Mathematics for A and AS level

Includes
4998 common core

ure mathematics

SMP CAMBRIDGE
Mathematics for A and AS level

Pure mathematics

1001 Mathematics Project

PUBLISHED BY THE PRESS SYNDICATE OF THE UNIVERSITY OF CAMBRIDGE
The Pitt Building, Trumpington Street, Cambridge CB2 1RP, United Kingdom

CAMBRIDGE UNIVERSITY PRESS

The Edinburgh Building, Cambridge CB2 2RU, United Kingdom
40 West 20th Street, New York, NY 10011-4211, USA
10 Stamford Road, Oakleigh, Melbourne 3166, Australia

© Cambridge University Press 1997

First published 1997

Printed in the United Kingdom at the University Press, Cambridge

Typeset in Sabon 10/124pt

A catalogue record for this book is available from the British Library

Adapted from the original 16-19 Mathematics material by Chris Belsom

Main authors Simon Baxter Barrie Hunt Paul Roder

Stan Dolan Mike Leach Jeff Searle
Tim Everton Tim Lewis David Tall
Doug French Lorna Lyons Brian Wardle
Andy Hall Tom Patton Thelma Wilson
Ron Haydock Richard Peacock Phil Wood

Team leaders Stan Dolan Barrie Hunt

Ron Haydock

16-19 project director Stan Dolan

ISBN 0 521 56617 7

Acknowledgements
The publishers would like to thank the following for supplying photographs:

page 78 (1) John and Irene Palmer/Oxford Scientific Films; BATH SPA UNIVERSITY
COLLEGE
(r) David Thompson/Oxford Scientific Films; NEWTON PARK LIBRARY
page 81 The Ancient Art and Architecture Collection Class No.
(Ronald Sheridan’s Photo Library);
page 117 (1) ESA/PLI/Science Photo Library;
aT Sis SCH
(r) | Simon Fraser/Science Photo Library;
page 310 George East/Science Photo Library

Ar CCHN OAS
Cf. TUD.CUUA

Notice to teachers
It is illegal to reproduce any part of this work in material form (including photocopying
and electronic storage) except under the following circumstances:
(i) where you are abiding by a licence granted to your school or institution by the
Copyright Licensing Agency;
(ii) where no such licence exists, or where you wish to exceed the terms of a licence, and
you have gained the written permission of Cambridge University Press;
(iii) where you are allowed to reproduce without permission under the provisions of
Chapter 3 of the Copyright, Designs and Patents Act 1988.
 Contents

Introduction

Foundations
el Graphs
1:2 Sequences 17
is Functions and graphs 40
1.4 Expressions and equations 7
te Numerical methods 76
Miscellaneous exercise 91

Introductory calculus 94
2.1 Rates of change 94
Dis Gradients of curves 102
Pas) Optimisation 124
2.4 Numerical integration 136
20) Algebraic integration 154
Miscellaneous exercise 174

Functions 178
Spi Algebra of functions 173°
Oe Circular functions 202
3.3 Growth functions 224
3.4 The number e 246
3D Radians 258
3.6 Transformations DHA
Miscellaneous exercise 278

Problem solving 283

4.1 Mathematical enquiries 283
4.2 Proof 298
4.3 Mathematical modelling 309
4.4 Mathematical articles 318
Case studies 321

Mathematical methods SE
oof ‘The power of Pythagoras B37
Je Vector geometry 364
BY) Binomials 395
Miscellaneous exercise 410
 Calculus methods 1 412
6.1 The chain rule 412
6.2 Parametric form 426
6.3 Further differentiation techniques 442
Miscellaneous exercise 456

Calculus methods 2 458

Well Integration techniques 458
Uh Polynomial approximations 481
Ts) Differential equations 498
Miscellaneous exercise 520

Supplementary tasksheets 523

Extension tasksheets 540

Solutions to exercises 570

Supplementary tasksheet commentaries 711

Extension tasksheet commentaries 725

Appendix
Core A and AS level formulas He|

Index 754
Introduction for the student

The material in this book provides a suitable preparation for the pure
mathematics content of most A level and AS level courses in mathematics, and
includes the core content of all new syllabuses, as implemented beginning in
1998. It is based on six units from the SMP 16-19 Mathematics course —
Foundations, Introductory calculus, Functions, Problem solving, Mathematical
methods and Calculus methods.

The book is a self-contained resource, consisting of explanatory text and

exercises. All exercises are provided with solutions at the end of the book. The
textual material is written in such a way that you yourself become involved in
the development of the ideas; it is a text to be worked through, rather than read
passively. You learn mathematics by actually doing it yourself, and this is
constantly encouraged through the text.

Throughout the body of the text, as material is being developed, you will meet
blocks of questions indicated as follows:

1 (a) Explain how the details of the graph...

The questions in these development sections are designed in such a way that
ideas are opened up, explored and developed before results or observations are
formalised. They are a crucial part of the learning process, making you more
familiar with the ideas that you will eventually apply in more straightforward
and conventional exercises. Answers, or more detailed solutions, are provided at
the back of the book both for these sections and the exercises. You should check
your work as you go along, correcting as necessary. Do not be tempted to look
at the solutions too readily when you encounter a problem — wrestling with a
difficulty is a better way of coming to terms with it than giving up too early.

At various points in the text you will also be directed towards a number of
tasksheets, located just after the body of the text. These contain either extension
material (and are labelled E tasksheets, for example 2.3 TS E1) or supplementary
support material (labelled 2.3 TS $1, for example). These are important in that
they provide extra material for you to study. However, they are additional to the
main flow of the text, and may be omitted if appropriate. Extension tasksheets
provide further enrichment material or ideas, and if you are able to cope with
‘the mainstream material confidently enough, then you should certainly attempt
them. Supplementary tasksheets provide extra help at certain points if needed,
for instance in solving inequalities or in extra differentiation practice. You need
not do these if you are confident with the ideas that they take you through.

There is also a complete revision guide for the 16-19 Mathematics course,
SMP 16-19 Revision, which you will find very helpful as you prepare for
examinations. The revision book summarises all of the material and important
results, chapter by chapter, and provides further revision examples (with
solutions).

Presentation and precision

Before you start your work in mathematics, a timely word about presentation is
in order! Mathematics is often described as a language, and it has its own
grammar which should be practised and rigorously applied. Mathematics is a
means of expressing and communicating ideas. You should develop good habits
of presentation and use of appropriate notation, so that your work can be easily
followed by others, with each statement correctly and logically following from
previous work. It is particularly important not to be careless with notation. For
example, sloppy use of the equals sign must be avoided: equals in mathematics
means precisely that! What is on one side of the equals sign exactly and precisely
equals what is on the other — no more, no less. Again, in using the implies
symbol >, you know for example that x = 2 implies that x* = 4, but it is not
true that x* = 4 implies x = 2 (x might be —2). So
: 2 :
x=2 => x =4 is correct, but Be ES OG is wrong.

Look at your presentation: can it be followed easily by someone else? Is it

mathematically precise, with the correct mathematical language? The ability to
communicate effectively is an important and valuable skill, in mathematics itself
and more generally.

Mathematics has been described as one of the greatest of all human intellectual
achievements; its study is a challenging but rewarding activity which requires
much hard work and practice. We hope that you will enjoy the challenge and
become confident with the skills developed, so that you may use mathematical
ideas in your further studies or in employment, either within mathematics itself,
or in the countless areas to which it can be applied.

vi
 Foundations
.1 Graphs
1.1.1 Introduction
Drawing and sketching graphs is an important and useful skill in mathematics,
often helping to illustrate an idea more effectively, or as an aid to solving
problems generally.

It is important, therefore, to be fully conversant with graphs of important and

common functions, and we concentrate initially on the graph of the straight line,
followed by the graphs of quadratic functions.

The following example considers a problem where the solution is greatly

simplified by drawing appropriate graphs.

@ 1
A market trader finds that she can sell 60 transistor radios each week if she
reduces her profit margin to zero, but sales drop when she increases her price.
In fact, at £6.00 profit per radio she sells none at all. What profit margin
should she choose to achieve the greatest possible total profit?

Number sold
per week «,,
60

To help decide what profit Bitsae

margin to choose, the trader Ea au aati
oN i
models her sales figures with
the straight-line graph shown.

Profit (£)
per radio

1 (a) Use the graph to copy and complete this table.

Profit (£) per radio 0 1 y) 3 4 Se 56)

Number of radios sold 60 0

Total profit (£) from sales 50

2 FOUNDATIONS
Ce ee ee

(b) Draw a graph from the table, plotting total profit against profit per radio.
(The solution to the trader’s problem should now be quite easy to see.)
(c) What profit per radio should she choose?
EESTI
2 Suppose the profit is £x per radio.
WASSER

(a) Find the number of radios that the trader sells per week in terms of x.
(b) What is her total profit from sales in terms of x?
(c) What are the equations of the straight-line graph shown on page 1
and the graph that you have drawn?
ee

When considering the graphs of functions it is very useful to recognise certain

properties of the graphs, which not only provide an aid to sketching graphs in
general (‘curve sketching’), but also give an insight into the properties of the
function and to relationships between functions. These ideas are explored more
fully throughout the course.

In the following work you should use a graph plotting calculator (or computer
package) to explore properties of the graphs of a variety of functions. Many of
the functions will be new to you but you will soon become more familiar with
them.

1B

| 1 Use a graph plotter to draw each of the following basic functions and their
| graphs. Sketch them and note any features which you think are interesting.
Look out for and note any features such as reflection or rotational
Sanna
(a) y=x (b) y= x" (c) y=x° (d) y= x"
(© y=ve y= @y=- Wy=5
(i) y=sinx () y=cosx (k) y=tanx (I) y=loex
(m) y=3 (nr) y= (0) y=) y= ine (x)
4
2 Plot the graphs of y = —x, y = —x*, y= —x°? and y=-x
How are they related to the graphs of question 1?
| 3} 3
(Note: —x° means square first, then change sign. : :
This can cause confusion
sometimes with a graph plotter, and it may be necessary to include
2 «
brackets, —(x*), to ensure the correct meaning.)

| 3 Compare and comment on the graphs of y = x* and y = /x, and y = x°

and y = </x.
@
 GRAPHS 3

1.1.2 Linear graphs

When a function has a straight-line graph, it is said to be linear.

Examples of straight-line graphs may be found in a wide range of subjects.

(a) The distance against time graph of a body moving with constant speed is a
straight line.
(b) Economists often assume that the demand for a commodity decreases linearly
with price.
(c) The amount by which a metal rod expands when it is heated is proportional
to its temperature and so the graph of length against temperature is linear.

Linear graphs are simple to deal with. Sometimes you can approximate more
complex graphs by linear graphs, as in the example which follows.

AiG

A long-distance walker aims to cover the 800 miles from John O’Groats to
Land’s End at the rate of 30 miles per day.

The graphs below illustrate his progress after t days.

Distance from Distance from

John O’Groats Land’s End
s (miles) s (miles)

800 800

600 600

400 400

200 : 200

s 10 20 30 0 10 20 30
t (days taken) t (days taken)

(a) What distance has the walker covered after one day and how far is he
from Land’s End?

(b) How far has he walked after two days and how far is he from Land’s End?
(er How far has he walked after t days and how far is he from Land’s End?

(d) What are the equations of the two graphs of distance against time?
4 FOUNDATIONS

Both graphs have the following features.

e They are both straight-line or linear graphs. For this reason, 30¢ and 800 — 30t
are called linear functions of t. ;
e The graph of s = 30¢ passes through the origin and has a gradient of 30.
e The graph of s = 800 — 30t crosses the s-axis at 800. The number 800 is called
the intercept on the s-axis. This function is decreasing, the graph having a
negative gradient of —30.

An introduction to straight-line graphs is provided in supplementary tasksheet S1.

Tasksheet $1 — The equation of a straight line (page 523)

There are a number of ways of finding the equation of a line if you know a
couple of points on the line. A method which is quick and easy to use is
illustrated in example 1.

Example 1
Find the equation of the straight line through (1,2) and (3, 8).

Solution

Gradient = aes or g aS
w= 1 ) y-2

So ae = g
Gaal Be 2
=> y—2=3(x—1) (= means ‘implies’) (, 2)
 GRAPHS 5

1.1 Exercise 1
1 Draw the straight lines whose equations are:

(a) y= 2x +4 (b) s+2t=7 (c) 2y—4x+3=0

(d) x+y=4 (eles (byeae=="3

2 Write down the equation for each of the following straight lines.
(jee a
Wee

i
=oSele DEAT) We too ae
3 Find the equations of the straight lines passing through the points:
(a) (0,3) and (2,7) (b) (1,4) and (2,6)
(c) (1,-5) and (—4,0) — (d) (1,2) and (—2, 1)
4 Which of the following equations have straight-line graphs?
(a) y=x-2 (b) 2y+x=4 (c) x+y =9
(d) xy
=5 (e) x= 5—2y (f) x-y=2x-—y+6

(e) y-3=4@41) (A) S45=1 0) vy=1tv2

5 Find the equation of the straight line:
(a) of slope 4 which passes through the point (2, 1)
(b) which passes through (2, 5) and is parallel to y= 3x+1
(c) which passes through (2,3) and is parallel to 2x + 3y +4 =0
6 FOUNDATIONS

1.1.3 Quadratic functions

\/
vy

<~ —_ —
FS SS Ee
iF

Parabolic reflector Parabolic path of body under gravity

You can see the curve known as a parabola in various everyday situations and it
; : : ‘ D
arises in mathematics most simply as the graph y = x’.

The graph of y = x” has line symmetry in

the y-axis, and a vertex at the origin. In this
case the vertex is the minimum point. For
inverted parabolas it would be the
maximum point.

0 4

Polynomial functions of x are functions formed by adding together powers of x.

Examples are:
3x1

x
2 + 5x —-—2

Sx 3x 2x 7e 5
(Note: the powers of x must be non-negative whole numbers.)

You have already looked at linear functions where the highest power of x
present is the first power, x! (or just x). In this section we will concentrate on
quadratic functions of the form

y=ax’+bx+c (a # 0)

where the highest power is x*. If a = 0, then this is a linear function.

You should use a graph plotter to confirm that the graph of ax* + bx +c is
always a parabola, whatever the values of a, b and c (a 4 0). Note, for example,
the effect of making a negative.
 GRAPHS 7

We will now look at some of the simplest transformations of y = x* and find

that y = ax’ + bx + c is not always the most helpful form of the equation of a
quadratic function.

el

1 Plot the graph of y = x and superimpose the graph of y = x + 3.

2 Superimpose various graphs of the form y = x* + q, for both positive and

negative values of g. Write down your conclusions.

3 Plot the graph of y = x* and superimpose the graph of y = (x + 4)’.

Describe carefully the relationship between the graphs.

4 Repeat question 2 for the family of curves with equations y = (x + p)’.

5 (a) Describe carefully how any curve of the form y = (x + p)’ +q is

related to the curve y = x’.

(b) What are the coordinates of the vertex of the parabola

y=(x+p) +@?
(c) What is the equation of its line of symmetry?

6 Suggest possible equations for the following curves. The curve y = x” is

shown dotted in each case and the coordinates of the vertex are given. Use a
graph plotter to check your equations.
8 FOUNDATIONS

7 Sketch the curves y = x” — 4 and y = 4 — x”. What are the coordinates

of the points where they meet?

8 Sketch the curves y = (x — 3)* + 2 and y = 11 and find the coordinates

of the points where they meet.

9E The curve y = ax’ passes through the vertex of y = (x + 2)’ + 3. What

is the value of a?

10E The curves y = (x — a)? +b and y = (x — cc)? +d never intersect. By

comparing their graphs, find what relationships exist between any or all
of a, b, ¢.and d,

Example 2
Sketch the graph of y = (x + 3)? —7.

Solution
The graph of y = (x + 3)? —7 is
obtained by translating the graph of
=>
y = x* through

As the vertex of y = x” is at (0, 0),

you can see that the vertex of
y= (x +3)" 7 1s aloe
 GRAPHS 9

1.1.4 Completing the square

We shall now investigate the relationship between the two forms y = x* + bx +c
andy=(x +p)? +4.
You may find it helpful to revise how to multiply brackets quickly and easily,
and in particular how to calculate squares. Practice is provided on tasksheet S2 if
needed.

.1 Tasksheet $2 — Multiplying brackets (page 526)

Example 3
Multiply out: (a) (« + 5)(x — 3) (Oy aye

Solution
(eNe ese Ne asee)
Sr
axer 5a 15
=x" + 2x = 15
(b) (+3) £1 = (x —3)\(x—3)41
=x? —6x+9+41
= x*— 6x+10
You have seen that the graphs of (x + p)* +q and x* + bx + c are always
parabolas. When written in the form (x + p)* + q, it is easy to locate the vertex
and sketch the quadratic. In the following questions you will see how the two
forms are related and how to express x* + bx +c in the form (x +p)’ + q. This
is known as completing the square.

LE

1 (a) Plot the graph of y = x” and superimpose the graph of y = x” + 2x.

(b) Consider the vertices to decide what translation maps y = x” onto

y = x” + 2x. Hence write x* + 2x in the form (x + p)’ + q.

2 Taking values of b as: (a) 10 (b) -6 (c) 7

(i) plot the graph of y = x” and superimpose the graph of y = x” + bx;

(ii) write down the translation which maps y = x” onto y = x” + bx, and
hence write x” + bx in the form (x + p)* + q.

3 (a) On the basis of your work in question 2 write x” + 4x in the form

(x+p) +4.
10 FOUNDATIONS

(b) Hence write x? + 4x +9 in the form (x + p) + q

(c) Check your answer to part (b) by plotting the graphs of x +4x +9
and (x +9)" £4 7
4 Write the following in the form (x + p)* +4
fa) x tl4x2 (bla 82s (ec) x — Se ed

5 (a) If x* + bx =(x+p)’ +4, express p and q in terms of b.

(b) Describe how you would express x” + bx +c by completing the
square.

6 (a) Express x” — 2x + 1 in completed square form.

(b) Explain why x? — 2x + 1 can never be negative.

7 Write in completed square form and hence sketch (without the aid of a
graph plotter) the graph of:

a) x 4x — 3, (b) x bast ea a Ss

8E Complete the square and hence solve the quadratic equation

2x* + 6x -9=0
Sketch a graph to illustrate your solution.

y=(x+p)+q and y = x* + bx +c are equivalent forms for the quadratic

function, and both give useful information — about the graph of the function, for
example. The relationship between the two forms can be shown algebraically.

(xt+ py +q=x°+2lpx+p
+4
Comparison with x*+bx+c>b=2p or p= 5b

i sioh not
beallJae ics ar
__ express oS. in ‘theform be+p)
 GRAPHS 11

Example 4
Complete the square on x* + 6x +2.

Solution
halve square

x
(ee 9
6x = (4 23)
x én +2 = (x1 3)-— 9 42
DeGroot)a7,

1.1 Exercise 2

1 For each of.the following equations:

(a) y=x? + 8x45 (b) y=x"
—4x —3
(c) y=x?-—Sx+6 8 (d) y=x*—7x—-3
(i) complete the square;
(ii) check your answer by multiplying out;
(iii) without using a graph plotter, sketch the corresponding graph.
2 (a) Write down a possible completed square form for the quadratic functions
whose graphs are shown below.
(b) Hence write down the equations of the quadratic functions in the form
y=x +bxt+e.

(i) (ii)

(2, —S)
12 FOUNDATIONS

1.1.5 Zeros of quadratics

The axis of symmetry and the vertex of the parabola are important features of
the graph, but they are by no means the only ones. Also useful are the points at
which the graph crosses the axes, if it does! The graphs of all quadratic functions
of x must cross the y-axis, but some graphs, such as those below, do not cross
the x-axis.

Vie

For the graph of any function, the point of

intersection with the y-axis is easily found
by putting x = 0 in the function. The values
of x at the points of intersection with the
x-axis are called the roots of the equation
y = 0, and are also known as the zeros of
the function because they make the function
equal to zero. The zeros of x” = x — 6 are
—2 and 3.

If the product of two numbers a and b equals zero, then either a or b (or both)
are zero. This means that it is possible to solve easily equations written in
factorised form (i.e. written as a product of factors).

Example 5
Solve the equations: (a) x(x+1)=
(b) (x —3)(2x+7) =
c) x + 6x =0

Solution
(@) x<@4
1) 20> S220 or x= 1
i.e. x = 0 and x = —1 are both solutions to the equation x(x + 1)=0
(orx?+x= 0).

(b) (x-—3)(2x+7)=0 3S x=3 or x=—$

(c)x" + 6 Om wae)
ee Oa
 GRAPHS 13

The ease with which equations may be solved when written in factorised form is
a major reason why finding the factors of algebraic expressions is important.
Equally, factorised form gives important information about the graph of a
function.

LF

1 (a) Plot the graph of y = (x + 1)(x +5). What is the significance of the
numbers 1 and 5 with respect to the graph?

(b) Investigate the graph of y = (x + a)(x + @) for various values of a and

G, including positive and negative values, zero and the case where
a = 2. What is the significance of a and ( for the graph?

2 What is the relationship between the graphs of

y=(xta)(x+6) and y=—(x+a)(x+ 8)?
- The quadra i ¢ expressions above are in factorised
Yi
form. _
Z,

3 Suggest possible equations for the following curves and use a graph plotter
to check your answers.

(a) (b) (c)

14 FOUNDATIONS

YR

= a Ss5 S < © 8 as S) 3 So~~8 a aN ray S £ S ~ ) make neat sketches of the graphs of:
~ ~_
_—
—

oy (x+2 )(x-1) y y =x +5x+6

 GRAPHS 15

1.1.6 Factorising quadratics

To find where the graph of a quadratic function crosses the x-axis it is helpful to
have the quadratic expression in a factorised form such as (x + 1)(x + 5). You
must therefore be able to factorise the form ax* + bx + c whenever this is
possible. Initially we shall only consider examples with a = 1. Tasksheet S3 gives
you practice at factorising quadratic expression if needed.

The relationships illustrated here for +18 = (3) x (=6)

the quadratic x* — 9x + 18 4 A A
enable you to factorise x = 9x £18 = (x= 3)(x —6)
quadratic expressions which y ‘f y
have integer roots. —9= (-3)+(-6)
=> x? = Ix + 18 = (x — 3)(x — 6)

i Tasksheet $3 — Further factorisation (page 528)

/.) Exercise3

1 Factorise the following. You may check your answers by using a graph plotter
to plot the graphs of each of the two forms.

(a) x +7x +12 (Dea = 23 (c) x*—

7x +10
(d) x —4 (e) x*—7x (f) x7 —6x+9
(g) x7 43x42 (hy) 4: x = 49

2 Use a graph plotter to plot the graphs of:

(a) y=x? +3x42 (b) y=x°+x-1 (c)ysx%—x+4
In each case determine whether

(i) the quadratic function has any zeros;

(ii) the quadratic expression can be factorised.

3 Solve the following equations by factorisation.

(a) 2x-—3x7=0 (b) x? +x=90
(c) a 9x (d) x4 3x" + 2x = 0

4 Simplify (x — 1)(x — 5) + (x — 1)(x + 2). Hence solve the equation

(x —1)(x —5)+ (x—1)(x+2) =0

5 Find the two times when a projectile is at a height of 140 m, if the height 4
metres, after time t seconds, is given by the expression h = 40t — St” + 80.
16 FOUNDATIONS
ee
eet
&

6E A rectangular garden 12m long and 10m wide consists of a rectangular lawn
bordered with flower beds of the same width round the two longer sides and
one of the shorter sides. If the area of the flower beds is half the area of the
garden, what is the width of the flower beds? :
7E 100m of fencing form three sides of a rectangular enclosure of area 1200 m°*.
Find the dimensions of the enclosure.
8E A rectangle has a perimeter of 46cm and an area of 120cm’. Using a
quadratic equation, find the dimensions of the rectangle.
 17

Foundations

1.2.1 Introduction — notation

Sequences, or lists of numbers, occur in many contexts. From an early age we
learn simple sequences such as the 5-times table
Sue tol as Oars. 30)
or we look for patterns in puzzles such as
real, uth Oy eRe Shan A
To aid the description of a sequence, certain notations are used. If U is the
sequencesy ome) TM 1S) TO. yee: then
u, = 3 is the first term,
u, =7 is the second term,
us = 19 is the fifth term, and so on.

@ 2
1 Uisthe sequence 2,-—8;° 14, 20,
~ (a) What are the values of u, and u4?

(b) What value would you expect for us?

Give an equation connecting u; and u4.

(c) Give an equation connecting u,;,, and 4;.

2 T is the sequence defined by t,; = 4 and tj,, =t;+9.

(a) What are the values of t), 3, t4, ts?

(b) What is the value of ty?

_ requires:
(i) a starti
ecurrence relation, i.e. a formula which
will generate any term fro:
or terms,for example 4j,,;=uj+6.
18 FOUNDATIONS
ee Saree Se nea ee ee ee ee ee ee eS

Example 1
Find an inductive definition for the series 1, 2, 6, 24, 120,

Solution
u, = 1 and the pattern is then
iy = 2 CU USS dD, Ug 4a, © 20d
60) OD,
So uy = il, Uj44 a (G+ 1)u;.

Some properties of sequences are investigated below.

2B

1 For each sequence write out the first five terms and the value of the 20th
term.
/
(a) i and W047 =U; +2 (b) iy = NS) and Ui, Su — 4

(Clipe van 4. Oe: (d) uw, =3 and 444 =—

u~
We can use diagrams to illustrate the behaviour of sequences.

: :
: 34S = 2
: x
x
2 x
|

|
i 1
:

| mae 2 7 a a Gia ae ae
| 0 1 2 3 4 5 i

/ Be, 24, 23, 22, 22, ... converges.

'
=

| 2h
i x
i 30

|
i 20
i x

i 10
i x
i x
: x
T Ta ah —>
| 0 1 2 3 4 5 i
i
2, 4558, 1 16," 32). eediverces:
 SEQUENCES 19

2 For each of the sequences below, sketch a diagram to illustrate its

behaviour (obtain sufficient terms to be able to describe the behaviour or
pattern of the sequence). You may find it interesting to’ use a variety of
starting values.

(a) Uj44 = —2u; (b) Uj44 = 2u; + 4 (c) Uj44 = (i + 1)u,;

1 pe
(d) i ace pa (e) fo Se (f) ujaq4
= 4j+ 2i+1
1

3E For different starting values, obtain a sufficient number of terms to enable

you to describe the behaviour or patterns of each sequence.

(a) Ujon = Uj + Uj44 (You will need starting values for both u, and u).)

20 33, (You will need to generate a considerable

(Bh eee
2u; number of terms to be sure of the pattern.)

4E Investigate the following sequence, using various whole-number starting

values.
3s; +1, when s; is odd
Sarg ei,
a =, when s; is even
2
You are recommended to begin with starting values of between 1 and 10.
What happens with numbers greater than 10?

The various types of sequence can be classified as follows.

Convergent

The values get closer and closer to a fixed value.

The values of this convergent sequence oscillate back

and forth about one value.
 20 FOUNDATIONS
a a i

Divergent
Any sequence which does not converge to a fixed value is called divergent.

Ui;

This sequence diverges to +00. The values grow in

size, eventually becoming infinitely large.

This sequence is periodic. A set of values is repeated at

regular intervals.

This sequence is both oscillatory and divergent.

1.2 Exercise 1

1 Write down the first five terms of the sequence U where u;4, = 2u, and u,; = 4
and describe the properties of the sequence.
2 Which of the following sequences converge?
1
(a) w441= ria u,=9 (b) eS uy =2 (c) Binary u,=1
~

3 %
2 x x

The diagram above illustrates the first five terms of the sequence U where
Uj414 = Ui; SF (4)' and “uy = 1

Describe fully the properties of the sequence.

 SEQUENCES 21

4 Investigate and describe the sequence T where t;,. = t; + t;,, and

ty = ik ty = My,

5 Investigate and describe the sequence T where

; wv
foal a ae) 5 and t; =
(t;)

6 Find inductive definitions for the sequences:

(a) by i, a - aes ee (b) ile —}NS) Ale co anlon

1.2.2 The general term

During a training run, a coach makes an athlete run at 6ms -. Let s, metres be
the distance covered in t seconds, so that s; = 6.

@ 2x
1 (a) What is S59?

(b) What is s,?

(c) Why is it inappropriate to use an inductive method to calculate s59?

Clearly, there are drawbacks if only inductive definitions are used to generate
the terms of a sequence. It can be very useful to have a formula for the
general term.

2 (a) Why is 2 x 3”~! the general term of the sequence T where

tj44 => 3t; and ty —

(b) What are the terms of the sequence U where

pl
u; = (=) 2?

Example 2
Find an expression for the ith term of the sequence —1, 3, —S, 7,
 22 FOUNDATIONS
nnn LEE eEEEEEE

Solution
It is helpful to think of the terms as
Sie | el sey eel ea SIL 7,

The. —I; +1, =1, -F1, ...- sequence is: generated: by (eae
The 1, 3, 5,7, see “sequences generated [by 27
So u; = (—1)'(2i — 1).

1.2 Exercise 2

LiyeThe set of patterns above can be made using matchsticks. How many
matchsticks would be required for the next pattern in the sequence? Copy and
complete the table.

No. of triangles bo ode OB: et 5 So) Os 220s SohOC Rees

No. of matchsticks

2 Copy and complete the table below for each of the following patterns of dots.

Position in pattern |SO? Pstae Pee,oe ae Re) Mary SeAV OE,

No.
of dots

e
e ®
(a) e, eee eesece : ececcee pa eee

ecce
ee sae
(b) eo. ece SOeen tr sae
e e
e e e e
(c) as or Soe ei49 . cia ve

r)
(d) e, oars ‘ cage , tacoVe

eo
(e) ee , wees ; eee ee ;
e0e e000 re a
eo

ee00e
(f) ee6 cece Sereeietu ee
 SEQUENCES 23

3 Write out the first five terms of the sequence whose ith term is:
(a) w=3i+2 (byw =SxZ (c) 4; = 30
4 (a) Write out the first five terms of the sequence whose ith term is:

(i) 4; = (—1)'
fin ey

(Gin) ea (221))"2
iva 12a

(b) Write down the general term of the sequence:

nse ea as haa oe
ee eee hes © hee
5 For each of the following sequences complete the table.

Term
1 2 By 4 5 6 9 100 1

A 2 6 8
B 2 5 ale
€ 2 geerhiC
D (ag 1 4 8 48
E peers (ee
F = yes 4
G ‘pee Sr a eA
H Date A Caneeee
I As il 1 1
2 ss 4 5

J eee or LG

1.2.3 Arithmetic series

A car is accelerating from rest.
In the first second it moves 3 m.
In the second second it moves 5 m.
In the third second it moves 7 m.
In the fourth second it moves 9m.
If this pattern continues, then the total distance travelled in the first ten seconds
will be
§=345474+94114+134+154+17+19+421
24 FOUNDATIONS

A sequence of numbers added together is called a series. In this case S will be the
sum of the series. To calculate S$ for the first ten seconds is straightforward, if a
little tedious, but had the sum been for the first thirty seconds, an algebraic
technique would have been useful.

oD)

As a schoolboy, the German mathematician Carl Friedrich Gauss (1777-1855)

spotted a simple fact which helped him to calculate the sum of the series
1 2p 34 99 100)
Can you find a simple way to sum this series?

Where consecutive terms of a sequence differ by a constant value, such a

sequence is known as an arithmetic sequence, or arithmetic progression (A.P.).

Gauss’s observation makes it straightforward to sum any such series.

Lalo eS ae Ora Ore lO

This series may be summed by noting how each pair of values linked below
has the same sum.

ieteh |
Poh 243 4S 67 Seo 10 10 aay
You can use this method to find the sum of any arithmetic series.

1 (a) Find the sum of:

(SS 2h Stati er 20
(ii) 1+-2-+3)-+4-b 59 (carefull)
(iii) 14+2+3+4+44+---+29
(b) Describe in your own words how to sum an arithmetic series.

2 Show that the series

57 Ole =e 0S
has 101 terms.

3 For the series below, state the number of terms and sum the series.

(a) 14+2+34+4+---+50
(b) 10+114+12+---490

(c) 200 + 199 + 198 + ---+ 100

 SEQUENCES 25

4 For the series below, state the number of terms and use the result above to
sum the series.

ee Se ee ee) se Uae O)aieAloe leis)

(b) 4+7+10+13+---+
100

Kc) 196 5 19d S86 ee 71

For some series, instead of being given the last term, you are given the
number of terms. If this is the case you need to find the last term.

5 For each of the series below, find a formula in terms of 7 for the ith term.

(ajo SP Soe 7 saeee

(by eoetnOee
(4-8 1ge1 8
(ch are
6 Bor the seties San 9 4. 13 icr b/s

(a) (i) calculate the 15th term;

(ii) calculate the sum of the first 15 terms.

(b) (i) Write down an expression for the ith term.

(ii) Find a formula for the sum of the first 7 terms.

You can generalise the method of question 6 to a series having first term a
and where each succeeding term is found by adding on d.

d is called the common difference for the sequence.

The first few terms are:

a, at+d, a+2d,

7 For the sequence'a, a+d, a+2d, ..., write down:

(a) the fifth term, (b) the SOth term,

(c). the mth term, (d) the sum of the first 50 terms.
26 FOUNDATIONS
Raper
Pa a

To obtain a general formula for the sum of m terms of the series, you need:
e the first term = a e the last term = a+ (n — 1)d

8 Show (by substitution into the earlier form for S,,) that the sum of m terms
of the series is
[2a+(n—1)d]_n "
Sem 5 = 5 Rat (n 1)d]

Example 3
Sum the series:

(a) 34+54+7+4+---+99 (b) 4+11+4+18+---

as far as the 50th term.

Solution
34+ 99
(a) The first term is 3, the last 99, so the average is = =15il
The difference between the first and last terms is 99 — 3 = 96.
With a common difference of 2, 99 is % = 48 terms on from 3.
Thus the total number of terms is 48 + 1 = 49 and the sum is
49 Ses 24993

(b) The first term a = 4, and the common difference d = 7, so from the
formula, Ssp = 25(8 + 49 x 7) = 8775.
 SEQUENCES” 27

1.2 Exercise 3

1 Find the sum of the arithmetic series:

(a)
a 4494144194 -..4.199
(b) 3+9+15+--- as far as the 50th term
(c) 1+15+2+234---+
100
(d) 99+97+4+95+---+25

2 Complete the following table.

First term Common Number of Last term Sum

difference terms

(a) 8 2 18
(b) 6 9 303
(c) 3) DS) 9S

3 A small terrace at a football

ground comprises a series of 15
steps, which are 50m long and
built of solid concrete. Each step
has a rise of 4m and a tread of $m.

By calculating the area under each step and forming a series, calculate the total
volume of concrete required to build the terrace.

4 A child builds a pattern with square building bricks using the sequence of
steps as shown.

cane
The total number of bricks used is 1+3+5+---

(a) How many bricks does the child use on the nth step?
(b) If the child has 60 bricks, how many steps can be completed?

(c) Use the formula for summing an arithmetic series to show that the total
; ; 2
number of bricks used will be x’.
28 FOUNDATIONS
ee
eee

SE Another child builds a square pattern using bricks that are twice as long as
they are wide.

(a) How many bricks does the child use on the nth step?

(b) Use the formula for summing an arithmetic series to find the total
number of bricks used.

6E Each year Mrs Smith gives her nephew a birthday present of money (in £)
equal to five times his age in years. The money is put into a bank account,
but unfortunately does not attract any interest and he is not allowed to
withdraw any money until he is 18. She makes the first payment on his first
birthday and continues until he is 18.

(a) How much does he have in the account on his 18th birthday?
(b) How old is he when the sum of money in the account first exceeds £500?

7E 220m of video tape are wound onto a reel of circumference 8.2 cm. Because
of the thickness of the tape, each turn is 0.1 cm longer than the previous one.
How many turns are required?

1.2.4 Finance — APR

When borrowing money or deciding
where to put savings, the interest
rate is not the only factor to be 21.7% APR
taken into account. When obtaining
a house mortgage, for example, it $
can be very important that the
Eee ; EASY FINANCE
building society or bank should be
Borrow £3000 under our
able to confirm the loan without a typical repayment scheme

long waiting period. Similarly, it and pay only £277.50 per

month for one year
might be necessary to accept a low
interest rate on savings in order to
be able to withdraw the money at
short notice.

In this diagram, three different ways

of advertising the same repayment
scheme for a loan are shown.
 SEQUBNGES 929

It is important that anyone considering borrowing should be able to compare

interest rates such as those above, all of which are commonly seen. Each refers to
the ‘easy finance’ payment described in the box. Notice that the customer must
begin repaying the loan almost immediately.

® .2F

11 per cent per annum

The customer repays

£27 1250 XG 1243330

and so has paid £330 in interest charges. Since

£3000 x 1 = £330
the stated rate of 11 per cent is easy to understand.
The 11 per cent is called a flat rate and, although it is easy to understand, it is
nevertheless misleading. The customer certainly pays 11 per cent interest on
the £3000 but does not have the use of the full £3000 for a full year!

1 A loan of £800 is repaid by 12 monthly instalments of £100. What is the

flat rate of interest?

1.65 per cent per month

This rate is also easy to understand. The customer pays 1.65 per cent interest
on the amount owed during the month.

After one month the customer owes the original £3000 and interest of
£3000x0.0165 = £49.50: a total of £3049.50. (A quick way of obtaining this
total is to calculate £3000 x 1.0165.) Since the customer repays £277.50, the
amount owing throughout the second month is £2772.

To find the amount owing after two months, the procedure is repeated. The
outstanding debt after two months is (£2772 x 1.0165) — £277.50 = £2540.24.

2 Use either a spreadsheet or a calculator to continue the procedure for 12

months.

The debt is repaid after 12 months. Surprisingly, the flat rate of 11 per cent
per annum and the rate of 1.65 per cent per month are equivalent!
30 FOUNDATIONS
SS ae Se ee eee eee

21.7 per cent APR

The concept of APR or annual percentage rate is used to give consumers a
simple way of comparing various methods of borrowing. Finance houses are
currently obliged by law to quote the equivalent APR.
The APR corresponding to the rate of 1.65 per cent per month can be found
by calculating the total interest on a year’s loan, assuming the entire
repayment is at the end of the year.

Suppose £100 is borrowed. After one month the amount outstanding is

£100 x 1.0165 = £101.65

After two months, the amount has become

£101.65 x 1.0165

and so on.

3 By finding the amount to be repaid after 12 months, show that the APR is
2a petmcent.

Remarkably, the three rates of 21.7 per cent APR, 11 per cent per annum and
1.65 per cent per month are all equivalent! When considering ways of
borrowing money, the APR enables sensible comparisons to be made.

4 Find the APR corresponding to monthly payments based on interest rates

per month of:

(a) 1 per cent (b) 2 per cent (c) 5 per cent

5 Describe an algorithm or procedure for converting any monthly interest

rate into an APR.

6 By trial and error using the method you found in question 5, find the
monthly interest rate that corresponds to an APR of 100 per cent.

7 Use a computer or programmable calculator to show the amounts

outstanding in successive months for any inputs of the initial loan, monthly
interest rate and monthly repayments.

8 A loan of £800 is repaid by 5 monthly instalments of £200. Use your

solution to question 7 and any appropriate method to find the monthly
interest rate. Hence find the APR.
 SEQUENCES 31

12.4% APR
CAR PURCHASE FINANCED
BY A HIRE PURCHASE AGREEMENT

CASH PRICE (ON THE ROAD) £7,292.86

DEPOSIT (30%)+ £2,187.86
BALANCE £5,105.00
36 EQUAL MONTHLY PAYMENTS OF peel 2a
1 TERMINAL RENTAL £1,786.75
TOTAL PAY ABLE £8,565.33

*36 monthly payments of £127.52 is equivalent to

approx £29.43 per week. + MINIMUM DEPOSIT 20%

The figures in this car advertisement are the result of certain calculations.
The ‘terminal rental’ is a final cash payment to clear any remaining debt.

(a) Explain how the figure of £2187.86 was obtained.

(b) How is the figure of £8565.33 obtained from the other figures?

(c) Show that £29.43 per week and £127.52 per month are equivalent.

(d) Find the monthly interest rate corresponding to an APR of 12.4 per
cent.

1.2.5 Sigma notation

If you invest a sum of £1000 per year into a savings account which pays 8 per cent
interest, then S,, the sum of money in the account after k years, and S, 4, are
related by
Spa1 = Sp X 1.08+ 1000
This inductive definition is useful for programming a spreadsheet, but is less
useful for generating a formula for total savings.

You can think of the savings over the ten-year period as being a set of ten
investments — the first gaining interest for the full ten years, the second gaining
interest for nine years, and so on.
32 FOUNDATIONS
a eee ee ee

1 How much is the first investment of £1000 worth after ten years?

2 How much is the second investment of £1000 worth at the end of the ten
year period?

3 Explain why the total investment is worth

£1000(1.08 + 1.08 + 1.08° + --- + 1.08"°)

after ten years.

Expressions such as £1000(1.08 + 1.08* + 1.08° + --- + 1.08"°) are cumbersome

to handle. However, since each term is of the same form, you can write it more
easily using a shorthand notation, known as sigma notation. Sigma, written %, is
a Greek letter, which to mathematicians means ‘the sum of’. Using this notation
you can write the above series as
10 10
S = 5/1000(1.08') or 1000 1.08'
i=1 i=]
10
Sometimes this is written simply as MS 1000(1.08’), where it is clear that the
H
summation is taken over the range of values of i.

Stated fully, this reads:

S is the sum of the series obtained by successively substituting the values of

i= 1toi= 10 in the general term i1000(1.08’).

Note that the 1000, which multiplies each term, may be taken outside the
summation as a common factor.

Example 4
Express the series 3+ 5+7+9+---+99 using © notation.

Solution
The general term of the series is 2i + 3.

The first term, 3, corresponds to i = 0 and the last, 99, corresponds to i = 48.
48 iw
Thus the series is SS (2i + 3). It could also be written as ye (2i + 1).
i=0 1
 SEQUENCES 33

Example 5
Sum the arithmetic series:
20 n

(a) $0Gi+4) — (b) S7Gi +4)

t='1 i=l

Solution
20
(a) S°(3i+4) =74104+13+---+64 (20 terms)
1
= 2 (7+ 64)
= Ae
n

b) So (3844) =74+10+---+(3n+4) — (# terms)

1 i
= 5{7+ (3n+4)}

= (3+ 11 11)
nN
= —(3

1.2 Exercise 4
1 Write out the terms of the pone series.
7 : 5 1
Won,
a
» 1—3
ic) Mier
1=1
8 6
(ay (1)
(21 3) fe) Gtiy ty Ser
1=4 :
M4
i
ii o il

2 Rewrite each of these using © notation.

(a) f/1+/24+ /34+---+ 50

(b) 27447 +67 +--+
+ 100°
(c) $+ $+5+-+- +55
(el 2 ee 197
(ec) 442434+44---4+75

3 Calculate the sum of the following arithmetic series.

24 20
(a) $/(45-2i) —(b) 5 (= 1)
i= i=1
 34 FOUNDATIONS
oe
Ee SS

4 Use sigma notation to express the following.

(a) 20+ 25+ 30+ 35+--- (25 terms)

b) 10000 + 1000 + 100 + 10 +.--- (10 terms)
c) 14+1.1+ 1.21 + 1.331 +--- (15 terms)
1 1 1 il
=r +.--- (8 terms
” eyo e rey error 44+5 ( )
6) 342x397? 43 x3? 44x 3° +. = 20 terms)

5 Write out the following sums in full.

6 4 5

CS: PG+1) (b) So(-'41) ©) SOx’

0 0

(d) (e) Sp f(x i)

Ge bated

ES 1.2 Tasksheet E1 — Using sigma (page 540)

1.2.6 Geometric series

An arithmetic series is a special type of series whose pattern enables a simple
formula to be developed.

The series on page 32 which represented total savings in the account,

£1000(1.08 + 1.08* + 1.087 + --- + 1.087°)

can also be summed using a simple formula.

This series is an example of a geometric progression (G.P.), in which each term

increases by a constant multiple or common ratio, which in this case is 1.08.

Summing a geometric series is not as straightforward as summing an arithmetic

series. To simplify matters ignore the 1000 and consider the sum of the series

S = 1.08 + 1.087 + 1.08 + --- + 1.08" ®

The trick is to multiply by 1.08

1.08S = 1.08* + 1.08° + 1.084 + --- + 1.08" @

and then subtract the original series S, so that
0.088 = 1.081! — 1.08 (all of the middle terms cancel out)

1:08) *=1:08
=>S=
0.08
 SEQUENCES 35

This method may be generalised for the general geometric series with first term a
and common ratio r. he
a : a .

S Sat ar ar ar ee are Or Ss ay are-

=I
Multiplying by the common ratio 7,
rS=artart+ar+---+ar
Subtracting the two series gives
rS —S=ar"—-a
S(r—1) = a(r" —- 1)
a(r”
— 1)
= Sie
r—1

Example 6 |
Find the sum of the series:

(yy Sa
a et OU
10

Solution
(a) The terms are 3, 3 x 2,3 x 27,...,3 x 2”.
The series is therefore a G.P. with first term 3, common ratio 2 and
11 terms. Its sum is
sf
2° -1
3 x —— = 6141
2-1
(b) The series is a G.P. with first term $s common ratio ; and 9 terms. Its sum
is therefore

ete:
5-1
— x ~“———_
w 0.1667
 36 FOUNDATIONS
eee

> Exercise 5
‘ he

1 Calculate the sum of the series to the number of terms stated.

(a) 2+6+184+54+-:-- (8 terms)
(b) 2+10+50+250+:-- (12 terms )
(Cio 7 eens (20 terms)
(d) 8+44+24+14+5+°:: (10 terms )
(e) 8—442-14+3---- (10 terms)

Calculate the sum of each series.

5 10 Uf. :

ay es be aes ea
i=1 i=1 i=1
8 20 :
OPI aoe Sie
i=3 Test
Legend tells that the Shah of Persia offered a reward to the citizen who
introduced him to chess. The citizen asked merely for the number of grains of
rice according to the rule:
1 grain for the first square on the chessboard,
2 grains for the second square,
4 grains for the third square,
8 grains for the fourth square, and so on.
(a) How many grains of rice did he request?
(b) If a grain of rice weighs 0.02 g, what weight of rice did he request?

Julius Caesar was born in 101 sc. If his mother had invested the Roman
equivalent of 1p for him in a bank account which paid interest per annum of
(a) 1 per cent (b) 5 per cent
how much would it have been worth in 1989?

The sum of £200 is invested annually at 5 per cent interest per annum. What is
the total sum of money in the account at the end of 50 years?

Using a typical figure for a school leaver’s salary and assuming that it will
increase by 5 per cent annually, estimate a person’s total earnings during their
working life.

The sum of £1000 is invested annually at 7.5 per cent interest per annum.
(a) What is the total sum of money at the end of n years?
(b) How long will it take for the total sum of money to be twice the total
amount invested?
 SEQUENCES 37

1.2.7 Infinity
So far, when you have summed series you have taken a finite number of terms.
On pages 19 and 20 you noticed that sequences show certain patterns of
behaviour (for example convergence or divergence) as you take more and more
terms. What happens to the sum as you take more and more terms in a series?

For a series with an infinite number of terms, the sum to infinity is the limit of
the sum to 7 terms as 1 tends to infinity.

Ft6-(9)
For example, the series 1 + ;a ele;ae (4)"~? has sum
pL cae Tee

wa D
As n — 00, ($)” > 0 and the sum of the series tends to 2.

On page 35 you found that, for a geometric series,

Se art =a (: )
aa p=

For what values of r will the series have a ‘sum to infinity’?

Clearly, if |r| < 1, then r” — 0 as n > co and so there is a ‘sum to infinity’.

(Recall that |r| < 1 means —1 < 7 < 1.)
n ; any 1 1 (tn n

The sum Di aa = a(? ; )can usefully be written as (=) for

r— —1r
examples where |r| < 1. As n — oo, r” — 0 if |r| < 1 and so there is a ‘sum to
1 . .

1 a
infinity’ o of e(—)
infinity’ = Rae :

If |r| > 1 then the series will not converge.

Example 7 —
Find the sum of the infinite series 1 + ZaL 3dt x le e560

Solution
First term a = 1 Common ratio r = 3
‘ 1
=> Sum to infinity $., = ane 3
3

a 1.2 Tasksheet E2 — Zeno’s paradox (page 542)

 38 FOUNDATIONS
nn EEE

1.2 Exercise 6

1 Where possible, calculate the sum of the infinite series:

9
(a) oO
9
10078 tooo
9 (b) 4s
PilsTeale
27 Ne,

(c) 1-24+4-8+--- (d) 54+$4+34+24+---

2 Calculate the sum of:

(a) Se a (b) SE (O25) (c) Se pa

i=1 i=] i=1

The diagram illustrates the infinite G.P.

ras bats
a La ue
(a) Find the sum of the G.P.
(b) How could you see this result directly from the
diagram?

4 Von Koch’s ‘snowflake’ curve is shown below in its various stages of

development. Fy is an equilateral triangle; F; is derived from Fp by trisecting
each side and replacing the centre third of each side of the triangle by two
sides of an equilateral triangle; F, is obtained in the same way from F, and
so on.

Fo I Fy Fo :
|

If each side of Fp is of length 1 and P,, is the perimeter of the mth snowflake
curve, write down:

(a)"Ro tbyPy wc) Pa (dy Pe

What happens to the perimeter of the curve as n — 00?

5 How many terms of the series

oe 4 ctipeBt cr ios
te aetna

must be taken before its sum to x terms differs from its sum to infinity by less
than 0.01?
————
ee ee eee
 SEQUENCES 39

i { }

ee
ee
a
40

Foundations
3 Functions and graphs
1.3.1 Function notation
A scientist performs an experiment to investigate the absorption of light by a
liquid. Light is shone through a coloured solution and the intensity of light
emerging is measured.
She finds that if she varies the concentration of the solution her readings are as
follows.

Concentration (mg cm) re 0 0.2 0.4 0.6 0.8 1.0

Intensity (lux) Le 200 ae 1S le as dm 00)

20
18
16

12
10

O 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 ©

The two variables are related by the formula

ee20 VE

This is known as Beer’s law of absorption.

The formula can be used to calculate the value of L for any given value of c.
It is often helpful to consider an expression from a different point of view and
think of it as a device which generates an output for any given input. In this case
the input is c and the output is L.
Input Output

c i fica 0
 FUNCTIONS AND GRAPHS 41

In other words, if the scientist inputs any value of c into her formula she will get
a corresponding output which tells her the value of L. This dependence on c can
be emphasised by use of the function notation L = f(c), where the function f is
given by

f(c) =—De

This gives a convenient shorthand, since you can immediately write f(0) = 20,
f(1) = 10 and so on.

Example 1
If f(x) = 2x” — 3x, then:
(a) find £(1) and £(2);
(b) find a simplified algebraic expression for f(x + 2);
(c) solve the saeionn fo —10:

Solution
:Mee 2—-3=-1 f2)=8-—6=2
f(x +2) = 2(x
+ 2)? — 3(x
+ 2)
=9(x" + 4x +4) —3x—6
= 2x* +5x'+2
(c) f(x) =0, so 2x*7-3x=0 => x(2x—3)=0
=x == 0) and a—= ~ are solutions to f(x) = 0

jes,€ ae Exercise 1
1 If f(x) = x* +3 for all values of x, find the values of:

(a) £0) — (b) f1)— (c) FCV2)_— (dd) (1)

2 ibe) = ;for all values of t, find:

(a) (2) (b) g(t) (<) 20) = (d) &(-1)_— (e) a(x)
3 (a) If f(x) = x* + 3x +2, find:
OL Ot a2 re) eB dy Avy Er)
(b) If f(y) = y° + 3y + 2, find:
COPEL) GUE 2) Gi) a0) ativ) t(—1) tv) E(n)
(c) Does f(x) differ from f(y)?
42 FOUNDATIONS
8 ee oe ee

4 If g(x) = (x+1)* and h(x) =x* +1

(a) find simplified algebraic expressions for

(i) e@—3) Gi) ho+2)

(b) solve the equation g(x) = h(x)
(c) solve the equation g(x — 3) = h(x — 3)

5 If g(x) =x* +2x +1 and f(x) =x+1

g(x)
(a) show that >~—=x+1
f(x)
(b) find simplified algebraic expressions for

(i) g(2x) (ii) g(x— 1)

1.3.2 Using function notation

The graphs of f(x) = x” and y = x* + 6x +5 are given below.
fe) = y

y=x7+6x+5
 FUNCTIONS AND GRAPHS 43

The expression for y can be written in completed squares form:

y= (x+3)?=-945 =(x+3)=4
Observe that this is the same as f(x +3) — 4.

Considering the two graphs, you can see that the graph of y = x* + 6x +5 isa
—3
translation of : of the graph of f(x).

Function notation provides a convenient and useful way of describing the

relationship between functions and their graphs. This idea is explored more fully
in the following questions.

1 (a) Sketch the graph of f(x) = x’.

(b) Rearrange the expression for f(x — 2) + 5 into the form ax” + bx +c.
(c) Superimpose the graph of f(x — 2) + 5. How is this related to the
graph of f(x)?

2 For each of the following functions:

(b) g(x) = 2* meee
(a) g(x) = x
(i)) sketch the graph of g(x);

(ii) write down the expression for g(x — 2) + 5;

(iii) what translation of the graph of g(x) would you expect?
(iv)) superimpose the graph of g(x — 2) + 5 and check your answer to part
(ii).
3 (a) Sketch the graph of f(x) = |x|.
(b) Without using a graph plotter, sketch the graph of |x + 3| — 4.

(c) Check your answer using a graph plotter.

4 (a) For any function f, what is the relationship between the graph of f(x)
and f(x +4) +b?
(b) Illustrate your answer to (a) by choosing your own function and
values for a and b.

(c) Will what you have described be true for any function?

SE (a) Confirm the relationship between the graphs of g(x) and g(x + a) +b
when g(x) = sinx.
(b) Write an illustrated account of your findings.
44. FOUNDATIONS
oo eee —————————e—Eeee

6E (a) Investigate the relationships between the graphs of

h(x) and ah(bx+c)+d
where h is any function, and a,b,c and d may take any values.

(b) Write an illustrated account of your findings.

1.3.3. Defining functions

You saw in section 1.3.1 that an example of Beer’s law of absorption can be
stated as
20
f(c) = Re

It would not be sensible to calculate f(—2), for example, because a concentration

of —2mgcm~~ would have no meaning. When defining a function you must be
specific about the values to which it applies.

For example, f(x) = /(x — 5), x > 5 defines a function f, where

e@ the rule is f(x) = V(x — 5)
e@ the domain is x > 5

In describing the domain of a function, the following notation is a useful

shorthand for defining important sets.
N — the natural numbers 1, 2, 3,...
Z:— the Ante gers...) oe on | Odeon ee
Q -— the rational numbers (or fractions, including, for example, $)

R — the real numbers (both rational and irrational such as \/2 and —7)
 FUNCTIONS AND GRAPHS 45

This notation can be extended using * and ~ signs. Thus R* means the positive
real numbers, Z means the negative integers.

You can also use the symbol € to mean ‘belongs to’. Thus x € Q* means
‘x belongs to the set of positive rationals’.

Note that N is a subset of Z, which is itself. a subset of Q, which is a subset of R.

When a function is written down, both the rule and the domain should be given.
However, in practice the domain is often omitted and the function is assumed to
be defined for all values that are valid in the rule. This is a common ‘misuse’ of
the description of a function.

: Exercise 2
1

(a) find:

(i £(0) — Gi) E(=1) = (iit)-#{(a) (iv). f(a —2) > (vw) (2)
(b) What is the largest possible domain for f?

(c) Sketch the graph of f(x).

i) (a) If h(x) = \/x, find the values of:

(i) h(2) ii) -H(9) Git) (V2) iv) Hr) () h(x")

(b) What is the largest possible domain for h?
(c) Sketch the graph of h(x — 3).

3 Write down the largest possible domain for g in each of the following.

@e@)=—=1 = WeW=a51
b ——————

1
(c) g(x) =4/x (d) g(x) SGD)

4 The modulus function |x|, meaning ‘the magnitude of x’, is defined as

Ke 0,
Os xe 10
(On some calculators it is called abs(x), which is an abbreviation for ‘the
absolute value of x’.)

(a) Write down the value of:

(isl iy a7 Guy Zh Gv elal a (9) 10]

 46 FOUNDATIONS
be ee ee eee

(b) What is the largest possible domain for the function?

(c) Sketch the graph of:

(i) y = |x|
(ii) y = |x —2|
Gi’) =e 224

1.3.4 To plot or to sketch?

In 1.3.2 and 1.3.3 you looked at many different functions. Now you will
investigate the general properties and techniques used for obtaining the graphs of
functions. You will be sketching graphs rather than plotting them. You can
illustrate why sketching is often better than plotting if you use the plotting
method to draw the graph of the function

Se 1
pam forx ER +

This gives the table of values and plotted points as follows.

Nigel 8 3 4 5)

Veo. 2.04 13,0), 16M 202

OB
4
1 What do you think the graph of y = 4x + F looks like?

2 What are the values of y when x = 2.45, 2.49, 2.499?

3 What happens when x = 2.5?

4 Use the graph plotter to draw the graph for values of x from 0 to 5.
 FUNCTIONS AND GRAPHS 47

You have seen that drawing a ‘smooth curve’ through selected points can lead to
major errors. In general, for functions it helps to have an idea of what the graph
will lock like before you attempt the sketch.

To obtain the overall shape of a graph you only need to consider some of the
major features.

Plot on the same screen and thus superimpose each of the following
graphs.

ORS AC) See Se gaa a Ses)

What points do all the graphs have in common?

Which function increases most rapidly, and which increases least

rapidly, as x becomes large?

What are the main differences between the graphs of the even powers
of x and the graphs of odd powers of x?

Plot on the same screen the following graphs.

G) y= | WGhy— se Gh) yx
What do you notice about the graphs of y = x* and y = x* + 4x when
x is a large positive or negative number?

What do you notice about the graphs p of y = 4x and yy= x* + 4x when

x is a small positive or negative number?

Plot on the same screen the following graphs.

(i) y=x° (ii) y= —4x* (iii) y = x° — 4x?

What do you notice about the graphs of y = —4x” and y = x? — 4x*
when x is a small positive or negative number?

What do you notice about the graphs of y = x° and y = x° — 4x” when

x is a large positive or negative number?

Plot the graph of the function

yarxrtx—2x41
48 FOUNDATIONS
ee

(b) Superimpose the graphs of:

(i) y=x (ii) y= -2e +1

(c) Compare the three graphs for:

(i) large positive and negative values of x,

(ii) very small positive and negative values of x.

(d) Suggest a reason for ignoring the terms x*, —2x and 1 when
considering the shape of the graph in part (a) for large values of x.
(e) Suggest a reason for ignoring the terms x° and x” when considering the
shape of the graph in part (a) for very small values of x.

5 Plot the graph of the function y = 3x + 2x — x?

What are the zeros of this function; that is, where does the graph cross the
x-axis?
Which term do you think determines the shape for:

(a) large positive and negative values of x,

(b) very small positive and negative values of x?

6 (a) What term is suggested by the shape of this graph when x is large?

Sa
elSet lO fa Safa an

Heeetee
ie ~ State +
{ i
se

tac
pap

(b) What term is suggested by the shape of the graph when x is small?

(c) (1) Suggest an equation for the graph above.

(ii) To confirm your suggestion, plot the graph of your equation.

(d) If necessary repeat (c) until you have found the equation of the graph.
 FUNCTIONS AND GRAPHS 49

In the questions in 1.3c, you developed some of the ideas that can be used to
build up an impression of a graph of a polynomial function.

Example 2
Sketch the graph of f(x) = 4x° — 6x? — 11x + 18.

Solution
(i) For large x the graph behaves like y = 4x°. This is the dominant term.
(11) For small x the graph behaves like —11x + 18.
Using just these features, you can draw part of the graph.
y We

which suggests
the basic shape:

However, the graph of example 2 might take any of these forms:

Va y Y

You need more information to establish which is the correct form. For example,
you may. be able to find the points of intersection with the x-axis, or the
coordinates of the turning points (i.e. points where the gradient changes sign).
50 FOUNDATIONS
a
eee A

When a polynomial function is given in factorised form, the roots give you extra
clues to help determine the shape of the graph.

Example 3
Sketch the graph of f(x) = (x — 2)?(2x +7).

Solution
(i) £(0) = 28.
(ii) For large x, 2x? is dominant.

(iii) The zeros of the function are at x = 2 (a repeated root) and x = —3 i,

The available information is plotted as shown.

f(x) i f(x)

This suggests the graph:

.3D

1 How can you obtain facts (i) and (ii) in the solution to example 3 without
multiplying out the brackets?

f(x)
2 Why is not possible?

x
 FUNCTIONS AND GRAPHS 51

(3 Exercise 3

1 For the function f(x) = (x — 2)(x + 3)(3x —7), which in expanded form is
f(x) = 3x° — 4x” — 25x 442
(a) on a partial sketch:

(i) mark any zeros on a set of axes;

(ii) mark the value of f(x) when x = 0;
(iii) sketch the dominant parts of the graph;
(b) complete the sketch.

2 For the function

f(x) = 5x" + 28x° — 33x" 42x +8 = —(w=1)* (x — 4)(Gx-+2)
(a) sketch the dominant parts of the graph;
(b) complete the sketch.

3 Sketch the graph of the function

y = x°(% + 4)(x — 7)

4 A girl standing on the edge of a cliff

throws a stone. The height of the
stone above the point of release,
after t seconds, is given by ~
h = t(12 — 5t) metres

(a) When is the stone level with the point of release again?

(b) The stone hits the sea after 4 seconds. Estimate the height of the cliff
above sea level. ”
(c) Sketch a graph of height against time.
52 FOUNDATIONS

1.3.5 Rational functions

The idea of dominance may be applied to sketching the graphs of rational
functions (i.e. algebraic fractions). However, your attention in rational functions
must always be drawn first to the denominator.
ewe
For example, consider the function

This function is not defined for x = —2. The denominator of the fraction equals
zero, and division by zero is not defined, so the function can have no value at
this point. On the graph of the function this means that there can be no y-value
when x = —2, and the graph will have a break where x = —2. Such a break in
the curve is called a discontinuity.

Example 4
Sketch the graph of -:*

Solution

(i) There is a discontinuity at

x= —2.

(1) As x becomes large positive,

the function is approximately
Re
ae 1 and approaches 1 from
below.

(iii) As x becomes large negative,

the function is again
approximately 1, but here
approaches 1 from above.
(iv) To the right of the
discontinuity at x = —2 (i.e.
x > —2), the value of the
function is very large and
negative. Just to the left of the
discontinuity (i.e. x < —2), the
value of the function is very
large and positive.
(v) When x = —1, f(x) = 0.
With this information you can
complete the sketch.
 FUNCTIONS AND GRAPHS 53

Example 5
Sketch the graph of the function f(x), where
x+1
(x) =
x7 + 5x +6

Solution
First try to factorise the denominator, so that f(x) becomes
CAS
(x + 3)(x
+ 2)
Note:

(i) There are discontinuities at x = —2 and x = —3.

x 1
(ii) As x becomes large positive, the function is approximatelyra
— i.e. —, and
6
approaches 0 from above.
1
(111) As x becomes large negative, the function is again approximately —,
—, but
here approaches 0 from below.

To the right of the discontinuity at x = —2 (i.e. x > —2), the value of the
function is dominated by the value taken by (x + 2) in the denominator.
f(x) is very large and negative for values of x just greater than —2 (e.g.
—1.9). Just to the left of the discontinuity (i.e. x < —2), the value of the
function is very large and positive.

To the right of the discontinuity at x = —3 (i.e. x > —3), the value of the
function is dominated by the value taken by (x + 3)-in the denominator.
f(x) is very large and positive here. Just to the left of the discontinuity
- (i.e. x < —3), the value of the function is very large and negative.

When x = —1, f(x) = 0. When x = 0, f(x) =2

With this information you can complete the sketch.
54 FOUNDATIONS
eee ee ee ee

In example 5, the lines x = —2, x = —3, and y = 0 (the x-axis) are called
asymptotes to the curve. The graph approaches these lines but does not (ever)
touch them.

In example 4, the asymptotes are the lines x = —2 and y = 1.

Example 6
» 80000
Sketch the curve y = 4x° + ——.
x

Solution Song
(a) y = 0 when 4 SS
5
so x = —20000
and x = —27.1
80 000
(b) When x is large, ——— is relatively small, and y & 4x*. For instance,
ae 80.000
when x = 100, 4x* = 40000 and ——— = 800. When x = 1000,
5 80.000 s
4x* = 4000000 and ——— = 80.
we
(c) y is undefined when x = 0, so the y-axis is an asymptote.
When x is small and positive, y is large and positive (for example x = 0.1
gives y = 800 000.04); small negative values of x make y large and negative.

This information is plotted, together with the complete curve.

NoMa th aac 4
20) ©40)5
 FUNCTIONS AND GRAPHS 55

The shapes of the graphs of some simple functions are worth knowing, as they
can be useful in sketching more complicated functions without the aid of a table
of values. The graphs of some functions, all of which are simple powers of x, are
shown below.

1 1 ;
Notice that y = — and y = — are not defined when x = 0, and that y = \/x is
2% x
not defined when x < 0.

1.3 Exercise 4

1 Sketch the graphs of:

ag car eos Oi

ee ee
5% x+1)(x-2 x + 5x +6

@ aae th xty Ulery

1
56 FOUNDATIONS
$$
-

2 A person who wishes to walk 20km in 4 hours proposes to walk half the
LOO
distance at vkmh’! and the remainder at ukm h!. Show that ee ral 4,
5 ,
and hence that v = = Uu —
Sketch the graph of v as a function of u, and
comment on the result.

3 The total resistance R ohms of two resistors of R; and Ry ohms in parallel is

1 1 1
given by ae + Re Sketch a graph of R as function of R,, assuming that
1 2
R, is fixed. Comment on the physical significance of very small and very large
values of Rj.

4 The surface area of a cylinder is given by the expression A = 2nr + 2arh.

Express / in terms of 7, and hence sketch a graph of h as a function of 7,
assuming that A is constant.

5 Suggest possible functions for the graphs below.

(a) y
 a7

Foundations
4 Expressions and equations
®

1.4.1 The language of algebra

The use of algebra in mathematics is extremely important. Many results are stated
algebraically and many problems must be expressed algebraically before they can
be solved — it is the language in which many mathematical ideas are expressed.

Letters appear in algebraic expressions in a variety of forms, and although you

will have met all of these in earlier work you may not have reflected on the
different uses of the letters in various cases.

Letters can stand for variables in expressions. In the quadratic expression

y= ax? +bxt+c
x and y are variables, while a, b and c are constants for a given quadratic
expression (although a, b and c may change for different quadratic expressions).
Additionally, in the quadratic expression given, x is called the independent
variable, while the variable y (which is calculated from the x-value) is the
dependent variable. This has an important parallel in science, where in a
scientific experiment the variable which you control is called the independent
variable, while other variables are the dependent variables.

When plotting graphs, the independent variable is plotted along the horizontal
(or x) axis.
V

Voltage V=IR

Current

Letters can also stand for unknowns in expressions. For example, if five adults’
tickets and three children’s tickets for the cinema cost £20, then
5x + 3y = 20
where x is the price of an adult’s ticket and y the price of a child’s ticket.
58 FOUNDATIONS
0 eS Se SS eee eee

As well as using letters to stand for unknowns, they may be used to express
general results such as

eee ee oe
n(n
+1

for the sum of a simple arithmetic progression, or to express general statements

such as
Numbers of the form 2” + 1, {n € Z}, are always odd.

The algebraic statement

x? —4=(x—2)(x+2)
is called an identity. Identities are also common in mathematics and differ from
equations in that they are true for any value of the variable. An equation is only
true for specific values of the variable, and these values are called the solutions of
the equation.

x* — 4=0 is true only for x = +2

Practice at setting up and solving equations is provided in the following exercise.

If you need to revise some of the basic methods of solving equations then you
should do tasksheet S1 first.

Tasksheet $1 — Review of equations (page 530)

Exercise 1

1 Solve the equations:

(a) 3x +9=0 (b) x* = 25 (c) x*+5x—-6=0
(d) n= (AG ey 60 Seam xi en es
(g) 2x +3y
= 22 (h) 2y—x=2 (i) 2x+3y+1=0
oye 6y + 5x = —2 XS

2 The triangular numbers are defined as

Aeet ys tee T,=1+2+3 and so on

If the mth triangular number is 210, formulate an equation for n and solve it to
find n.
 EXPRESSIONS AND EQUATIONS 59

3 (a) Find the sum of three consecutive numbers. Repeat for several other sets of
three consecutive numbers. What do you notice about the results?
(b) Any three consecutive numbers can be written algebraically as
n, (n+1), (n +2)
Find the sum of these three algebraic terms. How does this explain what
you have observed about the numerical results?

4 ‘The difference between the squares of two consecutive odd numbers is always
a multiple of 8.’

(a) Test this result in a few numerical cases. Is it true for the odd numbers
91 378 627 513 and 91 378 627 515?

(b) Prove the result algebraically.

5 ‘A bamboo 18 cubits high was broken

by the wind. Its top touched the
ground 6 cubits from the root. Tell the
lengths of the segments of the
bamboo.’

Solve this problem posed by mee =n 3

Brahmagupta (Hindu mathematician (tha 2 lem Mlr ata

PSOCIES Pc merenie oe
2

c. AD 630).
From H. Eves, Introduction to the History of Mathematics (Saunders)

6 Four right-angled triangles with sides a, b and c are arranged as shown in the
diagram.

Explain why the area of the large square can

be given as both (a+ b)” and c’ + 2ab.

Use these two results to prove Pythagoras’

theorem.
60 FOUNDATIONS

1.4.2 Quadratic equations

The path around a square lawn is 1 metre wide. If the area of the path is equal
to the area of the lawn, the dimensions of the lawn may be found as follows.

Let x be the length of a side of the lawn.

Area of lawn = x*

Area of path = 4(x + 1)

So x? = 4x +4

x’ —4x-—4=0

This equation does not factorise to give integer roots!

However, it may be solved using the method of completing the square developed
earlier (section 1.1.4).
x’ —4x-—4=0

[(«—2)? —4)=4=0
(x—
2/7 =8=0
(a) as
>x=244/8
The negative solution (2 — \/8) cannot be a solution in the context of this
problem, so the only possible value for x is 2 + \/8.

The lawn is a square of side 4.83 m (to 3s.f.).

Using the completed square form gives an alternative approach to solving a

quadratic equation. It can also be used to derive an important formula that can
be applied to any quadratic.

@ 4
1 Solve the equation x* + 6x + 4 = 0 using the method of completing the
square.

The solution of the equation x” + bx +c = 0 depends entirely on the two

numbers b and c. You can find a formula for the solutions in terms of these
two letters using the method of completing the square.
 EXPRESSIONS AND EQUATIONS 61

x +bx+c=0

<+8) -F4en0 ®

(x+$) =4-< ®@

eyes
pee ®

Zby/(b?
= 4c)
Zz
; 2
2 (a) Explain where the Gfhas come from in line @.

(b) Why has the c in line @) become 4c in line @)?

(c) Why has the + appeared in line @)?

3 Use the formula to solve:

(a) x7 +6x+4=0 = (b) x*—-6x4+4=0 8 (c) x =5—3x

By using a method similar to that above, the following result is obtained.

Example 1
Find (correct to 2 d.p.) the coordinates of the pos of intersection of the
straight line y = 2x + 3 with the parabola y = 5x* — x — 1.

Solution
The intersections occur when 5x” — x — 1 = 2x + 3, i.e. when
5x” —3x—4=0.

—b+./(b’ — 4ac) a— 3.6 ——3) ce= —4,

Using x = “whete
2a
 62 FOUNDATIONS
I

34J7(9-4x
5x (-4))
Ue
3+./89
=>
xX =>
10
=> x = 1.24 0r x = —0.64 (to2 decimal places)
Note that the solution may be given in one of two forms.
89 ; ;
ule ikare the exact solutions; 1.24 or —0.64 are the approximate solutions.

When x = 1.24, y = 5.48; when x = —0.64, y = 1.72. Hence the intersections

occur at (1.24, 5.48) and (—0.64, 1.72).
— ,/89
Numbers such as are known as surds. Further practice at dealing
10
with numbers written in surd form is provided on tasksheet S2.

1.4 Tasksheet S2 — Working in surd form (page 532)

Example 2
Solve the pair of simultaneous equations x + y° = 4 andx+y=1.

Solution
The problem is best solved by substitution, as follows.
y=1—«x (from the second equation)
et Ge x) =4 (substitute for y in the first equation)
= 2x? = 2% —'3'=0 (expand and simplify)
A Paes Diets Ose
— eG ~
4 2

i # il
When x= shave ; pee a eel
2 z
1—/7
When i= a ; yt

1.4 Exercise 2

1 Use the quadratic equation formula to solve the following, giving your
answers (a) in exact form, and (b) to 2 decimal places.
(i) x74 Sx S= 6 (ii) 3x? + 6x+2=0
(iii) 3x7 +4x-—-1=0 (iv) 5x* -8x -1=0
(v) x7+2x=4 (vi) 6x7 + 10x —4=x7+7x—3
2 Find the coordinates of the points of intersection of the graphs of:
(a) y= 5x? and y=3x+4 (b) y=x—x’ and y = 8 — 6x
(Cc) xy =4 andy — 2x =
 EXPRESSIONS AND EQUATIONS 63

Solve the following equations where possible, giving your answers to 2 decimal
places. Use factors, where appropriate, otherwise use either the formula or
completing the square.

(a) 8x—x7=0 (b) x +2x=5e44 (c) x? +x%-1=0

(d) x*= 25 (e) x*+x+1=0 (f) 25 = 10x — x”

Derive the formula

Mee J (b? — 4ac)

iz. 2a
for the roots of the quadratic equation ax* + bx +c = 0.

Solve the following simultaneous equations.

(a) y* = 4x (b) yy =4x +1 (c) xy = 64
ee e Vea L 4x — y = 60

Two numbers differ by 1 and have a product of 10. Let n be the smaller
number. Form an equation in 1 and solve it. Hence find the two numbers
exactly.

(a) Use the formula to solve, where possible, the equation f(x) =0 for each of
the functions:
(i) f(x) = x*—2x+4 © (ii) f(x) =x? -4e44 (iii) f(x) =x*-— 6x44
(b) The graph of each of the functions is given below. Use the information
about the roots that you found in part (a) to match each function to its
graph.

PA

BNWA BNW

o N w > n »< S TAR 2 Saatex

(c) Describe how the value of b? — 4ac (known as the discriminant of the
quadratic equation) relates to the number of roots.
 64 FOUNDATIONS

Tasksheet E1 — Regular pentagons and the Fibonacci sequence

(page 544)

1.4.3 Inequalities
Some problems may not lead to an equation, but may give rise to an inequality.
For example, what is the first triangular number greater than 50?

Trial and error may lead you to the result that m = 10, but if you were to take an
algebraic approach you would have to solve

ti 2 Oe 0

i n=100 =.0

which is an example of a quadratic inequality.

We shall give two examples of how to handle inequalities. The first demonstrates
\ how to solve a linear inequality.

Example 3
Find the solution set for the inequality t+ 2 > 6t + 7.

Solution
P+ 2 > 61-27
=>--S5t>5 Step 1: Gather together like terms.
atid Step 2: Divide both sides by a negative value
(note the change in the inequality sign).

The set of values of t which solve this inequality are known as the solution set
for the inequality.

The following example demonstrates how a graphical approach is helpful in

solving quadratic inequalities.
 EXPRESSIONS AND EQUATIONS 65

Example 4
Find the solution set for the following inequalities.
(a) (x +1)(x-2)>0
(b) (x+1)(x-2)>4
Solution
The graph of y = (x + 1)(x — 2) is as
shown.

(a) From the graph you can see that (x + 1)(x — 2) > 0 implies
xu 2) pores —1
(b) The graph of y = (x + 1)(x — 2) crosses y = 4 at x = —2. and x =3.
So (x + 1)(x — 2) > 4 implies
SS 3. GOD GR <= 2

When you solve a quadratic inequality you are seeking the set of values for
which it is true. It is usually simplest to solve the corresponding equation and
referto a sketch graph to find the ranges of values which satisfy the inequality.

Example 5
For what values of x is:

(a) x* >6—x
(b) x* <6—x?

Solution
Solving x* = 6 — x:
x +x-6=0
(x+ 3)(x—2) =0
x=-3 or x=2

From the graph you.can see that:

(a) x* > 6 — x when either x < —3 or x > 2.

(b) x” <6 —x when both x > —3 and x < 2. This solution set is usually
written as —3 << x < 2.
66 FOUNDATIONS

When dealing with algebraic fractions, the same ideas are employed but you
must be careful that when you simplify the fractions you multiply by positive
1
expressions. For example, to simplify at > 1 you might multiply by x + 1.
Xe
However, x + 1 may be either positive or negative, and this will cause problems
with the inequality. It is better to multiply by (x + 1)’, as (x +1)? > 0 for all x
(except —1, which cannot be part of the solution set anyway).

Example 6
Solve for x:

2x —1 x—1
aL b 1
Soi (0) x (c) CS Cea

Solution

(a) ae > Part Tae 14 [Multiply by (x + 1)7]

x+1>x°+2x41
0> x? +x

0 > x(x + 1)

f(x) =x(x+1)=0 whenx=0andx=-~-1

when x >'0,) He) S00
when = <0; f(x) <0
when wet FG 0

S Os ie 2 he Sl) < xi 0

x= 1
(b) a => x(2x—1)>x
2x”
= x > x?

x —x>0

x(x—1)>0

There are zeros at x = 0 and x =1.

when A> 1, ox(e ot) 0 J

when 0<x<1, x(x-—1) <0 x
when x<0, x(x — 1) 10 J
So the solution setis x>1 or x <0.
 EXPRESSIONS AND EQUATIONS 67

al
(c) > 0=> («—-2)(x-—D(x+1)
>0
(x — 2)(x +1)
f(x).= (x —2)(x —1)(x+1)=0 whenx = —1, 1,2
If x ld, (x0
Lice 2 het RO
Be Ae a 0)
w<—1~ (xe) <0 Ee
So the solution setis x>2 or -1<x< 1.

1.4 Exercise 3

1 Solve:
(a) 5x < —10 (b) 1—2x
< 3x +6
(c) 2(x-—3)
<8 (d) 3(x +5) <2x+4+3
(e) —3x < 6

2 Find the range of possible values for x if (2 — x)” > 0.

3 a= (x—1)(x+3)
y=2x+1

Mas

Use the graph to write down the

solutions of:
(a) (x —1)(x+3)<0
(b) (x —1)(x+3)>5
(c) (x —1)(x
+3) < 2x41

4 Use sketch graphs to help solve the inequalities:

(a) (x + 5)(x-—2)
>0 (b) (x —2)(3-—x)
<0
(c) (x +2)(x—2)(x-5)>0 —2)? 50
(d) (x +2)(x

5 Solve the inequalities:

(a) x? < 3x (b) 3x* + 2x < 2x? + 3x
(c) x(x—3)<10 9 (d) x +x>1
68 FOUNDATIONS
Oo —

6 Find the first triangular number which is greater than 1000.

JE Solve the following inequalities. You should use algebraic and not graphical
methods in all cases.
1 axa ‘aaa
nee (b) 7 >0 Orie ae

Smeal 58
Skint Seall Se 1 Se
ae eee

1.4.4 Inventing new numbers: ,/—1

The quadratic equations we have met so far have either had no roots or two
roots (counting equal or repeated roots as two roots). It would certainly be
attractive and convenient if all quadratics had two roots.

Consider the equation x” + 4 = 0. There are no values of x among the real

numbers that solve this equation. However, if we define a number j = /—-1, then
the equation has the two roots +2):
x +4=0
xv =-—4

x=+,/-4

= £ /(4 x (—1))
=+/4x /-1
x=+2)
It is clear that the roots of this equation are not real numbers (i.e. they are not
numbers that have a place on the real number line); they are called complex
numbers.

Extending the number system in this way has been a common feature of
mathematics throughout its development over the past 2500 years or so.
Problems associated with the solution of polynomial equations have absorbed the
attention of mathematicians since the earliest times. The simplest equations such
as x — 7 = 0 or 12x — 5 = 0 could be solved in the positive integers or rationals,
both of which were in common use by 1000 Bc (although modern notation for
fractions dates only from about Ap 1500). In the 5th and 4th centuries Bc, the
Greeks made the first important extension to the number system by laying the
foundations of irrational numbers, enabling equations like x* — 5 = 0 to be
solved. Since their interest came principally from a geometric viewpoint, the
Greeks did not appreciate that there might be two roots — indeed negative
numbers in their present form were not fully established until the 16th century.
 EXPRESSIONS AND EQUATIONS 69

At about that time also, the modern system of classification of polynomial

equations according to their degree was introduced and it was realised that there
was no neat theory to account for the number of roots. A quadratic might have
0 or 2 roots (counting a repeated root as 2), a cubic might have 1 or 3 roots, a
quartic might have 4, 2 or 0 and so on. Obviously, it would be ideal if every
quadratic had exactly 2 roots, every cubic 3 roots and every equation of degree n
had exactly n roots.

In 1545, Cardan was the first to attempt to deal with the solution of an equation
like x” + 4 = 0, but it was more than 250 years later that the invention of
complex numbers was completed by Gauss (who went on to show that no
further numbers need be invented to solve equations of any degree).

In fact, only one new number is needed, denoted by j, such that j? = —1.

j is combined with the real numbers to form numbers of the form z = a + bj, the
complex numbers. Notice that when b = 0, the complex numbers reduce to the
familiar real numbers. In fact, the real numbers are simply a special set of the
complex numbers. When a = 0, the complex numbers reduce to the form bj.
Such numbers are said to be imaginary numbers and are again a special set of the
complex numbers.

The arithmetic of the complex numbers is illustrated in the following example.

Example 7
If z, and z) are the complex numbers z, = 2 + 3), z, = 3 + j, then find:

(a) 24 + 2 (by:zq 1% (c) 2%

Solution
(Alea eo 2 a Sf Ot JS SA

(b) 21 — x = (2+3/)- 3 +4)

=-1+42

(c) 4% = (2+3/)3+/)
= 2(3+/)
+ 3j(3 +/)

=64+2+9+3/
Sorell;3. = 1)
=3411
70 FOUNDATIONS
og
eee

Within the complex numbers, every quadratic will have exactly two roots.
As described earlier, Gauss went on to show that every polynomial of degree
has exactly 7 roots. This is a considerable improvement on the previous position,
and is a very important result.

Example 8
(a) Solve the equation x —4x+9=0 (b) Factorise x +4

Solution
(a) x —4x+9=0
(= 2) 4-9-1)
(x-2)? =-S5=-1x5
x—-2=+,/-175
x—-2=+j/5

So x =2+4+j/5 and x = 2 —j,/5 are the two solutions. The solutions are
both complex numbers.
(b) x7 +4 = (x + 2j)(x — 2/)
You could factorise by finding the roots of x* + 4 = 0 (i.e. x = 42s) and
then writing the factorised form. This is often easier.

Introduction of the complex numbers is a crucially important extension of the

number system. The study of complex numbers is most important as they may
be applied to the solution of many problems, both within mathematics itself and
its applications, especially in science and engineering.

Exercise 4

1 Work out the squares of:

(a)37 ¢ ADOT Sle) 4) sp (Day a Ce) iam) ee

2 Write down the square roots of:

(a) -—9 (b) —16 (cy 12 (d) —20

3 Ifz; =2—j and z, =4+4Yj, find:

(a) 3z, + 22) (b) 2(z, — z) (Cc) 24%

(d) 2 (z1 + 2) (e) (z + 22)(% — 2)
 EXPRESSIONS AND EQUATIONS 71

4 Solve the following quadratics by completing the square. Leave your answers
in exact form.
(a) x —6x+10=0 § (b) x? +2x+10=0
(c) x*+4x+20=0 (d) 2x*-2x+1=0
5 Find the real roots of the following equations.
(a) x°—6x+6=0 (b)x*—-2kx+2=0 8 (c)x*+6x+9=0
State how many real roots you find for each equation.
6E Factorise:

(ert 94 Abita xe Niclas = 6x411

7E Write down a quartic (fourth-degree) polynomial equation which has no real
roots. Illustrate with a graph.

1.4.5 Polynomials
A polynomial is a function involving whole-number powers of a variable. In
general, you can write
al
P(x) at Pg ll (Pe oe ee ax bap as

where a, 41, 4),--.,4@, are constants, referred to as coefficients. The highest

power present is the degree of the polynomial. For example, a cubic polynomial
is a polynomial of degree 3, such as
P(x) = x° + 3x* — 6x — 8
The zeros of P(x) can be found by determining the values of x which make the
factors zero.
P(x) = (x + 1)(x — 2)(x + 4)
The value x = —1 makes the first factor zero and so —1 is a root. The other
roots are +2 and —4.

So x +3x* —6x -8=0 =.= — 1 24

The next tasksheet will give you extra practice at expanding brackets if you
need it.

md 1.4 Tasksheet $3 — Expanding brackets (page 534)

72 FOUNDATIONS
ee

4B

1 For the polynomial P(x) = x° — 13x — 12

(a) calculate the values of:
(i) P(1) (ii) P(2) (iii) P
(v) P(—1) (vi) P(—2) (vii) P(—3) (viii)

(b) write down three factors of P(x);

(c) confirm your answers to (b) by multiplying the three factors together.

2 The cubic polynomial P(x) is ei = 10 8.

(a) To check whether x + 2 is a factor of P(x), for which value of a
should you choose to calculate P(a)?
(b) Is x +2 factor of P(x)?

3 (a) If P(x) = (x — 2)(x* — x — 2), explain why P(2) = 0.

(b) More generally, if P(x) = (x — a)Q(x), where Q(x) is a polynomial,
explain why P(a) = 0.

In question 3(b) you proved that:

When you have found one linear factor, one approach to finding the remaining
factor or factors is to use the process of algebraic division. This method, illustrated
below, essentially evaluates the quotient function term by term. To divide
> _ 3x” — 10x + 24 by x — 2, you would proceed as follows:
> _ 3x? — 10x + 24 = ((x — 2)x? + 2x? — 3x* — 10x + 24 (match x°terms)
= (ex 0d
= (x— 2)x*— (x —2)x — 2x — 10x +24 (match remaining
(2 = (ee See paged)
= (x — 2)x?— (= 2)x= 12(¢ = 2)
= (x— 2)(x* —x = 12)
 EXPRESSIONS AND EQUATIONS 73

This process is more commonly set out in the style of numerical long division:
a x —12
x-2)x?— 3x? — 10x + 24
oye
—x* — 10x
= =x’ + 2x
—12x + 24
- —12x + 24
0

You should be able to see how this algorithm works by comparing it with the
Way it is set out earlier. It is usually easier to use the second of the two methods
of setting out the work.

Example 9
Divide x? + 2x — 1 by x —2.

Solution
x? +2x +6
x —2)x! + 0x? + 2x 1
ae eee

2x? + 2x
= 2x? — 4x
6x — 1
— 6x — 12
11
x? 42x —1= (x? +2x+6)(x
—2)4+11
It helps the setting out of the division if you include the ‘missing’ term
0x* — all missing terms should be included in this way. You should also check
your final answer by expanding brackets, whichever method you use.

The factorised form of a polynomial P(x) is very convenient for solving the
equation P(x) = 0 and for sketching the graph of P(x). The following example
illustrates another method of finding the remaining factors once the factor
theorem has provided the first. It can sometimes be used more easily than the
method of long division described above.
 74 FOUNDATIONS
a

Example 10
For the function f(x) = x° — 3x7 — 10x + 24:
(a) find the factors of f(x);

(b) find the roots of the equation f(x) = 0;

(c) sketch the graph of f(x).

Solution
(a) The possible factors of 24 are +1, +2, +3, +4, +6, +8, +12 and +24.
By trial,
{= 9 = 12 4 204240
so
x — 2isa factor

Instead of using algebraic division, you can proceed as follows. Write

x? — 3x? — 10x + 24 as (x — 2)(x? + ax — 12). Comparing the coefficients
of x* gives
—3=-2+4+a

a lL

and so
f(x) = (x — 2)(x* — x — 12)
= (x —2)(x
+ 3)(x — 4)

=> (x — 2)(x+3)(x
—4) =0
Sea)
(c) The graph is a cubic which cuts the
x-axis at —3, 2, 4 and the y-axis at 24
as shown.

1.4 Exercise 5

1 Using algebraic division, find the quotient and remainder when:

(a) ) x° + 2x” + 8x — 5 is divided byx+1
(b) x? — 2x* + 3x — 5 is divided
by x + 2
(c)) 2x? — 5x +3 is divided by x —1
(d) 4x° + x” — 1 is divided by 2x + 1
(e) x7 4+2x? — 4x? — 2x +1 is divided
by x* —1
 EXPRESSIONS AND EQUATIONS 75

2 TP (o:) a a SAA 12, show that x + 3 is a factor of P(x), and

find Q(x) if P(x) = (x + 3)Q(x). eer

If P(x) = x° — Sx*+2x+8
(a) use the factor theorem to find one factor of P(x);
(b) factorise P(x) completely;
(c)) write down the solutions of P(x) = 0;
(d) sketch the graph of P(x).

Solve the equation x° + 5x* + 3x —9 =0.

Solve the equation x° + 4x” + 2x — 4 = 0 by finding one simple factor and

then solving a quadratic equation.

(a) Solve the equation x —2x+4=0.

(b) Sketch a graph of the function to explain the solution.

(a) Find all the zeros of, and hence factorise, the function
P(x) = x° = x? —7x7
+ x46
(b) Sketch the graph of the function.
(c) Solve the inequality PG) 230:

(a) What happens if you try to factorise these expressions completely?

(i) 2 "8" Giese
(b) Sketch the graphs of these functions and comment on their zeros.
76

Foundations
_.5 Numerical methods

1.5.1 The golden ratio

In section 1.4 you looked at techniques for solving polynomial equations.
Although you found a formula that works for all quadratic equations, the
method that you used for cubic and quartic equations will only work if the
factors can be found easily. In fact there are general methods of solving both of
these types of equation, but they are beyond the scope of this book.

Here we look at alternative ways of solving equations, using numerical

techniques. These techniques do not give the exact solution to an equation,
but will often give good approximations in cases where exact methods break
down.

This section introduces two important numerical techniques for solving

equations, a decimal search method and an iterative procedure.

Consider the following problem. Given a square, can a rectangle be ‘added’ ...

... So that the shape of the new rectangle is the same as that of the added
rectangle, as shown below?
 NUMERICAL METHODS 77

A rectangle with this shape is called a golden rectangle and the ratio of the
lengths of its sides is called the golden ratio. Later you will see that there is only
one possible value for this ratio. It is denoted by the Greek symbol ¢ (phi) in
honour of the great sculptor Phidias who used it in his work. Like other famous
mathematical constants such as 7 and e, ¢ is found in many situations. Many
mythical and mystical properties were attributed to ¢, which may explain the use
of the term ‘golden’.

How can you find the golden ratio? Comparing the added rectangle with the new
rectangle,
ol

gp x

you can see that the ratio of the shorter sides is ¢. The ratio of the longer sides
must also be ¢ and so
ox$¢=o4+1
@ therefore satisfies the quadratic equation
x =x4t1
2
or, rearranging, x —x—1=0.

This equation can of course be

solved using the formula, but in
this instance you are going to see
how it can be solved numerically.

If you put x* — x — 1 into the

completed square form,
(x — 1)? — 3, the graph is easy to
sketch.

Possible values for @ occur at points where the graph cuts the x-axis. There is
just one useful solution, between 1 and 2, because the negative solution cannot
represent a length.

Now that you know an approximate value of ¢ you can find it more precisely by
using the method of decimal search.
78 FOUNDATIONS
fe ee ees

1 Begin the decimal search by considering values of x between 1 and 2 in

steps of 0.1 and show that there is a solution between 1.6 and 1.7.

2 Continue the search in steps of 0.01, from 1.6 to 1.7.

The cold facts discussed so far have not done justice to the history of the golden
ratio, from the serious school of Pythagoras to the fanciful theories of
enthusiasts. Perhaps the most important work was done by Leonardo of Pisa
(b.1175) otherwise known as Fibonacci. His study of natural phenomena led to
the sequence of numbers which bears his name
tS
Sie Ase LS Oo5

which can be defined inductively by

w=1, a1,

1 te age LOL 1 2 83,

Numbers from this sequence arise in many surprising contexts, including the
structure of a beehive, the population of a rabbit warren and the white and black
notes on a piano keyboard. Many plants have spiral patterns of petals or leaves
and the number of spirals is invariably a number from the Fibonacci sequence.
A detailed study shows that the sunflower head has 55 spirals in the clockwise
direction and 34 in the anticlockwise, while the pine cone has 8 clockwise and
13 anticlockwise parts.

Another major contribution Fibonacci made to mathematics was his promotion

of the Arabic system of numbers, which we use today.
 NUMERICAL METHODS 79

What is the connection between the golden ratio and Fibonacci numbers?
F . U; é
Consider the ratio +**, the values of which are plotted on the graph below.
u;
“i441

— x =e ee SS SS SS Oe

Uj, 89
= — =1.618
U0) Sy)

ale ima at
8 9 LOE

This sequence converges to the golden ratio, ¢, that is

Uj44 4
Be a as i > 00
U;

As the shell of the chambered nautilus grows,

the size of the chambers increases but their
shape remains unchanged.
80 FOUNDATIONS
je SS

The convergence of the sequence of ratios suggests that an iterative method of

solution could be used to evaluate the golden ratio. For such an approach, you
need a recurrence relation or iterative formula to set up a sequence of values.

The equation x” = x + 1 can be written as x = \/(x + 1). (The root given by

x = —,/(x +1) is rejected as you are only interested in the positive root.)

If the sequence generated by the formula

ign = V(x + 1)
converged, then the limiting value x would satisfy x = \/(x + 1) and would
therefore satisfy x7 =x +1.

From the sketch of the graph on page 77, x; = 1 seems an appropriate choice for
the first term of the sequence. Then
x) = 14142136 xg = 1.617 8513
x3 = 1.5537740 x9 = 1.6179775
x4 = 1.598 0532 X49 = 1.6180165
x; = 1.611 8478 x41 = 1.618 0286
xg = 1.616 1212 x1) = 1.618 0323
x7 = 1.617 4428
This sequence clearly converges to a solution for x which is 1.618 (to 3 decimal
places). Using an 8-digit display calculator, the value 1.618 0340 is obtained after
sixteen iterations and remains unchanged by any further iterations.

As well as fascinating mathematicians, the ‘golden’ property has also attracted

the attention of philosophers, architects and artists over the centuries.

It has been claimed that the golden rectangle

$=1.618

is the most artistically pleasing rectangle.

 NUMERICAL METHODS 81

An attempt to confirm or deny this was made by Gustav Fechner in 1876. His
extensive experiments did confirm that the most aesthetically pleasing shape for a
rectangle was something between that of a square and a rectangle with sides in
the ratio 1:2.

Architecturally, some very famous and beautiful structures, for example the
Parthenon, are said to be based on the golden ratio and golden rectangle.

It would be satisfying if an exact representation of the ‘golden’ number could be

given. For this it is necessary to use the quadratic formula for the solution of
‘x? —x—1=0.
feeds J/(1+4+ 4)
a as
a

=f41ys or 4-45
/5 is a number written in surd form. This means that the square root sign
remains, rather than replacing \/5 with its decimal value of 2.236 068 (to 7 s.f),
which would be clumsy to write out every time, and which of course has a small
error associated with it.

As seen earlier, the negative value is discarded, so the golden ratio is

— @ =4(1 + V5) © 1.618.034
82 FOUNDATIONS
Be
SS EE EE

We can link the golden ratio to the Fibonacci sequence algebraically, sterting
from the definition of the Fibonacci sequence.
Uj42 = Uj41 + Yj (u. =u, = 1)
Tae U;
a pe eet
Uj+1 Uj+1
; Uj+4
If we define v; = —*, then
u;

Uv:
t=
1 = 1+ —:
U;

With a suitable starting value, this gives ¢ iteratively and the limit satisfies
il :
x= 1+-—orx* =x +1, the equation for ¢ seen above.
x
If you would like to find out more about the golden ratio, many recreational
mathematics books have chapters devoted to it. A very readable book is
H. E. Huntley, The Divine Proportion (Dover, 1970).

1.5.2 Locating roots

Section 1.5.1 described two numerical methods for solving an equation — a
decimal search and an iterative process. Both methods had the same first step of
approximately locating any solutions by plotting a graph.

Two possible arrangements of the equation from section 1.5.1 are

x Sx+1 and #*—x-1=0

5B

1 In each arrangement:
(a) what graphs should you draw to solve the equation?
:
i (b) which points give the solutions?
:|

2 Sketch the graphs and find inequalities (bounds) for the roots.
3 Use the ‘zoom’ facility of a graph plotter to find the roots to 3 decimal

|
places.

4 What are the advantages and disadvantages of the two arrangements?

 NUMERICAL METHODS 83

Example 1
Find bounds for the solutions of
2 1
x =1+
Cea)

Solution
From the graph, it can be seen that roots lie
between —3 and —2, between —2 and —1
and between 1 and 2. ae wen 0 SU Le

Graphs can be drawn very easily using a graph plotter, but care is needed to
ensure that all the solutions are displayed on the screen.

1.5 Exercise 1
1 For each of the following, sketch appropriate graphs and find bounds for all
the possible solutions. ;
(a) x? —1=5,/x (b) x° +3x* S2x-2=0
(c) %=S5—x (d) 10 — x” = 2|x|
2 Use decimal search to solve each equation correct to 2 decimal places.

1.5.3 Iterative formulas

One method of solving an equation is to use a ‘zoom’ facility on a graph plotter.
An alternative method of reaching a solution was described in section 1.5.1. This
iterative method consists of taking an initial value and using a formula to obtain
a sequence of values which converges to the solution. Such formulas can be
obtained by rearranging the equation into the form
x = g(x)
which suggests the iterative formula x;,, = g(x;).

For example, to show that x” — 3x +2 =0 can be rearranged to give the

iterative formula
x;7 +2
Xi1 = 3

proceed as follows.
x7 +2
x —3x+2-0S3x°4+2=3xSx= 5

D
F ‘ : ; Xx; +2
A suitable iterative formula is x;,, = ;
3
84 FOUNDATIONS
fe a a a a

Alternatively, since the steps are reversible, it is sometimes easier to work

backwards,
oe as x +2
aor 3

= 3x =x? 42

— x? —3x+2=0

Example 2
Find the positive root of the equation x* — 8x — 7 = 0 correct to 3 decimal places.

Solution
Step 1: Obtain the iterative formula
x —8x-7=0
=e = 8x-F7
=> x = ¥/(8x +7)
which suggests the iterative formula

Xi41 = V (8x; +7)

Step 2: Sketch the graph to locate the

roots.

Step 3: From the graph, choose a

suitable value for x.

x, = 3 is nearest to
the solution.

Step 4: Use the formula to generate

the iterative sequence.

goes Se E30)
= 3.141 380652
X33. 0/9 129,979
 NUMERICAL METHODS 85

The solution to this problem can conveniently be obtained using a spreadsheet —

the results are shown below.

Iteration Input (x) Output (x;,,)

number

1 3 3.14138065
2 3.14138065 3.17912998
5) 3.17912998 3.18905898
4 3.18905898 3.19166031
5 3.19166031 3.19234114
6 3.19234114 3.19251928
? 3.19251928 3.19256589
8 3.19256589 3.19257808
9 3.1925 7808 3.1925812?
10 3.1925812? 3.19258211

x = 3.193 (to 3 decimal places)

1.5 Exercise 2

1 For the equation

2x? — 5x +1=0
find which of the following are possible iterative formulas and show how they
can be obtained.
Sx; —1 14 2x,
(a) X44 = /( 5 ) (b) x44 oar te

1 1 1
ic) Hid = 5 (s= ~) (d) xi41 = Ge,

(e) xin, = 4/(1 = Sx) (f) xig4 = 2x? — 4x, +1

Qi to oa (ee |) (h) xj41 = —4(1— 2x,’)
3 oe
(i) X44 = /(5x;-1-x;’) Cj) Aen =2+/( 5 *)

2 For the equation x° = 10

10
(a) (i) show that the equation can be arranged into the form x = V(=):
x
(ii) by letting x; = 2, and using an iterative formula, obtain the positive
solution for x° = 10, to 5 decimal places;
86 FOUNDATIONS
Oe
eS ne

(b) (i) show that the equation can be rearranged to x = \/(,/(10x));

(ii) by letting x; = 2 and using an iterative formula, obtain the positive
solution for x° = 10, to 5 decimal places.

3 Using an initial value of x; = 3 and an iterative formula, find a positive

solution of 2* = 3x to 4 decimal places.

4 (a) By sketching appropriate graphs, find an interval that contains the root of
1 ayx
(b) Show that x = \/(6,/x + 1) is a rearrangement of this equation.
(c) By choosing an appropriate starting value, solve the equation giving your
answer correct to 6 decimal places.

5 (a) Show that the equation x* + 2x = 1 has a root that lies between 0 and 1.
tan
(b) Show that x = si! is a rearrangement of the equation.

(c) Find the root between 0 and 1 correct to 5 decimal places using a starting
Yalueom, 2G).0" ane,
(d) What happens if you take a starting value of 2?

6 This question concerns the equation 2x* — 5x + 1 = 0 and three possible

rearrangements given in question 1. All answers should be given to 6 decimal
places.
(a) Show that the equation 2x* — 5x + 1 = 0 has one root in the interval (0, 1]
and another in the interval [2, 3].

(b) For the iterative formula x;,,; = (AS : )

(i) explain why the starting value x; = 0 cannot be used;

(ii) solve the equation using starting values of 1, 2 and 10. Record the
number of iterations used for each starting value.
(c) Solve the equation using the iterative formula
1 1

u i x;
and starting values: (i) 1 (ig)2 (iii) 10.
Record the number of iterations used for each starting value.
(d) Solve the equation using the iterative formula
1+ 2x7
carSalta es
and starting values: (i) ei (ri): (iii) 3.
(e) Comment on the suitability of each formula.
ee
EEE eee
 NUMERICAL METHODS 87

1.5.4E Convergence
In exercise 2 of 1.5.3 you solved the equation 2x* — 5x + 1 = 0 using a variety of
formulas and starting points, of which some converged rapidly, some converged
more slowly, some converged to a root in a different interval and some did not
converge at all!

It is plain that the choice of an iterative formula is critical if you are to obtain a
sequence which converges quickly.

SC

1 (a) Give a rough estimate for </10.

10
(b) Explain how the rearrangement x = — is obtained from x? = 10.
n
: 10
Use the iterative formula x;, 1= —, together with the starting value
xi
that you gave in part (a), to evaluate x7,x3,..., X49. What do you find?

It is helpful to know when a sequence is likely to converge before working

out all the values. These questions show how a graphical approach can help
to predict convergence.
10
2 Consider the rearranged equation x = —.
6
This is equivalent to the two simultaneous equations
10
y=x and y=
x
and its solution lies at the intersection of the two graphs.
10
(a) For.0' <x. <3; plot y = x and y= — on the same graph. You can now
x
illustrate the solution procedure for a particular starting value, say
x= Me

10
x4 aaa B(¥i) ge

Therefore x, = 2.5

(b) On your graph, plot and join the points (x1, x1), (x;, x2) and (x2, x2).
How could you have used your graph to locate the points (x;, x2) and
(x>, x) without doing any calculations?
88 FOUNDATIONS

(c) Without further calculation, plot y y=x

and successively join up the
points (x2, x3), (x3, Xa)5 (x3, Xa) .

(%4, x4) 0

The diagram you have obtained

is called a cobweb diagram.

10
y=2
x

3 What does the cobweb diagram of question 2 illustrate about the iteration
attempted in question 1?

4 (a) Draw a cobweb diagram for the a eo

function g(x) illustrated. (It is not
necessary to give g(x) an equation
— simply use the construction
described in question 2.)
Ne 508)
(b) What would happen in this case
to the sequence defined by
xX) x
BGs) g(x;)?

5 Draw similar diagrams for the following functions and describe their
behaviour.
(a) y eae (b) y y = g(x) ee

Vian o((X')

ae —>

a x Xi Xx

What happens in each case if x, is on the other side of the root?

6 In question 5, staircase diagrams should have been obtained. What

property of g(x) determines whether or not you get a staircase or a cobweb
diagram?
 NUMERICAL METHODS 89

7 By considering the staircase and cobweb diagrams above, explain how the
gradient of g(x) determines whether an iteration based upon
X41 = g(x;)
will converge or diverge.

8 For x° = 10
(a) show how to obtain the iterative formula

xX: =
ie
— xX:
a
—
i+1 a i x72

(b) using appropriate staircase or cobweb diagrams, investigate the

convergence of the iterative sequence obtained for different initial
values.

Example 3

wony(2)
Illustrate the convergence of the iterative formula

with starting value x, = 2 using a cobweb diagram.

Solution

Convergence occurs here because, near the root, the graph is sufficiently flat.

The convergence of iterative sequences is of interest in various areas of

mathematics. For example, the iterative process based upon the sequence
X41 = Ax;(1 — x;) is at the heart of the recently developed mathematical
theory of chaos.
 ONS
DATI
90 FOUN

_
 91

Foundations
Miscellaneous exercise 1

Find the equation of the straight line which passes through the point (2, 3)
and is parallel to the line y = 3x + 1.

Find the point of intersection of the lines having equations 3x + 2y = 1 and

2A OS Oe

Without using a graph plotter, sketch the graphs of:

(a) y=(x—-2)(e+2) — (b) y= x(x +2)

(c) y=x?(x —5) (d) y = (x —1)(« + 2)(x +3).

For a cylinder of radius rcm and height 4 cm, the volume V cm? and surface
area Acm” are given by the formulas
V=arh, A = 2nr + 2nrh.
If the volume of a cylinder is given to be 500m’, find h in terms of r, and
hence A as a function of r. Then use a graph of this function to estimate to
2s.f. the value of r which gives the smallest surface area.

Find the equation of the graphs obtained when the curve y = x° is translated:
“tl 0 2,
(a) through | (b) through | (c) through 7]

What is the equation obtained from y = x* — 6x by applying the following

transformations to its graph?
(a) A reflection in the x-axis (b) A reflection in the y-axis
—3
(c) A translation 2

Find the sums of the following arithmetic and geometric progressions:

(a) 7+15+23+...
to 20 terms
(b) 54+ (5+x)+(5+2x)+...
to 20 terms
fc) oO 124.884
(d) 1+0.14 0.014... to infinity

(e) 16—-4+4+1-—f+...
to infinity
92 FOUNDATIONS
ee ee

8 Find the sum of 1 terms of the geometric progression

0.23 + 0.0023 + 0.000023 +...

What happens to this sum as 1 becomes larger and larger? What fraction is
equivalent to the recurring decimal 0.23 (i.e. 0.232 323 232 323...)?

9 Evaluate:
15 20 n
(a) S°G+5) b) SOQt+1) «6 DOr +3) @) SS)
1 10 a1 t=

10 Evaluate:
20 20 20 20 40

(a) SoitS (5-4) b) S- 2+ 50 (8-1)- Soi

1 1 1 1 if

11 Factorise:

(a) x7+5x+4 8 (b) x*+3x = (c) 9x7 +9x —4

(d) 9x7 — 4 (e) 9x* — 9x —4

12 Solve the following quadratic equations by factorising:

(a) x7-—5x+6=0 (b) 4x*-—4x+1=0

(c) x +4.5%—-2.5=0. (d) (x—2)(x—3)
+ (x —2)(x — 4) =0

13 The distance d metres fallen by a stone thrown down a vertical shaft in

t seconds is given by d = 5t* + 2t. Find the time taken to fall 24 metres.

14 Express x* — 4x + 2 in the form (x — p)* + q. Hence sketch the curve

y =x’ — 4x + 2 and solve the quadratic equation x” — 4x +2 =0.

15 Find the solution sets of:

(a) (x = 4)(x+1)
>0 (b) x74x43<0
(c) (x-2)(x+3)(5—x)>0 (d) x —9x*
4+5x <0

16 Find the remainder when A(x) = 2x° + 3x* — 5x — 6 is divided by

(a)-x=4. (bee eee
Hence factorise A(x) and find the values of x which make A(x) = 0.

17 What number must be added to x° + 5x* + 10x +1 to make x + 1a factor?

 MISCELLANEOUS EXERCISE 93

18 Use the factor theorem to investigate whether

(a) x —2 isa factor of x° + 3x” —7x — 6,

(b) 2x — 1 is a factor of 4x° + 18x* + 6x — 5,
(c) x —1 isa factor of x° — 2x° + 11x* — 3x —7.

19 Express f(x) = x* — 4x7 + x + 6 as the product of three linear factors. Hence

solve f(x) > 0.

20 Show that x — 5 is a factor of x* — 3x” — 9x — 5. Hence solve

Me a 3K tO ES =);

21 Find the points of intersection of the curve y = x* — x — 3 and the straight

line y=,

22 The lengths in cm of the sides of a right-angled triangle are x, 2x + 1 and

2x — 1. Determine the lengths of the sides.

23 Sketch the curves:

@y=—1 y=}XG y=51

+ 2. 3(x — 2 1-2.
pe ee ee yee
94

Introductory calculus
.1 Rates of change

2.1.1 Introduction
Everything changes! Indeed, the rate at which things change may well be of very
great significance — the rate at which populations grow, the rate at which a
radioactive material decays or the temperature of an object cools are but a few
examples where the study of rates of change is important.

The area of mathematics developed to deal with problems concerning and

related to rates of change is known as the calculus. It is extremely important in
mathematics and in many related disciplines and forms a considerable part of
any study of mathematics at this and higher levels.

In this section we consider only rates of change for linear functions; we then
extend these ideas to more general functions.

Rates of change do not always involve time; for example, a conversion rate (like
the monetary exchange rate) enables us to convert from one unit to another.

For linear functions, a ‘rate of change’ is the gradient of a straight-line graph.

For example, you will be familiar with the

conversion relationship between oF
temperature in degrees fahrenheit (F) and in
degrees celsius (C). 200

F=2C +32
160
The graph of °F against °C is a temperature
conversion graph with °F on the vertical
axis and °C on the horizontal axis. eee

The gradient (slope) of the line is a = 2, 80

40

0 20 40 60 80 100 °C
 RATES OF CHANGE 95

The gradient of any linear graph of y against x ¥

can be found by choosing any two points on the
line and calculating
the difference in y-coordinates
the difference in x-coordinates

d
The symbol = is used to represent the gradient of such a straight line. The
dx
dF
gradient of the graph of fahrenheit against celsius would be written as IC:

9 dFi 19
2 gd aan
d
Equally, if s = 4t + 5, for example, then = = 4,

poet Exercise 1
1 For the line with equation y = 3x + 2, copy and complete these.statements.

dient of of line
(a) gradient line = b) —=
(b) co
dy y
6
2 For the graph of y = 6 — 3x :

(a) what are the ‘difference in y-coordinates’ and the 4

‘difference in x-coordinates’ from A to B? ‘

b) Find —.
dy .
am dx 2
1

0
d
3 Write down the gradient, a for each of the lines with equations
bs
(ayiy = 35 —7x (b) y=4+x

(c) y= —2x (d) y=3x-1

4 The graph of 2y + x = 4 is as shown.
(a) What are the ‘difference in y-coordinates’ and
the ‘difference in x-coordinates’ from A to B?
. dy

(b) Fin ae
96 INTRODUCTORY CALCULUS
:

5 Write down the gradient, 2 for each of the lines with equations:
be
(a) 2y=x+4 (bi yee x = 7

(c) x -—y=6 (d) 2y+-x=4

6 The cost of electricity consists of a standing charge of 702p and a charge of

2.87p for each unit of electricity used.

(a) Write the total cost, C pence, in terms of the number of units, 7.

(b) Find tsand explain what it means.

n
7 Copy and complete:
(2) con = aa

(b) y=4t+2>
(Cu—2
(dy y= Ses 1) > y = 10x 4-5

oor
dy

(e) y=4(3 -—5x) >

8 The circumference, C, of a circle of radius r is given by the formula C = 277.

dC
(a) Find a Explain what this rate of change represents.
r
(b) A wire is placed taut around the Earth’s equator. Approximately how
much extra wire would be needed to enable the wire to be pulled 2 metres
away from the surface at all points?

(c) Answer part (b) for a similar wire pulled taut around the Moon.

