BooksAudiobooksComicsSheet Music### Categories

### Categories

### Categories

Editors' Picks Books

Hand-picked favorites from

our editors

our editors

Editors' Picks Audiobooks

Hand-picked favorites from

our editors

our editors

Editors' Picks Comics

Hand-picked favorites from

our editors

our editors

Editors' Picks Sheet Music

Hand-picked favorites from

our editors

our editors

Top Books

What's trending, bestsellers,

award-winners & more

award-winners & more

Top Audiobooks

What's trending, bestsellers,

award-winners & more

award-winners & more

Top Comics

What's trending, bestsellers,

award-winners & more

award-winners & more

Top Sheet Music

What's trending, bestsellers,

award-winners & more

award-winners & more

Welcome to Scribd! Start your free trial and access books, documents and more.Find out more

**The Chemistry Maths Book
**

Second Edition

Erich Steiner

University of Exeter

without mathematics the sciences cannot be understood, nor made clear, nor taught, nor learned. (Roger Bacon, 1214–1292)

1

3

Oxford New York

Great Clarendon Street, Oxford OX2 6DP Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide in Auckland Cape Town Dar es Salaam Hong Kong Karachi Kuala Lumpur Madrid Melbourne Mexico City Nairobi New Delhi Shanghai Taipei Toronto With ofﬁces in Argentina Austria Brazil Chile Czech Republic France Greece Guatemala Hungary Italy Japan Poland Portugal Singapore South Korea Switzerland Thailand Turkey Ukraine Vietnam Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries Published in the United States by Oxford University Press Inc., New York © E. Steiner 2008 The moral rights of the authors have been asserted Database right Oxford University Press (maker) First published 1996 Second edition published 2008 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this book in any other binding or cover and you must impose the same condition on any acquirer British Library Cataloguing in Publication Data Data available Library of Congress Cataloging in Publication Data Data available Typeset by Graphicraft Limited, Hong Kong Printed in Great Britain on acid-free paper by Ashford Colour Press Ltd, Gosport, Hampshire ISBN 978–0–19–920535–6 1 3 5 7 9 10 8 6 4 2

Preface

This book describes the mathematics required for the full range of topics that make up a university degree course in chemistry. It has been designed as a textbook for courses in ‘mathematics for chemists’.

Structure of the book

The subject is developed in a logical and consistent way with few assumptions made of prior knowledge of mathematics. The material is organized in three largely independent parts: Chapters 1 to 15 on algebra, the calculus, differential equations, and expansions in series; Chapters 16 to 19 on vectors, determinants, and matrices; Chapters 20 and 21 are introductions to the big topics of numerical analysis and statistics. A feature of the book is the extensive use of examples to illustrate every important concept and method in the text. Some of these examples have also been used to demonstrate applications of the mathematics in chemistry and several basic concepts in physics. The exercises at the end of each chapter are an essential element of the development of the subject, and have been designed to give the student a working understanding of the material in the text. The text is accompanied by a ‘footnote history’ of mathematics. Several topics in chemistry are given extended treatments. These include the concept of pressure–volume work in thermodynamics in Chapter 5, periodic systems in Chapter 8, the differential equations of chemical kinetics in Chapter 11, and several applications of the Schrödinger equation in Chapters 12 and 14. In addition, the contents of several chapters are largely dictated by their applications in the physical sciences: Chapter 9, the mathematics of thermodynamics; Chapters 10 and 16, the description of systems and processes in three dimensions; Chapter 13 (advanced), some important differential equations and special functions in mathematical chemistry and physics; Chapter 15 (advanced), intermolecular forces, wave analysis, and Fourier transform spectroscopy; Chapters 18 and 19, molecular symmetry and symmetry operations, molecular orbital theory, molecular dynamics, and advanced quantum mechanics.

Global changes in this edition

1. An overall reorganization has been carried out to link the text and examples more closely to the exercises at the end of each chapter. The symbol 0 has been placed at appropriate places within the body of the text as a pointer to the relevant exercises. New examples and exercises have been inserted to give a more complete coverage of the development of the mathematics. 2. In addition to the solutions to the numerical exercises given at the back of the book, a full set of worked solutions of the end-of-chapter exercises has been placed on the book’s companion website at www.oxfordtextbooks.co.uk/orc/steiner2e 3. The opportunity has been taken to consolidate the several major and many minor corrections and improvements that have been made over the years since publication of the ﬁrst edition in 1996. A small number of new historical footnotes have been

vi

Preface

added to accompany new material. The material within some chapters has been reordered to make the development of the subject more logical.

Other principal changes

Chapter 1. A new section, Factorization, factors, and factorials, ﬁlls a gap in the coverage of elementary topics. The rules of precedence for arithmetic operations has been brought forward from chapter 2 and extended with examples and exercises, providing further revision and practice of the arithmetic that is so important for the understanding of the material in subsequent chapters. The biggest change in the chapter, reﬂected in the change of title to Numbers, variables, and units, is a rewritten and much enlarged section on units to make it a more authoritative and useful account of this important but often neglected topic. It includes new examples of the type met in the physical sciences, a brief subsection on dimensional analysis, and a new example and exercise on the structure of atomic units. Chapter 2. Parts of the chapter have been rewritten to accommodate more discussion of the factorization and manipulation of algebraic expressions. Chapter 7. Numerous small changes have been made, including an introduction to the multinomial expansion, and revision of the discussion of the Taylor series. Chapter 9. Section 9.8 has been rewritten to clarify the relevance of line integrals to change of state in thermodynamics. Chapter 13. The section on the Frobenius method has been revised, with new and more demanding examples and exercises. Chapter 19. Sections 19.2 and 19.3 on eigenvalues and eigenvectors have been rewritten, with new examples and exercises, to improve the ﬂow and clarity of the discussion.

Acknowledgements

I wish to express my gratitude to colleagues and students at Exeter University and other institutions for their often helpful comments on the previous edition of the book, for pointing out errors and obscurities, and for their suggestions for improvements. I particularly want to thank Anthony Legon for his valuable comments on the revised chapters 1 and 9. I also wish to thank the reviewers of this book for their generous response to the proposal of a second edition, and the staff of Oxford University Press for their patience and help. Above all, I want to thank my wife Mary, without whom nothing could have been done. Erich Steiner, Exeter, June 2007

Contents

1 Numbers, variables, and units

1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 Concepts Real numbers Factorization, factors, and factorials Decimal representation of numbers Variables The algebra of real numbers Complex numbers Units Exercises

1

1 3 7 9 13 14 19 19 29

2

Algebraic functions

2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 Concepts Graphical representation of functions Factorization and simpliﬁcation of expressions Inverse functions Polynomials Rational functions Partial fractions Solution of simultaneous equations Exercises

31

31 32 34 37 40 50 52 55 58

3

Transcendental functions

3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 3.10 Concepts Trigonometric functions Inverse trigonometric functions Trigonometric relations Polar coordinates The exponential function The logarithmic function Values of exponential and logarithmic functions Hyperbolic functions Exercises

62

62 63 72 73 77 80 83 86 87 89

4

Differentiation

4.1 4.2 4.3 4.4 4.5 4.6 4.7 Concepts The process of differentiation Continuity Limits Differentiation from ﬁrst principles Differentiation by rule Implicit functions

93

93 94 97 98 100 102 110

viii

Contents

4.8 4.9 4.10 4.11 4.12 4.13

Logarithmic differentiation Successive differentiation Stationary points Linear and angular motion The differential Exercises

111 113 114 118 119 122

5

Integration

5.1 5.2 5.3 5.4 5.5 5.6 5.7 5.8 5.9 Concepts The indeﬁnite integral The deﬁnite integral The integral calculus Uses of the integral calculus Static properties of matter Dynamics Pressure–volume work Exercises

126

126 127 132 142 147 148 152 157 160

6

Methods of integration

6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 Concepts The use of trigonometric relations The method of substitution Integration by parts Reduction formulas Rational integrands. The method of partial fractions Parametric differentiation of integrals Exercises

163

163 163 165 173 176 179 184 187

7

**Sequences and series
**

7.1 7.2 7.3 7.4 7.5 7.6 7.7 7.8 7.9 Concepts Sequences Finite series Inﬁnite series Tests of convergence MacLaurin and Taylor series Approximate values and limits Operations with power series Exercises

191

191 191 196 203 204 208 214 219

221 225

225 226 228 235 236 240

8

Complex numbers

8.1 8.2 8.3 8.4 8.5 8.6 Concepts Algebra of complex numbers Graphical representation Complex functions Euler’s formula Periodicity

Contents

ix

8.7 8.8

Evaluation of integrals Exercises

244 245

9

**Functions of several variables
**

9.1 9.2 9.3 9.4 9.5 9.6 9.7 9.8 9.9 9.10 9.11 9.12 Concepts Graphical representation Partial differentiation Stationary points The total differential Some differential properties Exact differentials Line integrals Multiple integrals The double integral Change of variables Exercises

247

247 248 249 253 258 262 272 275 281 283 285 289

10

Functions in 3 dimensions

10.1 10.2 10.3 10.4 10.5 10.6 10.7 Concepts Spherical polar coordinates Functions of position Volume integrals The Laplacian operator Other coordinate systems Exercises

294

294 294 296 299 304 307 312

11

**First-order differential equations
**

11.1 11.2 11.3 11.4 11.5 11.6 11.7 11.8 Concepts Solution of a differential equation Separable equations Separable equations in chemical kinetics First-order linear equations An example of linear equations in chemical kinetics Electric circuits Exercises

314

314 315 318 322 328 330 332 334

12

**Second-order differential equations. Constant coefﬁcients
**

12.1 12.2 12.3 12.4 12.5 12.6 12.7 12.8 12.9 12.10 Concepts Homogeneous linear equations The general solution Particular solutions The harmonic oscillator The particle in a one-dimensional box The particle in a ring Inhomogeneous linear equations Forced oscillations Exercises

337

337 337 340 344 348 352 356 359 363

365

x

Contents

13

**Second-order differential equations. Some special functions
**

13.1 13.2 13.3 13.4 13.5 13.6 13.7 13.8 Concepts The power-series method The Frobenius method The Legendre equation The Hermite equation The Laguerre equation Bessel functions Exercises

368

368 369 371 375 381 384 385 389

14

**Partial differential equations
**

14.1 14.2 14.3 14.4 14.5 14.6 14.7 14.8 Concepts General solutions Separation of variables The particle in a rectangular box The particle in a circular box The hydrogen atom The vibrating string Exercises

391

391 392 393 395 398 401 410 413

15

**Orthogonal expansions. Fourier analysis
**

15.1 15.2 15.3 15.4 15.5 15.6 15.7 Concepts Orthogonal expansions Two expansions in Legendre polynomials Fourier series The vibrating string Fourier transforms Exercises

416

416 416 421 425 432 433 441

16

Vectors

16.1 16.2 16.3 16.4 16.5 16.6 16.7 16.8 16.9 16.10 16.11 Concepts Vector algebra Components of vectors Scalar differentiation of a vector The scalar (dot) product The vector (cross) product Scalar and vector ﬁelds The gradient of a scalar ﬁeld Divergence and curl of a vector ﬁeld Vector spaces Exercises

444

444 445 448 453 456 462 466 467 469 471 471

17

Determinants

17.1 17.2 17.3 Concepts Determinants of order 3 The general case

474

474 476 481

Contents

xi

17.4 17.5 17.6 17.7 17.8

The solution of linear equations Properties of determinants Reduction to triangular form Alternating functions Exercises

483 488 493 494 496

18

**Matrices and linear transformations
**

18.1 18.2 18.3 18.4 18.5 18.6 18.7 18.8 Concepts Some special matrices Matrix algebra The inverse matrix Linear transformations Orthogonal matrices and orthogonal transformations Symmetry operations Exercises

499

499 502 505 513 516 521 524 529

19

**The matrix eigenvalue problem
**

19.1 19.2 19.3 19.4 19.5 19.6 19.7 Concepts The eigenvalue problem Properties of the eigenvectors Matrix diagonalization Quadratic forms Complex matrices Exercises

532

532 534 537 543 546 551 555

20

Numerical methods

20.1 20.2 20.3 20.4 20.5 20.6 20.7 20.8 20.9 20.10 20.11 Concepts Errors Solution of ordinary equations Interpolation Numerical integration Methods in linear algebra Gauss elimination for the solution of linear equations Gauss–Jordan elimination for the inverse of a matrix First-order differential equations Systems of differential equations Exercises

558

558 558 562 566 573 581 581 584 585 590 592

21

**Probability and statistics
**

21.1 21.2 21.3 21.4 21.5 21.6 21.7 21.8 Concepts Descriptive statistics Frequency and probability Combinations of probabilities The binomial distribution Permutations and combinations Continuous distributions The Gaussian distribution

595

595 595 601 603 604 607 613 615

xii

Contents

21.9 21.10 21.11 21.12

More than one variable Least squares Sample statistics Exercises

618 619 623 624

Appendix. Standard integrals Solutions to exercises Index

627 631 653

**1 Numbers, variables, and units
**

1.1 Concepts

Chemistry, in common with the other physical sciences, comprises (i) experiment: the observation of physical phenomena and the measurement of physical quantities, and (ii) theory: the interpretation of the results of experiment, the correlation of one set of measurements with other sets of measurements, the discovery and application of rules to rationalize and interpret these correlations. Both experiment and theory involve the manipulation of numbers and of the symbols that are used to represent numbers and physical quantities. Equations containing these symbols provide relations amongst physical quantities. Examples of such equations are 1. the equation of state of the ideal gas pV = nRT 2. Bragg’s Law in the theory of crystal structure nλ1= 2d1sin1θ 3. the Arrhenius equation for the temperature dependence of rate of reaction k = Ae

− Ea / RT

(1.1)

(1.2)

(1.3)

4. the Nernst equation for the emf of an electrochemical cell E = Eo − RT ln Q nF (1.4)

When an equation involves physical quantities, the expressions on the two sides of the equal sign1 must be of the same kind as well as the same magnitude.

1 The sign for equality was introduced by Robert Recorde (c. 1510–1558) in his The whetstone of witte (London, 1557); ‘I will sette as I doe often in woorke use, a paire of paralleles, or Gemowe (twin) lines of one lengthe, thus: , bicause noe.2. thynges can be moare equalle.’

2

Chapter 1 Numbers, variables, and units

EXAMPLE 1.1 The equation of state of the ideal gas, (1.1), can be written as an equation for the volume, V= nRT p

in which the physical quantities on the right of the equal sign are the pressure p of the gas, the temperature T, the amount of substance n, and the molar gas constant R = 8.31447 J K−1 mol−1. We suppose that we have one tenth of a mole of gas, n1=10.1 mol, at temperature T1=1298 K and pressure p1=1105 Pa. Then V= nRT 0.1 mol × 8.31447 J K −1 mol−1 × 298 K = p 105 Pa 0.1 × 8.31447 × 298 mol J K −1 mol−1 K = × Pa 105 = 2.478 × 10−3 m 3 The quantities on the right side of the equation have been expressed in terms of SI units (see Section 1.8), and the combination of these units is the SI unit of volume, m3 (see Example 1.17).

Example 1.1 demonstrates a number of concepts: (i) Function. Given any particular set of values of the pressure p, temperature T, and amount of substance n, equation (1.1) allows us to calculate the corresponding volume V. The value of V is determined by the values of p, T, and n; we say V is a function of p, T, and n. This statement is usually expressed in mathematics as V1=1f ( p, T, n) and means that, for given values of p, T and n, the value of V is given by the value of a function f( p, T, n). In the present case, the function is f( p, T, n)1=1nRT2p. A slightly different form, often used in the sciences, is V1=1V( p, T, n) which means that V is some function of p, T and n, which may or may not be known. Algebraic functions are discussed in Chapter 2. Transcendental functions, including the trigonometric, exponential and logarithmic functions in equations (1.2) to (1.4), are discussed in Chapter 3.

1.2 Real numbers

3

(ii) Constant and variable. Equation (1.1) contains two types of quantity: Constant: a quantity whose value is ﬁxed for the present purposes. The quantity R1=18.31447 J K−1 mol−1 is a constant physical quantity.2 A constant number is any particular number; for example, a1=10.1 and π1=13.14159 = Variable: a quantity that can have any value of a given set of allowed values. The quantities p, T, and n are the variables of the function f ( p, T, n)1=1nRT2p. Two types of variable can be distinguished. An independent variable is one whose value does not depend on the value of any other variable. When equation (1.1) is written in the form V1=1nRT2p, it is implied that the independent variables are p, T, and n. The quantity V is then the dependent variable because its value depends on the values of the independent variables. We could have chosen the dependent variable to be T and the independent variables as p, V, and n; that is, T =1pV2nR. In practice, the choice of independent variables is often one of mathematical convenience, but it may also be determined by the conditions of an experiment; it is sometimes easier to measure pressure p, temperature T, and amount of substance n, and to calculate V from them. Numbers are discussed in Sections 1.2 to 1.4, and variables in Section 1.5. The algebra of numbers (arithmetic) is discussed in Section 1.6. (iii) A physical quantity is always the product of two quantities, a number and a unit; for example T1=1298.15 K or R1=18.31447 J K−1 mol−1. In applications of mathematics in the sciences, numbers by themselves have no meaning unless the units of the physical quantities are speciﬁed. It is important to know what these units are, but the mathematics does not depend on them. Units are discussed in Section 1.8.

1.2

Real numbers

The concept of number, and of counting, is learnt very early in life, and nearly every measurement in the physical world involves numbers and counting in some way. The simplest numbers are the natural numbers, the ‘whole numbers’ or signless integers 1,12,13,1= It is easily veriﬁed that the addition or multiplication of two natural numbers always gives a natural number, whereas subtraction and division may not. For example 51−131=12, but 51−16 is not a natural number. A set of numbers for which the operation of subtraction is always valid is the set of integers, consisting of all positive and negative whole numbers, and zero: −3 −2 −1 0 +1 +2 +3 -

The operations of addition and subtraction of both positive and negative integers are made possible by the rules m1+1(−n)1=1m1−1n m1−1(−n)1=1m1+1n

2

(1.5)

The values of the fundamental physical constants are under continual review. For the latest recommended values, see the NIST (National Institute of Standards and Technology) website at www.physics.nist.gov

4

Chapter 1 Numbers, variables, and units

so that, for example, the subtraction of a negative number is equivalent to the addition of the corresponding positive number. The operation of multiplication is made possible by the rules (−m)1×1(−n)1=1+(m1×1n) (−m)1×1(+n)1=1−(m1×1n) Similarly for division. Note that −m1=1(−1)1×1m. EXAMPLES 1.2 Addition and multiplication of negative numbers 21+1(−3)1=121−131=1−1 (−2)1×1(−3)1=121×131=16 (−6)1÷1(−3)1=161÷131=12 0 Exercises 1–7 In equations (1.5) and (1.6) the letters m and n are symbols used to represent any pair of integers; they are integer variables, whose values belong to the (inﬁnite) set of integers. Division of one integer by another does not always give an integer; for example 61÷131=12, but 61÷14 is not an integer. A set of numbers for which the operation of division is always valid is the set of rational numbers, consisting of all the numbers m2n1=1m1÷1n where m and n are integers (m2n, read as ‘m over n’, is the more commonly used notation for ‘m divided by n’). The deﬁnition excludes the case n1=10 because division by zero is not deﬁned (see Section 1.6), but integers are included because an integer m can be written as m21. The rules for the combination of rational numbers (and of fractions in general) are m p mq + np + = n q nq m p mp × = n q nq m p m q mq ÷ = × = n q n p np where, for example, mq means m1×1q. (1.7) 21−1(−3)1=121+131=15 (2)1×1(−3)1=1−21×131=1−6 61÷1(−3)1=1−61÷131=1−2 (1.6)

(1.8)

(1.9)

EXAMPLES 1.3 Addition of fractions (1) Add 1 1 and . 2 4

1.2 Real numbers

5

The number one half is equal to two quarters and can be added to one quarter to give three quarters: 1 1 2 1 3 + = + = 2 4 4 4 4 The value of a fraction like 122 is unchanged if the numerator and the denominator are both multiplied by the same number: 1 1× 2 2 = = 2 2×2 4 and the general method of adding fractions is (a) ﬁnd a common denominator for the fractions to be added, (b) express all the fractions in terms of this common denominator, (c) add. (2) Add 2 4 and . 3 5

A common denominator is 31×151=115. Then 2 4 2 × 5 3 × 4 10 12 10 + 12 22 + = + = + = = 3 5 3 × 5 3 × 5 15 15 15 15 (3) Add 1 5 and . 4 6

A common denominator is 41×161=124, but the lowest (smallest) common denominator is 12: 1 5 3 10 13 + = + = 4 6 12 12 12 0 Exercises 8–13 EXAMPLE 1.4 Multiplication of fractions 2 4 2×4 8 × = = 3 5 3 × 5 15 This can be interpreted as taking two thirds of 425 (or four ﬁfths of 223). 0 Exercises 14–17

6

Chapter 1 Numbers, variables, and units

EXAMPLE 1.5 Division of fractions 2 4 2 5 10 ÷ = × = 3 5 3 4 12 The number 10212 can be simpliﬁed by ‘dividing top and bottom’ by the common factor 2: 102121=1526 (see Section 1.3). 0 Exercises 18–21

Every rational number is the solution of a linear equation mx1=1n (1.10)

where m and n are integers; for example, 3x1=12 has solution x1=1223. Not all numbers are rational however. One solution of the quadratic equation x 21=12 is x = 2 , the positive square root of 2 (the other solution is − 2 ), and this number cannot be written as a rational number m2n; it is called an irrational number. Other irrational numbers are obtained as solutions of the more general quadratic equation ax 21+1bx1+1c1=10 where a, b, and c are arbitrary integers, and of higher-order algebraic equations; for example, a solution of the cubic equation x 31=12 is x = 3 2 , the cube root of 2. Irrational numbers like 2 and 3 2 are called surds. The rational and irrational numbers obtained as solutions of algebraic equations of type a01+1a1x1+1a2 x 21+1a3 x 31+1-1+1an x n1=10 (1.11)

where a0, a1, =, an are integers, are called algebraic numbers; these numbers can be expressed exactly in terms of a ﬁnite number of rational numbers and surds. There exist also other numbers that are not algebraic; they are not obtained as solutions of any ﬁnite algebraic equation. These numbers are irrational numbers called transcendental numbers; ‘they transcend the power of algebraic methods’

1.3 Factorization, factors, and factorials

7

(Euler).3 The best known and most important of these are the Euler number e and the Archimedean number π.4 These are discussed in Section 1.4. The rational and irrational numbers form the continuum of numbers; together they are called the real numbers.

1.3

Factorization, factors, and factorials

Factorization is the decomposition of a number (or other quantity) into a product of other numbers (quantities), or factors; for example 301=121×131×15 shows the decomposition of the natural number 30 into a product of prime numbers; that is, natural numbers that cannot be factorized further (the number 1 is not counted as a prime number). The fundamental theorem of arithmetic is that every natural number can be factorized as a product of prime numbers in only one way.5

EXAMPLES 1.6 Prime number factorization (1) 41=121×121=122 (2) 121=121×121×131=1221×13 (3) 3151=131×131×151×171=1321×151×17 (4) 51201=121×121×121×121×121×121×121×121×121×121×151=12101×15 0 Exercises 22–25 Factorization and cancellation of common factors can be used for the simpliﬁcation of fractions. For example, in 12 6 × 2 2 = = 42 6 × 7 7

3 Leonhard Euler (1707–1783). Born in Switzerland, he worked most of his life in St Petersburg and in Berlin. One of the world’s most proliﬁc mathematicians, he wrote ‘voluminous papers and huge textbooks’. He contributed to nearly all branches of mathematics and its application to physical problems, including the calculus, differential equations, inﬁnite series, complex functions, mechanics, and hydrodynamics, and his name is associated with many theorems and formulas. One of his important, if unspectacular, contributions was to mathematical notation. He introduced the symbol e, gave the trigonometric functions their modern deﬁnition, and by his use of the symbols sin, cos, i, and π made them universally accepted. 4 The symbol π was ﬁrst used by William Jones (1675–1749) in a textbook on mathematics, Synopsis palmariorum matheseos (A new introduction to the mathematics) in 1706. Euler’s adoption of the symbol ensured its acceptance. 5 A version of the fundamental theorem of arithmetic is given by Propositions 31 and 32 in Book VII of Euclid’s Stoichia (Elements). Euclid was one of the ﬁrst teachers at the Museum and Library of Alexandria founded by Ptolemy I in about 300 BC after he had gained control of Egypt when Alexander’s empire broke up in 323 BC.

8

Chapter 1 Numbers, variables, and units

cancellation of the common factor 6 is equivalent to dividing both numerator and denominator by 6, and such an operation does not change the value of the fraction.

EXAMPLES 1.7 Simpliﬁcation of fractions (1) 4 22 2×2 1 = 3 = = 24 2 × 3 2 × 2 × 2 × 3 6 15 3 × 5 3 = 2 = 25 5 5 105 3×5×7 1 1 = = = 2 1470 2 × 3 × 5 × 7 2 × 7 14

(2)

(3)

0 Exercises 26–29

In general, the purpose of factorization is to express a quantity in terms of simpler quantities (see Section 2.3 for factorization of algebraic expressions).

Factorials

The factorial of n is the number whose factors are the ﬁrst n natural numbers: n!1=111×121×131×1-1×1n =1n1×1(n1−11)1×1(n1−12)1×1-1×121×11 (1.12)

(read as ‘n factorial’). Consecutive factorials are related by the recurrence relation (n1+11)!1=1(n1+11)1×1n! for example, 3!1=131×121×111=16 and 4!1=141×131×121×111=141×13!1=124. In addition, the factorial of zero is deﬁned as 0!1=11.

EXAMPLES 1.8 Factorials (1) 1!1=111×10!1=111×111=11 (2) 5!1=151×14!1=151×141×13!1=151×141×131×12!1=151×141×131×121=1120 (3) (4) 5! 5 × 4 × 3 ! = = 5 × 4 = 20 3! 3! 7! 7 × 6 × 5 × 4 ! = = 7 × 5 = 35 3!4! 3 × 2 × 4!

0 Exercises 30–36

1.4 Decimal representation of numbers

9

1.4

**Decimal representation of numbers
**

These are the nine ﬁgures of the Indians 9 8 7 6 5 4 3 2 1

With these nine ﬁgures, and with this sign 0 which in Arabic is called zephirum, any number can be written, as will below be demonstrated. (Fibonacci)6

In the decimal system of numbers, the ten digit symbols 0 to 9 (Hindu-Arabic numerals)7 are used for zero and the ﬁrst nine positive integers; the tenth positive integer is denoted by 10. A larger integer, such as ‘three hundred and seventy-two’ is expressed in the form 3001+1701+121=131×11021+171×1101+12 and is denoted by the symbol 372, in which the value of each digit is dependent on its position in the symbol for the number. The decimal system has base 10, and is the only system in common use. Although rational numbers can always be expressed exactly as ratios of integers, this is not so for irrational numbers. For computational purposes, a number that is not an integer is conveniently expressed as a decimal fraction;8 for example, 5241=11.25. The general form of the decimal fraction (integral part).(fractional part) consists of an integer to the left of the decimal point, the integral part of the number, and one or more digits to the right of the decimal point, the decimal or fractional part of the number. The value of each digit is determined by its position; for example 234.567 = 200 + 30 + 4 + 5 6 7 + + 10 100 1000

**=121×11021+131×11011+141×11001+151×110−11+161×110−21+171×110−3
**

Leonardo of Pisa, also called Fibonacci (c. 1170–after 1240). The outstanding mathematician of the Latin Middle Ages. In his travels in Egypt, Syria, Greece, and Sicily, Fibonacci studied Greek and Arabic (Muslim) mathematical writings, and became familiar with the Arabic positional number system developed by the Hindu mathematicians of the Indus valley of NW India. Fibonacci’s ﬁrst book, the Liber abaci, or Book of the Abacus, (1202, revised 1228) circulated widely in manuscript, but was published only in 1857 in Scritti di Leonardo Pisano. The ﬁrst chapter opens with the quotation given above in the text. 7 One of the principal sources by which the Hindu-Arabic decimal position system was introduced into (Latin) Europe was Al-Khwarizmi’s Arithmetic. Muhammad ibn Musa Al-Khwarizmi (Mohammed the son of Moses from Khorezm, modern Khiva in Uzbekistan) was active in the time of the Baghdad Caliph Al-Mamun (813–833), and was probably a member of his ‘House of Wisdom’ (Academy) at a time when Baghdad was the largest city in the world. Al-Khwarizmi’s Algebra was widely used in Arabic and in Latin translation as a source on linear and quadratic equations. The word algorithm is derived from his name, and the word algebra comes from the title, Liber algebrae et almucabala, of Robert of Chester’s Latin translation (c. 1140) of his work on equations. 8 The use of decimal fractions was introduced into European mathematics by the Flemish mathematician and engineer Simon Stevin (1548–1620) in his De Thiende (The art of tenths) in 1585. Although decimal fractions were used by the Chinese several centuries earlier, and the Persian astronomer Al-Kashi used decimal and sexagesimal fractions in his Key to Arithmetic early in the ﬁfteenth century, the common use of decimal fractions in European mathematics can be traced directly to Stevin, especially after John Napier modiﬁed the notation into the present one with the decimal point (or decimal comma as is used in much of continental Europe). It greatly simpliﬁed the operations of multiplication and division.

6

10

Chapter 1 Numbers, variables, and units

where 10−n1=11210n and 1001=11 (see Section 1.6). 0 Exercises 37–42 A number with a ﬁnite number of digits after (to the right of ) the decimal point can always be written in the rational form m2n; for example 1.2341=1123421000. The converse is not always true however. The number 123 cannot be expressed exactly as a ﬁnite decimal fraction: 1 = 0.333… 3 the dots indicating that the fraction is to be extended indeﬁnitely. If quoted to four decimal places, the number has lower and upper bounds 0.3333 and 0.3334: 0.3333 < 1 < 0.3334 3

where the symbol < means ‘is less than’; other symbols of the same kind are ≤ for ‘is less than or equal to’, > for ‘is greater than’, and ≥ for ‘is greater than or equal to’. Further examples of nonterminating decimal fractions are 1 = 0.142857 142857…, 7 1 = 0.083333 333333… 12

In both cases a ﬁnite sequence of digits after the decimal point repeats itself indeﬁnitely, either immediately after the decimal point, as the sequence 142857 in 127, or after a ﬁnite number of leading digits, as 3 in 1212. This is a characteristic property of rational numbers. EXAMPLE 1.9 Express 1213 as a decimal fraction. By long division, 0.07692307 ... 13) 1.00 91 90 78 120 117 30 26 40 39 100

The rational number 12131=10.076923 076923= is therefore a nonterminating decimal fraction with repeating sequence 076923 after the decimal point. 0 Exercises 43–46

1.4 Decimal representation of numbers

11

An irrational number cannot be represented exactly in terms of a ﬁnite number of digits, and the digits after the decimal point do not show a repeating sequence. The number 2 has approximate value to 16 signiﬁcant ﬁgures, 2 =11.414213 562373 095= and can, in principle, be computed to any desired accuracy by a numerical method such as the Newton–Raphson method discussed in Chapter 20.9

The Archimedean number π

The number π is deﬁned as the ratio of the circumference of a circle to its diameter. It is a transcendental number,10 and has been computed to many signiﬁcant ﬁgures; it was quoted to 127 decimal places by Euler in 1748. Its value to 16 signiﬁcant ﬁgures is

π1=13.14159 26535 897931=

The value of π has been of practical importance for thousands of years. For example, an Egyptian manuscript dated about 1650 BC (the Rhind papyrus in the British Museum) contains a prescription for the calculation of the volume of a cylindrical granary from which the approximate value 2562811≈13.160 can be deduced. A method for generating accurate approximations was ﬁrst used by Archimedes11 who determined the bounds 223 22 <π< 71 7 and the upper bound has an error of only 2 parts in a thousand.

9 A clay tablet (YBC 7289, Yale Babylonian Collection) dating from the Old Babylonian Period (c. 1800–1600 BC) has inscribed on it a square with its two diagonals and numbers that give 2 to three sexagesimal places: 2 = 1, 24, 51, 10 = 1 + 24260 + 512602 + 102603 ≈ 1.41421296, correct to 6 signiﬁcant decimal ﬁgures. 10 The proof of the irrationality of π was ﬁrst given in 1761 by Johann Heinrich Lambert (1728–1777), German physicist and mathematician. He is also known for his introduction of hyperbolic functions into trigonometry. The number π was proved to be transcendental by Carl Louis Ferdinand von Lindemann (1852–1939) in 1882 by a method similar to that used by Hermite for e. 11 Archimedes (287–212 BC) was born in Syracuse in Sicily. He made contributions to mathematics, mechanics, and astronomy, and was a great mechanical inventor. His main contributions to mathematics and the mathematical sciences are his invention of methods for determining areas and volumes that anticipated the integral calculus and his discoveries of the ﬁrst law of hydrostatics and of the law of levers.

all numbers are given with a ﬁxed number of decimal places.51+10.000198 +10. 12 Charles Hermite (1822–1901).358. More generally. 003.71828 18284 59045= The value of e can be computed from the series to any desired accuracy. with zeros on the left of a number not counted.32101×1104. is known for his work in algebra and number theory. 1.10 Show that the sum of the ﬁrst 10 terms of the series gives an approximate value of e that is correct to at least 6 signiﬁcant ﬁgures. e = 1+ 1+ 1 1 1 1 1 1 1 1 1 + + + + + + + + + 2 6 24 120 720 5040 40320 362880 3628800 8 ≈111+111+10. 62. French mathematician. professor at the Sorbonne. and units The Euler number e The number e is deﬁned by the ‘inﬁnite series’ (see Chapter 7) e = 1+ 1 1 1 1 + + + + 1! 2! 3! 4! =12.12 Chapter 1 Numbers.71828 The value is correct to the 6 ﬁgures quoted because every additional term in the series is at least ten times smaller than the preceding one. for example. The Hermite differential equation and the Hermite polynomials are important in the solution of the Schrödinger equation for the harmonic oscillator. The number was shown to be a transcendental number by Hermite in 1873.0000251+10.0083331+10.0013891+10.0000031+10.0416671+10.1666671+10. a number in the decimal system is approximated either with some given number of decimal places or with a given number of signiﬁcant ﬁgures. For example.142.32101×1101. 0. variables.12 EXAMPLE 1.0000003 ≈12. used more widely in the sciences. In the ﬂoating-point representation.0032101=10.000 have 3 decimal places. .013. and the result of an arithmetic operation is also approximate. 0. 3. arithmetic involving only integers gives exact answers (unless the numbers are too large to be written). 32101=10.32101×110−2 all have 4 signiﬁcant ﬁgures. In the ﬁxed-point representation.2101=10. Signiﬁcant ﬁgures and rounding In practice. the numbers are given with a ﬁxed number of ‘signiﬁcant ﬁgures’. His work on the algebra of complex numbers (‘Hermitian forms’) became important in the formulation of quantum theory.

1.14159 is truncated to 3. the number is rounded up. xn} is a set of objects.1416. If the domain of the variable x is an interval a to b. (ii) If the ﬁrst digit dropped is less than 5.1415.1x2. In some cases however a physical quantity can have values. 7. 7. 3. variables are used to represent both numbers and physical quantities. For example. If {x1. Truncation is not recommended because it can lead to serious computational errors. In the ideal-gas example discussed in Section 1. the preceding digit is increased by 1. the set forms the domain of the variable. Typically. the preceding digit is left unchanged. If the set consists of only one value then the variable is called a constant variable.2.1x3. 7. and T are continuous variables whose numerical values can in principle be any positive real numbers.5 Variables In the foregoing sections. The most widely accepted rules for rounding are: (i) If the ﬁrst digit dropped is greater than or equal to 5. a1≤1x1≤1b then x is a continuous variable in the interval. to 4 decimal places or 5 signiﬁcant ﬁgures. 0 Exercises 47–49 1. If the domain consists of a discrete set of values. If the domain consists of integers. and can have any value in the continuous range of values a to b (including a and b). 7. then a variable x can be deﬁned in terms of this set such that x can have as its value any member of the set. an integer variable is used for the counting and the counted objects form a sample of some discrete set. x2. 2. the physical quantities p.4 Errors arising from truncation and rounding are discussed in Section 20. that is. the objects of the set are real numbers. and a real variable can have as its domain either the whole continuum of real numbers or a subset thereof. n. In (real) number theory. or simply a constant. V. the number is rounded down. For example.1.1=. A quantity that can take as its value any value chosen from a given set of values is called a variable. Discrete variables are normally involved whenever objects are counted as opposed to measured.3628. for example the n numbers x1. and 1 decimal places. symbols (letters) have been used to represent arbitrary numbers. obtained by rounding up.363. 3.36284 is 7. =. by removing or replacing by zeros all superﬂuous digits on the right. then x is called a discrete variable.1xn. A more sensible (accurate) approximation of π to ﬁve ﬁgures is 3. In the physical sciences. x3. This is the case for the energy levels and the observed spectral frequencies of an atom or molecule. x is an integer variable. not necessarily numbers. for 4. .36.5 Variables 13 A number whose exact (decimal) representation involves more than a given number of digits is reduced most simply by truncation. some of which belong to a discrete set and others to a continuous set.

in which the motion of the electron is conﬁned to the vicinity of the nucleus. with all positive energies. is to place the dot on the line for multiplication (2.14 Chapter 1 Numbers. × and ÷ (or 2) are called arithmetic operators. . a(bc)1=1(ab)c 5. the result of multiplying two numbers. and units EXAMPLE 1. variables. are 1. a1+1b1=1b1+1a 2. The corresponding states of the atom are those of a free (unbound) electron moving in the presence of the electrostatic ﬁeld of the nuclear charge. for the decimal point (2. −. a(b1+1c)1=1ab1+1ac (commutative law of addition) (commutative law of multiplication) (associative law of addition) (associative law of multiplication) (distributive law) The operations of addition and multiplication and their inverses. The symbols +. are called arithmetic operations. (ii) Continuous energy levels. Let a. 3. 1. n1=11. a1+1b. The basic rules for the combination of real numbers. as it easily confounded with x = often I simply relate two quantities by an interposed dot’.5 = 522). and c be variables whose values can be any real numbers. a1+1(b1+1c)1=1(a1+1b)1+1c 4. still widely used. E1>10. and it allows the formulation of a set of rules for the manipulation of numbers. ab1=1ba 3. on the line.11 The spectrum of the hydrogen atom The energy levels of the hydrogen atom are of two types: (i) Discrete (quantized) energy levels with (negative) energies given by the formula (in atomic units. subtraction and division. = The corresponding states of the atom are the ‘bound states’. The set of rules is called the algebra. b. It is becoming accepted practice to place the ‘dot’ in the ‘high position’ to denote multiplication (21·15 = 2 × 5) and in the ‘low position’. Transitions between these energy levels and those of the bound states give rise to continuous ranges of spectral frequencies. The result of adding two numbers. the algebra of real numbers or the arithmetic. ab1=1a1×1b1=1a1·1b.5 = 2 × 5) and high for the decimal point (21·15 = 522).13 13 In 1698 Leibniz wrote in a letter to Johann Bernoulli: ‘I do not like × as a symbol for multiplication. is called the product of a and b.6 The algebra of real numbers The importance of the concept of variable is that variables can be used to make statements about the properties of whole sets of numbers (or other objects). 2. Transitions between the energy levels give rise to discrete lines in the spectrum of the atom.8) En = − 1 2 n2 . is called the sum of a and b. see Section 1. An alternative convention.

a1+101=101+1a1=1a 7.6 The algebra of real numbers 15 EXAMPLES 1. If it were a number then.1. and equal to −a if a is negative: + a if a > 0 |a|= − a if a < 0 For example. and @(21×13)1×141=161×141=124 !21×1(31+14)1=121×171=114. for positive values of a. The modulus of a real number a is deﬁned as the positive square root of a 2. and @(21+13)1+141=151+141=19 !21×1(31×14)1=121×1121=124. For example. | 3 | = 3 and | −3 | = 3. a1×111=111×1a1=1a (addition of zero) (multiplication by zero) (multiplication by unity) We have already seen that subtraction of a number is the same as addition of its negative. the equations 1201=1∞ and 2201=1∞ would imply 11=12. (1.12 Examples of the rules of arithmetic rule 1. It is the ‘magnitude’ of the number. a(b1+1c)1=1ab1+1ac examples 21+131=131+121=15 21×131=131×121=16 !21+1(31+14)1=121+171=19. a1+1b1=1b1+1a 2. However. and @21×1(31+14)1=1(21×13)1+1(21×14)1=161+181=114 −2(31+14)1=1(−21×13)1+1(−21×14)1=1−61−181=1−14 −2(31−14)1=1−21×131−121×1(−4)1=1−61+181=12 A corollary to rule 5 is (a1+1b)(c1+1d)1=1a(c1+1d)1+1b(c1+1d) (21+13)(41+15)1=12(41+15)1+13(41+15)1=1181+1271=145 Three rules deﬁne the properties of zero and unity: 6.13) . there is no number whose inverse is zero. | a | = + a 2 (read as ‘mod a’). by the laws of algebra. and that division by a number is the same as multiplication by its inverse. it is not a number. the number 12a. division by zero is not deﬁned. ab1=1ba 3. a1×101=101×1a1=10 8. a1+1(b1+1c)1=1(a1+1b)1+1c 4. equal to +a if a is positive. a(bc)1=1(ab)c 5. we say that 12a tends to inﬁnity as a tends to zero: 1 → ∞ as a → 0 a Although ‘inﬁnity’ is represented by the symbol ∞. becomes arbitrarily large as the value of a approaches zero.

a02a n1=112a n1=1a−n so that the inverse of a n is a−n. 12a1=1a−1. 1001=1102 with base 10 and exponent 2. Also. setting m1=10 in rule 10. 232231=123−31=1201=11 because 232231=11. In particular. a 3a 21=1(a1×1a1×1a)1×1(a1×1a)1=1a1×1a1×1a1×1a1×1a1=1a 51=1a 3+2 Three auxiliary rules are 10. a m2a n1=1a m−n 11. for example. a a 1=1a m n m+n examples (a) 231×1221=123+21=125 (b) 361×13−31=136−31=133 (c) 21221×121241=12122+1241=12324 10. In practice.16 Chapter 1 Numbers. a m is the mth power of a. the number am is read ‘a to the power m’ or ‘a to the m’. and 161=124 with base 2 and exponent 4. (a m)n1=1(a n)m1=1a m×n 12.13 The index rule rule 9. where a is called the base and m is the index or exponent. setting m1=1n. (ab)m1=1a mbm (index rule) Rule 10 deﬁnes numbers with zero and negative exponents. a m2a n1=1a m−n (d) 2324221241=12324−1241=12122 (e) 2422−21=124−(−2)1=124+21=126 (f ) 342341=134−41=1301=11 . (−a)31=1(−a)1×1(−a)1×1(−a)1=1(−1)31×1a 31=1−a 3 Numbers are also deﬁned with negative and non-integral exponent. EXAMPLES 1. a ma n1=1a m+n For example. a n2a n1=1a n−n1=1a01=11 and any number raised to power zero is unity. The rule for the product of numbers in base–index form is 9. for example. Thus. a 31=1a1×1a1×1a. even when m is not a positive integer. for m1=13. variables. When m is a positive integer. and units The index rule Numbers are often written in the form a m.

they apply to all numbers written in the base–index form. 43221=1(43)1221=1(4122)31=18 Although the index rules have been demonstrated only for integral and rational indices.6 The algebra of real numbers 17 11. For example.6 for real exponents and Chapter 8 for complex exponents). 2123 is a cube root of 2 because (2123)31=12(123)×31=1211=12. for positive integer m. When the exponent m is a variable. (ab)m1=1a mbm (k) (21×13)21=1221×132 (l) (−8)1231=1(−1)1231×181231=1(−1)1×121=1−2 0 Exercises 50–65 Example 1.13(h) shows that 21221×121221=12. a m n = n a m = ( n a )m so that a m2n is both the nth root of the mth power of a and the mth power of the nth root. equivalently. rule 11 gives a m2n1=1(a m)12n1=1(a12n)m or. If x1=1a m then m1=1loga x is the logarithm of x to base a (see Section 3. a m is called an exponential function (see Section 3.1. the square root of 2. The expression can be interpreted in two ways: (21+13)1×141=151×141=120 . a12m is the mth root of a: a1 m = m a Thus. Rules of precedence for arithmetic operations An arithmetic expression such as 21+131×14 is ambiguous because its value depends on the order in which the arithmetic operations are applied.7). (a m)n1=1(a n)m1=1a m×n (g) (22)31=1(22)1×1(22)1×1(22)1=122×31=126 (h) (2122)21=12(122)×21=1211=12 (i) (23)4231=1(2423)31=123×4231=124 (j) ( 2 2 ) 2 =2 2× 2 = 22 = 4 12. for rational exponent m2n. It follows that 21 2 = 2 . More generally. In general.

Multiplication and division take precedence over addition and subtraction. the convention is to use parentheses as the innermost brackets. increasing sizes of brackets can help to clarify the structure of the expression.14 Rules of precedence for arithmetic operations 1 21+131×141=121+1(31×14)1=121+1121=114 (1) 2 3but (21+13)1×141=151×141=120 1 21+131×141×151+161=121+1(31×141×15)1+161=121+1601+161=168 (2) 2 3but (21+13)1×141×1(51+16)1=151×141×1301=1600 1 21+1321=121+191=111 (3) 2 2 2 3but (21+13) 1=15 1=125 1 91+1161221=191+141=111 (4) 2 122 122 3but (91+116) 1=1(25) 1=15 1 31×1421=131×1(42)1=131×1161=148 (5) 2 2 2 3but (31×14) 1=1(12) 1=1144 1 21×161÷131=1(21×16)1÷131=1121÷131=14 (6) 2 3and 21×161÷131=121×1(61÷13)1=121×121=14 (rule 3) (rule 1) (rule 3) (rule 1) (rule 2) (rule 1) (rule 2) (rule 1) (rule 2) (rule 1) . If in doubt use brackets. Exponentiation (taking powers) takes precedence over multiplication2division and addition2subtraction. variables.18 Chapter 1 Numbers. Evaluation then proceeds from the innermost bracketed expressions outwards. Brackets take precedence over arithmetic operators. then square brackets. and units with the parentheses indicating that the addition is to performed ﬁrst. 3. Arithmetic expressions are generally evaluated by following the rules of precedence for arithmetic operations: 1. In case of more complicated expressions. then braces (curly brackets). 2. 4. Addition and subtraction are performed last. and 21+1(31×14)1=121+1121=114 in which the multiplication is performed ﬁrst. EXAMPLES 1. containing nested brackets. for example {(2 + 3) × 4 + 5} × 6 = {5 × 4 + 5} × 6 = {20 + 5} × 6 = 25 × 6 = 150 As shown. Ambiguities if this kind can always be resolved by the proper use of parentheses or other brackets.

0 Exercises 66–77 in case (4): (91+116)1221≠191221+116122 1. where ≠ means ‘is not equal to’. For example. The metre is a constant physical quantity . For example. Thus. magnitude and dimensions.7 Complex numbers The solutions of algebraic equations are not always real numbers. the quantity ‘2 metres’ has the dimensions of length and has magnitude equal to twice the magnitude of the metre. it is only necessary to remember to replace i2 by −1 whenever it occurs. 79 1. For example. in case (3): (21+13)21≠1221+132. EXAMPLE 1. They are discussed in greater detail in Chapter 8.8 Units A physical quantity has two essential attributes. They are incorporated into the system of numbers by deﬁning the square root of −1 as a new number which is usually represented by the symbol i (sometimes j) with the property i 21=1−1 The two square roots of an arbitrary negative real number −x2 are then ix and −ix. −16 = (16) × ( −1) = 16 × −1 = ±4i Such numbers are called imaginary to distinguish them from real numbers.15 Find the sum and product of the complex numbers z11=121+13i and z21=141−12i.8 Units 19 What not to do: (a1+1b)n1≠1a n1+1bn. Addition: Multiplication: z11+1z21=1(21+13i)1+1(41−12i)1=1(21+14)1+1(3i1−12i)1=161+1i z1z21=1(21+13i)(41−12i)1=12(41−12i)1+13i(41−12i) =181−14i1+112i1−16i21=181+18i1+161=1141+18i 0 Exercises 78.2. More generally. the number z1=1x1+1iy where x and y are real is called a complex number.1. the solutions of the equation x 2 = −1 are x = ± −1 and these are not any of the numbers described in Section 1. Complex numbers obey the same rules of algebra as real numbers.

velocity (or more precisely. examples are relative density. For example. conform with the dimensions. a number of quantities that are particularly important in the physical sciences have been given SI names and symbols. however.1.nist. and units that deﬁnes the dimensions of the quantity and provides a scale for the speciﬁcation of the magnitude of an arbitrary length. In general. and mole fraction. The dimensions of a physical quantity are independent of the system of units used to describe its value. In addition to the base units. many tailored to the needs of particular disciplines in the sciences.1 Base physical quantities and SI units Physical quantity length mass time electric current temperature amount of substance luminous intensity Symbol l m t I T n Iv Dimension L M T I Z N J Name of SI unit metre kilogram second ampere kelvin mole candela Symbol for SI unit m kg s A K mol cd The symbols in column 3 deﬁne the dimensions of the base physical quantities. relative molar mass.20 Chapter 1 Numbers. it is a unit of length.2). the corresponding SI unit of velocity is metre per second. variables. Every physical quantity has an SI unit determined by its dimensionality. The recommended system for the physical sciences. Every system of units must.16(i)). m. The SI units of length and time are the metre. IUPAC (International Union of Pure and Applied Chemistry) provides the standard on chemical nomenclature and terminology. 14 SI (Système International d’Unités) is the international standard for the construction and use of units (see the NIST website at www. and has dimensions of length divided by time.2. m2s1=1m s−1 (see Example 1.1.physics. a physical quantity is the product of a number and a unit. . and for chemistry in particular. is the International System of Units (SI)14 which is based on the seven base units whose names and symbols are listed in columns 4 and 5 in Table 1.org). LT −1. and on the measurement and evaluation of data (see www. and the dimensions of all other quantities (derived quantities) can be expressed in terms of them. A variety of systems of units are in use. and the second.iupac. All physical quantities can be expressed in terms of the seven ‘base’ quantities whose names and symbols are listed in the ﬁrst two columns of Table 1. s.gov). This is the case for a quantity that is the ratio of two others with the same dimensions. In addition. Table 1. A less obvious example is (plane) angle which is deﬁned as the ratio of two lengths (see Section 3. Some of these are listed in Table 1. speed) is distance travelled in unit time. l2t1=1lt −1. We note that some physical quantities have no dimensions.

yd min−1. and has dimensions of length2time: LT −1. 1 min1=160 s. the unit of velocity is yards per minute. with SI unit m s−2.01524 m s−1 (ii) Acceleration is rate of change of velocity with time. In SI. heat power electric charge electric potential electric capacitance electric resistance electric conductance magnetic ﬂux magnetic ﬂux density inductance plane angle solid angle Name hertz newton pascal joule watt coulomb volt farad ohm siemens weber tesla henry radian steradian Symbol Hz N Pa J W C V F Ω S Wb T H rad sr Description events per unit time mass × acceleration force per unit area force × distance work per unit time current × time work per unit charge charge per unit potential potential per unit current current per unit potential work per unit current magnetic ﬂux per unit area magnetic ﬂux per unit current angle subtended by unit arc at centre of unit circle solid angle subtended by unit surface at centre of unit sphere SI unit s−1 kg m s−2 N m−2 Nm J s−1 As J C−1 C V−1 V A−1 Ω−1 J A−1 Wb m−2 Wb A−1 1 1 0 Exercises 80–90 EXAMPLES 1. In a system in which. In general.16 Dimensions and units (i) Velocity is rate of change of position with time. the unit of velocity is meters per second.9144260) m s−11=10. and has dimensions of velocity2time: [LT −1]1×1[T −1]1=1LT −2. .8 Units 21 Table 1.2 SI derived units with special names and symbols Physical quantity frequency force pressure energy.1. the unit of a derived quantity is obtained by replacing each base quantity by its corresponding unit. the unit of length is the yard (yd) and the unit of time is the minute (min). for example. and 1 yd min−11=1(0.9144 m (exactly). This ‘non-SI’ unit is expressed in terms of the SI unit by means of conversion factors deﬁned within SI. Thus 1 yd1=10. m s−1.9144 m)1×1(60 s)−11=1(0. work.

665 Gal. J1=1N m1=1kg m2 s−2 and kinetic energy 1 2 mv 2 has dimensions of mass × (velocity)2: [M]1×1[LT −1]21=1ML2T −2. N1=1kg m s−2. Pa1=1N m−21=1kg m−1 s−2 Widely used alternative non-SI units for pressure are: ‘standard pressure’: atmosphere: torr: bar1=1105 Pa atm1=1101325 Pa Torr1=1(1013252760) Pa1≈1133. Dimensional analysis is the name given to the checking of equations for dimensional consistency. 0 Exercises 91–94 Dimensional analysis The terms on both sides of an equation that contains physical quantities must have the same dimensions. with SI unit the pascal. (iii) Force has dimensions of mass × acceleration: [M]1×1[LT −2]1=1MLT −2. with SI unit the joule.17 For the ideal-gas equation pV1=1nRT. For nRT. with the same dimensions and unit.80665 m s−21=1980. EXAMPLE 1. work has dimensions of force × distance: [MLT −2]1×1[L]1=1ML2T −2. the dimensions of pV (using Tables 1. energy and heat are quantities of the same kind. with SI unit the newton.22 Chapter 1 Numbers. Thus.1 and 1. and units The standard acceleration of gravity is g1=19. (iv) Pressure has dimensions of force per unit area: [MLT −2]2[L2]1=1ML−1T −2.1).322 Pa (v) Work. (mol)(J K−1 mol−1)(K)1=1J . variables. where Gal1=110−2 m s−2 (cm s−2) is called the galileo. The corresponding expression in terms of SI units is Pa1×1m31=1N m−21×1m31=1N m1=1J. equation (1.2) are those of work (or energy): [ML−1T −2]1×1[L3]1=1ML2T −2.

012 kg) of 12C. particularly when the properties of individual atoms and molecules are considered. the dimensions of V are energy2pressure.022141×11023) g1≈11. Pa N m −2 0 Exercise 95 Large and small units Decimal multiples of SI units have names formed from the names of the units and the preﬁxes listed in Table 1. For example. This number is given by Avogadro’s constant. NA1≈16. It follows that when equation (1. the mole is deﬁned as the amount of substance that contains as many elementary entities (atoms or molecules) as there are atoms in 12 g (0. with symbol Da). a picometre is pm1=110−12 m.3. These units of length are frequently used in chemistry. where u1=112(6.8 Units 23 as required. a decimetre is dm1=110−1 m. . as in Example 1.1. and concentrations in moles per decimetre cube.660541×110−27 kg is called the uniﬁed atomic mass unit (sometimes called a Dalton.022141×11023) g1≈121×110−26 kg or m(12C)1=112 u.3 SI preﬁxes Multiple 10 10 10 2 3 Preﬁx deca hecto kilo mega giga tera peta exa zetta yotta Symbol da h k M G T P E Z Y Multiple 10 10 10 −1 −2 −3 Preﬁx deci centi milli micro nano pico femto atto zepto yocto Symbol d c m µ 106 109 10 10 12 15 10−6 10−9 10 10 −12 −15 n p f a z y 1018 1021 10 24 10−18 10−21 10 −24 0 Exercises 96–103 The quantities that are of interest in chemistry often have very different magnitudes from those of the SI units themselves. with SI unit J Nm = = m 3 for the volume. Table 1.022141×11023 mol−1.1. The mass of an atom of 12C is therefore m(12C)1=1122(6. For example. molecular bond lengths in picometres.1) is written in the form V1=1nRT2p. mol dm−31=1103 mol m−3.

0105 g mol−11=10.12811×110−101m.75 pm. Atomic and molecular masses are often given as relative masses: Ar for an atom and Mr for a molecule.13851×110−26 kg)1×1(1.01801 kg mol−1. on a scale on which Ar(12C)1=112.660541×110−27 kg1=11. Thus for CO.18 Molecular properties: mass.13851×110−26 kg The bond length of CO is 112.811pm1=11.20751×110−10 m. Then µ (12 C16O) = 12 × 15. Ar(1H)1= 1. and these are the atomic masses in units of the uniﬁed atomic mass unit u. 0 Exercises 104. 105 . and the mass of the individual molecule is m(1H216O)1=1Mr(1H216O)1×1u1=12.8562 u 27. for O2. where µ is the reduced mass. variables. Thus. and the Ångström Å1=110−10 m or the Bohr radius a01=10. separated by distance R is I1=1µR2. so that the moment of inertia of the molecule is I1=1µR 21=1(1.0078 and Ar(16O)1=115. mA and mB.9948. and units EXAMPLES 1. given by 1 µ = 1 1 + .85621×11.99071×110−26 kg (ii) length.12811×110−10 m)2 =11. length and moment of inertia (i) mass. On this scale.0105. the molar mass is M(1H216O)1=118. mA mB µ= mA mB mA + mB Relative atomic masses can be used to calculate the reduced mass of a diatomic molecule. The corresponding relative molar mass of water is Mr(1H216O)1=121×1Ar(1H)1+1Ar(16O)1=118. (iii) reduced mass and moment of inertia.529177 Å in theoretical chemistry.44891×110−46 kg m2 The reduced mass and moment of inertia are of importance in discussions of vibrational and rotational properties of molecules. The moment of inertia of a system of two masses. Re1=11.9948 u = 6. and molecular dimensions are usually quoted in appropriate units such as the picometre pm1=110−12 m or the nanometre nm1=110−9 m in spectroscopy.5291771×110−10 m1=10.9948 u 2 = 6.2075 Å1=1120.9948.24 Chapter 1 Numbers. Ar(12C)1=112 and Ar(16O)1=115. The bond length of the oxygen molecule is Re1=11.

14) (see Example 3.76 nm1=15. By equation (1.99792 × 108 m s −1 5.1.997921×11081m1s−11≈131×11081m1s−1. where c1=12.022141×11023 mol−1) =196.4) and the SI unit of electric potential V1=1J C−1 (Table 1. respectively. and energy The wavelength λ and frequency ν of electromagnetic radiation are related to the speed of light by c1=1λν (1.15) In a spectroscopic observation of the transition between two states of an atom or molecule. Different spectroscopic techniques are used to study the properties of atoms and molecules in different regions of the electromagnetic spectrum.626081×110−34 J s: E1=1hν (1.8 kJ mol−1 Very often. this corresponds to frequency ν= c λ = 2. but a variety of units is used for energy. For the sodium example. the wavelength of one of the pair of yellow D lines in the electronic spectrum of the sodium atom is λ1=1589.2): eV1=11.8975 × 10−7 m = 5.16) where NA is Avogadro’s constant.89761×110−7 m.602181×110−19 J)1×1( 6. or as molar energies in units of kJ mol−1. The value of eV is the product of the protonic charge e (see Table 1.14).3681×110−19 J Energies are often quoted in units of the electron volt.602181×110−19 J. The corresponding molar energy is eV1×1NA1=1(1.15).7). The values of frequency and wavelength are usually recorded in multiples of the SI units of hertz (Hz1=1s−1) and metre (m). eV. the frequency of the radiation emitted or absorbed is given by hν1= |∆E|. where ∆E1=1E21−1E1 is the energy of transition between states with energies E1 and E2.3681×110−19 J1=12. ∆E1=13.08331×11014 s−1)1=13. frequency.17) .0833 × 1014 s−1 and by equation (1.486 kJ mol−1 (1. The energy of a photon is related to the frequency of its associated wave via Planck’s constant h1=16. For example.8 Units 25 EXAMPLES 1.102 eV1≡1202. and different units are used to report the characteristics of the radiation in the different regions. the characteristics of the radiation are given in terms of the wavenumber ν= 1 λ = ν c = ∆E hc (1.626081×110−34 J s)1×1(5. the corresponding energy of transition is ∆E1=1hν1=1(6.19 Molecular properties: wavelength.

if a length A is two orders of magnitude larger than length B then it is about 1021=1100 times larger.13851×110−261≈110−26 and R1=11.14) to (1. These quantities are related by equations (1.602181×110−19 J1≡196. as an aid to avoiding errors.5 cm−1 0 Exercises 106 Approximate calculations Powers of 10 are often used as a description of order of magnitude. and wavenumber 9 in units of cm−1. For the sodium example. the characteristics of the radiation observed in spectroscopy can be reported in terms of frequency ν in Hz (s−1).486 kJ mol−11≡18065. wavelength λ in (multiples of) m.478 × 10−3 p 105 Two estimates of the answer are (a) V ≈ 10−1 × 10 × 102 10 5 = 10−3 (b) V ≈ 10−1 × 8 × 300 10 5 = 2. and units This has dimensions of inverse length and is normally reported in units of the reciprocal centimetre.8976 × 10−5 cm = 16956 cm −1 The second line of the sodium doublet lies at 16973 cm−1.17). EXAMPLE 1. to calculate the order of magnitude of the answer before embarking on the full detailed calculation. The simplest way of performing such an ‘order of magnitude calculation’ is to convert all physical quantities to base SI units and to approximate the magnitude of each by an appropriate power of ten.12811×110−101≈110−10. In summary.31447 × 298 = = 2. for example. possibly multiplied by an integer.1 (ignoring units). µ =11. energy ∆E in units of eV.20 Order of magnitude calculations (i) For the calculation of volume in Example 1. Such calculations are often surprisingly accurate.1 × 8. In some calculations that involve a wide range of orders of magnitude it can be helpful. variables. cm−1.4 × 10−3 (ii) For the calculation of the moment of inertia of CO in Example 1.89761×110−5 cm and ν= 1 5. molar energy in units of kJ mol−1. 0 Exercise 107 .44891×110−46). V= nRT 0. and an order of magnitude estimate of the moment of inertia is I1=1µR21≈1(10−26)1×1(10−10)21=110−46 (accurate value 1. λ1=15. and the ﬁne structure splitting due to spin–orbit coupling in the atom is 17 cm−1.18 (ignoring units). Conversion factors for energy are 1 eV1=11.26 Chapter 1 Numbers.

19) Atomic unit me e A = h22π a0 = 4πε0A22mee2 2 Eh = mee4216π2ε 0 A2 Value in SI units 9. the dimensionless quantity r2a0 is replaced by r.72114 × 101 V 8. For example. e.1. for example. If this is done to equation (1. and ε0.10938 × 10−31 kg 1.35052 × 105 T 7.64878 × 10−41 F m2 1.05457 × 10−34 J s 5. mass.4. h. The atomic units of length and energy have been given names: the unit of length. the charge on the proton.41888 × 10−17 s 6. Planck’s constant.60218 × 10−19 C 1. is called the hartree.35974 × 10−18 J 2. and is the most probable distance of the electron from the nucleus in the ground state of the hydrogen atom (the radius of the ground-state orbit in the ‘old quantum theory’ of Bohr).89104 × 10−29 J T−2 A2Eh eEh2A Eh2e ea0 Eh2ea0 4πε0a3 0 eA2me A2ea 2 0 e2a 2 2me 0 . The unit of energy. is called the bohr. Some of the atomic units are listed in Table 1.8 Units 27 Atomic units The equations of motion in quantum mechanics are complicated by the presence of the physical quantities me. for a distance r.4 Atomic units Physical quantity mass charge angular momentum length energy time electric current electric potential electric dipole moment electric ﬁeld strength electric polarizability magnetic dipole moment magnetic ﬂux density magnetizability (1. Eh.47835 × 10−30 C m 5. and then to delete the unit from the expression.1).18) The four experimentally determined quantities can be used as base units for the construction of atomic units for all physical quantities whose dimensions involve length. a0.18) the resulting dimensionless equation is 1 1 − ∇ 2ψ − ψ = Eψ 2 r Table 1. the Schrödinger equation for the motion of the electron about the stationary nucleus in the hydrogen atom is − h2 8π 2 me ∇ 2ψ − e2 4 π ε0 r ψ = Eψ (1. The convention is to express each physical quantity in an expression in atomic units.85480 × 10−23 J T−1 2.29177 × 10−11 m 4. and is equal to twice the ionization energy of the hydrogen atom. the rest mass of the electron. the permittivity of a vacuum. and electric current (the ﬁrst four entries in Table 1. time.14221 × 1011 V m−1 1.62362 × 10−3 A 2. Atomic units are widely used in quantum chemistry.

35975 × 10−18 J = Eh . variables.29177 10−12 × 10−11 F m −1 m e2 =1(4. and units In this form the equation is often referred to as the ‘Schrödinger equation in atomic units’.18) is an energy. EXAMPLE 1. the potential energy of interaction of charges q1 and q2 separated by distance r is V= q1q2 4 πε0 r where ε01=18. use a01=14πε0A22mee2: e2 4 πε 2 e2 m e2 me e4 0 = = × e ÷ = 4 πε0 a0 4 πε0 m e2 4 πε0 4 πε 2 16 π 2 ε 2 e 0 0 e2 = Eh 2 (ii) To calculate the value of Eh in SI units.4. Therefore e2 4 πε0 a0 0 Exercise 108 = 4.854191×110−12 F m−1 is the permittivity of a vacuum. for all positive integers n.602182 × × = 4 πε0 a0 4 × 3.14159 × 8. Then 10−19 × 10−19 C2 1.359751×110−3)1×1(10−15)1×1(C21F−1) From the deﬁnitions of the coulomb C and farad F in Table 1.85419 × 5. Z Z e2 V = 1 2 R 4 πε a 0 0 (i) To show that the unit is the hartree unit Eh in Table 1.19) gives the numbers E1=1−122n2.2. and these numbers are then interpreted as the energies E1=1−122n2 Eh. use the values of e and a0 given in Table 1. The results of computations are then numbers that must be reinterpreted as physical quantities. F1=1C21J −1 so that C21F −11=1J. the quantity E in equation (1.28 Chapter 1 Numbers. For example. Solution of equation (1.4.21 The atomic unit of energy By Coulomb’s law. For charges q11=1Z1e and q21=1Z2e separated by distance r1=1Ra0.

8 25 32 Find the repeating sequence of digits in the nonterminating decimal fraction representation of: 38.1. 27. 12131=10. 2! 32.07692 3076923 48. + 12 16 18.9 Exercises Section 1. 16. (a2)−3 57. 3 4 ÷ 4 5 4.3 Factorize in prime numbers: 22. 36. a52a−4 60. 41÷121+171×12 76. 2 = 1. 41÷121×171−12 75. 46. 45. a32a2 59. − 4 7 1 3 14. 7! 3! 2! 6! 3! 33. 10! 5! 3!2! 10! 7 !3! Evaluate by cancellation: 33. 8223 53. + 4 8 11 3 13. 6. π1=13. 2 5 ÷ 3 3 5. 81×121÷141÷12 77. (71+13)1×141−15 74. 251+1144122 . 7. 6 23. 3. 27−423 69. (521+1122)122 55.9 Exercises 29 1. a3a−3 56. 32325 54.6 Simplify if possible: 50. 1 1 8. 9122 Evaluate: 66. − × − 3 4 12. 31−1(−4) 3. 21+131×110−41+151×110−6 40. 10−2 3 1 5 41. 4. (a 1+1b ) 62. (−3)1×1(−4) 3 5 9. Section 1. a2a3 51.2 Calculate and express each result in its simplest form: 1.414213 562373 49. 810 768 5120 Simplify by factorization and cancellation: 26. 1 2 + 14 21 2 5 × 3 6 1 1 ÷ 3 9 7. 9 11 21 17 Use the rules of rounding to give each of the following to 8. 71+131×141−15 73. a3a−4 58. 31+142 52. 31+141×152 68. (a3)4 67. 34. Section 1. 42. (71−13)1×12 72. 5. (−8)1÷1(−4) 1 2 − 18 27 2 3 17.4 Express as decimal fractions: 37. 2 5 − 9 6 3 15. 21×110−3 39. Section 1. 31×1(−4) 10. 80 24. 29. a122a123 63. (a2)322 64. 71−1(31×12) 71. 1 1 1 1 44. (−3) −1(−4) 6. 2 and 1 signiﬁcant ﬁgures: 47. 71−131×12 70. 31+1(−4) 2.141592 653589 43. 28. (12a2 )−4 3 3 123 61. Find the value of: 30. 21. × 2 4 19. 81÷1(− 4) 11. (a3b6)223 65. 2 × 4 2 4 ÷ 15 5 20. 35. 256 3 18 21 49 63 294 25.

Calculate (ii) the reduced mass of the molecule and (iii) its moment of inertia. in terms of the derived units in Table 1.9688 and Ar(1H)1=11. ML2 T −2 80. The root mean square speed of the particles of an ideal gas at temperature T is c1=1(3RT2M)122. now called the Bohr (radius).8 For each of the following dimensions give its SI unit in terms of base units (column 5 of Table 1. (ii) bar. m s−1. 108.45359237 kg. ML I T 89. 95. (iii) A snail travels at speed 1. identify a physical quantity for each: 81.7 Find the sum and product of the pairs of complex numbers: 79. 106. (i) What is the unit of velocity in a system in which the unit of length is the inch (in1= 2. Assuming that the acceleration of gravity is g1=19. The ‘weight’ of the mass in the presence of gravity is called the pound-force. ML−3 82. 94.30 Chapter 1 Numbers. In the kinetic theory of gases. (i) express 1 lbf in SI units. the pressure that is denoted (in some parts of the world) by psi1=11 lbf in−2.854191×110−12 F m−1. kN dm 103. 107. Express this in units yd h−1. express a speed of 60 miles per hour in (i) m s−1. and km h−1. (ii) km h−1. and (ii) the value of a0 in Table 1. ML T N Z 91. mmol dm 104. z21=1−51−14i 78. . the electron moves round the nucleus in a circular orbit of radius a01=14πε0A22mee2.80665 m s−2. where M is the molar mass. e and A given in Table 1.4 to conﬁrm (i) that a0 is a length. z21=141−17i Section 1. NL−3 83.27451×110−10 m and the relative atomic masses are Ar(35Cl)1=134.15 K (M1=128. dm−3 97.31447 J K−1 mol−1 and M is the molar mass. 105. (i) Perform an order-of-magnitude calculation of b for N2 at 298. z11=111−16i. z11=131+15i. Given relative atomic masses Ar(14N) = 14. cm ms−2 98. (b) Å and (c) a0. IT 88. (iii) calculate the work done (in SI units) in moving a body of mass 200 lb through distance 5 yd against the force of gravity.0078. The non-SI unit of mass called the (international avoirdupois) pound has value 1 lb1=10. where R1=18. Conﬁrm that c has dimensions of velocity. ML T 87.01 g mol−1). calculate (i) the relative molar mass of ammonia.0031 and Ar(1H)1=11. the mean speed of the particles of gas at temperature T is b1=1(8RT2π M )122. (ii) Calculate b to 3 signiﬁcant ﬁgures. (ii) wavelength. dg mm−1 ns−2 −3 101. Given ε01=18.2. Calculate the corresponding (i) frequency. The vapour pressure of water at 20°C is recorded as p(H2O. (iii) atm. use the units and values of me.5 Torr. mg pm µs−2 100. 20°C)1=117. and units Section 1. (ii) the molecular mass and (iii) the molar mass. lbf. The bond length of HCl is Re1=11.1) and. Express this in terms of (i) the base SI unit of pressure. g dm−3 99. Mr(14N1H3). (i) Express the bond length in (a) pm.2 in min−1. The origin of the fundamental aborption band in the vibration–rotation spectrum of 1 35 H Cl lies at wavenumber 91=12886 cm−1.4. L3 −1 −2 2 −1 −3 2 −2 −1 2 −2 −1 −1 86. MLT −1 84. ML T N 90. 92. where possible.541×110−2 m) and the unit of time is the hour (h)? (ii) Express this in terms of base SI units. 93. MLT −2 85. In the Bohr model of the ground state of the hydrogen atom. GHz µm 102. (ii) express. Express in base SI units 96. in SI units. Given that 1 mile (mi) is 1760 yd and 1 hour (h) is 60 min.0078. and (iii) energy in units of eV and kJ mol−1. variables.

in equation y1=12x 21−13x1+11 the expression on the right of the equal sign deﬁnes the function f. In the present chapter we are concerned with functions of only one variable. The word ‘function’ was ﬁrst used in this context by the German mathematician Gottfried Wilhelm Leibniz (1646–1716). Let the variable y be a function of the variable x. a dependent variable is said to be a function of the variable or variables on which it depends. x1=11. and x1=10 are f(2)1=121×1221−131×121+111=13 f(1)1=121×1121−131×111+111=10 f(0)1=121×1021−131×101+111=11 0 Exercises 1–3 The concept of function is more general than this however. temperature T.1 In this example. because the variable x can be replaced by another variable.1 Concepts When the equation of state (1. and amount of substance n (the independent variables). or by a more complicated quantity such as a differential operator or a matrix. f(x)1=12x 21−13x1+11 (2. V is a function of the three variables p.1) of the ideal gas is written in the form V= nRT p it is implied that the value of the volume V (the dependent variable) is determined by the values of the pressure p.1) whose value for any given value of x is to be assigned to the variable y (read f(x) as ‘f of x’). For example. EXAMPLE 2. The function f is the rule for calculating y from x.2) when x1=12. by a function.2) (2. and n. T.1 The values of the function (2. 1 . In general. A function has a numerical value when numerical values are assigned to the variables. the case of more than one independent variable is discussed in Chapter 9.2 Algebraic functions 2.

6 d EXAMPLE 2.5 By the equation of state of the ideal gas.2 Replace the variable x in (2.2) by the function h1+12.1 f(105 Pa.4781×110−3 m3 2 2. n)1=1nRT2p and by the calculation performed in Example 1. f(a)1=12a 21−13a1+11 0 Exercise 4 EXAMPLE 2. Consider the function y1=1f(x)1=1x 21−12x1−13 (2.4 Replace the variable x in (2.2 Graphical representation of functions A real function may be visualized either by tabulation or graphically by plotting. 298 K.3 Replace the variable x in (2.1 mol)1=12.2) by the variable a. T. temperature.2) by the differential operator (see dx Chapter 4). d d d d2 d f = 2 − 3 + 1 = 2 2 − 3 + 1 dx dx dx dx dx is a new differential operator. EXAMPLE 2. the volume is a function of pressure. V1=1f( p. 0 Exercises 5.3) . f(h1+12)1=12(h1+12)21−13(h1+12)1+11 =12(h21+14h1+14)1−13(h1+12)1+11 =12h21+18h1+181−13h1−161+11 =12h21+15h1+131 =1g(h) where g(x)1=12x21+15x1+13 is a new function of x that is related to f(x) by g(x)1=1f(x1+12). 0.32 Chapter 2 Algebraic functions EXAMPLE 2. and amount of substance.

. . . .. . . . .1. . He developed the relation between algebra and geometry in his Géométrie. . . .. . .. . . .2. . . y) in the table may be regarded as deﬁning the position of a point in a plane.. . . . . . ‘the root (x). . . . . . A table can be drawn. . . .. . . . . .. ... . .. . .. . . .. His work marks the beginning of modern algebra. the square.. and the ﬁrst use of the words ‘real’ and ‘imaginary’ in the context of complex numbers..1. . . . . . . −2 −1 0 1 2 3 4 5 5 0 −3 −4 −3 0 5 12 Figure 2. . . . . or Renatus Cartesius. . .. In addition.. . .. .. .. . .. . . . . .. . . . . . .. . .. . .. . . made important contributions to the theory of numbers and to the calculus. . are merely magnitudes in continuous proportion’. . . . .. 8 The cartesian coordinate system2 The position of a point in a plane is speciﬁed uniquely by its coordinates in a given coordinate system. . Pierre de Fermat (1601–1665). and letters at the end (x. . . . .. .. . . .. . . ... and can be plotted in a graph as in Figure 2. . . . .. . . . . . Descartes discarded this restriction.2. The Géométrie contains a formulation of the fundamental theorem of algebra. . . . .. . . French philosopher and mathematician. . .. . . . but his work was not published until 1679.. . .... . He attributed his search for a universal mathematics to a mystical experience in 1619 in which ‘full of enthusiasm. . . . . . .. . . . . . y Table 2. such as Table 2.. . . .. I discovered the foundation of a wonderful science’. the quantities x. . . . . . .1 The function given by equation (2. . . . . . . . . . . .. .. and volume. . . after his death... . area. .. . .. ... . . . . and x 3 were always associated with the geometric concepts of line. . Before Descartes.. . . . 1637. . . . . . . . .. . . . . . . . . . x 2. . . . polynomials and other algebraic functions are discussed in the following sections. . . . giving values of y corresponding to a set of values of x. . .. z) for unknowns and variables.. the cube. . quadrature is the act of squaring. . . . b.. .. . . It is an example of a general class of functions called polynomials. . . ... . . . . ﬁnding a square whose area is equal to that of a given ﬁgure).. .. .. .. . Coordinate geometry was also developed by Fermat at about the same time as Descartes. .. . . . . . . . . . . .. .. .. Descartes introduced the convention of using letters at the beginning of the alphabet (a. .. .. . .. . .. . . . ... ... .2 Graphical representation of functions 33 For each value of x there exists a value of y. . . . . . . .. . . . . . . . . . .. . . . lawyer at the provincial parliament of Toulouse.. René Descartes (1596–1650). . . . 0 Exercises 7. .. that is. The most generally useful system is the cartesian (rectangular) coordinate system shown in Figure 2. . . . y... . . . published as an appendix to the Discours de la méthode pour bien conduire sa raison et chercher la vérité dans les sciences (Discourse on the method of good reasoning and seeking truth in the sciences).. . .. . . . . . . . . .1 . . . etc.. . ... .. each pair of numbers (x. . . . .. . .. . . . . =) for constants and parameters. .. .3) is called a quadratic function because the highest power of x is the square (in plane geometry. . .. . . . . . ... .. .. . .. . . . .. 2 x 5 • 4 •3 • 2 • 1 • − 4 0 1 −3 12 • 2 1 8 6 4 2 1 2 3 •− − − • • x y .

suitable scales are marked on the coordinate axes.. and two perpendicular axes (directed lines). . .. The origin has coordinates (0. y) lies at perpendicular distance |x| from the y-axis and |y| from the x-axis.. . . ......... .. .x( •. .. ...... in the expression 3xy1+16x 2 each term of the sum can be written as the product of the common factor (3x) and another term: 3xy1+16x 21=1(3x)1×1( y)1+1(3x)1×1(2x) Therefore 3xy1+16x 21=1(3x)1×1( y1+12x)1=13x( y1+12x) and the algebraic expression has been written as the product of the two factors (3x) and (y1+12x). The curve is the graphical representation of the function...3 for factorization of numbers). . .. ... . O n igiro si x a - y . . ... . ..... A point with coordinates (x.. .. . .. . For example.. .. ... .. Figure 2. ... . . . ....... .2 the origin is labelled O..... .. .... ena lp - yx si x a - . . .. ..3 Factorization and simpliﬁcation of expressions The structure of an algebraic expression can often be simpliﬁed and clariﬁed by the process of factorization (see Section 1.... . and each pair of numbers in a table such as Table 2. If the function is known to vary smoothly between the plotted points (as it is in this example) then the points may be joined by a smooth curve. .. . . . where x is the x-coordinate or abscissa and y is the y-coordinate or ordinate. ... y . . . .... ... . called the coordinate axes.. which intersect at the origin.. . In an actual example...2 A frame of reference is deﬁned in the plane..... and the plane is called the xy-plane.. x ) y .. y). .. ..... . . .... . . .. ... it lies above the x-axis if y1>10 and below if y1<10... .. .34 Chapter 2 Algebraic functions . .. . consisting of a ﬁxed point called the origin of the coordinate system. . the coordinate axes are the x and y axes. x. ... The position of a point in the plane is speciﬁed by the ordered pair (x.. . . .... y x . ... .. . . ..... .. ....1 is plotted as a point on the graph... . In Figure 2.. . .. . 2. . It lies to the right of the (vertical) y-axis if x1>10 and to the left if x1<10. . ....... . . .. . ... . ... 0).... ..

to ﬁnd numbers a and b such that x21−15x1−161=1(x1+1a)(x1+1b) Expansion of the product gives (x1+1a)(x1+1b)1=1x(x1+1b)1+1a(x1+1b)1=1x 21+1bx1+1ax1+1ab and.6 Factorize: (i) 2xy 21−14x 2y1+16xy The expression 2xy is a common factor. so that b1=1−a and ab1=1−a 21=1−9. that is. Therefore x 21−15x1−161=1(x1−16)(x1+11) (iii) x 21−19 Let x 21−191=1(x1+1a)(x1+1b)1=1x 21+1(a1+1b)x1+1ab. EXAMPLE 2. Therefore a = 9 = ±3 and the factorization is x 21−191=1(x1+13)(x1−13) This is an example of the general form x 21−1a 21=1(x1+1a)(x1−1a). x 21−15x1−161=1x 21+1(a1+1b)x1+1ab For this equation to be true for all values of x it is necessary that the coefﬁcient of each power of x be the same on both sides of the equal sign: a1+1b1=1−5 and ab1=1−6.2. Therefore 2xy 21−14x 2y1+16xy1=1(2xy)1×1(y)1−1(2xy)1×1(2x)1+1(2xy)1×1(3) =12xy( y1−12x1+13) (ii) x 21−15x1−16 The aim is to express the quadratic function as the product of two linear functions. followed by factorization gives y 21−15y1+141=1( y1−11)( y1−14) . (iv) x41−15x 21+14 The quartic in x is a quadratic in disguise.3 Factorization and simpliﬁcation of expressions 35 The inverse operation of factorization is usually called expansion or multiplying out. In this case there is no term linear in x: a1+1b1=10. The two numbers whose sum is −5 and whose product is 6 are a1=1−6 and b1=11. Replacement of x2 by y. therefore.

..... ..... ..... . .. .3 The ﬁrst two equations of (2. ..... ..... . .......... . ....... .. . .. . . . .. . . ... . ..... . ..... .. . . ...... ... .. . ... . . .............. . . . .. . a131b1=1±c represents the pair of equations a1−1b1=1+c and a1+1b1=1−c... .. .... . ...... .. Propositions 4 and 7 are the geometric equivalents of the ﬁrst two equations (2. . ... . ... ..... .. ... .. . .. .. ........ . . ........ ...... ...... . .... ....... .. .... . . ................... ...... meaning ‘plus or minus’: (a1±1b)21=1a 21±12ab1+1b2 (2. . 3 ba xb b (a1−1b) 1=1a 1−12ab1+1b 2 2 2 xa x (a1+1b)21=1a 21+12ab1+1b2 ... . ... ... . ......... .. .. .. . ........ . ... . .. . .. ......... ... .. ...... ..... . . .. ... . ......... ... .. ...5) for the squares of (a1+1b) and (a1−1b)..... .... .. .......... . ... .......... . ..... .. . .. Sometimes the symbol 3 is used in a similar way. . ... . ....... . ...... ... . .. . . .... ...6) in which either the upper symbol is used on both sides of the equation or the lower symbol is used on both sides.. . .... .. . . ....... ........ .... . . . .... .. .. . ...... ‘The Elements’. ....5) (a1+1b)(a1−1b)1=1a 21−1b2 Figure 2.. ... ..... ..... ..... . as illustrated in Figure 2..... .. ... For example.. . .. ....... ...... ... ....... . ...... .3 Other important general forms are a 2x x ...... . . ..... .. .6..... ...... .......... . ..... . .............. .......... . .... . .. .. .... . ...... .... .. . ...... .. .. .. . ..... . . ..... ... ... ....... ...... . ... ... .. .. . . ...... ......... .... . ..... ..... . ........... . .. . .... Factorization can be used to simplify algebraic fractions. . ............ . ... ...... . ............ .... . ...... . .......... .. .. .... .... . .. ... .. .... . ...... .. .... .... ...... ........ ..36 Chapter 2 Algebraic functions Therefore x41−15x 21+141=1(x 21−11)(x 21−14) Both the quadratic factors have the form x 21−1a 21=1(x1+1a)(x1−1a) discussed in case (iii) above: x 21−111=1(x1+11)(x1−11) Therefore......4) square of side (a1+1b) square of side |a1−1b| difference of squares (2..... .. .... . . .. . in xy + 2 x 2 4 x + 6 xy Euclid... ... ... ..... Book II. ..... .. . . . . . . ... ... .. . . .... .. .. . . . . ... .. . ..... ... .. ..... . . ...5) can be combined by using the symbol ±... . ... . .. . .. ..... . .. ........... for example..... x41−15x 21+141=1(x1+11)(x1−11)(x1+12)(x1−12) 0 Exercises 9–16 and x 21−141=1(x1+12)(x1−12) The expansion (x1+1a)(x1+1b)1=1x 21+1(a1+1b)x1+1ab (2............ .. . . . . .. ........ ..... . ... .. . .. ........ .. ... .... . .. ....... . . .. . . .......... . ....... has geometric interpretation as the area of a rectangle of sides (x1+1a) and (x1+1b). ... ... . ...... used in Examples 2.... . .. . .... ... . ... . ..... . ...... .......... ... .. ................3.. ....... . .... ... .......... ...... ...... . . . .......... . . . . .... .. ............... . . .. .. .. .. .. . ... . .. .. ..

express x in terms of y.4 Inverse functions 37 both numerator and denominator have the common factor x. ﬁnd x1=1f −1(y). cx + d . and can be divided by this factor (when x1≠10) without changing the value of the fraction: xy + 2 x 2 x ( y + 2x) y + 2x = = 4 x + 6 xy 2 x ( 2 + 3 y ) 2( 2 + 3 y ) EXAMPLE 2.4 Inverse functions Given some function f and the equation y1=1f(x). To ﬁnd x in terms of y.8 If y1=1f(x)1=12x1+13.9 If y = ax + b . that is.7 Simpliﬁcation of fractions (1) 4 x 2 × (2 x ) 2 x = = 2y 2 × ( y) y 1+ 2 y 3+ 6y 3 (1 + 2 y ) = 2 = 9 + 18 x 3 (1 + 2 x ) 3(1 + 2 x ) a 2 − b2 a + 2 ab + b 2 2 (2) (3) = ( a + b) ( a − b) ( a + b) ( a + b) = ( a − b) ( a + b) 0 Exercises 17–22 2. it is usually possible to deﬁne. y is a single-valued function of x. a function g such that x1=1g(y). at least for some values of x and y. 0 Exercises 23–25 EXAMPLE 2. In this example.2. Similarly. for each value of x there exists just one value of y. (i) subtract 3 from both sides of the equation: y1=12x1+131→1y1−131 =12x (ii) divide both sides by 2: → y−3 =x 2 (2. This new function is the inverse function of f and is denoted by the symbol f −1 (not to be confused with the reciprocal 12f ): if y1=1f(x) then x1=1f −1( y) EXAMPLE 2.7) Therefore x1=1( y1−13)221=1f −1( y). x is a single-valued function of y.

. ..... ... . ... Graph (b) also shows the double-valued nature of the inverse function.. ..... . ........ ..... . ........ ..... ..... . .. ..... ..... . . .. . ... ... 1 − y √ ± = ) y( 1 − f = x ) b ( • • x− y y 1 • o .. . .. ... .. . ... .... ...... . ...... .... .. . . .... ... . To solve for x... ....... . .. ...... .. ...... ..... or by rotation around the line x1=1y. .. . . .... .. . ... .. . . ... .. ...4 shows how the graphs of the function and its inverse are related... ...... .... . .... ... ... .... ......... . . .... ........ . ... ...... 0 Exercises 26–29 We note that step (vi) is not valid if (cy1−1a)1=10 because division by zero has no meaning.. .. . . .... ........ .. ......... ........ .. .......... . Figure 2.. . .... .. ....... .. .. ..... ... . ......... .. ... . .. . .. .. ..... . .. . . (iii) subtract (ax1+1b) from both sides: We have (iv) collect the terms in x and x : (ii) expand the l...... . ..... .. .. .... . . .... ... . graph (b) is obtained from (a) by interchanging the x and y axes.. ... .. . ... . .. that is... . ..... .. ...... .. . .. . ......s. . ..... .. . ... ..... but x is a double-valued function of y (except for y1=11)... ............. .... .. ..10 If y1=1f(x)1=1x 21+11. . 1 • .. ... . ... . ..... .... . .. . . ..... ... ..... ...... .. ...38 y is a single-valued function of x. • y ........... ... . ..... . .. .. ........ .... . .. . .. .......... .... . . .... ... . .. . . . ...... .. .... . ... . . . ... ... ......... .............. .. .. ... . . . . .. .. .. . ...... . .... .. ... .. .... (cy1−1a)x1=1−(dy1−1b) cxy1+1dy1=1ax1+1b cxy1+1dy1−1ax1−1b1=10 (cx1+1d)y1=1ax1+1b (cy1−1a)x1+1(dy1−1b)1=10 x=− ( dy − b) = f −1 ( y ) ( cy − a ) ... . ..... . . .. .... ... .............. ..... ... .. ... ........ . . .. . .. . . . y=x x x (vi) divide both sides by (cy1−1a): Chapter 2 Algebraic functions EXAMPLE 2. .. . ... . ... . .. .. ...... ... .... . .... .... . ..... . . ... ..... ... ... . . .. x 21=1y1−11.. ... ....... ... .. . .. .... . .... .. x = ± y − 1 = f −1 ( y ) 1 0 Figure 2. .... .: (v) subtract (dy1−1b) from both sides: (i) multiply both sides of the equation by (cx1+1d): y1=1x 21+11.. Such complications can normally be ignored... . .. .. . . .... . . . . ... . ....... . .. .. ... ...... ... ........ . ... This example demonstrates the type of algebraic manipulation routinely used in the solution of real problems.. . .... .. ..... . .... .. .. . y ... ...... ... ...... .............. .... . express x in terms of y.. . ....... . . ........ . ....... . .... ..... . .. .... ...4 1 + 2 x = ) x( f = y ) a( x x o y=x . . .. . .... . ..... .. .. .... .... . .. .. ..... . .. . . ... .. ..... .. . .. ..... .... . ..... . . .. ...h.. ....... ....... . . .. . . for each real value of y1>11 there exist two real values of x (if y1<11 then x is complex).. ........ .. ... ......... ... .... .... .. . ... .. .... . .. .... . .. .........

.. . .... . ... .... ... .. ... .. . .. ... .. ... ..... ........ .... . . ..... ................. ... ... . .... . ... . ... .. ..... ......... . . .... ... .. ..... .... .... .. .... .. . .. ....... .............. . .. ..... .. ... .... .. ...... .. .... ... . ... −∞1<1x1<1+∞... .. In Example 2......... .. ..... . ...... ..... .... . .. As in Example 2. ....... ... ... . .... .......... . . ... .. . .. .. ... .... ... .. ..... ... . .............. ........ . ....... ......... . . ....... .. . . . ....... ... ... ....... .. ....... . .. . .. .. .. ... ... ... . .. . .. .. . .. ... ... .. .. .. .. . . ....... ... . . .. ..... . .... . .... ... ..... . .. .... . ....... .......... .. . ........ .... . ...... ......... . . . .. . .... . ... . . ....... ....... .. .. ........... . . .... . . ......... ......... ... .......11 the functional dependence of y on x is given explicitly by the right side of the equation... .... ..... ... .. ........4 Inverse functions 39 .... .. ... ... . .. ... . . . ..... .. y)1=10 Figure 2.......... .10... .. ... ... EXAMPLE 2. .. . . ..... ......... .... ....... . . .. .. ... ..... ...... ....... ....... .... ..... ..... On the other hand....... .. .. . ...... ... . .. . . .... . . ..... . ... .. and the equation deﬁnes x as an implicit function of y.... when x and y are real numbers... . . . ...... ..... ..... .... .......... . .. .... .... .. 2− y=x • 51 .. 2 .. ... .. . . .. ..... .... . . .. . .... . .... .. ..... .. ... ..... .. .. ..... .. . . gives either variable as a function of the other... .... .... .. ....... .... ...... ...... ... ... . .... .... .... ... ...... ... ....... . ..... ... ............. In general....... ... . although it can be tabulated and plotted as in Figure 2.. x is only deﬁned for y1≥11...... .. ....... ...... ... f(x....... ...... . ... . .. ...... ........... . ... .. ....... .. .... ... ... .... ... . . .. .... . ...... ... . . . ....5.... .... . ..... that whereas y is deﬁned for all values of x... ......... .. ........ . ...... ......... ..... . ... ......... .. .. . . .. .. . . ...... ..... . ... ..5 ) x( f = y ) a( .. ..... ... . . .. . .. . .. ......... .... . .. .. ... ... ....... . ...... ............ ..... .. ......... .. . .. . ... .... ..... ... ..... . . ........ .... . ..... .. . ..... .. ..... . ...... .. .. ... . y is an explicit function of x... . .. or it may not be convenient to do so. 31 In physical applications it is usually obvious from the context which value is to be chosen. . . . .. ..... .. ... .... .. . an equation of the form ) y( 1 − f = x )b ( 5 1− • y 2 .... .. . .. 2− • x 51 5 1− y=x • ....... ...... ..... .. x In this case the inverse function f −1 exists for all values of x.......... . .... . .... ....... .... ...... .. . . ... y 2... ......... ..... .. . ... ........ .. ..... ... ...... ........... ..... graph (b) has been obtained from graph (a) by rotation around the line x1=1y.... ..... ...... . .......... ...... ...... .... . ........ ...... .. . ..... ... .. . ...... .... .. . ..... .. . ....where f is a function of both x and y.. ...... . ......... ...... . .... .............. .... ...... .. ........ ... . .... . but it cannot be written in simple algebraic form...... .. .......... .. . ... ... . ........... . ............. . .... ...... .... .......11 y1=1f(x)1=1x 51−12x The ﬁnding of the inverse function is not always so straightforward...... ...... In many equations in the physical sciences it is either not possible to express one variable as an explicit function of the others.... . ..... .... ......... .. .. . . . .. . . ... .. .. ......... .... .. ..... ... ....... ... .... .. ...... . .. .. . . .... ... ...... .... . . ... .... . . ............ ... . . . .... .............. . .. .. . .... .. . ........... . .... ..... ....... 0 Exercises 30.. ... . ...... ..... . ....... . ... . . ... .. ....... . . .. .... . . ... . .. . ... ......... .............. It is also seen. .. . ..... .. ....... .. . ...... .......... .. ........ .. ........... ..... . .... .... . ...... .. .. .. ... .. ... ...... .. . x cannot be written as an explicit function of y.......... ........ .. .. ....... . . ... ... ..

0 Exercises 32–35 2. and n is a positive integer. It is possible to write down explicit solutions of a cubic equation. n: ∑ ai x i = ( a0 x0 ) + ( a1x1 ) + ( a2 x 2 ) + i=0 n + ( an x n ) = a01+1a1x1+1a2 x 21+1-1+1an x n (remembering that x01=11 and x11=1x). equation (2.1=1.8) In this case. equation (2.8) as deﬁning V as an implicit function of p. The notation tells us to add together the terms ai x i in which the integer variable i takes in turn the values 0.40 Chapter 2 Algebraic functions EXAMPLE 2.1a1. nR V2 p= nRT n2 a − 2 V − nb V For V. 1. If n1=10 the function is the constant a0.8) can be solved numerically by an iterative method such as the Newton–Raphson method described in Chapter 20.12 The van der Waals equation The equation of state for a ‘slightly imperfect gas’ is n2 a p+ (V − nb) − nRT = 0 V2 (2.5 Polynomials f(x)1=1a01+1a1x1+1a2 x 21+1-1+1an x n (2. and n. but these are complicated and seldom used. For any set of values of the independent variables and of the constants. both T and p are easily expressed as explicit functions of the other variables: T= n2 a 1 p+ (V − nb). it is most convenient to regard equation (2. The polynomial is often written in short-hand notation as f ( x ) = ∑ ai x i i=0 n (2.9) The general polynomial of degree n has the form where the coefﬁcients a0.8) can be rearranged into RT 2 n2 a n3ab V 3 − nb + =0 V− V + p p p which is a cubic equation in V. 2.1an are constants.1=. . T. In this case.10) where the symbol ∑ represents summation.

. ... ... ........ 2 2 ( ..... ... . ..... . .. . ...... ...... .. . ... y. .. 1 . . .... •......... ..... ... .....The graph of the function is a straight line.. . .. .. ....5 Polynomials (2.. . ........ .......... ..4 to involve no new principles.. . .. . . ... y1) and (x2......... . ..... ..... . .... .. ...... . . . .... .. .. .. . .. •........ ....... ... .. ..... . .. ... . . .... ....... ... . . . .... If we take any two points on the line.......... . ...... .. ... ..... ..... ... ... ......... ... . .... .. ......6.... . and is better known in the form (3) (2) (1) ∑ ( − x ) i = ( − x ) 2 + ( − x )3 + ( − x ) 4 = x 2 − x 3 + x 4 ∑ i x i = 0 × x0 + 1 × x1 + 2 × x 2 + 3 × x3 = x + 2 x 2 + 3x3 i= 2 x 2 n −1 x −1 x1 x 3 x5 1 x x 3 x5 = + + + = + + + ∑ 1 2 3 4 x 2 3 4 n= 0 n + 1 i=0 4 3 3 f(x)1=1a01+1a1x y21=1mx21+1c y11=1mx11+1c y1=1mx1+1c Figure 2. . .... . . x y 2... . .. ..... ..... . . ) y. ... .... .. . ... ....... ..... .. . ...... .. . ............. . x Degree n1=11: linear function 0 Exercises 36–39 EXAMPLE 2........... and intercepts the vertical y-axis (when x1=10) at the point y1=1c. . 1 2 .. .......... ... ........ ...... .... ... .. . ......... . ............ ... ..... ... .... ....... ... .. the case of complex coefﬁcients is shown in Section 8. . ... .. . .. This is the simplest type of function. ...... ... .... . .......... .. .. ... ... ........ ..... 1 1 ( ....... ... .. . ... . .. as shown in Figure 2.. ..... . .... .. . ... . . .... .. . .... . ...... .... ....... ....... .. . . ........ It has slope m.... ...... y2).. . . .... ..... ........... ........ ..... x x− x 1 y− y ......... . .12) (2.. m/c− ..... .... •. .. ..... . . .... ... ... then x 2 .... .. ............ ............ .. .... ... ... . .. ... .. ........ ....... . .... ...... ... . 2 .... ... ........ . . x c y y .. .6 1 .. .. ....... .. . .11) 41 . ... .... .. . .. ... ......13 Write out in full: Only real coefﬁcients are discussed here.. ......... . 2 . . . ..... .... .... ..... .. .. . .... .. .. . .... .. .. ... . . . .. .... ..... .. .. .... ... with coordinates (x1. . .. .... ) .... .. . ..... . .. ... . . . ......... .. . ..... •. ..

... . .. ..7. .............. .........1−( . ....... . ... 1 ... ..... − 0 . .. .. −61=1−m1+1c 21=13m1+1c Solution of the pair of simultaneous equations (see Section 2. ..... .... −6).......... .... ..... .... .... . ........ ..... ... .... x . ... . .. . .. Therefore y1=12x1−14 The graph of the line is shown in Figure 2.. . .... ....... . . . . ..... .. .. .... the roots of a polynomial function are those values of the variable for which the value of the function is zero. . the roots are the solutions of the polynomial equation f(x)1=10 (2...... ........... and crosses the x-axis at x1=12..... ... . . ... .... •. •. ..7 0 Exercises 40–43 )2 . ... .. 4 ..... Let the line be y1=1mx1+1c. .... .. 2 . .. . .. ... . . . ....... .. ..... .. ...15) EXAMPLE 2... − ... ...... . . . . • . . .. . ....... .. .. ...... . . ......... .... .... . 2 . . The line crosses the y-axis at y1=1c1=1−4. . .. .... . ..8) gives m1=12 and c1=1−4.. . .. y1)1=1(−1.. which means that the value of y increases twice as fast as that of x..... ... .... y2)1=1(3. ......14) This value of x is called the root of the linear function. . . that is... . . .......... .... .... ...... ... .......... ..... ..... . ... .. ... )6− . ......13) deﬁnes the constant slope.. . 3 . . ... 4 .. ........ 1 . at point (x2.... .... . ............. . . .. . . . ...... .. In general....... .. .. ...... . . . .... ....... − . −6) and (3. The line has slope m1=12....... ....3( •....... .. 2).. . ... .. The line crosses the horizontal x-axis at one point: y1=10 when x=− c m (2...... .. .... ... .. ........ . . . ... . . . Then: at point (x1.....42 Chapter 2 Algebraic functions and m= y2 − y1 x2 − x1 (2... ... . ... . 2 . ... y Figure 2... .. .... . . .... .. .... . . 2).. .. .14 Find the equation of the straight line that passes through the points (−1. ........... .... .

French lawyer.5 Polynomials 43 Degree n1=12: quadratic function f(x)1=1a01+1a1x1+1a2 x 2 The quadratic function is usually written as y1=1ax 21+1bx1+1c (2. cryptoanalyst. distinguishing between constants and variables. Modern algebra became possible with the development of a general abstract notation in the 15th to 17th centuries. Yale Babylonian Collection) of the Old Babylonian Period (c. when f(x)1=10. he made contributions to trigonometry and algebra. −4). For the function in Figure 2.16) (2. and amateur mathematician. These are the roots of the quadratic function. in his In artem analyticem isagoge (Introduction to the analytical art) of 1591.18) this minimum point has coordinates (x. The shape of the curve is that of a parabola. as in Examples 2. 1800–1600 BC) has inscribed on it in the Sumerian cuneiform script the following problem (in modern notation): given that xy1=160 and x1−1y1=17. f(x)1=1x 21−12x1−13 (2.2. and is symmetrical about a vertical line that passes through the point of minimum value. The method and prescriptive approach is almost identical to that used by Al-Khwarizmi two and a half millennia later. ﬁnd x and y.17) A typical graph is shown in Figure 2. One important step was taken by François Viète (1540–1603).19) (the symbol ⇒ means ‘implies’) Whilst it is possible to factorize a variety of quadratic functions by trial and error.1. and they are the solutions of the quadratic equation x 21−12x1−131=10 In this example the roots are easily obtained by factorization: x 21−12x1−131=1(x1+11)(x1−13) and the function is zero when either of the linear factors is zero: either x 2 − 2 x − 3 = 0 when or x + 1 = 0 ⇒ x = −1 x − 3= 0 ⇒ x = 3 (2.1 in Section 2. The prescription given for the (positive) solution corresponds to x = (7 2)2 + 60 + (7 2) and y = (7 2)2 + 60 − (7 2) . . y)1=1(1. at the two points x1=1−1 and x1=13.6. The graph crosses the x-axis.2. politician. He is best remembered as the man who. the roots can always be found by formula:4 ax 21+1bx1+1c1=10 when x= − b ± b2 − 4 ac 2a (2. When the constant a is positive. introduced the systematic use of symbols (letters) into the theory of equations.20) 4 A clay tablet (YBC 6967. the function has a single minimum value (turning point).

. .... .... .. . .... . .. . ......... ... ..... .. ..... .. .. ........20).. ... . . ... ... .. . . . b1=16.. .. .... .. . . .... .. .......... ... . .. Its value in Examples 2. . .. .. .. ........ ............. We have a1=11. .... . .......... . . .. .. ... . ..... .... ..... .. .. ..20) is called the discriminant of the quadratic function.... . . ........ .... . ... ... and the roots are The quantity 0 Exercises 44–46 EXAMPLE 2.. . .. .. .... . .... ........ . . x . ... .... . .. .. ..... . . • .. .. . and c1=13 in formula (2. .. y y .... .. .23) (2... . ... . .. . . . . .... .. 44 Chapter 2 Algebraic functions in (2...... ... ...16 is positive. . ... ..... . .. ...... . .. ..... .. . ... .. .. .. . .. .. .. ....... ..... ......... .... x ......... .. . ...... 2a ax 21+1bx1+1c1=1a(x1−1x1)(x1−1x2) ( Figure 2..... . . ....... .. ..... ....... ..8... .. ..21) (2...... .. .. .. . The roots are therefore x= x= −6 ± 36 − 24 1 = −3 ± 3 4 2 +2 ± 4 + 12 = 1 ± 2 = −1 or 3 2 x1 = − b + b2 − 4 ac . ...... . ...... . ... .15 and 2.. . . .. .. but in other examples it can have zero or negative value... .. . x ........ . ... ..8 st o or l a er tn r eﬀid 2 e 0 > ca 4 − 2b 2 x ... . . .... ....... .. .. ... b1=1−2... ... .. .. ..... ....... .. .... .... ..... .. . ..... .. .... .... .... ..... • . ..... .... . .... ..... .. .. and the function has two real roots... x . ......... .. .. .. ........... .. . . .......... .... .... .. . . ..... .22) . . ..... ... .. . . . ..... .. .. .. .... b21−14ac ) x2 = − b − b2 − 4 ac 2a (2.. ...... ... .. .. . ... .. We have a1=12.. ..... ...... .. ... . . ... .... ............... ......... ...... ... ..... .. ......... .. ... .. A graphical explanation of the three possible types of discriminant is shown in Figure 2..... . . .. .... . .. EXAMPLE 2... ... . ... .. . ..... ..... ... ......... .. .. y .. .. .. . .. ...... and c1=1−3 in formula (2. . ..... . ...20)... . .. ... . ... . . .... .. .......... ........... ... . ........ .... . ... . ......15 The roots of the quadratic function f(x)1= x 21−12x1−13. ..... . . .... ..st o or l a r on e 0 < ca 4 − 2b st o or l a er l au q e 2 0 = ca 4 − 2b 2x = 1x • . ... . ......... .. . . ..... .. . ........... 1 .... .. . . .......... . ...... .. .... . .......16 Find the roots of the quadratic function f(x)1=12x 21+16x1+13. .... and the quadratic has factorized form The two roots are and the factorized form of the function is x 21−12x1−131=1(x1+11)(x1−13)... ...... ....

In general for the polynomial of degree n.20) requires the taking of the square root of a negative number. the function behaves like the simpler function ax 2. and the result is not a real number. 50 0 Exercises 51.17 Case The quadratic b2 − 4 ac = 0.2. and can sometimes be replaced by it. equation (2. f ( x) x 2 =a+ b c + → a as x → ± ∞ x x2 This means that for large enough positive and negative values of x. 0 Exercises 47. 2x 21−18x1+181=12(x1−12)2 has zero discriminant and the double root (two equal roots) x1=12.18 Case b2 − 4 ac < 0. 48 When the discriminant is negative.9). Thus. formula (2. 52 For very large (‘large enough’) values of | x| the term in x 2 in the quadratic f(x)1=1ax 21+1bx1+1c is very much larger in magnitude than the other two terms. The quadratic x 21−14x1+113 has a pair of complex roots x1 and x2 given by x= 4 ± −36 = 2 ± 3i 2 The roots are x11=121+13i and x21=121−13i and the factorized form of the quadratic is (x1−1x1)(x1−1x2).5 Polynomials 45 EXAMPLE 2. . In this case the roots of the quadratic are a pair of complex numbers involving the square root of −1 : i = −1 (see Section 8. dividing the function by x 2.2) EXAMPLE 2. 0 Exercises 49.

. ... ... .. . . . . .. ...... ... F. .... .. . . .. .. .. . ... . .. . .. ... . .. . ... ..... . . . .. . .. ..... EXAMPLE 2.... .... ... .. .. a ball rolling forwards and backwards in a parabolic container (a ‘parabolic potential well’) performs simple harmonic motion. ...19 Behaviour of a quadratic for large values of the variable Other examples are the swings of a pendulum... . . .. .. ..1% for | x| 1=1103.. . ... .. ... . . ..... ...20 The classical simple harmonic oscillator For values of | x| larger than about 100 the function f(x)1=1x 21+1x1−11 differs from x 2 by less than 1%.... . . ... . .. . .... .. . ..10 Figure 2.... . .. . . .. . . ... .. .. .. . ... . ... .. . .... . .. .. .. . . . ... . . . .. . .... .. . . ... .. ... .. .. .. . . ... .... . .... . .. ... .. . . .. . . . . ... .. for | x|1=1105.. . . . .. . . . . . . . . ... . .. ... .... . . ..... . . ... .... .... ....... F1=1−kx E= 1 2 1 2 mv + kx 2 2 f ( x) x n = an + an −1 an − 2 + 2 + x x Figure 2.. .. . . . . ... .... . .... . ........ ..... . . . .. ... . .... . .. . .. .. . .... . . . .. . .. . . .. ... . . .. .. ..... . . . . ... The difference decreases like 12x as x increases: 0. ... . ........ . . . . . . . . .. . ... .... ... . .. .. .. . the energy of the system is Chapter 2 Algebraic functions The simple (linear) harmonic oscillator consists of a body moving in a straight line under the inﬂuence of a force x m ss am • . .. .. .... and whose direction is towards this point. . .. . ... .. ..... ... .. ... . . ... ... .. .. . . . . . . ... .. . .. .... . ..... .... The (positive) quantity k is called the force constant and the negative sign ensures that the force acts in the direction opposite to that of the displacement. .... ... .. . .. .. . ... .... .. . .. . ... .. . ... ........ ... .. .. .. .9 + xn a0 → an as x → ± ∞ (2...... .... .. . ... . .. .... ... . .... ........e.. ... . . . .. . .... . . . .. . .. .... .... .. . ... .. . ... . .. .. . ... . .. . ..... .. .. ... . ....... .... .. ...... ....... . . ... .. ....... . and 10−8% for | x| 1=11010. .. . ... .. . . . .... .. ... . . .. . ... . the oscillating electric and magnetic ﬁelds in electromagnetic radiation.. ...... . . . . ... . . .... .. ..... x muirbiliu q e O• EXAMPLE 2. . .. . . .. .. . . . . . f o .... .. .. . . . . .. ... . ...... . ... ... . . . . .... . .... . . ... ... . ... ... .. .... . . ... .. . the point of equilibrium. .. ...... .... .. . .. .. . . .. ...... .. .. . ... . . . . .. .. . ...... . the vibrations of atoms in molecules and solids.. ... . . ... .. .. ... . . . . . . .. .. . . . .... . . . .. .. . .. . . .. . . . for example.. .. . . ... ... . . .. . . .. . ...... ..... .001%.. .. . . .. .. . . ... .. . . .. . .... . .. .. . . .... . . .. . .. . . . .... .. . . . ... . . .. .. ... ..... . . .r.. The simplest kind of vibrational motion is simple harmonic motion and. ..... 0.. . .... . . For a body of mass m. . . . .. ... . . . ... ... . .. . . . ... .. . . . ....... .. . . .. . . .. ...... .... . . ...... .. .... . .. . . . . . .. .. . ... . . . ... .. .. . ... ... . . . .. . .. .. .. .. . .. .. .. ... . ... . ... .... .. . . .. . ....... ... . . . .. ... .... . . ....... ... . . .. . . . . . . . .... . . . . .. . .... . ... . .. . ... . . . Quadratic functions are important in the physical sciences because they are used to model vibrational motions of many kinds.... ...46 whose magnitude is proportional to the displacement x of the body from the ﬁxed point O.. .. .. .... ... . xk− =. . .24) . .. . .. .. . . ....... .. .. .. ... ... ... ... . . . .... .c . .... ... .... .... . . . ... .. .. . . .. . ... .. ... . . . ... ... .. ... . . . ...... . . .. .

Some of the roots may be equal (multiple roots) and some may be complex. In the absence of external forces the total energy is constant (see Section 12. and was ﬁrst proved by the great mathematician Gauss.5 Polynomials 47 where v is the velocity of the body.13. electromagnetic theory.11). x2. . The equation can be written in the form Hψ1=1Eψ where the Hamiltonian operator for harmonic motion H =− 2 d2 2 2 m dx + 1 2 kx 2 d d d2 . that formed the basis for much of 19th century mathematics.1=) always has at least one real root because its graph must cross the x-axis at least 5 Carl Friedrich Gauss (1777–1855). statistics. child prodigy and professor at Göttingen.5 The function is zero when any of the linear factors is zero. since = 2 . and the numbers x1. they are the solutions of the polynomial equation f(x)1=10. A polynomial of odd degree (n1=11.2.21 The simple harmonic oscillator in quantum mechanics The stationary states of the simple harmonic oscillator in quantum mechanics are given by the solutions of the time-independent Schrödinger equation − 2 d 2ψ 2 2 m dx + 1 2 kx ψ = Eψ 2 where ψ1=1ψ (x) is the wave function (see Example 13. including the theory of functions of a complex variable and the differential geometry. 1799. algebra. perturbation theory. Proof of the theorem that every rational integral function in one variable can be resolved into real factors of ﬁrst or second degree. geometry. The expression for the energy is a quadratic function of the variables v and x. he made substantial contributions to every important branch of pure and applied mathematics. that is. the theory of numbers. 1 mv 2 is the kinetic energy and 1 kx 2 is the potential 2 2 energy. EXAMPLE 2. He invented or initiated new branches of mathematics.5 for a more complete discussion of the harmonic oscillator). =. dx dx dx 2 is a quadratic function of x and of The general polynomial A polynomial of degree n can always be factorized as the product of n linear factors f (x)1=1a01+1a1x1+1a2 x 21+1-1+1an x n =1an(x1−1x1)(x1−1x2)1-1(x1−1xn) (2. He gave his ﬁrst proof (there were four) of the fundamental theorem of algebra in his doctoral thesis.15.25) This is called the fundamental theorem of algebra. xn are the n roots of the polynomial.

.. . .... ...... ...... . . . ......... ...... ... . ............ . . ........... x .. ... (a) three real roots: x 31−16x 21+111x1−161=1(x1−11)(x1−12)(x1−13) A polynomial of degree 3 can have all three roots real or it can have one real root and two complex roots....... ....... . ......... ..... .. .... −101=1−c Figure 2.. ..... ...... ........ ..... ......... .. ...... ........ ... ..... ... .... . ... . .... ..... .23 Given that x1−11 is a factor.. . ... . ... .. . . . ........ .. .. ... ..... . •...... . . .... .......... ... .. • 2 . .. . .. 3 ... .. .. .. ........ ......... ....... ......... . ........•... A polynomial of even degree (n1=12... . . .... ... ........ .. . ... .. x 21−12x1+121=1[x1−1(11+1i)][x1−1(11−1i)] =1ax 31+1(b1−1a)x 21+1(c1−1b)x1−1c 161=1c1−1b... for example. .... . .... .. ..... . . .. . ....... . .. . ..... ..... . ..... ... . . . .... . ....... ....... ......... .. . ....... .. ...... ...... x 31−13x 21+14x1−121=1(x1−11)(x1−111−1i)(x1−111+1i) x 31−17x 21+116x1−1101=1(x1−11)(ax 21+1bx1+1c) )b( . .. ... .......... .... .. . ... ....... .... .11 x ..... . ........... ............... . ...... EXAMPLE 2........ x .. .. ............ ... .. . ...... . .... ....... ........ ................................ ... . ........ ...... y .. .. ........ 1 . .. .... .......... .. .. ... ..... In general it has an odd number of real roots.. . . ...... ... .... ) ( a ....... .... .... ........ . ... ........... .. .. −71=1b1−1a... one double: x 31−15x 21+17x1−131=1(x1−11)2(x1−13) and the fully factorized form of the cubic is The roots of the quadratic factor are 11±1i.... . ....... ......... .. or no real roots if the curve does not cross the x-axis... where i = −1.. . ..... .... ..... =) has an even number of real roots.. .) ( c .... ﬁnd the roots of the cubic x 31−17x 21+116x1−110... . . 1 ..... . 6. ......11.. . ...... . .. ....... .. ... ......... .. .. ... . .. ... . ... ..... .... ......... ......... ......... ....... ............... y 11=1a.... ............ ..... ........ . .......... .. ..•. ............ ........................... . ......... . ... . • . . . ........ ..... .... .. . . ........ .. .. ...... ...... . .. . ... ....... .. .... .... ... ........ .. ...... • . .. ....... ...... in Figure 2........ .. ... ........ . . .. .... .. 4. ... (b) three real roots........ ... ... 1 .... ... ...... ... y 48 (c) one real root and two complex roots: x 31−13x 21+14x1−121=1(x1−11)(x 21−12x1+12) If x1−11 is a factor then the cubic function can be written as For this equation to true for all values of x it is necessary that the coefﬁcient of each power of x be the same on both sides of the equal sign: Chapter 2 Algebraic functions EXAMPLE 2. .... . .......22 Factorization of a cubic once. . .. ..... 3 .. ...... .. ..... .. .... . .....

.... 1 (see also Examples 2.. .. . . . .. .. . .... . .. . .. with roots . . ... see Section 4. one for each real root. . at most..... .. .. .... .. c1=110.. .. .. . • . . . . and.12... . .... ..........6 The theorem is used in Section 2. quadratic factors..... 6 This is the statement of the fundamental theorem of algebra given in Gauss’ ﬁrst proof of 1799.. . . .. ... .. .. ... .. ..... .. .. .5 Polynomials 49 Figure 2. . for example x41−1x 31−17x 21+1x1+161=1(x1−11)(x1+11)(x1+12)(x1−13) (ii) 2 real roots and 2 complex roots.. ..22... . . . 6 ± 36 − 40 = 3± i 2 ... .. .. . . ..10) at x1=12 and x1=1823.... . .22 to 2.. . ... . .. . . ... like types (a) and (b).. . if complex numbers are disallowed.. . .. . . . . a polynomial can always be factorized as the product of some linear factors..... ...25 and 2. • .. .. .12 0 Exercises 53–55 EXAMPLE 2. x11=12 and the two complex roots x21=131+1i and x31=131−1i. . ..2. . (i) 4 real roots. • . shows however that. .24 Factorization of a quartic Three cases may be considered... . . x 21−16x1+110 x= . ... .. .. .. .. . . ..... Figure 2. . y Then a1=11.... .. . . . ... . .. . b1=1−6. . .7 for the construction of partial fractions.... . ...... . . .... ... . . for example x41−12x31+13x 21−12x1+121=1(x 21+11)(x 21−12x1+12) =1(x1−1i)(x1+1i)(x1−111−1i)(x1−111+1i) 0 Exercise 56 Examples 2. . . . . .... .... all real. . ... . . with one real root..... . for example x41−12x 31+1x 21+12x1−121=1(x1−11)(x1+11)(x 21−12x1+12) =1(x1−11)(x1+11)(x1−111−1i)(x1−111+1i) (iii) 4 complex roots. .. ... .. . . .. .. . ..24 demonstrate that.. . .... . ... . .26). . . . the function has local maximum and minimum values (or turning points.. . .. . . ... ..... .. . . respectively. The graph of the function. . . .. x / 8 3 • • 2 . . . .... . ... . ... . . .. The cubic is therefore of type (c) in Example 2. . . and the quadratic factor is ..

since division by zero is not permitted. the function (i) in (2. 7 .6 Rational functions Let P(x) and Q(x) be two polynomials P(x)1=1a01+1a1x1+1a2 x 21+1-1+1an x n Q(x)1=1b01+1b1x1+1b2 x 21+1-1+1bm x m (2. 2. the exponential function. made early contributions to group theory. For example. an equation of the kind P(x)y n1+1Q(x)y n−11+1-1+1U(x)y1+1V(x)1=10 (2. More generally. In general. He gave the ﬁrst rigorous proof of the binomial theorem.27) A rational function. and (iii) is not deﬁned at x1=1−2. and did important and innovative work on the theory of elliptic and other higher transcendental functions. The Norwegian mathematician Niels Henrik Abel (1802–1829) proved in his On the algebraic resolution of equations (1824) that there does not exist an algebraic solution of the general quintic equation. The book also contains a description of a method of solving quartic equations due to Ludovico Ferrari (1522–1565). ‘Abel’s short life was ﬁlled with poverty and tragedy’. and can be solved for each value of x.50 Chapter 2 Algebraic functions Algebraic functions Polynomials are the simplest examples of algebraic functions. The general equation of the ﬁfth degree was solved in terms of elliptic functions by Hermite in 1858. Cardano showed that some solutions are complex. =. Q(x). the equation y 31+1(x1+11)y 21+1(x 21+13x1+12)y1+1(x 31+12x 21−1x1−11)1=10 is a cubic equation in y. or of any polynomial equation of degree greater than 4.26) where P(x).28) Examples of rational functions are ( i) 1 x ( ii) x+2 x +1 ( iii) 3x 2 + 2 x − 1 x+2 ( iv ) x −1 3x + 2 x − 1 2 (2. is an algebraic function that has the general form y = f ( x) = 2 P( x ) a0 + a1 x + a2 x + = Q( x ) b + b x + b x 2 + 0 1 2 + an x n + bm x m (2. For example. he died of consumption at the age of 27. or algebraic fraction. these functions are discussed in Chapter 3. Examples are the trigonometric functions.29) is not deﬁned at x1=10. The method of solution (Cardano’s method) was described by Girolamo Cardano (1501–1576) in his Ars magna of 1545.7 Functions that cannot be deﬁned in this way in terms of a ﬁnite number of polynomials are called transcendental functions. the rational function The formula for the general solution of the cubic equation was discovered in Bologna in the early 16th century by Scipio del Ferro and Nicolo Tartaglia. V(x) are polynomials of any (ﬁnite) degree in x. and the logarithmic function.29) In each case the function is deﬁned for all values of x for which the denominator is not zero. deﬁnes the variable y as an algebraic function of x.

....28) is deﬁned for all values of x with the exception of the roots of the polynomial Q(x) in the denominator... ... . .. ... ............2... . ... An improper rational function is reduced to a combination of proper functions by algebraic division.... . An improper fraction can always be reduced to a combination of proper fractions by division..... .. and all rational functions have at least one such point............ . .. . .. .. .... one for each root of Q(x)...29). . . .... . .. .. .......... Division of one polynomial by another The function (2........ ... .. . The point x1=10 is called a point of singularity.. ...... ...... . ... .... . we write x 2 − 6 x + 10 x into x 3 goes x 2 times i subtract x into − 6 x 2 goes − 6 x times subtract x into 10 x goes 10 times subtract remainder x3 − x 2 − 6 x 2 + 16 x − 11 − 6x2 + 6x 10 x − 11 10 x − 10 −1 It follows that 1 x 3 − 7 x 2 + 16 x − 11 = x 2 − 6 x + 10 − x −1 x −1 x − 1 x 3 − 7 x 2 + 16 x − 11 ) x ...... . . ........... .............. .. . .. . ...... .13 x1=10 asymptotically........... ....... EXAMPLE 2. .......... . .... The y-axis is the line x1=10 and is called an asymptote to the curve...... as in examples (ii) and (iii)...... ....... .... ....... we say that y1=112x tends to inﬁnity as x tends to zero. .... .... .. it is called improper. . ..6 Rational functions 51 (2...... ... . . ... ........ . .. Otherwise. . As x approaches zero from the right (x1>10) the value of 12x becomes arbitrarily large. .. ... In ordinary number theory.... . .. The graph of the function y1=112x in Figure 2. the curve approaches the y-axis arbitrarily closely but does not cross it..... .. we say that the curve approaches the line Figure 2... .. ...... .... .... y ..13 demonstrates some typical properties of rational functions...... By adapting the method of ordinary long division... ... for example 522 or 222. .... .......... ... ... ...... Similarly..... . . ..... The graph also shows that as x1→1∞ from either side.. ....28) is called a proper rational function if the degree n of the numerator P(x) is smaller than the degree m of the denominator Q(x).....25 Divide x 31−17x 21+116x1−111 by x1−11. . ..... .. The line y1=10 (the x-axis) is also an asymptote... .. .. ... . . for which Q(x)1=10... an improper fraction is one whose value is greater than or equal to 1.. .... .. ... ..... as in examples (i) and (iv) of (2. ........ y tends to minus inﬁnity as x tends to zero from the negative side. .. . . .. .. . ............

25. can always be decomposed into simpler partial fractions.23).26 Divide x31−17x21+116x1−110 by x1−11. It follows that the cubic can be factorized: x 3 − 7 x 2 + 16 x − 10 ( x − 1)( x 2 − 6 x + 10) = = x 2 − 6 x + 10 x −1 x −1 (see Example 2. The following examples demonstrate three of the important simple cases. and in integration in general.28 Two linear factors in the denominator x+2 5 2 = + ( x − 3)( x + 4) 7( x − 3) 7( x + 4) . In this case the fact that the rational function is not deﬁned at x1=11 either before or after cancellation of the factor (x1−11) has no practical consequences and can be ignored. A proper rational function P(x)2Q(x).52 Chapter 2 Algebraic functions EXAMPLE 2. We will see in Chapters 6 and 11 that the decomposition into partial fractions is an important tool in the solution of some differential equations in the theory of reaction rates. and the fraction has been decomposed into two simpler partial fractions.7 Partial fractions Consider 1 x + 3x + 2 2 = 1 1 1 = − ( x + 1)( x + 2) x + 1 x + 2 (2.30) The quadratic denominator has been expressed as the product of two linear factors. whose denominator can be factorized. EXAMPLE 2. and there is no remainder of the division. In this case it is not necessary to resort to long division: x + 2 ( x + 1) + 1 x + 1 1 1 = = + = 1+ x +1 x +1 x +1 x +1 x +1 0 Exercises 57–60 2. The cubic in this example is (number) 1 larger than that in Example 2.27 Divide x1+12 by x1+11. EXAMPLE 2. All others can be treated in the same way.

The values of A and B can be obtained from this equation of the numerators by the ‘method of equating coefﬁcients’ described in Example 2.29 Three linear factors in the denominator 3x 2 + 4 x − 2 A B C = + + ( x − 1)( x − 2)( x + 3) x − 1 x − 2 x + 3 = Then.23. they are obtained directly by making suitable choices of the variable x. Alternatively. write x+2 A B A( x + 4) + B( x − 3) = + = ( x − 3)( x + 4) x − 3 x + 4 ( x − 3)(x + 4) x It is required therefore that x1+121=1A(x1+14)1+1B(x1−13) for all values of x.2. 3x 21+14x1−121=1A(x1−12)(x1+13)1+1B(x1−11)(x1+13)1+1C(x1−11)(x1−12) so that when when x1=11: x1=12: 51=11−4A 181=15B and A1=1−524 and B1=11825 A( x − 2)( x + 3) + B( x − 1)( x + 3) + C ( x − 1)( x − 2) ( x − 1)( x − 2)( x + 3) when x1=1−3: Therefore 131=120C and C1=113220 3x 2 + 4 x − 2 1 25 72 13 − = + + ( x − 1)( x − 2)( x + 3) 20 x − 1 x − 2 x + 3 0 Exercise 64 .7 Partial fractions 53 To derive this result. Thus when when x1=13: x1=1−4: 51=17A and A1=1527 21=17B and B1=1227 0 Exercises 61–63 EXAMPLE 2.

If all the roots are different then P(x)2Q(x) can be decomposed into the sum of n simple fractions. if x11=1x21=1x3 in Q(x) then c c2 c3 c P( x ) = 1 + + + 4 + 2 3 Q( x ) x − x1 ( x − x ) x − x4 ( x − x1 ) 1 = ax 2 + bx + c ( x − x1 )3 c + 4 + x − x4 c + n x − xn + cn x − xn (2.30 A repeated linear factor in the denominator 3x + 1 ( x + 3) 2 = A B A( x + 3) + B Ax + (3 A + B) 3 + = = 2 2 x + 3 ( x + 3) ( x + 3) ( x + 3)2 and it follows that A1=13 and B1=1−8. x2. of the form ax + b x + px + q 2 (2.5 on the roots of the general polynomial).33) (ii) Some of the roots are complex. one for each real root. as in Examples 2.29: c c P( x ) = 1 + 2 + Q( x ) x − x1 x − x2 + cn x − xn (2. as in Example 2. if Q has degree n then Q(x)1=1a(x1−1x1)(x1−1x2)1-1(x1−1xn) (2. (i) All the roots are real. with powers of the linear factor in the denominator.34) .2). =. If complex numbers are disallowed then the denominator Q(x) can be factorized as the product of linear factors. If complex numbers are allowed then the discussion of (i) above is applicable. xn are the roots. the decomposition of a proper rational function P(x)2Q(x) in partial fractions depends on the nature of the roots of the denominator Q(x) (see Section 2.30. 0 Exercise 65 In the general case.28 and 2. one fraction for each quadratic factor. in addition to the terms discussed in (i) above. In this case Q(x) can be factorized as the product of real linear factors. and one or more real quadratic factors. The decomposition of the rational function in partial fractions then contains.31) where x1.54 Chapter 2 Algebraic functions EXAMPLE 2. one for each pair of complex conjugate roots (Section 8.32) If some of the roots are equal then there are additional terms. For example.

. . .... . ..... ... the equation x− 1 − x = y . .... . . ....... ... . . .. A second algebraic equation... ...... 1 ........ ... . ..... .... .. .... .. . .. . the two equations have the simultaneous solution x1=12...... ... . ......... ......14 shows that the graphs of these linear functions cross at the point (x..... . ..... ... ... ........ .. ...... .... .. .... .............. ... . ...... .. .. ........... ............. . ..... .. . ... ... .36) (2... .......... .. ..... . .. . . .... .. .... ... .. For example...... ........... ... .. ... ........In general....8 Solution of simultaneous equations 1 1− x + 2 x − 1 x − 2x + 2 2 am x + bm (2.... y1=11.. .... .... . . .. .. . ...... + = ( x + px + q ) ( x − 1)( x − 2 x + 2) 2 a2 x + b2 Figure 2... . . .. . . ......... ... ... ... . ... . .. ... . .... ... .......... ................ . .. ..... ... ........ ..... .... ... .... .......... For example (see Example 2.. . ..... ...... deﬁnes one function y1=1f(x). .. ....... ... ...... .... .. ... . . . ..... .. .•......... . .......... .... ........ ...... .... ... . . .... ... . y)1=1(2..... .... ...... ..... ....37) (2............... . ........ .22c).... ..... . ... .... . . ... ..... ... ...... Solution of simultaneous equations y1=131−1x y1=1x1−11 (2) x1−1y1=11 (1) x1+1y1=13 x − 3x + 4 x − 2 3 x + px + q 2 a1 x + b1 2 1 p′(x)y m1+1q′(x)y m−11+1-1+1u′(x)y1+1v′(x)1=10 p(x)y n1+1q(x)y n−11+1-1+1u(x)y1+1v(x)1=10 3 = y ..... ... .. . .. ... .. ............... .......... . ... . ....... ..... ........ ....... an algebraic equation in two variables x and y deﬁnes either variable as an algebraic function of the other... ... ..... ... . ...... .......... ...... . ...... ... .... Consider the pair of linear equations 2... .... .... ... . whereas equation (2) deﬁnes y as a second function of x Equation (1) deﬁnes y as one function of x Figure 2....... ...... .. .............. .. ..... ...... ... . . ... .... . .... . .. ......... .... .. ..... .... .. . ... ...... ... .. .35) 55 2 . .14 2 1 2 + + = ( x + px + q ) m 2...... ........... .8 or if the same quadratic factor occurs m times. ..... .. ..... 1). .... ..... ... .. ... .......... . .... ... ..... . ... ... ....... ......

Substitution for y in (1) then gives x1=15.39) is zero and. This is an example for which the denominator (a0b11−1a1b0) in equation (2. Graphically. and the solution is the point at which the lines cross. the equations (2. −2) 0 Exercises 66. EXAMPLE 2. graphically. at which the graphs of y1=1f(x) and y1=1g(x) cross. y)1=1(5. multiply equation (1) by 2: (1′) 2x1+12y1=16 (2) 2x1+13y1=14 and subtract (1′) from (2) to give y1=1−2. subtract twice (1) from (2): (1) x1+1y1=1 3 (2′) 01=1−2 The second equation is not possible. 0 Exercise 68 . The two equations have simultaneous solutions for those values of x for which f(x) and g(x) are equal. The lines therefore cross at point (x. it corresponds to parallel lines.39) (2. the two linear equations a0 x1+1b0 y1=1c0 a1x1+1b1 y1=1c1 can be solved to give the solution x= c0 b1 − c1b0 .56 Chapter 2 Algebraic functions deﬁnes a second function y1=1g(x). 67 EXAMPLE 2. if any. For example.32 Solve (1) x1+1 y1=13 (2) 2x1+12y1=14 To solve. The equations are said to be inconsistent and there is no solution.38) We note that this solution exists only if the denominator (a0b11−1a1b0) is not zero.38) then represent two straight lines. a0 b1 − a1b0 y= a0 c1 − a1c0 a0 b1 − a1b0 (2.31 Solve (1) x +1 y1=13 (2) 2x1+13y1=14 To solve for y. the simultaneous real solutions are those points. Graphically.

both equations give x1=131−1y for all values of y. both equations represent the same line. 0 Exercise 70 EXAMPLE 2.35 A line and an ellipse (1) 3x1+14y1 =1 4 2 (2) 2x 1+13xy1+12y 21=116 Equation (1) can be solved for y in terms of x and the result substituted in equation (2) of the ellipse (see Section 19.34 Three linear equations (1) x1+1 y1+1 z1=1 3 (2) 2x1+13y1+14z1=112 (3) x1−1 y1−12z1=1−5 To solve. It is discussed further in Chapter 20.2. ﬁrst eliminate x from equations (2) and (3) by subtracting 21×1(1) from (2) and (1) from (3): (1) x1+ y1+1 z1=1 3 (2′) y1+12z1=1 6 (3′) − 2y1−13z1=1−8 Now eliminate y from (3′) by adding 21×1(2′) to (3′): (1) x1+1y1+1 z1=13 (2′) y1+12z1=16 (3″) z1=14 The equations can now be solved in reverse order: (3″) is z1=14. y1=111−13x24. We will return to the general problem of solving systems of linear equations in Chapters 17 and 20. and (1) is x1−121+141=13 so that x1=11. and (2) becomes x 21−1161=1(x1+14)(x1−14)1=10 . The equations are said to be linearly dependent and it is only possible to obtain a partial solution. from (1).8 Solution of simultaneous equations 57 EXAMPLE 2. doubling equation (1) gives equation (2) and there is effectively only one independent equation. Thus. then (2′) is y1+181=16 so that y1=1−2. 0 Exercise 69 EXAMPLE 2. The method used in this example is a general systematic method for solving any number of simultaneous linear equations.33 Solve (1) x1+1 y1=13 (2) 2x1+12y1=16 In this case.5 on quadratic forms).

