MST124

Essential mathematics 1

Book C
This publication forms part of an Open University module. Details of this and other
Open University modules can be obtained from the Student Registration and Enquiry Service, The
Open University, PO Box 197, Milton Keynes MK7 6BJ, United Kingdom (tel. +44 (0)845 300 6090;
email [email protected]).
Alternatively, you may visit the Open University website at www.open.ac.uk where you can learn
more about the wide range of modules and packs offered at all levels by The Open University.
To purchase a selection of Open University materials visit www.ouw.co.uk, or contact Open
University Worldwide, Walton Hall, Milton Keynes MK7 6AA, United Kingdom for a brochure
(tel. +44 (0)1908 858779; fax +44 (0)1908 858787; email [email protected]).

The Open University, Walton Hall, Milton Keynes, MK7 6AA.

First published 2014.
c 2014 The Open University
Copyright !
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, transmitted
or utilised in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without
written permission from the publisher or a licence from the Copyright Licensing Agency Ltd. Details of such
licences (for reprographic reproduction) may be obtained from the Copyright Licensing Agency Ltd, Saffron
House, 6–10 Kirby Street, London EC1N 8TS (website www.cla.co.uk).
Open University materials may also be made available in electronic formats for use by students of the
University. All rights, including copyright and related rights and database rights, in electronic materials and
their contents are owned by or licensed to The Open University, or otherwise used by The Open University as
permitted by applicable law.
In using electronic materials and their contents you agree that your use will be solely for the purposes of
following an Open University course of study or otherwise as licensed by The Open University or its assigns.
Except as permitted above you undertake not to copy, store in any medium (including electronic storage or
use in a website), distribute, transmit or retransmit, broadcast, modify or show in public such electronic
materials in whole or in part without the prior written consent of The Open University or in accordance with
the Copyright, Designs and Patents Act 1988.
Edited, designed and typeset by The Open University, using the Open University TEX System.
Printed in the United Kingdom by Halstan & Co. Ltd., Amersham, Bucks.

ISBN 978 1 7800 7359 0

1.1
 Contents

Contents
Unit 7 Differentiation methods and integration 1

Introduction 3
1 Derivatives of further standard functions 3
1.1 Derivatives of sin, cos and tan 4
1.2 Derivatives of exp and ln 10
2 Finding derivatives of more complicated functions 14
2.1 Product rule 14
2.2 Quotient rule 19
2.3 Chain rule 23
2.4 Differentiating functions of linear expressions 31
2.5 Using the differentiation rules together 33

3 More differentiation 38
3.1 Optimisation problems 38
3.2 Differentiation using a computer 44
3.3 Derivatives of inverse functions 46
3.4 Table of standard derivatives 54

4 Reversing differentiation 55
4.1 Antiderivatives and indefinite integrals 56
4.2 Antiderivatives of power functions 61
4.3 Constant multiple rule and sum rule for
antiderivatives 64

5 Particular antiderivatives 67
5.1 Finding particular antiderivatives 67
5.2 Using integration to work with rates of change 68
5.3 Changes in the values of antiderivatives 74
6 Antiderivatives of further standard functions 78
6.1 An antiderivative of the reciprocal function 78
6.2 Table of standard indefinite integrals 81
Learning outcomes 83

Solutions to activities 84

Acknowledgements 101
Contents

Unit 8 Integration methods 103

Introduction 105

1 Areas, signed areas and definite integrals 106

1.1 Areas and signed areas 106
1.2 Definite integrals 119

2 The fundamental theorem of calculus 126

2.1 What is the fundamental theorem of calculus? 127
2.2 Using the fundamental theorem to find signed areas 133
2.3 Using the fundamental theorem to find changes in the
values of antiderivatives 146
2.4 Notation for indefinite integrals 151

3 Integration by substitution 155

3.1 Basic integration by substitution 156
3.2 Integration by substitution in practice 162
3.3 Integrating functions of linear expressions 167
3.4 Integration by substitution for definite integrals 172
3.5 Finding more complicated integrals by substitution 174
4 Integration by parts 176
4.1 Basic integration by parts 176
4.2 Integration by parts in practice 180
4.3 Integration by parts for definite integrals 185

5 More integration 189

5.1 Trigonometric integrals 189
5.2 Choosing a method of integration 192
5.3 Integration using a computer 194

Learning outcomes 196

Solutions to activities 197

Acknowledgements 216
 Contents

Unit 9 Matrices 217

Introduction 219

1 Matrices and matrix operations 221

1.1 Matrix notation and terminology 221
1.2 Matrix addition and subtraction 223
1.3 Scalar multiplication of matrices 226
1.4 Matrix multiplication 228

2 Matrices and networks 238

2.1 Networks 239
2.2 From networks to matrices 242
2.3 Combining networks 245
3 The inverse of a matrix 251
3.1 Identity matrices 251
3.2 Matrix inverses 253

4 Simultaneous linear equations and matrices 257

4.1 Solving simultaneous linear equations in two
unknowns 258
4.2 Non–invertible coefficient matrices 262
4.3 Simultaneous linear equations in more than two
unknowns 265

Learning outcomes 269

Solutions to activities 270

Acknowledgements 278

Index 279
Unit 7

Differentiation methods and integration

 Introduction

Introduction
The first half of this unit continues with the topic of Unit 6, differential
calculus. You’ll meet the formulas for the derivatives of five standard
functions, namely sin, cos, tan, exp and ln. You’ll also learn some further
ways of differentiating complicated functions by combining the formulas
for the derivatives of standard functions.
By the end of this part of the unit you should have acquired skills in
differentiating a wide variety of functions. You’ll need these skills in the
later parts of the module, and in any further modules that you study that
involve calculus. So it’s important that you become proficient in them.
The only way to do that is to practise, so you should aim to do all the
activities in this unit, and as many further activities in the practice quizzes
and exercise booklets as you have time for, until you feel confident with
your skills.
In the first half of the unit you’ll also learn how to use your computer to
find and work with derivatives, and you’ll see some further applications of
differentiation.
In the second half of the unit you’ll start to learn about integral calculus,
which can be thought of as the reverse of differential calculus. You’ve seen
that in differential calculus you start off knowing the values taken by a
continuously changing quantity throughout a period of change, and you
use this information to find the values taken by the rate of change of the
quantity throughout the same period. In integral calculus, you do the
opposite: you start off knowing the values taken by the rate of change of a
quantity throughout a period of change, and you use this information to
find the values taken by the quantity throughout the same period. This
process is called integration. The topic of integral calculus continues in
Unit 8.

1 Derivatives of further standard

functions
In this section you’ll meet the formulas for the derivatives of the five
standard functions sin, cos, tan, exp and ln.

3
Unit 7 Differentiation methods and integration

1.1 Derivatives of sin, cos and tan

In this first subsection we’ll obtain formulas for the derivatives of the
trigonometric functions f (x) = sin x, f (x) = cos x and f (x) = tan x.
When you’re working with these functions, it’s essential to remember that
the input variable x is the size of an angle measured in radians. This is
indicated by the formulas for the functions, because x represents a
number, and when the size of an angle is given just as a number, with no
units, then the units are assumed to be radians. You met this convention
in Unit 4. For example, sin 10 denotes the sine of 10 radians, whereas
sin 10◦ denotes the sine of 10 degrees.
When we’re dealing with the derivatives of trigonometric functions, we
always measure angles in radians. This is because it makes the formulas
for the derivatives of trigonometric functions simpler. If you’d like to know
more about this, then, once you’ve completed Sections 1 and 2 of this unit,
read the document Why do we use radians? on the module website.

Derivative of sin
Our first aim in this subsection is to obtain a formula for the derivative of
the sine function, f (x) = sin x.
As you know, the derivative of a function is itself a function, and therefore
the derivative has a graph. So, to get a rough idea of what the derivative
of the sine function might be like, let’s try to roughly sketch its graph, by
estimating the gradient of the graph of the sine function at various points.
Figure 1 shows the graph of the sine function, drawn on axes with equal
scales, and some tangents to the graph.

Figure 1 The graph of f (x) = sin x

When x = 0, the gradient of the graph of the sine function seems to be
about the same as the gradient of the line y = x; that is, it seems to be
about 1. As x increases, the graph gradually gets less steep – that is, its
gradient gradually decreases, until x = π/2, when the gradient seems to be
zero. The gradient decreases slowly when x is only a little larger than 0,
but decreases more rapidly as x gets closer to π/2.
In other words, for x between 0 and π/2 the value of the derivative seems
to decrease from 1 to zero, slowly at first, but then more rapidly. So its
graph must look something like the sketch in Figure 2.

4
 1 Derivatives of further standard functions

Figure 2 A sketch of the graph of the derivative of f (x) = sin x, for x

between 0 and π/2
Now look at the graph of the sine function for x between π/2 and π. The
gradient starts at zero, then it becomes negative. At first the graph is not
very steep – that is, the gradient has small negative values – but it
becomes steeper and steeper – that is, the gradient takes negative values of
greater and greater magnitude. Eventually, when x = π, the gradient
seems to be about the same as the gradient of the line y = −x; that is, it
seems to be about −1. The gradient decreases fairly rapidly when x is only
a little larger than π/2, but decreases more slowly as x gets closer to π.
So, for x between π/2 and π, the value of the derivative seems to decrease
from 0 to −1, fairly rapidly at first, but then more slowly. So the graph of
the derivative must look something like the sketch in Figure 3.

Figure 3 A sketch of the graph of the derivative of f (x) = sin x, for x

between 0 and π
In the next activity you’re asked to extend the sketch in Figure 3 to cover
more values of x.

Activity 1 Extending the sketch of the graph of the derivative of sin

By considering the gradients of the graph of the sine function (shown in

Figure 1) for x between π and 3π/2, and then for x between 3π/2 and 2π,
work out what the graph of the derivative of the sine function must look
like for these values of x. Hence extend the sketch of the derivative of the
sine function in Figure 3 to cover all values of x between 0 and 2π.

5
Unit 7 Differentiation methods and integration

In Activity 1 you should have worked out that the graph of the derivative
of the sine function for x between 0 and 2π must look something like the
sketch in Figure 4.

Figure 4 A sketch of the graph of the derivative of f (x) = sin x, for x

between 0 and 2π
Since the graph of the sine function repeats every 2π units, its gradients
also repeat every 2π units. So you can now extend the sketch of the graph
of the derivative to cover any interval of values of x that you wish. For
example, Figure 5 shows a sketch of the graph for values of x between −π
and 3π.

Figure 5 A sketch of the graph of the derivative of f (x) = sin x, for x

between −π and 3π
The graph in Figure 5 should look familiar to you – it looks very like the
graph of the cosine function! So it seems possible that the derivative of the
sine function is the cosine function – in other words, that
if f (x) = sin x, then f " (x) = cos x.

This is indeed true. It can be confirmed by using differentiation from first

principles, but this is quite tricky to do. If you’d like to see some details
about how it’s done, then have a look at the document The derivative of
the sine function on the module website.

6
 1 Derivatives of further standard functions

Derivative of cos
Let’s now look at obtaining a formula for the derivative of the cosine
function, f (x) = cos x. You could get a rough idea of what this derivative
must be like by using a similar procedure to the one that you’ve just seen
for the sine function. Figures 6 and 7 show the graph of f (x) = cos x, and
the sketch of its derivative that’s obtained from this procedure.

Figure 6 The graph of f (x) = cos x

Figure 7 A sketch of the graph of the derivative of f (x) = cos x

The graph in Figure 7 looks like the graph of the sine function, reflected in
the x-axis. That is, it looks like the graph of y = − sin x. So it seems
possible that the derivative of the cosine function is the negative of the
sine function – in other words, that
if f (x) = cos x, then f " (x) = − sin x.
Again, this is indeed true. One way to confirm it is to use differentiation
from first principles, but, as with the sine function, this is a tricky process.
A simpler way to confirm it is to deduce it from the fact that the
derivative of the sine function is the cosine function. As you know, you can
obtain the graph of the cosine function by translating the graph of the sine
function by π/2 to the left. It follows that you can obtain the graph of the
derivative of the cosine function by translating the graph of the derivative
of the sine function by π/2 to the left. (This is because the two derivatives
give the gradients of the two graphs.)

7
Unit 7 Differentiation methods and integration

The derivative of the sine function is the cosine function, whose graph is
shown in Figure 8. Translating this graph by π/2 to the left gives the
graph in Figure 9. Because of the symmetries of the graphs of sine and
cosine, this graph is the graph of y = − sin x. So the derivative of
f (x) = cos x is indeed f " (x) = − sin x.

Figure 8 The graph of y = cos x

Figure 9 The graph of y = cos x translated by π/2 to the left

Derivative of tan
The final derivative of a trigonometric function that you’ll meet in this
subsection is the derivative of the tangent function. In fact,
if f (x) = tan x, then f " (x) = sec2 x.
Remember that sec2 x means (sec x)2 , and also that sec x means 1/(cos x).
So another way to write the derivative of f (x) = tan x is
1
f " (x) = .
(cos x)2
However, we usually use the form f " (x) = sec2 x, for conciseness.
This formula for the derivative of the tangent function can be worked out
by using the fact that tan x = (sin x)/(cos x), together with the formulas
for the derivatives of f (x) = sin x and f (x) = cos x. The process involves a
rule for combining derivatives that you’ll meet later in the unit, and you’ll
see the justification of the formula there.
For now, you might like to compare the graphs of y = tan x and y = sec2 x,
shown in Figure 10, and convince yourself that the graph of y = sec2 x
does seem, at least roughly, to give the gradients of the graph of y = tan x.

8
 1 Derivatives of further standard functions

In particular, notice that the gradients of the graph of y = tan x are always
positive and that the graph of y = tan x is the least steep whenever x is a
multiple of π.

Figure 10 The graphs of (a) y = tan x (b) y = sec2 x

Here’s a summary of the three derivatives that you’ve met in this
subsection.

Derivatives of sin, cos and tan

d d d
(sin x) = cos x (cos x) = − sin x (tan x) = sec2 x
dx dx dx

In the next activity you’re asked to use these formulas, together with the
sum and constant multiple rules, to find the derivatives of some more
functions. In this activity, and in general, remember that if a function is
specified using function notation, then you should usually use Lagrange
notation for its derivative, whereas if it’s specified by an equation that
expresses one variable in terms of another, then you should usually use
Leibniz notation.

Activity 2 Differentiating functions involving sin, cos and tan

Write down the derivatives of the following functions.

(a) f (x) = sin x + cos x (b) g(u) = u2 − cos u (c) P = 6 tan θ
(d) r = −2(1 + sin φ)

9
Unit 7 Differentiation methods and integration

1.2 Derivatives of exp and ln

In this subsection you’ll meet the formulas for the derivatives of the
exponential function, f (x) = ex , and its inverse function, the natural
logarithm function f (x) = ln x. Remember that e is a special irrational
constant, whose value is about 2.718.

Derivative of exp
Let’s look first at the exponential function, f (x) = ex . Its graph is shown
in Figure 11.

Figure 11 The graph of y = ex

You saw in Unit 3 that the constant e has the special property that the
graph of f (x) = ex has gradient exactly 1 at the point with x-coordinate 0,
as shown in Figure 12. So you already know one value of f " (x) for this
function f : you know that f " (0) = 1.

Figure 12 The graph of y = ex has gradient 1 when x = 0

10
 1 Derivatives of further standard functions

You can find the general formula for the derivative f " (x) of the function
f (x) = ex by using differentiation from first principles. The difference
quotient for f (x) = ex is
f (x + h) − f (x) ex+h − ex
= .
h h
As usual, to find a formula for f " (x), you have to work out what happens
to the value of the difference quotient as h gets closer and closer to zero.
It’s not immediately obvious how to do this, but surprisingly you can do it
by using the fact that you already know what happens in the particular
case when x = 0, since you know that f " (0) = 1.
When x = 0, the difference quotient is
e0+h − e0 eh − 1
= .
h h
So you already know that, as h gets closer and closer to zero, the value of
this expression gets closer and closer to 1.
Now look again at the difference quotient for a general value of x. You can
rearrange it as follows:
) h -
ex+h − ex ex eh − ex x e −1
= =e .
h h h
The expression in the brackets is the difference quotient in the case when
x = 0. So as h gets closer and closer to zero, the expression in the brackets
gets closer and closer to 1, and hence the value of the whole expression
gets closer and closer to ex . That is, the formula for the derivative is
f " (x) = ex .
So the derivative of the exponential function is the exponential function!
To see that this makes sense, look at the graph of the exponential function
in Figure 11. As x increases, the gradient of the graph increases, and it
increases more and more rapidly. So as x increases, the value of the
derivative of the exponential function increases, and it increases more and
more rapidly. The exponential function itself has the property that as x
increases, the value of the function increases, and it increases more and
more rapidly.
The exponential function, and its constant multiples such as the function
f (x) = 3ex , are the only functions that are equal to their own derivatives
(except, of course, for functions obtained by restricting the domains of
these functions). It’s this property that makes the constant e such a
special number.

11
Unit 7 Differentiation methods and integration

Derivative of ln
Now let’s look at the natural logarithm function, f (x) = ln x. Since this
function is the inverse of the exponential function f (x) = ex , its graph is
the reflection of the graph of f (x) = ex in the line y = x, as shown in
Figure 13.

Figure 13 The graphs of y = ln x and y = ex

So the gradients of the two graphs are related, and hence you can use the
formula for the derivative of f (x) = ex to work out a formula for the
derivative of f (x) = ln x. If you’d like to see this done, then have a look at
the document The derivative of the natural logarithm function on the
module website.
However, a quicker way to work out the derivative of the natural logarithm
function is to use a general rule, called the inverse function rule, that
you’ll meet later in the unit. This rule tells you how to use the formula for
the derivative of any invertible function to work out a formula for the
derivative of its inverse function. When you meet this rule, you’ll see that
the formula for the derivative of f (x) = ln x turns out to be
1
f " (x) = .
x
For now, you might like to compare the graphs of y = ln x and y = 1/x,
shown in Figure 14, and convince yourself that the function y = 1/x does
seem to give the gradients of the graph of the function y = ln x. Only the
right-hand half of the graph of y = 1/x is shown. The left-hand half
doesn’t give gradients of the graph of f (x) = ln x, because ln x is defined
only for x > 0.

12
 1 Derivatives of further standard functions

Figure 14 The graphs of (a) y = ln x (b) y = 1/x (x > 0)

The two derivatives that you’ve met in this subsection are summarised in
the box below.

Derivatives of exp and ln

d d 1
(ex ) = ex (ln x) =
dx dx x

In the next activity you’re asked to use these formulas, together with the
sum and constant multiple rules and the formulas for the derivatives of
trigonometric functions, to find the derivatives of some more functions.

Activity 3 Differentiating functions involving exp and log

Find the derivatives of the following functions.

(a) f (x) = ex + ln x (b) h(r) = r − cos r − 3 ln r
1
(c) v = + ln t (d) w = 5 − 3eu (e) k = 4(ln v − tan v)
t

The formulas for the derivatives of standard functions that you’ve met in
this section are all included in the Handbook. However, it’s worth
memorising the formulas for the derivatives of sin, cos, exp and ln, at least,
as they occur frequently.

13
Unit 7 Differentiation methods and integration

2 Finding derivatives of more

complicated functions
In this section you’ll learn some more ways in which you can combine
formulas for the derivatives of functions to obtain formulas for the
derivatives of other, more complicated functions.

2.1 Product rule

Suppose that you know the formulas for the derivatives of two functions,
and you want to know the formula for the derivative of their product.
(Remember that the product of two functions is obtained by ‘multiplying
them together’.) For example, you already know the formulas for the
derivatives of the functions f (x) = x2 and g(x) = sin x, but suppose that
you want to know the formula for the derivative of the function
k(x) = x2 sin x.
You’ve seen that to obtain a formula for the derivative of the sum of two
functions, you just add the formulas for the derivatives of the two
individual functions – this is the sum rule. So you might hope that to
obtain a formula for the derivative of the product of two functions, you just
multiply the formulas for the derivatives of the two individual functions.
Unfortunately, this doesn’t work! For example, consider the two power
functions f (x) = x2 and g(x) = x3 . The derivatives of x2 and x3 are 2x
and 3x2 , respectively, and multiplying these two derivatives gives 6x3 .
However, the product of the functions f (x) = x2 and g(x) = x3 is the
function k(x) = x5 , and you already know that the derivative of x5 is 5x4 .
So multiplying the formulas for the derivatives of the two individual
functions doesn’t give the right answer for the derivative of their product.
In fact, to find a formula for the derivative of the product of two functions,
you need to use the rule in the box below. It’s given in Lagrange notation,
but you’ll see it in Leibniz notation shortly.

Product rule (Lagrange notation)

If k(x) = f (x)g(x), then
k " (x) = f (x)g " (x) + g(x)f " (x),
for all values of x at which both f and g are differentiable.

The product rule can be proved by using the idea of differentiation from
first principles, and you’ll see this done at the end of this subsection.
Your main task in this subsection is to learn how to use the product rule.
The next example demonstrates how it can be used to differentiate the

14
 2 Finding derivatives of more complicated functions

function k(x) = x2 sin x, which was mentioned at the beginning of this

section.

Example 1 Using the product rule

Differentiate the function
k(x) = x2 sin x.
Solution
Identify two functions f and g such that k(x) = f (x)g(x).
Here k(x) = f (x)g(x) where f (x) = x2 and g(x) = sin x.
Find their derivatives.
Then f " (x) = 2x and g " (x) = cos x.
Use the product rule.
By the product rule,
k " (x) = f (x)g " (x) + g(x)f " (x)
= x2 cos x + (sin x) × 2x
Simplify the answer.
= x2 cos x + 2x sin x.

An alternative way to write the final answer in Example 1 is as

k " (x) = x(x cos x + 2 sin x). Either form is acceptable.
Here are some similar examples for you to try.

Activity 4 Using the product rule

Use the product rule to differentiate the following functions.

(a) k(x) = x3 tan x (b) k(x) = xex

Activity 5 Using the product rule again

Check that if you find the derivative of the function k(x) = x5 by writing
x5 = x2 × x3 and using the product rule, then you obtain the same answer
as you obtain by differentiating it directly.

15
Unit 7 Differentiation methods and integration

The product rule can be written in Leibniz notation as follows.

Product rule (Leibniz notation)

If y = uv, where u and v are functions of x, then
dy dv du
=u +v ,
dx dx dx
for all values of x at which both u and v are differentiable.

(The phrase ‘u and v are differentiable’ in the box above is a condensed

way of saying that if we write u = f (x) and v = g(x) then f and g are
differentiable at x.)
However it’s written, in essence the product rule tells you the following:
     
derivative ) - derivative ) - derivative
 of product  = first second
×  of second  + ×  of first .
function function
function function function

For most people, the best way to remember and use it is as the
abbreviated informal version below.

Product rule (informal)

) - ) - ) -
derivative derivative derivative
= (first) × + (second) × .
of product of second of first

When you use this informal version of the product rule, you can usually
skip the steps of writing down the first and second functions and their
derivatives, and just go straight to applying the rule. To do this, you recite
the informal version of the rule in your head: as you think ‘first’, you write
down the formula for the first function, then as you think ‘derivative of
second’, you differentiate the second function in your head and write down
the result, and so on. This is illustrated in the next example.

Example 2 Using the product rule efficiently

Differentiate the function
k(x) = (2x + 1) ln x.

16
 2 Finding derivatives of more complicated functions

Solution
Identify the first and second functions: here first = 2x + 1 and
second = ln x. Use the rule
) - ) - ) -
derivative derivative derivative
= (first) × + (second) × .
of product of second of first
By the product rule,
1
k " (x) = (2x + 1) × + (ln x) × 2
x
Simplify the answer.
2x + 1
= + 2 ln x
x
2x + 1 2x ln x
= +
x x
2x + 1 + 2x ln x
= .
x

There are various ways to write the final answer in Example 2. For
example, the following are all acceptable:
1 1 2x(1 + ln x) + 1
2+ + 2 ln x, + 2(1 + ln x) and .
x x x

You can practise using the informal version of the product rule in the next
activity. Alternatively, you might prefer to continue using the formal
version for a little longer, and move to the informal version when you feel
more confident with it. You can use the exercises on the product rule in
the Unit 7 practice quiz and exercise booklet for further practice.
Whichever version of the product rule you use, remember to simplify your
final answers, where possible. You should always do this when you
differentiate functions.

Activity 6 Using the product rule efficiently

Use the product rule to differentiate the following functions.

(a) k(x) = (3x2 + 2x + 1)ex (b) k(x) = sin x cos x
(c) k(z) = z sin z (d) v = (2t2 − 1) cos t (e) m = (u2 + 3) ln u
√
(f) y = x sin x
Hint: In part (f), convert the square root symbol to index notation before
you apply the product rule.

17
Unit 7 Differentiation methods and integration

Activity 7 Using the product rule to find a gradient

Find dy/dx when y = x2 ex . What is the gradient of the graph of y = x2 ex

at the point with x-coordinate −1?

To finish this subsection, here’s a proof of the product rule, using

differentiation from first principles. It uses the Lagrange notation form of
the product rule, which is repeated below.

Product rule (Lagrange notation)

If k(x) = f (x)g(x), then
k " (x) = f (x)g " (x) + g(x)f " (x),
for all values of x at which both f and g are differentiable.

A proof of the product rule

Suppose that f and g are functions, and that the function k is given by
k(x) = f (x)g(x). Let x denote any value at which both f and g are
differentiable. To determine whether k " (x) exists, and find it if it does
exist, you have to consider what happens to the difference quotient for k
at x, which is
k(x + h) − k(x)
h
(where h can be either positive or negative, but not zero), as h gets closer
and closer to zero. Since k(x) = f (x)g(x), this difference quotient is equal
to
f (x + h)g(x + h) − f (x)g(x)
.
h
You can rearrange this expression so that it includes the difference
quotients for f and g at x. To do this, you first subtract and add
f (x)g(x + h) in the numerator. Then the expression becomes
f (x + h)g(x + h) − f (x)g(x + h) + f (x)g(x + h) − f (x)g(x)
,
h
which is equal to
) - ) -
f (x + h) − f (x) g(x + h) − g(x)
g(x + h) + f (x) .
h h
The expression in the first pair of large brackets is the difference quotient
for f at x, and the expression in the second pair of large brackets is the
difference quotient for g at x. So as h gets closer and closer to zero, the
values of these two expressions get closer and closer to f " (x) and g " (x),
respectively. Also, the value of g(x + h) gets closer and closer to g(x).

18
 2 Finding derivatives of more complicated functions

Hence the whole expression gets closer and closer to f " (x)g(x) + f (x)g " (x);
that is, f (x)g " (x) + g(x)f " (x). So
k " (x) = f (x)g " (x) + g(x)f " (x),
which is the product rule.

2.2 Quotient rule

Suppose now that you know the formulas for the derivatives of two
functions, and you want to know the formula for the derivative of their
quotient. (Remember that a quotient of two functions is obtained by
‘dividing one by the other’.) For example, you already know the formulas
for the derivatives of the functions f (x) = x2 and g(x) = sin x, but suppose
that you want to know the formula for the derivative of the function
k(x) = x2 / sin x.
You can find the derivative of the quotient of two functions by using the
rule in the box below. It’s given in Lagrange notation, but you’ll see a
version in Leibniz notation shortly.

Quotient rule (Lagrange notation)

If k(x) = f (x)/g(x), then
g(x)f " (x) − f (x)g " (x)
k " (x) = ,
(g(x))2
for all values of x at which both f and g are differentiable and
g(x) )= 0.

You’ll see a justification of the quotient rule in Subsection 2.5. Here’s an

example that illustrates its use.

Example 3 Using the quotient rule

Differentiate the function
x2
k(x) = .
sin x
Solution
Identify two functions f and g such that k(x) = f (x)/g(x).
Here k(x) = f (x)/g(x) where f (x) = x2 and g(x) = sin x.
Find their derivatives.
Then f " (x) = 2x and g " (x) = cos x.

19
Unit 7 Differentiation methods and integration

Use the quotient rule.

By the quotient rule,
g(x)f " (x) − f (x)g " (x)
k " (x) =
(g(x))2
(sin x) × 2x − x2 cos x
=
(sin x)2
Simplify the answer.
2x sin x − x2 cos x
= .
sin2 x

There are various ways to write the final answer in Example 3. For
example, the following are all acceptable:
x(2 sin x − x cos x) 2x x2 cos x 2x − x2 cot x
, − , .
sin2 x sin x sin2 x sin x

Here are two similar examples for you to try.

Activity 8 Using the quotient rule

Use the quotient rule to differentiate the following functions.

ex x3
(a) k(x) = (b) k(x) =
x 2x + 1

The quotient rule can be written in Leibniz notation as follows.

Quotient rule (Leibniz notation)

If y = u/v, where u and v are functions of x, then
du dv
dy v −u
= dx 2 dx ,
dx v
for all values of x at which both u and v are differentiable and v )= 0.

For most people the best way to remember and use the quotient rule is as
the informal version below.

20
 2 Finding derivatives of more complicated functions

Quotient rule (informal)

) - ) -
derivative derivative
) - (bottom) × − (top) ×
derivative of top of bottom
= .
of quotient (bottom)2

You can use this informal version of the rule in a similar way to the
informal version of the product rule. To differentiate a quotient of two
functions by using the quotient rule, you recite the informal version of the
rule in your head: as you think ‘bottom’, you write down the formula for
the bottom function, then as you think ‘derivative of top’ you differentiate
the top function in your head and write down the result, and so on. (Make
sure that you remember that the quotient rule contains a minus sign rather
than a plus sign.) This is illustrated in the next example.

Example 4 Using the quotient rule efficiently

Differentiate the function
2x + 1
k(x) = .
ln x
Solution
Identify the top and bottom functions: here top = 2x + 1 and
bottom = ln x. Use the rule
) - ) -
derivative derivative
) - (bottom)× − (top)×
derivative of top of bottom
= .
of quotient (bottom) 2

By the quotient rule,

1
(ln x) × 2 − (2x + 1) ×
"
k (x) = x
(ln x)2

Simplify the answer. To clear the fraction in the numerator,

multiply both the numerator and the denominator by x.
1
2x ln x − (2x + 1) × × x
= x
x(ln x)2
2x ln x − 2x − 1
= .
x(ln x)2

21
Unit 7 Differentiation methods and integration

You can practise using the informal version of the quotient rule in the next
activity. Alternatively, as with the product rule, you might prefer to
continue using the formal version for a little longer, and move to the
informal version when you feel more confident with it.

Activity 9 Using the quotient rule

Use the quotient rule to differentiate the following functions.

ex u−1 r+1
(a) f (x) = 3 (b) g(u) = u
(c) z =
x e r−1
θ ln t
(d) m = (e) p(t) = 2
cos θ t

In Subsection 1.1 you met the fact that the derivative of tan x is sec2 x, but
this result wasn’t justified there. It can be confirmed by using the quotient
rule, as shown in the next example. (Note, however, that this justification
is incomplete, as you haven’t yet seen a proof of the quotient rule!)

Example 5 Using the quotient rule to find the derivative of tan

Show that the derivative of tan x is sec2 x.
Solution
Use the fact that tan x = (sin x)/(cos x), and apply the quotient
rule.
By the quotient rule,
) -
d d sin x
(tan x) =
dx dx cos x
(cos x)(cos x) − (sin x)(− sin x)
=
(cos x)2
cos2 x + sin2 x
=
cos2 x
Simplify this expression by using the facts that cos2 x + sin2 x = 1
and 1/ cos x = sec x.
1
=
cos2 x
= sec2 x,
as required.

22
 2 Finding derivatives of more complicated functions

You can also use the quotient rule to find the derivatives of the
trigonometric functions cosec, sec and cot. These derivatives are given in
the box below, and you’re asked to confirm them in the next activity.

Derivatives of cosec, sec and cot

d
(cosec x) = − cosec x cot x
dx
d
(sec x) = sec x tan x
dx
d
(cot x) = − cosec2 x
dx

Activity 10 Finding the derivatives of further trigonometric

functions
Use the quotient rule to confirm the following formulas for derivatives.
d d
(a) (cot x) = − cosec2 x (b) (sec x) = sec x tan x
dx dx
d
(c) (cosec x) = − cosec x cot x
dx
Hint: Use the facts that cot x = (cos x)/(sin x), sec x = 1/(cos x) and
cosec x = 1/(sin x).

The derivatives of cosec, sec and cot can now be included in the list of
derivatives of standard functions that you have available to work with.
The Handbook includes a table of all the standard derivatives given in this
module.

2.3 Chain rule

Now that you know the product rule and the quotient rule, you can
differentiate a much wider range of functions than you could before.
However, there are still some fairly simple functions that you can’t
differentiate using any of the rules that you’ve met so far.
For example, consider the function k(x) = sin(x2 ). This function isn’t a
sum, a product or a quotient of functions that you can differentiate.
However, it’s still made up of functions that you can differentiate, just
combined in a different way. To work out k(x) for a particular value of x,
you have to first square x, then find the sine of the result. So the
function k is a composite of the functions f (x) = x2 and g(x) = sin x, as
shown in Figure 15. You met the idea of a composite of two functions in
Subsection 3.2 of Unit 3.

23
Unit 7 Differentiation methods and integration

Figure 15 A composite of the functions f (x) = x2 and g(x) = sin x

The functions f (x) = x2 and g(x) = sin x are functions that you can
differentiate.
In this subsection you’ll meet another rule for differentiation, called the
chain rule, which allows you to differentiate functions that are composites
of functions that you can already differentiate. You’ll see towards the end
of the subsection why it’s called the chain rule. It’s also sometimes called
the composite rule or the function of a function rule.
The first step in learning to use the chain rule is to learn to recognise when
a function is a composite of two functions, and how to decompose such a
function into its constituent functions. (Decomposing is the opposite of
composing.) To see how to do this, let’s consider again the function
k(x) = sin(x2 ).
While you’re learning about decomposing functions, it’s easier to work
with functions expressed using two variables rather than function notation,
so let’s first express this function as
y = sin(x2 ).
To recognise that it’s a composite of two functions, you observe that it’s of
the form
y = sin(something),
where the ‘something’ is an expression that contains the input variable x.
You can then decompose it into its two constituent functions.
A convenient way to do this is to set the ‘something’ equal to an extra
variable, say u. This gives
y = sin(u) where u = x2 .
These two equations specify the two constituent functions.
In general, suppose that you have an equation that specifies y as a function
of x. If the equation is of the form
y = a function of ‘something’,
where the ‘something’ is a function of the input variable x, then the
original function is a composite function. To decompose it into its two
constituent functions, you set the ‘something’ equal to u. In other words,
you write y as a function of u, where u is a function of x.
Here are some more examples that illustrate how to do this.

24
 2 Finding derivatives of more complicated functions

Example 6 Decomposing composite functions

For each of the following equations, write y as a function of u, where
u is a function of x.
(a) y = cos(ln x) (b) y = e2x (c) y = sin2 x
Solution
(a) y = cos(something), so set the something = u.
y = cos u where u = ln x.
(b) y = esomething , so set the something = u.
y = eu where u = 2x.
(c) Remember that sin2 x means (sin x)2 . Therefore
y = something2 , so set the something = u.
y = u2 where u = sin x.

Here are some examples for you to try.

Activity 11 Decomposing composite functions

For each of the following equations, write y as a function of u, where u is a

function of x.
(a) y = ln(x4 ) (b) y = sin(x2 − 1) (c) y = cos(ex )
√
(d) y = cos3 x (e) y = etan x (f) y = ln x
√ √ √
(g) y = (x + x )9 (h) y = 3 x2 − x − 1 (i) y = e x

The chain rule, which allows you to differentiate composite functions, is

given below. It’s given in Leibniz notation, because it’s easier to remember
in this form, but you’ll see it in Lagrange notation later in this subsection.

Chain rule (Leibniz notation)

If y is a function of u, where u is a function of x, then
dy dy du
= ,
dx du dx
for all values of x where y as a function of u, and u as a function of x,
are differentiable.

25
Unit 7 Differentiation methods and integration

As you can see, in Leibniz notation the chain rule looks like an obvious
statement about fractions, which makes it easy to remember. You should
keep in mind, though, that this is not what it is! The pieces of notation
dy/dx, dy/du and du/dx represent derivatives, not fractions.
As with the other rules for combining derivatives, the chain rule can be
proved by using the idea of differentiation from first principles.
Unfortunately, this is quite tricky to do, and the proof isn’t included in
this module.
However, here’s an informal way to see that the chain rule makes sense.
Suppose that y is a function of u, where u is a function of x, as in the
statement of the chain rule. Suppose that for a particular value of x, the
value of du/dx is 2, and for the value of u that corresponds to this
particular value of x, the value of dy/du is 3. Since y is increasing at the
rate of 3 units for every unit that u increases, and u is increasing at the
rate of 2 units for every unit that x increases, you’d expect y to be
increasing at the rate of 3 × 2 = 6 units for every unit that x increases.
This is what the chain rule tells you.
The next example shows you how to use the chain rule.

Example 7 Using the chain rule

Differentiate the function y = sin(x2 ).
Solution
Recognise that the given function is a function of ‘something’,
where the ‘something’ involves the input variable x. Hence decompose
the given function.
Here y = sin u where u = x2 .
Find the derivatives of the constituent functions.
This gives
dy du
= cos u, = 2x.
du dx
dy dy du
Apply the chain rule, = .
dx du dx
By the chain rule,
dy
= (cos u) × 2x
dx
Express the answer in terms of x, by substituting for u in terms
of x. Here we have u = x2 .

= (cos(x2 )) × 2x

26
 2 Finding derivatives of more complicated functions

Simplify the answer.

= 2x cos(x2 ).

Notice that when you apply the chain rule as illustrated in Example 7, you
first obtain an answer that’s expressed in terms of both u and x. You then
have to substitute for u, using the equation for u in terms of x that you
wrote down when you decomposed the function, to obtain a final answer in
terms of x.

Activity 12 Using the chain rule

Differentiate the following functions.

,x+
(a) y = e5x (b) y = sin(2x2 − 3) (c) y = sin
4
3
(d) y = cos(ex ) (e) y = ln(sin x) (f) y = sin x
Hint: In part (f), remember that sin3 x means (sin x)3 .

You’ve now seen how to apply the chain rule by first introducing an extra
variable, usually u, which you then remove again to obtain the final
answer. Once you’re familiar with using the chain rule, you can often
apply it without introducing this extra variable at all.
To see how to do this, consider again the equation y = sin(x2 ). Think of
the expression that you would have written as u, that is, the expression
that was referred to as ‘something’ earlier, as one ‘big variable’:
✎☞
x2 )
y = sin( ✍✌

To find dy/dx, first differentiate with respect to this big variable; that is,
with respect to the ‘something’. This gives
✎☞
x2 )
cos( ✍✌
By the chain rule, to obtain the final answer, you now need to multiply by
the derivative of the something with respect to x. This gives
dy
= cos(x2 ) × 2x.
dx
This is the final answer, but, as before, you should simplify it slightly to
give
dy
= 2x cos(x2 ).
dx

27
Unit 7 Differentiation methods and integration

Here are two more examples. Notice that in the second of these examples
the function is expressed using function notation rather than using two
variables. This makes no difference to how you apply the ideas above.

Example 8 Using the chain rule without introducing an extra

variable
Differentiate the following functions.
(a) y = ln(2x − 1) (b) k(x) = cos4 x
Solution
(a) Think of the ‘something’, 2x − 1, as a ‘big variable’.
Differentiate with respect to this variable. Then multiply by the
derivative of 2x − 1 with respect to x.
By the chain rule,
dy 1
= ×2
dx 2x − 1
Simplify the answer.
2
= .
2x − 1
(b) The function is k(x) = (cos x)4 , so think of the ‘something’,
cos x, as a ‘big variable’. Differentiate with respect to this
variable. Then multiply by the derivative of cos x with respect
to x.
By the chain rule,
k " (x) = (4 cos3 x)(− sin x)
Simplify the answer.
= −4 sin x cos3 x.

Try this quicker way of using the chain rule in the following activity.

Activity 13 Using the chain rule without introducing an extra

variable
Differentiate the following functions.
(a) f (x) = cos(4x) (b) f (x) = e2x+1 (c) r(x) = cos5 x
(d) g(x) = ln(x2 + x + 1) (e) v(t) = sin(et ) (f) v = ln(r5 )
(g) r = tan2 θ (h) R = etan q (i) A = (1 + p2 )9

28
 2 Finding derivatives of more complicated functions

You should try to get used to applying the chain rule in the quick way
demonstrated in Example 8, as you’ll find this skill useful in calculus.
However, sometimes, if a function that you want to differentiate using the
chain rule isn’t straightforward, then you might find it easier to revert to
introducing an extra variable. It’s fine to do that if you find it helpful.
Sometimes you can make it easier to apply the chain rule without
introducing an extra variable if you start by rearranging the formula for
the function. This applies in particular to composites that involve power
functions. Here’s an example.

Example 9 Applying the chain rule to a composite involving a

power function
Differentiate the function
1
f (x) = √ .
sin x
Solution
√
The function f is of the form f (x) = 1/ something, where the
‘something’ is sin x. So use the chain rule. Start by rearranging the
formula for f into the form ‘something n ’.
The function is
1
f (x) = √ ;
sin x
that is,
f (x) = (sin x)−1/2 .
Apply the chain rule.
By the chain rule,
f " (x) = − 12 (sin x)−3/2 (cos x)
cos x
=− .
2 sin3/2 x

With practice, you may be able to do the sort of initial rearranging needed
in Example 9 in your head.

29
Unit 7 Differentiation methods and integration

Activity 14 Applying the chain rule to composites involving power

functions
Differentiate the following functions.
√ 1 √
(a) f (x) = ln x (b) p(w) = 3 4
(c) f (x) = cos( x)
(w + 7)
1 1 2
(d) f (x) = √ x (e) a(t) = 2 1/3
(f) C(p) = e1/p
e (2 − 3t − 3t )

Now here’s an explanation of where the name ‘chain rule’ comes from.
Consider a function that’s a composite of three functions. Suppose that
y is a function of u,
u is a function of v,
v is a function of x.
You can find the derivative of y with respect to x by applying the chain
rule twice:
dy dy du du du dv
= and =
dx du dx dx dv dx
Gerald was never very good so, using the second equation to substitute in the first equation, you obtain
at using the chain rule. dy dy du dv
= .
dx du dv dx
As you can see, this gives a ‘chain’ of derivatives. You can keep adding
‘links in the chain’ indefinitely. One of the examples in Subsection 2.5,
Example 12, illustrates how you can apply the chain rule twice in this way.
Finally in this subsection, let’s translate the chain rule into Lagrange
notation. You can do this by thinking about the way of applying it
without introducing an extra variable. Suppose that f and g are functions,
and you want to differentiate the expression g(f (x)) with respect to x.
The chain rule says that you can do this by differentiating it with respect
to the ‘something’ f (x), then multiplying the result by the derivative of
f (x) with respect to x. This gives the version of the chain rule below.

Chain rule (Lagrange notation)

If k(x) = g(f (x)), where f and g are functions, then
k " (x) = g " (f (x))f " (x)
for all values of x such that f is differentiable at x and g is
differentiable at f (x).

30
 2 Finding derivatives of more complicated functions

The first use of the chain rule seems to be by Gottfried Wilhelm

Leibniz in a memoir√ of 1676, in which he calculated (with mistakes!)
the derivative of a + bz + cz 2 . Further examples of the use of the
rule appear in the first calculus textbook, written by Guillaume de
l’Hôpital and published in 1696, but nowhere is the rule actually
stated. (There is more about l’Hôpital’s book in Subsection 3.4.) It
was not until the 19th century that the chain rule was commonly
stated explicitly, and at that time it was generally described as a ‘rule
for differentiating a function of a function’. The term chain rule seems
to have originated as a translation of the German word Kettenregel,
which was used in German calculus books in the early 20th century.

2.4 Differentiating functions of linear expressions

This subsection is about a type of function that occurs frequently, and
which you can differentiate by using the chain rule.
Consider the functions
k(x) = sin(3x) and h(x) = e−2x .
Each of them is a composite function, and in each of them the ‘something’
is a constant multiple of the input variable x; the ‘somethings’ are 3x
and −2x, respectively. It’s worth noticing, and remembering, what
happens when you use the chain rule to differentiate functions like these.
For example,
if k(x) = sin(3x), then k " (x) = cos(3x) × 3 = 3 cos(3x),
if h(x) = e−2x , then h" (x) = e−2x × (−2) = −2e−2x .
You can see that to differentiate a composite function in which the
‘something’ is ax, where a is a constant, you differentiate with respect
to ax, and then multiply by a.
It’s worth remembering this special case of the chain rule, which can be
stated as below.

Derivative of a function of the form k(x) = f (ax)

If k(x) = f (ax), where a is a non-zero constant, then
k " (x) = af " (ax),
for all values of x such that f is differentiable at ax.

You can use this rule to quickly differentiate any function of the type
discussed above. For example, it tells you that
if k(x) = sin(5x), then k " (x) = 5 cos(5x).
Try using the rule above in the next activity.
31
Unit 7 Differentiation methods and integration

Activity 15 Differentiating functions of the form k(x) = f (ax)

Differentiate the following functions.

) -
θ
(a) k(x) = cos(7x) (b) r(θ) = sin (c) g(u) = e3u/2
2
(d) h(x) = e−x (e) p(α) = cos(−8α) (f) w(x) = ln(7x)
) -
−2t
(g) f (t) = sin (h) k(φ) = tan(3φ)
3

In fact you can slightly extend the special case of the chain rule above.
Notice that, by the chain rule,
if k(x) = sin(3x − 2), then k " (x) = cos(3x − 2) × 3 = 3 cos(3x − 2),
if h(x) = e−2x+5 , then h" (x) = e−2x+5 × (−2) = −2e−2x+5 .

You can see that, in general, to differentiate a composite function in which

the ‘something’ is ax + b, where a and b are constants, you differentiate
with respect to ax + b, and then multiply by a. This special case of the
chain rule is summarised below.

Derivative of a function of the form k(x) = f (ax + b)

If k(x) = f (ax + b), where a and b are constants with a non-zero, then
k " (x) = af " (ax + b),
for all values of x such that f is differentiable at ax + b.

For example, this rule tells you that

if k(x) = sin(5x − 8), then k " (x) = 5 cos(5x − 8).
Try using it in the next activity.

Activity 16 Differentiating functions of the form k(x) = f (ax + b)

Differentiate the following functions.

) -
θ−1
(a) k(x) = cos(7x + 4) (b) h(x) = e−x−3 (c) r(θ) = sin
2
) -
2−θ
(d) s(θ) = sin
3

32
 2 Finding derivatives of more complicated functions

Remember that an expression of the form ax + b, where a and b are

constants, is called a linear expression in x. So in this subsection you’ve
seen how to quickly differentiate functions of linear expressions.

2.5 Using the differentiation rules together

You can differentiate a wide range of functions by combining the various
differentiation rules that you’ve met. For example, you can use more than
one differentiation rule, or more than one application of the same rule.
This subsection starts with three examples that show you the kinds of
things that you can do.
When you’re combining differentiation rules in this way, it can be useful to
use the notation d/dx to indicate the derivative of a function that appears
as part of your working, then actually do the differentiation in the next
line. This is illustrated in the examples.

Example 10 Combining the differentiation rules

Find the derivative of
f (x) = x2 sin(x3 ).
Solution
The function is of the form something × something. So apply the
product rule. The second ‘something’, sin(x3 ), isn’t straightforward
to differentiate, so use the notation d/dx to indicate its derivative.
d
f " (x) = x2 (sin(x3 )) + (sin(x3 ))(2x) (by the product rule)
dx
Now differentiate sin(x3 ). It’s of the form sin(something), so apply
the chain rule.
= x2 (cos(x3 ))(3x2 ) + (sin(x3 ))(2x) (by the chain rule)
Simplify the answer.
= 3x4 cos(x3 ) + 2x sin(x3 ).

33
Unit 7 Differentiation methods and integration

Example 11 Combining the differentiation rules again

Find the derivative of
) 2 -4
x −2
y= .
x2 + 1
Solution
The function is of the form (something)4 . So apply the chain rule.
The ‘something’ isn’t straightforward to differentiate, so use the
notation d/dx to indicate its derivative.
) 2 -3 ) -
dy x −2 d x2 − 2
=4 × (by the chain rule)
dx x2 + 1 dx x2 + 1
Now differentiate (x2 − 2)/(x2 + 1). It’s of the form
something/something, so apply the quotient rule.
) 2 -3 ) 2 -
x −2 (x + 1)(2x) − (x2 − 2)(2x)
=4 ×
x2 + 1 (x2 + 1)2
(by the quotient rule)
Simplify the answer.
) 2 -3
x −2 (2x)(x2 + 1 − x2 + 2)
=4 ×
x2 + 1 (x2 + 1)2
) 2 -3
x −2 6x
=4 2
× 2
x +1 (x + 1)2
24x(x2 − 2)3
= .
(x2 + 1)5

Example 12 Combining the differentiation rules yet again

Find the derivative of
g(t) = cos(e2t+1 ).
Solution
The function is of the form cos(something). So apply the chain
rule. The ‘something’ isn’t straightforward to differentiate, so use the
notation d/dt to indicate its derivative.
d 2t+1
g " (t) = − sin(e2t+1 ) × (e ) (by the chain rule)
dt

34
 2 Finding derivatives of more complicated functions

Now differentiate e2t+1 . It’s of the form esomething . So apply the

chain rule again.
= − sin(e2t+1 ) × e2t+1 × 2 (by the chain rule again)
Simplify the answer.
= −2e2t+1 sin(e2t+1 ).

You can practise combining the differentiation rules in the next activity. If
you find it tricky to work out which rules you need to apply, then the
checklist below might help.

Checklist for differentiating a function

1. Is it a standard function (is its derivative given in the Handbook )?
2. Can you use the constant multiple rule and/or sum rule?
3. Can you rewrite it to make it easier to differentiate? For
example, multiplying out brackets may help.
4. Is it of the form f (ax) or f (ax + b), where a and b are constants
and f is a function? If so, use the rule for differentiating a
function of a linear expression.
5. Can you use the product rule (is it of the form
f (x) = something × something)?
6. Can you use the quotient rule (is it of the form
f (x) = something/something)?
7. Can you use the chain rule (is it of the form
f (x) = a function of something)?
When you use a differentiation rule, you usually have to find the
derivatives of simpler functions. You can apply the checklist above to
each of these simpler functions in turn.

Note also that when you need to apply more than one rule to differentiate
a function, the first rule that you should apply is the one that applies to
the overall structure of the rule of the function. For example, as you saw in
Example 10, the overall structure of the function f (x) = x2 sin(x3 ) is that
it’s a product, since x2 sin(x3 ) is the product of x2 and sin(x3 ). So, to
differentiate this function, you start by applying the product rule, ignoring
for the moment the fact that one of the two functions in the product is
itself a composite. After you’ve done that, you can deal with the
composite.

35
Unit 7 Differentiation methods and integration

Activity 17 Combining the differentiation rules

Differentiate the following functions.

x
(a) g(x) = (x2 + 1)(x3 + 1) (b) f (x) =
(x − 2)3
cos(4u)
(c) g(x) = cos(x ln x) (d) h(x) = ex/2 sin(3x) (e) q(u) =
e3u
e−2x
(f) f (x) = esin(5x) (g) h(x) = cos3 (x2 ) (h) z(x) =
5
(i) G(x) = 3 sin x cos x + 2 sin(3x) (j) p(t) = tet sin t
√
sin u
(k) r = sin(2θ) cos θ (l) h = ze3z (m) v = √
u

Sometimes you have to differentiate functions that involve constants

represented by letters. You can do this in the usual way, treating the
constants like any other numbers.

Activity 18 Differentiating functions involving constants

Differentiate the following functions.

(a) g(x) = e−kx , where k is a constant.
(b) f (x) = x cos(πx).
(c) h = sin2 (aθ), where a is a constant.
(d) y = cos(3x) + c, where c is a constant.

Here’s another technique that’s occasionally useful when you want to

differentiate a complicated function. Whenever you have a function that
you can differentiate using the quotient rule, an alternative way to
differentiate it is to rearrange its rule to express it as a product of two
functions, and then apply the product rule.
In the next activity you’re asked to differentiate a function in this way.
(You were asked to differentiate the same function using the quotient rule
in Activity 9(e).)

Activity 19 Differentiating a quotient by using the product rule

ln t
Differentiate the function p(t) = 2 by writing it as p(t) = t−2 ln t and
t
using the product rule.

36
 2 Finding derivatives of more complicated functions

In fact, the quotient rule is really just a combination of the product rule
and a special case of the chain rule. Here’s a proof of it, as promised in
Subsection 2.2.

A proof of the quotient rule

Suppose that the function k has rule k(x) = f (x)/g(x). This rule can be
written as
1
k(x) = × f (x);
g(x)
that is,
k(x) = (g(x))−1 f (x).

So, for each value of x at which both f and g are differentiable and
g(x) )= 0,
d % $
k " (x) = (g(x))−1 f " (x) + f (x) (g(x))−1 (by the product rule)
dx
% $
= (g(x))−1 f " (x) + f (x) −(g(x))−2 g " (x) (by the chain rule)
f " (x) f (x)g " (x)
= −
g(x) (g(x))2
g(x)f " (x) − f (x)g " (x)
= ,
(g(x))2
which is the quotient rule.