2.1.2 Linear functions

d
You know that, if, for example, <Y = 4 then the graph of y against x will be a
; ; dx
straight line of slope 4.

Example 1 q
Find the equation of the straight line through (5,2) with = — OE
G
Solution

oN a
Ae =) eee 2X
— GC

As (5,2) is on this line, then 2 = 10 +c >c = —8

So y = 2x — 8 is the equation of the line.

 RATES OF CHANGE 97

Many situations arise where variables are connected in some way, and you can
now investigate how their rates of change are related. This problem is illustrated
in the following example, and in the development section that follows it.

Example 2
A cylindrical water heater is 95 cm high and has a cross-sectional area of
2700 cm’. It is initially full of water. Water is run out so that the height of water
in the tank is reduced at 15cm per minute.
(a) Find expressions for the height (4) and the volume (V) of water in the tank
at time f¢.
dh dV
(b) Write down ae and ar and explain how they are related.

Solution
(a) After ¢ minutes
h=95 —15t

V = 2700(95 — 15t) — i.e. ~V =cross-sectional area x height

ey Oe gig OY S00 (—15)

di dees
5 Ve dh
O tp = 2700 x ‘dr

The rate of change of the volume is equal to the area of cross-section

times the rate of change of height. This idea is explored in the questions
which follow.

WA
1 A handbook issued with a microwave oven gives the following guide for
cooking a whole chicken from frozen.

‘Thaw on a low setting for 15 minutes per kilogram, then stand in cold
water for 30 minutes. Next cook on a high setting for 20 minutes per
kilogram, then let it stand for 16 minutes.’
Consider a chicken of weight w kg, so that the time taken in minutes for
the first stage is wu= 15w + 30.

(a) Write down a similar formula for v, the time in minutes for the second
‘stage.

(b) Hence find an expression for the total time taken, T = u + v, in terms
of w.
(c) Write down the values of sd ou and at What do these rates of
dw’ dw dw
change represent?
ay eee eld ed dv
(d) xplain why 7 aw nae
98 INTRODUCTORY CALCULUS
a nnn EEE

2 Let w= 3x +1 and v = 2x — 3.
du du
(a) Write down the values of — and —.
dx dx

(b) By substituting for uw and v, find an expression for y in terms of x for

each of the following:
(i) y=u+v (ii) y= 2u+v
(ii) y=a—v (iv) y = 4u —3v
d
(c) For each part of (b) write down the value of sed
dae
du dv d
(d) State the relationship between mei) “ and — in each case.
dx” dx dx
(e) What would the relationship be if y = au + bv, where a and b are
constants?

A firm charges a basic fee of £12 plus £5 per hour for each engineer
sent out on repair work. If one engineer is called out for t hours,
write down an expression for u, the charge in £, in terms of t.

A rival firm charges a basic fee of £9 plus £6 per hour for each
engineer sent. Write down an expression for the charge £v if one
engineer is called out for thours.

Comment on which firm is cheaper.

Write down es and we and explain their significance.

dt dt
In an emergency, a factory calls out three engineers from the first
firm and two from the second. Write the total cost £c in terms of u
and v.
de
(f) Deduce the value of — and explain its meaning.
dt

The work done so far should bring you to the following result.
 RATES OF CHANGE 99

2.1 Exercise 2
; d
1 A linear graph has — = 5 and passes through the point (—1, 2).
x
Find its equation.

2 Find the equation of each of the following lines:

(a) the line passing through (3,2) with - = —2;:

x

(b) the line passing through (4,3) with = = ;

(c) the line passing through (—6, —1) with - =

x

a
3 A line passes through the points (1,5) and (4,11). Find 7 and the equation of
x
the line.

4 (a) A plumber charges £5 for a call-out plus £7 per hour for labour.
(i) Write the charge £C as a formula in terms of t, the number of hours
taken to do the job.

(ii) What is the value of oe

(b) Another plumber charges £6 per hour for labour, and for a job lasting
3 hours the bill is £26.

(i) Write down the value of ae

dt
(ii) Hence obtain the charge £C as a formula in terms of t, the number of
hours taken to do the job.

5 The marks obtained in a test ranged from 25 to 50. They have to be rescaled
to range from 0 to 100. Copy and complete this table.

Test mark, T 25 26 50
Rescaled mark, R 0 96 100

dR
(a) Find
Fin —.
aT

(b) Hence express R in terms of T.

6 The growth of the population of Britain in the first half of the twentieth
century was approximately linear, rising from roughly 38 million in 1900 to
48 million in 1950.
100 INTRODUCTORY CALCULUS
nnn, EE aS

(a) Find an expression for the population P millions of people t years after
1900.
ec!
(b) Find ae What does it represent?

(c) What does your formula give for the population in 1998? Comment on
your answer.

7 Letu=44+2x andv=S5— 4x.

(a) Write down ae and a

dx dx

(b) Calculate = for each of the following functions, without expressing y as a

function of x.
(i) ee (ii) y=u-v (iii) y=3u+v
(iv) y=u—3v (v) y=3u+2v (vi) y = 2u—3v

(c) For each part of (b), substitute for u and v, then express y in terms of x.

Hence check the value of cy

dx

8 Making steamed puddings in a pressure cooker involves placing a basin

containing a dough made from flour and suet into boiling water in a pressure
cooker. To allow the raising agent to take effect, a short steaming time is
allowed at low heat and then the cooker is brought to the appropriate pressure
for the remaining cooking time.

One recipe for four portions recommends:

Amount of flour Steaming time Cooking time at pressure
200 grams 15 minutes 25 minutes

To adjust the recipe for other quantities, it is suggested that for every
additional 15 grams of flour an extra 1 minute should be added to both the
steaming time and the cooking time.

Let f grams be the amount of flour, s minutes the steaming time and p minutes
the cooking time at pressure.
: d
(a) Write s in terms of f. Hence find the value of = and describe the meaning
in words of this rate of change. f
d
(b) Write p in terms of f. Hence find the value of o and describe the meaning
in words of this rate of change. f
(c) Find T, the total cooking time in minutes. Hence find the value of iy and
describe the meaning of this rate of change. :
eeee eee eee
RATES OF CHANGE 101
102

Introductory calculus
2 Gradients of curves

2.2.1 Locally straight curves

Examining what is meant by the gradient or slope of a curve is a rather more
sophisticated problem than that of the straight line, although it is very closely
related. Consider the curve sketched below.

>
Be

If you imagine moving along the curve from left to right, then the steepness of
the curve is constantly changing (remember that for the straight line it was
constant). Because the gradient changes continuously we talk about the gradient
at a point on the curve. If you were to ‘zoom’ in more closely to the points
labelled on the graph, then you would see the following.

ra Ne

The gradient at A is positive, at B, C and D it appears to be zero, and at E it is

negative. In every case the curve is locally straight.

Since you are already familiar with the gradients of straight lines, the fact that
zooming in on the curve shown gives straight lines is an idea worth exploring
when trying to find the gradient of a curve at a point. The questions of 2.2a
consider what happens when you zoom in at various points on a number of
different curves.
 GRADIENTS OF CURVES 103

These questions require the use of a computer or graphical calculator to

enlarge and zoom in on part of a graph.

1 (a) Input the graph of y= x° — 7x* + 8x +7 for0<x <5.

(b) Zoom in to the point with x = 2 and redraw.

(c) Repeat, increasing the magnification. What do you notice?

(d) Zoom in at x = 4. Note your observation.

When you zoom in at some point on a sufficiently smooth curve, the curve
starts to look more and more like a straight line. The diagram below shows
this for the graph y = x” — 2.

2 (a) What would you expect to see if you zoomed in at (0, —2) on the graph
y=x’—2?
(b) Check your answer on the graph plotter.

3 Investigate the local straightness of the following graphs.

(a) y= |x| (|x| is entered on some graph plotters as ABs(x).)
(b) y= 100x7 (c) y= Int(x) = (d) y= |x" —4|
104 INTRODUCTORY CALCULUS
i UEEEEEIIIIISI EEE SEESSSSSSSSSSEESSSSSEnSS

4E (a) Use the graph plotter to obtain the graph of y = ¢sin 3x. Describe
fully in words the transformations from the graph of y = sinx to that
of y= ¢sin Shee

(b) Use the graph plotter to obtain the graph of:

(i) y==ssinis + sin 3x
(in) y= sina 3 sin 100x

(c) Can you invent a function whose graph looks just like that of
y = sinx under normal magnification but not under magnification
x 1000?

When you zoomed in on some of the graphs in 2.24, you found that some
functions, such as y = |x|, were not ‘locally straight’ everywhere. For example,
the graph of y = |x” — 4| is locally straight at all points except (2,0) and (—2, 0).

However, the cubic graph was locally straight everywhere. In fact, so are all
polynomial graphs. Suppose you could superimpose the ‘zoom’ view onto the
graph at a particular point; you would see the following.

The line AB is called a tangent to the curve at P.

 GRADIENTS OF CURVES 105

2.2.2 Gradient graphs

The previous section showed that the value of the gradient may be different for
different points on a graph. On the graph shown below, if you imagine moving
from left to right, you would experience:
@ a numerically large negative gradient which increases to zero at A;
@ a positive gradient which increases from A to B and then decreases to zero
ateG:

@ a gradient which finally decreases through numerically larger and larger

negative values.
If you plot the gradient at each point against x, you obtain (very roughly) a
gradient graph as shown.
106 INTRODUCTORY CALCULUS

A, B and C are special points; A and C are the only points where the gradient is
zero and B is the point where the gradient is maximum.

This local maximum is both a This is a local minimum.

stationary point and a turning
point.

This is a stationary point but This is a turning point but not

not a turning point. a stationary point.

Note that the point at D is not a stationary point because the graph at D is not
locally straight, and therefore the gradient is not defined at this point. At a
stationary point the gradient must be equal to zero.

The maximum and minimum points are generally described as local maximum
or local minimum points, because they are not generally the maximum or
minimum values of the function itself, simply the maximum or minimum value
around the particular point on the function.

Also note carefully the distinction between stationary points and turning points,
as illustrated at points B and C, for example.
 GRADIENTS OF CURVES 107

Example 1
Find the stationary points and/or
turning points on the graph of
y=
|x -4|

Solution
A and C are turning points (local
minima) but are not stationary
points. B is both a stationary point
and a turning point (a local
maximum).

-2.2 Exercise 1

1 Copy each of these graphs. Directly beneath each one, sketch the
corresponding gradient graph, using the same scale for x.

Mark any points you think are special and state the important features of
each graph.

(a) y

(c) y (d) v
108 INTRODUCTORY CALCULUS

=
2 Sketch each of these graphs and its gradient graph. Start by deciding what
happens to each gradient graph when x is near zero and also when x is
numerically large (either positive or negative).

ZEA
3 Two gradient graphs are sketched below.

) For each of them sketch a possible (x, y) graph.

) How many other possible (x, y) graphs are there for each of them?
eee
 GRADIENTS OF CURVES 109

2.2.3 Obtaining a gradient

You are already able to state some gradients exactly. For example,
d
Neer eG
dx
. yn lt:
For non-linear graphs the notation is still used, but now this represents the
Be
gradient along the tangent at the point P.

Seed a
A practical method for obtaining = at any point on a curve is introduced below.
dx

2B

1 Ona sheet of graph paper, draw accurately the graph of y = 1x? for values
of x between —3 and 3, using the same scale for both axes.
(a) Draw, as accurately as possible, the tangent at the point (1.5, 1.125),
; Seber! 4
and hence measure the gradient = of the curve at this point.
x
(b) By repeating this process as necessary, and using the symmetry of the
graph, copy and complete the following table.

x <= eoMlme\e alk el 1ES, 2

d
Gradient ie
dx

d
(c) Plot all the points («z) to obtain the gradient graph for y = 1x?
dx

Your points should be approximately on a straight line through the

origin (the line y = x). So,

=
Se dx

2 Plot the graph of y = 21 + 4x — x’.

(a) Find the gradient at a number of different points on the curve.
(b) Check that your answers satisfy the equation of the gradient graph
dy
—=4-2x.
dx
 110 INTRODUCTORY CALCULUS
Oe ee ae ea een

ead
You found that the graph of y = 1x? has a gradient graph with qe its
equation.

Soh
Sy ile) ih oy Bey es

In the following questions you will investigate, by drawing graphs, the gradient
functions for various quadratic functions.

eC

1 (a) Plot the graph y = 1x? + 1, letting 1 unit be represented by 2cm on

each axis.
d ;
(b) By drawing an accurate tangent, confirm that i =) at the pomt (253):
x
(c) Find sufficient gradients to sketch the gradient graph, and state its
equation.

(d) How are the graphs of y = ve +1,y= 1x? — 2 and, generally,

y= tx? + call related to that of y =1x’
xt P)

(e) What can be deduced about the gradient graph for any curve of the
form y = 1x? Ge

2 (a) Plot the graph of y = x’, letting 1 unit be represented by 4cm on each
axis.
(b) Find sufficient gradients to sketch the gradient graph, and state its
equation.
(c) Carefully explain why the gradient of y = x” is twice that of Vi 1x?
for corresponding values of x.
(d) What would you expect the gradient graph for y = 3x* to be?
(e) What can be deduced about the gradient graph for any curve of the
form y = ax’?

it 3 What is the equation of the gradient graph of y = ax” + c?

 GRADIENTS OF CURVES 111

You have seen some justification for the following result:

Example 2
Find the gradient of the graph of y = 3x* + 1 when x = 2.

Solution
d
ya3x +1 > =6x
x

At x — 2, the gtadient.is 12.

The gradient of a curve at a point may be found very accurately and much more
quickly by using the zoom facility on a calculator or computer, or by obtaining
its value numerically in the following way.

You have seen that when you zoom in on the graph of y = x” the curve is
everywhere locally straight.
y

24

If you zoom in at (3,9), you know the property of local straightness will mean
that the graph looks increasingly like a straight line which becomes more and
more like the tangent to the graph at (3,9). So, to find the gradient of the
tangent, you can use the curve itself.
112 INTRODUCTORY CALCULUS

Using the point (u,v) you will get an approximate gradient for y = x atx =3
by calculating
: v=
approximate gradient = ———
us
You could take the point (u,v) to be (3.1, 9.61), for example, and this would give
an estimate of the gradient of the tangent at (3,9). You could get a more
accurate value for the slope by taking u even closer to x = 3, for example, 3.01.

.2D

Find estimates for the gradient of y = x” at x = 3 by taking u = 3.2, 3.1, 3.01,

3.001 and so on. Describe what is happening to the estimate as you move u
closer to x == 3. State the gradient at x = 3.

It is often convenient to use function notation when finding gradients

numerically.

(a+ h, f(a + h))

 GRADIENTS OF CURVES 113

Example 3
Find the gradient of the graph of f(x) = 2x? — 3x* at (2,4).

Solution
£(2) =4 and £(2.000 01) = 4.000 120001...

and so the gradient of the graph is approximately

_ 4.000 120 001 — 4
f(2) = eae = 12,0001

Using smaller differences in x will result in values closer and closer to 12.
This limit is the gradient of the graph.

If f(x) = 2x° — 3x? then f’(2) = 12.

2.2 Exercise 2

Calculate numerical estimates of the gradients of the functions at the point

indicated.

1 y=? at (2,8)

2 y=2x?4+3 at 3,21)

3 y= x? 4+4x at (1,5)

4 y=x—x’ at (2,4)

5 y=x' + 5x at (1,6)

6 y=-x" at the points with x-coordinates:

(aes (bye (c) 0 (d) —2

Use your results, together with that for x = 1, to predict the gradient when
the x-coordinate is (e) 15, (f) 3.6, and confirm these guesses by calculation.

2.2.4 Gradient functions

The great speed and accuracy of numerical methods, compared for example
with the use of a gradient measurer, enables you to construct many gradient
functions.
 114 INTRODUCTORY CALCULUS

1 (a) Draw a rough sketch of

'
what you would expect for
the gradient graph of y = x
|‘
i (b) Use a numerical method to
calculate the gradient at
c= Oe 1 2 and 3:
SSS
SENEULUIES
NOELLE
(c) What is the gradient
' function?

EES
SESSLER
SEE
RONNIE

Many graph plotters will calculate gradients numerically for several values of
x and plot them on the graph. This gives you an accurate picture of the
gradient graph. You can then make a sensible guess at the equation of the
gradient graph and check your conjecture by superimposing. Use a graph
EYLTTE
ESSIEN
SNORE
COLIN
NMOL
plotter with this facility to check your answer to question 1(c) and to answer
the rest of these questions.

2 For a function of the form

3
y= ax
what does the gradient function appear to be?
ELLER
SSN
ESPRENOS
SNOINN
NESS
SA
/ A general polynomial is built up from
multiples of simple powers of x. For
example, x + x’ is built up as shown
_ below.

fe On 1 3
aaeexo Oka ee he
 GRADIENTS OF CURVES 115

When x increases from 1 to 2, x increases by 1, x” increases by 3 and x + x”

increases by both amounts, i.e. 4.

Does this remind you of a result you discovered in section 2.1?

In section 2.1 you found that if y = aw + bv, where uy, v are linear functions
of x, then

d du du
sea ayy easy re
dx dx dx
You might expect functions that are locally straight to behave in a similar
way to linear functions.
d
3 If y = ax + bx’, check that =.= a-+ 2bx for various values of a and b.
bd
4 If y=a+bx-+cx* + dx’, find the equation of the gradient graph.

SE If y = ax” + bx”, find the gradient function.

The questions of 2.2z above have provided considerable evidence for the
following result.

“A

- is called the derivative of y with respect to x and b + 2cx + 3dx’ is called the
id
derived function.

Example 4
(a) Find the gradient of the graph of y = 1 —3x+ 2x’ at the point (2, 3).

(b) Hence find the equation of the tangent at (2, 3).

Solution

(a) dy =
d
—3 + 4x. At the point (2, 3), ae ==3+4x2=5
dx dx cs
 116 INTRODUCTORY CALCULUS

(b) The tangent has gradient 5 and passes through (2, 3). If (x, y) is any other
point on the tangent, then
Ve)
= Sy ee
ey en
5 =D
The equation of the tangent is y = Sx — 7.

2.2 Exercise 3
1 Use the rules you have discovered to find the equation of the gradient graph
for each of the following.
(a) y = 3x" +4 (b) v = Su? — 2’
(c) y=6—x° (d) s=4¢-2
2 For the graph whose equation is y = 2 + 5x*
(a) write down the equation of the gradient graph;
(b) write down the gradients of the graph at:
Gy Us7) Gi) 0,22). Gi) the pome where x= — 1
3 Find the gradients of each of the following graphs at the given points.
(a) y= 3— 2x? at (0,3) and (2,-13) (b) y= 5x — x” at (2,6) and (4,4)
4 Find the equations of the tangents to each of the following graphs at (2, 3).
(a) y=3+2x—2x7
(b) y=x?—5
(c) y=7Px?
— x?

5 Find the equation of the tangent to y = x + 2x* at the point whose

x-coordinate is 3.

6 Find the equations of the tangents to each of the following graphs at (0, 5).
(a) y=S4+x- x?
+ 2x?
(b) y= 5 —3x
(c) y=5 44x" 4 3x°
eee

2 Tasksheet E1 — Tangents and normals (page 551)

2.2.5 Differentiation from first principles

The introductory study of calculus so far has been based upon the idea of
local
straightness. When you zoom in at a point on a locally straight curve,
the curve
appears to be a straight line and this enables you to find the gradient.
 GRADIENTS OF CURVES 117

However, zooming in does not always make a curve appear straight. From a
spaceship, the Earth appears to have a smooth, spherical surface, but from a
closer vantage point, enormous imperfections in the surface are apparent.

To be able to differentiate a function you require the graph of the function to be

locally straight, but so far you have no way of knowing whether a graph really
is locally straight.

One important aspect of mathematics concerns giving rigorous arguments to

prove results indisputably. It seems that you would have to zoom in forever
in order to confirm that graphs of relations such as y = sinx or y = x” really
are locally straight. We now consider the mathematical technique for doing this.

To be able to find the numerical gradient of —

y = x at the point (3,9) you have previously
considered the gradients of lines joining (3, 9)
to nearby points on the graph.

P is the point (3, 3°). Q, a nearby point, has

coordinates (3+, (3+4)’).

The gradient of the chord PQ is

yo-yp (3+h)/-3
bh+h ° oat OM f
xgo=xp (3+h)=3 > Bb
Sey
118 INTRODUCTORY CALCULUS

As h becomes smaller and smaller the gradient of PQ becomes closer and closer
to 6. The chord PQ becomes closer and closer to the tangent at P.

You can see that for any small h, the gradient is close to 6. You therefore know
that zooming in further (by taking smaller values of /) cannot cause any change
in this result: the graph is locally straight at x = 3 and has gradient 6 at that
point.

Making the value of / smaller and smaller is termed ‘letting / tend to zero’ and
the notation for this is
h—0
Finding the value of 6 + h as h tends to zero is termed ‘finding the limit of 6 +h
as h tends to zero’ and the notation for this is

eee

This idea is explored in the questions which follow.

.2F

1 Evaluate: (a) jim (Sh — 2) (b) jim (3 + 2h)

=> ==i0)

Limits can be obtained for ) tending to values other than 0.

ane h(h—2 -.
The limit of Ss) as h tends to 2 is written as lim plese),
h —2 h>2 h —2

This limit cannot be evaluated simply by putting / equal to 2, because Q is

undefined. However, for h # 2, the factor h — 2 can be cancelled.

Mb 2)
lim ime
pepe = ee
2 Use this method to evaluate:

h(h — 2) Sh* — 2h 4h°

—3
eer:
] UN Manes code

Pa
. ee re

era
hi a .

(d) ie h —2)(b
)(h + 2) ) i a
25-18
h—2 h—2 eae h+3
 GRADIENTS OF CURVES 119

You have found a number of limits of the type

lim
li 2(h + 3)
3) ==6
A limit of the form
Pe et eS
lim —— = =
h0 2+h 2

can be found simply by substituting zero for / in both numerator and

denominator. However, this method cannot be applied if the denominator is zero
when h is zero.

Example 5

ee TS
Ae a cree
Solution
For b £0 4h—h (4-h*)h 4-h
as "+h (9+h)b 9+h
4hp—h? 4
rire as
La Obs he 9
The notation developed for limits can be used to give a general definition for the
gradient of the graph of a function.

(a + h, f(a + h))
120 INTRODUCTORY CALCULUS
SSS 8 oe

Finding a gradient by means of the limit is called differentiating from first

principles.

Example 6
Differentiate y = 3x* — 4x at (3, 15) from first principles.

Solution
(3 +h) =3(3+h) —4(3+5)
= 3(9+
6h +h’) — 12
— 4h
= 15+ 14h + 3h’

The gradient is jim

18 + ee
14h hag
+ 3hie— 18
lim Poe i = jim (14
+ 3h) = 14

For any function f, a general formula for f'(x) can be found from first principles
using the definition

Example 7
d
If y = x’, find - from first principles.

Solution
dy |. (x +b)? =x?
dx a h
De aseae2
h-0 h

= lim (2x +h) = 2x

h-0
 GRADIENTS OF CURVES 121

Example 8
1 dy
Ify=-
be ae find —de fromfro first principles.
inci

Solution
1 il

Cel et) Pecans

dx ie h

. [x—(x+hb)
| ] BeSS NE se
lim,
|hx(x + h) |

bai lea Ma Si iow

~ p0|hx(xth)| boolx(x+h)] 2
toedy 1
sy => =
y oe Obs x?

The method of differentiating from first principles is important, since it is the

way in which mathematicians have proved all the known derivatives. It can also
be used to prove the correctness of the various rules of differentiation.

2.2 Exercise 4
1 Suppose y = 3x7.

(a) Find the gradient at (1,3) from first principles.

(b) Find a general formula for = from first principles.

2 Differentiate 5x* + 3x with respect to x, from first principles.

3 Differentiate 4x* — 2x +7 with respect to x, from first principles.
1 dy —2
ow, f from first
4E Show, first pringiples, thatiliy
principles, that if y= =~, , th then = Qr4 5p :

5E Differentiate x* from first principles.

2.2.6 Leibniz notation

Calculus is the study of the changes of a continuously varying function. In
section 2.2, you have looked at the gradients of locally straight curves and, in
particular, the gradients of ‘graphs of polynomial functions. At any point, the
gradient of a locally straight curve is the same as the gradient of its tangent at
that point. The concept of a derivative arose chiefly as the result of many
centuries of effort spent in drawing tangents to curves and finding the velocities
of bodies in non-uniform motion.
122 INTRODUCTORY CALCULUS
a SE eS eee ee

Isaac Newton was born in 1642 and entered Cambridge University in 1660,
quickly mastering all the mathematics known at that time. In 1665, the year of
the Great Plague, he invented his ‘method of fluxions’, which was a method of
dealing with varying quantities. If a quantity, say x, was a function of time ¢,
dx : ; :
then Newton used the notation x to represent —. In mechanics, the notations x
dt
and y are still used to represent velocities in the x and y directions respectively.
During the Great Plague, Newton retired from Cambridge to his home in
Lincolnshire. Here, he investigated various applications of his method,
including finding the equations of tangents. His treatise on calculus was written
in 1671 but its publication did not take place until 1736, nine years after his
death. It is interesting that in his Principia of 1687, in which he dealt with both
terrestrial and celestial mechanics, he relied on geometry and did not use
fluxions.

Because Newton’s results were published so late, a bitter controversy arose

between the supporters of Newton and of the German mathematician and
philosopher, Leibniz. Gottfried Leibniz (1646-1716) had not yet started his study
of mathematics in 1671 and it was not until 29 October 1675 that the first
mention of the calculus notation in use today appeared in his notes. Having
noticed that differentiation reduced the degree of a polynomial, he used the
; x d : :
notation 5 but soon changed this to dx. In 1676, in correspondence with the
Royal Society, Leibniz learnt that Newton had produced some important results.
Newton supplied him with a variety of theorems but referred to his method of
fluxions by means of two anagrams from which Leibniz deduced nothing. In a
reply he gave his rules for dy and dx. Newton, in his Principia, mentioned that
Leibniz had discovered independently a method ‘which hardly differed from mine
except in words and symbols’.

The greatest merit of Leibniz’s work was his creation of a mathematical

Pee ' d d
symbolism. Besides his introduction of dx, he invented the notations — and Te
be x
d
In the text you have seen that = represents
x
the difference in y-coordinates
the difference in x-coordinates
along the tangent to the curve. Using the notation invented by Leibniz,

a dy
y=lx >—=6x 2 or
d 3
—(2x) =6x 2
dx dx Cee
Leibniz also adopted a useful notation for the second derivative — the function
obtained when you differentiate the first derivative. In mechanics, the second
derivative of displacement with respect to time (which is the acceleration)
is
 GRADIENTS OF CURVES 123

; PR ude
often written as X or —. Using this notation, —
di?

x =ut+tat te a

or x=u podsa ges—_—=Uu fie ei

ZT,= a
Z dt di? 2
A further adaptation of this notation, extensively used later in the text, refers to
the first and second derivatives f’(x) and f(x) of a function f(x). For example,
f(x) =x° = 2x?
df bey ale tJ
a f'(x) = 3x° 2 — 10x 4

d’f
—, =f" (x) = 6x— 40x
124

Introductory calculus
Optimisation
2.3.1 Graphs and gradient graphs
Optimisation is the process of producing the most favourable outcome: the
greatest food supply or the least pollution, for example. Decision-making can
sometimes depend upon an analysis using calculus and stationary points. Before
tackling such optimisation problems we shall review the use of stationary points
in graph sketching.

& 3A y

Part of the graph of

yox—2x*+x41
is as sketched. >

1 (a) Explain how the details of the graph shown above are obtained.

d
(b) What features of the (2) gradient graph sketched below can you

relate to the shape of the (x,y) wien

The graph of
d
——_ 3x" — 4x41
dx
is shown here.

(c) Complete the gaps in the original (x, y) graph.

How can you be certain that your sketch is roughly correct?
 OPTIMISATION 125

2 Make a copy of this sketch graph and indicate on it:

d
(a) the parts of the graph where = is F
positive; me

d
(b) the parts of the graph where = is
negative; i

d
(c) the points on the graph where = is
x
Zero;

(d) any local maxima or minima.

} d
3 The points shown on an (x,y) graph and corresponding («2 graph
have the same x-coordinate. #

“ oo

Sketch a possible shape for the part of the (x,y) graph near the point
shown, carefully explaining your answer.

eee
d
4 One point of an (x,y) graph and a segment of the corresponding (2)
(
graph are shown. -

Sketch the (x,y) graph near x = 0, carefully explaining your answer.

d
The values and behaviour of = can be used to help sketch graphs.

Knowing the gradient of a graph at a point tells you what the graph is like near
the point and not just at the point itself.

d
Finding the stationary points (where = = (0) can help you to determine quickly
the overall shape of the graph. -
 126 INTRODUCTORY CALCULUS
ee a a Te

Example 1
(a) Sketch the graph of y = x” — 5x. :

(b) Indicate on your sketch where = is positive and where it is negative.

dy
(c) For what value of x does —— = 0?
dx
(d) Is this point a maximum or a minimum? State its coordinates.

Solution
(a) y= x" — 5x or y= x(x — 5)
(b) +, —, as on the graph
d
(c) yan — Sx =2x-5
dy
ae = 0 es
dx
(d) It is clear from the graph that it is a
minimum. When x = 2.5,
y =2.5° -5x 2.5 = —6.25.
The coordinates are (2.5, —6.25).

2.3 Exercise 1

1 Sketch y = (x — 1)(x — 2)(x — 4). What extra information about the graph
could be obtained using calculus? (There is no need to find this extra
information!)
 OPTIMISATION 127

eas
2 Three cubic graphs are sketched below.

2)
Sketch the comtesponding (~ =) graphs and relate features of the gradient
x
graphs to the shape of the (x, y) graphs.

3 A cubic graph passes through (0,2) and has

gradient graphas shown.
gle
Sketch the (x,y) graph.

4 The distance travelled by an object in time t is given by

c= Pf [6 Pit

(a) Find = and sketch the graph of S against ft.

(b) Sketch the graph of s against t.

(c) By considering the two graphs, describe the motion of the object.

2.3.2 Quadratics and cubics

It is easy to differentiate quadratic and cubic functions and hence to find the
stationary points.

@ »
d d
For y= x? — 12x +2, find eos and solve the equation oi 0.
dx dx
Hence find the stationary points and complete the graph.
128 INTRODUCTORY CALCULUS

du
2 For u = 3x* + 6x +5, solve the equation oeite 0.
x
Hence find the stationary point and sketch the quadratic graph.

3 Use your sketch graph for question 2 to show that y = x° + 3x* + Sx +7

has no stationary points. Hence sketch the cubic graph.

4 Use a graph plotter to obtain cubic graphs with equations of the form
y = x° + ax for various values of a. Describe how the value of a affects the
shape of the curve.
d
Relate what you discover to -- and stationary points.
x

Example 2
Sketch y = —x? + 27x — 2.

Solution
For large x, the graph has roughly the same shape as that of —x°.

At the stationary points, t

dy
eae
dx
—3x + 27 =0
3(9-— x”) =0
x=3o0r—3

When x = 3

oo
75
When
x = —3

y= = 3)) 27 (3) se
 OPTIMISATION 129

2.3 Exercise 2

1 Find the stationary points on the graph of y = x° — 12x + 5. Hence sketch the
graph.

2 Repeat question 1 if the graph is that of y = 2x? — 9x* + 12x —7.

3 For each function (a) to (f) given below:

(i) find all the values of x, if any, for which = = 0 and hence find the
x
coordinates of all the local maximum and minimum points;
(ii) sketch carefully the graphs of the functions below and indicate on your
d d
sketch the parts of the graphs where = is positive and the parts where =
ie a
is negative.
(a) y= Sx — x (b) y = (1—x)’
(c) y=x? — 3x7 +5 (d) y= 4x —x* —4
(ec) y=2x°-9x7
+12 (f) y= xt — 8x? +12
4 The sketch graphs below are those of y = x* — 6x and y = x° — 6x’.
(a) Find the x-coordinates of A, B and C. Explain the relationship between the
x-coordinate of B and the other two x-coordinates.
(b) Find the x-coordinates of D, E and F. Does the relationship you noticed in
(a) hold for this graph?

2.3.3 Maxima and minima

Decision-making often depends upon choosing the value of one variable so as to
maximise or minimise another variable. For example, it might involve
maximising profits, minimising the amount of material used in a design,
maximising the number of customers served each hour and so on. Calculus can
be of great help in this decision-making process. The stages in such a process can
be seen in the following simple example.
130 INTRODUCTORY CALCULUS
nnn EEE

A circular piece of paper is folded into a cylindrical paper case for a cake. Where
should the paper be folded to create the container of greatest volume?

What do you think happens to the volume of the cylinder as it changes from a
tall thin cylinder to a short fat cylinder as shown below?

| -S
To use calculus methods to solve this problem you can express the volume, V, in
terms of a suitable variable length.

Consider a circle of paper of fixed radius R, folded at radius r. The paper folds
up to form a cylinder, of volume V.

Volume = area of base x height

V=ar(R—7r)
=7Rr —orr

dV
papa, 2nRr — 37° (Note that R is fixed and does not vary with r.)

= m7r(2R — 3r)
 OPTIMISATION 131

dV
26 = 0 when r = 0 (zero volume) and when r = 2R (maximum volume). The
maximum volume is therefore

mR) 2 (R—2R) =47R°

V can, of course, be expressed in terms of other lengths, for example fixed
diameter D and variable height h. It could also be expressed in terms of an area,
such as the area of the base of the cake case. It is important to choose quantities
which make the calculations reasonably easy.

The questions below further illustrate the method.

.3C

The population density (number of residents

per unit area) of many cities depends roughly
on the distance from the city centre.

For a particular city, the population density P

in thousands of people per square kilometre
at a distance r kilometres from the centre is
given approximately by
P=5+30r—
15/7
1 What is the population density in the centre of the city?

2 Sketch a graph of P against r. For what values of r is the formula definitely

not valid?

dP '
3 Find 5 and calculate the rate of change of population density at a radius
r
of

(a) 0.5km (b) 1km (c) 2km from the city centre.

iP
Sketch a graph of - against r.
r
4 Use the two graphs to describe in words how population density varies
with distance from the centre.

Example 3
During a promotion drive, an electrical retailer sells a particular make of
television at cost price. She finds that, at this price, she sells twenty televisions a
week. However, according to a market survey the demand would fall to zero if
the price were increased by £40.

By what amount should the retailer increase the price to make the maximum
weekly profit?
 132 INTRODUCTORY CALCULUS
nee

Solution
For a price increase of £1 you can model the number sold by
N=20-4]
f£I is the increase above cost price and so the profit (in £) is

1P = ING P

= (20-41
= 201-17
dP 0 40\ 7
P is maximum when — = 0.
dl
dP dP
Since —, = 20 — 1, =, = 0 when I = 20

The retailer should increase the price by £20.

2.3 Exercise 3

1 The fuel economy of a new car was measured when it was test driven at
various speeds in top gear. For a speed of v miles per hour, the fuel
consumption, F miles per gallon, was found to be roughly modelled, for
30 < v < 80, by the formula F = 25 + v — 0.0120”.
dF
(a) Find F and a when v = 35 and when v = 60. What do these results
Vv
indicate about fuel economy?
(b) What speed is most economical for this car?

2 As a result of a survey, the marketing director of a company found that the

demand for its product was given approximately by the linear equation
n = 30 — 2P where n is the demand, or the number of items that will be sold
(in millions), at a price £P.

If 7 million are sold at £P each then the revenue (money taken) will be £R
million where
R = nP = (30 — 2P)P = 30P — 2P*
Pe railse
(a) Find ap and explain what it; means. What is the best selling price?

(b) Calculate the value of = when P = 5 and when P = 10.

(c) For what range of selling prices would the revenue rise if the price were
increased a little?
 OPTIMISATION 133

3 A rectangular strip of plastic of width

20cm is folded into a length of guttering
as shown.
Where should the folds be located to <>
20 cm
enable the gutter to carry as much water
as possible?

4 A new housing estate started with a population of approximately 500 people.

(a) It was planned that it should grow by roughly 100 inhabitants each year.
Find an expression for the intended population P of the estate t years after
its opening. Find ae and explain what it represents.

(b) For various reasons, the new estate did not grow as planned and the
population was better modelled by the quadratic expression
P= 100(5 +7 — 0.257)
What was the rate of change of the population after 1, 2 and 3 years?
What was the maximum population of the estate? What happened to the
estate?

5 To express 10 as a product of two numbers which have the least possible sum
10 Ay se 10
you can take the two numbers to be x and — and try to minimise x + —.
ae oe
10
(a) Use a graph plotter to sketch x + —.
x
(b) From the graph, estimate some possible answers to this minimising
problem, depending upon what types of numbers are allowed.

2.3.4 Graphical optimisation

Calculus has been described as the most powerful and useful invention of
mathematics. Applications of calculus range over many areas of mathematics,
physical science, engineering and the social sciences. Section 2.3 has concentrated
upon just one type of application — optimisation.
 134 INTRODUCTORY CALCULUS
ES

When tackling real problems, the expressions obtained may be difficult to

optimise algebraically. In such cases you can use a graph plotter to observe the
overall shape of the curve and hence obtain from the graph the approximate
positions of any local maxima and minima. ;

In question 3 of exercise 3 you should have obtained proportions for the

rectangular guttering of

1 @ 1

Gutters can, however, have splayed sides. Intuitively, do you think that the
amount of water which the gutter is able to carry can be increased by splaying its
sides while keeping the 1:2:1 ratio as above?

3D

| Find an expression for the cross-sectional area of the splay-sided gutter in

| terms of some chosen variable quantity. Use a graph plotter to check if your
intuition was correct.

2.3 Exercise 4

100

A cardboard box is to be made from a rectangular piece of card, 100 cm by

40cm, by cutting and folding as necessary along the dashed lines shown in the
right-hand diagram. The problem is to find the values of /, w and h which
maximise the volume, V.cm’.

(a) Explain why 2/ + 2w = 100. Hence express / in terms of w.

(b) Similarly, find / in terms of w.
(c) The volume of the box is given by V = whl. Use your answers to (a) and
(b) to show that V = w(40 — w)(50 — w).
(d) Find, approximately, the value of w corresponding to the maximum
possible volume. What approximate dimensions will the box then have?
 OPTIMISATION 135

2 Small open-topped boxes are to be made out of sheet steel. Each box is to be
made from a 6cm by 4cm rectangular piece of steel. A square will be cut from
each corner, as shown in the diagram, and the remainder made into the box
by bending along the dashed lines and welding.

z fe

——<—<—————
6cm
es idan

6cm
“et >x
eng

A rectangular piece ... with corners ... made up into an

of sheet steel. . . removed... open-topped box

(a) If the squares cut out have side x cm, show that the volume of the box is
Vcm>, where V = x(4 — 2x)(6 — 2x).
(b) What should be the approximate dimensions if the volume of the box is to
be as large as possible?
3 A box with a lid has a square base of side x cm and height / cm. If its total
surface area is 2040 cm* write down a formula for its volume Vcm? and
eliminate 4 to show that V = 510x — 1y?, Hence find the dimensions giving
the maximum possible volume, and calculate this volume.
4 The height, in metres, of a rocket t minutes after blast-off is given by
h =11(36 — 24¢ + 102° — 2°)
Calculate the maximum velocity and maximum acceleration attained.

ma 2.3 Tasksheet E1 — Optimisation problems (page 553)

136

Introductory calculus
4 Numerical integration
2.4.1 Areas under graphs
Suppose that water flows from a tap at a constant rate of 15 litres per minute.
This can be represented graphically.

Rate (litres per minute)

iS

0 20 Time (minutes)

The area under the graph represents the actual volume of water that has passed
in a given time interval. In the first 20 minutes, 15 x 20 = 300 litres flow from
the tap.

In this section we start to develop methods for finding the area under a graph, an
important problem with many applications. We shall consider two of the
possible methods for calculating such areas.

However, calculating an area is only part of the

problem; knowing what quantity that area _litres per minute
represents is also very important. You have
probably met the idea of distance travelled
being represented by the area under a (time,
speed) graph. In the example above the fact that
the area represents a volume of water in litres can
minutes
be deduced by considering the units of the axes of
the graph.

Area is calculated by multiplying a height by a width, so the units of the

area in
this case are
litres
Xx minutes = litres
minutes
Using the units of the axes of a graph in this way is an essential
step in solving
many problems.
 NUMERICAL INTEGRATION 137

@
1 State the dimensions of area for each of these sets of axes.
(a)
| kmh! miles per litre |

litres

.
g cm?

2 A graph shows how the speed of a

turntable, recorded in revolutions per
minute (r.p.m.), varies with time (seconds).
What would one unit of area under such a
graph represent?
seconds

We now consider specific methods for estimating area which will be useful in
later work.

@ +
Suppose the driver of a car leaving the motorway allows the car to decrease
in speed gradually over a 60-second time period. The speed is recorded at
10-second intervals to give the table below.

Time (s) 5 15 25 35 45 55
Speed (ms~') DO SiS, 02170, dS tS0

1 Plot the (time, speed) coordinates from the table on graph paper and draw
the graph for times ranging from 0 to 60 seconds. What does the area
under the graph represent?
You need to use the information in the table to estimate the distance the car
travels during the 60 seconds. Although common sense may tell you that the
138 INTRODUCTORY CALCULUS
Ea
ae

speed of the car is continuously changing with time, you can approximate the
motion in the following way. Suppose the car travels at a constant 29.9 m 5
during the time interval 0 — 10 seconds, then instantly changes speed and
travels at 23.1ms_/ during the time interval 10 — 20 seconds, and so on (call
this the ‘constant speed’ model).

2 Use this model to estimate the distance the car travels during the 60
seconds.

3 Superimpose the ‘constant speed’ model graph on the graph you drew for 1.

4 Shade in the area of the graph which corresponds to your answer to 2, and,
by considering this area, explain why your answer is a good estimate of the
actual distance travelled.

5 Do you think your answer to 2 over-estimates or under-estimates the actual

distance travelled? Explain why.

The method of estimation used above is called the mid-ordinate rule.

The mid-ordinate rule uses a series of Y

rectangles to estimate the area under
a graph. The height of each rectangle
is determined by the height of the
curve at the mid-point of the interval.

The diagram illustrates the mid-

ordinate rule with three strips. It is
customary to keep all strip widths the
same. This makes the calculation of
the area much simpler.

We now consider an alternative numerical method, known as the trapezium

rule.

AC

Suppose readings of the speed of the car considered in 2.48 were taken at
different times, with results as follows.

Time (s) 0 10 20 30 40 50 60
Speed (ms ~') 35.0: 26.0" 2097 21S:0e 16.5 eel aay,

1 Draw the (time, speed) graph for times ranging from 0 to 60 seconds.
 NUMERICAL INTEGRATION 139

Although the graph is obviously curved, you can approximate it with a series
of six straight line segments by joining the known points on the graph. In this
model of the car’s motion you assume that the car’s speed decreases
uniformly during each 10-second interval.

2 Superimpose this model as a graph on the (time, speed) graph you have just
drawn.

3 Use the model to estimate the distance travelled by the car during the 60
seconds.

4 Does the method over-estimate or under-estimate the actual distance

travelled?

The trapezium rule uses a series of

trapezia to estimate the area under a
graph.

Example 1
A research firm has a circuit which is used to evaluate the performance of
engines. A test car is fitted with a computer to record various information, and
in one trial it is driven for 8km around the circuit, gradually increasing in speed.
The fuel consumption in cm®? km‘ is recorded at each kilometre.

Distance driven (km) OY Vil Che Ree rea? Salen tahcs ate yt gtd
Fuel consumption OS ey OF 62s 57s 255-609 5724.87 109
(cm? km7!) :

(a) Sketch the graph, explain briefly the characteristics of the graph and state
what the area under the graph represents.

(b) Calculate the area under the graph using:

(i) the mid-ordinate rule with four strips,
(ii) the trapezium rule with four strips.
140 INTRODUCTORY CALCULUS
Neen eee eee ee eee

Solution
(a) Fuel used
(cm? km7!)

100

SO

o = N ios) ES n 6 if 8
Distance driven (km)

At low speed and at high speed the engine is less economical. As the speed
increases from the start, the fuel consumption, in cm? km‘, decreases until
the most economical speed is reached (after about 4km); then the fuel
consumption increases again.

The area under the graph represents the volume of fuel, in cm’, used during
the 8km circuit.

(b) (i) Mid-ordinate rule

Fuel used
(cm? km!)

100

50

0 1 2 3 4 5 6 iu 8
Distance driven (km)

Avéa = 270! 2 5792 G0 ee,

= 548

= 548 cm?
Fuel used
 NUMERICAL INTEGRATION 141

(ii) Trapezium rule

Fuel used
(cm3 km-!)

100

0 1 2) 3 4 5 6 if 8
Distance driven (km)

Area =1x 2 x (95 + 62) +4.x 2 x (62 +55) +1 x2 x (55472)

+45 x 2x(72+
109)
= 582
Here, the trapezium rule over-estimates the area while the mid-ordinate rule
under-estimates the area. The actual amount of fuel used is somewhere
between the two values calculated.

2.4 Exercise 1

1 Depth readings are taken across a river of width 18 metres. Depths at various
distances from the left bank are shown in the table.

Distance (m) 0 2 4 6 8 10 12 14 1G 1S

Depth (m) 0 O2e 7s Dae m1 OSS po 8ae ee oreo 0

Use the trapezium rule with nine strips to calculate an estimate of the
cross-sectional area.

2 A geologist does a survey of stalactitesand g@¢--—~-~~7~7—

stalagmites in a cave. In order to estimate
their volumes, she measures their
circumferences at different points along
their lengths. 100 cm

Her measurements for one particular

stalagmite are shown in the following
table.
142 INTRODUCTORY CALCULUS

Height (cm) 10 30 50 70 90

Circumference (cm) 50% 140" 530) 20 10

(a) Estimate the cross-sectional area of the stalagmite at each height and draw
a graph which shows how the cross-sectional area changes with the height
of the stalagmite. What assumptions have you made?
(b) Use the mid-ordinate rule to estimate the area under the graph. What does
this area represent?

3 A train is travelling at 20ms_' when the brakes are applied; t seconds later the
speed of the train is given by 20 — 0.2¢7 ms‘. Sketch the (time, speed) graph
and use the trapezium rule, with two-second intervals, to estimate the distance
travelled by the train before it comes to rest.

2.4.2 Integration
You have already seen that to solve some problems it is necessary to find areas
under curves.

The precise value of the area under an (x, y)

graph from x = a to x = b is denoted by
b
| y dx

The notation may look complicated but it is

very useful. It was introduced by Leibniz in
1684 and is one of the reasons why Leibniz’s
version of the calculus was more popular than Newton’s.

The symbol |is an old-fashioned form of the letter ‘s’ and indicates that Leibniz
thought of the area under a curve as being obtained by Jamming areas of lots of
very thin rectangles. y=x+4

The area under the graph of y = x” + 4 is shown in the

diagram. The precise value of this area is written as

j(x* + 4) dx
The process of finding the area under a graph is called
integration because it is, in essence, a process of combining
many small parts to form a whole.
 NUMERICAL INTEGRATION 143

Example 2
Consider the function y = \/(4 — x’).

(a) Draw a diagram to illustrate the area represented by

iJ(4— x2) dx
0

(b) What is the precise value of this integral?

(c) Use the mid-ordinate rule with two strips to estimate this integral.

Solution ;
(a) The shaded area under the graph of
y = /(4— x’) is one quarter of the area of
the circle with equation x* + y” = 4. The
radius is 2.

(b) The area of the circle is 47. So

[V(4 — x*) dx = 7.
0

Wien 10-5 eye 3-7 St 1986 (to 4 s.f.)

Whena-= 1.5, 74/175 21.323 (to 4 s.f.)

2 j
|J (4— x”) dx & (1 x 1.936) + (1 x 1.323) © 3.26 (to 3 s.f.)
0

This method could be used to find an approximate value for 7.

 144 INTRODUCTORY CALCULUS
Sennen
Ee ee eee ee ee

2.4 Exercise 2

1 Draw the graph of y = x and indicate the area represented by the integral
3
| x dx. Find the precise value of this integral.
0
4 3 4
2 Calculate: (a) | x dx (b) | Sidx (c) |(2x
+ 3) dx
1 1 1

3 To estimate, for example, the depth of a well or the height of a cliff, you can
use the fact that the downward speed of a dropped stone increases by
approximately 10ms~’ each second.
From the top of a particular cliff, a stone takes 5 seconds to reach the sea.
(a) Find the speed of the stone t seconds after being dropped.
(b) Express the height of the cliff as an integral.
(c) Find the height of the cliff.

3
The illustrated area is | /(10 — x”) dx.
1

Copy the diagram and draw in the two strips you use to estimate the area.

Calculate this area approximately using the mid-ordinate rule with two strips.

2.4.3 Numerical methods

In the previous section we obtained an approximation for 7 using the
mid-ordinate rule with just two strips. You can improve this estimate by using
more (thinner) strips.

To get a very accurate estimate of the integral you might have to use a very large
number of strips.
 NUMERICAL INTEGRATION 145

The diagram shows

the area represented
2
by |J/(4 — x”) dx
0
divided into ten strips.
(Only a few of the
strips are shown.)

The width of each strip = ) = ——————- = 0.2

no. of strips
Strip @ has a height of \/(4 — 0.17) = 1.9975 and an area of
0.2 x 1.9975 = 0.3995 (to 4 decimal places).
To obtain an accurate estimate of an area you need to perform a large number
of calcuiations. It is therefore very convenient to use either a spreadsheet or a
programmable calculator to perform calculations of this kind and to explore,
for example, the effect of using more and more strips.
The following are the results of using a spreadsheet to obtain an estimate for the
area with 10 strips.
Number of
strips = 10

Width of strip Mid y Areaof Cumulative

value (0) strip area
0.2 0.1 1.99750 0.3995 0.3995
0.2 0.3 1.9773? 0.3955 0.7950
0.2 0.5 1.93649 0.3873 1.1823
0.2 0.7? 1.87350 0.374? 1.5570
0.2 0.9 1.78606 0.3572 1.9142
0.2 1.1 1.67033 0.3341 2.2483
0.2 1.3 1.51987 0.3040 2.5523
0.2 1.5 1.32288 0.2646 2.8169
0.2 1.2 1.0535? 0.2107 3.0276
0.2 1.9 0.62450 0.1249 3.1525

Estimate of
area = 3.1525
 146 INTRODUCTORY CALCULUS

4D

| The mid-ordinate rule

Consider the area represented by the integral above to be split up into 1 strips
of equal width, 4. Using the mid-ordinate rule to approximate the integral,
the area of the first strip would be calculated as hy,, the second strip would
be hy, and so on.

1 (a) Express / in terms of a, b and n.

(b) Express x, in terms of a and h.

(c) By how much do you increase x each time you move up a strip?

The trapezium rule

y = f(x)

Xo xy X2 Xy-1 Xp

Again, consider the area represented by the integral above to be split up into
n strips of equal width, h. Using the trapezium rule, the area of the first strip
would be 5h(yo + y,), the area of the second strip would be Sh(y, + yz), and
so on.

‘ 2 Derive a formula for the total area of the n strips shown.

 NUMERICAL INTEGRATION 147 f

The trapezium and the mid-ordinate rules with n strips both involve the sum of
n areas.

The mid-ordinate rule

b n

|F(x) dx ~ S~ hy, =)
a r=1

where
b-a
n
x4 =a+$h

Xy41 = Xy +b

ieee f(x,)

The trapezium rule

% n
|f(x) dx ~ $7 4h(y,-1 + y,)
fS4 y

er where

——E———
eS

The two rules can be summarised as follows.

148 INTRODUCTORY CALCULUS

As you take more and more strips, the effect near the mid-point of any strip is
similar to the effect you observe when a computer or calculator is used to zoom
in on a small part of a graph.
y

— 2 x

The area estimate given by either the trapezium or the mid-ordinate rule will
become closer and closer to the true value, at least for all locally straight graphs.

For the graph of y = \/(4 — x”) you may have noticed that the trapezium rule
consistently under-estimates the true area whereas the mid-ordinate rule always
Over-est mates it.

The trapezium rule approximates a graph with a series of chords and it is easy to
tell whether it will over- or under-estimate the area.

Over-estimate Under-estimate

To predict if the mid-ordinate rule will over- or under-estimate an area, it is

helpful to think of it as producing trapezia rather than rectangles!

A = mid-ordinate
estimate

The area of trapezium B is precisely the same as the area of rectangle A. (Can
you see that this is so?)
 NUMERICAL INTEGRATION 149

Thus, you can think of the mid-ordinate rule

as approximating a graph with a series of
tangents to the graph.

2.4 Exercise 3

1 State, where it is clear, whether a mid-ordinate rule estimate of each of the

following areas will be too large or too small.
(a) (b) y

2 Repeat question 1 for the trapezium rule.

3 Among other things, a firm manufactures two types of metal alloy casting
which will be machined into components for use in the car industry. The
castings are in the form of prisms whose cross-sectional areas are shown
below. (Measurements are in centimetres.)
Type B
150 INTRODUCTORY CALCULUS

The firm received an order for 12000 castings of type A and 8000 of type B.
The production manager needs to calculate how much alloy to prepare.
He does this by calculating the combined volume of all the castings and then
adding on 5% for wastage. Calculate this volume and give your answer in
cubic metres.

1 1 ;
4 Evaluate |irae using the mid-ordinate rule with five strips.
0 be

2.4.4 ‘Negative’ areas

When calculating areas, note that the mid-ordinate rule assigns positive values
to areas above the x-axis and negative values to areas below the x-axis. This is
because it uses sums of terms, each of the form yh.

yh is positive yh is negative
if y is positive if y is negative h
h xX

For example, consider

b
|(3x7 — 18x + 24) dx
a

for intervals given by

C= ib 2) ad = 2b a
The graph of the function 3x” — 18x + 24 is shown below.

y = 3x? — 18x + 24
 NUMERICAL INTEGRATION 151

Using a numerical method (either the mid-ordinate or trapezium rule) we obtain

the following.

iD Between x = 1 and x = 2 the graph is

|(3x* — 18x
+ 24) dx = 4 above the x-axis and the integral is
1
positive.
4 Between x = 2 and x = 4 the graph is
|(3x? — 18x + 24) dx = —4 below the x-axis and the integral is
2
negative.
4 The area below the x-axis is equal to
|(3x? — 18x
+ 24) dx =0 the area above.
1

Note that area is always positive. The areas A and B indicated are both positive.

2.4 Exercise 4

1 In a balanced aquarium, the rate of change of the amount of carbon dioxide

dissolved in a litre of water is found to have the following daily pattern.

CO,
(mg
hour)
per

(a) During what part of the day is the amount of carbon dioxide dissolved in
the aquarium water increasing? Estimate the total increase during this part
of the day.
(b) Estimate the increase in the amount of carbon dioxide dissolved during a
full 24-hour cycle.
152 INTRODUCTORY CALCULUS
ee

2 (a) Sketch the graph of y = x7 — x —2.

(b) Estimate:
2 4.
(i) [=4 Sage DNGn [vex
-4
Gi) |1 y dx (iv) |y dx
2
(c) Estimate the area enclosed between the graph and the x-axis.

3 Water flows both into and out of a tank. The 15 litres

flow out is at a constant rate of 3 litres per Zaaee | |
per minute
minute, whereas the flow in starts off at 15 litres |
per minute but then decreases with time. The
supervising engineer decides that the flow in at
time t minutes may be modelled by
3 litres per
minute
litres per minute
el ’
The tank initially contains 50 litres of water.
(a) The net flow into the tank is given by the expression
Se
P+
Use a graph plotter to show the graph of this (time, flow) function. Why is
t = 2 an important point on the graph?
(b) What does area under the (time, flow) graph represent and what is the
significant difference between the areas above and below the axes?
(c) Estimate

| Role
over the time intervals:
@) P=0
tor = 2) 9) Gi) P= Wter=o) Ee 0S
Explain the meaning of your answer in each case.
(d) How much water is there in the tank after 2 minutes and after 5 minutes?
About how long will it take to empty the tank completely?

4 An alloy casting is in the form of a prism with cross-sectional area as shown.

5
Calculate |x dx precisely and use
0
the mid-ordinate rule to estimate
4 5
|(x? —4x)dx and |(x? — 4x) dx.
0 4

Explain how you can use these integrals to

evaluate the shaded area.
 NUMERICAL INTEGRATION 153

An alternative ‘profile’ for the prism in question 4 is obtained by translating

the graphs up 5 units, so that they become the graphs of aaa
y=xt5 and y=x*-—4x4+5
(a) Sketch the graphs of these two functions on the same axes and shade in
the area representing the cross-section of the alloy casting.
5 5
(b) Evaluate | (x + 5) dx precisely, estimate | (x* — 4x +5) dx and use your
0 0
answers to calculate the shaded area, confirming your answer to question 4.
154

Introductory calculus
5 Algebraic integration
2.5.1 The integral function
In section 2.4, you used the mid-ordinate rule with two strips to obtain the
approximation to the area under the graph of y = \/(4 — x’) given below.

jJ/(4— x’) dx & 3.26

This is a value for the area of a quarter of a circle of radius 2 units, which you
could use to estimate the area of the whole circle. While this is one way of
finding the area of the circle, it is unlikely to be the way you would choose
because you already know the formula
2
A=mTr
for the area of a circle of radius 7 units.

The formula A = rr’ for the area of a circle was known in classical times. In the
seventeenth century, the calculus developed by Leibniz and Newton enabled
formulas to be obtained for various areas. Today, calculators and computers can
be used to work out such areas by numerical methods but there are still
advantages in knowing simple formulas. This section will look at formulas for
the areas under the graphs of a few simple functions, starting with f(x) = x.

Example 1

(a) Explain why |ede Ly

0
(b) Use this formula to evaluate the integral
7
|xdx
3
4 ALGEBRAIC INTEGRATION 155

Solution

(a) y x dx = area of shaded panels

5 < base height

=lxuxu
| Uu
2
RIP
oN

if 7 3
S a.Rg I xdx—| xdx
3 0 0
rae 2 2
=3x7—3%«3
24.5
— 4.5

20

ae
———

} The following examples investigate the integral function for f(x) = x*.

| SA
; u f(x)
( A(u) = x” dx is the area shaded.
0

5 EWN
156 INTRODUCTORY CALCULUS

A(u) OU Lose On, iaianes 1s marie

Using the mid-ordinate rule, calculate an estimate of the value of A(z)

missing from the table above.

2 Use the completed table of question 1 to estimate:

4 3 y
(a) [seas (b) [sax
2 1

0 2 Belk
3 (a) Use the completed table from question 1 to draw the graph of A(u)
against u.

(b) Use suitable values from your graph to estimate:

Sad 3.8
(i) | x*dx (ii) | x* dx
2.5 0.7

(c) Suggest a formula for A(z) in terms of u.

You have some numerical evidence for the result that

f(x) = x? > A(x) = 1x?

This result, although true, has not of course been proven here, only suggested.
You can now use these area functions, called integral functions, to evaluate
integrals precisely and easily.

Example 2
Sketch the graph of y = x’. Calculate the area given by
4
| x” dx
29)

Solution
For f(x) = x”, A(x) = :
 ALGEBRAIC INTEGRATION 157

f(x)
16 So
4
| x’ dx = A(4) — A(2.5)
12 2.5
Csis «4 gees3 NOUS 3

6 4/208
4

0 1 a7 23 4 x

A special notation is used when writing out the evaluation of integrals. For
example,
4 4
| x’ dx = 1x?
) 2.5
1
where the new notation on the right-hand side shows the integral function, 1°,
and also shows the limits, 2.5 and 4. The full solution to example 2 would
therefore be written
4 4
| x dx Fa =1x
4 -1x 2.59 = 16.125
Tes) 25)

Example 3
An object starts from rest and its speed v ms_' at time t seconds is given by
v = t’. Calculate the distance travelled in the third second of its motion.

Solution
The (time, speed) graph shows that the” Bee
distance travelled in the third second, 6 :
as represented by the shaded area, will
be given by RZ
3 3
|? dt = ye 8
o 2
= Ps 4
2) 8}
_ 193
0 1 2 3 4 ¢
The distance travelled is 64m.

@ x
1 (a) Sketch the graph of y = x.

(b) Use the symmetry of the quadratic graph in part (a) to explain why
4 =)
|x dx = | x? dx
2 4
158 INTRODUCTORY CALCULUS
——

(c) Use the formula for the integral function to confirm that
4 2
|x dx = | x? dx
3) af

2 (a) Using the integral function, evaluate:

=505 1.5 se 5 at
(i) | x’ dx (ii) | x’ dx (iii) | x dx> (IV) | x dx
=; =15 1.5 a5
(b) Write down any relationships which connect two or more of these
integrals. Explain, with the aid of a sketch graph, why these
relationships are true.

Calculate the shaded area in terms of u.

(b) What is the formula for the integral function?

4 For the function f(v) = 3v:

(a) sketch and shade the area between the graph and the horizontal axis,
between the limits v = 0 and v = u;

(b) calculate the shaded area in terms of u.

5 Repeat question 4 using the following functions.

(i) f(x) =2 (ii) g(t) = —3
(itt) (Eee (iv) g(v) = —3v

6 Write down the integral function for:

(a) Oe) (b) g(x) = mx
when m is any constant and areas are measured from x = 0.
 ALGEBRAIC INTEGRATION 159

2.5.2 Integrals of polynomials

So far, you have met the following integral functions:

(a) If f(x) = m, then A(x) = mx

(b) If f(x) = mx, then A(x) = Linx?
(c) If f(x) = x”, then A(x) = 1x?

This might suggest that if, for example, f(x) = 2x”, then the integral function
would be A(x) = 2x?

What about A(x) if f(x) = x°? The results for f(x) = x and f(x) = x” might
suggest that, if f(x) = x°, then A(x) = 1x

These possibilities are explored in the questions which follow.

Be Be

1 (a) Sketch the graphs of y = x* and y = 2x’.

(b) Shade the areas represented by the integrals:

Ix*dx and j2x” dx

0 0
(c) What simple geometrical transformation connects the two regions?
: b b
(d) What will be the connection between |kx? dx and |x” dx?
a a

2 (a) Explain, with the aid of a copy of

y=2x+3
the graph of y = 2x + 3, and
considering the areas A; and A),
why
u

[2x Be 3) doe [ ~ an eee a eae

0 3

(b) Sketch the graph of y = x* + 4 and

use it to explain, by considering u x
suitable areas, why
u u

|(x? +4) dx = lax?+ tx

0 0
160 INTRODUCTORY CALCULUS

3 Choose two functions of the form ax” + bx +c, where a, b and c are
constants. (Do not always choose positive values for a, b and c.) Write
down probable integral functions and check these either with a graph
plotter, or by numerical integration with suitable limits.

4 Suggest a general formula for the integral function of any quadratic.

5 Repeat question 3 for a function of the form ax* + bx? + cx + d.

6 Suggest a general formula for the integral function of any cubic.

You now have the following rule for integrating:

Example 4
3
(a) Evaluate the integral |(x — 4)(x — 2)(x + 1) dx.
0
(b) Use a suitable sketch to interpret your answer.

Solution
3 3
(a) |j(@—A) -2)x-4 de = | a? — 5x4 +28 + 8) ae
0 0

wihES Gara Wael

ey es ae eet,
eS
(b) The integral is the sum of
a positive part from x = 0
to x = 2 and a negative y=(*+
1) (%—- 2) (x- 4)

pant from x = 2 to —3:

 ALGEBRAIC INTEGRATION 161

2.8 Exercise 1

1 (a) Write down the integral which

represents the shaded area.
(b) Calculate this area.

1
2 (a) Evaluate | (e42¢ —3)de.
2
(b) Using a suitable sketch, explain why your answer is negative.

3 (a) Sketch the graph of y = x° — 2x” — 5x + 6, showing clearly where the

curve cuts the x-axis.
(b) Calculate the total area enclosed by the curve and the x-axis.

4 A stone is projected vertically upwards such that its speed (vms_‘) after t
seconds is given by
v = 5(5 — 28)
(a) How far does the stone travel in the first two seconds?
(b) After how many seconds does it reach its maximum height?
(c) Calculate the maximum height the stone will reach.

5 Evaluate:
i
& |e =») dx (b) [ (et 1) de (c) [ee
0

1
(d) [ot -284 dx (e) [(? -284 Dax (f) [+ +2) dx
0

6 Sketch the graph of y = 3x — x’, and then find the total area between the
graph and the x-axis over the interval [0, 5].
a

7 Find the values of a for which |(S= 2x) dx= 0.

1
8 The velocity, in m s_', of a particle is given by v = 4t — 2’. Find the total
distance travelled in the first 6 seconds.
162 INTRODUCTORY CALCULUS

2.5.3 Numerical or algebraic integration?

Many problems of integration can be solved easily using an appropriate
algorithm on a hand-held calculator. So why have a formula for integral
functions?

If you just want a numerical answer then numerical integration is fine. But in this
section you will see that if you want to understand what is going on in a
problem, then having a formula can be very helpful.

Example 5
A fruit farmer estimates that the average apple tree yields 45 kg of fruit and that
it takes about 3 hours to pick all the apples. The fruit pickers are paid at a rate
of £6 per hour and the farmer sells the apples for 60p per kg. Productivity
decreases with time; in other words a person can pick more apples during the
first half hour than during the last as the apples become harder to reach. The
farmer assumes productivity decreases linearly with time.

(a) What weight of apples can be picked from the tree in t hours?
(b) What is the maximum profit that the farmer can make?

Solution
k
(a) p(kgh"') Area represents = x h= ke:
30
As 45 kg can be picked in 3 hours,
the area shown is 45 and the
height of the triangle must be 30.
3 t(h)

Hence the formula for productivity is:

p = 30 — 10¢
Then the weight of apples picked in
t hours is 30
t t

|(30 — 102) = [sor= se

0 0

= 30¢ — St? t
(b) If all the apples on a tree are picked, the farmer makes a profit of £9
(45 kg at 60p less 3 hours at £6). If, however, the person picking the
apples stops after t hours, then the profit, £P, is obtained by calculating
(30¢ — St”) kg at £0.60 less t hours at £6.
P = 0.6(30t — 5#*) — 6 = 12-32 (0 <t <3)
 ALGEBRAIC INTEGRATION 163

You can see that the farmer can increase the profit to ——p
£12 per tree by instructing the workers to spend only ‘12 ae
2 hours on each tree and leave the last few apples for 9 ’ .
the birds.
6

0 din a

The following extended example provides further illustration of the usefulness

of algebraic integration, this time in the industrial process of ‘catalyst
renewal’.

5D

You may know that a catalyst is often necessary to promote a chemical

reaction. As the reaction goes on, the efficiency of the catalyst diminishes,
until eventually it has to be renewed.

Suppose a chemical plant can produce a desired chemical at a rate of 100kg

per hour when the catalyst is fresh but that this productivity gradually
decreases in such a way that after t hours the productivity has fallen to a
value p kg per hour, where pike He)
pe ates) 100
7s
After 5 hours the catalyst has no
effect whatsoever, so the graph 50

does not apply beyond this point. 25

Oc 0 Sega eS ae O7LE (fi)

Maximising production

Clearly, what is important is the total amount of chemical produced in a

given period of time and not the rate at which it is being produced (the
productivity) at any particular time.

Suppose the catalyst is renewed after 3 hours. The production manager would
want to know how much chemical is produced in this time. You can see that
as one axis of the graph represents time in hours and the other represents
productivity in kg hour’, the area under the graph will represent amount in
kilograms.
kg
ate x hour = kg

The total production of chemical is therefore given by the area under the
graph.
164 INTRODUCTORY CALCULUS
a SS SS EE

The amount of chemical produced p (kg ho!)

during the first 3 hours is ee
75
3
|4(t — 5)? de 50
e 25
=4|hes — 10¢ + 25) dt 0 ee ae ae se)

1 Integrate algebraically to find the amount of chemical produced in the first

3 hours.

When the catalyst is renewed the whole plant is shut down. Suppose it takes
30 minutes to change the catalyst. In this case, the plant works on a 3.5-hour
production cycle which can be shown graphically.

P (kg h-')

OS yo 3. 41 S58 6 7 8 OF 10 i 12 a aS ra)

2 (a) What is the average output per hour (average productivity) for a
complete production cycle?

(b) Suppose the catalyst is changed after 1.5 hours (i.e. the plant works on
a 2-hour production cycle as it still takes 30 minutes to change the
catalyst). Use the integral function you obtained in question 1 to find
the average productivity.

(c) Suppose the catalyst is changed after 40 minutes. What would the
average productivity be now?

3 The production manager wants to maximise production (i.e. run the plant
so that as much chemical as possible is produced). Investigate different time
intervals and determine the frequency with which the catalyst should be
changed. Give your answer to the nearest minute.

Maximising profit

Suppose that the chemical is sold at a fixed price of £3 per kg and that there
is a cost of £150 each time the catalyst is renewed, with other production
costs more or less fixed at £50 per hour.
 ALGEBRAIC INTEGRATION 165

The company is not sure that maximising production is necessarily the same
as maximising profit.

4E (a) If the catalyst is renewed every 3 hours, what profit does the company
make on each kilogram of chemical sold?
(b) How much profit does the company make per hour in this case?

SE Investigate the profit per kilogram and profit per hour for other time
intervals between catalyst renewal. To maximise annual profit, should
the company maximise production, profit per kilogram, or profit per
hour?

2.5.4 The fundamental theorem of calculus

You have been able to calculate some integrals precisely by first finding an
integral function. For example, x° + x is an integral function for 3x* + 1 and so
5
[oe + 1)dx= [.- 7
3

There is a simple relationship between a function and its integral function. The
table below may help you to spot the connection.

f(x) 1 x 3x" Bx EA
A(x) Ds

It is clear that if you differentiate the integral function, you obtain the function
you are integrating, that is,

=d (AC) = fle)
The process of differentiation (e.g. finding gradients) is integrate
an inverse process to that of integration (e.g. finding pe EE Ce
areas).) a 3x2 x3

ate avin See

differentiate
166 INTRODUCTORY CALCULUS

This remarkable relationship is called the fundamental theorem of calculus.

The graphs of all differentiable functions are locally straight, so it is sensible to

investigate the connection between differentiation and integration by first
considering linear functions.

The gradient of a line segment joining two foints is constant.

For example:

ae g'(x)

0 0 1 4 x
End points (1, 3) and (4, 9) Area 6

JE

1 Draw diagrams similar to those above for line segments joining the points:

(a) (1,2),(5,6) — (b) (4,8), (6,2) (ce) (3,5), (6,5)

Can you spot a connection between the y-coordinates of the end points of
the function and the area under the graph of the derived function?

Explain this connection by considering the definition of the gradient of a

straight line.
 ALGEBRAIC INTEGRATION 167

When the graph of g(x) is a series of connected line segments, the diagrams
obtained are like those below.

&'(x)

as A nT Wie one ae US Maes ies Nall? WE. Wiad Seas c

7
End points g(1) = 3 and g(7) = 10 Area |g (x)dx =44+2+4+1=7

2 The following diagrams show two further ways of joining the end points
(1,3) and (7, 10) with three line segments.
g(x) WS 10)

(7, 10) 10 (7, 10)

a Ae ee oh ey eS Ole
Oh s3crdonS Oman Ore:

Sketch the graph of the function g’(x) for each of the examples shown
above, or any two similar examples of your own invention.
7
In each case, find |g (x) dx. Is it always true that
“7 1
|@) dx = 27) - 2(0)?
b
3 Is it always true that |g (x) dx = g(b) — g(a)? Test this conjecture with
a

any similar type of function of your own choice consisting of several line
segments.
168 INTRODUCTORY CALCULUS

4E (a) The graph shows a derived function g’(x)

g(x). Construct a possible graph of 3
the original function g(x). SG
(b) Explain why your answer to (a) is 2 Y
not unique. Construct another
possible graph of g(x).
1 GY

(c) In each case, check that

[8) dx = 8(6)~ 2(0).

° I
RENEE ae ee

|e@)4x = 2) - a
Any differentiable function g is locally straight and so can be approximated by a
series of straight-line segments.

The gradient function g’ can then be approximated by a step-graph.

 ALGEBRAIC INTEGRATION 169

This does not prove the fundamental theorem of calculus but at least indicates

The fact that integration and differentiation are inverse operations should not
come as a surprise when you think about the way in which (time, speed) and
(time, distance) graphs are related. Given a (time, distance) function you
differentiate to find speed, and given a (time, speed) function you integrate to
find distance.

The fundamental theorem gives a method of integrating which is easy to use.

You must find a function which, when differentiated, gives the function you wish
to integrate!

2.5 Exercise 2

1 (a) Evaluate:

(i) [ (x? — 3x” — x43) dx (ii) i (x? — 3x” — x +3) dx

=H) =i

(b) Sketch the graph of y = x° — 3x” — x + 3. Hence explain the connection

3
between your answers to (a) and the value of |(x? — 3x” — x +3) dx.
1
p)
2 Find |(3x7 — 10x) dx.
1
:
d
3 (a) i |g(e)ida = [ss?
Ae 3a , write down g(x).
c c

b b
(b) i |h(t) dt = [2- 7 , write down h(t).
a a
170 INTRODUCTORY CALCULUS

4 Some areas which cannot be calculated directly can be obtained by considering

a combination of integrals.

(a) (i) Explain why the shaded area is given by

a b
| dx + |(8 — x”) dx
0 a

where a and Bb are to be found.

(ii) Find the shaded area A.

y=8-x7 x

(b) Find the shaded areas illustrated in (i), (ii) and (iii) below.

(i) (ii) (iii)

ee

W/
5 Find c if the shaded area is 6 square units.
vax

6E Find a such that

|(3x* — 2x) dx =0
0
 ALGEBRAIC INTEGRATION 171

2.5.5 The indefinite integral

d(x? + 5x)
Since = 2x +5 you know that
x

fests dx = [P+5x]at
1 1

SF

Differentiate each of:

xe +Setd1, x t5e4+4, x7?4+5x-3.
Explain why each of these functions could be used as an integral function for
2x + 5 and why each gives the same answer for

[ex eae
J1

5
To evaluate an integral such as |(2x) dx you can use any integral function of
3
the form x* + c where c is a constant, known as the constant of integration.

A general integral function such as x” + c is called an indefinite integral.

An integral sign without any limits is used to denote indefinite integrals.

A constant term ‘+c’ should always be included, for example

ie dy =x, |

|?dx'= 1x? +c

Integrals between limits, for example

5
|(2x)dx = 16
3
are called definite integrals. A definite integral has a (definite) numerical value.

For definite integrals, there is no need to include the constant of integration

because it cancels out as shown in the following example.
 172 INTRODUCTORY CALCULUS
i

Example 6
Find y as a function of x given that y = 10 when x = 1 and that

d
es (3x — 1)(x +3)
dx

Solution

y= |Gx=1)( +3) dx

=| (30? + 8x — 3) dy

=x +4x* —3x+e¢

But y = 10 when x = 1

= 10=1+4+4—3
+6

=> C=

=> y=x+
4x? —3x+8

2.5 Exercise 3

1 Find the following:

3
(a) [ee = nas (b) | e+ 0-2) ax
il

2 Find y as a function of x, if:

dy | dy _ 2
(a) AS ie 4 (b) ame AP ae

dy dy
(c) apa bet (d) = (*+1)(x-2)

3 Express y as a function of x if:

d
(a) = = 3x* 4+ 4y and the (x,y) graph passes through (1, 5).

d
(b) = =x? 4++%+41 and the (x,y) graph passes through (0, 3).

b b
4 (a) | (3x* — 2x + 5) dx = A] , write down a possible f(x).
a

d d
(b) | (2¢+ 1)(t —4) dt = kc] , find a suitable k(t).
c c
 ALGEBRAIC INTEGRATION 173

5 The population of a certain country increases by approximately 0.2 million per

year. Find an expression for the population P million in year t, remembering
to include the constant of integration.
Given that the population was roughly 38 million in 1900, estimate the
population in 2000.

1
6 A ball is released from rest on a ramp. Its speed t seconds later is 4tms_
(a) Show that s = 2¢? +c, where c is the constant of integration. What
information is provided by the value of c in this case?
(b) Write s in terms of t if the ball is released 0.5 m from the top of the ramp.
Find the length of the ramp if the ball then takes 1 second to reach the
bottom.
(c) Write s in terms of t if the ball is released 1m from the top of the ramp.
How long will it then take to reach the bottom of the ramp?

Tasksheet E1 — Traffic (page 555)

174

Introductory calculus
Miscellaneous exercise 2

1 Find the derivative of each of the following:

(a) ya 7 (b) y = 2x* + 3x G4 ts

(d) y=6x-11 () y=Sx?—x-8 = (f) y=x—

2 Find the gradients of each of the following curves at the given points:
(a) y=2x?—xat(1,1) (b) y=4x*+3 at (-2,11)
(c) y=3—x—x? at (2,-7)

3 Find the equation of the tangent to the curve at the point given. Then sketch
both curve and tangent on the same diagram.

(a) y=x?+4at(-1,5) (b) y= 4x" at (1,4)

(c) y = 3x° — 1 at (0, —1)

4 Sketch the curve y = 2x* + 3. Calculate the points where the gradient is:

(a) 2 (b) 6 (c) 0

5 Find the point on y = x” — 2x + 8 where the tangent is parallel to the straight

line y = 4x +1.

6 Show that at (0,8) the gradient of the curve y = x° — 3x* + 8 is zero. Find
another point on the curve where the gradient is also zero.

7 Find the equations of the tangents to y = x” at the points (2,4) and (—1, 1).
Verify that the tangents intersect at (5, —2), and illustrate with a sketch.

8 Find the equations of the tangents to y = x* — 1 at (1,0) and (2,7). By

solving simultaneous equations, find where the tangents intersect.
 MISCELLANEOUS EXERCISE 175

2 A particle moves along a straight line so that after t seconds its position in
metres relative to a fixed point is given by
x =?f.—3¢ +10
(a) What is the formula for the velocity?

(b) When is the particle stationary?

(c) In which direction is the particle moving at time t = 1? For how long
overall is it moving in this direction?

10 A ball is thrown vertically upwards so that its height after t seconds is

(30t — St”) metres.
(a) What is its initial velocity?

(b) When is its velocity zero?

(c) How high does the ball go?

11 A stone is projected vertically upwards from the surface of the Moon, and its
height after t seconds is (20t — 0.8t7) m.
(a) What is its initial speed?
(b) To what height does the stone rise?

(c) After how long will the stone be descending at Sms '?

12 The curve y= x+ ax? for 0 < x < 10 is used to model the vertical
cross-section of a sea wall, the units being metres.

(a) Draw a sketch of the wall, and find its height.

(b) Find the gradient of the wall at the base and at the top.

13 Find the stationary points of each of the following:

(a) (x) =x’ ~4x43 (b) h(x) =44+x%—x7

(c) g=x' —32x4+ 50 (dd) g=8t+5¢—-—f for t>0

14 A faulty firework moves so that its displacement s metres from a fixed point
is given by s = 1? = 37 + 1 where t is the time in seconds and 0 < t < 4.
Find an expression for the velocity and hence find the two values of ¢ at
which the firework is stationary. Make a rough sketch of the graph of s
against t, marking in clearly the stationary points.

15 Find all stationary points of the following. Hence sketch their graphs.

(a) y= 7 — 8x 4+ 2x" (b) y = x* — 2x? (c) y=tx? —x° -—3x41

176 INTRODUCTORY CALCULUS
EE
SS Se

16 A flower bed is to be L-shaped as shown, and its perimeter is 36m. Write

down a formula for its area Am? and eliminate x to show that

A = 36y — 6y’.
Hence find the values of x and y giving maximum area.
ym

xm

xm

17 A body moves in a straight line such that its displacement in metres from a
fixed point O is related to the time t seconds by s = 127 —F for 0 St 10:
Find
(a) when the body is stationary,

(b) when its velocity is 36 ms‘,

(c) the maximum velocity.

18 Find the areas under the graphs of the following

(i) using the trapezium rule with six trapezia,
(ii) by algebraic integration.

(a) y=30+2x—x* fromx=1tox=4

(b) y=1+10x—x° fromx=O0tox=3

19 Find the areas of the regions bounded by the x-axis and the graphs of the
following:

(a) y=x'—3x (b) y= (x4+1)(6-x) — () y= (6-2)?

20 Find the areas bounded by parts of the graphs of y = 3x+5 and y =x? +1.

21 The following definite integrals clearly have a meaning, yet their values
cannot be found by the algebraic methods you have met so far. Find their
approximate values, to 3s.f.:
DP 1 1 1 0
(a) | — dx (b) | dx (c) | 2” dx
1x 41+? wes

5 4
22 Determine the values of | x dx and |(x? + 2x + 1) dx, and explain why
2 1
your answers are related.
 MISCELLANEOUS EXERCISE 177

Pe, Find these indefinite integrals:

(a) Jo + 6x”) dx (b) |x(x” + 1) dx (c) | x + bx + ¢) dx

24 The velocity (in ms‘) of a particle moving along the x-axis is given by
v = St — 3t’. Find an equation for the displacement x in terms of t, given that
x = 2 when t = 0. When is x greatest? When is v greatest?

25 The rate at which water was flowing in a stream was measured at hourly
intervals with the following results:

Time 12a.m. 1p.m. 2p.m. 3 p.m. 4p.m.

Rate (m?/min) 8 10 11 9.5 9

(a) Plot these points on a graph. Join the points with straight lines. What
does the area under the graph represent?
(b) Find a trapezium rule approximation for the volume of water flowing in
the stream between 12 noon and 4p.m.

26 Sketch graphs to illustrate the following integrals, and find their values:
3 4 2
(a) | x dx —(b) | x* dx (c) | x3 dx
i 1 il

2a Sketch graphs to illustrate the following definite integrals, and find

expressions for their values:
2t t+5
(a) | x'dx (t>0) wb | xdx (¢ >—3)
t t+3

28 Find b in the following:

b b 3b
(a) | xedx=60 (b) | x’ dx = 63 (c) | xdx= 05, (6 0)
2 3 b
178

Functions

3.1.1 Composition of functions

Temperatures are often measured in degrees celsius or degrees fahrenheit. On the
fahrenheit scale, water freezes at 32°F and boils at 212 °F. The celsius scale is
such that water freezes at 0°C and boils at 100°C.
The function f, given by
f(t) = 3(¢— 32)
converts a temperature of t degrees fahrenheit to celsius.
Temperature is a measure of the vibration of molecules and at —273 °C
molecules are no longer vibrating, so —273 °C is the lowest temperature that can
be obtained. This temperature is called 0 on the kelvin scale, or OK.
In order to convert from celsius to kelvin the function g is used.
g(t) =t+273
To convert a temperature measured on the fahrenheit scale to one on the kelvin
scale would require conversion to the celsius scale first — using f(t) — then to the
kelvin scale — using g(t). It would be helpful to have a function which does this
conversion directly.
The rule for converting from °F directly into K can be illustrated by using an
arrow graph or a flow diagram.

oF ee °C Pe K

22 —>— 100 a 373

122 —> 50 a 323

32 —_>— 0 —>- 273

t —»>— ._ {(t-32) —»—~_ R(t — 32) + 273

3(t
— 32) +273
 ALGEBRA OF FUNCTIONS 179

The resulting function is the composition of the two functions f and g.

Note that, since f(122)=50, you can write g(50) as g(f(122)) or, with fewer
brackets, as gf(122). So, contrary to what might be expected, the notation for f
followed by g is ef.

For fg(x), x is first put through the function g, and the output from g is fed
through f.

% fg(x)

Example 1 |
If f and g are the functions given by f(x) = x* and g(x) = 2x + 3, then find the
functions fg, gf and gg.

Solution
fe(x)\=f(e(x)) = f(2x + 3) =(2x +3)?

ef(x) = g(f(x)) = g(x") = 2x* +3

gg(x) = g(g(x)) = g(2x + 3) = 2(2x +3) +3

=4x+9
Tasksheet S1 gives further practice in working with functions, should you need it.

3.1 Tasksheet $1 — Functions of functions (page 535)

3.1 Exercise 1

1 For each of the functions f and g defined below, evaluate (i) fg(x), (ii) gf(x).

(a) f(x) =2x +3, g(x) =x

(b) f(x) =2x+1, g(x) = -
(c) (x) =3x+2, g(x)=5-—x

(d) f(x) =1-x*, g(x) =1-2x

180 FUNCTIONS

On a gas bill, the cost of x therms of gas used by a consumer is given by £c,
where
c(x) =9+0.4x
A gas meter indicates the amount of gas in cubic feet used by a consumer.
The number of therms of heat from x cubic feet of gas is given by the
function t:

t(x) = 1.034x

(a) Find the function ct.

(b) What does ct(x) represent?

es) Each of the following is of the form fg(x). Identify f(x) and g(x).

) : Gea (c) : (d) 2/x—1

is xD Be " 2x +3

(e) at3 (f) (Q2x+1)* (g) x® —4x*-3

ff(x) means the function f applied twice to x; this is often written as f*(x).
If each of the expressions below is f(x), write down an expression for f(x) in
each case.
(a) x +2 (b) x? (c) 2x —3 (d) x (e) sinx (f)
le
R

n If f(x) = x — 3 and g(x) = x’, it is possible to combine these functions in

many ways.
(a) Explain why (x — 3)’ — 3 = fgf(x).
(b) Express each of the following as combinations of f and g.
(i) x =3 Gi) @—3)) Gi)ie—6 “G)e 8. (wn 3) Se

lon Using s(x) = sinx and q(x) = x’, distinguish clearly between:
(a) sin? x, ice. (sin x) (b) sinx’, i.e. sin (x?) (c) sinsinx, i.e. sin (sin x)

N In each case, find fg(x) and gf(x), and then determine the set of values for
which fg(x) = gf(x).
(a) f(x) =x’, g(x) =x+4+3 (b) "f(y =x = 5, g(x) = x2

(c) f(x) =2x-1, g(x)=3x+1 (d) f(x) =— ee)

(e) (x) =2e+1, e(e)=4@e4+1) f) fe) =x, ex) =x-1
 ALGEBRA OF FUNCTIONS 181

8 Four functions, e, f, g and h, are defined by

e(x)=x f(x)=—x g(x) == h(x) =——

opener K
ee eee x

Complete the following table, where each entry is one of e, f, g or h.

First function
feeae oh
* The function to be entered
Second fae et here is fg, i.e. h.
function
Oo
e+
a7

3.1.2 Range and domain

You have seen that the function f(t) = 3 (t — 32) converts °F to °C.

The lowest attainable temperature is —273 °C and this places a corresponding

restriction on temperatures in °F.

Domain
of f
|
32 °F

In a Cartesian graph of a function, the domain is all or part of the x-axis and the
range is all or part of the y-axis.
182 FUNCTIONS
oe
ee et) ee

For example, the function

f(x) = x? — 2x
has domain all real numbers but the range consists
only of the real numbers greater than or equal to —1,
that is {y € R: y > —1}.

The function

F(x) = V(x-4)
has domain {x € R: x > —4} and range
{y eR: y > 0}.

When only the formula for a function is given, it is usual to take as the domain
all the numbers for which the formula can be worked out; for example, given the
function g such that

you would assume that the re

domain is all numbers except 1.
The range is then all numbers
except 0, as can be seen from
the graph.

@ 0
Give the natural domain and find the corresponding range for the function h
such that

Sketch the graph of y = x* and deduce the sketch for i= es

oe
 ALGEBRA OF FUNCTIONS 183

Two or more values in the domain

of a function can correspond to
the same value in the range. Such
a function is said to be many-to-
one. For example, the function
Dis
y = x” is many-to-one.

The function y = x° is an example of a

one-to-one function.

However, it is a requirement that a value in a function’s domain must correspond

to only one value in the range. Functions, therefore, cannot be one-to-many
(although they can be many-to-one). So, for example, x = y” does not define y as a
function of x because, if x = 9, y = +3 (i.e. there are two values in the range).

3.1.3 Inverse functions |

To return to the example of temperature, the inverse function reverses what was
done by f — in other words it converts °C to °F. In building up the function f,
two functions were used.

5
subtract 32 multiply by 9 f(t)

So f(t) = 8(t — 32)

To find the inverse function you need to ‘undo’ this.

9
f(t) multiply by 5

The inverse of f is denoted by f—'. The flow diagram above shows that
fo-(t) 20 32
As a quick check, note that
f(212)=100 and f-'(100) = 212
184 FUNCTIONS

As noted in the last section, if f(x) = x” then both f(3)

and f(—3) give f(x) =9 and consequently f~'(9) is
ambiguous, since it could mean both —3 and +3.
A function must be unambiguous — a single input
must give rise to a unique output. Thus f does not
have an inverse function if its domain is unrestricted.

The inverse function, f~'(x), does exist if the domain is restricted to

non-negative numbers, that is if f(x) = x” {x > 0}, then f~'(x) = \/x.

In general, a function must be one-to-one in order to have an inverse.

In order to avoid ambiguity, mathematicians

take ,/x to mean the positive square root of x,
and \/x, °/x, */x,..., to mean the positive
fourth, sixth, eighth, ... roots of x.

The same problem does not arise with ¥/x af

because x — x° is a one-to-one function.
y=Vx
—__— >»
x

Example 2
1
Find the inverse of f(x) = j {ett
—x
Solution
Flow chart for f 1

i 1
f"(y) =1--
y
As a quick check, note that (3) = —0.5, f-'(—0.5) = 3.

The letter y can be changed for any other and it is usual to write
i eee x
 ALGEBRA OF FUNCTIONS 185
a

If a function f is to have an inverse function, the situation must be as follows.

Since f is a function, any

vertical line cuts the graph
once only.

f-' is a function: any

horizontal line from a value on
the range of f cuts the graph
once only.

Domain of f

The following examples investigate the relationship between the graph of a

function f, and the graph of its inverse function f '.

®
1 (a) What is f~'(f(x))?
(b) Explain why the domain of f is the range of f~!.

(c) Sketch the graph of a function with no inverse function.

(d) If both f and f~' are functions, explain why f must be a one-to-one
function.

2 For each function given below, choose a suitable domain so that the
function has an inverse function and plot on the same axes (which should
have equal scales) the graphs of the function, its inverse and y = x. Define
the inverse function in each case.

(a) f(x) =3x+5 (b). g(x) = x7 —7

(c) h(x) = (x —7) (d) r(x) = /x+6

3 What simple transformation will map the graph of y = f(x) onto the graph
of y = f(x)?

4 The graph of:

(a) f(x) = («+543 . (b) f(x) =3Q«e—1)

is reflected in the line y = x. Find the equation of the image.
186 FUNCTIONS

5 (a) Investigate the sequences:

1
(1) X_¢1=— — (il) %y41 = —%Xy
Xn

for various different values of x, in each case.

(b) What is the inverse of:

1
(i) f(x) =-, the reciprocal function, and
x
(ii) f(x) = —x, the ‘change sign’ or ‘multiply by —1’ function?
Why do you think these functions are called self-inverse? Sketch the
graphs of the two functions and explain how they are related to what
you observed in question 3.

6 Find the inverse of each of the functions f defined as follows.

Bie =< (SoS eer) ere

il 8
(i) ie OF (d) f(x) = Pie {a> —1}
(pie) = /W@—27)) 10K 1) 4 CO) EG) =4 = Ge te
Which of these functions are self-inverse? What are the equations of the
lines of symmetry of the graphs of the self-inverse functions?
 ALGEBRA OF FUNCTIONS 187

Example 3
Th 6a) = (= 3)? 44 {x > 3}, find f~'(x). Sketch the graphs of f(x) and f~'(x).

Solution
To find the inverse function, represent f by a flow chart showing the simpler
functions which compose it.

Then reverse the flow chart.

a y Saal a :

F-'(x) = f(x -4)4+3 {x>4}

Note that x is usually chosen to represent the input variable for the inverse
function as well as the original function.

The graphs of the function and the inverse function are sketched below.

y = (x — 3)? +4

3.1.4 Rearranging formulas

The process of finding an inverse function is identical to that of rearranging a
formula. In the temperature example,
f= (632)
converts °F to °C, which could be written as
C= 4(F— 32)
 188 FUNCTIONS

f-"(t) = 2¢+32
converts °C to °F, which could be written as
F=2C+32
The formula for C in terms of F has been rearranged to give F in terms of C.
The process can be seen as applying the same function to both sides of the
formula and is often set out in this way.
C= 3(F — 32) Multiply both sides by 2
2C = F—32 Add 32 to both sides

2C432=F
This approach may be used for finding inverse functions and is equivalent to the
flow diagram as shown below:

Example 4
Make x the subject of the formula y = (3x + 1)’.

Solution
(Bee) Take the square root of each side.
y/ y= oe Subtract 1 from both sides.
Ey) yp lee Divide both sides by 3.

The ‘+’ is appropriate here as we are not defining functions, simply changing
the subject of the formula.

There is more practice on rearranging formulas on tasksheet $2.

3.1 Tasksheet $2 — Rearranging formulas (page 536)

A particular difficulty arises with the flow chart method when the letter that is to
be the subject of the formula appears more than once. A different strategy is then
required, which is illustrated by the next example.
 ALGEBRA OF FUNCTIONS 189 |

Example 5
Find the inverse of the function

fe @=25 ee)—2
% 1

Solution
Let y = f(x), then rearrange to find x in terms of y.
1
es zat Multiply both sides by x + 2.
Reape
y(x +2) =x+1 Multiply out the brackets.
yx +2y=x+1 Collect x terms together by subtracting 2y
and then subtracting x from both sides.
yx —x=1-2y Factorise the left-hand side.
x(y—1) =1-2y Divide both sides by y — 1.
T= 2y
x= ——
eos |

Thus the inverse of the function which maps x to e 52 is

x
o 1—2y
f'(y) = Viel
x is conventionally chosen as the input variable and so it is usual to write
1 — 2
fa ee x —_
eR. eet

3.1 Exercise 2

1 Make x the subject of the following formulas.

@) y= (b) y=3(x-5)
1
OE eee MI ieee
2 Taking the formulas in question 1 as of the form y = f(x), write down in each
case the formula for f~'(x), stating also the greatest possible domain and
range of f for which f~' can be defined.
3 A major chemical company researching crop yields tries out a new pesticide.
The results indicate that, per hectare, for akg of pesticide the extra yield ykg
of a crop is given by
90a
y
Pte
(a) What is the formula which gives the amount, akg, of pesticide needed to
return an extra yield of ykg?
(b) Explain why the values of y will lie between 0 and 900.
190 FUNCTIONS

4 Make x the subject of the formulas:

x—-1 2—x
(a) y =
eri (b)
b
y =
es

5 The graph of
os=:
a yey
is reflected in the line y = x. Find the equation of the image and use a graph
plotter to check your answer.

(Hint: how are the graphs of y = f(x) and y = f'(x) related?)

1+? Hi
(cin a eg ee {x <= 1}. “find f-7(@).

3.1.5 Parameters and functions

If a holidaymaker takes £x into her local bank, which offers an exchange rate of
8 francs to £1 and charges a commission of £4, the formula

Vi Bat)
gives y, the number of francs that she will receive.

The formula for an exchange rate of a francs to £1 and a commission of £b

would be y = a(x — db).

The formula y = a(x — b) is of a more general kind than those met in 3.1.4, and
the roles of a and b are different from those of x and y. a and b can vary, but for
any given function mapping x onto y they will act as constants. They are called
parameters.

y = 8(x — 4) gives the value of £x in francs, at the exchange rate given above. To
find the number of pounds required to buy y francs you first need to rearrange
the formula, giving
Y
=-+4+4
eo ak

Using the more general formula, the value of y francs in pounds is found by
rearranging y = a(x — b) to give

x=24b
a

The same techniques used in finding the inverse of a function will be suitable if
parameters replace numbers.
 ALGEBRA OF FUNCTIONS 191

Example 6
Make a the subject of the formula s = ut + 1gt’.

Solution
s—ut= Lat? subtracting ut from both sides

2s — 2ut = at’ multiplying both sides by 2

US 22
= =!= =% dividing both sides by 2
Ds 2;
soa — = 2 it
E is

3.1 Exercise 3
1 Make the variable shown in brackets the subject of the formula.

(a) P=aW+b (W) (b) C=2ar (7) (c) s=F(a+l) (1)

a 1k 4
(d) s= (r) (e) rane (x)

2 The driver of a car travelling at v m.p.h. sees an obstruction ahead of him

and immediately applies the brakes. The distance, d feet, that the car travels
from the time that the driver sees the obstruction until the car stops is
given by
as 10)?
5
Ge 20
(a) Find the stopping distance for a car travelling at:
(i) 30m.ph. Gi) SO mph. (iii). 70 mph.
(b) Rearrange the formula to find v in terms of d.
(c) The driver sees an obstruction 250 feet ahead of him. What is the greatest
speed at which he can be driving if he is to pull up before he reaches the
obstruction?
3 If a metal rod is heated, the length increases according to the equation

where / is the final length, Jp is the initial length, ¢ is the increase in

temperature and a is the coefficient of expansion of the rod.
Find a if a steel rod of length 1 metre expands to a length of 1.004 m when
heated through 230 °C.

4 Einstein’s famous equation E = me” gives the energy equivalent to a mass m,

where c is the speed of light. Rearrange this equation to find c in terms of E
and m.
192 FUNCTIONS

5 If a body of mass m moving with velocity u is later observed to be moving

with velocity v then the change in kinetic energy of the body is given by
seri ZB th 2
E=5;mv" —5mu

Rearrange this formula to give v in terms of E, m and u.

6 A pendulum consists of a light steel rod with a heavy metal disc attached to
the end. The time, T, taken for the pendulum to swing through a complete

ren)
cycle is given by

where / is the length of the pendulum and g is a constant.

Complete the following steps to find / in terms of T and g.

ED ery ea: = v(;) [square both sides]

g

For a grandfather clock, T = 2 and g = 9.81 in SI units. The equation then

gives a value for / in metres. Find this value.

7 (a) Show that the surface area of a solid cylinder of radius r and height h is
S =2nr(r +h)
(b) Rearrange the formula S = 27r(r + h) to give h in terms of S and r.

8 In rearranging formulas a student proceeds as follows. Simplify her working in

order to arrive at more elegant solutions.
(a) The volume V of a cone of radius r and height h is
Vi
V=t1nrbh >ic= rh

l(@
30

(b) The total interest £1 on £P invested at r% for n years is given by

pelt SS tegep
~ 100 Pais
I x 100
i
>Tr=
nN
 ALGEBRA OF FUNCTIONS 193

9 Locate and correct the errors in the following.

(a) The current I flowing in a circuit consisting of a resistance R and n
batteries of voltage E each and internal resistance r each is given by
jet Ei jah
SeRear eR nr
eR Ir
(b) The kinetic energy E of a body of mass m moving with speed v is
E= Lim?

vem) L

FE

3.1.6 Functions and transformations of graphs

You have seen that the graph of the inverse of a function is obtained by
reflecting the graph of that function in the line y = x. You can now look at other
transformations of graphs and find how the equations of the resulting graphs
relate to those of the original graphs.
194 FUNCTIONS

1 Which of these graphs can be mapped onto other graphs in the diagram?

2 What transformations would map:

(a) graph b onto graph c,
(b) graph a onto graph d,
(c) graph 6 onto graph e?

3 Is there more than one possible answer to any of parts (a) to (c)?

In chapter 1 you saw that the image of the graphof y = x* under the translation
7)
| is the graph of y = (x + 2)* + 3. Using function notation, this can be
expressed as
; —2 1h,
the image of y = f(x) under a translation a is y= f(x +2)+3

In general:

ee
.1D

1 The function f is defined by f(w) = w’*.

(a) Write down the expressions f(x), f(x) +2 and f(x + 3), then plot
the graphs of y = f(x), y = f(x) +2 and y = f(x + 3) on the same
screen.
(b) What translation would transform the graph of y = f(x) onto the
graph of:
(i) y=f(x)+2 (ii) y= f(x +3)?

2 The function g is defined by g(u) = S

u
(a) Write down the expressions g(x), g(x + 4) and g(x + 4) + 3, then plot
the graphs of y = g(x), y = g(x + 4) and y = g(x + 4) +3 on the same
screen.
(b) What translation transforms the graph of y = g(x) onto the graph of:
(i) y=g(xt+4) (ii) y= g(x + 4) +3?
 ALGEBRA OF FUNCTIONS 195

3 5
The graph of y = — is translated through | as shown above. Suggest an
x
equation for the new curve. Check your answer by plotting the graph of
your equation.

4 By ee the square, rewrite y = x* + 4x +3 in the —

y= (x+p)+
What eanlacon will map the graph of y = x’ onto the graph of
y = x* + 4x + 3? Check your answer by plotting both graphs on the same
screen.

Example 7
4
Find the image of the graph of y = 5,/x under a translation |]:

Solution
Taking
(x)= 5,/x% sand) p= —=4, g=3

then
f(x +p) +q = f(x -—4) +3 =5V(x —4) +3
so the image of

y= Sx

y = 54/(x
— 4) +3
 196 FUNCTIONS

aee Exercise 4

(Do not use a calculator in this exercise.)

Use translations of simple graphs to sketch the graphs of the following. In each
case give the equation of the basic graph and the translation used.

1y=x'49 4y=x-2

2 y=(x-1) 5 y=5(x
—3)? +6
3 ’
Pe
ay eae 61 y =x = 2x

3.1.7 Combining transformations of graphs

In 3.1.6 you saw how translations of graphs changed their equations. Here, we
shall consider other transformations and combinations of transformations of
graphs, together with their effect on the general equation y = f(x).

Graphs b, c and e can be mapped onto each other by translations, reflections or

rotations. Graphs a and d can be mapped onto each other by a reflection or a
half-turn. Some of the graphs can be mapped onto each other by a combination
of transformations.
 ALGEBRA OF FUNCTIONS 197

11
Graph b maps to c by a translation through | or a reflection in the y-axis.

Graph a maps to d by a reflection in the x-axis or a half-turn about the origin.

Graph c maps to e by a half-turn about the point (55,0) or a reflection in the

x-axis.

Graph b maps to e by a half-turn about the origin or reflections in the x-axis and
in the y-axis.

These ideas are explored below.

.1E

1 (a) If {(w)= w* — w, write down expressions for f(x), f(—x) and —f(x)
and draw their graphs.

(i) What transformation will map y = f(x) onto y = f(—x)?

te What transformation will map y = f(x) onto y = —f(x)?
) For any function f, will the transformations found in (b) always be the
same? Give reasons for your answer.

2 (a) f(x) =x* — 2x°. The graph of y = f(x) is reflected in the x-axis. Use
the ideas of question 1 to find the equation of the new graph. Plot both
graphs to check your answer.

(b) The original graph is now reflected in the y-axis. Write down the
equation of the new graph. Plot the graph of your equation to check
your answer.

3 (a) If f(x) = 3x* — x*, write down f(—x). Plot the graphs of f(x) and
f(—x). Explain what occurs.
(b) If f(x) = x° — 5x, write down f(—x) and —f(x). Plot the graphs of f(x),
f(—x) and —f(x). Explain what occurs.

|
_. ( = f(x), chessGi
is called an evenn function andi
itsie has lineor
reflection symmetry in the y- axis. _ _ -

- If f(—x) = —f(x), dee fis oo an odd function and its ne has

| rons — symmetry about the origin.

| 4 Classify the following functions f as odd, even or neither.

(a). (x) =5 (by £(x) = 2x° + 3x° (c) f(x) =x? +2

198 FUNCTIONS

5 Classify the following functions f as odd, even or neither.

| (a) (b)

|
f(x) f(x)

| y «Vv

|
| f(x) & f(x)

: !

| f(x) & f(x)

| LA. "
6 f(x) =x? +3x — 2. The graph of y = f(x) is first reflected in the x-axis and
then the new curve is reflected in the y-axis.

(a) Find the equations of these two new curves and plot the three graphs

|
to check your answers.

(b) How could you have transformed the first curve onto the third using a
single transformation?

Me!
7 £(x) = 2x° —-. The graph of y = f(x) is first reflected in the y-axis
x
| 4
and then translated through | . Find the equation of the final curve.

Check your answer by using the graph plotter.

 ALGEBRA OF FUNCTIONS 199

Example 8 :
ee i —2 Tne
Find the image of y = ve after a translation of 7 followed by a reflection in
the x-axis.

Solution
=)
Under a translation of Ae the image of y = f(x) is y = f(x +2) —7. So
3
the image of y = = is

(x +2)
df= a4 a
Under a reflection in the x-axis, the image of y = f(x) is
y = f(x)
23
So the image of y = et —7 is

ee (x +2)’ ety
ee
: 4
Ie;

Pe — ep(x+2)
ee
2 4
You have now seen the effectsof translations and reflections in the axes. The
general rules for these transformations are summarised below.

eS
 200 FUNCTIONS
a

wsPe wane, Exercise 5

1 For each of the following pairs of graphs, the equation of one graph is given.
Find the equation of the other.
Use the graph plotter to check your answers.

Ya (b) ya
6
: ALGEBRA OF FUNCTIONS 201

2 Find the images of the graphs of the following functions under the
transformations given.
1 -

(4) Sy
i
reflection in the x-axis followed by a translation through
~6.
— (ha
= oe
reflection in the y-axis followed by a reflection in the x-axis
1
(c) y a
Zz
translation through 3 followed by a reflection in the x-axis

4
202

Functions

22 Circular functions

3.2.1 Rotation
The mathematics of the circular functions describes the behaviour of many
physical systems. Those illustrated below are but a few examples; others include
the motion of the tides and the current flow in an electric circuit. There are
many more.

f te
& RY ae

Although some of these movements are obviously circular (the rotation of the big
wheel, for example), others are not (for example, tidal movements). However, the
same mathematics is used to describe their behaviour. This is the mathematics of
the elementary trigonometric functions — the sine, cosine and tangent functions.
 CIRCULAR FUNCTIONS 203

Our study begins with definitions of the functions themselves. Initially, we shall
concentrate on the sine and cosine functions.

Consider a point rotating around a unit circle (a circle of radius 1 unit) as

shown.

So h = sin @° and d = cos @°, where @° is the

angle through which the point has turned
from the horizontal position.

The values for d and /) are obtained from the elementary ratios in a right-angled
triangle.
P
In a right-angled triangle for which the
hypotenuse is of length 1,
d ’
sin® = 7 and con 1 a

i.e. sin 6° = and cos?° =d

Ait °
The definition of the sine and cosine in terms of the coordinates of a point
rotating around the unit circle apply for all values of the angle 0°. If the point P
moves in an anticlockwise direction then the angle generated is taken to be
positive. In a clockwise direction the angle is negative.

By considering the height / of the point P above the horizontal as @ varies, the
sine curve can be obtained. The cosine curve is obtained by considering the
horizontal distance, d metres, from the origin O.

A few values are calculated below.

P
hb = sin 30°

(0,b) = (30, 0.5) AN

a= cos 30

(0, d) = (30, 0.866)

 204 FUNCTIONS

b= sinl20"

(0,h) = (120, 0.866)

d—cos20"

(6, d) = (120, —0.5)

b= sin 210°

(0,b)= (210, —0.5)

d= cos 210°

(0,d) = (210, —0.866)

The sine and cosine functions are both periodic — that is, they repeat themselves
after a certain interval known as the period. In the case of both sine and cosine
the period is 360°.

Observe that sin (‘anything’) is always a number between +1 and —1 inclusive

(the range of the function); similarly, for cos (‘anything’).

3.2 Exercise 1

1 (a) Use your calculator to give the value of sin 50°.

(b) Write down, from the sine graph, six other angles which have the same
sine as 50°.
 CIRCULAR FUNCTIONS 205

2 (a) Use your calculator to find cos 163°.

(b) Write down, from the cosine graph, five other angles which have the
same cosine as 163° and which lie in the range —360 < 6 < 720.

3 (a) Use your calculator to find sin 339°.

(b) Write down, from the sine graph, five other angles which have the same
sine as 339° and which lie in the range —360 < 0 < 720.

3.2.2 Transformations
In this section we consider the effect of various transformations on the graphs of
the sine and cosine functions. This will help when you sketch graphs of related
functions, such as cos (2x + 60)°, for example. Consider first the sine and cosine
functions themselves.

ya y=sin& yA y=cos@

@ sin@° is an odd function whose graph has rotational symmetry about the
origin.

sin (0°) = — sin #°

so that sin (—50°) = — sin 50° = —0.766

@ cos @° is an even function, as the graph has reflection symmetry in the y-axis.

cos (—0°) = cos &°

so that cos (—50°) = cos 50° = 0.643

—90°
e The graph of cos @° may be obtained by a translation of 0° |of the sin@®
graph.
sin (9 + 90)° = cos #°
so that sin 120° = cos 30° = 0.866
Further relationships are explored in 3.2a, which follows.
206 FUNCTIONS

AN

1 Use a graph plotter, working in degrees, to plot the graph of

y= sind
Investigate the graph of y = asin @° for various values of a, including
negative values, and describe the transformations involved. a is called the
amplitude of the function.

2 (a) Investigate y = sin b@° for various values of b and comment on the
significance of the factor b.

(b) What is the period of sin b@° in terms of b?

3 Investigate y = sin(@+ c)° +d for various values of c and d and describe

the transformations involved.

4 Investigate:
(a) y=e0s 00° (b) y = cos (b6 + c)°
for various values of b and c and describe carefully the transformations
involved.

The diagram shows a part of the graph of

y = asin (b6 + c)°
Find a, b and c.

6 Describe fully a sequence of transformations which maps the graph of

‘Vi= (COS Om
onto the graph of
y =acos(b0+c)°+d
where a, b, c and d may take any values. Illustrate your conclusions with
appropriate diagrams.
“Ne
CIRCULAR FUNCTIONS 207

You should have noted the following results.

These results are entirely consistent with the results for the polynomial functions
with which you are familiar. This is an important observation and one we will
meet again.

Example 1
(a) For y = 3cos (60 + 180)°, what are the amplitude, period and phase shift?
Sketch the graph.
(b) Describe a sequence of transformations which maps the graph of y = cos 0°
onto that of y = 3 cos (60 + 180)°.
=

Solution . | + 180)°
= y = 3 cos (60
(a) wre

Amplitude 3.
Period 60.
: Phase shift —30°.
208 FUNCTIONS

(b) The transformations are:

@ a one-way stretch, parallel to the x-axis, of scale factor 2, followed by
@ a one-way stretch, parallel to the y-axis, of scale factor 3, followed by

; ee
@ a translation ae

It is informative to track some of the points from the graph of y = cos @° to

y = 3cos (66 + 180)°.

For example, point A (360°, 1)

(360°, 1) — (60°, 1) > (60°,3) — (30°,3) =A’
point B (270°, 0)
(270°,0) — (45°,0) — (45°,0) — (15°,0) = B’

y = 3cos (68 + 180)°

J Ae es
 CIRCULAR FUNCTIONS 209

3.2 Exercise 2

1 Suggest suitable equations for the following graphs. Check your answers
using a graph plotter.

180 360 0

360 6

>
100 6

A wheel of radius 0.2 metres rolls along a straight horizontal line. Initially, a
spot P on the rim is directly over the centre C. After turning through 6°, the
height of the spot P is / metres. Find an equation for / in terms of @ and
sketch the graph of h against 0.
 210 FUNCTIONS

3 (a) Describe a sequence of transformations that maps the graph of y = sin x°

to y = 2sin (3x + 60)°.
(b) Find the image point of (i) (180°, 0) (ii) (90°, 1)
under the transformations in (a).

3.2.3. Modelling periodic behaviour

The sine and cosine functions are ideal for modelling many situations which are
periodic. As you will see, the input to a sine or cosine function need not be an
angle.

:
A big wheel has a radius of 4.8m
| ¢
and a seat in the lowest position
is 0.8 m above ground level. One
h complete revolution takes 60
seconds.
/i
!
Y
WU
|
1 If a seat starts from the bottom in position A, and if after t seconds it has
turned through an angle 6°, express @ in terms of tf.
| 2 Draw a rough sketch to show
|
how the height in metres, h, will vary
with (a) 6, (b) ¢.
:
(There is no need to perform any detailed calculations.) In each case,
suggest a possible formula for h.

3 Plot a graph of the first part of the motion by completing the following
table of values of 4 for various values of 0.

D OFF 330" =60)-90 -120> aioOle 180

| 4 Repeat question 3, but this time using ¢ as the variable.

|| t Oia Ss, {10a . 20 25 30
'
@
 CIRCULAR FUNCTIONS 211

You may have noticed that a subtle change has occurred in this example. Until
now sine and cosine have been used exclusively with angles as input. In part (d),
however, the input to the function was t. There is no reason why, having drawn
the basic graphs of the circular functions, you should not use any variable you
choose as input.