.............. . .. . ....... .... ..... Find the values of y1=12x 21+13x1−11 for: (i) x1=10 (ii) x1=11 (iii) x1=1−1 (iv) x1=1−223 3... .. ......... ........ .. ..... .......... ... . . . ...... . ..... .. .. .... .... .. 4x 21−19 ... . . ... ........ . .... . .. . ......58 1..... ... ... . y1)1=1(−4.. x 21+16x1+15 13. ....... ... .. ......... . ... ........... .... .... .. 2x 21+1x1−16 Chapter 2 Algebraic functions Section 2. . . 4 − ...... ........ . ...... express f(g) as a function of x......... .. ... ..... ... ...... ... .... ... ..... ... ... ...... ...• .......... ... ... .. ... ... . ... . . .. In this case the two solutions are the points at which the straight line crosses the ellipse.. .... .... ... what is: (iii) f(x1+11) (iv) f(x 21−13x1−14) (i) f(a1+13) (ii) f(a 21+11) 6..... .. .. . ... y .. ... If f(x)1=12x1−11 and g(x)1=13x1+11.. .... . . .... .. . .. . . ..... . Exercises 10.. ...... y) values and sketch a fully labelled graph of the quadratic: 8. ..... ... .. . ..... . ....... ... .. ....15.......... ... .. ... . ...9 0 Exercises 71–73 This has roots x11=1−4 and x21=14.. .. . . .. .. .2 Section 2.... .. .. . . y11=14 and y21=1−2.. .. . .. ..... ..... ... .. . .............. . ...... .... ..... 4 ........ . . ... . . ... . . ....... ... ... ........ ... .... . . .... ... .. . ........... .. . . .... ... ........ .. .... . .. ... ...15 11. .... ... .... ....... .. .................. ... .... ... .. ..... ...... . ..... .... .. .... . x 21−18x1+115 15.. ..... (x1.... ... ......... .... .. . .......... .... ...... . ........ . .... ...... . . . .. ... If f(x)1=1x 1−13x1−14... . . .. . 4 ....... ........... ..................... .. ... ... ....... . .. ........... ... . .. ............ ... .. .. ...... .... .. ..... .. ... ... ....... ... what is: (i) f(a) (ii) f( y 2 ) 2 5.. .. ... ... . ..... . .... .. . ... .. .. . ............. y1=1−x 21−1x1+12 7. .. ... .. . . ..... . ..... ... ...... ....... .... .... ..... 6x 2y 21−12xy 31−14y 2 12.. ......... . ........ ...... . y1=1x 21−14x Section 2.... ... .... .. ... ........ .... ..... .. .. .. ........ .. . .. .. ..... .... . . 4) and (x2.. . ... ..... . ... ... . . ....... .......... ... ... . .... .... .. ... ...... .......... ... .. ..... ... .. ........... ... .. .... .......... ..... ........ ......1 2....... x41−110x 21+19 Figure 2.... ............ . ...... ..... . ...... ..... . .. . ....... .......... ..... ..... .. . ... . . . .... . .... .. ........ .. x 21−14 16. .... .. ..3 Make a table of (x.. ..... ... .... ... . y2)1=1(4... . .... ............ ..... . .•... .. .. .. Find the values of y1=121−13x for: (i) x1=10 (ii) x1=12 (iii) x1=1−3 (iv) x1=1223 2.... . .. ....... ... ... .. −2).. ..... . .......... as in Figure 2.. If f(x)1=12x 21+14x1+13... . .. . .... . .... ...... .......... 2 . ... .. ..... .. with corresponding value of y... .... x 21+1x1−16 14. . ... x ... . .......... Factorize: 9..... . .. .. ... .. . ﬁnd: (i) f(5) (ii) f(0) (iii) f(−2) (iv) f(−223) 4....... .. Given f(x)1=1x 31−13x 21+14x1−13... ... ...... . ....... . . . .... . . . .... ........... .. ........ . ... . ......... −.. . .. .

y = (3x + 1) 2 28. Express B as an explicit function of the other variables. 35. n= 0 ∑ n! x n 2 Find the equation and sketch the graph of the straight line that passes through the points: 40. x 3x + 2 x 2 18. (−1. 4) 41. and B is the second virial coefﬁcient. R is the gas constant. verify the explicit expressions given for T and p. y = x −1 2x + 1 1 25. y1=1(x 21−11)2 30. y 21=1x 21+11 32. y = 2x + 3 3x − 2 1 24.12 on the van der Waals equation. −5) and (1. x+2 x+4 19.9 Exercises 59 Simplify if possible: 17. Vm is the molar volume. 6) and (3. x2 − 9 x + 5x + 6 2 2 x 2 − 3x + 1 x 2 − 3x + 2 Section 2. y1=1x1−12 27. The virial equation of state of a gas can be approximated at low pressure as B pVm = RT 1 + V m where p is the pressure. Express c as m an explicit function of Λm. where K is a constant. In Example 2. y = x 1− x For y1=1f(x). Section 2. ∑ ix i−1 i=0 38. and the cubic equation in V. x2 − 4 x−2 20. T is the temperature. (ii) sketch graphs of y1=1f(x) and x1=1f −1(y): 31. x 2 + 3x + 2 x+2 21.4 Find x as a function of y: 23. (−2. 34. 22.5 Expand (write out in full): 36. The Langmuir adsorption isotherm θ= Kp 1 + Kp gives the fractional coverage θ of a surface by adsorbed gas at pressure p. 33.2. y = x2 − 1 x2 + 1 26. Express p in terms of θ. (i) ﬁnd x as a function of y. 2 3 3 3 n= 0 ∑ ( n + 1) x n 37. y = ( 2 − x ) 3 29. Kohlrausch’s law for the molar conductivity Λm of a strong electrolyte at low concentration c is Λ m = Λ0 − K c m where Λ0 is the molar conductivity at inﬁnite dilution and K is a constant. −2) . ∑ k ( k + 1) x − k k =1 39.

2 x 2 − 5x + 7 x ( x − 1)( x + 2) ( x − 1)2 ( x + 2) . 3x 21−13x1−11 47. Find the roots and sketch the graphs of the quadratic functions: 45. 3x 3 − 2 x 2 − x + 4 x+2 59. If y = 2x2 + x + 1 2x2 + x − 1 ﬁnd x as a function of y. (i) ﬁnd the roots. NA. and c are positive quantities). 43. −3x 1+13x1−11 48. m Λ m = Λ0 − K c m can be obtained graphically from the results of measurements of Λm over a range of concentration c. Given that x 1−11 is a factor of the quartic x 1−15x 31+15x 21+15x1−16. 1 ( x − 1)( x + 2) x + 2x − 1 2 62. The acidity constant Ka of a weak acid at concentration c is Ka = α 2c 1− α where α is the degree of ionization. (i) ﬁnd the roots. 60. x 1+1x1+12 50. and ε0 are constants. where ρ is the density at temperature T. Explain how µ and α can be obtained graphically from the results of measurements of εr and ρ over a range of temperatures. (ii) sketch the graph. −2x 21−13x1+12 46.7 Express in terms of partial fractions: 61. The Debye equation εr − 1 εr + 2 = ρ NA µ2 α + M 3ε0 3kT relates the relative permittivity (dielectric constant) εr of a pure substance to the dipole moment µ and polarizability α of the constituent molecules.60 Chapter 2 Algebraic functions 42. 4x 1+14x1+11 51. Given that x1−11 is a factor of the cubic. 65. x 31+14x 21+1x1−16 2 4 56. Explain how K and Λ0 in Kohlrausch’s law (Exercise 33). x−2 x + 3x + 2 2 64. k. Ka. and M. 2x − 1 x+3 58. −x 21+16x1−19 44. x 31−16x 21+19x1−14 55. x 3 + 2 x 2 − 5x − 6 x +1 2 x 4 − 3x 3 + 4 x 2 − 5x + 6 x2 − 2x − 2 Section 2.6 Use algebraic division to reduce the rational function to proper form: 57. x 21−13x1+12 2 2 2 49. 52. Section 2. x 31−13x 21+13x1−11 53. Express α in terms of Ka and c (remember that α. x+2 x ( x + 3) 63. (ii) sketch the graph: 54.

2x1−1y1=12. 6x1−14y1=12 2x1−1y1=11. x 21−1xy1+1y 21=11 67. x1+1y1=13. x1−1y1=11 68. x1−12y1+13z1=13. x 21−1xy1+1y 21=11 . 2x1+13y1=12 3x1−12y1=11.8 Solve the simultaneous equations: 66. 3x1−12y1=11. 3x1+13y1−1z1=11 72. 73. 69. 71. 2x1−1y1−12z1=18.2. 6x1−14y1=13 70.9 Exercises 61 Section 2. 3x1−12y1=11. x 21−1xy1+1y 21=11 2x1−1y1=13.

and a brief description of the hyperbolic functions is given in Section 3.1 Some applications are illustrated in Figure 3.1 Concepts The mathematical description of physical phenomena often involves functions other than the algebraic functions discussed in Chapter 2. and the exponential function and its inverse function. Trigonometry as a distinct branch of mathematics has its origins as a tool for the construction of astronomical tables. Figure 3.2 to 3. discussed in Sections 3. Other functions are deﬁned in terms of these ‘elementary functions’. the logarithmic function. Three of these involve trigonometric functions. discussed in Sections 3. It is concerned with the use of trigonometric functions for the solution of geometric problems involving .1.5.6 to 3.9. The most important of these transcendental functions are the trigonometric functions.8.3 Transcendental functions 3.

... . ﬂight behaviour of some birds and. .. .. . and is involved in the deﬁnition and representation of fundamental physical concepts in thermodynamics... .. . .. . focused for the ﬁrst time on the properties of the right-angled triangle. .. . known as the Almagest. . The trigonometric functions are important in the physical sciences for the description of periodic motion.. .. . .. European trigonometry was developed by Johann Müller (Regiomontanus) of Königsberg (1436–1476). .. .3) and (1. .. When k is negative.. . etis opp o B tn e c ajd a C θ esun t op h e y A .. .. .. . . . .2) in Chapter 1. . are the sine (sin). radioactive decomposition of nuclei and. Figure 3. Arab mathematicians (about 950 AD) added new tabulations and theorems. . . new formulas. the internal angle at A of the rightangle triangle in Figure 3. . .. The exponential function e x has the unique and deﬁnitive property that the slope of its graph at any point (its derivative... such as entropy and chemical potential. . . ... . was called by the Arabs al-magisti (the greatest) and whose tables were used by astronomers for over a thousand years. ... . . . Another example is given in Figure 3. the popularity of fashions. . .. . . . .. the shape of hurricanes and some galaxies. . .. .. including circular motion.. . . ... .1 and is important in structural and architectural design.3) Figure 3... . ...1(ii). .. . François Viète (1540–1603) built on this work with extensive new tables for all six common functions.. a student of Copernicus. . sin θ = cos θ = tan θ = 1 opposite BC = hypotenuse AC adjacent AB = hypotenuse AC opposite BC sin θ = = adjacent AB cos θ (3..1(iii) show exponential growth (with k positive) which provides a model of. . . . . The word ‘trigonometry’ came into use in about 1600.2 Trigonometric functions Geometric deﬁnitions The principal trigonometric functions of the angle θ. .. .. . . the cosine (cos). .... and navigation.. . . .. . 3.. and the use of trigonometric functions for the solution of algebraic equations. . as in ﬁgure 3.. see Chapter 4) is equal to the value of the function at that point.. . . . astronomy. They are also essential for the description of systems with periodic structure. more trivially... . Figure 3..2 The earliest ‘trigonometric tables’ were constructed in about 150 BC by the astronomer Hipparchus of Nicaea (modern Iznik in Turkey)... . ..2) (3. ....1) (3. The Arrhenius and Nernst equations...1(i).... . .. . . .3. . .. .2. .. . . . 100–178 AD) whose Syntaxis mathematica (Mathematical Synthesis). . the function e−r is the 1s orbital of the hydrogen atom if r is the distance of the electron from the nucleus. ... This makes it particularly useful for the modeling of all types of ﬁrstorder rate processes... . . .. for the interpretation of the diffraction of X-rays from the surfaces of crystal lattices. .. and by Georg Joachim Rheticus (1514–1576) of Wittenburg. in which the rate of change (growth or decay) of a property is proportional to the value of the property. . . ... In quantum theory. . .... ... The logarithmic function ln1x is the inverse function of e x... . .. . and by Claudius Ptolemy of Alexandria (c... . equation (1. .... provide an example of the dual relation of the inverse functions... .4).. .. 1596. .. . .. It is often used as an alternative to the exponential. . .. . . .... . ... as in the use of Bragg’s law. ..... .... on a larger scale. as in the growth of shells of molluscs.. . ... . population explosion. . exponential decay provides a model of ﬁrst-order chemical reactions. .. . and the tangent (tan): . .. .. ..... . .. .2 Trigonometric functions 63 triangles... . bacterial growth in a culture. .. and a nuclear chain reaction. . (1.. . . to whom we owe the 360° circle. . for example. . whose Opus palatinum de triangulis (The palatine work on triangles). . and wave motion... The tables in the Hindu Siddhantas (about 400 AD) are essentially tabulations of the sine function. ....1(iv) by alternative equations for the logarithmic spiral observed in natural phenomena.. . .

.. . . ..3 One of the best known properties of the right-angled triangle is the theorem of Pythagoras. . . .. . . . . .. .. . . . .. . number theory and geometry. . . .. ... B 3 A sin φ = θ 3 5 4 φ 5 cosec θ = .. . . .. ....... . . ... and the existence of irrational numbers was possibly a discovery of the later Pythagoreans in about 400 BC.. .... ... . sin θ = 4 5 5 4 cos θ = sec θ = cos φ = 5 3 sec φ = 3 5 5 3 4 5 5 4 tan θ = cot θ = tan φ = cot φ = 4 3 3 4 3 4 4 3 cosec φ = 0 Exercise 1 Figure 3.... . ... he travelled widely and was one of the principal importers of Egyptian and Babylonian mathematics and astronomy into the Greek world. . . The general form of the theorem of Pythagoras is: ‘In a right-angled triangle..... .. . . . The Pythagorean school introduced the systematic study of the principles of mathematics.. ‘The elements’.. Pythagoras’ theorem was known in the old Babylonian period. cosecant (cosec).. .. .. ... . . . . . . . ... ....4) EXAMPLE 3. .. . .5) + BC2 AC2 =1 (3.. ... . . . .. He is reputed to have coined the word ‘mathematics’.. . the most important being the secant (sec). . . .. .. 580–c.... He settled in Croton. . . 2 (3. . . . . ... . . . 500 BC). ..... ... . . . . ... .... C .. .. . . .. ..... . Born on the island of Samos. . Proposition 31 in the Heath translation)....... . ..... .. ... . and cotangent (cot): sec θ = cos θ 1 1 1 ...2 AB21+1BC21=1AC2 This can be written as a trigonometric equation by dividing both sides by AC2: AB2 AC2 or sin21θ + cos21θ1=11 (a quantity like (sin θ)2. . ..... .. . ..6) Pythagoras (c. .. . ...... . . . cosec θ = ... ..64 Chapter 3 Transcendental functions Several derived functions are also deﬁned. ... . . . . . . . .... ..... ..3.. . ..... . . ... .. in Southern Italy....... . ... .. . . the area of the ﬁgure on the hypotenuse is equal to the sum of the areas of the similar ﬁgures on the other two sides’ (Euclid.... . .... ..1 For the angles in Figure 3. the square of sin1θ.. .. . . .. .... . .. motto ‘all is number’.. .. .. is usually written as sin21 θ)... where he founded a religious and philosophical society with a strong mathematical basis. . . . .. ... Attributions to him of mathematical discoveries are traditional.. . . cot θ = = cos θ sin θ tan θ sin θ (3.... .. that which is learned. . . ..... . .. .. .... Book VI.. .

. .... .... ... .... . ... .............. .... .. .... ........ .. ...... . ... .... .. ..4 shows the right angle (90°).... . .. ............. ............ ..... the radian.. ...... .... . .. .... ..... . . ... . ..... The length of the arc is proportional to the size of the angle......... ... sometimes a superscript c is used.... .... .. doubling the angle θ doubles the length s...... .. ... . . ...... . ... .. ... ..... ... . .. ... . Figure 3. .. .. .. the angle on a line (180°) and the angle round a point (360°)... .. ... ... . .. .... ... ....• . ..... ....... . . ... .... .... ......... .... . .... ............. . .. .. ..... . . . .... .. ......... .... .. . for example s r θ r . Figure 3. .. ... .... .. ..... ... .. ................ .. . .. ... . ..... ....... .. . ... .. . .. .. ..... ... .. . .... . ... . . ..... .. . .. ........... .. .... . . .. ..... ..... . . so that 1c1=11 rad).. ... .. .. .. ...... . . . . ... ... .... ... . ... deﬁned in terms of the properties of the circle. .. ...... .. .. ... ..... ... ....5 shows a circle of radius r. . . .. ... In practice.. .... .... . . ... ...... ..... . . .. .......... . .. ...... . .. .... . . .. .7) 65 . ......... .......... .... ... . .. . ... . . ...... ... . ... . ... ...... . .. ............ ........ . ... Units of angle 0 Exercises 2 EXAMPLE 3....... .. . ... the symbol for the unit is omitted...... . . ... .. .. . .. .. ...... .... .... ....... .. . . ..... ..... .. ..... ...... ...... .. . ......... ..... . .... . ....... . .. . ..... . ..... .... .... .... .. .... .......2 For the triangle in Figure 3.... .. .. . ......... .... . ... ... . ... ... . .. ....... . ..3.. ... ... ....... . ..... ....... It follows that the ratio of s to the circumference of the circle (2πr) is equal to the ratio of θ to the complete angle around the centre (360°): The ordinary unit of angle is the degree... . ... 321+1421=152 and (rad is the SI unit of angle. .. ... ............. .. . . ......... .. .. ... . ... .... ... ........ ...... . ... ... ... ..... . .. . . ... .... .. ... ............... . .. . .. ... ... ... ... . ........ . .. .... . .. . ... .... . . . .... .... . .. .. .... .. . . .. .. . and an arc of length s subtending the angle θ at the centre. ..... .. .... ... ................. . .... . . ... ...... ....9) (3. .... ...... ....... ...... ....... ... ... . .. . ... ............ . ........ . . ....... ....... . ..... ... . .... .. .. . . .. .. .. ...... .. ..... . . .. . .... . . ... . .. ....... ... .. ..... . .... . . . .. ◦06 3 The unit of angle that is always used in mathematical and scientiﬁc applications is the radian (SI symbol rad)... ...... ... • ◦08 1 The unit of angle. ...... . .. .. ....... . .. . for example... ...... . .... ......5 3. . . ........ .. . . .... ...... ..... ..... . .. ... . ... .... . ...... ..... is deﬁned by • ◦09 so that 4 3 42 + 32 25 sin 2 θ + cos 2 θ = + = = =1 25 5 5 52 2 1 rad = 2 θ= s θ = 2 πr 360 º Figure 3.. . and an angle given without unit is assumed to be in radians. . . .. . ..... ....... ....2 Trigonometric functions (3... . ... .. . ...... ........ ............. .......... ... ..... ..4 360 º ≈ 57 º 18′ 2π s 360 º × r 2π Figure 3... .. .. . .. ... .... ..8) (3. . ....... . ........... .... ..... ...

.. ... The length of an arc of a circle is s1=1rθ when θ is in radians. . .. .. .. . . .... III . . . ..... . .. ..3 Radians and degrees (1) The angle 40° in units of the radian is 40 º = 40 º × 2π 2π = ≈ 0.... . .. .. .. .. . . ....... . . . .. ... ... .... .. .6 º 2π π (3) The length of the arc that subtends an angle of θ1=12 rad at the centre of a circle of radius r1=13 is s1=1rθ1=131×121=16 0 Exercises 3–5 Trigonometric functions for all angles The geometric deﬁnition of the trigonometric functions in terms of ratios of the sides of a right-angled triangle restricts the functions to values of angles in the range 0° to 90°. 180 º = π. .. . ..... ... .... .5 × 360 º 90 º = ≈ 28.... . .. . . ... Figure 3.. ... . . .. .. . . . . . ....... ...66 Chapter 3 Transcendental functions 90 º = π 2 .. . .5 rad in degrees is 0..... ... . or 0≤θ ≤ π 2 The deﬁnitions can be extended to all values of the angle by consideration of the coordinates of a point on a circle. . . . ... .. . .. . .. . . . . . . . .. ... . . . .. . ... . . . ... . .. . . . .x( . . ..... ... . .. ... . ) y . . . .. . r II .... . . .. . . ... ... ... . . ....... .. . . . .. . .. EXAMPLE 3.... .. . . ...... .. . . . .... . .. ... .. . .. . . ... .... .. .. .6. . .... . . . .7 rad 360 º 9 (2) The angle 0. . . . . ... . . . . . . .. . . ... . . ... . .... . . . .. . .. . ... . . .. ... ... . . as shown in Figure 3. .. .. ... .. . . .. .. .. . .. . .. .. . . . . . .... . . ... . . .. x . . . . ... . 360 º = 2 π. . ... . . ... sin π 6 = sin 30 º (angles are often quoted as multiples of π).. . . . .. . . ..... . .5 = 0. . ... . .. . . ..... .. . . . .. . . ... VI . . . ... I . ... . . .6 θ y • . .. . . ... ..

with angles 01<1θ1<1π22.10) 67 . ... . ... . ..... . ... .... ... 4π23 in the third... .. .. . ...... . .. . . . .. . . .. .. . . . ..... . . . ....... . .. .. . .. .. .. . . . . . . EXAMPLE 3.......... ..... .. . .. . ..... . ... . . . .. ... . . . I.4 By Table 3...... .. . . . ..... . ... . by Figure 3. ..... ... .....7.. . .... .. .... . . . . . . ..... . . .. . . ..... ....... ... . .1 Signs of the trigonometric functions In the second quadrant however. .. .. . .. . ... . .... . . . . .... . . 5π23 in the fourth. . . .. ... .. .. ...... .. .. . .. ... .. ... . .. .. . .... Therefore. . .. . x is negative and y is positive..... .... . .. .. . . . ..... . .. .. .. . . The signs of the trigonometric functions in all four quadrants are given in Table 3..... Angles sin12π231=1+ 3 2 sin15π231=1− 3 2 sin14π231=1− 3 2 01<1θ1<1π22 01<1θ1<190° + + + I sin θ = sin1θ1>10 π221<1θ1<1π 90°1<1θ1<1180° y r cos15π231=1+122 cos14π231=1−122 cos12π231=1−122 II − − + Figure 3.. ... . . . . What are the values of the sine. .7 cos θ = cos1θ1>10 x r π1<1θ1<13π22 180°1<1θ1<1270° tan1θ1>10 tan θ = tan15π231=1− 3 tan14π231=1+ 3 tan12π231=1− 3 III + − − y x 3....... ... ... . ..... .... .. .. ..... ..... . .. . .. . . ... .. . . . .. . ..2 Trigonometric functions 3π221<1θ1<12π 270°1<1θ1<1360° IV − + − (3. .. .. ... . .. .. . ... . .. .. .... . .. . . .... .. .. ... ... ... .. . . . .. ... . .. .... ... . . with π221<1θ1<1π... . .. . . ... . ... ... ... . ....... .... ... . . . ... .. . . .... . . . .. .. . . .. .. .... . .. . . 4π23. .. ...... ... . ... . .. ... .. . .. . . ... .. . . .. .. .......1.. . . ..... .... . . and 5π23? 0 Exercises 6 tan1θ sin1θ Quadrant In the ﬁrst quadrant.. . . . ./π 5 3 • • /π 4 3 •π 0 • − + + + • − − − + • . .. .. . . cos1π231=1122. .. . .2.. ....... . .. . . .. . . ... . . . . . . . ... . . ... . ... ... . . .. and tan1π231= 3 .. . . .. . . .. . ....... . . . . .......... . . ... . ...... ... . ... . .. ....... . . . .. .. cosine.. . ... .... ......... . ... . .. . . ........ ... .... .... ... . . ...... .. . .. . .. .. .. ...... sin1π231=1 3 2... ... ... ..... .... . .. .. .... .. . . and tangent of 2π23. .. . ..... the values of x and y are both positive and The trigonometric functions are deﬁned for all points on the circle as Table 3. . .. .. . .. . . . ... . ... . . ... ... .... . ..... 3/ π /π 2 3 cos1θ The angle 2π23 is in the second quadrant.. . .. . .. .. . . ... . .... . . . .. .. ... . . ... . .... ... . . .. .. ...... . ... . . . ... . . . .. .. .. . ... ... . .... ... ...... . . . ... . ........ ..... .. . . .. . ...... . .. . .. . . .. .. . .. . .

.. .... . . . .... .. .. . sin1θ1=1a2c. . cos1π261=1 3 2. . .. . . ... . .. .... . . .. . . . .. ..... ... . . ... . .. ... . .... . . . . . . ... . . ..... .9 0 Exercise 8 b θ As the angle θ increases to 90°... .. . ... . . . . ... .. ... .. . .. .... . . . .... .. . . . ..5 Demonstrate that sin1π221=11 and cos1π221=10. . .. . . . the magnitude of side b approaches zero whilst that of side a aproach the value of c. . .. . . .....2... . . .. . ... ... . . ... . . ..... .. ... . . . cos1θ1=1b2c sin1θ1→1c2c1=11 0 Exercises 7 and cos1θ1→102c1=10 as θ1→1π22 Figure 3. tan1π261=1 1 3 Figure 3. .. ... . . . . . .. . ... . . . . . . . . ... . . .. .. . . ..... . .. .. . . . . . . ... ... . . .. . .. . .. .. Therefore a c . . ..2 Some special values θ θ 2rad sin1θ cos1θ tan1θ 0° 0 0 1 0 90° π22 1 0 ∞ 180° π 0 −1 0 270° 3π22 −1 0 −∞ 360° 2π 0 1 0 30° π26 122 3 2 1 3 60° π23 3 2 122 3 45° π24 1 1 1 2 2 EXAMPLE 3. . ... . . . . . .. . . . . and sin1π261=1122. ..... . .9. .. ... . ... . .. ... . .. . .... .. . ...... h = 22 − 12 = 3 .. . are listed in Table 3. .. .. . . .. . . . . . .. .. . ... . .. . .. . . . . ... 2 1 h ◦03 1 ◦0 6 2 . . .. . .. . . ... .. . . . . . . . . . . . .. . . .. . . . .. . . .. . . . ... .. . . .. by Pythagoras.6 Verify the values of the trigonometric functions for θ1=1π26 in Table 3. .. . . ... . .. . . .. ... .. . . . . . . .... . . .. . ... . . .. . . . . .. By Figure 3. . .. . .. ... . . .. .. . ... . .. .. .. . . . . . . . .. .. . ...2.. . .8.. .... . ..... . . .... . ... . . ... . . . . . . . . .. ... ....... . . . . Then. .... and for a few other angles... . ...8 EXAMPLE 3. .. .. ... .. .. .. . . . .. . . . .... ..... . .... . Draw an equilateral triangle with sides of length 2 and bisect the triangle as in Figure 3.. . . .68 Chapter 3 Transcendental functions Special values The values for the angles on the boundaries of the four quadrants... . . . ... . . .. .. . .. . ... . . .. . .... .... . ... . .... .. . .. . . ... .. ... . . .. .. . .. . . .. .. . . Table 3... . . .. ..... . .. . ... . ... ..

... . . . . ....... .. have the same trigonometric values: sin(θ1±12π n)1=1sin1θ. ... .. . . . . ... . ... ... . and the angles θ1±12π n. . . . . ... . .. . .. .... .....12) In addition. . ... . . . . .. . .. . ... . . each point corresponds to an inﬁnite number of angles. tan(θ1±1π n)1=1tan1θ (3. . . ..3. .. ... . Each complete rotation adds or subtracts 2π... The graphs of sine. .... . . . .. . .. . .. . ..... . ... .2 Trigonometric functions 69 Negative angles Each point on the circle can be reached by either anti-clockwise rotation or by clockwise rotation. . .11) .. .. . whereas every angle corresponds to a point on the circle. ... ..... . . . .. for all values of the integer n1=10. . . . . .. ... . .. 1. . . . . . .... . . . cosine. ... . 3 =. . . .. . ... .. ........ .. . Figure 3.. . . .. .... .. . .... . ...... . .. .. ... . . .. 2. ... . .10 The point P in Figure 3. . .... . . .... .. . . . An angle is deﬁned to have positive value for an anti-clockwise rotation and negative for a clockwise rotation. θ (3.. . . . .. . . . . . . .. . . .. .13) We see that. . ... ... . . . .. the tangent repeats every half rotation. . . .... . ... corresponding to the negative angle −θ. . .. . ...3 3 The graph of the sine function was ﬁrst drawn in 1635 by Gilles Personne de Roberval (1602–1675). . . .. . ....... . .. can be reached by anti-clockwise rotation through angle (2π1−1θ). . and the two angles have the same trigonometric values: sin(−θ)1=1sin(2π1−1θ)1=1−sin1θ cos(−θ)1=1cos(2π1−1θ)1=1+cos1θ tan(−θ)1=1tan(2π1−1θ)1=1−tan1θ Further angles The range of allowed values of the angle can be extended further by allowing one or more complete rotations around the centre. . .. .. . and tangent are shown in Figure 3.....10. . . .. . . .... ... . .. . . . ...... . . . .. cos(θ1±12π n)1=1cos1θ (3... .. .. . . ... θ − π2 θ− P• • .... .. . ... ........ . . ... . .... . .. .. . . .. .. ... . . ..11. .. . ... . ........ .. . .. .

.. . . . . . .... ... . . . . . .. ..... .. . ... . .. .. . . . .. . . .. . .. ... . . . . .. . . . .... .. .. . ... . . . .. . . . . . .. . . .... . . . . . .. . ... . .. ... . . . . . . . .. ... . . . ... .. .. .. . . . . . .. ... . . the tangent is periodic with period π.. . . . . .. . ..... . . . . . . . . . . . . .. . . .... ... . . ... ..14) .... .. . .. . ... . . . .. . . . .. . .. . the value of the function is unchanged when x is replaced by x1+1a or x1−1a. . .. .. . ... .. ... . . . . . . . . .. . .. . .. .. . . . . . . . .. . . . . . .... . . ...... . ....... . .. . .. .. ... . .. . .. . . . . . .. . .. ..... .. .. . ..... . . .... . . . . .. . .. x nis = y Periodic functions π3 2 − )c( .. .. .. ... . .. . .. ... . . ..... . .. .. .. .... .. .. .. . . . . ... . 1+ y A function f(x) with the property π 2 − . . . ) b( .. .... .... . π −. . .. . . . .. . . . . . . .. . . ... . .. .. . .. . . . . .. . .. . . . .. ... .. . .. .. . . ...... . . . ..... .... . .. . . .. .. .. The sine and cosine curves in Figure 3. . . ... . .. . . . . ... . .70 Chapter 3 Transcendental functions is called a periodic function of x with period a. ... . .. .. . .... . .. .. .. . . . . .. . ... ... ... . . .. . . ... . . . 10 f(x1±1a)1=1f(x) Figure 3. . . .. . . . . . . . .. . . . . ...... .. . . . . . . . . . .. .. . . .. .. . . . ... .. .. .. . . . . .. .. . .. ... .. . .. . . . . .. ..... . .. . .. .. ... . . .. . .. . . . . .. .. . . .. ... . . . .. .. . .. . .. . . .. . ... . . 1 . .. .. . . .. . .. .... . .. .. ... .. .11 are called harmonic waves. .. .. . ... . . . . . . .. . . . .. . . . . . . . .. ... .. ... . .. .. . ...... .. ... . .. ...... .. .. . . . .. . . . . . .. ... . ..... ... . . x . . . . . . .. . 1 .... .. . . . . . . .. .. . . . . .. . . ... .. . . . .. .. . .. ....... . . . . . . . . . ... .. . . ... ... ...... . . . . . . .. . . . . .. . . . . ...... .. The sine and cosine functions are periodic functions with period 2π. . .. . . . . . .. . and the functions form the basis for the description of all forms of waves and other oscillatory motions. . .. . . . .. ..... .. .. ... . .. . .. .. .. . .. .. . . . .... . . . .. . . . . 1+ y . .. ..... ... . .. .... . . . . . . . . . . .. .. .. . .. π2 − . . .. . .. .. . . .. . .. . .. . . .. . . .... . . . . . . .. π 2 .. . . . .. . . . . . .. .. . . . . . . . . . .. . .... . .. . . .... . . . ... . . ..... . . . . .... . . . ... . π2 . ... ..... . .. .. . . . .. .. . .. . . . .. ...... . . . ... . . . .... .. .. . . . .. .... . . .. .... .. .. . . ... . .. . .. .... . .. . . ... . . . . . . . ... . . ... .... .. . . . ... . . x π3 2 π o . . .. .. . .. . π .. . .... . .. . .. .. . .. . . .. ... . .. .... .. .. . .... . .. .. .. . . ... .. . . . .. . . ... .. .. . .. .. .. . . .... . .. . . . .. ... . . . . . .. . . .. .... ... . . . ) a( 0 Exercises 9. . . . . . . .. . . . . . .. . .. .... o . .. . ....... .. . . . .. . .. . − o .. . . .11 (3. .. . . . . . . ..... ... .. .. . − . . . . .. . ... . .. .. . . . ..... π . x s oc = y . . .. .. .. . .. . . . .. . . . . . .. .. ... .. ..... . .... .. .. . . .. .... . . ... .. .. . . . .. .... . . ... .. .. . . . . .. .... . . . ... .. .... . . . .. .. .... . . .. . ... . . ... . .. .. . .. . . .. ... .. . . . .. ... .. . . . . π2− .. .. .. . . . .. .. .. . ... . .. . .. . . ... . . . . ...... x π2 . .... . . . ..... . . . . . .. . . . .. . .. . ... . .... . ... ... . ... . . .. . . ..... . . . . . . ...... . . . . . . ... .. ..... .. . . .. . .. ... .... .. . .. . ... . . .. .. . .. . . .. .. . . . .. . .. . ... . . . ..... . . .. .... . .. ....... ..... .. .... .. . . . .. . . . . .. ......... . .. . . ... . . . .. .. . .. ... . . . .. .... . . ... .. π− .. . .... .. ... . . ... .. . π− x n at = y . . ... . .. . . . . .. .. . y .. . .... . ... ..... . . . . ... . .. .. . . .. ... ..

.. .... . Newton’s second law of motion states that the force acting on a body is equal to the mass of the body times its acceleration: EXAMPLE 3.. .. .. ... . .. . . ..5)..... . ..htgnelevaw = λ .. . . .20 and Section 12... . . . . . .. .. .. the time of a complete oscillation........ . . . ... . A )0 .... and τ1=112ν is the (time) period....... ...... .... . .... . ...... ......... .. . . .... ....... .. . . . .... . ... . .. .. .. .. . ..... x( φ 0 Exercises 11 The shortest distance between equivalent points on the curve (the period with respect to changes in x) is the wavelength λ.... . ... . . A− . . .. .. .. . .... ..... .. .... ....... . .. ........ . ..12 3... .. . .... ..... . ... . . .... ..7 A harmonic wave travelling in the positive x direction (a plane wave) is described by the wave function (or by the equivalent cosine function)... . ....... . .......... .. ... .. . .. ..... .. . ... .... . .. . . ... . .... .. .... .. ... . ..... . . ... . ... ..... . . .. . ... ..... . . ........ .. ........ ... . . .. .... . .. . ... .. . .... .. ... . .12 at time t1=10....... . ... .. .. ... ... .. ... ... ..... .. ... .. . .. . . .. . t ) = A sin 2 π − kx = m d 2x − νt λ x Figure 3.. . ...... . For the simple harmonic oscillator (see Example 2. ..... .. . .. . .. ... .... .. ... .. . ..... . ... .. ... .. .... ..... . ....... . ........ . .. ... . f 1=1ma φ ( x .. .. . . . ..... .. . ... . ... ........ . .... ........ .... .. .... ... ... The number A is the amplitude of the wave. ... ...... ...... .. .... ... . ..... .. ... . .... . νλ = v •...... . .. . .. . . . . . .. . .... . ..... ... . . . .. . ... . .... . . λ . .. .. ... The speed of propagation of the wave is v1=1λν..... . ... ... .... . . . .... ... .... ..... ....... .... . ............ .... . . . ... .. . where ν is the frequency.. ....... . .. ..... ... .. ........... . .......... ... .. x λ2 . or number of oscillations per unit time. ....... ....... .. o .. . .... . .... . ........ ... . .... ........ ..... .. . . . . . and is shown in Figure 3. ..... . . . . ........... . . . .. .. . .. . ....... .. . the force is f 1=1−kx and the acceleration is the second derivative of distance x with respect to time (see Chapter 4 on differentiation): EXAMPLE 3. . . ... . .. . . . ..... ... .... . . ..... . ....... .. . . .. . .. ..2 Trigonometric functions 71 dt 2 ... ..... .. ... .. . ..... .... .. ... .. . . . .. ...8 In classical mechanics..... ... . . .... . . . ...... . .... .... .. ... .... . .. .. .. ..... .. .. . . .......... .. . ... ... .... .... ... ....

.72 Chapter 3 Transcendental functions This is one of the simplest second-order differential equations (Chapter 12) and a solution is x(t)1=1A1cos1ω t where A. .. ... ... . . as indicated in Figure 3.. ... ..15) Because of the possible confusion between the notation for the inverse sine. . . ...... arctan1y1=1tan−1 y (3.... and is given by the inverse sine function sin−1 y: if y1=1sin1x then x1=1sin−1 y (3.. . . sin−1 y. . . . . . . . . ... . ....... .. .. .. .. ... . . . . . ... ... . .... . ..... . .. ... .. .. . . . ..... ..... . . . . . .. . . ...... .. . . .. . . . . .. . . ... . .. . . x−π •. . .. .. .. . ....... .... .... . . . .. . ..... x . ....... ... . . ... . .. .. . .. . ..13 x nis = y x . . A plot of displacement against time 2π m is very similar to that in Figure 3..... The displacement x(t) is periodic with respect to time... ... . for example.. .. .. .. .. . .. ... . .. . . .. (sin1y)−1. . ... ........ . .... .. . .. . . . ... ... .... . ... . . ...16) The inverse functions are multi-valued functions.. • 2 .. . . . . . . ..... .. many angles have the same sine: sin x1=1sin1(π1−1x)1=1sin1(x1±12n π) (3.... ... .. ... ..... .. .... . . .. .. . .. .. . . .. ..... . .. .. ..... .. ... .. . .. .. . arccos1y1=1cos−1 y. .. .. ... . . . .... an alternative notation for the inverse trigonometric functions is often used: arcsin1y1=1sin−1 y.. .. ... . . ... . . with period τ= 2π ω = 1 ν where ν = 1 k is the frequency of vibration. . . .. .... ... .3 Inverse trigonometric functions If y1=1sin1x then x is the angle whose sine is y.. .... ... ...... ... ... . . . . .. .... .. .... . .. 3. .. . . .. . . . . .... . . .. .... . . . . .. . .13.. .. . . . . ... . ...... . • y o 1 + 1 − . . . and the inverse of the sine. .. . . .3)... .. . .... . . . . ... .. .. .... .... . .......... . . .. .... . .. is the maximum displacement from equilibrium and ω = k m is called the angular frequency. . . .. .. . the amplitude.... .. ... .12 (see Figure 12. ..17) Figure 3. .. ...... . .. ... . ... ...... ..... .... ... .. ... ... . . .. . . . x+π . ... . .. . ..

. .... .. .... .. ....9 To overcome this ambiguity. . .... . . ..15 Figure 3. ... .. . . .. .. . .... ... . . ...... .. . .. . .. .. ....... .. .. ... .. .. C ... .. .14) are related by two rules: The sine and cosine rules 3.. . .. ... . . . .... ............ .. . .. . ... . .. . .. .19) a c cosine rule: b For the sine rule the proof is. .... .... . .......... .... ... ... .. .. .... ... . ... . .... .. . . . ... ... .. .... . . . .... . . .. sin C = c a (3.... .. .. ....... . .. h Proofs a 21=1b21+1c21−12bc1cos1A h h . ..14 (3. . ... .. . . .. ... .... . .... ... using Figure 3.. .. . . . ... ... . ..... .... .... .. . ... . .. . . .. . ... ......... .. ..... ....... ...4 0 Exercises 12. . B .. . .. ... . . .. ... . . . . .. . ..... ....... ..... .. .. ... 13 Angle π26 lies in the ﬁrst quadrant and is the principal value of the inverse function: EXAMPLE 3. .... . . .. . .... . .... . ... . .. .......... .... ..... .......... .. . . . ... .A. .. . ........ .......... .. .. . . ..... . .. ...... ..... .. ... . .... ...... . a principal value has been deﬁned for each inverse function: They are the values computed. . . .. ... . ... . . ..... ... ... .. ..... . . .. . .. . ...... . ... .. . ....... .. ... . ... ... .... . ..... ...... . .. . ....x−b a c . . . . ... . sin A sin B sin C = = a b c Trigonometric relations 3 π cos−1 = 2 6 cos x = cos−1 y x = sin −1 y x = tan −1 y π 13π = cos − = cos = 6 6 6 π − − 2 π 2 π 0≤x≤π ≤x≤ ≤x≤ 2 π 2 π = (3.. . . ........... .4 Trigonometric relations Figure 3.. .. . . . on a pocket calculator or other computer..... ..... .. .. ..... .... . .. .. . .. . .... ... .... . ..... ... ...... . . ...... .15....... . ... ............ . ... .... .. . . . ..... . .. ..... . C ... .... .. .. .... .18) 73 ..... .. .....20) 3 2 x sin A = (quadrants I and IV) (quadrants I and II) (quadrants I and IV) 3.. . .. ... .. ................ ...... .. . . . .... sine rule: The angles and sides of a triangle (Figure 3. . A .. .. .. ..... ......... ..... .. . . . ... .. .. .... ... .. .. . .... .. . ...... .. . . ... ... ... for example....... .. ...

c1=13.10 Given the lengths a1=12.11°. 0 Exercises 14–17 Compound-angle identities The sines and cosines of the sum and difference of two angles are sin(x1+1y)1=1sin1x1cos1y1+1cos1x1sin1y sin(x1−1y)1=1sin1x1cos1y1−1cos1x1sin1y cos(x1+1y)1=1cos1x1cos1y1−1sin1x1sin1y cos(x1−1y)1=1cos1x1cos1y1+1sin1x1sin1y (3.14. Then sin A sin C = a c Similarly for the third angle and side. we use the cosine rule to ﬁnd b: b21=1a 21+1c21−12ac cos1B1=141+191−112 cos1π23 and cos1π231=1122.15 and Pythagoras’ theorem. EXAMPLE 3. from Figure 3. a 21=1h 21+1(b1−1x)21=1h 21+1b21+1x 21−12bx.89°1=179.74 Chapter 3 Transcendental functions so that h1=1c sin A1=1a sin C. a 21=1b 21+1c 21−12bc cos1A so that cos1A1=1(b21+1c 21−1a 2)22bc1 = 2 7 and 2 A = cos−1 ≈ 40. Given two sides and the included angle.22) (3. ﬁnd the third side and the other angles. and the angle B1=1π23 of the triangle in Figure 3. for example. To ﬁnd the other two angles we use the cosine rule for one.21) . Therefore b21=141+191−161=17 and b = 7. For the cosine rule. c 21=1h 21+1x 2 and the rule follows since x1=1c1cos1A.89° 7 Then C1≈1180°1−160°1−140.

23) (3.13 Express sin13θ1sin12θ and cos13θ1cos12θ in terms of cos1θ and cos15θ.26) . sin15θ1=1sin(3θ1+12θ)1=1sin13θ1cos12θ1+1cos13θ1sin12θ sin1θ1=1sin(3θ1−12θ)1=1sin13θ1cos12θ1−1cos13θ1sin12θ 0 Exercises 18–21 EXAMPLE 3. From equations (3. From equations (3.21).4 Trigonometric relations 75 Also.21) it follows that sin x cos y = and. From equations (3. therefore.11 Express sin15θ and sin1θ in terms of the sines and cosines of 2θ and 3θ. sin13θ1sin12θ = cos 3θ cos 2θ = 0 Exercises 23 1 cos θ − cos 5θ 2 1 cos θ + cos 5θ 2 (3. EXAMPLE 3. sin 3θ cos 2θ = 0 Exercises 22 EXAMPLE 3. therefore.25) 1 cos x cos y = cos( x − y ) + cos( x + y ) 2 and.24) where the alternative expressions for cos12x are obtained from sin2 x1+1cos2 x1=11. putting x1=1y.12 Express sin13θ1cos12θ in terms of sin1θ and sin15θ.3.22) it follows that sin1x sin1y = 1 cos( x − y ) − cos( x + y ) 2 1 sin 5θ + sin θ 2 1 sin( x + y ) + sin( x − y ) 2 (3. sin12x1=121sin1x1cos1x cos12x1=1cos x1−1sin x1 =111−121sin21x1=121cos21x1−11 2 2 (3.

EXAMPLE 3.5).21). using equations (3.15 The harmonic wave travelling in the x-direction described in Example 3.22).14 Express sin ± θ and cos ± θ in terms of sin1θ and cos1θ. Therefore π sin ± θ = cos θ 2 Similarly.76 Chapter 3 Transcendental functions π π EXAMPLE 3.11 shows that sin π 2 = 1 and cos π 2 = 0 (see also Example 3. 2 2 From equations (3. π π π cos ± θ = cos cos θ ∓ sin sin θ = ∓ sin θ 2 2 2 0 Exercises 24 The expressions for the sum and difference of angles are important for the calculation of integrals (see Chapter 6) and for the description of the combination (interference) of waves.12 has wave function φ+ = A sin 2 π x − νt λ The same wave travelling in the opposite direction is (replacing t by −t) φ− = A sin 2 π x + νt λ As the waves overlap they interfere to give a new wave whose wave function is a linear combination of the form ψ1=1aφ+1+1bφ− = a A sin 2 π x x − ν t + b A sin 2 π + ν t λ λ .7 and shown in Figure 3. π π π sin ± θ = sin cos θ ± cos sin θ 2 2 2 Figure 3.

.21) for the sines of the sum and difference of angles Polar coordinates ψ = 2 A sin ψ = a A sin = ( a + b) A sin 2π x 2π x cos 2 πν t + A cos sin 2 πν t + b A sin λ λ 2π x λ 2π x λ cos 2 πν t 2π x cos 2 πν t − A cos λ cos 2 πν t − ( a − b) A cos Figure 3.... .. ..... . . . ... .. . .. . .... .. . The wave oscillates with frequency ν.. . . . . . .. .. ... ... .. . ... .. . . . . . . .. . . . ... .... ..... . . . . .... . .. ... . . . . .. . ...... .... ......... . . . ... .. .. .. . . . .. ...... that is. . . .. .. . . .... ...... . .... .. ... . .. . . ... .. .. .... .... . the amplitude has doubled but there has been no change in wavelength. .... .. ... .. .17... ...... ...There is constructive interference.... .. .... ... . ......... . .... . .. . λ A2 ψ λ 3.. . ... The motion in time is give by the periodic function cos12πν t... . .. .. .. ..16 2π x sin 2 πν t λ 2π x λ sin 2 πν t 3. . . .. ...... ... . .. .... . . as shown in Figure 3. . .... . .. This is called a standing wave.. .. . ... .. . .. ... . . . ... . . ..... . .. .. . ... . . .. .. . . .. .. .. .. ...... ... . .. .. .. .. ..... .... .... .. . . ..... ... .. .... Its shape is given by the x-dependent factor 2π x 2 A sin ... .. . ... . .... . . . ... .. . . ....... . ... .. . .. . . ... . . .... . . .. . ... .. ... .. . .. . . . . . ... ... ...... . . .. .. ... . . . . . . .. .. . . ... . ... .. . . .... . . ... . ... . .. . .. . ... . .... . . ... . ... . . .. ... .. . . . . . .. .. .. .. .. ..... . ... . . .. .. ... The position of a point in a plane can be speciﬁed by its coordinates with respect to a given frame of reference (see Section 2.. .... .. . .. . . . . . ...... .... . .5 Polar coordinates 77 .. . ..... . .. . . . .. .. .... ... . . . .. ... ..... . . .. . . ... ... . .. .. ... ..... . . ... .. ... .. . . . . . . . .. . . . .. . ... .... ... . .. . .. . .. . ... .. .. . . . .. .. . . . ... . 26 An important special case is obtained for a1=1b1=11: Making use of the expressions (3........... .... ... ........ .. . .... . ... ... . .. .. ....... .. . ... . . . . ...... .. ...... .. . . . . . . . . . . . .. . ... . ...... .. ..... .. .. ..... .. .. .... .. . . .. .. .. . . . . .... .. .... .. ...... . .. . . . .. . . .. . .. ... .. .. ... . .. . .. ... . ..... . .... . ... .. . ..... ... . .. .. .. ... .. . .. .. . . . . .... ...... .. . . .. .. . . . ...... ...... . . ...... .. . . . .. .. the wave is stationary in space. . .. . ... . .... . .. . . . . .... ..5 0 Exercises 25. .16. . ...... .... .. . .. .. .. .. . ..... ... .. . ... .2).. .. .. . . ........ .. . . ... . .. . .. . .. ... . ... . . .. ... ... but the positions of the nodes (zeros) of the wave do not move.. . .. A2− x • • • •o edon yranoitats . . . ........... . ... .. ..... .... as shown in Figure 3.. . . . . .. . . .... . ..... . . . . ..... ..... .. . ... . . .. . .

.. Making use of the values displayed in Table 3.. ... . . . .. .. The numbers r and θ are called the polar coordinates of the point in the plane. . .. . . .. . . . . ... . .. . .. .. . . . .. .. . .... . .78 Chapter 3 Transcendental functions Figure 3.. . . and some care needs to be taken in 4 In his Methodus ﬂuxionum (Method of ﬂuxions). . . .. . . .. x . . .. .. . r is necessarily positive.. .. .. ... .. . .. .. ... . .. . . .... .. written in about 1671. .2.. .. . . .. ... . θ). .. ... . . ...... . . . . .. y O ... . .... . . .. . . . . for spirals’.. x = r cos θ = 2 cos y = r sin θ = 2 sin 0 Exercises 27. . .. ... . . . . . . ... .17 The ﬁgure shows that the position of the point P can be speciﬁed not only by its cartesian coordinates (x. . . . . . ... . because tan1θ is a periodic function. .. . . .. . period π... .. . . . . . y). . .. .. .. the angle θ is not uniquely determined by the inverse tangent tan−1( y2x).. .. so that x1=1r1cos1θ y1=1r1sin1θ (3. . .... . . The polar coordinates were his ‘seventh manner. .. .... .... .. .. . .. . . . .. . . . . the distance r of the point from the origin and the angle θ which gives the orientation of the line OP with respect to the positive x-direction (with the usual convention about sign).. y .. . .....27) and the conversion from polar to cartesian coordinates is straightforward.. . x r θ .. . . EXAMPLE 3.. . . ... . .. . .. ... . ... . .. ..... ... . . ... .. .. . . . P •... The two sets of coordinates are related by cos1θ1=1x2r and sin1θ1=1y2r. . 28 π 6 = 3 π 6 =1 The polar coordinates can be obtained from the pair of equations r 2 = x2 + y2 tan θ = y x However. . . .. . . . .. . . . . .. . . .. . .. . . . but also by the pair of numbers (r.... and the angle θ. . . .. .. .. . . . . . . . has values θ1=10 to 2π.. .16 Find the cartesian coordinates of a point whose polar coordinates are r1=12 and θ1=1π26.. .. ... .4 Whereas the cartesian coordinates can take any real positive and negative values... . Newton suggested eight new types of coordinate system. . . .. . . with values r1=10 to ∞. ...

and the angle is the principal value of the inverse tangent. 0 Exercises 29 EXAMPLE 3.1θ) of the point whose cartesian coordinates are (x. the point lies in the second or third quadrant. y θ = tan −1 x y θ = tan −1 + π x when x > 0 (3. 0 Exercises 30 0 Exercise 31 . θ1=1tan−1( y2x). 4) lies in the ﬁrst quadrant. But the point (−1. r = + x2 + y2 . and the angle is θ1=1[principal value1+1π].6). We have r 2 = x 2 + y 2 = 5. We have r 21=1x 21+1y 21=125. y tan θ = = −2.28) when x < 0 EXAMPLE 3. y)1=1(−1. x r=+ 5 θ = tan −1 ( −2) + π Use of the inverse tangent facility of a pocket calculator gives the principal value. and the correct angle is tan−1 (−2)1+1π1≈1−63°1+1180°1=1117° with the same tangent value. 2). The correct value of θ is determined by the quadrant in which the point (x. r1=15 θ1=1tan−1(423)1≈153° The point (3. y) lies.3. Therefore.17 Find the polar coordinates (r. and the angle is the principal value. When x1>10.5 Polar coordinates 79 converting from cartesian coordinates to polar coordinates. the point lies in the ﬁrst or fourth quadrant (see Figure 3.18 Find the polar coordinates (r. where 90°1<1θ1 <1180°. When x1<10. y)1=1(3. 4). tan−1(−2)1≈1−63°. 2) lies in the second quadrant. tan1θ1=1y2x1=1423. θ) of the point whose cartesian coordinates are (x.

the study of inﬁnite processes.021+10.1.000121−10. is exp (x)1=1e x in which the base e is the Euler number (see Section 1.31) The value of the exponential function can be computed from the series to any desired accuracy.19 Calculate exp (−2x2) to 8 signiﬁcant ﬁgures for x1=10. the exponential function. The exponential function that occurs in statistics and in the physical sciences.31×110−101+1- 5 In his inﬂuential textbook Introductio in analysin inﬁnitorum of 1748. The value of the exponential can be obtained directly by substituting (−0. and showed them to be inverse functions. The best known example is the function 10 x which. exp( −2 x 2 ) = 1 + ( −2 x 2 ) ( −2 x 2 )2 ( −2 x 2 )3 ( −2 x 2 )4 ( −2 x 2 )5 + + + + + 1! 2! 3! 4! 5! 4 6 2 8 4 10 x + x − x + 3 3 15 = 1 − 2x2 + 2x4 − Substituting for x1=10.02)1≈111−10. In this book Euler laid the foundation for the branch of mathematics called analysis. as in inﬁnite series and in the calculus. when x is an integer. Euler deﬁned the exponential and logarithmic functions in terms of inﬁnite series.02) for x in (3. Alternatively.4). .10 for x1=11). exp (−0.80 Chapter 3 Transcendental functions 3. although the number of terms required increases rapidly as | x| increases (see Example 1.0001000100617 −10.1.29) in which the number a is the base of the function and the variable x is the exponent.6 The exponential function An exponential function has the form f(x)1=1a x (3.31). and holding 2 signiﬁcant ﬁgures more than the required 8 for intermediate values. is used for the representation of numbers in the decimal system. EXAMPLE 3.00010011333131+10. 0 Exercise 32 The function is deﬁned by the inﬁnite series (see Chapter 7)5 ex = ∑ ∞ (3.30) n= 0 xn x x 2 x3 x 4 = 1+ + + + + n! 1! 2! 3! 4! (3.

..... . .... . . . .. . .... . . . .. . . . . . ... . . ... . . .. . . .. . .. . . .. .... a x1→10 as x1→1−∞ Figure 3. . .... .. . ... . . ... ... .. a x1→1∞ as x1→1∞. . .. . . . . .. ..... . . .. . . . ..... .. .. .. . ... . . . . ... ... ..33) (3... . ... . .. . and statistical mechanics. . . . .. . . .. . .. . . ... . . .. .... . . . . . .. . . .. .. . .. .. . . . .... . . ... . . .. . .. . . .. ... . . ... .. .. . . . .. . . .10): 0 Exercises 35 The exponential function occurs in nearly all branches of applied mathematics... ...... . . . . .. . . . . . .. . .. . .. and the ﬁrst ﬁve are sufﬁcient (compare Example 1.. . .. . . ..... .. . . . . ... . ... ... .... . . . .. .. .. ... .. . . . . .... . .9801198167 dx = ± kx (t ) dt a01=11... .. . .. . 1 ... . . ... . . .. . . ... . . . . .. .. . . . . It will be seen in Chapter 4 that the unique property of e x is that the slope of its graph at any point is equal to the value of the function at that point: exp (−0. . .. . . ... .. . . . . .. .. .. . . . . . . ...... . .20 Exponential growth and decay o . . . . . .. .. . .. . . kinetics.. . .. . ... .. . ... . .. If x(t) is the size at time t then (see Chapter 4) the rate of change of x is . . ...... .. ... . . . .. . .... . . . . .. including statistics. .. .. 34 The terms are decreasing rapidly in magnitude.. . ..18... . . . . . . . . .. .. .. . .. .. . .. . .. . . .. .. . ... .. . .. . . ... . . ... . ... . . . . .. . . .02)1≈10. .. . . . . quantum mechanics. . .. ... . . .. .. .. .. . . . .. . ..... . . . . . . .. ... ... . . . .. . .. . ..... with properties (for all a1>10) Exponential growth arises when the rate of growth of a system at any time is proportional to the size of the system at that time.. . . ..... . . . ... . .. . ... . . . . . .. ..32) 81 ... . .. ... electromagnetic theory. ....... .... . .. . .6 The exponential function (3. . . .. .. .. . . . . .. ... . . . . .. .... . . . . .The graphs of e x and its reciprocal e−x are shown in Figure 3... . . x−e 0 Exercises 33. . ... . .. . EXAMPLE 3. ... .. . . .. . ... .. . . . .. . .. . . . . . . . . .. The graphs of all the exponential functions a x are very similar. . . .. .18 de x = ex dx 3. . xe . . .. .... .

for example. z)e−ar where (x. consider a system whose size x is doubled after every time interval τ. The function f (x. Starting with size x1=1x0. and so on. y. All the orbitals for the hydrogen atom have the form ψ1=1f (x.82 Chapter 3 Transcendental functions with the1+1sign for growth and the1−1sign for decay. the size after time τ is 2x0. for example. As an example. after time 2τ it is 4x0. sample analysis for quality control in the manufacturing industry.21 Atomic orbitals The 1s orbital for an electron in the ground state of the hydrogen atom is ψ1=1e−r where r is the distance of the electron from the nucleus. EXAMPLE 3. and z. and a is a constant. 0 Exercise 36 . EXAMPLE 3. f 1=1z gives a pz orbital.8). After time t. after time 3τ it is 8x0. where ln12 is the natural logarithm of the number 2 (see Section 3.22 The normal distribution The normal or Gaussian distribution in statistics is described by the probability density function 2 1 − 1 x − µ p( x ) = exp σ 2π 2 σ where µ is the mean and σ is the standard deviation of the distribution (see Section 21. y. error analysis of the results of experiments in the physical sciences. x1=12t2τ x0 Equating this with the solution of the differential equation shows that the rate constant k is inversely proportional to the time interval τ : k1=1(ln12)2τ. sample analysis in population studies.7). and determines the shape of the orbital. The probability function forms the basis for the statistical analysis of a wide range of phenomena. z) is a polynomial in x. The proportionality factor k is called the rate constant. The solution of the differential equation is x(t)1=1x0e ± kt where x0 is the size at time t1=10. z) are the cartesian coordinates of the electron relative to the nucleus at the origin. y. y.

. .. . . . . y1=110 x.23 log2 231=13. .. .. ..36) that y1=1loga a y1=1log10 10 y1=1ln1e y (3. . . . ... . .. .. professor of geometry at Oxford. .. ... . .. .. . . Figure 3.. . .. .. ... . . . ... .. .. . .. ... Napier’s logarithms were based on a logarithm of 1071=10. . .. .. .. . . . . . .. The ﬁrst table of common logarithms... . . . .. . . .. ... . . .. . . . was published. . . .... . .3. ... .. . . ... ... . ... .. . . .. . ... ... ... .. ... .... .. . .. ... . . ... ... .. by Henry Briggs (1561–1630)..7 . . y1=1e x. . .. .. .. . Logarithms greatly simpliﬁed computations involving multiplication and division.. The graph of ln1x and of its inverse function e x are shown in Figure 3. . . .. . . . . . . . ...... . The most important logarithmic functions are the ordinary logarithm.. ... 0 Exercises 37 log10 1031=13. . . ... .. . . ... ... . . .. . .. ... . . .. .. . . . . . . . ... .. . ..19... . ... .. .... Scottish baron and amateur mathematician... . . . to base e. . . .. .. .... .. .. . .. . . . ... . . .. . .. . . . . . . . ... .. ... . . . . . . . ... .. ... . . .. . . . . .. . . . .37) EXAMPLE 3.. . .. It follows from equations (3. . . . . . . .. .35) and the natural logarithm (sometimes called the Napierian logarithm). . . . . . .. .34) The logarithmic function6 is the inverse function of the exponential: and loga1 y is called the logarithm to base a of y.. . .. . . .... . ... .7 The logarithmic function 83 3. .. . . . . . . .. . . 7 The graph of a log function was ﬁrst drawn in 1646 by Evangelista Torricelli (1608–1647)... . .. . x1=1log101y1=1lg y (3. .. . . to base 10. .. . . . .. ... . .. . . . . .. . The natural logarithm loge is nearly always given the symbol ln. .. .. ... .. .. . . . loga a01=1loga 11=10 We note that the logarithm of 1 to any base is zero. . . . . .... .34) to (3... . .. . . ln1e−1221=1−122.19 6 John Napier (1550–1617). . . . .. loge e31=13.. . . .. .36) The ordinary logarithm log10 is sometimes given the symbol lg. ..... . . .. . . . .. .. lg110−21=1−2.. . . . ...... . ...... ... . . x nl xe 1 1 .. . .7 The logarithmic function if y1=1a x then x1=1loga1y (3. ... . x1=1loge 1y1=1ln1y (3... .. ... . .. . .. in the Arithmetica logarithmica in 1624. . . ...... .. .. . . .... .. .. ... with log111=10 and log1101=11. . ..... . . . ... . . . .. .. .. ... . . published his invention of what he called logarithms in the Miriﬁci logarithmorum canonis descriptio (A description of the wonderful canon of logarithms) in 1614.. .. . . ... . . ... . .. ... . .. after a famous consultation with Napier. . ... .. . .. .. . . . . . . . . .......

so that xy1=1ea+b Then. by deﬁnition (3. EXAMPLE 3. log1x1→1+∞ as x1→1∞.39). From the rule ln1x n1=1n1ln1x. The combination properties of logarithms are the properties of the indices (exponents) of the corresponding exponentials. for the natural logarithm: ln1x1+1ln1y1=1ln1xy ln x − ln y = ln ln1xn1=1n1ln1x x y (3. log1x1→1−∞ as x → 0 (3.39) (3.26 Simplify the expression ln1(11−1x2)1+1ln1(11+1x)−11−1ln1(11−1x).25 Combinations of logarithms.36) ln1x1=1a.38) We note that (real) log1x is not deﬁned for negative values of x. it follows that ln1(11+1x)−11=1−ln1(11+1x). ln1xy1=1a1+1b Therefore ln1x1+1ln1y1=1ln1xy. Then .41) EXAMPLE 3.40) (3. with properties log111=10. ln1y1=1b. let x1=1ea and y1=1eb.84 Chapter 3 Transcendental functions The graphs of all the logarithmic functions are similar.24 Addition of logarithms. To prove equation (3. ln 2 + ln 4 = ln ( 2 × 4) = ln 8 6 ln 6 − ln 3 = ln = ln 2 3 ln 1 = ln 2−1 = − ln 2 2 ln 30 = ln ( 2 × 3 × 5) = ln 2 + ln 3 + ln 5 ln 23 = 3 ln 2 = ln 2 + ln 2 + ln 2 ln 1 = ln 1 − ln 2 = 0 − ln 2 = − ln 2 2 1 25 1 ln 27 = ln ( 27 )1 3 = ln 3 3 0 Exercises 38 − 2 ln 5 = ln 5−2 = ln EXAMPLE 3.

3.7 The logarithmic function

85

ln (1 − x 2 ) + ln (1 + x )−1 − ln (1 − x ) = ln (1 − x 2 ) − ln (1 + x ) − ln (1 − x ) = ln 1 − x2 (1 − x 2 ) = ln 1 − x2 (1 + x )(1 − x )

= ln1 = 0 0 Exercises 39 EXAMPLE 3.27 What not to do. A surprisingly common error is to put ln1(x1+1y)1=1ln1x1+1ln1y This is not in general true. For example, ln1(11+12)1=1ln13 but ln111+1ln121=1ln12 (ln111=10)

The only case for which ln1(x1+1y)1=1ln1x1+1ln1y is when x1+1y1=1xy; that is, when x1=1y2( y1−11).

Before the invention of the microchip and of the pocket calculator in the early 1970’s, the ordinary logarithm was used mainly as an aid to long multiplication and division; for example, the multiplication of numbers can be replaced by the addition of their logarithms. There are now only a few uses of log10 in the physical sciences; for example in the deﬁnitions of pH as a measure of hydrogen-ion concentration, and of pK where K is an equilibrium constant.

EXAMPLE 3.28 pH as a measure of hydrogen-ion concentration The pH of an aqueous solution is deﬁned as pH1=1−log10[H+] where [H+] is the ‘hydrogen-ion concentration’ in units of mol dm−3 (moles per litre). Then [H+]1=110−pH mol dm−3 For example, a pH of 7 (neutral) corresponds to [H+]1=110−7 mol dm−3.

86

Chapter 3 Transcendental functions

The pH is an example of the use of the logarithm as a scale of measure. Other examples of logarithmic scales are the Richter scale for the ‘strength of earthquake’, the bel scale of loudness, and the scale of star magnitudes. 0 Exercises 40–42

Conversion factors

The logarithm occurs in a number of expressions for physical properties and processes, and the ordinary logarithm is still sometimes found, instead of the natural logarithm, in the scientiﬁc literature and in textbooks. In general, the conversion factor from one base, a, to another, b, is given by logb x1=1logb a1×1loga x Thus, if x1=1a y, then loga x1=1y logb x1=1logb a y1=1y1logb a1=1(loga x)1×1(logb a) The conversion factors between ordinary and natural logarithms are therefore lg1x1=1log10 x1=1log10 e1×1loge x1≈10.434294481×1loge x ln1x1=1loge x1=1loge 101×1log10 x1≈12.302580931×1log10 x (3.42)

3.8

Values of exponential and logarithmic functions

**Table 3.3 shows values of ln1x, x1ln1x, e x, and e−x for a wide range of values of x. Table 3.3 x
**

0

ln1x

–∞ slowly

x1ln1x

0

ex

1

e−x

1

10−6 10−3 1 10 102 103

−13.8 −6.9 0 2.3 4.6 6.9 slowly

−0.00001 −0.007 0 23 460 6908

1.000001 1.001 2.7 21×1104 31×11043

0.9999990 0.9990 0.37 51×110−5 41×110−44

fast ∞ ∞

fast 0

∞

∞

3.9 Hyperbolic functions

87

The following conclusions can be drawn from the table. (a) ln1x varies slowly compared to x. In fact it varies more slowly than any power of x: as x1→10, as x1→1∞, ln1x1→1−∞ ln1x1→1∞ but but xa ln1x1→10 x−a ln1x1→10 (3.43) (3.44)

for any positive value of a, however small. (b) e x varies rapidly compared with x. In fact it varies more rapidly than any power of x: as x1→1∞, as x1→1∞, e x11→1∞ e−x11→10 and and x−ae x11→1∞ xae−x11→10 (3.45) (3.46)

for any positive value of a, however large. EXAMPLE 3.29 Plots of xe−x and x−1e x. Figure 3.20 shows that whereas e−x decreases monotonically with increasing value of x, the function xe−x ﬁrst increases and passes through a maximum before the exponential decay takes over. This is a characteristic behaviour of atomic orbitals. A 1s orbital has the form e−r, where r is the distance from the nucleus, but all other orbitals behave with distance like r le−r, where l is a positive integer.

Figure 3.20

Figure 3.21

0 Exercises 43

3.9

Hyperbolic functions

Hyperbolic functions have their origin in geometry in the description of the properties of the hyperbola. The properties of the functions are readily derived from the properties of the other transcendental functions described in this chapter, and only a brief discussion is presented here.

x/ 1

x/ e

................. .... ............ ............... .......... .............. .............. ... .. ....... ...... ..... .... ... ....... ....... . .. ..... ..... ..... .. . . .... .... .... ... .... .. .... .... .. .. .. ... ... .. ..... .. . ... .. .. . .. . .... ... .. ... .... .. . .. .. .. .. .. . .. . .. .. . . .. . .. . .. . . .. . .. . .. . .. . ... . . . . ... .. .............. . ................ . .. ... ... .. .. . .. ... . ... . .... .. . ..... ... . . . . . . . . . .... . . .... .. . .. . ... .. . .. . ... ... .. . . . .. . . ... . . .. ... . . .. . . . .. .. .. .. . .. . . .. .. . . .. . . .. .. . . . .. . . .. .. . . . . . .. .. .. . . . . . .. . .. .. . . . .. .. . .. . .. . .. . . . .. .. .. . .. .. . .. . .. .. . . . .. .. . .. .. . .. .. . . .. . . .. . .. . . .. .. . .. .. .. . .. . . . . .. . . . . .. .. .. . . .. . . .. .. . .. . . .. .. .. . . .. . .. .. . .. .. . .. .. . .. .. . .. . .. . .. . .. . . .

x

x

e

O

1

. . . . . . ........... ... ....... ..... . .. . . ........ ........ . .... . . .. . . . . . ...... ..... . .. . . .. .. . .. ..... ....... ....... .. ..... . ..... ...... ...... .. ..... .. . .. ...... .. . .. .. . ... .. ...... ..... .. ... .... ... .. .. . . . .. . . ..... ..... ... . .... ... .. .. . .. . ... ...... ...... ... .. .. ... . .. . .. . .. . .. .. ... .. ....... .... ....... .... . .... . ... ..................... . ... .. . .................. . . .... .. .. ... .. .. .. .. .. . . .. . . . . .. .. .. . . .. . .. . . . .. . ... .. . .. . .. . . . . .. . ... . .. . . .. .. .. . . . . . .. . .. . . . .. . . . . .. . . . . . . . .

x−ex

x ........... x−e............... 1 . . .

O

88

Chapter 3 Transcendental functions

The hyperbolic cosine and sine are deﬁned in terms of the exponential function as cosh x = 1 x 1 x −x −x e + e , sinh x = e − e 2 2 (3.47)

**(read as ‘cosh’ and ‘shine’). Their graphs are shown in Figure 3.22.
**

. ........ .

. .. . .. ... .... .... . . .. .. .. ... ..... . . . .. .. .. . . . .. .. .. . . . .. ...... .. . .. . . . . .. ...... .. . . . . . ........ . . . . ........ . . . . . ...... . . . . . . . . .. . . . . . . . . . . . . . . . .

Figure 3.22

Both the notation and the properties of the hyperbolic functions parallel those of the trigonometric (circular) functions; for example, cosh2 x1−1sinh2 x1=11 sinh(x1±1y)1=1sinh1x1cosh1y1±1cosh1x1sinh1y cosh(x1±1y)1=1cosh1x1cosh1y1±1sinh1x1sinh1y The functions are called ‘hyperbolic’ because, if x1=1a1cosh1t and y1=1a sinh1t are the coordinates of a point, where t is a parameter, then by the ﬁrst equation (3.48), x 21−1y 21=1a 2 and this is the equation of a hyperbola.

The inverse hyperbolic functions

The inverse hyperbolic functions are deﬁned in the same way as the inverse trigonometric functions; for example, if x1=1cosh1y then y1=1cosh−1 x. These functions can be expressed in terms of the logarithmic function:

cosh −1 x = ln x ± x 2 − 1 sinh −1 x = ln x + x 2 + 1

tanh −1 x =

1 1 + x ln 2 1 − x

. . . . . . . . . . .... . . . . . ........ . . . . . . . ........ . . . . . . ........ .. . . . . . ........ . .. . . . ......... . .. . . . . ........ . . .. . . .... ..... . . .. .. .. . . . . .. .. . . . ..... .... . .. ... . .. . . . . .... .... . . . . .. . . . .. . . ... .... ... . . .. . .. .. . . . .. .. .. ... ... ... .. . . . ... . . . . . . . . . . . ... ... ... . .. .. . . .. . .. . . .. . .. .. . . . . ... ... ...

( x ≥ 1) (3.49) (| x | < 1)

. ........

. . .......

xe

x hnis

O

x−e

−

xe

. . . . . . .. .. . . . .. . . . .. . . .. . .. .. . .. . . . .. . .. . . . .. .. .. .. .. . .. . .. . . . . . .. ... .. . .. ... .. . . .. . . . . .. .. .. .. . .. .. .. .. . . . . .... .... . . ...... . . . . . . . . . . . . . .. .. .. .. .. .. .. . . . ... . . . . . . ... .... . . . . . . ... ... ... . . . . . . . . . . . . ... ... . . . . ..... .. . . . . . .. . . . ...... . . . .. ..... .. . . . . . . . .. . . ......... . . .. . . . . . ... ... . . . ........ . ... . . . . ...... . . . . . . ........ . .. . . ... . . . . ......... . .. . . ... . . .... . . . . . . . . . .... . . . . . . . . . . . . . . . . . . . .

x hsoc

O

x −e

(3.48)

3.10 Exercises

89

The relation for cosh−1 x shows that there exist two values of the function for each value of x (>11). These two values differ only in sign, and the positive value is deﬁned as the principal value.

**EXAMPLE 3.30 To show that cosh −1 x = ln x ± x 2 − 1 .
**

**If x1=1cosh1y then, because cosh21y1−1sinh21y1=11, it follows that sinh y = ± x 2 − 1 and ln x ± x 2 − 1 = ln cosh y + sinh y = ln e y
**

from the deﬁnitions (3.47). The result follows since ln1e y1=1y1=1cosh−1 x.

3.10

Exercises

Section 3.2

1. In Figure 3.23, the right-angled triangle ABC has sides a1=112 and b1=15. Find c and the sin, cos, tan, cosec, sec and cot of the internal angles A and B. 2. For the triangle in Exercise 1, ﬁnd (i) sin2 A1+1cos2 A, (ii) sin2 B1+1cos2 B. 3. Express the following angles in radians: (i) 5° (ii) 87° (iii) 120° (iv) 260° (v) 540° Figure 3.23 (vi) 720° 4. Express the following angles in degrees: (ii) π24 (iii) π26 (iv) π23 (v) 3π28 (vi) 7π28 (i) π210 5. For a circle of radius r1=14, ﬁnd (i) the angle subtended at the centre of the circle by arc of length 6, (ii) the length of arc that subtends angle π210 at the centre of the circle, (iii) the length of arc that subtends angle π22 at the centre of the circle, (iv) the circumference of the circle. 6. Use Table 3.2 to ﬁnd the sine, cosine and tangent of (i) 3π24, (ii) 5π24, (iii) 7π24. 7. By considering the limit θ1→10 of an internal angle of a right-angled triangle, show that (i) sin101=10 (ii) cos101=11 8. Use the properties of the right-angled isosceles triangle to verify the values of the trigonometric functions for θ1=1π24 in Table 3.2. 9. Sketch diagrams to show that (ii) cos(π1−1θ)1=1−cos θ (iii) sin(π1+1θ)1=1−sin θ (i) sin(π1−1θ)1=1sin1θ (iv) cos(π1+1θ)1=1−cos θ (v) sin(π221−1θ)1=1cos θ (vi) cos(π221−1θ)1=1sin θ 10. Find the period and sketch the graph (−π1≤1x1≤12π) of (i) sin12x, (ii) cos13x. 11. Sketch the graph of the harmonic wave φ(x, t)1=1sin12π(x1−1t) as a function of x (−11≤1x1≤12) for values of time t, (i) t1=10, (ii) t1=1124, (iii) t1=1122.

a

B C

................... .................. . . . .................. . . . . . .. . .. . . . .. . .. . . .. . . .. ... . . .. . .. . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . .. .. . . .. . ... . .. . . ... . ... . .. .. ... .. .. . . . . . ..

b

c

A

90

Chapter 3 Transcendental functions

Section 3.3

12. Find the principal values of

−1 (i) sin

()

1 2

(ii) sin−1(1)

−1 (iii) cos

()

1 2

(iv) cos−1(−1)

13. The Bragg equation for the reﬂection of radiation of wavelength λ from the planes of a crystal is nλ1=12d1sin θ where d is the separation of the planes, θ is the angle of incidence of the radiation, and n is an integer. Calculate the angles θ at which X-rays of wavelength 1.51×110−10 m are reﬂected by planes separated by 3.01×110−10 m.

Section 3.4

.... . .. . . .. . . . . . .. . ............................................................. ..............C............. ........... ........... ........ ..... .... ........... ........... ........... ............. . ... .... ... ... .. A ... . .. .. .. . ... . .. . . . . . . .. ... ... ... .. . .. . . . . .. ... ... .. .. . . .. . . . . .. ... ... .. .. . .. .. . . . . .. ... .. .. .. ... . .. ... .. . .. .. B . . .. .. . . .. . . .. ... ... ... .. . ... .. ... .. . .. ... ........ ... ......... ... .. ..... . . . ... . .. ... .. .. . . ... .. .... .... .

14. Given the side a1=11 and angles A1=1π24 and B1=1π23 of a triangle ABC, Figure 3.24, ﬁnd the third angle and the other two sides. 15. Given the sides a1=12, b1=12.5 and c1=13 of a triangle ABC, ﬁnd the angles. 16. Given the sides a1=13, b1=14, and included angle Figure 3.24 C1=1π24 of triangle ABC, ﬁnd the third side and the other two angles. 17. Given the sides a = 2 , b = 3, and included angle C1=1π24 of the triangle ABC, ﬁnd the third side and the other two angles. 18. Express in terms of in terms of the sines and cosines of 2θ and 5θ: (ii) sin13θ, (iii) cos17θ, (iv) cos13θ. (i) sin17θ, 19. Express (i) sin13θ in terms of sin1θ, (ii) cos13θ in terms of cos1θ. 20. Express cos14x in terms of (i) sin12x and cos12x, (ii) sin12x only, (iii) cos12x only, (iv) sin1x only, (v) cos1x only. 21. Given sin110°1=10.1736, sin130°1=1122, sin150°1=10.7660, ﬁnd cos120° (without using a calculator). 22. Express in terms of the sines of 8x and 2x: (i) sin15x1cos13x, (ii) cos15x1sin13x. 23. Express in terms of the cosines of 8x and 2x: (i) sin15x1sin13x, (ii) cos15x1cos13x. 24. Express (i) sin(π1±1θ) and (ii) cos (π1±1θ) in terms of sin1θ and cos1θ. 25. The function ψ (x, t)1=1sin1π x1cos12π t represents a standing wave. Find the values of time t for which ψ has (i) maximum amplitude, (ii) zero amplitude. (iii) Sketch the wave function between x1=10 and x1=13 at (a) t1=10, (b) t1=1128. 26. The function

φ ( x ) = a sin

2π x

λ

+ b cos

2π x

λ

represents the superposition of two harmonic waves with the same wavelength λ. Show that φ is (i) also harmonic with the same wavelength, and (ii) can be written as

φ ( x ) = A sin

2π x λ

+ α

where A = a 2 + b2 and tan1α1=1b2a.

Section 3.5

27. Find the cartesian coordinates of the points whose polar coordinates are (ii) r1=13, θ1=15π23. (i) r1=13, θ1=1π23,

a

b

c

3.10 Exercises

91

28. Find the cartesian coordinates of the points whose polar coordinates are (ii) r1=13, θ1=14π23. (i) r1=13, θ1=12π23, 29. Find the polar coordinates of the points whose cartesian coordinates are (i) (3, 2), (ii) (3, −2). 30. Find the polar coordinates of the points whose cartesian coordinates are (i) (−3, 2), (ii) (−3, −2). 31. A solution of the equation of motion for the harmonic oscillator is given in Example 3.8 as x(t)1=1A1cos1ω t. Show that x(t) can be interpreted as the x-coordinate of a point moving with constant angular speed ω in a circle in the xy-plane, with centre at the origin and radius A.

Section 3.6

32. Simplify (ii) e3e−3 (iii) e3e−4 (iv) e32e2 (v) e52e−4 (i) e2e3 −x23 33. (i) Write down the expansion of e in powers of x to terms in x5. (ii) Use the expansion to calculate an approximate value of e−123. Determine how many signiﬁcant ﬁgures of this value are correct, and quote your answer to this number of ﬁgures. 3 34. (i) Write down the expansion of e− x in powers of x to terms in x15. 3 Use the expansion to calculate an approximate value of e− x that is correct to 12 signiﬁcant ﬁgures for the following values of x, in each case giving the smallest number of terms required: (ii) 10−1, (iii) 10−2, (iv) 10−3, (v) 10−4, (vi) 10−5. 35. Sketch the graphs of e2x and e−2x for values −1.51≤1x1≤11.5. 36. For a system composed of N identical molecules, the Boltzmann distribution ni − ε / kT =e i N gives the average fraction of molecules in the molecular state i with energy εi. (i) Show that the ratio ni 2nj of the populations of states i and j depends only on the difference in energy of the two states. (ii) What is the ratio for two states with the same energy (degenerate states)?

Section 3.7

37. Simplify: (i) log10 100 (ii) log2 16 (iii) ln1e−5 (iv) ln e x

2

(v) ln e− ( ax

2

+ bx + c )

(vi) ln1e−kt 38. Express the following as the log of a single number: (i) ln121+1ln13 (ii) ln121−1ln13 (iii) 51ln12 (iv) ln131+1ln141−1ln16 39. Simplify: (ii) ln1(2x31−13x2)1+1ln1x−2 (iii) ln1(x51−13x2)1+121ln1x−11−1ln1(x31−13) (i) ln1x31−1ln1x x x 2 +3 3 (v) ln e (iv) ln1e − ln e 40. The barometric formula p1=1p0e−Mgh2RT gives the pressure of a gas of molar mass M at altitude h, when p0 is the pressure at sea level. Express h in terms of the other variables. 41. The chemical potential of a gas at pressure p and temperature T is

µ = µ ο + RT ln

f pο

92

Chapter 3 Transcendental functions

where f 1=1γ p is the fugacity and γ is the fugacity coefﬁcient. Express p as an explicit function of the other variables. 42. In a ﬁrst-order decomposition reaction, A1→1products, the amount of substance A at time t is x(t)1=1x(0)e−kt where x(0) is the initial amount of A, and k is the rate constant. The time taken for the amount of A to fall to half of its initial value is called the half-life, τ122, of the reaction. Find the half-life for rate constants: (i) k1=13 s−1, (ii) k1=110−5 s−1.

Section 3.8

43. As in Example 3.30, sketch (i) x2, e−x, x2e−x, (ii) x−2, ex, x−2ex.

p

P∆ + p

•

P∆

p

O

V∆ + V V∆

T

T∆ + T

T∆

V∆

. . . . . . . . . . . . . . . . . . . . . . .......... ................ . ...... ...... .... . ..... .... . .. ... . ... .. .. .. .. .. .. .. ... .. .. .. .. .... .. .. .. .. .. .. .. .. .. .. .. .. . . . .. . . . . .. . .. . . .. . . ............. .......... ............ ........... . . . . . . . . ......... .......... ............ . . . . ....... ............... . . . . ...... . ........ . . . . . ..... .......... . . . ... ........... . . .. . . . .... . . . . .......... . . . . .. . . . . . .. ............. . . . .. . . .. ............ . . . .... ... . .... .. .. .. .. . . .. . . .. .. .. ..

..

.........

•

V

... ..

........... .

. . . . .......................

. . . . . . .............

. . . . . . . . ....................... . .......... . . . . . . . . . . . . . . .............. . . . . . . ...................... . . . . . . . . . . ....................... . . . ....................... . . . . . . . . .......... . . . . . . . . . . . . .............. . . . . . . . . . . . ....................... . . . . . . ...................... . . . . ........... . . . ... .... . . . .. . . . . .. . . .. .. .. .. .. .. .. .. . . . ........ .. .. .. .. .. .. .. .. . . .. .. . . .. . . . . . .. .. .. .. .. .. .. .. . . ....... ............. .......... . . .. .. .. .. .. .. .. .. .. ............ . . . . . . . . . ........... . . . . . . ...................... . . . . . . . . ....................... . . . ....................... . . . . . . ................ . . . . . . . . . . . . . ....... . . . . . . . . . ..................... . . . . . . . . ....................... . . ... .............. . ... .......... .. . . .... ...... .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . .. .. . . . . .. . . . . .. . . . . . .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. ..

•

. ....

V

Because any one of the four variables p, V, T, and n can be expressed as a function of the other three, the state of the system is determined by three of the four quantities. If the temperature T is changed by an amount ∆T (read as ‘delta t’), keeping the pressure p and the amount of substance n ﬁxed, the volume changes from

Figure 4.1 shows that the graph of V against T (at constant p and n) is a straight line with gradient, or slope,

and the change in volume is

The gradient is the change in V per unit change in T, or the rate of change of V with respect to T.

4.1

In the physical sciences we are interested in the value of a physical quantity and how it is related to other physical quantities. In addition, we are interested in how the value of the physical quantity changes on going from one state of the system to another, and in the rate of change with respect to time or with respect to some other physical quantity. Consider the equation of state of the ideal gas

4 Differentiation

Concepts

Figure 4.1

V=

nRT p

∆ V nR = = constant ∆T p

to

∆V =

pV1=1nRT

V + ∆V =

nR ∆T p

nR(T + ∆ T ) p

Figure 4.2

T

O

V V∆ + V

•

V

94

Chapter 4 Differentiation

If the pressure p of the gas is changed by an amount ∆p at constant T and n, the change in the volume is ∆V = nRT nRT ∆p − = − nRT p + ∆p p p( p + ∆ p )

Figure 4.2 shows that the graph of V against p (at constant T and n) is not a straight line. The gradient at any point on the curve is deﬁned as the gradient of the tangent to the curve at that point. The gradient changes from point to point, and is not given by ∆V2∆p. The branch of mathematics concerned with the determination of gradients and, therefore, with rates of change is the differential calculus.

4.2

The process of differentiation

Let the value of a variable x change continuously from p to q. The difference (q1−1p) is called the change or increment in x. In the differential calculus this change is denoted by* ∆x1=1q1−1p We note that ∆x1>10 if q1>1p, and ∆x1<10 if q1<1p.

..... . ...... ... ... . .

(4.1)

Figure 4.3

Let y1=1f (x) be a function of x that changes continuously and smoothly from point P to point Q (Figure 4.3). The values of y at P and Q are yP1=1f ( p) and yQ1=1f (q). The change ∆y in y corresponding to change ∆x in x is therefore yQ1−1yP1=1∆y1=1f (q)1−1f ( p) The quantity yQ − yP q− p = ∆y ∆x (4.2)

is the gradient of the line PQ, and can be interpreted as the average rate of change of y with respect to x between P and Q.

* Sometimes, if the change in x is supposed to be ‘small’, δ x is used instead of ∆x.

x

y

∆

.......... .

q

. ............. ..............

. . . . . . . . . . . . . . . . . . . . . . . ....................................................... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. .. .. .. .. .. .. .. .. . .. .. . .. . . ...... ........................... .. .. .. .. .. .. .. .. .. .. .. .. .. . .... .............. ... ...... .. .................. ...... . . .......... . .... .. .. .. .. . . . . .. . . . . . .. . .. . . ... . . . . . ................................... . . . . . . . .... .. . . . ....... . . . . . . . . . . ... .. . . . . . . . . . . . . . ........................ . . . . .. .. . . . . ......................... . . . . . . .. .. . . ......................... . . . . . . ................. . . .. . .. . . . ............ . . . . .. .. . . . . ..... . . . . .. .. . . . ................. . . . .. .. . . . ............... . . . .. .. . ................ . . . . . .. . .. . ................ . . . ............... .. .. . ........... . .. . .. .. . . . . . . .............. .. . .. . . .............. .. . ........... . .. . ... . . . .............. . ... .. ... .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . .. .. . ........... .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . . . .. . . . . .. . . . . .. ..

. . ......

•

Q

x

∆

P

p

•

y

Qy P y

a p t a tn e gn t

Q

•

. . . .. . . .. . . . .. . . . . .. . . .. . . . . .. . .. . . . . . . . .. .. . .. . . . .. . . . .. . . . . . .. . .. .. . . . .. .. . .. . . .. .. . . .. .. .. . . . . .. .. . . . . ..... .. .. .. .. . .. . .. .. .. . . . . . . . . . . . . . . ........................................................................................... . . . . . . .. . .. . .. . ....... ................. .. .... ......................................... . . . . . . . . . . . . . . .. . . . . . . .. . . .. . . . . . .. ... . . .... . ........................... .. . ... ........................... ............. .. . . . .. .. .. ... ... . .. . . ................... ....... ................. . . . .. . . . .. .. ............ . . . ... . .. . . . ................. . ... . . . . .. .. .. ...... . . . . . . . . .. .. .. . . . . . ................... .. .. . .. .. . . .. . .. .. . . ...................... . .. .. . . . .. .. .. . . .. .. . .................... . . .. .. . . . .. .. .. .. .. . . . ................. . . .. . .. . . .. .. .. .. . . . . . .. .. .. . . . . .. .. .. .. .. . . . . ............... .. . . . . . .. .... . .. . . . . ... .. . .... .. ..... .. . . . ............... .. . .. . . .. . . .. . ............... .. . . .. . .. . . .. . ............ . . .. . . . . . ...... . .. . . . .. . . ................. . . . . .. . .. . . . . . .. . .............. . . . . . .. . .. . . . . . .. .. ............... . . . . . . . . . . . .. . ............... . .. . . . .. . . . . . . . . . . ............... . .. .. .. .. . .. . . . . ............ . . . ... .. . ... . .. ... . .. . . . . ... .. . ................ . . . ... . .. . . .. .. .. .. . . . . .. ..... . . .. .. . ........ . . .. .. . . . .. . . .. .............. . . .. .. . . . .. . . . ............. . .. . . ... . . . ............... . .. . . ................. . . . . . ..... .. . ......... . . .... . ...... .. .............. ..... . . . .. ............... . . . .. . ..............

•

Q

..............

. .............

.

The change in the function on going from P to Q is therefore

Figure 4.4 shows how the quantity ∆y2∆x changes as the point Q is moved along the curve towards P (as ∆x is decreased in magnitude).

and the corresponding average rate of change is

More generally, if the variable changes by ∆x, from p1=1x to q1=1x1+1∆x, the corresponding change in the function is

0 Exercise 1

and the gradient of the line PQ is

EXAMPLE 4.1 The general quadratic function, y1=1ax 21+1bx1+1c:

At point Q

At point P

∆y = ( 2 ax + b) + a ∆ x ∆x

∆ y = yQ − yP = a( x + ∆ x )2 + b( x + ∆ x ) + c − ax 2 + bx + c

=1(2ax1+1b)∆x1+1a(∆x)2

yQ1=1f (x1+1∆x)1=1a(x1+1∆x)21+1b(x1+1∆x)1+1c

yP1=1f (x)1=1ax 21+1bx1+1c

∆ y f ( x + ∆ x) − f ( x) = ∆x ∆x

∆y1=1f (x1+1∆x)1−1f (x)

Figure 4.4

P

•

4.2 The process of differentiation

(4.4)

(4.3)

(4.5)

95

96

Chapter 4 Differentiation

As Q moves through Q′ towards P, the gradient of the line PQ approaches the gradient of the tangent at P. We express this as ∆y gradient at P = lim Q→ P ∆ x (4.6)

(read as ‘the limit as Q goes to P’). At the same time, the magnitude of ∆x goes to zero, ∆x1→10 as Q1→1P, and the limit can be expressed as1 ∆y gradient at x = lim ∆x→0 ∆x EXAMPLE 4.2 For the quadratic function y1=1ax 21+1bx1+1c in Example 4.1, ∆y = ( 2 ax + b) + a ∆ x so that ∆x

∆x→0 ∆x

(4.7)

∆y lim = 2 ax + b

The limit is a function of x, and it gives the gradient or slope of the curve at each value of x (each point on the curve). 0 Exercises 2, 3 The process of taking the limit in (4.7) is called differentiation. In the differential dy calculus, the limit is denoted by the symbol : dx f ( x + ∆ x) − f ( x) ∆y dy = lim = lim dx ∆ x → 0 ∆ x ∆ x → 0 ∆x (4.8)

(read as ‘dy by dx’).2 It is called the differential coefﬁcient of the function or, simply, dy the derivative of the function. We note that the symbol does not mean the quantity dx ‘dy’ divided by the quantity ‘dx’; the symbol represents the limit, and the taking of the limit as given by the right side of (4.8).

This method of ﬁnding the tangent at a point on a curve is essentially that given by Fermat in his Method of ﬁnding maxima and minima in about 1630. This work marks the beginning of the differential calculus. A method similar to Fermat’s but involving quantities equivalent to ∆x and ∆y was described by Barrow in Lectiones geometriae, published in 1670. In these lectures Isaac Barrow (1630–1677), theologian and professor of geometry at Cambridge, gave a ‘state of the art’ account of inﬁnitesimal methods. The formulation of the method of tangents was included ‘on the advice of a friend’, Newton, who succeeded him in his chair when Barrow became chaplain to Charles II in 1669. 2 The notation is derived from Leibniz’s formulation of the calculus. Gottfried Wilhelm Leibniz (1646–1716), philosopher, diplomat and mathematician, discovered his form of the calculus in the years 1672–1676 whilst on diplomatic service in Paris, where he came under the inﬂuence of the physicist and mathematician Christiaan Huygens, inventor of the pendulum clock. His ﬁrst account of the differential calculus was the Nova methodus pro maximis et minimis, itemque tangentibus (A new method for maxima and minima, and also for tangents), published in 1684.

1

c

. . . . . . . . . . . . . . . . . . . . . . . . .. . .. .. . .. . .. . .. .. . .. .. . .. . . . . . . . .. .. .... .. . . . . .. . . .. ...

•

......... . ........ . . ...... ..... . . . . .. . ........ . . . .. ... ... ... .

. . ...... . . . ..... . . . . ... . . . . . .. ....... . ... . .... . . ..... . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . .. . .. .. . .. . .. . .. .. . .. . .. ..

..... ..... .... . ..... ....

. . . . . . . . . . . . . . . . . . . . . . . .. .. . .. . .. .. . .. . .. .. . .. .. . ..

........

. .........

.......

. ........

........

The concept of differential and other operators is widely used in the physical sciences, and will be discussed in later chapters.

In the discussion of taking the limit in Section 4.2 it was assumed that the function y1=1f (x) is a continuous function of x, and that the limit deﬁned by equation (4.8) exists and is unique. Generally speaking, a function is continuous if its graph is an uninterrupted curve. A given function y1=1f (x) may be tested for continuity at a point, x1 say, by letting the independent variable x move continuously from the right side and from the left side towards the speciﬁed point x1 as shown in Figure 4.5.

Another symbol is Df, by which is implied that the derivative of f is obtained by acting (operating) on f with the differential operator D,

If this holds for every value of x1 in a certain interval a1≤1x11≤1b, then the function is continuous in the interval. Three types of discontinuity are illustrated in Figure 4.6.

If f (x11+1∆x) and f (x11−1∆x) approach the same value f (x1) as ∆x1→10 then the function is said to be continuous at x1, and we write

x

∆ + 1x

1x

x

∆ − 1x

•

. ................

. ...............

. . . ........

. . . . . . . . . . . . . . . . . . . . . ...................... . . . . . . . . . . . . . . . . . . . . . . . . . .............................. . . . . . . . . . . . ........................................... . . . . . . . . . .. . . ............ ....... ................... . . . . . . .. . . . .. . . .. . . . .. .. ......... . . . . . . ..... ...... ... .......... ....... .. .. .. ............ . . . . . . ............................ . . . . . . . . . . . . . . ......................... . . ......... ... .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . . . . . . . . ............... . . . . .................... . . . . . . ................... . . .................. . . . . . ............ . . . ........ . . . .. . . . ...... . ....................... ... .. . . . . . .. . . . . . . . . ................. . . . ......... . . . ...... . . . . ............. . ....

•

.. •

4.3

An alternative symbol for the derivative of the function f (x) is f ′(x):

Continuity

D=

d , dx

x → x1

lim f ( x ) = f ( x1 )

f ′( x ) =

Figure 4.5

Figure 4.6

..... ...... ....... . ..... ........ .. . .. ..... .

b

.. .... .. . . . .. . . .. ...

. ..

. . ...... . . . ..... . . . . ... . . . . ........ . . . .... . ..... . . ...... . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

•

. . ... ..

..... . . .

. ... .. ... .. . . . . .

. . . . . . . . .. . .. .. . .. . .. . .. .. . .. . .. . .. . .. . .. . .. .. . .. . .. . .. . .. . . . . . . .

a

•

. . . .... .... .

. . . . ... .

... .... . . . .

. . ... .... ... . .

. . . ... ... ... ...

. . . . . . . . . . . . . . .. .. . .. .. .. . . . . . . . . . . . . . . . . . . . . . . .. .. . . . . . . . .. . .. .. .. .. .. .. . . . . . . . . . .. ... .. .. .. .. . . . .. . . . .. .. .. ... .. . . . . . . . . . . . . . . . . . . . ..... . . . . . .. ... .. ... .. . . . . ... ... ... .. . . . . ... ... . . . . . . . . . . .. ... . . . . . . . . . . .

Df =

df ( x ) dx

d df f = dx dx

4.3 Continuity

(4.10)

(4.11) (4.9)

97

) 1x( f

98

Chapter 4 Differentiation

Point a. The function has a ﬁnite discontinuity at x1=1a. For example, the function x +1 f ( x) = x −1 if x > 0 if x ≤ 0

is discontinuous at x1=10 where its value is −1. However, if we approach the point x1=10 from the right, f (x) approaches the value +1. Point b. The function has an inﬁnite discontinuity at x1=1b. If we approach this point from the left, the value of the function tends to −∞; if we approach from the right the function tends to +∞. Inﬁnity is not a number, and the function is not deﬁned at x1=1b. An example is 12(x1−11), discontinuous at x1=11. Point c. The function tends to inﬁnity if we approach the point x1=1c from either side; for example, 12x 2 at x1=10. In these three cases, the nature of the discontinuities is obvious from the graphs; they are said to be essential discontinuities. In some cases however, the discontinuity is not obvious from the graph. For example, the function f (x)1=1x2x has constant value equal to 1 for all values of x1≠10, but it is not deﬁned at x1=10 because 020 is indeterminate and has no meaning. Such a discontinuity is said to be removable. If we redeﬁne the function such that f1=1x2x when x1≠10 and f (x)1=11 when x1=10, then the function becomes continuous for every value of x; the discontinuity has been removed with no change to the function except at an isolated point. 0 Exercises 4–6

4.4

Limits

Consider the rational function y= x2 − 4 x−2

which is continuous for all values of x except x1=12. Table 4.1 shows that, whereas both the numerator and the denominator go to zero at x1=12, their ratio approaches the value y1=14 as x1→12 from both sides: x2 − 4 =4 lim x→2 x − 2 Table 4.1 Values of y1=1(x 21−14)2(x1−12)

x y 2.1 4.1 2.01 4.01 2.001 4.001 2.0001 4.0001 = = 1.9999 3.9999 1.999 3.999 1.99 3.99 1.9 3.9

4.4 Limits

99

This is an example of a removable discontinuity; we have x 2 − 4 ( x − 2) ( x + 2) = =x+2 x−2 ( x − 2) if x≠2

and the discontinuity can be removed by redeﬁning the function to have value y1=14 when x1=12. This example is important because taking the limit in differentiation always involves letting the denominator go to zero. 0 Exercises 7–10 The ﬁnding of limits is necessary whenever a quantity becomes indeterminate. In addition to the case 020, the indeterminate forms most commonly met in the physical sciences are ∞2∞ and ∞1−1∞.

EXAMPLE 4.3 2 x 2 + 5 =2 lim 2 x → ∞ x + 3x Both numerator and denominator tend to inﬁnity as x1→1∞, but the ratio remains ﬁnite. Thus, dividing both numerator and denominator by x 2, which is allowed if x1→1∞, we obtain 2x2 + 5 x 2 + 3x = 2 + 5 x2 2 → =2 1+ 3 x 1 as x→∞

0 Exercises 11–13

EXAMPLE 4.4

2 2 1 1 lim 2 x + − 3x − = 10 0 x → 0 x x

Both squared terms tend to inﬁnity as x1→10, but the difference remains ﬁnite. By expanding the squared terms and simplifying, we get 2x + 1 − 3x − x

2

2 1 1 2 1 = 4x + 4 + 2 − 9x − 6 + 2 x x x =1−5x 21+1101→110 as x1→10

2

0 Exercises 14, 15

100

Chapter 4 Differentiation

EXAMPLE 4.5 x→∞ lim ln ( 2 x + 3) − ln ( x − 2) = ln 2

From the properties of the logarithm, 2 x + 3 ln ( 2 x + 3) − ln ( x − 2) = ln → ln 2 as x → ∞ x−2 0 Exercises 16, 17

4.5

**Differentiation from ﬁrst principles
**

∆y f ( x + ∆ x) − f ( x) dy = lim = lim dx ∆ x → 0 ∆ x ∆ x → 0 ∆x

A function is said to be differentiable at a point if the limit in equation (4.8), (4.8)

exists and is unique. A necessary condition for this to be true is that the function be continuous at the point, but not all continuous functions are everywhere differentiable. For example, the function

. ...... ... . . . . . ..... ... .

is continuous at all values of x but Figure 4.7 shows that its Figure 4.7 slope changes abruptly at x1=10, from value –1 when x1<10 to value +1 for x1≥10. The function has a cusp at x1=10 and the derivative is not deﬁned by equation (4.8). In general, a function is differentiable if it is continuous and ‘smooth’, with no essential discontinuities or cusps. For any such function, the taking of the limit in equation (4.8) is called differentiation from ﬁrst principles, and was demonstrated in Examples 4.1 and 4.2 for the general quadratic. All functions can be differentiated in this way. Let y1=1f (x) be a function of x, and let the function change from y to y1+1∆y1=1f (x1+1∆x) when the variable changes from x to x1+1∆x. The steps for differentiating from ﬁrst principles are: (1) subtract y from y1+1∆y to obtain ∆y as a function of x and ∆x, and simplify as much as possible, (2) divide both sides of the equation by ∆x, dy ∆y (3) ﬁnd the limit of as ∆x1→10; this gives the required derivative . dx ∆x 0 Exercises 18, 19

x

. . ..... ... .

. . . ...... ..

. . . .....

. ... . . .

.. ..... ... . .

. ... . . . ...... .. . . . .... .. .

. . . ..... .. ...

...... . . . .

. . .... .. .

. . .. ..... ..

.

if x < 0

. . . . ......

.. . . . .

...... ... . .

)x( f

. . . ...... .. ...

...... .. . . .

. ..... .. . .

. . . ...... .

x f ( x) = | x | = − x

if x ≥ 0

4.5 Differentiation from ﬁrst principles

101

EXAMPLE 4.6 Find

dy 1 from ﬁrst principles for y = . dx x 1 , x y + ∆ y = f ( x + ∆ x) = 1 x + ∆x

(1) y = f ( x ) =

∆ y = f ( x + ∆ x) − f ( x) = −1 ∆y = ∆ x x( x + ∆ x )

1 1 −∆ x − = x + ∆ x x x( x + ∆ x ) ∆ y dy 1 = lim =− 2 dx ∆ x → 0 ∆ x x

(2)

(3)

0 Exercise 20 EXAMPLE 4.7 Differentiate (1) y = x , x.

y +∆y = x +∆x ( x +∆ − x ) × ( x +∆x + x ) x +∆x + x ( x + ∆ x) − x = x +∆x + x ∆x x +∆x + x

∆y = x +∆x − x = =

(making use of the relation (a1−1b)(a1+1b)1=1a 21−1b2). (2) ∆y = ∆x 1 x + ∆x + x (3) ∆y dy 1 = lim = dx ∆ x → 0 ∆ x 2 x

0 Exercise 21 EXAMPLE 4.8 Differentiate e x. (1) y1=1e x, y1+1∆y1=1e x+∆x1=1e x1×1e ∆x

From the deﬁnition of the exponential function as an inﬁnite series, equation (3.31), e ∆x= 1 + ∆x + Therefore, y + ∆ y = ex × ( 1 + ∆ x + = ex + ex (∆ x + ( ∆ x ) 2 ( ∆ x )3 + + 2 6 ) ) ( ∆ x ) 2 ( ∆ x )3 + + 2! 3!

( ∆ x ) 2 ( ∆ x )3 + + 2 6

102

Chapter 4 Differentiation

and ∆ y = ex (∆ x + ( ∆ x ) 2 ( ∆ x )3 + + 2 6 ) (3) )

(2)

∆y ∆ x ( ∆ x )2 = ex (1 + + + ∆x 2 6

dy = ex dx

0 Exercise 22

4.6

Differentiation by rule

Whilst all functions can be differentiated from ﬁrst principles, differentiation is performed in practice by following a set of rules.3 These rules can be proved from ﬁrst principles, as in Examples 4.6 to 4.8. The derivatives of the more important elementary functions are listed in Table 4.2. Table 4.2 Differentiation of elementary functions Type

c1=1constant power of x trigonometric

Function

c xa sin1x cos1x tan1x

Derivative

0 ax a−1 cos1x −sin1x sec21x ex sinh1x cosh1x 12x

exponential hyperbolic logarithmic

e

x

cosh1x sinh1x ln1x

The ﬁrst example in Table 4.2 states that the rate of change of a constant is zero. In general, if a function is independent of a variable x then its derivative with respect to x is zero. Examples of the derivative of a ‘power’ of x, xa, have been given in Examples 4.6 for a1=1−1 and 4.7 for a1=1122. Other examples, using a variety of notations, are given in Examples 4.9.

Many of the rules of differentiation appeared in Leibniz’s 1684 paper on the differential calculus, with the d–notation and the name calculus differentialis for the ﬁnding of tangents. Isaac Newton (1642–1727) developed his ideas on the calculus in the year 1665–1666 (Trinity College, Cambridge, was closed because of the plague); he later maintained that during this time he discovered the binomial theorem, the calculus, the law of gravitation and the nature of colours. He wrote the ﬁrst of three accounts of the calculus in 1669 in De analysi per aequationes numero terminorum inﬁnitas, but published only in 1711. The ﬁrst account to be published appeared in 1687 in Philosophiae naturalis principia mathematica, probably the most inﬂuential scientiﬁc treatise of all time. In the ﬁrst edition of the Principia, Newton acknowledged that Leibniz also had a similar method. By the third edition of 1726, the reference to Leibniz had been deleted when questions of priority had led to a bitter quarrel between supporters of the two men.

3

4.6 Differentiation by rule

103

EXAMPLES 4.9 Differentiating powers (i) y1=1x5 (ii) f (x)1=1x−122 (iii) f (x)1=1x0.3 0 Exercises 23–26 EXAMPLE 4.10 For the ideal gas example discussed in Section 4.1, V= nRT , p 1 − nRT dV d 1 = nRT = nRT − 2 = p dp dp p p2 dy = 5x 4 dx f ′( x ) = − 1 −3 2 x 2

d f ( x ) = 0.3x −0.7 dx

since nRT is constant at constant T and n. We note that whereas V is inversely proportional to p, it is directly proportional to 12p; that is, V is a linear function of 12p, and the graph of V against 12p is a straight line with slope dV = nRT . d (1 p ) The rules for differentiating combinations of elementary functions are summarized in Table 4.3; in these rules, x is the independent variable, y, u and v are functions of x, and a is a constant. Table 4.3 Differentiation of combinations of functions Type

1. multiple of a function 2. sum of functions 3. product rule

Rule

d du ( a u) = a dx dx d du d v (u + v ) = + dx dx dx d dv du (uv ) = u +v dx dx dx d u du dv − u v2 = v dx v dx dx d df du f (u ) = × dx du dx dy dx =1 dy dx or dx dy × =1 dy dx

4. quotient rule 5. chain rule 6. inverse rule

104

Chapter 4 Differentiation

Linear combination of functions

A linear combination of the functions u, v, and w of x has the form y1=1au(x)1+1bv(x)1+1cw(x) (4.12)

where a, b, and c are constants. Such a function can be differentiated term by term; by Rules 1 and 2 in Table 4.3, dy du dv dw =a +b +c dx dx dx dx (4.13)

EXAMPLE 4.11 Differentiate y1=12x31+13ex1−1 1 1ln1x. 2 By Equation (4.13), 1 d 1 1 dy d d = 2 x3 + 3 ex − ln x = 2 × 3x 2 + 3 × e x − × dx dx dx 2 dx 2 x = 6 x 2 + 3e x − 0 Exercises 27, 28 1 2x

The product rule

The function y1=1(2x1+13x2)(51+17x3) can be differentiated by treating it as the product y1=1uv where u1=1(2x1+13x 2) and v1=1(51+17x 3). Then, by Rule 3 in Table 4.3, dv du dy =u +v dx dx dx = ( 2 x + 3x 2 ) d d (5 + 7 x 3 ) + (5 + 7 x 3 ) ( 2 x + 3x 2 ) dx dx

=1(2x1+13x 2)(21x 2)1+1(51+17x 3)(21+16x) This may now be simpliﬁed. In this example it is equally simple to multiply out the original product and differentiate term by term, but in many cases the brute force approach is more difﬁcult than use of the product rule.

4.6 Differentiation by rule

105

EXAMPLE 4.12 Product rule The function y1=1(2x1+13x2)1sin1x is easily differentiated only by means of the product rule. Let y1=1uv where u1=1(2x1+13x 2) and v1=1sin1x. Then dy dv du =u +v dx dx dx = ( 2 x + 3x 2 ) d d sin x + sin x ( 2 x + 3x 2 ) dx dx

**=1(2x1+13x 2)1cos1x1+1(21+16x)1sin1x 0 Exercises 29–32
**

The quotient rule

By Rule 4 in Table 4.3, d dx u du dv − u v2 = v dx v dx 2 x + 3x 2 5 + 7 x3

EXAMPLE 4.13 Differentiate y =

Let y1=1u2v where u1=1(2x1+13x 2) and v1=1(51+17x3). Then d d dy = (5 + 7 x 3 ) ( 2 x + 3x 2 ) − ( 2 x + 3x 2 ) (5 + 7 x 3 ) (5 + 7 x 3 )2 dx dx dx = (5 + 7 x 3 )( 2 + 6 x ) − ( 2 x + 3x 2 )( 21x 2 ) (5 + 7 x 3 )2

0 Exercises 33–36

The chain rule (function of a function)

The polynomial y1=1f (x)1=1(2x 21−11)3

106

Chapter 4 Differentiation

can be differentiated by ﬁrst expanding the cube and then differentiating term by term: y1=1(2x 21−11)31=18x61−112x41+16x 21−11 dy = 48 x5 − 48 x 3 + 12 x = 12 x ( 2 x 2 − 1)2 dx A simpler way is to use the chain rule. Let u1=12x 21−11 and rewrite y as a function of u: y1=1g(u)1=1u3 Then by Rule 5 in Table 4.3, dy dy du = × = (3u 2 ) × ( 4 x ) dx du dx and, substituting for u, dy = 12 x ( 2 x 2 − 1)2 dx In this example, y has been considered in two ways: (i) as a function of x: f (x)1=1(2x 21−11)3; (ii) as a function of u: g (u)1=1u3 where u is the function u1=12x 21−11. The substitution u1=12x 21−11 highlights the structure of the function, that of a cube, and makes the chain rule the natural method of differentiation.

EXAMPLE 4.14 Differentiate y1=1(2x 21−11)522. Put y1=1u522, where u1=12x 21−11. Then dy dy du 5 3 2 = × = u ( 4 x ) = 10 x ( 2 x 2 − 1)3 2 dx du dx 2 0 Exercises 37–40

Table 4.4 shows the generalization of Table 4.2 for elementary functions of a function, f(u) where u1=1u(x).

6 Differentiation by rule 107 Table 4. where u1=12x 21−11 dy dy du = × = ( − sin u ) × ( 4 x ) = −4 x sin( 2 x 2 − 1) dx du dx (iii) y = e2 x 2 −1 = eu .4.4 The chain rule Type power of u trigonometric Function ua sin1u cos1u tan1u Derivative a ua − 1 cos u du dx du dx du dx du dx − sin u sec 2 u eu du dx exponential logarithmic eu ln1u 1 du u dx EXAMPLES 4. where u1=12x 21−11 2 dy dy du = × = ( eu ) × ( 4 x ) = 4 xe2 x − 1 dx du dx (iv) y1=1ln(2x 21−11)1=1ln1u. where u1=12x dy dy du = × = (cos u ) × ( 2) = 2 cos 2 x dx du dx (ii) y1=1cos(2x 21−11)1=1cos1u. where u1=12x 21−11 dy dy du 1 4x = × = × (4 x ) = 2 dx du dx u 2x − 1 (v) y1=1ln(sin1x)1=1ln1u. where u1=1sin1x cos x dy dy du 1 = × = (cos x ) = = cot x dx du dx u sin x 0 Exercises 41–55 .15 The chain rule (i) y1=1sin12x1=1sin1u.

. the inverse function of f is deﬁned by x1=1f −1( y). dx Inverse functions d 2x e = 2e2 x dx If y1=1f (x). d cos 3x = −3 sin 3x . By Rule 6 in Table 4. for which du = a . (4. Then dx dy dy du dy = × =a dx du dx du For example.15(i) demonstrates the important special case of u(x)1=1ax. the derivatives of function and inverse function are related by d −1 dy dx d × = f ( x) × f ( y) = 1 dx dy dx dy The inverse rule dx 1 = dy dy dx is used when it is more difﬁcult to differentiate the function than its inverse. where the inverse trigonometric functions have their principal values.3. dy 0 Exercises 56–59 dx dy can be found as the inverse of : dy dx dy 1 = dx 5 y 4 − 2 Particularly important examples of the differentiation of inverse functions are given in Table 4.16 Use of the inverse rule If y is deﬁned implicitly by x1=1y 51−12y (see Example 2.14) EXAMPLE 4.5.108 Chapter 4 Differentiation Example 4. then dx = 5 y 4 − 2.11).

6 d −1 x = sinh dx a d −1 x = cosh dx a 1 x + a2 2 1 x − a2 2 d a −1 x = 2 tanh dx a a − x2 . and cos y is positive. 2 2 Then a cos y = a 1 − sin 2 y = a 2 − a 2 sin 2 y = a 2 − x 2 and dy = dx 1 a − x2 2 0 Exercises 60–62 The derivatives of the inverse hyperbolic functions (Section 3.17 Inverse trig functions x If y = sin −1 then x1=1a1sin1y and a dx = a cos y .5 d −1 x = sin dx a d −1 x = cos dx a 1 a − x2 2 −1 a − x2 2 d −1 x a tan = dx a a2 + x 2 EXAMPLE 4. Table 4.9) are given in Table 4. dy dy 1 = dx a cos y If y has its principal value then it lies between − π and π .6.6 Differentiation by rule 109 Table 4.4.

4) f (x. y) is zero for all allowed changes: d f ( x. y ) = 0 dx (4. y)1=1y 51−12y1−1x1=10.7 Implicit functions Every functional relationship between two variables x and y can be expressed in the implicit form (see Section 2.16) EXAMPLE 4.18 Inverse hyperbolic functions x If y = sinh −1 then x1=1a sinh1y and a dx = a cosh y . In some cases however neither variable can be so expressed and the rules must be applied to the implicit function itself.15) In all the examples of the applications of the rules of differentiation discussed in Section 4. Then the change in y that accompanies a change in x is such that equation (4. y)1=10 (4. and dy = dx 1 a + x2 2 0 Exercises 63. 64 4.110 Chapter 4 Differentiation EXAMPLE 4. We have dy dy d d df d = ( y5 ) − 2 ( y ) − ( x ) = 5 y 4 − 2 −1 dx dx dx dx dx dx = (5 y 4 − 2) dy −1= 0 dx . It follows that the rate of change of the implicit function f (x. Let equation (4.22). dy dy 1 = dx a cosh y Because cosh y1>10 (see Figure 3.6 it has been possible to express at least one of the variables as an explicit function of the other.15) is always true.19 Find dy2dx for f (x. a cosh y = a 1 + sinh 2 y = a 2 + x 2 .15) deﬁne y as a function of x.

c. v. because d ln y 1 dy = . 0 Exercises 65–68 4. For example. it reproduces the quotient rule. y1=1uv the method reproduces the product rule. = are constants. ln y = 1 ln(1 + x ) − ln(1 − x ) 2 . w.4. when y1=1u2v.6 by differentiation of the inverse function.8 Logarithmic differentiation For some functions it is easier to differentiate the natural logarithm than the function itself. EXAMPLES 4. dx y dx 1 dy a du b d v c d w = + + + y dx u dx v dx w d x (4.19) (4.17) This method of differentiating is called logarithmic differentiation. then ln1y1=1a1ln1u1+1b1ln1v1+1c1ln1w1+1and d d d d ln y = a ln u + b ln v + c ln w + dx dx dx dx Then. and a.20 Logarithmic differentiation 1+ x (i) y = 1− x Then 1 dy 1 1 1 1 d ln y = = + = dx y dx 2 1 + x 1 − x 1 − x 2 12 . b. = are functions of x. if y1=1uav bw c1where u. When.8 Logarithmic differentiation 111 Solving for 1 dy dy = gives dx dx 5 y 4 − 2 This is the result obtained in Example 4.18) (4.

. . . .. . d d d ln y = x ln x + ln x x = 1 + ln x dx dx dx and.. . . . . . ..20) x(t)1=1x0e−kt x nl 12 = 1 (1 + x ) 12 (1 − x )3 2 dy = a x ln a dx dy = x x (1 + ln x ) dx and. .8 t .... . .8. dx y dx (iii) y1=1x x.. . .112 Chapter 4 Differentiation and.. . . . . . .... .. . .. .... . . . ... . . . . . k−. . .. . .... . and intercept ln1x 0 with the axis t1=10. . . . . . . Figure 4... 1 dy = 1 + ln x and y dx 0 Exercises 69–72 EXAMPLE 4. o 0x nl . . .. . therefore. . . . . ... . . ... .. . .. . . . dy 1 1 + x = dx 1 − x 2 1 − x (ii) y1=1a x... ln1y1=1x1ln1a. . . . . . .. . . . . . . . ... .. ln1y1=1x1ln1x... .. . . Applying the product rule.... . . . .... . . This example is important because it demonstrates the standard way of calculating the rate constant k from a linear plot of experimental values of x and t.. . . ... multiplying by y.. . • . ln1x1=1ln1x01−1kt A plot of ln x against t.. 0 Exercise 73 Figure 4.. . .. . ... . . therefore. . .. .. . . . . . . . . .. . . the size of the system is given by (see Example 3. . . . .. gives a straight line with slope d(ln1x)2dt1=1−k.. . . . . . . ...... Then d 1 dy ln y = = ln a and . ... . .. . . . taking the logarithm of both sides.21 Logarithmic plots For a system undergoing exponential decay (ﬁrst-order kinetics). . ... .

d dy d 2 y = 6x + 2 = dx dx dx 2 Successive differentiation gives the third and fourth derivatives d3y dx 3 = 6.4. or Df. D2f.1=1. the cubic y1=1x 31+1x 21+1x1+11 has (ﬁrst) derivative dy = 3x 2 + 2 x + 1 dx and this may be differentiated to give the second derivative.9 Successive differentiation The derivative of a function can itself be differentiated if it satisﬁes the continuity and smoothness conditions discussed in Sections 4.22 Derivatives of sin1ax. f ′′′(x). Alternative notations (see Section 4. where D is the differential d operator .3 to 4. EXAMPLE 4. In particular. D3f.1=1for derivatives of f(x).9 Successive differentiation 113 4. or second differential coefﬁcient. For example. but other simple functions can be differentiated indeﬁnitely.2) are f ′(x). f ″(x). ( −1)( n −1) 2 a n cos ax ( x) = ( −1) n 2 a n sin ax if n is odd if n is even f ( n) . f (x)1=1sin1ax f ′(x)1=1a1cos1ax f ″(x)1=1−a 21sin1ax1=1−a 2f (x) In general. the exponential function e x has all its derivatives equal to e x. for the nth derivative. d4y dx 4 =0 and all higher derivatives are zero. dx 0 Exercise 74 A polynomial of degree n can have only up to the ﬁrst n derivatives nonzero.5.

..20) To distinguish between maximum and minimum values..... ...... The function goes to ±∞ as x1→1±∞. . ... ... . ... for example.. y1=1x(x1−13)2 . . . ... ... ...... .. To the right of A the gradient is negative and y is decreasing.. On moving through the maximum at A from left to right....... ... . . .•.... . . .... ... A ... The second derivative f ″(x) is the rate of change of slope.... . . ... ... ... . .. 3 B .. ....... . .. .. . O . ........ 1 . . ..... .. ... . .. .. .... the x ..... .. . ... .......... . ........... .... . • C . .... . .. .. . where its value is greater than at all neighbouring points.. ..... ... . . . . . . . ..... . ... .. . ........ .. .. ... but the graph shows that the function has a local maximum at point A. .. Figure 4. ..... . .... ..... . ....114 Chapter 4 Differentiation For example...... ... ... ... .25).. Points of maximum and minimum value are called turning points... ....... .. ...... . ..... .. .... .. ...... . .... .. whereby solutions are obtained as maxima or minima of some variational function (see Example 4... . ..... ... .... . .. To the left of A the gradient is positive and the value of the function is increasing.. ...... y . ... .10 Stationary points ....... ...... . .. .... .. . Consider the curve in the neighbourhood of the maximum at point A in Figure 4...... . and the general condition for a stationary point is that the ﬁrst derivative of the function be zero: dy =0 dx at a stationary point (4. . . The point is called a stationary point.. .... at x1=11.. . . . .. .. . or the slope of its graph at point x. ... . ... .... where its value is smaller than at all neighbouring points. .. .. .... ...... . At the point A itself the function has zero gradient (the tangent to the curve is horizontal)... ... ... ........ .9 The determination of the maximum and minimum values of a function is of importance in the physical sciences because. . .... Similar considerations apply for the minimum at B. .. .. . .. . .. .. ............ . . .. ... ... ..•. .... .... ....... . . ... (i) equations of motion are often formulated as ‘variation principles’.. ..... ... ..... . .. . . . ... Consider the cubic (Figure 4. .. . .. .. . . . . . ... ... it is necessary to consider the second derivative..................... .... as in the method of least squares discussed in Chapter 21.. ....... .. . . .. .... .. .. .... . ..... .9) 4 3 2 1 ..... ... . n1=13: n1=14: 0 Exercises 75–77 The ﬁrst derivative f ′(x) of a function f(x) is the rate of change of the function... 2 . ....... . . ........ The function also has a (local) minimum at point B. ..... . .... ... . . . f (3)(x)1=1f ′′′(x)1=1−a31cos1ax f (4)(x)1=1(−1)2a41sin1ax1=1a41sin1ax 4........... .. and the rate of change of y with respect to x is zero. . . ...... .. . ... . ... .9.... .... . and the value of the function at the point is called a stationary value.. ... .. ......... ....... . ... . at x1=13.. and is related to the curvature at x...... . .. . . ......... .... ..... (ii) the ﬁtting of a theoretical curve to a set of experimental points can be expressed in terms of a ‘minimum deviation’ principle........... .... . .. .......... ..... ...... .... ... .. .

It follows that the rate of change of the gradient is negative at A.4. with minimum value at C.24) . d4y dx 4 ≠0 (4. to negative values. through zero at A. y1=1x(x1−13)21=1x 31−16x 21+19x dy = 3x 2 − 12 x + 9 = 3( x − 1)( x − 3) = 0 when x = 1 or dx d2y <0 = 6 x − 12 > 0 2 dx =0 when x = 1. The slope of the curve decreases dx 2 (becomes more negative) between A and C and increases between C and B. a point of inflection x=3 dy = 0 and dx d2y dx 2 >0 (4.22) The point C. and is the case discussed in Example 4.9. Two examples are When (i) dy = 0. and this is a sufﬁcient condition for the function to have maximum value at this point: for a maximum: dy = 0 and dx d2y dx 2 <0 (4. dx d2y dx 2 = 0. is an example of a point of inﬂection.21) Similar considerations applied to the minimum at B show that for a minimum: For the cubic shown in Figure 4. with = 0. (ii) dy = 0. This is an example of a simple point of inﬂection with dy ≠ 0 and dx 0 Exercises 78–82 d2y dy = 0 and = 0 at a point then the nature of the point is determined by dx dx 2 the ﬁrst nonzero higher derivative.23) d2y dx 2 =0 This is a point of inﬂection which is also a stationary point (but not a turning point). a minimum i when x = 2. dx d2y dx 2 = 0. d3y dx 3 = 0. at x1=12 in Figure 4. d3y dx 3 ≠0 (4.9.23. at which the d2y gradient is a maximum or minimum. a maximum when x = 3.10 Stationary points 115 gradient decreases from positive values.

. dy = 12 x 3 − 12 x 2 = 12 x 2 ( x − 1) dx =10 when x1=10 or x1=11. . .. .. . ........ .. ..... . .... . .... ... . ... For the stationary values of ε. .. .. the possible values of the orbital energies of the π electrons of ethene (C2H4) are given by the stationary values of the quantity ε1=1α1+12c(11−1c 2)122β where α and β are constant ‘Hückel parameters’. y1=1(x1−11)2(3x 21+12x1+11) and has a double root x1=11. .. ... .. 1 • . .. .... d3y dx 3 = 72 x − 24 ≠ 0 when x1=10......... .... .... . . a minimum. . .. dε = 2(1 − c 2 )1 2 β − 2c 2 (1 − c 2 )−1 2 β = 0 dc ... .... at x1=11... . ... . ..... .. EXAMPLE 4... .. . . .. ... . . . .... . .... . ..... .. and a point of inﬂection at x1=10. ... . the function y1=1(x1−11)4 satisﬁes (4. . .. .... .. .. . .. . .. d2y =0 = 36 x − 24 x 2 >0 dx 2 when x = 0 when x = 1.. ... ... .. . . .. .. .. ................ ..... . when y1=10.. For the stationary values... • 1 o . .. .... a minimum m Figure 4. For example.. ... .10 To determine the kind of stationary point at x1=10. .. .... In fact..... . To determine the kinds of stationary points... ... . . . ..... .. 0 Exercise 83 EXAMPLE 4... the function can be factorized.. . with a minimum at x1=11... . ..... .... .. .. . . ... ... . ........116 Chapter 4 Differentiation This is a maximum or minimum. . . .. .. and c is a variable. depending on the sign of the fourth derivative. . ...... .... .. .. a point of inﬂection The function therefore has a single turning point.. when y1=11....... .... .... . .... ..24)..... ... .23 Find the stationary points of the quartic y1=13x41−14x 31+11... ... .. ... . . . ... .. ..24 In Hückel molecular orbital theory....... .... . ...... ... . .. .. .. ... and two complex roots. . . . . ... .... .. ............. ... . ..... . ... .. .. ....

.. .... .. . t= r1 v1 + r2 v2 The problem is to ﬁnd point O such that t is a minimum. . . ... .. .. . ... . . .... ..... .... ... .. . ... .. . EXAMPLE 4... . . .. . ..... ..... . .. . . . .. the speed of light is v11=1c2η1. .... .. .... where c is the speed of light in vacuum and η1 is the refractive index of the phase.... Then ε1=1α1±1β... .. . .. . .. .. ... . .. .... ..... .. . .. . . . . .... . ..... . .. .. Dutch mathematician and physicist. .... ... . . . . . .. ..5 The total time travelled from P to Q through point O is (distance2speed in each phase). . .. . . ... 5 This use of the principle of least time. . .. . .. . ... . ........ . .... 2 2 12 2 2 2 r2 = ( x2 + y2 )1 2 = ( X − x1 )2 + y2 4 Willebrord van Roijen Snell (1591–1626). . ... .. .. . .. . ... .. .. . . .. .. .. ... . . ..... . . ...... . . . . ... . . .. .... .. .10 Stationary points 117 Division by 2β and multiplication by (11−1c 2)122 gives (1 − c 2 ) − c 2 = 0.. .. .. ..... .. .. .. .. . .... .. . .. .. . .. Choosing x1 as the independent variable. .. .. .. .. . . . . . . ... . ... . . ... . . .. ... .. ..... .4.11 A ray of light travels between points P and Q across a phase boundary at O. .. . .. . formulated the law of refraction in 1621. . . ... was one of the examples used by Leibniz in his 1684 paper to demonstrate his method of ﬁnding maxima and minima.. .. . . .. .. .. ..... ..... . . .... ... ... . . . .4 or c=± 1 2 Figure 4...... .. ... ..... . . .. . . .. . . ... . .. ... . . ... ......... . . . . .... .... . . .. . . .. . . ... ... ... . ... .... . .... .... Snell’s law of refraction is sin θ1 v1 η2 = = sin θ 2 v 2 η1 The law can be derived from a ‘principle of least time’........ that the path followed is that of least time.. . . . . proposed by Fermat.. . .. .... . y r adnu ob es ahp 2y .. . . .. . . .. ....25 Snell’s law of refraction in geometric optics... .. ... In the upper region. .. .. .. . .. . .... .. we have r1 = ( x1 + y1 )1 2 . .... . .... .. ......... .. . . . .. ...... .. .... .. .. . ... . . . .. . P • 2r 2x 2θ o 1θ 1x 1r • Q 1y .. . . .... . . . .. ... In the lower region the speed of light is v21=1c2η2. . . . . .. .. .. ...... .

... The magnitude of the velocity is the speed......... . p • x o • .. ... motion in a straight line and motion in a circle. Let O be a ﬁxed point and let P be the position of the body at time t....... vectors are discussed in Chapter 16.......... ... .......... .......... ........ Then i dt 1 dr1 1 dr2 1 x1 1 x2 sin θ1 sin θ 2 = + = − = − dx1 v1 dx1 v 2 dx1 v1 r1 v 2 r2 v1 v2 =10 for a minimum Hence Snell’s law............................... along the x-direction say........ ................. ........ The derivative of the velocity is the acceleration.. ... ................... x is increasing and the body is moving to the right... .. The distance OP1=1x is then a function of time... having both magnitude and direction...... ....... or simply the velocity................. .... We consider here only the simplest kinds of motion........ y2........ and X1=1x11+1x2 are constant.......... ... velocity = v = dx dt (4.............. If the body moves from point x to point x1+1∆x in time interval ∆t.... When v is negative..... ......... ...............25) When v is positive. x1=1f (t)................. ....... x is decreasing and the body is moving to the left. 0 Exercises 84–86 4............. ...118 Chapter 4 Differentiation in which y1.........26) x ...12 Consider a body moving in a straight line.... .... then the average rate of change of x in the interval is ∆x = average velocity in interval ∆ t ∆t The limit of this as ∆t1→10 is the instantaneous rate of change of x with respect to t...............................11 Linear and angular motion The description of the motion of bodies in space is an important application of the differential calculus.. ... ..... Linear motion Figure 4.............. It is the linear velocity. at time t. ................ More general kinds of motion are discussed in Chapter 16. .. ........ ..... ............. Velocity is in fact a vector quantity..... acceleration = dv d 2 x = 2 dt dt (4................

