CK-12 FOUNDATION

CK-12 Calculus

Almukkahal

Cifarelli

Fan

Jarvis

CK-12 Foundation is a non-prot organization with a mission to reduce the cost of textbook materials
for the K-12 market both in the U.S. and worldwide. Using an open-content, web-based collaborative
model termed the FlexBook, CK-12 intends to pioneer the generation and distribution of high-quality
educational content that will serve both as core text as well as provide an adaptive environment for learning,
powered through the FlexBook Platform.
Copyright 2011 CK-12 Foundation, www.ck12.org
Except as otherwise noted, all CK-12 Content (including CK-12 Curriculum Material) is made available
to Users in accordance with the Creative Commons Attribution/Non-Commercial/Share Alike 3.0 Unported (CC-by-NC-SA) License (http://creativecommons.org/licenses/by-nc-sa/3.0/), as amended
and updated by Creative Commons from time to time (the CC License), which is incorporated herein
by this reference. Specic details can be found at http://www.ck12.org/terms.
Printed: March 23, 2011

Authors
Raja Almukkahal, Victor Cifarelli, Chun-Tak Fan, Louise Jarvis

www.ck12.org

Contents
1 Functions, Limits, and
Continuity

1.1

Equations and Graphs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1.2

Relations and Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

11

1.3

Models and Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

27

1.4

The Calculus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

37

1.5

Finding Limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

45

1.6

Evaluating Limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

51

1.7

Continuity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

57

1.8

Innite Limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

65

2 Derivatives

69

2.1

Tangent Lines and Rates of Change

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

69

2.2

The Derivative . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

76

2.3

Techniques of Dierentiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

82

2.4

Derivatives of Trigonometric Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

88

2.5

The Chain Rule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

92

2.6

Implicit Dierentiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

96

2.7

Linearization and Newtons Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101

3 Applications of Derivatives

109

3.1

Related Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

3.2

Extrema and the Mean Value Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116

3.3

The First Derivative Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124

3.4

The Second Derivative Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131

3.5

Limits at Innity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138

3.6

Analyzing the Graph of a Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142

3.7

Optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156

3.8

Approximation Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161

www.ck12.org

ii

4 Integration

167

4.1

Indenite Integrals Calculus

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167

4.2

The Initial Value Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171

4.3

The Area Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174

4.4

Denite Integrals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179

4.5

Evaluating Denite Integrals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183

4.6

The Fundamental Theorem of Calculus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187

4.7

Integration by Substitution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193

4.8

Numerical Integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197

5 Applications of Denite
Integrals

204

5.1

Area Between Two Curves

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204

5.2

Volumes

5.3

The Length of a Plane Curve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226

5.4

Area of a Surface of Revolution

5.5

Applications from Physics, Engineering, and Statistics . . . . . . . . . . . . . . . . . . . . . 234

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229

6 Transcendental Functions

250

6.1

Inverse Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250

6.2

Exponential and Logarithmic Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256

6.3

Dierentiation and Integration of Logarithmic and Exponential Functions . . . . . . . . . . 259

6.4

Exponential Growth and Decay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267

6.5

Derivatives and Integrals Involving Inverse Trigonometric Functions . . . . . . . . . . . . . 275

6.6

LHospitals Rule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281

7 Integration Techniques

285

7.1

Integration by Substitution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285

7.2

Integration By Parts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292

7.3

Integration by Partial Fractions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300

7.4

Trigonometric Integrals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 306

7.5

Trigonometric Substitutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314

7.6

Improper Integrals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319

7.7

Ordinary Dierential Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327

8 Innite Series

333

8.1

Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333

8.2

Innite Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 343

iii

www.ck12.org

8.3

Series Without Negative Terms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352

8.4

Series With Odd or Even Negative Terms . . . . . . . . . . . . . . . . . . . . . . . . . . . . 358

8.5

Ratio Test, Root Test, and Summary of Tests . . . . . . . . . . . . . . . . . . . . . . . . . . 363

8.6

Power Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 368

8.7

Taylor and Maclaurin Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372

8.8

Calculations with Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 380

www.ck12.org

iv

Chapter 1
Functions, Limits, and
Continuity
1.1 Equations and Graphs
Learning Objectives
A student will be able to:

Find solutions of graphs of equations.

Find key properties of graphs of equations including intercepts and symmetry.
Find points of intersections of two equations.
Interpret graphs as models.

