C ALCULUS Q UESTIONS

F OR M ATHEMATICS 101

Joel F ELDMAN Andrew R ECHNITZER Elyse Y EAGER

T HIS DOCUMENT WAS TYPESET ON M ONDAY 17 TH A PRIL , 2017.

IJ Legal stuff

• Copyright c 2016 Joel Feldman, Andrew Rechnitzer and Elyse Yeager

• In the near future this will be licensed under the Creative Commons Attribution-
NonCommercial-ShareAlike 4.0 International License. You can view a copy of the
license at
http://creativecommons.org/licenses/by-nc-sa/4.0/.

2
 H OW TO USE THIS BOOK

IJ Introduction

First of all, welcome back to Calculus!

This book is written as a companion to the CLP notes.

§§ How to Work Questions

This book is organized into four sections: Questions, Hints, Answers, and Solutions. As
you are working problems, resist the temptation to prematurely peek at the back! It’s
important to allow yourself to struggle for a time with the material. Even professional
mathematicians don’t always know right away how to solve a problem. The art is in
gathering your thoughts and figuring out a strategy to use what you know to find out
what you don’t.
If you find yourself at a real impasse, go ahead and look for a hint in the Hints section.
Think about it for a while, and don’t be afraid to read back in the notes to look for a key
idea that will help you proceed. If you still can’t solve the problem, well, we included the
Solutions section for a reason! As you’re reading the solutions, try hard to understand
why we took the steps we did, instead of memorizing step-by-step how to solve that one
particular problem.
If you struggled with a question quite a lot, it’s probably a good idea to return to it in a few
days. That might have been enough time for you to internalize the necessary ideas, and
you might find it easily conquerable. Pat yourself on the back–sometimes math makes you
feel good! If you’re still having troubles, read over the solution again, with an emphasis
on understanding why each step makes sense.
One of the reasons so many students are required to study calculus is the hope that it will
improve their problem-solving skills. In this class, you will learn lots of concepts, and
be asked to apply them in a variety of situations. Often, this will involve answering one

i
H OW TO USE THIS BOOK

really big problem by breaking it up into manageable chunks, solving those chunks, then
putting the pieces back together. When you see a particularly long question, remain calm
and look for a way to break it into pieces you can handle.

§§ Working with Friends

Study buddies are fantastic! If you don’t already have friends in your class, you can ask
your neighbours in lecture to form a group. Often, a question that you might bang your
head against for an hour can be easily cleared up by a friend who sees what you’ve missed.
Regular study times make sure you don’t procrastinate too much, and friends help you
maintain a positive attitude when you might otherwise succumb to frustration. Struggle
in mathematics is desirable, but suffering is not.
When working in a group, make sure you try out problems on your own before coming
together to discuss with others. Learning is a process, and getting answers to questions
that you haven’t considered on your own can rob you of the practice you need to master
skills and concepts, and the tenacity you need to develop to become a competent problem-
solver.

§§ Types of Questions

Q[1]: Questions outlined in blue make up the representative question set. This set of
questions is intended to cover the most essential ideas in each section. These questions
are usually highly typical of what you’d see on an exam, although some of them are
atypical but carry an important moral. If you find yourself unconfident with the idea
behind one of these, it’s probably a good idea to practice similar questions.
This representative question set is our suggestion for a minimal selection of questions to
work on. You are highly encouraged to work on more.

Q[2](˚): In addition to original problems, this book contains problems pulled from quizzes
and exams given at UBC for Math 101 and 105 (second–semester calculus) and Math 121
(honours second–semester calculus). These problems are marked with a star. The authors
would like to acknowledge the contributions of the many people who collaborated to
produce these exams over the years.
The questions are organized into Stage 1, Stage 2, and Stage 3.

§§ Stage 1

The first category is meant to test and improve your understanding of basic underlying
concepts. These often do not involve much calculation. They range in difficulty from
very basic reviews of definitions to questions that require you to be thoughtful about the
concepts covered in the section.

ii
H OW TO USE THIS BOOK

§§ Stage 2

Questions in this category are for practicing skills. It’s not enough to understand the philo-
sophical grounding of an idea: you have to be able to apply it in appropriate situations.
This takes practice!

§§ Stage 3

The last questions in each section go a little farther than Stage 2. Often they will combine
more than one idea, incorporate review material, or ask you to apply your understanding
of a concept to a new situation.
In exams, as in life, you will encounter questions of varying difficulty. A good skill to
practice is recognizing the level of difficulty a problem poses. Exams will have some easy
question, some standard questions, and some harder questions.

iii
H OW TO USE THIS BOOK

iv
 C ONTENTS

How to use this book i

I The questions 1
1 Integration 3
1.1 Definition of the Integral . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.2 Basic properties of the definite integral . . . . . . . . . . . . . . . . . . . . . . 5
1.3 The Fundamental Theorem of Calculus . . . . . . . . . . . . . . . . . . . . . 6
1.4 Substitution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.5 Area between curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.6 Volumes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.7 Integration by parts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.8 Trigonometric Integrals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
1.9 Trigonometric Substitution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
1.10 Partial Fractions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
1.11 Numerical Integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
1.12 Improper Integrals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
1.13 More Integration Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

2 Applications of Integration 29
2.1 Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
2.2 Averages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.3 Centre of Mass and Torque . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
2.4 Separable Differential Equations . . . . . . . . . . . . . . . . . . . . . . . . . 33

3 Sequences and Series 37

3.1 Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
3.2 Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
3.3 Convergence Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.4 Absolute and Conditional Convergence . . . . . . . . . . . . . . . . . . . . . 43

v
CONTENTS CONTENTS

3.5 Power Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

3.6 Taylor Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

II Hints to problems 51
1.1 Definition of the Integral . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
1.2 Basic properties of the definite integral . . . . . . . . . . . . . . . . . . . . . 54
1.3 The Fundamental Theorem of Calculus . . . . . . . . . . . . . . . . . . . . . 54
1.4 Substitution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
1.5 Area between curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
1.6 Volumes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
1.7 Integration by parts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
1.8 Trigonometric Integrals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
1.9 Trigonometric Substitution . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
1.10 Partial Fractions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
1.11 Numerical Integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
1.12 Improper Integrals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
1.13 More Integration Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
2.1 Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
2.2 Averages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
2.3 Centre of Mass and Torque . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
2.4 Separable Differential Equations . . . . . . . . . . . . . . . . . . . . . . . . . 62
3.1 Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
3.2 Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
3.3 Convergence Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
3.4 Absolute and Conditional Convergence . . . . . . . . . . . . . . . . . . . . . 65
3.5 Power Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
3.6 Taylor Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

III Answers to problems 69

1.1 Definition of the Integral . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
1.2 Basic properties of the definite integral . . . . . . . . . . . . . . . . . . . . . 71
1.3 The Fundamental Theorem of Calculus . . . . . . . . . . . . . . . . . . . . . 72
1.4 Substitution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
1.5 Area between curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
1.6 Volumes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
1.7 Integration by parts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
1.8 Trigonometric Integrals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
1.9 Trigonometric Substitution . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
1.10 Partial Fractions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
1.11 Numerical Integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
1.12 Improper Integrals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
1.13 More Integration Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
2.1 Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
2.2 Averages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
2.3 Centre of Mass and Torque . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

vi
CONTENTS CONTENTS

2.4 Separable Differential Equations . . . . . . . . . . . . . . . . . . . . . . . . . 81

3.1 Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
3.2 Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
3.3 Convergence Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
3.4 Absolute and Conditional Convergence . . . . . . . . . . . . . . . . . . . . . 84
3.5 Power Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
3.6 Taylor Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

IV Solutions to problems 87
1.1 Definition of the Integral . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
1.2 Basic properties of the definite integral . . . . . . . . . . . . . . . . . . . . . 95
1.3 The Fundamental Theorem of Calculus . . . . . . . . . . . . . . . . . . . . . 97
1.4 Substitution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
1.5 Area between curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
1.6 Volumes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
1.7 Integration by parts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
1.8 Trigonometric Integrals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
1.9 Trigonometric Substitution . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
1.10 Partial Fractions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
1.11 Numerical Integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
1.12 Improper Integrals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
1.13 More Integration Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
2.1 Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
2.2 Averages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
2.3 Centre of Mass and Torque . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
2.4 Separable Differential Equations . . . . . . . . . . . . . . . . . . . . . . . . . 178
3.1 Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
3.2 Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
3.3 Convergence Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
3.4 Absolute and Conditional Convergence . . . . . . . . . . . . . . . . . . . . . 206
3.5 Power Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
3.6 Taylor Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217

vii
CONTENTS CONTENTS

viii
Part I

T HE QUESTIONS

1
 Chapter 1

I NTEGRATION

1.1IJ Definition of the Integral

Exercises
Jump to HINTS, ANSWERS, SOLUTIONS or TABLE OF CONTENTS.

§§ Stage 1
ż7
Q[1](˚): Let f be a function on the whole real line. Express f ( x ) dx as a limit of Rie-
´1
mann sums, using the right endpoints.
4
ÿ
Q[2](˚): f (1 + k ) ¨ 1 is a left Riemann sum for a function f ( x ) on the interval [ a, b] with
k =1
n subintervals. Find the values of a, b and n.
Q[3](˚): Fill in the blanks with right, left, or midpoint; an interval; and a value of n.
3
ř
f (1.5 + k ) ¨ 1 is a Riemann sum for f on the interval [ , ] with
k =0
n= .

§§ Stage 2

Q[4](˚): The value of the following limit is equal to the area below a graph of y = f ( x ),
integrated over the interval [0, b]:
n   
ÿ 4 4i 2
lim sin 2 +
nÑ8 n n
i =1

3
I NTEGRATION 1.1 D EFINITION OF THE I NTEGRAL

Find f ( x ) and b.
Q[5](˚): For a certain function f ( x ), the following equation holds:
n
c ż1
ÿ k k2
lim 1´ 2 = f ( x ) dx
nÑ8 n2 n 0
k =1

Find f ( x ).
Q[6](˚): Use sigma notation to write the midpoint Riemann sum for f ( x ) = x8 on [5, 15]
with n = 50. Do not evaluate the Riemann sum.
ş5
Q[7](˚): Estimate ´1 x3 dx using three approximating rectangles and left hand end points.
n
ÿ 3 ´i/n
Q[8](˚): Express lim e cos(3i/n) as a definite integral.
nÑ8 n
i =1
n
ÿ iei/n
Q[9](˚): Let Rn = . Express lim Rn as a definite integral. Do not evaluate this
n2 nÑ8
i =1
integral.
ÿ
n 
´1´2i/n 2
Q[10](˚): Express lim e ¨ as a integral in three different ways.
nÑ8 n
i =1
ż3
Q[11](˚): Use elementary geometry to calculate f ( x ) dx, where
0
#
x, if x ď 1,
f (x) =
1, if x ą 1.

ż2
Q[12](˚): Evaluate |2x| dx.
´1

Q[13](˚): A car’s gas pedal is applied at t = 0 seconds and the car accelerates
continuously until t = 2 seconds. The car’s speed at half-second intervals is given in the
table below. Find the best possible upper estimate for the distance that the car traveled
during these two seconds.

t (s) 0 0.5 1.0 1.5 2

v (m/s) 0 14 22 30 40

§§ Stage 3

Q[14](˚): (a) Express

 2
d
n
ÿ 2 2i
lim 4 ´ ´2 +
nÑ8 n n
i =1

4
 I NTEGRATION 1.2 B ASIC PROPERTIES OF THE DEFINITE INTEGRAL

as a definite integal.
(b) Evaluate the integral of part (a).
Q[15](˚): Consider the integral:
ż3
(7 + x3 ) dx. (˚)
0

(a) Approximate this integral using the left Riemann sum with n = 3 intervals.
(b) Write down the expression for the right Riemann sum with n intervals and calculate
the sum. Now take the limit n Ñ 8 in your expression for the Riemann sum, to
evaluate the integral (˚) exactly.
You may use the identity
n
ÿ n4 + 2n3 + n2
i3 =
4
i =1

ş4
Q[16](˚): Using a limit of right–endpoint Riemann sums, evaluate 2 x2 dx. You may use
n n
i2 = n(n + 1)(2n + 1)/6.
ř ř
the formulas i = n(n + 1)/2 and
i =1 i =1
ż2
Q[17](˚): Find ( x3 + x ) dx using the definition of the definite integral. You may use the
0
n n
n4 +2n3 +n2 n2 + n
i3 =
ř ř
summation formulas 4 and i= 2 .
i =1 i =1
ż4
Q[18](˚): Using a limit of right–endpoint Riemann sums, evaluate (2x ´ 1) dx. Do not
1
n
ř
use anti-differentiation, except to check your answer. You may use the formula i =
i =1
n ( n +1)
2 .

1.2IJ Basic properties of the definite integral

Exercises
Jump to HINTS, ANSWERS, SOLUTIONS or TABLE OF CONTENTS.

§§ Stage 1

Q[1](˚): Decide whether each of the following statements is true or false. If false, provide
a counterexample. If true provide a brief justification. (Assume that f ( x ) and g( x ) are
continuous functions.)

5
 I NTEGRATION 1.3 T HE F UNDAMENTAL T HEOREM OF C ALCULUS

ż ´2 ż2
(a) f ( x )dx = ´ f ( x )dx.
´3 3
ż ´2 ż3
(b) If f ( x ) is an odd function, then f ( x ) dx = f ( x ) dx.
´3 2
ż1 ż1 ż1
(c) f ( x ) ¨ g( x ) dx = f ( x ) dx ¨ g( x ) dx.
0 0 0

§§ Stage 2
ż3 ż3 ż3


Q[2](˚): Suppose f ( x ) dx = ´1 and g( x ) dx = 5. Evaluate 6 f ( x ) ´ 3g( x ) dx.
2 2 2
ż2 ż2 ż2


Q[3](˚): If f ( x ) dx = 3 and g( x ) dx = ´4, calculate 2 f ( x ) + 3g( x ) dx.
0 0 0

Q[4](˚): The functions f ( x ) and g( x ) obey

ż ´1 ż2 ż0 ż2
f ( x ) dx = 1 f ( x ) dx = 2 g( x ) dx = 3 g( x ) dx = 4
0 0 ´1 0

ş2  
Find ´1 3g( x ) ´ f ( x ) dx.
ż2
Q[5](˚): Evaluate |2x| dx.
´1

§§ Stage 3

ş2  ? 
Q[6](˚): Evaluate ´2 5+ 4 ´ x2 dx.
ż +2012
sin x
Q[7](˚): Evaluate dx.
´2012 log(3 + x2 )
ż +2012
Q[8](˚): Evaluate x1/3 cos x dx.
´2012

1.3IJ The Fundamental Theorem of Calculus

Exercises
Jump to HINTS, ANSWERS, SOLUTIONS or TABLE OF CONTENTS.

6
I NTEGRATION 1.3 T HE F UNDAMENTAL T HEOREM OF C ALCULUS

§§ Stage 1
2 ´3)
Q[1](˚): Suppose that f ( x ) is a function
?
and F ( x ) = e( x + 1 is an antiderivative of f ( x ).
ż 5
Evaluate the definite integral f ( x ) dx.
1

Q[2](˚): For the function f ( x ) = x3 ´ sin 2x, find its antiderivative F ( x ) that satisfies
F (0) = 1.
Q[3](˚): Decide whether each of the following statements is true or false. Provide a brief
justification.
şπ
(a) If f ( x ) is continuous on [1, π ] then 1 f 1 ( x ) dx = f (π ) ´ f (1).
ş1
(b) ´1 x12 dx = 0.
şb şb
(c) If f is continuous on [ a, b] then a x f ( x ) dx = x a f ( x ) dx.

§§ Stage 2
ż2
Q[4](˚): Evaluate x3 + sin x ) dx.
0
ż2
x2 + 2
Q[5](˚): Evaluate dx.
1 x2
Q[6](˚): If
żx ż0
F(x) = log(2 + sin t) dt and G (y) = log(2 + sin t) dt
0 y



find F1 π
2 and G1 π
2 .
şx 2
Q[7](˚): Let f ( x ) = 1 100(t2 ´ 3t + 2)e´t dt. Find the interval(s) on which f is increasing.
ż cos x
1
Q[8](˚): If F ( x ) = 3
dt, find F1 ( x ).
0 t + 6
ş 1+ x 4 t2
Q[9](˚): Compute f 1 ( x ) where f ( x ) = 0 e dt.
ż sin x !
d
Q[10](˚): Evaluate (t6 + 8)dt .
dx 0
ş x3

Q[11](˚): Let F ( x ) = 0 e´t sin πt 1
2 dt. Calculate F (1).
ż 0 
d dt
Q[12](˚): Find 3
.
du cos u 1 + t
şx
Q[13](˚): If x sin(πx ) = 0 f (t) dt where f is a continuous function, find f (4).

7
I NTEGRATION 1.3 T HE F UNDAMENTAL T HEOREM OF C ALCULUS

ż x2 ż0
´t 2
Q[14](˚): Consider the function F ( x ) = e dt + e´t dt.
0 ´x

(a) Find F 1 ( x ).
(b) Find the value of x for which F ( x ) takes its minimum value.
żx
Q[15](˚): If F ( x ) is defined by F ( x ) = esin t dt, find F1 ( x ).
x4 ´x3
 ż ´x2 
d t


Q[16](˚): Evaluate cos e dt .
dx x5
ż ex ?
Q[17](˚): Differentiate sin t dt.
x

§§ Stage 3
ż5 #
3 if x ď 3
Q[18](˚): Evaluate f ( x ) dx, where f ( x ) = .
1 x if x ě 3
şx
Q[19](˚): Find f ( x ) if x2 = 1 + 1 f (t) dt.
ż2
1 1
Q[20](˚): If f (1) = 2 and f (2) = 3, find f 1 ( x ) f 2 ( x ) dx.
1
Q[21](˚): A car traveling at 30 m/s applies its brakes at time t = 0, its velocity (in m/s)
decreasing according to the formula v(t) = 30 ´ 10t. How far does the car go before it
stops?
ş2x´x2

Q[22](˚): Compute f 1 ( x ) where f ( x ) = 0 log 1 + et dt. Does f ( x ) have an absolute
maximum? Explain.
ş x2 ´2x dt
Q[23](˚): Find the minimum value of 0 1+ t4
. Express your answer as an integral.
ş x2 ?
Q[24](˚): Define the function F ( x ) = 0 sin( t) dt on the interval 0 ă x ă 4. On this
interval, where does F ( x ) have a maximum?
π ř
n

Q[25](˚): Evaluate lim sin jπ/n by interpreting it as a limit of Riemann sums.
nÑ8 n j=1

n
1ÿ 1
Q[26](˚): Use Riemann sums to find the limit lim j
.
nÑ8 n 1 +
j =1 n
ż x 3 +1
3
Q[27](˚): Define f ( x ) = x3 et dt.
0

(a) Find a formula for the derivative f 1 ( x ).

(b) Find the equation of the tangent line to the graph of y = f ( x ) at x = ´1.

8
 I NTEGRATION 1.4 S UBSTITUTION

1.4IJ Substitution

Exercises
Jump to HINTS, ANSWERS, SOLUTIONS or TABLE OF CONTENTS.
Recall that we are using log x to denote the logarithm of x with base e. In other courses it
is often denoted ln x.

§§ Stage 1

Q[1](˚): What is the integral which results when the substitution u = sin x is applied to
ż π/2
the integral f (sin x ) dx?
0

§§ Stage 2
ż1
2 2
Q[2](˚): Use substitution to evaluate xe x cos(e x ) dx.
0
ş8 ş2
Q[3](˚): Let f (t) be any function for which 1 f (t) dt = 1. Calculate the integral 1 x2 f ( x3 ) dx.
x2
ż
Q[4](˚): Evaluate dx.
( x3 + 1)101
ż e4
dx
Q[5](˚): Evaluate .
e x log x
ż π/2
cos x
Q[6](˚): Evaluate dx.
0 1 + sin x
ż π/2
Q[7](˚): Evaluate cos x ¨ (1 + sin2 x ) dx.
0
ş3 x2 ´x
Q[8](˚): Evaluate 1 (2x ´ 1) e dx.
( x 2 ´ 4) x
ż
Q[9](˚): Evaluate ? dx.
4 ´ x2

§§ Stage 3
ż2
2
Q[10](˚): Calculate xe x dx.
´2

j2

n
ř j
Q[11](˚): Calculate lim n2
sin 1 + n2
.
nÑ8 j=1

9
 I NTEGRATION 1.5 A REA BETWEEN CURVES

j2

n
ř j
Q[12](˚): Evaluate lim n2
cos n2
.
nÑ8 j=1

n b
ř j j2
Q[13](˚): Calculate lim n2
1+ n2
.
nÑ8 j=1

1.5IJ Area between curves

Exercises
Jump to HINTS, ANSWERS, SOLUTIONS or TABLE OF CONTENTS.

§§ Stage 1

Q[1](˚): Write down a definite integral that represents the finite area bounded by the
curves y = x3 ´ x and y = x for x ě 0. Do not evaluate the integral explicitly.
Q[2](˚): Write down a definite integral that represents the area of the region bounded by
the line y = ´ 2x and the parabola y2 = 6 ´ 5x
4 . Do not evaluate the integral explicitly.

Q[3](˚): Write down a definite integral that represents the area of the finite plane region
bounded by y2 = 4ax and x2 = 4ay, where a ą 0 is a constant. Do not evaluate the integral
explicitly.
Q[4](˚): Write down a definite integral that represents the area of the region bounded
between the line x + 12y + 5 = 0 and the curve x = 4y2 . Do not evaluate the integral
explicitly.

§§ Stage 2

1
Q[5](˚): Find the area of the region bounded by the graph of f ( x ) = (2x´4)2
and the x–axis
between x = 0 and x = 1.
Q[6](˚): Find the area between the curves: y = x and y = 3x ´ x2 , by first identifying the
points of intersection and then integrating.
?
Q[7](˚): Calculate the area of the region enclosed by y = 2x and y = x + 1.
?
Q[8](˚): Find the area of the finite region bounded between the two curves y = 2 cos(πx/4)
and y = |x|.
?
Q[9](˚): Find the area of the finite region that is bounded by the graphs of f ( x ) = x2 x3 + 1
and g( x ) = 3x2 .
Q[10](˚): Find the area to the left of the y–axis and to the right of the curve x = y2 + y.

10
 I NTEGRATION 1.6 V OLUMES

§§ Stage 3

Q[11](˚): The graph below shows the region between y = 4 + π sin x and
y = 4 + 2π ´ 2x.
y
8
y = 4 + 2π − 2x

6 y = 4 + π sin(x)
4

x
π π 3π 2π
2 2

Find the area of this region.

Q[12](˚): Compute the area of the finite region bounded by the curves x = 0, x = 3,
y = x + 2 and y = x2 .
?
Q[13](˚): Find the total area between the curves y = x 25 ´ x2 and y = 3x, on the interval
0 ď x ď 4.

1.6IJ Volumes

Exercises
Jump to HINTS, ANSWERS, SOLUTIONS or TABLE OF CONTENTS.

§§ Stage 1

Q[1](˚): Write down definite integrals that represent the following quantities. Do not
evaluate the integrals explicitly.
(a) The volume of the solid obtained by rotating around the x–axis the region between
? 2
the x–axis and y = x e x for 0 ď x ď 3.
(b) The volume of the solid obtained by revolving the region bounded by the curves
y = x2 and y = x + 2 about the line x = 3.
Q[2](˚): Write down definite integrals that represent the following quantities. Do not
evaluate the integrals explicitly.
(a) The volume of the solid obtained by rotating the finite plane region bounded by the
curves y = 1 ´ x2 and y = 4 ´ 4x2 about the line y = ´1.
(b) The volume of the solid obtained by rotating the finite plane region bounded by the
curve y = x2 ´ 1 and the line y = 0 about the line x = 5.

11
I NTEGRATION 1.6 V OLUMES

Q[3](˚): Write down a definite integral that represents the volume of the solid obtained by
rotating around the line y = ´1 the region between the curves y = x2 and y = 8 ´ x2 . Do
not evaluate the integrals explicitly.
Q[4](˚): Write definite integrals that represent the following quantities. Do not evaluate the
integrals.
(a) The area of the finite plane region bounded by y2 = 4ax and x2 = 4ay, where a ą 0 is
a constant.
(b) The volume of the solid obtained by rotating the finite plane region bounded by the
curves y = 1 ´ x2 and y = 4 ´ 4x2 about the line y = ´1.
(c) The volume of the solid obtained by rotating the finite plane region bounded by the
curve y = x2 ´ 1 and the line y = 0 about the line x = 5.

§§ Stage 2

Q[5](˚): Let a ą 0 be a constant. Let R be the finite region bounded by the graph of
? 2
y = 1 + xe x , the line y = 1, and the line x = a. Using vertical slices, find the volume
generated when R is rotated about the line y = 1.
? ?
Q[6](˚): Let R be the region between the curves T ( x ) = xe3x and B( x ) = x (1 + 2x ) on
the interval 0 ď x ď 3. (It is true that T ( x ) ě B( x ) for all 0 ď x ď 3.) Compute the volume
of the solid formed by rotating R about the x-axis.
Q[7](˚): Find the volume of the solid generated by rotating the finite region bounded by
y = 1/x and 3x + 3y = 10 about the x–axis.
Q[8](˚): Let R be the region inside the circle x2 + (y ´ 2)2 = 1. Let S be the solid obtained
by rotating R about the x-axis.
(a) Write down an integral representing the volume of S.
(b) Evaluate the integral you wrote down in part (a).
Q[9](˚): The region R is the portion of the first quadrant which is below the parabola
y2 = 8x and above the hyperbola y2 ´ x2 = 15.
(a) Sketch the region R.
(b) Find the volume of the solid obtained by revolving R about the x axis.
Q[10](˚): The region R is bounded by y = log x, y = 0, x = 1 and x = 2. (Recall that we
are using log x to denote the logarithm of x with base e. In other courses it is often
denoted log x.)
(a) Sketch the region R.
(b) Find the volume of the solid obtained by revolving this region about the y axis.
Q[11](˚): The finite region between the curves y = cos( 2x ) and y = x2 ´ π 2 is rotated
about the line y = ´π 2 . Using vertical slices (disks and/or washers), find the volume of
the resulting solid.

12
I NTEGRATION 1.6 V OLUMES

Q[12](˚): The solid V is 2 meters high and has square horizontal cross sections. The length
of the side of the square cross section at height x meters above the base is 1+2 x m. Find the
volume of this solid.
Q[13](˚): Consider a solid whose base is the finite portion of the xy–plane bounded by the
curves y = x2 and y = 8 ´ x2 . The cross–sections perpendicular to the x–axis are squares
with one side in the xy–plane. Compute the volume of this solid.
Q[14](˚): A frustrum of a right circular cone (as shown below) has height h. Its base is a
circular disc with radius 4 and its top is a circular disc with radius 2. Calculate the
volume of the frustrum.
2

h
4

§§ Stage 3

Q[15](˚): Let R be the bounded region that lies between the curve y = 4 ´ ( x ´ 1)2 and
the line y = x + 1.
(a) Sketch R and find its area.
(b) Write down a definite integral giving the volume of the region obtained by rotating R
about the line y = 5. Do not evaluate this integral.
(
Q[16](˚): Let R = ( x, y) : ( x ´ 1)2 + y2 ď 1 and x2 + (y ´ 1)2 ď 1 .
(a) Sketch R and find its area.
(b) If R rotates around the y–axis, what volume is generated?
?
Q[17](˚): Let R be the plane region bounded by x = 0, x = 1, y = 0 and y = c 1 + x2 ,
where c is a positive constant.
(a) Find the volume V1 of the solid obtained by revolving R about the x–axis.
(b) Find the volume V2 of the solid obtained by revolving R about the y–axis.
(c) If V1 = V2 , what is the value of c?
Q[18](˚): The region R is the portion of the first quadrant where 3 ď x ď 4 and
0 ď y ď ? 10 2 .
25´x

(a) Sketch the region R.

(b) Determine the volume of the solid obtained by revolving R around the x–axis.
(c) Determine the volume of the solid obtained by revolving R around the y–axis.

13
 I NTEGRATION 1.7 I NTEGRATION BY PARTS

Q[19](˚): The graph below shows the region between y = 4 + π sin x and
y = 4 + 2π ´ 2x.
y
8
y = 4 + 2π − 2x

6 y = 4 + π sin(x)
4

x
π π 3π 2π
2 2

The region is rotated about the line y = ´1. Express in terms of definite integrals the
volume of the resulting solid. Do not evaluate the integrals.

1.7IJ Integration by parts

Exercises
Jump to HINTS, ANSWERS, SOLUTIONS or TABLE OF CONTENTS.

§§ Stage 1

§§ Stage 2
ż
Q[1](˚): Evaluate x log x dx.
ż
log x
Q[2](˚): Evaluate dx.
x7
żπ
Q[3](˚): Evaluate x sin x dx.
0
ż π
2
Q[4](˚): Evaluate x cos x dx.
0
ż
Q[5](˚): Evaluate cos´1 y dy.

§§ Stage 3
ż
Q[6](˚): Evaluate 4y arctan(2y) dy.

Q[7](˚): A reduction formula.

14
 I NTEGRATION 1.8 T RIGONOMETRIC I NTEGRALS

sinn´1 ( x ) cos( x )
(a) Derive the reduction formula sinn ( x ) dx = ´ n´1
sinn´2 ( x ) dx.
ş ş
n + n

(b) Calculate 0 sin8 ( x ) dx.

ş π/2

Q[8](˚): Let R be the part of the first quadrant that lies below the curve y = arctan x and
between the lines x = 0 and x = 1.
(a) Sketch the region R and determine its area.
(b) Find the volume of the solid obtained by rotating R about the y–axis.
ż4
?

1
Q[9](˚): Let f (0) = 1, f (2) = 3 and f (2) = 4. Calculate f 2 x dx.
0

1.8IJ Trigonometric Integrals

Exercises
Jump to HINTS, ANSWERS, SOLUTIONS or TABLE OF CONTENTS.

§§ Stage 1
ż
Q[1](˚): Evaluate cos3 x dx.
ż
Q[2](˚): Evaluate sin36 t cos3 t dt.

§§ Stage 2
ż
Q[3](˚): Evaluate tan3 x sec5 x dx.
ż
Q[4](˚): Evaluate sec4 x tan46 x dx.
żπ
Q[5](˚): Evaluate cos2 x dx.
0

§§ Stage 3

Q[6](˚): A reduction formula.

(a) Let n be a positive integer with n ě 2. Derive the reduction formula
tann´1 ( x )
tann ( x ) dx = n´1 ´ tann´2 ( x ) dx.
ş ş

(b) Calculate 0 tan6 ( x ) dx.

ş π/4

15
 I NTEGRATION 1.9 T RIGONOMETRIC S UBSTITUTION

1.9IJ Trigonometric Substitution

Exercises
Jump to HINTS, ANSWERS, SOLUTIONS or TABLE OF CONTENTS.

Recall that we are using log x to denote the logarithm of x with base e. In other courses it
is often denoted ln x.

§§ Stage 1

Q[1](˚): For each of the following integrals, choose the substitution that is most beneficial
for evaluating the integral.

2x2
ż
(a) ? dx
9x2 ´ 16
x4 ´ 3
ż
(b) ? dx
1 ´ 4x2
ż
´5/2
(c) (25 + x2 ) dx

§§ Stage 2
ż
1
Q[2](˚): Evaluate dx.
( x2 + 4)3/2
ż4
1
Q[3](˚): Evaluate dx. Your answer may not contain inverse trigonometric
0 (4 + x2 )3/2
functions.
ż ż
ˇ ˇ
Q[4](˚): Evaluate ? dx . You may use that sec dx = log ˇ sec x + tan xˇ + C.
x2 +25
ż
Q[5](˚): Evaluate ?dx .
x2 x2 +16

ż 5/2
Q[6](˚): Evaluate ? dx .
0 25´x2
ż
dx
Q[7](˚): Evaluate ? . Do not include any inverse trigonometric functions in
x2 x2 ´ 9
your answer.

16
 I NTEGRATION 1.10 PARTIAL F RACTIONS

§§ Stage 3
ż?
Q[8](˚): Evaluate 4 ´ x2 dx.
ż
dx
Q[9](˚): Evaluate ? .
3 ´ 2x ´ x2
Q[10](˚): (a) Show that 0 cos4 θ dθ = (8 + 3π )/32.
ş π/4

ş1 dx
(b) Evaluate ´1 ( x2 +1)3 .
ż?
25x2 ´4
Q[11](˚): Evaluate x dx.

1.10IJ Partial Fractions

Exercises
Jump to HINTS, ANSWERS, SOLUTIONS or TABLE OF CONTENTS.
Recall that we are using log x to denote the logarithm of x with base e. In other courses it
is often denoted ln x.

§§ Stage 1

1 3x3 ´ 2x2 + 11
Q[1](˚): Find the coefficient of in the partial fraction expansion of 2 .
x´1 x ( x ´ 1)( x2 + 3)
x3 + 3
Q[2](˚): Write out the general form of the partial-fractions decomposition of .
( x 2 ´ 1)2 ( x 2 + 1)
You need not determine the values of any of the coefficients.

§§ Stage 2
ş2 dx
Q[3](˚): Evaluate 1 x + x2 .
ż
1
Q[4](˚): Calculate dx.
x4 + x2
ż
12x + 4
Q[5](˚): Calculate dx.
( x ´ 3)( x2 + 1)
Q[6](˚): Evaluate the following indefinite integral using partial fractions:
3x2 ´ 4
ż
F(x) = dx
( x ´ 2)( x2 + 4)

17
 I NTEGRATION 1.11 N UMERICAL I NTEGRATION
ż
x ´ 13
Q[7](˚): Evaluate dx.
x2´x´6
ż
5x + 1
Q[8](˚): Evaluate dx.
x2 + 5x + 6

§§ Stage 3

1.11IJ Numerical Integration

Exercises
Jump to HINTS, ANSWERS, SOLUTIONS or TABLE OF CONTENTS.
Recall that we are using log x to denote the logarithm of x with base e. In other courses it
is often denoted ln x.

§§ Stage 1

Q[1](˚): Decide whether the following statement is true or false. If false, provide a
counterexample. If true provide a brief justification.
When f ( x ) is positive and concave up, any Trapezoid Rule approximation for
żb żb
f ( x ) dx will be an upper estimate for f ( x ) dx.
a a

§§ Stage 2
şπ
Q[2](˚): Find the midpoint rule approximation to 0 sin x dx with n = 3.
Q[3](˚): A 6 metre long cedar log has cross sections which are approximately circular.
The diameters of the log, measured at one metre intervals, are given below:
metres from left end of log 0 1 2 3 4 5 6
diameter in metres 1.2 1 0.8 0.8 1 1 1.2
Use Simpson’s Rule to estimate the volume of the log.
Q[4](˚): The solid V is 40 cm high and the horizontal cross sections are circular disks. The
table below gives the diameters of the cross sections in centimeters at 10 cm intervals.
Use the trapezoidal rule to estimate the volume of V.
height 0 10 20 30 40
diameter 24 16 10 6 4
Q[5](˚): The circumference of an 8 metre high tree at different heights above the ground
is given in the table below. Assume that all horizontal cross–sections of the tree are
circular disks.

18
I NTEGRATION 1.11 N UMERICAL I NTEGRATION

height (metres) 0 2 4 6 8
circumference (metres) 1.2 1.1 1.3 0.9 0.2
Use Simpson’s rule to approximate the volume of the tree.
Q[6](˚): By measuring the areas enclosed by contours on a topographic map, a geologist
determines the cross sectional areas A in m2 of a 60 m high hill. The table below gives the
ş60
cross sectional area A(h) at various heights h. The volume of the hill is V = 0 A(h) dh.

h 0 10 20 30 40 50 60
A 10,200 9,200 8,000 7,100 4,500 2,400 100
(a) If the geologist uses the Trapezoidal Rule to estimate the volume of the hill, what will
be his estimate, to the nearest 1,000m3 ?
(b) What will be the geologist’s estimate of the volume of the hill if he uses Simpson’s
Rule instead of the Trapezoidal Rule?
Q[7](˚): The graph below applies to both parts (a) and (b).
y
10

x
1 2 3 4 5 6 7

(a) Use the Trapezoidal Rule, with n = 4, to estimate the area under the graph between
x = 2 and x = 6. Simplify your answer completely.
(b) Use Simpson’s Rule, with n = 4, to estimate the area under the graph between x = 2
and x = 6.
ş1
Q[8](˚): The integral ´1 sin( x2 ) dx is estimated using the Midpoint Rule with 1000
points. Show that the error in this approximation is at most 2 ¨ 10´6 in absolute value.
şb
You may use the fact that when approximating a f ( x ) dx with the Midpoint Rule using n
points, the absolute value of the error is at most K (b ´ a)3 /24n2 when | f 2 ( x )| ď K for all
x P [ a, b].
Q[9](˚): The total error using the midpoint rule with n subintervals to approximate the
integral of f ( x ) over [ a, b] is bounded by M (b ´ a)3 /(24n2 ), if | f 2 ( x )| ď M for all
a ď x ď b.

19
I NTEGRATION 1.11 N UMERICAL I NTEGRATION

ş1
If the integral ´2 2x4 dx is approximated using the midpoint rule with 60 subintervals,
what is the largest possible error between the approximation M60 and the true value of
the integral?
ş2
Q[10](˚): Both parts of this question concern the integral I = 0 ( x ´ 3)5 dx.
(a) Write down the Simpson’s Rule approximation to I with n = 6. Leave your answer in
calculator-ready form.
(b) Which method of approximating I results in a smaller error bound: the Midpoint
Rule with n = 100 intervals, or Simpson’s Rule with n = 10 intervals? You may use
the formulas
K ( b ´ a )3 L ( b ´ a )5
|E M | ď and |ES | ď ,
24n2 180n4

where K is an upper bound for | f 2 ( x )| and L is an upper bound for | f (4) ( x )|.
şb
Q[11](˚): Consider the Trapezoid Rule for making numerical approximations to a f ( x ) dx.
K (b´a)3
The error for the Trapezoid Rule satisfies |ET | ď 12n2 , where | f 2 ( x )| ď K for a ď x ď b.
If ´2 ă f 2 ( x ) ă 0 for 1 ď x ď 4, find a value of n to guarantee the Trapezoid Rule will
ş4
give an approximation for 1 f ( x ) dx with absolute error, |ET |, less than 0.001.
ş5
Q[12](˚): Find a bound for the error in approximating 1 1x dx using Simpson’s rule with
n = 4. Do not write down the Simpson’s rule approximation S4 .
şb
In general the error in approximating a f ( x ) dx using Simpson’s rule with n steps is
K (b´a)
180 ( ∆x ) where ∆x = b´a
bounded by 4
n and K ě | f (4) ( x )| for all a ď x ď b.
Q[13](˚): Find a bound for the error in approximating
ż1
 
e´2x + 3x3 dx
0

using Simpson’s rule with n = 6. Do not write down the Simpson’s rule approximation
Sn .
şb
In general the error in approximating a f ( x ) dx using Simpson’s rule with n steps is
K (b´a)
180 ( ∆x ) where ∆x = b´a
bounded by 4
n and K ě | f (4) ( x )| for all a ď x ď b.
ş2
Q[14](˚): Let I = 1 (1/x ) dx.
(a) Write down the trapezoidal approximation T4 for I. You do not need to simplify your
answer.
(b) Write down the Simpson’s approximation S4 for I. You do not need to simplify your
answer.
(c) ˇWithoutˇ computing I, find an upper bound for |I ´ S4 |. You may use the fact that if
ˇ f (4) ( x )ˇ ď K on the interval [ a, b], then the error in using Sn to approximate
şb 5 4
a f ( x ) dx has absolute value less than or equal to K ( b ´ a ) /180n .

20
I NTEGRATION 1.11 N UMERICAL I NTEGRATION

Q[15](˚): A function s( x ) satisfies s(0) = 1.00664, s(2) = 1.00543,

ˇ (k) sˇ(4) =k 1.00435,
s(6) = 1.00331, s(8) = 1.00233. Also, it is known to satisfy ˇs ( x )ˇ ď 1000 for 0 ď x ď 8
and all positive integers k.
(a) Find the best Trapezoidal Rule and Simpson’s Rule approximations that you can for
ş8
I = 0 s( x ) dx.
(b) Determine the maximum possible sizes of errors ˇ (kin the
ˇ approximations you gave in
)
part (a). Recall that if a function f ( x ) satisfies f ( x )ˇ ď Kk on [ a, b], then
ˇ
ˇż b ˇż b
ˇ K2 ( b ´ a ) 3 ˇ K ( b ´ a )5
ˇ ˇ
f ( x ) dx ´ Sn ˇˇ ď 4
ˇ ˇ
ˇ f ( x ) dx ´ Tnˇ ď
ˇ and ˇ
ˇ
ˇ
a 12n2 a 180n4

§§ Stage 3

Q[16](˚): A swimming pool has the shape shown in the figure below. The vertical
cross–sections of the pool are semi–circular disks. The distances in feet across the pool
are given in the figure at 2 foot intervals along the sixteen foot length of the pool. Use
Simpson’s Rule to estimate the volume of the pool.

8’
10’ 6’ 10’
10’ 8’
12’
2’

Q[17](˚): A piece of wire 1m long with radius 1mm is made in such a way that the
density varies in its cross–section, but is radially symmetric (that is, the local density g(r )
in kg/m3 depends only on the distance r in mm from the centre of the wire). Take as
given that the total mass M of the wire in kg is given by
ż1
´6
M = 2π10 rg(r ) dr
0

Data from the manufacturer is given below:

r 0 1/4 1/2 3/4 1

g (r ) 8051 8100 8144 8170 8190
(a) Find the best Trapezoidal Rule approximation that you can for M based on the data
in the table.

21
I NTEGRATION 1.11 N UMERICAL I NTEGRATION

(b) Suppose that it is known that |g1 (r )| ă 200 and |g2 (r )| ă 150 for all values of r.
Determine the maximum possible size of the error in the approximation you gave in
part (a). Recall that if a function f ( x ) satisfies | f 2 ( x )| ď K on [ a, b], then

K ( b ´ a )3
|I ´ Tn | ď
12n2
şb
where I = a f ( x ) dx and Tn is the Trapezoidal Rule approximation to I using n
subintervals.
ş2
Q[18](˚): Simpson’s rule can be used to approximate log 2, since log 2 = 1 1x dx.
(a) Use Simpson’s rule with 6 subintervals to approximate log 2.
(b) How many subintervals are required in order to guarantee that the absolute error is
less than 0.00001?
K (b´a)5
Note that if En is the error using n subintervals, then |En | ď 180n4 where K is the
maximum absolute value of the fourth derivative of the function being integrated
and a and b are the end points of the interval.
ż2
Q[19](˚): Let I = cos( x2 ) dx and let Sn be the Simpson’s rule approximation to I using
0
n subintervals.
(a) Estimate the maximum absolute error in using S8 to approximate I.
(b) How large should n be in order to insure that |I ´ Sn | ď 0.0001?
Note: The graph of f 4 ( x ), where f ( x ) = cos( x2 ) is shown below. The absolute error in
K (b´a)5
the Simpson’s rule approximation is bounded by 180n4
when | f 4 ( x )| ď K on the
interval [ a, b].
100

0 x
0.5 1.0 1.5 2.0

−100

−200

−300

Q[20](˚): Define a function f ( x ) and an integral I by

ż x2 ? ż1
f (x) = sin( t) dt, I= f (t) dt
0 0

Estimate how many subdivisions are needed to calculate I to five decimal places of
accuracy using the trapezoidal rule.

22
 I NTEGRATION 1.12 I MPROPER I NTEGRALS

M(b´a)3
Note that if En is the error using n subintervals, then |En | ď 12n2 where M is the
maximum absolute value of the second derivative of the function being integrated and a
and b are the end points of the interval of integration.
Q[21](˚):
A piece of wire 1m long with radius 1mm is made in such a way that the density varies
in its cross–section, but is radially symmetric (that is, the local density g(r ) in kg/m3
depends only on the distance r in mm from the centre of the wire). Take as given that the
total mass M of the wire in kg is given by
ż1
´6
M = 2π10 rg(r ) dr
0

Data from the manufacturer is given below:

r 0 1/4 1/2 3/4 1
g (r ) 8051 8100 8144 8170 8190
(a) Find the best Trapezoidal Rule approximation that you can for M based on the data in
the table.
(b) Suppose that it is known that |g1 (r )| ă 200 and |g2 (r )| ă 150 for all values of r. Deter-
mine the maximum possible size of the error in the approximation you gave in part
(a). Recall that if a function f ( x ) satisfies | f 2 ( x )| ď K on [ a, b], then

K ( b ´ a )3
|I ´ Tn | ď
12n2
şb
where I = a f ( x ) dx and Tn is the Trapezoidal Rule approximation to I using n subin-
tervals.

1.12IJ Improper Integrals

Exercises
Jump to HINTS, ANSWERS, SOLUTIONS or TABLE OF CONTENTS.

§§ Stage 1

Q[1](˚): Decide whether the following statement is true or false. If false, provide a
counterexample. If true provide a brief justification. (Assume that f ( x ) and g( x ) are
continuous functions.)
ż8 ż8
If f ( x ) dx converges and g( x ) ě f ( x ) ě 0 for all x, then g( x ) dx converges.
1 1
ż8
1
Q[2](˚): What is the largest value of q for which the integral dx diverges?
1 x5q

23
 I NTEGRATION 1.13 M ORE I NTEGRATION E XAMPLES

§§ Stage 2
ż1
x4
Q[3](˚): Evaluate the integral 5
dx or state that it diverges.
0 x ´1
ż2
1
Q[4](˚): Determine whether the integral 4/3
dx is convergent or divergent. If it
´2 ( x + 1)
is convergent, find its value.
ş8
Q[5](˚): Does the improper integral 1 ? 12 dx converge? Justify your answer.
4x ´x
ş8 dx?
Q[6](˚): Does the integral converge or diverge? Justify your claim.
0 x2 + x
ş +8
Q[7](˚): Determine (with justication!) whether the integral ´8 x2x+1 dx converges abso-
lutely, converges but not absolutely, or diverges.
ş8 | sin x|
Q[8](˚): Decide whether I = 0 x3/2 + x1/2
dx converges or diverges. Justify.
ż8
x+1
Q[9](˚): Does the integral 1/3
dx converge or diverge?
0 x ( x 2 + x + 1)

§§ Stage 3

sin4 x
ş8
Q[10](˚): Is the integral 0 dx convergent or divergent? Explain why.
x2
ş t ´x
Q[11](˚): Let Mn,t be the Midpoint Rule approximation for 0 1e+ x dx with n equal
ş8 ´x
subintervals. Find a value of t and a value of n such that Mn,t differs from 0 1e+ x dx by at
most 10´4 . Recall that the error En introduced when the Midpoint Rule is used with n
subintervals obeys

M ( b ´ a )3
|En | ď
24n2
where M is the maximum absolute value of the second derivative of the integrand and a
and b are the end points of the interval of integration.

1.13IJ More Integration Examples

Exercises
Jump to HINTS, ANSWERS, SOLUTIONS or TABLE OF CONTENTS.

§§ Stage 2

Recall that we are using log x to denote the logarithm of x with base e. In other courses it
is often denoted log x.

24
I NTEGRATION 1.13 M ORE I NTEGRATION E XAMPLES

x
ş
Q[1](˚): Evaluate x2 ´3
dx
Q[2](˚): Evaluate the following integrals.
ş4
(a) 0 ? x 2 dx
9+ x

cos3 x sin2 x dx
ş π/2
(b) 0
şe
(c) 1 x3 log x dx
Q[3](˚): Evaluate the following integrals.
ş π/2
(a) 0 x sin x dx

(b) 0 cos5 x dx
ş π/2

Q[4](˚): Evaluate the following integrals.

ş2
(a) 0 xe x dx
ş1
(b) 0 ? 1 2 dx
1+ x
ş5 4x
(c) 3 ( x2 ´1)( x2 +1) dx

Q[5](˚): Calculate the following integrals.

(a) 0 cos5 ( x ) dx
ş π/2

ş3 ?
(b) 0 9 ´ x2 dx
ş1
(c) 0 log(1 + x2 ) dx
ş8
(d) 3 ( x´1)x2 ( x´2) dx

Q[6](˚): Evaluate the following integrals. Show your work.

şπ
(a) 04 sin2 (2x ) cos3 (2x ) dx

´ 32
9 + x2
ş
(b) dx
dx
ş
(c) ( x´1)( x2 +1)
ş
(d) x tan´1 x dx
Q[7](˚): Evaluate the following integrals.
ż π/4
(a) sin5 (2x ) cos(2x ) dx
0
ż a
(b) 4 ´ x2 dx
ż
(c) log(1 + x2 ) dx

25
I NTEGRATION 1.13 M ORE I NTEGRATION E XAMPLES
ż
x+1
(d) dx
x 2 ( x ´ 1)
Q[8](˚): Calculate the following integrals.
ş8
(a) 0 e´x sin(2x ) dx
ş ?2
(b) 0 (2+ x12 )3/2 dx
ş1 2 ) dx
(c) 0 x log(1 + x
ş8 1
(d) 3 ( x´1)2 ( x´2) dx

Q[9](˚): Evaluate the following integrals.

ş
(a) x log x dx
ş ( x´1) dx
(b) x2 +4x+5
ş dx
(c) x2 ´4x +3
ş x2 dx
(d) 1+ x6
Q[10](˚): Evaluate the following integrals.
ş1
(a) 0 tan´1 x dx.
(b) x22x´1
ş
´2x +5
dx.
Q[11](˚):
ş
(a) Evaluate x log x dx.
x2
ż
(b) Evaluate dx.
( x3 + 1)101
(c) Evaluate cos3 x sin4 x dx.
ş
ş?
(d) Evaluate 4 ´ x2 dx.
Q[12](˚): Evaluate the following integrals.
x
(a) (ex +1e)(ex ´3) dx.
ş

ş4 2 ´4x +4
(b) 2
?x dx.
12+4x´x2

Q[13](˚): Evaluate these integrals.

sin3 x
ż
(a) dx
cos3 x
ż2
x4
(b) 10
dx
´2 x + 16
ż1
(c) log(1 + x2 ) dx
0

26
I NTEGRATION 1.13 M ORE I NTEGRATION E XAMPLES

Q[14](˚): Evaluate (with justification)

ż3 a
(a) ( x + 1) 9 ´ x2 dx
0
ż
4x + 8
(b) dx
( x ´ 2)( x2 + 4)
ż +8
1
(c) dx
´8 ex + e´x
Q[15](˚): Evaluate these integrals.
ż
(a) sin(log x ) dx
ż1
1
(b) dx
0 x2 ´ 5x + 6

27
I NTEGRATION 1.13 M ORE I NTEGRATION E XAMPLES

28
 Chapter 2

A PPLICATIONS OF I NTEGRATION

2.1IJ Work

Exercises
Jump to HINTS, ANSWERS, SOLUTIONS or TABLE OF CONTENTS.

§§ Stage 1

Q[1](˚): Find the work (in joules) required to stretch a string 10 cm beyond equilibrium, if
its spring constant is k = 50 N/m.

§§ Stage 2

Q[2](˚): A variable force F ( x ) = ?ax Newtons moves an object along a straight line when
it is a distance of x meters from the origin. If the work done in moving the object from
x = 1 meters to x = 16 meters is 18 Joules, what is the value of a? Don’t worry about the
units of a.
Q[3](˚): A force of 10N (newtons) is required to hold a spring stretched 5cm beyond its
natural length. How much work, in joules (J), is done in stretching the spring from its
natural length to 50cm beyond its natural length?
Q[4](˚): A 5-meter-long cable of mass 8 kg is used to lift a bucket off the ground. How
much work is needed to raise the entire cable to height 5 m? Ignore the weight of the
bucket and its contents. Use g = 9.8 m/s2 for the acceleration due to gravity.
Q[5](˚): A spherical tank of radius 3 metres is half–full of water. It has a spout of length 1
metre sticking up from the top of the tank. Find the work required to pump all of the
water in the tank out the spout. The density of water is 1000 kilograms per cubic metre.
The acceleration due to gravity is 9.8 metres per second squared.

29
 A PPLICATIONS OF I NTEGRATION 2.2 AVERAGES

1m

3m

Q[6](˚): A sculpture, shaped like a pyramid 3m high sitting on the ground, has been
made by stacking smaller and smaller (very thin) iron plates on top of one another. The
iron plate at height zm above ground level is a square whose side length is (3 ´ z)m. All
of the iron plates started on the floor of a basement 2 m below ground level.

Write down an integral that represents the work, in joules, it took to move all of the iron
from its starting position to its present position. Do not evaluate the integral. (You can
use 9.8 m/s2 for the force of gravity and 8000 kg/m3 for the density of iron.)

§§ Stage 3

2.2IJ Averages

Exercises
Jump to HINTS, ANSWERS, SOLUTIONS or TABLE OF CONTENTS.

Recall that we are using log x to denote the logarithm of x with base e. In other courses it
is often denoted ln x.

§§ Stage 1

§§ Stage 2

Q[1](˚): Find the average value of f ( x ) = sin(5x ) + 1 over the interval ´π/2 ď x ď π/2.

Q[2](˚): Find the average value of the function y = x2 log x on the interval 1 ď x ď e.

Q[3](˚): Find the average value of the function f ( x ) = 3 cos3 x + 2 cos2 x on the interval
0 ď x ď π2 .

Q[4](˚): Let k be a positive constant. Find the average value of the function f ( x ) = sin(kx )
on the interval 0 ď x ď π/k.