... ... as has been emphasized before.. . . . . ... .. ...... . . One of the aims of the Society was to promote ‘the principles of pure d-ism as opposed to the dot-age of the university’. . . . . ..... . . .. .. .. ... . . . . . .. . ... .. Let O be a ﬁxed point on the circle and let the position of the body at time t be given by the angle θ (Figure 4. . . . . .. .. .. ... . .... .. . . f ( x + ∆ x) − f ( x) ∆y dy = f ′( x ) = lim = lim ∆x→0 ∆x ∆x→0 dx ∆x and.. .. .. . .. . .. . .. The rate of change of θ with respect to time is called the angular velocity: ...27) angular velocity = ω = lim ∆t →0 ∆ θ dθ = =θ ∆t dt (4. . . . . . . .... .. or ﬂuents. .. ... . John Herschel (1792–1871).... .. . . the symbol dy2dx does not mean the quantity dy divided by dx... .. .. . .. . . . this is a useful way of describing changes.. . .... . is a better symbol for the derivative. . ... .... .. ... .... . ....13). . . . .... .. . ... because arc length s is related to subtended angle θ by s1=1rθ.. . .. . . .. .. .. . .. . . . . .. . .. . .. ... .. . . .4. . . .. . .. . ...... . ... . . . .. It is nevertheless tempting to write dy1=1f ′(x)1dx and.. . .. .. . . . . .. ...... dx .. ... . . .... . . .. ... . . ... . .... 6 In his 1671 paper on the calculus. .... . ..... but represents the value of the limit.13 (4. . ... .. . . . . . .. . . .. . . .. . . . . . Newton considered variables like x and y as ﬂowing quantities. . .. . . . . . .. . ... .. . . . . . ... . ... ... . . .. . or y′.. . . .. .8). . . The angular acceleration is 71=15.. ... .. . .. . .. . . . ... .28) In addition. . . . ..... . .. . . . . . . ...... . .. . . .. . .. . . ... and wrote H and I for their rates of change. .... .. ..... .. . . . . .. Methodus ﬂuxionum et serierum inﬁnitorum.. .. . .. .. . . .. .. .. ... .. .. . . . o t t • • ∆ +t θ θ • r • Consider a body moving in a circle of radius r. .. . .....12 The differential The ﬁrst derivative of a function y1=1f (x) is deﬁned by equation (4. . .. .... .. dt v=x= d 2x dt 2 (4... . . .. . and Charles Babbage (1792–1871) founded the Analytical Society in 1813... . .12 The differential 119 Derivatives with respect to time are sometimes written in a ‘dot notation’:6 v=x= 0 Exercise 87 Angular motion . . .. . .... ... . 0 Exercise 88 4. .... . . . . .. ..... .. ... . .. . . . we have K1=1r4 or v1=1rω Figure 4.. . . .... .. . .... ... . . Dotted ﬂuxions were still being used by English mathematicians when the Cambridge undergraduates George Peacock (1791–1858)... . . . . .. .. ......∆ . . . . or ﬂuxions.. in this sense f ′(x).. when properly interpreted. . . . . ... . . .. . .29) for the relation between linear and angular velocities for motion in a circle (see Chapter 16 for a more general discussion of velocity in terms of vectors).. . .. ... . .... . ... . .. . .. .

. dy = f ′( x ) = 3x 2 dx Let ∆y be the change in y accompanying the change ∆x in x: ∆y1=1f (x1+1∆x)1−1f (x)1=1(x1+1∆x)31−1x 3 =13x 21∆x1+13x(∆x)21+1(∆x)3 The derivative dy2dx is obtained by dividing this expression by ∆x and letting ∆x1→10. The corresponding change in y dy1=1f ′(x)1dx (4. Another way of looking at the limit is to consider ∆x as a ‘very small’ change.7 The use of the differential will become clear in later chapters. It is useful to consider an arbitrary small change dx. an ‘inﬁnitesimal change’. in particular. It is important in the physical sciences because fundamental theorems are sometimes expressed in differential form. If ∆x is made small enough then the term in (∆x)3 becomes much smaller than the term in (∆x)2 which in turn becomes much smaller than the term in ∆x. An approximate expression for the change in y is then ∆y1≈13x 21∆x1=1f ′(x)1∆x and this is often a useful way of approximating small changes.26 The differential area of a circle The area of a circle as a function of the radius is A(r)1=1π r 2 7 Leibniz’s formulation of the calculus was in terms of differentials. The quantity f ′(x)∆x would be the change in y if ∆x were small enough.30) is called the differential of y. such that terms in (dx)2 and higher can be set to zero. for the inﬁnitesimal change or differential of x n. (∆x)31<<1(∆x)21<<1∆x For example. 0 Exercises 89–91 EXAMPLE 4. and dxy1=1xdy1+1ydx for the product rule (see Example 4.120 Chapter 4 Differentiation Consider the cubic y = f ( x) = x3 . ∆x1=110−3. and (∆x)31=110−9.27). (∆x)21=110−6. His 1684 paper contains the formulas dx n1=1nx n−1 dx. the laws of thermodynamics are nearly always expressed in terms of differentials.

. .14 The corresponding ‘differential area’ is dA1=12π r1dr1=1circumference1×1width 0 Exercise 92 EXAMPLE 4. .32) (4..... ... . .. . .. . .. .. .. .. .. .. . . . u. ....... . . . . ... . . . . . . . . this expression is equivalent to the normal form of the product rule: ‘division by dx’ gives dv du dy =u +v dx dx dx An important application of the differential is as a formal procedure for changing the independent variable.. . . ..... . and v are functions of x.. .... . . . .... .. ....... . .. . . . .. . .... . . such that x1=1g(t) with differential dx1=1g′(t)dt Substitution of this in (4. ...... .. .4. .... . .. then the change in y accompanying changes in u and v is ∆y1=1(u1+1∆u)(v1+1∆v)1−1uv1=1u1∆v1+1v1∆u1+1(∆u)(∆v) and the differential form is dy1=1d(uv)1=1u1dv1+1v1du If y. .... ......... ..... ... .. ... .. . .... ...31) then gives dy1=1f ′(x)dx1=1f ′(x)g′(t)dt (4.. . . . . .. .. . . . . . ....... . . .... . .. . ... ..... ..... . .... . ...... .. . . .. .. . ... . . ... .... .. . . . . ... .. . . . .... .. . . .. . .. . .. . .. .. . . ... . t.. ... . . ... . . . . ..... .... . .. . ... . . .. . . . .. . . .... .. ... ... .. . .. . .. . .. . .... ... .. .. . . . . . . .. ... . . .. . . . . .. ....27 Differential form of the product rule If y1=1uv....... . . ... . . .. .... . . . . . .. ...... . .. .. .. . . .. . . . .... . .. ... . . . . . . .. ... .... .... .. . . ....... .. . .. . ... . .... .. .. .. . . . ... .. . .. .. .. . . . . .... . . .. . .. Consider y1=1f (x) and its differential dy1=1f ′(x)dx Let x be a function of some other variable... ... .31) r ∆ r • ... ∆ A ≈ 2π r ∆ r = dA ∆r dr Figure 4. . .... .. . .... ... .. . . ..... . .. . . . .. . . ...... . .. . . ... . .. . .. .... .. ... . . When ∆r is small enough... . . .... .. . .. .... . .. . . . .. the corresponding change in the area is ∆A1=1π (r1+1∆r)21−1π r 21=12π r1∆r1+1π(∆r)2 and this is the area of a circular ring of radius r and width ∆r.. . . . .. .. .12 The differential . .. . .... 121 If the radius is increased by amount ∆r. . ...... .. .. . ... . . . .. ... ..... . . .. . .... . .. ..... . ... .. . .. . .... ..

4 Find the limits: x2 7. 2. Sketch graphs to demonstrate your answers.13 Exercises Section 4. x + 1 lim x → 0 x + 3 8 In fact.33) This is the differential of y with respect to the variable t. 1996). 2x x − 3x 2 Section 4. (ii) lim ∆p→0 ∆p Section 4. For the Langmuir isotherm θ = Kp ﬁnd (i) the change ∆θ in θ that corresponds to 1 + Kp ∆θ change ∆p in p. . lim 2 x→0 x 9. 4.3 Find the discontinuities of the following functions and state which are essential and which removable. For y1=1x 3. lim x→0 x x 8.2 1.8 4. Division by dt then gives the Chain Rule: dy dy dx = × dt dx dt (4. developments in mathematical logic between 1920 and 1960 have led to the development of a ‘nonstandard analysis’ which involves an extension to the number system to include inﬁnitesimals (Abraham Robinson.122 Chapter 4 Differentiation or dy dx dy = × dt dx dt (4. ﬁnd ∆x →0 ∆x ∆y lim 3. Non-standard analysis. (ii) ∆y2∆x.34) Although there have been conceptual difﬁculties with this type of manipulation of differentials (inﬁnitesimals or ‘inﬁnitely small changes’). For y1=1x 3. ﬁnd (i) the change ∆y in y that corresponds to change ∆x in x. operations with differentials can always be shown to duplicate methods involving ﬁnite changes (∆’s instead of d’s) and limits. Princeton. x2 x 6. 1 x +1 5.

1 3− x 2 2 40. 3x 2(21+1x)122 55. x123 26.5 Differentiate from ﬁrst principles: 19. (11−14x 2)2sin1x 31. (21+13x)e x 34. e x1cos1x 35. ln 3− x 2 38. x 3 2 3 27. 2x 21+13x1+14 21. 22x 2 18.4. (11+1x)5 41. e2 x −3x +1 44. (1n1x)2x Chain rule Differentiate 37. lim 4 x 2 − 2 + 2 x − x → 0 x x e2 x − 1 15.6 Differentiate by rule: 24. x 2 e2 x 2 52. lim 2 x → 1 x − 1 x 2 − 1 lim x→∞ x +1 x→0 11. tan14x1cos 212x +3 Inverse functions 56. x 322 22. lim (ln x − ln 2 x ) 17. cos (2x 1−13x1+11) 2 + x 49. Products and quotients Differentiate 29. 3x 2 ( 2 + x 2 )1 2 50. lim x→0 x 16. ﬁnd Find dy . a and b dp are constants). dx dV at constant T and n for the following equations of state (assume that B. . x 524 25. 3 ( 2 x − 3x − 1)1 2 2 2 42. The virial equation of state of a gas at low pressure is pV = nRT 1 − . e− cos( 3x +2) 47. If x1=12y 21−13y1+11. Find V dV at constant T and n (assume B is also constant). ln(sin12x1+1sin21x) 54. x→∞ lim ln( x − 4) − ln(3x + 2) Section 4. sin14x 45. (11+12x1+13x 2)2(31+1x 3 ) 30. e 1 sin x 43. 2 + x2 39. x4 20. cos1x2sin1x 32. sin1x1cos12x 53. 2 1 1 14. ln(2x 21−13x1+11) 48. x → ∞ x + 3 x + 1 lim 12. e−x Section 4. x → ∞ x 2 − 1 x −1 lim 13.13 Exercises 123 x −1 10. 12x 3 23. 11−12x1+13x 1−14x 1+151sin1x1−161cos1x1+17e x1−181ln1x nB dp 28. x1ln1x 36. e−2x 46. (11−14x 2 )1cos1x 33. ln(cos1x) 51.

(1 + x 2 )( x − 1)1 2 ( 2 x + 1)(3x 2 + 2 x − 1)1 3 dp p∆H vap = dT RT 2 71. y1=1x 21−13x1+12 81.72 1877 0. the following concentrations of N2O5 were obtained: x1=1[N2O5]2mol1dm−3 t2s 2. x1=1y1ln1xy 65.0 1.36 867 1. x 21+1y 21=14 66. 77. tan−11x 2 58. sketch the graph and show the positions of these points. Section 4. sinh−112x 64. The decomposition of dinitrogen pentoxide in tetrachloromethane at T1=145°C has stoichiometry: 1 N 2O5 → 2 NO 2 + 2 O 2 and obeys ﬁrst-order kinetics. Section 4. 73. .55 2315 Plot a graph of ln x against t2s and determine the rate constant.9 74. 79. Find a general formula for the nth derivative of cos12x. 4 + x 13 70. Find dy d 2 y d 3 y d 4 y . cos −1 1+ x Section 4. In each case. p(V1−1nb)1−1nRT1=10 n2 a 59. y1=1xe−x (see Figure 3. Show that the equations d ln p ∆H vap = dT RT 2 and are equivalent expressions of the Clausius–Clapeyron equation. sin−112x 61.11 1196 0.20) . dx dx 2 dx 3 dx 4 76. y1=14x 31+16x 21+13 78. y 2 + − x 2 y 2 + 3x + 2 = 0 y Section 4. Find all the nonzero derivatives of the function y1=13x 51+14x41−13x 31+1x21−12x1+11. sin1221x1cos3(x 21+11)tan12312x 72.7 Find dy : dx 67. From the volumes of oxygen liberated after various times t. .124 Chapter 4 Differentiation nB 57. 75.10 Find the maximum and minimum values and the points of inﬂection of the following functions. Find a general formula for the nth derivative of e3x. pV = nRT 1 + V Differentiate 60. y1=1x 31−17x 21+115x1−19 80. tanh−11x 2 1− x 62. for the function y1=1ln x. y 31+13x1+1x 21−111=10 2 68.33 0.8 Differentiate: 3− x 69.91 319 1. p + 2 (V − nb) = nRT V 63.

(ii) U(R) in terms of R. y1=12x 90. and describe the motion of the particle. Derive the differential dV from ﬁrst principles. (ii) Sketch graphs of s and v as functions of t in the interval t1=101→12. in terms of the rate constants k1 and k2. and (ii) show that the maximum concentration is k [ B]max = [A]0 1 k 2 k2 ( k2 − k1 ) Section 4. 84. . The concentration of species B in the rate process A B → C . A particle moving on the circumference of a circle of radius r1=12 travels distance s1=1t 31−12t 21−14t in time t. Find the most probable speed (for which f (v) is a maximum). y1=13x 21+12x1+11 92. Conﬁrm that the cubic y1=1x 31−17x 21+116x1−110.4. (iii) ﬁnd the stationary values. 1 2 → 86. The probability that a molecule of mass m in a gas at temperature T has speed v is given by the Maxwell–Boltzmann distribution m 2 − mv 2 f (v ) = 4 π v e 2 π kT 32 2 kT where k is Boltzmann’s constant. Give a geometric interpretation of the result. 88. y1=1sin1x 89.12 Find the differential dy: 91. Sketch a graph to show the positions of these points. discussed in Example 2. (iii) Sketch graphs of θ. The Lennard-Jones potential for the interaction of two molecules separated by distance R is U ( R) = A R 12 − B R6 where A and B are constants. is given by (when k11≠1k2) k k [ B] = [A]0 k1 −k t −k t (e 1 − e 2 ) k2 − k1 (i) Find the time t. Section 4. 83. (i) Express the distance travelled in terms of the angle θ subtended at the centre of the circle. Find the maximum and minimum values and the points of inﬂection of y1=12x 51−15x41+13. (ii) ﬁnd the angular velocity ω and acceleration 7 around the centre of the circle. A particle moving along a straight line travels the distance s1=12t 21−13t in time t. (i) Find the velocity v and acceleration a at time t. (iv) ﬁnd the stationary values. 85. has local maximum and minimum values at x1=12 and x1=1823. consisting of two consecutive irreversible ﬁrst-order reactions. ω and 7 as functions of t in the interval t1=101→14. and describe the motion of the particle. at which B has its maximum concentration. The equilibrium separation Re is that value of R at which U(R) is a minimum and the binding energy is De1=1−U(Re ). Re and De. Express (i) A and B in terms of Re and De. The volume of a sphere of radius r is V1=14π r 323.13 Exercises 125 82.11 87.23.

. . . . . . .... .. ... Dean of Rouen Cathedral and Bishop of Lisieux. . . . . . .. . . . .... .. . .. . . . . . . . . . .....2. . .... Therefore.. . ... . . . . . ..... . .... . . . .. . ... . as illustrated in Figure 5.. B . if s(t) is a known function... . .. . . .. . . . . . . .1 the average velocity is v = (v A + v B ) 2 and the total distance travelled is 1 d = (v A + v B ) × (tB − tA ) 2 1 This graph of velocity as a function of time for a body moving with uniform acceleration appeared in Quaestiones super geometriam Euclides by Nicole Oresme (c... . . . . . ... . 1323–1382). .. .. . . . . v (t)1=1ds2dt is the gradient of the graph of the function s(t).. . . .. . . ..1) where v is the average (or mean) value of v(t) between A and B. . . .. . .. . . .... . . . . .. .. . v (t) is obtained by differentiation.... . . that is. . . . . . .. .. . .. ... . if v (t) is a known function then s(t) is that function whose derivative is v(t).. t . . . . .. .. . . .. The distance travelled between two points is equal to the average velocity multiplied by the time taken. . .. Let the distance from A along the curve at some intermediate time t be s(t). .. . . . ... . •.. . A )t(s t=t . ... .. . .1 If v(t) is the velocity along the curve at time t then. .. . . . . . ... .. • A ..1 Concepts Consider a body moving along a curve from point A at time t1=1tA to point B at time t1=1tB. . . ..•. ... . .. . . . ... For example. .. . . . if the body undergoes constant acceleration from v1=1vA at A to v1=1vB at B.... .. This was possibly the ﬁrst graph of a variable physical quantity.. emit . .. . For the motion shown in Figure 5....... to ﬁnd s(t) we need to reverse the differentiation ds2dt1=1v(t).. ... .. .. .. . . . . .. Figure 5. .. . Oresme also considered the extension of his ‘latitude of forms’ to the representation of the ‘quality’ of a surface by a body in three dimensions and ‘the quality of a body will no doubt be represented by something having four dimensions in a different kind of quantity’. .. .. . .. ...1.. . the distance AB along the curve is therefore d = v × ( t B − tA ) (5. . . . ... as illustrated in Figure 5. . . .. . .... . . .. . Conversely.11.. . . . . . . . .. . . by the discussion of Section 4. . . . . . . .1. t=t B ..5 Integration 5. .. . . .. .. .. .. ... .. ... .

..... . ..... . .. ......... a follower of Galileo. ............... ...... .. 408–355 BC).. . . . Archimedes in his Quadrature of the parabola attributed the method to Eudoxus of Cnidus (c.... . described in his inﬂuential Geometria indivisibilibus continuorum............ .... ..... . .... .. ....... . .. ...... . . .. . ........ ..... .... The case of n1=1−1 was treated by Gregoire de Saint Vincent (1584–1667).... ..... . .. . .. ... . .... ..... The second. . . ... .. . . . . ....... ... .. . .. .... ........... ..... . .... ... ...... ... .... ....... ... . .... ....... . .. . In the Method...... .... best known for his Astronomia nova of 1609. .. .. . and developed a geometric method for ﬁnding the integral of x n for positive integers n.2 The demonstration by Leibniz and by Newton that differentiation and integration are essentially inverse operations is one of the landmarks of the history of mathematics. .... .... ............ . .. ..... .. . .. .. ... ... ... ........ .......... ........ . .. .. This example demonstrates the two central problems of seventeenth-century European mathematics... ..2 It is readily veriﬁed that this distance is equal to the area.. 1635.... .. and Barrow.. . .. We shall see that this last result is valid for any velocity function v(t)............. .. A B v v .... ... . ... . .. .... .. ........ ...... ..... A t ....5. . We shall also see that the solution of a physical problem is often equivalent to ﬁnding the area enclosed by an appropriate curve..... computed areas and volumes by considering them to be composed of inﬁnitely many inﬁnitesimal elements.... ... ... ......... . ... The indeﬁnite integral dx Let y1=1F(x) be a function of x whose derivative is F ′( x ) = of the derivative is deﬁned by 2 Integration has its origins in the Greek ‘method of exhaustion’ for ﬁnding areas and volumes... . .... John Wallis (1616–1703). ........ .. ..... .. ... ............ ............ .. ....... . ..... ........ ......... . . .............. ......... ... The concept of integration as the inverse operation to differentiation leads to the deﬁnition of the indeﬁnite integral... . ..... how an area can be thought of as made up of lines or ‘indivisibles’ and a volume of areas...... discovered in Constantinople in 1906 after being ‘lost’ for over a thousand years.... . ... . . . . ...... . whose lectures Newton attended and a copy of whose Lectiones was bought by Leibniz when on a visit to London in 1673........ ....... • . . ....... .. . Roberval integrated the sine function in 1635....... ..... ..... .. Archimedes describes how a plane area can be regarded as a sum of line segments. ... . ... ... ... .... .... .... ... ........ . . . •... ... ........... . . .. ...... .. . .. ....................... .. .......... ......... ... ..... . .. ........ His work on volumes appeared in 1615 in Nova stereometria doliorum vinariorum (New solid geometry of wine barrels). .. . .......... .. .. ........ ... .. ....... The concept of the integral as an area leads to the deﬁnition of the deﬁnite integral.... ........... .. .. ................ ........... . ... ... ... . .. . ... ............ ......... ....... ..... .. . whose Traité des sinus du quart de cercle of 1658 Leibniz said inspired his discovery of the fundamental theorem....... Chapter 4) and by Leibniz. . . ... . ....... ... . At about the same time..... . . ... .... .. .. .. t B .. ... ... .. .. . ... ... . ... ...... ... .. . . . . .. to ﬁnd the tangent lines to an arbitrary curve. ..... .... .. ......... .... ........ .......... .. Johann Kepler (1571–1630)...... ..... 5...... . .. ......... ... .. .... who published the ﬁrst account of his integral calculus... .. ....... .... ....... ...... ......... . .. In 1586.. ... . . .. ... and Torricelli the log function in 1646........ . ..... led to the invention of the differential calculus... ....... . .. Galileo Galilei (1564–1642) made use of the inﬁnitely small in his work on dynamics...... ... .. ....... ... ........ ............ .... .. . . ....... .. . . . .......... led to the invention of the integral calculus.. . . .... .. .......... .. ........... . ..... ... ....... ..... .. .... . ......... . ...2 The indeﬁnite integral v 127 Figure 5.. .... . .. . .. . The ﬁnal step in the synthesis of the differential and integral calculus was taken by Newton (Footnote 3. ..... ....... ... .... . .... . . the Scot James Gregory (1638–1675) whose work on inﬁnite series and the calculus anticipated that of Newton............ Stevin described how the centroid of a triangle can be obtained by considering the area as made up of a large number of parallelograms... ..... ... ..... .............. to ﬁnd the area enclosed by a given curve................. ... ... .. ....... .... .... . ... . ........... . the ‘problem of tangents’ and the ‘problem of quadrature’.... . .. ... shaded in the ﬁgure. .......... ..... ... ... ..... ......... . . Fermat solved the same problem for positive and negative integers (except n1=1−1) and for fractions by dividing his areas into suitable rectangular strips... ... ... ..... ...... .......... ...... .. .. .. ... . ... ... ........... ... .... .......... .... . .. . .... ... .. .. .. .... .... . ... .. .. ....... ........ ... ... ..... ..... ....... ... .. ....... . . ........... . ... ....... .......... ...... .... ... ......... ......... .. bounded by the line v(t) and the t-axis between tA and tB. ...... ..... ... ........ . . ........ ...... ... ..... ........ .. .. .. .. .. .. ... ........ ........... . ....... .... .. ..... the subject of Chapter 4.. ...... .. ... ..... .. ... ........... ......... ... ... ... . .. .........2 The indeﬁnite integral dy . t ...... ......... .... .. . . . ... .... ... . ....... ..... Bonaventura Cavalieri (1598–1647)........ The ﬁrst of these...... ....... .... ..................... whose work on inﬁnite processes inﬂuenced Newton and to whom we owe the symbol ∞. ......... .. .... ..... .......... ....... Other contributors include Pascal..... ... . .... ...... .. . ... ............ .... ........... . ....... ..... .... ................ ... ......... .................... . .. . ................ .. Analysi indivisibilium atque inﬁnitorum (Analysis of indivisibles and inﬁnities) in 1686...... .. ........

2) dy where C is an arbitrary constant.2).128 Chapter 5 Integration Z F ′( x ) dx = F ( x ) + C (5. A more comprehensive list of standard integrals is given in the Appendix. if y1=1x 2 then 1=12x. Table 5.1 Elementary integrals 1. 3. d a ln( ax + b) + C = ax + b dx so that Z 1 1 dx = ln( ax + b) + C ax + b a General methods of integration and further standard integrals are discussed in Chapter 6.4 when we discuss the integral as the limit of a sum. 2. 5. C is an arbitrary constant called the integration constant. the function (y1+1C) also has derivative F′(x) dy dC dy d ( y + C) = + = dx dx dx dx (5. Z x a dx Z eax dx = = x a+1 +C a +1 1 ax e +C a 1 a a ≠ −1 Z sin ax dx = − cos ax + C Z cos ax dx = sin ax + C Z 1 1 dx = ln ( ax + b) + C ax + b a 1 a .2). The function to be integrated. for example. 4. Each entry in the list can be checked by differentiation of the right side of the equation. given y1=1F(x) with derivative F′(x).3) Table 5. and the dx indeﬁnite integral of the function 2x is Z 2 x dx = x 2 + C because d 2 ( x + C ) = 2x dx The symbol Z is called the integral sign. For example. x is the variable of integration and dx is called the element of x. its signiﬁcance will become clearer in Section 5. it is an elongated ‘S’ (for summation) and has its origins in Leibniz’s formulation of the integral calculus. is called the integrand.1 is a short list of ‘standard integrals’ involving some of the more important elementary functions (compare Table 4. It is included as part of the value of the indeﬁnite integral because. F′(x) in (5.

.2. Also. with property DF(x)1=1F′(x) The effect of D on F(x) is to transform it into its derivative F′(x). dx is to reverse the effect of differentiation. can be deﬁned whose effect is to reverse that of D. and a corresponding inverse operator D−1. an integral operator. the effect of the operation Z . differentiating both sides of equation (5. makes use of the differential operator D1=1d2dx introduced in Section 4.2) gives d d Z F ′( x ) dx = F ( x ) + C = F ′( x ) dx dx so that the derivative of the integral of a function retrieves that function. Thus D−1F′(x)1=1F(x)1+1C (5. the integral of the derivative of a function retrieves the function.5.. that does not involve Leibniz’s symbolism.1 Indeﬁnite integrals (i) Z Z dx x2 = Z x −2 dx = x −2 +1 x −1 1 +C= +C=− +C x −2 + 1 −1 (ii) dx x ( −1 2 ) +1 x1 2 = Z x −1 2 dx = +C= + C = 2 x1 2 + C = 2 x + C ( −1 2) + 1 12 x (iii) Z dx = Z 1 dx = x + C Z e2t dt = e2t + C Z cos 2θ dθ = sin 2θ + C Z 1 dx = ln (x + 3) + C x+3 1 dx = ln ( x + 3) + ln A = ln A( x + 3) x+3 1 2 1 2 (iv) (v) (vi) We note that if we put C1=1ln1A in (vi) then Z 0 Exercises 1–10 In every case.4) .2 The indeﬁnite integral 129 EXAMPLES 5. Differential and integral operators An alternative way of describing the operations of differentiation and integration.

..... ...... the value of the integration constant is determined by the nature and state of the system... . . . . .. . . . .... . .... ...... ...... . . . . . . . . .. ... . .. . .. . . .. .... ..5) is that y1=12 when x1=11... ... .. . .. . . . . .. .. ... . . .. .. ......... ...... . .... .. . . 2.... . .... .... . .... ..... . .... Then Chapter 5 Integration This equation represents a family of curves... ..... . . ... ..... Consider... ... . . .... .. . . ........ .. ... .... . .... . ........ . ..... . .. . ... .. ... .. . .. a curve whose gradient at every point is given by the function 2x. 1 + . . . ..... .. . . . . .... .. . ... ... . . .. ... .. . .. . . ... . . .... . ... ..... ... . .. ..... .. .. . .. .. . ... . . . . .. . ... . . . .... ... ....... ..... .. . .. .. ... .. ... . ..... ... .. . ... . ... ... . ...... ... . . . . ... .. ..5) . .. .. ........ ... .. ..... . ..... . ... . . . . . ..... .... . . .. ... ... .. . ... . . . ...... .. . . . ... .. .... ... ....... ..... . . .. . . ... . . . ..... . . in a physical problem... 1 ... . . .. . .... .... ..... ....... ... .. . .. ... ... . . .. .. .. . .. . .... .. .. . . ... ....... . .. . .•. .. . . C1=11. . .. . . ( 1 is equivalent to true of any pair of inverse operators...... ... .. .... .. .. . . .. .. . .... .. .. .... ...... .. ..... . . ... .. .. .. .. .. .. .. . . . . ...... . .12) say.... . . . ... and y1=1x21+11 is the equation of the particular curve whose gradient is given by dy2dx1=12x and that passes through the point (1.. ....... . the point (1. .. ... ) ...... .... .. .. . .. .. ....... ..... . . .. .. ..... .. . ..... ....... . . .... . . .. . .. .... .. . . for example... . . ... . .. . .. .. . . .. .. .. .. .2) is determined by some auxiliary condition..... . It is shown in later chapters how.. ... . .. .. ...... . . . .. 2 . . 0. .. .... The operators D and D−1 have the property D−1 D1=1D D−11=11 Z D −1 D F ( x ) + C = D D −1 F ( x ) + C = F ( x ) + C d d F ( x ) + C dx = dx Z F ( x ) + C dx = F ( x ) + C dx y = Z2 x dx = x 2 + C Figure 5.... ... . that is.. . . .... .. .. the auxiliary condition on (5. .. .... .. ... .... .130 If it is required that the curve pass through a particular point.. .. .. .. .. .. . .. ... ..... .. .. .... .. .. .. ... ... .... . . .. . . ... .3). In an actual problem the value of the integration constant C in (5. . . . .... . ... − .. ..... ..... ... ... y1=121=111+1C. . 1... .. . ... . .. . . . Therefore. .. .. ... ..... ........ .. one curve for each value of C (Figure 5.. .... .... . . .. .... ..... ..... .. 0 .. The equation of such a curve is Value of the integration constant is the operator form of equation (5. . .. . .. .. . . ... . . . . ..... . ...... . . . . ... .2). .. . .. .... ... ... 1 −= C (5.. .. .. . ... .. .. ... ...... .... .. .12). . . . . . ... . . .. ... .. .. .. . . . . .3 .. . .. . . . .. ... .. . .... . . .......

12 (ii) 101=1321+1C. For a multiple of a function: Z a u( x ) dx = aZ u( x ) dx 2. y1=1x 21+11 Two rules 1.8) Z f ( x ) dx = aZu( x ) dx + bZv ( x ) dx + cZw ( x ) dx + (5. For a sum of functions: (5. (ii) y1=110 when x1=13. C1=1−4. f (x)1=1a1u(x)1+1bv(x)1+1cw (x)1+1can be written as the sum of integrals (5.5. y1=1x 21−14 0 Exercises 11. We have y = Z2 x dx = x 2 + C .3 Integral of a linear combination of functions Z 3x 3 + 2 sin 3x − e− x dx = 3Z x 3 dx + 2 Z sin 3x dx − Z e− x dx = 3× 0 Exercises 13–15 1 3 2 x4 + 2 × − cos 3x − − e− x + C = x 4 − cos 3x + e− x + C o 3 4 4 3 ( ) ( ) .7) (5.9) EXAMPLE 5.10). C1=11.2 Find y = Z2 x dx subject to conditions (i) (x.6) Z u( x ) + v ( x ) dx = Z u( x ) dx + Z v ( x ) dx It follows from these rules that the integral of a linear combination of functions.2 The indeﬁnite integral 131 EXAMPLE 5.1y)1=1(2. Then (i) 01=1221+1C.

.. .... .. ....... . ... ..... . ............ ... .......... .... . In this section we introduce... . ... . . ........ . .. ........ ......................... ... . ... .. .. . . ... .... .... .. .... . ...... . ........... .. . .. .. ......... . and show how it is related to the indeﬁnite integral...... . ...... the average height is x ... .. .. .. ............ ....... . ... .... . ..... as in Figure 5... . .... . ... ........ . .. ... ..... . ... ..... . . . .... multiplied by the average height..... .. .. ... .. ... . . . .. ..... . .... .... .... .. . .... .... .. .......... .... . . ... .. .......... ... .... ............ ....... .. ..... .... ... .... ... . .. . . . ... .... ............... .... ... .. . . . (b1−1a)....... .... .. . ... .... . ... ... .. .. .. .. .. .. .. ......................... ............ . .. ...... ....... ... .. ......................... .. ... .... ... ........ . . . .. .. ... .. ... .. ...... . ... ... ............... .... ....... . ... .. .... .......... .. .. . .... ..... .... .... . . ... . . ..... ...... .. . ... ........... . .. ... .....4.. . .. .... . .... .. .. .... . . .. . .. .... .... . ........ . . ..... . . . .. ... .. . .. . .... . ........... . .... y . . . . ... ......... ..... . . ... .. .. .... . .. ... . ......... 132 The integral calculus was invented to solve the problem of ﬁnding the area enclosed by a given curve. ..... . . .... . ... . .... ..... ... . ... .... .......... . . ........ b )b( f )x( f = y . ..... ........ . ........ ..... ...... . ..... .. .. . .. .. ... .... .. . .. this is the area enclosed by the graph of y1=1f (x).. . ....... . .... .. .. ...... .. .. .. In general... ... .... . .... . . ... of the curve above the axis (the average value of the function between a and b).. .... . .x b . . . ... .. ...... . . .......... The shaded area in Figure 5. .... .. ... . . .......... ........ . ..... . ... ...... .... ........ ... . .... .. ... ........ .... ..... ..... ... o ... .. . . ... . .. ... ... ... .. .. .. . ... . . .... ... . .... .. . . .. ...... . .. .. . ... . ......... ...... ........ ... .. . . . ... ....... . . ..... .... .. ........ ... ....... .. ..... ..... . .... .... . ... ... ..... . .. ....... . .. ......... ...... ........ . ... ............... .... ...... . . .. .... ..... ... . . .. . .... .. . ..... . . ..... .... . ... . .. ... .. .. . .. ...... . . .... . .. .... . . ... ... ..... ........ . the deﬁnite integral as a measure of area........ .. . . ...... ..... .... .. ... .. ... ... . ................. .. ..... ....... . . . . ...... . . ..... ..... ... ... . . ... Other problems however require a more intimate understanding of the integral calculus.. .. .... . . . . ..... .. . . .. .. .. ............ . ...... . . . . .. .. ....... . . . .. .......... . ............ ..... . ... .... . . ... . ..... ......... .. .... ... . ... . .. ... ........ ..... and the verticals at x1=1a and x1=1b.... .. .. ..... ... . ..... ... ... ... . .. . ........ .. ... ... and we will return to this in Section 5... . ... ... ..... ....... ......... .... ............ ... .. .... . ... . ..... ... ..... In the special case of a linear function. .................... ..... ... ... . ... .... ... .. . . ....... . .... .. ... . .. y a )a( f o ...... .. .. ...... ....... ..... /)b + a( 2 y )b( f ........... ... ... . . .... ..... .... . ....... ...... ....... . . .... . ...... .. . this area is equal to the width......... .. ... .4 Figure 5. ..... . . ... .. . . . ......... .. .. ... .. .... . .. . . ... ..... ... .. ............. .. .. ... . . . .... ... ........ ... .. . ... . . . .. .... ... .... . .... ......4 is known as the area under the curve........ . ... ..... ......... .. .... ...... .. ..... .... ........ ... .. . . .. ..... .....5.. .. .. . ......... ... . . .. y (5... ..... . .. . ... ... . . . .. ... .... ... ... .... .. . ....5 a )a( f . ... . .. . . . . . .. . ... ... ............... ...... .... ... Let y1=1f (x) be a function of x.. .. .. . ..... Chapter 5 Integration 5... ...... . .... .. ..... .. . ... the x-axis... .3 The deﬁnite integral y= A = (b − a) y 1 f ( a ) + f ( b) 2 Figure 5... .. ....... ... ..... ... ..... . ............ ....... .. ... ........... .. . .............. .. .. . . ......... ... . ... ... .. .... . .. ... ... . ... .. ...... . ... . .. .. ....... . .. ......... . . . ... . .. . ... . ...10) . ......... .. . .... ... ... ... ..... . .. .. .. . . .... ... .. . ..... ...... ..... For many problems in the physical sciences.. .. ... ... ....... .......... .... . . continuous in the interval a1≤1x1≤1b. . . .. ... ... ... .. .. ... ......... .. ..... .. .......... . .... ....... .. .. .. and the area under the curve (straight line) is that of a rectangle of width (b1−1a) and height y ..... . .... .... ........ .... . ......... . ........ . .... . ... ..... .... ...... . .. ... . .. ...... . . . .. .... ... . . .. . . .. ............. . ... .... .. .... without proof..... .. this brief introduction is sufﬁcient. ...

so that a b Z f ( x ) dx = F ( x ) = F ( b) − F ( a) a a b (5. the integral calculus (see Section 5. and the interval a to b is called the range of integration. let y1=1f (x)1=12x1+13.5.11) is often denoted by F ( x ) .4 Deﬁnite integrals (i) Z ( 2 x + 3) dx = x 2 + 3x = ( 42 + 3 × 4) − (12 + 3 × 1) = 28 − 4 = 24 1 4 4 1 (ii) Z x 2 dx = 2 3 x 3 3 33 23 19 = − = 3 3 3 2 3 −2 (iii) Z 4 dx x2 =Z x 2 4 2 1 4 1 1 1 1 1 dx = − = − − − = − + = 4 2 4 x 2 4 2 . where F(x) is the function whose derivative is f ( x ) = F ′( x ) = b The difference F(b)1−1F(a) in equation (5. The indeﬁnite integral is Z f ( x ) dx = Z( 2 x + 3) dx = x 2 + 3x + C = F ( x ) The deﬁnite integral of f (x) in the range x1=1a to x1=1b (‘the integral from a to b’) is then Z ( 2 x + 3) dx = x 2 + 3x + C = ( b2 + 3b + C ) − ( a 2 + 3a + C ) a b b a =1(b21+13b)1−1(a 21+13a) We note that the constant of integration C cancels for a deﬁnite integral.11) dF . The numbers F(a) dx and F(b) are the values of F(x) at the limits of integration a and b.4) tells us that the area is given by the deﬁnite integral A = Z f ( x ) dx = F ( b) − F ( a ) a b (5. and can always be omitted. when y1=1f(x) is not necessarily a linear function. b the upper limit. For example. a is called the lower limit.3 The deﬁnite integral 133 In the general case.12) It follows that in order to calculate the value of the deﬁnite integral it is normally necessary ﬁrst to evaluate the corresponding indeﬁnite integral. EXAMPLES 5.

10) that the average value of the function y1=1f (x) in the interval a1≤1x1≤1b is A y= = b− a Z b a f ( x ) dx Z b a (5. by equation (5. The value of the integral does not depend on the symbol used for the variable of integration: Z f ( x ) dx = Z f (t ) dt = Z f (u) du = a a a b b b (5.5 Find the average value of sin1θ in the interval 01≤1θ1≤1π22. .4(v). the value of the integral is F(b)1−1F(a).11).134 Chapter 5 Integration (iv) Z dx = Z 1 dx = x = b − a a a a π 2 b b b (v) Z 0 π 2 π sin θ d θ = − cos θ = − cos − − cos 0 = (0) − ( −1) = 1 0 2 ( ) (vi) Z e −1 3 +1 −2t dt = − 1 −2t 1 1 1 e = − e−2 − − e2 = e2 − e−2 2 2 2 2 −1 +1 (vii) Z 2 3 dx 3 = ln x = ln 3 − ln 2 = ln 2 x 2 0 Exercises 16–25 Average value of a function Because the deﬁnite integral is identiﬁed as the area under the curve it follows from equation (5. sin θ = 1 2 Z sin θ dθ = π 2 0 π π 2 0 Exercises 26–28 Three properties of the deﬁnite integral Let f (x)1=1F′(x) so that. By Example 5. Z f ( x ) dx = F ( b) − F ( a) a b 1.14) In each case.13) dx EXAMPLE 5.

30 Negative areas Consider the integral Z sin x dx = − cos x 0 2π 2π 0 = ( − cos 2 π ) − ( − cos 0) = ( −1) − ( −1) = 0 .6 Properties 1 1 (i) A = Z x dx + Z x dx = (32 − 22 ) + (52 − 32 ) 2 2 2 3 1 = (52 − 22 ) = Z x dx . 3. If c is a third limit of integration. 3 3 2 ( ) Z x 2 dx = 5 2 ( ) 5 0 Exercises 29.15) is equal to the sum of the areas represented by the integrals on the right side. the value of the integral changes sign: a b Z f ( x ) dx = − Z f ( x ) dx a b (5. When the limits are interchanged. not necessarily between a and b: Z f ( x ) dx = Z f ( x ) dx + Z f ( x ) dx a a c b c b (5. 2 2 5 3 5 (ii) Z x 2 dx = 2 5 1 3 3 117 5 −2 = . EXAMPLE 5. 3 3 1 3 3 117 2 −5 = − = − Z x 2 dx.16) This follows because F ( a ) − F ( b) = − F ( b) − F ( a ) .15) This is true because the value of the integral can be written as F ( b) − F ( a ) = F ( c) − F ( a ) + F ( b) − F ( c) If c lies between a and b then the area represented by the integral on the left side of equation (5.5.3 The deﬁnite integral 135 2.

. . .. . . . . . . . . . . . A . . . .. . . . . . ... ... ...... . .... . .. . . . .. . .... . . . . .. . Figure 5.. ... . ... .. .... . . .. .. . .. . .. . . . . . .. . .... . . . .. .. .. 0 Exercise 31 Integration of discontinuous functions The function 2x f ( x) = 2x + 1 if x < 2 if x ≥ 2 .. .. . . . 2 ..... .. .. .. . . .. . . ... . . ... . . . . . . .. . . .. .... .. . .. . . . . .... . .. . .. . . . ....7 Z f ( x ) dx = Z f ( x ) dx + Z f ( x ) dx 1 1 2 3 2 3 2 3 = Z 2 x dx + Z ( 2 x + 1) dx = 3 + 6 = 9 1 2 This can always be done when the integrand has only a ﬁnite number of ﬁnite discontinuities within the range of integration. π . .. .. . .... . . . .. . . . . . . ........ .... . . . .. .. .. . . ... A x nis . .. ... . . ... .. .. . . . .. .. . .... .. .. . has positive values when 01<1x1<1π. . . ..... .. . .. . .. ..136 Chapter 5 Integration Figure 5. . . . .. . .. .. .. . .. . . . . . ... . . . ... .... . . .. .. . . .. .. . .. .. . .6 Figure 5. .. .. . .. . . . . . . .... . . In this case the positive and negative contributions cancel. . . A similar technique is used when the function is continuous but has a discontinuous derivative.. . . . .. . ... .. . .... .. .. . .. . . and the average value of sin1x in the interval is zero. Thus. .. .. .... .. . ... . . . . .... . .. .. . ... .... . . .. .. . .. . . ..... . . . .. .. .. . . . .. .. . . . ... . . . but Figure 5. . . . . . .. . . .. . . ... . ..7 shows that the function can be integrated across the discontinuity if the range of integration is split at the point of discontinuity... . . 1 . . ... .6 shows that the integrand. . . ... 3 2 1 0 is discontinuous at x1=12.. .. .. ..... .. .... π2 . ... ... ... .... .. .. . and the integral can be written as Z sin x dx = A1 + A 2 0 2π where A1 = Z sin x dx = +2 0 π and A2 = Z sin x dx = −2 π 2π This example shows that areas corresponding to negative values of the integrand make negative contributions to the total... .. .. . . . .. .... . . .. . The total ‘area under the curve’ is the sum of the two areas labelled A1 and A2 in the ﬁgure. . . . .... .. . and negative values when π1<1x1<12π. . .... . ..... O − 1 + 1 . .. ..... ... 2 4 6 8 x . .. . ..... ... . ... . . . . . ... . . ...... . . .. . sin1x..

. . . .. .. .. .. . ..... .. . . . ... .. .. . . . ... . . .. . .. ...... ..... .. . . .. . ..... . .. ... . . .. . . . . .. . .. .. .. . . .. .. . . .. . . . .. .. . where a1≤1c1≤1b. . . .18) 137 a . ... ... .. . b 0 a− Improper integrals 0 Exercises 32–34 If the discontinuity lies at one end of the range of integration. . .. ... . . . .. . . . . ... .. ..8 5.. .. . . . . . ... . . . ... . .. .. .. . . . .. . ...... . .. .. EXAMPLE 5.. . .. . . .. . .. . . . .. ... . . . the slope of its graph changing discontinuously from +1 to −1 on passing through x1=10 from left to right (Figure 5.. . ... .. . .. .. . ... ... .... .. .. . .. ... . .. . . . . .. . . .. . . .. . . .. . . ... . . .. . . .... . .. ...... ... .. . . for ε1>10. . . . . .. ..... . . . . ... .. .. . ... ... . .. ..... . .... . .. . . . . ... . . . . . As shown in Figure 5.. .. .... .. .. . . . . ... .. .... . ... . . a b ) ( if x ≤ 0 if x ≥ 0 0 b Figure 5.. .. the point c is excluded because the integrand is not deﬁned there. .. . . .. . . ... .. .... If the discontinuity is at the point x1=1c.... .. . .. . . . .. .. . . .17) is ﬁnite and unique then the value of the integral is the ‘area under the curve’. ... . .. .... . . .. . .. . ... ... . .. ... . ........ .. . .. . . . then the integral is deﬁned as the limit....... .. . .. .... . ... . . . . .. . .... . .. .. ... .. . .. . . .. . . . . .. .. .. .... .. . . .. . .. .... . .. .. . .. . . .. . .. . . .. .. . . . . .. .. ... . . . . .. ... . . . .. . . .. .9 ε →0 ) a+ε b 0 b f ( x ) dx Figure 5. .. ... .. . ..... .. . .. . . ... . . . ... ..... . . .. . .. ... .. . . . . . . .. ..... . .... . . . . . . .. . .. . . ... . . .. . . . . .. .. .... . . . .. . . . .. . . . .. . .. .... ... . . . .. ... .... . ... .. . . .3 The deﬁnite integral (5. .. .. ... ...... . . .. .... .. .. . . . . . . . ... . . . . .. .. ... . ... .. . . . . ...... .. . . . . . . ... . .. . . .. . .. . .. ... ... . ... . ... . . .. . . . . ..... . . ...8). . . .. . . .. .. .. . . . . . . ... .. . .... .. . . ... .. the integral is deﬁned as b .. . ....... +c . . . . . . .. . . ...... . ... .. ... ..... .... . .... . . . .. ... . . . . . .. . . .. ... . .. .. .. . . .. ..... . . at x1=1a say... .. . ... .... . .. . . . .. . .. . . .17) (5. . . .... . ... .. . .. . . . . ... . . .. .. . . . .. .. . .. . . . When the limit in (5.. . .. .. . . . . .. . . .. . . . .. .. .. . .. . but not smooth. ... ... ........ .. ... . ... . . .. .. ... . ....... .7 The function 1 is continuous at x1=10. .. .... . e− x f ( x ) = e− | x | = ex Z e−|x| dx = Z e x dx + Z e− x dx = e x + − e− x −a 0 −a b b c−ε Z f ( x ) dx = lim Z f ( x ) dx + Z f ( x ) dx ε →0 a a c+ε b = 1 − e− a + 1 − e− b = 2 − e− a − e− b ( −a 0 Z f ( x ) dx = lim Z c −c . . . .... . . . .. . . .. . .... ... . . ..... . ... .. ... ... ... .. . . ... .. . .. ... . ... . ... .... ... .9.... . .. .. .. . .. . . .. . .. . . .. .... .. .. .. . . . . .. .. .. .. . . . ... . .. . . .. ...... . .. . ...... ... .. . . ........ . .. .. .. The function can be integrated if the range of integration is split at the point of gradient discontinuity: A deﬁnite integral is called improper when the integrand has an inﬁnite discontinuity at a point within the range of integration... . . . .. .. . .. ... . . . . . . ... . . . .. . . . ..... .. .. ... ... .

but when a1>11 the limit is inﬁnite and the integral is not deﬁned. a1≠11 Z 1 0 dx 1 1 1 1 1 = lim a −1 = lim 1 − a −1 x a ε → 0 1 − a x ε 1 − a ε → 0 ε When a1<11 the limit has value 12(11−1a). letting ε1→10.138 Chapter 5 Integration EXAMPLES 5. If the upper limit is inﬁnite then the deﬁnite integral is deﬁned by ∞ a Z f ( x ) dx = lim Z f ( x ) dx b→∞ a b (5. (iii) In the general case of the integral of an inverse power. 0 Exercise 35 Inﬁnite integrals It often happens in applications in the physical sciences that one or both of the limits of integration are inﬁnite. Integrals with inﬁnite ranges of integration are called inﬁnite integrals. Z Z 1 0 1 0 dx =2 x 1 (ii) 1 dx dx = lim Z = lim ln x = lim ( − ln ε ) ε ε →0 ε →0 ε x ε →0 x The limit does not exist because ln1ε1→1−∞ as ε1→10.19) .8 Improper integrals (i) The function 1 x is not deﬁned at x1=10 and the deﬁnite integral between limits x1=10 and x1=11 is deﬁned as Z Then 1 0 dx dx = lim Z ε →0 ε x x 1 Z 1 ε dx 1 = Z x −1 2 dx = 2 x1 2 = 2 − 2ε1 2 ε x ε 1 and.

but some care must be taken when assigning the inﬁnite value to the variable. ( x − 1)( x − 2) The integrand can be written in terms of partial fractions (see Section 2.3 The deﬁnite integral 139 For example. the integral is said to be convergent. ∞ 0 Z e− r dr = lim Z e− r dr = lim − e− r b b 0 b→∞ 0 b→∞ = lim − e− b + 1 = 1 b→∞ using the property of the exponential that e−b1→10 as b1→1∞. For example.19) is ﬁnite and unique. ∞ Z 1 b dx = lim ln x = lim ln b 1 b→ ∞ x b→ ∞ is divergent because ln1b1→1∞ as b1→1∞.9 Find the value of the inﬁnite integral I = Z 4 dx . A more subtle example is Z cos x dx = lim sin x = lim sin b 0 b→ ∞ b→ ∞ 0 ∞ b . inﬁnite integrals are treated in exactly the same way as ordinary integrals. An integral is divergent when the limit is indeterminate.7) as 1 1 1 = − ( x − 1)( x − 2) x − 2 x − 1 Then I = lim b→ ∞ Z 4 b 1 b 1 ) − dx = lim ln( x − 2) − ln( x − 1) 4 b→ ∞ x − 2 x − 1 x − 2 b b − 2 2 3 = lim ln − ln = ln = lim ln b→ ∞ x − 1 b→ ∞ b − 1 3 2 4 because (b1−12)2(b1−11)1→11 as b1→1∞ and ln111=10 0 Exercises 36–39 When the limit in the deﬁnition (5.5. ∞ EXAMPLE 5. In practice.

.. ... ...... . .. .. . .. . On the other hand.. ... and no unique value can be assigned to the integral as deﬁned here. . . y . . . ... . .. . .. . . and it follows that ∞ =0 Z e− ax cos x dx = lim a→0 0 a → 0 1 + a2 lim We note that the limit must be taken after integration. .. .... . . . ...... . . .. .. . ... ... . . noitcnuf neve . Figure 5. . .. ... .. . . .. .8) and cos1x (shown in Figure 5. . .. ... . .. . • • .. . . . .. . .... . ..... . ... ... . .. . .. . . . .. . ..... .. ..10 π n o i t c n u f d d o . ... .... + 1 1 − x− • • . . .. . and divergent. and x1cos1x.. . .... .. • .... ...20) it is called an even function of x. ... . . . . .21) is said to be an odd function. Examples of even functions are x2... . .. . .. in each case the value of the function changes sign when x is replaced by −x.. Thus x1cos1x is an odd function... ... . . .. . ∞ Z e− ax cos x dx = 0 a 1 + a2 This result is valid for all positive values of the parameter a... .. . . . .. .... ..... . .. .. . . .. . .. . .. ... . . ... . .... ... .. . ..x soc = π . ... . .... ....... . .. .. .. . .. .. .. . .. it is shown in Example 6... .. Even and odd functions a When a function has the property f (−x)1=1f (x) (5. .. ... .. .. .. . ... . .. ........ . . ...... . .. ... . . . . . .. . . . ... it has odd parity (or parity −1) and is antisymmetric with respect to the axis x1=10. .. . • .. . ..... ... .. . . . .. ... ... . .. . . . .. ... .. .... . . ... .. The product of two even functions or of two odd functions is even. ...... . . .. . .. .. ... .. . .. . . . .... ) a( π− . ... . . . .. )b( . .... sin1x (shown in Figure 5. . .. .13 that.... . .. π− . . ... • ....... . . . . . ... e−| x| (ﬁgure 5. . ... ... . .x n i s = .. . .. . ... . . ... . . .... . . . . ...... . . . ..... . with x odd and cos1x even.. .. . . .. ... in each case the value of the function is unchanged when x is replaced by −x..... . .. . . x . .. .. . ... however small. ..... . . . . . . . . . . .... . ... .. . . . .. ... .. On the other hand... .. y . ... . ... . .. .. . .. . ... . . . ..140 Chapter 5 Integration In this case the value of sin1b oscillates between +1 and −1 as b increases.. . .... .. . . . . . .. .. . . . 0 − 1 + 1 . ... . ... .. .... .. . . ... . ... ... taking the limit before integration leads to a different. .. . . . ... . . . ..... . ... for a1>10.... ... .. . .. ... . . . Examples of odd functions are x3.. .. . . . . .. .. .. . . .... ... . . . . .. .. 0 ... ... . .. x .... .. . a function with the property f (−x)1=1−f (x) (5.... .. . . .. . . ..... .. . ... .. . integral.. . . x− . . • .. .. . . . . it has even parity (or parity +1) and is symmetric with respect to the axis x1=10 (the y-axis)... ......... . . .. .. ..10a).. ...10b).. ... the product of an even function and an odd function is odd..... . ... ...

and the value of the integral is zero: Z +a f ( x ) dx = 0 if f ( x ) odd (5. It is always possible however to express a function as the sum of an even component and an odd component. an arbitrary function is neither even nor odd. f (−x)1≠1± f (x). because every even power of x is an even function whilst every odd power of x is an odd function.3 The deﬁnite integral 141 In general.25) 0 On the other hand. we can write f ( x) = 1 1 f ( x) + f (− x) + 2 f ( x) − f (− x) 2 (5.5.23) −a is the sum of the areas to the left and right of the x1=10 axis: A=Z 0 −a f ( x ) dx + Z +a 0 f ( x ) dx = A < + A> (5. For example. if f (x) is an odd function of x. the ﬁrst set of terms forms the even component of the polynomial and the second set forms the odd component.24) If f (x) is an even function of x. the integral is equal to the integral of the even component: Z +a −a f ( x ) dx = Z +a −a f + ( x ) dx + Z +a −a f − ( x ) dx = 2 Z +a 0 f + ( x ) dx (5. if f ( x ) even (5.22) =1f+1(x)1+1f−1(x) where f+(x) is the even component of f (x) and f_(x) is the odd component.26) −a In the general case. then A< and A> are equal in magnitude but have opposite signs: A<1=1−A>. the two areas A< and A> are equal in magnitude and sign. The integral properties of functions of well-deﬁned symmetry are of great importance in the physical sciences.27) . an arbitrary polynomial f (x)1=1a01+1a1x1+1a2 x21+1-1+1an x n can be written as f ( x ) = a0 + a2 x 2 + a4 x 4 + + a x + a x 3 + a x5 + 3 5 1 and. The area represented by the integral A=Z +a f ( x ) dx (5. and the total area is twice each of them: Z +a −a f ( x ) dx = 2 Z +a f ( x ) dx. when f (x) has no particular symmetry.

.142 Chapter 5 Integration EXAMPLE 5.3 that the ‘area under the curve’. continuous in the interval a1≤1x1≤1b.10 (i) Find the even and odd components of f (x)1=1e x. In solid-state chemistry.28) and divide the area into n strips by vertical lines at these points as shown in Figure 5.4. 0 Exercises 40–47 The concept of the parity or symmetry of functions is widely used in the physical sciences. (ii) evaluate the deﬁnite integrals of f (x) and of its even and odd components over the range x1=1−1 to x1=1+1. and are used to explain and predict the allowed transitions between energy levels that are observed in molecular spectra. the relevant branch of mathematics is called group theory. is given by the value of the deﬁnite integral of f (x) from a to b. We now look at how this result is derived.4. the shaded region in Figure 5. We postulated in Section 5. In molecular chemistry. (i) even: odd: f+ ( x ) = f− ( x ) = 1 −x 1 x f ( x) + f (− x) = 2 e + e 2 1 −x 1 x f ( x) − f (− x) = 2 e − e 2 These are the hyperbolic functions cosh1x and sinh1x (Equation (3. the space groups describe the symmetry properties of lattices and the structure of X-ray diffraction spectra.47) and Figure 3. To obtain an estimate of the area.4 The integral calculus Let y1=1f (x) be a function of x. (ii) Z f ( x ) dx = Z e x dx = e x = e + e−1 −1 −1 +1 +1 +1 −1 +1 1 Z f+ ( x ) dx = Z e x + e− x dx = e x − e− x = e + e−1 −1 −1 +1 +1 2 2 −1 −1 +1 1 Z f− ( x ) dx = Z e x − e− x dx = e x + e− x = 0 −1 −1 +1 +1 Only the symmetric component makes a nonzero contribution to the integral of f (x). with a1=1x01<1x11<1x21<1-1<1xn−11<1xn1=1b (5. 5.11. we divide the interval a to b into n subintervals by choosing n1−11 arbitrary points on the x-axis. for example. as in Figure 5. the symmetry properties of molecular wave functions and of the normal modes of vibration of molecules.22). the point groups are used to describe.

..... . . . ... .. . . .... . .. . .... . .. . ... . . . . . .. ..... If we decrease the widths of all the strips by increasing the number of subdivisions. .. ..... . . . . .. . .... ... ..... . . . y . .... . . .. . . . .. . .. . ... .... . . . .. . . .. . . . . . ...... . . . . . .. . ..... ..... . . .... ... . . . . . ..... . .. .. ... . . . .... . . . .... .. . .. ... . ... . .. . .. . .. . . . .. .. . .. . . . ... . . . . .. .... . . . . .. .. . ..... . . .. . .. .. . . . ... . . . . . . . . .. . .. . . .. . ... .... .. . ..... ......... . . . . ... . .. . .. .. . . . .. . .. .... ... . .. . ... .. .. ...... .. ... . .. . . . . ..... . .30) converge to the same limit as the number of strips is increased indeﬁnitely. .. . .. .. . . . ..... . ... .... .. .. . .... . . ........ ... ... .. .. . . . . . . .. . ... . .. . ... . .29) (5.. . . .. ... . . .. .... .. . we have An estimate of the area is then obtained by replacing each strip by a rectangle. ..... ... . ..... .. . . . . . . . . . . .. . . . .. . .. ... ... . ... .... . . .... . ..... . . .. . . . . ...... .. . . .. . . . . ... . .... . . . .. .. . . . . . ... . . ... . . . .. ... . . . ... . .. .. .. .. . . .... . .. .. .. . . ..... ..... ...... ...... . . .. .. . ... . ..... . . ... . .... ... . . . . .. ... .. .. ..... . . . ..... The area of the rth rectangle is ∆Ar1=1yr ∆ xr ... .. . . . . ... . . .. . . . . . .. . . .. ... . . ... . . .. . . .. . .. . . .. . .. .. .... ... . ... .... . . . . . ... . .... . . .... .. .. . .... . . . ..... ... . .. . . .... . . . .. .. .. . . . . .. . . . . . .... . . . .. .. . . . .. . . . ... .. . . . .. .. .. . .. . . ... .. .. .. ... . ..... . .. . . .. ...... . ... . .. . .... . ... ... . . . ... . .. . . ....... . .. ... ... .. .. .. . .. .. . . . .. .. .... . . . . ... . . . .. . . .. . . . Therefore.. .. . . .... . .. .. .... ......... . .. . . . ... .. . . . .. . .. . ..... .. .. .. .... . . x Two ways of choosing the heights of the rectangles are shown in Figure 5.... . .. .. .... .... ... ..... . . .. . ... . . . ... . It is clear that the ﬁrst choice gives a value that overestimates the area under the curve. . . .. .. . .. .. . . . ... . .. ... . .. . . . ... . .. .30) 143 )a ( 1 −rx ··· 2x 1x a .. . .. . .... ... . .. .. ......... . . .. . .. ..... .. . . .. . .. . .. . . . . . . .. .. .. . . .. .. .. . . . . . . ..... .. .... .. . ..... . .. . . . .. ..... . .. ..... . .. . . . .. .. .. . ... .. ... .. . .. . . ... .. . .. .. . .... . .. .. . . .. . . .. . . . ... . ... . ... .... .. .. .. ... .. . ... ........ . .. . . .... . ... . .4 The integral calculus (5. . .. ...... . ... .. . . .. ... ... .. . . ... ... . . . . .. .. .. .... .... .. . . .. . . . ... . . . . .. .. ..... . as has been assumed) and the two sums in (5.. .... ... . .. .... .. . ..... . . ... . ..... . . .. . . ... .... . . ..... .. . . ... . . . . .. ... .. . .... . . ... . .. . . .. irrespective of the particular choice of heights of the rectangles and of how the divisions of the interval are made...... . . . .. . .......... . ... ... .... .. . . . . . . or. . . .. .. . . .... . . . . .. .... ... . . .. ..... . . .. . . . ..... .. . . .... . . ... . .. . . . . .. . . .. ... ... . . . . . . ....... . .... )x( f =y ... .. . ... .... . . .. . . . . .... . .. . . .... .. .. . . .. . . .. . ... .. . The width of the rth strip is ∆xr1=1xr1−1xr−1 and we can choose the height yr of the rectangle to be any value of the function in the strip.. .. ... . . . . .... ... .. . ........ ..... ... ... .. .. . .. In Figure (b) the height has been chosen to be the smallest value.... . .. . ... . . . .. . .. . ... ... . ... . ... ... . . ... ... .. . .. . . . . .... .... . .... . . . . ... . . . . .. .... . . . . ... .. . . .. ... . . ... . .. . .. .. In Figure (a) the height of each rectangle has been chosen to be the largest value of y1=1f(x) in the subinterval.... . . .. . .11 r =1 n r =1 r =1 n n (all ∆ xr → 0) 5.. . .... .... ... . ... . . .... . . we call this quantity yr(min) for the rth strip. ... .... . . . .. ..31) (5. ....... . . ... .. we call this quantity yr(max) for the rth strip. . . . . ... ... . ... . . . . . . .. ... .. ... . .. . . . .... . .. .. . . . ... . . ..... . . . . . .. . . . . .... . .. .. . .32) (5. .. .. . ... . ... . . .. . . ... . . . .. . ... . ...... . .. .... . whereas the second choice gives a value that underestimates the area: )b ( b 1 −nx ··· rx 1 −rx ··· 2 x 1x a b .... ... .. .. ... . . . . . . .. .. . .. .. . . . .. .. . . .. .. . .. ... . . ... . . . . .. . . as n1→1∞ each sum converges to the limit A. ... . .... . . ... . ... . . . . . .. ..11. .. ... ... . .. the area under the curve. . .. . .. . . . .. ..... . ..... ... . ... . .. . .. . .. . .... . .. . . .... . because we can choose yr1=1f (xr). . ... . . ... .. .. . ... . . .. .. ... ..... . . .. . .. .. ... ... .. ... . . ...... . ... .. .. .. the values of yr(min) and yr(max) in each strip approach each other (if the function is continuous. . .. .. . .. . . ....... . ... .. ... . ..... .. . ... . ... .. . .. .. . . .. .. . . ... ... ... . . .. . . ... and the total area is 1 −nx A = lim ··· ∑ r =1 rx n n → ∞ r =1 n yr (min) ∆ xr < A < ∑ yr (max) ∆ xr A ≈ ∑ ∆ A r = ∑ y r ∆ xr ∑ f ( x r ) ∆ xr A = lim r =1 n ∑ yr ∆ xr n→ ∞ Figure 5. .. .. ..

.. .... . . . . . . .... . by equation (5........... .. ..... . . .. . .... ... .. . . . .. ...... . . . . ... .. .. .... ... .......... .33) and is called the deﬁnite integral of the function f (x) from x1=1a to x1=1b.... . . . . . . .. . . ... ... . . .. . . . ..... . ... .. .. .. .... .. .. . .... .... . ..... . ... . .. ... . . . . . ..... ... .. ........ . . . .. professor of mathematics at Göttingen... . .... . . . . .... ... ... .. . . .. ∆ . ... . ..... .... .... . . . .... ... ...... .. .. . .... .. .. . .... . .. . .. .. . .. . .. . . ... . . ... This taking of the limit of the sum of small quantities is the characteristic feature of the integral calculus.. ... .. ...... . functions of a complex variable. .. . . .. . .34) This synthesis of the integral and differential calculus is called the fundamental theorem of the calculus. .. . . ... . ...... . .. . .... . ... . . . EXAMPLE 5. .. . . .... ... . . .. . .. . . . .. .. . . ... x1+1∆x is ∆ A ≈ y ∆ x = a 2 − x 2 ∆ x and. . . . . .. .. ... made independently by Leibniz and by Newton. . . ...... . . . . .. . .. . ... .. . . .. . .. .. The deﬁnite integral as deﬁned in this section is known as the Riemann integral. . .... ... .3 In calculating the area enclosed by a curve it is neither essential nor always convenient that the area be divided into linear strips in the way described above. was that if F(x) is a function whose derivative is f (x)1=1F′(x) then Z f ( x ) dx = Z F ′( x ) dx = F ( b) − F ( a) a a b b (5. . ... .. . ...12).. ..... . . . .. . .. . .. ... . . . ..... . .. .11 with two different ways of calculating the area of a circle. . . ... ... .... . . . . The essential discovery...11 The area of a circle The equation of the circle of radius a and centre at the origin of coordinates is x 21+1y 21=1a 2. .. the total area is a A=Z 0 a 2 − x 2 dx Figure 5.... .33) above.... . . . . .. . . . .. .. . .. . .. ... ... .33). . . .. .. .. . .. . . ..... .. ..... ... .. ... . .. . . ... . .. . made contributions to the theory of numbers. ... Method 1..... . ..144 Chapter 5 Integration Following Leibniz.. .. . . . ... . .... . ... . .... ... .... ... . .. . . ..... . . .. . . . In his 1854 lecture Über die hypothesen welche der geometrie zu grunde liegen (On the hypotheses that lie at the foundations of geometry) Riemann developed a system of non-Euclidean geometry and initiated the study of curved metric spaces that ultimately formed the basis for the mathematics of the general theory of relativity. ... .... . .... . ...... . ...... ... . . . . .. . .. . . . . . . .. .. ... .. .. this limit is written as Z f ( x ) dx = lim a b n → ∞ r =1 ∑ f ( xr ) ∆ xr n (5.... ... An approximate value of the area of the strip between x and .. . .. . . . . ... .. ... ... .. . . . .. . . .... .. . . . .... .. .. ..... .. . .. We illustrate this point in Example 5...... . .. Then y y = a2 − x 2 for 0 ≤ x ≤ a.. .. .. ... and differential equations..... . . .... . . . ..... .. .. ... .. ... . .... . . . . .12 3 Georg Friedrich Bernhard Riemann (1826–1866). .. .... .. . ... . in which both x and y are positive (Figure 5.. .. . .. .. ..... .. . .. Let A be the area of that quarter of the circle that lies in the ﬁrst quadrant.. . .. ..... . . . .. .... . ... .. . .. . . . .. . . ... . .. .. .. . .. . ... .. . . . . . . ..... .. .. . . .. . . .. just as taking the limit of the ratio of two small quantities is characteristic of the differential calculus.. . .. x a x +x x 0 a Divide the area into n vertical strips as described in the derivation of equation (5. ... . . ... . . .. ...... ..

. .. . ...... ... . . .. .... . ... .. ... .. .36) The use of differentials A convenient alternative approach to the deﬁnite integral as an area makes use of the concept of the differential introduced in Section 4.. and can be replaced by the symbol C: x Z f ( x ) dx = Z f ( x ) dx = F ( x ) − F ( a) = F ( x ) + C a (5. ... . .12.. ...10. .... ..... .. .. . . . .. .. .. ... . ... .. ....... .. .. . .. in the language of differentials. . . .. .. . . ... . . . ... Division by ∆r gives 2π r < and... Consider the expression (5... .. .......... .. .. ..... .. . .. . .... . .. . ..... .. A = πa 2 4... . . ... .. .. ... ..... . . .. .. ... .. ... .. ... equation (5... .. . . .... ... . ....... . The corresponding differential dA1=12πr1dr is.. . ... .. .......... . . ... .. . . . . ... .... .. .. .13.... . . . . . .. .... .... . in the limit ∆r1→10. . .. ... .. . .. .... . .. . . . . ..... . .. . ... ... .. .. . ..... ... . ... . .. .. . . . . . . . ... . . . Method 2... ..... . . . . .... . ... ..13 A = Z 2 π r dr = π a 2 0 0 Exercise 48 We note that the indeﬁnite integral can be viewed as a special case of the deﬁnite integral. . . . . ...... . . . and the area of the circle is four times this. ... ..... . .. . ..... ... .35) in Example 5. ........ ........ ... ..12). .. ... .... . . ... . .. ... .... . .. .. .. . ... . ∆ A1≈12πr∆r. ... . .. . . ..... Given that the circumference of a circle of radius r is 2π r (see Example 5. . . . ........ ....35) and.. .. . ..... .... ....... . . . ... . . . . . . ...... .. .. ... . .. it is an element of area or differential area. . ..... .... . . ... .... ...... . . ..... . ......... . .. has been replaced by the variable x... .. .. . . . .. .. ... ....... . . . .. . ........ . . . ...... The total area is then a Figure 5. . .. ... . ........... . . . . .. . . .. the area of a circular strip of radius r and ‘inﬁnitesimal’ width dr.. ... . . ... . .. . . . ...... . . . .. .... .. . .. . ... . . . ..... ..... .... .. . . .. . .. .. . . .... ... .. . . .. .. . .. .. . . ..... . .. .... . ..5.. . ...4 The integral calculus 145 The integral is evaluated in Example 6.. . . . . . .. . . . .. . . .. ...... . . .. .. .... . . ... . when ∆r is small enough. .. . .. . . .. . .. . . .. . . ... .. ... . .. .... . .. . . .. .. .... .. . . a. ... . ... . . ... .. .. ... . . .... ... b... ........ . .... .... .. . . . is arbitrary... .. ..... . .34). ... .. .. . .. .. in which the upper limit of integration... .... . . . . . .. . . . . .. .. ∆A dA → = 2π r ∆r dr The quantity dA2dr is the rate of change of the area A(r) of the circle with respect to the radius r... ... ... . ... . .. ... ... . . whilst the lower limit of integration.. .. . .. .. .. .. . .. . . . . . .. . . . . ... .. .... . .. . ... . . Divide the area of the circle into concentric circular strips as in Figure 5. .. .. . . .. . . .. . .... . .. ... .... . . . This approach is widely used in the application of the calculus to the formulation of physical problems. . . . The quantity F(a) is therefore an arbitrary constant... . . .. ... the area of the strip between r and r1+1∆r lies between 2π r∆r and 2π (r1+1∆r)∆r: 2π r∆r1<1∆A1<12π(r1+1∆r)∆r (5.. . .. . . .. . . . .. .. ... . .. . . . . .. . . . ..... .. . ......... .... . . . . ... .. . ... . .. . .. . . . .. . .. . ..11... . The ‘sum’ of these elements is the integral A = Z dA = Z 2π r dr 0 0 A a ∆A < 2π(r + ∆ r ) ∆r a ... .. . . .. .. .. ... ... . ........... ... . ... ....... .. ..... . . .. .. . . . .. .... ... .. .. . .. . . ... . .... .. ..... .. . ... . ... .... .. . . ..

.. . . .. ... . . .. . Therefore. . . . . .12 The circumference of a circle The equation of the circle of radius a and centre at the origin is x 21+1y 21=1a 2 so that y1=1±(a 21−1x 2)122 and dy2dx1=13x2(a 21−1x 2)122.14). . . . . . . . . . . . .. . . . . . . .. . . . . . ... .. . . . . . .. . . . . ... . . .. . .. . . .. an approximate value of the length of the segment between x and x1+1∆x is 2 (∆ x ) ( ∆ y )2 12 ∆s ≈ + 2 1 2 ∆y = 1 + ∆x ∆x The corresponding element of length is 2 1 2 dy ds = 1 + dx dx and the total length of arc AB is s = Z ds = Z 0 s b a 2 1 2 1 + dy dx dx EXAMPLE 5... .. .. .38) . ... .. . . ... .. . . .. . . . . .. . .. . . .. at x1=1b (Figure 5. .... . . . to point B.. . . .. .. . . . . . . . . . . . . . . . • y x ∆ +x • x s x • a • A (5. . .. . ... ... . . .. . . .... . .. . ... . . . . . .. . . ... .. . .. . . .. . . . .. .. . . .. . . . . . B Figure 5. . . . . .. .. ∆ . . . . ... . . . .. . . . .. .37) (5. .. . .. .. . . . . ... . . ... . . .... . .. . .. . .. we divide the arc AB into segments of length ∆s. . .. . . .. . . . . . . . . . . ...... . . . .. . . . . . .. . . . .. .. . . . ∆ . . . . . .. ... . .. . . ... ... .. .. . . . .. . ... . .. . .. . . . . . .. . .... . .. .. . ..... . Then. . . . . . by Pythagoras’ theorem. . . . .. . . ..... . . . . . . . . .14 Let y1=1f (x) be continuous in the range a1≤1x1≤1b. . ... .. . . .. . ... . . . .. . . .. . . ... at x1=1a. . ........ ....... . . . . To calculate this length. .. . . . ∆ . . . . ... . . . . . . . .... .. . .. . .. . . . . . .. . . .. . . dy x2 a2 1+ =1+ 2 = 2 dx a − x2 a − x2 and the circumference of the circle is a 2 s=4Z 0 a 2 1 2 1 + dy dx = 4 a Z dx 0 dx a − x2 2 b .. .. .... . ... . .. . ... . .. .. . . . . . . .. . . . .. . . . . . . . .. . . .... . . .. . . . . .. .. . ... . .... . . . . . .. . . . ... .. .146 Chapter 5 Integration The integral as a length .. .... .. . .. . .. .. .. .. . . . . .. . . . . .. . .. . and let s be the length of its graph from point A. .. . . ..

0 Exercise 49 5. a body moving with a given velocity at any time continues to move along a straight line with the same velocity if no external forces act on the body. By Newton’s ﬁrst law of motion.3: a x s = 4 a sin −1 = 4 a sin −1 1 − sin −1 0 a 0 ( ) The principal values of the inverse sine of 1 and of 0 are π22 and 0 respectively. the more general case of motion in two and three dimensions is discussed in Chapter 16. The application to pressure–volume work in thermodynamics is discussed in Section 5. This use of the integral calculus for the determination of static properties is introduced in Section 5. This use of the integral calculus is discussed in Chapters 11 to 14. The concept of the integral was introduced in Section 5. the three-dimensional case is discussed in Chapter 10. see Example 5. The integral is one of the standard integrals listed in Table 6. the centre of mass. When the function represents. The use of the integral calculus for the description of the dynamics of physical systems is introduced in Section 5.1 by considering the motion of a body along a line. The generalization to curves in three dimensions. the equations of motion in classical mechanics (derived from Newton’s second law for example). the integral calculus can be used to determine static properties of the body such as the total mass. for example. and the moment of inertia in terms of deﬁnite integrals involving the mass density. Similarly for a distribution of charge or of any physical property of a system that is distributed in space. The widest use of the integral calculus is for the solution (integration) of differential equations. On the other hand. v1=1constant when F1=10.5 Uses of the integral calculus 147 (four times the length of that quarter in the ﬁrst quadrant. and to volumes is discussed in Chapters 9 and 10. . Therefore s1=12πa. to nonplanar surfaces. The same methods are used in Chapter 21 for the analysis of probability distributions in statistics. that the deﬁnite integral provides the average value of a function. or rate of change of velocity a1=1dv2dt. in particular. the area of a plane ﬁgure and the length of a curve in a plane. the equations of chemical kinetics and of other rate processes.5. We saw in Section 5. equation (5. by Newton’s second law.3.6 for properties distributed along a line (the one-dimensional case). in particular. the distribution of mass in a physical body.8.11). and the equations of motion in quantum mechanics (such as the Schrödinger equation).7 for motion along a straight line.13). the acceleration.4 was concerned with the use of the integral calculus for the determination of geometric properties. experienced by a body of mass m in the presence of an external force F is given by F1=1ma.5 Uses of the integral calculus The discussion of Section 5.

. .......... . ... .. . .............. .... ... ........... . ....... ........ ......... .............. ....... . ............ .. This force Fi is what is commonly called the weight of the body................... ..................... . ... ........ ..... . ......... ............ (ii)...... ...40) by g gives NF 1− N F x .. .............. ................ ....... .......... ... .. ... ... .............39) 3F 2F 1F • • • ··· 3x 2x 1x 1 m • 1 x ...6 and there are similar deﬁnitions for all higher moments... The ﬁrst moment of mass....15........ .. . . .. . where g is the standard acceleration of gravity (‘gravitational constant’). ... .... o x (ii) ﬁrst moment of mass about O: 3 ··· m • 2 m • 3 x 2 x (i) total mass: Static properties of matter M = ∑ mi i =1 N T = ∑ Fi xi = MgX = FX • o x ··· Figure 5...... ......... ...... .148 Consider a set of N discrete masses m1....... . ..... .. ....... ... ........ .... .. .. ........ .. ...... . ... ..... . . .... .... ... .... m2.... .... .. . .......... ...... .............. . ................ ........................ . ..... ..... ........ .................... ...... ..... .. .... .. ........ . .16).. mN distributed along a (horizontal) straight line....... ... ... . =.................... ........... .... .... . ..... .......... ... ....... . .... . ... ......... ...... In the presence of gravity. . ....... .... ... ... ............................ ...... ... ... ... .. ....... ........ .......... .. .... . .............. . .......... .................. ...... .................. .. ....16 ∑ mi xi i =1 N ∑ mi xi2 i =1 N (5.... .. . ............................. ... ............ Multiplication of equation (5.. ............... .. ..... .15 i =1 N ∑ mi xi = MX i =1 N Figure 5... ... ..... .. .41) (5............ .40) (5... ....... ...... with mass mi at position xi with respect to a ﬁxed point O as in Figure 5........ ......... .. ..... .. .......... .......... .............. .. ..................... ............ ....... ... ................... . . .... .. .. . each mass experiences a force Fi1=1mi g directed downwards (Figure 5.......... .......... ... ...... ........... ................... .. ... .. . . ......... .. ...... deﬁnes the position X of the centre of mass of the system of masses: x Nm • 1− N ............. Nx 1− N The system of masses has the following properties: • m x ··· (iii) second moment of mass about O: The ﬁrst moment of the system of masses is therefore equal to the moment of the total mass M concentrated at the centre of mass.. . • Nx 1− Nx Chapter 5 Integration 5. .. ..... ..... ..... ... .......... ............... .... . ............. . ........... ....... .............. ...... .......... ... ...... . ......... .

.... ... .... The position X of the centre of mass is then also called the centre of gravity...... .. .... If the masses are attached to a uniform rigid rod (itself of negligible mass) then the torque is a measure of the tendency of the forces to rotate the system around O as pivot and. ....... .. Figure 5......... .. putting x11=1−r1 and x21=1r2 in equation (5.. . ....... .......... is the moment of inertia of the system of masses with respect to the point O........ ... .. ....... ... . .. .. ......17 shows two bodies..... F1r11=1F2r2 (law of levers) and the body is at equilibrium with respect to rotation about the pivot O... ........ . . ..... . .......... EXAMPLE 5.. .... . . .. .. .......... and the distances r1 and r2 can be written as r1 = m2 R.... ...40) so that m1r11=1m2r2.... masses m1 and m2......41)............... . .. If the point O is at the centre of mass (gravity) then X1=10 and T = ∑ Fi xi = 0 i =1 N (5............ The forces F1 and F2 are the weights of the bodies. .. .. ... . m1 + m2 r2 = m1 R m1 + m2 .. .......... (iii). .. .. ............ . .. .... . and F is a counter force acting at the pivot point O.. . .............. ....... ...... .... .. ... it is equal to the torque produced by the total force F concentrated at the centre of mass... The second moment of mass. The moment of inertia of the two masses is I1=1m1r 21+1m2r 2..... ..17 2F • 2r o F 1r 1F • .. and the quantity T = ∑ Fi xi is called the moment of force or torque of the system of forces about the point O.... joined by a rigid rod (of negligible mass).......... ...... . ... . . ........... ..6 Static properties of matter 149 where F = ∑ Fi is the total force acting on the system of masses.......... ....... ... The total vertical force acting on the body is F11+1F21−1F........... ... ...... . ... .42) so that the total torque about the centre of mass is zero. . .13 A system of two masses Figure 5.. .. . . so that the body is at equilibrium with respect to vertical motion if F1=1F11+1F2... .. . . . ....... If the pivot is at the centre of mass then.. .......... .. .... .......42).... ...... ... .. by equation (5. When O is at the 1 2 centre of mass then X1=10 in (5....... ..... . . ........ It is the property of the mass distribution that is most important in the description of the dynamics of rotating bodies....... ...... ....... .. ........ . .. . ..... ......5. . . .. . . ...... ........ ........ . . . ... ....

........................ .... .. ......... the sums of the discrete case are replaced by integrals...... ....... . .......... and its value depends on the position of the segment and on its length..................... In this case.. so that ∆m2∆x is the mass per unit length in the segment. .. ............ ......... then ∆m2∆x is independent of the choice of segment..... . .............. . Figure 5.... the total mass is M1=1ρl... the average density is M ρ= = l Z ρ ( x ) dx 0 l Z l 0 dx x (5... ........... ...... ....... .. ...... ... ......... ... a constant throughout the length.... . l x +x ∆ m ∆ x 0 . .. . ......... ....... when a discrete mass distribution is replaced by a continuous distribution... the ratio approaches the limit ρ ( x ) = lim ∆ m dm = ∆x→0 ∆x dx The value of the function ρ (x) at each point is the density at that point... .18)....... ........................44) (5................. The moment of inertia is then 2 2 I = m1r 1 + m2 r 2 = m1m2 R2 = µ R2 m1 + m2 where the quantity µ1=1m1m2 2(m11+1m2) is the reduced mass of the system of two masses (see also Example 1.......................... the linear mass density) ρ of the body. ... ................ ... ........... . .......18)........ If the mass is distributed evenly over the length........ ...........43) (5... .... .......18 Let the mass in the segment x to x1+1∆x be ∆m....... .. ............. ........ .. ............150 Chapter 5 Integration where R1=1r11+1r2 is the distance between the masses...... ............. ......... The total mass of the body is then M =Z M dm = Z ρ ( x ) dx 0 l 0 Therefore... .... ................. ................... ....... ....... then ∆m2∆x is the average density in the segment x to x1+1∆x... ..... ......... ......... As this length is reduced to zero. If the mass of the rod is not evenly distributed over its length.......... .................................. 0 Exercise 50 The continuous case Consider a continuous mass distribution such as a straight rod of matter of length l (Figure 5............. .......... .... ..... .... .... ................. The moment of inertia about the centre of mass of a system of two masses is therefore the same as the moment of inertia of a single mass µ at distance R from the centre of mass............. ... ....... ........... ... more correctly.... . .......... ............ and the differential quantity dm1=1ρ (x)dx is the mass of a segment dx at x. .... ........... Thus. .. ....... ...... ........45) .................. ...... ............... ......... ................. ........ ..... ................. ....... It is then the density (or.....

........... .. ...47) We note that.. ........ ... . . ................. ........ .... .. . .. ......... ....... ..... . ............ ... .. .... ..... .. . . Thus ba . 0 or X= Z ρ( x ) x 0 l l dx (5...........6 Static properties of matter 151 The position of the centre of mass is given by (see equation (5. .. .... . .. ...... . .. .. ..... . ... .. ........ .......... .... .... ........ ....... ....... ... . ... ...... .... .. ...... . for example.. a 0 a 0 M = Z ρ( x ) dx = Z b( a − x ) dx = The position of the centre of mass is X= 1 M a a2b 2 Figure 5. . . ......46) Z ρ ( x ) dx 0 and the moment of inertia with respect to an arbitrary point x0 on the line is I = Z ρ ( x )( x − x0 )2 dx 0 l (5.... .. . ......... . . .. ....... .... ..... 0 . ... ..... ..... ..... .. ... .... ... .. ....40)) l MX = Z ρ ( x ) x dx ....... . . .... .. ... ..... ..... .. .... .. .... .... . .. .. . .... ......... ... ..... ... .. . ..........14 Find the total mass............ ... ... ....... .. ..... ...... . ... .. . ... 0 Exercise 51 x a )x(ρ The total mass is equal to the area a2b22 of the triangle shaded in Figure 5.... . ........ .. .. ..... ... ..19.. . .. . .. ..... ....... .... ....5. centre of mass.. this is a general result. .. ... . ....... The moment of inertia therefore has its smallest value when computed with respect to the centre of mass. .... ..... .. ..... . ........ ..... . ....... .. .. .... and moment of inertia of the linear distribution of mass of length a and density ρ (x)1=1b(a1−1x).. .. .. ... ... ........ ....... . ..... .. .... .. ... . .. . .. . . .... ... the total mass can be interpreted as the ‘area under the curve’ of the graph of the mass density function ρ (x).... ... EXAMPLE 5. .. ...... ... . ....... .......... .. . . . . . .... .... ....... . ....... . ...... .... .. .... .....19 a Z 0 ρ ( x ) x dx = 2 a 2 Z ( a − x ) x dx = 0 a 3 The moment of inertia with respect to an arbitrary point x0 on the line is a 0 a 0 I = Z ρ( x )( x − x0 )2 dx = bZ ( a − x )( x − x0 )2 dx = ba 2 2 2 ( a − 4 ax0 + 6 x0 ) 12 The value of x0 for which I has its minimum value is given by dI ba 2 =0= ( − 4 a + 12 x0 ) dx0 12 so that x01=1a23..... . .. ........... .. .... . . .. ... . . . ..... ... ....... ... .... .. .... .. .. . . 01≤1x1≤1a.... .. .. . ....... .

by Newton’s second law. the corresponding element of work is dW1=1F(x)1dx and the total work done is the integral W = Z dW = Z 0 W xB F ( x ) dx xA (5. F1=1F(x). and the total distance is then s = Z ds = Z v(t ) dt 0 tA s tB (5. the quantity v(t)1=1ds2dt is the velocity of the body at time t. If the body falls from rest at t1=10.1. we consider a body moving along a curve from point A at time t1=1tA to point B at time t1=1tB.48) EXAMPLE 5. given that g1≈19. Then v(t)1=1gt1+1v(0).8 m s−2.7 Dynamics Velocity and distance As in Section 5. Work is done on the body by the application of the force. and let the distance from A along the curve be s(t) at some intermediate time t. If the force is constant then the work done is (work1=1force1×1distance) W1=1F(xB1−1xA).9 m.11. 0 Exercise 52 Force and work Consider a body moving along the x direction between points A and B with velocity v1=1dx2dt. If a force F acts on the body then.49) where m is the mass of the body. but is a function of position.91×1(221−112) m1=114.7 m. By the discussion of Section 4. The work done on the body between positions x and x1+1∆x is ∆W1≈1F(x) ∆x.50) . then v(t)1=1gt. the distance travelled in the ﬁrst second of fall is 4. and in the following second it is 4. the acceleration a experienced by the body is given by F = ma = m dv dt (5. In the limit ∆x1→10.152 Chapter 5 Integration 5.15 A body falling under the inﬂuence of gravity A body of mass m falling freely under the inﬂuence of gravity experiences the constant acceleration dv2dt1=1g (air resistance and other frictional forces are neglected). The distance travelled in the time interval t1=1tA to t1=1tB is then s = Z v(t ) dt = Z gt dt = tA tA tB tB 1 2 2 g t B − tA 2 Therefore. where v(0) is the velocity at time t1=10. then the work done is obtained by means of the integral calculus. The differential length ds1=1v dt is the distance travelled in the inﬁnitesimal time interval dt. If the force is not constant between A and B.

.. The work done by gravity on the body as it falls through a height h is therefore W = Z F dx = Z mg dx = mgh 0 0 h h This is also the work that must be done against the force of gravity to raise the body through the distance h. .................................. . The force −F must be applied to overcome the repulsion.. .............. Because the charges repel..........20......... Figure 5. ............. work must be done on the system to bring q2 from inﬁnity to the distance x from q1.. ....................... .....17 Electrostatic work The force acting between two electric charges q1 and q2 separated by distance x in a vacuum is given by Coulomb’s inverse-square law F ( x) = q1q2 4 π ε0 x 2 where ε0 is the permittivity of a vacuum..... 2q • x 1q • 0 ........7 Dynamics 153 EXAMPLE 5..... and the force on q2 is directed towards q1 (F is negative in Figure 5...... .. F ...... . . x ............................ .. and the work done is x W = − Z F ( x ) dx = − ∞ q1q2 4 π ε0 x Z ∞ dx x2 =− x q1q2 1 qq − = 1 2 4 π ε0 x ∞ 4 π ε0 x * The presence of the factor 4πε0 ensures that the force acting between two charges of one coulomb separated by one metre is one newton.............................. ......................20)....... ......... the force acting on q2 due to the presence of q1 acts in the positive x-direction.. .. in Figure 5............ .............. .......* Like charges (charges with the same sign. .. EXAMPLE 5............ so that............. ..... . .... away from q1. .. ....... ......... ...... ............ ...... initially inﬁnitely far apart............. .... ................... such as two nuclei or two electrons) repel.......................... Unlike charges (of opposite signs.......................... .... .... N1=1C 22(4πε0 m2) in terms of SI units.......... ....................... .................................. ..........16 A body falling under the inﬂuence of gravity A body of mass m falling freely under the inﬂuence of gravity experiences the constant force F1=1mg directed downwards.....20 Consider two like charges. ......................... such as the proton and electron in the hydrogen atom) attract.......... .. ....... ............. This is the total force acting on the body in the absence of air resistance and other frictional forces.5.. .....

53) where vA and vB are the velocities of the body at A and B respectively. arising from the position of the system in space and from the forces acting on the system at that position. 0 Exercise 53 Work and energy When work is done on a system by an external force. so that F1=1F(x). and (ii) the potential energy. but the work is then negative. This condition excludes time-dependent forces and. (i) Kinetic energy The work done on a body by an external force F(x) as the body travels from point A to point B is WAB = Z F ( x ) dx = m Z A B B A dv dx dt (5.54) Let the force acting on a body depend only on the position of the body.51) Because v1=1dx2dt. the kinetic energy is the sum of the kinetic energies of its constituent parts. . these parts are (i) the kinetic energy. Conversely.52) A in which use has been made of d(v 2)2dt1=12v dv2dt. or translational energy. (ii) Potential energy and total energy (5. arising from the motion of the system in space. It follows that the work done between A and B is WAB = 1 2 2 m v − vA 2 B (5.154 Chapter 5 Integration This work is positive for like charges. the energy of the system is increased by the amount of the work done. the energy of the system is decreased by the amount of the work done. when a system does work against an external force. The energy of a system is usually expressed as the sum of two parts. so that WAB = mZ B B dv 1 d 2 1 v dt = m Z (v ) dt = m v 2 A 2 2 dt A dt B (5. The same formula applies to the case of unlike charges. For a simple system. and the potential energy is the sum of the potential energies of its parts. more importantly. In the case of a system with internal structure. the element of length dx can be replaced by the differential v1dt. The work done on the body is therefore equal to the change in kinetic energy: WAB1=1TB1−1TA We note that the kinetic energy of a body at rest is deﬁned to be zero. The quantity 1 mv 2 is called the kinetic energy of the body and is usually denoted by the symbol T 2 (or K). with no internal structure. and the integration limits now refer to the times at A and B.

5.7 Dynamics

155

dissipative forces such as those due to friction. Then, by the fundamental theorem of the calculus (equation (5.34)), there is a function f (x) such that F(x)1= f ′(x) and WAB = Z F ( x ) dx = f ( B) − f ( A )

A B

(5.55)

The work done from A to B can therefore be expressed as the change in a quantity which depends only on the end points A and B, and not on the path A to B. This quantity is normally designated by −V, and V is called the potential energy of the body. The work from A to B is then WAB = Z F ( x ) dx = VA − VB

A B

(5.56)

where VA and VB are the values of the potential energy at A and B. When the work done by a force is independent of the path, the force is called a conservative force and is (–) the derivative of the potential energy (function): F ( x) = − dV , dx V ( x ) = − Z F ( x ) dx + C (5.57)

Three simple but important types of conservative force, and corresponding potential energy, are (a) F1=1constant, (b) (c) F1=1−kx, F=− 1 x

2

V (x)1=1−Fx1+1C V ( x) = 1 2 kx + C 2 1 +C x

(5.58) (5.59) (5.60)

,

V ( x) = −

where C is an arbitrary constant. The graphs of V(x), for C1=10, are illustrated in Figure 5.21. In each case, the gradient of the graph is dV2dx1=1−F so that the force acts in the direction of decreasing potential energy.

Figure 5.21

x

2x 1

. . . . . . . . . . . . . . .. ... .. .. ... .. .. ... . .

−= F) ( c

. . . . . . . . .....

...... . ...... . .... ...... ...... . . . . ... . . .... . ... ....... . . .. .. . . . .. .. .. . .... . . . . . .

....

x

. . .... .... .... . .. .

xk− = F )b(

...... .

......

. .. . ...... . ...... . ...... . . .... . ...... .. .. ..... ... . ...... . .....

... .

..... . .

. . ... ..

. . .. . ..

. . . . . ...

. ... .. .. .. .. .. ... ... ... . .

. . . . .. . ...

. ... . .

. .. .

. .....

.

. . .. ..

.... .

. ..

. ....

.

. . .. ..

.....

x

tn atsn o c =

.. . .. . . .

. .. .. . .

. . .. . . .

. . .. . ..

.

. . . .. .. . . ... .. ... . . .. .. . . . . . .. . .. .. . . . .. . . . . .. . . . .. .. . . . .. .

... ... ...

.. .... ...

. . ... . .... . . ... ... .... .... ... .

. . ..

... .

.. ... ... .

x

1

−= V

.. . .. .. ... .. . . . . . .

. .....

. ......

2

2 xk 1

.. . .. .. ... .. . . . . . .

=

V

. ...... .. ...

.. ...

....

... ... .... .... ... .. .

x F− = V F)

. . . . .. .. . ... .. . .. . .

.. . .. .. ... .. . . . . . .

. . .. . .

. . . .. .

( a

.

156

Chapter 5 Integration

We note that, whilst the kinetic energy has a well deﬁned absolute value, this is not true of the potential energy, for which only relative values are deﬁned by equations (5.56) and (5.57). Examples 5.18 and 5.19 below show how the zero of potential energy is chosen in two different physical situations. A system in which all the forces are conservative is called a conservative system. In such a system, the work done in moving a body round a closed loop A1→1B1→1A is zero: WABA1=1WAB1+1WBA1=1(VA1−1VB)1+1(VB1−1VA)1=10 (5.61)

Dissipative forces such as those due to friction are not conservative forces because the work done against friction is always positive. Combining the expressions (5.54) and (5.56) we have the result TA1+1VA1=1TB1+1VB (5.62)

and it follows that, in a conservative system, the quantity T1+1V is constant. This quantity is called the total energy of the system, E1=1T1+1V, and (5.62) is an expression of the principle of the conservation of energy: if the forces acting on a body are conservative, then the total energy of the body, T1+1V, is conserved.

EXAMPLE 5.18 A body moving under the inﬂuence of gravity Consider a body of mass m at height h above a horizontal surface, as in Figure 5.22. The force of gravity acting on the body is F1=1−mg (negative because the force acts in the negative x-direction) and the work done on the body as it falls freely from height x1=1h onto the surface at x1=10 is W = Z F dx = − mg Z dx = mgh

h h 0 0

Figure 5.22

This work is the change of potential energy, W1=1mgh1=1V(h)1−1V(0) where V(x) is the potential energy of the body at height x. The natural choice of zero of potential energy in this example is V(0)1=10, zero at the surface. Then V(x)1=1mgx is the potential energy of the body at height x, and the force is related to it by F1=1−dV2dx1=1−mg. Let the body fall from rest at x1=1h and let the kinetic energy at height x be T(x). Then T(h)1=10 and, by equation (5.54), the kinetic energy of the body when it reaches the surface is T(0)1=1mgh. In addition, because the force (a constant) is conservative, the total energy of the body is conserved and is equal to E1=1mgh, which is the potential energy at x1=1h (where the kinetic energy is zero) and is the kinetic energy at x1=10

5.8 Pressure–volume work

157

(where the potential energy is zero). At an intermediate height, E1=1mgh1=1T(x)1+1V(x) and the kinetic energy is T ( x) = 1 2 mv = mg ( h − x ) 2 (5.63)

If both the body and the surface are perfectly elastic, the velocity of the body is reversed on contact with the surface, and the body returns to its original height at x1=1h in an exact reversal of the falling motion. Thus, solving equation (5.63) for the velocity, v = ± 2g ( h − x ) and the velocity of the body is negative as it falls and positive as it rises. In the absence of dissipative forces the bouncing motion is repeated indeﬁnitely. EXAMPLE 5.19 Electrostatic potential energy By Example 5.17, the work that must be done against the internal force to bring two charges from inﬁnite separation to separation x is W1=1q1q224πε0x. This is the same as the work done by the internal force in separating the charges: W = Z F ( x ) dx =

x ∞

q1q2 4 π ε0 x

The force depends only on the relative positions of the charges and is conservative, so that a potential energy function V(x) exists such that F(x)1=1−dV2dx and W1=1V(x)1− V(∞). It is conventional, and convenient, to choose the zero of potential energy of interacting charges to be zero for inﬁnite separation: V(∞)1=10. Then V ( x) = q1q2 4 π ε0 x

is the electrostatic potential energy of the system of two charges.

0 Exercises 54, 55

5.8

Pressure–volume work

Consider a ﬂuid (gas or liquid) enclosed in a uniform cylindrical container, closed at one end and ﬁtted with a piston as shown in Figure 5.23. Let A be the internal crosssectional area of the cylinder. A ﬂuid with internal pressure p exerts a force of magnitude 1|F|1=1pA on the surface of the piston, and the piston moves in or out

158

in which the limits of integration now refer to the volume, and the minus sign is included to make the work positive for compression. It can be shown that this expression for the mechanical work done on a thermodynamic system is independent of the shape of the container. To compress the ﬂuid, it is necessary that the external pressure be greater than the internal pressure of the ﬂuid. Let pext1=1( p1+1∆p) where ∆p is a positive excess pressure that, for simplicity, can be assumed to be constant throughout the compression. Then, with Va1>1Vb,

Now, the external force has magnitude |Fext |1=1pext A, and a length of cylinder |dx| contains a volume |dV |1=1A|dx|. The work can therefore be written in ‘pressure– volume’ form as

Chapter 5 Integration

according as the external pressure pext is greater than or less than p. If pext1>1p then the ﬂuid is compressed, and the work done by the external force Fext in moving the piston from a to b is

Conversely, to allow the ﬂuid to expand from Vb to Va it is necessary that the external pressure be smaller than the internal pressure. If pext1=1( p1−1∆p) then

............................................................................................................................ ......................... ..... .. ........ ................................................................................ ... .. . . .. ... . .... .... . . . ................................................................................................................................ .. . . . . . .. ... .. ........................................... ........................................ . ........................................ . . . .. . . ...................................................................................................... . ..................................................................................................... . .................................................................................................... . ................................................................................................... . . . . . . ....... . . .... . ... .. . . .. .. .. . .. . . . .... . .. . .... . . . . .... . ..... .... ... .. . .. .. . . . . ..... . .... .. ... . .. . .. .. . . . . . . .... . .. . . .. .... ............... .... . .. ... . .... ...... .... . .. . . ... . . .. ... . .. . . . . .... . . . . . .. . .. . .. .... .. . .... . ......... . . ... ... . . . . . . . .. . . . . .. .. . . . . . . .. .. . .. .. .......... . ... .. . . ... . ......... .... .. . t................................................................................... x.................................................... ......................... e . ..... . . .. .. . .. . . . . .. . . . ..... .. . .... . . .... .. .. .. ..... . . .. . . . . .. .. .. . ... . . . .... .. . .. . . .. . . . .... . . . . .. ....... . .... . .. ....... . .. .. .. .. ... .. .. ............................................................................................................................ . .......................................................... .......................... ..... ............................... . . .... .. .. .. .. ................................................................. ................................ ........................ . . ... ........................................................ ............................................. ................ . ........... ................................................................................................................ ............................................................................................................................ .

Wab = − Z p dV − ∆ p Z dV = − Z p dV + ∆ p (Va − Vb )

b a x ......................................................................................................................................................................................................................................................................................................................................... .... .... .... .... . . . . . . . . . . .

. .

p

p

> − Z p dV

a

b

a

b

Wba = − Z p dV + ∆ p(Va − Vb )

**> − Z p dV Wab = − Z pext dV Wab = Z Fext dx
**

a b a

b

Figure 5.23

a

b

a

b

a

b

a

b

(5.65)

(5.64)

(5.66)

(5.67)

5.8 Pressure–volume work

159

The total work done on the ﬂuid in the cycle a1→1b1→1a is therefore Waba1=1Wab1+1Wba1=12∆p(Va1−1Vb)1>10 (5.68)

and the process is said to be irreversible. This is analogous to the case in ordinary dynamics when nonconservative dissipative forces are present. The process can be made reversible only by letting the excess pressure, ∆p in our example, approach zero. In this (ideal) limit the work is reversible and Wab = −Wba = − Z p dV

a b

(5.69)

Expansion of a gas

The equation of state of a gas is a relation of the form f ( p,1V,1T )1=10 amongst the three thermodynamic quantities p, V, and T (for a given amount of gas). For example, the equation of state for the ideal gas can be written as pV1−1nRT1=10. There is therefore some freedom in the choice of conditions under which the expansion of the gas can occur.

isobaric expansion

Expansion of a gas can occur at constant pressure; for example, the gas can be heated to expand against a constant external pressure such as atmospheric pressure. The reversible work done by the gas against the external pressure is then W = Z p dV = pZ dV = p(Vb − Va )

a a b b

(5.70)

**For the ideal gas, pV1=1nRT and W1=1nR(Tb1−1Ta )
**

isothermal expansion

(5.71)

Expansion of a gas can occur at constant temperature; for example, if the expansion occurs with the container immersed in a heat bath at a given temperature. To calculate the work it is necessary to know the equation of state. For the ideal gas, p1=1nRT2V and the (reversible) work is W = Z p dV = nRT

a b

Z

V p dV = nRT ln b = − nRT ln b Va pa a V

b

(5.72)

0 Exercise 56

160

Chapter 5 Integration

5.9

Exercises

Section 5.2

Evaluate the indeﬁnite integrals: 1. 6. dx x3 dx x −1

Z 2 dx Z sin 4x dx

2. 7.

Z x 3 dx Z e3x dx

3. 8.

Z x 2 3 dx Z e−2 x dx

4. 9.

Z Z

5. 10.

Z x −1 3 dx Z

dx 3− x

Evaluate the indeﬁnite integrals subject to the given conditions: 11. y = Z x 2 dx ; y1=10 when x1=13 13. I = Z(5x 4 + 2 x + 3) dx; I1=14 when x1=12 12. y = Z cos 4 x dx; y1=10 when x1=1π24 14. I = Z 3x 2 + 2 x + 1 x2 dx ; I1=13 when x1=11

1 15. I = Z −4 + 4 cos 2 x − e2 x dx; I1=10 when x1=10 2

Section 5.3

Evaluate the deﬁnite integrals: 16.

+1 5 2

Z ( 2 x 2 + 3x + 4) dx

−1

17.

Z dx

3

18.

Z

du u3

19.

Z Z

1

π

dv v+2 2

0

3

20.

Z e−3t dt

1

5

21.

Z cos θ dθ

0

π 2

22.

Z cos 3θ dθ

0

23.

sin 2 x dx

−π 4

24. (i) Show that the rate equation of a ﬁrst-order decomposition reaction dx = − kx dt can be written in the logarithmic form d ln x = −k dt (ii) Integrate this equation with respect to t over the range 0 to t, and show that x( t ) = − kt and x (t ) = x(0)e− kt ln x (0) 25. The Clausius–Clapeyron equation for liquid–vapour equilibrium is d ln p ∆ H vap = dT RT 2 If the enthalpy of vaporization, ∆Hvap, is constant in the temperature range T1 to T2 show, by integrating both sides of the equation with respect to T, that p ∆ H vap 1 1 − ln 2 = p T T R 1 1 2 where p11=1p(T1) and p21=1p(T2).

5.9 Exercises

161

Find the average values in the given intervals: 27. cos13θ ; 26. 2x 21+13x1+14; −11≤1x1≤1+1 Demonstrate and sketch a graph to interpret: 29.

3 1 2 3

01≤1θ1≤1π22

3

28. 1;

31≤1x1≤15

Z e− x dx = Z e− x dx + Z e− x dx + Z e− x dx

0 0

π 2

30.

Z e x dx = Z e x dx − Z e x dx

2 0 2

π 2

6

6

1

2

π π 2

3

**31. (i) Show that Z
**

π

cos x dx = − Z

**cos x dx. (ii) Calculate Z cos x dx , Z
**

−π

cos x dx,

0

−π

**Z cos x dx . (iii) Sketch a graph to interpret these results.
**

−π

Evaluate and sketch a graph to interpret: 32.

Z f ( x ) dx where f ( x ) =

−1

+3

x2 + 2 2 x

if x < 1 if x ≥ 1 if x > 0 if x ≤ 0 if x > 0 if x ≤ 0

1

33.

Z f ( x ) dx where f ( x ) =

−1 +a

+1

x − x e− x − e+ x

34.

Z f ( x ) dx where f ( x ) =

−a

35. (i) Show that Evaluate: 36.

d ( x ln x − x ) = ln x, (ii) evaluate Z ln x dx . dx 0

∞ ∞ ∞

Z e−3t dt

0

∞

37.

Z e− x 2 dx

0

38.

Z

2

dx x ( x − 1)

39.

Z

dx x ( x − 1)2

2

For each function, state if it is an even function of x, an odd function, or neither. If neither, give the even and odd components. 45. 3x21+12x1+11 40. sin12x 41. cos13x 42. sin1x1cos1x 43. x 44. x4 −x 2 −x 47. (3x 1+12x1+11)e 46. e

Section 5.4

48. The equation of an ellipse with centre at the origin is = 1, where, if a1>1b, a is a b2 the major axis and b the minor axis (if a1=1b, we have a circle). Use Method 1 in Example 5.11 to ﬁnd the area of the ellipse.

2

x2

+

y2

1 49. Find the length of the curve y = 2 x 3 2 between x1=10 and x1=11.

Section 5.6

50. Three masses, m11=11, m21=12 and m31=13, lie on a straight line with m1 at x11=1−4, m2 at x21=1−1 and m3 at x31=1+4 with respect to a point O on the line. Calculate (i) the position of the centre of mass, (ii) the moment of inertia with respect to O, and (iii) the moment of inertia with respect to the centre of mass. 51. The distribution of mass in a straight rod of length l is given by the density function ρ(x)1=1x2; 01≤1x1≤1l. Find (i) the total mass, (ii) the mean density, (iii) the centre of mass,

162

Chapter 5 Integration

(iv) the moment of inertia with respect to an arbitrary point x0 on the line, (v) the moment of inertia with respect to the centre of mass. (vi) Show that the moment of inertia has its smallest value when computed with respect to the centre of mass.

Section 5.7

52. A body moves in a straight line with velocity v1=13t 2 at time t. Calculate the distance travelled in time interval (i) t1=101→11, (ii) t1=111→12, (iii) t1=131→14. 53. A body of mass m moves in a straight line (the x-direction) under the inﬂuence of a force F1=1kx. What is the work done on the body between x1=1xA and x1=1xB? 54. A body of mass m moves in a straight line (the x-direction) under the inﬂuence of a force F1=1kx, where k is positive (see Exercise 53). (i) Find the potential energy V(x) (choose V(0)1=10). The body is released from rest at x1=11. (ii) Find (a) the total energy E and (b) the kinetic energy T(x) as functions of x. (iii) Sketch a graph showing the dependence of V(x), T(x), and E on x. (iv) Use the graph to describe the motion of the body. (v) What would be the motion if the body were released from rest at (a) x1=1−1, (b) x1=10? 55. Repeat Exercise 54 with F1=1−kx.

Section 5.8

56. A slightly imperfect gas obeys the van der Waals equation of state n2 a p+ (V − nb) = nRT V2 Find expressions for the work done by the gas in expanding reversibly from volume V1 to volume V2 at (i) constant pressure, and (ii) constant temperature (assume a and b are constant).

6 Methods of integration

6.1 Concepts

To calculate the value of a deﬁnite integral it is normally necessary to evaluate the corresponding indeﬁnite integral; that is, given

Z f ( x ) dx = F ( x ) + C

we need to ﬁnd the function F(x) whose derivative is f(x)1=1F′(x). We saw in Section 4.6 that every continuous function can be differentiated by the application of a small number of rules. Each such rule, on inversion, provides in principle a rule of integration. In particular, the chain rule and the product rule provide, on inversion, the two principal general methods for calculating integrals; the method of substitution (Section 6.3) and integration by parts (Section 6.4). The aim of these general methods of integration, and of the other particular methods discussed in the following sections, is to reduce a given integral to standard form; that is, to transform it into an integral whose value is given in a table of standard integrals. A standard integral is simply an integral whose value is known and whose form can be used for the evaluation of other ‘non-standard’ integrals. The most elementary standard integrals are those given in Table 5.1, and a more comprehensive list is given in the Appendix. Extensive tabulations of standard integrals have been published, and it is our purpose in this chapter to show how (some of) these standard integrals are obtained and, most important, how others can be reduced to standard form. It must be remembered, however, that there are many functions whose integrals cannot be expressed in terms of a ﬁnite number of elementary functions. In such cases, approximate values are obtained by numerical methods; with the development of computing machines, numerical methods of integration (numerical quadratures) have become routine and accurate, and some of the simpler numerical methods are discussed in Chapter 20. We consider ﬁrst the use of trigonometric relations (Section 3.4) for the integration of some trigonometric functions.

6.2

The use of trigonometric relations

Table 6.1 contains a number of integrals that can be evaluated by making use of the trigonometric relations (3.21) to (3.26) discussed in Section 3.4. The useful forms of these relations are cos2 x = sin 2 x = 1 1 + cos 2 x 2 1 1 − cos 2 x 2 (6.1)

(6.2)

164

Chapter 6 Methods of integration

1 sin x cos x = sin 2 x 2 sin x sin y = 1 cos( x − y ) − cos( x + y ) 2 1 cos( x − y ) + cos( x + y ) 2 1 sin ( x − y ) + sin ( x + y ) 2

(6.3)

(6.4)

cos x cos y =

(6.5)

sin x cos y =

(6.6)

More generally, the relations can be used to express a function sinm1x1cosn1x, where m and n are positive integers, in terms of simple sines and cosines, but alternative methods of integration are often simpler to use when m or n is greater than 2. Table 6.1

1. 2. 3.

Z cos2 ax dx = Z sin 2 ax dx =

1 ax + sin ax cos ax + C 2a 1 ax − sin ax cos ax + C 2a 1 sin 2 ax + C 2a

Z sin ax cos ax dx =

For a1≠1b: 4. 1 sin ( a − b) x sin ( a + b) x +C Z sin ax sin bx dx = − 2 a−b a+b 1 sin ( a − b) x sin ( a + b) x +C Z cos ax cos bx dx = + 2 a−b a+b 1 cos( a − b) x cos( a + b) x +C Z sin ax cos bx dx = − + 2 a−b a+b

5.

6.

EXAMPLES 6.1 The integrals in Table 6.1 Integral 1:

Zcos2 2x dx

1 1 + cos 4 x . Therefore 2

By equation (6.1), cos2 2 x =

6.3 The method of substitution

165

**Zcos2 2 x dx = Z 1 + cos 4 x dx = x + sin 4 x + C 2 2 4
**

= Integral 3: 1 2 x + sin 2 x cos 2 x + C 4

1

1

1

Zsin 2x cos 2x dx

1 By equation (6.3), sin 2 x cos 2 x = sin 4 x. Therefore 2

**Zsin 2 x cos 2 x dx = Zsin 4 x dx = − cos 4 x + C
**

1 1 1 = − 1 − 2 sin 2 2 x + C = sin 2 2 x + C − 8 4 8 = 1 2 sin 2 x + C ′ 4

1 2

1 8

where C′ is a new arbitrary constant. Most tabulations of indeﬁnite integrals omit the arbitrary constant, and this example shows why different tabulations sometimes give apparently different values for indeﬁnite integrals. Integral 4:

Zsin 2x sin 4x dx

1 1 cos( −2 x ) − cos 6 x = 2 cos 2 x − cos 6x . 2

By equation (6.4), sin 2 x sin 4 x = Therefore

**Zsin 2 x sin 4 x dx = Z cos 2 x − cos 6 x dx = 2 2 2
**

0 Exercise 1–10

1

1 sin 2 x

−

sin 6 x +C 6

6.3

The method of substitution

The polynomial f(x)1=1(2x1−11)3 can be integrated by ﬁrst expanding the cube and then integrating term by term (see the corresponding discussion of the chain rule in Section 4.6):

166

Chapter 6 Methods of integration

f(x)1=1(2x1−11)31=18x 31−112x 21+16x1−11

Z( 2 x − 1)3 dx = 8 Z x 3 dx − 12 Z x 2 dx + 6 Z xdx − Z dx

=12x41−14x 31+13x 21−1x1+1C′ 1 = ( 2 x − 1)4 + C 8 (C1=1C′1−1128 is an arbitrary constant). A simpler way of integrating the function is to make the substitution u1=12x1−11, du = du dx = 2 dx dx

where du is the differential of u(x). Then, dx = 1 du and 2

Z( 2 x − 1)3 dx = Zu3 du = u 4 + C = ( 2 x − 1)4 + C

The integral has been transformed into a ‘standard integral’ by changing the variable of integration from x to u. The method of substitution is also called integration by change of variable. In the general case, given the integral of a function f(x) whose form is non-standard, the method of substitution is to ﬁnd a new variable u(x) such that

1 2

1 8

1 8

Z f ( x ) dx = Z g (u ) du

(6.7)

where the integral on the right is a standard integral; that is, g(u) is easier to integrate than f(x). Differentiating both sides of (6.7) with respect to x gives: on the left side, by deﬁnition of the indeﬁnite integral, d Z f ( x ) dx = f ( x ) dx on the right side, by application of the chain rule, d du du d Z g (u ) du = du Z g (u ) du × dx = g (u ) dx dx Therefore f ( x ) = g (u ) du dx (6.8)

6.3 The method of substitution

167

and, substituting in (6.7),

Z f ( x ) dx = Z g (u)

du dx = Z g (u ) du dx

(6.9)

The essential skill in applying the method of substitution is the ability to recognize when an integrand can be written in the form (6.8). The transformation of the integral is then achieved by the substitution u1=1u(x), or, alternatively, x1=1x(u), dx = dx du. du du = du dx dx (6.10)

EXAMPLE 6.2 Show that

Z( ax + b)n dx =

1 ( ax + b) n +1 + C a( n + 1)

where the numbers a, b, and n are arbitrary, except that n1≠1−1. Let u1=1ax1+1b. Then du1=1a1dx, and

Z( ax + b)n dx = Zu n du =

0 Exercises 11, 12

1 a

1 u n +1 ( ax + b) n +1 +C= +C a n +1 a( n + 1)

EXAMPLE 6.3 Integrate Z f ( x ) dx = Z( x + x 2 )−1 2 (1 + 2 x ) dx . Because (1 + 2 x ) = du1=1(11+12x)1dx and f ( x ) = ( x + x 2 )−1 2 (1 + 2 x ) = u −1 2 Therefore du dx d ( x + x 2 ), dx we make the substitution u1=1x1+1x 2. Then

Z( x + x 2 )−1 2 (1 + 2 x ) dx = Zu −1 2 du = 2u1 2 + C = 2( x + x 2 )1 2 + C

The result is conﬁrmed by differentiation: d 2( x + x 2 )1 2 + C = ( x + x 2 )−1 2 (1 + 2 x ) dx 0 Exercises 13, 14

168

Chapter 6 Methods of integration

EXAMPLE 6.4 Integrate Z e− ( 2 x

2

+ 3x +1)

( 4 x + 3) dx .

Let u1=12x 21+13x1+11. Then du1=1(4x1+13) dx, and

Z e− ( 2 x

2

+ 3x +1)

( 4 x + 3) dx = Z e− u du = − e− u + C = − e− ( 2 x

2

+ 3x +1)

+C

0 Exercises 15, 16

There are no all-embracing rules for ﬁnding the correct change of variable that will transform an integral to standard form; proﬁciency in the art of integration is the result of a lot of practice. Some of the simpler types of substitution are summarized in Table 6.2. Table 6.2 Type

1.

Substitution

u1=1f(x), du1=1f ′(x)1dx u1=1f(x), du1=1f ′(x)1dx

Result

1 Z u du = 2 u 2 + C

Z f ( x ) f ′( x ) dx Z

f ′( x ) dx f ( x)

2.

Z

du = ln u + C u

3.

Z f (sin x ) cos x dx Z f (cos x ) sin x dx

u1=1sin1x, du1=1cos1x1dx

Z f (u ) du

− Z f (u ) du

4.

u1=1cos1x, du1=1−sin1x1dx

EXAMPLES 6.5 Integrals of type 1: Z f ( x ) f ′( x ) dx . (i) I = Zsin ax cos ax dx. In this case, cos1ax is proportional to the derivative of sin1ax (and vice versa); f(x)1=1sin1ax, f ′(x)1=1a1cos1ax

Therefore, putting u1=1sin1ax, du1=1a1cos1ax1dx, I= 1 1 1 Zu du = u 2 + C = sin 2 ax + C a 2a 2a

We note that this integral is identical to case 3 in Table 6.1, and this example demonstrates that there are often several ways of evaluating a particular integral.

6.3 The method of substitution

169

(ii)

I =Z Because

ln x dx. x

1 d 1 = ln x, we let u1=1ln1x and du = dx. Then x dx x

**1 1 I = Z u du = u 2 + C = (ln x )2 + C 2 2 0 Exercises 17–20 EXAMPLES 6.6 Integrals of type 2: Z I =Z x 2x + 3
**

2

f ′( x ) dx. f ( x)

(i)

dx.

In this case, x is proportional to the derivative of 2x 21+13: f(x)1=12x 21+13, f ′(x)1=14x. Therefore, putting u1=12x 21+13 and du1=14x1dx, I= 1 du 1 1 Z = ln u + C = ln( 2 x 2 + 3) + C 4 u 4 4 cos x dx. sin x

(ii)

I = Z cot x dx = Z

Let u1=1sin1x. Then du1=1cos1x1dx, and I =Z du = ln u + C = ln(sin x ) + C u 1 dx. x ln x 1 dx , and x

(iii)

I =Z

Let u1=1ln1x. Then du = I =Z

du = ln u + C = ln(ln x ) + C u

0 Exercises 21–26

170

Chapter 6 Methods of integration

EXAMPLES 6.7 Integrals of type 3: Z f (sin x ) cos x dx. (i) I = Zsin a x cos x dx, where a is an arbitrary number (but a1≠1−1). Let u1=1sin1x. Then du1=1cos1x1dx, and I = Zu a du = 1 a +1 1 u +C= sin a +1 x + C a +1 a +1

**(ii) I = Z esin x cos x dx . Let u1=1sin1x. Then du1=1cos1x1dx, and I = Z eu du = eu + C = esin x + C 0 Exercises 27, 28
**

Trigonometric and hyperbolic substitutions

The standard integrals listed in Table 6.3 can be evaluated by substituting appropriate trigonometric and hyperbolic functions for the variable x.

Table 6.3

1.

Z Z Z Z Z

x = sin −1 + C , a a −x dx

2 2

a 2 > x2

2.

x = cosh −1 + C = ln x + x 2 − a 2 + C , a x −a dx

2 2

x2 > a2

3.

x = sinh −1 + C = ln x + x 2 + a 2 + C a x +a dx

2 2

4.

dx a +x

2 2

=

1 −1 x tan + C a a x 1 1 a+ x + C, tanh −1 + C = ln a 2a a − x a a2 > x2

5.

dx a −x

2 2

=

6.3 The method of substitution

171

Integral 1 in the table is evaluated by means of the substitution x1=1a1sin1θ. Then dx1=1a1cos1θ1dθ and Therefore a 2 − x 2 = a 2 − a 2 sin 2 θ = a 1 − sin 2 θ = a cos θ .

Z

dx a −x

2 2

=Z

x a cos θ d θ = Z d θ = θ + C = sin −1 + C a cos θ a

Similarly, integral 2 is evaluated by means of the substitution x1=1a1cosh1u. Integral 4, evaluated by means of the substitution x1=1a1tan1θ, is used in Section 6.6 for the integration of rational functions. Integral 5 can be evaluated either by means of the substitution x1=1a1tanh1u or by expressing the integrand in terms of partial fractions to give the logarithmic form (see Section 6.6). Alternatively, all the integrals in Table 6.3 are readily obtained by integrating the standard derivatives listed in Tables 4.5 and 4.6. Such substitutions are useful when the integrand contains the square root of a quadratic function.

EXAMPLE 6.8 Evaluate Z a 2 − x 2 dx . Let x1=1a1sin1θ. Then dx1=1a1cos1θ1dθ, a 2 − x 2 = a cos θ , and

Z a 2 − x 2 dx = a 2 Zcos2 θ dθ

This is integral 1 in Table 6.1. Therefore

Z a 2 − x 2 dx =

a2 a2 Z(1 + cos 2θ ) dθ = (θ + sin θ cos θ ) + C 2 2

Now sin1θ1=1x2a, cos θ = a 2 − x 2 a, and θ1=1sin−11(x2a). Therefore,

Z a 2 − x 2 dx = a 2 sin −1 + x a 2 − x 2 + C

a 0 Exercises 29–32

1 2

x

1 2

172

Chapter 6 Methods of integration

Deﬁnite integrals

When the variable of integration of a deﬁnite integral is changed, from x to u(x) say, the limits of integration must also be changed. If the range of integration over x is from a to b then the range of integration over u(x) is from u(a) to u(b):

Z f ( x ) dx = Z

a

b

u ( b)

g (u ) du

(6.11)

u( a )

**EXAMPLE 6.9 Integrate I = Z cos2 x sin x dx .
**

0

π

**Substitute u1=1cos1x and du1=1−sin1x1dx. When x1=10, u1=1cos101=1+1; when x1=1π, u1=1cos1π1=1−1. Therefore I = − Z u du = + Z u 2 du
**

2 +1 −1 −1 +1

**since interchanging the limits changes the sign of the integral. Then u3 +1 2 Z cos x sin x dx = Z u du = = 3 −1 0 −1 3
**

π

+1

2

2

EXAMPLE 6.10 Find the area of the circle whose equation is x 21+1y 21=1a 2. As in Example 5.11, let A be the area of that quarter of the circle that lies in the ﬁrst quadrant, in which both x and y are positive (Figure 5.12). Then y = a 2 − x 2 and A=Z

a

a 2 − x 2 dx

0

As in Example 6.8, the integral is evaluated by means of the substitution x1=1a1sin1θ, and the new integration limits are θ1=10 when x1=10 and θ1=11π22 when x1=1a. Therefore, A= a

2

Z cos2 θ dθ =

0

π 2

π 2 a2 πa 2 θ + sin θ cos θ = 0 2 4

The area of the circle is four times this. 0 Exercises 33–39

6.4 Integration by parts

173

6.4

Integration by parts

In the integral y = Z x cos x dx the integrand is the product of two quite different types of function; the polynomial x and the trigonometric function cos1x. The value of the integral is y1=1x1sin1x1+1cos1x1+1C as can be veriﬁed by differentiation: dy d d cos x = x sin x + dx dx dx

(

)

(

)

=1(sin1x1+1x1cos1x)1+1(−sin1x)1=1x1cos1x The product rule has been used to differentiate the product x1sin1x, and the method of integration by parts is used to integrate products of this type. In general, let y1=1uv, where u and v are functions of x. Then dy dv du =u +v dx dx dx and, integrating both sides of this equation with respect to x, (6.12)

Z

du dv dy dx = Zu dx + Zv dx dx dx dx

(6.13)

The left-hand side is equal to y1=1uv by deﬁnition, and the equation can be rearranged as

Zu

dv du dx = uv − Zv dx dx dx

(6.14)

This equation, the inverse of the product rule, is the rule of integration by parts. Given an integrand like x1cos1x, one of the factors is identiﬁed with u in (6.14), the other with dv2dx. For example, let u1=1x, Then du = 1, dx v1=1sin1x dv = cos x dx

174

Chapter 6 Methods of integration

and (6.14) becomes

Zx

d d (sin x ) dx = x sin x − Z sin x ( x ) dx dx dx = x sin x − Zsin x dx =1x1sin1x1+1cos1x1+1C

The art of integration by parts is the ability to make the correct choice of u and v. Thus, the choice of u1=1cos1x and dv 2dx1=1x in our example gives

**Z x cos x dx = x 2 cos x − Z x 2 × ( − sin x ) dx
**

1 2 1 2

=

1 2 1 x cos x + Z x 2 sin x dx 2 2

and the problem has become more difﬁcult. This demonstrates the rule that if one of the factors is a polynomial then, with only one important exception, the polynomial must be chosen as the function u in equation (6.14).

EXAMPLE 6.11 Integrate by parts Z x 2 cos x dx . Let u1=1x 2 and dv du = cos x. Then = 2 x and v1=1sin1x, and dx dx

**Z x 2 cos x dx = x 2 sin x − 2Z x sin x dx
**

The degree of the polynomial under the integral sign has been decreased by one; x 2 has dv = sin x, gives been replaced by x. Integrating the new integral by parts, with u1=1x and dx

**Z x sin x dx = − x cos x + Zcos x dx = − x cos x + sin x
**

Therefore

**Z x 2 cos x dx = x 2 sin x − 2 − x cos x + sin x + C
**

=1x 21sin1x1+12x1cos1x1−12sin1x1+1C The results may be veriﬁed by differentiation. 0 Exercises 40–45

6.4 Integration by parts

175

In general, a polynomial of degree n can be removed by n successive integrations by parts. The exception to the rule is when the other factor is a logarithmic function.

EXAMPLE 6.12 Integrate by parts Z x n ln x dx , ( n ≠ −1) . In this case, choosing u1=1x n leads to a more complicated integral. The correct choice is u1=1ln1x and dv du 1 = x n . Then = and v1=1x n+12(n1+11), and dx dx x 1 n +1 1 1 x ln x − Z x n+1 × dx n+1 n+1 x 1 n +1 1 Z x n dx x ln x − n+1 n+1 1 n +1 1 x ln x − x n +1 + C n+1 ( n + 1)2 1 ( n + 1)2 x n +1 ( n + 1) ln x − 1 + C

Z x n ln x dx =

= = =

A special case of this integral is Z ln x dx = x ln x − x + C 0 Exercises 46–48

**Integration by parts is straightforward only if one of the factors is a polynomial.
**

∞

EXAMPLE 6.13 Integrate

Z e− ax cos x dx, ( a > 0) .

0

In this case either factor can be chosen as u in (6.14); for example, if u1=1e−ax and dv = cos x then, for the indeﬁnite integral, dx I = Z e− ax cos x dx = e− ax sin x + aZ e− ax sin x dx = e− ax sin x + a − e− ax cos x − aZ e− ax cos x dx = e− ax sin x − ae− ax cos x − a 2 Z e− ax cos x dx =1e−ax1sin1x1−1ae−ax1cos1x1−1a 2I

176

Chapter 6 Methods of integration

**Then, solving for I, I = Z e− ax cos x dx = and
**

∞

1 1 + a2

e− ax (sin x − a cos x ) + C

Z e− ax cos x dx =

0

a 1 + a2

0 Exercises 49–51

6.5

Reduction formulas

The method of integration by parts can be used to derive formulas for families of related integrals. Consider the integral I n = Z x n eax dx where n is a positive integer. Choosing u1=1x n and In = or In = 1 n ax n x e − I n −1 a a (6.15) dv = eax in the formula (6.14), dx

1 n ax n x e − Z x n −1eax dx a a

This result is called a reduction formula for In, or a recurrence relation between In and In−1. For example I3 = where I0 = Z eax dx = Then I3 = 1 ax 3 3 2 6 6 e x − x + 2 x − 3 + C a a a a 1 ax e +C a 1 3 ax 3 x e − I2 , a a I2 = 1 2 ax 2 x e − I1 , a a I1 = 1 ax 1 xe − I0 a a

Then dx v1=1sin1x. Therefore In = and it follows that ∞ n I . because sin21x1=111−1cos21x.14 Determine a reduction formula for I n = Z cos n x dx where n is a positive integer. the quantity e−ar1r n is zero at both integration limits.6. a n −1 n1≥11 I n = Z e− ar r n dr = 0 n! a n +1 (6. For example. Integration by parts gives − In = 1 − ar n n e r + I n −1 a a 0 ∞ When n1≠10. the integral ∞ I n = Z e− ar r n dr 0 (a1>10) (6.5 Reduction formulas 177 More important still. I n = cos n −1 x sin x + ( n − 1) Z cos n − 2 x dx − ( n − 1) Z cos n x dx =1cosn−11x1sin1x1+1(n1−11)In−21−1(n1−11)In . r1=10 and r1→1∞.16) where n is a positive integer or zero. and let u1=1cosn−11x and = cos x. occurs in the quantum-mechanical description of the properties of the hydrogen atom. the recurrence relation is ideally suited for the computation of one or several members of a family of integrals.17) where n!1=1n(n1−11)(n1−12)1-11 is the factorial of n. Reduction formulas are particularly simple for some important deﬁnite integrals in the physical sciences. dv Write the integrand as cosn−11x1cos1x. EXAMPLE 6. and I n = cos n −1 x sin x + ( n − 1)Z cos n − 2 x sin 2 x dx and.

56 ∞ EXAMPLE 6.178 Chapter 6 Methods of integration Solving for In then gives In = 1 n −1 cos n −1 x sin x + I n n n− 2 0 Exercises 52–54 EXAMPLE 6. ∞ Z e 0 2 − ar 2 n r dr = − 2 1 − ar 2 n −1 ( n − 1) e r + Z e− ar r n− 2 dr 2a 2a 0 0 ∞ ∞ Now e− ar → 0 as r1→1∞ and. I4 = 3 3 1 3π I = × I = . if n1>11. 4 2 4 2 0 16 I5 = 4 4 2 8 I = × I = 5 3 5 3 1 15 π 2 n−1 I . r n−11=10 when r1=10. Then.16 Determine a reduction formula for I n = Z e− ar r n dr . I0 = Z π 2 dx = π 0 0 2 0 Exercises 55. it follows that In = with I1 = Z For example. 2 0 Because 2 2 1 d − ar 2 e = −2 ar e− ar . the integrand is written as − −2 ar e− ar r n −1 . π 1 cos n −1 x sin x In = 0 n 2 + n−1 I n n− 2 Because sin 01=10 and cosn−1(π22)1=10 if n1>11. for n1≥12. by parts. 0 From the result of Example 6. 2a dr Then. n n− 2 n≥2 cos x dx = 1.15 Determine a reduction formula for the deﬁnite integral In = Z π 2 cos n x dx .14. .

∞ I1 = Z e 0 − ar 2 2 1 1 r dr = − e− ar = 2a 2a 0 ∞ When n1=10. the molecular orbitals are expressed in terms of ‘gaussian basis functions’.11).6 and 2. The method of partial fractions ∞ 2 ( n − 1) Z e− ar r n− 2 dr . and the use of these functions leads to integrals of the type discussed in this example. The integrals In are important in several branches of chemistry.7 that every such function can be expressed as the sum of a polynomial and one or more partial fractions of types (if complex numbers are excluded) ( i) 1 ( x + a) n .18) This last integral is a standard integral that cannot be evaluated by the methods described in this chapter (see Section 9. 0 Exercises 57–62 6.6. Such a 2 function is essentially the exponential e− ar multiplied by a polynomial.6 Rational integrands. ( ii) ax + b ( x + px + q ) n 2 (6.6 Rational integrands. its discriminant p21−14q is negative. in modern computational methods for the calculation of molecular wave functions. 2a 0 179 ∞ Z e− ar r n dr = 2 In = 0 ( n − 1) I 2 a n− 2 When n1=11. ∞ I0 = Z e− ar dr = 2 0 1 π 2 a (6. For example.20) .19) where n is a positive integer and the quadratic x 21+1px1+1q has no real roots. The method of partial fractions A rational algebraic function has the general form P(x)2Q(x) where P(x) and Q(x) are polynomials: 2 P( x ) a0 + a1 x + a2 x + = Q( x ) b + b x + b x 2 + 0 1 2 + an x n + bm x m It was shown in Sections 2. Integrals of type (i) These are the integrals that occur in the theory of elementary kinetic processes: ln( x + a ) + C Z = −1 n +C ( x + a) ( n − 1)( x + a ) n −1 dx if n = 1 if n > 1 (6. that is.