A proof of the formula for the derivative of a power function

Finally in this subsection, here’s a proof of the general formula for the
derivative of a power function, as promised in Unit 6. The formula states
that, for any constant n,
d
(xn ) = nxn−1 .
dx
The proof given here is valid only for x > 0, as you’ll see. However, for a
power function y = xn that’s defined when x < 0, you can use the fact that
the formula holds for x > 0, together with the symmetry of the graph of
y = xn , to deduce that the formula also holds for x < 0; the details are not
given here. To see that the formula also holds when x = 0, you can use the
fact that at x = 0 the gradient of the graph of y = xn is 0.

37
Unit 7 Differentiation methods and integration

To prove the formula for x > 0, you can use the fact that x = eln x for such
values of x to write the expression xn in the following form:
xn = (eln x )n = en ln x .
This gives
d d n ln x
(xn ) = (e )
dx dx
d
= (en ln x ) (n ln x) (by the chain rule)
dx
d
= (en ln x ) × n × (ln x) (by the constant multiple rule)
dx
1
= (en ln x ) × n ×
x
n 1
=x ×n×
x
n−1
= nx ,
which is the formula stated above.
Note that the two proofs that you’ve just seen, namely the proof of the
quotient rule and the proof of the formula for the derivative of a power
function, both depend on the chain rule. Since this module doesn’t include
a proof of the chain rule, strictly the proofs of these two results are
incomplete.

3 More differentiation
In this section you’ll see how to use differentiation to solve a type of
problem that occurs frequently in applications of mathematics, how to use
differentiation on a computer, and how to find the derivative of the inverse
of a function if you know the derivative of the original function.

3.1 Optimisation problems

In mathematics, and its applications, it’s often helpful to solve problems
that involve identifying the best possible option from a choice of suitable
possibilities. Problems of this type are known as optimisation problems.
For example, the manufacturer of an electronic device might want to
identify the best price that it should charge for its device, if its goal is to
achieve the maximum profit. Increasing the price might increase the profit,
but increasing it further might decrease the profit, by causing fewer of the
devices to be sold. This problem is an example of a maximisation
problem, because it involves identifying the circumstances under which the
maximum value of a quantity is obtained. Similarly, some optimisation

38
 3 More differentiation

problems are minimisation problems, which involve identifying the

circumstances under which the minimum value of a quantity is obtained.
You can often solve optimisation problems by modelling them
mathematically and then applying calculus. In particular, you can often
use the strategy below, which you met in Unit 6, or an adaptation of it.

Strategy:
To find the greatest or least value of a function on an
interval of the form [a, b]
(This strategy is valid when the function is continuous on the interval,
and differentiable at all values in the interval except possibly the
endpoints.)
1. Find the stationary points of the function.
2. Find the values of the function at any stationary points inside the
interval, and at the endpoints of the interval.
3. Find the greatest or least of the function values found.

The next example illustrates how you can use this strategy to solve an
optimisation problem. You might like to try guessing the answer to the
problem posed in the example, and then read on to see whether you’re
right.

Example 13 Solving an optimisation problem

A rectangular sheet of metal, of width 1 m, is to be bent along two
straight lines parallel to the sides of the sheet, to form an open
channel with a rectangular cross-section, as shown below. How far
from the sides of the sheet should the metal be bent to achieve a
channel of maximum volume?

39
Unit 7 Differentiation methods and integration

Solution
Simplify the problem if possible. Draw a diagram.
The volume of the channel is the area of its cross-section times its
(fixed) length. So to achieve a channel of maximum volume, we have
to achieve the maximum area of the cross-section.
The diagrams below show the cross-sections of the metal sheet and the
channel. We have to determine how far from the side the metal sheet
should be bent to achieve the maximum area of the cross-section.

Identify the quantity that you can change, and represent it by a

variable, noting the possible values that it can take. Identify the
quantity to be maximised, and represent it by a variable too.
Let the distance from the side of the metal to where it is bent be x
(in m). The value of x must be between 0 and 21 .
Let the area of the cross-section be A (in m2 ).
Find a formula for the variable to be maximised (the dependent
variable) in terms of the variable that you can change (the
independent variable). Start by annotating the diagram.

The area of the cross-section is given by

A = x(1 − 2x).
Formulate the problem in terms of the variables.
We have to find the value of x, between 0 and 21 , that gives the
maximum value of A.

40
 3 More differentiation

Use calculus to solve the problem. You can use the strategy in the
box given before this example.
The variable A is a polynomial function of x, so it is continuous on its
whole domain and differentiable at every value of x. So the maximum
value of A must occur either at an endpoint of the interval [0, 12 ], or at
a stationary point.
Find the stationary points.
The formula for A is
A = x(1 − 2x)
= x − 2x2 ,
so
dA
= 1 − 4x.
dx
At a stationary point, dA/dx = 0, which gives
1 − 4x = 0;
that is,
x = 41 .
So the only stationary point occurs when x = 14 .
Consider the values of the function at any stationary points inside
the interval, and at the endpoints of the interval.
The endpoints of the interval are 0 and 21 , so the maximum value of A
occurs when x = 0, x = 14 or x = 12 .
When x = 0, A = 0, and similarly when x = 21 , A = 0.
When x = 14 ,
1
A= 4 × (1 − 2 × 14 ) = 1
4 × 1
2 = 1
8 = 0.125.
Find the greatest of the function values found.
The greatest of the three values of A found is 0.125, which is achieved
when x = 14 .
So the maximum value of A is achieved when x = 14 .
State a conclusion that answers the original problem.
To achieve a channel of maximum volume, the metal should be bent
at a distance of 25 cm from each side.

41
Unit 7 Differentiation methods and integration

In fact, in Example 13 it isn’t necessary to actually work out the values

of A when x = 0, x = 14 and x = 21 . When x = 0 the height of the channel
is zero, and when x = 12 its width is zero, so in both these cases the area A
will be zero. When x = 14 , the height and the width of the channel are
both positive, so the area A will also be positive. So you can see
immediately that, of these three values of x, the value x = 14 will give the
greatest value of A.
The solution to Example 13 shows that the channel has maximum volume
when it is twice as wide as it is high.
When you’re solving a maximisation or minimisation problem, it can be
helpful to sketch a graph of the function involved, or plot it using a
computer. This will give you a rough idea of how the quantity that you
want to maximise or minimise changes as the quantity on which it depends
changes. For example, Figure 16 shows a graph of the equation
A = x(1 − 2x) from Example 13. You can see that, as x increases, the
value of A first increases and then decreases, as you’d expect, and the
maximum value of A seems to occur when x is about 0.25, which accords
with the answer found in Example 13.

1
Figure 16 The graph of A = x(1 − 2x) for 0 ≤ x ≤ 2
Notice that in fact you could have solved the problem in Example 13
without using calculus, because the function involved turned out to be a
quadratic function. You could have found the value of x that gives the
maximum value of A by using a method other than calculus to find the
vertex of the parabola that’s the graph of the function. You saw some
methods for doing this in Unit 2. However, once you’re familiar with
calculus, it’s usually easier to use it.
Each of the next two activities poses an optimisation problem for you to
solve by using calculus. The first activity involves maximising the volume
of a box, and the second one involves maximising the area of the
cross-section of an open channel, in a similar way to Example 13. In both
activities, you might like to try guessing the answers to the problems posed
before you start work on the solutions.

42
 3 More differentiation

Activity 20 Solving an optimisation problem

Suppose that you have a sheet of cardboard that measures 1 m by 1 m.

You intend to cut a square from each of the four corners and then fold the
resulting shape to form a box (without a top), as shown. What should the
side length of the squares be, to give a box of maximum volume?

In the next activity you should use the fact that the area of a triangle with
sides of lengths a and b and included angle θ (as illustrated in Figure 17) is
given by the formula A = 21 ab sin θ. You met this formula in Unit 4.

Activity 21 Solving another optimisation problem

Consider again the problem of the open channel in Example 13. Suppose Figure 17 A triangle with
sides of lengths a and b and
that instead of bending the metal along two straight lines, you intend to
included angle θ
bend it along one straight line in the middle, to form an open channel with
a v-shaped cross-section, as shown below. What should be the angle
between the two sides of the v-shape, to give a channel with maximum
volume?

43
Unit 7 Differentiation methods and integration

Here’s a summary of the main steps that you can use to solve an
optimisation problem. Of course, you should combine these steps with
other problem-solving techniques as appropriate, such as simplifying the
problem if possible, drawing a diagram and drawing a graph.

Strategy:
To solve an optimisation problem
1. Identify the quantity that you can change, and represent it by a
variable, noting the possible values that it can take. Identify the
quantity to be maximised or minimised, and represent it by a
variable. These variables are the independent and dependent
variables, respectively.
2. Find a formula for the dependent variable in terms of the
independent variable.
3. Use the techniques of differential calculus to find the value of the
independent variable that gives the maximum/minimum value of
the dependent variable.
4. Interpret your answer in the context of the problem.

The optimisation techniques that you’ve met in this subsection are

useful in the design of everyday objects. For example, suppose that
you want to design the shape of a food tin so that it has a certain
volume but uses the minimum amount of metal. You can use calculus
to determine the optimal shape of the tin. If you simplify the problem
by assuming that the tin is simply a hollow cylinder, as shown in
Figure 18, then it turns out that the optimal shape is for the diameter
of the cylinder to be equal to its height. (You might like to try to
prove this.) Food tins tend to have proportions close to this, though
to optimise accurately the shape of a food tin you would need to take
into account the extra metal needed for the joins, and any differences
in the thicknesses of metal needed for the sides, base and top.
Figure 18 A cylinder

3.2 Differentiation using a computer

Sometimes it’s convenient to use a computer to find the derivatives of
functions, and to work with these derivatives. The next activity shows you
how to use the module computer algebra system to do this.

Activity 22 Using a computer for differentiation

Work through Section 8 of the Computer algebra guide.

44
 3 More differentiation

To illustrate how computers can be used to solve calculus problems, in the

rest of this subsection we’ll use a computer to solve the following
minimisation problem.
A man is in a forest, 1 km away from the nearest point on a straight
road that passes through the forest, as shown in Figure 19. He wants
to get to a point 2 km further along the road, as quickly as possible, by
first running in a straight line through the forest and then running
along the road. He can run at 8 kilometres per hour through the
forest, and at 16 kilometres per hour along the road. At what point
along the road should he aim to join it?

1 km

2 km

Figure 19 The position of a man in a forest and his path to his

destination
Before we can use a computer, we need to carry out the first two steps of
the strategy given at the end of the last subsection, to convert the problem
from a word problem into a problem expressed in terms of variables and a
function.
To do this, let x be the distance in kilometres between the point where the
man joins the road and the point on the road that is closest to his initial
position, as shown in Figure 20. Then x can take any value in the interval
[0, 2]. Let T be the total time in hours taken by the man to reach his
destination.

1 km

Figure 20 An annotated diagram of the man’s position and path

45
Unit 7 Differentiation methods and integration

We now have to express T in terms of x. By Pythagoras’ theorem, the

distance in kilometres that the man runs through the forest is
(
12 + x2 .
The distance in kilometres that he runs along the road is simply
2 − x.
For each of the two parts of the man’s run, the time that he takes is equal
to the distance that he runs divided by the speed at which he runs. Hence
the total time T that he takes to reach his destination is given by
√
1 + x2 2 − x
T = + .
8 16
To solve the problem, we now need to use calculus to find the value of x in
the interval [0, 2] that gives the minimum value of T . That is, we need to
carry out the final step of the strategy.
You could do that without using a computer, and you might like to try
this, if you want a challenge. However, in the next activity you’re asked to
do it by using a computer. This is quicker and easier, and it will give you
practice in using a computer to solve problems like this. A by-hand
solution is provided at the end of the solution to the activity.

Activity 23 Solving an optimisation problem using a computer

Solve the problem discussed above, by following the steps below. Carry out
parts (a), (b), (d) and (e) using the module computer algebra system.
(a) Define a function f as follows:
√
1 + x2 2 − x
f (x) = + .
8 16
(b) Plot the graph of f , for values of x in the interval [0, 2] at least.
(c) Use your plot to estimate roughly the value of x in [0, 2] that gives the
minimum value of f (x).
(d) Find the derivative f " of f .
(e) Solve the equation f " (x) = 0.
(f) Hence state, to the nearest 10 metres, how far along the road the man
should join it.

3.3 Derivatives of inverse functions

In this subsection you’ll meet a rule, known as the inverse function rule,
which you can use to work out a formula for the derivative of an inverse
function when you know a formula for the derivative of the original
function.
The inverse function rule can be derived from the chain rule, as follows.
We’ll use Leibniz notation, since the inverse function rule is usually easier
to work with in Leibniz notation.
46
 3 More differentiation

Suppose that y is an invertible function of x. Then x is also a function

of y. As you know, the derivative of x with respect to x is 1, and we can
write this fact as
dx
= 1.
dx
However, by the chain rule,
dx dx dy
= .
dx dy dx
Therefore
dx dy
= 1.
dy dx
Hence, for values of x for which dx/dy exists and is non-zero,
dy 1
= .
dx dx/dy
So we have the following rule.

Inverse function rule (Leibniz notation)

If y is an invertible function of x, then
dy 1
= ,
dx dx/dy
dx
for all values of x such that exists and is non-zero.
dy

You can see that, as with the chain rule, the Leibniz form of the inverse
function rule looks like an obvious fact about fractions. As with the chain
rule, you should keep in mind that that’s not what it is.
Here’s an informal way to see that the inverse function rule makes sense.
Suppose that y is an invertible function of x, which means that x is also a
function of y. Suppose that, for a particular value of x and its
corresponding value of y, the value of x is increasing at the rate of 2 units
for every unit that y increases, as illustrated in Figure 21. Then, at these
values of x and y, you’d expect the value of y to be increasing at the rate
of 21 unit for every unit that x increases. This is what the inverse function
rule tells you.

Figure 21 A point (x, y) on the graph of a function

47
Unit 7 Differentiation methods and integration

You’ll see the inverse function rule in Lagrange notation later in this
subsection. First let’s look at some examples of how you can use it.

Using the inverse function rule

In the next example, the inverse function rule in Leibniz notation is used
to work out that the derivative of ln x is 1/x. This is a fact that you met
in Subsection 1.2.

Example 14 Using the inverse function rule

Use the inverse function rule, and the derivative of the exponential
function, to differentiate y = ln x.
Solution
Express x in terms of y.
We have y = ln x, so
x = ey .
Differentiate with respect to y.
Therefore
dx
= ey .
dy
Use the inverse function rule.
By the inverse function rule,
dy 1
= y.
dx e
Use the relationship between x and y to express the derivative in
terms of x.
Hence, since x = ey ,
dy 1
= .
dx x

As illustrated in Example 14, when you use the inverse function rule, you
obtain an expression for the derivative that’s expressed in terms of y
rather than x. You then have to use the relationship between x and y to
express the derivative in terms of x.
In the next example, the inverse function rule is used to find the derivative
of the function f (x) = sin−1 (x). You saw in Unit 4 that this function has
domain [−1, 1]. Its graph is shown in Figure 22.

48
 3 More differentiation

Example 15 Using the inverse function rule again

Use the inverse function rule, and the derivative of the sine function,
to differentiate y = sin−1 x.
Solution
Express x in terms of y.
We have y = sin−1 x, so
x = sin y.
Differentiate x with respect to y.
Therefore
dx
= cos y.
dy
Use the inverse function rule.
By the inverse function rule,
dy 1
= ,
dx cos y Figure 22 The graph of
provided that cos y )= 0. y = sin−1 x
Use the relationship between x and y to express the derivative in
terms of x. In this case the relationship is given by x = sin y (or
y = sin−1 x), and you need to express cos y in terms of x. You could
write cos y = cos(sin−1 x), but you can obtain a simpler expression by
using the identity sin2 y + cos2 y = 1, as follows.
The identity cos2 y + sin2 y = 1 gives
! (
cos y = ± 1 − sin2 y = ± 1 − x2 .
The + sign applies here, because y takes values only in the interval
[−π/2, π/2] (since y = sin−1 (x)) and so cos y is always non-negative.
Hence
(
cos y = 1 − x2 .
Therefore
dy 1
=√ ,
dx 1 − x2
provided that x )= ±1.

49
Unit 7 Differentiation methods and integration

Notice that the formula for the derivative of sin−1 x found in Example 15
doesn’t apply at the endpoints of the domain of this function, that is,
when x = −1 and x = 1. This is because the function f (x) = sin−1 x isn’t
right-differentiable at x = −1 and isn’t left-differentiable at x = 1, as you
can see, roughly, from the shape of its graph, in Figure 22. The ‘one-sided
tangents’ at these values of x are vertical.
Here’s another point to note about Example 15. Strictly, it involves the
sine function with its domain restricted to the interval [−π/2, π/2], rather
than the sine function with its usual domain, R. As you saw in Unit 4, the
sine function with its usual domain doesn’t have an inverse function.
In the next activity you’re asked to find the derivatives of the functions
f (x) = cos−1 x and f (x) = tan−1 x. You saw in Unit 4 that these functions
have domains [−1, 1] and R, respectively. Their graphs are shown in
Figures 23 and 24, respectively.

Activity 24 Using the inverse function rule

Use the inverse function rule, and the derivatives of the cosine and tangent
Figure 23 The graph of functions, to differentiate the following functions.
y = cos−1 x
(a) y = cos−1 x (b) y = tan−1 x
Hint: To express the derivatives in terms of x, use the identity
sin2 y + cos2 y = 1 in part (a), and the identity tan2 y + 1 = sec2 y in
part (b).

The derivatives found in Example 15 and Activity 24 are repeated below,

Figure 24 The graph of and can now be included in the list of derivatives of standard functions
y = tan−1 x that you have available to work with.

d 1
(sin−1 x) = √
dx 1 − x2
d 1
(cos−1 x) = − √
dx 1 − x2
d 1
(tan−1 x) =
dx 1 + x2

50
 3 More differentiation

In the next activity you’re asked to combine these standard derivatives

with some of the rules for differentiation that you’ve seen, to find the
derivatives of some more functions.

Activity 25 Finding derivatives of functions involving inverse

trigonometric functions

Differentiate the following functions.

(a) f (x) = sin−1 (3x) (b) f (x) = ex cos−1 x

The inverse function rule in Lagrange notation

Let’s now translate the inverse function rule into Lagrange notation. To do
this, we set y = f −1 (x). Then
dy
= (f −1 )" (x),
dx
where (f −1 )" denotes the derivative of f −1 , as you’d expect.
Also, the equation y = f −1 (x) is equivalent to the equation x = f (y), so
dx
= f " (y) = f " (f −1 (x)).
dy
This gives the following form of the inverse function rule.

Inverse function rule (Lagrange notation)

If f is a function with inverse function f −1 , then
1
(f −1 )" (x) = " −1 ,
f (f (x))
for all values of x such that f " (f −1 (x)) exists and is non-zero.

This is a formula for the derivative of f −1 in terms of the derivative of f ,

but it may look rather complicated and difficult to understand. Here’s a
helpful way to understand what it tells you, and why it holds.

51
Unit 7 Differentiation methods and integration

First, it’s important to realise the following fact. If two straight lines
(drawn on axes with equal scales) are reflections of each other in the line
y = x, then their gradients are reciprocals of each other. This is because if
you consider any pair of points on one line, and their reflections on the
other line, then the values of the run and the rise are interchanged for the
two pairs of points, as illustrated in Figure 25(a). The lines in this
diagram have gradients 2 and 12 .
Another fact, which you met in Unit 3 and will be useful here, is that if
two points (plotted on axes with equal scales) are reflections of each other
in the line y = x, then they’re obtained from each other by interchanging
x- and y-coordinates. For example, Figure 25(b) shows the point (3, 1) and
its reflection (1, 3) in the line y = x.

Figure 25 (a) Two straight lines that are reflections of each other in the
line y = x (b) Two points that are reflections of each other in the line
y=x
Now remember that the graphs of a function and its inverse function are
reflections of each other in the line y = x, when they’re drawn on axes with
equal scales. For example, Figure 26 shows the graphs of a function f and
its inverse function f −1 . (The function
√ is in fact f (x) = x2 + 2 (x ≥ 0),
−1
with inverse function f (x) = x − 2.) The two tangents shown are
reflections of each other in the line y = x, so their gradients are reciprocals
of each other. The gradient of the red tangent is the gradient of the graph
of f −1 at the point (3, 1), which is (f −1 )" (3). The gradient of the blue
tangent is the gradient of the graph of f at the point (1, 3), which is f " (1).
Hence
1
(f −1 )" (3) = " .
f (1)
This is a particular instance of the inverse function rule, with x = 3 and
f −1 (x) = f −1 (3) = 1.

52
 3 More differentiation

Figure 26 The graphs of a function f and its inverse function f −1 , and

their tangents at (1, 3) and (3, 1), respectively
In general, consider the graphs of any invertible function f and its inverse
function f −1 , as illustrated in Figure 27. Consider two (non-vertical)
tangents that are reflections of each other in the line y = x, as illustrated.
Their gradients are reciprocals of each other. The gradient of the red
tangent is the gradient of the graph of f −1 at the point (x, f −1 (x)), which
is (f −1 )" (x). The gradient of the blue tangent is the gradient of the graph
of f at the point (f −1 (x), x), which is f " (f −1 (x)). Hence
1
(f −1 )" (x) = .
f " (f −1 (x))
This is the inverse function rule.

Figure 27 The graphs of a function f and its inverse function f −1 , and

their tangents at (f −1 (x), x) and (x, f −1 (x)), respectively

53
Unit 7 Differentiation methods and integration

3.4 Table of standard derivatives

You’ve now reached the end of the material that introduces differential
calculus in this module, though you’ll use the ideas and methods again in
Unit 11, Taylor polynomials. In the next section, you’ll make a start on
learning about integral calculus. To round off your introduction to
differential calculus, here’s a table of the standard derivatives that you’ve
met. You should try to memorise at least the first six (those above the line
in the middle of the table), as they’re used frequently in mathematics.
This table of standard derivatives is also included in the Handbook.

Standard derivatives
Function f (x) Derivative f " (x)
a (constant) 0
xn nxn−1

ex ex
1
ln x
x
sin x cos x
cos x − sin x

tan x sec2 x
cosec x − cosec x cot x
sec x sec x tan x
cot x − cosec2 x
1
sin−1 x √
1 − x2
1
cos−1 x −√
1 − x2
1
tan−1 x
1 + x2

54
 4 Reversing differentiation

The first calculus textbook

In about 1690 the wealthy Marquis de l’Hôpital (1661–1704), a minor
French nobleman, became interested in calculus as a result of journal
articles by Gottfried Wilhelm Leibniz and the Swiss mathematicians
Jacob and Johann Bernoulli. However, these articles contained little
by way of explanation and so l’Hôpital paid Johann Bernoulli, then
living in Paris, to tutor him in the new subject. When Bernoulli
moved from Paris to Groningen to take up a professorship, they came
to an arrangement whereby for a considerable monthly sum Bernoulli
would send l’Hôpital material on calculus, including his new results,
and would deny access to the material to anyone else. In effect,
l’Hôpital bought the rights to Bernoulli’s discoveries.
By 1696 l’Hôpital felt he understood enough of the mathematics to Guillaume De l’Hôpital
publish a book on it, Analyse des Infiniment Petits (Analysis of (1661–1704)
Infinitely Small Quantities). Although in the book l’Hôpital
acknowledges his debt to Bernoulli, after l’Hôpital’s death Bernoulli
complained that l’Hôpital had won undeserved praise for the text.
But because l’Hôpital was known to have been a competent
mathematician and because Bernoulli was known to be a contentious
egotist, the latter’s complaints were generally dismissed. However,
documents discovered in the 20th century have revealed that
Bernoulli was in fact right to claim much of the book for himself.

4 Reversing differentiation
Johann Bernoulli (1667–1748)
In Unit 6 you learned that if you know the values taken by a continuously
changing quantity throughout a period of change, then you can use
differentiation to find the values taken by its rate of change throughout the
same period. For example, you saw that if you have a formula for the
displacement of a moving object in terms of time, then you can use
differentiation to find a formula for its velocity in terms of time. You saw
that the area of calculus that deals with this type of process is called
differential calculus.
In the rest of this unit, and in Unit 8, we’ll consider the opposite process.
We’ll consider situations where you know the values taken by the rate of
change of a continuously changing quantity throughout a period of change,
and you want to find the values taken by the quantity throughout the
same period. For example, you might have a formula or a graph that tells
you the values taken by the velocity of an object (such as the walking man
discussed in Unit 6) over some time period, and you want to use this
information to work out the values taken by the displacement of the object
over the same time period. As mentioned earlier, the area of calculus that
deals with this type of process is called integral calculus. The title page of Analyse des
Infiniment Petits (1696)
55
Unit 7 Differentiation methods and integration

As you’ll see, when you’re working with integral calculus, it can be useful
to consider whether a function is continuous. A continuous function is
one whose graph has no discontinuities (that is, ‘breaks’) in it. In other
words, informally, it’s a function whose graph you can draw without taking
your pen off the paper (though you might need an infinitely large piece of
paper, as the graph might be infinitely long!). For example, the function
f (x) = 2x is a continuous function, as shown in Figure 28(a), but the
function f (x) = 1/x2 isn’t, as its graph has a break at x = 0, as shown in
Figure 28(b).

4.1 Antiderivatives and indefinite integrals

In integral calculus you usually start with a function f , and you want to
find another function, say F , whose derivative is f . Such a function F is
called an antiderivative of the original function f . For example, an
antiderivative of the function f (x) = 2x is the function F (x) = x2 , because
the derivative of x2 is 2x.
The process of finding an antiderivative of a function is called
antidifferentiation, or, more commonly, integration. So integration is
the reverse of differentiation.
The first thing to realise is that a function can have more than one
antiderivative. To see this, try the following activity.

Figure 28 The graphs of Activity 26 Differentiating functions whose formulas are the same
(a) f (x) = 2x apart from a constant term
(b) f (x) = 1/x2
Differentiate the following functions.
5
(a) F (x) = x2 (b) F (x) = x2 + 3 (c) F (x) = x2 − 7

You should have found that all three functions in Activity 26 have
derivative F " (x) = 2x. In fact, you can see that any function of the form
F (x) = x2 + c,
where c is a constant, has derivative F " (x) = 2x. This is because the
derivative of a constant term is zero. So each function of the
form F (x) = x2 + c is an antiderivative of the function f (x) = 2x.
Another way to think about the fact that all the functions of the form
F (x) = x2 + c have the same derivative is to remember that adding a
constant to the formula of a function has the effect of translating its graph
vertically. So the graphs of all the functions of the form F (x) = x2 + c are
vertical translations of each other, as illustrated in Figure 29(b).
Translating a graph vertically doesn’t change its gradient at each x-value,
of course. In other words, it doesn’t change the derivative of the function
that the graph represents. So each of the functions in Figure 29(b) has the
same derivative, namely f (x) = 2x, whose graph is shown in Figure 29(a).

56
 4 Reversing differentiation

Another thing to appreciate about all the functions of the form

F (x) = x2 + c, where c is a constant, is that these functions are the only
antiderivatives of the function f (x) = 2x. That’s because any
antiderivative of f (x) = 2x has the same gradient at each x-value as the
function F (x) = x2 , and the functions of the form F (x) = x2 + c are the
only functions with this property.

Figure 29 The graphs of (a) a function (b) some of its antiderivatives

So the formula F (x) = x2 + c describes the complete family of
antiderivatives of the function f (x) = 2x. We call the general
function F (x) = x2 + c the indefinite integral of the function f (x) = 2x.
The word ‘indefinite’ refers to the fact that the constant c can take any
value.
In general, consider any function f that has an antiderivative. You can see
that you can add any constant that you like to the formula for the
antiderivative, and you’ll get the formula for another antiderivative of f .
Equivalently, you can translate the graph of the antiderivative vertically by
any amount, and you’ll get the graph of another antiderivative of f .
These antiderivatives are the only antiderivatives of the function f ,
provided that f is a continuous function.
So, if f is a function that has an antiderivative, and f is continuous, then
the general function
F (x) = (formula for any particular antiderivative of f ) + c,
where c represents any constant, and whose domain is the same as the
domain of f , describes the complete family of antiderivatives of f . We call
this general function F the indefinite integral of the function f .
The constant c in an indefinite integral is called an arbitrary constant,
or the constant of integration. It’s usually denoted by c, but you can
use any letter. Like the word ‘indefinite’ in ‘indefinite integral’, the word
‘arbitrary’ in ‘arbitrary constant’ refers to the fact that the constant c can
take any value.

57
Unit 7 Differentiation methods and integration

You don’t need to be concerned about what happens when the function f
isn’t continuous, because normally when we’re solving problems in integral
calculus we work only with continuous functions. These are usually the
only functions that we need for the sorts of calculations that we want to
perform. However, if you’d like to know what the problem is with
functions that aren’t continuous, then read the explanation at the end of
this subsection, when you reach it.
Before you go on, it’s important to make sure that you fully understand
the difference between an antiderivative and an indefinite integral of a
function. The definitions that you’ve seen are summarised below.

Antiderivatives and indefinite integrals

Suppose that f is a function.
An antiderivative of f is any specific function whose derivative is f .
If f has an antiderivative, and f is continuous, then the indefinite
integral of f is the general function obtained by adding an arbitrary
constant c to the formula for an antiderivative of f . It describes the
complete family of antiderivatives of f .

Example 16 Understanding antiderivatives and indefinite integrals

(a) Show that the function F (x) = x3 is an antiderivative of the
function f (x) = 3x2 .
(b) What is the indefinite integral of the function f (x) = 3x2 ?
Solution
(a) Since
d
(x3 ) = 3x2 ,
dx
it follows that F (x) = x3 is an antiderivative of f (x) = 3x2 .
(b) The indefinite integral of f (x) = 3x2 is F (x) = x3 + c.

58
 4 Reversing differentiation

Activity 27 Understanding antiderivatives and indefinite integrals

1
(a) Show that the function F (x) = 2 sin(2x) is an antiderivative of the
function f (x) = cos(2x).
(b) What is the indefinite integral of the function f (x) = cos(2x)?
(c) Write down an antiderivative of the function f (x) = cos(2x) other
than the antiderivative in part (a).

Even though the idea of an indefinite integral really applies only to

continuous functions, it’s convenient in practice to state indefinite integrals
of other functions. For example, consider again the function f (x) = 1/x2 ,
which isn’t continuous. Its graph is repeated in Figure 30. An
antiderivative of this function is the function F (x) = −1/x, as you can
check by differentiating this function F . Even though f isn’t continuous,
we do still say
the function f (x) = 1/x2 has indefinite integral F (x) = −1/x + c.
This is shorthand for
any continuous function f with rule f (x) = 1/x2 has indefinite
integral F with rule F (x) = −1/x + c.
For example, the continuous function f (x) = 1/x2 (x > 0), whose graph is
shown in Figure 31, has indefinite integral F (x) = −1/x + c (x > 0).
Figure 30 The graph of the
We use this shorthand for any function that isn’t continuous. function f (x) = 1/x2
As you’ve seen, to find an indefinite integral of a function, you just find a
formula for any antiderivative of the function, and add an arbitrary
constant, usually c.
You can find antiderivatives of many functions by using what you already
know about derivatives. In the next two subsections you’ll practise finding
antiderivatives, and hence indefinite integrals, of some simple functions in
this way. Before you make a start on that, here are a few things that you
need to know.
First, note that when we’re discussing antiderivatives and indefinite
integrals, we use conventions similar to those for derivatives. For example,
the terms ‘antiderivative’ and ‘indefinite integral’ can be applied directly
to expressions that could appear on the right-hand side of the rule of a
function, as well as to the function itself. For instance, rather than saying
that
the indefinite integral of f (x) = 2x is F (x) = x2 + c, Figure 31 The graph of the
function f (x) = 1/x2 (x > 0)
we can simply say that
the indefinite integral of 2x is x2 + c.
Also, as you’d expect, you can use any letters for functions and variables –
you’re not restricted to the standard ones, f and x.

59
Unit 7 Differentiation methods and integration

When we’re working with antiderivatives and indefinite integrals in the

rest of this unit, we’ll usually use a lower-case letter, such as f , to denote
the function that we start with, and the corresponding upper-case letter,
such as F , to denote any of its antiderivatives, or its indefinite integral.
You’ll meet a more adaptable, though more complicated, notation" for
indefinite integrals in Unit 8. It involves the integral sign, , which you
may have seen elsewhere.

Functions that aren’t continuous

If you’d like to know what the problem is with functions that aren’t
continuous, then read the discussion below. Otherwise just go on to
Subsection 4.2.
A function f that has an antiderivative but which isn’t continuous may
have other antiderivatives as well as those obtained by vertically
translating the graph of the original antiderivative. For example, consider
the function f (x) = 1/x2 , whose graph is shown in Figure 32(a). One
antiderivative of this function f is the function F (x) = −1/x, as mentioned
above. The graph of this function F is shown in Figure 32(b). Every
function obtained by vertically translating this graph is an antiderivative
of f , but so is every function obtained by vertically translating each of the
two pieces of the graph individually, since that doesn’t change the gradient
at each x-value. For example, Figure 32(c) shows an antiderivative of f
obtained by translating the left-hand piece of the graph of F (x) = −1/x
down by 2 units, and the right-hand piece up by 1 unit.

Figure 32 The graphs of (a) f (x) = 1/x2 (b) the antiderivative

F (x) = −1/x of f (x) = 1/x2 (c) another antiderivative of f (x) = 1/x2
So if f is a function that has an antiderivative, but which isn’t continuous,
then the general function
F (x) = (formula for any particular antiderivative of f ) + c,
where c represents any constant, and whose domain is the same as the
domain of f , usually doesn’t describe the complete family of antiderivatives
of f , as stated earlier. As mentioned earlier, this isn’t a problem, because
normally when we’re solving problems in integral calculus we work only
with continuous functions.

60
 4 Reversing differentiation

4.2 Antiderivatives of power functions

In this section you’ll see how to find antiderivatives, and hence indefinite
integrals, of power functions.
You can find antiderivatives of power functions by using what you already
know about derivatives of power functions. For example, suppose that you
want to find an antiderivative of the power function f (x) = x6 . Consider
what happens when you differentiate the power function whose exponent
is 1 more than 6, namely g(x) = x7 :
d % 7$
x = 7x6 .
dx
This equation tells you that x7 is ‘nearly’ an antiderivative of x6 . You can
deduce from it that if you multiply x7 by the constant 71 , then
differentiating will give the answer that you want, x6 :
) -
d 1 7 1
x = × 7x6 = x6 .
dx 7 7
So an antiderivative of f (x) = x6 is F (x) = 17 x7 . Hence the indefinite
integral of f (x) = x6 is
F (x) = 71 x7 + c.

You can find a general formula for an antiderivative of a power function by

applying the thinking above to the general power function f (x) = xn .
Differentiating the power function g(x) = xn+1 gives
d % n+1 $
x = (n + 1)xn .
dx
So if you multiply xn+1 by 1/(n + 1), then differentiating will give the
answer that you want, xn :
) -
d 1 n+1 1
x = × (n + 1)xn = xn .
dx n + 1 n+1
This equation gives a general formula for an antiderivative, and hence the
indefinite integral, of a power function, which is stated below. The formula
applies for every value of the exponent n, except n = −1. It doesn’t apply
for n = −1 because division by zero is undefined. (The case n = −1 is
more complicated, and is dealt with in Subsection 6.1.)

Indefinite integral of a power function

For any number n except n = −1,
1
the indefinite integral of xn is xn+1 + c.
n+1

Here’s a helpful informal way to remember this formula.

61
Unit 7 Differentiation methods and integration

Indefinite integral of a power function (informal)

To find an antiderivative of a power function (except the power
function f (x) = x−1 ), first increase the power by 1, then divide by the
new power.
As usual, to obtain the indefinite integral, add the arbitrary
constant c.

Example 17 Finding indefinite integrals of power functions

Find the indefinite integrals of the following functions.
1 √
(a) f (x) = x10 (b) g(x) = 5 (c) h(x) = x
x
Solution
(a) To find an antiderivative, increase the power by 1, then divide
by the new power. To obtain the indefinite integral, add the
arbitrary constant c.
The indefinite integral is
1
F (x) = x11 + c.
11
(b) First write the function in the form g(x) = xn .
The function is g(x) = x−5 .
Proceed as in part (a).
The indefinite integral is
1 −4
G(x) = x +c
−4
Simplify the answer.
1
= − 4 + c.
4x
(c) First write the function in the form y = xn .
The function is h(x) = x1/2 .
Proceed as in part (a).
The indefinite integral is
1 3/2
H(x) = x +c
3/2

62
 4 Reversing differentiation

Simplify the answer.

= 23 x3/2 + c.

Activity 28 Finding indefinite integrals of power functions

Find the indefinite integrals of the following functions.

1
(a) f (x) = x9 (b) p(u) = 3 (c) f (x) = x2/3
u
√ 1
(d) f (x) = 4 x (e) g(t) = t−2/3 (f) b(v) = √
v

Activity 29 Finding different antiderivatives of a function

Let f be the function with rule f (x) = x5 .

(a) Find the indefinite integral of f .
(b) Hence write down three different antiderivatives of f .

Notice that the formula for the indefinite integral of a power function tells
you that the indefinite integral of the function
f (x) = 1 (which is the same as f (x) = x0 )
is
F (x) = x + c.
(As in earlier units, we assume that 00 = 1, so the formula works when
x = 0.) Another way to obtain the indefinite integral above is, of course, to
use the fact that the derivative of x is 1.
In Subsection 6.1 later in the unit, you’ll see a formula for the indefinite
integral of the only power function that isn’t covered by the general
formula that you’ve met in this subsection. This is the power function
f (x) = x−1 , that is, the reciprocal function f (x) = 1/x.

63
Unit 7 Differentiation methods and integration

4.3 Constant multiple rule and sum rule for

antiderivatives
Just as with derivatives, once you have formulas for antiderivatives of some
functions, you can combine them to obtain formulas for antiderivatives of
further functions.
Unfortunately, though, it’s generally more tricky to combine
antiderivatives than it is to combine derivatives. For example, there’s no
straightforward rule that you can use to find an antiderivative of a product
of two functions if you already know an antiderivative of each of the two
individual functions. In other words, there’s no ‘product rule for
antiderivatives’. Similarly, there’s no ‘quotient rule for antiderivatives’ and
no ‘chain rule for antiderivatives’.
However, there is a constant multiple rule and a sum rule for
antiderivatives, and they work in the same simple ways as the similar rules
for derivatives, as follows.

Constant multiple rule for antiderivatives

If F (x) is an antiderivative of f (x), and k is a constant, then kF (x) is
an antiderivative of kf (x).

Sum rule for antiderivatives

If F (x) and G(x) are antiderivatives of f (x) and g(x), respectively,
then F (x) + G(x) is an antiderivative of f (x) + g(x).

(You might have noticed that the letter used for the constant in the
statement above of the constant multiple rule for antiderivatives, namely k,
is different from the letter used in Unit 6 for the constant in the constant
multiple rule for derivatives, which was a. It doesn’t matter which letter is
used, of course, but in integral calculus it’s sometimes convenient to
reserve the letter a to denote something else, as you’ll see in Unit 8.)
These rules follow directly from the similar rules for derivatives. To see
why, first suppose that F (x) is an antiderivative of f (x). Then
d
(F (x)) = f (x).
dx
It follows from the constant multiple rule for derivatives that, for any
constant k,
d
(kF (x)) = kf (x).
dx
This equation tells you that kF (x) is an antiderivative of kf (x), which is
what the constant multiple rule for antiderivatives says.

64
 4 Reversing differentiation

Now suppose that F (x) and G(x) are antiderivatives of f (x) and g(x),
respectively. Then
d d
(F (x)) = f (x) and (G(x)) = g(x).
dx dx
It follows from the sum rule for derivatives that
d
(F (x) + G(x)) = f (x) + g(x).
dx
This equation tells you that F (x) + G(x) is an antiderivative of
f (x) + g(x), which is what the sum rule for antiderivatives says.
Remember that, as always when a constant multiple rule and a sum rule
hold, the sum rule also holds if you replace the plus signs by minus signs.
That is, if F (x) and G(x) are antiderivatives of f (x) and g(x), respectively,
then F (x) − G(x) is an antiderivative of f (x) − g(x).
Here’s an example of how you can use the two rules above.

Example 18 Using the constant multiple rule and the sum rule for
antiderivatives
Find the indefinite integrals of the following functions.
6 √
(a) f (x) = 10x4 (b) g(t) = 3 − 5 t + 1
t
Solution
(a) Find an antiderivative by using the constant multiple rule,
together with the rule for finding an antiderivative of a power
function. To obtain the indefinite integral, add the arbitrary
constant c.
The indefinite integral is
F (x) = 10 × 15 x5 + c
= 2x5 + c.
(b) First express the function as a sum of constant multiples of
power functions.
The function is
g(t) = 6t−3 − 5t1/2 + 1.
By the sum rule, you can find an antiderivative by working
term by term. Use the method of part (a) to find an
antiderivative of each term. To obtain the indefinite integral, add
the arbitrary constant c.

65
Unit 7 Differentiation methods and integration

The indefinite integral is

1 −2 1 3/2
G(t) = 6 × t −5× t +t+c
−2 3/2
= −3t−2 − 5 × 23 t3/2 + t + c
3 10t3/2
=− − + t + c.
t2 3

Notice in particular that it follows from the constant multiple rule for
antiderivatives that, for any constant a, the indefinite integral of the
function
f (x) = a (which is the same as f (x) = a × 1)
is
F (x) = ax + c.
Another way to see this is, of course, to use the fact that the derivative of
ax is a.
This useful fact is summarised below.

Indefinite integral of a constant function

The indefinite integral of the constant a is ax + c.

You can practise using the constant multiple rule and the sum rule for
antiderivatives in the activity below. In parts (d)–(g), you have to begin
by using an algebraic technique, such as multiplying out brackets or
expanding a fraction, to express the function as a sum of constant
multiples of power functions.
Remember that you must include an arbitrary constant c in every
indefinite integral that you find.

Activity 30 Using the constant multiple rule and the sum rule for
antiderivatives
Find the indefinite integrals of the following functions.
√
(a) f (x) = 8x3 (b) f (x) = 7 (c) f (x) = x − x2
(d) f (x) = (x + 2)(x − 5) (e) g(t) = (t + 1)2
) -
x2 + 1 √
3 2 1
(f) g(x) = (g) f (r) = r 10r − 2
x4 r

You’ll meet some more ways to combine antiderivatives in Unit 8.

66
 5 Particular antiderivatives

5 Particular antiderivatives
Sometimes when you’re working with a function, you need to find a
particular antiderivative that it has, rather than any antiderivative, or its
indefinite integral. For example, this is often what you need to do when
the function is a mathematical model of a real-life situation. In this
section, you’ll learn how to find such particular antiderivatives, and see
some examples of how you can use them.

5.1 Finding particular antiderivatives

You can work out which of the infinitely many antiderivatives of a function
is the right one if you have some appropriate extra information. Usually
the extra information that you have is the value taken by the
antiderivative at some value of the input variable. Here’s an example.

Example 19 Finding a particular antiderivative

Find the antiderivative F of the function f (x) = x2 + 5 such that
F (3) = 20.
Solution
Find the indefinite integral.
The indefinite integral is
F (x) = 13 x3 + 5x + c.
Use the extra information to find the required value of the
constant c.
Using the fact that F (3) = 20 gives
1
3× 33 + 5 × 3 + c = 20
9 + 15 + c = 20
c = −4.
Hence the required antiderivative is
F (x) = 31 x3 + 5x − 4.

Here are some similar examples for you to try.

67
Unit 7 Differentiation methods and integration

Activity 31 Finding particular antiderivatives

(a) Find the antiderivative F of the function f (x) = x + 2 such that

F (1) = 29 .
(b) Find the antiderivative F of the function
f (x) = 1/x3 (x ∈ (0, ∞))
such that F (3) = − 19 .

5.2 Using integration to work with rates of change

As mentioned at the start of this section, it’s often useful to find a
particular antiderivative when you’re working with a function that models
a real-life situation. This is illustrated in the example below.
In this example, we’ll consider again a situation that you met in Unit 6,
namely the one where a man walks along a straight path, gradually
slowing down as he walks. In Section 3.1 of Unit 6, we started with the
equation for the man’s displacement in terms of time, and we used
differentiation to find an equation for his velocity (his rate of change of
displacement) in terms of time. Here we’ll do the opposite: we’ll start with
the equation for his velocity in terms of time, and we’ll use integration to
find an equation for his displacement in terms of time. Remember that
integration is the opposite process to differentiation – it’s the process of
finding an antiderivative of a function.

Example 20 Using integration to deduce displacement from velocity

Suppose that a man walks along a straight path, and his velocity v
(in kilometres per hour) at time t (in hours) after he begins his walk
is given by the equation
v = 6 − 45 t.
Let s be his displacement in kilometres from his starting point.

(a) Find an equation that expresses s in terms of t.

(b) Hence find the man’s displacement two hours after he began his
walk.

68
 5 Particular antiderivatives

Solution
(a) Write down the given equation for the man’s velocity.
The man’s velocity v at time t is given by
v = 6 − 45 t.
Use integration to find an equation for his displacement. It
will contain an arbitrary constant, because at this stage we don’t
know which antiderivative is the right one.
Hence his displacement s at time t is given by
s = 6t − 5
4 × 21 t2 + c,
that is,
s = 6t − 58 t2 + c,
where c is a constant.
Use extra information to find the value of the constant c.
When the man begins his walk, his displacement from his starting
point is 0 km. That is, when t = 0, s = 0. Substituting these
values into the equation above gives
0=6×0− 5
8 × 02 + c,
that is,
c = 0.
Hence the required equation for s in terms of t is
s = 6t − 85 t2 .
(b) When t = 2,
s=6×2− 5
8 × 22 = 12 − 5
2 = 19
2 = 9.5.
So the man’s displacement two hours after he began his walk is
9.5 km.

The calculation in Example 20 is the sort of calculation that integral

calculus allows you to carry out: if you know the values taken by the rate
of change of a quantity throughout a period of change, then you can use
integration to find the values taken by the quantity throughout the same
period.
Here’s something else that’s illustrated by Example 20. If the only
information that you have about the motion of an object is an equation for
its velocity in terms of time, then you don’t have enough information to
find an equation for its displacement in terms of time. The best that you
can do is to find an equation for its displacement in terms of time that
contains an arbitrary constant. To find a definite equation (that is, to

69
Unit 7 Differentiation methods and integration

choose the correct function from a family of functions whose graphs are
vertical translations of each other), you need to use further information,
which is usually a fact about the displacement of the object at some
moment in time.
This is just what you’d expect: if all you know is an equation for the
velocity of the object in terms of time, then you don’t have enough
information to determine its displacement in terms of time, because it
could have started anywhere!
In this sort of situation you do, however, have enough information to find
the change in the displacement of the object between any two points in
time. You’ll see some examples of this later in this subsection.
Here’s an example similar to Example 20 for you to try.

Activity 32 Using integration to deduce displacement from velocity

Suppose that a marble is rolled in a straight line down a long slope, and its
velocity v (in metres per second) at time t (in seconds) after it begins
rolling is given by the equation
1
v= 10 t.
The marble starts rolling from a position that is 0.3 metres down the slope.
Let s be its displacement in metres from the top of the slope.
(a) Find an equation that expresses s in terms of t.
(Hint: First find an equation for s in terms of t that contains an
arbitrary constant, then use the value of s when t = 0.)
(b) Hence find the displacement of the marble from the top of the slope
4 seconds after it begins rolling.

Let’s now consider another example of motion in a straight line from

Unit 6. You saw that if an object falls from rest under the influence of
gravity, and the effects of air resistance are negligible, then its
displacement s (in metres) at time t (in seconds) after it began falling is
modelled by the equation
s = −4.9t2 .
Here the displacement of the object is measured from the point from which
it starts to fall, and the positive direction along the line of motion is
upwards.
You saw that if you differentiate this equation, to obtain an equation for
the object’s velocity in terms of time, and then differentiate again to
obtain an equation for its acceleration in terms of time, then you obtain
the fact that the object is moving with a constant acceleration of
−9.8 m s−2 . The magnitude of this acceleration, 9.8 m s−2 , is known as the
acceleration due to gravity.

70
 5 Particular antiderivatives

In the next example, we’ll carry out this process in reverse. We’ll start by
assuming that the effect of gravity on any object falling from rest is to
produce a constant acceleration of −9.8 m s−2 . We’ll use this fact to write
down an equation for the object’s acceleration in terms of time, integrate it
to obtain an equation for the object’s velocity in terms of time, and then
integrate again to obtain an equation for the object’s displacement in
terms of time. You’ll see that the result is the equation stated above.

Example 21 Using integration to deduce displacement from

acceleration
Assume that an object falling from rest has a constant acceleration of
−9.8 m s−2 , where the positive direction is upwards.
Let s be the displacement in metres of the object from its initial
position and let t be the time in seconds since it began falling.
Find an equation that expresses s in terms of t.
Solution
Define the variables that you intend to use.
Let v be the velocity of the object in m s−1 and let a be its
acceleration in m s−2 .
Write down the equation for the acceleration of the object.
The equation for a in terms of t is
a = −9.8.
The variable t doesn’t appear on the right-hand side of this
equation because the acceleration is constant.
Use integration to find an equation for v in terms of t, containing
an arbitrary constant.
Hence the equation for v in terms of t is
v = −9.8t + c,
where c is a constant.
Use extra information to find the value of the constant c.
The object falls from rest, so when it begins falling its velocity is
0 m s−1 . That is, when t = 0, v = 0. Substituting these values into the
equation above gives
0 = −9.8 × 0 + c, that is, c = 0.
So the equation for v in terms of t is
v = −9.8t.

71
Unit 7 Differentiation methods and integration

Use integration again to find an equation for s in terms of t,

containing another arbitrary constant. Don’t use the letter c for the
arbitrary constant, as it’s been used already. Choose a different letter,
such as b.
It follows that the equation for s in terms of t is
s = −9.8 × 21 t2 + b,
that is,
s = −4.9t2 + b,
where b is a constant.
Use extra information to find the value of the constant b.
When the object begins falling, its displacement is 0 m. That is, when
t = 0, s = 0. Substituting these values into the equation gives
0 = −4.9 × 02 + b, that is, b = 0.
So the equation for s in terms of t is
s = −4.9t2 .
This is the required equation for the displacement of the object in
terms of time.

The next activity is about the motion of a ball thrown vertically into the
air. The ball is an example of a vertical projectile. A projectile is an
object that’s launched into the air by a force that ceases after launch, and
a vertical projectile is a projectile that’s launched vertically upwards.
Provided that the effects of air resistance are negligible, a vertical
projectile has a constant downwards acceleration due to gravity of about
9.8 m s−2 at all times after its launch, just as an object falling from rest
has. The only thing that makes its motion different from that of an object
falling from rest is that at the start of its motion its velocity is not zero. In
the next activity you’re asked to use a process similar to that in
Example 21 to find a formula for the displacement of a ball thrown
vertically into the air, in terms of time.

Activity 33 Using integration to deduce displacement from

acceleration
Suppose that a ball is thrown vertically upwards with initial speed
12 m s−1 . Assume that its subsequent motion is modelled as having a
constant acceleration of −9.8 m s−2 , where the positive direction along the
line of motion is upwards.

72
 5 Particular antiderivatives

Let the acceleration, velocity and displacement of the ball at time t (in
seconds) after it was thrown be a (in m s−2 ), v (in m s−1 ) and s (in m),
respectively, where displacement is measured from the point from which
the ball was thrown.
(a) Find an equation that expresses v in terms of t.
(b) Hence find an equation that expresses s in terms of t.
(c) Use the formulas that you found in parts (a) and (b) to find the
velocity and the displacement of the ball 1 second after it was thrown.
(d) Use the formula that you found in part (b) to determine how long it
takes for the ball to fall back to the point from which it was thrown.
Hint: At the time when the ball has fallen back to this point, what is
the value of s?

The next activity involves an economic model of a type that you met in
Unit 6. You saw there that marginal cost is the derivative of total cost.

Activity 34 Using integration to work with marginal cost and total

cost
A company that produces agricultural fertiliser has developed a model for
its production costs. According to the model, if the amount of fertiliser
that the company is currently producing each week is q (in tonnes), then
the marginal weekly cost m (in £ per tonne) of producing extra fertiliser is
given by the equation
m = 250 − 0.5q.
The model applies for values of q between 50 and 200.
Currently the company produces 80 tonnes of fertiliser each week, at a
total weekly cost of £30 400.
Let the total weekly cost of producing q tonnes of fertiliser be t (in £).
(a) By using the fact that marginal cost is the derivative of total cost, find
a formula for t in terms of q.
(b) Use the formula that you found in part (a) to find the total weekly
cost of producing 160 tonnes of fertiliser.
(c) What is the cost per tonne (in other words, the unit cost, not the
marginal cost) of producing the fertiliser if the company produces
80 tonnes of fertiliser each week? What is it if it produces 160 tonnes
each week?