30
 A PPLICATIONS OF I NTEGRATION 2.3 C ENTRE OF M ASS AND T ORQUE

Q[5](˚): The temperature in Celsius in a 3 m long rod at a point x metres from the left end
80
of the rod is given by the function T ( x ) = 16´x 2 . Determine the average temperature in
the rod.
log x
Q[6](˚): What is the average value of the function f ( x ) = x on the interval [1, e]?
Q[7](˚): Find the average value of f ( x ) = cos2 ( x ) over 0 ď x ď 2π.

§§ Stage 3

Q[8](˚):
A car travels two hours without stopping. The driver records the car’s speed every 20
minutes, as indicated in the table below:
time in hours 0 1/3 2/3 1 4/3 5/3 2
speed in km/hr 50 70 80 55 60 80 40
(a) Use the trapezoidal rule to estimate the total distance traveled in the two hours.
(b) Use the answer to part (a) to estimate the average speed of the car during this period.

2.3IJ Centre of Mass and Torque

Exercises
Jump to HINTS, ANSWERS, SOLUTIONS or TABLE OF CONTENTS.

§§ Stage 1

Q[1](˚): Express the x–coordinate of the centroid of the triangle with vertices (´1, ´3),
(´1, 3), and (0, 0) in terms of a definite integral. Do not evaluate the integral.

§§ Stage 2

Q[2](˚): Find the y-coordinate of the centroid of the region bounded by the curves y = 1,
y = ´e x , x = 0 and x = 1. You may use the fact that the area of this region equals e.
Q[3](˚): Find the y-coordinate of the centre of mass of the (infinite) region lying to the right
of the line x = 1, above the x–axis, and below the graph of y = 8/x3 .
1
Q[4](˚): Consider the region bounded by y = ? , y = 0, x = 0 and x = 2.
16´x2

(a) Sketch this region.

(b) Find the y–coordinate of the centroid of this region.

31
A PPLICATIONS OF I NTEGRATION 2.3 C ENTRE OF M ASS AND T ORQUE

Q[5](˚): Find the centroid of the finite region bounded by y = sin( x ), y = cos( x ), x = 0,
and x = π/4.
Q[6](˚): Let A denote the area of the plane region bounded by x = 0, x = 1, y = 0 and
y = ? k 2 , where k is a positive constant.
1+ x

(a) Find the coordinates of the centroid of this region in terms of k and A.
(b) For what value of k is the centroid on the line y = x?
Q[7](˚): The region R is the portion of the plane which is above the curve y = x2 ´ 3x and
below the curve y = x ´ x2 .
(a) Sketch the region R
(b) Find the area of R.
(c) Find the x coordinate of the centroid of R.
1
Q[8](˚): Let R be the region where 0 ď x ď 1 and 0 ď y ď 1+ x 2
. Find the x–coordinate of
the centroid of R.
Q[9](˚): Find the centroid of the region below, which consists of a semicircle of radius 3
on top of a rectangle of width 6 and height 2.
y

x
−3 −2 −1 1 2 3
−1

−2

?
Q[10](˚): Let D be the region below the graph of the curve y = 9 ´ 4x2 and above the
x-axis.
(a) Using an appropriate integral, find the area of the region D; simplify your answer
completely.
(b) Find the centre of mass of the region D; simplify your answer completely. (Assume it
has constant density ρ.)

§§ Stage 3

Q[11](˚): Let A be the region to the right of the y-axis that is bounded by the graphs of
y = x2 and y = 6 ´ x.
22
(a) Find the centroid of A, assuming it has constant density ρ = 1. The area of A is 3
(you don’t have to show this).

32
 A PPLICATIONS OF I NTEGRATION 2.4 S EPARABLE D IFFERENTIAL E QUATIONS

(b) Write down an expression, using horizontal slices (disks), for the volume obtained
when the region A is rotated around the y-axis. Do not evaluate any integrals; simply
write down an expression for the volume.
Q[12](˚): (a) Find the y–coordinate of the centroid of the region bounded by y = e x ,
x = 0, x = 1, and y = ´1.
(b) Calculate the volume of the solid generated by rotating the region from part (a) about
the line y = ´1.

2.4IJ Separable Differential Equations

Exercises
Jump to HINTS, ANSWERS, SOLUTIONS or TABLE OF CONTENTS.

§§ Stage 1

§§ Stage 2

Q[1](˚): Find the solution to the separable initial value problem:

dy 2x
= y, y(0) = log 2
dx e
Express your solution explicitly as y = y( x ).
dy xy
Q[2](˚): Find the solution y( x ) of dx = x 2 +1
, y(0) = 3.
y
Q[3](˚): Solve the differential equation y1 (t) = e 3 cos t. You should express the solution
y(t) in terms of t explicitly.
Q[4](˚): Solve the differential equation

dy 2 2
= xe x ´log(y )
dx

Q[5](˚): Let y = y( x ). Find the general solution of the differential equation y1 = xey .
yy1 1
Q[6](˚): Find the solution to the differential equation x
= that satisfies y(0) = 3.
e ´ 2x y
Solve completely for y as a function of x.
Q[7](˚): Find the function y = f ( x ) that satisfies

dy 1
= ´xy3 and f (0) = ´
dx 4

33
A PPLICATIONS OF I NTEGRATION 2.4 S EPARABLE D IFFERENTIAL E QUATIONS

Q[8](˚): Find the function y = y( x ) that satisfies y(1) = 4 and

dy 15x2 + 4x + 3
=
dx y

Q[9](˚): Find the solution y( x ) of y1 = x3 y with y(0) = 1.

Q[10](˚): Find the solution of the differential equation

dy
x + y = y2
dx
that satisfies y(1) = ´1.
Q[11](˚): A function f ( x ) is always positive, has f (0) = e and satisfies f 1 ( x ) = x f ( x ) for
all x. Find this function.
Q[12](˚): Solve the following initial value problem:

dy 1
= 2 y (1) = 2
dx ( x + x )y
a
1+ y2 ´ 4 1 sec x
Q[13](˚): Find the solution of the differential equation y = that satisfies
tan x y
y(0) = 2. You don’t have to solve for y in terms of x.
Q[14](˚): The fish population in a lake is attacked by a disease at time t = 0, with the
result that the size P(t) of the population at time t ě 0 satisfies

dP ?
= ´k P
dt
where k is a positive constant. If there were initially 90,000 fish in the lake and 40,000 were
left after 6 weeks, when will the fish population be reduced to 10,000?
Q[15](˚): An object of mass m is projected straight upward at time t = 0 with initial speed
v0 . While it is going up, the only forces acting on it are gravity (assumed constant) and a
drag force proportional to the square of the object’s speed v(t). It follows that the
differential equation of motion is

dv
m = ´(mg + kv2 )
dt
where g and k are positive constants. At what time does the object reach its highest point?
Q[16](˚): A motor boat is traveling with a velocity of 40 ft/sec when its motor shuts off at
time t = 0. Thereafter, its deceleration due to water resistance is given by

dv
= ´k v2
dt
where k is a positive constant. After 10 seconds, the boat’s velocity is 20 ft/sec.

34
A PPLICATIONS OF I NTEGRATION 2.4 S EPARABLE D IFFERENTIAL E QUATIONS

(a) What is the value of k?

(b) When will the boat’s velocity be 5 ft/sec?
Q[17](˚): Consider the initial value problem dx
dt = k (3 ´ x )(2 ´ x ), x (0) = 1, where k is a
positive constant. (This kind of problem occurs in the analysis of certain chemical
reactions.)
(a) Solve the initial value problem. That is, find x as a function of t.
(b) What value will x (t) approach as t approaches +8.
Q[18](˚): The quantity P = P(t), which is a function of time t, satisfies the differential
equation

dP
= 4P ´ P2
dt
and the initial condition P(0) = 2.
(a) Solve this equation for P(t).
(b) What is P when t = 0.5? What is the limiting value of P as t becomes large?
Q[19](˚): An object moving in a fluid has an initial velocity v of 400 m/min. The velocity
is decreasing at a rate proportional to the square of the velocity. After 1 minute the
velocity is 200 m/min.
(a) Give a differential equation for the velocity v = v(t) where t is time.
(b) Solve this differential equation.
(c) When will the object be moving at 50 m/min?

§§ Stage 3

Q[20](˚): An investor places some money in a mutual fund where the interest is
compounded continuously and where the interest rate fluctuates between 4% and 8%,
Assume that the amount of money B = B(t) in the account in dollars after t years
satisfies the differential equation

dB

= 0.06 + 0.02 sin t B
dt
(a) Solve this differential equation for B as a function of t.
(b) If the initial investment is $1000, what will the balance be at the end of two years?
Q[21](˚): An endowment is an investment account in which the balance ideally remains
constant and withdrawals are made on the interest earned by the account. Such an
account may be modeled by the initial value problem B1 (t) = aB ´ m for t ě 0, with
B(0) = B0 . The constant a reflects the annual interest rate, m is the annual rate of
withdrawal, and B0 is the initial balance in the account.

35
A PPLICATIONS OF I NTEGRATION 2.4 S EPARABLE D IFFERENTIAL E QUATIONS

(a) Solve the initial value problem with a = 0.02 and B(0) = B0 = $30, 000. Note that
your answer depends on the constant m.
(b) If a = 0.02 and B(0) = B0 = $30, 000, what is the annual withdrawal rate m that
ensures a constant balance in the account?
Q[22](˚): A certain continuous function y = y( x ) satisfies the integral equation
żx


y( x ) = 3 + y(t)2 ´ 3y(t) + 2 sin t dt (˚)
0

for all x in some open interval containing 0. Find y( x ) and the largest interval for which
(˚) holds.
Q[23](˚): A cylindrical water tank, of radius 3 meters and height 6 meters, is full of water
when its bottom is punctured. Water drains out through a hole of radius 1 centimeter. If
• h(t) is the height of the water in the tank at time t (in meters) and
• v(t) is the velocity of the escapingawater at time t (in meters per second) then
• Torricelli’s law states that v(t) = 2gh(t) where g = 9.8 m/sec2 . Determine how
long it takes for the tank to empty.
Q[24](˚): A spherical tank of radius 6 feet is full of mercury when a circular hole of radius
1 inch is opened in the bottom. How long will it take for all of the mercury to drain from
the tank?
Use the value g = 32 feet/sec2 . Also use Torricelli’s law, which states when the height
a of
mercury in the tank is h, the speed of the mercury escaping from the tank is v = 2gh.
Q[25](˚): Consider the equation
żx



f (x) = 3 + f (t) ´ 1 f (t) ´ 2 dt
0

(a) What is f (0)?

(b) Find the differential equation satisfied by f ( x ).
(c) Solve the initial value problem determined in (a) and (b).
Q[26](˚):
A tank 2m tall is to be made with circular cross–sections with radius r = y p . Here y
measures the vertical distance from the bottom of the tank and p is a positive constant to
be determined. You may assume that when the tank drains, it obeys Torricelli’s law, that
is
dy ?
A(y) = ´c y
dt
for some constant c where A(y) is the cross–sectional area of the tank at height y. It is
desired that the tank be constructed so that the top half (y = 2 to y = 1) takes exactly the
same amount of time to drain as the bottom half (y = 1 to y = 0). Determine the value of
p so that the tank has this property. Note: it is not possible or necessary to find c for this
question.

36
 Chapter 3

S EQUENCES AND S ERIES

3.1IJ Sequences

Exercises
Jump to HINTS, ANSWERS, SOLUTIONS or TABLE OF CONTENTS.

§§ Stage 1

§§ Stage 2
(
Q[1](˚): Find the limit, if it exists, of the sequence ak , where

k! sin3 k
ak =
( k + 1) !
!
1

)
Q[2](˚): Consider the sequence (´1)n sin n . State whether this sequence converges or
diverges, and if it converges give its limit.
h 6n2 + 5n i
2
Q[3](˚): Evaluate lim + 3 cos ( 1/n ) .
nÑ8 n2 + 1

§§ Stage 3
!  1 )
Q[4](˚): Find the limit of the sequence log sin + log(2n) .
n
(8 ?
Q[5](˚): A sequence an n =0
Ă R satisfies the recursion relation an+1 = 3 + sin an for
n ě 0.

37
 S EQUENCES AND S ERIES 3.2 S ERIES
?
(a) Show that the equation x = 3 + sin x has a solution.
(b) Show that lim an = L, where L is a solution to equation above.
nÑ8
?
(c) Show that the equation x = 3 + sin x has a unique solution.

3.2IJ Series

Exercises
Jump to HINTS, ANSWERS, SOLUTIONS or TABLE OF CONTENTS.

§§ Stage 1
8
ÿ 1
Q[1](˚): Evaluate
k =7
8k

1 1 1 1 1
Q[2](˚): To what value does the series 1 + + + + + + ¨ ¨ ¨ converge?
3 9 27 81 243
8  
ÿ 6 6
Q[3](˚): Show that the series ´ converges and find its limit.
k 2 ( k + 1)2
k =1
8 
ÿ π  π 
Q[4](˚): Find the sum of the convergent series cos ´ cos .
n =3
n n+1
8
ÿ 1 + 3n
Q[5](˚): The nth partial sum of a series an is known to have the formula sn = .
5 + 4n
n =1

(a) Find an expression for an , valid for n ě 2.

8
ÿ
(b) Show that the series an converges and find its value.
n =1

§§ Stage 2
8
ř 3¨4n+1
Q[6](˚): Find the sum of the series 8¨5n . Simplify your answer completely.
n =2

Q[7](˚): Relate the number 0.23̄ = 0.233333 . . . to the sum of a geometric series, and use
that to represent it as a rational number (a fraction or combination of fractions, with no
decimals).
Q[8](˚): Express 2.656565 . . . as a rational number, i.e. in the form p/q where p and q are
integers.

38
 S EQUENCES AND S ERIES 3.3 C ONVERGENCE T ESTS

Q[9](˚): Express the decimal 0.321 = 0.321321321 . . . as a fraction.

Q[10](˚): Find the value of the convergent series
8  n +1 
ÿ 2 1 1
n
+ ´
n =2
3 2n ´ 1 2n + 1

Simplify your answer completely.

Q[11](˚): Evaluate
8  n  2 n´1 
ÿ 1
+ ´
3 5
n =1

8
ÿ 1 + 3n +1
Q[12](˚): Find the sum of the series .
n =0
4n

§§ Stage 3

3.3IJ Convergence Tests

Exercises
Jump to HINTS, ANSWERS, SOLUTIONS or TABLE OF CONTENTS.

§§ Stage 1
8
ÿ n2
Q[1](˚): Does the series ? converge?
n =2
3n2 + n

Q[2](˚): Suppose that you want to use the Limit Comparison Test on the series 8
ř
n =0 a n
2n + n an
where an = 3n +1 . Write down a sequence tbn u such that lim bn exists and is nonzero.
nÑ8
(You don’t have to carry out the Limit Comparison Test)
Q[3](˚): Decide whether each of the following statements is true or false. If false, provide
a counterexample. If true provide a brief justification.
8
ř
(a) If lim an = 0, then an converges.
nÑ8 n =1
8
(´1)n an converges.
ř
(b) If lim an = 0, then
nÑ8 n =1
8
ř 8
ř
(c) If 0 ď an ď bn and bn diverges, then an diverges.
n =1 n =1

39
S EQUENCES AND S ERIES 3.3 C ONVERGENCE T ESTS

§§ Stage 2

8
ÿ 5k
Q[4](˚): Determine, with explanation, whether the series converges or di-
n =1
4k + 3k
verges.
8
ÿ 1
Q[5](˚): Determine whether the series 1
is convergent or divergent. If it is conver-
n =0 n + 2
gent, find its value.
8
ÿ 5
Q[6](˚): Show that the series converges.
n =3
n(log n)3/2
8
ÿ 1
Q[7](˚): Find the values of p for which the series p converges.
n =2
n ( log n )
?
8 ´ n
ÿ e
Q[8](˚): Does ? converge or diverge?
n
n =1

Q[9](˚): Use the comparison test (not the limit comparison test) to show whether the series
8 ? 2
ÿ 3n ´ 7
converges or diverges.
n =2
n3
?
3 4 8
k +1 ÿ
Q[10](˚): Determine whether the series ? concerges.
k 5+9
k =1
8
n4 2n/3
ÿ
Q[11](˚): Does converge or diverge?
n =1
(2n + 7)4
ř8 n2 ´sin n
Q[12](˚): Determine (with justication!) whether the series n =1 n6 + n2 converges abso-
lutely, converges but not absolutely, or diverges.
(´1)n (2n)!
Q[13](˚): Determine (with justication!) whether the series 8
ř
n=0 (n2 +1)(n!)2 converges ab-
solutely, converges but not absolutely, or diverges.
(´1)n
Q[14](˚): Determine (with justication!) whether the series 8
ř
n=2 n(log n)101 converges abso-
lutely, converges but not absolutely, or diverges.
Q[15](˚): Determine, with explanation, whether each of the following series converge or
diverge.
8
ÿ 1
(a) ?
n =1
n2 + 1
8
ÿ n cos(nπ )
(b)
2n
n =1

40
S EQUENCES AND S ERIES 3.3 C ONVERGENCE T ESTS

Q[16](˚): Determine whether the series

8 4
ÿ k ´ 2k3 + 2
k5 + k2 + k
k =1

converges or diverges.

Q[17](˚): Determine whether each of the following series converge or diverge.

8
ÿ n2 + n + 1
(a)
n =2
n5 ´ n
8 ?
ÿ 3m + sin m
(b)
m =1
m2
8
ÿ 6
Q[18](˚): Determine whether the series is convergent or divergent. If it is conver-
n =2
7n
gent, find its value.

Q[19](˚): Determine, with explanation, whether each of the following series converge or
diverge.

(a) 1 + 31 + 51 + 71 + 19 + ¨ ¨ ¨ .
8
ÿ 2n + 1
(b)
n =1
22n+1

Q[20](˚): Determine, with explanation, whether each of the following series converges or
diverges.
8 ? 3
ÿ k
(a) 2
.
k ´k
k =2
8 10 k
ÿ k 10 (k!)2
(b) .
(2k)!
k =1
8
ÿ 1
(c) .
k(log k )(log log k )
k =3
8
ÿ n3 ´ 4
Q[21](˚): Determine whether the series is convergent or divergent.
n =1
2n5 ´ 6n

N
ÿ (´1)n
Q[22](˚): What is the smallest value of N such that the partial sum approxi-
n ¨ 10n
n =1
8
ÿ (´1)n
mates within an accuracy of 10´6 ?
n ¨ 10n
n =1

41
S EQUENCES AND S ERIES 3.3 C ONVERGENCE T ESTS

ř8 (´1)n´1 π2
Q[23](˚): It is known that n =1 n2
= 12 (you don’t have to show this). Find N so that
2
S N , the N th partial sum of the series, satisfies | π12 ´ S N | ď 10´6 . Be sure to say why your
method can be applied to this particular series.
8
ÿ (´1)n+1
Q[24](˚): The series 2
converges to some number S (you don’t have to prove
n =1
( 2n + 1 )
this). According to the Alternating Series Estimation Theorem, what is the smallest value
1
of n for which the nth partial sum of the series is at most 100 away from S? For this value
th
of n, write out the n partial sum of the series.

§§ Stage 3

Q[25](˚): Determine, with explanation, whether the following series converge or diverge.
8
ÿ nn
(a)
9n n!
n =1

8
ÿ 1
(b)
n =1
nlog n
ş8 x +sin x
Q[26](˚): (a) Prove that 2 1+ x 2
dx diverges.
8
ř n+sin n
(b) Explain why you cannot conclude that 1+ n2
diverges from part (a) and the
n =1
Integral Test.
8
ř n+sin n
(c) Determine, with explanation, whether 1+ n2
converges or diverges.
n =1

8 ?
ř e´ n
Q[27](˚): Show that ?
n
converges and find an interval of length 0.05 or less that
n =1
contains its exact value.
8
ř
Q[28](˚): Suppose that the series an converges and that 1 ą an ě 0 for all n. Prove that
n =1
8
ř an
the series 1´an also converges.
n =1

8
ř
Q[29](˚): Suppose that the series (1 ´ an ) converges, where an ą 0 for n = 0, 1, 2, 3, ¨ ¨ ¨ .
n =0
8
2n an converges or diverges.
ř
Determine whether the series
n =0

8
ÿ nan ´ 2n + 1
Q[30](˚): Assume that the series converges, where an ą 0 for
n+1
n =1

42
 S EQUENCES AND S ERIES 3.4 A BSOLUTE AND C ONDITIONAL C ONVERGENCE

n = 1, 2, ¨ ¨ ¨ . Is the following series

8  a 
n
ÿ
´ log a1 + log
a n +1
n =1

convergent? If your answer is NO, justify your answer. If your answer is YES, evaluate
8

log ana+n 1 .
ř
the sum of the series ´ log a1 +
n =1
ř8
ř8 2 Prove that if an ě 0 for all n and if the series
Q[31](˚): n =1 a n converges, then the series
n=1 an also converges.

3.4IJ Absolute and Conditional Convergence

Exercises
Jump to HINTS, ANSWERS, SOLUTIONS or TABLE OF CONTENTS.

§§ Stage 1

Q[1](˚): Decide whether the following statement is true or false. If false, provide a
counterexample. If true provide a brief justification.
8
ÿ 8
ÿ
If (´1)n+1 bn converges, then bn also converges.
n =1 n =1

§§ Stage 2
8
ÿ (´1)n
Q[2](˚): Determine whether the series is absolutely convergent, conditionally
9n + 5
n =1
convergent, or divergent; justify your answer.
8
ÿ (´1)2n+1
Q[3](˚): Determine whether the series is absolutely convergent, condition-
1+n
n =1
ally convergent, or divergent.
8
ÿ 1 + 4n
Q[4](˚): The series (´1)n´1 either: converges absolutely; converges
n =1
3 + 22n
conditionally; diverges; or none of the above. Determine which is correct.
8 ?
ÿ n cos n
Q[5](˚): Does the series 2´1
converge conditionally, converge absolutely, or di-
n =5
n
verge?

43
 S EQUENCES AND S ERIES 3.5 P OWER S ERIES

§§ Stage 3
8
ÿ 3
Q[6](˚): Both parts of this question concern the series S = (´1)n´1 24n2 e´n .
n =1

(a) Show that the series S converges absolutely.

(b) Suppose that you approximate the series S by its fifth partial sum S5 . Give an upper
bound for the error resulting from this approximation.

3.5IJ Power Series

Exercises
Jump to HINTS, ANSWERS, SOLUTIONS or TABLE OF CONTENTS.

§§ Stage 1

§§ Stage 2

Q[1](˚): (a) Find the radius of convergence of the series

8
ÿ
(´1)k 2k+1 x k
k =0

(b) You are given the formula for the sum of a geometric series, namely:
1
1 + r + r2 + ¨ ¨ ¨ = , |r| ă 1
1´r
Use this fact to evaluate the series in part (a).
8
xkÿ
Q[2](˚): Find the radius of convergence for the power series
k =0
10k+1 (k + 1)!
8
ÿ ( x ´ 2) n
Q[3](˚): Find the radius of convergence for the power series
n =0
n2 + 1
8
ř (´1)n?( x +2)n
Q[4](˚): Consider the power series n
, where x is a real number. Find the
n =1
interval of convergence of this series.
Q[5](˚): Find the radius of convergence and interval of convergence of the series
 
(´1)n x + 1 n
ÿ8

n =0
n+1 3

44
S EQUENCES AND S ERIES 3.5 P OWER S ERIES

Q[6](˚): Find the interval of convergence for the power series

8
ÿ ( x ´ 2) n
.
n =1
n4/5 (5n ´ 4)

8
ÿ ( x + 2) n
Q[7](˚): Find all values x for which the series 2
converges.
n =1
n
8
ÿ 4n
Q[8](˚): Find the interval of convergence for ( x ´ 1) n .
n
n =1

Q[9](˚): Find, with explanation, the radius of convergence and the interval of
convergence of the power series
8
ÿ ( x ´ 1) n
(´1)n
n =0
2n ( n + 2 )

8
ÿ
Q[10](˚): Find the interval of convergence for the series (´1)n n2 ( x ´ a)2n where a is a
n =1
constant.
Q[11](˚): Find the interval of convergence of the following series:
8
ÿ ( x + 1) k
(a) 2 9k
.
k =1
k
8 8  
ÿ ÿ ak a a
(b) k
ak ( x ´ 1) , where ak ą 0 for k = 1, 2, ¨ ¨ ¨ and ´ k +1 = 1 .
a k +1 a k +2 a2
k =1 k =1
x3
Q[12](˚): Find a power series representation for 1´x .

§§ Stage 3

Q[13](˚): Determine the values of x for which the series

8
ÿ xn
n =2
32n log n

converges absolutely, converges conditionally, or diverges.

Q[14](˚): (a) Find the power–series representation for 1+1x3 dx centred at 0 (i.e. in
ş

powers of x).
ş1/4
(b) The power series above is used to approximate 0 1+1x3 dx. How many terms are
required to guarantee that the resulting approximation is within 10´5 of the exact value?
Justify your answer.

45
 S EQUENCES AND S ERIES 3.6 TAYLOR S ERIES

8
ÿ x
Q[15](˚): (a) Show that nx n = for ´1 ă x ă 1.
n =0
(1 ´ x )2
8
ÿ
(b) Express n2 x n as a ratio of polynomials. For which x does this series converge?
n =0

Q[16](˚): Suppose that you have a sequence tbn u such that the series 8
ř
n=0 (1 ´ bn ) con-
verges. Using the tests we’ve learned in class, prove that the radius of convergence of the
ÿ8
power series bn x n is equal to 1.
n =0
(
Q[17](˚): Assume an is a sequence such that nan decreases to C as n Ñ 8 for some real
number C ą 0
8
an x n . Justify your answer carefully.
ř
(a) Find the radius of convergence of
n =1

(b) Find the interval of convergence of the above power series, that is, find all x for which
the power series in (a) converges. Justify your answer carefully.

3.6IJ Taylor Series

Exercises
Jump to HINTS, ANSWERS, SOLUTIONS or TABLE OF CONTENTS.

§§ Stage 1

§§ Stage 2

8
ÿ
Q[1](˚): Find the coefficient c5 of the fifth degree term in the Maclaurin series cn x n for
n =0
e3x .
Q[2](˚): The first two terms in the Maclaurin series for x2 sin( x3 ) are ax5 + bx11 , where a
and b are constants. Find the values of a and b.
1
Q[3](˚): Find the Maclaurin series for f ( x ) = 2x´1 .
8 8
3 1 3
bn x n be the Maclaurin series for f ( x ) = bn x n =
ř ř
Q[4](˚): Let x +1 ´ 2x´1 , i.e. x +1 ´
n =0 n =0
1
2x´1 . Find bn .
2
e´x ´ 1
ż
Q[5](˚): Give the first two nonzero terms in the Maclaurin series for dx.
x

46
S EQUENCES AND S ERIES 3.6 TAYLOR S ERIES
ż
Q[6](˚): Find the Maclaurin series for x4 arctan(2x ) dx.

Q[7](˚): Express the Taylor series of the function

f ( x ) = log(1 + 2x )

about x = 0 in summation notation.

Q[8](˚): The Maclaurin series for arctan x is given by
8
ÿ x2n+1
arctan x = (´1)n
n =0
2n + 1

which has radius of convergence equal to 1. Use this fact to compute the exact value of
the series below:
8
ÿ (´1)n
n =0
(2n + 1)3n

8
ÿ (´1)n
Q[9](˚): Evaluate .
n =0
n!
8
ÿ 1
Q[10](˚): Evaluate k
.
k =0
e k!
8
ÿ 1
Q[11](˚): Evaluate the sum of the convergent series k k!
.
k =1
π
8
ÿ (´1)n´1
Q[12](˚): Evaluate .
n 2n
n =1
8
ÿ n+2 n
Q[13](˚): Evaluate e .
n!
n =1
df x
Q[14](˚): Suppose that dx = 1+3x3
and f (0) = 1. Find the Maclaurin series for f ( x ).

§§ Stage 3
şx 1
Q[15](˚): Let I ( x ) = 0 1+ t4 dt.
(a) Find the Maclaurin series for I ( x ).
(b) Approximate I (1/2) to within ˘0.0001.
(c) Is your approximation in (b) larger or smaller than the true value of I (1/2)? Explain.
Q[16](˚): Using a Maclaurin series, the number a = 1/5 ´ 1/7 + 1/18 is found to be an
ş1 2
approximation for I = 0 x4 e´x dx. Give the best upper bound you can for |I ´ a|.

47
S EQUENCES AND S ERIES 3.6 TAYLOR S ERIES

Q[17](˚): Find an interval of length 0.0002 or less that contains the number

ż 1
2 2
I= x2 e´x dx
0

Q[18](˚): Find the Taylor series for f ( x ) = log( x ) centred at a = 2. Find the interval of
convergence for this series.
ş x ´t
Q[19](˚): Let I ( x ) = 0 e t´1 dt.
(a) Find the Maclaurin series for I ( x ).
(b) Approximate I (1) to within ˘0.01.
(c) Explain why your answer to part (b) has the desired accuracy.
şx
Q[20](˚): The function Σ( x ) is defined by Σ( x ) = 0 sint t dt.
(a) Find the Maclaurin series for Σ( x ).
(b) It can be shown that Σ( x ) has an absolute maximum which occurs at its smallest
positive critical point (see the graph of Σ( x ) below). Find this critical point.
(c) Use the previous information to find the maximum value of Σ( x ) to within ˘0.01.
y

şx cos t´1
Q[21](˚): Let I ( x ) = 0 t2
dt.

(a) Find the Maclaurin series for I ( x ).

(b) Use this series to approximate I (1) to within ˘0.01
(c) Is your estimate in (b) greater than I (1)? Explain.
şx
Q[22](˚): Let I ( x ) = 0 cos t+tt2sin t´1 dt

(a) Find the Maclaurin series for I ( x ).

(b) Use this series to approximate I (1) to within ˘0.001
(c) Is your estimate in (b) greater than or less than I (1)?
żx
1 ´ e´t
Q[23](˚): Define f ( x ) = dt.
0 t
8
ÿ (´1)n´1 n
(a) Show that the Maclaurin series for f ( x ) is x .
n ¨ n!
n =1

48
S EQUENCES AND S ERIES 3.6 TAYLOR S ERIES

(b) Use the ratio test to determine the values of x for which the Maclaurin series
8
ÿ (´1)n´1 n
x converges.
n ¨ n!
n =1
x3
ş1
Q[24](˚): Show that 0 e x ´1 dx ď 13 .
1´cos x
Q[25](˚): Use series to evaluate lim x.
xÑ0 + x´e
1

3
sin x´x + x6
Q[26](˚): Evaluate lim x5
.
xÑ0

Q[27](˚): Use power series to evaluate

ż1
1 ´ x2 ´ cos x
dx
0 x5/2
with an error less than 0.001.
8
ÿ x2n
Q[28](˚): (a) Show that the power series converges absolutely for all real
n =0
( 2n ) !
numbers x.
8
ÿ 1
(b) Evaluate .
n =0
(2n)!
e x +e´x
Q[29](˚): Let cosh( x ) = 2 .
(a) Find the power series expansion of cosh( x ) about x0 = 0 and determine its interval
of convergence.
(b) Show that 3 32 ď cosh(2) ď 3 32 + 0.1.
1 2
(c) Show that cosh(t) ď e 2 t for all t.

49
S EQUENCES AND S ERIES 3.6 TAYLOR S ERIES

50
Part II

H INTS TO PROBLEMS

51
Hints for Exercises 1.1. — Jump to TABLE OF CONTENTS.
H-2: Write out the general formula for the left Riemann sum (with the sum symbol being
řn
) and choose a, b and n to make it match the given sum.
k =1

H-3: Write out the given sum explicitly without using summation notation. Also write
out the first few terms in the sums in the three bullets of Definition 1.1.11 in the CLP 101
notes. Then try to identify b ´ a, and n, followed by “right”, “left” or “midpoint” and a.
H-4:
řn The main step is to express the given sum as the right Riemann sum
i =1 f ( a + i∆x ) ∆x. Don’t be afraid to guess ∆x,
řn a and f ( x) (review Definition 1.1.11 in
the CLP 101 notes). Then write out explicitly i=1 f ( a + i∆x )∆x with your guess
substituted in, and compare the result with the given sum. Adjust your guess if they
don’t match.
H-5: The main step is to express the given sum as the right Riemann sum
n
f ( a + k∆x )∆x. Don’t be afraid to guess ∆x, a and f ( x) (review Definition 1.1.11 in the
ř
k =1
n
f ( a + k∆x )∆x with your guess substituted
ř
CLP 101 notes). Then write out explicitly
k =1
in, and compare the result with the given sum. Adjust your guess if they don’t match.
H-6: Review Definition 1.1.11 in the CLP 101 notes.
H-8: The main step is to express the given sum in the form in=1 f ( xi )∆x. Don’t be afraid
ř
to guess ∆x, xi (for either a left or a right or a midpointř
sum — review Definition 1.1.11 in
the CLP 101 notes) and f ( x). Then write out explicitly in=1 f ( xi )∆x with your guess
substituted in, and compare the result with the given sum. Adjust your guess if they
don’t match.
H-9: The main step is to express the given sum in the form in=1 f ( xi )∆x. Don’t be afraid
ř
to guess ∆x, xi (probably, based on the symbol Rn , for a right Riemann sum ř— review
n
Definition 1.1.11 in the CLP 101 notes) and f ( x). Then write out explicitly i=1 f ( xi )∆x
with your guess substituted in, and compare the result with the given sum. Adjust your
guess if they don’t match.
H-10: Try several different choices of ∆x and xi .
H-11: Sketch the graph of f ( x ).
H-12: Draw a picture. See Example 1.1.15 in the CLP 101 notes.
H-13: At which time in the interval, for example, 0 ď t ď 0.5, is the car moving the
fastest?
H-14: Sure looks like a Riemann sum.
H-15: For part (b): don’t panic!. Just take it one step at a time. The first step is to write
down the Riemann sum. The second step is to evaluate the sum, using the given identity.
The third step is to evaluate the limit n Ñ 8.

53
H-16: Don’t panic!. Just take it one step at a time. The first step is to write down the
Riemann sum. The second step is to evaluate the sum, using the given formulas. The
third step is to evaluate the limit n Ñ 8.
H-17: Don’t panic!. Just take it one step at a time. The first step is to write down the
Riemann sum. The second step is to evaluate the sum, using the given formulas. The
third step is to evaluate the limit n Ñ 8.
H-18: You’ve probably seen this hint before. It is worth repeating. Don’t panic!. Just take
it one step at a time. The first step is to write down the Riemann sum. The second step is
to evaluate the sum, using the given formula. The third step is to evaluate the limit
n Ñ 8.

Hints for Exercises 1.2. — Jump to TABLE OF CONTENTS.

H-2: Split the “target integral” up into pieces that can be evaluated using the given
integrals.
H-3: Split the “target integral” up into pieces that can be evaluated using the given
integrals.
H-4: Split the “target integral” up into pieces that can be evaluated using the given
integrals.
H-5: The evaluation of this integral was also the subject of question 5 in Section 1.1. This
time try using the method of Example 1.2.6 in the CLP 101 notes.
H-6: Split the integral into a sum of two integrals. Interpret each geometrically.
H-7: Hmmmm. Looks like a complicated integral. It’s probably a trick question. Check
for symmetries.
H-8: Check for symmetries again.

Hints for Exercises 1.3. — Jump to TABLE OF CONTENTS.

H-2: First find the general antiderivative by guessing and checking.
H-3: Be careful. Two of these make no sense at all.
H-4: Guess a function whose derivative is the integrand.
H-5: Split the given integral up into two integrals. Guess, for each of the two integrals, a
function whose derivative is the integrand.
H-7: There is a good way to test where a function is increasing, decreasing, or constant,
that also has something to do with topic of this section.
H-8: See Example 1.3.5 in the CLP 101 notes.
H-9: See Example 1.3.5 in the CLP 101 notes.
H-10: See Example 1.3.5 in the CLP 101 notes.

54
H-11: See Example 1.3.5 in the CLP 101 notes.
H-12: See Example 1.3.6 in the CLP 101 notes.
H-13: What is the title of this section?
H-14: See Example 1.3.6 in the CLP 101 notes.
H-15: See Example 1.3.6 in the CLP 101 notes.
H-16: See Example 1.3.6 in the CLP 101 notes.
H-17: See Example 1.3.6 in the CLP 101 notes.
H-18: Split up the domain of integration.
d
H-19: Apply dx to both sides.
H-20: It is possible to guess an antiderivative for f 1 ( x ) f 2 ( x ) that is expressed in terms of
f 1 ( x ).
H-21: When does the car stop? What is the relation between velocity and distance
travelled?
H-22: See Example 1.3.5 in the CLP 101 notes. For the absolute maximum part of the
question, study the sign of f 1 ( x ).
H-23: See Example 1.3.5 in the CLP 101 notes. For the “minimum value” part of the
question, study the sign of f 1 ( x ).
H-24: See Example 1.3.5 in the CLP 101 notes. For the “maximum” part of the question,
study the sign of F1 ( x ).
H-25: Review the definition of the definite integral and in particular Definitions 1.1.9 and
1.1.11 in the CLP 101 notes.
H-26: Review the definition of the definite integral and in particular Definitions 1.1.9 and
1.1.11 in the CLP 101 notes.
H-27: In general, the equation of the tangent line to the graph of y = f ( x ) at x = a is
y = f ( a ) + f 1 ( a ) ( x ´ a ).

Hints for Exercises 1.4. — Jump to TABLE OF CONTENTS.

H-2: What is the derivative of the argument of the cosine?
H-3: What is the title of the current section?
H-4: What is the derivative of x3 + 1?
H-5: What is the derivative of log x?
H-6: What is the derivative of 1 + sin x?
H-7: cos x is the derivative of what?
H-8: What is the derivative of the exponent?

55
H-9: What is the derivative of the argument of the square root?
H-10: There is a short slightly sneaky method — guess an antiderivative — and a really
short still more sneaky method.
H-11: Review the definition of the definite integral and in particular Definitions 1.1.9 and
1.1.11 in the CLP 101 notes.
H-12: Review the definition of the definite integral and in particular Definitions 1.1.9 and
1.1.11 in the CLP 101 notes.
H-13: Review the definition of the definite integral and in particular Definitions 1.1.9 and
1.1.11 in the CLP 101 notes.

Hints for Exercises 1.5. — Jump to TABLE OF CONTENTS.

H-1: Draw a sketch first.
H-2: Draw a sketch first.
H-3: Draw a sketch first.
H-4: Draw a sketch first.
H-5: See Example 1.5.2 in the CLP 101 notes.
H-6: Part of the job is to determine whether y = x lies above or below y = 3x ´ x2 .
H-7: Guess the intersection points by trying small integers.
H-8: Draw a sketch first. You can also exploit a symmetry of the region to simplify your
solution.
H-9: Figure out where the two curves cross. To determine which curve is above the other,
try evaluating f ( x ) and g( x ) for some simple value of x. Alternatively, consider x very
close to zero.
H-10: Think about whether it will easier to use vertical strips or horizontal strips.
H-11: You are asked for the area, not the signed area. Be very careful about signs.
H-12: You are asked for the area, not the signed area. Draw a sketch of the region and be
very careful about signs.
H-13: You have to determine whether
?
• the curve y = f ( x ) = x 25 ´ x2 lies above the line y = g( x ) = 3x for all 0 ď x ď 4
or
• the curve y = f ( x ) lies below the line y = g( x ) for all 0 ď x ď 4 or
• y = f ( x ) and y = g( x ) cross somewhere between x = 0 and x = 4.
?

One way to do so is to study the sign of f ( x ) ´ g( x ) = x 25 ´ x2 ´ 3 .

56
Hints for Exercises 1.6. — Jump to TABLE OF CONTENTS.
H-1: Draw sketchs. The mechanically easiest way to answer part (b) uses the method of
cylindrical shells, which we have not covered. The method of washers, which we do
know about also works, but requires you have more patience and also to have a good
idea what the specified region looks like. Look at your sketch very careful when
identifying the ends of your horizontal strips.
H-2: Draw sketchs.
H-3: Draw a sketch.
H-4: (a) Draw a sketch. (b) Draw a sketch. (c) Draw a sketch. See a pattern?
H-5: Sketch the region.
H-7: Sketch the region first.
H-8: You can save yourself quite a bit of work by interpretting the integral as the area of
a known geometric figure.
H-9: See Example 1.6.3 in the CLP 101 notes.
H-10: See Example 1.6.5 in the CLP 101 notes.
H-11: Sketch the region. To find where the curves intersect, look at where cos( 2x ) and
x2 ´ π 2 both vanish.
H-12: See Example 1.6.6 in the CLP 101 notes.
H-13: See Example 1.6.6 in the CLP 101 notes.
H-14: See Example 1.6.1 in the CLP 101 notes.
H-17: The mechanically easiest way to answer part (b) uses the method of cylindrical
shells, which we have not covered. The method of washers, which we do know about
also works, but requires you have enough patience and also to have a good idea what R
looks like. So it is crucial to first sketch R. Then be very careful in identifying the left end
of your horizontal strips.
H-18: The mechanically easiest way to answer part (c) uses the method of cylindrical
shells, which we have not covered. The method of washers, which we do know about
also works, but requires you have enough patience and also to have a good idea what R
looks like. So look at the sketch in part (a) very careful when identifying the left end of
your horizontal strips.
H-19: Note that the curves cross. Be very careful. The area of this region was found in
Problem 11 of Section 1.5. It would be useful to review that problem.

Hints for Exercises 1.7. — Jump to TABLE OF CONTENTS.

H-1: What is the title of this section?
H-2: See Example 1.7.7 in the CLP 101 notes.

57
H-3: See Example 1.7.5 in the CLP 101 notes.
H-4: See Example 1.7.5 in the CLP 101 notes.
H-5: You know, or can easily look up, the derivative of arccos. You also know the title of
this section.
H-6: See Example 1.7.9 in the CLP 101 notes.
H-7: What is the title of the current section?
H-8: See Examples 1.7.9 and 1.6.5 in the CLP 101 notes.
H-9: Think, first, about how to get rid of the square root in the argument of f 2 , and,
second, how to convert f 2 into f 1 . Note that you are told that f 1 (2) = 4 and f (0) = 1,
f (2) = 3.

Hints for Exercises 1.8. — Jump to TABLE OF CONTENTS.

H-1: See Example 1.8.6 in the CLP 101 notes.
H-2: See Example 1.8.6 in the CLP 101 notes.
H-3: For practice, try doing this in two ways, with different substitutions.
H-4: See Example 1.8.14 in the CLP 101 notes.
H-5: See Example 1.8.7 in the CLP 101 notes.
H-6: See Example 1.8.16 in the CLP 101 notes.

Hints for Exercises 1.9. — Jump to TABLE OF CONTENTS.

H-1: What is this section about?
H-2: Do question 1 in this section first.
H-4: Do question 1 in this section first.
H-5: Do question 1 in this section first.
H-6: Do question 1 in this section first.
H-7: Do question 1 in this section first.
H-8: See Example 1.9.3 in the CLP 101 notes.
H-9: Complete the square.
H-10: In part (a) you are asked to integrate an even power of cos x. For part (b) you can
use a trigonometric substitution to reduce the integral of part (b) almost to the integral of
part (a).
H-11: Do question 1 in this section first.

58
Hints for Exercises 1.10. — Jump to TABLE OF CONTENTS.
H-1: Review Example 1.10.1 in the CLP 101 notes. Is the “Algebraic Method” or the
“Sneaky Method” going to be easier?
H-2: Review (1.10.7) through (1.10.11) of the CLP 101 notes. Be careful to fully factor the
denominator.
H-3: What is the title of this section?
H-4: You can save yourself some work in developing your partial fraction expansion by
renaming x2 to y.
H-5: Review Steps 3 (particularly the “Sneaky Method”) and 4 of Example 1.10.3 in the
CLP 101 notes.
H-6: Review Steps 3 (particularly the “Sneaky Method”) and 4 of Example 1.10.3 in the
CLP 101 notes.
H-7: Fill in the blank: the integrand is a function.
H-8: The integrand is yet another function.

Hints for Exercises 1.11. — Jump to TABLE OF CONTENTS.

H-1: Draw a sketch.
H-2: See §1.11.1 in the CLP 101 notes.
H-3: See §1.11.3 in the CLP 101 notes.
H-4: See §1.11.2 in the CLP 101 notes. To set up the volume integral, see Example 1.6.6 in
the CLP 101 notes.
H-5: See §1.11.3 in the CLP 101 notes. To set up the volume integral, see Example 1.6.6 in
the CLP 101 notes.
H-8: The main step is to find an allowed K. It is not necessary to find the smallest possible
allowed K.
H-9: The main step is to find M. This question is unusual in that its wording requires you
to find the smallest possible allowed M.
H-10: The main steps in part (b) are to find the smallest possible values of K and L.
H-16: See §1.11.3 in the CLP 101 notes. To set up the volume integral, see Example 1.6.2
in the CLP 101 notes.
H-17: See Example 1.11.13 in the CLP 101 notes.
H-18: See Example 1.11.15 in the CLP 101 notes.
H-19: See Example 1.11.15 in the CLP 101 notes.
H-20: See Example 1.11.14 in the CLP 101 notes. You may also want to review the
fundamental theorem of calculus and, in particular, Example 1.3.5 in the CLP 101 notes.

59
H-21: See Example 1.11.14 in the CLP 101 notes.

Hints for Exercises 1.12. — Jump to TABLE OF CONTENTS.

H-1: Read both the question and Theorem 1.12.17 in the CLP 101 notes very carefully.
H-2: Review Example 1.12.8 in the CLP 101 notes.
H-3: First: is the integrand unbounded, and if so, where? Second: when evaluating
integrals, always check to see if you can use a simple substitution, before trying a
complicated procedure, like partial fractions.
H-4: Is the integrand bounded?
H-5: See Example 1.12.21 in the CLP 101 notes.
H-6: Which of the two terms in the denominator is more important when x « 0? Which
one is more important when x is very large?
H-7: Which of the two terms in the denominator is more important when x is very large?
H-8: Which of the two terms in the denominator is more important when x « 0? Which
one is more important when x is very large?
H-9: What are the “problem x’s” for this integral? Get a simple approximation to the
integrand near each.
H-10: What is the limit of the integrand when x Ñ 0?
ş8 e´x
şt e´x
H-11: First find a t so that the error introduced by approximating 0 1+ x dx by 0 1+ x dx
is at most 21 10´4

Hints for Exercises 1.13. — Jump to TABLE OF CONTENTS.

H-1: The integrand is a rational function. So it is possible to use partial fractions. But
there is a much easier way!
H-2: You should prepare your own personal internal list of integration techniques
ordered from easiest to hardest. You should have associated to each technique your own
personal list of signals that you use to decide when the technique is likely to be useful.
H-3: You should prepare your own personal internal list of integration techniques
ordered from easiest to hardest. You should have associated to each technique your own
personal list of signals that you use to decide when the technique is likely to be useful.
H-4: For the integral of secant, see See §1.8.3 or Example 1.10.5 in the CLP 101 notes.
H-5: Part (b) can be done by inspection – use a little highschool geometry! Part (d) is an
improper integral.
H-6: Note that in part (d), tan´1 x means the arctangent of x, not 1 divided by tan x. This
is standard notation.

60
H-8: For part (a), see Example 1.7.11 in the CLP 101 notes. For part (d), see Example
1.10.4 in the CLP 101 notes.
H-9: For part (b), first complete the square in the denominator. For part (d) use a simple
substitution.
H-10: For part (b), complete the square in the denominator. You can save some work by
first comparing the derivative of the denominator with the numerator.
H-11: For part (b), the numerator is the derivative of a function that appears in the
denominator.
H-12: For part (a), can you convert this into a partial fractions integral? For part (b), start
by completing the square inside the square root.
H-13: For part (b), the numerator is the derivative of a function that is embedded in the
denominator.
H-14: For part (a), split the integral in two. One part may be evaluated just by
interpretting it appropriately, without doing any integration at all. For part (c), multiply
both the numerator and denominator by e x and then make a substitution.
H-15: For part (a), the substitution u = log x goves an integral that you have seen before.

Hints for Exercises 2.1. — Jump to TABLE OF CONTENTS.

H-1: See Example 2.1.2 in the CLP 101 notes.
H-2: Review Definition 2.1.1 in the CLP 101 notes.
H-3: Be careful about the units.
H-4: Suppose that the bucket is a distance y above the ground. How much work is
required to raise it an additional height dy.
H-5: See Example 2.1.4 in the CLP 101 notes.

Hints for Exercises 2.2. — Jump to TABLE OF CONTENTS.

H-1: Review the definition of “average value” in §2.2 of the CLP 101 notes.
H-5: Review the definition of “average value” in §2.2 of the CLP 101 notes.
H-6: Review the definition of “average value” in §2.2 of the CLP 101 notes.
H-7:
H-8: Review the definition of “average value” in §2.2 of the CLP 101 notes, and the
trapezoidal rule in §1.11.2 of the CLP 101 notes

Hints for Exercises 2.3. — Jump to TABLE OF CONTENTS.

61
H-1: Which method involves more work: horizontal strips or vertical strips?
H-2: Review (2.3.1) and (2.3.2) in the CLP 101 notes.
H-3: Review (2.3.1) and (2.3.2) in the CLP 101 notes. For practice, do the computation
twice — once with horizontal strips and once with vertical strips. Watch for improper
integrals.
H-4: See Example 2.3.2 in the CLP 101 notes.
H-5: See Example 2.3.1 in the CLP 101 notes.
H-6: See Example 2.3.2 in the CLP 101 notes.
H-7: See Example 2.3.2 in the CLP 101 notes.
H-8: See Example 2.3.1 in the CLP 101 notes.
H-9: You can save quite a bit of work by, firstly, exploiting symmetry and, secondly,
thinking about whether it is more efficient to use vertical strips or horizontal strips.
H-10: Sketch the region. You can save quite a bit of work by exploiting symmetry.
H-11: Draw a sketch. In part (b) be careful about the equation of the right hand boundary
of A.
H-12: Draw a sketch.

Hints for Exercises 2.4. — Jump to TABLE OF CONTENTS.

H-4: Simplify the equation.
H-5: Be careful with the arbitrary constant.
H-7: Be careful about signs.
H-8: Be careful about signs.
H-10: Move the y from the left hand side to the right hand side. Be careful about the
signs. Remember that we need y = ´1 when x = 1.
H-11: The unknown function f ( x ) satisfies an equation that involves the derivative of f .
1
H-12: Try guessing the partial fractions expansion of x ( x +1)
.

H-14: The general solution to the differential equation will contain the constant k and one
other constant. They are determined by the data given in the question.
H-15: What is the velocity of the mass at its highest point? The answer will depend on
the (unspecified) constants v0 , m, g and k.
H-16: The general solution to the differential equation will contain the constant k and one
other constant. They are determined by the data given in the question.
H-17: Review the method of integration by partial fractions and in particular Example
1.10.5 in the CLP 101 notes.

62
H-18: Review the method of integration by partial fractions and in particular Example
1.10.5 in the CLP 101 notes. Be careful about signs.
H-19: The general solution to the differential equation will contain a constant of
proportionality and one other constant. They are determined by the data given in the
question.
H-20: You do not need to know anything about investing or continuous compounding to
do this problem. You are given the differential equation explicitly. The whole first
sentence is just window dressing.
H-21: Again, you do not need to know anything about investing to do this problem. You
are given the differential equation explicitly.
H-22: Differentiate the given integral equation.
H-23: Suppose that in a very short time interval dt, the height of water in the tank
changes by dh (which is negative). Express in two different ways the volume of water
that has escaped during this time interval. Equating the two gives the needed differential
equation.
H-24: Sketch the mercury in the tank at time t, when it has height h, and also at time
t + dt, when it has height h + dh (with dh ă 0). The difference between those two
volumes is the volume of (essentially) a disk of thickness ´dh. Figure out the radius and
then the volume of that disk. This volume has to be the same as the volume of mercury
that left through the hole in the bottom of the sphere. Toricelli’s law tells you what the
volume is. Setting the two volumes equal to each other gives the differential equation
which determines h(t).
H-25: The fundamental theorem of calculus will be useful in part (b).
H-26: For any p ą 0, determine first y(t) and then the times at which y = 2, y = 1 and
y = 0. The condition that “the top half takes exactly the same amount of time to drain as
the bottom half” then gives an equation that determines p.

Hints for Exercises 3.1. — Jump to TABLE OF CONTENTS.

(
H-1: Simplify
ř8 a k . Also note that you are asked about the sequence a k , not about the
series k=1 ak .
1
H-2: What happens to n as n gets very big?
1
H-3: What happens to n as n gets very big?
1
H-4: This is trickier than it looks. Write n = x and look at the limit as x Ñ 0.
H-5: Consider an ´ L.

Hints for Exercises 3.2. — Jump to TABLE OF CONTENTS.

H-1: You should recognize this series.

63
H-2: You should recognize this series.
ÿ 
H-3: When you see ¨ ¨ ¨ k ¨ ¨ ¨ ´ ¨ ¨ ¨ k + 1 ¨ ¨ ¨ , you should immediately think
k
“telescoping series”.
ÿ 
H-4: When you see ¨ ¨ ¨ n ¨ ¨ ¨ ´ ¨ ¨ ¨ n + 1 ¨ ¨ ¨ , you should immediately think
k
“telescoping series”. But be careful not to jump to conclusions — evaluate the nth partial
sum explicitly.
H-5: Review Definition 3.2.3 in the CLP 101 notes.
H-6: This is a special case of a general series whose sum we know.
H-7: Review Example 3.2.5 in the CLP 101 notes.
H-8: Review Example 3.2.5 in the CLP 101 notes.
H-9: Review Example 3.2.5 in the CLP 101 notes.
H-10: Split the series into two parts.
H-11: Split the series into two parts.
H-12: Split the series into two parts.