Example 2
When a particular tuning fork is
struck, each prong vibrates at a
frequency of 256 Hz (hertz, i.e. cycles
per second) with a maximum
displacement at the tip of 0.3 mm.
(a) Sketch a graph to show the
displacement of the tip of a
prong with time.

(b) Assuming that this is a sine’

graph, express d, the
displacement in millimetres, as a
function of t, the time in seconds
from the start of the motion.

Solution
(a) Assume the initial displacement is 0.3 mm.
da

(b) d = asin (bt'+c)°

Amplitude: a=0.3

360 1
Period
iod: 556
~—=—= 5) = 92160

—C il
Phase shift ft: —=-——3c=90
7 ie 9

So d = 0.3 sin (92 160¢ + 90)°

 212 FUNCTIONS

Example 3
(a) B A wave machine in a swimming pool comprises a
cylinder of radius 2m which rotates at 1 revolution
every 10 seconds. The cylinder starts with the bar,
4 B, uppermost and has rotated through an angle 6°
after t seconds. A is a fixed point just beneath the
cylinder. Express 0 in terms of t.

(b) Hence write down the height, after t seconds, of the bar above

(Hr; (ii) A

Solution
(a) There is a rotation of 36° in 1 second
So 0 = 36t

(b) (i) Height = 2 cos (36t)° (ii) Height = 2 + 2 cos (36t)°

3.2 Exercise 3

1 As the Moon circles the Earth, its gravitational force causes tides. The height
of the tide can be modelled by a sine or cosine function.
(a) Assuming an interval of 12 hours between successive high tides:
(i) sketch the graph of the height if it is 5.7 metres at low tide and 7.3
metres at high tide;
(ii) use the graph to help express the height of the tide, metres, as a
function of the time ¢ hours after high tide.
(b) Express / as a function of t if h is 3.6 at low tide and 4.9 at high tide.

2 The times for sunset at four-weekly intervals over a year are as follows.

Jan. 2 16:03 July 16 21:10 \

Jan. 30 16:45 Aug. 13 20:27
Feb: 27 17:36 Sep. 10 19:26
Mar. 26 18:24 Oct. 8 18:22
Apr. 23 20:11 Nov. 5 16:26
May 21 20:55 Dec. 3 15:54
June 18 OAS

Plot these data on a graph and, making any necessary adjustments, find a
suitable function to model the data approximately.
 CIRCULAR FUNCTIONS 213

A mass oscillates up and down at the end

of a spring. The unstretched length of the
spring is 12cm, and it is extended to
14.5cm and released. One complete
oscillation takes one second. Sketch a
graph to show the length of the spring as a
function of time. Assuming that this is a
cosine graph, express /, the length in
centimetres, as a function of t, the time in
seconds from the start of the motion.

4E The ‘science’ of bio-rhythms is based on the belief that an individual’s

behaviour is governed by three cycles which begin at birth. The physical cycle
(P), with a period of 23 days, governs such things as strength, confidence and
aggression; the emotional cycle (E), with a period of 28 days, affects feelings,
creativity and cooperation; and the intellectual cycle (I), period 33 days, covers
intelligence, concentration, memory and quickness of mind.

Critical days in an individual’s behaviour occur when the graph of a cycle

crosses the time axis.

Each cycle of the physical curve may be modelled by a sine wave using the
equation
2/360? Ve
P = sin | ——
25

(a) Suggest suitable models for the other two cycles.

(b)) Calculate your current values for P, E and I.
(c) When is your next critical day for each of the three cycles?
(d) On what days of your life are all three cycles critical?
214 FUNCTIONS
Neen
ee eee EE EEE EEE

3.2.4 Inverse trigonometric functions

In studying periodic behaviour it is often useful to be able to solve problems
involving the inverse functions. For example, a student who believes that his
intellectual bio-rhythm is governed by the equation

may wish to know on what days his value for I is greater than 0.9. In order to do
this he would need to solve the equation

360t \
sin {|—— (0.9
65

Before you can solve equations like this, you need to be able to solve equations
of the form sin x° = a.

180 360 5$40\ x

=I

1 (a) Use your calculator to find a solution of sinx°® = 0.4.

(b) Use the graph above to find three more solutions in the range
ORS 5405

(c) Write down two solutions in the range 3600 < x < 3960.

(d) Write down two solutions in the range 360” < x < 360(n + i}

(e) Does your formula apply if is negative?

Your solution to question 1(d) is known as the general solution of the

question. It is possible to write the general solution in a rather more elegant
form, as question 2 demonstrates.
 CIRCULAR FUNCTIONS 215

2 (a) Find a solution, p, of sinx° = 0.5 in the range 0 < x < 90.

(b) Two more solutions are 180 — p and 360 + p. Write down the next two
in the same form.
(c) Write down the 20th and 21st terms of the sequence starting 180 — p,
360 + p.
(d) Write down the nth term. (Hint: (—1)" equals +1 if 1 is even, —1 if n
is odd.)

3 Find the general solutions of:

(a) sinx: = 0.7 (b) sin x° = —0.7 (c) <os.%° =J0W7,

Example 4
Solve the equation sin x° = 0.6 where 0 < x < 360.

Solution
The calculator gives a single value of x = 36.9.

However, from the graph there is clearly an infinite number of solutions.

They include
180 — 36:9 = 143.1), 3604-36.9 = 396.9, 360.+ 14331 = 3503.1,

36.9 — 360 = —323.1, 143: 1— 360'= —216.9

The general solution is 180n + (—1)"36.9.

The solutions in the given range are

S09 velgond. (when n = 0 and n = 1)

When considering the inverse function of sinx° (x € R), you should recall that,
in general, a mapping from a set A to a set B is a function if and only if every
element a of set A has a unique image in set B. For continuous functions of real
numbers, this means that a vertical line drawn on the graph must cut the graph
at exactly one point.
216 FUNCTIONS

Inverse sine (the

graph of sinx
reflected in y = x)

sinx is a function. It is a many-to-one mapping.

From the graph you can see that inverse sine x is not a function. For example,
the image of 5 under inverse sine includes 30, 150, 390, ... It is a one-to-many
mapping.

The function called sin~! below is sometimes called arcsin to avoid confusion
with the one-to-many mapping.

The graphs of sin-' x and cos! x are shown below.

 CIRCULAR FUNCTIONS 217

Example 5
(a) Find the value of cose (0125),

(b) Solve cosx° = —0.25 for —360 < x < 360.

Solution
(a) A calculator gives cos '(—0.25) as 104.5°, which is the principal value.

(b) A graph shows there are four solutions. Angles having the same cosine
are found using the symmetry of the graph. They are:

A —360 + 104.5 = —255.5° Principal

value
—104.5°

104.5°

lez
* 360 — 104.5 = 255.5°
She),

3.2 Exercise 4
1 Give the (principal) values of the following.
(a) sin™' 0.2 (b) cos'0.9 — (c) sin '(—0.36)
(d) cos-'(—0.74) — (e) sin™*(1) (£)' cos (1)
2 Solve the following equations, giving solutions in the range —360 < x < 720.
(aye six <= 0.3 (b)\ cos x == 0:8 (c) cosx° = —0.3
(Masia == 0:5 (e) cosx) = —1 (G9 eines. Set

3 Find general solutions to the equations

(Ves cinx = 06 -(b)<cosx = —08
218 FUNCTIONS
Ceee

3.2.5 Solving equations

2D)

Suppose that the height of the tide, / metres, at a harbour entrance is

modelled by the function
h = 2.5 sin 300° + 5

where t is the number of hours after midnight.

ha
7.55 h = 2.5sin 30+ 5

6-4

25}

T = See
ees ee BO IG I eS US? te

1 When is the height of the tide 6 m?

2 If a boat can only enter and leave the harbour when the depth of water
exceeds 6 m, for how long each day is this possible?

Example 6
A girl is sitting on a big wheel which rotates once every 30 seconds. When the
wheel begins to rotate for the ride, she is sitting in the position marked A on the
diagram. The diameter of the wheel is 16 m.

(a) Show that her height y metres above the

lowest point of the wheel t seconds later
is given by
y = 8+ 8sin (12t + 30)°
(b) At what times is she

(i) 15 metres above the ground,

(ii) at the highest point?
 CIRCULAR FUNCTIONS 219

Solution
(a) AN = 8sin (12t + 30)°
So height of A = y = 8 + 8 sin (12¢ + 30)°
(b) (i) When y = 15, you need to solve the equation
8 + 8sin (12¢ + 30)° = 15
=> 8sin(12t+30)° =7
= sin (12t + 30)° = 0.875
Now solve sinx°® = 0.875, where x = 12t + 30.
The calculator gives x = 61.0 so the possible solutions are
F610) 18021610, 0.360-61 05
= 12t+30=61.0, 119.0, 421.0, 479.0,:..
=> $= 26) TAs. 326, 374.5.

asrst

61.0 x

(ii) At the highest point, y = 16, so you have to solve the equation
8 + 8sin (12¢ + 30)° = 16
= ssinti2e 4-30)" 8
= sin (127-230) = 1
Now solve sinx°® = 1, where x = 12t + 30.

This time there is sin x°

1
only a single solution
for each cycle and
you do not need a
calculator to tell you
that the basic solution
is x = 90; So the
possible solutions are
190.) 60H 90. 720 42.90.1080. 90, . ».
= 1273090, 450, 810,.-1170,...
= PIGS 185. gro5 195
 220 FUNCTIONS

The following tasksheet provides further practice at solving simple trigonometric

equations.

3.2 Tasksheet $1 — Solving equations (page 537)

3.2 Exercise 5
1 Find the values of t in the range 0 < t < 60 which satisfy the following
equations.
airssin 101i; = 5 (b) 4—7cos(t+35)° =0
(c) 3+4sin
(8t — 21)° =0 (d) 10cos$#°
=9
2 The height above ground of a chair in a big wheel is given by
h = 5.6 — 4.8 cos 6t°
where t is the time measured in seconds from the instant when the chair is at
the lowest point. For how many seconds during one complete revolution is
the chair more than 9 metres above ground level?
3 If the height of the tide is ) metres at time t hours, where
h =5+2.5 sin 30¢°
find all the times in the first 24 hours when the height is:
(a) 6.7 metres (b) 4.5 metres

4 A cowboy ties a handkerchief to a lasso which he then spins so that the

height in metres of his handkerchief above the ground after t seconds is
given by
b=2-- 1 Sisin 5007
Find at what times the height of the handkerchief above the ground is:
(a) 2.75 metres (b) 2 metres (c) 3.5 metres
 CIRCULAR FUNCTIONS 221

3.2.6 tan 0°
While sine and cosine are the most commonly used, they are certainly not the
only periodic functions. Another important periodic function is the tangent
function.

& .2E

A tennis umpire, U, is watching a rally between two players. The ball, B, is

hit straight down the court from P to Q over the centre of the net, C.
U
fs
tle
5m \
|
IeFold Ge ae ae ae ee a FE, Ec B ah Pare =@

SS i Se ee LES

1 What is (a) length y (b) angle CUQ?

2 Sketch a graph to show how y varies with @ as the ball travels

(a) from C to Q (b) from P to Q.

3 Using the sides of a right-angled triangle, show that if 0 < @ < 90 then
sin 0°
tan @° =
cos 0°
4 What is the greatest possible domain for tan?

5 Find a suitable set of principal values for tan” ‘x.

Although these results were illustrated for 0 < x < 90, they are true for values of
x outside this range.

The graph of y = tanx® can be transformed into the graph of

y = atan(bx + c)° +d in the same way as the graph of y = sin x° is transformed
into the graph of y = asin (bx + c)° +d.
222 FUNCTIONS
Sy
eee a

The graph of the tangent function is shown.

yc tain

The inverse tangent function has the following graph.

y = tan! (x)

Example 6
State the general solution to the equation tan x° = 0.8.

Solution
The calculator gives x = 38.66. However, it is clear from the graph above
that there is an infinite number of solutions. The solutions occur every 180°,
so the general solution to the equation is
x = 38.66+180n (n=0,+1,+2,...)
 CIRCULAR FUNCTIONS 223

SEE

a2 Exercise 6 -
1 Find the values of e
(gatane 2 (bi tan c=6)p nc eane0

2 Sketch the graph of

(a) y = tan 2x° (b) y = tan (x + 45)°_

3 Solve the equations for values of x in the ranges 0 < x < 360.
(aN EAN C= 1S (b) Stan (2x + 30)° =4
(c) tan? x° = 1 (d) 4sinx° = 3cosx°

4 State the general solutions to the equations in 3(a) and 3(b).

y
;
224

Functions

3 Growth functions

3.3.1 Exponential growth

Under favourable circumstances some organisms exhibit a particular type of
unrestricted growth. The graph shows the growth of a number of bacteria
starting with roughly 3000 at time t = 0 (hours).

N (thousand)
12

11

10

0 1 25 33° G45) 6) 2A VSO 10S (hours)

From the graph, there were 4000 bacteria after approximately 2 hours and 8000
after 7 hours. There were 5000 bacteria after approximately 3.7 hours and 10000
after 8.7 hours.

Notice that in each of the cases above it takes 5 hours to double the number of
bacteria. If the number of bacteria doubles in any period of 5 hours, you could
use this to estimate the time for 24000 bacteria. Doubling does appear to take
place in all S-hour intervals. Since there are 12000 bacteria after 10 hours it
is
reasonable to suggest that there will be 24000 after 15 hours. Similarly, since
there are 3000 bacteria after 0 hours, there would have been 1500 bacteria
5
hours earlier, when t = —S.

It can also be seen that the number of bacteria trebles approximately every
8 hours — for example, 3000 after 0 hours, 9000 after 8 hours; 4000 after
2 hours,
 GROWTH FUNCTIONS 225

12.000 after 10 hours. Given any fixed time period, the number of bacteria
increases by the same factor during that time, independent of the number at the
start of the period.

Example 1 |
The graph shows the growth of world oe
population from 1650 to 1950. Is the sets)
growth exponential?
3000

2000

1000

1650 1750 1850 1950 Year

Solution
The populations in 1650, 1750, 1850 and 1950 were approximately 500
million, 700 million, 1300 million and 2500 million.
The growth factors over successive hundred-year intervals are:

Time interval Growth factor

1650-1750 00 aah 14
=
1750-1850 ay = 1.857
1850-1950 2500— 1.923
2500 __

The growth factors for successive equal time intervals are not constant, so the
growth is not exponential.

Exponential decay occurs when the growth factor is less than 1.

Example 2
A scientist was analysing the decay of a radioactive form of lead, lead-214.
The mass of lead-214 remaining in a particular sample of lead was as follows:

Time 0 1 2 3 4 5 6 7 8
(minutes)
Mass (kg) 3.127 3.047 2.969 2.894 2.820 2.748 2.678 2.609 2.542

Was the radioactive lead decaying exponentially?

 226 FUNCTIONS
ee

Solution
3.047
In the first minute, the growth factor was 37 0.974.

As you can verify, the growth factors in succeeding minutes were all 0.974.

The lead decayed exponentially (the constant growth factor was less than 1).
nn nee EEE

3.3 Exercise 1

1 To attract new investors, a construction company publishes its pre-tax profit

figures for the last ten years.
Profit before tax
(millions of pounds)
1989) Gren 27/20)
1989 ee 24
1990 Ne
eee) 5Ono,
19949 ees ed) 4627.
1992) Se) 560
993 ee ee Ge
1994, See OOo
1995) eee)
SS Se 677,
199 63 Se reer ee arren| waGal
1997 SS ee eee eae RS

Was the growth of profits exponential?

2 £5000 is deposited in a fixed interest building society account. The amount in

the account increases as shown below.

End of year if Z 3 4 5
Amount £5450 £5940.50 £6475.15 LAOS £7693.12

What is the interest rate? Is the growth exponential?

3 A girl’s annual pocket money is £50 plus £10 for each year of her age. Does
her pocket money increase exponentially with age?

4 The following table shows the population of Latin America over a period of
24 years.

Year 1950 1954 1962 1966 1974

Population (millions) 164 183 DLE, 254 315

Is this exponential growth? Justify your answer.

 GROWTH FUNCTIONS 227

5 A capacitor is an electronic component which can store charge.

(a) A capacitor is initially charged to 9 volts. It is discharged. across a
particular circuit, the voltage dropping by one volt each second.
(b) In another circuit, the voltage would have dropped by one quarter of its
value each second.
Are either of the above examples of exponential decay? Find the growth
factors, if appropriate.

3.3.2 Indices

1 week ago ‘In lweek In 1% weeks In 2 weeks

A culture of algae doubles in area each week. Now it covers 1 cm’, so in a week
it will cover 2cm’, in a fortnight 4cm’, and so on. The growth of algae is
exponential with growth factor 2. In t weeks the area, A, will be:
A=2'
The number 2 is called the base and ¢ the index (plural: indices).

The following examples introduce definitions and laws for powers of 2 and for
other base numbers.

@ 3:
The area, A, covered by an algal growth is initially 1cm*. The growth
increases exponentially with time, t, in such a way that A = 2’, where t is
measured in weeks.
A (cm?)4
20
228 FUNCTIONS

So far 2’ has been defined only when t is a positive whole number. It

is nevertheless sensible to join the points on the graph with a smooth
curve.

1 (a) Explain why, for whole number powers,

(ya ee) ee Pie SSG ee

(b) Simplify (2”)”.

2 From your graph, find 2° and interpret 2° in terms of growth of algae.

3 (a) What was the area of algae a week before measurements started?
How would you define 27'?

(b) Use the same approach to define

Go oeeares Mi More (ie

4 You do not have to restrict yourself to doubling. Suggest values for:

(a) 377 (bys? (c) 10°° (dye (6) ame

You need to check that the laws of question 1 are still being obeyed.

5 (a) Use the laws to simplify 27* x 27°.

(b) Express 2~* and 2~° as fractions.

(c) Multiply the two fractions in (b) to check that the answer in (a) fits in
with the laws of indices.

(d) Verify that the laws hold for:

(2 DPE) GS Ae DMT ep ee ie!

6 Decide which of the following are equal.

(A
(a) (5) (B} Syo1py lel och O oyeee
8
7 (a) If 2? is to obey the laws, what must (22)? equal?
(b) What is (./2)??
How will you define 2 23 ’ 2m, a?
 y GROWTH FUNCTIONS 229

8 Find: (a) 92 (b) 83 —(c) 644 (d) 81°

9 (a) Explain why 47 = (42)? = (4°):
(b) Find the value of: (i) 83 (ii) 163
\

10 Check the laws of indices using the graph and your calculator. For
example, on your calculator find \/2 and 2°°.
Read off from your graph the value of A when t = 0.5.
You should experiment with various bases and both signs of indices.

We can use these laws for a < 0, where it is meaningful to do so. Note that,
while a° = 1, no value is assigned to 0°, which is undefined. Equally, 0~ is
undefined.

Example 3
ns 9 : A ee
By using a combination of several laws of indices, evaluate 83.

Solution

See =
 230 FUNCTIONS
a
eee Ee ee

Lad 3 Exercise 2

1 Express the following as single powers of 2.

(GOP woe che 2) (ne
2 Simplify the following.
AQ ba ca (cd ed doe
3 Evaluate:

es (b) 10°? (ys. (Eee

4 Simplify:
Glgy yee a(Bl ce == ca a uC) urcn oa (ras
r
5 Evaluate these, checking your answers using the x” or x» key on your
calculator.
(a) 42 (b) 25-2 ~— (c) 25-2 ~— (d).: 10000003 ~—_(e) (0.022
6 Use your calculator to solve the equation 2’ = 10 by trial and improvement,
correct to two decimal places.

Further practice in using the laws of indices can be found on tasksheet S1.

3.3. Tasksheet $1 — Laws of indices (page 538)

 GROWTH FUNCTIONS 231

3.3.3 Growth factors

In 3.3.2 you considered the growth function with equation y = 2”. In many fields
of study it can be useful to find equations which closely model given data. For
example, the population figures for England and Wales from 1841 to 1901 are as
follows.

Year TSF ed 85 te “thelr 1371 Sissi 1891, “1901

Population, 15.9 Weg 20.1 De, 26.0 29.0 B22):
P millions

30

25

20

15

10

1841 $1 61 71 81 Oo lea iear

If you could fit an equation to such data, then you could make reasonable
estimates for populations in years when a census was not taken, and you could
also project the figures beyond the years for which data are available. However,
great care must be taken in making such projections because changes in
conditions dramatically alter population trends.

The table gives the data in 10-yearly intervals. To find out whether it is suitable
for modelling using a growth function, you can check to see if the 10-yearly
growth factor is approximately constant.

Year Population Growth factor

1841 1529
1851 V9 Ills}
1861 20.1 Alt
1871 Zoe itsils}
1881 26.0 ileils)
1891 29.0 ‘i
1901 32.5 1:12
232 FUNCTIONS

The 10-yearly growth factor is roughly constant, so the data can be modelled
using a growth function. To do this, it is necessary to find an estimate for the
yearly growth factor.

In the sixty years from 1841 to 1901 the population grows by a factor of
SPS)
i595 2.044. If the yearly growth factor is a, this means that
a = 2.044 > a = 2.044% = 1.012
Some of the properties of the equations and graphs of growth functions are
investigated in the following questions.

3B

You have seen that y = a” may be used as a model for exponential growth.
Here you will see how, by changing the function to Ka”, you can model any
exponential data.

| 1 (a) Investigate the graph of K x 2* for various values of K. What is the

|
significance of the factor K?
(b) What is the significance of K if y = K x a?

| 2 (a) If y =a’, what is the initial value of y, i.e. the value of y when t = 0?
(b) If y = K x a’ what is the initial value of y?

3 Investigate the graph of K x (})* for various values of K. What is the

significance of the factor K?

||
4 Use the ideas developed in questions 1-3 to sketch the graphs of:

(a) y=Sx30 (bey 2G) (livsaxs ta linea:

|
Check your answers using a graph plotter.

S The population P of Great Britain has been estimated at 1.5 million in 1066
and 6.1 million in 1700. Assume that an exponential growth model is

| appropriate and that ¢t years after 1066, P = Ka’.

(a) Write down the value of K.

(b) Use the data for the population in 1700 to explain why

| 1
6.1 \ 634

|
a= |{—
il.3)

||
Evaluate this to 5 decimal places.
(c) Use this model to estimate the population of the UK in 1990.

(d) Comment on the model.

 GROWTH FUNCTIONS 233

Example 4
Model the population data for England and Wales given on page 231 with an
equation for P in terms of t, the number of years after 1841.

Solution
Assuming the growth is exponential, P = K x a’. K = 15.9, the initial value
and, since 32.5 = 15.9 x a® the annual growth factor can be estimated by
1
32.5 \ 60
-_— ~ 1.012
1559
The equation is then

P = 15.9 x 1.012°
A check on the suitability of this model can be made by comparing tabulated
values of the original data and populations predicted by the equation.

t 0 10 20 30 40 50 60
K 15.9 5Wg) 20.1 227 26.0 29.0 32.5
15.9% 1.012" 15.9 i7.9 20.2 22.7 25.6 28.9 325)

Although there is some variation, this model gives results close to the true
values.
The model predicts that the population in 1990 would be 15.9 x 1.012”,
that is, 94.0 million. This figure is much higher than the true value of about
49 million. Many factors have affected this, including the world wars, family
planning and a different social structure (mothers working outside the home,
and so on).

3.3 Exercise 3
1 A colony of bacteria has a growth factor of 6 per hour. Initially there are 400
bacteria.
(a) After how many hours will there be 14 400 bacteria?
(b) When will there be 1000000 bacteria?
(c) Write down an expression for the number of bacteria t hours after the
start.
234 FUNCTIONS

2 The compound interest on a savings account is 8% per annum.

(a) What is the growth factor?

(b) Explain why the number, n, of years before an initial investment of £4000
grows to £5000 is given by 1.08” = 1.25.
(c) Find an approximate value for n.

3 A radioactive element, bismuth-210, was observed every few days, and the
mass remaining was measured.

The following figures were obtained.

No. of days from start 0 2 3 6 7; 10

of experiment
Mass (kg) LOW e577 SOS 4534S eee at

(a) Estimate the growth factor.

(b) Write down an equation for M, the mass of bismuth remaining, in terms
of t, the number of days from the start of the experiment.
(c) Check how well your equation models the data.
(d) How much will remain after 3 weeks?

(e) What is the half-life of bismuth-210 (i.e. after how many days does only
half of the original amount remain), to the nearest whole day?
4 In an electronic circuit, the voltage V across a capacitor drops from 15 volts
to 6 volts in 12 seconds. Assuming that the process is one of exponential
decay, find a formula for V in terms of ¢, the time in seconds from the start.
—__—_—_—_———,T-_
«XKeee —_—_"::§:/972707??
 GROWTH FUNCTIONS 235

3.3.4 Logarithms
The graph shows the growth of aquatic plants starting with an initial surface
coverage of 1m’.
Surf:
ur ae The time taken to reach
2m* and 4m” are 1 and 2
weeks respectively.

eat The growth factor is 2 per

week.
S Area (A) =2’
A is the surface area
covered after t weeks.

0 1 2 3 Time (weeks)

The exponential equation expresses A as a function of t. Often though, you

require the inverse function, i.e. t in terms of A. This inverse function is called
the logarithm of A to base 2, written as log, A.

From the graph, the area is 8 m? after 3 weeks, i.e. 8=2°. Conversely, it takes
3 weeks before the area is 8m’, i.e. 3 = log, 8.

Since logy x is the inverse function of 2”, the graph of log,x is the reflection of
the graph of 2” in the line y = x. Since log, 1 = 0, the graph cuts the x-axis at
x= 1.

Note that log, x is not defined for x < 0.

236 FUNCTIONS
ee eee eee ee ee ee ES

Example 5
Write down the values of:

(a) log,32 (b) log;25 (c) logy 1000 (d) log; 27

Solution
(a) Since 32 = 2°, 5 is the logarithm of 32 to base 2, i.e. log) 32 = S.

(b) Since 5* = 25, log; 25 = 2.

(c) 10° = 1000 => logy 1000 = 3
(d) 3° = 27 = log; 27 =3
Properties of the ‘log’ function are developed in the questions which follow.

ReX@

You know that

= 2 exe = loepy

1 Use this result to find:

(a) log,64 = (b) logyg = (c)_ logy2_~— (d)_ logy 2

i
Just as you can find the logarithm of a number to base 2, you can find
logarithms to any positive base. The power of a which equals y is called
log, y. So,

= a =x = los y

|
y
2 Use this definition to find:

/ (a) log;9 (b) log; 125 (c) logs # (d) log;1 (e) loge x1

(f) log; ¥3 — (g) logy2 —(h)_ logy11. (i): logs -3

|
3 Find: (a) log,1 (b) log,a —(c) lone (d) log,a
a

4 Use your calculator to find: (a) logig 10°” (b) 1081037

|
Explain your findings.

5 (a) Writedown: (i) log,8 (ii) log) 16 (iii) log, 128

(b) Since 8 x 16 = 128, you can write this as 27 x 2” = 2°. What are a,
b and c? How is « related to a and b?

(c) Use this to explain why log, 8 + log, 16 = log, (8 x 16).

BHD
SESS
BBO
0
 GROWTH FUNCTIONS 237

Cs 3
Explain how this verifies that log; 9 + log3 27 = log; (9 x 27).

7 (a) Use your calculator to verify that 3 + 10S 10

(b) What is: (i) logyo3 (ii) logy5?

(c) Use these results to find logg 15.

8 Use your calculator to verify that logi) 9 + logy) 8 = logy 72.

Questions 5—8 suggest that logs are related by the law

log, m + log, n = log, mn

In fact, it is possible to prove that this result is true for any positive base a by
using the result
q)oBa* =

The proof is as follows.

a log, m+log,n a=
_ _log,m
xd log, n eb
(law of indices)
=i tt
=e q)°Ba (mn)

=> log, m+ log, n = log, (mn)

Question 9 extends this result to logs of quotients.

l
9 In the law above, replace n by —. Hence show that
m

log, / — log, m = log,—

m
Verify that this holds by choosing two arbitrary numbers for / and m.

10 logio 20 S010 logio 3 = 0.4771

Use the properties of logs and the result that log;) 10 = 1 to find (in any
order):

logins, logio1.5,. logio2.5, logio4, logio S,

login 6, logio8, logio9
238 FUNCTIONS
re

You have obtained the following results.

In pre-calculator days, tables of logarithms were used to help perform various

calculations. Part of a table of logarithms to base 10 is given below. From the
table, logjg 1.351 = 0.1306 and so on.

In 1615, the Scottish mathematician John Napier discussed the idea of using
logarithms with the Oxford professor Henry Briggs. Two years later, Briggs
published his first table of logarithms (to 14 decimal places!) and after much
further work published his Arithmetica Logarithmica in 1624.

Nowadays logarithms can be found using a calculator but originally their

calculation involved considerable hard work and ingenuity. ‘Log’ is by
convention usually taken to mean logj9, and you will therefore find that the [log |
key on calculators evaluates logarithms to the base 10.
 GROWTH FUNCTIONS 239

3.3 Exercise 4

1 You have seen that

2 =8= log, 8 =3
From these equations with indices, form equations using logarithms.
Gis =9 (By 4 = (ce) (0.5 = 4
(d4) 3=2 (ce) 7=9
2 Write down the values of:

(a) logs (3) (b) logs 125 (c) log; (4) (d) logs (3)

3 Simplify:

(a) log;9+ ee 27 — log; 81 (b) logs 15 — logs 3 (c) 2log, /7

4 Sketch, on the same axes, y = logig x, y = logjg 2x and y = logy 3x. How are
the graphs related? Use the laws of logs to explain this relationship.

5 (a) Use the log tables given earlier to calculate 1.05 x 1.267.

(b) Use the properties of logs to write down log;9 10.5 and logy 1267.
Hence use log tables to find 10.5 x 1267.

6 The notation 4! means 4 x 3 x 2 x 1 and is read as ‘4 factorial’.

If log; 4! = 1.9746, write down log; S!.

7 A colony of bacteria doubles every hour. Explain why the time ¢t hours for the
colony to increase in size 1000-fold is given by 2’ = 1000. Express t as a
logarithm to base 2 and explain why 9 < t < 10. Use a numerical method to
find t to two decimal places.

3.3.5 The equation a = b

In answering the problem in exercise 4 about a colony of bacteria you used a
numerical method to solve the equation

2 = 1000

Problems concerning growth often lead to such equations, in which the unknown
occurs as an index.
240 FUNCTIONS
oo a

Example 6
Suppose that a radioactive isotope decays by 10% each year.

m (g)
500

400

300

200

100

2 4 6 8 10 12 14 ¢ (years)

(a) Initially there is 500g of the isotope. Find an expression for the amount ¢
years later.

(b) The half-life of the isotope is the time taken for the amount present to
decrease by 50%. Use the graph to estimate this half-life.

(c) What equation must be solved to find the half-life more precisely?

(d) Use a decimal search to solve the equation in (c).

Solution
(a) If 10% of the isotope decays each year then 90% will remain, so the
growth factor must be 0.9. After t years, 500 x 0.9° will remain.

(b) The half-life will be the time taken for the amount to drop to 250 grams.
From the graph, this is approximately 6.3 years.

(c) To find the half-life more precisely, the equation

500 x 0,9° = 250

must be solved. This simplifies to

0.9
= 0.5

d
() t 6 a) 6.5 6.6 6.58
0.9" 0.531 0.478 0.504 0.499 0.49994

Equations of the form

Gb
can be solved by a numerical method. There is, however, a more direct way of
solving such equations, which is explored in the following questions.
 GROWTH FUNCTIONS 241

3D

1 (a) Use your calculator to find the relationship between

(i) logy 49 and logy 7 (ii) logy) 64 and logy, 2

(iii) logy) 125 and logy 5

(b) What is the relationship between log, m? and logyo m?

2 (a) Use the result log, mn = log, m + log, n to explain why

log, m” = 2log, m.

(b) Generalise this method to explain the law that you found in
question 1(b).

3 What is the relationship between log 2* and log 2? By taking logs of both
sides, use this relationship to solve the equation
ery:

4 £1000 is invested in an account which earns 1% interest per month, all

interest being reinvested. ;

(a) Explain why the number m of months taken for the total investment to
reach £2000 is given by the equation
012)

(b) Find m.

5 The halt lites t days, of bismuth-210 is given approximately by the

equation

10x (01872) = 5
Find its half-life in days, correct to 2 significant figures.

Example 7
Find the half-life of the radioactive isotope considered in example 6.
 242 FUNCTIONS
2
ene
ee

Solution
500 x 0.9° = 250
= O09" = 0:5 (divide both sides by 500)
=> log0.9 =log0.5 (take logs of both sides)

=> tlog0.9= log0.5 (using the property of logs above)

(You may use either of the

_ log 0.5
~ 6.58 logarithm buttons on your
~ log 0.9
calculator here.)

The half-life is approximately 6.58 years.

3.3 Exercise 5

1 Solve for x:

Gy.vias) (bo S28 a ee =e

(ds? 10.05) Clas =) (Osos
2 Explain how you could have obtained the answers to 1(a), 1(b) and 1(c)
without using a calculator.

3 A colony of bacteria has a growth factor of 3.7 per hour and initially there are
250 bacteria.
(a) Write down an expression for the number of bacteria after t hours.
(b) Find the time (to the nearest minute) after which there are 10000
bacteria.

4 A capacitor is discharging with a growth factor of 0.9 per second. After how
long will there be i of the original charge? (Give your answer in seconds, to
2 d.p.)

5S The number, 7, of years needed for an investment of £4000 to grow to £5000

at 8% per annum compound interest is given by 1.08” = 1.25. Find n using
logarithms.

6 In 1980, the population of Africa was 470 million and growing at a rate of
2.9% per annum. In what year will its population reach one thousand million
according to this model?

7 In 1980, the population of China was 995 million and growing at a rate of
1.4% per annum. After how many years will the population of China equal
that of Africa?
a
 GROWTH FUNCTIONS 243

3.3.6 Using logarithms in experimental work

A very important practical use of logarithms is in experimental work, and you
may have the opportunity in your other studies to employ the following
procedure.

Suppose that you obtained the following results while investigating the
relationship between the time of swing of a pendulum and its length.

Length of pendulum, / (m) 0.6 1.0 1.4 1.8 Des

Average time of swing, t (s) 1.54 2.03 209 2.67 LIT

It would be natural to plot the graph of t against I.

4
The shape of the graph does not make it easy to
predict what the relationship is, although it clearly
indicates a definite relationship which would be 5
convenient to represent algebraically.

1 2 l

If either the shape of the graph or other considerations lead you to suspect that
the relationship is that t is proportional to some power of /, then you can test
this with the help of logarithms.

he kl"
log
t = log (k/”)

logt = nlog!+ logk

Comparing this with the equation of the straight line, y = mx + c, indicates that,
if the power relationship is correct, then the graph of logt (the ‘y’ variable)
against log/ (the ‘x’ variable), should be a straight line having gradient » and
intercept on the log t-axis of logk.

For convenience (calculators have an appropriate button!), we have taken the

logs to base 10 and tabulated the values below.

lox ee 0222 C000 0146s 40055) «0.342

logiot 0487. ¥ 0307- 96/03387, 90426. -. 0.473
 244 FUNCTIONS
ee
ES

The graph is plotted here. You can read off the

values for 1 and logk.
logio k & 0.30 Ig?

>k=10°=2 (to2s.f.) Gradient = n

1050

Hence the relationship is

ae chess
Ey = 0:2 =0: 1200 WO, 10: 28010 logy
or =

A very similar method is employed if the relationship is thought to be a power

function. If
y= ROW
then taking logs of both sides of this equation gives
log y= log (kb*)
= log (b*) + logk
logy = xlogb+logk
A graph of logy against x would be a straight line of gradient logb and intercept
log k.

Wa Cae Exercise 6

1 In an experiment to determine how the sag of a beam varies with the distance
between its supports the following results were obtained

d (cm) 500 540 580 620 660 700

z (cm) see 13. LZ, 2.0 2.4 2.8

where d is the distance between the supports and z is the sag.

By drawing the graph of logz against log d, determine the power of d to which
z is proportional.
 GROWTH FUNCTIONS 245

2 The charge on a capacitor, t milliseconds after a switch is closed, is believed to

decay exponentially. The results of a set of measurements are as follows.

Time, ¢ 2 3 4 5 6
Charge, O 7.6 2.8 1.0 0.4 0.1

If O = kb’, by taking logs and plotting an appropriate straight line graph, find
the relationship between O and t.
3 The period of oscillation, P seconds, for bars of uniform material is thought to
be proportional to some power of their length, L metres (P = RL”). One set of
measurements is given in the table below.

is 2 3 4 5) 6
ie 45 Oye) 6.4 Ton. 8.7

However, one result has been incorrectly copied. Plot an appropriate straight
line graph. Give the corresponding formula connecting P and L and state what
you think the incorrect result might have been.
4 The rate of decay of a radioactive substance is usually measured in terms of its
half-life.

Suppose that the formula for its mass at time t is m = mo e * where mp is the
initial mass, e = 2.718 and k is a constant with a value particular to these
circumstances.

If m = 0.9m when t = 2, show that k = 0.0527; hence find the value of t

when m = 0.5mo.
246

Functions

4 The number e
3.4.1
You saw in section 3.3 how functions given by equations of the form y = Ka®™
can be used to model growth. In this section we shall look in detail at rates of
growth for these functions and see how all functions of this kind are very closely
related.

In the following questions you will investigate the gradient function for y = a’.

4A

You will need a graph plotter which can plot the gradient graph of a
function, so that you can check your results.

dy , the gradient function.

1 (a) Sketch the graphs of y = 2* and a
%
ye
(b) Suggest an equation for the function kdl
age and check your answers using
a graph plotter. 2

2 Repeat question 1 for each of the following functions.

(Dp mmm) peed she 0 (5) n5) all

d
3 Suggest a value for a for which —(a*) = a. Check your answer by
sketching appropriate graphs.

4 Suggest an appropriate gradient function for ke”, where e is the value for a
that you suggested in question 3.

The gradient function for y = 2* can be found by ‘zooming-in’ at any point

P (x,y) on the graph of y = 2” until the curve looks straight.
Q
The gradient of the graph at P is
approximately equal to the gradient of PQ,
where Q is a nearby point on the graph.
P (% y)
0.001
 THE NUMBER e 247

5 (a) Explain why the y-coordinate of Q is 2°12".

0.0015x _ 5%
b) Explain
(b) why the
Explainwhy thgradient
dient PQ
PQisis———_—_—_..
0001

(c) Show how the expression in (b) can be simplified to 0.693 x 2”.
d
(d) You have seen that y = 2” > a % 0.693 x 2*. How could you increase
the accuracy of this result?

6 Adapt the method of question 5 to find dy when y = 3”.

dx

The value of a for which k = 1 is denoted by the letter e. This gives the
important results

e = 2.718 281 828 4 to ten decimal places.

Like 7, e is an irrational number. The Swiss

mathematician Leonhard Euler (1707-1783) first “
used the letter e to represent this number. Euler /
introduced several other mathematical notations,
including that for functions, f(x). He was also the
first to use the summation sign )., the letter 7 for
the ratio of circumference to diameter of a circle
and i for ,/—1 (both i andj are in common use
today). He continued to work actively after
becoming totally blind in 1768.
248 FUNCTIONS

The shapes of the graphs of y = e* and

y =e ~ are typical of exponential growth
and decay respectively. Some calculators and
textbooks express e* as exp x. This notation
is also used in some computer languages.

Notice that = > 0 for all x for the exponential function y = e”. Such a function
Be
is called an increasing function of x.

Clearly e* is an increasing function of x, while e ~ is a decreasing function.

Exercise 1

1 (a) Use your calculator to find:

(i) & (ii) exp (5.1) (iii) e 7 (iv) 5 (v) exp (0.5)

(b) What is the largest power of e that your calculator can evaluate,
and why?

2 Make tables of values for —4 < x < 4 and draw the graphs of:
eee Dy ee
3 Draw the graph of
y= 5( 1— en)
for 0 < x < 5. Check the shape using a graph plotter.

4 (a) When certain drugs are injected into the body, the amount remaining in
the bloodstream decays exponentially. The amount of one drug in the
bloodstream is modelled by the equation
y = 5e0-2#

where t is the time in hours after the dose is administered, and y is the
amount remaining, in milligrams.
(i) What is the initial value of y?
(ii) What is the value of y when t = 10?
(iil) Sketch the graph of y against t.
(b) The amount of a second drug is modelled by the equation y= 5e°™,
Does it decay more or less rapidly than the first drug?
 THE NUMBER e 249

5 A colony of bacteria grows according to the law y = 4e’, where ¢ is measured

in hours and y is the population.
ae d
By differentiating, show that = = y. What does this tell you about the rate of
growth of the bacterial colony?

How rapidly is the colony growing at a time when it contains 500 bacteria?

h
—1
6E Show by a numerical method that lim (:i )= 1. Hence show, using
—0

aie ees d
differentiation from first principles, that —(e*) = e”.
dx

3.4.2 e*
One of the great benefits of introducing e as a base for the growth function is
that it can replace all the other bases. This simplifies subsequent work,
particularly in calculus.

Example 1
(a) Find alternative expressions
p of the form 2” for (i) 8° (ii) 5°

(b) Is it always possible to express b’ in the form 2”?

Solution

(ii) 5° = | 2" > 5 =2%, ie. a =~"

log 5
log2
=2.32. Therefore 5¢ = 227%
(b) When 6 > 0, it is clear that you can always write b’ in the form 2”, since
you can always solve the equation b = 2’. As has been noted already,
when b < 0 the meaning of b’ is not defined for some values of t (for
example, t = 5).

This idea of changing the base of an exponential function is explored below.

>> 4B

1 (a) Use a graph plotter to verify that the graphs of 9* and 3” coincide.

(b) Find the value of a so that the following pairs of graphs coincide.
(i) 5” and 3” Gi? sandes a (iti) -2%and 3
250 FUNCTIONS

It appears that, for any positive value of b, you could replace b* by 3”. In
other words, only one base is needed for all exponential functions. The base
used in practice is not 3 but e.

2 (a) Use a graph plotter to sketch the family of curves y = e*

(i) for a few positive values of a, of your own choice,

(ii) for a few negative values of a.

What shape is the graph if a = 0?

(b) If a> b > 0, describe the relationship of the graph of e” to that of e2”.
For what values of x is e% > e?*?

3 Using the same method as in question 1, find the value of a so that the
following pairs of graphs coincide.

(a) 5* and e” (b) 8% and e”* (c) 2” and e”

d
4 (a) You know from section 3.4.1 that, if y = 2”, ae = 0.69 x 2” and, from
x
question 3, that 2* ye°° Explain how these results can be combined
to show that
7d (ee) 2m 0.69e°-*

(b) Suggest a possible derivative for e”.

: d
The previous question suggests that ——(e**) = ae. In the next question you
investigate how this result arises.

2q x

(a) In the diagram, Q and R have the same y-coordinate. What is it?

(b) What is the x-coordinate of Q?

 THE NUMBER e 251

(c) The graph of y = e” can be obtained from the graph of y = e* by

squashing by a factor of 2 in the x-direction. What effect does this have
on the gradient of the graph?
(d) If the gradient at R is g, what is the gradient at Q?

(e) Write down, in terms of q, the gradient of y = e* at the point R. Hence

write down the gradient of y = e™ at the point Q.
(f) Complete

ans (e") =... Xe

dx

@ «
(a) What is [oman

(b) In general, if us (E(x) =e(), what are:

dx

(i) (lax)
aed
Ci)? [alan a :
Example 2
eX +1 , é;
lig =5 sz }» show that y is a decreasing function of x.
e
Solution
2x
e +1 ys
y=3/ oo )=545e

dy
PEE) az a 3 (e
dx
d
Since e ~ > 0 for all x, = = —10e** < 0 for all x, so y is a decreasing
function of x.
 252 FUNCTIONS
eS ee

3.4 Exercise 2

1 Differentiate:
1
(a) e* (by cen (c) (e*) (d) ie

cam | 5
set ye +, vere hes
2 Integrate the functions of x in question 1.

3 Sketch the graphs of:

3(1 — aa) (c) y= @ a: =
(a) = ere (b) y=

4 When a drug such as penicillin is prescribed, it is usual to take it 3 or 4 times a

day. One or two days may elapse before the drug starts to take effect. A
simple model of this process is known as the ‘Rustogi’ drug model.
Suppose that for a certain drug, the amount in milligrams in the bloodstream
t hours after taking a dose of size A mg is given by
aSip
x-= Ae ®

and that a dose of 10 mg is administered at 8-hourly intervals. On graph

paper, using a scale of 1cm to 2 hours on the t-axis and 1cm to 2 mg on the
x-axis, plot the drug level in the body over the first 32 hours as follows:
(a) For the first 8 hours calculate the drug level for t = 0, 2, 4, 6, 8 and plot
these values.
(b) Add the next dose of 10 mg and plot this point, also at t = 8.
(c) This gives the ‘effective dose’ at this time, i.e. the new value for
A. Recalculate the drug level for the next 8-hour period, again taking
t= 0; 25-43 6,.3:
(d) Repeat steps (b) and (c) to show how the drug level varies in the body for
the remainder of the period.
What value does the maximum drug level in the body approach?

5 In hospitals, when it is necessary for a patient to respond rapidly to treatment,

a doctor will often give a ‘booster’ dose, equivalent to 1.6 times the initial
dose. Repeat question 4, but with an initial dose of 16 mg and subsequent
doses of 10 mg. [It should only be necessary to consider the first 8 hours.]
Note: this method is not used for drugs available on prescription to the
general public; if subsequent doses of 16 mg were taken in error the drug level
would rise to 1.6 x 16 = 25.6 mg, with potentially dangerous consequences.
 THE NUMBER e 253

3.4.3 The natural log

You saw in 3.3.4 that the power of 2 that was equal to y was called log) y, i.e.
if2*=y, thenx =log,y
You also saw how this idea can be extended to other bases, so that, for example,
if 10° = y then x = log,o y.

Since e was chosen as the base for exponential functions in order to simplify
results in calculus, it is useful to consider e as a base for logarithms. Logarithms
to base e are called natural logarithms and log. x may be written as In x (n for
‘natural’). Some older texts simply use logx for Inx.

By analogy with base 2, Inx is defined so that, if y = Inx, then x = e’. Inx and
e’ are therefore inverse functions.

Since Inx and e”are inverse functions, the application of one followed by the
other restores the original value.

At this stage you may find it useful to recall the laws of logarithms, as applied to
logarithms with base e.

Example 3 1
Express in terms of Inx: (a) In4x° (b) In (=:)

Solution
(a) In4x° = In4+4Inx® =1n44+5Inx

(b) In (=;) =In1—InV/x =0-Inx? = —$lnx

48
In (s.) Pee Le
Natural logarithms are sometimes called Napierian logarithms, after John
Napier. However, this is a misnomer as they are not the logarithms originally
developed by Napier.

Some of the properties of Inx are explored in 3.4, especially its gradient function.
254 FUNCTIONS

.4D

The graph of Inx

1 (a) Iny (= log, y) is the power of e that equals y, that is y = e* <> x = Iny.
Use this result to find:

(i) Ind (ii) Ine (iii) Ine?

(iv) ie (v) Ree (vi) In (—1)

e e
(b) Use your results to sketch the graph of In x.

Derivative of Inx
The graphs of e* and of Inx are sketched below. Since they are inverse
functions, the graph of Inx is a reflection of that of e* in the line y = x. You
can use this idea to find the derivative of In x.

The figure shows the graph of

yee and y= Inx, Quis the
reflection of P (a,b) in the line
ee

You can use the relationship

between the two graphs to
find the gradient of y = Inx
at Q.

(a) Express b in terms of a.

(b) Write down the coordinates of Q.

(c) Use the fact that the triangle at Q is a reflection of the triangle at P to
il
explain why the gradient of the curve at © i¢ ———————— a
gradient of y = e* atP
(d) Since P lies on the curve y = e*, you know that the gradient at P (a, b)
is e*. What is the gradient of y = Inx at Q?

(e) Explain why (tn x) a

x x
 THE NUMBER e 255

As you have seen, physical situations involving growth are often modelled using
exponential functions. In solving the model, a useful step is that from an
equation of the form e“ = B to one of the form A = In B. An example of this is
in the use of the logistic curve, a model that is appropriate when growth is
limited by fixed resources. Its general equation is
> aA
y
~ 14+Ke-*
where A, K and X are constants, and its graph is as shown.

1+K

Example4
A highly infectious disease is introduced into a small isolated village, of
population 200. The number of individuals, y, who have contracted the disease
t days after the outbreak begins is modelled by the logistic equation
200
¥~ T+ 199e 02
After what time has half the population been infected?
 256 FUNCTIONS

Solution
When half the population has been infected y = 100.
200
1D -——=—=z => ee 1S
~0.2t SS
1+ 1992-94
—()>) Fae
=> e : = 709

= —0.2t = Ings
= t=—7 ln, = 26-5 days)

3.4 Exercise 3

1 Use your calculator to evaluate: (a) In3.5 (b) In 0.35 (c) In7

2 Use the laws of logarithms to prove that:

(a) In3+1n4 =I1n12 (b) In 10 — In2 = In5

In 20
(ce) 31ns
= In125 (cd) eens
In4

3 Use the laws of logarithms to express the following in terms of In x.

(ay inex; | (by In4e <(e) lee
d d 3 d
4 Find: (a) Gy (4 in) (b) q, (Inx ) (c) q, (ln 4x)

5 Use the population model of example 4 to find how long it takes for 90% of
the population to become infected.

6 A cup of coffee, initially at boiling point, cools according to Newton’s law of

cooling, so that after t minutes its temperature, T °C, is given by
Tio ee5en
Sketch the graph of T against t. How long does the cup take to cool to 40°C?

7 In the process of carbon dating, the level of the isotope carbon-14 (!*C) is
measured. When a plant or animal is alive the amount of *C in the body
remains at a constant level, but when it dies the amount decays at a constant
: _K : Sap:
rate according to the law m = mye ~’, where mp is the initial mass and m the
mass after t years.
(a) If the half-life of '*C is 5570 years, find the decay constant K.
(b) A piece of oak from an old building contains 4 of the level of *C that is
contained in living oak. How old is the building?
oo
eESFSFSFSFSFSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSsesse
THE NUMBER e 257
258

Functions

5 Radians

3.5.1 Rates of change

This section considers the rates of change of the sine and the cosine functions
and introduces an important new measure of angle. Consider first the graph of
y= sing.

y = sin 6°

T
270 360 450 540 46

d
The graph of y = sin @° has gradient a

dy

T ] ame
—360 —270 —180 270 360 450 540 46

You should be able to sketch this graph from the graph of sin 6° by noting the
points on sin @° where the gradient is zero (—270°, —90°, 90°), where it is at its
greatest value (at —360°, 0°, 360°) and at its least value (—180°, 180°,...). The
value of k is the gradient of the graph y = sin 6° at the origin.

Zooming in to the origin on the graph of y = sin 0°

and using the principle of local straightness, the ”
gradient of y = sin @°at the origin is
dy _ sin@
do 6
 RADIANS 259

Observe that:
@ the gradient graph appears to be a cosine curve, with amplitude k. It has
equation
d
- =kcos@°;

e k is the gradient of y = sin 6° at 0. Thus, from the diagram and argument

above
O°
kw for small 6

@ you can obtain a sequence of values which approaches k by taking smaller and
smaller values of 6. This is explored further in 3.54 below.

5A
For these questions you will need a graph plotter which can plot the gradient
graph of a function.
sy
1 (a) (i) Calculate the values of stor or 10s ee tad Os
sin 0°
(ii) To what value (to 5 d.p.) does your sequence of values of
converge, as 0 approaches zero?
(b) Use a graph plotter to obtain the gradient graph of y = sin 8°. Does
this give a value of k which agrees with your solution to part (a)? You
will need to be careful with the vertical scale.

Consider a sector AOB of a circle of

1 unit radius with angle 0° at the centre.
BC is the perpendicular from B to OA.
ZS Cy
2 Re A

(a) Por @ = 10; 5,2, 1 and 011 calculate:

(i) the length of BC (ii) the length of the arc BA

(b) What do you notice about the results?

3 Express the length BC and the length of arc BA as functions of 0.

From the previous question it is evident that these two lengths are
approximately equal for small values of 0. Use this fact to explain why
sing: | a
6 180
for small values of 6.
260 FUNCTIONS
i = Se) ee eS ee ee

T Sie:
ste
4 Calculate the value of —~ and compare it with the answer to
question 1(a)(11).

5 What is the gradient of y = sin 9° at the origin?

dy . ae
6 Suggest a suitable expression for a ify Sing.
do’

The questions of 3.5a demonstrate the result

i ea beeen BT -
0 (sti a mec”

At (0, 0),
T
gradient = —
180

0 90 180 @ (degrees)
If angles were measured in units other than degrees, with 7 of these new units
equivalent to 180°, then the diagram above would look like this:

At (0, 0), .
; us
gradient = — = 1
1

$7 a x(new units)

The new units would then give the simple result

ee
y = SIN X =
yZ COS X
dx

Scientific calculators provide the option of using these new units, which are
called radians and are defined by the relationship
m radians = 180°
 RADIANS 261
i

5B

(a) Sketch the gradient graph for y = cos x°.

(b) What do you think is the derivative of:

(i) cos x (x in radians) (ii) cos x°?

3.5.2 Radian measure

In most of the subsequent work on circular functions you will be using radians.
It is often convenient to express radians as multiples of 7, because the key
relationship between radians and degrees is

The symbol for a radian is 1° (‘c’ suggests circular measure); thus 7° = 180°.
Except when the radian unit is being stressed, the c is usually omitted.

Calculators will accept either degrees or radians as input for cifcular functions.
The following questions allow you to explore the different modes on your
calculator.

noe

1 Use your calculator in appropriate mode to find:

(a) sin 1° (by ‘sini f° (c) cos —5S° (d) cos —5°

(e) tang7 (f) tang 7°

2 (a) Find: (i) sin 30° (ii) sin 32°

What do you notice?

(b) You know 7 radians are equal to 180°. This can be used to establish a
number of other reference points between the two scales. Complete the
following table.

Radians NI-
>
Ale
=
yes)

Degrees £180 60 45 360

(c) What formula will convert 0° to radians?

(d) What formula will convert 60° to degrees?
262 FUNCTIONS

3 It is very easy to leave your calculator in the wrong mode! Suppose you are
asked to find sin $7 and you have your machine in degree mode.

(a) What is ait to 3 decimal places?

(b) What is sin ah (taking $7 in radians)?

(c) What answer does your calculator give if left in degree mode?

4 Suppose you try to evaluate sin 60°, but leave your calculator in
radian mode. What should you get in degree mode? What in fact do you
get?

5 Working in radians, plot on graph paper the graph of y = sinx for values
of x from 0 to 7, increasing in steps of 0.5. In addition, mark on the x-axis
i 5%,
the numbers 37, Sie 4%,
ee %, 57,
3 27.

1 Express these angles, measured in degrees, in radians.

(ajng0e” waibieg60° ewes: (d) 120°
() 60° 720) 0 ese
2 Express these angles, measured in radians, in degrees.
(a) im (b) 37 (c) —7
(d) 3x (e) —27

3 If the period of y = cos @° is 360, what is the period of y = cos 0°?

4 What are the periods of the functions with these equations?

(A) = sing (b) y = sin zt (c) y =sinwt
nnn
 RADIANS 263

3.5.3 Area and arc lengths

The angle 6 shown is measured in radians.

Since the circumference of a full circle of radius r is 277 units and the arc AB is
7) 0
— of the full circle, the length of arc AB is — x 2m7r = 16.
27 : 2a

This gives rise to the alternative definition of a radian as that angle which is
subtended at the centre by an arc of length 1 in a circle of radius 1. Observe that
1 radian is a little less than 60°. :

The area of the sector OAB is 2 of the area of the full circle. For a circle of
20
radius r, then, the sector of angle @ radians has area
0
— x ar =
Qi.
Radian measure for angles can be directly related to circles, and is therefore
often called circular measure.
 264 FUNCTIONS

3.5 Exercise 2

1 The wedge OAB is cut from a circle of radius 2cm.

(a) What is the area of the wedge?
(b) What is the length of arc AB?
(c) What is the perimeter of the wedge?

2 D
If the perimeter of sector CDE is numerically
r equal to the area of sector CDE, find r.

3 An area is to be fenced off for a crowd at a pop concert.

(a) Calculate the length required to fence off the perimeter.
(b) Calculate the maximum crowd if the police decide that the crowd density
should not exceed 1 person per 2 square metres.

400m

400m

4 OAB is a sector of a circle of radius r. Find in terms of r and 0:

(a) the length BC
(b) the area of triangle OAB
(c) the area of the sector OAB
(d) the area of the shaded segment
 RADIANS 265

5 A circular cake of diameter 20cm is cut along

AB, half-way from the centre to the rim. Show
that the angle @ is 120°. Calculate the areas of
the sector OAB, the triangle OAB, and hence
the area of cake cut off.

6E In the diagram show that the ratio of the area

of the triangle OAB to the area of the sector A
ind ind
OAB is a . Deduce that = tends to 1, and \

is always less than 1, as @ tends to 0. (Use the Z\ B

area formula $bc sin A for a triangle.)

3.5.4 More about derivatives

You have seen evidence to suggest that, when x is measured in radians,

y= sine > = cosx cos ee sinc

dx dx
These results will be proved when we do more work on these functions later in
the text. The results, however, are true and we can use them to find derivatives
of functions such as sin2x, 5cosx and 10sin0.17x. The graphs of such functions
are related to those of sinx and cosx by means of stretches in the x- and
Ne 1. Poe
y-directions. You have seen that a stretch of factor 5 in the x-direction and a
stretch of a in the y-direction maps y = sinx onto y = asin bx.

In the following questions you can use these ideas to find the derivative of
y= asin bx:

5D

It is important to remember that x is measured in radians throughout. You

will need a graph plotter which can plot the gradient graph of a function, to
check your findings.
266 FUNCTIONS

1 On the same axes sketch the graphs of y = sinx and y = Ssinx for
OSG:

(a) Describe the stretch which maps y = sinx onto y = Ssinx.

(b) What is the effect of this stretch on the gradient of the graph of
y — sine?

(c) Suggest what the derivative of y = Ssinx might be.

(d) Check your answer to (c) by using a graph plotter to obtain the
gradient graph for y = Ssinx.

2 On the same axes, sketch the graphs of y = sinx and y = sin 3x for
OS be Sie.

(a) Describe the stretch which maps y = sinx onto y = sin 3x.

(b) What is the effect of this stretch on the gradient of the graph of
viene,

(c) Suggest what the derivative of y = sin3x might be. Check your answer
by using a graph plotter.

3 Investigate the derivative of y = 5 sin3x in a similar way.

4 Using the ideas of the previous questions, find the derivatives of:

(a) y-=cos 2x (b) y = 10sin 2x (c) yy = sini

5 (a) Sketch graphs of y=3cos2x, y=3cos2x+4 and

y = 3cos2x — 1.

b) What are the derivatives of each of these functions?

6 (a) What are the derivatives of:

(1) y= a sinw. (iy asia x. (iii) y = asin bx?

(b) Write down the corresponding result for cosine functions for part (iii).
 RADIANS 267

3.8 Exercise 3

1 Find ay when:
dx
(a) y=$sinx (b) y=Scos x (c) y=0.1sinx
+ 0.5
(d) y = sin4x (e) y= cos2ax (f) y = sin 0.2x
(g) y'=3cos2x (h) y = 6sin}1x (i) y=443sin3x

2 (a) Ty =cos 2x5 what is an

dx
(b) Find |sin 2x dx.

(c) Find [cos3X ax:

>| 3.5 Tasksheet E1 — Derivative of sin? x (page 556)

3.5.5 Applications
You can formulate many problems in terms of radians so that it is easier to
answer questions about rates of change. For example, suppose the height of the
tide in a harbour at time ¢ is as shown.

|
|
|
|
I
|
|
|
|
|
|
|
PSS |
| |
| |
| |
| |
|
Be A A ee
OR ame Ou aS eenG 7 SOM Omelet Toei Seet amelie

In building up the equation which links 4 and t, observe that the graph could be
a sine function stretched by scale factor 2.5 in the direction of the y-axis and
translated 0|
:
5

h =2.5sin (wt) +5
268 FUNCTIONS
i
ee

; As
Since the period of y = sinwt is —,
Ww

—=12>w=tn
WwW

The amplitude is 2.5 and the mean height is 5 and so

p=) Seinene
Differentiating to find the rate at which the tide is changing at any time,

eee!
a os nt

dh
When t= 0, = r= 13 (to 2 s.f.)

The tide is rising most rapidly when t = 0 (and also when t = 12, 24, ...), and
dh
the rate of rise is 1.3 metres per hour. You can use the formula for ay to find the
rate of change at any time.

Example 1
A pendulum is pulled to one side and then released from rest, after which its
displacement x cm from the central position after t seconds is given by
x = 1.Scos mt. Write down the period and amplitude of the motion, and calculate
the first two times when x = 1, and its velocity at these times.

Solution oye
By inspection of the function, period = — = 2 seconds and
amplitude = 1.5 cm. o
1
x= when cosa?= ='0:666 4.4,
1S)
mt = 0.8410, 27 — 0.8410...

So t = 0.268 and t = 1.732 are the first two times, to 3 d.p.

dx
The velocity vy= — = —1.5msin zt.
dt
Substituting,

when t = 0.268..., v= —3.52cms/,

when t = 1.732..., v=3.52cms¢.

The negative sign in the first velocity indicates that the pendulum is then
moving in the opposite direction to that in which x is increasing.
 RADIANS 269

3.5 Exercise 4

1 A mass oscillates up and down at the end of a spring. The length of the
spring in centimetres after time t seconds is given by the equation
L=12+2.5 cos 2nt

Aen eek: dL
(a) Find the derivative, me Sketch the graphs of L and Ze against f.

(b) Calculate, when t = 0, 0.1, 0.25, 0.4 and 0.5,

(i) the length of the spring, (ii) the velocity of the mass.
Comment on the results.

The height in metres of the tide at a harbour entrance is given by

h=0.8 cos 2mt oS ee)
where ft is.the time in hours measured from high tide.
dh dh
(a) Find the derivative, ae Sketch graphs of h and oF against t for a 24-hour
interval. f Y
(b) Calculate the two times during the first 12 hours when the height of the
tide is 6 metres. Find the rate of change of height at both these times and
comment on the results.
(c) When is the tide falling most rapidly during the first 12 hours? Find the
rate at which it is then falling.
(d) When is the speed of the tidal current least and when is it greatest? What
factors are important in deciding when it is safe to enter or leave the
harbour?
The height in metres above ground level of a chair on a big wheel is given by
h=56—48cosdhnt
where t is the time measured in seconds.
dh
(a) Find the derivative, we and sketch graphs of h and a against ¢ for a
two-minute interval.
(b) Between what times is the chair descending at a rate greater than
0.4 metre per second? When is the chair descending most rapidly and at
what speed?
4 (a) The heights of the tide at Sheerness on a certain September day were
4.6m at high water and 1.7 m at low water.
(i) Assuming a period of 12 hours and measuring the time in hours from
high tide, sketch a graph of the height from 6 hours before to 6 hours
after high water.
(ii) Suggest a suitable formula for h, the height in metres, in terms of ¢,
the time in hours.
270 FUNCTIONS

(b) The currents in the Thames estuary near to Sheerness are given in the
following table.

Hours before high water = 4 3 ®) 1

Current in knots 0.7 0.9 tel £0 0.7

(i) Assuming that the currents after high water are the same, but in the
opposite direction, sketch the graph of current against time from
6 hours before to 6 hours after high water.
(ii) Suggest a suitable formula for the current in knots in terms of the
time in hours.
(iii) How is this related to the height equation found in part (a)?

(c) (i) Find the rate of change of / and sketch a graph of a against ¢ for
the same values as those in the previous graphs.
dt

(1) Comment on the relationship between this and the graph of the
current.
 271

Functions
.6 Transformations
3.6.1 Graph sketching
You will have noticed that the transformations of graphs have formed a central
theme throughout the text. You have seen several examples where a simple
algebraic transformation of the equation of a function has brought about a
geometric transformation of the graph.

GA

If the equation y = x* — 4x is transformed by replacing y with 3y, the new

equation is 3y = x* — 4x or y= 1x? - ox; By plotting the graphs of
y=x’ —4x and y= 4x°2 — $x you can observe the geometric effect of this
algebraic transformation.

1 Choose various values for k and describe what geometrical transformation

of the graph is produced by each of the four algebraic transformations
described by:

(a) replacing x withx+korx—k

(b) replacing y with y+kory—k

x
(c) replacing x with kx or —
k
(d) replacing y with ky or -

for the graph of:

(i) yaix?—4x (ii) y=sinx (iii) ye" — (iv) + = 1

(You may need to plot both y = \/(1 — x”) and y = —\/(1 — x’) for (iv).)

2 What is the geometrical transformation of a graph when:

(a) x is replaced with —x;

(b) y is replaced with —y;

(c) x and y are interchanged?

272 FUNCTIONS
ee eee eee eee SSS a

| 3 Describe the transformations which map the graphs of:

(a) |x| to: (i) |x +1] (ii) |2x + 1 (iii) 2|x + 1|

| (Illustrate each function with sketch graphs.)

ast e 5 tl eee ieee
(DieFOG Serer giapear au

scale factor —
k

Combining transformations can produce interesting effects. Suppose the

equation y = x* — 6x + 11 has the algebraic transformation ‘x is replaced
with —x’ followed by ‘y is replaced with —y’ applied to it. The new
equation is

(—y) = (=x)? — 6(-x) +11

> -y=x +6x+11
> y=-x’-6x-11
Understanding the relationship between algebraic and geometric transformations
can help you sketch the graphs of quite complicated functions.
 TRANSFORMATIONS 273
|
Pe:

Example 1
Describe the transformations which map
the graph of y = e* onto y = 2e**!

Solution
Replace x with x + 1.

Translate |
Fe,|;
0

Replace x with 3x.

One-way stretch from the y-axis,
scale factor ;:

Replace y with Sy.

One-way stretch from the x-axis,
scale factor 2.

.6B

In example 1 above, what would happen if you replaced x with 3x before you
replaced x with.x + 1?
ya

3.6.2 Stretching a circle

x* + y* =1 is the equation of a
circle, centre (0,0), radius 1.
Vv
x
274 FUNCTIONS

Rearranging the equation gives y = +,/(1 — x’), and plotting y = /(1 — x*)
followed by y = —1/(1 — x’) gives the graph of the full circle. Make sure
that the same scale is used on both axes so that your graph will look like a
circle.

Note that x* + y’ = 1 is not the equation of a function. If x is 0.6, for

example, then y = +0.8. A function can have only one output for any given
input.

Example 2
(a) Describe what effect replacing x with }x and y with (y — 2) has on the graph
of P+y=1.

(b) Sketch the graph of (4x)* + (y— 2)? =1.

(c) Rearrange the equation into the form y = ... so that you can plot the
graph.

Solution
(a) Replacing x with }x and y with (y — 2) gives (4x)* + (y— 2)? =1. The
unit circle, x7 + y* = 1, is stretched by a factor 3 from the y-axis and then
: . ; 0
the resulting ellipse is translated through " :

(b)
ya
3

(c) 9x +(y=2) =15 (y—2 =1-14 sy=24 /0—1e)

 Se,

TRANSFORMATIONS 275

.6C

1 What is the equation of a circle, radius r,

centre (0, 0)?

2 What is the equation of the ellipse shown in

the diagram?

3 What is the equation of a circle, radius 1,

centre (p,q)?

The shape of an ellipse is defined by the

lengths of its major and minor axes.
You can obtain the shape of an ellipse
by stretching a unit circle along the
x- and y-axes. You can then translate
the shape to any location on the grid.

Example 3
Give the equation of the ellipse shown in the
diagram.
 276 FUNCTIONS
ce
ee

Solution
The unit circle, x” + y* = 1, can be transformed onto the ellipse by a one-way
stretch factor 2 along the x-axis and a one-way stretch factor 3 along the
y-axis (replace x with ix and y with a)s giving

Gx) +Gy) =1
< 4
This is followed by a translation Be (replace x with x — 4 and y with

y — 3). The equation of the ellipse is |5(x — 4)]° + [$(y- 3)? its

The technique of taking a basic graph and transforming it to fit another has been
used several times throughout this chapter. You have, for example, seen how
both stretches and translations are used to fit the graph of y = sinx to the graph
of y = asin (bx +c) +d. You will also recall that it was understanding the effect
a stretch has on the gradient of a graph that enabled you to differentiate
i SAAS, y= silow and y= 6°.

You have also seen how transformations can be used to enhance your
understanding of some functions. Understanding how two distinct
transformations can be used to map the graph of y = Inx onto the graph of
y = Inax should, for example, have given you a greater insight into the log
function.

3.6 Exercise 1

1 The graph of y = Inx can be fitted to the graph of y = In3x by either a

one-way stretch or a translation. Describe the transformation in each case.

2 The graph of y = x* can be fitted to the graph of y = 4x7 by either:

(a) a one-way stretch along the x-axis from x = 0
or (b) a one-way stretch along the y-axis from y = 0
What is the stretch factor in each case?

3 The equation of the ellipse shown in a

the diagram can be obtained by
transforming the unit circle. The unit
circle can be fitted to the ellipse in two 0 fh 4
ways:
(a) a stretch followed by a translation
(b) a translation followed by a stretch

Specify the transformations in each case and hence obtain the equation of the
ellipse in two different ways.
ee a i
 ,

TRANSFORMATIONS 277
278

Functions

Miscellaneous exercise 3
3
1 The functions f and g are defined by f(x) = 3x — 2, g(x) =2—

(a) Write down formulae for (i) g (x), (ii) g‘£(x).

(b) Find a value of x for which ff(x) = —f(x).

2 Given the functions f(x) = 2x — 3, g(x) = 4 — 3x, find

HO) et), =fe(5)s- tele); era) t (5) eer
et 1G) atgaetn ahemat()
3 Find inverse functions of

(a) flz)=7x+10 (b) f(x) =os 2

= x + 1 = pe 2
(Cie) = ae (d) f(x) =4-x

State any restrictions on the domains.

4 Give the equation of the image of the graph of y = \/x under each of the
following transformations:
2m, —1
(a) the translation S| (b) the translation 4

(c) reflection in the y-axis (d) reflection in the x-axis

(e) reflection in y = x

Write down the image of (25,5) under each of these transformations, and
hence check your equations.

Pato
nn Express anes GG a) as a single fraction in its simplest form.

6 (a) Sketch a cosine graph with amplitude 3 and period 90°.

(b) Write down the equation of the graph you have sketched.
 —
MISCELLANEOUS EXERCISE 279

KOO

M
(a) (b)

(a) Calculate the shaded area in (a). (C is the centre of the circle.)

(b) Two circular ripples with centres C and D, each with radius 2 m, intersect
at L and M (b). Angle LCM is @ radians.
(i) Show that the area of the common region is (40 — 4sin 8).
(ii) Hence show that all three regions will have the same area when
6 —sin@ =4n.
(iii) Show that this occurs when angle LCM is approximately 132°.

Calculate the gradient of each of the following graphs at the given point:
(aly = 3 smd at O=1 (b) y = cos 26 at 0=i7

(c) y= 2 sin3@ at.0-— 2 (d) y =Scos$0 atO0=3n

The height of the water level at a jetty is shown on a post, marked in metres
above the mud bottom. One day the height / metres is given by the formula
i) SOLE
h = 5.5 + 4sin=,

where t is the number of hours after midnight.

(a) At what time during the day is the first high tide?
(b) Write down the values of h at high tide and low tide, and show that there
are 6 hours between them.

(c) Sketch the graph of h against t for the whole 24 hours.

(d) Calculate the rate at which the water is rising at 10 a.m.

Solve the following, giving solutions to the nearest degree in the range
0 sx = 180:

(a) .cos:x° =0 mY Ab \icos 2x - s=10 (ce). cos Sx” =:0

(d) 1+2cosx° =0 (e) 1+2cos3x°

=0
280 FUNCTIONS

11 The height above the road, ycm, of the pedal of a child’s cycle t seconds after
an instant when it is at its highest point is given by y = 20 + 10 cos 180¢°

(a) What is the height of the bottom

bracket O above the road?

(b) What is the length of the crank

AO?

(c) How long does it take to complete

one revolution of the pedals?
Rewrite the equation connecting y
and t if
(d) the cyclist speeds up so that the pedal rotates once every second,

(e) she changes gear so the pedal rotates once every 3 seconds.

12 For the following graphs, state their period and amplitude, and sketch them
for —360 < x < 360:

(a) y=2sinx (b) y=2-+3cosx° (c) y=0.5sin2x°

(d) YS Sen (e) y=6—3singx°

13 Write sine functions (angles in degrees) with

(a) period 90, amplitude 1 (b) period 720, amplitude 2.5

(c) period 60, amplitude 3 (d) period 1440, amplitude +

14 Solve forx: (a) 3*=4 8 (b) x'=3 §=(c) x = log,4

15 2% of a radioactive substance decays every hour. Give a formula for the

percentage remaining after T hours. Calculate

(a) the percentage which decays in one day,

(b) the half-life of the substance.

16 The stress y in a joint is believed to be proportional to some power p of the

strain x applied to it. Write down the form of equation connecting x and y.
In an experiment, the corresponding values of y are measured for five
different values of x. State what should be plotted against what to obtain a
straight line graph. How can the value of p and of the constant of proportion
be found from this graph?
 MISCELLANEOUS EXERCISE 281

17 The temperature of a cooling liquid falls exponentially. The table shows, to

the nearest degree, its temperature, 0°C, above its surroundings at 5 minute
intervals.

T (min) 0 >) 10 15 20 25) 30

6 (°C) 83 68 55 45 37 30 24
Plot a suitable straight line graph, and hence find the relation between 0
and T.

d
18 Show that x =e”, x =3e” andx = Ae” all satisfy ~,= 1K:

d
19 Show thatx=e ‘,x=Se ’andx=Ae “all satisfy 7 = 3%.

d
20 Show that x = Ae” satisfies = = 5x and find A if x = 33 when t = 0.1.

21 A bacteriologist finds that the rate of growth of a culture of a certain type of

bacteria is described by the differentia! equation

ety
ic 4m

where m grams is the total mass of the culture at time t hours after
observations began.
(a) Show that m = mp exp (0.4t) (where mpg is a constant) satisfies the
differential equation.
(b) Find mo if m = 10 when t = 0.

(c) With this value of mp find the mass at (i) t = 1, (ii) t = 2.

(d) Find the percentage increase in the mass during

(i) the first hour, (ii) the second hour.

22 The mass mkg of radioactive lead remaining in a sample ¢ hours after

observations began is given by m = 2 exp (—0.21).
(a) Find the mass left after 12 hours.
(b) Find how long it takes for the mass to fall to half its value at t = 0 (the
half-life).

(c) Find how long it takes for the mass to fall to

(i) one-quarter, (i1) one-eighth of its value at ¢ = 0.

(d) Express the rate of decay as a function of m.

282 FUNCTIONS

23 The velocity of a particle at time t seconds is ums __ | where v = 4t—¢°.

(a) Sketch the graph of v against t for0 <t < 4.

(b) Calculate the acceleration when t = 1.

(c) Find the distance travelled between t = 1 and t = 2.

24 (a) Simplify (x ty)? + (xty~2)3,

(b) Express as single fractions in their simplest form:
“i 1 if 1 iD 1 if 1
1 ——e —— ——F —

Ax 6x? ? x\2—-x 224+x

1
25 (a) Sketch the curve with the equation y= x+—.
%
(b) Suggest a suitable equation for the graph shown.

ay
 283

Problem solving
.1 Mathematical enquiries
4.1.1 Introduction
In this chapter we try to unravel some of the processes of mathematical activity.
Considerable independence of thought is necessary if you are to be able to tackle
unfamiliar problems with confidence. To develop an investigative attitude to
mathematics you must always pursue your own ideas!

The three problems below (4.14) may be attempted in any order. As you work
through them note the strategies you use — what methods you have of setting out
to crack an unfamiliar puzzle or investigating a new situation. The problems will
be referred to later, so keep a record of your solutions. All the extensions should
be attempted but they are quite demanding and you should feel satisfied if you
manage just one in depth.

It may be helpful to work with a small group of fellow students on some stages
of discussion or investigation. It is important that you should then write an
individual account and attempt to extend the work on your own.

@
1 Eleven players enter for a singles knockout tennis tournament. The two
organisers decide not to attempt any seeding but to pair at random. They
disagree about byes; one wants all byes to be in the first round but the
other insists that in each round as many players as possible should be
involved. (A player given a ‘bye’ in a round of a tournament is not required
to play in that round but carries straight on to the next round.)
Make a full analysis of the two suggestions, including in each case:
(a) the number of byes,
(b) the number of rounds,
(c) the total number of matches played.
Extend your investigation by considering numbers other than 11.

2 Think of a three-digit number, i.e. a whole number between 100 and 999.
Reverse its digits and find the difference between the ‘reversed’ number and
the original number. Check that it is a multiple of 99. Try one or two more
examples.
 284 PROBLEM SOLVING

Now try following the same procedure with numbers with two, four and
so
five digits. Comment on your findings.

Extend your investigation so that you can make and explain a general
statement about your findings.
UES

‘Everybody knows’ that equilateral triangles and squares will tessellate.

Qe

What
SSN
exactly is meant by this?

Show that:

LE
RRMULL
MRR (a) a tessellation can be made using any parallelogram,

(b) a tessellation can be made using any triangle;

| (c) no tessellation can be made using a regular pentagon.

Extend your investigation by considering other shapes. Is there any

pentagon which will tessellate? Can all quadrilaterals tessellate?

In attempting these problems you may have felt in turn frustrated, bored, then
interested and (it is hoped) finally satisfied that you had reached a pretty good
understanding. In this section, the mathematical activities which accompanied
these various emotions are analysed. A vocabulary is introduced which you
should use in future solutions.

The analysis will not be exhaustive; for example there is no specific mention of
the skill of using iterative techniques. This is because the text concentrates on
features common to many problems. Of course, you can use any method you
like to solve problems. The analysis of processes is there to help, not to
constrain.

You should read with pencil and notebook at the ready; to check the various
assertions made, work through the exercises and make notes. For each process a
brief explanation is given, followed by a few exercises. Do not get too carried
away by these exercises: just deal with the particular process under discussion. In
certain cases you will meet a problem more than once because it gives a good
illustration of more than one process.

The first example illustrates the use of mathematical processes which seem to be
particularly important. They are:

@ investigating particular cases;

e finding patterns;
@ generalising.
 MATHEMATICAL ENQUIRIES 285

Example 1
Investigate:

iP =
Pepa =
r+7+4+3?=
P+P4+3
44h =

Solution
A first reaction might be to work out these particular cases and look for a
pattern. You might then predict the next line in the sequence:

P+P4374+ 445 =
Investigating further particular cases seems to confirm the general
pattern:

P+P4R4+F4R =225=15
P+P4+3P4 F454 6 = 441 = 21°
The numbers on the right-hand side are squares of

193.6 10 dS: Gilad

You need to look for a connection between the numbers on the left-hand side
and the resulting square number on the right-hand side.

Epsiat ae

IMs leA on

fee ZePorhA = 10

eA nic Sidesai aa) = 15

SS a ea nok
In symbols, the pattern can be expressed as the generalisation

P4243? 4---+n =(14+24+34-:-+n)

It still remains to verify or prove that this general result holds for all
possible particular cases. The important process of proof is discussed later
(in section 4.2).

The two processes of investigating particular cases and finding patterns both
appear in this example but in no obviously systematic way. The processes are
286 PROBLEM SOLVING
A
ee ee

like tools used as and when the need arises. For the problem just considered, the
process of mathematical enquiry has appeared to be:

Explore particular Look for

cases ‘ , patterns

Obtain a generalisation

Find a convincing reason

for your general result

Prove your result

Often a problem asks for a general result, for example the number of byes in a
tournament with 7 players. If no solution immediately suggests itself then you
might decide to experiment with particular values of n. In the tennis tournament
problem of 4.14, you were invited to start with 2 = 11 and then try other
particular cases. Perhaps you tried the case of 12 players, or a number chosen at
random. Similarly, in the tessellation problem, a particular kind of quadrilateral
— a parallelogram — was considered initially. Although none of these particular
cases may have led directly to a general solution they can give you a hint on how
to approach the general case.

It is often (though not always) best to consider particular cases methodically in

some sort of order. Very often the smallest possible case is important but
sometimes particular cases are chosen for other reasons.

> 1B

1 In discussing the solution of equations of the form

x +ax+b=0

for various values of a and b, which particular values might you dispose of
before treating the general case?

2 In teaching a young child how to find the area of a triangle, which special
kind of triangle would you first consider?

; 1
3 In sketching the graph of y = x + —, which particular values of x would
you first consider? if
 MATHEMATICAL ENQUIRIES 287

4 (a) Find a particular set of values of the numbers a, b, c, d, e and f for

which the simultaneous equations
ax+by=c
dx +ey=f
have no solutions in x and y.
(b) Find another set for which the equations have an infinity of solutions.

In original mathematical investigations, guesswork can loom large, though this is

not always apparent in the final written account. At this level a guess is often
given the more respectable name of conjecture; for example, ‘conjecture —
knockout tournaments with 2” players have no byes’. If this is not immediately
obvious you can verify it in particular cases by drawing up tournament tables for
(1, 2,) 4, 8, 16, ... players.

Just because a conjecture can be verified for some particular values of n, it is not
necessarily true for all values of n. Having shown the results to be true in a few
cases you should then try to give a convincing argument that the conjecture is
true for all possible values of n; that is, you need to prove the result. The process
of proof will be discussed later.

The patterns formed in your investigations need not be of numbers; they may be
actual geometrical patterns, such as you made in answering the tessellation
problem of 4.14. Such patterns can give rise to further interesting problems.

Ac
1 (a) Investigate:
(1x2x%3xk4)41=
Qx3x
4% 5)ae t=
(3x4x5x6)4+1=

(b) Hence, find consecutive integers a, b, c and d such that

(axbxcxd)+1=
43681
2 The first seven terms of the Fibonacci sequence are 1, 1, 2, 3, 5, 8, 13.
(a) What are the next two terms?
(b) Continue the sequence
A Geet ieee s iS kod, $= (8.3)
PA
(c) The 3rd, 6th, 9th, 12th, ... terms are all even. List the 4th, 8th, 12th,
... terms. What do you notice?
(d) Make a conjecture about the Sth, 10th, 15th, ... terms and check it by
working them out.
 288 PROBLEM SOLVING
—
eee.

Very often the aim of an investigation is to arrive at a general result, commonly

(though not always) expressed in a formula. You may have had the satisfaction
of finding that in a knockout competition with n players the minimum number
of byes, number of rounds and number of matches played can all be given by
formulas.

The statement ‘no tessellation is possible using only a regular polygon with more
than six sides’ contains no formula but, being about a whole class of objects, is
still a generalisation.

——————

| Exercise 1

1 Generalise on this sequence of equations

1 = 1
Tae = 4
Tastes So
1+3+5+7
= 16

(a) in symbols, (b) in words.

2 The sum of two consecutive odd numbers is divisible by 4.

(a) Can you make similar statements about the sum of three consecutive odd
numbers and of four consecutive odd numbers?

(b) Generalise your findings.

3 How many diagonals has a quadrilateral? A pentagon? A hexagon? A polygon

with 1 sides?

4 (a) Of all triangles of equal perimeter, the equilateral triangle has the greatest
area.
(b) Of all quadrilaterals of equal perimeter, the square has the greatest area.

Make a general statement of which (a) and (b) are particular cases.

4.1.2 Organising your work

Organising your work is very important in rendering a problem manageable —
and its solution readable! Here, we consider the processes of choosing a notation
and classifying. In section 4.1.3 we focus on tabulation.
 MATHEMATICAL ENQUIRIES 289

The choice of notation in problem-solving is often routine. In the tennis

tournament problem of 4.14 you probably used the standard tournament table.

A. North

Semi-final Final Winner

In the digit-reversing problem of 4.14 you probably used letters to represent

the digits of a general three-digit number, the number looking like ABC having
the value 100A + 10B + C. This would have allowed you to tackle the problem
algebraically.

.1D

1 How good are you at describing objects or giving clear directions without
using a street plan?

Suppose you have to describe a diagram, such as the one below, over the
telephone. How could you do it?

2 A journey in a US city is described by the notation RSSRSL,

where

R means turn right

S means go straight on } at the next intersection.
L means turn left

How would you describe the inverse (return) journey?

 290 PROBLEM SOLVING

3 Three tumblers are arranged with just one upside-down.

Take any pair and turn them over, and continue doing this, a pair at a
time, in an effort to finish with all three right way up. Invent a simple
notation to explain why this is impossible.
|
euee\ccemamemnnnaaamammeeaanal
=

Sometimes an investigation is simplified if you deal with separate classes one at a

time. In the digit-reversing problem of 4.14, the class of numbers with an odd
number of digits behaves differently from that with an even number of digits. An
important class in the tessellation problem of 4.14 is that of pentagons with a
pair of parallel sides. The process of classifying can help in the solution of a
problem, or provide a good structure for an investigation.

.| Exercise 2

1 Name at least three different classes of:

(a) angle (b) triangle (c) solid shape

2 A strategy for the game of noughts and crosses hinges on the first move made.
How many essentially different first moves are possible?

3 Which classes of quadrilateral have exactly two lines of symmetry?

4 By classifying the integers a and b as being odd or even and by considering the
different combinations, show that the equation
G46 =4c43
has no solution in integers a, b and c.

5 Your friend has a pack of eight cards and invites you to choose one at
random.

ellialte
 MATHEMATICAL ENQUIRIES 291

He then asks three questions to which you answer ‘yes’ or ‘no’, after which
he identifies the card chosen. It later turns out that he can always do this
by asking the same three questions, in the same order. What might the
questions be?

4.1.3 Tabulation
You sometimes explore particular cases in a more or less haphazard way as you
feel your way into an investigation. To make progress you then need to display
your findings in a more orderly fashion. At this stage a table may suggest a way
ahead. Tables are important, too, in communicating information clearly to a
reader of your work. For both these reasons, in solving the tennis tournament
problem you may have felt the need to draw up a table like this.

No. of players
No. of byes
No. of rounds
No. of matches O11}
So
fF
0 OWN
BP Ww
RP
NY 4
Oo
won

Though there are other kinds of table, of properties of shapes for example, a
table is often useful when building up a sequence of numbers. In trying to find
the pattern in a sequence, a table of differences sometimes helps. Consider the
problem: what is the greatest number of points of intersection given by x lines?
Starting with small values of m you can quickly draw up a table.

No. of lines 1 2 3 4 5
No. of points 0 1 3 6 10

Perhaps you do not remember the sequence in the second line. Or perhaps you
vaguely recognise the triangular numbers but do not find this particularly
helpful. But if you subtract each number from the one on its right, the pattern
becomes clear.
Ue me Arle
First differences 1 a 3 4
This enables you to extend the table easily — the next ‘first difference’ is probably
5, so the next entry above will be 15, and so on. As a result the problem is
replaced by a much simpler one; the number required appears to be

1+2+4+3+---+(n-1)
292 PROBLEM SOLVING
ee
Se ee

Looking at differences can be extended into a useful method for finding a

generating formula for a polynomial sequence.

For example, to find the sequence generated by the quadratic expression

2n* —n+3

you can calculate its value for 2 = 1,2,3,... to obtain

Aamo 18) Ble 48.

The first differences are found by subtracting each term from the one to its right.

4 T 9 T 18 I 51 r 48...
) 9 13 Wire

The process may be repeated to find second differences.

5 g 13 Uaioe
ee il I
4 4 4...

You will see that the third differences would all be zero. This is investigated
further in the examples below.

1 Choose another quadratic sequence, i.e. one generated by an expression of

the form
an’ +bnt+c

where a, b and ¢ are integers. Find first and second differences.

2 Use a cubic sequence, generated by an expression of the form

an’ + bn? +en+d
and find first, second and third differences.

3 Use the method of differences to make a conjecture about the sequence

Se OG EO

4 Find the tables of differences given by sequences generated by the

expressions:

S If the fifth differences given by a sequence are all 30, how do you think the
sequence was generated? (Answer in as much detail as you can.)
 MATHEMATICAL ENQUIRIES 293

4. Exercise 3

1 Make a table showing the number of diagonals of an n-sided polygon for

values of n from 3 to 6. By finding first differences, conjecture the number of
diagonals of a decagon. Check your conjecture.

2 Devise a table showing the properties of bilateral and rotational symmetry for
parallelograms, rhombi, rectangles, kites and squares.

3
n ee
r(n) 1 2
s(n) 1 8

In the table, (7) is the number of ways of representing the number 7 as a sum,
using only the terms 1 and 2, the order of the terms not being significant;
for example

eee ep eat ep
s(m) is the number of ways when the order is significant; for example
4=lAFit lela teiplait2tia2e
t+ la242

Check the entries in the table, find 7(6) and s(6), and conjecture a formula for
r(n) and a rule for generating the sequence s().

4.1.4 Starting points

The following is a list of possible starting points for investigations of your own.
In considering these problems, you should use the ideas met in this section where
appropriate, and other ideas — especially that of proof — as you progress through
this chapter.

1 Chains — Divisors
The numbers which divide 12, and are smaller than 12, are 1, 2, 3, 4 and 6.
Add these divisors: 1+2+3+4+6= 16. Write 12 — 16.

Repeat the procedure for 16. Its divisors are 1, 2, 4, 8.

Add: 1+2+4+48 = 15. Write 12 — 16 — 15.

Repeat the procedure until you have

12> 16 — 15 ~ 9 > 4-3-1

Investigate with other numbers.

294 PROBLEM SOLVING

(a) Some numbers go to bigger numbers: for example 12 — 16. These are called
abundant numbers. Can you find more abundant numbers, and a pattern
within them?
(b) Numbers which go to smaller numbers are called deficient numbers (for
example 9 — 4). Investigate these.
(c) Note that 6 @ 6 and 28 @ 28. These are called perfect numbers. Investigate.

(d) 220 = 284. Can you find smaller pairs? (Pairs such as 220 and 284 are called
friendly numbers.)

2 Triangle-square numbers
The triangular numbers are 1,3,6,10,...,5(2+ 1),...
The square numbers are 1,4,9,16,...,7*,...

Which numbers are both triangular and square? The first two are 1 and 36.

3 Chains — Reordering digits

Any four-digit number is rewritten with its digits arranged in ascending and
descending order of size. The smaller is then subtracted from the larger and the
process is repeated.
Example: 7345 gives 7543 — 3457 = 4086

4086 gives 8640 — 0468 = 8172

Then write 7543 — 8640 — 8721—...
Complete this chain and then try again with other four-digit numbers.

Now try with three or five or six or n digits.

4 Stamps
A small mail-order firm finds that it is constantly running out of stamps of the
right denomination, so it calls on Professor Schomp for help. ‘Buy lots of 3p and
Sp stamps’ is his advice.
Examine this reply and consider related situations.

5 Solution in integers
In the equation
x —y =m"

x, y, m and 7 are all integers. Investigate the possibility of solutions in the cases
(ayiean'l (b) m>1, n>2 (Cage1am ee
 MATHEMATICAL ENQUIRIES 295

6 Cyclic arrangements of digits

Take any three-digit multiple of 37, 2
for example 7 x 37 = 259. ios ns
Arrange the digits of 259 in a cycle: 4 é
| aaa eg

and construct the other two numbers having the same three digits in the same
cyclic order, 592 and 925.

Both of these are also multiples of 37:

SING] and) (925: 25-37.

Explain why this happens and construct similar examples.

7 Recurring decimals
1 = 0.142857
Ta
2 = 0.285714
7

3 = 0.428571
Starting with some observations on these and other particular cases, write an
account of recurring decimals.

8 Integral part
For positive integers 1, let

p(n) =[n+ vn +3]

where the square brackets represent the integral part (written INT in computer
language).

Which values does p() not take, and why?

9 Divisors of N
12 can be written as the product of two positive divisors (factors) in six different
ways:
WS hx 2x 6= 3% 4=4%3 =6%2=12x1
You can look at it in a slightly different way and say that 12 has six divisors, 1,
2, 3, 4, 6 and 12.

Define d(N) to be the number of positive integers dividing N. So

di2)-=6
What can you say about d(N) for different integers N? Justify any statements
you make.
296 PROBLEM SOLVING

10 Average number of divisors

Let d(N) be the number of divisors of N, defined above. Now define D(N), the
average number of divisors for numbers up to N, by

N
By relating D(N) to the number of lattice points on or under the graph of y = —,
find the limit of D(N) for large N. (A lattice point is a point with integer
coordinates.)

11 Railway layouts
A railway set contains a large number of pieces of rail, each in the form of a
quadrant of a circle. These can be joined together to make layouts:

The layouts can be described by ‘words’ consisting of R for right and L for left.
If the layout above starts at S and finishes at F its word is RLRR.

Some layouts are closed.

Start ———_p»

These are called circuits. The diagram above shows the circuit RRLRRRLR.

Given a word consisting of a large number of Rs and Ls, how would you tell
(without drawing the layout) whether or not it represented a circuit?
 MATHEMATICAL ENQUIRIES 297

12 Wrong deliveries
A postman has 4 letters to deliver, one for each of 4 flats. In how many ways can
he deliver them so that each is in a wrong box (and no two go into the same box)?
Try the same problem for 5 letters for 5 flats and explore the possibility of
obtaining a general result.

13 Garden path
A garden path is to be laid with rectangular slabs, each of width 1 unit and
length 2 units. If the length of the path is 7 units, in how many ways can this be
done? (Consider various widths of path.)

14 Cubic graphs
What ‘different’ graphs are possible for the cubic function
y=atbx+cx*
+ dx?
Explain how the shape is related to the values of the coefficients a, b, c, d.
Investigate possible shapes of graphs of polynomial functions of degree more than 3.

15 Pebbles
Fifteen pebbles are placed on the table; two players, in turn, take one, two or
three pebbles until all pebbles are taken. The player who then holds an odd
number of pebbles wins. Investigate winning strategies and extend the problem.

After working through section 4.1 you should: _

1 feel content about one a Sa ee, enquiry;

2 beready:toadopt:an‘investigative atcitude to” mathematics;

3 be familiar “« cS use a meaning of

theterms pele cases,
conjecture and generalisation;

4ee the pee ~ Bed pee a presentation;

5 : een aei. use ‘nd pe of the terms ee and

. _
tabulation;

6 know io to use atableof Dice,

_7 be ableto find the degreeof the ol generating a oo sens

ae ee of problemSolune which will be ef use in your later ver

298

Problem solving
2 Proof

4.2.1 Introduction
Proof has been described as the very essence of mathematics. It gives permanence
to a result or theorem. It is proof, or the nature of mathematical proof, that
distinguishes mathematics from most other fields of human endeavour. It is an
area where great creativity, elegance and beauty in mathematics can be seen.
Some of the great proofs, for example those made by the Greek mathematicians,
are still regarded as classics. Euclid’s proof of the Theorem of Pythagoras, or his
proof that there are infinitely many prime numbers, are proofs of great beauty
which demonstrate conciseness, creativity and elegance, and have not been
improved upon. There is simply no doubt about the theorems they prove — they
are permanent, fixed for all time.

A major reason for studying mathematics is to understand something of

mathematical proof and the process of mathematical thought. Proof provides a
framework within which you are able to present a reasoned argument logically,
with clarity and precision, whether within mathematics or more generally.

Proof is often such a complicated process that whole books can be written about
it. Disproof, however, is often a very short and simple matter, as the following
example shows.

Suppose that a friend, having tried many particular pentagons, is convinced that
no pentagon will tessellate. To refute this generalisation, all you have to do is
produce a single counter-example.

An excellent book, which explores many of the great theorems of mathematics,

is Journey Through Genius by William Dunham (John Wiley, 1990;
ISBN 0471 50030 5).
 PROOF 299

2 Exercise 1

1 Disprove, by finding a suitable counter-example, the statement

x > y >x>y

2 If two lines are perpendicular to a third line then they must be parallel to
each other — true or false?
Chet Ore teat ee 0 8 ee OF
4
All numbers of the form (4n + 1) may be expressed as a sum of two squares
— true or false?
4 Find the first four terms, ft, t), t; and t, for the sequence defined by
t, = (n—1)(n—2)(n—3)(n—4) +n
What might you expect t; to be? Is it?

4.2.2 Making a proof

In the previous section, you considered several conjectures which could be
disproved by finding counter-examples.

Simply spotting a pattern is not the end of a mathematical investigation. You

then need either to look for a convincing reason for the pattern or to find a
counter-example.

Can you think Can you find

of a reason? a counter-example?

Write out the reason Further investigation 1s

so as to convince someone else. needed so that you can
The written reasoning make another conjecture.
is called a proof. -

The real challenge of a mathematical investigation is to break out of the

potentially endless ‘no—no’ cycle of not being able to find either a convincing
reason or a counter-example!
300 PROBLEM SOLVING

Example 1 ;
Prove that Geese) = 1 a) as ee

Solution

adding ‘shells’ provides a geometrical approach. 2

The area of the shaded shell is:
Gees) (iene 6 . 2
= 27 sq. units

The three L-shaped shells have areas 8, 27 and 64 square units respectively.

The diagram demonstrates that

(2434 = 1 ae
You need to show that, in general, the kth shell has area k’. For this, use the
result

1+24+3+4---+(k-1) =!k(k-1)
For instance, as you can see in the top diagram, the fourth shell has an ‘inside
arm’ measurement of 1+2+3= 3 x 4x 3.

The kth shell looks like this when

divided into three parts.

k(k — 1)
 PROOF 301

Consider the areas A, B and C.

A=kxtk(k—1) =4k(k-1)
B= 1h (k—-1)
Cre:
So A+B+C=2k'(k-1) 43° (R-1)
+2?
=k(k-1)+k =k
The area of a square of side (1+ 2+---+ 7) is the sum of the areas of n
shells. So

(1 sett ert ated EE Pe

and the proof is complete.

The next example illustrates a very powerful and useful method of proof known
as proof by contradiction. It has been used by mathematicians for centuries and
is at the heart of many classic and beautiful proofs.

Example 2
Prove that the product of two integers whose sum is 1001 cannot be divided by
1001.

Solution
Let the two numbers be » and m.

You want to prove that nm is not divisible by 1001.

The essence of this proof is to suppose 1m is divisible by 1001 and to show

that this leads to a contradiction.

Suppose nm is divisible by 1001, i.e. 2m = 1001k (k being an integer)

nm = 1001k
= AL 13k

=> n (or m or both) has a factor of 7.

But n+ m= 1001 (given)

nam =7 x TE KAS
=>n-+m has a factor of 7
=> both n and m have a factor of 7
An identical argument applies to 11 and 13.
 302 PROBLEM SOLVING

So.n and m have factors of 7, 11 and 13.

=>n- 1001 (eon = 711 * 13)

and m-> 1001

=>n+m => 2002

So if nm is divisible by 1001, then n+ m must be greater than 1001.

But 2 + m = 1001 (given) and so the product nm cannot be divisible by 1001

ifn +m = 1001.

4,2 Exercise 2
1 (a) Explain why 27 is always even for all integer values of n.

(b) Write down a number, in terms of 1, which you know is odd. Explain
how you know.

(c) “The sum of three consecutive odd numbers is always divisible by 3.’
Check this conjecture by considering a number of cases. Prove that the
statement is true.

2 Take any two-digit number, reverse its digits and add to the original number.
For example:
34
pe
[eile
A convincing explanation that the result will always be divisible by 11 might
start by letting the two digits be a and b, so that the original number is
10a + b.
(a) Write down the value of the ‘reversed’ number.

(b) Find the sum of the numbers and show that the sum is always divisible
by 11.
(c) Discover if the rule applies to three- or four-digit numbers. Explain your
findings.
3 A neat party trick is the following rule for multiplication by 11, illustrated by
(tS 23530

Sy PeA
WYwY
+ +

3 S Ww a
 PROOF 303

For 11 x 1325 you can therefore proceed by:

++ 4
ENN EEN
Ue ile Bs ie 575
(a) Check that this method always seems to work.

(b) Using the method of expressing a two-digit number used in question 2

(i.e. ‘ab’= 10a + b), explain how this method for multiplying by 11 works.
(c) Explain carefully how the method works for 11 x 392.
4 What is wrong with the following demonstration that the area of an 8 x 8
square is the same as the area of a 5 x 13 rectangle?
5) 3

nn nn

Ww w n o2)

5 (a) Prove that the sum of two consecutive odd numbers is divisible by 4.
(b) Prove that the sum of 1 consecutive odd numbers is divisible by 1 if 1 is
odd, and by 2n if 1 is even.

6E Using the method of ‘proof by contradiction’, show that ,/2 is an irrational

. j ‘ a
number (i.e. it cannot be written as —, where a and b are whole numbers).
b

4.2.3 Fermat and proof

In question 3 of exercise 1 you may have discovered that 21 cannot be expressed
as the sum of two squares. The seventeenth-century amateur mathematician
Pierre de Fermat (1601-1665) studied a refinement of the conjecture that you
looked at in exercise 1.

All primes of the form (4n + 1) can be expressed as the sum of two squares.

Of his attempts to prove this conjecture, Fermat wrote (Diophantus, page 268):

‘... when I had to prove that every prime number of the form (4n + 1) is
made up of two squares, I found myself in a pretty fix.’
304 PROBLEM SOLVING
ee

Fermat was in the ‘no-no’ cycle. However,

‘But at last a certain reflection many times repeated gave me the necessary
light, and affirmative questions yielded to my method, with the aid of some
new principles by which sheer necessity compelled me to supplement tt.’

Another problem with which Fermat wrestled is known as ‘Fermat’s last

theorem’. The problem is to prove that there are no positive integers x, y, z such
that, for some integer m greater than 2,

Keep
m

eth. ays 2 2 2
There are certainly positive integers satisfying x +y =z.

soy
=A z= Sand =), po baz — AO
5 : at CO et ee
are well-known cases. Fermat’s theorem is thatx +y =z,x +y =Z,
and so on, cannot be solved for positive integers x, y and z.

Of this problem, Fermat wrote (Diophantus, page 145):

‘I have discovered a truly marvellous proof of this, which however the margin
is not large enough to contain.’

Although it was demonstrated for many values of m, the full result remained
unproved from Fermat’s time and was regarded as one of the great unsolved
problems of mathematics until, in 1993, it was announced that a complete (and
very long) proof had been made by Professor Andrew Wiles of Princeton
University, USA. The story does not end even here, however; a tiny but crucial
gap in the proof, unnoticed until after the announcement, took a further year’s
work to repair and the proof is now accepted as correct.

Other attempts on this problem have stimulated many important advances

in number theory. When you are in the ‘no—no’ cycle, the ideas you think of and
try may be far more important than any eventual solution to the problem!

You may feel that proof is not really a part of problem-solving as such. After all,
you may argue, the problem has really been solved when you reach the point of
‘knowing’, and proving is then no more than icing on the cake. The trouble is
that you often do not ‘know’ until a proof has been found.
 PROOF 305

The following problem illustrates this point.

How many regions are formed when n points on the circumference of a circle are
joined?

It can be tackled step by step as follows:

1 point 2 points
1 region 2 regions

3 points 4 points Ze
4 regions 8 regions

Sete
ein

1 Tabulate this information, guess a pattern and write a general formula

based on your conjecture.

2 Check your formula in the cases n = 5 and n = 6.

You must draw the diagram for 1 = 6 very carefully if you are not to mis-
count the number of regions. To obtain the greatest number of regions for
n = 6, you must not draw, for example, a regular hexagon.
EER
PHORESIS

3 At this stage, after a recount, your investigation will need to take a fresh
direction. Extend your table up to 7 points.

4 Make a table of differences.

5 Make a final conjecture.

4.2 Tasksheet E1 — Prime number formulas (page 557)

306 PROBLEM SOLVING
eee

A simplified outline for the process of solving a problem can be represented by a

loop:

Explore particular Look for

cases +——__—__> patterns

Generalise

ce Counter-example
Can a convincing P
reason be found? obtained

Present a report

Even when you have written up your findings, the investigation may not be over.

For example, suppose you had solved the problem

of finding how many squares of various sizes are
bounded by the lines in a 4 x 4 grid. You might
then go on to look at what is essentially another
particular case, say that of the 5 x 5 grid. This is a
fairly trivial form of extension of a problem, though
it might lead ultimately to a generalisation. On the
other hand you might ask yourself the question:
what would happen if a triangular grid had been
used?

.2B

1 How many different sizes of square may be seen in the 4 x 4 square grid?

2 What is the total number of squares?

 PROOF 307

3 How many squares would there be in an m x n square grid? You will need
the result for the sum of the squares of integers here, i.e.

do? =F (nt IQn+1)

fl

4 Find the total number of triangles in

the 4 x 4 x 4 triangular grid.

5 Find the total number of triangles in:

(a) a3 x 3.x Serid,

(b)' a5 <5 i S-orid:

6 Find a useful way of dividing the

triangles innxnxn grids into two
classes and hence conjecture the
number of triangles in a 6 x 6 x 6 4x 4x 4 triangular grid
grid.

The original problem of finding the number of squares on a 4 x-4 grid can be
extended in a variety of ways:

|
Shape

Save. ol Dimension
,
ERTS,

An extension into an analogous problem is more interesting when the new

situation, though related, is really different from the old one.
308 PROBLEM SOLVING

The square and cubical grids also have a one-dimensional analogue — five
equally spaced points in a straight line. So there is a chain of analogous
problems.
——_+— ++ ‘How many line segments:

How many
cubes?

How many squares?

There are 10 line segments, but to see the analogy clearly you can write this as
142 P34
Then the number of squares is
P42°4+3?+4
and the number of cubes is
P+P+P+4
 309

Problem solving
.3 Mathematical modelling

4.3.1 Modelling processes

Modelling is concerned with using pure mathematical techniques and processes
to solve ‘real’ problems. There are several important stages concerned with
deciding what mathematical techniques need to be applied and making sense of
the answers obtained. In attempting to solve ‘real’ problems, we go from the real
world into the theoretical world of mathematics and back again.

The process of modelling can be represented by the following modelling loop.

Mathematical
Probless spat Meret wet thereh 95,
4 problem
|
]
I
|
|

Real Mathematical
Interpret
solution solution

Set up model Define the real problem.

Formulate the mathematical problem. (This often involves
deciding what variables are relevant and what connecting
relationships can be assumed.)

Analyse Solve the mathematical problem.

Interpret Interpret the solution in real terms.

Validate Compare the solution with reality. If the solution does not
compare well with reality, then you must go around the loop
again, refining the model.
310 PROBLEM SOLVING

4.3.2 An extended example — solar eclipses

Partial and total eclipses of the Sun have long both terrified and fascinated
humanity. We now know that an eclipse occurs when the Earth, Moon and Sun
are in a straight line with the Moon between the Earth and the Sun.

During a total eclipse, all that is visible of the Sun is a bright ring (called the
corona) surrounding a black disc (the Moon). The diagram below shows
successive stages in a total eclipse of the Sun. The difference in apparent size of
the discs representing the Sun and Moon is exaggerated.

Stage I Stage Stage Ill

I
First contact Second contact Third contact Fourth
(beginning of (end of totality) contact
| totality) ' I
|
| Partial
| |
| | Partial |
l eclipse l Total eclipse l eclipse

The period of total eclipse is in stage . During stages 1 and m, a partial eclipse is
observed.
 MATHEMATICAL MODELLING 311
————
ar

The total eclipse is only seen on a narrow band of the Earth’s surface known as
the zone of totality or the umbra.

Umbra or Penumbra or \
shadow half shadow Ne (Not to scale)

To an observer standing in the penumbra, the Moon passes in front of the Sun
but never completely obscures it and only a partial eclipse is observed.

The last total eclipse visible from Britain was in 1927 and the next one is due in
1999!

Mathematics has a vital role to play in our attempts to explain and understand
events we observe in the world. Mathematical tools enable us to predict that
there will be a total eclipse in 1999, to determine its duration and to pick the best
observation point.

Consider the following problem.

How long does a total eclipse last?

This is not an easy problem. To obtain an estimate for the total time for an
eclipse you need to make a number of simplifying assumptions. These are
considered in the various models discussed below.
312 PROBLEM SOLVING

The main dimensions needed are (from Encyclopedia Britannica):

Diameters in km: Sun 139 < 10°
Earth 18 10:
Moon 3.48 x 10°
Mean distances from Earth in km: Sun 1.50°< 10°
Moon 3.82 x 10°

The Moon orbits the Earth in 27.3 days, a lunar month.

First model
In a simple model, you might consider the Sun and Earth to be stationary and
the Moon’s orbit around the Earth to be circular. You can define an eclipse as
total when, for some time, the Moon completely obscures the main disc of the
Sun from some observer O on Earth.

(None of the
diagrams are
to scale.)

A partial eclipse begins with the Moon in position M, and ends in position My.
The total eclipse is between positions M, and M3.

M, - first contact
M, — second contact
M, — third contact
M, - fourth contact

To make further progress, you can simplify your model by assuming that the
Moon’s orbit is locally straight.

P
 MATHEMATICAL MODELLING 313

You may also make the assumption that PQ is the Sun’s diameter and BC is part
of the Moon’s orbit.
P

: : ; OP
Triangle OPQ is an enlargement of triangle OBC with centre O, scale factor OB
and so
OB 3.82 x 10°
BG] PO 139 10°
EQ OP 1.50 x 108
= 3.54x 10°km
The diameter of the Moon is 3.48 x 10° km so, according to the model, total
eclipses do not occur! Since 3.48 < 3.54, the Moon never completely obscures the
Sun. So you must refine the model in some way.

Second model
One simplification you made was that the orbits of the Earth around the Sun and
of the Moon around the Earth are circular. In fact, both are roughly elliptical;
instead of using mean distances, you should consider a range of distances.

Moon Sun

Nearest distance from Earth (km) 3.63 x10? 1.47 x 10°

Farthest distance from Earth (km) 4.06 x 10° 1.5210"

The eclipse of maximum duration will occur when the Moon is nearest the Earth
and the Sun farthest away. Applying the model in this case,
3.63 x 10°
BC = 1,39 x 10° x ——_——_,, = 3.32 x 10°k
i 1.52 x 108
The eclipse is total only when BC falls within the diameter of the Moon, for a
distance of
(3.48 — 3.32) x 10° = 1.6 x 10° km
The Moon’s speed is
circumference of total circular orbit —2aX 3.82 x 10°
lunar month ~ 27.3 x 24 x 60
= 61.1km per minute
314 PROBLEM SOLVING
pn
eee ea ee eS Se ee

(The elliptical orbit has a length about equal to that of a circle with radius the
mean distance.)

Using the relation

: distance
tine = ————
speed
oat eee: kris
time of total eclipse = AL

= 2.6 minutes

The maximum duration of total eclipse is observed to be about 7 minutes. The

model can be amended further to account for this.

Third model
Another main assumption made at the outset was that the Earth could be
regarded as stationary. In fact the observer O is moving because of the spin of
the Earth on its axis. The speed of an observer is greatest on the equator, being
circumference of Earth eh TX 1228 10°
day length F 24 x 60
= 27.9km per minute
Since the observer and the Moon move in the same direction, the speed of the
Moon relative to the observer is
Moon’s speed in orbit — observer’s speed = 61.1 — 27.9
= 33.2km per minute
Using this figure, you will find that the time of total eclipse is
1.6
x 10°
= 4.8 minutes
33.2
You still have not reached the target and a dutiful modeller would check
assumptions and consider other refinements.

You should at least consider the following relevant observations.

e@ Over a series of calculations, a considerable rounding error accumulates.
e The distances between bodies are probably ‘centre-to-centre’; this is not always
made clear in reference books.
e The Moon’s speed in elliptical orbit is not constant.
e@ Orbits are not quite elliptical anyway! In particular, the Sun perturbs the orbit
of the Moon.
e In observations not made from the equator, you must take into account the
latitude in calculating the speed of the observer.
 MATHEMATICAL MODELLING 315

Since 1963, scientists have used specially equipped aircraft or observation

platforms above the clouds. Supersonic machines flying at 2000 km per hour in
the direction of the Moon’s shadow prolong the period of total eclipse to an
hour or more.

For the eclipse problem, the various stages in the modelling cycle can be
identified as follows.

Set up model
The stage of setting up a model is at the heart of the mathematical modelling
process. It is important to choose suitable simplifying assumptions which
preserve the key features of the real situation but which avoid an unnecessarily
complicated model. Essentially, there are two key stages in setting up a model.