73
Unit 7 Differentiation methods and integration

5.3 Changes in the values of antiderivatives

There’s a fact about the antiderivatives of a continuous function that it’s
important to appreciate.
To understand this fact, consider again the function f (x) = 2x, and its
antiderivatives, the functions of the form F (x) = x2 + c, whose graphs are
shown in Figure 33.

Figure 33 The graphs of a particular continuous function and some of its

antiderivatives
Let’s choose any two numbers in the domain of the function f , say 1
and 2, as shown in Figure 34(a), and consider the amounts by which the
antiderivatives of f change as x changes from the first number to the
second number. Figure 34(b) illustrates the amount of the change for each
of two different antiderivatives of f . You can see that the amount of the
change appears to be 3 units in each case.
The important fact to appreciate is that the amount of the change is the
same for all the antiderivatives of f . That’s because the graphs of all the
antiderivatives are vertical translations of each other, and translating a
graph vertically doesn’t affect how much it changes from one value of x to
another.
Another way to see that all the antiderivatives of the function f (x) = 2x
change by the same amount as x changes from x = 1 to x = 2 is to use
algebra. Consider any antiderivative F of f . Then F is of the form
F (x) = x2 + c, where c is some constant. The value of F when x = 1 is
F (1) = 12 + c,
and the value of F when x = 2 is
F (2) = 22 + c.
So the change in the value of F as x changes from x = 1 to x = 2 is
F (2) − F (1) = (22 + c) − (12 + c)
=4+c−1−c
= 3.

74
 5 Particular antiderivatives

Figure 34 (a) The numbers 1 and 2 in the domain of f (x) = 2x (b) The
amounts that two different antiderivatives of f (x) = 2x change as x
changes from 1 to 2
You can see that no matter what the value of c is, the two occurrences of it
will cancel out when you calculate F (2) − F (1). Hence you’ll always get
the same answer for F (2) − F (1), namely 3.
In general, we have the following important fact.

Suppose that f is a function that has an antiderivative, and f is

continuous. If a and b are numbers in the domain of f , then all the
antiderivatives of f change by the same amount as x changes from
x = a to x = b. In other words, the value of F (b) − F (a) is the same
for every antiderivative F of f .

So if you want to work out the amount by which an antiderivative of a

continuous function f changes from x = a to x = b, then you just need to
find any antiderivative F of f , and calculate F (b) − F (a).
You can use this method to solve problems in which you know the values
taken by the rate of change of a quantity over some period of change, and
you want to find the amount by which the quantity changes from one point
in this period to another. Here’s an example.

75
Unit 7 Differentiation methods and integration

Example 22 Finding a change in the value of an antiderivative

Suppose that the rate of change of a quantity is given by the function
f (x) = 15 x2 (x ≥ 0), as shown in Figure 35. By what amount does the
quantity change from x = 3 to x = 6?
Solution
Find any antiderivative F of f and calculate F (6) − F (3).
An antiderivative of f is
F (x) = 1
5 × 31 x3 = 1 3
15 x .
Figure 35 The graph of So the change in the quantity from x = 3 to x = 6 is
y = 15 x2 (x ≥ 0)
1
F (6) − F (3) = ( 15 × 63 ) − ( 15
1
× 33 )
72 9
= 5 − 5
63
= 5 = 12.6.

Here’s a similar activity for you to try.

Activity 35 Finding a change in the value of an antiderivative

Figure 36 The graph of Suppose that the rate of change of a quantity is given by the function
f (x) = ex/8 (x ≥ 0) f (x) = ex/8 (x ≥ 0), as shown in Figure 36. Find the amount by which the
quantity changes from x = 2 to x = 5, to three significant figures.

In the next example and activity, the changing quantity is the

displacement of a moving object, and hence the rate of change of the
quantity is the velocity of the object.
The example is about a car whose velocity is increasing, as shown in
Figure 37.

Example 23 Finding a change in displacement

Suppose that a car begins to move along a straight road, and its
velocity v (in m s−1 ) during the first five seconds of its journey is
given by v = f (t), where
f (t) = 3t.
Here t is the time in seconds since the car began moving. How far
does the car travel from the end of the fourth second to the end of the
Figure 37 The graph of fifth second?
v = 3t (0 ≤ t ≤ 5)

76
 5 Particular antiderivatives

Solution
Find any antiderivative F of f and calculate F (5) − F (4).
An antiderivative of f is
F (t) = 23 t2 .
So the change in the displacement of the car from t = 4 to t = 5 is
F (5) − F (4) = ( 23 × 52 ) − ( 23 × 42 )
= 37.5 − 24
= 13.5.
That is, the car travels 13.5 m (in the positive direction) during the
fifth second.

The activity below is about an object whose velocity is decreasing, and

then becomes negative, as shown in Figure 38; that is, it starts to move in
the opposite direction, as shown in Figure 39.

Figure 39 The object whose motion is given by Figure 38 is moving in

the positive direction at time t = 0, then turns at time t = 2 to move in
the negative direction

Activity 36 Finding changes in displacement

Figure 38 The graph of
v = 3 − 32 t (0 ≤ t ≤ 6)
Suppose that an object moves along a straight line, and its velocity v (in
m s−1 ) at time t (in seconds) is given by v = f (t), where f (t) = 3 − 23 t
(0 ≤ t ≤ 6), as shown in Figure 38.
(a) By what amount does the displacement of the object change from time
t = 0 to time t = 1? What is the distance that the object travels in
that time?
(b) Repeat part (a) for the times t = 4 and t = 5.

The ideas that you’ve met in this subsection will be important in Unit 8.

77
Unit 7 Differentiation methods and integration

6 Antiderivatives of further standard

functions
In this section you’ll meet and use formulas for antiderivatives of some
further standard functions. These formulas are all found by using formulas
for derivatives that you met earlier in this unit.

6.1 An antiderivative of the reciprocal function

We’ll begin by finding a formula for an antiderivative of the reciprocal
function f (x) = 1/x, which is the only power function that isn’t covered by
the general formula for the indefinite integral of a power function that you
met in Subsection 4.2.
Here’s how you can find an antiderivative of this function, by using what
you already know about differentiation.
Remember that you saw earlier in this unit that
d 1
(ln x) = . (1)
dx x
Unfortunately this formula doesn’t immediately give you a formula for an
antiderivative of the function f (x) = 1/x, because the function f (x) = 1/x
has domain (−∞, 0) ∪ (0, ∞), but the formula applies only when x is
positive (because ln x is defined only when x is positive). However, the
formula does tell you that an antiderivative of the function
1
g(x) = (x > 0)
x
is
G(x) = ln x.
This fact is illustrated in Figure 40. The graph of g(x) = 1/x (x > 0), in
Figure 40(a), gives the gradients of the graph of G(x) = ln x, in
Figure 40(b).

Figure 40 The graphs of (a) g(x) = 1/x (x > 0) and (b) G(x) = ln x

78
 6 Antiderivatives of further standard functions

To deal with the ‘other half’ of the function f (x) = 1/x, you can use the
fact that equation (1), together with the chain rule, gives
d 1 1
(ln(−x)) = × (−1) = .
dx −x x
This formula applies for negative values of x, because ln(−x) is defined
when −x is positive, that is, when x is negative. So it tells you that an
antiderivative of the function
h(x) = 1/x (x < 0)
is
H(x) = ln(−x).
This fact is illustrated in Figure 41. The graph of h(x) = 1/x (x < 0), in
Figure 41(a), gives the gradients of the graph of H(x) = ln(−x), in
Figure 41(b).

Figure 41 The graphs of (a) h(x) = 1/x (x < 0) and (b) H(x) = ln(−x)
You can obtain an antiderivative for the function f (x) = 1/x, with its
whole domain (−∞, 0) ∪ (0, ∞), by putting together the antiderivatives
found for the two ‘halves’ of f . The graph of this antiderivative is shown in
Figure 42(b). The graph of f (x) = 1/x is shown in Figure 42(a): it gives
the gradients of the graph in Figure 42(b).

Figure 42 The graphs of (a) f (x) = 1/x and (b) F (x) = ln x and
F (x) = ln(−x)
79
Unit 7 Differentiation methods and integration

The formula for the antiderivative of f (x) = 1/x found above can be
written as
/
ln(−x) (x < 0)
F (x) = (2)
ln x (x > 0).
However, there’s a more concise way to write it, using the notation | | for
the modulus of a number. Notice that if x is a negative number, then
−x = |x|
(for example, −(−3) = 3 = | − 3|). Also, if x is a positive number, then
x = |x|
(for example, 3 = |3|). So formula (2) can be written concisely as
F (x) = ln |x|.
Hence the indefinite integral of f (x) = 1/x is
F (x) = ln |x| + c.

This formula is sometimes useful, but often when you’re working with the
function f (x) = 1/x, the variable x takes only positive values. If this is the
case, then you can just use the antiderivative of the function f (x) = 1/x
(x > 0), which has the simpler formula
F (x) = ln x + c.

The facts that you’ve seen so far in this section are summarised below.

Indefinite integral of the reciprocal function

1
The indefinite integral of is ln |x| + c.
x
1
(If x takes only positive values, then the indefinite integral of is
x
simply ln x + c.)

In the next activity you’re asked to use this formula, together with the
constant multiple rule and the sum rule for antiderivatives, to find the
indefinite integrals of some further functions.

Activity 37 Finding indefinite integrals of more functions

Find the indefinite integrals of the following functions.

4 x+1 1
(a) f (x) = (b) f (x) = 2
(c) f (x) = (x > 0)
x x 2x

80
 6 Antiderivatives of further standard functions

6.2 Table of standard indefinite integrals

You can use all of the derivatives of standard functions that you met in
Unit 6 and in the first three sections of this unit to deduce antiderivatives,
and hence indefinite integrals, of further functions.
You saw in the last subsection that it isn’t straightforward to use the fact
that the derivative of ln x is 1/x to deduce an antiderivative of 1/x, because
the domain of the function f (x) = 1/x contains numbers that aren’t in the
domain of the function F (x) = ln x. Luckily there are no similar problems
with any of the other derivatives of standard functions that you’ve met, so
it’s straightforward to use them to deduce antiderivatives.
For example, you can immediately deduce antiderivatives of the standard
functions sin, cos and exp, as follows. Since
the derivative of sin x is cos x,
it follows that
an antiderivative of cos x is sin x.
Similarly, since
the derivative of cos x is − sin x,
it follows that
an antiderivative of − sin x is cos x,
and hence (by the constant multiple rule for antiderivatives)
an antiderivative of sin x is − cos x.
Finally, since the derivative of ex is ex , it follows that an antiderivative of
ex is ex .
The table in the box below contains all the indefinite integrals of standard
functions that you’ve met so far, and all the further indefinite integrals
that can be deduced from the table of standard derivatives in
Subsection 3.4, in the way just described. You should try to memorise at
least the ones above the line in the middle of the table. For the others, you
should try to make sure that you can at least recognise the functions on
the left as functions that can be integrated easily, even if you continue to
refer to the table for their indefinite integrals. This table of standard
indefinite integrals is also included in the Handbook.

81
Unit 7 Differentiation methods and integration

Standard indefinite integrals

Function Indefinite integral
a (constant) ax + c
1
xn (n )= −1) xn+1 + c
n+1
1
ln |x| + c
x
or ln x + c, for x > 0

ex ex + c

sin x − cos x + c
cos x sin x + c

sec2 x tan x + c
cosec2 x − cot x + c
sec x tan x sec x + c
cosec x cot x − cosec x + c
1
√ sin−1 x + c
1 − x2 or − cos−1 x + c
1
tan−1 x + c
1 + x2

In the next activity you’re asked to use these formulas, together with the
constant multiple rule and the sum rule for antiderivatives, to find the
indefinite integrals of some further functions. In some of the parts you
have to start by using an algebraic technique to express the function in a
form that you can integrate.

82
 Learning outcomes

Activity 38 Finding indefinite integrals of more functions

Find the indefinite integrals of the following functions.

7
(a) h(θ) = 3 cos θ + 4 sin θ (b) g(φ) = 6 − 5 cosec2 φ (c) f (x) =
x
1 1 3
(d) g(t) = (e) f (x) = (x > 0) (f) f (x) =
3t πx 1 + x2
1 1 1
(g) h(t) = √ (h) p(x) = 2
(i) g(x) = √
4 1 − t2 5 + 5x 4 − 4x2
x−2 x−2
(j) q(x) = 5(x − 3)(2x − 1) (k) f (x) = (l) g(x) =
x x3
(m) f (x) = 8e x (n) f (x) = e 1+x (o) r(φ) = − cosec φ cot φ
(p) r(θ) = sec θ(sec θ + tan θ)

You’ll learn much more about integration in Unit 8. In particular you’ll

meet a surprising and useful alternative way to understand it. You’ll also
meet some further methods for combining antiderivatives of functions to
obtain antiderivatives of more functions.

Learning outcomes
After studying this unit, you should be able to:
• remember the formulas for the derivatives of some standard functions
• differentiate a wide variety of functions, including products, quotients
and composites
• use differentiation to solve simple optimisation problems
• find and use derivatives using a computer
• understand and use the relationship between the derivative of an
invertible function and the derivative of its inverse function
• understand that integration reverses differentiation
• understand what’s meant by an antiderivative and the indefinite
integral of a function
• remember the formulas for the indefinite integrals of some standard
functions
• use the constant multiple rule and the sum rule for antiderivatives
• use integration to work with the relationship between a changing
quantity and its rate of change
• in particular, use integration to work with the relationships between
displacement, velocity and acceleration.

83
Unit 7 Differentiation methods and integration

Solutions to activities
Solution to Activity 1 Solution to Activity 2
The graph of the derivative of the sine function (a) f (x) = sin x + cos x, so
for x between 0 and 2π must look something like the f " (x) = cos x − sin x.
sketch below.
(b) g(u) = u2 − cos u, so
g " (u) = 2u − (− sin u)
= 2u + sin u.
(c) P = 6 tan θ, so
dP
= 6 sec2 θ.
dθ
(d) r = −2(1 + sin φ), so
dr
(You can work this out as follows. = −2(0 + cos φ)
dφ
When x = π, the gradient of the graph of the sine = −2 cos φ.
function seems to be about the same as the gradient
of the line y = −x; that is, it seems to be about −1. Solution to Activity 3
As x increases, the graph gradually gets less steep – (a) f (x) = ex + ln x, so
that is, its gradient takes negative values of smaller
1
and smaller magnitude, until x = 3π/2, when the f " (x) = ex + .
gradient seems to be zero. The gradient changes x
slowly when x is only a little larger than π, but (b) h(r) = r − cos r − 3 ln r, so
changes more rapidly as x gets closer to 3π/2. 1
h" (r) = 1 − (− sin r) − 3 ×
In other words, for x between π and 3π/2 the value r
3
of the derivative seems to increase from −1 to zero, = 1 + sin r − .
slowly at first, but then more rapidly. r
1
When x increases from 3π/2, the gradient starts at (c) v = + ln t, so
zero, then it becomes positive. At first the graph is t
dv 1 1
not very steep – that is, the gradient has small =− 2 +
positive values – but it becomes steeper and steeper dt t t
1 t
– that is, the gradient takes larger and larger =− 2 + 2
positive values. Eventually, when x = 2π, the t t
t−1
gradient seems to be about the same as the gradient = 2 .
of the line y = x; that is, it seems to be about 1. t
The gradient increases fairly rapidly when x is only (d) w = 5 − 3eu , so
a little larger than 3π/2, but increases more slowly dw
= −3eu .
as x gets closer to 2π. du
So, for x between 3π/2 and 2π, the value of the (e) k = 4(ln v − tan v), so
) -
derivative seems to increase from 0 to 1, fairly dk 1 2
rapidly at first, but then more slowly.) =4 − sec v .
dv v

Solution to Activity 4
(a) The function is k(x) = x3 tan x.
Let f (x) = x3 and g(x) = tan x.
Then f " (x) = 3x2 and g " (x) = sec2 x.

84
 Solutions to activities

By the product rule, (e) m = (u2 + 3) ln u, so, by the product rule,

k " (x) = f (x)g " (x) + g(x)f " (x) dm 1
= (u2 + 3) × + (ln u) × 2u
= x3 sec2 x + (tan x) × 3x2 du u
u2 + 3
= x3 sec2 x + 3x2 tan x = + 2u ln u
u
= x2 (x sec2 x + 3 tan x).
u2 + 3 2u2 ln u
(b) The function is k(x) = xex . = +
u u
Let f (x) = x and g(x) = ex . u2 + 3 + 2u2 ln u
Then f " (x) = 1 and g " (x) = ex . =
u
By the product rule, 2
u (1 + 2 ln u) + 3
k " (x) = f (x)g " (x) + g(x)f " (x) = .
u
= xex + ex × 1 √
(f) y = x sin x = x1/2 sin x, so, by the product
= xex + ex rule,
= (x + 1)ex . dy 1
= x1/2 cos x + (sin x) × x−1/2
dx 2
Solution to Activity 5 √ sin x
= x cos x + √
The function is k(x) = x2 × x3 . 2 x
Let f (x) = x2 and g(x) = x3 . 2x cos x + sin x
Then f " (x) = 2x and g " (x) = 3x2 . = √ .
2 x
By the product rule,
k " (x) = f (x)g " (x) + g(x)f " (x)
Solution to Activity 7
= x2 × 3x2 + x3 × 2x
If y = x2 ex , then, by the product rule,
= 3x4 + 2x4 dy
= 5x4 . = x2 ex + ex × 2x
dx
This is the same answer that you obtain by = x(x + 2)ex .
differentiating k(x) = x5 directly, using the usual Substituting x = −1 gives
rule for the derivative of a power function. dy 1
= (−1)(−1 + 2)e−1 = − .
dx e
Solution to Activity 6 So the gradient is −1/e, which is approximately
(a) k(x) = (3x2 + 2x + 1)ex , so, by the product rule, −0.37.
k " (x) = (3x2 + 2x + 1)ex + ex (6x + 2) (The graph of y = x2 ex is shown below. You can see
= (3x2 + 2x + 1 + 6x + 2)ex that the gradient at the point with x-coordinate −1
2 x does seem to be approximately −0.37.)
= (3x + 8x + 3)e .
(b) k(x) = sin x cos x, so, by the product rule,
k " (x) = (sin x)(− sin x) + (cos x)(cos x)
= cos2 x − sin2 x.
(c) k(z) = z sin z, so, by the product rule,
k " (z) = z cos z + (sin z) × 1
= z cos z + sin z.
(d) v = (2t2 − 1) cos t, so, by the product rule,
dv
= (2t2 − 1)(− sin t) + (cos t) × 4t
dt
= (1 − 2t2 ) sin t + 4t cos t.

85
Unit 7 Differentiation methods and integration

Solution to Activity 8 r+1

(c) z = , so, by the quotient rule,
e x r−1
(a) The function is k(x) = . dz (r − 1) × 1 − (r + 1) × 1
x
x =
Let f (x) = e and g(x) = x. dr (r − 1)2
Then f " (x) = ex and g " (x) = 1. r−1−r−1
So, by the quotient rule, =
(r − 1)2
g(x)f " (x) − f (x)g " (x) 2
k " (x) = = =− .
(g(x))2 (r − 1)2
xex − ex × 1 θ
= (d) m = , so, by the quotient rule,
x2 cos θ
(x − 1)ex dm (cos θ) × 1 − θ(− sin θ)
= . =
x2 dθ (cos θ)2
x3
(b) The function is k(x) = . cos θ + θ sin θ
2x + 1 =
3 cos2 θ
Let f (x) = x and g(x) = 2x + 1.
Then f " (x) = 3x2 and g " (x) = 2. 1 + θ tan θ
So, by the quotient rule, = .
cos θ
g(x)f " (x) − f (x)g " (x) (The final expression here is obtained by
k " (x) = =
(g(x))2 dividing the top and bottom of the previous
(2x + 1) × 3x2 − x3 × 2 expression by cos θ. The previous expression
= would also be an acceptable final answer.)
(2x + 1)2
(3(2x + 1) − 2x)x2 ln t
= (e) p(t) = 2 , so, by the quotient rule,
(2x + 1)2 t
1
(6x + 3 − 2x)x2 t2 × − (ln t) × 2t
= "
p (t) = t
(2x + 1)2 (t2 )2
(4x + 3)x2 t − 2t ln t
= . =
(2x + 1)2 t4
1 − 2 ln t
Solution to Activity 9 = .
t3
ex
(a) f (x) = , so, by the quotient rule, Solution to Activity 10
x3
x3 ex − ex (3x2 ) (a) By the quotient rule,
f " (x) =
(x3 )2 d d , cos x +
(cot x) =
x2 (x − 3)ex dx dx sin x
=
x6 (sin x)(− sin x) − (cos x)(cos x)
(x − 3)ex =
= . (sin x)2
x4
sin2 x + cos2 x
u−1 =−
(b) g(u) = , so, by the quotient rule, sin2 x
eu 1
(eu ) × 1 − (u − 1)eu =− 2
g " (u) = sin x
(eu )2
(1 − u + 1)eu = − cosec2 x.
=
(eu )2
2−u
= .
eu

86
 Solutions to activities

(b) By the quotient rule, Solution to Activity 12

) -
d d 1 (a) y = e5x , so
(sec x) =
dx dx cos x y = eu where u = 5x.
(cos x) × 0 − 1 × (− sin x) This gives
=
(cos x)2 dy du
= eu , = 5.
sin x du dx
= By the chain rule,
cos2 x
) -) - dy dy du
1 sin x =
= dx du dx
cos x cos x = eu × 5
= sec x tan x. = e5x × 5
(c) By the quotient rule, = 5e5x .
) -
d d 1 (b) y = sin(2x2 − 3), so
(cosec x) =
dx dx sin x y = sin u where u = 2x2 − 3.
(sin x) × 0 − 1 × cos x This gives
=
(sin x)2 dy du
= cos u, = 4x.
− cos x du dx
= By the chain rule,
sin2 x
) -, dy dy du
1 cos x + dx
=
du dx
=−
sin x sin x = (cos u) × 4x
= − cosec x cot x. = (cos(2x2 − 3)) × 4x
= 4x cos(2x2 − 3).
Solution to Activity 11 ,x+
(a) y = ln(x4 ) (c) y = sin , so
4
so y = ln u where u = x4 . x
y = sin u where u = .
(b) y = sin(x2 − 1) 4
so y = sin u where u = x2 − 1. This gives
dy du 1
(c) y = cos(ex ) = cos u, = .
du dx 4
so y = cos u where u = ex . By the chain rule,
(d) y = cos3 x dy dy du
so y = u3 where u = cos x. =
dx du dx
(e) y = etan x 1
= (cos u) ×
so y = eu where u = tan x. 4
√ 1 ,x+
(f) y = ln√x = cos .
4 4
so y = u where u = ln x.
√ (d) y = cos(ex ), so
(g) y = (x + x)9 √ y = cos u where u = ex .
so y = u9 where u = x + x.
√ This gives
(h) y = 3 x√2−x−1
dy du
so y = u where u = x2 − x − 1.
3 = − sin u, = ex .
√
du dx
(i) y = e x √
so y = eu where u = x.

87
Unit 7 Differentiation methods and integration

By the chain rule, (Remember that cos5 x means (cos x)5 .)

dy dy du (d) g(x) = ln(x2 + x + 1), so, by the chain rule,
=
dx du dx 1
= (− sin u) × ex g " (x) = 2 × (2x + 1)
x +x+1
= (− sin(ex )) × ex 2x + 1
= 2 .
= −ex sin(ex ) x +x+1
(e) y = ln(sin x), so (e) v(t) = sin(et ), so, by the chain rule,
y = ln u where u = sin x. v " (t) = cos(et ) × et
This gives = et cos(et ).
dy 1 du (f) v = ln(r5 ), so, by the chain rule,
= , = cos x.
du u dx dv 1
By the chain rule, = 5 × 5r4
dr r
dy dy du 5
= = .
dx du dx r
1 (Here’s an alternative way to differentiate this
= × cos x
u function, without using the chain rule:
1
= × cos x v = ln(r5 ) = 5 ln r, so, by the constant multiple
sin x
cos x rule,
= dv 1 5
sin x = 5 × = .)
= cot x. dr r r
2
(g) r = tan θ, so, by the chain rule,
(f) y = sin3 x = (sin x)3 , so
dr
y = u3 where u = sin x. = (2 tan θ) × sec2 θ
dθ
This gives = 2 tan θ sec2 θ.
dy du
= 3u2 , = cos x. (h) R = etan q , so, by the chain rule,
du dx
By the chain rule, dR
= etan q sec2 q.
dy dy du dq
=
dx du dx (i) A = (1 + p2 )9 so, by the chain rule,
= 3u2 cos x dA
= 9(1 + p2 )8 × 2p
= 3(sin x)2 cos x dp
= 3 sin2 x cos x. = 18p(1 + p2 )8 .

Solution to Activity 13 Solution to Activity 14

√
(a) f (x) = cos(4x), so, by the chain rule, (a) f (x) = ln x = (ln x)1/2 , so, by the chain rule,
1
f " (x) = − sin(4x) × 4 f " (x) = 12 (ln x)−1/2 ×
x
= −4 sin(4x). 1
= √ .
(b) f (x) = e2x+1 , so, by the chain rule, 2x ln x
f " (x) = e2x+1 × 2 1
(b) p(w) = = (w3 + 7)−4 , so, by the
= 2e2x+1 . (w + 7)4
3

(c) r(x) = cos5 x, so, by the chain rule, chain rule,

r" (x) = (5 cos4 x)(− sin x) p" (w) = −4(w3 + 7)−5 × 3w2
= −5 sin x cos4 x. 12w2
=− 3 .
(w + 7)5
88
 Solutions to activities

√
(c) f (x) = cos( x) = cos(x1/2 ), so, by the chain (f) w(x) = ln(7x), so
rule, 7 1
w" (x) = = .
f " (x) = − sin(x1/2 ) × 12 x−1/2 7x x
√ (Here’s an alternative way to differentiate this
sin( x)
=− √ . function:
2 x
1 w(x) = ln(7x) = ln 7 + ln x, so
(d) f (x) = √ x = (ex )−1/2 = e−x/2 , so, by the 1 1
e w" (x) = 0 + = .)
chain rule, x x
) -
f " (x) = e−x/2 × (− 12 ) −2t
(g) f (t) = sin , so
3
= − 12 e−x/2 ) -
2 −2t
1 f " (t) = − cos .
= − √ x. 3 3
2 e
(This solution√uses the index laws (am )n = amn (h) k(φ) = tan(3φ), so
and am/n = n am .) k " (φ) = 3 sec2 (3φ).
1
(e) a(t) = = (2 − 3t − 3t2 )−1/3 , Solution to Activity 16
(2 − 3t − 3t2 )1/3
so, by the chain rule, (a) k(x) = cos(7x + 4), so
a" (t) = − 13 (2 − 3t − 3t2 )−4/3 × (−3 − 6t) k " (x) = −7 sin(7x + 4).
1 + 2t (b) h(x) = e−x−3 , so
= .
(2 − 3t − 3t2 )4/3 h" (x) = −e−x−3 .
2 −2 ) - ) -
(f) C(p) = e1/p = e(p )
, so, by the chain rule, θ−1 θ 1
(c) r(θ) = sin = sin − , so
C " (p) = e(p
−2
)
× (−2p−3 ) 2 2 2
) - ) -
2 1 θ 1 1 θ−1
= −2p−3 e1/p r" (θ) = cos − = cos .
2 2 2 2 2 2
2e1/p ) - ) -
=− . 2−θ 2 θ
p3 (d) s(θ) = sin = sin − , so
3 3 3
) - ) -
Solution to Activity 15 " 1 2 θ 1 2−θ
s (θ) = − cos − = − cos .
(a) k(x) = cos(7x), so 3 3 3 3 3
k " (x) = −7 sin(7x).
) - Solution to Activity 17
θ
(b) r(θ) = sin , so (a) g(x) = (x2 + 1)(x3 + 1)
2
) - = x5 + x3 + x2 + 1,
" 1 θ
r (θ) = 2 cos . so
2
g " (x) = 5x4 + 3x2 + 2x
3u/2
(c) g(u) = e , so
= x(5x3 + 3x + 2).
"
g (u) = 23 e3u/2 .
(An alternative method is to use the product
−x
(d) h(x) = e , so rule.)
h" (x) = −e−x .
(e) p(α) = cos(−8α), so
p" (α) = −8(− sin(−8α))
= 8 sin(−8α).

89
Unit 7 Differentiation methods and integration

x cos(4u)
(b) f (x) = , so (e) q(u) = , so
(x − 2)3 e3u
d % $ d d % 3u $
(x − 2)3 × 1 − x (x − 2)3 e3u (cos(4u)) − cos(4u) e
f " (x) = dx q " (u) = dx dx
(x − 2)6 (e3u )2
(by the quotient rule) (by the quotient rule)
(x − 2)3 × 1 − x × 3(x − 2)2 × 1 e3u × (−4 sin(4u)) − cos(4u) × 3e3u
= =
(x − 2)6 (e3u )2
(by the chain rule) (by the chain rule)
(x − 2)3 − 3x(x − 2)2 −4e3u sin(4u) − 3e3u cos(4u)
= =
(x − 2)6 e6u
3u
(x − 2) − 3x −e (4 sin(4u) + 3 cos(4u))
= =
(x − 2)4 e6u
−2x − 2 4 sin(4u) + 3 cos(4u)
= =−
(x − 2)4 e3u
2(x + 1) (f) f (x) = esin(5x) , so
=− .
(x − 2)4 f " (x) = esin(5x) × 5 cos(5x)
(c) g(x) = cos(x ln x), so (by the chain rule)
d = 5esin(5x) cos(5x).
g " (x) = − sin(x ln x) (x ln x)
dx
(by the chain rule) (g) h(x) = cos3 (x2 ), so
) - d % $
1 h" (x) = 3 cos2 (x2 ) cos(x2 )
= − sin(x ln x) x × + (ln x) × 1 dx
x (by the chain rule)
(by the product rule)
= 3 cos2 (x2 ) × (− sin(x2 )) × 2x
= − sin(x ln x)(1 + ln x). (by the chain rule again)
(d) h(x) = ex/2 sin(3x), so
= −6x cos2 (x2 ) sin(x2 ).
d d , x/2 +
h" (x) = ex/2 (sin(3x)) + sin(3x) e e−2x
dx dx (h) z(x) = = 51 e−2x , so
(by the product rule) 5
z " (x) = 15 × (−2e−2x )
= ex/2 × 3 cos(3x) + sin(3x) × 12 ex/2 2
(by the chain rule) = − e−2x .
5
= 3ex/2 cos(3x) + 12 ex/2 sin(3x) (i) G(x) = 3 sin x cos x + 2 sin(3x), so
= 12 ex/2 (6 cos(3x) + sin(3x)). G" (x) = 3((sin x)(− sin x) + (cos x)(cos x))
+ 2(cos(3x)) × 3
(by the product rule)
= −3 sin2 x + 3 cos2 x + 6 cos(3x).
(You can simplify this answer further by using
the trigonometric identity
cos(2θ) = cos2 θ − sin2 θ:
− 3 sin2 x + 3 cos2 x + 6 cos(3x)
= 3(− sin2 x + cos2 x) + 6 cos(3x)
= 3 cos(2x) + 6 cos(3x).)
90
 Solutions to activities

(j) p(t) = tet sin t = t(et sin t), so Solution to Activity 18

d t (a) g(x) = e−kx , where k is a constant, so
p" (t) = t (e sin t) + (et sin t) × 1
dt g " (x) = −ke−kx .
(by the product rule)
= t(et cos t + (sin t)et ) + et sin t (b) f (x) = x cos(πx), so, by the product rule,
(by the product rule again) f " (x) = x × (−π sin(πx)) + (cos(πx)) × 1
= −πx sin(πx) + cos(πx)
= tet cos t + tet sin t + et sin t
= cos(πx) − πx sin(πx).
= et (t cos t + t sin t + sin t).
(c) h = sin2 (aθ), where a is a constant, so, by the
(Alternatively, you could start by writing
chain rule,
p(t) = (tet ) sin t, and again apply the product
rule twice.) dh
= 2 sin(aθ) × a cos(aθ)
dθ
(k) r = sin(2θ) cos θ, so = 2a sin(aθ) cos(aθ).
dr
= sin(2θ)(− sin θ) + (cos θ) × 2 cos(2θ) (You can use the trigonometric identity
dθ sin(2θ) = 2 sin θ cos θ to write this answer above
(by the product rule) in a neater form, namely a sin(2aθ).)
= 2 cos θ cos(2θ) − sin θ sin(2θ). (d) y = cos(3x) + c, where c is a constant, so
3z
(l) h = ze , so dy
= (− sin(3x)) × 3 + 0
dh dx
= z × 3e3z + e3z × 1
dz = −3 sin(3x).
(by the product rule)
Solution to Activity 19
= 3ze3z + e3z ln t
= (3z + 1)e3z . The function is p(t) = 2 = t−2 ln t.
t
√ So, by the product rule,
sin u sin(u1/2 )
(m) v = √ = , so 1
u u1/2 p" (t) = t−2 × + (ln t)(−2t−3 )
% $ t
1/2 d 1/2 1/2 1 −1/2
1 2 ln t
dv u sin(u ) − sin(u ) × 2 u
= du = 3
−
du u t t3
(by the quotient rule) 1 − 2 ln t
= .
t3
1/2 1/2 1 −1/2 1/2 1 −1/2
u cos(u ) × 2 u − sin(u ) × 2 u
= Solution to Activity 20
u
(by the chain rule) Let the length of the sides of the squares be x
(in m), as shown below. The value of x must be
cos(u1/2 ) − u−1/2 sin(u1/2 ) between 0 and 21 .
=
2u
u1/2 cos(u1/2 ) − sin(u1/2 )
= 3/2
√ √ 2u √
u cos u − sin u
= √ .
2u u

91
Unit 7 Differentiation methods and integration

Let the volume of the box be V (in m3 ). The endpoints of the interval are 0 and 21 , so the
maximum value of V occurs when x = 0, x = 16 or
x = 12 .
When x = 0 the height of the box is zero, and when
x = 12 its length and width are zero, so in both these
cases the volume V will be zero. When x = 16 , the
length, width and height of the box are all positive,
so the volume V will also be positive.
So the maximum value of V is achieved when x = 16 .
Hence the sides of the squares should be 16 m; that
is, about 16.7 cm.
(A graph of V against x, for x between 0 and 12 , is
shown below. You can see that the maximum value
of V does indeed seem to occur when x is
approximately 61 = 0.166 . . ..)

Then the height of the box will be x, and the length

and width of the box will each be 1 − 2x. Hence
V = x(1 − 2x)2 .
We have to find the value of x, between 0 and 12 ,
that gives the maximum value of V .
V is a polynomial function of x, so it is continuous
on its whole domain and differentiable at every (Note also that an alternative way to differentiate
value of x. So the maximum value of V must occur the formula for V , rather than first multiplying out
either at an endpoint of the interval [0, 12 ], or at a the brackets, is to use a combination of the product
stationary point. rule and the chain rule, as follows:
The formula for V is dV d
= (x(1 − 2x)2 )
dx dx
V = x(1 − 2x)2 d % $
= x(1 − 4x + 4x2 ) =x (1 − 2x)2 + (1 − 2x)2 × 1
dx
= x − 4x2 + 4x3 , = x × 2(1 − 2x) × (−2) + (1 − 2x)2
so = −4x(1 − 2x) + (1 − 2x)2
dV
= 1 − 8x + 12x2 . = (1 − 2x)(−4x + 1 − 2x)
dx
At a stationary point, dV /dx = 0, which gives = (1 − 2x)(1 − 6x).)
1 − 8x + 12x2 = 0 Solution to Activity 21
12x2 − 8x + 1 = 0 The volume of the channel is the area of its
(6x − 1)(2x − 1) = 0 cross-section times its (fixed) length. So to achieve a
x = 61 or x = 21 . channel of maximum volume, we have to achieve the
1 maximum area of the cross-section.
So the stationary points occur when x = 6 and
x = 12 .

92
 Solutions to activities

Let the angle between the two sides of the v-shape (A graph of A against θ, for θ between 0 and π, is
be θ, as shown below. The value of θ must be shown below. You can see that the maximum value
between 0 and π. of A does indeed seem to occur when θ is
Let the area of the cross-section of the channel be A approximately π/2.)
(in m2 ).

Each side of the v-shape has length 12 m, so, by the

formula A = 21 ab sin θ for the area of a triangle,
A = 12 × 12 × 12 sin θ = 18 sin θ. (Notice that in fact you could have answered this
We have to find the value of θ, between 0 and π, question without using calculus, because, from what
that gives the maximum value of A. you saw in Unit 4, the maximum value of sin θ for
values of θ between 0 and π occurs when θ = π/2. It
Since the sine function is continuous on its whole
follows that the maximum value of A = 18 sin θ for
domain and differentiable everywhere, so is the
values of θ between 0 and π also occurs when
variable A as a function of θ. So the maximum value
θ = π/2.)
of A must occur either at an endpoint of the interval
[0, π], or at a stationary point. Solution to Activity 23
The formula stated above for A gives
(a) (You need to do this on a computer.)
dA
= 18 cos θ. (b)
dθ
At a stationary point, dA/dθ = 0, which gives
1
8 cos θ = 0
cos θ = 0
π
θ=
2
(since θ is in the interval [0, π]).
So the stationary point occurs when θ = π/2.
The endpoints of the interval are 0 and π, so the
maximum value of A occurs when θ = 0, θ = π/2 or
θ = π.
When θ = 0 the width of the channel is zero, and
when θ = π its height is zero, so in both these cases
the area A will be zero. When θ = π/2, the width (The symbol consisting of two short slant lines on
and height of the channel are both positive, so the the y-axis of this graph indicates that part of the
area A will also be positive. axis has been omitted.)
So the maximum value of A is achieved when (c) The value of x that gives the minimum value of
θ = π/2. f (x) seems to be about 0.6.
That is, to give a channel with maximum volume, (d) The derivative of f is
the angle between the sides of the v-shape should be x 1
π/2, which is 90◦. f " (x) = √ − .
8 1+x 2 16

93
Unit 7 Differentiation methods and integration

(e) The solution of the equation f " (x) = 0 is At a stationary point, f " (x) = 0, which gives
x = 0.577 (to 3 s.f.). x 1
√ − =0
(f) The man should join the road 0.58 kilometres 8 1+x 2 16
(to the nearest 10 metres) along from the point x 1
√ =
that is closest to his initial position. 8 1+x 2 16
(
(Because the graph of f indicates that the minimum 16x = 8 1 + x2
value of f (x) occurs at a local minimum rather than (
at one of the endpoints of the interval [0, 2], in the 2x = 1 + x2
solution above we don’t consider the values of f (0) (2x) = 1 + x2
2

and f (2).) 4x2 = 1 + x2

(Details of how to use the CAS for this question are 3x2 = 1
given in the Computer algebra guide, in the section 1
x2 =
‘Computer methods for CAS activities in 3
Books A–D’.) 1
x= √ .
By-hand solution 3
The problem in this question can be solved by hand (Here we have used the fact that x ≥ 0.)
√
as follows. So the only stationary point of f is x = 1/ 3. The
The function value of f at this stationary point is
√ to be minimised has the rule # , +2
1+x 2 2−x , +
f (x) = + ) - 1 + √13 2 − √1
8 16 1 3
1 2 1/2 1
f √ = +
= 8 (1 + x ) + 16 (2 − x). 3 8 16
!
The variable x can take values between 0 and 2. √
1 + 13 2 3−1
The function f is continuous on its domain [0, 2], = + √
8 16 3
and differentiable at every value of x in this interval, !
except at the endpoints 0 and 2. (Remember that a 4 √
3 2 3−1
function can’t be differentiable at an endpoint of its = + √
8 16 3
domain.) (These facts can be deduced from the fact √
that f is a simple combination of functions that 2 2 3−1
= √ + √
have these properties. You can learn more about 8 3 16 3
√ √ √ √
making such deductions in later modules. For now, 2 3 2 3 3− 3
you can see from the graph of f that it appears to = √ √ + √ √
8 3 3 16 3 3
have these properties.) √ √
2 3 6− 3
Hence the minimum value of f on the interval [0, 2] = +
24
√ 48√
occurs at a stationary point of f or at an endpoint
of the interval. 4 3 6 − 3
= +
48 √ 48
Differentiating f gives
6+3 3
f " (x) = 81 × 12 (1 + x2 )−1/2 × 2x + 16
1
× (−1) =
48√
x 1 2+ 3
= √ − . = ≈ 0.23.
8 1 + x2 16 16
The values of f at the endpoints of the interval [0, 2]
are as follows:
√
1 + 02 2 − 0
f (0) = +
8 16
1 1 1
= + = = 0.25
8 8 4

94
 Solutions to activities

√
1 + 22 2 − 2 Solution to Activity 25
f (2) = +
√ 8 √ 16 (a) f (x) = sin−1 (3x), so, by the chain rule,
5 5 1
= +0= ≈ 0.28. f " (x) = ( ×3
8 8 1 − (3x)2
So the √
minimum value of f (x) is achieved when
3
x = 1/ 3 = 0.577 (to 3 s.f.). Hence, as was found =√ .
above using a computer, the man should join the 1 − 9x2
road 0.58 kilometres (to the nearest 10 metres) along x −1
from the point that is closest to his initial position. (b) f (x) = e cos) x, so, by the-
product rule,
1
f " (x) = ex − √ + (cos−1 x)ex
Solution to Activity 24 1 − x2
) -
(a) We have y = cos−1 x, so 1
= cos−1 x − √ ex .
x = cos y. 1−x 2

Therefore Solution to Activity 26

dx
= − sin y. (a) F (x) = x2 , so F " (x) = 2x.
dy
By the inverse function rule, (b) F (x) = x2 + 3, so F " (x) = 2x.
dy 1 (c) F (x) = x2 − 75 , so F " (x) = 2x.
=− ,
dx sin y
provided that sin y =
) 0. Solution to Activity 27
2 2
The identity sin y + cos y = 1 gives (a) Since
( (
sin y = ± 1 − cos2 y = ± 1 − x2 . d %1 $
2 sin(2x) = 12 × 2 cos(2x) = cos(2x),
The + sign applies here, because y takes values dx
only in the interval [0, π] (since y = cos−1 (x)) it follows that F (x) = 21 sin(2x) is an
and so sin y is always non-negative. Hence antiderivative of the function f (x) = cos(2x).
(
sin y = 1 − x2 . (b) The indefinite integral of f (x) = cos(2x) is
F (x) = 21 sin(2x) + c.
Therefore
dy 1 (c) An antiderivative of f (x) = cos(2x), other than
= −√ , the antiderivative in part (a), is
dx 1 − x2
provided that x )= ±1. F (x) = 21 sin(2x) + 1.
(b) We have y = tan−1 x, so (You probably chose a different antiderivative.
Any function obtained by setting the constant c
x = tan y.
in the indefinite integral F (x) = 21 sin(2x) + c
Therefore equal to a particular number will do.)
dx
= sec2 y.
dy Solution to Activity 28
By the inverse function rule, (a) The indefinite integral of f (x) = x9 is
dy 1 1 10
= . F (x) = 10 x + c.
dx sec2 y
The identity tan2 y + 1 = sec2 y gives 1
(b) The indefinite integral of p(u) = = u−3 is
u3
sec2 y = 1 + tan2 y = 1 + x2 . 1 −2 1
Therefore P (u) = u + c = − 2 + c.
−2 2u
dy 1
= . (c) The indefinite integral of f (x) = x2/3 is
dx 1 + x2
F (x) = 53 x5/3 + c.

95
Unit 7 Differentiation methods and integration

√
4 x = x1/4 is
(d) The indefinite integral of f (x) = (f) We have
F (x) = 4 5/4
5x + c. x2 + 1 1 1
g(x) = = 2 + 4 = x−2 + x−4 ,
x 4 x x
(e) The indefinite integral of g(t) = t−2/3 is
which has indefinite integral
G(t) = 3t1/3 + c. G(x) = −x−1 − 31 x−3 + c
1 1 1
(f) The indefinite integral of b(v) = √ = v −1/2 is = − − 3 + c.
v x 3x
1/2
√
B(v) = 2v + c = 2 v + c. (g) We have
) -
√ 1
Solution to Activity 29 f (r) = 3 r 10r2 − 2
r
(a) The indefinite integral of f (x) = x5 is
= r1/3 (10r2 − r−2 )
F (x) = 16 x6 + c.
= 10r7/3 − r−5/3
(b) Three different antiderivatives of f are which has indefinite integral
F (x) = 61 x6 , 3 10/3
F (r) = 10( 10 r ) − (− 32 r−2/3 ) + c
F (x) = 16 x6 + 1,
= 3r10/3 + 23 r−2/3 + c
1 6
F (x) = 6x
− 2.
= 23 r−2/3 (1 + 2r4 ) + c.
(You probably chose different examples.)
Solution to Activity 31
Solution to Activity 30
(a) The function is f (x) = x + 2. Its indefinite
(a) The indefinite integral of f (x) = 8x3 is integral is
F (x) = 8 × 14 x4 + c F (x) = 12 x2 + 2x + c.
= 2x4 + c. Using the fact that F (1) = 9
gives
2
(b) The indefinite integral of f (x) = 7 is 1 2
×1 +2×1+c= 9
2 2
F (x) = 7x + c. 5
+c= 9
2 2
(c) The indefinite integral of c = 24 = 2.
√
f (x) = x − x2 = x1/2 − x2 is
Hence the required antiderivative is
F (x) = 23 x3/2 − 13 x3 + c F (x) = 21 x2 + 2x + 2.
= 13 (2x3/2 − x3 ) + c. (b) The function is f (x) = 1/x3 = x−3 with domain
(d) We have (0, ∞). Its indefinite integral is
f (x) = (x + 2)(x − 5) 1 −2 1
F (x) = x + c = − 2 + c.
= x2 − 3x − 10, −2 2x
which has indefinite integral Using the fact that F (3) = − 91 gives
1 1
F (x) = 13 x3 − 32 x2 − 10x + c. − +c=−
2×3 2 9
(e) We have 1 1
− +c=−
g(t) = (t + 1)2 18 9
= t2 + 2t + 1, 1 1 1
c=− + =− .
which has indefinite integral 9 18 18
Hence the required antiderivative is
G(t) = 31 t3 + t2 + t + c. 1 1
F (x) = − 2 −
2x 18
9 + x2
=− (x ∈ (0, ∞)).
18x2

96
 Solutions to activities

Solution to Activity 32 that is,

(a) The given equation for v in terms of t is s = 12t − 4.9t2 + b,
v = 10 1
t. where again b is a constant.
Hence the equation for s in terms of t is At the start of the motion, the displacement of
s = 10 1
× 21 t2 + c, the ball is 0 m. That is, when t = 0, s = 0.
that is, Substituting these values into the equation for s
s = 20 1 2
t + c, above gives
where c is a constant. 0 = 12 × 0 − 4.9 × 02 + b, that is, b = 0.
So the equation for s in terms of t is
The marble starts rolling from a position
0.3 metres down the slope, so when t = 0, s = 12t − 4.9t2 .
s = 0.3. (c) When t = 1,
Substituting these values of s and t into the v = 12 − 9.8 × 1 = 12 − 9.8 = 2.2,
equation above gives and
1
0.3 = 20 × 02 + c s = 12 × 1 − 4.9 × 12 = 12 − 4.9 = 7.1.
c = 0.3. So the velocity and the displacement of the ball
So the equation for s in terms of t is 1 second after it was thrown are 2.2 m s−1 and
1 2
s = 20 t + 0.3. 7.1 m, respectively.
(b) Substituting t = 4 into the equation from (d) When the ball has fallen back to the point from
part (a) gives which it was thrown, s = 0. Substituting this
value of s into the equation found in part (b)
1
s = 20 × 42 + 0.3 = 0.8 + 0.3 = 1.1.
gives
That is, the displacement of the marble from
0 = 12t − 4.9t2
the top of the slope after it has been rolling for
4 seconds is 1.1 metres. 12t − 4.9t2 = 0
t(12 − 4.9t) = 0
Solution to Activity 33 t = 0 or t = 12/4.9 = 2.4 (to 1 d.p.).
(a) The equation for a in terms of t is So the ball takes about 2.4 seconds to fall back
a = −9.8. to the point from which it was thrown.
Integrating this equation gives the following (The displacement–time graph for the ball is
equation for v in terms of t: shown below. You can see that the ball first
v = −9.8t + c, rises and then falls, as expected. You can also
where c is a constant. see that it reaches a height of about 7 m after
1 second, and that it returns to the point from
At the start of the motion, the velocity of the which it was thrown after about 2.4 seconds, as
ball is 12 m s−1 . That is, when t = 0, v = 12. you would expect from the answers found in
Substituting these values into the equation for v parts (c) and (d).)
above gives
12 = −9.8 × 0 + c, that is, c = 12.
So the equation for v in terms of t is
v = 12 − 9.8t.
(b) Integrating the equation found in part (a) gives
the following equation for s in terms of t:
s = 12t − 9.8 × 12 t2 + b,

97
Unit 7 Differentiation methods and integration

It was calculated in part (b) that the total

weekly cost of producing 160 tonnes of fertiliser
each week is £45 600. Hence the cost per tonne
of producing 160 tonnes of fertiliser each week is
£45 600
= £285.
160
(The graph of t against q is shown below. You can
see that the total weekly cost of producing
160 tonnes of fertiliser each week is about £45 000,
as expected.)

Solution to Activity 34
(a) The given equation for m in terms of q is
m = 250 − 0.5q.
Since marginal cost is the derivative of total
cost, the equation for t in terms of q is
t = 250q − 0.5 × 21 q 2 + c,
that is,
t = 250q − 0.25q 2 + c,
where c is a constant.
According to the information given in the
question, when q = 80, t = 30 400.
Solution to Activity 35
Substituting these values into the equation
An antiderivative of f is
found for t above gives
F (x) = 8ex/8 .
30 400 = 250 × 80 − 0.25 × 802 + c
So the change in the quantity from x = 2 to x = 5 is
30 400 = 18 400 + c
F (5) − F (2) = 8e5/8 − 8e2/8
c = 12 000.
= 4.67 (to 3 s.f.).
Hence the equation for t in terms of q is
t = 250q − 0.25q 2 + 12 000. Solution to Activity 36
(b) The total weekly cost of producing 160 tonnes An antiderivative of f is
of fertiliser each week is F (t) = 3t − 23 × 12 t2 = 3t − 34 t2 .
£(250 × 160 − 0.25 × 1602 + 12 000) (a) The change in the displacement (in metres) of
= £45 600. the object from t = 0 to t = 1 is
(c) The question states that the total weekly cost of F (1) − F (0) = (3 × 1 − 43 × 12 )
producing 80 tonnes of fertiliser each week is − (3 × 0 − 34 × 02 )
£30 400. Hence the cost per tonne of producing
= 2.25.
80 tonnes of fertiliser each week is
£30400 That is, the object travels 2.25 m in that time
= £380. (in the positive direction).
80

98
 Solutions to activities

(b) The change in the displacement (in metres) of Hence its indefinite integral is
the object from t = 4 to t = 5 is 1
G(t) = ln |t| + c.
F (5) − F (4) = (3 × 5 − 34 × 52 ) 3
− (3 × 4 − 3
× 42 ) (e) The function is
4
1
= (15 − 75
4 ) − (12 − 12) f (x) = (x > 0),
πx
= −3.75. which can be written as
That is, the object travels 3.75 m in that time 1 1
f (x) = × (x > 0).
(in the negative direction). π x
Hence its indefinite integral is
Solution to Activity 37 1
F (x) = ln x + c.
4 π
(a) The function f (x) = has indefinite integral
x 3
F (x) = 4 ln |x| + c. (f) The indefinite integral of f (x) = is
1 + x2
(b) The function F (x) = 3 tan−1 x + c.
x+1 1 1 1 1
f (x) = = + 2 = + x−2 (g) The indefinite integral of h(t) = √ is
x2 x x x 4 1 − t2
has indefinite integral
H(t) = 14 sin−1 t + c.
1
F (x) = ln |x| − x−1 + c = ln |x| − + c. (Or, alternatively, H(t) = − 41 cos−1 t + c.)
x
(c) The function (h) The function is
1 1
f (x) = (x > 0) p(x) = ,
2x 5 + 5x2
has indefinite integral which can be written as
1 1
F (x) = 21 ln x + c. p(x) = × .
5 1 + x2
(The answer F (x) = 12 ln |x| + c is also correct, Hence its indefinite integral is
but can be simplified, as x takes only positive
values.) P (x) = 51 tan−1 x + c.
(i) The function is
Solution to Activity 38 1
g(x) = √ ,
(a) The indefinite integral of h(θ) = 3 cos θ + 4 sin θ 4 − 4x2
is which can be written as
H(θ) = 3 sin θ + 4(− cos θ) + c 1
g(x) = √ √ ;
= 3 sin θ − 4 cos θ + c. 4 1 − x2
that is,
(b) The indefinite integral of g(φ) = 6 − 5 cosec2 φ is
1
G(φ) = 6φ − 5(− cot φ) + c g(x) = √ .
2 1 − x2
= 6φ + 5 cot φ + c. Hence its indefinite integral is
7 G(x) = 21 sin−1 x + c.
(c) The indefinite integral of f (x) = is
x
(Or, alternatively, G(x) = − 21 cos−1 x + c.)
F (x) = 7 ln |x| + c.
(d) The function is
1
g(t) = ,
3t
which can be written as
1 1
g(t) = × .
3 t
99
Unit 7 Differentiation methods and integration

(j) The function is (n) The function is

q(x) = 5(x − 3)(2x − 1), f (x) = e1+x ,
which can be written as which can be written as
q(x) = 5(2x2 − 7x + 3). f (x) = e × ex .
Hence its indefinite integral is Hence its indefinite integral is
Q(x) = 5(2 × 13 x3 − 7 × 12 x2 + 3x) + c F (x) = e × ex + c
= 10 3 35 2
3 x − 2 x + 15x + c.
= e1+x + c.
(It could also be written as (You can integrate the expression e × ex by
Q(x) = 56 x(4x2 − 21x + 18) + c, using the constant multiple rule for
antiderivatives, since e is a constant.)
for example.)
(o) The indefinite integral of r(φ) = − cosec φ cot φ
(k) The function is
is
x−2
f (x) = , R(φ) = −(− cosec φ) + c
x
which can be written as = cosec φ + c.
2 (p) The function is
f (x) = 1 − .
x r(θ) = sec θ(sec θ + tan θ)
Hence its indefinite integral is
which can be written as
F (x) = x − 2 ln |x| + c.
r(θ) = sec2 θ + sec θ tan θ.
(l) The function is
Hence its indefinite integral is
x−2
g(x) = , R(θ) = tan θ + sec θ + c.
x3
which can be written as
g(x) = x−2 − 2x−3 .
Hence its indefinite integral is
1 −2
G(x) = −x−1 − 2 × x +c
−2
1 1
=− + 2 +c
x x
x 1
=− 2 + 2 +c
x x
1−x
= + c.
x2
(m) The indefinite integral of f (x) = 8ex is
F (x) = 8ex + c.