The cubic in the denominator can be factorized as x31−13x1+121=1(x1−11)2(x1+12) so that the integrand can be expressed in terms of partial fractions as 5x + 1 x − 3x + 2 3 = 1 2 1 + − x − 1 ( x − 1)2 x + 2 Then Z 5x + 1 x − 3x + 2 3 dx = Z dx dx dx + 2Z −Z 2 x −1 x+2 ( x − 1) 2 − ln( x + 2) + C x −1 = ln( x − 1) − x − 1 2 = ln +C − x + 2 x − 1 0 Exercises 63–65 Integrals of type (ii) We ﬁrst consider two special forms.17 Integrate Z dx .21) . the integral is either of type 2 in Table 6. The numerator is the derivative of the quadratic In this case.180 Chapter 6 Methods of integration EXAMPLE 6.2 or it is a simple generalization thereof: Z 2x + p ( x + px + q ) 2 n dx = Z f ′( x ) [ f ( x )] n dx (6.18 Integrate Z dx. ( 2 − x )( 4 − x ) The integrand can be expressed in terms of partial fractions as 1 = ( 2 − x )( 4 − x ) Therefore 1 1 1 − 2 2 − x 4 − x Z dx 1 1 1 = Z − dx ( 2 − x )( 4 − x ) 2 2 − x 4 − x = 1 1 4 − x − ln( 2 − x ) + ln( 4 − x ) + C = 2 ln 2 − x + C 2 5x + 1 x − 3x + 2 3 EXAMPLE 6.

The method of partial fractions 181 If u1=1f(x)1=1x21+1px1+1q then du1=1f ′(x)1dx1=1(2x1+1p)1dx.27) which is one of the standard integrals in Table 6. . and u 21+1a 21=1a 2(tan21θ1+11)1=1a 22cos21θ.14. the integral (6. When n1>11. Because it has been assumed that the quadratic has no real roots.26) Z du u2 + a2 = u θ 1 1 Zdθ = + C = tan −1 + C a a a a (6. for example.3.24) where p21−14q is the discriminant of the quadratic. du (u 2 + a 2 ) n = 1 a 2 n −1 Z cos2 n− 2 θ dθ (6.23) Z dx ( x + px + q ) n 2 The integral is evaluated by ﬁrst transforming the quadratic in the denominator into the form u 21+1a 2. ln( x 2 + px + q ) + C dx = Z 2 −1 n +C ( x + px + q ) 2 n −1 ( n − 1)( x + px + q ) 2x + p 0 Exercises 66.6.26) can be evaluated. Then. Therefore Z When n1=11.22) if if n=1 n >1 (6. 67 The numerator is unity: if if n=1 n >1 (6. but 4q1−1p2 is positive: 2 p 4q − p 2 = u2 + a2 x + px + q = x + + 2 4 2 (6. Then du1=1a1sec21θ1dθ1= (a2cos21θ)1dθ.6 Rational integrands. Z dx ( x + px + q ) 2 n =Z du (u + a 2 ) n 2 We now use the trigonometric substitution u1=1a1tan1θ. by reduction as in Example 6.25) where u1=1x1+1p22 and a 21=1(4q1−1p2)241>10. because dx1=1du. and ln u + C dx = Z n = Z 2 −1 n +C ( x + px + q ) u n −1 ( n − 1)u 2x + p du Therefore. We write 2 p p 2 − 4q x 2 + px + q = x + − 2 4 (6. the discriminant is negative.

27). By equation (6. Z cos4 θ dθ = sin θ cos3 θ + Z cos2 θ dθ 1 3 3 = sin θ cos3 θ + sin θ cos θ + θ + C 4 8 8 We need to change variable from θ back to u (and then to x). and u1=1x1+11. by equation (6.20 Integrate Z dx ( x + 2 x + 5)3 2 .19 Integrate Z dx x + 2x + 5 2 . a1=12. If tan1θ1=1u2a then θ1=1tan−11(u2a) and it is readily veriﬁed that sin θ = Then u u +a 2 2 1 4 3 4 . Then. by reduction as in Example 6. Z dx x + 2x + 5 2 =Z dx ( x + 1) + 2 2 2 = 1 −1 x + 1 tan +C 2 2 0 Exercise 68 EXAMPLE 6. Z dx ( x + 2 x + 5) 2 3 = 1 Z cos4 θ dθ 32 where tan1θ1=1u2a1=1(x1+11)22. cos θ = a u + a2 2 Z cos4 θ dθ = and u 1 ua3 3 ua 3 + + tan −1 + C 4 (u 2 + a 2 ) 2 8 u 2 + a 2 8 a Z dx ( x 2 + 2 x + 5)3 = x + 1 1 2( x + 1) 3( x + 1) 3 + + tan −1 + C 32 ( x 2 + 2 x + 5)2 4( x 2 + 2 x + 5) 8 2 0 Exercise 69 . with n1=13. The quadratic function x21+12x1+15 can be written as (x1+11)21+122.14.182 Chapter 6 Methods of integration EXAMPLE 6. with a1=12 and u1=1x1+11.26). Then.

31) The trigonometric functions sin1θ and cos1θ are rational functions of t.6.6 Rational integrands. Every such function can be integrated by the methods described earlier in this section. dividing the numerators and denominators by cos21θ22 and putting t1=1tan1θ22.29) 0 Exercises 70.32) .30) θ θ − sin 2 = 2 2 cos 2 θ 2 2θ cos 2 − + sin 2 θ 2 2θ sin 2 Then. The method of partial fractions 183 The general form: Z ax + b ( x + px + q ) n 2 dx The numerator can be written as ax + b = a ap 2x + p + b − 2 2 ( ) (6. the substitution θ t = tan . cos θ = 1 − t2 1+ t 2 . t = tan θ 2 (6. A trigonometric function of θ that becomes a rational (algebraic) function of t when we make the substitution t1=1tan1θ22 is called a rational trigonometric function of θ. if the integrand in the integral Z f (θ ) d θ is a rational trigonometric function of θ.28) so that the integral is expressed in terms of the special cases discussed above: Z ax + b ( x + px + q ) 2 n dx = a 2x + p dx Z 2 2 ( x + px + q ) n ap dx + b − Z 2 2 ( x + px + q ) n (6. 71 Rational trigonometric integrands By trigonometry. Thus. we obtain sin θ = 2t 1+ t 2 . 2 dθ = 2 1 + t2 dt (6. 2 sin θ cos θ θ θ 2 2 sin θ = 2 sin cos = 2 2 cos 2 θ + sin 2 θ 2 2 cos θ = cos2 (6.

3. 0 Exercises 73–75 1 2 + tan θ 2 +C ln 4 2 − tan θ 2 This method can be applied in all cases but is not always the simplest in practice.184 Chapter 6 Methods of integration gives Z f (θ ) dθ = Z 2 f (θ ) 1 + t2 dt (6. For example.7 Parametric differentiation of integrals Consider the indeﬁnite integral Z e−α x dx = − e−α x + C .33) where f(θ) in terms of t is a rational function of t. the application of the method to the integration of the elementary trigonometric functions sin1θ and cos1θ is considerable more complicated than the use of the standard integrals. (i) Z Z 2 dθ = Z 1 + t2 sin θ 2t 1 1 + t 2 dt = Z t dt = ln t + C = ln tan θ 2 + C 2 1 − t2 3 + 5 dt = Z dt 1 + t2 8 − 2t 2 =Z dt 4−t 2 (ii) 2 dθ = Z 1 + t2 3 + 5 cos θ = 1 2 + t +C ln 4 2 − t = making use of standard integral 5 in Table 6. differentiation then gives 1 x d d 1 Z e−α x dx = − e−α x + C = 2 + e−α x dα dα α α α .21 Examples of the substitution t1=1tan1θ22. 0 Exercise 72 EXAMPLE 6. 6. α 1 (α ≠ 0) The integral can be treated as a function of the parameter α .

α ) dx = F ( x. by integration by parts. α ) are continuous functions of x and α : dα d d d f ( x . α) and d f ( x . α ) dx = Z F ( a. α ) − Z f ( x. This result is true in the general case Z f ( x. α ) dx = F ( x. α ) dx = F ( b. ∞ EXAMPLE 6.22 Integrate Z e− ax x n dx. α ) dx = Z dα dα dα (6. 1 a ( a ≠ 0) The nth derivative of e−ax with respect to a is (−1)n1xn1e−ax so that ∞ Z x n e− ax dx = ( −1)n 0 dn da ∞ 0 Z e− ax dx = ( −1) n n d n 1 n! = n +1 n a da a .34) (6.5 from a reduction formula derived by successive integrations by parts. An alternative method is to differentiate the simple standard integral ∞ 0 Z e− ax dx = . α ) + C if f(x.6. 0 The integral was evaluated in Section 6.7 Parametric differentiation of integrals 185 In addition. α ) dα dα a dα dα a (6. b b d d d d f ( x . Z It follows that 1 x d −α x e dx = − Z xe−α x dx = 2 + e−α x α α dα d d Z e−α x dx = Z e−α x dx dα dα so that the order of integration with respect to x and differentiation with respect to α can be interchanged in this case.35) For the corresponding deﬁnite integral. if the limits are independent of the parameter α .36) When one or both of the limits of integration are inﬁnite. it is necessary to ensure that the integral of the function and that of its derivative are both convergent. α ) Z f ( x.

37) ∞ ∞ −α x e 0 sin x dx x To obtain the value of the constant of integration. α ) Z f ( x. we note that I(α)1→10 as α1→1∞.186 Chapter 6 Methods of integration ∞ EXAMPLE 6. We consider instead the related integral I (α ) = Z Then d I (α ) = − Z e−α x sin x dx dα 0 and the new integral can be integrated by parts as in Example 6. integrating with respect to α. the method of integration by parts does not work.38) 0 Exercise 76 When the limits of integration also depend on the parameter. α ) dx = Z f ( x . x The integral cannot be evaluated by the standard methods discussed in this chapter. I (α ) = − Z dα 1 + α2 = − tan −1 α + C (6.39) .23 Integrate Z 0 sin x dx. Then d 1 I (α ) = − dα 1 + α2 and.13. α ) dx + f ( b. so that C = lim tan −1 α = α →∞ π 2 To retrieve the original integral we now set α1=10: ∞ I (0) = Z 0 sin x π π dx = − tan −1 0 + = x 2 2 (6. d db da d − f ( a. the result of differentiating the integral is given by Leibniz’s theorem: if a(α) and b(α) are continuous functions of α. α ) dα dα d α a (α ) a (α ) d α b(α ) b(α ) (6.

17. 25. Z sin 3x cos 2 x dx Z cos 5x cos 2x dx 4. 20. + 2x) dx (u1=1x31+12x) 16. Z sin 3x cos 3x dx Z sin 3x sin x dx 3. 6.3 Evaluate the indeﬁnite integrals (use the substitutions in parentheses. dx (u1=1x21+1x1+12) 22. The wave functions for a particle in a box of length l are ψ n ( x) = 2 nπx sin . l l n = 1.6. Evaluate the deﬁnite integrals: 7. 23. Z sin 2 3x dx Z sin x cos 3x dx 2. 26. 3. Z tan x dx Z tan x dx ln (cos x ) .8 Exercises Section 6. Z sin 2x cos 2 x dx 0 π 2 9. 15. cos x dx 1 − sin x x 4 − x2 dx (u1=111−1sin1x) 24. 5. Z (3x + 1)5 dx (u1=13x1+11) Z (3x 2 + 2 x + 5)3 (3x + 1) dx (u1=13x21+12x1+15) Z (3x 2 + 2)e− ( x 3 12. 21. Z cos x e2 sin x dx Z x cos (3x 2 − 1) dx Z 3x 2 − x 2x3 − x2 + 3 dx 19. … Show that the functions satisfy the orthonormality conditions Z ψ nψ m dx = 0 l 1 0 if m = n if m ≠ n Section 6. Z ( 2 x − 1)1 2 dx Z ( 2 x 3 + 3x − 1)1 3 ( 2 x 2 + 1) dx Z (1 − x )e4 x−2 x dx 2 13. Z x 4 − x 2 dx (u1=141−1x2) Z e x (1 + e x )1 2 dx (u1=111+1ex) Z Z Z 2x + 1 x +x+2 2 18.8 Exercises 187 6. 14. Z cos2 3x dx 0 π 2 8.2 Evaluate the indeﬁnite integrals: 1. Z sin x cos 2 x dx 0 π 10. when given): 11. 2.

(i) Use the substitution x1=1a1sinh1u to show that Z (ii) Use the substitution u = x + x 2 + a 2 to show that Z Evaluate the deﬁnite integrals: 33. 45.4 Evaluate the integrals: 40. a x +a dx 2 2 32. Z e− x sin 2x dx 50. An approximate expression for the rotational partition function of a linear rotor is qr = Z ( 2 J + 1)e 0 ∞ − J ( J +1)θ R T dJ where θR1=1A222Ik is the rotational temperature. x +a dx 2 2 Z Z 2 x dx 3x 2 − 2 dx 2− x 2 34. Z x 2e2 x dx Z ln x x2 dx 44. Z x 2 ln x dx 0 1 49. Z ( x + 1)2 cos 2x dx Z x ln x dx Z eax cos bx dx 43. 51. Z ln (cos x ) sin x dx Z x 2 dx 1 − x2 (x1=1sin1θ) 31 x dx (u = x ) 1+ x x = sinh −1 + C . ω0 39. 47. 48. Z π2 1 1 0 ∞ sin ( x + π ) dx x 2 35. Z sin3x cos x dx (u1=1sin1x) Z Z dx 4 + x2 28. 30.188 Chapter 6 Methods of integration 27. = ln x + x 2 + a 2 + C . 37. Z π 2 sin θ cos θ dθ 0 36. Z x sin x dx Z xe x dx 0 1 41. I is the moment of inertia. Z π 2 e−2 x cos 3x dx 0 . 46. Z xe− x dx 0 0 38. Section 6. Z x 3 sin x dx Z x 2e−2 x dx 0 ∞ 42. Line shapes in magnetic resonance spectroscopy are often described by the Lorentz function g (ω ) = ∞ 1 T π 1 + T 2 (ω − ω 0 )2 . and k is Boltzmann’s constant. 29. Evaluate the integral. Find Z g (ω ) dω .

8 Exercises 189 Section 6. Z r 3 e−2r dr 2 ∞ 0 60. For the Maxwell–Boltzmann distribution in Exercise 60. for integers m1≥10 and n1≥11. where v 2 = Z v 2 f (v ) dv . Use the results of Exercises 52 and 53 to evaluate Z sin5 x cos 4 x dx. Z . ﬁnd the root mean square speed v 2 . 0 π 2 57. 0 ∞ 62. Z r e−2r dr 2 ∞ 0 2 58.6. Show that. Z sin m θ cos n θ dθ = sin m+1 θ cos n−1 θ n − 1 + Z sin m θ cos n−2 θ dθ m+ n m+ n 54. Evaluate the integral for the gaussian curve g (ω ) = 2 1 T exp − T 2 (ω − ω 0 )2 2 π Section 6. 53. The second moment of a signal centred at angular frequency ω0 is Z (ω − ω 0 )2 g (ω ) dω ω0 ∞ where g(ω) is a shape function for the signal. where n is a positive integer. ( x + 2) dx ( x + 3)( x + 4) ( x 2 − 3x + 3) dx ( x + 1)( x + 2)( x + 3) Z dx ( 2 x − 1)( x + 3) 64. Z Evaluate 56.5 52. 55. Determine a reduction formula for Z sin n x dx. 0 ∞ 61. for integers m1≥10 and n1>11. The probability that a molecule of mass m in a gas at temperature T has speed v is given by the Maxwell–Boltzmann distribution m 2 − mv 2 f (v ) = 4 π v e 2 π kT 32 2 kT where k is Boltzmann’s constant. Line shapes in spectroscopy are sometimes analysed in terms of second moments. Z 65. Show that. Z r 2 e−2r dr 0 ∞ 59. π 2 sin m θ cos n θ dθ = 0 n−1 Z sin m θ cos n−2 θ dθ m+ n 0 π 2 Z sin5 x cos5 x dx.6 Evaluate the indeﬁnite integrals: 63. Find the average speed v = Z v f (v ) dv .

show that dθ = dt (Equation (6. Z Z x+2 x2 + 4x + 5 dx dx 67. show that Z x 2 ne− ax dx = 2 ∞ 1⋅ 3⋅ 5 0 2 n+1 a n ( 2 n − 1) π a . ( x 2 + 4 x + 5)2 70. By differentiation of the integral Z e− ax dx = 2 ∞ 0 1 π 2 a with respect to a.190 Chapter 6 Methods of integration 66.7 76. Z Z x ( x 2 + 3)( x 2 + 4) x x2 + 4x + 5 dx dx 68. Z Z dx x2 + 4x + 5 4x + 3 ( x 2 + 4 x + 5)2 dx 69. If t = tan . dθ cos θ dθ 5 − 3cos θ dθ 1 + sin θ + cos θ Z 74. 71. Z Section 6. Z 75.33)) 2 1+ t2 Evaluate by means of the substitution t1=1tan1θ22: 73. θ 2 72.

We will see in Chapters 12–14 that the differential equations that are important in the physical sciences often have solutions that can only be represented as series. for example. u11+1u21+1u31+1u41+1where the dots mean that the sum is to be extended indeﬁnitely (ad inﬁnitum). such as the exponential function and other transcendental functions. 3. both in formal theory and in approximate methods such as the method of ‘linear combination of atomic orbitals’ in molecular-orbital theory (LCAO-MO). and in 7. ﬁnite series in 7. variables.8. The terms can be numbers. = . containing a ﬁnite number of terms. = with a rule that speciﬁes each term. as are some important physical quantities. the numbers 1. and the representation of functions as series is an essential tool for the solution of many physical problems.7 how they are used to obtain approximate values of functions.6 we discuss how the MacLaurin and Taylor series can be used to represent certain types of function as power series (‘inﬁnite polynomials’). A series can be ﬁnite.2. solutions of the Schrödinger equation are often represented as series.2 Sequences A sequence is an ordered set of terms u1. Sequences are discussed in Section 7. u2. Approximate and numerical methods of solution of problems are often based on series.3. An important application of series is in the analysis of wave forms in terms of Fourier series and Fourier transforms.1 Concepts A series is a set of terms that is to be summed. inﬁnite series and tests of convergence in 7. For example. 5. 7. the partition function in statistical thermodynamics. functions.7 Sequences and series 7. u3. In Section 7. Series occur in all branches of the physical sciences. For example.4 and 7. are deﬁned as series. Some properties of power series are described in Section 7. The terms themselves form a sequence. Some functions. u11+1u21+1u31+1-1+1un or it can be inﬁnite.5. Fourier analysis is discussed in Chapter 15. or more complex quantities. 7.

2. 343 mice. 343. . with each loaf were 7 knives. … 3 9 27 1 ur +1 = − ur . u11=10 ur1=111−12(r1−11). r1=11.192 Chapter 7 Sequences and series form a sequence deﬁned by the general term ur1=111+12(r1−11). = 1. u11=1a or ur1=1a1+1(r1−11)d. 3. r1=11.1= (7.2) (7. but important. −5. 10. This sequence is an example of the arithmetic progression a. 2. … 2 4 8 ur+11=17ur. 3. ur = 1 2r . 7. 1. each woman had 7 mules and each mule carried 7 sacks. 2. . r1=11.4) (7. . u11=11 r = 0. 3. −3.1) Another simple. u11=11 so that. the rule can be expressed as a recurrence relation plus an initial term: ur+11=1ur1+12. and each knife was put up in 7 sheaths’. = with rule ur+11=1xur. = (ii) Geometric progression:1 1. each sack contained 7 loaves. a1+12d. ax 3. EXAMPLES 7. a1+13d. = 1. − .1= 1 1 1 1. = with rule ur+11=1ur1+1d. 49 cats. 3 1 The Rhind papyrus (c. −1. 2. Ives problem’ in Fibonacci’s Liber abaci (1202 AD) is ‘7 old women went to Rome. 31= (7. 5. The version of the ‘St. 16807 hekats’. a1+1d. − .1 Sequences (i) Arithmetic progression: 0. … u1 = 1 ur+11=1ur1+15. 2.3) A sequence of terms ur is denoted by {ur}. . 1 1 1 . for example. u11=1a or ur1=1ax r−1. sequence is the geometric progression a. ax. 1650 BC) contains a problem concerning ‘7 houses. 15. ax 2. 2401 ears of grain. 3. 49.1= Alternatively. r1=11. u51=1u41+121=171+121=19.

… 2 3 4 5 has general term ur1=1(r1−11)2r and limit lim ( r − 1) r = 1 . or rabbit problem. The quantity 1 lim = 0 r →∞ r is called the limit of the sequence.7. 2. . 2. . … (iv) Fibonacci sequence (‘series’):2 1. and in this case the limit is ﬁnite and unique. 1 1 1 . = and when r1=110n.9. u1001=10. the sequence 1 2 3 4 . r 1. . 8. and they begin to breed in the second month after birth?’ The Fibonacci sequence has been linked with various patterns of growth and behaviour in nature. . Similarly. … 2 3 4 1 ur = .999. . r →∞ u101=10. = 0 Exercises 1–10 ur+21=1ur+11+1ur. 1. 5. 3. When the 2 This is the solution to the paria coniculorum. the term ur has n 9’s after the decimal point.2 Sequences 193 (iii) Harmonic sequence: 1. 13.99. Thus. 1 1 1 . r r = 1. u01=1u11=11 Limits of sequences 1 The terms of the harmonic sequence . and approach the value zero as r tends to inﬁnity. . 3. u10001=10. given in Fibonacci’s Liber abaci: ‘how many pairs of rabbits can be bred from one pair in one year if each pair breeds one other pair every month. … 2 3 4 decrease in magnitude as r increases. . When the limit u of a sequence is ﬁnite and unique the sequence is said to converge to the limit u. .

. .. . .. . . ... ... . ..... . ..... . .. . . . .. . .... . ..... ... . . .. ) f( )e( )d( )c( The examples shown are (for a1=11): )b( (f) x1<1−1 (e) x1=1−1 (d) −11<1x1<10 (c) 01<1x1<11 (b) x1=11 (a) x1>11 x r +1 ( r + 1)! e3 = 1 + )a ( 3 32 33 34 35 + + + + + 1! 2! 3! 4! 5! 1 1 1 1.. ... . . .... . . .. .. ...... . . … 2 4 8 xr x = → 0 as r → ∞ for all x r! r +1 Figure 7. . . ... .... . . . .. ... .... .. ... . . .. .... . . .. . . . ..... .. . . .. .. .. . . . .. . .. … 2 4 8 1. 1. . ... .. . It shows the six possible types of behaviour. .. ....... . . ........ ... ...... . ... . . .. . . ... ... ...... . . .. . .. . . . . . ... ... = 1.. . . .... ..... . .. .... .. . . .. . . .... ... ... . . .. .... . ... . ... .... .... .. ... ............ . . . .. . . .... .. ..... .. . . . ..... . ... . . .. ... . ... . ... . ... . ...... ...... .. ....... . . ... . ..... . . .... ... . . .. . . ... . ... . . .... . . ... .... . . . . .. .... . ... ... . . ... .. ... .. . . . .... .. .. ..... .. . .... ........1) is necessarily divergent for all values of d: Chapter 7 Sequences and series r →∞ lim a + d ( r − 1) = ± ∞ . . .. . .. . ... . ... . . ... ... .. . . ... ... . ......... .. ...375 + 2. ..... thus. .. .. . . . . ........ .. . .. .. . ... ... . . .... . .... . .. .. .. .. .... .194 The geometric progression (7. . . ..... . .. ... ....... . For example. . .. .. . .. ..... .. .. .... . .. . . .... . .. . . r The sequence is convergent only for x1=1+1 and | x|1<11.. . ..3) is interesting because it is convergent for some values of x and divergent for others. ..... .... .. . ... . ... . ...... .. . ..... ...025 + . . .. ..... . 4. −1... ... . . . . .. . . .. . . ... .......... . ... = 1. . .. ... . . 1. . .. . . . ....... ..... .... . . . . .... . .. ....... ... .. ... ... ... . .. .. .. .. . . limit is not ﬁnite or not unique the sequence is divergent. .. .. −2.. .. . . . . .. . .. .... . . . .. . .. − .. . .... . ... .. .. . .. ... . ... ..... . . ... . . . . ... ....... ... .. .. .. . . .... .... . .. .. . . . ........... . ... . .. decrease when r1+111>1|x|..... . ... .... ... .... . .. .... .... . . .... ..... . ... ... .. .. . . . .. .. ....... . . .. . ... . . ... ... ... .. . . ... . .... ... . .. 0 0 1 ru . . . ........ . . . .....6). For example. . .... ... .... ... ........ .. .. . .... .... .. ... . 1. . . .. .. . . . ... .... .. . . ... ... . .. . . 8.. . ... ... .. . . .... ........ ...... . ..... . . .. .... −8. . .. . . .. . .... . .. . . . . . .. −1. − .. ... .. .. . .... . . . . .. . ... ... . . . . . .1 = 1 + 3 + 4. .. .1... .. ... . . .. types (b). . 1.. .. ...... . . ...... = 1 1 1 .. 1.. ... .. .. . ....... .... . ... ..... . .. . . . ... . .. .. .. .. illustrated in Figure 7. ...... .. . .. . 2....... 4. .. and (d).. . . . ... .... .. . . . . ..... .... .. ... .5 + 4. ... . .... .. .. .. . .. . .. .. .. . .... . . . ... .. .. .. .. . . ......... . . . ..... . . . ... . ...... . ... . . .... ... .. .. .. . .. ......... .. ... . . An important example is the sequence of the terms {xr2r!} in the expansion of the exponential function (Section 3. .. . ... characteristic of many sequences. ..... .. . .. . .. ... .... . . . .. . . . . .. . . .. .. . . .. ... (c). .... .... . . . .. . .. . ... . . ... . . . . .. .. . . . . .. 0 1 0 1 1 1 0 1 . This sequence has limit zero for all values of x... .. ..... . . .. .. ..5 + 3. = 1. .... .. . . . .. . .. . . ..... .. .. ... The behaviour of a sequence in the limit does not necessarily depend on the behaviour of a ﬁnite part of the sequence. . . . . . ....... .... . .... . . ...... . .... .. .. ...... . .. .. . . .... .. .... . ... .. .. . .. .. ... the ratio of consecutive terms is but the terms increase in magnitude when r1+111<1|x|. ... the arithmetic progression (7. .... ... .. .

2 Sequences 195 EXAMPLES 7. Giorgio Vasari (1511–1574) accuses Pacioli of plagiarising the mathematical works of Piero della Francesca (c. Then φ 21−1φ1−111=10. proportioni et proportionalita (Venice.7.6190.618034 . This is identical to the quantity known in geometry as the ‘golden section’ (ratio).6181. 1. In his ‘Lives of the artists’.3 0 Exercises 11–17 3 The golden section was known to the Greeks as ‘the division of a straight line in extreme and mean ratio’ (Euclid. Book II. Its present name originated in 15th Century Italy when it was taken up by artists as a ‘divine proportion’ and used in painting and architecture.6350. with positive solution φ = (1 + 5 ) 2 . 1. 2. 1. ur+12ur. 2 r 2 We have ur = (ii) ur = 2r 2 + 2r + 1 r2 − r +1 2 + 2 r +1 r2 1− 1 r + 1 r2 → 2 as r → ∞ Dividing top and bottom by r 2. Pacioli’s Summa de arithmetica. r →∞ (i) ur = r−2 2r 1 1 1 − so that ur → as r → ∞ .6176. Luca Pacioli (1445–1517) wrote De divina proportione (1509). 1. 1. with illustrations thought to be by Leonardo da Vinci. it contained the ﬁrst published description of doubleentry bookkeeping. . r →∞ ( ) Thus. Book VI.2 Limits lim (ur ) . 1. Proposition 11.1(iv)) is divergent. but a convergent sequence is obtained from the ratios of consecutive terms. and the limit of the sequence is lim ur +1 ur = φ = (1 + 5 ) 2 ≈ 1. 1. Proposition 30). ur+2 2ur+11=111+1ur 2ur+1 and taking the limit r1→1∞ gives φ1=111+112φ. ‘The elements’. The ﬁrst 10 terms are 1.6666. dividing the recurrence relation ur+21=1ur+11+1ur by ur+1. ur = (iii) 1 ur = sin → sin 0 = 0 as r → ∞ r (iv) The Fibonacci sequence (Example 7. geometrica. 1494) was one of the ﬁrst comprehensive mathematics book to be printed.5.6153. 1420–1492).6. 1.

the limit n S = lim S n = lim ∑ ur n→ ∞ n → ∞ r =1 (7. The arithmetic series The sum of the ﬁrst n terms of the arithmetic progression (7. = the partial sums S11=1u1 S21=1u11+1u2 S31=1u11+1u21+1u3 with general term S n = u1 + u2 + u3 + + un = ∑ ur r =1 n (7. The word series is commonly also used for the sums of terms themselves. u2. S2.196 Chapter 7 Sequences and series 7.3 Finite series Given a sequence u1. A ﬁnite series of n terms is then ∑ ur = u1 + u2 + u3 + r =1 n + un with sum Sn . when the sequence converges. u3. S3. S1.5) also form a sequence.6). = A sequence obtained in this way is called a series and.6) is called the sum of the series. Sn1=1[a1+1(n1−11)d]1+1[a1+1(n1−12)d]1+1-1+1a .1) is S n = ∑ [ a + ( r − 1) d ] = a + [ a + d ] + [ a + 2 d ] + r =1 n + [ a + ( n − 1) d ] The value of the ﬁnite series is obtained by considering the sum in reverse order. and the inﬁnite series is ∑ ur = u1 + u2 + u3 + r =1 ∞ with sum (if it exists) given by (7.

xSn1=1ax1+1ax 21+1ax 31+1ax41+1-1+1ax n and subtract the two series term by term: Sn1−1xSn1=1a1−1ax n1=1a(11−1x n ) Therefore4 1− xn . . Proposition 35. term by term.8) The sum of the ﬁrst n terms of the geometric progression (7.7. 1− xn = 1 + x + x 2 + x3 + 1− x + x n −1 (7.10) ( x ≠ 1) (7. addition of the two forms. Sn = a 1− x Then.3 Finite series 197 Then. multiply by x.7) +n= 1 n ( n + 1) 2 (7. 19 The geometric series (7.3) is S n = ∑ ax r = a + ax + ax 2 + ax 3 + r =0 n −1 + ax n −1 To obtain the value of the series. Book IX.9) 4 A discussion of the sum of the geometric series is given in Euclid’s ‘Elements’. for a1=11. gives 2Sn1=1[2a1+1(n1−11)d]1+1[2a1+1(n1−11)d]1+1-1+1[2a1+1(n1−11)d] =1n[2a1+1(n1−11)d] so that n S n = [2 a + ( n − 1) d ] 2 In particular. the sum of the ﬁrst n natural numbers is (a1=1d1=11) Sn = 1 + 2 + 3 + 0 Exercises 18.

0 Exercises 20–23 The binomial expansion The binomial expansion is the expansion of the function (11+1x)n in powers of x when n is a positive integer. (1 + x ) n = 1 + nx + with general term n( n − 1)( n − 2)…( n − r + 1) r x r! n( n − 1) 2 n( n − 1)( n − 2) 3 x + x + 2! 3! + xn (7. (1 + x )6 = 1 + 6 x + 6×5 2 6×5× 4 3 6×5× 4×3 4 x + x + x 2 ×1 3× 2 ×1 4 × 3× 2 ×1 + 6 ×5× 4 × 3× 2 5 6 ×5× 4 × 3× 2 ×1 6 x + x 5× 4 × 3× 2 ×1 6 ×5× 4 × 3× 2 ×1 = 1 + 6 x + 15x 2 + 20 x 3 + 15x 4 + 6 x5 + x 6 0 Exercises 24. Examples of such expansions are (11+1x)21=111+12x1+1x 2 (11+1x)31=111+13x1+13x 21+1x 3 (11+1x)41=111+14x1+16x 21+14x31+1x 4 In the general case. 25 . This concept of the expansion of one function in terms of a set of (other) functions provides an important tool for the representation of complicated (or unknown) functions in the physical sciences.11) EXAMPLE 7.3 Expand (11+1x)6 in powers of x. with n1=16.198 Chapter 7 Sequences and series This equation can be regarded in two ways: (i) the value of the sum (11+1x1+1x 21+1-1+1x n−1) is (11−1x n)2(11−1x). By equation (7.11). (ii) the series (11+1x1+1x 21+1-1+1x n−1) is the expansion of the function (11−1x n)2(11−1x) in powers of x.

14) EXAMPLE 7.13) A more general form of the binomial expansion is n n ( x + y )n = ∑ x r y n− r r r =0 (7. Therefore.3 Finite series 199 The coefﬁcient of x r in the expansion (7. 2. we shall see in Chapter 21 that n is the number of combinations of n r r objects taken r at a time. 5 5 5! = = 0 5 0! 5! = 1. the expansion is n n n (1 + x ) n = ∑ x r r r =0 (7. 5 5 5! = = 2 3 2! 3! = 10 r = 0. The binomial coefﬁcients are important in probability theory. In terms of binomial coefﬁcients.7. 5 5 5! = r 5 − r = r!(5 − r )! . 5 5 5 5 5 5 (1 + x )5 = x 0 + x1 + x 2 + x 3 + x 4 + x5 0 3 1 4 2 5 = 1 + 5x + 10 x 2 + 10 x 3 + 5x 4 + x5 5 5 5! = = 1 4 1! 4! = 5. By the deﬁnition of the binomial coefﬁcients. 5 .13). (i) By equation (7. remembering that 0!1=11. where they are often given the symbol Cr (or C n).4 Calculate the binomial coefﬁcients 5 . and use them to expand r 5 5 (i) (11+1x) and (ii) (x1+13) in powers of x. 1. ….12) and is called a binomial coefﬁcient (sometimes read as ‘n choose r’).11) is n n( n − 1)( n − 2)…( n − r + 1) n! = = r r! r!( n − r )! (7.

with applications in games of chance. The multinomial expansion This is the generalization of the binomial expansion (7.14). The Pascal triangle appears in the Traité du triangle arithmétique. The triangle was known long before.15) in which the k-fold sum is over all positive integer and zero values of n1. it appeared in a book by the Chinese mathematician Yang Hui in 1261. (c. ( x1 + x2 + + xk ) n = ∑ ∑ n1 n2 ∑ nk n! n n x 1x 2 n1 ! n 2 ! nk ! 1 2 xk nk (7. and each interior number is the sum of the two numbers diagonally above it. 1425). avec quelques autres petits traités sur la même manière. 5 5 5 5 5 5 ( x + 3)5 = x 0 35 + x1 34 + x 2 33 + x 3 32 + x 4 31 + x5 30 5 0 3 1 4 2 = 243 + 405x + 270 x 2 + 90 x 3 + 15x 4 + x5 0 Exercises 26–33 The binomial coefﬁcients form a pattern of numbers called the Pascal triangle5 1 1 1 1 1 1 1 = 5 4 10 3 6 10 2 3 4 5 1 1 1 1 1 Each row begins and ends with the number 1 for the coefﬁcients of x n and y n in the expansion of (x1+1y)n. The multinomial coefﬁcients n n! = n1 n 2 nk n ! n ! n ! 1 2 k 5 Blaise Pascal (1623–1662). with Fermat. French philosopher and mathematician who made contributions to geometry. n2. published posthumously in 1665. the calculus and. nk .1=1.14). and the properties of the binomial coefﬁcients were discussed by the Persian Jamshid Al-Kashi in his Key to arithmetic. developed the mathematical theory of probability (at the instigation of the gambler Antoine Gombard.200 Chapter 7 Sequences and series (ii) By equation (7. The work contains a discussion of the properties of the binomial coefﬁcients. Chevalier de Méré). subject to the constraint n11+1n21+1-1+1nk1=1n. .

(1. 1. 2).5 Expand the trinomial (a1+1b1+1c)3. EXAMPLE 7.6) and are used in a popular derivation of the Boltzmann distribution in statistical thermodynamics (Example 21. 0). (1. (2. ∑ r (r + 1) = 1⋅ 2 + 2 ⋅ 3 + r =1 1 1 1 + 1 . n3) are (3. (0.3 Finite series 201 are important in combinatorial theory (Section 21. 2). 0. =6 3!0!0! 2!1!0! 1!1!1! Therefore (a1+1b1+1c)31=1(a 31+1b31+1c 3)1+13(a 2b1+1a 2c1+1ab 21+1ac 21+1b 2c1+1bc 2)1+16abc 0 Exercises 34. 0). (0. n( n + 1) . 0). (1. (0. 1). 0. 0). therefore. 2. = 3. 3). (2.7.6 Find the sum of the series The general term can be written as 1 1 1 = − r ( r + 1) r r + 1 and. 0. 35 The method of differences Many simple ﬁnite series can be summed if the general term ur can be written as the difference ur1=1vr1−1vr−1 Then ∑ ur = ∑ v r − ∑ v r −1 r =1 r =1 r =1 n n n =1(v11+1v21+1-1+1vn−11+1vn)1−1(v01+1v11+1-1+1vn−1) =1vn1−1v0 n EXAMPLE 7. 1. 1) with distinct multinomial coefﬁcients 3! 3! 3! = 1. 1). The possible values of (n1. 2.12). 3. 1. n2. (0. 0.

With his brother Johann (1667–1748). and the calculus of variations.1 Some ﬁnite series ∑ r = 1+ 2 + 3+ r =1 n r =1 n n +n = 1 2 1 1 n + n = n( n + 1) 2 2 2 ∑ r 2 = 1 + 22 + 32 + ∑ r 3 = 1 + 23 + 33 + r =1 n 1 1 1 1 + n2 = n3 + n2 + n = n( n + 1)( 2 n + 1) ( 3 2 6 6 + n3 = 1 4 1 3 1 2 1 2 n + n + n = n ( n + 1)2 4 2 4 4 ∑ r 4 = 1 + 24 + 34 + r =1 n 1 1 1 1 + n4 = n5 + n4 + n3 − n 5 2 3 30 ∑ r (r + 1) = 3 n( n + 1)( n + 2) r =1 n 1 ∑ r (r + 1)(r + 2) = 4 n( n + 1)( n + 2)( n + 3) r =1 n 1 ∑ r (r + 1) = n + 1 r =1 n 1 n ∑ r (r + 1)(r + 2) = 4 − 2( n + 1)( n + 2) r =1 1 1 1 The sums of many other series can be derived from these. It was Jakob Bernoulli who ﬁrst used the word ‘integral’. Newton’s method of ﬂuxions was never well known outside England.8) is the simplest example of the family of series ∑ r where m is a positive integer.202 Chapter 7 Sequences and series ∑ r (r + 1) = ∑ r − ∑ r + 1 r =1 r =1 r =1 n 1 n 1 n 1 1 1 1 = + + + 1 2 3 = 1− 0 Exercises 36–39 Some ﬁnite series 1 1 1 + − + + n 2 3 + 1 1 + n n + 1 1 n = n +1 n +1 m The arithmetic series (7. that is. the sum of powers of the natural numbers. Table 7. theory of differential equations. Professor of mathematics at Basel. professor of mathematics in Groningen and at Basel.6 some of these sums and other ﬁnite series are given in Table 7. he also made contributions to the calculus. . by 1700 most of the elementary calculus (described in this book) had been developed. 6 The general method for generating these sums is due to the Swiss mathematician Jakob Bernoulli (1654–1705). series.1. and in collaboration with Leibniz. It was this collaboration that led to the success of Leibniz’s formulation of the calculus.

that is. and 1− xn 1 = lim n→ ∞ 1 − x 1 − x .1 (type (b) only for the trivial case x1=10). x n1→10 as n1→1∞.4 Inﬁnite series 203 EXAMPLE 7.7. The sequence of sums shows all six types of behaviour illustrated in Figure 7.16) where the dots mean that the sum is to be extended indeﬁnitely. 2. 3. and converges only when | x|1<11. when the sequence of partial sums converges. when | x|1<11.4 Inﬁnite series The limit of a sequence of partial sums is the (sum of the) inﬁnite series n un n → ∞ r =1 S = lim ∑ = u1 + u2 + u3 + (7. The geometric series The geometric series is the limit of the sequence of partial sums Sn = 1 + x + x 2 + + x n −1 = 1− xn .7 Find the sum of the ﬁrst n terms of the series 21·151+131·171+141·191+1The general term is ur1=1(r1+11)(2r1+13) for r1=11. 1− x x ≠1 (when x1=11 the sum is n and the series diverges). Thus.1= Then ur1=12r(r1+11)1+13r1+13 and ∑ ur = 2∑ r (r + 1) + 3∑ r + 3∑ 1 r =1 r =1 r =1 r =1 n n n n 1 1 = 2 × n( n + 1)( n + 2) + 3 × n( n + 1) + 3 × n 3 2 = 0 Exercise 40 1 n( 4 n2 + 21n + 35) 6 7. The series has a sum only if the limit is ﬁnite and unique.

204 Chapter 7 Sequences and series so that the geometric series is the expansion of the function 1 = 1 + x + x 2 + x3 + 1− x 1 in powers of x: 1− x (7. and smallest. 7 . 1350). 0 Exercises 41–45 The harmonic series S = 1+ 1 1 1 + + + 2 3 4 Despite appearances. It can be written as a sum of partial sums.5 Tests of convergence The example of the geometric series demonstrates that it is straightforward to show that a series converges.17) . has value 122n+1. if a closed formula is known for the partial sums.7 7. 1 1 1 1 s1 = + > + . this series diverges. Each of these sums is therefore larger than 2n1×1(122n+1)1=1122. |x| < 1 The series diverges for all other values of x (| x|1≥11). This demonstration of the divergence of the harmonic series and discussions of other inﬁnite series are found in Oresme’s Quaestiones super geometriam Euclidis (c. for example. S = 1+ 1 1 1 1 1 1 1 + + + + + + 2 3 4 5 6 7 8 1 1 1 1 1 1 1 1 + + + + + + + + + 9 10 11 12 13 14 15 16 = 1+ 1 +s +s +s + 2 1 2 3 in which sn contains 2n terms of which the last. 3 4 4 4 It follows that S > 1+ 1 1 1 1 + + + + 2 2 2 2 1 1 1 1 1 1 1 1 s2 = + + + > + + + 5 6 7 8 8 8 8 8 and the series diverges. or otherwise.

the example of the harmonic series shows that when such a formula is not known less direct procedures must be followed. (i) p1=11: S is the harmonic series. (iii) p1>11: write the series as 1 1 1 1 1 1 S = 1+ p + + p + p + p + p + p + 2 3 4 5 6 7 =111+1s11+1s21+1- . (ii) If series B diverges. On the other hand.7.8 The series S = 1+ 1 2 p + 1 3 p + 1 4p + converges if p1>11 and diverges if p1≤11. the series can be tested for convergence in a number of ways. then series A converges if ar1≤1br. Then: (i) If series B converges. (ii) p1<11: each term of S (after the ﬁrst) is larger than the corresponding terms of the harmonic series. Given a series ∑ ar = a1 + a2 + a3 + r =1 ∞ a necessary ﬁrst condition for convergence is that the limit of the sequence {ar} be zero: ar1→10 as r1→1∞ If this condition is satisﬁed. and S diverges. The comparison test Let A1=1a11+1a21+1-1+1ar1+1B1=1b11+1b21+1-1+1br1+1be two series of positive terms. and diverges.5 Tests of convergence 205 The sum of the inﬁnite series is then the limit of the sequence of partial sums. EXAMPLE 7. then series A diverges if ar1≥1br.

is 122np. and largest. des arts et des métiers (1751–1772) he ﬁrst gave the derivative as the limit of a quotient of increments but. it was probably ﬁrst given in 1776 by the Cambridge mathematician Edward Waring (1734–1793). and further tests are necessary. Secretary of the French Academy. He gave the complete solution of the precession of the equinoxes. and sometimes to Cauchy. because of the conceptual difﬁculties associated with the limit as a process consisting of an inﬁnite number of steps. independent of r. In his Différentiel (1754) in the Encyclopédie des sciences. EXAMPLES 7. Each sum sn is then less than 2n 2 np 1 = p −1 and 2 2 3 n 1 1 1 S < 1 + p −1 + p −1 + p −1 + 2 2 2 The series on the right is a convergent geometric series 0 Exercises 46. he is best known for his contribution to the post-Newtonian development of mechanics with his principle of virtual work (d’Alembert’s principle) published in his Traité de Dynamique (1743). . and inﬁnite series. Although the ratio test is usually ascribed to him. 47 d’Alembert’s ratio test8 ∑ xr with x1=1122 p −11<11.9 The ratio test (i) The geometric series 11+1x1+1x21+1x31+1The general term is ar1=1x r so that ar+11=1x r+1 and lim ar +1 ar =|x| ar +1 ar = x. He contributed to the theory of partial differential equations. and the test fails for x1=1±1. In this case it is readily shown by inspection that the series diverges when x1=1±1. 8 Jean LeRond d’Alembert (1717–1783). Then r →∞ and the series converges if | x|1<11.206 Chapter 7 Sequences and series in which sn contains 2n terms of which the ﬁrst. diverges if | x|1>11. the calculus. The series a11+1a21+1-1+1ar1+1ar+11+1(i) converges if lim ar +1 ar <1 r →∞ (ii) diverges if lim ar +1 ar r →∞ >1 (iii) may do either if the limit equals 1. the deﬁnition was not accepted until the work of Cauchy (1821).

0 Exercises 48. inﬁnite series. Then r x Z 1 ∞ 1 dx = ln x 1 x and ln1x1→1∞ as x1→1∞. and limits (see d’Alembert). The integral test is sometimes named for MacLaurin. The series then ∞ converges if Z a( x ) dx 1 ∞ converges ( is finite and unique) diverges if Z a( x ) dx 1 diverges EXAMPLE 7. .be a series of decreasing positive terms. Then ar r +1 ( r + 1)! r! r →∞ lim = lim r →∞ x =0 r +1 and the series converges for all values of x. Let a(x) be a function of the continuous variable x such that a(x) decreases as x increases and a(r)1=1ar .5 Tests of convergence 207 (ii) The exponential series 1 + x x 2 x3 + + + 1! 2! 3! The general term is ar = ar +1 ar ar +1 x xr x r +1 = so that ar +1 = and . He made contributions to. The harmonic series therefore diverges.10 The harmonic series 1 + In this case. ar+11<1ar .7. The leading French mathematician of the ﬁrst half of the nineteenth century. with the Cauchy integral theorem and the calculus of residues. he is best known for his work on the theory of functions of a complex variable. 0 Exercises 50. partial differential equations. ar = ∞ 1 1 + + 2 3 . He invented the word determinant for his class of alternating symmetric functions (1812). the theory of elasticity. 51 9 Augustin-Louis Cauchy (1789–1857). amongst others. 49 Cauchy’s integral test9 Let a11+1a21+1-1+1ar1+1. 1 1 and a( x ) = .

It also diverges when x1=11. The convergence properties of such series can be investigated by the methods described in the previous section.11 Radius of convergence (i) The coefﬁcient of xn in the series xn ∑n n=1 ∞ is cn = 1 .6 that the sum of this series is ln12. c1.18) n →∞ where R is called the radius of convergence of the series. The series therefore converges when |x|1<11 and diverges when | x|1>11. but converges to ln12 when x1=1−1 (see the MacLaurin series for the logarithmic function). The geometric and exponential series are examples of power series. |cn2cn+1|1=1(n1+11)2n1→11 as n1→1∞ and the radius of n convergence is R1=11. it diverges when | x|1>1R. applying the ratio test. We will see in Section 7. a power series converges when lim cn +1 x n +1 cn x n n →∞ = | x | lim cn +1 cn n →∞ <1 or.become progressively smaller and alternate in sign then the series converges. the exponential series has R1=1∞ (see Examples 7. equivalently. 7. Thus. when it is the harmonic series. The series is therefore convergent when | x|1<1R. c2.9). By the ratio test. For example. .6 MacLaurin and Taylor series Power series A power series in the variable x has the form of an ‘inﬁnite polynomial’ c01+1c1x1+1c2 x 21+1c3 x 31+1where c0.1= are constants.208 Chapter 7 Sequences and series Alternating series If the terms of the series a11+1a21+1a31+1. EXAMPLES 7. and the case x1=1±R has to be tested by other methods. the alternating harmonic series 1− 1 1 1 + − + 2 3 4 converges. when | x | < lim cn =R cn +1 (7. The geometric series has radius of convergence R1=11.

6 MacLaurin and Taylor series 209 (ii) The coefﬁcient of x n in the series ∑ ∞ ( −3x ) n n2 n =1 is cn1=1(−3)n2n2. The series called after him appeared in his Treatise of ﬂuxions (1742). and was known to the Scottish mathematician James Gregory (1638–1675).19) The coefﬁcients c0. Similarly.8). f ′(x)1=1f(x). f(0)1=1c0. d n f ( x) f ( n ) (0) = = n! cn . The derivatives of the function are f ′(x) = f ″(x) = f ′′′(x) = Then. letting x1=10. f ′(0)1=1c1. for the nth derivative.1= can be obtained in the following way. . but the more general Taylor series was published in 1715.7. = df = c + 2 c2 x + 3 c3 x 2 + 4 c4 x 3 + dx 1 d2 f dx 2 d3 f dx 3 = 2 c2 + 6 c3 x + 12 c4 x 2 + = 6 c3 x 2 + 24 c4 x + and. Many other functions can be expanded in this way. The series therefore converges when | x|1<1123. dx n x =0 1 ( n) f (0) n! cn = 10 Colin MacLaurin (1698–1746). Let f(x) be a function of x that can be represented as a power series f(x)1=1c01+1c1x1+1c2 x 21+1c3 x 31+1c4 x 41+1(7.4 that the geometric series can be regarded as the expansion of the function 12(11−1x) in powers of x. It also converges when | x|1=1123 (see Example 7. f ′′′(0)1=13!c3. c1. c2. 0 Exercises 52–57 The MacLaurin series10 We saw in Section 7. in general. The Treatise contains also the method of deciding the maximum2minimum question by investigating the sign of a higher derivative. f ″(0)1=12!c2. professor of mathematics at Edinburgh. By the ratio test. |cn2cn+1|1=1(123)(n2n1+11)21→1(123) as n1→1∞ and the radius of convergence is R1=1123. the exponential series can be regarded as the expansion of that function f(x)1=1e x whose derivative is equal to itself.

. . . . . the expansion is valid only within the radius of convergence of the series. . . . . .. . . . ... . .... . . . . .... ... . . .. . ... The binomial series11 The binomial series is the MacLaurin expansion of the function (11+1x)a for arbitrary values of a.... . .. . . . .. .. . . The theorem was published by Wallis in his Algebra of 1685... . .. . . The MacLaurin series is an expansion of the function f(x) about the point x1=10 (Figure 7.. .. . .. secretary of the Royal Society..20) Figure 7. . . ... . .... . . .21) 11 The binomial theorem (series) was discovered by Newton in 1665 and described in 1676 in letters written to Henry Oldenburg. . . .. . . .. . .. . . . For this to be possible the function and its derivatives must exist at x1=10 and throughout the interval 0 to x... . . ... . . . . . . ... . . . . ..... . . .. . . ) ( . . . .... .... .. . .. . . . . . . . . .. x .. . . .. . . . .. . that is.... the factor (11+1x)a−n is replaced by 1. the value of the function at point x is expressed in terms of the values of the function and its derivatives at x1=10. . . . . .. .. .. ... . ... . . In addition. .. f ′(0)1=1a.19) then gives f ( x ) = f (0) + x x2 x3 f ′(0) + f ′′(0) + f ′′′(0) + 1! 2! 3! =∑ ∞ n= 0 x n ( n) f (0) n! (7. . . .. . . . . .. .. .. .. . .... . . .. . . . . . .. . .. .. f ″(0)1=1a(a1−11). . . . . 0 .. . .. . . . .. . . . .. . . f ′′′(0)1=1a(a1−11)(a1−12). We have f(x) =1(11+1x)a f ′(x)1 =1a(11+1a)a−1 f ″(x)1 =1a(a1−11)(11+1x)a−2 f ′′′(x)1 =1a(a1−11)(a1−12)(11+1x)a−3 f (n)(x)1=1a(a1−11)(a1−12)=(a1−1n1+11)(11+1x)a−n When x1=10..... .. . . . . .. .. for transmission to Leibniz... . ) ( . . . . .. . . . . .. .. . . = and the MacLaurin expansion is (1 + x ) a = 1 + a x + a( a − 1) 2 a( a − 1)( a − 2) 3 x + x + 2! 3! (7. . .. . . . . .. . . .. . . . .. .2 Examples of MacLaurin series 1.. . .. . .. . . .. . .210 Chapter 7 Sequences and series Substitution into (7. . . . . .. . .. . . . . . . . . . . . . . . ... .. f(0)1=11.. . . . . . . . .... . . .. . .. . . . . . . .. ... . ... .. . x • • x f 0 f This power series is called the MacLaurin series expansion of the function f(x).. . . .. .. . . . . . . . . . . . . . .. .. .2). .. . . . . . . . . . . .. . ... ... . . .. . . . . . . . . .. . .

12 (i) ( −1)( −2) ( −1)( −2)( −3) 1 = (1 − x )−1 = 1 + ( −1)( − x ) + ( − x )2 + ( − x )3 + 1− x 2! 3! =111+1x1+1x21+1x31+1 1 (ii) (1 + x ) 1 2 = 1 + ( x ) + 2 = 1+ ( )(− ) ( x) + ( )(− )(− ) ( x) 1 2 1 2 2 1 2 1 2 3 2 3 2! 3! + x x 2 x 3 5x 4 − + − + 2 8 16 128 .23) EXAMPLES 7. to arbitrary powers. When a is not a positive integer. the expansion of (x1+1y)a. the (n1+11)th derivative and all higher derivatives are zero. the series has radius of convergence R1=11.6 MacLaurin and Taylor series 211 with general term a( a − 1)( a − 2)…( a − n + 1) n x n! The series (7. Equation (7.21) is called the binomial series. and (7. when a is a positive integer n. Thus. applying the ratio test.7. It may also converge when x1=1±1. The more general form. and is the generalization of the binomial expansion. is obtained from (7.21) by factorizing out the term xa if | x|1>1| y| or y a if | y|1>1| x|.21) reduces to the binomial expansion.11).22) 2 3 y a( a − 1) y a( a − 1)( a − 2) y = x a 1 + a + + + 3! 2! x x x and the series converges since | y2x|1<11. In fact. cr r = lim =1 cr +1 r → ∞ a − r R = lim r →∞ and the binomial series converges for values −11<1x1<1±1. y ( x + y )a = x a 1 + x a (7. equation (7.22) can also be written as ( x + y ) a = x a + a x a −1 y + a( a − 1) a − 2 2 a( a − 1)( a − 2) a − 3 3 x y + x y + 2! 3! (7. Thus. if | x|1>1| y|.

the function ln(11+1x) is well behaved at x1=10: f(x)1 =1ln(11+1x) f ′(x) = f ″(x) = f ′′′(x) = f ′′′′(x) = 1 1+ x −1 (1 + x )2 2 (1 + x )3 −6 (1 + x )4 f(0)1 =1 0 f ′(0)1 =1 1 f ″(0)1 =1−1 f ′′′(0) =1 2 f ′′′′(0)1=1−3! - . However. For example. Then sin x = x − x 3 x5 x 7 + − + 3! 5! 7! Comparison with the exponential series shows that the sine series converges for all values of x. f ′(x)1=1cos1x.212 Chapter 7 Sequences and series (iii) (8 + x ) 12 x = 8 1 + 8 12 1 x = 8 1 + + 2 8 + 1 2 ( )(− ) x 1 2 1 2 2 2! 8 3 ( )(− )(− ) x 1 2 3 2 3! 8 + x x2 x3 5x 4 = 8 1 + − + − + 16 512 8192 524288 For example. f ″(x)1=1−sin1x.999998 2. f(x)1=1sin1x. for the sine function. using the ﬁrst ﬁve terms of the series. if x1=11. Trigonometric functions The trigonometric functions sin1x. f ″(0)1=10. 1 1 1 5 9 = 8 1 + − + − + 16 512 8192 524288 ≈1 8 ×11.060659411≈12. f(0)1=10. and can be expanded as MacLaurin series. f ′′′(0)1=1−1. f ′(0)1=11. f ′′′(x)1=1−cos1x. The logarithmic function The function ln1x cannot be expanded as a power series in x because the function and all its derivatives are discontinuous at x1=10. cos1x and tan1x have continuous derivatives at x1=10. 3.

. sinh x = x + all x 8. but had been known to James Gregory. the expansion of a function f(x) about the point x1=10. secretary of the Royal Society. and ln(1 + x ) = x − x 2 x3 x 4 + − + 2 3 4 The series converges when | x|1<11. e x = 1 + x + x 2 x3 x 4 + + + 2! 3! 4! x 3 x5 x 7 + + + 3! 5! 7! x 2 x 4 x6 + + + 2! 4! 6! all x 7.7. is a special case of the more general expansion of the function about the point x1=1a: 12 Brook Taylor (1685–1731). sin x = x − x 3 x5 x 7 + − + 3! 5! 7! x 2 x 4 x6 + − + 2! 4! 6! x 3 2 x5 17 x 7 + + + 3 15 315 x 2 x3 x 4 + − + 2 3 4 all x 3. the radius of convergence is R1=11. Thus. tan x = x + − 2 <x< 2 5. (1 + x ) a = 1 + a x + a( a − 1) 2 a( a − 1)( a − 2) 3 x + x + 2! 3! | x|1<11 2. In addition. the series is the (convergent) alternating harmonic series when x1=11: ln 2 = 1 − 4. cr 2cr+11=1(r1+11)2r1→11 as r1→1∞ and by the ratio test (7. cos x = 1 − all x π π 4.6 MacLaurin and Taylor series 213 The nth derivative at x1=10 is f (n)(0)1=1(−1)n−1(n1−11)!. 1 1 1 + − + 2 3 4 A list of some useful series 1. The series called after him appeared in his Methodus incrementorum (1715). cosh x = 1 + all x 0 Exercises 58–67 The Taylor series12 The MacLaurin series.18). ln(1 + x ) = x − −11<1x1≤1+1 6.

for example. f(x)1=1cos1x.24) ( x − a)n ( n) f ( a) =∑ n! n= 0 This power series in (x1−1a) is called a Taylor series expansion of the function f(x). = Therefore cos x = −( x − π 2) + 0 Exercises 68–71 1 1 1 (x − π 2)3 − (x − π 2)5 + (x − π 2)7 + 3! 5! 7! 7. f ′(x)1=14x3. − 1 < x ≤ +1 .214 Chapter 7 Sequences and series f ( x ) = f ( a) + ∞ ( x − a) ( x − a )2 ( x − a )3 f ′′′( a ) + f ′( a ) + f ′′( a ) + 3! 1! 2! (7. Consider.7 Approximate values and limits The MacLaurin and Taylor series provide a systematic tool for approximating functions in the form of polynomials. f ′(π22)1=1−1. f ′′(1) = . f ″(x)1=1−cos1x. f ′′′(1) = . The veriﬁcation and conditions of existence of the Taylor series are as for the MacLaurin series. f ″(π22)1=10. f ″(x)1=141×13x2. f (n)(x)1=10 if n1>14 f (1) = 1. f ′′′′(1)1= 4! 3! 2! 1! + 4 ( x − 1)3 3 + ( x − 1)4 (ii) Expand cos1x about x1=1π22. f ′′′′(x)1=14!. f ′′′(x)1=1sin1x. We have f(x)1=1x4. the logarithmic series ln(1 + x ) = x − x 2 x3 x 4 + − + 2 3 4 . x4 = 1 + 4 4 ( x − 1) + ( x − 1)2 1 2 4! 4! 4! . f ′(1) = Therefore. f ′′′(π22)1=11.13 Taylor series (i) Expand x4 about the point x1=11. = f(π22)1=10. EXAMPLE 7. f ′(x)1=1−sin1x. f ′′′(x)1=14!x.

The term Rn(x) is called the remainder term and is the error involved in approximating the function by a polynomial of degree n.2 u11=1x 0 0.7 Approximate values and limits 215 If the series is terminated after a ﬁnite number of terms.1826 66 0.0009 995 0.1 and that the series converges rapidly for these small values of x.0009 9950 0333 0. 2 3 Some values of these are shown in Table 7. The inﬁnite series is obtained in the limit n1→1∞ if Rn(x)1→10 as n1→1∞. f (n)(x) in the interval a to x. Then f ( x ) = f ( a) + ( x − a) ( x − a )2 f ′( a ) + 1f ′′(a)1+11! 2! (7.01 0.69 The table shows that u11=1x is a good approximation to ln(11+1x) when x1≤10. The series therefore provides a sequence of such approximations: u1 = x .0009 9950 0333 0.0000 9999 5000 0.2 1.1 0.0099 5033 33 0. The smallest and largest values of Rn(x) are lower and upper bounds to the error.0953 333 0. f ″(x).0953 310 0. Table 7.0 Values of ln(1 + x) u21=1x1−1x 222 0 0.26) and a1<1b1<1x is some point in the interval.83 ln(11+1x)1=1 lim (un) n→ ∞ 0 0.7. 2 u3 = x − x 2 x3 + . and eight terms are needed to give 10% accuracy when x1=11. and let f (n+1)(x) exist within the interval.095 0. u2 = x − x2 . The theoretical basis for this use of the series is Taylor’s theorem.25) ( x − a)n ( n) f ( a ) + Rn ( x ) + n! where Rn ( x ) = ( x − a ) n +1 ( n +1) f ( b) ( n + 1)! (7.0000 9999 5 0.2. =. the result is a polynomial approximation to the function.1823 21 0. Taylor’s theorem Let f(x) be a continuous single-valued function of x with continuous derivatives f ′(x).0099 5 0.0000 9999 5000 0.001 0.18 0. .0001 0. Convergence is less good for larger values of x.0099 5033 08 0.5 u31=1x1−1x 2221+1x 323 0 0. each term providing at least one additional ﬁgure of accuracy.

21<11.000067e0.221415 . e0. For the upper bound.21<11.0000671<1R31<10.2 The lower bound is 1. ex = 1 + x + where Rn = x n +1 b e ( n + 1)! x2 + 2! + xn + Rn ( x ) n! for some point 01<1b1<1x. When x1>10.2213331+10.2 ≈ 1 + 0.2)3 + = 1.21=10.2 + (0.2 e0.221333 Therefore e0.2 and n1=13 (and rounding all numbers to six decimal places).21<11.999933e0.9999331=11.221400. the smallest and largest values of Rn are given by x n +1 x n +1 x < Rn < e ( n + 1)! ( n + 1)! Then 1+ x + x2 + 2! + xn x n +1 x2 + < ex < 1 + x + + 2! n! ( n + 1)! + x n +1 x xn + e n! ( n + 1)! For example.0000671<1e0.21−10.216 Chapter 7 Sequences and series EXAMPLE 7.000067e0.14 The exponential series By Taylor’s theorem (for a1=10).221333 2 6 and the error bounds are given by 0.22133320. the cubic approximation has value e0.2)2 (0.2213331+10.000067e0.21<11. when x1=10.2213331+10.000067e0.2 so that 1.

. .... .. ... . .... ....... .. . . . .. . . . . . ... . ... . . . . ... . ........ .... .. . . 1. . .. . . . .... .. .. .28) More generally.2214001<1e0.. .. ... . . . .. .. . .. ... . sin x = x − cos x = 1 − x 3 x5 + − 3! 5! x2 x4 + − 2! 4! ≈x ≈ 1− x2 2 Another way of expressing the same results is in terms of limits. .. . .. . Figure 7. . . ... . . ... . . . . It is not possible to substitute x1=1a in this case because the result would be the indeterminate 020..... . . . .. . . lim f ( x) g( x) x →a when f(x)1→10 and g(x)1→10 as x1→1a... .. . .. . . we have x =y x nis Similarly. ...24)... the use of power series provides a systematic way of determining the limit of the quotent of two functions. . .. . ... . .. . . 0 Exercise 72 Limits The MacLaurin series shows how a function behaves when the variable is very small.. . . . . . .. . .. .... . . .. . .. . . . ..3 x . ..... . . . .221403. ... . .. . . ... ... . ... . . .. .. .. . . .... .. if both functions are expanded as Taylor series. .. . .. . . . . ... .. . . Thus ln(11+1x)1≈1x when x is small enough. . ..... . y ... . .. .. . .. . Similarly. ..27) . .. . . Thus (Figure 7. . . . . . . π . .. .... . .. .. . . ...... .. . .. .. . . .... . . . ..221415 The exact value is 1... ..... However.. . . . . ... . . .. . . . .. . .. . .. . 1 − cos x 1 lim = x →0 x2 2 (7.. 0 . . .. .. . .. ..... .. .. . ..3) sin x x2 x4 = 1− + − x 3! 5! so that in the limit x1→10. .21<11. . .... . . . .. . . .....7 Approximate values and limits 217 and 1. so that in the immediate vicinity of x1=10 the function y1=1ln(11+1x) can be approximated by the straight line y1=1x.. . . . . .. .. . . . . . .. . ....... ... . . . . . ... .. . . .... when x is small enough. .. . .... . .. .. .. π− . .. ..... .. ..7.. .. by equation (7. ... . . . .. sin x lim =1 x →0 x . .. .. (7. .. .. .. . ... .... .

he was tutored by Johann Bernoulli in the new calculus. lim f ( x ) f ′( a ) f ′( x ) = = lim g ( x ) g ′( a ) x → a g ′( x ) (7.30) x →a and so on.29) x →a This method of ﬁnding limits is called l’Hôpital’s rule. if g′(a) is not zero. x →0 x3 sin x − x x 3 = = ( x − x 3 23! + x5 25! − x x 3 3 )− x 1 x2 + − 3! 5! 1 as x → 0 6 − x 3 23! + x5 25! − =− →− (ii) x 2ex . lim x x → a ( e − 1)2 13 Guillaume François Antoine de l’Hôpital (1661–1704).15 Find the limits. (i) sin x − x lim . EXAMPLE 7. ( x − a) f ′′( a ) + 2! ( x − a) g ′( a ) + g ′′( a ) + 2! f ′( a ) + f ( x) = g( x) and. If f ′(a) and g′(a) are both zero the process is repeated to give lim f ( x ) f ′′( a ) = g ( x ) g ′′( a ) (7. .13 If g′(a)1=10 but f ′(a)1≠10 the limit is inﬁnite. Analyse des inﬁniment petits (1696). French nobleman and amateur mathematician.218 Chapter 7 Sequences and series ( x − a )2 f ′′( a ) + f ( x) 2! = g( x) ( x − a )2 g ( a ) + ( x − a ) g ′( a ) + g ′′( a ) + 2! f ( a ) + ( x − a ) f ′( a ) + Then because f (a)1=1g(a)1=10. The rule ascribed to him was contained in a letter from Bernoulli in 1694 and appeared in l’Hôpital’s inﬂuential textbook on the calculus.

The power series may be added or subtracted term by term to give a power series C ( x ) = A( x ) ± B( x ) = ∑ ( ar ± br ) x r r =1 ∞ (7. (ii) Multiplication. In Einstein’s theory of the heat capacity of simple atomic solids.8 Operations with power series 219 The numerator is x 2(11+1x1+1x 2221+1-). (i) Addition and subtraction.7. each atom in the solid is assumed to vibrate with the same frequency ν. where h is Planck’s constant. x 2 ex ( e − 1) x 2 = x 2 (1 + x + x (1 + x + 2 ) ) → 1 as x → 0 This example is important in the statistical mechanics of solids. The product of A(x) and B(x) is the double inﬁnite sum C ( x ) = A( x ) B( x ) = ∑ ∑ ar bs x r + s r =0 s=0 ∞ ∞ . The denominator is [(11+1x1+1x 2221+1-)1−11]21=1(x1+1x 2221+1-)21=1x 21+1x 31+1Therefore. The molar heat capacity of the solid is then Cν = 3R x 2 ex ( e x − 1)2 with x1=1hν2kT. Then Cν1→13R. respectively. k is Boltzmann’s constant. The taking of the limit x1→10 corresponds to letting T1→1∞.31) whose radius of convergence is at least as large as the smaller of RA and RB. R is the gas constant and T is the temperature. 0 Exercises 73–77 7.8 Let Operations with power series A( x ) = ∑ ar x r r =0 ∞ and B( x ) = ∑ br x r r =0 ∞ be two power series with radii of convergence RA and RB.

220 Chapter 7 Sequences and series and this can be rearranged as the Cauchy product C ( x ) = a0 b0 + ( a0 b1 + a1b0 ) x + ( a0 b2 + a1b1 + a2 b0 ) x 2 + r = ∑ cr x r r =0 ∞ (7.16 Multiplication of power series (3x )2 e3x × e−2 x = 1 + 3x + + 2! 2 1 + ( −2 x ) + ( −2 x ) + 2! ( −2)2 32 = 1 + 1 × ( −2) + 3 × 1 x + 1 × + 3 × ( −2) + × 1 x 2 + 2! 2! = 1+ x + 0 Exercise 78 (iii) Differentiation and integration. .33) Z A( x ) dx = ∑ ar Z x r dx = ∑ r =0 ∞ ar r +1 x +c r =0 r + 1 a1 2 x2 + a2 3 x3 + ∞ (7. termwise integration of the series A(x) gives (7.34) = c + a0 x + The radius of convergence of both the derived series and integrated series is R1=1RA. EXAMPLE 7. x2 + 2! = ex A power series may be differentiated or integrated term by term to give a power series whose radius of convergence is the same as that of the original series. Thus. term by term differentiation of A( x ) = ∑ ar x r = a0 + a1 x + a2 x 2 + a3 x 3 + r =0 ∞ gives dA ∞ = ∑ ra x r −1 = a1 + 2 a2 x + 3a3 x 2 + dx r =1 r Similarly.32) where cr = ∑ ai br − i i=0 The radius of convergence of the product is equal to the smaller of RA and RB.

… wn . u11=13 (see Exercise 8). u01=1u1 9. u11=13 10. un+21=13un+11−12un. 7. 2 16. u01=11. 11+151+191+1131+119.9 Exercises Section 7. v n = . x 31+1x 51+1x 71+1- . ux = 1 . u01=11. 1. w1 = 1 n 8. − . . = 2.17 Differentiation and integration of power series d 1 d 2 3 4 = (1 + x + x + x + x + dx 1 − x dx = 1 + 2 x + 3x 2 + 4 x 3 + = ) 1 (1 − x )2 ) dx = − ln (1 − x ) + c Z 1 2 3 4 dx = Z(1 + x + x + x + x + 1− x = c+ x+ x 2 x 3 x 4 x5 + + + + 2 3 4 5 0 Exercises 79. 2. Section 7.9 Exercises 221 EXAMPLE 7. Find the limit of the sequence {un+12un} for un+21=1un+11+12un. 11+131+191+1271+11 1 1 21.… 5 25 125 n Find the ﬁrst 6 terms of the sequences: 1 4. 31−121−171−1121−120. un+21=1un+11+12un.2 Find (a) the general term and (b) the recurrence relation for the sequences: 1. 80 7. 10. … 2 5. 3. un+21=13un+11−12un. 2 6. 1. 14. ur +1 = ur + . u01=11. x1+12x 21+14x 31+122. − . (ii) the ﬁrst 10 terms: 18. 9.3 Find the sum of (i) the ﬁrst n terms. 1 + + + + 3 9 27 Find the sum of the ﬁrst n terms: 23. x ( x + 2) u1 = 0 x = 1.7. 2r 13. 3. 27. w n+1 = n = 0. 1. u11=1122 Find the limit r1→1∞ for: r 1 r 3r 2 + 3r + 1 1 11. 1. r 12. 15. r+2 r+2 r + r +1 5r 2 − 6r − 1 3 17. 2. = 1 1 1 3. 4. 3 7.

222 Chapter 7 Sequences and series Use equation (7.13) or (7. r1=10. (11−14x)5 32. and ﬁnd its sum.11) to expand in powers of x: 25. Conﬁrm that the series is a convergent geometric series. (i) Use the geometric series to express the number 12(1061−11) as a decimal fraction. 1. then (ii) ﬁnd the sum of the series − 2 r r + 2 ∑ r ( r + 2) .4). (11+1x)5 n Calculate the binomial coefﬁcients . 36. then (ii) show that ∑ r 2 = 6 n( n + 1)(2n + 1) r =1 n 1 40. 3! 35. r =1 n Section 7. (i) Verify that 1 = r ( r + 2) 11 1 . 1 2+ x 44.14) to expand in powers of x: 29. n=1 10 1 37. . 45. then (ii) show that r ( r + 1)( r + 2) 1 1 1 ∑ r (r + 1)(r + 2) = 4 − 2( n + 1)( n + 2) r =1 39. n. (i) Express n 1 in partial fractions. (i) Verify that (11+1r)31−1r31=13r21+13r1+11. (i) Calculate the distinct coefﬁcients . (ii) Use the coefﬁcients to n1 !n2 !n3 ! 4 expand (a1+1b1+1c) . (ii) Find the values of x for which the series converge: 41.4 (i) Expand in powers of x to terms in x6. (31+1x)6 Use equation (7. r =1 n 1 38. (11+1x)7 24. (i) Expand (11+1r)61−1r6. n1=14 28. (ii) Use the coefﬁcients to expand n1 !n2 !n3 !n4 ! (a1+1b1+1c1+1d)3. (ii) Show that the decimal representation of 127 can be written as 1428572(1061−11) (see Section 1. 1 1 + 5x 2 43. n1=13 27. for r 26. then (ii) use the series in Table 7. Find ∑ n( n + 1) .1 to ﬁnd the sum of the series ∑ r5. (11−1x)3 30. (31−12x)4 4! 34. n1=17 33. (11+13x)4 31. The vibrational partition function of a harmonic oscillator is given by the series qv = ∑ e n= 0 ∞ − nθ v /T where θv1=1hνe 2k is the vibrational temperature. =. (i) Calculate the distinct trinomial coefﬁcients . 1 1 − 3x 42.

p + 2 (V − nb) = nRT V (ii) the Dieterici equation. π22 71. 1 3+ x ex − 1 x 2 66. 1 1+ x2 ln (1 − 2 x ) + 2 x x2 60. 65. ∑ ∞ r =1 ∞ ln r r3 1 r =1 48. 2 70.5 Examine the following series for convergence by Comparison test (use ln1n1<1n): D’Alembert ratio test: 46. ∑ (s + 1)! s=0 ∞ 49.1 x 69. ex.7. 67.6 Find the radius of convergence of each of the following series: 52. 63. sin1x. ∑ ra r =1 1 51. m=1 ∑ mm x m 57. 1 . (11+1x)123 62. p(V − nb) = nRTe− an RTV (i) Expand each of the following functions as a Taylor series about the given point. ∑ ra n= 2 Cauchy integral test: 50. Express T as a power series in v and show that the series reduces to the nonrelativistic kinetic energy in the limit v2c1→10. ∑ nx n n=1 ∞ 55. ∑ ∞ ( −1) n x 2 n 3n Write down the ﬁrst 5 terms of the MacLaurin series of the following functions: 58. Find the ﬁrst three coefﬁcients for n2 a (i) the van der Waals equation. ∑ n2 n=1 ∞ xn 56. ∞ ∞ n= 2 ∞ ∑ ln n sa 1 47. 2 . e−3x 61. (11−1x)−122 64. and (ii) ﬁnd the values of x for which the series converges: 68. ∑ n ln n 1 Section 7. sin12x 2 59. A body with rest mass m0 and speed v has relativistic energy E = mc 2 = m0 c 2 1 − v 2 / c2 and kinetic energy T1=1E1−1m0c2.9 Exercises 223 Section 7. ∞ m= 0 ∞ ∑ 4m xm 53. The equation of state of a gas can be expressed in terms of the series ∞ n pV = nRT ∑ Bi (T ) V i=0 i where the Bi are called virial coefﬁcients. ln1x. ∑ (−1)r x 2r r =0 n= 0 ∞ 54.

Show that the formula reduces to the classical Rayleigh–Jeans law ρ1=18πkT2λ4 (i) for long wavelengths (λ1→1∞). . Find the Cauchy product of the power series expansions of sin1x and cos1x. (ii) if Planck’s constant is set to zero (h1→10). (ii) Use this expansion to ﬁnd an approximate value of 3 9 . lim ln x x →1 x2 − 1 77. The energy density of black-body radiation at temperature T is given by the Planck formula ρ( λ ) = 8πhc λ5 [e hc λ kT − 1]−1 where λ is the wavelength. Find the limits: 73. where C is a constant. and show 1 that it is equal to 2 sin 2 x.224 Chapter 7 Sequences and series Section 7. lim tan x − sin x x 3 x→0 75. Differentiate the power series expansion of sin1x and show that the result is cos1x. Section 7. (iii) Use this value and Taylor’s theorem for the remainder to compute upper and lower bounds to the value of 3 9 . 80. Integrate the power series expansion of sin1x and show that the result is C1−1cos1x. lim e x + e− x − 2 x → 0 cos x − 1 76.7 72. 79. lim ex − 1 x→0 x 74.8 78. (i) Find the MacLaurin expansion of the function (81+1x)123 up to terms in x4.

Descartes used the words ‘real’ and ‘imaginary’.2) where x and y are real numbers. −1. The ﬁrst serious consideration of complex numbers was by Albert Girard (1595–1632) who in his L’invention nouvelle en l’algèbre (New discovery in algebra). Powers of i Every integer power of i is one of the numbers i.8 Complex numbers 8. =.7 and in Chapter 2 that the solutions of algebraic equations are not always real numbers.1) A number containing i = −1 is called a complex number. and the square root of a negative number is not a real number.1 Such numbers are incorporated into the system of numbers by deﬁning the square root of −1 as a new number which is usually denoted by the symbol i (or j in engineering mathematics) with the property i 21=1−1 (8. i 4n+21=1−1.4) Square roots of negative numbers are mentioned in Cardano’s Ars Magna of 1545 in connection with the solution of quadratic and cubic equations. The number x is called the real part of z. If y1=10 then z1=1x is real.1 Concepts We saw in Section 1. 1 i 4n+11=1+i. 1629. 1. The general complex number has the form (the letter z is usually used to denote a complex number) z1=1x1+1iy (8. Cardano called such a result ‘as subtle as it is useless’. Rafael Bombelli (1526–1572). and 21+15i. −3i. and this was adopted by Gauss in his Disquisitiones arithmeticae of 1801. for integers n1=10. i 4 = ( i 2 )2 = 1. −i. . Euler proposed the symbol i for −1 in 1777. ±1. 1 i i i −1 = = 2 = = −i i i −1 In general. called ± −1 ‘più di meno’ (plus of minus) and ‘meno di meno’ (minus of minus). Italian engineer. for example i 3 = i 2 × i = − i.3) If x1=10 then z1=1iy is called pure imaginary. He called the complex solutions of equations ‘impossible’ but ‘good for three things: for the certainty of the general rule (the theorem). and y is called the imaginary part of z: x1=1Re (z). so that the set of complex numbers includes the real numbers as subset. and for their use’. i 4n1=1+1. and presented the arithmetic of complex numbers in his Algebra of 1572. for the fact that there are no other solutions. Leibniz factorized x 4 + a 4 = ( x + a i )( x − a i )( x + a − i )( x − a − i ). in particular. ±2. examples of complex numbers are 2i. y1=1Im (z) (8. published the ﬁrst statement of the fundamental theorem of algebra. the solutions of the equation x 21=1−1 are x = ± −1 . i 4n+31=1−i (8.

(8. EXAMPLES 8. the basic equations of quantum mechanics necessarily involve i = −1 . It is only necessary to remember to replace i 2 by −1 whenever it occurs. subtracted.2 Algebra of complex numbers Complex numbers can be added. the solutions of the differential equations of motion in both classical and quantum mechanics often involve complex numbers.6) z11+1z21=1(x11+1x2)1+1i( y11+1y2) (8. Let two complex numbers be z11=1x11+1iy1. 8.5) Two complex numbers are equal if their real parts are equal and if their imaginary parts are equal: z11=1z2 Addition if x11=1x2 and y11=1y2 (8.8) .1 Addition and subtraction (i) (31+12i)1+1(41−13i)1=1(31+14)1+1(21−13)i1=171−1i (ii) (31+12i)1−1(31−12i)1=1(31−13)1+1(21+12)i1=14i (iii) (31+12i)1+1(41−12i)1=17 0 Exercises 1–3 Multiplication z1z21=1(x11+1iy1)(x21+1iy2) =1x1(x21+1iy2)1+1iy1(x21+1iy2)1=1(x1x21+1ix1 y2)1+1(iy1x21+1i 2y1 y2) =1(x1x21−1y1 y2)1+1i(x1 y21+1y1x2) using i 21=1−1. multiplied. for example. Complex numbers also occur in the formulation of physical theory.226 Chapter 8 Complex numbers Apart from their role in extending the concept of number. and divided in much the same way as ordinary real numbers. Equality z21=1x21+1iy2 (8. the imaginary parts are added to give the imaginary part of the sum.7) The real parts of z1 and z2 are added to give the real part of the sum. complex numbers occur in several branches of mathematics that are important in the physical sciences.

10) (ii) (8. 2 1 ( z − z*) = − i. The conjugate pair of complex numbers z and z* has the following properties: (i) 1 1 ( z + z*) = ( x + iy ) + ( x − iy ) = x = Re ( z ) 2 2 1 1 ( z − z*) = ( x + iy ) − ( x − iy ) = iy = i Im ( z ) 2 2 (real and positive) (8. z= 1 3 ± 9 − 16 1 = 3 ± −7 = 3 ± i 7 2 2 2 .11) (8. 2 z z* = 1 + 1 = 2 (iii) Solve the quadratic equation z21−13z1+141=10 (see Example 2.2 Algebra of complex numbers 227 EXAMPLES 8.12) (iii) zz*1=1(x1+1iy)(x1−1iy)1=1x 21+1y 2 EXAMPLES 8.8.3 Conjugate pairs of complex numbers (i) If z1=121+13i then z*1=121−13i and 1 ( z + z*) = 2.2 Multiplication (i) (21+13i)(31+14i)1=12(31+14i)1+13i(31+14i)1=161+18i1+19i1+112i 21=1−61+117i (ii) i(21−13i)1=12i1−13i 21=131+12i 0 Exercises 4–7 The complex conjugate If z1=1x1+1iy is an arbitrary complex number then the number obtained from it by replacing i by −i is z*1=1x1−1iy (8.9) and is called the complex conjugate of z (sometimes G is used instead of z*). 2 z z* = 22 + 32 = 13 (ii) If z1=111−1i then z*1=111+1i and 1 ( z + z*) = 1. 2 1 ( z − z*) = 3i.18). z is then also the complex conjugate of z*.

The simpler method is to make use of property (8.1.228 Chapter 8 Complex numbers and the roots of the quadratic are the complex conjugate pair z= 1 3 + i 7 . discussed the graphical representation of complex numbers in his On the analytical representation of direction of 1797. Caspar Wessel (1745–1818). Swiss bookkeeper. by which time he believed mathematicians were comfortable enough with complex numbers to accept it. .2 The plane is called the complex plane. multiplying top and bottom by z *1=1x21−1iy2. Gauss used the same interpretation of complex numbers in his fourth and ﬁnal proof of the fundamental theorem of algebra in 1848. that is. EXAMPLES 8. x21≠10 and y21≠10.13) 0 Exercises 12–15 8. 2 0 Exercises 8–11 Division z1 ÷ z2 = z1 ( x1 + iy1 ) = z2 ( x2 + iy2 ) The division can be performed by the rules of ordinary long division.12) of conjugate pairs to transform the denominator into a real number. Norwegian surveyor. Real 2 John Wallis (1616–1703) ﬁrst suggested that pure imaginary numbers might be represented on a line perpendicular to the axis of real numbers.1y). The complex plane is also called the Gaussian plane. Thus. we have * ( x + iy )( x − iy ) ( x + iy )( x − iy ) z1 z1 z2 1 2 2 1 2 2 = = 1 = 1 2 2 z2 z z* ( x2 + iy2 )( x2 − iy2 ) x2 + y2 2 2 x x + y y y x −x y 1 2 1 2 = 1 2 + i 1 2 x2 + y2 x2 + y2 2 2 2 2 The division is deﬁned only if z21≠10. and Jean Robert Argand (1768–1822). in his Essai of 1806.3 Graphical representation The complex number z1=1x1+1iy is represented graphically by a point in a plane.4 Division (i) (ii) i 2 + 3i ( 2 + 3i )(3 − 4i ) 18 + i 18 = 2 = = + 2 3 + 4i (3 + 4i )(3 − 4i ) 3 + 4 25 25 1 + i (1 + i )(1 + i ) 0 + 2i = = =i 1 − i (1 − i )(1 + i ) 1 + 1 (8. with coordinates (x. as in Figure 8. 2 z* = 1 3 − i 7 . 2 the complex conjugate of the denominator.

. .. . The representation is called the Argand diagram. .. .... . .. . . .. . . . .. .. . . . . .. . . . .. . . . . . . . . . . ... .. . . . . . . ... . . ... (8..... .. . thus.. . .. . . . . . . .. .. . .. .. ..... . .. . . . . . . . . . . .. . . . . . .... with y1=10. . A unique value can be computed from x and y by the prescription given in Section 3. .. .. .. . . . with x1=10. . . .. . ..17) if x1<10 in which the inverse tangent has its principal value. . .. . . .. . ... . y1=1r1sin1θ and the complex number can be written in the polar form z1=1r (cos1θ1+1i1sin1θ) The angle θ is called the argument or angle of z. .. . . . . . . .. .. .. . . . . .... . . . . .. . . . .. . . .. . . When | z |1=11 the point lies on the unit circle.. . . .. .. . ... . . .. . . . . . . . .. . . . . .... . is called the modulus or absolute value of z and is written r1=1mod1z1=1| z | (8. . .... as discussed in Chapter 3. . . . ... . )si x a l a er( x yi +x = z . . .. .. The polar representation The position of the point z1=1x1+1iy in the plane can be speciﬁed in terms of the polar coordinates r and θ.. . . . the trigonometric functions in (8.. .. . . . . . . ... . . .. .15) θ1=1arg1z The angle can have any real value but..1. . . . ..... . .. . . .. .. .. ... . ... .. ... .. Then x1=1r1cos1θ.... . . . . .. .. . . .. ... . . ..5 for the transformation from cartesian to polar coordinates.. . . .. .. . . . ... .. ..14) Complex numbers with the same modulus r1=1| z | lie on the circle of radius r in the plane. . . . .. . . .. as in Figure 8.... .. . ..15) are periodic functions of θ so that the number z is unchanged when a multiple of 2π is added to θ. . . . . . . . ... . . .3 Graphical representation 229 numbers. . lie on the y or imaginary axis.. .1 The distance of point z from the origin. . r = x 2 + y 2. . . . . . . . ..16) . . . .. . . . . ... . . . . • )si x a r ani g ami( y y x r θ o (8. .. . .. ... . .. .. . . . ..... .8.... . Figure 8. .. given that tan1θ1=1y2x. . . . . . . .. . . .. . ... . y arg z = tan −1 x y = tan + π x −1 if x1>10 (8.. . .. . y . . . ... . . .. .. are represented by points on the x or real axis and pure imaginary numbers.. . . . . . . . .. . ..

. ..... .. . .. . . because x1<10. . .. . .. . . tan −1 = tan −1 (1) x The principal value of tan−1(1) is π24. . . . . . . Therefore ... . . . .. Figure 8. the numbers z11=1x11+1iy1 and z21=1x21+1iy2 are represented by points P and Q.. .. . . . ... .... . . .. . ..5 Express z1=1x1+1iy in polar form... .. . ... . . . . .. .. ... .. ........ . . . .. .... .. . . ... . . . . . . ... . ... . . . . .. . . . .... . .. . ... . . . . •.. . . . . . . .. . .. . . ... ... .. . . . . .. . .. .. . .. .. . . . ....... . .. .. . .. . .... . . ... .. . .. . ... ... ... . .. . ... . ... .. . . ... . ... . .. . . .. .. . . . . . . .. . . . . . . .... . . . . .. y We have x1=1−122 and y = − 3 2 so that r o (ii) z=− Figure 8.. ... . . . . . ... . . . ... .. .. . . ... . . . . . . . .. . . . .... . . . .230 Chapter 8 Complex numbers EXAMPLES 8. . ... . .. respectively. . .. . .. . ... . . . . ... .. . .. . .. ... . . it follows that z = cos 0 Exercises 16–22 Representation of arithmetic operations Addition and subtraction In Figure 8. ... . .. . .. ...2 x 1 3 − i 2 2 r π π z = 2 cos + i sin 4 4 i + 1 = z. .. .. . .. ... . .. ... . . .. .. ... . . . ... . .. .... ..... . ... . . . .. . . .. . ... . .. . . . .. . . . .. .. . ..... . . . . . Therefore 4π 4π + i sin 3 3 ( 3) is π23 and. . . . . . . . . . . ... ... θ y . ... . . . . . . . . . ... . . .. ...... ... . . .. .. . .4.. . . .. . ....3 The principal value of tan −1 arg1z1=1tan −1 ( 3) + π = 4π 3. (i) z1=111+1i We have x1=11 and y1=11 so that y r = | z | = x 2 + y 2 = 2 .. and arg1z1=1π24 because the point lies in the ﬁrst quadrant. .. .. . . . . .. . ... The representation of the sum z11+1z21=1(x11+1x2)1+1i( y11+1y2) • 2 3 2 − 1− ... . y tan −1 = tan −1 ( 3 ) x x .. .. . o θ 2 1 3 r = + =1 2 2 √ 2 . . .. ..... . .....

. . . . . . . . ... . ... .5 2θ 1θ 2z o • 2r 2r 1r 2z 1 z • y Figure 8... . .. . . . .. . ... .. . . .. . . . . . . . . . . . . . . . ... . .. . .... .. . ... . .. ... .... .. .. . .. .. .. . .. . . . .. .. . . ... ..18) 231 . . . . . ... . .. . . . ..... . .. . .... . .... . . . . . ... .. . ....... . .. ......... . . .. . . .4 arg1(z1z2)1=1arg1(z1)1+1arg1(z2) z21=1r2(cos1θ21+1i1sin1θ2) 8. . If Multiplication and division is obtained by completing the parallelogram OPSQ. . . Similarly for subtraction. .. .... . . .... . . . ... . ... . ... . .. . . . . ... . ... .. .. .... . ... . ..... .. .... .. . . .. ... . .. . .. .. .. . . . ... . . . . .. .... ...... . . ... . . ... . . . . . . .. . ... ... ... . . . . .. . . . ....... . .... . .. . .3 Graphical representation (8... .. .. . .. . . .. . . . . .. . . . . .. ..... . . .. . .. . and has argument equal to the sum of the arguments: z1z21=1r1r2(cos1θ11+1i1sin1θ1)(cos1θ21+1i1sin1θ2) =1r1r2[cos(θ11+1θ2)1+1i sin(θ11+1θ2)] =1r1r2[(cos1θ11cos1θ21−1sin1θ11sin1θ2)1+1i(cos1θ11sin1θ21+1sin1θ11cos1θ2)] (8. . . .... ) (Q . . . ... z11=1r1(cos1θ11+1i1sin1θ1).. . .... ... .. .. . .... . . .. . . .. .. .... .. . . .. . .. . ... . ) .. . . .. .. .. . . . 2x 1x • 1z 2y 2z • ) 2 z + 1 z(S .. . .. ... . . ..... ..... .... .... .. . . .. . .. . . .... . . .. .. .. . . . . . .. . .. ... . .. . . . . O . .. ...... . . . . . .. .. . . ... .. . . ... . .. .. . . . ... . .. . . . . .. . . . . . . . ..... . It follows that the product of two complex numbers has modulus equal to the product of the moduli of the numbers. .. . . .... . . . . ... y11+1y2) and therefore represents the sum. ... .. . . . . .. . . .. .. . .. .. . .. . . . . . . . .... .. ......... .. . . . . .. .... ...... . . ... .. .. .. . .. . ... ... .. .. . . . . .. . . . . ... ... .. . . . . .. . . . ...... ... .. . . . .. . . the difference z11−1z2 is the sum of z1 and −z2.. . .. ... . .. ... . . .. ... .... .. . .. . . ... .. .. . . . . ... . .. . . .... . . .. .. . ... .. .. . . .. .. . . . ... . . .. . . . . . .. . .19) | z1z2 |1=1| z1 |1×1| z2 |. . .... .... . .... . . .. ..... ... .... ..... ... .. . . .. . ... . .. ... .. . .. . . .. .. ..... ... . ... . . ... . . .. . . . . .. . . ... . . .. . . . ..... . . .. . . .... .. . .. .. . ... . . ... . .. .. .. . . . . .. . .. .. . . .... . . . ... . .... . . . . . . . .. . ..... .. .. . . . . ... . . . .. .. ... . .. . . . . ..x . . .. . . . . . . . . . . .... . . . . ........ . .. .. .. . . ... . . .. ... .. .. . . ..... . .. . .. . . . . . . . . . . .. ... . ... .... . . .... . .. .. . . . .... .. . . .. . . . x . . . . .. .. . .. . .. . . . . . ..... . . ... ... . .. ... . .. . .. .. ... .. . . .. .. . . .. . .... . . . ..... . ... . ...... .. .. . .. . . . . ... . . . . .. . ... .... . . .. . . . . .... . . . . . ... . .. .. . . . .. ..... . . . . . . . ....... . . . . . . .. . ... .. .. . . . . . . ...20) (8. ... . .. ... .. . (P .. . . . .. Point S has coordinates (x11+1x2. . .. . .. . . . . . .. . ..... .. . . . 2θ + 1θ 1r 1z • .. . . • 1y y then The operations of multiplication and division are the more easily described when the numbers are expressed in polar form.. . . .... .. . .. ... . . . . . . . . .... .. .. . .. . . . . . . . . . . ..... ... .. . . .. .... . . . . . .. . ... . ... . . . . . ...... . . . . .. .. ... ..... . . . . . .. . ... . .. .. .. .. . ... . . . . . .. .. . . .. . . .. . .. . .. . . . .. . .. .. . . . .. . . .. . ... ... . . ... . . . .. .. .. .. . . .. Figure 8... .. . . . . . . ...... . . ... . . . . ... ... . .... . . .. .. .. .. ....... . . . .

22) that 1 = z 1 cos( −θ ) + i sin ( −θ ) r (8. 1 1 = (cos θ − i sin θ ) z r (8. In the case of division. z1=1x1+1iy and z*1=1x1−1iy. and θ21=14π23. θ11+1θ21=119π212 and.22) z arg 1 = arg ( z1 ) − arg ( z2 ) z 2 arg1zz*1=10 arg1z1=1−arg1z* (8. it follows from (8. because cos (−θ )1=1cos1θ and sin (−θ )1=1−sin1θ. z z1 = 1 . z12z2 and z22z1 as a single complex number for π z1 = 2 cos + i sin 4 (see Examples 8. 19 π 19 π z1 z2 = r1r2 cos(θ1 + θ 2 ) + i sin (θ1 + θ 2 ) = 2 cos + i sin 12 12 . then | z | = | z *| = x 2 + y 2 .6 Express each of z1z2. and zz*1=1| z |21=1x 21+1y 2. by equation (8. 4 π z2 = cos 4π 4π + i sin 3 3 . z2 z2 and z1 r1 = cos(θ1 − θ 2 ) + i sin (θ1 − θ 2 ) z2 r2 For the inverse of a complex number.232 Chapter 8 Complex numbers If the two numbers are a complex conjugate pair. Therefore (i) r1r2 = 2 .21) and. r21=11.24) EXAMPLES 8.23) (8.5) We have r1 = 2 . θ11=1π24.19).

24 de Moivre’s formula It follows from the relations (8. by equation (8. z1=1cos1θ1+1i1sin1θ and (cos1θ1+1i1sin1θ)n1=1cos1nθ1+1i1sin1nθ (8. in general. for the product of three numbers.22).27) . z n1=1r n(cos1nθ1+1i1sin1nθ) (8.8. For example.25) For a number on the unit circle in the complex plane (r1=11).24). θ11−1θ21=1−13π212 and.3 Graphical representation 233 (ii) r1 r2 = 2 .26) (8. when all the numbers are equal to z1=1r (cos1θ1+1i1sin1θ ). z2 1 13π 13π cos = + i sin 12 12 z1 2 0 Exercises 23. by equation (8. z1z2-1zn1=1r1r2-1rn[cos(θ11+1θ21+1-1+1θn)1+1i1sin(θ11+1θ21+1-1+1θn)] In the special case.20) that the product of three or more complex numbers has modulus equal to the product of the moduli of the numbers and has argument equal to the sum of the arguments. | z1z2 z3 |1=1| z1z2 |1×1| z3 |1=1| z1 |1×1| z2 |1×1| z3 | arg1(z1z2z3)1=1arg1(z1z2)1+1arg1(z3)1=1arg1(z1)1+1arg1(z2)1+1arg1(z3) Therefore z1z2z31=1r1r2r3[cos(θ11+1θ21+1θ3)1+1i1sin(θ11+1θ21+1θ3)] and. z1 r1 = cos(θ1 − θ 2 ) + i sin (θ1 − θ 2 ) z2 r2 13π 13π = 2 cos − + i sin − 12 12 13π 13π = 2 cos − i sin 12 12 (iii) z22z11=1(z12z2)−1 and.

(a1+1b)51=1a 51+15a4b1+110a 3b31+110a 2b31+15ab41+1b5 so that (cos1θ1+1i1sin1θ )51=1(cos51θ1−1101cos31θ1sin21θ1+151cos1θ1sin41θ ) +1i(51cos41θ1sin1θ1−1101cos21θ1sin31θ1+1sin51θ ) 3 Abraham de Moivre (1667–1754).3 The formula is also valid for other values of n. Newton would tell visitors who came to him with questions on mathematics to ‘go to Mr. EXAMPLE 8.7 Express cos15θ and sin15θ in terms of sin1θ and cos1θ. cos21θ1−1sin21θ1=1cos12θ.5. Similar formulas.234 Chapter 8 Complex numbers This is de Moivre’s formula for positive integers n. For example. two complex numbers are equal only when the real parts are equal and the imaginary parts are equal.28) n = 1 = cos( − nθ ) + i sin ( − nθ ) cos nθ + i sin nθ de Moivre’s formula can be used to derive many of the formulas of trigonometry (see Section 3. de Moivre. For example. the expression on the left side of equation (8.23).24)). expressing sin1nθ and cos1nθ in terms of powers of sin1θ and cos1θ. de Moivre was a friend of Newton.27) is expanded by means of the binomial formula. Therefore. 1 (cos θ + i sin θ ) so that (cos1θ1+1i1sin1θ )−n1=1cos(−nθ )1+1i1sin(−nθ )1=1cos1nθ1−1i1sin1nθ (8. are obtained in this way for any positive integer n. By the binomial expansion (7. equation (7. by equations (8. .27) with n1=12 is (cos1θ1+1i1sin1θ )21=1cos12θ1+1i1sin12θ Expansion of the left side of this gives (cos21θ1−1sin21θ )1+1i(21sin1θ1cos1θ )1=1cos12θ1+1i1sin12θ A single equation between two complex numbers is equivalent to two equations between real numbers. The ﬁrst form of the formula occurs in a Philosophical Transactions paper of 1707. 21sin1θ1cos1θ1=1sin12θ (see equations (3.4). In his later years. ﬂed to England in 1688 from the persecution of the French Huguenots.14). equation (8.27) and (8.14). and its real and imaginary parts equated to the corresponding terms on the right side of the equation. or by using Pascal’s triangle. The generalization and signiﬁcance of de Moivre’s formula are discussed in Section 8.23) and (3. he knows these things better than I do’.

Such a function differs in no essential way from a real function of one variable.4 Complex functions Let g(x) and h(x) be (real) functions of the real variable x. except that the possible values of the function are complex numbers. 0 Exercises 25–27 8. where g(x) and h(x) are real.4 Complex functions 235 Therefore. The quantity f f * plays an important role in wave theories. (iii) Find | f(x) |2 (i) f (x)1=1(2x21+17x1−14)1+1i(2x1−11) (ii) g(x)1=12x21+17x1−141=1(2x1−11)(x1+14)1=10 when x1=1122 and x1=1−4 h(x)1=12x1−111=10 when x1=1122 Therefore f(x)1=10 when x1=1122 (iii) | f(x) |21=1g(x)21+1h(x)21=14x41+128x31+137x21−160x1+117 0 Exercise 28 . then f (x)1=10.29) where i = −1 . The discussion of the properties of complex numbers applies to complex functions. the complex conjugate function f *(x) is deﬁned by f *(x)1=1g(x)1−1ih(x) with property f(x) f *(x)1=1| f (x) |21=1g(x)21+1h(x)2 (8. cos15θ1=1cos51θ1−1101cos31θ1sin21θ1+151cos1θ1sin41θ1=1161cos51θ1−1201cos31θ1+151cos1θ sin15θ1=1sin51θ1−1101sin31θ1cos21θ1+151sin1θ1cos41θ1=1161sin51θ1−1201sin31θ1+151sin1θ and the ﬁnal expressions have been obtained by using the formula sin21θ1+1cos21θ1=11. h(x)1=10. when the wave function is complex. (ii) Solve g(x)1=10. A complex function of x is then f (x)1=1g(x)1+1ih(x) (8.31) (8. EXAMPLE 8.30) where | f(x) | is the modulus of the function. for example.8.8 (i) Express the complex function f (x)1=12x 21+1(71+12i)x1−1(41+1i) in the form f (x)1= g(x)1+1ih(x).

forcing it to introduce a rule. so that eiθ1=1cos1θ1+1i1sin1θ (8.4 Thus. still in force. restricting publications to four pages. if x and y are real variables then z1=1x1+1iy is a complex variable.33) 4 The ﬁrst discussion of functions of a complex variable appeared in a letter by Gauss to Bessel in 1811. y) where g(x. respectively (see Section 7. and is discussed in the following section.236 Chapter 8 Complex numbers A more advanced application of complex numbers is in the extension of the concepts of variable and function to include complex variables and functions of a complex variable. the function f(z)1=1z 21+1z1+11 can be written as f(z)1=1(x1+1iy)21+1(x1+1iy)1+111=1(x 21−1y 21+1x1+11)1+1i(2xy1+1y) =1g(x. The theory was developed independently by Cauchy from about 1814. and f(z) is a function of the complex variable z. . can be expanded in the familiar inﬁnite series ez = 1 + z + If z is the imaginary number z1=1iθ then eiθ = 1 + ( iθ ) + ( iθ )2 ( iθ )3 + + 2! 3! θ3 θ5 θ7 + i θ − + − + 3! 5! 7! z2 z3 + + 2! 3! (8. the real and imaginary parts of f (z).32) θ2 θ4 θ6 = 1 − + − + 2! 4! 6! The real and imaginary parts of this function are the series expansions of the trigonometric functions cos1θ and sin1θ. For example. y)1+1ih(x.6).1y). He swamped the weekly bulletin of the Paris Academy of Sciences. 0 Exercise 29 8. together with a description of the geometric interpretation of complex numbers.1y) and h(x. For many purposes in the physical sciences only one such function is of importance. Cauchy was the most proliﬁc mathematician of the 19th century. Comptes Rendus. The properties of f (z) as a function of the single complex variable z are more general than the properties of real functions.5 Euler’s formula It is known from the theory of functions of a complex variable that the exponential function e z. are real functions of x and y. where z is a complex number.

..38) EXAMPLE 8.. . .. .36) From the polar form of a complex number... ... . .. . . ... and as θ varies from 0 to 2π the function deﬁnes the unit circle (Figure 8. . . . .5 It forms the basis for the uniﬁed theory of the elementary functions. .5 Euler’s formula 237 This relation between the exponential function and the trigonometric functions is called Euler’s formula. .... .35) sin θ = (8. . it now follows that every complex number can be written as z1=1x1+1iy1=1re iθ (8.5 that π π z = 1 + i = 2 cos + i sin 4 4 The formula.. . ... . .. .6 (8. .. ... ... and others for the logarithmic and trigonometric functions of z. ... . . .. . . . . .. . .. .. . . . ..... .. ... ..... z −1 = 1 1 = e− iθ x + iy r (8. . . .. ... .. . . .. . . .. . . ... .. . . .... . . . ..... . . . . . . . . . . .. Then zz*1=1zz−11=1eiθe−iθ1=1e01=11 (8.. . . .. ... . . . .. . .. . . . .9 The number z1=111+1i.. . . z*1=1e−iθ1=1cos1θ1−1i1sin1θ and this is also the inverse.... . . . . ... equation (8. . . . . .... . .34) The pair of equations (8.. .. . .... ...... .. . . . . ... . .. The complex conjugate and inverse functions of z are z* = x − iy = re− iθ . .. ... .. 5 x −e e • • θ θ − 1 1 y o .. .. ... . ... .. . .. . . . . .. . .. . For each value of θ. . .. . . . .. . ... .. .. .. ... . . . .. ...... . . . . . . . . ... .. . . .. .. .. .... . . .. . z−11=1z*1=1e−iθ... . ... ... . . θi . .. . . . .. . . the number lies on the unit circle of the complex plane........ . . . ....8.. . . . . . . . .. .. .37) where r1=1|z | and θ1=1arg1z. . . ....... .. ... .. . . . .. .. ... . .. ... . . . ... . .. . ... It was shown in Example 8. ..... . .... . ... .... . . . . . The function z1=1e iθ has modulus | z | = cos2 θ + sin 2 θ = 1 and argument arg z1=1θ.. ..6).. .. .... .33) and (8. ... .. .. . .15). .. . . .... .. appeared in Euler’s Introductio of 1748..... ... .. .. . ..... . .. . .34) can be inverted to give relations for the trigonometric functions in terms of the exponentials: cos θ = 1 iθ − iθ e + e 2 1 iθ − iθ e − e 2i Figure 8.. . .... and it is one of the important relations in mathematics. . .. .. . .. . . ... .. . θi ... . ... .. The complex conjugate of z1=1eiθ is .. ... ...

sin π 4 = 1 iπ 4 e − e− iπ 2i 4 0 Exercises 30–33 EXAMPLE 8.238 Chapter 8 Complex numbers The number can therefore be written as z = 2 eiπ 4 with complex conjugate and inverse π π z* = 2 e− iπ 4 = 2 cos − i sin 4 4 z −1 = Also. involving the transcendental numbers e and π.11 The number e iπ.10 Express in cartesian form x1+1iy: (i) 2eiπ 3 = 2 cos π 3 + i sin π 3 = 2 1 2 + i 3 2 = 1 + 3 i (ii) e−iπ221=1cos(−π22)1+1i1sin(−π22)1=1−i 0 Exercises 34–36 EXAMPLE 8. cos π 1 − iπ 4 1 π π e = cos − i sin 4 4 2 2 4 = 1 iπ 4 e + e− iπ 2 4 . and the imaginary unit i.39) This relation. eiπ1=1cos1π1+1i1sin1π Because cos1π1=1−1 and sin1π1=10 it follows that eiπ1=1−1 (8. By Euler’s formula. 0 Exercises 37–39 0 Exercises 40–42 . is probably the most remarkable relation in mathematics. the negative unit −1.

. . .. . .. .. .. .. . ... .. .. ...... . .... . . . . . . .. . .. . .. . ... . . . .. . .. .. . . .. ...... . .. ... ....... ... .. .. ..8.... .. . .. .. . .... .. . . . . .. . . . . . .. . . ... . ... ....7 y . . . . . .. . .12 The square root of i: We have i1=1eiπ22. . . (8.. . .. . . .. . ... . .... . .. .. . . .. from z1=1x1+1iy1=1r1cos1α1+1ir1sin1α ) ) with coordinates x′1=1r1cos(α1+1θ)1=1r(cos1α1cos1θ1−1sin1α1sin1θ) y′1=1r1sin(α1+1θ)1=1r(sin1α1cos1θ1+1cos1α1sin1θ) Figure 8. . EXAMPLE 8. . .... .. . .. . .... ... .. . . . . .. . the result is einθ1=1cos1nθ1+1i1sin1nθ But because einθ1=1(eiθ )n. . . . . . .. .. . . . .... .. .... . .. . .. x( z . .. y1=1r1sin1α to o . .. .. . .. .... ..5 Euler’s formula 239 de Moivre’s formula When θ in Euler’s formula... .. .. . . . .. . ... ... . .. .. ... . Therefore Check: ± i = ± eiπ 4 = ± cos π 4 + i sin π 4 = ± 1 2 1 + i ( 1 2 1 + i ) 2 = 1 (1 + i 2 + 2i ) = i 2 0 Exercise 43 Rotation operators When a complex number z1=1reiα is multiplied by eiθ the product is a number with the same modulus as z but with argument increased by θ : e iθz1=1e iθre iα1=1re i(α+θ) Graphically.. ... (8. .. .. it follows that (eiθ )n1=1(cos1θ1+1i1sin1θ)n1=1cos1nθ1+1i1sin1nθ (8...... ... . .... . . .. . .... . . is replaced by nθ where n is an arbitrary number. .. . ... .... . generalized for arbitrary numbers n (that can themselves be complex).... ... . . . . ..... . .... . .. ... . . .. .... . . . . .. . .. . .. ... . • with cartesian coordinates x1=1r1cos1α1.... ..... . . . . .. . .. . .. . . .. . . . . . . .. ... . . .. . . . .. . .. . . . . . .. .. . . as shown in Figure 8.. . ... .. . . ... .... . . ... . .40) This is de Moivre’s formula. . .. . . . . . . ... .33).x( z • α θ eiθz1=1z′1=1x′1+1iy′1=1r1cos(α1+1θ)1+1ir1sin(α1+1θ) y .... .. ......7. .. . .. . ... . .... .... . . . . . .... the multiplication corresponds to the (anticlockwise) rotation of the representative point through angle θ about the origin of the complex plane. . . . .. .. . . . .. .. . ... .. . .. . .27)...

. . .. . . . ....... ... . . ... . ... n1=10... . . . The n nth roots of 1 . . . . .. . . .. .. .12 0z zk1=1e(2πk23)i. . . . .. . .1= (8. . ... . .... ... .. .. . . ... . . .. . . . . . . . . . . . ... . ... . . . . .. . . .... . . ... . . .. . .. . . . . . .. These numbers have the property Figure 8. . . . .. |n| clockwise rotations if n is negative)... .... . . . .. . .42) Equations (8. ..240 Chapter 8 Complex numbers The coordinates of the rotated point are therefore related to the coordinates of the unrotated point by x′1=1x1cos1θ1−1y1sin1θ y′1=1x1sin1θ1+1y1cos1θ (8.. . . ...... . The function eiθ occurs in the physical sciences whenever periodic motion is described or when a system has periodic structure. . . . . the function is unchanged when a multiple of 2π is added to θ : ei(θ +2πn)1=1eiθ. ... at the vertices of an equilateral triangle.. .. ..... . . .... .. . . . .. . ±1.. . .41) play an important role in the mathematical formulation of rotations (see Chapter 18).... . . . .. . . . . . .... . y′) (8. . ... . . .... . . . It follows that the exponential function eiθ is also periodic with period 2π. . . . .. . .. . . . . . . . . y) of a point z into the coordinates (x′. .... .8 shows three equidistant points on the unit circle... .. . 0 Exercise 44 8... . . .. .. We consider here three important representative classes of physical situations exhibiting periodic behaviour.. .... .. . . . ... . . ..43) Graphically. Thus.. . .. . . .. .. if θ is increased by 2π. . .11.. .. . it follows that e 2πi1=11 and ei(θ +2π)1=1eiθ More generally. .. . . .. . . .. .. .. . ...6 Periodicity The trigonometric functions cos1θ and sin1θ are periodic functions of θ with period 2π (see Section 3. ... . . ( z k ) 3 = e( 2 π k ( 3) i ) =e 3 2 π ki =1 2z • with unit modulus and arguments 2πk23. . .. .... .... .. . .. .. . . ... .... .. .... . . ei(θ +2π)1=1eiθ1×1e 2πi1=1eiθ(cos12π1+1i1sin12π) Therefore. changing the argument θ by 2πn corresponds to moving the representative point on the unit circle through n full rotations back to its original position (n anticlockwise rotations if n is positive. .... ..8 • k1=10. . ... .. . .. .. ....... .. . . . .. . . . . . . ... . ... .. ... .. ... .. ..... . .... .... . . .. . . .. because cos12π1=11 and sin12π1=10... .. . . ... .2). .. . .. . . . ... . 1z • Figure 8. . .... Periodicity on a circle. .. . .. ±2... .. . . .. The points correspond to complex numbers .... . . .. .. .. ..... . . .. . .. .... . ... ... . . .. . . . ... .. . . .. ...... ... . . ... . . . . .. ... y)1=1(x′. .. ... . ... .. .41) The function eiθ can be regarded as a representation of the rotation operator Rθ which transforms the coordinates (x. . . .... . . y′) of the rotated point z′: Rθ (x. . . ... . .. . .. ..... . ... . ..

.. . ...... . . . . . . .. . . . . . .. .. . .. . ... .. ... ..... .. ... . ± ( n 2 − 1).. . . .. 3 ... . .. .. . . ...... ....z • 2... .. . . . .. . . The six roots are zk1=1e2π ki26. .... ... . .6 Periodicity 241 so that each is a cube root of the number 1: z01=1e01=11 z1 = e2 π i 3 = cos z2 = e4 π i 3 = cos 2π 2π 1 3 + i sin =− + i 3 3 2 2 4π 4π 1 3 + i sin =− − i 3 3 2 2 We note that z1 and z2 are a complex conjugate pair..13 The six sixth roots of 1. .8. . .. .. .. n 2 if n is odd if n is even zk = e2 π ki n (8. ±2.. . n22). ... conveniently as zk1=1e 2πki23... . . . . . ... . ±1. . . ..... .. . ... . . . . . . . ...... . . . . . . . .. .. . .... ..... . .. . .. ... .. . ± ( n − 1) 2 for k = 0. . . ... . . .. . .. The n representative points lie on the vertices of a regular n-sided polygon... ±1.. .. the only real root is +1 (for k1=10) and the rest occur as complex conjugate pairs. ... . . . .. .. . .. ...... 3 3 z31=1eπi1=1cos1π1+1i1sin1π1=1−1 0 Exercises 45–47 6 Figure 8. .. ... ± 2. . . .. .. . ... . . ... .. .... .. . . ..... . . ... . .. ... ....... . or z01=1e01=11 • k1=10.. ..... ... .. .. .. .... ... . .... .. ±1 (for k1=10.. . .. . . . ..... . .. . . . ... . .. the three roots can be speciﬁed by any three consecutive values of k.. . . .. . . ….... .. .. . .... . .. ....... . . . ... ... . . ±1 such that z01=11. . . .. ... . ... . . . ... .. ... …. ... .. ... ......... . . ..... ... .44) Thus when n is odd... . . ... ..... ...... . two of the roots are real. . ..... . . ... .... .... ± 2. . ... ..... ....... .... ... . . . .. ... . . . .. . . .. When n is even.... .. . . .. . . . ... .. with e4πi231=1e−2πi23... EXAMPLE 8. . .. .. .. . 1. ..... . . . .. . .. .... .. ±1. . .... ..... .. .. k1=10... ..z • z±2 = e±2 π i 3 = cos 2π 2π 1 3 ± i sin =− ± i 3 3 2 2 • z±1 = e± π i 3 = cos ± i sin = 0z 3z π π 1 3 ± i 2 2 1z • 2z • ... .. .. .. .. .. . .. . Because of the periodicity of the exponential... .... .. ... . . z±11=1e±2πi23... .9 The nth roots of a complex number were discussed by de Moivre in a Philosophical Transactions paper of 1739... ... . .... . .. In general the n nth roots of the number 1 are6 0.. .

... .. . ....... • m . 1... ...... ...... . . . Thus f(θ1+12π2n)1=1ein(θ +2π2n)1=1einθ1×1e 2πi1=1e inθ1=1f(θ ) Such functions are important for the description of systems with circular periodicity... .. ... .13) the system has moment of inertia I1=1µ r2 where µ1=1m1m22(m11+1m2) is the reduced mass and r1=1r11+1r2 is the distance between the masses......... . .. and c in the z-direction..... .... .10 Figure 8... Such a x (8. . (8.. . ............. ...... . ... Periodicity on a line (8.. .... . ... .. m1 and m2..... ... .. ... ... . . As discussed in Section 5........ .... .. .... ......... ... . ...... .. . .. A function of x that has the same periodicity as the lattice must satisfy the periodicity condition f (x1+1a)1=1f(x) The simplest such function is f (x)1=1e 2πxi2a Thus. . . ... ........ .242 Chapter 8 Complex numbers A function that has the same circular periodicity as the ﬁgure with n equidistant points on a circle is f(θ )1=1e inθ This function is periodic in θ with period 2π2n............ ...... ................ ... ...... . ... .... .. joined by a rigid rod (of negligible mass) rotating about the centre of mass at O... Figure 8..... .... .... ... ........ ..... ....... . ........ The functions are readily generalized for three-dimensional periodic systems: the function f(x. . .. ... . .. ...... . . . 2 o 2 r ..... ..45) Figure 8...... . .. ..... .... . . .. .... .... • ... . .. .. . ...... .. ..... ... .. ... . ...... . .. . ....... .......... y. .. .... .... ......... ....47) Rotation in quantum mechanics ........ b in the y-direction. . .. .... ..... . Figure 8..... for example. ............. ... ......... ... ..... . ...47) are important for the description of the properties of periodic systems such as crystals... ... .. . a linear lattice.... . .. . .10 shows a simple linear array of equidistant points representing........... .... ... 1m ... ... .. ..11 θ .......... z)1=1e 2πxi2ae 2πyi2be 2πzi2c has period a in the x-direction.46) (8. .... ... ··· r . . . . . .. . .... ... f (x1+1a)1=1e 2π(x+a)i2a1=1e 2πxi2a1×1e 2πi1=1f(x)e 2πi1=1f(x) Functions like (8... .11 shows a system of two masses.... .. ..48) ...... ... . . . • a • a • a • a • a • a • ··· ............. .............. . ..... . ... ...... ........ .....6 (Example 5... ... . .... . ....... .... . ... . ..

7 for a more complete discussion). and it is readily veriﬁed that a solution of this equation is ψ (θ )1=1Ceiaθ (8. dψ = iaCeiaθ . dθ d 2ψ dθ 2 = ( ia )2 Ceiaθ = − a 2 Ceiaθ = − a 2ψ For the solution (8. The corresponding values of the energy E1=1A2a222I are En = n 2I 2 2 (8. Thus. ψ (θ1+12π)1=1Ceia(θ +2π)1=1Ceiaθ1×1e2πai1=1ψ (θ)1×1e2πai and the periodicity condition is satisﬁed if 2πa is a multiple of 2π. it must satisfy the periodicity condition ψ (θ1+12π)1=1ψ (θ ) (8.53) where the ‘quantum number’ n has been used to label the solutions.1= The physically signiﬁcant solutions of the Schrödinger equation are therefore ψn(θ)1=1Ceinθ. ±1. Equation (8.49) can be written d 2ψ dθ 2 = − a 2ψ (8.6 Periodicity 243 system is called a ‘rigid rotor’ and the equation of motion in quantum mechanics (the Schrödinger equation) for a rigid rotor in a plane is − 2 d 2ψ 2I d θ 2 = Eψ (8. if a1=1n for n1=10. ±2. and represent rotation.51) where C is an arbitrary constant (see Section 12. The rigid rotor is used in chemistry to model the rotational motion of a molecule.49) where the wave function ψ1=1ψ (θ) is a function of the orientation variable θ.51) to be physically signiﬁcant.54) . and E is the (positive) kinetic energy of rotation. ±2.50) where a 21=12IE2A21>10.52) For the function (8. it is necessary that the wave function be unchanged when θ is replaced by θ1+12π. that is. ±1.8. n1=10.51).1= (8.

0 Exercise 48 8.45) for the possible periodicities around a circle. Ordinary integration over complex functions obeys the same rules as integration over real functions.244 Chapter 8 Complex numbers We note that the quantization of the energy of the system has arisen as a consequence of applying the periodicity condition (periodic boundary condition) to the solutions. We consider the general form I = Z eax cos bx dx Because cos1bx is the real part of e ibx it follows that the integral I is the real part of the integral obtained from I by replacing cos1bx by e ibx: I = ReZ eax eibx dx = ReZ e( a + ib) x dx The complex integral is evaluated by means of the ordinary rule for the integration of an exponential function. it is shown in Example 6. For example. how the integral Z e− ax cos x dx can be evaluated by the method of integration by parts.7 Evaluation of integrals Integration with respect to a complex variable is an important part of the theory of functions of a complex variable. In addition. An alternative. method is to express the trigonometric function in terms of the (complex) exponential function. complex functions can be used to simplify the evaluation of certain types of integral. Z e( a + ib) x dx = e( a + ib) x eibx = eax a + ib a + ib and this can be resolved into its real and imaginary parts: Z e( a + ib) x dx = eax = cos bx + i sin bx a + ib ( a cos bx + b sin bx ) + i( a sin bx − b cos bx ) ( a + b2 ) 2 eax . Thus (ignoring the constant of integration). but is used only in advanced applications in the physical sciences. more elegant.13. We note also that the set of wave functions includes all the functions (8.

(21+13i)1+1(21−13i) 3. If z1=131−12i. (51+13i)(31−1i) 6. Section 8. 8 8 26. Solve the equations: 10. (11+12i)5 7. 11−1i 19. For z = 3 cos + i sin ﬁnd (i) z4. 50 eax ( a sin bx − b cos bx ) a 2 + b2 (8.5 Exercises 245 The integral I is the real part of this: Z eax cos bx dx = The imaginary part is a bonus: eax ( a cos bx + b sin bx ) a 2 + b2 (8. 1 5 + 3i 14.2 Express as a single complex number: 1.55) Z eax sin bx dx = 0 Exercises 49. z1 = 5 cos + i sin . (iii) z22z1 as a single complex number for π π π π 23. (11−13i)(11+13i) 5.8 Exercises Section 8. 2i 22. 3+i 20. (ii) z−4. (21+13i)1−1(21−13i) 4. Use de Moivre’s formula to expand cos18x as a polynomial in cos1x. 15. (iii) Express in polar form r (cos1θ1+1i1sin1θ): 16. z 21−12z1+141=10 1− i 1+ i 11. −61+16i 21. 9. (ii) ﬁnd the modulus and argument. −2 − 12i Given z1 and z2.8. (iii) Express the real and imaginary parts of z in terms of z and z*. −3 18. z1 = 2 cos + i sin . 12i 17. z2 = 3 cos + i sin 2 2 3 3 3π 3π 2π 2π 24. (21+13i)1+1(41−15i) 2. 13. . (ii) z12z2. Use de Moivre’s formula to show that (i) cos14θ1 =1cos41θ11−161cos21θ1sin21θ1+1sin41θ (ii) sin14θ1=141sin1θ1cos1θ (cos21θ1−1sin21θ) 27. express (i) z1z2. z 31+181=10 3 + 2i 3 − 2i 1 3 − 4i − 5 3 + 4i Express as a single complex number: 12.3 (i) Plot as a point in the complex plane. ﬁnd (i) z* and (ii) zz*.56) 8. Find z such that zz*1+14(z1−1z*)1=151+116i. z2 = cos + i sin 4 4 3 3 π π 25. (11−13i)2 8.

y)1=10 and h(x. 29. Show that ln1z1=1ln1| z |1+1i1arg1z 43. (ii) Find the (real) solutions of the pair of equations g(x.1= (i) Calculate the ‘normalization constant’ C for which (ii) Show that Z 2π ψ n (θ ) dθ = 1. ∞ ∞ Z e− x cos 2 x dx 0 50. (ii) sin1ix1=1i1sinh1x. e3πi22 39. 3eiπ24 35. (iii) Solve f(z)1=10 directly in terms of z to conﬁrm the results of (ii). Section 8.246 Chapter 8 Complex numbers Section 8. (iii) tan1ix1=1i1tanh1x 41. (iii) z−1 in exponential form reiθ: 30.7 Use complex numbers to integrate: 49. 5 1 47. Use Euler’s formulas for cos1x and sin1x to show that (i) cos1ix1=1cosh1x. 42. Locate them on the complex plane. 2 0 Z 2π ψ * (θ ) ψ n (θ ) dθ = 0 if m1≠1n. e−iπ23 36. Use de Moivre’s formula to ﬁnd the square roots of −i. Z e−2 x sin3 x dx 0 . z1=12i 33. (ii) z*. ±1. (i) Express the complex function f(z)1=1z 21−12z1+13 in the form f(z)1=1g(x. ±2. n1=10. eiπ22 38.5 Express (i) z. where g(x) and h(x) are real. Section 8. then f(x)1=10. and hence of f(z)1=10. y)1=10. y)1+1ih(x. z = 3 + i 32. 2eπi26 37. 4 1 46. y) where g(x. Express cos (a1+1ib) in the form x1+1iy. The wave functions for the quantum mechanical rigid rotor in a plane are ψn(θ)1=1Ceinθ. (ii) clockwise rotation through 30° about the origin of the complex plane. (iii) ﬁnd | f (x) |2. 8 1 48. h(x)1=10. (i) Express the complex function f(x)1=13x 21+1(11+12i)x1+12(i1−11) in the form f(x)1=1g(x)1+1ih(x). 44. y) and h(x. Find the number obtained from z1=131+12i by (i) anticlockwise rotation through 30°. e3πi 40. m 0 Section 8. y) are real functions of the real variables x and y. (ii) solve g(x)1=10. z1=1−3 Express in cartesian form x1+1iy: 34.4 28. z1=111−1i 31.6 Find all the roots and plot them in the complex plane: 45.

and (x.1 The values of the function (9. The expression on the right of the equation deﬁnes the function f(x. (x. EXAMPLE 9. y)1=1(0. Let the variable z be a function of the two variables x and y. the position along a line. that is. 2 . The variables x and y are called independent variables if no relation exists between them such that the value of one depends on the value of the other. 0). and the amount of substance n. More generally. the temperature T. For example. 0)1=1121−121×111×101−131×1021=11 f(0. examples are the thermodynamic functions of state. V is a function of the three variables p.1) when (x.9 Functions of several variables 9. n) = nRT p it is implied that the volume V of the gas is determined by the values of the pressure p. For example. y)1=1x 21−12xy1−13y 2 (9. mass density and potential energy were discussed in Chapter 5 as functions of one variable only. Functions of more than one variable occur widely in the physical sciences. and all those physical properties of a system whose values depend on position. y)1=1(2. 1) are f(2. 1)1=1221−121×121×111−131×1121=1−3 f(1.1 Concepts When the equation of state of the ideal gas is written in the form V = f ( p. T . 1)1=1021−121×101×111−131×1121=1−3 0 Exercises 1. functions of position are functions of the three coordinates of a point in ordinary three-dimensional space. the equation z1=1x 21−12xy1−13y 2 gives z as a particular function of x and y. 1).1) whose value for a given pair of values of x and y is to be assigned to the variable z. as in the above example. and n. T. y)1=1(1.

... . .. . . .. . ... .. . . ... . . . .. ...... .. . . .. . .. .. ...... . .. ...... . .. . .. . ... . .. . . ..... ... .. .. .. .. ... . . . and the value of the function is represented by the point P at height z above the plane (coordinate systems in three dimensions are discussed in Chapter 10). .. .. . . . . . .. . .... .. . . .. a function of several variables can be visualized. ..... . ... ..... ... .. .... .. . .. .. . . . .. ..... .... ... .. . . ... . .. ...... .. . ... ..... . . ..... .. . . ... ... . . .. .. . . .... ....... . ... . . .... .. . ... . .. Figure 4.. ....... .. .. .. .... ... . ... . . .. . ...... .. ... . . ...... ..... ... . . .. . ........ .. . .. . .. .. ...... .. .... ... . . . . . ... .. . .. . . . . . .. .... ... ... ... . . .... . whilst Figure 4. .. .... .. . . . ... . ... .. ... ... .. . . ... .. .. . ..... .. .. . . . . . .. .... .. Oy.. ....... . .. ....... ..... and Oz are three perpendicular axes. .. .. . . . . ... .. . . . .. Examples of such plots are Figures 4.. . . . . .. . .. .. ... .. .. . ...2 that a (real) function of one variable deﬁnes a curve in a plane... . ... ... . . ...... ..... .. . .. .. . .. ... . .. .. .. . ... . . ... .. .. .. ... ... . .... . . . . . .. .. .... . the pair of values (x... ... . .. .. . ........ .. .... .. In the general case.... ..... .. ... ... .... ... .. . x y x • y z o .. ... . . . . . . ... . ... .... . . . .. .. .. .. ... . ... ..... .. . . . Figure 9..... . ... . .. .. . . .. ..... . ... ..... . . .... ..... . .... .. . . ...... .. ...... . ........ . . ... . .. .... .. .... . . . .. ... .. . .. .. . . . ... . .. . .. . .. ..... .... . .... ... . ... ... .... ... . .... .. . . ...... .. .. . . .. y . ........ .. . . .. .. . .. ...... ... . .. . ... . . ... . .. .... . .. . .. .. . . ....... ..... ... ... . ....... .. . .. .... . .. . ....... . ... .... .. .. .. .... .. .. . . .. . . . .... . ... . .. . ... . .. . .. ...... . . .. for example.... . .. .. ... . . . ... .... .. . . . ............ . ... ... .. . .. .... .. ... . ... .... . . . .. .. . ..... .. . the graph of y1=1x21−12x1−13 is shown in Figure 2. . .. V1=1f( p.... ... .. .. . .... . ... .1) as a function of x for several values of y.. .... . . . . ... . . ... ..... . . .. .. . .. .. . .. .. . . .... .. .. .. .... ... . . . . . . . .. . .. ... .. .. .. . .. . .. .. ....... . ... ..... . . . .. . .... .. Ox... ... .. . and the surface is the representation of the function. .. . . . . .. . ........ ... .. .. that is. .. .. . ... .... .. ... ... . .. . ...... .. . .. . .. . ........ ... .. . . ..1... .... . ... . .. . . . y) moves in the xy-plane. . ... . . . . .. . . ... but such complete physical representations are not possible for functions of three or more variables... . .. . . . ... ... . . .... . .. . . . . . ... ... . .... .. . .. . .. ....... .... . ... ...... ... . . . the locus of the point P maps out a surface..1 is the graph of V as a function of T with p and n held constant. .... .. .. .... . ...... . . .... .. .. ...... ... ... . As the point (x.. . . .... . .. . .. ... . . . . . . ........ . . . . ..... . . . . . ... .. . . . . . .. . . ... . .. ...... .. . . . . p • z 9. . . .... ... . .. .. . . .. . .. ..... . .... . . . . ... ... . . . . ... . . . . .. ..1. .. . . ..... . ...... ... . . .. ... .. ... .. . . .. .... .. . . ... .... . ... ... .. .. .. .. .. . . ...... . .... ... . . . .... .. .. . . ... . . .. . ... .. . .... ..1 and 4. ..... ... . . . .. . . ... ... ..... . . .. .. . . . ..1 Figure 9. ... . ... . .. ... . y) speciﬁes a point in the (horizontal) xy-plane. at least in part.. .. ..... . Such simple graphs are often the most useful representation of a function..... . .... .. .. . and plotting the resulting function of the one variable only.. . ... ..... ... . .. . ... .. ... .... ..... . . . . .. . . . .. . .... . .. ... . . .. .. . . . . . . .. ...... .. . . . ..... . .. . .. .... . ........ .. .... . . . .. .. ... . . . .. ... . . .. the point P moves on the surface. .2 . ... . .... .. . . . .. . ..... . . . . .. ... .. . . .. .. . . . .. ... ....... .. .. . . .. . . . . .... . . .. .... . ... . ... ... . . .... .... . .. ....... . .. . ... . .. .. . ..... . . ..... .. . ..... ...... ..... ..... .. ... ... . .. .. . ...... .. . .. . .. . . ... .. . . .. ... .. . deﬁnes a surface in a three-dimensional space. . . . . . .. ...248 We saw in Section 2.. .. . ...... z1=1f(x.. ... . . . . .. . . ..... . . ..2 1 Graphical representation Figure 9.. .. .... .... .. .. .... .... ... ... .. .. . .. Chapter 9 Functions of several variables The representation of surfaces by functions of two variables and the concept of partial derivatives were ﬁrst considered by Leibniz in the 1690’s. . .... ... .... .. .. . ... . . ... ..... .. . .. . .. . . .. . . . It is possible to draw beautiful three-dimensional representations of functions of two variables by means of modern computer graphics. ... .. . . . .. . . .... .. . . . .. ... .. . .. .. . . . . . . . . . ... ... . . ... . . . ) y ... ... ... y .. . ...... . .1 In Figure 9. ... . .... . ... .. . . ..... . .... . .. ... .. . . ...... . . .... . .... .. . .. .. .... . ... ..... .. ..... .. .. . ... .. . ............ . .. .. . ... .. .. .. ... .. . .. . .. . . . .. . . ... . . .. . . ... n)1=1nRT2p. . . . .2 for the volume of the perfect gas. . . . . .. . . . . .... ... . .. .. . . ...... . . ... . .. ....... .. . y y . . . .. . .... . ... .. . . .... . ... .. . ..... . . . .... . ...... .. . ... . y).. . . .. ... . .. .. .. . . . A function of two independent variables. .... . . . . .. ... . .. .. ... .... . .. .. .. . ... ...... . .. ... . . . . .. . .... .. . ... . . . . .. .. . ... .. .... . T.. by assigning values to all the variables except one.... . . ....... ..2 shows graphs of the function (9. . ... .... . .. ... .. . .. ... . .. . ... . . ..... . .2 is the graph of V as a function of p at constant T and n.. ... .. .. ..... .. ... ..... .... .. .x( f .. .... . . ... . ........ .... .... . . .. . ....... ...... . . . . ... . ... . . .... . . . . . . . . .. .... .. . . .... .. . ..... . ... . ...... ... .. .. . ... ... ... 3= 2= 1= 0= x . . ..

and dynamically as the rate of change of the function with respect to the variable.3. y ) = lim ∆y→0 ∂y ∆y (9. The plane ABC is parallel to the xz-plane. the plane DEF is parallel to the yz-plane. If these conditions are satisﬁed then the partial derivatives of a function of two variables are deﬁned by the limits (compare equation (4. y)1=1x 21−12xy1−13y 2 can be differentiated with respect to variable x.9. so that y1=1constant in the plane and the values of ∂z2∂x for this value of y are the slopes of the tangent lines to the curve APB.3) The geometric interpretation of these quantities is shown in Figure 9. In the same way. For example. but began to be accepted only after Jacobi used it in his theory of determinants in 1841. and the values of ∂z2∂y are The notation ∂z2∂x was ﬁrst used by Legendre in 1788. 9. In the general case. the function z1=1f(x. For a function of two or more variables there exist as many independent ﬁrst derivatives as there are independent variables. y + ∆ y ) − f ( x.3 Partial differentiation We saw in Chapter 4 that the ﬁrst derivative of a function of one variable is interpreted graphically as the slope of a tangent line to its graph.2 and with respect to y at constant x for the partial derivative with respect to y ∂z = − 2x − 6 y ∂y The existence of partial derivatives and the validity of the operation of partial differentiation are subject to the same conditions of continuity and smoothness as for the ordinary (total) derivative.8)) f ( x + ∆ x. with y treated as a constant.3 Partial differentiation 249 Each of these graphs is a planar ‘cut’ through the three-dimensional surface. to give the partial derivative of the function with respect to x ∂z = 2x − 2 y ∂x (read as ‘partial dz by dx’). a function of n variables deﬁnes a ‘surface’ in an (n1+11)-dimensional space. y ) − f ( x.2) (9. 2 . y ) ∂z = lim ∂x ∆ x → 0 ∆x ∂z f ( x. and its graph as a function of one of the variables is obtained by taking a planar cut through the representative (n1+11)-dimensional surface.