@ Decide on the.relevant variables, keeping your list as short and simple as

possible.
e Look for relations connecting the variables on your list.

For the total eclipse problem, initial models were chosen in which the relevant
variables were the distances on the diagram below, together with the orbital
speed of the Moon and time of the eclipse.

Cera Observer

Moon

In this case, a relationship between the distances can be obtained from the
diagram and a further relationship is
distance = speed x time

Analyse
For the eclipse problem a further simplification was made, producing the similar
triangles shown below.

The mathematical analysis was then simply the use of similar triangles, followed
later by a use of the distance—speed relationship.
316 PROBLEM SOLVING

Interpret
When modelling ‘real’ problems it is important to remember that ‘real’
answers are required. If you are trying to estimate the annual cost of running
the family car, then an answer such as £4000 is required and not an algebraic
formula.
An analysis of the initial eclipse model had a very simple interpretation — total
eclipses cannot occur!

Validate
It will be possible to check the solutions to many problems by direct comparison
with reality.
You know that total eclipses do occur and so the initial model needed refining.
Even after two successive refinements the eclipse model still predicted a much
shorter eclipse than can actually occur.

The acid test of a model is always how well it represents reality but, generally,
over-elaboration should be avoided.

Normally a model will need improvement and you will have to go around the
loop in the flow chart a second or even a third time. At some point, however, the
model will be judged to be good enough and you stop.

Mathematical modelling occurs when solving pure problems as well as ‘real’

ones.

Example 1
The product of the ages of Ann and Mike is 300. Find their ages if Ann is 20
years older than Mike.

Solution
Let Mike be m years old; then Set up model Appropriate notation is
Ann is m + 20 years old. introduced for the variables.

m(m + 20) = 300 Relationships are listed.

=> m + 20m — 300 = 0 Analyse The standard method of solving

=> (m+ 30)(m — 10) = a quadratic equation by factorisation is
=> m = —30 or 10 applied.

m cannot be —30 and so Interpret/validate The two mathematical

i= 10: solutions yield only one ‘real’ solution.
Mike is 10 and Ann is 30. This solution is then stated in everday
language.
 MATHEMATICAL MODELLING 317

You will have plenty of opportunity to use your modelling skills in your work in
mechanics or statistics. a
 318

Problem solving
.4 Mathematical articles

4.4.1 Introduction
The purpose of this section is to help you both to read and to write mathematics.
It contains some pure and some ‘real’ case studies. These provide models for
writing mathematics and give an idea of what is expected when you write up
your own investigations.

In reading the best mathematical writing a considerable effort of self-discipline is

needed to keep going through to the end. (The same applies to the worst writing,
but for different reasons!) In a good article, ideas will arise which suggest
investigations not followed in the text and you may find it difficult to keep your
mind on the job in hand. You can comfort yourself with the thought that side-
lines can always be followed up later. Mathematical articles cannot be read as
you would read a novel; the information content is usually very dense and you
should take time to check calculations, verify assertions made and understand
thoroughly the definitions and the argument as it develops.

The two case studies below are interspersed with questions of the kind you
should be asking yourself when you read a piece of mathematics. The questions
should help you to develop your understanding and to check that the writer’s
argument is correct.

Case Studies
1 — The Platonic solids (page 321)
2 — The gravity model in geography (page 325)

Case studies 1 and 2 include questions interspersing the text at various stages. In
the remaining case studies the exposition will be uninterrupted, all questions
being set at the end. To help to bridge the gap between the two styles, here is an
exercise based upon the beginning of an article published in Mathematical
Spectrum (volume 19, number 1 (1986/87)).

—_—_—_—_—_—
SSS SSS

4.4 Exercise 1
Read the following article carefully and carry out the necessary thinking and
writing at all the points indicated by numbered asterisks.
 MATHEMATICAL ARTICLES 319

On doodles and 4-regular graphs

J. C. Turner, University of Waikato
The next time you are doodling ona piece of paper, let your pen come back to its
starting point and then consider what kind of mathematical figure you have drawn.

If you have always let your pen trace a smooth continuous curve, and whenever
you have met a previous point on the curve you have crossed directly over the
old curve, it is probable that you have drawn what graph-theorists call a
4-regular graph: figure 1 is an example.*! In this case n, the number of crossing
points = 9, m, the number of edges (an edge is a curve joining two adjacent
crossing points) = 18 and 7, the number of regions (one is shown shaded,
34589) = 11, including the region of the plane exterior to the graph.” Such
diagrams are called 4-regular graphs because every crossing point (these are
usually called vertices, or nodes) has four edges adjoined to it.

if

You might like to think what you would have to do whilst doodling if the result
were not to be a 4-regular graph,”* and conversely, whether there are 4-regular
graphs which are not doodles!** You might like to think of how many small
regions (ones with few edges) you have managed to draw in your doodle.

Let us say a region is a k-gon if it is bounded by k edges. Then the smallest

region possible is a 1-gon or loop; then a 2-gon; then a 3-gon; and so on. In the
doodle shown in figure 1 there are four 2-gons and three 3-gons. There are also
two 4-gons, one 5-gon, and the outer-region 6-gon”™.

The purpose of this article is to discuss the possibility of bounds on the

frequencies of small regions in 4-regular graphs. Simple questions to ask are:
(i) Can one draw a doodle without a 2-gon?
(ii) Can one draw a doodle without a 3-gon?
(iii) Can one draw a doodle without either a 2-gon or a 3-gon?
320 PROBLEM SOLVING

Further Case Studies

3 — The game of Hex (page 331)
4 — Archimedes and 7 (page 333)
The Platonic solids

For n = 3, 4, 5,... there is precisely one shape of polygon with 1 equal sides and
n equal angles. We say that there are infinitely many regular polygons.

1 What do you understand by the term ‘regular polygon’?

What is meant by ‘infinitely many’ regular polygons?
STUDY
CASE
1
A regular polygon is a two-dimensional shape. The analogous
three-dimensional shape is a solid whose faces are congruent
regular polygons with the same number of these polygons meeting
at each vertex (or corner) of the solid.

Such solids were considered in Plato’s Timaeus and became known as the
Platonic solids.

2 Name a three-dimensional shape which fits the description of a Platonic solid.

Given that there are infinitely many regular polygons, you might expect there
to be infinitely many Platonic solids. We shall investigate whether or not this is
the case.

To try to classify the Platonic solids systematically, we can start with solids with
the ‘simplest’ possible faces: equilateral triangles.

Case 1 — Equilateral triangular faces

To form a corner, at least three faces meet at any vertex. For triangular faces,
at most five such faces meet at a vertex because if the angles at a vertex totalled
360° or more, then the faces could not form a corner.

3 Explain fully why six triangular faces cannot meet at a vertex.

We need to consider the cases of three, four or five faces

meeting at each vertex. As an example, let us examine the
second possibility.

Four faces meet at each vertex

It is easy to form the required solid by taping together
equilateral triangles, four at each vertex. The shape
obtained is known as the regular octahedron and has eight
faces.
From a model of an octahedron we can verify the figures in the following
table:

Number of faces
Number of vertices
Number of edges

4 What are the values of F, V and E for a cube?

The mathematician Euler found that the equation F + V = E +2 is true for all
polyhedra.

5 Find F+ V —E for: (a) the octahedron (b) the cube.

Remarkably, we could have used Euler’s formula to obtain the number of faces,
vertices and edges of the octahedron without ever making the model!

We are looking for a shape formed by F triangular faces.

These have 3F sides and 3F corners. When they are taped together, each edge of
the solid is formed from a pair of these sides and so

Each vertex of the solid is formed from four corners and so

3F
V=—4
Substituting these results into Euler’s formula, we obtain

pu chs 3h
Aen)
=> 4F4+3F=6F+8

=> |r 49}

Then E = 12 and V = 6, as required.

This kind of analysis can be very helpful in showing precisely what possibilities
there are for particular solids. By a similar analysis, we can show that the only
Platonic solids with three or five triangular faces at a vertex are the tetrahedron
and the icosahedron.

CASE
STUD
1

Tetrahedron
Icosahedron

6 Use Euler’s formula to show that a Platonic solid with three triangular faces at
each vertex must have precisely four faces.

Case 2 — Square faces

In this case three faces must meet at each vertex.

4F 4F
7 Explain the assertion above. Explain also why E + SF and V= Za

; 4F 4F
Substituting E = os and V = — into Euler’s formula gives

4F 4F
gee)
ea
from which we obtain F = 6, V = 8 and E = 12. So the solid must be a cube.

Case 3 — Faces with five or more edges

Three hexagons or four pentagons at a vertex would give an angle of 360° or
greater — impossible for a three-dimensional solid. So the only remaining case is
that of three regular pentagons meeting at each vertex, because 3 x 108° < 360°.

8 What is the significance of the 108° mentioned above?

A similar analysis to that carried out in cases 1 and 2 shows that a Platonic solid
with pentagonal faces must have F = 12, V = 20, E = 30.
 Dodecahedron

9 Obtain the given values of F, V and E.

It is interesting that there should be infinitely many two-dimensional regular

shapes yet only five regular three-dimensional shapes, the five Platonic solids:
tetrahedron, octahedron, icosahedron, cube and dodecahedron. Mathematicians
have shown that there are six regular four-dimensional ‘solids’ and, curiously,
only three in any space of more than four dimensions — the n-dimensional cube,
tetrahedron and octahedron.
The gravity model in geography
Newton’s law of universal gravitation states that two bodies with masses m, and
my attract each other with a force
Gm\m)
da

where d is the distance between the centres of mass of the bodies and G is
constant. This equation has been used successfully in modelling planetary DD
6)
saree
t)
Oy.
Gy
motion, by Newton and others.

The term ‘gravity model’ is applied in geography in a looser way. Some authors
use it of any model where the intensity of an interaction between places
decreases with distance. The interaction could be one of trade or transportation
or migration, for instance. Others insist upon a stricter analogy with Newton’s
model in which, if the two places had measures of attractiveness W, and W, and
were separated by a distance d, then if T were some measure of the intensity of
interaction between the places,
_ kW1W
=
Si

where k is a constant and ( a positive number.

1 What value does 3 have in Newton’s model of gravitation?

Telephone calls
Start with an example using a model closely resembling Newton’s. The
populations and distances separating towns are given below.

Population Distances (km) from

Town (in thousands) Sheffield Derby

Sheffield : 537 - 52
Derby 216 52 ~
Nottingham Z7A 50 20

Suppose that the number of telephone calls made in a given time between town
A and town B can be found using the equation
kP Pp
DS
dp
where k& is a constant, P, and Px, are the populations of towns A and B
respectively and dap is the distance between them.
 Using this model, the number of telephone calls between Sheffield and Derby is
k x 537 x 216 _ 43k
Tsp =
Se a
If the model were to be used to predict numbers of calls between towns, it would
first be tested with known data. This would enable the ‘calibration’ of the
model, in which the value of k would be found. At the same time, a check would
be made to find whether the inverse square model was most appropriate or
same
We)
DD
Oya
LO)
Gy whether some other exponent should be substituted for 2 in the formula for the
number of telephone calls.

2 (a) Use the model to find Tpy and Tsn.

(b) Which pair of the towns of Sheffield, Derby and Nottingham has the most
telephone calls between them according to the model?

Two competing shopping centres

If, in a rural area, there is a good network of roads and two towns satisfy the
shopping needs of the scattered community, then the application of a simple
gravity model may be considered.

Suppose towns A and B have populations P, and Px respectively. An individual

lives at C which is distance dy km from A and dg km from B, as shown in the
diagram.
C

The gravity model suggests that the attractiveness of A to the individual is of the
form

dy
3 What is the attractiveness of B to the individual?

Now you can introduce the idea of an area of influence. If A is more attractive
than B, then the individual is in A’s area of influence, and vice versa.
First you should investigate this idea in a simple theoretical case. Suppose that A
and B have populations of 40000 and 10000 respectively. Then for a point P, the
40 000k 10 000k
attractiveness of A is pe and that of B is ape

If A and B are equally attractive then

40 000k =" 10 000k
Ap Bee
AP? = 4BP* STUDY
CASE
2
AP =2BP

7)
ees=- (a iam: = laa
A P, B P,

If P, divides AB in the ratio 2: 1, then P, will be on the locus of points where A

and B are equally attractive. Similarly, the point P, dividing AB externally in the
ratio 2: 1 will be on this locus. °

If AB = 3a then it is not difficult to verify that

AP, = 6a, AP, = 2a, P,P, = 4a

Now let AB be the x-axis and take as origin O the midpoint of PP).

Let P (x,y) be any point on the locus of equal attractions. You know that
AP = 2BP.

=> AP? = 4BP?

From the diagram,
AM? + PM? = 4(BM? + PM’)
(x + AO)? + y* = 4(x + BO)? + 4y”
(x + 4a)? + y? = 4(x +a)" + 4y”
3x* + 3y* = 12a*
x ay aS ee
 This last equation will be recognised as that of the circle with P,P, as diameter.
All points inside the circle are in B’s area of influence. Points outside the circle
are in A’s area of influence.

STUDY
CASE
2

You can now apply the theory to an actual case.

The map on the next page shows the area around. Wisbech and King’s Lynn,
whose populations are respectively 17000 and 30000. You may assume that they
are the only shopping attractions in the area.

To find the areas of shopping influence of Wisbech and King’s Lynn, consider
someone for whom they are equally attractive, who lives distance dy from
Wisbech and distance d; from King’s Lynn.
k x 17000 oe k x 30000
dy dt
So 17 000 dr = 30000 dy
You can take square roots to obtain 130d, = 173dy.

People for whom this equation holds live on a circle whose centre lies
somewhere along the line through Wisbech and King’s Lynn. Two points on the
circle’s circumference, X and Y, are marked on the diagram below.
 - THE WASH

°
CASE
STUDY
2
Sandringham

S Great @
Massingham

If the distance from Wisbech to King’s Lynn is 13 miles, then in mile units
a = 0.043 and the point X is 5.6 miles from Wisbech and 7.4 miles from King’s
Lynn.

4 Explain the calculation that leads to a = 0.043.

To locate the point Y, find the distance x as follows:

al wo medLIfe
a ee
43% = 130. x13

x = 39.3
Therefore the point Y is 39.3 miles beyond Wisbech and the diameter of the
circle is 44.9 miles.

5 (a) Explain the statement

x+13 173
xg) 130
(b) Shrewsbury and Welshpool are 19 miles apart and have populations of
56000 and 7000 respectively. Using the same gravity model, write down
expressions for the attractiveness of the two towns to an individual who
lives distance d; from Shrewsbury and dy from Welshpool.

(c) Using the method outlined above, carry out the necessary calculations and
describe as fully as possible Welshpool’s area of influence.

(d) In areas of the USA, it is easy to travel long distances by car. What
difference would this make to the value of 3 taken in the model?
The game of Hex
The game of Hex was invented in 1942 by Piet Hein at Niels Bohr’s Institute for
Theoretical Physics in Copenhagen. Blockbusters, a popular television quiz, used
a variation of this game.

The original game, for two players, is played on a diamond-shaped board made
up of hexagons. One player has a set of white counters, the other a set of black STUDY
CASE
3
counters. The players, White and Black, alternately place one of their counters
on any vacant hexagon. The object of the game is for White to try to complete a
chain of white counters between the two edges marked ‘White’, while Black tries
to form a chain of black counters joining the ‘Black’ edges. The game ends as
soon as one player completes such an unbroken chain.

Here is a diagram showing a game won by Black on a 4 x 4 board.

The game went as follows:

White Black

3 10
7 15
if 14
16 12
ys) $3

Black wins because of the chain 13—14—15-—12.

In this game, Black’s first two moves, at 10 and 15, illustrate a useful tactic when
playing Hex.

Result 1
Black cannot be prevented from connecting 10 and 15 because if White plays on
11 Black can play on 14, whereas if White plays on 14 Black can play on 11.
Therefore, Black can always complete the chain 10—11/14-15.
 332 PROBLEM SOLVING

When playing the game, it is also helpful to know another type of connecting
result.

Result 2 SS
Black, to play, cannot prevent White
connecting her counter on 1 to the White edge. SS
For example, if Black plays on 2 or 4, White

3
STUDY
CASE 36)
can simply play on 3 and then form the chain

White

On small boards, the first player to move has such a strong advantage that she
should always win. For example, White to play first on a 4 x 4 board will win
easily, especially if she knows the following result, which is easy to prove with
the help of result 2.

Result 3
An opening play by White on
4,7, 10 or 13 wins. Any other
opening play loses.

On larger boards it is much more difficult to make such a full analysis of the
game and in practice an 11 x 11 board is used for serious games.

Questions
1 For result 2, show how White can connect 1 to the white edge when Black’s
initial move is 3, 6 or 7.

After the sample game of the article, White claimed that her third move, 11,
was the losing one. She believed that she could have won the game by playing
another move instead. Sketch the position immediately before her third move,
and prove or disprove her claim.

For result 3, prove that White can force a win by playing at either 7 or 10.
Archimedes and 7
m may be defined as the ratio of the circumference of a circle to its diameter.
circumference
T=
diameter
The aim is to estimate the value of 7 by finding sequences of converging upper
and lower bounds for the circumference of a circle of radius 1 unit. This method
was used by the Greek mathematician Archimedes. STUD
CASE
4
P

Q R

In the diagram, perimeter of — circumference _ perimeter of

< <<
‘triangle ABC ~ of the circle triangle PQR
The two triangles are equilateral. The circle has radius 1 unit, so its
circumference is 27 units. Let O be the centre of the circle.

As AC is a chord of the circle, the line OD bisects it at

right angles and so
3
DG@e7cos 50. — “

The perimeter of the triangle ABC is then 3/3.

For triangle PQR, the sides PQ, QR and PR

form tangents to the circle and so, for example,
line OS bisects PR at right angles.
il
Mier Shs ene = ,/3 and the perimeter of
triangle PQR is 6,/3.

So the first upper and lower bounds for 7 are given by aa

 The figure shows a regular hexagon ABCDEF,
inscribed within the circle, and a regular
hexagon PQRSTU, circumscribed around the
circle.

STUDY
CASE
4

You require the lengths of the perimeters of both these

hexagons.

Since the circle has radius 1, you know that both OM and
AB are 1 unit long, so the perimeter of the inscribed
hexagon is 6.

The circumscribed hexagon is an enlargement of the

OM
inscribed hexagon by scale factor ——. By Pythagoras’ theorem,
ANG NewS Loge
O= 1-—[{ = = — and so the scale factor of the enlargement is —-.
2 2 V3
12
It follows that the perimeter of the circumscribed hexagon is Fy ot 4,/3, and so
ieee 3: v

To bring the upper and lower bounds for 7 still closer together you need to
consider polygons with more than 6 sides.

A polygon with n sides is called an ‘n-gon’. You can demonstrate that it is

possible to calculate the length of the side of a regular 2”-gon inscribed in the
circle from the length of the side of the regular inscribed n-gon.

You can use an enlargement argument, similar to that for the hexagon case
above, to find the perimeters of the circumscribing n-gon and 2n-gon, so you can
extend your sequences of converging upper and lower bounds for 7.
Let x, be the length of the side of a regular inscribed n-gon,
I,, be the perimeter of a regular inscribed n-gon, I,, = nx,,,
C,, be the perimeter of a regular circumscribed n-gon.

An enlargement argument, similar to that used in the hexagon case, gives

I,
oe
ra
4 STUD
CASE
4
You can obtain a regular inscribed 2n-gon from the regular inscribed n-gon by
drawing the line of symmetry of each triangle OAB, OBC, and so on, to obtain
triangles OAM, OMB, OBN, ONC, and so on, as shown in the diagram below.

AM? = AL? + LM? (Pythagoras’ theorem)

2
(TOD

2
Xn
But OL? = OA? — AL?’ = 1— a

x 2 x 2 D

So AM? = 1 (mee, (ee

al ang 5
eZ 2 2

Thatis cn =) (—24E
FR |4 = wee4
x x.
at 1S Xp,

These equations make it possible to find successive approximations for 7 as

shown in the table.

ies OF Inequality
3/3
By ous 64/3 BS cn <3y3
6 ARO a3 an 2/8
Questions
1 Explain what is meant by ‘converging upper and lower bounds’.

2 Give a detailed explanation of why (in the diagrams on page 333)

1
@) DC =scos30 (b) SR= “an 30"

Show how the two results from question 2 lead to the inequality
3/3
ye ats Av2

Explain the meaning of the terms (a) inscribed, (b) circumscribed.

Justify the statement ‘you know both OM and AB are 1 unit long, so the
perimeter of the inscribed hexagon is 6’.

(a) Give the details of the calculations leading to the value of 4,/3 for the
perimeter of the circumscribed hexagon.
(b) Show how this results in the inequality 3 < @ < 2,/3.

Give the details of the argument that

I,

(a) Write down the formula for x1) in terms of x¢.

(b) Hence find x1).

Showing all details of the working, complete the row of the table for n = 12.

Explain how continuing the table would lead to a value of z of any desired
accuracy, provided that sufficiently accurate computational equipment is used.
 337

Mathematical methods
.1 The power of Pythagoras
5.1.1 Pythagoras and right-angled triangles
The great mathematician Pythagoras is best remembered for stating a
relationship (now known as Pythagoras’ theorem, although it was certainly not
his only theorem) connecting the lengths of the sides of right-angled triangles.

Little is known for certain about the Greek mathematician Pythagoras. Born on
the island of Samos, just off the coast of Asia Minor, around 570Bc, he travelled
much throughout his life, notably in Egypt, Asia Minor and Italy. He eventually
established a community of scholars at Croton, a Greek settlement in what is
now southern Italy. They formed a select
brotherhood sharing philosophical and political
ideals as well as their study of mathematics,
and lived under a strict code of discipline —
temperance, purity and obedience were their
watchwords. They were a secret society, and
the pentagram was used as a sign of
recognition. Although women were forbidden
by law from attending public meetings, many
went to Pythagoras’ lectures, and a few were
even admitted to the inner circle. The
Pythagoreans believed that mathematics was The sign of the Pythagoreans
the key to the world. Everything was assigned
a number. For example, 1 represented reason,
2 man, 3 woman, 4 justice and 5 marriage (the
union of 2:and 3).
338 MATHEMATICAL METHODS
een
eee

Pythagoras’ influence continued after his death and his followers developed a
custom of assigning all work to ‘the Master’ (Pythagoras). This has made it very
difficult to know how much of the work assigned to him was his own, and how
much his students.

It is believed that the theorem for which Pythagoras is best remembered

genuinely is his own work. It can be stated as

The square on the hypotenuse of a right-angled triangle is equal to the sum of

the squares on the other two sides.

Pythagoras’ theorem has countless practical applications.

Pythagoras and his students searched for sets of whole numbers which satisfied
the relationship a +b? =’. The smallest such values are 3, 4 and 5, since
9+ 16 = 25, ie. 3? +47 = 5’,

Sets of numbers like this are known as Pythagorean triples. Since they satisfy the
relationship, any triangle with these measurements must be right-angled. This
fact is still used for marking out sportsfields: a rope marked in the correct
proportions and arranged in a triangle will form a right-angle.

(QM gg L LY
a™

3m 5m

Malas

Since any triangle with these proportions must be right-angled, it follows that
(6, 8, 10), (9, 12, 15) and (12, 16, 20) must also be Pythagorean triples.

The Pythagoreans could see that the hypotenuse of an isosceles right-angled

triangle with two equals sides of length 1 unit would not be a whole number.

Using h* = a’ + b’, the hypotenuse must be nae

 THE POWER OF PYTHAGORAS 339

You will find this triangle useful for remembering the values of the circular
functions at 45°.
il
sin 45° = = cos 45° = 7) tan 45° = 1

Similarly, if an equilateral triangle is divided along a line of symmetry, _

two right-angled triangles are formed. Using half of an equilateral triangle of side
2 units, the values of sin 30°, cos 30°, tan 30° and sin 60°, cos 60°, tan 60° can all
be obtained in exact form.

By Pythagoras’ theorem BD* = 4 — 1 = 3

ie. BD = 3
You can then read off the required
trigonometric ratios, for example,
1 -
SiO — 77 608 50) a

It is very useful to know the ete.

trigonometric ratios.

Pythagoras’ theorem can easily be extended into three dimensions.

>*
et
Cf
340 MATHEMATICAL METHODS

Example 1

(a) A frame is to be constructed from thirteen metal rods. Twelve of the rods
are welded together to form a cuboid. The thirteenth is fitted as a crosspiece
between opposite corners, thus increasing the rigidity of the structure.

If the cuboid is made from rods measuring

18 cm, 24cm and 11cm, what is the length of
the crosspiece?

(b) The frames are made to a variety of sizes. If the rods used to form the
cuboid measure x cm, ycm, and zcm, find an expression for the length of the
crosspiece.

Solution
(a) . F The diagonal AG is the hypotenuse of the right-
angled triangle AGC
Ax ie. AG? = AC? + CG?
a G AC is the hypotenuse of the right-angled
24 aes 11. ~«triangle ADC

C5 ie ACS ADA Ope

=A Cre)

and AG = ,/(900 + 121) = 31.95 cm

(b) Generalising this result, if AD =x, CD=y, CG=z
then AC* = x” +. y* F
Sop AG AC =CG"
(x7 + y°) +27 A
\H__ S>\6
Zz

= AG = / (9° +" +7) mea

 THE POWER OF PYTHAGORAS 341
a

S.1 Exercise 1

1 Use Pythagoras’ theorem for right-angled triangles to find the hypotenuse of

(a). | (b) 4.9cm

10cm 16.8cm

24cm

2 Two pigeons fly off Nelson’s Column. The first flies 3.7 km due east and
lands on the Tower of London. The other flies 0.9km due south and lands
on Westminster Abbey. How far apart are Westminster Abbey and the
Tower of London?

3 A ramp is to be built to allow wheelchairs and pushchairs to enter a hotel

more easily. The height of the step is 10cm. The planks to be used are 0.7m
long. How much space will be needed in front of the step?

4 A flag-pole is supported half-way up by four guy ropes, each of length 12 m.

The ropes are tethered at the four corners of a rectangle measuring 5m by
8m. How tall is the flag-pole?

5 -Find eight Pythagorean triples which use only numbers less than 30.
Group them to show which triples represent triangles with the same
proportions.
342 MATHEMATICAL METHODS

5.1.2 The equation of a circle

Pythagoras’ theorem can be used to find the equation of a circle. The necessary
ideas are developed in the following questions.

1A

1 (a) Use Pythagoras’ theorem to find the distance of the point (3,3) from
the origin.
(b) Use Pythagoras’ theorem to find the distance of the point (4,6) from
the point (1,2).

A circle is made up of all the points that are a fixed distance from its centre.
The distance from the centre is the radius r of the circle.

2 (a) Use Pythagoras’ theorem to decide which of the following points lie on
a circle with centre (0,0), radius 5.

(4x3), (2255255), (354) eon) (eS)

(b) Write down an equation connecting x and y which is satisfied by all
points (x,y) which lie on this circle.

(c)" Use a graph plotter to draw the graph of this equation.

“Note that for some graph plotters, you may need to rearrange the equation
to give y in terms of x. This form of the equation will involve a square root,
so the graph will be drawn in two sections, one part using the positive square
root, and one using the negative square root.

3 (a) Use Pythagoras’ theorem to decide which of the following points lie on
a circle with centre (2,5), radius 25.

(27,5), (17,25), (—5,29), (—22,—2), (—18, —10)

(b) Write down an equation connecting x and y which is satisfied by all
points (x,y) which lie on this circle.

(c) Use a graph plotter to draw the graph of this equation.

4 Use Pythagoras’ theorem to prove that the equation of a circle

(a) centre (0,0) and radius 7 is x* + y=r

(b) centre (a,b) and radius r is (x — a) +(y- b)* ae
 THE POWER OF PYTHAGORAS 343
eee

Example 2
Find the equation of the circle radius 4, centre (2,3). Does the point (5, 5) lie
inside or outside the circle?

Solution
The equation of the circle is (5, 5)
(x — 2)? + (y — 3)? = 16.

The distance between the points (2, 3) 2

and (5,5) is+/(3" + 2”) = /13.
DS 3
! Since 1/13 is less than the radius of the Bey
circle, (5,5) must lie inside the circle.

: Example3
Find the radius and coordinates of the centre of the circle having equation
x 2x4
4+4y41=0

Solution
You need to write the equation in the form (x — a)? + (y—b)’ =r’. The
method of completing the square is used to do this.
(x1)? -1+y'+4y+1=0 Complete the square on the
AteTIAGs
(x — 1)? + (y+2)?-4=0 ...and
on the y terms
ives (x1)? +(y+2)? =4
The circle has radius 2 and centre (1, —2).

Again, we can easily extend these ideas into three dimensions and obtain the
: equation for a sphere.
344 MATHEMATICAL METHODS

@
1 A point in space can be represented by the use of three-dimensional
coordinates (x,y,z). Write down the length of OP if P is:

(a) (4,5, 6)
(b) (3, 0,4) : a!)
(c) (-3,0,4)
(d) (—2, —1, 3) O ;

2 If P is the point (3, 4,5) and Q is (6,6, 9), write down the difference
between their: é
i
3 7
(a) x-coordinates, Z c eT
7

(b) y-coordinates, P

(c) z-coordinates.

Hence find the distance PQ.

3 Find the distance PQ in each of the following cases.

(a)eR(S31529) QOx(85,2, 3)

(b) P (—1, 3, 0) Q (—5, —8, 1)

(c) P (x,y,z) — Q (a,b, ¢)

4 (a) P is a point on a sphere of radius r,

centre O, with coordinates (x, y, z).
Write down the length OP in terms of
x,y and z and hence write down the
equation of a sphere (i.e. find a
relationship between x, y,z and 7).

& Generalise the method of part (a) to

write down the equation of a sphere
of radius r whose centre is at the point
with coordinates (a, b,c).
 THE POWER OF PYTHAGORAS 345

an . yo Exercise 2

1 Find the equations of the following circles with centre the origin:
(a) radius 15 units (b) diameter 8 units
(c) circumference 10 units (d) passing through the point (12, 16)

Find the equations of the following circles:

(a) radius 3 units, centre (1, 1)

(b) diameter 16 units, centre (—4, 6)

A lighthouse has a grid reference (20,85) on a map, where each unit

represents one nautical mile. The light is powerful enough to be seen from up
to 18 nautical miles away. Write an equation involving E, the easterly map
reference, and N, the northerly map reference, to show the boundary beyond
which ships cannot see the lighthouse.
How might this equation be misleading?

Find the equation of the sphere, centre (2,3, 1), radius 4 units.

Do the following points lie inside or outside the figures given by the
equations?

A324), =i +4) 9
DING 1)) © Geet)
EG =2)? = 30
e) (=1,3,5); ~~ (@=—1) ++ 1% + (@—3)? = 24
dy*(— 1,355); C=) SiGe i) (e383) = 25
Find the radius and the coordinates of the centre for the circles having
equation:

(a) x —4x+y —2y+1=0

(b) x = 2x +y —8=0
(c) 4x7 —4x + 4y° +2y+1=0
346 MATHEMATICAL METHODS
Se eee

7E Two aerial fireworks are timed to go off at the same time. Relative to an
observer on the ground, their centres are at (120, 150,30) and (160, 180, 40).
The radius of the first firework extends to 20 units, that of the second to
30 units. Assuming that both fireworks make a spherical pattern in the sky,
will the patterns intersect?
8E Find the centre, radius and equation of the circle passing through the points
Gioia 8\rand (4, 15),
[Hint: You could let the equation be (x — a)’ + (y — 6)” = 7, or plot the
points and consider a geometric approach.]
9E Find the possible equations of circles, radius 10 units, which pass through the
points (10,9) and (8, —5). Which of these equations describes a circle which
also passes through the point (—6, —3)?

5.1.3 Trigonometric identities

Points on a circle of unit radius, centre the origin, e (cos @, sin 6)
have coordinates given by (cos 6, sin 8) where @ is the
angle measured anticlockwise from the x-axis.

The diagram can be used to illustrate two useful

identities.

By Pythagoras’ theorem,
x+y’ =1, where (x,y) are the coordinates of P on the circle.
Since x = cos@ and y = sin@, then cos” 6 + sin* @ = 1

sin 8
Also, as tan? = a then tan@ =
6 cos 0
Both of these results are true for any value of 6. They are trigonometric identities.

Example 4
In the triangle shown above, if sin@ = is find: (a) cos6 (b) tan 6

Solution

(a) Since cos” @ = 1 — sin’0, cos0 = VC -5) = aN

3

(b) Using tan 6 = pine. ian

1 _ 2y/ 1
suze.
c os 6” Sy 3S
oe Se
 THE POWER OF PYTHAGORAS 347

SN

You can make use of the identities above to solve some trigonometric equations.

Example 5
Solve the equation 2sin? x = 3cosx for 0 < x < 2rn.

Solution
2sin* x= 3cosx
=> 2(1- cos” do) = S1COSx replace sin” x with 1 — cos” x
=> 2—2cos’ x= 3cosx
=> 2cos’x +3cosx —2=0 rearrange and note that this is a
quadratic equation of the form
2c? + 3c —2 = 0 with c = cosx
= (2cosx — 1)(cosx +2) =0 factorising to give
Cele 42) =0
= COS — ; OF -<CosS.Y —— 2 cosx = —2 gives no solutions

T Sm |
ier or oe in the range 0 < x < 27

The solutions can be illustrated by sketch graphs.

a OA “ F R tu)
es and — are sensible solutions to the equation 2 sin” x = 3 cos x.
 348 MATHEMATICAL METHODS

Example 6
Solve the equation 3 sin@ = 4cos 6 for 0° < 6 < 360°.

Solution
As cos @ = 0 is not a possible solution, you can divide through by cos @.
3sinO _
4 (dividing both sides by cos 6)
cosO
Sie,
=Ss tang —4 («sing—— = tan 6)
cos 0
= tang — n= 1.33

=> = SB oe D3. i1°

5.1 Exercise 3
1 (a) By replacing sin?x by 1 — cos” x, show that the equation
1+ cosx =3sin* x is equivalent to 3cos’x + cosx —2=0.
(b) By writing c = cosx, factorise the left-hand side of this equation.
(c) Solve the equation to find all values of x between 0° and 360°.

2 Solve the following equations for 0° < 6 < 360°.

(a) 3sin@ =2cosé (b) 0.5 sin? = 0.8 cos 0
(ec) “S:sin 20 = 7 cos 26

3 Solve the following equations for 0 < 6 < 2r.

(a) 2cos*@=cos6+1 (b) sin@d— /3tos6 =O
(c) 8sin?6 =7 —2cos8
4 Solve the following equations for 0 < x < 27:
(a) sin? x = 0.25 (b) cos? x — sin? x = 1
(c) cos’x —4sin*x =0 (d) 3cos*x =2+2sinx

SE A cyclist, C, cycles around a circular track, centre O and

of radius 100 m. A photographer is at P, 30m from the
edge of the track. G
100
(a) If angle COP = @ show that vox *
PC? = (100 sin 8)? + (130 — 100cos 6)? we aes
m
(b) Hence show that PC = \/(26 900 — 26 000 cos 6).
(c) If the photographer has a lens which can focus on objects at between
30m and 70m, for what range of values of @ is he able to take
photographs of the cyclist?
 THE POWER OF PYTHAGORAS 349

6E Find y as a function of x, given that

x = 3cos8, Vi St

5.1.4 rsin (0+ a)

In the last section you saw how to solve equations of the form asin # = bcos 8.
This section looks at equations of the form asin@ + bcos@ = c.

Two men are trying to carry a wardrobe through a doorway which is too low to
allow them to carry it upright. The wardrobe is 2.5 metres high and 1.5 metres
wide and the doorway is 2 metres high. If the men tip the wardrobe, as shown in
the diagram, they will be able to carry it through the doorway.

From the diagram, a= 1.5cos0

b= 2.5sin@
If the wardrobe is to pass through the doorway, a+b <2

=> 1.5cos#+2.5sin@ <2

You do not yet know how to solve the equation 2.5 sin + 1.5 cos @ = 2 using
analytic methods. In the questions which follow you will look at alternative ways
of writing this equation in order to solve it.
350 MATHEMATICAL METHODS
wo A ee ee ees

1c

The graphs of 1.5 cos and 2.5 sin @ are illustrated. The function
1.5 cos 6 + 2.5 sin @ is the sum of the two separate functions.

y= 1.5 cos 6

1 (a) In order to find the sum of two graphs, you may use the method of
‘pointwise addition’. The value of the graph at A is added to the value
at B in order to give the value at C. By adding pointwise the two
graphs shown above, obtain a sketch of the graph of

y=2.5sin9?+1.5cos@ for —180° < 6 < 360°

(b) Check your sketch using a graph plotter, and write down an
approximate solution to the equation.

2.5sin6+1.5cos@=2 between 0° and 90°

(c) The wardrobe on page 349 can be rotated through any angle between
0° and 90°.

(i) What is the greatest height of the top corner above the ground, and
for what value of @ is this height achieved?

(ii) Through what range of angles can the wardrobe be tipped so that
it fits through the door?
 THE POWER OF PYTHAGORAS 351

2 (a) Use a graph plotter to examine the graph of y = 3sin@ + 4cos 0.

(b) The resulting graph should be of the form y = rsin (0 ay Find the
values of r and a from your graph.
(c) Repeat parts (a) and (b) jor one or two more graphs of the form
y = asin6+ bcos@. Z

3 The first two questions suggest that the graph of y = asin6 + bcos@ is
| identical to a graph of the form y = rsin (6+ a). The diagram shows how
the two expressions are connected.

(a) Explain why

(i) h =asin@+ bcosé

(ii) b= rsin(9+a)
(b) sore r and a in terms of a and b.

4 (a) Use the result in question 3 to express 4sin@ + 7 cos @ in the form
rsin(0+ a).
ee
a

(b) Verify your answer using a graph plotter.

tee
————

esee
aa
cS
352 MATHEMATICAL METHODS
eaeSS

Example 7
Solve the equation
6sin@+ 9cos@ = 7

for values of 6 in the range 0° < 0 < 360°.

Solution
6 sin @ + 9cos@ is equivalent to the expression rsin (9 + a), where r and a are
found from the triangle shown below.

r= /(6 +9") =10.82 and a=tan 2=5631°

So,
6 sin @ + 9cos@ = 10.82 sin (6 + 56.31)
=> 10.82sin (6 + 56.31) =7

> sin (8 + 56.31) = 0.6472

Solving the equation sin x = 0.6472, where x = 0 + 56.31, gives x = 40.33°.

sinx

40.33° 180° 360° x

From the sketch graph, you can see that the solutions are
x = 40.33°, 139.67°, 400.33°, 499.67°, and so on
Since x = 0+ 56.31,
6 = —16.0°, 83.4°, 344.0°, 443.4°, and so on
The solutions in the range 0° < 6 < 360° are
6 = 83.4°,344.0° correct to 1 decimal place
 THE POWER OF PYTHAGORAS 353

§.1 Exercise 4

1 (a) Express 3sin@+ 2cos6@ in the form rsin (9+ Q).

(b) Solve the equation 3 sin@ + 2cos 8 = 3 for 0° < 6 < 90°.

(a) Explain why the maximum value of 5 sin@ + 12.cos@ is 13.

(b) Solve the equation 5 sin@ + 12cos@ = 9 for 0° < @ < 360°.

Any point on the ellipse

has coordinates of the
form (4 cos @, 3 sin 6).

(a) In terms of 6 what are the lengths of:

(i) AD (ii) AB (iii) the perimeter of the rectangle ABCD?
(b) If the rectangle ABCD has a perimeter of 14 units, explain why
3sin9 + 4cos@ = 3.5.
(c) Solve the equation 3 sin @ + 4cos @ = 3.5, and hence find the lengths of
sides of the rectangle whose perimeter is 14.
(d) What is the largest possible perimeter, and for what value of @ does it
occur?

4E The octopus ride is a common feature of fun-fairs. There are various designs:
the one illustrated moves in a combination of horizontal circles.

Wawa
fl pis 2S;
x
 354 MATHEMATICAL METHODS
aT

The diagram below shows the position after an arm has moved through an
angle of @°.

To simplify matters, assume that the arm rotates about O, but that the chairs
(at B, etc.) do not rotate about A.

(a) Explain why the chair at B will move in a circle, and find the centre and
radius of this circle.
(b) Calculate, in terms of 6, the distance, d, of B from the x-axis.

(c) Write the expression for d in the form rsin (6 + a).

(d) For what values of 6 is the chair at B farthest from the x-axis?

5.1 Tasksheet E1 — Extending the method (page 559)

5.1.5 Addition formulas

In section 5.1.4 you considered the use of the expression r sin (9 + a). Here we
are concerned with general formulas for the sine and cosine of the sums of
angles. Consider a rotated rectangle with a diagonal of length 1 unit.

if

>
 THE POWER OF PYTHAGORAS 355

We can obtain a formula for sin (A + B) as follows.

c+d=1sin(A+B)
also, c+d=asinA+bcosA

=> sin(A+B)=asinA+bcosA
Then a= cos 8, b= sinB

=> sin(A +B) = sin AcosB

+ cos AsinB

Similarly, for the cosine of angle (A + B):

From the diagram, c = d—e

| d=acosA, e=bsinA

c=1cos(A+B)
=> cos(A+B)=acosA—bsinA
=> cos(A+B) = cos AcosB
— sin Asin B

If we let B = A in each of the expressions we obtain important results known as

the double angle formulas.
sin
(A + B) = sinAcosB+
cos Asin B
Let B = A; then

sin2A = sinAcosA+cosAsinA =2sinAcosA

356 MATHEMATICAL METHODS

Similarly,
cos (A + B) = cos Acos B — sin Asin B
Let hi -A:

cos2A = cos” A — sin? A

or, using sin? A + cos’ A = 1, so that sin’ A = 1 — cos” A, we obtain

cos2A
= 2cos*A—1

Other formulas for sin(A — B) and cos (A — B) can be obtained from the
formulas for sin (A + B) and cos (A + B) simply by replacing B by —B. You
should familiarise yourself with the following identities, which are known as the
addition formulas, together with the double angle formulas for sin2A and cos2A
which are derived from them.

Example 8

; 1 1
Show that sin a =a ve sinx
+ cosx).

Solution

Se a Ee oh BOS &gees
Nes ne
SID

We. if 1 3
= oe ee cost =O and Ror

1
= 5 (v3 sin x + cos x)
 THE POWER OF PYTHAGORAS 357

S.1 Exercise 5

1 (a) Use the formula for sin (A + B) to show that

1 3
sin (x + 60°) = 5 sinx ae Weoss

(b) Check this result using a graph plotter.

(a) Use an addition formula to simplify sin (x + 7).

(b) Explain your result graphically. (Note that x is in radians here.)

(a) Show that cos (A + B) + cos (A — B) = 2cos Acos B.

(b) Simplify cos (A — B) — cos(A + B).
341
(a) By writing sin 75° as sin (45° + 30°), show that sin75° = ot
(b) Use the method of part (a) to express sin 15° in surd form (i.e. using
square roots).

If A and B are acute angles with sin A= 4 and cos B= 7, find sin(A + B)
without using a calculator. [The Pythagorean triangles with sides 3, 4, 5 and
5, 12, 13 may be useful in finding cosA and sin B.]

Solve the following equations for x, where 0° < x < 360°.

(a) cos2x =7cosx+3

(b) 4cos?x
+ 2cosx = Ssin2x
)— 1+tanx
TE Show that tan (:+ a ee :
4 1 —tanx

8E (a) By writing sin 3x = sin (2x + x) show that sin3x = 3 sinx — 4 sin? x.
(b) Express cos 3x in terms of cos x. Verify your answer by plotting
appropriate graphs.
2 tan x
9E Show that tan24 =.
1 —tan*x

10E Solve sin 3x = sin’ x for x, where 0° < x < 360°.

11E (a) Show by a numerical method that

; sinh ‘ cosh — 1 "i
lim, ( : )= and jim(2=*) =o
(b) By expanding and factorising the expression sin (x + 4) — sinx, show
‘from first principals that — (sin x) = cos x.
dx
(c) Show that as(cos x) = —sinx.
dx
358 MATHEMATICAL METHODS
ee
eee

5.1.6 Solution of non-right-angled triangles: the cosine rule

To ‘solve a triangle’ means to find all the angles and lengths of sides in the
triangle. It is easy to do this for right-angled triangles. In the rest of this section,
methods for other triangles will be considered.

.1D

A triangle with sides a,b,c and angles A, B, C can be divided into two
right-angled triangles:

: pisos!
From triangle (@), Pythagoras’ theorem gives
a =p + x?

1 From triangle @, express both h and y in terms of b and angle A.

2 Explain why x = c — y. Using this and the previous result, express x in

terms of b, c and angle A.

3 Replace h and x in a’ = h* + x” by the expressions found for / and x in

terms of b, c and A.

4 Simplify the expression for a” found in question 3 by multiplying out the

brackets and using the fact that cos” A + sin’ A = 1.

The result is known as the cosine rule for triangles.

5 How does the cosine rule relate to Pythagoras’ theorem for right-angled
triangles?

6 By treating side b as the base of the triangle and using properties of

symmetry, find a similar expression for:

7 So far, you have only considered acute-angled triangles. Modify the

diagram and extend the proof to obtuse-angled triangles.
 THE POWER OF PYTHAGORAS 359

Example 9
The hands of a clock are 10cm and 7 cm long. Calculate the distance between
their tips at 2 o’clock.

Solution |
The angle between the hands is 60°. Using the cosine rule
with the triangle labelled as shown:
a@ =10°+7°—2x 10 x7 x cos 60°
= 100+ 49— 70

=79

a= /79=8. 9cm (to 1 decimal place)

5.1 Exercise 6

1 Triangle ABC is such that AB = 4cm, AC = 8cm and angle A = 43°.

Calculate the length of BC.

(a) For triangle ABC, with sides BC = a, AC = b, AB = c, express cos A in

terms of a,b,c.
(b) A triangle has sides 4cm, 5cm, 7 cm. Calculate its angles.

The hands of a clock have lengths 10cm and 7 cm. Calculate the distance
between the tips of the hands at:
(a) 4:30 (b) 8:00 (c) 6:00

ABC is a triangle in which BC = 8, CA = 6, AB = 7, and M is the midpoint

of BC. Suppose that angle AMC = 6. Write down by the cosine rule
expressions for AB? and AC? in terms of cos @ and add the results. Hence
calculate the length of AM.

A is 2.1km due north of B; C is 3.7km from B on a bearing 136°. Find the

distance from C to A.
360 MATHEMATICAL METHODS

5.1.7 Solution of non-right-angled triangles: the sine rule

@ x
The area of a triangle can be found using area = ; x base x height.

Complete the following:

1 In the triangle with sides a,b,c and angles A, B, C as shown, the height h,
can be expressed as
Lh = SSG.
Thus the area of the triangle is
5bh, —

Since triangles are normally described using lengths of sides and angles, the
formula area = }base x height cannot be applied directly. The new formula
gives the area in terms of the lengths of two sides and the angles between
them, i.e.
Area = tbe sin A

= $x product of two sides x sine of included angle

2 Treating a as the base, the height /) of the triangle can be expressed as

ins Site
Thus the area of the triangle is }ah) =

3 Use these two expressions for the area of the triangle to form an equation.
Simplify it and write it in the form
sinA

4 Treating side c as the base, find an expression for the height of the triangle
in terms of a and B, and hence find an expression for the area of the
triangle.

5 Use the expression obtained in question 4 with each of the previous

expressions to obtain two more equations simplified to the form
b Cc
sin B sin C

6 Find the area of triangle ABC such that AC = 7cm, BC = 4cm and angle
G50
 THE POWER OF PYTHAGORAS 361
-—_e———"::2_ _ mn

Example 10
In a triangle with sides a = 7, b = 12 and angle A = 23°, angle A is opposite side
a and angle B is opposite side b. Find angle B.

Solution
The first step in solving this triangle is to calculate the value of

Ae
si sina
Sy ais
! : b
Using the sine rule, i795
sin B
17
and since b= 12, —~= 17.915
sin B

Rearranging this equation gives

12)
Sintb
17.915 ~ 0-670 = B= 42.1" (to 1 decimal place)
However, a sketch of the triangle shows two possible positions for B,
corresponding to the two solutions.

This is sometimes called the

‘ambiguous case’ of the sine rule.

There are many angles which have sine

equal to 0.670. Two of these angles are
in the range 0° < 0 < 180°.

From the graph, @ could be either 42.1°

or 180° — 42.1° = 137.9°.
 362 MATHEMATICAL METHODS

5.1 Exercise 7

Remember that to ‘solve a triangle’ means to find all the unknown sides and
angles.

1 Use the sine rule to solve these triangles:

(a) (b)

12mm 20mm

Ts SN

2 Use the cosine rule and then the sine rule to solve these triangles:

(a) (b)
“>>
18cm 10cm om

Aas
80m

3 Find the areas of the two triangles in question 2.

4 Solve these triangles:

(a) (b)

25mm
7mm
24m

32m
 THE POWER OF PYTHAGORAS 363

38cm 9cm

5 <A plane flies north-east with an airspeed of 400 kmh_!. If the wind is
blowing at a steady speed of S0kmh ' from the west, calculate the distance
covered over the ground in one hour, and the direction in which the plane
has travelled.
364

Mathematical methods
.2 Vector geometry
5.2.1 Vector and position vectors

An aircraft A takes off from an airport O.

After 1 minute it is 1km east and 5km north
of O at a height of 0.8 km.

A second aircraft B is then 5km east and 5km

north of O at a height of 4km.

A situation of this kind, concerning the relative position of two aircraft and how
this may change with time, is handled most conveniently using the notation and
mathematics of vectors. Many problems in mathematics, and in the application
of mathematics, are best handled using vectors — you may have already done
vector work in mechanics, for example. In this section we extend this work to
consider the vector equations of lines and planes.

i]
Taking the vector b | to represent a position akm east and bkm north of O,
4 |

at a height of ckm, aircraft A has position vector Guan; and B has

5 0.8
position vector OB = | §
4
SS ———_ —
OA and OB are the position vectors from the point O of the points A and B
respectively.

They are given in column vector form above. The position vector OA of a point
. .
. . arg .

is often denoted by a or a.
 VECTOR GEOMETRY 365

A vector diagram shows the relationship between the vectors.

—-_—
sw —=—s >
A+ AB = OB A 5

—_—> ar >
or AB=OB-O b

The vector AB gives the displacement from A to B, that is the distance and the
——? . . . .

direction of a movement from A to B.

@ 2
1 Using squared paper, draw the position vectors from the origin (0,0) for
each of the two points A and B and find the vector AB in each case.

Sea S| 4) wa=| 4 b= |)
— — : —.
2 If p= OP and q = OQ, write down PQ in
terms of p and q. 4

3 It is not always necessary to use a square grid for vectors.

Vectors a and b are as shown on . : -D
this isometric grid. B
Fee ;: 5

° O ita 5 A e Cc

Fe

(a) Write down the position vectors of C, D, E and F in terms of a

and b.
(b) Find the vectors AB, Cp;
CD DE, EF and FC in terms of a and b.

(c) Check that CD + DE + EF + FC =0. Why is this so?

=
(d) Compare AD and EF. Explain what you notice.
366 MATHEMATICAL METHODS

4 On squared paper plot the points (x, y) whose position vectors are given by
the following vector equations. Take 0, +1, +2 and +3 as values of the
parameter f.

of)
=a)
© [= [le
What do you notice about each set of points?
‘ Bab oe 2
What is the significance of the vector is?

5 On squared paper draw the triangle OAB where A and B have position
12 0
vectors and :
0 6

Taking 0, 1, 2 and 3 as the values of each of the parameters s, t and u in

turn, draw the lines whose vector equations are:

© [l= [e]+2]
=f) a]
© [3]-[o] ++[7]
What do you notice about the three lines? Where is their point of
intersection?

OABCDEFG is a cube with edges of length 6

and axes as shown. P, Q, R are the midpoints
of the edges EF, FG and FB.

(a) Find the position vectors of the points P, Q and R.

(b) Demonstrate that PQ + QR “eR a0)
 VECTOR GEOMETRY 367

7E The vector equation lee : + t zm in question 4can be written

in terms of its ah ene as:
x=342 @
y=-14+3t @ 7
(a) Express ¢ in terms of x from equation @.
(b) Substitute this for t in equation @ to find the Cartesian equation of
the line in the form y = mx + ¢.
2
(c) What is the gradient of the line? How is this related to the vector wu

8E Find Cartesian equations for the lines:

. aoe] = 3 —2 EG d Z
(a) : = a =Fs 1 (b) ; II = af =

Position-vector
Position of a This vector is
of any point particular
on the line the direction of
point on the line the line
368 MATHEMATICAL METHODS
Seen

Example 1
Find a vector equation of the line joining the points A (2,3) and B (5, 4).

Solution
The position vector of a point on the line is

|| and the direction of the line

3
is given by the vector eS ||
. Thus the
equation of the line AB is

Pio bledi]
Note that this answer is not unique.

Exercise 1

1 Find vector equations for:

(a) the line joining the points (4,1) and (7,7),
(b) the line joining the points (2,1) and (—1, 5), kG
(c) the line through the point (5,1) parallel to the vector 7 |
;
(d) the line y = x,
(e) the y-axis.
2 The position vectors of four points P, Q, R and S are

ict L-aleL_a}
;
‘| bs 4 m4[a]
‘
="
and s

(a) Find the vectors PQ and SR.

What does this tell you about the quadrilateral PQRS?
(b) What can you say about the vectors PS and QR?

3 OABCDE is a triangular prism with

6 0 0
OA =-0'l SOB = 181, «OC Sele
[0 0 10
(a) (i) Find the position vectors of D and E.
(ii) Find the vectors AB, AD, RES AE and DE.
(b) M is the midpoint of AB and N is the midpoint
of DE.
(i) Find the position vectors of M and N.
(ii) Find the vectors AN and ME.
(iii) Explain what you notice about the results.
 VECTOR GEOMETRY 369

OABCDE is the roof of a house. OABC is

a rectangle with OA of length 8m and
OC of length 10 m. The ridge DE of
length 6 m is positioned symmetrically,
3m above the rectangle.

(a) With axes as shown, find the position vectors of A, B, C, D and E.

(b) Find the vectors AD, OD, BE and (oe. representing the slant edges of the
roof.

(c) What is the length of a slant edge?

5.2.2 Equations of lines

The ideas of section 5.2.1 can easily be extended to three dimensions.

Consider the motion of the two aircraft described.

The position of aircraft A, t minutes after take-off, is given by the equation

x 1
Mi Sell S
Poll 2 0.8

The position of aircraft B at the same time, t minutes, is given by the vector
equation
% be 0
y= TO) tS
x 4 0

0
Aircraft B is flying in the direction of the vector | 5
0

The position vector of each aircraft after 2 minutes is

2 5
OA = | 10 and B = | 10
1.6 4
370 MATHEMATICAL METHODS

—_—=s >
The displacement vector ‘AB at this moment is given by OB — OA

3
—-
Le. Bes 0
2.4

Notice that when t = 4 the aircraft have the same position vectors — a collision
occurs!

Each aircraft is moving in a straight line. The following examples examine in

more detail the vector equation of a line.

@ »
Questions 1 to 6 refer to the cuboid CFGEOADB with OA = 4, OB = 6 and
OC:

al tl

2ee) II

al
Sr
1S
Boars
Gio)

1 Find the position vectors of the points D, E, F and G.

2 Points are given by the vector equation

x 4
y| =A]6], where A is a parameter.
z

(a) Which points correspond to \ = 0 and \ = 1?

(b) On a large copy of the diagram, mark points corresponding to

\= i, + and 3. What do you notice about them?

(c) What can you say about the positions of the points where \ = 2 and
A= -1?
 VECTOR GEOMETRY 371

X. 4
Sey =X] 6| is the vector equation of the line OD.
z 0

Identify the lines whose vector equations are:

Re 0

(a) |y |=A] 0
3

(b) ==IN|

Give vector equations for the lines:

(c) OB (d) OF (e) OG

4 Which points correspond to \ = 0 and A = 1 on the line with vector

equation

x 0 4
wee 0) +A/6]?
z 3 0

On your copy of the diagram, mark points corresponding to A = 4, 5 and 3.

Where would be the points corresponding to \ = 2 and A = —1?

5 Identify the lines whose vector equations are:

= 0
avi nea Gh eNO
& 0 3

BE 4 —4
(by) hey (P= ONE A 6
z 3 0

0 23
What is the significance of the vectors | 0} in (a) and 6 | in (b)?
3 0

6 Give vector equations for the lines:

(a) AD (b) AG (c) AE

372 MATHEMATICAL METHODS

7 (a) On squared paper, draw the straight lines whose vector equations are
given below by plotting points, taking 0, +1, +2 as the values of the
parameters 4 and pu.
5% Aesa ks cal eda). x
yl Lo 1] nr aac ane ce
(b) Where do the lines intersect?

(c) What are the values of A and p at this point?

(d) By equating components, write down the two simultaneous equations
given by
POC ea fee
Lo Tabak) Claes
(e) Solve the simultaneous equations and confirm your values of A and wp.

8 Use simultaneous equations to find the intersection of

iG ) 1 1 2
= +X and ies + pb
A 2 Zz y 3 ={

Position vector Position vector of This vector is

of any point a particular point the direction
on the line on the line of the line

Example 2
Find the point of intersection of the lines with equations
% 8) 2 x 4 1
= aN and = +p
y 5 1 y 1 —2
Note that it is necessary to have different parameters, \ and yp,
for the two lines
as it would otherwise be impossible to generate points indepen
dently.
 VECTOR GEOMETRY 373

Solution
At the point of intersection the two position vectors will be equal.

—34+2’\=4+ 4p
and ~“S-+A=1—=2
or, 2A = [b= 7
A+ 2p = —4

Solving these two equations for A and yu gives A = 2, yp = —3 and the position
: ! =3 4 1
vector of the point of intersection as + piled

The method of example 2 can easily be extended to lines in three dimensions.

The x- and y-coordinates can still be equated to find \ and yu. Then, when these
values of and yu are substituted into the z-coordinates, two possibilities exist:
(i) the coordinates are equal and so the lines meet;

(ii) the coordinates are unequal and so the lines do not meet.

In two dimensions, the lines in case (ii) would be parallel. In three dimensions
there is a further possibility, because a pair of distinct lines may have no point in
common and yet not be parallel. In this case they are called skew lines.

Example 3
2 4 4 2

Show that the lines r= |3| +t] —1] andr= |7J| +s] —2] meet and
5 3 y) 3
find the point of intersection.

Solution
If the point (x,y,z) lies on both lines then
x pee. 4 4 iz
al MN) | a Uc Van FA Wc RZ
z by e) Ji 3

for suitable values of s and t, which must satisfy the three equations

2+ 4t=4+2s

Bh i I lg

§+3t=2+43s

Adding the first two of these equations gives

Sesh == v=? “and 3s= 3

374 MATHEMATICAL METHODS
eee
ee

there is
The third equation is also satisfied by t = 2, s = 3, which means that
Substitu ting ¢ = 2 in the equation of the
a point common to the two lines.
first line gives the coordinates of the point, (1051510).

Example 4
2 [ 4 4 2
Show tharthe lines rz 3 ¢|—1 | and r= |74 45:21 earerskewalimess
6 5 8 1
Solution
Following the method of example 3 to find a point of intersection gives the
equations

24+ 4t=4+42s

ots

6+5¢=8-+ s

As before, the first two equations are satisfied simultaneously by t = 2 and

s = 3. But these values do not satisfy the third equation so there is no point
common to the two lines. The lines are not parallel, since the directions of
4 ?)
—1]| and | —2| are different. The lines are therefore skew.
5 il

1 Write down vector equations for the two lines shown

in the diagram.

Calculate the point of intersection of the two lines.

 VECTOR GEOMETRY 375

2 OABCDEFG is a cuboid with edges OA, OC

and OD of lengths 4, 5 and 3 respectively.

Which lines have the following vector equations?

83 0

a =r) 5
z 3

oe 4 —4
(Diets .0)] dS
z 3 0

% 0 4
(c) Jy} = 15S] +A] —S
z 0 B

4 —4
(d) }y| =|]5] +A] —-5
%, 0 3

3 For the cuboid of question 2, find vector equations for the lines:
(a) AB (b) AC (ce) AF (d) AG

4 Two slant edges of a square-based pyramid, with its base on the xy plane,
have equations:

x =i ae 4 =)
y|= +A|-1 and |y|=]-3}+p| 1
z 1 z 0 1

Find values of and py such that the y and z components are equal.
Check that these values both give the same value for the x-coordinate.
Hence write down the position vector of the vertex of the pyramid.
376 MATHEMATICAL METHODS
ee
ee ESS

5 Find whether the following pairs of lines meet. If they meet, find the
coordinates of the common point; if they do not, find whether they are parallel
or skew lines.
yi 1 5
(ay r= | 3 +r Ol, r= |0p +s
5 Mu, 4

1 i y,
(UG), ele ea BS Peres Sei be
3 4 6

Z 4 —3 3
(Cees , r= 6) +¢t|-1
3 —$§ —8 2

4 5 2 3
(Apes) SN a ee ee al
1 4 6 2

6 Prove that the two lines with equations

0 [ 6 =4 2
i 2| +s] -1 anda 6| +t
=3 = a4 =|
have a common point.

5.2.3 Scalar products

You are familiar with the idea of adding and subtracting vectors. Here we
consider one way in which a meaning can be given to multiplication of vectors.
This problem is closely related to finding the angle between two vectors.

-[] E
Consider vectors a and b as shown.

How could you find the angle, 0, between the two vectors?
 VECTOR GEOMETRY 377

One way would be to use the cosine rule on triangle OAB. The lengths OA and
; d —> >
OB can be found using Pythagoras’ theorem and since AB = OB — OA its
length can also be found.

OA=3/6 t= 10) OB =O 3°) = 4/13

AB =b-a=| INS OCGisoe ae) ety 0)

So, by the cosine rule AB? = OA? + OB? — 2 OA OBcos 6

> $=10+413 —2,/10,/13 cos0

e)
= cos d= = 017894 = 0 = 37.9°
130
The aim of the following questions is to develop a more convenient method of
finding the angle between two vectors.

& 2c

3 1
For the triangle illustrated above, a = ,] and b= |:

(a) Find the magnitudes (or lengths) of vectors a and b.

(b) Write vector c in terms of a and b and then express it as a column

vector. Find the magnitude of vector c.

(c) Use the cosine rule with your values of a, b and c (the magnitudes of a,
b and c) to calculate cos @ and find the angle 0.
378 MATHEMATICAL METHODS
ee

; ay by
2 For any triangle, a = and b= 4
ay b,

ss 2
(a) Explain why a? = aj + a, and write down a similar expression for Bis
Oia : :
(b) As before, c = b—a= : 1] Since the magnitude of c is c, you
; bo -—a
can write 2 2

2 = (bj — a) + (by — m)”

By multiplying out the brackets and using the results of (a), show that
C= a +b* — 2(a,b, + ab)
(c) By comparing with the cosine rule, explain why
a,b; + ayb, = abcosé
3 Use the result in 2(c) to calculate # in question 1.

The word scalar is used to emphasise that the product is not a vector quantity.
In fact, a second product, known as a vector product, exists but is beyond the
scope of this book.

: } ay by oe ha
The scalar product is written as . or as a+b which is pronounced
ay 2
‘adotb’. For obvious reasons some writers refer to it as the ‘dot product’.

Consider the following three situations.

eer ie
ze) VAN
 VECTOR GEOMETRY 379

Applying the scalar product definition in each case:

: b
(i) a+b= |a| |b| cos 0 .
(ii) b is in the opposite direction so ab is »
|a| |b| cos (180 — 6) = —|a| |b] cos0
a
(i) Both a and b are pointing towards the origin @
so a+b = |a| |b| cos 0 b

ay by
In three dimensions, if the two vectors are a = | a> and. Sb =.|165"\.,
a3 b3
the definition extends naturally to a,b, + a,b, + a3b3 or abcos 6, where, in order
to find a, the length of a, it is now necessary to use the three-dimensional form
of Pythagoras’ theorem.

Example 5

Find the angle between the vectors 2) and

—3

Solution
a-b=5SxX242x4+4+(-3)x1=15
a=J/(254+449=/38 b= V/(44+164+1)=
721
iS)
So cos 738/21
C—O

A* Exercise3

f=) f=
1 Find the angles between:

2 Intriangle ABC, A= (3,2), B=(-—1,3), C= (1,7).

— —
(a) Find vectors AB and AC.
(b) Explain why, to calculate angle A, you should find AB : KC and not
—~>> >
AB: CA. Hence calculate angle A.
3 Find the angles between:
2 8 4
(a) 1 and mA (b) | —1 and
—12 1 —8 —4
 380 MATHEMATICAL METHODS
ee ee a en a

4 Intriangle PQR, P=(5,—3,1), Q=(-2,1,5), R=(9,5,0).

Find the angles of the triangle.
5 1k Ons,2), B13, 113), C= (7,12, - 1) sD toa
show that ABCD is a parallelogram and find its sides and angles.
6E Use the scalar product method to find the angle made by a longest diagonal
of a cube with:
(a) an edge of the cube, (b) a face diagonal,

(c) another longest diagonal.

5.2.4 Properties of the scalar product

The usefulness of the scalar product extends beyond providing a convenient
method for finding the angle between vectors. Some of its properties are explored
further in the following questions and applied later to three-dimensional
geometry. Many of the properties of the scalar product are similar to those of
ordinary algebra. Here we examine the similarities and differences.

@ 2»

-() 20) (a
Consider the vectors:

1 (a) Draw a diagram showing these five vectors.

(b) Which pairs of vectors are parallel and which are perpendicular?

2 (a) Calculate the magnitude of each vector.

(b) Calculate the scalar products: aa, ib=b, tc-comd-dumece:

(c) What do you notice?

3 (a) Calculate a-b and ba.

(b) Calculate a-c and c-a.

(c) What do you notice? Explain why this occurs.

4 (a) Calculate a-b+a:c.

(b) Calulate a-(b +c).

(c) What do you notice?
 VECTOR GEOMETRY 381

5 (a) Calculate the scalar products a+b, a-d, are.

(b) What can you say about two vectors which are perpendicular?
(c) Find a vector which is perpendicular to b.

(d) Does it follow that, if two vectors have scalar product zero, then they
are perpendicular?

: : 4 ; P
GE A’A is the diameter of a circle centre O and P is NS
any point on the circumference.
—s
(a) With a= OA and P= OP, express OA’, “AP
Tyee .

and A’P in terms of a and p. A

(b) Calculate AP - AP.

(c) What is the geometrical significance of the

value of this scalar product?

5.2 Exercise 4
5 —2
ea — b= C=
y) 3 3

(a) Calculate the magnitudes of a, b, and c.

(b) Calculate the scalar products a+b, b-c, cea.

(c) Which pairs of vectors are perpendicular?

(d) Find a vector which is perpendicular to b.

1 4
Dea 2 b= 0 c=]—-S5
il —2 Dy)

(a) Calculate the scalar products a+b, b-c, c-a.

(b) Which pairs of vectors are perpendicular?

382 MATHEMATICAL METHODS
ea EE ee

aes
(a) Find the position vector c of the midpoint of AB.

(b) Find the vector d = BA.

(c) Calculate the scalar product c-d.

(d) Draw a diagram of the triangle OAB with the point C included. What
can you deduce from the value of the scalar product c+ d?

4E OAB is a triangle with the altitudes B

from A and B intersecting at H as
shown.

[The altitude of a triangle is the

perpendicular from a vertex to the b
opposite side.]

OA, OB and OH are denoted by a,

b and h.
O a A

(a) Why is a-(b —h) = 0?

(b) Write down a similar equation involving a — h.
(c) By subtracting the two equations, show that (a — b)-h=0.
(d) Explain how this proves that the altitudes of a triangle are concurrent.
(Concurrent lines all pass through a single point.)

SE (a)

OPRQ is a parallelogram.
; SS Se ———>
Write down the vectors OR and QP in terms of p and q.

(b) Explain why (p + q)-(p—q) =p? — q’.

(c) If OR: Qe = 0 what can you say about
(i) the lines OR and QP, (ii) the sides of the parallelogram?

a
 VECTOR GEOMETRY 383

5.2.5 Vector equations of planes

In three dimensions, as well as the equations
of lines, a description ot planes can be
important. In describing crystals, for example,
you might be interested in the various faces
and the way they are related. Suppose A, B
and C are three points with position vectors
5 0 0
On); 5 and |0|] respectively.
0 0 4

Example 6
(a) What are the vectors AB and AC?

(b) Where are the points with the following position vectors?
— — —
(i) OA +5 AB +3AC
— — —
(ii) OA +4 AB +4AC
(iii) OA +2AB — AC
>> — —
(iv) OA +AAB + pAC

(c) Can every point in the plane ABC be found from a suitable choice of \ and
win (b) (iv)? :
Solution
—3 —3
— =
(a) AB=b-—a= S C=e—a= 0
0

(b) (i) M, the midpoint of BC.

(ii) D, the midpoint of AM.

(iii) OA + 2AB — AC is the— point F— in

the plane ABC, where AE = 2AB
——>
and. EF =AG,
— — —.
(iv) OA + AAB + AC is a general z
point in the plane ABC.
, —
(c) Every point in the plane ABC can be reached by the vector OA together
with some combination of vectors AB and AC.
384 MATHEMATICAL METHODS

1 a Points A and B have position vectors a and b relative

to an origin O as shown.

b B Copy the diagram and mark on it the points having

the following position vectors:

EOC 24 (b) OD =4b

(c) OF =2a¢b (d) OF =2a—1b
(e) OG =3at+2b~ = (f) OH = 4a +-2b
2. . . . . A and B are described by the position
Fix : «D «Gx vectors a and b as shown. Express the
B , position vectors of the points C, D, E, F,
bY «x! G, H, I in terms of vectors a and b.
A ° Cx

Questions 1 and 2 illustrate that it is possible for position vectors of all points
in the plane OAB to be expressed in the form Xa + pb for some values of
and p. The remaining questions illustrate how this technique may be used to
find the equation of any plane, not necessarily through O.

The diagram shows a cube with sides of

length 6cm. The origin is at O.
The following questions all refer to this cube.

3 Consider the plane DEF. Find the vectors DE and DE.

Explain why the position vector of any point in the plane DEF can be
written as

a3 6 —6 0
y| = ]6} +A} O}| +p] -6
z 0 6 6
This equation is known as the vector equation of the plane.
 VECTOR GEOMETRY 385
——

4 State the points with the following values of \ and py and find the
coordinates of each point.
(ene 0.0 7 =0 (b) A

(Oe oD Aa 2 (e) A=,

5 Consider the equilateral triangle ABC.

s , —>- ——>,.
What are the direction vectors CB and CA?

Using OC as the position vector write down the vector equation of the
. ee Che . .

plane ABC.
What do you notice about the planes ABC and DEF?

6 Repeat question 4 for the plane ABC.

7 (a) Suggest a possible vector equation for a plane parallel to ABC which
passes through the origin.
(b) Likewise suggest an equation for a parallel plane through G.

8 If H is the midpoint of the line CF, find a vector equation for the plane
DEH.

Position vector Position vector Vectors parallel

of any point of a particular to the plane, but
on the plane point on the plane not parallel to
each other
 386 MATHEMATICAL METHODS

Example 7
—6), parallel to the
Find the vector equation of the plane through the point (2,5,
4
vector | 1 | and to the z-axis.
3

Solution
0

Since the z-axis is parallel to the vector | 0 |, the vector equation of the
plane is 1

G 2 4 0
yi S51 +Al1) +p] 0
z —6 3 i

Mee eee 8 ee eee

Lat N Exercise 5

1 Find a vector equation for the plane through the three points
OMe 1) 4B (e104) and. C= 4a).
2 A plane cuts the x-, y- and z-axes at x = 2, y = —1 and z =3. Find a vector
equation for the plane.

3 OABCD is a regular square-based pyramid with

base edges of length 4 and height 3.
(a) What is the position vector of the vertex D?

(b) Find vector equations for the faces OAD and

BCD.

4 Explain why it is not possible to find the vector equation of the plane through
the three points A (2,3, 1), B (—1,2,4)and C (—4, 1,7).

5.2.6 Cartesian equations of planes

Section 5.2.5 showed you how to write the equation of a plane in vector form.
Here we consider what the equation of a plane will look like in Cartesian form,
and the relationship between the two alternative forms of the equation.

Consider the Cartesian equation x + y + z = 4. The point (1, 3,0) satisfies this
equation. Other obvious points which satisfy the equation are, for example,
 VECTOR GEOMETRY 387

(4, 0,0), (0, 4,0), (0,0, 4), (3, 1,0), (2,2,0) and so on. Clearly, there are
infinitely many such points. If these points are plotted, the following diagram is
obtained.
vh

~~

This suggests that x + y + z = 4 represents a plane.

@ 2»
1 The vector equation of the plane DEF is:
6 =6 0
= |6/+A! O| +p] -6
\
AQ
see
SS “1-0 6 6

Considering the components separately, you can write the x components as

x= 6— 6
Write down the corresponding equations for the y and z components.

Add the three equations together and simplify. The result should involve x,
y and z only and not A and yp.

This equation is known as the Cartesian equation of the plane.

2 (a) What is the Cartesian equation of the plane ABC? Check your
suggestion by working from the vector equation as in question 1.

x 0 0 6
In 5.2£ you found that this was }y| = |O/| +A] 6/+yp| O
. Zz 6 6 ~6
388 MATHEMATICAL METHODS
SS TS NE CAE ee ne

(b) In 5.2r, the vector equations of the planes through O and G parallel to
ABC and DEF were found to be
% 0 6
yl A| 6) pw] O| and
z —6 —6
x 6 0 6
A= 6 =A 16 | ane GO
ze 6 26 —6
respectively. What are the Cartesian equations of these planes?

3 The vector equation of the plane DEH, where H is the midpoint of CF, is
x 6 6 2
Vi oO lee Onl al 6
z 0 6 6
Write down equations for the x, y and z components.
By eliminating \ from the first and third equations and then eliminating p,
find the Cartesian equation of the plane DEH.

x 5 —3 2
4 A plane has vector equation |y} = |2] +A] OO] +p] 3
z 4 =6 1

Find the Cartesian equation of the plane.

5 The Cartesian equation of the plane DEF is x + y+ z= 12.

il x
Using scalar product notation, this could be written as |1 | - |y | = 12.
1 ie

To investigate the significance of the vector | 11, first note that it is in the i
6 1
same direction as OG. which is | 6
6

How is this vector | 1} related to the plane DEF?

1
1
Calculate the scalar products of | 1] with DE and DFE. What do the
results tell you? ql
 VECTOR GEOMETRY 389

Using these ideas, the equation x + y + z = 12 can be written, using scalar

product notation, as

1 x
1G) Roa ba" [raced
es
1
You saw in 5.2F that the vector | 1 | is at right angles to the plane. This vector
i
is known as the normal vector to the plane and is usually written n. The result
suggests that n-r = k, a constant, for any point r in the plane.

Since a plane is two-dimensional it is not possible to choose a single direction in

which to measure the gradient. However, since the plane has a unique direction
to which it is perpendicular, it is possible to specify its orientation using the
normal vector.

The vector equation of any plane can be written in the form

r=a+Ab+ uc

where r is a general point on the plane, a is a particular point on the plane and
b and c are two vectors parallel to the plane.

Choosing a vector n, which is perpendicular to the

plane, as normal vector and considering the scalar
product nr, you obtain
ner=n-a+An-b+yun-c

Since the vector n is perpendicular to the plane, the dot products n+ b and n-c
will be zero. The product n-a will be a constant because a is a fixed point in the
plane, so n-a is the product of two fixed vectors.
 390 MATHEMATICAL METHODS
ep ES =

The equation therefore reduces to

n-r=n-a or n-r=d_ a constant
d is re
whe

In the case of ax + by + cz = d,
a x

no b ir = y

Cc z

Example 8
Zz 5)
A plane through a point with position vector | 1] has normal vector | —1}.
Find the Cartesian equation of the plane. 3 4

Solution
3 x =) 2
Using n-r=n-a, hes | <2) «ae le
4 z 4 3

Se
Vite tee

5.2 Exercise 6

1 Find the Cartesian equations of the following:

2
(a) a plane through the origin with normal vector | —3 ’

1
(b) a plane through the point (3, 1, —2) parallel to the plane of part (a);

5
(c) a plane through the point (3,1, —2) with normal vector | —2
0
 VECTOR GEOMETRY 391

2 OABCDEFG is a unit cube with x-, y- and z-axes

as shown.
For each of the following planes, use inspection to
write down a normal vector and give the Cartesian
equation of the plane.
(a) ACGE (b) OBFD (c) ADC

(d) EGB (e) OABC

3 A, Band C are 2, 3 and 4 units from the origin z

along the x-, y- and z-axes. 4KC
(a) What is the position vector of C?

(b) Write down the vectors CA and CB.

(c) What is the vector equation of the plane ABC?

B
(d) Write down equations for x, y and z in terms is
of the parameters \ and wp.
(e) Eliminate 4 and yp to find the Cartesian A 2
equation of the plane ABC. x
(f) What is the normal vector to the plane ABC?

4E A regular octahedron ABCDEF is placed with

its vertices on the x-, y- and z-axes, each at
one unit from the origin.
(a) Find the Cartesian equations of the planes
AEB and DCF.

(b) What do you notice about these two

planes?
(c) Find the Cartesian equations of the planes
ECB and FAD.

> |5.2 Tasksheet E1 — Intersections (page 561)

392 MATHEMATICAL METHODS

5.2.7 Finding angles

When describing a polyhedron or a crystal, it is often useful to know the angle
between adjacent faces, the dihedral angle.

It is conventional to consider the angle between two

planes to be the angle formed by the two perpendiculars
to the line of intersection of the plane. 7

Since it is the normals that specify the direction between

the planes, the angle between the planes can be found
from the angle between the two normals.

Example 9
Find the angle between:

(a) the planes 3x +y—2z=4 and 2x—y+5z=1

x 1 =
(b) the plane 3x+ y—2z=4 andtheline | y| =|]2]+A]| 3
z 3 —5

Solution

(a) 3 2
ny, — Ny — —1

nj:ny =6—1-—10=-—5

Im|= /(9+1+4) = 14
jm.| = /(4+1+425) = /30
Ny - Ny = || |22| cos 0 > —S = v/14,/cos
30 8
=> 6=104°
The acute angle between the planes is 76°.

ait
(b) The direction of the line is given by the vector b = 3
=)
 VECTOR GEOMETRY 393

3
Thus the angle between the line and the normal n = 1
=
is given by b-n = |b| |n| cos0
=> 10 = /35./14 cos 8 :
> G=63.A-
Thus the angle between the line and the plane is
90° — 63.1° =:26.9°

§.2 Exercise 7
a+b
1 Use the result, cos 9 = —— for the angle between two vectors to calculate the
ab
angle between the normal vectors for each pair of planes. What is the angle
between the planes in each case?
Cr ee ee Se 2x + 39-2
=4
(b) x — 3y — 2z = 1; 5x +2z=—-—5
(Cex 22 14s y+ 3z=6

“E-ChoR Gl
2 Find the angles between these pairs of lines.

x 1 —1 Ss 0 —2
(b) |}y| = ]2] +A] 04; yal e—er ale ee hires
z 3 3 & 1

3 Calculate the angle between the line

5% 1 2
y= P24 aX) 1
z il 1

and the normal vector to the plane 2x + 3y — z= 6.

What is the angle between the line and the plane?

4 The equations of the faces AEB and ECB of the regular octahedron in question
4E of exercise 6 are

xtytz=1 and -x+y+z=1

Calculate the angle between the planes.

What is the dihedral angle of a regular octahedron?
394 MATHEMATICAL METHODS
a

SE OABC is a tetrahedron where O is the origin and A, B, C are the points

(112:0) (0,451) and (15041):

(a) Draw a diagram to show the tetrahedron.

(b) Calculate the lengths of the six edges to show that the tetrahedron is
regular.
(c) Find the Cartesian equations of the faces OAB and OAC and calculate
the angle between them. What is the dihedral angle of a regular
tetrahedron?
(d) Write down the vector equations of the edges BC and OA and find the
angle between them. What can be said about opposite pairs of edges of a
regular tetrahedron?

(e) Calculate the angle between the edge BC and the face OAC and use the
result to calculate the height of the tetrahedron, taking OAC as base.
 395

Mathematical methods
3 Binomials

5.3.1 Binomial expansions

Algebraic expressions which have two terms, for example a + b, 2x — 3y and
a} . . cs F
p’ + 2p, are known as binomials. In the same way, an expression like a+ b+ c,
with three terms, is referred to as trinomial.

If you expand the brackets for (a + b)(a +b), you obtain the identity
(a+by =a 4+2ab 4 b°
& 3A

1 Expand (a+ b)(a° + 2ab + b’) to show that

(a+bpsa@ +3ab4+3ab +b
2 Find a similar expansion for (a+ b)*.
If you include the obvious results that (a+ b)° = 1 and (a+b)' =a+b and
also include the (usually unnecessary) coefficient of 1, you can tabulate your
results as follows.
(a+b) = 1
(Geb la + 1b
(a+by = les ee aba 1b"
Gaby = la 94 3ab + 3ab? .-- 1b

3 (a) Can you spot the pattern produced by the coefficients of the various
terms?
(b) Check whether your answer for (a + b)* fits the pattern.
(c) Assuming the pattern continues, write down what you would expect
for the expansion of (a +b)’.
: ; Ay» FS ee Sen
[The answer should have six terms, involving a, a'b, ab’, ab’, ab
and b°.]
You should have spotted that the coefficients of the various terms are the
binomial coefficients you may already have met in Pascal’s triangle. The
pattern continues for all positive integer powers of a+ b.

4E Find (a + 2b)’.
396 MATHEMATICAL METHODS

(a+b)* =a'+4a°b+ 6a°b” + 4ab> + b*

The coefficients of a particular power of a + b are formed by adding pairs of
coefficients of the previous power, just as in Pascal’s triangle. This is illustrated
below.
(a + b)(1a® + 3a*b + 3ab* + 1b*) = at + 3a°b lee + ab?+
ab|+3|a*b? + 3ab? + b*
=> (a+b) = ---[+6]a7b?---

The result for (a + 6)” can be extended to any binomial expression.

Example 1
Expand (2x — 3y)’.

Solution
((2x) + (—3y))? = 1(2x)? + 3(2x)?(—3y) + 3(2x)(—3y)* + 1(—3y)?
= 8x? — 36x’y + 54xy* — 275°

Exercise 1
1 Expand:
(ay(atb) ~—(b) (—ayY (c) Geto)" id) re),
2 (a) Expand (a+b) and (a—b)?.
(b) Show that (a+ b)’ + (a— b)> = 2a(a* + 3b’).
(c) Find a corresponding result for (a + b)? — (a— b)?.
3E (a) By pus p =a+b and q =a-— b in question 2(b), find the factors of
p+qd.
(b) Hence, or otherwise, factorise p*® — q°.

4E By writing 11 as 10 + 1, explain the pattern of powers of 11.

 BINOMIALS 397

5.3.2 Binomial coefficients

If a binomial expansion such as (a + b)’ is expanded to give
1a’ + 3a’b + 3ab? + 1°, then the coefficients are referred to as binomial
coefficients. (For example, the binomial coefficient of a*b is 3.)

; n
The notation
i )is used for binomial coefficients, where and r are the rows

and columns of Pascal’s triangle as indicated below.

0 1
1 al
Z es 1
3 Vel t3
4 1 at 6 4 -1
5 tle ee) 10. <lOe eo

(5 5
So, for example, =o, = 10 and so on.
\o 3

()-0).2)-C ane
It is also useful to notice the symmetry of the table, so, for example,

It is sensible to link ” to the power of the binomial expansion as follows.

1

line number 1 (a+b)! has.

line peer 2 (a+b) Loe
line number 3. (a +b)° tage nk aaa
line number 4 (a + b)* 1456 4°31
The top line therefore corresponds to n = 0.

Similarly, the first term in each row corresponds to the coefficient of a”b° and is
therefore the term corresponding to r = 0.

It is easy enough to write down the first few lines of Pascal’s triangle. However,
if you want the 20th line it is a hard task to write down the preceding 19 lines!
The following questions develop a general formula for the binomial coefficients.
398 MATHEMATICAL METHODS

@ x
You can always write down the first two terms of any line of Pascal’s
triangle. The 10th line certainly starts 1 10, but it is not immediately
obvious how to write down the next term unless you already know the 9th
line.

1 The4thlineis1 4 6 4 1. Each number is related to the previous

number as shown below.
x4 x reo Ne colt % Dl

1 ee ee. Prine ait ae 6 pe Lam it4 SS 1

How are the multipliers 4 and 3 related to the multipliers 3 and 2?

2 (a) Find multipliers in a similar form for the Sth line of Pascal’s triangle.
(Express the multipliers with denominators 1, 2, 3,4 and 5.)

(b) What patterns do you notice?

3 (a) Use the pattern you have found to generate the 6th line.
(b) Check that your result is correct by using the Sth line to generate the
6th line in the usual way.

4 Use the pattern of multipliers to generate the 10th line of Pascal’s triangle.

5E Add up the terms you have generated in question 4. How does the sum
act as a check that the terms are correct?

6 Find the first four terms of the 80th line of Pascal’s triangle.

The next step is to consider how any individual binomial coefficient can be
found independently of any others.

Consider how the method of multipliers is used to generate the 12th line.

xi x 2 xP <<
jes 2 Be Pa De ze
1x2 Xe exes TexO2e
asi ae:

= 2 = 66 = 220 = 495

Usi n : f2 12 12 2S 1
sing notation, ; eine : S pieaae:
r x
DID e1tcei0
Bion “ithe sae san
 BINOMIALS 399

12
7 Ina similar way, write down (é ):

Results such as these can be simplified using factorials. For example

4 x 3 x 2 x 1 is written 4!, which is read as ‘4 factorial’. Most calculators
have keys labelled x!

8 (a) By writing out the factorials and cancelling, explain why

12!
ST ate ee UE Saas

i 12!
(b) Show that =—.,
5 5!7!

12
(c) Use your calculator to evaluate (é i:

12 12
9 Use factorial notation to explain why Deh hash
/

We 12 12
10 Evaluate: (a) ( ) b) (a (c) ee
y)

11 Suggest a formula, using factorial notation, for ( )in terms of n and r.

. . . . n .

12 12
12 (a) What are the values of (‘ and Cae

(b) How are these expressed in factorial notation, assuming that 0! has a
meaning?

(c) How should 0! be defined?

The notation n! (called 1 factorial) is used to denote

n(n—1)(n—2)x---x2x1
0! is defined to be equal to 1.
n!\
The binomial coefficients are then op ie hecy ie
r r\(n — r)!
For example,
(;) WA LI POKS KA XSL Ts 7X6.
=—_ = = =— wll
2 DSW DSS ASE SSK ISS BSS Se A A ea

Binomial coefficients may be found directly using some scientific calculators.

 400 MATHEMATICAL METHODS
i

The binomial expansion can now be summarised in terms of binomial

coefficients. The result is known as the binomial theorem.

5.3 Exercise 2
Rees 8 b 5 9 4 100
valuate: (a) : (b) 5 (c) 6 (d) 98

2 Expand (a+ b)’ using the binomial theorem.

3 Find the first four terms of the expansions of:

6
(a) (a—b)® ~— (b) (2a—3b)" =e) ae
“

ilk 15
4 If = find a.
4 a

| |
5 Evaluate: (a) peas ila (b) a = Be
80! 99! 20 19

10 9 9
6E (a) Show Show that
tha ‘ = + i ;

(b) Generalise the result in (a) and prove your result.

SSS
 BINOMIALS 401

5.3.3 Binomial series

When the binomial theorem is applied to the function (1 + x)”, the resulting
series is particularly useful and important. Sir Isaac Newton saw that this result
could be extended to powers other than positive integers. The binomial series
was Newton’s first major discovery, which he published in 1676 in a letter to the
Royal Society.

We can use the binomial theorem to obtain the expansion of (1+ x)” as follows.
n nN n
+5)" = 1" + ( Jacket (Glatee( Jie ete
1 3
n! a n!\ 2 a3 n! 3 a + n

(eee = eal). *3inss3)l- =

Se eG ——___—_ @, 6 18,

=1+nx+ = Marat
n(n—1) , _
n(n—1)(n—-2) Ns3 4-4"
There are m + 1 terms in the expansion.

A number of possibilities arise which you should now explore. Can you use the
binomial theorem when the value of n is not an integer (for example, (1+ x)?)
or when 7 is negative (for example, (1+ x)~)?

@ 3
1 (14x)? =143x4 3x7+°
(a) Use a graph plotter to plot the graph of the function (1 +x)’.
(b) Plot the function 1 + 3x, taken from the first two terms of the
expansion. What do you notice about the line that is produced?
(c) Calculate the values of 1 + 3x and (1+)? for x from 0.05 to 0.25 at
intervals of 0.05. What do you notice about the results?

1 + 3x is a linear approximation to (1+.x)* and you will notice from your

graphs and your numerical calculations that the approximation is good for
small values of x.

2 (a) Compare the graphs of the functions (1+ x)* and 1+ 3x + 3x’.
(b) Calculate values of 1+ 3x + 3x” for the same values of x as before
and compare them with the values obtained for (1 + x) and 1+ 3x.

1+ 3x +3x* is a quadratic approximation to (1+ x)’. This is a better

approximation than 1 + 3x for small values of x.
402 MATHEMATICAL METHODS

3. (a) Find a quadratic approximation to (1+x)*.

(b) Use a graph plotter to compare the graph of your quadratic
approximation with that of y = (1+x)*.

The binomial expansion of (1 + x)” can be written in the form

(1 x)= 1 bax ell) a eee

2! 3! coe
:
4 (a) By putting 2 = —1 in the series above, show that a possible quadratic
approximation to (1+x)' is1—x4+x’.

| (b) Use a graph plotter to compare the graphs of y = (1+ x)~' and
1—x-+.x* using a domain of —2 < x <2 (and —5 < y <5).
i For what range of values of x is the comparison a good one?

| 5 (a) By putting m = 5 into the binomial ERD anCION show that a possible
(ee
quadratic approximation to V/(1+ x) is 1+3x — 432°.
(b) Use a graph plotter to compare the graphs of ,/(1 +x) and
1+4x—- 1x? For what range of values of x is the comparison a good
i one?
|
6 Use the binomial series to show that 1 + 5x me Lye + tx is a possible
! cubic approximation to \/(1+~). cme your par using a graph

|
plotter.

Questions 4 to 6 suggest that the following is true for all values of n,

provided x is small.

| n(n — 1)x? a n(n — 1)(n — 2)x*

(14+x«)"=14+nx+
eee
a1 x

The result is known as the binomial series.

7E Further evidence to support the use of the binomial series may be found
by summing a geometric series.

(a) Find the sum to infinity of 1—x+x*—374.-.

(b) Expand (1+ x)~' from the binomial series. How does your answer
relate to the sum in (a)?

8E Work out (1+ 3x —+x*)* and comment on your answer.

 BINOMIALS 403

You have seen that the binomial expansion appears to generalise to values of n
which are rational and/or negative, though with the restriction that the result
only works for -1 <x <1.

The following result, although not proven here, is always true.

Example 2
y) 4 40
Show that ,/(1 8) Aaa 5” - a”

Solution
1 dee ye eee
(1- 2x)3 =1+ 3 (24) + ee3) (—2x)? _ (3)( ah 3) (=2x)? +

Bigalll ae
tls5° he Acs
Te SF

The result is valid for -—1< —2x <1

=> -}<-x<}
=> -lex<§

Example 3 1
Show that, for small values of x, -_—, © 1 — 2x
(1+ x)

Solution
1 ; =) (2)(=3) 2
Se ae ANaie) Ra AN ae acreme iOo

= 1—2x (where x* and higher powers

of x are ignored)

5.3 Exercise 3

1 Use the formula for the binomial series to expand the following as far as the
term:in x’.

(a) (1+x)) (b) (1+x)°

404 MATHEMATICAL METHODS

x)”. (There is
2 Use the laws of indices to write the following in the form (1+
no need to expand the functions.)
il 1
() Ya+x) @ res)

3 Expand the following as far as the term in x.

1 iD
(a) (ieee): (b) f(l-2x) — (e)
J(1 +x’)

4 (a) Show that \/(9 — 18x) = 3,/(1 — 2x).

Ee ar raiey
eli isy.: for
(b) Hence show that \/(9 — 18x) © 3 — 3x — tanita
1 1
any << 6G 7

5 Find the first three terms of the series expansion of:

1
(a) J/(4 :+ 4x) (b) (3 + 3x)?
x x x
6 The binomial expansion for /(1+ x) = 1+ se a je

(a) Why would it be incorrect to conclude that

49 49° 49°
J/50 = /(1+ 49) 2
=1+—- 8 cs 16
2
1
(b) Show that ./50 = 74](:
ae 3) and hence find an approximate value for
/50 using the binomial expansion for ,/(1 + x).

7E Einstein’s theory of relativity predicts that if a stick of length / moves with

2\2
velocity v in the direction of its length it will shrink by a factor (:— “)
where c is the speed of light.
2
(a) Show that for low speeds this factor is approximately 1 — wee
c
P c . : : :
(b) Hence show that if v = 3 the stick shrinks to approximately 94% of its
original length.

5.3.4 Error and relative error

Much of science and engineering is dependent on measurement. Since all
measurement is subject to error, it is important to understand how to deal
with these errors when making calculations. You may have met some of these
ideas before — the following example will serve as a reminder. Later, you will
see how the binomial theorem can extend your understanding of the theory
of errors.
 BINOMIALS 405

The accuracy of a measurement depends on the instrument used for measuring.

For example, using a ruler you might measure the length of a line to be between
43 and 44mm. This can be written as 43.5 + 0.5 mm.

Example 4
The measurements of the length and width of a rectangle are as shown.
26.5 + 0.5mm

oy 13.5 +0.5mm

Vo,
(a) What are the greatest and least possible values for
(i) the perimeter (ii) the area?

(b) (i) Express the perimeter of the rectangle in the form p te.
(ii) Express the area in the form a +e.

Solution
(a) (i) Minimum perimeter = 2(26 + 13) = 78mm
Maximum perimeter = 2(27 + 14) = 82mm
(ii) Minimum area = 26 x 13 = 338mm?
Maximum area = 27 x 14 = 378 mm”
(b) (i) Perimeter = 80+ 2mm
(ii) Area = 358+ 20mm?»
Note that the perimeter is found by adding four numbers, each with an ‘error’ of
0.5mm. The error in the perimeter is then found to be 4 x 0.5 = 2mm.

However, multiplying numbers with errors is not as straightforward. This is

investigated later, where the idea of relative error is introduced.

Expressing a result in the form a +e is a way of stating that the result lies
between a — e and a+e. The ‘error’, e, measures the largest possible difference
between the actual value and the number a.

Example 5
An isosceles triangle has perimeter 72 + 1.5 mm
and base 18 + 0.5 mm.
a a p=2a+b

Express a in the form ate.

406 MATHEMATICAL METHODS
SES a
ee

Solution
2Za=p—b
Maximum value of 2a = 72 + 1.5 — (18 — 0.5) = 54+2mm
Maximum value of a= 3(54+2) =27+1mm
Minimum value of 2a = 72 — 1.5 — (18+ 0.5) = 54 -— 2mm
Minimum value of a= 3(54—2) =27—-1mm
Cheep
besens IqqNit)

In example 5, the error in p — b was found by adding the errors in p and b. The
error in a was then found by halving the error in 2a.

These are particular cases of the following general result.

We now investigate the effect of multiplying and dividing measurements.

@ 2»
1 A rectangle has area 350 + 10 mm” and base 14 + 0.5 mm. Find the greatest
and least possible values for the height /. Hence express / in the form a +e.

When measurements are multiplied or divided, the connection between their

errors and the error in the resultant is by no means obvious. However, the
binomial series can be used to show that simple connections are possible
when the original measurements are in the form 1 +e.

2 (a) Use the binomial series to show that, for small r,

(147)? 2142r
il
x1-r

(b) The length of the side of a square is 1+ 0.05 m. Use the results of
part (a) to find approximate bounds for
(i) the area of the square,
(ii) the reciprocal of the length of the side.
 BINOMIALS 407

More generally, any two numbers of the form 1+ e can be os or

divided easily as is shown in question 3. ~

3 (a) Show that (1+7)(1+s) © [1+ (r+s)] if rs is small enough to be

ignored.

(b) By expressing =sas (Q+r\(1+s)", show that

pau ~1+(r+s)
1#s

4 Express the area in question 1 as 350(1 +) and the base length as

14(1 + 3). The height, /, is therefore =

a pa
on jee 5)
Use the result from question 3 to write he height, ) mm, in the form h +e.
~ How does this compare with your answer to question 1?

Unlike absolute errors, relative errors combine in a straightforward way when

measurements are multiplied or divided.

Example 6
A piece of wire, length /cm, is bent to form three sides of a rectangle.

a a l=2a+b

If ]= 20+0.4cm and b = 8+0.2 cm, calculate:

(a) the value of a (b) the area of the rectangle

 408 MATHEMATICAL METHODS

Solution
(a) 2a=l1—b

2a = (20+ 0.4) — (8 £ 0.2) = 12 + 0.6

4=—6-- 03 em
(b) ab = (6 + 0.3)(8 + 0.2)
= 6(1+0.05) x 8(1 + 0.025)
~ 48(1 + 0.075)
~ 48 +3.6cm?
————

5.3 Exercise 4

1 (a) Calculate the area of the rectangle in example 6 (to 1 decimal place) if the
wire is bent so that b = 19+ 0.2.cm.
(b) Use the idea of relative error to explain why the calculation is so
inaccurate.

2 Calculate the height of the trapezium shown below if the area is 125 + 2.5 cm’.

8+0.5cm

17 +0.5cm

3 In a mechanics practical, a ball (travelling at a constant speed) is measured as

travelling 50cm in 1.32 seconds. Calculate its speed if the student can measure
distance to the nearest centimetre and time to +0.1 second. (Give your answer
in cms ' to 1 decimal place.)
; ; BINOMIALS 409
410

Mathematical methods
Miscellaneous exercise 5

1 Use the sine and cosine rules to find the remaining sides and angles of
triangles PQR and XYZ, given that

(a) PQ=6.1cm, QR = 7.4cm, angle QPR = 43°

(b) XY = 19km, XZ = 33km, angle YXZ-= 27°

2 If the angle a is obtuse and cos 2@ = i calculate cos @ without using a

calculator.

3 If sin @= cos 26 show that 2s” + s — 1 = 0 where s = sin@. Solve the equation
for 0 <0 < 2n.

4 If 2sin? ¢ = Scos¢@ — 1, deduce a quadratic equation for cos ¢ and find all
solutions in the domain 0° < d < 360°.

5 (a) Without using a calculator show that sin 15° = (esi

Ps,
(b) Use a double-angle formula to show that cos 225° =4,/(2 + /2).

6 Find the vector equation of the straight line

(a) through the origin and the point (3, 2, —1),

(b) through (6, 1, 3) parallel to the straight line

in part (a).

7 Write down a vector equation

(a) of the line AB

(b) of the plane ABC

where A is (1, 2, 3), B is (4, —5, 6) and C
is (7, 8, —9).
Does the point (12, —8, —6) lie in the plane
ABC?

8 Find the equations of the planes through

the given points and parallel to the
given planes.
(a) 2x —3y—z=3, Cle Ox} (b) 4x — y = 6, (5, 1, —2)
 MISCELLANEOUS EXERCISE 411

So Find the angle between the lines

(@) 3x +y=2 and 4x%-—3y=1

(b) Sx+12y+3=0 and 12x+S5y=11

10 Find the angle between the planes

(a) 3x+6y—2z=3 and 8x-—4y+z=1

(b) 2x+3y—z=0 and x-—y-—z=4

11 Use the binomial theorem to find approximations in the form of cubic

polynomials to the following functions for small values of h:

(a) near b) J+h) ~() V2)

12 (a) Writing /(4 +4) as 2,/(1+ 4h), use the binomial theorem to find a
cubic polynomial which approximates to \/(4 +4) for small h.
(b) Evaluate \/4.2 to three decimal places.
(c) Writing /15 as \/(16 — 1) = 4\/(1 — #) and using the binomial theorem,
find ,/15 to three decimal places.

13 Simplify:
sin’ 0
(a) cosGa+6),
1 sin(a—@),
i — cos(@+7) (b) Fake)
Rae g
. tan 20 — tan0
(c) sin 36 cos 6 + cos 36 sin 0 (d) een Db ca

(a) Sketch the curves y = tanx and y = 2sin2x on the same diagram for
14
Behi ae eS iite

(b) Show that at the points where they meet, apart from the origin,
cos’ x = 4, and determine the coordinates of these points of intersection.
412

Calculus methods 1
.1 The chain rule

6.1.1 Functions of functions

You already know how to differentiate polynomial functions such as
x° + 2x” — 3 and other simple functions such as sinx and e*. Here we extend
these methods to more complicated functions. We consider composite functions,
which you can think of as functions of other functions. Previously you have used
function notation to represent such relationships.

Example 1
If f(x) = 3x +7 and g(x) = x’, find fg(x) and gf(x).

Solution

fg(x) = £(2’)
= 3xecb 7
gf(x) = g(3x +7)

= (3x 4-7)
We now examine the rates of change of composite functions. Initially, only linear
relationships will be considered.

1A

A rod with initial temperature 50°C is being heated so that its temperature
increases by 2°C per minute. What is C, the temperature in degrees celsius,
after ¢ minutes?

To convert from degrees celsius to degrees fahrenheit, multiply by 1.8 and

add on 32. Express F, the temperature in degrees fahrenheit after t minutes, in
terms of C and then in terms of t.

dF dC
What are av de and ra Can you find a connection between these rates of
change? Think of other examples which involve two linear functions and see
if there is a similar relationship.
 THE CHAIN RULE 413

If z is a function of y and y is a function of x, then the relationship

is easy to prove for linear functions such as z = my +c and y= nx +d.

Z—
ny +

Then z=m(nx+d)+c

=> z=mnxt+mdt+c

Eye ga bale

dzu) dz vedy
and so eae

This relationship is called the chain rule.

If z is a linear function of y and y is a linear function of x, then to find — it is

dx
not in fact necessary to first express z in terms of x. This would be an especially
useful result if the chain rule were true for non-linear functions as well. It may
very well extend to non-linear functions; as you saw earlier, if you zoom in at
any chosen point on a locally straight function, then it will look increasingly like
a linear function. You might, therefore, expect the chain rule to be true for any
locally straight functions.

For example, y = sin x? is a composite of the locally straight functions y = sinu

and#=x°.3
414 CALCULUS METHODS 1
rr
rrr

If the chain rule does work for non-linear functions then

d
LN where dy = COs, Sand ages
Hig dae dx du dx

== (cos u) x (3x”)
dx
= (cos
x*) x (3x’)

= 3x” cos x’

The example above shows that if the chain rule works, then

f(x) = sinx’® = f/(x) = 3x? cosx®

You can check this for any particular value of x by using a numerical method for
differentiating the function and comparing it with the above formula for the
derivative. Alternatively, you could use a graphical calculator or a computer to
numerically differentiate the function for several different values of x and then
plot the (x, f/(x)) points as a graph. You can then superimpose y = 3x* cos x°
and check that the graphs are the same.

1B

1 Assume that the chain rule holds for any locally straight functions and use
a it to find = if:

) y=wandu=sinx (ie. y = sin’ x)

(b)
i ae

(c) y=u*
and u=e*

2 Check the answers you have obtained for question 1 by a numerical

M method. (When you check (a), note that x is in radians, of course.)

Example 2
Water being poured into a paddling pool spreads at such a rate that the area in
square metres covered after ¢ minutes is S = (5 + 4)’.

(a) Find the rate at which the area is increasing after 2 minutes by multiplying
out (5 + 4t)? and differentiating with respect to ¢.

Verds.
(b) Alternatively, let R = 5 + 4¢ so that § = R2 . Find ae by considering
Sead Rog a
ak x aa Check that your results agree.
 THE CHAIN RULE 415

Solution
(a) S=25+4 40¢+ 162
ds
= —- = 40 + 32
dt

eee ite
dt

ds dR
(b) ae = 2R, <a

Se soa 40 ae
di. de ede Sued os i
ds
When t = 2, R = 13 and so sy = 104;

Example 3

Find ou when:
dx

(a) y=e™ Yb) y= In (3x? +1)

Solution |
(a) y=e” es composite function of the two locally straight functions
y =e" and u = x’ +1. Using the chain rule,

iy dy , ae
du
dx du dx
wes 2x

=> ERI
dy x +1
cP 28

(b) y = In (3x? + 1)

Let y=Inu where

u = 3x* +1

dy dy | du
dx du dx
it
=X OX
u

sid d 6
are
 416 CALCULUS METHODS 1

Expressions such as (x* + 3x)* and sin (x”) can be differentiated rapidly once the
stages of their composition have been recognised.
ay)
For example, to differentiate (x* + 3x)*, let y = u*, where u = x* + 3x.
dy dy du
Ne ‘dx du “aor
dx

The derivative is therefore

4(x? + 3x)> x (derivative of x” + 3x) = 4(2x + 3)(x? + 3x)?
To differentiate sin (x7), let y = sinu, where u = x’.
The derivative is therefore
cos (x*) x (derivative of x”) = 2x cos (x”)

6.1 Exercise 1
1 Find ay for each of the following by (a) using the chain rule and
dx
(b) multiplying out the brackets.

Qy=(@4+1" Gi) y=Qx—1" iy Ga

Show that your answers agree in each case.

2 Find = for each of the following. You do not need to multiply out the
x
brackets in your answers.

(a) y=(x7+3)' = (b) y= (5 + 2x)

(c) y= (2x* = 3x)° (d) y= (x? — 3x*)4

3 Differentiate each of the following. Hence find the gradient of each graph at
the point (0, 1).

(a) y = cosx* (b) y=sin2x+1 (c) y=e*

4 Differentiate: (a) cos x? (b) sin? x (c) 2cos*x

5S You have already discovered by numerical methods that the derivative of

sin 2x appears to be 2 cos 2x.
It is possible to obtain this result using the chain rule and making a
substitution for 2x.
If u = 2x, then y = sinu. Write down gy and Ge
— and so find
2—
du dx dx’
6 Use the chain rule to show that the derivative of cos 3x is —3 sin 3x.
 THE CHAIN RULE 417
i

7 Use the chain rule to obtain an expression for the derivative of sin ax, where
a is any constant.

8 Differentiate:
(a) e* (b) sin?x (c) e&* —(d) 3cos2x (e) 2(x2 +1)

9 A balloon is inflated at a rate of 200cm? per second.

After t seconds, when the balloon has radius rcm and volume Vcm’, the
following formulas apply.

V = 2001 and) V =a"

, dV dV
(a) Write down is and ar

; dV dV dr
(b) By the chain rule, waiaaae x rs

’ : dr
Use this to work out an expression for a and so find the rate at which
the radius is changing when ¢ = 1.

10 When a hot-air balloon is being inflated, the balloonist finds that a good rule
of thumb is that after ¢ minutes the radius, r metres, is given by
r = 3+0.04t’. The balloon can be assumed to be roughly spherical.

d dv
(a) Work out expressions for = and for —.
dr
dV
(b) Combine these two expressions to find
dt’
(c) How fast is the volume increasing after 2 minutes?

11 An ice cube is melting, and at time t hours it has the form of a cube of side
3
xcm and volume Vcm’.

(a) Find ou in terms of x.

dx
(b) If x = 4 —0.5t, write down o and so find =

(c) At what rate is the volume changing when t = 2?

12 Differentiate: (a) sin? 2x (b) 3cos*4x (c) e°S*

 418 CALCULUS METHODS 1

6.1.2 Applications to integration

The derivative of sin 2x is 2.cos 2x. It follows that

|2008 2x ds = sin2x + ¢

and

|cos ax a = tsin2x +.

Being able to differentiate using the chain rule greatly increases the number of
functions you are able to integrate. You need to know what type of integral
function you are looking for.

Example 4
1
Find |(2x — 1)* dx.
0

Solution
First, try differentiating (2x — 1)”.
d
y = (2x 1) > = S(2x 1) x 2 = 10(2x — 1)"
1
So |10(2x — 1)* dx = (ox= 1)
1
= [cx —1)*dx= = (2s- 1)

= alIS- (1) =02

6.1 Exercise 2
1 Write down the integrals of:

(a) cos3x (b) singx (c) 2sinSx (d) oe

2 Find:
2 0)
(a) Ieo * dx (b) | sin 2x dx
=i

2 1
(c) |3 cos}x dx (d) |(Qx=3\4 dx
0 0
 THE CHAIN RULE 419
a eee ee ee eee

3 Work out the coordinates of the points A, B and C and then evaluate the
shaded areas.

d
4E If y = sinx’, YS Ose.
dx
az 1 2
Use a numerical method to check if |cosx* dx = |—sinx?] .
1 2m 1
Try to explain what you find.

5E Some of these functions can be integrated by the methods of this section.

Integrate as many of them as possible.
(a) cos3x bye (c) (x* -3)
(d) (Sx +3)* = (e) sinSx —()_ sinx?

: d : 2 2 : 2
GE Since — (sin x*) = 2x cos x", xcos x’ dx =4sinx 1G
dx

Find the following integrals.

(a) |xsin sds (b) ee ax (c) |2x(a0* + 1)* dx

1
7E Use the identity cos 2x = 1 — 2sin’ x to find |sin? x dx.
0
420 CALCULUS METHODS 1

6.1.3 Small increments

One application of calculus is the calculation of small changes in one variable
which are brought about by small changes in a related variable. For example,
suppose the radius of a balloon is increased by 1%; by what percentage is the
volume of the balloon increased?

Suppose that y = f(x) and x is

subjected to a small change 6x, that
is, it increases from x to x + 6x. This
dy
will cause a change in y, from y to
y + dy. You need to calculate the
change, dy, in y brought about by
the small change in x, as illustrated
in the graph.

a sel 6:
The gradient of the curve at P is = but is also approximately equal to = (for
small changes). ‘

or

Example 5
A sphere of radius 5 cm increases its radius by 0.1. cm as a result of heating.
Calculate the change in the surface area and the volume of the sphere.

Solution
Let A be the surface area and V the volume of the sphere.
A=4rr
dA | 3
drew

When r = Scm, a = 407

dr

So,
dA
6A & ($F) 6 = 40r x 0.1
r

= 4r
The area increases by about 47cm’.
 THE CHAIN RULE 421

Similarly,
Vi tn

- = 4nr°

When r = 5cm, ey = 1007

dr

So, 6V = (<= 1007

x 0.1
= 107%

The volume increases by about 107 cm’.

Example 6
It is suspected that there is an error of 1% in the measurement of the sides of a
cube. Calculate the percentage error in the calculated values of the surface area
and the volume of the cube.

Solution
Suppose the cube has side x, area A and volume V.
6
Aeiord aud = 001 (i.e. a 1% error in x) |
a3
ah =-125 oso. OAS sie Ox = bx
dx dx
Dividing by A (= x) 5

oF iy oe
i 6x2 aaa

=0.02 (ie. 2%)

There is a 2% error in the area.

V=x'

dV
——=3x* and 6V & 3x’*6x
dx
6V 6 6
eS 3x7 = ae
x x
oe) 0il

== )503

There is a 3% error in the calculated volume.

This confirms the result in section 5.3 that an r% error in x leads to an

approximate error of nr% in x”.
 422 CALCULUS METHODS 1

’ Exercise 3
—
1 A cylinder has a height of 10cm and a base radius of 6 cm.
in the volume of the
(a) If the radius increases by 0.1 cm, calculate the change
cylinder. (The height remains fixed.)
calculate the
(b) If the radius remains fixed but the height increases by 0.2 cm,
change in the volume.

mean
2 The time T taken for a planet to revolve around the Sun is related to the
distance, r, of the planet from the Sun by the formula

T=kr?

If the Earth’s orbit changed so that its mean distance from the Sun were to be
increased by 2%, calculate how much longer the year would become.

3 A gravel heap is in the shape of a cone with its radius roughly equal to its
height.
(a) The height of the heap is 10m. If 0.5 m> of gravel is added to the heap,
calculate the increase in height of the cone.
(b) If gravel is being added to the cone at a rate of 8 m°h_’, find the rate at
which the height of the gravel heap is increasing.

4 The radius of a spherical balloon is decreased by 1.5%. Calculate the

percentage reduction in its volume.

5 Each week a factory produces N thousand topsy-turvies. The cost of

production is reckoned to be £1000C, where C = (N* + 5). At a certain time,
N = 2 and N is increasing at the rate 0.1 per week.
dC
(a) Calculate aN: What does this represent?