100
 Acknowledgements

Acknowledgements
Grateful acknowledgement is made to the following sources:

Page 30: Mark Hobbs, for the idea for the cartoon
c gallica.bnf.fr / Bibliothèque nationale de France
Page 55: '

Every effort has been made to contact copyright holders. If any have been
inadvertently overlooked the publishers will be pleased to make the
necessary arrangements at the first opportunity.

101
Unit 8

Integration methods
 Introduction

Introduction
In this unit you’ll continue your study of integration. In Unit 7 you saw
that integration is the reverse of differentiation. You saw that you can use
it to solve problems in which you know the values taken by the rate of
change of a continuously changing quantity, and you want to work out the
values taken by the quantity, or how much the quantity changes over some
period. For example, you saw that if you have a formula for the velocity of
an object in terms of time, then you can use integration to find a formula
for the displacement of the object in terms of time, or work out the change
in the displacement of the object over some period of time.
In Sections 1 and 2 of this unit, you’ll meet a different way to think about
integration. This way of thinking about it involves areas of regions of the
plane, rather than rates of change. You’ll see that the link between the two
ways of thinking about integration helps you to solve problems both about
areas and about rates of change.
In Sections 3 to 5 of the unit, you’ll learn some more methods for
integrating functions, which you can use in addition to the methods that
you learned in Unit 7. These further methods will allow you to integrate a
much wider variety of functions than you’ve learned to integrate so far.
Many students find some of the material in this unit quite challenging
when they first meet it, so don’t worry if that’s the case for you too. As
with most mathematics, it will seem more straightforward once you’re
more familiar with it. If you find a particular technique difficult, then
make sure that you watch the associated tutorial clips, and use the
practice quiz and exercise booklet for this unit for more practice.
This unit is likely to take you more time to study than most of the other
units, so the study planner allows extra time for it.

105
Unit 8 Integration methods

1 Areas, signed areas and definite

integrals
In this section you’ll look at a type of mathematical problem that you
might think has little connection with the mathematics of rates of change
that you studied in Units 6 and 7. In Section 2 you’ll see that in fact the
two areas of mathematics are closely related.
Throughout Sections 1 and 2, we’ll work with continuous functions.
Remember that a continuous function is a function whose graph has no
breaks in it, over its whole domain. Informally, it’s a function whose graph
you can draw without taking your pen off the paper (though you might
need an infinitely large piece of paper, as the graph might be infinitely
long!).

1.1 Areas and signed areas

Sometimes it’s useful to calculate the area of a flat shape with a curved
boundary. For example, suppose that an architect is designing a large
building shaped as shown in Figure 1(a). The curve of the roof is given by
the graph of the function
1 2
f (x) = 9 − 36 x (−18 ≤ x ≤ 18),
where x and f (x) are measured in metres. This graph is shown in
Figure 1(b).

Figure 1 (a) The shape of a building (b) the graph that gives the curve
of the roof

106
 1 Areas, signed areas and definite integrals

Suppose that the architect wants to calculate the area shaded in

Figure 1(b), so that she can determine the amounts of materials needed to
build the end wall of the building.
Here’s a way of calculating an approximate value for this area, which you
can make as accurate as you wish. You start by dividing the interval
[−18, 18], which is the interval of x-values that corresponds to the curve,
into a number of subintervals of equal width.
For example, in Figure 2 the interval has been divided into twelve
subintervals. The right endpoint of the first subinterval is equal to the left
endpoint of the next subinterval, and so on.
For each subinterval, you approximate the shape of the curve on that
subinterval by a horizontal line segment whose height is the value of the
function at the left endpoint of the subinterval, as shown in Figure 2. You
calculate the areas of the rectangles between these line segments and the
x-axis (simply by multiplying their heights by their widths in the usual
way), and add up all these areas to give you an approximate value for the
required area. The more subintervals you use, the better will be the
approximation.

Figure 2 A collection of rectangles whose total area is approximately the

area in Figure 1(b)

In the next example this method is used, with four subintervals, to find an
approximate value for the area discussed above. Notice that, in both the
example and in Figure 2 above, the rectangle corresponding to the first
subinterval has zero height and therefore zero area – this happens because
the value of the function at the left endpoint of this subinterval is zero.

107
Unit 8 Integration methods

Example 1 Calculating an approximate value for an area

Use the method described above, with four subintervals as shown
below, to find an approximate value for the area between the graph of
1 2
the function f (x) = 9 − 36 x and the x-axis.

Solution
We divide the interval [−18, 18] into four subintervals of equal width.
The whole interval has width 36, so each subinterval has width
36/4 = 9.
The left endpoints of the four subintervals are
−18, −18 + 9, −18 + 2 × 9, −18 + 3 × 9,
that is
−18, −9, 0, 9.
So the heights of the rectangles are
f (−18), f (−9), f (0), f (9),
that is,
9− 1
36 × (−18)2 , 9− 1
36 × (−9)2 , 9− 1
36 × 02 , 9− 1
36 × 92 ,
which evaluate to
27 27
0, 4 , 9, 4 .
So we obtain the following approximate value for the area:
405
(0 × 9) + ( 27 27
4 × 9) + (9 × 9) + ( 4 × 9) = 2 = 202.5.

The calculation in Example 1 shows that a (rather crude) approximate

value for the area of the cross-section of the roof discussed at the
beginning of this subsection is 202.5 m2 .

108
 1 Areas, signed areas and definite integrals

Note that the units for the area of a region on a graph are, as you’d
expect, the units on the vertical axis times the units on the horizontal axis.
For example, if the units on both axes are metres, then the units for area
are square metres (m2 ). If a graph has no specific units on the axes, then
the units for area are simply ‘square units’, and we usually omit them
when we state an area.
In the next activity you’re asked to find another approximation for the
area of the cross-section of the roof discussed earlier, by using the
subinterval method with six subintervals instead of four.

Activity 1 Calculating an approximate value for an area

Use the method described above, with six subintervals as shown below, to
find an approximate value for the area between the graph of the function
1 2
f (x) = 9 − 36 x and the x-axis.

Table 1 shows the approximate values for the area between the graph of
1 2
the function f (x) = 9 − 36 x and the x-axis that were found in Example 1
and in the solution to Activity 1. It also shows, to three decimal places,
some further approximate values that were found in the same way, but
using larger numbers of subintervals. The calculations were carried out
using a computer.

Table 1 Approximate values for the area in Figure 1(b)

Number of subintervals 4 6 12 50 100 500
Approximation obtained 202.5 210 214.5 215.914 215.978 215.999

You can see that as the number of subintervals gets larger and larger, the
approximation obtained seems to be getting closer and closer to a
particular number. We say that this number is the limit of the
approximations as the number of subintervals tends to infinity. It’s the
1 2
exact value of the area between the graph of the function f (x) = 9 − 36 x
and the x-axis. From Table 1, it looks as if the exact value of this area is
216, or perhaps a number very close to 216.

109
Unit 8 Integration methods

In general, if you have a continuous function f whose graph lies on or

above the x-axis throughout an interval [a, b], then you can use the method
demonstrated above to find, as accurately as you want, the area between
the graph of f and the x-axis, from x = a to x = b. Figure 3(a) illustrates
an area of this type, and Figure 3(b) illustrates the approximation to the
area that’s obtained by using 8 subintervals.

Figure 3 (a) The area between a graph and the x-axis, from x = a to
x = b (b) A collection of rectangles whose total area is approximately
this area
Here’s a summary of the method that you’ve met for finding an
approximate value for an area of this type. You start by dividing the
interval [a, b] into a number, say n, of subintervals of equal width. The
width of each subinterval is then (b − a)/n. For each subinterval you
calculate the product
, -
left endpoint b−a
f × , (1)
of subinterval n
and you add up all these products. In general, the larger the number n of
subintervals, the closer your answer will be to the area between the graph
of f and the x-axis, from x = a to x = b.
Now suppose that you have a continuous function f whose graph lies on or
below the x-axis throughout an interval [a, b], and you want to calculate
the area between the graph of f and the x-axis, from x = a to x = b, as
illustrated in Figure 4(a).
You can use the method above, with a small adjustment at the end, to
calculate an approximate value for an area like this. Consider what
happens when you apply the method to a graph like the one in
Figure 4(a). You start by dividing the interval [a, b] into n subintervals of
equal width, then you calculate all the products of form (1) and add them
all up. For a graph like this, because the value of f at the left endpoint of
each subinterval is negative (or possibly zero), each product of form (1)
will also be negative (or zero). In fact, each product of form (1) will be the
negative of the area between the line segment that approximates the curve
and the x-axis, as illustrated in Figure 4(b). So when you add up all the
products of form (1), you’ll obtain an approximate value for the negative of
the area between the curve and the x-axis, from x = a to x = b.

110
 1 Areas, signed areas and definite integrals

Figure 4 (a) The area between another graph and the x-axis, from x = a
to x = b (b) A collection of rectangles whose total area is approximately
this area
That’s not a problem, because you can simply remove the minus sign to
obtain the approximate value for the area that you want. However, to help
us deal with situations like this, it’s useful to make the following
definitions. These definitions will be important throughout the unit.
Consider any region on a graph that lies either entirely above or entirely
below the x-axis. The signed area of the region is its area with a plus or
minus sign according to whether it lies above or below the x-axis,
respectively. For example, in Figure 5 the two shaded regions above the
x-axis have signed areas +4 and +6, respectively, which you can write
simply as 4 and 6, and the shaded region below the x-axis has signed
area −3.

Figure 5 Regions on a graph

If you have a collection of regions on a graph, where each region in the
collection lies either entirely above or entirely below the x-axis, then the
total signed area of the collection is the sum of the signed areas of the
individual regions. For example, the total signed area of the collection of
three regions in Figure 5 is 4 + (−3) + 6 = 7.
The units for signed area on a graph are, as you’d expect, the same as the
units for area on the graph. That is, they’re the units on the vertical axis
times the units on the horizontal axis. If there are no specific units on the
axes, then we don’t use any specific units for signed area.

111
Unit 8 Integration methods

Example 2 Understanding signed areas

The areas of some regions on a graph are marked below.
In each of parts (a)–(c), use these areas to find the signed area
between the graph and the x-axis, from the first value of x to the
second value of x.
(a) From x = −3 to x = −2. (b) From x = −2 to x = 1.
(c) From x = −3 to x = 1.

Positive signed area

Negative signed area Solution

(a) The signed area from x = −3 to x = −2 is −2.2.
(b) The signed area from x = −2 to x = 1 is +4.8 = 4.8.
(c) The signed area from x = −3 to x = 1 is −2.2 + 4.8 = 2.6.

Of course, the signed area value found in Example 2(c) doesn’t correspond
to any actual area on the graph, as it’s the sum of a positive signed area
and a negative signed area.

112
 1 Areas, signed areas and definite integrals

Activity 2 Understanding signed areas

Consider again the graph in Example 2. In each of parts (a)–(d) below, use
the given areas to find the signed area between the graph and the x-axis,
from the first value of x to the second value of x.
(a) From x = 1 to x = 5. (b) From x = 5 to x = 7.
(c) From x = 1 to x = 7. (d) From x = 1 to x = 9.

With the definition of signed area that you’ve now seen, the method that
you’ve met in this subsection can be described concisely as follows.

Strategy:
To find an approximate value for the signed area between
the graph of a continuous function f and the x-axis, from
x = a to x = b
Divide the interval between a and b into n subintervals, each of
width (b − a)/n. For each subinterval, calculate the product
, -
endpoint of subinterval b−a
f × ,
nearest a n
and add up all these products.
In general, the larger the number n of subintervals, the closer your
answer will be to the required signed area.

Notice that the box above uses the phrase ‘endpoint of subinterval
nearest a’, rather than ‘left endpoint of subinterval’, which means the same
thing. This will be convenient later in this subsection. (Note that, in this
phrase, it’s the endpoint of the subinterval, not the subinterval itself,
that’s nearest a!)
Figures 6 and 7 illustrate the strategy above. If you add up the signed
areas of all the rectangles in Figure 6(a), then you’ll obtain an
approximate value for the signed area in Figure 6(b).

Figure 6 (a) A collection of rectangles whose total signed area is

approximately the signed area shown in (b)
113
Unit 8 Integration methods

Similarly, if you add up the signed areas of all the rectangles in Figure 7(a),
then you’ll obtain an approximate value for the signed area in Figure 7(b).

Figure 7 (a) Another instance of a collection of rectangles whose total

signed area is approximately the signed area shown in (b)

Activity 3 Calculating an approximate value for a signed area

Use the method described in the box above, with six subintervals as shown
on the left below, to find an approximate value for the signed area between
the graph of the function f (x) = 3 − x2 and the x-axis from x = −3 to
x = 3, as shown on the right below.

In the next activity you can use a computer to explore approximations to

signed areas found by using subintervals in the way that you’ve seen.

114
 1 Areas, signed areas and definite integrals

Activity 4 Calculating more approximate values for signed areas

Open the applet Approximations for signed areas. Check that the function
is set to f (x) = 3 − x2 , the values of a and b are set to −3 and 3
respectively, and the number of subintervals is set to 6. The resulting
approximation for the signed area between the graph of f and the x-axis,
from x = a to x = b, should then be the value found in the solution to
Activity 3.
Now increase the number of subintervals, and observe the effect on the
approximation. What do you think is the exact value of the signed area?
Experiment by changing the function, the values of a and b, and the
number of subintervals, to find approximate values for some other signed
areas.

So far in this subsection, whenever a signed area from x = a to x = b has

been discussed, the value of b has been greater than the value of a.
The value of b can also be equal to the value of a, and this gives a signed
area of zero, as illustrated in Figure 8. In cases like this, the method that
you’ve seen for using subintervals to find approximate values for signed
areas gives the answer zero, since the width (b − a)/n of each subinterval is
zero.

Figure 8 A signed area equal to zero

In fact, for reasons that you’ll see later in the unit, it’s useful to extend the
definition of signed area a little, to give a meaning to the phrase ‘the
signed area between the graph of f and the x-axis from x = a to x = b’
when b is less than a. It’s not immediately obvious what the phrase means
in this case, but it turns out that the natural meaning is as follows. This
definition will be important throughout the unit.

115
Unit 8 Integration methods

Suppose that f is a continuous function whose domain includes the

interval [b, a], as illustrated in Figure 9. Then the signed area between
the graph of f and the x-axis from x = a to x = b is defined to be the
negative of the signed area between the graph of f and the x-axis from
x = b to x = a.

Figure 9 The graph of a continuous function f whose domain includes

the interval [b, a]
For example, consider the graph in Figure 10. The signed area from x = 1
to x = 3 is 4, so the signed area from x = 3 to x = 1 is −4. Similarly, the
signed area from x = 6 to x = 8 is −5, so the signed area from x = 8
to x = 6 is −(−5) = 5.

Figure 10 Areas on a graph

Because of this extended definition of signed area, you now have to think a
little more carefully when you read or use the phrase ‘signed area’ (and the
signed area isn’t zero). If there’s no mention of ‘from x = a to x = b’, then
to decide whether the signed area is positive or negative you just need to
consider whether it lies above or below the x-axis. However, if the text is
of the form ‘the signed area from x = a to x = b’, then to decide whether
the signed area is positive or negative you also have to consider whether b
is greater than or less than a.
This is illustrated in the next example. A helpful way to think about
whether b is greater than or less than a is to think about whether x moves
forward or backward as it changes from x = a to x = b.

116
 1 Areas, signed areas and definite integrals

Example 3 Understanding the extended definition of signed area

The areas of some regions on a graph are marked below. In each of
parts (a)–(d), use the given areas to find the signed area between the
graph and the x-axis, from the first value of x to the second value
of x.
(a) From x = 3 to x = 3. (b) From x = 6 to x = 10.
(c) From x = 10 to x = 6. (d) From x = 6 to x = −1.

Solution
(a) From 3 to 3 is from a number to the same number.
The signed area from x = 3 to x = 3 is 0.
(b) From 6 to 10 is forward.
The signed area from x = 6 to x = 10 is −5.4.
(c) From 10 to 6 is backward.
The signed area from x = 6 to x = 10 is −5.4,
so the signed area from x = 10 to x = 6 is 5.4.
(d) From 6 to −1 is backward.
The signed area from x = −1 to x = 6 is −2.4 + 4.0 = 1.6,
so the signed area from x = 6 to x = −1 is −1.6.

In each part of the next activity, start by thinking about whether x moves
forward or backward (or stays fixed) as it moves from the first value of x
to the second value of x.

117
Unit 8 Integration methods

Activity 5 Understanding the extended definition of signed area

The graph in Example 3 is repeated below. In each of parts (a)–(g), use

the given areas to find the signed area between the graph and the x-axis,
from the first value of x to the second value of x.
(a) From x = 2 to x = 6. (b) From x = 6 to x = 2.
(c) From x = 2 to x = 10. (d) From x = 10 to x = 2.
(e) From x = 8 to x = 8. (f) From x = 2 to x = −3.
(g) From x = 10 to x = −1.

The subinterval method for finding approximate values for signed areas,
which is summarised in the box on page 113, applies no matter whether
the value of b is greater than, less than, or equal to the value of a. The
reason why it works in cases where b is less than a is that in these cases
the ‘width’ (b − a)/n of each subinterval is negative.
You might like to use the applet Approximations for signed areas to see
examples of the method giving approximate values for signed areas in cases
where b is less than a.
Of course, if you want to work out an area between a curve and the x-axis
over an interval [a, b] by using the method that you’ve seen in this
subsection, then normally you would work out the signed area from a to b,
rather than the signed area from b to a. However, it’s important to
understand what’s meant by the signed area from b to a, for reasons that
you’ll see later in the unit.

118
 1 Areas, signed areas and definite integrals

1.2 Definite integrals

Before you learn more about why signed areas are important, it’s useful for
you to learn some terminology and notation that are used when working
with them.
If f is a continuous function and a and b are numbers in its domain, then
the signed area between the graph of f and the x-axis from x = a to x = b
is called the definite integral of f from a to b, and is denoted by
! b
f (x) dx.
a
This notation is read as ‘the integral from a to b of f of x, d x’. The
numbers a and b are called the lower and upper limits of integration,
respectively.
For example, for the function f in Figure 11,
! 7 ! 9
f (x) dx = −9, f (x) dx = 2,
3 7
! 9
and f (x) dx = −9 + 2 = −7.
3
Similarly, for the same function,
! 4 ! 3
f (x) dx = 0, and f (x) dx = −(−9) = 9.
4 7

Figure 11 The graph of a function, and some areas

You’ll see a little later in this subsection where this notation for a definite
integral comes from.
!
The symbol is called the integral sign. It sometimes appears in a
/
smaller form, , when it’s in a line of typed text.

119
Unit 8 Integration methods

As with Leibniz notation for derivatives, the ‘d’ in the notation for definite
integrals has no independent meaning. In many texts, including this one,
it’s printed in upright type, rather than italic type, to emphasise that it’s
not a variable. You don’t need to do anything special when you
handwrite it.
/b
When you use the notation a f (x) dx, remember that you must include
/b
not only the a at the beginning, but also the dx at the end. Try not to
forget the dx!
You’ve now seen that a definite integral is a type of signed area, and you
saw in Unit 7 that an indefinite integral is a type of general antiderivative.
These two seemingly unrelated concepts are, as you probably suspect,
closely related. You’ll see how in Section 2.
The box below lists some standard properties of definite integrals, which
come from their definition as signed areas. These properties hold for all
values of a, b and c in the domain of the continuous function f . The
second property comes from the extended definition of signed area, to
cases where b is less than a. The third property expresses the fact that the
signed area from a to c is equal to the signed area from a to b plus the
signed area from b to c. Figure 12 illustrates this property in a case where
a < b < c, but the extended definition of signed area ensures that the
property holds even when a, b and c aren’t in this order. This is one reason
why the definition of signed area is extended in the way that you’ve seen.

Standard properties of definite integrals

! a
f (x) dx = 0
a
! a ! b
f (x) dx = − f (x) dx
b a
! c ! b ! c
f (x) dx = f (x) dx + f (x) dx
a a b

Figure 12 Two adjacent signed areas

120
 1 Areas, signed areas and definite integrals

Activity 6 Understanding definite integrals

Consider the function f whose graph is shown below. The areas of some
regions are marked. By using these areas, write down the values of the
following definite integrals.
Hint: notice that in some of these definite integrals the upper limit of
integration is less than the lower limit of integration.
! −3 ! 2 ! 7
(a) f (x) dx (b) f (x) dx (c) f (x) dx
−5 −3 2
! 2 ! 7 ! 7
(d) f (x) dx (e) f (x) dx (f) f (x) dx
−5 −3 −5
! 5 ! −3 ! 2
(g) f (x) dx (h) f (x) dx (i) f (x) dx
5 2 7
! −5 ! −3 ! −5
(j) f (x) dx (k) f (x) dx (l) f (x) dx
−3 7 7

Activity 7 Using a standard property of definite integrals

Consider again the graph in Activity 6.

! 9 ! 9
Given that f (x) dx = −15, find f (x) dx.
2 7

121
Unit 8 Integration methods

As you’d expect, you can replace the expression f (x) in the notation for a
definite integral by the formula for a particular function. For example, the
signed areas in Figure 13 are denoted by
! 1 ! 1
x2 dx and (x3 − 1) dx,
−1 0
respectively.

Figure 13 Signed areas between the graphs of particular functions and

the x-axis
As always with algebraic notation, the notation for a definite integral can
be used with letters other than the standard ones. For example, if g is a
continuous function whose domain contains the numbers p and q, and you
use t to denote the input variable of g, then the definite integral of g
from t = p to t = q is denoted by
! q
g(t) dt.
p

In fact, the input variable of the function in a definite integral is what’s

known as a dummy variable – you can change its name to any other
variable name that you like, without affecting the value of the definite
integral. For example, if f is a continuous function whose domain contains
the numbers a and b, then
! b ! b ! b
f (x) dx = f (t) dt = f (u) du,
a a a
and so on. As a particular example,
! 1 ! 1 ! 1
2 2
x dx = t dt = u2 du,
−1 −1 −1
since all these definite integrals denote the signed area shown in
Figure 13(a).
If f is any continuous function, and a and b are numbers in its domain,
/b
then you can find an approximate value for the definite integral a f (x) dx,
as accurately as you like, by using the method that you met in the last
subsection. Here’s the method expressed algebraically.

122
 1 Areas, signed areas and definite integrals

/b
Suppose that you want to find a f (x) dx, as illustrated in Figure 14(a).
You divide the interval between a and b into n subintervals, each of
width (b − a)/n, as illustrated in Figure 14(b). We’ll denote (b − a)/n
by w here, for conciseness.

/b
Figure 14 (a) A definite integral a f (x) dx (b) A collection of n
rectangles whose total signed area is approximately this definite integral
The endpoints nearest a of the subintervals are
a + 0w, a + 1w, a + 2w, ..., a + (n − 1)w.

(Remember that the strategy on page 113 specifies ‘endpoint of subinterval

nearest a’, rather than ‘left endpoint of subinterval’. The reason for this is
that it ensures that the algebraic expressions for the endpoints stated
above are correct when b is less than a, as well as when b is greater than a
or equal to a.)
So an approximate value for the definite integral of f from x = a to x = b
is given by
f (a + 0w) × w + f (a + 1w) × w + f (a + 2w) × w + · · ·
· · · + f (a + (n − 1)w) × w.
This expression can be simplified slightly by taking out the common
factor w. Doing this (and writing the common factor at the end) gives
. #
f (a + 0w) + f (a + 1w) + f (a + 2w) + · · · + f (a + (n − 1)w) w.
As the value of n gets larger and larger, the value of this expression gets
/b
closer and closer to the value of a f (x) dx.
The limit of the value of the expression, as the value of n tends to infinity,
/b
is the exact value of a f (x) dx. This idea is summarised below. The
notation ‘ lim ’ means ‘the limit, as n tends to infinity, of’.
n→∞

123
Unit 8 Integration methods

Algebraic definition of a definite integral

Suppose that f is a continuous function and a and b are numbers in
its domain. Then the definite integral of f from x = a to x = b is
given by the equation
! b .
f (x) dx = lim f (a + 0w) + f (a + 1w) + f (a + 2w) + · · ·
a n→∞
#
· · · + f (a + (n − 1)w) w

where w = (b − a)/n.

If f is a particular continuous function, and a and b are numbers in its

/b
domain, then you can sometimes calculate an exact value for a f (x) dx by
using an algebraic method to find the limit in the box above. In some
/b
cases you might even be able to obtain a general formula for a f (x) dx, in
terms of a and b. However, this is usually a complicated and difficult
process, so normally we don’t do it. Instead, if you want to evaluate a
/b
definite integral a f (x) dx, then you have two main options.
The first option is to obtain an approximate value by using subintervals in
the way that you’ve seen. We usually use a computer to carry out the
calculations. The computer algebra system that you’re using in this
module includes a command that instructs the computer to find an
approximate value for a definite integral, to a particular level of precision,
by using a version of the subinterval method that you’ve seen. You’ll learn
about this command in the final subsection of this unit.
/b
The other option for evaluating a definite integral a f (x) dx is to use an
algebraic method that arises from the fundamental theorem of calculus,
which you’ll meet in Section 2. This algebraic method involves much less
calculation than using subintervals, and it has the advantage of giving an
exact answer. However, sometimes it’s not possible or not practicable to
use this method, as you’ll see.
Now here’s an explanation of where the notation and terminology for
definite integrals come from. They arise directly from the method that
you’ve seen for calculating approximate values for definite integrals. To see
how, consider applying the method to a function f , as illustrated in
Figure 15(a).
Consider any one of the subintervals. You could denote its endpoint
nearest a by x, and its width by δx, as shown in Figure 15(b). As you saw
in Unit 6, the notation δx represents a small change in x.

124
 1 Areas, signed areas and definite integrals

Figure 15 (a) The subinterval method applied to a function f from

x = a to x = b (b) One of the subintervals and its corresponding rectangle
With this notation, the signed area of the rectangle corresponding to the
subinterval, from x to x + δx, is
f (x) δx. (2)
So the total signed area of all the rectangles is the sum of a number of
terms, each of form (2). As the number of subintervals gets larger and
larger, the size of δx gets smaller and smaller, and the total signed area of
the rectangles gets closer and closer to the definite integral of f from a
to b. So, loosely, you can think of this definite integral as the sum of
infinitely many terms of form (2), where the quantity δx is infinitely small.
Historically, as you saw in Unit 6, the quantity δx was denoted by dx when
it becomes infinitely small, so the sum described above was denoted by
! b
f (x) dx,
a
!
where the symbol is an elongated ‘S’, which stands for ‘sum’.
Johann Bernoulli (1667–1748)
To see where the term ‘integral’ comes from, remember that the verb ‘to
integrate’ means ‘to join together’. Loosely, you can think of a definite
integral as being obtained by joining together infinitely many signed areas,
each infinitely narrow, into a single signed area.

The name and principal symbol for integral calculus were discussed
by Gottfried Wilhelm Leibniz and the Swiss mathematician
Johann Bernoulli.
/ Leibniz preferred ‘calculus summatorius’ as the
name, and , the elongated S, as the symbol. Bernoulli preferred
‘calculus integralis’ as the name, and the capital letter I as the
symbol. Eventually they agreed to compromise, / adopting Bernoulli’s
name ‘integral calculus’ and Leibniz’s symbol . The first appearance
in print of the word ‘integral’ was in a work by Johann’s brother,
Jacob Bernoulli, in 1690, though Johann insisted that he had been
the one to introduce the term.
Jacob Bernoulli (1654–1705)

125
Unit 8 Integration methods

The modern notation for a definite integral, with limits at the bottom
and top of the integral sign, is due to the French mathematician
Joseph Fourier, who introduced it in his pioneering book on heat
diffusion Théorie analytique de la chaleur (The analytic theory of
heat) of 1822. He first published the notation four years previously in
an announcement for the book. Fourier accompanied
Napoleon Bonaparte on his Egyptian expedition of 1798, and later
supported Jean-François Champollion, the decipherer of the Rosetta
Stone. He is remembered today for initiating the investigation of
Fourier series. These are infinite trigonometric series, which are now
named after him, and which arose in the context of his work on heat
diffusion. You’ll meet the idea of a series in Unit 10.

Joseph Fourier (1768–1830)

To finish this section, here’s a summary of the main idea that you’ve met.

Definite integrals
Suppose that f is a continuous function, and a and b are numbers in
its domain. The signed area between the graph of f and the x-axis
from x = a to x = b is called the definite integral of f from a to b,
and is denoted by
! b
f (x) dx.
a

2 The fundamental theorem of

calculus
In this section you’ll discover how the ideas about signed areas that you
met in Section 1 are connected with the ideas about rates of change that
you met in Units 6 and 7. The two areas of mathematics are linked by an
important fact known as the fundamental theorem of calculus.
This theorem is stated in the first subsection of this section, preceded by
an explanation of why it’s true. It’s a good idea to read the explanation,
to increase your understanding, particularly if you intend to study more
mathematics after this module. However, if you find the explanation hard
to understand, then you can skip it, and go straight to the box headed
‘Fundamental theorem of calculus’, near the end of Subsection 2.1. Read
the contents of this box, and continue reading to the end of the subsection.
Make sure that you understand the fundamental theorem of calculus, and
the other important facts stated there, even if you don’t know why they’re

126
 2 The fundamental theorem of calculus

true. The two subsections that follow, Subsections 2.2 and 2.3, show you
how to use the fundamental theorem of calculus. You might like to try
reading Subsection 2.1 again later, as some of the ideas that you found
difficult the first time might seem easier once you’re more familiar with the
ways in which you can use the fundamental theorem of calculus.

2.1 What is the fundamental theorem of calculus?

The first step in obtaining a link between signed areas and rates of change
is to define, for any continuous function f , a new function whose values are
signed areas between the graph of f and the x-axis. Here’s how we do that.
Consider any continuous function f , as illustrated in Figure 16.

Figure 16 The graph of a continuous function f

Let’s choose any number, say s, in the domain of f , and define a new
function, A, to have the same domain as f and the following rule:
A(x) = signed area between the graph of f
and the x-axis, from s to x.
This definition is illustrated in Figure 17.

Figure 17 A continuous function f and a signed area A(x)

We call the function A the signed-area-so-far function for the
function f , with starting point s. Its rule can be expressed concisely as
follows:
! x
A(x) = f (t) dt.
s

127
Unit 8 Integration methods

(Here the variable t has been used in the definite integral instead of the
usual variable x, to avoid confusion with the input variable x of the
function A.)
Note that if a and b are any two numbers in the domain of f , then the
signed area between the graph of f and the x-axis, from x = a to x = b, is
given by
A(b) − A(a).
In other words,
! b
f (x) dx = A(b) − A(a).
a
This fact is illustrated in Figure 18, in a case where s < a < b. The third
property of definite integrals in the box on page 120 tells you that the fact
holds even when s, a and b aren’t in this order.

/b
Figure 18 An illustration of the fact that a
f (x) dx = A(b) − A(a)
Now let’s consider how the value of a signed-area-so-far function changes
as the value of the input variable x changes. Let’s start by looking at what
happens when the function f is a constant function.

128
 2 The fundamental theorem of calculus

For example, consider the constant function f (x) = 3, whose graph is

shown in Figure 19. Let’s take the starting point s for the
signed-area-so-far function A to be 1, as indicated in Figure 19.

Figure 19 The graph of the constant function f (x) = 3

Figure 20 shows some values of A(x).

Figure 20 Some values of A(x) when f is the constant function f (x) = 3

and the starting point s is 1

You can see that the signed area A(x) changes at the rate of 3 square units
for every unit by which x changes. For example,
if x increases by 1, then A(x) increases by 3
if x increases by 0.5, then A(x) increases by 0.5 × 3 = 1.5
if x increases by −1, then A(x) increases by −1 × 3 = −3.
(Remember that a negative increase is a decrease.)
In other words, the rate of change of A(x) with respect to x is 3.

129
Unit 8 Integration methods

More generally, you can see that for any constant function f (x) = k, and
any signed-area-so-far function A for f , the value of A(x) changes at the
rate of k square units for every unit by which x changes. In other words,
the rate of change of A(x) is k.
There are two more examples of this in Figure 21. In Figure 21(a), the
function is f (x) = 1.5, so the signed area A(x) changes at the rate of
1.5 square units for each unit by which x changes. In Figure 21(b), the
function is f (x) = −2, so the signed area A(x) changes at the rate of
−2 square units for each unit by which x changes.

Figure 21 Signed-area-so-far functions for constant functions, with

starting points s = −1 in (a) and s = 3 in (b)
Now let’s look at what happens for a function f that isn’t a constant
function. Consider, for example, the function f whose graph is shown in
Figure 22. The starting point s for the signed-area-so-far function A has
been chosen to be 2.

Figure 22 A signed-area-so-far function for a function f that isn’t a

constant function
For this function, as the value of x changes, the value of A(x) doesn’t
change at a constant rate. For example, as illustrated in Figure 23(a),
when x is close to 5 the value of A(x) changes at the rate of about 3 square
units for every unit by which x increases. Similarly, as illustrated in

130
 2 The fundamental theorem of calculus

Figure 23(b), when x is close to 7 the value of A(x) changes at the rate of
about 4 square units for every unit by which x increases.

Figure 23 (a) When x is close to 5, the rate of change of A(x) is about 3

(b) When x is close to 7, the rate of change of A(x) is about 4
More precisely, when x = 5 the instantaneous rate of change of A(x) is 3,
and when x = 7 the instantaneous rate of change of A(x) is 4. In fact –
and this is the crucial thing to realise – at any value of x, the
instantaneous rate of change of A(x) is f (x), as illustrated in Figure 24.

Figure 24 The instantaneous rate of change of A(x) at x is f (x)

In other words, the derivative of the function A is the function f , and so
the function A is an antiderivative of the function f .
You can see that this will be true for any continuous function f and any
signed-area-so-far function A for f . That is, we have the important fact
below. Because of its importance, it deserves to be called a theorem.

Theorem
Suppose that f is a continuous function, and s is any number in its
domain. Let A be the signed-area-so-far function for f with starting
point s. Then A is an antiderivative of f .

It’s this fact that leads to the fundamental theorem of calculus. Consider
any continuous function f . Let s be any number in its domain, and let A
be the signed-area-so-far function for f with starting point s.

131
Unit 8 Integration methods

Suppose that a and b are any numbers in the domain of f . Then, as you
/b
saw earlier in this subsection, the value of a f (x) dx, the signed area
between the graph of f and the x-axis from x = a to x = b, is given by
! b
f (x) dx = A(b) − A(a). (3)
a

Now let F be any antiderivative of f . Then, since A and F are both

antiderivatives of f , the value of F (b) − F (a) is the same as the value of
A(b) − A(a). That’s because, as you saw in Unit 7, the value of
F (b) − F (a) is the same no matter which antiderivative of f you take F
to be.
So you can replace A(b) − A(a) by F (b) − F (a) in equation (3) above.
This gives the crucial fact that’s known as the fundamental theorem of
calculus. It’s stated concisely below.

Fundamental theorem of calculus

Suppose that f is a continuous function whose domain contains the
numbers a and b, and that F is an antiderivative of f . Then
! b
f (x) dx = F (b) − F (a).
a

This theorem can be stated in words as follows.

Fundamental theorem of calculus (in words)

Suppose that f is a continuous function whose domain contains the
numbers a and b, and that F is an antiderivative of f . Then the
signed area between the graph of f and the x-axis from x = a to
x = b is equal to the change in the value of F as x changes from x = a
to x = b.

The equation in the fundamental theorem of calculus is true no matter

whether the value a is less than, equal to or greater than the value b. This
is the main reason why the definition of signed area is extended in the way
that you saw earlier in this unit.
Here’s another important fact that’s come out of the discussion in this
subsection. The theorem in the box on page 131 tells you immediately that
the following fact is true.

Every continuous function has an antiderivative.

132
 2 The fundamental theorem of calculus

In fact there’s some disagreement among mathematicians about exactly

which theorem should be called ‘the fundamental theorem of calculus’.
Everyone agrees that this name should be given to a theorem that links
signed areas with rates of change. However, while some mathematicians
use the name for the theorem that’s given this name in this module, other
mathematicians use it for the theorem in the box on page 131. Still other
mathematicians take the view that these two theorems are two
fundamental theorems of calculus, or two parts of a single fundamental
theorem of calculus. So if you search for ‘fundamental theorem of calculus’
on the internet, or look it up in books, then you’re likely to find a variety
of different results.

2.2 Using the fundamental theorem to find signed

areas
The fundamental theorem of calculus gives you a quick way to find the
signed area between the graph of a continuous function f and the x-axis,
from x = a to x = b, in cases where you know a formula for an
antiderivative F of f . The theorem tells you that this signed area, which is
the definite integral
! b
f (x) dx,
a
is given simply by F (b) − F (a).
For example, suppose that you want to find the area between the graph of
the function f (x) = cos x and the x-axis, from x = 0 to x = π/2, as
illustrated in Figure 25. This area is the definite integral
! π/2
cos x dx.
0
Since an antiderivative of f (x) = cos x is F (x) = sin x (as you saw in
Unit 7), the fundamental theorem of calculus tells you that
! π/2 .π#
cos x dx = sin − sin 0 = 1 − 0 = 1.
0 2
Figure 25 The area between
So the area in Figure 25 is 1 square unit. the graph of f (x) = cos x and
You may find it magical that a complicated area like the one in Figure 25 the x-axis, from x = 0 to
x = π/2
can be found from such a simple calculation. In particular, it may seem
amazing that although the area in Figure 25 depends on infinitely many
values of cos x (namely all the values of cos x for x = 0 to x = π/2), it can
be calculated from the values of an antiderivative of cos x at just two
values of x, namely x = 0 and x = π/2.
Notice that in the calculation above it doesn’t matter which of the
infinitely many antiderivatives of f (x) = cos x you use, because they all

133
Unit 8 Integration methods

differ by a constant and hence they all lead to the same final answer. For
example, using the antiderivative F (x) = sin x + 7 gives
! π/2 . .π# #
cos x dx = sin + 7 − (sin 0 + 7) = 1 + 7 − 0 − 7 = 1.
0 2
It’s usually best to use the simplest antiderivative in calculations of this
kind.
You’ll see another example of the fundamental theorem of calculus used to
find a signed area shortly, and you’ll then have the opportunity to find
some signed areas in this way yourself. First, however, it’s helpful for you
to learn a type of notation that’s convenient in calculations of this kind.
For any function F , the expression
F (b) − F (a)
can be denoted by
) 'b
F (x) a .
For example,
) 'π/2 .π#
sin x 0 = sin − sin 0 = 1 − 0 = 1.
2
You can practise evaluating expressions of this type in the next activity.

Activity 8 Evaluating expressions of the form [F (x)]ba

Evaluate the following expressions. In part (c), give your answer to three
significant figures.
) '5 ) '2π ) '1
(a) 12 x2 3 (b) cos x 0 (c) ex −1

With the square bracket notation introduced above, the fundamental

theorem of calculus can be restated as follows.

Fundamental theorem of calculus (square bracket notation)

Suppose that f is a continuous function whose domain contains the
numbers a and b, and F is an antiderivative of f . Then
! b
) 'b
f (x) dx = F (x) a .
a

In the next example, the fundamental theorem of calculus is used, together

with the square bracket notation, to find a signed area – that is, to
evaluate a definite integral.

134
 2 The fundamental theorem of calculus

Example 4 Evaluating a definite integral

Evaluate the definite integral
! 2
x3 dx.
1
(This definite integral is shown in Figure 26.)

Solution
Use the fundamental theorem of calculus. By the general formula
for the indefinite integral of a power function, an antiderivative of x3
is 14 x4 .
! 2
) '2
x3 dx = 41 x4 1
1

Evaluate this expression.

= ( 41 × 24 ) − ( 14 × 14 )
1 15
=4− 4 = 4 .
Figure 26 The definite
integral in Example 4

You’ve now seen that, when you want to find a signed area, it’s much
quicker and simpler to use the fundamental theorem of calculus than to
use the method involving subintervals that you saw in Subsection 1.1.
Another advantage of using the fundamental theorem of calculus is that it
gives you an exact answer, rather than an approximate one. However,
remember that you can use the fundamental theorem of calculus only when
you can find a formula for the antiderivative that you need. Sometimes it’s
difficult or impossible to find such a formula.
In the next activity, you can try evaluating some definite integrals by using
the fundamental theorem of calculus. Use the square bracket notation, as
in Example 4, since you need to get used to this notation. You can find the
antiderivatives that you need by using the table of standard indefinite
integrals given near the end of Unit 7. This table is also included in the
Handbook.

135
Unit 8 Integration methods

Activity 9 Evaluating definite integrals

Evaluate the following definite integrals (which are shown in Figure 27).
Give exact answers, and (except in parts (a) and (d), where the answer is a
simple fraction or an integer) also give answers to three significant figures.
! 1 ! 2 ! 4
1
(a) x2 dx (b) et dt (c) du
−1 0 1 u
! π/4 ! 1
2 1
(d) sec θ dθ (e) 2
du
−π/4 −1 1 + u

Figure 27 The definite integrals in Activity 9

The next example demonstrates how to evaluate another definite integral
by using the fundamental theorem of calculus. The only difference from
the examples that you’ve seen so far is that the antiderivative that’s
needed for the method can’t be found directly from the table of standard
indefinite integrals. Instead, we have to start by manipulating the given
function to express it as a sum of constant multiples of functions that we
can integrate, then use the constant multiple rule and sum rule for
antiderivatives, together with the table of standard indefinite integrals.
You practised finding antiderivatives in this way in Unit 7.
As shown in the example, a convenient way to carry out these sorts of
manipulations is to manipulate the expression inside the notation
/b
a . . . dx. This expression
/1 2
is known as the integrand. For example, the
definite integral 0 x dx has integrand x2 .

136
 2 The fundamental theorem of calculus

Example 5 Evaluating a more complicated definite integral

Evaluate the definite integral
! 4
(2x + 1)(x − 4)
dx.
2 x
(It is shown in Figure 28.)
Solution
Manipulate the integrand to get it into a form that you can
integrate.
! 4
(2x + 1)(x − 4)
dx
2 x
! 4 2
2x − 7x − 4
= dx
2 x
! 4, -
4
= 2x − 7 − dx
2 x
! 4, -
1
= 2x − 7 − 4 × dx
2 x
Apply the fundamental theorem of calculus, and evaluate the
result.
) '4 Figure 28 The definite
= 2 × 12 x2 − 7x − 4 ln x 2
) '4 integral in Example 5
= x2 − 7x − 4 ln x 2
= (42 − 7 × 4 − 4 ln 4) − (22 − 7 × 2 − 4 ln 2)
= 16 − 28 − 4 ln 4 − 4 + 14 + 4 ln 2
= −2 − 4(ln 4 − ln 2)
= −2 − 4 ln(4/2)
= −2 − 4 ln 2
There’s no need to find a decimal approximation, as the question
didn’t ask for one.

There are other acceptable ways to leave the final answer in Example 5.
For example, you could write it as −2(1 + 2 ln 2).

137
Unit 8 Integration methods

Activity 10 Evaluating more complicated definite integrals

Evaluate the following definite integrals (which are shown in Figure 29).
Give exact answers.
! 4 ! 0 ! 4
√ r+5
(a) 3 x dx (b) x(1 + x) dx (c) dr
1 −1 2 r

Figure 29 The definite integrals in Activity 10

When you use the square bracket notation, you can sometimes simplify
your working by using the following rules.

Constant multiple rule and sum rule for the square bracket
notation
) 'b ) 'b
kF (x) a = k F (x) a , where k is a constant
) 'b ) 'b ) 'b
F (x) + G(x) a = F (x) a + G(x) a

For example, the constant multiple rule for the square bracket notation
tells you that
) '2 ) '2
5 sin x 1 = 5 sin x 1
and the sum rule for the square bracket notation tells you that
) '2 ) '2 ) '2
sin x + cos x 1 = sin x 1 + cos x 1 .
To see why these rules hold, you just need to use the definition of the
square bracket notation. For example,
) '2
5 sin x 1 = 5 sin 2 − 5 sin 1
= 5(sin 2 − sin 1)
) '2
= 5 sin x 1 .

138
 2 The fundamental theorem of calculus

Similarly,
) '2
sin x + cos x 1 = (sin 2 + cos 2) − (sin 1 + cos 1)
= (sin 2 − sin 1) + (cos 2 − cos 1)
) '2 ) '2
= sin x 1 + cos x 1 .

As always when a constant multiple rule and a sum rule hold, the sum rule
for the square bracket notation also holds if you replace the plus sign on
each side with a minus sign. To prove this, you combine the sum rule with
the case k = −1 of the constant multiple rule. You saw this done for the
constant multiple and sum rules for derivatives in Unit 6, Subsection 2.3.
The next example illustrates how you can use the constant multiple rule
and the sum rule for the square bracket notation when you’re evaluating a
definite integral.

Example 6 Using the constant multiple rule and sum rule for the
square bracket notation
Evaluate the definite integral
! 2
(1 + x3 ) dx.
1
(It is shown in Figure 30.)
Solution
Use the fundamental theorem of calculus.
! 2
) '2
(1 + x3 ) dx = x + 14 x4 1
1

Use the sum rule.

) '2 ) '2
= x 1 + 14 x4 1
Figure 30 The definite
integral in Example 6
Use the constant multiple rule, and simplify the result.
) '2 ) '2
= x 1 + 14 x4 1

= (2 − 1) + 14 (24 − 14 )
1
=1+ 4 × 15
19
= 4 .

You can practise using the constant multiple rule and sum rule for the
square bracket notation in the next activity.

139
Unit 8 Integration methods

Activity 11 Using the constant multiple rule and sum rule for the
square bracket notation

Evaluate the following definite integrals. (They’re shown in Figure 31.)

Give your answer to part (a) to three significant figures.
! 5 ! 1
(a) x3/2 dx (b) (u − 2u2 ) du
3 −1/2

Figure 31 The definite integrals in Activity 11

Note that when you’re evaluating a definite integral it’s sometimes more
convenient not to use the constant multiple rule and/or the sum rule for
the square bracket notation, even if there’s an opportunity to use them.
You can choose whether or not to use them, as seems convenient.
A constant multiple rule and a sum rule also hold for definite integrals, as
follows.

Constant multiple rule and sum rule for definite integrals

! b ! b
k f (x) dx = k
f (x) dx, where k is a constant
a a
! b ! b ! b
(f (x) + g(x)) dx = f (x) dx + g(x) dx
a a a

These rules follow from the fundamental theorem of calculus and the
constant multiple rule and sum rule for antiderivatives, which you met
in Unit 7. To see this for the constant multiple rule, suppose that f is a
continuous function whose domain contains the numbers a and b, and
that k is a constant. Let F be an antiderivative of f . Then, by the constant
multiple rule for antiderivatives, kF (x) is an antiderivative of kf (x).

140
 2 The fundamental theorem of calculus

Hence, by the fundamental theorem of calculus and the constant multiple

rule for the square bracket notation,
! b
) 'b
k f (x) dx = k F (x) a
a
) 'b
= k F (x) a
! b
=k f (x) dx.
a
You can prove the sum rule for definite integrals in a similar way.
Remember that the sum rule for definite integrals also holds if you replace
the plus sign on each side with a minus sign, because, as mentioned earlier,
this is always the case when a constant multiple rule and a sum rule hold.
The constant multiple rule for definite integrals tells you that, for example,
the area in Figure 32(b) is twice the area in Figure 32(a).

Figure 32 The area in (b) is twice the area in (a)

Similarly, the sum rule for definite integrals tells you that, for example, the
area in Figure 33(c) is the sum of the areas in Figure 33(a) and (b).

Figure 33 The area in (c) is the sum of the areas in (a) and (b)
The next example illustrates how you can use the constant multiple rule
and the sum rule for definite integrals when you’re evaluating a definite
integral.

141
Unit 8 Integration methods

Example 7 Using the constant multiple rule and sum rule for
definite integrals
Evaluate the definite integral
! π
(x + 2 sin x) dx.
0
(It is shown in Figure 34.)
Solution
Use the fundamental theorem of calculus. Use the constant
multiple rule and sum rule, for definite integrals and for the square
bracket notation, as convenient.
! π ! π ! π
(x + 2 sin x) dx = x dx + 2 sin x dx
0 0 0
! π ! π
= x dx + 2 sin x dx
Figure 34 The definite 2 0 1π 0
) 'π
integral in Example 7 = 21 x2 + 2 − cos x 0
0
1
) 2 'π ) 'π
= 2 x 0 − 2 cos x 0
= 21 (π2 − 02 ) − 2(cos π − cos 0)
= 21 π2 − 2(−1 − 1)
= 21 π2 + 4.

You can practise using the constant multiple rule and sum rule for definite
integrals in the next activity.

Activity 12 Using the constant multiple rule and sum rule for
definite integrals

Evaluate the following definite integrals. (They are shown in Figure 35.)
Give exact answers.
! √7 ! π/4
1 3
(a) √ 2 x dx (b) (sin x + cos x) dx
3 −π/4

142
 2 The fundamental theorem of calculus

Figure 35 The definite integrals in Activity 12

As with the sum rule and constant multiple rule for the square bracket
notation, you can choose whether or not to use the sum rule and the
constant multiple rule for definite integrals in your working, as seems
convenient. Often it’s more convenient not to use them, even if there’s an
opportunity to do so.
In the next activity, you’re asked to use the fundamental theorem of
calculus to find the exact area of the cross-section of a roof that was
discussed at the beginning of Subsection 1.1.

Activity 13 Using the fundamental theorem for a practical problem

The curve of the roof of a building is given by the graph of the function
1 2
f (x) = 9 − 36 x ,
where x and f (x) are measured in metres. Use the ideas that you’ve met
in this section to find the area between this graph and the x-axis from
x = −18 to x = 18, as shown below. Hence state the area of the
cross-section of the roof.

You may remember that in Subsection 1.1 we used subintervals to work

out that the area in Activity 13 is approximately 216 m2 . The solution to
Activity 13 confirms that the area is exactly 216 m2 .
Here’s another practical example for you to try. In this example the graph
lies below the x-axis, so the signed area that you’ll obtain by using the
fundamental theorem of calculus will be negative. You have to remove the
minus sign to obtain the required area.
143
Unit 8 Integration methods

Activity 14 Solving another practical problem about area

An engineer is planning the construction of an open channel with a

parabolic cross-section, as illustrated below. The shape of the cross-section
is given by the equation
1 2
y= 32 x − 200 (−100 ≤ x ≤ 100),
where x and y are measured in centimetres, as shown in the graph below.

The maximum height of liquid permissible in the channel is the height of

the x-axis on the graph; that is, 200 cm above the lowest point of the
channel. To ensure that the channel is adequate for the expected flow of
liquid, the engineer needs to know the area shaded on the graph.
(a) Find the x-intercepts of the graph.
(b) Hence calculate the shaded area, to the nearest square centimetre.

