Series ISSN: 1938-1743

ASHLOCK
Series Editor: Steven G. Krantz, Washington University in St. Louis

Fast Start Integral Calculus

Daniel Ashlock, University of Guelph

FAST START INTEGRAL CALCULUS

The three books in the fast start calculus series were developed over a period of five years with classroom
testing each year at the University of Guelph. The development team included two physics professors and
two senior teaching assistants in math in addition to the author. The books cover dierential and integral
calculus with a selection of topics from advanced calculus including partial derivatives, multiple integrals,
and sequences and series up to Taylor series. The topics track the use of calculus in first year physics and can
be paired with a physics course that provides an anchor for the mathematical concepts. Over the first five
years of use, including three revisions to improve clarity and accuracy, no more than 3% of a class dropped
or flunked the joint math and physics course. This contrasts with a 12%-20% rate in the standard calculus
course. The topics in the course are ordered in a novel fashion, moving difficult theoretical material to the later
part of the first semester, and moving quickly to calculus topics with applications in the physical sciences.
The second in the fast start calculus series, this book focuses on the integral calculus, while also
extending the dierential and integral calculus to parametric, polar, and vector functions. The material on
alternate coordinate systems is encapsulated for easy course design. The text starts with the basic definitions
for integrals and moves rapidly to the fundamental theorem and its applications for finding areas and
solving initial value problems. The second chapter extends calculus to alternate coordinate systems that
arise in the physical sciences and provides additional points of view that simplify problem solving. The text
contains a novel chapter on the qualitative analysis of polynomial functions that emphasizes the analytic
geometry of functions in a manner useful in physical science. This chapter introduces L’Hôpital’s rule and
extends the students’ algebra skills substantially. The book then concludes with two chapters on methods
of integration covering substitution, integration by parts, integration with trig identities, trigonometric
substitution, partial fractions, and practice identifying workable methods for integration problems.

About SYNTHESIS
This volume is a printed version of a work that appears in the Synthesis
Digital Library of Engineering and Computer Science. Synthesis

MORGAN & CLAYPOOL

books provide concise, original presentations of important research and
development topics, published quickly, in digital and print formats.

store.morganclaypool.com
Fast Start Integral Calculus
 Synthesis Lectures on
Mathematics and Statistics
Editor
Steven G. Kranz, Washington University, St. Louis

Fast Start Integral Calculus

Daniel Ashlock
2019

Fast Start Differential Calculus

Daniel Ashlock
2019

Introduction to Statistics Using R

Mustapha Akinkunmi
2019

Inverse Obstacle Scattering with Non-Over-Determined Scattering Data

Alexander G. Ramm
2019

Analytical Techniques for Solving Nonlinear Partial Differential Equations

Daniel J. Arrigo
2019

Aspects of Differential Geometry IV

Esteban Calviño-Louzao, Eduardo García-Río, Peter Gilkey, JeongHyeong Park, and Ramón
Vázquez-Lorenzo
2019

Symmetry Problems. Thne Navier–Stokes Problem.

Alexander G. Ramm
2019

An Introduction to Partial Differential Equations

Daniel J. Arrigo
2017
iv
Numerical Integration of Space Fractional Partial Differential Equations: Vol 2 –
Applicatons from Classical Integer PDEs
Younes Salehi and William E. Schiesser
2017

Numerical Integration of Space Fractional Partial Differential Equations: Vol 1 –

Introduction to Algorithms and Computer Coding in R
Younes Salehi and William E. Schiesser
2017

Aspects of Differential Geometry III

Esteban Calviño-Louzao, Eduardo García-Río, Peter Gilkey, JeongHyeong Park, and Ramón
Vázquez-Lorenzo
2017

The Fundamentals of Analysis for Talented Freshmen

Peter M. Luthy, Guido L. Weiss, and Steven S. Xiao
2016

Aspects of Differential Geometry II

Peter Gilkey, JeongHyeong Park, Ramón Vázquez-Lorenzo
2015

Aspects of Differential Geometry I

Peter Gilkey, JeongHyeong Park, Ramón Vázquez-Lorenzo
2015

An Easy Path to Convex Analysis and Applications

Boris S. Mordukhovich and Nguyen Mau Nam
2013

Applications of Affine and Weyl Geometry

Eduardo García-Río, Peter Gilkey, Stana Nikčević, and Ramón Vázquez-Lorenzo
2013

Essentials of Applied Mathematics for Engineers and Scientists, Second Edition

Robert G. Watts
2012

Chaotic Maps: Dynamics, Fractals, and Rapid Fluctuations

Goong Chen and Yu Huang
2011

Matrices in Engineering Problems

Marvin J. Tobias
2011
 v
The Integral: A Crux for Analysis
Steven G. Krantz
2011

Statistics is Easy! Second Edition

Dennis Shasha and Manda Wilson
2010

Lectures on Financial Mathematics: Discrete Asset Pricing

Greg Anderson and Alec N. Kercheval
2010

Jordan Canonical Form: Theory and Practice

Steven H. Weintraub
2009

The Geometry of Walker Manifolds

Miguel Brozos-Vázquez, Eduardo García-Río, Peter Gilkey, Stana Nikčević, and Ramón
Vázquez-Lorenzo
2009

An Introduction to Multivariable Mathematics

Leon Simon
2008

Jordan Canonical Form: Application to Differential Equations

Steven H. Weintraub
2008

Statistics is Easy!
Dennis Shasha and Manda Wilson
2008

A Gyrovector Space Approach to Hyperbolic Geometry

Abraham Albert Ungar
2008
Copyright © 2019 by Morgan & Claypool

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in
any form or by any means—electronic, mechanical, photocopy, recording, or any other except for brief quotations
in printed reviews, without the prior permission of the publisher.

Fast Start Integral Calculus

Daniel Ashlock
www.morganclaypool.com

ISBN: 9781681736501 paperback

ISBN: 9781681736518 ebook
ISBN: 9781681736525 hardcover

DOI 10.2200/S00946ED1V01Y201908MAS029

A Publication in the Morgan & Claypool Publishers series

SYNTHESIS LECTURES ON MATHEMATICS AND STATISTICS

Lecture #29
Series Editor: Steven G. Kranz, Washington University, St. Louis
Series ISSN
Print 1938-1743 Electronic 1938-1751
Fast Start Integral Calculus

Daniel Ashlock
University of Guelph

SYNTHESIS LECTURES ON MATHEMATICS AND STATISTICS #29

M
&C Morgan & cLaypool publishers
ABSTRACT
This book introduces integrals, the fundamental theorem of calculus, initial value problems,
and Riemann sums. It introduces properties of polynomials, including roots and multiplicity,
and uses them as a framework for introducing additional calculus concepts including Newton’s
method, L’Hôpital’s Rule, and Rolle’s theorem. Both the differential and integral calculus of
parametric, polar, and vector functions are introduced. The book concludes with a survey of
methods of integration, including u-substitution, integration by parts, special trigonometric in-
tegrals, trigonometric substitution, and partial fractions.

KEYWORDS
integral calculus, polynomials, parametric functions, polar functions, vector func-
tions, methods of integration
 ix

Contents
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi

Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii

1 Integration, Area, and Initial Value Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1.1 Anti-derivatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 The Fundamental Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.2.1 Even and Odd Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
1.3 Initial Value Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
1.4 Induction and Sums of Rectangles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

2 Parametric, Polar, and Vector Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

2.1 Parametric Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
2.1.1 The Derivative of a Parametric Curve . . . . . . . . . . . . . . . . . . . . . . . . . 55
2.2 Polar Coordinates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
2.2.1 Polar Calculus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
2.3 Vector Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
2.3.1 Calculus with Vector Curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

3 The Arithmetic, Geometry, and Calculus of Polynomials . . . . . . . . . . . . . . . . . . 91

3.1 Polynomial Arithmetic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
3.2 Qualitative Properties of Polynomials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
3.2.1 Multiplicity of Roots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
3.3 L’Hôpital’s Rule; Strange Polynomials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
3.3.1 Strange Polynomials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113

4 Methods of Integration I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

4.1 u-Substitution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
4.1.1 Substitution in Definite Integrals . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
4.2 Integration by Parts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
4.3 Integrating Trig Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
x
5 Methods of Integration II . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
5.1 Trigonometric Substitution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
5.2 Partial Fractions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
5.3 Practicing Integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176

A Useful Formulas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181

A.1 Powers, Logs, and Exponentials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
A.2 Trigonometric Identities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
A.3 Speed of Function Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
A.4 Derivative Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
A.5 Vector Arithmetic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
A.6 Polar and Rectangular Conversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
A.7 Integral Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185

Author’s Biography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
 xi

Preface
This text covers single-variable integral calculus, presuming familiarity with differential calculus
from the precursor text Fast Start Differential Calculus. The text was developed for a course that
arose from a perennial complaint by the physics department at the University of Guelph that
the introductory calculus courses covered topics roughly a year after they were needed. In an
attempt to address this concern, a multi-disciplinary team created a two-semester integrated
calculus and physics course. This book covers the integral calculus topics from that course as
well a material on the behavior of polynomial function. The philosophy of the course was that
the calculus will be delivered before it is needed, often just in time, and that the physics will serve
as a substantial collection of motivating examples that will anchor the student’s understanding
of the mathematics.
The course has run three times before this text was started, and it was used in draft form
for the fourth offering of the course, and then for two additional years. There is a good deal of
classroom experience and testing behind this text. There is also enough information to confirm
our hypothesis that the course would help students. The combined drop and flunk rate for this
course is consistently under 3%, where 20% is more typical for first-year university calculus. Co-
instruction of calculus and physics works. It is important to note that we did not achieve these
results by watering down the math. The topics covered, in two semesters, are about half again
as many as are covered by a standard first-year calculus course. That’s the big surprise: covering
more topics faster increased the average grade and reduced the failure rate. Using physics as a
knowledge anchor worked even better than we had hoped.
This text, and its two companion volumes, Fast Start Differential Calculus, and Fast Start
Advanced Calculus make a number of innovations that have caused mathematical colleagues to
raise objections. In mathematics it is traditional, even dogmatic, that math be taught in an order
in which no thing is presented until the concepts on which it rests are already in hand. This is
correct, useful dogma for mathematics students. It also leads to teaching difficult proofs to stu-
dents who are still hungover from beginning-of-semester parties. This text neither emphasizes
nor neglects theory, but it does move theory away from the beginning of the course in acknowl-
edgment of the fact that this material is philosophically difficult and intellectually challenging.
The course also presents a broad integrated picture as soon as possible. The text also emphasizes
cleverness and computational efficiency. Remember that “mathematics is the art of avoiding
calculation.”
It is important to state what was sacrificed to make this course and this text work the
way they do. This is not a good text for math majors, unless they get the theoretical parts of
calculus later in a real analysis course. The text is relatively informal, almost entirely example
xii PREFACE
driven, and application motivated. The author is a math professor with a CalTech Ph.D. and
three decades of experience teaching math at all levels from 7th grade (as a volunteer) to graduate
education including having supervised a dozen successful doctoral students. The author’s calculus
credits include calculus for math and engineering, calculus for biology, calculus for business, and
multivariate and vector calculus.

Daniel Ashlock
August 2019
 xiii

Acknowledgments
This text was written for a course developed by a team including my co-developers Joanne M.
O’Meara of the Department of Physics and Lori Jones and Dan Thomas of the Department of
Chemistry, University of Guelph. Andrew McEachern, Cameron McGuinness, Jeremy Gilbert,
and Amanda Saunders have served as head TAs and instructors for the course over the last six
years and had a substantial impact on the development of both the course and this text. Martin
Williams, of the Department of Physics at Guelph, has been an able partner on the physics side
delivering the course and helping get the integration of the calculus and physics correct. I also
owe six years of students thanks for serving as the test bed for the material. Many thanks to
all these people for making it possible to decide what went into the text and what didn’t. I also
owe a great debt to Wendy Ashlock and Cameron McGuinness at Ashlock and McGuinness
Consulting for removing a large number of errors and making numerous suggestions to enhance
the clarity of the text.

Daniel Ashlock
August 2019
 1

CHAPTER 1

Integration, Area, and Initial

Value Problems
The two main concepts in calculus are derivatives and integrals. We assume you have already
studied derivatives. This chapter introduces the concept of integrals. Derivatives and integrals
were developed independently, and later it was discovered that they are closely related. We will
start with this relationship and define integrals as a sort of backward derivative. Then, we will in-
troduce some of the primary uses of integration – finding areas under curves and solving bound-
ary value problems. Finally, we will show how integrals were originally defined. This chapter does
not cover how to integrate most functions. That we’ve saved for later in Chapter 4.

1.1 ANTI-DERIVATIVES
The anti-derivative of a function is another function whose derivative is the function you started
with. More precisely:
Definition 1.1 If f .x/ and F .x/ are functions so that

F 0 .x/ D f .x/;

we say that F .x/ is an anti-derivative of f .x/.

The terminology anti-derivative is fairly modern. The original terminology is to call an anti-
derivative of f .x/ an integral of f .x/. Integrals also have a special notation.
Knowledge Box 1.1

Integral notation
If F .x/ is an anti-derivative of f .x/, we write
Z
F .x/ D f .x/
dx

and call F .x/ an integral of f .x/.

2 1. INTEGRATION, AREA, AND INITIAL VALUE PROBLEMS
Example 1.1 Find an anti-derivative of f .x/ D 2x .

Solution:

We know that the derivative of F .x/ D x 2 is f .x/ D 2x . So, one possible answer is:

F .x/ D x 2

˙
One of the issues that arises when we work with anti-derivatives is that they are not unique.

The derivative of
G.x/ D x 2 C 5

is also 2x. This means that anti-derivatives (integrals) are known only up to some constant value.
We will develop techniques for dealing with this in Section 1.3, but for now we will simply use
unknown constants or constants of integration.

Example 1.2 Use integral notation, with an unknown constant C , to represent all the
anti-derivatives of 2x .

Solution:
Z
2x
dx D x 2 C C

We also need to explain dx . The job of dx is to tell us which symbol is the variable.

Definition 1.2 The symbol dx is spoken “the differential of x ” and is used to designate the active
variable in an integral.

dy
We have seen dx and other differentials before. The symbol was an alternate way of saying
dx
0
y , for example. There is another use of differential symbols – as an operator.
 1.1. ANTI-DERIVATIVES 3
Knowledge Box 1.2

The differential operator

d
Another way to say “take the derivative” is with the symbol .
dx
This symbol denotes the derivative with respect to x . So:
d
f 0 .x/ D f .x/
dx

We will need this symbol, the differential operator, in the next section.
Z
d 2
In order to compute 2x
dx , we used the fact that we knew that x D 2x . In fact, each
dx
derivative rule is also an anti-derivative rule, just used backward. Let’s start by giving the reverse
of the power rule for derivatives.
Knowledge Box 1.3

The power rule for integration

Z
1
x n
dx D x nC1 C C
nC1

The rules for constant multiples and sums for derivatives also apply for integrals.
Knowledge Box 1.4

Integrals of constant multiples and sums

Z Z
a
f .x/
dx D a
f .x/
dx

Z Z Z
.f .x/ C g.x//
dx D f .x/
dx C g.x/
dx

This gives us enough machinery to be able to compute the integrals of polynomial functions.
4 1. INTEGRATION, AREA, AND INITIAL VALUE PROBLEMS
Example 1.3 Find Z


x 3 C 4x 2 C 5x C 3
dx

Solution:

Z Z Z Z Z
3 2

3 2 1
x C 4x C 5x C 3
dx D x
dx C 4 x
dx C 5 x
dx C 3 dx

1 4 1 1
D x C 4
x3 C 5
x2 C 3
x C C
4 3 2

1 4 4 3 5 2
D x C x C x C 3x C C
4 3 2

Notice that an anti-derivative of a constant a is ax and that, since the sum of several unknown
constants is some other unknown constant, we can get away with one “+C”.
To check an integral rule, you take the derivative of the result and see if you get back where you
started.

Example 1.4 Check the preceding example by taking the derivative.

Solution:
 0
1 4 4 3 5 2 1 3 4 2 5 1
x C x C x C 3x C C D 4x C 3x C 2x C 3.1/ C 0 D x 3 C 4x 2 C 5x C 3
4 3 2 4 3 2
and we see the integral in the last example was correct.

All the derivative rules from Fast Start Differential Calculus have corresponding integral rules.
Let’s go through them.
 1.1. ANTI-DERIVATIVES 5
Knowledge Box 1.5

Integrals of logs and exponentials

Z
1

dx D ln.x/ C C
x
Z
ex
dx D ex C C

One problem with the laundry-list of integrals in this section is that, until we get the techniques
in Chapter 4, the integrals we can do are a bit contrived. Nevertheless, let’s do an example.

Example 1.5 Compute

Z  
4
3ex C
dx
x

Solution:

Z   Z Z
x 4 x dx
3e C
dx D 3 e
dx C 4
x x

D 3ex C 4 ln.x/ C C

Next are the integrals arising from the trigonometric derivatives.

6 1. INTEGRATION, AREA, AND INITIAL VALUE PROBLEMS
Knowledge Box 1.6

Integrals of trigonometric derivatives

Z
sin.x/
dx D cos.x/ C C
Z
cos.x/
dx D sin.x/ C C
Z
sec2 .x/
dx D tan.x/ C C
Z
csc2 .x/
dx D cot.x/ C C
Z
sec.x/ tan.x/
dx D sec.x/ C C
Z
csc.x/ cot.x/
dx D csc.x/ C C

In the next example, the integral we are asked to perform doesn’t look like one of our known
forms, but it can be rearranged into one of the known forms with a few trig identities.

Example 1.6 Compute

Z
sin.x/

dx
cos2 .x/

Solution:

Z Z
sin.x/ 1 sin.x/

dx D

dx
cos2 .x/ cos x cos.x/
Z
D sec.x/
tan.x/
dx

D sec.x/ C C

˙
 1.1. ANTI-DERIVATIVES 7
Trigonometric identities can be used to make really easy integrals look really hard.

Example 1.7 Compute

Z


sin2 .x/ C 2 sin.x/ C cos2 .x/
dx

Solution:

Z Z Z
2 2

2 2


sin .x/ C 2 sin.x/ C cos .x/
dx D sin .x/ C cos .x/
dx C 2 sin.x/
dx

Z Z
D 1
dx C 2 sin.x/
dx

D x C 2. cos.x// C C

Dx 2 cos.x/ C C
˙

We also have some integrals arising from the inverse trigonometric functions.

Knowledge Box 1.7

More integrals of trigonometric derivatives

Z
1
p
dx D sin 1 .x/ C C
1 x 2
Z
1
2

dx D tan 1 .x/ C C
Z 1 C x
1
p
dx D sec 1 .jxj/ C C
x x 2 1

This example is another one in which we can set up a known form with a little bit of algebra.
Example 1.8 Compute Z  
1

dx
3x 2 C 3
8 1. INTEGRATION, AREA, AND INITIAL VALUE PROBLEMS
Solution:

Z Z  
1 1 1

dx D
dx
3x 2 C 3 3 x2 C 1
Z
1 dx
D
3 1 C x2

1 1
D tan .x/ C C
3

It is quite common for integrals to require some algebraic setup before it becomes apparent
how to do them. Let’s do another.

Example 1.9 Compute

Z  
x2

dx
1 C x2
Solution:

Z Z
x2 1 C x2 1

dx D
dx
1 C x2 1 C x2
Z Z
1 C x2 dx
D
dx
1 C x2 1 C x2
Z Z
dx
D dx
1 C x2

1
Dx tan .x/ C C

˙
 1.1. ANTI-DERIVATIVES 9
Example 1.10 Compute p
Z
x 1 x2 C 1
p dx
1 x2
Solution:

Z p Z p !
x 1 x2 C 1 x 1 x2 1
p dx D p Cp dx
1 x2 1 x2 1 x2
Z  
1
D xCp dx
1 x2

1 2 1
D x C sin .x/ C C
2
˙

All of the integrals we’ve done in this section lead to formulas, not numbers. That will change
in the next section. But before we’re done, let’s get the terminology for these “formula only”
integrals.
Definition 1.3 The integral formulas given in this section of the form:
Z
f .x/
dx D F .x/ C C

are called indefinite integrals in honor of the unknown constant C .

One way to think of indefinite integrals is they store patterns that we will use in the integrals
that compute specific quantities.
The integrals presented in this section are all simple anti-derivatives, except for the ones that
have been lightly disguised by the use of algebra. The next three sections explore the applications
and origins of integration without expanding the tool set for actually performing integration.
Chapter 4 develops u-substitution, integration by parts, partial fractions, and many other clever
methods of doing integrals. It also develops some methods based on trig identities that turn
calculus into a puzzle solving activity.
First introducing integration and later presenting the starter tool-kit is something that is done
to hand you integration as a useful tool for physics as soon as possible. The later exploration
of integration techniques is mind-expanding, but the fundamental concept that permits you to
manipulate the formulas that describe natural law is more important in preparing you to study
physics.
10 1. INTEGRATION, AREA, AND INITIAL VALUE PROBLEMS
PROBLEMS

Problem 1.11 Find the indefinite integral of each of the following polynomial functions.

1. f .x/ D x 2 C 1 5. s.x/ D 7x 7 C 6x 6 C x 2

2. g.x/ D .x C 1/2 6. q.x/ D .x C 1/.x C 2/.x C 3/

3. h.x/ D x 4 3x 3 C 5x 2 7x C 6 7. a.x/ D .x 2 C x C 1/3

x3 1 8. b.x/ D .x C 2/5
4. r.x/ D
xC1

Problem 1.12 Perform each of the following integrals. Hint: for many of these, trig identites
may be a huge help.

Z Z  
x4 1 sin.x/ sec.x/
1. dx 5. C dx
x2 C 1 cos2 .x/ cos.x/
Z
Z 6. sin2 .x/ C cos2 .x/C
x 2 C 3x C 5
2. dx sin.x/ C cos.x/ C 1/
dx
x

Z Z
dx 7. 2 sin.x=2/ cos.x=2/
dx
3.
cos2 .x/
Z 
Z 
dx 8. sin x C
dx
4. 2
3
sin .x/ C 4 C 5x 2 C cos2 .x/

Problem 1.13 For each of the following statements, verify that the integral is correct, by taking
a derivative.

Z
2 2x
1. ln.x C 1/ C C D
dx
x2 C1
Z
1
2. sin.2x/ C C D cos.2x/
dx
2
 1.1. ANTI-DERIVATIVES 11
Z
3. ln.sec.x// C C D tan.x/
dx

Z
1
5
4. x2 C 1 C C D x
.x 2 C 1/4
dx
10
Z
5. ln.sec.x/ C tan.x// C C D sec.x/
dx

Z
x
6. .x 1/ e C C D xex
dx
  Z
1 x 3 dx
7. ln CC D
2 x 1 x24x C 3
Z
8. .x 3 3x 2 C 6x 6/e x C C D x 3 e x
dx

Problem 1.14 Verify that

Z
dx 1
D ln.ax C b/ C C
ax C b a

Problem 1.15 Verify that Z

1 ax
eax D e CC
a

Problem 1.16 Verify that

Z
1
cos.ax C b/ D sin.ax C b/
a

Problem 1.17 Verify that

Z
1
sec2 .ax C b/ D tan.ax C b/
a
12 1. INTEGRATION, AREA, AND INITIAL VALUE PROBLEMS
Problem 1.18 Demonstrate that if p.x/ is a polynomial, then we can compute
Z
p.x/

dx
xk

with the techniques in this section.

Problem 1.19 Compute

Z
x2 C 2

dx
x2 C 1

Problem 1.20 Compute

Z p
1 x2

dx
.1 x/.1 C x/

1.2 THE FUNDAMENTAL THEOREM

Before we can formally state the relationship between the integral and the derivative, we need
to define the definite integral.
Z
Definition 1.4 Suppose that F .x/ D f .x/
dx . In other words, F .x/ is an anti-derivative of
f .x/. Then the definite integral from x D a to x D b of f .x/ is defined to be:
Z b
f .x/
dx D F .b/ F .a/
a

One nice thing about the definite integral is that it removes the unknown constant. If we write
F .x/ C C for the anti-derivative, then

.F .b/ C C / .F .a/ C C / D F .b/ F .a/ C C C D F .b/ F .a/

With the definite integral defined we can now state the first form of the fundamental theorem.
 1.2. THE FUNDAMENTAL THEOREM 13
Knowledge Box 1.8

The First Fundamental Theorem of Calculus

Z x
d
f .t/
dt D f .x/
dx a
for any constant a.

This form of the fundamental theorem tells us that the derivative of the integral of a function
is the same function, although the variable of integration (t in the above statement) may be
different from the variable appearing in the final expression.

Example 1.21
Z x
d
.t 2 C 1/
dt D x 2 C 1
dx 0

The integral of the derivative is also the same function – almost. The ubiquitous unknown
constant causes us to answer: except for the “+ C.” Later, we will see that this unknown constant
is where we place the starting point (position, velocity, etc.) into the formula when solving an
applied problem.

The second form of the fundamental theorem has more applications. It relates integrals to the
area under the graph of a function.