Introduction
In this lesson we will review what you have learned in previous classes about mathematical equations of
relationships and corresponding graphical representations and how these enable us to address a range of
mathematical applications. We will review key properties of mathematical relationships that will allow us
to solve a variety of problems. We will examine examples of how equations and graphs can be used to
model real-life situations.
Lets begin our discussion with some examples of algebraic equations:
Example 1: y = x2 The equation has ordered pairs of numbers (x, y) as solutions. Recall that a particular
pair of numbers is a solution if direct substitution of the x and y values into the original equation yields a
true equation statement. In this example, several solutions can be seen in the following table:

www.ck12.org

y = x2 + 2x 1

14

We can graphically represent the relationships in a rectangular coordinate system, taking the x as the
horizontal axis and the y as the vertical axis. Once we plot the individual solutions, we can draw the curve
through the points to get a sketch of the graph of the relationship:

We call this shape a parabola and every quadratic function, f (x) = ax2 +bx+c, a , 0 has a parabola-shaped
graph. Lets recall how we analytically nd the key points on the parabola. The vertex will be the lowest
point, (1, 2). In general, the vertex is located at the point (b/2a, f (b/2a)). We then can identify points
crossing the x and y axes. These are called the intercepts of the equation. The yintercept is found by
setting x = 0 in the equation, and then solving for y as follows:
y = 02 + 2(0) 1 = 1. The yintercept is located at (0, 1).
The xintercept is found by setting y = 0 in the equation, and solving for x as follows: 0 = x2 + 2x 1

Using the
quadratic formula, we nd that x = 1 2. The xintercepts are located at (1 2, 0) and
(1 + 2, 0).
Finally, recall that we dened the symmetry of a graph. We noted examples of vertical and horizontal line
symmetry as well as symmetry about particular points. For the current example, we note that the graph
has symmetry in the vertical line x = 1. The graph with all of its key characteristics is summarized below:
www.ck12.org

Lets look at a couple of more examples.

Example 2:
Here are some other examples of equations with their corresponding graphs:
y = 2x + 3

x 2 + y2 = 4

www.ck12.org

y = x3 9x

Example 3:
We recall the rst equation as linear so that its graph is a straight line. Can you determine the intercepts?
Solution:
xintercept at (3/2, 0) and yintercept at (0, 3).
Example 4:
We recall from pre-calculus that the second equation is that of a circle with center (0, 0) and radius r = 2.
Can you show analytically that the radius is 2
Solution:
Find the four intercepts, by setting x = 0 and solving for y, and then setting y = 0 and solving for x.
Example 5:
www.ck12.org

The third equation is an example of a polynomial relationship. Can you nd the intercepts analytically?
Solution:
We can nd the xintercepts analytically by setting y = 0 and solving for x. So, we have
x3 9x = 0
x(x2 9) = 0
x(x 3)(x + 3) = 0
x = 0, x = 3, x = 3.
So the xintercepts are located at (3, 0), (0, 0), and (3, 0). Note that (0, 0) is also the yintercept. The
yintercepts can be found by setting x = 0. So, we have
x3 9x = y
(0)3 9(0) = y
y = 0.
Sometimes we wish to look at pairs of equations and examine where they have common solutions. Consider
the linear and quadratic graphs of the previous examples. We can sketch them on the same axes:

We can see that the graphs intersect at two points. It turns out that we can solve the problem of nding the
points of intersections analytically and also by using our graphing calculator. Lets review each method.
Analytical Solution
Since the points of intersection are on each graph, we can use substitution, setting the general ycoordinates
equal to each other, and solving for x.
2x + 3 = x2 + 2x 1
0 = x2 4
x = 2, x = 2.
We substitute each value of x into one of the original equations and nd the points of intersections at
(2, 1) and (2, 7).

www.ck12.org

Graphing Calculator Solution

Once we have entered the relationships on the Y= menu, we press 2nd [CALC] and choose #5 Intersection from the menu. We then are prompted with a cursor by the calculator to indicate which two graphs
we want to work with. Respond to the next prompt by pressing the left or right arrows to move the cursor
near one of the points of intersection and press [ENTER]. Repeat these steps to nd the location of the
second point.
We can use equations and graphs to model real-life situations. Consider the following problem.
Example 6: Linear Modeling
The cost to ride the commuter train in Chicago is $2. Commuters have the option of buying a monthly
coupon book costing $5 that allows them to ride the train for $1.5 on each trip. Is this a good deal for
someone who commutes every day to and from work on the train?
Solution:
We can represent the cost of the two situations, using the linear equations and the graphs as follows:
C1 (x) = 2x
C2 (x) = 1.5x + 5

As before, we can nd the point of intersection of the lines, or in this case, the break-even value in terms
of days, by solving the equation:
C1 (x) = C2 (x)
2x = 1.5x + 5
x = 10.
So, even though it costs more to begin with, after 10 days the cost of the coupon book pays o and from
that point on, the cost is less than for those riders who did not purchase the coupon book.
Example 7: Non-Linear Modeling
The cost of disability benets in the Social Security program for the years 2000 - 2005 can be modeled as
a quadratic function. The formula
www.ck12.org

Y = 0.5x2 + 2x + 4

indicates the number of people Y, in millions, receiving Disability Benets x years after 2000. In what year
did the greatest number of people receive benets? How many people received benets in that year?
Solution:
We can represent the graph of the relationship using our graphing calculator.