144
 2 The fundamental theorem of calculus

The history of the fundamental theorem of calculus

You’ve seen that the fundamental theorem of calculus relates rates of
change – that is, gradients of tangents – to signed areas.
One of the earliest mathematicians to relate the problem of
calculating tangents to the problem of calculating areas was
Isaac Barrow, the first Lucasian professor at the University of
Cambridge. However, his method was presented in the geometrical
style of the day and was not computationally useful for solving
problems. Isaac Newton was one of Barrow’s students, and in 1669
Barrow resigned his professorship so that Newton could take it up.
Newton based his definition of quadrature (determining area) on the
idea that it is the converse of finding tangents. So what for some Isaac Barrow (1630–1677)
people was, and for modern-day mathematicians is, a theorem that
finding areas and tangents are inverse processes – the fundamental
theorem of calculus – was taken by Newton as the very definition of
quadrature. Newton’s way of regarding area questions as the opposite
of tangency questions is set down clearly in his De analysi of 1669.
Gottfried Wilhelm Leibniz published his first accounts of differential
and integral calculus in the newly founded journal Acta Eruditorum
Lipsiensium (Acts of the scholars of Leipzig) in 1684 and 1686
respectively, and his proof of the fundamental theorem of calculus
followed in 1693.
The first rigorous proof of the fundamental theorem of calculus was
provided in the early decades of the 19th century by the French
mathematician Augustin Louis Cauchy, who included it in his course
at the École Polytechnique in Paris. Cauchy was one of the most
productive and influential mathematicians of his generation. His
reformulation of the foundations of calculus, while very thorough, was
also very hard to understand – so much so that not only his students
but also his fellow professors protested! Augustin Louis Cauchy
(1789–1857)

145
Unit 8 Integration methods

2.3 Using the fundamental theorem to find

changes in the values of antiderivatives
The fundamental theorem of calculus isn’t useful just for solving problems
related to finding areas. It’s also useful for solving problems involving rates
of change.
As you saw in Subsection 5.3 of Unit 7, in integral calculus you often want
to solve problems of the following type. You have a continuous function f ,
and two numbers a and b in its domain, and you want to work out the
amount by which an antiderivative of f changes from x = a to x = b.
(Remember that all the antiderivatives of f change by the same amount.)
If you can find a formula for an antiderivative F of f , then it’s
straightforward to do this: you just calculate F (b) − F (a). You saw
examples like this in Unit 7.
If you can’t find a formula for an antiderivative, then you can use the
fundamental theorem of calculus to help you find an approximate answer.
The theorem tells you that, for any antiderivative F of f ,
! b
F (b) − F (a) = f (x) dx.
a
So the answer that you want is equal to the signed area between the graph
of f and the x-axis, from x = a to x = b. You can use the method that you
met in Subsection 1.1 to find an approximate value for this signed area.
You would normally do that by using a computer, as mentioned earlier.
For example, suppose that you have a velocity-time graph for an object
moving along a straight line, such as the graph in Example 8 below. Since
displacement is an antiderivative of velocity, the fundamental theorem of
calculus tells you that the change in the displacement of the object, from
any point in time to any other point in time, is equal to the signed area
between the graph and the time axis, from the first point in time to the
second point in time. This fact is used in Example 8.
As with any graph, the units for signed area on a velocity-time graph are
the units on the vertical axis times the units on the horizontal axis. So, for
example, the units for signed area on the graph in Example 8 are
m s−1 × s = m;
that is, metres. These are units for displacement, as you’d expect.

Example 8 Using the fundamental theorem of calculus for a

velocity-time graph
The graph below is the velocity–time graph of an object moving along
a straight line. The areas (in m) of some regions are marked on the
graph.

146
 2 The fundamental theorem of calculus

In each of parts (a) to (c), use the given areas to find the amount by
which the displacement of the object changes from the first time to
the second time.
(a) From t = 0 to t = 4. (b) From t = 4 to t = 10.
(c) From t = 0 to t = 10.

Solution
(a) The change in displacement from t = 0 to t = 4 is 9 m.
(b) The change in displacement from t = 4 to t = 10 is −7 m.
(c) The change in displacement from t = 0 to t = 10 is
9 m + (−7) m = 2 m.

Notice that the velocity of the object in Example 8 changes from positive
to negative at time t = 4. That is, at this time the object changes from
moving in the positive direction to moving in the negative direction, as
illustrated in Figure 36. The negative displacement that occurs after this
time cancels out most of the positive displacement that occurred initially.
Since the object was displaced by 9 m in the positive direction, and then
by 7 m in the negative direction, its final displacement from its starting
position is 2 m, as found in Example 8(c). The object travels a total
distance of 9 m + 7 m = 16 m during the 10 seconds of its motion.

Figure 36 The object in Example 8 starts moving in the positive

direction, then turns and moves in the negative direction

147
Unit 8 Integration methods

Activity 15 Using the fundamental theorem of calculus for a

velocity–time graph

The velocity–time graph of an object moving along a straight line is shown

below. The areas of some regions are marked on the graph. Use these
areas to find the answers in parts (a) to (c) below.

(a) In each of parts (i) to (iv), find the amount by which the displacement
of the object changes from the first time to the second time.
(i) From t = 0 to t = 1. (ii) From t = 3 to t = 4.
(iii) From t = 0 to t = 3. (iv) From t = 3 to t = 6.
(b) What is the displacement of the object from its starting position at
each of the following times?
(i) t = 2 (ii) t = 6
(c) What is the total distance travelled by the object during the 6 seconds
of its motion?

If you want to solve a problem that involves finding a change in the value
of an antiderivative, and you can find a formula for an antiderivative, then
it’s usually convenient to use the notation for a definite integral, and the
square bracket notation, in your working. Here’s Example 22 from
Subsection 5.3 of Unit 7 again, but with the working carried out using this
notation.

148
 2 The fundamental theorem of calculus

Example 9 Finding a change in the value of an antiderivative

Suppose that the rate of change of a quantity is given by the function
f (x) = 15 x2 (x ≥ 0), as shown in Figure 37. By what amount does the
quantity change from x = 3 to x = 6?
Solution
The change in the quantity from x = 3 to x = 6 is
! 6 2 16
1 2 1 1 3
5 x dx = 5 × 3 x
3 3
2 16
1 3
= 15 x
3
1
) '
3 6
= 15 x 3
1 3 3
= 15 (6 − 3 )
1
= 15 × 189 Figure 37 The graph of
63
= 5 y = 15 x2 (x ≥ 0)
= 12.6.

In the next activity, you’re asked to repeat Activity 36 from Subsection 5.3
of Unit 7, but using the notation for a definite integral.

Activity 16 Finding a change in displacement

Suppose that an object moves along a straight line, and its velocity v
(in m s−1 ) at time t (in seconds) is given by v = 3 − 23 t (0 ≤ t ≤ 6)
(as shown in Figure 38).
(a) Write down a definite integral that gives the change in the
displacement of the object from time t = 0 to time t = 1. Evaluate it,
and hence state how far the object travels in that time.
(b) Repeat part (a), but from time t = 4 to time t = 5.

Try using the notation for a definite integral to solve the problem in the
next activity.

Figure 38 The graph of

v = 3 − 32 t (0 ≤ t ≤ 6)

149
Unit 8 Integration methods

Activity 17 Finding another change in displacement

The driver of a car travelling at 28 m s−1 along a straight road applies the
brakes. It takes four seconds for the car to stop, and its decreasing
velocity v (in m s−1 ) during that time is given by the equation
v = 28 − 7t,
where t is the time in seconds since the brakes were applied (as shown in
Figure 39). How far does the car travel during those four seconds?

In this subsection you’ve seen examples of the following two useful facts.

For the velocity-time graph of any object, and for any two points in
Figure 39 The graph of time t = a and t = b in the time period covered by the graph, the
v = 28 − 7t (0 ≤ t ≤ 4) following hold.
• The change in the displacement of the object from t = a to t = b
is the total signed area between the graph and the time axis from
t = a to t = b.
• The total distance travelled by the object from t = a to t = b is
the total area between the graph and the time axis from t = a to
t = b (this applies when a ≤ b).

Note that if a graph is a straight line, or consists of several line segments,

then integration may not be the easiest way to work out an area or a
signed area between the graph and the horizontal axis. For example, you
can work out the area shaded in Figure 39 by using the formula for the
area of a triangle, as shown at the end of the solution to Activity 17. Try
using a similar geometric method in the next activity, which is a repeat of
Example 23 in Unit 7.

Activity 18 Finding yet another change in displacement

Suppose that a car begins to move along a straight road, and its velocity v
(in m s−1 ) during the first five seconds of its journey is given by the
equation
v = 3t,
where t is the time in seconds since it began moving, as shown in
Figure 40. How far does the car travel from the end of the fourth second to
the end of the fifth second?
Figure 40 The graph of
v = 3t (0 ≤ t ≤ 5)

150
 2 The fundamental theorem of calculus

2.4 Notation for indefinite integrals

Until now, when we’ve been working with antiderivatives and indefinite
integrals, we’ve been using a lower-case letter, such as f , to denote the
function that we start with, and the corresponding upper-case letter, such
as F , to denote any of its antiderivatives, or its indefinite integral.
For example, some antiderivatives of the function f (x) = x2 are
F (x) = 31 x3 , F (x) = 13 x3 + 7, and F (x) = 13 x3 − 92 ,
and the indefinite integral of this function f is
F (x) = 13 x3 + c.

However, there’s a standard notation for indefinite integrals, which arises

from the link between antiderivatives and definite integrals. For any
function f that has an antiderivative, the indefinite integral of f (x) is
denoted by
!
f (x) dx.

(This notation is read as ‘the integral of f of x, d x’.)

For example,
!
x2 dx = 31 x3 + c,

and similarly
!
sin x dx = − cos x + c.

(You saw in Unit 7 that the indefinite integral of sin x is − cos x + c.)
Note that this notation is used only for indefinite integrals – it’s never used
for antiderivatives. So it’s important to remember the following fact.

/
If an equation has the notation . . . dx on one side only, then there
must be an arbitrary constant on the other side.

For example, the equation

!
x2 dx = 31 x3

is incorrect. The correct equation is

!
x2 dx = 31 x3 + c,

as you saw above.

We’ll continue to use the capital letter notation for antiderivatives, when
it’s convenient to do so.

151
Unit 8 Integration methods

The notation for indefinite integrals introduced above, and the similar
notation for definite integrals, are together known as integral notation.
We refer to either an indefinite integral or a definite integral as an
integral. In the indefinite integral
!
f (x) dx,

the expression f (x) is called the integrand, just as in a definite integral.

The next example illustrates how integral notation is used for indefinite
integrals in practice. It’s similar to the examples and activities on finding
indefinite integrals in the second half of Unit 7, except that it uses integral
notation.

Example 10 Finding indefinite integrals, using integral notation

Find the following indefinite integrals.
! !
2 x−5
(a) sec x dx (b) dx
x
Solution
(a) The indefinite integral of sec2 x is given in the table of
standard indefinite integrals, so you can simply write it down.
!
sec2 x dx = tan x + c

(b) Manipulate the integrand to get it into a form that allows you
to use the constant multiple rule and sum rule for antiderivatives,
together with standard indefinite integrals.
! ! , -
x−5 5
dx = 1− dx
x x
! , -
1
= 1−5× dx
x
= x − 5 ln |x| + c

152
 2 The fundamental theorem of calculus

You can practise using integral notation for indefinite integrals in the
following activity. Don’t forget to add the ‘ + c’ in each part.

Activity 19 Finding indefinite integrals, using integral notation

Find the following indefinite integrals.

! ! !
1
(a) ex dx (b) (x + 3)(x − 2) dx (c) dx
x3
! ! !
1 −3
(d) (sin θ − cos θ) dθ (e) 2
dx (f) du
1+x 1 + u2
! !
1 (1 + x2 )
(g) dr (h) dx
2(1 + r2 ) 2

Since an indefinite integral is just an antiderivative with an arbitrary

constant added on, the constant multiple rule and sum rule for
antiderivatives, from Unit 7, can be written in terms of indefinite integrals.
These forms of the rules are stated in the following box, in integral
notation.

Constant multiple rule and sum rule for indefinite integrals

! !
k f (x) dx = k f (x) dx, where k is a constant
! ! !
(f (x) + g(x)) dx = f (x) dx + g(x) dx

As with the other constant multiple rules and sum rules, you can choose
whether or not to use these rules in your working, as seems convenient.
The next example repeats Example 10(b), but uses the constant multiple
rule and sum rule for indefinite integrals in the forms stated above.

153
Unit 8 Integration methods

Example 11 Using the constant multiple rule and sum rule for
indefinite integrals
Find the indefinite integral
!
x−5
dx.
x
Solution
Manipulate the integrand to get it into a form that allows you to
use the constant multiple rule and sum rule for indefinite integrals,
together with standard indefinite integrals.
! ! , -
x−5 5
dx = 1− dx
x x
! !
1
= 1 dx − 5 dx
x
= x − 5 ln |x| + c

Here’s something to notice about Example 11. You might have thought
that each of the two indefinite integrals in the second line of the working
should have been replaced by an expression containing an arbitrary
constant. That would lead to a solution something like this:
! ! , -
x−5 5
dx = 1− dx
x x
! !
1
= 1 dx − 5 dx
x
= (x + c) − 5(ln |x| + d)
= x − 5 ln |x| + c − 5d,
where c and d are arbitrary constants.
However, since the arbitrary constants c and d can take any value, the
expression c − 5d can also take any value, so we may as well replace it by a
single letter that can take any value. It’s convenient to use the usual
letter, c. In general, if an expression consists of a sum of constant multiples
of indefinite integrals, then it needs only a single arbitrary constant.
You can practise using the constant multiple rule and sum rule for
indefinite integrals in the next activity.

154
 3 Integration by substitution

Activity 20 Using the constant multiple rule and sum rule for
indefinite integrals

Find the following indefinite integrals.

! , - !
3 1
(a) − dx (b) (cosec x)(cosec x + cot x) dx
x 1 + x2

You’ve now reached the end of the material in this module that tells you
what integration is all about. In the rest of this unit you’ll concentrate on
actually finding indefinite and definite integrals.
Before you go on to that, you might like to watch the 25-minute video
The birth of calculus, which is available on the module website. It tells you
more about the history of the development of calculus.

3 Integration by substitution
To allow you to fully exploit all the ideas about integration that you’ve
met, it’s important that you’re able to find formulas for the indefinite
integrals of as wide a range of functions as possible. In the rest of this
unit, you’ll learn to integrate a much wider variety of functions than you
learned to integrate in Unit 7. In the final subsection of the unit, you’ll
learn how to use the module computer algebra system to find indefinite
and definite integrals.
Unfortunately, as mentioned in Unit 7, it’s generally more tricky to
integrate functions than it is to differentiate them. That’s because there
are no rules for antiderivatives that are similar to the product, quotient
and chain rules for derivatives, so you can’t usually integrate functions in
the systematic way that you can differentiate them.
However, there are still many useful techniques that you can use to
integrate functions. Each of these techniques applies to functions with
particular characteristics. You’ll meet some of these techniques in the rest
of this unit. You’ll learn how to recognise functions to which each
technique can be applied, and how to use the techniques.
The technique that you’ll learn in this section is integration by substitution.
This technique might seem quite complicated when you first meet it, but
you should find it more straightforward once you’ve had some practice
with it. If you find it difficult, then make sure that you watch the tutorial
clips, and try some extra practice. Remember that there are more practice
exercises in the practice quiz and exercise booklet for this unit.

155
Unit 8 Integration methods

3.1 Basic integration by substitution

Integration by substitution is based on reversing the chain rule for
differentiation.
Consider what happens when you differentiate an expression by using the
chain rule. Remember that you start by recognising that the expression is
a function of ‘something’. You differentiate the expression with respect to
the ‘something’, then you multiply the result by the derivative of the
‘something’ with respect to the input variable (usually x). For example, by
the chain rule,
d & + & +
sin(x2 ) = cos(x2 ) (2x).
dx
Since differentiation is the reverse of integration, this equation tells you
that
!
& +
cos(x2 ) (2x) dx = sin(x2 ) + c.

Whenever you differentiate an expression by using the chain rule, you

always obtain an expression of the form
f (something) × the derivative of the something,
where f is a function that you can integrate. The essential idea of
integration by substitution is to recognise an expression that you want to
integrate as having this form, and then perform the integration by
reversing the chain rule.
To help you reverse the chain rule, it’s helpful to denote the ‘something’ by
an extra variable, in the way that you did when you first learned to use the
chain rule in Unit 7. The usual choice of letter for the extra variable is u.
Here’s how you can use this method to find the integral
!
& +
cos(x2 ) (2x) dx,

if you hadn’t first seen the integrand obtained as an ‘output’ of the chain
rule.
The first step is to recognise that the integrand has the form
f (something) × the derivative of the something,
where f is a function that you can integrate. You can see that it does,
since it is
cos(something) × the derivative of the something,
where the ‘something’ is x2 .
Then, since the ‘something’ is x2 , you put u = x2 . This gives du/dx = 2x.
Hence, in the integral you can replace cos(x2 ) by cos u, and 2x by du/dx,
to give
!
du
cos(u) dx.
dx

156
 3 Integration by substitution

Now here’s the crucial step that you need to apply at this stage. Any
integral of the form
!
du
f (u) dx
dx
is equal to the simpler integral
!
f (u) du.

This is the step that uses the chain rule, and you’ll see an explanation of
why it’s correct later in this subsection. Notice that it’s easy to remember,
because it looks like we’ve simply cancelled a ‘dx’ in a denominator with a
‘dx’ in a numerator. (Of course that’s not what we’ve done, since du/dx
isn’t a fraction.)
So the integral that we’re trying to find here can be expressed in terms
of u as the simpler integral
!
cos u du,

where u = x2 . You can now do the integration, which gives

sin u + c.
The final step is to express this answer in terms of x, using the fact that
u = x2 . This gives the final answer
sin(x2 ) + c.
So we’ve now worked out that
!
& +
cos(x2 ) (2x) dx = sin(x2 ) + c,

as expected.
Here’s a summary of the method used above, which is the method known
as integration by substitution.

Integration by substitution
1. Recognise that the integrand is of the form
f (something) × the derivative of the something,
where f is a function that you can integrate.
2. Set the something equal to u, and find du/dx.
3. Hence write the integral in the form
!
f (u) du,
! !
du
by using the fact that f (u) dx = f (u) du.
dx
4. Do the integration.
5. Substitute back for u in terms of x.

157
Unit 8 Integration methods

As illustrated by the example above, the effect of substituting the

variable u in place of the ‘something’ in the integral is to reduce the
integral to a simpler integral in terms of u, which you may be able to find
more easily.
You’ll see another example of integration by substitution shortly, and
you’ll then have a chance to try it for yourself, but first it’s useful for you
to learn about a convenient way to carry out step 3 of the method.
Consider once more the integral above:
!
& +
cos(x2 ) (2x) dx.

Remember that we put u = x2 , which gave du/dx = 2x, and this allowed
us to write the integral in terms of u as
!
cos u du.

It’s convenient to think of this new form of the integral as being obtained
from the original form by making two replacements, as shown below:
!
& +
cos(x2 ) (2x) dx .
* %$ 3 * %$ 3
cos u du

It’s straightforward to work out that you can replace cos(x2 ) by cos u, as
that just comes from the equation u = x2 .
A helpful way to work out that you can replace (2x) dx by du is to imagine
‘cross-multiplying’ in the equation
du
= 2x,
dx
to obtain
du = (2x) dx.
This isn’t a real equation, of course, since du and dx don’t have
independent meanings outside the notation for a derivative or integral.
But it tells you immediately that you can replace (2x) dx by du. (In fact,
it’s possible to attach a meaning to du and dx, and hence to the equation
above, but this is outside the scope of this module.)
Here’s another example of integration by substitution, which illustrates all
the ideas above.

Example 12 Integrating by substitution

Find the integral
!
etan x sec2 x dx.

158
 3 Integration by substitution

Solution
Since the derivative of tan x is sec2 x, the integrand is of the form
esomething × the derivative of the something.
So set the something equal to u.
du
Let u = tan x; then = sec2 x.
dx
Imagine ‘cross-multiplying’ in the second equation to obtain
du = (sec2 x) dx.
So
! !
tan x 2
e sec x dx = eu du

Do the integration.
= eu + c
Substitute back for u in terms of x.
= etan x + c.

Here are some examples for you to try. Make sure that you have your
Handbook to hand, so you can consult the tables of standard derivatives
and standard integrals.

Activity 21 Integrating by substitution

Find the following indefinite integrals.

! !
& +
(a) ecos x (− sin x) dx (b) sin(x3 ) (3x2 ) dx
! , - ! , -
1 1
(c) cos x dx (d) (− sin x) dx
sin x cos2 x
! ! , -
4 1
(e) sin x cos x dx (f) cos x dx
1 + sin2 x

In the examples that you’ve seen so far, it’s been fairly straightforward to
decide what the ‘something’ should be, so that the integrand is of the form
f (something) × the derivative of the something.
However, sometimes you need to think about this more carefully, as
illustrated in the next example.

159
Unit 8 Integration methods

Example 13 Choosing the ‘something’ when you integrate by

substitution
Find the integral
! , -
1
cos x dx.
2 + sin x
Solution
The integrand is of the form
1
× the derivative of the something,
2 + something
where the something is sin x.
It’s also of the form
1
× the derivative of the something,
something
where the something is 2 + sin x.
Choosing the first option would lead to the first part of the integrand
being replaced by 1/(2 + u), which isn’t straightforward to integrate.
Choosing the second option would lead to the first part of the
integrand being replaced by 1/u, which you can integrate using a
result from the table of standard integrals.
So choose the second option: take the something to be 2 + sin x, and
set it equal to u.
du
Let u = 2 + sin x; then = cos x.
dx
Imagine ‘cross-multiplying’ to obtain du = cos x dx.
So
! , - !
1 1
cos x dx = du
2 + sin x u
= ln |u| + c
= ln |2 + sin x| + c

In the particular case here you can remove the modulus signs,
since 2 + sin x is always positive.

= ln(2 + sin x) + c.

In each part of the next activity it’s useful to take the ‘something’ to be an
expression that includes a constant term, in a similar way to Example 13.

160
 3 Integration by substitution

Activity 22 Choosing the ‘something’ when you integrate by

substitution
Find the following indefinite integrals.
! ! "
7
(a) (4 + cos x) (− sin x) dx (b) 1 + x2 (2x) dx
! ! , -
5 10 4 1
(c) (x − 8) (5x ) dx (d) ex dx
ex + 5
!
& +
(e) sin(5 + 2x3 ) (6x2 ) dx

In the next activity, some of the integrals are similar to those in

Activity 22, whereas others are similar to those in Activity 21. You have
to decide in each case what the ‘something’ should be. Remember that the
idea is that when you replace the something by u in the first part of the
integrand, you obtain an expression in u that you can integrate.

Activity 23 Choosing the ‘something’ when you integrate by

substitution
Find the following indefinite integrals.
! !
x x
(a) (cos(e )) e dx (b) e1+sin x (cos x) dx
! , - !
1 9
(c) (10x ) dx (d) (cos3 x)(− sin x) dx
x10 + 6

Finally in this subsection, as promised near the start, here’s an

explanation of why
! !
du
f (u) dx = f (u) du,
dx
where the variable u is a function of x, and f is a function that you can
integrate. To justify this, we let F be an antiderivative of f , and apply the
chain rule to F (u). This gives
d & + du
F (u) = f (u) .
dx dx
It follows from this equation that
!
du
f (u) dx = F (u) + c.
dx
However, because F is an antiderivative of f , we also have
!
f (u) du = F (u) + c.

Comparing the final two equations above proves the result.

161
Unit 8 Integration methods

3.2 Integration by substitution in practice

Usually when you have an integrand that you can integrate by
substitution, it won’t initially be exactly in the form
f (something) × the derivative of the something.
You have to start by recognising that it can be rearranged into this form,
then rearrange it accordingly. Here are two examples that illustrate some
things that you can do.

Example 14 Rearranging an integral before using substitution

Find the integral
!
ex cos(ex ) dx.

Solution
The derivative of ex is ex , so the integrand is in the required form,
except that the two factors cos(ex ) and ex are in the ‘wrong’ order. If
you swap them, then the integrand will be exactly in the required
form.
du
Let u = ex ; then = ex .
dx
Imagine ‘cross-multiplying’ to obtain du = ex dx. Then start by
swapping the two factors in the integrand.
So
! !
x x
e cos(e ) dx = (cos(ex )) ex dx
!
= cos u du
= sin(u) + c
= sin(ex ) + c.

162
 3 Integration by substitution

Example 15 Rearranging another integral before using substitution

Find the integral
!
& +8
x2 1 + x3 dx.

Solution
The derivative of 1 + x3 is 3x2 , which is ‘nearly’ x2 (it just has the
wrong coefficient). So if you swap the order of the two factors x2 and
(1 + x3 )8 , then the integrand will be nearly in the required form.
du
Let u = 1 + x3 ; then = 3x2 .
dx
Imagine ‘cross-multiplying’ to obtain du = 3x2 dx. Then start by
swapping the two factors in the integrand.
So
! !
& +8 & +8
x 1 + x3 dx =
2
1 + x3 x2 dx

The only problem is that in place of x2 you need 3x2 . To achieve

this, multiply by 3 inside the integral, and divide by 3 outside, to
compensate. (The constant multiple rule for indefinite integrals tells
you that it’s okay to do this.)
!
& +8
=3 1
1 + x3 (3x2 ) dx

Now do the substitution.

!
= 3 u8 du
1

1 1 9
= 3 × 9u + c
1 9
= 27 u + c
1 3 9
= 27 (1 + x ) + c.

Here’s something to notice

/ about Example 15. You might have thought
that when the integral u8 du in the fourth-last line of the solution was
replaced by 91 u9 + c, the solution should have proceeded like this:
!
= 3 u8 du
1

& +
= 13 91 u9 + c
1 9 1
= 27 u + 3 c
1 3 9
= 27 (1 + x ) + 31 c.

163
Unit 8 Integration methods

However, since the arbitrary constant c can take any value, the expression
1
3 c can also take any value, so we may as well replace it by a single letter
that can take any value. It’s convenient to use the usual letter, c.
This sort of thing occurs frequently when you integrate. You never need to
multiply or divide a term that is an arbitrary constant by a number – you
just replace it by another arbitrary constant, and you usually use the
standard letter, c.
Another point to remember from the working in Example 15 is the
strategy of multiplying by a constant inside an integral, and dividing by
the same constant outside to compensate. It’s often helpful to do this
when you use integration by substitution. It’s okay to do it because the
constant multiple rule for indefinite integrals tells you that for any
function f that has an antiderivative, and any non-zero constant k,
! ! !
1 1
f (x) dx = kf (x) dx = kf (x) dx.
k k
Remember that the constant k can be either positive or negative. In
particular, it can be −1.
Remember too that k must be a constant. You can’t do the same thing
with an expression in x.
Here are some examples for you to try.

Activity 24 Rearranging integrals before using substitution

Find the following integrals.

! ! " !
2 3
(a) x cos(x ) dx (b) x 1 + x2 dx (c) cos4 x sin x dx

The next example illustrates another way in which you can sometimes
rearrange an integrand to get it into the right form for integration by
substitution.

Example 16 Rearranging another integral before using substitution

Find the integral
!
x
dx.
1 − x2

164
 3 Integration by substitution

Solution
The derivative of the denominator 1 − x2 is −2x, which is ‘nearly’
the numerator x (it just has the wrong coefficient). So if you split the
numerator from the rest of the fraction, then the integrand will be
nearly in the required form.
du
Let u = 1 − x2 ; then = −2x.
dx
Imagine ‘cross-multiplying’ to obtain du = −2x dx. Then start by
splitting the numerator from the rest of the fraction in the
integrand.
So
! ! , -
x 1
dx = x dx
1 − x2 1 − x2
The only problem is that in place of x you need −2x. To achieve
this, multiply by −2 inside the integral, and divide by −2 outside to
compensate.
! , -
1 1
= −2 (−2x) dx
1 − x2
Now do the substitution.
!
1 1
= −2 du
u
= − 21 ln |u| + c
= − 12 ln |1 − x2 | + c.

The type of integral in Example 16 occurs quite frequently, so it’s worth

learning to recognise it. You can use integration by substitution to find
any integral of the form
!
the derivative of the something
dx,
something
or
!
the derivative of the something
dx, where n is a constant.
(something)n
The constant n can be any nonzero real number. Here are some integrals
of this type for you to find.

165
Unit 8 Integration methods

Activity 25 Rearranging more integrals before using substitution

Find the following integrals.

! ! !
x ex cos x
(a) dx (b) dx (c) dx
1 + 2x2 (2 + ex )2 3 − sin x

Here are three more integrals for you to find by using substitution.
They’re a mixture of the types in Activities 24 and 25.

Activity 26 Rearranging even more integrals before using

substitution
Find the following integrals.
! ! !
ex
(a) ecos x sin x dx (b) dx (c) (1 + 12 sin x)2 cos x dx
3 − ex

You can use integration by substitution to find the indefinite integral of

the tangent function. You’re asked to do that in the next activity.

Activity 27 Finding the indefinite integral of tan

By writing
sin x
tan x = ,
cos x
and using integration by substitution, find
!
tan x dx.

Remember that when you’ve found an indefinite integral, you can always
check your answer by differentiating it. This can be particularly helpful
when you found the integral by using integration by substitution.

166
 3 Integration by substitution

3.3 Integrating functions of linear expressions

This subsection is about a particular type of integral that occurs
frequently, and which you can find by using integration by substitution.
The integrals
! ! !
1
sin(5x − 2) dx, dx, and e−9x dx
4−x
all have integrands of the form f (ax + b), where f is a function, and
a and b are constants with a (= 0. In other words, each of the integrands is
of the form f (a linear expression in x).
You can always find an integral of this form by using integration by
substitution, provided that you can integrate the function f . Here’s an
example.

Example 17 Integrating a function of a linear expression

Find the integral
!
sin(5x − 2) dx.

Solution
The derivative of 5x − 2 is 5, so the integral will be in the required
form if you rearrange it to obtain 5 inside.
du
Let u = 5x − 2; then = 5. So
dx
! !
sin(5x − 2) dx = 15 sin(5x − 2) × 5 dx
!
= 15 sin(5x − 2) × 5 dx
!
1
= 5 sin u du

= − 15 cos u + c
= − 15 cos(5x − 2) + c.

167
Unit 8 Integration methods

Here are some similar integrals for you to find.

Activity 28 Integrating functions of linear expressions

Find the following integrals.

! ! !
6x−1
(a) e dx (b) sin(4x) dx (c) e−9x dx
! ! !
−x/3 1
(d) e dx (e) cos(3 − 7x) dx (f) dx
4−x
! ! !
1 1
(g) dx (h) dx (i) sec2 (6x) dx
(2x + 1)2 (1 − x)3

The next example uses exactly the same ideas as Example 17 and
Activity 28, but is a little more complicated.

Example 18 Integrating another function of a linear expression

Find the integral
!
1
dx.
1 + 9x2
Solution
1
You can write the integrand as , which is of the
1 + (3x)2
1
form f (3x) where f (x) = . The integral of this function f is a
1 + x2
standard integral. Since the derivative of 3x is 3, you need to
rearrange the given integral to obtain 3 inside.
du
Let u = 3x; then = 3. So
dx
! !
1 1
2
dx = dx
1 + 9x 1 + (3x)2
!
1
= 13 × 3 dx
1 + (3x)2
!
1 1
=3 du
1 + u2
= 31 tan−1 u + c
= 1
3 tan−1 (3x) + c.

168
 3 Integration by substitution

Here are some similar integrals for you to find.

Activity 29 Integrating more functions of linear expressions

Find the following integrals.

! ! !
1 1 1
(a) √ dx (b) dx (c) dx
1 − 4x2 1 + 2x2 4 + 9x2
Hint for part (c): start by writing
1 1 1
as × .
4 + 9x2 4 1 + 49 x2

The answers to Activities 28 and 29 all follow the same general pattern, as
follows. Suppose that you have an integral of the form
!
f (ax + b) dx,

where a and b are constants with a (= 0, and you know an antiderivative F

of the function f . To find the integral by substitution, you put u = ax + b,
du
which gives = a. Then
dx
! !
1
f (ax + b) dx = f (ax + b) a dx
a
!
1
= f (u) du
a
1
= F (u) + c
a
1
= F (ax + b) + c.
a
So we have the following general fact.

Indefinite integral of a function of a linear expression

If f is a function with antiderivative F , and a and b are constants
with a (= 0, then
!
1
f (ax + b) dx = F (ax + b) + c.
a

It’s worth remembering this fact, and using it to deal with integrals where
the integrand is a simple function of a linear expression, rather than using
integration by substitution every time. For example, the fact above tells
you immediately that
!
cos(3x − 2) dx = 31 sin(3x − 2) + c.

169
Unit 8 Integration methods

The following special case of the fact in the box above is useful especially
often.

Indefinite integral of a function of a multiple of the variable

If f is a function with antiderivative F , and a is a non-zero constant,
then
!
1
f (ax) dx = F (ax) + c.
a

For example,
!
e−2x dx = − 12 e−2x + c.

Try using the facts in the boxes above in the following activity.

Activity 30 Integrating functions of linear expressions efficiently

Find the following integrals.

! ! !
(a) cos(10x) dx (b) sin(2 − 5x) dx (c) e3t+4 dt
! ! !
1
(d) cos( 12 θ) dθ (e) e−4p dp (f) dx
7x − 4
! ! !
1 2
(g) dr (h) sec (5x − 1) dx (i) sin( 51 x) dx
1−r
! ! !
−x x/2
(j) e dx (k) e dx (l) sec2 (2 − 3p) dp
! , -
2−x
(m) sin dx
3

The next example is of a type that you met in Unit 7, but it involves a
function that you’ve learned how to integrate in this section.

170
 3 Integration by substitution

Example 19 Integrating in a practical situation

The velocity of an object moving along a straight line is given by the
equation
& +
v = 12 sin 32 t − 2 ,
where t is the time in seconds since the object began moving, and v is
the velocity of the object, in m s−1 . The displacement of the object at
time t = 0 is 3 m.
Find an equation for the displacement s (in metres) of the object, in
terms of t. Give the constant term to two decimal places.
Solution
Use the fact that displacement is an antiderivative of velocity.
The velocity v (in m s−1 ) of the object at time t (in seconds) is given
by
& +
v = 12 sin 32 t − 2 .
Hence the displacement s (in m) of the object at time t (in seconds) is
given by
!
1
&3 +
s= 2 sin 2 t − 2 dt
1 1 & & ++
= × − cos 23 t − 2 + c
2 3/2
& +
= − 13 cos 32 t − 2 + c,
where c is an arbitrary constant.
Use information given in the question to find the value of c.
When t = 0, s = 3. Hence
& +
3 = − 31 cos 32 × 0 − 2 + c
3 = − 31 cos(−2) + c
1
c=3+ 3 cos(−2)
1
c = 3 + cos 2
3
c = 2.86 (to 2 d.p.).
So the equation for the displacement of the object in terms of time is
& +
s = − 31 cos 23 t − 2 + 2.86.

171
Unit 8 Integration methods

Activity 31 Integrating in a practical example

The velocity of an object moving along a straight line is given by the

equation
& +
v = 2 cos 21 t + 5 ,
where t is the time in seconds since the object began moving, and v is the
velocity of the object, in m s−1 . The displacement of the object at time
t = 0 is 4 m.
(a) Find an equation for the displacement s (in metres) of the object, in
terms of t. Give the constant term to two decimal places.
(b) Find the displacement of the object at time t = 10. Give your answer
in metres to two significant figures.

3.4 Integration by substitution for definite integrals

Once you’ve used integration by substitution to find an indefinite integral,
you can find any definite integral with the same integrand by using the
fundamental theorem of calculus in the usual way. Here’s an example.

Example 20 Evaluating a definite integral by substitution (first

method)
Evaluate the definite integral
! 3π/2
esin x cos x dx.
π/2

(It’s shown in Figure 41.) Give your answer to three significant

figures.
Solution
Use integration by substitution to obtain the indefinite integral of
the integrand.
Figure 41 The definite du
integral in Example 20 Let u = sin x, then = cos x. So
dx
! !
esin x cos x dx = eu du
= eu + c
= esin x + c.

172
 3 Integration by substitution

Apply the fundamental theorem of calculus.

Hence
! 3π/2
) '3π/2
esin x cos x dx = esin x π/2
π/2

= esin(3π/2) − esin(π/2)
= e−1 − e1
1
= −e
e
= −2.35 (to 3 s.f.).

An alternative way to use integration by substitution to evaluate a definite

integral is to work with the definite integral from the start, rather than
first finding an indefinite integral. If you do this, then because the limits of
integration are values of x, when you change the variable from x to u you
must also change the limits of integration to the corresponding values of u.
This method is demonstrated in the next example. The definite integral in
this example is the same as the one in Example 20.

Example 21 Evaluating a definite integral by substitution (second

method)
Evaluate the definite integral
! 3π/2
esin x cos x dx.
π/2

Give your answer to three significant figures.

Solution
Use integration by substitution.
du
Let u = sin x, then = cos x.
dx
Find the values of u that correspond to the values of x that are
the limits of integration.
.π#
Putting x = π/2 into u = sin x gives u = sin = 1.
2
, -
3π
Putting x = 3π/2 into u = sin x gives u = sin = −1.
2

173
Unit 8 Integration methods

Change the variable from x to u, remembering that you also have

to change the limits of integration.
So
! 3π/2 ! −1
sin x
e cos x dx = eu du
π/2 1
) '−1
= eu 1
= e−1 − e1
= −2.35 (to 3 s.f.).

You can try using this method in the next activity.

Activity 32 Evaluating definite integrals by substitution

Evaluate the following definite integrals. Give your answers to three

significant figures.
! 1 ! π ! 0
x 3 2
(a) 2
dx (b) cos x sin x dx (c) xe−3x dx
0 1 + 2x π/2 −1

3.5 Finding more complicated integrals by

substitution
You can sometimes use integration by substitution to find an integral even
if the integrand isn’t obviously of the form
f (something) × the derivative of the something.
In this subsection, you’ll see some examples where the integrand is nearly
of this form, except that it also contains a further factor, which is an
expression in x. If you can express this further factor in terms of the
variable u that you use for the substitution, then you may be able to
simplify the integrand into a form that you can integrate. Here’s an
example.

174
 3 Integration by substitution

Example 22 Expressing a further factor in terms of u when

integrating by substitution
Find the integral
!
(x + 1)(3x + 1)9 dx.

Solution
If the integral were just
!
(3x + 1)9 dx,

then we could use the substitution u = 3x + 1, and find the integral

by first writing it as
!
1
3 (3x + 1)9 × 3 dx.

However, the integrand contains a further factor, x + 1. Try using the

substitution, expressing the further factor x + 1 in terms of u.
du
Let u = 3x + 1. Then = 3.
dx
Also, rearranging the equation u = 3x + 1 gives
u = 3x + 1
3x = u − 1
x = 31 (u − 1).
Hence
x + 1 = 13 (u − 1) + 1
= 31 (u + 2).
So
! !
9
(x + 1)(3x + 1) dx = 1
3 (x + 1)(3x + 1)9 × 3 dx
!
= 1
3
1
3 (u + 2)u9 du
!
= 1
9 (u + 2)u9 du
!
= 1
9 (u10 + 2u9 ) du
&
1 1 11 1 10
+
= 9 11 u + 2 × 10 u +c
1 11 1 10
= 99 u + 45 u + c
1 11 1 10
= 99 (3x + 1) + 45 (3x + 1) + c.

175
Unit 8 Integration methods

Here are two similar integrals for you to try.

Activity 33 Expressing further factors in terms of u

Use integration by substitution to find the following indefinite integrals.

! !
√ x−2
(a) x 1 + x dx (b) dx
(x + 3)3

You’ll learn more about integration by substitution in the module MST125

Essential mathematics 2, if it’s in your study programme.

4 Integration by parts
In this section you’ll meet another integration technique, integration by
parts. Like integration by substitution, this technique is useful for
integrating functions that have certain characteristics. It’s based on the
product rule for differentiation.

4.1 Basic integration by parts

Suppose that f and g are functions, and that the function G is an
antiderivative of the function g. Consider the equation that you obtain
when you use the product rule to differentiate the product f (x)G(x):
, - , -
d d d
(f (x)G(x)) = f (x) G(x) + G(x) f (x) ,
dx dx dx
that is,
d
(f (x)G(x)) = f (x)g(x) + G(x)f ! (x).
dx
If you swap the order of G(x) and f ! (x) in the final term, and subtract this
term from both sides of the equation, then you obtain
d
(f (x)G(x)) − f ! (x)G(x) = f (x)g(x).
dx
Swapping the sides of this equation gives
d
f (x)g(x) = (f (x)G(x)) − f ! (x)G(x).
dx

176
 4 Integration by parts

If you now integrate both sides of this equation with respect to x, then you
obtain
! !
f (x)g(x) dx = f (x)G(x) − f ! (x)G(x) dx.

There’s no need to add an

/ explicit arbitrary constant here, because it’s
included/in the integral f (x)g(x) dx on the left-hand side, and in the
integral f ! (x)G(x) dx on the right-hand side.
The formula above is known as the integration by parts formula. It’s
stated again in the box below, for easy reference. If you’ve met it before,
then you may have seen it stated in a different form. This is explained at
the end of this section.

Integration by parts formula (Lagrange notation)

! !
f (x)g(x) dx = f (x)G(x) − f ! (x)G(x) dx.

Here G is an antiderivative of g.

The integration by parts formula is useful for integrating some products of

the form f (x)g(x). You can see that it changes the problem of integrating
f (x)g(x) into the problem of integrating f ! (x)G(x), where G is an
antiderivative of g.
So it’s useful for integrating products of the form f (x)g(x) that have the
following characteristics.
• You can find an antiderivative G of g; in other words, you can
integrate the expression g(x).
• The product f ! (x)G(x) is easier to integrate than the product
f (x)g(x). For example, this may be the case if f ! (x) is simpler than
f (x).
When we integrate by using the integration by parts formula, we say that
we’re integrating by parts. Here’s an example.

177
Unit 8 Integration methods

Example 23 Integrating by parts

!
Find the integral x sin x dx.

Solution
The integrand is a product of two expressions, x and sin x. You
can integrate the second expression, sin x, and differentiating the first
expression, x, makes it simpler. So try integration by parts.
Let f (x) = x and g(x) = sin x.
Then f ! (x) = 1, and an antiderivative of g(x) is G(x) = − cos x.
So the integration by parts formula gives
! !
x sin x dx = f (x)g(x) dx
!
= f (x)G(x) − f ! (x)G(x) dx
!
= x × (− cos x) − 1 × (− cos x) dx
!
= −x cos x + cos x dx

Find the integral in this expression.

= −x cos x + sin x + c.

You can see that in Example 23 the use of the integration by parts formula
changed the problem of integrating x sin x into the problem of integrating
cos x, which is easier. Here are two similar examples for you to try.

Activity 34 Integrating by parts

Find the following integrals.

! !
(a) x cos x dx (b) xex dx

As you become more familiar with integration by parts, you’ll probably

find that you can carry it out more quickly and conveniently if you
remember the informal version of the formula stated below. You can recite
this version in your head as you apply integration by parts, in a similar
way to the informal versions of the product and quotient rules for
differentiation that you met in Unit 7. It’s useful for applying integration

178
 4 Integration by parts

by parts in fairly simple cases, like the ones that you’ve seen so far, which
is usually all that you need to do.

Integration by parts formula (informal)

, -
antiderivative
integral of product = (first) ×
of second
,, - , --
derivative antiderivative
− integral of × .
of first of second

In the next example, the integration in Example 23 is repeated, using the

informal version of the integration by parts formula.

Example 24 Integrating by parts, using the informal formula

!
Find the integral x sin x dx.

Solution
The integrand is a product of two expressions, x and sin x. You
can integrate the second expression, sin x, and differentiating the first
expression, x, makes it simpler. So try integration by parts.
Recite the informal version of the formula in your head. As you think
‘first’, write down the first expression, then as you think
‘antiderivative of second’, write down an antiderivative of the second
expression, and so on.
Integrating by parts gives
! !
x sin x dx = x × (− cos x) − 1 × (− cos x) dx
!
= −x cos x + cos x dx
= −x cos x + sin x + c.

You can practise using the informal version of the integration by parts
formula in the next activity. Alternatively, you might prefer to continue
using the formal version for a little longer, and move to the informal
version when you feel more confident with it. You can use the exercises on
integration by parts in the practice quiz and exercise booklet for this unit
for further practice.

179
Unit 8 Integration methods

Activity 35 Integrating by parts

Find the following integrals.

! ! !
(a) x sin(5x) dx (b) xe2x dx (c) xe−x dx

When you use integration by parts, remember that it matters which

expression in the integrand you take to be the ‘first expression’, and which
you take to be the ‘second expression’. If you’re trying to use integration
by parts, but you find that it leads to an integral that’s more complicated
than the original integral, then try swapping the two expressions in the
original integrand.

4.2 Integration by parts in practice

There are several frequently occurring types of integrals that often have
the right characteristics for integration by parts. In this subsection you’ll
meet three of these types. It’s helpful for you to learn to recognise them.

Integrands of the form xn g(x)

You can sometimes find an integral of the form
!
xn g(x) dx,

where n is a positive integer and g(x) is an expression that you can

integrate, by using integration by parts n times.
All the integrals in the previous subsection were of this form, with n = 1,
so you could find them by using integration by parts once. Here’s an
example where you have to use integration by parts twice.

Example 25 Integrating by parts twice

!
Find the integral x2 e4x dx.

Solution
The integral is of the form described above, so try integration by
parts.

180
 4 Integration by parts

Integrating by parts gives

! !
x2 e4x dx = x2 × 41 e4x − 2x × 14 e4x dx
!
= 41 x2 e4x − 12 xe4x dx

/
The integral in this expression, xe4x dx, looks easier to find than
the original integral, but it’s still not straightforward to find.
However, it is itself of the form described above, so try integration by
parts again.
, ! -
= 14 x2 e4x − 12 x × 14 e4x − 1 × 14 e4x dx
, ! -
1 2 4x 1 1 4x 1 4x
= 4 x e − 2 4 xe − 4 e dx

Multiply out the brackets. Take care with the minus sign in front
of the brackets – remember that it turns the minus sign inside the
brackets into a plus sign.
!
= 14 x2 e4x − 18 xe4x + 18 e4x dx

Now find the integral in this expression, and simplify the resulting
expression.

= 14 x2 e4x − 18 xe4x + 1
8 × 41 e4x + c

= 14 x2 e4x − 18 xe4x + 1 4x
32 e +c
1 4x 2
= 32 e (8x − 4x + 1) + c.

Activity 36 Integrating by parts twice

Find the following integrals.

! !
2 x
(a) x e dx (b) x2 cos(2x) dx

The method introduced above for dealing with integrands of the form
xn g(x) works just as well if the expression xn is replaced by any
polynomial
/ expression in x of degree n. For example, you can find the
integral (3x2 − 1)e4x dx by integrating by parts twice, and you can find
the integral in the next activity by integrating by parts once.

181
Unit 8 Integration methods

Activity 37 Integrating an expression of the form (ax + b)g(x)

!
Use integration by parts to find the integral (5x + 2)e−5x dx.

Integrands of the form xr ln x

Suppose that you want to integrate an expression of the form
xr ln x,
where r is a constant (any real number). Your first thought, particularly if
r is a positive integer, might be that this is similar to the expressions that
you saw how to integrate earlier in this subsection. However, you can’t
apply the integration by parts formula with ln x as the second expression,
because ln x isn’t an expression that you’ve seen how to integrate.
In fact, you can find an integral of the form above by applying the
integration by parts formula with ln x as the first expression. Here’s an
example.

Example 26 Integrating an expression of the form xn ln x

Find the integral
!
x2 ln x dx.

Solution
The integral is of the form discussed above, so swap the
expressions x2 and ln x, then integrate by parts.
! !
x ln x dx = (ln x) x2 dx
2

!
1 3 1 1 3
= (ln x) × 3 x − × x dx
x 3
!
= 31 x3 ln x − 31 x2 dx
& +
= 13 x3 ln x − 31 13 x3 + c
= 31 x3 ln x − 19 x3 + c
= 91 x3 (3 ln x − 1) + c

182
 4 Integration by parts

Activity 38 Integrating expressions of the form xr ln x

Find the following integrals.

! !
3
(a) x ln x dx (b) ln x dx

Hint for part (b): write ln x as (ln x) × 1.

You can use the ideas that you used in Activity 38 to deal with other,
similar integrals that involve the natural logarithm function. Try the
following activity.

Activity 39 Integrating another expression containing the function ln

!
Use integration by parts to find the integral (x − 2) ln(3x) dx.

Integrands of the form eax sin(bx) or eax cos(bx)

You can integrate an expression of the form
eax sin(bx) or eax cos(bx),
where a and b are nonzero numbers, by integrating by parts twice. When
you do this, you obtain an equation that expresses the original integral in
terms of itself, and you can rearrange this equation to find the original
integral. The method is demonstrated in the example below.
When you use this method, you can arrange the exponential expression
and the trigonometric expression in the integrand in either order before
you integrate
/ by parts for/ the first time. For example, you can start with
either eax sin(bx) dx or sin(bx)eax dx. However, you must choose the
same order (exponential before trigonometric, or trigonometric before
exponential) when you integrate by parts for the second time. If you don’t
do that, then the equation that you’ll obtain will reduce to 0 = 0, which
isn’t helpful!

183
Unit 8 Integration methods

Example 27 Finding an integral by expressing it in terms of itself

!
Find the indefinite integral ex cos(2x) dx.

Solution
The integral is of the form discussed above, so use integration by
parts twice. Start by swapping the expressions ex and cos(2x), as this
makes the working slightly easier (it avoids introducing fractions in
the early stages).
We use integration by parts twice.
! !
e cos(2x) dx = cos(2x) ex dx.
x

Integrate by parts.
! !
cos(2x) e dx = cos(2x)e − (−2 sin(2x))ex dx
x x

!
= cos(2x)ex + 2 sin(2x)ex dx

Integrate by parts again.

, ! -
x x x
= cos(2x)e + 2 sin(2x)e − (2 cos(2x))e dx
!
= cos(2x)ex + 2 sin(2x)ex − 4 cos(2x)ex dx

The final term here contains the original integral. Move this term
to the other side of the equation. When you do this, the right-hand
side no longer contains an indefinite integral, so you have to add an
arbitrary constant to that side.
So
! !
x
cos(2x)e dx + 4 cos(2x)ex dx = cos(2x)ex + 2 sin(2x)ex + c;

that is,
!
5 cos(2x)ex dx = cos(2x)ex + 2 sin(2x)ex + c.

To obtain the final answer, divide both sides by 5. (Remember

that you leave the arbitrary constant as c rather than c/5.) Also, it’s
slightly tidier to write the exponential expression before the
trigonometric expression in each term, and to take out a common
factor.

184
 4 Integration by parts

This gives
!
ex cos(2x) dx = 51 ex cos(2x) + 25 ex sin(2x) + c

= 15 ex (cos(2x) + 2 sin(2x)) + c.

Activity 40 Finding integrals by expressing them in terms of

themselves
Find the following integrals.
! !
(a) ex sin(3x) dx (b) e2x cos x dx

4.3 Integration by parts for definite integrals

Once you’ve used integration by parts to find an indefinite integral, you
can find any definite integral with the same integrand by using the
fundamental theorem of calculus in the usual way. Here’s an example.

Example 28 Evaluating a definite integral by parts (first method)

Evaluate the definite integral
! 1
xe7x dx.
0
Solution
Use integration by parts to obtain the indefinite integral of the
integrand.
Integration by parts gives
! !
xe dx = x × 7 e − 1 × 17 e7x dx
7x 1 7x

!
= 7 xe − 7 e7x dx
1 7x 1

= 17 xe7x − 1
7 × 17 e7x + c
1 7x
= 49 e (7x − 1) + c.

185
Unit 8 Integration methods

Apply the fundamental theorem of calculus.

It follows that
! 1 2 11
xe7x dx = 49 1 7x
e (7x − 1)
0 0
2 11
1
= 49 e7x (7x − 1)
0
& 7×1 +
= 49 e (7 × 1 − 1) − e7×0 (7 × 0 − 1)
1

1 7
= 49 (6e − (−1))
1 7
= 49 (6e + 1).

An alternative way to use integration by parts to evaluate a definite

integral is to apply the limits of integration ‘as you go along’. To do this,
you use the following formula, which is just the usual integration by parts
formula with the limits of integration included.

Integration by parts formula for definite integrals

! b ! b
) 'b
f (x)g(x) dx = f (x)G(x) a − f ! (x)G(x) dx.
a a
Here G is an antiderivative of g.

You can remember this formula as an informal version, by applying the

limits of integration a and b in the appropriate way to the informal version
of the usual integration by parts formula.
The next example demonstrates how to apply the limits of integration as
you go along when you integrate by parts. The definite integral in this
example is the same as the one in Example 28.

Example 29 Evaluating a definite integral by parts (second method)

Evaluate the definite integral
! 1
xe7x dx.
0
Solution
Integrate by parts, applying the limits of integration.