Knowledge Box 1.9

The Second Fundamental Theorem of Calculus

Suppose that f .x/  0 on the interval [a,b]. Then, if A is the area under
the graph of f .x/ between a and b ,
Z b
AD f .x/
dx
a
14 1. INTEGRATION, AREA, AND INITIAL VALUE PROBLEMS
Example 1.22 Find the area A under the curve of y D x 2 between x D 0 and x D 3.

Solution:
Guided by the picture, compute the definite integral.
Start with a picture:

9 Z ˇ
3
1 3 ˇˇ3
x 2
dx D x ˇ
0 3 0

1 3 1 3
D 3 0
3 3

1 1
D 27 0
3 3

A D9 0

D 9 units2

-1 4

-1

Notice the vertical bar notation, used to hold the limits until we plug them into the anti-
derivative.

At this point let’s check the intuition on this one. Why would the anti-derivative of a function
be the area under it? The first form of the fundamental theorem tells us that a function is the
derivative of its integral – but that means that a function is the rate of change of its integral.
The larger a function is, the faster the area under it is changing. The smaller a function is, the
slower the area under it is changing. So, a function is the rate of change of the area under the
function. If you’re unconvinced, wait for Section 1.4. We will use another approach to show
that the integral gives the area under the curve.
 1.2. THE FUNDAMENTAL THEOREM 15
What meaning does the restriction f .x/  0 in the second fundamental theorem have? The
short answer is: the area below the x axis, for which f .x/  0, comes out negative. This actually
makes sense if we remember that the derivative is a rate of change. Positive derivatives represent
increases, negative ones represent decreases. Since there is no such thing as negative area, we
have to be careful when computing the total area between a graph and the x -axis.

Example 1.23 Compare the definite integral and the area between the curve and the x -axis
for f .x/ D x 2 1 from x D 1 to x D 2.

Solution:

This picture shows the areas above and below the x -axis.

+
-2 2
-

-2

Notice the points where f .x/ crosses the axis are x D ˙ 1.

16 1. INTEGRATION, AREA, AND INITIAL VALUE PROBLEMS
First the integral:
Z 2
ˇ2

1 ˇ
x 2
1
dx D x 3 x ˇˇ
1 3 1

 
1 1
D 23 2 . 1/3 . 1/
3 3

8 1
D 2C 1
3 3

9
D 3
3

D0

So, even though they are different shapes, the areas above and below the curve are equal. Now
we need to compute the areas separately and take the positive area minus the “negative” one:

Z 2 Z 1  ˇ2  ˇ1


1 3 ˇ 1 3 ˇ
x 2
1
dx x 2
1
dx D x x ˇˇ x x ˇˇ
1 1 3 1 3 1

   
8 1 1 1
D 2 C1 1C 1
3 3 3 3

7 2
D 1 C2
3 3

8
D units2
3

Notice that the integral is a number and so has no units, while the area between the curve and
the x -axis has Cartesian units squared as its units. It is very important to keep clear in your mind
the context in which you are using an integral. The meaning of the result is different for different
procedures.
 1.2. THE FUNDAMENTAL THEOREM 17
Also notice that we could have found the total area as
Z 2
AD2 .x 2 1/
dx
1

once we knew the areas above and below the curve were equal. Since it’s not obvious until after
you do the integral, this isn’t all that useful in this case. We will look at a useful version of this
phenomenon later.

1
Example 1.24 Find the area under y D from x D 1 to x D 4.
x
Solution:

3
A

1
yD
x
A
-1 4
xD1

-1

Z 4
ˇ4
1 ˇ

dx D ln.x/ ˇˇ
1 x 1

D ln.4/ ln.1/

D ln.4/ 0

D ln.4/

˙
18 1. INTEGRATION, AREA, AND INITIAL VALUE PROBLEMS
This example pays off on explaining a mystery – why we use e Š 2:71828 : : : as the base of the
1
“natural” logarithm. It is because the area under y D from x D 1 to x D a is ln.a/. This gives
x
us a method of computing logs, and it shows a place where logarithms arise naturally from the
rest of mathematics. A much better way to choose a base than “we have ten fingers.”

Example 1.25 Find the area bounded by y D 4 x 2 and the x -axis.

Solution:

4 yD4 x2

-3 3
xD 2 xD2

-1

This problem does not give explicit limits – instead it tells us what the objects bounding the area
are. Since the x -axis is where y D 0, we need to solve 0 D 4 x 2 , which gives us x D ˙2 as the
points at the left and right end of the area. This then gives us a nice everything-above-the-axis
integral.
 1.2. THE FUNDAMENTAL THEOREM 19

Z 2

AD 4 x 2
dx
2

ˇ
1 3 ˇˇ2
D 4x x ˇ
3 2

1 1
D 4.2/ 8 4. 2/ C . 8/
3 3

8 8
D8C8
3 3
 
2
D8 2
3

4
D8

3

32
D units2
3

Finding the area bounded by two different curves requires solving for those x where

curve 1 D curve 2

in order to find the limits of integration. This will come up a lot in the future. This problem can
be rephrased as curve 1 curve 2 D 0 making it a root finding problem.
20 1. INTEGRATION, AREA, AND INITIAL VALUE PROBLEMS
p
Example 1.26 Find the area bounded by y D x 2 and y D x.

Solution:
2 y D x2

p
yD x

-1 2

-1
p
This one is a little tricky. The picture shows us that if we solve x 2 D x to get the points where
the area begins and ends, then those points are (0,0) and (1,1). The area we want is shown in
Z 1
p 2
p
dark gray. It is the area under y D x that is not under y D x . If we computed x dx ,
0
we would get the dark gray area and the light gray area. The light gray area is the area under
y D x 2 . This means the dark gray area is:
Z 1 Z 1
p
x dx x 2
dx
0 0
Compute
Z 1 Z 1 Z 1  
p
x dx x 2
dx D x 1=2 x 2
dx
0 0 0

ˇ
2 1 3 ˇˇ1
D x 3=2 x
3 3 ˇ0

2 1
D 0C0
3 3

1
D units2
3
˙
 1.2. THE FUNDAMENTAL THEOREM 21
One thing that might be a little tricky in this example is the slightly odd version of the power
rule:

Z
2 3=2
x 1=2
dx D x CC
3

1 3 1 2
All that is going on is that C 1 D and D .
2 2 3=2 3

To find the area bounded by the curves we first found their intersections and then subtracted
the area under the lower curve from the area under the upper curve. Another point of view
on this is that we integrated the upper curve minus the lower curve. Let’s formulate this as a rule.

Knowledge Box 1.10

Area between two curves

If f .x/ > g.x/ on an interval [a,b], then the area between the
graphs of the two functions on that interval is:
Z b Z b Z b
f .x/
dx g.x/
dx D .f .x/ g.x//
dx
a a a

This rule is handy, but it does leave you with the problem of finding the appropriate intervals.
22 1. INTEGRATION, AREA, AND INITIAL VALUE PROBLEMS
Example 1.27 Find the area between y D x 2 and y D x C 2.

Solution:

5 y D x2

-3 3
y DxC2

-1

Solving

x 2 D x C 2;

or equivalently

x2 x 2 D 0;

we see that .x 2/.x C 1/ D 0. So x D 1; 2, making the intersection points . 1; 1/ and .2; 4/.
This also gives us the limits of integration and so
 1.2. THE FUNDAMENTAL THEOREM 23

Z 2

AD xC2 x 2
dt
1

Z 2

D 2Cx x 2
dt
1

ˇ
1 1 3 ˇˇ2
D 2x C x 2 x
2 3 ˇ 1

4 8 1 1
D4C C2
2 3 2 3

24 C 12 16 C 12 3 2
D
6

D 27=6

D 9=2 units2
˙

Let’s do an example that moves beyond polynomial functions.

Example 1.28 Find the area under the sine function from x D 0 to x D  .

Solution:

y D sin.x/
1

A
xD0 xD
-1 4

-1
24 1. INTEGRATION, AREA, AND INITIAL VALUE PROBLEMS
The sine function is non-negative on the entire interval so we may compute

Z 
AD sin.x/
dx
0

ˇ
ˇ
D cos.x/ˇˇ
0

D cos./ C cos.0/

D . 1/ C 1

D 2 units2

1.2.1 EVEN AND ODD FUNCTIONS

Even and odd functions have some useful properties, relative to integration. These definitions
appear in the book on the differential calculus, but it is worth repeating them here.

Definition 1.5 A function is even if, for x where the function exists,

f .x/ D f . x/:

A good example of an even function is f .x/ D x 2 . Even functions forget signs.

Definition 1.6 A function is odd if, for x where the function exists,

f . x/ D f .x/:

A good example of an odd function is f .x/ D x 3 . Odd functions remember signs.

It turns out that we can save some effort when integrating these functions, sometimes, because
of special geometric properties of these functions.
 1.2. THE FUNDAMENTAL THEOREM 25
Knowledge Box 1.11

Integrating an even function on a symmetric interval

If f .x/ is an even function, then
Z a Z a
f .x/
dx D 2 f .x/
dx
a 0

Example 1.29 Find the integral of f .x/ D x 2 on Œ 2; 2 .

Solution:

Remember that x 2 D . x/2 . So f .x/ is an even function.

4
f .x/ D x 2

-2 2

-1

The area on either side of the y -axis is the same so:

Z 2 Z 2 ˇ
2 2 1 3 ˇˇ2 16
x
dx D 2 x
dx D 2
x ˇ D units2
2 0 3 0 3
Not having to plug in the negative number avoids chances to make arithmetic mistakes.

˙
26 1. INTEGRATION, AREA, AND INITIAL VALUE PROBLEMS
Knowledge Box 1.12

Integrating an odd function on a symmetric interval

If f .x/ is an odd function, then
Z a
f .x/
dx D 0
a

Example 1.30 Find the integral of g.x/ D sin.x/ on Œ 2; 2 .

Solution:

Remember that sin.x/ D sin. x/. So g.x/ is an odd function.

1
g.x/ D sin.x/

-2 2

-1

The area on either side of the y -axis is the same, but half is above the x axis, and half is below
the x -axis. So:
Z 2
sin.x/
dx D 0
2

Wow, is that easier than plugging in numbers.

˙
 1.2. THE FUNDAMENTAL THEOREM 27
The odd and even function results are fairly special purpose. One of the big applications of odd
and even function is writing test questions that can be done much faster if the student notices
something is an odd function on a symmetric interval, for example. Symmetric intervals do also
show up in some application problems. If you’re good at shifting functions sideways, you can
also extend the application of these rules.
The set of even functions and the set of odd functions are closed under addition and also under
multiplication by a constant. This fact is occasionally useful. If you’re feeling ambitious, try and
prove it’s true. Also remember that you can break up an integral into a symmetric piece and
leftovers, to change a definite integral into a possibly simpler integral. Finally, if you see an
integral that you have no idea how to do, consider the possibility that the odd function shortcut
gives you a way of solving the problem without integrating.

PROBLEMS

Problem 1.31 Simplify each of the following. Assume Snerp.x/ has a domain of . 1; 1/.

Z x Z x
d t d es C 1
1. 2

dt 4.
ds
dx 1 t C1 dx 6 es 1
Z x Z x
d d ey
2. cos.6t C 1/
dt 5.
dy
dx  dx 2 ln.y/ C 4
Z x Z x
d 2 d
3. tet
dt 6. Snerp.t C 2/
dt
dx 0 dx 0

Problem 1.32 Compute the following.

Z 1 Z 4


1. x 2 C x C 1
dx 4. x n
dx
1 0
Z =2 Z 1
2. cos.x/
dx 5. .x C ex 2/
dx
=2 0
Z Z p
3
3
1 2 dx
3. x C 1
dx 6.
1 x 0 x2 C 1
28 1. INTEGRATION, AREA, AND INITIAL VALUE PROBLEMS
Problem 1.33 Compute
Z 5

x 5 C 3x 3 C 7x 2 x C 1
dx
5

Problem 1.34 Compute the area between the following curves and the x axis on the stated
interval. Be careful, some of these have area above and below the x -axis. Pictures may help.

p p
1. f .x/ D x 2 on Œ 1; 1  4. r.x/ D tan 1 .x/ on Œ 1= 3; 1= 3 
2. g.x/ D x 3 on Œ 1; 1  5. s.x/ D e x
on Œ0; 5 

3. h.x/ D cos.x/ on Œ =2; 3=2 6. q.x/ D 4 x 2 on Œ 3; 3 

Problem 1.35 Compute the area bounded by the specified curves. You will need to know which
curve is higher. Again: pictures may help.

1. y D x 2 and y D x 3 4. y D x 2 and y D 4x 3

2. y D x 2 and y D 6 x 5. y D x 2 and y D 9 x2

3. y D sin.x/ and y D cos.x/ on Œ 0; 3=2 6. y D x 3 and y D 4x

Problem 1.36 Find the area bounded by

y D x n and y D x m

for all positive whole numbers m < n. There will be different categories of answer based on
whether n and m are even or odd. That’s four categories: CC, C , C, and .

Problem 1.37 Compute the following definite integrals.

Z 3

1. x 5 C x 3 C x C 1
dx
3
Z 2
2. .cos.x/ C sin.x//
dx
0
 1.2. THE FUNDAMENTAL THEOREM 29
Z 1
3. x 6 C x 5 C x 4 C x 3 C x 2 C x C 1/
dx
1
Z p
3 3
1
4. p tan .x/
dx
3 3
Z 2
5. .sin.x/ C sin.2x/  C sin.3x//
dx
0
Z a
6. x 2n
dx
a
Z a
7. x 2nC1
dx
a
Z a
1


8. sin.x/ C tan .x/C x 5
dx
a

Problem 1.38 Classify each of the following functions as being odd, even, or neither.

1. y D sin.x/ x 2 =2
6. y D e
2. y D cos.x/
7. y D x 2 C x C 1
1
3. y D tan .x/
8. y D sin.x 2 /, and
4. y D ln.x 2 C 1/
5. y D x
ln.x 2 C 1/ 9. p.x 2 /, where p.x/ is any polynomial.

Problem 1.39 Suppose that f .x/ is an even function. Prove that y D x
f .x/ is an odd func-
tion.

Problem 1.40 Suppose that f .x/ is an odd function. Prove that y D x
f .x/ is an even func-
tion.

Problem 1.41 Suppose that f .x/ is a function and g.x/ is an even function. Is f .g.x// an
even function? Demonstrate your answer is correct.

Problem 1.42 Find b when Z b

x 2
dx D 14
0
30 1. INTEGRATION, AREA, AND INITIAL VALUE PROBLEMS
Problem 1.43 Find b when
Z b
x 4
dx D 1
0

Problem 1.44 If
Z 
sin.x/
dx D 2;
0

what is the smallest possible value for  ?

Problem 1.45 Find a constant c so that the area bounded by y D c and y D x 2 is exactly
4 units2 .

Problem 1.46 Compute the slope m so that the area bounded by the curves y D x 3 and y D
mx is exactly 4 units.

Z b
Problem 1.47 Find the largest possible value of cos.x/
dx .
a

Problem 1.48 Compute

Z 5

x 3
ln x 2 C 1 dx
5

Problem 1.49 Suppose

Z x
2
F .x/ D et
dt
0

What is F 0 .x/?
 1.3. INITIAL VALUE PROBLEMS 31
1.3 INITIAL VALUE PROBLEMS
In this section we come to grips with the constant of integration and figure out what its value is.
This requires that we have a bit of additional information. Our motivating example is to build
up the position function s.t/ from the acceleration function a.t / in steps, with the velocity
function v.t / as an intermediate object. The mathematical model of motion in one dimension
under constant acceleration is:
1 2
s.t / D at C v0 t C s0
2
In English, the distance an object is from a reference point is equal to half the acceleration
times the time squared plus the initial velocity times the time plus the initial distance from that
reference point. If we break this into integrals we get:

Z t
v.t / D a
dx
t0

D a
.t t0 / C C

v.t0 / D a
0 C C

v0 D C
So the constant of integration when we transform acceleration into velocity is the initial velocity.
Similarly:
Z t
s.t / D v
dx
t0

D v
.t t0 / C C

s.t0 / D v
0 C C

s0 D C

The constant of integration for velocity is initial distance. This shows how the constant of inte-
gration can be solved for if we know the initial value of the quantity we are calculating. In fact,
all we need is the value of the thing we are calculating anywhere in the interval we are integrating
on. The initial value just has neater algebra.
32 1. INTEGRATION, AREA, AND INITIAL VALUE PROBLEMS
Example 1.50 Suppose we fire a cannonball directly upward with a velocity of 120 m/sec with
a gravitational acceleration of 10 m/sec2 . If the cannon is at a height of 160 m above sea level,
find an expression for the distance above sea level of the cannonball after the cannon is fired at
t D 0.

Solution:

Plug into the equation of motion given above.

1
s.t / D 10t 2 C 120t C 160 D 160 C 120t 5t 2
2

Since we have a model for this situation, the calculus is all done. Let’s look at a situation where
we need calculus.

Example 1.51 Suppose a missile has an acceleration that builds gradually so that
a.t/ D 100 C 0:1t . If it is launched from a fixed position with a charge that gives it an
initial velocity of 20m/sec, find an expression for the distance the missile has traveled t seconds
after launch and find its position and velocity after t D 20 sec when its fuel runs out.

Solution:

First we find the velocity function. Assume that we start at t D 0.

Z
v.t/ D a.t /
dt

Z
D .100 C 0:1t /
dt

D 100t C 0:05t 2 C C
 1.3. INITIAL VALUE PROBLEMS 33
Now solve for C :

20 D v.0/

20 D 100.0/ C 0:05.0/ C C

20 D C

And we obtain:

v.t/ D 0:05t 2 C 100t C 20

This gives us the velocity after t D 20 sec: V .20/ D 2040 m/sec. Now we find the distance func-
tion.
Z
s.t / D v.t/
dt

Z


D 0:05t 2 C 100t C 20
dt

1 3
D t C 50t 2 C 20t C C
60

As before...

s.0/ D 0 C 0 C 0 C C D 0

C D0

So...
1 3
s.t/ D t C 50t 2 C 20t
60

This is the expression for the the distance the missile has traveled t seconds after launch, and its
position after t D 20 sec is s.20/ Š 20; 533 m.

˙
34 1. INTEGRATION, AREA, AND INITIAL VALUE PROBLEMS
Example 1.52 Suppose that
Z


f .x/ D 3x 2 C 1 dx

and we know that f .2/ D 3. Find an expression for f .x/ with no unknown constants.

Solution:

Z


f .x/ D 3x 2 C 1 dx

D x3 C x C C

Now use the added information.

f .2/ D 23 C 2 C C

3 D 10 C C

7DC

Combine

f .x/ D x 3 C x 7

Knowledge Box 1.13

Solving initial value problems

When a function resulting from integration has the form
f .x/ C C;
an additional piece of information is needed to determine a value
for C .
 1.3. INITIAL VALUE PROBLEMS 35
Z
Example 1.53 Suppose that g.x/ D ex
dx . For g.1/ D 12:2, find an expression for g.x/
that does not involve any unknown constants.

Solution:

The integral is trivial – ex is its own integral – and so

g.x/ D ex C C

Plug in the additional information and solve.

g.1/ D 12:2

e1 C C D 12:2

C D 12:2 e

C Š 9:48

If we need to do more than one integral, we will need one piece of added information per
unknown constant that arises.

Example 1.54 Suppose that the second derivative of h.x/ is h00 .x/ D 1:2x 1. For h.0/ D 4
and h0 .2/ D 2, find an expression for h.x/ that is free of unknown constants.

First we find the first derivative of h.x/:

Z
0
h .x/ D .1:2 x 1/dx

D 0:6x 2 x C C1
36 1. INTEGRATION, AREA, AND INITIAL VALUE PROBLEMS
Plug in the value of the derivative:

2 D h0 .2/

D 0:6.2/2 2 C C1

D 0:4 C C1

1:6 D C1

h0 .x/ D 0:6x 2 x C 1:6

Now we move on to the function itself:

Z
h.x/ D .0:6x 2 x C 1:6/dx

D 0:2x 3 0:5x 2 C 1:6x C C2

4 D h.0/

4D0C0C0CC

4DC

h.x/ D 0:2x 3 0:5x 2 C 1:6x C 4

And we have our answer. Notice that, since two unknown constants appeared, we gave them
different names: C1 and C2 . Also notice that it is much easier to deal with added information
or initial conditions that happen at time zero.

˙
 1.3. INITIAL VALUE PROBLEMS 37
PROBLEMS

Problem 1.55 Assume we are describing the upward motion of a projectile. Given the constant
acceleration, initial velocity, and initial position, determine the greatest height of the projectile
and the time it has that height.

1. a D 5 m/s2 , v0 D 20 m/sec, and s0 D 5 m

2. a D 5 m/s2 , v0 D 10 m/sec, and s0 D 4 m
3. a D 5 m/s2 , v0 D 12 m/sec, and s0 D 3 m
4. a D 8 m/s2 , v0 D 30 m/sec, and s0 D 4:1 m
5. a D 2 m/s2 , v0 D 11 m/sec, and s0 D 1:4 m
6. a D 14 m/s2 , v0 D 14:2 m/sec, and s0 D 7:33 m

Problem 1.56 Find s.t/ given the following rates of acceleration and initial velocity and posi-
tion.

1. a.t / D 3:2 C 1:1
t m/s2 , v0 D 3 m/s, s0 D 0 m

p
2. a.t / D t m/s2 , v0 D 1 m/s, s0 D 5 m
3. a.t / D t 1:5 m/s2 , v0 D 4 m/s, s0 D 2 m
4. a.t / D t 2 t m/s2 , v0 D 2 m/s, s0 D 0:4 m

5. a.t / D cos.t / m/s2 , v0 D 1 m/s, s0 D 0 m

1
6. a.t / D m/s2 , v0 D 1 m/s, s0 D 0 m
t

Problem 1.57 Solve the following initial value problems.

Z
1. f .x/ D x2 x
dx when f .0/ D 3
Z
2. g.x/ D sin.x/
dx when g.0/ D 0:7071
Z


3. h.x/ D 0:04x 3 2x C 4
dx when h.1/ D 2:4
38 1. INTEGRATION, AREA, AND INITIAL VALUE PROBLEMS
Z  
2:3
4. r.x/ D 0:08x C
dx when r.2/ D 0:4
x
Z
dx
5. s.x/ D 2
when s.1/ D 1:2
x C1
Z 2
x C1
6. q.x/ D
dx when q.1/ D 8:2
x

Problem 1.58 Suppose that

f 00 .x/ D 2x C 1
If f .1/ D 3 and f 0 .1/ D 0:5, find an expression for f .x/ with no unknown constants.

Problem 1.59 Suppose that

g 00 .x/ D 0:03x 2 4x C 1
If g.2/ D 7 and g 0 .1/ D 3, find an expression for g.x/ with no unknown constants.

Problem 1.60 Suppose that

h00 .x/ D 1:4x C 5:6
If h.1/ D 2 and h.2/ D 4, find an expression for h.x/ with no unknown constants. Notice that
you will have to integrate an unknown constant before you can solve.

Problem 1.61 Suppose that

r 00 .x/ D 2ex
If r.0/ D 2 and r 0 .0/ D 8, find an expression for r.x/ with no unknown constants.

Problem 1.62 Suppose that

s 00 .x/ D ex
If s.0/ D 1:2 and s.1/ D 5, find an expression for s.x/ with no unknown constants. Notice that
you will have to integrate an unknown constant before you can solve.

Problem 1.63 Suppose that

q 00 .x/ D sin.x/ C cos.x/
If q.0/ D 1:0 and q.=2/ D 0:5, find an expression for q.x/ with no unknown constants. Notice
that you will have to integrate an unknown constant before you can solve.
 1.4. INDUCTION AND SUMS OF RECTANGLES 39
1.4 INDUCTION AND SUMS OF RECTANGLES
In this section we will study the theory of integration that was developed before the fundamental
theorem that doesn’t need the concept of the derivative. Let’s start with an example that could
be solved with calculus but need not be.
Example 1.64 Find, without using calculus, the area under f .x/ D 2x C 1 from x D 0 to
x D 3.

Solution:

7
f .x/ D 2x C 1

-2 3
-0.5

-2
1 1
The area of the large triangle is bh D 3:5  7 D 12:25. The area of the light gray triangle is
2 2
1 1
also bh D 0:5  1 D 0:25. The area we want, the dark gray, is thus 12:25 0:25 D 12 units2 .
2 2
˙

The reason we don’t need calculus for this example is that the shape of the area is really simple.
As long as we remember that the area of a triangle is one-half base times height, we’re fine. As
soon as the integral is under a function that is curved, we will need calculus. Where does this
calculus come from? At this point we need a digression to create the mathematical machinery
that permits us to synthesize the calculus.
40 1. INTEGRATION, AREA, AND INITIAL VALUE PROBLEMS
The symbol X

means “add up the things following the symbol.” So

5
X
i
iD1

is a symbolic way of saying “add up the numbers i from i D 1 to i D 5. The name of the symbol
P
is uppercase sigma – not to be confused with  which is called lowercase sigma. Applying
we get
X5
i D 1 C 2 C 3 C 4 C 5 D 15
i D1

which isn’t too bad. The problem is when we get something like
200
X
i D 1 C 2 C


C 199 C 200 D 20; 100
i D1

The incomparable German mathematician Carl Friedrich Gauss found the following shortcut:
n
X 1
iD n.n C 1/
2
i D1

How do you prove a formula like that is correct?