(b) At what rate (in £ per week) is the weekly cost of production rising?

6.1.4 Inverse functions and x”

: : 2 i 3 :
You know how to differentiate x” but not the inverse of this function, \/x.
The chain rule enables you to find the derivative in such cases, using the
fact that

dx dy
dy * dx
This can be shown easily as follows.
 THE CHAIN RULE 423
rare
eS ee ee ee Ee ee eee

b
For the locally straight curve shown, Soa Pate graph of the inverse
Ghe
function ts a reflection of the graph of y = f(x) in the line y = x. Its gradient
ee
at the equivalent point in P will be =
d Since,
dy
in the reflection, the x- and y-coordinates are
interchanged,
dx a
dy b

Gan Cys eae,

H ence faGeode
a |6 oes

This result may be written as

dx me dy
dy dx

Example 7
Find the derivative of y = \/x.

Solution
x=y
da cya 3 at 1 dy tied = 4
=> —_ = 2 => = = — = — 2
dy : di 7 2y 2 2A/x ro dah O.

You already know that if 7 is a positive integer, then

d 8
ae ea :
dx
The working above shows that the rule is also applicable when n = 5. In fact the
result is generally true.

6.1 Exercise 4
1 2
1 (a) Show that the derivative of — is ——.
x x
(b) Find the derivative of (1+ x)/x.

2 Differentiate:

A 1
(a) /x (i.e. x3) (b) (c) x? (d) ie J/x
424 CALCULUS METHODS 1

Pe ae
Qe Find the derivative of x/x (i.e. x") with respect to x.
> If = Inx; then x =e”.
d heats
of
Write down a and use this to find 2 Hence explain why the derivative
1 y «
In x is —.
x
5 Use the chain rule to find the derivative of In2x. (Start by putting u = 2x.)
6 (a) Work out the derivatives of In 3x and In Sx.
(b) What is the derivative of In ax, where a is any constant?
(c) Use the laws of logarithms to explain the result above.
1
7 (a) Integrate mae
a

; er :
(b) What is the integral of — with respect to x, where a is any constant?
ax

8E If y = \/x then x = y’. Find ia and hence find pi

dy dx
Check that your answer agrees with that for question 2(a).

9E If y= sin 'x, then x = siny.

(a) Write down 3 and a

(b) Express the third side of the triangle illustrated

in terms of x. Hence explain why the derivative
Ofsin
= is —————— for, lex
v(1 — x*)
(c) By a similar method, find the derivative of cos 'x.
 ; THE CHAIN RULE 425

he

a
426

Calculus methods 1
22 Parametric form

6.2.1 Curves which vary with time

Consider the computer simulation of a game, where you need to describe the
position of a ball on the computer screen. The x- and y-coordinates, giving the
position of the ball, will vary with time t. The screen can be thought of as a
Cartesian (x,y) plane, with the origin at the bottom left-hand corner of the screen.
Suppose that the position of the ball at some time, ¢, is given by the equations
DEAN eeyim t
These equations are called parametric equations, and the time, t, which
determines the x- and y-coordinates, is called the parameter.

You can plot the position of the ball at various times.

af
5
Time t Oe sinensis
EG OF A Gr 8 al
y Oe ai PY Be

0
0 Oe,

The equation of the path of the ball across the screen is y = 5x. This may be
obtained by eliminating the parameter t from the two parametric equations.
i — ot a)
ar 1)
From@), ¢=
and so 7
> y= 7x

Most graph-plotting calculators will allow you to enter equations in parametric

form and to plot their graphs. However, you do need some practice at drawing such
graphs for yourself, and this is considered in the following examples. Later on we
shall consider differentiation when relationships are expressed in parametric form.

To plot a parametric curve, it is sometimes sufficient to ‘plot some suitable

points and join the dots’, as in the following example.
 PARAMETRIC FORM 427
a
enemeneenmeneermeree
ee. ey

Example 1
Plot the curve given by the parametric equations
Be aps op y=r tor
0 <f< 3

Solution

t ON OES 1 1.5 De Des) 3

OES aD a Oy a We eee
Nae Wi O25 1 e252 TAP 6.25. 29

In this example, the (x,y) graph appears to be parabolic. You could confirm this
by finding y as a function of x. This is done by eliminating the parameter ¢ as
follows.

x= or = —
2: 9

? 2 9
(y is a quadratic function of x. The graph is a parabola.)

This idea of eliminating the parameter to form the Cartesian equation is

considered again in section 6.2.2.

6.2 Exercise 1

1 (a) For the curve given by the parametric equations

x=20t, y=90-St

Complete the following table of values t OEE. =i26 tS eae s

and plot the points on an (x,y) graph. x

(b) On the same graph, plot the points which would arise if t were to take the
values —1, —2, —3, —4 and —5. You should not need to recalculate the
values — look for symmetry with your answers to part (a).

(c) What is the general shape of the curve?

 428 CALCULUS METHODS 1

2 By choosing suitable values of t and drawing up a table, plot the following

parametric curves using any properties of symmetry or general shape to obtain
the complete sketch.
2
@)x=2r, y=4r (b)
x = 2F, =

Check your results using a graph plotter.

6.2.2 Circles and ellipses

Pythagoras’ theorem can be used to show that the equation of a circle of radius r
centred on the origin is
xt+y=r @®

Introducing 0, the angle between OP and the x-axis, gives a different view of the
problem and leads to alternative parametric equations for the circle.

Consider a circle of radius 3.

You know that x” + y* = 9 is the

Cartesian equation.
P(x, y)
Using trigonometry in triangle OAP
gives the parametric equations

x=3cos#, y=3sin0

You can eliminate @ from these

equations to show that they give the
correct Cartesian equation for this
circle as follows.

x* = (3cos@)* =9cos*@ and y’ = 9sin? 6

x? + y* = 9cos* 6+ 9 sin?6

= 9(cos 6 + sin? 6)
x+y =9 (as cos’ 6 + sin? @ = 1)
The following questions develop the parametric form of the
equation of an
ellipse.
 PARAMETRIC FORM 429

1 Suppose a circle of radius 3 is stretched by a factor of 2 in the x direction,

so that the point P is transformed to P’ and the circle becomes an ellipse.

TQ’ is the major axis of the ellipse. SR is the minor axis.

(a) Write down the coordinates of P in terms of 0.

(b) Write down the y-coordinate of P’ in terms of 0.

Write down the x-coordinate of P’ in terms of 0.

Write down the parametric equations of the ellipse.
Write down the coordinates of R, Q and Q’.
Write down the area of:
(i) the circle, (ii) the ellipse.

Sketch, on the same diagram, the curves:

(i) x =4cos#, y=4sin@ (ines = Scos@, y= 47sind

The ellipse in (ii) can be obtained from the circle in (i) by means of a
one-way stretch. What is the scale factor for this transformation?

What is the area of:

(i) the circle, (ii) the ellipse?

3 Sketch the graphs of the following curves, indicating the lengths of the
major and minor axes.

(ajex = Sicos0, sy = 6 sit (b) = ="2cos'?, -y =sin 8

(Cex acos0, y= )sind

430 CALCULUS METHODS 1
SS
Oe

n
4 (a) Complete the following argument which leads to the Cartesia
equation of an ellipse.

x=acos0, y=bsin@

=> €0S.0 = sin@ =?

So, since cos” 9 + sin’ @ = ?, it follows that

2 2 ? 2

Cele
A 2
x y
or oad yh

2 2

(b) Sketch the ellipse > Be_ = il,

2, 2
%
5 Consider the ellipse a + 5 =n

(a) This ellipse can be obtained from the circle x” + y* = 1 by means of a

two-way stretch. What are the scale factors of this in the x and y
directions?

(b) Write down the area of the circle x” + y> = 1. Hence write down the
area of the ellipse.
 PARAMETRIC FORM 431

».2 Exercise 2
1 Copy and complete the following table.

Ellipse Cartesian equation Parametric equations Area

Dee? :
menace
A —+—=1
; y - 16
B X= 3:C0s 0,1) i=
5 sing
2 2
x y
Cc ——+——=1
0.25 4 0.16

ay
2 Rewrite 9x* + 4y” = 36 in the form Sez Bo 1. Hence write down the
a
parametric equations and the area of this ellipse.

3 Repeat question 2 for the ellipse with equation 4x” + 25y” = 100.
,

.2 Tasksheet E1 -— Drawing parametric curves (page 562)

5
432 CALCULUS METHODS 1
eee
ee es A

6.2.3 Conversion
You have seen how to convert the parametric equations

x=acos0, y=bsin@

into the Cartesian equation

iS) i)

S|
R
=z= 1
by using the trigonometric identity cos’ @ + sin’ 6 = 1.

We now introduce two other trigonometric identities which are sometimes

encountered in work on parametric equations.

@ »
You know that

sin@ = —
G b

a
cos@= —
Cc
b VAN |
fang. — a
a

There are three other trigonometric ratios which are sometimes used. They
are secant (usually abbreviated to sec), cosecant (cosec) and cotangent (cot).

In the triangle above,

ve bekC a
sec Ais
0is -,— cosec 6Ois is —D cot@ isis —b

1 Show that:

il
(a) secO = (b) cosec@ = ae.
cos 0 sin 6
1
(c) cot? = (d) cot? = cose
tan@ sin @
 PARAMETRIC FORM 433
KE

2 The diagram shows the graph of cos 0

for 0 < 0 < 27 and part of the
1
graph of sec @ = —_.
cos @
Copy the diagram and complete the
graph of sec 0.

3 (a) Sketch on one diagram the graphs of sin @ and cosec 6.

(b) Sketch on one diagram the graphs of tan 6 and cot 0.

4 You know that, for any value of 0,

sin? 9 + cos” = 1 @) |

(a) By dividing each term of equation @ by cos” 6, show that

tan’6+1 = sec’ @)

(b) By dividing each term of equation @ by sin* 6, show that

1+ cot? @ = cosec’ 6 @

If x = 2sec0 and y = 3tan 8, identity @) above can be used to write x in

terms of y.

secO = 5 and ran@ =2

Substituting in Q), »
2 2
y x
2 1=[(=
(2) it &)
D) 2
y x
=>—+1= —
9 4
=> 4y* + 36 = 9x"

5 Use identity @) to convert the parametric equations

= cot. and! 2y = cosec?
into the Cartesian equation.
 434 CALCULUS METHODS 1

You have used trigonometric identities to remove parameters from pairs of

simultaneous equations. In general, conversion from parametric to Cartesian
form involves the removal of a parameter using simultaneous equation
techniques. While these may involve indirect elimination using trigonometric
identities, simpler direct methods are often used. This is illustrated in example 2.

Example 2
Find the Cartesian equation of the curve given by the parametric equations
07 ®

y = 90 — 5¢* ®

Solution
Using equation © to find ¢, t= a
x \? x?
Substituting for t in Q, = 90 — 5{| —{|—= 90 — —
2 @ : 20 80
a et toe 9 Die Obote tn ite OOS Wate eR a fee 2oe
6.2 Exercise 3

1 Find the Cartesian equations of:

(a) x=2-—-3f, y=4—52 (bite 3r aces ee
t
(Cie 47, y = 10t — S¢* (di = le y=4-2t

2 Use appropriate trigonometric identities to find the Cartesian equations of:

(ard == sec0, 5= tand (b) x =4sint, y=3cost
1 38
(c)) x = cosec@, P= coe) (d) 3 = sec, y=tan@+1

3E (a) Obtain the Cartesian equation for the parametric equatio

ns
x=1+V7t, y=44+/t
(b) The curve given by the Cartesian equation is not quite
the same as the
curve given by the parametric equations. Why not?
 PARAMETRIC FORM 435

4E For the curve given by y

VE ee Dyaaee \v
: z 10
suggest two different possible sets of
parametric equations. :
6
4

6.2.4 Differentiating parametric equations

You have used the chain rule to find rates of change in cases where there were
two equations involving three variables.

For example, if y = sin@ and 6 = 3x* + 2, then

dy_ dy a
dx d@° dx
= C0Si0) 16x

= 6xcos@

= 6x cos (3x” + 2)
Parametric equations can also give two equations with
three variables. For example, a circle with centre the
origin and radius 3 units has parametric equations
<= 3:c0s0,, y= 3sing
d
You can find = using the chain rule.
x

dy _ dy | do
dx dO dx
‘ dy dx
From the parametric equations, you can write down 70 and @ but not ae
d d
However, you should remember that — = 1 + ae
dx dé
’ dys fdas dx
This means that the chain rule can be rewritten as ax db dé

Using this result for the circle x = 3cos0, y = 3 sin,

dy _ ae
ap 08? and 707 —3sin0

Then es ae, ae eae! 3 cot@

dx iode 9 3600
436 CALCULUS METHODS 1
NN

Example 3
dy and the
For the curve defined parametrically by x = 10t, y= St’, find ——
dx
:
equation of the tangent to the (x,y) graph at t = 3.

Solution
d d
Differentiating, “ = 10) and = = 107

dye cpa
dxe lO
dx pad? “di 210

When t = 3,
dy =e = 50) aranidies yyon
dx

The equation of the tangent is zs =3 (or 3 y— 3445.

be SU)

The process of differentiating expressions given in parametric form is known as

parametric differentiation. It may have occurred to you that you could have
found some of the gradients by first converting the parametric equations to
Cartesian equations. This may occasionally be an easy method. For example, if
eto On

you may spot at once that y = 3x, so — = 3.

dy
dx

Generally, however, parametric differentiation will be quicker.

Example 4
A curve is defined by x = 2t+1, y=?’.

(a) Use parametric differentiation to find the gradient at the point (5, 4).
d
(b) Find y in terms of x and so write down = and find the gradient at the point
(5,4). %
Solution
dy dx dy 2¢
(a) wy),
a t and —_
7 = 250 —-
mS.= — = t

When x = 5 and y = 4, t = 2, so the gradient is 2.

(b) Sincere= 27 = (x — 1)
2
Gat) o; dy _2%(x-1)_ («-1)
50)—
4 dx ey
4
When x = 5 the gradient is a 2.
 PARAMETRIC FORM 437

Even with this simple conversion, it is clear that parametric differentiation leads
to a quicker solution, and there are many cases where the conversion to
Cartesian equations is difficult or impossible.

6.2. Exercise 4
: 1 ; dx d
1 A curve has parametric equations x =t,
y= a Waltetown sand and
dt dt
d
hence find = Calculate the gradient of the curve at t = 2.
%
A curve has parametric equations x = 3cos#, y= 4sin@.
Work out the Cartesian coordinates and gradient at the point where 0 = 37.

d
For the curve defined by x =4u, y =u’, find oaand the equation of the
b
tangent to the curve at the point where u = 2.

Find the equation of the tangent to the curvex =u’, y = 2u? at the point
where u = 1.

For the curve x =2v, y=v> —3v

d
(a) work out ae
die
(b) write down the two values of v for which w =.0;
dx
(c) write down the x- and y-coordinates of the turning points on the curve.
d <<
Given that x = 3s and y=’, find first by parametric differentiation and
dx
then by conversion to a Cartesian equation.

Check that both methods give the same value for the gradient of the curve at
$= I;

For each part of this question, find e by the method of your choice.
dx
=(t+2)*, y=r-3
on y= br =A
=2cos6— sing,” y— 3 sin’
= sin26,. y= sine

A curve has parametric equations x = +4 and y = 2t + 4t.

2
(a) Find a in terms of t and show that a, > 24.
dx dx
(b) Sketch the curve.
438 CALCULUS METHODS 1
pin ke Lr ee ee ee

9E Find the equation of the tangent to the curve x = 6 —cos@, y=sin@ at the
point where 6 = }. Write down the coordinates of the points A and B at
which this tangent cuts the x- and y-axes respectively and hence find the area
of the triangle OAB.

6.2.5 Velocity vectors

The ideas encountered in this section so far may be usefully employed in
describing the motion of a body. We have used parametric equations to describe
the position of a body at various times; here we use the method to obtain the
velocity vector.

Consider an ice hockey puck, moving across the surface of the ice with constant
velocity. Its position at half-second intervals is given (in terms of its position
vector at that time) in the table.

Time, t seconds 0 0.5 1 1S) Dy

Position vector, metres (from 0 6 12 18 24
origin in corner of pitch) 6 10.5 15 19S 24

The graph shows the motion of the puck

and the arrow indicates its direction.

0 12 24 x
Starting from x = 0, the puck moves 12 m in the x direction every second. At
time ¢ it will have moved 12¢m in the x direction (i.e. x = 12t). Similarly,
starting from y = 6 on the y-axis, it moves 9 m every second in the y direction
Ge. = 6 198),

So the velocity of the puck isaba:| ms bs: ' and its speed is the magnitude of this

vector, which is given by \/(12*+ 9°) = 15ms7!.

The parametric equations for the position of the puck are

x=12t, y=6+9% -or r=

12t |
Oe es
 PARAMETRIC FORM 439

By differentiating, you can obtain the rates of change of the position of the puck
along the x- and the y-axes, and hence obtain the velocity vector. ~

x=12t, y=6+4+9%
dx
Differentiating, os 12 is the rate at which the x-coordinate is increasing, a
constant 12ms~! (the speed at which it is going
across the pitch)
d
and - =9 is the rate at which the y-coordinate is increasing, a
constant 9ms_~ (the speed at which it is going up the
pitch).

Notice that these are the components of the velocity vector in the directions of
the x- and y-axes. That is,
dx

(oe a then v= “ = fa
y dy y
dt
The notation x indicates differentiation with respect to time.

The gradient of the velocity vector is 2s as illustrated on the

be
diagram. This may also be obtained using the chain rule. v A

de) idx. dt a ak” dt x. x

While the example considered above is one of motion in a straight line, the
results are also true for non-linear motion.

Example 5
A particle moves so that its position vector is given by
2b 2
r
“|e +42t
Plot the positions of the particle over the first three seconds and sketch its path.
Calculate the velocity of the particle and mark the velocity vectors on the graph
att —0 and t= 2.
 440 CALCULUS METHODS 1
- ie oh Re ey ES Ee

Solution

i 0 1 2 3
2 4 6 8
i 0 3 8 15

ee
Oneifferentiating, iley| lees y
v= = |ae+2

ey]
When t = 0, v= [3]. When t = 2, lA

Notice that the velocity vectors are in the direction

of the tangent to the curve. The lengths of the
arrows indicate their magnitudes.

6.2 Exercise 5
2
t ; :
1 For the position vector r = a , find the coordinates of the points when

t = 0, 1, 2 and 3, and sketch the path. Differentiate to find the velocities when
t = 0, 1, 2 and 3, and calculate their magnitudes. Mark each velocity on the
curve using an arrow of appropriate length and direction.

2 The displacement in centimetres from the origin of a particle after t seconds is

given by

r=
era
3t—
(a) Find the velocity after t seconds.

(b) Calculate its initial speed and direction (i.e. when t = 0).

(c) Find at what time it is travelling in the direction fT‘

(d) When is it travelling parallel to the x-axis?

3 A particle moves along a straight line y = 3x + 1 with a constant speed of /10

units. What is the velocity vector?

If the particle starts at (0,1) write down the position vector at time t.

ee EEEEESsFSFMMFsFsSsSmsSSSSsSsSsSs—
PARAMETRIC FORM 441
442

Calculus methods
3 Further differentiation
mr

echniques
6.3.1 The product rule
It is always possible to estimate numerically the gradient at any point of a locally
straight curve and you also know how to work out the gradients of many such
curves algebraically.

You know how to deal with functions of functions (like sin x”) by using the
chain rule.

You also know that, to differentiate compound functions which have been
obtained by addition or subtraction (like x” — sin x), you merely add or subtract
the separate derivatives.

It is unfortunate that derivatives of products (like x sin x) cannot be dealt with by

multiplying the separate derivatives.

The dae shows the graphs of the two linear functions y = 1 — x and
y = 3x — 2, together with the graph of their product, y = (1 —x)(dx—2).
It is Ree. that the two linear graphs have gradients —1 and 45 respectively for any
value of x. However, the cee o the product graph is ae and so
cannot have the value —1 SS5 Stor every value of x.

You could, of course, differentiate the product function by first multiplying out
the PEELS; but this method will be lengthy for functions like
(2 + 3x)’(3 — 2x)? and it is not possible to ‘multiply out’ a product like x sinx.
It would therefore be very useful to find a formula for the derivative of a
product.
 FURTHER DIFFERENTIATION TECHNIQUES 443

Example 1
Let
y = uv, where u = ax+bandv=cx-4d.

d d d
Work out = and show that it is equal to = - ie

Solution
y = (ax + b)(cx + d) = acx* + adx + bcx + bd
d
SE cet aie
dx
dv du
th lt (ax
+ b)c+ (cx +d)a = 2acx
+ ad + be

So, for a function y = uv, where uw and v are linear functions of x,

dy du dv
dx dx dx
This rule is called the product rule. It would be of limited use if it could only be
used for products of linear functions. Here, we consider its use for other functions.

Any function which is differentiable has a graph which is locally straight. Since
the product rule can be proved to be true for products of linear functions, you
would expect the rule to be true for any two differentiable functions.

The following questions provide some evidence that the product rule works for
any two differentiable functions.

6 A

(a) Use the product rule to find the gradient of y = x* sinx at x = 1 and
check that your answer seems reasonable in view of the shape of the
graph of y = x’ sinx.
; : De
(b) Use a numerical method to evaluate the gradient of y = x“ sinx at
x = 1 and check that it agrees with your answer to (a).
 444 CALCULUS METHODS 1
ee

2 (a) Use the product rule to obtain the equation of the gradient graph for
ed
y = uv where u = cos; x and w= 4x —=3x°.

(b) Use a program for numerical gradients to check your answer to (a).

3 Repeat question 2 for any two functions you choose.

‘ 2
4 (a) x° can be written as x x x",

Use the product rule with u = x? and v = x” and check that you do
obtain the derivative of x’.

(b) Write x® as a product in at least two different ways. In each case,

differentiate using the product rule and check that you obtain 8x".
(c) nt tO = x4 x x

Differentiate x” x x° using the product rule. Do you obtain the

expected answer?

You have seen some evidence for the following result.

6.3 Exercise 1
1 Use the product rule to work out the derivatives of:
(a) e* sinx (b) x e* (c) x cos x

2 Work out the gradients of:

(a) the tangent to y= xe" atx =2

(b) the tangent to y = 2x*e* atx =1

3 Use the product rule to differentiate x sinx and hence work out the
equation of the tangent at x = 1.5 on the graph of y = xsinx.
(Work to 2 s.f.)
 FURTHER DIFFERENTIATION TECHNIQUES 445

4 A rectangle on a computer screen has width w, height / and area A.

w and h are programmed to be functions of time, f.

(a) (i) Ifw =? andh = sint, use the product rule to find 2 and so work

out the rate at which the area of the rectangle is increasing when
pale

(i) What is happening to the area when t = 2.5?

(b) (i) If w = sint and h = cost, how fast is the area increasing when
p= 05?

(ii) At what value of t does the area of this rectangle first stop increasing?

Find the equation of the tangent to the graph of y = 0.25xe* at the point
where x = 1. (Work to 2d.p.)

Differentiate x e* and so work out the coordinates of the turning point on the
graph of y= xe’.

(a) Differentiate x* e* and explain how this shows that the graph of y = x’ 2 e*
must have a stationary point at (0,0).

b) How do vfyou know that there is only y one other stationary yar point on the
graph? Work out the coordinates of this stationary point.

1
(a) Let y = uv, where u = x andv =-.
a
1 d
It follows that y = x x —= 1 and Sy
x dx
dy du dv
aay a te ds
du Pe, ibe 1
Use the above to find —. Hence show that the derivative of — is ——.
dx 5 x
dv 1 ,
(b) Show that the answer Ge gh Ete with the one obtained by using
x o
the nx”! rule.

9E Show that there is a stationary point on the curve y = x sinx when

x +tanx = 0. Show graphically that x + tanx = 0 has three solutions in the
region —3 < x < 3.

One of these three solutions should be obvious. Use any method you wish to
find the other two solutions and so work out the coordinates of the
stationary points of y = xsinx in the region —3 < x < 3.
 446 CALCULUS METHODS 1
a eee ee

6.3.2 Product rule and chain rule

rule and when
It is very important to be clear when you need to use the chain
you need to use the product rule.
together
e* sinx means e* X sinx, so two simple functions are being multiplied
and the product rule is needed.

e"* is a composite function fg(x) where g(x) = sinx and f(x) = e*, so the chain
rule is needed.

It is sometimes necessary to use both rules.

Example 2
d
Find where y =e” sin 0.5x.
dx

Solution
y = uv, where
u=e and v=sin0.5x
d
By the chain rule, a Oe amand de = 0.5 cos 0.5x
dx dx
d d
By the product rule, == v ue

dv : De Dx
So) =sin0 sx x 2e7"+e> xX 0S cos05x
dx

d
Or = = e* (2 sin0.5x + 0.5 cos 0.5x)
a

6.3 Exercise 2

1 Differentiate these products, using both the chain rule and the product rule.
Set out your working as in example 2.

(a) 2e3* sin 2x (b) e** cos 3x (c) e* sindx

2 Use the product rule or the chain rule or both in order to differentiate the
following functions.

(a) In (x* +1) (b) xInx (c) x sin? x

(d) x sin x” (e) (x + sinx)* (f) e* cosx + xsinx

(g) (2x+3)' — (h) x*e**

 FURTHER DIFFERENTIATION TECHNIQUES 447

3 Work out the gradient of each of these graphs at x = 2.

y = 0.5x sin? x

0 1 2 3 4 x
4 The tip of a tuning fork moves so that its distance in centimetres from a
central position is given by

s =0.4sin5127t, where t is the time in seconds

(a) What are the displacement and the velocity of the tip of the tuning fork
after 1 second?
(b) How many vibrations per second does the fork make?
x+2
5 Differentiate x
78 by writing the function as (x + 2)(x +1)’.
.

6.3.3 Differentiating quotients

x +2)
When tackling exercise 2, you probably found the derivative of
(x + 1)
to be
Gri) + —1 (4-1) 4&4 2)
1 Xoo
You may have rearranged the answer as ——— — ————..
: e xt1. (x41)
448 CALCULUS METHODS 1
a

This is still not a very neat answer, and it can be simplified further.
(ee)
ea 2) el
Cat) DG sty
It is possible to differentiate any quotient by rewriting the function with negative
indices and using the product rule, but the process of writing the answer in a
neat form is tedious. It is therefore worthwhile to try to find a formula for the
derivative of a quotient.
u
Suppose y = — where both u and v are functions of x.
Vv
Start by writing y as a product and use the product rule.

TU

dy _4du d(v')
dae. te eds
d —1
ae may be evaluated using the chain rule.

d(v™') = d(v~') . dy
dx = dv dx
Paty Laas,
v* dx

dy 1du udv
dx udx vwdx
du dv
Cad ae

Example 3
(a) Use the quotient rule to differentiate ——_.
(2x + 3)
(b) Show that the function has a local maximum at x — —3.
 FURTHER DIFFERENTIATION TECHNIQUES 449

Solution
u
(a) Sm where
u =x" and p= 244-3

du dv
dye aye eda. xe 3) 2x a xi. 2x(x +3)
dx v? (Cay. = Omer3y (223)

d
(b) At stationary points, == 0
x
x(x
+ 3) = 0
i () OTS

There are stationary points at (0,0) and (—3, —3).

Examine the nature of the

stationary point at
N= Se

When x is just less than

dy .
—3 (say —3.1), — is
dx
positive.

When x is just greater ; A local maximum at'x = —3

than —3 (say —2.9),

i ae )
— is negative.
dx

= (—3, —3) is a local maximum.

3 Exercise 3
: a5 3x
1 Differentiate: (a) yas (b) Se = =
x o sin x sin 2x

2 Work out the gradients of:

2x
(a) y=—— atx =-1 (b) y=— atx =0.8
e x
Use a graph plotter to check that your answers look reasonable.
1
3 You can differentiate (x +3) = (2x + 3)’ by using either the quotient rule

or the chain rule. Work out the derivative using each method in turn.
Which method do you prefer?
450 CALCULUS METHODS 1
SE

4 (a) The graph of y = A fas has a local minimum and a local maximum.

Work out the coordinates of these points, showing clearly how you know
which is the maximum and which is the minimum.
2
(b) Repeat (a) for y = eae

sin x
5 Use the quotient rule to differentiate tanx = ie
Oo
6 Work out the derivative of cot x by writing:
cos x 1
(@)iecotha— - (b) cotx =
sin x tanx

(c) Check that your answers to parts (a) and (b) are consistent by writing
each one in as simple a form as you can.

7E The first part of an alternative proof of the quotient rule, assuming that the
product rule is true, is given below. Try to complete the proof.
u
y=— where uw and v are functions of x
v
Then u=vy
d
By the product rule, ee
dx

6.3.4 Implicit differentiation

You know how to differentiate functions of functions, sums, differences,
products and quotients of functions.

However, all the functions which you have

been asked to differentiate have been stated
explicitly, i.e. as y = f(x).

However, functions are sometimes stated

implicitly. For example, the equation of the
circle, centre (0,0) and radius 3 units, is
usually given in the implicit form

x Py 9
 FURTHER DIFFERENTIATION TECHNIQUES 451

d
You can find ~ from this implicit equation by finding the derivative of both
x
sides of the equation, with respect to x.

£(@)+ £0) =£0)

The only difficulty here is with <( ), and this can be overcome by using the
chain rule.
d 3 jibe d 9 IES

d PB dy
ase(y) = aI,

If x+y=9

Then 2x+ 2y® = 0 (differentiate throughout with respect to x)

x
Spt
dx sy
The process used here is called implicit differentiation.

Example 4 ;
The point (1, 2) lies on the graph of x° + 3y? — 4x +y = 11.
(a) What is the gradient of the tangent at the point (1, 2)?

(b) Work out the equation of this tangent.

Solution
Rebate os kdl dies > od
(a) Ae alte oe (1h)

dy dy
3x7 + 6y
ee ea 0

d
(6 40) 43x 2420
dx
d
(6y
+1) =4— 3x’
dx
dy 4 —3x"
dx 6y+1
dye cet
Whenrx —1 and y= 2, awa

1
The tangent at (1, 2) has gradient EY
 452 CALCULUS METHODS 1

(b) The equation of the tangent is

(y-2)_1
Gat 3
13(y—2) =x-1
or 13y—x=25

a ——

6.3 Exercise 4
d
1 Use implicit differentiation to find = when:

(a) 27 —3y+4x7=2 = (b) +4y° —7x4+3y=3

2 The graph of 9x” + 4y* = 45 is as shown.

d
(a) Use implicit differentiation to find oe
dx
(b) Work out the y-coordinates of the two x
points where x = 1 and calculate the
gradients at these points.
(Check that your results look reasonable.)

3 The circle with centre (—3, 1) and radius 5 units has equation
Gar
1) —25
ie: x +6x+y" —2y = 15

d
(a) Work out an expression for > and use it to find the equation of the
be
tangent to the circle at the point (1,4).

(b) The point (1, —2) lies on the circumference of the circle. Work out the
gradient of the tangent at this point and the gradient of the radius of the
circle to this point. Check that your results are in agreement with the
property of circles that a tangent is always perpendicular to the radius at
that point.

(c) Repeat (b) for any other point on the circumference of the circle.

4 You have seen that the circle x* + y” = 9 can be written in parametric form as
X25 COS, y = 3sin0
Parametric differentiation gives the result
dy — cos
dx sin
 FURTHER DIFFERENTIATION TECHNIQUES 453

; ; d 6
Explain why this result is equivalent to the result eee Alcs explain why
dx y}
the results show that the angle between tangent and radius must always be a
right angle.

In section 3.4 you obtained the following results by numerical methods.

d
qe)
— (2*) = 0.693 xx 2*

d
ae)
— x 3”
(3*) = 1.10 x

You can now obtain these results using implicit differentiation and the fact
that
y=2*
s-hiry = In 2°
in y= x lin
1 dy |
(a) Explain why sa(In y) = — — and hence find ay (i.e. the derivative of 2°).
dx y dx dx
Show that your answer agrees with the numerical result.

(b) Work out the derivative of 3* using implicit differentiation.

(c) What is the derivative of a*, where a is any constant?

6.3.5 Implicit differentiation and the product rule

The equation considered in the previous section,
x+y =9
could have been rewritten as

y=+y(0-2)
and differentiated in the usual way. However, implicit differentiation is much
easier. In some cases, just rewriting an equation in the form y = f(x) is itself very
difficult or even impossible.

Example 5
d
Find = when y* + xy +x” = 8.
oe

Solution
Consider the xy term first.
Let z = xy and use the product rule with u = x andv = y.
 454 CALCULUS METHODS 1
nee ET

du dv dy
Then 7 = 1 and eral Ac

gee an du
ES ae de
d
=yx tex

me
dy
ay

Now differentiate each term of the original expression.

yp xy a eas
Ser
ee
dy dy + 0x10)
=e
(v+x2) 2x

Then (2y + x) = +y+2x=0

dy
dx
dy
=> —(2y+x) = —(y+ 2x)
dx
dy d
_ of
=(y + 2x)
dx (2y + x)

6.3 Exercise 5

1 Work out a for:

dx
(a) 2x*+3xy-—4yt+y=5 (b) yy —2xy+3x—x* =1
(c) x7 +4 —4x4+8y=28 (d) y4+xetxy? =4
Ve Olen yi=i.
d
(a) differentiate implicitly to find 2
ie
d
(b) express y as an explicit function of x and differentiate to find -
x
(c) show that your answers to (a) and (b) are the same.

3 For x*y+2x° =4,

(a) use implicit differentiation to find > and so shéw-that (—2,3) isa
be
stationary point on the curve;
(b) rewrite the equation as a function of y and use the quotient rule to find
dy . Show that this also indicates a stationary point x = —2.
dx
 FURTHER DIFFERENTIATION TECHNIQUES 455

d !
4 (a) Find if 3xy — 2x” = 8 and so show that the graph of 3xy — 2x? = 8 has
stationary points at (2,24) and (—2, —23).
(b) Rewrite 3xy — 2x” = 8 in the form y = f(x) and use the quotient rule to
check the coordinates of the stationary points.
(c) Do youyou p prefer implicit
Pp differentiation or rearrangement
g and use of the
quotient rule for checking the coordinates of stationary points? Which
would you use if you had to find the coordinates of stationary points?

:(3 6.3 Tasksheet $1 — Differentiation practice (page 539)

456

Calculus methods 1
Miscellaneous exercise 6

1 Use the chain rule to differentiate the following; then check your answers by
applying an addition formula to y and differentiating each term separately:
(a) y = cos (3x 4+ 2) (b) y = cos (4x — 1) (c) y = sin (Sx — 11)

2 Differentiate: (a) sin? x (b) sin (x?) (c) cos? x (d) cos” (3x)

3 Use the chain rule to differentiate:

(a) (3x +5)? (b) sin 4x (c) 6x2

(even a) (e:) 4cos(2x—7) — (£) 3(x? — 5)

(g) cos” x (h) cos” 5x (i) /(x* — 3x)
D 1
eee4/(wei T)s i(k) Gx +4)

4 The radius of a spherical balloon is 2m and its volume is increasing at a rate

of 0.1 m* min7'. At what rate are

(a) the radius, (b) the surface area increasing?

5 Ata certain moment each edge of a block of ice measures 10cm and the
surface area is decreasing at 20cm*h~!. At what rates are

(a) the length of an edge, (b) the volume changing?

(Assume that the ice remains cubical in shape.)

D
6 Sketch the graph of y = x + 5 and determine the coordinates of the
Be ap Il
maximum and minimum points.

7 A circular cylinder of radius rcm, height h cm and constant volume 100

cm?
is being rolled in a machine to reduce the radius. Write down a formula
for h
in terms of r.
(a) At a certain moment r = 2. If the height is increasing at
0.2cmmin~!, at
what rate is the radius being reduced?
(b) Under the same conditions, at what rate is the total surface
area
changing?
 MISCELLANEOUS EXERCISE 457

Find the equation of the tangent to the ellipse 3x” + 4y” = 12 at the point
(—1, 15).

Differentiate both sides of the equation x” + y* = 4x + 6y + 3 with respect

dy .
to x, and hence express Pa terms of x and y.
x
10 A particle moves along a straight line so that its distance from a fixed point
A on the line is x metres after t seconds where x = ¢’(t — 3)”. When is it at
rest? What is its acceleration when t = 3?
asin’ t
11 The position vector of a particle at time t is ( ),Find the velocity
os’ t
vector at time f, and hence the position and velocity when t = 0.

12 A cone is to-be made from a sector of a circular piece of card, radius 30 cm.
Obtain an expression for the volume of the cone in terms of @ where the
angle at the vertex of the cone is 20. What is the maximum value of this
volume?

13 Find ay in terms of x and y if:

dx
(a) 6x* + Sy’ =7 (b) x7(y +1) =4 (c) 3x= cosy

14 Find the gradient of the hyperbola x* — y* + y = 1 at (1, 1).

15 Use the product and quotient rules to differentiate:

(a) 3x?(2x — 1) (b) 2x cosx (c) x(x +5)°
3 2 3
Cpe
4x —1
CsXe. (f) (2x+1) sin 3x
(g) x” tan 2x (h) sin 3x cos 4x (1) sin’ x cos x

16 Differentiate:

(a) cot3x (b) eed (c) sinx cos x (d) 1+ tan? x

iP cosx

d d
17 (a) Given that y = pes = find — and verify that cos x ae y.
1 — sinx dx dx
Os x ;
(b) Show that Fis secx + tanx and hence verify your answer to (a).
— sinx

cos x = aos
18 Differentiate cot x by treating it as . Also show that the derivative of
x
secx is secx tan x, and find the derivative of cosecx in a similar form.
458

Calculus methods
1 Integration techniques
7.1.1 Volumes of revolution y=?
Consider the section of pipe illustrated here . 2m
The curved surface is bounded by the function
y = x* and the dimensions are as shown. 1m

The problem is to calibrate a stick — called a

dipstick — which indicates the volume of
liquid in the container.

One way to solve this is to consider the liquid as a number of thin horizontal
slabs, each of thickness 6h.

The width of a slab at distance h

from the bottom will be 2,//,...

the horizontal cross-sectional area

ill b py hee

and the volume of the slab will 3m

therefore be 6,/h 6h. ae
 INTEGRATION TECHNIQUES 459

The total volume of the container will be the sum of the volumes of the thin
slabs.

S > 6h bb
In the limit as 64 — 0, the exact volume is
pl
V= |6/hdh NBwWh=h°»
0
1
ae [4b]
0

=4m? (or 4000 litres)

The (h, V) graph for the container is as shown.

V (m3)

4 V =4p15

!
|
I
|

j
x 1 h (m)

The container is half full at a height x as shown on the graph.

Via Ale

= b= (4)
bh= (41)}
~0.63
= 63cm
460 CALCULUS METHODS 2

@
1 At what depth will the container be:
(a) a quarter full, (b) three-quarters full?

Another container is shown below.

qu |:
2 Explain why a thin slab at height / will have the dimensions shown below.

2+h
h a

3 Calculate the volume of the container and sketch the (h, V) graph.

4 For what value of h is the container half full?

5 The depth of liquid in the container increases from 1.1m to 1.4m. What is
the increase in volume?

This technique, often used to calculate the volume of a solid, is to imagine the
solid as a large number of thin slabs. If the horizontal cross-sectional area can be
expressed as a function of the vertical height, then the volume of the solid can be
calculated.

Shaded area
Area = A(h) represents the
volume ofa
thin slab.

oh h
 INTEGRATION TECHNIQUES 461
i

Example 1
A container and a horizontal cross-section are shown below.
3m l+h

\Z2F V (2h)
ae
2m

The depth of liquid increases from 4 = 0.6m to h = 1.2m. Calculate the increase
in volume.

Solution
The horizontal cross-sectional area at depth h is
Geb) 20 > 2h me.
im + b)s/(2b) dh = 2 [0° +h!) db
0.6 6

ey
ay v2|34" as ad
0.6

= 1.536m? (or 1536 litres)

You can apply these ideas when finding the volumes of shapes which are formed
by rotating areas. For example, a solid with the same shape as a wine glass can
be produced by spinning the area between the graph of y = x’, the line y = 4
and the y-axis about the y-axis.

y
462 CALCULUS METHODS 2

The horizontal cross-sectional areas you would use to calculate the volume of the
glass are particularly simple, as they are all circles. Solids formed in such a way
are called solids of revolution.

& 1B

1 Thegraph of y = x” for 0 <x < 2 is

rotated about the x-axis.
(a) Explain why the volume obtained by
rotating the shaded strip is
approximately y* 6x.

(b) If you consider the area as being made up of a large number of thin
strips of this kind, the volume is
2
|my” dx
0
To be able to integrate this you must write y* in terms of x. What will
this give?

(c) Evaluate the volume.

2 Work out the volumes obtained by rotating each of these shaded areas
about the x-axis.

(a) y (b)
 INTEGRATION TECHNIQUES 463

3 Work out the volume obtained by rotating y

y= x* for 0< y < 4 about the y-axis.
4
Work as in question 1 by first writing down
the volume obtained by rotating the shaded |
strip about the y-axis. by

Work out the volume obtained by rotating the area bounded by the curve
1
y = -, the y-axis and the lines y = 1 and y = 2 about the y-axis. (First
te *
draw a sketch to show the area being rotated.)

The two shaded areas shown on the graph are

equal.
Would you expect the volume generated by
rotating area A about the
x-axis to be the same as that obtained by rotating B
about the y-axis?
Justify your answer.
 464 CALCULUS METHODS 2
__ ee
SE
ee

Example 2
The shaded area shown is part of a circle with centre
at the origin and radius r.

Rotate this area about the x-axis and hence prove that
the volume of a sphere is given by the formula:
v= 4a?

Solution
The solid generated will be a hemisphere and its volume is

van! y’ dx
0
2 2
But y° =r xX

ova|P-ade
0

=1|Px—42)]
: 0

ss
aa it Bi

The volume of a sphere is twice the volume of a hemisphere and so the

formula is proved.

7.1 Exercise 1

1 Find the volume formed when the area between the curve y = x’, the x-axis
and the lines at x = 1 and x = 2 is rotated about the x-axis.
2 The region between the curve y = 2 — x* and the x- and y-axes is rotated about:
(a) the y-axis, (b) the x-axis.
Find the volumes of the solids formed.

3 Calculate the volumes formed by rotating about the x-axis the areas bounded by:
(a) the lines y = x, x = 1, x = 2 and the x-axis
(b) y=x
and y= x?
(c) y=x—2x? and
y=0

4 Find the points of intersection of the curve y = x(4 — x) and the line y = 2x.
Find the volume generated when the area enclosed between the curve and the
line is rotated through 27 radians about the x-axis.
OO — — —
 INTEGRATION TECHNIQUES 465

7.1.2 Integration by inspection

You have now used integration to find areas and volumes. You also know that
to integrate a function you only need to find a function whose derivative is the
original function. Unfortunately, this is far from simple in practice, as many
functions cannot be integrated algebraically. Definite integrals can always be
evaluated numerically, but having an algebraic solution is often more convenient
and may even be essential in obtaining a complete solution to a problem.
When integrating, it is important to know how to choose the best method.

Example 3

Find [sin3x — x) dx.

Solution
Always try to integrate by inspection first, since this is likely to give the
answer more quickly.

[sin —x’)dx = —}cos3x = 1x? ae

Example 4

Find [ow— 3x”)(4x* — 3) de.

Solution
In some cases it is necessary to multiply out brackets first.

|(x? — 3x*\(4x° — 3) dx = |(4x°® — 15x° + 9x”) dx

= 45° —Bx° + 3x? +c

To integrate a function by inspection, you must first think of a function which is

likely to differentiate to the original function. You then compare the derived
function with the function you want and make any small adjustment as
necessary.

Example 5

Find [>eo dx:

Solution ;
Differentiating the function e** seems to be a sensible starting point.
d 2 2 3x2 3x2
(“l= one = ea(ic ) xe
dx dx
2 zy
=> be dx =te™ +¢
466 CALCULUS METHODS 2

Ac
What functions might you try to differentiate to solve the following integrals?
Find the integrals where possible.

(a) |xcos x” dx (b) |xcos Dax

(c) |cosas dx (d) [cos dx

The questions above illustrate the difficulty of integration compared with

differentiation. In (a) and (b) the function being integrated is the product of
two functions. (a) looks more complicated, but proves to be an easy ‘backward’
chain rule, while (b) requires the product rule and some very clear thinking.
(c) and (d) both look straightforward, but (d) cannot be integrated
algebraically.

Exercise 2

1 Find: (a) [cos3x dx (b) Jo+ 2)(x — 2) dx

(c) fe dx (d) |ae

XM

2 Write down the derivatives of:

(a) sinx® (b) cos2x* — (c). (x — 3)?
3 Use your answers to question 2 to write down the integrals of:
(a) x? cos x? (b) x sin 2x? (c) x(x* a3)?

4 The nose-cone of a rocket is obtained by rotating the area between the graph
of y = 3(1 — x’) and the axes about the y-axis.
(a) Draw a sketch showing the area being rotated.
(b) Calculate the volume of the nose-cone.

5 (a) Sketch the parabolas y* = 4x and y* = 5x — 4 on the same axes and find
their points of intersection.
(b) A bowl is made by rotating the area enclosed by the curves about the
x-axis. Find the volume of the material used to make the bowl.
eee Se
 INTEGRATION TECHNIQUES 467

7.1.3 Integrating trigonometric functions

You have seen that not all functions can be integrated algebraically. However,
many functions which may look impossible to integrate can be integrated if they
are first rewritten. The trigonometric identities developed in Chapter 5 are
particularly useful.

Example 6

Find |cos’ x dx

Solution
The identity 2 cos* x — 1 = cos 2x can be written as cos” x= 5oi 5.cos 2x

> |cos? x dx = |(44+Jeos2s) dx

= 4x + }sin 2x +¢

The following identities are the addition formulas, developed in section 5.1.5.
sin (A + B) = sin Acos B + cos Asin B @
sin (A — B) = sin Acos B — cos Asin B @
cos (A + B) = cos Acos B — sin AsinB @
cos (A — B) = cosAcosB + sin Asin B @
These may be used to prove other useful results. For example,
a eas Not
ot (3-8) sin (F)coso ~ co (5)sin 0 (from @)

= cos

=> sin (4-8) = COS)

It may also be demonstrated easily that

cos (5-8) = sind

sec (4-2) = cosec0

tan (5-8) = cot8

You should ensure that you are able to prove each of the results above.

You may have wondered about the significance of the prefix ‘co’ in the names of
trigonometric functions. It comes from complementary angles, which are angles
whose sum is 90° (or 3m radians).
 468 CALCULUS METHODS 2

The diagram shows that, when @ is acute,

cos 6= sin ($7 — 8)

The cosine of an angle is the sine of the complementary
angle.
cos 6 = sin ($7 — @)

The addition formulas may also be used to prove the sum and difference
formulas.
2cos AcosB = cos(A + B) + cos (A — B)
2sin Asin B = —cos(A +B) + cos (A — B)
2 sin Acos B = sin(A + B) + sin (A — B)
For example, using formulas @) and @ on page 467 and adding them, you obtain
cos(A+B)= cosAcosB—sinAsinB ©
cos(A—B)= cosAcosB+sinAsinB @
cos (A + B) + cos (A — B) = 2cos AcosB @+@

@ wv
1 Adapt the method used above to prove the remaining two sum and
difference formulas.

2 Use the sum and difference formulas to prove that:

(a) 2cos*x =1+cos2x

(b) 2sin? x= 1—cos2x

(c) 2sinx cos x = sin 2x

All of these results, especially the double angle results proved in 7.1p, are useful
in integration.

Example 7

Find |sin 5x cos 2x dx

Solution
Using the identity 2 sin A cos B = sin(A + B) + sin(A — B),

|sin sxcos2edx = |(Gsin7x + 4sin3x) dx

= — 7qcos7x —1cos3x 4c
 INTEGRATION TECHNIQUES 469

a Exercise 3

1 Find (a) |sin x cos x dx (b) |sin 3x cos 3x dx

2 Find (a) |cos Sx cosx dx (b) [sin3x sin 7x dx

1qr

3 Find | cos’
x dx
0

4 Calculate the volume generated by rotating

the shaded area about the x-axis.

7.1.4 Integration by parts

You know that not all functions can be integrated using algebraic techniques
and that those that can are not always straightforward. The first technique you
should try when searching for an algebraic solution is ‘trial and improvement’
or ‘inspection’. Think of a function which looks as though it might differentiate
to the function you need to integrate, then differentiate it and make adjustments
as necessary (for example, multiply or divide by a constant). We now develop
some standard techniques which often prove useful when ‘trial and improvement’
fails.

The formula for the product rule for differentiation can be rearranged into a
form which is helpful when integrating certain functions.

a ele aves =| Sede + [uS dx

Geo” oo de tos ae dx

dv du
= |ugeae = uw — [eGhax

This is called the formula for integration by parts. It provides an efficient method
for integrating some products of functions.
470 CALCULUS METHODS 2
ee eR ee ee

Example 8

Find |xcos2x dx

Solution
du
Letu=x, so—=1.
dx
du :
Let rr =—TCOS.AmeES Ot — +sin Mae
55

|xcos 2x dw = $x sin 2x — [sin 2x a

= x sin 2x + 4cos2x + ¢

Integration by parts is only applicable to functions written as the product of two

functions. When using the formula, always start by deciding which part of the
dv u
product should be u and which should be ie Then write down an and v. Only
x x
: du dv
when you have the functions u, v, ez and ne clearly written down can you be
x x
reasonably sure of substituting correctly into the formula.

; . ; dv
For instance, in example 8, if you had chosen to let u = cos 2x and ae = x, then
you would obtain Hy

du d
u=cos2x
> —=-—2sin2x and Sax syaix
dx dx

= |xcos2xdx = 1x? cos 2x + |?sin 2x dx

This rearrangement of the integral does not make it simpler to evaluate. If you
make the wrong choice as to which part of the product should be u and which
should be then you will generally find that the integral has become more,
rather than less, complicated. Experience will enable you to spot which part of
the product to integrate and which to differentiate.

Integration by parts will not always prove successful for functions written
as the product of two functions. Experience will help you decide when this
method will work. The method works for all the functions in the next
exercise.
 INTEGRATION TECHNIQUES 471

7A Exercise 4
1 Find: (a) lee dx (b) ire dx — (c) xem dx

2 Find: (a) |cos. dx (b) |xcos3rde (c) |xcos ax dx

3 Use integration by parts twice to evaluate:

(a) [2e* dx (b) |?sinxdx

4 Work out each of these definite integrals. Sketch diagrams to show the areas
you have found and check that your answers seem reasonable.
0 0
(a) | x sin 2x dx (b) | 2x €°* dx
=f =3

Tasksheet E1 — By parts (page 563)

7.1.5 Integration by substitution

Another useful technique of integration can be obtained from the chain rule for
differentiation.

d
To find = if y= sin (3x* +5), let w = 3x" +5 and y = sinu.
x
dy dy du
afm aX dg 7 OOS X OH = 6x cos (3x + 5)

In the chain rule, the variable u is substituted for the original variable x and then
x is substituted back at the end of the answer. The same technique can be used
for integrating certain types of function. Changing the variable of the integral
can be done using the chain rule.
ay. 2 dy dx \2 dx dy
7 x 7 du but y= |Pax
du dx ie
 472 CALCULUS METHODS 2
eee

Example 9

Find y = |cos (3x* + 5) dx

Solution
du dx 1
Let u = 3x +5 =>7 = 6x This gives =o

if dx
>y= —_—
|xcos (3x7 + 5) re du
2 —.
Replacing dx by rs du.
i =

Shy = 1 [cos(3x +5) du This cannot be solved as it stands. You must

express the integral entirely in terms of the
new variable u.

== 1|cosudu and so y = dsinu+c = sin (3x7 +5) +e

You may have been able to solve the integral in this example by inspection.
While using substitution is not wrong in such a case, it is unnecessarily
complicated. Substitution only works for some functions and even then it is
sometimes no more than a rather slow method fo. finding integrals which can be
found by inspection. However, there are many cases where it considerably
simplifies the integral. Some of these cases are considered on tasksheet E2.

Le 7.1 Tasksheet E2 — Integrating the circle (page 564)

It is important to choose the right substitution when using this method of

integration. All the integrals in the exercise below lend themselves to integration
by substitution, although you may feel that some of them could be done by
inspection. Remember that whatever substitution you make, your final solution
must be in terms of the original variable.

7.1 Exercise 5

1 Evaluate the following integrals by using the suggested substitution.

(a) [oc+3)>dx (let u =x +3) (b) |xx —5)®dx (let u = 2x — 5)

(c) exe —2)’dx (letu =x —2) (d) |xo? —4)®dx (let u = x” — 4)

(e) [eve —2)dx (let u =x’ — 2)

 INTEGRATION TECHNIQUES 473

2 Integrate the following functions using the suggested substitutions.

(a) sin*xcosx - (let u = sin x) (b) cos*xsinx (let w= cos x)

3 Integrate the following functions using a suitable substitution.

(alex (x +3) (b) x(x — 3) (c) x4/(x — 2)
; N
(d) COS X sin? x (e) cos> x sin xX (f) aye

7.1.6 The reciprocal function

You know that
faba n= 1
gas

and that
b b
|in [Inx| ce hy ee
ax

Negative values of x, however, present a problem

as the function In x is defined only for x > 0.

1
It is possible to get around this as follows, using the symmetry of the graph of —.
x
=a 4 b 1 b
| —dx = -| —dx = —[Inx| =Ina—Inb
—bXx aXx a

This can be rewritten in a more conventional form using the modulus

function, |x].

i ae Ina—Inb
Bs
I In [—a| — In |—6|

[inlet]
where the symbol |x| denotes the absolute (or positive) value of x. For example,
|-3.5| = [3.5] = 3.5.
474 CALCULUS METHODS 2
i

You should be aware of integrals whose limits extend across the discontinuity of
2
1 , Pee: i = dx or 1 dx:
the — function. For example, it is not possible to evaluate (es=
x —a
A sketch of the graph will usually reveal any problems.
2 1
| dx cannot be evaluated.
ox—1

The function must be continuous

over the range of the integration.

Care must be exercised; however, this forms the basis of a very useful integration
technique.

Any integral of the form laa can be integrated algebraically, as follows.

f(x)
du dx 1
Let w= f(x). Then rea f'(x) and ee xy

The variable of the integral can be changed from x to u by replacing dx by

dx
ait So

onan eter ronaed coke fe

= In|ul+c
= In|f(x)| +c
 INTEGRATION TECHNIQUES 475

Example 10
3 5
Find:
in ——.
(a) |e ——;
(b) Fesexe (c) [ana dx

Solution
(a) |e
MeaS se=i"
Glee sePipes
2s c
x rr ree i 2
(b) | Rae = 3) Sea tine +1i)+c¢

(c) |eanxdx = [aa = -[& ae = —In|cosx|+c

COS Xx COS X

7.1 Exercise 6
; 3 6 1
1 Find: (a) |;g 5 dx (b) lee (c) lz A dx

ee 2 5S ees
Evaluate: (a) |—— dx (b) | dx
13x -—2 23x —2

4
3
Explain why you cannot evaluate |Faia
(x
Show that:

(a) [tans dx = In|secx| +c (b) |cots dx = In|sin| +c

: x? Mp
Find: (a) | 3 dx (b) | ue

6E Find: a) [2 de (b) |FS te) lane @) | bg

14x? 1+ x? xInx 1+ cosx

(For (a) and (b) you will need to use the substitution tan“ = x with the
identity 1 + tan? u = sec” u.)
476 CALCULUS METHODS 2
I

7.1.7 Partial fractions

Example 11
m h < il 1
Dee eR (sD), = 8 x +2
2 5 d
(b) Hence evaluate iCLIC x.

Solution
1 1 (x= 2) (x — 3)
(a) $23 290 Cle Gaocee
(x +2) — (x — 3)
(x — 3)(x + 2)
3 Ss
(x — 3)(x + 2)

(b) j,(
1
ee
=e
dx = [Ins ~3]—In +2),
II In (1) — In (4) — In (2) + In (3)

In @) = —0.981 (to 3\s.f.)

Example 11 illustrates an important technique for integrating certain polynomial

fractions — splitting a single polynomial fraction into simpler fractions which can
then be integrated. This process is known as splitting or resolving the fraction
into partial fractions.

Example12
3 1 B
Express oy in the form
A+ - and hence find|=" de.
Pw ar Os ar

Solution
Diicinhs wagd ects Oe ra Gat <0 5 ay 5
RD eS) ee Oe ee) ee
3 1
Hence [=~ ibe || 3- 2 dx
x +2 foee
= 3x —SIn|x+2)/+c

Example 13
SE aoe A B
Express Gree)
cerayors Sri in the form ame Pat

3x — 8
Hence find | dx.
@40)Q2=1)
 INTEGRATION TECHNIQUES 477

Solution
ae Bs AGx 1) Bx 6)
SoS se (x + 6)(2x — 1)
if Ax B(x Oe. 3x— 8
(x+6)(2x—-1) § (x+6)(2x —1)
then A(2x —1)+ B(x +6) =3x—-8 @
Putting x = + in © to eliminate A gives
61B=-615B=-1
CASE Sail is

Putting x = —6 in (@ to eliminate B gives

SISA
26 = Ae)

Thus

lisence at)
= 21n |x +6 —$1n|2x
—1|+¢

The following example illustrates how to deal with the situation where the
denominator consists of a repeated linear factor.

@
5) A B
1 By trying to find A and B so that me
gaan
nie 82 end”
explain why it is not possible to do so.

B G B
2 Show that by using einen instead of ———] you can obtain a possible
(x + 1) (x +1)
set of partial fractions.

3 Complete the process by finding A, D and E such that

Oye ae A D E
= D@ 1) eg x2 2 x1 ay (x41) 2 (1)
Hence show that
ares E 7x 19
ime ar te 2 (2)
(x—2)(x+1)? %%-2) 9%(x+1)
You now have two equivalent sets of partial fractions. You may find (2)
simpler to calculate, although (1) is easier to integrate.
478 CALCULUS METHODS 2

Example 14
3x" — 3x”
—x
Express ese in partial fractions and hence find es dx.
Dien | Koa Lf

Solution
This example illustrates an important preliminary to the partial fraction
process. When the degree of the numerator is equal to or greater than that of
the denominator, you will need to divide out until the degree of the
numerator is less than that of the denominator:
5
emilee 1
3x7 —3
3—Xx

3x* — x (3 — x)
S2 SS
x —1
Soe
eee
kee eis A B DAG) FBG)
x°—-1 x-1 x+1 (x — 1)(x
+ 1)
== 3—x = A(x+1)
+ B(x -1)
Put x = 1 to give A = 1, and
x = —1 togive
B = —2. Sofinally
Se tom ie 1 —2
x1 aah i

Keale ee

and
3x7—
SE dg sulle Naess dx
Neel x Xap

=3x+In|x—1)—2In|x+1)
+c
 INTEGRATION TECHNIQUES 479

7.1 Exercise 7

1 Split into partial fractions:

[eee a

(x + 2)(x +3) (b) ee

: + 1)
(2x + 1)(x yee
(&

2x?
nT

+ 3x -2
Tal Sx 4 ee |
erieeo) oo GennGr2)
d). Se
") kGan)
es

5 1 A B
2 (a) Express in the Oe ees eae.

5) 1
(b) Hence find |e
Baia |W)
be
8 9
3 Rewrite =~" * in the form eae ee
Orta std, NSD,
B
and then in the form
A + ——-+ e :
x+1 x«+2
1x7 48%49
Hence evaluate a eee
0x*+3x+2

4 By first resolving into partial fractions, find:

(a) loca (b) | us dx (c) | ue,

x — 1)(x — 2) x? — 4 x +5x4+6—

aoe ee ae, Ae re C
a) Expres ———————.,in the form
esa): Sil) LCT):
(b) Express Week
eo hein partial fractions and hence find
(x + 1)(x — 1)?

ene
3 x d
(c) praluate ee ea: x.

6 Resolve into partial fractions:

x (b) 3x"
x? — S5x+4
480 CALCULUS METHODS 2

Ce
ee
 481

Calculus methods 2
.2 Polynomial approximations
7.2.1 Taylor’s first approximation
You can determine the value of a function such as sinx or e” for whatever value
of x you choose, merely by pressing the appropriate button on a calculator. You
may, however, have wondered how the calculator computes these values. When
a computer evaluates a function such as sin x, it does so by evaluating a
polynomial approximation to the function. This chapter will look at how you
can obtain and make use of polynomial approximations in order to solve
equations and handle complicated functions.

The simplest polynomial approximation to a function is linear. You will recall

that a differentiable function always appears to be locally straight under
sufficient magnification and it is because of this that the first approximation of a
function at a point is taken to be the tangent to the graph of the function at that
point.

Such an approximation is known as Taylor’s first approximation, after the

English mathematician, Brook Taylor (1685-1731).

For example, —
a,

is Taylor’s first approximation to

y =sinx near the origin.

With x in radians, — (sin x) = cos x.

dx
When x = 0 (at the origin) the gradient of sinx is 1 (cos0 = 1).

The equation of the tangent to sinx at (0,0) is therefore y = x.

This linear approximation is only valid near (0,0). To illustrate this, the values
of x and sinx are tabulated below.
 482 CALCULUS METHODS 2

x 1 0.5 0.2 0.1 0.01

sin x 0.84 0.479 0.199 0.0998 0.009 999 8

If you plot the graphs of y = x and

y = sinx on the same axes on a
graph plotter, you can see the ‘error’.

The percentage error is given by Percentage

isthe error
100 aac 20
sin x

The error function can also be

plotted on a graph plotter.

What constitutes a reasonable

approximation is open to
discussion. The approximation
is good for —0.5 < x < 0.5 if an
error of no more than 5% is
considered reasonable.

Taylor’s first approximation is often used as a ‘rule of thumb’.

For example, you
will sometimes find it useful to say that ‘sinx = x when x is
small’, although
sinx is actually equal to x only when x = 0.

You should confirm for yourself that similar linear approximatio

ns for cos x and
tanx atx = 0 are cosx © 1 and tanx x x.
 POLYNOMIAL APPROXIMATIONS 483

It is possible to find a Taylor approximation at a point other than the origin.

The process simply involves finding the equation of the tangent at the given
point.

Example 1
Find Taylor’s first approximation to the function y = sinx when x = eT.

Solution
The point P on y = sinx has coordinates (27, }), i.e. (0.52, 0.5).

Since = cos x, the gradient of the graph at P is cos!a = 0.87.

dx
The equation of the tangent at P is given y
by
ye Ore
0 See oe 1 y =sin x

=> y—0.5 = 0.87(x — 0.52) (0.52, 0.5)

= y= 0.87x + 0.05
which is Taylor’s first approximation.

7.2 Exercise 1

1 Find Taylor’s first approximations to the following functions at the points

given. Check your answers using a graph plotter.
(a) y=x?+5x—2at (2,16) (b) y=e*at (0,1) (c) y=Inxat (1,0)
(d) y= x* — x? at (2, 12) (e) y= x* 4+Inx at (1,1)

2 (a) Find Taylor’s first approximation to y = cosx at x =47.

(b) Use your approximation to estimate cos 1.5 to 6 decimal places.
(c) What is the percentage error in your estimate?
484 CALCULUS METHODS 2

7.2.2 The Newton—Raphson method

Taking a tangent as an approximation to a function has an extremely useful
application. The contemporary and colleague of Isaac Newton, Joseph Raphson
(1648-1715) realised that it was possible to use a tangent approximation to a
curve in order to solve virtually any equation quickly and rapidly using an
iterative approach.

Up to now you have used iterative methods for solving equations. While in some
cases they can be quick and easy to use, in other cases they are slow and
unreliable. The Newton—Raphson method, although using a more complex
formula, is usually both fast and reliable.

All iterative methods require a first ye

f(x) = x? -3 sinx
approximation to a root, which will normally
be obtained by doing a quick sketch of the
graph. The method can be illustrated by
looking at a specific problem, such as solving
x = 3sinx
= 0.

From the graph, it can be seen that x = 2 is a reasonable first approximation to

the root. At x = 2, f(2) = 4 — 3 sin2 = 1.27 and, since f’(x) = 2x — 3cosx, the
gradient of the tangent is f'(2) = 4— 3cos2 = 5.25.

Using the method of section 7.2.1, Taylor’s first approximation is

Va SELON ELS

By enlarging the region around the root it can be

seen that a better approximation can be found
where the tangent at x = 2 cuts the x-axis, i.e.
where 5.25x — 9.23 = 0. This gives x = 1.758. 2

This method will generalise to give a formula for the improved approximation.

Suppose the equation to be solved is f(x) = 0

and the root you are tying to find is a.

If x = a is the first guess and the tangent at

x = a crosses the x-axis at x = b, then x = b
will be closer to the actual solution a than was
i = ih

You can find the value of b as follows.

 POLYNOMIAL APPROXIMATIONS 485

Since f'(a) is the gradient of the graph atx =a, f(x)

ap See ; f(a)
= fo

=
oy
ae
le)
acai

You can now use this ‘improved guess’ as the starting value in the process and
hence obtain a value for x which is even closer to the root. So if you repeat the
process several times you can get closer and closer to the solution of f(x) = 0.

2A
You have seen that, if x = a is an approximation to the solution of f(x) = 0,
f(a
then b =a — ule), appears to be a better approximation.
f(a)
1 The equation x* — 3 sinx = 0 has a solution near x = 2.
(a) If f(x) = x” —3sinx, write down f'(x).
Use the formula, b = a —
f(a) with a = 2, to obtain a better
f(a)
approximation to the root.
Taking your improved approximation in (b) as your new value for a,
find a new approximation.
Continue this process until you have an estimate of the root which is
accurate to 6 decimal places.

1.76 Zi

The diagram shows how the second approximation, 1.76, is obtained

geometrically from the first approximation, a = 2. Copy the diagram and
show how the third and fourth approximations can be constructed.
486 CALCULUS METHODS 2 ee
___
ee
ee eee

The Newton—Raphson process generates an iterative sequence,

X1, X24 X3, X4,--+)Xpy, and the equation

f(a)
ey ee
"(a)
can be written as

Xnt1 = Xn —

3 (a) Sketch the graph of f(x) =x —6+6e *.

(b) Find a suitable first approximation x; to the positive root of the

equation

x? -6+6e*=0
(c) Write down f’(x) and hence give the Newton—Raphson formula.

(d) Use the Newton—Raphson formula to find the root correct to 4

decimal places.

Although the Newton—Raphson process is generally efficient, it can give rise

to problems. The next example illustrates one such problem.

Copy the diagram and use the method of question 2 to show how to
construct the approximations x7, x3, from the starting value of x, given.
Why is Newton—Raphson not appropriate in this case?

SE (a) Use a computer or calculator with function f(x) = 30 — 15x — 2x* + x°

and x; = 5 to solve the equation 30 — 15x — 2x” + x° = 0.
(b) Do other starting values work? Can you find any that do not? What
happens if x; = 3, 3.1, 2.8, 2.9?
 POLYNOMIAL APPROXIMATIONS 487

(c) What range of starting values converges to the root x = 2?

(d) What happens when: (i) x, = —0.68 (ii) x; = —0.69?

(e) Change the function and solve 5 cos x— x = 0 with starting values
q (en Sie tier —0Ns. Giles? Gv) aj == 6:5
Comment on your results.

(f) Change the function to 2,/x — 1 = 0. Can you find a suitable starting
point?

6E Try to solve sinx — ¥sin (100x) = 0, with starting values x = 1, 2 or 3.

Zoom in on the curve and explain what is happening.

Providing the first estimate, x, is ‘good’, the method will usually converge to a
zero of the function very quickly. A good first estimate is one such that the
graph is locally straight at all points between the zero and the estimate, and has
no turning points. If the initial estimate is itself near a turning point, the
Newton—Raphson method will usually take you further from the zero, and
becomes unpredictable.

In practice, not all equations are of the form

f(x) = 0
Sometimes an equation will take the form

h(x)
= g(x)
488 CALCULUS METHODS 2
Ne ee SE eee

Such an equation can always be rearranged into the form

h(x) = g(x) =0
and so the problem becomes one of finding the zero of the function h(x) — g(x).

Example 2
2
Solve sin2x = x*.

Solution y HEN Se
(Note that you must
work in radians here.)

This equation cannot be Wee

solved algebraically, so a
numerical method must
be used.

The sketch shows that \

there are two solutions.

x = 0 is one solution, and the other is somewhere between x = 0 and x = 3%.

This suggests a possible value of x, = 1.

The equation can be expressed in the form f(x) = 0 by writing

x” —sin2x = 0

Then f(x) =x?—sin2x and f'(x) = 2x —2cos2x

Hence the iteration formula will be

2 ‘
Gi SI 2k,
MEAS ER 2%. 2 COS 2X,
Taking x, =1, then x, = 0.967976, x; = 0.966878, x4 = 0.966877 and
x5 = 0.966 877. Since x4 and x5 agree to 6 decimal places, you can conclude
that x = 0.966 877 to 6 decimal places.

The Newton—Raphson method is usually extremely efficient. You will often find
that if x; is accurate to 1 decimal place, then x 18 accurate to 2 decimal places,
x3 to 4 decimal places, x4 to 8 decimal places and x5 to 16 decimal places!
 POLYNOMIAL APPROXIMATIONS 489

Exercise 2

1 Solve x = cos x correct to 6 decimal places. How can you be sure there is only
one root to this equation?

2 Use the Newton—Raphson method to find the three roots of x* = 3x —1 to

6 decimal places.
; Sis 5 2 ;
3 Find the positive solution of e“ = 3 cosx correct to 6 decimal places.

4 A circular disc, centre O, is divided by a straight cut AB so that the smaller

area ACB is 7 the area of the whole circle.

Show that, if angle AOB = @ radians, then 0 — 7h ='sin@:

Solve the equation and find @ correct to 3 decimal places.

5 Two circles of radius r intersect as shown.

The angle subtended by the common chord at
the centre of each circle is 20.

(a) Find an expression for the shaded area in

terms of r and @.

(b) If the shaded area is equal to i of the area

of one of the two circles, show that
80 — 4sin 20 = x and hence find @ to 4 decimal places.
6 Two savers each have a regular income to invest. Tom invests in a
savings account and calculates that after ¢ years his savings will be worth
£5000(e° — 1). Jerry invests in life assurance and he estimates that after
t years his savings will be worth £700¢(1 + 0.06t).
(a) Who has more savings:

(i) in the short term; (ii) in the long term?

(b) After how long will Tom’s and Jerry’s savings be equal? (Give your
answer in years and months.)
490 CALCULUS METHODS 2

7.2.3 Maclaurin’s series

To calculate the equation of a linear function, you need to know either two
points on the graph or one point and the gradient of the graph at that point. You
need two pieces of information to define a linear graph. It seems logical to
assume that you need three pieces of information to define a quadratic graph,
and this is indeed the case.

If, for example, you know that the graph passes through three points, (—2, 14),
(0, 30) and (3, 24), and if you let the equation of the graph be
f(x) = a+ bx + cx’, you can find three simultaneous equations in the three
unknowns a, b and c.

If, on the other hand, you know that the graph passes through two specified
points and if you also know the gradient of the graph at one of the points, again
you have three pieces of information and can therefore deduce the equation of
the graph.

These are two ways in which a set of three conditions can define a quadratic
function. An alternative set of three conditions is the values of y and its first and
Meee ;
second derivatives, 78 and on at a single point.
;
De Wa

Two pieces of information are never sufficient. For example, there are infinitely
many different quadratic functions, f(x) = a4 + bx + cx’, which pass through the
point (0, 10) with gradient 4.

Since £(0) =10, a+bx0+cx0=105a4=10

f'(x)
= b + 2ex

and since f’(0) = 4,

ble
x 0 =A = b= 4

So f(x) = 10 + 4x + cx” satisfies the given conditions. Since any value of c

will do, there are infinitely many quadratics which satisfy the stated conditions.

If you plot your equations on a graph plotter, they all pass through the same
point (0, 10) and they all have the same gradient (4) at that point. However, you
will notice that some of the graphs are more curved than others. In fact, it is the
rate at which the gradient is increasing or decreasing which is different for each
of them. The rate at which a gradient is increasing or decreasing is easily
calculated. You simply find the gradient of the gradient graph. In other words,
differentiate the function twice.
 POLYNOMIAL APPROXIMATIONS 491

Example 3
Find the quadratic approximation to f(x) = cosx at x = 0.

Solution
Suppose f(x) = cosx % a + bx + cx’, ) a oe |
Then f'(x) = —sinx & b + 2cx, fa O) =O a= 0)
£" (x)= — cos x 2c, f"(0) =-1>c=-}
The quadratic approximation to cos x at x = 0 is
cosx % 1—4x* &

Taylor’s first approximation to a function at a point is a linear function which

passes through the point with the same gradient as the function. A better
approximation to a function can be found by using a quadratic graph. For
example, the graph of p(x) = 1— 1x? is a good approximation to f(x) = cos x,
492 CALCULUS METHODS 2
eee ee

while p(x) =1+x*+ 1x? is a good approximation to f(x) = e*.

Comparing the properties of the function f(x) = cos x with its approximating
quadratic p(x) =1-— 1x? gives:
o£(O)-—1, p(o)}=1
i.e. they ate equal at x = 0.

e f(x) =—sinx > f/(0) = 0, p(x) =—x > pO) =0

i.e. their gradients are equal at x = 0.

e £"(x) = —cosx > f"(0) = -1, p’(x) =-1=> p’(0) =-1

i.e. their second derivatives are equal at x = 0.

Thus a quadratic approximation to cosx at x = 0 is a quadratic which passes

through (0, 1) and has the same gradient and second derivative as cos x.

The cubic approximation will be a function which passes through the point and
has the same gradient, the same second derivative and the same third derivative
as the function.

Higher order polynomial approximations are possible if the function being

approximated can be differentiated repeatedly at a point. You would, for
example, have to differentiate a function five times at a point to calculate the
fifth degree approximation.

_ This idea is developed in the questions which follow.

: > .2B
_ You have seen that when you approximate to a function f(x) using the
| . 2
| quadratic p(x) = a+ bx + cx”, you have to solve
£(0) aun (Obi a0) ae
To extend this method to general polynomials, you have to solve more
equations of this form.
 POLYNOMIAL APPROXIMATIONS 493

1 If p(x) =4+bx + cx’ + dx’ and p(0) = 12, p’(0) = 11, p”(0) = 10 and
p®)(0) =.6, find a, b, ¢ and d.

2 If p(x) = a+ bx + cx + dx’ express a, b, c and d in terms of p(0), p’(0),

p’(0) and p“)(0) and hence show that
2
p(x) = p(0) + p'(0)x + p"(0)— +p (0) >
Ww

3 Explain why, if p(x) is a polynomial of degree four, then

2 3 4
p(x) = p(0) + p'(0)x + p"(0) 5+p" (0) 5+ p% (0) S
4 (a) Ina similar way, if p(x) is a polynomial of degree five write down p(x)
in terms of its derivatives at x = 0.

(b) Generalise this result to a polynomial of degree n.

5 If f(x) =e”, find#(0), f'(0), £"(0), £0), £0).

>» 4x° 2x?
Hence explain why 1+ 2x + 2x* + maeae ar is a good approximation
to e for values of x near x = 0.

Plot both of these functions on a graph plotter and suggest a range of

x-values for which the approximation is good.

If a function f(x) can be differentiated n times at x = 0, then a polynomial of

degree n which has the same derivatives as the function will be a good
approximation to the function for values of x near zero.

The result developed in 7.28 above was published by Colin Maclaurin

(1698-1746). Maclaurin was a Professor of Mathematics at Aberdeen by the age
of nineteen and a Fellow of the Royal Society at twenty-one. In 1742 he
published his book Treatise on Fluxions, which included a description of the
‘Maclaurin series’. He took part in opposing the march of the Young Pretender
when the Jacobites attacked Edinburgh in 1745 and when the city fell he fled to
York, where he died. Despite its name, Maclaurin’s series was first used by
James Stirling.
494 CALCULUS METHODS 2

Example 4
(a) Find the Maclaurin’s series for sin x.

(b) Use the series found in (a) to find the series for sin (x),

Solution
(a) If f(x)=sinx then £(0)=0
f(x) =cosx =» f(0)=1
f"(x)=—sinx => £f"(0)=0
£9) (x) = —cosx => £9) (0) = -1
f(x) =sinx => £4 (0) =0
f(x) =cosx => f(0)=1 and the cycle repeats itself.

Thus, using the values for f(0), f'(0), f”(0),..., Maclaurin’s series
will be
£"(0) x2 £(3) 0 3 (4) 4 (7) 7
Oe ee
0x2 3 4 5 6 7
See ee i eae
2! 3! 4! Si 6! Hs
3
x x x! x?
See SE fag eiec open a

Sees
Vas
 POLYNOMIAL APPROXIMATIONS 495

(b) You can use the series

;
hag 2 ieeee xaE x x?
a RO
aa es eae
to find the series for sin x” by replacing x by x’.

i
= x2
Thus sin(x*) 2 = a
te)
7 a ee)
F ——,
Le) in = 26

a osar wld = 1A
Fe AleeT
& -2C
ee DP Sa ae 6
aAagree
1 (a) Show that CEES E oe rere

(b) Find the Maclaurin’s series for e*.

For the following questions you will need to use the results given below.

For example,
Lf 1 SE saan te rete
= (eae )") = =2(1+ x)

2 (a) If f(x) =Inx, explain what happens if you attempt to evaluate f(0).

(b) Explain why it is not possible to find a Maclaurin’s series for In x.

For the reason given in question 2, Inx does not have a series expansion.
The series for In (1+ x) is found instead.

3 (a) If f(x) =In(1+-~x), write down f’(x).

1
(b) By writing apne (1+.x)~', write down f’(x).

(c) Show that f° (x) = 2(1+<x)~? and find £(4) (x) and £5) (x).

(d) Find £(0), £'(0), £”(0), £°(0), £0) and £° (0).

(e) Show that the Maclaurin’s series for In (1+ x) is
2 3 4 >
fleas he Sek

Snes CRP EST

 496 CALCULUS METHODS 2

Searcr
So

eee
~~"

SS

Can
Se
Se

eat eS
BS
Aras

Sate
So

mae
aS

nee
+

Cee

aos
ea
z+

5 =

—
=e

Oo
Oy,
Ww

age) ey
es
i,
Ss

fe

s&s |
Nae

g
ae
a+

ie
=<
ee
oC

-&
Fw

=
&
-S
a

—
ef

er

|)
a

Ce
+

oe
8
S
2

+
&

3
u
o

a
2
ies
Rel
esr

wast
sy
4
Gy

<i
ot
sou

Swa
LS

—
“N
Se

eS
Ss
-
&
ae

|
|
ESCH)
aE

==
———

ey
Oo
—

wn
wo
4=
SS
ee

nN
R
R
i

es

cy
12)
ao
ai

Not all series are valid for all values of x. For example 2) although

3
pea tg gs ee Sea ore
14+x

for values of x between —1 and 1 , when x is put equal to 2 the result

1
i Dad STs
Ie oy
1 Se

is clearly untrue!

The following standard results are important.

|

7.2, Exercise 3

1 Find the first five terms of the Maclaurin’s series for &

(a) by differentiation and evaluation of £(0), (0)

x
(b) by substituting into the series e* = 1+ x + a x?
ae
 POLYNOMIAL APPROXIMATIONS 497

2 (a) Use your calculator to find cos 0.5.

(b) How many terms of the cosine series are needed to give an answer that is
accurate to +0.001?
(c) How many terms of the cosine series are needed to give cos 1.0 to an
accuracy of +0.001?
3 Write down the nth term of the series for each of the functions:
2
x x
(a) e (b) cos x (c) e
4 (a) Find the Maclaurin’s series for In (1 — x).
(b) Use the result 1 — x* = (1 — x)(1 +x), together with a property of logs, to
find the series for In (1 — x”).
5 (a) Show thate * #1—~x.
(b) Use your result in (a) to show that ,/(1 —e *) © \/x.
(c) What is the P percentage & error in using \/x as an approximation to
/(1—e°*) when x = 0.1?
498

Calculus methods 2
ferential equations
7.3.1 Introduction
Many situations where mathematics is used to model the real world lead to the
formulation of differential equations. Any equation involving a derivative, such as
d :
or S is called a differential equation.
dx “di

For example, the cooling of an object which is hotter than its surroundings is
described by Newton’s law of cooling, which states that

The rate of cooling at any instant is directly proportional to the difference in

temperature between the object and its surroundings.

Using calculus notation, this may be expressed as

dy_ —ky
dt
where y is the difference in temperature between the object and its surroundings.
Va : : ; : 4
as the rate at which the temperature difference is changing with respect to

time. The constant of proportionality is k and the negative sign indicates that the
object is cooling. The equation expressing Newton’s law is therefore a
differential equation.

Because of the frequency with which differential equations occur and their
importance both within mathematics and in its areas of application, the
understanding and solution of differential equations is a substantial element in
any course in mathematics. In this section we consider some of the situations
which give rise to these equations, and consider ways to solve them.

Finding a solution to a differential equation means expressing the relationship

between the variables in a form which does not contain any derivatives. In the
case of the equation for Newton’s law,

dy
Eo
it means expressing y as a function of t. We shall return to this particular
equation as the necessary methods to solve it are developed.
 DIFFERENTIAL EQUATIONS 499

Some differential equations can be solved algebraically by inspection. For

d
example, if you know that = = x’, then integrating the function gives
x
y= 1x? aes

There is a family of solution curves

which all satisfy the differential
equation.

If, for example, you know that the

graph passes through the point (3, 1),
then you can specify the particular
solution as y =4x° — 8.

Example 1
Find the solution to the following differential equation, for which y = 1 when
x= 0,

sa
—_

dx
=> 35x

Solution .
The equation may be solved directly by integration with respect to the
variable x.

| y= |(Gx+2) de
3x?
SNA er oe at

Veale when =i) = cal

me
Son y= a +2x-+1 is the particular solution required.

Notice that in this case the solution involves integration once only, and that this
yields a single constant of integration. In the solution of differential equations,
the constants of integration are a most important part of the solution process,
and must not be overlooked.

Many differential equations can be solved only by numerical methods, and we

consider some methods for doing this later in this section. Equally, there are a
number of techniques for the solution of particular types of differential equation
which do not require numerical methods. Example 2, together with exercise 1
500 CALCULUS METHODS 2
JOG yee en ee ee ee

which follows, will give you some practice with differential equations which can
be solved by ‘simple’ algebraic methods.

Example 2 ‘
For the differential equation “Y — 2x +4 sinx, find the value of y when x = 2 for
dx
the particular solution through (3, 4).

Solution

= [a=+ sin x) dx

=>y=x'—cosx+c, buty=4whenx
=3

>4=3?—cos3+e

=> 4=9+4+0.99+¢
Ep SB. in OD
=> y =x’ —cosx — 5.99 and so whenx=2, y=—1.57 (to 2 dp.)

Exercise 1

1 Solve the following differential equations and in each case make a sketch
showing some particular solutions.
Cy Ea d
(a) —=e
Se (b) Sage er (c) ph Se ea
dx dx

2 For each of the differential equations given below, find y when x = 3 for the
particular solutions which pass through the point (1,0).
dy
(a) oe dy = 3x
(b) a, 2x +2

yan! dy _ ae
(c) ee (d) ae oe

3 A can of water is heated at a rate which decreases steadily with time. The
temperature y °C after ¢ minutes satisfies the differential equation
dy ae #47
dt 3
(a) Find y when t = 4 if y = 32 initially.
(b) Do you think the model will still be valid at time t = 10?

4E For each of the differential equations given below, find y when x = 2 for the
particular solutions which pass through the origin.
d : d 3 d
(a) =, asin (3x + 2) (b) ae (c) a FOS (x? +1)

eee
 DIFFERENTIAL EQUATIONS 501

7.3.2 Direction diagrams

From Newton’s law of cooling, you know that the rate at which a cup of coffee
cools is proportional to the number of degrees, y, that it is above room
temperature.

d
For example, the differential equation oe —0.2y expresses this in symbols, for
dt
a particular cup for which the constant of proportionality is 0.2. The negative
sign indicates that the coffee is cooling rather than heating up.

d
The equation — —0.2y also determines the shape of the (t, y) graph.
dt

At any point, the gradient is —0.2 times the value of y

the y-coordinate of the point. For example, at any
point with y-coordinate 40, the gradient is —8. This 40 \ N
gives the direction of the graph at such points.
You can show this with small line-segments of oe
gradient —8.

2 4 t

BAN VN ONE IN WAY

NaN VINON NEUTAC AeWEAN NNR
RONRABARNEN ABA NES WN EN
AOS NNTAG NGNSN NENG XUNNN Nic NN
NPA WeNGNEAGN A NEA ATAGNAAEN IN Ae shen
KORUNA AOACAANCN, \CNGACA NENA a Similarly, you can calculate the direction at
BL NENEN NOS
ee AN NENT NPN
Oo
NS NES ENENS NNN
no every
.
point. Uhis-is shownuin this direction
.
NNN NNNNNNNANANNA\SNY diagram for some regularly spaced points
SONI Ne AN N NUNN
BLN SNS NNSNISASON SANS in the first quadrant.
SSS ao SS Bas Ses SS SS OSS SS
SS Se aA bh) Da ass
SS SSS Sa SS SN ES BS RES
10 ~~S~SsnSN SSS SS SSS
Ria aa aly ig aN, ah es ae Rey Sy i) Ras a ay Sas

i a i

2 4 6 8 t

By looking at the direction diagram, it is clear that the graph is not a single
curve but is a whole family of curves. There is a different cooling curve for each
different starting temperature the cup could have.

The direction diagram can also be extended into the other three quadrants.
502 CALCULUS METHODS 2
FD twee sin ay cone Pvc os Pore Were SU ie ae ee

The second quadrant represents cooling

SRA
VV VW WS NWA SRO
ASS
OATIS BWA
SRA
ADDN A WANS SAN curves
before the (arbitrarily y chosen) time
EAGAN NTE TEATS INGAA AGASEAN AGEEAGLES ATEN
SNA ANS i#=0
SS WS NAGS AAW WSS VIN NSS NW SRA GSS ANN USN a
DS SES AS WAS SSSA SS SSNIN WSIS WY ASS SAS NSS
SSAVIAN WAU ES SSUES AWN SORES ASIAY SAGAN SSS ARIA SSSI BUA SSIS
POR Se ea

NAC SEeR SENN SOAS mS SS Tin the'third and tourchaquadramcorene

SEN ESS NSENENE AG NE NENT RENIN ENE CENCE ON EOENE NLS EEN, NON NING 6
SSC OS OEN SE EES CONE TSS NTN TENG OTE EROTIC SNUG ENSUE NONGIEN

ES
SSS SNS SSSA Rein SSS SN
temperature difference, y °C, is negative
and this difference gets smaller with
time. These curves represent objects
ae ee ee ge
heating up, for example iced drinks.
ee es Ne ee sere
ee a ee
A A A EC CCAD NVA LUAU CD UAE ATA CGE
Le Ae OA ACA yA AAU? CAA AE CAND
LOE CL Ce NLL CA Aa ENE CAD CREA CEE CAO (A
VOLE DE AMAA GRIN PAL EE GEE. OB OL
EBL MA AE TY DME LID TB TEE OE EAD OU EEG F
ALE MA DEE AU DAA IIE PPE EY
PALOMA DLA AOGAL UT SNL EMRE TEE DAA AQ UME G
VL APM EMT TAPE AMV AEA, PLETE LEE CELL
ZUM MRM ANAM CD UNNI MIMD LDA LLL

This is a direction diagram for the

differential equation
dye
dx 1+x+

Each segment represents part of a solution through its centre. When sketching a
solution curve it is important to try to imagine what the path of the solution
might be as it follows its own path in the same general direction as the nearby
line-segments. A curve which follows the direction of the line-segments is
sketched on the direction diagram.

In the case above, the gradient is

dependent only on x, so the line-
segments in a vertical column (with
fixed value of x) are all parallel.
SN
SS)
SSS
We
The family of solutions for

dx >
are parallel in a horizontal direction.

er
OS
NSIS
EON
RN
nee
ae
 DIFFERENTIAL EQUATIONS 503

The following questions are designed to give some insight into how the
direction diagram is calculated and how it can be used for hand-sketching
of solutions.

@ 3
1 Sketch by hand a direction diagram for
dy
a x

(Use ranges x = —S to 5, y = —5 to 5.) Sketch a solution through (0, 1).

d
2 Sketch the direction diagram for = = —y. What happens to solutions
be
which start with y negative as x increases?

d
3 One of the following pictures is the direction diagram for ra y; the
d dx
other is the direction diagram for = x+y. Which is which? In the case
d x
of — =x — y, sketch a few solution curves and suggest what happens to y
x
as x increases.

P Uae ll ATI LA ANE Ny

Mgl enay aif \ Cy A Aye AN SSS 2k
/ fil \ NAY eX Ney,
/ he} \ Nie es NNN SS and
NWA ie ate CV VS VAMOS SeZ GY UY
IS Lp yy) (PAYA YN Vilas
WAN Cue Way fay,
NEN AENEAN Lier NGVENENEN / i
VPN 72/7 \-2N\ 42/1 !
Nye eee ek \iota NeANN ate I yf NENTS allel ia}
NPN SE SNPANI NS Nes i Hp Ahfi IfIi
VA Wh Sy WEES / af atl I]
VS oa ala Vrsah aA a Ale hl
eo ey NES RA ih fea
WS WS Ae EN haha lanl
ieeeVeVi NU Noe NY eg tall zal
 504 CALCULUS METHODS 2
SS ee

¢ Neb Exercise 2
1 With —5 <x <5, —-4<y <4, sketch those solutions to the differential
d nee
equation Se shich pass through the following points:
dx 2y
(a) (0, 1) (b) (0, 2)
Describe the family of curves.

2 With —4 <x <4,0<y <6, sketch solutions to the differential equation

d é :
= =— — which pass through the following points:
iG
(a) (0,1) — (b) (0,2)
Describe the family of curves.

3 Wilhelm’s law states that, in a chemical reaction, the rate of change of mass is
proportional to the mass, m, of the reacting substance present at any instant.
d
Explain briefly how this leads to the differential equation = = —km where k
is a constant.
d
Consider the case a = —0.5m.

With 0 < t < 5,0 <m < 10, sketch the solution which passes through the
point (0, 8).

Explain the significance of the value 8, and investigate other curves in the
family of solutions.

7.3.3 Numerical methods

Numerical methods for solving differential equations depend upon
approximating the solution curve with a series of straight line-segments.

If (x, y) is a point on a graph, then the point (x + 6x, y + dy) is on the tangent.
For a locally straight graph, and for small 6x, the point (x + 6x, y + dy) is close
to the graph.
A

i ra
fo aii 2od
 DIFFERENTIAL EQUATIONS 505

An approximate numerical solution can therefore be obtained by fixing a small

numerical value for 6x, then calculating a point farther along the curve as
(x + dx, y + dy) and taking this as a new starting point. Consider a particular
situation for the simple differential equation = = 2x and starting from the
point (1, 4). si

(1.1, 4.2)

At (1,4) the gradient is 2, so a step along of 6x = 0.1 by =0.2

will give a step up of dy = 0.2. The move along the
tangent is from (1,4) to (1.1, 4.2).
yA)
6x=0.1

The gradient at (1.1, 4.2) is 2 x 1.1 = 2.2.

Since 6x = 011, then éy =.0.22 (6= = x ix)

Ea

So the next point will be x+6x = 1.2

y + by = 4.42

It may help to set the calculation out in a table, working down column by
column.

d
x y cus 6x by x + 6x y + dy
dx
First step 1 4 Wd 0.1 0.2 ils 4.2
Second step icf 4.2 Upp 0.1 0) ee 4.42

Much numerical work of this sort is ideal for working with a spreadsheet or
calculator. The next few steps in the solution are set out in the spreadsheet on the
following page.
 506 CALCULUS METHODS 2

x +dx/y + Oy
(lel 4.2
1.2 | 4.42
aah | 4.66 |
0.26| 1.4 | 4.92
ec EN 0)
0.3 | 1.6 |5.5
RA EEY
34] 1.8 | 6.16
[1.9 |6.52

4 fs ie sit TAS
1 1.2 1.4 1.6 a

In this example, the actual solution curve is v=o 3,

So y = 4.21 when x = 1.1, whereas y = 4.2 is

obtained by moving along the tangent instead of
the curve.

The accuracy of the step-by-step method could be

improved by using smaller step sizes for dx.

Example 3
For the differential equation

Cy Blt
die tens
find the value of y when x = 2 for the particular
solution through (1, 6).
 DIFFERENTIAL EQUATIONS 507

Solution
The table below shows the start of the calculation, using five steps of
6x = 0.2 and working to 3 significant figures.

x y
dy
Ae 6x by x + 6x y + dy

il 6 3 0.2 0.6 i 6.6

daz 6.6 PES) 0.2 0.43 1.4 703)
1.4 7.03 1.45 0.2 0.29 1.6 ESO
IEG) Hay OO 0.2 Opi, 1.8 Higayll
1.8 Sew tl 0.65 0.2 0.13 2.0 7.64
Z.0 7.64

The step-by-step calculations are straightforward with a calculator but it is more

efficient to use a program or spreadsheet as shown opposite.

Exercise 3
d
1 (a) What is the equation of the solution curve for ES ecoseeawiich passes
dx
through (0, 0)?
d
(b) Calculate the numerical solution of the differential equation a cose
dx
starting at x = 0, y = 0 with step 0.1. Record the values of x and y for
x = 0, 0.5, 1 and so on, giving y values to 1 decimal place. How accurate is
the numerical solution for different values of x?

d
2 (a) What is the equation of the solution curve for “3= 4x° which passes
OS
through (1,0)? Find the value of y when x = 2.
(b) Use a numerical method, with a step size of 0.2, to find the value of y
when x = 2.

3 (a) What is the equation of the solution curve for = = sin 2x which passes
through (0, —0.5)? ‘
(b) Using a step size of 0.1, find the percentage error in the numerical solution
for yy. at.x— We
508 CALCULUS METHODS 2
eee ee eS ES Ee eee eee

d
(a) eee ae
dx

For = = 2x, what is the particular solution curve through (1,2)?

Be
Plot this solution curve on graph paper for —1 < x < 3.
d
S Calculate a numerical solution of Oy
dx
Startins ata 1 ya

Using steps of size 0.2, record the values from x = 1 to 3 and round off
the y values to 1 decimal place. Record also the values from x = 1 to —1.
Plot the numerical solutions for —1 < x < 3 on the same diagram as your
graph of the solution curve and comment on the accuracy of the
numerical method.

eo Repeat part (b) with step size 0.1. Comment on the improvement in
accuracy obtained with the smaller step value.

SE For equations where you can find a solution curve by integration, it is

possible to write a program to check the accuracy of the numerical method
for different-sized steps. The program could either print out both the
numerical solution and the correct value of y as obtained by integration, or it
could plot the numerical solutions and then superimpose the correct graph
over them. Use one of these methods to investigate the accuracy of numerical
dy
solutions for — = e* for different step values, starting at x = 0, y = 1.
dx

6E Write a table of values for a numerical solution to

Cy ig A
dx 1+4+x

starting at x = 0, y = 0, with step 6x = 0.1, from x = 0 to x = 4. Record

selected values of x and y sufficiently accurately to plot a graph.

Also write a table of selected values starting at x = 0, y = 0, with step

6x = —0.1 and sketch the graph for the solution through x = 0, y = 0 from
x = —4to +4.

Does the graph have a shape you recognise? If not, reflect it in the line y= x
to consider x as a function of y. Suggest a possible relationship between x
and y and check a few values to see if your graph is (approximately) in
agreement with your suggestion.
Write a brief report on what you find.
eee
 DIFFERENTIAL EQUATIONS S09

7.3.4 Growth and decay

eas:
The differential equation See: y generates a
dx
family of solution curves which look like
the graphs of growth functions.

ne
The equation - = y cannot be solved by algebraic integration as it stands.
4
However, a simple rearrangement is helpful.

i lgcea
dea dy y
|;
= chy
y
=>x=In|y|+c
>x+k=Inly| (k = —c)
Big hue

>y=tee’
Sie (A = +e*)
The solution curves are indeed the graphs of growth functions.

The solution a be generalised easily to solve = = Xy as follows.

Ke
dy
sce | gx
Sess a
dx “ee dy Ay
1 |il
Sy
AJ y
=> Ax = Inly| +c
=> dx +k = In|y| (k = —c)

=>y=tete*
>y=Ae*“ (A=+e')
d a
Differential equations of the form = Ay generate families of solution curves
dx
which are the graphs of growth functions.
510 CALCULUS METHODS 2

Example 4
dm
Earlier in the chapter, you considered the differential equation Baa —0.5m,
where mkg was the mass of a reacting chemical present ¢ hours after the start of
the reaction. Initially there was 8 kg of the substance. Find the mass of the
substance after 1 hour.

Solution
dm
—— = 0.5m >m=Ace°™
dt

Whend — 0.71 =o. A — 5

Then m= 8e °” and, after 1 hour,

m=8e°° © 4.85
After 1 hour there is approximately 4.85 kg of the substance remaining.

Example 5
A murder victim was discovered by the police at 6:00 a.m. The body temperature
of the victim was measured and found to be 25°C. A doctor arrived on the
scene of the crime 30 minutes later and measured the body temperature again.
It was found to be 22°C. The temperature of the room-had remained constant
at 15°C. The doctor, knowing normal body temperature to be 37°C, was able
to estimate the time of death of the victim. What would be your estimate for
the time of death?

Solution
Assume Newton’s law of cooling:

The rate of cooling at any instant is directly proportional to the difference in

temperature between the object and its surroundings.

t hours after 6 a.m., let the body temperature be y°C above the temperature
of the surroundings.
dy X
Be
AY = Jy We”
= Ae

When!
? = Olindiy 310;
10= Ae? > A=10
 DIFFERENTIAL EQUATIONS 511

When ¢ = 0.5 and y =7,

T= Wer = 058= Inve
EES O13" tos of)
So y = 106073

You can find the time, t, when the temperature was 22 degrees above room
temperature.
y=22>22=10e °’
=> In2.2 = —0.713t
=> t= —1.106

1.106 hours is 1 hour 6 minutes. The time of death was about 1 hour
6 minutes before 6:00 a.m., or 4:54.a.m.

7.3 Exercise 4

1 Find the equation of the solution curve of

dy _
Te
which passes through the point (0, 2).
2 The mass, mkg, of a substance satisfies the differential equation

Sa
where ¢ is the time in hours after the start of a chemical reaction.
(a) Initially there was 2kg of the substance. Find a formula for m in terms of t.
(b) Hence calculate the time taken for the mass of the substance to be halved.

3 Boiling water is left in a room and cools to 90°C in 5 minutes. If the room
temperature is 20°C, how long will the water take to cool to 60 °C?

4 A colony of insects initially has a population of 100 and is growing at the rate
of 50 insects per day. If the rate of growth at any time is proportional to the
population size at that time, how many insects will there be after 10 days?
5 A radioactive substance decays at a rate proportional to its mass. When the
mass of a sample of the substance is 0.020 it is decaying at a rate of 0.001 g
per day. There are m grams left after ¢ days.
(a) Formulate a differential equation connecting m and t.
(b) How long does the sample take to decay to 0.010 g?

7.3 Tasksheet E1 — Carbon dating (page 566)

512 CALCULUS METHODS 2
See

7.3.5 Formulating differential equations

You have studied some of the numerical and analytical methods for solving
differential equations. The geometrical picture of the family of solution curves
is also valuable. In particular, it indicates the effect of different initial
conditions.

The relative advantages and disadvantages of solving a differential equation by

inspection or by a step-by-step method can be summarised as follows:

To be able to apply either of these methods you must be able to formulate the
differential equation correctly. This section considers a few examples of
formulation.

Example 6
High on the moors, perched on a rocky crag, lies a most curious boulder. It has
fascinated the locals and tourists alike for years, for it is almost perfectly
spherical in shape. Over the years it has gradually been eroded by the action of
the winds, but it has retained its basic shape. In fact, according to the locals, it
now has half the diameter it had 100 years ago. ‘Be gone in another 100 years’
they say. Are they right?

Solution
Set up a model
The locals will be right only if there is a linear relationship between the
radius of the boulder and time. To decide if this is the case you need to make
some assumptions about the rate of erosion and decide upon the variables
and units to be used.

Volume of boulder Vem?

Radius of boulder rom
Time from 100 years ago tyears
 DIFFERENTIAL EQUATIONS 513
—_ereaeeeee” OO ee

It is reasonable to assume that the rate of erosion is proportional to the

hare Fades ads
surface area, A = 4m’. Since the rate of erosion is — =e the important
: : ‘ ? t
differential equation is

dV ed A
TS “dp = kA ies ae = —4krr

Analyse the problem

4
Since — 3 Tr,

dv
melee 4 ae9 dr
aReen! eSr a by the chain rule

dr
=4rr
soe—

So 4nr° eS= —4krr°

dt
dr
= —=—k
dt

Interpret/validate
The radius decreases at a constant rate and so the boulder will be gone in
another 100 years. You cannot validate this conclusion directly although you
could find out known facts about erosion rates. In fact, it is unlikely that the
boulder will retain its shape as it is eroded.

Example 7
A child makes his way to school at a speed which is proportional to the distance
he still has to cover. He leaves home, 2km from school, running at 10km Hott
How long will it be before he has gone nine-tenths of the way?

Solution
Let x km be the distance from home that he has travelled at time ¢ hours. At
that instant he will be (2 — x) km from school.

First record the initial conditions.

dx
<= 0 “and -——-=10 when
t=0
dt
Next, write down the differential equation which translates into symbols the
statement ‘the rate of change of distance from home equals a constant
multiplied by the distance from school’.
dx. 5
epee ar?)
 514 CALCULUS METHODS 2
ee

Substitution into
Then use the second initial condition to find the constant k.
the differential equation gives
10=2he= k=)
and the differential equation becomes
dx
eric 5(2
— x)

or

ag. gee . &=12(£))

dx 5(2—x) a de eae
The general solution of this equation is
i —fln |2—x|+c¢ (check this carefully)

Using the first initial condition, you can find the particular solution which fits
the problem. x = 0 when t = 0 gives
O=—{ln2+¢

SBS Ny

=> t=! [In2—In|2 —~x|] =4ln é

5 5 2-—x

When the child has travelled nine-tenths of the way, x = 1.8 and
¢== (In2.—In0.2]
= ¢In 10
=~ 0.46

So the child has gone nine-tenths of the way about 28 minutes after he leaves
home.

Although the model we have been using is plausible enough for the earlier
part of the journey, it is inappropriate for the whole journey to school. Why?

7.3 Exercise 5

1 The volume of a large spherical snowball decreases as it melts at a rate

proportional to its surface area at any instant.
(a) Express this statement in symbols.

(b) Given that a snowball of radius 30cm takes 10 days to melt, find an
expression for the radius 7 in terms of the time t.
(c) After how many days will:

(i) the radius be halved, (ii) the volume be halved?

 DIFFERENTIAL EQUATIONS 515

2 Ina lake, about 2000 newly hatched fish survive each year. However, about
10% of the fish in the lake die each year as the result of disease, predators or
old age. These observations lead to the hypothesis that
d
Ee + By where y is the number of fish present.
dt
(a) Explain why @ = —0.1 and state the value of a.
d
(b) In what units are y and “=measured?

(c) The lake had a stable population of 20000 fish before a careless discharge
of chemicals killed 5000 fish. Estimate how long it will take for the
population to reach 19000.
(d) Write down the differential equation which will apply if 12% of the fish
die each year and 2500 newly hatched fish survive.
3E A full tea urn contains a hundred cupfuls.
Nine cups are filled from the urn in the first minute. The rate of flow is
believed to be proportional to the square root of the height of the liquid in
the urn.
Explain the relationship in symbols, stating their units and meaning.

How long does.it take to fill four dozen cups?

[This question can be tackled either analytically or numerically. Use
whichever method you prefer.]

7.3.6 Separating the variables

With the differential equations that we have so far solved by algebraic means, it
has always been possible to do so with a single direct integration. Many other
differential equations may be rearranged so that they are in a form suitable for
direct integration. We examine here one important procedure for doing this for
certain forms of differential equation.

d
Consider the problem of finding the general solution to the equation pee xy.
dx

The solution requires one integration with respect to the variable x. However, on
the right-hand-side this leads to the integral

| |xvdx

Since y is a function of x which is not known (that would be the solution

we are trying to find!) then it is not possible to evaluate this integral.
516 CALCULUS METHODS 2
se ee Pea eee ee ee ee

However, we may proceed as follows

Integrating with respect to x,

|G) dx [vas
y dx

|54y= [vas
y
2
In|y|
oP =—+
=—+¢
ier

y= peer ee FP

or y= Ae? (where A is a constant)

The technique illustrated above depends on your being able to ‘separate’ the two
variables (here x and y) so that one side of the equation consists of a function of
x alone, and the other side has a product of dy/dx and a function of y.

In other words, you must have an equation which can be rearranged in the form

f(y) = = g(x) ®
Then, integrating with respect to x, you obtain

[t0)
f(y ied2 ex=[emdr
Peay
= ©
or

|f(y) dy = |g(x) dx @
Provided the two integrations can be performed, the differential equation has
then been solved. The method is called separating the variables.

In practice @) is usually omitted and you can proceed directly from @ to @).
However, you should bear in mind that it is integrating both sides of @ with
respect to x which leads to @), in which one integral is with respect to x and the
other with respect to y.

Example 8 i
Find the solution to the differential equation ae x(y + 1), given that y = 2
when x = 0. at
 DIFFERENTIAL EQUATIONS 517

Solution
dy
eee ji ek ectp sdy,
ee e
dx a0) ) y+1 dx _
Integrating with respect to x,
1 dy
|gee|ee

1
| aay |vee

x2
Injy+ I= +e

Wihentx
= (0, V2.
ln =
2
So, In |y + | — In3 ==

yo x?
In =—
8) 2

7.3 Exercise 6

1 Find the general solution to the following differential equations.

VG dyera
_ ers
Ore (c) din
=m _
aye dys yer]
eee y te) dx x+1

2 Find the solution to the differential equation

for which y = 0 when x = 0.

518 CALCULUS METHODS 2
GPO cee OER A ee Ra eS) Eee ee ee

3 The rate of increase of the population of a colony of insects ¢ days after the
beginning of an experiment is proportional to the population P at that time.
If the colony initially contains 100 insects and at that timeis growing at a rate
of 50 insects per day, find how many there are after ten days.

4 A radioactive substance decays at a rate proportional to its mass. When the

mass of a sample is 0.020, the decay rate is 0.001 g per day.
(a) t days later the mass is mg. Write down a differential equation and solve it
to give m in terms of t for these initial conditions.
(b) How long does the sample take to decay to 0.010 g?
(c) How long does it take to decay from 0.012g to 0.003 g?
(d) What is the rate of decay after 10 days?

5 Bacteria in a tank of water increase at a rate proportional to the number

present. Water is drained out of the tank, initially containing 100 litres, at a
steady rate of 2 litres per hour. Show that if N is the number of bacteria
present at time ¢ hours after the time at which the draining starts, then
dN 2N
ee SS IN eee
dt 100 — 2¢

If k = 0.7 and, at t = 0, N = Nog, find in terms of No the number of bacteria

after 24 hours.
DIFFERENTIAL EQUATIONS 519
520

Calculus methods 2
Miscellaneous exercise 7

1 Calculate the volume of the cone formed by rotating the area between
yi 5X, the line x = 4 and the x-axis, about the x-axis.

i) Find the volumes of the solids formed when the following areas are rotated
about the x-axis:
y= sina fer 0 =...
<7 (b) y=2e* for—=la
x7<1
y)
(c) y=— forl<x<4
Ve
Qe Find the values of:
1 14m

(a) | (can? x +1) de wb) | tan? x dx

0 0

4 Try to integrate the following, using your experience of chain-rule

differentiation. Alternatively, use a substitution.

(a) [axa —~x*)>dx — (b) lea dx — (c) [eva ~ x°) dx

3 5
x sin x
(d) |e dx (e) |sin’ xcosx dx (f) |S = dx

5 Using the suggested substitutions, integrate the following:

3 2
(a) |xe -1)*dx saa al (b) [2evG +») dx u=1+x
0
4
2 ‘(ese
(c) [+ )vGx +1) de, u=3x+1 @) | as, u=1+x
0 0 (1+ <x)
6 Evaluate:

(a) |xsin x dx (b) lecos x dx (c) |xsin? xde

(For (c) use the formula for cos 2%.)

 MISCELLANEOUS EXERCISE 521

Evaluate: (a) | x sin x dx (b) |x sin;x

dx
0 0

(a) EERE Ree ee

(x—2)(xn+5) x-2 x+5
4x +2 A B G
b) Find A, B and C when —————__~_
= — + ——
Re a Peel) Ger) GAS ee

Use the method of partial fractions to find:

(a) |g ae (b) lo AG
x — 3)(x +1) x?
— 16
1 x + Sx*—3x
+3
eee
10 Use the Newton—Raphson method to find to 3s.f. the greatest and least roots
of x —3x+1=0.

11 Make an estimate of the root of x — cosx = 0, and improve it by a single

application of the Newton—Raphson formula.

12 Draw a graph to show the approximate locations of the roots of

x’ — 5x +3 =0. Use the Newton—Raphson method to find the largest root
correct to 4s.f.

13 Using the series expansion for e* or otherwise, obtain the power series
: ae 3
expansion, as far as the term containing x’, for
(ay ex a E(bie (c) In (1 + 2x)

14 Use the Maclaurin Series to find cubic approximations to the following:

(a) (1+x)? — (b)

15 Find, as far as the term in x*, Maclaurin Series for

(al tatxe pub) csecx (c) e* sinx

is ;
16 Using the chain rule, find i when
dy
x
(a) y = sin (x? +1) _ (b) y = cos (1 — 2x) (c) y=
e*
(d) y= esinx (e) y= e(sinx + cos x)
522 CALCULUS METHODS 2

ds
17 The size S of a population at time ¢ satisfies the differential equation — = kS
dt
where k is a constant.

Find S as a function of t.

The population was 32000 in the year 1900 and had increased to 48 000 by
1970. Estimate what its size will be (correct to the nearest 1000) in the year
2000.

18 Solve the following differential equations; in each example, use the

information given to find y as a function of x.

(a) —=y
dy and y=10whenx=0
dx

Gees, and) i= 2 when—=i1

dx

(c) ye
oe 2 and 1
y=1whenx= 5
 equation of a straight line
Use a graph plotter to sketch the graph of y = 2x + c where
é= 2, —1, 0, 1 2:

Describe the effect of varying c.

Use a graph plotter to sketch the graph of y = mx + 3 where

m = —2, —1, 0, 1, 2.

Describe the effect of varying m.

A gradient of 1 in 5, i.e. a rise of 1 unit for every 5 horizontal units travelled, is

described mathematically as a gradient of or 0.2.

Example
Find the gradient of the line AB.

Solution
ie tee
The gradient is ROT

3 Plot the points A (1,2) and B (3, 10) and calculate the gradient of the line AB.

4 Find the gradient of the line AB for:

 (a) Find the gradient of the line.
(b) Where does the line cross the y-axis?

(c) The equation of the line is y = 2x + 1. Show

HH that the points (0, 1) and (2, 5) satisfy this
equation.

HHiat (d) How can you relate the equation y = 2x + 1 to

itei
your answers to (a) and (b)?
TASKSHEET
S1
1.1

6 (a) Find the gradient of the line.

(b) Where does the line cross the y-axis?

(c) The equation of the line is y = —3x — 1.

Choose two points, and show that their
coordinates satisfy this equation.
(d) How can you relate the equation
y = —3x — 1 to your answers to (a)
and (b)?

7 Find the equations of the lines.

| is oa |
4
}
ase
CoE
}
HHan ro aae8
a
Tr
H : 217
A nee aeesaeae SN | H : EEECEHEEH
rH
AERA
SSPE
Senn’ oy peaeaeseeeeenes
iase aes seststreres
8 There are many alternative ways of writing y = mx + c. Find the gradients
and y-intercepts of the following straight lines.
(a) y=9-—4x (b) 2y = 6x +3 (c) 2y-—4x =5 S1
TASKS
1.1
iS )
(d) 3x+5y+10=0 (e) y-3=S(x-2) § (f) y= 4
3
=f

You can use the form y = mx + ¢ to find the equation of a straight line which
passes through any two given points.

Example Z
Find the equation of the straight line which passes through the points (2,5) and
(55,11).