186
 4 Integration by parts

! 1 2 11 ! 1
xe7x dx = x × 17 e7x − 1 × 71 e7x dx
0 0 0
! 1
) '1
= 71 xe7x 0 − 17 e7x dx
0

Evaluate the first term, and do the integration in the second

term.
2 11
= 17 (1 × e7×1 − 0 × e7×0 ) − 17 17 e7x
0
) '1
= 17 e7 − 49
1
e7x 0
= 17 e7 − 1
49 (e
7×1
− e7×0 )
1 7 1 7
= 7 e − 49 (e − 1)
1 7 7
= 49 (7e − (e − 1))
1 7
= 49 (6e + 1)

Activity 41 Evaluating a definite integral by parts

Use the integration by parts formula for definite integrals (the method of
Example 29) to evaluate the definite integral
! 1
xe3x dx.
0
Give your answer to four decimal places.

In the next two activities you’re asked to use integration by parts to find
areas.

Activity 42 Using integration by parts to find an area

This question is about the function f (x) = x sin(3x).

(a) Explain why the graph of f lies above the x-axis for values of x in the
interval [π/12, π/6].
(b) Write down a definite integral that gives the area between the graph
of f and the x-axis from x = π/12 to x = π/6.
(c) Use the integration by parts formula for definite integrals to find the
area described in part (b). Give your answer to four decimal places.

187
Unit 8 Integration methods

Activity 43 Using integration by parts to find another area

This question is about the function f (x) = (x − 2)(e−x − 1).

(a) Explain why the graph of f lies above the x-axis for 0 < x < 2 and
below the x-axis for x > 2. You might find it convenient to use a table
of signs as part of your explanation.
(b) Find the total area (not the total signed area) between the graph of f
and the x-axis, from x = 0 to x = 4. Give your answer to four
significant figures.

As mentioned earlier, in some texts the integration by parts formula is

expressed in a different form.
Here’s the formula that you’ve seen in this section:
! !
f (x)g(x) dx = f (x)G(x) − f ! (x)G(x) dx.

It involves a function G and its derivative g.

If you denote this pair of functions by g and g ! instead, then you obtain
the alternative version below.

Integration by parts formula (Lagrange notation, alternative

version)
! !
f (x)g ! (x) dx = f (x)g(x) − f ! (x)g(x) dx

As with all calculus formulas, the integration by parts formula can be

stated in Leibniz notation.
If you put u = f (x) and v = g(x) in the formula in the box above, then
you obtain the following form of the formula.

Integration by parts formula (Leibniz notation)

! !
dv du
u dx = uv − v dx
dx dx

188
 5 More integration

5 More integration
You’ve now met several integration techniques, and practised using them
on a variety of integrals.
In this section, you’ll start by meeting one further approach, which can be
helpful for integrals that contain trigonometric expressions. Then you’ll
have an opportunity to practise identifying which of the integration
techniques that you’ve met are appropriate for which integrals, for a
variety of different types of integrals.
Finally, you’ll learn how to use a computer for integration.

5.1 Trigonometric integrals

In Section 3 you saw that you can integrate some expressions that contain
trigonometric functions by using integration by substitution. Another
approach that’s often helpful when you want to integrate an expression
that contains trigonometric functions is to start by using one of the
trigonometric identities that you met in Unit 4 to write the expression in a
different form. This can give you an expression that’s easier to integrate.
For example, you can use the identities below to help you integrate the
expressions sin2 x, cos2 x and sin x cos x, as illustrated in the next example.
The first two identities are the half-angle identities, and the third identity
is one of the double-angle identities.

Three trigonometric identities

sin2 θ = 12 (1 − cos(2θ))
cos2 θ = 12 (1 + cos(2θ))
sin(2θ) = 2 sin θ cos θ

189
Unit 8 Integration methods

Example 30 Integrating by using a trigonometric identity

Find the integral
!
sin2 x dx.

Solution
By the half-angle identity for sine,
! !
sin2 x dx = 1
2 (1 − cos(2x)) dx
!
1
= 2 (1 − cos(2x)) dx

= 12 (x − 1
2 sin(2x)) + c
1 1
= 2 x − 4 sin(2x) + c
1
= 4 (2x − sin(2x)) + c.

In the next activity you’re asked to use two of the trigonometric identities
stated near the beginning of this subsection to integrate another two
expressions that contain trigonometric functions.

Activity 44 Integrating by using trigonometric identities

!
(a) Use a half-angle identity to find the integral cos2 θ dθ.

(b) Use
! the double-angle identity in the box above to find the integral
sin x cos x dx.

The trigonometric integral in Example 30 and the one in Activity 44(a)

can’t easily be found without first using a trigonometric identity to write
them in a different form.
However, there’s an alternative way to find the integral in Activity 44(b).
Since the derivative of sin x is cos x, you can use integration by
substitution, taking u = sin x. You’re asked to do that in the next activity.

190
 5 More integration

Activity 45 Integrating a trigonometric expression by using

substitution
!
Use integration by substitution to find the integral sin x cos x dx.

In fact there are even two different ways to find the integral in
Activity 44(b), namely
!
sin x cos x dx,

by using integration by substitution. One way is as suggested above: you

use the fact that the derivative of sin x is cos x and hence make the
substitution u = sin x. Alternatively, you can use the fact that the
derivative of cos x is − sin x, write the integral as
!
− (cos x)(− sin x) dx,

and make the substitution u = cos x. The working for each of these
methods is given in the solution to Activity 45.
So you’ve now seen three different ways to integrate the expression
sin x cos x: one way by using a trigonometric identity, and two ways by
using substitution. These three different ways lead to three answers that
look different, as you can see in the solutions to Activities 44(b) and 45.
This illustrates that there can be more than one way to integrate an
expression, and different ways of integrating it can lead to answers that
look different, though of course these answers will be equivalent.
If an answer that you obtain for an integral is different from the answer
given in a book or by a computer, but you think that you haven’t made a
mistake, then you may be able to check that the two answers are equivalent
by using algebraic manipulation. You may need to use trigonometric
identities – there are examples of this in the solution to Activity 45.
A rougher, but sometimes quicker, check is to plot the graphs of the two
different answers on a computer (taking the arbitrary constants to be zero,
for example), and check that they seem to be vertical translations of each
other. This shows that the two different answers seem to be equivalent
once the arbitrary constants are taken into account. Again, there’s an
example of this in the solution to Activity 45.
In this subsection you’ve seen just a few examples of how you can use
trigonometric identities and integration by substitution to help you
integrate expressions that contain trigonometric functions. There are many
more possibilities, some of which are covered in the module Essential
mathematics 2 (MST125).

191
Unit 8 Integration methods

5.2 Choosing a method of integration

When you have an integral that you want to find, it may not be obvious
which method to try. Also, sometimes you might need to combine two or
more techniques – for example you might be able to use algebraic
manipulation, the constant multiple rule and the sum rule to express an
integral in terms of simpler integrals, then use integration by substitution
or integration by parts to find the simpler integrals.
Here are some things to think about when you have an integral that you
want to find.

Choosing a method for finding an integral

• Is it a standard integral? Consult the table in the Handbook.
• Can you rearrange the integral to express it as a sum of constant
multiples of simpler integrals?
• Is the integrand of the form f (ax) or f (ax + b), where a and b
are constants and f is a function that you can integrate? If so,
use the rule for integrating a function of a linear expression,
or use integration by substitution, with u = ax or u = ax + b as
appropriate.
• Can the integrand be written in the form
f (something) × the derivative of the something,
where f is a function that you can integrate? If so, use
integration by substitution. Start by setting the ‘something’
equal to u.
• Is the integrand of the form
f (x)g(x),
where f is a function that becomes simpler when differentiated,
and g is a function that you can integrate? (For example, f (x)
might be x or x2 or ln x.) If so, try integration by parts.
• If the integrand contains trigonometric functions, can you use
trigonometric identities to rewrite it in a form that’s easier to
integrate?

When you’re finding an indefinite integral, try to always remember to add

the arbitrary constant ‘ + c’. It’s easy to forget it!
In the next activity you can practise identifying which method of
integration to use.

192
 5 More integration

Activity 46 Choosing a method of integration

For each integral below, suggest a method of integration to use. Where you
suggest substitution, also suggest a suitable equation for u in terms of x.
You are not asked to actually find the integrals, but solutions are included
so you can do so if you wish.
! ! !
(a) x3 cos(9x4 ) dx (b) x cos(5x) dx (c) (x + cos(5x)) dx
! ! !
1 x2
(d) x(x2 + 3) dx (e) dx (f) dx
1 + 3x2 (7 − x3 )7
! ! !
−x/3
&1 + &1 +
(g) xe dx (h) sin 2 x cos 2 x dx (i) (x − 1)4 dx
! ! !
x x 2
(j) e (1 − 2e ) dx (k) x(1 + sin(x )) dx (l) x(3x2 − 2)8 dx

You’ll learn further useful strategies for integration in the module Essential
mathematics 2 (MST125), if it’s in your study programme. In particular,
you’ll learn more about integration by substitution and integration by
using trigonometric identities, and you’ll study partial fractions, which are
useful when you want to integrate rational functions.

193
Unit 8 Integration methods

5.3 Integration using a computer

There are many situations in which it’s helpful to use a computer for
integration, such as when you want to integrate an expression that’s
difficult to integrate by hand. In the first activity of this subsection you
can learn how to use the module computer algebra system to find
indefinite integrals and evaluate definite integrals.

Activity 47 Using a computer for integration

Work through Section 9 of the Computer algebra guide.

You can use the skills that you learned in Activity 47 in the next activity.

Activity 48 Using a computer to work with a complicated function

This activity is about the function

x2 − 10x + 15
f (x) = √ .
1+ x+1
Do parts (b) to (d) below by using the module computer algebra system.
Give your answers to parts (c) and (d) to three significant figures.
(a) What is the domain of f ?
(b) Plot the graph of f .
(c) Find the two x-intercepts of f .
(d) Find the area between the graph of f and the x-axis, between the two
x-intercepts.

The next two activities are the final ones of the unit. In them, you’re
asked to use your knowledge of integration to solve practical problems.
The definite integrals involved in the problems can’t easily be evaluated by
using the fundamental
√ theorem of calculus, as it’s difficult to find
3 2
antiderivatives of x + 1 and sin(t /150). However, you can use a
computer to find approximate values.

194
 5 More integration

Activity 49 Using integration to solve a practical problem

Consider the plastic edging shown below. The shape of its cross-section is
given by the equation
"
y = 13 x3 + 1 (−1 ≤ x ≤ 1.5),
where x and y are measured in centimetres.
Calculate the volume of plastic that is required to make one metre of the
edging. Give your answer in cubic centimetres to three significant figures.

Activity 50 Using integration to solve another practical problem

Suppose that an object moves in a straight line, and its velocity v

(in m s−1 ) at time t (in seconds) is given by the equation
, 2 -
t
v = sin (0 ≤ t ≤ 20),
150
as shown in Figure 42.
If the displacement of the object from a particular reference point on the
line at time t = 0 is 6 m, as shown below, what is the displacement of the
object from this reference point at time t = 10? Give your answer to the
nearest centimetre.

Figure 42 The graph of

v = sin(x2 /150) (0 ≤ t ≤ 20)

195
Unit 8 Integration methods

You’ve now come to the end of the introduction to basic calculus in this
module. You can use the skills that you’ve acquired to solve many different
types of problems in mathematics, engineering and science, for example.
These skills also equip you to study many further topics in these areas.
You’ll make a start on that in Unit 11, in which you’ll use calculus in your
study of Taylor polynomials. If your study programme includes further
mathematics modules, then in some of these you may study more
advanced calculus, with even more powerful applications.

Learning outcomes
After studying this unit, you should be able to:
• understand what is meant by a definite integral
• understand how to use subintervals to find an approximate value for a
definite integral
• use integral notation
• understand the fundamental theorem of calculus
• evaluate some definite integrals by using the fundamental theorem of
calculus
• integrate by substitution
• integrate by parts
• appreciate how to use trigonometric identities and integration by
substitution to help you integrate some trigonometric expressions
• identify a suitable method of integration for some types of integrals
• use a computer for integration.

196
 Solutions to activities

Solutions to activities
Solution to Activity 1 So the heights of the rectangles are:
If we divide the interval [−18, 18] into six f (−3) = 3 − (−3)2 = −6
subintervals of equal width, then each subinterval f (−2) = 3 − (−2)2 = −1
has width 36/6 = 6.
f (−1) = 3 − (−1)2 = 2
The left endpoints of the six subintervals are
f (0) = 3 − 02 = 3
− 18, −18 + 6, −18 + 2 × 6,
f (1) = 3 − 12 = 2
− 18 + 3 × 6, −18 + 4 × 6, −18 + 5 × 6
f (2) = 3 − 22 = −1.
that is
So we obtain the following approximate value for
−18, −12, −6, 0, 6, 12. the signed area:
So the heights of the rectangles are: (−6 × 1) + (−1 × 1) + (2 × 1)
1
f (−18) = 9 − 36 × (−18)2 = 0 + (3 × 1) + (2 × 1) + (−1 × 1) = −1.
1 2
f (−12) = 9 − 36 × (−12) =5
Solution to Activity 4
f (−6) = 9 − 36 1
× (−6)2 = 8
The exact value of the signed area between the
1
f (0) = 9 − 36 × 02 = 9 graph of the function f (x) = 3 − x2 and the x-axis
1
f (6) = 9 − 36 × 62 = 8 from x = −3 to x = 3 is 0.
f (12) = 9 − 361
× 122 = 5.
Solution to Activity 5
So we obtain the following approximate value for
the area: (a) The signed area from x = 2 to x = 6 is 4.0.
(0 × 6) + (5 × 6) + (8 × 6) (b) The signed area from x = 2 to x = 6 is 4.0,
+ (9 × 6) + (8 × 6) + (5 × 6) = 210. so the signed area from x = 6 to x = 2 is −4.0.
(c) The signed area from x = 2 to x = 10 is
Solution to Activity 2 4.0 − 5.4 = −1.4.
(a) The signed area from x = 1 to x = 5 is −6.3. (d) From part (c), the signed area from x = 2 to
(b) The signed area from x = 5 to x = 7 is 1.3. x = 10 is −1.4, so the signed area from x = 10
to x = 2 is 1.4.
(c) The signed area from x = 1 to x = 7 is
−6.3 + 1.3 = −5. (e) The signed area from x = 8 to x = 8 is 0.
(d) The signed area from x = 1 to x = 9 is (f) The signed area from x = −3 to x = 2 is
−6.3 + 1.3 − 5.9 = −10.9. 4.1 − 2.4 = 1.7, so the signed area from x = 2 to
x = −3 is −1.7.
Solution to Activity 3 (g) The signed area from x = −1 to x = 10 is
If we divide the interval [−3, 3] into six subintervals −2.4 + 4.0 − 5.4 = −3.8, so the signed area
of equal width, then each subinterval has width from x = 10 to x = −1 is 3.8.
6/6 = 1.
The left endpoints of the six subintervals are
−3, −2, −1, 0, 1, 2.

197
Unit 8 Integration methods

Solution to Activity 6 Solution to Activity 8

! −3 )1 '
2 5 1 1
(a) 2x 3 = × 52 − × 32 = 12 (25 − 9) = 8.
(a) f (x) dx = −5. 2 2
−5 ) '2π
! (b) cos x 0 = cos(2π) − cos 0 = 1 − 1 = 0.
2
(b) f (x) dx = 6. ) '1
(c) ex −1 = e1 − e−1 = e − (1/e) = 2.35 (to 3 s.f.).
−3
! 7
(c) f (x) dx = −11. Solution to Activity 9
!
2 (a) An antiderivative of x2 is 31 x3 , so
2 ! 1
(d) f (x) dx = −5 + 6 = 1. ) '1
−5
x2 dx = 13 x3 −1
! −1
7
(e) f (x) dx = 6 − 11 = −5. = 1
3 × 13 − 1
3 × (−1)3
2
−3 = 3.
! 7
(f) f (x) dx = −5 + 6 − 11 = −10. (b) An antiderivative of et is et , so
! 2
−5 ) '2
! 5 et dt = et 0
(g) f (x) dx = 0. 0
5 = e2 − e0
! −3 ! 2
= e2 − 1 = 6.39 (to 3 s.f.).
(h) f (x) dx = − f (x) dx = −6.
2 −3 (c) An antiderivative of 1/u (for u > 0) is ln u, so
! 2 ! 7 ! 4
1 ) '4
(i) f (x) dx = − f (x) dx = −(−11) = 11. du = ln u 1
7 2 1 u
! −5 ! −3 = ln 4 − ln 1
(j) f (x) dx = − f (x) dx = −(−5) = 5.
−3 −5
= ln 4 = 1.39 (to 3 s.f.).
! −3 ! 7 (d) An antiderivative of sec2 θ is tan θ, so
(k) f (x) dx = − f (x) dx ! π/4
) 'π/4
7 −3
sec2 θ dθ = tan θ −π/4
= −(−5) = 5, −π/4
π . π#
by part (e). = tan − tan −
! −5 ! 7 4 4
(l) f (x) dx = − f (x) dx = 1 − (−1)
7 −5 = 2.
= −(−10) = 10,
(e) An antiderivative of 1/(1 + u2 ) is tan−1 u, so
by part (f). ! 1
1 ) '1
du = tan−1 u −1
Solution to Activity 7 −1 1 + u
2
! 9 ! 7 ! 9
= tan−1 (1) − tan−1 (−1)
f (x) dx = f (x) dx + f (x) dx π . π#
2 2 7 = − −
so 4 4
! 9
= π/2
−15 = −11 + f (x) dx
7 = 1.57 (to 3 s.f.).
and hence
! 9
f (x) dx = −15 − (−11) = −4.
7

198
 Solutions to activities

! 1
Solution to Activity 10
! ! (b) (u − 2u2 ) du
4 √ 4
−1/2
(a) 3 x dx = 3x1/2 dx 2 11
1 2
1
(
1
04 = − 2 × 13 u3
2u
−1/2
1 3/2 2 11
= 3× x
3/2 1 = 21 u2 − 32 u3
2 14 −1/2
= 2x3/2 2 11 2 11
1 = 12 u2 − 32 u3
−1/2 −1/2
= 2 × 4 − 2 × 13/2
3/2
2 11 2 11
=2×8−2 = 12 u2 − 32 u3
−1/2 −1/2
= 14. & 2 + 2& 3 +
! = 2 1 − (− 2 ) − 3 1 − (− 21 )3
1 1 2
0
(b) x(1 + x) dx = 12 (1 − 14 ) − 23 (1 + 18 )
−1 1 3 2 9
! 0 = 2 × 4 − 3 × 8
= (x + x2 ) dx = 3
8 − 3
4
−1
)1 2 1 3
'0 = − 83 .
= 2x + 3x −1
= ( 12 ×0 + 2 1
× 03 ) − ( 12 (−1)2 + 13 (−1)3 ) Solution to Activity 12
3 √ √
! 7 ! 7
= −( 21 − 31 ) 1 3
(a) √ 2x dx = 1
2 √ x3 dx
= − 61 . 3 3
2 1√7
! 4 ! 4, - 1 1 4
r+5 5 = 2 4x √
(c) dr = dr 1+ 3
2 r 2 r ) '√
) '4 1 4 √ 7
= 8 x
= r + 5 ln r 2 .√ 3
√ #
= (4 + 5 ln 4) − (2 + 5 ln 2) = 1
8 ( 7)4 − ( 3)4
= 2 + 5(ln 4 − ln 2) = 18 (72 − 32 )
= 2 + 5 ln(4/2) = 18 (49 − 9)
= 2 + 5 ln 2. = 5.
! π/4
Solution to Activity 11 (b) (sin x + cos x) dx
! 5 ( 05 −π/4
1 5/2 ! !
(a) x3/2 dx = x π/4 π/4
3 5/2 3 = sin x dx + cos x dx
2 15 −π/4 −π/4
2 5/2
= 5x ) 'π/4 ) 'π/4
3 = − cos x −π/4 + sin x −π/4
2 15
= 25 x5/2 ) 'π/4 ) 'π/4
3 = − cos x −π/4 + sin x −π/4
= 2 5/2
− 35/2 ) . .π# . π ##
5 (5
= − cos − cos −
= 16.1 (to 3 s.f.). . π #4 . π #4
+ sin − sin −
, 4 - 4 , -
1 1 1 1
=− √ −√ + √ − −√
2 2 2 2
2
=√
2
√
= 2.
199
Unit 8 Integration methods

Solution to Activity 13 Hence the shaded area is 21 333 cm2 , to the

The required area is given by nearest cm2 .
! 18 . #
1 2
9 − 36 x dx Solution to Activity 15
−18
2 118 (a) (i) From t = 0 to t = 1 the displacement of
1 1 3 the object changes by 2.3 cm.
= 9x − 36 × 3x
−18
2 118 (ii) From t = 3 to t = 4 the displacement of
1 3
= 9x − 108 x the object changes by −1.8 cm. That is, it
−18
)'18 2 118 changes by 1.8 cm in the negative direction.
= 9x −18 − 108 1
x3
−18 (iii) From t = 0 to t = 3 the displacement of
) '18 1
) 3 '18 the object changes by
= 9 x −18 − 108 x −18
3 (2.3 + 5.9 + 2.7) cm = 10.9 cm.
= 9(18 − (−18)) − 1
108 (18 − (−18)3 )
1 3 3 (iv) From t = 3 to t = 6 the displacement of
= 9 × 36 − 108 (18 − (−18 )) the object changes by
1 3
= 9 × 36 − 108 (2 × 18 ) (−1.8 − 2.0 − 0.3) cm = −4.1 cm.
= 324 − 108 That is, it changes by 4.1 cm in the
= 216. negative direction.
So the area of the cross-section of the roof is 216 m2 . (b) (i) The displacement of the object from its
starting position at time t = 2 is
Solution to Activity 14
(2.3 + 5.9) cm = 8.2 cm.
(a) The x-intercepts are given by
1 2 (ii) By (a) parts (iii) and (iv), the
32 x − 200 = 0 displacement of the object from its
1 2
32 x = 200 starting position at time t = 6 is
2
x = 6400 (10.9 − 4.1) cm = 6.8 cm.
√
x = ± 6400 (c) By parts (a)(iii) and (a)(iv), the object travels
x = ±80. 10.9 cm in the positive direction, then 4.1 cm in
So the x-intercepts are −80 and 80. the negative direction, so the total distance that
it travels is
(b) The shaded area is
! 80 . # (10.9 + 4.1) cm = 15 cm.
1 2
x − 200 dx
32
−80 Solution to Activity 16
2 180
= 32 1
× 13 x3 − 200x (a) The change in the displacement of the object
−80 from time t = 0 to time t = 1 is given by
2 180 ! 1 2 11
1 3
= 96 x − 200x (3 − 23 t)dt = 3t − 32 × 12 t2
−80 0
1
) 3 '80 ) '80 0
2 11
= 96 x −80 − 200 x −80
= 3t − 43 t2
1 3 3
= 96 (80 − (−80) ) − 200(80 − (−80)) 0
) '1 3 ) 2 '1
= 1 3 =3 t 0−4 t 0
96 (2 × 80 ) − 200 × 160
80 3 = 3(1 − 0) − 43 (12 − 02 )
= − 32 000 3
48 =3− 4
64 000 = 49 = 2.25.
=−
3
That is, the object travels 2.25 m in that time
= −21 333 31 . (in the positive direction).

200
 Solutions to activities

(b) The change in the displacement of the object Solution to Activity 19

from time t = 4 to time t = 5 is given by !
! 5 2 15 (a) ex dx = ex + c.
(3 − 23 t)dt = 3t − 23 × 12 t2 ! !
4 4
2 15 (b) (x + 3)(x − 2) dx = (x2 + x − 6) dx
= 3t − 34 t2
4
) '5 3 ) 2 '5 = 31 x3 + 12 x2 − 6x + c.
=3 t 4−4 t 4 ! !
1
= 3(5 − 4) − 43 (52 − 42 ) (c) dx = x−3 dx
x3
=3− 3
×9 1 −2
4 = x +c
= − 15 −2
4
1
= −3.75. = − 2 + c.
2x
That is, the object travels 3.75 m in that time !
(in the negative direction). (d) (sin θ − cos θ) dθ = − cos θ − sin θ + c.
!
Solution to Activity 17 1
(e) dx = tan−1 x + c.
The change in the displacement of the car during 1 + x2
!
those four seconds is given by −3
! 4 2 14 (f) du = −3 tan−1 u + c.
1 + u2
(28 − 7t) dt = 28t − 27 t2 !
0 0 1
) '4 ) '4 (g) dr = 12 tan−1 r + c.
= 28 t 0 − 27 t2 0 2(1 + r2 )
!
= 28(4 − 0) − 72 (42 − 0) (1 + x2 )
(h) dx = 12 (x + 13 x3 ) + c
= 112 − 56 2
= 56. = 16 (3x + x3 ) + c.
That is, the car travels 56 m during those four
seconds.
Solution to Activity 20
(In fact, since the required answer is the area of the ! , - ! !
shaded triangle in Figure 39, you can calculate it in 3 1 3 1
(a) − dx = dx − dx
the following more straightforward way. x 1 + x2 x 1 + x2
! !
The change in the displacement of the car during 1 1
=3 dx − dx
the four seconds is x 1 + x2
( 21 × 4 × 28) m = 56 m.) = 3 ln |x| − tan−1 x + c.
!
Solution to Activity 18 (b) (cosec x)(cosec x + cot x) dx
The change in the displacement of the car is the !
area shaded in Figure 40, which is given by = (cosec2 x + cosec x cot x) dx
1 1 1 ! !
2 × 5 × 15 − 2 × 4 × 12 = 2 (75 − 48) 2
= cosec x dx + cosec x cot x dx
= 12 × 27 = 13.5.
So the car travels 13.5 m during the fifth second. = − cot x − cosec x + c.
(There are various ways to work out the area
shaded in Figure 40. It is worked out above by
subtracting the area of one triangle from the area of
another, but alternatively you could add the areas
of a rectangle and a triangle, or use the formula for
the area of a trapezium.)
201
Unit 8 Integration methods

Solution to Activity 21 du
(f) Let u = sin x; then = cos x. So
du ! , - dx ! , -
(a) Let u = cos x; then = − sin x. So 1 1
! dx ! cos x dx = du
1 + sin2 x 1 + u2
ecos x (− sin x) dx = eu du
= tan−1 u + c
= eu + c = tan−1 (sin x) + c.
= ecos x + c.
du Solution to Activity 22
3
(b) Let u = x ; then = 3x2 . So du
! dx ! (a) Let u = 4 + cos x; then = − sin x. So
& 3
+ 2 ! dx !
sin(x ) (3x ) dx = sin(u) du
(4 + cos x) (− sin x) dx = u7 du
7
= − cos u + c
= − cos(x3 ) + c. = 81 u8 + c
du = 18 (4 + cos x)8 + c.
(c) Let u = sin x; then = cos x. So
! , - dx ! du
1 1 (b) Let u = 1 + x2 ; then = 2x. So
cos x dx = du ! " dx !
sin x u √
1 + x2 (2x) dx = u du
= ln |u| + c
!
= ln | sin x| + c. = u1/2 du
du
(d) Let u = cos x; then = − sin x. So 1 3/2
! , - dx ! = u +c
1 1 3/2
(− sin x) dx = du
cos x
2 u2 = 23 (1 + x2 )3/2 + c.
!
du
= u−2 du (c) Let u = x5 − 8; then = 5x4 . So
! dx !
1 −1
=u +c (x − 8) (5x ) dx = u10 du
5 10 4
−1
1 = 1 11
=− +c 11 u +c
u 5
1 = 1
11 (x − 8)11 + c.
=− +c du
cos x (d) Let u = ex + 5; then = ex . So
= − sec x + c. ! , - dx!
1 1
(A quicker way to find this indefinite integral, x
e dx = du
x
e +5 u
without using substitution, is to write the
integrand as − sec x tan x and use the table of = ln |u| + c
standard indefinite integrals.) = ln |ex + 5| + c
du = ln(ex + 5) + c.
(e) Let u = sin x; then = cos x. So
! dx! du
(e) Let u = 5 + 2x3 ; then = 6x2 . So
sin4 x cos x dx = u4 du ! dx !
& 3
+ 2
sin(5 + 2x ) (6x ) dx = sin u du
= 15 u5 + c
= 1
sin5 x + c. = − cos u + c
5
= − cos(5 + 2x3 ) + c.

202
 Solutions to activities

Solution to Activity 23 du
(b) Let u = 1 + x2 ; then = 2x. So
du ! " !dx." #
(a) Let u = ex ; then = ex . So
! dx ! x 1 + x2 dx = 1 + x2 x dx
x x
(cos(e )) e dx = cos u du ! ." #
= 21 1 + x2 × 2x dx
= sin u + c !
√
= sin(ex ) + c. = 12 u du
du !
(b) Let u = 1 + sin x; then = cos x. So
! dx
! = 12 u1/2 du
e1+sin x (cos x) dx = eu du = 1 2 3/2
2 × 3u +c
= eu + c = 1 3/2
3u +c
= e1+sin x + c. = 1 2 3/2
3 (1 + x ) + c.
du du
(c) Let u = x10 + 6; then = 10x9 . So (c) Let u = cos x; then = − sin x. So
! , - dx ! dx !
1 1 !
9
10
(10x ) dx = du cos4 x sin x dx = − (cos4 x)(− sin x) dx
x +6 u
= ln |u| + c !
= − u4 du
= ln |x10 + 6| + c
= ln(x10 + 6) + c. = − 15 u5 + c
du = − 51 cos5 x + c.
(d) Let u = cos x; then = − sin x. So
! dx !
Solution to Activity 25
(cos3 x)(− sin x) dx = u3 du
du
1 4 (a) Let u = 1 + 2x2 ; then = 4x. So
= 4u + c ! ! ,dx -
= 1 4 x 1
4 cos x + c. dx = x dx
1 + 2x2 1 + 2x2
! , -
Solution to Activity 24 1 1
=4 (4x) dx
du 1 + 2x2
(a) Let u = x3 ; then = 3x2 . So !
! dx ! 1
& + = 41 du
2 3
x cos(x ) dx = cos(x3 ) x2 dx u
! = 14 ln |u| + c
= 3 (cos(x3 ))(3x2 ) dx
1
= 1
4 ln |1 + 2x2 | + c
! = 1
ln(1 + 2x2 ) + c
1 4
= 3 cos u du (since 1 + 2x2 is always positive).
1
= 3 sin u + c
= 1
3 sin(x3 ) + c.

203
Unit 8 Integration methods

du du
(b) Let u = 2 + ex ; then = ex . So (c) Let u = 1 + 12 sin x; then = 21 cos x. So
! dx
! , - ! dx
ex 1
dx = ex dx (1 + 21 sin x)2 cos x dx
(2 + ex )2 (2 + ex )2 !
!
1 = 2 (1 + 21 sin x)2 ( 12 cos x) dx
= du
u2 !
!
= u−2 du = 2 u2 du

= −u−1 + c = 2 × 31 u3 + c
1 = 23 u3 + c
=− +c
u
1 = 23 (1 + 12 sin x)3 + c.
=− + c.
2 + ex Solution to Activity 27
du du
(c) Let u = 3 − sin x; then = − cos x. So Let u = cos x; then = − sin x. So
! dx
! , - dx
cos x 1 ! !
dx = − (− cos x) dx sin x
3 − sin x 3 − sin x tan x dx = dx
! cos x
! , -
1 1
=− du = (sin x) dx
u cos x
= − ln |u| + c ! , -
1
= − ln |3 − sin x| + c =− (− sin x) dx
cos x
!
= − ln(3 − sin x) + c 1
=− du
(since 3 − sin x is always positive). u
= − ln |u| + c
Solution to Activity 26 = − ln | cos x| + c.
du
(a) Let u = cos x; then = − sin x. So
! dx ! Solution to Activity 28
ecos x sin x dx = − ecos x (− sin x) dx du
(a) Let u = 6x − 1; then = 6. So
! ! ! dx
=− eu du e6x−1 dx = 1
6 e6x−1 × 6 dx
!
= −eu + c
= 1
eu du
= −ecos x + c. 6

du = 16 eu + c
(b) Let u = 3 − ex ; then = −ex . So
! dx
! , - = 61 e6x−1 + c.
ex 1
dx = − (−ex ) dx du
3 − ex 3 − ex (b) Let u = 4x; then = 4. So
! ! dx!
1
=− du sin(4x) dx = 41 (sin(4x)) × 4 dx
u
= − ln |u| + c !
1
= − ln |3 − ex | + c. = 4 sin u du

= 14 (− cos u) + c
= − 41 cos u + c
= − 41 cos(4x) + c.

204
 Solutions to activities

du du
(c) Let u = −9x; then = −9. So (g) Let u = 2x + 1; then = 2. So
! dx
! ! dx
! , -
1 1
e−9x dx = − 19 e−9x × (−9) dx dx = 1
2 × 2 dx
! (2x + 1)2 (2x + 1)2
!
= − 19 eu du 1 1
=2 du
u2
!
= − 91 eu + c
=2 1
u−2 du
= − 19 e−9x + c.
du = − 12 u−1 + c
(d) Let u = −x/3; then = − 31 . So 1 1
! !dx =− × +c
2 u
e−x/3 dx = −3 e−x/3 × (− 31 ) dx 1
! =− + c.
2(2x + 1)
u
= −3 e du
du
(h) Let u = 1 − x; then = −1. So
= −3eu + c ! dx
! , -
= −3e−x/3 + c. 1 1
dx = − × (−1) dx
du (1 − x)3 (1 − x)3
(e) Let u = 3 − 7x; then = −7. So !
dx 1
! ! =− 3
du
u
1
cos(3 − 7x) dx = − 7 (cos(3 − 7x)) × (−7) dx !
! = − u−3 du
= − 17 cos u du
= −(− 21 u−2 ) + c
= − 71 sin u + c 1
= 2 +c
1
= − 7 sin(3 − 7x) + c. 2u
1
du = + c.
(f) Let u = 4 − x; then = −1. So 2(1 − x)2
! ! dx
, - du
1 1 (i) Let u = 6x; then = 6. So
dx = − × (−1) dx dx !
4−x 4−x !
!
1 sec2 (6x) dx = 16 (sec2 (6x)) × 6 dx
=− du !
u
= − ln |u| + c = 6 sec2 u du
1

= − ln |4 − x| + c. = 1
tan u + c
6
1
= 6 tan(6x) + c.

205
Unit 8 Integration methods

!
Solution to Activity 29 (c) e3t+4 dt = 13 e3t+4 + c.
du !
(a) Let u = 2x; then = 2. So
! dx ! (d) cos( 12 θ) dθ = 2 sin( 21 θ) + c.
1 1
√ dx = " dx !
1 − 4x2 1 − (2x)2
! (e) e−4p dp = − 41 e−4p + c.
1 1
=2 " × 2 dx !
1 − (2x)2 1
! (f) dx = 17 ln |7x − 4| + c.
1 7x − 4
= 12 √ du !
1 − u2 1
(g) dr = − ln |1 − r| + c.
= 21 sin−1 u + c 1−r
!
= sin−1 (2x) + c.
1
2 (h) sec2 (5x − 1) dx = 15 tan(5x − 1) + c.
√ du √ !
(b) Let u = 2 x; then = 2. So
! dx ! (i) sin( 15 x) dx = −5 cos( 15 x) + c.
1 1 1 √
dx = √ &√ + × 2 dx !
1 + 2x2 2
2 1+ 2x (j) e−x dx = −e−x + c.
!
1 1 !
=√ du
2 1 + u2 (k) ex/2 dx = 2ex/2 + c.
1 !
= √ tan−1 u + c
2 (l) sec2 (2 − 3p) dp = − 13 tan(2 − 3p) + c
1 .√ #
= √ tan−1 2 x + c. = 31 tan(3p − 2) + c.
2
du (The final expression was obtained by using the
(c) Let u = 32 x; then = 23 . So trigonometric identity tan(−θ) = − tan θ.)
! dx ! ! , - !
1 1 1 2−x & +
dx = 4 dx (m) sin dx = sin 32 − 13 x dx
4 + 9x2 1 + 94 x2 3
! & & ++
1 = −3 − cos 32 − 13 x + c
= 14 & 3 +2 dx , -
1 + 2x 2−x
! = 3 cos + c.
1 2 1 3 3
=4×3 & 3 +2 × 2 dx
1 + 2x Solution to Activity 31
!
1
= 61 3
& 3 +2 × 2 dx (a) The velocity v (in m s−1 ) of the object at time t
1 + 2x (in seconds) is given by
! & +
1 1 v = 2 cos 21 t + 5 .
=6 du
1 + u2 Hence the displacement s (in m) of the object
= 61 tan−1 u + c at time t (in seconds) is given by
& + !
= 61 tan−1 23 x + c. & +
s = 2 cos 21 t + 5 dt
Solution to Activity 30 & +
! = 2 × 2 sin 21 t + 5 + c
& +
(a) 1
cos(10x) dx = 10 sin(10x) + c. = 4 sin 12 t + 5 + c,
! where c is an arbitrary constant.
(b) sin(2 − 5x) dx = − 15 (− cos(2 − 5x)) + c
1
= 5 cos(2 − 5x) + c.

206
 Solutions to activities

When t = 0, s = 4. Hence du
& + (b) Let u = cos x; then = − sin x. Putting
4 = 4 sin 12 × 0 + 5 + c dx
x = π/2 gives u = 0 and putting x = π gives
4 = 4 sin(5) + c u = −1. So
! π ! π
c = 4 − 4 sin(5) 3
cos x sin x dx = − cos3 x(− sin x) dx
c = 7.835 . . . π/2 π/2
! −1
c = 7.84 (to 2 d.p.).
=− u3 du
So the equation for the displacement of the 0
object in terms of time is 2 1−1
& + = − 14 u4
s = 4 sin 12 t + 5 + 7.84. 0
1
) 4 '−1
(b) When t = 10, = −4 u 0
& + & +
s = 4 sin 12 × 10 + 5 + 7.835 . . . = − 14 (−1)4 − 04
= 4 sin(10) + 7.835 . . . = − 41
= 5.7 (to 2 s.f.). = −0.25.
So the displacement of the object at time (Notice that in the second line of the
10 seconds is 5.7 m (to 2 s.f.). manipulation above we obtain an integral,
/ −1 3
0
u , in which the lower limit of integration is
Solution to Activity 32 greater than the upper limit of integration. An
du alternative way to proceed from this line is to
(a) Let u = 1 + 2x2 ; then = 4x.
dx use the fact that, in general,
/a /b
Putting x = 0 gives u = 1 and putting x = 1 b
f (x) dx = − a f (x) dx. Then the
gives u = 3. So manipulation continues like this:
! 1 ! 1, - ! −1 ! 0
x 1
2
1
dx = 4 (4x) dx u3 du = u3 du
0 1 + 2x 0 1 + 2x2 0 −1
! 3 2 10
1 = 1 4
4u
1
=4 du
1 u
−1
) '3 1
) '
4 0
= 41 ln |u| 1 = 4 u −1
& 4 +
= 14 (ln 3 − ln 1) = 1
4 0 − (−1)4
= 14 ln 3 = − 41
= 0.275 (to 3 s.f.). = −0.25.)
du
(c) Let u = −3x2 ; then = −6x.
dx
Putting x = −1 gives u = −3 and putting x = 0
gives u = 0. So
! 0 ! 0. #
−3x2 2
xe dx = − 61
e−3x (−6x) dx
−1 −1
!0
= − 16 eu du
−3
) '0
= − 61 eu −3
& +
= − 61 e0 − e−3
& +
= 61 e−3 − 1
= −0.158 (to 3 s.f.).

207
Unit 8 Integration methods

Solution to Activity 33 Hence

! !
du x
xe dx = f (x)g(x) dx
(a) Let u = 1 + x; then = 1. Also, x = u − 1. So
! !dx !
√ √
x 1 + x dx = (u − 1) u du = f (x)G(x) − f " (x)G(x) dx
! . # !
= u3/2 − u1/2 du = xe − ex dx
x

u5/2 u3/2 = xex − ex + c

= − +c = ex (x − 1) + c.
5/2 3/2
= 25 (1 + x)5/2 − 23 (1 + x)3/2 + c. Solution to Activity 35
!
du & +
(b) Let u = x + 3; then = 1. Also, x = u − 3, so (a) x sin(5x) dx = x − 15 cos(5x)
dx
x −! 2 = u − 5. Hence !
! & +
x−2 u−5 − 1 × − 51 cos(5x) dx
dx = du
(x + 3)3 u3 !
! 1 1
= − x cos(5x) + cos(5x) dx
= (u−2 − 5u−3 ) du 5 5

= − 51 x cos(5x) + 51 × 51 sin(5x) + c
= −u−1 + 25 u−2 + c
1 5 = − 15 x cos(5x) + 25 1
sin(5x) + c
=− + 2 +c = 251
sin(5x) − 15 x cos(5x) + c.
u 2u
1 5 ! !
=− + + c. 2x
& 1 2x + & +
x + 3 2(x + 3)2 (b) xe dx = x 2 e − 1 × 12 e2x dx
!
1 2x
= 2 xe − 2 e2x dx
1
Solution to Activity 34
(a) Let f (x) = x and g(x) = cos x. Then f " (x) = 1 = 12 xe2x − 1 1 2x
2 × 2e +c
and an antiderivative of g(x) is G(x) = sin x. = 1
2 xe
2x
− 1 2x
4e +c
Hence
! ! 1 2x
= 4 e (2x −1) + c.
x cos x dx = f (x)g(x) dx ! !
& + & +
! (c) xe−x dx = x −e−x − 1 × −e−x dx
= f (x)G(x) − f " (x)G(x) dx !
! = −xe−x + e−x dx
= x sin x − sin x dx
= −xe−x + (−e−x ) + c
= x sin x − (− cos x) + c = −xe−x − e−x + c
= x sin x + cos x + c. = −e−x (x + 1) + c.
(b) Let f (x) = x and g(x) = ex . Then f " (x) = 1
and an antiderivative of g(x) is G(x) = ex .

208
 Solutions to activities

Solution to Activity 36 Solution to Activity 38

! ! ! !
2 x 2 x x 3
(a) x e dx = x e − 2xe dx (a) x ln x dx = (ln x)x3 dx
! !
1 1 4
= x2 ex − 2 xex dx = (ln x) × 1 4
4x − × x dx
x 4
, ! - !
2 x x x
= x e − 2 xe − 1 × e dx = 14 x4 ln x − 1
4 x3 dx
!
= 14 x4 ln x − 1
× 14 x4 + c
= x e − 2xe + 2 ex dx
2 x x 4
1 4
= 16 x (4 ln x − 1) + c.
2 x x
= x e − 2xe + 2e + c x ! !
= ex (x2 − 2x + 2) + c. (b) ln x dx = (ln x) × 1 dx
! !
1
(b) x2 cos(2x) dx = (ln x)x − × x dx
x
! !
= x2 ( 21
sin(2x)) − 2x( 12 sin(2x)) dx = x ln x − 1 dx
!
1 2 = x ln x − x + c
= 2 x sin(2x) − x sin(2x) dx
= x(ln x − 1) + c.
= 12 x2 sin(2x)
, ! - Solution to Activity 39
1 1 !
− x(− 2 cos(2x)) − 1 × (− 2 cos(2x)) dx
(x − 2) ln(3x) dx
!
!
= 12 x2 sin(2x) + 12 x cos(2x) − 12 cos(2x) dx
= (ln(3x))(x − 2) dx
= 12 x2 sin(2x) + 12 x cos(2x) − 21 × 12 sin(2x) + c !
1 2 1 1 2
1 2 1
= 4 (2x − 1) sin(2x) + 2 x cos(2x) + c. = ln(3x) × ( 2 x − 2x) − ( x − 2x) dx
x 2
!
1
Solution to Activity 37 = 2 x(x − 4) ln(3x) − ( 12 x − 2) dx
!
!
(5x + 2)e−5x dx 1 1
= 2 x(x − 4) ln(3x) − 2 (x − 4) dx
, - ! , -
1 1
= (5x + 2) e−5x − 5 e−5x dx = 21 x(x − 4) ln(3x) − 12 ( 12 x2 − 4x) + c
−5 −5
! = 21 x(x − 4) ln(3x) − 14 x(x − 8) + c.
−5x −5x
= − 51 (5x + 2)e + e dx

= − 51 (5x + 2)e−5x − 1 −5x

−5 e dx +c
= − 15 e−5x (5x + 2 + 1) + c
= − 15 (5x + 3)e−5x + c.
(An alternative way to find this integral is to start
by writing it as
! !
−5x
5 xe dx + 2 e−5x dx.
The first integral in this expression has an integrand
of the form xg(x), where g(x) is an expression that
you can integrate, so you can find this integral by
using integration by parts. The second integral is
straightforward to find.)

209
Unit 8 Integration methods

Solution to Activity 40 and

! hence
(a) We write e2x cos x dx = 15 e2x sin x + 52 e2x cos x + c.
! !
e sin(3x) dx = sin(3x) ex dx.
x
That is,
!
Integrating by parts twice gives e2x cos x dx = 15 e2x (sin x + 2 cos x) + c.
!
sin(3x) ex dx
! Solution to Activity 41
! 1 2 11 ! 1
= sin(3x)e − (3 cos(3x))ex dx
x
xe 3x 1 3x
dx = x × 3 e − 1 × 13 e3x dx
! 0 0 0
!
= sin(3x)e − 3 cos(3x)ex dx
x
1
) 3x '1 1 1 3x
= 3 xe 0 − 3 e dx
0
= sin(3x)ex 2 11
, ! -
x x
= 13 (e3 − 0) − 13 13 e3x
− 3 cos(3x)e − (−3 sin(3x))e dx 0
) '1
! = 31 e3 − 19 e3x 0
& +
= sin(3x)ex − 3 cos(3x)ex − 9 sin(3x)ex dx. = 13 e3 − 19 e3 − 1
So ! = 91 (3e3 − e3 + 1)
10 sin(3x) ex dx = 19 (2e3 + 1)
= 4.5746 (to 4 d.p.).
= sin(3x)ex − 3 cos(3x)ex + c
and! hence Solution to Activity 42
x (a) If x ∈ [π/12, π/6], then 3x ∈ [π/4, π/2].
sin(3x) e dx
Now
= 101
sin(3x)ex − 3
10 cos(3x)ex + c.
sin θ > 0 when θ ∈ [π/4, π/2],
That is,
! so
ex sin(3x) dx sin(3x) > 0 when x ∈ [π/12, π/6].
1 x 3 x Also
= 10 e sin(3x) − 10 e cos(3x) + c
= 1 x x > 0 when x ∈ [π/12, π/6].
10 e (sin(3x) − 3 cos(3x)) + c.
(b) Integrating by parts twice gives Hence
! x sin(3x) > 0 when x ∈ [π/12, π/6].
e2x cos x dx
That is, the graph of f (x) = x sin(3x) lies above
! the x-axis for values of x in the interval
= e sin x − (2e2x ) sin x dx
2x
[π/12, π/6].
!
(b) It follows from part (a) that the area between
= e2x sin x − 2 e2x sin x dx
this graph and the x-axis from x = π/12 to
= e2x x = π/6 is
, sin x ! - ! π/6
2x 2x
− 2 e (− cos x) − (2e )(− cos x) dx x sin(3x) dx.
! π/12

= e2x sin x + 2e2x cos x − 4 e2x cos x dx.

So !
5 e2x cos x dx = e2x sin x + 2e2x cos x + c

210
 Solutions to activities

(c) Integrating by parts gives The function y = e−x is decreasing on its whole
! π/6 domain, so the function y = e−x − 1 is also
x sin(3x) dx decreasing on its whole domain. The x-intercept
π/12
! of its graph is given by e−x − 1 = 0, which gives
2 1π/6 π/6
1
= x(− 3 cos(3x)) − (− 13 cos(3x)) dx e−x = 1
π/12 π/12 − x = ln 1
! π/6
) 'π/6 −x=0
= − 31 x cos(3x) π/12 + 13 cos(3x) dx
π/12 x = 0.
. .π# . π ##
1 π π So its x-intercept is x = 0.
= −3 cos − cos
6 2 12 4
2 1π/6 So we have the following table of signs, for
+ 31 13 sin(3x) x > 0.
π/12
.π π 1 # ) ' x (0, 2) 2 (2, ∞)
π/6
= − 31 ×0− × √ + 19 sin(3x) π/12
6 12 2 x−2 − 0 +
π . .π# . π ## e−x − 1 − − −
= √ + 19 sin − sin
36 2 2 4 (x − 2)(e−x − 1) + 0 −
π . 1 #
= √ + 91 1 − √ The table shows that the graph of the function
36 2 2
= 0.0943 (to 4 d.p.). f (x) = (x − 2)(e−x − 1) lies above the x-axis for
0 < x < 2 and below the x-axis for x > 2.
(The area found in this solution is shown below.)
(b) Because the graph of f lies above the x-axis for
0 < x < 2 and below the x-axis for x > 2, to
find the total area between the graph of f and
the x-axis from x = 0 to x = 4 we have to work
out the signed area from x = 0 to x = 2 and the
signed area from x = 2 to x = 4 separately.
The signed area between the graph of f and the
x-axis from x = 0 to x = 2 is given by
! 2
(x − 2)(e−x − 1) dx
0
2 12 ! 2
−x
= (x − 2)(−e − x) − (−e−x − x) dx
0 0
! 2
= 0 − (−2)(−e−0 ) + (e−x + x) dx
0
2 12
−x 1 2
= 2(−1) + −e + 2 x
0
−2 −0
= −2 + (−e + 2) − (−e + 0)
Solution to Activity 43 = −2 − e−2 + 2 + 1
(a) To help us construct a table of signs for the = 1 − e−2 .
function f (x) = (x − 2)(e−x − 1), first we Since the graph of f lies above the x-axis for
consider the two factors of the expression 0 < x < 2, it follows that the area between the
(x − 2)(e−x − 1) separately. graph of f and the x-axis from x = 0 to x = 2 is
The function y = x − 2 is increasing on its 1 − e−2 .
whole domain, and the x-intercept of its graph
is x = 2.

211
Unit 8 Integration methods

! !
Similarly, the signed area between the graph (b) sin x cos x dx = 1
sin(2x) dx
2
of f and the x-axis from x = 2 to x = 4 is given
by = 1
× 21 (− cos(2x)) + c
! 4 2

(x − 2)(e−x − 1) dx = − 14 cos(2x) + c.
2
2 14 ! 4 Solution to Activity 45
= (x − 2)(−e−x − x) − (−e−x − x) dx du
2 2 Let u = sin x; then = cos x. So
! 4 ! dx
!
= 2(−e−4 − 4) − 0 + (e−x + x) dx sin x cos x dx = u du
2
2 14
= −2e −4
− 8 + −e−x + 12 x2 = 12 u2 + c
2
= −2e−4 − 8 + (−e−4 + 8) − (−e−2 + 2) = 12 sin2 x + c.
= −3e−4 + e−2 − 2. (There’s an alternative approach to finding this
integral, as follows. Let u = cos x; then
Since the graph of f lies below the x-axis for du
2 < x < 4, it follows that the area between the = − sin x. So
dx! !
graph of f and the x-axis from x = 2 to x = 4 is
−(−3e−4 + e−2 − 2) = 3e−4 − e−2 + 2. sin x cos x dx = cos x sin x dx
!
So the total area between the graph of f and = − (cos x)(− sin x) dx
the x-axis from x = 0 to x = 4 is !
(1 − e−2 ) + (3e−4 − e−2 + 2) = − u du
= 3 − 2e−2 + 3e−4
= − 21 u2 + c
= 2.784 (to 4 s.f.).
= − 21 cos2 x + c.
(The area found in this solution is shown below.)
This approach gives a different answer, but of
course it’s equivalent to the first answer. You can
confirm this by using the identity
sin2 θ + cos2 θ = 1, as follows:
− 21 cos2 x + c = − 12 (1 − sin2 x) + c
= 1
2 sin2 x − 1
2 +c
1 2
= 2 sin x + d,
where d = − 21 + c is an arbitrary constant.
Notice also that the solution to Activity 44(b) gives
a third
! different answer:
sin x cos x dx = − 14 cos(2x) + c.

Solution to Activity 44 You can confirm that this answer is equivalent to

! ! the first answer above by using the identity
(a) cos2 θ dθ = 1
2 (1 + cos(2θ)) dθ cos(2θ) = 1 − 2 sin2 θ, as follows:
! − 41 cos(2x) + c = − 14 (1 − 2 sin2 x) + c
1
= (1 + cos(2θ)) dθ
2 = 1
2 sin2 x − 1
4 +c
1 1 1 2
= 2 (θ + 2 sin(2θ)) + c = 2 sin x + d,
1
= 4 (2θ + sin(2θ)) + c. where d = − 41 + c is an arbitrary constant.)