The usual technique is called mathematical induction.

Knowledge Box 1.14

Mathematical induction
Suppose we wish to prove a proposition P .n/ is correct for all n  c .
Then the following steps suffice
• Verify that P .c/ is true.
• Assume that, for some n, P .n/ is true.
• Show that, if P .n/ is true, then so is P .n C 1/
 1.4. INDUCTION AND SUMS OF RECTANGLES 41
The assumption in the second step is easy, because P .n/ is true when n D c , due to our work
on the first step. The key step is the third one. Once we’ve got it, P .c/ implies P .c C 1/ is true,
which in turn implies P .c C 2/ is true, and so on, until we hit any particular n  c .
Example 1.65 Use mathematical induction to prove Gauss’ formula.

Solution:

The proposition is
n
X 1
P .n/ W iD n.n C 1/
2
i D1
Let’s start with c D 1.
1
X 1 1
P .c/ D P .1/ W i D1D c.c C 1/ D .1/.1 C 1/ D 1
2 2
i D1

Since 1=1, this is true, and we have the first step of the induction. We now assume that P .n/ is
true (for some n) and look at P .n C 1/.
nC1
X n
X
iD i C .n C 1/
iD1 i D1

1
D n.n C 1/ C .n C 1/ Apply P .n/
2

n.n C 1/ C 2.n C 1/
D
2

n2 C 3n C 2
D
2

1
D .n C 1/.n C 2/
2

1
D .n C 1/..n C 1/ C 1/ Which is P .n C 1/
2
So if P .n/ is true, we can show using algebra that P .n C 1/ is true. This tells us, by mathematical
induction, that Gauss’ formula is true for all n  1.

˙
42 1. INTEGRATION, AREA, AND INITIAL VALUE PROBLEMS
Some other examples of proof with mathematical induction appear in the homework problems.
Now we know enough to calculate an integral or a curved function without using the funda-
mental theorem of calculus. Our goal is to approximate the area under the curve with shapes
we can compute.
Z 2
2
Examine pictures of f .x/ D x in Figure 1.1. The gray area is x 2
dx ; the four rectangles
0
are a (bad) approximation of the area under the curve. How can we make the approximation
better? Use more rectangles! The more rectangles, the better the area is approximated.

7 7

-2.5 2.5 -2.5 2.5

-1 -1
7 7

-2.5 2.5 -2.5 2.5

-1 -1

Figure 1.1: Approximating an intgral with increasing number of rectangles.

 1.4. INDUCTION AND SUMS OF RECTANGLES 43
It turns out that summing and integrating share two algebraic properties.

Knowledge Box 1.15

P
Algebraic properties of
b
X b
X b
X
f .i/ C g.i/ D f .i/ C g.i/
i Da i Da i Da
b
X b
X
c
f .i / D c
f .i/
i Da i Da

for any constant c

There are an infinite number of summation formulas – but a few of them are especially useful.
All of these can be proved by induction (and you are asked to do so in the homework).

Knowledge Box 1.16

Useful sum formulas

n
X
• 1Dn
i D1
n
X n.n C 1/
• iD
i D1
2
n
X n.n C 1/.2n C 1/
• i2 D
i D1
6
n
X n2 .n C 1/2
3
• i D
i D1
4

Z 2
2
If we use n rectangles to approximate x 2
dx , then each rectangle has a width of w D
.
1 n
The right side of each rectangle (which determines its height) is xi D i
w for i D 1; 2; : : : n.
44 1. INTEGRATION, AREA, AND INITIAL VALUE PROBLEMS
 2
2i
The height of the i th rectangle is .xi /2 D .iw/2 D . Summing the areas we get that:
n

n
X n
X  
2 2i 2
AŠ W H D
n n
i D1 i D1

Let’s simplify this with the algebraic rules for sums.

n
X   n
2 2i 2 X 2 4i 2
D
n n n n2
i D1 i D1

n
X 8i 2
D
n3
i D1

n
8 X 2
D i
n3
i D1

8 n.n C 1/.2n C 1/ P
D Use i 2 formula
n3 6

16n3 C 24n2 C 8n
D
6n3

8n3 C 12n2 C 4n
D
3n3

Now we have a formula for the approximate area with n rectangles – the approximation gets
better as n grows. This means that

Z 2
8n3 C 12n2 C 4n 8
x 2
dx D lim 3
D units2
0 n!1 3n 3

which is the same result we get if we do the integral in the usual way.
This is a very cumbersome method of computing integrals – not used in practice – but it shows
that there is a theory for integrals, just as there is for derivatives. The fundamental theorem is a
godsend. Imagine if you did not know that integrals and anti-derivatives were the same thing.
Every integral would be a limit of sums of rectangles (or some other shape).
 1.4. INDUCTION AND SUMS OF RECTANGLES 45
Many integrals cannot be done symbolically – with formulas and algebra. The discipline of
numerical analysis studies how to use things like rectangle-sum approximations to get useful
values for integrals that cannot be calculated with pure calculus.

PROBLEMS

Problem 1.66 Compute the following sums. You may use the formulas you are asked to prove
in Problem 1.67.

40
X 100
X 28
X
1
1. i2 4. .i C 2/ 7. i3
2
i D1 i D1 i D14
60
X 101
X 37
X
2. i 5. i 8. .2i 1/
i D20 i D61 i D18
30
X 23
X
3. .2i C 5/ 6. 2i
i D1 i D5

Problem 1.67 Use mathematical induction to demonstrate that the following formulas are
correct.

n
X n
X
1. .2i 1/ D n2 5. 2i D 2nC1 1
i D1 i D0
n
X
2. 1Dn n
X 1 nC1

i D1
6. 3i D 3 1
n
X 2
n.n C 1/.2n C 1/ i D0
3. i2 D
6
i D1
n
X
n
X n2 .n C 1/2 x nC1 1
4. 3
i D 7. xi D
4 x 1
i D0
i D1
46 1. INTEGRATION, AREA, AND INITIAL VALUE PROBLEMS
Problem 1.68 Explain why the area under a line y D mx C b can always be found without
calculus.

Problem 1.69 A formula for approximating

Z 2
x 2
dx
0

with n rectangles was computed in this section. For n D 4; 6; 8; 12; 20; and 50 rectangles, com-
pute the error of the approximation.

Problem 1.70 Find the formula for the sum of rectangles for y D x 3 to approximate the in-
tegral of y D x 3 from x D 0 to x D c . Having found the formula, find the integral by taking a
limit.

Problem 1.71 The sum of rectangles used in this section was based on the right side of the
intervals. How would an approximation that used the left side of the interval be different? Could
it still be used with a limit, to compute integrals? Explain.

-2.5 2.5

-1

Problem 1.72 The approximation sketched above uses trapezoids instead of rectangles to ap-
proximate the area under the curve. The area of each trapezoid is the average of the height of
the points on either side of the interval. Write out the approximation formula for n trapezoids
instead of n rectangles. Make it as simple as you can.
 1.4. INDUCTION AND SUMS OF RECTANGLES 47
Problem 1.73 Is the trapezoid method more accurate than the rectangle method? Explain or
justify your answer.

Problem 1.74 Approximate

0 2
x
dx
2
using 20 intervals using rectangles and trapezoids. The area of a trapezoid with left side of height
h and right side of height k and width w is

1
AD w.h C k/
2
 49

CHAPTER 2

Parametric, Polar, and Vector

Functions
Consider the plot in Figure 2.1. By the rules we’ve come up with so far this is not even close to
being a function. It also looks like the graph of nothing we have done so far. It is a spiral. In this
chapter we will look at methods for using calculus – which wants things to act like functions
at least locally – on curves like this. The underlying idea is to treat the position of a point as
having coordinates that are individually functions of a parameter. These are called parametric
functions.

2.1 PARAMETRIC FUNCTIONS

The spiral shown in Figure 2.1 is not a function of the form y D f .x/. It violates the vertical line
test and, if we had shown the entire spiral instead of only its beginning, would have intersected
every vertical line an infinite number of times. In order to make the spiral – and many other

5.0

-5.0 5.0

-5.0

Figure 2.1: Example of a parametric function – a spiral.

50 2. PARAMETRIC, POLAR, AND VECTOR FUNCTIONS
curves – into a function, we will make the x and y coordinates of the curve functions of a
parameter t in their own right. This is how parametric curves are created.

Knowledge Box 2.1

Parametric curves
If we specify a set of points by
.x.t/; y.t//;
where x D x.t / and y D y.t/, then the resulting structure is called
a parametric curve. The variable t is called the parameter.

Example 2.1 Graph the curve .x.t/; y.t // if x.t / D 3 cos.t/ and y.t / D 2 sin.t/.

Solution:
4

-4 4

-4

In order to graph this curve, assuming we don’t know ahead of time that it is an ellipse
with major axis of length 6 and minor axis of length 4, we plot points. We know that
1  sin.x/; cos.x/  1. So, it’s not hard to figure out where the bounds are. The points on the
curve where the extreme values are have been plotted with dots.

˙
 2.1. PARAMETRIC FUNCTIONS 51
4

-4 4

-4

Figure 2.2: Orientation of ellipse from Example 2.1.

Definition 2.1 The major axis of an ellipse is its largest diameter. The minor axis of an ellipse is its
smallest diameter.

The major axis of the ellipse in Example 2.1 is 6; the minor axis is 4. In general,

.a
cos.t/; b
sin.t //

is an ellipse with major and minor axes in the direction of the coordinate axes, centered at the
origin. The major and minor axes have size 2a and 2b with the major/minor order determined
by which is larger.

One thing that is very different about a parametric curve is that the points are ordered by the
parameter. Points generated by a larger value of the parameter are thought of as coming after
points generated by a smaller value of the parameter.

Definition 2.2 The positive orientation of a parametric curve is the direction, along the curve, in
which the parameter increases.

The orientation of the ellipse in Example 2.1, shown in Figure 2.2, is counterclockwise.
52 2. PARAMETRIC, POLAR, AND VECTOR FUNCTIONS
It turns out that any function of the usual sort can be put into the form of a parametric curve
by the simple technique of starting with y D f .x/ and defining x.t/ D t and y.t/ D f .t/. This
will make the parametric function exactly trace the graph of y D f .x/.

One disadvantage of parametric functions is that they give us many, many ways to specify the
same function. The lines .t; 2t C 1/ and .3t C 1; 6t C 3/ are two different parametric forms of
the line y D 2x C 1.

If x.t / and y.t / are both linear functions, then .x.t /; y.t // is the parametric form of
some line.

Example 2.2 Put the line

.2t C 5; 4t 1/

into the standard y D mx C b form.

Solution:

x D 2t C 5

2x D 4t C 10

2x 11 D 4t 1

2x 11 D y

So we see y D 2x 11 is the standard form of this parametric line.

A big advantage of parametric curves is that they let us specify things that are hard to specify
in other ways and that they let us turn things that were not functions into functions of the
parameter. Once something is a function, most of the formalisms of calculus apply and we can
differentiate, integrate, and optimize.
 2.1. PARAMETRIC FUNCTIONS 53
Example 2.3 Consider the parametric curve .sin.t/; t cos.t //. Graph the curve for 0  t  2 .

Solution:

-4 4

-4

This one is done by plotting points.

˙

Parametric curves can intersect themselves, possibly many times, at least in the x y plane. If
we consider the parameter, the “intersecting” points are not intersections, because their values
of the parameter differ. If we graph x.t/ D 1:8 cos. 34  t /, y.t / D 1:8 sin. 20
3
 t /, we get the curve
in Figure 2.3.

We have not yet explained how to make a spiral. If f .t / is a function that is increasing on Π0; 1/,
then the parametric curve

.f .t / cos.t /; f .t / sin.t//

is a spiral with f .t / controlling how fast it spirals outward. If we take f .t / D ln.t C 1/, for
t  0, then we get the curve in Figure 2.4.
54 2. PARAMETRIC, POLAR, AND VECTOR FUNCTIONS

2.0

-2.0 2.0

-2.0

Figure 2.3: Parametric curve that intersects itself. Given by x.t/ D 1:8 cos. 43  t/, y.t / D
1:8 sin. 20
3
 t /.

4.0

-4.0 4.0

-4.0

Figure 2.4: Spiral parametric curve. Given by x.t / D t cos.t/, y.t / D t sin.t/
 2.1. PARAMETRIC FUNCTIONS 55
2.1.1 THE DERIVATIVE OF A PARAMETRIC CURVE
In our study of differentials, we learned that we could algebraically cancel differentials. The
dy
ordinary derivative of a curve is , which lets us calculate the derivative of a parameteric curve
dx
by doing this:

dy dy=dt y 0 .t/
D D 0
dx dx=dt x .t /
Let’s highlight that in a Knowledge Box.

Knowledge Box 2.2

Derivatives of parametric curves

The slope m of the tangent line to a parametric curve .x.t/; y.t// at
parameter value t is:
y 0 .t/
mD 0
x .t/

dy
Example 2.4 Find the derivative of the parametric curve with x.t/ D 5 cos.t/ and
dx
y.t / D 3 sin.t/.

Solution:

First compute x 0 .t / D 5 sin.t/ and y 0 .t/ D 3 cos.t /.

dy y 0 .t /
D 0
dx x .t/

5 sin.t /
D
3 cos.t /

5
D tan.t/
3
˙
56 2. PARAMETRIC, POLAR, AND VECTOR FUNCTIONS
Example 2.5 Find the tangent line to .t 2 ; 2t 3/ at (4,1).

Solution:

Our first job is to check if the point is on the curve. If 2t 3 D 1, then t D 2, and the
point for t D 2 is (4,1). So, no problem there.

dy y 0 .t / 2 1
D 0 D D
dx x .t/ 2t t

1 1
so we have m D D . Computing the line using the point-slope formula, we get:
t 2

1
y 1D .x 4/
2

1
yD x 2C1
2

1
yD x 1
2

Let’s do that again for the practice.

p
Example 2.6 Find the tangent line to .4 cos.t /; 2 sin.t// at .2 3; 1/.

Solution:

p is on the curve. If 2 sin.t/ D 1, then sin.t/ D 1=2,

Our first job is to check if the point
 
 3 p
and so t D . Then 4 cos D4 D 2 3. So the point is on the curve, no problem.
6 6 2
Compute the slope of the tangent line:

dy y 0 .t/ 2 cos.t/ 1
D 0 D D cot.t /
dx x .t/ 4 sin.t / 2
 2.1. PARAMETRIC FUNCTIONS 57
p
 1p 3
At t D this is m D 3D . Apply the point-slope formula and simplify:
6 2 2
p
3 p
y 1D .x 2 3/
2
p
3
yD xC3C1
2
p
3
yD xC4
2
and we have the tangent line.
˙

Example 2.7 Find the parameter values for which the tangent line to .2 sin.t/; cos.t// has
slope m D 1=2.

Solution:

dy y 0 .t/ sin.t / 1
D 0 D D tan.t/
dx x .t/ 2 cos.t/ 2
Solve:
1 1
tan.t/ D
2 2

tan.t/ D 1


tD C n n D 0; ˙1; ˙2; : : :
4
Remember that the angles with tangent equal to 1 are the ones with equal sine and cosine.

This section has been a small sampler platter of the many sorts of parametric curves that are
possible. We’ve also restricted ourselves to only two coordinates. In Section 2.3 we will revisit
this with the formalism of vectors rather than parameters.
58 2. PARAMETRIC, POLAR, AND VECTOR FUNCTIONS
PROBLEMS

Problem 2.8 Graph each of the following parametric curves.

1. .cos.t /; sin.t/ C 2/, 0  t  2 4. .t cos.t/; t sin.t//, 0  t  4

2. .1 t; 2t 4/, 5  t  5 5. .2 cos.3t/; 3 sin.2t //, 0  t  2
3. .sin.t/; t cos.t//, 2  y  2 6. .t 2 ; t/, 3  t  3

Problem 2.9 Find a parametric equation for an ellipse with major axis of length 12 in the
direction of the y -axis and minor axis of length 2 in the direction of the x -axis.

Problem 2.10 Find a parametric equation for an ellipse with major axis of length 5 in the
direction of the x -axis and minor axis of length 3 in the direction of the y -axis.

Problem 2.11 Find a parametric equation for an ellipse with major axis of length 14 in the
direction of the y -axis and minor axis of length 7 in the direction of the x -axis.

Problem 2.12 Find the standard y D mx C b form for each of the following parametric lines.

1. .t; 2t C 1/ 4. .t C 7; 7 t/

2. .2t; 1 t/ 5. .3t 1; 3 2t/

3. .3t C 1; 5t 1/ 6. .2 3t; 3 2t /

Problem 2.13 Graph and carefully describe the parametric curve

.2 sin.t/; 3 sin.t / C 1/
 2.1. PARAMETRIC FUNCTIONS 59
0
Problem 2.14 Find y for each of the following parametric curves.

1. .sin.2t /; cos.3t// 3. .t
sin.t/; cos.t// 5. .cos.5t/; 2 sin.3t//

2. .1 t 2 ; 3t C 1/ 4. .t sin.t/; t cos.t // 6. .t 3 ; t 2 /

Problem 2.15 Find the tangent line to .t 3 ; t 2 / at . 8; 4/.

Problem 2.16 Find the tangent line to .sin.t/; t / at .1; =2/.

Problem 2.17 Find the tangent line to

 
1 1
2
;
t C1 t
at the point . 12 ; 1/.

Problem 2.18 For which values of t does

.cos.t /; sin.2t //
have a tangent line with a slope of 2?

et
Problem 2.19 Let f .t/ D . Describe carefully the parametric curve
et C 1
.f .t / cos.t /; f .t / sin.t //
for 1 < t < 1.

Problem 2.20 The parametric curve .af .t /; bf .t // on 1 < t < 1 is a line or a line segment.
Give explicit directions on how to figure out which line segment it is.

Problem 2.21 Plot carefully the parametric curve:

.sin.t / C cos.2t/=2; cos.t/ C sin.2t/=2/

Problem 2.22 A function cannot assign multiple values to the same point. In light of this, why
are points where a parametric curve intersects itself not a problem?
60 2. PARAMETRIC, POLAR, AND VECTOR FUNCTIONS

p
. 3; 1/

r D2

 D =6

Figure 2.5: Point in both polar and standard coordinates.

2.2 POLAR COORDINATES

Polar coordinates are an alternate coordinate system for doing business, based on direction
p and
distance instead of x -coordinate and y -coordinate. Examine Figure 2.5. The point . 3; 1/ can
also be specified by going a distance (radius) of two from the origin in the direction of the angle

D .
6

The circle of radius 2 is shown to supply clarity – polar graphs use circles and radial lines the
way that a standard graph uses grids. Figure 2.6 shows an example of polar graph paper.
 2.2. POLAR COORDINATES 61


2 r D5
2 
3 3
3 r D4 
4 4
r D3
5 
6 6
r D2

r D1

 2

7 11
6 6

5 7
4 4
4 5
3 3
3
2

Figure 2.6: Polar graph paper.

62 2. PARAMETRIC, POLAR, AND VECTOR FUNCTIONS
Knowledge Box 2.3

Polar to rectangular conversion formulas

In order to transform between .x; y/ and .r; / coordinates we use the
following formulas:
• x D r
cos./
• y D r
sin./
p
• r D x2 C y2
•  D tan 1 .y=x/
Where the formulas don’t work, when x is zero and so  is an odd mul-
tiple of =2, use common sense – the directions are vertical.

For historical reasons there are two names for the standard coordinate system when we are com-
paring it to the polar coordinate system – rectangular coordinates and Cartesian coordinates.
We will use the terms interchangeably.

Example 2.23 Find the polar version of the point .2; 1/.

Solution:

p p
rD 22 C 12 D 5; and

 
1 1
 D tan Š 0:4636 rad
2

p
So the polar point is .r;  / D . 5; 0:4636/.

˙
 2.2. POLAR COORDINATES 63
 
3
Example 2.24 Find the rectangular coordinates for the polar point 4; .
4
Solution:
3
In this case r D 4 and  D so:
4
  p
3 2 p
x D r
cos./ D 4
cos D4
D 2 2
4 2
  p
3 2 p
y D r
sin. / D 4
sin D4
D2 2
4 2
p p
and so the corresponding point in Cartesian coordinates is .x; y/ D . 2 2; 2 2/.

One of the major uses for converting between the two coordinate systems is to permit us to plot
polar points on normal graph paper when we are graphing a polar function. If you have a good
eye, or a protractor, it is possible to plot polar points directly, but typically a person just learning
polar coordiantes have far more practice plotting .x; y/-points.
One of the nice things about polar coordinates is that they let us deal very easily with circles
centered at the origin. Circles centered at the origin are constant functions in polar coordinates.
The next example demonstrates this.
Example 2.25 Graph the polar function r D 2.

Solution:
(0,3)

r D2

(-3,0) (3,0)

(0,-3)
˙
64 2. PARAMETRIC, POLAR, AND VECTOR FUNCTIONS
We usually write polar functions in the form
r D f . /

making the angle the independent variable and the r the dependent variable. This makes it very
easy to give polar functions as parametric functions.
Knowledge Box 2.4

Parametric form of polar curves

If r D f ./ on 1    2 is a polar curve, then a parametric form
for the same curve is:
.f .t/
cos.t/; f .t/
sin.t //

for t 2 Œ1 ; 2 .

So far in this section we have established the connections between polar coordinates and the
rest of the systems developed in this text. It is time to display polar curves that have unique
characteristics that are most easily seen in the polar system.
Example 2.26 Graph the polar function r D cos.3/ on Π0; /.

Solution: (0,1)

(-1,0) (1,0)

(0,-1)

The arrows show the drawing direction.

˙
 2.2. POLAR COORDINATES 65
Definition 2.3 Petal curves are curves with equations of the form:

r D cos.n/ or r D sin.n /;

where n is an integer.
If no restriction is placed on n, then the curve is traced out an infinite number of times. For odd
n, a domain of  2 Π0; / traces the curve once; when n is even,  2 Π0; 2/ is needed to trace
the entire curve once.

Example 2.27 Compare the curves r D sin.5 / and r D cos.5 / on the range  2 Π0; /.

(0,1)

(-1,0) (1,0)

(0,-1)
r D cos.5 /
(0,1)

(-1,0) (1,0)

(0,-1)
r D sin.5 /
˙
66 2. PARAMETRIC, POLAR, AND VECTOR FUNCTIONS
The odd fact, that the minimal domain (to hit all the points) is twice as large when n is even,
is to some degree explained by the fact that, while petal curves with odd parameter n yield n
petals, when n is even we get 2n petals.

Example 2.28 Plot the polar function r D cos.4 /.

Solution:
(0,1)

(-1,0) (1,0)

(0,-1)

See? We have n D 4 but 8 petals.

˙

It is possible to use values of the petal-determining parameter for polar curves that are not
integers, but then figuring out the minimal domain to plot the curve becomes problematic.

Example 2.29 Plot the polar function r D cos.1:5  / for  2 Π0; 4/.

Solution:
(0,1)

(-1,0) (1,0)

(0,-1)
˙
 2.2. POLAR COORDINATES 67
A fractional parameter yields more and fatter petals on the curve. If you have polar curve
plotting software, experiment with these parameters and see what you can find.

2.2.1 POLAR CALCULUS

There are many interesting shapes that can be made with polar coordinates, but at this point we
dy
are going to look into the calculus of polar curves. First, computing at a point .r;  /.
dx
We already have a parametric form for polar curves:

x./ D f . /
cos. /; y. / D f . /
sin./

This is the parametric curve:

.x D f ./ cos./; y D f . / sin.//

We can then apply the technique for finding the derivative of a parametric curve and get:

dy y 0 ./ f 0 ./
sin. / C f ./
cos. /

D 0 D 0
dx x ./ f ./
cos./ f . /
sin./

That’s not a compact or elegant formula, but it is a formula.

Let’s put this formula in a Knowledge Box.

Knowledge Box 2.5

dy
Computing for a polar curve
dx
dy f 0 ./
sin./ C f ./
cos./
D 0
dx f ./
cos./ f ./
sin./
68 2. PARAMETRIC, POLAR, AND VECTOR FUNCTIONS

Example 2.30 Find the tangent line to r D cos.3 / at  D .
12
Solution:
To use the formula in Knowledge Box 2.5 we need f . / D r D cos.3 / and f 0 . / D 3 sin.3 /.
Computing the derivative we get:
dy 3 sin .3/ sin ./ C cos .3 / cos ./
D
dx 3 sin .3/ cos . / cos .3 / sin ./










3 sin 3 12 sin 12
C cos 3 12 cos 12
D 







Š 0:06
3 sin 3 12 cos 12
cos 3 12 sin 12
p !
      2 
When  D we get the point in polar coordinates cos ; D ; . Now we
12 4 12 2 12
need
p the Cartesian
p version of the point. Applying the polar-to-rectangular formulas we obtain:
2  2  
. cos ; sin / Š .0:683; 0:183/
2 12 2 12
Find the line:

y 0:183 D 0:06.x 0:683/

y D 0:06x C 0:224

Let’s actually take a look at the graph with the curve and the tangent line. The tangent line hits
the curve at three points, but the point of tangency is also plotted and is clearly a tangent line;
the other two intersections cut through the curve at sharp angles.