Hints for Exercises 3.3. — Jump to TABLE OF CONTENTS.

H-1: Always try the divergence test first (in your head).
H-2: Review Theorem 3.3.11 and Example 3.3.12 in the CLP 101 notes.
H-3: Don’t jump to conclusions about properties of the an ’s.
H-4: Which test should you always try first (in your head)?
H-5: Review the integral test, which is Theorem 3.3.5 in the CLP 101 notes.
H-6: Review the integral test, which is Theorem 3.3.5 in the CLP 101 notes.
H-7: Review the integral test.
H-8: Try the integral test.
H-9: Review Example 3.3.9 in the CLP 101 notes.
H-10: What does the summand look like when k is very large?
H-11: What does the summand look like when n is very large?
H-12: What does the summand look like when n is very large?
H-13: This is a trick question. Be sure to verify all of the hypotheses of any convergence
test you apply.
H-14: See Example 3.3.7 in the CLP101 notes.

64
H-15: cos(nπ ) is a sneaky way to write (´1)n .
H-16: What is the behaviour for large k?
H-17: What is the behaviour for large n/m?
H-18: You should recognize this series.
H-19: Build up your own personal list of convergence tests ordered from easiest to
hardest. With each test you should also build up a list of signals that you can use to guess
whether or not it is useful to apply that test.
H-21: What does the summand look like when n is very large?
H-22: Review the alternating series test, which is given in Theorem 3.3.14 in the CLP 101
notes.
H-23: Review the alternating series test, which is given in Theorem 3.3.14 in the CLP 101
notes.
H-24: Review the alternating series test, which is given in Theorem 3.3.14 in the CLP 101
notes.
H-26: For part (a), see Example 1.12.23 in the CLP 101 notes.
For part (b), review Theorem 3.3.5 in the CLP 101 notes.
For part (c), see Example 3.3.12 in the CLP 101 notes.
H-27: Review the integral test, which is Theorem 3.3.5 in the CLP 101 notes.
8
ř
H-28: What does the fact that the series an converges guarantee about the behavior of
n =0
an for large n?
8
ř
H-29: What does the fact that the series (1 ´ an ) converges guarantee about the
n =0
behavior of an for large n?
8
ÿ nan ´ 2n + 1
H-30: What does the fact that the series converges guarantee about the
n+1
n =1
behavior of an for large n?
H-31: What does the fact that the series 8
ř
n=1 an converges guarantee about the behavior
of an for large n? When is x2 ď x?

Hints for Exercises 3.4. — Jump to TABLE OF CONTENTS.

H-1: What is this section about?
H-3: Be careful about the signs.
H-4: Does the alternating series test really apply?
H-5: What does the summand look like when n is very large?

65
H-6: For part (a), replace n by x in the absolute value of the summand. Can you integrate
the resulting function?

Hints for Exercises 3.5. — Jump to TABLE OF CONTENTS.

H-1: Review the discussion immediately following Definition 3.5.1 in the CLP 101 notes.
H-2: Review the discussion immediately following Definition 3.5.1 in the CLP 101 notes.
H-3: Review the discussion immediately following Definition 3.5.1 in the CLP 101 notes.
H-4: See Example 3.5.11 in the CLP 101 notes.
H-5: See Example 3.5.11 in the CLP 101 notes.
8  
ÿ ak a
H-11: Start part (b) by computing the partial sums of ´ k +1
a k +1 a k +2
k =1
1
H-12: You should know a power series representation for 1´x . Use it.
H-13: n ě log n for all n ě 1.
H-14: See Example 3.5.21 in the CLP 101 notes. For part (b), review §3.3.4 in the CLP 101
notes.
1
H-15: You know the geometric series expansion of 1´x . What (calculus) operation(s) can
you apply to that geometric series to convert it into the given series?
H-16: First show that the fact that the series 8
ř
n=0 (1 ´ bn ) converges guarantees that
limnÑ8 bn = 1.
H-17: What does an look like for large n?

Hints for Exercises 3.6. — Jump to TABLE OF CONTENTS.

H-1: You should know the Maclaurin series for e x . Use it.
H-2: You should know the Maclaurin series for sin x. Use it.
1
H-3: You should know the Maclaurin series for 1´x . Use it.
1
H-4: You should know the Maclaurin series for 1´x . Use it.
H-5: You should know the Maclaurin series for e x . Use it.
H-6: You should know the Maclaurin series for arctan( x ). Use it.
H-7: Review Example 3.5.20 in the CLP 101 notes.
π
H-8: ?
2 3

H-9: There is an important Taylor series, one of the series in Theorem 3.6.5 of the CLP 101
notes, that looks a lot like the given series.

66
H-10: There is an important Taylor series, one of the series in Theorem 3.6.5 of the CLP
101 notes, that looks a lot like the given series.
H-11: There is an important Taylor series, one of the series in Theorem 3.6.5 of the CLP
101 notes, that looks a lot like the given series. Be careful about the limits of summation.
H-12: There is an important Taylor series, one of the series in Theorem 3.6.5 of the CLP
101 notes, that looks a lot like the given series.
H-13: Split the series into a sum of two series. There is an important Taylor series, one of
the series in Theorem 3.6.5 of the CLP 101 notes, that looks a lot like each of the two
series.
1
H-14: You should know the Maclaurin series for 1´x . Use it.
H-15: See Example 3.5.21 in the CLP 101 notes. For parts (b) and (c), review §3.3.4 in the
CLP 101 notes.
H-16: Look at the signs of successive terms in the series.
H-17: The magic word is “series”.
H-18: You know the Maclaurin series for log(1 + y). Use it! Remember that you are asked
for a series expansion in powers of x ´ 2. So you want y to be some constant times x ´ 2.
H-19: See Example 3.6.10 in the CLP 101 notes. For parts (b) and (c), review §3.3.4 in the
CLP 101 notes.
H-20: See Example 3.6.10 in the CLP 101 notes. For part (b), review the fundamental
theorem of calculus in §1.3 of the CLP 101 notes. For part (c), review §3.3.4 in the CLP 101
notes.
H-21: See Example 3.6.10 in the CLP 101 notes. For parts (b) and (c), review §3.3.4 in the
CLP 101 notes.
H-22: See Example 3.6.10 in the CLP 101 notes. For parts (b) and (c), review §3.3.4 in the
CLP 101 notes.
H-24: Use the Maclaurin series for e x .
H-25: See Example 3.6.17 in the CLP 101 notes
H-26: See Example 3.6.17 in the CLP 101 notes
8
ÿ xn
H-28: Can you think of a way to eliminate the odd terms from e x = ?
n =0
n!

67
68
Part III

A NSWERS TO PROBLEMS

69
Answers to Exercises 1.1 — Jump to TABLE OF CONTENTS
ş7 řn  
8i 8
A-1: ´1 f ( x ) dx = limnÑ8 i =1 f ´1 + n n

A-2: n = 4, a = 2 and b = 6
A-3: It is a midpoint Riemann sum for f on the interval [1, 5] with n = 4. It is also a left
Riemann sum for f on the interval [1.5, 5.5] with n = 4. It is also a right Riemann sum for
f on the interval [0.5, 4.5] with n = 4.
A-4: f ( x ) = sin2 (2 + x ) and b = 4
?
A-5: f ( x ) = x 1 ´ x2
50 
8
5 + i ´ 1/2 15 15
ř
A-6:
i =1

A-7: 54
ş3
A-8: 0 e´x/3 cos( x ) dx
ş1
A-9: 0 xe x dx
ş2 ş3 ş1
A-10: Here are three ways: 0 e´1´x dx, 1 e´x dx, 2 0 e´1´2x dx.
ş3
A-11: 0 f ( x ) dx = 2.5
A-12: 5
A-13: 53m
ş0 ?
2
A-14: (a) There are many possible answers. Two are
ş2 a ´2 4 ´ x dx and

0 4 ´ (´2 + x )2 dx. (b) π

A-15: (a) 30 (b) 41 14
56
A-16: 3
A-17: 6
A-18: 12

Answers to Exercises 1.2 — Jump to TABLE OF CONTENTS

A-1: (a) False. For example, the function
#
0 for x ă 0
f (x) =
1 for x ě 0

provides a counterexample.
(b) False. For example, the functions f ( x ) = g( x ) = x provide a counterexample.

71
(c) False. For example, the functions f ( x ) = g( x ) = x provide a counterexample.
A-2: ´21
A-3: ´6
A-4: 20
A-5: 5
A-6: 20 + 2π
A-7: 0
A-8: 0

Answers to Exercises 1.3 — Jump to TABLE OF CONTENTS

A-1: e2 ´ e´2
x4
A-2: F ( x ) = 4 + 21 cos 2x + 21 .
şb şb
A-3: (a) True (b) False (c) False, unless a f ( x ) dx = a x f ( x ) dx = 0.
A-4: 5 ´ cos 2
A-5: 2
A-6: F1 ( x ) = log(3) G1 ( x ) = ´ log(3)
A-7: f ( x ) is increasing when ´8 ă x ă 1 and when 2 ă x ă 8.
x
A-8: F1 ( x ) = ´ cossin
3 x +6

4 )2
A-9: 4x3 e(1+ x
A-10: sin6 x + 8) cos x
A-11: F1 (1) = 3e´1
sin u
A-12:
1 + cos3 u
A-13: f (4) = 4π
2
A-14: (a) (2x + 1)e´x (b) x = ´1/2
4 3

A-15: esin x ´ esin( x ´x ) 4x3 ´ 3x2
2
5

A-16: ´2x cos e´x ´ 5x4 cos e x
a a
A-17: e x sin(e x ) ´ sin( x )
A-18: 14
A-19: f ( x ) = 2x
5
A-20: 2

72
A-21: 45 m
2

A-22: f 1 ( x ) = (2 ´ 2x ) log 1 + e2x´x and f ( x ) achieves its absolute maximum at x = 1,
because f ( x ) is increasing for x ă 1 and decreasing for x ą 1.
ş´1 dt ş x dt
A-23: The minimum is 0 1+ t 4 . As x runs from ´8 to 8, the function g ( x ) = 0 1+ t4
decreases until x reaches 1 and then increases all x ą 1. So the minimum is achieved for
x = 1. At x = 1, x2 ´ 2x = ´1.
A-24: F achieves its maximum value at x = π.
A-25: 2
A-26: log 2
ş x 3 +1 3 3 +1)3
A-27: (a) 3x2 0 et dt + 3x5 e( x (b) y = ´3( x + 1)

Answers to Exercises 1.4 — Jump to TABLE OF CONTENTS

?f (u) du
ş1
A-1: 0 2
1´u
1


A-2: 2 sin(e) ´ sin(1)
1
A-3: 3
1
A-4: ´ 100 +C
300( x3 +1)

A-5: log 4
A-6: log 2
A-7: 4/3
A-8: e6 ´ 1
A-9: 13 (4 ´ x2 )3/2 + C
A-10: 0
A-11: 21 [cos 1 ´ cos 2] « 0.478
1
A-12: sin 1
2
?
A-13: 31 [2 2 ´ 1] « 0.609

Answers to Exercises 1.5 — Jump to TABLE OF CONTENTS

ş?2  
A-1: 0x ´ ( x3 ´ x ) dx
ş4 h i
A-2: ´3/2 45 (6 ´ y2 ) + 2y dy
ş4a ? x2

A-3: 0 4ax ´ 4a dx

73
 ş25  1 1?

A-4: 1 ´ 12 ( x + 5) + 2 x dx
1
A-5: 8

4
A-6: 3

5
A-7: 3 ´ log1 2
4 
A-8: 2 π ´ 12
20
A-9: 9

1
A-10: 6
h i
A-11: 2 π ´ 14 π 2

31
A-12: 6 = 5.16̇
26
A-13: 3

Answers to Exercises 1.6 — Jump to TABLE OF CONTENTS

ş3 2
A-1: (a) π 0 xe2x dx
ş1  ?
2 ?
2  ş4 
2 ?
2 
(b) 0π 3+ y ´ 3´ y dy + 1 π 5 ´ y ´ 3 ´ y dy
ş1  2 2 ş0  a
2 a
2 
A-2: (a) ´1 π (5 ´ 4x2 ) ´ (2 ´ x2 ) dx (b) ´1 π 5+ y+1 ´ 5´ y+1 dy
ş2  2 2
A-3: π ´2 (9 ´ x 2 ) ´ ( x 2 + 1) dx
ş4a ? x2
 ş1  
2 )2 ´ (2 ´ x 2 )2 dx
A-4: (a) 0 4ax ´ 4a dx (b) ´1 π ( 5 ´ 4x
ş0  a 2 a 2 ş0 a
(c) ´1 π (5 + y + 1) ´ (5 ´ y + 1) dy = ´1 20π y + 1 dy
π  2a2 
A-5: e ´1
4
 
17e18 ´4373
A-6: π 36

h i
38
A-7: π 3 ´ 514
34
= π 512
81

ş1 ?
A-8: (a) 8π 2
´1 1 ´ x dx (b) 4π 2

A-9: (a) The region R is the region between the blue and red curves, with 3 ď x ď 5, in
the figures below.

74
 y y 2 = x2 + 15
y 2 = 8x
√ √
√ (5, 40) (5, 40)
(3, 24) R
√ R
(3, 24)

(b) 43 π « 4.19
A-10: (a) The region R is sketched below.
y
y = log x
x=2
x
h i
(b) π 4 log 2 ´ 32 « 3.998

8π 6
A-11: π 2 + 8π 3 + 5
8
A-12: 3
256ˆ8
A-13: 15 = 136.53̇
28
A-14: 3 πh
9
ş2 
2
2 
A-15: (a) 2 (b) π ´1 4´x ´ 1 + ( x ´ 1)2 dx
π π2
A-16: (a) 2 ´1 (b) 2 ´ π « 1.793
 ? 
A-17: (a) V1 = 43 πc2 (b) V2 = π3c 4 2 ´ 2 (c) c = 0 or c = 21 [23/2 ´ 1]
A-18: (a) The region R is

y y= √ 10
25−x2

3 4 x
(b) 10π log 94 = 20π log 23 (c) 20π
A-19:    
2 ´ (5 + 2π ´ 2x )2 dx + 3π/2 π (5 + 2π ´ 2x )2 ´ (5 + π sin x )2 dx
şπ ş
π/2 π ( 5 + π sin x ) π

75
Answers to Exercises 1.7 — Jump to TABLE OF CONTENTS
x2 log x x2
A-1: 2 ´ 4 +C
log x 1
A-2: ´ 6x6
´ 36x 6 +C

A-3: π
π
A-4: 2 ´1
a
A-5: y cos´1 y ´ 1 ´ y2 + C
A-6: 2y2 arctan(2y) ´ y + 12 arctan(2y) + C
35
A-7: (a) See the solution for the derivation. (b) 256 π « 0.4295

A-8: (a) π
4 ´ ln22
y y = tan−1 x

x=1

x
π2
(b) 2 ´π
A-9: 12

Answers to Exercises 1.8 — Jump to TABLE OF CONTENTS

3 x
A-1: sin x ´ sin3 +C
sin37 t 39
A-2: 37 ´ sin39 t + C
1
A-3: 7 sec7 x ´ 15 sec5 x + C
tan49 x tan47 x
A-4: + +C
49 47
π
A-5: 2
13 π
A-6: (a) See the solution for the derivation. (b) 15 ´ 4 « 0.0813

Answers to Exercises 1.9 — Jump to TABLE OF CONTENTS

4 1
A-1: (a) x = 3 sec θ (b) x = 2 sin θ (c) x = 5 tan θ
1? x
A-2: 4 x 2 +4 +C
1
A-3: ?
2 5

76
 ˇb ˇ
x2 xˇ
A-4: log ˇ 1 + 25 + 5 ˇ + C
ˇ

b
1
A-5: ´ 16 1 + 16
x2
+C
π
A-6: 6
?
d  2
x2 ´9 1 3
A-7: 9x +C = 9 1´ x +C
b
x2
A-8: 2 arcsin 2x + x 1´ 4 +C

A-9: arcsin x+ 1
2 +C
8+3π
A-10: (a) See the solution. (b) 16
?
A-11: 25x2 ´ 4 ´ 2 arcsec 5x2 +C

Answers to Exercises 1.10 — Jump to TABLE OF CONTENTS

A-1: 3
A B C D Ex + F
A-2: x´1 + ( x´1)2
+ x +1 + ( x +1)2
+ x 2 +1

A-3: log 43

A-4: ´ 1x ´ arctan x + C
A-5: 4 log |x ´ 3| ´ 2 log( x2 + 1) + C
A-6: F ( x ) = log |x ´ 2| + log |x2 + 4| + 2 arctan( x/2) + D
A-7: ´2 log |x ´ 3| + 3 log |x + 2| + C
A-8: ´9 log |x + 2| + 14 log |x + 3| + C

Answers to Exercises 1.11 — Jump to TABLE OF CONTENTS

A-1: True. Because f ( x ) is positive and concave up, the graph of f ( x ) is always below the
top of the trapezoids used in the trapezoidal rule.
2π
A-2: 3

A-3: 4.377 m3
A-4: 5403.5 cm3
A-5: 0.6865 m3
A-6: (a) 363,500 (b) 367,000
49 77
A-7: (a) 2 (b) 3

77
A-8: See the solution.
3
A-9: 100
  5  5  5  5 
1/3 1 2 4 5
A-10: (a) 3 (´3)5 +4 3 ´3 +2 3 ´3 + 4(´2)5 +2 3 ´3 +4 3 ´3 + (´1)5

(b) Simpson’s Rule results in a smaller error bound.

A-11: Any n ě 68 works.
8
A-12: 15
1 1
A-13: 180ˆ34
= 14580
 
A-14: (a) T4 = 18 1 + 2 ˆ 45 + 2 ˆ 46 + 2 ˆ 47 + 21 ,
1
 
(b) S4 = 12 1 + 4 ˆ 45 + 2 ˆ 46 + 4 ˆ 47 + 12
ˇ ˇ
24 3
(c) ˇ I ´ S4 ˇ ď 180ˆ4 4 = 5760
ˇ ˇ

A-15: (a) T4 = 8.03515, S4 = 8.03509

ˇş ˇ ˇş ˇ
ˇ b 2 83 ˇ b 4 85
(b) ˇ a f ( x ) dx ´ Tn ˇ ď 1000 = 0.00533, f ( x ) dx S = 0.00284
ˇ ˇ
12(4)2
ˇ a ´ nˇ ď 1000 180(4)4

A-16: 494 ft3

A-17: (a) 0.025635 (b) error ď 1.8 ˆ 10´5
A-18: (a) 0.6931698 (b) n ě 12 with n even
A-19: (a) 0.01345 (b) n ě 28 with n even
A-20: n ě 259
A-21: (a) 0.025635 (b) 1.8 ˆ 10´5

Answers to Exercises 1.12 — Jump to TABLE OF CONTENTS

A-1: False. For example, the functions f ( x ) = e´x and g( x ) = 1 provide a
counterexample.
1
A-2: q = 5
A-3: The integral diverges.
A-4: The integral diverges.
A-5: The integral does not converge.
A-6: The integral converges.
A-7: The integral diverges.
A-8: The integral converges.
A-9: The integral converges.

78
A-10: The integral converges.
A-11: t = 10 and n = 2042 will do the job. There are many other correct answers.

Answers to Exercises 1.13 — Jump to TABLE OF CONTENTS

1
ˇ ˇ
A-1: 2 log ˇx2 ´ 3ˇ + C
2 3e4 1
A-2: (a) 2 (b) 15 (c) 16 + 16
8
A-3: (a) 1 (b) 15
?
A-4: (a) e2 + 1 (b) log( 2 + 1) (c) log 15
13 « 0.1431
8
A-5: (a) 15 « 0.53333 (b) 49 π (c) log 2 ´ 2 + π
2 « 0.264 (d) 2 log 2 ´ 12 « 0.886
A-6: (a) 151
(b) 91 ? x2 + C (c) 12 log |x ´ 1| ´ 41 log( x2 + 1) ´ 12 tan´1 x + C
x +9
 
(d) 12 x2 tan´1 x ´ x + tan´1 x + C
b
2
A-7: (a) 121
(b) 2 sin´1 2x + x 1 ´ x4 + C (c) x log(1 + x2 ) ´ 2x + 2 tan´1 x + C

(d) ´2 log |x| + 1x + 2 log |x ´ 1| + C

ş8
A-8: (a) 0 e´x sin(2x ) dx = 25 (b) 1
?
2 2
(c) log 2 ´ 12 « 0.193

(d) log 2 ´ 12 « 0.193

A-9: (a) 12 x2 log x ´ 14 x2 + C (b) 21 log[( x + 2)2 + 1] ´ 3 arctan( x + 2) + C
(c) 21 log |x ´ 3| ´ 21 log |x ´ 1| + C (d) 13 arctan x3 + C
x´1
A-10: (a) π
4 ´ 21 log 2 (b) log |x2 ´ 2x + 5| + 21 tan´1 2 +C
x2 2 sin5 x 7
A-11: (a) log x ´ x4 + C
2 (b) ´ 300( x31+1)100 + C (c) 5 ´ sin7 x + C
?
(d) 2 arcsin 2x + 2x 4 ´ x2 + C
?
A-12: (a) ´ 14 log |e x + 1| + 14 log |e x ´ 3| + C (b) 4π
3 ´2 3
A-13: (a) 21 sec2 x + log | cos x| + C (b) 1
10 tan´1 8 « 0.1446 (c)
log 2 ´ 2 + π2 « 0.2639
A-14: (a) 94 π + 9 (b) 2 log |x ´ 2| ´ log( x2 + 4) + C (c) π2
 
A-15: (a) 12 x sin(log x ) ´ cos(log x ) + C (b) log |x ´ 3| ´ log |x ´ 2| + C

Answers to Exercises 2.1 — Jump to TABLE OF CONTENTS

1
A-1: 4 J
A-2: a = 3

79
A-3: 25
A-4: 196 J
A-5: 904,050π joules
ş3
A-6: 0 9.8 ˆ 8000(2 + z)(3 ´ z)2 dz joules

Answers to Exercises 2.2 — Jump to TABLE OF CONTENTS

A-1: 1
h i
1 2 3
A-2: e´1 9 e + 19
4
A-3: π +1
2
A-4: π
10
A-5: 3 log 7
1
A-6: 2(e´1)
1
A-7: 2

A-8: (a) 130 km (b) 65 km/hr

Answers to Exercises 2.3 — Jump to TABLE OF CONTENTS

ş0
A-1: x̄ = ´ 31 ´1 6x
2 dx

A-2: ȳ = 3
4e ´ 4e
8
A-3: ȳ = 5

A-4: (a)
y
y= √ 1
16−x2

x=2

x
3 log 3
(b) 8π
π
?
2´1
A-5: x̄ = 4?
and ȳ = ?1
2´1 4( 2´1)

k
?  k2 π 8
? 
A-6: (a) x̄ = A 2 ´ 1 , ȳ = 8A (b) k = π 2´1

80
A-7: (a) The sketch is the figure below.
y y=x−x2

(2, −2)
y=x2 −3x

8
(b) 3 (c) 1
2
A-8: π log 2 « 0.44127
12
A-9: x̄ = 0 and ȳ = 24+9π

A-10: (a) 49 π (b) x̄ = 0 and ȳ = π4

8
ş4 ş6
A-11: (a) x̄ = 11 , ȳ = 166
55 (b) π 0 y dy + π 4 (6 ´ y)2 dy
2

A-12: (a) ȳ = 4e ´ 4e 3
(b) π e2 + 2e ´ 32

Answers to Exercises 2.4 — Jump to TABLE OF CONTENTS

A-1: y = log( x2 + 2)
?
A-2: y( x ) = 3 1 + x2
A-3: y(t) = 3 log C+´3
sin t .
b
A-4: y = 3 32 e x + C.
2

x2

x2
A-5: y = ´ log C ´ 2 . The solution only exists for C ´ 2 ą 0, i.e. for C ą 0 and
?
|x| ă 2C.
A-6: y = (3e x ´ 3x2 + 24)1/3
A-7: y = f ( x ) = ´ ? 21
x +16
?
A-8: y = 10x3 + 4x2 + 6x ´ 4
4 /4
A-9: y( x ) = e x
1
A-10: y = 1´2x
2
A-11: f ( x ) = e1+ x /2
b
A-12: y( x ) = 4 + 2 log x2x +1 . Note that, to satisfy y (1) = 2, we need the positive square
root.
y2  3/2
A-13: 2 + 31 y2 ´ 4 = sec x + 1
A-14: 12 weeks

81
 b
m
b
k


A-15: t = kg tan´1 mg v0
1
A-16: (a) k = 400 (b) t = 70sec
3´4ekt
A-17: (a) x (t) = 1´2ekt
(b) As t Ñ 8, x Ñ 2.
4
A-18: (a) P = 1+e´4t
(b) At t = 12 , P = 3.523. As t Ñ 8, P Ñ 4.
dv 400
A-19: (a) dt = ´kv2 (b) v = t +1 (c) t = 7
A-20: (a) B(t) = C e0.06t´0.02 cos t with the arbitrary constant C ě 0. (b) $1159.89
A-21: (a) B(t) = t30000 ´ 50mu et/50 + 50m (b) $600
4´e1´cos x
A-22: y( x ) = 2´e1´cos x
. The largest allowed interval is

´ cos´1 (1 ´ log 2) ă x ă cos´1 (1 ´ log 2) « 1.259

b
A-23: 180, 000 3g « 99, 591sec « 27.66hr
b
4ˆ144 125
A-24: t = 15 2g « 2, 394 sec « 0.665 hr

4´e x
A-25: (a) 3 (b) y1 = (y ´ 1)(y ´ 2) (c) f ( x ) = 2´e x
1
A-26: p = 4

Answers to Exercises 3.1 — Jump to TABLE OF CONTENTS

A-1: lim ak = 0.
kÑ8

A-2: The sequence converges to 0.

A-3: 9
A-4: log 2
A-5: See the solution.

Answers to Exercises 3.2 — Jump to TABLE OF CONTENTS

1
A-1: 7ˆ86
3
A-2: 2
A-3: 6
 
A-4: cos π
3 ´ cos(0) = ´ 12

A-5: (a) an 16n2 +1124n+5 (b) 3

4
24
A-6: 5

82
 7
A-7: 30
263
A-8: 99
321 107
A-9: 999 = 333

A-10: 3
1
A-11: 2 + 75 = 17
14
40
A-12: 3 = 13 13

Answers to Exercises 3.3 — Jump to TABLE OF CONTENTS

A-1: No. It diverges.
2n
A-2: bn = n
3
8
ř 1
A-3: (a) In general false. The harmonic series n provides a counterexample.
n =1
8 8
(b) In general false. If an = (´1)n n1 , then (´1)n an is again the harmonic series 1
ř ř
n,
n =1 n =1
which diverges.
(c) In general false. Take, for example, an = 0 and bn = 1.
A-4: It diverges.
A-5: The series diverges.
A-6: See the solution.
A-7: p ą 1
A-8: It converges.
A-9: It converges.
A-10: The series converges.
A-11: It diverges.
A-12: It converges absolutely.
A-13: It diverges.
A-14: It converges absolutely.
A-15: (a) diverges (b) converges
A-16: diverges
A-17: (a) converges (b) diverges
1
A-18: 7

83
A-19: (a) diverges by comparison with the harmonic series
(b) converges by the ratio test
1
A-20: (a) Converges by the limiting comparison test with b = k5/3
.
(b) Diverges by the ratio test.
(c) Diverges by the integral test.
A-21: It converges.
A-22: N = 5
A-23: N ě 999
1
A-24: We need n = 4 and then S4 = 32
´ 512 + 1
72
´ 912
A-25: (a) converges (b) converges
A-26: (a) See the solution.
x +sin x
(b) f ( x ) = 1+ x 2
is not a decreasing function.
(c) See the solution.
A-27: The sum is between 0.9035 and 0.9535.
A-28: See the solution.
A-29: It diverges.
A-30: It converges to ´ log 2 = log 12 ,
A-31: See the solution.

Answers to Exercises 3.4 — Jump to TABLE OF CONTENTS

1
A-1: False. For example, bn = n provides a counterexample.
A-2: conditionally convergent
A-3: The series diverges.
A-4: It diverges.
A-5: It converges absolutely.
3
A-6: (a) See the solution. (b) |S ´ S5 | ď 24 ˆ 36e´6

Answers to Exercises 3.5 — Jump to TABLE OF CONTENTS

1 2 1
A-1: (a) R = 2 (b) 1+2x for all |x| ă 2
A-2: R = 8
A-3: 1

84
A-4: The interval of convergence is ´1 ă x + 2 ď 1 or (´3, ´1].
A-5: The interval of convergence is ´4 ă x ď 2, or simply (´4, 2].
A-6: ´3 ď x ă 7 or [´3, 7)
A-7: The given series converges if and only if ´3 ď x ď ´1. Equivalently, the series has
interval of convergence [´3, ´1].


A-8: The interval of convergence is 43 ď x ă 45 , or 34 , 54 .

 radius of convergence is 2. The interval of convergence is ´1 ă x ď 3, or

A-9: The
´ 1, 3 .


A-10: The interval of convergence is a ´ 1 ă x ă a + 1, or a ´ 1, a + 1 .
A-11: (a) |x + 1| ď 9 or ´10 ď x ď 8 or [´10, 8] (b) This series converges only for
x = 1.
8 8
x n +3 = xn
ř ř
A-12:
n =0 n =3

A-13: The series converges absolutely for |x| ă 9, converges conditionally for x = ´9 and
diverges otherwise.
8 3n+1
(´1)n x3n+1 + C
ř
A-14: (a) (b) We need to keep two terms (the n = 0 and n = 1 terms).
n =0

A-15: (a) See the solution.

8
x (1+ x )
n2 x n =
ř
(b) (1´x )3
. The series converges for ´1 ă x ă 1.
n =0

A-16: See the solution.

A-17: (a) 1. (b) The series converges for ´1 ď x ă 1, i.e. for the interval [´1, 1)

Answers to Exercises 3.6 — Jump to TABLE OF CONTENTS

35
A-1: c5 = 5!

A-2: a = 1, b = ´ 3!1 = ´ 61 .
8
2n x n
ř
A-3: ´
n =0

A-4: bn = 3(´1)n + 2n
ş ´x2 2 4
A-5: e x ´1 dx = C ´ x2 + x8 + ¨ ¨ ¨ . It is not clear from the wording of the question
whether or not the arbitrary constant C is to be counted as one of the “first two nonzero
terms”.
8 2n+1 2n+6 8 2n 2n+6
(´1)n (2n2 +1)(x2n+6) + C = (´1)n (2n2+1x)(n+3) + C
ř ř
A-6:
n =0 n =0

85
 8
ÿ 2n +1 x n +1
A-7: (´1)n for all |x| ă 1
2
n =0
n+1

A-8:
1
A-9: e

A-10: e1/e
A-11: e1/π ´ 1
A-12: log(3/2)
A-13: (e + 2)ee ´ 2
8 n
(´1)n 3n3+2 x3n+2
ř
A-14: f ( x ) = 1 +
n =0
8
x4n+1
(´1)n
ř
A-15: (a) 4n+1 (b) 0.493967
n =0

(c) The approximate value of part (b) is larger than the true value of I (1/2)
1
A-16: 66
A-17: Any interval of length 0.0002 that contains 0.03592 and 0.03600 is fine.
8
ř (´1)n´1 n
A-18: log( x ) = log 2 + n 2n ( x ´ 2 ) . It converges when 0 ă x ď 4.
n =1
8
xn
(´1)n
ř
A-19: (a) n n! (b) ´0.80 (c) See the solution.
n =1
8
x2n+1
A-20: (a) Σ( x ) = (´1)n
ř
(2n+1)(2n+1)!
(b) x = π (c) 1.8525
n =0
8 2n´1
(´1)n (2nx)!(2n´1) (b) I (1) = ´ 21 + 1 1
ř
A-21: (a) I ( x ) = 4!3 ˘ 6!5 = ´0.486 ˘ 0.001
n =1

(c) I (1) ă ´ 21 + 1
4!3
1 1 3
A-22: (a) 2! x ´ 4! x + 6!1 x5 ´ 8!1 x8 + ¨ ¨ ¨ (b) 0.460 (c) I (1) ă 1
2! ´ 4!1 + 1
6! ă 0.460
A-23: (a) See the solution. (b) The series converges for all x.
A-24: See the solution.
A-25: ´1
1 1
A-26: 5! = 120

A-27: ´ 61
60
1 1


A-28: (a) See the solution. (b) 2 e+ e
8 8
ř xn ř x2n
A-29: (a) cosh( x ) = n! = (2n)!
for all x.
n =0 n =0
n even

86
Part IV

S OLUTIONS TO PROBLEMS

87
Solutions to Exercises 1.1 — Jump to TABLE OF CONTENTS

S-1: In the given integral, the domain of integration runs from a = ´1 to b = 7. So we

have ∆x = (b ´ a)/n = (7 ´ (´1))/n = 8/n. The left hand end of the first subinterval is
at x0 = a = ´1. So the right hand end of the ith interval is at xi = ´1 + 8i/n. So
ż7 n  
ÿ 8i 8
f ( x ) dx = lim f ´1 +
´1 nÑ8 n n
i =1

şb
S-2: In general the left Riemann sum for the integral a f ( x ) dx is of the form
n  
ÿ b´a b´a
f a + ( k ´ 1)
n n
k =1

So
• To get the limits of summation to match the given sum, we need n = 4.
• Then to get the factor multiplying f to match that in the given sum, we need
b´a
n = 1 or b ´ a = 4.
• , Finally, to get the argument of f to match that in the given sum, we need
b´a b´a b´a
a + ( k ´ 1) = a´ +k = 1+k
n n n
Subbing in n = 4 and b ´ a = 4, gives a ´ 1 + k = 1 + k, so a = 2 and b = 6.

3
ř
S-3: f (1.5 + k ) ¨ 1 is a midpoint Riemann sum for f on the interval [1, 5] with n = 4. It
k =0
is also a left Riemann sum for f on the interval [1.5, 5.5] with n = 4. It is also a right
Riemann sum for f on the interval [0.5, 4.5] with n = 4.

S-4: We identify the given sum as the right Riemann sum in=1 f ( a + i∆x )∆x, with a = 0
ř
(that’s specified in the statement of the question), interval ∆x = 4/n, xi = a + i∆x = 4i/n
and f ( x ) = sin2 (2 + x ). So b = a + n∆x = 4.

S-5: The given sum is of the form

n n n
c c
ÿ k k2 ÿ k k2 1 ÿ
lim 2
1 ´ 2 = lim 1 ´ 2 = lim f ( xk˚ )∆x
nÑ8 n n nÑ8 n n n nÑ8
k =1 k =1 k =1
?
with ∆x = n1 , a = 0, xk˚ = nk = a + k∆x and f ( x ) = x 1 ´ x2 . Since x0˚ = 0 and xn˚ = 1, the
ş1
right hand side is the definition (using the right Riemann sum) of 0 f ( x ) dx.

89
S-6: In general the midpoint Riemann sum is
n
ÿ 
 b´a
f a + i ´ /2 ∆x ∆x
1 ∆x =
n
i =1

In this problem we are told that f ( x ) = x8 , a = 5, b = 15 and n = 50, so that

∆x = b´a 1
n = 5 and the desired Riemann sum is
50 
ÿ
1 8 1
5+ i ´ 1/2
5 5
i =1

S-7: The given integral has interval of integration going from a = ´1 to b = 5. So when
we use three approximating rectangles, all of the same width, the common width is
∆x = b´a
n = 2. The first rectangle has left hand end point x0 = a = ´1, the second has left
hand endpoint x1 = a + ∆x = 1, and the third has left hand end point x2 = a + 2∆x = 3.
So
ż5
   
x3 dx « f ( x0 ) + f ( x1 ) + f ( x2 ) ∆x = (´1)3 + 13 + 33 ˆ 2 = 54
´1

S-8: As i ranges from 1 to n, 3i/n range from 3/n to 3 with jumps of ∆x = 3/n, so this is
n n żb
ÿ 3 ´i/n ÿ
lim e cos(3i/n) = lim f ( xi )∆x = f ( x ) dx
nÑ8 n nÑ8 a
i =1 i =1

where xi = 3i/n, f ( x ) = e´x/3 cos( x ), a = x0 = 0 and b = xn = 3. Thus

n ż3
ÿ 3 ´i/n
lim e cos(3i/n) = e´x/3 cos( x ) dx
nÑ8 n 0
i =1

S-9: As i ranges from 1 to n, the exponent i/n range from 1/n to 1 with jumps of
∆x = 1/n. So let’s try xi = i/n, ∆x = n1 . Then
n n n n
ÿ iei/n ÿ i i/n 1 ÿ
xi
ÿ
Rn = = e = x i e ∆x = f ( xi )∆x
n2 n n
i =1 i =1 i =1 i =1

with f ( x ) = xe x , and the limit

n
ÿ żb
lim Rn = lim f ( xi )∆x = f ( x ) dx
nÑ8 nÑ8 a
i =1

where a = x0 = 0 and b = xn = 1. Thus

ż1
lim Rn = xe x dx
nÑ8 0

90
S-10: choice #1: If we set ∆x = n2 and xi = 2in , i.e. xi = a + i∆x with a = 0, then
ÿ n  ÿ n 
´1´2i/n 2
lim e ¨ = lim e ´1´xi
∆x
nÑ8 n nÑ8
i =1 i =1
ÿ n 
= lim f ( xi )∆x with f ( x ) = e´1´x
nÑ8
i =1
żb
= f ( x ) dx with a = x0 = 0 and b = xn = 2
a
ż2
= e´1´x dx
0

choice #2: If we set ∆x = n2 and xi = 1 + 2in , i.e. xi = a + i∆x with a = 1, then

ÿ n  ÿ n 
´1´2i/n 2
lim e ¨ = lim e ∆x
´xi
nÑ8 n nÑ8
i =1 i =1
ÿ n 
= lim f ( xi )∆x with f ( x ) = e´x
nÑ8
i =1
żb
= f ( x ) dx with a = x0 = 1 and b = xn = 3
a
ż3
= e´x dx
1

choice #3: If we set ∆x = n1 and xi = ni , i.e. xi = a + i∆x with a = 0, then

ÿ n  ÿ n 
´1´2i/n 2 ´1´2xi
lim e ¨ = lim e 2∆x
nÑ8 n nÑ8
i =1 i =1
ÿ n 
= lim f ( xi )∆x with f ( x ) = 2e´1´2x
nÑ8
i =1
żb
= f ( x ) dx with a = x0 = 0 and b = xn = 1
a
ż1
=2 e´1´2x dx
0

S-11: Here is a sketch the graph of f ( x ).

y
1 y = f (x)

1 3 x

91
There a linear increase from x = 0 to x = 1, followed by a constant. Using the
ş3
interpretation of 0 f ( x ) dx as the area between y = f ( x ) and the x–axis with x between 0
and 3, we can break this area into:
ş1
• 0 f ( x ) dx: a right-angled triangle of height 1 and base 1 and hence area 0.5.
ş3
• 1 f ( x ) dx: a rectangle of height 1 and base 2 and hence area 2.
ş3
Summing up: 0 f ( x ) dx = 2.5.

S-12: Recall that

#
´x if x ď 0
|x| =
x if x ě 0
so that
#
´2x if x ď 0
|2x| =
2x if x ě 0
To picture the geometric figure whose area the integral represents observe that
• at the left hand end of the domain of integration x = ´1 and the integrand
|2x| = | ´ 2| = 2 and
• as x increases from ´1 towards 0, the integrand |2x| = ´2x decreases linearly, until
• when x hits 0 the integrand hits |2x| = |0| = 0 and then
• as x increases from 0, the integrand |2x| = 2x increases linearly, until
• when x hits +2, the right hand end of the domain of integration, the integrand hits
|2x| = |4| = 4.
ş2
So the integral ´1 |2x| dx is the area of the union of the two shaded triangles (one of base
1 and of height 2 and the other of base 2 and height 4) in the figure on the right below and
y
4
ż2
1 1
|2x| dx = ˆ1ˆ2+ ˆ2ˆ4 = 5
´1 2 2 2

−1 2x

S-13: The car’s speed increases with time. So its highest speed on any time interval
occurs at the right hand end of the interval and the best possible upper estimate for the
distance traveled is given by the right Riemann sum with ∆x = 0.5, which is
   
v(0.5) + v(1.0) + v(1.5) + v(2.0) ˆ 0.5 = 14 + 22 + 30 + 40 ˆ 0.5 = 53 m

92
S-14: (a, solution #1) Set xi = ´2 + 2in . Then a = x0 = ´2 and b = xn = 0 and ∆x = n2 . So
 
d
n n
ÿ 2 2i 2 ÿ a 2
lim 4 ´ ´2 + = lim f ( xi )∆x with f ( x ) = 4 ´ x2 and ∆x =
nÑ8 n n nÑ8 n
i =1 i =1
ż0 a
= 4 ´ x2 dx
´2

(a, solution #2) Set xi = 2in . Then a = x0 = 0 and b = xn = 2 and ∆x = n2 . So

 
d
n n
2 2i 2 2
ÿ ÿ b
lim 4 ´ ´2 + = lim f ( xi )∆x with f ( x ) = 4 ´ (´2 + x )2 , ∆x =
nÑ8 n n nÑ8 n
i =1 i =1
ż2b
= 4 ´ (´2 + x )2 dx
0
ş0 ? ?
(b) For the integral ´2 4 ´ x2 dx, y = 4 ´ x2 is equivalent to x2 + y2 = 4, y ě 0. So the
integral represents the area between the upper half of the circle x2 + y2 = 4 (which has
radius 2) and the x-axis with ´2 ď x ď 0, which is a quarter circle with area 14 π22 = π.

S-15: (a) The left Riemann sum is defined as

n
ÿ
Ln = f ( xi´1 )∆x with xi = a + i∆x
i =1

We subdivide into n = 3 intervals, so that ∆x = b´a 3´0

n = 3 = 1, x0 = 0, x1 = 1 and
x2 = 2. The function f ( x ) = 7 + x3 has the values f ( x0 ) = 7 + 03 = 7,
f ( x1 ) = 7 + 13 = 8, and f ( x2 ) = 7 + 23 = 15, from which we evaluate
   
L3 = f ( x0 ) + f ( x1 ) + f ( x2 ) ∆x = 7 + 8 + 15 ˆ 1 = 30

(b) We divide into n intervals so that ∆x = b´a 3 3i

n = n and xi = a + i∆x = n . The right
Riemann sum is therefore:
n n   n  
ÿ ÿ (3i )3 3 ÿ 21 81 i3
Rn = f ( xi )∆x = 7+ 3 = + 4
n n n n
i =1 i =1 i =1
To calculate the sum:
! !
n n
21 ÿ 81 ÿ 3
Rn = 1 + i
n
i =1
n 4 i =1
21 81 n4 + 2n3 + n2 81
= ˆn+ 4 ˆ = 21 + (1 + 2/n + 1/n2 )
n n 4 4
To evaluate the limit exactly, we take n Ñ 8. The expressions involving 1/n vanish
leaving:
ż3
81 1
(7 + x3 ) dx = lim Rn = 21 + = 41
0 nÑ8 4 4

93
S-16: In general, the n slice, the right–endpoint Riemann sum approximation to the
şb
integral a f ( x ) dx is

n
ÿ
f ( a + i∆x )∆x
i =1

where ∆x = b´a n . In this problem, a = 2, b = 4, and f ( x ) = x , so that ∆x =

2 2
n and the n
slice, the right–endpoint Riemann sum approximation becomes

n 
2i 2 2
n n n
ÿ ÿ 8 ÿ 16i ÿ 8i2
2+ = + +
n n n n2 n3
i =1 i =1 i =1 i =1
8 16 n(n + 1) 8 (n + 1)(2n + 1)
= n+ 2 + 3n
n n 2 n 6
 1  4 1  1
= 8+8 1+ + 1+ 2+
n 3 n n

So
ż4 h  1  4 1 1 i 4 56
x2 dx = lim 8 + 8 1 + + 1+ 2+ = 8+8+ ˆ2 =
2 nÑ8 n 3 n n 3 3

S-17: We’ll use right Riemann sums with a = 0 and b = 2. When there are n strips
∆x = b´a 2
n = n and xi = a + i∆x = 2i/n. So we need to evaluate

  !
n   n
ÿ ÿ 2i 3 2i 2
lim xi3 + xi ∆x = lim +
nÑ8 nÑ8 n n n
i =1 i =1
n  
2 ÿ 8i3 2i
= lim +
nÑ8 n n3 n
i =1
!
n n
16 ÿ 3 4 ÿ
= lim i + 2 i
nÑ8 n4 n
i =1 i =1
 
16(n4 + 2n3 + n2 ) 4(n2 + n)
= lim +
nÑ8 n4 ¨ 4 n2 ¨ 2
    
16 2 1 4 1
= lim 1+ + 2 + 1+
nÑ8 4 n n 2 n
16 4
= + = 6.
4 2

S-18: We’ll use right Riemann sums with a = 1, b = 4 and f ( x ) = 2x ´ 1. When there are

94
n strips ∆x = b´a
n = 3
nand xi = a + i∆x = 1 + 3i/n. So we need to evaluate
n n n  
ÿ ÿ ÿ 6i 3
lim f ( xi )∆x = lim (2xi ´ 1) ∆x = lim 2+ ´1
nÑ8 nÑ8 nÑ8 n n
i =1 i =1 i =1
n  
3 ÿ 6i
= lim +1
nÑ8 n n
i =1
!
n n
18 ÿ 3 ÿ
= lim i+ 1
nÑ8 n2 n
i =1 i =1
 
18 ¨ n(n + 1) 3
= lim + n
nÑ8 n2 ¨ 2 n
   
1
= lim 9 1 + +3
nÑ8 n
= 9 + 3 = 12.

Solutions to Exercises 1.2 — Jump to TABLE OF CONTENTS

S-1: (a) False. For example if

#
0 for x ă 0
f (x) =
1 for x ě 0
ş´2 ş2
then ´3 f ( x )dx = 0 and 3 f ( x )dx = 1.
(b) False. For example, if f ( x ) = g( x ) = x, then
ż ´2 ż ´2  2 ´2
x 4 9 5
f ( x ) dx = x dx = = ´ =´
´3 ´3 2 ´3 2 2 2
while
ż3 ż3  3
x2 9 4 5
f ( x ) dx = x dx = = ´ =
2 2 2 2 2 2 2
ş1 ş1 1
(c) False. For example, if f ( x ) = g( x ) = x, then 0 f ( x ) ¨ g( x ) dx = 0 x2 dx = 3 and
ş1 ş1 ş1 ş1 1 1
0 f ( x ) dx ¨ 0 g ( x ) dx = 0 x dx ¨ 0 x dx = 2 ¨ 2 .

S-2: The operation of integration is linear (that’s part (d) of the “arithmetic of
integration” Theorem 1.2.1 in the CLP 101 notes), so that:
ż3 ż3 ż3
[6 f ( x ) ´ 3g( x )] dx = 6 f ( x ) dx ´ 3g( x ) dx
2 2 2
ż3 ż3
=6 f ( x ) dx ´ 3 g( x ) dx = (6 ˆ (´1)) ´ (3 ˆ 5) = ´21
2 2

95
S-3: The operation of integration is linear (that’s part (d) of the “arithmetic of
integration” Theorem 1.2.1 in the CLP 101 notes), so that:
ż2 ż2 ż2
[2 f ( x ) + 3g( x )] dx = 3g( x ) dx
2 f ( x ) dx +
0 0 0
ż2 ż2
=2 f ( x ) dx + 3 g( x ) dx = (2 ˆ 3) + (3 ˆ (´4)) = ´6
0 0

S-4: Using part (d) of the “arithmetic of integration” Theorem 1.2.1, followed by parts (c)
and (b) of the “arithmetic for the domain of integration” Theorem 1.2.3 in the CLP 101
notes,
ż2 ż2 ż2
 
3g( x ) ´ f ( x ) dx = 3 g( x ) dx ´ f ( x ) dx
´1 ´1 ´1
ż0 ż2 ż0 ż2
=3 g( x ) dx + 3 g( x ) dx ´ f ( x ) dx ´ f ( x ) dx
´1 0 ´1 0
ż0 ż2 ż ´1 ż2
=3 g( x ) dx + 3 g( x ) dx + f ( x ) dx ´ f ( x ) dx
´1 0 0 0
= 3 ˆ 3 + 3 ˆ 4 + 1 ´ 2 = 20

S-5: Recall that

#
´x if x ď 0
|x| =
x if x ě 0

so that
#
´2x if x ď 0
|2x| =
2x if x ě 0

Also recall, from Example 1.2.5 in the CLP 101 notes that
żb
b2 ´ a2
x dx =
a 2

So
ż2 ż0 ż2 ż0 ż2
|2x| dx = |2x| dx + |2x| dx = (´2x ) dx + 2x dx
´1 ´1 0 ´1 0
ż0 ż2
02 ´ (´1)2 22 ´ 02
= ´2 x dx + 2 x dx = ´2 +
´1 0 2 2
= 1+4 = 5

96
S-6: We first use additivity:
ż2  a  ż2 ż2 a
2
5 + 4 ´ x dx = 5 dx + 4 ´ x2 dx
´2 ´2 ´2

The first integral represents the area of a rectangle of height 5 and width 4 and so equals
? second integral represents the area above the x–axis and below the curve
20. The
y = 4 ´ x2 or x2 + y2 = 4. That is a semicircle of radius 2, which has area 12 π22 . So
ż2  a 
2
5 + 4 ´ x dx = 20 + 2π
´2

sin x
S-7: Note that the integrand f ( x ) = log(3+ x2 )
is an odd function because

sin(´x ) ´ sin x
f (´x ) = = = ´ f (x)
log(3 + (´x )2 ) log(3 + x2 )

The domain of integration ´2012 ď x ď 2012 is symmetric about x = 0. So, by Theorem

1.2.11 of the CLP notes,
ż +2012
sin x
2
dx = 0
´2012 log(3 + x )

S-8: Note that the integrand f ( x ) = x1/3 cos x is an odd function because

f (´x ) = (´x )1/3 cos(´x ) = ´x1/3 cos x = ´ f ( x )

The domain of integration ´2012 ď x ď 2012 is symmetric about x = 0. So, by Theorem

1.2.11 of the CLP notes,
ż +2012
x1/3 cos x dx = 0
´2012

Solutions to Exercises 1.3 — Jump to TABLE OF CONTENTS

S-1: The fundamental theorem of calculus tells us that

ż ?5 ?
f ( x ) dx = F ( 5) ´ F (1)
1
? 2
2

= e( 5 ´3
+ 1 ´ e(1 ´3) + 1
= e5´3 ´ e1´3 = e2 ´ e´2

97
 x4
S-2: One function with derivative x3 is 4. One function with derivative sin 2x is
4
cos 2x. So the general antiderivative of f ( x ) is x4 + 21 cos 2x + C. To satisfy F (0) = 1,
´ 21
we need1
h x4 1 i 1 1
+ cos 2x + C = 1 ðñ + C = 1 ðñ C =
4 2 x =0 2 2
x4
So F ( x ) = 4 + 12 cos 2x + 12 .

S-3: (a) This is true, by part 2 of the fundamental theorem of calculus, Thereom 1.3.1 in
the CLP 101 notes, with G ( x ) = f ( x ) and f ( x ) replaced by f 1 ( x ).
(b) This is not only false, but it makes no sense at all. The integrand is strictly positive so
the integral has to be strictly positive. In fact it’s +8. The fundamental theorem of
calculus does not apply because the integrand has a singularity at x = 0.
şb şb
(c) This is not only false, but it makes no sense at all, unless a f ( x ) dx = a x f ( x ) dx = 0.
The left hand side is a number. The right hand side is a number times x. For example, if
ş1
a = 0, b = 1 and f ( x ) = 1, then the left hand side is 0 x dx = 21 and the right hand side is
ş1
x 0 dx = x.

S-4: By the fundamental theorem of calculus,

ż2  4 ˇ2
3 x ˇ 24
x + sin x ) dx = ´ cos x ˇˇ = ´ cos 2 + cos 0 = 4 ´ cos 2 + 1 = 5 ´ cos 2.
0 4 0 4

S-5: By part (d) of our “Arithmetic of Integration” theorem, Theorem 1.2.1 in the CLP 101
notes, ż2h
2i
ż2 2 ż2 ż2
x +2 1
2
dx = 1 + 2 dx = dx + 2 2
dx
1 x 1 x 1 1 x
Then by the fundamental theorem of calculus,
ż2 ż2
1 h i2 h 1 i2   h 1 i
dx + 2 2
dx = x + 2 ´ = 2´1 +2 ´ +1 = 2
1 1 x 1 x 1 2

S-6: By the fundamental theorem of calculus,

d x
ż
1
F (x) = log(2 + sin t) dt = log(2 + sin x )
dx 0
 ży 
1 d
G (y) = ´ log(2 + sin t) dt = ´ log(2 + sin y)
dy 0

1 The symbol ðñ is read “if and only if”. This is used in mathematics to express the logical equivalence
of two statements. To be more precise, the statement P ðñ Q tells us that P is true whenever Q is
true and Q is true whenever P is true.