The vertex is the maximum point on the graph and is located at (2, 6). Hence in year 2002 a total of
6 million people received benets.

Lesson Summary
1.
2.
3.
4.

Reviewed
Reviewed
Reviewed
Reviewed

graphs of equations
how to nd the intercepts of a graph of an equation and to nd symmetry in the graph
how relationships can be used as models of real-life phenomena
how to solve problems that involve graphs and relationships

Review Questions
In each of problems 1 - 4, nd a pair of solutions of the equation, the intercepts of the graph, and determine
if the graph has symmetry.

1. 2x 3y = 5

www.ck12.org

2. 3x2 y = 5

3. y = x3 x

www.ck12.org

4. y = x3 + x2 6x

5. Once a car is driven o of the dealership lot, it loses a signicant amount of its resale value. The
graph below shows the depreciated value of a BMW versus that of a Chevy after t years. Which of
the following statements is the best conclusion about the data?

(a) You should buy a BMW because they are better cars.
(b) BMWs appear to retain their value better than Chevys.
(c) The value of each car will eventually be $0.
6. Which of the following graphs is a more realistic representation of the depreciation of cars.

www.ck12.org

7. A rectangular swimming pool has length that is 25 yards greater than its width.
(a) Give the area enclosed by the pool as a function of its width.
(b) Find the dimensions of the pool if it encloses an area of 264 square yards.
8. Suppose you purchased a car in 2004 for $18, 000. You have just found out that the current year
2008 value of your car is $8, 500. Assuming that the rate of depreciation of the car is constant, nd
a formula that shows changing value of the car from 2004 to 2008.
9. For problem #8, in what year will the value of the vehicle be less than $1, 400?
10. For problem #8, explain why using a constant rate of change for depreciation may not be the best
way to model depreciation.

Review Answers
1. (1, 1) and (4, 1) are two solutions. The intercepts are located at (0, 5/3) and (5/2, 0). We have a
linear relationship between x and y, so its graph can be sketched as the line passing through any two
solutions.
2. by solving (fory, we) have y = 3x2 5, so two solutions are (1, 2) and (1, 2). The xintercepts are
located at

5
3, 0

and the yintercept is located at (0, 5). The graph is symmetric in the yaxis.

3. Using your graphing calculator, enter the relationship on the Y= menu. Viewing a table of points,
we see many solutions, say (2, 6) and (2, 6), and the intercepts at (0, 0), (1, 0) and (1, 0). By
inspection we see that the graph is symmetric about the origin.
4. Using your graphing calculator, enter the relationship on the Y= menu. Viewing a table of points,
we see many solutions, say (2, 0), and (1, 6), and the intercepts located at (0, 0), (3, 0), and (2, 0).
By inspection we see that the graph does not have any symmetry.
5. b.
6. c. because you would expect (1) a decline as soon as you bought the car, and (2) the value to be
declining more gradually after the initial drop.
7. The pools surface area as a function of width:
8. (a) A(w) = w2 + 25w
(b) The pool has area 264 when w = 8, l = 33.
9. The rate of change will be (9500/4) = 2375. The formula will be y = 2375x + 18000.
10. At the time x = 8, or equivalently in the year 2112, the car will be valued at $1375.
11. A linear model may not be the best function to model depreciation because the graph of the function
decreases as time increases; hence at some point the value will take on negative real number values,
an impossible situation for the value of real goods and products.
www.ck12.org

10

1.2 Relations and Functions

Learning Objectives
A student will be able to:

Identify functions from various relationships.

Review function notation.
Determine domains and ranges of particular functions.
Identify key properties of some basic functions.
Sketch graphs of basic functions.
Sketch variations of basic functions using transformations.
Compose functions.