(0,1)

y D 0:06x C 0:224
(-1,0) (1,0)

(0,-1)

The curve r D cos.3 / and the tangent line at  D .
12
˙
 2.2. POLAR COORDINATES 69
2

1

Figure 2.7: Area “under the curve” in polar coordinates.

The formula for the Cartesian derivative of a curve in polar coordinates is pretty cumbersome,
but the Cartesian form of r D cos.3/ is really just not manageable at all. So, if we need tangent
lines to these curves, this formula is our only hope.
What about integrals? If we have a sector of a circle subtending an angle of  , then the area is
just the fraction of the full circle that  covers times the area formula of  r 2 so we get

 
Area D
 r 2 D r 2:
2 2

This means that the farther from the origin an area is, the larger it gets. This leads to a really
different integral formula for area. First of all, it is area enclosed by the polar curve. Second, we
need to be very careful about r < 0 because it’s so much less obvious than y < 0.
The shaded area between the polar curve shown in Figure 2.7 and the origin, in the angular
range 1    2 is
Z
1 2
Area D r. /2
d
2 1
70 2. PARAMETRIC, POLAR, AND VECTOR FUNCTIONS
When we did integrals in Cartesian space, we approximated areas with rectangles. In polar space,
we use pie-shaped slices, and so the area depends on the square of the functional value instead of
just its value. The constant of one-half carries over from the area formulas for sectors of a circle.

Knowledge Box 2.6

Finding the area between a polar curve and the origin

Z
1 2
Area D r./2
d
2 1

In Chapter 1, we had to be careful about area above and below the x -axis. When working
with polar curves, the analogous problems are areas with r  0 and r  0. If we know r stays
positive, then we still need to know the minimal angular domain that sweeps out the entire
shape once. Let’s do some examples.

4
Example 2.31 Find the area enclosed by r D  from  D 0 to  D .
3
Solution: r is positive on the domain of integration so we can just integrate.
Z 4=3
1
AD r 2
d
2 0

Z 4=3
1
D  2
d
2 0

ˇ
1 3 ˇˇ4=3
D  ˇ
6 0

 3 !
1 4
D 0
6 3

32 3
D units2
81

Š 12:25 units2
 2.2. POLAR COORDINATES 71
Let’s look at a picture of the area:

(0,3)

(-4,0) (2,0)

(0,-4)
4=3

Now for a more challenging example. The petal curve function from Example 2.26 was
r D cos.3/ on Œ0; /. It has both positive and negative r in the course of the graph.
72 2. PARAMETRIC, POLAR, AND VECTOR FUNCTIONS
Example 2.32 Find the area enclosed by r D cos.3/.

Solution:

Let’s graph this again, using different shades for the positive and negative r .

(0,1)

(-1,0) (1,0)

(0,-1)

Notice the first dark gray arc of the graph is from  D 0 to  D =6. This arc encompasses half
1
of one of the three petals and so is of the area. Since on this angular interval r is positive,
6
computing six times the integral of r D cos.3/ on Œ 0; =6  will give us the full area of the
curve.
 2.2. POLAR COORDINATES 73

Z =6
1
Area D 6
cos2 .3/
d
2 0

Z =6
1 C cos.6/
D3
d
0 2
Z =6 Z =6
3 3
D d C cos.6 /
d
2 0 2 0

Z =6 Z 
3 3
D d C cos. /
d Cosine was rescaled
2 0 2 0

ˇ ˇ
3 ˇˇ=6 3 ˇ
D ˇ C sin./ˇˇ
2 0 2 0

3   3
D 0 C .0 0/
2 6 2

D units2
4

PROBLEMS

Problem 2.33 For each of the following points in Cartesian coordinates, give the correspond-
ing points in polar coordinates.

p
1. (1,1) 3. .4; 4 3/ 5. (-5,-5)
2. (3,4) 4. (-3,7) 6. (0,8)
74 2. PARAMETRIC, POLAR, AND VECTOR FUNCTIONS
Problem 2.34 For each of the following points in polar coordinates, give the corresponding
points in rectangular coordinates. The points are in the form .r;  / with the angular coordinate
in radians.

1. .4; =3/ 3. . 5; =4/ 5. .0:2; 3=4/

2. .2; =5/ 4. .7; 4=3/ 6. .2:6; 1:14/

Problem 2.35 Plot the following polar curves on the given domain.

1. r D cos.2 / on Π0; 2/ 4. r D 4.cos./ C 1/ on Π0; 2/

2. r D sin.7/ on Π0; / 5. r D =4 on Π0; 6/
3. r D sin.3/ C 2 on Π0; 2/ 6. r D 5 cos. / on Π0; /

Problem 2.36 Find a parametric form for 3 D 5 cos./ on Π0; /.

3
Problem 2.37 Find the tangent line to r D 6 at  D .
4


Problem 2.38 Find the tangent line to r D  at  D .
4


Problem 2.39 Find the tangent line to r D cos.4 / at  D .
16

Problem 2.40 Using the polar integral, verify that a circle of radius R has area equal to R2 .

Problem 2.41 Graph r D sin.5/ on Œ0; / using two colors or line styles to show where r is
positive or negative.

Problem 2.42 Find the area enclosed by r D sin.5 /. Worry about the issue of negative radii.
 2.2. POLAR COORDINATES 75
Problem 2.43 Graph r D cos.2/ on Œ0; 2/ using two colors or line styles to show where r is
positive or negative.

Problem 2.44 Find the area enclosed by r D cos.2 /. Worry about the issue of negative radii.

Problem 2.45 Using software of some sort, graph r D cos.1:1
/ on Œ0; 20/.

Problem 2.46 Using software of some sort, graph r D sin.5:1
/ on Œ0; 20/.

Problem 2.47 Describe completely all polar curves of the form r D C
sin. / in simple En-
glish.

Problem 2.48 For the polar curve r D  find the angle ˛ so that the area between the curve
and the origin on the interval Œ0; ˛ is 2.

Problem 2.49 Carefully graph the polar curve r D cos2 . /.

Problem 2.50 Carefully graph the polar curve r D sin2 . /.

Problem 2.51 Why is the curve

r D3 3 sin. /

called a cartiod?

Problem 2.52 Write a paragraph accurately explaining where the term “Cartesian,” as used in
describing a coordinate system, comes from.

Problem 2.53 For the polar curve r D cos A where A is a decimal number with a positive,
finite number of non-zero decimals, find the rule for the domain for  that permits the curve to
be graphed once.
76 2. PARAMETRIC, POLAR, AND VECTOR FUNCTIONS
2.3 VECTOR FUNCTIONS
One of the most unsatisfying things about the treatment of position, velocity, and acceleration
in single-variable calculus was that these things happened in one dimension. This situation can
be amended by using vectors. A vector is a lot like a point – it has coordinates and a dimension
– but a vector is thought of as specifying a direction and a magnitude rather than just a point in
space.
Example 2.54 A vector:
(0,4)
(2,3)

(-4,0) (4,0)

(0,-4)
The vector vE D .2; 3/

We use a small arrow over a variable name to denote that an object is a vector. So, the vector
shown above is vE D .2; 3/. In a way, vectors are like multi-dimensional numbers that let us
work in as many dimensions as we need to. Vectors have their own arithmetic. It’s called vector
arithmetic.
Knowledge Box 2.7
Vector Arithmetic
If vE D .v1 ; v2 ; : : : ; vn / and wE D .w1 ; w2 ; : : : ; wn / are vectors in n-dimensions
and c is a constant, then we define the following.
• c
vE D .cv1 ; cv2 ; : : : ; cvn / (scalar multiplication)
• vE C wE D .v1 C w1 ; v2 C w2 ; : : : ; vn C wn / (vector addition)
• vE w
E D .v1 w1 ; v2 w2 ; : : : ; vn wn / (vector subtraction)
• vE
wE D v1 w1 C v2 w2 C : : : C vn wn (dot product)
 2.3. VECTOR FUNCTIONS 77
Example 2.55 If vE D .1; 2; 1/ and wE D .0; 2; 4/, compute 5
wE vE.

Solution:

5
.0; 2; 4/ .1; 2; 1/ D .0; 10; 20/ .1; 2; 1/ D .0 1; 10 2; 20 . 1// D . 1; 8; 21/

Example 2.56 If vE D . 2; 1; 3/ and wE D .1; 2; 1/, compute vE
wE .

Solution:
vE
w
ED 2
1C1
2C3
1D 2C2C3D3
˙

There are a number of useful algebraic rules for vector arithmetic.

Knowledge Box 2.8

Vector Algebra
• c
.E
v C w/
E D c
vE C c
w
E
• c
.d
vE/ D .cd /
vE
• vE C wE D wE C vE
• uE
.E
v C w/
E D uE
vE C uE
w
E

Definition 2.4 The length or magnitude of a vector vE D .v1 ; v2 ; : : : ; vn /, denoted jE

v j, is given by
q
v j D v12 C v22 C


C vn2
jE

Notice that this is similar to the formula for calculating distance. The length of a vector is the
distance it spans, so this is a natural definition.
It turns out that there is a really useful property of one way we multiply two vectors, the dot
product in Knowledge Box 2.7, that also uses the length of vectors. Suppose we have two vectors
vE and w
E as shown in Figure 2.8. Then, we can use the dot product together with the lengths of
the vectors to calculate the angle between them.
78 2. PARAMETRIC, POLAR, AND VECTOR FUNCTIONS

w
E

vE

Figure 2.8: Two vectors.

Knowledge Box 2.9

Formula for the angle between vectors

vE
w
E
cos./ D
jvjjwj

What makes this property so useful? If two vectors are at right angles, then the cosine of the
angle between them is zero. This means we can detect right angles in a new and easy way and in
any number of dimensions. In particular, two vectors are at right angles to one another if and
only if their dot product is zero.
Example 2.57 Which of the following pairs of vectors are at right angles to one another?
1. vE D .1; 1/ and wE D .1; 1/
2. sE D .1; 2; 1/ and tE D . 1; 2; 3/
3. aE D .3; 1; 2/ and bE D .2; 2; 3/
 2.3. VECTOR FUNCTIONS 79
Solution:

Compute dot products:

vE
w
E D .1; 1/
.1; 1/ D 1 1D0

So vE and wE are at right angles to one another.

sE
tE D .1; 2; 1/
. 1; 2; 3/ D 1C4 3D0

So sE and rE are at right angles to one another.

aE
bE D .3; 1; 2/
.2; 2; 3/ D 6 2 C 6 D 10

So aE and bE are not at right angles to one another.

Definition 2.5 Objects at right angles to one another are also said to be orthogonal to one another.

So with our new word we can re-state this property of the dot product as follows. Two vectors
are orthogonal if and only if they have a dot product of zero.

Definition 2.6 The cross product of vE D .v1 ; v2 ; v3 / and wE D .w1 ; w2 ; w3 / is defined to be:

vE  w
E D . v 2 w3 v3 v2 ; v3 w1 v1 w3 ; v 1 w2 v2 w1 /

The cross product of two vectors is only defined in three dimensions. It is a very special purpose
operator, but it is quite useful in physics. The following property is used to construct systems of
directions.

Knowledge Box 2.10

Orthogonality of the cross product

If vE and wE are non-zero vectors in three dimensions that are not scalar
multiples of one another, then vE  wE is at right angles to (orthogonal to)
both vE and wE .
80 2. PARAMETRIC, POLAR, AND VECTOR FUNCTIONS
Example 2.58 Find a vector at right angles to both (1, 1, 2) and (3, 0, 4). Check your answer
by taking the relevant dot products.

Solution:

Use the cross product:

.1; 1; 2/  .3; 0; 4/ D .1
4 2
0; 2
3 1
4; 1
0 1
3/ D .4; 2; 3/

Check:
p
.1; 1; 2/
.4; 2; 3/ D 4 C 2 6D0
p
.3; 0; 4/
.4; 2; 3/ D 12 C 0 12 D 0

So the vector .4; 2; 3/ is orthogonal to .1; 1; 2/ and .3; 0; 4/.

Definition 2.7 A unit vector is a vector vE with jE

v j D 1.

This is an almost trivial definition, but, coupled with the next Knowledge Box rule, it captures
an important feature of vectors.

Knowledge Box 2.11

The unit vector in the direction of a given vector

If vE is not zero, then
1

vE
jE
vj
is the unit vector in the direction of vE.

Unit vectors have a number of applications, but the one we will use the most often is that they
capture a notion of direction. There are an infinite number of vectors in any given direction, but
only two unit vectors, and they point in opposite directions from one another.
 2.3. VECTOR FUNCTIONS 81
Example 2.59 Find the unit vector in the direction of vE D .1; 2/.

Solution:
p p
Start with jE
vj D 12 C 22 D 5. Then the desired unit vector is

 
1 1 2
p
.1; 2/ D p ;p
5 5 5

We’ve now reviewed enough vector algebra to start working with vector functions.

Definition 2.8 A vector function is a vector vE.t / whose coordinates are functions of a parameter t .

This is very close to the notion of a parametric curve. In fact, for some applications the two
notions are interchangeable.

Example 2.60 Identify the points that appear on the graph of the vector function

vE.t/ D .cos.t/; sin.t//

Solution:

The unit circle centered at the origin.

˙

Vector functions permit us to specify a large number of different types of curves through space.
They share with parametric curves the property that they can encode systems that are not
Cartesian functions at all.
82 2. PARAMETRIC, POLAR, AND VECTOR FUNCTIONS
Example 2.61 Plot the vector curve w.t/
E D .sin.t / C t=5; cos.t / C t=8/.

Solution:

3.0

-3.0 3.0

-3.0

It is not too hard to understand this curve. The sine and cosine terms make a simple circle while
the t =5 and t=8 terms create a line that moves the center of the circle. We can play a similar
trick by combining two circles. We will make the circles both have distinct radii and have the
particle travel around them at different speeds.
 2.3. VECTOR FUNCTIONS 83
Example 2.62 Plot the vector curve w.t/
E D .2 sin.t=4/ C cos.t /; 2 cos.t=4/ C sin.t//.

Solution:

3.0

-3.0 3.0

-3.0

This curve is tracing a circle of radius 2 that moves at one fourth the rate of a circle of radius one
that is circling the current point of the circle of radius 2. It is potentially instructive to look at
the previous example for several different speeds of motion of the particle on the smaller circle.
84 2. PARAMETRIC, POLAR, AND VECTOR FUNCTIONS
Example 2.63 Plot the vector curves:

E / D .2 sin.t =4/ C cos.t/; 2 cos.t=4/ C sin.t//

• w.t

• vE.t / D .2 sin.t =4/ C cos.2t /; 2 cos.t =4/ C sin.2t //

• sE.t / D .2 sin.t=4/ C cos.3t/; 2 cos.t=4/ C sin.3t //

• tE.t / D .2 sin.t=4/ C cos.4t/; 2 cos.t=4/ C sin.4t //

Solution:

3.0 3.0

-3.0 3.0 -3.0 3.0

-3.0 -3.0
w
E vE

3.0 3.0

-3.0 3.0 -3.0 3.0

-3.0 -3.0
sE tE

˙
 2.3. VECTOR FUNCTIONS 85
2.3.1 CALCULUS WITH VECTOR CURVES

A big advantage of vector functions is that they free us from the tyranny of abstract one-
dimensional position, velocity, and acceleration.

Definition 2.9 If vE.t/ D .f1 .t/; f2 .t /; : : : ; fn .t // then the derivative of vE.t/ is

vE0 .t / D .f10 .t/; f20 .t/; : : : ; fn0 .t//:

Here is the payoff. If sE.t / gives the position of a particle, then its velocity vector is vE.t/ D sE0 .t/,
and its acceleration vector is aE .t / D vE0 .t/. The derivative-based relationships between position,
velocity, and acceleration hold for vector functions just as they do for ordinary functions. The
difference is that we may now describe motion in complex two- and three-dimensional paths.

Example 2.64 Suppose the path of a particle is given by the vector function:

sE.t / D .sin.t/; cos.2t //

Plot the particle’s path and find its velocity and acceleration vectors.

Solution:

The plot looks like this:

1.0

-1.0 1.0

-1.0

Take derivatives to find

vE.t / D .cos.t/; 2 sin.2t // and aE .t / D . sin.t/; 4 cos.2t//

˙
86 2. PARAMETRIC, POLAR, AND VECTOR FUNCTIONS
What do the vector velocity and acceleration mean? They give (as their magnitude) the amount of
velocity or acceleration, but they also tell us which direction in space the velocity or acceleration
is going.

Example 2.65 If a particle has position sE.t / D .2 cos.t /; sin.t//, plot the position curve and
show the velocity vectors – starting at the curve – at times t D =6; 2=3; 7=6, and 5=3.

Solution:

Taking a derivative we get that the velocity vector is vE D . 2 sin.t/; cos.t//. So, let’s plot
the curve and the vectors at the specified times.

2.0

t D 2=3

t D =6

-3.0 3.0

t D 7=6

t D 5=3

-2.0

Notice that these vectors are tangent to the curve and also show which direction the curve is
oriented. They show the instantaneous velocity of the particle following the curve.

˙
 2.3. VECTOR FUNCTIONS 87
Example 2.66 Find the times when the position curve sE.t/ D .2 sin.t/; cos.t / C 1/ has
velocity of the greatest magnitude.

Solution:

First, take a derivative to obtain the velocity function:

vE.t/ D .2 cos.t/; sin.t //

The magnitude of the velocity is then

q p
j.2 cos.t/; sin.t//j D 4 cos2 .t/ C sin2 .t / D 3 cos2 .t/ C 1

Note that the second step above used the Pythagorean identity, sin2 .t / C cos2 .t/ D 1. Now
we need to know when this is largest: : : Since 0  cos2 .t /  1 it will simply happen when
cos2 .t/ D 1. This means that cos.t/ D ˙1. So the times are t D k for any whole number k .

Note that the optimization of the function

p
veloci ty D 3 cos2 .t / C 1

was not performed by solving a derivative equal to zero. Rather, reasoning based on well know
propeties of the function cosine were used to extract the maximum. Again: mathematics is the
art of avoiding calculation.

PROBLEMS

Problem 2.67 For each possible pair of the following four vectors, say which are orthogonal.
• vE D .1; 1/ • rE D .1; 2/ • sE D .2; 1/ • wE D .5; 5/

Problem 2.68 For each possible pair of the following four vectors, say which are orthogonal.
• vE D .1; 1; 1/ • rE D . 2; 2; 0/ • sE D .3; 1; 4/ • wE D .1; 2; 1/
88 2. PARAMETRIC, POLAR, AND VECTOR FUNCTIONS
Problem 2.69 Plot the following vector curves. You are free to use software.

1. vE.t / D .3 sin.t =3/ C cos.t/; 3 cos.t=2/ C sin.t //

E / D .3 sin.t =3/ C cos.2t /; 3 cos.t=2/ C sin.2t //
2. w.t
3. sE.t / D .2 sin.t=3/ C 2 cos.2t/; 2 cos.t =2/ C 2 sin.2t //
4. qE.t / D .sin.t=2/ C 3 cos.3t/; cos.t=2/ C 3 sin.3t//
5. aE .t / D .3 cos.2t/; 3 sin.t //
E / D .t C sin.t/; cos.t//
6. b.t

Problem 2.70 Example 2.64 shows the graph of the position function

sE.t / D .sin.t /; cos.2t//

and it looks like a part of a parabola. Demonstrate that it is a parabola by finding the Cartesian
form, including the domain on which the curve is defined.

Problem 2.71 If sE.t/ D .2t; cos.t// is the position of a point, then plot the curve traced by the
point on Œ0; 2 and show the velocity vectors at each multiple of =6 in the interval.

Problem 2.72 For each of the following position vectors, find the velocity and acceleration
vectors.

1. sE.t / D .3t C 5; 2t C 6/ 4. sE.t/ D .tan 1 .t/; t 2 /

2. sE.t / D .t 2 C t C 1; 5 t/ 5. sE.t/ D .1 C t C t 2 ; 1 t C t 2/

3. sE.t / D .sin.t/; cos.2t // 6. sE.t/ D .t cos.t/; t sin.t//

Problem 2.73 Find when a particle whose position is given by

sE.t / D .sin.t C =3/; cos.t//

is traveling parallel to the y -axis.

Problem 2.74 Find a Cartesian function with the same graph as:

w.t/
E D .3t C 1; 5t C 2/
 2.3. VECTOR FUNCTIONS 89

Problem 2.75 Find a Cartesian function with the same graph as

uE .t/ D .t 2 ; 1 3t/

or give a reason it is impossible.

Problem 2.76 Show, by argumentation, that a vector function, all of whose components are
linear functions of t , has the same graph as a Cartesian line.

Problem 2.77 The vector function .cos.t/; sin.t /; t / is a twisting path. First explain what curve
this vector function traces out if we ignore the third coordinate and then do your best to explain
what the shape of the curve is.

Problem 2.78 If f .t/ is a function with domain 1 < t < 1, then describe the vector func-
tion .f .t /; f .t /; f .t // as best you can.
 91

CHAPTER 3

The Arithmetic, Geometry, and

Calculus of Polynomials
Polynomials are a rich family of functions. They include lines and quadratic equations –
functions that we’ve studied in a good deal of detail because they are useful for so many things.
Polynomials are defined everywhere on the real line; they are continuous, differentiable, and
just generally nice to work with. As we will see in our later study of sequences and series,
polynomials can be used to approximate any continuous function. In this chapter we will learn
to work with polynomials more closely as well as introducing a couple of techniques that are
generally useful in the context of polynomials.

In Section 3.1 we present Newton’s method which is a general technique for finding the roots of
an equation. In Section 3.3 we use polynomials to demonstrate a rule, La’Hospital’s rule, that is
useful for evaluating limits in general. This section should be covered even if the other material
on polynomials is not of interest.

3.1 POLYNOMIAL ARITHMETIC

Consider the problem of computing .x 2 C x C 2/2 . We might do this in the following fashion:
.x 2 C x C 2/.x 2 C x C 2/ D x 2 .x 2 C x C 2/ C x.x 2 C x C 2/ C 2.x 2 C x C 2/

D x 4 C x 3 C 2x 2 C x 3 C x 2 C 2x C 2x 2 C 2x C 4

D x 4 C 2x 3 C 5x 2 C 4x C 4
Which is a lot of algebra! If we made the polynomial higher degree it would get worse. Let’s
try this instead.

1 1 2 (1x 2 C 1x C 2)
1 1 2 (1x 2 C 1x C 2)
2 2 4
1 1 2 .
1 1 2 . .
1 2 5 4 4 (1x 4 C 2x 3 C 5x 2 C 4x4)
92 3. THE ARITHMETIC, GEOMETRY, AND CALCULUS OF POLYNOMIALS
Which also shows .x 2 C x C 2/2 D x 4 C 2x 3 C 5x 2 C 4x C 4. But how? In the “long multi-
plication” table the positions correspond to powers of x . The rightmost place is the 1’s place,
the next is the x ’s place, then the x 2 ’s place and so on.

It is very similar to the algorithm for multiplying numbers with pencil-and-paper, but there is
no carry. Why no carry? Because when you add up a whole lot of terms of the form ax m you
just get many x m ’s; it never piles up to the point where you get some x mC1 ’s. In other words,
this is actually easier than the multiplication algorithm for numbers.

Knowledge Box 3.1

Fast Polynomial Multiplication

• Place the coefficients of the polynomials in the first two rows.
• In the following rows, scale the coefficients of the first polynomial by each of
the coefficients of the second, one per row.
• The first new row lines up with the ones above it; after that shift left one
place for each row.
• Add up the columns for the rows you just generated.
• This creates a last row that is the coefficients of the product.

Example 3.1 Find

.x 3 C 4x 2 C 2x 3/  .x 2 C 3x C 2/

Solution:
1 4 2 -3
1 3 2
2 8 4 -6
3 12 6 -9 .
1 4 2 -3 . .
1 7 16 11 -5 -6

So:
.x 3 C 4x 2 C 2x 3/  .x 2 C 3x C 2/ D x 5 C 7x 4 C 16x 3 C 11x 2 5x 6

˙
 3.1. POLYNOMIAL ARITHMETIC 93
This technique is enormously faster than imposing the distributive law and collecting terms, and
because it’s structured, it is easier to avoid errors. There is one potential sand trap – when a term
is missing, you must fill in a zero for it. Let’s do another example that demonstrates this.

Example 3.2 Find .x 2 C x C 1/  .x 2 C 4/.

Solution:
1 1 1
1 0 4
4 4 4
0 0 0 .
1 1 1 . .
1 1 5 4 4
So:
.x 2 C x C 1/  .x 2 C 4/ D x 4 C x 3 C 5x 2 C 4x C 4

A similar technique – that is more likely to be familiar to you – can also be used to divide
polynomials. It is sometimes called synthetic division.