.. . .... . . ..... . ... .. .. ... . .. ... .... . ... . . ... .. . .. . . .. . . ... .... . . ... . .. .. . ... .. .. .. . . .... ... .. . .. . . . . ... . .. . . . . .. .. .. .. . .... .. .. ..... ... .. ..... . The two tangent lines at P in Figure 9....... ... . .. . ....... . .... . .... . .. . . .. . . . .. . .. ... .. . .. ... . .... ..... . . . . .. . .... .. .. . . .. ...... . .. . .. . .. ..... .. .. Every other line through P in the tangent plane is then also a tangent line and its slope is the derivative of the function in some direction.... .. .. .. ... ... . . .. . .. ...... . . .. .. .. . . .. .... .. . . ... . ... .... . . . . .. . ..... .. .... ... .. . .... . ∂z ∂z ∂z cos θ + sin θ = ∂y ∂r ∂x Figure 9. ...... . .... .. .... . ..... .. . .. .. ..... .. .. . . . ... . . . . .. . . .. . . .... . ..... . . . .... ... Such a derivative can be expressed in terms of the ‘standard’ derivatives ∂z2∂x and ∂z2∂y. . . .. .. .. ... .. ... . .. . . . . . . . . . .. . .... . . . . . . . . . .. ... . ... . . .. ... .. .. .. ........ . .. ............ .... ..... ... . . . .... .. .. ... . ...... ... . . ... .. . .... .. .. . .. .... . ... . .. . . .... . . . .. . ... ..... . . . ... . . .... . .... ... . .... . ... .. .. ..... . ... ... . . . . . . .. . . . .... .... ... .. .... . . . .. . . . ... .. . . .. .... .... .... ... . ..... .. .17)... . ... . x ∂/ z ∂ θ o y ∂/ z ∂ r .... . . . .... . .. .. .. ..... ...... .... .. .. . .. . x1=1r1cos1θ... .. . the distance r along the direction at angle θ to the x-direction is given by the pair of ‘parametric equations’ x ... .. .. . . . .. ........ . .. . .. . . .. . . .. .. . . . . . .. ... . . . .. . . .. . . .. .. ... .... . . . ... . . .. .. . .... . .. . ..... . ... . ... .... . .. . . . .. . .. . . . . ... . .. .. .... . . .. . .. .... .. . . . . . ... . .. . . .. . . .. . . ... . . ... . ... ..... .. . .... . ... . .. The two tangent lines at point P in Figure 9... ... .. . . . . . ... ...... . .... . .... . .. . . . . ... . .. . . . .. .. . ... . . . . . .. . ..... . . . . . . . .. ... .... .. . . . . . .. . . ... .. . . . . . .. . . .. .. . . . . .. .. .. ... .... . .... ... ... ...4.. .. . . ... .. .... . . . ..3 deﬁne the tangent plane at P.. . . . . .. . .... ... .. . .... that is. ..... . .. . .. . .... . .. .. .. .. . .. ... .... . . .... ... .... ... ..... ...... . .. . .. . ... ... .. .. . . . .... . .... . . . ... . . . ..... .. . .... ... . . . . .. . .. . .. ... ..... .. . . .. ... . ... . ... . .. . .. .. . ... . . . .... ... . .. . . . .... . . . . ... . . . . . . . . . .. .... . . ...... . . . . ... with ∂z2∂x for the rate of change along the x-direction and ∂z2∂y for the rate of change along the y-direction. . ......5) x . . . ... . . .. . . ...... . .. ... . . . ... .3 therefore represent the rates of change of the function z1=1f(x.. . . . . . .... . ... .. .. . . . ..... .. .. b d z (see Example 9. . . . . ... .. . . ..... ..... .4 Figure 9. .. . .. .. . .. .. . . .. . . ....... . ... . .... .. ... . ... .. .. .. .. . ... . . . y) along this r-direction is Chapter 9 Functions of several variables In Figure 9... .. .. . . ... . . .. ... . . .. .. . . .. . . ...... ... .. . ... .. . .... ... .. ..... .. .. .. . ... a e y x • y f c z o •p ... . .... .. .... . . ... . .... ...250 and the derivative of the function z1=1f(x. ..... . y). . .. . . . .. . ... . .. ... . . .. .. . .. .. .. .. .... . . .... . . . .... ... . .. . ..3 y1=1r1sin1θ (9. ... . .. . . ..... . .. . ... . ... .. . .. ... . . .... .. . ... .... . . .... . .. . . . . ... .. .. . . . . . .. . . .. .. .... . .. .. .. . ... ... .. . .. . . .. .... . ... .. .... . . .... . . . . . . .. . .. . . .... . ... . .. . . ..... . . .. ..... . .. . . .. .. .. .... ... . ....... . . . .. . ... ..... .... . . . . .. . ... .. . ... ...... .. ... . .... . ... .. .... . ... .... .... .. .. ..... . . . .. .. .... . . . . .. ... . ... ...... . y the slopes of the tangent lines to the curve DPE. . .... .. .. . .. . .. . . .. . ..... . .. ...... . . .. ..... . . . . ...... .. .. . . ... .. .. the plane that touches the representative surface of the function only at the point P.. ... . . ... . . . ... . .. . . .. .... . . .. ... .. .. . . ... . . . . ... .. ... . .. .. . . . .. .. ... .. . .. .. .. . ..... . .. . ... . .... ... ....... . . . ..... . . . .. . . . . . . . .... ... . .. . .. .. .. .. ..... . . . . ... . .... . . ... .... . ... .. . . .. . . ..... ... . . . . ... . ... .... . ... . . . .. ... .. . ... ... ... . .. ...... . . . . .. ... . .... . ... . . ..... ... . . . ... .. . . . . . ... .. . . . . ... ... . ..... . . . .... . ... . . . . .. . . ... . .. ..... . .... ... . . . . . .. .. . . .. .. .... .. .. . ... .. . ... .. .. . . .. . . . ... . .. .. ... . . . .. ......4) (9.. . . .. . .. ..... . ... . .. . .. . .. . .. ... ... . .. .. . . . .. . .... . . . . . . ... . . . .. .. . . . .... .. .. .. ... . . ... . . .. .. .. . . . . . . .. . . .. .. . .... .. . . .. .... . . . .. . ...... ... ... . .. . . . . . . .. . ... . ... ... . . . . ... ........ ..... .. ......... . .. .. .... . .

y. ∂f df ∂u 1 −1 2 = × = u × 2 x = x ( x 2 + 2 y 2 )−1 2 ∂x du ∂x 2 ∂f df ∂u 1 −1 2 = × = u × 4 y = 2 y ( x 2 + 2 y 2 )−1 2 ∂y du ∂y 2 (iii) f(x. ∂f ∂v ∂u =u× +v × = y × 2 y cos( x 2 + y 2 ) + sin ( x 2 + y 2 ) × 1 ∂y ∂y ∂y = 2 y 2 cos( x 2 + y 2 ) + sin ( x 2 + y 2 ) 0 Exercises 3–7 ∂f = 2 xy cos( x 2 + y 2 ) ∂x ∂f = 4 y + 4x + 6z. Then. ∂y ∂f = 6 z + 5x + 6 y ∂z ∂f . the cubic function in two variables z1=1x 31+12x 2y1+13xy 21+14y 3 has partial ﬁrst derivatives ∂z = 3x 2 + 4 xy + 3 y 2 . by the product ∂y Higher derivatives Like the derivative of a function of one variable (Section 4. the partial derivative of a function of more than one variable can itself be differentiated if it satisﬁes the necessary conditions of continuity and smoothness. Then. To ﬁnd rule. y)1=1(x 21+12y 2)122 Let f1=1u122 where u1=1x 21+12y 2.3 Partial differentiation 251 EXAMPLES 9. y)1=1y1sin(x 21+1y 2) By the chain rule.2 Partial differentiation (i) f(x.9. For example. ∂x ∂z = 2 x 2 + 6 xy + 12 y 2 ∂y . by the chain rule.9). z)1=1x 21+12y 21+13z 21+14xy1+15xz1+16yz ∂f = 2 x + 4 y + 5z . ∂x (ii) f(x. let f1=1u1×1v where u1=1y and v1=1sin(x 21+1y 2).

for the cubic function. fxxy1=1fxyx1=1fyxx (subject to the relevant continuity conditions). for a function of three variables f(x. Differentiation of ∂z2∂x gives ∂ ∂z ∂2 z = 6 x + 4 y. and the following more compact notation is often used: fx = ∂f . for example. In some applications. and ∂2z2∂x 2 is the rate of change of this gradient as the point P moves along the curve.3. the two mixed second derivatives are identical. Alternative notations (9.6) becomes fxy1=1fyx and.252 Chapter 9 Functions of several variables and each of these can be differentiated with respect to either variable to give four partial second derivatives. on the curve APB. ∂x∂y f xyz = ∂3 f . and so on.8) . there are a possible 8 partial third derivatives. ∂x f xx = ∂2 f ∂x 2 . in general a possible 2n nth derivatives. This is achieved by adding the constant variables as subscripts to the ordinary symbol for the partial derivative. (9. the symbol ∂f ∂x y . In terms of Figure 9. We note that. and is therefore (very nearly) always true in practice: ∂2 z ∂2 z = ∂x∂y ∂y∂x Similar results are obtained for higher derivatives. particularly in thermodynamics.7) In this notation. the ‘mixed’ second derivative ∂2z2∂y∂z1=1∂(∂z2∂x)2∂y is the rate of change of the gradient ∂z2∂x (in the x-direction) as the point P moves along the curve DPE (in the perpendicular y-direction). z means the derivative of f with respect to x at constant y and z. … ∂x∂y∂z (9. For example. ∂y ∂y ∂ ∂x ∂z ∂2 z = 4x + 6 y = ∂y ∂x∂y ∂ ∂z ∂2 z = 4x + 6 y = ∂y ∂x ∂y∂x For a function of two variables.6) The above symbols for partial derivatives become unwieldy for the higher derivatives. P say. f xy = ∂2 f . = ∂x ∂x ∂x 2 and differentiation of ∂z2∂y gives ∂ ∂y ∂z ∂2 z = 2 = 6 x + 24 y . y. equation (9. z). On the other hand. the ﬁrst derivative ∂z2∂x is the gradient at a point. 16 fourth derivatives. This is true for functions whose ﬁrst derivatives are continuous. it is necessary to specify explicitly which variables (or combinations of variables) are to be kept constant.

b)1=10. ∂x y ∂2u ∂x 2 uy = ∂u x = − sin x sin y − 2 ∂y y ∂2u ∂y 2 uxx = = − sin x cos y . . ∂x∂y y u yx = ∂2u 1 = − cos x sin y − 2 ∂y∂x y and uxy1=1uyx (iii) For the ideal gas.9. that is. = = . its tangent line is ‘horizontal’.3 More partial derivatives (i) The nonzero partial derivatives of u1=1x1+1y 21+12y 3 are ∂u ∂u ∂2u ∂3u = 1. y)1=1(a. = 12 ∂x ∂y ∂y 2 ∂y 3 (ii) The ﬁrst and second partial derivatives of u1=1sin1x1cos1y1+1x2y are ux = ∂u 1 = cos x cos y + . the graph of the function has zero slope at the stationary point. u yy = = − sin x cos y + 2x y3 uxy = 1 ∂2u = − cos x sin y − 2 . = 2 y + 6 y2 .5).9) or fx(a.4 Stationary points We saw in Section 4. = 2 + 12 y . b) ∂x ∂y (9.10 that a function f(x) of one variable has a stationary value at point x1=1a if its derivative at that point is zero. A function f(x. these are sufﬁcient for all the ﬁrst derivatives of a continuous function to be zero at a point. The corresponding condition for a function of two variables is that the tangent plane be horizontal. b)1=1fy(a. n p 0 Exercises 8–20 9. Geometrically.4 Stationary points 253 EXAMPLES 9. b) if its partial ﬁrst derivatives are zero: ∂f ∂f = = 0 at ( a. n ∂n p. y) then has stationary point at (x. V1=1nRT2p and ∂V nRT ∂V nR ∂V RT =− 2 . T ∂T p. In view of equation (9. p p ∂p T . if f ′(a)1=10.

(−2. −423) 0 Exercises 21–23 For a function of one variable.10a) (9. ∂x and these are zero when x 21+12y 21=14. y)1=1x 31+16xy 21−12y 31−112x are (2. 0). f ″(a)1<10. The stationary points of the function f(x. y(2x1−1y)1=10 ∂f = 12 xy − 6 y 2 ∂y The second equation is satisﬁed when either y1=10 or y1=12x.10b) for either a maximum or a minimum (9.4. and substitution of these in the ﬁrst equation gives the four stationary points (2.254 Chapter 9 Functions of several variables EXAMPLE 9. The corresponding conditions for a function of two variables are fxx1<10 fxx1>10 and 2 fxx fyy1−1f xy1>10 and and fyy1<10 fyy1>10 for a maximum for a minimum (9. y)1=1x 31+16xy 21−12y 31−112x We have ∂f = 3x 2 + 6 y 2 − 12.5 The nature of the stationary points of Example 9. 423). 0). (223. a stationary point at x1=1a is a local maximum if the second derivative is negative.10c) 2 If the quantity ∆1=1fxx fyy1−1f xy is negative then the point is a saddle point. If ∆1=10 then further tests are required to determine the nature of the point. 423). EXAMPLE 9. a local minimum if f ″(a)1>10. 0). (223. 0). The corresponding conditions are more complicated for functions of more than two variables. (−223. (−2. and may be a point of inﬂection if f ″(a)1=10. −423) .4 Find the stationary points of the function f(x. (−223. a maximum in one direction and a minimum in another.

0 Exercises 27(i). In the absence of the constraint. F″(x)1=12 so that F(x) has minimum value for x1=1−4. Table 9. These constraints usually take the form of one or more relations amongst the variables. however.6 Find the extremum value of the function f(x. the function has a saddle point at x1=1y1=10. and no maxima or minima. The extremum point of f(x.4 the variables x and y are independent variables. EXAMPLE 9. In the present case.9.1 x 2 −2 2 3 − 2 3 − y 0 0 4 3 4 3 fxx 12 −12 4 −4 fyy 24 −24 −8 8 fxy 0 0 16 −16 2 fxx fyy1−1f xy nature minimum maximum saddle point saddle point >10 >10 <10 <10 0 Exercises 24–26 Optimization with constraints The ﬁnding of the maxima and minima (extremum values) of a function is called optimization. y)1=13x 21−12y 2 subject to the constraint x1+1y1=12. with no constraints on their values. and the extremum value of the function is f(−4. 28(i) . y)1=1F(x)1=13x 21−12(21−1x)21=1x 21+18x1−18 Then. F′(x)1=12x1+181=10. the search for a stationary value is restricted to the line y1=121−1x. the optimization may be subject to one or more constraints. y) is therefore a minimum point at (x. for a stationary value. Thus. y)1=1(−4. substituting y1=121−1x in the function gives f(x. 6)1=1−24. In many applications in the physical sciences. 6).4 Stationary points 255 The determination of the nature of these points is summarized in the following table. In Example 9. The constraint is a relation between the variables x and y that reduces the number of independent variables to 1. we have constrained optimization.

so that the three equations to be solved are ∂ ( f − λ g ) = 6 x − λ = 0. xn) subject to m (<1n) constraint relations gk(x1. ∂x ∂x ∂f ∂g −λ = 0. x2. xn)1=1ak . and the constraint relation g(x. x3. z). k1=11. particularly for functions of more than two variables or when there are several constraint relations. ∂y g=x+ y=2 The solution is x1=1−4. Napoleon Bonaparte). Lagrange invented the name ‘derived function’ (hence ‘derivative’) and notation f ′(x) for the derivative of f(x).1=. These three equations and the constraint relation are sufﬁcient to determine the values of the stationary points and of the corresponding values of the Lagrange multiplier λ. 6)1=1−24 as in Example 9. however.256 Chapter 9 Functions of several variables The optimization problem in this example has been simpliﬁed by using the constraint to eliminate one of the variables. He emphasized that problems in mechanics can generally be solved by reducing them to differential equations.1=. 0 Exercise 27(ii). m (9. and f(−4. and has been called the greatest mathematician of the 18th century (‘Lagrange is the lofty pyramid of the mathematical sciences’.12a) 3 Joseph-Louis Lagrange (1736–1813). ∂y ∂y ∂f ∂g −λ =0 ∂z ∂z (9. such a simpliﬁcation is either difﬁcult or impossible. 3. 28(ii) In the general case.11) at the stationary points. ∂x ∂ ( f − λ g ) = − 4 y − λ = 0. made important contributions to many branches of mathematics.12b) (9. . A general procedure for solving many of the constrained optimization problems in the physical sciences is the method of Lagrange multipliers.6.6).1=. y. By the method of Lagrangian multipliers there exists a number λ such that ∂f ∂g −λ = 0. y1=16. f(x. the optimization is that of a function of n variables. 2. z)1=1contant. We have g1=1x1+1y and f1−1λg1=13x 21−12y 21−1λ (x1+1y). y. In general.3 We consider a function of three variables.7 Use the method of Lagrange multipliers to ﬁnd the extremum value of the function f1=13x 21−12y 2 subject to the constraint x1+1y1=12 (Example 9. EXAMPLE 9. born in Turin. f(x1. x3. x2. His greatest achievement was the development of the calculus of variations. and in his Mécanique analytique of 1788 he extended the mechanics of Newton and Euler. λ1=1−24.

Another application. xn. the second by y. The procedure is to construct the auxiliary function φ = f − λ1 g1 − λ 2 g2 − λ3 g3 − − λm gm = f − ∑ λk gk k =1 m (9. and z is V1=1xyz and its surface area is A1=12(xy1+1yz1+1zx). x1. The following examples demonstrate one application of the method of Lagrangian multipliers in geometry (or packaging) and one important application in chemistry. m g k = ak . x2 . 3.4 Stationary points 257 where a1.1=. = xz − 2λ ( x + z ) = 0. y.9 Secular equations Variation principles in the physical sciences often lead to the problem of ﬁnding the stationary values of a ‘quadratic form’ in the n variables. 3. and the third by z gives xyz1−12λ (xy1+1xz)1=1xyz1−12λ (xy1+1yz)1=1xyz1−12λ (xz1+1yz)1=10 It follows that x1=1y1=1z.14) k = 1.1=. 2. x2.9. …. ….13) and to solve the n1+1m equations m ∂g ∂φ ∂f = − ∑ λ k k = 0. By the method of Lagrangian multipliers we form the auxiliary function φ1=1V1−1λ A. and the box is a cube of side 0 Exercise 29 EXAMPLE 9. n (9. ∂x ∂y ∂φ = xy − 2λ ( x + y ) = 0 ∂z Multiplication of the ﬁrst equation by x. f ( x1 .8 Find the dimensions of the rectangular box of largest volume for given surface area. ∂xi ∂xi k =1 ∂xi i = 1.1am are constants. EXAMPLE 9. xn ) = ∑ ∑ Cij xi x j i =1 j =1 n n A 6. for the n variables xi and the m multipliers λk. is discussed in Chapter 21. . the derivation of the Boltzmann distribution. ….1a2. The volume of a box of sides x. and solve the set of equations ∂φ ∂φ = yz − 2λ ( y + z ) = 0. The problem is therefore to ﬁnd the maximum value of V subject to the constraint A1=1constant. 2.

2 3 By the method of Lagrange multipliers the stationary values of the function are obtained by forming the auxiliary function φ1=1f1−1λg and solving equations (9. ∆z is the change of ‘height’ on the surface. x2. x3)1=1C11x1 1+12C12 x1x21+12C13 x1x31+1C22 x21+12C23 x2 x31+1C33 x 2 2 3 2 with constraint g1=1x1 1+1x21+1x21=11. y1+1∆y)1−1f(x. xn provide a representation of the state (of an orbital in molecular orbital theory). x2.258 Chapter 9 Functions of several variables subject to the constraint g ( x1 . x2. y) be a function of the variables x and y.1=.5. when the quadratic form represents the energy of the system (or an orbital energy in molecular orbital theory) and the numbers x1. For the case n1=13. y). x2 . and let the values of the variables change continuously from (x. y) and is shown in the ﬁgure as the displacement P to Q on the representative surface of the function. to (x1+1∆x. The signiﬁcance and solution of such systems of equations are discussed in Chapters 17 and 19. y1+1∆y) at point q. for example. in the ‘method of linear combinations’ in quantum chemistry. 0 Exercise 30 9. and x3 and setting each derivative to zero then gives the set of simultaneous equations (C111−1λ)x11+1 C21x11 C31x11 C12 x2 +1 C13x31 C23x31 =10 =10 +1(C221−1λ)x21+1 +1 C32 x21 +1(C331−1λ)x31=10 Equations of this kind are often called secular equations. 2 f (x1. at point p in Figure 9.5 The total differential Let z1=1f(x. . …. They occur. we have 2 φ1=1(C111−1λ)x1 1+12C12 x1x21+12C13 x1x31+1(C221−1λ)x 21+12C23 x2 x31+1(C331−1λ)x 2 2 3 Differentiation with respect to x1. For example. The corresponding change in the function is ∆z1=1zq1−1zp1=1f(x1+1∆x. for n1=13. xn ) = ∑ xi2 = 1 i =1 n where the Cij are constants (with Cij1=1Cji).14).

.... ... .. . . ... . ..... . ..... . . . . .. .. . .... ... . ... .. . . . . ... ... . . ..... .... ..... .... .. . . . .. . .This expression is exact for the present case. . . . ... . .. . ... 2 ∂x 2 Figure 9... . .. .. .. ... . . .. . .... .5 ∂2 z 1 ∂2 z 1 ∂2 z ( ∆ y )2 ( ∆ x ) (∆ y ) + + 2 ( ∆ x )2 + ∆ ∂x∂y 2 ∂x 2 ∂y 2 ∂2 z = b. .... .. .. . . . . . .. .. . . . . .. . . . .. . ... . . . . . . . . ... . ... . ...... ... .. .... .. .. .. . . .. . ... . .. .. .. .. . . . . .. .. ... .. .. .. . .... . . . . . .. .. ... .. .. .. .. ... . . ...... .... For a general function f(x.. . .. . . . . ..... . .. .. .... . . . ... . ... . . . . . .. .. . .. . . . .. ... . . ... . .. . . .... . .... .. . .. .. ... . . .. . .. ..... ... . . .. .. . . . . .. . . .... . . .. . .. . . .. ... ... . .. ... ....... .. .. .. . .. . .. .. . .. .. .. ... ... . . .. . . . . .... .. .5 The total differential (9. .... .... .. .15) 259 . . . . . .. ... .. .. . ... . .. . . . . .. . .... . .. ... .. . . . .. . . .. . . ... ... . . .. ..... . . .. . . ..... ... . y1+1∆y) about the point (x. ... .... .... . . .. . . . ... . . .. . ... . .... ... . .. .. . . . ... .. . . .. . ... . ... ... ... . . . . . . .... . . . . . ......... ..... .... .. . . . . . . . ... y).. . . . .. . . . . .... . .. . . ... . . .. .. .. .. . . ... .. ... . ..... . ... . . ....... ... . . .... .. . . .. . .. . . ..... . . . . . . . . ... ... . .. .. . .... .. . .... .. . .. . ... ... . . .. ... . . . ... . . . . ... . . . . ... . . . . . . . . . ... . . . ... . .. . ... .. .. . . .. . ... .. .. . . .. .. .. ..... ...... y) (compare with equation (7. ... .. .. .. . .. . . .. . . for the quadratic function ax 21+1bxy1+1cy 2... . . ... . . .... . .. . . .. . . . . . (9. .. . ... ... .. . . ... . . .. . .. .... . ... .. . . . ..... . . . .. . ..... . .. .. . . ... .. ... ... .... . .. .. .. .. the change in the function can be written as x y x •r •p pz o •R P• z Now. . .. .. ... . . . . . . ..... . .. . . .. . . . . . . . .... .... . . ..... .. . .. .. .... . . . . . . . . . .. ... ..15) represent the ﬁrst few terms of a Taylor expansion of the function at point (x1+1∆x. . .. ... . . .. ..... ... . . . . ... .. . .... . .. . . . y1+1∆y)1=1a(x1+1∆x)21+1b(x1+1∆x)( y1+1∆y)1+1c( y1+1∆y)2 zp1=1f(x..... . . because and For example. . ... ... . . .. .. . .. . .. .. .. .... ... .. ... . . .... . ... q • y qz Q • .. . .... . . . .. .. . . and for all functions whose third and higher derivatives are zero.. . .. . ..... .... .. .. .. .. .. ∆ . ... ........ .. ∆ ... . .... . . ..... . ..24) for a function of one variable). .. ...... ... .... .. . . .... . . .. .. ∂x 1 ∂2 z = a. .. ... . . ... . . . . . ... . . .. . . .. ... . .. . . . . . . . .. . .. . . . . . . ... . . .. .... . . .. . ... ... . . . .. . . .. . ... .. .. . . .. . .. ... ...... .. . . ... . . .... .. .... .... . . . . . .. . . . .. .. . . . ... . . .. . ... .. .. . . . . . .. .. .. y)1=1ax 21+1bxy1+1cy 2 ∂z ∂z ∆z = ∆x + ∆ y ∂x ∂y ∂z = 2 ax + by .. .. . .. .. .... .. . . .. .. .. .. . ... . . ..... ... . . . .... .. .... . .. . . .. ∂x∂y ∂z = bx + 2cy ∂y 1 ∂2 z =c 2 ∂y 2 9. .. . ...... .. . . . .. . . . . . . . ... .. .. . ..... ... . . .. . .. . . ∆z1=1zq1−1zp1=1(2ax1+1by)∆x1+1(bx1+12cy)∆y1+1a(∆x)21+1b(∆x)(∆y)1+1c(∆y)2 zq1=1f(x1+1∆x. ... .... ..... .. . . . .... .. . . ... . . .. . .. . .. . .. .. ... ..... . . . .. . .... .... .. . .. .. . . . ... . . .. . .... . . . ...... . . . ..... ..

The nephew of Johann (I) and Jakob (John and James. . or Jean et Jacques). for a function of n variables. nor with Nicolaus (III).1=. xn. and an approximate value of ∆z is ∂z ∂z ∆z ≈ ∆x + ∆y ∂x y ∂y x (9.4 The concept of the total differential is readily generalized for functions of any number of variables. the son of Johann (I) and brother of Daniel and Johann (II). z1=1f(x1. shows that when the changes ∆x and ∆y are small enough the total change in z is approximately equal to the change in z due to change ∆x alone (the ﬁrst term of the expression on the right of (9. ∂z2∂x1 is the partial derivative with respect to variable x1 with all other variables. kept constant. x3.16)) plus the change in z due to change ∆y alone (the second term). the accuracy of the expression improves as ∆x and ∆y approach zero.10 Find the total differential: (i) z= x y ∂z x =− 2.18) where. we consider inﬁnitesimal changes dx and dy. valid for all continuous functions of two variables. the terms quadratic in ∆ are small compared with the linear terms.17) as the limiting case of (9. his father. xn) the total differential is ∂z ∂z dz = ∂x dx1 + ∂x dx2 + 1 2 n ∂z ∂z dxn = ∑ dxi + ∂x ∂x n i =1 i (9. for example.260 Chapter 9 Functions of several variables If ∆x and ∆y are small enough. and deﬁne the quantity ∂z ∂z dz = dx + dy ∂x y ∂y x (9.12. It is Daniel Bernoulli (1700–1782) who wrote the Hydrodynamica of 1738 which contains the concept of ‘Bernoulli’s Theorem’ and a development of the kinetic theory of gases. As in Section 4.16). x2. In addition.1=. he is not to be confused with Nicolaus (I). ∂y y ∂z ∂z x 1 dz = dx + dy = dx − 2 dy y ∂x y ∂y x y ∂z 1 = . x2.16) This result. This quantity is called the total differential of z with respect to x and y. ∂x y 4 The total differential and the equality of the mixed second derivatives were discovered in 1719 by Nicolaus (II) Bernoulli (1687–1759). EXAMPLES 9. x3.

= − r sin θ sin φ ∂φ ∂r ∂θ ∂x ∂x ∂x dx = dr + d θ + d φ ∂φ r . ∂y u ∂u z = ∂z u ∂u ∂u ∂u 1 dz = ( x dx + y dy + z dz ) du = dx + dy + u ∂z x . T ∂T p. θ ∂r θ . φ = sin θ cos φ dr + r cos θ cos φ d θ − r sin θ sin φ d φ 0 Exercises 31–35 EXAMPLE 9. n 1 ∂V the isothermal compressibility V ∂p T .5 The total differential 261 (ii) u1=1(x 21+1y 21+1z 2)122 ∂u x = x ( x 2 + y 2 + z 2 )−1 2 = . ∂x u ∂u y = . n ∂p T . x (iii) x1=1r1sin1θ1cos1φ ∂x ∂x ∂x = sin θ cos φ . temperature T. y ∂x y .11 Differential volume The volume of a one-component thermodynamic system is a function of pressure p.T 0 Exercise 36 . and amount of substance n: V1=1V( p. φ ∂θ r . n = αV dT − κV dp + Vm dn where α= 1 ∂V the thermal expansivity (coefﬁcient of thermal expansion) V ∂T p. T. The total differential volume is ∂V ∂V ∂V dn dV = dT + dp + ∂n p.9. n). n the molar volume κ =− ∂V Vm = ∂n p. = r cos θ cos φ . z ∂y z .

y) and differentiating the resulting function of t. dz ∂z dx ∂z dy + = dt ∂x y dt ∂y x dt (9.6 Some differential properties The total derivative In the function of two variables. This derivative is called the total derivative of z with respect to t. In the following sections we use (9. Thus. x3.17) to derive a number of differential and integral properties of functions.17) by (inﬁnitesimal) dt. For a function of n variables. Equation (9. x2. u1=1f(x1.19) Then z1=1f(x(t). y1=1y(t) (9. by dividing the expression (9.16) by ∆t and taking the limit ∆t1→10.21) is a generalization of the chain rule (see Section 4.22) .20) Alternatively. y). in which the variables are all functions of t. let x and y be functions of a third variable t: x1=1x(t). xn). z1=1f(x. du ∂u dx1 ∂u dx2 = + + dt ∂x1 dt ∂x2 dt ∂u dx = ∑ i i = 1 ∂xi dt n ∂u dx n + ∂x dt n (9.1=.16) and its limiting form (9.6). the same result is obtained by dividing the total differential (9. and there exists an ordinary derivative dz2dt. and can be obtained directly by substituting the functions x(t) and y(t) for the variables in f(x. y(t)) is essentially a function of the single variable t.22 and 9. in the limit ∆t1→10.21) (9. 9.27).262 Chapter 9 Functions of several variables One of the principal uses of the total differential in the physical sciences is in the formulation of the laws of thermodynamics (see Examples 9. It can also be obtained indirectly. division of ∂z ∂z ∆z ≈ ∆x + ∆ y ∂x y ∂y x by ∆t gives ∆ z ∂z ∆ x ∂z ∆ y ≈ + ∆ t ∂x y ∆ t ∂y x ∆ t and.

Then. ﬁnd dz2dt. y). with t replaced by θ. by equation (9. such equations are called the parametric equations of a curve. In general. where x1=1e t and y1=1e−t.12 Given z1=1x 21+1y 3. Let z1=1f(x. and = x2 − y2 ∂x ∂y dθ .19). since dx2dθ1=1−y and dy2dθ1=1x. 0 Exercises 37–39 EXAMPLE 9. A displacement on the circle is most easily described when the equation of the circle is expressed in terms of the polar coordinates r and θ : x1=1a1cos1θ. we have ∂z = 2 x.21).21). dz = 2e2t − 3e−3t = 2 x 2 − 3 y 3 dt (ii) By the chain rule (9.6 Some differential properties 263 EXAMPLE 9. y1=1a1sin1θ ∂z = 3y2 ∂y and dx = et = x. (i) By substitution: z = e2t + e−3t . dt dy = − e− t = − y dt These have the form of the pair of equations (9. ∂x Therefore. dz ∂z dx ∂z dy = + = ( 2 x ) × ( x ) + (3 y 2 ) × ( − y ) = 2 x 2 − 3 y 3 dt ∂x dt ∂y dt and this is identical to the result obtained by substitution. ∂z ∂z dz ∂z dx ∂z dy = + = −y + x d θ ∂x y d θ ∂y x d θ x ∂x y ∂y x For example.9.13 Walking on a circle The equation of a circle in the xy-plane with centre at the origin and radius a is x 21+1y 21=1a 2. if z1=1xy then 0 Exercise 40 ∂z ∂z dz = y. = x .

∂z dy dz ∂z = + dx ∂x y ∂y x dx (9.21). Then.12. 0 Exercise 41 Walking on a contour Consider changes in x and y that leave the value of the function z1=1f(x. the plane ABC is parallel to the xy-plane. The displacement P to Q therefore lies on a contour of the surface. since x1=112y.6.25) . In Figure 9. ﬁnd dz2dx. where y1=112x. from (9.23). ∂z ∂z ∆ z = 0 ≈ ∆x + ∆y ∂x y ∂y x (9.21) with respect to t by the chain rule: dz dz dx ∂z dx ∂z dy dx = = + dt dx dt ∂x y dt ∂y x dx dt (9. y) unchanged.21) is obtained if t is replaced by x. (i) By equation (9. y(x)) can be treated as a function of the single variable x.14 (i) Given z1=1x 21+1y 3. and dt dz dz dx = = 2 x 2 − 3xy 4 dt dx dt This is identical to the result of Example 9.23) We note that if x1=1x(t) then this total derivative with respect to x is related to the total derivative (9. Then.16).264 Chapter 9 Functions of several variables A special case of the total derivative (9. (ii) if x1=1et. Then y1=1y(x) is an explicit function of x and z1=1f(x. by (9. so that all points on the curve APB on the representative surface are at constant value of z. since dy 1 = − 2 = − y2.24) EXAMPLE 9. Then. ﬁnd dz2dt. dx x dz ∂z ∂z dy = + = 2 x + (3 y 2 ) × ( − y 2 ) = 2 x − 3 y 4 dx ∂x ∂y dx (ii) If x1=1et then dx = et = x .

. .. . . .... . ... ..... .. .... ... .. .. . ..... ....... ... . .. . ... . .... . . . . . .. ∆ .. ...... . . . . .. .. after division by ∆x.. .. . .. .. .... .. ... . .... ... .. . . . ..... . .. ..... . . . . . .... . .. .. . . ... . ... .. .... . ... ..... because In the limit ∆x1→10... . . ..... ... . . . . .. ... ..... ... . .. ......... . . . .. .. ..... .. . . . . .......... .. ... ... . . .... .. . .. . ..... . . ... .. . . .. .... .. .. .. . .... .. .. ... . .. ... .. . .. .... and x • y • y b • q p o ..... .. . .... .... . ... .... .. ...... .. ... . .. . . ........ . . .... . .... .. ...... ... .. . . .. .. . .. .. .......... .. . . ... .. ........ . . . . ........ .. ... .. . . . . . ..... .. . . .. . . ..... . .. . ........ .. ... . . .. . ... . . .. . . .... .. . . . . ......... . .. ... ..... ....... . ........17) by considering inﬁnitesimal changes dx and dy such that dz1=10: To obtain (9. . . .. ..... . The same result is obtained directly from the total differential (9.. ... ... .. . .. .. . . .... . .. ........ . ... . . . . .. .. . .26) (9. .. .... .. ... . . . .. ...... . ... . . ... ... . .. . . . .. .... .. .. . . ... . . .. ... .. .. ... . . . . .. . . . .. . ..... .. .... .. .. ... . .. .... . . . .. .. ... . ... ....... .. ... . . .. . ........ ... . .. ... ... ..... . . ... . .... . .... .. . ......... .. .... . . . .. . ...... ... . .. . ... . . .... .. .. ... . . . . . . . .. .. .. .. .. .. .. . . .... .. ..26) we divide (formally) by dx and.... . . ... . . ...... ..... .... .... ..... .... .. ..... ... . ....... ... ... .. .... .. . .... . . ..... . .. ..... .... .. . .. . .. . .... . ... .... ... .. . ... ... ... . . . ............ ... ..... . ...... ...... . ... ... . . .. ......... . . .. . ... . .... . .... .. . ..... ..... .. . ... ..... . .. . .. ..... . .. . .. ... . ..... .. . .. ........ . .. . . ... .... .. . . . .. ... .... . . ... ...... .. . .. ... . .. ......... . .... .. . B Q • P• C z and.. .. ..... . ... .... . . .. ...... ..... .. . ..... . . .. .. ... ... . . . ...... ... . ... . . .. ........ . . .. . ... ..... because the process is at constant z... ... . .6 ∂z ∂z ∆ y 0≈ + ∂x y ∂y x ∆ x ∂z ∂y x 9... .. .. . ... . . . ... . . .. .. .. ... .... .. ... . .. . ... .. ..... ..... . . . ... . . .... . . . . .... . . . ... .. ... ....... ... .. . . .. .. . ....... . . .. . ... .. ... . . . . ... .. . . .. . .... . . .. . . .. ... . . . ...... ... . .. .. .... ... ... .... . . . . . .... .. .. . ... .. ....... ...... . . ... . .......... . . Equation (9..26) can be rearranged as ∂z ∂z ∂y 0= + ∂x y ∂y x ∂x z ∂z ∂z dz = 0 = dx + dy ∂x y ∂y x ∂y ∂z = − ∂x z ∂x y Figure 9.. . ... that is. . . . .. . . .. . .... . .... ........ .. . .... ... . ... .. .. . .... . .. . . .. . . ... ... . ..... .. ..... . .. . ... .. . . . . ..... . . .. ....... .... . . . ... . .. .. .. . .. .. . .. . . . . .. ..... .... ..... . .. . ... ..... . ..... ... .. .... .... . . ... .. ....... . .. . .. . . .. .. .. . ..... . . .. ∆ . ... ....... ........ .. .. ... . . . . .. . .... .. . ... . .. ... . .. . .... . . ... .... .. . .... ... In addition... . . . .. .. .. ..... . .. ... .... . . .... ... ... .. ..... . ..... ..... ..... .. . . . ... . ..... .. . .... .. . . .... .. . .. ... ... ..and this is the gradient of the graph of the function y(x) for which z is constant. .......... .... .. ... . ... . . ... ... ... . .... . . .. . . .. . . .. ..... .. ... . ..... .. ... . . . ... . . . ... . . . . ... .. .. ... .. ... . ... .. .. .. .... .. .... ........ .... . ... .. ... . .. the gradient of the curve ab in the xy-plane in Figure 9. . . . .. . . .. . . . .. .. . .... . . ... .. . ... .. . ..... . . ... .. . . .. . .. ...... ... .. . .. ... ..... ... ...... . . . . . . . . . . .... . ......27) 265 x a A .. ... . ... . .. we replace dy2dx by the partial derivative at constant z.. .. . ......... .. ....... . .. ...... .. . ... .. ........ ... .. . .. . .. ...6 for which AB is a contour.. .. .. ........ ..... . .... ...... . ... . . . ....... .. . . . . ....... ... . .... .. ........ the ratio ∆y2∆x approaches the derivative of y with respect to x at constant z.. .. ........ . .......6 Some differential properties (9... . . ... .. . . .... .. . ... ...... .....

∂x z ∂z 3 = 2 xy . EXAMPLES 9. ∂x r ∂y x ∂r x ∂r y = . We note also that whereas (9.27) is positive.29).26) to (9. any one variable can be regarded as a function of the other two. ﬁnd .6 the slopes along the x and y directions are both negative at P (the gradient lines slope ‘downward’).26) is readily generalized for sets of more than three variables.266 Chapter 9 Functions of several variables ∂z =1 ∂x y the equation can also be written as ∂x ∂z y (9. However. ∆y is necessarily positive but ∆x negative (or vice versa for motion Q to P) so that (∂y2∂x)z is negative.27).3 and 9.27) and (9.29) This form is sometimes called the ‘−1 rule’.27) and (9.28) ∂x ∂y ∂z −1 = ∂y z ∂z x ∂x y (9. ∂y =− ∂x z ∂z ∂x y ∂z 2y 2 xy 3 =− 2 2 =− 3x ∂y x 3x y ∂z 2 2 = 3x y ∂y x (ii) For r = x 2 + y 2 ( ) 1/ 2 ∂y . and = − ∂x r ∂y r ∂x r y . ﬁnd . so that (∂z2∂x)y and (∂z2∂y)x are negative and their quotient in equation (9. for the particular z-surface shown in Figures 9. Thus.15 ∂y (i) Given z1=1x 2y 3. but it can be explained geometrically. ∂x y By equation (9. Hence the – sign.29) sometimes causes unease.29) are true for only three variables at a time (all others being kept constant). Figure 9. We note that the distinction between dependent and independent variables has disappeared from equations (9. Similar considerations apply to the other three possible pairs of signs of the slopes. The presence of the – sign in equations (9. = .6 shows that for motion along the contour from P to Q. equations (9.

∂z ∂z x ∂z = − ∂x u ∂x y y ∂y x . y) be a function of the variables x and y. y) be some other function of x and y. y) and u1=1x 21+1y 2. by equation (9. t ∂T p ∂p T ∂V ∂p ∂T By equation (9.11) ∂V ∂V dV = dT + dp (for ﬁxed amount n). by equation (9. division of the total differential dz by inﬁnitesimal change dx at constant u gives ∂z ∂z ∂z ∂y = + ∂x u ∂x y ∂y x ∂x u (9. (iii) The differential volume of a thermodynamic system is (see Example 9.30). y) of x and y is kept constant.30) This equation shows how the partial derivative with respect to x at constant y is related to the partial derivative with respect to x when some function u(x.9. 43 ∂V α expansivity n = = ∂p T κ compressibility Change of constant variable Let z1=1f(x.6 Some differential properties 267 The representative surface of the function r1=1(x21+1y2)122 is a vertical cone whose contours (r1=1constant) are circles of radius r parallel to the xy-plane. Consider changes in the variables such that u is constant.29). EXAMPLE 9.16 Let z1=1f(x.27). and let u1=1g(x. −1 = ∂p T ∂T V ∂V p so that ∂p ∂V =− ∂T p ∂T V 0 Exercises 42. Then. The quantity −x2y is the gradient at (x. y) on the circle. Then. therefore. ∂y ∂u =− ∂x u ∂x y ∂u x =− y ∂y x and.

1v) (9.33a) (9.17) by du at constant v to give equation (9.33b) . y) be a function of the independent variables x and y with total differential of z with respect to x and y. Thus.17) Let the variables x and y be themselves functions of two other independent variables. Divide the total differential (9. and its total differential with respect to the new variables is ∂z ∂z dz = du + d v ∂v u ∂u v (9.32) and those in (9. and by dv at constant u to give (9.33b): ∂z ∂z ∂x ∂z ∂y = + ∂u v ∂x y ∂u v ∂y x ∂u v ∂z ∂z ∂x ∂z ∂y = + ∂v u ∂x y ∂v u ∂y x ∂v u (9. say). if z1=1xy then = y − x 2 y and dθ ∂x u 0 Exercises 44–46 Change of independent variables Let z1=1f(x. when u is constant (=1a 2.31) Then z can be treated as a function of u and v.33a).17) can be obtained in the following way.32) The relationships between the partial derivatives in (9. ∂z dz ∂z dx = = − y d θ ∂x u d θ ∂x u ∂z dz = x2 − y2 For example. y1=1y(u. and displacements at constant u are therefore constrained to the circle.268 Chapter 9 Functions of several variables This expression is closely related to that in Example 9.1v). Then. u and v: x1=1x(u. u1=1x 21+1y 21=1a 2 is the equation of a circle of radius a. with x1=1a1cos1θ and dx2dθ1=1−y. ∂z ∂z dz = dx + dy ∂x y ∂y x (9.13 (walking on a circle) for dz2dθ.

where x1=1x(r.9. 48 Laplace’s equation in two dimensions The Laplace equation in two dimensions is ∂2 f ∂x 2 + ∂2 f ∂y 2 =0 (9.17 From cartesian to polar coordinates Let z1=1f(x.36) ∂z ∂z ∂r ∂z ∂θ ∂z cos θ ∂z = + = sin θ + r ∂θ r ∂y x ∂r θ ∂y x ∂θ r ∂y x ∂r θ 0 Exercises 47.34b) EXAMPLE 9. The inverse relationships are.34). ∂z ∂z ∂x ∂z ∂z ∂z ∂y = + = cos θ + sin θ ∂r θ ∂x y ∂r θ ∂y x ∂r θ ∂x y ∂y x (9.35) ∂z ∂y ∂z ∂x ∂z ∂z = + = −y + ∂θ r ∂x y ∂θ r ∂y x ∂θ r ∂x y ∂z x ∂y x The ﬁrst of these is identical to equation (9. by equations (9. θ )1=1r1sin1θ. Then.13 for motion around a circle. The position of the point is speciﬁed equally well in terms of the polar coordinates r and θ.32): ∂z ∂z ∂u ∂z ∂v = + ∂x y ∂u v ∂x y ∂v u ∂x y ∂z ∂z ∂u ∂z ∂v = + ∂y x ∂u v ∂y x ∂v u ∂y x (9. ∂z ∂z ∂r ∂z ∂z ∂θ sin θ ∂z = + = cos θ − r ∂θ r ∂x y ∂r θ ∂x y ∂θ r ∂x y ∂r θ (9.33). y) be a function of the cartesian coordinates of a point in the xy-plane. θ )1=1r1cos1θ and y1=1y(r.34a) (9.5) and has the same graphical interpretation.37) . replacing u by r and v by θ in equations (9.6 Some differential properties 269 The inverse relationships are obtained in the same way from (9. The second is identical to the result obtained in Example 9.

His Traité de mécanique céleste (Treatise on celestial mechanics) in 5 volumes (1799–1825) marked the culmination of the Newtonian view of gravitation. The Laplace equation occurs in many branches of the physical sciences. where he taught elementary mathematics to the cadets. The equation in two dimensions is important in ﬂow theories. Example 9. and is the fundamental equation in ‘potential theory’. for example. f1=1f(r. the position of a point in a plane can be speciﬁed in terms of the polar coordinates r and θ.18 shows how this is done.39) EXAMPLE 9. the gravitational and electrostatic potential functions in a region free of matter satisfy the Laplace equation in three dimensions (see Chapter 10). (9. ∂f sin θ ∂f ∂f = cos θ − ∂x ∂r r ∂θ 5 Pierre Simon de Laplace (1749–1827). and passed. Napoleon in 1785. with result ∂2 f ∂x 2 The differential operator ∇2 = ∂2 ∂x 2 ∂2 + ∂2 ∂y 2 (9. for example in the theory of ﬂuid ﬂow and of heat conduction.18 The two-dimensional Laplacian in polar coordinates By the ﬁrst of equations (9. when a physical system is described in terms of a potential function. Legend has it that whilst at the École Militaire. where x1=1r1cos1θ and y1=1r1sin1θ. .36) in Example 9. he examined. see Chapter 10). The Laplace equation is then ∇2 f1=10 and a solution f of the equation as known as a harmonic function.17.5 As in Example 9.37) can be transformed from an equation in cartesian coordinates to an equation in polar coordinates by the method described in Example 9.17.38) + ∂2 f ∂y 2 = ∂2 f ∂r 2 + 1 ∂f 1 ∂2 f + 2 =0 r ∂r r ∂θ 2 1 ∂ 1 ∂2 = 2 + + 2 r ∂r r ∂θ 2 ∂r (read as ‘del-squared’ or ‘nabla-squared’) is called the Laplacian operator (although the symbol and the name are usually reserved for the three-dimensional form. The function f can therefore be treated as a function of r and θ. and equation (9.17. θ ). y) is a function of the cartesian coordinates of a point in a plane.270 Chapter 9 Functions of several variables where f1=1f(x.

∂θ ∂2 f ∂r ∂θ 2 = 2 cos 2θ = 2f r2 ∂2 f 2 = − 4r 2 cos 2θ = − 4 f . ∂r r ∂f = −2r 2 sin 2θ . ∂x Therefore. ∂2 f ∂x 2 ∂2 f ∂x 2 + =2 and ∂f = −2 y . 2f ∂f = 2r cos 2θ = . ∂f = 2 x. f1=1r 2(cos 2θ1−1sin2θ)1=1r 21cos12θ. ∂2 f ∂y Therefore ∂2 f ∂x 2 0 Exercise 49 + ∂2 f ∂y 2 = 2 = sin 2 θ ∂2 f ∂r 2 + cos2 θ ∂f cos 2 θ ∂2 f 2 sin θ cos θ + − r ∂r r r 2 ∂θ 2 1 ∂f ∂2 f − r ∂θ ∂r ∂θ ∂2 f ∂r 2 + 1 ∂f 1 ∂2 f + 2 r ∂r r ∂θ 2 EXAMPLE 9. ∂y ∂2 f ∂y 2 = −2 ∂2 f ∂y 2 =0 (ii) In polar coordinates.19 Show that the function f1=1x 21−1y 2 satisﬁes the Laplace equation (i) In cartesian coordinates.9.6 Some differential properties 271 Then ∂2 f ∂ sin θ ∂ ∂f sin θ ∂f = cos θ − − cos θ ∂r r ∂θ r ∂θ ∂r ∂x 2 = cos θ ∂f sin θ ∂f ∂ ∂f sin θ ∂f sin θ ∂ − − cos θ − cos θ ∂r r ∂θ r ∂θ r ∂θ ∂r ∂r ∂2 f sin θ ∂f sin θ ∂2 f = cos θ cos θ 2 + 2 − r ∂r ∂θ r ∂θ ∂r − sin θ r ∂2 f ∂r 2 ∂f ∂2 f cos θ ∂f sin θ ∂2 f − sin θ + cos θ − − r ∂θ r ∂θ 2 ∂r ∂θ ∂r + sin 2 θ ∂f sin 2 θ ∂2 f 2 sin θ cos θ + + r ∂r r r 2 ∂θ 2 1 ∂f ∂2 f − r ∂θ ∂r ∂θ = cos2 θ Similarly.

and T are the pressure. This function occurs in potential theory in two dimensions. and p.40) and (9. volume.20 Show that the function f1=1ln1r. ∂f2∂θ1=10 and the derivatives with respect to r are total derivatives: ∇2 f (r ) = d2 f dr 2 + 1 df r dr Now df2dr1=112r and d 2f2dr 21=1−12r 2. ∂2 f ∂r 2 + 1 ∂f 1 ∂2 f 2 f 2 f 4 f + 2 = 2 + 2 − 2 =0 r ∂r r ∂θ 2 r r r 0 Exercises 50–52 EXAMPLE 9. the equation applies to each separate phase. More generally. In general.40) where U is the internal energy of a thermodynamic system.41) The expression on the right side of (9.272 Chapter 9 Functions of several variables Therefore. Therefore ∇2f1=10.40) is called an exact differential. 9. and temperature. combining both the ﬁrst and second laws. . with additional terms if the amounts of substance are not constant. S is its entropy. V ) as a function of S and V. This example is important because it can be shown that the only solution of the Laplace equation in two dimensions that depends on r = x 2 + y 2 alone has the general form f1=1a1ln1r1+1b. a differential F(x. equating (9.43) * For a single closed phase with constant composition.7 Exact differentials One of the fundamental equations of thermodynamics.42) (9. satisﬁes the Laplace equation. y) dy (9. It can therefore be written as ∂U ∂U dU = dS + dV ∂S V ∂V S so that. y)dx1+1G(x. where r = x 2 + y 2 . Because the function depends on r only. is dU1=1TdS1−1pdV (9. ∂U T = .41). ∂S V ∂U −p = ∂V S (9. V.* The quantity dU is the total differential of U1=1U(S.

called Maxwell relations.44) The general condition that a differential in two variables be exact is that the functions F and G satisfy ∂F ∂G = ∂y x ∂x y (9.21 Test of exactness (i) F dx1+1G dy1=1(x 21−1y 2) dx1+12xy dy. and was a member at 18 (a younger brother.22). only one of whom survived the father. The condition (9. He read a paper on geometry to the Académie des Sciences at the age of 13. and is used in thermodynamics to derive a number of relations. 0 Exercises 53–55 6 The equality of the mixed second partial derivatives was used by Clairaut in 1739 to test a differential for exactness (and also by Euler at about the same time). published a book on the calculus in 1731 at the age of 15. . known as ‘le cadet Clairaut’.44). y) such that ∂z ∂z F = and G = ∂x y ∂y x The differential can then be equated to the total differential of z: ∂z ∂z Fdx + Gdy = dz = dx + dy ∂x y ∂y x (9.9.45) (9. Alexis Claude Clairaut (1713–1765) was one of a family of 20.6 The signiﬁcance of exactness is discussed in the following section. ∂y x ∂x y It is readily veriﬁed that it is the total differential of ax 21+1bxy1+1cy 2. and died of smallpox a year later). ∂2z2∂x∂y. ∂F ∂G We have = −2 y and = +2 y . His Recherches sur les courbes à double courbure (Research on curves of double curvature) in 1731 marked the beginning of the development of a cartesian geometry of three dimensions. EXAMPLE 9. each of these partial derivatives is equal to the mixed second derivative of z.7 Exact differentials 273 is exact when there exists a function z1=1z(x. and the differential is not exact. ∂F ∂G We have = b = . and the differential is exact.46) By (9. ∂y x ∂x y (ii) F dx1+1G dy1=1(2ax1+1by) dx1+1(bx1+12cy) dy. amongst thermodynamic properties (see Example 9.46) is sometimes called the Euler reciprocity relation.

and the Gibbs energy G1=1H1−1TS. for the enthalpy (and using (9. the Helmholtz energy A1=1U1−1TS. The other three are derived from the differentials of the enthalpy H1=1U1+1pV. For example.274 Chapter 9 Functions of several variables EXAMPLE 9. ∂S p ∂H V = ∂p S ∂p ∂2U − = ∂S V ∂S ∂V and the corresponding Maxwell relation is ∂T ∂V = ∂p S ∂S p Similarly.42) that ∂T ∂2U .22 Maxwell relations It follows from equations (9. dH1=1dU1+1pdV1+1Vdp1=1TdS1+1Vdp The quantity dH is the total differential of H as a function of S and p. ∂S ∂p → = ∂V T ∂T V ∂S ∂V → = − ∂p T ∂T p dA = − SdT − pdV dG1=1− SdT1+1Vdp 0 Exercise 56 . so that ∂H T = .40) for dU ). = ∂V S ∂V ∂S so that ∂T ∂p =− ∂V S ∂S V This is one of four Maxwell relations of importance in thermodynamics.

. and let the function F(x.. the conduction of heat. .. ... . .. . Similarly (see Section 5. .. ... .. . ... .. ..... . . ....... .. ... . . . . . . . . ..7 In these examples.... .. .. . .... . .. .. . ....... y ) dy C 7 The concept and notation of the line integral was used by Maxwell in 1855 in his studies of electric ﬁelds.. . . ... . .. .. . ... . .. chosen to be the x-axis.... ..... Maxwell presented his ﬁeld equations in his Dynamical theory of the electromagnetic ﬁeld in 1864. .. . . The line integral notation appeared in a physics text by Charles Delaunay (1816–1872) in 1856 for the work done along a curve.. the general line integral in the plane is I = Z F ( x .. y ) dx C is called a curvilinear or line integral.... ... The Dynamical theory of gases of 1859 describes the Maxwell distribution. and the diffusion of gases. . .. . . . . .. . ..... .... . . . .. .. . . .... .. . . ... . .. .. .. . .... . . .. . . . . .... . . . .... . . . ... .. . . . . . . .... . . ... .. . ... .. . .. . ... . .... James Clerk Maxwell (1831–1879). . . y ) dx + G ( x.. Building on the work of Michael Faraday. ... Figure 9. ... ... . . . the differential work is dW1=1F(x)dx and the total work from a to b is again the deﬁnite integral (9.. . ... . . . . . .. .... .. ........ ... . with applications to the theory of viscosity. . . .. ranks with Newton and Einstein in pre-quantal theoretical physics. if F(x) is the force acting on a body at a b point x on a line. . ... .. .47) The integral was interpreted in Section 5. as shown in Figure 9... . . .... .. . let F(x) be the mass density of a straight rod of matter of length b1−1a (see Section 5.6). .. . ... . . . . More generally.. ..... .... .. .. born in Edinburgh. . ... .. ... .. . ....48) . .. . . .... . .. .. y) is a second function deﬁned on the curve. and the curve C is called the path of integration. .. ..... .. y) be some property associated with the points on the curve. . .. .... . . .. . . . . . . . . .. .... .. .. .. . ..... . . . .. . .7).. . ....8 Line integrals 275 9. . ... . . . .... ... .. ... The quantity Z F ( x... .. . .. .. .. ... . . .. .. .. . .7. ...47)... An alternative interpretation is obtained by considering the function F(x) as some property associated with the points x on a line..7 If G(x...... . ...... . . . x b b • c a a • y (9. .. . . ... . ... ... . . .. .. ... .. .. . . . . . . . ..... . The differential mass in element dx is then dm1=1F(x)dx and the total mass is M = Z F ( x ) dx . ... ... . .. . .. ....... . .. . ....4 as the ‘area under the curve’ of the graph of F(x) between x1=1a and x1=1b..... . .. .. . ..... . and quickly became standard in physics.9. . . .... For example. .8 Line integrals Consider the function F(x) and the integral Z F ( x ) dx a b (9.. . . . . . . . . . ....... . ... .. let y1=1f(x) represent a curve C in the xy-plane in the interval a1≤1x1≤1b. .... . a function is deﬁned on a straight line. . .. ... . ... ... .. .

49)... . ........ . .. ... . .. ... This is demonstrated in the following example... . • . we return to this topic in Chapter 16. . . .. by equation (9. . .. . . .. . . .... .. .8 Figure 9......9 = Z (1 − x 2 ) dx = 0 1 2 3 0 Exercises 57. . . . . dy dx gives replacement of dy in (9..... .. .. . Then dy1=1−dx.... .. ... . . 1 • 1 .. . .48) when F1=1−y. .... . . but C is now the circular arc shown in Figure 9. 1 1 0 y ......48) when F and G are as in Example 9... ... and C is the line in Figure 9... y ) dx dx a (9... . . ... .. . ....49) EXAMPLE 9. Z − y dx + xy dy = Z − (1 − x ) − x(1 − x ) dx 0 c . . . C 1 a .... . . . ..276 Chapter 9 Functions of several variables Line integrals are important in several branches of the physical sciences.. . For example.. G1=1xy....24 Find the value of the line integral (9. .. .. 58 In general... . • b.. Thus....... . . ..... . ....... .. given the curve y1=1f(x).9..... Then −x x dy = dx = − dx and 2 y 1− x I = Z − y dx + xy dy C 0 π 1 = Z − 1 − x 2 − x 2 dx = + 4 3 1 x x a .48) by dx b dy I = Z F ( x .. . .. ....23 Find the value of the line integral (9. they provide a method for representing and calculating the work done by a force along an arbitrary path in ordinary space. . . y ) + G ( x .. 0 . c . .... . . ... . . ... ..8 from A to B (x1=11 to x1=10).. •.... .. . ... .48) can be converted into an ordinary integral over either variable when the equation of the curve C is known... ... .. ... .. The line integral (9.. .. The equation of the circular arc is y = + 1 − x 2 . ..... and. the value of a line integral depends on the path of integration between the end points..... . . . ... ... ... EXAMPLE 9... .. . ..... .. .. . .... .... when generalized to curves in three dimensions. ... b .23... y Figure 9... .. .... . . . The equation of the line is y1=111−1x.

. . . .. when work is done against friction.... 2T c . ... T1)... .. . .72).. From the discussion of work and potential energy in Section 5.23. . . . .... .. . ... . ..... The work done along path C2 is at constant temperature T1=1T2 and. . . . .. .. .... . . .. .. . . . .. ... .. . ... .. ... .. EXAMPLE 9. . . . . . .. and pdV with the (mechanical) work done by the system. . dU1=1TdS1−1pdV is identiﬁed with the heat absorbed by the system. . .. . .. . .. .. . . . . .. 4 .. .. . .. ... . . .. ... ..... . . . . for example... 1 c 1T ... .. .. • . . . ... . •. . .10 for Example 9.. . . . . . ...... . . ... .. . . .. 0 Exercises 59. . . . .. . . .. .. . .7. . .. ..c... . . We consider the work done by the ideal gas on (i) expansion from V11=1V( p1.. . . . . ... at point A in Figure 9. .. .. ... . ... . .. .. .... . .... . .. .. .. .. ... . . . .. T 2 b •... . W11=1nR(T21−1T1)..... ..8 Line integrals 277 (see Example 6... ...... .. ... ..71). . .. .. . for example... .. .. . . . . . ....... . .. .. . . . .... .. .. .25 Work in thermodynamics When changes in a thermodynamic system are reversible (Section 5.. .40). . ... . 60 Z 1 0 1− x 2 dx). ... ...10 (i) Path C11+1C2 The work done along path C1 is at constant pressure p1=1p1 and. ... . by equation (5. ..... . .. . . .. . .. . . .. we expect the work done to depend on the path when the forces acting on the body are not conservative forces.. along path C11+1C2 to V21=1V( p2. ... . . . .. . . . in terms of the work done on a body in moving it from A to B along the path. . . . . .... .. . . . . The result is different from that obtained in The dependence of the line integral on the path can be understood...... .. ..... . . .. . . net work is done in moving the body around a closed path.... . . . .. . the quantity TdS in equation (9. . ... . .. .. A dependence on path is also found in thermodynamics.. .. ... . ..3.. . . In particular.. . .. . . . ... . .... .... ... .. .. .. . .. ..... .. . W21=1−nRT21ln( p22p1). . and (ii) the return to A along C31+1C4.. . . .... ... ...... . 1 . . .. . . . . .. . . . . . . . ... ... . . . . .... . .. ... .... a c 2p 0 p .. . . ..10. . . . .... . . . . ..8).. . ... . . .. .. . . . by equation (5. . .... . . . . . . . . . . .. . . . . . . . ...... . ... .. ... .. . .. . . .. . .. . . .. . . .. . .. ... .... . ... .... . T2) at point B...9. . . . . .. . . . . p Figure 9. .. .. . .. .. .. . ...

In terms of 8 This independence of the path for the line integral of an exact differential was observed by Riemann in 1857. y ) dx + G ( x . In general.7. Z dz = z = z ( x2 .50) (9. there then exists a function z(x. y) such that ∂z ∂z F ( x .51) y x Then.7. the work is independent of the path. y ) dy = dz = dx + dy ∂x y ∂y x and the line integral (9. and no net work is done around a closed path. y1) to B at (x2. 0 Exercise 61 When forces are conservative and can be derived from a potential-energy function.52) and this depends only on the values of z at the end points (we note that the value of the integral changes sign if the direction of integration is changed to B to A). Therefore W31=1nR(T11−1T2). the value of a line integral is independent of the path if the quantity in the square brackets in equation (9. y2). and along path C4 it is at constant temperature T1=1T1. W41=1−nRT11ln( p12p2). y2 ) − z ( x1 .8 By the discussion of Section 9. for a path C from A at (x1. as in equation (5. . y1 ) A C B (9.48) is an exact differential.278 Chapter 9 Functions of several variables Therefore Z C1 + C2 pdV = W1 + W2 = nR (T2 − T1 ) − T2 ln( p2 p1 ) (ii) Path C31+1C4 The work done along path C3 is at constant pressure p1=1p2. and Z C3 + C4 pdV = − nR (T2 − T1 ) − T1 ln ( p2 p1 ) The total work done by the gas around the closed path C1=1C11+1C21+1C31+1C4 is then I pdV = − nR(T2 − T1 ) ln ( p2 p1 ) C and this is not zero unless either T21=1T1 or p21=1p1.48) can be written as ∂z ∂z Z dz = Z dx + dy ∂x ∂y C C (9.57) of Section 5.

and ∂V ∆ V2 = Z dp = C2 ∂p T Therefore ∆ VA → B = V ( p1 . EXAMPLE 9.8 Line integrals 279 the graphical representation of z(x. T ) = V ( p1 . T2 ) − V ( p1 . the line integral is the change in ‘height’ above the xy-plane on the representative surface of z.25. T2 ) − V ( p1 . Path C3 is at constant p2 and path C4 is at constant T1=1T1. T2 ) p1 p1 ∂p T =T 2 * We note however that the line integral (9. T1 ) T 1 ∂T p ∂T p = p T1 C1 1 Path C2 is at constant T1=1T2 (dT1=10).11 for ﬁxed amount of substance) ∂V ∂V dV = dT + dp ∂T p ∂p T As in Example 9. ∆ VB→ A = V ( p2 . V1=1V( p. T2 ) = V ( p2 . T ).48) around a closed path C is zero in general only if the differential Fdx1+1Gdy is exact at all points on and within the closed path. y) discussed in Section 9. T1 ) − V ( p2 . T1 ) = V ( p1 . further consideration of this point is beyond the scope of this book. T2 ) = V ( p2 . we consider the paths C11+1C2 and C31+1C4 shown in Figure 9. as above.9. T2 ) Therefore ∆VB→A1+1∆VA→B1=10 and the change in volume around the closed path is zero.26 Independence of path for an exact differential Volume is a function of pressure and temperature. T1 ) − V ( p2 .10. so that T2 ∂V ∂V T2 ∆ V1 = Z dT = Z dT = V ( p1 . . and this cannot depend on the path between the end points. (i) Path C1=1C11+1C2: The change in volume along path C1 is at constant pressure p1=1p1 (dp1=10). T2 ) + V ( p1 .2. T2 ) − V ( p1 . 63 Z p2 ∂V p2 dp = V ( p. T2 ) − V ( p1 . T1 ) + V ( p2 . T2 ) − V ( p1 .* 0 Exercises 62. T1 ) − V ( p2 . and the total differential volume is (see Example 9. Then. T1 ) (ii) Path C1=1C31+1C4.

...... . .... .. 0 .. . . . .... . T2) = Z dS = Z C2 p2 p1 Figure 9... . . . . .. ... ... .. . ... . 1 3 •... 2p .. .. . . .... .. . .. .... ... .22)..... . ... .. . . c . ... . .. . T2)1−1S( p1. . . ... .. . .. p 1 ∆S1→21=1∆S1→31+1∆S3→2 where ∆S1→3 is at constant p1=1p1: ∆S1→31=1S( p1. .. ...... .. .. . . .. . . .. . .... ∂T p 1 ∂V V ∂T p α= in the following way. . ..... . 1T . Treating the entropy as a function of p and T. . . T1) to a state with pressure and temperature ( p2.. ... . .. . . .. . . . ... .. . . ...11 ∂S dp ∂p T = T 2 Now (Example 9. . . . .... .... . . . . .. . ...... . . .. .•. .11... .. . .... T1 ) = Z dT + dp ∂p T C ∂T p Choosing the path C1=1C11+1C2 shown in Figure 9... T1) T2 ∂S = Z dS = Z dT T1 ∂T p = p C1 and ∆S3→2 is at constant T1=1T2: ∆S3→21=1S( p2. . . .280 Chapter 9 Functions of several variables EXAMPLE 9. ..... . ... .. .. . . the total differential of entropy is ∂S ∂S dS = dT + dp ∂T p ∂p T Then. . .. . . T2 ) − S ( p1 . ∂H Cp = . ... . .1.. . ... . . .. ... ... .. .... . . . T2) can be obtained from calorimetric measurements of the heat capacity and thermal expansivity. . . T2)1−1S( p1..... p 2T c . . .... ..27 Change in entropy in thermodynamics The change in the entropy when a thermodynamic system goes from a state with pressure and temperature ( p1. the change on going from one state to another is independent of the path: ∂S ∂S ∆S1→ 2 = S ( p2 .. .. . . .. . .. because entropy is a function of state. .. . .. . . 1 . . . .... . . .. . .. . .. . . .. . .. 2 •. ... .... .. . . .. ... . .. . . . .... .. .. .... ... .. .. . . . . ... . .. it follows from the differential enthalpy dH1=1TdS1+1Vdp that (dividing dH by dT at constant p) T 2 . . ....... . . . .. ..... .

y).9. 9. or by using a numerical integration method (see Chapter 20). y ) dy c EXAMPLE 9.9 Multiple integrals 281 ∂H ∂S = T = Cp ∂T p ∂T p Also. f(x. The integrals are evaluated either by plotting Cp 2T against T and αV against p and measuring the area under the curve in each case. whilst keeping the other constant: Z f ( x. and the expansivity is measured at several pressures between p1 and p2 at temperature T2. the heat capacity is measured at several temperatures between T1 and T2 at pressure p1. y ) dy dx d b d b b d c a c a a c (9. y ) dx dy = Z Z f ( x. and with respect to y in the interval c1≤1y1≤1d. y ) dx a b d or Z f ( x. the integral with respect to both x and y is also deﬁned: Z Z f ( x.1y)1=12xy1+13y 2 with respect to x in the interval a1≤1x1≤1b. ∆ S3→ 2 = − Z αV dp (at T = T2 ) p1 p2 Experimentally. the Maxwell relation derived from the differential Gibbs energy is ∂S ∂V = − = −αV . can be integrated with respect to either.9 Multiple integrals A function of two independent variables. ∆ S1→3 = Z T2 T1 Cp T dT (at p = p1 ).53) . ∂p T ∂T p Therefore. y ) dx dy = Z Z f ( x.28 Integrate f(x. (i) Z ( 2 xy + 3 y 2 ) dx = x 2 y + 3 y 2 x = y( b2 − a 2 ) + 3 y 2 ( b − a ) a b b a (ii) Z ( 2 xy + 3 y 2 ) dy = xy 2 + y 3 = x( d 2 − c2 ) + ( d 3 − c3 ) c d d c In addition to these ‘partial integrals’.

The double integral is discussed in some detail in the following sections. y. The special case of the triple integral and its importance for the description of physical systems in three dimensions are discussed in Chapter 10.282 Chapter 9 Functions of several variables This is called a double integral and is evaluated by integrating ﬁrst with respect to one variable and then with respect to the other. and (e.28(ii). 65 When the order of integration is not given explicitly. using the result of Example 9. f ) for z: f d b I = Z Z Z f ( x . Z Z ( 2 xy + 3 y 2 ) dy dx = Z x( d 2 − c2 ) + ( d 3 − c3 ) dx a c a b d b 1 = ( d 2 − c 2 )( b2 − a 2 ) + ( d 3 − c3 )( b − a ) 2 0 Exercises 64. d) for y. with (a. b) the limits for x. some care must be taken to associate each pair of integration limits with the appropriate variable. with some changes in interpretation. (i) Integrating ﬁrst with respect to x. y . .53) when f(x. using the result of Example 9. if a triple integral is written as I =Z f Z Z f ( x. but some care must be taken when discontinuities are present. z ) dx dy dz e c a The concepts and methods of integration discussed in Chapters 5 and 6 for integrals over one variable apply to multiple integrals. Z Z ( 2 xy + 3 y 2 ) dx dy = Z c a d b d c y( b2 − a 2 ) + 3 y 2 ( b − a ) dy 1 = ( d 2 − c 2 )( b2 − a 2 ) + ( d 3 − c3 )( b − a ) 2 (ii) Integrating ﬁrst with respect to y. EXAMPLE 9.28(i). y)1=12xy1+13y 2. The value of the integral does not depend on the order in which the integrations are performed if the function is continuous within the ranges of integration.29 Evaluate the integral (9. z ) dx dy dz c a d b e the convention is that the integration is to be carried out from the inside outward. (c. Thus.

. . . . . . . ... .... . .. .... . .. ... .. . .... . ... . .... .... ... . . . . ... .. x b . .. . . ... .. ... . ..... .. . . ... . . . .. . .. . . .. . ....... ..... . .... . . . .. Let f(x. . ... .. .. ...... ....... ..... . .. . . . . ..... . . ... ... ... . . . .. . ... .. .. . .... ... . .55) (9. . The integral (9. .. . .. . . ..... .. .. ... .. .. . . ..... ..... .. . .... ... .... ... .... . ... ... . ... . .... Just as the integral in one variable was interpreted in Section 5. .... . .... . .. .... . .... .. ... .. .. . ys )∆ xr ∆ ys = ∑ hs ∆ ys ∑∑ m n r =1 s =1 s =1 r =1 n m f ( xr .... . .. . . .. .. .. . . . .. . ...54) can be performed in either order: either as where hs is the sum over the elements in a horizontal strip and the second sum is over the horizontal strips. ... ..Divide the interval x1=1a1→1b into m subintervals of width ∆xr1=1xr1−1xr−1 and the interval y1=1c1→1d into n subintervals of width ∆ys1=1ys1−1ys−1. . . .. .. ..10 The double integral (9.. . . . ..... .. .. . ....... . .... . . or as (when the limit exists). ..53) is then deﬁned as the limit The double integral can be deﬁned as the limit of a (double) sum in the same way as the Riemann integral was deﬁned in Section 5. . . . . .. .. .. ... . . . ... . . .. ... . . . . ..... The double summation (9. .. .. .. . . . .... ... .......12). . . .. . . . ....... . ... . .. ... . ..... .... . .. . .. . .. . . . .. . .. .. . . .. .. .. . .... .. .... .. . .. .. . .... .. .... ..... .. .. .... . . ... ........ . .... ... .. . . . .. .... ..... .. ... where vr is the sum over the elements in a vertical strip and the second sum is over the vertical strips... .... .. .4... .. .. . .. ... . . . .... . ... These two orders of summation correspond to the two orders of integration in (9. . .. . .... . . . ... . .. . .. .... ... . .10 The double integral Z Z f ( x. . ... ..... . . .. .. . ........ .. . . . .. . ... . . .. .. . . . . ... .. .. ...... ... . .. .. .. . y) by a continuous function of x and y in a rectangular region of the xy-plane. . .. ........ .... .... .. . ... ... ... . .. ..... . .. .... . . .. .. . ... ...... . .. . .. ... . . .. . ....... ...... ... ...... ... ... ... .. .. . . .. ... .. ... ........ . .. . .. .. . divide the rectangle into small rectangles of area ∆Ars1=1∆xr ∆ys. . . ... ... . .. . ys )∆ xr ∆ ys n m s =1 n m r =1 9...... . . . . .. ..... . .. .. .... the double integral can be interpreted as the ‘volume under the surface’... . . ... .. .. .......... .. ... ∆ ..... .. . .. .. ... .... .... ... . . ..... ... . .. . .... ... ... ... .. . .4 as the area under the curve.... .. . . . . ... .... ..... .... .. . . . .... .. .... . . . . . . .. .... ... .. .. .. . ... . .... . . ..... .. . . . ... . . ... . . .. y ) dx dy = lim a . ... . ... . . .. ..... ...56) 283 ... ..... s ..... .... . ... . . ... . . .. for a1≤1x1≤1b and c1≤1y1≤1d (Figure 9..54) (9. . .. ... ... . . ... . . .. . .53)... . . . .. ..... . . .. .. .... c y ∆ d y c d a b ∑ ∑ f ( xr . .. ... ...... . ..... ys ) ∆ ys ∆ xr = ∑ v r ∆ xr Figure 9.. .... .. . . .. ... ..... . . . . ... .... .... . . .. ....... rx . ....... . .... ... .. .. . . .. ........ .. . .. . . ... . .... .... . . . . ... . ... . . . . .. . that is. ...... . .. .. ..... ...... ..... . ... . . . . . .. .. .. . . .. ............... . .. . ...... 9. .. .. .. ..12 m → ∞ s =1 r =1 n →∞ ∑ ∑ f ( xr ..... . .. ....... ..... ..... . ... .. . . . ... ..... . . ..... .. . .. . . ......... ....... ... ....... ... . . .. .. ... . .... . . . . .. . .. ... .. . . .. . .... .... .. . .. .. .. . . .... . ..

. . . . . .. . . . . . . In Figure 9.... . . .. . .. . . .. . .. .. .. . . . . . . . . .... . .. .. ... . . . . and x1=1q(y) to the ‘right’..... . . . .. . . .. .. .. ... .. . . . .. . . .. .. . . .. .... . . ....... . .. . . . . . . ..... .. . ... ... . .... . . . . . .... . .. . . .. . . . .... ... . .. . . ....... . .. . .... . .. .... .. .. . ... ... .. . . . .. ... .... .. . .. ... . .. . ... . ... . . . . ... .. . .. . .... .... .... .. ... . . . . ....... .. .. ..... . .... . Then.. .. .. . ... the integral can be interpreted as a property of the rectangular region in the xy-plane... ... .. .. . ... ... . .. .. . . . ... . . that is... . .. The double integral sign notation (∫∫) was ﬁrst used by Lagrange in 1760.. . ... .. . . ... .. . . .. .. .... . . ... .......... . .... ...... . ... ..... .. . . . . . ... . . .. . ....... .. .. . ... .. . . . . ..... .... . . . ...... .. .. . .. .. . ... . .... .. . . . . . . . . . .... .. ... . .. . ..... .... . .. .... . ... y ) dy dx R a g(x) M = Z f ( x .. . . . .. .. . .... . ... . . ........ . . . .... . ..... .. .. .. ... . .. . . . .. . . . .. ..... . . . ... ... .. ..... y) be continuous within and on a boundary of a region R of the xy-plane.. . . .... . .. .. and y1=1h(x) ‘above’ these points.. For example.... .. . . ..... .... . .. . . .. .. . . . ... . ... the differential mass in element of area dA1=1dx1dy is dm1=1f(x..... ..... .. . . ... .. . ... . within R.... . . ... . ..13 a b for c d for c1≤1y1≤1d Z f ( x..... .. . ... .. .. . Then.. . the integral is the area of the rectangle: where In Figure 9. .. .. ... .. . .. .. . . . . ..... . .. . .. .. . .. .. .. ...... . . ... .. ... . .. . .. ... .. . . . ... . . ...58a) (9.... . . ... . . . ........ .. . . . ... ... .. .. ... ..... .. . . .. . .. . . . . . . . .. . . . .. ... . .. .. . .... .. .... .. ... . . . .. . . ... ..... . .. .x ) y( q 284 The double integral as a volume was discussed by Clairaut in his Recherches of 1731.... .. .. . . . .. . .. . . .... .... .. ... . .. . .. . . ... . .. . . • and the integral over the region (the sum of vertical strips) is x c yd b xd a y x g • • d . ..... ...... . . . .. . . . . .... . . .. . . . .....57) y . .. ........... ....... . .. ... .. ... .. .. ... . .. . . ... . . ... . . .. . . .. . ... . . .. .... .. . . . . . .. ... . .. . . .. . . . . . .. . ... ......... . . . . . ......... . ) ( .... .. . . . . .. . . . . .. ... . ... .. .. . . . . .. ... =x . ... . More generally.. . .. .. . . .. ... of f(x. ...... . ... . .. . ... . . ... . . . .... ..... .. . . ... .. .. . .. . .. .. y)1=11. ..... .. . .. .. .. .. . . . . .. . . . .. . .... ... ... . . . . ... . . . y) is a surface mass density. .... .. .. .. ... . . . y) and the xy-plane.... ... .... . ... . . . . . . .. .. . . . . . . .. .. .. .. ) y(p = x .. . . . ..... .. . . . . ... ... .. . .. . . . . .. . .. . . .. .... . ... ..... . .. . .. .. ...... . . .. . . . .. ... ... . .. .. ... .. ... . . . .. .. . . . ..... . ... . . . y ) dx dy a b c d a1≤1x1≤1b (b) R (9.... .. . .. . . .. .... .. . . . . ... . . .. Let f(x. .. .... .... .. .. . .. ... . within R. ..... . . . .. ..... . y ) dA = Z p( y)1≤1x1≤1q( y) R c g(x)1≤1y1≤1h(x) d a b Figure 9... . . . . . .. . ..... .. .. ... . . .. . ... ... .. .. . ..... . . . . ..... . .. .. ... . .. .. . ........ . . if f(x.. . .. . ... .. . . . . .. . . . . .. ... ... ... .. . . . . . ..... . .... .. .. . ... ... . .. ... ... . .... . .. . .. ... . . . ..... . ... . .. ... ... ... . .. .. . .. .... y )x(h . . . .. ..... . ... . .. ... .. .. . . . . .. . .. ... .. y ) dA = Z Z f ( x . ... ... ...... ... y)dA and the total mass is Chapter 9 Functions of several variables The form of the integral over the region depends on how the boundary is deﬁned.. . .. . . .. . . ... . .. ..... .. .. .13b the boundary is made up of the two sections: x1=1p( y) to the ‘left’ of the extreme points y1=1c and y1=1d. .. .. = ... . . .. ... . .. . ... ... . ..... . . . In the special case R Z dA = Z Z dx dy = Z dx Z dy = ( b − a )( d − c) (a) R b h( x ) I = Z f ( x ... .. . .......... . .... .. =y 9 Z R ⋅⋅⋅ dA means integration over the (rectangular) region... . . ..... . . . . . .9 Alternatively.... .. ... . . . .. ... .. .... ... . ... . .. . .. . ..13a the boundary is made up of two sections: y1=1g(x) ‘below’ the extreme points x1=1a and x1=1b.. ... .. . .... .. ... the region in the xy-plane need not be rectangular... . . ... .. . .. . . . .. . . . .. . .. . ...... . . ..... the volume between the representative surface of the function f(x. .. ... .. ... .. . . .. .. .. . . .... . . . . . .. . .... . .. . . .. .... . . . .. .... ... ..... . .. . . ... ... . ..... ..... .... .. .. • .. . . . ... .. . .. . . .. . . . .. .. . .. .. . .. . . .. . . . . ... . . . ..

. . .... . . . .. y ).. . given a deﬁnite integral over the variable x.... .. .. .. . R x=y y1=1x 2 (or x = area of R.14 x 1 = Z y + xy 2 2 dx = Z x + x 3 − x 2 − x5 dx = x 4 0 0 1 1 (ii) using equation (9... ... . •. ....58b) Figure 9. .. . .......... . .. ... ... . .... . . . .. .. ... . . . . .. . . . .. . . ..... . ..... . .. . . ... . ... . ....... y EXAMPLE 9.. y 1 = Z x + x 2 y dy = Z y + y 2 − y − y 3 dy = y 4 0 0 1 1 The area of the region R is 1 1 x 1 A = Z Z dy dx = Z x − x 2 dx = 6 0 x2 0 ( ) 0 Exercises 66–68 9....11 Change of variables We saw in Section 6... . .... . . a new variable of integration can be introduced by setting x = x (u ).... .. . .. . ... . ... . . .... .. ... . ... ..3 that one of the principal general methods of evaluating integrals is the method of substitution whereby.. . . ... ...... . .. . ... . .30 Evaluate the integral of f(x.. . .. .... ... . .... .. . . . . . . . ... . .. .. .. . .. ..58b) with p( y)1=1y and q( y ) = 1 y I = Z Z (1 + 2 xy ) dx dy y 0 y.. . .... ..... ... . . .. ....... . . y ) dA = Z Z f ( x . .. . . .... ..... .. dx = dx du du 1 0 x I = Z Z (1 + 2 xy ) dy dx 2 0 x 1 x x=y (i) using equation (9..... . .. .. . . ... y)1=111+12xy over the region R bounded by the line y1=1x and the curve . . ....9... . . . . . .. . . . . ...... .. . 2 .. ........ . . . ...... .. ... ...... . .. .... .. ... ... . . ... ... ......... . ... . as in Figure 9.. . . ... .. •.. ..58a) with g(x)1=1x 2 and h(x)1=1x. . . .. . .. .. .. ..... . .. . 1 .. . .. ... . ... .. .... . . . . ... . .... . ... .. .. . . ....... .. . . .. . . ..14..11 Change of variables 285 and the integral (the sum of horizontal strips) is q( y ) I = Z f ( x .. . . .. ... ... . . . . .. .. . . . . . .. .. . . . . . . .. . . . Calculate also the . . .. . .. . .. . . . .. ..... y ) dx dy R c p( y ) d (9.. ...... .. .. .. ..

.. .. . . . . . ......... ..... .. both the function and the boundary are expressed more simply in terms of polar coordinates (r. . . . . . .. . . dA.. . .. . . .. . . ..... . . . . . ... . . . . .. . .. .. ... . .. . ..58a). .. .. . . . . . ... . .. . .. . .. . ∆ .. . .. . Then to change to polar coordinates... .. ... . . .. . . ... .. . Consider.... . . .. . .. . .. .... . . . . . . . . .. . . . . . . . . . .. . .. . . .. . .. . .. .. .. . . ... . . . . .. .. .. . . .. .... . .. .... .. . .. . . .. ... dA1=1rdrdθ. . . .. .. . .. . . . . . . . . . . . y ) = e− ( x + y ) over a circular region (Figure 9. . . .. .. .. ... .. . .. . .. . . .. . . .. . ... . .. .. .. . .... . .... .. . . . .... The shaded region has area 1 ∆ A = ∆ r × r ∆θ + ( ∆ r ) 2 ∆θ 2 Z f ( x.. . .. . .. .. .. . . . .... .. . .. ... . . ....15 −a a u( a ) u ( b) R R + Z − f x (u ) ( a2 − x 2 a2 − x 2 dx ) du du e− ( x 2 + y 2 )1 2 dy dx Figure 9. r sin θ ) r dr dθ Figure 9.. can be deduced from Figure 9. x . . . . .. ... .. . .. ... . .... . . .. ... . . . r1sin1θ)1=1e−r x1=1r1cos1θ. .. .... . .. ... . . . . .. . . . .. . . . . ... . . . . . . . ... . . . . .. ... . ... . . . . ..16. ... ... .. ...... ... .. . .. .. ... .. . . .. .. .. .. .. . . .... . . . ... . . . . .... . . .. .. 2x − 2a − = y √ 2x − 2a + = y √ 286 Chapter 9 Functions of several variables and the equation of the circle of radius a is r1=1a. ... . . ... .... . ... .. . . . . . . . . . . .. . . . ... . . ... .. . . ...... .. . ... . . .. .. ... .. ... .. a− y By equation (9. .. ... . ... ... the 2 2 2 integral of f ( x . ... .. . .. . . . .... . . .. . . . . .... . .. . . ... . ..... . .16 (9. .. .. . ... ... . .. . . .. .. ... .. . . . . ..... .60) (9. ... ... . . .. .. . . ... .. .. . . . . . . . .. . . .. . . Changes of variable are even more important for double (and higher) integrals because both the function and the boundary of the region may be sources of difﬁculty when expressed in terms of12inappropriate variables. .. . . ... .. . . . . . ... .15). . .. . . . ... .. . . . .. .. .. . .. ... .. . .. .. ... .. . .. . ..... . . . . .... . . .. . . . . .. . . .. ... .... . .. .. ... . .. . ..... .. ... . . . ... . . .. . .... . . . .... .. ....59) . .. .. .. .... .. . ..... ...... . . . . . ... ... .. ..... . . . . . . . . .. . . .. .. . . . . . .. . . . . .. . . . . . .. . . ... . . .. .... . . . . ... . . . . ... . . ... for inﬁnitesimal quantities. .. . .. . .. . .. .. ∆ ... .. . .. ..... . . ... .. . . . . . . .. ... . ... . .. . . .. ....... . .... .. ... . . .. ....... .. .. . .. . a . .. .. . ... . . .. .. . . . . . .. . ... ... .. . .1θ). . .... .. . .. . . . . .. .. . . .. . .. .... . . .. . such that f (x.... ... . . . .... . .. .. . .. It is sensible therefore to make the substitution r In this example.. .. ... ... .. .. .. . .. . . .. ..... .. .. . . . . . .. . . . .. . . . ... . . . .. . .. .. . . .. ... ........ .. . . . . . . . . . . . . . . I = Z f ( x . .. . .. . .θ r . . ...... . . . . ..... ... .. . . . y ) dxdy = Z f ( r cos θ . ... ... . . ... . for example. . ... .... . . ... .. ... . . .. . y ) dA = Z R Z f ( x ) dx = Z a b y1=1r1sin1θ θ so that. . . .. . . .. ..... . ... .... . .. ..... . . . . . ... . .. ... . . .. . ... .. .. . .. .. . . .. . ... . .. ... ...... .... . . .. . ... . . . . . .. . . .. The corresponding element of area.. .... .. .. . . . . .. . .. .. . . . . . . .. .. . ... . Thus... . .. .. .. ... .. . . . ...... . . ... . .. . . .. . . . .. . . . . .. . .. .. . . .... . . . . . .. . . ...... y)1=1f (r1cos1θ. .. ... .. . . ... . . . . . . .. ... .. ... . . . . . .... . . .. .. ... . . ...... .. .. . . .... .. . . . .. ... .. . . .. ... .

59) is then I =Z 2π 0 Z e r dr dθ = Z dθ × Z e− r r dr = 2π 1 − e− a (a + 1) a −r 0 0 0 a 2π a EXAMPLE 9.11 Change of variables 287 The integral (9. the double integral can be factorized: I = Z e− r r dr × Z sin 2 θ d θ 2 a 2π 0 0 Because sin 2 θ = 1 (1 − cos 2θ ). 2 (i) Z e R −r2 sin θ dA = Z 2 2π Z e− r sin 2 θ r dr dθ 2 a 0 0 Because the integrand is the product of a function of r and a function of θ. 2 1 1 1 Z sin 2 θ dθ = Z (1 − cos 2θ ) dθ = θ − sin 2θ = π 0 2π 2π 2 π 0 2 0 2 2 and. using the substitution u1=1r 2. Z e 0 a −r2 a 2 1 1 1 r dr = Z e− u du = − e− u = 1 − e− a 0 2 0 2 2 a2 2 ( ) Therefore. 1− e ) 2( π − a2 Z Z e− r sin 2 θ r dr dθ = 2 2π ∞ π 0 0 2 0 Exercises 69–71 .9. e−r sin2 θ r1=1(e−r r)(sin2 θ ) 2 2 and because the limits of integration are constants.1θ)1=1e−r sin21θ over (i) the area of a circle of radius a and (ii) the whole plane. I = (ii) Letting a1→1∞. du1=12r dr.31 Evaluate the integral of f(r.

63) For example. . y1=1y(u. Then ZZ f ( x.11 We have I =Z e 0 ∞ − x2 2 1 dx = Z e− x dx 2 −∞ ∞ because the integral is an even function of x (see Section 5.1v) (9. It is also the most important in practice.16 of Section 6. following Euler.10 J= ∂( x . if x1=1r1cos1θ and y1=1r1sin1θ then J= ∂x ∂y ∂x ∂y − = r cos 2 θ + r sin 2 θ = r ∂r ∂θ ∂θ ∂r and dA1=1dxdy1→1dA1=1rdrdθ. used the method described here in 1774. The integral Z e− x dx 2 ∞ 0 It was stated in Example 6.5 that this integral cannot be evaluated by the methods described in Chapters 5 and 6. 11 The integral was ﬁrst evaluated by de Moivre in 1733. y ) dx dy = ZZ f ( x(u. The method can be generalized for three or more variables. y(u.3). It can however by evaluated by means of a ‘trick’ involving the transformation of a double integral from cartesian to polar coordinates.16.62) where | J| is the modulus of the Jacobian of the transformation. as obtained geometrically from Figure 9. Also I= 2 1 Z e− y dy 2 −∞ ∞ 10 Carl Gustave Jacobi (1804–1851). y ) ∂x ∂y ∂x ∂y = − ∂(u. Laplace. born in Berlin. v )) J du dv R R* (9. and only a brief discussion of the general case is given here.1v) in the corresponding region R* of the uv-plane. y) in a region R of the xy-plane there corresponds a unique point (u.288 Chapter 9 Functions of several variables The general case The transformation from cartesian to polar coordinates is a special case of the general transformation of variables (of coordinate system). Let x1=1x(u. The Jacobian is used in advanced formulations of thermodynamics for the transformation of thermodynamic variables. v ). such that for each point (x.61) be continuous and differentiable functions of the variables u and v. The general transformation of variables in a double integral was given by Euler in 1769. v ) ∂u ∂v ∂v ∂u (9. discussed the quantities called Jacobians in his De determinantibus functionalibus (On functional determinants) of 1841.1v).

If φ1=1f(x1−1ct)1+1g(x1+1ct).64) 9. 0) Section 9. π22. 2) 2. 1 Z 4 0 2π Z e− r r drdθ = Z dθ Z e− r r dr = 2 2 ∞ 0 1 4 2π ∞ π 0 0 4 Z e− x dx = 2 ∞ π 0 2 (9. φ)1=1(1. y)1=1(3. f1=1cos (x1+12y1+13z) 16. z1=1x 21−13x 2y1+14xy 2 9. z1=1e2x+3y Find all the nonzero partial derivatives of 8. f = xy x + y2 2 17. A transformation to polar coordinates then gives I2 = Therefore. Find the value of the function f(x. If r1=1(x 21+1y 21+1z 2)122 ﬁnd 18. z1=12x 2y1+1cos (x1+1y) Show that fxy1=1fyx for 12. f1=1xye yz ∂r ∂r ∂r . φ)1=1(2. f1=1x 2 cos (y1−1x) 14.12 Exercises Section 9. .3 ∂z ∂z and for ∂y ∂x 3. π23). φ)1=1(0. z1=1sin (x1+1y)ex−y 10. z1=1e x cos (xy) 5. 0) (iii) (x. where c is a constant. f1=1x 31−13x 2y1+1y 3 Show that fxyz1=1fyzx1=1fzxy for 15. Find the value of f(r. z1=12x 21−1y 2 6. θ.9. z1=1x 21+12y 21−13x1+12y1+13 2 7. y)1=12x 21+13xy1−1y 21+12x1−13y1+14 for (i) (x. ∂x ∂y ∂z ∂ 2φ ∂x 2 = 1 ∂ 2φ c 2 ∂t 2 . θ. φ)1=1r 2 sin21θ1cos2 φ1+121cos21θ1−1r 3 sin12θ1sin1φ for (ii) (r. y)1=1(2. u1=13x 21+1y 21+12xy 3 Find all the ﬁrst and second partial derivatives of 11. π. show that ∂y 19. ∂x z . θ. θ. (i) (r. 1) (ii) (x. z1=1sin (x 21−1y 2) Find 4. π24.1 1. π26) (iii) (r.12 Exercises 289 because the value of a deﬁnite integral does not depend on the symbol used for the variable of integration. 13. ﬁnd . . If xyz1+1x 21+1y 21+1z 21=10. Then I2 = 2 2 2 2 1 1 Z e− x dxZ e− y dy = Z Z e− ( x + y ) dxdy 4 −∞ 4 −∞ −∞ −∞ ∞ ∞ ∞ ∞ and the double integral is over the whole xy-plane. y)1=1(0.

y. and amounts nA and nB of the components A and B. θ. (i) Find the stationary points of the function f1=1(x1−11)21+1( y1−12)21+1(z1−12)2 subject to the constraint x 21+1y 21+1z 21=11. 2) from the surface of the sphere x 21+1y 21+1z 21=11. z)1=1a(x 21+1y 21+1z 2)1+12b(xy1+1yz) subject to the constraint x 21+1y 21+1z 21=11 (a and b are constants) is equivalent to solving the secular equations (a1−1λ)x1+ 1by =0 bx +1(a1−1λ)y1+1 bz =10 by +1(a1−1λ)z = 0 These equations have solutions for three possible values of the Lagrangian multiplier: λ1 = a. Find the stationary value of the function f1=12x 21+13y 21+16z 2 subject to the constraint x1+1y1+1z1=11. (ii) by the method of Lagrange multipliers. Use the full notation with subscripts for constant variables. Find the maximum value of the function f1=1x 2y 2z 2 subject to the constraint x 21+1y 21+1z 21=1c 2.5 Find the total differential df: 32. y)1=1x 3y 21+1ln1y 35. 31−1x 21−1xy1−1y 21+12y 23. f ( x . f(r. 4x 31−13x 2y1+1y 31−19y 27.9). CH2CHCH2. n ∂p (iii) . ∂p T . 31−1x 21−1xy1−1y 21+12y 25. f(x. (iii) Show that the three stationary values of E are identical to the corresponding values of λ. 28. (i) by using the constraint to eliminate z from the function. pressure p. (i) by using the constraint to eliminate z from the function. y)1=1x 21+1y 2 34. f(x. (ii) Show that these lie at the shortest and longest distances of the point (1. φ)1=1r1sin1θ1sin1φ 36. For the van der Waals equation n2 a (V − nb) − nRT = 0 p+ V2 ∂V Find (i) . 2. x 31+1y 21−13x1−14y1+12 26. x 31+1y 21−13x1−14y1+12 21. (i) Show that the problem of ﬁnding the stationary values of the function E(x. 30. ∂T V . n ∂p (iv) . ∂T p. (ii) by the method of Lagrange multipliers. λ3 = a − 2b (ii) Find the stationary point corresponding to each value of λ (assume x is positive). see also Example 17. y . (This is the Hückel problem for the allyl radical. Section 9. z ) = 1 x2 + y2 + z2 33. λ 2 = a + 2b. 4x 31−13x 2y1+1y 31−19y Determine the nature of of the stationary points of the functions in Exercises 21−23: 24. 29. n Section 9. f(x. ∂V T . .4 Find the stationary points of the following functions: 22. Write down the total differential of the volume of a two-component system in terms of changes in temperature T.290 Chapter 9 Functions of several variables 20. y)1=13x 21+1sin(x1−1y) 31. n ∂V (ii) .

ﬁnd by (i) substitution. ∂v u 48. If z1=1x5y1−1sin1y.6 37. (ii) equation (9. use the expressions for and from Exercise 20 ∂T p. (ii) the chain rule (9. Given f1=1sin (u1+1v). ∂x z df df .9. (iii) if f1=1u 1−1v . ﬁnd . (i) ﬁnd ∂y 42. For the van der Waals gas.30). If U1=1U(V. Given x1=1au1+1bv and y1=1bu1−1av. y1=1a1sin1θ. If x1=1au1+1bv. n ∂z 44. where a and b are constants. dt dz . Given u1=1ln(x1+1y1+1z). T ) are functions of V and T and if H1=1U1+1pV. (ii) if f1=1x21+1y2. show that ∂U ∂H ∂U + = − ∂T p ∂T V ∂V T ∂V p . x1=1a1cos1θ. where n is a constant. Given z1=1ln (2x1+13y). ﬁnd . ∂y x 49. express and in terms of and . ∂y u 46. where x1=1a1cos1t. dθ dz by dt 40. ∂y u ∂z 45. Hence. = ( a 2 + b2 ) 2 + 2 . y1=1bu1−1av. express ∂u v ∂x y 2 ∂v u ∂y x ∂f ∂f ∂f ∂f ∂f 2 2 and in terms of and . Given z1=1x 21+12xy1+13y 2. and f is a function of x and y (see Exercise 47(i)). ∂x ∂u 1 ∂y ∂v (i) show that = = . ﬁnd (i) substitution.21). ∂T p 47. (ii) = ab 2 − 2 + ( b2 − a 2 ) ∂x ∂x ∂u∂v ∂x∂y ∂v 2 ∂y ∂y . Given u1=1e x−y. . Given z1=1x1sin1y and u1=1x 21+12xy1+13y 2. (ii) if u1=1e−t. and z1=1ct. dt du . ﬁnd du dt ∂V ∂V 43. 38.12 Exercises 291 Section 9. ﬁnd du . ﬁnd ∂x y ∂y x ∂u v ∂v u ∂x y ∂f and in terms of x and y. n ∂p T . y1=1b1sin1t. use the chain rule to ﬁnd 41. n ∂p to ﬁnd ∂T V . ﬁnd ∂v u ∂x y ∂y x ∂u v ∂u v ∂f and in terms of u and v. If z1=1x 21+1y 2 and u1=1xy. where x1=121cos1t and y1=13t. where v1=1cos1u. (ii) If f is a function of x and y. (i) if f is a function of x and ∂f ∂f ∂f ∂f ∂f y. T ) and p1=1p(V. use the chain rule to ﬁnd 39. Given u1=1x n1+1y n and v1=1x n1−1y n. show that (i) ∂2 f ∂u 2 + ∂2 f ∂2 f ∂2 f ∂2 f ∂2 f ∂2 f ∂2 f . where x1=1(11+1t)122 and y1=1(11−1t)122.

the line integral can be evaluated as dx dy Z F dx + G dy = Z F + G dt. 1). Evaluate the line integral Z ( x 2 + 2 y ) dx + ( y 2 + x ) dy on the curve with parametric C equations x1=1t 2.7 Test for exactness: 53. When the path of integration is given in parametric form x1=1x(t). = p 2T ∂T p T ∂p T ∂T p show that the (reversible) heat absorbed by the ideal gas round the closed path shown in Figure 9.292 Chapter 9 Functions of several variables Show that the following functions of position in a plane satisfy Laplace’s equation: B 51.10 is equal to the work done by the gas. Evaluate the line integral Z xy dx + 2 y dy on the line y1=12x from x1=10 to x1=12. 2). 1) (see Exercise 58). y1cos1x1dx1+1sin1x1dy 54. C 60. 0) to B(1. C 58. that is. ∂p T ∂T p Section 9. Given the total differential dG1=1−S1dT1+1V dp. The total differential of entropy as a function of T and p is (Example 9. 59. 2. (6x1+15y1+17)dx1+1(4x1+110y1+18)dy ∂S ∂V 56. (ii) By choosing an appropriate path. evaluate Z [ F dx + G dy ] from C (x. = and = − = − for 1 mole of ideal gas. C y1=1t from A(0. y1=1t from A(0. C tB tA dt dt Evaluate 2 Z ( x 2 + 2 y ) dx + ( y 2 + x ) dy on the curve with parametric equations x1=1t. 0) to (1. show that = − . = 50. y1=1y(t) from t1=1tA to t1=1tB. 3. (4x1+13y)dx1+1(3x1+18y)dy 55. 62. Evaluate Z xy dx + 2 y dy on the curve y1=1x 2 from x1=10 to x1=12 (see Exercise 57). (i) Show that F dx1+1Gdy for F1=19x21+14y21+14xy and G1=18xy1+12x21+13y2 is an exact differential. x 51−110x 3y 21+15xy 4 r Section 9.8 57. y)1=1(0. .27) ∂S ∂S dS = dT + dp ∂T p ∂p T C p 5R ∂S ∂S ∂V R Given that. 61. (iii) Show that the result in (ii) is consistent with the differential as the total differential of z(x. I T dS = I pdV (see Example 9. y)1=13x 31+14xy 21+12x 2y1+1y 3. Ar + sin θ 52. 0) to B(1. n1=11.25). r n1cos1nθ.

1 1− x 2 ∞ ∞ 2 2 12 ZZ 0 ∞ 0 ( x 2 + 2 xy ) dy dx ∞ 2 2 70. (i) from A(1. Evaluate Z 2 xy dx + ( x 2 − y 2 ) dy on the circle with parametric equations x1=1cos1θ. (i) Show from a sketch of the region of integration that Z Z 0 2 ( 2− y )/ 2 x 3 dx dy = Z 1 Z x 3 dy dx + Z 0 Z ( x + 4 )/ 2 x 3 dy dx. 2 2x a a2 − x2 Z Z ( x 2 + y 2 ) dy dx 0 x 67.9 64. Z Z e− ( x + y ) x 2 dx dy 0 0 . ZZ 0 0 2− 2 x xy 2 dy dx 68. 65.12 Exercises 293 63. Evaluate the integral Z Z ( x 2 y + xy 2 ) dx dy and show that the result is independent of 0 1 π 3 2 the order of integration.11 Transform to polar coordinates and evaluate: 69. 0 R 0 Sections 9.9. 1) and (ii) around a complete circle (θ11=101→12π). Z Z e−2( x + y −∞ −∞ ) ( x 2 + y 2 )3 dx dy 71.10 Evaluate the integral and sketch the region of integration: 66. 2 y−4 0 0 −4 0 (ii) evaluate the integral. Evaluate the integral Z Z e− r cos2 θ sin θ dr dθ . C y1=1sin1θ. 0) to B( 0. Section 9. Sections 9. (iii) Conﬁrm that the differential 2xy dx1+1(x 21−1y 2) dy is exact.

a function of position. y. In the cartesian system shown in Figure 10. the temperature at each point in a body deﬁnes a temperature ﬁeld. whose textbook on the calculus was translated into English in 1816 and was inﬂuential (with other texts from the École) in bringing European methods to England and the United States. Functions of position are discussed in Section 10. Lagrange. a point function. the method was classiﬁed for a time as a military secret. The most important and most widely used coordinates. a function and the region in which it is deﬁned may sometimes (in fact. and 19. y.1 Concepts A function of three variables. He showed how the coordinates of a point are determined by the perpendiculars from three coordinate planes. the discussion is restricted to the cartesian and spherical polar coordinate systems. The general coordinate system is the subject of Section 10. f(x. in which he gave x 1+1y 21+1z 21=1a 2 as the equation of a sphere of radius a. often) be expressed more simply in terms of coordinates other than cartesian.3 and volume integrals in 10. and Sylvestre François Lacroix (1765–1843). and in the second volume of Euler’s Introductio of 1748. for points in a plane. When a function of position is expressed in the form f(x. 18. In the textbook Application de l’analyse à la géométrie. developed a ‘descriptive geometry’ that formed the basis for modern engineering drawing. The use of vectors and matrices for the description of systems and processes in three dimensions is discussed in Chapters 16.2. equality. Monge discussed the analytical geometry of two and three dimensions. are the spherical polar coordinates. z). in which the variables are the coordinates of a point in ordinary three-dimensional space is called a function of position. other than the cartesian.1 The value of a function at a given point cannot however depend on the particular system of coordinates used to specify its position. and a hatred of tyrants’). Teachers at the École included Laplace.6. z) it is implied that it is a function of the cartesian coordinates of a point. In these. The analytical geometry of three dimensions has its origins in Clairaut’s Recherches of 1731. The systematic theory was developed by Monge in papers from 1771 and in two inﬂuential textbooks written for his students at the École Polytechnique. Gaspar Monge (1746–1818). We saw in Chapter 9 that. The École became a model for colleges throughout Europe and the United States. They are presented in Section 10. Electric and magnetic ﬁelds are vector functions of position. and respecter of education.1. 1807. the distribution of mass in a physical body is described by the (volume) mass density. and published in 1799 in the textbook Géométrie descriptive. and the velocity at each point in a mass of moving ﬂuid is a vector function of position. Monge was the ﬁrst Director of the École Polytechnique formed in 1794 by the National Convention of the Revolution for the training of engineers and scientists (who show ‘a constant love of liberty. A great variety of physical quantities are described by such functions.5. 10.10 Functions in 3 dimensions 10. is discussed in Section 10. For example.2 Spherical polar coordinates The position of a point in a three-dimensional space is speciﬁed uniquely by its three coordinates in a given coordinate system. a velocity ﬁeld (see Chapter 16). 2 1 . atomic and molecular wave functions are scalar functions of position. peddler. The Laplacian operator. knife-grinder. so important in the physical sciences. the son of Jacques Monge.4. or a ﬁeld.

. . . ...... .. .. .. . . . . . . . ... . it has possible values 0 to π. ... . .. . . .. the cartesian coordinates... . . . . . .. . .. . The distance r of the point from the origin is called the radial coordinate. .. . . . ... .. .. ... . . ... .. .. ... .. 2 2 z1=1r1cos1θ Figure 10. .. . .. .. .. .. . . . . . . . .. .... . .. .. . ... . .. .1φ)1=1(2. .. . .. . .... .. ..... .. .. .. ... . .. .. ... . ... . .. ....... ...... .. .. . . . . . . .... . ... . .... . .. . .... . . ... ..... . . ..... . .. ....... . . . . . ... ... . .. . . .. .1 z1=1r1cos1θ1=121cos(π26)1=1 3 x1=1r1sin1θ1cos1φ. ... . . . . . . .. . . .. . . . .. . . .Other coordinate systems are normally deﬁned in terms of the cartesian system. . .. .... . . .. .. .. . .... . .. .. .. .... ... . . . . .. . . ... . . .... ... ... ... ... . .. . . . . .. . . .. . ) z ..... ..... . .... x ... φ . .. .. .. ... . . ....x(P . . ... . ... . . .. .. . ...... . . ...... . .... . . . . .. ... . .. . . .. . . ... .. .. . In this context the z-axis is called the polar axis.1π26.. . .. and z is the distance from the xy-plane. . The coordinate x is the distance of P from the yz-plane. . .... .. . .... . the colatitude. . . . . . ... .... ... . . . ..... . . .. ... . .. .. . . . . . . . . ... ... . ..... . . .. ... . . .. .. .. .. y .... . . .. . .... . . . . . . .. y1=1r1sin1θ1sin1φ1=121sin(π26)1sin(π24)1=11 x1=1r1sin1θ1cos1φ1=121sin(π26)1cos(π24)1=11 Figure 10.. . . . .. ..... .. .. . . . . .. ... .. . . . . . .. . . ... . . y1=1r1sin1θ1sin1φ. .. .. .. ..... O . Changes in φ describe rotation around the polar axis.... .. .. ... ....... .... . . . . .. • z 0 Exercises 1–3 EXAMPLE 10. . . . .... . .. . . .. .... .. .. . .. .1π24). .. . . is the angle between the x-axis and the line OQ.... . .. ... . ... . . .. .. .... . . . .. .. z ... . ..... . . .... . . . the projection of OP in the xy-plane. . the position of the point P is speciﬁed by the ordered triple (x.. . ... . . . . .. . .. y . . ..θ . . . . .. . .... . . . The system of spherical polar coordinates is shown in Figure 10..... ... . .. . . . ... . ) ( .. ...... ..5). ... .. . ..... . ..... .. .. .. .. . .. .. . conversion from spherical polar to cartesian coordinates is straightforward.. . .. ... .. . .... ... ... .. .. ... ..... . .. .. . .. . . .. ... .. . . z). . . . . .. .. ... .. .... . . .. .... . . ... ... . ... . . . ... . ..... .. y is the distance from the zx-plane... it has possible values 0 to 2π.. . .. . . . . .. ... .... . .. .. . As for polar coordinates in a plane (Section 3. . ... . . .... . . .. . .. .. .. . . . . . .... . . .. . . . ... . ... . . .. . ... . ..... .. . .r p • z )si x a r l op( a z .. . . . .. . ... . . . . .. .. . ... . . .. the longitude. . . .. .... . .. . ... ..1 Find the cartesian coordinates of the point whose spherical polar coordinates are (r... ... .1) 295 x .. ... ... .. .. ...... . . . .. . .. .. The spherical polar and cartesian coordinates are related by the equations q x • y φ x y o y r θ ... . ..... .. .. . . is the angle between the radial line OP and the zaxis. . ....... . ... . . . . . . .. . . . ... . . .. The angle θ... .. ... . ...... . .. . ...... .... . .. ......... ... . .. . ...2 10. .. ...... . ... . . ... . ... .. . . .... . ... . . .. .. ... . . . . ... . . .. . . ...... . .... ..1θ..... . . . .. .. . ... .. . .. ...... .... The angle φ. . y... . ...... .. . .. .. .. ... . .. ... it has possible values from 0 to +∞. . ... . .. ... . .. . . .... .. . .. .... .... .. . . . . . ... ... . . ..2.. . ... . .. . .... .... . . .. .. . . .... .. . . .. .. .. .2 Spherical polar coordinates (10... .. . . .. . .. .. ... . ... . ... .. . .... . . ... . .... . ... ... . . ... .. . .. ..... ....... .. ..

.... if f(x... . .. yp.1y........... .. ... .. . .... .. . ... . . . . . . .... . . .. .. . . ... . . . . .1θp. . . . .. .. .. . .. ..... . . . . .. r −1 in which the inverse functions have their principal values (see Section 3... .. . ... . z)1=1(−1. If the spherical coordinates at P are (rp.. . .3 The temperature at a point cannot depend on the particular system of coordinates used to specify the position of the point... . a body with non-uniform temperature..... . .. ... . . .. .. . . .. .. . ... .. . .. . . . . ... . . ...296 Chapter 10 Functions in 3 dimensions The conversion from cartesian coordinates to spherical polar coordinates makes use of the following relations: −1 y tan x φ= −1 y tan + π x if x > 0 (10. .. ..... 1r1=1 14 θ1=1cos−1(z2r)1=1 cos−1 ( − 3 14 ) 1≈1143..... .. . .. . z T1=1f (x. . . .. ... .... . ..... .. EXAMPLE 10. .. .. . . ... .. . . ... . .5). .2 Find the spherical polar coordinates of the point (x. .. . . ... .... . ..... ... ... . ... . .. .... . . .. ... ... .. .. . . . .. is a function of the three coordinates within some region of three-dimensional space... . . . .1z) Then. ...... .. . . .. .. Tp1=1r 21cos21θp1−1r 21sin21θp1cos21φp1−1r 21sin21θp1sin21φp1=1r 2(cos21θp1−1sin21θp) p p p p y p • V ..... .... . . zp).....1)... 2... ... . .. . .. . . .. .. . . .. .. . . y. the temperature at this point is Tp1=1f (xp.. yp.. . . ... .. .... . ..... . . ... ... .. .... . . .. .. . .3 Functions of position A function of position. . ...... −3). . ... ... .. . . .. ... . .. ... .... . . . .. .....3° φ1=1tan−1( y2x)1+1π1=1tan−1(−2)1+1π1≈1116. . . ... .. .. .. . .. . . ... .... .... . . .. ... . ... .1φp) then. .. or ﬁeld.. .. ... .. .. ... .1z)1=1z 21−1x 21−1y 2 then Tp1=1z 21−1x 2 1−1y 2 p p p Figure 10.. . Let the region V (for volume) in Figure 10.. .. 2 2 2 2 z θ = cos . .. .. if the cartesian coordinates of the point P are (xp. ..... .. ... ... . . . .. zp) For example. . . .3 represent. . . . .. . . ..... . ... . ..... r 21=1x 21+1y 21+1z 21=114. .. . .. . ...... . ... .2) if x < 0 r =x +y +z . .. . . . . . .. .... . by equations (10. .... . ... .. . .. .1y. .... . the temperature is a function of position..... . .. .. . .... . . . . . . . . .. .. . . .. . . x . ... . ... . . ... .. . ... .. .. . ... .. . .. . . ... . ... . . .... . . ..... . .. .6° 0 Exercises 4–9 10... for example.. . . ..... ... .... . .. .. .. . ..... .. . ..... ... . .. . . . . .. .. ....

that is. . .. ........ .. . ..... ... . ..... .. . .. .. . . ...... and is therefore cylindrically symmetric about the polar (z) axis. . . . ... . ... .. ... . . ... .. ...... . . .. .......... ... . or the average mass density. . The ratio ∆m2∆v is then the mass per unit volume. .. .. .............. .... . . .. the function g(r.. . .. . ...... . ......... .. . . . .. ... ... (iii) ( x 2 + y 2 + z 2 )1 2 ∂x ∂r (i) (x 21+1y 21+1z 2)1221=1r. .... .. .. ... . z .. ..... .. the ﬁeld is independent of the value of the angle φ.. ..... ... ..... . . . .. . .. . ......... the value of this ratio depends not only on V the position of the volume ∆v but also on the size and shape of ∆v.. . . .......... .. . ..... .. ... . . . . .. . .... ... . . . ..... . ... .. ∆v1→10... T1=1f (x.. ... . ... . .. . . as shown in Figure 10. . .. .... . . it only depends on two of the spherical polar coordinates.. ..... ... (ii) ∂ 2 ∂ ( x + y 2 + z 2 )1 2 .. . ..4.. .10.. .. . . If the mass is not distributed uniformly throughout the body. .. . ..... φ) In this example.. . . ... . . . . . ... ... . ... ... . . .. . .. . .. . .... ...3 Functions of position 297 The temperature ﬁeld is therefore described equally well by the function g(r... ... . ........ ... . . ... ....... . .1φ)1=1r 2(cos21θ1−1sin21θ ) such that. . .. .1θ. ... . ... . ... . ... . . ........ ... that is. .. . ....4 y .... . ... . .... ....... . The second simpliﬁcation is that the variables in the function g have been separated... .. ... ..... . . . .. .... y.. ..... ... ..... . .... .. . ..... ... (ii) x ∂ 2 1 ( x + y 2 + z 2 )1 2 = ( x 2 + y 2 + z 2 )−1 2 × 2 x = = sin θ cos φ ∂x 2 r ∂ 2 ∂ ( x + y 2 + z 2 )1 2 = r = 1 ∂r ∂r (iii) 0 Exercises 10–13 Density functions Consider the distribution of mass in a three-dimensional body (see Section 5.. x.. . ..3) x ∆ m ∆ v →0 ∆v Figure 10... .1θ) has been factorized as the product of a function of r and a function of θ... .. . . ... . .. ..... θ. . .... ... . . . Such simpliﬁcations are important for the evaluation of multiple integrals and for the solution of partial differential equations. y. ... . .. . .. ..... . . . .. .. .. . .. . . .... .. Firstly. and let ∆m be the mass in a volume ∆v surrounding the point P. . ...... . .... . .... ..... . .. . .. .. . ..... ... ..... ... .. . the transformation from cartesian to spherical polar coordinates has led to two simpliﬁcations of the representation of the ﬁeld.... . ... ... .. . . . .. ... . ... . . .. ... . ........ . .. ... whereas T depends on all three cartesian coordinates...... and z.. .. .. .... .... . ... . .... ... .. . . . .. . .... . . . . . . ... . . ... . . r and θ. . . . .. . . ... .. .. ... . .. .. ...... . . ... ... ... . .. P • . ... .....6 for a linear distribution of mass). . .. ..... . .. ...... .... in ∆v.. . ... .. . .. As the size of the volume is reduced to zero. . .. ... ..... . .... . . . Let P be some point within the body... .. z)1=1g(r.. .. . .. . .. . . ...... . ......... .. . ............. . . .. . .. . . . . .. . .3 Express in spherical polar coordinates: (i) (x 21+1y 21+1z 2)122...... at any point in V.. ... EXAMPLE 10. ... .... ... ..... .. .. . . .. .... .. ... .. . . . . . . .. . ....... . .. ..... the ratio approaches a limit ρ = lim (10.... .. ..

298 Chapter 10 Functions in 3 dimensions and the value of this limit is the density at the point P. one for each coordinate: ψ (r. The concept of density is applicable to any property that is distributed over a region. M = Z dm = Z ρ d v V V (10.4 Atomic orbitals and electron probability density The solutions of the Schrödinger equation for an atom. We note that although ρ is deﬁned as the limit of a ratio it is not a derivative in the normal sense.1 (a0 is the bohr radius). We can however consider the differential mass dm1=1ρdv in volume element dv. Some of the orbitals of the hydrogen atom are listed in Table 10. The total mass of the body is then the integral over the volume of the body.5) The radial function R(r) determines the size of the orbital. One important example is the probability density of a distribution in probability theory and statistics.4) Such volume integrals are triple integrals over the three coordinates of the space and are discussed in Section 10.1φ)1=1R(r)1⋅1Θ (θ)1⋅1Φ (φ) (10.4.1 Atomic orbitals of the hydrogen atom ψ 1s = 1 3 π a0 e − r 2 a0 ψ 2s ψ2p = = 1 4 3 2 π a0 ( 2 − r 2 a0 )e − r 22 a0 − r 22 a0 1 5 4 2 π a0 re x sin θ cos φ ψ2p y = 1 4 5 2 π a0 re − r 22 a0 sin θ sin φ ψ2p z = 1 5 4 2 π a0 re − r 22 a0 cos θ . the atomic orbitals. and the radial or ‘in-out’ motion of the electron in the orbital. Each orbital is the product of three functions.1θ. The angular functions Θ (θ) and Φ (φ) determine the shape of the orbital and the angular motion of the electron (its angular momentum). These functions are discussed in greater detail in Chapter 14. Probability densities are also used in quantum mechanics for the interpretation of wave functions and for the computation of the properties of atoms and molecules. discussed in Chapter 21. EXAMPLE 10. Table 10. are nearly always expressed in spherical polar coordinates.

...5.1φ). . . . . .... . .. . . . . . When the variables are the coordinates of a point in ordinary space the integral is often called a volume integral... . . .. .. . .. ..4 Volume integrals 299 The physical interpretation of an orbital is in terms of an electron probability density.. . . . ... .. . ..... . .. .. .... . . ... . integral has the general form Z f ( x.. .. . . ....... ... . . ... .. . .7) is c o x a ..4 Volume integrals A triple.. . ...... ..... .. .. . ..1y. ... .. ..7) are constants then the region V is a rectangular box of sides x21−1x1.... ... ..... . ...6) is interpreted as the probability of ﬁnding the electron in the volume element dv at position (r. .. y.... . .. . . .. ... .. ...1θ.. . . ... . ... . ... .. .. ... ... ... ... .. .. . . . ... ... . . ........ .... . ... ....... .... .. ...... z ) dv = Z Z Z f ( x. ... . . .. Z f ( x. .. . . .. z ) dv = Z dv + Z xyz dv V V V Then c b a 0 a 0 b c 0 Figure 10... . .. . . . .... z21−1z1.. .. ....... ... ... . The probability of ﬁnding the electron in a region V is then the volume integral Z V |ψ | 2 dv ... . . .. ... ... .. . |ψ |21=1ψψ *. . . . ... . .... ...... . ....1z)1=111+1xyz over the rectangular box of sides a.. . The square modulus.. .5 (i) Z dv = Z Z Z dx dy dz = Z dx Z dy Z dz V 0 0 0 = a × b× c =V This is a general result. . ....... . . .. .. .... .. . .. . ... . ..5 Evaluate the integral of the function f(x... y. If the limits of integration in (10.... ....... .... .. .. ... .. . . ... . . . . .. . . ... EXAMPLE 10.. ...10.. z ) dx dy dz V z1 y1 x1 z2 y2 x2 (10... .. .... b. is used because wave functions are in general complex functions. ..... . . . ... . ... . . .. ........... .. . c shown in Figure 10.. . . . . . .. .. . . ... .. . . .... .... y.. .. y21−1y1. . ..... ........... ... . ..... .. ... ...... ..... .... . .. .1θ. .. . ..1φ)|2 dv (10... . ..... . .. for an electron in orbital ψ the quantity |ψ (r.. . .... . 10.. . .. .. ..... . . . . .. .. . . or three-fold. ..... . .. . ..7) where V is a region in xyz-space. . .. . . . . . .. ... .. ...... . . . a 0 b c 0 (ii) Z xyz dv = Z Z Z xyz dx dy dz = Z x dx Z y dy Z z dz V 0 0 0 y b z The integral (10. . . . the integral c b a 0 Z V dv is the volume of the region V........ ... . ... ..

. .. . . .. . . . ...... . . ..... .... . ... ... .. ....... .. .. . . . . . . .... φ ) r 2 sin θ dr dθ d φ a 2 b2 c 2 V 2 × × = 2 2 2 8 dv1=1r 21sin1θ dr dθ dφ (10... .. . . ... .. .. . ... . .. y . .. . .. .. ... . . .. .... . . ... ... . . ... .... . .. ... . . . .. .. .. .. . ... . . . .. . . . ... . .. ... . .. .. . r1sin1θ ∆φ... . . .. . . . . . . . . ..... .. . . . .. . .. .. .. .. ... . . . . . . .. . . ... . . . ... .. . . ... .... . . .... .6 (10. ..8) .. . . . . . . .. .. . . ..... ... . . .. . . . . . .... ... .... .. .. .. .. . ......... . . ... . . . .6. . . .. ... . .. . .... ∆ . .. .... .. . .... . . .. ... . .. . . .. . . .. .. . .. . .. . . .. .. .. . ..... . .. ... . . . . . . .. .. ... . .. ... . .. . ... . .. . ... . . . . . .. . . .. . .. . . .. ... .. . . . . . .. . . .. . . .. .... ... . . ...31). . . . . .. . . .. . . .. . . . .... .... . .. . . ...... ............ . .. . . . . .. . φ ) dv = Z Z Z f ( r .... .. Figure 10. ... .. r∆θ.. .. .φ φ .. ...... .. ... .. .... . .. . . .... . .. .... ... .. ...... . ... .. . . ... . .. .... .. .. .... .... .. ... . ... .... .. . ... . . . ...... .. . . . .. ...9) (10.. .. .. .... . . ... . .. . . ..... ... . ... ... . ... .. . . .. . . . .. .. ..... . . .. . .... . . ... . ..... . .. . . .. . . .. .. .. . . . . . .. . .... . .. ...... .. .. . . . . .. . .. .. ... ... . .. .. . .... . . ... .. .. .. . . ... ........ . .... . . .. ...... .. . . ..... .. . ∆.. .. ..... . .. . .. .. . ..... .... . . .. ... . . . ........... . . ... . . .. . ...... . . . . . The region ∆v is a section of spherical shell of thickness ∆r between radii r and r1+1∆r. . . .. .. .. .. .. .. .. . . . .. . .. .... .. . . .. ..... .... ..... and angles φ and φ1+1∆φ. . .. . . .. . . . . .... . ..... . .... . .. ... . . . . ..... . . ....... . . . .. ... . . . ... and the limits of integration are constants (see Example 9.. . .. It can be shown (see Example 10. .... ..... .... . . ... . . ... .. ... . . .. .. . . ... .. . .... .. .... . .. .. ... .... . . .. .. ... ..... . .... ............ ..... .. .... . . . .. .. . ... . . . . . . .. .. .. .. .. .θ . . . . . . ... . .... .... . ... . ... . .. .. . .. . ... . .. . .. . . . .. ...... θ θ For small ∆-values this volume is approximately equal to the volume of a rectangular box of sides ∆r.. .. .... ... .. . .. . .. . ..... one for each variable. ... . . . . . .. . ... .. . . .... . .. . .. . . .... . . . .. . . . . . . . . .. ... . . . . .. . . .... . .. ...... ... .... .. ... .. . . . .. . ..... ... ... . . ... ... .. . .. . .. . ...... .. ... . .... .. . . . . . . .. angles θ and θ1+1∆θ... .... .. . .. . .... . . . .. .... .. . .. . .. . ..... .. . ... .. ...... .. . ... .. . . . ..... . .. ... . ... .. .... . . . ..10) θ1 θ2 r1 r2 x and the volume element (10...... . . ... .. .. .. .. . .. .. . .. . ... . . . . . ... . . . . . ... . .... . . . .. . ... . . .. ...θ ... . . . . . ... r ∆ . . . . . ... r . ....... .. . . .. Then 1 ∆ v = r 2 ∆ r + r ( ∆ r ) 2 + ( ∆ r )3 3 Z xyz dv = × cos θ − cos(θ + ∆ θ ) × ∆ φ . . . 300 Chapter 10 Functions in 3 dimensions can be understood in terms of Figure 10. .. .... . . .. .. . .. . .. .. .... . ...... ...... .. . ..... . ... ... . .. . ... . . . o V ∆v1≈1r 21sin1θ1∆r1∆θ1∆φ V φ1 φ2 Z f ( r .... .... . .. . .. . .... . .. .. .. ... ... . . ... ... . ...9) is obtained for inﬁnitesimal quantities.. . . ... .. . . .... . . .. .... ... . .... . .. .. . . ... .. ... .. . . .. . .. ... . .... ... . .. . v ∆ .. .. ......... . ...... . ... .. .. ... ... ..... ... .. ..... . . ... . .. . ... . . . . .. . .. . .. ... .. .... .. . . .. .. ....... . . ... .. . . . . . . .. . ... . . .. . . ... . .. ....... .. . . .... .. . . .. .. . . .. .. . . . .. ... . ....... . .. .. ... . .. ... ... .. . . . .. . . . ... .6) that the volume of the region is z The form of the volume element The volume integral in spherical polar coordinates is Spherical polar coordinates 0 Exercises 14–17 The factorization of the triple integral is allowed because the integrand is the product of three functions.. ... . .... ... .. . . .. ... ... .. ..... .. . . ... .. . ...... ... .. .... . . . . . ... ... .. . ... . . . . . . .... . .. . . . .. .. . ... . . . ... .