Solution
The slope is g = 2)
aS The Miaion isy=2x+c.
The line passes through (2,5),
so JS 45-6

=> 64

_ The line has equation y = 2x + 1.

9 Find the equation of the straight lines through

(a) (5,4) and (10,19) (b) (—3, -2) and (1,2)
 Multiplying brackets
You have met expressions like 3(x + 2) before.
3(x + 2) means ‘3 lots of (x + 2)’ or 3 lots of x’ plus ‘3 lots of 2’.

ie. 3(x +2) =3x+6

1 Multiply out:
TASKSHEET
1.1
S82
(a) 5(x +3) (b) 2(x — 4) (c) 8(2x +5)
(d) —2(x + 6) (e) —4(x — 7) (f) 6(x — 2y)

Complicated expressions can sometimes be simplified by multiplication of any

brackets, for example,
6x(x — 1) — 2(2x — 3) = 6x” — 6x — 4x +6
6x" — 10x + 6
Multiply out and gather together like terms for:

(a) 34+2(x+3) (b) 3(x —4) (<) @5(8 =a)

(ee 4) (ec) p—1+2(3p—8) (f) 5—6(5x —9)
(g) y-9%(y—2) — (h) 4x —x(2— x) (i) 2 — 3x(1 + 2x)

You have seen how to multiply out expressions of the form 2x(x + 1). You can
now consider expressions like (x + 2)(x + 4).
(x + 2)(x +4) means ‘x lots of (x + 4)’ plus ‘2 lots of (x + 4)’.
So: (x + 2)(x + 4) becomes x(x + 4) + 2(x + 4)
and the problem has been reduced to the earlier form with which you are
familiar.

3 Multiply out:

(a) (x +2)(x +4)

(d) (x — 5)(x — 2)
(uC 2)(x = 9)
 CN
Expressions of the form (x + 2)* are called perfect squares. (x + a means Sa)
(x + 2)(x + 2) and can be multiplied out in the standard way. ee
ae
Ll
4 (a) Multiply out: a0
Gy eee) ei OS ne) ia)

(b) If (x +p)? =x? + bx + c, express: —

(i) b in terms of p (ii) c in terms of p aa
 Further factorisation

1 (a) Write in the form x +bx +c:

(Geese 2G 3). a ibn 2) (3)
(iii) (x +4)(x+5) (iv) (x — 4)(x — 5)
(b) How is the constant term c related to the numbers in the brackets?

(c) How is the coefficient b related to the numbers in the brackets?

‘1.1
$3
TASKSHEET
When you multiply the brackets (x + 4)(x +7), you find
the coefficient of x by adding 4 to 7,
the constant term by multiplying 4 by 7.

(x +4)(x +7) =x7 + (447)x+4x7

x’ + 11x + 28
When you factorise x* + 11x + 28 you have to do the opposite and find two
numbers

(a) whose product is +28 and (b) whose sum is +11.

Since 28 has a limited set of factors, i.e. (428, +1), (+14, +2), (+7, +4), it is not
hard to see that the numbers must be 7 and 4.

2 Factorise:

(a) x7 +9x+14 (b) x7 +13x4+40 (c) x*-9x +14

(d) x*+12x4+36 (e) x?-7x-8 (f) x” + 3x — 28
(g) x? —8x+4+12 (h) x” — Sx — 36 (i) x” — 2x — 48
(ie 2 da 24

You should be familiar with the following important special cases.

e If the constant term is missing, x will be a factor, for example

x? + 6x = x(x +6)

e If the expression has the form x* — a’, then it factorises into (x — a)(x +a),
for example
x — 16 =x° — 4° = (x —4)(x +4)
e Expressions of the form x* + a” will not factorise.
 on
3 Factorise where possible: y)
(a) x +2x 7
i
jaa
ea
(d) x* +25 (e) x7 +25x | an
Y)
(eee (h) x*-1 na
7
(j) 6x7 +5x+1 — (k) 2x*+7x+6 <x
(m) 9x* — 4 a
Review of equations
You have already met several types of algebraic equation. This tasksheet reviews
methods of solving such equations and gives you an opportunity for further
practice.

Linear equations

Example
Solve 6x + 14 = 2x — 6.

Solution
2x — 20 (subtracting 14 from both sides)
—20 (subtracting 2x from both sides)
—5 (dividing both sides by 4)
1 Solve:
(a) dx +30) =0 (b) 4x +3
= 2x (c) 5x +2=3x—-—7

(d) 4(x —3)

= 6x (e) 3(2x +5) = 3(% +2) (f) x+4=3-2x

Quadratic equations

Example
Solve the equation x” + 6x — 7 = 0.

Solution
x +6x—-7=0 (Try to factorise first.)
Se al (a) 0
= either) ge 1-0) or x +7—0
— ee Ot 7

2 Solve:
(ae (x —S\ia+5)= 0° *(b) x(x2) = 0. (ome yea ano
(d) x* —3x-—10=0 (e) x* — 4x =0 (f) x7 -6x+9=0
Other methods
Some equations can be solved by inspection, though care sometimes needs to be
taken not to miss solutions — for example, x* = 9 clearly has x = 3 as a solution,
but it also has the solution x = —3.

Other equations may need some rearrangement before solving.

Example
Solve the equation
6
x= —<—_—
x+1

Solution x(x 1) = 6 (multiplying throughout by x + 1)

x +x =6 (expanding)
Lae = b= 0 TASKS
1.4
S1
(x —3)(x+2)=0 — (factorising the quadratic)
=>x=3 or x
3 Solve:

2 1
<=
ans 1
7 oie
CEs
x eR
Rl
a
SO RWl
Bo
ee
Nl
laced

Simultaneous equations may be solved by a variety of methods; the method

shown here is substitution from one equation into the other.

Example
Solve the simultaneous equations 2x + y = 13 and 3x — 2y = 9.

Solution y=13-—2x (from the first equation)

34 —2(13 = 2x) = 9 (substitute into the second)
Ti — 26 = 9

x5

eh i 2X

SON =) See

4 Solve the simultaneous equations:

(a) x+y=8 (b) 2x + 3y = 17
3x —2y = 19 ; 2x-y=5
 Working in surd form
1 The diagram shows a spiral of triangles based
around an initial right angled triangle. i

1.4
TASKSHEET
$2

Find the lengths of the hypotenuses labelled 41, u. and uw; in the diagram.

It is best in this example to work in exact form, where you observe the lengths
to be \/2, /3 and \/4. This suggests that the next might be \/5, which it is, and
that the mth length u,, will be \/n.

This pattern would never be apparent if you worked in approximate form,

where the first few lengths would be
1414) (e732) 2h 2936. 249)
Numerical expressions which contain one or more irrational roots of numbers,
such as 3 + \/5, are known as surds. Familiarity with the methods of working
with surds often means that patterns can be easily spotted, and that numbers
containing irrational parts can be written and dealt with in exact form.

2 Evaluate, leaving your answers in exact form

(Qe 5) eee) ae. 13). (cGy Se
diGaw2) Nels —v/2) 2 SO) G73 3./5)5
: D
3 A circle has an area of 50cm’. Show that the radius is 52 cm and the
circumference is 10,/(27) cm. si

4 Solve, leaving your solutions in simplified surd form:

(a) x°+5x+1=0 (b) x* —3x-9=0 (c) 3x7 +7x —3 =0

It can easily be shown, by squaring each expression, that

= a_ Ja
Jab= a/b and ar

These results are useful when simplifying surds.

5 Express each of the following in terms of the simplest possible surds.
/5
V/8.. (b) /0 . ©) 2 @ Wie
6 (a) Show that if a and b are rational (i.e. contain no surds), so is
(a+ /b)(a— Vb).
(b) Evaluate: (i) (2+ \/3)(2 — V3)
pee 2 Ae2,/5)
Gee eto 4/3)
The results obtained in question 6 help in dealing with fractions which contain
surds. It is usual to simplify them so that the denominator of the fraction does
not contain any surds — it is rational. Such a process, illustrated in the following
example, is known as rationalisation (of the fraction).

Example
2/2—/3
Rationalise the denominator of ————"—
vt Se
aoe Jf/2+/3

Solution
2/23, 2/8 _4= 22/3 =/3/2 +3
pha en Saka 2=3
=3/6-7
7 Rationalise the denominators of the following.
NO Ne Ge een
(a) ee Mee aS
(d) MAT lag ane SABE
Vi+l res
Write down the exact values, using surd form as necessary, for the sine, cosine
and tangent of 30°, 45° and 60°.
Hence express in surd form, and rationalise, the following expressions:
i her 1
(a) 1+ sin. 45 ° (b)
sin 60° — cos 45 °
cr °

1 + sin 60°
(d) (1 + tan 30°)*
1 — sin 60°
Expanding brackets
When multiplying out more than two brackets, a well-organised, methodical and
careful approach is important if careless mistakes are to be avoided.

With three sets of brackets

(x — 1)(x +3)(x — 4)

= (6 Wer == 1D) _ Expand brackets in pairs only.

= x(x* — x — 12) —(x°—x-—12) This may be omitted, but errors

with signs are common and this
is when they usually occur.

ee
1x x xe 12
= Dx 11x 2 Gather together like terms.
1 Expand:
(a) (x-1)(x7>+x+1) — (b) (x —2)(x? 42% +4)
(c) («+ 1)(x7-x+1) — (d) (x +.2)(x? — 2x +4)

2 Expand:
(a) (x+1)(x+2)(@—4) — (b) («= 2)(x— 3)(x — 4)
() @=-H@tLDe@+s) dd @a1~e43)
With more than three sets of brackets

(x + 1)(x — 2)(x — 3)(x +2) The pairs of brackets may be

expanded in any order. Your
= (x + 1)(x* — 4)(x — 3) experience might help you to speed
up the process by choosing a
= (0 $1) (x? — 3x7 4h 12) convenient pair.

= x" — 3x? — 4x7 4 12x Careful setting out makes any

Oe ee rp) simplification easy.

=*' — 2x? — 7x? + 8x + 12

3 Expand:
(a) (w+ 1)(x—1)(@+2)(x-2) — (b) (x +2)?(« — 2)?
(c) (x =1)(x +3) (d) (x + 1)(x — 2)(x
+ 3)(x — 4)
Functions of functions

1 The functions f and g given by f(x) = x* and g(x) = 3x + 1 can be described

by the following flow charts.

square

TASK
3.1
S1
multiply by 3 3x + 1 = g(x)

(i) What is £(4)? (ii) What is g(16)?

If the output from f(x) is used as input to g(x), you can write g(f(x)).
What is g(f(4))?

Write down: (i) g(f(2)) (ii) g(£(—3)) (iii) £(g(—2))

Draw a flow chart to describe the composite function gf(x) and hence find
a formula for gf(x).
Check whether your formula is correct by substituting a few numbers
in gf(x) and in your formula.
Now find a formula forfe(x).
Use the ideas above to find gf(x) and fg(x) for:

(i) f(x) =-, e(x)=x-3 (ii) £(x)

ae des
(v) f(x)
(b) Comment on the cases where fg(x) = gf(x).

Each of the expressions below is of the form fg(x) where f(x) = 1 — x*. What
is g(x) in each case?
(aie a2) i(bad as" ic) es (d) 1—x
x

4 Each of the expressions below is of the form fg(x) where g(x) = x°. What is
f(x) in each case?
(a) 2° +8 (b) x® (c) 3x2 +1 (d) = (e) x (f) 4x? — x®
 earranging formulas
O ° 2 Qu—oO eas)oO : & I ts |
| y G oO 2)rus)o Quycp) Pp° is}~ oO R — oO i) ° el Te)a o
ras)G oO D =) 9) uwle) uy = —iss} a
wy Ww i Ta
a.

y =5(x—7)
= (x —7)* divide both sides by 5
LV

=i) find the square root of each side

OY

add 7 to both sides

7
yy

2 Make x the subject of the formula:

3
3 Complete the steps to find x in terms of y if y 4
x

3 — 4x multiply both sides by x

add 4x to both sides to collect terms in
x together

factorise the left-hand side

4 Make x the subject of the formula:

Solving equations
This tasksheet is designed to emphasise and give extra practice in the
methodology needed to solve equations involving sine and cosine functions.

1 To solve the equation 4cosx° = 3:

(a) Write down the value of cos x°.

(b) Use your calculator to give one solution for x. TASKS
3.2
S1
(c) Sketch the graph of y = cosx°.

(d) (i) Mark on the graph the solution from the calculator.
(ii) Mark on the graph the other solution between 0 and 360.
(iii) Write down the value of this other solution.

Find all the solutions between 0 and 360 for the following equations,
illustrating your answers with sketch graphs.
(a) cosx = 056 (b) sinx® = —0.23 (c) cosx® = —0.5

Find all the solutions between —180 and 180 for these equations, illustrating
your answers with sketch graphs.
(a) “sin. x? = 0:65 (b) cosx° = —0.38 (c) sinx® = —0.47

Find all the solutions between 0 and 360 for the following equations.
(al asisiiwx == 2 (b)- S:cosx”
+2 =0 (c) 2cosx° +5=0

Copy and complete the solution of 5 sin (3¢ + 40)° = 4.

5 sin (3t + 40)° = 4 => sin (3t+ 40)° =...

This is equivalent to sinx®° = ..., where x = 3t + 40

From the Sienna 5 Oe

So (using a sketch of the graph of sin x°) six possible solutions are
ae
=> 3t+ 40
=>

Solve these equations for values of t between 0 and 360.

(a) sin 2¢° = 0:7 (by 2eos3e> = 1 (c) 3cos (0.5t + 20)° = 2
Laws of indices

1 Express the following as single powers of 2.

(ee a2 eb 2 )
Che Br): (g) (2°)
Simplify the following.
(a) y xy — (b) BY +B’
(e) yt+y!
Evaluate:

(an?
(e) 5° x54

Simplify:

Evaluate:

(a) 8 (c) 273

(g) 25°
kKWeNs
(0) 10000003
Differentiation practice
Work through this tasksheet if you feel that you need more practice at
differentiation techniques and choosing which technique to use. You will need to
use the chain rule, product rule and quotient rule, as well as parametric and
implicit differentiation.
dy
Find — when:
dx
TASKS
6.3
S1
y = sin’ x x(2x — 3)*

y— 3'c0s 4x x'(2x — 3)

sin.20; = =)2 cosnG

v (Sx)
1
x=-;i?
3
y =t

x’ +3xy+2y =8
sin x
cos 2x

x — 40s0;
Using sigma
When manipulating series, sigma notation can be a very powerful and useful
notation. To exploit it to the full you need to become confident in its use.

1 For the series wu, +4, +u3+---+4u,

Write down an expression for the sum of the series in sigma notation.
Write down a simple series and investigate the effect on the sum of the
series when each term of the series is multiplied by the same constant.
n n

Show that y au;=a ) u; | for any constant a.

i=1 i

Investigate how the sum of a series changes when you add a constant to
each term of the series.
n

Show that cs (u; +b) = Sou; + nb

t=1 i=1
n

Show that Sy (au, +b) =a De u; |}+nb

i=1

Example
n

Evaluate 3 (41 + 2).

i=1

Solution

YS) (4i+2) =4( $0: )+20

i=] i=1

. n(n + 1)
Le Saar using the formula for the sum of an
arithmetic series

pil
4n(n+1
a
6oFind: (@)>. (2i-3)
i=1

4 Generalise the result of question 2 by showing that

n n n

So (uj +0) = Sou; a ory

c— tal ca

5 (a) Write down the result of subtracting 1° + 2? +---+7° from

P43? +---+(n41).
(b) Hence show that
n

Sei yee n+1)>-1

i=1 =A

(c) Simplify (i+ 1)? — # and show that

1
on
— yay
SDie aeegae t=1 i— i=1

(d) Hence obtain the formula

SPF +Hen+1)
il

6 Use the result you have obtained in question 5 to find the sum of the first 99
squares.

7 Use the method of question 4 and the result from question 5 to find:

(a) SS (2? - 61 +4) (Bie ee So tll Steal)?

fat
You should now be familiar with the following results.
Zeno’s paradox
In a paradox, two different, seemingly sound arguments lead to contradictory
conclusions. The Greek, Zeno of Elia (c.450Bc), expounded a famous set of
paradoxes on the subject of motion. The following is an illustration of one of
them.

DISTANCES IN METRES

The dog chasing the rabbit is running at 8ms_! and the rabbit at 4ms_'. Both
are running in a straight line. When the dog first spotted the rabbit the distance
between them was 128 m.

Now consider Zeno’s argument.

1 (a) When the dog has run 128 m the rabbit will have moved away. How far
will it have moved?

(b) When the dog has run the next 64m, the rabbit will have moved away
again. How far will it have moved?
(c) When the dog has run the next 32 m, the rabbit will have moved away
again. How far will it have moved?

Each time the dog arrives at a position previously occupied by the rabbit it will
have moved away. Thus the rabbit will always be in front of the dog and it will
never be caught.
2 The distance in metres travelled by the dog in the first five intervals is
128 + 644324 16+8.
(a) Work out the sum for the first 1 intervals.

(b) What is the value of the sum for an infinite number of intervals?

3 The distance in metres travelled by the rabbit in the first five intervals is
64+ 32+ 164+8+4
Di
oe
fo
ws
Oc)
Wat
iPS
oP
(a) Work out the sum for the first 1 intervals.

(b) What is the value of the sum for an infinite number of intervals?

It appears to take an infinite number of steps for the dog to catch the rabbit.

4 (a) How far will the dog run in 40 seconds?

(b) How far will the rabbit run in 40 seconds?

(c) After 40 seconds, which animal will be in the lead?

5 From your answer to question 4 it is clear that the dog will catch the rabbit.
Can you explain the fallacy in the arguments which led to the conclusions
stated after questions 1 and 3?

Other entertaining paradoxes are outlined in the book Riddles in Mathematics

by Eugene P. Northrop (Penguin).
Regular pentagons and the Fibonacci sequence
1 Why is the interior angle of a regular pentagon 108°?

If you have ever tried to draw a regular pentagon of given edge length, you will
have found that it is difficult to get an accurate diagram if you use a method that
involves measuring the interior angles of 108° with a protractor.

A better method would involve measuring lengths only, using compasses to

construct the diagram. To do this you need to know the length of the diagonals
of a regular pentagon. The procedure for drawing the pentagon is then as shown
below, where the edge length is 1 and the diagonal length is denoted by the
Greek letter @ (‘phi’).
 TASKS
1.4
E1

@ can be calculated using the fact that triangle CDE is an enlargement of triangle
ABD with ¢ as the scale factor. Since it is a diagonal, AC = ¢. Also CD = 1, so
AD = ¢ — 1. By comparing the sides AD and CD in triangles ABD and CDE, it
then follows that

¢(@-1)=1 @

2 Explain the values for the angles in triangle ABC.

3 Explain why CD = 1 and ¢(¢ — 1) = 1.

Rearranging () gives:

M14 V3) or 6 =4(1- v5)

~ 1.618 or @—0.618 (to 3 decimal places)

As far as the diagonal of the regular pentagon is concerned, the negative root is
meaningless, so the diagonal length is 1.618, or, perhaps more usefully, the
diagonal length is the edge length multiplied by 1.618. You now have a simpler
means of calculating the diagonal for a given edge and then drawing an accurate
regular pentagon.

4 Using a suitable diagonal length, draw accurately a regular pentagon with

5m edges.
 ns can be enlarged
The story does not end at this point, because regular pentago
as
by extending pairs of sides and joining the five points of intersection,
illustrated below.

E1
TASKSHEET
1.4

Two of the outer triangles making up the shape above are shown below.

, U
L >

R
fo fra S

These triangles are similar to those used on page 545 and so

SU=¢?xRS=¢@ UT=¢xSU=¢

5 Explain why SU = ¢ and UT

= @’.

The edge length of the larger pentagon is ¢”. Moreover, the diagonal of the
larger pentagon must be ¢°. So you can see the striking fact that the sequence of
lengths

edge, diagonal, edge, diagonal,...

for a succession of regular pentagons, formed by extending the sides, takes the
form of a simple geometric sequence

TR nme vee
6 Why is a diagonal of the larger pentagon given by ¢°?
Previously, the value of ¢ has been calculated as }(1 + 4/5). It is interesting to
examine the sequence of powers of ¢ expressed in surd form.

e=k(tys) df=1(7+3¥5)
¢ =18+V5) ¢w= 1 + 5/5)
@ =1(442/5) ¢° =1(18 + 8/5)

¢* may be calculated from ¢ = $(1+ \/5) by multiplying out brackets and

TASKS
1.4
E1
simplifying as follows.

¢ =4(1+ V5) x4(14+ v5)

& =1(1+ /5+/54+5)

go =7(6 + 25)
$ =3(3+ v5)
However, it is easier to observe that ¢* = ¢ +1.

So
Bes) et
2(3 +4/5)
Similarly, ¢ = ¢° + ¢, ¢ = ¢° +¢’, and so on.

7 Check the values for ¢° up to ¢° in the same way, and calculate ¢’ and ¢°.

Two important sequences are known as the Lucas sequence

to os 4, 7, (iil 28s

and the Fibonacci sequence

Ripe ay Ono) 8,
In both sequences, successive terms are found by adding the two previous
terms.
 We next consider how you can find formulas for the general terms of these two
sequences.

When ¢ was calculated there was a second root, }(1 — /5), which will now be
referred to as w (‘psi’). This generates a sequence similar to that for ¢. The two
sequences are placed alongside each other for comparison.

= ~
TASKSHEET
F1
1.4
2
y
3
~
yy I

~
ha
RIF
NI
IR
VI

If you add corresponding members of the two sequences you obtain the Lucas
’ sequence. The Fibonacci sequence is obtained by subtracting corresponding
members and then dividing by ,/S.

o+yp =1 o-py=V5
¢+y =3 ey =/5
Ptyp=4 2s
+p =7 o340
gt+we=il g—W =5/5
e+ =18 go — of =8,/5
You now have formulas for the general or nth terms, L,, and F,,, of the Lucas
and Fibonacci sequences:
L,=¢" +4" = [311+ v5)]" + [40 - V5)]”

Fs = 3(8" — 8") = 1G( + v5)" - Gt = v5)"

; 1
Verify that ¢’ + W’ and Vs (oe ~’) give the seventh terms of the Lucas
and Fibonacci sequences and verify the eighth terms in the same way.

Use a calculator or computer to tabulate 6”, w", L, and F,, for a range of
values of n. In particular observe the behaviour of y”.
It is instructive to look at the four sequences $”, 7", L,, and F,, numerically,
using a calculator or a spreadsheet program on a computer. In particular, it will
be noted that the terms of =” rapidly become small because 7 is numerically less
than 1. The first terms in the formulas for L, and F,, give approximations to the
sequences, which improve in accuracy as n gets larger.

The approximations are

L, © (5(1 + 5))”
1 TASKS
1.4
E1
F,, © ys at + /5))"

10 Calculate the ratios of successive pairs of terms in each sequence. In other

words, calculate
IL F
|e lope

What do you notice and how can this be explained in relation to the
approximations of L,, and F,,?

The Greeks derived ¢ from the golden rectangle, which was thought to display
particularly pleasing proportions, and is, for example, the shape used for the
frontage of the Parthenon in Athens. ¢ is known as the golden ratio, the ratio of
the sides of a golden rectangle. A golden rectangle is such that when a square of
the same width is removed the remaining rectangle is also golden, as shown
below.
ey

Ol

Since the length of the golden rectangle is the width multiplied by ¢, it follows
that
o(6-1)
¢—-1
which is precisely the equation derived earlier in relation to the diagonal length
of a regular pentagon.
|
and the golden
-i These two sources of the golden ratio — the regular pentagon
r very nicely in the regular icosahe dron.
se rectangle — are brought togethe
robel with twenty faces in the form of
The regular icosahedron is a polyhedron
ft] are fitted together so that they are
ae equilateral triangles. If three golden rectangles
vertices of a regular
iS) mutually perpendicular, their twelve vertices form the
w icosahedron.
ie)
ae
Fs
a5
we

In this icosahedron, five equilateral triangles meet at a vertex. Such a set of

triangles forms a pyramid with a regular pentagon as its base. One of these
pentagonal pyramids is shown by the dotted lines above. Note that the edges of
the icosahedron are the same length as the shorter edges of the golden rectangles
and that the one diagonal of the regular pentagon that is shown is the longer
edge of one of the golden rectangles. So the icosahedron provides a link between
regular pentagons and golden rectangles.
Tangents and normals
A line with gradient —4 makes a right angle
with a line with gradient 2.

Each line is said to be normal to the other.

TASKS
2.2
E1
1 Ifa line is drawn with a gradient g, what can you say about the gradient of a
line which is normal to it?

Tangent

A line is said to be a normal to a curve at a given

point if it is normal to the tangent at that point.

The graph of y = 1x? has tangents drawn at x = 1 and at x = —1. These two
tangents and the normals to the tangents form a rectangle.
(a) Sketch the graph with the tangents and normals shown.
(b) Find the equations of the two tangents.
(c) Find the equations of the two normals.

(d) Use your.answers to (b) and (c) to find the coordinates of all four corners
of the rectangle.
(e) What is the area of the rectangle?

Tangents and normals drawn on the graph

of y = x” form a square as shown.
(a) Find the coordinates of the four corners
of the square.
(b) Show that the area of the square is ;
The normal to the graph of y
cuts the graph at A as shown.

Find A.
Optimisation problems
1 A mathematical ornament consists of a cone inside a sphere of radius 5 cm,
such that the top and the perimeter of the base of the cone touch the sphere.
Design the ornament so that the cone has maximum volume.

To use calculus to solve this design problem, you must first express the
volume, Vcm’, of the cone in terms of another quantity which can then be
varied to maximise V. The depth of the base of the cone below the centre of
the sphere, x cm, seems a suitable quantity for this purpose.

(a) Why can x be neither greater than 5 nor less than —5?

(b) Without doing any calculations, write down what you think happens to
the volume of the cone as x gradually changes from —S to 5.

x=-4 ——> x=-2 —— > 1=0

oO OO
(c) Draw a rough sketch of the (x, V) graph which more or less fits your
answer to part (b). Mark the value of x which you think will result -in the
greatest volume.

(d) The volume of a cone is 1ieh where / is the height of the cone and r is
the radius of its base. Calculate the volume of the cone for the value of x
which you think gives the greatest volume.
 where ycm is the
(e) Suppose you choose as your variable not x, but y,
point of the sphere as
height of the base of the cone above the lowest
shown in the diagram.

Again without doing any calculations, sketch

what you think the graph (y, V) will look like.

TASKSHEET
E1
+3 ics
(f) Whether you choose x or y is unimportant, except that the mathemat
involved may be easier for one rather than the other.
Express V in terms of x, sketch the graph of V against x and use calculus
to find the maximum value of V.

(g) Now express V in terms of y and again use calculus to find the maximum
value of V.
Check that this results in exactly the same shape for the ornament.
(h) You could also have expressed V in terms of 0, where
6 is the angle shown in the diagram.

Without doing any calculations, sketch what you think the (6, V) graph
looks like.
Express V in terms of 6, use a graph plotter to plot the graph and
determine the value of @ which maximises the volume. Check that this
results in the same shape as before.

A bicycle manufacturer has designed a new model and has the problem of
fixing the price in such a way that profits are maximised. After an initial cost
of £50000 to set up the production line, it will cost £85 in labour, raw
materials and components to produce each bike. Market research suggests
that the firm can hope to sell 5000 bikes if the price is fixed at £100, but they
can only expect to sell 1000 if the price is £200 per bike. They assume the
relationship between price and demand is linear between these two extremes.
How many bikes would you advise the company to manufacture and at what
price should they be sold?
Traffic
Car B is being driven at a speed of 45 m.p.h. (20ms_'). Car A can accelerate
from 0 to 20ms_‘ in 10 seconds in such a way that its speed after t seconds is
given by the formula:
t(20 — t)
spee d= 5

The (time, speed) graphs for the two cars are sketched below.
Speed (m s_!) Speed (m s~!)
20

Time (s) 10 Time (s)

State clearly the assumptions you make when answering each of the following
questions.

1 Evaluate the integral

10+(20 — t)
dt
0 5
and explain what information it gives you about the motion of car A.

2 What is the least distance B can be from A at the start of the manoeuvre if B
is to avoid slowing down?

3 If 1200 cars per hour use the road, is A likely to find a gap of sufficient length?
 5 ; aa)
Derivative of sin’ x

Sketch the graph of y = sin’ x.

: t a? j
By sketching the gradient graph of y = sin” x, suggest an appropriate
expression for the derivative of sin” x.

sin’ x may be written as acos bx + c. What are the values of a, b, c?

TASKSHEETF1
3.5
; : : : Ree
Use your results in (a) to write down the derivative of sin” x. Compare
your answer with that found in question 1.

Find the derivative of cos” x, checking your answer.

(a) Find the derivative of sin’ x + cos” x.

(b) Sketch the graph of y = sin? x + cos” x.

(c) Explain the result in (a), using your graph.

Another way to find the derivative of sin* x uses an important identity, which
can be developed as follows. Consider the triangle OAB shown.
AD is perpendicular to OB.
Lei @a—all
Then. OD = ‘cos 2x (in triangle OAD)
Angle DAB = 90° — angle ABD = x
AG — ssi wi AB = 2Sitex

DB = (2sinx) sinx = 2(sin x)* = 2 sin? x

Since OB = 1, then OD = 1— DB = 1 — 2sin?x

But OD cos 2x

So cos2x = 1—2sin?x

Although this result is proven here for acute angles, it is in fact true for all angles
x. (This will be proven in a later chapter.)
d
5 Use this result to show that ae (sin* x) = sin 2x.
x
6 Using also the result sin* x + cos? x = 1, show that = (cos* x) = —sin 2x.
Prime number formulas
Many quadratic formulas generate long strings of prime numbers. The formula
n’ —n +41 is a much quoted example.

First conjecture * — n+ 41 is always prime.

This is not true; for example when n = 41, n* — n+ 41 = 41’, which is not
prime.
TASKS
4.2
E1»
After trying many other expressions you might make a totally different type of
conjecture.

Second conjecture No quadratic expression in n is prime for all integral values

of n.

1 Write down various quadratic expressions in 1, for example

2
jE a Wiper
5
ibe le Sia
2

Can you always find a value of n for which the expression is not prime?

It is likely that the more quadratic expressions you try, the more you will
become convinced of the truth of the second conjecture. But, as you know, it is
not sufficient simply to try lots of examples — you may miss the one example
which turns out to be a counter-example!

This is the exciting phase of the solution of a mathematical problem where all
sorts of ideas must be tried out as you search for either a convincing proof or a
counter-example. Two attempts at proof are given below.

First attempted proof

The general quadratic is of the form an” + bn +c.

Putting n = c gives ac’ + bc +c = c(ac +b +1), which is not prime because it is

divisible by both ac + b + 1 and c. Therefore no quadratic expression in ” can be
prime for all integral values of n.

2 Check over the ‘proof’ above carefully. Which particular cases spoil the
‘proof’?
 Second attempted proof (‘by contradiction’)
Suppose an” + bn +c to be prime for all integers x. In particular, for n= 1,
a+b+c must bea prime. Letta+b+c=p.

Forn
= 1+),
an +bn+c=a(p+1)+b(pt+1)
+e
ap’ +2ap+bp+at+bt+c
TASKSHEET
E1
4.2 ap’ +2ap+bp+p-
an’ + bn +c is therefore (ap + 2a+b+1)p.

3 Find a similar expression for an” + bn +c when n = 1+ 2p.

an’ + bn +c is divisible by pwhenn = 1, n=1+pandn=1+4 2p. If it is

prime for each of these values then an’ + bn +c must equal p itself and we
would have three points on a quadratic graph as shown. :

an?
+ bn+c

ERMA
RARE IE ae ee
1

This is not possible for a quadratic.

2: : ;
So an“ + bn +c is not prime for at least one of these three values.

There may be some points on which you are still not convinced. If so, first try to
fill in the necessary details yourself and then discuss the unclear points with
fellow students and your teacher. Do not be convinced too easily!
 TICAL METHODS

Extending the method

The method of section 5.1.4, although useful, has only been developed for the
expression asin@ + bcos @ on the assumption that a and 8 are positive. In this
tasksheet, other possibilities are considered.

(a) Use the diagram to find an expression

for x
(i) in terms of r, 6 and a, TASKS
5.1
E1
(ii) in terms of a, b and @.

(b) Use your result from (a) to obtain an

alternative expression for
acos@ — bsin@.

By expressing y in two different ways,

explain why
asin
@ — bcos@ = rsin
(6 — a)

3 Using the diagram of question 2, explain why

acos6+ bsin0 = rcos (9 — a)

The expression you have just found is an alternative to the expression used in
section 5.1.4 where you wrote asin@ + bcos@ = rsin(0+ a).

4 (a) Express 7sin@ + 4cos@ in the form r; sin(@+ a;).

(b) Express 4cos 6 + 7 sin @ in the form r, cos (8 — a).
(c) By plotting the two graphs show that these give the same result.
(d) What is the relationship between a, and ay?
5 Express each of the following as a phase-shifted sine or cosine wave.
(a) 7cos@+ 24sin8 (b) 12sin@
+ 5cos@

(c) 9sin@
— 40cos@ (d) 4sin@
+ 2cos6

It is not necessary for a@ to be acute, although practically it is much easier to

work with if it is. The next question demonstrates alternative forms.

6 This graph may be regarded either as a sine graph or as a cosine graph, phase-
shifted either to the right or to the left. .

Express the graph in the form:

(a) rsin(@+a) (b) rcos(@— a)

(c) rsin(@—a) (d) rcos(0+ a)
Intersections

Three points have coordinates:

A (2, 2, 2); Bi-1,. 16); C (0, 2, 5)

(a) Find the vector equation of the plane ABC in the form:
(i) r= OC+ACA
+ wCB
o =—_ — —
(ii) r= OA +AAB + pAC TASKS
5.2
E1
(b) Show that both vector equations give the same Cartesian equation:
SY eS

Find two points which both lie on the line of intersection of the planes
3x — y+ 2z = 8 and x — 2y+z= 1. Hence find a vector in the direction of
the line of intersection and write down the vector equation of this line.

By substituting for x, y and z in the equation of the plane, find the value of
the parameter ¢ at the point of intersection of the line with vector equation

x ee 3
Sr ecbacta elle el
Z 5| ==
and the plane with Cartesian equation 2x + 3y +z =7.
Hence, write down the coordinates of the point of intersection.

Find the point of intersection of the three planes

x—-ly+z=9, xty+2z=8 and x—3y+3z=2.

What happens if you try to find the point of intersection of the three planes
x—2ytz=1, 3x—y+2z=8 and 4x—-3y+3z=5?
In question 2, you found the direction vector of the line of intersection of the
first two of these planes. In the same way, find the direction vectors of the
lines of intersection of the other two pairs of planes.
How does this explain what has happened?
How are the planes related geometrically?
How is your result modified if the third equation is 4x — 3y + 3z = 9?
 Drawing parametric curves
1 (a) For the curve given by the parametric equations x = cos’ 6 and y = sin’ 0,
complete the table and sketch the branch of the curve which is formed.

0 0 1 T 1 1
x

y
TASKSHEET
6.2
F1
ff
How are the values of x and y for }m < 6 < x related to the values in the
table? Use these relationships to sketch the part of the curve 7 < 0 < 7.
Use symmetry properties to complete the sketch of the curve.
Check your curve using a graph plotter.

Sketch the curve given by

x=0-sin0, y=1-—cosé

It is often possible to sketch a curve without plotting a large number of

points. Consider, for example, the curve given by
ee 2+
i“ = a 5
By = ——_
2-t 4-t

(a) Write down where the curve cuts the axes.

As t — 2 from above, 2 — t is a small negative quantity, so x — —oo.

As t > 2, y > 2, so y = 2 is an asymptote.

(b) Explain what happens as t > 2 from below (i.e. through values smaller
than 2).

(c) Write down any further asymptotes.

(d) Write down any obvious points.

(e) Sketch the curve. (Do not plot!)

Use the method developed so far to sketch the curve

x =>—
2t =
-
ae eee
In particular you should find any asymptotes and examine what
happens as
[ets BECO
By parts
1 To find Jecos x dx you could start as follows.

dv
Let uw = e and ae cosx
x

So |e* cos. dx = esinx — [e*sinz dx

Veg Jecos
x dx, then TASKS
7.1
E1

l=€ sinx— |et sind

Using integration by parts again, you should find that I appears on the right-
hand side. Show how this enables you to find I.
fe Gre es
Find | e* cos x dx by first putting u = cos x and aaa
x
d
Discuss whether the choice of u and a makes any difference to integrating
this product. 22

Integrate each of these functions:

(a) e* sinx (b) e* cosx (e)-e sin 2x (d) e°* cos 2x

(a) Use integration by parts to find |sin cos. dx.

(b) You could integrate sin xcosx by using the chain rule in reverse or by first
using a trigonometric formula to rewrite sinx cos x in a simpler form.
Work out the integral by these methods and check that the answers are
consistent with the one you obtained in part (a).

To find Jinseds,write Inx = Inx x 1 and integrate by parts, letting u = Inx

dv
an di
a — ih
3
Evaluate |Inx dx and draw a sketch to show the area you have worked out.
2
Integrating the circle
1
1 Explain why | (1 —x*)dx =47
0

2
It is not so easy to evaluate | \/(1 — x*) dx. This integral can, however, be
0
integrated using the method of substitution.

2 (a) The obvious substitution to try would be wu= 1— x”. What happens
when you try this substitution?
(b) Try the trigonometric substitution, sin @ = x, and show that

J/(1 — x*) dx cos’ 6d0

Ix /(1—x°) +4sin"'x +c
(Hint: Use the identity 2 cos” @ = 1 + cos 26.)
1 2
(c) Confirm that | /(1—x*) dx =17 and evaluate | /(1 — x”) dx.
0 0
There is a more elegant way of evaluating a definite integral than using the
method of substitution, and that is to substitute the limits at the same time as
you substitute the function.

3 If sin@
= x, explain why x = 0 if 8 = 0 and why
x=
1

Thus iJ/(1 — x’) dx = ” cos? 6d8

0 0
 Lrus

4 Evaluate | cos’ 6.d6 and confirm your answer to 2(c).

0

5 (a) Give a geometrical reason why the triangle

OAB has area = }\/3 and why the sector
OBC has area = a5
(b) Do these results agree with the answer you
found by integration?
TASKS
7.1
E2

O a
2

3
Use the method of substitution to evaluate | J/(9 — x*) dx.
1.5

(a) Use the trigonometric substitution, tanu = x, to evaluate F

(Hint: Use the identity 1+ tan? u = sec’ 2 u.)

(b) Find |————;

: dx.
1+ 4x? 5

Use the trigonometric substitution sinu = x to evaluate

Pcehe
Repeat question 8 but use the substitution cosv = x.

Use a graph plotter to graph, on the same axes, the graphs of:
(a) y=sin' x (b) y= —cos 'x

How do these graphs explain the apparent discrepancy in your answers to

questions 8 and 9?
 Carbon dating

Radioactive chemical elements

The isotopes (see below) of many of the chemical elements are radioactive.
This means that the structures of their atoms are unstable and the atoms readily
decay to form other stable elements, releasing radiation in the process. ‘There are
three main types of radiation: alpha particles, which are essentially helium
7.3 nuclei; beta particles, which are very fast moving electrons; and gamma rays,
TASKSHEET
E1
which are electromagnetic radiation of very high intensity

For example, uranium-238 is radioactive and decays to eventually become

lead-207. (238 and 207 are the atomic weights of uranium and lead
respectively.)

Half-life
The half-life of a radioactive isotope is the time it takes for half the radioactive
atoms in a sample to decay. In other words, if a sample of radioactive substance
contains N atoms at some particular time, then the time it takes for this to decay
N
so that oiatoms remain is the half-life.

Isotopes
Isotopes are different atoms of the same element, in that the nuclei of the atoms
are different. All nuclei of a certain element contain the same number of protons,
thus defining the atomic number and the characteristic chemical properties, but
isotopes contain different numbers of neutrons in the nuclei.

Half-lives vary from a fraction of a second to millions of years. Some examples

ane.

uranium-238 4.5 x 10° years iodine-128 25 minutes

carbon-14 5730 years lawrencium-257 8 seconds
radium-226 1600 years polonium-214 1.64 x 10°* seconds

Representation as a differential equation

The principles of radioactivity were discovered and developed at the beginning
of the twentieth century by the New Zealand physicist Ernest Rutherford
(1871-1937) while he was at McGill University in Canada. It was for his work
in this field that he was awarded the Nobel Prize for chemistry in 1908.
Rutherford discovered that if at a particular time, t, a sample of radioactive
substance contains N atoms of the radioactive element per unit mass of the
substance, then the number of atoms decreases with time according to the
differential equation

dN
ac —kN

Radio carbon dating

One of the applications of this differential equation and its solution is found in TASKS
7.3
E1
archaeology. In a scientific paper published in 1949, the American chemist
Professor Willard F. Libby first proposed a figure for the half-life of the
radioactive isotope of carbon, carbon-14. He showed, further, how it could be
used to date wooden artefacts and other remains containing carbon found on
archaeological sites. This was a breakthrough of great significance for
archaeology and, in 1961, Libby received the Nobel Prize for his work.

Libby first calculated the half-life of carbon-14 as 5568 years, but the accepted
value is now about 5730 years. This is equivalent to about 1% of the carbon-14
atoms decaying every 83 years. Carbon-14 emits beta particles, becoming
nitrogen in the process.

Carbon has three isotopes. Carbon-12 accounts for roughly 99% of carbon in the
world and carbon-13 the other 1%. So the occurrence of carbon-14 is tiny, the
ratio of carbon-14 to carbon-12 atoms being about 1:10. It is the fact that it is
possible to measure the radioactivity of carbon-14 that makes it possible to
date ancient remains, whether they are of wood, flesh or bone. Carbon-14 in
living organisms is radioactive, and its radioactivity level is measured at 6.68
pico-curies per gram (1 pico-curie is equivalent to 3.7 x 10" disintegrations per
second), but the loss is made up by natural processes and there is only a net
decay after the carbon-containing organism dies.

Carbon-dating is usually accepted as a dating method valid up to about 40000

years, when carbon-14 levels become too low to measure. Some error is
inevitable, not least because radiation itself is a random process. Great care must
be taken to avoid contamination of artefacts with fresh carbon when preparing
to measure their radioactivity, otherwise gross errors could occur. However, the
carbon dating method has been verified back to about 5000 8c using
dendrochronology. Dendrochronology is the counting of tree rings, one new ring
being formed in the trunk for each year of growth. This check is possible since
some trees are incredibly old and yet still growing. For instance, the bristle-cone
pines of California are over 4000 years old and still growing.
 Solution of the differential equation

—-
dN
= iN = N=Noe —kt
dt :
where No is the number of atoms of
carbon-14 per unit mass in the
substance at time t = 0. first’ half-life

7.3 1 Show that

TASKSHEET
E1 if the half-life of ‘second’ half-life
carbon-14 is 5730 years then ‘third’ half-life
1
BS
8300
0 4 8 12 16 20 24 28 32 36 40
Age (thousands of years)

If you know how many carbon-14 atoms there are per gram for a dead organism
(for example, a piece of wood) then knowing how many there should have been
when the organism was alive (i.e. a tree) would enable you to estimate the time
that has elapsed since its death. It would be difficult to establish the number of
carbon-14 atoms in one gram of substance. What can be measured with
accuracy is the radioactivity level of any carbon-14 atoms present in a gram of
substance.

Suppose R(t) represents the measured radioactivity of carbon-14 at time f.

Because radioactivity is proportional to the total number of carbon-14 atoms
present it follows that

R(0)_ Noe’
R(t) Nye-#/8300

2 Explain how this equation can be rearranged to give

t = 83001n (2)

How old is Pete Marsh?

Pete Marsh is a name popularly given to the remains of a man whose body was
found in a bog in Lindow Moss in Cheshire. The man had apparently been
murdered by garrotting before being thrown in the bog; but when did it happen?
Readings of the carbon-14 radioactivity levels from the body were about 5.3
Ppico-curies per gram.
It is reasonable to assume that his radioactivity would have been 6.68 when he
was alive and so the time that has elapsed since the death of Pete Marsh can be
estimated by
6.68
t = 83001n (S$) = 1920 years

This suggests that the murder took place nearly 2000 years before the body was
discovered. Latest estimates from the Radiocarbon Unit at Oxford University
suggest that Pete Marsh died between 2Bc and Ap 119, at the time of the Roman
occupation of Britain. TASKSH
7.3
E1

3 Historical records indicate that the Egyptian king, Sneferu, died some time
between 2700 Bc and 2550 Bc. Radioactivity levels from carbon-containing
artefacts in his tomb gave a reading of about 3.8 pico-curies per gram.
Does this reading agree with the historical records?

For a long time, historians believed that the origins of agriculture were in the
Near East around 4500 sc. Archaeological investigations at the ancient city of
Jericho (in modern day Israel) found farming implements that gave a
radiocarbon reading of 2.8 pico-curies per gram.

Why did this lead to a storm in historical circles?

570

S OLUTIONS (c) The graph on page 1 shows the number c

radios sold at a given profit per radio. If:
1 FOUNDATIONS stands for the number of radios sold, ther
(a) gives

1 .1 Graphs y = 60 — 10x

i 1.1 Introduction This is the equation of the first graph.

The graph that you have drawn for 1(b)

shows the total profit for a given profit
per radio. If ¢ stands for the total profit,
(a) then (b) gives

Profit (£) per ORD Paesa eee 6 t = x(60 — 10x)

radio
Number of 60 50 40 30 20 10 O This is the equation of the second graph.
radios sold
Total profit (£) 0 50 80 90 80 50 O
from sales

—_= For each graph, you might consider:

e general impression of shape;

7044
e whether or not it passes through the
604
Total profit (£) <9] origin;
405 @ steepness of the graph in different places;
305 @ symmetries — reflection, rotation,
translation;
@ values of x for which the function is
undefined;
Profit (£) per radio @ what happens when x is close to the
undefined values;
@ what happens when x is very large (positive
(c) The maximum total profit is £90, which is or negative);
obtained with a profit per radio of £3. e whether or not there are any restrictions ot
the values of y;
(a) The expression 60 — 10x or 10(6 — x) can @ similarities to graphs of other functions;
be established fairly easily. e how the functions might be classified.

The list is not exhaustive, nor is it necessary

(b) The table was completed by multiplying
for you to record all features at this stage.
‘profit per radio’ x ‘number of radios sold’.
The graphs are related by reflection in the
The answer given in part (a) leads to x-axis.

x x (60 — 10x) = x(60 — 10x)

You should notice that they are related by
= 60x — 10x* reflection in the line y = x. This is true for
both pairs of graphs.
or any equivalent expression.
 FOUNDATIONS 571

1.1.2 Linear graphs 3 (a) y=2x+3 (b)-y

= 2x +2

(Cy 4 (d) 3y=x+5

p.1 C 4 (a) no (b) yes, gradient — 7 intercept 2

No. of days Distance Distance (c)) no (d) no

walked from Land’s (e) yes, gradient — 5, intercept 3
(miles) | End (miles)
(f) yes, parallel to the y-axis
(a) 1 30 770
(b) 2 60 740 (g) yes, gradient 4, intercept 7
(c) t 30t 800 — 30t (h) yes, gradient — 3, intercept 3
(i) no
(d) In the first graph, s represents the
distance, in miles, walked from John An equation will only have a straight-line
O’Groats, which is 30t. Therefore, graph if it can be written in the form
s = 30t is the equation of the graph. Vi= Ti Ge

In the second graph, s represents the 5 (a) y-4x +7 =0 (b) y—3x+1=0

distance, in miles, from Lz nd’s End,
which is 800 — 30¢. Therefore,
(C\3y--
2% — 13'= 0
s = 800 — 30t is the equation of the
graph.,
1.1.3 Quadratic functions
1.1 Exercise 1
1D
(b) You have seen that the graph of the quadratic
function y = ax” + bx +c may be obtained from
the graph of y = x” through a series of scalings
and translations. The questions on pages 7—8
explore the translations involved in obtaining
the graph of y = (x +p)? + q from that of
Ve we
It should be apparent that the graph of
y =(x+p)*+q is a parabola with vertex at
(—p,q). It may be drawn by translating the
graph of y = x” by ny :

1 The graph of y = x* +3 is that of y= x*

translated upwards by 3 units.

2 Similarly, the graph of y = x” +q is that of

y = x’ translated upwards by q units. This is
not surprising since to get from y = x” to
y = x* +q you simply add q to the
2 (a) y=
2x 22. (b) y= —2x+4
y-coordinates.
(c) s=it+3 or 2s=t+6
3 The graph of y = (x + 4) is that of y = x?
(d) y=3x-—2
5 or 2y = 3x—4 translated 4 units to the left.
572 SOLUTIONS

| 4 Similarly, the graph of y = (x + p)’ is that of | 1 (b) It can be seen from the graph that
| y =x’ translated p units to the left. Notice y = x* + 2x has its vertex at (—1, —1), sc
that the vertex occurs when x = —p, not at =1
aD. a translation of i maps y = x” onto
y=x? +2x.
| 5 (a) To obtain the graph of y = (x +p)’ +4, Hence y = x* + 2x is equivalent to
the graph of y = x? is translated through y= (ei) —1.
p units to the left and g units upwards.
(b) The vertex of the resulting parabola is at | 2 (ii) (a) las x? + 10x = (x + 5)? — 25
235)
(—P, 4).
(c) Its line of symmetry has equation x = —p. (b) fae x = 6x =(«—3)—9
| 6 (a) y=x' +3 (b) y= (x — 2)?
—3.5 7 2 ye

(c) y=x-4 (d) y= (x +5) (c) eed x ie 7x == (2 43)5)

3.5!) ale

(e) y=(x—2)°4+3 (f£) y=—x?

+4
| 3 (a) x + 4x = (x42)? —4
C= a3), y= ae2) +4 (b) x? + 4x = (x +2)? —4
(i) y= (x +2)+5
et aie 159 Ie ede
| 7 The curves meet at (—2,0) and (2,0). Sx ag = (yD) 5

| 8 The curves meet at (0,11) and (6, 11). 1 4 (a) x7 + 14x +2=(x4+7)2-4942
9E The vertex of y = (x + 2)* +3 is at (—2, 3). =(x+7)?
—47
So y = ax’ passes through (—2, 3). (b) x* — 8x +5 =(x—4)
-—11
Substituting in the equation gives (c) x’ —3x+1=(x—3)?-§
Si a2
5 (a) x +bx = (x +4b)° —1b°
=>3=44a
>p=4b and q = —1b
>a=3 4
(b) x°+bx+c=(x+lby—-1P+e
_ 10E The curves must be parallel, soa =c and it t
bed. Halve 6 Subtract ib? from

| 6 (a) (x-1)
(b) (x — 1)? > 0 for all x.
1.1.4 Completing the square Hence x” — 2x +1 > 0.

17 (a) (x +2 -6
iby higsh
If you wish to sketch the graph of a quadratic
function given in the conventional form
y=x° +bx +c, then it is helpful to be able to
xy
rewrite it in the completed square form
y = (x +p)’ +4, so that the translation from
y = x* is obvious. These questions develop an
(-2, -6)
appropriate strategy.
 FOUNDATIONS 573

(b) (i) (x—2)*-7

(li) x" ~ 4% 447
Se 43

(-6, -41)

|SE 2[(@+3)?-2]<0

+20
| (-3/2, - 22)

Mi rvercieo
id xercise
| 2 (a) (i) y= (x +2)? +3 (ii) y=(x-2P-5
(ae ie (z + =s pte tics eae

(ii) x° + 8x + 16-11 ee Se ae
=x +8x+5

(iii) y

1 (a) The graph shows that 1 and 5 give the

(ees
ee dl)) points x = —1 and x = —5 where the
574 SOLUTIONS

curve cuts the x-axis. The values are also 4 (a) y=(x+2)(x-1)
the zeros of the function because
(x + 1)(x +5) =0
= Olorxs- 5 = 0 ;
= Xt Vertex at (—55= 9

=>x=-lorx=-—S5

(b) When a and @ are distinct real numbers,

the graph will cross the x-axis at two
points, irrespective of whether a and 8 2 2
are positive, negative or zero. If a and ( INS (c) Ye
are equal then the two crossing points vA
coincide and the graph touches the x-axis.
a and £ give the points x = —a and 6
x = —6 where the graph cuts the x-axis.
The values —@ and —{ are also the zeros
of the function with that graph.

bf ST etc ei
Vertex at (2,0)
y= (x+a)(x+8) Vertex at (—$,—1
2g
() y=-(x+1% — (@) y=a?-4
| = (x —2)\(eee

Vertex at (—1, 0)
Vertex at (0, —4)

y= —(x + a)(x
+ 6)
is the reflection in the x-axis of 1.1.6 Factorising quadratics
y = (x+0)(x +B)
QO NGes) 1.1 Exercise 3
(b) y= (x +3)(x
—9) 1 (a) (x+3)(x+4) = (b) (x —3)(x +1)
(c) y= (x + 10)(x + 2) (d) y=x(x
—4) — 2)
(c) (x — 5)(x (d) (+ 2)(x — 2)
(e) y= —(x =1)(x —5) (e) x(x —7) (f) («—3)
= y zion5) ee (g) (x+1)(e+2) — (h) (x +2)?
g) y=—(x- y = (x +2)
() y= (x +5)(e-5) pues
 FOUNDATIONS 575

eZ Sequences
es 1 Introduction — notation

dees
i (en) = 2, u4 = 20

(b) Us = 26, Us =U4t+6

(i) Zeros at —1 and —2

(c) Uj41 =U; +6
(ii) (x + 1)(x + 2)
2 (a) t =44+9=13, 313-9 22s

(b) y % =31; ts = 40

uy= —5

Putting 7 = 1 gives
(i) Zeros at roughly 0.6 and —1.6 Wy =m +2=—-54+2=-3,
(ii) There are no factors of the form putting 7 = 2 gives
ax +b, a and b whole numbers. U3 =U, +2=—3+2=-—1 and so on.
Ww=—-5, wm =-3, uw=-1,
U4 = ile us = 3
Ux = —5 + (19 x 2) = 33

Wt 15; uu = itil, u3 = Hs U4 = 3s
Us =-1

ung = 15 —(19
x 4)= -61
U2, eaten—10,
u3=3X6=18, u,=3X18=54,
(i) There are no zeros. us = 3X 54 = 162
Uy = 3? x 2% 2.3 x 10°
(ii) Hence there are no factors.
Note how multiplying by 3 gives a rapid
3 (a) 0,4 (SpA (0) 3,03 increase in magnitude.

(d) —2,—1,0

4 2x?
—5x43 The sequence oscillates.
Mail's ns
The sequence alternates in sign. The
magnitude of each term is twice that of
5 2s,
6s
the previous term.
With u, = 3 for example, the sequence is
6E 2m 3, —6, 12, —24, 48, —96,...

7E 30a or 20 x 60 The sequence diverges rapidly unless

Wy = 4;
8E 8 and 15 If u; < —4 the sequence approaches
negative infinity.
576 SOLUTIONS
SSS eee

If uw, > —4 the sequence approaches 5 The sequence is

2, OHS, WseVXS, 76.88, 0.0046,
infinity.
1294 319.3,
(c) If u; > 0 the sequence approaches infinity.
If wu, < 0 the sequence approaches negative Odd terms approach zero, whilst the
infinity. remaining terms become very large.

(d) The sequence approaches zero.

(e) The sequence alternates between the
values u, and u, for example
te
hy = I
i)
Un — 7, te
2 any
U4=F,---

(f) The sequence increases. The difference

between successive terms increases by 2
each time.

3E (a) If uw, = u) = 1 the Fibonacci sequence

RS rect Narae Pee aca teas i216
is obtained. Otherwise, a similar (1 (a) Ss = 300 (b) s, = 6¢
sequence where each term is the sum of (c) The inductive method requires every term
the two preceding terms is obtained. in turn to be calculated, in this case a
(b) The sequence eventually converges to a further forty-nine terms!
value of —4 unless u, = 2.5.
2 (a) Each term is 3 times the previous term, so
| 4E The sequence always appears to settle into
theicycle= 4.2 -2 hase 1459-2) he
This has never been proved and is known as i= (2% 3) 63 = 2 3"
Thwaite’s conjecture.
= (2%3 x3) x3=2xS°
To obtain ¢,,, t; must be multiplied by 3
a total of nm — 1 times. Therefore,
1.2. Exercise 1 i, = 203"-*,
1uw=4, m=8, w=16, um =32,
(b) y= (P= —1, y= (Pd =4,
maak w= CN Rad
us = 64. The sequence is diverging. 3

2 (a) Ui— 9; U7, = 6, uz

=
= 4, U4
=
— 33
and
so on.
converging The (—1)’ causes the sign to change for
(b) u,=2, u.=4, U3 = 16; diverging alternate terms.

(c) uy = ils, uz = 5 U3 = il U4 — D3

oscillating
1.2 Exercise 2
oat gu 3 Pasi Wis
3 U4 i uz = 15, uz = 14, u4 = 14,
1 No. of 12.3 4° SHOmORIOOR
UuS P=16 triangles
The sequence is converging, approaching the No. of 3579 11 21 41.20
limit of 2. matchsticks

4 The sequence is In this case the idea of adding on two match

1, 23503, 5,8, (se id each time helps you to find the first few term
which diverges. of the sequence. In order to find the later
 FOUNDATIONS 577

terms you need to spot that the number of

matches is found by doubling the number of
triangles and adding 1.

otto: & 2 SB 2 5 10 2D WD: | The numbers from 1 to 100 can be paired up

In pattern
into 50 pairs, each with a total of 101, as shown.
No. ofdots
(Nth ZS 1) TG 8s 5S OS ey De a3 Ee 100
(b) 4 8 12 16 20 40 80 400 4 ie
eecececcemers
ee
@ “i Be Goa OD, TO SB OD eal
(4d) 1 4 9 16 25 100 400 10000 7 The total is therefore 50 x 101 = 5050.
(ec) 2 8 18 32 50 200 800 20000 27 &
(f) 3 8 15 24 35 120 440 10200 i(i+2)

@) p= ie) Hoe
1s BSS 1 (a) (i) Each pair of terms adds up to 21,
(SIR Vy 3 6 US so the total is 10 x 21 = 210.
4, =3x34+2=11 (1 a Note that, because there is an odd
4W,=3X442=14 Ss
S\N
number of terms, not all terms pair
uy =3X5+4+2=17 up. However, the same result may be
obtained by finding the average of
(b) uw,=5x2'=10, uw =5x 2° =20,
the first and last terms and
u3=5x2>=40, u,=80, us = 160
multiplying by the number of terms.
pe
(pasx t= 3) wm —3 x 2— 12, The average of the first and last
Ux = De us = 48, te
terms is

(1+ 9) aS
eg
(a) (i) y
ii)
— | so the total is 9 x 5 = 45
es te
— eee ae se
Gye «2m 41018, (111)
(1+ 29)
iil) ————
5 xX 29
= 435

(b) (i) 4 =3 x (-1)'t" Gi) 4, = 3 x (-1)' (b) There are many possible ways. One way
is to find the average of the first and last
terms and multiply by the number of
Term terms. This works for series with either an
5 6 9 i even or an odd number of terms.
Aen 10n Mes tS 200°" 25
There are many ways of doing this. For
Breen Hig) 26 299 31 = 1 example, subtracting 4 from each term gives
po) Ot a tO0- apt 1B ren ee ea. a(8) with 101 terms.
DEC ape be ame ie era ce a
(a) 50;1275 (b) 81;4050 (c) 101; 15150
(2 es eee | 1 Se ye
Pes 6 —-9 100 = (-1)' The same principle applies here as in question 1.
Golins= =6 9 2 f000tme (—1)t%s You can find the average of the first and last
Heat =12,. 130" S200 (S1742 terms, and then multiply by the number of terms.

oie 6 = OiaTae 40
otaaiet101° meget
fa
16
(a) —x8=64
2
(b) ae x 33 = 1716
2 cs
reese 36. Sl 10000. (a1 r 267
(c) 5 ~26 = 3471
578 SOLUTIONS
a ee ES ere

5 (a) 142i (b) 24+ 4i (ies) 6E (a) £(5 +10 +--+ +90) = £855
(b) §n(n +1) > 500
:6 (a) (i) 61 (ii) 495
=> n(n +1) > 200
b) i) 4241 (ii) #2743)
ay (Peal
[7 (a) at4d — (b) a+ 49d 7E 8.2+8.3+--+= 22000
| (at (n—1)d 5n(16.4 + (n — 1) x 0.1) = 22000
fi (gy Se 50 258 + 49) n(n + 163) = 440000
é n & 587

1.2 Exercise 3
1 (a) 4060 (b) 7500 2.4 Finance — APR
(c) 10049! — (d) 2356
2 (a) Last term = 8+ 17 X 2 = 42;
8+ 42 £1200 is repaid, which includes £400 interest.
sum = (=) x 18 = 450 This is a rate of 50%.

03 —6 The outstanding debt after 12 months is £0

(b) Number of terms = +1 = 34;
(approximately!) and so the debt is fully
sum = (°1) x34 = 35253 repaid.

SS £100 x 1.016 5"? = £121.699

(c) Common difference = =85 The original £100 must be repaid together
24
Sor IS) with interest of approximately 21.7%.
sum = )x 25 = 2475
2
(a) 12.7% (b) 26.8% (c) 79.6%
3 The volume under the bottom step is
= Wey Sica
V=S0x4xX zm Possibly the simplest algorithm is:
Successive steps up have volumes divide rate by 100,
IV BV eal, add 1,
The total volume is therefore raise to the power 12,
subtract 1,
V+15V
15 ee
De Sse
120V = 1125 m? multiply by 100.

4 (a) 2n-1 The reversed algorithm is:

(b) 7 (This uses 49 bricks.) divide APR by 100,

add 1,
(c) S= $n(2+ (n — 1)2)
find the 12th root,
= In(2n) subtract 1,
= 7 multiply by 100.
The rate is 5.95%.
5E (a) 42 — 2

(b) S = $n(4+ (n—1)4) = 2n? oo Monthly interest rate = 7.93% (7.930 83%);
This is twice as many as in question 4. APR = 149.9%
 FOUNDATIONS 579

| 9 (a) iq X 7292.86 (iri 13-45

— 179
| (b) 2187.86 + 36 x 127.52 + 1786.75 (ec) 1+8+27+4
64 + 125 + 216
(c) 12 months and 52 weeks are assumed to (Ae n7 64a 16
be equivalent with 29.43 = 4 x 127.52
Note that (e) and (f) are different
(d) 0.978 8745% (approximately!) representations of the same sum.
' For normally structured financial facilities, 50 50

_ companies have printed charts for APR. More PCS SEN 3),
=
unusually structured facilities like that for the t=

| car are often calculated using special financial

_ calculators, with functions embodying the
approved methods of calculating APR. The
_ regulations and financial formulas relating to
_ APR calculation are governed by the Consumer
_ Credit Act 1974 and form a very detailed body
03h i=1

| of information with prescribed mathematical 3 (a) 35+334---+(-3)

_ techniques.
Ss
= 20% (20 terms)

= 320
(b) 400
1.2.5 Sigma notation 25 10
4 (a) $545.45). (6) 10100070
1 1
G 14 } 8 1
| 1 £1000 x 1.08"° = £2158.92
| 2 £1000 x 1.08° = £1999.00 20 F

(e) SNaix 3
3 Continuing this pattern, the third investment 1

| will be worth £1000 x 1.08° since it earns

5 (a) 1x2+4x3+9x4+16x5
interest for eight years.
ado Ot OOlx
The final investment earns interest for one
year and will be worth £1000 x 1.08". (by(— 1) d+ Cl) eee xs
+(-1)? x 44 (-1)* x5
The total investment is therefore (ie. 1— 243-445)
£(1000 x 1.08’ + 1000 x 1.087 (c) t4+xtx? tr taxt+x°
eS. 4-1000:% 1:08"")
= £1000(1.08 + 1.08” + 1.08°
a) Septet ge!
(e) £(x1) + £(x2) + £(x3) + +++ + £(%6)
Here aee 085)

1.2 Exercise 4
1.2.6 Geometric series

1 (a) 1+4+44343
1.2 Exercise 5
(b) 9+ 16 + 25 +36 +49
its
htath
8
ey cee 6560 (b) 122070312
580 SOLUTIONS
Oe ee ee

(c) 1743392200 (d) 15.984 375

(ec) 5.328 125
he Exercise 6
2 (a) 13 9 27 Si 12t
2 4 16
(b) 1+84+87> +--- +8" = 153391 689 1 (a) 1-%
—“,=1 (b) (ees—3=>
~ 1.53 x 10°, which is a more practical
} 5
expression. (c) The sum diverges (d) ae 10
5 2 Os re)
ie) Dee ae) ee er = 254

: ie Pires ea ME
ie) (5) +t (G) |= 3 (a) The sum is ;
(b) If you consider the diagram to be made u
L(y
(e) nae = 0.570 (to.3's-f.) from a sequence of nested L shapes,

3 (a) He requested 1+2+4+4+8+4---+2°

iga
~ 1.84 x 10!” grains!
oa
(b)eoex 10" g or 3.7 x 10"! tonnes!

4 (a) The value increases by a factor of 1.01 per

annum. After 2090 years it would be
worth 1 x 1.0170? = 1075 650 555p or it can be seen that each L shape is made
approximately £10.8 million.
up of one shaded and two unshaded
(b) Replacing 1.01 by 1.05 in (a) gives squares, i.e. $ of the diagram is shaded.
ake) esti
4(a)3 (b) 4 ()* (d) 3x)”
200 x 1.05(1.05°° — 1) 3 > 1, therefore (3)” — co as 7 — oo. This
= £43 963
105 — 1 means P,, — 00 as
n — .
(to the nearest pound)
The limiting curve is of infinite length, yet
encloses a finite area!
6 Taking the school leaver’s salary as £8000, the
total earnings over a 45-year period would be 5 The sum to # terms is

£8000(1 + 1.05 + 1.057 + --- + 1.05%)

20-0 _wa=@y)n
21 SIGNS as
i=
(1.05° —+)
= £8000 = £1.28 million The sum to infinity is 10ae
1.05 =1
The difference, 4? (2)”, is less than 0.01 when
1000 x 1.075(1.075” — 1) 17
7 (a)
OS

(b) The conditions lead to the equation

1000 x 1.075(1.075" — 1)
= 2000n
1.075 —1
The conditions are met after 17 years.
 FOUNDATIONS 581

1.3 Functions and graphs , 2 (a) Gi) s(x —2) HS = (x = 2)? £5

= x* — 6x* + 12x —3
1.3.1 Function notation
(b) (ii) g(x -—2)+5=27 7 +5
1.3 Exercise 1 (c) (ii) g(x -—2)+5= J/(x—2)+5

1 (a) 3 (b) 4 () 5 (d) 4 2)

(iii) In each case, a translation of | will
5)
2 (a) 9 (b) epadenaee= superimpose g(x) onto g(x — 2) + 5.
Wwiln
y= |x
3 (a) (i) 6 (ii) 0 (iii) 2 (iv) 0 y= |x+3|-4
(v) n’ +3n+2
(b) (i) 6 (ii) 0 (iii) 2 (iv) 0
(v) n+3n+2 (53, =4)

(c) No, except that a different symbol is used

4 The graph of f(x) is translated onto the graph
for the variable.
of f(x +a) + b by a translation of By , for
wAay (iy (x — 2) . (i) ot 4+4e45 any function f.

(b)aec==0 (Cea —3 _ SE The effect of a is to translate the graph

through —a parallel to the x-axis.
Z
ye 12
pe gee The effect of'b is to translate the graph
x+1 x+1
through b parallel to the y-axis.
(b) (i) 4x7 +4x+1 (ii) x? The graph of sin (x + a) + b will be identical
to the graph of sinx if b = 0 and a is any
multiple of 27.

1.3.2 Using function notation | 6E The effect of a is a stretch, parallel to the

y-axis.

The effect of b is a stretch, parallel to the

1.3 A x-axis.
4 <(b) fo — 2) +5 =(x—2)7
+5 The effect of c is a translation, parallel to the
x-axis.
=x)
4x +445
The effect of d is a translation, parallel to the
=x’
— 4x49 y-axis.

1.3.3 Defining functions

1.3 Exercise 2

1 (a) (i ; (ii) 1 (iii) 1

Gi) :
a+2
The graph of f(x — 2) + 5 is obtained from
d
(v) £(—2) is undefined.
the graph of f(x) by a translation of B :
(b) All values of x, except x = —2
582 SOLUTIONS

Z 1
| 3 When x = 2.5, 2x —5 =0 and is not
: 2x —S§
defined.
At this point there is a discontinuity in the
graph.

| 4 The graph plotter shows the discontinuity at

N= 2.5.
y

725)

xv

3 (a) All values of x, except x = —5

(b) x > 3; note that x 4 3 because division by

zero is undefined. 1 2 3 4 5) x

(c) All values of x ; Important features of the graph can be missed if

you simply plot points. It is therefore necessary
(d) All values of x, except x = —2 | to be able to recognise these features of a graph
_ from its equation.
aay S97 Ga) 7) Sa av) ae 20

(b) x ER

Chiar (i), te
_ 1 (b) All the graphs pass through (0,0) and
@y 1);
> (c) y=<x° increases most rapidly and y = x”
x 74 x
least rapidly. The graphs for the higher
powers of x become steeper more rapidly
than those for the lower powers of x as x
increases above 1 or decreases below —1.
(d) The graphs of the even powers of x have
line symmetry in the y-axis, and the
graphs of odd powers have rotational
symmetry of order 2 about the origin.

1.3.4 To plot orto sketch? | 2 (b) When x is a large positive or negative

"4 : number the graphs of y = x* and
y =x’ + 4x are similar, in that, although
13 B they are not very close to each other, they
1 There is a temptation here just to join up the increase at a similar rate. The x” term is
crosses with a smooth curve; this would said to be dominant.
produce an incorrect graph. (c) When x is a small positive or negative
2y=-0.2, y=-—40.04, y= —490.004 number the graphs of y = 4x and
y=x* + 4x are very close together; the x 2
These values show that y decreases very term has very little effect and the 4x term
rapidly as x approaches 2.5 from below. is said to be dominant.
 FOUNDATIONS 583

3 The graph of y = x° — 4x” is similar to that of 2 The zeros are at —2, 1,4 and the graph is as
y = —4x? for small values of x and is similar illustrated here.
to that of y = x° for large values of x.
f(x)

| 4 The graph of y= x° + x* — 2x + 1 is similar

_ to that of y = —2x + 1 for small values of x
and is similar to that of y = x° for large
values of x.

5 The graph crosses the x-axis at x = 0, —1

| and 3. These values are related to the factors
of 3x + 2x* — x’, ie. x, (x +1) and (3 — x). 3 x(x +4)(x—7) =x — 3x4 — 28x?
They are the solutions of x = 0,x+1=0 x° dominates for large x.
and 3—x=0. —28x> dominates for very small x.
In the expansion of x(x + 1)(3 — x) the term
of highest degree is —x* and the term of va
lowest degree is 3x. So the graph of A>

y = x(x + 1)(3 — x) is similar to the graph of

y = —x? for large positive and negative values
of x, and similar to the graph of y = 3x for
small positive and negative values of x near
the origin.
4 (a) The stone is level with the point of release
| 6 (a) —x? (b) x?
when h/ is zero. This occurs when t = 0 or
(c) 2x*-—x? (You might have thought of 2.4, and the relevant answer is 2.4
trying x” — x° first.) seconds.

Ss When t = 4 then ) = —32. The point of

release is 32 m above sea level and the
p> D
height of the cliff will be a little less than
1 The relevant terms from each bracket are this.
multiplied.
f(0) = (—2)> x 7=28 and
x x (2x) = 2x?

2 From the factorised form for f(x), the graph can

only cross (or touch) the x-axis at —3} and 2.

1.3 Exercise 3.
f(x)
1 The zeros are at —3, 2,4. re
The dominant parts of / 2\
the graph are indicated i \
by the line segments. i "
The dashés indicate the soraay aan
completed sketch. i
584 SOLUTIONS

1.3.5 Rational functions 2ROsee

v,u >2.5kmh7!
1.3 Exercise 4

| x

! |
| = RiRy
(R; + R32)
For small R,, RR © R,;_ for large Ry, R © Ro.

(c) oe (d) Ly , , Sees 5

Ss 2ar
7s)

! ! 2n 0
| | ! F
1 =I
5
2 ieee a
>Gaeta
(e) (f) (so pete ee OE ee ee eee
ve ee
=.
|
: .
oes i REE eet | d ‘
1.4 Expressions and ;

sae sri ia equations

! if} \ 1.4.1 The language of algebra
: prone, aod

= 1.4 Exercise 1
g
1 (a)x=-3 (b)x=45 () x=-6,1
(d)x=+2 (ce) x=0 (f)x =0,4
(g) 354 (In lls: (i) —2,1

2 (a) 1+2+---+n=210
in(n
+ 1) = 210
nw +n—420=0

(b) (x — 20)(n+ 21) =0

n= 20 (positive root)

3 (a) The sum of three consecutive numbers is

always 3 times the middle number.
(b) n+ (n+ 1) + (n+2) =3n4+3=
3(n + 1), giving the sum as 3 times the
middle number (7 + 1) as expected.
 FOUNDATIONS 585

4 (a) For example,

a = 61 49 = 30 = 4 8
1.4.2 Quadratic equations
(b) Any two consecutive odd numbers can be
14 A
written algebraically as 2n — 1, 2n +1.
The difference between their squares is }1 x 4+6x+4=05 (x+3)?-5=0
given by
>x+3=-/5 or V5
(2n
+ 1)? — (2n —1)° =>x2--—5.24 or —0.76

= 4n? + 4n +1 - (4n* — 4n +1)

2 2
= 4n? + 4n4+1-—4n? +4n-1 2 (a) 46x47 =(245)

= 8n Ib 2 be

>x 2
+bx= ——
(=+5) —. _
4
which is a multiple of 8.

5 The problem can be (b) -e=-=4c

represented by the
2.

diagram. (c) The squares of both + a are

b* — 4c
equal lt to 5 :

By Pythagoras, Patio 04, 076 (b) 0.76, 5.24

ie: =4:19% “a9
(18=hY
=h +6
324 — 36h + h* = h’ + 36
1.4 Exercise 2
36h
= 288 —§+,/13
124i) ee --4.30 or —0.70
h=8
The bamboo is divided into pieces of lengths 8
6 4/12
(11) 76 12 = —1.58 or —0.42
and 10 cubits.
1 SAE /28
Gi))§ == = 155 or 022
6 Treating the area as one large square, it is
(a+b). 8+ ,/84
(iv) seEAES —0.12 or 1.72
Treating the area as one small square (area c’)
and four triangles (each of area 5ab), it is (v) x = —3.24 or 1.24
o+4xtab=c’ +2ab.
(vi) —0.84 or 0.24
Equating these two expressions for the area
gives
2 (a) (1.24, 7.73) or (—0.64, 2.07)
(a+b) = +2ab
(b) (5.56, —25.37) or (1.44, —0.63)
=> @+b'4+2ab=c
+2ab
(c) (1.19, 3.37) or (—1.69, —2.37)
> a+b ac
Since a, b and c are the three sides of each of
3 (a) 0, 8 (b) —1, 4 (Cc) e620 62:
the right-angled triangles in the diagram, this
proves Pythagoras’ theorem. (ah) ge = seh) (e) no solutions (E)P 5
586 SOLUTIONS

5) (2) eeSU
VySO, eS
4 as 5 (a) nO) ors (by tO

(DROME s (C= 2 SS
—1—f5 —1 R
(c) 16,4; —1,—64 = or a

6 The numbers are m and n+ 1. Therefore

nw+tn
n(n + 1) = 10, so 6 > 1000 > 77 +n—2000>0; n= 45

pecteval
atekey4
2
| teval 2
7E (a) O<x<1 (b)
x >0 or x<-}t
(c) -1<x<0
7 (a) 1G) no solutions (i) = 2
(ii1) 0.76, 5.24 (d) —2<x<-1 or x>0

(b) From left to right, the graphs are for (i), (iy <——) Sorel
(iii), (ii).
(c) If b? — 4ac < 0, the equation has no
solutions.
1.4.4 Inventing new numbers:
If b* — 4ac = 0, the equation has a single
(repeated) solution. ee
If b? — 4ac > 0, the equation has two
solutions. 1.4 Exercise 4
1 (a) -9 (b) —49 (COR =16 (d) —144

(e) 2) (f) —5 =12)

1.4.3 Inequalities
2 (a) 3) b) 4 (O) f/12
= jy
1.4 Exercise3 (d) 2j/S
1 (a) x < -2 (b) -5<5x>-1<x
3 (a) 14+; (b) —4- 6 (c) 10
(C) P26 Khas p< 7
(d) 13-4) (C= 92.07)
(d) 3x+15<2x+3>x<
—-12

(ce) x —2 4 (a) 3+), 3-j (b) -1+j/8

2 The square of any number except 0 is

(c) —2+ 4) (diye ss
positive. Therefore x may take any value
except 2. S57 (aos Sten, (b) no real roots

(c) —3,—-3; 2 real roots (repeated roots)

3 (a) -3<x<1 (b) x< —4 or x>2
(c) -2<x<2 GE (a) (x — 3/)(x
+3/)
(b) (w— (-1+ 2)))(* — (-1 - 2/))
4 (a) x<-5 or x>2
= (x+1—2j)(x +14 2;)
(bi x<2 of x>3

(©) 2S re 5
(c) (x— 3 +)v2))(x- 3 =jv2))
(d)
x > -2 =a x — 3 — jr/2)(x
—3+jx/2)
 FOUNDATIONS 587

7E For example, a 1.4 Exercise 5

(x* + 1)(x? +1) sere:
Ae oF 4 1 (a) x + x +7
x+1 )x apae Se eh)5)
—~ x+ x
x” + 8x
1 = x+ x
a ee)
. Tx+7
Sly
1.4.5 Polynomials el caeeas Setonee
remainder: —12

1.4 B (b) quotient: x* —4x +11

1 (a) (i) —24 (ii) —30 (iii) —24 remainder: —27
(iv) 0 (v) 0 (vi) 6
(vii) 0 (viii) —24 (c) 2x” +2x —3
x—1)2x* + 0x? = 5x43
(b)
x +1, x+3, x-4 ie ee

(c) («+ 1)(x

+ 3)(x — 4) 2x* — 5x
LYN= (x* +
My4x + eds3)(x— oe4) - 2x*
=
ae
2x

=x? — 13x —12 > —3x +3

0
2 (a) a= —2. The value is chosen so that quotiene= 2 4
eat = 0)
remainder: 0 (so x —1 isa factor)
(b) P(—2) = 0, therefore x + 2 is a factor of 2x? = $x ag 5
(d) quotient:
remainder: — :
a
2)-=0
3 (a) P(2) = (2—2)(2-2’ (c) Pues
Ghee ae v1 \tboe 4 oe
= x' 40x? = x
4 (x —3)(x — 1)(x +1) since P(3) = 0, ee ye
P(1) = 0, P(—1) =0 = 2x? + Ox” — 2x
=3x7 + 0x1
~ —3x? + Ox +3
=

quotient: x + 2x—3
remainder: —2
588 SOLUTIONS
ep

2 (a) P(-—3)=0 (b) Q(x)

=x? —Sx 44 8 (a) (i) x°—8 = (x —2)(x* +2x +4)
(ii) x* — 16 = (x* + 4)(x — 2)(x +2) .
3 (a) eg. P(—1) = 0'= x +-1Us a factor
(11) y
b) (x + 1)(x — 2)(x
— 4) (c) —1, 2,4

Each linear factor gives rise to a zero.

Quadratic factors which cannot be factorised
contribute no further zeros. Hence there will
4 1, —3 (repeated root) always be at least one real root. (Note that if
we consider complex numbers then there will
5 (%+2)(x7-+4
2x =2) = 0 always be three roots — some of which may be
repeated roots, of course.)
0B 273

6 (a) (x+2)(x* —2x+2) =0

= x==2 1.5 Numerical methods
(b) yA
1.5.1 The golden ratio

2 161<x < 1.62

The quadratic equation x” — 2x + 2 = 0 has

no real roots, hence the cubic equation has
only one solution.
1.5.2 Locating roots

7 (a) Zeros are —2, —1, 1 and 3. B

P(x) = (x — 3)(x — 1)(x + 1)(x +2) ES 1 For x? = x +1, draw the graphs y = x* and
y =x +1. The solutions are at the points of

ie
y&
(b) intersection of the two graphs.
For x* — x — 1 = 0, draw the graph of
T i iS Ze >
y = x’ —x—1. The solutions are the points
= ie Sel | | N73 x where the graph cuts the x-axis.
2 Either
(Ce 2, ore lr a Orcas
 FOUNDATIONS 589

or ya abe ¥e

One root lies between —1 and 0. The second

root lies between 1 and 2.

The roots are —0.618 and 1.618. (e) Not possible

Without a graph plotter, x* = x + 1 gives two

(f) 2x* —4x-—x+1=0
simple graphs to sketch, but care must be te de + 1%
taken to ensure that all solutions are obtained.
(g) Not possible (h) Not possible
Note that approximate roots of any quadratic
equation can be found by sketching y = x*
(i) 2x7 =5x—-1
and a straight line.
With x? — x — 1 =0 it is simple to check that x? = 5x—-1-—x

no solutions have been missed, but it is not so x = /(5x

—1— x’)
easy to sketch the graph without a graph
plotter. (j) As is the case for several of the other
formulas, it is easier to demonstrate
correctness in reverse.

1.5 Exercise 1
1 (a) The interval [3, 4]
ao NN)
(b) x = 1 is one root; the other two lie in (x- Ear2 == 38

[—1, 0] and [—4, —3]. 7 —3x

x —4x+4=
(c) [1,2] - (d) [-3,-2], [2,3] 2
2x* — 8x +8 =7—3x
2 (a) 3.14 (0) Sh, =O, ALO
2x? — 5x+1=0
(c) 1.72 (d) —2.32, 2.32
10
2 (a) (i) P=10Sxr
=—

4)
x

1.5.3 Iterative formulas

(0) Ba Sy SP STOO OL
1.5 Exercise 2 x = 2.15443 to 5 decimal places
1 (a) 2x7 =5x-1 (b) Sx = 2x*
+1 (b) (i) 2 =10 > x* = 10x
poet 1+ 2x? => x’ = ,/(10x)
xe x=
2 »)
=> x = /(/(10x))
(res See eaoleie Ene
x = 2.15443 to 5 decimal places
590 SOLUTIONS
aaa se a dn

In part (b) the convergence to the solution is (e) The iterations in (b) and (c) converge to
much faster. the root in the interval [2,3] but not to
| the root in [0,1], even with a starting
3 Using the iterative formula x;,; = }(2™) value of 1. ©
“4 = 3 The convergence of (b) is very slow.

fea The iteration in (d) will converge to the

x = 0.4578 to 4 decimal places root in [0,1] but not to the one in [2, 3].

4 (a) [3,4]

(b) x? -1=6,/x
sx =6/x+1 1.5.4E Convergence
=> x= /(6/x~+1)
(c) With x, = 3; x =3.37525 and ior
x = 3.495 358 to 6 decimal places. 11 (a) ¥8=250 9102.1

5 (b) x +2x-1=0 b) x”
(b)x" == 10
10 = xx"
2 == 10, giving x=>
10, giving =>
=>2x=1-x° If x, = 2.1, x) = 2.26757,
x3 = 1.94481, x, = 2.6439,
Sree $(1 = x?)
x; = 1.43057, x6 = 4.88633,
(c) (i) 0.453 40 (ii) 0.453 40
x7 = 0.41883, xs = 57.0073,
x9 = 0.00308, x49 = 1056 139.
(Gl) 2) = AGNES eh SHS Seis AS.
X4 = —5278 Odd terms form a subsequence tending to
zero, but even terms increase without
The sequence diverges and x, = 2 is
limit. Other starting values will give a
clearly an unsuitable starting value.
similar pattern. The sequence does not
converge.
6 (b) (i) x; = 0 would give /(=) , which
cannot be found. 2 : 10
| 2 (b) (x1, x2) is the point on the graph of y = —
(il) x; = 1 gives 2.280776 after 27 x
ierahone. located by moving vertically from (x,, x;).

x, = 2 gives 2.280776 after 24 (x2, 22) is the point on the line y = x

iterations. located by moving horizontally from
(x1, %2).
x, = 10 gives 2.280776 after 30
This method can be continued to generate
iterations.
further approximations. It can be
(c) x1 = 1 gives 2.280776 after 8 iterations. described geometrically.

x, = 2 gives 2.280776 after 7 iterations.

x, = 10 gives 2.280776 after 10 iterations.

(d) x; = 1 gives 0.219 224 after 11 iterations.

x1 = 2 gives 0.219 224 after 14 iterations. | 3 The diagram illustrates that the sequence
x, =3 diverges. diverges away from the root.
 FOUNDATIONS 591

NU, ioi BOF ak eps ares Can ere OR oo

Miscellaneous exercise |

iP — shes)

x x

(b) The sequence converges.

(b
y=)
va
i=
>
a5) Jara =) 0 x

104)
-4

can (c)
T >

ys

(a) converges to the root, whichever side of

the root x, is on; 0
5) Se
(b) diverges from the root, whichever side of
the root x, is on.

A cobweb diagram is obtained if g(x) is

decreasing and a staircase diagram is obtained
if g(x) is increasing.
4 A=2nr +—, 1r=43
A rearrangement will converge to a root a if r
g(a) lies between —1 and +1. In practice,
since @ is unknown, it is necessary to ensure
that g’(a) lies between —1 and +1 on an (c) y= (x—-2)2 +4
interval which contains the root.

3 10
(a)ars Se ae
(OS 2 Eire fh
10
= 3x = 2% =F
x 7 (a) 1660 (b) 100 + 190x (c) 765
: 1 10
== 3 (2+) GIs Cae
3 Ss

= PG) 5 (Le 23
(b) If ~ = 2.1 then g’(@) is approximately 8 S, = 95(1 — (o6)")» 53
zero and so the sequence converges very :
rapidly. This particular kind of 9 (a) 195 (b) 341 (c) nr+35n(n + 1)
arrangement is very useful in finding roots nin — 1)
of numbers. (d) <3)

10 (a) 100 (b) 0

592 SOLUTIONS

11 (a) (x+ 1)(x +4) (b) x(x

+ 3)
(c) (3x+4)(3x-1) — (d) (3x — 2)(3x + 2)
(e) (3x — 4)(3x + 1)

12a) ees (b) ; (CROPS —9) (d) 2,

SF)

13 2 seconds

$40 — 2)? = 2; 341, 07586

155 (Qe Onn —all

(Die 3

(Cie <— 3" or 2x

(da ON oT ONO <1 << 8°41

‘Celthy a 0) ieee
A = (x + 2)(x+ 1)(2x
— 3)
x== =2;—1,23

1 5)

18 (a) Yes (b) No (c) Yes

19 (x+ 1)(x — 2)(x — 3);

gS Oe Sih Sac W

20s — 9s — —1 (al tepeated! root)

21 (-1,-1) and (3,3)

228; losli7,
 593

2 INTRODUCTORY (b) Using the result of (a), an extra

2 xX 2m = 4m metres would be needed

CALCULUS (neglecting any deviation of the equator

from a perfect circle).
24
Rates of change —2) oe Surprisingly, this would still be 47 metres.

2.1.1 Introduction
2 Linear function
oom Exercise 1

1 (a) 3 (b) 3
ndPs Sone

aes: vy =20w + 16 (b) T = 35w + 46

2 (a) Difference in y-coordinates= —6;
difference in x-coordinates = 2 du dv dT
ie)
—i)
Se a
SOS
dw
20
dw
eas
dw
dy —6
b) ==—=-—
(b) Che : ESERIES
OES
LUISE
SE du .
For the first stage, ws the change in the
w
ea e7 Yb) i (2 ee time taken in minutes per kilogram of
Te

ERTS
weight (that is, the extra number of
4 (a) Difference in y-coordinates= —2; minutes needed for an increase in weight
difference in x-coordinates = 4 SLOT
of 1 kg).
dy -2 1 sesecnerras
du IE
Oe ir amigo — and — represent the corresponding
dw dw
~_ dy aan! changes for the second stage and the
overall process.

(b) y= —-x+7>5>—=-1
dy For each extra kilogram of weight, the
dx total extra cooking time is the sum of the
d extra times needed for each of the two
(c) —y=-x4+6>y=x-635=1
dx stages.

a) 2 = — apes pee ee eae du _ dv be

ee Rants a PERE dx 2
ae? dx
67 (a) C= 702. 2.8/n
(i) y= (3x+1)+ (Qx-3) Sy=5x-2
(b) ia
dc Saat
2.87. This represents the extra cost
(ii) y = 2(3x + 1) + (2x — 3)
n = cre — Il
(change in cost) for each unit used. (ii) y=uxt+4
(hig) sy eee 1S)
SK dy dz dy 4. cE ie
7 (a) ar 8 (b) (c) ay (i) ee (i1) ore

(@) dy= 10 (6) y=ie 12-20

>dy
7 = -20 ili ce
es iv a
a)

edy deh dy. sd ae ae)

8 (a) = = 2m. This is by how much the (d) ee ee ee
r ESI
circumference increases when the radius Pads e au adv. * dy " du a dv
increases by one unit. ay dx dx dx uy dx dx wae
594 SOLUTIONS
eS eee

(e) The rule should now be apparent. If 6 (a) P=38+8%t or P=38+0.2t

y = au + bv, then
dP
dy du dv (b) atk 0.2. The population is increasing by
dx * dx ‘2 dx 0.2 million each year.
(y is called a linear combination of u
and v.) P = 38+0.2 x 98 = 57.6 million
This may be reasonably close to the actual
u=12+4+5St population. However, 1998 is outside the
first half of the century so the simple
(b) v=9+6t
linear model is being applied
(c) The first firm is cheaper for jobs lasting inappropriately. During the second half of
longer than 3 hours. The second firm is the twentieth century the population
cheaper for jobs lasting less than 3 hours growth has been more irregular.
because of its smaller basic fee.
d d
muaes 5 and “2 6 are the two firms’
dt dt
respective extra charges in £ per hour
worked.
de du du
(e) a Ist a cet Apes

(f) Each extra hour that the job takes costs an

extra £27.

2.1 Exercise 2
d
1 ss (gradient)
dx

So the line has equation y = Sx + c.

(—1, 2) is a point on the line, so
(Sie
= S46 SNe =e — Hf
The change in steaming time is e minutes
The equation is y = 5x +7.
(4 seconds) for each extra gram of flour.

(a) y= —2x +8 (b) s=3r+1 f=200nh) > dpeee'4

P 15 df 15
= 25) + —_—_
=
(¢) p=sx+3

ay G
_ chy The change in pressure time is * minutes
Se ay (4 seconds) for each extra gram of flour.
oo ee ae
dC dT. “ds. wdp
T=s+p>
(a) G@) C=5+7t Wea ST Def>= aa
mee a dee
(b) (i) —-=6 (ii) C= 8+ 6t
dt df = 45
Test mark, T 26 49 50 The change in total cooling time is *
minutes (8 seconds) for each extra gram
Rescaled mark, R 0 4 96 100 of flour.
dR
(a) ao (b) R= 4T — 100
 INTRODUCTORY CALCULUS 595

2.2. Gradients of curves 2.2.2 Gradient graphs

2.2.1 Locally straight curves
2.2 Exercise 1
1 Any points where the tangent to the curve is
A horizontal correspond to a point where the
1 With an increasing magnification, the curve gradient graph meets the x-axis.
looks more and more like a straight line. At
x = 4 the graph looks more and more like a
horizontal straight line.
Zero gradient
2 You should expect to see the curve looking
more and more like a horizontal straight line.

3 (a) y = |x| is locally straight near all values of

XV except — 0;
The gradient decreases to zero and becomes
(b) y = 100x? is locally straight everywhere.
negative. The gradient then increases and
(c) y = Int (x) is locally straight near any becomes positive again.
value of x which is not an integer but has
a discontinuity at each integer value.
(d) y = |x* —4| is locally straight near all
values of x, except % = £2.

4E (a) y= ¢sin 3x has ; of the amplitude

(y stretch; scale factor 1) and oscillates
three times as frequently (x stretch, scale
factor 5) as y = sinx. The gradient never becomes negative but there
is just one point where it is zero.
y

sin x
(c) :

The gradient is constant for all x.

(b) Both graphs look like the graph of

y = sin x with ‘wrinkles’. When
magnified the wrinkles become more
obvious.
(c) y = sinx+ zgsin1000x or any similar
function. The gradient is always positive for all x. It
decreases to a minimum gradient but never to
zero (over the range shown).
596 SOLUTIONS

(c) wt
Gradient locally
a maximum
xv

Gradient locally
a minimum

There are two points where the gradient is

zero. (At one of them, the gradient is also
locally a minimum.)
~¥
dy
(f) dx
Gradient locally
a maximum

> (b) There are infinitely many possible original

Gradient locally
a minimum
graphs — formed by translating the above
graphs parallel to the y-axis.

The three points where the gradient is zero 2.2.3 Obtaining a gradient
correspond to stationary points on the
original graph.
2.2 B
| Your results for the gradients may vary slightly
_ from the answers given.

1 (a) 1.5

(b) x =) 215) = One atone

dy 2 215) STO ie
dx

(c) ie dy _
dx ~*~

ig

As x becomes numerically large, the gradients

a 4.2/0.2. 4.6
tend towards zero. As x tends to zero, the
gradient reaches its maximum value in (a), (a) Gradient 12 8 4 0 -—4 —8 =12
whereas in (b) the gradient becomes very large (b) 4 — 2x 2 84.0 —4 =39 0
and negative.
 INTRODUCTORY CALCULUS 597

E.2,C 22°.
_ 1 (c) You should find that, whatever points you
u Sy) Sell SAIL SOOM
chose, the gradient is the same as the
x-coordinate value.
Estimate of gradient 6.2 6.1 6.01 6.001

d
The equation is es, The gradient is approaching 6.
dx
dy The gradient of y = x” at x = 3 is 6.
dx
ENA
Qc,

2.2 Exercise 2
il WZ Pd ll72 3 6 4 8 Sy

6 (a) 6 (b) 10 (c) 0 (d) —4

(d) The graphs are related by translations up
or down. For any particular value of x the (e) 30 (E72
graphs will all have the same gradient.

(e) All the graphs have the same gradient

d
graph = =e

S ele
xv

The equation of the gradient graph is

— = 2x.
dx’
(b) Value of x 0 1 —2 3
(c) For any value of x, the value of y on the Gradient 0 3 12; 27
graph of y = x” is twice the value of y on
the graph of y= Ly? So for any change in
value of x the corresponding change in the
value of y will be twice as great.

dy = &
(d) 5

(e) = = 2ax. The gradient graph of any 14 —=b+2cx+3dx’

2 x
quadratic with equation y = ax* isa
For example,
straight line through the origin.
d
y=345x +x) > = = 10x + 3x7
| 3 The graph of y = ax’ +c is the translation of aS
0
that of y = ax” by |It has the same
c
gradient graph, with equation — = 2ax.
dx
598 SOLUTIONS
er UEEEEEEEEEESEDEEEEESEENTTTSST

d
5E You should have investigated the gradient 6 (a) a aa
functions of graphs such as, for example, cy
At (0,5) the gradient is 1.
y = 3x" + 4x3, y = 5x? — 3x4 7
The equation of the tangent is
The general conclusion you should have y—-S5=x-0,ory=xt+5.
reached is that, if a and b are constants,
(b) y=5—3x
= ae 4 bx
(c) y=S
=
dy = anx + bmx!
dx
The gradient of the sum is the sum of the
2.2.5 Differentiation from
gradients.
first principles
2.2 Exercise 3
1 (a) ers (b) eee ay ara F
dx du \ it @ =2.) big
(es Eye lewen lim (+2) =2
pr aee ss alee h—0 .
(b) lim (5h — 2) = —2
Mey) =10e
dx
ae
(c) lim (4h —h?) =0
(b) (i) 10 (ii) 20 (iii), — 10 h—0
(d) im (+2) =4
1G eS —6x" ae
_ Uh—-3)(b+3
lim) —————— = li -
When x = 0, gradient = 0 a oe h+3 pas a
When x = 2, gradient = —24 =.

(b) dy = 5-2
When x = 2, gradient = 1 2.2 Exercise 4
When x = 4, gradient = —3 ‘ i
_ 31 +h) -3 ,. £+6b+37 =e
dy 1 (a) lim =A = Jim A—___
4 (a) qt 27 h-0 h h-30 h
x
= lim (6+ 3h) = 6
At (2,3), 5-=2-2x2=-2
dy h-0

dy. 3(x +h)? — 3x”

The tangent passes through (2,3) and has (b) = = (a(S
gradient —2. Its equation is therefore .
y —3 = —2(x —2), or y = —2x +7. = lim (6x + 34) = 6x
(b) a 12x — 21 (c) Y= —8x +19 ae [5(x at bh) + 3(x at h)] ‘gtd [52x 4. 3x]

dy mt h-0 h

ap ee = 10x +3
Tene oat
=I),— = an and = yo
dx ‘ 3 tim MOH = 2(x +h) +7) = (ax? - 2x +7]
The equation of the tangent is Pamngy hb
y — 21 = 13(x — 3), or y = 13x — 18. = 8x-2
 INTRODUCTORY CALCULUS 599

s 1
AOEAeng 1 oe,
ae 2t Pip. 2r 5
dt h=0 h

a —2p
a ee Eoesineer + 5)
—2
- (2¢ + 5)? Consideration of the gradient graph
confirms the general shape of the (x, y)
dy (x + hy? ee graph. Final confirmation would come
SE tack jim [ear] from calculating the positions of the two
stationar ints.
x3 + 3hx* + 3h’x +h — x? pene ge
= Ph h | | 2 The local maxima and minima consist of the
3hx” + 3h’x +h | three points marked with dots.
oa
40 h
= jim [3x” + 3hx + h?] = 3x?

d
Soyer? > 2 = 3x?
dx

| 3 - s negative at the given point, so the

2.3 Optimisation x
gradient of the graph is also negative,
2.3.1 Graphs and gradient as shown.

graphs
SS z
Ao A
1 (a) For large positive and negative values of x, :
the graph has a shape similar to y = x’.
When x is small, the graph has a shape
similar to y= x + 1. 4 y gy
dx

(b) The zeros of the | x, ie

dy graph

correspond to the x-coordinates of the x x

two stationary points.

dy , ;
The value of z is negative between the As x increases through x = 0,—
dychanges from
dx
two zeros, corresponding to the negative negative to positive, actually becoming zero at
gradient of the graph between the two x=0.
ion ints. here the gradient
gO eee Ee
rsew here cre erecuch Since the gradient of the graph is negative for
is positive.
x <0 and positive for x > 0, there is a
(c) The cubic graph has at most two minimum on the graph at x = 0.
stationary points.
600 SOLUTIONS

>.4 Exercise 1
1 y= (x —1)(x —2)(x — 4)

The gradient graph has no zeros because

there are no stationary points on the
d oe
(x,y) graph. is always positive, and s
oi
is the gradient of the (x,y) graph.
The coordinates of the stationary points could
be determined precisely by using calculus.
(Note that the stationary points are not at 3 ya
oa —sleosandiaa—— 32)

2 (a) dy4
dx
2

zy
x

ds 5
4 (a) qf 7 12k + 12 = 3 — 2) 2
Zeros of the gradient graph correspond to
stationary points on the (x,y) graph. The A ds
* dt
dy
sign of — corresponds to the sign of the
dx \

gradient on the (x,y) graph.

>
2 t
i
ele

(b) ey
xv

The stationary point on the (x,y) graph

corresponds to the single zero on the /
Ne 7
/
d
gradient graph. = is positive for all other
%
values of x because the gradient of the
(x,y) graph is positive everywhere except (c) The particle is initially moving with spee
at the stationary point. ds
oF = 12. It slows down until it comes to

rest instantaneously when ¢ = 2. It then

speeds up again and increases in speed
forever.
 INTRODUCTORY CALCULUS 601
EY

dy
2.3.2 Quadratics and cubics 4 a>0: —=0has
dx
no solutions, and so there

are no stationary points.

B a=0: gives the cubic y = x°, with a

dy , stationary point at the origin.
= =3x’— 12
deo
<A): ey = 0 has two solutions,
dy
—=0 whenx= +2
dx
dx x = +,/(—}a), and so there are two
Nex. y=? —(12x2)4+2=—i14
stationary points.
Aex—=—2, y= (—2) —(12x—2)42=18
2.5 Exercise 2

d
bad
The stationary points occur when i == 0) athat
x
is when x = 2 or —2. The stationary points
are (2, —11) and (—2, 21).
ys
du
2 re
—=6 b6 ae 6 (=2, 21)

d
Shae fe whenx = —1
dx

u= 3x (-1)?+6~x (-1)+5=2
ua Ce =10))

d
(-1, 2)
by tay eh.Si|ronan
dx
= 6(x* — 3x +2)
' d
3 ~ = 3x* + 6x +5. From the graph for
Sloe= Nie)
| x
question 2 you can see that the least value of dy
— = 0 when x = 1 or x = 2. The stationary
d : id dx
< is 2 and so the equation EO Wasino points are (1,—2) and (2, —3).
dx dx
solutions. if

“Vv

The (x,y) graph therefore has no stationary

point and its gradient is minimum when
C= lk
602 SOLUTIONS

3 (a) (i) y= Sx—x* =x(5—x) (ii)

ee? 6 ng
dx

Maximum at (2.5, 6.25)

()G) you = 8x 412

dy
==) fone 0
dx

Maximum at (0, 12),

(b) (i) y=(1—x)? =1-2x 4x? minima at (2,—4) and (—2, —4)
dy
= —| (fons.
=1
dx

Minimum at (1, 0)

4 (a) +6, +3, 0

The quadratic has reflection symmetry
AE ee, A 2 OO
(c) (i) oe x — 3x" +5 and so B’s x-coordinate is
2 = 0 for x =0,x =2
dex (b) +6, +4, 0. The cubic does not have
reflection symmetry.
Maximum at (0,5),
minimum at (2, 1)

2.3.3. Maxima and minima

PIAS
(d) (i) y=4x—x? -—4 1 5000 people per square kilometre. (Note:
dy _ 2 putting r = 0 is an idealisation; there has to be
—=0forx=2 :
dx a small radius for there to be space in which
l live.
Maximum at (2, 0) asa

(i) 2 Not valid for r < 0 or r > 2.15 since P < 0 is

meaningless.

(e) (i) y = 2x? — 9x” + 12

d
== Oforx = i243
dx

Maximum at (0, 12),

minimum at (3, —15)
 INTRODUCTORY CALCULUS 603

dP At the higher speed, the number of miles

apa) meet wl SY) per gallon is lower.
At 35 m.p.h., the number of miles per
r 0.5 il Z, gallon increases with increasing speed. At
60 m.p.h. it decreases.
= 15 0 5)
dF
(b) 7a 0 when v = 41.7. The most
v
economical speed is 41.7 m.p.h.

2n=30—2P, R=30P—2P"
dR
(a) ap 30 — 4P gives the rate at which

revenue from the items changes with their

price. The best selling price is £7.50.
dR
The population density rises to a maximum of At this1s price
price, —
dP = 0.

20000 people per square kilometre at a

dR
distance of 1km from the city centre. by:
(b) Wheask en = 5; ap = 10)
The population density then decreases as you dR
When P = 10,— = —10.
go farther out from the centre. dP

dP dR :
(c) ae 30 — 4P must be positive and so
The (2) graph shows that the rnaaximum
r
P</250:
value of P corresponds to the value of r where
dP
—=(), If b is the variable height of the gutter, then
dr
the cross-sectional area A is given by
For r < 1, the population density increases
with r.
A = (20 — 2h)h = 20h — 2h°
dA
For r > 1, it decreases with r. aay)
ah 20 (Nat
— 4h

.3 Exercise 3
PF = 254y=—0.0121", 30<
1 < 80
(The inequality for v suggests that the car is in
top gear.)

1 00nd
du

y 35 60 (a) P = 500 + 100¢

oe 45.3 41.8 dP
= 100
a 0.16 —0.44
dt
dv This represents the rate at which the
population is increasing each year.
604 SOLUTIONS
NN

(b) P = 100(5 + t — 0.257) | Whichever way is chosen should lead to a

| maximum value for A when the shape is as
dP’=100
—502 | shown.
dt

dP dP
b= 1, 47= 50; t= 2,=0;

peas eee
dt
The population stopped increasing eo: 2
years, and then it decreased. * Lad Exercise 4
The maximum population occurred when 1 (a) The length of the net is
t = 2 and was w+l+w+l= 100.
100G 4. = 0 25 2) — 1600 2] + 2w = 100
The population decreased to zero and so l+w= 50
the estate was abandoned.
l= SO-w
5 If integers only are permitted, the minimum is Sc The width of the net is
7 when the numbers are 2 and 5. Sw+h+5w = 40.
If all positive numbers are permitted, the h+w=40
minimum is approximately 6.3, when both h=40-—w
numbers are just less than 3.2.
(c) V=whl
If negative numbers are permitted, there is no V = w(40 — w)(50 — w)
limit to the minimum value that the sum
takes. (d) The maximum volume (of approximately
13 130cm®) occurs when w ~ 14.7.
The dimensions are then approximately
14.7 cm by 35.3cm by 25.3.cm.
2.3.4 Graphical optimisation
2 (a) The dimensions of the box are
5
Zo DD
to
length = 6—2x, width = 4 — 2x,
| Two possible ways of expressing the cross- height = x
_ sectional area in terms of a variable are given
so the volume is
_ below, but several others are possible.
V = (4 — 2x)(6 — 2x)x
(b) V is maximum when x & 0.8.
The approximate dimensions are 0.8 cm
by 2.4cm by 4.4m.

A = (2+ cos 8) sin 6 3 The maximum volume is given by a cube of

side 18.4cm.
V = 6270cm’*

4 Maximum velocity = 17mmin!

Maximum acceleration = 6.75 mmin~~

A=(2+VJV(1—b°))h
 INTRODUCTORY CALCULUS 605

2.4 Numerical integration 4 This is an over-estimate. Because of the way

the graph curves, the six straight-line segments
2.4.1 Areas under graphs are all above it.

2.4 A 2.4 Exercise 1

1 (a) km (b) miles (c) cm’ (d) g 1 Cross-sectional area © 42.6 m*

2 ath of a revolution 2 (a) Assuming that the cross-section is always

circular, you can first calculate the cross-
sectional area, A, from the circumference,
C, by the formula
2.4 B D
1,3 The area under the graph represents the peel4r
distance travelled in the 60 seconds.
Speed
(ms) Height, 10 0 Ws
30
h (cm)
20
Area, USS) WAT sr Tlie stilts) 0)
10
A (cm’)
0 10 20 30 40 SO 60
Time (s) A (cm?)

200

2 10 x (29.94 23.14+-+-4+ 15.0) =

10 x 119.9 = 1200m
100
4 The area under the ‘steps’ is close to the area
under the curve of the graph.
5 The ‘step’ area is an under-estimate because
on i) 30 50 70 100 h(cm)
of the way the graph curves (‘concave
upwards’). (b) Area under graph
ee 20[198.9 + 127.3 + 71.6 + 31.8 + 8.0]
& 8750
C This area represents the volume of the
: 1,2 Speed stalagmite in cm>.
(ms)
40

30 3 The train comes to rest when t = 10.

20
Speed v(ms_!)

20 v= 20- 0.27
0 10 20 30 40 SO 60
Time (s)

SEL oS SplaemOlNOm
J 3 10x15 a ‘5 ee
S 10 ‘t(s)

ISasiqe Wisy
\= 1210im
Dy
606 SOLUTIONS

By the trapezium rule, the area under the 2.4.3 Numerical methods
graph from t = 0 to t = 10 is approximately
1x 2x (20+ 19.2)
+4 x 2 x (19.2
+ 16.8)
+3x 2x (16.8
+ 12.8)
+4x 2x (12.8
+7.2)
+4x 2x (7.2+0)
= 132 (b) m=at5
The distance travelled is about 132 m.
(c) x is increased by h each time.

2 Area = Th(yo +91) +34(11 + yo) ++

2.4.2 Integration
+ 3h(yn-1 + Yn)

2.4 Exercise 2 = 1h(yo + 29, + 2y2 +++ + 2p 1 + In)

1 The integral represents the area under the
graph of y = x between x = 0 and x = 3. b

Y |f(x)dx © 4h (yo + 29) +-++

+ 2¥n—1 + Yn)
y=x

2.4 Exercise 3
1 (a) Too small
(b) Neither — exactly the same
This is the area of a triangle (shaded), (c) Neither clearly too large nor too small
$x3x3=45. (d) Too large
3
Thus |reek == 41S 2 (a) Too large (b) Exact
0
(c) Not clear (d) Too small

(area of trapezium)
3 Type A: Shaded area © 63cm?
(b) 10 (area of rectangle) Volume © 63 x 80 © 533cm*
(c):3 x $(5+ 11) = 24 (area of trapezium) Type B: Shaded area © 42cm?
Volume © 42 x 15 = 70cm”
3 (a) 10tms' (assuming it is dropped from Total needed = (533 x 12 000) + (70 x 8000) cm
rest)
Speed
~ 6.96 x 10° cm?
(ms)
5 6.96 m®
(b) |10t dt
0 Adding 5% gives 6.96 x 1.05 =7.3m’.
—>
5 t(s)

(c) Area of triangle shaded, 3 x 50 = 125m oe Ti 0.3 0.59597 “Soe

4 ,/(10 — 1.57) + (10 — 2.5) = 4.72 y 0.9901 0.9174 0.8000 0.6711 0.5525

Area & 0.786 (to 3s.f.)

 INTRODUCTORY CALCULUS 607

2.4.4 “Negative’ areas 4

so |(x* — 4x) dx = —10.6
0
2.4 Exercise 4 Shaded area © (12.5 — 2.3) + 10.6
1 (a) From 5 p.m. until 9 a.m. The increase is ~ 21 square units
about 70 mg.

(b) The ‘negative’ area between t = 9 and

t = 17 is also about 70. So the amount
dissolved stays roughly constant.

2 (a) y= x? —x—2=
(x +1)(x—-2) eee
RBQA4 5

5
5 [,@+5) dx = (*) x 5 = 37.5
0

5
(il) 223 (iii) —43 (iv) 85 |(x* — 4x + 5) dx © 16.7
0

3 (a) At t =2 the net flow into the tank is zero.

So this is the moment when the water
level is highest (it stops rising and begins
to fall).

So Area represents change in volume of water

in the tank during the appropriate time
interval. ‘Positive’ area represents increase
in volume and ‘negative’ area represents Shaded area © 37.5 — 16.7
decrease in volume.
~~ 21 square units
(c) (i) About 10.6. This represents a net
This confirms the value obtained in question 4.
increase of 10.6 litres during the first
2 minutes.
(ii) About —5.0. This represents a net
decrease from t=2tot=S.
(iii) About 5.6. This represents a net
increase from t=0tot=5S.

(d) 60.6; 55.6 litres. About 24.5 minutes.

5
4 |Cdl)
0
The mid-ordinate estimates should be
approximately
S
|(x? — 4x) dx & 2.3
4
608 SOLUTIONS

2.5 Algebraic integration

4
(b) |x’ dx and | x” dx are the areas of the
2 a4
regions shaded above. By symmetry these
are equal.

4 4
(c) |e aly = Beak = 183. Similarly,

=)
| x? dx = 183.
4

3) 715
| 2 (a) (i) le x* ie dx= [4x] begs
—1.5
tala

(it) 2.25 (111) 7.875 (iv) 10.125

lex? dx = A(3.1) — A(2.5)

I) ey IE

A(u)
(il)
| 3 (a) Shaded area = ; Xuxu=uw

(b) ie
Drache = eae = 7

f(v) = 3u
iex? dx = A(3.8) — A(0.7)
0.7
8-3) Ole 1822

0 u v

(b) Shaded area =} SS Sie 3°

 INTRODUCTORY CALCULUS 609

5 (i) (a) (c) The graph of y = 2x* is obtained by a

one-way stretch x2 from the y-axis. Areas
f(x) =2
are therefore increased by a factor of 2.
N
oem
u x b b
(d) |kx? dx =k |x* dx
(b) Shaded area = 2u

(ii) (a) 8
1
0
aa

=3 g(t)=—3

(b) Shaded area = —3u

(ili) (a).