Example 3.3 Compute .x 3 C 6x 2 C 11x C 6/  .x C 2/.

Solution:
1 4 3
1 2 1 6 11 6
1 2
4 11 6
4 8
3 6
3 6
0
So:
.x 3 C 6x 2 C 11x C 6/  .x C 2/ D x 2 C 4x C 3
˙
94 3. THE ARITHMETIC, GEOMETRY, AND CALCULUS OF POLYNOMIALS
These techniques for multiplication and division of polynomials are essentially book-keeping
devices. They don’t give you new capabilities, but they make existing capabilities more reliable.

Remember that we often want to find the roots of a function. For instance, when optimizing or
sketching curves, we want to to be able to solve f .x/ D 0, f 0 .x/ D 0, and f 00 .x/ D 0. Being able
to rapidly multiply and divide polynomials makes these tasks easier. The root-factor theorem
tells us that “If f .x/ is a polynomial and f .c/ D 0 for some number c , then .x c/ is a factor
of f .x/.” One thing these new techniques do not do is tell us what value of c to try. Plugging in
values of c and looking for zeros often works – if there is some whole number c that is a root.
Sir Isaac Newton worked out a formula that, given a value close to a root, can move it closer.

Knowledge Box 3.2

Newton’s method for finding roots

If x0 is close to a root of f .x/, then the sequence generated by using the formula
f .xi /
xi C1 D xi will approach the nearby root.
f 0 .xi /

Example 3.4 Use Newton’s method with an initial guess of x0 D 1 to approximate a root of
f .x/ D x 2 2.

Solution:

To start with, we need to find the Newton’s method formula. Since f 0 .x/ D 2x we get
that
xi2 2
xi C1 D xi
2xi

Now compute

1 2
x1 D 1 D 1:5
2

2:25 2
x2 D 1:5 Š 1:4166666667
3
 3.1. POLYNOMIAL ARITHMETIC 95
x3 Š 1:4142157

x4 Š 1:4142126

x5 Š 1:4142126
p
So by the fifth updating the approximation has stabilized atpa value that agrees with 2 up
to seven decimals. Since the roots of f .x/ D x 2 2 are ˙ 2 this is a nice demonstration
that the technique works. Although we are demonstrating Newton’s method on polynomials,
it will work for any function that is continuous and differentiable near a root we are trying to find.

Newton’s method – or other more sophisticated root finding methods – are often built into a
calculator. By showing the values for x1 , x2 , and so on, coding the formula into a spreadsheet is
a relatively low work method of doing the calculations. Once you code the formula you can just
let the spreadsheet perform the iterations.

Example 3.5 Find the Newton’s method formula for approximating roots of:

f .x/ D x 3 C 3x 1

Solution:

Just plug into the form for the Newton’s method formula:

xi3 C 3xi 1
xi C1 D xi
3xi2 C 3
If we start with x0 D 1:0 we get:
x0 D1

x1 D0.5

x2 D0.36111111

x3 D0.32917086

x4 D0.32335786
96 3. THE ARITHMETIC, GEOMETRY, AND CALCULUS OF POLYNOMIALS
x5 D0.32237954

x6 D0.32221744

x7 D0.32219065

x8 D0.32218623

x9 D0.32218549

x10 D0.32218537

x11 D0.32218535

x12 D0.32218535

At which point the number has stopped changing and so is the root to 8 decimals. You can verify
that f .x/ has only one root by graphing it.

The picture on the cover of this book demonstrates how Newton’s method can be used to make
pretty pictures. It plots points on the complex plane with the real part of a number plotted on
the x -axis, and the imaginary part on the y -axis. It uses the polynomial:

x 6 C 2:43
x 4 5:8644
x 2 10:4976

This polynomial has six roots. Points are colored based on which root they converge to when
used as a starting guess for Newton’s method.

PROBLEMS

Problem 3.6 Compute the following polynomial products.

1. .x 3 C 2x 2 C x C 3/  .x C 5/ 4. .x 3 C 3x C 7/  .x 4 C x C 2/
2. .x 2 C 4x 1/  .x 2 4x C 1/ 5. .x C 1/3  .x C 2/3
3. .x 3 C 6x 2 C 11x C 6/  .x 3 6x 2 C 11x 6/ 6. .x 2 C x C 1/3
 3.1. POLYNOMIAL ARITHMETIC 97
Problem 3.7 Perform the following polynomial divisions.

1. .x 4 C 4x 3 C 6x 2 C 4x C 1/  .x C 1/

2. .x 4 x 3 C 2x 2 C x C 3/  .x 2 C x C 1/

3. x 4 C x 3 C 2x 2 C x C 1/  .x 2 C 1/

4. .x 5 C 8x 4 C 21x 3 C 35x 2 C 28x C 15/  .x 2 C 2x C 3/

5. .x 6 x 5 C 2x 4 x 3 C 2x 2 x C 1/  .x 2 x C 1/

6. .x 6 1/  .x 1/

Problem 3.8 Use Newton’s method to approximate the roots of the polynomial

f .x/ D x 3 3x C 1

using initial guesses of x0 D 2; 0; and 2. These should generate three distinct roots.

Problem 3.9 Use Newton’s method to find a root of:

g.x/ D x 5 5x 2 6

Problem 3.10 Earlier in the text it was asserted that a polynomial of odd degree must have at
least one root. Explain why.

Problem 3.11 The polynomials

f .x/ D x 2 2x C 1

and
g.x/ D xr 3x C 2
both have a root at x D 1. Use a spreadsheet to apply Newton’s method to both of these problems
with an initial guess of x0 D 0:4. Both these intial guesses should converge to x D 1 – what is
different about the convergence to one for these two polynomials. Why? Pictures may help.
98 3. THE ARITHMETIC, GEOMETRY, AND CALCULUS OF POLYNOMIALS
Problem 3.12 Find all real roots of the following polynomials.

1. f .x/ D x 4 2x 3 2x 2 2x 3 4. q.x/ D x 4 5x 2 C 4

2. g.x/ D x 3 6x 2 C 11x 6 5. r.x/ D x 4 8x 2 C 5x C 6

3. h.x/ D x 6 64 6. s.x/ D x 3 6x 2 C 12x 8

Problem 3.13 Find a polynomial of the form

x 4 C ax 3 C bx 2 C xc C d
with none of a; b; c; or d zero that has no real roots at all.

Problem 3.14 Suppose that p.x/ is a polynomial. How many roots does p.x/2 C 1 have?
Justify your answer with one or more sentences.

Problem 3.15 Consider the polynomial

q.x/ D x 3 25x
First find its three roots. Then figure out how many roots q.x/2 1 has.

Problem 3.16 Find a polynomial with roots at x D 1; 2; 3; 4; and 5.

Problem 3.17 Find a polynomial with roots at x D 1; 2; 3; 4; and 5 that does not take on any
negative values.

Problem 3.18 Given the Newton’s method formula for a polynomial:

2xi3 xi2 C 4
xi C1 D
3xi2 2xi 4
what is the polynomial?

Problem 3.19 If the Newton’s method formula for a polynomial is

3xi4 C xi2 7
xi C1 D
4xi4 C 10xi
what is the polynomial?
 3.2. QUALITATIVE PROPERTIES OF POLYNOMIALS 99
3.2 QUALITATIVE PROPERTIES OF POLYNOMIALS
Much of the work we do in a calculus class consists of calculating things. Some of the examples
and homework problems are chosen to demonstrate the power of instead deducing things. In this
section we will learn a number of properties of polynomials that help us deduce things about
polynomials. Our first task is understanding how the roots of a polynomial affect the shape of
its graph.

Knowledge Box 3.3

Rolle’s Theorem
Suppose a continuous, differentiable function y D f .x/ has roots at x D a and
at x D b . Then for some c , a  c  b , f 0 .c/ D 0.

The best way to understand Rolle’s theorem is with a pictorial example. The function in Figure
3.1 has a root at x D 2 and one at x D 3. At c D 31 it has a critical value and so a horizontal
tangent line.

3.0

-4.0 4.0

-1.0

Figure 3.1: Illustration of Rolle’s Theorem.

Colloquially, Rolle’s theorem says there is (at least) one horizontal tangent between two roots.
While we use it to describe polynomials, Rolle’s theorem actually applies to all continuous dif-
ferentiable functions.
The pictures in Figures 3.2 and 3.3 show examples of even and odd degree polynomials with the
smallest possible number of roots.
100 3. THE ARITHMETIC, GEOMETRY, AND CALCULUS OF POLYNOMIALS
4.0 4.0

-4.0 4.0 -4.0 4.0

-4.0 -4.0

Figure 3.2: Fourth degree, no roots. Figure 3.3: Third degree, one root.

Knowledge Box 3.4

The number of roots of a polynomial

• An odd degree polynomial has at least one root.
• An even degree polynomial can have no roots.
• A polynomial of degree n has at most n roots.
• An exception to the previous rule is f .x/ D 0 which has roots ev-
erywhere.
• Any number of roots between the minimum and maximum number
of roots are possible.

The fact that a polynomial of odd degree must have at least one root follows from the way its
limits at infinity act. An odd degree polynomial must go to both of ˙1, which means that, on
its way from one infinity to the other, it must pass through zero.

What does Rolle’s theorem tell us about polynomials? Between any two roots there must be a
horizontal tangent. Whenever there is a horizontal tangent, the function changes direction, so
 3.2. QUALITATIVE PROPERTIES OF POLYNOMIALS 101
Rolle’s theorem tells us that the polynomial must change directions between roots.

A polynomial of degree n has at most n 1 horizontal tangents. This follows from the fact that
the derivative has degree n 1 and, so, at most n 1 roots. The hill-tops and valley-bottoms of
a polynomial correspond to the roots of its derivative.

Example 3.20 What is the smallest number of horizontal tangents a polynomial p.x/ may
have?

Solution:

Since horizontal tangents are roots of p 0 .x/, the odd/even degree rules tell us that an
odd degree polynomial may have no horizontal tangents, while an even degree polynomial must
have at least one.
In fact, p.x/ D x n exemplifies this minimum number of horizontal tangents. The even powers
of x have a single critical value at x D 0 that yields a sign chart of

.C1/ .0/ C C C .C1/

102 3. THE ARITHMETIC, GEOMETRY, AND CALCULUS OF POLYNOMIALS
while the odd ones have a sign chart of

. 1/ C C C .0/ C C C .C1/:

3.2.1 MULTIPLICITY OF ROOTS

What makes it possible for a polynomial of degree n to have less than n roots? Part of this is we
can add constants to a polynomial so that it intersects the x -axis as little as possible. Another
factor is that roots can be repeated. Look at the polynomial shown in Figure 3.4:
1 1 3 1 2
p.x/ D .x C 2/.x 2/2 D x x xC2
4 4 2
The fact that the root at x D 2 is squared makes the graph bounce off of the x axis instead of
passing through. This leads to a definition.
Definition 3.1 If .x a/ divides a polynomial p.x/, then the highest power k so that .x a/k di-
vides p.x/ is the multiplicity of the root x D a in p.x/.
 3.2. QUALITATIVE PROPERTIES OF POLYNOMIALS 103
4.0

xD2
-4.0 4.0
xD 2

-4.0

1
Figure 3.4: p.x/ D .x C 2/.x 2/2
4

The multiplicity of the root x D 2 in Figure 3.4 is two.

Knowledge Box 3.5

The effect of root multiplicity

If p.x/ is a polynomial, and x D a is a root of p.x/, then
• If the root at x D a is of odd multiplicity, the graph of p.x/ passes
through the x -axis at x D a.
• If the root at x D a is of even multiplicity, the graph of p.x/
touches the x -axis and then bounces back on the same side it was on
before at x D a.
104 3. THE ARITHMETIC, GEOMETRY, AND CALCULUS OF POLYNOMIALS
Notice that, since the root at x D 2 in Figure 3.4 has multiplicity two, there is a local minimum
and a critical value at x D 2. This effect is fairly general and can be summarized as in Knowledge
Box 3.6.
Knowledge Box 3.6

Roots with multiplicity and derivatives

If p.x/ is a polynomial, and x D a is a root of p.x/ with multi-
plicity k > 1, then x D a is also a root of f 0 .x/ with multiplicity
k 1.

Example 3.21 Verify that if

f .x/ D .x 2/2 .x C 2/ D x 3 2x 2 4x C 8;

then f 0 .2/ D 0.

Solution:

Compute
f 0 .x/ D 3x 2 4x 4
Then f 0 .2/ D 12 8 4 D 0, and 2 is a root of the derivative.

The examples in this section show us how we can construct polynomials with particular prop-
erties quite easily by putting in the roots we want with the correct multiplicities. This sort of
information is made even more useful by the fact that polynomials can be used to approximate
other functions on bounded intervals.

PROBLEMS

Problem 3.22 Show that the number of horizontal tangents between two roots – that do not
have another root between them – must be odd.

Problem 3.23 Construct a polynomial that has more than one horizontal tangent between two
roots that do not have another root between them.
 3.2. QUALITATIVE PROPERTIES OF POLYNOMIALS 105
Problem 3.24 Prove that a non-constant polynomial can have zero inflection values; do this
by giving an example of one.

Problem 3.25 Show that a polynomial of degree n can have at most n 2 inflection values.

Problem 3.26 Give an example of a non-constant polynomial with two roots that never takes
on negative values. Explain why your example is correct.

Problem 3.27 Suppose a polynomial has three distinct roots, r1 , r2 , r3 . What can you deduce
from this fact alone about the degree of the polynomial?

Problem 3.28 Suppose we have two polynomials p.x/ and q.x/ and that all roots of p.x/ are
roots of q.x/ but at least one root of q.x/ is not a root of p.x/. If p.x/ has two roots, provide
three examples. In the first, p.x/ should have a degree less than q.x/; in the second, p.x/ and
q.x/ have equal degree; in the third, p.x/ should have a higher degree than q.x/.

Problem 3.29 Deduce as much as you can from the graph above. The grids are of length one
unit. In particular, what is the minimum degree and what can you deduce about the roots and
their multiplicity?
106 3. THE ARITHMETIC, GEOMETRY, AND CALCULUS OF POLYNOMIALS

Problem 3.30 Deduce as much as you can from the graph above. The grids are of length one
unit. What is the minimum possible degree and what can you deduce about the roots and their
multiplicity?

Problem 3.31 Deduce as much as you can from the graph above. The grids are of length one
unit. In particular what is the minimum degree and what is possible to deduce about the roots
and their multiplicity?

Problem 3.32 What is the smallest possible degree for a polynomial that is an answer to Ques-
tion 3.26.
 3.3. L’HÔPITAL’S RULE; STRANGE POLYNOMIALS 107
Problem 3.33 Using the product rule, prove the statement: If p.x/ is a polynomial and x D a
is a root of p.x/ with multiplicity k > 1, then x D a is also a root of f 0 .x/ (from Knowledge
Box 3.6).

Problem 3.34 Using material from this section, prove that

y D cos.x/

is not a polynomial function.

Problem 3.35 Suppose two polynomials share the same roots and, for those roots, the same
multiplicities. If the polynomials are not equal, how do they differ?

3.3 L’HÔPITAL’S RULE; STRANGE POLYNOMIALS

During our study of curve sketching, we developed a rule of thumb about computing the limit
at infinity of the ratio of two polynomials. If p.x/ D ax n C


is a polynomial of degree n and
q.x/ D bx m C


is a polynomial of degree m the rule said:

8
p.x/ < ˙1 n > m
lim D a=b n D m
x!1 q.x/ :
0 n<m

This rule is a consequence of L’Hôpital’s rule, a useful tool for computing limits that cannot be
resolved other ways.

Knowledge Box 3.7

L’Hôpital’s rule at x D c
If lim f .x/ D 0 and lim g.x/ D 0
x!c x!c
or
if lim f .x/ D ˙1 and lim g.x/ D ˙1, then
x!c x!c

f .x/ f 0 .x/
lim D lim
x!c g.x/ x!c g 0 .x/
108 3. THE ARITHMETIC, GEOMETRY, AND CALCULUS OF POLYNOMIALS
The hypotheses that the limits of the numerator and denominator are both going to zero or
both going to some sort of infinity are critical. L’Hôpital’s rule will give the wrong answer if
the hypotheses are not satisfied.

Let’s do an example.

Example 3.36 Compute

x2 4
lim
x!2 x 2
with L’Hôpital’s rule.

Solution:

We could resolve this limit by factoring and canceling, but, since the top and bottom of
the fraction are both 0 at x D 2, L’Hôpital’s rule applies.

x2 4 2x 4
lim D lim D D4
x!2 x 2 x!2 1 1

Knowledge Box 3.8

L’Hôpital’s rule as x ! 1
If lim f .x/ D 0 and lim g.x/ D 0
x!1 x!1
or
if lim f .x/ D ˙1 and lim g.x/ D ˙1, then
x!1 x!1

f .x/ f 0 .x/
lim D lim
x!1 g.x/ x!1 g 0 .x/

This version of L’Hôpital’s rule is the one that gives us the rule-of-thumb for the limit at
infinity of ratios of polynomials.
 3.3. L’HÔPITAL’S RULE; STRANGE POLYNOMIALS 109
Let’s do an example.

Example 3.37 Compute

x 2 C 2x C 3
lim :
x!1 3x 2 C 1
Solution:

Both the numerator and denominator are going to infinity, so:

x 2 C 2x C 3 2x C 2
lim D lim L’Hôpital’s rule still applies
x!1 3x 2 C 1 x!1 6x

2
D lim
x!1 6

1
D
3
1
And so the limit at infinity is – exactly what the rule of thumb gives us.
3
˙

The reason the rule of thumb works for polynomials of equal degree is that L’Hôpital’s rule
applies until we have taken so many derivatives that the numerator and denominator are con-
stants, at which point the accumulated values from bringing the power out front cancel, leaving
the ratio of the highest degree coefficients as the limit.
At this point, we add a name to our mathematical vocabulary with the following definition.

Definition 3.2 A rational function is a function that is the ratio of polynomial functions.

We have been working with and graphing rational functions for some time now, so they are not
new. We just have a name for them now. While L’Hôpital’s rule was introduced to back-fill
the rule of thumb for limits at infinity of rational functions (to use their new name), it in fact
applies to all continuous, differentiable functions.

Let’s do a couple of examples.

110 3. THE ARITHMETIC, GEOMETRY, AND CALCULUS OF POLYNOMIALS
Example 3.38 Compute
sin.x/
lim :
x!0 x

Solution:

The numerator and denominator both go to zero, so L’Hôpital’s rule applies.

sin.x/ cos.x/
lim D lim D1
x!0 x x!0 1

Example 3.39 Compute

cos.x/ 1
lim :
x!0 x

Solution:

The numerator and denominator are both going to zero, so L’Hôpital’s rule applies.

cos.x/ 1 sin.x/
lim D lim D0
x!0 x x!0 1

As with every other technique we have learned, we may use algebraic rearrangement to permit
us to extend the reach of L’Hôpital’s rule.

Example 3.40 Compute

lim x
ln.x/:
x!0

Solution:

This problem does not fit the form for L’Hôpital’s rule, but we can coerce it into that
 3.3. L’HÔPITAL’S RULE; STRANGE POLYNOMIALS 111
form:
ln.x/ 1
lim x
ln.x/ D lim Now in correct form for L’Hôpital
x!0 x!0 1 1
x

1
D lim x Use L’Hôpital
x!0 1
x 2

x2 x
D lim D lim Simplify
x!0 x x!0 1

D0 Done

So x moves toward zero faster than ln.x/ moves toward negative infinity as x ! 0.

It is possible for L’Hôpital to take you to a limit that clearly does not exist. As before, the rules-
of-thumb developed for curve sketching can also supply the answer, but in mathematics a solid
method is preferred.

x3 C 1
Example 3.41 Find lim .
x!1 x2
Solution:

The rule-of-thumb tells us this diverges to 1, but let’s work through this with L’Hôpital’s rule
and see where it goes.

x3 C 1 3x 2 3
lim 2
D lim D lim x!1
x!1 x x!1 2x x!1 2

L’Hôpital’s rules also tells us this diverges to infinity.

˙
112 3. THE ARITHMETIC, GEOMETRY, AND CALCULUS OF POLYNOMIALS
Since this chapter is about polynomials, Knowledge Box 3.9 gives a number of useful polyno-
mial identities that did not really fit before this point.

Knowledge Box 3.9

Useful polynomial identities

• x2 a2 D .x a/.x C a/
• x3 a3 D .x a/.x 2 C ax C a2 /
• x 3 C a3 D .x C a/.x 2 ax C a2 /
• xn an D .x a/.x n 1
C ax n 2
C


an 2 x C an 1 /

That last identity is quite useful for adding up powers of a variable. Another form of the identity,
with a D 1 is
xn 1
D xn 1 C xn 2 C


C x C 1
x 1
How can this be applied?

Example 3.42 Compute

20
X
1:5k
kD0

Solution:
 21
3
20
X 1
k 1:521 1 2 321 221 10458256051
1:5 D D D D Š 9973:8
1:5 1 1 2 20 1048576
kD0
2
˙

Let’s do another with slightly less insane numbers.

Example 3.43 Find
11
X
3k
kD0
 3.3. L’HÔPITAL’S RULE; STRANGE POLYNOMIALS 113
Solution:
11
X 312 1 1
3k D D
531440 D 265720
3 1 2
kD0

This material is very important later, when we study sequences and series.

3.3.1 STRANGE POLYNOMIALS

The goal of this section is to let us broaden the reach of what we know about polynomials to
other functions. Suppose that we wish to solve the equation:

e2x 3ex C 2 D 0

This is clearly not a polynomial equation, but we can find a polynomial with the following trick.
Let u D ex . Then, since e2x D .ex /2 D u2 , we transform the original problem into:

u2 3u C 2 D 0

This factors into .u 2/.u 1/ D 0 so u D 1; 2. Reversing the transformation ex D 1; 2. Take

the log of both sides and the result becomes x D ln.1/; ln.2/ or x D 0; ln.2/.
This trick, called u-substitution, lets us turn one type of equation into another. This technique
will become very useful in Chapter 4, but for now it permits us to solve a wider variety of
equations. Some care is required.

Example 3.44 Find the solutions to the equation:

2 sin2 .x/ 5 sin.x/ C 2

Solution:

Try u D sin.x/ changing the problem to:

2u2 5u C 2 D 0

As in the previous example, this factors, giving us .2u 1/.u 2/ D 0, and we get that
u D 1=2; 2.

So far, so good.
114 3. THE ARITHMETIC, GEOMETRY, AND CALCULUS OF POLYNOMIALS
But when we reverse the transformation we get sin.x/ D 1=2; 2 and sin.x/ D 2 is im-
possible! We know that 1  sin.x/  1.

So even though there are two values for u, only one leads to a solution of the original
equation.
1 
sin.x/ D or x D
2 6
Giving the sole solution to the problem.
˙

Example 3.45 Solve ln.x/2 3 ln.x/ 3 D 0.

Solution:

Here the obvious substitution is u D ln.x/, and we get u2 3u 3 D 0. Apply the quadratic
equation and reverse the transformation:

p
3˙ 9 4.1/. 3/
uD
2
p
3˙ 21
D Only one positive root
2
p
3C 21
ln.x/ D Š 3:79
2

x Š 1:33

We dismissed the negative root because only positive numbers have logs.

This trick sometimes helps with polynomials as well.

 3.3. L’HÔPITAL’S RULE; STRANGE POLYNOMIALS 115
4 2
Example 3.46 Solve x 6x C 8 D 0.

Solution:

Take u D x 2 . Then
u2 6u C 8 D 0
p
Factor: .u 2/.u 4/ D 0, and we get u D 2; 4. So x 2 D 2; 4 and thus x D ˙ 2; ˙2.

PROBLEMS

Problem 3.47 For each of the following limits, use L’Hôpital’s rule to resolve the limit.

x 3 x3 C 8
1. lim 4. lim
x!3 x 2 9 x! 2 x3 C x2 xC2
x3 x 6 x 5 32
2. lim 5. lim
x!2 x2 4 x!2 x 2 4
x6 1 cos.x/ C 1
3. lim 6. lim
x!1 x 2 1 x! x 

Problem 3.48 Before, when we had to resolve a limit of the form 00 by algebra using
cancellation, we had to know the factorization of the numerator and denominator.

Discuss: does L’Hôpital’s rule make this process easier?

Problem 3.49 Find, in general, using L’-Hôpital’s rule

xn 1
lim
x!1 x2 1
116 3. THE ARITHMETIC, GEOMETRY, AND CALCULUS OF POLYNOMIALS
Problem 3.50 For each of the following limits, use L’Hôpital’s rule to resolve the limit.

x 2 C 3x C 5 4. lim .x 1/
ln.x 3 1/
1. lim x!1
x!1 x 2 3x C 4
2e x sin.3x/
2. lim x 5. lim
x!1 e C 4 x!0 2x
x C ex cos.x/ 1
3. lim 6. lim
x!1 x C x 2 C 3ex x!0 x2

Problem 3.51 Compute the following sums with a polynomial identity. Some of these may
require using the identity twice.

