire is uniform and the current steady, what p.d.s
can be tapped off between (7) A and P, (ii) A and Q,
(iii) A and R, (iv) Q and B?

Accumulator

—- Potentiometer
wire

asa tal
A B
TS
ie nillimetre

Fig. 4.32
Fig. 4.30
In practice the unknown p.d. V, is connected with
(c) Measurement of resistivity using a metre bridge.
its positive side to X in Fig. 4.33a if the positive terminal
The resistivity p of a material can be determined by
of the driver cell is jo