(b) Shaded area = wv”

(iv) (a) a

3,4 f(x) =ax*

+ bxe+c
3 b 2
=> A(x) == Jee Bex

5,6 f(x) =ax? + bx +ex+d

4 OC
GX;
3 2
= A(x) = —i = — 3 + —5 + dx
d.

2.5 Exercise 1
2.5.2 Integrals of polynomials 4
1 (a) |(3x2—5)dx
4
(b) [> a 5s] 346
9) 2

(a), (b) ”4 4 y — 2x? 1 1

2 (a) | (fo 2 = 3) di = [et +30 — 34] ;
=) =

= -63
jax

0 Saeee 0 Sex
610 SOLUTIONS

2.5.3 Numerical or algebraic

integration

i 3 3
(1 |4(e—syrade = 4] (@ — 108-425)
0

Between t = —2 and t = 1, the graph lies

= 4]423 — 52 2 +254]
3

completely below the t-axis and hence the

shaded area is negative. = 156 (kg)
| 2 (a) Average productivity is ate
— per hour
= 44.6kg per hour.
1S 15
(b) |
0
Me 5 dee 4]ieSsr + 25¢|0
— 1098S
This is the number of kilograms produce
in 2 hours, so the average productivity is
54.75 kg per hour.

) ,4(t — 5)* dt = 58.17kg produced in

1 0
) [Le — 2x* — 5x + 6) dx = 8 1 hour 10 minutes.

Average productivity is 49.86 kg per hour

fe — 2x* — $x +6) dx = —183
1
| 3 If the catalyst is changed after t hours, the
The total area is 8
9 +46 =— Br~ 21.08.
output in kg per hour is:

4 (a) 30m (b) ¢ = 2.5 seconds (when v = 0)

(e)esile25mn
4(ir — St + 25¢)
a (Oss)

5 (a) —} (b) 0 (c) =23 (d) j Output per hour

(e) 13 (f) 38 60 (1.29, 55.07)

40
6 Area= 132
20
7 a=1and 4

8 213 metres

Using a graph plotter you can see that

maximum output per hour is achieved when
the catalyst is renewed after approximately
1.29 hours (1 hour 17 minutes), giving avera:
productivity of 55.07 kg per hour.
 INTRODUCTORY CALCULUS 611

4E In a 3} hour production cycle 156 kg is

produced.
Profit in £s is
(156 x 3) — (50 x 34) — 150 = £143
(a) Profit per kg sold = £143 + 156 = £0.92
(b) Profit per hour = 143 + 34 = £40.86

5E Profit per kg is a maximum £0.94 when

t = 2.53, i.e. the catalyst is changed after
2 hours 32 minutes.
Profit per hour is a maximum £47.13 when
t = 2.155, i.e. the catalyst is changed after
2 hours 9 minutes.
The company should maximise profit per
hour to achieve maximum profit per annum. Area of shaded rectangle
difference in : ;
= se ieatte x difference in x
difference in x

2.5.4 The fundamental = difference in y = g(b) — g(a)

theorem of calculus

g(7) = 10 and g(1) = 3, so g(7) — g(1) =7

7
The result |g (x) dx = g(7) — g(1) is always
true, ,

b
| 3 |g (x) dx = g(b) — g(a) for any function g
Area = 0
whose graph consists of a series of connected
The difference in the y-coordinates of the end
line segments.
points equals the area under the gradient graph.
612 SOLUTIONS

2V8S =

ae ae Ge a
Oneal 2a SP SoS kG Ny) ox

(i)P A= [423]
3]? + [8x-425]"
eRe ee
= 4.42
Any of the graphs shown should give the
(x, g'(x)) graph shown on page 168. (i) The graphs intersect at (0,0) and
(b)
(1,1). The shaded area is
Each graph is a simple translation (up or
1
down) of one of the other graphs. There
pxixi-| x dx =4-1=1
is an infinite number of such graphs, all 0
sharing the same gradient graph.
From the graph on page 168 you can see
6
that |g (x) dx =9
1
6
|g (x) dx = g(6) — g(1) for any correct
1
(x, g(x)) graph.

2.5 Exercise 2

peai
a
SSeS

(ii=e AreaA+B=6x4=24
Mteac@
= 244

4
Area B+ C= |
2
157 dx = 9 13
The graph has rotational symmetry about
(1,0). The areas shaded are equal. So area A= 2444-9}= 182
y Required area = 373
2 Ea — 52 +7] ——>

3 (a) 10x+3 = (b) 627

Note that h(t) = 6¢* for any integral function
of the type 2¢° + c where c is constant.

4 (a) (i) The graphs intersect where

x? =8—x’ > x=2 (since x > 0).
The graph of y = 8 — x’ intersects the
x-axis where x” = 8 > x = 1/8.
The integrals arise from splitting the
areas as shown.
 INTRODUCTORY CALCULUS 613
NS

(iii) The graphs y = 3x” — 12 and 2.5 Exercise 3

y = 12 — 3x” both intersect the
Keamdsvat xe 1 (a) Ix4@_x+4¢

By symmetry, the shaded area is

twice the area enclosed by the curve
(b) fe Ge = Dye
y = 12 — 3x’ and the x-axis from
et Onye—— = | t-x-2)a

-|
yy
3
Shaded area = 2 (12 — 3x”)dx Bh
so eh Da
5% Ps 2x) ent
=3
=i)

3 2
= 2[12x - 23] a
Daa) y=4tx—4x4+¢

(b) y=xrtixrte
5 Shaded area = |
0
(c) yatxe+hxt+xte
> =18
(yaa) 2)) =x? =x=2
c=
V/18 = 2.62 lye?
tex 7 _
Sey
2x —2x+¢

6 |x?-x'|" =o 3 (a) y= x? 42x42

=>a-a=0
(b) y=txr tix? +x43
= a= 0 (which does not apply here)
Oma —
4 (a) x —x? 45x
(b) Qt 1G — 4) S27 — 714
2.5.5
The indefinite integral k(t) =$0 3 —fr 2 — 4¢

5 P=0.2t+c
* ‘AtF
_ They each differentiate to 2x + S. 58 million

/ |@ + 5) dx =x? + 5x +k. When evaluating 6 (a) s= |aede=2? +c

the integral between x = 1 and x = 2, the

c is the value of s when t = 0, i.e. the
_ constant k will ‘cancel out’:
distance of the sphere from the top of the
2 2 ramp when it is released.
|(2x +5) dx = [x + 5x +],
1 (b). s = 2 + 0.5
= (14+k)—(6+k) The ramp is 2.5 m in length.

= (c) t= 0.87 seconds (to 2 d.p.)

Hence any of.the functions could have been

_ used.
614 SOLUTIONS

Miscellaneous exercise 2 18 (a) (i) 835 (ii) 84

(b) (i) 27% (ii) 272
1 (a) 2x (b) 8x°+3 ©©=— (c) — 12x"
(d) 6 (nlOx— 1. (6) 15x" 19 (a) 44. (b) 57%. (ce) 108

20a) ee a e168 ac) — 13 20 202

3 (a) y= —2x+3 (b) y = 12x — 8

21 (a) 0.693 (b) 1.57 (c) 1.26

(c) y=-1 22 39, 39

x* + 2x + 1 is obtained from x” +1 by a
4 (a) G,33) — (b) (13,73) ~—() (0,3)
tr anslation
0

5 (3,11)
23 (a) txt 42x +k — (b) xt+ix?
+k
6 (2,4)
(c) Lax? +4bx? +cx+k

UW ee oy N= kg Il
24 x=25°—-P42;

8 y=3x—-3, y=12x-17, (¥,¥ x is greatest when t = 14, v when t = 3.

9 (a) v=3t°-—6t = (b) t=0or2 25 (a) The volume of water flowing between
12 noon and 4 p.m.
(c) Backwards, 2s
(b) 2340 m?
10 (a) 30ms? (b) t=3
26 (b) 21
(c) 45m
27
11 (a) 20ms' (b) 125m _—(c) 19.6255
28
12 (a) 30m —~— (b)'11,:- 7

13 @ji2;—1) — Mb)- G4)

(e). (252) (d) (4,48)

14 ya Sanit dy tO

15 (a) (2,—-1) (minimum)

(b) (0,0), (3, -1#2) (minimum)

(c) (3,—8) (minimum), (—1,23) (maximum)

16 wes, i

17 (a) O and 8s (b) 2 and 6s

 615

3 FUNCTIONS (c) fg(x) =6x+1,

no values of x
gf(x) = 6x —2,

3.1 Algebra of functions (d) fg(x)=-,

i
gf(x)= = all values of x
x x
3.1.1 Composition of (e) fg(x)=x+2; ef(x) =x + 1,
functions no values of x

(f) fe(x)=V(x—- 1), gf(x) = V(x) -1,

Exercise 1 ell

1 (a) (i) 2x°+3 = (i) (22 +3)?

ne MoS
(b) (i) es+1 (ii) ea

(c) (i) 3(5—x)+2=17-3x

(ii) 5 — (3x+2) =3—3x
(d)- (i) 1 — (1 — 2x)? = 4x — 4x”
(ii) 1— 2(1 — x”) = 2x* -1

2 (a) ct(x) =9+0.4 x (1.034x) = 9 + 0.4136x

(b) The cost in pounds of x cubic feet of gas.
3.1.2 Range and domain
3 There are alternative answers for all of the
following questions.

(a) f(x) =<. SexyH x2

(b) f(x)=2+2, g(x) ==

fe =i = 2x3

f(x) = 2x g(x) = Vx
ae)==+3, ore
(f) f(x) =x", g(x) =2x+1 Domain {x € R: x £ 0}
(g) £(x) = x* — 4x — 3, g(x) =x
Range
{y € R: y > 0}
4 (a)x+4 (b)x* (c) 4x-9
(d) x (e) sin (sinx) (f) x

5 (a) A(g(A(x))) = Flex —3))

= £((#-3)*)= (x33 3.1.3 Inverse functions

1 B
1 (a) £1 reverses the effect of f and so
6 (a) qs(x) — (b)_sq(x) (c) s*(x) f*(£(x)) = x.
7 (a) fg(x) = (x+3), gf(x)=x° +3, x=-1 (b) If £(x) = y, then f-'(y) = x. Any point in
the domain of f is therefore in the range
(b) te(x) =x —3, gi(xs=x—3,
of f"' and vice versa.
all values of x
616 SOLUTIONS

ho!
(x) = /x+7

Domain of h= {x € R: x >7}

(d) If a function is many-to-one (as is y = x

in (c)), then the inverse is not a function
because it is one-to-many.
YA
r(x) = (x - 6)
y= Vx
Domain of r= {x € R: x > 0}
|
|

| 3 Reflection in y = x (resulting in interchange «

~~ > >
| x

|
|
x and y in the equation)

| 4 Ge) y=4vle43)-5 tb) y= 4xtt

y=-Vx

5 (a) (i) The reciprocal sequence alternates

between two values, except when
x1 = 0 (x undefined) or x; = +1
(constant sequence).

(ii) The ‘change sign’ sequence behaves

similarly; it is constant for x; = 0.
fax) = aie)
(b) Both functions are self-inverse and hence
Domain of f = {x € R} f?(x) = x in each case.

The graphs of both functions reflect in t!

g "(x)= v(x +7) line y = x onto themselves.

Domain of g = {x €R: x > 0} 1-x 1

ox (b) 12—x (c) poe

(a) /(2) ihe eieaieaee

(Ey 14 ee?
(b) and (e) are self-inverse and have y = x as
line of symmetry.
 FUNCTIONS 617

Coe.
3.1.4 Rearranging formulas

Exercise2

1 (a) x =1(2y +3)

(c) x=S+VJ/(y—-4)
(b) x =7y +5

id) x= 3+
8
aS in}

n 2 lI
Need ee)
N (a) £1 (x) = 3(2x + 3);
f has domain and range {x € R}
(b) f"(x) =4x45;
f has domain and range {x € R}
Of (Ge) =5 +./(% =4):
f has domain x > 5, and range y > 4
or f(x) =5—/(x — 4);
f has domain x < 5, and range y > 4

hat a(x) = 3 nee

x
f has domain x # 3, and range y # 0 (gee)
ae a ee
Sma)ina bs cea
~ 900 —y
7 (a) Area of cylindrical surface = 2mrh
(b) Since a and y are both positive, the Area of ends = 217°
denominator (900 — y) must be positive.
Total surface area = 2arh + 2nr*
pealory. aay S = 2ar(hb +7)
ae is, ee |
ies,
5 The image will be y = f(x)= ; —
te
Se.
5 a ee
x+1
eee een S
(b) ie omeant

3.1.5 Parameters and 3V 100]

functions -

3.1 Exercise3
9 (a) ==e
Fae I(R+hielo
nr) = nE
—b 1
1 (a) W="—"
C a
or =(p—6)
a
=> IR =nE
— ntl

(b) Sale
n(E — rl)
Sk =
I
aay: Gp es I ay ae
n we s (b) E=4mv* => 2E = mv
R ft iP —R
Oe (since 2 = 3-5 = 755) rons

~o-y(2)
=> —_ = V
m

Dilarwuleowect. (iv) 175. feet. Gil) 6315 feet

(b) v = ./[20(d + 5)] — 10 (c) 61.4 m.p.h.

618 SOLUTIONS

parallel to the x-axis; and that adding 2 to the

3.1.6 Functions and
function was equivalent to a translation of 2
transformations of parallel to the y-axis.

graphs So it is reasonable to suggest that the image o

3 ; Sales
y = — under a translation is
96 2
a1 EC
3
1 Graphs b, c and e can be mapped onto each ee et
other. Graphs a and d can be mapped onto 2) Gar
each other. —2
1 4 y= (x +2)? —1, a translation by lea
5i
2 (a) The translation through |5 2
3
4

‘ uz 3.1 Exercise 4
(b) The translation through ,
1 y

1i
(c) The translation through |4

3 b can be mapped onto e by a reflection in the

y-axis.

0
y = x’ translated through S
Bele 1D)
1 (a) f(x)=x*, £(x)+2=x*+2, 2 iy
f(x+3) = (x +3)*
(b) The three graphs are congruent and the
transformations needed are:

>
(i) translation B
»

(ii) translation ee y ee
= x translated through A1

1 1
2
g(x)
(x =
x
—S5
g(x+ 4 4)= ear 3 ae
3 y =~—translated through
1 oy 0
+4) +3=——_4 3
a(x Bs ae.
4 y=»? translated through |4
The three graphs are congruent and the
transformations needed are:
5 y

(i) translation GA
0

(ii) translation is

3 In questions 1 and 2 you saw that replacing

‘x’ by ‘x — S’ resulted in a translation of § y = 5x* translated through i
 FUNCTIONS 619

(b) The equation of the reflected graph will

be y =f(—x) =x*+ 2x°.

3 (a) f(—x) =3x* — x*

eae Ox (x el) 1
; =I
i.e. y = x” translated through | i

The graphs of f(x) and £(—x) coincide,

since (—x)? = x* and (—x)* = x*.
3.1.7 Combining
(b) f(—x) = —x? + 5x = —f(x)
transformations of y= f(x)

graphs

\
i y = f(—x) = -f(x)

Here f(—x) = —f(x), since (—x)? = —x°

and 5(—x) = —5x.

| 4 (a) Even (b) odd (c) neither

5 (a) Even (b) neither (c) neither

(d) odd (e) odd (f) even

va | 6 (a) After the first reflection,

f(x)
y= —f(x)= —x*? —3x4+2= g(x)
After the second reflection,
y = g(—x) = —x? + 3x +2
—f(x)

(b) (i) reflection in y-axis

(ii) reflection in x-axis

(c) Yes
(i) If (a, 6) is a point on the graph of
y = f(x), then (—a, b) will be a point
on the graph of y = f(—x). (b) A single equivalent transformation is a
180° rotation about the origin.
(ii) Similarly, if (a,b) is on the graph of
y = f(x), then (a, —b) is on the graph
7 The equation of the curve obtained by
of y = —f(x).
reflection is

2 (a) The equation of the reflected graph will be

y = —f(x) = —x* + 2x? y = f(—x) = 2x" +—= g(x)
620 SOLUTIONS
a eee

The equation of the curve obtained after 3.2.2 Transformations

translation is

y=eg(x—4)+3
3.2 A
| 1 y=sin0° maps onto y = asin@ by a stretch
of factor a in the y-direction.

\ y=asin&®

3.1 Exercise 5
1
1@y=@-4" by=
(x—3)
(c) y=—-2-|x| (d) y= 3 -2(x
—4)?
(ec) y=(x+3)?
+2

(f) y= (x +3 —4)J/(x
+3) 41
= (x-1)y(x+3)+1 _ 2 (a) y= sin@° maps onto y = sinb@ by a
ft. t :
stretch of factor b in the 6-direction.
(b) y= 3x+7

y = sin 6°

(x2) y = sin (26)°

>=y=sin (26):

3.2 Circular functions 360

(b) The period of sin b0° is (=) ‘
3.2.1 Rotation
y =sin(9+c)°
3.2 Exercise 1

ay eee a
ov

(b) e.g. 130°, 410°, 490°, 790°, —230°, —310°

aS y=sin@
+d
2 (a) cos 163° = —0.96

(b) —197°, —163°, 197°, 523°, 557°

3 (a) sin339° = —0.36 y = sin(@ + c)° is obtained by a translation «

—c in the @-direction. —c is known as the
(b) 201°, 561°, 699°, —21°, —159°
phase shift.
y = sin@° + d is obtained by a translation of
in the y-direction.

y = sin(6 + c)° +d is obtained by a

—c
translation
lati o f | ‘|:
 FUNCTIONS 621

_ 4 (a) As in question 2, y = cos b@° has period aah

360
h = 0.2 + 0.2 cos 0°
oe
(b) y = cos @° is mapped onto 0.24
y = cos (b8 + c)° by a stretch of —
followed by a translation in the
@-direction.

Thus y = cos (b@ + c)° has period =, 3 (a) A one-way stretch of scale factor : parallel
and phase shift “5 to the x-axis, followed by a translation
—20
|i |and then a one-way stretch of
NB This is not surprising, since the
maximum value of cos (b@ + c)° is 1, scale factor 2 parallel to the y-axis.
which occurs when b@+ c = 0,
c
(b) (i) (180°,0)— (60°,0)> (40°,0)
ego ee —* (407,,0)
(ii) (90°, 1) — (30°, 1) = (10°, 1)
: SEOne combinations a— 2,003.) jo— 30 — (10°, 2)

| 6 y=cos@° is mapped onto 1

y = acos (b0+c)° +d by stretches of a and — 29
in the y- and 6-directions followed by a b
;
ae 3 Modelling
}
periodic
.
translation | : |
|—ofb
. Note that the stretches
behaviour2
ak;

have been done befoxe the translation.

B
(i @=6@

3.2 Exercise 2 12h h

; ~ 104 10.4
~1 In each case there are alternative correct
answers. oy ey
08 08
(a) y=2cos (b) y= —3sin@ 180 ae 0 30 Bye

(c) y = 10 cos6 (d) y= 4cos@°

+4

(ec) y=sin@?+2 = (f) y= 3sin‘ 26° fi 5161458

ae Ge b=56—48
a: 6t°

(g) y= 20s
60° + 1 |3
0 0 50 COU LOIS Ose 0
(h) y = 4sin (36 + 30) + 3 hh 08 14 3256 80-98> 104
my or

PN = PM+ MN aie toa 10.5 152 220005 80

: h OPS 2256S On Seam
h Soh=0.2+0.2 cos 0° ee
eee ee ee

Pee
622 SOLUTIONS

of SOULE NG
3.2 Exercise 3 4E (a) E=sin (=)
28

1 S/360EN:
3 Sl n i
SS33

(c) Critical days occur every 11.5, 14 and

16.5 days respectively for the three
cycles.
(d) All these cycles are critical about once
(ii) b = 0.8 cos 30t° + 6.5 every 74 years.

(b) hb = 0.65 cos 30t° + 4.25

2 From your first graph you will see that the

points from April to October, inclusive, do 3.2.4 Inverse trigonometric
not conform because of the British Summer functions
Time adjustment. To make the data easier to
& raph you might proceed as follows:
® subtract 1 hour from sunset times in BST,
e change from hours and minutes to hours
written in decimal form,
count days from 12 December (when sunset
is earliest).
This should give the table shown below,
where t is the number of days after 12
December and s is the sunset hour.

20 48 6a LO Same 3 eS Oa 87
s 16.05 16.75 17.60 18.40 19.18 19.92 20.35
From the graph the other solutions are:
PAS ANB) PKI PE) SP SSS) 180 — 23.6 = 156.4, 360+ 23.6 = 383.6,
s 20.17 19.45 18.43 17.37 16.43 15.90 540 — 23.6 = 516.4

) 3600 + 23.6 = 3623.6,

The graph of s against t is an approximate 3600 + 156.4 = 3756.4
sine or cosine curve, having amplitude about
360n + 23.6, 360n + 156.4
2 .28 (hours). A possible equation is
) Yes. For example, if 1 = —1, the solutions
eStigite os cos (ee 23.6 — 360 and 156.4 — 360 are generated.
365 The graph has period 360 in both positive
and negative directions.
Bhd = 12+ 2.5 cos 360t°
p = 30
540 — 30, 720+ 30
wa 180 x 20 + 30 = 3600 + 30,
180 x 21 — 30 = 3780 — 30

&
9:5
180n + 30 if n is even, 180n — 30 if n is
odd.
0.5 1 t
Using the (—1)” notation,
x = 180n + (—1)"30.
 FUNCTIONS 623

3 (a) The principal value is 44.4 You could find subsequent times by extending
The general solution is 180” + (—1)"44.4 the range of values of x beyond 360. The next -
(b) The principal value is —44.4 orate
The general solution is 180” — (—1)"44.4 x = 383.58 and 516.42
=> 30t = 383.58 and 516.42
(c) The principal value is 45.6 Be Pa pirece nay ote
Other solutions are 360 — 45.6, 360 + 45.6,
720 + 45.6, 720 — 45.6, etc. giving, as expected, 1247 and 1747.
The general solution is 360n + 45.6
Height (m)

3.2 Exercise 4 :
fia) edt.5° (b) 258% = le)i--21,1°
(d) 137.7° (e) 90° (f) 180° 0047 0513. ~‘Time

TAG) i7leoyy ICP2Sys SYA eis Say) EM Sey 84 Pap) 2 From the graph, the depth is greater than 6m
between 0047 and 0513, i.e. for 4 hours
(6) 36.95 3231, 396.9, 683.1, —36.9, —323.1 26 min, twice each day.
(c)) 107.5; 252-55 467.5, 6125; —107.5, —252.5

(d) —30, —150, 210, 330, 570, 690 3.2 Exercise 5

(e) —180, 180, 540 1a) Sisin 10° = 5 | sin 10 = 0625

‘ => sinx° = 0.625 where x = 10¢
(f) 19.5, 160.5, 379.5, 520.5, —199.5, —340.5 1-0 <1 < 60, then 0 <7 = 600

x = 38.7 or 141.3 or 398.7 or 501.3

3 (a) sinx® = 0.2 => t = 3.87, 14.13, 39.87 or 50.13
Calculator value for x = 11.5 (to 3 s.f.)
Solutions are 90 + 78.5 + 360n. — (b)Saaeee 7 cos (t+ 35)° = 4. So, if x = t+ 35,
COS = Ol
(b) Calculator value for x = 143 (to 3 s.f.)
=> x = 55.2 or 304.8
General solution x = £143 + 360n
=> t+35 = 55.2 (or 304.8)
Other forms are possible in both (a) and (b). = t = 20.2, since remaining solutions are
not in the required interval.
(c) ¢= 31.2 or 41.6
3.2.5 Solving equations
ey x “ ca i Le * 4
ie ees
234
2 5.6 — 4.8 cos 6t° = 9 > cos 6t® = ——
3.2 D a
praeee ay = eS
6 = 2.5 sin 30t + 5 = sin 30t = 0.4 The chair is above 9 metres for
Using a calculator, sinx° = 0.4 > x = 23.58 37.5 — 22.5 = 15 seconds.
and, using the symmetry of the sine graph, ha
x = 156.42 is also a solution. So

30t = 23.58 or 156.42

= t=0.786 hours or 5.214 hours

The height is 6 m at 0047 hours and 0513

hours.
624 SOLUTIONS

3) (a) eee sina 0Ee— 1647 —> Sines 0t — x 3.2 Exercise6

1 (a) 45° (b) —80.5° (Cn 0:

SS = 1B AO, ISAS IG S7/

Times: 0126, 0434, 1326, 1634 2a) ya

(b) 5+ 2.5 sin 308° = 4.5 => sin30f° = =

T T >
90 180 x
— 16. oerlIkO2 lS. 35,929-02

Times: 0623, 1137, 1823, 2337

y= tan 2a
BSS —~ aS 2+ 1.5 sin 500¢° = 2.75 = sin S00t° = 0.5
— i — OL060053
0.06 and 0.3 seconds from the start and (by 74
repeatedly every 0.72 seconds

(b) 2+ 1.5 sin 500¢° = 2 = sin 500t° = 0

T T a ieee
= (p= 0 O30 Was 46 90, 180
Every 0.36 seconds from the start

(Ce leSisino 0Of ==. 9st 00% wal

Si = OG
0.18 seconds from the start and every 0.72 y = tan (x + 45)°
seconds thereafter
3 (a) x =71.6 or 180 + 71.6 = 251.6
(b) tan (2x + 30)° = 0.8
=> 2x + 30) = 38-7 o1 21857 on 398e7,
or 578.7
=> x = 4.4 or 94.4 or 184.4 or 274.4

(GC) tan scn set

(b) the acute angle whose (ange) — eis bo == 45 axe DUS
tangent is 3; tan'3 = 71.6° tan = = = 1S Son 16:

(Gd) =4isinies = Sicosics

2 (a) y (b)
15 4 sin x°
a cos x°
-71.6
= tanx =—
4
71.6 0
OR Ote low
LALA
ALSSAPRA
LESCOL
LSE
AOL
a
3 tan = —? : 4 (a) x=71.6+180n (n=0,1,2,...)
i

sin 0° _4 b oe a 4
(b) 2x +30 = 38.7 + 180n
COSURN CN CaNeD b
2x = 8.7 = 180n
4 All real numbers @ for which cos 6° 4 0 x=44+90n (to 2s.f.)
|

—90° < tan! x < 90° (n= 0,1,2,...)

 FUNCTIONS 625

3.3 Growth functions 5 (a) Time (s) 0 1 2 3 4 5

3.3.1
ee enn
Exponential
es ee bee
growth
se Mee
ce ast ee 0.89 0.88
Growth factor
ae 0.80
ey 0.83
ie 0.86
q : Time (s) ad ee wa
> Exercise 1 Ghaeve (v) on ao Pes de
1 Although the growth factors show slight Growth factor 0.75 0.67 0.50 0
variations, they are all 1.20 to 3 s.f. Allowing
for the error introduced in rounding the The decay is not exponential.
profit figures, the profit is growing
exponentially. (b) Time (9 ory aes :
Charge (V) QD 6.75 5.063 3.797 2.848 2.136
2 In each case the annual growth factor is 1.09. Grownicron 0.75 0.75 0.75 0.75 0.75
This means that the amount in the account at
the end of the year is 109% of that at the , :
beginning, so the increase or interest rate The decay is exponential with growth
is 9%. factor 0.75.

Since the growth factor is constant, the

growth is exponential.
3.3.2 Indices
3 Age Pocket money (£) | Growth factor
0 Oem 3.3 A
1 0 : i
é fe of (ah 1G) 22" = oe)
2, 70 IU 5
3 80 i m factors

4 90 113 C222)
ce See
5 100 int ] pe
=e Dae ea 2
Since the yearly growth factor is not constant, | SS ae
this is not exponential growth. (m+n) factors

(ii) A similar argument may be used, or

4 Notice that the time intervals are sometimes the observation that, from (i),
4 years and sometimes 8 years.
gn ytaa
Over the 4-year intervals, the growth factor is
approximately 1.12 in each case. (b). (2)? =2" x27 = 70+ = 7m
: (2% = (are seg — gam +m = gem

Over the 8-year intervals, the growth factor is The nth term in this sequence of results is
approximately 1.24 in each case. (xe = Dae

Over two 4-year intervals the population

would increase by a factor of 1.12 twice, 2 2° =1. Initially (when ¢ = 0) the area of algae
giving an 8-year factor of 1.12 x 1.12 © 1.25. was 1cm’.

The result may be obtained in many ways

The growth is therefore approximately from the rules found in part (a); for example,
exponential.
Ded eee aed
626 SOLUTIONS

1
Som,
(a) 0.5cm 2b=- 5 (b) Similarly,
re
(i) 3 = (s') =?=4
(b) @) 27 == Gi) 2? = Git) r= 1 3
Gite = (16) Oe
1
(a) ge ei (b) S°=1 | 10 You should investigate all the laws and
include cases in which an index is negative
1 1
ke LON oe Taam te) = IC ae ie
eA a and others in which an index is not a whole
number or a simple fraction.
(aa2n 22"
—4 1 a3 1 Exercise 2
(b) Te
a De ee
;
cele SI. cine eae ae 1 (a) 2 (b) 2°) se ee
Pe aiee
¥ 27 =5S
GY BT 1Gx58 =2*x2 Zax (b) a” iodides
1 lade s re
2-3+0
(det) 2 x2? gxiager? 3(a)i (b) 5 © 27 ~~ (d) 125
1 4 (ayy (b) ee) x = ee
(i) 2? x2=-x2=1=2°=2"'1!
a ha 5 (a) 2 b)? (4X
(iii) le ene
Jee gate
(d) 100 = (e) 0.1
= =)
6 3.32

(a), (b), (d) and (e) are equal.

(a) (22)? S222(or 2222) = 2! =

(b) Since (./2)? = 2, 2? is defined as \/2
Since (23)3 = 2, 2} is defined as 6)
Similarly, 2" = ¥/2 and a* = Va

(a) 9 = /9 =3 The family of graphs is as shown. The

(b) 8 = 78 =2 graph of the function cuts the y-axis
where y = K.
(c) 64 = 64 = \/(,/64) = \/8 (or 22)
(b) If y= K x a*, y= K whenx =0
(d) 81°° =9
2 (a) y=1 whent=0 (b) y= K when
t=
(a) (4")" = 4; x45 Xeon 43 (p factors) 3 y

earn
Signet erie +7aye (p terms)

Sih
x

(4°): = (4x4x...x 4)s (p factors)

This time the graph shows exponential decay
1 ug
= 47x 47x...x43
aL
The graph of y = K x (4)* is obtained by
reflecting that of y = K x 2* in the y-axis.
= 4a
Again, K is the value of y when x = 0.
 FUNCTIONS 627

(c) m & 2.899. In practice, there would be

£5000 at the end of the third year.

3 (a) The growth factor over the 10 days is

EEE Ss
10m ce e.
The daily growth factor is (0.248) =I0,87
(b) M = 10 x 0.87°
y=2x (4)!
(c) t (days) On: 3 6 # - iy
Mass (kg) 10 7.57 6.57 4.34 3.77 2.48
5 (a) K =1.5 x 10°, ice. the initial value. M 10 7.57 6.59 4.34 3.77 2.48
(b) In 1700, P = 6.1 x 10° = Ka®*, since
t = 634 (d) After 3 weeks, when t = 21 the mass is
ie. 6.1 x 10° = 1.5 x 10°a* 0.54kg.
A ee (e) Approximately 5 days
1.5

Bee, he
6.1\*
Wh a OF OD
4 If V= Ka’,
E (3) Vi= 15 when t= 0 => K=15
(c) In 1990, t = 924
P = Ka’ = 1.5 x 10° x 1.002227" Growth factor a = (£)3Bk 2 0.93

= 11.59 x 10° V=15 x (£)# ~ 15 x 0.93!

(d) The annual growth factor since 1700 has
been considerably greater than 1.002 22,
since the population of the United
Kingdom is approximately 55 million.
3.3.4 Logarithms

3.3 Exercise 3 33°C

1 (a) There will be 14400 bacteria after 2 hours. 1 (a) 2° = 64= log,
64 =6
(b) There will be 1000000 bacteria after 4-5 (b) 2° =4=> log)? = -3
hours.
(c) 2'=2=> log,2=1
(c) Since the growth factor is constant, a
growth function can be used. (d) 2 = /2 > log, /2=1
Number of bacteria = 400 x 6°
2 (a) 3? =9>log,s9=2 = (b) 3
Z (a) 1:08 (p=?) PdeOn wie) 3
(b) The value of the investment after 1 years (f) 4, since 1/3 = 34
is £40001.08”. When the value is £5000,
(g) 4,since2=4 (h) 1
5000 = 4000 x 1.08”

Dividing by 4000, (i) cannot be found since 3” is always

positive.
125° 1.08”
628 SOLUTIONS
eee ee

| 3 (a) a =1=>
log, 1=0 logio 4= 0.6020 logy 5 = 0.6990

(b) a =a=>log,a=1 logy = 0.7781 logig 8 = 0.9030

1 1
(c) a =—=>log,-=-1 logy 9 = 0.9542
a a
(d) log, a” = 2

aeayiloegl0 (=a (binge = 37 Exercise 4

This is to be expected, since logy x and 10° 1 (a) log,9 = 2

are inverse functions. In each case 3.7 is
(c) logys4=—-2 (d) logs2=—3
operated on by the function and its inverse
and so is its own image. (e) logy7 9 = 3

| 5 (a) (i) los 8=3 (ii) log, 16 =4

(iii) log, 128 = 7
(b) 8 x 16
= 128 becomes 2? x 2* = 2’ so 3 (a) By the laws of logarithms,
b= eb S4) 6=7 ancl G4 =e log; 9 + log; 27 — log; 81
(c) Since a = log, 8, b = log, 16, c = log, 128 2) & U7
it follows that log, 8 + logy 16 = log, 128. = logs ( 31 )= tog3 = 1

| 6 As with question 5, 2 +3 = 5, but 5

| 2=log;9,3 = log; 27 and (b) logs 15 — log; 3= log; (=) = log; 5 =a
5 = log; 243 = log; (9 x 27) so
log; 9 + log; 27 = log; (9 x 27). (c) 2logy V7 = log; \/7+ log; /7
= logy 4/7 X../7/) = log,7 = 1
7 (b) (i) logy3 = 0.4771
(ii) logy) 5 = 0.6990 y; |
(c) logo 15 = logy (3 x S) 0810 3x
= Josip loz = 047712 0.6990 logi9 2x
= 1.1761
logio x

fo) logo 9 = 0.954, logy) 8 = 0.903,

logyo 72 = 1.857 = 0.954 + 0.903
l
| 9 log, m+ log,— = log,m x a = log,!
7 m m

=> log, ! — log, m = log, ze

m
1 The graphs of log;) 2x and logy) 3x can be
: 10 logio 2 = logio 1—- logio 2 = 0 — 0.3010
considered either as scalings of logy) x by
= —0.3010 factors of 343 from the y-axis or as
3 translations in the y direction.
logio Se logio 5 = logio 3- logio 2D
By the laws of logs,
= 0.4771 — 0.3010 = 0.1761
logi9 2x = logy 2 + logip x
10 10
0810 2 iy 0810 —=1
4 0810 ——
x2 = logig x + 0.3010
= logy 10 — logyg 2 — logy 2 logig 3x = logi9 3 + logy x
el 025010 07398 = logy x+ 0.4771
 FUNCTIONS 629

5 (a) log 1.05 = 0.0212, log 1.267 = 0.1028 (iii) logyp 125 = 2.0969, logy)5= 0.6990,
log (1.05 x 1.267) = 0.0212 + 0.1028 logy 125 =X logy 5

= 0.1240 (b) The results above suggest that

Using the table in reverse, logig m? = p log, m.

log 1.330 = 0.1240,

| 2 (a) log, m* = log, (m x m)
somnOSix< 1°.267.— 12330
= log, m+ log, m = 2log,m
(b) logy 10.5 = logy (10 x 1.05)
(b) For any positive index,
= logy 10 + logio 1.05

= 1+ 0.0212 =1.0212 log,Sa m’ =log,

Sa .
mXmxX---xXm ,

p factors
logy 1267 = logyo (10° x 1.267)
= log,m+log,m+-:-+log,m
= 3+ 0.1028 = 3.1028 a
p terms
logy (10.5 x 1267) = 1.0212 + 3.1028
= 4.1240
= plog,m
Now, since log 1.330 = 0.1240 | 3 log2* = xlog2
4.1240 = 4 + log 1.330 If 2” =.7, then log 2” =log7
= log 10* + log 1.330 = log 13 300 => xlog2 =7
10.5 x 1267 = 13300 log7
= eS ee Sy
log2
6 logs 5! = logs (5 x 4x 3 x 2x 1)
_ 4 (a) 1% interest represents a monthly growth
lI logs
5 + logs (4 x 3 x 2 x 1).
factor of 1.01. After 7 months the amount
= logs 5 + log; 4! in the account will be 1000 x 1.01”. There
will be £2000 in the account when
e976
1000 x 1.01” = 2000. Dividing by 1000,
= 2.9746
Ogee)

7 t= log, 1000 (b) Glee")

512, 2°
= 1024 . 9<14<10 log (1.01") = log2
t= 9.97 mlog 1.01 = log2

log2
m= = 69.66
log 1.01
3.3.5 The equation @ = b The amount will be a little over £2000
(£2006.76) after 70 months.

3 D 1s 2= 4.98
1 fa) (i) log949-= 1.6902, logyy7= 0.8451, The half-life is 5.0 days.
logig 49=2 logio #

(ii) logy 64 = 1.806, logy 2 = 0.301,

logio 64 =6 x logio D)
630 SOLUTIONS
nn ne rE ESSERE

3.3 Exercise 5 3.3.6 Using logarithms in

1 (a) x=5 (b) x = 2.5 (c) x = 2.67
experimental work
(ie — 210 (Cie 1238 (i) ee ll

2 (a) 2x2x2x2x2=3252
=32 3.3 Exercise 6
13
(b) 9x9 xe = 243
397x% = 243 > 9° = 243
2 O=65
x 0.35°
(c) 8x 8x2x2 =256
> §8x2?=256
3 P = 3.2,/L;.8.7 should) be 7.8.
=> 8? x (s')” 156 (There may be some variation in answers to
questions 2 and 3, depending on the graphs
BONS 8! = 256 drawn.)
=> Bi 256
4 13.2
3 (a) The number of bacteria after ¢ hours is
given by 250 x 3.7’.
(b) There will be 10000 bacteria when
250 x 3.7 = 10000
= 3.7' = 40
34 The number e
— eo 3.4.1 &
i.e. after 2 hours 49 minutes.

4 The time when there is i of the original charge 34 A

is given by the solution of (a) Vie

0:9) =02
dy
i.e. £ = 15.28 (seconds) dx

5 The equation used was 1.08” = 1.25.

E,%
n = 2.90 (years).
dy 0.7 x 2
(b) aa
6 1000 = 470 x 1.029’
1000
=> 1.029% = —— = 2.128
470
K log 212870 64
» logd.029 7 ~
i.e. the population will reach one thousand
million in 2006.

7 They will be equal t years after 1980, where dy

— 2 1.1 x 3%
dx -
ne log 995 — log 470 31
~ Jog 1.029 — log 1.014 —
 FUNCTIONS 631

79.001 x 9 cas 2* s pale 2: 1)2*

(c)
0.001 i 0.001
0.693 x 2*
(d) You could zoom in even more so that the
difference in x between P and Q is even
less than 0.001. For example,
79.000 01 ay
0.00001 Shs 0.6931 x 2*

50:00001 4
0.00001 x 5 1.6094 x $
6 Se. ae eS eh x ~ ; x

3.4 Exercise 1
1 (a) (i) 20.09 (ii) 164.0 (iii) 0.1353
(iv) 0.249 (v) 1.649
dy
—= 2.3 x 10 re
(b) For many calculators it is 230.2 (to 4 s.f.)
because e%?3 > 101”,
3
2 (a)
d |x
gi ra )

ie) 0.4 x 1.5*

2 OVX Ze
3 Dl a3
10 2 S10”

The value of a will lie somewhere between

2 and 3, probably around 2.7.

4 Since multiplication by k represents a scaling

by a factor of k in the y-direction, the
gradient will also be multiplied by k.

d
Gy (ke Jic=ake

5 (a) The x-coordinate of Q is x + 0.001 4 (a) : (i) Whent#=0; y= 5 (ii) 0.677

=> the y-coordinate of Q is
(iii) ¥

S)
(b) The gradient is

change in y _ papel ay >

change inx 0.001 t

The graph exhibits exponential decay.

(b) Since 0.5 > 0.2, it decays more rapidly.
632 SOLUTIONS
NN eee

d 3 (a) 1.61 (b) 2.08 (c) 0.69

5 y=4e! > = 4c = y

The rate of growth is equal to the size of the

colony. It is growing at a rate equal to its size, 4 (a) =(
d 0.69x
)=pecs
a, 2")
Nw
= 0.69 x2 2
i.e. at 500 bacteria per hour. =0.69 ~ en

6E y=e&" (b) As in (a), oa(e*) = Se?

dy ae eth aes

dx 4-0 h
(c) It multiplies the gradient of the graph by

(d) 2g, from part (c)

(e) The gradient is e*, which at R is ene

2 Therefore the gradient at Q is 2 x e4.

f) ae =2xe™*

B Avie 3.4 C

(a) Je
ax
ake = ae
1 ax
+ constant
3.4 B
1 (b) (i) a=1.46
(iii) a = 0.63
= (ii) a= 1.77
(6) () (lax) = aaax)
The graph of f(ax) is mapped onto the
2elaGea 06 + y=e" graph of f(x) by a one-way stretch,
factor —, from the y-axis. This squashit
a
transformation increases the gradient k
a factor of a.

6 (ii) |elax) dx *f(ax) + constant

(ii) a<0 Ya

3.4 Exercise 2
ft (a) 4" (by ea

_— ie (c) ad ca)
Oe
= d
a SEV cas
y=se Vapnoks

d NGS sete a ae
If a=0, y= e° = 1. The graph is a (d) als) ple ) = —3e
straight line.
1
(b) If a> b > 0 then e™ lies above e” for (e) 200% — (f) e* =
positive values of x and below e* for
negative values.
 FUNCTIONS 633

(a) e* (b) — te (c) 1, The maximum level of the drug in the body
ear a se approaches 16 mg, which is approximately 1.6
ige— se (e) ge (Bye = times the administered dose. (In fact, you can
o 5 £ ies 10e os
(g) 2e (h) — re show that it is sae 15.82 mg.)

The constant of integration has been omitted 5

in all cases. t 0 2) 4 6 8
ais seul’ acrieanish a tom UZ6e 5.9 apt

, |
>

(c) y

2 Thus, with a single booster dose, the required

4 . level is reached immediately and subsequent
doses of 10 mg give a stable level of 16 mg.
(a), (b)

Se ae gape cp OS 3.4.3 The natural log

3.4
(b) 3.7 becomes 13.7 mg. ‘¢ y fe
1a) 1G) eines 1)
(c) (ii) Ine=Ine' =1
(iii) Ine? =2
1
t 0 2 a 6 8 (iv) In-=Ine7! = -1
ie 57d O78 a 265) a0 4
=5
(v) In==Ine~ =—5
€

(vi) In(—1) = x > e&* = —1, which has no

solution. That is, In (—1) is not defined.
634 SOLUTIONS
eee
Oe

5 90% represents 180 individuals.

180
200
(c) From the triangle / y the gradient of Pie. =
a = 180(1 + 199e°*) = 200
y =e at P is e From the triangle
a = 199e7°* = 0.1111
B
Oe the gradient of y = Inx at Q is (ce ° 4 ='5.58 > 100:

@ With x and y interchanged, the gradient = —0.2t =In(5.58 x 107%)

at Q is the reciprocal of the gradient at P. =St= ~sin (5.58 x 10~*) = 37.5 days
1
(d) Using a result in (c), the gradient is me 6 T

100
1 1
(e) The gradient of y = Inx at Q is ae

The x-coordinate of Q is b and so

ai (ay) .
dx Se f

t = 9.8 minutes
3.4 Exercise 3 7 (a) Te moe 35%
1 (a) 1.25 (b) —1.05 (@) 1k)
bs ee SK =1

2 (a) In3+1n4=In(3
x 4) = In12 = —9570K =In}
(b) In 10 — In2 = In? = In => K = qin} = 1.24 x 107
(c) 3InS =InS?
= In 125 (b) am = moe**

In 20 = Orem 4
(d) In4 # In5 since this does not correspond
= f= = Dee 0.9 = 847 years
to one of the laws of logs.

3 (a) Inx® =3lnx

(b) In4x
= 1n44+ Inx
3.5 Radians
(c) In¢x = Inx —1n3 3,5:1 Rates of change
4
aa) 3.5 A
1 (a) (i)
3
(b) ae (In x?) = = (3Inx) =~
6 10 8) 2 i 0.1
sin 6°
d d 1 0.01736 0.017 34 0.01745 0.01745 0.0174:
(c) q, (in 4x) a (In 4 + In x) =
x

(ii) The sequence tends to a value of

0.017 45 rounded to five decimal place:
 FUNCTIONS 635

2 (a) So, if x is measured in radians,

0 10 5 mL, 1 0.1
— (cosx) = —sinx
(i) BC 0.1736 0.0872 0.0349 0.01745 0.001745 dx
(il) arc BA 0.1745 0.0873 0.0349 0.01745 0.001745
If x is in degrees, then

(b) The lengths of BC and arc BA become

closer in value as @ gets smaller. Both tend
nc
api COS X* = eae
130°"

to 0.01745 x 6, the same limit as question 1.

3 Length BC = sin @°

OAS 6 ee70 3.5.2 Radian measure

a 360 IED
Since the lengths are approximately equal for
small values of 6, eae
nd?
Sin
70 am sin@®
YS SS SS
ot 1 (a) 0.841 (b) 0.017 (c) 0.996
180 0 180
(d) 0.284 (e) 1.000 (f) 0.014

4 = = 0.017 45, rounded to five decimal places. 2 (a) sin30°\= sin(2x)° = 0.5

(b)
5 The gradient of sin 6° at the origin is =

d ; Radians 7 ius in in a 30 2
6 j— = sini
gj e => —46-180
y = aes OF Degrees’ 180) 905 160) 5455 30.270) 360

T
(c) To convert 6° to radians, multiply by
180°
Few5
:
B
180
(a) Sketching the gradient graph of y = cos x°® (d) To convert 6° to degrees, multiply by —.
2 T
ed
gives the graph of = hsin x” 3 (a) 1.047 (b) 0.866 (c) 0.018
dx

4 sin 60° = 0.866. In radian mode, this will

probably give an error on a calculator. In fact,
sin 60° = —0.305.

5 Plotting a graph of y = sinx with x in

radians, using simple numerical values rather
than multiples of 7, helps to emphasise that
multiples of 7 are not the only way to express
angles measured in radians.

(b) Working in radians, you can make the

maximum gradient (i.e. the value of k) equal 3.5 Exercise 1
tOnle-S0;

d ;
y= cosx => Bre cinx
dx
636 SOLUTIONS
SD
ee ee a

2 (a) 45° (b) 540° (c) —180° (a) y = sinx is mapped onto y = Ssinx by a
stretch of factor 5 in the y-direction.
(d) 270° (e) —360°
(b) The gradient is multiplied by a factor of S.
8) Pig
(c) y=Ssinx> I= N5cos x
2m
dx
4 (a) 20 (b) 2 (c) —
WwW

ine LAY
3.5.3 Area and

Exercise2
arc lengths
ES(a) y = sinx is mappe’ onto y = sin3x bya
stretch of factor4
3 in the x-direction.
LQ) x 2dr ata
(b) The gradient is multiplied by a factor of 3.
(b) 2x 4a=$30 (c) 4457
(Cy sito
dy COS OY
2 The ee - CDE =r. dx
Area ee 3 y = Ssin3x is obtained from y = sinx by
Sonia 1? ? ine i ou(cince a= ()r applying both stretches from the previous
questions.
3 (a) 1600m dy
= sine 15 cos 3x
(b) Area = 160000m? dx
The largest crowd is 80000.
4 (a) —2sin2x (b) 20 cos 2x
4 (a) BC=rsind (b) 4rsind = (c) 1r8 (c) 0.5 cos 0.5x
(d) Area of segment = area of sector OAB
— area of triangle OAB 77¥ = 3008 2x +4

=1/76- r sind

=17/(6 — sin 6)

Nn 104.7cm’, 43.3cm*, 61.4cm?

(b) These curves are mapped onto each other

by translations in the y-direction.
3.5.4 More about deriv atives The derivatives are all the same:

a
dy
ty
AD
we
(a) 1@) asin = acosee
dx
e : d
(ii) y = sinbx Seon
dx
(Ci) Ay = wisn aoe Ss dy = ab cos bx
dx
d
(by aicosibaa— = oe —ab sin bx
dx
 FUNCTIONS 637

3.5 Exercise 3 This represents the motion between the

maximum and minimum positions. The
1 (a) $cos x (b) —Ssinx
velocity is momentarily zero at the extreme
(c) 0.1 cosx (d) 4cos 4x positions, and the speed is greatest at the
midway position.
(e) —2msin 21x (£) 0.2. cos 0.2x
dh :
Daa 10:8 cos3 nt + 6.5 > —=
dt —2nsinint
5 6
(g) —6sin 2x (h) 37 cos iTx
ha
(i) cos ix
73

d
D(a) Sagi —2sin2x \L\f
S/S
dx =s
T T >
12 24 t

(b) |sin2x ake = —4Fcos 2x

dh
dt
2
(c) Since the derivative of sin 3x is 3 cos 3x, sian

|cos 3x dx = isin 3x. >

12 24 ‘3

-2-74
7

3.5.5 Applications (b) 0.8 cosiat + 6.5 =6

=> cosint= —0.625
3.5 Exercise 4 => t= 4.29,7.71 (0417 hours and
dL : 0743 hours)
1 (a) L=12+ 2.5 cos2at > =— = —Srsin2at
dx h
Hs
When t = 4.29, “ = —0.33 (mh);
14.5
12 when t = 7.71, 2 = 0.33 (mh“')
9.5
The rates of change are numerically the
T=
0.5
pease
Wie
same but opposite in sign because in one
case the tide is falling and in the other it
is rising.

(c) The tide is falling most rapidly at t = 3.

dh 4
— >
Fp = 70-42 (mb)
t

(d) Tidal current is greatest when the tide is

rising or falling most rapidly and is least
(b) near high and low tides.

_ Depth of water and strength of tidal

t : 0 Os 10:255 5 50:4810:5:
current are the two important factors in
(i) Length (cm) 14.5 140 12 10.0 9.5 deciding when it is safe to enter or leave
(ii) Velocity (cms~') Ol = 9255 920 harbour.
638 SOLUTIONS

dh h
3 (a) Fp = 0: 16m sin 35
q me @r a) a — —0.76 sini at

10.4

5.6

0.8
120 ¢
(ii) As indicated above, this is directly
proportional to the current,
dh demonstrating the direct relationship
dt between the speed of the current and
0.1677
the rate of rise and fall of the tide.

—0.167 Transformations
3.6.1 Graph sketching
) 0.167 sin=a is
> t = 38.8, ea as TeGe?
3.6 A
The speed is over 0.4ms_! between
t = 38.8 and t = 51.2 and again between 1 (a) Replacing x with x + k results in the
t = 98.8 and t = 111.2. The chair descends
most rapidly at t = 45; speed = 0.5ms_! t ransilation
lati a0 6

4 (a) (i) (i) y=f@etkY

—4(x +h)
46 (ii) ysin (x +k)
(iii) y = e* t*
17 (iv) ys (l= (x + k))
<6 =? 3 6 (b) Replacing y with y + results in the
T T | ii T

(ii) b = 3.15 + 1.45 cos rt translat 1 on .

(b) (i) Current (knots) i) yrR= aa — 4x

el
=> yaar —4x—k
>
(ii) y+ kR=sinx
> y=sinx—k
6 Number of hours (iii) ytkR=e’ >y=e—k
before high water
(iv) yt kR=+/(1
—x’)
=>y=-k+.,/(1—x*)
Since the period is 12 hours,
aS
w= prt
sb24i |
T
(c) Replacing x with kx results in a one-way
1
(a= — Tel sind at stretch from the y-axis with scale factor zk

(111) Since current represents the rate at (i) y= au —4kx (ii) y= sinkx
which water is entering or leaving the (iii) y= (iv) y >/(1 — (kx)?)
harbour, this is proportional to the
rate of change of the height of the tide.
 FUNCTIONS 639

(d) Replacing y with ky results in a one-way (b) The transformations are identical to those
eer 1 described in (a)(i), (ii) and (iii).
stretch from the x-axis with scale factor —Ee
2 2
x 6 4x
i oe 3.6 B
: sinx The equation would become y = ee
ky =scinx
> =
k instead of y = 2e**!.
en

k = —s =—
Jie. y k
y= 23 +1)

ky=+/(1-x)>y=+ (t=k 2)
2 (a) Replacing x with —x reflects the graph in
the y-axis.
(b) Replacing y with —y reflects the graph in
the x-axis.
(c) Interchanging x and y reflects the graph in
the line sy x:

| 3 (a) (i) Translation fe 3.6.2 Stretching a circle

3.6 C
1 If you transform the unit circle, x7 + y* = 1,
with a stretch factor r from the y-axis and a
stretch factor 7 from the x-axis then you
obtain the circle with radius r shown in the
(ii) One-way stretch from y-axis of scale diagram on page 275.
factor 3, followed by a translation

a]
x : :
Replacing x with — and y with ” in the
r r
equation x + y = 1 gives
ag!
CS jot 2 2 2
Soap LSS ay =r
pip :
2 Transforming the unit circle, x7 + y” = 1, by a
stretch factor a along the x-axis and by a
stretch factor b along the y-axis gives
—1
(iii) Translation a followed by a one- tw 2
+ Ae
way stretch from the x-axis of scale S|
R mat
factor 2
3 Giving the circle in question 1 a translation

| by replacing x with x — p and y with

q
y — q, you obtain

(x—p) +(y-a) =r
RW
640 SOLUTIONS
eS

3.6 Exercise 1 Miscellaneous exercise 3

1 The equation y = Inx can be transformed 3 . 3 :
onto y = In3x by replacing x with 3x. The 1 (a) (i) ee (ii) ae (b) x =%
graph of y = Inx can therefore be fitted to the
graph of y = In3x by a one-way stretch from
= Otactor dt De £17, NTs 425,55
ae 6x,
eon 13-6
x 3
2 1(5 — x), 4(7 — 3x), }(7 — 2x)
The equation y = In3 can be rewritten
v= Inse—y— ls ne efi ear (pees
i. 8
=>y—In3=Inx
5 1
The equation y = Inx can be transformed (c) aap (x #3)
: ; 4x —1
onto y = In 3x by replacing y with y — In3.
The graph of y = Inx can be fitted to the (d) /(4—x) (x <4)

graph h of of y y=
= In3x
In3x by
b a translation
lati ||,
: |. ia) BO)

Day 40 Sy — (2x) (b) y=S+V/(x+1), (24,10)

The graph of y = x’ can be fitted to the (c) y= V(—x), (—25,5)
sae ee one-way stretch igo eee hoses,

. een aes eli Say 1125)

The graph of y = x’ can be fitted to the nt:
graph of y = 4x” by a one-way stretch 5 x(x +3)(x +4)
from y = 0, factor 4.
6(b)ege y= sicos4x~
3 (a) A one-way stretch from x = 0, factor 2,

followed by a translation 3 3 7 (a) 2—2sin1°

Replace x with }x, then replace x with 8 (a) 162 (b) -—2 (¢) 5.76 (d) =o
Ge De
Phy 1 > (bey? yee OF (a)e3ra.im: (b) 9.5m and 1.5m

(c) The graph oscillates about = 5.5 with

Sr Le Dey
ere ata cat period 12h and amplitude 4.
1 =i]
(b) A translation 3 followed by a one-way id) 1.05 mh
stretch from x = 0, factor 2. 10 (a) 90 (b) 45,135 = (c) 18, 54, 90, 126, 162
Replace x with x — 1, then replace x (d) 120 (e) 40, 80, 160
with 3x.
e+y¥=13(x-1)4+/=1 11 (a) 20cm (b) 10cm (c) 2s
+ @x-12+y=1 (d) y = 20+ 10 cos 360¢°
The two equations are equivalent. You are, (e) y = 20 + 10 cos 1202°
however, less likely to make a mistake with
the translation if you stretch the circle before 12 (a) 2, 360 (b) 3, 360 (c) 0.5, 180
you translate
translate 1it.
(d) 5, 1080 (e) 3, 540
 FUNCTIONS 641

13 Many possible answers, for example

(a) y = sin 4x° (b) y =$sin$x°

(oy = Ssin 6x~ (d) y= $sin5x°

14 (a) 1.262 (b) 1.316 (c) 1.262

tS) P = 100 x (0.98)"

(a) 38.4% (b) 34.3 hours

16 y=kx?; plotting logy against log x, the

gradient gives p and k is given by 10° where
c is the intercept with the vertical axis.

17 6 = 83 x (0.96)"

20 A = 20.0

21 (b) mp = 10
(c) (i) 14.92¢ (ii) 22.26
(d) (i) 49.2% increase (ii) 49.2% increase

22 (a) 0.18 kg (b) 3.47h

(c) (i) 6.93h (ii) 10.4h

dm

23 (b) 2ms-* — (c) 33m

-270 270

(b) y= (x - nN
1)°(2—
x)
-360 360
642

4 PROBLEM Under the second system there can be at mos;

one bye per round. You could try to find a

SOLVING more precise rule.

Differences formed from numbers with an od

4.1 Mathematical enquiries number of digits are divisible by 99.
4.1.1 Introduction Differences from numbers with an even
number of digits are divisible by 9.

By looking at particular examples such as

41 A
1 : 365 = (3 x 100) + (6x 10) +5
First Second
organiser organiser you can see the general result that a number
(a) byes 5 2; looking like ABC actually has the value
(b) rounds 4 4 100A + 10B + C. So the difference between
(c) matches 10 10 this number and the one with digits reversed.

100A + 10B + C — (100C + 10B + A)

The type of notation used can be an
important aid in solving the problem. The
= 99(A —C)
first organiser’s method of running the
(You may have managed to give an
tournament could be illustrated like this.
explanation for these results without using
algebra.)

One fruitful line of further investigation

concerns using ‘differencing’ to form chains o
numbers. All chains from non-repeating two-
digit numbers end in the cycle:
63 ——> 27

ene
vi \

For example, 63 — 36 = 27, so 63 — 27, and

=. so on.

What happens with three-digit numbers, four

Are there any better ways of recording the digit numbers, ... ?
investigation?

The number of matches played is always one

less than the number of competitors. An
amusingly brief proof of this can be given —
everybody loses once except the winning
finalist!

The number of byes is zero only if there are

2” players.
A tessellation originally was the result of
For P players under the first system, the number covering an area with tesserae, the small
of rounds is given by n where 2"~! < P < 2". square blocks used by the Romans to make
The number of byes is then 2” — P. mosaics. Nowadays, from the primary school
 PROBLEM SOLVING 643

onward, it is generally accepted as any Further extensions might include tackling

covering or ‘tiling’ of a plane surface of questions such_as the following:
indefinite extent by a regular pattern of one or
more congruent non-overlapping shapes. For
example:

(a) A parallelogram tessellation can be

obtained from one for rectangles by a
shear: Do all pentagons with one pair of parallel
sides tessellate?

Are there any other pentagons which will

tessellate?

For extensionsit might be more useful to eeeSo

think of the parallelograms forming strips
which can then be fitted together easily. 5)
TE
(SYETT Sia Quadrilateral B has been obtained from
quadrilateral A by rotating it through 180°
about O, the midpoint of a side. Repeated
Ss Perhaps the most obvious method is to fit
copies of the triangle together in pairs to applications of this principle give a
form parallelograms tessellation of quadrilaterals.

As an extension you could check that the

method works for re-entrant
quadrilaterals, such as

but finding other patterns might be a

profitable exercise.

The 108° corners cannot be fitted together

to form 360° and so regular pentagons
cannot tessellate. and explain why the method always
works.

4.1 B
1 If a =0 then x is the cube root of —b.

If b = 0 then
xe +ax=0

This idea might be used to establish the => x(x? +a) =0

fact that only three regular shapes will
tessellate on a plane surface — an = x= 0. *or x +a=0
equilateral triangle, a square and a regular
hexagon. and further roots can be found only if a < 0.
644 SOLUTIONS

| 2 The right-angled triangle might be a useful (d) The Sth, 10th and 15th terms are all
| first example, since such a triangle is half a multiples of 5. Generally, every nth term
rectangle and since any triangle may be is divisible by u,.
dissected into two right-angled triangles.

|3 x =0 (y is not defined) Al Exercise 1

n

x small and positive (y is large and positive) 1 (a) S (Qi-1)=e

x small and negative (y is large and negative) i=1
x large (y is approximately equal to x) (b) The sum of the first 2 odd numbers is n.

| 4 Two such examples are: 2 (a) Divisible by 3; divisible by 8.

@) Gs BHA G 4,d 336=
6, | =v (b) The sum of 1 consecutive odd numbers is
(b)a il, D=2ye 4,d 20 =6,7/7 =i divisible by 1 if 1 is odd, and by 2n if n is
even.
Write out the equations with these coefficients
to see why. You will also see that there is an 3 2, 5, 9,4n(n — 3)
infinity of possible answers.
4 A polygon of given perimeter has greatest area
if it is regular, i.e. all its angles are equal and
all its sides are equal.
41C
| 1 (a) The numbers in the sequence are
25,121, 361 and these are all perfect
squares bee nee 197). 4.1.2 Organising your work
Further terms give 29 and 41”.
4,1 D
If the consecutive digits start at 1, then the
general term will be _ 1 Labelling the points is a useful first step.
B
[n(n + 1)(n + 2)(n+3)] +1

If you expand and factorise, looking for a

perfect square, the expression becomes
(n? + 3n+1)*. You should confirm this. A E

43 681 = 209. So, assuming that d is the

greatest of a, b, c and d, d = (n+ 3) You would need to say which pairs of points
where are linked, for example A—B, A—C, and so on.
You might also give distances and directions
nw +3n+1
= 209 between points or, alternatively, you might give
the approximate coordinates of each point.
The numbers are 13, 14, 15 and 16.
| 2 RSLSSL
2 The recurrence relation is
The route is shown in this grid.
Un+2 = Uny1 + Uy
(a) 21, 34
(by 87413. <5) — fi, oe = (O08 ea
and so on.

(c) The 4th, 8th and 12th terms are 3, 21 and

144. They are all multiples of 3. Finish
 PROBLEM SOLVING 645

3 A simple pictorial notation might be LJ for 5 Is it shaded? Is it made up of straight lines

(rectilinear)? Has it more than two lines of
an upright tumbler, and [] for one which is symmetry?
upside down.
However, the explanation hinges on ideas of
oddness and evenness (parity) and so a
numerical notation is better. 4.1.3 Tabulation
Let 1 represent an upright tumbler and 0 one
which is upside down. Start with a sum
1+ 0+ 1 = 2, which is even. Because you
change two at a time the sum will remain 1 As an example, you might take
even, i.e. always be 0 or 2, and never 3 as E(n) =n? —n4+2
required.
Then E(1) = 2; E(2)=4+-£(6)= 8; and soon.

4.1 Exercise 2 ele ang ee Pils T 22

2 a: 4 6 8
1 (a) Acute, right, obtuse, straight, reflex, ... 2 re|
& ve
im
(b) Scalene, isosceles, equilateral,
Whichever quadratic expression was chosen
right-angled, ...
the second differences should be equal.
(c) Prisms, shapes having one curved surface,
shapes having a plane of symmetry, ... You may have realised that there was an
unnecessary restriction in this question.
Using the notation for cells as shown: a, b and c can be any numbers but, of
course, it simplifies calculations if they are
ORAS integers.

4 [5 [is The third differences should be equal.

7 NST 9
Sequencem —1y —6" 33 168) 2555 654 1379
first moves in cells 1, 3, 7 and 9 are essentially
First
alike, as are those in cells 2, 4, 6 and 8. There
differences 5), 9 65) 1S SIO 72S
are therefore three essentially different first
Second
moves.
differences 147 156) 1227 212s 26 ee
Third
Rectangles and rhombi
differences 42 66 90 114...
Fourth
The possible cases are:
differences 24 24 = 24...
e aand b both even, then a” +b’ is a
multiple of 4; On the basis of the pattern established in
@ one odd, one‘even, then a + 6? = 4k +1 questions 1 and 2, you should conjecture that
(see below); the sequence was generated by a quartic
e aand b both odd, then a + b* = 4k +2. expression.
If, for example,
(a) First differences are all 1.
a = (2m +1) and b = 2n,then
(b) Second differences are all 2.
a? +b = 4m? +.4m +14 4n?
(c) Third differences are all 6.
=4(m? +m+n)+1
(d) Fourth differences are all 24 (see next
= 4k + 1, say page).
646 SOLUTIONS
=o

Sequence 5(n a 1) if m is odd

(n*) da) 16) eB1e 0256 06250111296) 2401.2. r(n) = Int1 — ifmiseven
First E
differences 15 65 175 369 671 1105... s(n) =s(n—1) +s(n—2) forn> 3
aa with s(1)=1, s(2) =2
differences SOLON 194 5 3020 434e ie
Third
differences 60) 845) 108 132/55:
Fourth
differences 24 24 24... 4) P root

From the answers to question 4, a 4.2.1 Introduction

generalisation is that n* generates a sequence
giving kth differences all k! (and hence ’
(k + 1)th differences all 0). So a polynomial 4.2 Exercise 1
with leading term n° would generate a 1 x =—1, y = 0 disproves the statement.
sequence with fifth differences all 5! = 120.
UNG IESey SOU lcs UES ees ie 2 False. The three-dimensional Cartesian axes
sequence was generated by a polynomial of
; are a counter-example.
fifth degree, leading term In. P

3 False. 21 is the first counter-example.

4.1 Exercise 3
44, = 10 — 2s i — 4 2 ine
value of t; is not 5 as might be expected if yor
ln 3) 4 8 considered only the first four terms.
D(n) Cee ea) She 9)

D(n) is the number of diagonals of an n-gon.

A decagon has 35 diagonals.
4.2.2 iMaking% aa pre
proof

4.2. Exercise 2
2 Shape No. of lines Order of ; :
Biv mecin aeoeonel 1 (a) 2n is 2x n and is therefore even because i
symmetry is divisible by 2.

Parallelogram 0 D) (b) 22+ 1. This number is 2 x n+ 1 and is

Rhombus D) 2 therefore odd because it has a remainder
Rectangle 2 2 of 1 when divided by 2.
Kite 1 1
S (c) (2n+ 1) + (2n + 3) + (2n+ 5) = 6n4+9
quare 4 4 Per
bite aseee babe ae en = 3 x (2n + 3) is divisible by 3.

2 (a) 10b+a
By ib 1 2, 3 4 5 6 b lla =
+11b6=11x (a+b
r(n) 1 2 2 3 3) “ wo tie
s(n) 1 2 3 5 8 13 (c) The rule does not work for three-digit
numbers, for example 102 + 201 = 303.

The rule does work for four-digit

numbers.
 PROBLEM SOLVING 647
—_—..?.?}.kTF —v—————K—KKXxyPprooooo

(10004 + 100b + 10c + d) 6E There are a number of ways in which this

proof can be made. The following is based
+ (1000d + 100c + 10b + a) on the classic method attributed to Euclid,
= 1001a + 110b + 110c + 1001d which employs contradiction.
= 11 x (91a + 10b + 10c + 91d) Suppose /2= ~ where a and b have no
common factors.
3 (b) (10a +b) x 11 = 110a + 11b Vi=F
3 Wad
= 100a+ 10a+ 10b+b6
= qa’ is even (has a factor a 2)
= 100a+10(a+b)+b
=> ais even
So ‘ab’ x 11 =[a][a+][b]
Let_ a=2m
(c) For examples of this kind you have to a = 4m
‘carry’ digits in the addition.
2
The method works, but you have to be
Since 5? = es b? =
more careful! 7 Dy

=> b* is even
dh 3923
(2 i
=> b is even
=4e2 iit 2
SANS 2 If /2= -then we have shown that both a
and b are even and consequently have a
4 Inthe 5 x 13 ‘rectangle’ a very thin common factor of 2.
parallelogram of area 1 is missing.
This contradicts our original statement, so
8
a
24#—
v b
5 ~ 5 and so 1/2 is an irrational number.

S 8

A ‘similar triangles’ argument shows

es
this: ei
347/aesy
4.2.3 Fermat and proof

5 (a) The numbers may in general be written as

4.2 A
(2a — 1) and (2a + 1), with sum 4a.
a eee ee
(b) The sum of the numbers may be written as eek
he tees
(2a + 1) + (2a + 3) + (2a +S)
n is the number of points, and r,, is the
+++++(2a+2n-—1)
corresponding number of regions.
= (2a+ 2a+2a+2a+---+2a) Apparently the number of regions doubles
ea
Ses 2 a) every time a point is added, i.e.
fo44=2 x *#,,. in which case
=2na+n

= n(2a+n) (bn = UE =4

This expression is divisible by 1. For n even, it 2 rs; = 16, which agrees with the conjecture.
is divisible by 2n. However, 7¢ = 31, so the conjecture is false.
648 SOLUTIONS

making separate counts for the two classes,

you will find patterns occurring in which
sequences of triangular numbers appear. For
example in the 4 x 4 x 4 grid there are
| 4 First differences 1 2 4 8 15 #426 1+3+6+4 10 pointing upwards
Second differences 1 De ANT 11
Third differences 1 eS 4 and
Fourth differences ileal Sagal 1 +6 pointing downwards.

5 The formula for 7, is probably a quartic For the 6 x 6 x 6 grid, using the patterns
| polynomial. found, there should be
1+3+6+4+10+15 +21 upwards
The polynomial
t, =an'+bn>+cn’>+dn+e and
1+6+15 downwards,
can be found by solving five equations (for the
five unknown coefficients), obtained by giving a total of 78 triangles.
substituting corresponding values of n and r,,.
The first is
at+b+ct+d+e=1

and the second is

16a+ 8b+4c+2d+e=2 4.4 Mathematical articles
The study could be taken further, either as a 4.4.1 Introduction
class or by an individual student.

4.4 Exercise 1

B 4) At this stage you should draw a few doodles

| 1 4; namely 1 x 1,2 x 2,3 x3 and4x4 such that your pen comes back to its startin;

= Bo @s
| squares. point each time. For example:

2 There are

16 1x 1 squares
9 2x 2 squares
4 3x 3 squares.
When the author of the article makes
So the total is 1+4+9+
16 = 30. definitions about such graphs you can then
use your own drawings as illustrations.
_ 3 Using a similar method of counting, the total
number is At this point you would probably read on a
14+44+94.--147 2 = in(n +
few lines to check that the author is going t
1)(2n + 1)
define the phrase ‘4-regular’.

4 There are 27 in all, including 16 of edge 1,

Each of the author’s definitions and
7 of edge 2 and 3 of edge 3.
statements should be checked.

(a) Number of crossing points, n = 9.

| 6 Classifying into triangles pointing upwards (b) Number of edges, m = 18.

(A) and those pointing downwards (v7) and (c) Number of regions, r = 11.
 PROBLEM SOLVING 649
SS

(d) 34589 defines the shaded region.

(e) There is a region exterior to the graph!

(a, b,c) For the doodles given in answer to

question 1,

jo 7 ee
tf | 2 it
— NA eS
ee sa

(d,e) The region exterior to the graph of

figure 1 is defined by 167 892.

*“3 It could be argued that in the doodle

Co) the only crossing point has 2

edges adjoined to it. In fact, mathematicians

would still count this figure-of-eight doodle
as 4-regular. One simple way to prevent the
doodle being 4-regular is to retrace the path
at some point, for example,

QO
*4 To answer this you must be clear about
precisely what is allowable as a ‘doodle’.
In fact, it is a result of graph theory that any
connected 4-regular graph can be drawn
without retracing your path or lifting your
pen off the paper.

(A graph such as CS Oe)

is 4-regular but disconnected.)

*5 The assertions should be checked on the

doodle of figure 1. The values of k for the
k-gons are shown below.

OG
 The Platonic solids
1 (a) A ‘polygon’ is a closed plane figure bounded by
straight edges.

A ‘regular polygon’ is a polygon all of whose edges are

equal and all of whose angles are equal.
CASE
STUDY
4.4
1
(b) There is no limit to the number of different regular polygons that can be

Kh ee
As will be seen, there are actually five Platonic solids. In view of the picture
in the text, the cube is the one most likely to be chosen to answer this
question.

The triangular faces must all be equilateral triangles and so the

angles at the vertex from the six faces would total precisely
360°. The six triangles would then each be part of a single flat
hexagonal face and would not in fact be individual faces.

F=6, V=8, E=12

(a) F+V—E=8 +6— 12 =2(octahedron)

(b)
F+ V—E=6+8-—12 =2(cube)

The shape is formed by F triangular faces with a total of 3F sides and 3F

corners. As before,
3F
ha
y
Each vertex of the solid is at a corner of each of three faces and so
spel
rue)
Then

V+F
Therefore

Vi pha

The solid is a tetrahedron.

As before, there must be at least three faces at each vertex. Four or more faces
would contribute an angle count of at least 4 x 90°, in which case the faces
would not form a corner. There are therefore precisely three faces at each
vertex.
There are F square faces, having 4F sides and 4F corners. When joined
together, each edge is formed from a pair of sides, so
4F
|
Sees i
2
and each vertex is formed from three corners, so
4F
a 3

Each angle of a regular pentagon is 108°.

Sh OF
Es: and Wels,
From Euler’s formula,

Pe
SF SF Sr
Eine”
Multiplying both sides by 6,
6F + 102 = 15 -- 12-> F = 12

Mbhen) WV = a 0 and E=7 = 30

 The gravity model in geography
The motivation for an inverse square model may be worth examining, especially
if Newton’s law is not familiar. You might compare the intensity of gravitational
attraction with the intensity of light.

STUDY
CASE
24.4 P ee

Think of a cine projector P illuminating in turn two screens S and S’. If S is three
times as far from P as S then the area of light on S’ is nine times that on S. But
the same amount of light falls on both S and S’. So the intensity of light on S’ is
only 3 of that on S. In fact, the intensity is inversely proportional to the square of
the distance from P. (Replace ‘cine projector’ by ‘candle’ to give the analogy the
right Newtonian flavour!)

k x 216 x 271 2 Se Soi <A

(a) Tpn = Se eT Se
502
(b) Derby and Nottingham

3
kPs
aoe

ds,
WK = 130a + 173a = 303a

So in mile units,
303¢ — 15

a = 0.043

(a) Y divides WK externally in the ratio 130: 173, i.e.

WE 180
Bac We
Since WY = x and YK = (x + 13)
ae 130
x+13 173
(b) Expressions for the attractiveness of Shrewsbury and Welshpool are
respectively
56 000k 7000k
a and ody

(c) The expressions for attractiveness are equal when

7d = 56dyy
dy = 8dy
dy 2.83dy STUDY.
4.4.CA
2

Welshpool’s area of influence is the circle diameter XY where X, Y divide

SW internally and externally in the ratio 2.83: 1. In mile units,
a al SY.
SW = 19: Wigs <1 46, vay ee

Soult
YW =:
x+19
ae 2.83

19
Fe 0 38
* > 63
XY = 10.38 + 4.96 = 15.3
Welshpool’s area of influence is a circle of diameter 15.3 miles, centred at
a point 2.7 miles from Welshpool on the side opposite Shrewsbury.
The ‘distance decay’ would be less marked, making (3 smaller.

The value @ = 1 is often used in US models.

The game of Hex
1 If Black plays at 7, White plays at 5, and connects 1—2/4-5—8/9.
If Black plays at 3 or 6, White plays at 4 and connects 1-4—7/8.

(b) The claim is false. Black connects 9/13—10—11/14—15—12/16.

If White plays at 7
White 7 can be connected to the top left edge by 7-3/4. It can be connected to
the bottom right edge by Result 1. White therefore cannot be stopped from
completing a winning chain.
If White plays at 10
The fact that White cannot be prevented from completing a winning chain
follows from the result above by symmetry (rotate the board 180°).
Archimedes and 7
An upper bound for the circumference is a value above which the
circumference cannot possibly lie. Similarly, a lower bound is a value below
which the circumference cannot possibly lie.
The actual value lies between the upper and lower bounds. The actual value is
known with greater precision as the lower and upper bounds are brought
closer together — precision improves as the bounds converge.

(a) © OC is a radius of the circle, so has length

1 unit.

DG—=1OCcos30 a— wl cos 30s cos30n

PR is a tangent to the circle, with point of

contact at S. OS is a radius of the circle
and so has length 1 unit.
OS if
5 ie [a R
te sae ee ne
tan30° — tan 30°

Perimeter of AABC = 6 x DC = 6co0s 30° = 3/3

Perimeter of APQR = 6 x SR = 6cot 30° = 6,/3
Thus 3/3 < circumference < 6,/3
Or 3/3 r= 6y3
3/3
or M3 <n <3y3

(a) An inscribed figure is a figure that is totally within the circle but touches
the circle at its vertices.

(b) A circumscribed figure is a figure that is totally outside the circle but such
that its edges form tangents to the circle.

M is the point of contact of the tangent PQ with the circle, so OM is a radius

and has length 1 unit. OA and OB are also radii so have length 1 unit. AOAB
is equilateral since ABCDEF is a regular hexagon, i.e. OA = OB = AB. Hence
AB has length 1 unit. The perimeter of ABCDEF is 6 x AB = 6 units.
 6 (a) In AOLB, OB= 1 unit, since it is a radius. LB = 51 unit, since L is the
mid-point of AB.
Angle OLB = 90°, so by Pythagoras’ theorem,

4
STUDY {aye Wea
ipa ee EN ieee
a)

3
$0 OM = 1and OL =
_ OM
CASE
4.4 The scale factor of the enlargement is OL

2 2
Hence PQ = —- x AB = —

and dth ve offP PQRee

the perimeter = =iV3
Ae a 4y3
Perimeter of ABCDEF < circumference < perimeter of PQRSTU

6 < 2m < 4,/3

Sera 2/3

Let AB be a side of the regular

inscribed n-gon so that
ae

Then LB = —,
5 since L is the mid-point
of AB.
OB has length 1 unit.
Q
By Pythagoras’ theorem in AOLB,

OL=V/0 =Gx,))=V0 =i.)

The scale factor of the enlargement is
OM _ 1
Oh wate)
AB ae oe,
so PO!—
V(i aa 5X,") ic]V(1 ae 5x")

Multiplying by the number of sides n, in order to obtain the perimeter, gives

he
ON Beer
" Va-%
8 (a) xp = V(bx6+ [1 — V(1 — 4x6"))’)
(b) xs = 1 since x¢ is the length
of the side of the inscribed hexagon.
yy = 4/(14 [1-/(1—-1)]*) ~ 0.518
9 Ip = 12x15 DSO 8 ="6.012:

I2
Cy = —— > = 6.658
: V(l 7 4X12")

Eels It, (G. Inequality

IZ 0.518 6.212 6.658 3.106 < m < 3.329

(All calculations are quoted to 3 decimal places, having been evaluated using
x42 = OS 638.)

10 As the number of sides of a regular m-gon increases, the n-gon becomes a

closer and closer approximation to the circle itself. So the inscribed and
circumscribed n-gons both have perimeters that approach the circumference
of the circle, getting arbitrarily close as n increases without limit. By
increasing n, increasingly accurate approximations to 7 are obtained. The
difference between the approximation and 7 itself can be made arbitrarily
small by taking a sufficiently large value of n.
658

5 MATHEMATICAL
METHODS
5.1 The power of Pythagoras
. P ¢ : © Further points which lie on the circle are
5.1.1 Pythagoras and right- (—3,4) and (—5, 0).
a ngled tria ngles (b) From the FO above, the equation ¢
i the circle is x + y~ = 25.

5.1 Exercise 1 3 (a) If the point P lies on the circle, then

1 Calling the hypotenuse x, CQ? + PQ’ = 257.
P(x, y)
(a) x” = 10° + 247 = 676 > x = 26cm
[This is simply a scaled up 5, 12, 13
triangle.]

(b) io —=17-5'cm

All the given points satisfy this equation.

2 Westminster Abbey and the Tower of London
are 3.8 km apart. (b) Since CQ = x —2 and PQ=y—S, the
equation of the circle is
3 The space needed in front of the step is 69 cm. (x — 2) +(y- Sie = 257,

4 Height == IA
4 (b) By Pythagoras’ theorem
Sea) (x,y) (x fi a) ae (y — b)? iss r.

5 By ‘trial and error’ you can find

Sy GPS oy 2 SS ehill eA Ry A) Lil)

6 810 10 24 26
12 155
12 16 20
15 20 25
5.1°B
1 (a) If Q is the foot of the perpendicular from
q , ; P to the x—y plane, then PQ has length 6
5.1.2 The equation of a circle z P(4, 5, 6)

At * A
1 (a) /G> 437) = 18
(b) V((4— 1) + (6—2)’) = y25 =5
2 (a) If a point lies on the circle, its coordinates
will satisfy the equation x* + y* = 25.
Thus, since 4° + 3” = 25, the point (4, 3)
(ee
. oe . S 0, applying Pythagoras to triangle OPQ
’ i i

OP? = PQ? + OQ? = 6 + OQ?

 MATHEMATICAL METHODS 659

You can find OQ by applying Pythagoras 3 By Pythagoras,

to triangle ORQ.
(E — 20)* + (N — 85)? = 18”
OR = 5,RQ=43 0 =544
(E, N)
Thus OP? = 67 4574+ 47 =77
=> OP & 8.77
You can use a similar method to give:
(b) 5 (c) 5 (d) /14
= 3.74

fala 3, 4. (b) 2. (c) 4 The equation fails to take account of varying

levels of visibility.
PQ? = 37 +2? +47 =29
So PQ = 29 & 5.39 4 If (x,y,z) are the coordinates of a point on
the sphere, the (distance) of this point from
3 (a) PQ= 46 © 6.78 the centre (2, 3, 1) is:
(b) PQ = \/138 © 11.75 (x92)? G3) ie)?
(c) PQ V(x —4) +y— by Fe —2) The equation of the sphere is therefore:
4 (a) OP= V(x? +y° +2’) (x — 2)? + (y— 3)? + (<-1P = 4 $16
Since OP = r, the equation of the sphere is
V(xetyte)=r 5 (a) The square of the distance from (3,2) to
(1,4) is:
This is more usually written as
(3—4)7
+ 2=4) =8
x + y a ee
Since 8 < 9, the (radius)’, it follows that
(b) If the centre is at (a,b,c), the distance of
(3,2) lies inside the circle.
a point on the sphere from (a, b,c) is

Vile — 4)? + (y= by + (= 0] A similar argument works for (b), (c), (d).

(b) outside (c) the point lies on the sphere

Hence the equation is

V(x — 4)-+ (yb) + - 6) = 4 (d) inside

or (x— a)? +(y—b +(e-ch =r 6 (a) radius 2 units, centre (2,1)

(b) radius 3 units, centre (1,0)

5.1 Exercise 2 (c) radius 5 unit, centre Gs, =~ +)

1 Using x? + y* = ? the equations are: 7E The distance between the centres is:

(a) x+y? =225 = (b) x” +-y* = 16 V((120 — 160)? + (150 — 180)" + (30 — 40)*]
(c) x7 +57 = 2.53 since 2mr = 10, sor = 1.59 = /2600 = 51
(d) x7 +” = 400 since the radius is the
distance of (12, 16) from the origin

2 Using the formula (x — a)’ +(y—byY =P,

the equations are:

(a) (x-1)7 + (y-17 =9 Since this is greater than the sum of the two
(b) (x +4)? + (y-6)* = 64 radii, they do not intersect.
660 SOLUTIONS
eee eRe

8E Suppose the equation is Henceb = 1 or b =3

Ipoh ai 6a andtto:—3 a
(x—a)’+(y—by
=r
So the two equations are
Each point which lies on this circle will
satisfy the equation. (x — 16)? + (y — 1)* = 100
(6—a)’?+(9—b’
=r and = (x — 2)* + (y—3)? = 100
(13-4)? +(-8-by=7
(—6, —3) lies on the second circle.
(=4= a) +(-15-b’ =r NB You may have looked at the problem
These simplify to geometrically and spotted a 6, 8, 10 triangle
a+b —12a-18b4117=r @ pattern.

a +b —26a+16b4+233=r @
ity pee

a+b? +8a+30b4+241=7r © 5.1.3 Trigonometric identities

Eliminating 7’,
So P *« Exercise 3
@ — @ => 14a — 34b —-116 = 0
1 (a) 1+ cosx = 3(1— cos” x)
@- @ => 34a4 14b4+8=0
=> 3cos*x +cosx —2=0
giving a — 1, O39) n= 13
(b) (3¢ — 2)(c + 1)
So the centre is (1, —3), radius 13 and
equation (x — 1)* + (y + 3)? = 169. (c) 48.2°, 180°, 311.8°

9E Suppose the equation is 2 (a). tan = #60 33-7 oo

(x — a)? + (y — b)* = 100.
(b) tan'@'= 1:6, 7A 58.074238.0-
Since it passes through (10,9) and (8, —5) it
follows that (c)"tan2e = 1-4, .6-= 27.255 117.2207
297 .2°
(10 — a)’ + (9— b)* = 100
(8 — a)’ + (—5 — 6)? = 100 2m 4a
3 (a)
0,? —3) —-
3 2,
2 TT
which become T An
(aay ise
a +b’ —20a-—18b+81=0 @ 3 3
5
(c) 1.05 . 5)1.82, 4.46, 5.24 le+)
and
a+b’—16a+10b—11=0 @ 4A) ORS25 622625. 6 moO
Firstly, eliminate a* + b* by subtracting
(>) Os re 1
@® from ®
4a + 28b = 92 (c) 0.46, 2.68, 3.61, 5.82

i.e. 4=23—7b (d) 0.34, 2.80, 4.71

Now substitute for a in @ to obtain a

SE (a) Let N be the foot of the perpendicular
quadratic in b
from C to OP.
(23 — 7b)’ + b? — 20(23 — 7b) — 186 + 81 =0 Then PC? = NC? + PN?,
where NC = 100sin@ and
which reduces to 6* — 4b +3 =0 PN = 130 — 100 cos 6.
 MATHEMATICAL METHODS 661

(b) PC? = 100° sin? 6 + 1002 cos? 6

—2 x 130 x 100cos 6+ 130°
= 26 900 — 26000 cos 6

meceroie « (3) (3)

(GC) S22 SOS Bae
pe Z

The curve is a stretched circle (i.e. an

ellipse). (a) (i) h=c+d=asin0+bcosd

h
(ii) > = sin (0 + a) =>bh= rsin (8 + a)
5.1.4 rsin(@+ a)
(b) 2 =a’ +0’ and coh aig!
a

Thus, you can write asin@ + bcos @ in the

form rsin(@ +a), where ? = a’ +b’ and
b
tana =-.
a

4 (a) r= /(4 +7’) = 8.06

-
tana efi => a = 60.3°

=> 4sin@ +7 cos @ = 8.06 sin (6 + 60.3°)

(b) The approximate solution is 10°.
[The value predicted by theory is 12.4°.]
(c) (i) The greatest height above ground is S.1 Exercise 4
2.9m, which occurs at an angle of
1 (a) 3sin?+2cos6= rsin/6 +a)
approximately 60° (59° in theory).
where r = \/(3° + 2”) = 3.61
(ii) The wardrobe may be tipped at any and tane== 2=> a= 33,7"
angle between 0° and 12°.
(b) 3.61 sin (6 + 33.7°) =3
GEE OOH = Soe OF 123. 7.
= 6 = 22-62 01 90>

2 (a) 5sin@ + 12cos@ is equivalent to

rsin(@+ a) where r and a can be found
from the triangle below.

P rT = 13;
tana = 12
12

=> a = 67.4°
The graph is a phase-shifted sine wave. S

(b) y = Ssin (x + 53°) since the graph has Since 5 sin@ + 12 cos @ = 13 sin (@ + 67.4°),
amplitude 5 and is y = 5 sinx translated the maximum value is 13, which occurs
53 when @ + 67.4° = 90°, i.e. at 0 = 22.6°.
throug h |‘
662 SOLUTIONS

5.1.5 Addition formulas

5.1 Exercise 5
1 (a) sin (x + 60°) = sin x cos 60° + cos x sin 60
=-sinx aye Cae
13 sin (9 + 67.4°) =9 2

=> 6+467.4° = 43.8°, 136.2°, 403.8", etc. 2 (a) sin(x +7) = sinxcosm
+ cos x sint
> 6 = 68.8° or 336.4° SIE

3 (a) (i) AD=6sin@ (b) i sin x

(ii) AB = 8cos@
xv
(iii) perimeter = 12 sin@ + 16 cos @

(b) 12 sin + 16 cos 6 = 14

=> 3sin8+4cos6 = 3.5 —sin x

(c) 3sin@ + 4cos@ = Ssin (6 + 53.1°) xv

=> 0 = 82°
The sides have length 5.9 and 1.1. A translation of the sine graph by —7 in
the x-direction has the same effect as. a
(d) Largest perimeter is 20, when 0 = 36.9°.
reflection in the x-axis.

4E (a) B will move in a circle because OB

remains constant. 3 (a) cos(A +B) +cos(A —B)
OB = / (67 +3?) = 45 =6.71m = cos Acos B — sin AsinB
(b) B + cos AcosB+
sin AsinB

= 2cos Acos B

] (b) cos (A — B) — cos(A+B) = 2sin AsinB

4 (a) sin (45° + 30°)

2) DE
= sin 45° cos 30° + cos 45° sin 30°
d=CD+BC= AE+BC
= 6sin6
+ 3cos@ 1 Sten eas
Se
OS aR ae Os SE
V2 AO
(c)
: sy ae:
6 me Ny
(b) sin 15° = sin (45° — 30°) = v3-1
3 2/2

63
6sin@
+ 3cos@ = rsin(@ + a)
where r= /(6 +3”) =6.71m 65
tanna=2=4>0=26.6°
6 (a) 120°, 240°
(d) The maximum distance is 6.71 m, when
sin (0+ 26.6°) = 1, Le. 0 = 63.4°. (by 30°590° S150" .270°
 MATHEMATICAL METHODS 663

’ T
- sin («+2) 4 a =b' sin? A+ —2bccosA
+b’ cos’ A
7E tan («+7) Nee = b* + ¢ —2becos A,
4 we sincecos* A + sin? A = 1
co _
>@=b6+c—2becosA
75 (cosa + sin)
—~ (cosx + sinx
J2 5 If A = 90°, cos A = 0 and the cosine rule gives
1 ; a’ = b’ +’, which is Pythagoras’ theorem.
—~ (cos x — sin x)
We If A < 90°, then a” < b? +c’, which
cosx + sinx
corresponds with the result given by the
cos x — sinx cosine rule.
Dividing the numerator and the denominator
by cos x, | 6 (a) Co =a +b —2abcosC
T 1+tanx (b) 6? =a* +? — 2accosB
in |sese= |) =——————
4 1 —tanx

8E (a) sin3x = sin 2x cos x + cos 2x sinx

= (2sinx cos x) cosx

+ (1 —2sin’ x) sinx
= 2sinx(1— sin’ x) MB = ccosB
MC =ccosB—a
+ sinx— 2sin?x
In triangle ACM,
= 3sinx — 4sin’x
b?-=b? + (ccosB— a)? ©)
(b) cos3x = 4cos’ x — 3cosx
and in triangle ABM,
sin 2x 2sin x cosx
retail. = - co = hb’ + (ccos B)* @
cos2x cos 2 x — sin’ 2 x
ere 2 Eliminating 4” from @ and @leads to the
Dividing by cos’ x,
Datramies 2tanx result.
sinZ x 1 —tan* x
i
cos” x
WS?8 Exercise 6
10E 0°, 48.6°, 131.4°, 180°, 270°, 360°
1 5.76cm
e+e e
5.1.6 Solution of non-right- aa) oi

angled triangles: (b) 34.0°, 44.4°, 101.5°

the cosine rule
3 (a) a =10° +7 —2 x 10 x 7608 45°
G— 7\-em
iD
1 b=bsinA (b) a? = 107 +77 —2 x 10 x 7cos 120°
y=bcosA a= 14.8cm

2 x=c—y, becausex+y=c (Ca —s e-em

x=c—bcosA
4 5.15
3a@=h +x
(bsin A)” + (c — bcos A) 5 5.4ikm
664 SOLUTIONS

Hy Solution of non-right- %.1 Exercise 7

angled triangles: ; Bie aces

(a) in80° Sine
the sine rule in 80°
2 gp eis
giving 6 = 38°
ks
€
| 1h, =csinA
3bh, = sbcesinA Thus the third angle is 62° and
b 8 ee ce
sin 80° sin 62°
Note: This is a more useful expression for the 8 sin 62°
area of a triangle than 5b/,, since it is given giving x = atpa aa = 7A Gan
sin 80
in terms of sides and angles.
(b) 6 = 27.4°, the third angle is 22.6° and
Area of triangle = }bcsinA x = 10.05 mm

= ; x product of two sides

2 (a) x? = 187 + 10? — 2 x 18 x 10co0s 35°
x sine of included angle.
=>x=114cm

Dap — sine sin@ — sin35°

— ee)
| Area= $ab sinC 108 Peas
(@ must be acute)
The remaining angles are 30.2° and 114.!

(bx = 1003
The remaining angles are 49.3° and 58.7°
3 {bcsinA = fab sin C
= esinA—
asin © 3 (a) 51.6cm?* (b) 3420m* (to 3s.f.)
a é
sinA sinC 4 (a) 43.2 m, 33.2° and 46.8°

| 4 By symmetry, if c is the base, area = $ac sinB (b) 24mm, 16.3° and 73.7°

(c) 41.9° and 73.1°, 40.1 cm

Nn Sac sin B = sbcsin A
b a Cc b (d) 48.2°, 58.4°, 73.4°
sinB sin A’ sinC sinB
5 437km, 050°
So, combining this result with that obtained in
question 3 gives the sine rule:

Sak b cat?
sinA sinB sinC

| 6 Area =1x 4x7 x sin30° =7 cm?

 MATHEMATICAL METHODS 665

5.2 Vector geometry _ 4 (a) Values of t generate the position vectors:

5.2.1 Vectors and position

vectors

You can apply a similar method to (b) and (c)

giving:

(b)

c=3a
d=2b
e=b-a
f= —b
—
In each case the graph is a straight line.
(b) AB=b—a 2
CD =d—c=2b—3a The vector ;|gives the direction of the line.
DE =e—d=(b—a)—2b——a—b
EF =f—e=-—b-—(b—a)=a-—2b
—
FC =e—f=3a+b The three lines join each vertex to the mid-
point of the opposite side. [Such a line is
ic) CB ADE + BF + FC called a median.]
=2b—-3a—a—b+a-—2b+3a+b
= (0 A

The sum of the vectors is zero because the B

net displacement around the quadrilateral

CDEEFC is zero, i.e. the vectors have
returned to their starting point.
— —
(d) —AD = 2b—a=—EF ad
EF is of the same magnitude as AD but
in the opposite direction. They intersect at the point (4, 2).
666 SOLUTIONS

§.2 Exercise 1
1 (a) The direction vector is

)-(1)-(8
ae de SENG
he line
so the line isi Nee
liek rf+(e
3|
3 6
Similarly OO = |6| and OR = |6
6
23
3
mle diesel
6 fT-G)(4]
=
Re
3
0
=
RP = | -3
0
© bls
=> PQ +
a
—3

QR-+
eo
RP
3
ae ied
—3 3 0 0

= 3) + Oo;}+ 4-3) =1]0

—3 3 0

: 1
| 7E (a) t=5(*-3)
1 a att
(b) y Se
1+ x5 (x Goa
Ne 5% 5

Since PQ = SR the vectors are of the

3 2
(c) The gradient is 5 The vector | has same length and direction, i.e. PQ is
parallel to SR and equal in length, so
radient —. PQRS is a parallelogram.
e 2
(b) It follows that PS must also equal QR.
| 8E (a) x=3-2s (b) x=142 6 0
= =
: y=—2+s Vi eee ee Sy ayy (()) y= 0 OE = 8

> y= -—5x-5 >y=—-—fx-4 10 10

—6
The two vector equations give the same (ii) ‘ABP bie a= 8
straight line.
0
—2
They have the same direction, | or 0 —6
— —
2 D= 0 AG = 0
‘lrand appear to be different since
10 10

different points on the line have been —6 —6

chosen. AE eae DE |e
 MATHEMATICAL METHODS 667

3)
5.2.2 Equations of lines
(b) (i)OM=OA+
4B = |4
0
B
3
4 0
iL ch= |G c= |h6
10 0 3
4 4
(i) AN=| 4] ME=| 4 f= 0 g=1|6
3 3
10 10
x 0
(111) AN = ME, which is to be expected
2 (a) \=0S|y| =| 0
since AN is parallel to ME and of
equal length. z 0

5 4
8 0 eS Sl 99) =|] 6
4 (a) a= b= |10 c= r10 z 0
0 0 So A\=0 givesO, A= 1givesD.

4 4
dia) e= (58
3 3

NB You can find d and e by viewing the

roof from above:

O 10

(b) is I, J, K and L correspond to A = ;; ;‘

, 2 and —1 respectively. H, I and J are on
hell ) ae line between O and D.
A B
(c) K and L are on the same line but outside
OD.
_4 4
(b) AD=d-—a=| 2 OD= 2 0

3 3 (a) Since OC = | 0| the vector equation

3
a4 4 corresponds to the line OC.
Bee |=o Chale?
3 3
(c) By Pythagoras, the lengthofAD is
n/(40-2 8")= 4/29= 5.4m
668 SOLUTIONS
‘esp fe nS ea

| Be 0 NB These are not unique equations. For

| 4\=0> Vale, Ol|peieepoint. © example, AE also passes through E and has
5 x 0 4
equation | y| = |6| +m] —6]. Thus the
x 4
z 3 =3
NS On| ea OLE taGr
point (—4, 12,6) is found either by starting
3
from A and putting \ = 2, or by starting fro
E and putting « = —1.

7a) SU te PO 1 2

<V

The points lie on the line CG.

H, I, J, K and L correspond to A = 4, 5, 3, 2
and —1 respectively.

: a 0
O(a) Wiheny\— 0s |Eyal— |pOaliouikes point Bs

x 0
Ware \ = iL, ||sy ||= |LG | was presto 122
3
(b) The lines intersect at (3, 2).
Hence the line is BE.

(b) X = 0 gives F, \ = 1 gives E, so the line (c) A=2, p= 1at the point of intersectior
is FE. The two vectors each specify the
(d) Since at the point of intersection both x-
direction of the line.
and y-coordinates are equal,
1+ A= 1424 and, A=3 =p
6 (a) The line AD passes through A and has
0 (e) Solving these gives \=2, w= 1.
—
direction AD = | 6] so it has
0
8 The intersection is at (24,22).

DG 4 0
equation |y| = |0] +A16
z 0 0 5.2 Exercise 2

(a)
x
yO
|4
NG
0
me bees
bl= lo)[2
ez r 3

XG 4 —4
(c) |y]| =]O] +A] 6 => 3-—3A=1+24 and 3A=2p
giving A= 3 u =}
& 0 3
and a point of intersection (2, 1).
 MATHEMATICAL METHODS 669

OG) OG
4
(b) A = 0 gives | 0];
i
3 =
(a) The length of a is \/13.
0 The length of b is \/17.
X= 1 gives | 5 i.e. EG
3 bic= b= a= ce
<; “=|
(G)}CE (d) BD
So the length of c is
J ((—2)7 4.27) = 9/8 = 2.828.
%
3 Writing r for | y | gives:
V8
z
4 0
— —
(a) r= OA+XAAB= ]/0] +A] 5
0 0
4 -4
=: —=
(b) r=OA+AAC=]0/+A] 5 (V8)° = (V/17)" + (/13)"
0 0
— 2/17/13 cos 0
4 0
=—> — if
(c) r=OA+AAF = ]0] +A] 5S = oe) = === & WIB8S
Woot
0 3
=> C= 423
4 af
=_— —> 4
(d) r=OA+AAG=]0] 4A] 5S
! 2 (a) As in question 1, by Pythagoras,
0 3
a=J(aj +43) > a =a, +a)
4 When the y components are equal,
3 —X = -—3+ p, and when the z components Similarly b? = 6? + b5
are eGUalay \i——/i mite:
(b). 7 = (by — a,)° + (by — a)?
Neh =3)
N= 3S x =4—3
= | = bj — 2a,b; + aj + b5 — 2ayb, +. a5
L=3=>xex=—4—-3=)
= (a7 +45) + (61 +05) = 24,6, — 24,6,
Hence the vertex is at (1,0, 3).
= q’ +b’ —2(a,b, + ab)

5 (a) They meet at (5, 3, 11). (c) By the cosine rule,

(b) Parallel lines — no intersection. C =a +b? —2abcosd
(c)) They-meet at (6, 3; —2), So a,b, + a,b) = ab cos 8, since the two
(d) Skew lines — no intersection. expressions are otherwise identical.

6 They meet at (—9; 3.5, —1.5).

670 SOLUTIONS
EEE

jab |
gy 4
4PQ=| 4
PR=| 8
Gi = 3 a bye b, =4 4 er
a=J/(9+4 =V13 b=V(1+16)
= 17 Ag 130 24
=>3x14+2x4=,/13,/17
cosé
=> cost = “749-4 1616),/(16 2k
= (P= oe”
11
cos Yoon
7 1
and @ = 42.3° as before. Q@=-PG=/-4| QR=| 4
4 at
TE 6c 20
5.2 Exercise 3 = cos = Tag 16 + 10) (inl = eee
1 (a) Using cos 0= —— gives
a-b . 81 9 1
ab ~ 916d eum a2
2}-(3]
5) [3
> O=45° and R=180°—
90° — 45° =45
cos 6 =
V(25 +4) V9 +4) 1
SS Roce 19 5
—
AB=| 6
nd
DC=| 6
a /Os Ow 377 —5 —5
=<) = ili,e"
Since AB = DC, AB is parallel to DC and of
(DCO Ae
equal length. Hence ABCD is a parallelogram
D, C
4
—
iN Ds || Al
A B
2

|AB| = /(1 +36 +25) = /62 =7.87

|AD| = /(16 +1+4) = 21 = 4.58
cosA =
AB-AD
—=—>«>
4+6-10_
|ABJ|AD|—-/62,/21
=> A= 90° and the parallelogram is a
rectangle.

6E Consider a unit cube as shown.

(b) Both vectors must be pointing away
from A.
=_> i>
epee EAC
cos A= —————_
|AB| |AC|
ere
V17,/29 ./493
— A Aeon
1
3 (a) 0= 54,9° (b) 0 = 46.3° The longest diagonal is OF = |1
1
 MATHEMATICAL METHODS 671
a

(a) You require DOF.

5.2.4 Properties_of the scalar
—=—= —_=>
cos DOL ———
OD: OF product
|OD| |OF|

Ole te Cy*

ONFSApa
pes[SUele Ahls eee
ieee 3
=> angle DOF = 54.7°

(b) One face diagonal is OE.

=> cos EOF = ———_

OE: OF
ae " JOE] [OF] (b) d is parallel to e
a is perpendicular to d and e
1 1

CO} eon al Dey [ene beeso, aaa

ale 1 eee
ld] =5 |e| = 10
232 6 (b) aca = 25 b-b = 50 c-c= 40
d-d=25 e-e
= 100
angle POR — 39.3
(cWw a-a=|al’
(c) One other longest diagonal is CE, The scalar product of a vector with itself
represented by is equal to the square of its modulus.
1 0 1 3 (a) a:b
= 35 b-a=35
Seay paren
(b) asc = 30 Cra 30)
1 0 1
(c) a-b=b-ea a-c=c-a
If the angle between the diagonals is 0, You can conclude that scalar
~=a— So>
multiplication is commutative as a
eel: consequence of the commutativity of
noe ~ ICEL [OF] ordinary multiplication.

il 1 1 4 (ay a-b+a-c=35+30=65
—1 1 3
il 1 a
by an(b+e) =| |-| | =214+44=65
11
Oa eee
(c) So, the distributive law appears to hold,
iG om ie.a-b+a+:c=a-(b+c). In fact, it can
be shown that the distributive law holds
for all vectors a, b, c.

5 (a) a:b = 35 a-d=0 a-e=0

(b) If the vectors are perpendicular then their

scalar product is zero.
=5
(c) ;|is perpendicular to b.
672 SOLUTIONS

(d) Since a:b = abcos0,a:b=0>a=0o0r (d) Sincec+d=0, c is perpendicular to d,

= 0 or cos 6 = 0. Thus either the vectors i.e. OC is perpendicular to AB.
are perpendicular, or one of the vectors is
the zero vector.

OC is called an altitude of the triangle.

E (a) b —h is the vector HB and since HB is

4 = perpendicular to OA, a+(b—h) =0
(as OAs —— ©OAl— a
= ; =
— (b) HA is perpendicular to OB so
APS—3p— a b-(a—h)=0
—
(c) b-(a—h) —a-(b—h) =0
=>b-a—b-h—a-b+a-h=0
(b) AP-A'P = (p—a)-(p+a) =>a-h—b-h=0 (since b-a=a-b)
=> (a—b)-h=
But, since |p| = |a|, p? — a* =0
(d) a—b is the vector BA and this is
—
(c) AP is perpendicular to A’P, i.e. the angle perpendicular to OH.
in a semi-circle is a right angle. Thus OH is also an altitude of the
triangle, i.e. in any triangle, the three
altitudes are concurrent. (They intersect
in a single point.)
5.2 Exercise 4
— > — >
a) jaj=VJ13 [BJ = V34— el = 13 5E (a) OR = OP + PR = OP +00 =p+q
(b) pari b-c=-1 c-a— 0 |eee SES
(c) ¢ is perpendicular to a. (b) Using the result that
a-(b+c) =a-b+a-c,
3
(d) By inspection is perpendicular to (p+ q)-(p—q)
=5) =P Pin, Bide died
3: Itiple oof ea:
or any multiple i — Pa D ape)
(since p:p
= p2 and p-q
= q-p)

2 (a)ra- b= 0 b-c=0 Cra () (c) (i) OR- QP = 0 = OR is perpendicular

ton@Pe
(b) All three sets of vectors are mutually
(ii) If OR. QP =0
perpendicular.
then (p + q)-(p—q) =
ie. p>? —q? =0
Bc eel => p=q and the parallelogram is a
rhombus.
16
(b) d=a—b= | This proves the well-known result that
—8
the diagonals of a rhombus intersect at
(c)se == 32 32'=— 0 right angles.
 MATHEMATICAL METHODS 673

5.2.5 Vector equations of ee ason

slanes ‘3 . : :
z 6 6 e,
The planes are parallel.

67a (0, On onites Ci (blaOner Oe ess

Le) (6 JON Ob uetia
(da 735.0) (ee (25 2502)

: 7 (a) If the plane is parallel to ABC then its

direction will still be specified by vectors
CB and CA, but will pass through O
instead of C.
The equation will therefore be:
i, fC= SE ' d—=2b e= —2b
‘ 59 0 0 él
=-—2b+a (B= alae 2
i=sFa+5b 1 ON aa = 86 ty 0
h=2b)— 2a
ki 0 —6 —6

— eS ae (b) Similarly the parallel plane through G will

poRDE >" |6"1 = Op =ar 0 be:
° 2 : x 6 0 6
6 6 0 piles, oMeler WG. leno
—-
Pe — sO hea |0On |e z 6 =6 —6
6 0 6
In order to reach any point in the plane DEF, bre! 3 pea 3 ia ee
it is necessary to go from O to D, followed by | 8 OHis | 0 HD is 6 HE is 6
some combination of DE and DE. Thus, if P 6 —6 0
is some point on the plane, Plane DEH 4s
> > > —_ —_ >
OP = OD+ DP = OD+ ADE + pDE x 3 3 = 3
for some \ and J, i.e. VE) ol ORIG Gale leviia| ee)
6 —6
x 26 0 :
y| = |6|/ +A) O} +p] —6

‘ : : 6 5.2 Exercise 5
| 4 The coordinates of each point are: 1 Taking A as a particular point on the plane,
(a n(G52600) sateen (b) (0, 6,6) ic. E —3 —6
; 3 —
(c) (6,0,6) 1. F (d) (3, 3, 6) B= |-1 G2
(ce) (4,4, 4) 1 4

2
—==> =>
| Sa Pau ees CA had) y| =]|3]+Al—-1} +y| -2
z 1
674 SOLUTIONS

2 The three points at which the plane cuts the 5.2.6 Cartesian equations of
axes are
planes
(
a 0 0

Fev Ole By (a, KEN)

F
0 0 3
| 1 x=6-6) ®
—2 —2 fF y=6 6 ®
—_ — Oh 6A + 6 @
AB = | -1 AG 0
@+@+@O>x«+y+7=12
0 3

so |2 (a) A is (6, 0, 0), B is (0, 6, 0), C is (0, 0, 6

which suggests that x + y+ z= 6.
By —2 —2 x= ou =@
y= 6A @
Z=6-—6\— 6p @
z 0 3
@+@+@Ox>x*ty+z72=6
—2 (b) For the plane through O,
—
Ba) ©) y) x= 6
i 6X
3
Z= —6\— 6p
=>xtyt+z=0
OAD For the plane through G,
a1) + 6u
G —2 0
y=6+6A
y) =A] 2) +p) 4 Z=6—6r—6p
z 3 0 >x+y+z=18

—2] —2 3 x=6-—6\ — 3p ®
= —
DG =n —2) IDI} <= 2 y=6 ale KO)
—3 —3 a 6A + 6 )
@+ @>x+z2=6+3p @
Be —2 —2 —2 Eliminating 4 between equations @ and @
Ft = | A SRN el Raia @+2x @
Zz 3 —3 =3 => y+2(x+z)
=6 —6u+2(6 + 3p)
=> 2x+y+2z
= 18
for BCD

| 4 Using the method of question 3,

—3 —6
— — x=5-3A+2y @)
AUAB =o AC = | -2 y=2 + 3p @
3 6 z=4-6A+ w ®
—! — 2x @- © 3 2x-z=64+3p @
i.e. AC = 2 AB and the three points are
@- @>2x-y-z=4
collinear.

|5 |1| is perpendicular to the plane DEF. It:

1
called the normal to the plane.
 MATHEMATICAL METHODS 675

1 1 —6
=
=a) Bee et ==1() > Be 4— 2

1 il 0

1 6

1
So | 1] is perpendicular to DE and DFE,
il
two vectors which lie in the plane.
2

(b) CA=| 0 Cpl

5.2 Exercise 6 =4 =
1 (a) n-r=n-a 0 2 0
2 Xs 2 0 (c) r= ]O;] +A] O}] +p] 3
=> |—3]/°+|y)-= |—-3 4 4 4
il Zz 1
(Qe NX @
Le. 2x —3y+z7=0
y = 3p @
(b) 2x-3y+z=1 (CSS z=4-4\-44 ©
(e) 6x + 4y + 3z = 12
1
2a)en=) | 6
0
1 58 1 1
1 = 0
0 z 0 0 4E The coordinates of the vertices are
Leoxty= A (1, 0, 0), B (0, O}S G (0510);

D (O10); HS (0, 0, il) R(ONOR he

—1
(a) The vector equation of AEB is:
(b) n= 1
53 il —1 —1
0
—1 x —1
z 0 1 0
tl e|(eyal = 1}+°10
ne ed 0} [0 giving the Cartesian equation
xty+z=1
ie. -—x+y=O0Oory=x
The vector equation of DCF is
Although this looks like the equation of a
line, it represents a plane, formed by the oe 0 —1 0
lines y = x corresponding to various Vile |]Sai Sed wil) em! al
values
of z.
& 0 0 —1
1
givingx+y+z=-—1
(c) n= ]1], xty+z=1
[Or both planes may be found by
1 inspection.]
676 SOLUTIONS

1
(b) Since both have normal vector | 1 | the
planes are parallel. 1

(c) ECB isx —y—z=-—1

and FAD isx-—y—z=1

The required angle between the line and the

F angies
plane is 90° — 49.1° = 40.9°

1 =i

1 D; 4n= 1 ho — 1

(2) ole m= 3 1 1

1 1 Sse ilsei i
cos 6 = ———
Vaue =->
3 0=70.5°
9 2+3 +1 6
cos 8 = = — However, the dihedral angle is obtuse in this
VA BND
case, So:
= ee dihedral angle = 180° — 70.5° = 109.5°
1 5
(b) ny = —3 Ny — 0

—2 2

SS DS OV

0
(Cant 0 iy = || tl
—2 3

= 0 = 148.1° or, taking the acute angle,

Si?