19
X 8
X 12
X
1. 0:9k 3. 7k 5. 0:1k
kD0 kD0 kD0

25
X 15
X 10
X
2. 1:2k 4. 1:5k 6. . 0:5/k
kD0 kD5 kD0

Problem 3.52 Give an example of a polynomial p.x/ with two roots for which

p.sin.x// D 0

has no solutions. Explain why your solution is correct.

Problem 3.53 If p.x/ is a polynomial with k roots, what are the possible number of solutions
to
p.ln.x// D 0‹
 3.3. L’HÔPITAL’S RULE; STRANGE POLYNOMIALS 117
Problem 3.54 Solve the following equations. Note that they are, in a sense, polynomials.

1. 2 sin2 .x/ C sin.x/ D 0 5. e3x 6e2x C 11ex 6D0

2. 4 cos2 .x/ 3D0 6. ln2 .x/ 8 ln.x/ C 15 D 0

3. 4 sin3 .x/ sin.x/ D 0 7. cos2 .x/ sin2 .x/ cos2 .2x/ D 0

4. e2x 7ex C 12 D 0 8. tan3 .x/ tan.x/ D 0

Problem 3.55 Factor x 6 729 as far as you can.

Problem 3.56 Factor x 8 256 as far as you can.

Problem 3.57 If we know that p.x/ is a polynomial with n roots and q.x/ is a polynomial with
m roots, with n  m, then what do we know about the number of roots of

h.x/ D p.x/
q.x/‹

 119

CHAPTER 4

Methods of Integration I
The integration methods we have learned thus far are based on the fact that integrals are reversed
derivatives. To some extent we can increase the reach of the “reverse derivative” technique by
setting things up with algebra. The chapters on methods of integration will introduce the most
useful and broadly applicable of thousands of integration methods that have been found over
the last several centuries.

4.1 u-SUBSTITUTION
Every integral method is a reversed derivative, and u-substitution is the reverse of the chain rule.
We introduced u-substitution in Chapter 3 to permit us to solve more complex equations with
polynomial techniques. For integrals the technique is much the same, except that we need to
worry about the differential. Let’s start with an example.
Z
2
Example 4.1 Find 2xex
dx:

Solution:

This is not an integral for which we already have a form. Set u D x 2 and then compute
du D 2x
dx . These pieces make up all of the integral and we can now solve the problem as
follows:
Z Z
x2 2
2xe
dx D ex
.2x
dx/ Set it up

Z
D eu
du Substitute

D eu C C Integrate

2
D ex C C At the end, reverse the substitution.

˙
120 4. METHODS OF INTEGRATION I
Recall that you can check an integral by taking a derivative, so
 2 0 2 2
ex C C D ex
2x D 2xex

and we see the substitution worked to let us do this integral. Notice that the chain rule re-created
du for us.
Let’s do some more examples.
Z
Example 4.2 Find cos.2x/
dx .

Solution:

We know that the integral of cosine is sine. So the only problem is the 2x . This is where
to make our u-substitution.
1
If we let u D 2x then du D 2dx so dx D du.
2
Now we substitute and integrate:

Z Z
1
cos.2x/
dx D cos.u/ du
2
Z
1
D cos.u/
du
2

1
D sin.u/ C C
2

1
D sin.2x/ C C
2

This last example has a very general version that can let you do some types of u-substitution in
your head.
 4.1. u-SUBSTITUTION 121
Z Z
Example 4.3 Suppose that F .x/ C C D f .x/
dx . Then what is f .ax C b/
dx‹
Solution:
1
Let u D ax C b and so du D a
dx and dx D
du. Then:
a
Z Z
1
f .ax C b/
dx D f .u/

du
a
Z
1
D f .u/
du
a

1
D F .u/ C C
a

1
D F .ax C b/ C C
a
˙

This is important enough to put in a Knowledge Box. What we are doing is correcting for
composing a function with a linear function.
Knowledge Box 4.1

u-substitution to correct a linear composition

Z
If f .x/
dx D F .x/ C C , then
Z
1
f .ax C b/
dx D F .ax C b/ C C
a

Z
Example 4.4 Find e0:5x 1
dx .

Solution:
2e0:5x 1
CC
by directly applying the linear correction with a D 0:5 and b D 1.
˙
122 4. METHODS OF INTEGRATION I
This example shows how much calculation the linear correction technique saves.
Sometimes a u substitution is not obvious.

Example 4.5 Find

Z
dx
x ln.x/
Solution:
d 1
The key to this one is remembering that ln.x/ D making u D ln.x/ a natural choice to
dx x
dx
try. Then du D and we can integrate.
x
Z Z
dx 1 dx
D

x ln.x/ ln.x/ x
Z
1
D
du
u

D ln.u/ C C

D ln.ln.x// C C

Sometimes the substitution is so unobvious that people clearly figured out the integral another
way. The following example has this property.

Example 4.6 Find Z

sec.x/
dx:

Solution:
We begin with an algebraic transformation.
Z Z
sec.x/ C tan.x/
sec.x/
dx D sec.x/
dx
sec.x/ C tan.x/
Z
sec2 .x/ C sec.x/ tan.x/
D
dx
tan.x/ C sec.x/
 4.1. u-SUBSTITUTION 123
At this point let u D tan.x/ C sec.x/ so that


du D sec2 .x/ C sec.x/ tan.x/ dx

which makes this all:

Z Z
du
sec.x/
dx D
u

D ln.juj/ C C

D ln.j tan.x/ C sec.x/j/ C C

The absolute value bars are needed because the value of these functions can be negative.
˙

We also now know enough to capture the integral of another of the basic functions.
Z
Example 4.7 Compute tan.x/
dx .
Solution:

sin.x/
Since tan.x/ D we let u D cos.x/ obtaining du D sin.x/dx or du D sin.x/dx .
cos.x/
Plugging in the substitution we get
Z Z Z
du du
tan.x/
dx D D
u u

D ln.juj/ C C
 
1
D ln CC
juj
 
1
D ln CC
j cos.x/j

D ln.j sec.x/j/ C C

Taking the minus sign inside the log is a negative first power or reciprocal, which is how we get
to the secant.
˙
124 4. METHODS OF INTEGRATION I
Knowledge Box 4.2

Basic trig integrals from u-substitution

Z
• tan.x/
dx D ln.j sec.x/j/ C C
Z
• sec.x/
dx D ln.j sec.x/ C tan.x/j/ C C

It is possible to have fairly obvious u-substitution problems.

Example 4.8 Find

Z
x.x 2 C 3/6
dx:

Solution:
1
Let u D x 2 C 3; then du D 2x
dx and so du D x
dx . Integrate:
2
Z Z
x.x 2 C 3/6
dx D .x 2 C 3/6
.x
dx/

Z
1
D u6 du
2
Z
1
D u6
du
2

1 1 7
D  u CC
2 7

1 2
D .x C 3/7 C C
14

˙
 4.1. u-SUBSTITUTION 125
It is possible to need algebra to set up a u-substitution, often in less apocalyptic ways than the
integral of secant, earlier in this section.

Example 4.9 Compute

Z
x 2 C 2x C 1

dx
x2 C 1

Solution:

Z Z Z
x 2 C 2x C 1 x2 C 1 2x

dx D dx C dx
x2 C 1 x2 C 1 x2C1
Z Z
2x
D dx C dx
x2C1
Z
2x
DxC dx C C
x2C1

At this point let u D x 2 C 1 so that du D 2x
dx and finish

Z Z
x 2 C 2x C 1 du

dx D x C CC
x2 C 1 u

D x C ln.u/ C C

D x C ln.x 2 C 1/ C C

Done! If you have a rational function whose denominator is not higher degree, dividing first is
a good way to simplify things.
˙
126 4. METHODS OF INTEGRATION I
Another use of u-substitution is to integrate odd powers of sine and cosine.

Example 4.10 Find

Z
cos3 .x/
dx:

Solution:

Z Z
cos3 .x/
dx D cos2 .x/
cos.x/
dx

Z
D .1 sin2 .x//
cos.x/
dx

Z Z
D cos.x/
dx sin2 .x/ cos.x/
dx

Z
D sin.x/ sin2 .x/ cos.x/
dx

Let u D sin.x/ so that du D cos.x/
dx

Z
D sin.x/ u2
du

1 3
D sin.x/ u CC
3

1
D sin.x/ sin3 .x/ C C
3

˙
 4.1. u-SUBSTITUTION 127
We have enough machinery to integrate even powers of sine and cosine now. We will learn how
to integrate odd powers in Section 4.3.

Example 4.11 Find

Z
sin2 .x/
dx:

Solution:
Z Z
2 1 1 1
sin .x/
dx D .1 cos.2x//
dx D .x sin.2x// C C
2 2 2

1 1
We are using a trig identity for the second step and the answer is x sin.2x/ C C .
2 4
˙

4.1.1 SUBSTITUTION IN DEFINITE INTEGRALS

An issue we have not dealt with is substitution in definite integrals. There are two viable ap-
proaches to this topic.

• Substitute back to the original variable before plugging in the limits.

• Apply the substitution to the limits to get new limits.

So far we have always substituted back to the original variable to finish the integral.

Example 4.12 Compute

Z 3 p
x x 2 C 4
dx
0

Solution:
1
This is a simple u-substitution. Let u D x 2 C 4, so that du D 2x
dx and du D x
dx .
2
Applying the substitution to the limits we get that 0 to 3 becomes u.0/ D 4 and u.3/ D 13.
128 4. METHODS OF INTEGRATION I

Z 3 p Z 13
2
p 1
x x C 4
dx D u
du
0 4 2
Z 13
1
D u1=2
du
2 4

1 2 3=2 ˇˇ13
D
u ˇ
2 3 4

1  3=2 
D 13 43=2
3

1 p 
D 2197 8 DŠ 12:96
3
˙

Example 4.13 Compute

Z 1
ex

dx
0 e2x C 1
Solution:

Let u D ex so that du D ex
dx . The limits become u.0/ D 1 and u.1/ D e .

Z 1 Z e
ex du
2x C 1

dx D 2C1
0 e 1 u
ˇe
ˇ
D tan 1
.u/ˇˇ
1

1 1
D tan .e/ tan .0/

1
D tan .e/

Š 1:218

˙
 4.1. u-SUBSTITUTION 129
PROBLEMS
Problem 4.14 Integrate the following by using u-substitution. Explicitly state u.

Z Z 3
1. sin.3x/
dx 4. 4x
.2x 2 C 1/5
dx
1
Z Z 4
2. sec2 .5x C 1/
dx 1
5.
dx
1 x
ln2 .x/
Z 4 Z
x
3. 2

dx 6. .sin.x/ C cos.x//2
dx
0 x C1

Problem 4.15 Integrate the following.

Z Z 1
1. cos.x/
dx 4. .3x C 4/12
dx
0
Z Z
2. e3xC2
dx 5. sec.2x/ tan.2x/
dx
Z Z 1
dx
3. sin.14x 5/
dx 6.
0 3x C 5
Z Z
Problem 4.16 Show that if F .x/ C C D f .x/
dx , then x
f .x 2 C a/
dx D
1
2
F .x 2 C a/ C C .

Z
Problem 4.17 Compute cot.x/
dx:

Z
Problem 4.18 Compute csc.x/
dx:

Z =4
Problem 4.19 Compute sin4 .x/ cos.x/
dx:
0

Z =2
Problem 4.20 Compute cos4 .x/
dx:
0
130 4. METHODS OF INTEGRATION I
Z
Problem 4.21 Compute cos5 .x/
dx:

Z
Problem 4.22 Compute sin3 .x/ cos3 .x/
dx:

Z
Problem 4.23 Compute tan3 .x/ sec2 .x/
dx:

Z
sin3 .x/
Problem 4.24 Compute
dx:
cos5 .x/

Problem 4.25 Integrate the following by using u-substitution, using algebra as needed. Ex-
plicitly state u.

Z Z 2
1. xex
dx
2 2x C 1
6.
dx
0 .x 2 C x C 1/5
Z 1 2
x C 4x C 2
2.
dx Z
0 x2 C 2 7. cos7 .2x/ sin.x/ cos.x/
dx
Z
x5
3.
dx
x2 C 1 Z
Z ex
2
ln .x/ C 4 ln.x/ C 2 8.
dx
4.
dx ex C 1
x
Z Z
sin2 .x/ tan 1 .x/
5.
dx 9.
dx
cos2 .x/ C 1 x2 C 1
 4.2. INTEGRATION BY PARTS 131
Problem 4.26 Perform the following definite integrals.

Z 4 Z 2
ln.x/
1. .2x 3/8
dx 4.
dx
0 1 x
Z p Z 2
 x5
2. 2
x sin.x /
dx 5.
dx
0 2 x2 C 1
Z Z p
4
1 3
2 C6xC1 2x
3. .3x C 1/e 9x
dx 6.
dx
0 0 x4C1

4.2 INTEGRATION BY PARTS

Probably the easiest of the complex derivative rules is the product rule. The reversed product
rule is a technique called integration by parts. Let’s derive it.

.U V /0 D V
d U C U
dV This is just the product rule

U
dV D .U V /0 V
dU Rearrange
Z Z Z
0
U
dV D .U V / V
dU Integrate both sides

Z Z
U
dV D U V V
dU This is the formula

Knowledge Box 4.3

Integration by parts
Z Z
U
dV D U V V
dU

The technique for using this rule is not obvious from its statement. The “parts” are U and V
together with their derivatives. Examples are needed.
132 4. METHODS OF INTEGRATION I
Example 4.27 Compute
Z
x
cos.x/
dx:

Solution:

Choose U D x and dV D cos.x/
dx .

Z
Then d U D dx and V D cos.x/
dx D sin.x/.

When doing integration by parts we add the CC as the last step of the integration pro-
cess.

Now that we have the parts and their derivatives, we apply the integration by parts for-
mula.
Z Z
U
dV D U V V
dU

Z Z
x
cos.x/
dx D x
sin.x/ sin.x/
dx

D x
sin.x/ . cos.x// C C

D x
sin.x/ C cos.x/ C C

Let’s check this result by taking the derivative:

.x
sin.x/ C cos.x/ C C /0 D sin.x/ C x
cos.x/ sin.x/ D x
cos.x/

So the method worked.

˙

There is a substantial strategic component to choosing the parts U and dV when doing
integration by parts. You take the derivative of U , and you must integrate dV , and when you’re
done it would be lovely if the result could be integrated without too much difficulty. In general
you choose U so that differentiation will make a problem go away.

Let’s do another simple example.

 4.2. INTEGRATION BY PARTS 133

Example 4.28 Find Z

x ex
dx:

Solution:

We can integrate ex , and x goes away if we take its derivative.

So, choose U D x , dV D ex
dx . This means d U D dx and V D ex .

Apply the formula:

Z Z
U
dV D U V V
dU

Z Z
x x
xe
dx D xe ex
dx

D xex ex C C

Sometimes the choice of parts is not obvious. Let’s pick up another basic integral using integra-
tion by parts.

Example 4.29 Compute Z

1
tan .x/
dx

Solution:

At first it might look like there are no parts, but there are.

dx
Choose U D tan 1 .x/ and dV D dx . Then d U D and V D x .
x2 C 1
Apply the formula: Z Z
U
dV D U V V
dU
134 4. METHODS OF INTEGRATION I
Z Z
1 1 x
tan .x/
dx D x
tan .x/
dx This is a substitution integral.
x2 C1

1
Let r D x 2 C 1. Then dr D x
dx .
2
Z
1 1 dr
D x
tan .x/
2 r

1 1
D x
tan .x/ ln.r/ C C
2

1 1
D x
tan .x/ ln.x 2 C 1/ C C
2

Notice that, since U and d U are used in integration by parts, we used “r -substitution” instead
of u-substitution for the part of the integral that needed substitution.

Knowledge Box 4.4

Integral of the arctangent function

Z
1
tan 1 .x/
dx D x
tan 1 .x/ ln.x 2 C 1/ C C
2

Z
Example 4.30 Compute x 2
ln.x/
dx:

Solution:

Tactically, we know that ln.x/ turns into a (negative) power of x when we take its derivative.
So a natural choice to try is U D ln.x/, dV D x 2
dx .

dx 1
This makes d U D and V D x 3 .
x 3
Apply the formula.
Z Z
U
dV D U V V
dU
 4.2. INTEGRATION BY PARTS 135
Z Z
1 3 1 3 dx
x 2
ln.x/
dx D x
ln.x/ x

3 3 x
Z
1 1
D x 3
ln.x/ x 2
dx
3 3

1 3 1 3
D x
ln.x/ x CC
3 9
˙

Sometimes integration by parts must be applied more than once to finish a problem.
Z
Example 4.31 Compute x 2
ex
dx:

Solution:

Since taking the derivative of a power of x at least makes it a smaller power of x , choose
u D x 2 , dV D ex
dx .

Then d U D 2x
dx and V D ex .

Integrate by parts:
Z Z
x 2
ex
dx D x 2
ex 2x
ex
dx

Z
2 x
Dx
e 2 x
ex
dx Integrate by parts again.

U D xI d V D ex
dx

d U D dxI V D ex
 Z 
2 x
Dx
e 2 x
ex x
e
dx

D x 2
ex 2x
ex C 2ex C C
˙
136 4. METHODS OF INTEGRATION I
Notice that the second integration by parts in this example was very similar to the one in Example
4.28. We could have simply plugged that result in. The example was worked in full to show how
two integrations by parts are needed.
Let’s consider for a minute what happened in the last example. To deal with the x 2 part of the
integral we integrated by parts twice – reducing the power by one in each step. That means that,
for example,
Z
x 5
ex
dx

would require integration by parts five times.

Fortunately, there is a shortcut. What happens if we integrate

Z
f .x/
ex
dx

Choose U D f .x/ and d V D ex
dx .

Then d U D f 0 .x/ and V D ex .

Integrate by parts and we get:

Z Z
f .x/
ex
dx D f .x/ex f 0 .x/ex
dx

Which, used correctly, is a remarkable shortcut.

Knowledge Box 4.5

Z
Shortcut for p.x/ex
dx

Suppose that p.x/ is a polynomial. Then

Z


p.x/ex
dx D p.x/ p 0 .x/ C p 00 .x/ p 000 .x/ C


ex C C

Z
The formula in Knowledge Box 4.5 comes from applying the formula for p.x/ex
dx many
times. Let’s use the shortcut in an example.
 4.2. INTEGRATION BY PARTS 137
Example 4.32 Find
Z
x 5
ex
dx:

Solution:
Applying the shortcut we get that
Z


x 5
ex
dx D x 5 5x 4 C 20x 3 60x 2 C 120x 120 ex C C

All we need to do is take alternating signs of x 5 and its derivatives. If the polynomial is not just
a power of x this is a little more complicated.
˙

Z
Example 4.33 Find .x 2 C 2x C 2/ex dx:
Solution:

p.x/ D x 2 C 2x C 2

p.x/0 D 2x C 2

p.x/00 D 2

So
Z
.x 2 C 2x C 2/ex D .x 2 C 2x C 2 .2x C 2/ C 2/ex C C

D .x 2 C 2/ex C C

The key fact is that, if you keep taking derivatives of a polynomial, then at some point you
get to zero. With a non-polynomial function the shortcut can produce an infinite object.
˙

Now we look at another technique, circular integration by parts. You need this technique when,
in the course of integrating by parts, you arrive back at some version of the integral you started
with. This sounds a bit awful, but it is actually good news. When this happens the integral may
be finished using only algebra.
138 4. METHODS OF INTEGRATION I
Example 4.34 Compute
Z
sin.x/ex
dx:

Solution:

There is no natural choice of parts. So pick anything and plunge in.

Let U D sin.x/ and d V D ex
dx . Then d U D cos.x/
dx and V D ex :

Z Z
sin.x/ex
dx D sin.x/ex cos.x/ex
dx

Let: U D cos.x/ and dV D ex
dx (Second integration by parts.)

Then: d U D sin.x/
dx and V D ex

 Z 
x
D sin.x/e cos.x/ex x
sin.x/e
dx

Z Z
x x
sin.x/e
dx D .sin.x/ cos.x//e sin.x/ex
dx

Notice that the remaining integral equals the left-hand side.

This is the circular part.

Z
So, 2 sin.x/ex
dx D .sin.x/ cos.x//ex
Divide by two and we get:
Z
1
sin.x/ex
dx D .sin.x/ cos.x//ex C C
2

Notice that we had to separately remember to put in the C C , because we finessed the final
integral where we would normally have generated it.
˙

The power of these methods of integration are expanded in the homework problems. We will
also see a horrific example of circular integration by parts in Section 4.3.
 4.2. INTEGRATION BY PARTS 139
PROBLEMS
Problem 4.35 Do each of the following integrals using integration by parts.

Z Z
1. x
sin.2x/
dx 4. x
ln.x/
dx
Z Z
5x
2. x
e
dx 5. x
sin.x/ cos.x/dx
Z Z
3. ln.x/
dx 6. ln.x 2 C 1/
dx

Problem 4.36 For Examples 4.28-4.31, verify the formulas by taking an appropriate derivative.

Z
Problem 4.37 Compute x n ln.x/
dx for n  1

Z
Problem 4.38 In this section we develop a shortcut for p.x/ex
dx where p.x/ is a polyno-
mial. Find the corresponding shortcut for
Z
x
p.x/e
dx

Z
Problem 4.39 In this section we develop a shortcut for p.x/ex
dx where p.x/ is a polyno-
mial. Find the corresponding shortcut for
Z
p.x/ sin.x/
dx

Z
Problem 4.40 In this section we develop a shortcut for p.x/ex
dx where p.x/ is a polyno-
mial. Find the corresponding shortcut for
Z
p.x/ cos.x/
dx
140 4. METHODS OF INTEGRATION I
Problem 4.41 Compute the following integrals using integration by parts.

Z Z
4 x


1. x e
dx 4. x2 C 1 e x

dx
Z Z
3 2 x
2. .x C 2x C 3x C 4/e
dx 5. x 2 sin.x/
dx
Z Z
3 x
3. x e
dx 6. x 5 cos.x/
dx

Problem 4.42 Compute the following integrals using integration by parts.

Z Z
x x
1. cos.x/e
dx 4. sin.x/e
dx
Z Z
x
2. sin.2x/e
dx 5. .cos.x/ C sin.x// ex
dx
Z Z
3x
3. cos.x/e
dx 6. cos.2x/e5x
dx

Problem 4.43 Compute Z

1
x
tan .x/
dx

Problem 4.44 Compute Z

x 3
.x 2 C 1/10
dx

Problem 4.45 Compute Z

x .cos.ax/ C sin.bx//
dx

Problem 4.46 Compute Z

sin.ax/ebx
dx
 4.3. INTEGRATING TRIG FUNCTIONS 141
4.3 INTEGRATING TRIG FUNCTIONS

For completeness we start with a Knowledge Box of the integrals of trigonometric functions
we have already obtained. Most of this section is about how to deal with products of powers of
trig functions, and these integrals are building blocks.

Knowledge Box 4.6

Trig-related integral forms

Z
• sin.x/
dx D cos.x/ C C
Z
• cos.x/
dx D sin.x/ C C
Z
• tan.x/
dx D ln j sec.x/j C C
Z
• sec.x/
dx D ln j tan.x/ C sec.x/j C C
Z
• sec2 .x/
dx D tan.x/ C C
Z
• csc2 .x/
dx D cot.x/ C C
Z
• sec.x/ tan.x/
dx D sec.x/ C C
Z
• csc.x/ cot.x/
dx D csc.x/ C C

Odd powers of sine and cosine

Functions of this sort are integrated by exploiting the Pythagorean identity. We leave
one power of the trig function to be d U and transform the remaining even powers via one of
142 4. METHODS OF INTEGRATION I

cos2 .x/ D 1 sin2 .x/

or
sin2 D 1 cos2 .x/

as appropriate, transforming the problem into an integral of a polynomial function.

Z
Example 4.47 Find cos5 .x/
dx:

Solution:

Z Z
5

2
cos .x/
dx D cos2 .x/
cos.x/
dx

Z

2
D 1 sin2 .x/
cos.x/
dx

Let u D sin.x/, du D cos.x/
dx

Z
D .1 u2 /2
du

Z


D 1 2u2 C u4
du

2 3 1 5
Du u C u CC
3 5

2 1
D sin.x/ sin3 .x/ C sin5 .x/ C C
3 5

Odd powers of sine work the same way, but with the other function in a starring role.
 4.3. INTEGRATING TRIG FUNCTIONS 143
Even powers of sine and cosine

These are much messier and rely on the power reduction identities:
1 cos.2x/
sin2 .x/ D
2
1 C cos.2x/
cos2 .x/ D
2
They can always be used to transform an even power into a bunch of much lower powers, both
odd and even. This means that even powers of trig functions often end up as furballs, but they
can be done with patience and persistence.
Example 4.48 Compute Z
sin4 .x/
dx
Solution:
Z Z

2
sin4 .x/
dx D sin.x/2
dx

Z  2
1 cos.2x/
D
dx
2
Z
1

D 1 2 cos.2x/ C cos2 .2x/
dx
4
Z  
1 1 C cos.4x/
D 1 2 cos.2x/ C
dx
4 2
Z
1
D .2 4 cos.2x/ C 1 C cos.4x//
dx
8
Z
1
D .3 4 cos.2x/ C cos.4x//
dx
8
 
1 1 1
D 3x 4 sin.2x/ C sin.4x/ C C
8 2 4

3 1 1
D x sin.2x/ C sin.4x/ C C
8 4 32
˙
144 4. METHODS OF INTEGRATION I
Even powers of secant

These are pretty easy, again by hitting them with Pythagorean identities. Save one sec2 .x/ to
be du and apply

sec2 .x/ D 1 C tan2 .x/

to set up the u-substitution u D tan.x/.