5: Z f dv = Z dv + Z r 2 cos2 θ sin 2 φ dv V V V The ranges of integration are r1=101→1a.1θ. as in Example 10.6 and show that it reduces to the approximate expression (10. Therefore. When ∆θ is small.9).1φ)1=111+1r 21cos21θ1sin21φ over a sphere of radius a and centre at the origin.10) for small ∆-values.10. θ1=101→1π. Then 2π π a (i) Z dv = Z Z Z r 2 sin θ dr dθ d φ V 0 a 0 0 π 2π 0 = Z r 2 dr 0 Z sin θ dθ Z d φ = 0 a3 4 × 2 × 2 π = πa3 3 3 and this is the volume of the sphere. EXAMPLE 10. The integral can be evaluated in two parts.7 Evaluate the integral of the function f(r. the terms in (∆r)2 and (∆r)3 can be neglected. cos(θ1+1∆θ)1=1cos1θ1cos1∆θ1−1sin1θ1sin1∆θ1≈1cos1θ1−1sin1θ ∆θ since cos1∆θ1→11 and sin1∆θ1→1∆θ as ∆θ1→10. and φ1=101→12π. 19 . ∆v = Z =Z φ +∆ φ Z 2 θ +∆θ φ θ Z r +∆ r r 2 sin θ dr d θ d φ sin θ d θ r r +∆ r r dr Z θ +∆θ r θ Z φ +∆ φ φ r 3 r +∆ r − cos θ θ + ∆θ φ φ + ∆φ dφ = φ θ 3 r 1 = r 2 ∆ r + r ( ∆ r )2 + ( ∆ r )3 cos θ − cos(θ + ∆ θ ) ∆ φ 3 When ∆r is small enough. 2π π a (ii) Z r cos θ sin 2 2 V 2 φ dv = Z 0 a Z Z r 4 cos2 θ sin θ sin 2 φ dr dθ d φ 0 4 0 π 2π 2 0 = Z r dr 0 Z cos θ sin θ dθ Z sin 2 φ d φ 0 a5 2 = × ×π 5 3 0 Exercises 18.6 Find the volume ∆v in Figure 10. since the volume element is used for the construction of the volume integral. ∆v1≈1r 2 sin1θ ∆r1∆θ1∆φ We note that this does not provide a proof of (10.4 Volume integrals 301 EXAMPLE 10. for small ∆-values.

is 2 Zψ 1s dv = 1 π a0 Z Z Z e−2r 2 a0 r 2 sin θ dr dθ d φ 3 0 0 0 2π π ∞ = 1 3 π a0 Z e 0 ∞ −2 r 2 a0 2 π 2π 0 r dr Z sin θ d θ Z d φ 0 The integral over the angles π 2π Z sin θ dθ Z d φ = 4π 0 0 (10. the coefﬁcients of the orbitals in Table 10. and must by unity. θ . 0 Exercises 20–22 .11) −∞ ∞ =Z 0 Z Z f ( r . z ) dx dy dz in cartesians (10. making use of the standard integral Z ∞ − ar 0 e r n dr = n!2 a n +1 . In fact. The total probability.8 The integral over all space of the electron probability density of 2 3 −2 r 2 a0 the 1s orbital of the hydrogen atom. is the integral over all space. EXAMPLE 10. the probability of ﬁnding the electron somewhere.12) o These integrals are important.1 have been chosen for this to be true. in quantum chemistry for the evaluation of atomic and molecular properties from wave functions obtained as solutions of the Schrödinger equation. Then. for example.1).13) is the complete solid angle around a point. φ ) r 2 sin θ dr dθ d φ 0 0 in spherical polars (10.302 Chapter 10 Functions in 3 dimensions Integrals over all space When the region V is the whole three-dimensional space. y. 1 3 π a0 2 Zψ 1s dv = × 2 ( 22 a0 )3 × 4π = 1 This result is consistent with the interpretation of | ψ |2 dv as the probability of ﬁnding the electron in element dv. The orbitals are said to be normalized (to unity). ψ 1s = (12 π a0 ) e (see Table 10. the volume integral is Z f dv = Z Z Z V −∞ 2π −∞ π +∞ +∞ +∞ f ( x .

and that the z-coordinate have value between z and z1+1dz. For example. and let p(x.θ . The average value of the function is then the volume integral over all space f =Z +∞ Z Z −∞ +∞ +∞ f ( x . then the quantity f = Z f ( x ) p( x ) dx a b (10. z ) dx dy dz (10.1z) dx dy dz be the probability that the x-coordinate have value between x and x1+1dx. z ) p( x . (10. y.10.1y.1z) be a function of position in three dimensions.4 Volume integrals 303 Average values We shall see in Chapter 21 that if x is a continuous variable in the interval a1≤1x1≤1b.15) −∞ −∞ The corresponding expression in spherical polar coordinates is 2π π f =Z 0 Z Z f ( r . and r = Z r | ψ |2 dv 1 3 π a0 −2 r 2 a0 (i) 2 ψ 1s = e . y. In these cases. let f(x.14) is the average value of the function f(x) in the interval.θ . that the y-coordinate have value between y and y1+1dy.1φ)1=1r in equation (10. φ ) r 2 sin θ dr dθ d φ 0 0 ∞ (10. p1=1| ψ |2. ∞ r1s = 1 π a0 Z Z Z e−2r 2 a0 r 3 sin θ dr dθ d φ 3 0 0 0 2π π . and if p(x)1dx is the probability that the variable have value between x and x1+1dx.16) If the probability density is the modulus square of a wave function in quantum mechanics. φ ) p( r .4).1 in Example 10.17). f(r.1y.9 Find the average distance of the electron from the nucleus in (i) the 1s orbital and (ii) the 2pz orbital of the hydrogen atom (see Table 10. The generalization to functions of more than one variable involves the corresponding multiple integral.1θ.17) EXAMPLE 10. the average value f = Z f | ψ | 2 dv is usually called the expectation value of f in the state ψ.

19b) is obtained by use of the product rule of differentiation. r1s = 4 3 a0 Z e−2r 2 a0 r 3 dr = 0 ∞ 4 3 a0 × 3! ( 22 a0 ) 4 = 3 a 2 0 (ii) 2 ψ2p = z 1 5 32 π a0 r 2e ∞ − r 2 a0 cos2 θ π r2 p = z 1 1 32 π a0 Z r 5 e− r 2 a0 dr Z cos2 θ sin θ dθ Z d φ 5 0 0 0 6 × 5!a0 × 2π = 5 32 π a0 2 × 2 π = 5a0 3 0 Exercises 23. ∇2 = = 1 ∂ 2 ∂ 1 1 ∂2 ∂ ∂ r + 2 sin θ + 2 2 2 ∂r ∂θ r sin θ ∂φ 2 ∂r r sin θ ∂θ r ∂2 ∂r 2 + 2 ∂ 1 ∂2 cos θ ∂ 1 ∂2 + 2 + 2 + 2 2 r ∂r r ∂θ 2 r sin θ ∂θ r sin θ ∂φ 2 (10. For example.6. ∇2 = in spherical polar coordinates.20) The Laplacian operator occurs in the equations of motion concerned with wave motion and potential theory in both classical mechanics (as in Maxwell’s equations .5 The Laplacian operator The Laplacian operator in two dimensions was discussed in Section 9.19a).18) (10.19b) The transformation from cartesian to spherical polar coordinates is achieved in the same way as that described in Example 9. 24 10.13)). The operator in spherical polar coordinates is usually quoted in the slightly more compact form (10. after integration over the angles (see equation (10. and the expanded form (10.304 Chapter 10 Functions in 3 dimensions and.19a) ∂2 ∂x 2 + ∂2 ∂y 2 + ∂2 ∂z 2 (10. The operator in three dimensions is: in cartesian coordinates.18 for the two-dimensional case. 1 ∂ 2 ∂f 1 r = r 2 ∂r ∂r r 2 ∂2 f ∂f ∂2 f 2 ∂f r2 + 2r = 2 + 2 ∂r ∂r ∂r r ∂r (10.

∂r dr ∂2 f ∂r 2 = d2 f dr 2 = e− r + 2 df 2 = e− r − e− r = 1 − r dr r 2 −r e r (ii) In cartesian coordinates. ∇2 f = 2 df d 2 f 2 − r + = − + 1 e r dr dr 2 r We note that the use of cartesian coordinates in this case naturally involves. EXAMPLE 10. ∇2 f = ∂2 f ∂x 2 + ∂2 f ∂y 2 + ∂2 f ∂z 2 ∂r x = . 0 Exercises 25–27 . Then ∇2 f = ∂2 f ∂x 2 + ∂2 f ∂y 2 + ∂2 f 3 ( x 2 + y 2 + z 2 ) df x 2 + y 2 + z 2 d 2 f = − + dr dr 2 ∂z 2 r r3 r2 so that.10. ∂f df ∂r x df = = . ∂x dr ∂x r dr ∂2 f ∂x 2 = ∂ ∂x x df 1 df x 2 df x 2 d 2 f − + = r dr r dr r 3 dr r 2 dr 2 The derivatives with respect to y and z are obtained in the same way. a transformation to spherical polar coordinates. ∂φ ∂f df = = − e− r .5 The Laplacian operator 305 of electromagnetic theory) and quantum mechanics (as in Schrödinger’s wave mechanics. via the chain rule. by the chain rule. see Chapter 14). ∂θ Then ∇2 f = d2 f dr 2 ∂f = 0.10 Evaluate ∇2f for f(r)1=1e−r in (i) spherical polar coordinates and (ii) cartesian coordinates. because x21+1y21+1z21=1r2. ∂x r ∂r y = . (i) Because f(r) is a function of the radial coordinate only it follows that ∂f = 0. Then. ∂y r ∂r z = ∂z r r = ( x 2 + y 2 + z 2 )122 .

1 2 2 cos 2 θ − sin 2 θ 1 1 ∇2 f = + 2 − + − 2 2 RΘ Φ = − 2 2 4 r r 4 r sin θ r sin θ 2 f r .1φ)1=1re−r22 sin1θ1cos1φ. 1 d 2 dR 1 d dΘ RΦ d 2Φ RΘ ∇2 f = 2 + r ΘΦ + sin θ r dr dr sin θ d θ d θ r 2 d φ 2 r 2 sin 2 θ Now R = re− r 22 .11 Evaluate ∇2 f for f (r. therefore.1φ)1=1(re−r22)(sin1θ )(cos1φ)1=1R(r)1×1Θ (θ )1×1Φ (φ) so that ∂f dR( r ) = Θ (θ ) Φ (φ ) ∂r dr dΘ (θ ) ∂f = R( r ) Φ (φ ) ∂θ dθ d Φ (φ ) ∂f = R( r ) Θ (θ ) ∂φ dφ and.306 Chapter 10 Functions in 3 dimensions EXAMPLE 10. Therefore. Using the form (10.1θ. − r 22 1 2 2 1 d 2 dR r 2 = + 2 − R( r ) r = + − 2 e r 4 r r 2 dr dr 4 r 1 d dΘ cos 2 θ − sin 2 θ cos 2 θ − sin 2 θ = Θ (θ ) sin θ = dθ sin θ d θ sin θ sin 2 θ d 2Φ dφ 2 = − cos φ = −Φ (φ ) Θ = sin θ.1θ. Φ = cos φ. ∇2 f = 1 ∂2 f ∂ ∂f 1 ∂ 2 ∂f 1 sin θ + 2 2 r + 2 2 ∂r ∂θ r sin θ ∂φ 2 ∂r r sin θ ∂θ r The function f has the factorized form f(r.19a).

12 Show that the function f (r)1=112r satisﬁes the Laplace equation in three dimensions.10 is the same equation for the 1s orbital. with f the 2px orbital. made contributions to the theory of the elasticity of solids and was involved in the construction of the ﬁrst railways in France.1). The most important of these are of the type called orthogonal curvilinear coordinates. the function does not depend on the angles θ and φ.2 In the cartesian system (Figure 10. other systems of coordinates have been found useful for the description of physical systems.1y. This example shows that the gravitational and Coulomb potential functions satisfy the Laplace equation (see Example 9.20 for the two-dimensional case). The result of Example 10. Gabriel Lamé (1795–1870).10.6 Other coordinate systems In addition to the cartesian and spherical polar coordinates. The Laplace equation is ∇2f1=10 As in Example 10. Therefore d2f dr 2 + 2 df 2 2 = − =0 r dr r 3 r 3 ( r ≠ 0) We note that this demonstration is not valid when r1=10. .6 Other coordinate systems 307 This result can be rearranged to give the equation 1 2 1 1 − ∇ − f = − f r 8 2 and this is essentially the Schrödinger equation of the hydrogen atom. the position of a point P(x.1z) is deﬁned by the intersection of three mutually perpendicular surfaces (planes). Therefore. engineer and professor at the École Polytechnique. 0 Exercises 34–37 10. and this singular point requires special treatment in physical applications. d 2f2dr 21=122r 3. x1=1constant 2 Curvilinear coordinates were introduced by Lamé in his Sur les coordonnées curvilignes et leurs diverses applications.10. ∇2 f = d2f dr 2 + 2 df r dr Now df2dr1=1−12r 2. 0 Exercises 28–33 EXAMPLE 10.

.. .... .. . ... .. ... . . .. . . q3 = q3 ( x. or orthogonal. q1. . . .. . . . . and are also perpendicular. . θ1=1constant..... ... .. .. .. . .... . ... . The surface φ1=1constant is the (vertical) plane APA′ in the ﬁgure.. y ... . ...... . ... .. .. .. .. q2. .. .. q2 = q2 ( x ... . .... .22) represents a surface. . . ... .. ... . .... .. .. .. . . . . . q3 )...308 Chapter 10 Functions in 3 dimensions (parallel to the Oyz–plane)...... . .. .. . . ... .. .. . . ... . . .. . ... ............... ... . . .. . . . .. . .. . . . . . q3....... ... . .. .. . ... . . .. . . .. ... and z1=1constant (parallel to the Oxy–plane). . . . . .. . .... . . . .. .. . . . .. . z ) (10. . .... . . ..... . . . .. .. . ... . ..... . .. at their point of intersection P..... ......... .. .. . .. .... . . . . . . .. . . ... . q21=1constant. .. .. .. ... .. .. . .. .. . .... .. .. .... . .. . . . . .. .. . . . . . . . .... .. ... . . .. ..... . . ... . .. . . . by the equations x = x ( q1 .. . . ..... . The coordinate axes through the point are the lines of intersection of these planes. .. .. . .... the φ -coordinate line is the circle BPB′. . the curves of intersection in pairs are the coordinate lines. . .. .. ... . ... . .. . and z. . . . . .... . . . ... .21) The position of a point is then deﬁned by specifying either x. . .. . . . .. . . q3 )... let the cartesian coordinates. . . . . y. . .1θ... .. .. .. .1φ) is again deﬁned by the intersection of three coordinate surfaces. ... . .. .... . . . . ... . .. . . .. .. . Each of the equations (10. . ..... . z ).. . .... . . .. .. . .. . . . . .. . ... . .. . b r φ θ • p a o a b x .. .. .. .... .... . ... y1=1constant (parallel to the Ozx–plane). .. ........ . ..... ..... ... .. .. . . ... . . ... . . . .. . . . .. . . .. ... . .... .. ...... . ... .... . .. .. ... ... .. . .. . ... . . .. . .. . .. .. . .. .... ... .. .. . .... .. .. ...... . . y.. .. . ... ... . . .. . .. . .. . . . . . ...... ... .... .. .. ..... . .. . ... q2 .. . . .... . . . .. .... . .. ... z or q1... . . . .... . .. . .. . . ... . .. ...... . . .. ..... . . . . .. ... . . . . .. . The surfaces q11=1constant... . Orthogonal curvilinear coordinate systems are those y . r1=1constant. .. . . .. ...... ..... . . . ... . .. . . ... . . . .... . . . .. . .. .. . they are nonorthogonal in general. The quantities (q1. .. .... . ..... ... .. ... . . .. . . .... .. .. .. . . . . .. . . .... . .. .... .. .. .. .. . . . ... . . . .... .. .. .. . .. . . .... q3) are the curvilinear coordinates of a point P(x.. . . .. . . .. .. . . .. .. .. .. . . z = z ( q1 . . .. ....... . . ... . . .. .. .... ....... .. . . . . ... .. .. . .22) y = y ( q1 . .... . ...7. . . ... . . ... .. .. . .. In the general case.. ....... .. .... . . . ..... q2... . . . .... .. ... ..... .. .. . ..... . .. .. . ....... .. . . ...... .. . .. . ...... .. .. . ... ... y . . . q2.. . y.. . .. . . . q2 . ... . .. .. . . . .... . .. . ... .......... . . . . . ... z Figure 10. . . . . . z )..... .. .... .. . .... . ... .... . . . . and φ1=1constant... . . ... .. .. . .. . . . .. .. and q31=1constant are the coordinate surfaces... ... . ... . be related to three new quantities. ..... . . . .. .. .. Curvilinear coordinates lines do not necessarily intersect at right angles... .. .. . .. . ... . . . . ... . . .. . . . ... ... the r-coordinate line is a radial line... . ... . . . . . . ..... . ....... . ... ... . . . . .. x. .. ... .. ..... . . . . ... . . . .. . ..... .. . ... . .. .... . .. z). .... . . .. ... These curves (and the surfaces) intersect at right angles.. .. . . .. and q3. .. . . .. . . .. q3 ) (10.... ........ . ..... ... q2 . . ... . ... ... .. . .. .. .. . they are perpendicular. .. . .. . . As shown in the ﬁgure.... . .. . ..... .. . . .. This is the characteristic property of the orthogonal curvilinear coordinates. .. .. . ... The surface r1=1constant is that of a sphere of radius r. . .7 The corresponding ‘axes’ through the point P are the curves of intersection of pairs of these surfaces.. . ......... . .. ... . . . ... the θ-coordinate line (curve) is the semicircle APA′. .. .. . . . .. . . . .. . .. . . . . .. .. . ... . . .... . ....... . ... .. . . with inverse relations q1 = q1 ( x .. .. .. . . .... ..... . .. . . In spherical polar coordinates. .... .... the position of P(r. . . . .. The surface θ1=1constant is that of the right circular cone with apex O and circular base BPB′ in Figure 10. . . and the intersection of three such surfaces locates a point.. . . .. ... . ... .. .... ..... ...... . . . .. .. . .. .. . . .. . y.. . . . . . ...

. . ... .... ... . . . .. . . . . .. .. .. .. . y1=1ρ1sin1φ.... . . cartesian: hx1=1hy1=1hz1=11 spherical polar: hr1=11.. . . . ..... . .. . . .. . .... . ... . ..8 ρ φ y x hρ1=11. ... .. . . . . dv1=1ρ1dρ1dφ1dz ∇2 = x ∂2 1 ∂ ∂ 1 ∂2 ρ + 2 2 + 2 ρ ∂ρ ∂ρ ρ ∂φ ∂z Figure 10.. . . . . . . .. .. . . ... .... .. . . . . . ... .. . ... ... . ... ... ... . .. .. ... ... . .... .. . . . .. . .. . . . . . .. .. z1=1−∞1→1+∞ z x1=1ρ1cos1φ. .. . ... .... ... . .... .. . ... .... .24) dv1=1ds1 ds2 ds31=1h1 h2 h3 dq1 dq2 dq3 (iii) The Laplacian operator is 1 ∂ h1h2 h3 ∂q1 h h ∂ ∂ ∂ h3h1 ∂ 2 3 h ∂q + ∂q h ∂q + ∂q 1 1 2 2 2 3 (10..... ... .... .... . . ... ... .. . ...... . .. .... .... .. .. ..23) hi2 (ii) The volume element is (10... ... . ..... ...... . Cylindrical polar coordinates . . .. . ) (P .. . . . . . . ... .. ... ... . . .. . ... . . .... .10. ... ... ..... ... ... . . . . . . .. . . ... . .......... . ... . . .... . . ...... .. . . hφ1=1r1sin1θ The following are two of the more widely-used alternative coordinate systems.. .. ... . .. .. .. . ..... . ... .. .. . . ....... .... . .. .. . .... ... . ... .... .. ...... .. . .... ..... .. ... . .... .. .. .. . . ... . .. . The cartesian and spherical polar systems are the bestknown examples.ρ • z These coordinates are the plane polar coordinates in the xy-plane plus the z-coordinate.... ... . ... z1=1z z . .. ... ... ... ....... ..... . . .6 Other coordinate systems 309 whose coordinate lines (and surfaces) are mutually perpendicular (orthogonal) at their point of intersection. . .. . . ..... .. . . . . .. .. ... . (i) The (inﬁnitesimal) distance between two points on a coordinate line is dsi1=1hi1dqi where ∂x ∂y ∂z = ∂q + ∂q + ∂q i i i 2 2 2 (10.... ... hz1=11 y ρ1=101→1∞.. . . .. ... .. ..... .. .. .. . .. .... .. .... hθ1=1r. .. . . . . . . .. .. . . . .. .. . . . ... ......... .φ . . . ..... . .. .26) 3 3 For the cartesian and spherical polar coordinates.. . .. ...... . . . . . .. ....... .... . ..... ... . .. ..... . . .. . . . . . . . . .. hφ1=1ρ.... . . . . . Such coordinate systems have the following properties.... . . . . . ... . .. . ... . . ..... . φ1=101→12π.. ... .. .25) ∇2 = h h ∂ 1 2 h ∂q (10.. . . .. .. .... ... ......... . . .... . . .. . ... ... ..... . . . .. . . . .....O .... . ..... .. . and are useful for the description of systems with cylindrical symmetry... . .... . ... . .. . .. . .. . . .. . . . ...

. .. . . . .. .. .. . .. .... .. . . . . y1=1a1sin1t. . . ..... . .. .. .. . . ... . ..... . . . . ∂ (ξ 2 − η 2 ) ∂2 ∂ ∂ 2 2 ∂ (1 − η ) + 2 (ξ − 1) + 2 2 2 ∂ξ ∂ξ ∂η ∂η (ξ − 1)(1 − η2 ) ∂φ 2 a (ξ − η ) ξ 2 − η2 ξ2 − 1 1 η = −1 → +1. ...... .. . .... . . . . ...... . .. . . . ... . .. ........ .. . .... ...... .... .... .... .. .. . ... . . . . and z1=1bt gives the displacement parallel to the axis.... .. .... . . ..... .. . ... . . . .. . . .. ... . ... . ...... ..... . . . . ... .. . . . . . . . ... . . .... .. . . ..... . . . .. . . .. . .. .. .. . z1=1bt.... ... .. . . . . .... .. . . .... .. . . ... in terms of the distances rA and rB from the foci A and B of an ellipse (the ellipse is deﬁned by rA1+1rB1=1constant): Chapter 10 Functions in 3 dimensions These coordinates (also called prolate spheroidal coordinates) are useful for twocentre potential problems..... . .. .. ... .310 The ﬁgure might be a representation of a diatomic molecule with nuclei at A and B...... . ... . . . . . . . . ... .. . .. . .... . ... ..... ...9. ....9 shows a right-handed helix. y = a ξ 2 − 1 1 − η2 sin φ .... . . .... .... . . . .. . . . .... .. ... .. ... . .. .. . .. . . .. .. ....... the displacement parallel to the axis made in one revolution about the axis.. . . . ... .. .. as shown in Figure 10. . ...... .. . .. . .. ... . ... .. .. ... .. . .. .. .. and an electron at P... . .. . . ... . ...... . .. . .. ...... . . . ... .. . ... . .. .. .. . .. . . ..... .... ..... ..... .. .. .. .. . .. ... .. .. .. .10... . . . .... .. .. ..... . . . .. .. .. .... .. .. . ... . . . .. . . .. ... . .. . . .. . . .... . . . .. . .. . ....... . .. In cylindrical polar coordinates. . .... . .... ... 2a .. .. . .. . . .... .... . .. . ..... . . . . ... .. .. .. .. . ... ... ..... .. . . ... .. ... . . . ... ..... . ....... The coordinates ξ and η are best visualized... .... .. . ... ..... . .. . . .... ...... . ... .... . .. . ...... .. ..... . .. . ........ . .. .. . .. .. ... .... .... x y . . . . ... ... . . ∇2 = ξ = 1 → ∞. . ... .... . . . ... 39 EXAMPLE 10. .... . .. .. . . . . . . .......... . .. The quantity 2πb is the pitch of the helix. . ... . . .. ....... .. ... . .. . ... . .. ... . .. and t is a parameter. . . .. . .. .. ... . .. in parametric form. . . . . . . . .. . ...... ...... .... ...13 A circular helix lies on the curved surface of a (right) circular cylinder.. ... . ... ... . . . ...... .... .. .. .... .. .. . . .. .. ...... .. .. . . . ........ . z Confocal elliptic coordinates 0 Exercises 38. . . . . .... .... . . .. . . .... . . .. .. . . . . .. .. .. . .. ........ . . ....... . ... ... .. .. .. . .. . . . .. . .. . . .. . .. .... ... .. .. ... . ... . . . ... . . . ... . .. ... . ... .. . ...... . . . .. ...... . .. .. ... .. .. . ... .. . .... .... . . . .. .... .... .. ....... .. .. . .. . ........ . ...... .... .. .. ... . . ... . . ... . are x1=1a1cos1t. .. The cartesian coordinates.. .. .. . . ... . .. ... .. .. . . . . . ... .. .... . . . .. ... . .. .. . . ... . ... .. . . .. .. . ρ1=1a is the constant radius of the cylinder. . . . .. . φ = 0 → 2π hφ = a (ξ 2 − 1) (1 − η2 ) η= rA − rB 2a z = aξη Figure 10. . as in Figure 10.. . . .... ... .. . .. ... . . .. ..... .. . ..... .. . .... . . . .. . . . . ... . .. . .. . . . . . .. .... . . . hη = a ξ= ξ 2 − η2 1 − η2 rA + rB .. .. ...... . .. . . . . with clockwise circulation around the axis. . . . . . ... . .. . ... . ... . . . ... .. . φ1=1t describes the rotation around the axis of the cylinder.. . . . . . Figure 10...... ... .. . .9 . . .... . . . ... ... .. . .. . dv = a3 (ξ 2 − η2 ) d ξ d ηd φ hξ = a x = a ξ 2 − 1 1 − η2 cos φ . . .. .. ... . .... . . . . . ... where a and b are constants. . . . .. . . .. . .. ..... . . . ... . .

.. Thus and 1sB = (1 π ) e B are normalized 1s orbitals centred on the protons at A and B (see Figure 10.. . .. . ..... Then C = 1 The integrals SAA and SBB are the normalization integrals for the 1s orbital...... . . ... . . . . . .... ... ........ .. . . ...... . . ..... .. ... .. ... ..... .. ... . .... ... .. .. . ...... ...10 2 ∞ 2(1 + SAB ) ....... . .. . ....... .. ... ....... ........ ... .. . .. .. . . . . .... . . . . . ....... . .. ..... . . .. To evaluate the overlap integral ( π)e − rA 10.. ........ .. .. . . ........ . .. .. ... . .. ... . ... .... .. . . ...... . ..... . if R1=12a is the bond length. .. . .... .. . . ... ..1η. z B r B • a o .... ... .... .... ... . .. ...... .... ... ...... . . . ... ........ . . ....... . .. .. . ........ ... .... .. . .. ... . .... .. ... .. . .. ... .. . . .. . .. ... ...... . . . .. .. .. ...... . .... . . . . .... . .. . . .... . .. η= A B R R R3 2 (ξ − η 2 ) d ξ d η d φ 8 − Rξ π 1 Z e− ( rA + rB ) dv R3 dv = Z Z Z e− Rξ (ξ 2 − η2 ) d ξ d η d φ 8π φ = 0 η =−1 ξ =1 2π +1 Figure 10.. . . ... ... ... .. . . .... ... as in constant chosen to normalize the orbital : Zψ 2 dv = 1 ..... .. .. ... . . ... . .. ... ... ... .......... and equal unity.......14 The hydrogen molecule The volume integral over all space is then a • A A r and the volume element is SAB = = 1 −r 1 − rB C2 1= Z e A+ e dv = Z e−2rA + e−2rB + 2e− ( rA + rB ) dv π π π ξ= dv = SAB = ∞ ∞ +1 +1 2π 2π R3 Z e− Rξξ 2 d ξ d η d φ − Z Z Z e− Rξη2 d ξ d η d φ Z Z 8π φ = 0 η =−1 ξ =1 φ = 0 η =−1 ξ =1 π 1 = C 2 SAA + S BB + 2 SAB Ze rA + rB r −r .. . . . .. . ..... .. ... .. . ... ...... . ......... . ..... ... ....... ... ... ... . .. . . ..... ...we use confocal elliptic coordinates (ξ...... .. .. .. P • x Example 10.. ..... . .. ... . .. .. ...... . . . ... .. .. ..... ...... .. . .... .... ... .... . . . . ..... .. .. .. .. .. . ... . ... ..... .. . . .... . ....... . . . .. ...... ... .. . . . .. . . . . ........ ... ... ... .... ... C is a ψ1=1C(1sA1+11sB)... where (in atomic units) 1sA = (1 In the method of ‘linear combinations of atomic orbitals’ (LCAO) for the ground state of the hydrogen molecule. .10). . ... . ....... .. .. . .. . ...... .. ..... ... .... . . .6 Other coordinate systems ) −r 311 ... . .. . .. .. . . ..... .... . . ... .. ... ..... .. ... . .. the simplest form of the occupied molecular orbital is EXAMPLE 10. .. ... ........ . . ... .1φ)....... . ... . .. ... . .... .. . . .. . .. .......... ...... .. . .... . . . ..... .. .. ... .. . .. .. .. . ..... ......... . ... .. .... .... . .. .. .. ... . .. . . ... .... . .. . ... Then. . . .8. . ... .. . .. . ... .... . .. .... .. . . .. . ...... .. . . . . .... .... .. . .... ... . . . . . ... .. .... ... .. . .

16) 9. centre at the origin r V: all space 20. z) of the points whose spherical polar coordinates are: 2.1φ)1=1(1. Find the value of C that normalizes the 3pz orbital.1π22.2 Find the cartesian coordinates (x. 41 ( ) 10.1−4.13π24) 1.12. (x. The 3pz orbital of the hydrogen atom is ψ3p 1=1C(61−1r)re−r231cos1θ (in atomic units) z where C is a constant.1y. 01≤1y1≤11. y.1−8) 8.1z)1=1(1. The 2s orbital of the hydrogen atom is ψ 2 s = 2 integral of ψ 2 s over all space is unity. (x.14.1φ)1=1(2. separating variables. 1 3 4 2 π a0 ( 2 − r 2 a0 )e − r 22 a0 . Express in spherical polar coordinates: (i) (x 21+1y 21+1z 2)−122.3 Express in spherical polar coordinates: 11. Show that the 22. (r.1π22) 3.1z)1=1(1.1y. z1≥10 17.4 Find the total mass of a mass distribution of density ρ in region V of space: V: the cube 01≤1x1≤11.10) 5. V: the inﬁnite region x1≥10. V: the box 01≤1x1≤1a.10) 6. centre at the origin 18. (x. 21. . ρ1=1e−(x+y+z).10.11. V: the sphere of radius a. (x.10) Find the spherical polar coordinates (r. ρ1=1x 21+1y 21+1z 2.1y. (x.1φ) of the points: 4.1−3. SAB = ∞ ∞ +1 +1 2π 2π R3 Z d φ Z d η Z e− Rξ ξ 2 d ξ − Z d φ Z η2 d η Z e− Rξ d ξ 8π 0 1 −1 −1 1 0 = e− R R3 2 2π × 2 × 3 R + 2 R + 2 R 8π ( ) − 2π × 2 × eR 3 − R = 1 + R + R 2 3 e− R 0 Exercises 40. (r.1θ. 01≤1z1≤11 14. (r.1z)1=1(−3.1y. 01≤1z1≤1c 15.312 Chapter 10 Functions in 3 dimensions Then.1z)1=1(1.1θ. 2z 21−1x 21−1y 2 Section 10. 01≤1y1≤1b. x 21−1y 2 13.1y.1−12) Section 10.7 Exercises Section 10.1y. 01≤1y1≤11. V: the region 11−1y1≤1x1≤11. ρ1=1x 2.1z)1=1(−2. ρ = . (x. (x 21+1y 2 )2z 2 10.1θ.12) 7. (ii) ∂ 2 ( x + y 2 + z 2 )−1 2 ∂x 12. ρ1=1r3e−r. 01≤1z1≤12 16. V: the sphere of radius a. ρ1=1xy 2z 3.10.12π23.1θ. ρ1=1x 21+1y 21+1z 2.1φ)1=1(2. sin 2 θ cos 2 φ 19.1z)1=1(0. y1≥10.

10.1y.1t)1=1f (x. Show that if the function f (x. z) satisﬁes Laplace’s equation then the function g(x.1y. Integrate the function f1=1y 2z 3 over the cylindrical region of radius a. If f1=1(21−1r)e−r22 show that ∇ 2 f + 31. t is the time. (ii) cylindrical polar coordinates? 39.6 38. Calculate the average value of r 2 for the 3pz orbital (see Exercise 22).13. y. satisﬁes ∇2f1=1a 2f. Section 10. 4 r 6f = f. 4 r 6f 9f = . Show that f ( r ) = 2f f = .7 Exercises 313 23. f1=1x 2y 3z4 26. φ. η = A B . What are the parametric equations of a left-handed helix in (i) cartesian coordinates. r ne−ar 27.1= = ∂2 ∂x 2 + ∂2 ∂y 2 + ∂2 ∂z 2 − 1 ∂2 c 2 ∂t 2 where x. and c is the speed of light. n1=11. xze−r22 30. y. z are coordinates. 24. If f1=1xye−r show that ∇ 2 f + 33. r n1sinn1θ1cos1nφ. The d’Alembertian operator is 2 35. cos 2φ 36.12. e − ( rA + rB ) rA + rB r −r . Calculate the average distance from the nucleus of an electron in the 2s orbital (see Exercise 21). Use confocal elliptic coordinates.1z. and symmetric about the z-axis. r e ar . e−r231sin1θ1sin1φ 29. where a is an arbitrary number. e− r r 28. to integrate the following R R rA 41. e − ( rA + rB ) rB cos 2 φ . between z1=10 and z1=11. r Show that the following functions satisfy the Laplace equation: 34. ξ = functions over all space: 40. 2x 31−13x( y 21+1z 2) r 2 sin 2 θ 37.1z)eikct satisﬁes the equation 2 g = k 2 g . If f1=1ze−3r22 show that ∇ 2 f + 32. Section 10.5 Find ∇2f: 25.

dx = kx dt dx = k ( a − x ) (b − x ) dt d 2h dt 2 d 2x dt 5. 2. and higher derivatives is a differential equation. The order of a differential equation is the order of the highest derivative in the equation. equations 1 and 2 are ﬁrst-order equations because they contain a ﬁrst derivative only. and the solution of physical problems provided much of the motivation for the further development of the calculus in the 18th century. ψ (x) in 5. and 5 are equations of motion. 2. engineering. For example. and also in biology. 4.11 First-order differential equations 11. but an equation that also contains one or more of dy2dx. the Maxwell equations in electromagnetism.1 Concepts A differential equation is an equation that contains derivatives. d 2y2dx 2. x(t) in 1. 2 ﬁrst-order rate process second-order rate process body falling under the inﬂuence of gravity classical harmonic oscillator wave equation for the harmonic oscillator = −g = −ω 2 x − 2 d 2ψ 1 2 + kx ψ = Eψ 2 m dx 2 2 The ﬁve examples describe physical processes. with representative uses in the physical sciences. equations 3 to 5 are secondorder equations. economics.1 Equations 1 and 2 are used to describe (to model) some simple rate processes in physics and chemistry. 4. Almost all the important differential equations in the physical sciences are ﬁrst order (such as elementary rate processes) or second order (equations of motion such as Newton’s second law. Examples of differential equations. The derivatives are ordinary derivatives and the equations are ordinary differential equations. and the Schrödinger equation). Equations containing partial derivatives are discussed in Chapter 14. and the social sciences. are 1. 1 Both Newton and Leibniz recognized that physical problems can be formulated in terms of differential equations. but for our purposes we consider only the order. 3. h(t) in 3. Equations 3. . In all of these the unknown function is a function of one variable only. There exists a system of classiﬁcation of the possible types of differential equations. and 4. if y is a function of x then an equation that contains y and x only is an ordinary equation.

3) (11. integration of both sides of (11.11.2) dy = 2x dx has the solution y1=1x 21+1c (11. y. because d(x 21+1c)2dx1=12x. . the solution is called the general solution. or complete integral.2) with respect to x gives Z dy dx = Z 2 x dx.3). 2 . … = 0 dx dx (11.2 that the general solution (11. The result is a particular solution of the differential equation. For example. the ﬁrst-order equation (see also Section 5. 11.4) In the form (11. and that a particular curve of the family is obtained by assigning a particular value to c. one for each value of c. To solve. and the choice of particular solution is determined by the nature and state of the system. of the differential equation.2 Solution of a differential equation 315 In this chapter we consider only ﬁrst-order differential equations. y. The differential equation is solved by performing an (indeﬁnite) integration.1) in which some combination (a function) of x. or integrate. with the arbitrary constant.2 Solution of a differential equation A differential equation containing an unknown function y(x) of the variable x has the general form dy d 2y f x . We saw in Section 5. Thus.3) represents a family of curves.2) where c is an arbitrary constant. . and the derivatives of y is equal to zero. the general solution represents a whole family of possible physical situations. second-order equations are discussed in Chapters 12 to 14. dx y = x2 + c (11. We consider ﬁrst what is meant by solving a differential equation. In the physical context. the differential equation is meant to ﬁnd the function (or family of functions) y(x) that satisﬁes the equation.

6) =2 (11. so that two arbitrary constants (constants of integration) appear in the general solution. and two conditions are required to specify the constants. If x01=1x(0) is the amount at time t1=10 then a1=1x0 and x(t)1=1x01e−kt is the particular solution.* * Some differential equations have no general solution. containing a second derivative. We note that the application of the initial condition speciﬁes the signiﬁcance of the arbitrary constant as well as its value. In this case. the second-order equation d2y dx 2 is integrated once to give dy = 2x + a dx and a second time to give y1=1x21+1ax1+1b (11.5) This is the general solution. .7) (11. where x(t) is the amount of reacting substance at time t. It is shown in Section 11.1 The equation dx = − kx dt provides a model for a ﬁrst-order rate process such as radioactive decay or a ﬁrstorder chemical reaction. the constant is speciﬁed by the amount of substance at any particular time.4 that the general solution of the equation is x(t)1=1ae−kt where a is the arbitrary constant. In the case of a second-order equation. For example. and is a necessary step in the solution of a physical problem. some have one or more particular solutions only. The general solution of an nth-order differential equation contains n arbitrary constants.316 Chapter 11 First-order differential equations EXAMPLE 11. Such cases are not often met in the physical sciences. some have a general solution and some particular solutions not obtained from the general solution. two integrations are required to remove the second derivative.

dx We have dy = − e− x.3 A body falling under the inﬂuence of gravity Let h be the height of the body at time t. We need to integrate twice: dy d 2y 1 = Z 2 dx = Zsin 2 x dx = − cos 2 x + a dx 2 dx y=Z 1 1 dy dx = Z − cos 2 x + a dx = − sin 2 x + ax + b 2 4 dx 0 Exercises 1–8 EXAMPLE 11.11. Then a1=1v0. The acceleration experienced by the body is therefore d 2h dt 2 = −g The ﬁrst integration gives the velocity v(t ) = dh = − gt + a dt and the second integration gives the height at time t 1 h(t ) = − gt 2 + at + b 2 A particular solution in this case is given by the initial conditions for the system. and v(t)1=1−gt1+1v0. 0 Exercise 9 1 h(t ) = − gt 2 + v 0t + h0 2 . Therefore dx dy + y − 1 = − e− x + e− x + 1 − 1 = 0 dx (ii) Find the general solution of the second-order differential equation d 2y dx 2 = sin 2 x . The (downward) force acting on the body is F1=1−mg and.2 Solution of a differential equation 317 EXAMPLE 11. F1=1md 2h2dt 2.2 (i) Show that y1=1e−x1+11 is a solution of the ﬁrst-order differential equation dy + y − 1 = 0. b1=1h0. the height h0 and the velocity v0 at time t1=10. by Newton’s second law.

More formally. There are however several important standard types whose solutions can be expressed in terms of elementary functions and that frequently occur in mathematical models of physical systems.12) Z g( y) dy dx = Z f ( x ) dx + c dx (11. writing (11. We discuss here two important special types of equations that can be solved by elementary methods. The method of separation of variables and the method of reducing a homogeneous equation to separable form discussed in this section were invented by Leibniz in 1691.318 Chapter 11 First-order differential equations All ordinary differential equations can be solved by the numerical methods discussed in Chapter 20. . The general ﬁrst-order differential equation has the form dy = F ( x. Such a differential equation together with an initial condition. separable equations and linear equations.10) (11. y(x0)1=1y0 (y1=1y0 when x1=1x0).8) can be solved by straightforward integration if it has.9) with all the terms involving y on one side of the equal sign and all the terms involving x on the other. It is these standard types that are discussed in this and the following chapters. or if it can be reduced to.8) where F(x.9) as g( y) integration with respect to x gives dy = f ( x) dx (11.13) 2 Much of Leibniz’s formulation of the calculus involved equations containing differentials (differential equations).3 Separable equations Equation (11. y ) dx (11.11) with the arbitrary constant included on one side. the form dy f ( x ) = dx g ( y ) This equation can be written in the differential form2 g( y) dy1=1f(x) dx (11. and y is a function of x. 0 Exercises 10–12 11. is called an initial value problem. Such an equation is called a separable differential equation and is solved by (indeﬁnite) integration of each side with respect to the relevant variable: Z g ( y ) dy = Z f ( x ) dx + c (11. y) is a function of x and y.

4. By equation (11. Therefore. (ii) dy = ky . by the method of substitution for the left side (Chapter 6).3 Separable equations 319 and. y(x0)1=1y01=1ae kx0 so that a1=1y0e −kx0. Then y1=1ae kx By the initial condition. On the left side. y ln y = kx + c An alternative form of the general solution is obtained by taking the exponential of each side of the equation. dx Separation of the variables and integration gives dy = k dx.9) with solution (11.10). The solution of the differential equation is completed by the evaluation of the integrals in (11.4 Separable equations Each of the following equations has the form (11. y( x ) = ax n +1 +c n +1 This is the general solution. on the right side. (i) dy = ax n . dy1=1ax n1dx. y Z dy = k Z dx. EXAMPLE 11. The following examples demonstrate the method. eln y1=1y. An initial condition is given in each case. Therefore c1=10 and y( x ) = ax n +1 n +1 y ( x0 ) = y0 . this is identical to (11. and the particular solution is y(x)1=1y0e k(x−x0) . that y1=10 when x1=10. y1=1y0 when x1=1x0. dx y (0) = 0. and integration gives (when n1≠1−1) Z dy = a Z x n dx. otherwise numerical methods must be used) and by the application of the appropriate initial condition. The required particular solution is obtained by applying the initial condition.11). e kx+c1=1e c1e kx1=1ae kx.11) by the methods discussed in Chapters 5 and 6 (if possible.11.11). Applications of separable differential equations are discussed in Section 11.

−Z dy y2 = 3 Zx 2 dx. y) (11. dx y (1) = 1 .14) where λ is an arbitrary number or function. A function of x and y has this property if it can be expressed as a function of y2x. 2 Separation of the variables and integration gives − dy y2 = 3x 2 dx . y) on the right of the general form (11.320 Chapter 11 First-order differential equations (iii) dy + 3x 2 y 2 = 0. y ) = Then λ y y F (λ x. y1=1122 when x1=11. λ y ) = f = f = F ( x. y (1) = 1 1 = 2 1+ c so that c1=11 and the required particular solution is y( x ) = 1 x +1 3 0 Exercises 13–22 Reduction to separable form Some ﬁrst-order differential equations that are not separable can be made so by means of a suitable change of variables. For example. F ( x. 1 = x3 + c y and the general solution is y( x ) = 1 x +c 3 By the initial condition. that is. λy)1=1F(x. y ) λx x (11. ‘scaling’ both x and y by the same factor leaves the function unchanged.8) has the property F(λ x. One such example is when the function F(x.15) y y x2 + y2 y = +1 = f xy x x x .

Then y = xu. (11. dx u Then. =) The function (11. Z u du = Z dx . dy du =u+x = f (u ) dx dx (11.11.17) (11.18) Z du = ln x + c f (u ) − u (11.16) so that. = dx xy The substitution y1=1xu gives 1 du dy =u+x =u+ dx u dx so that x du 1 = . y.14) is therefore homogeneous of zero degree.19) An equation of this type is often called a homogeneous equation because a function of type (11. x u2 = ln x + c 2 . du dx = f (u ) − u x and the solution of the equation is (11.20) EXAMPLE 11. separating the variables and integrating. y ) = f dx x is made separable by means of the substitution u1=1y2x.5 Solve dy x 2 + y 2 .3 Separable equations 321 The differential equation y dy = F ( x. λy. =)1=1λnf(x. In general.14) is a homogeneous function. a function of one or more variables is a homogeneous function of degree n if f(λ x. separating the variables x and u.

22) in which the sum is over all the species present. are the numbers a. called stoichiometric numbers. but with a change of sign for the reactants. = in (11. q. b. = and p. 2 x 2 dx y dy y 2 1 − = x 2 dx x 3 x 11. q molecules of Q. is therefore y2 2x2 = ln x + c The result can be checked by implicit differentiation: d dx and dy x 2 1 y 2 x y x 2 + y 2 + = + = = dx y x x3 y x xy 0 Exercises 23–25 0 Exercises 26–28 y2 d = (ln x + c ). B. and the νB.21). = in the equation means that a molecules of A are reacting with b molecules of B. Q.21) in which the symbols A. q. both reactants and products. after substituting u1=1y2x.4 Separable equations in chemical kinetics A chemical reaction is denoted by a ‘stoichiometric’ equation of the form aA1+1bB1+1-1=1pP1+1qQ1+1(11. = The presence of the numbers a.322 Chapter 11 First-order differential equations The general solution of the equation. = For example 2H21+1O21=12H2O means that two molecules of hydrogen react with one molecule of oxygen to give two molecules of water.21) is 01=1−aA1−1bB1+1pP1+1qQ1+1=1νA1A1+1νB1B1+1νP1P1+1νQ1Q1+1- . p. = (in practice. = to give p molecules of P. chemical formulas) represent the chemical species that are reacting to give the products P. b. In this form equation (11. the general chemical reaction is written 0 = ∑ νB B B (11. In thermodynamics and in kinetics.

the partial pressure pA for a gas.21) we have v=− 1 d [A] 1 d [ B] 1 d [ P] 1 d [Q] =− = = = a dt b dt p dt q dt (11. and this is also the unit of ξ.11.* The concentration of species A in volume V is given by [A]1=1nA2V. deﬁned by nA1=1nA01+1νA1ξ (11. . but omit the units. and division of equation (11. so that r= 1 dnA dξ = dt ν A dt (11.26) by the volume gives the rate of reaction (in terms of the concentration of A) v= r 1 d [A] = V ν A dt (11. We follow the normal practice in physical chemistry texts of using concentrations for our discussion of kinetics.25) For many purposes in physical chemistry it is more convenient to express the rate in terms of concentrations instead of amounts. The ‘amount’ of reaction that has occurred in time t is given by the extent of reaction ξ. r= dξ dt (11.24) This rate can also be expressed in terms of the rate of change of the amount of each species taking part in the reaction.28) * By convention. at time t1=10.26) (11.23). Other quantities used instead of amount are the concentration [A]. dnA dξ = νA dt dt (since nA0 and νA are constants). and let nA be the amount at time t.23) and a measure of the rate of a reaction is then the rate of change of the extent of reaction.27) For the general reaction written in the form (11.4 Separable equations in chemical kinetics 323 and the H22O2 equation is 01=1−2H21−1O21+12H2O Rate of reaction Let nA0 be the initial amount of species A. and the dimensionless mole fraction xA. or rate of conversion. the symbol nA is the amount of species A in the (SI) unit mol. differentiating (11.

324 Chapter 11 First-order differential equations and for the H22O2 reaction.4(ii). β. or both.32) This is the differential equation discussed in Examples 3. the pressure. =.33) .1. that may be consecutive. we represent such a reaction by aA1+1bB1+1-1→ products It is found from experiment that the rate of the reaction has the form v1=1k[A]α[B]β= (11. separating the variables and integrating. β is the order with respect to B.30) (11. as well as on the temperature. x ln1x1=1−kt1+1c or ln[A]1=1−kt1+1c (11. and the numbers α.29) where k is called the rate constant of the reaction. The cases considered are (1) A1→1products (2) 2A1→1products (3) A1+1B1→1products (1) ﬁrst order second order second order The ﬁrst-order process: A1→1products The rate equation is v=− d [A] = k [A] dt (11. dx = − kx dt (11. and the nature of the container. both reactant and product. It is found by experiment that the overall rate may depend on the amounts of all the chemical species present at any time. v=− d [O 2 ] 1 d [ H 2O] 1 d [H 2 ] =− = dt 2 dt 2 dt The mechanism of a chemical reaction is in general complex. competitive. proceeding through several elementary kinetic steps. if the concentration [A] is represented by the variable x. those whose rates depend only on the amounts of the reactants. Thus. We consider here only the very simplest reactions. dx = − k dt . deﬁne the order of the reaction. =.20 and 11. and the solution is that given in Example 11. and it is the task of the kineticist to unravel the mechanism and to relate the rate of the overall chemical reaction to the rates of the elementary steps.is called the order of the reaction. and the sum α1+1β1+1.31) or. The exponent α is called the order with respect to reactant A.

. . . . .. ... ... . . . k 0 Exercise 29 (2) The second-order process: 2A1→1products The rate equation is v=− or.. .... . . .. dx = −2 kx 2 dt (11. ... .. . .. from (11... . . k− • 0] 0 A [nl A plot of ln[A] against t is a straight line (Figure 11. . ...... . . . ... .. . . . .35). . ..... . . .. .. .34) by taking the exponential of each side. . . . The more conventional expression for the integrated rate equation is obtained from (11. . .... . . . then c1=1ln[A]0 and the solution of the differential rate equation is the integrated rate equation ln[A]1=1−kt1+1ln[A]0 ] A [nl (11... .. ..1) whose slope is −k and whose intercept with the ln[A] axis is ln[A]0. ... . . . . . . . . . .. . ..... . . .36) t . ..1 and shows that the ﬁrst-order process is an example of exponential decay... . ..4 Separable equations in chemical kinetics 325 If the initial concentration of A is [A]0.... .. . . .. .. .. . ... . .. . . . . . τ1 2 = ln 2 . [A]0 e Therefore e − k τ1 2 − k ( t + τ1 2 ) 1 = [A]0 e− kt 2 = 1 2 and. . .. . . .... . . . .34) [A]1=1[A]01e−kt (11.. The half-life The half-life τ122 of an exponential decay is the interval in which the amount of reactant is halved: x (t + τ1 2 ) = 1 x (t ) 2 or. taking logs. . .11. ... .37) 1 d [A] = k [A]2 2 dt (11.. . .. . .... . . . .. . . . .. . .. .35) Figure 11.. .. . . at time t1=10. . . . ..... .

. .. . .. . The rate equation is then − d ( a − x ) dx = = k ( a − x )( b − x ) dt dt (11. .. . . . .. a1=1b.38) / 0 1 Figure 11.. . . (a) The initial concentrations of A and B are equal. . .2 (3) The second-order process: A1+1B1→1products The rate equation is v=− d [A] d [ B] =− = k [A][ B] dt dt (11... .. .42) = k dt . ..43) Then c1=112a1=112[A]0 and the integrated rate equation is 1 1 − = kt [A] [A]0 (11.. ... 1 = 2 kt + c x If x0 is the initial concentration of reactant then c1=112x0 and 1 1 − = 2 kt x x0 or 1 1 − = 2 kt . . .. .. . . .. k 2 . .. separating the variables and integrating.. . ... . ...... .. . ... ] A[ (11.. . . ... The rate equation (11. . . .. ... ... .. and let [A]1=1a1− x and [B]1=1b1− x at time t. ... .. . . . . . . . .. 1 1 = = kt + c a − x [A] (11. [A]01=1[B]0.40) Let the initial concentrations of A and B be [A]01=1a and [B]01=1b. .326 Chapter 11 First-order differential equations Separation of the variables and integration gives − dx x 2 = 2 k dt .... . .41) is dx = k ( a − x )2 dt and. dx (a − x) 2 (11.. . . .. ..... • 0] A[ / 1 .. . .. . .44) t A plot of 12[A] against t is a straight line (Figure 11... respectively. [A] [A]0 [A] = [A]0 1 + 2 kt[A]0 (11. .. ...39) .. . .41) Two cases need to be distinguished. .2) whose slope is 2k and whose intercept with the vertical axis is 12[A]0. ... .. ..... . .

which distinguishes case A1+1A from case A1+1B.11.6. Write 1 A B A( b − x ) + B( a − x ) = + = ( a − x )( b − x ) a − x b − x ( a − x ) (b − x ) It is required that 11=1A(b1−1x)1+1B(a1−1x) be true for all values of x. Separating the variables in (11. dx = k dt . and c = .45) At time t1=10. Setting x1=1b gives B1=112(a1−1b) and setting x1=1a gives A1=112(b1−1a). ( a − x )( b − x ) Z dx = kt + c ( a − x )( b − x ) The integral over the variable x is evaluated by expressing the integrand in partial fractions (see Sections 2. and Examples 2. Therefore 1 1 1 1 = − ( a − x ) (b − x ) b − a a − x b − x and Z 1 dx = ( a − x ) (b − x ) b − a = Z 1 1 − dx a − x b − x b− x 1 1 − ln ( a − x ) + ln ( b − x ) = ln b−a b − a a − x ( ) Therefore b − x 1 ln = kt + c b − a a − x b 1 ln .41) and integrating.28 and 6.7 and 6. x1=10. (b) The initial concentrations of A and B are not equal.17).4 Separable equations in chemical kinetics 327 This is the same as in case (2) above except for the factor 2. so that b − a a 1 a( b − x ) ln = kt b − a b( a − x ) (11.

47) is multiplied throughout by the function F ( x) = e ∫ which. has the differential property3 dF ( x ) = F ( x ) p( x ) dx (11.48) and. given that d dx p ( x ) dx (11.50) Z p( x ) dx = p( x ) . When the right side.47) The equation is linear in both dy2dx and y. r(x). y ln y = − Z p( x ) dx + c (11. otherwise it is an inhomogeneous equation. taking the exponential of each side. The homogeneous equation dy + p( x ) y = 0 dx The equation is separable: dy = − p( x ) dx . the general solution is − p ( x ) dx y = ae ∫ (11.51) 3 This method of solving the general ﬁrst-order differential equation was discovered by Leibniz in 1694. The inhomogeneous equation The general linear equation is transformed into a form that can be integrated directly if (11.49) where a is the arbitrary constant (if a is chosen as zero then y1=10 is called the trivial solution). is zero the equation is called homogeneous.46) 11. 0 Exercises 30–32 (11. .5 First-order linear equations dy + p( x ) y = r ( x ) dx A ﬁrst-order differential equation is called linear if it can be written in the form (11.328 Chapter 11 First-order differential equations and the integrated rate equation is [A]0 [ B] 1 = kt ln [ B]0 − [A]0 [ B]0 [A] A plot of ln[B]2[A] against t is a straight line.

53). by virtue of (11. and the integrating factor is = e− x (the constant of integration need not be include in the exponent because it cancels in (11.6 Linear equations (i) dy − y = 3e2 x dx In this case. e− x y = Z e− x × 3e2 x dx + c = 3 Z e x dx + c = 3e x + c y1=1ex(3ex1+1c)1=13e2x1+1cex (ii) dy 2 + y = 3x 3 dx x The integrating factor is F ( x) = e ∫ ( 2 / x ) dx = e2 ln x = x 2 .52) can therefore be written as d F ( x ) y = F ( x )r ( x ) dx and can be integrated to give F ( x ) y = Z F ( x )r ( x ) dx + c (11.5 First-order linear equations 329 Then.51). F ( x) dy + F ( x ) p( x ) y = F ( x ) r ( x ) dx (11.47) by F(x).52) and. Then. The function F(x) is called the integrating factor of the differential equation. multiplying (11.53) This is the general solution of the inhomogeneous equation.53)).11. p(x)1=1−1. the left side of this is the derivative of the product F(x)y: dy dF ( x ) dy d F ( x ) y = F ( x ) dx + dx y = F ( x ) dx + F ( x ) p( x ) y dx Equation (11. by formula (11. F ( x ) = e∫ p ( x ) dx Z p( x ) dx = − x . EXAMPLES 11.

The process is modelled by the two ﬁrst-order equations d [A] = − k1[A] dt d [ B] = k1[A] − k2 [ B] dt (11. and numerical methods must otherwise be used.54) (11. [B]01=10. x 2 y = Z x 2 × 3x 3 dx + c = y= x4 c + 2 x2 x6 +c 2 0 Exercises 33–40 11.56) (11.57) gives dy −k t = ak1 e 1 − k2 y dt (11. Such a system of equations can be solved only in simple cases without resorting to approximations.58) (11. and [C]01=10. Then d (a − x) = − k1 ( a − x ) dt dy = k1 ( a − x ) − k2 y dt The ﬁrst of these equations is the separable equation (11. at time t1=10. and a description of the overall rate process then involves several simultaneous ﬁrst-order differential equations. are [A]01=1a. most) chemical reactions proceed through several elementary steps. One of the simplest solvable systems is that of two consecutive ﬁrst-order processes.53).4 that many (in fact. by formula (11.330 Chapter 11 First-order differential equations Therefore. and that the concentrations at time t are [A]1=1a1−1x and [B]1=1y. 1 2 A B C → → k k in which A is transformed into C through the intermediate species B.6 An example of linear equations in chemical kinetics We saw in Section 11.32). with solution a1−1x1=1ae−k1 t Substitution of this result into equation (11. so that [C]1=1x1−1y.57) .55) We assume that the initial concentrations.

shows how the system behaves when k21>>1k1. We have ak1 − c = k2 − k1 0 and the required particular solution is ak −k t −k t 1 e 1 −e 2 y = k2 − k1 − k1t ak1te if k1 ≠ k2 if k1 = k2 ( ) if k1 ≠ k2 if k1 = k2 (11. and the . the rate-determining step is A1→1B. y1=10 when t1=10.61) [C]1=1[A]01−1[A]1−1[B] Figure 11. r(t)1=1ak1e−k1 t. Figure (a).59) is therefore F (t ) y = Z F (t )r (t ) dt + c e k2 t y = ak1Ze ( k2 − k1 )t dt + c = ak1 ( k2 – k1 ) t e +c k2 − k1 ak1t + c if k1 ≠ k2 if k1 = k2 By the initial conditions. and integrating factor F (t ) = e ∫ p ( t ) dt (11. with k22k11=110.3 illustrates the three types of behaviour exhibited by the system. In each case [A]01=11 and k11=11 (in some appropriate units).59) =e k2 t The solution of equation (11.6 An example of linear equations in chemical kinetics 331 or dy −k t + k2 y = ak1 e 1 dt This is a linear equation with p(t)1=1k2.60) The concentration of each species present at time t is therefore [A]1=1[A]0e−k1 t [A] k − k t −k t 0 1 e 1 −e 2 k −k [ B] = 2 1 [A] k te− k1t 0 1 ( ) if k1 ≠ k2 if k1 = k2 (11.11.

+.... .. . . . . . . .. ... .. . .. ..... . .. . . . . .. .. . ...... ..... ... . ..... The total drop in voltage is equal to the e.. .. . .. ... .. ... ..... +.. . .......... .. . .... .. . .. .. . . ........ .. .......... . .. .... . ... . . .... . ... . .. .. . .... . . ..... . .... . .. . ... .... . .. .... . . . . . ........ ..... . .. .. ... . . ..... .. . . . . ... ... ....... . .. . .. ... . ..... . . .. . .. . . ... .. . .. .. .. .. • +.3 EL = L dI dt (11. . . .. .. 1=1 a k )( a b c tnu om A amount of intermediate B remains small throughout the reaction because B is rapidly converted into C.... . . . .. ... ... ... ... . . ... .. .. ... .. .. .... ..... ...... ... .. .. ... ... .. ... ... . of the source is equal to the difference of electric potential between the two points at which it is attached to the circuit. . . .. . .. .... . .. . .... ... . .. . . . .. ....4 a drop in voltage (electric potential) in the circuit. .. .. ... ... .. . . .. . ... .. .. .... ... .... . .. . . .. .. . . . . . ... . .... . ..... . ..... .. ..... ..... . .. ... .. ... . .... . . ........... . .. ..... .. .. .... ... .. .. .. .. .. . ...... ... ..f. .. .. .... ..... . ... ..... . .. ... . .. .. .... .. .. .. . . . .. .. . ... ...... . . . ... . . . .. . .. .. .. . .. .. . .... .. ... . . .. . ... . .. .... ... . . . .. . .. .. .... ... .. c b c a ....f. .. .. . . .... . +.. .. ... . . .. .. . ....... . . . .. .. . .... Figure (c).. . . ... ... .... .. .. . . ........ . . . .. . .... .... .. . such as a generator or a battery... . . ... ... . .. . ..... .. . . ..... ... . .. . . . . .. .. .... ..... . .... .. . . .. .. ... . . . . . ... . . ..4) connected in series..... ...... .. .... . . .. .. .. . .. . . .... . . .... ...... .... . .. . .. . .. .. . .. .. .... . . . .. ... . . . . .. . ... .. . .. . .. . . .... ....... . .. . ... .. .m... .. . . . .. . ... . ... .. .. . .. . .. . . . . ... ... .. .. . .. . .. ....... . .. .... . .... ........ . . ... .. . .... .. .... .. .. . . . . .. . . . .. .. .. .. . . . . .... . ..... . . .. . .... .. . . ... .... .. . ...... .. . .. .. .... . .. . ....... ... . • • L . .. ......m.. .. . . .. . .. ... . . . . ......... .... . .. . . . .. . . . . .. . . . ... . . .. .. ... . .. ....... . . .... . ... . ... .. .. .. .. . ... . . . .. .. ........f......... . . .. . . ..... ... .. . . .... .. .. .. . ... R Electric circuits ER1=1RI (Ohm’s law) Figure 11.. .. .. ... . ....... . .. ..... . . ... . . ... . ....... . .... . .. . .... . .. ..... .... .. ... . ..... . . . ... .. . . . . .... . ... .. ... .. .. . . .. ... .. . .. .. .. . .. ......... . .. ... . .. . .... .. . . .... with k12k21=110.. .. . . ... and the amount of B becomes large because the process B1→1C is slow.. .... . ... .... ... . . .. . .. ..... .. . ... . . ...... .. .. .. .. ...... . .. . . . a current I ﬂows and each element of the circuit causes Figure 11... .. .. An inductor is a coil which builds up a magnetic ﬁeld within itself when a current ﬂows.. . .. . . .. . . . . ..... .... ... . . . . ...... .. .... . . ... . .... . . ...... . . The voltage drop ER across a resistor is proportional to the current.. ... ..... . .. ... .. .. ... .. . . .... . ... .. .. .. . ... ..... ...62) (11... .... . .. . .. ... .. .... .. . ...... .. .... ... .. ...... . .. . . . .. . . ... . .. . . . ... . . .. .... . .. ..... that generates an electromotive force (e.... . .. . and one or more elements (Figure 11... ... ...... . . . . ... ... . . .........1 and 1. . . .. . • .... . . . .. .. . . .. . . .. .... . . .. . . .. ... and B decreases only when the supply of A has been exhausted. .... ... . .... .. ..... . . ... .. ... .. .. . . . .. ..... .. .. . . .. .. . . .. . . . .. ... .... . . . . . . .. ... .. . .... . . .... . ... .. . .... ... . . . ... .. ..... .. .... . ... . . .. .. ... .. ... ... .. . . . .... . . . . ..... . .. .. . .... .... . . . .. . .. 1 0=2 t k . .. .. . .. .. .. .......... .. . . . . .... ... .... .. .... . .. . ... .. . ... and capacitors (C). . .. .. .. .. ....... .. ... . . . . ... .. . . . . . . ... . . .... .. +.. .. ........ ... . .... . ... ...... . . . ... ...7 Chapter 11 First-order differential equations and the constant of proportionality R is called the resistance of the resistor (see Tables 1.. . ... The e. . . ... ... .... ..... .. .. . .... . . ... .. .. . .... ....... ..... . ........ .. . .. . .. ..... .. . ... . . . . . ... ..... .. ... . .. . .. .... . .. .. .. ... .. ... . ..m.. .... . . . .. .... . 1=1 k )( c 1=2 =1 k k )b( 0 1=2 k . .. .. . .. .... .... .. .... .. ... . . . . ... . . .. ... . .... . ... .. .... . .. . . .. .. ..... .... ..... . . ...... .. .. . . .... . .... .... . . ... . . ..... . . ... . .. .... . . ... .. . .. . .. . . . ..) E....... . A resistor is a circuit element that offers resistance to the ﬂow of charge in the circuit. . . .. .. .. .. ..... . . .. .. ... .. .. .... . . . .. . ..... . ...... . . ...... ..... . .. ... . .... ... . . . . .. . . . ........ .... .. .. .... .. . . . .... . . . .. . 42 The modelling of electric circuits is an important application of linear differential equations. .. . . .. ... .... ... ..... . . .... ... .... ... ... .. ... .... ... . .. .. .... .... .. . ..2 for the appropriate SI units).. .. . ..... A simple circuit consists of a source of electric energy.. ... ... .. . . .. .. . . . ... .... ....... inductors (L).... .... A changing magnetic ﬁeld develops an induced voltage across the coil that is proportional to the rate of change of current and opposes the current ﬂow: Q ...... .. . ..... .. . . ...... resistors (R).... . .... . . .. .. ...... .. .. . .. ... ... ... . . .. .... .. ... ....... . . ... ... . .. ... . . .. .... .. . . ... . . ... . . . ..... ... . . ... .. . . . . .. . .... . ... .. .. .. . .. .. ....... ... . . .. ... .... .. ..... . . . . . .... . ...... .. .. . ........ . . . . . .. . .. ... .... . . ... ..... .... .. .... . .. . ..... . ..... .. . .. ..... ... .. . .. .. .. .. .. . .. . . .. . ...... ......... . ............... .. ... ......... .. ... . .. ........ . ... .. 11.. . .. Q .. . ... .. . . .... . .... ... .... . . . . .. . • • C . . ...... .. ... . When the circuit is closed. ... .. ... . . ... . . .... . .. . ... . . . ... ... . . . . . . .. .. . ... . .. .... .. .. . . .... .... . . . b a 332 0 Exercises 41. . .... .. ...... .. . . . ...... .. the rate-determining step is B1→1C. . . .. . .. .. . .... .... shows the behaviour when k11>>1k2.. . ... . . . .... (Kirchhoff’s voltage law)... . . . . . . . .. .. . . . .. . .. . .. ... .. .... ... . ....... ... .. . .. . .. . .. . .... .. . . .. .... .... ..... . .. . . .. .63) 0 1 . . . .. ...... . . ... . . . . .. .. ..... .

.. . . . .11.... .7 Electric circuits 333 where L is the inductance. is constant..... . ..m..... . .5 ( ) When the circuit is closed the current increases from zero to its steady value E02R in a time that depends on the inductive time constant τL1=1L2R.. .. The ( R L ) dt integrating factor is F ( x ) = e ∫ = e Rt L and the general solution of the equation is E I (t ) = e− Rt / L Ze Rt / L dt + c L If the e.... . . .. . ... . . . . ... . . .. .. . . . . .... . ...... .. .. .. .... . . . . ..... .... . . E1=1E0... . . . . . . . . .. .64) EXAMPLE 11. . By Kirchhoff ’s law. The plates become charged dQ = I . . . E (11. .. ......... .. We have p1=1R2L (constant) and r1=1E2L in equation (11.. .. ... . . . . .. ... . . ... . .. . . .. .. . ..... ..7 RL circuit .66) (11... .. . .. ..67) (11... .. .. . . . ..... . .... . .65) R • • An RL-circuit containing one resistor and one inductor is illustrated in Figure 11. .. . . .. .. . . . . . the accumulated charge is when a current ﬂows.. . . . . . ...... .... . . . .. . L and this is a ﬁrst-order linear differential equation..... 44 L dI + RI = E dt (11. ... .. . . ... .. ... . Q C (11. . .. . ... .. ... ... . . .. . .5... ... . . . . ... . .. ....... then E E I (t ) = e− Rt / L 0 Ze Rt / L dt + c = 0 + ce− Rt / L R L The particular solution for the initial condition I(0)1=10 is then I (t ) = E0 1 − e− Rt / L R Figure 11....... . . . . . . . . . . . and the potential difference between the plates is EC = where C is the capacitance. . 0 Exercises 43... .. ..f.. .... .. . .. .. . . . . .. ...68) .. ... . . . . . .. . . . ... . .. ... .... .. ... . .. ..... .. . .47). . The rate of charging is dt Q = Z I dt.. ... . . . .. . . ... ... ..... . .. . A simple capacitor is made up of two parallel metallic plates separated by a thin layer of insulating material. . . . .. . .. . .... . . . . ... .. . . . . ...

4. Verify that the given function is a solution of the differential equation.2 State the order of the differential equation and verify that the given function is a solution: 1. dx dy + 2 y + 2 = 0. 7. y(1)1=12 y(2)1=10 24. dx x − 3 y(0)1=11 y(2)1=11 20. 18. dy = 3x 2 y dx dy y = dx x 16. dx d3y dx 3 y = e2 x − 1 2. y 2 Solve the initial value problems: 19. y1=12x 31+13x 21+14x1+15 dy 3 y + = 3x 2. dy x 2 − 1 = . 2 xy dy = − ( x 2 + y 2 ). (ii) Find the solution that satisﬁes the initial conditions x(0)1=10 and H(0)1=11. dy − 2 y = 2. y(0)1=14 Section 11. . dy = 2 y. dx 26.3 Find the general solution of the differential equation: 13. 12. dy 3x 2 = dx y dy = ex dx 14. x 11. dx 2 y + 1 dy = e x+ y . dx x Solve the initial value problems: 23. dx x ( x − 1) dy x + y = . xy 3 dy = x 4 + y 4. dy y + 2 = . A body moves along the x direction under the inﬂuence of the force F(t)1=1cos12πt.334 Chapter 11 First-order differential equations 11. d2y dx 2 + 4 y = 0. dx y1=1ce−2x1−11. and determine the particular solution for the given initial condition: 10. 22. dx y(1)1=10 25. dx y(0)1=1−1 y(0)1=10 dy y ( y + 1) = . 6. dx x d3y Find the general solution of the differential equation: = 24 x dx 2 dx 3 9. 3. (i) Write down the equation of motion. y1=12 when x1=12 2 dy = x2 dx dy = e−3x dx d2y = cos 3x dy + 2 xy = 0. y1=124 when x1=12 y = ce− x . dy = 4 xy 2 dx dy = y ( y − 1) dx 15. dx y1=1cx 2. 5. y1=1A1cos12x1+1B1sin12x y= x3 c + 2 x3 = 12. 21. where t is the time. 17.8 Exercises Section 11. Show that a differential equation of the form dy = f ( ax + by + c ) dx is reduced to separable form by means of the substitution u1=1ax1+1by. 8.

ﬁnd the amount of C as a function of t. Given the initial conditions x1=1y1=1z1=10 at t1=10. and the system is modelled by the equations → → k k k d (a − x) = − k1 ( a − x ). dt dz = k2 y . 37. and C. 39. B. k21≠1k3.4 29. at time t. dy + 3 y = e−3x dx dy + ( 2 tan x ) y = sin x dx dy y + a = xn dx x Section 11. 1 → 42.8 Exercises 335 Use the method in Exercise 26 to ﬁnd the general solution: 27. dx 34. dt dy = k1 ( a − x ) − k2 y .11. and z are the amounts of A. The system of three consecutive ﬁrst-order processes A B C D is modelled by the set of equations k k k d (a − x) = − k1 ( a − x ). dy x− y = dx x − y + 2 Section 11. 32. dt dz = k 2 y − k3 z dt where (a1−1x).6 1 2 3 → → → 41. k11≠1k3. dt du = k3 y dt . Show that the solution that satisﬁes the initial condition x(0)1=10 is given by kt = 1 ( 2 a − b)2 ln a( b − 2 x ) 2x + b( a − x ) b( 2 a − b) ( b − 2 x ) Section 11.5 Find the general solution: 33. 40. respectively. 38. y. has rate equation dx = k1 ( a − x ) − k−1x dt Find x(t) for initial condition x(0)1=10. Find the interval τ12n in which the amount of reactant in a ﬁrst-order decay process is reduced by factor n. dt dy = k1 ( a − x ) − ( k2 + k3 ) y. 30. Solve the initial value problem for the nth-order kinetic process A1→1products dx = − kx n dt x(0)1=1a (n1>11) 31. The reversible reaction A1j1B. Assume k11≠1k2. respectively. dy = 2x + y + 3 dx 28. (n1≠1−1) dy − 4 xy = x dx dy y 4 − = dx x 2 x 2 35. dy + 2y = 4 dx dy 2 y + = 2 cos x dx x dy + ax n y = bx n . A third-order process A1+12B1→1products has rate equation dx = k ( a − x ) ( b − 2 x )2 dt where a and b are the initial amounts of A and B. ﬁrst order in both directions. The ﬁrst-order process A B is followed by the parallel ﬁrst-order processes 2 3 B C and B D . 36.

z and u are the amounts of A. 44.f.336 Chapter 11 First-order differential equations where (a1−1x). respectively. E(t)1=1E01sin1ω t. The current in an RL-circuit containing one resistor and one inductor is given by the equation L dI + RI = E dt Solve the equation for a periodic e.7 43.m.. B. and D. E1=1E0. (ii) a periodic e.. E(t)1=1E01sin1ω t. y. C. . with initial condition I(0)1=10.m.f. Given the initial conditions x1=1y1=1z1=1u1=10 at t1=10 ﬁnd the amount of C as a function of time.m. at time t. Show that the current in an RC-circuit containing one resistor and one capacitor is given by the equation R dI I dE + = dt C dt Solve the equation for (i) a constant e.f. Section 11.

Examples of such equations are the classical equations of motion for the harmonic oscillator and for forced oscillations in mechanical and electrical systems.1) When the function r(x) is zero the equation d2y dx 2 + p( x ) dy + q( x ) y = 0 dx (12. between particular and general solutions. and for the solution of second. 12.and higher-order linear differential equations with constant coefﬁcients. Constant coefﬁcients 12.2) is called homogeneous.1 Concepts The second-order differential equations that are important in the physical sciences are linear equations of the general form d2y dx 2 + p( x ) dy + q( x ) y = r ( x ) dx (12. otherwise it is inhomogeneous.2 Homogeneous linear equations The general homogeneous second-order linear equation with constant coefﬁcients is d2y dx 2 +a dy + by = 0 dx (12. Differential equations with non-constant coefﬁcients are discussed in Chapter 13. and it is these linear equations with constant coefﬁcients that are discussed in this chapter. . and the Schrödinger equations for the particle in a box and in a ring.1 The functions p(x) and q(x) are called the coefﬁcients of the equation. 1 Many of the methods and examples found in textbooks can be traced to Euler’s Institutiones calculi differentialis (1755) and Institutiones calculi integralis (1768–1770.3) where a and b are constants. Euler was responsible for the distinction between homogeneous and inhomogeneous equations. When these coefﬁcients are constants the solutions of the equation can be expressed in terms of elementary functions.12 Second-order differential equations. the use of integrating factors. It is always possible to ﬁnd a solution of this equation that has the form eλx where λ is a suitable constant. 3 volumes).

and d 2 y2 dx 2 2 2 −5 dy + 6y = 0 dx dy1 = 2e2 x .1 Show that y11=1e 2x and y21=1e 3x are two solutions of the equation d2y dx We have y1 = e2 x .2). d 2 y1 dx 2 + p( x ) dy1 + q( x ) y1 = 0 dx and let y3(x)1=1cy1(x). dx dx dx and d 2 y3 dx 2 2 d y dy dy3 + q( x ) y3 = c 21 + p( x ) 1 + q( x ) y1 = 0 dx dx dx d 2 y3 dx 2 =c d 2 y1 dx 2 + p( x ) . dx d 2 y2 dx 2 = 9e3x −5 dy2 + 6 y2 = 9e3x − 15e3x + 6e3x = 0 dx 0 Exercises 1–3 This example demonstrates the general result that it is always possible to ﬁnd two particular solutions y1(x) and y2(x) of a homogeneous linear equation. For example. dx d 2 y1 dx 2 = 4e2 x −5 dy1 + 6 y1 = 4e2 x − 5 × 2e2 x + 6e2 x = e2 x ( 4 − 10 + 6) = 0 dx dy2 = 3e3x . y2 = e3x . Therefore d 2 y1 dx Similarly. If either of these solutions is multiplied by a constant the new function is also a solution. Then dy3 d ( cy1 ) dy = =c 1. Constant coefﬁcients EXAMPLE 12. where c is a constant.338 Chapter 12 Second-order differential equations. let y1(x) be a solution of the general homogeneous equation (12.

The functions y1 and y3 are not regarded as distinct solutions however because each is merely a multiple of the other. The solutions y1 and y2 in Example 12. and neither is a multiple of the other. If (12. This important property of linear homogeneous equations is called the principle of superposition (it is not true for inhomogeneous equations or for nonlinear equations).5). We have dy dy dy = c 1 + c2 2 . substituting into the general homogeneous equation (12.4) is true only when a11=1a21=10 then the functions are linearly independent. In general. two functions. the function (12. the functions are said to be linearly dependent.12. y1 and y2. when y1 and y2 are linearly-independent solutions.2).5) Therefore. containing two arbitrary constants. In particular. dx 1 dx dx d2y dx 2 = c1 d 2 y1 dx 2 + c2 d 2 y2 dx 2 (12.2 Homogeneous linear equations 339 so that y31=1cy1 is also a solution.4) such that a1 and a2 are not zero. are said to be linearly dependent if there exists a relation a1 y1(x)1+1a2 y2(x)1=10 (12. We now show that if y1(x) and y2(x) are two solutions of a linear homogeneous equation then any linear combination of them. is the general solution of the homogeneous equation (see Section 11.1 are linearly independent. .2). d2y dx 2 d2y d 2 y2 dy + q( x ) y = c1 21 + c2 dx dx dx 2 dy dy + p( x ) c1 1 + c2 2 + q( x ) c1 y1 + c2 y2 dx dx + p( x ) ( ) = c1 2 d y 1 dx 2 + p( x ) 2 d y dy1 + q( x ) y1 dx dy2 + q ( x ) y2 dx + c2 =10 dx 2 2 + p( x ) and the result is zero because both sets of terms in square brackets are zero since y1 and y2 are solutions. is also a solution. y1=1c1 y11+1c2 y2 where c1 and c2 are arbitrary constants.

3). Constant coefﬁcients EXAMPLE 12.2 The general solution of the equation in Example 12. dx and substitution into (12. I happened on my solution entirely unexpectedly. +a dy + by = 0 dx (12.3) if λ21+1aλ1+1b1=10 (12.6) d2 dx 2 eλ x = λ 2 eλ x or e λ x(λ21+1aλ1+1b)1=10 The function (12.1 is y(x)1=1c1e2x1+1c2e3x Particular solutions are obtained by assigning particular values to c1 and c2.340 Chapter 12 Second-order differential equations.3) gives λ2e λ x1+1aλe λ x1+1be λ x1=10 (12. 0 Exercises 4–6 12.3 The general solution d2y dx 2 Let one solution of equation (12.3) be y1=1e λ x Then d λx e = λ eλ x . before that I had no suspicion that the solution of algebraic equations had so much importance in this matter’.7) This quadratic equation is called the characteristic equation or auxiliary equation of the differential equation.8) 2 In a letter to Johann Bernoulli in 1739 Euler described his discovery of the solution of linear equations with constant coefﬁcients by means of the characteristic equation: ‘after treating the problem in many ways. .2 The possible values of λ are the roots of the quadratic: λ1 = 1 − a + a 2 − 4b . 2 ( ) λ2 = 1 − a − a 2 − 4b 2 ( ) (12.6) is therefore a solution of equation (12.

y2 = e λ2 x (12. 8 In Example 12.8). Two particular solutions of the differential equation are therefore y1 = e λ1 x = e2 x . y2 = e λ2 x = e3x and. λ21−15λ1+161=1(λ1−12)(λ1−13) and its roots are λ11=12 and λ21=13.10) . If a and b are real numbers.3 the characteristic equation has two distinct real roots so that y1 and y2 are linearly independent and the general solution is a linear combination of them.3 The characteristic equation of the differential equation d2y dx is λ21−15λ1+161=10 2 −5 dy + 6y = 0 dx The quadratic can be factorized. because these functions are linearly independent. the general solution is y1=1c1 y11+1c2 y21=1c1e2x1+1c2e3x 0 Exercises 7.3 The general solution 341 and the possible solutions of type (12.9) EXAMPLE 12.12. The nature of the roots (12.3.8) depends on the discriminant a 21−14b. as in Example 12. the general solution is y ( x ) = c1 y1 ( x ) + c2 y2 ( x ) = c1e λ 1x + c2 e λ 2x (12. the particular solutions by (12.9) and.6) are y1 = e λ1 x . the three possible types are (i) a 21−14b1>10 (ii) a 21−14b1=10 (iii) a 21−14b1<10 (i) two distinct real roots one real double root a pair of complex conjugate roots Two distinct real roots The roots are given by (12.

The characteristic equation λ21−14λ1+141=1(λ1−12)21=10 (12. and the general solution is y(x)1=1(c11+1c2 x)e 2x 0 Exercises 9.11) Given one particular solution of a homogeneous equation it is always possible to ﬁnd a second. dx so that d 2 y2 dx 2 +a dy2 a 2 x − ax 2 ax e + by2 = − a + + a 1 − e− ax 2 + bxe− ax dx 4 2 a2 x a2 x = −a + +a− + bx e− ax 4 2 2 2 ( ) d 2 y2 dx 2 = − a + a 2 x 4 e− ax ( ) 2 a2 x x = − + bx e− ax 2 = − a 2 − 4b e− ax 4 4 ( ) 2 =10 since a 21−14b1=10. Constant coefﬁcients (ii) A real double root When a 21−14b1=10 the two roots (12. 10 . with λ11=1λ 21=1−a22. Thus dy2 = 1 − ax 2 e− ax 2 .3). Only one particular solution is therefore obtained from the characteristic equation: y1 ( x ) = e λ 1x = e− ax 2 (12. linearly independent of y1.13) has the double root λ1=12. The general solution in this case is therefore y(x)1=1c1e−ax221+1c2 xe−ax221=1(c11+1c2 x)e−ax22 EXAMPLE 12. a second solution.8) are equal. In the present case.342 Chapter 12 Second-order differential equations.12) This is shown to be a solution by substitution into the differential equation (12. is y2(x)1=1xy1(x)1=1xe−ax22 (12. Two particular solutions are therefore e 2x and xe 2x.4 Solve the differential equation y″1−14y′1+14y1=10.

the roots are a pair of complex conjugate numbers.34). 0 Exercises 11.7) are. A trigonometric form is obtained by expressing e±ix in terms of sin1x and cos1x by means of Euler’s formula. equations (8. EXAMPLE 12. or 2 λ 21=1λ*1=111−1i 1 ( ) Two particular solutions are therefore y1 ( x ) = e λ 1x = e(1+ i ) x .8). the roots of the characteristic equation (12. y2 ( x ) = e λ 2x = e(1− i ) x These are linearly independent. from (12.5 Solve the differential equation y″1−12y′1+12y1=10.12. and the general solution is y(x)1=1c1e(1+i)x1+1c2e(1−i)x =1e x(c1e ix1+1c2e−ix) This is the exponential form of the solution.7) is negative. 12 In the general complex case. e ±ix1=1cos1x1±1i1sin1x Then y ( x ) = e x c1 (cos x + i sin x ) + c2 (cos x − i sin x ) = e x ( c1 + c2 ) cos x + i( c1 − c2 ) sin x =1ex(d11cos1x1+1d21sin1x) where d11=1c11+1c2 and d21=1i(c11−1c2) are (new) arbitrary constants. 1 2 ± 4 − 8 .33) and (8.3 The general solution 343 (iii) Complex roots When the discriminant a21−14b of the characteristic quadratic equation (12. λ=− a a ± −b 2 2 2 . The characteristic equation is λ21−12λ1+121=10 and the roots are λ = λ11=111+1i.

where ω is real. Initial conditions are usually associated with applications in dynamics. and (when ω1≠10) the general solution is y(x)1=1c1 y1(x)1+1c2 y2(x) =e − ax 2 y2(x)1=1e(−a22−iω)x (12. y(x)1=1c1 y1(x)1+1c2 y2(x) of a homogeneous second-order differential equation are normally determined in physical applications by the application of two conditions (two for two constants) in either of the following two forms. Let (a22)21−1b1=1−ω 2.15) (c e 1 iω x + c2 e − iω x ) (12.18) A second-order differential equation with initial conditions is called an initial value problem.5) y(x)1=1e−ax22(d11cos1ω x1+1d21sin1ω x) where d11=1c11+1c2 and d21=1i(c11−1c2). 2 λ2 = − a − iω 2 (12. (12.344 Chapter 12 Second-order differential equations. y′(x0)1=1y1 (12.16) The trigonometric form of the general solution is (see Example 12. Constant coefﬁcients where (a22)21−1b1<10. Then a 2 − b = −ω = −1 ω = iω 2 and the roots are λ1 = − 2 a + iω . . (a) Initial conditions The values of the function y(x) and of its derivative y′(x) are speciﬁed for some value of x: y(x0)1=1y0.4 Particular solutions The two arbitrary constants c1 and c2 in the general solution.17) 12.14) The particular solutions of the differential equation are therefore y1(x)1=1e(−a22+iω)x. when x is the time variable.

y′(0)1=151=12c11−13c2 with solution c11=11. and x1 and x2 are the end-points of the interval x11≤1x1≤1x2 within which the differential equation is deﬁned. y(π22)1=12 The differential equation is that solved in Example 12. The conditions y(x1)1=1y1.12.19) are called boundary conditions. c21=1−1. The general solution is y(x)1=1c1e 2x1+1c2e−3x with ﬁrst derivative y′(x)1=12c1e 2x1−13c2e−3x Then y(0)1=101=1c11+1c2. y(x2)1=1y2 (12. The solution of the initial value problem is therefore y(x)1=1e 2x1−1e−3x 0 Exercises 13–16 (b) Boundary conditions In many applications in the physical sciences the variable x is a coordinate and the physical situation is determined by conditions on the value of y(x) at the boundary of the system. y(0)1=10.5. with general solution (in the trigonometric form) y(x)1=1e x (d11cos1x1+1d21sin1x) . y(0)1=11. A differential equation with boundary conditions is called a boundary value problem. with roots λ11=12 and λ 21=1−3. EXAMPLE 12.6 Solve the initial value problem y″1+1y′1−16y1=10.7 Solve the boundary value problem y″1−12y′1+12y1=10. y′(0)1=15 The characteristic equation is λ21+1λ1−161=1(λ1−12)(λ1+13)1=10.4 Particular solutions 345 EXAMPLE 12.

6. y(0)1=11. when the differential equation itself contains parameters to be determined.346 Chapter 12 Second-order differential equations. The general solution is given by either (12. Several examples of such equations are discussed in Chapter 13. y(x)1→10 as x1→1∞ The equation is that discussed in Example 12. y(x)1=1c1eiω x1+1c2e−iω x (12. Then c21=11 and the solution of the boundary value problem is y(x)1=1e−3x 0 Exercise 20 The equation y″1+1ω 2y1=10 In Examples 12. the two conditions on the solution of the differential equation are sufﬁcient to specify a particular solution. d11=11 and d21=12e−π22.21) . with general solution y(x)1=1c1e 2x1+1c2e−3x The ﬁrst boundary condition gives y(0)1=111=1c11+1c2 The second condition requires that the solution go to zero as x goes to inﬁnity. one or more additional conditions are required to completely specify the solution.16) with a1=10. In some cases however. and the solution of the boundary value problem is y(x)1=1e x (cos1x1+12e−π221sin1x) 0 Exercises 17–19 EXAMPLE 12.8. The condition therefore requires that we set c11=10.6 to 12.8 Solve the boundary value problem y″1+1y′1−16y1=10. One of the most important simple examples in the physical sciences is the equation d2y dx 2 + ω2 y = 0 (12. The function e−3x has this property but the function e 2x must be excluded.20) where ω is a real number. Constant coefﬁcients Application of the boundary conditions gives y(0)1=111=1e0 (d11cos101+1d21sin10)1=1d1 y(π22)1=121=1eπ22 (d11cos1π221+1d21sin1π22)1=1d2e π22 Therefore.

However. y(x)1=1d1 cos1ω x1+1d21sin1ω x (12. In this case. ±1.12. EXAMPLE 12. The general solution is y(θ)1=1c1eiωθ1+1c2e−iωθ.1= (12. d 2y dθ 2 + ω 2 y = 0 for the cyclic boundary .26) The constants c1 and c2.24) The possible values of ω are then (using n to label the values) ωn1=12πn2λ . This is the case for the quantummechanical ‘particle in a box’. replacement of x by x1+1λ in the general solution (12. or d1 and d2. this is the case when ωλ is a multiple of 2π. ±2. ±3. ωλ1=12πn. ±1. ±3. The case of the periodic boundary condition is exempliﬁed in Section 12. and the corresponding solutions are yn(x)1=1c1e(2πnx2λ)i1+1c2e−(2πnx2λ)i.1= (12.4 Particular solutions 347 or by the equivalent trigonometric form (12.25) + d 2 sin 2 π nx λ (12. as a boundary value problem. if ω is a parameter to be determined. This is exempliﬁed in Section 12.7 for the quantum-mechanical problem of the ‘particle in a ring’.17). n1=10. then an additional condition is required.22) If ω is a given constant then two initial or boundary conditions are normally sufﬁcient to specify a particular solution.21) gives y(x1+1λ)1=1c1e iω (x+λ)1+1c2e−iω (x+λ)1=1c1e iω xe iωλ1+1c2e−iω xe−iωλ The condition (12.6). A third possibility arises when the physical situation requires the imposition of a periodic (cyclic) boundary condition (see Section 8.6.6. As discussed in Section 8. y(x1+1λ)1=1y(x) (12. discussed in Section 12. are not determined.23) is then satisﬁed if both e iωλ1=11 and e−iωλ1=11.9 Solve the equation conditions y(θ1+1π)1=1y(θ).23) where λ is the period.5 for the classical harmonic oscillator as an initial value problem. The trigonometric form of these solutions is yn = d1 cos 2 π nx λ n1=10. ±2.

....... ................... Therefore......................... .•...... .................. x o Figure 12..........1= and y(θ)1=1c1e2inθ1+1c2e−2inθ 0 Exercise 21 12. .. ...... ±1.. ω1=12n for n1=10... and whose direction is towards this point.......... ........................ ............................... . ..... the point of equilibrium.........•....1 xk− = F e r f c o m ss am ............. important in chemistry.. the oscillations of a spring balance (when equation (12..... x .27) is called Hooke’s law)..1) consists of a body moving in a straight line under the inﬂuence of a force F1=1−kx (12............ the vibrations of the atoms in a molecule or crystal. The quantity k is called the force constant and the negative sign ensures that the force acts in the direction opposite to that of the displacement.27) whose magnitude is proportional to the displacement x of the body from the ﬁxed point O............. By Newton’s second law of motion........................... ±2....348 Chapter 12 Second-order differential equations.........5 The harmonic oscillator The simple (linear) harmonic oscillator (Figure 12................... the swings of a pendulum............ .................. the oscillations of a diving springboard and..... the acceleration experienced by the body is given by F=m d 2x dt 2 (12.................. muirbiliu q e ...28) and simple harmonic motion is therefore described by the differential equation m d 2x dt 2 = − kx (12....... ... Constant coefﬁcients Then y(θ1+1π)1=1c1eiω θeiω π1+1c2e−iω θe−iω π and y(θ1+1π)1=1y(θ) if πω1=12πn for integer n..................... ....29) A variety of physical systems can be modelled in terms of simple harmonic motion...... ........

. . ... . ....... ...... ......... . . . . . . . .... . Then..... ... .. . . .. .... . .7. . .... . ....57)).. . . . . . . . . . ... . . .... . ... .... .. ... . . . d 2x dt 2 + k x=0 m (12. . .. ... .. ... .... .. ... .... ...... . ... .. .. e . .2 V = De a 2 ( R − Re )2 − a3 ( R − Re )3 + ≈1a De(R1−1Re) 2 2 (12. .. . . .... . .. . ... . .. . .. .... . ... ... . . .. ... .. . ..... ... .5 The harmonic oscillator 349 EXAMPLE 12. .. . .... .... .30) Figure 12....... . . . Re is the distance at equilibrium (the equilibrium bond length).. ... ... . .. . .. ... ... . ...33) If k1=12a 2De and x1=1(R1−1Re).. . .. .... . .. . . R1−1Re. . . ...... R ) ( RV e 0 D where (Figure 12. . . . . .. .... ...... .. . ... . .... ... the force is F1≈1−kx. .3) of a homogeneous linear equation with constant coefﬁcients. ... . .... .. .. .. . .. .. .. . De is the dissociation energy of the molecule and a is a constant (the vibrations of the molecule can be visualized in terms of a ball rolling forwards and backwards in the ‘potential well’ in Figure 12.. .. . . . . .. .. . .... .. .. ... . ..... . ..31) gives F1≈1−2a 2De(R1−1Re) (12.... . . .. .. . .... . . . . ... .. .. .. . .. .. . . . ....... ..... . Equation (12... . .. . . .10 The vibrations of diatomic molecules The vibrations of a diatomic molecule are often modelled in terms of the Morse potential 2 − a ( R − Re ) V ( R) = De 1 − e (12. ... . and the vibrations of the molecule are (approximately) simple harmonic. . ..... . .. . .. .. . .. F=− dV dR (12. . .. . . .. . . .. . . . .. . .. .. .. . .2). .. .. .. A stable molecule in its ground or low-lying excited vibrational states undergoes only small displacements. equation (5. ... . . from equilibrium. . .. ... .. .. . .. . . .. .. for small displacements.. .... .. ... differentiation of (12..... .. . ...... .... .. ... ..... ... .. .... .34) R ... . . .. . . . ...... ..2) R is the distance between the nuclei. .. .. expanding the potential-energy function V(R) as a power series in (R1−1Re)..32) Therefore.. ... . ...... ...29) can be written in the standard form (12. .. .. . . . .. .. . .. .. . ..31) The force acting between the nuclei of the molecule is (see Section 5.. .... . .. .. .... . . .. . ..12.. . . ... ... . . . . . . . . . . .. . .. ... . . . . . .. ......... . . . . . . .. .. .. .....

....... .. .. ..... .. . ... .. . Then. . . .. . .... . .... . . . . . .. .. . . is called the frequency of oscillation. ..3. ....350 with general solution... .. ... ..... .. ........ .. . . . ..... .... . . The frequency of oscillation is related to the force constant k by ω 21=1k2m: t .. . . . .. .... . ... .. ... . .... . .. . ... . . the time taken for one complete oscillation.. A ) ( .. .. . . . . . .. ... . . . . .. ... ... ...... . . ... .. .. . .. .. . ... .. .. ........... .. . ... ... . . The inverse of the period. .. .. . . ..... . . .. ......... x(t)1=1A1cos1ω t d 2x dt 2 k . .. ... .. ... . d21=10.. .... . . . . .. . . x′(0)1=1−d1ω1sin101+1d2ω1cos101=1d2ω 1=10 x(0)1=1A1=1d11cos101+1d21sin101=1d1 x′(t)1=1−d1ω1sin1ω t1+1d2ω1cos1ω t x(t)1=1d11cos1ω t1+1d21sin1ω t ω= ....... ... . . ...37) (12.. .. .. .. .. . . .. ... .. . . . .... .. ... ... .. . ... .... . . .......... .... . setting k2m1=1ω 2. . . . .... .. . . . . . . . ..... . ........... ... . . . . ... . m Figure 12. .. ..20).. for the initial displacement. . .. . . ........ ... . . ...... . . . . . .. ...... . .... . . .. . .. . ... .. ... . . .. . . .. ... .... ... . ...... .. . .. . .. . .. ω/π . .. . ... .... .. .35) A− 0 ... ....... . ...... ...... .. .36) (12.. .. .. . . . . . ...... ... . .. . . .. . . .. . . . .. . . .. .... . .. x(0)1=1A. ... . . .. .. . .. . . The quantity ω1=12πν is called the angular frequency. . . ... . .... . .. . ..... ......... . .. .. . . . .. ....... . . The wavelength of the representative curve is the period τ 1=12π2ω ... . .. ..... .... . For example.. ....... .. ... . ... .. . . .. . . .. .... .. ... .. .. . ......... . . .. . ...... ... . .... .. . .... ... ... . tx .... . .. .. . ..... ... ..39) (12... ν1=112τ 1=1ω22π.. ... . in trigonometric form... . ....... ... and the solution of the initial value problem is The state of the system is determined by the initial conditions. the number of oscillations in unit time... . The graph of the solution is shown in Figure 12.. . . . ..... . .. .. .. . ... ... .. ..... The maximum displacement of the body from equilibrium is the amplitude A. ... . .. . . .. . .. .... . ..... .. . . .. ........ . . . ...... ... .. . . . . .. . .. ... .. .. . . . . .. . . . . ... .. .. . .... ... .. .. .3 + ω2x = 0 ν= x′(0)1=10 1 k 2π m (12. ... ... . .... . ... .. . ... .... . ... . . . Chapter 12 Second-order differential equations..... .... . ... . ..38) (12.. . .. . .. . . . .. .... .. . . ... .. . ... .. ... ... .... .. .. Constant coefﬁcients Therefore d11=1A... ω/π 3 ... .. .. ... let the displacement and velocity at time t1=10 be For the velocity or. ω/π 2 ........... ....... ... . . . . .. ... .... . ω/π 2= τ .. in the form of equation (12.

..x A x V xT . .. . . Then The harmonic oscillator is a conservative system (see Section 5... ... .. . . .. ... ....... ... ..... . . . . .. .. . ....... . . .... . ..... ..... . and the total energy is E1=1T1+1V.. . ... . . . . the potential energy has its maximum value.. .. + = .. ..... . .... ....... The potential energy is therefore the integral The kinetic energy is T1=1mv 222.. .. ... ...5 The harmonic oscillator (12. ..... .. . ... ..... ...... ..... .. .. ...... ... . ... ....... ...... .. . .. . .. . .. . ........ . .... ...... .. ...... .... . .. .. ... . The total energy is therefore the constant and so that Energetics T= V ( x) = Z x1=1A1cos1ω t..... .. . ..... . . . .......... . .... . ... ... 1 2 1 1 mv = mω 2 A2 sin 2 ω t = k A2 sin 2 ω t 2 2 2 V= 1 dV dx = − Z F dx = kx 2 + c 2 dx E =T +V = 1 2 1 2 kx = k A cos2 ω t 2 2 Figure 12. .7) and the force F1= −kx is the derivative of a potential-energy function. . .. ..... .. ... .. .. . .. . . . .. For the state of the system described by solution (12. ... .......... .. ...... . .. .... ..... ... ... . .. . . . . ...... . . . . .. . . . .. ...... ... . . . ... .... ... ..40) (12......... . ....... . . .. .. . . ... . . . ...... .. . ...... . .. . . .. . . ....... . ... ..... .. ........ . . .. .... .. . . .. ..... ........ . . . ........ . . ... . .. .... . .. by convention. . .... ... ....... ............ x1=1A... . .... . ... .. ... . ..... . .. . . .. .. . .... . . . ........... ....... .4.. . ) ( .. ... . .... ..... .... . .. . ... .... . .. . .. .... .. .... F1=1−dV2dx... . . . . . ... . ..... .. ... ....41) 351 ...... . . . ...38)..... .... .. . .. .. . ... ... ... ...... . . ........ . .. .. .. .. .... .. . .... . ... . . . .. . . . . ... .. ...... .. ..... .... .. . ...... . .. .. . .... . ... ... ... ... ....... . . . ...... . . . .. . .. .. . .. ..... ... . ... ... . .... . . .. . ....... ........ .. . . .. At maximum displacement from equilibrium..... ........ .. ..... . .. .... ..... ..... .. . . ... .. . ... . .. .... . ... . . . . .. . . . .. . . .. .... .. ... ... . ... . . ..... . .. .... . and the kinetic energy is zero...... ..... ... . .. ...... ... .... .. . .... . ... . . . .. . ..... .. . .. .. . . . ... . ..... .. .. .... . ... .......... . ........ .. ... .. .... ... .. .... ........ ... .. . ..... . . ... . ... ... ... .. .. ... .. ... . .. . .. . .. .... . .. . ..... . ..... . .. .. .. .. ..... . . .. . .. . ) ( . . since ω 21=1k2m... . . .. . ... . . . .. ...... ... ... ... .. . . . . ...... .. .... V1=1E. .......... ...... . ... . ... ....... .. ... where v is the velocity. . . . ... . . ..... . .... . . .. ... . . .. ..... .. . .. .. .. . . . .... . ..... . ........... . ........ . .... . .. . . ..... .. ... ....... . ...... ... . .. . ....... . .. .. ... when x1=10.... ...... ........ . ....... ............ ... . ..... .. ..... As the body approaches the The potential energy is chosen.... .. ..... .. ... . ... .. .. ... ... ... . .. ..... . ..... . V T E The relation between the potential and kinetic energies is shown in Figure 12..... ... . .. . .... .. ... . . . ... .. . .. ... ... .. . . . ...... to be zero at the equilibrium position. .. ... .... . ... . .. .. ...... ......... ... . ....4 0 A− y r n e g e V= v1=1x′1=1−Aω1sin1ω t 1 2 kx 2 1 2 kA 2 12. ... .. ... ..... ... ... . . . . .. .... . ..... . . .. .. . .. ......... .. ...... .... . .... . . ...... .... .. . . .. ... . ..... . ........ ..... . . .. .. ..

...... The inﬁnite value of V at the ‘walls’ and outside the box ensures that the particle cannot leave the box............ potential energy is converted into kinetic energy and...... .......... ........... ...... 0 Exercises 22. in quantum mechanics this means that the wave function is zero at the walls and outside the box.... .. ............ V1=10 and the kinetic energy is a maximum................. (12...... ...... ............................. .35) for the harmonic oscillator... ............20)......... .... . ... ........ ... ... .. .... The Schrödinger equation for a particle of mass m moving in the x-direction is − 2 d 2ψ ( x ) dx 2 2m + V ( x )ψ ( x ) = Eψ ( x ) (12........ ................. ..... setting V1=10 means that the energy E is the (positive) kinetic energy of the particle......... ............352 Chapter 12 Second-order differential equations...44) is identical in form to equation (12............ ......... .................... . ........................... ∞ = V ........... .... ...... ... . Constant coefﬁcients equilibrium position........................ ........... ................43) Figure 12................... 0 = V ..... ..................5) 0 V ( x) = ∞ . For the particle within the box.................... E is the (constant) total energy.... this is one of the simplest systems that demonstrate the quantization of energy...... ........ ....... Thus......44) (12...... ........... ...................................... ...... .............5 The constant value of V inside the box ensures that no force acts on the particle in this region.......... .... T1=1E.. . ........................ In quantum mechanics.... .. ..42) where V(x) is the potential energy of the particle at position x. and ψ is the wave function.................... .......... .. ....45) (12..46) ......... ........ or to equation (12..... . we therefore have the boundary value problem − with boundary conditions ψ (0)1=1ψ (l)1=10 2 d 2ψ ( x ) dx 2 2m = Eψ ( x ) Equation (12............... at x1=10.. .... .. l . for for 0<x<l x ≤ 0 and x ≥ l ... For the present system the potential-energy function is (Figure 12...... 0 ∞ = V . . setting ω2 = 2 mE 2 x (12...... . ................. ............... .... 23 12.. ..... ........ .....6 The particle in a one-dimensional box The ‘particle in a one-dimensional box’ is the name given to the system consisting of a body allowed to move freely along a line of ﬁnite length... ... ..................

48) We now apply the boundary conditions (12. For each unique solution ψn there exists an energy En = n2 h2 8ml 2 (12.51) where the permitted solutions have been labelled with the quantum number n.47) with general solution.4): ψ 2(x)dx1=1probability that the particle be in element dx at position x . The value n1=10 has been discounted because the trivial solution ψ0(x)1=10 is not a physical solution.12.6 The particle in a one-dimensional box 353 we have d 2ψ dx 2 + ω 2ψ = 0 (12. ψ (x)1=1d11cos1ω x1+1d21sin1ω x (12. l n = 1. because the sine function is zero only when its argument is a multiple of π. … (12. … (12. The wave functions (12.49) It follows.48)). ± 1. 2. they are valid for any continuous curve of length l if the variable x is measured along the curve. ψ (l)1=1d21sin1ω l1=10 (12. ψ (0)1=1d11cos101+1d21sin101=1d11=10 and at x1=1l (having set d11=10 in (12.52) (using E1=1A2ω222m1=1h2ω228π2m). We note that these results are not restricted to motion in a straight line. At x1=10. l n = 0. that ωl1=1n π where n is an integer: ω= nπ . and negative values of n have been discounted because ψ−n(x)1=1−ψn(x) is merely ψn with a change of sign.51) are not yet fully speciﬁed because the coefﬁcient d2 is not deﬁned. this is a general result in quantum mechanics. in trigonometric form.45). We see that quantization of the energy is a consequence of constraining the motion of the particle to a ﬁnite region of space by the application of the boundary conditions. To determine d2 we invoke the quantum-mechanical interpretation of the square of the wave function as a probability density (see also Example 10. ± 2. 3.50) The solutions of the boundary value problem are therefore ψ n ( x ) = d 2 sin nπ x .

. .. . .. . l l l 0 ) (3 x ψ ) (2 x ψ ) (1 x ψ 354 The graphs of the ﬁrst three of these solutions are shown in Figure 12. the number of zeros (nodal points. ... . ........ . .. . . ..... ..... .. . .. . ... .. ... .. .. . . .. ... . . . ... .......... .... ... .... . . .. . .. .. .. .. is used in statistical thermodynamics to derive the thermodynamic properties of the ideal gas.. . ....... . ... ...... . . .. . ..... . .... . . ... .. . . ..... .... .. .... .. .. . . . . .. ... . . . .. Chapter 12 Second-order differential equations... .. .. . .... . .. .. .. ..3= n 2= n 1= n x . ... . ... ... ... . .... .. ... .... ... .. .. . .. ... .. . ...... ........ . .... .... .... . .. . .... ........... . ... . . . .. .. ... . . ...... .. .. . .... ..53) to demonstrate an important property of solutions of the Schrödinger equation.. . . . . . . .... . . .. .. . . . .. ..... 25 1 = Z ψ 2 ( x ) dx = ( d 2 )2 Z sin 2 0 l ψ n ( x) = 2 d2 − ψ = Eψ 2m dx 2 Figure 12. . .. . ... . .... ... ..... .... .. ... . .. . ... .. .... . These graphs show that the wave function ψn has (n1−11) zeros.. . ...... ......... .... .. . .. .. .. . . . . .. 0 Exercises 24. . ..... ... ....... . ... ..53) (12. when generalized to three dimensions.. . . ..... ..... ...... . ... .. . . . . ... .. ..... .. . . .... . ... . ..... ..... . .. .. . .... .... .. . . ... .. . . . . . . ... . ... .. .. ... . .. . ..... . . .. . ...... ...... ..... .. .... . .. .. .. . .. ........ . .. . . . .. . .. .. . .. .. .. . . . . ... .. .. . . .. .. ... ..... ..6 H ψ 1=1Eψ nπ x 2 sin l l 0 l nπ x l dx = ( d 2 )2 2 l (12.... .. . ....... ... ... . . . . . . ... . between the end points.. .... .... ..... .. Constant coefﬁcients or The Schrödinger equation for the particle in a box can be written as Orthogonality Therefore d 2 = 2 l . . . . ...... . .. ....... .. . ... .. . .... . ...... .. ... .... . ..... .... . . .. or nodes.. . . ... ...... . .... .... . .... . .. ... . ..... . . .. .. .. integrating over the length of the box..... . . .. . . ...6. .. curves or surfaces) increases as the energy of the system increases. ... . . . . . . .. ... . ... . .... . . .. . . ... . . .. .. . ... .. . .. ... . . . .. ... .. .. . . ..... . ... ... . .. . . .. .. ...... . ... . . ...... . .. . . . .. ...... . ...... .. . ... ... . ... .. ... . .... ... . . ....... ........... ... ... . . . ... ... . . . . .. .. .. .. .. .. . ......... .... .. ..... .. . . ....... ... .. . . .. The particle in a box is important in the physical sciences not only as one of the simplest solvable problems that demonstrate the phenomenon of quantization.... .... . .. ..... .... .. . ... . . . . .. . . ..... .. ..... .... . .. . . . .... . . .... .. . . ... . .... ... . . ... .. . ..... ..... . .. . .... .. ... . .. ... .......... ...... ... . . .. .. . .. . .. .... ... ...... .... ... . ..... .. .. .. .... . ..... ... . ... .. ...... ... ... ... .. ...... . . and the normalized solutions are The total probability that the particle be in the box is unity so that. .. ... . . .. .. . .... . .. .. . . .... ... . . . ... . . .. . . . .. ....... but also because the system... .. . . . 0 0 . .. . ..... ... . . ... .. . . .. ... . . . ..... .. ... .. .......... ........ . We now use the solutions (12. .54) ... . . .. . .. ... .. ... .. . ...... . . . This is a demonstration of a general property of wave functions... .. . . . ... .... .. .. ... . . .. ..... .. . ... .... . . . . . . ... . . .. . . .. .. ...... .. .. . ...... . . .. .. ... .. .. . ..... ..... .. ... . .... ... ........... ... . .. .. . . . .. ... ....... . . . .... .. . . ... . . . .. . ...... . . . . ...... ... .. . . .. ... . ... . . . ... . ... . . .... ... ... ... .. .. ... . ....... .. . ... . ... . . . ..... . ........ . .. .. ...

and the corresponding energies E1=1En are the eigenvalues of H . Then I= 2 mπ x nπ x Z sin sin dx l 0 l l when m1≠1n l (12. consider the integral I = Z ψ m ( x ) ψ n ( x ) dx 0 l (12. The permitted wave functions ψ1=1ψn are called the eigenfunctions of the Hamiltonian. Thus. m1±1n. of π is also zero.54) as an eigenvalue equation. or simply the Hamiltonian.2.12.57) =10 Proof. l l ( m − n) π x ( m + n) π x 1 1 Z cos dx − Z cos dx 2 0 l 2 0 l l ( m ± n) π x ( m ± n) π x 1 Z cos sin dx = ( m ± n) π l l 0 0 l = l sin ( m ± n) π − sin 0 = 0 ( m ± n) π since sin101=10 and the sine of a multiple. An important property of eigenfunctions is that of orthogonality. The integral is evaluated by making use of the addition properties of trigonometric functions. The (time-independent) Schrödinger equation can always be written in the form (12. with the system speciﬁed by the Hamiltonian operator and appropriate boundary conditions.56) where ψm and ψn are two different eigenfunctions (12. In the present case. .53). as discussed in Section 6.6 The particle in a one-dimensional box 355 where the differential operator H =− 2 d2 2m dx 2 (12.55) is called the Hamiltonian operator. The effect of operating with H on ψ is to generate a multiple of ψ. ( m + n) π x mπ x nπ x 1 ( m − n ) π x sin = cos − cos l l l l 2 sin so that I= Now. of the system.

... .. ... . . .62) with general solution. . .. .. . .. equal to 1 if m1=1n (normalization) and equal to 0 if m1≠1n (orthogonality). ... . .. .. . . 0 Exercise 26 12. in exponential form ψ (θ )1=1c1eiωθ1+1c2e−iωθ (12. . . . . . . . . .. . . Figure 12.. . . is called the Kronecker delta. . .. ... .. .. . . .. . . ... . . ..59) are said to be orthonormal (orthogonal and normalized). . .60) . . .. ....... . ... •.. we can write Z l ψ m ( x )ψ n ( x ) dx = δ mn = 0 1 0 if if m= n m≠n (12. Constant coefﬁcients It follows therefore that Z l ψ m ( x ) ψ n ( x ) dx = 0 when m≠n (12..... .7 The particle in a ring ... .... . . ... The Schrödinger equation for a particle of mass m moving freely in a circle of radius r is. .. . 2 d 2ψ (θ ) . .. . ... . .. .... .. .. . .64) . . . .. .. . it must satisfy the periodic boundary condition ψ (θ1+12π)1=1ψ (θ ) (12.. . . . . .. .. ... .. . .. .... ... . . . . .. . .. . . .. . For normalized wave functions.... ... ............. . . ... . ... .. . . . .. .. .. . .59) The quantity δmn... . . ..... . .. . . ... .. ..... ... . ..... . . ..61) we have d 2ψ dθ 2 + ω 2ψ = 0 (12.. . .7. . .. . .... . ...... . Functions that satisfy (12. . . ..63) For the wave function to be continuous around the circle..... . .. . .. . . .. . .. o . . . θ ... ... . ... . . .... .. . . ....... . .. . .. .. . . when Z ψn 0 l ( x ) ψ n ( x ) dx = 1 . . .. ... .... . .58) 0 The functions are said to be orthogonal.. . H ψ (θ ) = − = Eψ (θ ) (12. .. . . .. . . . ...... . .. .. . ... .. .. . .. . . .. r .. .7 where E is the (positive) kinetic energy of the particle and I1=1mr 2 is its moment of inertia with respect to the centre of the circle.. . Setting ω2 = 2I E 2 (12.. . .356 Chapter 12 Second-order differential equations. ... .. ... .. .. . ... . . ... . . . . ... .. 2I dθ 2 Figure 12... . .. .. .. .. .. . .. .

66) We note that the states of the system with quantum number n1≠10 occur in degenerate pairs. The allowed values of ω are therefore ωn1=1n. In the absence of such a force. therefore. H (aψn1+1bψ−n)1=1En(aψn1+1bψ−n) where a and b are arbitrary. ± 1.7 The particle in a ring 357 Equations (12. ψn and ψ−n. ±2. H ψ −n1=1Enψ−n By the principle of superposition (Section 12. = (12. It is conventional to choose c21=10 in (12. ±1. with x replaced by θ and λ by 2π. The normalization condition for the complex functions is Z 2π ψ n θ ) ψ n (θ ) d θ = 1 *( (12. ± 2. n1=10.12.65). every linear combination of a pair of degenerate eigenfunctions is itself an eigenfunction with the same eigenvalue. they are orthogonal as well as normalized: Z 2π ψ n θ ) ψ m (θ ) d θ = 0 *( if n≠m (12.68) 0 where ψ *(θ)1=1c*e−inθ is the complex conjugate function of ψn(θ). to give the set of eigenfunctions ψn(θ)1=1c1einθ.64) are the periodic boundary value problem discussed in Section 12.70) 0 . 2π n = 0. = (12. every choice of coefﬁcients c1 and c2 is equally good.2). and the corresponding solutions are the eigenfunctions ψn(θ)1=1c1einθ1+1c2e−inθ. Then n 1 Z 2π ψ *(θ ) ψ n (θ ) d θ = c1 n 2 Z e− inθ einθ dθ = c1 Z dθ = 2π c1 0 0 2π 2 2π 2 0 and this is unity if c1 = 1 2 π .69) ψ n (θ ) = These functions form an orthonormal set in the interval 01≤1θ1≤12π. The normalized eigenfunctions are therefore 1 inθ e . ±1.4. It is physically possible to distinguish degenerate states of a quantum-mechanical system only by the application of an external force to break the degeneracy.65) with eigenvalues En = n 2I 2 2 (12. H ψ n1=1Enψn. n1=10.62) and (12.67) and to choose c1 to normalize these functions. ±2. … (12.

. . . . . . ..... . . . .. .. .. . .69) can be interpreted as representating clockwise rotation of the particle when n1>10 and anticlockwise rotation when n1<10 (see Section 8. . . . . . . .... . ... . . . . . ... .. . . .. . ..... . ...... . ... . . ... . ... . ...... . ... . . . . . . . . .. . .. ... .. and so on.. . . . . . ... . .. . . .. .. . .. .. .. . . .. .. . . ... . . . .... . . . . .... ... . . They are complex functions when n1≠10. .. . . . . . . . . .. ... . . . . . . . .... . . .. . .... . .... .. ... . . .... .. . .. . . . . . 2± = n . . ... . ... . .. .. . .. . . ... . ... . . . . . .... . .. . ... .. . . .... . . . . . Constant coefﬁcients 0 Exercise 28 0 Exercise 27 and this is zero because e2πi1=11. . .. .. . . . .. . . ......... ..... .. .. . . .. .... .. . 2 2 π ψ *(θ ) ψ 2 (θ ) d θ = 1 ψ ± n (θ ) = + – 1 1 iθ 1 2π i = = e − 1 e i 2π 2π i 0 1 2π Z e− iθ e2iθ dθ = Figure 12.. .. ..... . . .... . . . ... .. .. . . . . . . . . . . ... .. . . . . . .... . . ... .. ..358 We note that these symmetries are also the symmetries around the molecular axis of the molecular orbitals of a linear molecule.. . . . .. . . . .. . . . . .. ..... . . . . ... . .. .. .. . . ... . .. .. . . . . ... .. ... .. . . .... . . .. . ... ... ...... .. .. . . . . . . . .. .. . .. ... .. .. .. .. . .... . . . .. . . .. .. ... ...... . . . . . . .. .. .... . .. . . .... ... ... . ..... ... . . . . . ... .. . .. . .. . . .. .... . .. .. .... . ... . ... . ..... . .. .. .. . .. . .. .. . .. . The symmetries of these functions are illustrated in Figure 12. . ..... . . .. . . . .. . . . .. . ... . . . . . . . . .. . . . . . ...... . . . .. . . ... .. . .... and for some purposes it is more convenient to have solutions of the Schrödinger equation that are real functions. . .... . .. n1=1±2 for δ-orbitals. . ... .. . . .. . . . ... ... . .... . . .. with n1=10 for σ-orbitals. . . .. ... . . ... .. . . ... . . .. . ... . . The convention is to use the trigonometric functions obtained by means of Euler’s formula. .. We have EXAMPLE 12....... ..... . .. ... . . . . . .. .. .. and the combinations The functions (12. . .... . . . . . ... ........ ... ... ... .. . . . . . . ... . . ... .. . .. ...... .. ..... . ... . ... . .. . ..... .. . . . ... ... . . .. .... .. .. . . . . . . . . . .. .. . . .. ... . . .... . .. . .. ... . . . ..... . .. .. . .. .. .. . . ... . . .. . .8. .... . . ... . We have Chapter 12 Second-order differential equations. .... ......6 for the rigid rotor). .... . .. .. .. . .. . . .. They are also the symmetries of the normal modes of vibration of a circular drum... ... ...... ... .. . .. ... .. .. . . . . . .. . Z + 0 2π 1 1 (ψ n + ψ − n ) = cos nθ . . . . ... ... . . . .. ... ....... .. .. . . . . .. .. 1± = n are the alternative orthonormal set. . . . . . ... . . . .. .... .. ... . . .. .. . ... . .. . .... . .. . n1=1±1 for π-orbitals... . . . ...... ... .. . . .. .... ... . . .... ... .. .. .. ... . ..... .. . .. .. ..8 0=n 1 (cos nθ ± i sin nθ ) 2π 0 2π – 2π 1 1 (ψ n − ψ − n ) = sin nθ i 2 2 π + 1 Z eiθ dθ 2π 0 – + + – 2π + – – (12.71) + . . .. . . ........ . . ... .. .. . . . .... ....... . .. . .. . . .. .70). . .... .. .. . ... .... . . .. .. . ... .. . ... . . . . . . . .. ... .. . . . .... .. .. . .. . . . . .. . . . ... .. . . . .. . .. . . ... . ... ...11 Show that the functions ψ1 and ψ2 satisfy the orthogonality condition (12.. . . ...

75) where a and b are constants.8 Inhomogeneous linear equations 359 The general case If the end-points of the linear box discussed in Section 12.12 Find a particular solution of the equation y″1+13y′1+12y1=12x2.73) are the periodic boundary problem discussed in Section 12. the Schrödinger equation is unchanged. but the two boundary conditions (12.72) and (12. and y″1+13y′1+12y1=1(2a21+13a11+12a0)1+1(6a21+12a1)x1+12a2 x 2 y″1=12a2 . are given by equation (12.12.26). and the variable x by rθ. The form of the function on the right of the equation suggests a solution of type y1=1a01+1a1x1+1a2 x 2 Then y′1=1a11+12a2 x. ω2 = 2 mE 2 (12.72) if the variable x is measured along the loop. ψ n ( x ) = d1 cos 2 π nx 2 π nx + d 2 sin l l (12.45) are replaced by the single periodic condition ψ (x1+1l)1=1ψ (x) (12. The allowed values of ω are ωn1=12πn2l.8 Inhomogeneous linear equations d2y dx 2 The general inhomogeneous second-order linear equation with constant coefﬁcients is +a dy + by = r ( x ) dx (12. Particular solutions of this equation can be found by elementary methods for several important types of inhomogeneity r(x).6 are joined to form a simple closed loop. The results for the circle of radius r are then obtained by replacing l by the circumference 2πr. and the solutions. d 2ψ dx 2 + ω 2ψ = 0.73) Equations (12.74) These results are valid for a simple closed loop with any shape. in trigonometric form.4 (with λ replaced by l). 0 Exercise 29 12. EXAMPLE 12.

(6a21+12a1)1=10. We consider ﬁrst how. given a particular solution.79) contains two arbitrary constants and is the general solution of the inhomogeneous equation. a1 and a2 being the coefﬁcients to be determined in this case.12.75). y(x)1=1yh(x)1+1yp(x) is also a solution of the inhomogeneous equation: d2y 2 2 dy p d y d yp dyh dy h +a + by = 2 + a + byh + 2 + a + by p = r ( x ) dx dx dx dx dx dx 2 (12. The function yh is often called the complementary function and yp the particular integral: general solution1=1complementary function1+1particular integral EXAMPLE 12.13 Find the general solution of the equation y″1+13y′1+12y1=12x 2.79) using equations (12.5) of the corresponding homogeneous equation (12. Let yp be a particular solution of the inhomogeneous equation (12.76) Let yh be the general solution (12.78). so that d 2 yp dx 2 7 − 3x + x 2 2 +a dy p dx + by p = r ( x ) (12. a general solution may be obtained. a21=11 so that a21=11.77) (12.78) +a dyh + byh = 0 dx It follows that the sum of the functions yh(x) and yp(x). By Example 12. A particular solution is therefore y= 0 Exercise 30 The method used in this example is the method of undetermined coefﬁcients. a0. called the reduced equation in this context: yh(x)1=1c1 y1(x)1+1c2 y2(x) d 2 yh dx 2 (12. the particular integral is y p ( x) = 7 − 3x + x 2 2 .76) and (12. Constant coefﬁcients This is equal to 2x2 if (2a21+13a11+12a0)1=10.12). a11=1−3 and a01=1722.360 Chapter 12 Second-order differential equations. The function (12.

with roots λ11=1−1 and λ 21=1−2. multiply yp by x before substitution in (12. =) αx c1cos1ω x or c1sin1ω x ce 1cos1ω x or ce 1sin1ω x The table gives the initial choice of particular integral yp corresponding to each function r(x).14 Find the general solution of the equation y″1+13y′1+12y1=13e−2x. Table 12. Example 12. 4. and is summarized in Table 12. EXAMPLE 12. 32 The method of undetermined coefﬁcients 7 − 3x + x 2 2 This method can be used for many elementary functions r(x) in (12. 1. 2. The general solution of the homogeneous equation.1 for some of the more important types. The coefﬁcients in yp are determined by substituting yp into the inhomogeneous equation. the total particular solution is the corresponding sum of the corresponding yp’s.8 Inhomogeneous linear equations 361 The reduced (homogeneous) equation is y″1+13y′1+12y1=10 and has characteristic equation λ21+13λ1+121=1(λ1+11)(λ1+12)1=10. the general solution of the homogeneous equation is yh(x)1=1c1e−x1+1c2e−2x .75). the complementary function. 2. is therefore yh(x)1=1c1e−x1+1c2e−2x and the general solution of the inhomogeneous equation is y ( x ) = yh ( x ) + y p ( x ) = c1e− x + c2 e−2 x + 0 Exercises 31. In that case (a) if the characteristic equation of the homogeneous equation has two distinct roots. multiply yp by x 2 before substitution in (12. 3.1 Term in r(x) 1.11 demonstrates the method for case 2 in the table. In addition.13.75). ce αx Choice of yp keα x a01+1a1x1+1a2 x 21+1-1+1an x n k1cos1ω x1+1l1sin1ω x eα x(k1cos1ω x1+1l1sin1ω x) cx n αx (n1=10. By Example 12.12. This is sufﬁcient unless a term in yp is already a solution of the corresponding homogeneous equation.75). if r(x) is the sum of two or more terms. or (b) if the characteristic equation has a double root.

By Example 12. Constant coefﬁcients By Table 12. and the complementary function is yh(x)1=1(c11+1c2 x)e 2x In this case. the function yp1=1ke 2x is multiplied by x 2. The general solution is therefore 1 y ( x ) = yh ( x ) + y p ( x ) = c1 + c2 x + x 2 e2 x 2 0 Exercise 35 0 Exercises 36–38 y′ 1=12kxe 2x1+12kx 2e 2x.362 Chapter 12 Second-order differential equations. but this is already a solution of the homogeneous equation. the choice of particular integral should be yp1=1ke−2x. p so that y ″1+13y′ 111+12yp1=1−ke−2x p p and this is equal to yp1=13e−2x if k1=1−3. p y ″1=12ke 2x1+18kxe 2x1+14kx 2e 2x p y ″1=1−4ke−2x1+14kxe−2x p . Then yp1=1kx 2e 2x.15 Find the general solution of the equation y″1−14y′1+14y1=1e 2x. so that y ″1−14y′ 111+14yp1=12ke 2x p p and this is equal to e 2x when k1=1122. we use yp1=1kxe−2x Then y′ 1=1ke−2x1−12kxe−2x. by prescription (b).4. the characteristic equation of the homogeneous equation has the double root λ1=12. Then yp1=13xe−2x and the general solution is y(x)1=1yh(x)1+1yp(x)1=1c1e−x1+1c2e−2x1−13xe−2x 0 Exercises 33. case 1. 34 EXAMPLE 12.1. By prescription (a) therefore.

The term F01cos1ω t is an external periodic force that interacts with the motion of the oscillator. is −eE01cos1ω t. with charge q1=1−e. the force acting on an electron. equation (12. a charge q in the presence of an electric ﬁeld experiences a force qE in the direction of the ﬁeld. In the case of an alternating ﬁeld. we have the particular integral (case 3 in Table 12. If the electron is undergoing simple harmonic motion (in the absence of the ﬁeld) the total force acting on it is3 F1=1−kx1−1eE01cos1ω t (12.80) This is the equation of motion for a body moving under the inﬂuence of the force F=m d 2x dt 2 = − kx − c dx + F0 cos ω t dt (12. The general solution of the homogeneous equation corresponding to (12. E1=1E0 cos1ω t. and it is this case that we consider here.84) 2 + ω0 x = 0 For the inhomogeneity r(t)1=1−A1cos1ω t in (12. The term −c1dx2dt is a ‘damping force’ proportional to the velocity representing.83).4) xh(t)1=1d11cos1ω0t1+1d21sin1ω0t (12.83).81) with no damping. 3 . When c1=10. the drag experienced by the body moving in a ﬂuid. d 2x dt 2 is (see Section 12. For example. the quantity ν01=1ω0 22π is called the natural frequency of the oscillator.80) has the form d 2x dt 2 2 + ω 0 x = A cos ω t (12.1) xp(t)1=1c1cos1ω t1+1d1sin1ω t Euler submitted his solution to the problem of the forced harmonic oscillator to the St Petersburg Academy of Sciences on March 30. for example.9 Forced oscillations An important equation in the theory of forced oscillations in mechanical and electrical systems is the inhomogeneous differential equation m d 2x dt 2 +c dx + kx = F0 cos ω t dt (12. 1739.81) The term −kx tells us that we have a harmonic oscillator.12.82) This is an example of (12.9 Forced oscillations 363 12.83) where ω 0 = k m is the angular frequency of the oscillator in the absence of the ﬁeld.

In particular. The particular integral is 0 therefore x p (t ) = A cos ω t 2 ω0 − ω 2 (12. and the particular integral in this case is xp (t ) = A 2ω 0 t sin ω 0 t (12. In the case of resonance. In mechanical systems resonance may be avoided by the application of a suitable damping force. . The required function is (by prescription (a) in Section 12. This phenomenon of large oscillations is called resonance. and tends to inﬁnity as ω1→1ω0. and is an important factor in the study of vibrating systems in both classical and quantum mechanics. is large.87) Substitution in (12. in Figure 12.83) then gives c1=10 and d1=1A22πω0. d 2 xp dt 2 = −ω 2 ( c cos ω t + d sin ω t ) This is equal to A1cos1ω t when d1=10 and c1=1A2(ω 21−1ω 2).83) for the oscillator in the external ﬁeld is x(t)1=1xh(t)1+1xp(t) = d1 cos ω 0 t + d 2 sin ω 0 t + A cos ω t 2 ω0 − ω 2 (12.85) is not the required particular integral because it is already a solution of the homogeneous equation. shows how in the case of resonance the amplitude of the oscillation increases indeﬁnitely with time.9.8) xp(t)1=1t(c1cos1ω0t1+1d1sin1ω0t) (12. when ω1=1ω0. the maximum amplitude of oscillation (the maximum value of x(t)).364 Chapter 12 Second-order differential equations. Constant coefﬁcients Then dxp dt and d 2xp dt 2 2 2 + ω 0 xp = (ω 0 − ω 2 ) ( c cos ω t + d sin ω t ) = −ω c sin ω t + ω d cos ω t . the function (12.86) The solution shows that the behaviour of the system depends strongly on the relative values of ω and ω0.88) The graph of this function. when the frequency ω of the external force is close to the natural frequency of the oscillator.85) and the general solution of equation (12.

... ......... .... . . .. . ...... . . ..... Exercise 3.. ... .. . . ... ...2 12.... .. .... ... . . .. .. . ..... . ....... .. Show that e3x and xe3x are particular solutions of the differential equation y″1−16y′1+19y1=10. . . ... . .. . Solve the boundary value problems: Solve the initial value problems: Section 12... . . .. ... ... .. . ....... ... .. . . ..... ... .. dx dy + 8 y = 0. ... . . . ..4 Find the general solutions of the differential equations: 7. .. . . .......... . .. ... y (1) = 1 . ..... . ... . ... ... .... .......10 0 Exercise 39 d 2x d 2x d2y d2y d2y d 2x d 2x dx 2 dx 2 dt 2 dt dt dt 2 2 2 dx 2 + −2 +6 + 9 x = 0.. . .. ........ .......... . . 13.. .. ... .. .. ... ..... ........ . ..... ... . . . .. (1) = 1 dt dt dx dx + 2 x = 0... ... ... .. .. ... ..... ..... . . .. . . .. . .. . .. . .. . x (1) = 0..... . . .. . . .. . . .... .. .. (0) = 0 dt dt Exercises dx dx + 9 x = 0.... . . . 14............. Show that e−2x and e2x23 are particular solutions of the differential equation 3y″1+14y′1−14y1=10....... . . ... .. ... . .. ... .. . ..... . ...... . ......... .. ..... .. .... .. ..... .. . .... . ... ..... ...... . ... .. ................ .. ... Write down the general solution of the differential equation in 4.. ... ..... . . ... .. . (0) = 0 dt dt dy + 16 y = 0.. ... .. . .. . ....... . .. .. .. .. .. . . .. .. .. ... . . . ... .... . ... . ... . .. . ........ . ... . . . ......... . . ..... . . .. . .. ... . .... .. y = 1 when x = π 2 y ( π 2) = −1... . . .... ... ....... 15. . . .............. . .... . . +4 +8 dx dx − 2 x = 0. . ... .. . . . . . . ... ..... ...... .9 12. 17.. . . . . . 5.... t . x (0) = 1. .... . ......... ..... .......... ... .. .. .. . . ... .. .. ....... ... ... 6. ... . . ..... ....... ...... .. .. ... ...... ....... . . .. . .... .... .. . . . ......... .. ... .. ..... ... . ...1.... . . . . ...... ... . ..... . . . . ..... . . .. .. . ..... ........ .. . ... . . .. ... .. .. .. .. . ... . 3.. .. .. . )t( p x 19.. . ..... .. .... . .. . ... y″1+14y′1+15y1=10 12. .... .......... . ... . .. .. . . .3 Section 12... .. . ...... ..... .. .. Show that cos12x and sin12x are particular solutions of the differential equation y″1+14y1=10. .... . ...... . . .. ... Exercise 2... . . 2y″1−18y′1+13y1=10 9. . .. . .... .. ....... .. .. .. .. . . . ...... .. 18.. .. .. y″1+13y′1+15y1=10 Section 12... . .. ....... . ... .... .. .. ........ dx y = 0 when x = 0. .. ... . .. ...... .. 4y″1+112y1+19y1=10 11..... ....... .. . .. . . ... . ...... .. ... ..... + 9 y = 0... ... ....... ... .... 16. . . y (3π 4) = 1 dx ( π 3) = −1 dt Figure 12... ... . .. .. ... ....... ....... .... . .. ... .... 2. .. ... .... ... . ... .. . ..10 Exercises 365 y (0) = 0... ..... . .. y″1−1y′1−16y1=10 8. ... .. ... .. ..... . ... ... . .. y″1−18y′1+116y1=10 10.... x (0) = 1. . . . . .. . ..... ... Exercise 1... ... .. .... .. .. .. ...... . ..... . ... x ( π 3) = 0.... .... .. .. . . .. . . .