98
So π π
F1 = log 3 G1 = ´ log(3)
2 2

S-7: By the fundamental theorem of calculus,

2 2
f 1 ( x ) = 100( x2 ´ 3x + 2)e´x = 100( x ´ 1)( x ´ 2)e´x
2
As f ( x ) is increasing whenever f 1 ( x ) ą 0 and 100e´x is always strictly bigger than 0, we
have f ( x ) increasing if and only if ( x ´ 1)( x ´ 2) ą 0, which is the case if and only if
( x ´ 1) and ( x ´ 2) are of the same sign. Both are positive when x ą 2 and both are
negative when x ă 1. So f ( x ) is increasing when ´8 ă x ă 1 and when 2 ă x ă 8.
żx
1 1
S-8: Write G ( x ) = 3
dt. By the fundamental theorem of calculus, G1 ( x ) = x 3 +6
.
0 t +6
Since F ( x ) = G (cos x ), the chain rule gives
sin x
F1 ( x ) = G1 (cos x ) ¨ (´ sin x ) = ´
cos3 x + 6

şx 2 2
S-9: Define g( x ) = 0 et dt. By the fundamental theorem of calculus, g1 ( x ) = e x . As
f ( x ) = g(1 + x4 ) the chain rule gives
4 )2
f 1 ( x ) = 4x3 g1 (1 + x4 ) = 4x3 e(1+ x

şx
S-10: Define g( x ) = 0 (t6 + 8)dt. By the fundamental theorem of calculus, g1 ( x ) = x6 + 8.
We are to compute the derivative of f ( x ) = g(sin x ). The chain rule gives
ż sin x !
d

(t6 + 8)dt = g1 (sin x ) cos x = sin6 x + 8 cos x
dx 0

şx

´t sin πt dt. By the fundamental theorem of calculus,
S-11: Let G ( x ) = 0
e 2
3 πx3


G1 ( x ) = e´x sin πx
2
and, since F ( x ) = G ( x3 ), F1 ( x ) = 3x2 G1 ( x3 ) = 3x2 e´x sin 2 and
F1 (1) = 3e´1 sin π2 = 3e´1 .

ż0 żx
dt 1 1
S-12: Define g( x ) = 3
=´ 3
dt, so that g1 ( x ) = ´ by the
x 1+t 0 1+t 1 + x3
fundamental theorem of calculus. Then by the chain rule,
ż 0 
d dt d d 1
3
= g(cos u) = g1 (cos u) ¨ cos u = ´ ¨ (´ sin u).
du cos u 1 + t du du 1 + cos3 u

99
 d
şx
S-13: Apply dx to both sides of x sin(πx ) = 0 f (t) dt. Then, by the fundamental theorem
of calculus
żx
d ( d d (
x sin(πx ) = f (t) dt = f ( x ) ùñ f ( x ) = x sin(πx ) = sin(πx ) + πx cos(πx )
dx dx 0 dx
ùñ f (4) = sin(4π ) + 4π cos(4π ) = 4π

S-14: (a) Write

ży ży
2 ´t 2
F ( x ) = G ( x ) ´ H (´x ) with G (y) = e dt, H (y) = e´t dt
0 0

By the Fundamental Theorem of Calculus,

2
G1 (y) = e´y H 1 (y) = e´y

Hence, by the chain rule,

2 2 2
F1 ( x ) = 2xG1 ( x2 ) ´ (´1) H 1 (´x ) = 2xe´( x ) + e´(´x) = (2x + 1)e´x

(b) Observe that F1 ( x ) ă 0 for x ă ´1/2 and F1 ( x ) ą 0 for x ą ´1/2. Hence F ( x ) is

decreasing for x ă ´1/2 and increasing for x ą ´1/2, and F ( x ) must take its minimum
value when x = ´1/2.

S-15: Write
żx ż0 żx ż x4 ´x3
sin t sin t sin t
F(x) = e dt + e dt = e dt ´ esin t dt
0 x4 ´x3 0 0
4 3
= G(x) ´ G(x ´ x )

with ży
G (y) = esin t dt
0
By the Fundamental Theorem of Calculus,

G1 (y) = esin y

Hence, by the chain rule,

d

F1 ( x ) = G1 ( x ) ´ G1 ( x4 ´ x3 ) x4 ´ x3
dx
= G ( x ) ´ G ( x ´ x ) (4x3 ´ 3x2 )
1 1 4 3

4 3

= esin x ´ esin(x ´x ) 4x3 ´ 3x2

100
S-16: Write
ż ´x2 ż ´x2 ż0
t

t



F(x) = cos e dt = cos e dt + cos et dt
x5 0 x5
ż ´x2 ż x5



= cos et dt ´ cos et dt
0 0
2 5
= G (´x ) ´ G ( x )
with ży


G (y) = cos et dt
0
By the Fundamental Theorem of Calculus,


G1 (y) = cos ey

Hence, by the chain rule,

d
d

F1 ( x ) = G1 (´x2 ) ´ x2 ´ G1 ( x5 ) x5
dx dx
= G1 (´x2 ) (´2x ) ´ G1 ( x5 ) (5x4 )
2
5

= ´2x cos e´x ´ 5x4 cos e x

S-17: Write
ż ex ?
F(x) = sin t dt
x
ż ex ? ż0? ż ex ? żx?
= sin t dt + sin t dt = sin t dt ´ sin t dt
0 x 0 0
x
= G (e ) ´ G ( x )
with ży?
G (y) = sin t dt
0
By the Fundamental Theorem of Calculus,
a
G1 (y) = sin y

Hence, by the chain rule,

d x

F1 ( x ) = G1 (e x ) e ´ G1 ( x )
dx
= e x G1 (e x ) ´ G1 ( x )
b b
= e x sin(e x ) ´ sin( x )

101
S-18: Splitting up the domain of integration
ż5 ż3 ż5
f ( x ) dx = f ( x ) dx + f ( x ) dx
1 1 3
ż3 ż5
= 3 dx + x dx
1 3
ˇ x =3 ˇ x =5
ˇ x2 ˇˇ
= 3xˇ
ˇ + ˇ
x =1 2 x =3
= 14

d
şx
S-19: Applying dx to both sides of x2 = 1 + 1 f (t) dt gives, by the Fundamental Theorem
of Calculus, 2x = f ( x ).

S-20: By the chain rule

d
2
f 1 (x) = 2 f 1 (x) f 2 (x)
dx

so 21 f 1 ( x )2 is an antiderivative for f 1 ( x ) f 2 ( x ) and, by the fundamental theorem of

calculus,
ż2 ˇ x =2
1 1 2 ˇˇ 1 1 5
1 2
f ( x ) f ( x ) dx = f ( x ) ˇ = f 1 (2)2 ´ f 1 (1)2 =
1 2 x =1 2 2 2

S-21: The car stops when v(t) = 30 ´ 10t = 0, which occurs at time t = 3. The distance
covered up to that time is
ż3 ˇ3
2
v(t) dt = (30t ´ 5t )ˇ = (90 ´ 45) ´ 0 = 45 m.
ˇ
0 0

şx

S-22: Define g( x ) = 0 log 1 + et dt. By the fundamental theorem of calculus,


g1 ( x ) = log 1 + e x . But f ( x ) = g(2x ´ x2 ), so, by the chain rule,

d 2

f 1 ( x ) = g1 (2x ´ x2 ) ¨ (2x ´ x2 ) = (2 ´ 2x ) ¨ log 1 + e2x´x
dx
2 2 2

Observe that e2x´x ą 0 for all x so that 1 + e2x´x ą 1 for all x and log 1 + e2x´x ą 0
for all x. Since 2 ´ 2x is positive for x ă 1 and negative for x ą 1, f 1 ( x ) is also positive for
x ă 1 and negative for x ą 1. That is, f ( x ) is increasing for x ă 1 and decreasing for
x ą 1. So f ( x ) achieves its absolute maximum at x = 1.

102
 ş x2 ´2x dt ş x dt 1 1
S-23: Let f ( x ) = 0 1+ t 4 and g ( x ) = 0 1+t4 . Then g ( x ) = 1+ x4 and, since
f ( x ) = g( x2 ´ 2x ), f 1 ( x ) = (2x ´ 2) g1 ( x2 ´ 2x ) = 2 1+( xx´1
2 ´2x )4 . This is zero for x = 1,

negative for x ă 1 and positive for x ą 1. Thus as x runs from ´8 to 8, f ( x ) decreases

until x reaches 1 and then increases all x ą 1. So the minimum of f ( x ) is achieved for
ş´1 dt
x = 1. At x = 1, x2 ´ 2x = ´1 and f (1) = 0 1+ t4
.

şx ?
S-24: Define G ( x ) = 0 sin( t) dt. By the fundamental theorem of calculus
?
G1 ( x ) = sin( x ). Since F ( x ) = G ( x2 ) we have F1 ( x ) = 2xG1 ( x2 ) = 2x sin x. Thus F
increases as x runs from to 0 to π (since F1 ( x ) ą 0 there) and decreases as x runs from π
to 4 ă 2π (since F1 ( x ) ă 0 there). Thus F achieves its maximum value at x = π.

S-25: The given sum is of the form

n
ÿ π  jπ  n
ÿ
lim sin = lim f ( x˚j )∆x
nÑ8 n n nÑ8
j =1 j =1

jπ
with ∆x = πn , x˚j = n and f ( x ) = sin( x ). Since x0˚ = 0 and xn˚ = π, the right hand side is
the definition (using the right Riemann sum) of

żπ żπ
ˇπ
f ( x ) dx = sin( x ) dx = ´ cos( x )ˇ0 = 2
0 0

S-26: The given sum is of the form

n n
1ÿ 1 ÿ
lim j
= lim f ( x j )∆x
nÑ8 n nÑ8
j =1 1 + n j =1

j
with ∆x = n1 , x j = n and f ( x ) = 1
1+ x . The right hand side is the definition (using the
right Riemann sum) of

ż1 ż1 ˇ1
1
f ( x ) dx = dx = log |1 + x|ˇ = log 2
ˇ
0 0 1+x 0

S-27: (a) Using the product rule, followed by the chain rule, followed by the fundamental

103
theorem of calculus,
ż x 3 +1 ż x 3 +1
t3 d 3
1
f ( x ) = 3x 2
e dt + x 3
et dt
0 dx 0
ż x 3 +1 " ż #
  d y
2 t3 3 2 t3
= 3x e dt + x 3x e dt
0 dy 0
y = x 3 +1
h i
ż x 3 +1
2 t3 3
 2 y3
= 3x e dt + x 3x e
0 y = x 3 +1
ż x 3 +1
3   3 3
= 3x2 et dt + x3 3x2 e( x +1)
0
ż x 3 +1
3 3 +1)3
= 3x2 et dt + 3x5 e( x
0

(b) In general, the equation of the tangent line to the graph of y = f ( x ) at x = a is

y = f ( a) + f 1 ( a) ( x ´ a)
Substituting in the given f ( x ) and
a = ´1 a3 + 1 = (´1)3 + 1 = 0
ż0
3 3
f (´1) = (´1) 3
et dt = 0 f 1 (´1) = 3(´1)2 ˆ 0 + 3(´1)5 e0 = ´3
0

the equation of the tangent line is y = ´3( x + 1)

Solutions to Exercises 1.4 — Jump to TABLE OF CONTENTS

a ?
S-1: We substitute u = sin x, du = cos x dx, cos x = 1 ´ sin2 x = 1 ´ u2 ,
du ?du . When x = 0 we have u = sin 0 = 0 and when x = π2 we have
dx = cos x = 2
1´u
u = sin π2 = 1 so that
ż x=π/2 ż u =1
du
f (sin x ) dx = f (u) ?
x =0 u =0 1 ´ u2

2 2 2
S-2: We write u( x ) = e x and find du = u1 ( x ) dx = 2xe x dx. Note that u(1) = e1 = e
2
when x = 1, and u(0) = e0 = 1 when x = 0. Therefore:
1 x =1
ż1 ż
x2 x2
xe cos(e ) dx = cos(u( x ))u1 ( x ) dx
2
0
ż x =0
1 u=e
= cos(u) du
2 u =1
1 ˇe 1

= sin(u)ˇ1 = sin(e) ´ sin(1) .
2 2

104
S-3: Substituting y = x3 , dy = 3x2
ż2
1 8
ż
2 3 1
x f ( x ) dx = f (y) dy =
1 3 1 3

S-4: Setting u = x3 + 1, we have du = 3x2 dx and so

x2 dx
ż ż
du/3 1 1
101
= 101
=´ 100
+C = ´ 100
+C
( x 3 + 1) u 3 ˆ 100u 300( x3 + 1)

S-5: Setting u = log x, we have du = 1x dx and so

ż e4 ż x = e4 ż u =4
dx 1 1 1
= ¨ dx = du,
e x log x x =e log x x u =1 u

since u = log(e) = 1 when x = e and u = log(e4 ) = 4 when x = e4 . Then, by the

fundamental theorem of calculus,
ż4 ˇ4
1
du = (log |u|)ˇ = log 4 ´ log 1 = log 4.
ˇ
1 u 1

S-6: Setting u = 1 + sin x, we have du = cos x dx and so

ż π/2 ż x=π/2 ż u =2
cos x 1 du
dx = cos x dx =
0 1 + sin x x =0 1 + sin x u =1 u

since u = 1 + sin 0 = 1 when x = 0 and u = 1 + sin(π/2) = 2 when x = π/2. Then, by

the fundamental theorem of calculus,
ż u =2 ˇ2
du
= log |u|ˇ = log 2
ˇ
u =1 u 1

S-7: Setting u = sin x, we have du = cos x dx and so

ż π/2 ż x=π/2 ż u =1
2 2
cos x ¨ (1 + sin x )dx = (1 + sin x ) ¨ cos x dx = (1 + u2 ) du,
0 x =0 u =0

since u = sin 0 = 0 when x = 0 and u = sin(π/2) = 1 when x = π/2. Then, by the

fundamental theorem of calculus,
ż1 ˇ1
(1 + u2 ) du = (u + u3 /3)ˇ = (1 + 1/3) ´ 0 = 4/3.
ˇ
0 0

105
S-8: Substituting t = x2 ´ x, dt = (2x ´ 1) dx and noting that t = 0 when x = 1 and t = 6
when x = 3,
ż3 ż6
x2 ´x
 6
(2x ´ 1)e dx = et dt = et 0 = e6 ´ 1
1 0

S-9: We use the substitution u = 4 ´ x2 , for which du = ´2x dx,:

ż 2
1 4 ´ x2
ż
x ´4
? x dx = ? (´2x ) dx
4 ´ x2 2 4 ´ x2
ż
1 u
= ? du
2 u
ż
1 ?
= u du
2
1 u3/2
= +C
2 3/2
1
= (4 ´ x2 )3/2 + C
3

d x 2 2 2
S-10: The slightly sneaky method: We note that dx e = 2x e x , so that 12 e x is a
x 2
antiderivative for the integrand xe . So
ż2  2
x2 1 x2 1 4 1 4
xe dx = e = e ´ e =0
´2 2 ´2 2 2

2
The really sneaky method: The integrand f ( x ) = xe x is an odd function (meaning that
f (´x ) = ´ f ( x )). So by Theorem 1.2.11 in the CLP 101 notes every integral of the form
şa x2
´a xe dx is zero.

S-11: The given sum is of the form

n
ÿ j  j2 
n
ÿ
lim sin 1 + 2 = lim f ( x˚j )∆x
nÑ8 n2 n nÑ8
j =1 j =1
j
with ∆x = n1 , x˚j = n and f ( x ) = x sin(1 + x2 ). Since x0˚ = 0 and xn˚ = 1, the right hand
side is the definition (using the right Riemann sum) of
ż1 ż1
f ( x ) dx = x sin(1 + x2 ) dx
0 0
1 2
ż
= sin(y) dy with y = 1 + x2 , dy = 2x dx
2 1
1 2
= ´ cos(y) 1
2
1
= [cos 1 ´ cos 2] « 0.478
2

106
S-12: The given sum is of the form
n
ÿ j  j2  n
ÿ
lim 2
cos 2 = lim f ( x˚j )∆x
nÑ8 n n nÑ8
j =1 j =1

j
with ∆x = n1 , x˚j = n and f ( x ) = x cos( x2 ). Since x0˚ = 0 and xn˚ = 1, the right hand side
is the definition (using the right Riemann sum) of
ż1 ż1
f ( x ) dx = x cos( x2 ) dx
0 0
1 1
ż
= cos(y) dy with y = x2 , dy = 2x dx
2 0
1h i1
= sin(y)
2 0
1
= sin 1
2

S-13: The given sum is of the form

n n
c
ÿ j j2 ÿ
lim 1 + 2 = lim f ( x˚j )∆x
nÑ8 n2 n nÑ8
j =1 j =1

j ?
with ∆x = n1 , x˚j = n and f ( x ) = x 1 + x2 . Since x0˚ = 0 and xn˚ = 1, the right hand side
is the definition (using the right Riemann sum) of
ż1 ż1 a
f ( x ) dx = x 1 + x2 dx
0 0
1 2?
ż
= y dy with y = 1 + x2 , dy = 2x dx
2 1
1 h y3/2 i
2
=
2 3/2 1
1 ?
= [2 2 ´ 1] « 0.609
3

Solutions to Exercises 1.5 — Jump to TABLE OF CONTENTS

S-1: The curves intersect when y = x and y = 3 3 2

? x ?´ x so that x = x ´ x or x ( x ´ 2) = 0.
For x ě 0, the curves intersect at (0, 0) and ( 2, 2). Using vertical strips as in the sketch

107
 y y=x
√ √
( 2, 2)

y = x3 − x

the top and bottom boundaries of the specified region are y = T ( x ) = x and
y = B( x ) = x3 ´ x, respectively. So
ż ?2 ż ?2
   
Area = T ( x ) ´ B( x ) dx = x ´ ( x3 ´ x ) dx
0 0

S-2: The curves intersect at (´8, 4) and (3, ´ 32 ). Using horizontal strips as in the sketch
y
(−8, 4)

y = −x/2 or x = −2y y 2 = 6 − 45 x or x = 54 (6 − y 2 )
x
(3, −3/2)

we have
ż4 h4 i
Area = (6 ´ y2 ) + 2y dy
´3/2 5

S-3: The curves intersect at (0, 0) and (4a, 4a). Using vertical strips as in the sketch
y
y 2 = 4ax
(4a, 4a)

x2 = 4ay
x

we have
ż 4a h?
x2 i
Area = 4ax ´ dx
0 4a

108
S-4: The curves intersect when x = 4y2 and 0 = 4y2 + 12y + 5 = (2y + 5)(2y + 1). So the
curves intersect at (1, ´ 21 ) and (25, ´ 52 ). Using vertical strips as in the sketch
2
√
y x = 4y or y = ± x/2
1
x + 12y + 5 = 0 or y = − 12 (x + 5)
(1, − 1 ) x
2

(25, − 52 )

we have
ż 25 h
1 1? i
Area = ´ ( x + 5) + x dx
1 12 2

1
S-5: We are asked for the area between the top curve y = T ( x ) = (2x´4 )2
and the bottom
curve y = B( x ) = 0 with x running from a = 0 to b = 1. So, by (1.5.1) in the CLP 101
notes, the area is
żb
 
ż1
dx h 1 1 i1 h 1 1 i 1
T ( x ) ´ B( x ) dx = 2
= ´ ¨ = ´ =
a 0 (2x ´ 4) 2 2x ´ 4 0 4 8 8

S-6: The two curves y = f ( x ) = x and y = g( x ) = 3x ´ x2 , intersect when

f ( x ) = g( x ) ðñ x = 3x ´ x2 ðñ 2x ´ x2 = 0 ðñ x (2 ´ x ) = 0
ðñ x = 0, 2

Furthermore g( x ) ´ f ( x ) = 2x ´ x2 = x (2 ´ x ) is positive for all 0 ď x ď 2. That is, the

curve y = 3x ´ x2 lies above the line y = x for all 0 ď x ď 2. We therefore evaluate the
integral:
ż2 ż2  2  
 2
 2 2 x3 8 4
(3x ´ x ) ´ x dx = [2x ´ x ] dx = x ´ = 4´ ´0 =
0 0 3 0 3 3

S-7: The two curves cross when x = 0 (y = 1 for both curves) and x = 1 (y = 2 for both
x
curves). Since 2x = (elog 2 ) = e x log 2 , the area is
ż1
 ?  h 2 3/2 1 x i1
x log 2
Area = (1 + x ) ´ e dx = x + x ´ 2
0 3 log 2 0
2 1 5 1
= 1+ ´ [2 ´ 1] = ´
3 log 2 3 log 2

109
S-8: Here is a sketch of the specified region.
y √
y= 2 cos(πx/4)

(−1, 1) (1, 1)

y = |x| y = |x|

It is symmetric about the y–axis.

? So we will compute the area of the part with? x ě 0 and
multiply by 2. The curves y = 2 cos(πx/4) and y = x intersect when x = 2 cos(πx/4)
or cos(πx/4) = ?x2 , which is the case2 when x = 1. So, using vertical strips as in the
figure above, the area (including the multiplication by 2) is

ż1 ?  1  
  ? 4 x2 4 1
2 2 cos(πx/4) ´ x dx = 2 2 sin(πx/4) ´ = 2 ´
0 π 2 0 π 2

?
S-9: For our computation, we will need an antiderivative of x2 x3 + 1, which can be
found using the substitution u = x3 + 1, du = 3x2 dx:

1 u3/2
ż ż ż
2
a ? 1 1 2
x x3 + 1 dx = u ¨ du = u1/2 du = + C = ( x3 + 1)3/2 + C.
3 3 3 3/2 9

The two functions f ( x ) and g( x ) are clearly equal at x = 0. If x ‰ 0, then the functions
are equal when
a
3x2 = x2 x3 + 1
a
3 = x3 + 1
9 = x3 + 1
8 = x3
2 = x.

The function g( x ) = 3x2 is the larger of the two on the interval [0, 2], as can ?
be seen by
plugging in x = 1, say, or by observing that when x is very small f ( x ) = x2 x3 + 1 « x2
and g( x ) = 3x2 .

2 The solution x = 1 was found by guessing. To guess a solution to cos(πx/4) = ?x just ask yourself
? 2
what simple angle has a cosine that involves 2. This guessing strategy is essentially useless in the real
world, but works great on problem sets and exams.

110
 y
(2, 12)

y = 3x2
√
y = x2 x3 + 1

The area in question is therefore

ż2  ˇ2
2 2
a
2 3 3
3/2 ˇ
ˇ
3x ´ x x3 + 1 dx = x ´ ( x + 1)
0 9 ˇ
 0  
3 2 3 3/2 3 2 3 3/2
= 2 ´ (2 + 1) ´ 0 ´ (0 + 1)
9 9
   
2 20
= 8´6 ´ 0´ = .
9 9

S-10: A point ( x, y) on the curve x = y2 + y = y(y + 1) has x = 0 for y = ´1, 0, has x ă 0

for ´1 ă y ă 0 (the factors y and y + 1 have opposite signs) and has x ą 0 for y ă ´1 and
y ą 0 (the factors y and y + 1 are either both positive or both negative). This leads to the
figure below. So, using horizontal slices, y

(0, 0) x
ż0  0
2


y3 y2 1 1 1
area = 0 ´ (y + y) dy = ´ + =´ + =
´1 3 2 ´1 3 2 6 (0, −1)
x = y + y2

S-11: We will compute the area by using thin vertical strips as in the sketch
y
8
y = 4 + 2π − 2x

6 y = 4 + π sin(x)
4

x
π π 3π 2π
2 2

The line y = 4 + 2π ´ 2x intersects the curve y = 4 + π sin x when

2 π 3π
4 + 2π ´ 2x = 4 + π sin x ðñ sin x = 2 ´ x ðñ x = , π,
π 2 2

111
These solutions were guessed by looking at the sketch above, but then verified by
substituting them back into the equation. From the sketch we see that
• when π2 ď x ď π, the top of the strip is at y = 4 + π sin x and the bottom ofthe strip
is at y = 4 + 2π ´ 2x. So the strip has height (4 + π sin x ) ´ (4 + 2π ´ 2x ) and
 
width dx and hence area (4 + π sin x ) ´ (4 + 2π ´ 2x ) dx.
• when π ď x ď 3π 2 , the top of the strip is at y = 4 +
 2π ´ 2x and the bottom of the

strip is at y = 4 + π sin x. So the strip has height (4 + 2π ´ 2x ) ´ (4 + π sin x ) and
 
width dx and hence area (4 + 2π ´ 2x ) ´ (4 + π sin x ) dx.
So the total
ż 3π/2
   
żπ
Area = (4 + π sin x ) ´ (4 + 2π ´ 2x ) dx + (4 + 2π ´ 2x ) ´ (4 + π sin x ) dx
π/2 π
ż 3π/2
   
żπ
= π sin x ´ 2π + 2x dx + 2π ´ 2x ´ π sin x dx
π/2 π
 π  3π/2
= ´ π cos x ´ 2πx + x2 π/2
+ 2πx ´ x2 + π cos x π
 3   5 
= π ´ π2 + π2 + π2 ´ π2 + π
4 4
h 1 2i
=2 π´ π
4

S-12: First, here is a sketch of the region. We are not asked for it, but it is still a crucial for
understanding the question.
y = x2
y

y = x+2

x
2 3
The two curves y = x + 2 and y = x2 cross at x = 2, y = 4. The area of the part between
them with 0 ď x ď 2 is
ż2
  h 1 3 i2 8 10
2 1 2
x + 2 ´ x dx = 2 x + 2x ´ x = 2+4´ =
0 3 0 3 3

The area of the part between them with 2 ď x ď 3 is

ż3
  h1 i3 9 8 11
2 3 1 2
x ´ ( x + 2) dx = x ´ 2x ´ 2x = 9´ ´6´ +2+4 =
2 3 2 2 3 6

112
 10
The total area is 3 + 11
6 =
31
6 = 5.16̇.

?
S-13: The curve y = f ( x ) = x 25 ´ x2 lies above the line y = g( x ) = 3x at all values of x
for which f ( x ) ě g( x ), i.e. f ( x ) ´ g( x ) ě 0. Now
a a

f ( x ) ´ g( x ) = x 25 ´ x2 ´ 3x = x 25 ´ x2 ´ 3

The first factor is positive for all x ě 0. The second factor is positive whenever
a a
25 ´ x2 ´3 ě 0 ðñ 25 ´ x2 ě 3 ðñ 25 ´ x2 ě 9 ðñ x2 ď 16
ðñ |x| ď 4
?
So y = f ( x ) = x 25 ´ x2 lies above y = g( x ) = 3x for all 0 ď x ď 4. The area we need to
calculate is therefore:
ż4 a ż4 a ż4
A = [ x 25 ´ x2 ´ 3x ] dx = x 25 ´ x2 dx ´ 3x dx = A1 ´ A2 .
0 0 0

To evaluate A1 , we use the substitution u( x ) = 25 ´ x2 , for which

du = u1 ( x ) dx = ´2x dx; and u(4) = 25 ´ 42 = 9 when x = 4, while u(0) = 25 ´ 02 = 25
when x = 0. Therefore
ż x =4 a ˇ9
1 u =9 ?
ż
2
1 3/2 ˇˇ 125 ´ 27 98
A1 = x 25 ´ x dx = ´ u du = ´ u ˇ = =
x =0 2 u=25 3 25 3 3

For A2 we use the antiderivative directly:

ż4 ˇ4
3x2 ˇˇ
A2 = 3x dx = = 24
0 2 ˇ0

Therefore the total area is:

98 26
A= ´ 24 =
3 3

Solutions to Exercises 1.6 — Jump to TABLE OF CONTENTS

S-1: (a) When the strip shown in the figure

y √ 2
y= x ex

3 x

113
 ? 2
is rotated about the x–axis, it forms a thin disk of radius xe x and thickness dx and
2 2
hence of cross sectional area πxe2x and volume πxe2x dx So the volume of the solid is
ż3
2
π xe2x dx
0

(b) The curves intersect at (´1, 1) and (2, 4).

y y = x + 2 or x = y − 2

(2, 4)

√ x=3
y = x2 or x = ± y
y=1
(−1, 1)
x

We’ll use horizontal washers as in Example 1.6.5 of the CLP 101 notes.

• We use thin horizontal strips of width dy as in the figure above.

• When we rotate about the line x = 3, each strip sweeps out a thin washer
?
– whose inner radius is rin = 3 ´ y, and

– whose outer radius is rout = 3 ´ (y ´ 2) = 5 ´ y when y ě 1 (see the red strip

in the figure on the right above), and whose outer radius is
? ?
rout = 3 ´ (´ y) = 3 + y when y ď 1 (see the blue strip in the figure on the
right above) and

– whose thickness is dy and hence




2 ´ r2 )dy = π 5 ´ y 2 ´ 3 ´ ?y 2 dy when y ě 1 and
– whose volume is π (rout in



2 ´ r2 )dy = π 3 + ?y 2 ´ 3 ´ ?y 2 dy when y ď 1
whose volume is π (rout in
and

• As our bottommost strip is at y = 0 and our topmost strip is at y = 4, the total

volume is
ż1 ż4
 ?
2 ?
2  
2 ?
2 
π 3 + y ´ 3 ´ y dy + π 5 ´ y ´ 3 ´ y dy
0 1

S-2: (a) The curves intersect at (1, 0) and (´1, 0). When the strip shown in the figure

114
 y

y = 4 − 4x2

(−1, 0) (1, 0)
x
y = −1
y = 1 − x2

is rotated about the line y = ´1, it forms a thin washer of

• inner radius (1 ´ x2 ) ´ (´1) = 2 ´ x2 ,
• outer radius (4 ´ 4x2 ) ´ (´1) = 5 ´ 4x2 and
• thickness dx and hence of
 2 2
• cross sectional area π (5 ´ 4x2 ) ´ (2 ´ x2 ) and
 2 2
• volume π (5 ´ 4x2 ) ´ (2 ´ x2 ) dx.
So the volume of the solid is
ż1
 2 2
π (5 ´ 4x2 ) ´ (2 ´ x2 ) dx
´1

(b) The curve y = x2 ´ 1 intersects y = 0 at (1, 0) and (´1, 0).

y

(−1, 0) (1, 0)
x
√
y = x2 − 1 or x = ± y + 1
(0, −1)
x=5
We’ll use horizontal washers.
• We use thin horizontal strips of width dy as in the figure above.
• When we rotate about the line x = 5, each strip sweeps out a thin washer
a
– whose inner radius is rin = 5 ´ y + 1, and
a a
– whose outer radius is rout = 5 ´ (´ y + 1) = 5 + y + 1 and
– whose thickness is dy and hence

2 ´ r2 ) dy = π 5 +
a
2 a
2 
– whose volume is π (rout in y+1 ´ 5´ y+1 dy

115
 • As our topmost strip is at y = 0 and our bottommost strip is at y = ´1 (when
x = 0), the total volume is
ż0
 a
2 a
2 
π 5+ y+1 ´ 5´ y+1 dy
´1

S-3: (a) The curves intersect at (´2, 4) and (2, 4). When the strip shown in the figure

y = x2 y

(−2, 4) (2, 4)
x
y = −1
y = 8 − x2

is rotated about the line y = ´1, it forms a thin washer (punctured disc) of
• inner radius x2 + 1,
• outer radius 9 ´ x2 and
• thickness dx and hence of
 2 2
• cross sectional area π (9 ´ x2 ) ´ ( x2 + 1) and
 2 2
• volume π (9 ´ x2 ) ´ ( x2 + 1) dx.
So the volume of the solid is
ż2
 2 2
π (9 ´ x 2 ) ´ ( x 2 + 1) dx
´2

S-4: (a) The curves intersect at points ( x, y) which satisfy both y2 = 4ax and x2 = 4ay.
x2
Substituting y = 4a (from the second equation) into the first equation gives

x4
2 2
= 4ax ðñ x4 = 43 a3 x ðñ x ( x3 ´ 43 a3 ) = 0
4 a

This has two solutions: x = 0 and x = 4a. The corresponding values of y are y = 0 and
y = 4a. So the curves intesect at (0, 0) and (4a, 4a). The strip shown in the figure
y (4a, 4a) y 2 = 4ax

x2 = 4ay
x

116
 x2
?
runs from y = B( x ) = (gotten by solving x2 = 4ay for y) to y = T ( x ) = 4ax (gotten
4a ? x2
by solving y2 = 4ax for y) and hence has height T ( x ) ´ B( x ) = 4ax ´ 4a and width dx.
So the desired ż 4a h?
x2 i
Area = 4ax ´ dx
0 4a

(b) The curves intersect at points ( x, y) which satisfy both y = 1 ´ x2 and y = 4(1 ´ x2 ).
That is, where
1 ´ x2 = 4(1 ´ x2 ) ðñ 3(1 ´ x2 ) = 0 ðñ x = ˘1
Thus the curves intersect at (1, 0) and (´1, 0). When the strip shown in the figure
y

y = 4 − 4x2

(−1, 0) (1, 0)
x
y = −1
y = 1 − x2

is rotated about the line y = ´1, it forms a thin washer (punctured disc) of
• inner radius (1 ´ x2 ) ´ (´1) = 2 ´ x2 ,
• outer radius (4 ´ 4x2 ) ´ (´1) = 5 ´ 4x2 and
• thickness dx and hence of
 2 2
• cross sectional area π (5 ´ 4x2 ) ´ (2 ´ x2 ) and
 2 2
• volume π 5 ´ 4x2 ) ´ (2 ´ x2 ) dx.
So the volume of the solid is
ż1
 2 2
π (5 ´ 4x2 ) ´ (2 ´ x2 ) dx
´1

a
(c) Note that solving y = x2 ´ 1 for x gives x = ˘ y + 1. When the strip shown in the
figure
x=5
y
x
y = x2 − 1
(0, −1)

is rotated about the line x = 5, it forms a thin washer (punctured disc) of

a
• inner radius 5 ´ y + 1,
a
• outer radius 5 + y + 1 and

117
 • thickness dy and hence of
 a 2 a 2 a
• cross sectional area π (5 + y + 1) ´ (5 ´ y + 1) = 20π y + 1 and
 a 2 a 2 a
• volume π (5 + y + 1) ´ (5 ´ y + 1) dy = 20π y + 1 dy.
So the volume of the solid is
ż0 ż0
 a 2 a 2 a
π (5 + y + 1) ´ (5 ´ y + 1) dy = 20π y + 1 dy
´1 ´1

? 2
S-5: Let f ( x ) = 1 + xe x . On the vertical slice a distance x from the y-axis, sketched in
the figure below, y runs from 1 to f ( x ). Upon rotation about the line y = 1, this thin slice
sweeps out a cylinder of thickess dx and radius f ( x ) ´ 1 and hence of volume
π [ f ( x ) ´ 1]2 dx. The full volume generated (for any fixed a ą 0) is
ża ża
2 2
π [ f ( x ) ´ 1] dx = π xe2x dx. √ 2
0 0
y y = 1 + xex
Using the substitution u = 2x2 , so that du = 4x dx:

π  2a2 
ż 2a2
du π ˇˇ2a2
Volume = π eu = eu ˇ = e ´1
0 4 4 0 4 y=1

x=a x
S-6: For a fixed value of x, if we rotate about the x-axis, we form a washer of inner radius
B( x ) and outer radius T ( x ) and hence of area π [ T ( x )2 ´ B( x )2 ]. We integrate this
function from x = 0 to x = 3 to find the total volume V:
ż3
V= π [ T ( x )2 ´ B( x )2 ] dx
0
ż3
? ?
= π ( xe3x )2 ´ ( x (1 + 2x ))2 dx
0
ż3


=π xe6x ´ ( x + 4x2 + 4x3 ) dx
0
ż3 h x2 4x3 i3
6x 4
=π xe dx ´ π + +x
0 2 3 0
ż3 h 32 4 ¨ 33 i
=π xe6x dx ´ π + + 34
0 2 3
For the first integral, we use integration by parts with u( x ) = x, dv = e6x dx, so that
du = dx and v( x ) = 61 e6x :
ż3 ˇ3 ż 3
6x xe6x ˇˇ 1 6x
xe dx = ´ e dx
0 6 0 ˇ
0 6
ˇ3  18 
3e18 1 6x ˇˇ e18 e 1
= ´0´ e ˇ = ´ ´ .
6 36 0 2 36 36

118
Therefore, the total volume is
      
e18 e18 1 32 4 ¨ 33 4 17e18 ´ 4373
V=π ´ ´ ´π + +3 = π .
2 36 36 2 3 36

S-7:
10
The curves y = 1/x and 3x + 3y = 10, i.e. y = 3 ´ x intersect when

1 10
= ´ x ðñ 3 = 10x ´ 3x2 ðñ 3x2 ´ 10x + 3 = 0
x 3
ðñ (3x ´ 1)( x ´ 3) = 0
1
ðñ x = 3 ,
3

y
(1/3, 3)

y = T (x)

(3, 1/3)
y = B(x)
y=0

When the region is rotated about the x–axis, the vertical strip in the figure above sweeps
out a washer with thickness dx, outer radius T ( x ) = 10 1
3 ´ x and inner radius B ( x ) = x .
This washer has volume

 100 20 1
2 2 2
π T ( x ) ´ B( x ) dx = π ´ x + x ´ 2 dx
9 3 x

Hence the volume of the solid is

ż3 
100 20 1 h 100x 10 1 1 i3
π ´ x + x2 ´ 2 dx = π ´ x2 + x3 +
1/3 9 3 x 9 3 3 x 1/3
h 38 514 i 512
=π ´ 4 =π
3 3 81

?
? the bottom of the circle have equations y = T ( x ) = 2 +
S-8: (a) The top and 1 ´ x2 and
y = B( x ) = 2 ´ 1 ´ x2 , respectively.

119
 y = T (x)

y = B(x)

y=0
x = −1 x=1

When R is rotated about the x–axis, the vertical strip of R in the figure above sweeps out
a washer with thickness dx, outer radius T ( x ) and inner radius B( x ). This washer has
volume

2 2



a
π T ( x ) ´ B( x ) dx = π T ( x ) + B( x ) T ( x ) ´ B( x ) dx = π ˆ 4 ˆ 2 1 ´ x2 dx

Hence the volume of the solid is

ż1 a
8π 1 ´ x2 dx
´1

?
(b) Since y = 1 ´ x2 is equivalent to x2 + y2 = 1, y ě 0, the integral is 8π times the area
of the upper half of the circle x2 + y2 = 1 and hence is 8π ˆ 21 π12 = 4π 2 .

S-9: (a) The two curves intersect when x obeys 8x = x2 + 15 or

x2 ´
?8x + 15 = (? x ´ 5)( x ´ 3) = 0. The points of intersection, in the first quadrant, are
(3, 24) and (5, 40). The region R is the region between the blue and red curves, with
3 ď x ď 5, in the figures below.

y y 2 = x2 + 15
y 2 = 8x
√ √
√ (5, 40) (5, 40)
(3, 24) R
√ R
(3, 24)

? x coordinate between
(b) The part of the solid with ? x and x + dx is a “washer” shaped
region with inner radius 2
x +?15, outer radius 8x and thickness dx. The surface area
?
of the washer is π ( 8x )2 ´ π ( x2 + 15)2 = π (8x ´ x2 ´ 15) and its volume is

120
π (8x ´ x2 ´ 15) dx. The total volume is
ż5 h 1 3 i5 h 125 i
2 2
π (8x ´ x ´ 15) dx = π 4x ´ x ´ 15x = π 100 ´ ´ 75 ´ 36 + 9 + 45
3 3 3 3
4
= π « 4.19
3

S-10: (a) The region R is sketched in the figure on the left below.
y y
y = log x x = ey
x=2 x=2
x x

(b) We’ll use horizontal washers as in Example 1.6.5 of the CLP 101 notes.
• We cut R into thin horizontal strips of width dy as in the figure on the right above.
• When we rotate R about the y–axis, i.e. about the line x = 0, each strip sweeps out a
thin washer
– whose inner radius is rin = ey and outer radius is rout = 2, and
– whose thickness is dy and hence


2 ´ r 2 )dy = π 4 ´ e2y dy.
– whose volume π (rout in
• As our bottommost strip is at y = 0 and our topmost strip is at y = log 2 (since at
the top x = 2 and x = ey ), the total
ż log 2

 log 2 h 1i
Volume = π 4 ´ e2y dy = π 4y ´ e2y /2 0 = π 4 log 2 ´ 2 +
0 2
h 3 i
= π 4 log 2 ´ « 3.998
2

S-11: Here is a sketch of the curves y = cos( 2x ) and y = x2 ´ π 2 .

y

(−π, 0) (π, 0)

y = cos( x2 )

y = x2 − π 2

y = −π 2

The curves meet at x = ˘π where both cos( 2x ) and x2 ´ π 2 take the value zero. We’ll use
vertical washers as specified in the question.

121
 • We cut the specified region into thin vertical strips of width dx as in the figure
above.
• When we rotate about the line y = ´π 2 , each strip sweeps out a thin washer
– whose inner radius is rin = ( x2 ´ π 2 ) ´ (´π 2 ) = x2 and outer radius is
rout = cos( 2x ) ´ (´π 2 ) = cos( 2x ) + π 2 , and
– whose thickness is dx and hence


2 ´ r 2 )dx = π (cos( x ) + π 2 )2 ´ ( x2 )2 dx.
– whose volume π (rout in 2

• As our leftmost strip is at x = ´π and our rightmost strip is at x = π,

the total volume is
żπ  
π cos2 ( 2x ) + 2π 2 cos( 2x ) + π 4 ´ x4 dx
´π
żπ  
1 + cos( x ) 2 x 4 4
=π + 2π cos( 2 ) + π ´ x dx
´π 2
żπ 
1 + cos( x ) 2 x 4 4
= 2π + 2π cos( 2 ) + π ´ x dx
0 2
 
1 1 2 x 4 1 5 π
= 2π x + sin( x ) + 4π sin( 2 ) + π x ´ x
2 2 5 0
 5 
π π
= 2π + 0 + 4π 2 + π 5 ´
2 5
8π 6
= π 2 + 8π 3 +
5
In the middle line, we used the fact that the integrand is an even function and the
interval of integration [´π, π ] is symmetric, but one can also compute directly.

S-12: As in Example 1.6.6 of the CLP 101 notes, we slice V into thin horizontal “square
pancakes”.
• We are told that the pancake at height x is a square of side 1+2 x and so

2
• has cross-sectional area 1+2 x and thickness dx and hence

2
• has volume 1+2 x dx.
Hence the volume of V is
ż2h
2 i2 h1 i
ż3 ˇ3
4 u´1 ˇˇ 8
dx = 2
du = 4 = ´4 ´ 1 =
0 1+x 1 u ´1 1 ˇ 3 3

We made the change of variables u = 1 + x, du = dx.

S-13: Here is a sketch of the base region.

122
 y
y = x2

y = 8 − x2
x

Consider the thin vertical cross–section resting on the heavy red line in the figure above.
It has thickness dx. Its face is a square whose side runs from y = x2 to y = 8 ´ x2 , a
2 2
distance of 8 ´ 2x2 . So the face has area (8 ´ 2x2 ) and the slice has volume (8 ´ 2x2 ) dx.
The two curves cross when x2 = 8 ´ x2 , i.e. when x2 = 4 or x = ˘2. So x runs from ´2 to
2 and the total volume is
ż2 ż2 ż2
2 2 2 2
 
(8 ´ 2x ) dx = 2 4(4 ´ x ) dx = 8 16 ´ 8x2 + x4 dx
´2 0 0
h 8 1 i 256 ˆ 8
= 8 16 ˆ 2 ´ 23 + 25 = = 136.53̇
3 5 15

S-14: Slice the frustrum into horizontal discs. When the disc is a distance t from the top of
the frustrum it has radius 2 + 2t/h. Note that as t runs from 0 (the top of the frustrum) to
t = h (the bottom of the frustrum) the radius 2 + 2t/h increases linearly from 2 to 4.

t
h

2
Thus the disk has volume π 2 + 2t/h dt. The total volume of the frustrum is
żh żh ˇh

2
2 (1 + t/h)3 ˇˇ 4 28
π 2 + 2t/h dt = 4π 1 + t/h dt = 4π = πh ˆ 7 = πh
0 0 3/h ˇ
0 3 3

S-15: (a) The curve y = 4 ´ ( x ´ 1)2 is an “upside down parabola” and line y = x + 1 has
slope 1. They intersect at points ( x, y) which satisfy both y = x + 1 and y = 4 ´ ( x ´ 1)2 .
That is, when x obeys

x + 1 = 4 ´ ( x ´ 1)2 ðñ x + 1 = 4 ´ x2 + 2x ´ 1 ðñ x2 ´ x ´ 2 = 0
ðñ ( x ´ 2)( x + 1) = 0 ðñ x = ´1, 2

Thus the intersection points are (´1, 0) and (2, 3). Here is a sketch of R

123
 y
y = x+1
(2, 3)

y = 4 − (x − 1)2

(−1, 0)
x

The red strip in the sketch above runs from y = x + 1 to y = 4 ´ ( x ´ 1)2 and so has area
[4 ´ ( x ´ 1)2 ´ ( x + 1)] dx = [2 + x ´ x2 ] dx. All together R has
ż2
 
Area = 2 + x ´ x2 dx
´1
 2
x2 x3
= 2x + ´
2 3 ´1
3 9 9
= 6+ ´ =
2 3 2

(b) We’ll use vertical washers as in Example 1.6.3 of the CLP 101 notes.
y
y=5

y = x+1
(2, 3)

y = 4 − (x − 1)2

(−1, 0)
x

• We cut R into thin vertical strips of width dx like the red strip in the figure above.
• When we rotate R about the horizontal line y = 5, each strip sweeps out a thin
washer
– whose inner radius is rin = 5 ´ [4 ´ ( x ´ 1)2 ] = 1 + ( x ´ 1)2 , and
– whose outer radius is rout = 5 ´ [ x + 1] = 4 ´ x and
– whose thickness is dx and hence
 2  


2 dx = π 4 ´ x 2 ´ 1 + ( x ´ 1)2 2 dx
– whose volume is π rout ´ rin
• As our leftmost strip is at x = ´1 and our rightmost strip is at x = 2, the total
ż2

2
2 
Volume = π 4´x ´ 1 + ( x ´ 1)2 dx
´1

124
S-16: (a) The curves ( x ´ 1)2 + y2 = 1 and x2 + (y ´ 1)2 = 1 are circles of radius 1
centered on (1, 0) and (0, 1) respectively. Both circles pass through (0, 0) and (1, 1). They
are sketched below.
y

y=x
(1, 1)
x2 + (y − 1)2 = 1

(x − 1)2 + y 2 = 1

The region R is symmetric about the line y = x, so the area of R is twice the area of the
part of R
ato the left of the line y = x. The red strip in the sketch above runs from
x = 1 ´ 1 ´ y2 to x = y

ż1
 b

Area = 2 1 ´ y2 dy
y´ 1´
0
!h y2 i1 ż 1 b )
=2 ´y + 1 ´ y2 dy
2 0 0
π
= ´1
2
ş1 a
Here the integral 0 1 ´ y2 dy was evaluated simply as the area of one quarter of a
cicular disk of radius 1. It can also be evaluated by substituting y = sin θ.

(b) We’ll use horizontal washers as in Example 1.6.5 of the CLP 101 notes.

• We cut R into thin horizontal strips of width dy like the blue strip in the figure
above.

• When we rotate R about the y–axis, each strip sweeps out a thin washer
a
– whose inner radius is rin = 1 ´ 1 ´ y2 , and
a
– whose outer radius is rout = 1 ´ (y ´ 1)2 and

– whose thickness is dy and hence

 a
2 a
2 
– whose volume is π 1 ´ ( y ´ 1)2 ´ 1 ´ 1 ´ y2 dy
a 
2
= 2π 1 ´ y + y ´ 1 dy

125
 • As our bottommost strip is at y = 0 and our topmost strip is at y = 1, the total

ż1 b
  hπ 1 i
Volume = 2π 1 ´ y2 + y ´ 1 dy = 2π + ´1
0 4 2
π2
= ´ π « 1.793
2
ş1 a
Here, we again used that 0 1 ´ y2 dy is the area of a quarter circle of radius one.

S-17: (a) Let V1 be the solid obtained by revolving R about the x–axis. The portion of V1
with x–coordinate between x and x + dx is obtained by rotating the red vertical ? strip in
2
? portion is a disk of radius c 1 + x
the figure on the left below about the x–axis. That
2 2 2 2
and thickness dx. The volume of this disk is π (c 1 + x ) dx = πc (1 + x ) dx. So the
total volume of V1 is

ż1 h x 3 i1 4
V1 = πc2 (1 + x2 ) dx = πc2 x + = πc2
0 3 0 3

y √ y q
y2
y =c 1+ x2 x= c2
−1

c c

x=1 x=1
x x

(b) We’ll use horizontal washers as in Example 1.6.5 of the CLP 101 notes.

• We cut R into thin horizontal strips of width dy as in the figure on the right above.