Example 4.49 Compute

Z
sec8 .x/
dx

Solution:

Z Z
8

3
sec .x/
dx D sec2 .x/
sec2 .x/
dx

Z

3
D 1 C tan2 .x/
sec2 .x/
dx

Let u D tan.x/, du D sec2 .x/
dx

Z

3
D 1 C u2 du

Z


D 1 C 3u2 C 3u4 C u6 du

3 1
D u C u3 C u5 C u7 C C
5 7

3 1
D tan.x/ C tan3 .x/ C tan5 .x/ C tan7 .x/ C C
5 7

˙
 4.3. INTEGRATING TRIG FUNCTIONS 145
Odd powers of secant

These are typically a nightmare!

Example 4.50 Compute

Z
sec3 .x/
dx

Solution:
Z Z
3


sec .x/
dx D 1 C tan2 .x/ sec.x/
dx

Z Z
D sec.x/
dx C tan2 .x/ sec.x/
dx

Z
D ln j sec.x/ C tan.x/j C tan.x/
sec.x/ tan.x/
dx

U D tan.x/I dV D sec.x/ tan.x/
dx Use parts

du D sec2 .x/
dxI V D sec.x/

Z
D ln j sec.x/ C tan.x/j C sec.x/ tan.x/ sec3 .x/
dx

Z
2 sec3 .x/
dx D ln j sec.x/ C tan.x/j C sec.x/ tan.x/ Circularize!

Z
1
sec3 .x/
dx D .ln j sec.x/ C tan.x/j C sec.x/ tan.x// C C
2

As with all circular integrations by parts, we need to remember to finish with a C C . Higher
odd powers are worse.
˙

Powers of tangent

Even powers of tangent can be transformed into even powers of secant by using
tan2 .x/ D sec2 .x/ 1.
146 4. METHODS OF INTEGRATION I
Example 4.51 Compute

Z
tan4 .x/
dx

Solution:

Z Z

2
tan4 .x/
dx D sec2 .x/ 1
dx

Z


D sec4 .x/ 2 sec2 .x/ C 1
dx

Z Z Z
4 2
D sec .x/
dx 2 sec .x/
dx C dx

Z Z Z
D .tan2 C1/ sec2
dx 2 sec2 .x/
dx C dx

Let: u D tan.x/, du D sec2 .x/
dx

Z Z Z
2 2
D .u C 1/
du 2 sec .x/
dx C dx

1 3
D u Cu 2 tan.x/ C x C C
3

1
D tan3 .x/ C tan.x/ 2 tan.x/ C x C C
3

1
D tan3 .x/ tan.x/ C x C C
3

˙
 4.3. INTEGRATING TRIG FUNCTIONS 147
Odd powers of tangent can be reduced to lower odd powers of tangent. Assume 2n C 1 > 1.
Then:
Z Z
tan2nC1
dx D .sec2 .x/ 1/ tan2n 1 .x/
dx

Z Z
2n 1 2
D tan .x/ sec .x/
dx tan2n 1
.x/
dx

Let: u D tan.x/, du D sec2 .x/
dx

Z Z
D u2n 1

du tan2n 1
.x/
dx

Z
1 2n
D u tan2n 1
.x/
dx
2n
Z
1
D tan2n .x/ tan2n 1
.x/
dx
2n

Since we also know that Z

tan.x/
dx D ln j sec.x/j C C

we can integrate any odd power of tangent by applying the above formula until the power is
down to one.

Z
Example 4.52 Find tan5 .x/
dx:
Solution:
Z Z
1
tan5 .x/
dx D tan4 .x/ tan3 .x/
dx
4
 Z 
1 1
D tan4 .x/ tan2 .x/ tan.x/
dx
4 2

1 1
D tan4 .x/ tan2 .x/ C ln j sec.x/j C C
4 2

˙
148 4. METHODS OF INTEGRATION I
Mixed functions
When you wish to integrate a mixed product of trig functions, one strategy is to turn everything
into sines and cosines and look for a good trig substitution. The fact that u D cos.x/ H) du D
sin.x/
dx and u D sin.x/ H) du D cos.x/
dx permits us to transform the trig integral
into the integral of a rational function.

Example 4.53 Find Z

sin.x/ tan2 .x/
dx
Solution:
Z Z
2 sin3 .x/
sin.x/ tan .x/
dx D
dx
cos2 .x/
Z
1 cos2 .x/
D
sin.x/
dx
cos2 .x/

Let u D cos.x/, du D sin.x/
dx

Z
1 u2
D
. du/
u2
Z  
1
D 1 du
u2
 
1
D u CC
u

1 1
D CuCC D C cos.x/ C C
u cos.x/

D sec.x/ C cos.x/ C C

The mainstay of the techniques for integrating trig functions presented in this section are the
application of trigonometric identities. There are literally an infinite number of problems and
identities that can be used to solve them. The techniques of u-substitution and integration by
parts also appear, arising naturally when we try to integrate trigonometric functions.
Many of the transformations created by u-substitution make the trig function integrals into
rational and even polynomial function integrals. What, however, is the point of learning to
integrate these functions? In the next section we will learn to transform integrals involving square
 4.3. INTEGRATING TRIG FUNCTIONS 149
roots into trigonometric integrals. Integration is one of the best examples of how mathematics
builds on itself.

PROBLEMS

Problem 4.54 Perform the following integrals.

Z Z Z
3 7
1. cos .x/
dx 3. cos .x/
dx 5. sin6 .x/
dx
Z Z Z
5 4
2. sin .x/
dx 4. cos .x/
dx 6. sin2 .x/ cos2 .x/
dx

Problem 4.55 Perform the following integrals.

Z Z Z
6 3 2
1. sec .x/
dx 3. tan .x/ sec .x/
dx 5. cot6 .x/
dx
Z Z Z
6 7
2. tan .x/
dx 4. tan .x/
dx 6. sec3 .x/ tan3 .x/
dx

Problem 4.56 Compute Z

sec5 .x/
dx

Problem 4.57 Find the power reduction integral for

Z
cot2nC1 .x/
dx

that is analogous to the odd powers of tangent technique in this section.

Problem 4.58 Compute Z

csc 3 .x/
dx
150 4. METHODS OF INTEGRATION I
Problem 4.59 Perform the following integrals.

Z Z
1. cos4 .x/ sin3 .x/
dx 4. sec3 .x/ sin3 .x/
dx
Z Z
6 2 tan2 .x/
2. cos .x/ tan .x/ sin.x/
dx 5.
dx
sin.x/
Z Z
3. cot.x/ csc3 .x/
dx 6. tan5 .x/sec.x/
dx

Z
Problem 4.60 Find .cos.x/ C 1/5
dx

Z
Problem 4.61 Find sin2 .x/
cos5 .x/
dx

Z
Problem 4.62 Find sin2 .x/
cos4 .x/
dx

Z
Problem 4.63 Find sin4 .x/
cos5 .x/
dx

Z

8
Problem 4.64 Find sin2 .x/ C 1 cos3 .x/
dx

Z

8
Problem 4.65 Find cos2 .x/ C 1 sin3 .x/
dx

Z
Problem 4.66 Find sin.x/ cos.x/
dx
 151

CHAPTER 5

Methods of Integration II
The integration methods in this chapter use more sophisticated geometry and algebra than those
in the previous chapter. Both trigonometric substitution and partial fractions come up in the
natural sciences and so merit inclusion in this text. They are placed in their own chapter be-
cause, with the possible exception of the material on circular integration by parts, they are more
mathematically challenging.

5.1 TRIGONOMETRIC SUBSTITUTION

We have the following integrals that use inverse trig functions:
Z
dx
2

dx D tan 1 .x/ C C
x C1
Z
dx
p
dx D sin 1 .x/ C C
1 x2
and Z
dx
p
dx D sec 1 .x/ C C
jxj x 2 1
The goal of this section is to enlarge our library of integrals of this type. Let’s start with an
example that shows how to use trigonometric substitution.

Example 5.1 Compute Z

dx
p
x2 C 1
Solution:

Examine the following picture of a carefully labeled right triangle.

p
x2 C 1 x

1
152 5. METHODS OF INTEGRATION II
x
We see that D x D tan./, which means dx D sec2 . /
d .
1
p
2
x C1 p
Also, D x 2 C 1 D sec./. So, we can substitute and integrate.
1
Z Z
dx sec2 . /
p D
d
x2 C 1 sec./
Z
D sec./
d

D ln.j tan./ C sec./j/ C C

p
D ln.jx C x 2 C 1j/ C C

For the last substitution – back to x – we need to refer back to our substitution triangle at the
beginning of the integral.
˙

When making the substitution triangle remember that it has to obey the Pythagorean theorem.
If at all possible, the triangle should include all the parts in your integral.
p p
So what just happened? Whenever you have a radical like ax 2 C b or b ax 2 in an integral,
it is possible to draw a right triangle that structures a substitution for you. This substitution
transforms the integral into a trigonometric integral – which goes a long way toward explaining
why we worked so hard on integrating trig functions in the last chapter.

There are several types of right triangles. The next example uses a different type of right triangle
from the last one. Rather than having the square root of the sum of squares
p
x2 C 1

on the hypotenuse of the right triangle, it will have the difference of squares
p
1 x2

on one of the legs of the triangle. One of the major indicators of the type of triangle needed for
substitution is the presence of the square root of a sum, as opposed to a difference, of squares.
 5.1. TRIGONOMETRIC SUBSTITUTION 153
Example 5.2 Compute
Z p
1 x 2
dx

Solution:

Here is the natural triangle for this problem:

1
1 x2


x
p
Then the substitutions are x D cos./ and 1 x 2 D sin. /. Taking a derivative, we see that
dx D sin./
d . Substituting all this into the problem yields:
Z p Z
1 x 2
dx D sin./. sin. //
d

Z
D sin2 ./
d

Z
1
D .1 cos.2//
d
2

1 1
D sin.2/  CC
4 2

But now we need to substitute back. sin.2/ D 2 sin. / cos./ so that’s not too bad. We also
know that x D cos./ so cos 1 .x/ D  . This is enough, and we get that:

1 1 1 1
sin.2/  C C D sin. / cos. /  CC
4 2 2 2

1 p 1

D x 1 x2 cos .x/ C C
2

Done!
˙
154 5. METHODS OF INTEGRATION II
Notice that we have a choice of which formula goes on what leg of the right triangle. Sometimes
one choice leads to a slightly easier problem than the other, so it is worth thinking about which
substitution you will use.
Z
x2
Example 5.3 Compute p
dx
4 x2
Solution:

Here is the natural picture:

2
x


p
4 x2
p
x 4 x2
This makes the substitutions D sin./ and D cos./ so that
2 2
p
4 x 2 D 2 cos. /

x D 2 sin./

dx D 2 cos. /d

Plug all this in and integrate:

Z Z
x2 4 sin2 ./2 cos./
p D d
4 x2 2 cos. /
Z
D4 sin2 . /
d

Z
D2 .1 cos.2//
d

 
1
D2  sin.2 / C C
2

D 2 sin.2/ C C
 5.1. TRIGONOMETRIC SUBSTITUTION 155
x  p
1
As before,  D sin and sin.2/ D 2 sin./ cos. / D 2x 4 x 2 so
2
Z x  p
x2 1
p
dx D 2 sin 2x 4 x2 C C
4 x2 2
˙

Example 5.4 Directly verify the area formula for a circle by integration using Cartesian
coordinates.

Solution:

It would be very easy to just verify the formula using polar coordinates, but for this example we
are using Cartesian coordinates. Since a circle of radius R centered at the origin has the form
x 2 C y 2 D R2 , our strategy is to take the function that graphs as the upper half of the circle
p
y D R2 x 2
integrate it and double the result. Since it is an even function, we can actually do the problem
Z Rp
Area D 4 R2 x 2
dx
0

Here is the substitution triangle:

R


p
R2 x2

So the substitutions are:

p
R2 x 2 D R cos./

x D R sin./

dx D R cos. /
d
156 5. METHODS OF INTEGRATION II

When x D R, we have that  D . When x D 0, we have that  D 0. This gives us the following
2
definite integral.
Z =2
Area D 4 R cos./
R cos./
d
0

Z =2
D 4R2 cos2 ./
d
0

Z =2
2
D 2R .1 C cos.2//
d
0

 ˇ=2
1 ˇ
D 2R2  C sin.2/ ˇˇ
2 0

 
2  1
D 2R C sin./ 0 0
2 2

D R2
Which verifies the formula.
˙

The next example will draw on a formula we already have – and paid for with blood:
Z
1
sec3 .u/
du D .sec.u/ tan.u/ C ln.j sec.u/ C tan.u/j/ C C
2

Example 5.5 Compute Z p

x 2 C 1
dx

Solution:

For this one we get to re-use the first triangle picture:

p
x2 C 1 x

1
 5.1. TRIGONOMETRIC SUBSTITUTION 157
So the substitutions are:
p
x 2 C 1 D sec./

x D tan./

dx D sec2 ./d

Substituting we get:
Z p Z
x 2 C 1
dx D sec./
sec2 . /
d

Z
D sec3 ./
d

Use the known result

D sec. / tan. / C ln.j sec./ C tan. /j/ C C

p p
D x x 2 C 1 C ln.jx C x 2 C 1j/ C C

Without the known form, this one is a bear.

Example 5.6 Compute

Z p
9 x2

dx
x2
Solution:

Use the triangle:

p
3

9 x2

x
158 5. METHODS OF INTEGRATION II
This gives us the following substitutions.

p
9 x 2 D 3 sin./

x D 3 cos. /

dx D 3 sin. /
d

Which permits us to integrate.

Z p Z
9 x2 3 sin./
. 3 sin. //

dx D
d
x2 9 cos2 . /
Z
sin2 . /
D
d
cos2 ./
Z
D tan2 . /
d

Z


D sec2 ./ 1
d

D tan./ C  C C

We can pull tan./ directly off the triangle and, by the usual trick,  D cos 1 .x=3/ making the
result of substituting back to x :

p
9 x2 1
C cos .x=3/ C C
x

˙
 5.1. TRIGONOMETRIC SUBSTITUTION 159
The next example looks at a power of a radical.

Example 5.7 Compute

Z

3=2
1 x2
dx

Solution:

We get to re-use one of our earlier triangles:

1
1 x2

x

Making the substitutions:

p
1 x 2 D sin./

x D cos. /

dx D sin./
d

Moving on to the integral,

Z Z

3=2
1 x2
dx D sin3 ./
. sin.//
d

Z
D sin4 ./
d

Which is Example 4.48.

3 1 1
D  sin.2/ C sin.4 / C C
8 4 32

3 1 1
D  sin./ cos./ C sin.2 / cos.2/ C C
8 2 16

3 1 1

D  sin./ cos./ C sin./ cos./ cos2 ./ sin2 . / C C
8 2 8
160 5. METHODS OF INTEGRATION II
Using  D cos 1 .x/

3 1 1 p 1 p
D cos .x/ x 1 x2 C x 1 x 2 .x 2 .1 x 2 // C C
8 2 8

3 1 1 p p 
D cos .x/ 4x 1 x2 C x 1 x 2 .2x 2 1/ C C
8 8

3 1 1 p p 
D cos .x/ 4x 1 x2 C 1 x 2 .2x 3 x/ C C
8 8

3 1 1 p 
D cos .x/ 1 x 2 .4x C 2x 3 x/ C C
8 8

3 1 1p

D cos .x/ 1 x 2 2x 3 C 3x C C
8 8

3 1 xp

D cos .x/ 1 x 2 2x 2 C 3 C C
8 8
˙

The really challenging part of this example was the back-substitution. We used two trig identities
to get to where we could back-substitute:

sin.2 / D 2 sin./ cos./

cos.2/ D cos2 . / sin2 . /

Trigonometric substitution is useful for square roots of sums and differences of squares. This
is because these fit well onto the three sides of a right triangle. This permits us to translate to
the domain of trigonometry where we can use identities to manipulate the form of the integral,
often into something doable.

The possible types of triangles

Shown below are the three possible types of trianges used in trigonometric substitution. The
second and third are equivalent, in a sense, but they spawn complementary trig functions when
used in substution.
 5.1. TRIGONOMETRIC SUBSTITUTION 161

p
x 2 C a2
x


a

a
a p x
a2 x2


 p
x a2 x2

PROBLEMS

Problem 5.8 For each of the following, show a substitution triangle that might work well. Do
not do the integrals.

Z p Z
x2
1. x 5 1 C 4x 2
dx 4. p
dx
2 3x 2
Z p Z
4 9x 2 x
2.
dx 5. p
dx
x3 x2 C1
Z p Z p
2 x2 C 1
3. .x C 2x C 1/ x 2 C 2x
dx 6.
dx
x3
162 5. METHODS OF INTEGRATION II
Problem 5.9 Perform the following integrals. Not all require trig substitution, but for those
that do, show the substitution triangle.

Z p Z
x3
1. x x 2 C 9
dx 4. p
dx
9x 2 C 16
Z p Z p
2
2. x x2 C 1
dx 5. 1 C 4x 2
dx
Z p Z
3
x2
3. x x2 C 4
dx 6.
dx
.x 2 C 1/3=2

Z
dx
Problem 5.10 Compute p where a; b are non-zero constants.
ax 2 C b 2

Z p
Problem 5.11 Compute ax 2 C b 2
dx where a; b are non-zero constants.

Problem 5.12 Perform the following integrals.

Z p Z p
1 x2 4. x3 1 x 2
dx
1.
dx
x
Z p Z
4 x2 x2
2.
dx 5. p
dx
x2 16 x2
Z Z
dx x3
3. p
dx 6. p
dx
9 4x 2 9 x2

Z
dx
Problem 5.13 Compute p
dx where a; b are non-zero constants.
a2 b2x2

Z p
Problem 5.14 Compute a2 b 2 x 2
dx where a; b are non-zero constants.
 5.2. PARTIAL FRACTIONS 163
Problem 5.15 Perform the following integrals.

Z q Z p
1. sin2 .x/ C 1
cos.x/

dx 4. 1 e2x
dx
Z p Z p
3x tan2 .x/
1
2. e e2x C 4
dx 5. sec2 .x/
dx
tan.x/
Z p Z p p
3. .x C 1/ x 2 C 2x C 2
dx 6. x
1 x 2 1 C x 2
dx

Problem 5.16 Compute Z

1
P
x sin .x/
dx

5.2 PARTIAL FRACTIONS

This section introduces a novel algebraic technique for setting up integrals. The first thing we
need is a useful fact about polynomials over the real numbers.

Knowledge Box 5.1

Factorization of polynomials
The only polynomials that do not factor over the real numbers are
quadratics with no roots. Any polynomial of degree 3 or more factors
into linear polynomials and quadratics that do not factor.

What can we do with this fact? Let’s look at an example to set the stage.

Example 5.17 Compute Z

dx
x2 3x C 2
Solution:

Notice that
1 1 .x 1/ .x 2/ 1
D D 2
x 2 x 1 .x 1/.x 2/ x 3x C 2
164 5. METHODS OF INTEGRATION II
This means that:
Z Z Z
dx dx dx
D
x2 3x C 2 x 2 x 1

D ln jx 2j ln jx 1j C C

The problem with this solution is that “notice” is not a recognized algebra technique. In order to
gain the ability to “notice” in this fashion we need one additional fact. Equal polynomials have
equal coefficients.
Suppose that
A B 1
C D
x 1 x 2 .x 1/.x 2/
Then, if we cross-multiply on the left side of the equation, we get

A.x 2/ C B.x 1/ 1
D
.x 1/.x 2/ .x 1/.x 2/

which simplifies to
.A C B/x 2A B D 1 D 0x C 1
Since these polynomials are equal we get the system of equations:

ACB D0
2A B D 1
AD 1 Second equation plus first.
BD1 Substitute AD 1 into the first equation.

So A D 1 and B D 1 tell us the thing we noticed:

1 1 1
D
x 2 x 1 x2 3x C 2
˙

This is the basis of integration by partial fractions. The fact that a polynomial factors into
linear and unfactorable quadratic terms means that we can always break a rational function into
(i) polynomial terms and (ii) rational functions of the form
A Bx C C
or
x u x2 C sx C t
Integration by partial fractions is used to integrate rational functions.
 5.2. PARTIAL FRACTIONS 165
Knowledge Box 5.2
Steps of integration by partial fractions
1. If the numerator of the rational function is not lower degree than the denom-
inator, divide to transform the problem into a polynomial plus a rational
function with a higher degree denominator.
2. Factor the denominator.
3. Set the rational function equal to the sum of partial fraction terms.
4. Clear the denominator by cross multiplication.
5. Solve for the coefficients of the partial fraction terms based on the equal poly-
nomials resulting from cross multiplication.
6. Perform the resulting integrals.

If a factor of the denominator is repeated, it gets one partial fraction term for each power of the
denominator. Let’s work some examples.

Example 5.18 Find

Z
dx
x2 4x C 3

Solution:

The numerator is already lower degree. Factor the denominator. x 2 4x C 3 D .x 3/.x 1/

which gives us
1 A B
D C
x2 4x C 3 x 1 x 3

meaning that
1 A.x 3/ C B.x 1/
D
x2 4x C 3 x 2 4x C 3

yielding the polynomial equality 1 D .A C B/x .3A C B/. We get two equations – one for
the coefficient of x and one for the constant term:
166 5. METHODS OF INTEGRATION II

3A C B D 1 for constant term

ACB D0 for coefficient of x
AD B
2B D 1
B D 1=2
A D 1=2

So  
1 1 1 1
D
x2 4x C 3 2 x 3 x 1
Setting up the integral:
Z Z  
dx 1 1 1
D dx
x2 4x C 3 2 x 3 x 1
1
D .ln jx 3j ln jx 1j/ C C
2
˙

Example 5.19 Compute Z

2x C 5
x2 x
Solution:

The numerator is already lower degree. Factor the denominator and set up the partial
fractions.
2x C 5 A B
D C
x.x 1/ x x 1
Cross multiply and extract the polynomials equations.
2x C 5 D A.x 1/ C Bx

which gives us the linear system

ACB D2
5D A
AD 5
BD7
 5.2. PARTIAL FRACTIONS 167
So
Z Z Z
2x C 5 7 5
D dx
dx
x2 x x 1 x

D 7 ln jx 1j 5 ln jxj C C

The next example will be our first with a repeated root, letting us showcase the fact that partial
fractions have one part for each point of multiplicity of a root.
Example 5.20 Compute Z
dx
x3 x2
Solution:

We begin, as always, by factoring: x 3 x 2 D x 2 .x 1/. The root x D 0 has multiplicity

two. This means that both x and x 2 are factors. So, the partial fraction decomposition is:
1 A B C
D C 2C
x3 x2 x x x 1
So, once we clear the denominator (D x 2 .x 1/) we get

1 D Ax.x 1/ C B.x 1/ C C x 2

or
.A C C /x 2 C .B A/x B D 0x 2 C 0x C 1
The resulting system of equations is:

ACC D0
B AD0
BD1

So B D 1, A D 1, and C D 1. Now we can integrate:

Z Z  
dx 1 1 1
D C
dx
x3 x2 x x2 x 1
1
D ln jxj C C ln jx 1j C C
x
˙
168 5. METHODS OF INTEGRATION II
Next we look at a shortcut that permits us to, in some cases, avoid having to solve the system of
equations for the coefficients of the partial fraction decomposition. Instead, we plug the roots
into the equations. May terms zero out and, often, we get the values of the partial fraction
coefficients directly.

Example 5.21 Compute

Z
dx
x3 6x C 11x 6
Solution:

Start by factoring: x 3 6x C 11x 6 D .x 1/.x 2/.x 3/ which gives us a partial

fractions form:
1 A B C
D C C
x3 6x C 11x 6 x 1 x 2 x 3
Clearing the denominator we get

1 D A.x 2/.x 3/ C B.x 1/.x 3/ C C.x 1/.x 2/

which sets us up for the shortcut:

1 D A. 1/. 2/ D 2A When x D 1

1 D B.1/. 1/ D B When x D 2

1 D C.2/.1/ D 2C When x D 3

So A D 1=2, B D 1, and C D 1=2. Integrate:

Z Z  
dx 1=2 1 1=2
3
D C
dx
x 6x C 11x 6 x 1 x 2 x 3

1 1
D ln jx 1j ln jx 2j C ln jx 3j C C
2 2
˙

What happened? Let’s summarize our findings in a Knowledge Box.