212
 Solutions to activities

(The diagram below shows the graphs of the three second expression is a simple function of a
different antiderivatives of f (x) = sin x cos x that linear expression.
!
were found in the solutions to Activity 44(b) and
Activity 45. The arbitrary constant has been taken (x + cos(5x)) dx = 12 x2 + 51 sin(5x) + c.
to be zero in each case. You can see that, as you’d
(d) You could use integration by substitution
expect, the three graphs appear to be vertical
(taking u = x2 + 3) or integration by parts, but
translations of each other.)
the simplest method is to start by multiplying
out! the brackets. !
x(x2 + 3) dx = (x3 + 3x) dx

= 14 x4 + 32 x2 + c
= 41 x2 (x2 + 6) + c.
1
(e) The integrand can be written as &√ +2 ,
1+ 3x
√
so it’s a function of the linear expression
√ 3 x.
So you can use the substitution u = 3 x,
1
together with the indefinite integral of ,
1 + x2
which is a standard integral.
√ √
Let u = 3 x; then du/dx = 3. So
! ! √
1 1 1
dx = √ &√ + × 3 dx
Solution to Activity 46 1 + 3x2 3 1+ 3x
2
!
(a) Since the derivative of 9x4 is 36x3 , which is 1 1
‘nearly’ x3 , you can use the substitution =√ du
3 1 + u2
u = 9x4 . 1
du = √ tan−1 u + c
Let u = 9x4 ; then = 36x3 . So 3
! dx ! 1 .√ #
= √ tan−1 3 x + c.
x3 cos(9x4 ) dx = 1
36 cos(9x4 ) × 36x3 dx 3
! (f) Since the derivative of 7 − x3 is −3x2 , which is
1
= 36 cos u du ‘nearly’ the numerator x2 , you can use the
1 substitution u = 7 − x3 .
= 36 sin u + c
du
= 1
sin(9x4 ) + c. Let u = 7 − x3 ; then = −3x2 . So
36
! dx !
(b) The integrand is of the form xg(x), where g is a x2 1
function that you can integrate, which suggests dx = − 1
3 (−3x2 ) dx
(7 − x3 )7 (7 − x3 )7
that you should try integration by parts. To !
1 1
integrate the ‘second’ expression, you can use = −3 du
u7
the fact that it is a simple function of a linear !
expression. = − 13 u−7 du
! !
1 1
x cos(5x) dx = 5 x sin(5x) − 5 sin(5x) dx 1 1 −6
=− × u +c
& + 3 −6
= 51 x sin(5x) − 15 − 15 cos(5x) + c 1
= +c
= 15 x sin(5x) + 1
25 cos(5x) + c. 18u6
(c) The integrand is the sum of two expressions 1
= + c.
that are straightforward to integrate. The 18(7 − x3 )6

213
Unit 8 Integration methods

(g) The integrand is of the form xg(x), where g is a du

Let u = 1 − 2ex ; then = −2ex . So
function that you can integrate, which suggests ! dx !
that you should try integration by parts. To
ex (1 − 2ex ) dx = − 12 (1 − 2ex )(−2ex ) dx
integrate the ‘second’ expression, you can use
!
the fact that it’s a simple function of a linear
expression. = − 21 u du
!
xe−x/3 dx = − 12 × 21 u2 + c
. # ! . # = − 41 u2 + c
−x/3
= x −3e − 1 × −3e−x/3 dx = − 41 (1 − 2ex )2 + c
!
−x/3 = − 14 (1 − 4ex + 4(ex )2 ) + c
= −3xe + 3 e−x/3 dx
= − 14 (1 − 4ex + 4e2x ) + c
= −3xe−x/3 − 9e−x/3 + c = ex − e2x − 1
+c
4
= −3e−x/3 (x + 3) + c. x x 2
= e − (e ) − 1
+c
4
(h) You can use the trigonometric identity x x
= e (1 − e ) − + c 1
4
sin(2θ) = 2 sin θ cos θ.
! ! = ex (1 − ex ) + d,
&1 + &1 + 1
sin 2 x cos 2 x dx = 2 sin x dx where d = − 14 + c is an arbitrary constant.)
! (k) You can start by multiplying out the brackets
= 21 sin x dx and applying the sum rule. Then you have two
integrals to find. One of them is
= − 21 cos x + c.
straightforward, and you can find the other by
(i) You can use the fact that the integrand is a using integration by substitution, taking u = x2 .
!
simple function of a linear expression, or you
can use the substitution u = x − 1. x(1 + sin(x2 )) dx
! !
(x − 1)5
(x − 1)4 dx = + c. = (x + x sin(x2 )) dx
5
! !
(j) You could use integration by substitution
(taking u = 1 − 2ex ), but the simplest method = x dx + x sin(x2 ) dx
is to start by multiplying out the brackets. !
! ! = 2 x + x sin(x2 ) dx
1 2
e (1 − 2e ) dx = (ex − 2(ex )2 ) dx
x x

!
Let u = x2 ; then du/dx = 2x. The expression
= (ex − 2e2x ) dx
above becomes
!
= ex − 2 × 12 e2x + c 1 2
(sin(x2 )) × 2x dx
1
2x + 2
= ex − e2x + c !
1 2 1
= ex − (ex )2 + c = 2 x + 2 sin u du
= ex (1 − ex ) + c. = 12 x2 + 21 (− cos u) + c
(Here’s how the integration goes if you use
= 12 x2 − 1
2 cos(x2 ) + c
integration by substitution.
= 21 (x2 − cos(x2 )) + c.

214
 Solutions to activities

(l) Since the derivative of 3x2 − 2 is 6x, which is Solution to Activity 49

‘nearly’ x, you can use the substitution The area of the cross-section, in square centimetres,
u = 3x2 − 2. is
du ! 1.5 "
Let u = 3x2 − 2; then = 6x. So 1
x3 + 1 dx.
! dx! −1
3

2 8
x(3x − 2) dx = 1 2 8
(3x − 2) × 6x dx A computer gives the value of this definite integral
6
! as 0.939 (to 3 s.f.). So the volume of plastic
= 1
u8 du required to make one metre of the edging is
6
approximately 93.9 cm3 .
1 1 9
= 6 × 9u (Details of how to use the CAS to find the indefinite
= 1 2
2)9 + c. integral are given in the Computer algebra guide, in
54 (3x −
the section ‘Computer methods for CAS activities
Solution to Activity 48 in Books A–D’.)
(a) The domain of f is the interval
√ [−1, ∞) Solution to Activity 50
(because the expression x + 1 is defined only
when x is in this interval). The change in the displacement of the object from
time t = 0 to time t = 10 is
(b) The graph of f is shown below. ! 10 , 2 -
t
sin dt.
0 150
A computer gives the value of this definite integral
as 2.15 (to 2 d.p.).
So the displacement of the object at time t = 10 is
(6 + 2.15) m = 8.15 m, to the nearest centimetre.
(Details of how to use the CAS to find the indefinite
integral are given in the Computer algebra guide, in
the section ‘Computer methods for CAS activities
in Books A–D’.)

(c) The x-intercepts of f are 1.84 and 8.16 (to

3 s.f.).
√
(The exact values are 5 ± 10.)
(d) The area between the graph of f and the
x-axis, between the two x-intercepts, is
! 8.162... 2
x − 10x + 15
− √ dx.
1.837... 1+ x+1
A computer gives the value of this expression as
12.4 (to 3 s.f.).
(Details of how to use the CAS for this activity
are given in the Computer algebra guide, in the
section ‘Computer methods for CAS activities
in Books A–D’.)

215
Acknowledgements

Acknowledgements
Grateful acknowledgement is made to the following source:

Page 193: Mark Hobbs, for the idea for the cartoon

216
Unit 9

Matrices
 Introduction

Introduction
This unit introduces mathematical objects called matrices (the singular
form is matrix ). At their simplest, matrices are arrays of numbers used to
hold information in an orderly way. In particular, they provide a useful
way of storing data that can be organised in rows and columns. Everyday
examples of such data include transport timetables and tables holding
survey results.
In this unit you will discover that arrays of numbers can be thought of as
‘multi-dimensional numbers’. For example, you can add, subtract and
multiply them, just as you can add, subtract and multiply numbers.
You’ve met a similar idea before: in Unit 5, you saw that column vectors
are numerical entities made up of more than one number, and they can be
added, subtracted and multiplied by scalars, for example.
One advantage of arranging large amounts of numerical data in matrix
form is precisely that a block of data can be treated as a single entity.
Many problems that involve large amounts of numerical data are first
reduced to matrix problems before being solved with a computer. So
matrices are an important topic in mathematics, and find widespread use
in mathematical modelling, as well as in computer graphics (where they
are used to move and rotate objects), medical imaging, economic models,
electrical networks, data encryption and many other applications. A magic square (an example
of an array of numbers) on
As a simple practical example, consider a publisher that supplies its books the facade of the Sagrada
to a number of bookshops. The publisher’s sales department might collect Familia basilica in Barcelona
the weekly order quantities for three of its bestselling books in a grid such
as the one in Table 1.
Table 1 A publisher’s order quantities for three books in a certain week
Bestseller 1 Bestseller 2 Bestseller 3
Bookshop 1 10 25 12
Bookshop 2 5 10 5
Bookshop 3 7 5 0
Bookshop 4 10 8 10

The information in Table 1 can be stored more succinctly by omitting the

row and column headings and enclosing the data in brackets, if it is
understood that each row corresponds to a bookshop and each column
corresponds to a book:
 
10 25 12
 5 10 5 
 
 7 5 0 .
10 8 10
This is an example of a matrix. Information about order quantities can be
collected every week for the life of the three books and stored in this form.

219
 For example, the information for two consecutive weeks can be recorded in
two separate matrices:
   
10 25 12 15 30 20
 5 10 5  10 20 10
   .
 7 5 0  , 20 8 5
10 8 10 15 0 10
The order quantities for a fortnight can now be calculated by adding
corresponding numbers in the two arrays, which gives
   
10 + 15 25 + 30 12 + 20 25 55 32
 5 + 10 10 + 20 5 + 10   
  , that is, 15 30 15 .
 7 + 20 5+8 0+5   27 13 5 
10 + 15 8 + 0 10 + 10 25 8 20
This calculation amounts to adding the two matrices, in a sense that will
be made precise in Section 1.
You might wonder what the advantage is in carrying out calculations in
this way. After all, the total number of orders for each bookshop can be
worked out in a straightforward way without resorting to matrices. In a
real-life situation, however, a large publisher will supply hundreds of
bookshops with hundreds of books. Calculating order quantities and total
charges requires careful accounting, and it is precisely in this sort of
situation that matrices become a useful way to store data (especially
electronically) and carry out calculations with them.

Matrices were first introduced in Western mathematics in their

current form by the British mathematician Arthur Cayley
(1821–1895). However, they appeared as early as 300–200 bc in the
Chinese text The nine chapters on the mathematical art, where
matrix methods are used to solve simultaneous equations. You’ll meet
a modern version of these methods in Section 4.

There are other useful operations that can be carried out when data are
stored in matrix form, such as the multiplication of matrices by single
numbers, and the multiplication of two matrices. You’ll learn about these
operations in Section 1, and see an application of matrix multiplication in
Section 2. In Section 3 you’ll learn about matrix inverses, and in Section 4
Pages from The nine chapters you’ll meet a connection between matrices and systems of linear equations.
on the mathematical art

220
 1 Matrices and matrix operations

1 Matrices and matrix operations

In this section you’ll meet matrices and standard matrix notation, and
learn about matrix addition, multiplication of matrices by single numbers
(known as scalar multiplication of matrices), matrix multiplication and
matrix powers.

1.1 Matrix notation and terminology

A matrix (pronounced ‘may-tricks’) is a rectangular array of numbers,
usually enclosed in brackets. Here are some examples:
1 
. / . / 4
1 2 −1 1 2 3
, ,  7 −6 .
5 3 0 3 1
22 0
Some texts use square brackets for matrices, like this:
) 2
1 2
.
3 1
In this module we’ll always use round brackets.
We often use capital letters to denote matrices. For example, we might
write
1 
. / . / 4
1 2 −1 1 2 3 First rule of consulting: never,
A= , B= , C =  7 −6 . never call this a “table”.
5 3 0 3 1
22 0 We can bill four times as much
In the text in this module, the capital letters that denote matrices are in when we call it a “matrix”.
bold font, but you don’t need to do anything special when you handwrite
them. In particular, you don’t need to underline them, unlike letters that
denote vectors.
The numbers in a matrix are called its elements (or its entries, in some
texts). For example, matrices A and C above each have six elements, and
matrix B has four elements. In this module, the elements of matrices are
always real numbers.
A horizontal line of numbers in a matrix is called a row, and a vertical line
of numbers $is called a -column. For instance, $in the matrix
- A above the
first row is. 1/ 2 −1 and the second.row / is 5 3 0 . Likewise, the . first/
1 2 −1
column is , the second column is and the third column is .
5 3 0
The matrix A has two rows and three columns: we say that it is a 2 × 3
matrix. In general, a matrix with m rows and n columns is described as an
m × n matrix, or a matrix of size m × n. The first number in this
notation is always the number of rows, and the second number is always
the number of columns. The size m × n of a matrix is read as ‘m by n’.
A matrix with the same number of rows as columns is called a square
matrix. Thus, for example, matrix B above is square.

221
Unit 9 Matrices

The column notation for vectors that you met in Unit 5 is a particular
instance of matrix notation: two-dimensional column vectors are 2 × 1
matrices, and three-dimensional column vectors are 3 × 1 matrices. In
general, the word vector is used to mean any matrix with a single column,
even if it has more than 3 elements and hence has no geometric
interpretation in the plane or three-dimensional space. For instance, the
following matrices are vectors:
√ 
  2  
1 −3 w
3  1  x
  ,   and   , where w, x, y, z are variables.
5  2  y 
 0 
7 z
7
The elements of a vector are often called its components, as in Unit 5.
A vector with n components is called an n-dimensional vector.
There is a useful notation for the elements of a matrix in terms of their
row and column positions. The element in row i and column j of a
matrix A is denoted by aij . Similarly, the element in row i and column j
of a matrix B is denoted by bij , and so on.
For instance, if
0 1
1
3 4
A= ,
7 −6

then a11 = 31 , a12 = 4, a21 = 7 and a22 = −6.

This notation is used for matrices of any size. Thus, for instance, a general
3 × 4 matrix can be denoted by
 
a11 a12 a13 a14
A = a21 a22 a23 a24  .
a31 a32 a33 a34

Activity 1 Using matrix notation

Write down the elements a33 , a34 , a43 , a14 , a44 and a42 of the matrix
 
2 8 0 −1
4 7 5 −2
A= 9 −1
.
3 −4
6 −5 −7 −9
What is the size of A?

Two matrices are equal if they have the same numbers of rows and
columns, and all corresponding elements are equal. For example,
. / . /
1 2 1 2
= ,
3 4 3 4

222
 1 Matrices and matrix operations

but
 
. / 1 2 . / . /
1 2 1 2 1 3
'= 3 4 and '= .
3 4 3 4 2 4
5 6

You can carry out operations on matrices in a similar way to numbers. For
example, you can add, subtract and multiply them, as you’ll see in this
section. There’s also a way to manipulate matrices that in certain
circumstances plays the role of division, which you’ll learn about in
Section 3. While matrix addition is similar to addition of vectors, which
you met in Unit 5, matrix multiplication is more surprising. We’ll start by
looking at matrix addition.

Arthur Cayley gave the rules for matrix operations in a paper

presented to the Royal Society in 1858. The paper also contained the
conditions under which a matrix has an inverse (you’ll meet matrix
inverses in Section 3). However, it was Cayley’s friend and fellow
mathematician James Joseph Sylvester who coined the term ‘matrix’,
in 1850, and he also contributed to the theory of matrices.
Before being appointed to a mathematics professorship at the
University of Cambridge, Cayley worked for 14 years as a lawyer in
London. At the time, Sylvester was employed as an actuary. The two
worked together at the courts of Lincoln’s Inn, which gave them the
opportunity to discuss mathematics.
Sylvester had a more varied mathematical career than Cayley,
including professorships at University College London, Johns Hopkins
University in the USA, and the University of Oxford. It is said that
he tutored Florence Nightingale in mathematics, though firm evidence Arthur Cayley
for this has yet to be found. (1821–1895)

1.2 Matrix addition and subtraction

You can add two matrices only if they have the same size. The sum of two
matrices is obtained by adding their corresponding elements. For example:
       
−2 4 3 1 −2 + 3 4+1 1 5
 1 5 + −2 10 = 1 + (−2) 5 + 10  = −1 15 .
0 −3 −6 4 0 + (−6) −3 + 4 −6 1

The rule for adding two matrices can be expressed formally as follows.

James Joseph Sylvester

(1814–1897)

223
Unit 9 Matrices

Matrix addition
If A and B are m × n matrices, then A + B is the m × n matrix
whose element in row i and column j is aij + bij .
Addition of two matrices of different sizes is not defined.

For instance, you cannot add a 3 × 3 matrix and a 3 × 5 matrix.

The rule for adding matrices extends to the sum of any number of matrices
of the same size: you just add corresponding elements.

Activity 2 Adding matrices

For each of the pairs of matrices below, work out the matrix sum A + B.
. / . /
2 1 4 6
(a) A = , B=
3 0 2 −1
. / . /
2 0 −4 1 0 1
(b) A = , B=
−1 6 0 0 1 1
. / . /
a b 3 −1
(c) A = , B=
c d −1 2

In Unit 5, you saw that vector addition shares many properties with the
addition of numbers. This happens for matrix addition, too. For example,
you can see from the definition of matrix addition that
A + B = B + A,
for all matrices A and B of the same size. As you saw with vectors, this
property is expressed by saying that matrix addition is commutative.
You can also see from the definition of matrix addition that
(A + B) + C = A + (B + C),
for all matrices A, B and C of the same size. In other words, when you
add three matrices A, B and C of the same size, it doesn’t matter whether
you add A and B and then add C to the result, or add B and C and then
add A to the result, as you get the same overall result either way. This
property is expressed by saying that matrix addition is associative. It
tells you that you can write the expression
A+B+C
without ambiguity: it doesn’t matter whether it means (A + B) + C or
A + (B + C), as both are equal.
There are other respects in which matrix operations behave like number
operations. You will discover more similarities between matrices and
numbers as you meet other matrix operations.

224
 1 Matrices and matrix operations

You can also subtract matrices of the same size by subtracting their
corresponding elements. For example,
. / . / . /
4 5 3 1 3 4 4−1 5−3 3−4
− =
7 −6 0 3 9 1 7 − 3 −6 − 9 0 − 1
. /
3 2 −1
= .
4 −15 −1

Here is the formal definition of matrix subtraction.

Matrix subtraction
If A and B are m × n matrices, then A − B is the m × n matrix
whose element in row i and column j is aij − bij .
Subtraction of two matrices of different sizes is not defined.

Like the subtraction of numbers, matrix subtraction is neither

commutative nor associative. For example, it is not true that
A − B = B − A for all matrices A and B of the same size, as illustrated
by parts (a) and (b) of the next activity.

Activity 3 Subtracting matrices

. / 0 1 . /
1
2 −2 2 4 0 −x
Let A = ,B= and C = . Work out the
1 0 −1 2 x 0
following matrices.
(a) A − B (b) B − A (c) B − C

Since vectors are single-column matrices, they are added and subtracted
by adding or subtracting corresponding components, as in the examples
below.
       
1 7 1+7 8
 −3   −2  −3 + (−2) −5
       
 4  +  1  =  4 + 1  =  5
 1  3  1 3   
       2
2 2 2 + 2
0 −1 0 + (−1) −1
       
10 −2 10 − (−2) 12
 3  3  3 − 3   0
 − =   
 1  2  1 − 2  = −1
−1 1 −1 − 1 −2
In other words, the ideas of addition and subtraction of column vectors
that you met in Unit 5 extend in the obvious way to vectors with more
than three components.

225
Unit 9 Matrices

A matrix each of whose elements is zero is called a zero matrix, and is

usually denoted by 0. For example, the 2 × 2 zero matrix is
. /
0 0
0= .
0 0
Also, as you’d expect, if A is any matrix, then the matrix formed by
changing each element of A to its negative is called the negative of A,
and denoted by −A. For example,
. / . /
0 1 0 −1
− = .
−2 −4 2 4
It follows from the definitions of matrix addition and subtraction that for
any matrices A and B of the same size,
A − B = A + (−B),
as you’d expect.

1.3 Scalar multiplication of matrices

Any matrix can be added to itself, and a matrix sum A + A can be
written as 2A. Each element in 2A is twice the corresponding element
in A, as in the following example:
     
1 0 5 1 0 5 2 0 10
−1 −2 3 + −1 −2 3 = −2 −4 6  .
1 1
3 2 1 3 2 1 6 1 2
Similarly, adding A to itself n times, for any natural number n, gives a
matrix in which each element is the corresponding element in A multiplied
by n.
This idea can be generalised to any real number k, so kA is the matrix
obtained by multiplying each element of A by k. For example,
. / 01 1
1 0
3
1
1 4 3 2 ×4 2 ×3 2 2
= 1 = .
2 0 −1 1
2 × 0 2 × (−1) 0 − 21

The matrix kA is called the scalar multiple of the matrix A by the real
number k, and the operation of forming kA is called scalar
multiplication.

Scalar multiplication
If A is an m × n matrix and k is any real number, then kA is the
matrix whose element in row i and column j is kaij .

The number k in a scalar multiple kA of a matrix A is sometimes referred

to as the scalar k in this context, to emphasise that it’s a number rather
than a matrix.

226
 1 Matrices and matrix operations

Activity 4 Multiplying matrices by scalars

(a) Calculate the following products of scalars and matrices, simplifying

your answers.
. / . / . /
4 3 1 4 6 2 6 x
(i) 3 (ii) (iii)
2 1 2 2 −1 3 3 y
(b) Simplify each of the following matrices by writing it as a scalar
multiple of a matrix with integer elements.
. / 031 . /
4.5 3 2 2x 3x
(i) (ii) 2 (iii)
2 1.5 0 5x
3

Notice that it follows from the definition of scalar multiplication that, for
any matrix A,
(−1)A = −A,
as you’d expect.
In the next activity you can practise the matrix operations that you’ve
learned so far.

Activity 5 Combining matrix addition, subtraction and scalar

multiplication
0 1 0 1 0 1
1 1 3
2 1 2 2 −1 2 0 −1
Let A = ,B= and C = .
4 −2 0 −2 2 −1 3 5
Where possible, evaluate A − C, B + 2A and C − 2A.

Here’s a summary of some useful properties of matrix addition and scalar

multiplication of matrices. As you’d expect (since vectors are single-column
matrices), these properties are the same as the properties of vector algebra
that you met in Unit 5, but with vectors replaced by matrices.

227
Unit 9 Matrices

Properties of matrix addition and scalar multiplication

The following properties hold for all matrices A, B and C for which
the sums mentioned are defined, and for all scalars m and n.
1. A+B=B+A
2. (A + B) + C = A + (B + C)
3. A+0=A
4. A + (−A) = 0
5. m(A + B) = mA + mB
6. (m + n)A = mA + nA
7. m(nA) = (mn)A
8. 1A = A
(In properties 3 and 4, 0 is the zero matrix of the same size as A.)

1.4 Matrix multiplication

In Subsection 1.3 you learned how to multiply a number and a matrix.
Under certain conditions, it is also possible to multiply two matrices. The
way that this is done is less obvious than matrix addition or scalar
multiplication.
Recall the example in the introduction to this unit, where a matrix was
used to store the numbers of copies of three bestselling books that four
bookshops ordered from a publisher in a certain week:
Bestseller 1 Bestseller 2 Bestseller 3
 
Bookshop 1 10 25 12
Bookshop 2 5 10 5 
.
Bookshop 3 7 5 0 
Bookshop 4 10 8 10
Now suppose that the publisher charges the bookshops £4 per copy for
Bestseller 1, £7 per copy for Bestseller 2 and £9 per copy for Bestseller 3.
According to the data in the matrix, Bookshop 1 ordered 10 copies of
Bestseller 1, 25 copies of Bestseller 2 and 12 copies of Bestseller 3.
Therefore the total amount, in £, that Bookshop 1 owes the publisher for
that week’s order is
10 × 4 + 25 × 7 + 12 × 9 = 323.
The expression on the left of this equation is obtained by multiplying each
element in the first row of the matrix by the price of the corresponding
book, and then adding the results.

228
 1 Matrices and matrix operations

If we write the prices (in £) for the three bestsellers as a three-dimensional

vector, that is, as a 3 × 1 matrix, then we can picture this procedure as
shown in Figure 1.

10 25 12 4
5 10 5 7 gives 10 × 4 + 25 × 7 + 12 × 9
7 5 0
10 8 10 9

Figure 1 The procedure for calculating the amount owed by Bookshop 1

If we want to calculate the amount owed by Bookshop 2, then we can carry
out the same procedure with the second row of the matrix, and likewise for
Bookshops 3 and 4 with the third and fourth rows, respectively. We can
write the four amounts that we obtain as a four-dimensional vector. The
procedure described above combines the two matrices
 
10 25 12  
 5 10 5  4
  and 7
7 5 0
9
10 8 10
to give
   
10 × 4 + 25 × 7 + 12 × 9 323
 5 × 4 + 10 × 7 + 5 × 9  135
   
 7 × 4 + 5 × 7 + 0 × 9  =  63  .
10 × 4 + 8 × 7 + 10 × 9 186
This calculation shows that the amounts owed by the four bookshops are
£323, £135, £63 and £186, respectively.
This procedure for combining matrices comes up so often in applications
that it is used to define matrix multiplication. We say that
   
10 25 12   323
 5 10 5  4 135
the product of  
 7 5 0  and
7 is   .
 63 
9
10 8 10 186

229
Unit 9 Matrices

Now suppose that as well as calculating the amount owed by each of the
four bookshops, the publisher also wants to calculate the profit made from
sales to each individual bookshop. Suppose that the profits made on each
copy of Bestsellers 1, 2 and 3 are £1, £2 and £2, respectively. You can use
exactly the same procedure as above to calculate the profit made from the
sales to each bookshop. You write the profits (in £) made on the three
books as a three-dimensional vector, and form
 
10 25 12  
 5 10 5  1
the product of  7 5 0
 and 2 ,
2
10 8 10
which is
   
10 × 1 + 25 × 2 + 12 × 2 84
 5 × 1 + 10 × 2 + 5 × 2  35
   
 7 × 1 + 5 × 2 + 0 × 2  = 17 .
10 × 1 + 8 × 2 + 10 × 2 46
So the profits made from sales to the four bookshops are £84, £35, £17
and £46, respectively.
In fact you can view the two matrix multiplications above – the one for the
amounts owed, and the one for the profits – as a single matrix
multiplication. To do this, you put the two vectors containing the prices
and the profits together as a single matrix with two columns, and you put
the two vectors that contain the results of the calculations together as a
single matrix with two columns, and say that
   
10 25 12   323 84
 5 10 5  4 1  
the product of   and 7 2 is 135 35 . (1)
7 5 0  63 17
9 2
10 8 10 186 46

This calculation is an example of how matrices are multiplied in general.

In Section 2 you’ll meet another context that shows that it makes sense to
multiply matrices in this way. Before that, you’ll see a more detailed
explanation of the procedure for multiplying matrices, and have a chance
to practise it yourself.
Let’s start by considering which pairs of matrices can be multiplied
together. Remember that the element in the first row and first column of
the product matrix in calculation (1) above was obtained by multiplying
each element in the first row of the first matrix by the corresponding
element in the first column of the second matrix, and adding the results:
10 × 4 + 25 × 7 + 12 × 9 = 323.
It was possible to do this because the number of columns in the first matrix
is the same as the number of rows in the second matrix, so each element in
the first row of the first matrix has a corresponding element in the first
column of the second matrix. You can multiply two matrices together only
if this condition holds.

230
 1 Matrices and matrix operations

Thus, you can multiply together a 4 × 3 matrix and a 3 × 2 matrix. On

the other hand, it is not possible to multiply together a 3 × 4 matrix and a
2 × 3 matrix. When it is not possible to multiply two matrices, we
sometimes say that their product is undefined.
A convenient way to determine whether two matrices, of sizes m × n and
p × q, say, can be multiplied together is to write their sizes next to each
other as follows:
m×n p × q.
The matrices can be multiplied only if the two numbers in the middle are
equal; that is, if n = p. If these numbers are equal, then the size of the
product matrix is given by the remaining numbers. So the product of an
m × n matrix and an n × q matrix is a matrix of size m × q.
For example, a 4 × 3 matrix and a 3 × 2 matrix can be multiplied together,
and the product has size 4 × 2. Here’s a useful way to picture this fact:

4× 3 3 ×2 gives 4 × 2.

Now let’s look at the procedure for multiplying two matrices step by step.
It’s demonstrated in the next example.

Example 1 Multiplying matrices

   
−2 1 3 1 0
Let A =  2 4 −2 and B =  21 2.
0 −1 0 −1 4
Check that the product matrix AB can be formed, determine the size
of AB, and calculate it.
Solution
Write down the sizes of A and B.
A has size 3 × 3 and B has size 3 × 2.
3× 3 3 ×2
The numbers in the middle are equal, so the product AB can be
formed. It has size 3 × 2.

231
Unit 9 Matrices

To obtain the element in the first row and first column of AB,
multiply each element in the first row of A by the corresponding
element in the first column of B, and add the results.
    
−2 1 3 1 0 (−2) × 1 + 1 × 21 + 3 × (−1)
 2 1
4 −2   2 2 =  
0 −1 0 −1 4
 9 
−2
= 

To obtain the element in the first row and second column of AB,
apply the same procedure to the first row of A and the second column
of B.
    9 
−2 1 3 1 0 − 2 (−2) × 0 + 1 × 2 + 3 × 4
 2 4 −2   12 2  =  
0 −1 0 −1 4
 9 
− 2 14
= 

To obtain the element in the second row and first column of AB,
apply the same procedure to the second row of A and the first column
of B.
    
−2 1 3 1 0 − 29 14
 2 4 −2   12 2 = 2 × 1 + 4 × 21 + (−2) × (−1) 
0 −1 0 −1 4
 9 
− 2 14
= 6 

In general, to obtain the element in the ith row and jth column
of AB, multiply each element in the ith row of A by the
corresponding element in the jth row of B, and add the results.
  
−2 1 3 1 0
 2 4 −2  12 2
0 −1 0 −1 4
 
− 29 14
= 6 2 × 0 + 4 × 2 + (−2) × 4
1
0 × 1 + (−1) × 2 + 0 × (−1) 0 × 0 + (−1) × 2 + 0 × 4
 9 
− 2 14
= 6 0
− 21 −2

232
 1 Matrices and matrix operations

Here’s a summary of the procedure for matrix multiplication.

Matrix multiplication
Let A and B be matrices. Then the product matrix AB can be
formed only if the number of columns of A is equal to the number of
rows of B.
If A has size m × n and B has size n × p, then the product AB has
size m × p.
The element in row i and column j of the product matrix AB is
obtained by multiplying each element in the ith row of A by the
corresponding element in the jth column of B and adding the results.
In element notation, if cij denotes the element in the ith row and
jth column of AB, then
cij = ai1 b1j + ai2 b2j + · · · + ain bnj .

Figure 2 illustrates how row i of a matrix A and column j of a matrix B

are combined to give the element in row i and column j of the product
matrix AB.
column j column j
b1j
row i ai1 ai2 . . . ain b2j = row i
...

bnj

ai1 b1j + ai2 b2j + · · · + ain bnj

Figure 2 The element in row i and column j of a product matrix AB

expressed in terms of elements from A and B
Matrix multiplication may seem quite complicated at first, but it is a very
useful technique, and with practice you should become proficient at it.

233
Unit 9 Matrices

Activity 6 Multiplying matrices

In each of
parts (a)–(e) below, calculate the matrix product if it exists.
   
29 . / 9
−2 3 1 $ -
(a)  71 (b) 1 2 3 8
4 0 5
−6 8 7
    
2
9 0 10 9 $ -
(c)  7
1   4 1 (d) 8 1 2 3
−68 −2 7 7
 
. / 1 0 1
1 2 3 
(e) 1 0 −1
7 6 1
2 1 3

Matrix multiplication shares many of the properties of multiplication of

numbers, as you’ll see later in this subsection and later in the unit.
However, there’s an important difference between the properties for
matrices and those for numbers, as you can find out in the next activity.

Activity 7 Investigating matrix multiplication

(a) Determine whether the products AB and BA are defined for the
following pairs of matrices.
. / . /
−1 4 3
(i) A = and B =
2 1 5
. / . /
1 1 0 1
(ii) A = and B =
−1 0 2 3
(b) For each of parts (a)(i) and (a)(ii), if the products AB and BA are
both defined, evaluate them. What do you notice?

Activity 7 illustrates the important fact that matrix multiplication is not

commutative. First, there are matrices A and B for which the product AB
exists but the product BA is not defined. Second, even in cases where
both the products AB and BA are defined, these two products can be
different matrices.
In the next activity, you can investigate whether another property that
holds for multiplication of numbers also holds for multiplication of
matrices.

234
 1 Matrices and matrix operations

Activity 8 Investigating matrix multiplication further

. / . /
$ - 1 −1 1 3 0
Let A = −1 0 , B = and C = .
2 −1 −3 4 1
(a) What are the sizes of the products AB and BC? Calculate these
matrices.
(b) Do the products (AB)C and A(BC) exist? If so, calculate them.
What do you notice?

In Activity 8 you should have found that, for the particular three matrices
A, B and C in the activity, (AB)C = A(BC). This does not of course tell
you that this property holds for matrices in general. However, this
property does hold in general: whenever A, B and C are matrices such
that the products AB and BC are defined, we have
(AB)C = A(BC).
In other words, matrix multiplication is associative. In this respect matrix
multiplication behaves like multiplication of numbers.
In the next activity you can investigate how matrix multiplication
interacts with multiplication by a scalar.

Activity 9 Combining matrix multiplication and multiplication by a

scalar
. / . /
1 2 1 0
Let A = and B = . Show that
3 4 2 1
A(2B) = 2(AB).

In Activity 9 you verified a particular instance of a general property of

matrices: if the matrix product AB exists, then, for any scalar k,
A(kB) = (kA)B = k(AB).

In the next activity you’re asked to investigate how matrix multiplication

interacts with matrix addition.

235
Unit 9 Matrices

Activity 10 Combining matrix addition and multiplication

. / . / . /
1 −1 1 −1
Let A = ,B= and C = .
3 4 2 1
By calculating A(B + C) and AB + AC, show that
A(B + C) = AB + AC.

In Activity 10 you verified a particular instance of another general

property of matrices: if the matrix product A(B + C) can be formed, then
A(B + C) = AB + AC.
It’s also true that if the matrix product (A + B)C can be formed, then
(A + B)C = AC + BC.
In other words, matrix multiplication is distributive over matrix addition.
So, for example, if you encounter a matrix expression of the form
AB + AC,
then you can factorise it to give
AB + AC = A(B + C).

Here’s a summary of the properties of matrix multiplication that you’ve

seen so far.

Some properties of matrix multiplication

The following properties hold for all matrices A, B and C for which
the products and sums mentioned are defined.
• (AB)C = A(BC)
• k(AB) = (kA)B = A(kB), for any scalar k
• A(B + C) = AB + AC
• (A + B)C = AC + BC

Remember also the following important fact.

Matrix multiplication is not commutative

• There are matrices A, B such that the product AB exists but
the product BA does not.
• Even when both products are defined, it can happen that
AB '= BA.

236
 1 Matrices and matrix operations

Matrix powers
Just as the square a2 of a real number a is defined to be the product a × a,
so we define the square A2 of a matrix A to be the matrix product AA,
if it exists.
For example, if
. /
2 1
A= ,
3 0
then
. /. / . /
2 1 2 1 7 2
A2 = AA = = .
3 0 3 0 6 3

In the next activity you’re asked to investigate which matrices can be

squared.

Activity 11 Investigating which matrices can be squared

. /
1 0 3
(a) Let M = . Why is it not possible to form the product MM?
0 2 4
. /
2 −1
(b) Let N = . Is the product NN defined? If, so, evaluate it.
0 2
(c) Describe which matrices can be squared.

In Activity 11 you saw that if A is a square matrix (a matrix with the

same number of rows as columns), then its square A2 exists. The
matrix A2 has the same size as the original matrix A.
Be careful not to confuse the two different concepts square matrix and
square of a matrix.
If A is a square matrix, then since A2 is a square matrix of the same size
as A, we can also form the products (A2 )A and A(A2 ). Since matrix
multiplication is associative, we know that
A2 A = (AA)A = A(AA) = AA2 .
Hence either of these products can be written unambiguously as AAA.
This matrix is called the cube of A and is denoted by A3 . For example,
for the matrix
. /
2 1
A=
3 0
discussed at the beginning of this subsection, we have
. /. / . /
3 2 7 2 2 1 20 7
A =A A= = .
6 3 3 0 21 6

237
Unit 9 Matrices

In general, for any square matrix A, the nth power of A is

An = ,AA"!
· · · A3 .
n times

Activity 12 Calculating matrix powers

(a) Calculate the squares of the following matrices:

 
. / 1 0 1
−1 4
, 0 2 0 .
2 0
0 0 3
. /
2 1
(b) Let A = . Calculate A6 .
3 0
Hint: the cube A3 of this matrix A was calculated before the activity.

Matrix multiplication with a computer

Matrix multiplication can be time-consuming to carry out when the
matrices involved are large. In the next activity you can find out how to
use a computer to carry out matrix operations, including multiplication.
This work will increase your familiarity with the way matrices behave, and
help you distinguish between circumstances when it is worth using a
computer to perform matrix calculations and when it is more efficient to
do the calculations by hand.

Activity 13 Working with matrices on a computer

Work through Subsection 10.1 of the Computer algebra guide.

2 Matrices and networks

Matrices can be used to represent and analyse various flows in networks. A
network is a collection of objects connected by links – either physical or
abstract – such as cities connected by roads, telephone exchange points
connected by telephone lines, or people connected by acquaintance. A
foremost example is the internet, which is a network of communicating
computers linked by a range of technologies. In this section you’ll look at
flows in networks of a particular type, and learn how to use matrices to
work with them.
A network of world internet
traffic

238
 2 Matrices and networks

2.1 Networks
Figure 3 shows a simple example of a network of water pipes. Water can
be poured into the system at position A or B, and it then flows down the
pipes to come out at positions U , V and W . The pipes have different
capacities: the three pipes that come out of position A are labelled with
numbers that indicate the proportions of water that flow through these
pipes when water is poured in at A, and the three pipes that come out of
position B are labelled in the same way. For example, if a litre of water is
poured in at position A, then 0.4 litres flows down the first pipe from A,
0.2 litres flows down the second pipe from A, and 0.4 litres flows down the
third pipe from A.

A B

0:2 0:1
0:4 0:6

0:3 0:4

U V W

Figure 3 A network of water pipes

A network like this can be represented more simply as a network
diagram, which is a mathematical representation of a physical network.
The network diagram in Figure 4 represents the network in Figure 3.

A B

0:2 0:1
0:4 0:6

0:3 0:4
U V W

Figure 4 A network diagram representing the pipe system in Figure 3

The dots in a network diagram are called nodes, and the lines are called
arcs. The arrows on the arcs in Figure 4 are included because this network
diagram represents a network in which flow along an arc takes place in
only one direction. When no confusion arises, we use the word ‘network’ to
mean either ‘network diagram’ or ‘physical network’. We’ll use network
diagrams to represent networks from now on in this section.

239
Unit 9 Matrices

Example 2 Calculating network outputs

The network diagram in Figure 4 is repeated below.

A B

0:2 0:1
0:4 0:6

0:3 0:4
U V W

How much water is output at each of nodes U , V and W if the only

input is 1 litre of water at node B ?
Solution
The label on the arc joining node B to node U indicates the
proportion of the water input at node B that reaches node U .
The label on the arc from node B to node U is 0.3. So 0.3 litres of
water is output at node U when 1 litre is input at node B.
Read the labels on the other arcs from node B.
Similarly, 0.1 litres of water is output at node V , and 0.6 litres of
water is output at node W .

Now consider what happens if water is input at both node A and node B
in the network in Example 2. The amount of water reaching an output
node is the sum of the water reaching the node from node A and from
node B. For example, if 1 litre of water is input at node A and another
1 litre is input at node B, then node U will collect 0.4 litres of water from
node A and 0.3 litres from node B. So, in total, 0.7 litres of water will be
output at node U . Similarly, 0.2 + 0.1 = 0.3 litres of water will be output
at node V , and 0.4 + 0.6 = 1 litre of water will be output at node W .
In general, if x litres of water is input at node A and y litres of water is
input at node B, then
0.4x + 0.3y litres of water is output at node U,
0.2x + 0.1y litres of water is output at node V,
0.4x + 0.6y litres of water is output at node W.
In the next activity you can practise calculating network outputs yourself.

240
 2 Matrices and networks

Activity 14 Calculating network outputs

The following network diagram represents a network of water pipes.

U V W

0:6 0:4
0:8 0:3
0:2 0:7

C D

(a) How much water is output at nodes C and D if 1 litre of water is

input at node U , 2 litres of water is input at node V , and there is no
input at node W ?
(b) How much water is output at nodes C and D if x litres of water are
input at node U , y litres of water is input at node V , and z litres of
water is input at node W ?

Network diagrams like those that you’ve seen in this subsection can be
used to model many other kinds of flows, such as the movement of money
in and out of companies. For example, consider three corporate donors, D,
E and F , say, that every month make donations to two charities, J and K.
Donor D donates 20% of its monthly designated sum to charity J and 80%
to charity K, donor E donates 30% of its sum to charity J and 70% to
charity K, and donor F donates 50% of its sum to charity J and 50% to
charity K.
The flow of money from the three donors to the two charities can be
represented by the network in Figure 5. For example, a total donation of
£1000 from donor D in a given month is modelled by an input of £1000 at
node D. The label of the arc connecting D to J tells you that 0.2, or 20%,
of that sum, which is £200, reaches the output node J .

D E F

0:3 0:7
0:8 0:5
0:2 0:5

J K

Figure 5 A network diagram representing the flow of money from three

corporate donors to two charities

241
Unit 9 Matrices

Activity 15 Calculating more network outputs

The network diagram in Figure 5 is repeated below.

D E F

0:3 0:7
0:8 0:5
0:2 0:5

J K

It represents the flow of money from donors D, E and F to charities J

and K, as described above. How much money is output at each of nodes J
and K in each of the following situations?
(a) The only input is £2000 at node D.
(b) The input at node D is £2400, the input at node E is £1700, and the
input at node F is £1100.
(c) The input (in £) at node D is x, the input (in £) at node E is y, and
the input (in £) at node F is z.

2.2 From networks to matrices

In this subsection you’ll see how network output calculations of the kinds
that you saw in Subsection 2.1 can be expressed succinctly using matrices.
Consider again the first network representing water pipes that you saw in
Subsection 2.1. It is repeated in Figure 6.

A B

0.2 0.1
0.4 0.6

0.3 0.4
U V W

Figure 6 A network representing water pipes

The amounts of water (in litres) input at nodes A and B can be
represented as a two-dimensional vector (a 2 × 1 matrix), whose first and

242
 2 Matrices and networks

second components are the inputs at A and B, respectively. For example,

the vector
. /
0
1
represents 0 litres of water input at node A and 1 litre of water input at
node B.
The outputs from the network can also be represented by a vector. For
example, the vector
 
0.4
0.2
0.4
represents an output of 0.4 litres of water at node U , 0.2 litres at node V ,
and 0.4 litres at node W .
The network itself can be represented by the matrix
 
0.4 0.3
0.2 0.1 .
0.4 0.6
Here the elements in each column are the labels of the arcs from a node
(A or B), and the elements in each row are the labels of the arcs to a node
(U , V or W ), as shown below:
A B
 
U 0.4 0.3
V 0.2 0.1 .
W 0.4 0.6
For example, the element in the column labelled B and the row labelled U
is the label of the arc from node B to node U . You’ll see the reason for
arranging the arc labels in this way in a moment.
You saw in Subsection 2.1 that if x litres of water is input at node A and
y litres of water is input at node B, then
0.4x + 0.3y litres of water is output at node U,
0.2x + 0.1y litres of water is output at node V,
0.4x + 0.6y litres of water is output at node W.
In other words,
 
. / 0.4x + 0.3y
x
if the input vector is , then the output vector is 0.2x + 0.1y  .
y
0.4x + 0.6y
This output vector can be expressed as the product of the matrix that
represents the network and the input vector:
   
0.4x + 0.3y 0.4 0.3 . /
0.2x + 0.1y  = 0.2 0.1 x .
y
0.4x + 0.6y 0.4 0.6

243
Unit 9 Matrices

So we can say that

 
. / 0.4 0.3 . /
x x
if the input vector is , then the output vector is 0.2 0.1 .
y y
0.4 0.6
So, for any input vector, you can work out the corresponding output vector
by multiplying the input vector by the matrix that represents the network.
This is illustrated in the next example.

Example 3 Using matrices to calculate network outputs

Use the matrix that represents the network diagram in Figure 6 to
calculate the outputs at nodes U , V and W if 2 litres of water is
input at node A and there is no input at node B.
Solution
The matrix representing the network was found before this
example.
The matrix representing the network is
 
0.4 0.3
0.2 0.1 .
0.4 0.6
The input vector (in litres) is
. /
2
.
0
The output vector is the product of the matrix representing the
network and the input vector.
Hence the output vector is
     
0.4 0.3 . / 0.4 × 2 + 0.3 × 0 0.8
0.2  2
0.1 = 0.2 × 2 + 0.1 × 0 = 0.4 .
  
0
0.4 0.6 0.4 × 2 + 0.6 × 0 0.8
So the outputs at nodes U , V and W are 0.8 litres, 0.4 litres and
0.8 litres, respectively.

In general, to represent a network by a matrix in the way described above,

you need a column for each input node and a row for each output node. So
if the network has n input nodes and m output nodes, then the matrix has
size m × n.
When you find a matrix representing a network of the type considered in
this section, it’s useful to check that the elements in each column have
sum 1. The matrix must have this property because the elements in each

244
 2 Matrices and networks

column are the labels on an arc from a particular node. You can see that
the matrix representing the network in Example 3 has this property.

Activity 16 Using matrices to calculate network outputs

The network in Activity 14, which represents a network of water pipes, is

shown again below.

U V W

0:6 0:4
0:8 0:3
0:2 0:7

C D

(a) Write down the matrix that represents this network.

(b) Write down a vector that represents an input of 1 litre of water at each
of nodes U , V and W .
(c) Use your answer to part (a) to calculate the output vector for the
input vector in part (b), and interpret your answer in terms of the
amounts of water output at nodes C and D.

2.3 Combining networks

Networks can be combined by making the outputs from one network
become the inputs to another network. In this section you’ll see how to use
matrices to calculate the outputs from a combined network using the
information about the original networks.
The network diagram in Figure 4 and the one in Activity 14 are repeated
in Figure 7. Both diagrams represent networks of water pipes.

A B U V W

0:6 0:4
0:2 0:1 0:8 0:3
0:4 0:6
0:2 0:7
0:3 0:4
U V W C D
(a) (b)

Figure 7 Two network diagrams

245
Unit 9 Matrices

The first network has three output nodes, and the second network has
three input nodes. We can join these nodes so that, for instance, the water
output at U in the first network flows down the pipes that leave U in the
second network. The combined network is shown in Figure 8.

A B

0:2 0:1
0:4 0:6

0:3 0:4
U V W
0:6 0:4
0:8 0:3
0:2 0:7

C D

Figure 8 The combined network

The network obtained in this way can be considered as equivalent to a
simpler network with arcs going directly from the input nodes A and B to
the output nodes C and D, as illustrated in Figure 9. In general, we
regard two networks of the type considered in this section as equivalent if
they have the same input and output nodes, and all choices of inputs
produce the same outputs in both networks.

A B

C D

Figure 9 A network equivalent to the one in Figure 8

How can we find the labels for the arcs in the new, simpler network
diagram? Here’s how you can find the label for the arc from A to C, for
example.
Suppose that 1 litre of water is input at node A in the combined network
diagram, and there are no other inputs. Then, from the arc labels in the
top half of the diagram, you know that the amounts of water passing
246
 2 Matrices and networks

through nodes U , V and W are 0.4, 0.2 and 0.4 litres, respectively. So,
from the arc labels in the bottom half of the diagram, you can work out
the following.
• Of the 0.4 litres of water that passes through node U , the amount that
reaches node C is 0.2 × 0.4 = 0.08 litres.
• Of the 0.2 litres of water that passes through node V , the amount that
reaches node C is 0.6 × 0.2 = 0.12 litres.
• Of the 0.4 litres of water that passes through node W , the amount
that reaches node C is 0.3 × 0.4 = 0.12 litres.
To find the total amount of water output at node C, we add the
contributions from the three different routes, which gives
0.08 + 0.12 + 0.12 = 0.32 litres. So the proportion of the water input at
node A that’s output at node C is 0.32, and hence the arc from A to C
should be labelled with the number 0.32.
You can find the labels for the remaining arcs in a similar way. However,
these calculations are quite tedious, and it turns out that they can be
simplified by using the matrices that represent the two original networks,
which are
 
0.4 0.3 . /
0.2 0.1 and 0.2 0.6 0.3
,
0.8 0.4 0.7
0.4 0.6
respectively. These matrices were found in Subsection 2.2.
Suppose that the input and output vectors (in litres) for the combined
network in Figure 8 are
. / . /
a c
and ,
b d
respectively. Suppose also that the output vector (in litres) of the top part
of the network, which is also the input vector for the bottom part, is
 
u
v  .
w
Then
     
u 0.4 0.3 . / . / . / u
 v  = 0.2 0.1 a and
c
=
0.2 0.6 0.3  
v .
b d 0.8 0.4 0.7
w 0.4 0.6 w
Using the first of these equations to substitute in the second gives
  
. / . / 0.4 0.3 . /
c 0.2 0.6 0.3  a
= 0.2 0.1 .
d 0.8 0.4 0.7 b
0.4 0.6
You know from Section 1 that matrix multiplication is associative, so
  
. / . / 0.4 0.3 . /
c  0.2 0.6 0.3   a
= 0.2 0.1 ;
d 0.8 0.4 0.7 b
0.4 0.6

247
Unit 9 Matrices

that is,
. / . /. /
c 0.32 0.3 a
= . (2)
d 0.68 0.7 b
The conclusion of this discussion is that the matrix that represents the
combined network in Figure 9 is the product of the matrices that represent
the original networks. So matrix multiplication provides a neat way to
calculate outputs of the combined network.
Notice that the matrix representing the top network is the second matrix
in the product that represents the combined network.
The labels of the simpler network that is equivalent to the combined
network can be read off the product matrix. This is easier if we label the
rows and the columns with the nodes as below:
A B
. /
C 0.32 0.3
.
D 0.68 0.7

Notice that each column of this matrix has sum 1, as expected in view of
the comment after Example 3.
The original combined network and the equivalent simpler network, with
the arc labels, are shown in Figure 10.

A B A B

0:2 0:1
0:4 0:6
0:68 0:3
0:3 0:4
U V W 0:32 0:7
0:6 0:4
0:8 0:3
0:2 0:7

C D C D

Figure 10 The two equivalent networks

248
 2 Matrices and networks

Activity 17 Finding the matrix representing a combined network

The network below is the combination of two smaller networks, one with
input nodes U and V and output nodes P and Q, and the other with input
nodes P and Q and output nodes A, B, and C.

U V

0.5 0.8
0.2 0.5

P Q

0.3 0.6
0.3 0.4
0.3 0.1
A B C

. /
p
(a) Write down the matrix equation that gives the output vector at
. / q
u
nodes P and Q when the input vector is at nodes U and V .
v
 
a
(b) Write down the matrix equation that gives the output vector b  at

c
. /
p
nodes A, B and C when the input vector is at nodes P and Q.
q
(c) Use your answers to parts (a) and (b) to find the
. matrix
/ equation for
u
the combined network when the input vector is at nodes U
v
and V .
(d) Draw a network with input nodes U and V and output nodes A, B
and C that is equivalent to the given network.
(e) Suppose that the given network represents a network of water pipes.
Determine the outputs at nodes A, B and C when 1 litre of water is
input at U and 2 litres of water is input at V .

Joining networks is also useful in other contexts. For example, Figure 11

repeats the network diagram from Figure 5, which represents the flow of
money from three corporate donors D, E and F to two charities J and K.

249
Unit 9 Matrices

D E F

0:3 0:7
0:8 0:5
0:2 0:5

J K

Figure 11 A network diagram representing the flow of money from three

corporate donors to two charities
Suppose that donors D, E and F are the only donors to the two charities,
and that the two charities spend their money in two different geographical
regions, R and S, as indicated in Figure 12. So, for example, charity J
spends 60% of its money in region R and 40% in region S.

J K

0:6 0:4 0:9 0:1

R S

Figure 12 A network diagram representing the flow of money from two

charities into two regions
In the next activity you’re asked to combine the networks in Figures 11
and 12 to obtain a network that represents the flow of money from the
three corporate donors to the two regions.

Activity 18 Finding the matrix representing another combined

network
(a) Write down the matrices that represent the networks in Figures 11
and 12.
(b) Hence find the matrix that represents the network obtained by
combining these two networks.
(c) Draw a simplified network that represents the combined network
described in part (b).
(d) If, in a particular month, donor D donates £1700, donor E donates
£1400 and donor F donates nothing, how much money is spent in
region S?

250
 3 The inverse of a matrix

In a real-life situation, networks like the one in Activity 18 would have

many more input and output nodes, and the matrices representing them
would be manipulated on a computer.

3 The inverse of a matrix

In Section 1 you met four matrix operations: addition, multiplication by a
scalar, subtraction and matrix multiplication. No division was mentioned:
it is not possible to divide a matrix by another matrix. However there is a
way to manipulate matrices that in certain situations plays the role of
division. To understand this method, you need to know about special
matrices called identity matrices, which are introduced next.