29. y″1−1y′1−16y1=121+13x1+12e−3x1+121cos13x Section 12. nπ x 2 sin 26. and express A and δ in terms of a and b. The diagrams of Figure 12. 23. y″1−1y′1−12y1=13e−x 35.44) for the particle in a box of length l with potential-energy function V1=10 for −l221≤1x1≤1+l22.8 30.53) if x is replaced by x1+1l22. d2y dx 2 + dy − 2 y = 0.m. y″1−1y′1−16y1=121+13x 32. Section 12. Section 12.366 Chapter 12 Second-order differential equations. ﬁnd the nodes and sketch the graph of the wave function ψn for (i) n1=14 and (ii) n1=15. 28. For the particle in the box in Section 12. where A is the amplitude of the vibration and δ is the phase angle. an inductor (inductance L). Given that the general solution of the equation of motion mP1=1−kx for the harmonic oscillator is x(t)1=1a1cos1ω t1+1b1sin1ω t. and a capacitor (capacitance C) connected in series with a source of e.72) and its solutions (12. dx d 2θ dt 2 y (0) = 2.f.5 22. except for a possible change of sign. Section 12. Find a particular solution of the differential equation y″1−1y′1−16y1=121+13x. For the particle in a box. Section 12. (i) Solve the Schrödinger equation (12. y → 0 as x → ∞ 21. H(0)1=1ω. (iii) Show that the solutions are the same as those given by (12. show that wave functions ψ n ( x ) = l l for n1=11 and n1=12 are (i) normalized.65) by means of the change of variable θ1=1x2r.9 39. V1=1∞ for x1≤1−l22 and x1≥1+l22. Solve + a 2θ = 0 subject to the condition θ (t1+12πτ)1=1θ (t). Find the general solutions of the differential equations: 31. (ii) orthogonal.8 are maps of the signs and nodes of some real wave functions (12. (i) show that the solution can be written in the form x(t)1=1A1cos (ω t1−1δ ). Draw the corresponding diagrams for (i) n1=1±3. y″1−18y′1+116y1=111−14x 3 −3x 33. y″1−1y′1−16y1=12e 34. Constant coefﬁcients 20.71) for the particle in a ring. . (ii) orthogonal. y″1−1y′1−16y1=121cos13x 37. H(0)1=1u0. For the particle in a ring show that wave functions ψ n (θ ) = 1 2 π einθ for n1=13 and n1=14 are (i) normalized. An RLC-circuit contains a resistor (resistance R). y″1+14y1=131sin12x 38.74) are transformed into (12. where ω = k m .6 24.6. (ii) ﬁnd the amplitude and phase angle for the initial conditions x(0)1=11. Solve the equation of motion for the harmonic oscillator with initial conditions x(0)1=10. 25.7 27. y″1−18y′1+116y1=1e4x 36. E. Verify that equation (12. (ii) Show that the solutions ψn are even functions of x when n is odd and odd functions when n is even.62) and (12. (ii) n1=1±4.

12. ωC R . tan δ = 1 S . E(t)1=1E0 sin1ω t is Ip(t)1=1I01sin (ω t1−1δ ) where I 0 = E0 R + S2 2 . and S = ω L − .10 Exercises 367 (i) Use Kirchhoff ’s voltage law (Section 11.f.7) to show that the current I(t) in the circuit is given by the inhomogeneous equation L d 2I dt 2 +R dI I dE + = dt C dt (ii) Find the solution of the homogeneous equation (for dE2dt1=10).m. and conﬁrm that it decays exponentially as t1→1∞. (iii) Show that the particular integral for the periodic e.

outlined in Section 13. The former is used in Section 13.1.47).5 to 12.62) for the quantum-mechanical particle in a ring as a periodic boundary value problem. The standard methods used to solve the equations in Table 13. . and (12.2.7 how three very different physical problems are modelled in terms of the same differential equation: equation (12. and they have particular solutions that play an important role in the mathematics of the physical sciences.1 are all second-order homogeneous linear equations with non-constant coefﬁcients. described in Section 13.3. These solutions are called Legendre polynomials and they occur whenever a physical problem in three dimensions is formulated in terms of spherical polar coordinates. the same equation. Some of these are listed in Table 13. Table 13.1.35) for the classical harmonic oscillator as an initial value problem.4 to obtain the physically-signiﬁcant particular solutions of the Legendre equation. These solutions are often called special functions. and the more general Frobenius method. This is a common phenomenon in the mathematical modelling of physical systems. Subsequent sections are devoted to brief descriptions of the other special functions that are solutions of the equations listed in Table 13.13 Second-order differential equations. (12. Some special functions 13. for the quantum-mechanical particle in a box as a boundary value problem.1 and many other linear differential equations in the sciences are the power-series method. and a number of equations are important enough to have been given names.1 Name Legendre Associated Legendre Hermite Laguerre Associated Laguerre Bessel Equation (11−1x 2)y″1−12xy′1+1l(l1+11)y1=10 (11−1x 2)y″1−12xy′1+1[l(l1+11)1−1m22(11−1x 2)] y1=10 y″1−12xy′1+12ny1=10 xy″1+1(11−1x)y′1+1ny1=10 xy″1+1(m1+111−1x)y′1+1(n1−1m)y1=10 x 2y″1+1xy′1+1(x 21−1n2)y1=10 The equations in Table 13.1 Concepts We saw in Sections 12.

substituting in the differential equation.1 and 13. It can therefore be used to solve the Legendre. we express the solution as the power series y = a0 + a1 x + a2 x 2 + a3 x 3 + Then dy = a1 + 2 a2 x + 3a3 x 2 + dx = = m= 0 ∑ am x m ∞ m =1 ∑ mam x m−1 ∞ and.1 Use the power-series method to solve the equation dy + y=0 dx By equation (13.1) m= 0 ∑ am x m ∞ (13.13. =.2 The power-series method 369 13. and the particular solutions for a physical system are obtained by the application of the appropriate boundary or initial conditions. q(x) and r(x) are polynomials or if they can be expanded as power series in x.2).2 by solving two equations for which we already know the solutions. The power-series method can be used when p(x).2 The power-series method Many important second-order linear differential equations y″1+1p(x)y′1+1q(x)y1=1r(x) have at least one particular solution that can be expressed as a power series. EXAMPLE 13. a1. a ﬁrst-order equation in Example 13.2.1 and a second-order equation in Example 13. y ( x ) = a0 + a1 x + a2 x 2 + a3 x 3 + = (13. associated Legendre and Hermite equations (the ﬁrst two are transformed into the standard form (13.1) by division by (11−1x 2)). dy + y = ( a1 + 2 a2 x + 3a3 x 2 + dx ) + ( a0 + a1 x + a2 x 2 + ) =1(a11+1a0)1+1(2a21+1a1)x1+1(3a31+1a2)x21+1- .2) The series is substituted into the differential equation to determine the numbers a0. We demonstrate the power-series method in Examples 13. a2.

He then solved the equation by a method equivalent to that given in this example to obtain the power-series representation of the function. The ability to represent functions as power series was a vital element in the development of the calculus by both Leibniz and Newton. = and. 3a31+1a21=10. so that a1 = − a0 .370 Chapter 13 Second-order differential equations. 2 EXAMPLE 13. Some special functions For this to be equal to zero for all values of x (within the radius of convergence of the power series) the coefﬁcient of each power of x must be zero: a11+1a01=10. 0 Exercises 1. am1=1(−1)ma02m!.4(ii) with k1=1−1). a3 = − a2 3 =− a0 3! . . ∞ y ′′ = m= 2 a ∑ m( m − 1)am x m−2 ∞ m= 2 ∞ ∑ m( m − 1)am x m− 2 = (2 × 1)a2 + (3 × 2)a3x + (4 × 3)a4 x 2 + ∑ ( m + 2) ( m + 1)am+ 2 x m ∞ m= 0 1 Leibniz discovered the differential equation for the sine function in 1693 by a geometric argument. ∞ y′ = m =1 ∑ mam x m−1 .2 Use the power-series method to solve the equation1 d2y dx 2 + y=0 By equation (13. y= m= 0 ∑ ∞ am x m = ∞ a0 ( −1) m x m (− x )m ∑ m! = a0 ∑ m! m= 0 m= 0 ∞ We recognize the inﬁnite series as the power-series expansion of the function e−x: e− x = 1 − x + x 2 x3 − + 2! 3! = (− x )m ∑ m! m= 0 ∞ Therefore y1=1a0e−x where a0 is an arbitrary constant (see Example 11. a2 = − a1 2 =+ a0 2! .2). we have y= We can write y ′′ = = m= 0 ∑ am x m . 2a21+1a11=10. Therefore. in general.

22) with ω 1=11). and for the odd values of m. a7 = − a5 7×6 =− a1 . a2 = − a0 2 ×1 . is also known for his work in matrix algebra. . a6 = − a4 6×5 =− a0 6! . a4 = − a2 4×3 =+ a0 4! .3 The Frobenius method The Frobenius method2 is an extended power-series method that is used to solve second-order linear equations that can be written in the form y ′′ + 2 b( x ) c( x ) y′ + 2 y = 0 x x (13.3) Georg Frobenius (1849–1917). y = a0 + a2 x 2 + a4 x 4 + + a x + a x 3 + a x5 + 3 5 1 a1 3× 2 =− a1 3! .13. a3 = − Therefore. 0 Exercises 3–5 13. German mathematician.3 The Frobenius method 371 Then y ′′ + y = = m= 0 ∞ ∑ ∞ am x m + m= 0 ∑ ( m + 2) ( m + 1)am+ 2 x m ∞ m= 0 ∑ am + ( m + 2) ( m + 1)am+ 2 x m and this is zero if the coefﬁcient of each power of x is separately zero: am1+1(m1+12)1(m1+11)am+21=10 For the even values of m. Therefore y1=1a01cos1x1+1a11sin1x where a0 and a1 are arbitrary constants (see equation (12. a5 = − a3 5× 4 =+ a1 5! . 7! x 2 x 4 x6 = a0 1 − + − + 2! 4! 6! 7 3 5 + a x − x + x − x + 1 3! 5! 7! We recognize the two series in square brackets as the power-series expansions of cos1x and sin1x. In a monograph in 1878 he organized the theory of matrices into the form that it has today.

the equation has been multiplied by x 2). c(x)1=1c0. Some special functions in which b(x) and c(x) are polynomials or can be expanded as power series in x. for the coefﬁcient of x r (m1=10).7) The coefﬁcient of each power of x in this equation must be zero so that. and for many of the linear differential equations in the sciences.1. c(x)1=1c01+1c1x1+1c2 x 21+1c3 x 31+1- but the indicial equation (13. We consider ﬁrst the case of b(x)1=1b0.8) This is called the indicial equation.8) remains valid.3) are polynomials or can be expanded as power series in x: b(x)1=1b01+1b1x1+1b2 x 21+1b3 x 31+1-. In general. (13. We have y1=1x r[a01+1a1x1+1-] y′1=1x r−1[ra01+1(r1+11)a1x1+1-] y″1=1x r−2[r(r1−11)a01+1(r1+11)ra1x1+1-] and substitution into (13.4).3) has at least one solution that can be expressed as y ( x ) = x r ( a0 + a1 x + a2 x 2 + a3 x 3 + ) = x r ∑ am x m m= 0 ∞ (13.5) (for convenience.4) where r (which may be zero) is an indicial parameter to be determined. It includes the simple power-series method discussed in the previous section as a special case and can be used for all the equations in Table 13. in full. where b0 and c0 are constants: x 2y″1+1b0 xy′1+1c0 y1=10 (Euler– Cauchy equation) (13. and a01≠10. Every differential equation of the form (13.372 Chapter 13 Second-order differential equations. . and the roots are the possible values of the parameter r in (13. r 21+1(b01−11)r1+1c01=10 (13.6) m= 0 ∑ (r + m)2 + (b0 − 1)(r + m) + c0 am x m+ r = 0 ∞ (13. b(x) and c(x) in (13.5) gives xr[r(r1−11)a01+1(r1+11)ra1x1+1-] +1b0xr[ra01+1(r1+11)a1x1+1-]1+1c0 xr[a01+1a1x1+1-]1=10 or.

and y1(x) and y2(x) are independent particular solutions.4). y1(x)1=1x r(a01+1a1x1+1a2 x 21+1-) and the second solution is y2(x)1=1y1(x)1ln1x1+1x r+1(A01+1A1x1+1A2 x 21+1-) (13. c01=1122. and the indicial equation is r 21−13r221+11221=10 with roots r11=11 and r21=1122.10a) (13. and the other depends on the solutions. 1 Distinct roots not differing by an integer Both y1 and y2 have the form (13.10b) . 0 Exercises 6–9 Once the values of the indicial parameter have been determined.4). Then r 21−11241=1(r1−1122)(r1+1122)1=10 and the indicial parameters are r1=1±122.11a) (13. The general solution is y(x)1=1c1 y1(x)1+1c2 y2(x) (13. the solution of the differential equation can proceed as in the power-series method. (ii) x 2y″1+1xy′1+1(x 21−1124)y1=10 This is the Bessel equation (see Table 13. We have the following three cases. At least one of y1 and y2 has the form (13. (iii) xy″1+12y′1+14xy1=10 We have x 2y″1+12xy′1+14x 2y1=10.3 The Frobenius method 373 EXAMPLE 13. b(x)1=121=1b0 and c(x)1=1x 2 so that c01=10.11b) (13. one solution has the form (13. r1 and r2.9) in which c1 and c2 are arbitrary constants. with roots r11=10 and r21=1−1.1) for n1=1122. of the indicial equation.3 Indicial equations (i) x 2 y ′′ − 1 1 xy ′ + y = 0 2 2 We have b01=1−122.13. We have b(x)1=111=1b0 and c(x)1=1x 21−1124 so that c01=1−124. y1(x)1=1x r1(a01+1a1x1+1a2 x 21+1-) y2(x)1=1x r2(A01+1A1x1+1A2 x 21+1-) 2 Double root If r11=1r21=1r.4). The indicial equation is r 21+1(b01−11)r1+1c01=1r 21+1r1=10. Therefore.

12a) EXAMPLE 13.12b) (13.12).3(ii). 2 ∞ y ′′ = m= 0 ∑ ( m 2 − 1 ) am x m − 3 2 4 ∞ = x1 2 2 a1 x + ( a0 + 2 × 3a2 ) x 2 + ( a1 + 3 × 4 a3 ) x 3 + ( a2 + 4 × 5a4 x 4 ) + . equations (13. In the present case. y1(x)1=1x r1(a01+1a1x1+1a2 x 21+1-) and the second solution is y2(x)1=1ky1(x)1ln1x1+1x r2(A01+1A1x1+1A2 x 21+1-) in which r11>1r2 and the constant k may be zero. and we show here that the particular solution with indicial parameter r1=1+122 is the Bessel function J1 2 ( x ) = We have y = x1 2 ∑ am x m = m= 0 ∞ 2 sin x πx m= 0 ∑ am x m+1 2 ∞ Then y′ = and 1 x 2 y ′′ + xy ′ + ( x 2 − 4 ) y = x1 2 ∑ m( m + 1) x m + x m+ 2 m= 0 ∞ m= 0 ∑ ( m + 1 ) am x m−1 2 . (13. without a logarithmic term.4). there is no logarithmic term (see Exercise 13). however.4). By Example 13. Some special functions 3 Roots differ by an integer One solution has the form (13.4 The Bessel equation x 2y″1+1xy′1+1(x 21−1n2)y1=10 for n1=1±122.374 Chapter 13 Second-order differential equations. the indicial roots are r1=1±122 and the solutions are nominally of type 3. 0 Exercises 10–12 Many of the particular solutions of second-order differential equations that are of interest in the physical sciences are generally series of type (13.

as in Example 13.13) can be solved by the series method in the way described in Section 13. 2 a4 = − ( l − 2) ( l + 3) ( l − 2)l ( l + 1) ( l + 3) a2 = + a0 . The series y = ∑ am x m = a0 + a1 x + a2 x 2 + m (13. Of interest in the physical sciences are then those solutions that are ﬁnite in the interval −11≤1x1≤1+1 (corresponding to 01≤1θ1≤1π). in which case the variable x is replaced by cos1θ. 0 Exercises 13–15 13. The same procedure with r1=1−122 gives the Bessel function of order −122 (Exercise 13).14) and its derivatives y′ and y″ are substituted into (13. Then. a2 = − l ( l + 1) a0 .15) For the even values of m. Equation (13. 3! 0 a4 = − 1 1 a = + a0 .4 The Legendre equation 375 This is zero if the coefﬁcient of each power of x is separately zero.4 The Legendre equation (11−1x 2)y″1−12xy′1+1l(l1+11)y1=10 (13.13).2. The coefﬁcient of x 1 is 2a1 so that a11=10 and it follows that am1=10 for odd values of m. setting the coefﬁcient of each power of x equal to zero gives the recurrence relation am+ 2 = m( m + 1) − l ( l + 1) ( l − m) ( l + m + 1) a =− a ( m + 1) ( m + 2) m ( m + 1) ( m + 2) m (13.13. This equation arises whenever a physical problem in three dimensions is formulated in terms of spherical polar coordinates.13) The Legendre equation is where l is a real number. For the even values. a2 = − Therefore. 5×4 2 5! a6 = − 1 1 a = − a0 .2. x 2 x 4 x6 y ( x ) = a0 x1 2 1 − + − + 3! 5! 7 ! =1a0 x−1221sin1x By convention a0 = 2 π and the function J1 2 ( x ) = 2 πx 1 a . r. θ and φ (see Section 10. … 3× 4 4! .2). 7×6 4 7! 3 5 7 = a x −1 2 x − x + x − x + 0 3! 5! 7 ! sin x is the Bessel function of order n1=1122.

The choice a11=10 in (13.17) (13. It is also possible to ﬁnd a solution involving inverse powers of x that is valid for |1x1|1>11. the ratio of consecutive terms is. the series y2 only odd powers. and they are the solutions of the Legendre equation that are of . … 5! A power-series solution of the equation is therefore y(x)1=1a0 y1(x)1+1a1 y2(x) where a0 and a1 are arbitrary constants and y1 ( x ) = 1 − l ( l + 1) 2 ( l − 2)l ( l + 1) ( l + 3) 4 x + x − 2! 4! ( l − 1) ( l + 2) 3 ( l − 3) ( l − 1) ( l + 2) ( l + 4) 5 x + x − 3! 5! (13.18) reduces to a polynomial of degree l when l is an odd integer.18) for power series. but the resulting set of polynomials is identical to that obtained for positive values.17) reduces to a polynomial of degree l when l is an even integer or zero. a3 = − ( l − 1) ( l + 2) a1 .376 Chapter 13 Second-order differential equations.16).16) gives a particular solution that is valid for all values of x. both series converge if |1x1|1<11 and diverge if |1x1|1>11 (unless l is an integer. Both series have radius of convergence R1=11 for arbitrary values of l. The function (13. is therefore the general solution of the Legendre equation in the interval −11<1x1<11. Convergence We ﬁrst consider the case of non-integer values of l. when l1=12 the series terminates after the second term to give y1(x)1=111−13x2. The same considerations apply for negative integer values of l.15). Similarly. Some special functions Similarly for the odd values of m.18) The series y1 contains only even powers of x. For example.16) y2 ( x ) = x − (13. 3! a5 = + ( l − 3) ( l − 1) ( l + 2) ( l + 4) a1 . am+ 2 am = m( m + 1) − l ( l + 1) →1 ( m + 1) ( m + 2) m→∞ as so that. Thus. These particular solutions are called Legendre polynomials Pl (x). with two arbitrary constants. by d’Alembert’s ratio test (7. and the choice a01=10 in (13.16) therefore gives a particular solution for even values of l that is ﬁnite for all values of x. the function (13. The Legendre polynomials The function (13. Both series also diverge when |1x1|1=11. by equation (13. see below).

and an American edition. the nonzero arbitrary constant in (13. 2 y ′′ = 15x 3 Adrien-Marie Legendre (1752–1833) has the doubtful distinction of the following footnote in E. was inﬂuential in the United States.4 The Legendre equation 377 interest in the physical sciences. The polynomials appeared in Legendre’s Recherches sur l’attraction des sphéroïdes homogènes of 1785 as the coefﬁcients in the expansion of the potential function (11−12h1cos1θ1+1h2)−122 in powers of h. The ﬁrst few of these polynomials are P0(x)1=11 1 P2 ( x ) = (3x 2 − 1) 2 1 P4 ( x ) = (35x 4 − 30 x 2 + 3) 8 P1(x)1=1x 1 P3 ( x ) = (5x 3 − 3x ) 2 1 P ( x ) = (63x5 − 70 x 3 + 15x ) 5 8 (13. Then. for l1=13. much of his best work was absorbed or circumvented by younger mathematicians’. 1. 3. (11−1x 2)y″1−12xy′1+1l(l1+11)y 3 1 = (1 − x 2 ) × 15x − 2 x × (5x 2 − 1) + 3 × 4 × (5x 3 − 3x ) 2 2 =1(−151−1151+130)x31+1(151+131−118)x1=10 0 Exercise 16 3 y ′ = (5x 2 − 1). Davies’ Legendre (1851). T. 2.19) ( 2l − 1) and the series is to be continued down to the constant term. =.20) EXAMPLE 13. Pl ( x ) = 1· 3· 5 l! l ( l − 1) l − 2 l ( l − 1) ( l − 2) ( l − 3) l − 4 x − × xl − x + 2( 2l − 1) 2 · 4( 2l − 1) ( 2l − 3) (13. for l1=10. 2 Therefore. We have 1 y = P3 ( x ) = (5x 3 − 3x ).13) for l1=13.13.3 By convention.16) is chosen so that Pl (1)1=11.5 Show that the polynomial P3(x) is a solution of the Legendre equation (13. . Bell’s Men of mathematics: ‘Considerations of space preclude an account of his life. His textbook Éléments géométrie of 1794 did much to reform the teaching of geometry (till then based on Euclid).

21) The associated Legendre equation is (see Table 3. the variable x is identiﬁed with cos1θ in physical applications.378 Chapter 13 Second-order differential equations.23) The corresponding particular solutions are called the associated Legendre functions Pl|m| (x). 1 1 P2 = (3xP − P0 ) = (3x 2 − 1) 1 2 2 For l1=12 the relation (13. = m1=10. ±1. =. Some special functions The Legendre polynomials satisfy the recurrence relation (l1+11)Pl+1(x)1−1(2l1+11)xPl (x)1+1lPl−1(x)1=10 so that given P0(x)1=11 and P1(x)1=1x.22) As with the Legendre equation (m1=10). 1.6 Use the recurrence relation (13. Therefore. l EXAMPLE 13. 1 1 1 1 P3 = (5xP2 − 2 P ) = 5x × (3x 2 − 1) − 2 x = (5x 3 − 3x ) 5 1 2 3 3 2 0 Exercise 17 The associated Legendre functions (13. EXAMPLE 13. 2.1) m2 y=0 (1 − x 2 ) y ′′ − 2 xy ′ + l ( l + 1) − (1 − x 2 ) (13.24) to derive the associated Legendre functions P m for m1=11. For l1=11 the relation (13.121and13. Therefore. ±l (13. and express these as functions of θ when x1=1cos1θ and 3 (11−1x 2)1221=1sin1θ.7 Use the formula (13.24) . In this case. all higher polynomials can be derived. They are related to the Legendre polynomials by the differential formula Pl|m| ( x ) = (1 − x 2 ) |m| 2 with P 01=1Pl. d |m| dx|m| Pl ( x ) (13. 3. with |1m1|1≤1l: l1=10.21) to ﬁnd P2(x) and P3(x). ±2.21) is 2P21−13xP11+1P01=10.21) is 3P31−15xP21+12P11=10. the solutions obtained by the power-series method converge in the interval −11≤1x1≤11 when both l and m are integers.

8 Show that P1 is orthogonal (a) to P2 and (b) to P3. The Legendre polynomials are orthogonal in the interval −11≤1x1≤11: Z Pl ( x ) Pl ′ ( x ) dx = 0 −1 +1 when l ≠ l′ (13. . The integrand is odd and the integral is zero (see Section 5.25) EXAMPLE 13. P2 ( x ) = (3x 2 − 1). P3 ( x ) = (5x 3 − 3x ).4 The Legendre equation 379 We have 1 y = P3 ( x ) = (5x 3 − 3x ).58)).13. We have 1 1 P ( x ) = 1. 2 Then 3 P1 ( x ) = (1 − x 2 )1/ 2 (5x 2 − 1) 3 2 P 2(x)1=115(11−1x 2)x 3 P 3(x)1=115(11−1x 2)322 3 0 Exercise 18 Orthogonality and normalization 3 y ′ = (5x 2 − 1).6 that the solutions of the Schrödinger equation for the particle in a box are orthogonal functions (equation (12. 2 y ′′ = 15x.3). 1 2 2 Then +1 +1 +1 1 3x 4 x 2 1 3 P ( x ) P2 ( x ) dx = Z (3x − x ) dx = − 1 2 4 2 −1 2 −1 (a) Z −1 = 1 3 1 3 1 − − − = 0 2 4 2 4 2 In this case. P1(x) is an odd function of x whereas P2(x) is an even function. Orthogonality is a property of the solutions of a class of second-order linear differential equations called Sturm– Liouville equations that includes the Legendre equation. y ′′′ = 15 3 P1 (cos θ ) = sin θ (5 cos2 θ − 1) 3 2 P 2(cos1θ )1=1151sin21θ1cos1θ 3 P 3(cos1θ )1=1151sin31θ 3 We saw in Section 12.

Some special functions (b) Z P1 ( x ) P3 ( x ) dx = Z (5x 4 − 3x 2 ) dx = −1 −1 +1 1 2 +1 +1 1 1 5 3 0 − 0 = 0 x − x = −1 2 2 In this case. Z +1 −1 (P | m| x l ( ) ) dx = (2l2+ 1) ((ll + || m |)!! − m |) 2 (13.27) and this result is used to construct the set of normalized associated Legendre functions (and normalized Legendre polynomials when m1=10). (ii) P 2 is orthogonal to P 2. EXAMPLE 13. 1 3 2 3 (i) 3 1 P1 ( x ) = (1 − x 2 )1 2 .6).26) In addition when l1=1l′. P3 ( x ) = (1 − x 2 )1 2 (5x 2 − 1) 1 2 I =Z +1 1 P1 ( x ) P3 ( x ) dx = 1 −1 3 Z (1 − x 2 )(5x 2 − 1) dx 2 −1 +1 .9 Show that (i) P 1 is orthogonal to P 1.29) When x1=1cos1θ these functions form part of the solutions of the Schrödinger equation for the hydrogen atom (Section 14. both P1(x) and P3(x) are odd functions so that the orthogonality of the functions is a new property.380 Chapter 13 Second-order differential equations. m ( x )Θl ′. 0 Exercise 19 The corresponding property of the associated Legendre functions is Z Pl|m| ( x ) Pl|m| ( x ) dx = 0 ′ −1 +1 when l ≠ l′ (13. m ( x ) = with property (13. not a consequence of even2odd parity. m ( x ) dx = δ l .28) Z Θl . ( 2l + 1) ( l − | m |)! |m| P ( x) 2 ( l + | m |)! l Θl . l ′ = −1 +1 1 0 if if l = l′ l ≠ l′ (13.

The equation arises in the solution of the Schrödinger equation for the harmonic oscillator. It is solved by the power-series method to give a general solution y(x)1=1a0 y1(x)1+1a1 y2(x) in which the particular solution y1 contains only even powers of x and y2 only odd powers of x. and I1=10 (Section 5. 2. P 2(x)1=115x(11−1x2) 3 2 I =Z +1 2 2 P2 ( x ) P3 ( x ) dx −1 = 45Z +1 x (1 − x 2 )2 dx −1 The integrand is an odd function of x in the interval −11≤1x1≤11. and similarly for y2 when n is an odd integer. These polynomials are called Hermite polynomials Hn(x). = The ﬁrst few of these polynomials are H0(x)1=11 H2(x)1=14x 1−12 2 H1(x)1=12x H3(x)1=18x 31−112x (13.3. Therefore (Section 5.13. equation (5.25)) I = 3Z (1 − x 2 )(5x 2 − 1) dx = 3Z ( −5x 4 + 6 x 2 − 1) dx 0 1 = 3 − x5 + 2 x 3 − x = 0 0 0 +1 +1 (ii) P 2 (x)1=13(11−1x2).5 The Hermite equation The Hermite equation is y″1−12xy′1+12ny1=10 (13. 3. 1. equation (5. y1 reduces to a polynomial of degree n when (positive) n is an even integer.5 The Hermite equation 381 The integrand is an even function of x in the interval −11≤1x1≤11. As in the case of the Legendre equation. and they are those solutions of the Hermite equation that are of interest in the physical sciences: H n ( x ) = (2 x )n − n( n − 1) n( n − 1) ( n − 2) ( n − 3) (2 x )n− 2 + (2 x )n− 4 − 1! 2! (13.31) for n1=10.3.32) .26)) 0 Exercise 20 13.30) where n is a real number.

31) to ﬁnd H4(x).10 (i) Use the series expansion (13.35) with second derivative y″1=1e−x 22[u″1−12xu′1−1(11−1x 2)u] 2 Substitution of y and its second derivative in (13. Some special functions The Hermite polynomials satisfy the recurrence relation Hn+1(x)1−12xHn(x)1+12nHn−1(x)1=10 so that. all higher polynomials can be derived.34) gives e−x 22[u″1−12xu′1+12nu]1=10 2 .382 Chapter 13 Second-order differential equations. (13. the Hermite equation is y ′′ − 2 xy ′ + 8 y = 192 x 2 − 96 − 2 x 64 x 3 − 96 x + 8 16 x 4 − 48 x 2 + 12 = 0 (iii) Use the recurrence relation (13. H4′(x)1=164x31−196x.34) (13.33) to ﬁnd H5(x). H51=12xH41−18H31=12x[16x41−148x21+112]1−18[8x31−112x]1=132x51−1160x31+1120x 0 Exercise 21 Hermite functions A differential equation that is related to the Hermite equation is y″1+1(11−1x 21+12n)y1=10 and this is solved by making the substitution y(x)1=1e−x 22u(x) 2 (13. H4″(x)1=1192x21−196 Therefore.30) that H4(x) is a solution of the Hermite equation. for n1=14. given H01=11 and H11=12x. H 4 ( x ) = ( 2 x )4 − 4×3 4 × 3× 2 ( 2 x )2 + ( 2 x )0 = 16 x 4 − 48 x 2 + 12 1! 2! (ii) Verify by substitution in (13.33) EXAMPLE 13.

37) 383 .. .. ... . .. . ... ...... . . is a solution of the Schrödinger equation for the harmonic oscillator −∞ +∞ αψ ′′ + d 2ψ ym ( x ) yn ( x ) dx = Z 2 + k 2 z ψ = Eψ 2α 1 2 kx ψ = Eψ 2 yn(x)1=1e−x 22Hn(x). . .. . . . .... ... ... .. . .. . ...... ...... . ) (0 x H 2/ x−e ... . . .. . . . . .. . . ... ..... .. ... . . .... . . . .. . . .. .. .. . . .. .. ...... .. ........ . . .. . .. .... . .. ...... ... ... . . .. .. .. . . ... .. ... ... .. ... ..... . . .. . ....36) (13.. .... .. .... ....... ... .... . . ... ...... . . . ... . .. α= and Let z = α x... . .. . .... . .... .. .. .... ﬁnite for all values of x.. 3. .. ... . . . ....... . . ... . are orthogonal. . . ... dx dz dx dz Z 2m 2 2 2 2 m dx . . for ψ = ψ ( α x ) = ψ ( z ).. . .. .... . ... . .... .. 0 .. .. . ...... ... . ..... ........... ) (2 x H. . .. . .... .. ) (1 x H.. .. ... The graphs of the ﬁrst three functions are shown in Figure 13. .5 The Hermite equation 2 (13.. . ... ... ... ..... ....... .. ...... . .... .. ... .. ... ... ...... .... ... .. . .. ... . .. .. ... . .. .... . ...... ..x .. . .. . 2...... . . .... .... .. ..... ....... .. .... . .... . .. ....... ........ . ......... ... ........ ..... ...... ... . . .. .. 1.... . . .. ..... . ... .. .... . ... .. . . .. ... . . .1 e− x H m ( x ) H n ( x ) dx = 2 n n! π δ m. .... ... . .. .... .. ... .. ... . .. .. .. .. .... .. . . . . .... .. ... . .. . . . . ...... . . ... . . EXAMPLE 13. . . . .. . . ... . . .. ... .2/x−e ... ..... . . . ... . .. ... .. . .. . .... .....2/x−e .. . so that particular solutions of (13. . . . .. ...... ... ... km − − d ψ d ψ dz dψ = = α = αψ ′. ... ... ... . . .. ...34) are the Hermite functions These functions.. ...... . . . ... ...... . = d 2ψ dx 2 = αψ ′′ 13. .. . ... ..11 Show that the Hermite function e−α x 22 H n ( α x ). . 0 ... . . ... .. ... . .. . .... .. . .... . . .. n 2 n1=10.... ... ...... . ... . . where The expression in square brackets is the left side of the Hermite equation (13.. . ... 0 . ...... .. . .. .. . ...... .... ... . . . . .. .. ..... .. . . . .... . .. .. . ... . . .. .. .. .. .... . . . ...... . with property 0 Exercise 22 The Hermite functions are of interest in chemistry because they are the eigenfunctions of the quantum-mechanical problem of the harmonic oscillator. . . .. ... ..... . . . Then. ...... .. ... 2 −∞ +∞ Figure 13.. .... . . ... . . .... ... .. ... . . . . . ....... . .. .. .. . .. . ..... .. ... .. .... . .. . ...... .... .. . .. .. ..30)... . .. .. . ...1. . ...... ... ...... .. .... . ..

2. when n is a positive integer or zero.41) The associated Laguerre equation is xy″1+1(m1+111−1x)y′1+1(n1−1m)y1=10 (13. 0 Exercises 23. 3.6 The Laguerre equation (13. given L0 and L1. The Hermite functions are therefore the eigenfunctions of the quantum-mechanical harmonic oscillator problem. 1. = The ﬁrst few of these are L0(x)1=11 L2(x)1=121−14x1+1x 2 + ( −1) n n! (13. n = 0. where ω = k m is the angular frequency of the oscillator.40) The Laguerre polynomials satisfy the recurrence relation Ln+1(x)1−1(11+12n1−1x)Ln(x)1+1n 2Ln−1(x)1=10 from which. They are related to the Laguerre polynomials by the differential formula Lm ( x ) = n 4 dm dx m Ln ( x ) (13. .38) The Laguerre equation4 xy″1+1(11−1x)y′1+1ny1=10 where n is a real number.42) and has polynomial solution when both n and m are positive integers or zero.43) Edmond Laguerre (1834–1886). is a polynomial of degree n called a Laguerre polynomial Ln(x): n2 n −1 n2 ( n − 1)2 n − 2 Ln ( x ) = ( −1) n x n − x + x − 1! 2! for n1=10.384 Chapter 13 Second-order differential equations. with m m1≤1n. 3. 13. all higher polynomials can be found.34). we obtain ψ ′′ + (1 − z 2 + 2 n)ψ = 0 if E= n+ ( 1 2 ) ω. These solutions are the associated Laguerre polynomials Ln (x). … This differential equation is identical to equation (13. has a power-series solution that. 1. Some special functions By dividing throughout by −A2α22m and noting that α1=1mω2A. 24 Associated Laguerre functions (13. 2.39) L1(x)1=111−1x L3(x)1=161−118x1+19x 21−1x 3 (13.

of whom one is only moderately prepared. This equation ranks with the Legendre equation in its importance in the physical sciences. 1.7 Bessel functions The Bessel equation is5 x2y″1+1xy′1+1(x 21−1n2)y1=10 (13. and they occur there in the form of associated Laguerre functions f n. The functions are important in the formulation of the theory of scattering processes. when divided by x2. is of type (13. =. and the same solutions are found for the Schrödinger equation for the particle in a circular box and in a spherical box. Bessel functions are involved. 3. l ( x ) x dx = Z e− x x 2l L2l++l1 ( x ) L2l′+1 ( x ) x 2 dx n n l 0 (13. Such are the burdens =’ 5 . and the third lacks both preparation and ability. Equation (13. in the solution of the classical wave equation for the vibrations of circular and spherical membranes. (n1−11). Examples of Bessel functions were discussed by Daniel Bernoulli. by expressing the solution in the form (13. n′ 13. 1. German astronomer. for example.47) where n is a real number. the other is less than moderately. Bessel is known as the recipient of numerous letters from his close friend Gauss. 2.4) y(x)1=1x r(a01+1a1x1+1a2 x 21+1-) ‘At this junction It is time to wrestle With a well-known function Due to Herr Bessel’.13.3) and is solved by the Frobenius method.44) for n1=10.47). and Lagrange. In 1810 Gauss wrote: ‘This winter I am giving two courses of lectures to three students. that is.45) and they are orthogonal with respect to the weight function x2 in the interval 01≤1x1≤1∞: ∞ 2 0 ∞ + Z f n. although it is met less frequently in chemistry than in physics and engineering.46) = 2n ( n + l ) ( n − l − 1)! 3 δ n. l1=10. but the ﬁrst systematic study appeared in 1824 in a paper by Bessel on perturbations of planetary orbits.6). l ( x ) f n′ . =. 2. Friedrich Wilhelm Bessel (1784–1864). These functions satisfy the differential equation f ′′ + n l ( l + 1) 1 2 f′+ − − f =0 x 4 x x2 (13. Euler. l ( x ) = e− x / 2 x l L2 l++l1 ( x ) n (13.7 Bessel functions 385 The associated Laguerre polynomials arise in the solution of the radial part of the Schrödinger equation for the hydrogen atom (Section 14.

. .... .... . 1 2 n ∞ ( −1) m x ∑ m = 0 m!( n + m)! 2 5 4 2m 7 6 (13. .. .. . .. . .. . . . ... . . .. .. . ... .. .. .. .50) (13. .. . ........ .. ... . . . . .... . ... . .. . ..... .. .. . ... . . . and the two particular solutions for these values of r are and their graphs are shown in Figure 13. ... ... .. . ... . .. .... . .. . .. . . .. .. . . . . . .. ... . ... .. .. .. . ..8) are r1=1±n. . . . .. .. . . .. ... .. .... . . ..... .. .. . . .. .. ...... .... ..... .... . .. The function is ﬁnite and converges for all values of x.. .. .. . .. . .. . .48) is Bessel functions Jn(x) for integer n The solutions of the indicial equation (13.. . . .. . . ......... ... . .. .. . . .... . . .. . . .. . . . . .. . ... ... . .... ... .. ... ......... . . . . . ... .. . .... .. . .. . . . . .... ... ..... . . . .. . . ... . .. . .. . .x . .... . . . . it converges very fast because of the pair of factorials in the denominator.... ....... .. . .... .... . ... . .... .. . . . .. . .. . . ... . . ... . . . .. ...... .. . . ...... 1 . .. Some special functions in which the constant a0 has been given its conventional value of a01=112(2n1n!).. ... the particular solution (13. .. ... . .....2 0 J . .. . . .. . . . .. .. . .. . . .. .. . ... . . .. . . . .. . ... . .... ... . ... . ... .. .. .49) (13. .... . ..... . . ... . . ... . . ... . . .. . . . . . . . .. ....... .. . .. . .. . . . . .... .... .. . . . . . . . This is the Bessel function of the ﬁrst kind of order n.... . .. . ... ... . . . .... .. .... . . .. .. . . ... . . ...... .... . . ..... ......... . .. . The functions for n1=10 and n1=11 are When n is a positive integer or zero. . . .... . .. . ... . . ... . .. . . ... . 386 Chapter 13 Second-order differential equations. . ... .. . .. . . .. . . . ... . .. ... ..... . .. . ..... . . . ... . .. .. . . . .. . .. . . .. . .. . . ... . . ... . .52) (13. . . .. ..... . . . . .. . . . . . .. . . . .. ... . . . . .... . .. .. .. ... . .... . .. Important values of the parameter n are the integer and half-integer values.. .... .. 01 . . . .. . . . . . . .. ........ .... . . .... .. .. . . .... . . .... . .. .. . ... ... . ... . . .. . ...... .. . . .... .. . x2 x4 y2 ( x ) = a0 x − n 1 + + + 2( 2 n − 2) 2 · 4( 2 n − 2) ( 2 n − 4) x2 x4 y1 ( x ) = a0 x n 1 − + − 2( 2 n + 2) 2 · 4( 2 n + 2) ( 2 n + 4) 1 x 1 x 1 x x J1 ( x ) = − + − + 2 1!2! 2 2!3! 2 3!4! 2 1 x 1 x 1 x J0 ( x) = 1 − + − + 2 2 2 (1!) ( 2!) 2 (3!)2 2 x J n ( x) = 2 3 Figure 13. ... .. .. .. . .... . . . . . . . . . . . .. .51) (13. . .. . . ... . .. . . ... . . . .. . . . .. . ... .. .. .. . ... . . .. . ..... . . .. . . .. .. ... . .... ... ... ..... . . .... .. 0J . . . . . . . . . .. . .... . . .. .. . . . . . .. ..2.. ... .. . . . . . . ..48) . ......... . . . .. .. .. .. ..... .. . .... . . .... . .. .. . . .

520. the zeros of the functions.173.56) and the formulas (13. 11. 7.931. 14. the values of x for which Jn(x)1=10.654. A second solution can be found.792. 13. 8. 10. 5. are not equally spaced so that the functions do not have a ﬁxed wavelength.49) is J−n(x) and it can be shown that this is related to Jn(x) by J−n(x)1=1(−1)nJn(x). πx 2 4 J n ( x) ∼ for x large (13. J −1 2 ( x ) = 2 πx cos x (13. 0 Exercise 25 Bessel functions Jl+122(x) of half-integer order Bessel functions of half-integral order can be expressed in terms of elementary functions. 3.53) In addition. The functions for n1=1±122 are (Example 13. Yn(x).832.55) to derive the Bessel functions J±322(x). However.55) and all others can be obtained by means of the recurrence relation (true for Bessel functions in general) J n +1 ( x ) − 2n J ( x ) + J n −1 ( x ) = 0 x n (13. = x1=10. = (13. The power-series method therefore gives only one independent solution for integral values of n.12 Use the recurrence relation (13. this is the Bessel function of the second kind.56) EXAMPLE 13. of less importance for physical applications.7 Bessel functions 387 Both the expansions and the graphs show that the Bessel functions have properties similar to the trigonometric functions. The behaviour for large values of x is that of a damped sine function.56) with n1=1122. We have. 2 nπ π sin x − + .4) 2 πx J1 2 ( x ) = sin x .324. from (13.405. the amplitude of the waves decreases as x increases. Some of the zeros are (to 3 decimal places) J0(x)1=10 J1(x)1=10 for for x1=12. J3 2 ( x) − 1 J ( x ) + J −1 2 = 0 x 12 .54) For positive values of n the particular solution (13.13.016.

59) .13 For l1=10. 1 2 J −3 2 ( x ) = − J −1 2 ( x ) − J1 2 ( x ) = − πx x 0 Exercise 26 cos x + sin x x Series expansions for Jl+122(x) and J−l−122(x) are given by equations (13. Some special functions so that 1 J ( x ) − J −1 2 ( x ) = x 12 2 sin x − cos x πx x J3 2 ( x) = Similarly.56). The functions of half-integral order are the Bessel functions that occur in the partial wave method in the theory of scattering processes.58) EXAMPLE 13.48) and (13. J−l−122(x)1→1∞ as x1→10. for l1≥10. jl ( x ) = π 2x J l +1/ 2 ( x ). ηl ( x ) = ( −1)l +1 π 2x J − l −1/ 2 ( x ) (13.49). J1 2 ( x ) = x2 x4 + − x1 2 1 − 3! 5! π 2 x2 x4 2 1 − + − 2! 4! πx = = 2 x 3 x5 x − + − 3! 5! πx 2 cos x πx = 2 sin x πx J −1 2 ( x ) = Equations (13.57) and (13.388 Chapter 13 Second-order differential equations. respectively.57) J − l −1 2 ( x ) = (13. They occur there in the forms. with n1=1−122 in (13.58) show that whereas Jl+122(x)1=10 at the origin (x1=10) and is therefore ﬁnite for all values of x. Thus x2 x4 x l +1 2 1 − + − 2( 2l + 3) 2 · 4( 2l + 3)( 2l + 5) π 2 x2 x4 x − l −1 2 1 + + + 2( 2l − 1) 2 · 4( 2l − 1)( 2l − 3) π 2 J l +1 2 ( x ) = (13.

15. Conﬁrm that the solution can be expressed as y1=1ae3x1+1be−3x. Solve the differential equations: 1 1 11. Conﬁrm the solution can be expressed as y1=1a2(11−1x) when | x|1<11. 0 Exercise 27 13. .3 For each of the following.8 Exercises Section 13. (i) Use the expansion method to ﬁnd a particular solution y1(x) of xy″1+1(11−12x)y′1+1(x1−11)y1=10. y″1−19y1=10. x 2y″1−1xy′1+1y1=10 2 1 13.8 Exercises 389 The functions jl (x) are called spherical Bessel functions of order l. x 2y″1+13xy′1+1y1=10 8.13.4 for r1=1122). the general solution is y1=1(a1+1b ln x)xr. 2 1+ x 5.60) These functions often occur in conjunction with the Legendre polynomials when a physical system is formulated in spherical polar coordinates. ﬁnd and solve the indicial equation 7. (11−1x 2)y″1−12xy′1+12y1=10 (the Legendre equation (13. x 1− x Show that the solution can be written as y = a1x + a0 1 + ln when | x|1<11. xy″1+1(11−12x)y′1+1(x1−11)y1=10 9. Show that the polynomial Pl (x) is a solution of the Legendre equation (13. Section 13. (ii) Show that for a double initial root r.2 Use the power-series method to solve the equations: 1. 3. y″1−1xy1=10 (Airy equation). 4.4 16. r11≠1r2. Find the general solution of xy″1+12y′1+14xy1=10. (ii) conﬁrm that y2(x)1=1y1(x)1ln1x is a second solution. (i) Solve the Bessel equation x y″1+1xy′1+1(x 21− 4 )y1=10 for indicial root r1=1−122 (see Example 13. x 2 y ′′ − 2 xy ′ + 2 y = 0 12. (i) Find the general solution of the Euler–Cauchy equation x 2y″1+1b0 xy′1+1c0 y1=10 for distinct indicial roots. (ii) Conﬁrm that the solution can be written as y( x ) = a0 x cos x + a1 x sin x = a J −1 2 ( x ) + b J1 2 ( x ) 14.13) for l1=11). Assume that there is no logarithmic term in the solution. They satisfy the differential equation x 2y″1+12xy′1+1[x 21−1l(l1+11)]y1=10 (13. x 2y″1+1xy′1+1(x 21−1n 2)y1=10 (Bessel equation) 6. y′1−13x 2y1=10 2.13) for (i) l1=12 and (ii) l1=15. (11−1x)y′1−1y1=10. Section 13. the functions ηl are the spherical Neumann functions. x 2y″1+16xy′1+1(61−1x 2)y1=10 10.

39) when n is a positive integer or zero and when the arbitrary constant is given its conventional value n!. Use the formula (13. (i) Use the series expansion (13. (i) Use the power series method to ﬁnd a solution of the Laguerre equation (13. .5 21. 27.24) to ﬁnd the associated Legendre functions (i) P1.41). Find L4(x) (i) from equation (13. (ii) from the general expression (13.50). Use the recurrence relation (13.39). Conﬁrm that the spherical Bessel function jl (x) satisﬁes equation (13. 20. Show that (i) P1 is orthogonal to P4 and P5. Section 13.7 25. Express the functions in terms of cos1θ1=1x and sin1θ1=1(11−1x 2)122. Find the Legendre polynomial P6(x) (i) by means of the recurrence relation (13. (ii) from L2(x) and L3(x) by means of the recurrence relation (13. 26.30) that H5(x) is a solution of the Hermite equation.21).31) to ﬁnd H5(x). 2. 18. Some special functions 17. (ii) Verify by substitution in (13.56).390 Chapter 13 Second-order differential equations. Show that P 1 is orthogonal to P 1 and P 1.6 23. 3. 2 22.60). (i) Find the Bessel function J2(x) (i) from the series expansion (13. (iii) Use the recurrence relation (13.33) to ﬁnd H6(x). (ii) Show that this solution reduces to the polynomial (13. 4 2 1 Section 13.19) for Pl (x). (ii) from J0(x) and J1(x) by means of the recurrence relation (13. 24.38). (ii) P m(x) for 1 4 m1=11. 4. Section 13. 19. Sketch the graph of the Hermite function e−x 22H3(x).56) to ﬁnd J522(x) and J−522(x). (ii) P2 is orthogonal to P0 and P3.

2. ∂f2∂y and higher partial derivatives. the unknown function f is a function of the coordinate x and of the time t.4) and in a circular box (Section 14.1 Concepts An equation that contains partial derivatives is a partial differential equation. The equations discussed in this chapter are equation 5 above for the particle in a rectangular box (Section 14. Equation 4 is inhomogeneous.14 Partial differential equations 14. the function depends on the three coordinates of a point in ordinary space.5). as well as f. z )ψ = Eψ ∇ 2ψ + V ( x . z) 5. the others are homogeneous equations. t )ψ = i 2m These are all second-order linear equations. y. z . y. − 2m 2 ∇ 2ψ + V ( x . In 3. ∂2 f ∂x 2 ∂2 f ∂x 2 = = + 1 ∂2 f v 2 ∂t 2 1 ∂f D ∂t ∂2 f ∂y 2 + ∂2 f ∂z 2 = ∇2 f = 0 1-dimensional wave equation 1-dimensional diffusion equation 3-dimensional Laplace equation 3-dimensional Poisson equation time-independent Schrödinger equation ∂ψ ∂t time-dependent Schrödinger equation ∂2 f ∂x 2 2 4. These examples demonstrate a number of important principles in the solution of partial differential equations. and whose solutions can be expressed in terms of known functions. . and equation 1 as applied to the vibrations of an elastic string. such as a guitar string (Section 14. Examples of such equations that are important in the physical sciences are 1. that the symmetry properties of the system being modelled can lead to the phenomenon of ‘degeneracy’. − 6.1t). f1=1f(x. ∇2f1=1g(x. 3. In equations 1 and 2. there are several important standard types of partial differential equation that frequently occur in mathematical models of physical systems. and that. x and y. is a partial differential equation. and in 6 it is also a function of the time.6). the solutions are expressed in the form of the ‘orthogonal expansions’ discussed in Chapter 15. and for the hydrogen atom (Section 14. For example. 4 and 5.7). y. if f is a function of the independent variables x and y then an equation that contains one or more of ∂f2∂x. As for ordinary differential equations. They show that different boundary and initial conditions can lead to very different types of particular solution. in some cases.

392

Chapter 14 Partial differential equations

14.2

General solutions

We have seen that the general solution of an ordinary differential equation contains a number of arbitrary constants, usually n for an nth-order equation, and that, in an application, the values of these constants are obtained by the imposition of appropriate initial or boundary conditions. The general solution of a partial differential equation, on the other hand, contains a number of arbitrary functions, often n such functions for an nth-order equation. For example, the 1-dimensional wave equation, equation 1 in the list given in Section 14.1, ∂2 f ∂ x2 has solution f(x,1t)1=1F(x1+1vt) where F is an arbitrary function of the variable u1=1x1+1vt. Thus, ∂f dF ∂u dF = = , ∂x du ∂x du ∂f dF ∂u dF , = =v du ∂t du ∂t ∂2 f ∂ x2 ∂2 f ∂t 2 = d 2F du 2 d2F du 2 (14.2) = 1 ∂2 f v 2 ∂t 2 (14.1)

= v2

and equation (14.1) is satisﬁed. The general solution of the equation is f(x,1t)1=1F(x1+1vt)1+1G(x1−1vt) (14.3)

where F and G are both arbitrary functions.1 The particular functions in any application are determined by appropriate initial and boundary conditions, and one important example is that discussed in Section 14.7. EXAMPLE 14.1 Verify that the function f(x,1t)1=13x 21−12xvt1+13v 2t 2 is a solution of the wave equation (14.1), and has the general form (14.3).

1 This form of the general solution was ﬁrst given by d’Alembert in 1747. The origins of the wave equation lie in the discussion De motu nervi tensi (On the motion of a tense string) by Brook Taylor in 1713. In 1727 Johann Bernoulli suggested to his son Daniel that he take up Taylor’s problem again: ‘Of a musical string, of given length and weight, stretched by a given weight, to ﬁnd its vibrations’. The equation was derived by d’Alembert in Recherches sur la courbe que forme une corde tendue mise en vibration, 1747, by considering the string to be composed of inﬁnitesimal masses and applying Newton’s force law to each element of mass. Euler published his own solution in Sur la vibration des cordes in 1750, and Daniel Bernoulli explored the idea of the superposition of normal modes in his Réﬂexions et éclaircissements in 1755. The debate over the kinds of functions acceptable as solutions of the wave equation was continued for about thirty years by these three, without any one of them being convinced by the others. The problem was resolved through the work of Fourier, Dirichlet, Riemann, and Weierstrass over the next hundred years, and involved a reconsideration of the meaning of function, continuity, and convergence.

14.3 Separation of variables

393

**The partial derivatives are ∂f = 6 x − 2v t, ∂x Therefore, 6= ∂2 f ∂ x2 = 1 ∂2 f v 2 ∂t 2 ∂2 f ∂x
**

2

= 6;

∂f = −2 xv + 6v 2t, ∂t

∂2 f ∂t

2

= 6v 2

as required. The function can be written as f(x,1t)1=1(x1+1vt)21+12(x1−1vt)21=1F(x1+1vt)1+1G(x1−1vt) 0 Exercises 1, 2 EXAMPLE 14.2 Verify that the function f(x,1t)1=1a exp [−b(x1−1vt)2] is a solution of the wave equation (14.1). We have ∂2 f = 2 ab −1 + 2b ( x − v t )2 exp − b ( x − v t )2 ∂x

2

∂2 f ∂t

2

= 2 abv 2 −1 + 2b ( x − v t )2 exp − b ( x − v t )2

2 2 Therefore ∂ f = 1 ∂ f . ∂x 2 v 2 ∂t 2

0 Exercise 3

14.3

Separation of variables

When a partial differential equation, involving two or more independent variables, can be reduced to a set of ordinary differential equations, one for each variable, the equation is called separable. The solutions of the partial differential equation are then products of the solutions of the ordinary equations. All the examples discussed in this chapter are of this type. We demonstrate the essential principles of the method of separation of variables by considering the simplest ﬁrst-order equation in two variables

394

Chapter 14 Partial differential equations

∂f ∂f + =0 ∂x ∂y We show that a solution of this equation can be written as f(x,1y)1=1X(x)1×1Y( y)

(14.4)

(14.5)

in which the solution, a function of the two variables x and y, is expressed as the product of a function of x only and a function of y only. We have ∂f ∂( XY ) dX = =Y ∂x ∂x dx because Y( y) does not depend on x. Similarly, ∂f dY =X ∂y dy because X(x) is a constant with respect to y. Substitution of f1=1XY and its derivatives into the differential equation (14.4) then gives Y ( y) dX ( x ) dY ( y ) + X ( x) =0 dx dy (14.6)

and, dividing throughout by f1=1X(x)1×1Y( y), 1 dX ( x ) 1 dY ( y ) X ( x ) dx + Y ( y ) dy = 0 (14.7)

The ﬁrst set of terms in square brackets on the left side of (14.7) depends only on the variable x, the second set only on y. If the total is constant (zero here) it follows that each of these sets of terms must be separately constant if x and y are independent variables. Thus a change in the value of x cannot lead to a change in the value of the terms that depend on y only, and vice-versa. Therefore, if the ﬁrst set of terms equals the constant C then the second set is equal to −C (for the total to be zero): 1 dX = C, X dx and C is called a separation constant. Then dX = CX dx dY = − CY dy (14.9) 1 dY = −C Y dy (14.8)

(14.10)

14.4 The particle in a rectangular box

395

and the partial differential equation (14.4) in two variables has been reduced to two ordinary differential equations, both of which we can solve. The equations, (14.9) in the variable x and (14.10) in the variable y, are separable ﬁrst-order equations of the kind discussed in Section 11.3, and they have general solutions X(x)1=1AeCx, Y( y)1=1Be−Cy (14.11)

A solution of the equation (14.4) in two variables is then the product f(x,1y)1=1X(x)1×1Y(y)1=1AeCx1×1Be−Cy1=1DeC(x−y) 0 Exercises 4–7 Particular solutions, including the possible values of the separation constant, can often be obtained by the application of initial and boundary conditions, as exempliﬁed by the important problems discussed in Sections 14.4 to 14.6. In other cases, as discussed in Section 14.7 for the vibrating string, it may be necessary to consider more general solutions that are linear combinations of products. (14.12)

14.4

The particle in a rectangular box

**The Schrödinger equation for a particle of mass m moving in the xy-plane is − where ∇2 = ∂2 ∂x
**

2 2

2m

∇ 2ψ ( x , y ) + V ( x , y )ψ ( x, y ) = Eψ ( x, y )

(14.13)

+

∂2 ∂ y2

(14.14)

is the two-dimensional Laplacian operator (see Section 9.6). For the present system, the potential energy function is (Figure 14.1) 0 for 0 < x < a and 0 < y < b V ( x, y ) = ∞ elsewhere

. . . . . . . . . . . . . . . . . . . . . . .. ... . ......................................................................... ..... ...... ....... ...... ...... ...... ...... ...... ...... ....... . .. .. . ....................................................................... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ............................. ............................ ............. .............. . . . . . . . . . . . . . . . . . . . . . . . . . .... ... . ... . ... .

(14.15)

Figure 14.1

x

∞

=

V

a

0

∞ ∞

=

= =

V

V V

y

b

0

∞

=

V

396

Chapter 14 Partial differential equations

As for the one-dimensional case, the constant value of V inside the box ensures that no force acts on the particle in this region; setting V1=10 means that E is the kinetic energy of the particle. The inﬁnite value of V at the ‘walls’ and outside the box ensures that the particle cannot leave the box; ψ1=10 at the walls and outside the box. For the particle within the box, we have the boundary value problem − with conditions

ψ (0,1y)1=1ψ (a,1y)1=10 ψ (x,10)1=1ψ (x,1b)1=10

∂2ψ ∂2ψ = Eψ + 2m ∂ x 2 ∂ y 2

2

(14.16)

(ψ1=10 when x1=10 or x1=1a) (ψ1=10 when y1=10 or y1=1b)

(14.17)

The partial differential equation is reduced to two ordinary differential equations by writing the wave function as the product

ψ (x,1y)1=1X(x)1×1Y(y)

(14.18)

Then, as in Section 14.3, substitution of this product and its derivatives into (14.13), and division by ψ1=1XY, gives 1 d 2 X 1 d 2Y 2 mE + =− 2 X dx 2 Y dy 2 (14.19)

The term involving x only must be constant and the term involving y only must also be constant. Therefore, 1 d2X = Cx , X dx 2 1 d 2Y = Cy Y dy 2 (14.20)

where Cx and Cy are constants, and Cx1+1Cy1=1−2mE2A2. The two-dimensional boundary value problem has been reduced to two onedimensional boundary value problems: d2X dx 2 d 2Y dy 2 = Cx X , = C yY , X (0) = X ( a ) = 0 (14.21a)

Y (0) = Y ( b) = 0

(14.21b)

Both of these are the same problem as that of the particle in a one-dimensional box discussed in Section 12.6. Equation (14.21a) describes the motion of the particle along the x-direction, and has normalized solutions given by (12.53), with appropriate changes of symbols: Xp ( x) = pπx 2 sin , a a p = 1, 2, 3, … (14.22a)

14.4 The particle in a rectangular box

397

and Cx1=1−p2π 22a 2. Similarly, (14.21b) describes the motion of the particle along the y-direction, and has normalized solutions Yq ( y ) = qπ y 2 sin , b b q = 1, 2, 3, … (14.22b)

and Cy1=1−q 2π 22b2. The total solutions (eigenfunctions) of the two-dimensional Schrödinger equation (14.13) are the products

ψp,q(x,1y)1=1Xp(x)1×1Yq(y)

=

pπx qπ y 2 2 sin sin × , a b a b

p, q = 1, 2, 3, …

(14.23)

and the corresponding total energies (eigenvalues) are E p, q = We note that the quantities Ep = h2 p 2 8ma 2 and Eq = h2 q 2 8mb2 (14.25) h2 p 2 q 2 + 8 m a 2 b2 (14.24)

**are to be identiﬁed with the kinetic energies of motion along the x and y directions, respectively.
**

A square box. Degeneracy

When the sides of the box are not equal and neither is an integer multiple of the other, the eigenvalues (14.24) are all distinct; the states of the system are then said to be nondegenerate. For a square box however, with a1=1b, Ep, q = h2 8ma 2 ( p2 + q2 ) (14.26)

and the eigenvalues for p1≠1q occur in pairs with Ep,q1=1Eq, p ; for example, E1,21=1E2,11= 5h 228ma 2. States of the system with equal energies are called degenerate states. The occurence of degeneracy for the square box is a consequence of the symmetry of the system. The eigenfunctions (14.23) are

ψ p ,q ( x , y ) =

pπx qπ y 2 sin sin a a a

(14.27)

**and an interchange of the x and y coordinate axes gives
**

ψ p ,q ( y , x ) =

pπ y qπx 2 sin sin = ψ q, p ( x, y ) a a a

(14.28)

398

where r is the distance from the origin at the centre of the box. The equation is not separable in cartesian coordinates because of the functional form of V at the boundary of the box, but it becomes so when the equation is expressed in the plane polar coordinates r and θ. In these coordinates, the two-dimensional Laplacian operator is (see equation (9.38) and Example 9.18)

The dashed lines are nodal lines, where the function is zero. We see that, for example, ψ1,2 and ψ2,1 are identical except for orientation.

that is, the degenerate eigenfunctions are interchanged (when p1≠1q).* The symmetries of some of the wave functions are illustrated in Figure 14.2.

Chapter 14 Partial differential equations

* An ‘accidental’ degeneracy may also occur that is not an obvious consequence of symmetry. For example, the states (7, 1) and (1, 7) are degenerate with the state (5, 5).

1=q 3=p

3=q 1=p

+

....................... ...................... . . ....................... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..................... . . . ....................... ....................... . . .

....................... ...... ............... ....................... . . . . . . . . . . . . . . . . . . . . . . . . .. .. .. .. .. .. . . . . . .. . . .. . . ... .. .. .. .. .. .. . . . . . . . . . . . . . . . . . . . . . . . . ... .. .. .. .. .. .. .. . . .. . . .. . . .. .. .. .. .. .. . . . . . . . . . . . . . . . . . . . . ..................... . ....................... . ....................... .

**For the particle within the box, with V1=10, equation (14.29) is then
**

....................... ...................... . . . ....................... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... .. .. .. .. .. .. . . . .. .. .. .. .. .. . . . . . . .. . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..................... . . ....................... ....................... . .

as in Section 14.4, but with potential-energy function

The motion of a particle in a circular box of radius a is described by the Schrödinger equation

2=q 2=p

1=q 2=p

2=q 1=p

1=q 1=p

....................... ............ . ............ .......... . . ............ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..................... . . ....................... ....................... . .

....................... ....... .............. ....................... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... .. .. .. .. .. .. . . . .. .. .. .. .. .. . . . . . .. . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..................... . ....................... . .......................

....................... ...................... ....................... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..................... . ....................... ....................... .

14.5

0 Exercises 8–10

+

The particle in a circular box

0 V ( x, y ) = ∞

−

+

–

2m

2

−

∇ 2ψ ( x , y ) + V ( x , y )ψ ( x, y ) = Eψ ( x, y )

∂2ψ 1 ∂ψ 1 ∂2ψ = Eψ + 2 + 2 m ∂ r 2 r ∂r r ∂θ 2

2

∇2 =

for r = x 2 + y 2 < a, (inside the box) elsewhere

– +

∂r 2

∂2

Figure 14.2

+

1 ∂ 1 ∂2 + 2 2 r ∂r r ∂θ

– + + –

+

–

+

–+

(14.31)

(14.30)

(14.29)

(14.32)

14.5 The particle in a circular box

399

**or, multiplying throughout by −2mr 22A2, setting
**

α2 =

2 mE

2

(14.33)

and rearranging, r2 ∂2ψ ∂r 2 +r ∂ψ ∂2ψ + α 2 r 2ψ + 2 = 0 ∂r ∂θ (14.34)

This equation can now be reduced to two ordinary equations by writing the wave function as the product

ψ (r,1θ)1=1R(r)1×1Θ (θ)

(14.35)

**Substitution in (14.34) and division by ψ1=1RΘ then gives
**

2 r R

d2R

2 r dR 2 2 1 d Θ =0 + +α r + 2 Θ dθ dr 2 R dr

(14.36)

Each set of terms in square brackets must be constant so that, with separation constant C, r2 d2R dr 2 +r dR + α 2 r 2 R = CR dr (14.37)

for the radial motion of the particle in the box, and d 2Θ dθ 2 = −CΘ (14.38)

**for the angular motion of the particle. We consider the angular equation ﬁrst.
**

The angular equation

The function Θ (θ) is deﬁned in the interval 01≤1θ1≤12π and must satisfy the periodic boundary condition

Θ (2π )1=1Θ (0)

(14.39)

for continuity round the circle. We therefore have the same boundary value problem as that discussed in Section 12.7 for the particle in a ring. The normalized solutions are (equation (12.69))

Θ n (θ ) =

1 inθ e , 2π

n1=10, ±1, ±2, =

(14.40)

and substitution in (14.38) gives the separation constant C1=1n2.

400

Chapter 14 Partial differential equations

The radial equation

**With C1=1n 2, the radial equation (14.37) is r2 d2R dr
**

2

+r

dR + (α 2 r 2 − n2 ) R = 0 dr

(14.41)

The equation is transformed into the Bessel equation (Section 13.7) by means of the change of variable x1=1α r. Then dR dR dx dR = =α , dr dx dr dx d2R dr 2 = α2 d2R dx 2

**and (14.41) becomes the Bessel equation (13.47), x2 d2R dx
**

2

+x

dR + ( x 2 − n2 ) R = 0 dx

When n is a positive integer or zero, the solution of this equation is the Bessel function Jn(x) given by (13.50), so that the solutions of the radial equation are Rn(r)1=1Jn(α r), n1=10, 1, 2, 3, = (14.42)

These solutions are subject to the condition that the wave function vanish at the boundary of the box, when r1=1a. Therefore Rn(a)1=1Jn(α a)1=10 (14.43)

and the possible values of α are determined by the zeros of the Bessel function, examples of which are given in (13.53) of Section 13.7. If the zeros of Jn(x) are labelled xn,1, xn,2, xn,3, =, the allowed values of α are

α n, k =

xn, k a

,

k1=11, 2, 3, =

(14.44)

and the solutions of the radial equation that satisfy the boundary condition are Rn,k(r)1=1Jn(α n,k r) (14.45)

By equation (14.33), the energy of the system is given by E1=1α 2A222m, so that the energy is quantized, with values En , k =

2 α n, k 2

2m

=

2 xn, k

2

2 ma 2

,

n1=10, ±1, ±2, =

(14.46)

**and the corresponding total wave functions are
**

ψn,k(r,1θ)1=1R|n|,k(r)Θn(θ )

(14.47)

We ﬁrst simplify the equation by expressing all physical quantities in atomic units (see Section 1.8); the Schrödinger equation ‘in atomic units’ is

where ∇2 is the three-dimensional Laplacian and, in terms of cartesian coordinates, the potential energy function is

A hydrogen-like atom is a system of two charges, a nucleus with charge +Ze (Z1=11 for the hydrogen atom itself) and an electron with charge −e, interacting through a Coulomb force. The potential energy of the system is (see Examples 1.21 and 5.17)

0=n 2=k

–

... .. .... ..... . .. ....... ... . . .. ... ... .. ... .. . .. . .. . . . .. . .. . .. . .... . .. .. . .... . . . . . . . . . . . . . . .. . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . .. . . . .. . .. .. . .. . .. . ... . . .. . .. .. . . . .. .. . .. .. .. ... .. ..... ...... ............. ........ .. .. .. . .

2± = n 1=k

... .. .... ..... . .. ....... ... . . .. ... ... . ... .. . .. .. .. .. . .. . . ... . ... .. . .. . . . . . . . . . . . . .. . .. .. . .. . . . . . .. .. . . . .. . . . . . . . . . . . . .. .. . . . . . . . . ... . . . . . . . . . .. .. .. . . . . . . . .. .. .. . .. . .. . .. .. . . .... .. . .. ... ... .. .. .. . . .. .. ... .. ..... ...... ............. ........ .. .. .. . .

1± = n 1=k

... .. .... ..... . .. ....... ... . . .. ... ... .. ... .. . .. . .. . . .. . .. . .. .. . . . . . . . . . . . . . . . . . . . . . ... .. .. .. .. .. .. . . . . .. . . . .. .. .. .. .. .. . . .. . . . . . . . . . . . . . . . . . . . . .. . .. . .. . . . . .. .. .. .. .. .. ... .. ..... ...... ............. ........ .. .. .. . .

0=n 1=k

... .. .... ..... . .. ....... ... . . .. ... ... .. ... .. . .. . .. . . .. . .. . .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . .. . .. . . . . .. .. .. .. .. .. ... .. ..... ...... ............. ........ .. .. .. . .

... .. .... .. . .. .. . . . ... . . .. ... . ..... .. ... . .... .. . .. . .. . . . . .. . .. . . .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . ... .. .. .. .. .. .. . . . . .. . . . .. .. .. .. .. .. . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . .. . .. . . . . . . . .. .. .. .. .. .. ... .. ..... . ...... ............. ........ .. .. .. .. . .

... .. .... .. . .. .. . . . ... . . .. ... . ..... .. ... . .... .. . .. . .. . . . . .. . .. . . .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . .. . .. . . . . . . . .. .. .. .. .. .. ... .. ..... . ...... ............. ........ .. .. .. .. . .

(the radial function depends only on the modulus of n). The symmetries of these wave functions, using the real forms (12.71) of the angular functions, are illustrated in Figure 14.3.

+

+

– +

– + + –

+

+

14.6 The hydrogen atom

–

+

+

14.6

0 Exercise 11

We note that these diagrams also represent the normal modes of motion, the standing waves, of a circular membrane.

where r is the distance between the charges. Let the nucleus be ﬁxed at the origin of a coordinate system, with the electron at position (x,1y,1z). The Schrödinger equation for the motion of the electron about the stationary nucleus is then

The hydrogen atom

Z 1 ∂2ψ ∂2ψ ∂2ψ − 2 + 2 + 2− 2 ψ = Eψ ( x + y 2 + z 2 )1/ 2 2 ∂x ∂y ∂z

–

V ( x, y, z ) =

−

2me

2

V=

Figure 14.3

∇ 2ψ + V ψ = Eψ

4 πε0 ( x 2 + y 2 + z 2 )1/ 2

− Z e2 4 πε0 r

− Ze2

–

(14.51) (14.50) (14.48) (14.49)

401

402

Chapter 14 Partial differential equations

Separation of variables

The ﬁrst step in the solution of the partial differential equation in three variables is to separate the variables; that is, to reduce it to three ordinary equations. This is not possible in cartesian coordinates because the potential function V cannot then be expressed as a sum of terms, one in each variable only. There exist a number of other coordinate systems, however, in terms of which the separation is possible, and one of these is the system of spherical polar coordinates discussed in Chapter 10. The Laplacian operator in these coordinates is (Section 10.5) ∇2 = 1 ∂ 2 ∂ 1 ∂ ∂ 1 ∂2 r + 2 sin θ + 2 2 ∂θ r sin θ ∂φ 2 r 2 ∂r ∂r r sin θ ∂θ

and the Schrödinger equation is then (after multiplication by −2 and rearrangement) ∂ ∂ψ 1 ∂2ψ 2 Z 1 ∂ 2 ∂ψ 1 ψ + 2 Eψ = 0 + r + 2 sin θ + 2 2 ∂θ r sin θ ∂φ 2 r r 2 ∂r ∂r r sin θ ∂θ (14.52) We ﬁrst separate the radial terms from the angular terms by writing the wave function as the product

ψ (r, θ, φ)1=1R(r)1×1Y(θ, φ)

(14.53)

**Substitution into (14.52), division throughout by ψ1=1RY, and multiplication by r 2 then gives
**

1 R 1 d 2 dR 2 r + 2 Zr + 2 Er + dr dr Y

1 ∂ 1 ∂2Y ∂Y =0 sin θ + 2 sin θ ∂θ ∂θ sin θ ∂φ 2 i (14.54)

We call the separation constant l(l1+11) for reasons that will become clear. Then 1 d 2 dR 2 r + 2 Zr + 2 Er = l ( l + 1) R dr dr or 1 d 2 dR l ( l + 1) 2 Z + + 2 E R = 0 r + − 2 dr 2 r dr r r is the radial equation of the hydrogen atom, and 1 ∂ ∂Y 1 ∂2Y + l ( l + 1) Y = 0 sin θ + 2 ∂θ sin θ ∂φ 2 sin θ ∂θ is the angular equation. (14.56) (14.55)

14.6 The hydrogen atom

403

To separate the angular variables we now write Y(θ, φ)1=1Θ (θ )1×1Φ (φ) (14.57)

Substitution of this product and its derivatives in the angular equation, division by Y1=1ΘΦ, and multiplication by sin21θ gives sin θ d 1 d 2Φ dΘ 2 =0 sin θ + l ( l + 1) sin θ + Θ dθ Φ d φ2 dθ so that, with separation constant −m 2 d 2Φ (φ ) dφ2 = − m2Φ (φ ) (14.59) (14.58)

1 d dΘ m2 sin θ + l ( l + 1) − 2 Θ = 0 dθ sin θ d θ sin θ

(14.60)

The separation of the variables is complete. It is now only necessary to solve the three boundary value problems represented by equation (14.55), (14.59), and (14.60), with appropriate boundary conditions.

The Φ equation

d 2Φ (φ ) dφ2

= − m2Φ (φ )

(14.59)

The function Φ is deﬁned in the interval 01≤1φ1≤12π and must satisfy the condition

Φ (2π )1=1Φ (0)

(14.61)

for continuity round the circle. We therefore have the same boundary value problem as that discussed in Section 12.7 for the particle in a ring and in 14.5 for angular motion of the particle in a circular box. The normalized solutions are

Φm (φ ) =

1 imφ e , 2π

m1=10, ±1, ±2, =

(14.62)

or, in real form (equations (12.71)), 1 1 (Φm + Φ− m ) = cos mφ , 2 2 π 1 1 (Φm − Φ− m ) = sin mφ i 2 2 π (14.63)

for m1=10, 1, 2, =. The functions form an orthonormal set, with property

Z Φ *(φ ) Φm′ (φ ) d φ = δ m, m′ m

0

2π

(14.64)

404

Chapter 14 Partial differential equations

The Θ equation

1 d dΘ m2 sin θ + l ( l + 1) − 2 Θ = 0 dθ sin θ d θ sin θ

(14.60)

The equation is transformed into the associated Legendre equation (13.22) by means of the substitution x1=1cos1θ. We have dΘ dΘ dx dΘ = = − sin θ dθ dx d θ dx d 2Θ dθ 2 Therefore, 1 d dΘ d 2Θ cos θ dΘ d 2Θ dΘ + = (1 − x 2 ) 2 − 2 x sin θ = 2 dθ d θ sin θ d θ sin θ d θ dx dx and equation (14.60) becomes d 2Θ dx 2 dΘ m2 Θ=0 + l ( l + 1) − dx (1 − x 2 ) = − cos θ d 2Θ dΘ d 2Θ dΘ + sin 2 θ 2 = (1 − x 2 ) 2 − x dx dx dx dx

(1 − x 2 )

− 2x

(14.65)

and this is identical to the associated Legendre equation. The ﬁnite solutions in the interval −11≤1x1≤11 (01≤1θ1≤1π ) are the (normalized) associated Legendre functions (equation (13.28))

Θ l , m (cos θ ) =

( 2l + 1) ( l − | m |)! |m| P (cos θ ), 2 ( l + | m | )! l

l = 0, 1, 2, 3, … m = 0, ± 1, ± 2, …, ± l (14.66)

These functions form an orthormal set, with property (equation (13.29))

**Z Θl , m (cos θ )Θl ′ , m (cos θ ) sin θ dθ = δ l , l ′
**

0

2π

(14.67)

Spherical harmonics

The products of the angular functions Θl,m and Φm, 2l + 1 ( l − | m |)! |m| Yl , m (θ , φ ) = Θ l , m (θ )Φm (φ ) = Pl (cos θ ) eimφ 4 π ( l + | m |)! (14.68)

14.6 The hydrogen atom

405

are called spherical harmonics. They occur whenever a physical problem in three dimensions is formulated in spherical polar coordinates.* Some of these functions are listed in Table 14.1. The functions are complex when m1≠10 and (see equations (14.63)) it is sometimes more convenient to use the corresponding real functions 1 (Y + Y ), 2 l , m l ,− m 1 (Y − Yl ,− m ), i 2 l, m m>0 (14.69)

Spherical harmonics multiplied by the factor r l are called solid harmonics, and the real forms of these are also listed in the table, with their conventional names in atomic structure theory. By virtue of the orthonormality relations (14.64) and (14.67), the spherical harmonics form an orthonormal set over a complete solid angle (θ1=101→1π , φ1=101→12π ):

Z Z Yl*m (θ , φ ) Yl ′, m′ (θ , φ ) sin θ dθ d φ = δ l , l ′ δ m, m′ ,

0 0

2π

π

(14.70)

**Table 14.1 Spherical harmonics
**

1 Y0, 0 = 4π 3 Y1,0 = 4π

12

**Solid harmonics (real)
**

1 s= 4π

12

12

cos θ

12

3 pz = z 4π 3 px = x , 4π 5 d z2 = 8π

12 12

12

3 Y1,± 1 = 8π

sin θ e± iφ

12

3 py = 4π

12

y

5 Y2, 0 = 16 π 15 Y2,± 1 = 8π

**(3 cos 2 θ − 1) sin θ cos θ e± iφ
**

12

(3z 2 − r 2 )

12

12

15 d xz = xz , 4π 15 d x2 − y2 = 16 π

12

15 d yz = 4π ( x 2 − y 2 ),

12

yz 15 d xy = xy 4π

12

15 Y2,± 2 = 32 π

sin 2 θ e±2 iφ

Angular momentum

**The angular equation (14.56) can be written as the eigenvalue equation (with appropriate units) −
**

2

1 ∂ ∂ 1 ∂2 Y = l ( l + 1) 2Y sin θ + 2 l, m sin θ ∂θ ∂θ sin θ ∂φ 2 l , m

(14.71)

* The spherical harmonics are often deﬁned with an additional ‘phase factor’ (−1)(m1+1| m |)22 that multiplies the function by (−1) when m is odd and positive.

406

Chapter 14 Partial differential equations

or

L2Yl,m1=1l(l1+11)A2Yl,m

(14.72)

in which L2 is the quantum-mechanical operator for the square of angular momentum. The spherical harmonics are therefore the eigenfunctions of L2. They describe the possible states of angular momentum of the system, and the eigenvalues l(l1+11)A2 are the allowed values of the square of angular momentum. In addition, it follows from (14.62) that −i and, therefore, that −i or ∂Yl , m ∂φ = m Yl , m (14.73) d Φm = mΦm dφ

LzYl,m1=1mAYl,m

(14.74)

of this component. The number l is called the angular momentum quantum number and m the component of angular momentum (or magnetic) quantum number.

The radial equation

Lz is the quantum-mechanical operator representing the component of angular momentum in the z-direction, and the eigenvalues mA of Lz are the allowed values

1 d 2 dR l ( l + 1) 2 Z + + 2 E R = 0 r + − 2 dr 2 r dr r r

(14.55)

The equation has two sets of solutions for the hydrogen atom. In the bound states of the atom the electron is effectively conﬁned to the vicinity of the nucleus by application of the boundary condition R(r)1→10 as r1→1∞. The energies of these states are negative (the zero of energy is for the two charges at rest and at inﬁnite separation) and energy must be supplied to the system in order to ionize the atom. States with positive energy are unbound states (continuum states) in which the electron moves freely in the presence of the nucleus but is not bound to it. We consider here only the bound states, with E1<10. We set

α 21=1−2E,

λ=

Z

α

(14.75)

**and introduce the new variable
**

ρ1=12α r

(14.76)

14.6 The hydrogen atom

407

Then dR dR = 2α , dr dρ and the radial equation becomes d2R d ρ2 2 dR l ( l + 1) λ 1 + − + − R=0 ρ dρ ρ 4 ρ2 d2R dr 2 = 4α 2 d2R d ρ2

+

(14.77)

This is identical to equation (13.45) for the associated Laguerre functions when λ1=1n, a positive integer. The solutions of the radial equation are therefore given by equation (13.44) e− ρ 2 ρ l L2 l++l1 ( ρ ) n These functions are ﬁnite and continuous for all positive values of ρ, and therefore of r, and they satisfy the boundary condition for bound states. The functions may be normalized, using equation (13.46), and the resulting normalized radial wave functions are* 3 ( n − l − 1)! 2Z Rn,l ( r ) = − n 2 n{( n + l )!}3 where, because α 1=1Z2n when λ1=1n,

ρ=

12

e− ρ 2 ρ l L2 l++l1 ( ρ ) n

(14.78)

2Z r n

(14.79)

The allowed values of the quantum numbers are n1=11, 2, 3, = l1=10, 1, 2, =, (n1−11) (14.80)

The radial functions form an orthonormal set with respect to the weight function r 2 in the interval 01≤1r1≤1∞:

Z Rn,l ( r ) Rn′,l ( r ) r 2 dr = δ n, n′

0

∞

(14.81)

Some of the radial functions are listed in Table 14.2.

* By convention the radial functions (14.78) are deﬁned with a – sign as part of the normalization constant.

408

Chapter 14 Partial differential equations

The total solutions of the Schrödinger equation for the bound states of the hydrogenlike atom are

r

1 2 0

r

4 1

d 3

p 3

p 2

s 3

s 2

. . . . . . .. ... ... . . .... . . . .. . .. .. ... .. . . . . ...... . . . . .. . . . ...... ........ . .. . .................. .... . . . .... ...... . . . . . ........... . . . ... . . ........... . .. ... ........ .............. .............. ........................................................................ ...................................................................... ........................... ....................... ...... .. ....... ..... ... .. .. . . . . . . .... . . .. . .. ....... . .. . .................. . . . . .. . . . ... . . . .......................... .. . .. .. .... ... ... . ... ............. . ........ .. .... ....... ................ .......... . . . .. . .............. ..... ... . ... . .. ..... .. . . . ...... . ....... ... ... . .. . ..... .. .... ... . . .. .. . .. .. ............... . . . .............. . .. ............ . .. .. . . . . .. . .. . .. ... . .. ... . ....... . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . .. . .. .. .. .. .. .. .. . .. .. .. .. . .. .. .. .. . .. .. .. .. .. .. .. .. . .. .. .. . .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. .. . . . . . . . . .

3 √2 3

. . . . . ... . .... ..... . .. . .......... .... ... . . .. . . . . .... .... .... . . . . . . . . ... . ...... . . . ...... . . .. ... . . ....... .................................................... ......................... ........... ............ .................. ........... ........ ... ...... ... . ...... . .. . . ....... .. . . . ......... .... ... . . ..... .... ......... . . ..... . . ... . . .. .. . .. . . . . .. . . . .. ... . .. .. . . .. . .. . .. .. .. . ... .. . .. . .. .. .. .. ... ... . ... .. . .. . . . . . . ...... . ..... . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. . .. .. . . .. .. .. . .. .. .. . .. .. .. .. .. .. .. . .. .. .. .. .. . .. .. .. . .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. . . . . . . . . .

The total wave function

The allowed values of the energy are obtained from (14.75); when λ1=1n, α 1=1Z2n and E1= −α 222. Therefore

0

r

7

2 √ 1

. . . . . . . . . . . . . . . . ............................ .............. ....... . . . . ................... .......................... . . .. ...... . .. . . . . . .. .. . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. . .. .. . .. . .. .. . .. .. .. . . .. .. .. .. .. .. . .. .. .. .. .. .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. . . . . . . . . .

The energy

The graphs of these are shown in Figure 14.4. The number of radial nodes (spherical nodal surfaces of the total wave function) is n1−1l1−11, excluding the zero at r1=10 when l1>10.

**The number n is called the principal quantum number.
**

3 3 3 2 2 1

n

Table 14.2 Radial functions Rn,l ( ρ), ρ = 2Zr2n

l

2

1

0

1

0

0

En = −

0

Name

3d

3p

3s

2p

2s

1s

2 n2 Z2

Figure 14.4

,

n1=11, 2, 3, =

2Z 322e−ρ22

Rn,l ( ρ)

1 Z 3 2 (6 − 6ρ + ρ 2 ) e− ρ 9 3

1 Z 3 2 ( 4 − ρ ) ρ e− ρ 9 6

1 Z 3 2 ρ 2 e− ρ 9 30

1 3 2 −ρ Z ρe 2 6

1 Z 3 2 ( 2 − ρ ) e− ρ 2 2

2

2

2

2

2

(14.82)

s 1

2

14.6 The hydrogen atom

409

ψn,l,m(r, θ, φ)1=1Rn,l (r)Θl,m(θ)Φm(φ)

(14.83)

and the numbers n, l and m are called the principal quantum number, (n), the angular momentum quantum number, (l), and the component of angular momentum (or magnetic) quantum number (m). From the orthormality relations (14.64) for Φm, (14.67) for Θl,m and (14.81) for Rn,l it follows that the total wave functions form an orthonormal set in three-dimensional space:

Zψ *,l ,m ψ n′,l ′,m′ dv = Z Z Z n

0 0

2π

π

∞

0

ψ *,l , m ( r , θ , φ ) ψ n′,l ′, m′ ( r , θ , φ ) r 2 sin θ dr d θ d φ n

=1δn,n′δl,l′δm,m′

(14.84)

Some of the total wave functions of the hydrogen atom, the atomic orbitals, are illustrated in Figure 14.5 by contour diagrams in an appropriate plane containing the nucleus.* Solid contours represent positive values of the orbitals, dashed contours represent negative values, dotted contours are nodes. We note that wave functions ψn,l,m with energy En have a total of (n1−11) nodal surfaces. It is these diagrams that have led to the conventional pictorial representation of atomic orbitals. 0 Exercises 12–15

Figure 14.5

* The diagrams in Figure 14.5 are to scale within boxes of side 20 (a0). The contour values are ±0.0021×12i, i1=10, 1, 2, 3, =

yxd3

xp2

xp3

s2

s3

s1

410

Chapter 14 Partial differential equations

14.7

**The vibrating string
**

. . . . . . . . . . . . . . . . . . . .... . . . . . . . . ... .. . . . . . . .. . . . . . .. .. . . . . . . . . .... ...... . . . . . . .... ... . . . .. . ... ... ........................................................ ....................................................... .. .. . . ........ . . ...... ........... .......... ... .......... ......... . . ... ...... ... .. . . . ....... ... . .. . . . .. . . ........ . . . . .. . . .. . ....... . . . ...... . . . ....... . . . .. . . . . ........ . . . ........ . . ........ . . . . . .. .. . . ........ . . . .. . . . . . . ........ . . . ... . ........ . . . ... . . .. . . .. . .. . .. .

Figure 14.6

y1=1y(x,1t)

(14.85)

and when the vibrations are small the motion of the string is described by the wave equation ∂2 y ∂x 2 = 1 ∂2 y v 2 ∂t 2 (14.86)

where v 21=1T2ρ. The boundary and initial conditions on the solutions of the equation are y(0, t)1=1y(l, t)1=10 for no displacement at the ends of the string, and ∂y = g( x) ∂t t = 0 (14.88) (14.87)

y(x,10)1=1f(x),

**for the initial displacement and velocity (functions).
**

Separation of variables

We write the displacement function (a wave function) as the product y(x,1t)1=1F(x)1×1G(t) Substitution in the wave equation and division by y1=1FG then gives 1 d2F 1 d 2G = F dx 2 Gv 2 dt 2 (14.90) (14.89)

Both sides of the equation must be constant so that, with separation constant −λ2, the problem in two variables reduces to the boundary value problem d2F dx

2

+ λ 2 F = 0,

F(0)1=1F(l)1=10

(14.91)

x

•

. . ...... . . ..

. . . .... ... . . . . . . . . . . .. .. . . . . . . .... ... .......... . . . . . .. . .. . .. .. . . . . . .

......... .

•

y

We consider an elastic string, such as a guitar string, of length l and uniform linear density ρ that is stretched and ﬁxed at both ends under tension T. The string is distorted transversely (pulled sideways), released and allowed to vibrate (Figure 14.6). The transverse displacement of the string is then a function of position x and time t,

14.7 The vibrating string

411

and the initial value problem d 2G dt 2 + λ 2v 2G = 0 (14.92)

with initial conditions (14.88), where f(x) and g(x) are given functions. The boundary value problem (14.91) is identical to that discussed in Section 12.6 for the particle in a box. The allowed values of the separation constant are given by

λn =

nπ , l

n1=11, 2, 3, =

(14.93)

**and the corresponding (unnormalized) particular solutions are Fn ( x ) = sin For each value of λn, equation (14.92) is d 2G dt where
**

ω n = λ nv =

2 2 + ω nG = 0

nπ x l

(14.94)

(14.95)

n πv l

(14.96)

and has solution Gn(t)1=1An1cos1ωnt1+1Bn1sin1ωnt (14.97)

where An and Bn are constants determined by the initial conditions. A set of solutions of the wave equation for the vibrating string is therefore yn ( x , t ) = sin nπ x A cos ω nt + Bn sin ω n t , l n n1=11, 2, 3, = (14.98)

These solutions are called the eigenfunctions of the system. The quantities ωn1=1n π v2l are called the eigenvalues. The set of values {ω1, ω2, ω3, =} is called the eigenvalue spectrum.

Normal modes of motion

Each eigenfunction yn(x,1t) is a periodic function of time with period 2π 2ωn; the motion is transverse harmonic motion with frequency νn1=1ωn 22π . This motion is called the nth normal mode of the vibrating string. The ﬁrst mode, with n1=11, is called the fundamental, the second, with n1=12, is the ﬁrst overtone, and so on. The motions

412

Chapter 14 Partial differential equations

Each of the normal modes contains two constants, An and Bn, that are to be determined by the initial conditions (14.88); that is, by the way the motion is initiated. It is possible to choose the initial conditions so that the actual motion is one of the normal modes; for example, the nth mode is produced if

In general however, a single mode will not satisfy the initial conditions; the motion is not a pure normal mode but is a mixture or superposition of normal modes. To obtain the solution for the general case, we invoke the principle of superposition discussed in Section 12.2; if y1,1y2,1y3,1= are solutions of a homogeneous linear equation then any linear combination of them is also a solution. Thus, for each normal mode we have

enotrevo dnoces 3=n

. .. . .. .. . . . . .. .. ... . . . . .. . . . . ... . . .. . . . . . .. . .. . . .. . . . . . . . . . . . . . . . . ... . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . .. . . . . . . . . . .. . . . .. .. . . .. .. . . .. . .. . . . . ..................................... . .................... .. ........... . . . . . . ................................. ... . . . . . . . . . .. . . .. .. . . . . . . . . .. .. . . .. . . . .. . . . . . . . . . . .. . . . . .. . . .. . . . . . . . . . .. .. .. . .. .. . . .. .. .. . .. . . . . . . .. . . . . . . . .. . . . . . . . . . . . . . .. . .. . .. . .. . . . .. . . .. . . . ... ... .. . .. ... . . . . ... ... . . .... ... . .. ... . . .. .. .. . . .

enotrevo tsrﬁ 2=n

.. . . . . . .. .. . . ... .. . . . . . .. .. . .. .. . .. .. . . . . .. . . .. . . . .. . . . . .. .. . . . . .. .. . . . . ... .. . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . .. ............... ... ................. ..................................... .................................... . .. . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . .. . . .. . . . . . . . . . . .. .. . . .. .. .. . . . .. . . . . . . .. . .. . . . . .. . . . . .. . . .. . .. . .. . ... . ... .. . .. . . .. . .. . .... .. . ... .. . . . .

The complete solution

The nth mode has n1−11 nodes between the end points; that is, points of zero displacement that do not move; the solutions (14.98) represent standing waves.

latnemadnuf 1=n

. . . . . .. .. .. . .. .. .. .. . .. . . .. .. . . . . . .. .. .. . . . . . . .. .. .. .. . . . . .. . . . . . ..................................... ..................................... .. . . ............................. ...... . . . . .. .. .. .. .. .. . . .. .. .. .. .. .. .. .. .. .. .. .. .. .. . . . .. .. .. . .. .. . .. .. .. .. . .. .. .. . .. .. ... ... . . .. . .. . ..... . .. . .. . ..... .......... ........ ..... . ....

in space of the ﬁrst few of these normal modes are illustrated in Figure 14.7 (see also Figure 12.6 for the particle in a box).

so that, if

then

∂y nπ x = g ( x ) = Bnω n sin l ∂t t = 0

y ( x , 0) = f ( x ) = An sin

y1=1c1 y11+1c2 y21+1c3 y31+1-

∂ 2 yn

∂x 2

∂2 y

∂x 2

Figure 14.7

=

=

v 2 ∂t 2

v 2 ∂t 2

2 1 ∂ yn

1 ∂2 y nπ x l

(14.100)

(14.102) (14.101) (14.99)

14.8 Exercises

413

The general solution of the wave equation that satisﬁes the boundary conditions is therefore a superposition of normal modes y ( x , t ) = ∑ sin

n =1 ∞

nπ x A cos ω nt + Bn sin ω nt l n

(14.103)

(the constants c1,1c2,1c3, = in (14.101) have been absorbed in the A’s and B’s in (14.103)). This solution satisﬁes the initial conditions when the A’s and B’s are determined from the equations y ( x , 0) = f ( x ) = ∑ An sin

n =1 ∞

nπ x l (14.104)

∞ ∂y nπ x = g ( x ) = ∑ Bnω n sin l ∂t t = 0 n =1

The two series in (14.104) are examples of the Fourier series discussed in Chapter 15. We return to this problem in Section 15.5 to determine the A’s and B’s for typical initial functions f(x) and g(x). 0 Exercises 16–18

14.8

Exercises

Section 14.2

1. Show that the function f(x, t)1=1a1sin1(bx)1cos1(vbt) (i) satisﬁes the one-dimensional wave equation (14.1), (ii) has the form f(x, t)1=1F(x1+1vt)1+1G(x1−1vt). 2. The diffusion equation ∂f ∂2 f =D 2 ∂t ∂x provides a model of, for example, the transfer of heat from a hot region of a system to a cold region by conduction when f(x,1t) is a temperature ﬁeld, or the transfer of matter from a region of high concentration to one of low concentration when f is the concentration. Find the functions V(x) for which f(x, t)1=1V(x)ect is a solution of the equation. 3. (i) It is shown in Example 14.2 that the function f(x, t)1=1a exp[−b(x1−1vt)2] is a solution of the wave equation (14.1). Sketch graphs of f(x, t) as a function of x at times t1=10, t1=122v, t1=142v (use, for example, a1=1b1=11), to demonstrate that the function represents a wave travelling to the right (in the positive x-direction) at constant speed v. (ii) Verify that g(x, t)1=1a exp [−b(x1+1vt)2] is also a solution of the wave equation, and hence that every superposition F(x, t)1=1f(x, t)1+1g(x, t) is a solution. (iii) Sketch appropriate graphs of f(x, t)1+1g(x, t) to demonstrate how this function develops in time.

using the real forms of the angular functions.23) satisfy the orthonormality conditions b a Z Z 0 ψ p . 9. Show that the wave functions (14.1 and 14. Some zeros of the Bessel functions Jn(x) are: J0(x)1=10 for x1=12. ∂2 f ∂x 2 + ∂2 f ∂y 2 =0 7.4048.5 11.6537 x1=13.0 and R2. 13.0 for the hydrogen-like atom. 0 < z < c elsewhere (ii) What are the possible degeneracies of the eigenvalues for a cubic box? Section 14. 8. (ii) The six diagrams in Figure 14. and sketch an appropriate energy-level diagram. 8.1z)1=10 for r = x 2 + y 2 + z 2 < a and ∞ elsewhere. . 10.2 are maps of the signs and nodes of the wave functions (14. Sketch the corresponding diagrams for the next four states.5201.0. (i) Make use of Tables 14.6 12. and sketch an appropriate energy-level diagram. 0 < y < b.3 Find solutions of the following equations by the method of separation of variables: 4.52). 7. Sketch the corresponding diagrams for the next ﬁve lowest states. Section 14.1356.0. with potential energy function V(x. y.1736 J1(x)1=10 x1=15.0156.1. (ii) Substitute this wave function into the Schrödinger equation (14.1y.0 in Table 14. using the real forms of the angular functions. y ∂f ∂f −x =0 ∂x ∂y 6. 5.4172 J2(x)1=10 x1=16. Repeat Exercise 12 for the wave function ψ2.3 are maps of the signs and nodes of the wave functions (14. ∂2 f + f =0 ∂x∂y Section 14.81). (i) Find the energies (in units of h228ma2) of the lowest 11 states of the particle in a square box of side a. The Schrödinger equation for the particle in a spherical box of radius a is −(A222m)∇2ψ1+1Vψ1=1Eψ.2 satisfy the orthogonality condition (14.7610 J3(x)1=10 x1=17.27) for the lowest six states. z ) = ∞ for 0 < x < a. 2 ∂f ∂f + =0 ∂x ∂t 5.4 8.q ( x .47) for the lowest six states. 15.8317. (ii) The six diagrams in Figure 14.414 Chapter 14 Partial differential equations Section 14.3802. 14.2 to write down the total wave function ψ1.82). (i) Solve the Schrödinger equation for the particle in a three-dimensional rectangular box with potential energy function 0 V ( x. s ( x . y ) ψ r. 10.5883 J4(x)1=10 (i) Find the energies (in units of A222ma 2) of the lowest 10 states of the particle in a circular box of radius a. and conﬁrm that it is a solution of the equation with E given by (14. Show that the radial functions R1. y ) dx dy = 0 1 if p = r and q = s 0 otherwise 9.

01≤1y1≤1b subject to the boundary conditions u(0.10)1=131sin1π x2l. y)1=10 u(x. (iv) Find the wave function and energy of the ground state. (ii) Find the particular solution for f ( x ) = sin 3π x . (i) Find the general solution of the Laplace equation + = 0 in the rectangle ∂x 2 ∂y 2 01≤1x1≤1a. (∂y2∂t)t=01=10. 0)1=10. y)1=10.14.5 for the particle in a circular box.8 Exercises 415 (i) Show that the equation is separable in spherical polar coordinates. A homogeneous thin bar of length l and constant cross-section is perfectly insulated along its length with the ends kept at constant temperature T1=10 (on some temperature scale). (iii) Use the boundary condition to ﬁnd an expression for the quantized energy in terms of the zeros of the Bessel functions. (ii) Show that the radial equation reduces to the Bessel equation (13. Find the solution of the wave equation for the vibrating string that satisﬁes the initial conditions y(x.1t)1=10. as in Section 14. The boundary conditions are T(0. x = 2 mE 2 r . ∂ 2u ∂ 2u 18. as for the hydrogen atom. u(x. Section 14. with the same angular wave functions. and satisﬁes the heat-conduction (diffusion) equation = D 2 where D is ∂t ∂x the thermal diffusivity of the material. t) of position x (01≤1x1≤1l) and ∂T ∂ 2T of time t. The temperature proﬁle of the bar is a function T(x.68). a .1t)1=1T(l.10)1=131sin1πx2l. u(a. b)1=1f(x) where f(x) is an arbitrary function of x.7 16. Find the solution of the equation for initial temperature proﬁle T(x.60) for spherical Bessel functions jl (x) where. 17. the spherical harmonics (14.

Two examples of expansions in Legendre polynomials. important in potential theory and in scattering theory. essential for the analysis of the results of diffraction experiments and in Fourier transform spectroscopy. The expansion is then f ( x ) = a0 g0 ( x ) + a1 g1 ( x ) + a2 g2 ( x ) + = ∑ al gl ( x ) l =0 ∞ (15. Fourier series are developed in Section 15. In general.3.6 that many functions can be expanded as power series and that. g0(x)1=1x01=11.4. Consider the function f ( x ) = a0 + a1 x + a2 x 2 + = ∑ al x l l =0 ∞ (15. a function f(x) of the variable x can be expanded in powers of x as a MacLaurin series if the function and its derivatives exist at x1=10 and throughout the interval 0 to x.1) can be written as a linear combination of Legendre polynomials.2) This formalism suggests that it may be possible to ﬁnd other sets of functions {gl} that can be used instead of simple powers.6. g1(x)1=1x. 15. Fourier transforms. are discussed in Section 15.1) We can regard each power of x as the function gl (x)1=1x l. The power series expansion of a function is a special case of a more general type of expansion. some functions are deﬁned by such series. The expansion is then valid within the radius of convergence of the series.2 Orthogonal expansions We introduce the concept of expansions in sets of orthogonal functions by demonstrating that the power series (15. that is. and that it may be possible in this way to expand functions that cannot be expanded as power series. and used in Section 15.15 Orthogonal expansions. with an application in electrostatics.1 Concepts We saw in Section 7.5 to solve the wave equation for the vibrating string for a given set of initial conditions. indeed.2.3) . f ( x ) = c0 P0 ( x ) + c1 P ( x ) + c2 P2 ( x ) + 1 = ∑ cl Pl ( x ) l =0 ∞ (15. The theory of orthogonal expansions is introduced in Section 15. are discussed in Section 15. The sets of functions that are of particular importance for expansions in series consist of functions with the property of orthogonality. Fourier analysis 15. g2(x)1=1x 2 and so on.

The Legendre polynomials were discussed in Section 13.5) The power series (15.4) that x 0 =1P 0 1 x 2 = ( 2 P2 + P0 ) 3 x4 = 1 (8 P + 20 P2 + 7 P0 ) 35 4 x1 =1P 1 1 x 3 = ( 2 P3 + 3P ) 1 5 x5 = 1 (8 P + 28 P3 + 27 P ) 1 63 5 (15. it follows from (15. and the ﬁrst few are P (x) =11 0 1 P2 ( x ) = (3x 2 − 1) 2 1 P4 ( x ) = (35x 4 − 30 x 2 + 3) 8 P1(x) =1x 1 P3 ( x ) = (5x 3 − 3x ) 2 1 P ( x ) = (63x5 − 70 x 3 + 15x ) 5 8 (15.3). For example.7) .3) in terms of the coefﬁcients al in (15. The polynomials are orthogonal in this interval (see equations (13.6) (8 P4 + 20 P2 + 7 P0 ) + a a = a0 + 2 + 4 + 3 5 3a 3a P0 + a1 + 3 + 5 + 5 7 3a 4 a P2 + 3 + 5 + 5 9 P + 5 2a 4a + 2 + 4 + 3 7 8a + 4 + 35 8a P4 + 5 + 63 and this has the required form (15.4.27)). and it is possible to express every power x l as a linear combination of Legendre polynomials of degree up to l.4) The function Pl (x) is a polynomial of degree l in x. the orthogonality of the Legendre polynomials can be used to derive a general formula for the coefﬁcients cl in (15.5)).1) can therefore be written (using only terms from (15. f ( x ) = a0 P0 + a1 P + 1 + a4 35 a2 3 ( 2 P2 + P0 ) + a3 5 ( 2 P3 + 3P ) 1 a5 63 (8 P + 28 P3 + 27 P ) + 5 1 P 1 P3 (15.l 2l + 1 if if k ≠l k=l (15.25) and (13.15.1). +1 Z −1 0 2 Pk ( x ) Pl ( x ) dx = δ = 2 2l + 1 k . The Legendre polynomials occur in the physical sciences with the variable x1=1cos1θ.2 Orthogonal expansions 417 and that. when the value of the variable is restricted to the interval −11≤1x1≤1+1. so that we are interested only in the interval −11≤1x1≤1+1.

9) for k1=10 to ﬁnd c0. We multiply equation (15. Fourier analysis and this can be used to obtain a general formula for the coefﬁcients cl in (15. Then.19). Then Z Pk ( x ) f ( x ) dx = Z Pk ( x ) ∑ cl Pl ( x ) dx = ∑ cl Z Pk ( x ) Pl ( x ) dx −1 −1 +1 +1 ∞ ∞ +1 l =0 l =0 −1 and because.3) by Pk (x) and integrate.8) This important formula gives the values of the coefﬁcients cl in the expansion of a ﬁnite continuous function of x.7).9) Replacement of Pk (x) in the integrals by its expansion in powers of x. by (15. EXAMPLE 15.1 Use equation (15.9) is 0 c0 = where +1 1 ∞ ∑a 2 l =0 l Z x l dx −1 +1 Z −1 x l +1 +1 2 = l +1 x dx = l + 1 −1 0 l if l is even if l is odd . l =0 ∞ ck = 2k + 1 ∞ ∑a 2 l =0 l Z Pk ( x ) x l dx −1 +1 (15. for the power series f ( x ) = ∑ al x l . only that term contributes to the sum: Z Pk ( x ) f ( x ) dx = ck Z Pk ( x ) Pk ( x ) dx = −1 −1 +1 +1 2 c 2k + 1 k Therefore ck = 2k + 1 Z Pk ( x ) f ( x ) dx 2 −1 +1 (15.418 Chapter 15 Orthogonal expansions. We have P 1=11 so that equation (15. then gives the coefﬁcient ck in terms of the coefﬁcients al.4) or from the general formula (13. from tabulations like (15.3). for | x|1≤11. each integral on the right is zero except for the one with l1=1k. in terms of the Legendre polynomials.

2 Orthogonal expansions 419 Therefore c0 = a0 + a2 a4 a6 + + + 3 5 7 =∑ l =0 ∞ a2 l 2l + 1 and this is in agreement with the coefﬁcient of P shown in (15. f ( x ) = ∑ cn g n ( x ) n= 0 ∞ (15. cn = Z g*( x ) f ( x ) w ( x ) dx n Z g*( x ) g n ( x ) w ( x ) dx n a a b b (15.10) (g* 1=1gm for real functions). 0 0 Exercises 1. be a set of functions. Let f(x) be an arbitrary function.11) Then. n1=11. 2. it is the square of the norm gn = Z g*( x ) g n ( x ) w ( x ) dx n a b (15. that are orthogonal in the interval a1≤1x1≤1b with respect to the weight function w(x): Z g m x ) g n ( x ) w ( x ) dx = 0 *( a b if m≠n (15.15. that can be expanded in the set {gn(x)}. deﬁned in the interval m a1≤1x1≤1b. possibly complex. 2 The general case Let gn(x).6). multiplication by g* (x) and integration with respect to weight function w(x) m in the interval a to b gives *( *( Z g m x ) f ( x ) w ( x ) dx = ∑ cn Z g m x ) g n ( x ) w ( x ) dx a n= 0 a b ∞ b *( = cm Z g m x ) g m ( x ) w ( x ) dx a b so that (replacing m by n).13) . =. 3.12) The denominator in this expression is the normalization integral of the function gn(x).

n a b (15.17) and (15. These sets share the property of completeness: An orthonormal set {g(x)} deﬁned in the interval a1≤1x1≤1b with respect to weight function w(x) is called complete if ‘every’ function f(x) deﬁned in the interval can be represented by a linear combination of the functions of the set. . The coefﬁcients in the expansion (15. Completeness of orthogonal sets A number of orthogonal sets have been discussed in previous chapters.18) The concept of orthogonal expansions is readily generalized for functions of more than one variable.16) The expansion (15. f ( x ) = ∑ cn g n ( x ).19) This expression of completeness is sufﬁcient for most purposes. *( Z g n x ) g n ( x ) w ( x ) dx = δ m. g n ( x) = 1 g ( x) gn n (15.11) then becomes f ( x ) = ∑ cn g n ( x ) n= 0 ∞ (15. We consider two qualiﬁcations.15) The resulting set of functions is an orthonormal set.14) It is often more convenient if the functions of the expansion set are normalized. as well as orthogonal. in particular.14) for the expansion coefﬁcients becomes *( cn = Z g n x ) f ( x ) w ( x ) dx a b (15. Normalization is achieved by dividing each function by its norm.11) of f(x) are then given by cn = 1 gn 2 Z g*( x ) f ( x ) w ( x ) dx n a b (15.420 Chapter 15 Orthogonal expansions. n= 0 ∞ cn = Z g *( x ) f ( x )w ( x ) dx n a b (15. the several ‘special functions’ in Chapter 13 and the solutions of the Schrödinger equation (eigenfunctions of the Hamiltonian) in Chapters 12 and 14. Fourier analysis The norm of a function is sometimes interpreted as the ‘magnitude’ or ‘length’ of the function.

23) converges when | x|1<11 for all values of t.21) The series is said to converge to f(x) in the mean. The series (15. 15. and its derivation is discussed in Example 15.15. Two examples in the physical sciences are (1 − 2 xt + t 2 )−1/ 2 = ∑ t l Pl ( x ). but care must sometimes be taken when equating derivatives and integrals of the series to the derivatives and integrals of the function it represents. 1 This was ﬁrst proved by Dirichlet for the Fourier series (‘Dirichlet conditions’). Gustav Peter LejeuneDirichlet (1805–1859).23) of importance in scattering theory.3 Two expansions in Legendre polynomials 421 (1) By every function is meant not only continuous functions but the more general class of piecewise continuous functions that have a ﬁnite number of ﬁnite discontinuities and a ﬁnite number of maxima and minima in the interval. and the function can in practice be replaced by the series even though (15.3.59)). It is this convergence in the mean that allows the representation of some discontinuous functions.22) of importance in potential theory (electrostatics and gravitational theory).21) means that the series need not converge to f(x) for every x. and is derived in Example 15.20) (15. . l =0 ∞ −1 ≤ x ≤ +1 (15. German mathematician.3 Two expansions in Legendre polynomials We saw in the previous section that a power series in x can be represented as a linear combination of Legendre polynomials. where the jl (t) are spherical Bessel functions (equation (13.1 (2) By the representation of the function is meant that the function can be approximated arbitrarily closely by a series f(x)1≈1fk(x)1=1c0 g0(x)1+1c1g1(x)1+1-1+1ck gk(x) and the limit of the series is such that 2 Z f ( x ) − f k ( x ) w ( x ) dx → 0 as k → ∞ a b (15. The series (15.22) converges when | x|1<11 and | t|1<11. was inﬂuenced by the work of Fourier whilst a student in Paris in the early 1820’s.2. and eitx = ∑ ( 2l + 1) i l jl (t ) Pl ( x ) l =0 ∞ (15.

. .q.... . . . Comparison with the Legendre polynomials listed in (15.... ... .... . ... .. . .. . ... .... . . ... . Figure 15.... .. . .. . f ′′′(0)1=115x 31−19x. ... . . . ...... .2 Derive the expansion (15.... ....... .. .. .... ... . . . .. . ..55) and (13.. ... ...24) o q 4 πε0 Rq .. . ...59) for Bessel functions of half-integral order that c0 = 0 Exercises 3.. . . .. .. ... ... We have f(t) =1(11−12xt1+1t 2)−122 f ′(t) =1(x1−1t) (11−12xt1+1t 2)−322 f ″(t)1=13(x1−1t)2 (11−12xt1+1t 2)−5221−1(11−12xt1+1t 2)−322 f(0) =11 f ′(0) =1x f ″(0)1=13x 21−11 and similarly. We treat the function to be expanded as a function of t.22).. .. ... ... .. The potential of a single charge q . . .. ... .. . .... The MacLaurin series f (t ) = f (0) + t f ′(0) + is then f(t)1=1(11−12xt1+1t 2)−1221=1P 1+1tP (x)1+1t 2P2 (x)1+1t 3P3 (x)1+1t 4P4 (x)1+10 1 and this is equation (15....23).. 4 The expansion of electrostatic potential π 2t J1/ 2 (t ) = j0 (t ) V= (15.... .... and expand it as a MacLaurin series. ..• .... . ... .1 R θ r The electrostatic potential at point P due to the presence of a point charge q is (see Figure 15..•.... .. . . f ′′′′(0)1=1105x 41−190x 21+19... .. ... . .4) shows that.. ... ... . ...1) • ... ... .. .. . . .. p . R . Therefore −1 +1 1 c0 = 2 Z e −1 +1 itx 1 dx = 2 eitx +1 1 eit − e− it +1 = 1 sin t = −1 t it −1 2it It follows from equations (13. f (l )(0)1=1l!Pl (x)... ... . If eitx = ∑ cl Pl ( x ) then cl = l =0 ∞ t2 t3 t4 f ′′(0) + f ′′′(0) + f ′′′′(0)1+12! 3! 4! 2l + 1 2 Z Pl ( x ) eitx dx ...3 Find the ﬁrst term in the expansion (15..... for the lth derivative. Fourier analysis EXAMPLE 15...... . ..422 Chapter 15 Orthogonal expansions.. . . .. ..22).. .. ...... EXAMPLE 15.. and so on...............

..3 Two expansions in Legendre polynomials 423 where Rq is the distance of point P from the charge (if a charge q′ is placed at P.. . .... . .. q1 at (x1.... . . ... . . .. . .... ... . .. . .. . .. .. .. ... ...... . . ... ... .. .. ... .. ..... .. . . . . . .. . . ..... . ...... . .... ...... . . . .. . by equation (15.. qn at (xn. . the same expansion is valid..... .... .. . .. and the total potential at P is the sum of these contributions. ....... ... q2 at (x2......... ...... . ... . . .. . . .. ... ...... . We write (15..... . . . for all i.. y2..... .. . .. . . The potential of a distribution of charges We consider a system of N charges. . .. ... ............ ...... but with r and R interchanged. . . . ..... zn) as illustrated in Figure 15. . .. . .22) with t1=1r2R and x1=1cos1θ. . ..... . .... . z2). ... .. .... . ..... ... ... V =∑ i =1 N If the point P is exterior to the distribution of charges.......25) We consider the case R1>1r. .. .. this can be expanded in Legendre polynomials as V= r ∑ R Pl (cos θ ).. . ... . 4 πε0 R l = 0 q ∞ l r <1 R (15.. . . .. so that R1>1ri .... ...15.... ..26): r i V =∑ ∑ R Pl (cos θ i ) 4 πε0 R l = 0 i =1 N Figure 15......... .. . .. ... . ..2 (for 2 charges). .... By the cosine rule 2 for the triangle OPq.... .24) can be written as V= q 4 πε0 ( R 2 − 2r R cos θ + r 2 )−1/ 2 (15....... .. .... .. .. . . . ... . . . . ..2 qi ∞ l = 1 4 πε0 ∑ l =0 ∞ 1 N ∑ q r l P (cos θ ) i i l i R l +1 i =1 (15.. z1). .. .27) 2R p .. . .... . . . ... ... . . ... ... =..... .... .. . . ..... .. ...... y1..... . . ... . .... .. . .. ..28) 2q • 2r • o 4 πε0 Ri 2θ qi (15. . . ........ ......... .. ..... ... .. .... we have R q1=1r 21+1R 21−12r1R1cos1θ. .. so that (15.... ... .. ... .. .. .. . . . .. ....... . . .. ... . .. . . .......... . yn...25) as −1/ 2 2 1 − 2 r cos θ + r V= 4 πε0 R R R q and. .. .26) When r1>1R.. .... Each charge makes its individual contribution to the electrostatic potential at point P. . ... . the potential energy of the system of two charges is Vq′1=1qq′24πε0Rq)... .. the potential can be expanded term by term as in (15... .. .. . ... .. . .. . . • 1R R • 1q 1θ 1r .......

..... and the total charge is zero. . .. .... ...29) in (15...... ... . .. . .. . Q2 = q ⋅ (3 cos2 θ − 1) − q ⋅ (3 cos2 ( π + θ ) − 1) = 0 2 2 2 2 and..... . .3 • p R q +• θ o 2/r For the system shown in Figure 15. ....... Q1 = 1 1 qr cos θ − qr cos( π + θ ) = qr cos θ = µ cos θ 2 2 where µ1=1qr is the (scalar) dipole moment of the pair of charges. .. ....... ... ... ... . . because P (cos1θ) is an even function of cos1θ. ...... .. .. The expansion (15... . ... . .. and µ1cos1θ is the component of the dipole moment in the direction OP. . ....... ........ .. .28) are called the electric multipole moments of order l of the charge distribution (strictly.. The ﬁrst few moments are N Q0 = ∑ qi i =1 N total charge Q1 = ∑ qi ri cos θ i i =1 N 1 Q2 = ∑ qi ri2 (3 cos2 θ i − 1) 2 i =1 dipole moment (15. . The quantities Ql = ∑ qi ri l Pl (cos θ i ) i (15.. ..... ... r 1 r 1 3... ... ... . .. .. .. .. . Ql is the component of the multipole in the direction OP.30) quadrupole moment EXAMPLE 15.... for example in the theory of intermolecular forces. . . . ... .. . .. .... . .... . Q01=1+q1−1q1=10 Figure 15. ...28) is called the multipole expansion of the potential.... . . .. ....... see Chapter 16). . Fourier analysis This analysis of the potential is important for the description of the electrostatic interaction of charge distributions....... . . .....424 Chapter 15 Orthogonal expansions..... .. every even-order multipole l moment is zero.. ............4 The ﬁeld of a dipole . . 2 2 • q− 2/ r 1...... . . ... 2.. .. ... . ....... . .3 of two unlike charges +q and −q separated by distance r: . . ... . .. ..

These functions are orthogonal in the interval (and in every interval of width 2π). = (15. 0 Exercises 5. m ≠ n −π +π (15. 6 15. m ≠ n −π +π (15.4 Fourier series The Fourier-series representation of a function f(x) is the expansion of the function in terms of the set of trigonometric functions cos1nx. = n1=11.15.32) Z sin mx sin nx dx = 0.34) and the corresponding normalization integrals are Z cos nx cos nx dx = −π +π 2 π if n = 0 π if n > 0 if n > 0 (15.4 Fourier series 425 The potential at point P is then V= µ cos θ 4 πε0 R 2 + Q3 4 πε0 R 4 + Q5 4 πε0 R6 + When R is large enough (R1>>1r).31) in the interval −π1≤1x1≤1π (we consider intervals of arbitrary width later in this section). n −π +π (15. the expansion of the potential can be truncated after the leading term. n1=10. V≈ µ cos θ 4 πε0 R 2 This is the potential that dominates the long-range interactions of a neutral polar molecule.36) .35) Z sin nx sin nx dx = π −π +π (15. 2. 2. sin1nx. 3. Z cos mx cos nx dx = 0.33) Z cos mx sin nx dx = 0. 1. all m.

35). and provided a new and powerful tool for the analysis and solution of physical problems. Fourier analysis The (trigonometric) Fourier series is usually written in the form2 f ( x) = a0 2 + a1 cos x + a2 cos 2 x + a3 cos 3x + + b1 sin x + b2 sin 2 x + b3 sin 3x + = a0 ∞ + ∑ ( a cos nx + bn sin nx ) 2 n =1 n (15. because of the orthogonality of the expansion functions. .5 Conﬁrm the relations (i) (15.37) and. and published in ﬁnal form in 1822 in the Théorie analytique de la chaleur (Analytic theory of heat). cos mx cos nx = so that 1 cos( m + n) x + cos( m − n) x 2 Z +π −π 1 cos mx cos nx dx = Z cos( m + n) x + cos( m − n) x dx 2 −π = 1 2 sin( m + n) x sin( m − n) x + π =0 + m − n −π m+ n +π and the zero is obtained because the sine of an integer multiple of π is zero. Z +π −π +π dx = x = 2 π −π 2 Jean-Baptiste Joseph Fourier (1768–1830) studied the series in connection with his work on the diffusion of heat.39) −π EXAMPLE 15. The work was ﬁrst presented to the French Academy in 1807. (ii) For m1=1n1=10. when cos1mx1=1cos1mx1=11.426 Chapter 15 Orthogonal expansions. when m1≠1n. the Fourier coefﬁcients are given by an = 1 π Z +π f ( x ) cos nx dx +π (15. It generated new developments in the mathematical theory of functions.32) and (ii) (15.38) −π bn = 1 π Z f ( x ) sin nx dx (15.22). (i) By the trigonometric relations (3.

.. a function that satisﬁes the periodicity relation (15.. .. ... .. . . . .... . .... . . . . . . ... .... . .. . ......... . .... . .. . . ... . ... ... .... . .... . . ... . .. . . . . ..... ... ... .. .. ... . . In general. . . . . . .. ...... . . .. . ... ... . . .. .. .. .. .. . . . therefore. ... . . ..... . .. . . .. that is.. . . . .. ... . . . .. . . . .. ....... ... . . . ... . . . . ... . .31) are periodic functions of x with period 2π and.. . . .. .. ... .. .. . .. . .. .. .. . .. . . . . . .. . . .. . . . .. . .. . . . . .... . .... . .. Despite the discontinuities (of the gradient) exhibited by the function.. . ..40) can be expanded as a Fourier series if it is single-valued and piecewise continuous. continuous except for a ﬁnite number of ﬁnite discontinuities and only a ﬁnite number of maxima and minima in any ﬁnite interval (that is... .. . . ... .5... .. . . .. .. .. ... . .40) (15. .. . . .. . . ... .. ... . . .. ...... . . . ..... . . the function sin x ... .. . . . . . .. . . .. . .. . ... .. .. . .. . . .. . .. . . . .. .. .... ... . .. . .. ... . .6.. .. . . .5 Periodic functions occur in the mathematical modelling of physical systems that exhibit periodic phenomena. ... . .. ... . . .. . .. This means that the function f(x) can be extended to values outside the base interval −π1≤1x1≤1π by means of the periodicity relation )x( f f(x1+12π)1=1f (x) For example.. ... . . . . .. . .. . . .. .. .. .. . .. ..... . .. .. . . . .15... .. .... .. .. . . .. ..... . . .. . . . .. . . . . ...41) Figure 15. .. . . .. ..... ... ... ..... . .. . ... . ... ... . . .. .. . . .. . might represent the result of passing an alternating electric current through a rectiﬁer that allows current to ﬂow in one direction only (a half-wave rectiﬁer)... ... .... .. . .... . . .. . . . . . . . . ... .. .. . .. An example of such a function is discussed in Example 15. ... . . . . . .. ... ..... ... .. ... . ... For example...... . . . . . ..... ... .. . ... ... . . .... . . . .. . .. . . . . . ... . x π+ .. . . .. ... . .. . . . ..... .. . . ..... ... . ......4 shown in Figure 15. . . . . ..... . . 1 ... ... . . ... .. Z cos nx dx = Z 2 −π +π +π −π +π 1 1 sin 2 nx =π x+ (1 + cos 2 nx ) dx = 2 2 2n − π 0 Exercise 7 Periodicity The trigonometric functions (15. . ... .. .. . .. . . .. ... .... . . . .. the function shown in Figure 15. .. . it is one of an important class of functions that can be represented as Fourier series. . . ... . . . . .. .. . . .. .. . .. . ... . . . .... . . ... .. ... .5. . . .. . . ... .. .... . .. . .. ... .. ... . ..... . . . ... ... . . .. . . . . . . . ... . . . . . . .. ..... .. .. . . .. ... .. . . . . .... . .. .. . . . .. .. . every linear combination of them is also periodic. .. . . .. .. . ... .. . ... . . .. .. with f(x) as current and x as time. . ... ... .. . ... 3 + 0 π π− 2 π+ )x( f + 1 0 π− π − 2 π − 3 ... .. . . . . . . .. .. . . . . .. . . . . x π . . ... . . .. .. . every reasonably well-behaved function). .. . . . .. . . . .....4 Fourier series 427 For m1=1n1>10.. . .... f ( x) = 0 for for 0≤x≤π −π ≤ x ≤ 0 (15. . . . .. . . . . . . . .4 can be extended by means of (15.. . .. .. .. .. ..... . . . . Figure 15.. .. .. . .. . ... . . . .. ..40) to give the periodic function shown in Figure 15.... . ..... .. . . .. . .. . .

.......... .......................... ......... . . ... .. .... In the present case...... ... .... ... . ..... ...... and is equal to unity in the second.... . ...... . f ( x) = 0.. this mean value is 122....... . .......44) .. .. .... .... Fourier analysis EXAMPLE 15.. .......6 The coefﬁcient a0.. .......... .. ... .. . .. .... .. 2 π+ ... .. . .... .... . .. . .. . ... ..... ... . ..... . .... . .. ....37)....... shows that the function is discontinuous when x is a multiple of π.. for for 0<x<π −π < x < 0 (15.. ... ... ....428 Chapter 15 Orthogonal expansions. ... ... ... .. ........ .. .. . ........ .. ..................... By equation (15.... .......... ... f ( x) = a0 ∞ + ∑ ( a cos nx + bn sin nx ) 2 n =1 n everywhere other than at these points of discontinuity.......... ......6...... ... ... . It is a property of Fourier series that the value of the series at a point of discontinuity is the mean value of the function at the point..... .. .... .... .. . ............. ............38)... .. ............. ...... .... Figure 15.. ... ... .... .... ...38) with n1=10.... ............. ................ .. .. 3 π ............. ... ... .... ............... ... ....... ................. x π . .6 The Fourier series of the function 1. ... .... . ... ........... .... .. ......... ............. n1>10...... .. ... ........... ... ... − 2 π − 3 (15...... ... ...... ..... .42) The graph of the (extended) function............... . and the function can therefore be expanded as the Fourier series (15.......... . ..... . . .. ....... .. . π ... ..... ....... ....... a0 = 1 π Z +π f ( x ) dx = 1 π −π Z 0 f ( x ) dx + 1 π −π Z +π f ( x ) dx 0 The function f(x) is equal to zero in the ﬁrst integral on the right.............. ..... .... ............. . .. ..... . .. ........ ..... . an = 1 π Z +π f ( x ) cos nx dx = 1 π −π 1 sin nx =0 Z cos nx dx = 0 +π π 0 π n All the Fourier coefﬁcients an are zero except for a01=11........... .......... . Figure 15... .... .. . ...... .... ...... ... Therefore..... ..........43) (15......... .... ......... ...... .. . ............. . .. ... ........ .. ....... )x( f ... ...... 0 + 1 π− ....... ...... By equation (15... a0 = 1 π Z dx = 1 0 +π The coefﬁcients an..

.. ... .. ... .... . . ... . . . .... .. .. . ...... . ... ... . .. .. .... . ......... .. 2S.. ... π+2 2 1 .... ........ ... . ... .. .. ..... ........ . ... . . . .. ... ................ . π+ ... ..... ........... ..... .... . .. .. ................ ... .. .. ... ..... .. . .. ......x5 i s n 5 + x33 ni s + x nis π+ . and cos1nπ1=1−1 when n is odd. . . .... . ........... ...... ... . .. . . ........... ..... .... .... .. .. .. . . ..... bn = - bn = S3 = S1 = f ( x) = π 1 Z −π +π 1 2 2 ...... .... . .. .. .. ... ... . π+ 2 2 1 = 4S ... . .. ...... ... ... .... ... ... ..... ....... . . . ..... .. .. ...... . π− .. . ..... . .... .. .. .... ... ........... .... .... . .... .... ... ..... . .... .... ...... .. ...... .. . π+ ........ .. .. .... .. ... . ...... . ...... . ........ ..... ...... . . .. . . ... . .... .. ... .. ...... .... ....... .. .. ... . . .... .... . ........ . ................ ... . ..... .... ..... .. . ... ... ......... ..... ... .... ............... . .... .. .. . ....... ...... . . . ... ...... ..... ........ ....... .. . .... ...... .. . . .. ... ... . . . .. .. . ....... . . ............ .... .. ......... .. ... . . ... ... ... ..... ... ........ ........... . . .... ..... ... ... ... ........ . . . . .. . .. . ... nπ 1 2 sin 3x + sin x + 2 π 3 sin 3x sin 5x 1 2 = + sin x + + + 2 π 3 5 a0 f ( x ) sin nx dx = 2 + b1 sin x + b3 sin 3x + b5 sin 5x + n odd π 1 1 cos nx s 1 = (1 − cos nπ ) Z sin nx dx = − Figure 15......... . .. ... . .... .... .. .. .. . By equation (15. .... . ...... ..... ........ ... . .. .... . .... ....... . . ... . .. ..... ..... ..... .. . ... . . . .. . . . ... .. .......... . .. 0 ... . . ..... . . . . .. . ...... . 4S.. . . . .. . .. . .. ... ... . 1 + 2 x nis π + 1 2 = 2S ... ............. .. .... ...... ...... .. ..... .. . .. ..... ..... ... . ...... ........ ... . ... .. .. .. .. ....... . ... .... . . .. .. . .... ... . .. .... . .... ... The graphs of the ﬁrst four partial sums are shown in Figure 15... ... . .... . ... ....... ... .. ....... .... ....... ... . . .. .... . .... ... . ... .. ...... ...7. ........ .... .. .. .. ...... + 1 To illustrate the behaviour of the series we consider the n-term approximations (partial sums)...... = 3S .... . ... .. . ............. . . .. .... .. .... . . . .. . . .. .. . ......... .... .... It follows that bn is nonzero only when n is odd: The coefﬁcients bn.. . .. . .. . . + 1 0 +π S4 = S2 = 1 2 sin 3x sin 5x + sin x + + 2 π 3 5 1 2 + sin x 2 π π n 0 π nπ 15. .. . ... . . . .... . .. .. 1 + 2 1 = 1S 0 π− 1S π+ ....... ... .. ... ........ 3S. ... . the Fourier series of the function (15. ..... . .... ....... . .. . . .. .. . .... . . .45) 429 . .. .... . .... 0 .... ... . . ...... . ... ..... .. .... .. ..46) (15.... .. By the results (15. ... ......... ... .. .. . . ...... .. . .... ......... .... .. .... .. .. .. ..42) is therefore We have cos1nπ1=1+1 when n is an even integer.. .. ..... .. .. .. .. ..45).. . ... . . .... π− .... .. ... .......7 x3 n i s 3 + x nis 0 ....... . .. . . .... . .. ..... .. .... .... ... ...... ... ..43) to (15....... . .. .. . . .. .. ......... . . ..39)...... ..... . π− .4 Fourier series (15...... . . .. .

.. ........... ........... ....... .... . .... . ........39)...... ........... .. ..... .. . ... ... ............ .. .. . .. . .... .. ... ... . .......49) (15.. ....47) a0 ∞ nπ x nπ x + ∑ an cos + bn sin l l 2 n =1 (15... .. .. .. 0 ................. 05S π− .. .... ... .... . .... The constant term S1 is the mean value of the function in the interval.... .. ....... . ...... . ... . ...37) to (15.. ...... . ... ... .. .......... .... .. ............ and is periodic in z with period 2l. .... .......... .... . ........ includes the ﬁrst correction to this mean value..... .8 and S3......... a function f(x) that is periodic in x with period 2l. Thus.. ..... .. .8.. .... .... ..... Thus...... . ... . .......... Figure 15. + 1 ........ .............. . ... .. 0 Exercises 8–11 Change of period A periodic function f(x) with period 2π is transformed into a periodic function g(z) with period 2l when the variable x in f(x) is replaced by πz2l... . ... .... .. .... ........ .... ..... .. ..... if πz f ( x) = f = g( z) l then π f ( x + 2 π ) = f ( z + 2l ) = g ( z + 2l ) l For example. .......... .... ... equations (15. ....... . ............. ..... Fourier analysis This example shows how an arbitrary wave is built up by interference of harmonic waves.. .. ..... .. The graph of the partial sum S50 is shown in Figure 15..... . ... . .......... . .. ... ... ..48) (15............ .. .. .. .. for the expansion of functions with periods other than 2π... .. with constructive (additive) interference in some regions and destructive (subtractive) interference in others..... ... . ....... . . deals with the discontinuities of the function. This simple change of variable is used to generalize the Fourier method... .... . the function sin1x1=1sin(πz2l) is periodic in x with period 2π. S4...... ......... .... .. .. . ... .50) π+ .. ........ ........... ..... .. . . and demonstrates clearly how the Fourier series... . The two-term approximation.... .... f (x)1=1f (x1+12l) can be expanded as the Fourier series f ( x) = with Fourier coefﬁcients nπ x 1 an = Z f ( x ) cos dx l −l l bn = nπ x 1 dx Z f ( x ) sin l l −l +l +l (15. continuous at all points in the interval.. . ..... . .. = include successively smaller corrections. .....430 Chapter 15 Orthogonal expansions....... ....... .. S2... .