• When we rotate R about the y–axis, i.e. about the line x = 0, each strip sweeps out
a thin washer

– whose outer radius is rout = 1, and

b
y2 ?
– whose inner radius is rin = c2 ´ 1 when y ě c 1 + 02 = c (see the red strip
in the figure on the right above), and whose inner radius is rin = 0 when y ď c
(see the blue strip in the figure on the right above) and

– whose thickness is dy and hence

2

2 ´ r2 )dy = π 2 ´ y dy when y ě c and whose
– whose volume is π (rout in c2
2 ´ r2 )dy = π dy when y ď c and
volume is π (rout in
?
• As our bottommost strip is at y = 0 and our topmost strip is at y = 2 c (since at

126
 ?
the top x = 1 and y = c 1 + x2 ), the total
ż ?2 c 
y2 
żc
V2 = π 2 ´ 2 dy + π dy
c c 0
?
h y3 i
2c
= π 2y ´ 2 + πc
3c c
h 4?2 5 i
= πc ´ + πc
3 3
π c ?  2π c  3/2 
= 4 2´2 = 2 ´1
3 3

(c) We have V1 = V2 if and only if

4 2 2
πc = πc[23/2 ´ 1] ðñ c = 0 or c = 21 [23/2 ´ 1]
3 3

S-18: (a) The region R is

y y= √ 10
25−x2

3 4 x
(b) Let V1 be the solid obtained by revolving R about the x–axis. The portion of V1 with
x–coordinate between x and x + dx is obtained by rotating the red vertical strip in the
figure on the left below about the x–axis. That portion is a disk of radius ? 10 2 and
25´x
10

2
thickness dx. The volume of this disk is π ? 2
dx. So the total volume of V1 is
25´x
ż4  2 ż4 ż4
10 1 1
π ? dx = 100π 2
dx = 100π dx
3 25 ´ x2 3 25 ´ x 3 (5 ´ x )(5 + x )
ż4h
1 1 i h i4
= 10π + dx = 10π ´ log(5 ´ x ) + log(5 + x )
3 5´x 5+x 3
h i 9 3
= 10π ´ log 1 + log 9 + log 2 ´ log 8 = 10π log = 20π log
4 2
y y= √ 10 y q
25−x2 100
x= 25 − y2
5
2

3 4 x 3 4 x

127
(c) We’ll use horizontal washers as in Example 1.6.5 of the CLP 101 notes.
• We cut R into thin horizontal strips of width dy as in the figure on the right above.
• When we rotate R about the y–axis, each strip sweeps out a thin washer
– whose outer radius is rout = 4, and
b
– whose inner radius is rin = 25 ´ 100y2
when y ě ? 10 = 10
4 = 5
2 (see the red
25´32
strip in the figure on the right above), and whose inner radius is rin = 3 when
y ď 52 (see the blue strip in the figure on the right above) and
– whose thickness is dy and hence
2 ´ r2 )dy = π 100

5
– whose volume is π (rout in y2
´ 9 dy when y ě 2 and whose
2 2 )dy 5
volume is π (rout ´ rin = 7π dy when y ď 2 and
10
• As our bottommost strip is at y = 0 and our topmost strip is at y = 3 (since at the
top x = 4 and y = ? 10 2 = ? 10 2 = 10 3 ), the volume is
25´x 25´4

ż 10/3   ż 5/2
100
π ´ 9 dy + 7π dy
5/2 y2 0
h 100 i10/3 35
=π ´ ´ 9y + π
y 5/2 2
h 45 i 35
= π ´ 30 + 40 ´ 30 + + π
2 2
= 20π

S-19: We will compute the volume by rotating thin vertical strips as in the sketch
y
8
y = 4 + 2π − 2x

6 y = 4 + π sin(x)
4

x
π π 3π 2π
2 2

about the line y = ´1 to generate thin washers. The line y = 4 + 2π ´ 2x intersects the
curve y = 4 + π sin x when

2 π 3π
4 + 2π ´ 2x = 4 + π sin x ðñ sin x = 2 ´ x ðñ x = , π,
π 2 2
These solutions were guessed by looking at the sketch above, but then verified by
substituting them back into the equation. From the sketch we see that

128
 • when π2 ď x ď π, the top of the strip is at y = 4 + π sin x and the bottom of the strip
is at y = 4 + 2π ´ 2x. So when the strip is rotated we get a thin washer with outer
and inner radii R( x ) = 1 + 4 + π sin x = 5 + π sin x and
r ( x ) = 1 + 4 + 2π ´ 2x = 5 + 2π ´ 2x, respectively.
• when π ď x ď 3π 2 , the top of the strip is at y = 4 + 2π ´ 2x and the bottom of the
strip is at y = 4 + π sin x. So when the strip is rotated we get a thin washer with
R( x ) = 1 + 4 + 2π ´ 2x = 5 + 2π ´ 2x and r ( x ) = 1 + 4 + π sin x = 5 + π sin x,
respectively.
So the total
ż 3π/2
   
żπ
2 2
Volume = π R( x ) ´ r ( x ) dx + π R( x )2 ´ r ( x )2 dx
π/2 π
 
żπ
= π (5 + π sin x )2 ´ (5 + 2π ´ 2x )2 dx
π/2
ż 3π/2
 
+ π (5 + 2π ´ 2x )2 ´ (5 + π sin x )2 dx
π

Solutions to Exercises 1.7 — Jump to TABLE OF CONTENTS

dx x2
S-1: By integration by parts with u = log x and dv = x dx, so that du = x and v = 2,

x2 log x
ż 2
x2 log x 1
ż ż
x dx
x log x dx = ´ = ´ x dx
2 2 x 2 2
x2 log x x2
= ´ +C
2 4

dx
S-2: By integration by parts with u = log x and dv = x´7 dx, so that du = x and
x´6
v=´ 6 ,

x´6
ż ż ´6 ż
log x x dx log x 1
dx = ´ log x + =´ 6 + x´7 dx
x7 6 6 x 6x 6
log x 1
=´ 6 ´ +C
6x 36x6

S-3: We integrate by parts, using u = x, dv = sin x dx so that v = ´ cos x and du = dx:

żπ ˇπ ż π h iπ
x sin x dx = ´x cos xˇ ´ (´ cos x ) dx = ´ x cos x + sin x = ´π (´1) = π
ˇ
0 0 0 0

129
S-4: We integrate by parts, using u = x, dv = cos x dx so that v = sin x and du = dx:
ż π
2
ˇπ ż π
2
h i π2 π
ˇ2
x cos x dx = x sin xˇ ´ sin x dx = x sin x + cos x = ´1
0 0 0 0 2

dy
S-5: We integrate by parts, using u = cos´1 y, dv = dy so that v = y and du = ´ ? :
1´y2
ż ż
´1 ´1 y
cos y dy = y cos y+ a dy
1 ´ y2
b
´1
= y cos y ´ 1 ´ y2 + C

S-6: We integrate by parts, using u = arctan(2y), dv = 4y dy so that v = 2y2 and

2 dy
du = 1+(2y)2 :

4y2
ż ż
2
4y arctan(2y) dy = 2y arctan(2y) ´ dy
(2y)2 + 1
4y2 4y2 +1
We now notice that 4y2 +1
= 4y2 +1
´ 4y21+1 . We therefore have
ż  
4y2
ż
1 1
dy = 1´ 2 dy = y ´ arctan(2y) + C
4y2 + 1 4y + 1 2

The final answer is then

ż
1
4y arctan(2y) dy = 2y2 arctan(2y) ´ y + arctan(2y) + C
2

S-7: (a) Integrate by parts with u = sinn´1 x and dv = sin x dx, so that
du = (n ´ 1) sinn´2 x cos x and v = ´ cos x.
ż ż
n n´1
sin x dx = ´ sin x cos x + (n ´ 1) cos2 x sinn´2 x dx
ż
n´1
= ´ sin x cos x + (n ´ 1) (1 ´ sin2 x ) sinn´2 x dx
ż ż
n´1 n´2
= ´ sin x cos x + (n ´ 1) sin x dx ´ (n ´ 1) sinn x dx

Moving the last term on the right hand side to the left hand side gives
ż ż
n n´1
n sin x dx = ´ sin x cos x + (n ´ 1) sinn´2 x dx

130
Dividing across by n gives the desired reduction formula.
(b) By the reduction formula of part (a)
ż π/2 ż π/2
n n´1
sin ( x ) dx = sinn´2 ( x ) dx
0 n 0

for all integers n ě 2, since sin 0 = cos π2 = 0. Applying this reduction formula, with
n = 8, 6, 4, 2,
ż π/2 ż ż ż
8 7 π/2 6 7 5 π/2 4 7 5 3 π/2
sin ( x ) dx = sin ( x ) dx = sin ( x ) dx = sin2 ( x ) dx
0 8 0 8 6 0 8 6 4 0
ż π/2
7531 7531π 35
= dx = = π « 0.4295
8642 0 8642 2 256

S-8: (a) The sketch is the figure on the left below. By integration by parts with
u = tan´1 x, dv = dx, v = x and du = 1+1x2 dx,
ż1 ˇ1 ż 1 ˇ1
´1 ´1 x ´1 1 2 ˇ
A= tan x dx = x tan xˇ ´ dx = tan 1 ´ 2 ln(1 + x )ˇ
ˇ
0 0 0 1 + x2 0
π ln 2
= ´
4 2
y y = tan−1 x y
x = tan y

x=1 x=1

x x
(b) We’ll use horizontal washers as in Example 1.6.5 of the CLP 101 notes.
• We cut R into thin horizontal strips of width dy as in the figure on the right above.
• When we rotate R about the y–axis, each strip sweeps out a thin washer
– whose inner radius is rin = tan y and outer radius is rout = 1, and
– whose thickness is dy and hence
2 ´ r 2 )dy = π (1 ´ tan2 y )dy.
– whose volume π (rout in
• As our bottommost strip is at y = 0 and our topmost strip is at y = π
4 (since at the
top x = 1 and x = tan y), the total
ż π/4 ż π/4
2
 π/4
Volume = π (1 ´ tan y) dy = π (2 ´ sec2 y) dy = π 2y ´ tan y 0
0 0
π2
= ´π
2

131
S-9: To get rid of the square root in the argument of f 2 , we make the change of variables
x = t2 , dx = 2t dt.
ż4 ż2
2
?

f x dx = 2 t f 2 (t) dt
0 0

Then, to convert f 2 into f 1 ,we integration by parts with u = t, dv = f 2 (t) dt, v = f 1 (t).
ż4 ˇ2 ż 2
?

" *
2 1 1
f x dx = 2 t f (t)ˇ ´ f (t) dt
ˇ
0 0 0
h i2
= 2 t f 1 (t) ´ f (t)
0
 1   
= 2 2 f (2) ´ f (2) + f (0) = 2 2 ˆ 4 ´ 3 + 1
= 12

Solutions to Exercises 1.8 — Jump to TABLE OF CONTENTS

S-1: Make the substitution u = sin x, so that du = cos x dx and

cos2 x = 1 ´ sin2 x = 1 ´ u2 :
ż ż ż
cos x dx = (1 ´ sin x ) cos x dx = (1 ´ u2 ) du
3 2

u3 sin3 x
= u´ + C = sin x ´ +C
3 3

S-2: Make the substitution u = sin t, so that du = cos t dt and cos2 t = 1 ´ sin2 t = 1 ´ u2 :
ż ż ż
sin t cos t dt = sin t (1 ´ sin t) cos t dt = u36 (1 ´ u2 ) du
36 3 36 2

u37 u39 sin37 t sin39 t

= ´ +C = ´ +C
37 39 37 39

S-3: First solution: Substituting u = cos x, du = ´ sin x dx, sin2 x = 1 ´ cos2 x = 1 ´ u2 ,

gives

sin3 x (1 ´ cos2 x ) sin x 1 ´ u2

ż ż ż ż
3 5
tan x sec x dx = dx = dx = ´ du
cos8 x cos8 x u8
h u´7 u´5 i 1 1
=´ ´ + C = sec7 x ´ sec5 x + C
´7 ´5 7 5

132
Second solution: Alternatively, substituting u = sec x, du = sec x tan x dx,
tan2 x = sec2 x ´ 1 = u2 ´ 1, gives
ż ż ż
tan x sec x dx = tan x sec x (tan x sec x ) dx = (u2 ´ 1)u4 du
3 5 2 4

h u7 u5 i 1 1
= ´ + C = sec7 x ´ sec5 x + C
7 5 7 5

S-4: Use the substitution u = tan x, so that du = sec2 x dx:

ż ż ż
4 46
sec x tan x dx = (tan x + 1) tan x sec x dx = (u2 + 1)u46 du
2 46 2

u49 u47 tan49 x tan47 x

= + +C = + +C
49 47 49 47

1+cos(2x )
S-5: Using the trig identity cos2 x = 2 , we have
1 π  1h iπ
ż ż
2 1 π
cos xdx = 1 + cos(2x ) dx = x + sin(2x ) =
2 0 2 2 0 2

S-6: (a) Using the trig identity tan2 x = sec2 x ´ 1 and the substitution y = tan x,
dy = sec2 x dx,
ż ż ż ż
n n´2 n´2
tan x dx = tan 2
x tan x dx = tan x sec x dx ´ tann´2 x dx
2

yn´1
ż ż ż
n´2 n´2
= y dy ´ tan x dx = ´ tann´2 x dx
n´1
tann´1 x
ż
= ´ tann´2 x dx
n´1

(b) By the reduction formula of part (a)

ż π/4 ż π/4
1
tann ( x ) dx = ´ tann´2 ( x ) dx
0 n´1 0
π
for all integers n ě 2, since tan 0 = 0 and tan = 1. Applying this reduction formula,
4
with n = 6, 4, 2,
ż π/4 ż π/4 ż π/4 ż π/4
6 1 4 1 1 2 1 1
tan ( x ) dx = ´ tan ( x ) dx = ´ + tan ( x ) dx = ´ + 1 ´ dx
0 5 0 5 3 0 5 3 0
1 1 π 13 π
= ´ +1´ = ´ « 0.0813
5 3 4 15 4

Solutions to Exercises 1.9 — Jump to TABLE OF CONTENTS

133
 4
substitution x =
S-1: (a) The ? 3 sec θ, combined with sec2 θ ´ 1 = tan2 θ eliminates the
square root 9x2 ´ 16.
1
? substitution x =
(b) The 2 sin θ, combined with 1 ´ sin2 θ = cos2 θ eliminates the square
root 1 ´ 4x2 .
2 2
? substitution x = 5 tan θ, combined with 1 + tan θ = sec θ eliminates the square
(c) The
2
root 25 + x .

S-2: Let x = 2 tan θ, so that x2 + 4 = 4 tan2 θ + 4 = 4 sec2 θ and dx = 2 sec2 θ dθ. Then
ż ż
1 1
dx = ¨ 2 sec2 θ dθ
( x + 4)3/2
2 2
(4 sec θ ) 3/2

2 sec2 θ
ż
= dθ
8 sec3 θ
ż 4
=
1
cos θ dθ ? x2 +
x
4 θ
1 1 x 2
= sin θ + C = ? 2 +C
4 4 x +4

The fact that sin θ = ? x2 when tan θ = 2x can be read off of the right angled triangle
x +4
above. In that triangle, we have chosen the lengths of the right hand and bottom sides so
that tan θ = 2x and then we determined the length of the hypotheneuse by using
Pythagorous.

S-3: Substitute x = 2 tan u, so that dx = 2 sec2 u du. Note that when x = 4 we have
4 = 2 tan u, so that tan u = 2.
ż4 ż arctan 2
1 1
3/2
dx = 3/2
2 sec2 u du
0 (4 + x2 ) 0 (4 + 4 tan2 u)
ż arctan 2
1 sec2 u
= du √
4 0 sec3 u
5
ż arctan 2 2
1 u
= cos u du 1
4 0
ˇarctan 2
1 ˇ
= sin uˇˇ
4 0
1
1
= sin(arctan 2) ´ 0 = ?
4 2 5

That the sin of arctan 2 is ?25 has been read off of the triangle above. The lengths of the
right hand side and bottom of the triangle were first chosen so that tan u = 2. Then the
hypotenuse was determined by using Pythagorous.

134
S-4: Substitute x = 5 tan u, so that dx = 5 sec2 u du.
ż ż
1 1
? dx = a 5 sec2 u du
2
x + 25 2
25 tan u + 25
ż
sec2 u
ż 25
= du = sec u du ? x2 + x
sec u u
5
ˇ ˇ
= log ˇ sec u + tan uˇ + C
ˇc x2 x ˇˇ
= log ˇ 1 + + ˇ+C
ˇ
25 5
b
x2
The fact that sec u = 1 + 25 when tan u = 5x can be read off of the right angled triangle
above. In that triangle, we have chosen the lengths of the right hand and bottom sides so
that tan u = 5x and then we determined the length of the hypotheneuse by using
Pythagorous.

S-5: Substitute x = 4 tan u, so that dx = 4 sec2 u du.

ż ż
1 1
? dx = a 4 sec2 u du
2 2
x x + 16 2 2
16 tan u 16 tan u + 16
sec2 u
ż ż
1 sec u
= du = du
16 tan2 u sec u 16 tan2 u
ż
1 cos u
= du
16 sin2 u

To finish off the integral, we’ll substitute v = sin u, dv = cos u du.

16
ż ż ż
1 1 cos u 1 dv 1
? dx = du = = ´ +C ? x2 + x
x2 x2 + 16 16 sin2 u 16 v2 16v
c u
1 1 16 4
=´ +C = ´ 1+ 2 +C
16 sin u 16 x
The fact that sin u = ? 2x when tan u = 4x can be read off of the right angled triangle
x +16
above. In that triangle, we have chosen the lengths of the right hand and bottom sides so
that tan u = 4x and then we determined the length of the hypotheneuse by using
Pythagorous.

S-6: Make the change of variables x = 5 sin θ, dx = 5 cos θ dθ. Since x = 0 corresponds to
θ = 0 and x = 52 correponds to sin θ = 12 or θ = π6 ,
ż 5/2 ż π/6 ż π/6
dx 5 cos θ dθ π
? = a = dθ =
0 25 ´ x2 0 25 ´ 25 sin2 θ 0 6

135
S-7: Substituting x = 3 sec u, so that dx = 3 sec u tan u du and
x2 ´ 9 = 9 sec2 u ´ 9 = 9 tan2 u, gives
ż ż
dx 3 sec u tan u du
? = ?
x x ´ 9 ż 9 sec2 u 9 sec2 u ´ 9
2 2
3 sec u tan u du
= ?
9 sec2 u 9 tan2 u
ż
1 du
x √ =
2
x −9 9 sec u
u 1
ż
1
3 = cos u du = sin u + C.
9 9

The bottom and hypotenuse of the right-angled triangle above have been
? chosen so that
x
? x2 ´9
sec u = 3 . By Pythagorous the right hand side is x2 ´ 9. So sin u = x and
?
x2 ´ 9
ż
dx
? = + C.
x2 x2 ´ 9 9x

There are of course equivalent ways to write this answer—for example,

 2
d
1 3
1´ +C
9 x

S-8: Substitute x = 2 sin u, so that dx = 2 cos u du.

ż a ż a
2
4 ´ x dx = 4 ´ 4 sin2 u 2 cos u du
 
ż ż
2
= 4 cos u du = 2 1 + cos(2u) du
= 2u + sin(2u) + C 2 x
= 2u + 2 sin u cos u + C u
b ?
2
= 2 arcsin 2 + x 1 ´ x4 + C
x 4 ´ x2

b
2
The fact that cos u = 1 ´ x4 when sin u = 2x can be read off of the right angled triangle
above. In that triangle, we have chosen the lengths of the right hand side and of the
hypotheneuse so that sin u = 2x and then we determined the length of the bottom side by
Pythagorous.

S-9: This integrand looks very different from those above. But it is only slightly
disguised. If we complete the square
ż ż
dx dx
? = a
3 ´ 2x ´ x2 4 ´ ( x + 1)2

136
and then make the substitution y = x + 1, dy = dx
ż ż ż
dx dx dy
? = a = a
3 ´ 2x ´ x 2 4 ´ ( x + 1) 2 4 ´ y2
we get a typical trig substitution integral. So we substitute y = 2 sin θ, dy = 2 cos θ dθ to
get
ż ż ż ż
dx dy 2 cos θ dθ y
? = a = a = dθ = θ + C = arcsin + C
3 ´ 2x ´ x 2 4 ´ y2 4 ´ 4 sin2 θ 2
x+1
= arcsin +C
2
An experienced integrator would probably substitute x + 1 = 2 sin θ directly, without
going through y.

S-10: We’ll use the trig identity cos 2θ = 2 cos2 θ ´ 1. It implies that
cos 2θ + 1 1  1 h cos 4θ + 1 i
cos2 θ = ùñ cos4 θ = cos2 2θ + 2 cos 2θ + 1 = + 2 cos 2θ + 1
2 4 4 2
cos 4θ cos 2θ 3
= + +
8 2 8
So
ż π/4 h
cos 4θ cos 2θ 3 i
ż π/4
4
cos θ dθ = + + dθ
0 0 8 2 8
h sin 4θ sin 2θ 3 iπ/4
= + + θ
32 4 8 0
1 3 π
= +
4 8 4
8 + 3π
=
32
as required.
(b) We’ll use the trig substitution x = tan θ, dx = sec2 θ dθ. Note that when θ = ˘ π4 , we
have x = ˘1. Also note that dividing the trig identity sin2 θ + cos2 θ = 1 by cos2 θ gives
the trig identity tan2 θ + 1 = sec2 θ. So
ż1 ż1
dx dx
3
=2 3
´1 ( x2 + 1) 0 ( x 2 + 1)

sec2 θ dθ
ż π/4
=2 3
0 (tan2 θ + 1)
sec2 θ dθ
ż π/4
=2
0 (sec2 θ )3
ż π/4
=2 cos4 θ dθ
0
8 + 3π
=
16

137
by part (a).

S-11: Substitute x = 25 sec u, so that dx = 25 sec u tan u du and

25x2 ´ 4 = 4(sec2 u ´ 1) = 4 tan2 u.
ż ?
25x2 ´ 4
ż
2 tan u 2
dx = 2 sec u tan u du
x 5 sec u 5
 
ż ż
2
= 2 tan u du = 2 sec2 u ´ 1 du
5x ?
25x2 ´ 4
= 2 tan u ´ 2u + C u
a
= 25x2 ´ 4 ´ 2 arcsec 5x 2
2 +C
?
The fact that tan u = 12 25x2 ´ 4 when sec u = 5x 2 can be read off of the right angled
triangle above. In that triangle, we have chosen the lengths of the bottom side and of the
hypotheneuse so that sec u = 5x 2
2 , i.e. cos u = 5x , and then we determined the length of
the right hand side by Pythagorous.

Solutions to Exercises 1.10 — Jump to TABLE OF CONTENTS

S-1: The partial fraction expansion has the form

3x3 ´ 2x2 + 11 A
2 2
= + various terms
x ( x ´ 1)( x + 3) x´1
When we multiply through by the original denominator, this becomes

3x3 ´ 2x2 + 11 = x2 ( x2 + 3) A + ( x ´ 1)(other terms).

Evaluating both sides at x = 1 yields 3 ¨ 13 ´ 2 ¨ 12 + 11 = 12 (12 + 3) A + 0, or A = 3.

S-2:
x3 + 3 x3 + 3
=
( x 2 ´ 1)2 ( x 2 + 1) ( x ´ 1)2 ( x + 1)2 ( x 2 + 1)
A B C D Ex + F
= + + + +
x ´ 1 ( x ´ 1)2 x + 1 ( x + 1)2 x2 + 1

S-3: This is a section on partial fractions. So of course we are going to use partial
fractions. We start by expressing the integrand, i.e. the fraction x+1x2 = x(11+ x) , as a linear
1 1
combination of the simpler fractions x and x +1 (which we already know how to
integrate). We will have
1 1 a b a( x + 1) + bx
= = + =
x + x2 x (1 + x ) x x+1 x (1 + x )

138
The fraction on the left hand side is the same as the fraction on the right hand side if and
only if the numerator on the left hand side, which is 1 = 0x + 1, is equal to the numerator
on the right hand side, which is a( x + 1) + bx = ( a + b) x + a. This in turn is the case if
and only of a = 1 (i.e. the constant terms are the same in the two numerators) and
a + b = 0 (i.e. the coefficients of x are the same in the two numerators). So a = 1 and
b = ´1. Now we can easily do the integral
ż2h
1 i h i2
ż2 ż2
dx dx 1 3
2
= = ´ dx = log x ´ log ( x + 1 ) = log 2 ´ log
1 x+x 1 x ( x + 1) 1 x x+1 1 2
4
= log
3

S-4: We’ll first do a partial fractions expansion. The sneaky way is to temporarily rename
x2 to y. Then x4 + x2 = y2 + y and
1 1 1 1
= = ´
x4 +x 2 y ( y + 1) y y+1
Now we restore y to x2 giving
ż h
1 i
ż
1 1 1
dx = ´ dx = ´ ´ arctan x + C
x4 + x2 x2 x2 + 1 x

S-5: The integrand is of the form N ( x )/D ( x ) with D ( x ) already factored and N ( x ) of
lower degree. We immediately look for a partial fractions decomposition:
12x + 4 A Bx + C
2
= + 2 .
( x ´ 3)( x + 1) x´3 x +1
Multiplying through by the denominator yields
12x + 4 = A( x2 + 1) + ( Bx + C )( x ´ 3) (˚)
Setting x = 3 we find:
36 + 4 = A(9 + 1) + 0 ùñ 40 = 10A ùñ A = 4
Substituting A = 4 in (˚) gives
12x + 4 = 4( x2 + 1) + ( Bx + C )( x ´ 3) ùñ ´4x2 + 12x = ( x ´ 3)( Bx + C )
ùñ (´4x )( x ´ 3) = ( Bx + C )( x ´ 3)
ùñ B = ´4, C = 0
So we have found that A = 4, B = ´4, and C = 0. Therefore
ż ż  
12x + 4 4 4x
dx = ´ dx
( x ´ 3)( x2 + 1) x ´ 3 x2 + 1
= 4 log |x ´ 3| ´ 2 log( x2 + 1) + C
Here the second integral was found just by guessing an antiderivative. Alternatively, one
could use the substitution u = x2 + 1, du = 2x dx.

139
S-6: The integrand is of the form N ( x )/D ( x ) with D ( x ) already factored and N ( x ) of
lower degree. We immediately look for a partial fractions decomposition:

3x2 ´ 4 A Bx + C
2
= + 2
( x ´ 2)( x + 4) x´2 x +4
Multiplying through by the denominator gives

3x2 ´ 4 = A( x2 + 4) + ( Bx + C )( x ´ 2) (˚)

Setting x = 2 we find:

12 ´ 4 = A(4 + 4) + 0 ùñ 8 = 8A ùñ A = 1

Substituting A = 1 in (˚) gives

3x2 ´ 4 = ( x2 + 4) + ( x ´ 2)( Bx + C ) ùñ 2x2 ´ 8 = ( x ´ 2)( Bx + C )

ùñ ( x ´ 2)(2x + 4) = ( x ´ 2)( Bx + C )
ùñ B = 2, C = 4

Thus, we have:

3x2 ´ 4 1 2x + 4 1 2x 4
= + = + +
( x ´ 2)( x2 + 4) x ´ 2 x2 + 4 x ´ 2 x2 + 4 x2 + 4
The first two of these are directly integrable:
ż
2 4
F ( x ) = log |x ´ 2| + log |x + 4| + dx
x2 +4
(The second integral was found just by guessing an antiderivative. Alternatively, one
could use the substitution u = x2 + 4, du = 2x dx.) For the final integral, we substitute:
x = 2y, dx = 2dy, and see that:
ż ż
4 1
2
dx = 2 2
dy = 2 arctan y + D = 2 arctan( x/2) + D
x +4 y +1
for any constant D. All together we have:

F ( x ) = log |x ´ 2| + log |x2 + 4| + 2 arctan( x/2) + D

S-7: This sure looks like a partial fractions problem. So let’s go through our protocol.
• The degree of the numerator x ´ 13 is one, which is strictly smaller than the dergee
of the denominator x2 ´ x ´ 6, which is two. So we do not long divide to pull out a
polynomial.
• Next we factor the denominator.

x2 ´ x ´ 6 = ( x ´ 3)( x + 2)

140
 • Next we find the partial fractions expansion of the integrand. It is of the form

x ´ 13 A B
= +
( x ´ 3)( x + 2) x´3 x+2

To find A and B, using the sneaky method, we cross multiply by the denominator.

x ´ 13 = A( x + 2) + B( x ´ 3)

Now we can find A by evaluating at x = 3

3 ´ 13 = A(3 + 2) + B(3 ´ 3) ùñ A = ´2

and find B by evaluating at x = ´2.

´2 ´ 13 = A(´2 + 2) + B(´2 ´ 3) ùñ B = 3

(Hmmm. A and B are nice round numbers. Sure looks like a rigged exam or
problem set problem.) So our partial fraction expansion is

x ´ 13 ´2 3
= +
( x ´ 3)( x + 2) x´3 x+2

As a check, we recombine the right hand side and make sure that it matches the left
hand side
´2 3 ´2( x + 2) + 3( x ´ 3) x ´ 13
+ = =
x´3 x+2 ( x ´ 3)( x + 2) ( x ´ 3)( x + 2)

• Finally, we do the integral

ż ż  
x ´ 13 ´2 3
dx = + dx = ´2 log |x ´ 3| + 3 log |x + 2| + C
x2 ´ x ´ 6 x´3 x+2

S-8: Again, this sure looks like a partial fractions problem. So let’s go through our
protocol.
• The degree of the numerator 5x + 1 is one, which is strictly smaller than the dergee
of the denominator x2 + 5x + 6, which is two. So we do not long divide to pull out
a polynomial.
• Next we factor the denominator.

x2 + 5x + 6 = ( x + 2)( x + 3)

• Next we find the partial fractions expansion of the integrand. It is of the form

5x + 1 A B
= +
( x + 2)( x + 3) x+2 x+3

141
 To find A and B, using the sneaky method, we cross multiply by the denominator.
5x + 1 = A( x + 3) + B( x + 2)
Now we can find A by evaluating at x = ´2
´10 + 1 = A(´2 + 3) + B(´2 + 2) ùñ A = ´9
and find B by evaluating at x = ´3.
´15 + 1 = A(´3 + 3) + B(´3 + 2) ùñ B = 14
So our partial fraction expansion is
5x + 1 ´9 14
= +
( x + 2)( x + 3) x+2 x+3
As a check, we recombine the right hand side and make sure that it matches the left
hand side
´9 14 ´9( x + 3) + 14( x + 2) 5x + 1
+ = =
x+2 x+3 ( x + 2)( x + 3) ( x + 2)( x + 3)

• Finally, we do the integral

ż ż  
5x + 1 ´9 14
dx = + dx = ´9 log |x + 2| + 14 log |x + 3| + C
x2 + 5x + 6 x+2 x+3

Solutions to Exercises 1.11 — Jump to TABLE OF CONTENTS

S-1: True. Because f ( x ) is positive and concave up, the graph of f ( x ) is always below the
top of the trapezoids used in the trapezoidal rule.

şb
S-2: By (1.11.2) in the CLP 101 notes, the midpoint rule approximation to a f ( x ) dx with
n = 3 is
żb
 
f ( x ) dx « f ( x̄1 ) + f ( x̄2 ) + f ( x̄3 ) ∆x
a

where ∆x = b´a
3 and
x0 = a x1 = a + ∆x x2 = a + 2∆x x3 = b
x0 + x1 x1 + x2 x2 + x3
x̄1 = 2 x̄2 = 2 x̄3 = 2
For this problem, a = 0, b = π and f ( x ) = sin x, so that ∆x = π
3 and
π 2π
x0 = 0 x1 = 3 x2 = 3 x3 = π
π π 5π
x̄1 = 6 x̄2 = 2 x̄3 = 6
and
 π 1 
żπ
sin x dx « sin π6 + sin π2 + sin 5π 1 π
6 3 = 2 +1+ 2 3 =
2π
3
0

142
S-3: Let f ( x ) be the diameter a distance x from the left end of the log. The cross sectional
f ( x )
2
area a distance x from the left end of the log is then π 2 = π4 f ( x )2 . The volume is
π 1h i
ż6
π
V= f ( x )2 dx « f (0)2 + 4 f (1)2 + 2 f (2)2 + 4 f (3)2 + 2 f (4)2 + 4 f (5)2 + f (6)2
0 4 43
πh 2 i
= 1.2 + 4(1)2 + 2(0.8)2 + 4(0.8)2 + 2(1)2 + 4(1)2 + 1.22
12
= 4.377 m3
where we have approximated the integral using Simpson’s Rule with ∆x = 1.

S-4: Let f ( x ) denote the diameter at height x. As in Example 1.6.6 of the CLP 101 notes,
we slice V into thin horizontal “pancakes”, which in this case are circular.
• We are told that the pancake at height x is a circular disk of diameter f ( x ) and so
f ( x )
2
• has cross-sectional area π 2 and thickness dx and hence
f ( x )
2
• has volume π 2 dx.
Hence the volume of V is
ż2 h i2
f (x) π h i
π 2 dx « 10 12 f (0)2 + f (10)2 + f (20)2 + f (30)2 + 12 f (40)2
0 4
π h1 2 2 2 2 1 2
i
= 10 2 24 + 16 + 10 + 6 + 2 4
4
= 688 ˆ 2.5π = 1720π = 5403.5
where we have approximated the integral using the trapezoidal rule with ∆x = 10.

c( x )
S-5: Call the circumference at height x, c( x ). The corresponding radius is 2π and the
c ( x )
2 x )2
corresponding cross–sectional area is π 2π = c(4π . Hence the total volume is
1 2h i
ż8
c ( x )2
dx « c(0)2 + 4c(2)2 + 2c(4)2 + 4c(6)2 + c(8)2
0 4π 4π 3
1 2h 2 i
= 1.2 + 4(1.1)2 + 2(1.3)2 + 4(0.9)2 + 0.22 = 0.6865
4π 3

S-6: (a) The Trapezoidal Rule gives

ż 60 h i
V= A(h) dh « 10 21 A(0) + A(10) + A(20) + A(30) + A(40) + A(50) + 12 A(60)
0
= 363,500

(b) Simpson’s Rule gives

20 h i
ż 60
V= A(h) dh « A(0) + 4A(10) + 2A(20) + 4A(30) + 2A(40) + 4A(50) + A(60)
0 6
= 367,000

143
S-7: Call the curve in the graph y = f ( x ). It looks like

f (2) = 3 f (3) = 8 f (4) = 7 f (5) = 6 f (6) = 4

(a) The Trapezoidal Rule gives

1
( 49
T4 = 2 3+2ˆ8+2ˆ7+2ˆ6+4 ˆ1 =
2

(b) Simpson’s Rule gives

1 ( 77
S4 = 3+4ˆ8+2ˆ7+4ˆ6+4 ˆ1 =
3 3

S-8: Let f ( x ) = sin( x2 ). Then f 1 ( x ) = 2x cos( x2 ) and

f 2 ( x ) = 2 cos( x2 ) ´ 4x2 sin( x2 ).

Since |x2 | ď 1 when |x| ď 1, and |sin θ| ď 1 and |cos θ| ď 1 for all θ, we have
ˇ ˇ
ˇ2 cos( x2 ) ´ 4x2 sin( x2 )ˇ ď 2| cos( x2 )| + 4x2 | sin( x2 )| ď 2 ˆ 1 + 4 ˆ 1 ˆ 1 = 2 + 4 = 6
ˇ ˇ

We can therefore choose K = 6, and it follows that the error is at most

K [ b ´ a ]3 6 ¨ [1 ´ (´1)]3 2
2
ď 2
= 6
= 2 ¨ 10´6
24n 24 ¨ 1000 10

S-9: Setting f ( x ) = 2x4 and b ´ a = 1 ´ (´2) = 3, we compute f 2 ( x ) = 24x2 . The largest

value of 24x2 on the interval [´2, 1] occurs at x = ´2, so we can take M = 24 ¨ (´2)2 = 96.
Thus the total error for the midpoint rule with n = 60 points is bounded by

M ( b ´ a )3 96 ˆ 33 3
2
= =
24n 24 ˆ 60 ˆ 60 100

S-10: (a) Since a = 0, b = 2 and n = 6, we have ∆x = b´a 2´0 1

n = 6 = 3 , and so x0 = 0,
x1 = 13 , x2 = 23 , x3 = 1, x4 = 43 , x5 = 35 , and x6 = 2. Since Simpson’s Rule with n = 6 in
general is
∆x  
f ( x0 ) + 4 f ( x1 ) + 2 f ( x2 ) + 4 f ( x3 ) + 2 f ( x4 ) + 4 f ( x5 ) + f ( x6 ) ,
3
the desired approximation is
 1 5 2 5 4 5 5 5 
1/3 5 5 5
(´3) + 4 ´ 3 + 2 ´ 3 + 4(´2) + 2 ´ 3 + 4 ´ 3 + (´1)
3 3 3 3 3

144
(b) Here f ( x ) = ( x ´ 3)5 , which has derivatives

f 1 ( x ) = 5( x ´ 3)4 f 2 ( x ) = 20( x ´ 3)3

f (3) ( x ) = 60( x ´ 3)2 f (4) ( x ) = 120( x ´ 3).

For 0 ď x ď 2, ( x ´ 3) runs from ´3 to ´1, so the maximum absolute values are found at
x = 0, giving K = 20 ¨ |0 ´ 3|3 = 540 and L = 120 ¨ |0 ´ 3| = 360. Consequently, for the
Midpoint Rule with n = 100,

K ( b ´ a )3 540 ˆ 23 180
|E M | ď 2
= 4
= 4;
24n 24 ˆ 10 10
whereas for Simpson’s Rule with n = 10,

360 ˆ 25 64
|ES | ď 4
= 4.
180 ˆ 10 10
Since 64 ă 180, Simpson’s Rule results in a smaller error bound.

S-11: In this case, a = 1, b = 4 and we may take K = 2. So we need n to obey

2(4 ´ 1)3 2 2(3)3 27000 9000

2
ď 0.001 ðñ n ě 1000 = = = 4500
12n 12 6 2
One obvious allowed n is 100. Any n ě 68 works.

şb
S-12: In general the error in approximating a f ( x ) dx using Simpson’s rule with n steps
K (b´a)
180 ( ∆x )
where ∆x = b´a
4 (4) ( x )| for all a ď x ď b. In this
is bounded by n and K ě | f
case, a = 1, b = 5, n = 4 and f ( x ) = 1x . So

1 2 6 24
f 1 (x) = ´ f 2 (x) = f (3) ( x ) = ´ f (4) ( x ) =
x2 x3 x4 x5
and
ˇ (4) ˇ
ˇ f ( x )ˇ ď 24 for all x ě 1

So we may take K = 24 and ∆x = 5´1

4 = 1 and

24(5 ´ 1) 4 24 8
Error ď (1) = =
180 45 15

şb
S-13: In general the error in approximating a f ( x ) dx using Simpson’s rule with n steps
K (b´a)
180 ( ∆x ) ∆x = n and K ě | f (4) ( x )| for all a ď x ď b. In this
4 b´a
is bounded by where
case, a = 0, b = 1, n = 6 and f ( x ) = e´2x + 3x3 . So

f 1 ( x ) = ´2e´2x + 9x2 f 2 ( x ) = 4e´2x + 18x f (3) ( x ) = ´8e´2x + 18 f (4) ( x ) = 16e´2x

145
and
ˇ (4) ˇ
ˇ f ( x )ˇ ď 16 for all x ě 0

So we may take K = 16 and ∆x = 1´0

6 = 1
6 and

16(1 ´ 0) 16 1 1
Error ď (1/6)4 = 4
= 4
=
180 180 ˆ 6 180 ˆ 3 14580

1 b´a
S-14: Set a = 1, b = 2, n = 4, f ( x ) = x and h = n = 14 . Then
(a)

h 
T4 = f ( x0 ) + 2 f ( x1 ) + 2 f ( x2 ) + 2 f ( x3 ) + f ( x4 )
2
h 
= f (1) + 2 f (5/4) + 2 f (6/4) + 2 f (7/4) + f (2)
2
1h 4 4 4 1i
= 1+2ˆ +2ˆ +2ˆ +
8 5 6 7 2

(b)

h 
S4 = f ( x0 ) + 4 f ( x1 ) + 2 f ( x2 ) + 4 f ( x3 ) + f ( x4 )
3
h 
= f (1) + 4 f (5/4) + 2 f (6/4) + 4 f (7/4) + f (2)
3
1h 4 4 4 1i
= 1+4ˆ +2ˆ +4ˆ +
12 5 6 7 2

(c) In this case, a = 1, b = 2, n = 4 and f ( x ) = 1x . So

1 2 6 24
f 1 (x) = ´ f 2 (x) = f (3) ( x ) = ´ f (4) ( x ) =
x2 x3 x4 x5

and
ˇ (4) ˇ
ˇ f ( x )ˇ ď 24 for all x ě 1

So we may take K = 24 and

K ( b ´ a )5 24(2 ´ 1)5 24 3
Error ď ď = =
180 ˆ n4 180 ˆ 44 180 ˆ 44 5760

b´a
S-15: Set a = 0, b = 8, n = 4, f ( x ) = s( x ) and h = n = 2. Then

146
(a)

h 
T4 = f ( x0 ) + 2 f ( x1 ) + 2 f ( x2 ) + 2 f ( x3 ) + f ( x4 )
2
= 1.00664 + 2 ˆ 1.00543 + 2 ˆ 1.00435 + 2 ˆ 1.00331 + 1.00233
= 8.03515
h 
S4 = f ( x0 ) + 4 f ( x1 ) + 2 f ( x2 ) + 4 f ( x3 ) + f ( x4 )
3
2 
= 1.00664 + 4 ˆ 1.00543 + 2 ˆ 1.00435 + 4 ˆ 1.00331 + 1.00233
3
= 8.03509

(b)
ˇż b
ˇ K2 ( b ´ a ) 3 83
ˇ
ˇ 2
ˇ f ( x ) dx ´ T n
ˇď ď = 0.00533
ˇ
a
ˇ 12n2 1000 12(4)2
ˇż b
ˇ K4 ( b ´ a ) 5 85
ˇ
ˇ 4
ˇ f ( x ) dx ´ S n
ˇď ď = 0.00284
ˇ
a
ˇ 180n4 1000 180(4)4

S-16: Denote by f ( x ) the width of the pool a distance x from the left hand end. Thus
f (0) = 0, f (2) = 10, f (4) = 12, f (6) = 10, f (8) = 8, f (10) = 6, f (12) = 8, f (14) = 10
and f (16) = 0. The volume of the part of the pool with x–coordinate running from x to
f ( x )
2
x + dx is 12 π 2 dx. So the total volume is

π 16
ż
V= f ( x )2 dx
8 0
π ∆x h i
« f (0)2 + 4 f (2)2 + 2 f (4)2 + 4 f (6)2 + 2 f (8)2 + 4 f (10)2 + 2 f (12)2 + 4 f (14)2 + f (16)2
8 3
π 2h i
= 0 + 4(10)2 + 2(12)2 + 4(10)2 + 2(8)2 + 4(6)2 + 2(8)2 + 4(10)2 + 0
83
« 494 ft3

S-17: (a) The Trapezoidal Rule gives

1h1 i
ż1
´6 1 1
1 1
3 3
1
M = 2π10 rg(r ) dr « 2π10´6 0g(0) + g + g + g + g (1)
0 4 2 4 4 2 2 4 4 2
= 0.025635

(b) In this case, the integrand f (r ) = 2π10´6 rg(r ) obeys

d    
f 2 (r ) = 2π10´6 g(r ) + rg1 (r ) = 2π10´6 2g1 (r ) + rg2 (r )
dr

147
and hence, for 0 ď r ď 1,
ˇ 2 ˇ  
ˇ f (r )ˇ ď 2π10´6 2 ˆ 200 + 1 ˆ 150 = 1.1π10´3

So we may take K = 1.1π10´3 and, as a = 0, b = 1, and n = 4,

1.1π10´3 (1 ´ 0)3
error ď 2
ď 1.8 ˆ 10´5
12(4)

S-18: (a) Let f ( x ) = 1x , a = 1, b = 2 and ∆x = b´a 1

6 = 6 . By Simpson’s rule

∆x h 7 8 9  10   11  i

ż2
1
dx « f (1) + 4 f + 2f + 4f + 2f + 4f + f (2)
1x 3 6 6 6 6 6
1 h 24 12 24 12 24 1 i
= 1+ + + + + + = 0.6931698
18 7 8 9 10 11 2

(b) The integrand is f ( x ) = 1x . The first four derivatives of f ( x ) are f 1 ( x ) = ´ x12 ,

f 2 ( x ) = x23 , f (3) ( x ) = ´ x64 , f (4) ( x ) = 24
x5
. On the interval 1 ď x ď 2, the fourth derivative
is never bigger in magnitude than K = 24. So
K ( b ´ a )5 24(2 ´ 1)5 4
|En | ď 4
= 4
=
180n 180n 30n4
b
4 4
This is no more than 0.00001 = 10´5 if n4 ě 30 105 or n ě 4 30 105 = 10.75 or n ě 12
(since n must be even for Simpson’s Rule).

S-19: (a) From the figure, we see that the magnitude of f 4 ( x ) never exceeds 310 for
0 ď x ď 2. So the error is bounded by
310(2 ´ 0)5
= 0.01345
180 ˆ 84

(b) We need to choose n so that

c
310(2 ´ 0)5 ´4 4 310 ˆ 25 4 4 310 ˆ 32
ď 10 ðñ n ě 10 ðñ n ě 10 = 27.2
180 ˆ n4 180 180
For Simpson’s rule, n must be even so choose an even integer obeying n = 28.
şx ? ?
S-20: Let g( x ) = 0 sin( t) dt. Since g1 ( x ) = sin( x ) and f ( x ) = g( x2 ),
f 1 ( x ) = 2xg1 ( x2 ) = 2x sin x f 2 ( x ) = 2 sin x + 2x cos x
Since | sin x|, | cos x| ď 1, we have | f 2 ( x )| ď 2 + 2|x| and, for 0 ď t ď 1, | f 2 (t)| ď 4. When
the trapezoidal rule with n subintervals is applied, the resulting error En obeys
4(1 ´ 0)3 4
En ď 2
ď 0.000005 ðñ n2 ě ðñ n ě 259
12n 12 ˆ 0.000005

148
S-21: The Trapezoidal Rule gives

1h1 1 1 1 1 3 3 1 i

ż1
´6
M = 2π10 rg(r ) dr « 2π10´6 0g(0) + g + g + g + g (1)
0 4 2 4 4 2 2 4 4 2
= 0.025635

(b) In this case, the integrand f (r ) = 2π10´6 rg(r ) obeys

d    
f 2 (r ) = 2π10´6 g(r ) + rg1 (r ) = 2π10´6 2g1 (r ) + rg2 (r )
dr
and hence, for 0 ď r ď 1,
ˇ 2 ˇ  
ˇ f (r )ˇ ď 2π10´6 2 ˆ 200 + 1 ˆ 150 = 1.1π10´3

So
1.1π10´3 (1 ´ 0)3
error ď ď 1.8 ˆ 10´5
12(4)2

Solutions to Exercises 1.12 — Jump to TABLE OF CONTENTS

ş8
S-1: False. For example if f ( x ) = e´x and g( x ) = 1 then 1 f ( x ) dx converges but
ş8
1 g ( x ) dx diverges.

S-2: Notice that

$ 1 1´5q ´ 1) with 1 ´ 5q ą 0, if q ă 1 ,
żt
1 & 1´5q (t
’ 5
dx = 1
5q log t, if q = 5 ,
1 x % 1 1
if q ą 15 .
’
5q´1 (1 ´ t5q´1 ) with 5q ´ 1 ą 0,

Therefore
$ 1
ż8 ż t  & 1´5q (limtÑ8 t
’ 1´5q ´ 1) = 8, if q ă 51 ,
1 1
dx = lim dx = limtÑ8 log t = 8, if q = 15 ,
1 x5q tÑ8 1 x5q % 1 1 1
if q ą 15 ;
’
5q´1 (1 ´ limtÑ8 t5q´1 ) = 5q´1 ,

The first two cases are divergent, and so the largest such value is q = 15 . (Alternatively,
we might recognize this as a “p-integral” with p = 5q, and recall that the p-integral
diverges precisely when p ď 1.)

S-3: The denominator is zero when x = 1, so the integrand has a singularity at x = 1. So

ż1 żt
x4 x4
dx = lim dx
0 x5 ´ 1 tÑ1´ 0 x5 ´ 1

149
To evaluate this integral we use the substitution u = x5 , du = 5x4 dx. When x = 0 we
have u = 0 and when x = t, we have u = t5 , so
ż x =t ż u = t5 ˇt5
x4 1 1 1
dx = du = log 1| = log |t5 ´ 1|
ˇ
5
|u ´ ˇ
x =0 x ´ 1 u =0 5 ( u ´ 1 ) 5 0 5
This diverges as t Ñ 1´ , so the integral diverges.

S-4: The integrand has a singularity at x = ´1. So

ż2 żt ż2
1 1 1
4/3
dx = lim 4/3
dx + lim 4/3
dx
´2 ( x + 1) tÑ´1´
´2 ( x + 1) tÑ´1 +
t ( x + 1)

Since żt
1 3 ˇt 3 3
dx = ˇ =´ +
ˇ
4/3
´ 1/3 1/3
´2 ( x + 1) ( x + 1) ´2 ( t + 1) (´1)1/3
diverges as t Ñ ´1´ , the integral diverges. (A similar argument shows that the second
integral diverges, which is also enough to conclude that the original integral diverges.)
? ?
S-5: For all x ě 1, 4x2 ´ x ď 4x2 = 2x so
ż8 żt żt
1 1 1 1 ˇt 1
? dx = lim ? dx ě lim dx = lim ln xˇ1 = lim ln t = 8
1 2
4x ´ x tÑ8 1 2
4x ´ x tÑ8 1 2x tÑ8 2 tÑ8 2

So the integral does not converge.

S-6: The integrand is positive everywhere. So either the integral converges to some finite
number or it is infinite. Since
ż1
1 1 dx
? ď ? and the integral ? converges by Example 1.12.9
x2 + x x 0 x
ż8
1 1 dx
? ď and the integral converges by Example 1.12.8
x2 + x x2 1 x
2

the integral converges by the comparison test. (The examples are in the CLP 101 notes.)

S-7: You might think that, because the integrand is odd, the integral converges to 0. But
you would be wrong. There are two “sources of impropriety”, namely x « +8 and
x « ´8. So we split the integral in two
ż +8 ż0 ż +8
x x x
2
dx = 2
dx + dx
´8 x + 1 ´8 x + 1 0 x2 + 1
and treat the two halves separately
ż +8 żR ˇR
x x 1 2 1
dx = lim dx = lim log ( x + 1 ) ˇ = lim log( R2 + 1) = +8
ˇ
2
x +1 2
RÑ8 0 x + 1 RÑ8 2 0 RÑ8 2
0
ż0 ż0
x x 1 ˇ0 1
ˇ
2
2
dx = lim 2
dx = lim log ( x + 1 ) ˇ = lim ´ log( R2 + 1) = ´8
´8 x + 1 RÑ8 ´R x + 1 RÑ8 2 ´R RÑ8 2

150
Both halves diverge, so the whole integral diverges. Don’t make the mistake of thinking
that 8 ´ 8 = 0. That can get you into big trouble. 8 is not a normal number. For
example 28 = 8. So if 8 were a normal number we would have both 8 ´ 8 = 0 and
8 ´ 8 = 28 ´ 8 = 8.

S-8: Since
ż1
| sin x| 1 dx
ď and the integral converges by Example 1.12.9
x3/2 + x1/2 x1/2 0 x
1/2
ż8
| sin x| 1 dx
ď and the integral converges by Example 1.12.8
x3/2 + x1/2 x3/2 1 x
3/2

the integral converges by the comparison test. (The examples are in the CLP 101 notes.)

S-9: The integrand is positive everywhere. So either the integral converges to some finite
number or it is infinite. There are two potential “sources of impropriety” — a possible
singularity at x = 0 and the fact that the domain of integration extends to 8. So we split
up the integral.
ż8 ż1 ż8
x+1 x+1 x+1
dx = dx + dx
0 x ( x 2 + x + 1)
1/3
0 x ( x 2 + x + 1)
1/3
1 x1/3 ( x2
+ x + 1)

When x « 0, the integrand

x+1 1 1
« 1/3 = 1/3
x1/3 ( x2
+ x + 1) x (1) x

When x is very large

x+1 x 1
« =
x1/3 ( x2 + x + 1) x1/3 ( x2 ) x4/3
In fact
ż1 ż1 ˇ1
x+1 2 x2/3 ˇˇ 2
dx ď dx = 2 = =3
0 x1/3 ( x2 + x + 1) 0x1/3 2/3 0 ˇ 2/3
ż8 żR żR
x+1 x+1 x+1
dx = lim dx ď lim dx
1 x [ x 2 + x + 1]
1/3 RÑ8 1 1/3 2
x [ x + x + 1] RÑ8 1 x 1/3 [ x ( x + 1)]
żR ˇR
1 x´1/3 ˇˇ 1
= lim dx = lim = =3
RÑ8 1 x4/3 RÑ8 ´1/3 ˇ
1 1/3
ż8
x+1
ùñ dx ď 6
0 x1/3 ( x2
+ x + 1)

So the integral converges.

S-10: The integrand is positive everywhere. So either the integral converges to some
finite number or it is infinite. There are two potential “sources of impropriety” — a

151
possible singularity at x = 0 and the fact that the domain of integration extends to 8. So
we split up the integral.
ż1
sin4 x sin4 x sin4 x
ż8 ż8
dx = dx + dx
0 x2 0 x2 1 x2

We’ll treat the first integral first. By l’Hôpital’s rule (or recall Example ?? in the CLP100
notes)
sin x cos x
lim = lim = cos 0 = 1
xÑ0 x xÑ0 1
Consequently

sin4 x   sin x  sin x 

2
lim = lim sin x lim lim = 0ˆ1ˆ1 = 0
xÑ0 x2 xÑ0 xÑ0 x xÑ0 x

and the first integral is not even improper.

Now for the second integral. Since | sin x| ď 1

sin4 x
ż8 ż8
dx x´1 ˇˇ8
dx ď = ˇ =1
1 x2 1 x
2 ´1 1
and second integral converges by the comparison test. So the original integral converges
too.

e´x e´x
ş8
S-11: Let’s first find a t such that t 1+ x dx ď 21 10´4 . For all x ě 0, 0 ă 1+ x ď e´x , so
ż8
e´x
ż8
1 1 
dx ď e´x dx = e´t ď 10´4 if t ě ´ log 10´4 = 9.90
t 1+x t 2 2
Choose, for example, t = 10.

e´x e´x e´x e´x e´x e´x

f (x) = ùñ f 1 ( x ) = ´ ´ ùñ f 2
( x ) = + 2 + 2
1+x 1 + x (1 + x )2 1+x (1 + x )2 (1 + x )3
Since f 2 ( x ) is positive and decreases as x increases

5(10 ´ 0)3 5000 625

| f 2 ( x )| ď f 2 (0) = 5 ùñ |En | ď 2
= 2
= 2
24n 24n 3n
and |En | ď 21 10´4 if
c
625 1 ´4 2 1250 ˆ 104 1.25 ˆ 107
ď 10 ðñ n ě ðñ n ě = 2041.2
3n2 2 3 3
So t = 10 and n = 2042 will do the job. There are many other correct answers.

Solutions to Exercises 1.13 — Jump to TABLE OF CONTENTS

152
S-1: Substitute u = x2 ´ 3, du = 2x dx. This gives
ż ż
x du/2 1 1 ˇ 2 ˇ
dx = = log |u| + C = log ˇ x ´ 3ˇ + C
x2 ´ 3 u 2 2

dy
S-2: (a) Substituting y = 9 + x2 , dy = 2xdx, xdx = 2 , y(0) = 9, y(4) = 25
ż4 ż 25 ?
x 1 dy 1 y ˇˇ25
? dx = ? = ˇ = 5´3
0 9 + x2 9 y 2 2 1/2 9
=2

(b) Substituting y = sin x, dy = cos x dx, y(0) = 0, y(π/2) = 1, cos2 x = 1 ´ y2

ż π/2 ż π/2 ż1 ż1
cos x sin x dx = cos x sin x cos x dx = (1 ´ y )y dy = (y2 ´ y4 ) dy
3 2 2 2 2 2
0 0 0 0
 1
y3 y5 1 1
= ´ = ´
3 5 0 3 5
2
=
15

(c) Integrate by parts with u( x ) = log x and dv = x3 dx so that du = 1x dx and v = x4 /4.

Then
że ˇe ż e 4 że 3 ˇe
3 x4 ˇ x 1 e4 x e4 x4 ˇˇ
x log x dx = log xˇ ´
ˇ dx = ´ dx = ´ ˇ
1 4 1 1 4 x 4 1 4 4 16 1
3e4 1
= +
16 16

S-3: (a) Integrate by parts with u = x and dv = sin x dx so that du = dx and v = ´ cos x.
Then
ż ż
x sin x dx = ´x cos x ´ (´ cos x ) dx = ´x cos x + sin x + C

so that
ż π/2 h iπ/2
x sin x dx = ´ x cos x + sin x =1
0 0

(b) Make the substitution u = sin x, du = cos x dx.