Introduction
In our last lesson we examined a variety of mathematical equations that expressed mathematical relationships. In this lesson we will focus on a particular class of relationships called functions, and examine their
key properties. We will then review how to sketch graphs of some basic functions that we will revisit later
in this class. Finally, we will examine a way to combine functions that will be important as we develop
the key concepts of calculus.
Lets begin our discussion by reviewing four types of equations we examined in our last lesson.
Example 1:
y = x2

y = 2x + 3

11

www.ck12.org

x 2 + y2 = 4

y = x3 9x

www.ck12.org

12

Of these, the circle has a quality that the other graphs do not share. Do you know what it is?
Solution:
The circles graph
where a particular xvalue has two points associated with it; for example,
includes points
the points (1, 3) and (1, 3) are both solutions to the equation x2 + y2 = 4. For each of the other
relationships, a particular xvalue has exactly one yvalue associated with it.)
The relationships that satisfy the condition that for each xvalue there is a unique yvalue are called
functions. Note that we could have determined whether the relationship satised this condition by a
graphical test, the vertical line test. Recall the relationships of the circle, which is not a function. Lets
compare it with the parabola, which is a function.

If we draw vertical lines through the graphs as indicated, we see that the condition of a particular xvalue
having exactly one yvalue associated with it is equivalent to having at most one point of intersection with
any vertical line. The lines on the circle intersect the graph in more than one point, while the lines drawn
on the parabola intersect the graph in exactly one point. So this vertical line test is a quick and easy way
to check whether or not a graph describes a function.
We want to examine properties of functions such as function notation, their domain and range (the sets of
x and y values that dene the function), graph sketching techniques, how we can combine functions to get
new functions, and also survey some of the basic functions that we will deal with throughout the rest of
this book.
Lets start with the notation we use to describe functions. Consider the example of the linear function
y = 2x + 3. We could also describe the function using the symbol f (x) and read as f of x to indicate the
yvalue of the function for a particular xvalue. In particular, for this function we would write f (x) = 2x+3

13

www.ck12.org

and indicate the value of the function at a particular value, say x = 4 as f (4) and nd its value as follows:
f (4) = 2(4) + 3 = 11. This statement corresponds to the solution (4, 11) as a point on the graph of the
function. It is read, f of x is 11.
We can now begin to discuss the properties of functions, starting with the domain and the range of a
function. The domain refers to the set of xvalues that are inputs in the function, while the range refers
to the set of yvalues that the function takes on. Recall our examples of functions:
Linear Function g(x) = 2x + 3
Quadratic Function f (x) = x2
Polynomial Function p(x) = x3 9x
We rst note that we could insert any real number for an xvalue and a well-dened yvalue would come
out. Hence each function has the set of all real numbers as a domain and we indicate this in interval form
as D : (, ). Likewise we see that our graphs could extend up in a positive direction and down in a
negative direction without end in either direction. Hence we see that the set of yvalues, or the range, is
the set of all real numbers R : (, ).
Example 2:
Determine the domain and range of the function.

f (x) = 1/(x2 4).

Solution:
We note that the condition for each yvalue is a fraction that includes an x term in the denominator. In
deciding what set of xvalues we can use, we need to exclude those values that make the denominator
equal to 0. Why? (Answer: division by 0 is not dened for real numbers.) Hence the set of all
permissible xvalues, is all real numbers except for the numbers (2, 2), which yield division by zero. So
on our graph we will not see any points that correspond to these xvalues. It is more diicult to nd the
range, so lets nd it by using the graphing calculator to produce the graph.

From the graph, we see that every y , 0 value in (, ) (or All real numbers) is represented; hence the
range of the function is {, 0} {0, }. This is because a fraction with a non-zero numerator never equals
zero.
Eight Basic Functions
We now present some basic functions that we will work with throughout the course. We will provide a list
of eight basic functions with their graphs and domains and ranges. We will then show some techniques
that you can use to graph variations of these functions.
Linear
f (x) = x
Domain = All reals
Range = All reals
www.ck12.org

14

Square (Quadratic)
f (x) = x2
Domain = All reals
Range = {y 0}

Cube (Polynomial)
f (x) = x3
Domain = All reals
Range = All reals

15

www.ck12.org

Square Root

f (x) = x
Domain = {x 0}
Range = {y 0}

Absolute Value
f (x) = |x|
Domain = All reals
Range = {y 0}

www.ck12.org

16

Rational
f (x) = 1/x
Domain = {x , 0}
Range = {y , 0}

Sine
f (x) = sin x
Domain = All reals
Range = {1 y 1}

17

www.ck12.org

Cosine
f (x) = cos x
Domain = All reals
Range = {1 y 1}

Graphing by Transformations
Once we have the basic functions and each graph in our memory, we can easily sketch variations of these.
In general, if we have f (x), and c is some constant value, then the graph of f (x c) is just the graph of
f (x) shifted c units to the right. Similarly, the graph of f (x + c) is just the graph of f (x) shifted c units to
the left.
Example 3:
f (x) = x2

www.ck12.org

18

f (x) = (x 2)2

In addition, we can shift graphs up and down. In general, if we have f (x), and c is some constant value,
then the graph of f (x) + c is just the graph of f (x) shifted c units up on the yaxis. Similarly, the graph
of f (x) c is just the graph of f (x) shifted c units down on the yaxis.
Example 4:

f (x) = x

19

www.ck12.org

f (x) =

x+3

We can also ip graphs in the xaxis by multiplying by a negative coeicient.

f (x) = x2

www.ck12.org

20

f (x) = x2

21

www.ck12.org

Finally, we can combine these transformations into a single example as follows.