... ... .9 shows that the function is an odd function of x.. or Figure 15.. Of the expansion functions in (15.. ...... .... . .9 if it contains an odd component. . .. .3 that an integral . +l . . ...... ... .4 Fourier series 431 The following example demonstrates the Fourier series for an arbitrary period.. we then have 2 Z l 0 l/2 l nπ x nπ x dx + Z f ( x ) sin dx l l l/2 bn = f ( x ) sin = l/2 l nπ x 4A nπ x dx + Z ( l − x ) sin dx Z x sin 2 l l l 0 l/2 A− .. . Z x .. .. ...51). It also follows from the discussion of Section 5... .. ..... .. l− 0 . ..... . .. ... . . . ... ..7. . .. . ... equation (5... ..7 The Fourier series of the function 2A − (l + x ) l 2 Ax f ( x) = l 2A (l − x ) l for for for 1 2 1 1 − ≤x≤+ 2 2 1 + ≤ x ≤ +l 2 −l ≤ x ≤ − ) ( (15....51) is zero unless g(x) is also an odd function of x..... . ... ........... . .52) (15. . It follows from the discussion of even and odd functions in Section 5. .. .... A+ xf . . ... All the integrals (15. .... . . .. .........48).. .... that the integral (15. ....3.. ..... the sine functions are odd but the cosine functions are even...... . . .25). .53) l+ . ..... .... ............ EXAMPLE 15. Figure 15. ... .. . . −l .. .. .... . and it also prepares the ground for the completion of the solution of the wave equation for the vibrating string discussed in Section 14.... . ..... .... f ( x ) g ( x ) dx (15. . ... ..49) for an are therefore zero. .. and the Fourier series reduces to the Fourier sine series for the expansion of an odd function: f ( x ) = ∑ bn sin n =1 ∞ nπ x l with coefﬁcients given by (15... . . .... ... .. . ... ..... .. .50)..... .. ..15. . . ...... . .. . ..50) over the whole interval −l1≤1x1≤1l is equal to twice the integral over the half-interval 01≤1x1≤1l: bn = 2 nπ x Z f ( x ) sin dx l 0 l +l For the function (15... with f(−x)1=1−f(x).. ...

. .. .. .. . ... .57) ∞ ∂y nπ x = g ( x ) = ∑ Bnω n sin l ∂t t = 0 n =1 y ( x . 15. .. . . . . . . .5 The vibrating string It was shown in Section 14. . . .104)) y ( x . . . Fourier analysis and.. .. 0 8A bn = + 2 2 n π − 8A n2 π 2 Therefore f ( x) = πx 8A 1 1 3π x 1 5π x sin − 2 sin + 2 sin − l l l 3 5 π 2 12 if if if n even n = 1. 0) = f ( x ) = 8A 1 sin π x − 1 sin 3π x + 1 sin 5π x − 2 2 l l l π 1 32 52 (15.55): l ... .. .... and then let go... .10 discussed in Example 15. . .. . 13 15... The initial shape of the string is shown in Figure 15. .. 13. . .. .7 that the solution of the wave equation for the vibrating string of length l is (equation (14. . . .. We suppose that. . . .... . . . . . ..... . .. . . . . . . . .. .. .... . . . ... . .. ) ( .... . 0) = f ( x ) = ∑ An sin n =1 ∞ nπ x l (15... A xf • 0 . .56) in which the coefﬁcients An and Bn are determined by the initial conditions (equations (14. .. . .. . . . and the displacement function is therefore identical to the function Figure 15..432 Chapter 15 Orthogonal expansions. ... 9. .. ... .. . ..54) (15.. 5. The Fourier representation of the function is then given by equation (15. ..103)) y ( x ... . .. . .55) 0 Exercise 12. . .. .. . ... at t1=10. . except that it is deﬁned only within the half-interval 01≤1x1≤1l. .... . ... after integration by parts. . ... … n = 3.... .. .. . . . .. .. . ... . . ....10.7. … (15. .. . t ) = ∑ sin n =1 ∞ nπ x A cos ω nt + Bn sin ω n t l n (15. . . .. .... . .. we pull the string sideways at its centre. . 7. . . .. . . . 11.58) x where f(x) and g(x) are given displacement and velocity functions that describe the state of the system at time t1=10.. . .. ..... . . . ...• ....... . .

.. .. . . . It follows that... . . .. . . ... . .. . .. ..... .. . .. .. .. .. . .. ... ... . . the graphs have been obtained from the 25-term approximation to the wave function (terms up to n1=149).. . ....... ... with the motion determined by the instantaneous force acting at the two points where the gradient changes (or at the single point at the turning points of the motion).. .... .. .. . . . .. . .. . . . . . .... . . ..... . . .. .... . . . .......... .. ... ... ... .... . . . . . . .......60) is periodic in t with period τ1=12π2ω11=12l2v.. . . . ... .. . ...... .. .... ... ......... . .. . . .. ... ......6 and the coefﬁcients An in (15.. .... . The Fourier analysis of nonperiodic functions is achieved by letting the width of the base interval become indeﬁnitely large (l1→1∞). ...... ..... ... . ... . ..... . ... .. ..... ... . . . In Section 15.... but several important applications of Fourier analysis in the physical sciences involve functions that are not periodic.... . .. . ..... ..... ... . .. . ..... . 61/ τ3 = t •....... .... .. . ... . .. . .. .... .. .... .. .. ......... . . . ...11 shows how the string behaves over the ﬁrst quarter of a period... . ... .. .... .... ωn1=1nπv2l with v constant..4/ τ = t •. ... . ...... . ... .... ... ... . .. .. . .. . ... .. .. ... ... ....... .. ... .... The essential shape of the string is therefore maintained.... . ... ..... ....... . .. no net force acts at a point on a straight-line section.. .... ......... . .. .... . and Figure 15. 0=t • .. . .... . ... .. .... . and all the coefﬁcients Bn in (15.. .. . .. . .. .. ..... . .... .. . .. ...... . by equation (14... . ...•. ..... . . . ... . ... . .... . . . . 0 Exercise 14 This startling behaviour can be understood by a consideration of the forces acting at each point on the string. ... ... .. . ... .. .. . .. ...... ....... .... .. .. . .. .. .. . . .. .... ... .. .... ... .... . ... . . ....... ... .... ... .. .. and by transforming the Fourier series into an inﬁnite integral... . .. so that y( x.. . .. . . ...... .. . .. . .... .. Because the tension is uniform throughout.. .. .. . . . where... . ... ... . ... . . . .... ... .. . .. .. .. .. . ...... .......... . . . ..... 8/ τ = t • . ... . ... ......... .. .. .. .... ...... .... . .. ... . ... .. . .. . . . .. . .. . . . .. . .. .. . .... ... .... ... . .. .. . .6 Fourier transforms (15. .. . . ... . . . ... . .. ...... ... . .. .. .. .... .. .... .. . . . ...... . . . .•... . . called a Fourier integral or a Fourier transform.... . ... . .. . .. ..... .11 15... .. . .. . . . ... .. . ... . ........96)... ......... . . . ....... ..... . .. . . ..... . .. . . .. the motion of the string is given by the wave function 15... .4 we were concerned with the use of Fourier series for the representation of functions that are periodic. . . ..... .. . ... ... . .. . . . .... .. . .. ... ......... .. . In addition. .... t ) = Fourier transforms πx 8A 1 3π x 1 5π x sin cos ω1t − sin cos ω 3t + sin cos ω 5t − l 9 25 l l π2 ∂y = g( x) = 0 ∂t t = 0 Figure 15... .. . ... ......56) are zero. .. . . . .. . .. . . . ...... .. . .... ....... ... . The function (15. . . .. . .. ... . ..... .. .. . .. ... ...... ...... • • .. . . . . ... ..... . . ...... ... .. ...... . .. . ... ... . .... the initial velocity of the string is zero at all points along its length. . . .. .... .. • • . .. . . ... . .. . .... ... .. . . . . . .. ..... ... . ..... ... . ..... . ... . .. .... .... .. .. . ....... ..60) (15.59) 433 .. ...... . . . ... . . .. .. .. ..... .. . ... ... . . .. . . . . . . 61/ τ = t . ... ...... . . .. . .. ... ...... .. . . . . .. .. . .. .. . ..... . . . ...... ..... . ...... .. for the given initial conditions.. .... . .... . .. . . ... ..54).. ....56) are identical to the Fourier coefﬁcients bn given by (15. . ... ... .. . .. . ..... ... ... .... . ... ..... . ... . .... .. .... . .. .

. ... . .. .66) (15. . ... . .49) and (15. . .. . . .. . . . . ··· ... . . .. . .61) (15. . .12 We now make the the substitutions yn = π l n... . . ... .. .. . . .. .50) for the Fourier coefﬁcients... ...67) (15. 0 .. .. . .. . . .64) (15. . .. .. .62) where c0 = a0 . ... .65) has the form f ( x ) = ∑ F ( yn ) ∆ yn n= 0 ∞ ··· 4 3 2 ∞ (15. .. . .. . .. .48) for arbitrary interval 2l. . ..... ... . . . .. . . . . Fourier analysis The inﬁnite interval The Fourier series (15.68) n 3c 2c 1c 0c A sum like (15.. . . . .. . . . .63) Figure 15. .... . . . . .. . ... . . f ( x ) = ∑ cn ∆ n n= 0 (15.... . . . ∆ yn = π l ∆ n.. . . . . . . . . . .. . . . . . .. .. . . . ... . . .. ..... . u( yn ) = v( yn ) = lan π = 1 π Z +l −l +l f ( x ) cos xyn dx lbn π = 1 π Z −l f ( x ) sin xyn dx We are now in a position to let l1→1∞ and ∆yn1→10.. . .. . . . . .. ... . . . . . . . . . . . .. . Equation (15. .. .. . . .. .... . . .. . . . ... . . . .. . . . 2 cn = an cos nπ x nπ x + bn sin . .. . . . . by (15. . . . . .. . . l l n>0 . . .. .. . .. .. .. as illustrated in Figure 15. . .. .61) is then u( y ) ∞ f ( x ) = 0 + ∑ u( yn ) cos xyn + v ( yn ) sin xyn ∆ yn 2 n =1 where... ... . . . . . .. ... . . .. . . . v ( y n ) = l π bn Equation (15... . .. . . . . . .. .62) can be considered as the sum of the areas of rectangles of width ∆n1=11 and height cn. .. . 1 )n( f . . . . .. . f ( x) = can be written as f ( x ) = ∑ cn n= 0 ∞ a0 ∞ nπ x nπ x + ∑ an cos + bn sin 2 n =1 l l (15.. . . . . . . . .434 Chapter 15 Orthogonal expansions. ..12. . . .. . .. . .. .65) (15. . .. .. .. .. . .. . .. . u( yn ) = l π an . . ..

. .. . . . .. .. . ..... . . . .. . . . . l+ k+ 0 k− l− f ( x) = x . .. .. . . . . . . . . . but is sufﬁcient for most purposes. . . u( y ) = v( y ) = 1 π Z +∞ f ( x ) cos xy dx +∞ (15. . .. . . . . .. .* EXAMPLE 15. . . ... . .. . . . . .65) is then f ( x ) = Z u( y ) cos xy + v ( y ) sin xy dy 0 ∞ l → ∞ n= 0 ∑ ∞ F ( yn ) ∆yn = Z F ( y ) dy 0 ∞ (15... (ii) the Fourier integral in the limit l1→1∞ (i) The Fourier series Figure 15.... and it can therefore be represented in the interval −l1≤1x1≤1+l by the Fourier cosine series Figure 15. . .. . . .69) of f(x) is called a Fourier integral (the name Fourier transform is normally reserved for a slightly different form. . .. f ( x ) = lim Equation (15. .69) given here is only an outline... . .. . . . . ... . . .72) Find (i) the Fourier series expansion in interval −l1≤1x1≤1+l with k1<1l. . . .. ... . .. . . . .. . .6 Fourier transforms 435 so that. . for n1>10. . . . .. . .. . .. after taking the limit in (15... .. . . ..13 where a0 = an = 2 Ak and. . .. . . .. . ... .. . all problems associated with discontinuities of the function and with the existence of the limit have been ignored. .66) and (15.... . a0 ∞ nπ x + ∑ a cos 2 n =1 n l .15. . .. . . .... . . . . . . ... .4 of the Riemann integral as the limit of a sum. . . ... . .. ..... ... . . . . .. . .69) where. ... . ... .. . . . . .. . . . . .. .. . . . . . . ... by the discussion in Section 5. . .. .... . as described below). . ..13 shows that the function is an even function of x.. ) ( . . . . .. . . . . . .. . .. . . . . . . . . . . .. . ... . .67). . l l k 2 nπ x 2A nπ x 2 Ak sin ( nπ k l ) Z f ( x ) cos dx = Z cos dx = ( nπ k l ) l 0 l l l l 0 * The ‘derivation’ of (15.8 For the square wave A f ( x) = 0 for − k < x < + k otherwise (15.. . .70) −∞ 1 π Z f ( x ) sin xy dx (15. . . A xf . . ... ..71) −∞ The representation (15. . . .

69) to (15. This Fourier cosine integral representation of the function can be written in the form (15.14(a). equations (15. f ( x) = 2 Ak π ∞ Z 0 sin( ky ) cos xy dy ( ky ) (15.70).75) with u( y) given by (15.74) (the constant term.76) as required by (15.74). does not contribute).73) and. corresponding to n1=10. Thus u( y ) = 1 π Z +∞ f ( x ) cos xy dx = = 2 π ∞ Z f ( x ) cos xy dx 0 −∞ 2A π Z cos xy dx = 0 k 2 Ak sin ( ky ) π ( ky ) (15.71). we make the substitution yn1=1nπ2l. can be written in exponential form by making use of the Euler relations cos xy = 1 ixy ( e + e− ixy ).69) as ∞ f ( x ) = Z u( y ) cos xy dy 0 (15. in the limit l1→1∞. 2 sin xy = 1 ixy ( e − e− ixy ) 2i (15.436 Chapter 15 Orthogonal expansions. The exponential form The Fourier-integral representation of a function.77) . Fourier analysis The Fourier-series representation of the function in the ﬁnite interval is therefore f ( x) = 2 Ak l 1 ∞ sin( nπ k l ) nπ x +∑ cos 2 n =1 ( nπ k l ) l (ii) The Fourier integral To ﬁnd the value of this expression in the limit l1→1∞. ∆yn1=1π2l Then f ( x) = 2 Ak 1 ∞ sin( kyn ) + cos xyn ∆yn ∑ π 2 n =1 ( kyn ) (15. For the corresponding pair of Fourier transforms see Figure 15.

71) replaced by w( y ) = 1 2π Z +∞ f ( x ) e− ixy dx (15. we have u(−y)1=1u( y). f ( x) = 1 Z u( y ) − iv ( y ) eixy dy 2 −∞ +∞ ∞ ∞ 0 v (−y)1=1−v( y) =Z where w( y) = w( y ) eixy dy −∞ 1 1 u( y ) − iv ( y ) = 2 π 2 Z +∞ f ( x ) e− ixy dy −∞ .79) −∞ with equations (15. replacing y by −y in the second integral on the right.69) can then be written as f ( x) = Z +∞ w ( y ) eixy dy (15.80) −∞ EXAMPLE 15.9 that equation (15.6 Fourier transforms 437 We deﬁne the function w( y) = 1 u( y ) − iv ( y ) 2 (15. Therefore. 1 1 f ( x ) = Z u( y ) − iv ( y ) eixy dy + Z u( − y ) + iv ( − y ) eixy dy 2 0 2 −∞ From equations (15.77) into (15.78) It is shown in Example 15.70) and (15.9 Derivation of the exponential form Substitution of the Euler relations (15.70) and (15.69) gives f ( x) = = 1 Z 2 0 ∞ u( y ) ( eixy + e− ixy ) − iv ( y ) ( eixy − e− ixy ) dy ∞ 1 1 Z u( y ) − iv ( y ) eixy dy + Z u( y ) + iv ( y ) e− ixy dy 2 0 2 0 ∞ or.15.71).

. . .. . .. . . 438 Chapter 15 Orthogonal expansions. . .. ... . . . . . ... ... . . . . .. . .. . .... . . . .. .. .... . ... . .. . .. . .... . . . . ... . .. . .... . . . . . ..... ...... . . .... .. . . . . . ... .. . . . . . .. ... .. .. . .. . .... . . . . . .81) and (15. ..... .. . . . .. . . .... . .. ... . . . . .. .. . . .. . . . . . . ... ... ... . ....... .. . . . .. .... .. .. . .. ... . .. . . .. . . . . . ... . . . .. .. ... . . . ... .. . .. .. . .. .. .. .... . ... . . . . . .. .. . . . .. . . . .. . . . . .82) are called a pair of Fourier transforms.. . . . ... .. .. . . . . .... .. . . .. ... .. . .. ... . . . . . . .. ... ...79) and (15. ... .. . . .. . . . xa−e = )x( f . . .. .. . . .. . ... . . . . . . .. .. . .. . . . .. .. . .. a+ a− ) ( a Z Z −∞ +∞ −∞ +∞ g ( y ) eixy dy f ( x ) e− ixy dx exists (15.. . . .. .. ... . . . ..... . . . . . . . .. ) ( c 0 > a r f o . . . . . . .. . . ... .. . . .. . . . .... . .. .. . . . . . . .. . . . . . .. . .. .. . .. . . .. .. . ... . .. . . ...... .... .. . . . ... .. .. . . . .. .. . . . .. . ... .. .. .. . .. .. . .. . .. . . Then Equations (15. .. . . ... . . . . ... .. .. .. . . ... . .. . . . . . .. . .. .. . ... ... . . . .. . . . . . . .. . . . . . ... . . .. .. . . .. .. ... . ..... . . . .. . . . . .. .. .. . . . . .. . . . .. . . .. .. . . .. . .. ... ..14. ... . .. .. . . . .. . . . . . .. . . . .. .. . .. . . . .. .. . .. . ... . . ... .. .. .. .... . . . . . ... ... . .. .. .. .. ... .. ...... . . . .. .. . . .. ... . . a a4/ 2y− e 2 √ =) 1 y( g . .. .. . .. ... .. . ... . . . . . .. . . .. .. . ... . . .. . . . .... . . . .... ... . . . and f is called the (inverse) Fourier transform of g.. .... . . .. .. .... . .. . . .. .. . .. . .... . .. . . ... . .. . . . . . . . . . .. ... . ... ..... . . .. . . . . .. . . . .. . . .. ..2 y + 2a a π 2 √ = )y(g 1 . . .. . . . .. .. . . . . . . . . . . .. .... . ... . .. .. . ... .. ... . . . . . . .. . . .. .... .. .. . . ..... . .. . . .. . . .. . . . . . . . ..... . . . . . . .. .. . . . .. . . . . . ... .. .. . . ..... . . . .. . ... . .. . .. . . . . . . 0>x r f o . . .... .. . ... . ... . . .. ... .. .. . . . ... . ... . )b( 0 esi r eht o w r f o a < x < a− A 0 =) ( xf . . . ... . . . . .. ... . .. . . . .. . .. . . . .... . .. . . ... . . ... . ... . . . .. . .. .. .. . . .. . . .. . .. . ... . .. . . . . . . .80) are conventionally written in the more symmetrical Fourier transform pairs Two functions that are related by equations (15. . . .. .. . . ... ... .. .... .. ... . . . . . . . .. .. . . . . ... . . . . . ... ... . . .... ...... .. ... . that is.. ... . ... .. . . ..... . . ... ... .. . . . .. ... . .. . . .. .. .. . .. . . . . . .. . . . . . .. . . ... . . . . .. .. . . . . .. . . . . . . . . . . . ... .. . . . . . . ....... . . .. .. . .. .. .. . . . . ... . . ... ... . . .. .. . . . . . . ... . . . .. . .. . . . . ya π 2 ya nis aA√2 = )y(g .. . . . . .. . ..... . . . . . . . In general. . ... .. . . .. . . .. . ... .. . ... .. ... . . . . . .. . .. . . .. .. .... . . . . . .. . .. .. . ... . . . . . R e . .. . .. . . .. . . . . . .. . ... . . . ..... . . .. . . . . . . . . . . . .. .. . .. . ..... . ... . . . . . . .. . . .. . .. . . .. . . . . . . ... . . . ... . . .. . ... . . . . ... . ... . .. . . .. ... . . . .. ... . if Three of the important elementary Fourier transform pairs are illustrated in Figure 15. . . . . .. .. . . . .. . .. . .. .. ... . . . .. . .... . . . . . . . . . . . . .... . ... . . . . . .. . . . ... . . . .. . ... . ... . . ... . .. . . .. . . .. . . . . . . . ... . . . . .. . . .. . . ... . .. . .. . . ... . . .. . . . ... .. . . . . . .. . . .. Fourier analysis form by means of the substitution g ( y ) = 2 π w ( y ).. . . . .. . . . ... .. .. . ... . . . .. . ... .. .. . . . .. . . . ... . . . . .. . . .. . .. .. . .. . . . . . .. .. . . . ..... The function g is called the Fourier transform of the function f. .. .... ... .. . . . . . .82) (15.. . . . . . . . .. . . . . .. .. . ... . . . . ... . .. . . . . . . .. ... . .. .. .... . . . . . . . . . 1 g( y) = 2π f ( x) = Z −∞ +∞ | f ( x ) | dx 1 2π Figure 15. . .. .. . . .. . . . . . . . ... . .. . . . . . .. . . . . . .. .. . . .. . .81) .. . . . . . . ... . ....... ..... . . . . .. . ..... .. ... .. .. ... . . ... . the Fourier transform of a function f(x) exists if the function is piecewise continuous and if it is absolutely integrable... . . .. ... ... . . . . . . . .. . . . ... . . . . . . . .. . .. . . . ... . . . .14 0 >a .. . . . .. . .. . . .. .. . . 2xa −e = )x( f .... . .. . . .

Thus 1 g( y) = 2π − e− ( a + iy ) x ∞ = 1 1 a + iy 2 π a + iy 0 because e−(a+iy)x1=1e−axe−iyx1→10 as x1→1∞. A Fourier transformation is then a transformation of the description of a physical system from one in the space (or domain) of one of the variables of a conjugate pair to that of the other. such as the x coordinate and its conjugate momentum. The momentum to coordinate Fourier transformation is an essential tool in the analysis . the linear momentum px. The Fourier transform is complex and to ﬁnd the real part we write g( y) = Then Re g ( y ) = 1 2π a a2 + y 2 1 a − iy 1 a − iy = 2 π ( a + iy ) ( a − iy ) 2π a2 + y 2 (15. and gives the shapes of spectral lines in Fourier transform spectroscopy.14(c). Figure 15. f(x)1=1e−ax By equation (15.83) This function is called a Lorentzian. and the time–frequency (or time–energy) pair.6 Fourier transforms 439 EXAMPLE 15. 1 g( y) = 2π = 1 2π for x1>10 and a1>10 Z Z +∞ f ( x )e +∞ − ixy −∞ 1 dx = 2π ∞ Z e− ax e− ixy dx 0 e− ( a + iy ) x dx 0 The presence of a complex exponent does not alter the normal rule for the integration of the exponential.10 Find the Fourier transform of the exponential function. The most important of these are the coordinate–momentum pair. 0 Exercises 15–17 In applications in the physical sciences.82). the variables x and y are usually a pair of conjugate variables.15.

. . . . . . .11 Find the real part of the Fourier transform of the exponentially damped harmonic wave...... . . . .. . .. . . . . the intensity of the output signal shows exponential decay as a function of time: I(t)1=1e−t2T (15. . . ... .. . ......... and the system is observed as it relaxes to its thermodynamically stable state.. .. . .. . .. ... .. . ...... . .... . . ... . . .... Fourier analysis of the results of diffraction experiments.. .... ..... . . .. ...14.. . .. . . . . ...... . . . . . . . . ...... . .. . . .... . . .... . . ... .. . . . . .. ... Relaxation is essentially a ﬁrst-order kinetic process and in the simplest case.... .. . .. .. ... ... . ..... . . .. . ... . ... ω0 say.. .... .. . . .. with x as time t and y as frequency ω... . .84) where T is called the relaxation time. . when the molecules can undergo transitions at only one frequency. ... . .. .. . .. . ..... the observed diffraction pattern is the representation of the structure of the system in momentum space and the Fourier transformation gives the structure in ordinary (coordinate) space. ... +∞ Figure 15. . .. .. . This is an example of case (c) in Figure 15. . . ..... ..85) The width of this ‘spectral line’ is inversely proportional to the relaxation time T... .. . . because cos1bx1=1(eibx1+1e−ibx)22.. .... .440 Chapter 15 Orthogonal expansions. . . ............ ..... .. ..82).. .... ... . . . . ... . .. . .. ..10) 1 F (ω ) = Re 2π ∞ Z I (t )e− iω t dt = 0 1 2π T 1 + ω 2T 2 (15. . .. .. . . . .. .... . . .... ... . .. g( y) = 1 2 2π 1 2 2π ∞ Z 0 − ( a + ib− iy ) x e + e− ( a − ib− iy ) x dx = 1 1 + a + ib − iy a − ib − iy . The Fourier transform of the decay curve is therefore a Lorentzian curve (see Example 15. . Its position (at ω0) is determined in practice by the ﬁne structure of the output... ... . . .. .. . . . f(x)1=1e−ax1cos1bx By equation (15... . .... . ... ..15 . ... . .. Of even greater importance to the chemist is the time to frequency transformation because it forms the basis for Fourier transform magnetic resonance spectroscopy..... ... ... the molecules of a sample (or other species with excitable degrees of freedom) are excited by a short pulse of radiation. .. . .. . g( y) = 1 2π for x1>10 and a1>10 Z f ( x )e− ixy dx −∞ ∞ 1 = Z e− ax cos bx e− ixy dx 2π 0 or. . . .... . . . . ... .. .. . . as illustrated in Example 15. In such an experiment. . .. ....11. ... .. EXAMPLE 15... . . . . . . . . . .. . ... .. . . ...... ...

....... ..... .. ..... t t .. ... ... .3) of f(x) in Legendre polynomials.. .... Given the power series f(x)1=1a01+1a1x1+1a2 x 21+1-..... .... .. ... .. . ......... ........ .7 Exercises 441 The real part of this is (see Example 15. ..... .... ....... .. b) by (t.. .. ..... . . .. ... Figure 15.. ω0) in (15. (i) Find the ﬁrst three terms of the expansion of the function 0 f ( x) = x if if −1 < x < 0 0 < x <1 .... . .. ... .. .. .... .. . ..... ... ..... ..... ...... . 5.. q+ .. . .. . 1 ... . . .. ... use equation (15.15...3 3. .. ..... and the function then reduces to a Lorentzian centred at ω1=1ω0. . .... ...... .9) to ﬁnd the coefﬁcient c1 of P1(x) in the expansion (15.. ...... . ... ..7 Exercises Section 15.. .....16) in Legendre polynomials. ........ . . .10) 1 2 2π a a + a 2 + ( b − y )2 a 2 + ( b + y )2 Re g ( y ) = (15.. . .... .. .... .. Figure 15.. ....... . . ... . . . . .. .... ........16 Section 15. y. . . ... .. ... ...... .. .17 p • R θ o r • q2− r 4........... .... . ... . .... .. . .. . two-term.. .......... . . ... .. .. .. and three-term representations of f(x). 1+ 0 • q+ 1− (Figure 15... •.. .... ω.... . . . . .. .. . ... . . ... a... .. .. .. . .. ... Find the ﬁrst nonzero term in the expansion in powers of 12R of the potential at point P for the system of three charges shown in Figure 15. 15. ....86)......... . .... .. .. ....... (ii) sin1tx... .. .. . ..... ......... . .. (ii) Sketch graphs of the one-term.. . .... ......... . .. .. ... . . g (ω ) = The second term of g(ω) is in practice small compared with the ﬁrst. + .... .. ... we obtain the pair of Fourier transforms 1 2 2π T T + 1 + (ω − ω )2 T 2 1 + (ω + ω )2 T 2 0 0 f (t ) = e− t /T cos ω 0t .... . ..... . . . ... ..... ... .. . Expand in terms of Legendre polynomials (i) cos1tx. 2.86) If we replace (x.. 12T. . .. . .. ...... . .. . . . .. . ...... .. .. ....17...2 1. . Show that the coefﬁcient of P1(x) in the expansion eitx = ∑ cl Pl ( x ) l =0 ∞ 1 sin t is c11=13ij1(t) where j1 (t ) = − cos t .

. . . .. . .. .... . . . .. .... .. . . . ..... . . .. .... ... .. . . . . . . 8. .. .. .] (iii) Draw the graphs of the ﬁrst four partial sums of the series. . .. . . .. .. . .. ...... .. .. . ... . . .. ... .. .... . .. .. .... ... . .. . ..... .. . . . .. .. .. . ..... . ... .. .. ... . .. (iii) Use the series to show that π2 6 π2 =∑ =∑ ∞ ∞ 1 n 2 = 1+ n=1 1 1 1 + + + 4 9 16 1 1 1 + − + 4 9 16 ( −1) n+1 n2 12 = 1− n=1 11... . ... .. . . . ..... . . ... .. . . .... 10... . . .. .... . . . . . . −π1<1x1<1π (i) Draw the graph of the function in the interval −3π1≤1x1≤13π. ... . ... . . . .. ...... .... . .. . . . ... .. .... . . . . .. . ... .. . . ... . ... . . . . .. ... . ... .... . .. . .... .. A function with period 2π is deﬁned by f(x)1=1x2. .. . . ...... . .. . .... .... . . . . . . ...... .. .... .. . . ... ..... . .. . .. ... . .. .... ... .... . ...... . . (ii) (15...... .. . .. . . . . . . . .. .. . . ... . . .. show that the leading term in the expansion in powers of 12R of the electrostatic potential at P (in the plane of the square) is V1=1qr 224πε0R3. .. .. . . . . . .. .. .. . ...... .... . ... . . . .. . .4 7... . .. (iii) Use the series to show that π 4 1 1 1 = 1− + − + 3 5 7 [Hint: substitute a suitable value for x in the series]. .... .442 Chapter 15 Orthogonal expansions. . .. .. . ... . . ........ .... ... . .. A function with period 2π is deﬁned by f(x)1=1x. .. . ... . . • q+ ..... . .. . .. .. ... . . . . ... .. .... . .. Section 15.. . . . . . .. .. .... . . . .. . . .. .. .. . ...... . independent of orientation. ... . .......... . ... . . A periodic function with period 2π is deﬁned by 1 f ( x) = 0 if if − π Figure 15. .. . . . .. .. . .... . . .. . . . .. (ii) Find the Fourier series of the function [Hint: f(x) is an even function of x]. (ii) Find the Fourier series of the function. . .. .. . ... . . ........ ...... .. . . ... . .... . . . .... .. .. .. . .. .. . .. .. . .. .... . . .. . . . .. . . Conﬁrm the relations (i) (15. . . .. 9. .. ..... .... ... .. . .. . Show that the Fourier series of the periodic function deﬁned by (see Figure 15.. . [Hint: f(x) is an odd function of x.. ........ . . .. ... ... . ... . ...33). . . .. . ... . .. . .. .. . . . .. . ... . . ... . .. . . . . ... . .. ....... . . .... ... −π1<1x1<1π (i) Draw the graph of the function in the interval −3π1≤1x1≤13π.. .. .. .. . ...... . . . . . ..... ..18 2 π <x< π 2 2 <|x|< π (i) Draw the graph of the function in the interval −3π1≤1x1≤13π. .. ... Fourier analysis .. .... . ... .. . . . . ... . (ii) Find the Fourier series of the function.. .. . . .. . .. . . ......... . ... . ... . .. . .... . ... ..... . . .. . .. ..... . .. .18. .. .... .. .. .. . .. ... . .. ..... ... .. .... . ... . .... ... .. ... ... ... For the square system of ﬁve charges in Figure 15. . ... . . .. .. .. . ... .. . . . ... .... ...... ... .. .. ..... . . . ... .. ...... . .. .. .4) sin t f (t ) = 0 is 1 if 0 ≤ t ≤ π if − π ≤ t ≤ 0 1 2 cos 2t cos 4t cos 6t + sin t − + + + π 2 π 1⋅ 3 3⋅ 5 5⋅7 • P q +• r q • θ + • q 4− q • + 6. . .... . . ... .34) and (iii) (15. ... .... .. . . ... . ... .. .... .. . . . . ... . .. .. ...... . .. . .36).

0)1=1x(l1−1x). 13.14 b) ∞ iy Hint : change variable to t = a x + and use 2 a Z e−αt dt = 2 −∞ π α .7 Exercises 443 12. A function with period 2l is deﬁned by x(l − x ) f ( x) = − x ( l + x ) if 0<x<l if − l < x < 0 (i) Draw the graph of the function in the interval −l1≤1x1≤1l. if a < x < b otherwise 16. t)1=10 and initial condition T(x. Show that the Fourier transform of e−ax (a1>10) is 1 − y 2 /4 a e 2a ( Figure 15. e−a|x| (a1>10) 2 17. t)1=1T(l.15. 15. (ii) Find the Fourier series of the function. f ( x ) = 0. and that decreases exponentially with time (See Exercise 17 of Chapter 14). Section 15. Find the solution of the diffusion equation ∂T ∂ 2T =D 2 ∂t ∂x in the interval 01≤1x1≤1l that satisﬁes the boundary conditions T(0. Section 15.6 Find the Fourier transform: 1. (ii) Find the Fourier series of the function.5 14. A function with period 2l is deﬁned by x(l − x ) f ( x) = x(l + x ) if 0<x<l if − l < x < 0 (i) Draw the graph of the function in the interval −l1≤1x1≤1l.

. . In (a). .. Persian.... . and displacement. .. He is best known for the reformulation and generalization of the mechanics of Newton. .. .. Such quantities. . .. .. ... .. .1 Figure 16.. . is reputed to have been ﬂuent in Latin. . .. and Lagrange that became important in the formulation of statistical and quantum mechanics... . temperature. and 121m. ... . . A vector is represented graphically by a directed line segment. ....... . . 2731K. in (b).... . together with an arrowhead.. . .. the modern European languages.. . . .. .. magnitude.. .. . and others at the age of ten... . .. .... ... . . ... Some of these uses of vectors.. ... . . are called scalar quantities and obey the rules of the algebra of real numbers. .. . . . . ... . . .. the vector a is represented by an arrow....... . . A vector of zero length is call the null vector 0... .... .. . To qualify as vectors these quantities must obey the rules of vector algebra.. .. . . . . . . . . . . .. .... ... .. . a . . The length. gives the direction. magnetic ﬁeld. in appropriate units.. . . . require both magnitude and direction for their speciﬁcation. .. important in molecular dynamics... . . .... . .... . . A vector of unit length is called a unit vector.. .. that is. . 1 Vector algebra has its origins in the algebra of quaternions discovered by Hamilton in 1843 and in Grassmann’s theory of n-dimensional vector spaces of 1844. and two ways of representing it in print. .. and whose orientation in space. . .. . .. . ﬂuid dynamics. Dublin. .. . 31kg.. called vectors... ..1 shows two graphical representations of the same vector. and no direction is deﬁned in this case. . electromagnetic theory.. Other physical quantities. born in Dublin... in mechanics. the speed being the magnitude of the velocity vector. Arabic. . .. ..... a ) a( .. The modern notation for vectors in three dimensions is due to Gibbs.. ... .. b− →a b = )b( a .. . . velocity is speed in a given direction.. . .. Hebrew.. .. .... .. ... William Rowan Hamilton (1805–1865).. .. and distance have values that are speciﬁed by single real numbers.1 Concepts Physical quantities such as mass.... .. A and B are the initial and terminal points of the vector.. . electric ﬁeld.. . are discussed in examples throughout this chapter. . a segment of line whose length is the magnitude of the vector. . .. . .. .. ... . .. ..... .... and theoretical chemistry.1 Vector notation and vector algebra are important for the formulation and solution of physical problems in three dimensions. in 1823 and became Astronomer Royal of Ireland and Professor of Astronomy in 1827 without taking his degree.... . ... and engineering design... . and the notation AB is particularly useful when the vector is a displacement in space. He entered Trinity College. ... Sanskrit. Figure 16. . for example. . . or modulus of a is written as | a | or simply as a (a scalar). . .... having magnitude only. Greek.. Other examples are force. .. . Euler. For example.. ... spectroscopy... . .. . in appropriate units..16 Vectors 16.. .. . .

. . ... . a • z . ... . .. ... . . . . .1) if they have the same length and the same direction. . ... .. . but one whose initial point is bound to a ﬁxed point. . ... ........ . ... ... ... b a • a . . . .. ... .... .. .. ... .. . ..... . . . . .. . .... .. . . . ... .. . Bv . . . .. . . . . . . . . . . . .. .. ..... . .. . .. . that is. . .... .. . .. . ... . . . .... . ... . . . . . .. ... .. . . . ... . . . .. ... ... .... . .. .. . . . .. . . . . . . . . . .. .. . .. . .. ... .... .. . ... . . . .. . . . . . .. ... ... . . . . . . . .. ... along curve PQ. AF .. . . . . . .. .. . . . . . . . .. .. .. . . . . ... .... . .. . .. ... ..... . the origin of a coordinate system. ..2 At each point along the curve PQ. . . .. . .. . ... . . . ... .. . . . . . . . Vector addition The sum.. ... . . . Figure (b) shows the velocities at points A and B.. .. . . . .. .... .. .. •b Av )b( .. . . .. . .. .2) y q .2 Vector algebra 445 Figures 16. . ... c . the position of point A on the curve is given by a = OA whose initial point is at the origin of a coordinate system and whose terminal point is at A. . .. .. . . with a1+1b1=1b1+1a (16.. . . ... . . In Figure (a). .. .. . .. . . ... .. . . . . .. .. .. . .. . . ..... ... . ..16... . ..... .. .. ... .. . . .. . .. . .. . . . . . . ..2 Vector algebra Equality Two vectors a and b are equal. . . .. . ... 16.. ... .. ..... .. .. . . ..... . .. . .. . ... . .. . . .... . Figure (b) also shows that addition is commutative. . .. ... .. .... .... . ..... . . Thus two separated bodies moving at the same speed and in the same direction have equal velocities. . . . . ...... . .. .. .... .. . . . BF .. . .. . .. . .. . . .. ..... .. .... . ... . .2 (a) and (b) illustrate several types of vector associated with the motion of a body in space. .... . . . . . . . . . . .. .. . . . . . .. .. . . .. .. ....... .. that is. .. . ... .. .. ... . . .. z . . ... .. . .. .. . . .. .... . . . . .. .. . .. . .. ... . . . . . ...... . .. . ..... . ... ... . . . . ... . ...... .. .... .. . .. . ... . .. . .. . . . When the body has moved from A to B the position vector has changed from a to b.. . . .. .... . ...... . . . .... . . . . . . . .. ... . We will see that vector algebra gives c1=1b1−1a.... . ... .. . ..... . o p x y q . .. .. .. . o p x . . .. . . . . . . . ... .. . .... b • ) a( .. . .. .. ... a1=1b (16. ... ... .. .. . ...... ... . ... .. . The vector a is called a position vector. .. . . . .. . . . .. .. ... . . ... . Figure 16. . . . .. and its value (magnitude and direction) changes as the body moves along the curve. . .. .. . . . .... . .. a1+1b of vectors a and b is deﬁned geometrically in Figure 16. .. . . .... . .... .. . ... . . . ... . . .. . . .. ... .. . .. . . . ..... . .. ..... ... .. the body has velocity v along the direction of motion at that point. . .. . .. . .... .. .. the displacement A to B is equal to the change of position vector. ... .. ... .. . . . . . .. . . . . the direction of v at every point is tangential to the curve. . .. . .... . .. .. .. . ... . . . .. .. . . . .. . . . . . . .. . . .. . . .... .. .. . . ... ... such as position vectors.... . . . . In (a) the vectors are positioned with the initial point of b coincident with the terminal point of a.... . .. The sum a1+1b is then equal to the vector drawn from the initial point of a to the terminal point of b... under the inﬂuence of external forces. .. .3... . . .... ... ... ...... . ...... . or resultant. . . .. . . .. ... . . . .. . ... . We note that a position vector is a type of displacement vector... .. . . .. ... .... . . . .... . . . . and also forces acting on the body at these points. .. . .... . .. and the displacement A to B is given by the vector c = AB. . . . ... . . ... The initial points of equal vectors need coincide only if they are bound vectors. .... ...

. . . .. .... . . . .. .. . .. . ... . . .. . ... . . ... .... . . . . . .. .... . ... . ... . .. . . ..4) b .. . . .. . where 0 is the null vector. . . .... . . .. . . . ... .. ... . . ... . . . .. a− b− a b−a )b( a b .... .. . .. . . .. ... . ... ... . .. .... .. . . ... . . .. . . .. ... ..... .... . .. ... . . . . . . . . . ... . .. . . . .. .. . . .. . .. . .. . ........ . . . .. . . . . .. .. . . . . . . .. . . . . ... ... . ... . . .. ... . ... .. . .. . . . . .. | b |1=1|a |. ... ...... .... .... . . . . . . . .. . ..... . . . ... ... .. .. .. .. . .. .. .. .. . . .. . . . . . ... .. .. . .. .. ... .. . ... .. . . . .. . ... .... . . . ... . .. . . .. . .. In general. . ...... ..... . . . . . .. . . . ... .. .. . .. . . ........ . . ...... .3) Figure 16. .. . .... . ... . . ... . . .. .. ...... .. . ... . .. . .. . . . .. . . . .. . . ... ........ . . . . .. . . . .. . .. but also appears in a work attributed to Aristotle (384–322 BC). . . .. .... . .. . .. . . . . . .. . .. . .... . ...... . .. ....... .... ... . . . .. . . ... . . the parallelogram law is then called the ‘parallelogram of forces’. . . . + .. .. .. . . .. . . . . . . . .. .. ... . .. then a1+1b is the total (or resultant) force.... . . .. ... .. . . . . ... . . ..... . .. .. . . ... . . . .. . . .. . .. .. . .. . . . .. . .. .. . . .... . . . . . . .... .. . . .. . . ... ... .. .. . . . .. . . . .. . ... .. ... . . .. . .... . the product of a scalar (number) c and a vector a is written as ca. . . . . .. ... . . .. . .... .. .. The subtraction of vectors is then deﬁned by (Figure 16. ..... . . .... . .. . . . .. . . ... .. .. . . . .. .. . . . .. ..... ... . .. . . . a . .. . . ... . . .. . . . . . .... .... ... . ... .. .. . . .. . . . .4) ... . .. . . ... .... ... .. .. . . .. . . . . . . . .. .. . . .. . . .... . and the direction is not deﬁned. ... . .. . . . .... .. ...... . . . . . .. . . . .. . . ... .. . .. ... ... . . . .. ..5 Figure 16. .. . .. . and vector addition is said to obey the parallelogram law. . . ... . .. . . . . . . ... . .. . .. .. ....... . . . . . . has the same direction as a if c is positive. If a and b are displacements of a body then their sum a1+1b is the total displacement.. . . . . . . . .. . . .. . . . ... .. . ..... .. . ... ... . .. .... . .... ... . .. .. . ... . . . .. . . ... .4 (16. .. .. . .. .. .. ... .. .... . .. .. . .. . . .. ... . .... .. .. .... . .. . . . . .. ... . ... .. ... ........ . . . .. .... ... . ..... .. . . . . . ..... . . . .. .. . .. .. .. .. . . . a2 . . .. b a a 446 Chapter 16 Vectors If a1+1b1=10.. . . . . ... . . . ... .. .... . . . . .. .. . ....... . . .. . . . .. . .3 a= ˆ a |a| (16... . . . . but points in the opposite direction.. ... .. then b1=1−a is a vector that has the same length as a. .... . Scalar multiplication a b a Subtraction In this case a and b deﬁne the pairs of parallel sides of a parallelogram. . .. .... . . . .. . . . . .. . .. .. . . . ............ . .. .. .. . . .. .... ... .. . ... . . ... . It has c times the length of a. .. . . ... b .... . . . .. ... . ... . . . ... .. . . .. . . . . . ... . ... . . .. . ... . . .. ... .. If a and b are two forces acting on a body.. ... ... . . . .. . . . . .. . ... . . ....... .... .. . . . .. . . . . .. . . . . . . . . . . . ... . . . . ... . . .. . .... . . . . .. . . .. . .... .. . .. ... ...... .. .. . . .. . . .... . . . . . . .. . . ... . . )a ( The parallelogram of velocities was described by Heron of Alexandria (1st century AD) in his Mechanics. ... . .... .. .. .. . . . . .. . ... . .. . . ... . . . . . . .. . ... ... .. . .. . .. ... .. . .. . . . . . . . ... . . . ... . .... . . . ..... . . . ..... . .... .... ...... ... .. . . . . .. . .. . . .. . . ... .. . . ... .. . . .. . .. .a 2 1 − . ... ... . .... . . . ... .... A vector a divided by its length | a | is the unit vector â that has the same direction as a: The vector a1+1a1=12a has twice the length of a and has the same direction. . .. . . . . .. . .. . . .. ... .. . . + .. .. .. .. . .. . . ... . . ... . . . . . . .. . . . .. .... . .. . . . .. .. . . . . . ...5). ... . .. ... . ... . .. .. . ... .. . . . . .. . . .2 2 a1+1(−b)1=1a1−1b Figure 16.. and has opposite direction if c is negative (Figure 16. . . . . ... . . ... . . .. . . . ... . . . . .. . ... . . .... .... . . ... . .. . If c1=10 then ca1=10.. . .. . b+a b Unit vectors are often used to deﬁne direction.. . . . . ... . ... . . .... . ... .. . . . . .. .. .. . . . . ... ... . .... .. . . ...... . ..

. .......... .. . . .. .. and C′ is the midpoint of AB. .. . .. . .. . ... . ..... ... .. .. . .... . . .. ... . . ....... .. .. ..... .. . .. . ... .. ... . .. ..... .. .. . ... .. . C is the midpoint of diagonal OD. . .. .... ... . ... . . ... ... . . .. ... ... ... . .... c x • b b In Figure 16.. . . ... . . .. .. . . . EXAMPLE 16. .... .. . . .. . . . .. .. .... .. ... . ... . .. . ... . ...... . . . .. . . ........... . .. ......... . ... . . . . . ... ... ..... . . . ... . . . .. .. .. ... .... .. the vertices A and B have positions (position vectors) a and b relative to vertex O (with null position vector 0).. .... .. . ... ... .. . . .. ..... . .. . .. ..... .. . . .6 16. ... .. . .... .. . .. . .. . . ..... .. . . ... .. . ... ... . . ... . ... ... . . ... .. . ... .. . . . . .... .. . . . . . Therefore a a o The midpoints C and C′ therefore coincide.. . . . ..... . . . . . .... ... . . .. .... .. .. . .. . Similarly for the position with respect to the two other vertices.... ... .. . . . . .. ... . . ..... ......... .. . . .... . .. .... . .... ..... .. . . .... ..... . . .... . ...... . .. ... .. . .. . . . .... ... ... .. . .. ... ... . . .. .. . . . . It is shown in Example 16. ... . .. .. . .. .. . . .. .. . b In Figure 16. ... . . . ... .. .. .. ..... . . .... . ... . ..... . . .. and this is the centroid of the triangle. . ... . . .... . . Then EXAMPLE 16. ... ..... ....2 Show that the mean of the position vectors of the vertices of a triangle is the position vector of the centroid of the triangle.. . . ... . .. so that X lies at the point of intersection of these lines... .. . . .. . . . ..... .. . ..... . . . .. . . . ..... ..... . ....... . . ..... . .. .. . ... . .. . . .. . ... . . . . ... . . ... . . . . .. ...... .. .. ... . . ... .. . ... .. .. . .. .. ... .. . ...... .. .. . . .. . .. .. .. .1 that if C is the mid point of AB then OC = ( a + b ) 2.. . ..... . ..... . . .. ... .. . .... . . .. .... .... . .. . .. .. ... . .. ... .d . .. . .. .... .. . . . . .. .... .. ... . . . .. .. . ... . . ... . .. . ... .. ... . . .. . .. . .... ... . . ... .. .. . ..... . ..... . .. . . .. . ... . .. .... . .. . .. ... . . .. . .... .. . .. . ... .. . . .. ......... . ... .... . . . . .. . .. . . .. . .. ... ... ... .... . . . . .. . .. . . .. .... . . .. . . . .... . . .... . . .. .. . .. .... .6. . . . . . .. . .... .. .. . .. . . The mean of the position vectors of the vertices is therefore a a o 0 Exercise 1 and the mean lies on the line joining the vertex O to the midpoint of the opposite side......... . .. .. .. ... . . . .. ... ... .... . . ... .. ..... ... .. .... ... .. ... . ..... ... . . .. ... . .. ... .. . .. ..... .. . .... . ... .. ... .. . . ... .. . ... . . . . .. . . . ... . . . . .. .. .... .. . . .. . . ... . ... .. .. .... .. . .... ... . .... Also OC′ = OA + AC′ = a + 1 1 1 OC = OD = (OA + AD) = ( a + b ) 2 2 2 1 AB 2 c• • c b But AB = b − a. ........ . . . ... .. ..... . . ... . ...7 447 . .. ...... .. .... ..7... . . Therefore 1 1 1 OX = (OO + OA + OB) = ( 0 + a + b ) = ( a + b ) 3 3 3 1 1 OC′ = a + (b − a ) = ( a + b ) = OC 2 2 Figure 16. . . ..1 Show that the diagonals of a parallelogram bisect each other. ..... .. . . . . ... . . .... . .. . . . ..2 Vector algebra OX = 2 OC 3 Figure 16.. .. .. .... ... . . ... . . . .. ...... . .. ... . .. ... . . . . .... ... ... .. .. . .. . .. . . ... . .. ........ . . .. . . . . .

. .. .... .. . ... .. .. .. x i . .. . ... . .. . . . . ... .. . .. . .... . . . . ... . .11) can be speciﬁed by its components.. .. . . . .. . . .. . .. .... . . .. .. .. .. . . . . . . ... .. .. . . . .... . . .. . . . . .. . .. . . .. . .... ... . a . . . . .. .. . . .. o . .. . .. . ...... .. . ... .... We write the vector in terms of its cartesian components as The concept of component is essential for the practical use of vectors for the solution of physical problems in three dimensions.. . . 2x 1x − 2x = xa . ...... ... . .. . .. . ... . . . . . . ... . 2y . .. . .. ... . . .. . Thus the length of the vector is | a | = ax + a 2 . . .. ...... .. .. . . ...... . . . . . ... . y2). . .. . . . . . . . ... . . . . . ... . .... ... . . . ..... .. . o . .. ... . p . . ... . . .. ... . . . ..... .. . . ...... . .. . . . . .. . . Let the initial and terminal points of a in the xy-plane be (x1.. . . . . . . . . We write 2 specify the vector uniquely. .. . ... . .. . .. . .. ... . . . . . . . a vector in three dimensions (Figure 16... . .. . .. j....... . .7) o 1y .. . .. .. . . . . .. .. . ... . .. . . .. .. . .. as shown in Figure 16.. . .. ... These two components are sufﬁcient to Also.. . ... . .. .. .. . ... .. . . . . ... . . . . ... ...... . . .. . ay and az. . . ... .. . . . .. . .. Components of vectors Figure 16... . . .. . and k (k is the unit vector in the z-direction). .. .. . ... .. . . ... . ......8 the component of a along the direction OP is the length 16. . ... .. ... . . .. . . .. . .. . as in Figure Chapter 16 Vectors More generally.. ... . .. .. ... . .. a 1y − 2y = ya . . .. . and the component in the y-direction is ay1=1y21−1y1. . ..... . . . ... . . . . .. . . .6) (16. . . .. ... . . az)1=1axi1+1ay j1+1az k a1=1axi1+1ay j a1=1(ax.. .. . . . . .. .. . . . y . . . . . ... .. . .9 ON1=1|a |1cos1θ a1=1(ax... . . . .. .... . . . . . . . The component of a vector in a given direction is the length of its projection in that direction. .8) (16.. .. .. . . .. .. . . ... . . . . . ... ... . .. .. .... .. . . . . . . . y p n θ . . . .. . . ...... ... .. ... ... .. .. .. .. . . . . . .. . .. ... ... ax.... .....9. . . .. .... . . . .. . .. . ... . . . ... .. . .. . . .. ...... . . .. . .. ... ...... .. ...... OQ = ax i in the x-direction (ax times the unit vector i in the x-direction) and QP = a y j in the y-direction.. .. . . . . ... . . . . . . y1) and (x2.. . ... .. ... . . . .. . . . . . in the three cartesian directions i. .. . .. . .. .. ..... ... ... . . . . ... . . . . ..... ... . . .. .... ... . . . . . . . .. ..5) Figure 16... . .. ......... . .. . .10 1x Figure 16. .. ... ... . . .3 16. . . . . and the y direction is given by the slope ay 2ax.. .. ... . . .. .. . ... .. . . ... .. .. . . . . .. . . .. .. .. . . .. .. .. ..... .. .. .. .. . . . .. . . ...10. . ..... . .. . . . ... . . . . . . . . ... In Figure 16. . .. . j . . . ... . ..... ... .. . .. .. . . ... .. .. . . . . q j ya i xa x ... .. .. . . . . . . . . . . ... . .... .. . . . .. .. . . . . . .... . . if the x and y directions are described by the unit vectors i and j..... ... .. . . . . . . . The (cartesian) component of the vector in the x-direction is ax1=1x21−1x1.. . . ... . . .. . .. . . . .. .. .. . . ..8 (16. . . . . . ... .. .. ... . . .. .. . ... . . .. . . . .. . . . .. .. ... . . .. . .. ... ... . . .. .. .. . .... ..... . . . .. ..448 We consider ﬁrst the simpler case of vectors in a plane. .. . .. . .. . ... ay) (16.. .... . .. ...... . .. .. . .. .. . .. .... . . a ... then a = OP can be expressed as the sum of the two vectors. . .... . .. . . ... .. ... . . . .. . . . .. . . . . . .. .. . .. . . ay . . ... .. ... . . . ... . ... ... .. .... .. . .. . . .... .. . . ..... .. . . . .

.. ... . .. . . .. . . . . . . . . . ... ... .. . . The vectors are equal when their corresponding components are equal. .. . . . . .. ... . . ... . . ... . . ... . . . . .... .. .. .. ... . .. . . . .... ... . . .. . . . . . . . . . .. . . . .. . . . . . .. ...... . ..... . . .. ...... . ......... . . . b (ii) Addition. . .. ... . . . ........... ... . . .. . . . ... . . . . . .. . .. ..... . .. . . . . .. . . . . ... .... . . . .. ....... . ... .. .. . . . .. . . . . . . .. .. .. . . . .... . . . .. . . . They have the form a1+1bi1+1cj1+1dk. .. ...3 Components of vectors (16. . .... . .. .. .. .. . . ..... .. . . ... . .. .. ... .. .. . . .... . . . .. . . ... . . . . . .... . ... .. . . . . ... .. . . . . . . . . . . ..... . . .. .. .. . . .. .. . . .. . . . . . . . . .. . . az) a1+1b1=1(ax1+1bx.. .. . . . . ... . . . . .. .. . . .. . .. . .. .. .. . .. . .. . . . . . . .. . .. . . . .. . ...12 for vectors in a plane. . ... . . . . . .... . .... . . .. ... . . .. .. .. . ..... .. ..... .. .. . ... ... . ...... . . .. .. . .. . .. .. .. . . . ....... . . . . . a Monday..3 We consider the vectors 3 Hamilton discovered quaternions in his search for an algebra of 3-dimensional complex numbers. . The sum a1+1b is obtained by adding the corresponding components. .. . . . . .. . .. ... . . . . . . . ... . . ... .. .. .. . . ... . . ... . . .... . ... . . .. .. .... .. .. . .. .. .. . . ... . . .. .. . . . .. . .. .. .. . . .. . . . .. .. (i) Equality. . a1=1(ax. .. . ... .. .. . .. . . . ....12 xa a ya yb b+a y Figure 16. . ... . . .. . . . .... ... .. . . .. . .... ... . . . ... .... .. . ... .. . .. . . . ... .... . . . . . . . . . .... . ..... . ... .. . .. . . and it is in this way that vectors are used for the solution of physical problems. .. ... . . ..... .. .... . . .. . . . ... .. . .... .. .. .. .. .... . . . . . ... . . . . .. . . .... .. . .. ... . . . .. ... . . ay . . .. . . . .... .. .... . ... .... . . . ... .. .. . . . ij1=1k1=1−ji.. . . .. . . .... ... . . . .. . . .. . .. . . . ... .. . .. .. .. .. . . . .. . .. .. . . . .. . ... .. . .. . . .. .. . ... . .. . . . ... . . .. . . .. . .... .. . . ... .. . . ...... . ... .. . . .. . .... . . . . .. .... ....... . . ... . .. . . . . ... .. . .. . . .. .. . ... . . .. .. . . .. .. .. ... . .... . . . . . . . .. . . ... . ... . .. . . . .. .. .. . . . ........ ... . . . .. . . . . . . . . . . . ... . . . ... . ... .. ..... . .10) (16.... .. . ... . .. . . . .. . ... . . .. .... . . . . by . . . ... ..... . ...... . . .. .. . . . . . . . ..... .. . . . .. . .... .... ..9) 449 x i xa k za z .. .. ..... . . .. . . .. . ..... . with rules of combination i 21+1j 21+1k 21=1−1..... ay1+1by. . . . . . . . . . . ... . ... . . . . . .. . . . ... ... . . . . .. . . .. . .. . ... . . . .. .. .... . . . .. . . . . . ... j ya . . .. .. .. .. . . ... ... . .. . . ... . .. . . x xb . .. . . .. . . . .. . . .. and he carved them on a stone on Brougham Bridge. . . . ... . .. bz) 16. . . . .. ... . .. . ... .. .. . ... ... . . ... . . .. .... .. . ... . . . .. . .. ... . ... . . . . . ...... . .... . . . .. . .. . ..The rules of vector algebra can then be formulated in terms of vector components.. . .. . . .... .. ... . . . . . . a a1=1b if ax1=1bx. . .. . .11 ay1=1by and az1=1bz b1=1(bx .. .. .. .... ..... az1+1bz) Figure 16. .. ...... . ... . .. . .. . .. ...... .. ... jk1=1i1=1−kj. . . ... .... . . .. . .. . .. .. . . . . .... .. . ... .. . . . . .. . .. . . . .. .. . ... .. . .. . . . .. . ... . . .. .. . . . . ... .. .. ... . ... .... .. . . . .... . .. ... ... . ... .. ... .. . . . . . . .. . .. .. ... .. . . ... .. y This is illustrated in Figure 16... . . . . ... .. . .... . . . . . . . .. Hamilton called a the scalar part and bi1+1cj1+1dk the vector part. . . . . ki1=1j1=1−ik.... ... . . . .. . .. .. ... . He discovered the rules whilst walking with his wife along the Royal Canal on 16 October 1843. .. . . ... ... . .. ... . .... . . .. . . .. .. . . ..... . . . . . . . .... ... . . .. . . . .. . ... . . . ... .... . .. .. ... ...... . ........ . . . . . . . .. .. .

. 3). 0). 0)1=1(−1. The resolution into the component equations is often necessary only when the solution of a problem needs to be completed by the insertion of numerical values for the components. −223) 0 Exercises 2–5 EXAMPLE 16. the three scalar equations are expressed by the single vector equation. −2)1−1(2. (ii) What is the length of a? (iii) Find the unit vector parallel to a. The product ca of the scalar c and the vector a is obtained by multiplying each component of a by c. −2). (i) a1=1q1−1p1=1(1. −2) (ii) | a | = ( −1)2 + 22 + ( −2)2 = 3 (iii) â1=1a2|a |1=1(−123. −2. 1). a vector equation is equivalent to three simultaneous scalar equations. 2 2 2 (b) | d | = d x + d y + d z = 22 + ( −2)2 + 32 = 17 0 Exercises 6–10 Equation (16. ca1=1(cax. 3. caz) (16. one for each component. 3.3 (i) Find the vector a = PQ whose initial point P is p1=1(2. 2. 2. ﬁnd (a) d1=12a1+13b1−1c and (b) | d | . 1. Vector algebra then provides a powerful method for the formulation and solution of physical problems that involve vector quantities. dy. b1=1(1. for vectors in three dimensions. −1).4 Given a1=1(2. 3. and c1=1(5. (a) If d1=1(dx. 223.450 Chapter 16 Vectors (iii) Scalar multiplication. dz) then dx1=12ax1+13bx1−1cx1=1(21×121+131×111−15)1=12 dy1=12ay1+13by1−1cy1=1(21×131+131×1(−2)1−12)1=1−2 dz1=12az1+13bz1−1cz1=1(21×111+131×101−1(−1))1=13 and d1=1(2. Conversely. −2.11) EXAMPLE 16. 1. cay .4 show that. 0) and whose terminal point Q is q1=1(1.9) for the equality of vectors and Example 16.

. . .. ... .. . ... . .... the position of charge qi with respect to some point R is This quantity depends on the position of the reference point if the total charge 2r . .. . . r .. . . .. .... ... .. . .. ... . •.. . . . .. with −q at r1 and q at r2. . ... ... .. ... . ... ... . .. . . . . .. .. . . .. . . . The vector equation (16. .. ... . . .. .. .. .... ... .. . ... . ..40) for the one-dimensional case). . .. .. . . . .. . . .. deﬁnes an electric dipole with vector dipole moment q − . ... .. . . . .. . •... . . .13) (16. .... ... .... . . .5 The centre of mass of a system of N masses. .. . .12) therefore corresponds to three ordinary (scalar) equations. . ... . .. ..... . .. . . ....1y. q1 at r1. . .. . .. . R= N . .. .. . . . ... ... ..... . ... ... ... .. Nr .. . . . . .1 m . .... . . EXAMPLE 16.. . . . .. .. .. ... ......... .. . . ..... . . . .. .. . . .....12) 451 .. ... ... ... .. . ... 2m •.. o .. ..... q2 at r2..... . . .. . .. ..... . .. .. . . ..... . . . .... ..... . . . ... ... .. . .. . . .. .... ... . . ... ..... . ... . . ... . . . .. .. . ... .. . . . . .... . . a system of N charges. . . mN at rN (Figure 16. . .. . 1. . Z ): ..... . . ..3 Components of vectors Figure 16. . . .... . . . . . .13 (16.. .... . . . .... .. ... . . .. . ..... 2 .. .. . .... and µ (R)1=1µ (0) only if Q1=10. . . . . .. .... . .... .... .. ... one for each coordinate of R1=1(X. . ... .. . .. . .1z). ... . ....... .. . .... .. . . . ... .. .. . .. .. .... ... . . ...... .. . Y.. .16) (16. .. .. . . . . . . ... ... . .. .. . . . .. ....... ... ... .. .. . . . ... . .. .... . . Nm • . .. . . . . .. . .. .. . . ..... . .. . .. . . .. .... .. . .. ... .... . . . . .. . .... m1 with position vector r1 (‘at position r1’). .. . . . .. . . . . . .. . r . r EXAMPLE 16. .. . .. . .. . . .. . .. .. .. .. . . . .. . . . ... . . . ...... ... . ... The cartesian components of a position vector r of a point are the cartesian coordinates of the point...13) is where M = ∑ mi is the total mass. . .. ... . . .. . . .... .. . .... . .. r1=1(x. . . . ... . . . q •. . . m2 at r2. . .. . . . ... ..... ... .. .... .. .. .. .. . .14 Figure 16. ... . . .. . . .. . . .. .15) (16.. .. . . . ...... .. ... .... . . .. . =.. . . . . .. . . ..... ... . ... . .. . . .. ..... .. ...... . and the dipole moment of the system of charges with respect to R is µ ( R ) = ∑ qi ( ri − R ) = ∑ qi ri − ∑ qi R = µ ( 0) − Q R µ = ( q1r1 + q2 r2 + N i =1 + qN rN ) = ∑ qi ri i =1 N i =1 N i =1 N Z= ∑ mi ri i =1 N 1 M ∑ mi zi i =1 N 16... ..... .. . . ... . 1 r i =1 . . . .. .. . . qN at rN has dipole moment with respect to the origin O as point of reference.. . . . . . .. . . .. . . ... .. .. .... .. . . . . .. . ri1−1R... . .. . .. The system of two charges shown in Figure 16. ... ... ... .. . . . .. .. . . . .. . . . .. .. ..... .. ... ... .... . . .... ..14) o More generally...... . .. . ..... . . ..6 Dipole moments µ1=1−qr11+1qr21=1q(r21−1r1)1=1qr X= 1 ( m r + m2 r2 + M 11 1 M ∑ mi xi i =1 N Y= 1 M + mN rN ) = ∑ mi yi i =1 N 1 M (16.. . .. .... . . . . . .. . . . . . .. . . ... . . ...... Thus. . ... . ..14. .. ... .. .... =. ... . . . ..... . . .... .. ..... .. . ... ....... . .... . . ... ..... .. Q = ∑ qi is not zero. .... . . .. . . . .. . .. . . . . ... . . . .. . . ... . 0 Exercise 11 (see equation (5.. . . . .. .... .. . ... .. . .. . . . . . . .... .

µi1=1kri (i1=11. k1=1(0. 0) =10 0 Exercise 12 Base vectors The cartesian unit vectors i. 0.7 Given a1=12i1+13j1+1k and b1=1i1−12j. 2. with r11=1(a. For example. 1.18) The rules (16. az) =1(ax. 4). and z directions and have components. 0). y. −a. i1=1(1. az) (16. a. 0. 0. 0. 1) (16. . a.19) EXAMPLE 16. µ2. with the carbon at the centre. µN . and the dipole moment of each bond lies along the direction of the bond. the CH bondlength.11) then conﬁrm that every vector in the three-dimensional space can be expressed as a linear combination of these three base vectors: a1=1axi1+1ay j1+1az k =1ax(1. In some cases of high symmetry. =. 1.17) The total dipole moment of a molecule is sometimes interpreted as the (vector) sum of ‘bond moments’. ay . (ii) a vector perpendicular to b. j1=1(0. 3. and is therefore a multiple of the bond vector. a dipole moment is associated with each bond. a1−1a1+1a1−1a. a). ﬁnd (i) d1=12a1+13b. If one of the vertices is placed at the ‘111–position’. 0)1+1(0. −a). the positions of the other vertices are r21=1(a. 1) =1(ax.452 Chapter 16 Vectors A system of electric dipoles with moments µ1. 0)1+1ay(0. The length of each of these bond vectors is 3a. −a). a). 0). a1−1a1−1a1+1a) =1k(0. 0)1+1az(0. 0. 0. ay . 0)1+1(0. the methane molecule in its stable state has its four hydrogens at the vertices of a regular tetrahedron. that is. −a. r31=1(−a. these bond moments may cancel to give zero total dipole moment. has total dipole moment µ = µ1 + µ 2 + + µ N = ∑ µi i =1 N (16.10) and (16. 0. and r41=1(−a. The total dipole moment is then µ1=1µ11+1µ21+1µ31+1µ4 =1k(r11+1r21+1r31+1r4) =1k(a1+1a1−1a1−1a. (iii) a vector perpendicular to d. j and k lie along the x.

. . .. .. . ..... . ..... . . . ...... ..... .. .. .. . . . . . (iii) Vector d lies in the xz-plane. . . . .. . . ...... •b • .... ..21) (16. .. ... . . .. .. ... .... . ... . . . .. . . . . . . a • • • . . . . . .. .. . . .. . ... .. ..... .. . In Figure 16.. .... . . .. .. . . .. .. . ... . . and uc are the components of u along the directions of the base vectors. ...... .. . ...... .. .. . . . ...... . • • (ii) Vector b1=1i1−12j lies in the xy-plane. . .. .. . ∆ .... . ... .. .. . . . .... .. .15 for a planar lattice...... In this case a. . ... . . . .. .. . . then any other vector in the space can be written 0 Exercises 13.. . .. . . . .. .. . ... ... .. . . . .. . .... .. . . . . . . . ... .. . . . ... .. . . . ... ... . .. .... . . . . . .. .... . a • b• td a ad ) ( a • • . . .. . . ... .. .. ... ... ... .. . . . . Therefore λ k (any λ ) is perpendicular to b. . . ... ..... . ... ....16... ....4 Scalar differentiation of a vector (16.. . ....16 16. .... . . . . .. 14 (i) d1=12a1+13b1=12(2i1+13j1+1k)1+13(i1−12j)1=1(4i1+16j1+12k)1+1(3i1−16j) Scalar differentiation of a vector =17i1+12k ∆a1=1a(t1+1∆t)1−1a(t) ) + ( o and the derivative of the vector with respect to t is deﬁned in the usual way by the limit (if the limit exists) u1=1uaa1+1ubb1+1ucc ∆a a(t + ∆ t ) − a(t ) da = lim = lim ∆t dt ∆ t → 0 ∆ t ∆ t → 0 Figure 16. . .. . . . . ...... . . .... . . ... . . ... . ... . ... The numbers ua.. .. .. . .. .. . . .. ....... .. .... . . .. .. .. . .. ... .. . .. . . ... .. .. . . .. .. . .. ... . .. ... . ... . .. .. . . .. A vector a1=1a(t) is a function of the scalar variable t if its magnitude or its direction.... . .. . . . ..... . . . . .. . ... . .20) 453 ... . ... . .. any three noncoplanar vectors can be used as base. .. .. . .. ... ..... . .. or both. . .. .. .. . .. . . 2 • .... . ... . . . ... . .. ... ... .... ....... ..... . . . .. . . . .. . . . . . .. ..... .... .. . .. . . . ... . . . . . . . . b3 + a •... .. . . .... ... .. .... .. . . .. . .. . . ...... .. . ... . ... . . . ... . . .......... . .. . . . .... .. ... ... . .. .. .. b and c are three such vectors... .. and λ j is perpendicular to d.... . b2 +.. . ... ... . . ..... ....... . . .. .... .... . ub..... . ...... .. .. . .. . ......... . ..... . .. . . .. . .. . .... . . ..... .. . .. .. . .. . . . . . ... . .. . . . .. . . .. . . .. . ..4 Although the cartesian unit vectors provide the most widely used representation of vectors in three dimensions. . ... . ... .. .. .. ... . . . . . . . ... .. .. .15 Figure 16. and every lattice point has position vector (16.. .. . . . . . . .. . .... ...... . . . .. .. . .. ... . .... . . . ... .. The change in the vector in interval ∆t is . . . .... .. . ... .. . ........ .... .. . . .. .. . .. . . .. . ... . ... . . . . .. . . ... . .Taking the limit corresponds to letting point B approach point A along the curve deﬁned by a(t).. . .... . ... . . . An example of the use of non-cartesian bases is found in crystallography in the description of the properties of regular lattices. . . .. . ...... . .. . .. .. .. .. .. . and this is the direction of the derivative at A.... . . .. ..20) with components that are integers. . . ... .. .. .. . .. depends on the value of t.. . . . .. . .... . . .. ... .. ...... . . . .. .. . . . . . . a.. . . .. ... This is illustrated in Figure 16. . . .. .. . ... . .... . ... . . ....... .. . . . . .. . . .. . .. .. . . .. .. . . . . . . ...... .. . not necessarily unit vectors or perpendicular.... ..... . .. .. . .. . .. .. . . . .. ..... . .. . .. . .. .. . . ... . . . . . .. . . . . . .. . . .. ...... . .. Thus. . . . . . . . . ... ... . .. a • 16.. . ....... . • . . . .. . The direction of AB = ∆ a approaches that of the tangent to the curve at A. ... . . . . . .. . . .. ... .. ... . . . .. ... . .. . other representations are sometimes more useful... . . . ..... . .. OA = a(t ) and OB = a(t + ∆ t ) are the vectors for values t and t1+1∆t of the variable.. ... If a.. . ..... .. .. ... . ... .. . .. . . . . . . . .. . . . . ... . . b and c deﬁne the crystal axes and the unit cell. . t ∆ t . . .... . .. .. . .. . ..... ... ... ...... . . . . .. . .. .. . . . .. .. . . .. . ... .. .. ... .. . .... .. .. .. ... ... .... .. .. . . ... .... vector k lies along the z-direction.. ... . .. . . ... . .... . . . .. .. . . .. . .. . .. . . ... .. . . ..... ... . . . . . .. . .. . . . .. t . .. . . . .. . . . . . .. .. . ... . ....... . . .. .... . . . ... . . ... ... ...... . .. ... ... . . . ... . . ... .. . ..... .. . .. . ... . . .. . .. . . .... .. .

... . .... . . . .. .. )t( .... . and the parameter t is the time variable.17). . .. ... . .... . . . . . . . . . .. . .. .. . . z . . . . dt 0 Exercises 15. . .8 Find da2dt and d 2a2dt 2 for a(t)1=13t 2i1+121sin1t1j1+1e−tk. The sense of increasing values of t is called the positive sense on C.. .. . ... . . . ... . . . . .. . . acceleration. . .... ... ... .... . .... . .... . . . . . .. . . ... . . ..454 Chapter 16 Vectors In terms of the components of the vector.. . ..... .... . . .. • r . and t is the parameter of the representation (Figure 16. .. .. . .. .. ... . . ..17 EXAMPLE 16.... . ... .. . where i.. momentum and force Velocity and acceleration If r1=1r(t) is the position vector of a body in space. . . . . j. . .. . and k are (constant) base vectors. t. . . v= dy dz dr dx = i + j + k = vx i + v y j + vz k dt dt dt dt (16.. derivatives of position vectors provide a general method for the description of the mechanics of dynamic systems.... .. .. . . . . . . . .. . . . . ... . .. . ... . . . . . .9 Velocity.. .. . ... ... ...22) (16. . . .. given a cartesian coordinate system... . ...25) C y . the dependence of a on t is that of the components: a(t)1=1ax(t)i1+1ay(t) j1+1az(t)k and da da da dax y j+ z k i+ = dt dt dt dt EXAMPLE 16. . . .. ... .. ... .. .. .. . . . da = 6t i + 2 cos t j − e− t k . . . .. . .. . . .. . ..23) d 2a dt 2 = 6 i − 2 sin t j + e− t k If r1=1r(t) is the position vector of a point for each value of t in some interval then.. if a1=1axi1+1ay j1+1az k. . . .. . .. . . ... . .. .24) . . . .. . . . then dr2dt is the velocity of the body. o x is a parametric representation of a curve C in three-dimensional space.. . .. 16 We note that the base vectors may also depend on t if the coordinate system itself (the frame of reference) is undergoing changes Parametric representation of a curve (16. . . . . When the parameter t is the time variable. .. . . .. . . . .. . . . and deﬁnes a direction on the curve... . .. .... . .. . . . . . .. Figure 16. . . r(t)1=1x(t)i1+1y(t) j1+1z(t)k (16. .

(ii) a = = − 4 cos 2t j . The direction of p lies along the direction of the line of motion.29) Equation (16. (i) v = dr dv = 4i − 2 sin 2t j . Velocity is therefore the rate of change of position. is then 2 2 v = |v | = v x + v 2 + v z y (16. Fy = m d 2y dt 2 . the principle of conservation of linear momentum.4 Scalar differentiation of a vector 455 where vx. (ii) its acceleration. and p is a constant vector. (v) in the y-direction and (vi) overall. Linear momentum and force In mechanics the momentum p of a body of mass m moving with velocity v is deﬁned as p1=1mv.16.26) and the kinetic energy of the body is T= 1 2 1 2 2 mv = m v x + v 2 + v z y 2 2 ( ) (16. (iii) the force acting on the body. (iii) F = ma = − 4 m cos 2t j dt dt .28) = x i+ dt dt dt 2 dt 2 dt 2 dt The vector form of Newton’s second law of motion is then F1=1ma. the speed. one for each component. By Newton’s second law of motion. Find (i) the velocity of the body.29) shows that when no external forces act on a system whose linear momentum is p then dp2dt1=10. This is Newton’s ﬁrst law of motion.10 A body of mass m moves along the curve r1=14t1i1+1cos12t1j. The magnitude of the velocity. equivalent to three scalar equations. EXAMPLE 16.27) The acceleration of the body is the rate of change of velocity. vy and vz are the components of velocity in the three cartesian directions. a= dv d 2 r = 2 dt dt dv 2 2 2 dv dv y j + z k = d x i + d y j + d z k (16. and p is often referred to as the linear momentum. Fx = m d 2x dt 2 . Fz = m d 2z dt 2 . Describe the motion of the body (iv) in the x-direction. the force acting on a body is given by the rate of change of momentum that it induces: F= dp dt (16.

.. ... .. . . . .. .. . .... . . . .... . . ... .. ..... .. ... .. . . . .. . .. . .. . . The unpublished work.. .. . .4 The scalar product or dot product of a and b is deﬁned as a1·1b1=1ab1cos1θ (16. .... independently. .. .. . . and ‘among the most renowned of theoretical physicists of all time’ (Max Planck). . . .. . . . ... ...... . ....18... . . . . .. . . . . . . .... Connecticut.. . . .. .. .. .. . .... . ... . .. . . .. . ...... . .. . . . .. .. . . . . ... .. . English electrical engineer... ... . .. . . Figure 16. ......... .... .. .. . .. .......... . . ..29) in Section 12.. . . . . . .. by equation (12.. . ... .. . . . . ..19 4 Clerk Maxwell used Hamilton’s quaternions in the Treatise on Electricity and Magnetism of 1873. . . .... . .. . . . .30) ... .. ... .. . . ... .. but it was observed.. .... . . .... .. . . .. . .. .. .. . . . . . ..... ..... .. . . . .. . . . (v) The acceleration is in the y-direction with ay1=1d 2y2dt 21=1−4y and... ... . . .. . .. . . the motion is simple harmonic in this direction. . . . is noted for his development of Laplace transform techniques. ... ... .. . . . .. . ... .. ... . .. . .. .. .. .... ... . .. ... . .. Wilson’s Vector analysis... . . .. . . .... . .. . .. .. . .. . . . . . .. . . . .. ... ...... .. . .... .. . . .. . . . .. . .. . . . . y .. ... (vi) The body moves to the right with constant speed vx1=14 whilst undergoing simple harmonic motion in the perpendicular y-direction. ... .. .. . .. . . .. ... .. . .. . .. . . . . . ... . ... . .. . .... .. .... ... . . ... . .. .. . bz) then a x .. . .6). 18 16... . . .. ... .. . .. . ...... . if a1=1(ax. .. . ... .. . . .. .. . .. .. . . .. . ... ... ... ....31) Figure 16........ . .. . . American engineer..... . . ... . .. . . . . .5. .456 Chapter 16 Vectors (iv) The velocity in the x-direction is vx1=14 so that the body is moving in the positive x-direction with constant speed 4.. by .. . . .5 The scalar (dot) product The application of vectors to physical problems has led to the deﬁnition of two ways of multiplying a vector by a vector. the other gives a vector product (Section 16. ... . circulated as lecture notes.... .. . . . . .. .. ........ .. . . . . .. .... . .. . . π 6 .. ... ... . . ... ... ... . . .. . . . . . as shown in Figure 16. ... . .... . . .. .. ... . . .. .. .... . ....... . . .. .. . . . . . . . .. .. . . . .. . .... . . . .. . ... . . . ..... .. . . . . .. .. chemist. .. . . . . . .. . . ... . . . ... . ... .. .. .. . and died in New Haven.... . . ... . . .. . . .... . . .. . az) and b1=1(bx. . . . .. .. ... b a θ o (read as ‘a dot b’) where a1=1|a | and b1=1|b | are the lengths of the vectors and θ (01≤1θ1≤1π) is the angle between the directions of a and b. . . His introduction of the concept of chemical potential in On the equilibrium of heterogeneous substances in 1876–8 marks the foundation of modern chemical thermodynamics. . In terms of components. .. by Heaviside and Gibbs that the full algebra of quaternions was not necessary for the representation of physical quantities..... . . . . .. . . .... .. . was popularized by Edwin B... . Gibbs is best known for his work in thermodynamics and statistical mechanics. . . . .. . . − 1 1 0 . . . He developed the modern notation and the concepts of scalar and vector products in the early 1880’s to simplify the mathematical treatment in Maxwell’s Treatise. . ..... .... . . ... .... .. . . . .... . .. . .. . . . . . ... . . . ... .. . a1·1b1=1axbx1+1ay by1+1azbz (16.. . .. .. . . ... .. . .. ay ... .. . . . . . . ... . .... ... .. .. lived.. ... . . .... .. . π 4 .. . . ... one way gives a scalar as product. . .. Willard Gibbs in 1901. . . .. ..... . ... . . . . .. . ... ... ... . .. . .. ... . .. . . ... Josiah Willard Gibbs (1839–1903).. . .. . ..... . .. . .. . b ... . .. . . . ... .. .. .. . Oliver Heaviside (1850–1925)...... .... . . . .. .. .. . . π 2 . .. . . . . . .... . . ..... . .. ... ...... . . .. . was born. . founded upon the lectures of J. . . .18 0 Exercises 17. . . . ..

. . . ... ... . ...86° 154 a 1·1b . . and its length is given by To show the equivalence of the deﬁnitions (16.. . . 16.. . . . .. . . . .. . .. −1) and b1=1(1. . .31). .. and b1=1|b |1=1 14.. .. . .. ..8702 ≈ 49. . . . =1a 21+1b21−12(axbx1+1ay by1+1azbz) 2 2 2 2 2 2 =1(ax 1+1ay 1+1a z )1+1(bx 1+1by 1+1bz )1−12(axbx1+1ayby1+1azbz) 8 . .. . . .. .... or negative: 0 Exercises 19–21 EXAMPLE 16..... . .. . . .. . .. . . . ... .. . . .. b π 2 <θ (16.. .... .. ... .. . .. . . . . . . . . . . ... .. . . Therefore This example demonstrates that scalar multiplication is commutative. . . . ..... . . .. . . . .. ... .. .. . ..... . . . ...... .. . . . . .. so that the scalar product can be positive. .. .. . .... . . . . . .. . . . . . .. .. . .. . cos θ = Therefore ab1cos1θ1=1axbx1+1ay by1+1azbz. .. a . . .. . . . . . . . .. . a . . . . ... ... π 2 >θ 0=b Figure 16.. . a1·1b1=1b1·1a (AB)21=1(OA)21+1(OB)21−12(OA)(OB)1cos1θ1=1a 21+1b21−12ab1cos1θ a1·1b1=131×111+111×121+1(−1)1×1(−3)1=18.. . .. . . .. . ... .. .5 The scalar (dot) product . . .... .. . . .. . . . .. . .. . . ... . . .. . .... . . . . . .. .. . . .. ...... .30). . .. . . . . .. . . . . . The sign of the scalar product is the sign of the cosine... . .. ... .. . .. ... . .. . .. .. .. By equation (16.. . .. . .. .. ... .. .. . . .. ...... .. . .20 a.. . . . . . . ... .. . . .. . . ...... ... . .. .. ... . 2.... 1.. . . . . ...... . . . . .. . .. . . . we apply the cosine rule to the triangle in Figure 16. .. . . . . . ... . . . . . π 2 =θ 0>b a.. .. ... . ... .. .. . . .. a . . . ... ... . . .. .... .. ... .. . . . . . . . .. . . .... ... . . .. .. −3) ﬁnd a1·1b. . . . .... . .... . . . . . . . .. . ..11 Given a1=1(3. .. ..... .. .. .. θ . . .. . . . . . .. zero. .. .... .. . .. .... . . . . .. The vector AB is b1−1a... ... . 154 8 θ = cos−1 ° ≈ 0. . . . . .. . . .. . .31). . .. .. ..32) 457 . . . . and the lengths of the vectors are a1=1|a |1=1 11 ab .. . θ. . b1·1a1=111×131+121×111+1(−3)1×1(−1)1=18 (AB)21=1(bx1−1ax)21+1(by1−1ay)21+1(bz1−1az)2 cos θ = .... . .. . . ... .. . .. .. ...... . b θ b ..... . . . . .... . . ... .19: Proof By equation (16. ... . .. . . . .. .. ... .. .. . ... . . .. ... . . ..30) and (16.. ... .. . . . .. .. . . ... . .. .. . .. . . . . . .. . .. .. . .. .. . . .. . ... . . b1·1a and the angle between the vectors. .. .. .. ... .. . . .. 0<b a.. .. . .

(16. =1ax bx i1·1i1+1ax by i1·1j1+1ax bz i1·1k1+1ay bx j1·1i1+1ay by j1·1j1+1-1+1azbz k1·1k . expressing a and b in terms of the base vectors.35) The expression (16. j. −2) are orthogonal.* We note that equation (16. a1·1b1=1(ax i1+1ay j1+1az k)1·1(bx i1+1by j1+1bz k) =1ax bx1+1ay by1+1az bz 0 Exercises 24–27 * Orthogonal means perpendicular for vectors in ordinary space. λ . 1) and b1=1(4. 23 When a and b are the same vector.31) for the scalar product follows from these properties of the base vectors. (iii) a is orthogonal to (b1−1c).33) and (16. and b orthogonal to a. −2.33) The vectors are then said to be orthogonal.33) shows that it is not possible to cancel vectors in a vector equation in the same way as is possible for scalars. a1·1b1=101=121×141+1λ 1×1(−2)1+111×1(−2)1=161−12λ .458 Chapter 16 Vectors For nonzero vectors a and b. the scalar product is zero when the vectors are perpendicular. Thus. EXAMPLE 16. with a orthogonal to b.34) The base vectors i.12 Find the value of λ for which a1=1(2. a1·1b1=10 (16. by equations (16. and k are orthogonal and of unit length so that.34). (ii) b1=1c. i1·1j1=10 j1·1k1=10 k1·1i1=10 i1·1i1=11 j1·1j1=11 k1·1k1=11 (orthogonality) (unit length) (16. For orthogonality. 0 Exercises 22. but the deﬁnition of orthogonality applies generally to vectors in ‘vector spaces’ of arbitrary dimensions.31) gives 2 a1·1a1=1ax 1+1a 21+1a 21=1|a| 2 y z The length of a vector is therefore given in terms of the scalar product by | a | = a 1·1a The use of cartesian base vectors (16. Therefore λ 1=13. Thus the equation a1·1b1=1a1·1c has the three possible solutions: (i) a1=10.

. . .. . . . . .. .. .16.21 where Fd acts along the line of d and F acts at right angles to d. as in Figure 16..... .. . We note that the work done by the body against the force is −W1=1−F1·1d. The displacement of the body is d1=1r21−1r11=1(x21−1x1.. . . . y2. . the component of F in the direction of d can be written in terms of the scalar product as Fd = F cos θ = F 1·1d ˆ = F 1·1d d ˆ where d = d d is the unit vector in direction d. .. . . ... . . . . . .. . . . .. .. .. ... .. .21... .. . .. .. .36) . . . . . .. .. . ... .. .. F1=1F 1+1F d ⊥ Figure 16...... . .. . . . .. the work done is W1=1F1·1d1=121×121+111×1(1−13)1+101×111=11 In addition. .. . . ... .. .37) where.... ... . 1). . . .. ... . . . . . . y21−1y1.. . ... . . . . . . . . ... . . . ... . .. . . . .. .. . . . . .. . .. . .. θ... .. ..... ... . . F is the component of F in the x-direction and Wx1=1F dx is the x x work done in moving the body through distance dx1=1x21−1x1 in this direction (‘work along the x-direction’).. . .... .. ... . . 1.. . .. . . ... .. ....... . . . . .. .. . . .. z1) to position r21=1(x2. . . . . . . .... . . . .. ...... . . . For example. for example.. . ... . .. . . . . . .. ... .. .. . ..... .. Thus. .. . .5 The scalar (dot) product 459 The following examples demonstrate two applications of scalar products in the physical sciences. . .. . In terms of cartesian components. . y1. .. . .... .. ..... ... . . . . and this is the only component of the force that contributes to the work. . .. . . in appropriate units... . . . .. .. . z21−1z1) and the work done by the force is W1=1F1·1d1=1Fd1cos1θ . . . . ... . . . −3. . . .. . . . . ... . 0 Exercise 28 d d F F ⊥F The quantity F1cos1θ is the component of the force in the direction of d. .... . . ..... ... . . .. . (16.. .. .. . ..36) is W1=1F dx1+1F dy1+1Fz dz1=1W 1+1W 1+1W x y x y z (16. . .. . . . . .. ... . the work (16. . .. . 0) and the displacement is d1=1(2. . . . . In the present case.. .. . The latter cannot ⊥ cause a displacement along d. . . ... . ... .. . . .. the force can be written as ... . . . . ..... . d = 14 so that Fd = 1 14 .. ... z2) under the inﬂuence of a constant force F. . ...... . .. EXAMPLE 16. . . . . . . .. .. . . . ... .. . ..13 Force and work Let a body be displaced from position r11=1(x1.. . . . . if the force is F1=1(2. . . ..... .. . .. .. .... .

.... . . .. .. . .... . .... . ... .. and the discussion of Section 9.. .. the components of the force can be expressed as (partial) derivatives of a potential-energy function V (see equation (5.. .. ... .. ... ....... in the limit |∆r | 1→10 on the curve. ... ......... . ... . . . . ..... . . .. ...8).. . . ..... .. ... . . ....... . ..38) is independent of the path... . .. ..... .... .... F ) and x y z dr1=1(dx...... . .. r . . ..8 applies with only minor changes.......22 WAB = Z F ( r )1·1 dr C (16. ..... . .. . . . ........ . ...... .. . ... . .. ∆ .. . . the line integral (16.. ... .. . . ..... . .. ... . ... . .......38) and the integral is a line integral. . . . ... . . . . . .. . . ... .... .. . . . Fx = − Then ∂V ∂V ∂V dz = − dV Fx dx + Fy dy + Fz dz = − dx + dy + ∂z ∂x ∂y (16... The total work done from A to B on C is then ... . ...... . ... . . ... .... ...... . Figure 16. . .. .... .. . ...... and x y z WAB = Z Fx dx + Fy dy + Fz dz C (16. . .. .. .. . . . .. .... ... .. . .... .. . .. ... . In terms of components... ... ....460 Chapter 16 Vectors The general case . ..39) This is a generalization of equation (9. .... .... ... ... . ..... . .... . . . . ... .. . .... .. . .... . . . .. . ..... . .... by a generalization of the discussion of conservative forces in Sections 5.... ......... . . . . . F1=1(F . .48) for a curve in three dimensions.. . .. .. .... .. dz).... .. .... . ... . . ... . . . .. .57)).... . . ... .. .40) and dV is the total differential of the potential-energy function V(r)1=1V(x. .. .. ... .. . . .... .. dy. . ... . .. . . ... for a conservative force.. Consider a body moving from point A at r1 to point B at r2 along the curve C under the inﬂuence of a force F whose value varies from point to point on C (Figure 16.. . ... . . . . . .. .... .... ... .. . .... .. F . • . z).. .. . . .. . .... ....22).. ... . .. .... ... . . .... . .. . and is equal to the difference in potential energy between A and B: WAB = − Z dV = VA − VB A B (16. . ... Let F(r) be the force at point r on the curve... . . z . . .. . ...... ... ... ... . .. the element of work is dW1=1F(r)1·1dr.42) 0 Exercise 29 b• y r • r ∆ +r . . ... . . ......... if the force is a conservative force then. .. y.. . In particular.. . . .8.. . The work done on the body between positions r and r1+1∆r is ∆W1≈1F(r)1·1∆r and.. .7 and 9. .. ....41) ∂V ∂x Fy = − ∂V ∂y Fz = − ∂V ∂z (16.. .. ..... .. .. .. .. .. . It follows that. .. ....... ...... . . . .... . . .. .. .. o •a C x If the force is not constant along the path r1 to r2 then the work has the form of a line integral (Section 9... ... F1·1dr1=1F dx1+1F dy1+1F dz. ..

.. . . .. If the ﬁeld is an electrostatic ﬁeld (constant in time) then this force is conservative so that... ... ... .... .. .... .... ...... ..... E1=1F2q.. . .... .... . . .. . . . . .. ..... . ∂x Ey = − ∂φ ... .. ... ..45) .. ... .. by equation (16.. The potential energy of charge q at r in the ﬁeld is then V1=1qφ (r)1=1−qr1·1E1+1qC (16. .. . V = q1 φ ( r1 ) + q2φ ( r2 ) + + qN φ ( rN ) = ∑ qiφ ( ri ) i =1 N Figure 16.. . are (minus) the derivatives of a function φ1=1V2q: Ex = − ∂φ . . . .. ... . If the ﬁeld E is a uniform ﬁeld.. .5 The scalar (dot) product 461 EXAMPLE 16. .. ..... . then integration of these equations gives (see (5.. is the sum of the energies of the individual charges.58) for a constant force) φ (r)1=1−(xEx1+1yEy1+1zEz)1+1C1=1−1r1·1E1+1C (16.46) =1−µ1·1E1+1QC where µ is the dipole moment of the system of charges and Q is the total charge.. qN at rN . q1 at r1.23 For the potential (16.. ... .... .. ..14 Charges in an electric ﬁeld The force experienced by a charge q in the presence of an electric ﬁeld E is F1=1qE. .... .16. .. . y... q2 at r2 =. .... . .. ..... . .... . . .43) The function φ is the electrostatic potential function (potential energy per unit charge) of the ﬁeld..40). . constant in space.. . . .. .... ...... .... z) and C is an arbitrary constant.... . ... . Then V1=1−µ1·1E1=1−µE1cos1θ 0 Exercise 30 µ . θ ... The term QC is zero for an electrically neutral system or if the potential φ is chosen to be zero at the origin (the usual choice). ∂y Ez = − ∂φ ∂z (16... ......... ... .... .. . . ........ ... we therefore have N N N V = ∑ ( − qi ri ·1 E + qi C ) = − ∑ qi ri 1·1 E + ∑ qi C i =1 i =1 i =1 (16.. . .......44).. . . . E The potential energy of a system of charges. .. . the components of the ﬁeld...44) where φ (r) is the electrostatic potential at position r1=1(x..

.. .... . ... .... . . ... .. . . . .. . . . θ. .. . . ......... . . ... . . the vector product or cross product of a and b. . . . . .. . .. . ... .. .. . . . . . .. .. ... . . that is. . . . . .... . .25 Figure 16. ... . . .. ... ... . . .. . . . ..... . ... . . .. . . . . . . . . . . . . . . . . . .... .. . ... .... .. . . . .. . .. ... ... ... . . ... . . .. . .... . . . . .. .. .. .. ... .. . . . . . . ... . . .. ... . . ... . . . .... . . . .. . . In particular. . .... so that Every pair of non-parallel vectors. . .. . .. . . .48) that the vector product is zero if θ1=10 or θ1=1π... .. .. . . . . .. .. . . . . . . . .. .... . . ... . . ..24. .. or ‘direction down’. . . .. . . .. .. . ..... .. . . .. .. ... . . . .. . .. .. ... . . .. . . . .... . . .. . . . . . . . .. . .. . . ... .. .. . θ nis b . . ... . . . . . . . . .. ...... . . . ... . they deﬁne a sense of direction of rotation. .. . .. . . .. . .. . . .. . .. .. . . .. . . . . .. . .. .. . . . .. . . . . . . .. ... . .. . .. .. . . . . . .... .. . . . . .. . .. . . .... . . ... . ..... . . v)... . . ... . .. .. . . . . ..... . . . . .... whose area is ab1sin1θ (base × perpendicular height). . .. .. . .49) (16. . . .. . . . . . . . ... .. ... . the vector product is not commutative. . . ... . . . . . . . . .. ... . . . v b . . . ... . . . . ... . . .. b a (read as ‘a cross b’)... . .. .. . ... . ... . .. .. . .. . .. ... . . . . . ... . . .24 v1=1a1z1b a1z1a1=10 )i( (16. .47) is reversed.. . ... . .... b. . .. ... .. ... . . . . . .. . . .. .. . . . . . .. .. . . .. . .. .. whose length is These properties of a pair of vectors deﬁne a new vector. . . .. . .. . .... .. . ... . . .. . . . ... . anticlockwise. . . . in (i) for a into b. . . .. . . Chapter 16 Vectors and whose direction is perpendicular to both a and b (to the parallelogram). . . v b . . .. . . ... . . Figure 16... . . .. . . . .. . . . . .. . . .. ........ . ... . . .. . . . ...... . . .... . . .. ... . . .... ..... ... . . . . . if a and b have the same or opposite directions.. . .. . . . .. ... ... . and is such that the triple of vectors (a. . . ..... .. . . . . .... .... . . ... . . ..... . . . . .. .. . . . ... . . . . . .. .. . . .. ... . . . .. .. . ... . . ... . . ... . .. .. . . ... .. .. . . ... . . . . . . . . .25(i)).... . .. . .. . .462 Unlike the scalar product therefore. .. .. . . . . . . . . . . . . ..... . .. . .. .. .. . .. . . .. . .. . .. . ... . .. .... .. . ..... .. ... . ... .. . .... . .. ..... ... . .. ... ... b .. . . ... . .. .... . . . . . . . . or ‘direction up’. . ... .. . . . . . . . . . . . . .. ... ..... . . ... .. . . . . . . ... .. . .. .. ... . .. . .. . . ... . . . . . .... . .. ... . .. . . .. .. .. .... . .. . . . .. .... .. .... . . .. .... . ... . . . . . . . .. . . . . . . . . . . . . . . . ... . .... . ... ... . . in this order. .. .. .47) (16. 16. . . .. . . . . . . .. . .. .. it is anti-commutative.. . .. . . . . .. . .. ... . .. . . . . . . . .. . . . . ... .. . .. . . .. . .. . .... .. . .. . .. .... ........ . . . .. . . . . . . .. .. .. .. . . . . .. . . . ..... ... .. .. . . . . . . . . .. . ... . . in (ii) for b into a. . . ..... . . . .. .. . . . . . . . ... ... .. . . . . . . . ... .... deﬁnes a parallelogram.. . . .. . . .... . .... . . . ..... . .. . .. . ... .48) (16.. . . .. . . . .. . . .. . . . ... .. .. ... . . . .. . . . . .6 In addition.. . It follows from (16.. .. . . . . . .. . .... ... . .. . ..... . . ... . Figures 16. . . ... . . ... .. .... . . . . . .. . ... .. . . . . . . . . . . ... .. . . .. .. . . ... ... . . . . . . . . . . ... .. . .. .. . .. . .. .50) a ... . . . . .. . . . .. . . ... .. . . . ... ... ... . . . . .. . ....25 show that when the vectors are set head to tail.. .. .. . . . ... ... . .. .... ... .. . The vector (cross) product b1z1a1=1−a1z1b |v | 1=1ab1sin1θ Figure 16. .. ... .. ... . ... . . ... . . . .. . .. . . ... . form a right-handed system (as in Figure 16.. . . . .. . ... ... .... . . . . .. .. . . .. . ... ... . ... . . . . .... . . .... . . . . . .... .... . . a and b. . . ... .. .. . . . . .. . . . . .. . . . .... . . . . . ... .. .. .. ... .. ) ii( . . . . . . . a a . . .. ... . . . Figure (ii) shows that this direction is reversed when the order of a and b in (16. ... .. .. . . . . and clockwise. . . . .. . . . . . . . .. ... . .. . .. . . .. .

.51) The vector product a1z1b can then be expressed in terms of cartesian components: a1z1b1=1(axi1+1ay j1+1az k)1z1(bxi1+1by j1+1bzk) =1ax bxi1z1i1+1ax byi1z1j1+1ax bzi1z1k1+1ay bx j1z1i1+1ay by j1z1 j +1ay bz j1z1k1+1az bxk1z1i1+1az byk1z1j1+1az bzk1z1k =1ax by k1+1ax bz(−j)1+1ay bx(−k)1+1ay bzi1+1azbx j1+1az by(−i) (remembering that. 5) Also.53) EXAMPLE 16. −8. j and k form a right-handed system of vectors and have properties i1z1j1=1k i1z1i1=10 j1z1k1=1i j1z1j1=10 k1z1i1=1j k1z1k1=10 (16. i a 1 z 1b = a x bx j ay by k az bz (16. 1. Therefore a1z1b1=1(aybz1−1azby)i1+1(azbx1−1axbz)j1+1(axby1−1aybx)k (16. 8. b1z1a1=1−a1z1b1=1(1. in the form of a determinant (Chapter 17). for example. j1z1i1=1−i1z1j). and more easily remembered. −5) | a 1z1b | = 12 + 82 + 52 = 90 0 Exercises 31–40 Vector products are used for the description of surfaces in geometry. b1z1a. and in the vector integral calculus for the evaluation of surface integrals. ﬁnd a1z1b.6 The vector (cross) product 463 In terms of cartesian base vectors The unit vectors i.16. −1) and b1=1(1.52) This form of the vector product can be written. 2. −3). the vector product is used for the description of properties associated with torque and angular motion. By equation (16.15 Given a1=1(3. and the area | a1z1b| of the parallelogram deﬁned by a and b.52). In mechanics. a 1z 1b = 1 × ( −3) − ( −1) × 2 i + ( −1) × 1 − 3 × ( −3) j + 3 × 2 − 1 × 1 k =1−i1+18j1+15k1=1(−1.

.. .... . . . .. .... . . . or torque T. . ... ..... . . .. ... . ... .. ..... If r is the vector from O to a point A on the line. ....... . . . .. . ... ... .. .. . .. . ...... ..... . . . . . ...... .55) Figure 16...... ... .54) . ....... ... . ... . .. .. 0 Exercise 41 T1=1r11z1F11+1r21z1F2 T1=1|r || F | sin1θ1=1| r1z1F | T1=1r1z1F v1=1ω r T1=1µ1z1E =1q(r11−1r2)1z1E1=1qr1z1E1=1µ1z1E (16... ...... .. Thus... .... . . about a point O is deﬁned as the product T1=1|F | d.. . ... .. ... .. . . ... .... . . . . ... . . . . .. .... . . . . .. . ........... . . The corresponding vector An electric dipole µ1=1qr (see Example 16.... . .... ... .. from Figure 16. . .. . . ... . . . .. .... . .. .... ... . . .... .... . . . ... .. . .. . ... . . . .. . . .. .. .. . 43 is perpendicular to r and F.. .. . . . . . . . ... ... . .. .. . ..... ... . . . .. .... ... ... .. . . ... .. . ... . . .. .. ... ... .. ... . . .. . ....27. . .. . . ... . ...... .. . E Angular velocity EXAMPLE 16.. .. . . .. .. .. ... ..26 (16... . . . . .. . ... . ..... . .. .27 Figure 16.. . . .. ..... . ... .. . . . . . . . .. .... . . . ... . .... ... . .. that tends to align the dipole along the direction of the ﬁeld. .. .. . . . . .. . .... .. .. .. ... ... . . .. .. .... . .. . .. . .. . . .. . . .. . . . .. .. .. .. .. .... . . .... ... .. .... and its direction is that of the axis through O about which the force tends to produce rotation.. . . .. ..17 An electric dipole in an electric ﬁeld EXAMPLE 16.26).. .. ... . .. . . . .. ... the total torque about O is A particle moving in a circle of radius r with speed v has angular velocity y (angular speed ω) about an axis through the centre with magnitude given by o• E q− = 2F r 2 1 r • q− .... .. .. .. . . . . .... .. . . . .. .. .. . . . .. ... ... . .... . . . ..... . . . . . . .. ..... . .. ... ... .. . ..6) in an electric ﬁeld E experiences a torque E q = 1F ... ... . . . . .... .. ... . . .. . .... . . . .. . .... .. . . ..... . .. .. .... . .. . .. ....... .. .... ... .. .... .... . .. .. . .. q• . . .... ... . ... ... .. . . .. .. .. . ... ... . ..16 Moment of force (torque) and direction at right angles to the plane of motion (out of the page for the motion shown in Figure 16. . ..... .. ..... .. . ... . .. .56) (16. ... ..28)... where d is the perpendicular distance from O to the line of action of the force (Figure 16. .. ... ... . .. . ... . . . .. .... . .. .. .. ....... . .. . . . .... .. . .. . . ... ... . . .. ... . ... .. .. .. ...... ... . .. .. . . . . . .. .... .. ....... θ a• ....18 Angular velocity and angular momentum EXAMPLE 16......... r •o 464 Chapter 16 Vectors In mechanics the magnitude of the moment of force F.. . . . . . .. ... . . ... . .... .. . . r ..... . ... d . 0 Exercises 42.. . .. . ... . .. . . . . then d1=1|r | 1sin1θ and F. . . ... .. ...

.. . .. .... .. . ..... . . .. .. . . .. . . nis .. ... ...... .61) (16.. . .. . . ...... . ... . .. . . .. . .... ... . .... ... .. . . ... ... . ... . .. ..... .. . . .. ........ . .. . . .... ... . .... . ... .. . . ... .. . . . .. ... .. . ... as can be seen by reference to Figure 16.... ... .. .... .... . ... .29 16. ... . ..... . . .. equation (16.. . ... . . . .. ........ ..... . . ... . .. . .. .... .. . . . .. ... a Figure 16... ..... .. .... . . ..... . . ..... . ..... . .. . . . . .... This is the case for motion in a circle with the reference point O at the centre of the circle. .... .58) gives the velocity at each point r in the body....... . . ... .. . .. ... ... . . .... . .. .. ...... . so that l and y have the same direction only if r and y are perpendicular vectors. . . ... ... ...... ... . .. .. . . .... .. .... . ... .. .. . ... . ... .... . ......... .. ... ... ...... . .. .......... . .. .. .. . .... .. .. . r • .... ..... .... ... . .. ... ..... . . ... . . ....... .29.... . ... ... . .... . .. . ... .. .. ... ... . . v ω . . . . . . . . • . . ... In this case. ..... .... ... .. . . . .......... θ ... .56) is replaced by which also correctly deﬁnes the direction of each of the three vectors with respect to the others.. ..... . . . .... . .. . • r ... ... ... .. . .. .. . . . .. . .... .. . .30 (16.. . ... . ... . .... .. . .57) (16. ... .. ... ..... . .. . .... ..... . ... .. .... .. . .. ..... . . .. ... ... ..60) (16.. ........ . ..... . . .... .... . . . .... . .. . . . . .. ........ . . as illustrated in Figure 16.. θ .. . .. .... . .. . .. . . ........ Then Its magnitude about the point O in Figure 16...... . .... .... . ... .... .. . . . .... . ... .... . .. . .. ... .... . ..... .. . . . • o Angular momentum l1=1mr 2ω 1=1Iω l1=1mr 2ω1−1m(r1·1y)r v1=1ω r1sin1θ l1=1r1z1p v1=1y1z1r Figure 16...... .. .. . The angular momentum l of a mass m about a point is deﬁned as the moment of the linear momentum p1=1mv about the point... ...... . . when r1·1y1=10. . . ... ... .... . . . . . .... . .. . ... ...... . ... . .... ... . .. ... p .. .. . .... .. . .. . ... . ... . . . ... . ....... . ... .. ... ... .. . ..... . . ... . . ..... . . . . . . . . .......... .. .... . . . .. .. . .. .... ...... . .. .. ...... . ... ......... . For a rigid body rotating with angular velocity y..... ... . It can be shown that (see Exercise 44) and this is just the magnitude of the vector product More generally. ...... .... .... . . .. .. . .... ... .... . . . ... .... . .......6 The vector (cross) product Figure 16.... . .. . . . ..... .... we consider a particle (that may be part of a rotating body) rotating about the axis OA through the ﬁxed point O. .. ... . . . . .. ...... ... . . .. ... .. .... . .. .. .. .. . . ...... . .. . . . . ... .. . .... ........... . .. ... . . ... .... .. . . . . .. ... . . .. . and its direction is perpendicular to the plane deﬁned by r and p. . .. . ... .. .... .. ... .. .... ..... . . . .. . .. . . ... . .. . . . . ..... . .. ... ..... .. . . . .... .. ...... . ....58) 465 o ωr .. .. . . ..... .. .... .. .. .. ... . . . . .... . ...... ... .. .. .. .. . .29 (with v replaced by p).. . ...... . . ... . ...... . . . .. . .. .... .. ... ... d •o v θ r .. . .. This direction is not in general the same as that of the angular velocity y.. .... .. ... .. . ..30 is l1=1pd1=1pr1sin1θ. . ... .. equation (16....... .... ..28 .. . .. . .. . .. . .. . ... ..... ..... . . .59) (16... . . . ....

y. 0 Exercises 45–47 16. The general relation is l1=1Iω (16. y. v1=1v(r)1=1v(x.466 Chapter 16 Vectors where I1=1mr 2 is the moment of inertia about the axis of rotation.9. Newton’s second law of motion can be written as dp F= .61) for arbitrary motion of a particle or for the rotation of a rigid body. y.62) where I is a quantity with nine components called the moment of inertia tensor (or matrix). A vector function of position. Therefore. 0 Exercise 44 Conservation of angular momentum As discussed in Example 16. using the product rule of differentiation (valid for both scalar and vector products). a vector ﬁeld. z). shows that the angular momentum of a system is constant in the absence of external torques. z) (16. and is discussed in Example 19. dt T= dl dt (16. a scalar ﬁeld.63) This form of Newton’s second law. Examples of scalar ﬁelds have been discussed in Chapter 10. The relation between l and y is less simple than that given by (16. The right side is. Taking the cross product of r with both sides of this equation gives dt r 1z1 F = r 1z1 dp dt The left side is the torque T acting on the system (Example 16. since r1z1p1=1l.7 Scalar and vector ﬁelds A function of the coordinates of a point in space is called a function of position or ﬁeld.15. A scalar function of position.16). that the rate of change of angular momentum of a system is equal to the applied torque. r 1z1 But dp d dr = ( r 1z1 p ) − 1z1 p dt dt dt dr = v is parallel to p. f1=1f(r)1=1f(x.65) .64) has a value at each point r1=1(x. that is. z) (16. and their vector product is zero. a scalar (a number) is associated with each point.

is discussed in the following section. We have ∂V ∂V ∂V = 1. The concept of the gradient of a scalar ﬁeld.8 The gradient of a scalar ﬁeld The gradient of a scalar function of position f1=1f(x. 16.66) This quantity can be interpreted as the result of operating on the function f with the vector differential operator ∇= (read as ‘del’ or ‘nabla’). ∂x ∂y ∂z ∂ ∂ ∂ i+ j+ k ∂x ∂y ∂z (16.9. with some applications. z) is deﬁned as the vector grad f = ∂f ∂f ∂f i+ j+ k ∂z ∂x ∂y (16. An example of a vector ﬁeld is the velocity ﬁeld used to describe ﬂuid ﬂow in hydrodynamics. and only a very brief description is given here.16. y. a velocity is associated with every point. Electric and magnetic ﬁelds are vector ﬁelds.19 Find grad V for V1=1x1+12yz1+13z2. The theory of vector ﬁelds is an essential tool in hydrodynamics and electromagnetism.68) The symbol ∇ was introduced by Hamilton. Therefore ∇V1=1i1+12zj1+1(2y1+16z)k. = 2 z . and k are assumed to be constant vectors). and = 2 y + 6 z.67) (16. The divergence and the curl of a vector ﬁeld are introduced in Section 16.8 The gradient of a scalar ﬁeld 467 deﬁnes a vector associated with each point r. These quantities are used in more advanced applications of the vector calculus in the theory of ﬁelds. and (possibly) called nabla after a harp-like musical instrument used in Palestine in Biblical times. 5 . j. so that ∂ ∂ ∂ grad f = ∇f = i + j + k f ∂y ∂z ∂x ∂f ∂f ∂f = i+ j+ k ∂x ∂y ∂z (i.5 EXAMPLE 16. A basic feature of vector ﬁeld theory is the use of vector differential operators to describe how a ﬁeld changes from point to point in space.

.. ... .. ... .. .. . . ... ... . and acts along the line from q1 to q2 for like charges.20 Force and potential energy By a generalization to three dimensions of the discussion of conservative forces in Section 5. .. 1 q The potential energy of interaction of charges q1 and q2 separated by distance r is V1=1q1q224πε0r (see Example 5. Therefore ˆ r F= q1q2 r 4 πε0 r 3 = q1q2 4 πε0 r 2 q1q2 r ˆ . . .. ...13). .... .. . . .. . 4 πε0 r 2 from q2 to q1 for unlike charges (see also Example 5. . (16.. . .19).468 Chapter 16 Vectors EXAMPLE 16. .. ..69) qq F = −∇V = − 1 2 4 πε0 = ∂ ∂x 1 ∂ 1 ∂ 1 i+ j+ ∂y r ∂z r r k Figure 16.31 q1q2 x qq y z 3 i + 3 j + 3 k = 1 2 3 ( x i + y j + z k) 4 πε0 r r r 4 πε0 r q1q2 r 4 πε0 r 3 = r The unit vector from q1 to q2 is r = . . .. . Fx = − ∂V .. . . ..7 (see also Example 16.. . .. . .21 Coulomb forces 2 q •. .70) .. the components of a conservative force are derivatives of a potential-energy function V. .. .... . . . . Then The force has strength . .. ... . . . . ∂V ∂V ∂V F = −∇V = − i+ j+ k ∂y ∂z ∂x 0 Exercises 48–50 EXAMPLE 16..... ∂y Fz = − ∂V ∂z The force is therefore (minus) the gradient of V. . .. In addition. r . ... .. . .... . .... . .. . .. ... .. If r is the position of q2 relative to q1 (Figure 16.. . .... . . . . . . the force per unit charge acting at point r due to the presence of charge q1 is E1=1F2q2.. ∂x Fy = − ∂V . ... . ..... .. .... . ..17)... .. .. . . . •. ..... ...31) then the force acting on q2 due to the presence of q1 is (16...

74) The quantity ∇f is therefore the generalization to three dimensions of the ordinary derivative df2dx. the vector ﬁeld is the velocity ﬁeld of a ﬂuid. y. is the total differential df = ∂f ∂f ∂f dx + dy + dz ∂z ∂x ∂y (16. the vector operator ∇ is the generalization of the differential (gradient) operator d2dx.72) The corresponding change in the function.68) and (16.72)). dy. of ﬂuid out of a volume element dV at the point.9 Divergence and curl of a vector ﬁeld 469 E= q1 4 πε0 r 2 r = −∇φ ˆ (16. 16.73) This can be written as the scalar product of the gradient ∇f and the displacement dr (Equations (16. the value of div1v(r) dV at a point r is a measure of the net ﬂux. and is sometimes written as d2dr. and φ1=1V2q21=1q124πε0r is the corresponding (scalar) electrostatic potential ﬁeld. df1=1f(r1+1dr)1−1f(r). We consider a differential (inﬁnitesimal) displacement r1→1r1+1dr or.16. The meaning of the gradient of a function of position f(r) is clariﬁed by considering how the value of the function changes from point to point in space.71) is the electrostatic ﬁeld of charge q1.75) In hydrodynamics. dz)1=1idx1+1jdy1+1kdz (16. z1+1dz): dr1=1(dx.9 Divergence and curl of a vector ﬁeld The divergence The divergence of a vector ﬁeld v(r) is deﬁned as the scalar quantity div v = ∇ · v = ∂v x ∂v y ∂v z + + ∂x ∂y ∂z (16. z)1→1(x1+1dx. or rate of ﬂow. (x. . y1+1dy. ∂f ∂f ∂f ∂f ∂f ∂f j + k 1·1( i dx + j dy + k dz ) = dx + dy + dz i+ ∂y ∂z ∂x ∂y ∂z ∂x so that df1=1∇f1·1dr (16. and the divergence is a ﬂux density. in cartesian components.

Then ∂ ∂f ∂ ∂ ∂f ∂f div v = ∇ · ∇f = i + j + k · i + j + k ∂z ∂y ∂z ∂x ∂y ∂x = ∂2 f ∂ x2 + ∂2 f ∂ y2 + ∂2 f ∂z2 (16. The curl The curl (or rotation rot1v) of a vector ﬁeld v(r) is deﬁned as the vector ∂v ∂v ∂v ∂v y ∂v ∂v i+ x − z j+ y − x k curl v = ∇ 1z1v = z − ∂z ∂y ∂x ∂z ∂x ∂y i ∂ = ∂x vx j ∂ ∂y vy k ∂ ∂z vz (16. On the other hand. the curl of the velocity ﬁeld at a point is a measure of the circulation of ﬂuid around the point. When a scalar ﬁeld f(r) satisﬁes Laplace’s equation ∇2f1=10 at a point then the derived vector ﬁeld ∇f has zero divergence at the point.52) and (16. 0 Exercises 51–55 . v1=1∇f. B1=1∇1z1A.79) Thus. curl1v1=10 if v1=1grad1f (16. In hydrodynamics. This is the case for an incompressible ﬂuid in a region containing no sources (of ﬂuid) or sinks and for an electrostatic ﬁeld in a region free of charge. A vector ﬁeld that is the gradient of a scalar ﬁeld has zero curl. so that ∇1z1E1=10. E1=1−∇φ.78) Thus. div1v1=10 if v1=1curl w (16. where A is called the magnetic vector potential. the electrostatic ﬁeld is the gradient of the scalar electrostatic potential.53)).470 Chapter 16 Vectors We consider here only the special case of a vector ﬁeld that is the gradient of a scalar ﬁeld. a vector ﬁeld that is itself the curl of a vector ﬁeld has zero divergence. the magnetic ﬁeld can always be expressed as the curl of a vector function. and there is no net ﬂux out of a volume element at the point.76) It follows that ∇1·1∇1=1∇2 is the Laplacian operator.77) (see Equations (16. Then ∇1·1B1=10.

a1·1b1=1a1b11+1a2b21+1a3b31+ . an inner (scalar) product is associated with each pair of vectors. 1. 0.10 Vector spaces The vectors discussed in this chapter are three-dimensional vectors. 0). =.81) These quantities obey the rules of vector algebra. and d. Express the following quantities in terms of a.11 Exercises 471 16. 16.2 and 16. 0. y1. 0). and d: (i) AB and BA . a3) with the given initial point P(x1. Two vectors are orthogonal (‘perpendicular’) if their inner product is zero. (ii) the length of a.+1anbn1=1b1·1a (16. b. Find BC Find (i) the vector a1=1(a1.+1anen (16. Section 16. Two sides of the triangle ABC are AB = ( 2. P(−3. z1) and terminal point Q(x2. In particular. Q(4. −3. 0. =. 3. ein neuer Zweig der Mathematik (The theory of linear extension.82) The inner product is often denoted by (a.1b). a new branch of mathematics) of 1862. 0. 2) 5. (iv) the position of an arbitrary point on the line BC (the equation of the line). The work included the algebras of Hamilton’s quaternions and Gibbs’ vectors as special cases. German mathematician. z2). B. −2. in his Die lineale Ausdehnungslehre. =. c.6 For the vector space so deﬁned. 0. e11=1(1. . an)1=1a1e11+1a2e21+1a3e31+ . 1) 6 A detailed treatment of n-dimensional vector spaces was given by Hermann Günther Grassmann (1809–1877). P(0. c. P(1. The concept of vector can be extended to any number of dimensions by deﬁning vectors in n dimensions as quantities that have n components and that obey the laws of vector algebra described in Sections 16. 1) (16. or vectors in a three-dimensional vector space. y2. 0) . (ii) the position of the centroid of the points.83) A vector space with these properties is called an inner product space. Q(−1.80) Every vector in the space can then be expressed as a linear combination of these. an n–dimensional vector space can be deﬁned by means of n orthogonal unit vectors. C and D are a. 0) en1=1(0. =. 1).3 2. a2. 2. a1=1(a1. b.2 1. a3. 2. −1) and AC = (3. (iii) the position of the midpoint of BC . (iii) the unit vector parallel to a. Q(2. 2. 3. 0) e21=1(0.3. a2. but was largely unknown during his lifetime. and the length or norm of a vector is 2 2 2 | a | = a 1·1a = a1 + a2 + a3 + 2 + an (16. 0) 4. 1.16.11 Exercises Section 16. 0. The position vectors of the points A. The particular space described in this section is the n–dimensional Euclidean space R n.

3a1+12b1−13c 9. r21=1(0. respectively. a1z1c1+1c1z1a . 3a. a1z1b. Find f such that f. a1+1b. Find the value of λ for which a1=1(λ. −1. Section 16. 0) to r21=1(2. 3. Find (i) the position vector of the centre of mass. r21=1(2. r31=1(0. f11=12i1−13j1+1k and f21=12j1−1k are in equilibrium. Section 16. a1=1i. 3). a1·1c1−1b1·1c 21. (i) Find the velocity and acceleration at time t. Forces are said to be in equilibrium if the total force is zero. Calculate the work done (i) by the force on the body when d1=12i1−1j. 2. −2. c1=1−k. (iv) in the xy-plane. 3). 1). For 19. 3. 1).directions. 16.directions. c1=1(1. b1+1a 7. (v) in the z-direction. 12. m31=11. and the applied electric ﬁeld E1=1−2k. b1·1a Show that a1=1(1. y-. For the vectors a1=12i1−1j1+13k. (ii) the position vectors of the masses with respect to the centre of mass. (iv) in the y-direction. have position vectors r11=1(2. (v) overall. a1z1c1+1b1z1c 31. 1). A body undergoes a displacement from r11=1(0. c1=1(0. q21=1−3 and q31=11 at positions r11=1(3. 1.472 Chapter 16 Vectors For a1=1(1. Find (i) the dipole moment of the system of charges with respect to the origin. (iii) by the body against the force when d1=12k. (iii) a vector perpendicular to b and c. 3. 23. 28. 3. 2). a23 8. −3. 1. 3(a1−1c) 10. (ii) Find the potential-energy function V(r) of which the components of the force are (–) the partial derivatives. ﬁnd (i) x1=12a1+1b1+14c. a1·1c a1·1b Find the angles between the direction of a = i − j + 2 k and the x-. 3a1−13c. ﬁnd: 32. −4. −3). −2). 30. Describe the motion of the body (iii) in the x. 1). −2). −2). −1. 2. −1). b1=1(−2. 13.6 For a1=1(1. 27. ﬁnd: 20. (ii) Find the force acting on the body. 1. 2) and c1=1(−13. (ii) the position of the point with respect to which the dipole moment is zero. For 24. A body undergoes the displacement d under the inﬂuence of a force f1=13i1+12j. c1=12i1−13j1+1k. −2. 2ti1+13t 2j 1 17. A body of mass m moves along the curve r(t)1=1x(t)i1+1y(t)j. respectively. (a1z1c)1·1b 36. 1) under the inﬂuence of the conservative force F1=1xi1+12yj1+13zk. (i) Calculate the work W(r11→1r2) done on the body. c1=1(0. a1z1(b1z1c). 3. b1·1c 26. ﬁnd 6. where x1=1at and y1=1(at1− 2 gt 2) at time t. b1=1(0. Three charges. (i) Find the velocity and acceleration at time t. (ii) by the body against the force when d1=1i1−13k. (a1·1c)b a1·1b. 14. 3). m21=13. −2. b1=12j1−12k. b1z1c. q11=13. 22. r21=1(2. respectively.and y. (vi) overall. b1z1a 35. ﬁnd 25. where x1=121cos13t. 1). |a1+1b|. (ii) a vector perpendicular to c and x. −a. 2. |b1z1c| 33.5 a1=1(1. 1) and b1=1(2. 2. m11=12. (a1−1b)1·1c. −4). 29.4 Differentiate with respect to t. b1=1(0. have position vectors r11=1(3. 31sin1t. Section 16. 3. A body of mass m moves along the curve r(t)1=1x(t)i1+1y(t)j1+1z(t)k. 3) are orthogonal vectors. 0. (ii) Find the force acting on the body. y1=121sin13t and z1=13t at time t. r31=1(0. −1) are orthogonal. (a1z1b)1z1c 34. 2) and r31=1(0. −3). b1=1j. 1). and z. 0). q21=1−2. −1. (a1+1b)1z1c. |a | 1+1|b| 11. Calculate the energy of interaction between the system of charges q11=12. Describe the motion of the body (iii) in the x-direction. 4. 2. b1=1(0. 18. Three masses. 2t) 15. q31=11. (iii) Conﬁrm that W(r11→1r2)1=1V(r1)1−1V(r2). (cos12t. 3.

in the electric ﬁeld E1=1−k. (iii) Find the angular momentum of the particle in terms of ω. ﬁnd the total angular momentum l1=1l11+1l2. f1=1xy1+1zx1+1yz 48. 45. 0). . −2). 1. −1). r21=1(0. The total angular momentum of a system of particles is the sum of the angular momenta of the individual particles.58) and (16. y(t)1=1R1sin1ω t. 3). Use the property of the triple vector product (Exercise 39) to derive equation (16. 2) (iii) F1=1(1. 38. The force F acts on a line through the point A. 1). 2). x. q21=1−3 and q31=11 at positions r11=1(3. 53. 1. 0) (ii) F1=1(0. and r31=1(0. 0). 0. Show that div1v1=10 if v1=1curl1w. Find the moment of the force about the point O for (i) F1=1(1.11 Exercises 473 37. 2). Calculate the torque experienced by the system of charges q11=12. f1=1(x 21+1y 21+1z 2)−122 Section 16. and z. v1=1xi1+1yj1+1zk 52. If the system in Exercise 45 is replaced by a system of two particles of mass m with positions r11=1x(t)i1+1y(t)j1+1zk and r21=1−x(t)i1−1y(t)j1+1zk. show that l1=1Iy when z1=10. 1. z1=1constant. 41. 44. 47.59). −2). O(1. 6). 3. (iv) Conﬁrm equation (16. 3. (5. 3. 5). Show that a1z1b is orthogonal to a and b. A charge q moving with velocity v in the presence of an electric ﬁeld E and a magnetic ﬁeld B experiences a total force F1=1qE1+1qv1×1B called the Lorentz force. 1. 1). 3) 42. For the system in Exercise 45. 2).60) in this case.60) from equations (16. (4. 2. 0).9 Find div1v and curl1v for 51. v1=1yzi1+1zxj1+1xyk 55. O(1. −3. O(0. c1=1(0. Show that curl v1=10 if v1=1grad1f. where I is the moment of inertia about the axis of rotation. This example demonstrates that l1=1Iy when the axis of rotation is an axis of symmetry of the system. −2. Calculate the force acting on the charge q1=13 moving with velocity v1=1(2. 0. 43. 0. −1. (i) By expanding in terms of components. 0).16. Find the area of the parallelogram whose vertices (in the xy-plane) have coordinates (1. The quantity a1z1(b1z1c) is called a triple vector product. (i) What is the angular velocity y about the z-axis? (ii) Find the velocity v of the particle in terms of ω. Section 16. and z. x. and show that l1=1Iy where I is the total moment of inertia about the axis of rotation. respectively. The position of a particle of mass m moving in a circle of radius R about the z-axis with angular speed ω is given by the vector function r(t)1=1x(t)i1+1y(t)j1+1zk where x(t)1=1R1cos1ω t. b1=1(0. The quantity a1·1(b1z1c) is called a triple scalar product. Show that ax (i) a1·1(b1z1c)1=1c1·1(a1z1b)1=1b1·1(c1z1a) (ii) a · (b z c ) = bx cx ay by cy az bz cz (determinant ) 39. A(2. (8. f1=12x 21+13y 21−1z 2 50. 0. 1). A(1. 1) in the presence of the electric ﬁeld E1=12i and magnetic ﬁeld B1=13j. 46. v1=1zi1+1xj1+1yk 54. show that a1z1(b1z1c)1=1(a1·1c)b1−1(a1·1b)c (ii) Conﬁrm this formula for the vectors a1=1(1.8 Find the gradient ∇f for 49. A(0. y. 40. y. 0.

and c2 are constants. Thus.2) (17.1 Concepts Many problems in the physical sciences.1) 1 The earliest descriptions of the method of solving sets of linear equations by determinants.3) (17. The branch of mathematics concerned with the theory of linear systems is matrix algebra. Some of the practical methods of solution. c1. The concept of determinants has its origin in the solution of simultaneous linear equations. They are used in quantum chemistry to construct electronic wave functions that are consistent with the requirements of the Pauli Exclusion Principle. we multiply equation (1) by b2 and equation (2) by b1 to give (1′) a1b2 x1+1b1b2 y1=1c1b2 (2′) b1a2 x1+1b1b2 y1=1b1c2 so that. (a1b21−1b1a2) y1=1a1c21−1c1a2 (17. as discussed in Section 2. The theory of determinants is discussed in this chapter as a separate topic. to solve for x. the subject of Chapters 18 and 19.17 Determinants 17. and alternative methods are then required both for the numerical solution of the large ‘linear systems’. and can be solved by the elementary methods of algebra. partly in preparation for the more general matrix algebra of Chapters 18 and 19. and for the formulation and theoretical analysis of the problems that give rise to them. (a1b21−1b1a2) x1=1c1b21−1b1c2 Similarly for y. and partly because determinants have certain symmetry properties that have made them an important tool in quantum mechanics.1 We consider the pair of equations (1) a1x1+1b1 y1=1c1 (2) a2 x1+1b2 y1=1c2 where a1. b1. b2. The equations are linear in the ‘unknowns’ x and y. subtracting (2′) from (1′). . The ﬁrst published account appeared in MacLaurin’s Treatise of algebra (posthumously in 1748). in engineering. but several of the more important and useful results in the theory of linear equations can be derived independently from quantities called determinants. In some cases however the number of equations can be large. the ‘numerical methods’. two or three. known as Cramer’s method. and by Leibniz in a letter to l’Hôpital in 1693 (published in 1850) in which he also gave the condition for the consistency of the equations.8. The methods of elementary algebra are adequate when the number of such equations is small. and in statistics give rise to systems of simultaneous linear equations. were by the Japanese Seki Kowa (1642–1708) in a manuscript of 1683. a2. are discussed in Chapter 20.

the required (unique) values of x and y are x= c1 b2 − b1 c2 .6) EXAMPLE 17. a1 b2 − b1 a2 y= a1 c2 − c1 a2 a1 b2 − b1 a2 (17.n) to represent the array with which a determinant is associated.4) The quantity in the denominators of (17.1 Concepts 475 When a1b21−1b1a2 is not zero. He introduced the concept of minors and the expansion along any row or column. D y= D2 D (17.2 the expression on the right side deﬁnes its value.4) is a property of the coefﬁcients in equations (17.4) of the system of equations (17.1) can now be written as x= where D= a1 a2 b1 b2 . 5 D2 = 2 1 5 = 2 × 9 − 5 × 1 = 13 9 Therefore x1=1D12D1=14 and y1=1D22D1=11. and used the abbreviation (a1.17. In the general case.1). and for the quantity now called the Jacobian.5) The symbol on the left is called a determinant. D2 = a1 a2 c1 c2 (17. he arranged the n2 different quantities in a square array. 0 Exercises 1–4 2 The name determinant was coined by Cauchy in 1812 in the ﬁrst of a long series of papers on the subject of a class of alternating symmetric functions such as a1b21−1b1a2. Cauchy used determinants for problems in geometry and in physics.1 Use determinants to solve the equations 2x1−13y1=15 x1+15y1=19 We have D= D1 = 2 1 5 9 −3 = 2 × 5 − ( −3) × 1 = 13 5 −3 = 5 × 5 − ( −3) × 9 = 52. and is written in the form a1 a2 b1 b2 = a1 b2 − b1 a2 (17. D1 = c1 c2 b1 b2 .7) D1 . . The solution (17.

x1.9) The system of three linear equations in three unknowns.476 Chapter 17 Determinants The determinant (17. for example. a23 lies in the second row and third column.2 Determinants of order 3 (1) a11x11+1a12 x21+1a13 x31=1b1 (2) a21x11+1a22 x21+1a23 x31=1b2 (3) a31x11+1a32 x21+1a33 x31=1b3 (17. and is written a11 a21 a31 an1 a12 a22 a32 an 2 a13 a23 a33 an3 a1n a2 n a3n ann (17. the element aij lies in the ith row and jth column of the array. Determinants of order n arise from the consideration of systems of n linear equations in n unknowns.1). The determinant is sometimes denoted by the symbol | aij |. x2 and x3.8) In this notation. The coefﬁcient of x1 in the resulting expression deﬁnes the determinant of order 3: a11 D = a21 a31 = a11 a12 a22 a32 a22 a32 a13 a23 a33 a23 a33 − a21 a12 a32 a13 a33 + a31 a12 a22 a13 a23 (17. 17. In the general case. can be solved for x1 by multiplying equation (1) by a22a331−1a23a321= a22 a32 a12 a32 a12 a22 a23 a33 a13 a33 a13 a23 (2) by −(a12a331−1a13a32)1=1− (3) by a12a231−1a13a221= and adding the three results. It is a determinant of order 2.10) . a determinant of order n is a property of a square array of n 2 elements.5) is a property of a square array of 41=122 elements. the coefﬁcients of x and y in the system of equations (17.

y1=1D22D1=12.17.2 Determinants of order 3 477 and. z1=1D32D1=13. and b1 D1 = b2 b3 a12 a22 a32 a13 a23 .12) where D1≠10 is the determinant of the coefﬁcients. D2 = 0 2 1 8 −7 11 4 2 −3 = 42. D2.10).13).11) The solution of the system of three equations can then be expressed in terms of third-order determinants as x1 = D1 . expanding the second-order determinants. 8 D1 = −7 11 −3 1 2 4 2 −3 = 21. 2 D= 0 1 −3 1 2 4 1 −3 = 2 2 2 −3 −3 −0 2 2 4 −3 +1 2 1 4 −3 =121×181−101×1(−14)1+111×151=121 The determinants D1. D x3 = D3 D (17. a33 a11 D2 = a21 a31 b1 b2 b3 a13 a23 . by equation (17.2 Use determinants to solve the equations 2x1−13y1+14z1=18 y1−13z1=1−7 x +12y1+12z1=111 The determinant of the coefﬁcients is.13) EXAMPLE 17. D3 = 0 2 1 −3 1 2 8 −7 = 63 11 Therefore x1=1D12D1=11. D x2 = D2 . by equations (17. and D3 are.10)