ż1 ż1

2


ż π/2 ż π/2
5 2 2 2
cos x dx = 1 ´ sin x cos x dx = 1 ´ u du = 1 ´ 2u2 + u4 du
0 0 0 0
h 2 1 i1 2 1 8
= u ´ u3 + u5 = 1 ´ + =
3 5 0 3 5 15

153
S-4: (a) By integration by parts, with u = x and dv = e x dx, so that du = dx and v = e x ,
ż2 ˇ2 ż 2 ˇ2
x xˇ x xˇ
xe dx = xe ˇ ´ e dx = 2e ´ e ˇ = e2 + 1
2
0 0 0 0

(b) Substitute x = tan y, dx = sec2 y b dy. When x = 0, tan y = 0 so y = 0. When x = 1,

? a
tan y = 1 so y = 4 . Also 1 + x2 = 1 + tan2 y = sec2 y = sec y, since sec y ě 0 for all
π

0 ď y ď π4 .
ż1
sec2 y dy
ż π/4 ż π/4
1 ˇπ/4
? dx = = sec y dy = log | sec y + tan y|ˇ0
0 1 + x2 0 sec y 0
?
Since sec π4 = 2, tan π4 =?1, sec 0 = 1, tan 0 = 0, the top evaluation
log | sec π4 + tan π4 | = log( 2 + 1), the bottom evaluation log | sec 0 + tan 0| = log 1 = 0
and
ż1 ?
1
? dx = log( 2 + 1)
0 1 + x2

(c) We use partial fractions.

4x 4x a b cx + d
= = + +
( x2 ´ 1)( x2 + 1) ( x ´ 1)( x + 1)( x2 + 1) x ´ 1 x + 1 x2 + 1

a( x + 1)( x2 + 1) + b( x ´ 1)( x2 + 1) + (cx + d)( x ´ 1)( x + 1)

=
( x ´ 1)( x + 1)( x2 + 1)
provided a( x + 1)( x2 + 1) + b( x ´ 1)( x2 + 1) + (cx + d)( x ´ 1)( x + 1) = 4x. Setting
x = 1 gives 4a = 4 or a = 1. Setting x = ´1 gives ´4b = ´4 or b = 1. Substituting in
a = b = 1 gives

( x + 1)( x2 + 1) + ( x ´ 1)( x2 + 1) + (cx + d)( x ´ 1)( x + 1) = 4x

ðñ 2x ( x2 + 1) + (cx + d)( x ´ 1)( x + 1) = 4x
ðñ (cx + d)( x2 ´ 1) = ´2x3 + 2x = ´2x ( x2 ´ 1)
ðñ (cx + d) = ´2x ðñ c = ´2, d = 0

So
ż5h
2x i
ż5
4x 1 1
dx = + ´ 2 dx
3 ( x ´ 1)( x2 + 1)
2
3 x´1 x+1 x +1
h i5
= log |x ´ 1| + log |x + 1| ´ log( x2 + 1)
3
= log 4 + log 6 ´ log 26 ´ log 2 ´ log 4 + log 10
6 ˆ 10 15
= log = log « 0.1431
26 ˆ 2 13

154
S-5: (a) Make the substitution y = sin x, dy = cos x dx and use the trig identity
cos2 x = 1 ´ sin2 x = 1 ´ y2 .

ż1 ż1
 
ż π/2 ż π/2
5 4 2 2
cos ( x ) dx = cos ( x ) cos( x ) dx = (1 ´ y ) dy = y4 ´ 2y2 + 1 dy
0 0 0 0
h y5 y3 i1 1 2 8
= ´2 +y = ´ +1 = « 0.53333
5 3 0 5 3 15

ş3 ? 2 2
2
(b) 0 9 ´ x dx is the area of the portion of the disk x + y ď 9 that lies in the first
9
quadrant. It is 14 π33 = π . Alternatively, you could also evaluate this integral using the
4
substitution x = 3 sin y, dx = 3 cos y dy.

ż3a ż π/2 b ż π/2

2
9 ´ x dx = 2
9 ´ 9 sin y (3 cos y) dy = 9 cos2 y dy
0 0 0
9h sin(2y) iπ/2
ż π/2
9
= [1 + cos(2y)] dy = y+
2 0 2 2 0
9
= π
4

2x
(c) Integrate by parts, using u = log(1 + x2 ) and dv = dx, so that du = 1+ x 2
, v = x.

ż1 ˇ1 ż 1 ż1
2 2 2x x2
log(1 + x ) dx = x log(1 + x )ˇ ´ x dx = log 2 2 dx
ˇ
2
´ 2
0 0 0 1+x 0 1+x
ż1h
1 i  1
= log 2 ´ 2 1´ dx = log 2 ´ 2 x ´ arctan x
0 1 + x2 0
π
= log 2 ´ 2 + « 0.264
2

(d) Use partial fractions.

x a b c a( x ´ 2) + b( x ´ 1)( x ´ 2) + c( x ´ 1)2
= + + =
( x ´ 1)2 ( x ´ 2) ( x ´ 1)2 x ´ 1 x ´ 2 ( x ´ 1)2 ( x ´ 2)
ðñ a( x ´ 2) + b( x ´ 1)( x ´ 2) + c( x ´ 1)2 = x

Setting x = 1 gives ´a = 1. Setting x = 2 gives c = 2. Substituting in a = ´1 and c = 2

gives

b( x ´ 1)( x ´ 2) = x + ( x ´ 2) ´ 2( x ´ 1)2 = ´2x2 + 6x ´ 4 = ´2( x ´ 1)( x ´ 2)

ùñ b = ´2

155
Hence
żMh
2 i
ż8
x 1 2
dx = lim ´ ´ + dx
3 ( x ´ 1)2 ( x ´ 2) MÑ8 3 ( x ´ 1)2 x ´ 1 x ´ 2
h 1 iM
= lim ´ 2 log |x ´ 1| + 2 log |x ´ 2|
MÑ8 x ´ 1 3
 ˇ ˇ M
1 ˇx ´ 2ˇ
= lim + 2 log ˇˇ ˇ
MÑ8 x ´ 1 x ´ 1ˇ 3
 ˇ ˇ  ˇ ˇ
1 ˇ M ´ 2ˇ 1 ˇ3 ´ 2ˇ
= lim + 2 log ˇˇ ˇ ´ + 2 log ˇˇ ˇ
MÑ8 M ´ 1 M ´ 1ˇ 3´1 3 ´ 1ˇ
1
= 2 log 2 ´ « 0.886
2
since
M´2 1 ´ 2/M
lim log = lim log = log 1 = 0
MÑ8 M´1 MÑ8 1 ´ 1/M

S-6: (a) Make the substitution u = sin(2x ), du = 2 cos(2x ) dx.

π π ż1
   
ż ż
4
2 3
4
2 2 1
sin (2x ) cos (2x ) dx = sin (2x ) 1 ´ sin (2x ) cos(2x ) dx = u2 1 ´ u2 du
0 0 2 0
1
ż1

h
1 1 3 1 5 i 1
1
= u2 ´ u4 du = u ´ u =
2 0 2 3 5 0 15

(b) Make the substitution x = 3 tan t, dx = 3 sec2 t dt and use the trig identity
9 + 9 tan2 t = 9 sec2 t.

3
3
´3
ż ż ż
2 ´2 2 ´2 2
9+x dx = 9 + 9 tan t 3 sec t dt = 3 sec t 3 sec2 t dt
ż
1 1 1 x
= cos t dt = sin t + C = ? +C
9 9 9 x2 + 9

To convert back to x, in the last step, we used the triangle below, which is rigged to have
tan t = 3x .

√
x2 + 9 x
t
3
(c) We use partial fractions.

1 a bx + c a( x2 + 1) + (bx + c)( x ´ 1)
= + =
( x ´ 1)( x2 + 1) x ´ 1 x2 + 1 ( x ´ 1)( x2 + 1)

156
provided a( x2 + 1) + (bx + c)( x ´ 1) = 1 for all x. Setting x = 1 gives 2a = 1 or a = 21 .
Substituting in a = 12 gives
1 2
( x + 1) + (bx + c)( x ´ 1) = 1
2
1 1
ðñ (bx + c)( x ´ 1) = (1 ´ x2 ) = ´ ( x ´ 1)( x + 1)
2 2
1 1
ðñ (bx + c) = ´ ( x + 1) ðñ b = c = ´
2 2

So
ż h
1 x+1 i
ż h
1 1 i
ż
dx 1 1 1 1 1 2x
= ´ dx = ´ ´ dx
( x ´ 1)( x2 + 1) 2 x ´ 1 2 x2 + 1 2 x ´ 1 4 x2 + 1 2 x2 + 1
1 1 1
= log |x ´ 1| ´ log( x2 + 1) ´ tan´1 x + C
2 4 2

(d) Integrate by parts with u = tan´1 x and dv = x dx so that du = 1+1x2 dx and v = 21 x2 .

Then
x2
ż ż
´1 1 2 ´1 1
x tan x dx = x tan x ´ dx
2 2 1 + x2
1 1 + x2
ż ż
1 2 ´1 1 1
= x tan x ´ 2
dx + dx
2 2 1+x 2 1 + x2
1 
= x2 tan´1 x ´ x + tan´1 x + C
2

π


S-7: (a) Substituting y = sin(2x ), dy = 2 cos(2x ) dx, y( x = 0) = 0 and y x = 4 = 1,
ż π/4 ż1 ˇ1
5 5 dy 1 1 6 ˇˇ 1
sin (2x ) cos(2x ) dx = y = y ˇ =
0 0 2 26 0 12

(b) Substituting x = 2 sin y, dx = 2 cos y dy,

 
ż a ż b ż ż
2 2 2
4 ´ x dx = 4 ´ 4 sin y 2 cos y dy = 4 cos y dy = 2 1 + cos(2y) dy

= 2y + sin(2y) + C = 2y + 2 sin y cos y + C

c
x x2
= 2 sin´1 + x 1 ´ +C
2 4
b b
x 2
since sin y = 2 and cos y = 1 ´ sin y = 1 ´ x4 .
2

2x
(c) Integrate by parts, using u = log(1 + x2 ), dv = dx, v = x and du = 1+ x2
dx
żh
2x2 2 i
ż ż
2 2 2
log(1 + x ) dx = x log(1 + x ) ´ dx = x log ( 1 + x ) ´ 2 ´ dx
1 + x2 1 + x2
= x log(1 + x2 ) ´ 2x + 2 tan´1 x + C

157
(d) The partial fractions expansion
x+1 A B C Ax ( x ´ 1) + B( x ´ 1) + Cx2
= + + =
x 2 ( x ´ 1) x x2 x ´ 1 x 2 ( x ´ 1)
is true provided
Ax ( x ´ 1) + B( x ´ 1) + Cx2 = x + 1
for all x. Setting x = 1, gives the requirement C = 2. Setting x = 0, gives the requirement
B = ´1. As well, the net coefficient of x2 on the left hand side, namely A + C, must be
the same as the net coefficient of x2 on the right hand side, namely 0. So A + C = 0 and
A = ´2. Checking,
´2x ( x ´ 1) ´ ( x ´ 1) + 2x2 = ´2x2 + 2x ´ x + 1 + 2x2 = x + 1
as desired. Thus
ż h
2 i
ż
x+1 2 1 1
2
dx = ´ ´ 2
+ dx = ´2 log |x| + + 2 log |x ´ 1| + C
x ( x ´ 1) x x x´1 x

S-8: (a) Define

ż8 ż8
´x
I1 = e sin(2x ) dx I2 = e´x cos(2x ) dx
0 0

Integrating by parts, with u = sin(2x ) or cos(2x ) and dv = e´x dx. That is, v = ´e´x .
ż8 żR
´x
I1 = e sin(2x ) dx = lim e´x sin(2x ) dx
0 RÑ8 0
 ˇR żR 
´x ´x
= lim ´ e sin(2x )ˇ + 2 e cos(2x ) dx = 2I2
ˇ
RÑ8 0 0
ż8 żR
I2 = e´x cos(2x ) dx = lim e´x cos(2x ) dx
0 RÑ8 0
 ˇ8 ż8 
´x ´x
= lim ´e cos(2x )ˇ ´ 2 e sin(2x ) dx = 1 ´ 2I1
ˇ
RÑ8 0 0
ş8
Substituting I2 = 21 I1 into I2 = 1 ´ 2I1 gives 52 I1 = 1 or 0 e´x sin(2x ) dx = 52 .
? ?
(b) Subsitute x = 2 tan y, dx = 2 sec2 y dy.
ż ?2
sec2 y
ˇπ/4
1 ? ż π/4
1
ż π/4
1 ˇ 1
dx = 2 dy = cos y dy = sin yˇˇ = ?
0 (2 + x2 )3/2 0 (2 + 2 tan2 y)3/2 2 0 2 0 2 2

2x x2
(c) Integrate by parts, using u = log(1 + x2 ) and dv = x dx, so that du = 1+ x 2
, v= 2.
ż1h
x i
ż1 ˇ1 ż 1 x 3
2 1 2 2 ˇ 1
x log(1 + x ) dx = x log(1 + x )ˇ ´ 2
dy = log 2 ´ x´ dx
0 2 0 0 1+x 2 0 1 + x2
1 h x2 1 i1 1
= log 2 ´ ´ log(1 + x2 ) = log 2 ´ « 0.193
2 2 2 0 2

158
(c) (Second solution) First substitute y = 1 + x2 , dy = 2x dx.
ż1
1 2
ż
2
x log(1 + x ) dx = log y dy
0 2 1

Then integrate by parts, using u = log y and dv = dy, so that du = y1 , v = y.

ż1 ż2 ˇ2 1 ż 2 1
2 1 1 1
x log(1 + x ) dx = log y dy = y log yˇ ´ y dy = log 2 ´ « 0.193
ˇ
0 2 1 2 1 2 1 y 2

(d) Use partial fractions.

1 a b c a( x ´ 2) + b( x ´ 1)( x ´ 2) + c( x ´ 1)2
= + + =
( x ´ 1)2 ( x ´ 2) ( x ´ 1)2 x ´ 1 x ´ 2 ( x ´ 1)2 ( x ´ 2)
ðñ a( x ´ 2) + b( x ´ 1)( x ´ 2) + c( x ´ 1)2 = 1

Setting x = 1 gives ´a = 1. Setting x = 2 gives c = 1. Substituting in a = ´1 and c = 1

gives

b( x ´ 1)( x ´ 2) = 1 + ( x ´ 2) ´ ( x ´ 1)2 = ´x2 + 3x ´ 2 = ´( x ´ 1)( x ´ 2) ùñ b = ´1

Hence
żMh
1 i
ż8
x 1 1
dx = lim ´ ´ + dx
3 ( x ´ 1)2 ( x ´ 2) MÑ8 3 ( x ´ 1)2 x ´ 1 x ´ 2
h 1 iM
= lim ´ log( x ´ 1) + log( x ´ 2)
MÑ8 x ´ 1 3
h 1 M ´ 2i h 1 3 ´ 2i
= lim + log ´ + log
MÑ8 M ´ 1 M´1 3´1 3´1
1
= log 2 ´ « 0.193
2
since
M´2 1 ´ 2/M
lim log = lim log = log 1 = 0
MÑ8 M´1 MÑ8 1 ´ 1/M

S-9: (a) Integrate by parts with u = log x and dv = x dx, so that du = dx 1 2

x and v = 2 x .
ż ż
1 2 1 1 1 1
x log x dx = x log x ´ x2 dx = x2 log x ´ x2 + C
2 2 x 2 4

(b)
ż ż ż ż
( x ´ 1)dx x´1 x+2 1
= dx = dx ´ 3 dx
x2 + 4x + 5 ( x + 2)2 + 1 ( x + 2)2 + 1 ( x + 2)2 + 1
1
= log[( x + 2)2 + 1] ´ 3 arctan( x + 2) + C
2

159
(c) We use partial fractions.
1 1 a b a ( x ´ 1) + b ( x ´ 3)
= = + =
x2 ´ 4x + 3 ( x ´ 3)( x ´ 1) x´3 x´1 ( x ´ 3)( x ´ 1)
provided a( x ´ 1) + b( x ´ 3) = 1. Setting x = 3 gives a = 21 . Setting x = 1 gives b = ´ 21 .
So
ż h
1 1 i
ż
dx 1 1 1 1
2
= ´ dx = log |x ´ 3| ´ log |x ´ 1| + C
x ´ 4x + 3 2x´3 2x´1 2 2

(d) Substitute y = x3 , dy = 3x2 dx.

ż 2 ż
x dx 1 dy 1 1
6
= 2
= arctan y + C = arctan x3 + C
1+x 3 1+y 3 3

dx
S-10: (a) Integrate by parts with u = tan´1 x, dv = dx, du = 1+ x 2
and v = x. This gives
ż1
ˇ1
ż1
x h1 i1 π 1
´1 ´1 ´1 2
tan x dx = x tan xˇ ´ dx = tan 1 ´ log ( 1 + x ) = ´ log 2
0
0
0 1 + x2 2 0 4 2

(b) Note that the derivative of the denominator is 2x ´ 2, which differs from the
numerator only by 1. So
ż ż ż
2x ´ 1 2x ´ 2 1
2
dx = 2
dx + 2
dx
x ´ 2x + 5 x ´ 2x + 5 x ´ 2x + 5
ż ż
2x ´ 2 1
= dx + dx
x2 ´ 2x + 5 ( x ´ 1)2 + 4
1 x´1
= log |x2 ´ 2x + 5| + tan´1 +C
2 2

x 2
S-11: (a) Integrating by parts with u = log x, dv = x dx, du = dx
x and v = 2

x2 x2 x2
ż ż ż
x
x log x dx = uv ´ v du = log x ´ dx = log x ´ +C
2 2 2 4

(b) Substituting u = x3 + 1, du = 3x2 dx

x2 u´100 1
ż ż
1 du 1
3 101
dx = 101
= +C = ´ +C
( x + 1) u 3 ´100 3 300( x + 1)100
3

(c) Substituting du = cos x dx, u = sin x, cos2 x = 1 ´ sin2 x = 1 ´ u2 ,

u5 u7
ż ż ż
cos x sin x dx = cos x sin x cos x dx = (1 ´ u2 )u4 du =
3 4 2 4
´ +C
5 7
sin5 x sin7 x
= ´ +C
5 7

160
(d) Substituting x = 2 sin u, dx = 2 cos u du,
ż a ż a ż ż
2 2 2 1 + cos(2u)
4 ´ x dx = 4 ´ 4 sin u 2 cos u du = 4 cos u du = 4 du
2
h 1 i h i
= 2 u + sin(2u) + C = 2 u + sin u cos u + C
2
We still need to express our answer as a function of x. Recall that sin u = 2x . So
u = arcsin 2x . Our answer also contains cos u, so we also need to express it as a function
of x. We can do this either by using the trig identity
c
a
2 x2
cos u = 1 ´ sin u = 1 ´
4
or by reading cos u off of the triangle below, which has u as an angle and whose sides
have been chosen so that sin u = 2x .

2 x
u
√
4 − x2

So
ż a
x xa
4 ´ x2 dx = 2 arcsin + 4 ´ x2 + C
2 2

S-12: (a) If the integrand had x’s instead of e x ’s it would be a rational function, ripe for
the application of partial fractions. So let’s start by making the substitution u = e x ,
du = e x dx:
ex
ż ż
du
x x
dx =
(e + 1)(e ´ 3) (u + 1)(u ´ 3)
Now, we follow the partial fractions protocol, starting with expressing

1 A B
= +
(u + 1)(u ´ 3) u+1 u´3

To find A and B, the sneaky way, we cross multiply by the denominator

1 = A ( u ´ 3) + B ( u + 1)

and find A and B by evaluating at u = ´1 and u = 3, respectively.

1
1 = A(´1 ´ 3) + B(´1 + 1) ðñ A = ´
4
1
1 = A(3 ´ 3) + B(3 + 1) ðñ B =
4

161
Finally, we can do the integral:
ż h
ex 1/4 i
ż ż
du ´1/4
dx = = + du
(e x + 1)(e x ´ 3) (u + 1)(u ´ 3) u+1 u´3
1 1
= ´ log |u + 1| + log |u ´ 3| + C
4 4
1 1
= ´ log |e x + 1| + log |e x ´ 3| + C
4 4

(b) The argument of the square root is

12 + 4x ´ x2 = 12 ´ ( x ´ 2)2 + 4 = 16 ´ ( x ´ 2)2

Hmmm. The numerator is x2 ´ 4x + 4 = ( x ´ 2)2 . So let’s make the integral look

somewhat simpler by substituting u = x ´ 2, du = dx. When x = 2 we have u = 0, and
when x = 4 we have u = 2 so
ż x =4 ż u =2
x2 ´ 4x + 4 u2
? dx = ? du
x =2 12 + 4x ´ x2 u =0 16 ´ u2
This is perfect for the trig substitution u = 4 sin θ, du = 4 cos(θ ) dθ. When u = 0 we have
4 sin θ = 0 and hence θ = 0. When u = 2 we have 4 sin θ = 2 and hence θ = π6 . So
ż u =2 ż θ =π/6
u2 16 sin2 θ
? du = a 4 cos θdθ
u =0 16 ´ u2 θ =0 16 ´ 16 sin2 θ
ż π/6
= 16 sin2 θ dθ
0
 
ż π/6
=8 1 ´ cos(2θ ) dθ
0
 π/6  ? 
1 π 1 3
= 8 θ ´ sin(2θ ) =8 ´
2 0 6 2 2
4π ?
= ´2 3
3

S-13: (a) Substituting y = cos x, dy = ´ sin x dx, sin2 x = 1 ´ cos2 x = 1 ´ y2

sin3 x sin2 x 1 ´ y2

ż ż ż ż
´3 ´1
dx = sin x dx = ( ´dy ) = ´ y ´ y dy
cos3 x cos3 x y3
y´2 1
=´ + log |y| + C = sec2 x + log | cos x| + C
´2 2

(b) Substituting x5 = 4y, 5x4 dx = 4 dy, and using that x = 2 ùñ 25 = 4y ùñ y = 8,

ż2 ż2
x4 x4 4 8 1 8 1
ż ż
1
dx = 2 dx = 2 dy = dy
10
´2 x + 16
10
0 x + 16 5 0 16y2 + 16 10 0 y2 + 1
1
= tan´1 8 « 0.1446
10

162
 2x
(c) Integrate by parts, using u = log(1 + x2 ), dv = dx, v = x and du = 1+ x 2
dx
ż1 ˇ1 ż 1 2x2
2 2
log(1 + x ) dx = x log(1 + x )ˇ ´ dx
ˇ
2
0 0 0 1+x
ż1h
2 i
= log 2 ´ 2´ dx
0 1 + x2
h i1
= log 2 ´ 2x ´ 2 tan´1 x
0
π
= log 2 ´ 2 + « 0.2639
2

S-14: (a) Split the specified integral into

ż3 a ż3a ż3 a
2
( x + 1) 9 ´ x dx = 2
9 ´ x dx + x 9 ´ x2 dx
0 0 0
?
The first piece represents the area above the x–axis and below the curve y = 9 ´ x2 , i.e.
x2 + y2 = 9, with 0 ď x ď 3. That’s the area of one quadrant of a disk of radius 3. So
ż3a
1 2 9
9 ´ x2 dx = π3 = π
0 4 4

For the second part we substitute u = 9 ´ x2 , du = ´2x dx, u( x = 0) = 9 and

u( x = 3) = 0. So

1 h u3/2 i0 1h 27 i
ż3 a ż0
? du
2
x 9 ´ x dx = u =´ =´ ´ =9
0 9 ´2 2 3/2 9 2 3/2

All together
ż3 a 9
( x + 1) 9 ´ x2 dx = π + 9
0 4

(b) The integrand is of the form N ( x )/D ( x ) with D ( x ) already factored and N ( x ) of
lower degree. We immediately look for a partial fractions decomposition:

4x + 8 A Bx + C
2
= + 2 .
( x ´ 2)( x + 4) x´2 x +4

Multiplying through by the denominator yields

4x + 8 = A( x2 + 4) + ( Bx + C )( x ´ 2) (˚)

Setting x = 2 we find:

8 + 8 = A(4 + 4) + 0 ùñ 16 = 8A ùñ A = 2

163
Substituting A = 2 in (˚) gives

4x + 8 = A( x2 + 4) + ( Bx + C )( x ´ 2)
ùñ ´2x2 + 4x = ( x ´ 2)( Bx + C )
ùñ (´2x )( x ´ 2) = ( Bx + C )( x ´ 2)
ùñ B = ´2, C = 0

So we have found that A = 2, B = ´2, and C = 0. Therefore

ż ż  
4x + 8 2 2x
dx = ´ dx
( x ´ 2)( x2 + 4) x ´ 2 x2 + 4
= 2 log |x ´ 2| ´ log( x2 + 4) + C

Here the second integral was found just by guessing an antiderivative. Alternatively, one
could use the substitution u = x2 + 4, du = 2x dx.
(c) The given integral is improper and so is
ż +8 ż R1 ż R1
1 1 e x dx
x
dx = lim x
dx = lim
´8 e + e´x R,R1 Ñ8 ´R e + e´x R,R1 Ñ8 ´R e2x + 1
We substitute u = e x , du = e x dx, giving
ż +8 ż e R1
1 du   e R1 π
x
dx = lim 2
= lim arctan u e´R
=
´8 e + e´x R,R1 Ñ8 e´R u +1 R,R1 Ñ8 2

S-15: (a) Substituting u = log x, du = 1x dx, dx = x du = eu du

ż ż
sin(log x ) dx = sin(u) eu du

We have already seen, in Example 1.7.11 of the CLP 101 notes, that


ż
1
sin(u) eu du = eu sin u ´ cos u + C
2
So
1  
ż
sin(log x ) dx = x sin(log x ) ´ cos(log x ) + C
2

(b) The integrand is of the form N ( x )/D ( x ) with N ( x ) of lower degree than D ( x ). So we
factor D ( x ) = ( x ´ 2)( x ´ 3) and look for a partial fractions decomposition:
1 A B
= + .
( x ´ 2)( x ´ 3) x´2 x´3
Multiplying through by the denominator yields

1 = A ( x ´ 3) + B ( x ´ 2)

164
Setting x = 2 we find:

1 = A(2 ´ 3) + 0 ùñ A = ´1

Setting x = 3 we find:

1 = 0 + B(3 ´ 2) ùñ B = 1

So we have found that A = ´1 and B = 1. Therefore

ż ż  
1 1 1
dx = ´ dx
( x ´ 2)( x ´ 3) x´3 x´2
= log |x ´ 3| ´ log |x ´ 2| + C

Solutions to Exercises 2.1 — Jump to TABLE OF CONTENTS

S-1: By Hooke’s Law, the force exerted by the spring at displacement x m from its natural
length is F = kx, where k is the spring constant. Measuring distance in meters and force
in newtons, the total work is
ż 0.1 m ˇ0.1 m
1 2 ˇˇ 1 1
kx dx = kx ˇ = ¨ 50 ¨ (0.1)2 J = J.
0 2 0 2 4

S-2: By definition, the work done in moving the object from x = 1 meters to x = 16
meters by the force F ( x ) is
ż 16 ż 16
a ? ˇˇx=16
W= F ( x ) dx = ? dx = 2a x ˇ = 6a
1 1 x x =1

To have W = 18, we need a = 3.

As a side remark, F ( x ) = ?a
has to have units Newtons. As x, a distance, has units
x ?
meters, a has to have the bizarre units Newtonsˆ meters.

S-3: First note that Newtons and Joules are MKS units, so we should measure distances
in meters rather than centimeters. Next recall that a (linear) spring with spring constant k
exerts a force F ( x ) = kx when the spring is stretched xm beyond its natural length. So in
this case 0.05k = 10, or k = 200. The work done is
ż 0.5 ż 0.5 h i0.5
F ( x ) dx = 200x dx = 100x2 = 25
0 0 0

165
S-4: Note that the cable has mass density 85 kg/m. When the bucket is at height y, the
y

cable that remains to be lifted has length (5 ´ y) m and mass 58 (5 ´ y) = 8 1 ´ 5 kg. So,
y

at height y, the cable is subject to a downward gravitational force of 8 1 ´ 5 ¨ 9.8; to raise
y

the cable we need to apply a compensating upward force of 8 1 ´ 5 ¨ 9.8. So the work
required is

ż5 
y  y2 
ˇ5
ˇ
8 1´ ¨ 9.8 dy = 8 y ´ ¨ 9.8ˇˇ = 8 ¨ 2.5 ¨ 9.8 = 196 J.
0 5 10 0

Alternatively, the cable has linear density 8 kg/5 m = 1.6 kg/m, and so the work
required to lift a small piece of the cable (of length ∆y) from height y m to height 5 m is
1.6∆y ¨ 9.8(5 ´ y). The total work required is therefore
ż5
1.6 ¨ 9.8(5 ´ y) dy = 1.6 ¨ 9.8 ¨ 12.5 = 196 J
0

as before.

S-5: Imagine slicing the water into horizontal pancakes as in the sketch

1m

x 3m

Denote by x the distance of a pancake below the surface of the water. So x runs from 0 to
3. Each pancake
?
˝ has radius 32 ´ x2 (by Pythagorous) and hence

˝ has cross–sectional area π (9 ´ x2 ) and hence

˝ has volume π (9 ´ x2 ) dx and hence

˝ has mass 1000π (9 ´ x2 ) dx and hence

˝ is subject to a gravitational force of 9.8 ˆ 1000π (9 ´ x2 ) dx and hence

˝ requires work 9800π (9 ´ x2 )( x + 4) dx to raise it to the spout. (It has to be raised

xm to bring it to the height of the centre of the sphere, then 3m more to bring it to
the top of the sphere, and then 1m more to bring it to the spout.)

166
So the total work is
ż3 ż3
2


9800π (9 ´ x )( x + 4) dx = 9800π 36 + 9x ´ 4x2 ´ x3 dx
0 0
h 9 4 1 i3
= 9800π 36x + x2 ´ x3 ´ x4
2 3 4 0
369
= 9800 π = 904,050π joules
4

S-6: The plate at height z has

˝ has side length 3 ´ z and hence
˝ has area (3 ´ z)2 and hence
˝ has volume (3 ´ z)2 dz and hence
˝ has mass 8000(3 ´ z)2 dz and hence
˝ is subject to a gravitational force of 9.8 ˆ 8000(3 ´ z)2 dz and hence
˝ requires work 9.8 ˆ 8000(2 + z)(3 ´ z)2 dz to raise it from 2m below ground level to
zm above ground level.
So the total work is ż3
9.8 ˆ 8000(2 + z)(3 ´ z)2 dz joules
0

Solutions to Exercises 2.2 — Jump to TABLE OF CONTENTS

S-1: By definition, the average value is

ż
1 π/2
sin(5x ) + 1 dx
π ´π/2
We now observe that sin(5x ) is an odd function, and hence its integral over the
symmetric interval [´ π2 , π2 ] equals zero. So the average value of f ( x ) on this interval is 1.
Alternatively, the average equals, by the fundamental theorem of calculus,
 π/2 "   *
1 ´ cos(5x ) 1 ´ cos(5π/2) π ´ cos(´5π/2) ´π π
+x = + ´ + = =1
π 5 ´π/2 π 5 2 5 2 π

S-2: By definition, the average is

że  3  x =e  3 
1 2 1 x x3 1 e e3 1
x log x dx = log x ´ = ´ +
e´1 1 e´1 3 9 x =1 e´1 3 9 9
 
1 2 3 1
= e +
e´1 9 9

167
 3 3
The indefinite integral x2 log x dx = x3 log x ´ x9 + C was guessed and then verified by
ş

d x3 x3
 2
checking that dx 3 log x ´ 9 = x log x. The same indefinite integral can be found by
x3
using integration by parts with u = log x, dv = x2 dx, v = 3.

S-3: By definition, the average value in question equals

ż π/2  ż π/2 ż π/2 
1 3 22 3 2
(3 cos x + 2 cos x ) dx = 3 cos x dx + 2 cos x dx
π/2 ´ 0 0 π 0 0

For the first integral we use the substitution u = sin x, du = cos x dx,
cos2 x = 1 ´ sin2 x = 1 ´ u2 . Note that the endpoints x = 0 and x = π2 become u = 0 and
u = 1, respectively.
ż π/2 ż π/2
3
3 cos x dx = 3 cos2 x cos x dx
0 0
ż1
= 3(1 ´ u2 ) du
0
ˇ1
= (3u ´ u3 )ˇ = 2.
ˇ
0

1+cos(2x )
For the second integral we use the trigonometric identity cos2 x dx = 2 .

ż π/2 ż π/2
2
2 cos x dx = 1 + cos(2x ) dx
0 0
 π/2
1 π
= x + sin(2x ) =
2 0 2

Therefore the average value in question is

 ż π/2 ż π/2   
2 3 2 2 π 4
3 cos x dx + 2 cos x dx = 2+ = + 1.
π 0 0 π 2 π

S-4: By definition, the average value in question equals

ż π/k
1
Ave = sin(kx ) dx
π/k ´ 0 0

To evaluate the integral, we use the substitution u = kx, du = k dx. Note that the
endpoints x = 0 and x = π/k become u = 0 and u = π, respectively. So

1h iπ
żπ
k du 2
Ave = sin(u) = ´ cos(u) =
π 0 k π 0 π

168
S-5: By definition, the average temperature is
1 3 1 3 80 1 3 1 3 h 10 10 i
ż ż ż ż
80
T ( x ) dx = dx = dx = + dx
3 0 3 0 16 ´ x2 3 0 (4 ´ x )(4 + x ) 3 0 4´x 4+x
1 3h 10 i 10 h i3
ż
10
= ´ + dx = ´ log |x ´ 4| + log |x + 4|
3 0 x´4 4+x 3 0
ˇ3
10 ˇ x + 4 ˇˇˇ 10
ˇ ˇ
= log ˇ ˇˇ = [log 7 ´ log 1]
3 x´4 0 3
10
= log 7
3

S-6: By definition, the average value is

że ż1 ˇ1
1 log x 1 1 u2 ˇˇ 1
dx = u du = =
e´1 1 x e´1 0 e´1 2 0ˇ 2( e ´ 1)
1
where we made the change of variables u = log x, du = x dx.

S-7: By definition, the average value is

1 2π 1 1 2π   1 1 h sin(2x ) i2π
ż ż
2 1 1 1
cos x dx = cos(2x ) + 1 dx = +x = 2π =
2π 0 2π 2 0 2π 2 2 0 2π 2 2

S-8: (a) a) Let v(t) be the speed of the car at time t. Then, by the trapezoidal rule with
a = 0, b = 2, ∆t = 1/3, the distance traveled is
ż2 h i
v(t) dt « ∆t 12 v(0) + v(1/3) + v(2/3) + v(3/3) + v(4/3) + v(5/3) + 21 v(2)
0
1h1 1
i
= 50 + 70 + 80 + 55 + 60 + 80 + 2 40 = 130 km
3 2
ş2
(b) The average speed is 12 0 v(t) dt « 65 km/hr.

Solutions to Exercises 2.3 — Jump to TABLE OF CONTENTS

S-1: We use vertical strips, as in the sketch below. (To use horizontal strips we would
have to split the domain of integration in two: ´3 ď y ď 0 and 0 ď y ď 3.)

(−1, 3)
y = −3x

(0, 0)
y = 3x
(−1, −3)

169
The equations of the top and bottom of the triangle are

y = T ( x ) = ´3x and y = B( x ) = 3x.

The area of the triangle is A = 12 (6)(1) = 3. Using vertical slices,

ż0 ż0 ż0
1  1  1
x̄ = x T ( x ) ´ B( x ) dx = x (´3x ) ´ (3x ) dx = ´ 6x2 dx
A ´1 3 ´1 3 ´1

S-2: The equation of the top of the region is y = T ( x ) = 1, the equation of the bottom of
the region is y = B( x ) = ´e x and x rus from a = 0 to b = 1. So the y-coordinate of the
centre of mass is

 1
1 1  1 1 
ż ż
2 2 2x 1 1 2x
ȳ = T ( x ) ´ B( x ) dx = 1´e dx = x´ e
2A 0 2e 0 2e 2 0
1 h e 2 1 i 3 e
= 1´ ´0+ = ´
2e 2 2 4e 4

S-3: The area of the region is

ż8 ż t   t  
8 8 4 4 4
A= dx = lim dx = lim ´ 2 = lim ´ 2 + 2 = 0 + 4
1 x3 tÑ8 1 x3 tÑ8 x 1
tÑ8 t 1

We’ll now compute ȳ twice, once with vertical strips, as in the figure in the left below,
and once with horizontal strips as in the figure on the right below.

y y
(1, 8) (1, 8)

2
y= 8 x= y 1/3
x3

x x
1 1

Vertical strips: The equation of the top of the region is y = T ( x ) = x83 and the equation of
the bottom of the region is y = B( x ) = 0. So, using vertical strips, as in the figure on the

170
left above, the y-coordinate of the centre of mass is
1 8 
ż
ȳ = T ( x )2 ´ B( x )2 dx
2A 1
ż  2
1 8 8
= dx
8 1 x3
ż t 
8
= lim 6
dx
tÑ8 1 x
 t
8
= lim ´ 5
tÑ8 5x 1
 
8 8 8
= lim ´ 5 + 5
=
tÑ8 5t 5ˆ1 5
b
8 8
Vertical strips: Since y = x3
is equivalent to x = 3
y, the equation of the right hand side
2
of the region is x = R(y) = y1/3
and the equation of the left hand side of the region is
x = L(y) = 1. The x– and y–coordinates of the point at the top of the region obeys both
x = 1 and y = x83 = 8. Thus y runs from 0 to 8. So, using horizontal strips, as in the figure
on the right above, the y-coordinate of the centre of mass is
1 8  
ż
ȳ = y R(y) ´ L(y) dy
A 0
1 8  ´1/3 
ż
= y 2y ´ 1 dy
4 0
1 8  2/3 
ż
= 2y ´ y dy
4 0
 8
1 6 5/3 y2
= y ´
4 5 2 0
   
1 6 ˆ 32 8 ˆ 8 6 8
= ´ = 8 ´1 =
4 5 2 5 5

S-4: (a) The sketch is the figure on the left below.

y y
y= √ 1 y= √ 1
16−x2 16−x2

x=2 x=2

x x

(b) The part of the region with x coordinate between x and x + dx is a strip of width dx
running from y = 0 to y = ? 1 2 . It is illustrated in red in the figure on the right above.
16´x

171
So the area of the region is

ż2 ż sin´1 1/2
1 1 1 π
A= ? dx = 4 cos t dt = sin´1 =
0 16 ´ x2 0 4 cos t 2 6
?
where we made the substitution x = 4 sin t, dx = 4 cos tdt, 16 ´ x2 = 4 cos t. On the
(red) strip with x coordinate between x and x + dx, the average value of y is ?1/2 2 . The
16´x
y–coordinate of the centroid is the weighted average of ?1/2 with the strip counted as
16´x2
1
having weight ? dx.
16´x2

ż2
1 2 1 2
ż ż
1 1/2 1 1 1
ȳ = ? ? dx = dx = dx
A 0 16 ´ x2 16 ´ x2 2A 0 16 ´ x2 2A 0 (4 ´ x )(4 + x )
1 2 h 1/8 1/8 i
ż2h
1 i
ż
1 1
= + dx = ´ dx
2A 0 4 + x 4 ´ x 16A 0 x + 4 x ´ 4
1 h i2 6  
= log |x + 4| ´ log |x ´ 4| = log 6 ´ log 2 ´ log 4 + log 4
16A 0 16π
3 log 3
=
8π

S-5: The top of the region is y = T ( x ) = cos( x ) and the bottom of the region is
y = B( x ) = sin( x ). So the area of the region is
ż π/4


ż π/4

h iπ/4
A= T ( x ) ´ B( x ) dx = cos( x ) ´ sin( x ) dx = sin( x ) + cos( x )
0
?0 0
= 2´1

and region has centroid ( x̄, ȳ) with

ż π/4 ż π/4
1 1
x̄ = x T ( x ) ´ B( x ) dx = x cos( x ) ´ sin( x ) dx
A 0 A 0
h iπ/4 π
?
1 4? 2 ´ 1
= x sin( x ) + cos x + x cos( x ) ´ sin x =
A 0 2´1



ż π/4 ż π/4
1 1
ȳ = T ( x )2 ´ B( x )2 dx = cos2 ( x ) ´ sin2 ( x ) dx
2A 0 2A 0
1
ż π/4 h
1 1 iπ/4 1
= cos(2x ) dx = sin(2x ) = ?
2A 0 2A 2 0 4( 2 ´ 1)

S-6: (a) Imagine that the plane region is a metal plate of density one unit per unit area.
Then the part of the plate with x–coordinate between x and x + dx has width dx and

172
 y
y= √ k
1+x2

x=1
x
height ? k .
So it has area, and hence weight, ? k dx. The x–coordinate of the
1+ x 2 1+ x 2
centroid is the weighted average of x or
?
k h u i2
ż1 ż2
1 k 1 k du k ? 
x̄ = x? dx = ? = = 2´1
A 0 1+x 2 A 1 u 2 2A 1/2 1 A

We made the substitution u = 1 + x2 , du = 2xdx. The average value of y on the part of

the plate with x–coordinate between x and x + dx is ? k 2 . The y–coordinate of the
2 1+ x
centroid is the weighted average of ?k or
2 1+ x 2

k2 h i
ż1 ż1
1 k k k2 1 k2 π k2 π
ȳ = ? ? dx = dx = arctan 1 ´ arctan 0 = =
A 0 2 1 + x2 1 + x2 2A 0 1 + x2 2A 2A 4 8A

(b) We have x̄ = ȳ if and only if

k ?  k2 π 8 ? 
2´1 = ùñ k = 2´1
A 8A π

S-7: (a) The sketch is the figure on the left below.

y y=x−x2 y y=x−x2

x x

(2, −2) (2, −2)

y=x2 −3x y=x2 −3x

(b) The curves cross when x2 ´ 3x = x ´ x2 ùñ 2x2 = 4x ùñ x = 0, x = 2. The

corresponding values of y are y = 0 and y = 2 ´ 22 = ´2. Using vertical strips, as in the
figure on the right above, the area is
ż2 ż2
 2 2
    2 2 16 8
( x ´ x ) ´ ( x ´ 3x ) dx = 4x ´ 2x2 dx = 2x2 ´ x3 0 = 8 ´ =
0 0 3 3 3

(c) The x–coordinate of the centroid of R, i.e. the weighted average of x over R, is

3 h 4 3 1 4 i2 3 h 32 i
ż2 ż2
3  2 2 3   
x̄ = x ( x´x ) ´ ( x ´3x ) dx = 4x2 ´ 2x3 dx = x ´ 2x = ´8
8 0 8 0 8 3 0 8 3
=1

173
S-8: By definition the x–coordinate of the centroid is
2 ) ˇ1
ş1 1 1
ˇ
1
0 x 1+ x2 dx log ( 1 + x 2 log 2 2
x̄ = ş1 = 2 1
0
= = log 2 « 0.44127
1
ˇ
2 dx tan´1 xˇ π/4 π
0 1+ x 0

S-9: By symmetry, the centroid lies on the y–axis, so x̄ = 0. We’ll use vertical strips as in
the sketch
y

3
√
y= 32 − x2
2

x
−3 −2 −1 1 2 3
−1

−2
y = −2

to compute? the y–coordinate of the centroid. The strip with x–coordinate

? x has y running
2 2 1 2 2
from ´2 to 3 ´ x . So the?average value of y on the strip is 2 [ 3 ´ x + (´2)]. The
stripe has “weight” (area) [ 32 ´ x2 ´ (´2)] dx. Thus, as the area of the region is
1 2
2 π3 + 2 ˆ 6 = 12 + 9π/2, the y–coordinate of the centroid is
ż3
1 1 a 2 a
ȳ = [ 3 ´ x2 ´ 2] [ 32 ´ x2 + 2] dx
12 + 9π/2 ´3 2
ż3 ż3
1 2 2 1
= [(3 ´ x ) ´ 4] dx = [5 ´ x2 ] dx
24 + 9π ´3 24 + 9π ´3
2
ż3
2 h x3 i
3
2
= [5 ´ x ] dx = 5x ´
24 + 9π 0 24 + 9π 3 0
2   12
= 15 ´ 9 =
24 + 9π 24 + 9π

S-10: (a) Notice that when x = 0, y = 3 and as x2 increases, y decreases until y hits zero at
x2 = 94 , i.e. at x = ˘ 23 . For x2 ą ?
9 3 3
4 , y is not even defined. So, on D, x runs from ´ 2 to + 2
and, for each x, y runs from0 to 9 ´ 4x2 . Here is a sketch of D.
y

√
y= 9 − 4x2

(−3/2, 0) (3/2, 0)
x

174
 ?
As an aside, we can rewrite y = 9 ´ 4x2 as 4x2 + y2 = 9, y ě 0, which is the top half of
the ellipse which passes through (˘a, 0) and (0, ˘b) with a = 23 and b = 3. The area of
the full ellipse is πab = 92 π. The area of D is half of that, which is 94 π. But we are told to
use an integral, so we will do so.
The area is
ż 3/2 a
Area = 9 ´ 4x2 dx
´3/2
3 3
We can evaluate this integral by substituting x = 2 sin θ, dx = 2 cos θ dθ and using
3
x=˘ ðñ sin θ = ˘1
2
So ´ π2 ď θ ď π
and
2

2 3
ż π/2 b ż π/2 a
3 3
Area = 9´4 2 sin θ cos θ dθ = 9 ´ 9 sin2 θ cos θ dθ
´π/2 2 ´π/2 2
ż π/2 ż π/2
9 9 cos(2θ ) + 1
= cos2 θ dθ = dθ
2 ´π/2 2 ´π/2 2
9 h sin(2θ ) iπ/2 9
= +θ = π
4 2 ´π/2 4
(b) The region D is symmetric about the y axis. So the centre of mass ? lies on the y axis.
9
That is, x̄ = 0. Since D has area A = 4 π, top equation y = T ( x ) = 9 ´ 4x2 and bottom
equation y = B( x ) = 0, with x running from a = ´ 23 to b = 32 ,
1 b  2 3/2   4 3/2  
ż ż ż
2 2 2
ȳ = T ( x ) ´ B( x ) dx = 9 ´ 4x dx = 9 ´ 4x2 dx
2A a 9π ´3/2 9π 0
4 h 4 i3/2 4 h 3 4 33 i 4 h 3 1i
= 9x ´ x3 = 9 ´ 3 = 9 ´9
9π 3 0 9π 2 3 2 9π 2 2
4
=
π

S-11: (a) The two curves cross at points ( x, y) that satisfy both y = x2 and y = 6 ´ x, and
hence
x2 = 6 ´ x ðñ x2 + x ´ 6 = 0 ðñ ( x + 3)( x ´ 2)
So we see that the two curves intersect at x = 2 (as well as x = ´3, which is to the left of
the y-axis and therefore irrelevant). Here is a sketch of A.
y
y =6−x
(2, 4)
A
y = x2

175
The top of A has equation y = T ( x ) = 6 ´ x, the bottom has equation y = B( x ) = x2 and
x runs from 0 to 2. So, using vertical strips,
ż2
1  
x̄ = x T ( x ) ´ B( x ) dx
area 0
ż2
  3 2
ż
1 2
= x (6 ´ x ) ´ x dx = (6x ´ x2 ´ x3 ) dx
22/3 0 22 0
 3 4 2
3 x x
= 3x2 ´ ´
22 3 4 0
3 h 8 i 3 16 8
= 12 ´ ´ 4 = =
22 3 22 3 11
and
ż2
1  
ȳ = T ( x )2 ´ B( x )2 dx
2(area) 0
ż2  2
1 1 2 4

3 (6 ´ x )3 x 5
= (6 ´ x ) ´ x dx = ´ ´
2 22/3 0 44 3 5 0
 
3 64 ´ 216 32 3 664 166
= ´ ´ = =
44 3 5 44 15 55

The integral was evaluated by guessing an antiderivative for the integrand. It could also
be evaluated as
 2
3 2
x3 x5
ż
2 4 3 2
36 ´ 12x + x ´ x dx = 36x ´ 6x + ´
44 0 44 3 5
  0
3 8 32 3 664 166
= 72 ´ 24 + ´ = =
44 3 5 44 15 55

(b) The question specifies the use of horizontal slices (as in Example 1.6.5 CLP 101 notes).
We start by converting both equations y = 6 ´ x and y = x2 into equations of the form
?
x = f (y). To do so we solve for x in both equations, yielding x = y and x = 6 ´ y.
y
(0, 6) x=6−y
(2, 4)

√
x= y

x
• We use thin horizontal strips of width dy as in the figure above.
• When we rotate about the y–axis, each strip sweeps out a thin disk

176
 – whose radius is r = 6 ´ y when 4 ď y ď 6 (see the blue strip in the figure
?
above), and whose radius is r = y when 0 ď y ď 4 (see the red strip in the
figure above) and
– whose thickness is dy and hence
– whose volume is πr2 dy = π (6 ´ y)2 dy when 4 ď y ď 6 and whose volume is
πr2 dy = πy dy when 0 ď y ď 4.
• As our bottommost strip is at y = 0 and our topmost strip is at y = 6, the total
volume is
ż4 ż6
π y dy + π (6 ´ y)2 dy
0 4

S-12: (a) Here is a sketch of the specified region, which we shall call R.
y
y = ex

R
x
(0, −1) y = −1
(1, −1)

The top of R has equation y = T ( x ) = e x , the bottom has equation y = B( x ) = ´1 and x

runs from 0 to 1. So, using vertical strips, we see that R has
ż1 ż1 ż1
   x
    1
area = T ( x ) ´ B( x ) dx = e ´ (´1) dx = e x + 1 dx = e x + x 0 = e
0 0 0

and
ż1
1  
ȳ = T ( x )2 ´ B( x )2 dx
2(area) 0
 1
1 1  2x  1 e2x
ż
= e ´ 1 dx = ´x
2e 0 2e 2 0
 
1 e2 1 e 3
= ´1´ = ´
2e 2 2 4 4e

(b) To compute the volume when R is rotated about y = ´1

• We use thin vertical strips of width dx as in the figure above.
• When we rotate about the line y = ´1, each strip sweeps out a thin disk
– whose radius is r = T ( x ) ´ B( x ) = e x + 1 and
– whose thickness is dx and hence

177
 – whose volume is πr2 dx = π (e x + 1)2 dx.
• As our leftmost strip is at x = 0 and our rightmost strip is at z = 1, the total volume
is
ż1 ż1  1
x x e2x
π 2 2x
(e + 1) dx = π (e + 2e + 1) dx = π + 2e x + x
0 0 2 0
 2   
e 1
=π + 2e + 1 ´ +2+0
2 2
 e2 3
=π + 2e ´
2 2

Solutions to Exercises 2.4 — Jump to TABLE OF CONTENTS

S-1: Rearranging, we have:

ey dy = 2x dx.

Integrating both sides:

ey = x2 + C.

Since y = log 2 when x = 0, we have

elog 2 = 02 + C
2 = C,

and therefore

ey = x2 + 2
y = log( x2 + 2)

S-2: Using separation of variables

dy xy dy x 1
= 2 ðñ = 2 dx ðñ log |y| = log(1 + x2 ) + C
dx x +1 y x +1 2
1
To satisfy y(0) = 3, we need log 3 = 2 log(1 + 0) + C or C = log 3. Thus

1 a a a
log |y| = log(1 + x2 ) + log 3 = log 1 + x2 + log 3 = log 3 1 + x2 ùñ y( x ) = 3 1 + x2
2
The other sign of y would violate y(0) = 3 and so is unacceptable.

178
S-3: The given differential equation is separable and we solve it accordingly.
y ´3
y1 = e 3 cos t ðñ e´y/3 dy = cos t dt ðñ ´3e´y/3 = sin t + C ðñ ey/3 =
C + sin t
´3
ðñ y(t) = 3 log
C + sin t
for any constant C. The solution only exists for C + sin t ă 0.

S-4: The given differential equation is separable and we solve it accordingly.

2 c
dy 2 2 xe x x2 y3 1 2 3 x2
= xe x ´log(y ) = 2 ðñ y2 dy = xe dx ðñ = e x + C1 ðñ y =
3
e +C
dx y 3 2 2

for any constant C.

S-5: The given differential equation is separable and we solve it accordingly.

dy 1 2 x2

y1 = xey ðñ = x dx ðñ ´e ´y
= x ´ C ðñ y = ´ log C ´
ey 2 2
x2
?
for any constant C. The solution only exists for C ´ 2 ą 0, i.e. for C ą 0 and |x| ă 2C.

S-6: The given differential equation is separable and we solve it accordingly.

Cross–multiplying, we rewrite the equation as

dy
y2 = e x ´ 2x
dx
y2 dy = (e x ´ 2x ) dx.

Integrating both sides, we find

1 3
y = e x ´ x2 + C
3

Setting x = 0 and y = 3, we find 13 33 = e0 ´ 02 + C and hence C = 8; therefore the

solution is
1 3
y = e x ´ x2 + 8
3
y = (3e x ´ 3x2 + 24)1/3 .

S-7: This is a separable differential equation that we solve in the usual way.

y´2 x2
ż ż
dy dy
= ´xy3 ùñ =´ x dx ùñ = ´ + C ùñ y´2 = x2 ´ 2C.
dx y3 ´2 2

179
To have y = ´ 41 when x = 0, we must choose C to obey

 1 ´2
´ = 0 ´ 2C ùñ ´2C = 16 ùñ y´2 = x2 ´ 2C = x2 + 16.
4
1
So y = f ( x ) = ´ ? . We need to take the negative square root to have f (0) = ´ 14 .
x2 +16

S-8: This is a separable differential equation that we solve in the usual way.
Cross-multiplying and integrating,

y dy = (15x2 + 4x + 3) dx
ż ż
y dy = (15x2 + 4x + 3) dx

y2
= 5x3 + 2x2 + 3x + C.
2
42
Plugging in x = 1 and y = 4 gives 2 = 5 + 2 + 3 + C, and so C = ´2. Therefore

y2
= 5x3 + 2x2 + 3x ´ 2
2 a
y = 10x3 + 4x2 + 6x ´ 4.