2 (a) The direction vectors for the two lines are =s)

fle {i}
—
BS +0
fl
6 = 108.4° or 71.6°
= |AB] = y/(1? 403 ii) = 9/2
=
0
(b) Taking direction vectors 0} and ==
AG=|—1 | SAC =.
=) 3
1
3), @=34.7-
4 1
—=>
BC =| =1 |= |BCh 4/2
 MATHEMATICAL METHODS 677
RR

(c) OAB is x — y+ z= 0 with normal

1
5.3 Binomials
=) 5.3.1 Binomial expansions
1

OAC is x — y— z = 0 with normal

1
|1 (a+b) =(a+b)(a +2ab +07)
—1
—1
a +2a°b + ab’

+a’b +2ab? +6
Sao at ed a +3a’b +3ab* +b?
J3/3 3
=e == 0.5.
which is the (acute) dihedral angle of the 2 (a+b)* (a+b)(a+b)
tetrahedron.
= (a+ b)(a’ + 3a’b + 3ab* + b*)
0 1 = a + 4a°b + 6a°b? + 4ab? + b'
(DEBGisi— Jel =o
1 0 |3 (a) The coefficients form the rows of Pascal’s
1 triangle.
direction | —1 (b) The nextrowis 1 4 6 4 1,asin
0 the answer to question 2.

1 (c) (a+b) =a? + 5a*b + 10a°b? + 10a*b*

OMis 1 — jp 1 + Sab* +b?

0
| 4E a! + 4a°(2b) + 6a” (2b)? + 4a(2b)? + (26)*
= a’ + 8a°b + 24a*b* + 32ab° + 16b*
cos = a = 0 8 = 90

i.e. opposite pairs of edges of a regular

tetrahedron are perpendicular.
os 3 Exercise 1
1
1 (a) (a+ b)® =a® + 6a°b + 15a*b? + 20a°b°
(e) The direction of BC is | —1 + 15a°b* + 6ab° + b°
0
1
(b) (p-q)? =p - Sp*4 + 10p°q° — 10p*q°
+5pq'-—q@
The normal to OAC is | —1
-1 (c) (3x +-y)* = (3x)? + 4(3x)>y + 6(3x)?y7
+ 4(3x)y> + y4
= cos = = D= 35.3"
= 81x* + 108x°y aE 54x77

B + 12xy? + y4
Thus the angle between
the edge and the face is (d) (1 4.2z)° = 1+ 6z + 152° 4.202 + 15,7
90° — 35.3° = 54.7°. + 62 +2°
Hence the height 4 of
' Aad
ve .
the tetrahedron is C 2 (a) (a+b) =a + 3a°b 4 3ab 4b?
BCsin 54.7° = /2 sin 54.7° & 1.15. (a@—b)? =a@ —3a’b + 3ab’ — b°
678 SOLUTIONS
en ee

(b) Adding the results in part (a), i 3 (a) The multipliers for the 6th line are:
6 3») a 3 2 1
(a+b)? + (a—b)? = 2a + 6ab? 7 BE PS eee
1 => 6 —2_15 — 20 — 15 — 6) ——
= 2a(a’ + 3b’)
(b) This agrees with the result from Pascal’s
(c) Subtracting the results in part (a),
triangle.
(a+b)? — (a—b) = 6a°b+ 26°
14x x2 x$ xf «xt x2 xp) xe
= 2b(3a* + b’)
1a 10> ASme 1205 210252 21 Oe
3E (a) p=at+b and gq=a-b
>a=3(pt+q) and b=3(p-4@) | 5E The sum is 1024 = 2°. The elements of
Then the mth row of Pascal’s triangle always sum
tore.
p+ =23(p +49)
x[(4(p +9) +36 (v - 9)’ |j 6 x 80 x 72 Fe78
=(p+q)(b -pq+7) tf (=> 80° > 3160782160

(b) p> — 4g = (p+ (-4)) 12 _ (2x1 x10x9xs

x(p” — p(—4) + (-4)’) 5) SONA
= (p—q)(p' +09 +q°)
12!|
4E Writing 11 as 10+ 1, rr == P< lil <1 Se eB

11° ='(10 + 1)? S10? +2 x10 4.1 = 121

11° = (10 +1) (b (12Ale IPAS<IbSeIHLDS Selo Sse fe}
BK ASS S
= 10 3 10 <i o 10,501- Ie
= 1331 _Wxilx10x9xs8
se 5!
11°40 +1)"
12!
=10'+4x10?x14+6x
10? x 1? from (a)
Arg 10 C1 sir
= 14641
(c) we S79)
11° = 15 51 = 161051
Pascal’s triangle still applies, but you need to
carry. 12 II
12! 12 12!
|° tae ae (3) =a
ea) Ss)
5.3.2 Binomial coefficients os —
8 4

os Sa) | 10 (a) 220 (b) 792) 12

1 The reciprocals of#and 3 are and ,

2 (a) The multipliers for the Sth line are:

SipP) A 4 la 3aie 2Arle 1

We So WO SIO GN)
12.\ aig ae 12)
ee () (et >) ora =
(b) The sequence of multipliers is +, 5, 3, 3,4,
. . . 5 4

(cc) Oem
where the numerators decrease by 1
and the denominators increase by 1.
 MATHEMATICAL METHODS 679
ES

3 Exercise 2
1 (a) 56 = (b) 10 = (c) 84 ~— (d):4950

2 (a+b) =a’ +7a°b + 21a°b* + 35a‘?

+ 35a°b* + 21a7b* +. 7ab® +b’

3 (a) (a— b)§ = a® — 84’b + 28a°b*

— 56a°b* +---
(b) (2a — 3b)'® = (2a) — 10(2a)?(3b) ss3.3 Binomial series
+ 45(2a)*(3b)?
— 120(2a)’ (3b)? + ---
5,3.-€
= 1024a"° — 15 360a’b
fi (b) y = 1+ 3x is the tangent of y = (1+ x)?
+ 103 680a°b*
at x = 0, y= 1 and is a good
— 414720a’b? + .-- approximation to the function near the
origin.
(c) (=-4) = (x*)° - a(s')5(<) (c)
x 0.05 0.1 0.15 0.2 0.25
lars 1.15 i153) 1.45 1.6 IS)
(1+ x) 1157-625, 1.330) 1.520875) 15728) 1.953125

The results are close for small values of x but

=x? 6x + 15x = 204-2. diverge as x increases.

(b)
4 a=11 from the symmetry of the binomial
coefficients. x 0.05 01 0s 7 02°17 0.25
{eax 3x7 A1575 | Leo e115 A172 etoo7s
5 (a) 0.0158 (b) 3.05 1+ 3x + 3x? gives a better approximation

wu(s)-m ()-m
than 1 + 3x.

(a) 44 + x)® =1+4+8x+28x"?4+.-..

(")=m=()+()
so 1+ 8x + 28x" is a quadratic
approximation to (1+ x)°.

H(CP)-Oe(2)
(b) The approximation looks good for small
positive and negative values of x. It is
within about 10% of the correct value for
—0.1<x < 0.17.
—1
- ~ (n See ~ (n—r+1)ir (a) With n = —1,1+nx ee
becomes j
—1)(—2

a aa eee eeu er
2!
a Es ©

nlr (b) For —0.464 < x < 0.464, the value of

(a—r+1)!nA the quadratic is within 10% of the true
value.
680 SOLUTIONS
<==

5 (a) Jl tx) =(1+x) &.3 Exercise 3

Li i the first three terms of the series 1 yh ae ee
Ata eae oe
are
b) (1 +x)7? = 1—3x + 6x? — 10x? +---
Dele a, ft ee)
Pe ei ee 5 2 {ah (eee
= ese
abe Oe
D) 8 c (1 +x)

1 1
(b) The value of 1+ i a” is within 3 (a) 1—4x%
+: 10x? = 20x?
1
10% of the value of \/(1 + x) for Bb) leer 5*
WL Kae < Ile 1
c) peax
2
6 The next term in the series is
n(n—1)(n—2)
phi heeTS EE 3 4 (a) /(9 — 18x) = Y(9(1 — 2x)
3! = /9./(1 — 2x) = 3/1 =e

0, putting
So, putting 5 g gives
= —
Nero)
31 + Mea (b) /(9 — 18x)
: 3 = 3(1 — 2x)?
83 — ~ x3 — —»? and a sensible cubic
6 148 F 16
would labore
be ~x ee
3” 3 are+—x°.
—-x° en: rhe
ht This is3 a =3 (1+ 5( 2x) + a 2) (—2x)?
better approximation than the quadratic, but
again is good only for small values of x. G)(—3)(—9) 3
+
fag
TN
: 7E (a) The series is a geometric series, common
+ OG GSGaye
ri 2 (—2x) --)
ratlo —x.

The sum is
1 ey! 3 Bee
f(s) Tx =3=—3x
pa a) 8
x x GOH Go

provided -1<x <1. 1 1

=) oe 2 = = ee
(b) Substituting 7 = —1 gives ss 9 ES
(+x) '=1oxtx =x +--.,
lyn
whose sum is known to be
x
from (a). tx)? a=
Z° 4 =a16
If the binomial series is summed to infinity, 1D iRes
(V4+x)?
then the series is equal to (1+ x)” for go tage
—1 <x <1 and is not merely an
approximation. 6 (a) The binomial series is only valid for
Slee I,
3 4
/ a
RES ee
x

-¥
: “iQ es) R Vie +3)
rme(es(8)-1(8)
Stee S Seed
Note that 1+4x—1x7 x V/(1+x);
(/(1 +x))? is of course 1+ x.
 MATHEMATICAL METHODS 681

1 Minimum value:
7E (a) (l+x)}he1+5x
{
Ones 2
name erst) 2 tC)
2 1 U

thas(
S Onn. Fe 1+(r+s)
i 22
350 (1#2 ie ail
{eA \ 17 pe 35)2s95 fs | Seer
By elas" |e aero nO Dak 14 (144) Sones
te) (3) 18 oe
= 25 gees
140

§.3.4 Error and relative error = 25+ 1.61 mm

This compares well with the more accurate

answer given in question 1.
350+ 10
~ 14+0.5
53 Exercise 4
360
hmax = —— © 26.67 mm 1 (a) 2a = (20+ 0.4) — (19 + 0.2)
Sis)
4 = hea OG
aa = a & 23.45 mm
14.5 = a 05 08,
bh = 25.06 + 1.61 mm ° => area =.(0,5' 0.3)(1910.2)

£2 (a) (+r)? =142r4+7 2142, if 7 can = 0.5(1+0.6) x 19(1 £0.01)

be ignored. ~ 9.5(1+0.61)
1 9.5 + 5.8cm?
ae (il(7) =!
(b) The calculation is inaccurate because the
Sf yy
ae ase relative error of a is large.
w1-r
_(17£0.5+8+0.5)
(b) (i) 142 x 0.05 =1+0.1m*
2 h = (125 +2.5) ‘
(ii) 140.05 m7! (25 +1)
= (125 +2.5) +
2
_ 3 (a) Maximum value:
= 125(1 + 0.02) + 12.5( 0.04)
Pa) (ls) Se res rs
~1+4+(r+s) = 10(1 + 0.06)
Minimum value:
~10+0.6cm
(earls) = dr — sexs
~1-—(r+s) 50+0.5 50(1+0.01)
d= eS
Sond a(t es) Aloe (7s) 3 Speed "32-200 1.350 0.076)
= 37.9(1 + 0.086)
(b) Maximum value:
& 37.943.3cms !
1+r
=~ (1+r)(1+s)%1+(r+s)
1-s
682 SOLUTIONS
es

13 (a) —sin@, sin@, —cosé

Miscellaneous exercise §
(b) sin@ (c) sin40 (d) tan@

1 (a) 10.6cm, 102.8°, 34.2°

(b) 18.2km, 124.8°, 28.2°

2 cosa=-— BIW

4 2cos’¢+5cos¢—3=0

ob = 60°, 300°

3 6
62 (a) |e) (b) erie ele ete
—1 3 —1

1 3
7 (a) r= }]2)] +A] —-—7
3 3

1 3 6
(b) r= ]2]}]+A}]—-7|/ +p 6
3 3 —12
No

8 (a) 2x -—3y-z=1 (b) 4x —y = 19

Sh (@)) S60" (b) 44.8°

10) (a), 88:2° (or 91.8") (b) 90°

11 (a) 1—3h+6h* —10/°

(b) 1+54-4h° +559

(c) 1—-4h-17°-1p3

12 (a) 2+4h-APr+
SP
(b) Let = 0.2 in the expansion.
/4.2 = 2.049 (to 3 d.p.)

(c) 3.873
 683

6 CALCULUS (c) y= (8) =e

dy MTS du
METHODS 1 dx ie dx
= ye
6.1 The chain rule
=2"xe&
6.1.1 Functions of functions =)

A 2 The chain rule does in fact apply to any

locally straight functions and you should have
= 50-27 found that your results were confirmed by
F= 32+ 1.8C whatever numerical methods you tried.

=> F = 32 + 1.8(50 + 22)

| OF
6.1 Exercise 1

dF dF dC 1 (a) (i) 2(x +1) (ii) 4(2x— 1)

(iii) 4x(x* — 2)
dF dF ae (b) (i) 2x +2 (ii) 8x-—4 (iii) 4x? — 8x
dt ate dt
For all your examples you should find that 2 (a) 8x(x” + 3) (b) 10(5 + 2x)*
the rate of change of the composite function (c) 3(4x — 3)(2x* — 3x)?
is the product of the other two rates of
change. (d) 12x’(x — 2)(x — 3)°

3 (a) —2xsinx’; 0 (b) 2cos2x; 2

(c) 36%; 3

3
4 (a) —3x* sin x (b) 3 sin? x cos x
d d
SY 31 and SeeCos x (c) —8 cos® x sin x
du dx
dy
dx
dy du
du.
du
dx 5
y
—=cos4,
du
du_
dx
= 3u’ x cos x: d
pl 2cosu = 2cos2x
ey
= 3(sinx)" < cosx
dx

= 5) sin” x cos x Greeti 35x y = cOsSu

d
2
5 xX —sinu = —3sin3x
(b) y =e" dx
dy dy du
7h etwas,
dx du dx y =sinu

=ex 2x —
dy = dacosu
= acosax

2
dx
= ¢e™ x 2% 2

2
8 (a) 3e* (b) 2sinx cos x (c) 2x &*
= 2x e*
(d) —6 sin 2x (e) 12x(x* “Pal
684 SOLUTIONS
—$———————
OOOO LS

9 (a)
dv
aa”
dV
pe ae
5 6.1.2 Applications to
integration
(b) 200 = ‘ne
6. Exercise 2
dr_ 50
dt mr 1a) $sin 3x + ¢
Winenia—e Ve 2.00)
This is an indefinite integral and so the
= 200=3nr
4 3
constant of integration must be included.

> +7=3.63 (b) —2cos$x +e (c) —2eos 5x + ¢

A dr =i 50
dt 7x 3.632
=121cems (to2d.p.) 2
= (2e) — (2e) = 2.14
d , (to 2 d.p.)
10 (a) r=3+0.042 @=> — = 0.08¢
0
Assuming the balloon is spherical, (b) |- bcos 2x|ie (— 3) — (—$cos(—2))

V see ya An’ =-—0.71 (to 2 d.p.)

3 dr
dV dV dr Note: the angle is in radians.
dt dr dt
2
dV
a =4rr* x 0.08t (c) csin = (6sin 1) — (0) = 5.05
Y (to 2 d.p.)
(When —2) A= S16 trom. @)
(d) [E(2x a 3y'|; = (tx 125) — (x 27)
1

dV 2
ate 0.327!x ((3:16)- «2

dV 3 . IB = 16 Wie

an =20.1m” per minute (to 3 s.f.)

B (ane (5.0),
11 (a) V=x
m/2 x/2
dv _ 3x2 | Sina [-0.5 cos 2x" =al
anew 0

dx dV 2
(b) B is (z, 0),
7
tad
5cossxdx age
= |'sinsx) = 1
oS
ee 1.5(4
.5(4 —— 0.5¢)
0.5t) 0 Fr 0

dV i
(c) When t = 2, ar = —13.5cm? per hour (GGnis (=,0),

12 (a) 2 x sin2x x 2cos2x = 4sin2x cos 2x 7/4 1/4

= 2sin4x | Bend |-1.5 cos 2x] = 45
0
(b) —24 cos 4x sin4x = —12sin 8x

(c) —sinx ee”

 CALCULUS METHODS 1 685

4E The result is not true. 6.1.4 Inverse functions and x

2
|cos x’ dx = —0.44 (to 2d.p.)
1
Exercise 4
by the trapezium rule
1
2:
1 . dy x Me,
and sina? ==0.61 ‘(to 2d.p.) 1 (a) y= =x Sees ae
De 1

ale pe (b) An expression such as (1 + x),/x should

oo (<esinx*) # cosx”
be multiplied out before differentiation.

SE (a) 2singxt+e (b) 2e°*%

+c y= (14x) /x = Jxtxyx =x tx

(c) Not possible by the methods of this dyn

Then == 5% meg: ee
pei ose gil pes
chapter. The expression would have to
be multiplied out before it could be
integrated. 2 (a) $x. (b) =x? = (e) 43x
(d) £(Sx+3/P
+e (ec) —tcosSx+c
(£) not possible
i d ib ail
GE (a) —teosx* +c (b) ee +¢ Sy en ee Spel
dx n
(c) + (2x7 ~ iy +e
aoe
1
dy d Nags) 1 Pa 1
7E =| (1 — cos 2x) dx
0 Nile Then dx dx/dy x
1 1
=
F-jsinas],
er
Sele
= 26 ==
du
dx
2
0 27— (to2rd-p»)
dy 1
y — |spats
—_ = —
‘s

dy eis Pe yl
CER wre
6.1.3 Small increments
d if
Exercise 3 jay =he
ake 3%
d il
1 (a) 12mcm? ~—(b) -7.20 cm? je laone
ake Be
d
oe £
2 It would increase by 3% (about 11 days).
(b) v=
dx x
3 (a) 0.16cm —(b) 2.6cmh™! () y= ars
=>y=Ina+Inx
4 4.5% [because In (AXB) =In A + In B]

dy
5 (a) 0.308 2 =04+-51
= dx
This is the marginal cost in £1000 per
1000 articles. [Ina is a constant value and so has a zero
derivative.
(b) £31 per week
686 SOLUTIONS

faa) lz Avcineeee (b) —Inx +c

dx
8E Ga

epee
ake Shy
Note the symmetry in the x-axis. Note
dy 1 1 = also that, since x = 2¢’, x will never be
hy TCL mee negative.

9E (a) eee
z tye
iia :
dx cosy
(b) perce +
EE + } Ha
(b) The third side is /(1 — x’).
Therefore cosy = \/(1 — x’) and Heras rH x
yee ee ete H
dx \/(1—27)

(CBR ee = costa, eeie ae

OLiy ie)
It is necessary to consider fractional
values of t, for example 5,is is in order to
generate the arms of the hyperbola.
Negative values of t result in a reflection
in the origin.

6.2. Parametric form

6.2.1 Curves which vary 6.2.2 Circles and ellipses
with time
6.2 A
6.2 Exercise 1 (a) (3cos@, 3 sin6)
@) e 0 1 2 3 4 5)
(b) 3sin0
x 0 20° 40) 60) 80 100
ye 20 85 70 45, 10, =35 (c) 6cos@

(d) x =6cos6, y=3sin

(e) RO,3), Q/(3,0), Q'6,0)

(f) (i) 97
(ii) Since the curve is stretched by a factor
2 in the x direction, the area of the
ellipse is 2 x 9x = 187.

(b) Since x = 20t, negative values of t will

simply give negative values of x and since
y = 90 — St’, the values of y will be
unchanged. The result will be a graph that
is symmetric about the y-axis.

(c) The curve is a parabola.

 CALCULUS METHODS 1 687

(b) The ellipse can be obtained from the circle 6.2 Exercise 2
by means of a one-way stretch, factor 3 in
the x direction. 1 Ellipse Cartesian Parametric Area
equation equations
(c) (i) 167 2 2
ee ai x =3cosé 120
(ii) 207 s caer ee y = 4sin0

fh 2 7
Sma) y * ies x =3cosé 157
£ b é Cp save y = Ssind

be2 2)
eee x=0.5cos@ 0.27
m Om oe - yaoacne

X= S'COS 0 3 — 2 cos, x =acos@

Bom.
y = 6sin8 y = sind y = bsin#é x y
2—+—=1
4 9
| 4 (a) x= gcos, y=bsind M2 COSO a yi—iS.Ging
eG Area
= 67
=> cos@=-, sn =
a b
i y
So, since cos” 6 + sin’ 6 = 1, it follows 3 —+—=1
25. 4
that P= ICOSIO me = SIG
D 2 Area= 107
x
oh y
Zs 4
& +t)
en)
x
eee Ae
oe
or Play
6.2.3 Conversion
the Cartesian equation of the ellipse
ane 6.2 B
(b) Comparing with = + es il,
a Cc 1 A eC
1 (a secO = — =1+-=-
a=2, b=3 (a) a cos 0 Cane
c 1 De
=—
b) cosec@ —_=1+-=
(b) b sin 0 @
a i Dae
c cot@ = =i) Qe
(<) b tan 0 @ 1D
COS|U OL
d cot? = — aide ee
(d) SNC Wes ye We

The scale factors are:

a in the x direction,
b in the y direction.

(b) rx V=7
Since the circle is stretched by a factor a,
followed by b, the area of the ellipse is
mxaxb=nab.
688 SOLUTIONS

(c) 4" +1=x°

/ — sing | 3
—— cosecd | x
l (d) i+0-1= (3)
|
| 2
|
2a
>
? _ay42=—
Py
0

3E (a) y=34+%

(b) Since \/t > 0, it follows that x > 1,

\ y > 4. Thus the parametric equations

(b)
define a half-line.
Kane)

Tes 4Ex=t+2, y=r"4;

gS 2S, ey tt
Many others are possible.
: é;
7)

6.2.4 Differentiating
pai LSE ric Cd Uatlons

ee?
Simm
2
COSs O, 1 :
- — 6.2 E i
cos*@ cos?@ cos*0 aEICISe A
0
So, since as tan @ and = sec 0, 1 dy dy, oe dx
cos @ cos Ago aap de
tan?
@ +1 = sec’0 il
t
(b) sin?@ cos?@ ee
i 1
sin?6 sin2@ sin?6 $e Os
0 il
So, since ee = cot 0 and —— = cosec 6,
sin8 sin® dy 4cos@
1 + cot” 6 = cosec” 6 dx —3sin@

| 5 cot?@=x and cosec@ =2y Ge

|| So, using : 1+ cot? @ = cosec * 6, oO yd d =A
ey
1+x°=(2y)’ or 4y°—-x?=1 dx
d

6.2 Exercise 3 "4

1 (a) 5x 43y=22 ina ooo wie =

x
Sx Sx? 2
pre
Sis 16 (d) x+y=ae Sa 3(u- - 1
(byusesee

2 (a) 4x7 =1+—y (b) in (c) (2, =2) and (282)

 CALCULUS METHODS 1 689

akhy 2s dy cos6
pehees 9E = = ——_—
(cE aa dx 1+sin@
At s = 1 the gradient is 2, At 6 =i, (x, y) = (0.0783, 0.707)

and 2d ~ 0.414
Bor a \ 31) gem
eae SO => e j x
dx
dy . 2x y — 0.707 = 0.414(x — 0.0783)
dx 9 y © 0.414x + 0.675
When s = 1, x = 3, so the gradient at x = 3 is
6Oa or 2ee
The area of
7 (a) Parametric differentiation is the only the triangle is
sensible method. 0.550 square units.
cy, MBE
dx 2(t+ 2)
(b) It is easy to spot that y = 3x — 4, so
dy
8
dx
(c) Parametric differentiation is the only
sensible method.
We 1 2) 3
dy _ —3cos0
oe pail 4 9
dx (2sin@ + cos6)
y 0 3 6 9
& Parametric differentiation is the only 0 D 4 6
method. Vv
3 3 3 3
dy _3cos30 v B) V13 5) V/45
dx 2cos20
ya
dy dy | dx
124
di de
94

64

ee T T = —
3 6 9 12 15 x

3 a 2t
Ns st
2 2
Since (-7) > 0 for all t, (2) > 24 0
t dx (b) When t = 0, v= SI: The speed is

3cms_ |, parallel to the y-axis.

2t 1
=X
0 ,= = |
xv a3:
SS Np 3) Ship bos4
(d) t=5
690 SOLUTIONS

3 The line has gradient 3. It is parallel to the 2 (a) 148 (b) 16.3
il
vector | , which has magnitude 1/10, so
3 The gradient is 1.1.
The equation of the tangent is
y = 1.1x — 0.16.

ea , 2
4 (a) (i) aa eee cost

At t = 1, the rate of change of area is

pd hphi

6.3 Further differentiation dA

(ii) When t = 2.5, a —2.01, so the
techniques
area is decreasing.
6.3.1 The product rule
(b) (i) A=sintcost
dA eee
— =cos*t— sin is
dt
1 (a) = 2x sinx + x? cos x When t = 0.5 the rate of increase of
dx area is 0.540.
—— 2 ated
(ii) The area stops increasing when
/ 2. (a) ner
yeti a x) eae day ;
=5sinG 5x5)
diseriate
+ cos ($x) x (4—x)
So cos*t =sin*t
= (bx — 2x) sin (1x) + (4— x) cos (5x)
tan*t=1
3 Whatever functions you choose, you should
This first occurs when tant = 1.
find that the product rule gives correct
gradient functions.
t = in = 0.785

d 5 The gradient at x = 1 is 1.36.

4 (a) read x x*) = 3x? x x? +x? x Ix
x The equation of the tangent is
= 3x‘
+ 2x‘ y = 1.36x — 0.68.
= 5x* as expected.
6 At the turning point e*(1+ x) =0, sox =—
(c) a d a b
ene a-1
axe = Keer b a
Dx oA So the coordinates of the turning point are
(Sipe037r
= gxtth-1 4 pyeto-1

= (a+ b)x**?-1 as expected. 7 (a) At the stationary point, e*(2x + x*) =0 >

so x = 0 or —2.
When x = 0, y = 0, so (0,0) is a
stationary point.
6.3 Exercise 1
S e* cannot equal 0, so the only stationary
1 (a) e*(sinx + cos x) (b) xe*(2+ x) points are atx = 0 and x = —2.
(c) x?(3cosx — xsinx) The coordinates of the other stationary
point are (—2, 0.54).
 CALCULUS METHODS 1 691

8 (a) O=-tee 5 (x +1) — (x +2)(x+1)?

ll du Leese?)
S == — x—
- x ces (x + 1) (x + 1)°

or ue. !
(x +1)
— (x +2)
This is equal to
dx x (x +1)?
dy : —1
SE ——=sinx+xcosx, so at the stationary or
dx (x +1)?
point, sinx + xcosx = 0.
2 1
Dividing by cosx gives tanx +x =0. Note that nia =1+
x+1 x+1
One stationary point is (0, 0).
The others are (—2.03, 1.82) and (2.03, 1.82)
to 3s.f. 6.3.3 Differentiating quotients

6.3.2 Product rule and chain 6.3 Exercise 3

tule 1 (a)
x cOosx
— sinx
z
2
6.3 Exercise 2 1—~x) 1—x
(b) a or e*

1 (a) e*(6sin 2x + 4cos 2x) e* (sin x — cos x)

(b) e* (2 cos 3x — 3 sin 3x) sin* x |

(c) e* (2x sin 4x + 4cos 4x) (d) e**(3sin 2x — 2. cos 2x)

sin* 2x
2%
2 (a) (b) Inx+1
(x? + 1) dy e*cosx—e* sinx
2 (a)
(c) sin’ x + 2x sin x cos x dx e*
cos x — sinx
(d) 2x*.cos x? + sin x”
e*

(e) 2(x + sin x)(1.+ cos x) At x = —1, the gradient is 3.76.

(f) e*(cosx — sinx) + sinx + xcosx
aye. 2x*e* —2xe* 2e™(x — 1)
(g) —2(2x + 3)~* (b)
dx xt ie x
(h) 2x e™ + 3x” e** = xe*(2 + 3x) At x = 0.8, the gradient is —3.87.

3 (a) dy— =e°*(0.5 sin x + cos x) 3 The derivative obtained by using the quotient
dx
The gradient at x = 2 is 0.105. —2
rule is z and by using the chain rule it
(2x + 3)
(b) The gradient at x = 2 is —0.343. is —2(2x + 3)7?.
The two answers are equivalent.
4 (a) Displacement = 0;
: ds
velocity =— dy (1+x7) — 2x’ 1 — 2
dt
5 reaa Ee Va a~ (1422)?
= 0.4 x 512m cos (5127#) At the stationary points, 1 — x’? =0s0
= 643.4cms"' (when t = 1) Ses ceils
When x = —1, y = —0.5;
12
TL when x = 1, y=0.5
21
692 SOLUTIONS
i

aT Se
d
201048 and (by) Setdy 2 — ——
dy
ae
a

dx

C= —019- => ey = 0.058, dy (7 — 3x")

dx
dx (vy+3)
so there is a local minimum at the point
(—1, —0.5).
d
d
x = 0.9 = 2dx = 0.058 and (a) 18x 4+8y—
dx
=0,
dy _ —18x re —9x
d
aii = =-0.043; aon Pe 8y 4y
dx
so there is a local maximum at the point (b) When x = 1, y=9, so y=+43
(1, 0.5). At (1,3) the gradient is —0.75.
At (1, —3) the gradient is 0.75.
Ss There is a local minimum at the point
(0, 0).
There is a local maximum at the point dy . =(2x 56)
(—8, —16). dx (2y — 2)
At (1,4) the gradient is — 3.
2 ay) The equation of the tangent is
cos x + sin" x 1 2
=
aaa ee z = sec Xx
cos’ x cos’ x 3y + 4x = 16.

—sin* x — cos” x 0 — sec” x

Ss At (1, —2) the gradient is 3.
6 (a) 5 (OS ea The radius joins (1, —2) to (—3, 1), so ha:
sin’ x tan* x gradient — 3.
—1 Since this is the same as —1 +4 the
(c) Both simplify to —; or —cosec’ x.
sin? x tangent and radius are perpendicular.

du du dy & Other points on the circumference with

integral coordinates are:
ze dx x dx dx
(0, 5) (0, =3)) (=6, 5), (=%, 4); (6 =)

and (—7, —2)

At all these points the gradient of the
curve should be equal to
=i
gradient of radius

Since x = 3cos@
and y= 3sin8,
=x —3cos6
y 3sin0
6.3.4 Implicit differentiation Sy ey
y sin6

6.5 Exercise 4

The radius joins (0,0) to (x,y), so has

1a) (a)
dy
4y—
Pe peak:3dy oe = 0, : y —1
radient —. Hence ; :
8 x (gradient of radius) oe E
dy —8x to the gradient of the tangent, so the radius
so —=
dx (4y—3) and tangent must be perpendicular.
 CALCULUS METHODS 1 693

d
d d
(In y) = by the chain rule
3G) dxpshx4 ar 0
5 (a) Gy (ny) = dx
dy dx dy — —(2xy + 3x’)
ey, dx oe
yy dx When x = —2 and y =3, a as(i)

Now Iny
= x|In2 So (—2,3) is a stationary point on the
1 dy curve.
By the quotient rule,

— SEK les I Oe" dy — x°(—3x") — 2x(4 — x’)

dx dx a
d dy (—4844
(b) — (3%) = In3 x 3* When x = —2, == ( - = ge)
dx
which is 0. 4
d
(c) an) == ina <a dy (4x — 3y)
4 (a) dx he A
When x = 2 and y = 22,2 =0,
dx
so (2, 24) is a stationary point.
6.3.5 Implicit differentiation When x = —2 and y = Te
d
0,
dx
and the product rule so (—2, —24) is a stationary point.
3xy = 2x" = 8
3 Exercise 5
—

3xy = 8 + 2x*
dy, dy dy |
a
8 + 2x*
She
dy
(3x = 4+ 2y) —= = — (4x
(4x+4+3 3y)
dy 12x” —3(8+2x") 6x — 24
dy’ 4x+ 3y dx 9x? Ox
dx » 3x+2y—4 At the stationary points, 6x? = 24,
SOM ase
ay 2x ty — 3 C= 2S y= 25% == ys
(b) dx —2y—2x so (2, 2% ) and (—2, —23 ) are stationary
points.
Daa
(c) 4+ 4y
a It is clear that implicit differentiation is
quicker for checking stationary points,
3x7 + y but it is not well designed for finding
2xy + 3y* stationary points.
694 SOLUTIONS
ST ee eee

3a sin’ tcost
L
Miscellaneous exercise 6 1 v=
—3a cos? tsint

hs, fet O ; ae
1 (a) —3sin (3x + 2) (b) —4 sin (4x — 1) Initial position is i with velocity a
(c) Scos (Sx — 11)
12 9000rsin*Ocos@; 10900cm? (to 3s.f.)
(a) 2sinxcosx = sin2x (b) 2x cos (x?)
— 2b —o
(c) —3 cos’ x sinx 13 (a) =e (b) ~2(1 +y) c) —=
Sy a siny
(d) —6 cos (3x) sin (3x) = —3 sin (6x)
14 2
(a) 6(3x + 5) (b) 4cos 4x

(d) (2x -7)? 15 (a) 6x(3x — 1) (b) 2cosx — 2x sin

if
(©) shew
—i11
(ce) —8sin(2x—7) — (f) —3x(x*
—5)72 (c) (x + 5)*(4x
) (4x +5 ) Bs—
( ) (4x 1)?

(g) —2sinxcosx (h) —10 sin 5x cos Sx —3x sin3x — cos3x

ae
(i) 4(2x — 3)(x* — 3x)7 2
1

(f) 2sin3x + 3(2x + 1) cos3x

(j) 1+ x(x? +1) 2 (k) —9(3x +4)~4
(g) 2xtan2x + 2x” sec? 2x
(a) 0.00199 m min“! (b) 0.1m? min! (h) 3cos 4x cos 3x — 4sin 3x sin 4x

(a) —0.167 cmh7! (b) —50cm? h7! (i) 2 cos’ x sinx — sin? x

1
16 (a)
(a) —3cosec* cosec” (3
(3x) b) ————
(b) Ey

2
(c) cos? x — sin? x (d) 2tanx sec” x

dy 1
17 (a) = = ———__
(a) dx 1-—sinx

2
18 —cosec” x; —cosecx
cot x

maximum at (—1},—23)
minimum at (5, 13)

100
h=—
mr
(a) —0.0251cm min™'
(b) Increasing at 0.625 cm? min7!

Vi ix ap

dy —2-%
9
dx y—3

10 t=0, 3, 3p 18ms*
 695

7 CALCULUS _ 4 The container is half full when V = 23.

hr +tp= 23>hx146m
METHODS 2 The equation can be solved by decimal search
‘1 Integration techniques
Sd
/ or by plotting the graph and reading off the

solution.
7.1.1 Volumes of revolution
| 5 The Hg in volume is
[P++4] = 0.9855 m?
1 A

1 (a) h= 40cm (b) b © 83cm

ak B
2 The width, w, of the slab increases uniformly
from 0 to 2 as the height, 4, increases from 0 1 (a) If the rectangle of height y
to2.Sow=h. and width 6x is rotated
completely about the x-
axis, it describes a cylinder
of radius y and width 6x
and hence will have volume
my” 8x.

(b) Ley — x’, > = (x”)? =x‘

The length, /, of the slab increases uniformly 2 2
from 2 to 3 as the height, 4, increases from 0 (c) Volume = | my’ dx = | ax’ dx
0 0
to 2.Sol=2+5h.

3 The volume of a slab at height / is = ils) Does

h (2 +h) 6h. Hence, the volume of the = [n(4x + 5%

container is
The graph y = x* — 2x cuts the x-axis
v=|0 h(2 + 4h)j db
2
where x* — 2x =0
=e2) = 0, Se =O Of a
=| (2h +4h?) dh nN

Volumeee
= my” dx
696 SOLUTIONS

When a rectangle, of length x and height dy, 7.1 Exercise 1

is rotated about the y-axis, it describes a cost
cylinder of volume 7x 6y. It follows that the 5 i
volume of the solid obtained by rotating the vee | dye
area about the y-axis 0
4 y2
=| mx dy =n| (2—y)c
0 0
4 4 > 2742
==|" ny dy =|rly?|
= [n}y"] = 81 x = n[2y= |

2 ae sae
Pe hay (b) 9.48 (to 3 sf.)
Volume s
2 “ 2 3 (a) 30

=| mx dy ae Yo if
1 (b)
2: 1 a
=|| m™— dy A
a. = A= (1h)
— $1

s >
Many people would expect the two volumes a
to be equal. It is easy to prove otherwise. Both
volumes formed will be cones. ve | 1 ee | 1 ORE
0 0
1 1
=n| dx | x' dx
0
cone Pata)
=i 1 =
aii Ca Sig
The volume of a cone is given by the formula
V =!nrb.
3 (c) be
30

The two cones formed will have volumes

nba for area A and 47a°b for area B. These 4 (0, 0) and (2, 4)
volumes will only be equal if a = b. Ve
5
You could have calculated the volumes
formed using integration. The volume formed Be ae . ; ‘
by rotating area A about the x-axis is 7.1.2 Integration by inspectio
a a

| my dx = | mmx? dx
0 0
7 qa

-- [}nm?x'| | (a) Differentiating the function sin x” seems to

0 . . .

be a sensible starting point.

2)
Lim?a?
3 sre: 2
2 — (sinx”) = 2x cosx
31 7b a because m?a* = b* dx
using the chain rule
This confirms the formula for the volume of a
cone. 2 :
so |xcoss alne = Tsin x? +c
 CALCULUS METHODS 2 697

(b) Differentiating the function x sin 2x by the

product rule,
d
— (xsin2x) = 2x cos 2x + sin 2x
dx

= x sin 2x =2| xc0s2x dx

+ |sindx dx

— 2| cos 2x dx xsi |sinax a

= 2 xcos2x dx = x sin 2x +4cos 2x +

= |xcos 2x dx ss $x sin 2x +4cos2x +c

_ (c) Differentiating the function sin 2x by

inspection, At the point of intersection,
b> =Sa—4, b =4a
sae (sin 2x) = 2 cos 2x = 44a=5a—4
= Sa— 4 “and b—4

so |cosas dx = 5sit 2x +c Be
(b) van y? dx
0
4
: (d) It is not possible to find an algebraic
=n| Axi o2ih
0
solution to [cosx? dx.

2
a Exercise 2

1 (a) tsin3x+c¢

b) |@?-4)dx = 4a — 44 +c

ease he (d) In|x| +c

2 (a) 3x* cos x? (b) —4x sin 2x”

Beye Sra

(b) —4cos 2x* +¢

The volume of material is

32m — 25.6m = 6.47 cubic units.
698 SOLUTIONS

7.1.3 Integrating (b) |sin 3x sin 7x dx

trigonometric functions
= | $cos 10x + 5cos 4x) dx

PAD = —Fsin 10x + ¢sin 4x + c

1 —cos(A+B) = —cosAcosB + sin AsinB
cos(A—B)= cosAcosB+sinAsinB
—cos(A
+ B)+cos(A
— B) = 2sinAsinB

sin(A + B) = sin Acos B + cos Asin B

sin(A — B) = sin Acos B— cos AsinB
sin(A
+ B) + sin(A
— B) = 2sin AcosB

2 (a) 2cos Acos B = cos(A +B) + cos (A — B)

LetA=B=<%
= 2cos’ x= cos
2x + cos0
7.1.4 Integration by parts
=cos2x+ 1
7.1 Exercise 4
(b) 2sin Asin B = —cos(A +B) +cos (A — B) 1 (a) xe —e* +c
Ket A= B= <<
=> 2sin? x = —cos2x + cos0 (b) 1x _ 1% +c
=-—cos2x+1 (c)
1
Ne See
1 oak

a
(c) 2sinAcos B= sin(A +B) +sin(A
— B)
EetAc— B= 2 (a) xsinx +cosx+c
=> 2sinxcosx = sin2x
+ sin0
; 1
=i 2< (b) ~ sin3x+—cos3x +c
3 9
aa 1
(c) —sinax_+
— cos ax + ¢
a a
7.1 Exercise 3
3 (a) [Petar = ate — [ret dx
1 (a) |sin cos dx = [sin2d = x*e* —2xe*
+ 2e* +c

=—cos2x +c (b) |?sinx dx = —x* cosx + |200s xd

= —x* cosx + 2x sinx

(b) |sin 3x cos3xdx = [sin6x dx
+2cosx+ec
ered!
=— 7 cos6x+c
Av(a) Put wed
dx
d
2 (a) [cosSos xd =|(}cos 6x + 4cos 4x) dx and < = sin2dx + v = —1cos 2x
dx
9 0
xsin2x dx = |-1x cos 2x
77sin 6x +1sin 4x +c =<Al 2 =i

als 1(—}cos 2x) dx

4sin2x] 0=4
~ 0 +0.20807 + | = 0.44
 CALCULUS METHODS 2 699
RR

y=xsin2x
a 7.1.5 Integration by
substitution

7.1 Exercise 5
1 (a) PCE ans iSe,
dx

The area found is shaded in the diagram. [ufdu = Lu +c

When x = —1, xsin2x & 0.91
=}(x+3)%+e
The shaded area is close to that of triangle
AOB ® 3 x 0.91 x 1 = 0.455, which (b) Pg er Rake
indicates that the answer is reasonable. dx

fies) uotdu=1 |e? + su) du

(b) ee
dx EAE Se See
and d
de AOS 8 trae 9 GONE
= 374 +3g4 +e

0
=4(2x-
+5 (2x—-5)
5)%+e
| 2x eo ™ dx = [2x x 2e°*]?,
3 (c) [(+ 2)? u’du = fe + 4u® + 4u’) du
0
— | 2 x 2e°*dx
3

~ 0+ 2.6776 — [8e°*|°,
= 2.6776 — 8 + 1.7850 © —3.54
2 (a) tsin?’x +c (b) —tcos’ x +c

3 du 2
Sma) eta— in ol a
dx

2 Viet +3)dx = | Je +3) 5a

ba
du
= [jutdu = 300 +3)! +c
Sf

(b) 3(x— 3)’ +4(x-3)°+c

The area found is shaded
in the diagram.
du
The minimum value, —1.47, of 2x e° is (OQ) Via 3 SS = IL,
dx
between x = —3 and x = 0.

An estimate of the integral is approximately |=V(e -2)dx = 34-2) Rin+$(x-2)?+6¢

Nie

—2} x 1.47 © —3.67 (negative because it is

below the x-axis), which indicates that the (d) tsin* Core
answer is reasonable. du
(c)) Let uv ="cosx = ——— _— sinx:
dx
[costxsin ds =-— « du

6
1 Ue
6 aG

—tcos’ x +c
700 SOLUTIONS

en ee
du
(f) Letu=x+2>—=1. 7.41 Exercise 7
dx
3 MD 6 3
1 (a) = (b) a
Ce sate Ga) x+3 x«>+2 2x+1 x+1

3 1 5 2
= d
5 ate (| ee
7.1.6 The reciprocal function 3 2 2. 1
(e) 2x+1
= 3x+2
(f) Peah.
— Ge
7.1 Exercise 6
2 (a) 2 3
1 (a) 3In|x
—2) +c (b) 3In|2x+7|
+c
as eee
(c) ¢In|3x —1)+c (b) 2In|x +3|+3In|x—4|+
2 (a) $1n|4| — 3In |1| = 0.9242 (to
4 d.p.)
p x t3xt2+ 5x47 _ Sx +7
(b) In |—5| — $1n |—8] = —0.3133 x? + 3x +2 a x? + 3x +2
(to 4 d.p.)
(|
xt xt2
3 The function is not defined for x = 2.
: 2 3
4 (a) |ee In |cosx| + ¢ 1 d
cosx j ( fee |‘
1
= In|secx| +c = |x+2In|x +1] +3]n|x+2)),

(b) |oral dx = In |sinx| + ¢ =3.6027 (to 4d.p.)

sin x

5 (a) —4Inj4—x*|+e (b) In|34+2x7]+c¢ 4 (a)

(a) In esi? (x -2) c

6E (a) 4In|1+x*|+tan} x+c = 2

b) 41
(b) In|1 + x7| —tan x+e oes 2 ee a

(c) In|Inx| +c
(c)
Cae
, ames
(d) —In|1+ cosx| +c

Arial fractions

| 1 The numerator of the partial fractions will

; 2 : ag!
contain an x” term which does not exist in the
original fraction.

3 A=3,D=-LE=—
WIE
 CALCULUS METHODS 2 701

7.2. Polynomial
approximations
7.2.1 laylor’s first
approximation X4 X3 X27 x1 x

2 Exercise 1
a 3 (a) a
Pea = 35? 5
dx
The gradient at x = 2 is 17.

Te
$2
re ee
WADI) Acre es
(b) y=x+1

(c) y=x-1
(b) From the graph, the root lies between 2
(d) y = 28x — 44
andes
(e) y=3x-2
Any value greater than 1 will work, for
example x, = 3.
2 (a) y=—x+ 51
(c) f'(x) = 2x—6e™*
_ (b) 0.070 796
x —6+6e™
(c) 0.084% Xn+1 = Xp
DX hae OC
(d) x)= 3 x)= 2.421406 x3 = 2.329530
x4 = 2.326890 x; = 2.326890
7.2.2 The Newton—Raphson => x =2.3269 (to 4 decimal places)
bi %
mernod

JA
| 1 (a) f'(x)
=2x —3cosx
2 ;
a —3sina
b) 6b =a —-————__. X2 X3 x1
(b) ‘ 2a —3cosa

2? — 3sin2
= =—2 = (17576
Cait 2x2—3cos2 i The Newton—Raphson process will not
(Remember to work in radians.)
converge to the nearest root.
(Gimme — le 122,89 IS)
SE (a) x = 3.8730 to 4 decimal places
(d) Now take a = 1.722 8915
giving b = 1.7221255 (b) All values greater than 2.87 will give the
then a = 1.7221255 and b = 1.7221251 root 3.8730.
and finally ¢@ = 1.722 125 1
gives b = 1.7221251 (c) Any start position between —0.67 and
2.87 will give the root x = 2.
Thus, two successive values agree to 7
decimal places and you can conclude that (d) (i) x, = —0.68 gives the root 3.87.
the root is 1.722125 (to 6 decimal places). (ii) x, = —0.69 gives the root —3.87.
702 SOLUTIONS

Newton—Raphson method to solve it. It

ee y = 30 - 15x — 2x? + x3
does, however, illustrate how the methox
30
can fail to work even if the starting valu:
is not near a turning point. In this case,
a starting value must be chosen near the
root 4.
>
~3.87 2 3.87 x

6E The graph looks like a standard sine curve,

You cannot always predict to which root but when you zoom in you see that it has
an iteration will converge. If the starting several small turning points very close to
position is reasonably close to the root, each other. To solve this equation you must
you will usually home in on the root have a starting point very close to the root
very quickly. However, if the iteration you are trying to find.
takes you near a turning point, then the
method becomes very unpredictable. In
this case, the method takes you near the
turning point at x = 3.
7.2 Exercise 2
f(x) &

(Using nc ee with x, = 1,
USrasines:
then x, = 0.750364... and x = 0.739085...
You can see that there is only one solution by
plotting the graphs of y = x and y = cosx on
the same axes. They will only intersect at one
Scosx-x=0
point.

From an initial value of —1, the iteration x? = 3x —1=> x & 1.532089, 0.347 296
quickly converges to the root —1.977. or —1.879 385
However, a starting value of —0.5 is near
a turning point and the iteration
x = 0.487 404
converges to the root —3.837 and not
—1.98, the nearest root. Starting values
of —7 and —6.5 are some way from any The area of sector OACB is 167.
of the roots, as well as being near Triangle OAB is isosceles and has area
turning points. In each case, the root r Ded 1 pene
sin; @cos5@ = 5r sin6.
which is eventually reached is difficult to
predict and depends upon the accuracy (0) oO

of your calculator.

(f) f(x) rcos 30

B A U A
The function is rsin 30
undefined at 6

Hence the area of the segment is

Way
297%ke = LP) Ao
—5r sinrey i
= 4aree?
> 60-—in=sind
6 = 1.627 to 3 decimal places

This is an equation which is easily

(a) 207° — sin20
solved by simple algebraic manipulation
and you would not normally use the (b) 6 = 0.8832
 CALCULUS METHODS 2 703
———————————

6 (a) (i) Jerry Hence

(11) Tom — exponential growth will
/ " cs (3) x
always outperform quadratic growth p(x) = p(0) +p (O)x + p > Pp >
eventually.
4 ca
(b) 19.03 years, i.e. 19 years 0 months +p OF
2
x
(a) p(x) = p(0) + p’'(0)x + p"(0) TI

7.2.3 Maclaurin’s series 3 5)

+p Oi) (O)+P
ag ERG,
Os as Os
OF

x2

p0) =12254 =12 (b) p(x)=p(0) + p'(0)x + p"(0)el

=
DO pe pe
SO)y 10 22:40
p10) = 6>cd= 6 ua ae
nage (4) Breck
x

Hencea=12, b=11, c=5, d=

A pe(O)s=
n a

nh:

If p(x) = a+ bx + cx* + dx?

p(0) =a =>a=p(0)
p(0) =b +b=p'(0)
P.O) = 20 = = ©) 2
p80) =6d > d= p)(0) +6

Hence Pp os
p(x) = p(0) + p'(0)x + p"(0) + (0) St(O)pea
(O)-=2
fa(0) o=e2-
w If p(x) is a polynomial of degree 4, then
£90) =23
p(x) =a+ bx + cx’ +. dx’ + ex’
=> p(x) =b+2cx + 3dx* + 4ex?
2c +6dx + 12ex? The polynomial
os
)=& | - 6d + 24ex
~~
o Aw ox!
p(x) = 24e Lie Deca e 3
i3
will pass through the same point with the
same gradient, and with the same second
derivative, third derivative and fourth
derivative, as the function e** does when
se = OL

Hence it is a good approximation to e** for

But 24 = 4.x 3 x 2 x 1 is usually written as 4! values of x near x = 0 and this is confirmed
(called ‘factorial four’). by plotting the functions. The approximation
So e = p“ (0) +4! gives a percentage error of within +1% for
In a similar way, d = p(0) +3! and —0.45 < x < 0.64.
= p'(0) +2!
704 SOLUTIONS

1x" 2x?
(e) In(i pa) = 05" Ia 7 see
Vat (30 eos 5, a= it(Oi oe f
beer ae tbta SOUR tan
i") = eee Fe 4! 5!
fe) sinx =f (O)i= ce > 1
£ (x) = cosx = fO(0) = moran x cate)
and the cycle repeats itself. an See >< Il nal
i ‘ 4) 4x3 42x14
=
es =1+0 ae gene)a 2 pigOh!
cl
Fr Ree
2 SS)
0. =—1 :
es us Sues gives
5! 6! at 3 ee 5
a In(1 a x
sap cra aalghs dee eeture” 5
an Al= Gl
| 4B (a) n(1t+x)"~!
(Dit (lense = at (0) = 1
i () =< = £7(0) 11 (b) £"(x) = n(n — 1)(1+x)"~*
and it is clear that all derivatives will take
f(x) = nln — (m2) +x)"
the value 1 when x = 0. () (0) = 15 £0) =n £"(0) = n(n—-1)
£3)(0) = n(n — 1)(n — 2)
xspy LS %a
> Sat ay,
MET “a ee n(n — 1) 2
3! (1+ x) Sen) pee apes x

| 2 (a) In0 is undefined. n(n —1)(n—2) 3

y=Inx

>

7.2 Exercise 3

1 a) (is Pit (neh ie ae oft

(b) Since £(0) cannot be found, it is not , 1a 1
possible to find a Maclaurin’s series. Ose td (0) » Ea 4 ec
Geometrically, it is not possible to
approximate a function at a point that is oo Leia ee
not defined! ! all
anes
Seb
2 a 4 2! + 8 3! “<
So

Bees
1 Be
3 Ie‘(x)= Sor =1
eat
(b) Replacing x by 4x in
(b) "Gy = 101 ha) ee ea oe
(c) £9(x) = -(-2)(1 +x)? = 201 + x)? a
£4) (x) =2 x —3(1+x)7* x\?* x\?
= —3!(1 + x)o* - Ly S 1 Xe a 2)

£) (x) = 41 +x) Ene FT oe

(d) 1 = S f’(0) = 1; i) =; _i.% x uk?
£9)(0)=2; 2 £90) =-3!; ¢ (0) = 4! == {eo
7. Bi is
ag
 CALCULUS METHODS 2 705

2 (a) 0.877 582 5619

0.57
baad “~y “a " : 3 q

/.3.1- Introduction

fen Ow : Mae Exercise1

a 0.8776 oo yee
x
aeil
A (a) yy "eG va ae
ye
so three terms are needed.
Vyeaw =I
y=e"-2
(c) cos 1.0 = 0.540 302 305 9

aitay Odd
>
1 : : = 0.541 67 x

1
re
L J
tar! =ae0.540 28

sO four terms are needed.

(b) y =F Sin 2x re
x
n—-1 i
xX
2n—2 y
3 ae a b 1 n—1 y= sin
2x +1
ray. Ohare oe em y =3sin 2x
y =1lg
= 7sin2x = —41
2n—2
(c) Seow
(n— 1)!

4 (a) Replacing x by —x in the series for

In (1 + x) gives

iS 3 eS
Se
| 8 iS) I
—
In(1 —x)(14+ x)

= In(1—x)+]n(1+x)

2 (a) y=x* = 1 When

x = 3, y = 80
(b) y=x?
—x? +2x-2
When
x = 3, y= 22

(c) y=3- x?
wile
When x = 3, y=3

(d) y=e*—e? When

x = 3, y & 396
(Se ane
ul —
canary eee

2t 2

For small values of x 3 (a) y= 81- > +32

(—-Lee<i1),e° s1—*x
so when t = 4, y= 534
(b) /—e*) & V1 -(1—x)) = vx
d
fe Agr hee eA Gente aes te 00
(ec) 0b = 03162 dt dt
hes) = 0.3085 and so the water is no longer being
=> percentage error = 2.5% heated. The model is probably no longer
valid.
706 SOLUTIONS

4E (a) y = —}cos (3x + 2) — 0.139 (to 3 d.p.)

dy
Solution curves for — = x — y
dx
When x = 2, y } —0.090

S
(b) y=te* -1 = Whenx =2, y ~ 993
(c) y= +sin (x? +1) — 0.421 (to 3 d.p.)
When x = 2, y % —0.900

7 ee OO recionwdiae rains

As x increases, y tends more and more to the

value of x. (The line y = x — 1 is an
asymptote to each solution curve.)

xv

7.3 Exercise 2

1 The curves are a family of ellipses.

2 The curves are similar to that of the Normal

distribution in statistics.

LX : x

If y is initially negative, then y increases as x

increases. However, y never becomes positive; 3 m represents the mass of reacting substance at
it just gets closer to y = 0. time tf.

m
d “Gz TePresents the rate at which the mass is
_ 3 The direction diagram for ee y is
shown below. dx changing, i.e. the rate of reaction.

d dm
(Note how = = 0 for all points with y = x.) Pie —km represents the fact that the rate of
loss of mass is proportional to the mass.
 CALCULUS METHODS 2 707

4 (a) y=x 41

0 t
0 1 2 3 4 5)

8 is the value of m at time t = 0, i.e. the initial

mass of the substance.
(b) With a step of 6x = 0.2,

ED) Ges 182: O22 AO ee SLO)

7.3.32 Numerical methods
oe
y 2A 2 SAA VAS) 156. Oecd. (oss ueFO)

7.3 Exercise 3 With a step of 6x = —0.2,

(a) Sy ssinie
x 1 0.8 0.6 0.4 0.2 0.0 —0.2 —0.4 —0.6 —0.8 —1.0
YZ 1-63) 1 ONO595) OFS 0:87 0,0 unl.Olme le oem eG
(b)
x OS OS 0m 5210829 3:0) 3.9) 4:0) 94.5 520
Numerical y 0 0.5 0.9 1.0 1.0 0.7 0.2 —0.3 —0.7 —0.9 —0.9 SE The solution to the equation is y= e* +k.
Exact y 0 0.5 0.8 1.0 0.9 0.6 0.1 —0.4 —0.8 —1.0 —1.0
The particular solution through (0, 1) is
V=He.
From the table you can see that the The values for a step size of 0.1 will be
numerical and exact values appear to (to 3 s.f.):
differ by at most 0.1.
x 0 OA 02103 0:4
OO TOMA 2 eS Sara Gee and so on.
(a) yex*t+eosy=x'-1
Wihtenexi—
52) —ail5)

(b) To 1 decimal place, successive points are

(ARO) G@e2= 058) (dey 22) \(k6s-4-45 6E
(LE SHW/ai)) sana lees) x O05) 80 Ss 20S 2 so Ones 5. 420
Wiheniai—=ny 1223 y OOS OSs SO. ree 2am Sole

3) (2) 90 —icos 2x
x 0 —O05 —1.0 —1.5 —2.0 —2.5 —3.0 —3.5 =—4.0

(b) An exact solution is — 5cos 2 = 0.208. y 0 0.5 —0.8 —1.0 —1.1 —1.2 —1.3 —1.3 —1.4
The numerical solution is 0.160.

The percentage error is

0.048
x 100 & 23%.
0.208
708 SOLUTIONS

dN
4 The differential equation is ae XN.

N is the number of insects and t is the time ir

days. |
dN |
When N = 100, sap et a
dN |
—05N>3>N=Ae™
dt
When f=10) N==100/=sAe—s
100
Therefore N = 100e°*' and so when t = 10,
N = 14841
There will be nearly 15000 insects after 10
| days.

The shape of the ‘turned round’ graph Be:

suggests the tangent function. 5. (a) ae Am
The relationship is y = tan”! x. When w= 0.020

dm
ae : =—0.0 = —0.05
+ A01
7.3.4 Growth and decay t
peng

d
The differential equation is oP = 0.
- dt
7.35 Exercise 4
Ppa
( ) dt
Vp m ores m Pree e ba
Beaty
1 y= Aen

y= 2 when — 0; therefore Al— 2 A is the mass when t = 0, so

ee A = 0.020 > m = 0.020e-0°

When m= 0.010, 0.010 = 0.020e °°
Dla) SA ea Sena Der |
=> t=—201n0.5
(b) i= 2ee
Pe ine — 15:86

The substance will decay to half its mass

=> —0.1t = 1n0.5 after 13.86 days. This is called the half-lif
=> —0.1t = —0.693 of the substance.

a t & 6.93 hours

Itminutes.
takes approximately 6 hours 56 7.3.5 FormulatingSe differential
equa tions
3 Assume Newton’s law of cooling.
y is the temperature in °C above room oie :
temperature, ¢ is the time in minutes after /.o Exercise 5
boiling. 1 (a) For a snowball of radius rcm and time ¢
dy At days
ee AY = ee
dV Pees
StS 9 —4krr Sia hea —k
t=0, y=80> A=80
t=5, y=70370=80e" (b) r=c—kt
1 r = 30 when t = 0 and r = 0 when t = 1
=>.=A In au )= SWLW2Sr
») 80 so r = 30 — 3t
The water cools to 60°C when y = 40. (©) GS 30237— 5
40 = 80°67! _s + — 26 minutes The radius will be halved after 5 day:
 CALCULUS METHODS 2 709

30
(11) pe a sage el The simultaneous equations give
X = 0.015 35, ¢ = 1363:
The volume will be halved after 2.1
days. t 1303 — 130.3,/V
When V=52, tw 363
2 (a) 0.1 = 10%; —0.1 is used because there is a
reduction of 10%. a = 2000. It takes 6 minutes 3 seconds.

(b) y is the number of fish and = is the

change in the number of fish per year.

7.3.6 Separating the variables
(c) About 16 years
dy
d) ==
(d)=,=2500 =.
—0.12y 73 Exercise 6

dy >
(a) Saas
' 3E V is the volume of liquid in the urn in
cupfuls.
t is the time in seconds. [5ay=| xae
y
dV In|y| = 4x7 +k
The rate of change of volume, —,, is
dt
proportional to the square root of V ets el +h)
(assuming that the urn has a uniform ve Ae™
cross-section).
dv (b) p =t?+k (c) m= Ae”
So — = —A,/V, where 2 is a constant.
dt (d) y =2e*+k (ec) y+ 1=k(x+1)
Numerical method
2 y=2(1-—e ”)
dV 9
When V = 100, ar yy Zp Cups Per second
9 3 14800
=> —-— & -),/100
60
=> 0.015 4 (a) 0.02e2 — (b) 13.9 days
You need the solution curve through (0, 100) (c) 27.7 days (d) 6.07 x 10°“ per day
and require t when V = 52 (when 48 cups
have been filled). 5 1.03 x 10’No
With a step of dt = 1 (second), this occurs
after 6 minutes 12 seconds.

Analytical method
Miscellaneous exercise 7
dV
Sa Sp aay
dt AV 1 18x

Baia tls
dV me A/V 2 (a) 4m (b) 45.6 ~—(c) 17.4
2
=> ¢ ae 5 Vetee
Sea) 1° (b) 142
When t=0, V = 100 and when
60-5 Ve= 91.
710 SOLUTIONS
we ce

a
4
1
2,
(Cee eee 16 (a) 2xcos (x? +1) (b) 2 sin (1 — 2x)
1 x in x
epeenc (eb re (Rveure (d) (cos)e°
- (sin
x +cos x)
1 — sinx)-€
e) 1sin® x +c (f) eee )

17 S=Ae™; 57000
ay (e)) ay) GOS SP SRS Gl) # =—4()() ag
18 (a) y=10e& (b)) y= 31297 ene
6 (a) —x’ cosx +2xsinx +2cosx +c len 2
2x -F 1
(b) x? sinx + 3x? cos. x — 6x sinx
—6cos x + €

(c) 1x? — ix sin 2x — §cos 2x + C

8 (a) A=1,B=2
(b) A=—1,B=2,C=-1

9 (a) 7In|x —3)/+2In|x+1|

+c

+.
x—4
(b) In

(c) —$ln|x +4] +2In|x — 1] +c

(d) x+In| x— 2| + == a4
x+3 .

10 1.53, —1.88

11 Exact root
= 0.739

ZeS3S4

13 (a) 1—x+4x°-12°
(b) 1+ 2x + 2x7 + 42°
(c) 2x — 2x7 +8’

14 (a) 14+}x-4%° |x| <1

2
(b) 1—x+x pelt
(c) $-—4x44x lool

i5ei(a) x+4x°

(b) 1+4x°+5x'
(c) xtx° +4?
The equation of a straight line
This tasksheet is intended to introduce or revise the concepts of gradient,
intercept and the equation y = mx +c.

The tasksheet explains that the equation y = mx + c represents a straight line

with gradient m, passing through the point (0,c), the intercept with the y-axis.

As well as noting how to calculate gradient precisely, it should be observed that:

& TASKS
1.1
S1
e@ a numerically large gradient means a steeply sloping line;

e@ a numerically small gradient means a flatter line;

@ a gradient of zero means that the line is flat or horizontal (parallel to the
x-axis);

a positive gradient slopes upwards to the right;

a negative gradient slopes downwards to the right.

(a), (b) y=2x+2 Varying c translates the line up and

down. The value of c is where the
line crosses the y-axis.

Varying m changes the

gradient of the line. When m
is positive, the line slopes up
as x increases. When is
negative the line slopes down
as x increases.
 BC a4
Gradient = ne = )

S1
TASKSHEET
1.1

11-2 9 :
4 (a) Gradient = —_—_— 3 Gradient =
2+1
¢ 0
(c) Gradient = Sag

4
(a) Gradient = 2
y
(b) The line crosses the y-axis at (0,1) so the y-intercept is 1.

(c) When ==0 When

x= 2
Meo 10a y=2x2+1
y=1 y=5
(d) For'y =2x-+1

wh 1 represents the y-intercept.

2 represents the gradient.

Choose any two points on the line, for example (—1, 2) and (—3, 8), which
give the gradient —3.
The line crosses the y-axis at (0, —1), so the y-intercept is —1.

For any point on the line, substituting the x-coordinate in the equation
will give the y-coordinate; for example, take the point (1, —4).
Whenx = 1
(d) For y= —3x-1

x—1 represents the y-intercept.

\
—3 represents the gradient.

(a) The equation of the line is y = —2x + 6 .

es
gradient y-intercept
TASKSHE
1.1
S1
(b) The equation of the line is y = 5x — 2
PALES
gradient —_—y-intercept
 Multiplying brackets Commentary
5x +15 (b) 2x —8

16x + 40 (d) —2x —12

—4x + 28 (f) 6x —12y

Ore
O= Le
lad
od
wf)
Oc)
Wao)
By
jy 3x
= 12

a+ 40—-—Sa=40—-—4a
t—4+4+4¢=S5t—4
p—-1+6p—16=7p—17
5 — 30x + 54 = 59 — 30x

y — 9y + 18 = 18 — 8y
(h) 4x —2x +27 = 2x4 x?

2 — 3x — 6x"

x + 6x +8 b
(b) x*—2x -3

x +3x—4 (d) x*—7x+10

Rn 35 (f) x? +10x +16
x’ = 11x + 18 (h) x? +3x—28

(i) x*+6x+9
(ii) x? + 14x + 49
(iii) x* — 18x + 81
(iv) x? — 12x + 36
(b) GF b= 2p
(ii) c=p 2
Further factorisation
This tasksheet provides a background from which a strategy for factorising
x” + bx + c can be developed. The relationship between the constant term c and
the factorised form is examined and this relationship is used to help select
appropriate pairs of factors.

1 (a) (i) x7 +5x+6 (ii) x? — Sx +6

(iii) x? + 9x + 20 (iv) x7 — 9x + 20 S3
TASKSH
1.1
(b) The constant term c is obtained by multiplying together the numbers in
the brackets, together with their signs (+ or —). For example,
(x = 2)(x +3) =x? +x-6
je
ee
os a=
The coefficient b is obtained by adding together the numbers
brackets. For example,
(x — 2)(x +3) =x? +x-6
|
tae
ie 1
3-3
(x + 2)(x +7) | (x + 8)(x+5)

(x +6)? (x — 8)(x +1)

(x — 6)(x — 2) (x — 9)(x
+ 4)
(x.+ 6)(x — 4)

x(x + 2) (x + 3)(x — 3)
Not possible, x(x + 25)
Not possible (x + 1)(x — 1)

Gi) x1) Gx 1) (2x + 3)(x

+ 2)
(m) (3x — 2)(3x + 2)
 Commentary
ron

| Qo
DN
who
niw

Qo}
Working in surd form Commentary
1 (a) Uw =/2 S/O Hae 4

2 (a) 144+ 6,/5 (b) 17 — 44/13 (c) 29 — 12,/5

‘(d) 45—29,/2 (e) 5-2/6 (f) 120 — 30/15

5 5/21 _ -7£/85
4 (a) x= 5 (b) x == VS (c) x= ; TASKSH
$2
1.4
16/2
S (a) 2V2_—(b) 2VIS_() 4Y2_—d)
(i) +19 (ii1) 1

1+ </2+/3+ 6 is aylOde 15
(b) = D 5
2h Sey 2
(dj 4/10 24,2 <5 = 4 )
es
42 2,/3
Smal 2a Oto rey) Pad)
3
 Expanding brackets Commentary
1 (a) x —-1
(b) x(x? + Dw + 4) —2(x? + 2x +4) = x) + 2x? + 4x — 2x? — 4 — 8
x — 8

(c) x +1
(d) x°4+8
Gc)
oe
Wa)
I2
lo)
Be)
2 (a) (x+1)(x* — 2x — 8) =x? — x* — 10x —8
(b) (x —2)(x* — 7x + 12) = x° — 7x” + 12x ~ 2x* + 14x — 24
x? — 9x” + 26x — 24
(c) (x% — 1)(x? + 6x +5) = 2° + 5x* —x—5
(d) (x* — 2x +1)(x +3) =x? — 2x* +x + 3x7 — 6x +3
x tx? —5x+3

BG) (2 1G? 4) x = Se
aye
(Dy Ge 4)* = x* — 8x* + 16
(c) (x —1)(x +3)(x* + 6x +9) = (x? + 2x — 3) (x? + 6x + 9)
= x7 4 6x? + 9x? + 2x? + 12x? + 18x — 3x?
=1xo7 |
x* + 8x° + 18x” — 27
(d) (x + 1)(x — 2)(x* — x — 12) = (x + 1)(x> — 3x” — 10x + 24)
x* — 2x? — 13x” + 14x + 24
 TASKS
3.1
S1

3x41 (3x + 1) i C=Geen

ese fg(30 :
Os ee
Bee Sone
=(x+5)-9=x-4, fe(x)=(x-9)+5=x-4

matter.

In (iii), the order in which successive addition and subtraction is carried

out does not matter.
In (iv) and (v), the composite function is in both cases the identity
function. Since functions f and g are the same in each example, they are
both self-inverse functions.

3 (a) x+2

4 (a) x+8
 0)
R earranging
ing formulas 23= =< = sa) fon>>
ao]

i Wee

a (ey)88
a ven
One solution is x = 41.4

The other solution between 0 and 360

1s = 360'— 41-4 = 318%

COS ® =05565 = xe 95.9.) S04 TASKS

3.2
S1
— 025 ea — 19595. 346.70

—J = 1205 240

Sine = 0:65 2=a 10 9S

cosx =—0.38 ~S -x1= 1123, —112:3

sinx® = —0.47 => x=-—28.0, —152.0

2sinx = 2-5 41 35, 1382

SCOSK at ae— 0. x — 1526, 2464

(G)) 2cosx, 5 = 0 No solution for x

5 sin (3t + 40)° =4

=> sin (3¢+40)° =?
=> sinx® = 0.8, where x = 3t+ 40

From a calculator, x = 53.1

x= 53.1, 126.97 413.1, 4869, 773.1, - 846.9

t= 4.37, 28.97, 124.37, 148.97, 244.37, 268.97

Citcinor = Ope G78) 200) 478

(ye cosro = On 100 140) 9220, 260,
(ec) 3cos (0.52420) =2 => £= 56.4
=
Wn
‘ae
aa)
jee
an
7)
na
Y)
<
i
oa
on
Differentiation practice Commentary
ye
1 By the chain rule: a 2 sinxcos x
x

By the chain rule: ay —12sin 4x

=I - or
TASKS
6.3
S1
By the chain rule: (2x +5)"

=—2(2x+5)? or

; oe 5 Bs. d 6
Using parametric differentiation: eee
dae "227,

Using the product rule and the chain rule: =e” sin3x +4 ea cos5

é7 2% (9 sind x:+ i5cos 152)

(2 sins

By the chain rule:

By the chain rule:

d
Or:. y=In4x =In4+In~x. eS
dx

dy —2xy -—2y
B y implicit
i differentiation
li it diff ti ti 5 ee =(0 toe
SS Ss =Ss -

Or: beh
36 36x *: =a
=)

dy 3cos0
By parametric differentiation: Ot OS SIGOEU.
dx —4sin0

Using the product rule and the chain rule: dy = (2x — 3)* + 8x(2x — 3)?
dx
 dy —6sin39 —3sin36
13 By parametric differentiation: de. Debsoh ee

14 By implicit differentiation:

15 By the chain rule:

S1
TASKSHEET
6.3 Dh)
Vv(Sx)
e ee 1
16 By parametric differentiation: = aris of —3h)

d d
17 By implicit differentiation: 2x + 3y + 3x eee+ 4y sea)
dx dx
dy) a 3)
dx (3x+4y)
dy _cos2xcosx +2sin x sin 2x
18 By the quotient rule: Ee
y qu dx cos? 2x

3% dy _
19 By implicit differentiation: 3e"y+e
dx

dy
dx
D
x
y = x and use the quotient rule:
e
dy Dx ee ax es ae 3x?
dx Aan eC om et

20 y= 12%)
 Qa aus,

a(uy +--+ Uy) TASKSHEE

1.2
E1

Sou1

(a) The sumis increased by 1 times the constant.

(b) So(uj +b) =u +b +m +b+---+u,+b

1
uy +-:-+u,+nb

Sn(n+1)+n

DP Naw
aa +5n

Wy ty ty tig te + Uy, tv,

(uy + uy + +++ uy) + (oy +o, +--- +r)

Sane
1 1
 5 (a) (n+ i) — 1°. The other terms cancel out in pairs.
(b) This is part (a) written in }> notation.

() G41 —#® =37 +3i+1

n n

SC See = ey) = Fa
1 1

E1
TASKSHEET
1.2 SG? F371)
1

ee
1 1

(dy 3) 07 +3) i+n=(n+1)-1

1 1

“2. 3n(n+1) A 5
ee
BN DS: eee eS
| ey a +3n° +3n

. are, A) anonn
3 > t=n ae 5

ysOe
> (2n? +3n? +n)
1 6

So? _ n(n +1)(2n +1)

1 6

99 x 100 x 199
= 328 350
6

Me SON
25. ies
Dae lasers)
: — 3n(n +1) + 4n

_ n(2n’
=6n+4)
3

_ 2n(n =1)(n ~2)

3
 n 7

4\°P -4) itn

1 r 1

2n(n + 1)(2n + 1)
—2n(n+1)+n
3
n(4n* — 1)
TASKSH
1.2
E1
3
n(2n — 1)(2n +1)
 Zeno’s paradox
1 (a) Since the rabbit is running at half the speed of the dog, the rabbit will
have run 64m.

(b) By the same argument, the rabbit will have cun a further 32 m.

(c) 16m.

The argument appears to suggest that the dog will never catch the rabbit
TASKSHEET
E2
1.2 because the dog always has to reach the spot last occupied by the rabbit. In
the meantime, the rabbit will have moved on ahead again. However, you
know that, in practice, the dog will catch the rabbit!

This is a G.P., first term 128, common ratio ;.

p = 128(1N=— ta)
($)”)
2
as6(1 — .
(4)"
(b) As 2 approaches infinity, (5)" — 0 and so D — 256.

1
This is a G.P., first term 64, common ratio 5.

(a) R= 64x he 128(1— (4)”)

(b) As 1 approaches infinity, (5)” — 0 and so R = 128.

(a) 40 x 8 = 320m

(b) 40
x 4= 160m

(c) Since the dog will have run 160 m farther than the rabbit it will have
caught up with it.

The answer to question 2 suggests that the dog will never travel more than
256m. You know this must be false! The paradox is resolved when you realise
that time must be taken into account. The distances in question 2 are taken
over decreasing periods of time and so give a false impression of the motion.

The dog runs 128 m in 16 seconds, the next 64m in 8 seconds, the next 32 m
in 4 seconds and so on. The total time is therefore
16
16+84+442+-:-= ; 7 = 32 seconds
>
In 32 seconds the dog will have run 256 m, as predicted. However, the dog
does not stop moving after 32 seconds and will now overtake the rabbit, who
has run 4 x 32 = 128m and is level at this point.
Regular pentagons and the Commentary
Fibonacci sequence

360° + 5 = 72°
180° — 72°
>y = 54° (isosceles triangle)
mu,
=> interior angle = 2y = 108° TASKSH
1.4
E1

/ ABC = 108° (interior angle)

Since triangle ABC is isosceles,

IRCA = /BAC = a = 36°

BC = CD (triangle BCD isosceles)

=> (CD -1

CA=CD+DA=>DA=¢-1
But ¢ x DA = CD ;
oe)

(a) Draw and measure a line

segment AD= 8.1cm. This is
5¢ to the nearest 0.1 cm.
Use compasses to draw an
arc with a radius of 5cm
and centre at A.

Use compasses to draw an

arc with a radius of 5cm and
centre at D.
 Join the point of intersection of the arcs at E to A and D, to give two sides,
AE and DE, of the pentagon.

(d) Use compasses to draw an arc with a radius of 5cm and centre at A.

(e) Use compasses to draw an arc with a radius of 8.1 cm and centre at D.

Join the point of intersection of the arcs at B to A, to give the side AB of the
pentagon.
TASKSHEET
1.4
E1 Two 5cm arcs with their centres at B and D will give the position of C and
hence the other two sides of the pentagon.

RUS is an enlargement of BCD, scale factor ¢, hence SU = ¢.

i
6 UT is the edge of the new pentagon.

The new pentagon is an enlargement of the original pentagon, scale factor TASKS
1.4
E1
¢’, hence
x= xgad

7 g” =1(29+13/5) o° =1(47 +2175)

o =1(29+ 13/5)’ = 4(29 — 13/5)

g + = 1(29 + 13/5) +4(29 — 13/5)

= 1(29 + 29 + 13/5 — 13/5) = 29

B (¢” - W’) = == [4(29 + 13/5) - 429 - 13/5)]

=1(47+4+21/5) = 4(47-21/5)
oF+ w= 1(47 + 21/5) +4(47 — 21/5)
(47 447 4214/5 — 21/5)
47
(g — ¥8) = <;[$(47+ 21/5) — 4(47 — 215
=: )
1
= WE; iB 47 — 47 + 21,/5 + 21,/5)|

=21
 3 ¢" yf les=

1.618 034 —0.618 034

2.618 034 0.381 966
4.236 068 —0.236 068
6.854 102 0.145 898 7.
11.090 170 —0.090 170 11
17.944 272 0.055 728 18
TASKSHEET
1.4
E1 29.034 442 —0.034442 29
Ss
US
AS
Gs
GY
S|
eS 46.978 714 0.021 286 47 NR
RPWmoOmnNWN
Re

As n increases, ~” becomes numerically smaller and oscillates between

positive and negative values.

1.33355 175, 15571, 1.636.. Gite 162 13e.

feet nlisd s 660755912650 VILO25a Ons ae

LB,al

Both ratios get closer to the golden ratio, because that is the ratio of
successive terms given by the approximate formulas in both cases.
Tangents and normals
1
1 The normal line has gradient —-.
&

2.2
E1
TASKS

==%—3,
4
y=
=

(d) (1,3), (351) (-1,-3), (-3,-1)

(ec) The rectangle has edges of lengths $./2 and #,/2. Its area is 48 square
units.

(a) The tangents and normals must have gradients of +1. At the corners of
the square on y=x*, _
d
EEes eet ee
dx
The corners are (4,4), (0,3), (-4,1 )» (0-4).
(b) The square has edge of length 5\/De ilts area is therefore
G /2)° =4 square unit.

The normal has gradient —} and equation y = —}x +3. At A,

—ly 43x > 2x*+x-3=053x=1or—-3
A has coordinates (- 3,2).
 Optimisation problems | Commentary
1 (a) Ifx > 5 or x < —S, then the base of the cone lies outside the sphere and
the design constraint has been violated. x must be smaller than the radius
of the sphere if the cone is to be inside it.
The diagrams indicate that the volume increases to a maximum value,
then gets smaller again.

TASKSHEET
2.3
E1

The maximum volume appears to correspond to x = 2.

V =1nrh
Ife =2; then
ie eh = and mg
Voom
x 21x 7 = 154cm°

The maximum value of V appears to correspond to y = 3.

ye Ves 3(5° —x°)(5 +x) =40(5 —x)(54+-x)
Writing the equation in this form makes the graph easier to sketch.

TASK
3E1

dV
—=0=>x=-5 or x=?
dx ;
The minimum is at (--5,0). The maximum is at (1.67, 155.1).
The maximum volume is 155 cm?.

(g) Using y as the variable, the height of the cone / = 10 — y and the radius is
given by 7 = 5* — (5 — y)’:

V = $n{5° — (5 —y)"}(1—0y)= }(° — 20y* + 100y)

dv
ae }m(3y° — 40y + 100) = 12(3y — 10)(y — 10)

=0>y=2 or y=10

The minimum is at (10,0). The maximum is at (3.33, 155.1).

Note: 1.67 + 3.33 = 5.00, as would be expected.

(h) V= 1nrh, where r = /[sin 6, h = 1cos@

Hence

Va inp sin? 0 cos 0

 In finding this expression you have introduced a new variable, the slant height
!. This must be eliminated from the expression for V, so that the expression
for V is in terms of 6 only. This can be done easily by substituting
T= 10 cos 8.

(s
This result follows directly from the well-known
geometrical result that the angle subtended on a
TASKSHEET
E1
2.3 diameter is a right angle.

So

Vie 71000 cos’ @ sin? 6

which is a maximum when 6 = 35.3°, giving a volume of 155.1cm’.

Let the price per bike be £P. The number sold drops by 40 for each increase
of £1 in the price, and so the number sold is
5000 — 40(P — 100) = 9000 — 40P
Total revenue = £(9000 — 40P)P
Total costs = £50000 + 85(9000 — 40P)
Profit = revenue — costs = £(—815 000 + 12 400P — 40P7)
d(Profit)
= 0 when 12400 — 80P=0, ie. P= 155
dP
Number sold = 9000 — 40P = 2800

Approximately 2800 should be manufactured and they should be sold at a

price of £155 each.
Traffic Commentary

104(20 — t)
1 6 dt = 13313

Car A travels about 133 m during the 10 seconds in which it accelerates from
rest.

During these 10 seconds, car B will travel 20 x 10 = 200 m. So car B must be TASKS
2.5
E1
at least 67m from car A at the start if it is to avoid slowing down.
(In practice, rather more than 67 m will be required since it would be
dangerous for the cars to get very close to each other. The actual length of the
cars has not. been considered, nor has the fact that A initially has to travel
around a bend.)

If the traffic is moving at 20ms_’, then in one hour the length of traffic
passing is 20 x 3600 = 72000 m.
72.000
Th e average space bet
between cars Is1S ==
1200 = 00 m mi.

So car A is unlikely to find a gap of sufficient length if the traffic is evenly

spaced. The traffic is very unlikely to be evenly spaced and so A is likely to
find a gap without having to wait too long.
 eo : 5 9
Commentary
Amal
aa
bo
Derivative of sin” x
a
ea 1 (a) y
ae
Y
1

Y
<x
>

es

cosZ xi a=-}, a— i2

= —}(—2sin2x) = sin2x
d
3 —(cos’ x) = — sin 2x, using a method similar to that used in question 1.
dx

cos’ Hee x =5+5c0s

eB cds 2x => =x (ais
) =3(—jsin 2x) == — sinj 2x

— (sin? x + cos” x) = sin2x — sin2x = 0

y = sin2x + cos?x = 1

ee) 3) 3 é 5
(c) sin” x + cos” x = 1, a constant whose derivative is zero.
 NCOs
5 cos2x =1—2sin*x= sin’
x = ae
yD
dns Daerad cos 2x
qn sin )=5(- 5 )

Lis. ;(—2 sin 2x) = sin2x

cos 2x = 1—2sin’x

= 1 —2(1 — cos”x)
=2cos*x—-1

pee cos 2x 4 1
x= =
2 2
d ey beh eos2x
qn (°° y= = 5 )

= —sin2x
 Prime number formulas Commentary

1 It is worth spending some time in exploring particular cases and establishing

subsidiary results such as that if an” + bn +c is always to be odd then c must
be odd and a and b must have the same parity.

In the first attempt at a proof, the cases c = +1 and ac+b+1= +1 all

invalidate the method. It is worth remembering that any one of these suffices
E1
4.2 as a counter-example.
TASKSHEET
When
n = 1+ 2p,
an’ +bn+c=a(1+2p)>+b(1+2p)+¢
4ap’ + (4a+2b)p+tatbt+e
(4ap + 4a +2b+1)p
The second attempt is sound and the method may be extended to show that
no polynomial form P() of degree one or more generates only primes. If the
degree of the polynomial is k it is sufficient to show that P(n) takes the same
value for (k + 1) different values of n.
Extending the method
1 -(a) (i) x =rcos(9+a)
(ii) x =c—d=acos@—
bsin#é

(b) acos@
— bsin@ = rcos (6+ a)

+ « — + J>
m — ¢ ——>
TASKSH
5.1
E1
c—ad

asin@
— bcos@

Also y = rsin (@ —@)

=> asin0 — bcos @ = rsin (6 — a)

Considering the projections on the x-axis,

x = acos@+ bsin@
Also x = rcos (0 — a)
=> acos@+ bsin@ = rcos (6 — a)

V(49+16)=/65 a= = 297°
= J/(16+49)= 65. ay 60.3°
You can consider the function either as a sine graph with a phase shift of
—29.7° or as a cosine graph of phase shift +60.3°.
a, + @, = 90°. This relationship may be seen clearly from the triangle.

Gi

21)
if

(ay 5nd (0 — 7317). BN (osin (0 2 22.6°)

4isin(@—77.3°) (a) 4.47sin (6 + 30°)
3 sin at (b) 3cos oS

(ike
3sin | 0+ (d) 3cos
6
 Intersections Commentary

1 (a) *ar=

TASKSHEET
E1
5.2

y= 2 = @ p> 3x-y+2z=8
5 ON a)
(i) <=2-3A-2w @
Si Oren @ >>3x-—y+2z=8 as before
z=2+4A+3u ©@

3x-— y+2z=8 @
x—l’y+ z=1 @

lf = Oe 2560)
=O npan—5.5 oy 2
lf'y= 0% "OreOne = —5, X= 6
so two points are A (0, 2, 5) and B (6, 0, —5).
(There are many others and many ways of finding just two points!)
6
Hence, a vector in the direction of the line of intersection is AB = | —2
. . . . . . . =,

10

x 0 6
and the equation of the line of intersection is | y 2)}+A| —-2
& 5 10)

At the point of intersection, the point (x, y,z) satisfies the equations of both
the line and the plane.
eax Tor
So, 2x +3y+z2=7 > 2(3t) +3(2—21) + 5—St=7
=> 2

Thus the point of intersection is (6, 0, —5).

The point of intersection is (5, —1, 2).

Ma Die al @
Save 22 Oar) .2
TASKSH
E1
Axi SN A 23k SamnO) 5
2X) = OF 5x 4-32 = 15 @
3 xOi— 6) = Sx 32— 19 6)

But equations @) and ©) are inconsistent — it is not possible for 5x + 3z to

be equal to 15 and 19 simultaneously.
Using the method of question 2, the line of intersection of @ and @) is
x at — 35
y O0| +A 1 (taking points (2,0, —1), Gs;,0) on the line)
5
z —l 1
and the intersection of @ and @) is
19 21
5
i+) a (taking (2,42
,0) and (0,#, ®))
o| Le 3
But the direction vector of each line is | —1
=)
Thus, the three planes intersect in three parallel
lines — a prism.

If the third equation is 4x — 3y + 3z = 9,

3x @-@ = 5x + 3z = 15, which is the same
as @) above.

In this case all three planes intersect in a single straight line.

 Commentary

16 14 137 Lp
0.65 0.35 0.125 0
0125 0.35 0.65 1

As 6 increases from $7 to 7, the values of x become negative, but equal in

magnitude to the values in the table. The values of y remain positive and
equal to the values in the table, i.e. x decreases from 0 to —1 while y
decreases from 1 to 0.
For 7 < @< 3r, x and y are both negative, while for nr <0 < 2n, 1s
positive and y is negative.

as :1
0.02 0.18
0.13 0

As @ increases, x will continue to increase, though not steadily. y will oscillate

between 0 and 2 (since 1 — cosa = 1 — (—1) = 2, after which cos @ increases
again).

The resulting curve is a cycloid, the path taken

by a point on a moving circle.
 i.e. the curve cuts the axes at (0, 3) and (— 5,0).
(b) Ast—-2°, x—++400, yS2 -(t—=32> meanst approaches2 from
below.) i.e. y = 2 is an asymptote.
(c) t — 4 also yields an asymptote since it makes the denominator of y zero.

Ast — 4", x—+—3 9 J CO

ie. — 3, CO

(d) Whent=0, x= $s s (It may be necessary to find a few more

points to increase your confidence.)

C=) EO
Ast
— 1", x —-=00, whoo
AS 1 5) 60,8 oe
Ast oo, x—-2, y—-oo
Ast—-—coo, x—-2, y—-+00

Again, plotting a few points will help.

 Commentary

e* sinx— | e* sinxdx

e* sinx — |—e* cosx + | e* cosxdx

e* sinx +e’ cosx —I

TASKSHEET
7.1
E1 e* sinx + e* cosx

= } e* (sinx + cos x)

du ; dv . ie
LCOS
> — — sin ands =——«. =v =e
dx dx

I= |e” cosx dx = cosx xXe& — | —sinx

x & dx
x bs : Oe

du
For | sinx x e” dx, let u = sinx > —=cosx and
dx

sinxX e* dx = sinx x e& — |cosx x e dx +c

=—[=e cosx+e sinx—I1 +c

=> 21 =e cosx+e’
sinx +c

=> I1=}(e* cosx+e*sinx) +K (where K = $c)

du
In this case, the choice of u and — does not matter.
dx

du dv
(a) Oe agree and qn = SIX = v= —cosx

I= |e sinxdx = —e* cosx + | e” cosxdx

du du
For lecosxdx.6 lew
¢ = — =e" and —=tosy
= pe
dx dx

e cosxdx =e” sinx — | e* sinxdx +c

=>I=-—e"*cosx
+e’ sinx —I+¢

=> 2] =e” sinx — & cosx +c

=> 1=}(e sinx — e* cosx) +K (where K = 4c)

 fi du dv ee
4 (a) Putu=sinx >—=cosx and —=cosx >v=sinx
dx dx
: ao) ;
I =| sinxcosxdx
= sin’ x — |cosxsinxdx
+c
SHEET
E1
=sin?x -—I+c

=> 2] =sin’x+c

=> 1 =sin’x+K K = $c)

(where
7.1
TASK
se:
b
(b) a
d jh eo.
sin
eet
) =}
Nit
= sinx cos x
2sinxcosx

=> | sinxcosxdx =tsin’x+c

Also, sinx cosx = $sin 2x

Hence, | sinxcosxdx = 5sin 2x dite —tcos 2x Fe

= —1(1-2sin?x)+c= —141sin?x +c =4sin’x+K

(where K =c= )

5 |In(x
x 1)dx =Inx
x x— xx dx

xInx— |1dx

xInx—-—x+e

xInx—<x

~ 0.910
 Integrating the circle
Fon)
1 A circle of radius 1 unit has area 7. A quarter of the circle therefore has area
147.

d
2 (a) page —2 and the integral therefore becomes:
dx
avn a U
2) Jao
The integral has not been simplified.
dx
(b)
b x Sa
=¢ — = cos@

J/(1 — x”) dx = |\/(1 — sin’6)x cos 0d

cos’ 6d

+ |(1 + cos 26) do

(0 +4sin 20) +c
=40+4sin@cosO +c

=tsin"'x +4x/(1 —x)+e

1
(c) Ae ~ x’) dx = qsing = in

V(t = #)de = fsin“ $+} EV =2) — sin

1
2

=i yo

Se sin
0!— Oh eand sing m = 4

cos’ 0d0 =} (1 + cos 26) da

0
am
il
5 res 20 ,
6 +5sin

perl
=n sve
asin ele
3
=5n +33
5 (a) For triangle OAB, } x base x height =} x} x /2=1/3—
angle BOC = i
and so sector BOC is ath of the unit circle. Its area is therefore TT.
(b) The shaded area is therefore iil +4/3, as established in questions 2(c)
and 4.

6 Let x= 3sind +“ = 3c0s0

in
TASKS
7.1
E2
yJ (9 — x*) dx /(9
— 9sin? 0) x 3.cos dO
6"

Z 2
9| cos 6d0

=3 0 +4sin 20

2
sec udu
———a 1d
(a) 1+ tan? u -

utc=tanixtc

(b) Let tan ay

dx =
ecru aed
8 |1du=utc=sin
'x+c

~1dv = —v+C=-—cos
!x+C

oo , jl
sin "x = 41 —cos x

The answers to questions 8 and 9 are the same if the arbitrary constants are
related by c + im = (G:
 Carbon dating Commentary

N=0.5Ny) and N=No ¢75730k

1 When t = 5730,
=> 05 = @3730k

=> In0.5 = —5730k

= k=121x10* (to3s.f.)
TASKSHEET
E1
7.3 ih
8300

3 2700 Bc is about 4700 years ago.

2550 Bc is about 4550 years ago.
6.68
Assuming a radioactivity level of 6.68 when 8300 In
3.8
4680 years
This agrees with historical records.

4 4500Bc is about 6500 years ago.

Assuming an original radioactivity level of 6.68: 8300 In pee

7200 years

There was strong evidence that the origins of agriculture were even earlier
than at first thought.
 751

Appendix
Core A and AS level formulas
The formulas listed here are assumed to be memorised by students taking any
A or AS level examination after 1998. You would not be provided with such a list
in an exam.

Quadratic equations
—b+
Vb? — 4ac
ax’ +.bx + c = 0 has roots
2a

Laws of logarithms
log, (xy) = log, x + log, y

log, (=) = log, x — log, y

y

log, (x*) = klog, x

Arithmetic series
w= a(n — lid
S, =4n(a +1) =5n(2a + (n— 1)d)

Geometric series
u, = ar"|

5 tr
a=?)
ay,

SS g for |r| < 1

l= %

Binomial series
r, n(n—1) 4 PN A level only
Og iat a rer ncen’ aan xe
752 APPENDIX

Trigonometry
In the triangle ABC
a Opie
sinA sinB sinC
a=b’+c—2becosA

area = 5ab sin C

cos’ A+sin*A=1

sec’ A=1+tan?A

cosec’ A=1+cot?A

sin (A + B) = sin AcosB + cos Asin B A level only

cos
(A + B) = cos AcosB ¥ sin A sin B
tan
A + tan B
B)=
cae) f=etrancAttanB

Differentiation
Function Derivative
at nx

sin kx kcoskx
cos kx —ksinkx
tan kx k sec? kx
ok kek

Inx
x
f(x) + g(x) £'(x) + g(x)
f(x)g(x) £"(x)g(x) + £(x)g'(x)
f(x) £'(x)g(x) — £(x)g'(x)
A level only
g(x) (g(x))?
f(g(x)) £"(g(x))g'
(x)
 APPENDIX 753

Integration
Function Integral

n 1
x x tl cast —1
n+1
1
cos kx —sinkx
+¢
k

sin kx ee coskx +c A level only

k
2 1
sec Rx planks +c

1
eae

In|x| +c, x #0

f(x) + g(x) +e A level only

f(g(x)) +¢

Area
b
Area under a curve = |y dx
a
(y > 0)

Vectors
2.6 a
A level only
y|- |b = xa+
yb + zc
z C
Index

abundant numbers 294 notation for 122-3

accuracy, see error see also differentiation, integration
addition formulas 354-5, 356 calibration of models 326
algebra 57-8 carbon dating, differential equation for 566-8
algebraic division 72-3 Cartesian equation of plane 386-7
algebraic transformation, and graph Cartesian equations, from parametric 427, 428,
transformation 272 434
amplitude of functions 206-7 chain rule 413
analysis of models 315 parametric differentiation 435
annual percentage rate 30 and integration 418, 471
approximation and inverse functions 422-3
of area under graph 138-9 and product rule 446
for differential equation solutions 504-7 circle
anderror 482 equation of 431
and graph sketching 52, 53 parametric equations for 428, 431
linear 3-4, 481 Pythagoras’ theorem 343
Newton-Raphson 484-8 circle stretching 273-4
of x by convergence 333-5 ellipse 275
polynomial 492, 493-4 circuit 295
by quadratic 491-2 circular function ratios 339
Taylor’s first 481-3 circular measure 261, 263, 267-8
see also numerical methods see also radians
APR, see annual percentage rate classification 290
arc length 263 of Platonic solids 321-2
arcsin (sin) 216-17 cobweb diagrams 88
area coefficients 71, 74
under graph 136, 138-9 binomial 397, 399-400
negative 150-1 column vectors 364
of sector 263 common difference 25
area of influence 326-8 common ratio 34, 35
arithmetic series 23-4 complementary angles 467-8
sum of 26 completing the square 9, 10, 60
articles, mathematical 318 complex numbers 68-9
asymptotes 54 roots of quadratics 68-70
composition of functions 178-9, 412-14
base 227 constant 57
belongs to (€) 45 constant of integration 171, 499
binomial coefficients 397, 399-400 construction of models 315
binomial expressions 395, 396, 526, 534 contradiction, proof by 301-2, 558
binomial series 401, 403 convergence 87
binomial theorem 400 approximation of nm by 333-5
bounds, of roots 82, 83 to golden ratio 79
graphical prediction of 87-9
calculus 94 and iterative methods 87, 89
fundamental theorem of 165-6, 169 of Newton-Raphson method 487
of linear functions 94-5, 98 of sequences 18, 19
 INDEX 755

eos, 217 with respect to time 439

cosecant (cosec) 432 see also gradient
cosine curve 203-4 dihedral angle 392-3
cosine function 468 dimension, and integration 136
derivative of 265,266 direction diagrams 501-3
range of values 204 discontinuities 52, 54
as series 496 and integration 474
tranformations of 205, 207-8 discriminant of quadratic 63
values commonly used 339 divergence, of sequences 18, 20
see also trigonometric identities division, algebraic 72-3
cosine rule 359 domain
for scalar products 377 of functions 44, 181-2
cotangent (cot) 432 of inverse sine and cosine 217
counter-example 298, 299, 557 of inverse tangent 221
cubics, stationary points of 128 of inverses 184, 185
domain restriction, and inverse functions 184
decay, exponential 225, 248 dominant term 49
decimal search 78 dot (scalar) product 376-7, 378, 381
decision-making, and optimisation 129-31 double angle formulas 355-6
deficient numbers 294
definite integral 171 e 247
degree-radian relationship 261 see also e*
derivative e* (exponential function)
graphical representation 125-6 gradient of 247
of inverse functions 422-3 graph of 248
Leibniz notation 121-3, 491 relationship to In(x) 253-5
of In(x) 254-5 as series 496
of product 442-3, 444 eclipse prediction (modelling example) 310-16
of quotient 447-8 ellipse 275
of trigonometric functions 265, 266 equations for 428, 431
see also chain rule, differentiation equations
derived functions 115 of circles 428, 431
difference formulas, sum and 468 and identities 58
differences, table of 291-2 of linear graphs 4
differential equation 498-500 parametric 426
algebraic solution 499, 500 simultaneous 62
direction diagrams 501-3 solution of 530-1
formulation 512-14 error 404-6
and growth functions 509-10 and approximation 482
numerical solution 504-7 relative 407, 421
for radioactive decay 566-8 error notation 405
separating variables 515-17 estimation, numerical
differentiation of area under graph 138-9
of cubics 128 of f'(a) 112
from first principles 120, 121 Euler, Leonhard 247
and gradient 109-12, 117-18, 120 Euler’s formula for polyhedra 322
implicit 450-1 even function 197
and limits 118, 120 exponential decay 225, 248
and local straightness 117-18, 120, 443 exponential function, see e*
multiple 491 exponential growth 224-5, 231-2, 233, 248
parametric 435-6 exponential series 496
of quadratics 128
relationship to integration 165-6, 169 f'(a), numerical estimation 112
756
ee INDEX

f(x) notation 491 of exponential functions 247

f', see inverse functions and graph sketching 125-6
factor theorem 72 graphical method 109-10
factorials 399 and incremental change 420
factorisation 12-13, 15,528 limit notation 120
Fechner, Gustav 81 linear functions 3-4, 94-5, 523, 525
Fermat, Pierre de 303-4 and local straightness 102, 104
Fermat’s last theorem 304 numerical method 111-12
Fibonacci (Leonardo of Pisa) 78 of sine function 258-60
Fibonacci numbers 78-9, 82, 547-8, 549 of velocity vector 439
first principles, differentiation from 120, 121 see also differentiation
formulas, re-arrangement of 187-9 gradient graph 105
4-regular graphs 319 of polynomial 115
fractions of sine function 259
partial 476-8 graph sketching 46-7, 125-6
rationalisation of 533 and algebraic transformation relationships 272
friendly numbers 294 cubics 128
function notation 41, 183, 491 polynomials 49-50
using in graphs 42-3 quadratics 14,128
functions rational functions 52, 54
amplitude of 206-7 graphical optimisation 133-4
composition of 178-9, 412-14 graphical prediction of convergence 87-9
conditions for 183, 184, 215-16 graphs
derived 115 area under (integration) 136, 138-9
domain 44, 181-2, 183 of exponential function 248
even 197 4-regular 319
graphs of 2, 55 linear 3-4
implicit 450 locally straight 104
inverse 183-5, 186 of parametric equations 426-7
linear 3-4, 94-8, 115, 154-5, 523, 525, 530 plotting or sketching 46-7
odd:1975221 of quadratics 14
range 181-2 shapes of functions 2, 55
as rule and domain 44 summed 350
self-inverse 186 translation 44, 194-5, 199
trigonometric 214-16, 467-8 transformation 193, 196-9
fundamental theorem of calculus 165-6, 169 units of 136
using function notation in 42-3
game strategy 331-2 gravity model 325
Gauss, Carl 69, 70 in geographical modelling 325-30
general term 21 growth, exponential 224-5, 231-2, 233, 248
generalisation 285-7 growth, modelling 231-3
generating formula 292 growth functions
geographical modelling 325-30 base 227
geometric series 34-5 base e 249
infinite 37 and differential equations 509-10
golden ratio (¢) 77, 81, 544-50
algebraic calculation 81 Hex 331-2
and Fibonacci numbers 79, 82
numerical estimation 77-8, 80 t(j) 68, 247
golden rectangle 81-2, 549 see also imaginary numbers
gradient identities
and differentiation 117-18, 120 and equations 58
in direction diagrams 501-3 trigonometric, see trigonometric identities
 INDEX 757

image of function graph 194 classifying 290

imaginary numbers 69, 247 extension of 307
implicit differentiation 450-1 notation 289
and product rule 453-4 use of tables 291-2
increment, small 420 irrational numbers 68
indefinite integral 171 iterative methods 80, 83-4
indices 227,229 and convergence 87, 89
as unknown parameters 239-40 Newton-Raphson 484-5, 487-8
inductive definition 17, 21
inequalities 64-6 i (i) 68,247
algebraic approach 64, 66 see also imaginary numbers
graphical approach 65
infinite series 37 lattice point 295
inspection, integration by 465-6 Leibniz, Gottfried 122
integers 44 Leibniz notation 121-3, 491
integrals limit notation 120
_ definite 171 limits
indefinite 171 definite integrals 171
of linear functions 154, 155 differentiation 118, 120
notation 157 linear approximation 481
of polynomials 160 linear functions 96-7, 115, 525
positive and negative 151 calculus of 98
integration 142-3 gradients of 3-4, 94-5, 523, 52
algebraic 162-4 integrals of 154,155
area under graph 136, 138-9 solving 530
and chain rule 418, 471 linear graphs, equations of 3-4
complex areas 170 linear inequalities 64
constant of 171,499 lines
and discontinuities 474 intersection point 372-4
fundamental theorem of calculus 165-6, skew 373, 374
169 vector equation of 369-70, 372
by inspection 465-6 In(x) 253-5
and negative area 150-1 derivative of 254,255
numerical estimation 144-9, 465 In(1+x), as series 496
by parts 469-70 relationship to e* 253-5
of reciprocal function 473-4 local maxima and minima (turning points) 106,
relationship to differentiation 165-6, 126
169 determination 129-31, 163-4
by substitution 471 local straightness 102, 104
as summation 142 chain rule 413-14
of trigonometric functions 467-8 differentiation 117-18, 120, 443
volume of revolution 458-9, 460-3 logarithm 235, 238
interaction, modelling 325-30 base of 235, 238
intercept 4 to base e (In) 253-5
interest rate comparison 28-30 logarithmic function, see In(x)
inverse functions 183-5, 186 logarithmic graph 243-4
differentiation by chain rule 422-3 logarithmic relationship 234-5
formula re-arrangement 187-9 logistic curve 255
with parameters 190-1 Lucas sequence 547-9
trigonometric 214-16
investigation Maclaurin’s series 493-4
basic processes 284-6 many-to-one relationship 183, 216
758 INDEX

mathematical articles 318 optimisation 124, 129-31, 133-4

maxima and minima, local 106, 126 oscillating sequences 19
maximum or minimum, determination 129-31,
163-4 parabolas 6
mid-ordinate rule 138, 146-7 see also quadratics
errors 148 parameters 190, 426
and negative area 150 elimination of 427, 428, 434
model, gravity 325 parametric curves 426-7
model analysis 315 parametric differentiation 435-6
model calibration 326 parametric equations 426
model construction 315 for circles 428, 431
model validation 309, 316 partial fractions 476-8
modelling parts, integration by 469-70
example (solar eclipse) 310-16 Pascal’s triangle 396-7
formulating differential equations 512-14 patterns and problem solving 285, 286, 299,
game strategy 331-2 300
geographical 325-30 pentagon drawing 544-5
of growth 231-3 perfect numbers 294
interaction 325-30 perfect squares 527
logistic curve 255 periodic function 207
of periodic behaviour 210-12 inverse 214, 215-16
process of 309 modelling of 210-11
periodic sequences 20
N (natural numbers) 44 phase shift 207, 351
Napier, John 238, 253 phi (@), see golden ratio
natural numbers 44 pi (Z), approximation 143, 144-5, 333-5
negative area 150-1 planes
negative numbers 64, 68 angles between 392-3
Newton, Isaac 122,401 Cartesian equation of 386-7
Newton-Raphson approximation 484-5, 487-8 vector equation of 383, 385, 389-90
normal 551 Platonic solids 321-4
normal vector 389-90 plotting graphs 46-7
notation pointwise addition 350
function 41, 183,491 polyhedra, Euler’s formula 322
integral 157 polynomial graphs
Leibniz 121-3, 491 locally straight 104
limit 120 sketching 49, 50
in problem solving 289 polynomials 6, 71
sigma 32, 540, 541 approximation 492, 493-4
vector 8, 367 construction of 114-15
number system 68-9 dominant term 49
numerical integration 144-5, 147-9, 465 factor theorem 72
numerical methods 76 generating formula 292
decimal search 78 gradient graph of 115
iteration 80, 83-4 integrals of 160
locating roots 82 roots of 68-9, 71
mid-ordinate rule 138, 146-8, 150 position vector 364, 367
trapezium rule 138-9, 146-8 power function 244
see also approximation see also indices
numerology 337 principal values
of inverse sine 2G Ay,
odd functions 197,221 of sin’ and cos! 217
one-to-one relationship 183, 184, 186, 216 of tan’ 221
 INDEX 759

problem-solving of linear functions 94, 97-8

algorithm 306 see also calculus; differentiation
notation 289 rational functions, graph sketching 52, 54
see also investigation rational numbers 44
product rule 442-3, 444, 469 rationalisation of fractions 533
and chain rule 446 real numbers 44, 69
and implicit differentiation 453-4 reciprocal function, integration of 473-4
proof 298-300 recurrence relation 17, 80
by contradiction 301-2, 558 reflection
Fermat’s approach 303-4 of function graphs 199
methods of 300-2 oflog graphs 235
Pythagoras’ theorem 337, 338 reflection symmetry 186
equation of circle 343 of cosine curve 205
three-dimensional applications 340-1 of even functions 197
Pythagorean triples 338 relative error 407, 421
resolving into partial fractions 476-8
Q (rational numbers) 44 revolution, solids of 458-9, 460-2, 463
quadratic inequalities 64, 65 roots
quadratics 6 ; bounds of 82, 83
algebraic methods 60, 61 complex 68, 70
approximation by 491-2 finding by iterative method 83-4
completing the square 9, 10, 60 numerical methods 82
complex roots 68, 70 of polynomials 68-9, 71
defining conditions 490 of quadratics 12
differentiation of 128 see also quadratics
discriminant 63 rotation 202
factorisation of 12-13, 15 rotation symmetry
gradient of tangent to 111 of odd functions 197
graph sketching 14 of sine curve 205
parabolas 6 rule (function) 44
roots 12
solving 530 scalar (dot) product 376-7, 378, 381
stationary points 128 secant (sec) 432
transformation of 7-8 sector, area of 263
zeros of 12, 50 self-inverse functions 186
quotient rule 447-8 separating variables 515-17
sequences 17
R (real numbers) 44, 69 convergence of 18, 19
radians 260, 261, 263 divergence of 18, 20
circular measure 261, 263 Fibonacci 547-9
in iterative methods 488 general term 21
for problem formulation 267-8 generating formula 292
relationship to degrees 261 Lucas 547-9
and trigonometric derivatives 265, 266 oscillating 19
radioactive decay, differential equation for 566-8 periodic 20
range types of 19-20
of functions 181-2 series 24
of inverse functions 185 arithmetic 23-4, 26
of inverse sine and cosine 217 binomial 401, 403, 496
of inverse tangent 221 cosine 496
single value requirement 183 geometric 34-5
rate of change infinite 37
and circular measure 267-8 logarithmic 496
760 INDEX
eee ee

Maclaurin’s 493-4 table of differences 291-2

sine 496 tan 222
series summation 32 tangent function 104-5, 221-2
sigma (XZ) notation 32, 540, 541 values commonly used 339
simultaneous equations 62 and velocity vector 440
solving 531 see also trigonometric identities
sin’ (arcsin) Taylor’s first approximation 481-3
domain and range 217 transformations
principal values 216 algebraic 272
sine curve 203-4 of circle 273-5
sine function combined 196-7, 199
derivative of 265, 266, 556 of quadratics 7-8
gradient of 258-9 relationship to original graph 193, 196-9
range of values 204 of sine and cosine 205, 207-8
as series 496 translation 8
of summed angles 354-5, 356 of graphs 44, 194-5, 199
tranformations of 205, 207-8 sine to cosine 205
values commonly used 339 vector notation 8, 367
see also trigonometric identities trapezium rule 138-9, 146-7
sine rule 361 errors 148
sketching graphs, see graph sketching triangular numbers 58, 294
skew lines 373, 374 trigonometric equation solving 218-20
solar eclipse prediction (modelling example) trigonometric functions
310-16 derivatives of 265, 266
solids integration of 467-8
Platonic 321-4 inverse 214, 215-16
of revolution 458-9, 460-2, 463 trigonometric identities 346-7, 434
solution set 64 addition formulas 354-5, 356
sphere, equation for 345 cosine rule 359
square numbers 294 double angle formulas 355-6
staircase diagram 88 sine rule 361
stationary points 106, 124-5 sum and difference formulas 351, 468, 560
graphical optimisation 133 trigonometric ratios 339, 432
quadratics and cubics 128 turning points 106
stepwise approximation, differential equations
504-7 validation of models 309, 316
stretch variables 57
circle 273-4 optimisation, by calculus 129-31
sine and cosine functions 207 separation of 515-16, 517
substitution, integration by 471-2 see also parameters
sum and difference formulas 468 vector diagrams 365
summation 32, 247 vector equations 367, 368
arithmetic series 26 vectors 364
geometric series 34—S angle between 378, 379
to infinity 37 notation 8, 367
and integration 142 scalar product 376-7, 378, 381
sigma (XZ) notation 32, 540, $41 of velocity 438-9
surds 62, 532 of lines in three dimensions 369-70, 372
symmetry of planes 383, 385, 389-90
and function type 197 velocity vector 438-9, 440
reflection 186, 197, 205 vertex 6
rotation 197,205 volume of revolution 458-9, 460-3
 INDEX 761

Wiles, Andrew 304

Z, (integers) 44
Zeno’s paradox 542
zero
division by 52
powers of 229
zero gradient, see stationary points
Zeros
and graph sketching 50
of polynomials 68-9, 71
of quadratics 12, 50
 i) a

ae 503. tieon - ao ne ny ign Fe 2 a a

ie ans : Ah 06s), ahah , pote
es. ay
. - cos 1 Selig A
<3 -
~r C7

a" aa
 — ,

a a

'
_—

iy

if
/

pe

a)

'

i §

Z
~ TA
 Mathematics for A and AS level

Pure mathematics
The School Mathematics Project
This self-contained book can be used as a course
book for the pure content of any A or AS level
mathematics syllabus. It adapts and combines the
existing 16-19 Mathematics units Foundations,
Introductory calculus, Functions, Problem solving,
Mathematical methods and Calculus methods
in a single volume.

The aim of 16-19 Mathematics has been to

produce a course which students find challenging,
but accessible and enjoyable. This book aims to
develop ability and confidence in mathematics and
its applications, together with an appreciation
of how mathematical ideas help in understanding
the world in which we live.

A resource pack (ISBN 0 521 56614 2) for this book

contains additional exercise material in the form
of photocopiable review sheets.

Also available in the Mathematics for A and AS level

series:

Statistics ISBN 0 521 56616 9

Mechanics ISBN 0 521 566150

Statistics resource pack ISBN 0 521 56613 4

Mechanics resource pack ISBN 0 521 56612 6

CAMBRIDGE
UNIVERSITY PRESS
ISBN 0-521-56617-7

+ el 8 0521°566179