3.1 Identity matrices

The number 1 has a special property among real numbers: multiplication
by 1 leaves any real number unchanged. An identity matrix is a matrix
that behaves like the number 1, in the sense that if a matrix A is
multiplied by an identity matrix of an appropriate size then the result is
again A. You will meet an identity matrix in the next activity.

Activity 19 Multiplying by an identity matrix

. / . / . /
3 4 2 7 −3 1 0
Let A = ,B= and I = . Evaluate each of
−2 1 8 1 2 0 1
the following products, where possible:
AI, IA, BI, IB.

The matrix I in Activity 19 has the following two properties.

• If A is any matrix such that the product AI exists, then AI = A.
• If A is any matrix such that the product IA exists, then IA = A.
Any matrix I with these two properties is called an identity matrix.
For a matrix to be an identity matrix, it has to be square. To see this,
suppose that A is an m × n matrix. Then in order for the product AI to
be defined, I must have n rows. Moreover, in order for AI to have size
m × n (which it must if it is to be equal to A), I must have n columns:
m× n times n × n gives m × n.

So I must have size n × n.

251
Unit 9 Matrices

It’s straightforward to verify that the following 2 × 2, 3 × 3 and 4 × 4

matrices are identity matrices:
 
  1 0 0 0
. / 1 0 0
1 0 0 1 0 0
, 0 1 0 ,  0 0
.
0 1 1 0
0 0 1
0 0 0 1

In fact, for every natural number n, the n × n matrix that has ones down
the leading diagonal (the diagonal that starts at the top left element and
ends at the bottom right element), and zeros everywhere else, is an
identity matrix.
Furthermore, for every natural number n, this matrix is the only n × n
identity matrix. To see this, let I be the n × n identity matrix described
above, and let E be an n × n identity matrix. Then IE = I and IE = E,
so I = E.
When we’re working with matrices, we usually reserve the letter I to
denote the n × n identity matrix.

Activity 20 Multiplying by an identity matrix

 
. / x
a b c
Let A = and B = y  be general matrices, and let I be the
d e f
z
3 × 3 identity matrix. Verify that AI = A and IB = B.

The facts that you’ve seen about identity matrices are summarised in the
following box.

Identity matrices
An identity matrix is a square matrix I such that
• for any matrix A for which the product AI is defined, AI = A
• for any matrix A for which the product IA is defined, IA = A.
Each identity matrix has the form
 
1 0 0 ··· 0
0 1 0 · · · 0
 
0 0 1 · · · 0
 .
 .. .. .. . . .. 
. . . . .
0 0 0 ··· 1
That is, it has ones down the leading diagonal and zeros elsewhere.

252
 3 The inverse of a matrix

3.2 Matrix inverses

As mentioned earlier, it isn’t possible to divide matrices, but it is
sometimes possible to perform a manipulation with a similar effect.
Remember that, in the arithmetic of numbers, dividing by a number is the
same as multiplying by its reciprocal. For example, dividing by 2 is the
same as multiplying by 21 . The reciprocal of a non-zero number a is the
number b such that ab = 1.
There is a notion in matrix arithmetic that is analogous to the notion of a
reciprocal in ordinary arithmetic. Let A be a square matrix. If there is
another matrix B of the same size with the property that
AB = I and BA = I,
where I is an identity matrix, then we say that A is invertible and that B
is an inverse of A. Multiplying by an inverse of a matrix is useful in many
situations where you might want to use a matrix operation analogous to
division.

Activity 21 Checking matrix inverses

For each of the following pairs of matrices A and B, check that B is an

inverse of A, by finding the products AB and BA.
. / . /
8 5 2 −5
(a) A = , B=
3 2 −3 8
   
2 3 1 0 1 −1
(b) A = −1 1 0 , B = 0 2 −1
−2 1 0 1 −8 5

If a matrix A is invertible, then it has exactly one inverse. To see this,

suppose that the matrices B and C are both inverses of the matrix A.
Then
AB = I and BA = I, and also AC = I and CA = I.
Now consider the matrix BAC. Since matrix multiplication is associative,
BAC = (BA)C = IC = C,
and also
BAC = B(AC) = BI = B.
Hence B and C are the same matrix. That is, A has only one inverse.
We usually write the inverse of an invertible matrix A as A−1 . So the
following facts hold.

253
Unit 9 Matrices

Inverse of a matrix
If A is an invertible matrix, then
AA−1 = I and A−1 A = I,
where I is the identity matrix of the same size as A.

It follows from the definition of the inverse of a matrix that if A is an

invertible matrix, then not only is A−1 the inverse of A, but also A is the
inverse of A−1 . In other words, A and A−1 are inverses of each other.
In the rest of this subsection, you’ll learn how to establish whether any
particular 2 × 2 matrix has an inverse, and how to work out the inverse
when it exists. You’ll also use a computer to calculate inverses of larger
matrices.

Inverses of 2 × 2 matrices
For 2 × 2 matrices, there is a useful formula for finding inverses.

Inverse of a 2 × 2 matrix
. /
a b
The inverse of the 2 × 2 matrix is given by
c d
. /
1 d −b
, provided that ad − bc '= 0.
ad − bc −c a
In other words, to obtain the inverse when ad − bc '= 0, swap the two
elements on the leading diagonal and multiply the other two elements
by −1, then multiply the resulting matrix by the scalar 1/(ad − bc).

You’ll be asked to verify this general formula later in this subsection, but
for now let’s check that it works for the matrix
. /
8 5
A=
3 2
in Activity 21(a). Here ad − bc = 8 × 2 − 5 × 3 = 1 '= 0, so the scalar
1/(ad − bc) is 1. You can see that the matrix
. /
2 −5
B=
−3 8
in Activity 21(a) is indeed obtained from A by swapping 8 and 2,
multiplying the other two elements by −1, and multiplying the resulting
matrix by the scalar 1.
Notice that the box above includes the condition ad − bc '= 0. This is
because the formula involves dividing by ad − bc, and division by 0 is not
defined. If the elements of a 2 × 2 matrix do not satisfy this condition,

254
 3 The inverse of a matrix

then the matrix does not have an inverse. You’ll see a proof of this fact in
the next section.
In fact the quantity ad − bc that appears in the formula is so important
that it deserves a name, as follows.

Determinant of a 2 × 2 matrix
. /
a b
Let A = . Then the number ad − bc is called the
c d
determinant of A, and written as det A.
If the determinant of A is not zero, then A is invertible.
If the determinant of A is zero, then A is not invertible.

Some texts denote det A by |A|, which looks like the notation for the
modulus of a real number, but has a different meaning in this context.
This notation is also used when a matrix is written out in full; for example,
* *
* 2 1 *
* *
* −3 0 * = 2 × 0 − 1 × (−3).

When seeking the inverse of a 2 × 2 matrix, you should first check that the
matrix is invertible by calculating the determinant. If the determinant is
not zero, then you can use the rule for finding the inverse.

Example 4 Finding inverses of 2 × 2 matrices

For each of the matrices below, check whether its inverse exists, and
find it if it does.
. / . /
4 7 3 4
(a) A = (b) B =
3 6 6 8
Solution
. /
4 7
(a) Let A = .
3 6
Check whether det A = 0.
Here det A = 24 − 21 = 3. Since det A '= 0, the matrix has an
inverse.
Use the formula to write down the inverse of A.
The inverse is
. / 0 7
1
1 6 −7 2 − 3
A−1 = = .
3 −3 4 −1 4
3

255
Unit 9 Matrices

If you wish, check the answer by verifying that AA−1 = I and

A−1 A = I.
Check: we have
. /. / . /
2 − 73 4 7 1 0
4 =
−1 3
3 6 0 1
and
. /. / . /
4 7 2 − 37 1 0
4 = .
3 6 −1 3
0 1
. /
3 4
(b) For B = , we have
6 8

det B = 3 × 8 − 4 × 6 = 24 − 24 = 0.

The determinant of B is zero.

Thus B is not invertible; that is, B−1 does not exist.

You can practise finding the inverses of 2 × 2 matrices in the next activity.

Activity 22 Finding inverses of 2 × 2 matrices

Determine which of the following matrices are invertible. For each matrix
that is invertible, write down its inverse.
. / . /
13 5 3 0
(a) A = (b) B =
5 2 −5 2
0 1 . /
1
−3 5 1.5 2.5
(c) C = (d) D =
5 − 13 0.5 1.5

In the next activity you’re asked to verify that the formula for the inverse
of a 2 × 2 matrix works in general.

256
 4 Simultaneous linear equations and matrices

Activity 23 Verifying the formula for the inverse of a 2 × 2 matrix

. / . /
a b 1 d −b
Let A = , with ad − bc '= 0, and B = .
c d ad − bc −c a
Work out the products AB and BA, and deduce that B = A−1 .
Hint: use the fact that P(kQ) = k(PQ) for any matrices P and Q for
which the product PQ is defined and for any scalar k.

A matrix that does not have an inverse is called a non-invertible matrix.

In some texts, non-invertible matrices are called singular matrices, and
invertible matrices are called non-singular matrices.

Determinants and inverses of larger matrices

You have seen that a 2 × 2 matrix has an inverse only if its determinant is
not zero.
The determinant is defined for square matrices of any size, and it plays the
same role as in the 2 × 2 case: any square matrix A such that det A '= 0
has an inverse A−1 . If det A = 0, then no inverse can be found.
It is possible to use hand calculations to work out the determinant and the
inverse (when it exists) of a square matrix whose size is larger than 2 × 2,
but the process is rather complicated, and beyond the scope of this
module. However, you can use the computer algebra system to calculate
determinants and inverses of larger matrices.

Activity 24 Working with matrix inverses on a computer

Work through Subsection 10.2 of the Computer algebra guide.

4 Simultaneous linear equations and

matrices
In Section 3 of Unit 2, you met simultaneous linear equations and you saw
two methods (substitution and elimination) for solving them. Here you’ll
meet a third method, which uses matrices.

257
Unit 9 Matrices

4.1 Solving simultaneous linear equations in two

unknowns
Consider the following simultaneous linear equations in the two
unknowns x and y:
0.4x + 0.2y = 4
(3)
0.6x + 0.8y = 11.
There’s a useful way to write equations (3) as a single equation involving
two-dimensional vectors (which are 2 × 1 matrices). To do this, you write
the sides of each equation as the components of a vector, as follows:
. / . /
0.4x + 0.2y 4
= .
0.6x + 0.8y 11
You can now write the left-hand side of this equation as the product of a
2 × 2 matrix and a two-dimensional vector:
. /. / . /
0.4 0.2 x 4
= .
0.6 0.8 y 11
The 2 × 2 matrix that appears in this equation is called the coefficient
matrix of the simultaneous linear equations (3). Its elements are the
numbers that appear on the left-hand sides of equations (3). Any pair of
simultaneous linear equations in two unknowns can be written in this
form.

Matrix form of simultaneous linear equations

The simultaneous linear equations
ax + by = e
cx + dy = f
can be written as the matrix equation
. /. / . /
a b x e
= .
c d y f
. /
a b
The matrix is called the coefficient matrix.
c d

Note that the coefficient matrix of simultaneous linear equations depends

on how the equations are written. For example, if you write equations (3)
as
0.6x + 0.8y = 11
0.4x + 0.2y = 4,
then you obtain the coefficient matrix
. /
0.6 0.8
,
0.4 0.2
which is different from the coefficient matrix obtained above for the same
equations.
258
 4 Simultaneous linear equations and matrices

Example 5 Writing simultaneous linear equations in matrix form

Write the following pair of equations in matrix form.
y = 2x − 1
3y = 2

Solution
Rearrange each equation so that the unknowns are on the
left-hand side and in the same order in each equation, and the
constant terms are on the right-hand sides.
The equations can be rewritten as follows:
2x + (−1)y = 1
0x + 3y = 2.
Now read off the matrix form from the equations.
Therefore the equations can be written in matrix form as
. /. / . /
2 −1 x 1
= .
0 3 y 2

There are other possible solutions to Example 5. For example, the

equations can be rewritten in the form
−2x + y = −1
0x + 3y = 2,
which gives the matrix form
. /. / . /
−2 1 x −1
= .
0 3 y 2

In the next activity you can practise finding matrix forms of pairs of
simultaneous linear equations.

Activity 25 Writing simultaneous linear equations in matrix form

Write each of the following pairs of equations in matrix form.

3 4
(a) 2x + 3y = 3 (b) 2x − 6 = 0 (c) 5x − 5y = 18
x + 4y = −1 3x + 6y = 15 4 3
5x + 5y = −1

259
Unit 9 Matrices

The matrix form of a pair of simultaneous linear equations is not just

convenient shorthand: it is the initial step in a method for solving the
equations. Let’s see how this method works for equations (3) from the
beginning of this subsection, which have the matrix form
. /. / . /
0.4 0.2 x 4
= .
0.6 0.8 y 11
Let A be the coefficient matrix; that is,
. /
0.4 0.2
A= .
0.6 0.8
Then the matrix form of the equations can be written as
. / . /
x 4
A = .
y 11
Since det A = 0.4 × 0.8 − 0.2 × 0.6 = 0.2, the matrix A has an inverse A−1 .
If we multiply both sides of the equation above by A−1 , then the result is
. . // . /
−1 x −1 4
A A =A .
y 11
By the associativity of matrix multiplication, this equation can be
rearranged as:
. / . /
$ −1 - x −1 4
A A =A .
y 11
Since A−1 A = I, where I is the 2 × 2 identity matrix, the equation above
can be rewritten as
. / . /
x 4
I = A−1 .
y 11
Since I is an identity matrix, our equation becomes
. / . /
x −1 4
=A . (4)
y 11
Therefore, if we work out A−1 and calculate the product on the right-hand
side of this matrix equation, then we’ll find a solution to the original
equations. Now
. / . /
−1 1 0.8 −0.2 4 −1
A = = .
0.2 −0.6 0.4 −3 2
So substituting for A−1 in equation (4) gives
. / . /. / . /
x 4 −1 4 5
= = .
y −3 2 11 10

In other words,
x=5 and y = 10.

260
 4 Simultaneous linear equations and matrices

The example above illustrates a general method for solving simultaneous

linear equations, which can be summarised as follows.

Strategy:
To solve a pair of simultaneous linear equations in two
unknowns using matrices
Write the simultaneous linear equations in matrix form Ax = b,
where A is the coefficient matrix, x is the corresponding vector of
unknowns, and b is the vector whose components are the
corresponding right-hand sides of the equations.
If the matrix A is invertible, then the solution is given by
x = A−1 b.

Example 6 Using matrices to solve simultaneous linear equations

Use matrices to solve the following pair of simultaneous linear
equations.
x − 5y = −3
x + 3y = 13
Solution
Find the matrix form of the equations.
The matrix form of the equations is
. /. / . /
1 −5 x −3
= .
1 3 y 13
Find the inverse of the coefficient matrix.
The determinant of the coefficient matrix is
1 × 3 − (−5) × 1 = 8,
so the inverse of the coefficient matrix is
. / 0 3 51
1 3 5 8 8
= .
8 −1 1 −8
1 1
8

261
Unit 9 Matrices

Use the inverse matrix to solve the simultaneous equations.

Hence
. / 0 1. /
x 1 3 5 −3
=
y 8 −1 1 13
. /
7
= .
2
The solution is x = 7, y = 2.
In the calculation above it is convenient to use the form of the
inverse matrix with the fraction 1/8 outside the matrix, in order to
simplify the working.

As usual, you can check the answer to Example 6 by substituting back into
the original equations.
You can practise the method in Example 6 in the next activity.

Activity 26 Using matrices to solve simultaneous linear equations

Use matrices to solve the following pairs of simultaneous linear equations.

(You were asked to write these pairs of equations in matrix form in
Activity 25.)
3 4
(a) 2x + 3y = 3 (b) 2x − 6 = 0 (c) 5x − 5y = 18
x + 4y = −1 3x + 6y = 15 4 3
5x + 5y = −1

4.2 Non–invertible coefficient matrices

You saw in Unit 2 that a pair of simultaneous linear equations in two
unknowns can have no solutions, exactly one solution or infinitely many
solutions. If it has exactly one solution, then we say that it has a unique
solution.
In this subsection you’ll see that a simple way to tell whether a pair of
simultaneous linear equations in two unknowns has a unique solution is to
work out the determinant of its coefficient matrix.
If the determinant is non-zero, then the coefficient matrix has an inverse
given by the formula that you saw in Subsection 3.2, and hence the
equations have a unique solution, by what you saw in Subsection 4.1.
Let’s now look at what happens when the determinant is zero. Let’s start
by looking at two examples.

262
 4 Simultaneous linear equations and matrices

First, consider the following pair of simultaneous equations:

2x − 3y = 5
(5)
−4x + 6y = 7.
The coefficient matrix is
. /
2 −3
,
−4 6
which has determinant 2 × 6 − (−3) × (−4) = 0. Let’s look for a solution
to the equations using the elimination method. If we multiply both sides of
the first one of equations (5) by −2, then we get
−4x + 6y = −10
−4x + 6y = 7.
These equations have identical left-hand sides, but different right-hand
sides. So they have no solution: no values of x and y can make −4x + 6y
equal to both −10 and 7.
Now consider the following pair of simultaneous equations:
x + 2y = −6
(6)
−3x − 6y = 18.
The coefficient matrix is
. /
1 2
,
−3 −6
which has determinant 1 × (−6) − 2 × (−3) = 0. Again, let’s look for a
solution to the equations using the elimination method. If we multiply
both sides of the first of equations (6) by −3, then we get
−3x − 6y = 18
−3x − 6y = 18.
These equations are identical. So they have infinitely many solutions: all
values of x and y that satisfy the first equation also satisfy the second
equation.
You’ve now seen two cases where the coefficient matrix has determinant
zero. In the first case, the simultaneous equations had no solution. In the
second case, they had infinitely many solutions. In fact, whenever the
coefficient matrix of a pair of simultaneous linear equations has
determinant zero, one of these two possibilities must occur.
To see why, consider any two simultaneous linear equations
ax + by = e
cx + dy = f,
and suppose that the determinant of the coefficient matrix is zero; that is,
ad − bc = 0. We can assume that in each equation at least one of the
coefficients is non-zero; otherwise, the equation is of the form 0x + 0y =
constant, so the pair of equations has either no solutions or infinitely many
solutions.

263
Unit 9 Matrices

Suppose that in the first equation a '= 0; a similar argument holds if b '= 0.
Since ad − bc = 0, we have d = bc/a. Hence c '= 0, since if c = 0 then d = 0
also. Multiplying the first equation by c/a gives
cx + (bc/a)y = ce/a
cx + dy = f.
Since d = bc/a, these two equations can be rewritten as
cx + dy = ce/a
cx + dy = f.
These equations have identical left-hand sides. So they (and hence the
original equations) have either no solution (if ce/a =
' f ) or infinitely many
solutions (if ce/a = f ).
So, in conclusion, if the coefficient matrix has determinant zero, then the
equations have either no solution or infinitely many solutions.
We can deduce from this fact that if a 2 × 2 matrix has determinant zero,
then it is not invertible. This is a fact that was mentioned in
Subsection 3.2. To see how, consider any 2 × 2 matrix with determinant
zero. If the matrix were invertible, then any pair of simultaneous equations
with this matrix as its coefficient matrix would have a unique solution,
which cannot be the case, since you saw above that if the determinant of
the coefficient matrix is zero, then the equations have no solution or
infinitely many solutions.
So you have learned the following important payoffs from the matrix
method for simultaneous linear equations.

For a pair of simultaneous linear equations in two unknowns:

• if the determinant of the coefficient matrix is non-zero, then this
matrix is invertible and the equations have a unique solution
• if the determinant of the coefficient matrix is zero, then this
matrix is not invertible and the equations have no solution or
infinitely many solutions.

Thus, calculating the determinant of the coefficient matrix gives a quick

test for the existence of a unique solution.

Activity 27 Determining whether simultaneous equations have

unique solutions

Which of the following pairs of simultaneous linear equations has a unique

solution? (You are not asked to find the solution.)
(a) 5x − 3y = 5 (b) 3x − y = −6
−x + 4y = 7 −9x + 3y = 18

264
 4 Simultaneous linear equations and matrices

4.3 Simultaneous linear equations in more than two

unknowns
The matrix method that you met in Subsection 4.2 can be used to find a
solution for more than two simultaneous equations in more than two
unknowns.
In Subsection 5.3 of Unit 1, you learned that a linear equation is one in
which, after you’ve expanded any brackets and cleared any fractions, each
term is either a constant term or a number times a variable. For example,
the equation
x + 2y − z = 3
is linear, but none of the equations
x + y 2 = 2, x + yz = 1 and x + 2y = 1
are linear. In this subsection we’ll usually write a linear equation in a form
similar to the linear equation above, that is, with its left-hand side as a
sum of terms, each of which is a number times a variable (and with each
variable appearing just once), and with the right-hand side as a number.
We’ll refer to a collection of linear equations in a given set of unknowns as
a system of linear equations. For example, here’s a system of three
linear equations in three unknowns, x, y and z:
2x + 2y + 3z = 2
2x + 3y + 4z = −1
4x + 7y + 10z = 3.
A solution to a system of linear equations is an assignment of values to
the unknowns that makes all the equations hold simultaneously. As in
Subsection 3.1 of Unit 2, the process of finding these values is known as
solving the equations simultaneously. We’ll look only at systems where the
number of unknowns is equal to the number of equations.
Numerical problems that can be modelled by systems of linear equations
often arise in applications. Consider, for example, a bank that offers three
types of investment to its customers: a financial product A that pays
guaranteed interest of 2.5% over two years and is quite safe, and two
financial products B and C that in the same time period are likely to yield
interests of 5% and 6% respectively, but are more risky.
Suppose that a customer is seeking to invest £9000 over two years, and is
advised to spread his investment across the three types of financial
products. He is risk-averse, so he wants to invest a total of only £4000
across the two riskier products B and C. Of this £4000, he wants to invest
just enough in the higher-yielding product C to ensure that his total
£9000 investment is likely to yield an overall return of 4%, with the
remainder of the £4000 invested in the lower-yielding product B, to spread
his risks. How should he spread his investment across the three products in
order to achieve his desired return?

265
Unit 9 Matrices

First of all, the amounts invested in each product must add up to £9000.
So, if x, y and z represent the amounts invested (in £) in products A, B
and C, respectively, then
x + y + z = 9000.
Moreover, the investor wants to invest a total of £4000 in products B
and C, so
y + z = 4000.
Since the interest on his investment in product A is 2.5%, the investor’s
expected interest payment from his investment in product A is 0.025x.
Likewise, his expected interest payments from his investments in products
B and C are 0.05y and 0.06z, respectively. The investor is seeking an
overall return of 4%, that is, an overall interest payment of
0.04 × £9000 = £360. Thus the amounts invested in the three products
must also satisfy the equation
0.025x + 0.05y + 0.06z = 360.
Therefore, the amounts x, y and z must simultaneously satisfy the three
linear equations
x+ y+ z = 9000
y+ z = 4000 (7)
0.025x + 0.05y + 0.06z = 360.
Such systems of linear equations can be solved efficiently by methods
similar to elimination and substitution – you can probably see that
equations (7) can be solved fairly easily for x, y and z by
• first subtracting the second equation from the first to give x = 5000
• then substituting this value for x into the third equation and solving
the resulting two simultaneous equations for y and z.
Methods of this sort are covered in detail in other modules. Systems of
linear equations like those above can also be solved by using matrices, and
the rest of this subsection covers this approach.
Any system of n linear equations in n unknowns can be written as a single
matrix equation
Ax = b,
where A is an n × n coefficient matrix, x is an n-dimensional vector whose
components are the unknowns, and b is an n-dimensional vector, as you
saw for the case n = 2 in Subsection 4.2. If the coefficient matrix A is
invertible, then the matrix equation can be solved by multiplying both
sides by the inverse of A, giving x = A−1 b, just as in the 2 × 2 case.
In Section 3 you learned how to use a computer to find the inverses of
matrices of size 3 × 3 and larger. This enables you to use the matrix
method to solve systems of three or more linear equations. In the next
example, this method is used to solve equations (7).

266
 4 Simultaneous linear equations and matrices

Example 7 Solving a system of three linear equations

Use the matrix method to solve the following system of three linear
equations.
x+ y+ z = 9000
y+ z = 4000
0.025x + 0.05y + 0.06z = 360
Solution
The matrix form of the system of three equations is
    
1 1 1 x 9000
 0 1 1  y  = 4000 .
0.025 0.05 0.06 z 360

Use the computer algebra system to find the inverse of the

coefficient matrix.
The inverse of the coefficient matrix is
 
1 −1 0
 2.5 3.5 −100 .
−2.5 −2.5 100
Multiply each side of the matrix equation by the inverse of the
coefficient matrix.
Therefore the solution is given by
      
x 1 −1 0 9000 5000
y  =  2.5 3.5 −100 4000 =  500  .
z −2.5 −2.5 100 360 3500

Note that you could use the computer algebra system to solve the
simultaneous equations (7) directly, that is, without first finding the
inverse of the coefficient matrix, and so check the answer to Example 7.
The solution to equations (7) found in this example tells our investor that
the best way to spread his investment in order to achieve a 4% return is as
follows:
£5000 on investment A
£500 on investment B
£3500 on investment C.
You can practise finding inverses of matrices and using them to solve
systems of more than two linear equations in the next activity.

267
Unit 9 Matrices

Activity 28 Solving systems of three linear equations

Solve each of the following systems of simultaneous linear equations by

using the computer algebra system to find the inverse of the coefficient
matrix.
(a) 2x + 2y + 3z = 2 (b) 3x + 2y + z = 1
2x + 3y + 4z = −1 4x + 3y + z = 2
4x + 7y + 10z = 3 7x + 5y + z = 1

In fact using matrix inverses is not usually the most efficient way to solve
systems of linear equations, but the method has theoretical importance.
For example, it can be shown that, for all n ≥ 2,
• if the determinant of the coefficient matrix of a system of n equations
is non-zero, then this matrix is invertible and the equations have a
unique solution
• if the determinant of the coefficient matrix of a system of n equations
is zero, then this matrix is not invertible and the equations have no
solution or infinitely many solutions.
You saw the case n = 2 of this result in Subsection 4.2.

Example 8 Determining whether a system of linear equations

has a unique solution
Does the system
x + 3y − 2z = 12
1
2x − 8y + z = 0
7x − z=2
have a unique solution?
Solution
The coefficient matrix of this system is
 
1 3 −2
A =  12 −8 1 .
7 0 −1
Using the computer algebra system we find that det A = −81.5.
Since det A '= 0, the system has a unique solution.

268
 Learning outcomes

The following activity provides further practice on the significance of the

determinant of the coefficient matrix for systems of three or more linear
equations.

Activity 29 Determining whether systems of linear equations have

unique solutions

Which of the following systems of linear equations have a unique solution?

(a) x + 2y + 3z = 5 (b) 2x + 2y + 3z = 5
2x + 3y + z = 6 2x + 3y + z = 6
4x + 7y + 7z = 1 4x + 7y + 7z = 1
(c) x + y − w + z =0
x− y+ w+z =2
2x + 3w − z =1
6x + 2y + 4w =0

Learning outcomes
After studying this unit, you should be able to:
• use matrix notation
• evaluate sums, differences and scalar multiples of matrices
• establish whether two matrices can be multiplied together
• evaluate a matrix product both by hand and by using the computer
algebra system
• determine whether a given n × n matrix has an inverse
• work out the determinant and inverse (if it exists) of a 2 × 2 matrix by
hand
• use the computer algebra system to work out the determinant and
inverse (if it exists) of a square matrix
• write a system of linear equations in matrix form and solve it using
matrix methods where possible.

269
Unit 9 Matrices

Solutions to activities
Solution to Activity 1 Solution to Activity 4
. ./ /
We have a33 = 3, a34 = −4, a43 = −7, 4 3 12 9
a14 = −1, a44 = −9, a42 = −5. (a) (i) 3 =
2 1 6 3
The matrix A has four rows and four columns, so . / . /
1 4 6 2 3
its size is 4 × 4. (ii) =
2 2 −1 1 − 21
Solution to Activity 2 . / 0 1
2 6 x 4 23 x
. / . / . / (iii) =
2 1 4 6 6 7 3 3 y 2 23 y
(a) A + B = + =
3 0 2 −1 5 −1 . / . /
. / . / 4.5 3 9 6
2 0 −4 1 0 1 (b) (i) = 0.5
(b) A + B = + 2 1.5 4 3
−1 6 0 0 1 1 0 1 . /
. / 3
1 9
3 0 −3 (ii) 2
=
= 2
6 4
−1 7 1 3
. / . /
. / . / 2x 3x 2 3
a b 3 −1 (iii) =x
(c) A + B = + 0 5x 0 5
c d −1 2
. / In each of the answers in this part, there is
a+3 b−1
= more than one way to choose the scalar
c−1 d+2
multiple. For example, another correct
solution to part (i) is
Solution to Activity 3 . / . /
4.5 3 18 12
. / . 1 / .3 / = 0.25 .
2 −2 2 4 2 −6
2 1.5 8 6
(a) A − B = − =
1 0 −1 2 2 −2
. 1 / . / . 3 / Solution to Activity 5
2 4 2 −2 −2 6 We have
(b) B − A = − = .1 / . 3 /
−1 2 1 0 −2 2 1 2
2 2 0 −1
. 1 / . / A−C= −
2 4
0 −x 4 −2 0 −1 3 5
(c) B − C = − . /
−1 2 x 0 −1 1 3
. / = .
1
4+x 5 −5 −5
2
= Next,
−1 − x 2 . /
1 2 4
2A = .
8 −4 0
The matrix 2A has size 2 × 3 and B has size 2 × 2;
hence the sum B + 2A is not defined.
Finally,
. 3 / . /
2 0 −1
1 2 4
C − 2A = −
−1 3 5 8 −4 0
. 1 /
2 −2 −5
= .
−9 7 5

270
 Solutions to activities

 
Solution to Activity 6 . / 1 0 1
1 2 3 
For each pair of matrices in this question, the sizes (e) 1 0 −1
7 6 1
of the matrices are written underneath the matrices, 2 1 3
and the middle two numbers are boxed in order to 2× 3 3 ×3
use the method described in the text.
  The middle numbers are the same, so these
2 9 . /
  −2 3 1 matrices can be multiplied together. Their
(a) 7 1
4 0 5 product is
−6 8  
. / 1 0 1 . /
1 2 3  9 3 8
3× 2 2 ×3 1 0 −1 = .
7 6 1 15 1 4
2 1 3
The middle numbers are the same, so these
matrices can be multiplied together. Their Solution to Activity 7
product is (a) (i) Here A has size 2 × 2 and B has size 2 × 1.
   
2 9 . / 32 6 47 For AB, the size check is
 7 1 −2 3 1 = −10 21 12 .
4 0 5 2× 2 2 × 1.
−6 8 44 −18 34
  The middle numbers are the same, so the
product AB is defined.
$ - 9
(b) 1 2 3 8 For BA, the size check is
7
2× 1 2 × 2.
1× 3 3 ×1
The middle numbers are different, so the
The middle numbers are the same, so these product BA is undefined.
matrices can be multiplied together. Their
(ii) Here both A and B have size 2 × 2, so the
product is
  products AB and BA are both defined.
$ - 9 (b) For the matrices in part (a)(i), the product BA
1 2 3 8 = (46).
is not defined.
7
   For the matrices in part (a)(ii),
2 9 0 10 . /. / . /
1 1 0 1 2 4
(c)  7 1  4 1 AB = =
−1 0 2 3 0 −1
−6 8 −2 7 and . /. / . /
3× 2 3 ×2 0 1 1 1 −1 0
BA = = .
2 3 −1 0 −1 2
The middle numbers are different, so these
Therefore AB '= BA.
matrices cannot be multiplied together.
 
9 $ - Solution to Activity 8
(d)  8 1 2 3
(a) The size of A is 1 × 2 and the size of B is 2 × 2.
7 The size check for the product AB is
3× 1 1 ×3 1× 2 2 × 2.
The middle numbers are the same, so these Thus AB has size 1 × 2.
matrices can be multiplied together. Their
product is The size of C is 2 × 3. The size check for the
    product BC is
9 $ - 9 18 27
8 1 2 3 = 8 16 24 . 2× 2 2 × 3.
7 7 14 21 Thus BC has size 2 × 3.
271
Unit 9 Matrices

We have . / Solution to Activity 10

$ - 1 −1 $ - We have
AB = −1 0 = −1 1 , and . / . / . /
2 −1 1 −1 0
. /. / B+C= + = ,
1 −1 1 3 0 2 1 3
BC = so
2 −1 −3 4 1 . /. / . /
. / 1 −1 0 −3
4 −1 −1 A(B + C) = = .
= . 3 4 3 12
5 2 −1
Also,
. /. / . /
(b) From part (a), the size of AB is 1 × 2. Since C 1 −1 1 −1
has size 2 × 3, the size check for (AB)C is AB = = ,
3 4 2 11
. /. / . /
1× 2 2 × 3. 1 −1 −1 −2
AC = = ,
3 4 1 1
So the product (AB)C exists and has size 1 × 3. so . / . / . /
From part (a), the size of BC is 2 × 3. Since A −1 −2 −3
AB + AC = + = .
has size 1 × 2, the size check for A(BC) is 11 1 12
1× 2 2 × 3. Thus A(B + C) = AB + AC, as required.

So the product A(BC) exists and has size 1 × 3. Solution to Activity 11

Using the answers for the matrices AB and BC (a) The product of two matrices A and B can be
found in part (a) we obtain formed only if the number of columns of A is
. /
$ - 1 3 0 equal to the number of rows of B. Since M has
(AB)C = −1 1 two rows and three columns, the number of its
−3 4 1
$ - rows is not equal to the number of its columns.
= −4 1 1
Thus the product MM cannot be formed.
and . /
$ - 4 −1 −1 (b) The number of rows of N is equal to the
A(BC) = −1 0 number of columns of N, so the product NN is
5 2 −1
$ - defined. We
. have / . / . /
= −4 1 1 . 2 −1 2 −1 4 −4
We can see that (AB)C = A(BC). NN = = .
0 2 0 2 0 4

Solution to Activity 9 (c) A matrix A must have the same number of

We have . rows as columns in order for the product AA to
/ . / be defined. So only matrices that have the same
1 0 2 0
2B = 2 = , number of rows as columns can be squared.
2 1 4 2
so (You might remember that such matrices are
. /. / . / called square matrices.)
1 2 2 0 10 4
A(2B) = = .
3 4 4 2 22 8
Solution to Activity 12
Also, . /2 . /. / . /
. /. / . / −1 4 −1 4 −1 4 9 −4
1 2 1 0 5 2 (a) = =
AB = = , 2 0 2 0 2 0 −2 8
3 4 2 1 11 4
 2   
so . / . / 1 0 1 1 0 1 1 0 1
5 2 10 4 0 2 0 = 0 2 0  0 2 0
2(AB) = 2 = .
11 4 22 8 0 0 3 0 0 3 0 0 3
 
Therefore A(2B) = 2(AB), as required. 1 0 4

= 0 4 0
0 0 9

272
 Solutions to activities

(b) From the text

. preceding
/ this activity, we know Solution to Activity 16
20 7 (a) Since there are three input nodes and two
that A3 = . Therefore
21 6 output nodes, the network can be represented
A6 = A3 A3 = (A3 )2 by a 2 × 3 matrix.
. /2 . /. /
20 7 20 7 20 7 The number that labels the arc from the first
= =
21 6 21 6 21 6 input node (U ) to the first output node (C)
. / is 0.2, and this goes in the first row and first
547 182
= . column. The number that labels the arc from
546 183
the second input node (V ) to the first output
Solution to Activity 14 node (C) is 0.6, and this goes in the first row
and second column.
(a) Inputting 1 litre of water at node U gives
0.2 litres of water output at node C and Proceeding similarly for all the inputs and
0.8 litres of water output at node D. outputs of the network gives the following
matrix:
Inputting 2 litres of water at node V gives
U V W
2 × 0.6 litres of water output at node C and . /
2 × 0.4 litres of water output at node D. C 0.2 0.6 0.3
.
D 0.8 0.4 0.7
So, in total, we have 0.2 + 2 × 0.6 = 1.4 litres of
water output at node C, and 0.8 + 2 × 0.4 (b) The vector that represents an input of 1 litre of
= 1.6 litres of water output at node D. water at nodes U , V and W is
 
(b) Using a method similar to that of part (a) gives 1
1 .
0.2x + 0.6y + 0.3z litres of water output at
node C and 0.8x + 0.4y + 0.7z litres of water 1
output at node D. (c) The output vector for the input vector found in
part (b) is
Solution to Activity 15  
. / 1
(a) If £2000 is input at node D, then the output at 0.2 0.6 0.3  
1
node J is 0.2 × £2000 = £400. 0.8 0.4 0.7
1
Similarly, the output at node K is . / . /
0.2 × 1 + 0.6 × 1 + 0.3 × 1 1.1
0.8 × £2000 = £1600. = = .
0.8 × 1 + 0.4 × 1 + 0.7 × 1 1.9
(b) If the input at node D is £2400, at node E is This vector represents an output of 1.1 litres of
£1700, and at node F is £1100, then the water at node C and 1.9 litres of water at
output (in £) at node J is node D.
0.2 × 2400 + 0.3 × 1700 + 0.5 × 1100 = 1540,
and at node K is
0.8 × 2400 + 0.7 × 1700 + 0.5 × 1100 = 3660.
(c) If the input (in £) at node D is x, at node E
is y, and at node F is z, then the output (in £)
at node J is
0.2x + 0.3y + 0.5z,
and at node K is
0.8x + 0.7y + 0.5z.

273
Unit 9 Matrices

Solution to Activity 17 (d) Labelling the matrix that represents the

(a) The matrix that represents the network with combined network gives
nodes U , V , P and Q is U V
 
U V A 0.3 0.3
. / B 0.2 0.14 .
P 0.5 0.2
. C 0.5 0.56
Q 0.5 0.8
. / A simplified network equivalent to the original one
p
The matrix equation giving the output for is given below.
q
. /
u
the input is
v U V
. / . /. /
p 0.5 0.2 u
= .
q 0.5 0.8 v
(b) The matrix that represents the network with
nodes P , Q, A, B and C is 0.3 0.56
P Q
 
A 0.3 0.3 0.2 0.14
B 0.3 0.1 . 0.3 0.5
C 0.4 0.6
  A B C
a
The matrix equation giving the output  b  for
c (e) An input of 1 litre at node U and 2 litres at
. /
. /
p 1
the input is node V is represented by the input vector .
q 2
    Using the equation for the combined network
a 0.3 0.3 . /
 b  = 0.3 0.1 p . found in part (c), the corresponding output is
q given by
c 0.4 0.6      
  a 0.3 0.3 . / 0.9
 b  = 0.2 0.14 1
a = 0.48 .
(c) The matrix equation giving the output  b  for 2
c 0.5 0.56 1.62
c Therefore the output is 0.9 litres at node A,
. /
u 0.48 litres at node B and 1.62 litres at node C.
an input in the combined network is
v
obtained by using parts (a) and (b) as follows. Solution to Activity 18
   
a 0.3 0.3 . / (a) The matrices (with row and column headings)
 b  = 0.3 0.1 p are
q
c 0.4 0.6 D E F J K
  . / . /
0.3 0.3 .. / . // J 0.2 0.3 0.5 R 0.6 0.9
0.5 0.2 u and ,
= 0.3 0.1 K 0.8 0.7 0.5 S 0.4 0.1
0.5 0.8 v
0.4 0.6 respectively.
  
0.3 0.3 . / . /
(b) The matrix that represents the combined
0.5 0.2  u
= 0.3 0.1 network is / .
0.5 0.8 v . /
0.4 0.6 0.6 0.9 0.2 0.3 0.5
 
0.3 0.3 . / 0.4 0.1 0.8 0.7 0.5
u . /
= 0.2 0.14 0.84 0.81 0.75
v = .
0.5 0.56 0.16 0.19 0.25

274
 Solutions to activities

(c) The simplified network is shown below. Also,

    
1 0 0 x x
D E F IB = 0 1 0 y  = y  .
0 0 1 z z
0:81 0:19 Therefore AI = A and IB = I, as required.
0:16 0:75
Solution to Activity 21
0:84 0:25 . /. / . /
8 5 2 −5 1 0
(a) AB = = ,
3 2 −3 8 0 1
. /. / . /
R S 2 −5 8 5 1 0
BA = = .
−3 8 3 2 0 1
  
(d) If the input vector (in £) to the combined 2 3 1 0 1 −1
network
 is
 (b) CD = −1 1 0 0 2 −1
1700 −2 1 0 1 −8 5
1400 ,  
1 0 0
0 =  0 1 0 ,
then the output vector (in £) is 0 0 1
 
. / 1700 . /   
0.84 0.81 0.75  2562 0 1 −1 2 3 1
1400 = .
0.16 0.19 0.25 538 DC = 0 2 −1 −1 1 0
0
1 −8 5 −2 1 0
So £538 is spent in region S.  
1 0 0
Solution to Activity 19 = 0 1 0 .

0 0 1
The products AI, IA and IB are defined, and they
give the following results:
. /. / . /
3 4 1 0 3 4 Solution to Activity 22
AI = = = A,
−2 1 0 1 −2 1 (a) We have det A = 13 × 2 − 5 × 5 = 1 '= 0.
. /. / . / Thus A is.invertible/and
1 0 3 4 3 4
IA = = = A, 2 −5
0 1 −2 1 −2 1 A−1 = .
. /. / . / −5 13
1 0 2 7 −3 2 7 −3 (b) We have det B = 3 × 2 − 0 × (−5) = 6 '= 0.
IB = = = B.
0 1 8 1 2 8 1 2 Thus B is invertible
. / and
.1 /
The product BI is not defined because B has −1 1 2 0 3 0
3 columns and I has 2 rows. B = = 5 1 .
6 5 3 6 2

Solution to Activity 20 (Either of these two forms of the inverse is

acceptable as your final answer.)
We have  
. /1 0 0 (c) We have det C = −3 × (− 31 ) − 1
5 × 5 = 0.
a b c  Thus C is not invertible.
AI = 0 1 0
d e f
0 0 1 (d) We have det D = 1.5 × 1.5 − 2.5 × 0.5 = 1 '= 0.
. /
a b c Thus D is.invertible and
/
= . 1.5 −2.5
d e f −1
D = .
−0.5 1.5

275
Unit 9 Matrices

Solution to Activity 23 is 0 1. / . /
3
Using the property of matrix products given in the 5 − 45 x 18
= .
hint, we have 4 3 y −1
. /. . // 5 5
a b 1 d −b
AB = Solution to Activity 26
c d ad − bc −c a
. /. / (a) Activity 25(a) gives the matrix form of the
1 a b d −b
= equations as
ad − bc c d −c a . /. / . /
. / 2 3 x 3
1 ad − bc 0 = .
= 1 4 y −1
ad − bc 0 ad − bc
. / The determinant of the coefficient matrix is
1 1 0 2 × 4 − 3 × 1 = 5,
= × (ad − bc)
ad − bc 0 1
so the inverse of the coefficient
1 matrix is
. / . / 0 4 3
1 0 1 4 −3 5 −5
= = I. = .
0 1 5 −1 2 −51 2
5
Moreover, Hence
. . // . / . / . /. /
1 d −b a b x 1 4 −3 3
BA = =
ad − bc −c a c d y 5 −1 2 −1
. / . /
1 ad − bc 0 3
= = I. = .
ad − bc 0 ad − bc −1
Hence AB = BA = I, and it follows that B = A−1 . The solution is x = 3, y = −1.

Solution to Activity 25 (b) Activity 25(b) gives the matrix form of the
equations as
(a) The coefficient matrix of the equations . /. / . /
2 0 x 6
2x + 3y = 3 = .
3 6 y 15
x + 4y = −1 The determinant of the coefficient matrix is
. /
2 3 2 × 6 − 0 × 3 = 12,
is . Thus, the matrix form of the
1 4
so the inverse of the coefficient
1 matrix is
equations is
. /. / . / . / 0 1
1 6 0 0
2 3 x 3 = 2
.
= . 12 −3 2 − 14 16
1 4 y −1
(b) First rearrange the first equation as Hence
. / . /. / . /
x 1 6 0 6 3
2x = 6. = = .
y 12 −3 2 15 1
The matrix form of the equations
The solution is x = 3, y = 1.
2x =6
(c) Activity 25(c) gives the matrix form of the
3x + 6y = 15
equations as
is . 0 1. / . /
/. / . / 3 4
2 0 x 6 5 −5 x 18
= . = .
3 6 y 15
4 3 y −1
5 5

(c) The matrix form of the equations The determinant of the coefficient matrix is
3 4 $ 4- 4
5 x − 5 y = 18
3 3
5 × 5 − − 5 × 5 = 1,
4 3
5 x + 5 y = −1 so the
0 inverse1of the
0 coefficient
1 matrix is
3 4 3 4
1 5 5 5 5
= .
1 − 54 53 − 54 35

276
 Solutions to activities

Hence We have
. / 0 3 41 . /  −1  
x 5 5 18 3 2 1 2 −3 1
= 4 3 1 = −3 4 −1 .
y − 54 35 −1
0 1 . / 7 5 1 1 1 −1
50
5 10 Therefore
= = .       
− 75 −15 x 2 −3 1 1 −3
5
The solution is x = 10, y = −15. y  = −3 4 −1 2 =  4 .
z 1 1 −1 1 2
Solution to Activity 27 The solution is x = −3, y = 4, z = 2.
(a) The matrix form of the equations is (Details of how to use the CAS to find the
. /. / . /
5 −3 x 5 inverses of the matrices are given in the
= .
−1 4 y 7 Computer algebra guide, in the section
The determinant of the coefficient matrix is ‘Computer methods for CAS activities in
5 × 4 − (−3) × (−1) = 17. Books A–D’.)
This is non-zero, so the coefficient matrix is Solution to Activity 29
invertible and the equations have a unique
solution. (a) The matrix form of the equations is
    
1 2 3 x 5
(b) The matrix form of the equations is 2 3 1 y  = 6 .
. /. / . /
3 −1 x −6 4 7 7 z 1
= .
−9 3 y 18
The determinant of the coefficient matrix is
The determinant of the coefficient matrix is zero, so the coefficient matrix is non-invertible
3 × 3 − (−1) × (−9) = 0. and the equations do not have a unique
This is zero, so the coefficient matrix is solution.
non-invertible and the equations do not have a (b) The matrix form of the equations is
unique solution.     
2 2 3 x 5
2 3 1 y  = 6 .
Solution to Activity 28
4 7 7 z 1
(a) The matrix form of the equations is
     The determinant of the coefficient matrix is 14.
2 2 3 x 2 This is non-zero, so the coefficient matrix is
2 3 4  y  = −1 .
invertible and the equations have a unique
4 7 10 z 3 solution.
We have (c) The matrix form of the equations is
 −1  1 1     
2 2 3 1 2 −2 1 1 −1 1 x 0
2 3 4  = −2 4 −1  . 1 −1 1 1   y  2
   =  .
4 7 10 1 −3 1 2 0 3 −1 w 1
6 2 4 0 z 0
Therefore
   1 1     The determinant of the coefficient matrix is
x 1 2 −2 2 0
y  = −2 zero, so the coefficient matrix is non-invertible
4 −1  −1 = −11 .
z 3 8 and the equations do not have a unique
1 −3 1
solution.
The solution is x = 0, y = −11, z = 8.
(Details of how to use the CAS to find the
(b) The matrix form of the equations is determinants are given in the Computer algebra
    
3 2 1 x 1 guide, in the section ‘Computer methods for
4 3 1 y  = 2 . CAS activities in Books A–D’.)
7 5 1 z 1

277
Acknowledgements

Acknowledgements
Grateful acknowledgement is made to the following sources:

Page 219: Taken from: http://en.wikipedia.org/wiki/File:Ms sf 2.jpg. This

file is licensed under the Creative Commons Attribution-Share Alike
Licence http://creativecommons.org/licenses/by-sa/3.0/
Page 221: Courtesy Rob Cottingham. www.robcottingham.ca
c Unknown
Page 238: "

Every effort has been made to contact copyright holders. If any have been
inadvertently overlooked the publishers will be pleased to make the
necessary arrangements at the first opportunity.

278
 Index

Index
acceleration due to gravity 70 determinant of a matrix 255
addition of matrices 223 differentiation
properties of 228 of a composite function 25, 27
antiderivative 56 of a function of a linear expression 31–32
change in the value of 75, 146 of a product 14
antidifferentiation 56 of a quotient 19
arbitrary constant 57, 151, 154, 164 of an inverse function 51
arc of a network 239 using a computer 44
dimension of a vector 222
Barrow, Isaac 145 displacement 146
Bernoulli, Jacob 125 dummy variable 122
Bernoulli, Johann 125
element (entry) of a matrix 221
Cauchy, Augustin Louis 145 equivalent networks 246
Cayley, Arthur 223 exp, derivation of derivative of 10–11
chain rule
in Lagrange notation 30 Fourier, Joseph 126
in Leibniz notation 25 function of a linear expression
without introducing an extra variable 27–30 differentiating 31–32
coefficient matrix of simultaneous equations 258 integrating 169–170
column of a matrix 221 function, continuous 56, 106
component of a vector 222 fundamental theorem of calculus 132, 134
composite rule see chain rule history 145
constant function, indefinite integral 66
constant multiple rule greatest value of a function on an interval 39
for antiderivatives 64
for definite integrals 140 identity matrix 251
for indefinite integrals 153 identity, trigonometric 189
for square bracket notation 138 indefinite integral 57, 151
constant of integration 57, 151, 154, 164 notation 151
continuous function 56, 106 of a constant function 66
cos, derivation of derivative of 7–8 of a function of a linear expression 169–170
cube of a matrix 237 of a power function 61
of ln 183
decomposing a composite function 24–25 of tan 166
definite integral 119, 124, 126 of the reciprocal function 80
evaluating 124, 135 standard 82
properties of 120 integral (see also indefinite integral) 152
derivative definite 119, 124, 126
of a composite function 25, 27 indefinite 151
of a function of a linear expression 31–32 standard 82
of a product 14 integral notation 119, 152
of a quotient 19 integral sign 119
of an inverse function 51 history 125
of exp and ln 10–13 integrand 136, 152
of sin, cos and tan 4–9
standard 54

279
Index

integration 56 negative of a matrix 226

and rates of change 68–73 network 238, 239
by parts 176–188 combining 245
by parts, for definite integrals 186 equivalent 246
by substitution 155–176 matrix representing 242
by substitution, for definite integrals 173 node 239
choosing a method 192 network diagram 239
different correct answers 191 Newton, Isaac 145
of a function of a linear expression 169–170 node of a network 239
of a power function 61 non-invertible matrix 255, 257
of a trigonometric expression 189 non-singular matrix 257
using a computer 194
integration by parts formula optimisation problem 38
for definite integrals 186 strategy for solving 44
in Lagrange notation 177, 188
in Leibniz notation 188 parts, integration by 176–188
stated informally 179 for definite integrals 186
inverse function rule power function
in Lagrange notation 51 indefinite integral of 61
in Leibniz notation 47 proof of derivative of 37
inverse of a matrix 253, 254 power of a matrix 238
invertible matrix 253, 255 product rule
in Lagrange notation 14
leading diagonal of a matrix 252 in Leibniz notation 16
least value of a function on an interval 39 proof of 18
Leibniz, Gottfried Wilhelm 125 stated informally 16
limit 109 projectile, vertical 72
limits of integration 119
quotient rule
ln, derivation of derivative of 12–13, 48
in Lagrange notation 19
ln, indefinite integral 183
in Leibniz notation 20
proof of 37
matrix 221
stated informally 21
addition 223
determinant of 255 reciprocal function, indefinite integral 80
element (entry) of 221 row of a matrix 221
identity 251
inverse of 253, 254 scalar 226
invertible 253 scalar multiplication of matrices 226
multiplication of 228–233 properties of 228
negative 226 signed area 111, 115, 116
power of 238 signed-area-so-far function 127, 131
product 228 simultaneous equations
properties of 228, 236, 254 matrix form of 258
scalar multiplication of 226 solution of 265
size of 221 solving using matrices 261, 266
square 221 system of 265
square of 237 unique solution of 262, 264, 268
subtraction of 225 sin, derivation of derivative of 4–6
zero 226 singular matrix 257
matrix form of simultaneous equations 258, 266 size of a matrix 221
maximisation/minimisation problem 38–39 solution of simultaneous equations 265
multiplication of matrices 228–233 square bracket notation 134
properties of 236 square matrix 221
square of a matrix 237
280
 Index

standard derivatives 54 system of linear equations 265

standard indefinite integrals 82
subinterval 107 tan, derivation of derivative of 8–9
using to approximate a definite integral 113 tan, indefinite integral 166
substitution, integration by 155–176 trigonometric expressions, integrating 189
for definite integrals 173 trigonometric identities 189
subtraction of matrices 225
sum rule unique solution of simultaneous equations 262
for antiderivatives 64
for definite integrals 140 vector 222
for indefinite integrals 153 vertical projectile 72
for square bracket notation 138
Sylvester, James Joseph 223 zero matrix 226

281