We must choose the positive square root since y(1) is positive.

S-9: The given differential equation is separable and we solve it accordingly.

dy x4
y1 = x3 y ðñ = x3 dx ðñ log |y| = +C
y 4

04
We are told that y = 1 when x = 0. So log 1 = + C which gives C = 0 so that
4
x4 4
log |y| = or |y( x )| =
4 e x /4 .
Since y(0) = 1 ą 0, we should drop the absolute values.
x 4 /4
(Since |y( x )| = e ě 1, the magnitude of y( x ) is always at least one. As y( x ) must be
4
continuous, it cannot change sign.) Thus y( x ) = e x /4 .

S-10: This is a separable differential equation, even if it doesn’t quite look like it. First
move the y from the left hand side to the right hand side.

dy dy
x + y = y2 ðñ x = y2 ´ y = y ( y ´ 1)
dx dx
dy h 1 1i dx
ðñ = ´ dy =
y ( y ´ 1) y´1 y x
ðñ log |y ´ 1| ´ log |y| = log |x| + C
|y ´ 1|
ðñ log = log |x| + C
|y|

180
To determine C we set x = 1 and y = ´1.

| ´ 2|
log = log |1| + C ñ C = log 2
| ´ 1|

So the solution is
|y ´ 1| |y ´ 1|
log = log |x| + log 2 = log 2|x| ðñ = 2|x|
|y| |y|
y´1
For x near 1 and y near ´1, both x and y are positive and we may drop the absolute
value signs.
y´1 1
= 2x ðñ y ´ 1 = 2xy ðñ y =
y 1 ´ 2x
As a check, we compute

dy 2 1 2x + (1 ´ 2x ) 1
x +y = x 2
+ = 2
= 2
= y2
dx (1 ´ 2x ) 1 ´ 2x (1 ´ 2x ) (1 ´ 2x )
and
1
y (1) = = ´1
1´2ˆ1

S-11: We are told that y = f ( x ) obeys the separable differential equation y1 = xy.

dy dy
= xy ðñ = x dx
dx y
ż ż
dy
ðñ = x dx
y
x2
ðñ log |y| = +C
2

To determine C we set x = 0 and y = e.

02
log e = +C ñ C = 1
2
So the solution is
x2
+1
log |y| =
2
We are told that y = f ( x ) ą 0, so may drop the absolute value signs.

x2 2
log y = + 1 ðñ y = e1+x /2
2

181
S-12: This is a separable differential equation.
dy 1 dx h1 1 i y2 x
= 2 ðñ y dy = = ´ dx ðñ = log +C
dx ( x + x )y x ( x + 1) x x+1 2 x+1
To satisfy the initial condition y(1) = 2 we must choose C to obey
22 1 1
= log + C ùñ C = 2 ´ log
2 1+1 2
So
c   c
x 1 2x
y( x ) = 2 log ´ log + 2 = 4 + 2 log
x+1 2 x+1
Note that, to satisfy y(1) = 2, we need the positive square root.

S-13: This is a separable differential equation.

a
1 + y2 ´ 4 1 sec x  b 
y = ðñ y 1 + y2 ´ 4 dy = sec x tan x dx
tan x y
 
ż b ż
2
ðñ y 1 + y ´ 4 dy = sec x tan x dx

y2 1  2 3/2
ðñ + y ´4 = sec x + C
2 3

To determine C we set x = 0 and y = 2.

22 1  2 3/2
+ 2 ´4 = sec 0 + C = 1 + C ñ C = 1
2 3
So the solution is
y2 1  2 3/2
+ y ´4 = sec x + 1
2 3

S-14: The given differential equation is separable and we solve it accordingly.

?
dP ? dP P
= ´k P ùñ ? = ´k dt ùñ = ´kt + C
dt P 1/2
At t = 0, P = 90, 000 so
a 1
2 90, 000 = ´k ˆ 0 + C ùñ C = 2 ˆ 300 = 600 ùñ P(t) = (600 ´ kt)2
4
The constant of proportionality is determined by
1 1
P(6) = 40, 000 ùñ 40, 000 = (600 ´ 6k)2 ùñ 200 = (600 ´ 6k) ùñ 600 ´ 6k = 400
4 2
200
ùñ k =
6

182

2
Subbing in the value of k, P(t) = 14 600 ´ 200
6 t so that P(t) = 10, 000 when
1 200 2 1 200 
10, 000 = 600 ´ t ùñ 600 ´ t = 100
4 6 2 6
ùñ t = 12 weeks

S-15: The given differential equation is separable and we solve it accordingly.

dv dv  k 2
ż
dv
ż
2
m = ´(mg + kv ) ùñ = ´g 1 + v ùñ k 2
= ´g dt
dt dt mg 1 + mg v
b b
k k
Substitute u = mg v, du = mg dv
c ż ż c
mg du mg
2
= ´g dt ùñ tan´1 u = ´gt + C
k 1+u k
c d 
mg ´1 k
ùñ tan v = ´gt + C
k mg
b
mg ´1
b
k


At t = 0, v = v0 so C = k tan mg v0 . At its highest point, the object has v = 0.
This happens when t obeys
c d  c d 
mg ´1 k C m ´1 k
tan 0 = ´gt + C ùñ ´gt + C = 0 ùñ t = = tan v0
k mg g kg mg

S-16: (a) The given differential equation is separable and we solve it accordingly.
ż ż
dv 2 dv dv 1
= ´k v ùñ 2 = ´k dt ùñ 2
= ´ k dt ùñ ´ = ´kt + C
dt v v v
At t = 0, v = 40 so
1 1 1 1 40
´ = ´k ˆ 0 + C ùñ C = ´ ùñ v(t) = = =
40 40 kt ´ C kt + 1/40 40kt + 1
The constant of proportionality is determined by
40 1 1
v(10) = 20 ùñ 20 = ùñ = ùñ 400k + 1 = 2
40k ˆ 10 + 1 2 400k + 1
1
ùñ k =
400

(b) Subbing in the value of k,

40 40 40 t
v(t) = = ùñ v(t) = 5 when 5 = ùñ +1 = 8
40kt + 1 t/10 + 1 t/10 + 1 10
ùñ t = 70sec

183
S-17: (a) The given differential equation is separable and we solve it accordingly.
ż h
1 i
ż
dx dx 1
= k(3 ´ x )(2 ´ x ) ðñ = kdt ðñ ´ dx = kdt
dt ( x ´ 2)( x ´ 3) x´3 x´2
ˇx ´ 3ˇ
ðñ log ˇ ˇ = kt + C
ˇ ˇ
x´2
x´3
ðñ = Dekt
x´2

where D = ˘eC . When t = 0, x = 1, forcing

1´3
= De0 ùñ D = 2
1´2
Hence
x´3 kt kt kt kt 3 ´ 4ekt
= 2e ðñ x ´ 3 = 2e ( x ´ 2) ðñ x ´ 2e x = 3 ´ 4e ðñ x (t) =
x´2 1 ´ 2ekt

(b)
3 ´ 4ekt 3e´kt ´ 4 ´4
lim x (t) = lim kt
= lim ´kt
= =2
tÑ8 tÑ8 1 ´ 2e tÑ8 e ´2 ´2

S-18: (a) The given differential equation is separable and we solve it accordingly.

dP dP dP 1h 1 1 i
= 4P ´ P2 ùñ = dt ùñ = dt ùñ + dP = dt
dt 4P ´ P2 P (4 ´ P ) 4 P 4´P
1 i
ùñ log |P| ´ log |4 ´ P| = t + C
4
 
When t = 0, P = 2, so 41 log |2| ´ log |2| = C ùñ C = 0. So

1 ˇ P ˇ
log ˇ ˇ=t
ˇ ˇ
4 4´P
P
At time t = 0, 4´P = 1 ą 0. The ratio may not change sign at any finite time, because this
could only happenˇ if atˇ some finite time P took either the value 0 or the value 4. But at
P ˇ P
this time t = 14 log ˇ 4´P would have to be infinite. So 4´P ą 0 for all time and

1 P P P
log = t ùñ log = 4t ùñ = e4t
4 4´P 4´P 4´P
ùñ P = (4 ´ P)e4t ùñ P + Pe4t = 4e4t
4e4t 4
ùñ P = 4t
=
1+e 1 + e´4t

(b) At t = 12 , P = 4
1+e´2
= 3.523,. As t Ñ 8, e´4t Ñ 0 and P Ñ 4.

184
S-19: The rate of change of speed at time t is ´kv(t)2 for some constant of proportionality
dv
k (to be determined). So v(t) obeys the differential equation = ´kv2 . This is a
dt
separable differential equation, which we can solve in the usual way.
ż
dv 2 dv dv 1
= ´kv ùñ 2 = ´kdt ùñ 2
= ´kt + c ùñ ´ = c ´ kt
dt v v v
1 1 1 400
At time t = 0, v = 400, so c = ´ 400 and v = 400 + kt or v = 400kt+1 . At time t = 1,
v = 200, so

400 1 400
200 = ùñ 400k + 1 = 2 ùñ k = ùñ v =
400k + 1 400 t+1
400
The speed is 50 when t +1 = 50 or t + 1 = 8 or t = 7 .

S-20: (a) The given differential equation is separable and we solve it accordingly.
dB dB
= (0.06 + 0.02 sin t) B ùñ = (0.06 + 0.02 sin t) dt
dt B
ùñ log B(t) = 0.06t ´ 0.02 cos t + C1
ùñ B(t) = Ce0.06t´0.02 cos t
1
for arbitrary constants C1 and C = eC ě 0. (Note that the function B(t) = 0 obeys the
1
differential equation so that C = 0 is allowed, even though it is not of the form C = eC .)
(b) We are told that

B(0) = 1000 ùñ Ce0.06ˆ0´0.02 cos 0 = Ce´0.02 = 1000 ùñ C = 1000e0.02

so that

B(2) = 1000e0.02 e0.06ˆ2´0.02 cos 2 = $1159.89

to the nearest cent. Note that cos 2 is the cosine of 2 radians.

S-21: (a) The given differential equation is separable and we could solve it accordingly. In
fact we have already done so. If we rewrite the equation in the form
dB  m
= a B´
dt a
it is of the form covered by Theorem 2.4.4 in the CLP 101 notes. So that theorem tells us
that the solution is
! m ) at m
B ( t ) = B (0) ´ e +
a a
1
In this problem we are told that a = 0.02 = 50 , so

B(t) = tB(0) ´ 50mu et/50 + 50m = t30000 ´ 50mu et/50 + 50m

185
(b) The solution of part (a) is independent of time if and only if 30000 ´ 50m = 0. So we
need
30000
m= = $600
50

S-22: By the Fundamental Theorem of Calculus

d x


ż
1
y (x) = y(t)2 ´ 3y(t) + 2 sin t dt = y( x )2 ´ 3y( x ) + 2 sin x
dx 0


So y( x ) satisfies the differential equation y1 = y2 ´ 3y + 2 sin x = (y ´ 2)(y ´ 1) sin x
and the initial equation y(0) = 3 (just substitute x = 0 into (˚)). For y ‰ 1, 2

dy dy
= (y ´ 2)(y ´ 1) sin x ðñ = sin x dx
dx (y ´ 2)(y ´ 1)
ż h
1 i
ż
1
ðñ ´ dy = sin x dx
y´2 y´1
ðñ log |y ´ 2| ´ log |y ´ 1| = ´ cos x + c
ˇ ˇ
ˇy ´ 2ˇ
ðñ ˇˇ ˇ = ec´cos x
y ´ 1ˇ

ˇ = ec´1 or ec = 1 e and ˇ y´2 ˇ = 1 e1´cos x . Observe that,

ˇ ˇ ˇ ˇ
The condition y(0) = 3 forces ˇ 3´2
3´1 2 y´1
y´2 1 1 1´cos x
ˇ y´2 ˇ 2
when x = 0, = ą 0. Furthermore e
y´1 2 2 , and hence ˇ y´1
ˇ, can never take the
value zero. As y( x ) varies continuously with x, y( x ) must remain larger than 2.
y´2
Consquently, y´1 remains positive and we may drop the absolute value signs. Hence

y´2 1
= e1´cos x
y´1 2

Solving for y,

y´2 1

= e1´cos x ðñ 2(y ´ 2) = e1´cos x (y ´ 1) ðñ y 2 ´ e1´cos x = 4 ´ e1´cos x
y´1 2
4 ´ e1´cos x
ðñ y =
2 ´ e1´cos x
To avoid division by zero in the last step, we need

e1´cos x ă 2 ðñ 1 ´ cos x ă log 2 ðñ cos x ą 1 ´ log 2

The largest allowed interval is ´ cos´1 (1 ´ log 2) ă x ă cos´1 (1 ´ log 2) « 1.259. As x

approachs the end points of this interval, e1´cos x approachs 2 and y approachs infinity.

S-23: Suppose that in a very short time interval dt, the height of water in the tank changes
by dh (which is negative). Then in this time interval the amount of the water in the tank

186
decreases by ´π (3)2 dh. This must be the same as the amount of water a that flows through
the hole in this time interval, which is π (0.01)2 v(t) dt = π (0.01)2 2gh(t) dt. Thus
b
dh  0.01 2 a
2 2
´π (3) dh = π (0.1) 2gh(t) dt ðñ 1/2 = ´ 2g dt
h 3
?  0.01 2 a
ðñ 2 h = ´ 2g t + C
3
?
At time 0, the height is 6, so C = 2 6 and
?  0.01 2 a ?
2 h=´ 2g t + 2 6
3
The height drops to zero at time t obeying
 0.01 2 a  3 2 3
d d
? 3
0=´ 2g t + 2 6 ðñ t = 2 = 180, 000 « 99, 591sec « 27.66hr
3 0.01 g g

S-24: Suppose that at time t, the mercury in the tank has height h, which is between 0 and
12 feet. At that time, the top surface of the mercury forms a circular disk of radius

h−6 6

a
62 ´ (h ´ 6)2 . Now suppose that in a very short time interval dt, the height of mercury
in the tank changes by dh (which is negative). Then in this time interval the amount of
a
2
the mercury in the tank decreases by ´π 62 ´ (h ´ 6)2 dh. (That’s the volume of the
red disk in the figure above.) This must be the same as the amount of mercury that flows


1 2

a
1 2
through the hole in this time interval, which is, π 12 v dt = π 12 2gh dt. Thus
b
2  1 2 a
1 a
´π 2
6 ´ ( h ´ 6) 2 dh = π 2gh dt ðñ h2 ´ 12h dh = 2gh dt
12 144

1 a
ðñ h3/2 ´ 12h1/2 dh = 2g dt
144
h5/2 h3/2 1 a
ðñ ´ 12 = 2g t + C
5/2 3/2 144
5/2 3/2

At time 0, the height is 12, so C = 125/2 ´ 12 123/2 = 125/2 52 ´ 23 = ´ 15 4
125/2 and
h5/2 h3/2 1 a 4
´ 12 = 2g t ´ 125/2
5/2 3/2 144 15
The height drops to zero at time t obeying
d d
1 a 4 5/2 4 ˆ 144 125 124416
0= 2g t ´ 12 ðñ t = = 38.4 « 2, 394 sec « 0.665 hr
144 15 15 2g g

187
S-25: (a) Setting x = 0 gives
ż0



f (0) = 3 + f (t) ´ 1 f (t) ´ 2 dt = 3
0

(b) By the fundamental theorem of calculus

d x




ż
1
f (x) = f (t) ´ 1 f (t) ´ 2 dt = f ( x ) ´ 1 f ( x ) ´ 2
dx 0

Thus y = f ( x ) obeys the differential equation y1 = (y ´ 1)(y ´ 2).

(c) If y ‰ 1, 2,
ż h
1 i
ż
dy dy 1
= (y ´ 1)(y ´ 2) ðñ = dx ðñ ´ dy = dx
dx (y ´ 1)(y ´ 2) y´2 y´1
ðñ log |y ´ 2| ´ log |y ´ 1| = x + C
dy
Observe that dx = (y ´ 1)(y ´ 2) ą 0 for all y ě 2. That is, f ( x ) is increasing at all x for
which f ( x ) ą 2. As f (0) = 3, f ( x ) increases for all x ě 0 and f ( x ) ě 3 for all x ě 0. So we
may drop the absolute value signs.

f (x) ´ 2 f (x) ´ 2
log = x + C ðñ = eC e x
f (x) ´ 1 f (x) ´ 1
f ( x )´2 1 1
At x = 0, f ( x )´1
= 2 so eC = 2 and

f (x) ´ 2 1
= e x ðñ 2 f ( x ) ´ 4 = [ f ( x ) ´ 1]e x ðñ [2 ´ e x ] f ( x ) = 4 ´ e x
f (x) ´ 1 2
4 ´ ex
ðñ f ( x ) =
2 ´ ex

S-26: Suppose that at time t (measured in hours starting at noon), the water in the tank
has height y, which is between 0 and 2m. At that time, the top surface of the water forms
a circular disk of radius r = y p and area A(y) = πy2p . Thus, by Torricelli’s law,

dy ? π 1
πy2p = ´c y ðñ ´ y2p´ 2 dy = dt
dt c
1
π y2p+ 2
ðñ ´ +d = t
c 2p + 12
1
π 22p+ 2
At time 0, the height is 2, so d = c 2p+ 1 and
2

 1 1 
π 22p+ 2 y2p+ 2
t= ´
c 2p + 12 2p + 12

188
The time at which the height is 1 is obtained by subbing y = 1 into this formula and the
time at which the height is 0 is obtained by subbing y = 0 into this formula. Thus the
condition that the top half (y = 2 to y = 1) takes exactly the same amount of time to
drain as the bottom half (y = 1 to y = 0) is
 1 1   1 1   1 1 
π 22p+ 2 12p+ 2 π 22p+ 2 02p+ 2 π 22p+ 2 12p+ 2
´ = ´ ´ ´
c 2p + 12 2p + 12 c 2p + 12 2p + 21 c 2p + 12 2p + 21
or
1 1
1 1
1 1

22p+ 2 ´ 12p+ 2 = 22p+ 2 ´ 02p+ 2 ´ 22p+ 2 ´ 12p+ 2

or
1 1 1
22p+ 2 = 2 ùñ 2p + = 1 ùñ p =
2 4

Solutions to Exercises 3.1 — Jump to TABLE OF CONTENTS

ˇ ˇ
S-1: Since ˇ sin3 kˇ ď 1 and (k + 1)! = (k + 1)k!,

k! sin3 k sin3 k
lim ak = lim = lim =0
kÑ8 kÑ8 ( k + 1) ! kÑ8 k + 1

1
S-2: As n Ñ 8 we have n Ñ 0 and hence sin n1 Ñ 0. So the sequence (´1)n sin n1
converges to 0.

S-3:
h 6n2 + 5n i 6 + n5 6+0
2
lim 2
+ 3 cos ( 1/n ) = lim 1
+ 3 lim cos(1/n2 ) = + 3 cos(0)
nÑ8 n +1 nÑ8 1 +
2
nÑ8 1+0
n
=9

1
S-4: Write n = x. Then
 1  1  1
log sin + log(2n) = log sin + log(n) + log(2) = log n sin + log(2)
n n n
 sin x 
= log(2) + log
x
Note that x Ñ 0 as n Ñ 8. Since, by l’Hôpital,
sin x cos x
lim = lim =1
xÑ0 x xÑ0 1

189
we have
!  1 ) !  sin x )
lim log sin + log(2n) = lim log(2) + log
nÑ8 n xÑ0 x
 sin x 
= log(2) + log lim
xÑ0 x
= log(2) + log(1) = log(2)

?
S-5: (a), (c) Write f ( x ) = x ´ 3 + sin ?x. We ?want to show?that f ( x ) has exactly
? one zero.
Because ´1 ď sin x ď 1, we have that 2 ď 3 + sin x ď 4 = 2. So if x ă 2, we have
f ( x ) ă 0 and if x ą 2, we have that ?
f ( x ) ą 0. As f ( x ) is continuous, it must take the value
0 at least once (for some x between 2, where f ( x ) ď 0, and 2, where f ( x ) ě 0). Since
cos x
f 1 (x) = 1 ´ ?
2 3 + sin x
and ˇ ˇ
ˇ cos x ˇ 1
ˇ 2 3 + sin x ˇ ď 2?2 ă 1
ˇ ? ˇ

we necessarily have f 1 ( x ) ą 0 for all x. That is, f ( x ) is a strictly increasing function, and
so can take the values 0 for at most one value of x.
? ?
(b) Subtracting L = 3 + sin L from an+1 = 3 + sin an gives

? ha ? i ?3 + sin a + ?3 + sin L
n
a
an+1 ´ L = 3 + sin an ´ 3 + sin L = 3 + sin an ´ 3 + sin L ? ?
3 + sin an + 3 + sin L
sin an ´ sin L
=? ?
3 + sin an + 3 + sin L
Now
ˇ ˇ an
ˇż ˇ
ˇ ˇ
ˇ sin an ´ sin Lˇ = ˇ cos t dt ˇ ď |an ´ L| since | cos t| ď 1
ˇ ˇ
L
and a ? ? ?
3 + sin an + 3 + sin L ě 2 + 2 since sin an , sin L ě ´1
so that ˇ ˇ
ˇ ˇ ˇ
ˇ a n +1 ´ L ˇ = ˇ ? sin a n ´ sin L ˇ |an ´ L|
ˇ 3 + sin a + 3 + sin L ˇ ď 2?2
? ˇ
n
1
Thus the distance from an to L decreases by a factor of at least ?
2 2
every time the index n
increases by one and ˇ ˇ
lim ˇan ´ Lˇ = 0
nÑ8

Solutions to Exercises 3.2 — Jump to TABLE OF CONTENTS

190
 1 1 1
S-1: This series is 7
+ 8 + 9 + ¨ ¨ ¨ . We recognize that this is a geometric series
8 8 8
a + ar + ar2 + ar3 + ¨ ¨ ¨ , with ratio r = 1/8 and first term a = 817 . We know this converges
to:
a 1/87 1
= =
1´r 1 ´ 1/8 7 ˆ 86

S-2: We recognize that this is a geometric series a + ar + ar2 + ar3 + ¨ ¨ ¨ , with ratio
r = 1/3 and first term a = 1. We know this converges to:

a 1 3
= =
1´r 1 ´ 1/3 2

S-3: We recognize this as a telescoping series. When we compute the nth partial sum, i.e.
the sum ˇ n terms, successive terms cancel and only
 of of the first  the firstˇ half of the first
term, k62 ´ (k+61)2 ˇ , and the second half of the nth term, k62 ´ (k+61)2 ˇ , survive.
ˇ ˇ
k =1 k=n

n  
ÿ 6 6 6 6
sn = 2
´ 2
= 2´
k ( k + 1) 1 ( n + 1)2
k =1

Therefore, we can see directly that the sequence of partial sums tsn u is convergent:
 
6 6
lim sn = lim ´ =6
nÑ8 nÑ8 12 ( n + 1)2

By definition the series is also convergent with limit 6.

S-4: The Nth partial sum is

N 
ÿ π  π   π  π   π  π 
sN = cos ´ cos = cos ´ cos + cos ´ cos
n =3
n n+1 3 4 4 5
 π  π  
+ ¨ ¨ ¨ + cos ´ cos
N N+1
π  π 
= cos ´ cos .
3 N+1
As N Ñ 8, the argument Nπ+1 converges to 0, and cos x is continuous at x = 0. By
definition, the value of the series is
 π  π  π 1
lim s N = lim cos ´ cos = cos ´ cos(0) = ´
NÑ8 NÑ8 3 N+1 3 2

191
S-5: (a) Since

sn´1 = a1 + a2 + ¨ ¨ ¨ + an´1
sn = a1 + a2 + ¨ ¨ ¨ + an´1 + an

we can find an by subtracting:

an = sn ´ sn´1
1 + 3n 1 + 3(n ´ 1) 3n + 1 3n ´ 2
= ´ = ´
5 + 4n 5 + 4(n ´ 1) 4n + 5 4n + 1
(3n + 1)(4n + 1) ´ (3n ´ 2)(4n + 5)
=
(4n + 1)(4n + 5)
11
= 2
16n + 24n + 5

(b) Since,

1 + 3n 1/n + 3 0+3 3
lim sn = lim = lim = =
nÑ8 nÑ8 5 + 4n nÑ8 5/n + 4 0+4 4

the series converges to 34 , by definition.

S-6:

3 ÿ  4 n
8 8 8
ÿ 3 ¨ 4n +1 3 ÿ n ˇˇ
n
= = r ˇ
n =2
8 ¨ 5 2 n =2
5 2 n =2
r =4/5

This is a geometric series whose first term (in this case the n = 2 term) is a = 32 r2 and
whose ratio r = 54 . So
8
ÿ 3 ¨ 4n +1 3 r2 ˇˇ 3 16/25 24
n
= ˇ = =
n =2
8¨5 2 1 ´ r r=4/5 2 1/5 5

3 3
+ 1034 + ¨ ¨ ¨ = 1
+ 1032
ř8 ´n
S-7: The number is 0.2 + 100 + 1000 5 n=0 10 . The geometric series
sums to
3 1 3 1
2 1
= =
10 1 ´ 10 10(10 ´ 1) 30

so the fraction is
1 1 7
+ =
5 30 30

192
 8
65 65 65 65 ř
S-8: The number is 2 + 102
+ 10 4 + 106 + ¨ ¨ ¨ = 2 + 102 10´2n . The geometric series
n =0
sums to
65 1 65
2 1
=
10 1 ´ 100 99
so the fraction is
65 263
2+ =
99 99

S-9: The number

8
321 321 321 ÿ
0.321 = 0.321321321 . . . = + 6 + 9 +¨¨¨ = 321 ˆ 10´3n
1000 10 10 n =1
321 1
This is a geometric series with first term a = 1000 and ratio r = 1000 and so sums to
a 321 1 321 1 321 107
= 1
= 999
= =
1´r 1000 1 ´ 1000 1000 1000 999 333

S-10: We split the sum into two parts. The first part is a geometric series with first term
2+1
a = 232 = 89 and ratio r = 23 .
8
ÿ 2n +1 a 8/9 8
n
= = =
n =2
3 1´r 1 ´ 2/3 3
The rest telescopes.
n =2
hkkkkikkkkj n =3
hkkkkikkkkj n =4
hkkkkikkkkj
8        
ÿ 1 1 1 1 1 1 1 1 1
´ = ´ + ´ + ´ +¨¨¨ =
n =2
2n ´ 1 2n + 1 3 5 5 7 7 9 3
So
8  n +1 
ÿ 2 1 1 8 1
n
+ ´ = + =3
n =2
3 2n ´ 1 2n + 1 3 3

S-11: We split the sum into two parts.

8  n  2 n´1  8  n 8 
ÿ 1 ÿ 1 ÿ 2 n´1
+ ´ = + ´
3 5 3 5
n =1 n =1 n =1

n =1
Both are geometric series. The first has first term a = 13 = 13 and ratio 31 . The second

n´1 ˇ
ˇ
has first term a = ´ 25 n =1
= 1 and ratio ´ 25 . So
8  n  2 n´1 
ÿ 1 1/3 1 1 5 17
+ ´ = +
= + =
3 5 1 ´ 1/3 1 ´ ´ 2/5 2 7 14
n =1

193
S-12: We split the sum into two parts.

8 8 8
ÿ 1 + 3n +1 ÿ 1 ÿ 3n +1
= +
n =0
4n n =0
4n n =0 4n
ˇ
1ˇ
Both are geometric series. The first has first term a = = 1 and ratio r = 41 . The
4n ˇ
ˇ n =0
n +1 ˇ
second has first term a = 34n ˇ = 3 and ratio r = 34 . So
n =0

8
ÿ 1 + 3n +1 1 3 4 1
n
= + = + 12 = 13
n =0
4 1 ´ 1/4 1 ´ 3/4 3 3

Solutions to Exercises 3.3 — Jump to TABLE OF CONTENTS

S-1: The limit

n2 1 1
lim ? = lim ? = ‰0
nÑ8 3n2 + n nÑ8 3 + 1/ ( n n ) 3

is nonzero, so the series diverges by the divergence test.

S-2: When n is very large, the term 2n dominates the numerator, and the term 3n
n
dominates the denominator. So when n is very large an « 32n . Therefore we should take
2n
bn = n . Note that, with this choice of bn ,
3

an 2n + n 3n 1 + n/2n
lim = lim n = lim =1
nÑ8 bn nÑ8 3 + 1 2n nÑ8 1 + 1/3n

as desired.

8
ř 1
S-3: (a) In general false. The harmonic series n diverges by the p–test with p = 1.
n =1

(b) Be careful. You were not told that the an ’s are positive. So this is false in general. If
8 8
an = (´1)n n1 , then (´1)n an is again the harmonic series 1
ř ř
n , which diverges.
n =1 n =1

(c) In general false. Take, for example, an = 0 and bn = 1.

S-4: This precise question was asked on a 2014 final exam. Note that the nth term in the
k
series is an = 4k5+3k and does not depend on n! There are two possibilities. Either this was
intentional (and the instructor was being particularly nasty) or it was a typo and the

194
 5n
intention was to have an = 4n +3n . In both cases, the limit

5k 5k
lim an = lim = ‰0
nÑ8 nÑ8 4k + 3k 4k + 3k
5n (5/4)n
lim an = lim n = lim = +8 ‰ 0
nÑ8 nÑ8 4 + 3n nÑ8 1 + (3/4)n

is nonzero, so the series diverges by the divergence test.

1
S-5: Let f ( x ) = . Then f ( x ) is positive and decreases as x increases. So, by the
x + 12
integral test, which is Theorem
ş8 1 3.3.5 in the CLP 101 notes, the given series converges if
and only if the integral 0 1 dx converges. Since
x+ 2

ż8
1
żR
1
"
1 ˇˇ 
ˇx= R *
1
dx = lim 1
dx = lim log x +
0 x+ 2
RÑ8 0 x+ 2
RÑ8 2 ˇ x =0
"  1 1
*
= lim log R + ´ log
RÑ8 2 2

diverges, the series diverges.

5
S-6: Let f ( x ) = . Then f ( x ) is positive and decreases as x increases. So the
x (log x )3/2
8
ÿ ż8
sum f (n) and the integral f ( x ) dx either both converge or both diverge, by the
3 3
integral test, which is Theorem 3.3.5 in the CLP 101 notes. For the integral, we use the
substitution u = log x, du = dx
x to get
ż8 ż8
5 dx 5 du
=
3 x (log x )3/2 log 3 u3/2

which converges by the p–test (which is Example 1.12.8 in the CLP 101 notes) with
p = 23 ą 1.

S-7: Since
ż8 żR ż log R
1 1 dx 1 dx
dx = lim = lim du with u = log x, du =
2 x (log x ) p RÑ8 2
p
(log x ) x RÑ8 log 2 up x

converges if and only if p ą 1, the same is true for the series, by the integral test, which is
Theorem 3.3.5 in the CLP 101 notes.

195
 ?
e´ x
S-8: Set f ( x ) = ?
x
. For x ě 1, this function is positive and decreasing (since it is the
?
product of the two positive decreasing functions e´ x and ?1x ). We use the integral test
?
with this finction. Using the substitution u = x, so that du = 2?1 x dx, we see that

ż8 żR ?
x 
e´
f ( x ) dx = lim ? dx
1 RÑ8 1 x
 ż ?R 
´u
= lim e ¨ 2 du
RÑ8 1
 ˇ? R 
´u ˇ
= lim ´2e ˇ
RÑ8 1
 ? ? 
´ R ´ 1
= lim ´2e + 2e = 0 + 2e´1 ,
RÑ8

and so this improper integral converges. By the integral test, the given series also
converges.

S-9: We first develop some intuition. For very large n, 3n2 dominates 7 so that
? ? ?
3n2 ´ 7 3n2 3
3
« 3
= 2
n n n
8
ÿ 1
The series 2
converges by the p–test with p = 2, so we expect the given series to
n =2
n
converge too.
To verify that our intuition is correct, it suffices to observe that
? ? ?
3n2 ´ 7 3n2 3
0 ă an = 3
ă 3
= 2 = cn
n n n
8
ř 8
ř
for all n ě 2. As the series cn converges, the comparison test says that an
n =2 n =2
converges too.

S-10: We first
? some intuition. For very large k, k4 dominates 1 so that the
develop ?
3 4 3 4 4/3 5
numerator
? ? k + 1 « k = k , and k dominates 9 so that the denominator
k5 + 9 « k5 = k5/2 and the summand
?
3 4
k +1 k4/3 1
? « 5/2 = 7/6
k5 + 9 k k
8
ÿ 1 7
The series converges by the p–test with p = 6 ą 1, so we expect the given series
n =1
k7/6
to converge too.

196
To verify that our intuition is correct, we apply the limit comparison test with
?
3 4
k +1 1 k4/3
ak = ? and bk = =
k5 + 9 k7/6 k5/2

which is valid since

?
3 4
a
3
ak k + 1/k4/3 1 + 1/k4
lim = lim ? = lim a =1
kÑ8 bk kÑ8 k5 + 9/k5/2 kÑ8 1 + 9/k5
8
ř 7
exists. Since the series bk is a convergent p–series (with ratio p = 6 ą 1), the given
k =1
series converges.

S-11: We first develop some intuition. For very large n, 2n dominates 7 so that

n4 2n/3 n4 2n/3 1
4
« 4
= 2n/3
(2n + 7) (2n) 16
8
ÿ
The series 2n/3 is a geometric series with ratio r = 21/3 ą 1 and so diverges. We
n =1
expect the given series to diverge too.
To verify that our intuition is correct, we apply the limit comparison test with

n4 2n/3
an = and bn = 2n/3
(2n + 7)4

which is valid since

an n4 1 1
lim = lim 4
= lim 4
= 4
nÑ8 bn nÑ8 (2n + 7) nÑ8 (2 + 7/n ) 2

8
ř
exists and is nonzero. Since the series bn is a divergent geometric series (with ratio
n =1
r = 21/3 ą 1), the given series diverges.

(It is possible to use the plain comparison test as well. One needs to show something like
4 n/3 n4 2n/3
an = (n2n2+7)4 ě (2n +7n)4
= 914 bn .)

Alternatively, one could show

2n/3
lim an = lim =8
nÑ8 nÑ8 (2 + 7/n )4

directly, and thus conclude that the given series diverges by the divergence test.

197
Alternatively, one can apply the ratio test:
ˇ (n + 1)4 2(n+1)/3 /(2(n + 1) + 7)4 ˇ
ˇ ˇ ˇ ˇ
ˇ a n +1 ˇ
lim ˇ ˇ = lim ˇ ˇ
nÑ8 ˇ an ˇ nÑ8 ˇ n4 2n/3 /(2n + 7)4 ˇ
(n + 1)4 (2n + 7)4 2(n+1)/3
= lim
nÑ8 n4 (2n + 9)4 2n/3
(1 + 1/n)4 (2 + 7/n)4 1/3
= lim 4
¨2 = 1 ¨ 21/3 ą 1.
nÑ8 (2 + 9/n)

S-12: We first develop some intuition. For very large n, n2 dominates sin n and n6
dominates n2 so that
n2 ´ sin n n2 1
6 2
« 6
= 4
n +n n n
8
ÿ 1
The series 4
converges by the p–test with p = 4 ą 1. We expect the given series to
n =1
n
converge too.
To verify that our intuition is correct, we apply the limit comparison test with
n2 ´ sin n 1
an = and bn =
n6 + n2 n4
which is valid since
an (n2 ´ sin n) n4 n6 ´ n4 sin n 1 ´ n´2 sin n
lim = lim = lim = lim =1
nÑ8 bn nÑ8 n6 + n2 1 nÑ8 n6 + n2 nÑ8 1 + n´4
8
ř
exists and is nonzero. Since the series bn converges, the given series converges
n =1
absolutely.

S-13: You might think that this series converges by the alternating series test. But you
would be wrong. The problem is that the nth term does not converge to zero as n Ñ 8, so
that the series actually diverges by the divergence test. To verify that the nth term does
(2n)!
not converge to zero as n Ñ 8 let’s write an = (n2 +1)(n!)2 (i.e. an is the nth term without
the sign) and check to see whether an+1 is bigger than or smaller than an .
a n +1 (2n + 2)! (n2 + 1)(n!)2 (2n + 2)(2n + 1) n2 + 1
= =
an ((n + 1)2 + 1)((n + 1)!)2 (2n)! ( n + 1)2 ( n + 1)2 + 1
2(2n + 1) 1 + 1/n2 1 + 1/2n 1 + 1/n2
= =4
(n + 1) (1 + 1/n)2 + 1/n2 1 + 1/n (1 + 1/n)2 + 1/n2
So
a n +1
lim =4
nÑ8 an
and, in particular, for large n, an+1 ą an . Thus, for large n, an increases with n and so
cannot converge to 0. So the series diverges by the divergence test.

198
S-14: This series converges by the alternating series test. It also converges absolutely by
the integral test. For the details, see Example 3.3.7 (with p = 101 ą 1) in the CLP101
notes.
? ?
S-15: (a) For large n, n2 " 1 and so n2 + 1 « n2 = n. This suggests that we apply the
limit comparison test with an = ? 12 and bn = n1 . Since
n +1
?
an 1/ n2 + 1 1
lim = lim = lim a =1
nÑ8 bn nÑ8 1/n nÑ8 1 + 1/n2
8
ř 1
and since n diverges, the given series diverges.
n =1

(b) Since cos(nπ ) = (´1)n , the given series converges by the alternating series test. To
check that an = 2nn decreases to 0 as n tends to infinity, note that

a n +1 ( n + 1 ) 2´ ( n + 1 )  1 1
= = 1+
an n2´n n 2
3
is smaller than 1 (so that an+1 ď an ) for all n ě 1, and is smaller than 4 (so an+1 ď 43 an ) for
all n ě 2.

k4 ´2k3 +2 4
S-16: For large k, k4 " 2k3 ´ 2 and k5 " k2 + k so k5 + k2 + k
« kk5 = 1k . This suggests that we
k4 ´2k3 +2
apply the limit comparison test with ak = k5 + k2 + k
and bk = 1k . Since

ak k4 ´ 2k3 + 2 k k5 ´ 2k4 + k2 1 ´ 2/k + 1/k3

lim = lim 5 = lim = lim
kÑ8 bk kÑ8 k + k2 + k 1 kÑ8 k5 + k2 + k kÑ8 1 + 1/k3 + 1/k4
=1
8
ř 1
and since k diverges, by the p–test with p = 1, the given series diverges.
k =1

S-17: (a) For large n, n2 " n + 1 and so the numerator n2 + n + 1 « n2 . For large n, n5 " n
2 n +1 2
and so the denominator n5 ´ n « n5 . So, for large n, n n+5 ´n « nn5 = n13 . This suggests that
n2 + n +1 1
we apply the limit comparison test with an = n5 ´n
and bn = n3
. Since

an ( n2 + n + 1) / ( n5 ´ n ) n5 + n4 + n3 1 + 1/n + 1/n2
lim = lim = lim = lim
nÑ8 bn nÑ8 1/n3 nÑ8 n5 ´ n nÑ8 1 ´ 1/n4
=1
8
ř 1
and since n3
converges, by the p–test with p = 3 ą 1, the given series converges.
n =1
? ?
3m+sin m
(b) For large m, 3m " | sin m| and so 2
m?
« 3m
m2
= 3
m. This suggests that we apply
3m+sin m 1 3
the limit comparison test with am = m2
and bm = m. (We could also use bm = m .)

199
Since
? ? ?
am (3m + sin m)/m2 3m + sin m sin m
lim = lim = lim = lim 3 +
mÑ8 bm mÑ8 1/m mÑ8 m mÑ8 m
=3
8
ř 1
and since m diverges, by the p–test with p = 1, the given series diverges.
m =1

6
S-18: This is a geometric series with first term a = 72
and ratio r = 71 . As |r| ă 1, the series
converges and takes the value
a 6/72 6/72 1
= = =
1´r 1 ´ 1/7 6/7 7

S-19: (a) The given series is

8
1 1 1 1 ÿ 1
1+ + + + +¨¨¨ = an with an =
3 5 7 9 2n ´ 1
n =1
1
First we’ll develop some intuition by observing that, for very large n, an « 2n . We know
8
ř 1
that the series n diverges by the p–test with p = 1. So let’s apply the limit comparison
n =1
test with bn = n1 . Since
an n 1 1
lim = lim = lim 1
=
nÑ8 bn nÑ8 2n ´ 1 nÑ8 2 ´ 2
n
8
ř 8
ř
the series an converges if and only if the series bn converges. So the given series
n =1 n =1
diverges.
(a, again) The series
1 1 1 1 1 1 1 1 1
1+ + + + +¨¨¨ ě + + + + +¨¨¨
3 5 7 9 2 4 6 8 10
1  1 1 1 1 
= 1+ + + + +¨¨¨
2 2 3 4 5
The series in the brackets is the harmonic series which we know diverges, by the p–test
with p = 1. So the series on the right hand side diverges. By the convergence test, the
series on the left hand side diverges too.
(2n+1)
(b) We’ll use the ratio test with an = 22n+1
. Since

a n +1 (2n + 3) 22n+1 1 (2n + 3) 1 (2 + 3/n) 1

= 2n+3 = = Ñ ă 1 as n Ñ 8
an 2 (2n + 1) 4 (2n + 1) 4 (2 + 1/n) 4
the series converges.

200
 ?
3 ?
3
k
S-20: (a) For very large k, k ! k2 so that an = k2 ´k « k2k = k5/3
1
. So we apply the limiting
1
convergence test with bk = k5/3 . Since
? 3
ak k/(k2 ´ k) k2 1
lim = lim 5/3
= lim 2
= lim =1
kÑ8 bk kÑ8 1/k kÑ8 k ´ k kÑ8 1 ´ 1/k
8
ř 1 5
and k5/3
converges by the p–test with p = 3 ą 1, the given series converges by the
k =1
limiting comparison test.
k10 10k (k!)2
(b) The kth term in this series is ak = (2k)!
. So

a k +1 (k + 1)10 10k+1 ((k + 1)!)2 (2k)!  k + 1 10 ( k + 1)2

= = 10
ak (2k + 2)! k10 10k (k!)2 k (2k + 2)(2k + 1)
 1 10 (1 + 1/k)2
= 10 1 +
k (2 + 2/k)(2 + 1/k)
1
As k tends to 8, this converges to 10 ˆ 1 ˆ 2ˆ2 ą 1. So the series diverges by the ratio test.
1
(c) We’ll use the integal test. The kth term in the series is ak = k(log k)(log log k)
= f (k) with
1
f (x) = x (log x )(log log x )
, which is continuous, positive and decreasing for x ě 3. Since
ż8 ż8 żR
dx dx
f ( x ) dx = = lim
3 3 x (log x )(log log x ) RÑ8 3 x (log x )(log log x )
ż log R
dy dx
= lim with y = log x, dy =
RÑ8 log 3 y log y x
ż log log R
dt dy
= lim with t = log y, dt =
RÑ8 log log 3 t y
h ilog log R
= log t
log log 3

diverges as R Ñ 8, the series is divergent.

S-21: For large n, the numerator n3 ´ 4 « n3 and the denominator 2n5 ´ 6n « 2n5 , so the
n3 1
nth term is approximately 2n 5 = 2n2 . So we apply the limit comparison test with
n3 ´4 1
an = 2n5 ´6n
and bn = n2
. Since
4
an (n3 ´ 4)/(2n5 ´ 6n) 1´ n3 1
lim = lim = lim = ,
nÑ8 bn nÑ8 1/n2 nÑ8 2 ´ 6 2
n4
8
ř 8
ř
the given series an converges if and only if the series bn converges. Since the series
n =1 n =1
8 8 ˇ
ř ř 1 ˇ
bn = np ˇ is a convergent p-series, both series converge.
n =1 n =1 p =2

201
S-22: By the alternating series test, the error introduced when we approximate the series
8 N
ÿ (´1)n ÿ (´1)n
by is at most the magnitude of the first omitted term
n ¨ 10n n ¨ 10n
n =1 n =1
1
. By trial and error, we find that this expression becomes smaller than
( N + 1)10( N +1)
10´6 when N + 1 ě 6. So the smallest allowable value is N = 5.

S-23: The sequence t n12 u decreases to zero as n increases to infinity. So, by the alternating
π2
series error bound, which is given in Theorem 3.3.14 in the CLP 101 notes, 12 ´ S N lies
(´1) N 1
between zero and the first omitted term, ( N +1)2
. We therefore need ( N +1)2
ď 10´6 , which
is equivalent to N + 1 ě 103 and N ě 999.

S-24: The error introduced when we approximate S by the nth partial sum Sn lies
(´1)n+2
between 0 and the first term dropped, which is (2n+3)2
. So we need the smallest positive
integer n obeying
1 1 7
2
ď ðñ (2n + 3)2 ě 100 ðñ 2n + 3 ě 10 ðñ n ě
(2n + 3) 100 2
So we need n = 4 and then
1 1 1 1
S4 = 2
´ 2+ 2´ 2
3 5 7 9

nn
S-25: (a) There are plenty of powers/factorials. So let’s try the ratio test with an = 9n n! .

a n +1 (n + 1)n+1 9n n! ( n + 1 ) n +1 (1 + 1/n)n e
lim = lim n+1 n
= lim n
= lim =
nÑ8 an nÑ8 9 ( n + 1) ! n nÑ8 n 9 ( n + 1) nÑ8 9 9

Here we have used that lim (1 + 1/n)n = e. See Example ?? in the CLP100 notes, with
nÑ8
1
x= n and a = 1. As e ă 9, our series converges.
1
(b) We know that the series 8
ř
n=1 n2 converges, by the p–test with p = 2, and also that
1 1
log n ě 2 for all n ě e2 . So let’s use the limit comparison test with an = nlog n and bn = n2 .

an 1 n2 1
lim = lim log n = lim log n´2 = 0
nÑ8 bn nÑ8 n 1 nÑ8 n

So our series converges, by the limit comparison test.

S-26: (a)
• Our first task is to identify the potential sources of impropriety for this integral.
• The domain of integration extends to +8. On the domain of integration the
denominator is never zero so the integrand is continuous. Thus the only problem is
at +8.

202
 • Our second task is to develop some intuition about the behavior of the integrand
for very large x. When x is very large:
– | sin x| ď 1 ! x, so that the numerator x + sin x « x, and
– 1 ! x2 , so that denominator 1 + x2 « x2 , and
– the integrand x1++sin
x2
x
« xx2 = 1x
ş8 ş8 x +sin x
• Now, since 2 dxx diverges, we would expect 2 1+ x 2
dx to diverge too.
• Our final task is to verify that our intuition is correct. To do so, we set
x + sin x 1
f (x) = g( x ) =
1 + x2 x
and compute
f (x) x + sin x 1
lim = lim ˜
xÑ8 g( x ) xÑ8 1 + x 2 x
(1 + sin x/x ) x
= lim ˆx
xÑ8 (1/x2 + 1) x2
1 + sin x/x
= lim
xÑ8 1/x2 + 1
=1
ş8 ş8
• Since 2 g( x ) dx = 2 dx 3
x diverges, by Example 1.12.8 in the CLP 101 notes , with
şp8= 1, Theorem ş81.12.22(b)
x +sin x
in the CLP 101 notes now tells us that
2 f ( x ) dx = 2 e´x + x2 dx diverges too.
ş8 ş8
(a again) Since 2 1+1x2 dx ď 2 x12 dx converges, by the p–test with p = 2, and
| sin x| 1
ş8 sin x ş8 x+sin x
1+ x 2 ď 1+ x 2 , the integral 2 1+ x 2 dx converges. Hence 2 1+ x2 dx converges if and
ş8 x
only if 2 1+ x2 dx converges. But
ż8
x
żr
x h ir
1 2
dx = lim dx = lim log ( 1 + x )
2 1+x
2 rÑ8 2 1 + x2 rÑ8 2 2
ş8 x+sin x
diverges, so 2 1+ x2 dx diverges.
x +sin x
(b) The problem is that f ( x ) = 1+ x 2
is not a decreasing function. To see this, compute
the derivative:
(1 + cos x )(1 + x2 ) ´ ( x + sin x )(2x ) (cos x ´ 1) x2 ´ 2x sin x + 1 + cos x
f 1 (x) = =
(1 + x 2 )2 (1 + x 2 )2
If x = 2mπ, the numerator is 0 ´ 0 + 1 + 1 ą 0.
(c) Set an = n1++sin
n2
n
. We first try to develop some intuition about the behaviour of an for
large n and then we confirm that our intuition was correct.

3 To change the lower limit of integration from 1 to 2, just apply Theorem 1.12.20 in the CLP 101 notes.

203
 • Step 1: Develop intuition. When n " 1, the numerator n + sin n « n, and the
denominator 1 + n2 « n2 so that an « nn2 = n1 and it looks like our series should
diverge by the p–test (Example 3.3.6 in the CLP 101 notes) with p = 1.
• Step 2: Verify intuition. To confirm our intuition we set bn = n1 and compute the
limit
n+sin n
an 2 n[n + sin n] 1 + sinn n
lim = lim 1+1n = lim 2
= lim =1
nÑ8 bn nÑ8 nÑ8 1 + n nÑ8 1 + 1
n 2 n
8 8
ř ř 1
We already know that the series bn = n diverges by the p–test with p = 1.
n =1 n =1
So our series diverges by the limit comparison test, Theorem 3.3.11 in the CLP 101
notes.
ˇ nˇ 8
(c again) Since ˇ 1sin 1 1
ř
+ n2
ˇ ď n2
and the series n2
converges by the p–test with p = 2, the
n =1
8 8 8
ř sin n ř n+sin n ř n
series 1+ n2
converges. Hence 1+ n2
converges if and only if the series 1+ n2
n =1 n =1 n =1
x
converges. Now f ( x ) = 1+ x 2
is a continuous, positive, decreasing function on [1, 8) since

(1 + x2 ) ´ x (2x ) 1 ´ x2
f 1 (x) = =
(1 + x 2 )2 (1 + x 2 )2
ş8
is negative for all x ą 1. We saw in part (a) that the integral 2 1+xx2 dx diverges. So the
ş8 8
integral 1 1+xx2 dx diverges too and the sum n
ř
1+ n2
diverges by the integral test. So
n =1
8
ř n+sin n
1+ n2
diverges.
n =1

?
e´ x
S-27: Note that ?
x
decreases as x increases. Hence, for every n ě 1,
? żn ? żn ?
e´ n e´ n e´ x
? = ? dx ď ? dx
n n´1 n n´1 x

and, for every N ě 1,

? ? ? ?
8 n 8 żn x ż N +1 x ż N +2 x
ÿ e´ ÿ e´ e´ e´
EN = ? ď ? dx = ? dx + ? dx + ¨ ¨ ¨
n x x N +1 x
n = N +1 n= N +1 n´1 N
?
e´ x
ż8
= ? dx
N x
? 1?
dx
Substituting y = x, dy = 2 x,
?
ż8
e´ x ż8 ˇ8 ?
dx = 2 e´y dy = ´2e´y ˇ? = 2e´ N
? ˇ
?
N x N N

204
 ? ?
e´ n
N.
ř8
This shows that n = N +1
?
n
converges and is between 0 and 2e´ Since
?
E14 = 2e´ 14 = 0.047, we may truncate the series at n = 14.
? ?
8 ´ n 14 ´ n
ÿ e ÿ e
? = ? + E14
n n
n =1 n =1
= 0.3679 + 0.1719 + 0.1021 + 0.0677 + 0.0478
+ 0.0352 + 0.0268 + 0.0209 + 0.0166 + 0.0134
+ 0.0109 + 0.0090 + 0.0075 + 0.0063 + E14
= 0.9042 + E14

The sum is between 0.9035 and 0.9535. This even allows for a roundoff error of 0.00005 in
each term.

8
ř
S-28: Since an , converges an must converge to zero as n Ñ 8. In particular there must
n =1
1
be some integer N such that 2 ą an ě 0 for all n ą N. Hence

8 N 8 N 8
ÿ an ÿ an ÿ an ÿ an ÿ an
= + ď +
1 ´ an 1 ´ an 1 ´ an 1 ´ an 1/2
n =1 n =1 n = N +1 n =1 n = N +1
N 8 N 8
ÿ an ÿ ÿ an ÿ
= +2 an ď +2 an
1 ´ an 1 ´ an
n =1 n = N +1 n =1 n =1

is finite.