 5.2. PARTIAL FRACTIONS 169
Knowledge Box 5.3

Finding the coefficients of a partial fractions decomposition

• Equal polynomials have equal coefficients for equal power terms.
• Equal polynomials are equal for any specific value of their variable.

The shortcut for finding the partial fraction constants is to plug in values of x that zero out
all but one term. Warning: this isn’t always possible. Terms with multiplicity above one or non-
factorable quadratic terms can mess up the shortcut. The next set of examples involve integrating
reciprocals of quadratics that don’t factor. These will arise as part of partial fraction decomposi-
tions.

Example 5.22 Compute

Z
dx
x 2 C 4x C 5
Solution:

The denominator doesn’t factor. Let’s try completing the square.

Z Z
dx dx
2
D
x C 4x C 5 x2 C 4x C 4 C 1
Z
dx
D
.x C 2/2 C 1

Let u D x C 2, so du D dx
Z
du
D 2
u C1

1
D tan .u/ C C

1
D tan .x C 2/ C C

˙
170 5. METHODS OF INTEGRATION II
Knowledge Box 5.4

Integrating the reciprocal of unfactorable quadratics

1. Complete the square to place the quadratic in the form a.x b/2 C c
1 1 1
2. D  a  D r 2 !
a.x b/2 C c c 2
.x b/ C 1 a
c c .x b/ C 1
c
r r r
a a c
3. Let u D .x b/ so that du D dx and dx D du
c c a
Z r
1 1 c
4. 2

dx D tan 1 .u/ C C
a.x b/ C c c a r 
1 1 a
D p tan .x b/ C C
ac c
With this particular type of u-substitution in place we can do the next batch of integrations by
partial fractions.

Example 5.23 Compute Z

dx
x3Cx
Solution:

Factor and we see that x 3 C x D x.x 2 C 1/: So the partial fraction form is:
1 A Bx C C
D C 2
x3 Cx x x C1

We need Bx C C in the numerator of the second part instead of just B because x 2 C 1 is

quadratic. Clearing the denominator we get the polynomial equation:

1 D A.x 2 C 1/ C x.Bx C C /

Plug in x D 0 and we see that 1 D A. This tells us that

1 D .x 2 C 1/ C Bx 2 C C x

or
x 2 C 0x D Bx 2 C C x
 5.2. PARTIAL FRACTIONS 171
so B D 1 and and C D 0. We are now prepared to integrate
Z Z  
dx 1 x
3
D 2

dx
x Cx x x C1

1
D ln jxj ln jx 2 C 1j C C
2

The second part of the integral is a u-substitution with u D x 2 C 1; we have done it before.

Let’s try a less neat integral of this sort.

Example 5.24 Compute

Z
3x
dx
x3 9x 2 C 25x 25

Solution:

Factor, and we get x 3 9x 2 C 25x 25 D .x 5/.x 2 4x C 5/ D .x 5/..x 2/2 C 1/.

That makes the partial fraction form

3x
dx A Bx C C
D C
x3 9x 2 C 25x 25 x 5 x2 4x C 5

Clear the denominator and we get

0x 2 C 3x C 0 D A.x 2 4x C 5/ C .Bx C C /.x 5/ D Ax 2 4Ax C 5A C Bx 2 5Bx C C x 5C

yielding the simultaneous equations

ACB D0
4A 5B C C D 3
5A 5C D 0

So, A D B , A D C , and 4A C 5A C A D 2A D 3. So, A D C D 3=2 and B D 3=2.

172 5. METHODS OF INTEGRATION II
This permits us to integrate:
Z Z  
3x
dx 3 1 x 1
3 2
D 2

dx
x 9x C 25x 25 2 x 5 .x 2/ C 1
Z  
3 1 x 2 1
D 2
C
dx
2 x 5 .x 2/ C 1 .x 2/2 C 1

Let u D x 2 with du D dx for the second and third terms

 
3 1
D ln jx 5j ln.u2 C 1/ C tan 1
.u/ C C
2 2
 
3 1
D ln jx 5j ln.x 2 4x C 5/ C tan 1
.x 2/ C C
2 2

Example 5.25 Compute

Z
dx
x3 1
Solution:

Factor: x 3 1 D .x 1/.x 2 C x C 1/. Then

1 A Bx C C
D C
x3 1 x 1 x2 C x C 1

Clear the denominator and we get

1 D A.x 2 C x C 1/ C .Bx C C /.x 1/

yielding: ACB D0

A B CC D0

A C D1

Solving the linear system we get A D 1=3, B D 1=3, C D 2=3.

 5.2. PARTIAL FRACTIONS 173
Z Z  
dx 1 1 xC2
D
dx
x3 1 3 x 1 x2 CxC1
Z  
1 1 1 2x C 4
D
2
dx
3 x 1 2 x CxC1
Z  
1 1 1 2x C 1 C 3
D

dx
3 x 1 2 x2 C x C 1
Z  
1 1 1 2x C 1 1 3
D
2
2
dx
3 x 1 2 x CxC1 2 x CxC1

The middle term is a u-substitution that leads to a log.

Z
1 1 1 dx
D ln jx 1j ln jx 2 C x C 1j
3 6 2 x2 CxC1
Z
1 1 1 dx
D ln jx 1j ln jx 2 C x C 1j
3 6 2 .x C 1=2/2 C 3=4
Z
1 1 1 dx
D ln jx 1j ln jx 2 C x C 1j  
3 6 2 3 4
.x C 1=2/2 C 1
4 3
Z
1 1 1 4 dx
D ln jx 1j ln jx 2 C x C 1j
 
3 6 2 3 4 2
.x C 1=2/ C 1
3
Z
1 1 2 dx
D ln jx 1j ln jx 2 C x C 1j  2
3 6 3 2
p .x C 1=2/ C1
3
2 2
Let u D p .x C 12 / and so du D p dx
3 3
p Z
1 1 2 3 du
D ln jx 1j ln jx 2 C x C 1j
2
3 6 3 2 u C1

1 1 1
D ln jx 1j ln jx 2 C x C 1j p tan 1
.u/ C C
3 6 3
174 5. METHODS OF INTEGRATION II
  
1 1 1 2 1
D ln jx 1j ln jx 2 C x C 1j p tan 1
p xC CC
3 6 3 3 2

See! Easy!
˙

One of the fairly obvious conclusions that follows from this section is that you need to not only
know algebra, you need to be a master of it to mess with partial fractions in all but its simplest
form. Some of the homework problems develop general formulas for this sort of super messy
inverse tangent integral. Like Knowledge Box 5.4, but with more parameters.

PROBLEMS

Problem 5.26 Perform the following integrals.

Z Z Z
dx dx dx
1. 3. 5.
x2 x 2
x C 4x C 3 x2 x 1
Z Z Z
dx dx dx
2. 4. 6.
x2 6x C 8 6x 2 5x C 1 x2 11x C 30

Problem 5.27 Perform each of the following integrals. Partial fractions are not needed.

Z Z Z
1 1 2x C 2
1. 2

dx 3. 2

dx 5.
dx
x C 2x C 2 x CxC2 x2 C 3x C 4
Z Z Z
1 xC1 x 1
2.
dx 4.
dx 6.
dx
x 2 C 3x C 4 x 2 C 2x C 2 x2 C x C 2

Problem 5.28 Perform each of the following integrals.

Z Z
dx x2 C x C 1
1. 3 2 4.
dx
x C 6x C 11x C 6 x3 6x 2 C 11x C 6
Z Z
dx x2 C 1
2. 5.
dx
x4 16 x 3 C 4x
Z Z
dx 3x 2
3. 6.
dx
x3 5x 2 2x C 24 x3 C 2x 2 5x 6
 5.2. PARTIAL FRACTIONS 175
Problem 5.29 For each of the following, compute the partial fractions form for decomposing
the integral, but do not solve for the partial fractions coefficients or perform the integral.

Z Z
dx dx
1. 4.
x 5 C 4x 3 C x x 6 5x 5 C6x 4 x 3 C5x 2 5x C1
Z Z
dx dx
2. 5.
x 4 C 4x 3 C 7x 2 C 6x C 3 x6 3x 5 C 3x 4 C x 3
Z Z
dx dx
3. 6.
x C x3
5 x 6 C2x 5 C5x 3 C6x 2 C3x C2

Problem 5.30 Compute

Z
dx
ax 2 C bx C c
There are three cases each of which requires its own method of integration.

Z
dx
Problem 5.31 Compute where a is a positive constant.
x3 a3

Z
dx
Problem 5.32 Compute where a is a positive constant.
x4 a4

Problem 5.33 Perform the integrals in Problem 5.29.

Problem 5.34 Compute

Z
x5

dx
x2 C x C 1

Problem 5.35 Compute

Z
dx

dx
.x a/.x b/.x c/
176 5. METHODS OF INTEGRATION II
Problem 5.36 Compute
Z
x
dx

dx
.x a/.x b/.x c/

Problem 5.37 Compute

Z
x 2
dx

dx
.x a/.x b/.x c/

5.3 PRACTICING INTEGRATION

One of the important skills for performing integrals is figuring out which method of integration
to use. Knowledge Box 5.5 may be some help (it continues on the next page). This section is no
more than a collection of problems – but, unlike all the other sections, you cannot guess which
method is useful by looking at the section the problem appears in. Practice! Practice! Practice!

Knowledge Box 5.5

Which method of integration do I use?

1. Polynomials are integrated with the power rule, one term at a time.
2. If the thing you are integrating is the derivative of something you
recognize, use the fundamental theorem – the integral is the thing
you recognize, plus “C”.
dx
3. Look at the known derivatives, e.g., Dx arctan.x/ D x 2 C1
. If
you find one, apply rule “b”.
4. Will an algebraic re-arrangement of the terms of the integral set up
an integral you can do?
5. Look through the examples in this chapter and in Chapter 1. Are
any of them similar enough to your problem to help?

(continued)
 5.3. PRACTICING INTEGRATION 177

6. Check if there is a u-substitution that makes a simpler integral.

Remember that du needs to be in there in an appropriate form.
7. Are there natural parts for integration by parts? Remember you
may need to integrate by parts several times.
8. Is your integral packed with trig functions? Flip through Section
4.3 which has special methods for many different combinations of
trig functions. Also remember that rule “c” includes applying trig
identities to set up one of these special trig methods.
p p
9. Does the integral contain x 2 ˙ a2 or a2 ˙ x 2 ? Consider
trigonometric substitution. This may yield integrals covered by any
of rules “b”, “c”, or “d”.
10. Is the integral a ratio of polynomials? If the numerator is not lower
degree, divide to get a remainder that does have a lower degree top.
Integrate the resulting ratio of polynomials with partial fractions.
11. If one of these rules seems to make some progress, keep going, you
may need many steps to finish an integral.
12. Try rule “d” again. Really.
13. There are a lot of integrals that you can’t do, there are many that
no one can do. If you are really stuck, check with someone who has
more experience than you. They may be able to recognize impossible
integrals or integrals well above your level.

Remember that a problem may use several methods of integration to complete one intergral: a
u-substitution sets up a partial fractions decomposition that leads to a trig-substitution integral.

PROBLEMS

Problem 5.38 Pick and state a method of integration and then perform the integration for
each of the following problems.
178 5. METHODS OF INTEGRATION II
Z Z
3 x x
1. x e
dx 4.
dx
x3 1
Z Z
2 3
2. x
.x C 8/
dx 5. sin.x/ cos.x/
dx
Z Z
x
3. 2

dx 6. sin4 .3x/
dx
x C1

Problem 5.39 Pick and state a method of integration and then perform the integration for
each of the following problems.

Z Z
1. sin.2x/e3x
dx ln3 .x/ C 1
4.
dx
x
Z p Z
2. 25x 2 C 1
dx 5. csc4 .x/
dx
Z Z
x2 1
3. cos.x/ sin.2x/
dx 6.
dx
x2 C 1

Problem 5.40 Pick and state a method of integration and then perform the integration for
each of the following problems.

Z Z
dx ex
1. p 4.
dx
1 4x 2 1 C e2x
Z Z
x ex
2. p
dx 5.
dx
1 9x 2 e2x 3ex C 2
Z p
Z
x
3. x
e
dx 6. x
ln.x/
dx
 5.3. PRACTICING INTEGRATION 179
Problem 5.41 Pick and state a method of integration and then perform the integration for
each of the following problems.

Z Z
x2 sin.2x/
1.
dx 4.
dx
x4 1 cos2 .x/ 9
Z p Z
3 x
2. x
e
dx 5. x 5 .x 2 C 2/4
dx
Z Z
cos.x/

3.
dx 6. cos2 .x/ sin2 .x/ esin.2x/
dx
sin2 .x/ 1

Problem 5.42 Compute Z

.sin.x/ C cos.x//2
dx

Problem 5.43 Compute Z

x n
ln.x/
dx

Problem 5.44 Compute Z p

n x
xe
dx

Problem 5.45 Compute Z

.sin.ax/ C cos.bx// ex
dx

Problem 5.46 Compute Z

.1 C ln.x//x x
dx

Problem 5.47 Find a problem in this section that can be done in two ways and demonstrate
them.
 181

APPENDIX A

Useful Formulas
A.1 POWERS, LOGS, AND EXPONENTIALS
RULES FOR POWERS

n 1 an
• a D • D an m
an am
• an  am D anCm • .an /m D anm

LOG AND EXPONENTIAL ALGEBRA

• b logb .c/ D c • logb .x y / D y
logb .x/

• logb .b a / D a
logb .x/
• logb .xy/ D logb .x/ C logb .y/ • logc .x/ D
  logb .c/
x
• logb D logb .x/ logb .y/
y • If logb .c/ D a, then c D b a

A.2 TRIGONOMETRIC IDENTITIES

TRIG FUNCTION DEFINITIONS FROM SINE AND COSINE

sin./ 1 1
• tan./ D • tan. / D • csc. / D
cos./ cot. / sin. /

cos./ 1
• cot. / D • sec./ D
sin. / cos./

PERIODICITY IDENTITIES



• sin.x C 2/ D sin.x/ • sec.x/ D csc x C 2
• sin.x C / D sin.x/
• cos.x C 2/ D cos.x/ • cos. x/ D cos.x/ • cos.x C / D cos.x/


• sin.x/ D cos x 2 • sin. x/ D sin.x/


• tan.x C / D tan.x/
• tan.x/ D cot x 2 • tan.x/ D tan.x/
182 A. USEFUL FORMULAS
THE PYTHAGOREAN IDENTITIES

• sin2 ./ C cos2 ./ D 1 • tan2 . / C 1 D sec2 ./ • 1 C cot2 ./ D csc2 . /

THE LAW OF SINES, THE LAW OF COSINES

B A

˛ ˇ
C
The Law of Sines The Law of Cosines

A B C
D D
sin.˛/ sin.ˇ/ sin. / C 2 D A2 C B 2 C 2AB
cos. /

The laws refer to the diagram.

SUM, DIFFERENCE, AND DOUBLE ANGLE

• sin.˛ C ˇ/ D sin.˛/ cos.ˇ/ C sin.ˇ/ cos.˛/ • cos.2/ D cos2 . / sin2 ./

• cos.˛ C ˇ/ D cos.˛/ cos.ˇ/ sin.˛/ sin.ˇ/
1 C cos. /
• sin.˛ ˇ/ D sin.˛/ cos.ˇ/ sin.ˇ/ cos.˛/ • cos2 .=2/ D
2
• cos.˛ ˇ/ D sin.˛/ sin.ˇ/ C cos.˛/ cos.ˇ/
1 cos. /
• sin.2/ D 2 sin./ cos./ • sin2 .=2/ D
2

A.3 SPEED OF FUNCTION GROWTH

• Logarithms grow faster than constants.
• Positive powers of x grow faster than logarithms.
• Larger positive powers of x grow faster than smaller positive powers of x .
• Exponentials (with positive exponents) grow faster than positive powers of x .
• Exponentials with larger exponents grow faster than those with smaller exponents.
 A.4. DERIVATIVE RULES 183
A.4 DERIVATIVE RULES

• If f .x/ D x n then • .sec.x//0 D sec.x/ tan.x/

f 0 .x/ D nx n 1
• .csc.x//0 D csc.x/ cot.x/

0 1
• .f .x/ C g.x//0 D f 0 .x/ C g 0 .x/ • sin 1 .x/ D p
1 x2
• .a
f .x//0 D a
f 0 .x/

0 1
1 • cos 1 .x/ D p
• If f .x/ D ln.x/, then f 0 .x/ D 1 x2
x
1
0 1
• If f .x/ D logb .x/, then f 0 .x/ D • tan 1 .x/ D
x ln.b/ 1 C x2
• If f .x/ D ex , then f 0 .x/ D ex

0 1
• cot 1 .x/ D
• If f .x/ D ax , then f 0 .x/ D ln.a/
ax 1 C x2
• .sin.x//0 D cos.x/
0 1
0
• sec 1 .x/ D p
• .cos.x// D sin.x/ jxj x 2 1

• .tan.x//0 D sec2 .x/
0 1

• csc 1 .x/ D p
• .cot.x//0 D csc2 .x/ jxj x 2 1

The product rule

.f .x/
g.x//0 D f .x/g 0 .x/ C f 0 .x/g.x/

The quotient rule

 0
f .x/ g.x/f 0 .x/ f .x/g 0 .x/
D
g.x/ g 2 .x/

The reciprocal rule

 0
1 f 0 .x/
D
f .x/ f 2 .x/

The chain rule

.f .g.x///0 D f 0 .g.x//
g 0 .x/
184 A. USEFUL FORMULAS
ALGEBRA OF SUMMATION

b
X b
X b
X b
X b
X
• f .i / C g.i / D f .i / C g.i / • c
f .i/ D c
f .i/
i Da i Da i Da i Da i Da

CLOSED SUMMATION FORMULAS

n
X n
X n.n C 1/.2n C 1/
• 1Dn • i2 D
6
i D1 i D1

n
X n
X
n.n C 1/ n2 .n C 1/2
• iD • i3 D
2 4
i D1 i D1

A.5 VECTOR ARITHMETIC

VECTOR ARITHMETIC AND ALGEBRA

• c
vE D .cv1 ; cv2 ; : : : ; cvn / • c
.E

v C w/
E D c
vE C c
w
E
• vE C wE D .v1 C w1 ; v2 C w2 ; : : : ; vn C
wn / • c
.d
vE/ D .cd /
vE

• vE wE D .v1 w1 ; v 2 w2 ; : : : ; vn • vE C wE D wE C vE
wn /
• vE
wE D v1 w1 C v2 w2 C : : : C vn wn • uE
.E
v C w/
E D uE
vE C uE
w
E

CROSS PRODUCT OF VECTORS

• vE  wE D . v2 w3 v3 v2 ; v3 w1 v1 w3 ; v1 w2 v2 w1 /

Formula for the angle between vectors

vE
w
E
cos./ D
jvjjwj

A.6 POLAR AND RECTANGULAR CONVERSION

p
• x D r
cos./ • rD x2 C y2

• y D r
sin./ •  D tan 1 .y=x/

 A.7. INTEGRAL RULES 185
A.7 INTEGRAL RULES
BASIC INTEGRATION RULES
Z Z Z
1
• x n
dx D x nC1 C C • a
f .x/
dx D a
f .x/
dx
nC1
Z Z Z
• .f .x/ C g.x//
dx D f .x/
dx C g.x/
dx

LOG AND EXPONENT

Z Z
1
•
dx D ln.x/ C C • ex
dx D ex C C
x

TRIG AND INVERSE TRIG

Z
• sin.x/
dx D cos.x/ C C Z
1
Z • p
dx D sin 1 .x/ C C
1 x2
• cos.x/
dx D sin.x/ C C Z
1
Z •
dx D tan 1 .x/ C C
1 C x2
• sec2 .x/
dx D tan.x/ C C Z
1
Z • p
dx D sec 1 .jxj/ C C
x x 2 1
• csc2 .x/
dx D cot.x/ C C Z
Z • tan.x/
dx D ln j sec.x/j C C
• sec.x/ tan.x/
dx D sec.x/ C C Z
Z • sec.x/
dx D ln j sec.x/ C tan.x/j C C
• csc.x/ cot.x/
dx D csc.x/ C C

INTEGRATION BY PARTSZ Z
U
dV D U V V
dU

EXPONENTIAL/POLYNOMIAL
Z SHORTCUT


p.x/ex
dx D p.x/ p 0 .x/ C p 00 .x/ p 000 .x/ C


ex C C
 187

Author’s Biography
DANIEL ASHLOCK
Daniel Ashlock is a Professor of Mathematics at the University of Guelph. He has a Ph.D. in
Mathematics from the California Institute of Technology, 1990, and holds degrees in Computer
Science and Mathematics from the University of Kansas, 1984. Dr. Ashlock has taught math-
ematics at levels from 7th grade through graduate school for four decades, starting at the age
of 17. Over this time Dr. Ashlock has developed a number of ideas about how to help students
overcome both fear and deficient preparation. This text, covering the mathematics portion of an
integrated mathematics and physics course, has proven to be one of the more effective methods
of helping students learn mathematics with physics serving as an ongoing anchor and example.
 189

Index

Algebra of summation, 43 of a root with multiplicity, 104

Algebra of vectors, 77 Dot product
Angle between vectors, 78 of vectors, 76
Anti-derivatives, 1
Area between curves, 21 Even function, 24
Area formula for polar curves, 70 integral of, 24
Arithmetic of vectors, 76 Factoring
polynomials, 163
C (unknown constant)
Fast polynomial multiplication algorithm, 92
in initial value problems, 35
Finding partial fractions coefficients, 168
of integration, 2
Function
Calculus with vector curves, 85
even, 24
Circles
integral of, 24
in polar coordinates, 63
odd, 24
Circular integration by parts, 137
integral of, 26
Constant
parametric, 49
of integration, 2
rational, 109
unknown, 2
Fundamental theorem of calculus I, 13
Constant of integration, 2
Fundamental theorem of calculus II, 13
Conversion
polar to rectangular, 62 Graph paper
Coordinates polar, 62
polar, 60
Cross product, 79 Identities
Curves polynomial, 112
parametric, 50 Indefinite integral, 9
Induction
Definite integral, 12 mathematical, 40
Derivative Initial value problems, 31
of a parametric curve, 55 Integral, 1
of a polar curve, 67 definite, 12
190 INDEX
indefinite, 9 Linear correction integral, 121
linear correction, 121 Linearity of integrals, 3
of cos.x/, 5
of csc.x/ cot.x/, 5 Magnitude of a vector, 77
of csc2 .x/, 5 Major axis
1 of an ellipse, 50
of ,7 Mathematical induction, 40
1 C x2
1 Methods of integration, 119
of p ,7
1 x2 Minor axis
1 of an ellipse, 50
of p ,7
x x2 1 Multiplicity
1
of , 4 of a root, 103
x
of ex , 4
Newton’s Method, 94
of sec.x/, 124
Number of roots of a polynomial, 99
of sec.x/ tan.x/, 5
of sec2 .x/, 5 Odd function, 24
of sin.x/, 5 integral of, 26
of tan.x/, 124 Orientation
of a sum, 3 of a parametric curve, 51
of an even function, 24
of an odd function, 26 Parameter, 49
of arctangent, 134 Parametric curves, 50
of constant multiples, 3 derivative of, 55
of polynomial times ex , 136 Parametric form
of trigonometric functions, 141 of polar curves, 64
Integration Parametric function, 49
by u-substitution, 119 Partial fractions
by partial fractions, 163 finding coefficients, 168
by parts, 131 integration, 163
circular, 137 steps of integration, 165
by trigonometric substitution, 151 Parts
methods of, 119 integration by, 131
steps of partial fractions, 165 Polar
which method?, 176 area formula, 70
Integration Methods, 176 integral, 70
Polar circles, 63
L’Hôpital’s Rule, 107 Polar coordinates, 60
for limits at infinity, 108 Polar curve
Length of a vector, 77 derivative, 67
 INDEX 191
Polar graph paper, 62 of calculus, fundamental I, 13
Polar to rectangular conversion, 62 of calculus, fundamental II, 13
Polynomial, 91 Rolle’s, 99
arithmetic, 91 root-factor, 94
factorization, 163 Trigonometric functions
fast division, 93 integrals, 141
fast multiplication, 92 Trigonometric substitution
identities, 112 integration by, 151
number of roots of, 99
Putting polar curves in parametric form, 64 u-Substitution, 119
Unit vector, 80
Rational function, 109 in the direction of, 80
Rolle’s Theorem, 99 Unknown constants
Root in initial value problems, 35
finding, 94 in integration, 2
multiplicity, 103
multiplicity and derivatives, 104 Vector
Root-factor theorem, 94 addition, 76
algebra, 77
Scalar multiplication, 76
angle between, 78
Substitution
arithmetic, 76
in definite integrals, 127
curves
Summation
calculus of, 85
algebra, 43
Sums of powers of n, 43 length of, 77
magnitude of, 77
Theorem subtraction, 76
fundamental, 12 unit, 80