8
ř
S-29: By the divergence test, the fact that (1 ´ an ) converges guarantees that
n =0
lim (1 ´ an ) = 0, or equivalently, that lim an = 1. So, by the divergence test, a second
nÑ8 nÑ8
time, the fact that

lim 2n an = +8
nÑ8

8
2n an diverges too.
ř
guarantees that
n =0

8
ÿ nan ´ 2n + 1
S-30: By the divergence test, the fact that converges guarantees that
n+1
n =1
+1
lim nann´2n
+1 = 0, or equivalently, that
nÑ8

n 2n ´ 1
0 = lim an ´ lim = lim an ´ 2 ðñ lim an = 2
nÑ8 n + 1 nÑ8 n + 1 nÑ8 nÑ8

8 
ř 
The series of interest can be written ´ log a1 + log( an ) ´ log( an+1 ) which looks like
n =1

205
a telescoping series. So we’ll compute the partial sum
N
ÿ  
S N = ´ log a1 + log( an ) ´ log( an+1 )
n =1
     
= ´ log a1 + log( a1 ) ´ log( a2 ) + log( a2 ) ´ log( a3 ) + ¨ ¨ ¨ + log( a N ) ´ log( a N +1 )
= ´ log( a N +1 )
and then take the limit N Ñ 8
8
ÿ   1
´ log a1 + log( an ) ´ log( an+1 ) = lim S N = ´ lim log( a N +1 ) = ´ log 2 = log
NÑ8 NÑ8 2
n =1

ř8
S-31: We are told that n =1 a n converges. Thus we must have that lim an = 0. In
nÑ8
particular, there is an index N such that 0 ď an ď 1 for all n ě N. Then a2n ď an for all
n ě N and
ÿ8 N´1
ÿ ÿ8 N´1
ÿ ÿ8
2 2 2 2
an = an + an ď an + an ă 8
n =1 n =1 n= N n =1 n= N
ř8 2
Thus n =1 a n converges.

Solutions to Exercises 3.4 — Jump to TABLE OF CONTENTS

8 8
S-1: False. For example if bn = n1 , then (´1)n+1 bn = (´1)n+1 n1 converges by the
ř ř
n =1 n =1
8
ř 1
alternating series test, but n diverges by the p–test.
n =1

8
ř (´1)n
S-2: The series 9n+5 converges by the alternating series test. On the other hand the
n =1
8 ˇ nˇ
ˇ (´1) ˇ = 8 1
diverges by the limiting comparison test with bn = n1 . So
ř ř
series 9n+5 n=1 9n+5
n =1
the given series is conditionally convergent.

S-3: Note that (´1)2n+1 = (´1) ¨ (´1)2n = ´1. So we can simplify

8 8
ÿ (´1)2n+1 ÿ 1
=´
1+n 1+n
n =1 n =1
8
1 1 1 ÿ 1
Since ě = , diverges by the comparison test with the
1+n n+n 2n 1+n
n =1
8
ř 1
divergent harmonic series n. The extra overall factor of ´1 in the original series does
n =1
not change the conclusion of divergence.

206
S-4: Since
1 + 4n 1 + 4n
lim = lim =1
nÑ8 3 + 22n nÑ8 3 + 4n

1 + 4n
n´1
the alternating series test cannot be used. Indeed, lim (´1) does not exist (for
n
nÑ8 3 + 22n
very large n, (´1)n´1 31++242n alternates betwen a number close to +1 and a number close to
´1) so the divergence test says that the series diverges. (Note that “none of the above”
cannot possibly be the correct answer — every series either converges absolutely,
converges conditionally, or diverges.)

S-5: First, we’ll develop some intuition. For very large n

ˇ? ˇ ˇ? ˇ ˇ ˇ
ˇ n cos(n) ˇ ˇ n cos(n) ˇ ˇ cos(n) ˇ
ˇ=ˇ ˇď 1
ˇ n2 ´ 1 ˇ « ˇ
ˇ ˇ ˇ
n 2 ˇ ˇ n 3/2 ˇ n3/2
8
ÿ 1
since |cos(n)| ď 1 for all n. By the p–test, the series converges for all p ą 1. So we
n =5
np
would expect the given series to converge absolutely.
ˇ? ˇ
ˇ n cos(n) ˇ
Now, to confirm that our intuition is correct, we’ll set an = ˇ n2 ´1 ˇ and apply the limit
comparison test with the comparison series having nth term bn = n1p . We’ll choose a
specific p shortly. Since
ˇ? ˇ
an ˇ n cos n/(n2 ´ 1)ˇ n p+1/2 | cos n| | cos n|
lim = lim p = lim 2 2
= lim 3/2´p
nÑ8 bn nÑ8 1/n nÑ8 n (1 ´ 1/n ) nÑ8 n (1 ´ 1/n2 )
3
= 0 if p ă
2
8
the limit comparison test says that if p ă 23 and the series
ř
bn converges (which is the
n =5
8 ˇ? ˇ
ˇ n cos(n) ˇ
ˇ n2 ´1 ˇ also converges. So choosing any 1 ă p ă 23 , for
ř
case if p ą 1) then the series
n =5
5
example p = 4, we conclude that the given series converges absolutely.

8 3
24n2 e´n converges. If we replace n by x in the
ř
S-6: (a) We need to show that
n =1
3
summand, we get f ( x ) = 24x2 e´x , which we can integate. (Just substitute u = x3 .) So
let’s try the integral test. First, we have to check that f ( x ) is positive and decreasing. It is
certainly positive. To determine if it is dereasing, we compute
df 3 3 3
= 48xe´x ´ 24 ˆ 3x4 e´x = 24x (2 ´ 3x3 )e´x
dx
which is negative for x ě 1. Therefore f ( x ) is decreasing for x ě 1, and the integral test
applies. The substitution u = x3 , du = 3x2 dx, yields
ż ż ż
2 ´x3 3
f ( x ) dx = 24x e dx = 8e´u du = ´8e´u + C = ´8e´x + C.

207
Therefore

ż8 żR  R
´x3
f ( x ) dx = lim f ( x ) dx = lim ´8e
1 RÑ8 1 RÑ8 1
´R3 ´1 ´1
= lim (´8e + 8e ) = 8e
RÑ8

8 3
24n2 e´n converges and the series
ř
Since the integral is convergent, the series
n =1
8
ÿ 3
(´1)n´1 24n2 e´n converges absolutely.
n =1

3
Alternatively, we can use the ratio test with an = 24n2 e´n . We calculate

ˇ ˇ
ˇ ˇ
ˇ a n +1 ˇ ˇ 24(n + 1)2 e´(n+1)3 ˇ
lim ˇ ˇ = lim ˇˇ ˇ
3 ˇ
nÑ8 ˇ an ˇ nÑ8 ˇ 24n e2 ´n ˇ
3
!
( n + 1)2 e n
= lim
nÑ8 n2 e ( n +1)
3

 
1 2 ´(3n2 +3n+1)
= lim 1 + e = 1 ¨ 0 = 0 ă 1,
nÑ8 n

and therefore the series converges absolutely.

Alternatively, alternatively, we can use the limiting comparison test. First a little intuition
8 3
24n2 e´n converges. The nth term in this
ř
building. Recall that we need to show that
n =1
series is
3 24n2
an = 24n2 e´n = 3
en

It is a ratio with both the numerator and denominator growing with n. A good rule of
thumb is that exponentials grow a lot faster than powers. For example, if n = 10 the
numerator is 2400 = 2.4 ˆ 103 and the denominator is about 2 ˆ 10434 . So we would
guess that an tends to zero as n Ñ 8. The question is “does an tendř8 to1zero fast enough
with n that our series converges?”. For example, we know that n=1 n2 converges (by the
p–test with p = 2). So if an tends to zero faster than n12 does, our series will converge. So
3 24n2 1
let’s try the limiting convergence test with an = 24n2 e´n = 3 and bn = n2
.
en

3
an 24n2 e´n 24n4
lim = lim = lim
nÑ8 bn nÑ8 1/n2 nÑ8 en3

208
By l’Hôpital’s rule, twice,

24x4 4 ˆ 24x3
lim = lim by l’Hôpital
xÑ8 e x3 xÑ8 3x 2 e x3
32x
= lim x3 just cleaning up
xÑ8 e
32
= lim 2 x3 by l’Hôpital, again
xÑ8 3x e
=0
ř8
That’s it. The limiting convergence test now tells us that n =1 a n converges.
3
(b) In part (a) we saw that 24n2 e´n is positive and decreasing. The limit of this sequence
equals 0 (as can be shown with l’Hôpital’s Rule, just as we did at the end of part (a)).
Therefore, we can use the alternating series test, so that the error made in approximating
8 8 3
N
(´1)n´1 24n2 e´n by the sum, S N =
ř ř ř
the infinite sum S = an = an , of its first N
n =1 n =1 n =1
terms lies between 0 and the first omitted term, a N +1 . If we use 5 terms, the error satisfies
3
|S ´ S5 | ď |a6 | = 24 ˆ 36e´6

Solutions to Exercises 3.5 — Jump to TABLE OF CONTENTS

S-1: (a) We apply the ratio test for the series whose kth term is ak = (´1)k 2k+1 x k . Then
ˇ (´1)k+1 2k+2 x k+1 ˇ
ˇ ˇ ˇ ˇ
ˇ a k +1 ˇ
lim ˇ ˇ = lim ˇ ˇ
kÑ8 ˇ ak ˇ kÑ8 ˇ (´1)k 2k+1 x k ˇ
= lim |2x| = |2x|
kÑ8

Therefore, by the ratio test, the series converges for all x obeying |2x| ă 1, i.e. |x| ă 12 , and
diverges for all x obeying |2x| ą 1, i.e. |x| ą 21 . So the radius of convergence is R = 12 .
Alternatively, one can set Ak = (´1)k 2k+1 and compute
ˇ (´1)k+1 2k+2 ˇ
ˇ ˇ ˇ ˇ
ˇ A k +1 ˇ
A = lim ˇˇ ˇ = lim ˇ ˇ = lim 2 = 2
kÑ8 Ak ˇ kÑ8 ˇ (´1)k 2k+1 ˇ kÑ8

1
so that R = A = 21 , again.
(b) The series is
8 8 8
ÿ
k k +1 k
ÿ
k
ÿ ˇ
kˇ 1 2
(´1) 2 x =2 (´2x ) = 2 r ˇ = 2ˆ =
r =´2x 1´r 1 + 2x
k =0 k =0 k =0

for all |r| = |2x| ă 1, i.e. all |x| ă 21 .

209
 xk
S-2: We apply the ratio test for the series whose kth term is ak = 10k+1 (k+1)!
. Then

x k +1 10k+1 (k + 1)! ˇˇ
ˇ ˇ ˇ ˇ
ˇ a k +1 ˇ ˇ
lim ˇ ˇ = lim ˇ k+2
ˇ
kÑ8 ˇ ak ˇ kÑ8 10 ( k + 2) ! xk ˇ
1
= lim |x| = 0 ă 1
kÑ8 10( k + 2)

for all x. Therefore, by the ratio test, the series converges for all x and the radius of
convergence is R = 8.
ˇA ˇ
Alternatively, one can set Ak = 10k+1 1(k+1)! and compute A = limkÑ8 ˇ Ak+1 ˇ = 0, so that R
k
is again +8.

( x´2)n
S-3: We apply the ratio test with an = n2 +1
. Since

ˇ ( x ´ 2 ) n +1 n 2 + 1 ˇ
ˇa ˇ ˇ ˇ
ˇ n +1 ˇ
lim ˇ ˇ = lim ˇ
ˇ ˇ
nÑ8 an nÑ8 ( n + 1)2 + 1 ( x ´ 2)n ˇ

n2 + 1
= lim |x ´ 2|
nÑ8 ( n + 1)2 + 1

1 + 1/n2
= lim |x ´ 2|
nÑ8 (1 + 1/n )2 + 1/n2

= |x ´ 2|
the series converges if |x ´ 2| ă 1 and diverges if |x ´ 2| ą 1. So the radius of convergence
is 1.

(´1)n?( x +2)n
S-4: We apply the ratio test for the series whose nth term is an = n
. Then
?
ˇ ( x + 2 ) n +1
ˇ ˇ ˇ ˇ
ˇ a n +1 ˇ n ˇˇ
lim ˇ ˇ = lim ˇ ?
ˇ
nÑ8 ˇ an ˇ nÑ8 n + 1 ( x + 2) n ˇ
?
n
= lim |x + 2| ?
nÑ8 n+1
1
= lim |x + 2| ?
nÑ8 1 + 1/n
= |x + 2|
So the series must converge when |x + 2| ă 1 and must diverge when |x + 2| ą 1. When
x + 2 = 1, the series reduces to
8
ÿ (´1)n
?
n
n =1

which converges by the alternating series test. When x + 2 = ´1, the series reduces to
8
ÿ 1
?
n
n =1

210
which diverges by the p–series test with p = 21 . So the interval of convergence is
´1 ă x + 2 ď 1 or (´3, ´1].

 n
(´1)n x +1
S-5: We apply the ratio test for the series whose nth term is an = n +1 3 . Then

 
ˇ ˇ ˇ (´1)n+1 x+1 n+1 ˇ
ˇa ˇ n +2 3
lim ˇˇ n+1 ˇˇ = lim ˇˇ
ˇ ˇ
 n ˇˇ
nÑ8 an nÑ8 (´1) n
x +1
n +1 3
ˇ ˇ
n + 1 ˇ x + 1 ˇˇ |x + 1|
ˇ
= lim = .
nÑ8 n + 2 ˇ 3 ˇ 3

|x +1|
Therefore, by the ratio test, the series converges when 3 ă 1 and diverges when
|x +1|
3 ą 1. In particular, it converges when

|x + 1| ă 3 ðñ ´3 ă x + 1 ă 3 ðñ ´4 ă x ă 2

(´1)n
and the radius of convergence is R = 3. (Alternatively, one can set An = ( n +1)3n
and
ˇA ˇ
compute A = limnÑ8 ˇ n+1 ˇ = 1 , so that R = 1 = 3.)
An 3 A
Next, we consider the endpoints 2 and ´4. At x = 2, i.e. x + 1 = 3, the series is simply
ř8 (´1)n
n=0 n+1 , which is an alternating series: the signs alternate, and the unsigned terms
decrease to zero. Therefore the series converges at x = 2 by the alternating series test.
At x = ´4 the series is
 
(´1)n ´4 + 1 n
8 8 8
ÿ ÿ (´1)n n
ÿ 1
= (´1) = ,
n =0
n + 1 3 n =0
n + 1 n =0
n + 1

n
since (´1)n ¨ (´1)n = (´1)2n = (´1)2 = 1. This series diverges, either by comparison
ř p-series
or limit comparison with the harmonic series (the with p = 1). (For that matter, it
1
is exactly equal to the standard harmonic series 8n =1 n , re-indexed to start at n = 0.)
In summary, the interval of convergence is ´4 ă x ď 2, or simply (´4, 2].

( x´2)n
S-6: We first apply the ratio test with an = n4/5 (5n ´4)
. Since

( x ´ 2 ) n +1 n4/5 (5n ´ 4) ˇˇ
ˇa ˇ ˇ ˇ
ˇ n +1 ˇ ˇ
lim ˇ ˇ = lim ˇ
ˇ
nÑ8 an nÑ8 ( n + 1)4/5 (5n+1 ´ 4) ( x ´ 2)n ˇ

n4/5 (5n ´ 4)
= lim |x ´ 2|
nÑ8 ( n + 1)4/5 (5n+1 ´ 4)
(1 ´ 4/5n )
= lim |x ´ 2|
nÑ8 (1 + 1/n )4/5 (5 ´ 4/5n )
|x ´ 2|
=
5

211
the series converges if |x ´ 2| ă 5 and diverges if |x ´ 2| ą 5. When x ´ 2 = +5, i.e. x = 7,
8 8
ř 5n ř 1
the series reduces to n4/5 (5n ´4)
= n4/5 (1´4/5n )
which diverges by the limit
n =1 n =1
1
comparison test with bn = n4/5 . When x ´ 2 = ´5, i.e. x = ´3, the series reduces to
8
ř (´5)n 8
ř (´1)n
n4/5 (5n ´4)
= n4/5 (1´4/5n )
which converges by the alternating series test. So the
n =1 n =1
interval of convergence is ´3 ď x ă 7 or [´3, 7).

( x +2) n
S-7: We apply the ratio test with an = n2
. Since
ˇ ( x +2 ) n +1 ˇ
ˇa ˇ
ˇ n +1 ˇ
ˇ
ˇ ( n +1)2 ˇ
ˇ n2 1
lim ˇ ˇ = lim ˇ ( x+2)n ˇ = lim 2
|x + 2| = lim |x + 2| = |x + 2|
nÑ8 an nÑ8 ˇ ˇ nÑ8 (n + 1) nÑ8 (1 + 1/n )2
n2

we have convergence for

|x + 2| ă 1 ðñ ´1 ă x + 2 ă 1 ðñ ´3 ă x ă ´1
and divergence for |x + 2| ą 1. For |x + 2| = 1, i.e. for x + 2 = ˘1, i.e. for x = ´3, ´1, the
8 8
ř (˘1)n ř 1
series reduces to n2
, which converges absolutely, because n p converges for
n =1 n =1
p = 2 ą 1. So the given series converges if and only if ´3 ď x ď ´1.

n
S-8: We apply the ratio test with an = 4n ( x ´ 1)n . Since
ˇ n +1
ˇ 4 ( x ´ 1 ) n +1 / ( n + 1 ) ˇ
ˇ ˇ ˇ
ˇ a n +1 ˇ
lim ˇ ˇ = lim ˇ ˇ
nÑ8 ˇ an ˇ nÑ8 ˇ 4n ( x ´ 1)n /n ˇ
n
= lim 4|x ´ 1|
nÑ8 n+1
n
= 4|x ´ 1| lim = 4|x ´ 1| ¨ 1.
nÑ8 n + 1

the series converges if

1 1 3 5
4|x ´ 1| ă 1 ðñ ´1 ă 4( x ´ 1) ă 1 ðñ ´ ă x ´ 1 ă ðñ ăxă
4 4 4 4
and diverges if 4|x ´ 1| ą 1. Checking the right endpoint x = 54 , we see that
8  n 8
ÿ 4n 5 ÿ 1
´1 =
n 4 n
n =1 n =1

is the divergent harmonic series. At the left endpoint x = 43 ,

8  n 8
ÿ 4n 3 ÿ (´1)n
´1 =
n 4 n
n =1 n =1

converges by the alternating series
test. Therefore the interval of convergence of the
original series is 43 ď x ă 54 , or 43 , 54 .

212
 ( x´1)n
S-9: We apply the ratio test with an = (´1)n 2n (n+2) . Since

ˇ ( x ´ 1 ) n +1 2n ( n + 2 ) ˇ
ˇ ˇ ˇ ˇ
ˇ a n +1 ˇ
lim ˇ ˇ = lim ˇ n+1
ˇ ˇ
nÑ8 ˇ an ˇ nÑ8 2 ( n + 3) ( x ´ 1) n ˇ
|x ´ 1| n + 2
= lim
nÑ8 2 n+3
|x ´ 1| 1 + 2/n |x ´ 1|
= lim =
2 nÑ8 1 + 3/n 2
the series converges if
|x ´ 1|
ă 1 ðñ |x ´ 1| ă 2 ðñ ´2 ă ( x ´ 1) ă 2 ðñ ´1 ă x ă 3
2
and diverges if |x ´ 1| ą 2. So the series has radius of convergence 2. Checking the left
endpoint x = ´1, so that x´12 = ´1, we see that
8 8
ÿ (´1 ´ 1)n ÿ 1
(´1)n =
n =0
2n ( n + 2 ) n =0
n+2

x´1
is the divergent harmonic series. At the right endpoint x = 3, so that 2 = +1 and
8 8
ÿ (3 ´ 1) n ÿ (´1)n
(´1)n =
n =0
2n ( n + 2 ) n =0
n+2

converges by the alternating series test. Therefore the interval of convergence of the
original series is ´1 ă x ď 3, or ´ 1, 3 .

S-10: We apply the ratio test with an = (´1)n n2 ( x ´ a)2n . Since

ˇ (´1)n+1 (n + 1)2 ( x ´ a)2(n+1) ˇ
ˇ ˇ ˇ ˇ
ˇ a n +1 ˇ
lim ˇ ˇ = lim ˇ ˇ
nÑ8 ˇ an ˇ nÑ8 ˇ (´1)n n2 ( x ´ a)2n ˇ
( n + 1)2
= lim |x ´ a|2
nÑ8 n2

2
= |x ´ a|2 lim 1 + 1/n = |x ´ a|2 ¨ 1.
nÑ8

the series converges if

|x ´ a|2 ă 1 ðñ |x ´ a| ă 1 ðñ ´1 ă x ´ a ă 1 ðñ a ´ 1 ă x ă a + 1

and diverges if |x ´ a| ą 1. Checking both endpoints x ´ a = ˘1, we see that

8
ÿ ˇ 8
ÿ
n 2 2n ˇ
(´1)n n2
ˇ
(´1) n ( x ´ a) ˇ =
n =1 x´a=˘1 n =1

fails the divergence test — the nth term does not converge to zero as n Ñ 8. Therefore

the interval of convergence of the original series is a ´ 1 ă x ă a + 1, or a ´ 1, a + 1 .

213
 ( x +1) k
S-11: (a) We apply the ratio test for the series whose kth term is Ak = k 2 9k
. Then

ˇ ( x + 1 ) k +1 2 9k ˇ
ˇ ˇ ˇ ˇ
ˇ A k +1 ˇ k
lim ˇ ˇ = lim ˇ ˇ
kÑ8 ˇ Ak ˇ kÑ8 ˇ ( k + 1)2 9k+1 ( x + 1)k ˇ
1 k2
= lim |x + 1|
kÑ8 9 ( k + 1)2
1 1
= lim |x + 1|
kÑ8 9 (1 + 1/k )2
|x + 1|
=
9
So the series must converge when |x + 1| ă 9 and must diverge when |x + 1| ą 9. When
x + 1 = ˘9, the series reduces to
8 8
ÿ (˘9)k ÿ (˘1)k
=
k 2 9k k2
k =1 k =1

1
which converges (since, by the p–test, 8
ř
k=1 k p converges for any p ą 1). So the interval of
covnergence is |x + 1| ď 9 or ´10 ď x ď 8 or [´10, 8].
(b) The partial sum

N   a
ak a a2   a2 a3   a
N a  a a
´ N +1 = 1 ´ N +1
ÿ
´ k +1 = 1 ´ + ´ +¨¨¨+
a k +1 a k +2 a2 a3 a3 a4 a N +1 a N +2 a 2 a N +2
k =1

8  
ÿ ak a a
We are told that ´ k+1 = 1 . This means that the above partial sum
a k +1 a k +2 a2
k =1
a1
converges to a2 as N Ñ 8, or equivalently, that

a N +1
lim =0
NÑ8 a N +2

and hence that

|ak+1 ( x ´ 1)k+1 | |a |
lim k
= |x ´ 1| lim k+1
kÑ8 |ak ( x ´ 1) | kÑ8 |ak |

is infinite for any x ‰ 1. So, by the ratio test, this series converges only for x = 1.

8
1
xn =
ř
S-12: Using the geometric series 1´x ,
n =0

8 8 8
x3 ÿ ÿ ÿ
= x3 xn = x n +3 = xn
1´x n =0 n =0 n =3

214
 xn
S-13: We apply the ratio test for the series whose nth term is either an = 32n log n
or
ˇ n ˇ
an = ˇ 2nx ˇ. For both series
3 log n

n +1 2n log n ˇ
ˇ ˇ ˇ ˇ
ˇ a n +1 ˇ ˇ x 3
lim ˇ ˇ = lim ˇ ˇ
nÑ8 ˇ an ˇ nÑ8 ˇ 32(n+1) log( n + 1) xn ˇ
ˇ ˇ ˇ ˇ
ˇ x log n ˇ ˇ x log n ˇ
= lim ˇ 2
ˇ ˇ = lim ˇ ˇ
nÑ8 3 log( n + 1) ˇ nÑ8 ˇ 32 [log( n ) + log(1 + 1/n )] ˇ
ˇ ˇ
ˇ x ˇ
= lim ˇˇ 2 ˇ
nÑ8 3 [1 + log(1 + 1/n ) / log( n )] ˇ
|x|
=
9
Therefore, by the ratio test, our series converges absolutely when |x| ă 9 and diverges
when |x| ą 9.
8 8
ÿ xn ÿ (´1)n
For x = ´9, 2n log n
= which converges by the alternating series test.
n =2
3 n =2
log n
8 8 8 ˇ
ÿ xn ÿ 1 ÿ ˇ (´1)n ˇ
ˇ
For x = +9, 2n log n
= which is the same series as ˇ ˇ. We shall
n =2
3 n =2
log n n =2
log n
shortly show that n ě log n, and hence log1 n ě n1 for all n ě 1. This implies that the series
8 8 ˇ
ÿ 1 ÿ 1 ˇˇ
diverges by comparison with the divergent series pˇ . This yelds both
n =2
log n n =2
n p = 1
divergence for x = 9 and also the failure of absolute convergence for x = ´9.
Finally, we show that n ´ log n ą 0, for all n ě 1. Set f ( x ) = x ´ log x. Then f (1) = 1 ą 0
and
1
f 1 (x) = 1 ´ ě 0 for all x ě 1
x
So f ( x ) is (strictly) positive when x = 1 and is increasing for all x ě 1. So f ( x ) is (strictly)
positive for all x ě 1.

8
1
(´1)n r n , with r = x3 , gives
ř
S-14: (a) Applying 1+r =
n =0
ż 8 ż 8 3n+1
1 n x
ÿ ÿ
n 3n
dx = (´1) x dx = (´1) +C
1 + x3 n =0 n =0
3n + 1

(b) By part (a),

ż 1/4 8 3n+1 ˇ1/4 8
ˇ
1 n x 1
ÿ ÿ
3
dx = (´1) ˇ = (´1)n
0 1+x n =0
3n + 1 0ˇ
n =0
(3n + 1)43n+1

This is an alternating series with successively smaller terms that converge to zero as
n Ñ 8. So truncating it introduces an error no larger than the magnitude of the first

215
dropped term. We want that first dropped term to obey

1 1
3n + 1
ă 10´5 = 5
(3n + 1)4 10

So let’s check the first few terms.

ˇ
1 ˇ 1 1
ˇ = ą
(3n + 1)43n+1 n=0 4 105
ˇ
ˇ
1 ˇ 1 1
3n + 1
ˇ = 5 ą 5
(3n + 1)4 ˇ 4 10
ˇn =1
1 ˇ 1 1 1 1 1
3n + 1
ˇ = 7
= 14
= = ă 5
(3n + 1)4 ˇ
n =2 7ˆ4 7ˆ2 7 ˆ 16 ˆ 1024 112 ˆ 1024 10

So we need to keep two terms (the n = 0 and n = 1 terms).

S-15: (a) Differentiating both sides of

8
ÿ 1
xn =
n =0
1´x

gives
8
ÿ 1
nx n´1 =
n =0
(1 ´ x )2
Now multiplying both sides by x gives
8
ÿ x
nx n =
n =0
(1 ´ x )2

as desired.
(b) Differentiating both sides of the conclusion of part (a) gives
8
ÿ (1 ´ x )2 ´ 2x ( x ´ 1) (1 ´ x )(1 ´ x + 2x ) 1+x
n2 x n´1 = = =
n =0
(1 ´ x )4 (1 ´ x )4 (1 ´ x )3

Now multiplying both sides by x gives

8
ÿ x (1 + x )
n2 x n =
n =0
(1 ´ x )3

We know that differentiation preserves the radius of convergence of power series. So this
series has radius of convergence 1 (the radius of convergence of the original geometric
series). At x = ˘1 the series diverges by the divergence test. So the series converges for
´1 ă x ă 1.

216
 8
ř
S-16: By the divergence test, the fact that (1 ´ bn ) converges guarantees that
n =0
lim (1 ´ bn ) = 0, or equivalently, that lim bn = 1. So, by equation (3.5.2) in the CLP 101
nÑ8 nÑ8
notes, the radius of convergence is
 ˇb ˇ´1  1 ´1
ˇ n +1 ˇ
R= lim ˇ ˇ = =1 (3.6.1)
nÑ8 bn 1

S-17: (a) We know that the radius of convergence R obeys

1 a n ( n + 1 ) a n +1 C
= lim n+1 = lim =1 =1
R nÑ8 an nÑ8 n + 1 nan C

because we are told that lim nan = C. So R = 1.

nÑ8

(b) Just knowing that the radius of convergence is 1, we know that the series converges
for |x| ă 1 and diverges for |x| ą 1. That leaves x ˘ 1.
8
ř
When x = +1, the series reduces to an . We are told that nan decreases to C ą 0. So
n =1
8
C ř 1
an ě n. By the comparison test with the harmonic series n, which diverges by the
n =1
p–test with p = 1, our series diverges when x = 1.
8
(´1)n an . We are told that nan decreases to C ą 0.
ř
When x = ´1, the series reduces to
n =1
8
(´1)n an converges by the
ř
So an ą 0 and an converges to 0 as n Ñ 8. Consequently
n =1
alternating series test.
8
an x n converges when ´1 ď x ă 1.
ř
In conclusion
n =1

Solutions to Exercises 3.6 — Jump to TABLE OF CONTENTS

S-1: Substituting y = 3x into the exponential series

8
y
ÿ yn
e =
n =0
n!

gives
8 8
ÿ (3x )n ÿ 3n x n
e3x = =
n =0
n! n =0
n!
35
so that c5 , the coefficient of x5 , which appears only in the n = 5 term, is c5 = 5!

217
S-2: We just need to substitute y = x3 into the known Maclaurin series for sin y, to get the
Maclaurin series for sin( x3 ), and then multiply the result by x2 .

y3
sin y = y ´ +¨¨¨
3!
x9
sin( x3 ) = x3 ´ +¨¨¨
3!
x11
x2 sin( x3 ) = x5 ´ +¨¨¨
3!

so a = 1 and b = ´ 3!1 = ´ 16 .

8
1
yn gives
ř
S-3: Substituting y = 2x into 1´y =
n =0

8 8
1 1 ÿ
n
ÿ
f (x) = =´ =´ (2x ) = ´ 2n x n
2x ´ 1 1 ´ 2x n =0 n =0

8
1
yn gives
ř
S-4: Substituting first y = ´x and then y = 2x into 1´y =
n =0

8 8
1 ÿ
n
ÿ
= (´x ) = (´1)n x n
1 ´ (´x ) n =0 n =0
8 8
1 ÿ ÿ
= (2x )n = 2n x n
1 ´ (2x ) n =0 n =0

Hence

8 8
3 1 3 1 ÿ
n n
ÿ
f (x) = ´ = + =3 (´1) x + 2n x n
x + 1 2x ´ 1 1 ´ (´x ) 1 ´ 2x n =0 n =0
8
ÿ

= 3(´1)n + 2n x n
n =0

So bn = 3(´1)n + 2n .

S-5: Recall that

8
y
ÿ yn y2 y3
e = = 1+ y + + +¨¨¨
n =0
n! 2 3!

218
Setting y = ´x2 , we have

x4 x6
2
e´x = 1 ´ x2 +
´ +¨¨¨
2 3!
2 x4 x6
e´x ´ 1 = ´x2 + ´ +¨¨¨
2 6
2
e´x ´ 1 x3 x5
= ´x + ´ +¨¨¨
x 2 6
ż ´x2
e ´1 x2 x4 x6
dx = C ´ + ´ +¨¨¨
x 2 8 36

S-6: Recall that

8
ÿ y2n+1
arctan(y) = (´1)n
n =0
2n + 1
Setting y = 2x, we have
8 2n+1 8
22n+1 x2n+5
ż ż ż
4 (2x )
ÿ ÿ
4 n n
x arctan(2x ) dx = (´1) x dx = (´1) dx
n =0
2n + 1 n =0
2n + 1
8 8
ÿ 22n+1 x2n+6 ÿ 22n x2n+6
= (´1)n +C = (´1)n +C
n =0
(2n + 1)(2n + 6) n =0
(2n + 1)(n + 3)

S-7: Since
8
d 2 ÿ 1
1
f (t) = log(1 + 2t) = =2 (´2t)n if |2t| ă 1 i.e. |t| ă
dt 1 + 2t n =0
2

and f (0) = 0, we have

żx 8 żx 8
ÿ ÿ x n +1 1
f (x) = 1
f (t) dt = 2 n n n
(´1) 2 t dt = (´1)n 2n+1 for all |x| ă
0 n =0 0 n =0
n+1 2

?
2n ?
2n+1
S-8: Since 3n = 3 = ?1 3
3
8 8 8 2n+1 ˇ
(´1)n (´1)n
ˇ
ÿ ? ÿ ? ÿ n x
? 1
= 3 ?
= 3 ( ´1 ) ˇ = 3 arctan ?
n =0
(2n + 1)3n n=0 (2n + 1) 3
2n+1
n =0
2n + 1 x= ?1
ˇ 3
3
? π π
= 3 = ?
6 2 3

219
 8
xn
S-9: Recall that e x =
ř
n! . So
n =0

(´1)n h ÿ x n i h i
8
ÿ 8
= = ex = e´1
n =0
n! n =0
n! x=´1 x =´1

8
xk
S-10: Recall that e x =
ř
k! . So
k =0
ÿ8
1 hÿ8
xk i h i
k
= = ex = e1/e
e k! k! x=1/e x =1/e
k =0 k =0

8
xk
S-11: Recall that e x =
ř
k! . So
k =0
ÿ8
1 hÿ8
xk i h i
k k!
= = ex = e1/π
π k! x=1/π x =1/π
k =0 k =0
This series differs from the given one only in that it starts with k = 0 while the given
series starts with k = 1. So
8 8
ÿ 1 ÿ 1
k
= k
1 on = e1/π ´ 1
´ loomo
k =1
π k! k =0
π k!
k =0

S-12: Recall that

8 8
ÿ x k +1
k
ÿ xn
log(1 + x ) = (´1) = (´1)n´1
k+1 n
k =0 n =1
(To get from the first sum to the second sum we substituted n = k + 1. If you don’t see
why the two sums are equal, write out the first few terms of each.) So
(´1)n´1 h ÿ ni h i
8 8
n´1 x
ÿ
= (´1) = log(1 + x ) = log(3/2)
n 2n n x=1/2 x =1/2
n =1 n =1

S-13: Write
8 8 8
ÿ n+2 n ÿ n n ÿ 2 n
e = e + e
n! n! n!
n =1 n =1 n =1
8 8 n
ÿ en ÿ e
= +2
n =1
( n ´ 1) ! n!
n =1
8 8
ÿ en´1 ÿ en
=e +2
n =1
( n ´ 1) ! n =1
n!
8 n 8 n
ÿ e ÿ e
=e +2
n =0
n! n!
n =1

220
 8
xn
Recall that e x =
ř
n! . So
n =0

8
ÿ n+2 n hÿ8
xn i hÿ
8
xn i h i h i
e =e +2 = e ex + 2 ex ´ 1 = e e +1 + 2 ( e e ´ 1 )
n! n =0
n! x =e n! x =e x =e x =e
n =1 n =1
e
= ( e + 2) e ´ 2

8
1
S-14: Substituting y = ´3x3 into yn gives
ř
1´y =
n =0

8 8
df 1 ÿ

3 n
ÿ
=x = x ´ 3x = (´1)n 3n x3n+1
dx 1 + 3x3 n =0 n =0

Now integrating,
8
ÿ x3n+2
f (x) = (´1)n 3n +C
n =0
3n + 2

To have f (0) = 1, we need C = 1. So, finally

8
ÿ 3n
f (x) = 1 + (´1)n x3n+2
n =0
3n + 2

S-15: (a) Using the geometric series expansion with r = t4 ,

8 8
1 ÿ 1 ÿ n
= r n ùñ 4
= (´t4 )
1´r n =0
1+t n =0

Substituting this into our integral,

żx 8 żx 8 4n+1
1 n x
ÿ ÿ
n 4n
I (x) = dt = (´1) t dt = (´1)
0 1 + t4 n =0 0 n =0
4n + 1

(b) Substituting in x = 12 .
8
ÿ 1
I (1/2) = (´1)n
n =0
(4n + 1)24n+1
1 1 1 1
= 2 ´ + ´ +¨¨¨
5 ˆ 25 9 ˆ 29 13 ˆ 213
= 0.5 ´ 0.00625 + 0.000217 ´ 0.0000094 + ¨ ¨ ¨ = 0.493967 ´ 0.0000094 + ¨ ¨ ¨

See part (c) for the error analysis.

221
(c) The series for I ( x ) is an alternating series (that is, the sign alternates) with
successively smaller terms that converge to zero. So the error introduced by truncating
the series is between zero and the first omitted term. In this case, the first omitted term
was negative (´0.0000094). So the exact value of I (1/2) is the approximate value found
in part (b) plus a negative number whose magnitude is smaller than 0.00001 = 10´5 . So
the approximate value of part (b) is larger than the true value of I (1/2).

S-16: Expanding the exponential using its Maclaurin series,

ż1 8 ż1 2 )n 8
(´1)n 1 2n+4
ż
4 ´x2 4 (´x
ÿ ÿ
I= x e dx = x dx = x dx
0 n =0 0
n! n =0
n! 0
8
ÿ (´1)n 1 1 1 1
= = ´ + ´ +¨¨¨
n =0
n!(2n + 5) 5 on loomo
loomo 7 on loo18 3!(11)
moon loomoon
n =0 n =1 n =2 n =3

The signs of successive terms in this series alternate. Futhermore the magnitude of the
nth term decreases with n. Hence, by the alternating series test, I lies between 51 ´ 17 + 18
1

and 15 ´ 17 + 18
1
´ 3!(111) . So
1 1
|I ´ a| ď =
3!(11) 66

S-17: Expanding the exponential using its Taylor series,

ż 1 8 ż 1
2 )n 8 ż1
(´1)n 2 2n+2
2 (´x
2 2
2 ´x2
ÿ ÿ
I= x e dx = x dx = x dx
0 n =0 0
n! n =0
n! 0
8
ÿ (´1)n 1
=
n =0
n!(2n + 3) 2 +3
2n

The signs of successive terms in this series alternate. Futhermore the magnitude of the
nth term decreases with n. Hence, by the alternating series test, I lies between
N +1
Nř
ř (´1)n 1 (´1)n 1
n!(2n+3) 22n+3
and n!(2n+3) 22n+3
, for every N. The first few terms are, to five decimal
n =0 n =0
places,

n 0 1 2 3
(´1)n 1
n!(2n+3) 22n+3
0.04167 -0.00625 0.00056 -0.00004

Allowing for a roundoff error of 0.000005 in each of these, I must be between

0.04167 ´ 0.00625 + 0.00056 + 0.000005 ˆ 3 = 0.035995

and
0.04167 ´ 0.00625 + 0.00056 ´ 0.00004 ´ 0.000005 ˆ 4 = 0.035920

222
S-18: We use that
8
ÿ yn
log(1 + y) = (´1)n´1 for all ´ 1 ă y ď 1
n
n =1

x´2
with y = 2 to give

 x ´ 2
8
ÿ (´1)n´1
log( x ) = log(2 + x ´ 2) = log 2 + log 1 + = log 2 + ( x ´ 2) n
2 n 2n
n =1

x´2
It converges when ´1 ă y = 2 ď 1, or equivalently, 0 ă x ď 4.

S-19: (a) Using the Taylor series expansion of e x with x = ´t,

8 8 8
ÿ (´t)n ÿ
nt
n e´t ´ 1 ÿ tn´1
e ´t
= ´t
ùñ e ´ 1 = (´1) ùñ = (´1)n
n =0
n! n! t n!
n =1 n =1

Substituting this into our integral,

żx 8 ż x n´1 8
e´t ´ 1 ÿ
n t ÿ xn
I (x) = dt = (´1) dt = (´1)n
0 t 0 n! n n!
n =1 n =1

(b) Substituting in x = 1.
8
ÿ 1
I (1) = (´1)n
n n!
n =1
1 1 1 1
= ´1 + ´ + ´ +¨¨¨
2 2! 3 3! 4 4! 5 5!
= ´1 + 0.25 ´ 0.0556 + 0.0104 ´ 0.0017 + ¨ ¨ ¨ = ´0.80

See part (c) for the error analysis.

(c) The series for I ( x ) is an alternating series (that is, the sign alternates) with
successively smaller terms that converge to zero. So the error introduced by truncating
the series is no larger than the first omitted term. So the magnitude of ´ 515! + ¨ ¨ ¨ is no
larger than 0.0017. Allowing for a roundoff error of at most 0.0001 in each of the two
terms ´0.0556 + 0.0104

I (1) = ´1 + 0.25 ´ 0.0556 + 0.0104 ˘ 0.0019 = ´0.7952 ˘ 0.0019

S-20: (a) Using the Taylor series expansion of sin x with x = t,

8 8
ÿ t2n+1 sin t ÿ t2n
sin t = (´1)n ùñ = (´1)n
n =0
(2n + 1)! t n =0
(2n + 1)!

223
So
żx 8 żx 8
sin t ÿ
n t2n ÿ x2n+1
Σ( x ) = dt = (´1) dt = (´1)n
0 t n =0 0 (2n + 1) ! n =0
(2n + 1)(2n + 1)!

(b) The critical points of Σ( x ) are the solutions of Σ1 ( x ) = 0. By the fundamental theorem
of calculus Σ1 ( x ) = sinx x , so the critical points of Σ( x ) are x = ˘π, ˘2π, ¨ ¨ ¨ . The absolute
maximum occurs at x = π.
(c) Substituting in x = π,
8
ÿ π 2n+1
Σ(π ) = (´1)n
n =0
(2n + 1)(2n + 1)!
π3 π5 π7
= π´ + ´ +¨¨¨
3 3! 5 5! 7 7!
= 3.1416 ´ 1.7226 + 0.5100 ´ 0.0856 + 0.0091 ´ 0.0007 + ¨ ¨ ¨
The series for Σ(π ) is an alternating series (that is, the sign alternates) with successively
smaller terms that converge to zero. So the error introduced by truncating the series is no
larger than the first omitted term. So

Σ(π ) = 3.1416 ´ 1.7226 + 0.5100 ´ 0.0856 + 0.0091 = 1.8525

with an error of magnitude at most 0.0007 + 0.0005 (the 0.0005 is the maximum possible
accumulated roundoff error in all five retained terms).

S-21: (a) Using the Taylor series expansion of cos x with x = t,

8
t2 t4 t6 ÿ t2n
cos t = 1 ´ + ´ + ¨ ¨ ¨ = (´1)n
2! 4! 6! n =0
(2n)!
8
cos t ´ 1 1 t2 t4 ÿ t2n´2
= ´ + ´ +¨¨¨ = (´1)n
t2 2! 4! 6!
n =1
(2n)!
żx 8
cos t ´ 1 x x3 x5 ÿ
n x2n´1
I (x) = dt = ´ + ´ + ¨ ¨ ¨ = ( ´1 )
0 t2 2! 4!3 6!5
n =1
(2n)!(2n ´ 1)

(b), (c) Substituting in x = 1,

1 1 1
I (1) = ´ + ´ +¨¨¨
2 4!3 6!5
= ´0.5 + 0.0139 ´ 0.0003 ´ ¨ ¨ ¨
= ´0.486 ˘ 0.001
The series for I (1) is an alternating series with decreasing successive terms that converge
to zero. So approximating I (1) by ´ 21 + 4!31 1
introduces an error between 0 and ´ 6!5 .
1 1
Hence I (1) ă ´ 2 + 4!3 .

224
S-22: (a) Using the Taylor series expansions of sin x and cos x with x = t,

t2 t4 t6 t8
cos t = 1 ´ + ´ + +¨¨¨
2! 4! 6! 8!
t3 t5 t7
sin t = t´ + ´ +¨¨¨
3! 5! 7!
t4 t6 t8
t sin t = t2 ´ + ´ + ¨ ¨ ¨
3! 5! 7!
 1 2 1 1 4 1 1 6 1 1 8
cos t + t sin t ´ 1 = 1´ t ´ ´ t + ´ t ´ ´ t +¨¨¨
2! 3! 4! 5! 6! 7! 8!
1 2 3 4 5 7 2 1 4 1 6 1 8

= t ´ t + t6 ´ t8 + ¨ ¨ ¨ use 1 = , = , = , = , ¨¨¨
2! 4! 6! 8! 2! 3! 4! 5! 6! 7! 8!
cos t + t sin t ´ 1 1 3 2 5 4 7 6
= ´ t + t ´ t +¨¨¨
t2 2! 4! 6! 8!
1 1 3 1 5 1 8 1 ´ cos x
I (x) = x ´ x + x ´ x +¨¨¨ =
2! 4! 6! 8! x

(b) I (1) = 2!1 ´ 4!1 + 6!1 ´ 8!1 + ¨ ¨ ¨ = 0.5 ´ 0.0416̇ + 0.00139 ´ 0.000024 + ¨ ¨ ¨ = 0.460 . The
error analysis is in part (c).

(c) The series for I (1) is an alternating series with decreasing successive terms that
convege to zero. So approximating I (1) by 2!1 ´ 4!1 + 6!1 introduces an error between 0 and
´ 8!1 . So I (1) ă 2!1 ´ 4!1 + 6!1 ă 0.460.

x2 x3 x4
S-23: (a) Substituting x = ´t into the known power series e x = 1 + x + 2! + 3! + 4! +¨¨¨,
we see that:

t2 t3 t4
e´t = 1 ´ t +´ + ´¨¨¨
2! 3! 4!
t 2 t3 t4
´t
1´e = t´ + ´ +¨¨¨
2! 3! 4!
1 ´ e´t t t2 t3
= 1´ + ´ +¨¨¨
t 2! 3! 4!
´t x2 x3 x4
ż
1´e
dt = C + x ´ + ´ +¨¨¨
t 2 ¨ 2! 3 ¨ 3! 4 ¨ 4!

Finally, f (0) = 0 (since f (0) is an integral from 0 to 0) and so C = 0. Therefore

żx
1 ´ e´t x2 x3 x4
f (x) = dt = x ´ + ´ +¨¨¨ .
0 t 2 ¨ 2! 3 ¨ 3! 4 ¨ 4!

225
 8
xn
We can also do this calculation entirely in summation notation: e x =
ř
n! , and so
n =0

8 8
´t
ÿ (´t)n ÿ (´1)n tn
e = = 1+
n =0
n! n!
n =1
8 8
ÿ (´1)n tn ÿ (´1)n´1 tn
1 ´ e´t =´ =
n! n!
n =1 n =1
8
1 ´ e´t ÿ (´1)n´1 tn´1
=
t n!
n =1
żx 8
1 ´ e´t ÿ (´1)n´1 x n
f (x) = dt =
0 t n ¨ n!
n =1

(´1)n´1 n
(b) We set an = An x n = n¨n! x and apply the ratio test. Since

ˇ (´1)n x n+1 /((n + 1) ¨ (n + 1)!) ˇ

ˇa ˇ ˇ ˇ
ˇ n +1 ˇ
lim ˇ ˇ = lim ˇ
ˇ ˇ
nÑ8 an nÑ8 (´1)n´1 x n /(n ¨ n!) ˇ
 n +1 
|x| n ¨ n!
= lim
nÑ8 |x|n (n + 1) ¨ (n + 1)!
 
n
= lim |x| since (n + 1)! = (n + 1) n!
nÑ8 ( n + 1)2
=0

This is smaller than 1 no matter what x is. So the series converges for all x.

S-24:

x2 x3
ex = 1 + x + + +¨¨¨ ě 1+ x for all x ě 0
2! 3!
ùñ e x ´ 1 ě x
x3 x3
ùñ ď = x2
ex ´ 1 x
ż1 ż1
x3 1
ùñ x
dx ď x2 dx =
0 e ´1 0 3

226
S-25: Using the Maclaurin series expansions of cos x and e x ,
x2 x4
cos x = 1 ´ + +¨¨¨
2! 4!
x2 x4
1 ´ cos x = ´ +¨¨¨
2! 4!
x2 x3
ex = 1+x+ + +¨¨¨
2! 3!
x2 x3
1 + x ´ ex =´ ´ +¨¨¨
2! 3!
x2 x4 1 x2
1 ´ cos x 2! ´ 4! + ¨ ¨ ¨ 2! ´ 4! + ¨ ¨ ¨
= x2 x3 = 1
1 + x ´ ex ´ 2! ´ 3! + ¨ ¨ ¨ ´ 2! ´ 3!x + ¨ ¨ ¨
we have 2
1 x 1
1 ´ cos x 2! ´ 4! + ¨ ¨ ¨ 2!
lim = lim 1 = 1 = ´1
xÑ0 1 + x ´ e x xÑ0 ´ ´ x + ¨ ¨ ¨ ´ 2!
2! 3!

S-26: Using the Maclaurin series expansion of sin x,

x3 x5 x7
sin x = x ´ + ´ +¨¨¨
3! 5! 7!
x3 x5 x7
sin x ´ x + = ´ +¨¨¨
6 5! 7!
3
sin x ´ x + x6 1 x2
= ´ +¨¨¨
x5 5! 7!
we have
sin x ´ x + x3 1 x2  1 1
6
lim = lim ´ +¨¨¨ = =
xÑ0 x5 xÑ0 5! 7! 5! 120

S-27: Using the Maclaurin series expansions of cos x,

x2 x4 x6 x8
cos x = 1 ´+ ´ + ´¨¨¨
2 4! 6! 8!
2 x2 x4 x6 x8
ùñ 1 ´ x ´ cos x = ´ ´ + ´ +¨¨¨
2 4! 6! 8!
1 ´ x2 ´ cos x x´1/2 x3/2 x7/2 x11/2
ùñ = ´ ´ + ´ +¨¨¨
x5/2 2 4! 6! 8!
ż1
1 ´ x2 ´ cos x 2 2 2
ùñ dx = ´1 ´ + ´ +¨¨¨
0 x5/2 5 ˆ 4! 9 ˆ 6! 13 ˆ 8!
This is an alternating series with successive terms decreasing. So the error introduced by
truncating is between 0 and the first term dropped. So
ż1
1 ´ x2 ´ cos x 1 1 1 1 61
5/2
dx = ´1 ´ + ´ + ¨ ¨ ¨ = ´1 ´ +E=´ +E
0 x 60 3240 262080 60 60

227
 1
with the error E between 0 and 3240 ă 0.00031 and ´ 61
60 = ´1.017 with an additional error
of at most 0.00034.

x2n
S-28: (a) The naive strategy is to set an = (2n)!
and apply the ratio test. Since
ˇ 2n+2 ˇ
ˇa ˇ ˇ x ˇ
ˇ n +1 ˇ ˇ (2n+2)! ˇ
lim ˇ ˇ = lim ˇ x2n ˇˇ
ˇ
nÑ8 an nÑ8 ˇ ˇ (2n)!
x2
= lim since (2n + 2)! = (2n + 2)(2n + 1) (2n)!
nÑ8 (2n + 2)(2n + 1)

=0
This is smaller than 1 no matter what x is. So the series converges for all x.
8 8
ÿ xn ÿ (´x )n
Alternatively, the sneaky way is to observe that both ex = ´x
and e =
n =0
n! n =0
n!
are known to converge for all x. So
8
1 x ´x

ÿ xn ÿ x2n
e +e = =
2 n even
n! n =0
(2n)!

also converges for all x.

8
xn
(b) Recall that e x =
ř
n! , Hence
n =0

8
ÿ 1
e=
n =0
n!
8
´1
ÿ (´1)n
e =
n =0
n!
8 8 8
ÿ 1 + (´1)n ÿ 1 ÿ 1
e + e´1 = =2 =2
n =0
n! n even
n! n =0
(2n)!
8
ř 1 1 1


Hence (2n)!
= 2 e+ e .
n =0

8
xn
S-29: (a) We know that e x =
ř
n! for all x. Replacing x by ´x, we also have
n =0
8
ř (´x )n
e´x = n! for all x and hence
n =0

1 h ÿ xn (´x )n i
8 8 8 8
1 x ´x
 ÿ ÿ xn ÿ x2n
cosh( x ) = e + e = + = =
2 2 n=0 n! n=0 n! n =0
n! n =0
(2n)!
n even

for all x. In particular, the interval of convergence is all of R.

228
(b) Using the power series expansion of part (a),
8 8
22 24 ÿ x2n 2 ÿ 22n
cosh(2) = 1 + + + =3 +
2! 4! n=3 (2n)! 3 n=3 (2n)!

ř8 22n 22n
So it suffices to show that n=3 (2n)! ď 0.1. Let’s write bn = (2n)!
. The first term in
ř8 22n
n=3 (2n)! is
26 26 4
b3 = = =
6! 6ˆ5ˆ4ˆ3ˆ2 45
ř8 22n
The ratio between successive terms in n=3 (2n)! is

bn + 1 22n+2 /22n 4 4 1
= = ď = for all n ě 3
bn (2n + 2)!/(2n)! (2n + 2)(2n + 1) 8 ˆ 7 14

Hence
b3 kj
hkkik b4 ď
hkkkikkkj b5 ď
hkkkkikkkkj b6 ď
hkkkkikkkkj
8
ÿ 22n 4 4 1 4 1 4 1 4 1 4 14 56 1
ď + ˆ + ˆ 2+ ˆ 3 +¨¨¨ = 1
= = ă
n =3
(2n)! 45 45 14 45 14 45 14 45 1 ´ 14 45 13 585 10

(c) Comparing
n n n
ÿ8
t2n
8
ÿ ( t2 ) 1 2 ÿ8
( 21 t2 ) ÿ8
( t2 )
cosh(t) = = and e 2t = =
n =0
(2n)! n=0 (2n)! n =0
n! n =0
2n n!

we see that it suffices to show that (2n)! ě 2n n!. Now. for all n ě 1,

n factors
hkkkkkkkikkkkkkkj n factors
hkkkkkkkkkkkkkkkkkikkkkkkkkkkkkkkkkkj
(2n)! = 1 ˆ 2 ˆ ¨ ¨ ¨ ˆ n (n + 1) ˆ (n + 2) ˆ ¨ ¨ ¨ ˆ 2n
n factors
hkkkkkkkikkkkkkkj n factors
hkkkkkkkikkkkkkkj
ě 1ˆ2ˆ¨¨¨ˆn2ˆ2ˆ¨¨¨ˆ2
= 2n n!

229