Example 5:
f (x) = (x 2)2 + 3. The graph will be generated by taking f (x) = x2 , ipping in the yaxis, and moving
it two units to the right and up three units.

www.ck12.org

22

Function Composition
The last topic for this lesson involves a way to combine functions called function composition. Composition of functions enables us to consider the eects of one function followed by another. Our last example
of graphing by transformations provides a nice illustration. We can think of the nal graph as the eect
of taking the following steps:
x (x 2)2 (x 2)2 + 3
We can think of it as the application of two functions. First, g(x) takes x to (x 2)2 and then we apply
a second function, f (x) to those yvalues, with the second function adding +3 to each output. We would
write the functions as
f (g(x)) = (x 2)2 + 3 where g(x) = (x 2)2 and f (x) = x + 3. We call this operation the composing of f
with g and use notation f g. Note that in this example, f g , g f. Verify this fact by computing g f
right now. (Note: this fact can be veried algebraically, by showing that the expressions f g and g f
dier, or by showing that the dierent function decompositions are not equal for a specic value.)

Lesson Summary
1.
2.
3.
4.
5.
6.
7.

Learned to identify functions from various relationships.

Reviewed the use of function notation.
Determined domains and ranges of particular functions.
Identied key properties of basic functions.
Sketched graphs of basic functions.
Sketched variations of basic functions using transformations.
Learned to compose functions.

Review Questions
In problems 1 - 2, determine if the relationship is a function. If it is a function, give the domain and range
of the function.

23

www.ck12.org

1.

2.

In problems 3 - 5, determine the domain and range of the function and sketch the graph if no graph is
provided.
2

3. f (x) = x3x
2 1
4. y = x + 3

www.ck12.org

24

5. f (x) = |2x 3| 2

In problems 6 - 8, sketch the graph using transformations of the graphs of basic functions.

6.
7.
8.
9.

f (x) = (x + 2)2 + 5
1
f (x) = x2
+3

y = x 2 + 3
Find the composites, f g and g f for the following functions.
f (x) = 3x + 2, g(x) =

10. Find the composites, f g and g f for the following functions.

f (x) = x2 , g(x) =

Review Answers
1. The relationship is a function. Domain is All Real Numbers and range = {2 y 2}.
2. The relationship is not a function.
3. The domain is {x , 1, 1}. Use the graph and view the table of solutions to determine the range.
Using your graphing calculator, enter the relationship on the Y= menu. Viewing the table shows
that range {y > 3} {y 0}.

25

www.ck12.org

4. Domain = {x 3}, range = {y 0}.

5. This is the basic absolute value function shifted 3/2 units to the right and down two units. Domain
is All Reals and range is {y 2}.
6. Reect and shift the general quadratic function as indicated here:

7. Reect and shift the general rational function as indicated here:

www.ck12.org

26

8. Reect and shift the general radical function as indicated here:

9. f g = 3 x + 2, g f = 3x + 2.
10. f g = x, g f = x; any functions where f g = g f = x; are called inverses; in this problem f and
g are inverses of one another. Note that the domain for f g is restricted to only positive numbers
and zero.

1.3 Models and Data

Learning Objectives
A student will be able to:
Fit data to linear models.
Fit data to quadratic models.

27

www.ck12.org

 Fit data to trigonometric models.

Fit data to exponential growth and decay models.

Introduction
In our last lesson we examined functions and learned how to classify and sketch functions. In this lesson we
will use some classic functions to model data. The lesson will be a set of examples of each of the models.
For each, we will make extensive use of the graphing calculator.
Lets do a quick review of how to model data on the graphing calculator.
Enter Data in Lists
Press [STAT] and then [EDIT] to access the lists, L1 - L6.
View a Scatter Plot
Press 2nd [STAT PLOT] and choose accordingly.
Then press [WINDOW] to set the limits of the axes.
Compute the Regression Equation
Press [STAT] then choose [CALC] to access the regression equation menu. Choose the appropriate
regression equation (Linear, Quad, Cubic, Exponential, Sine).
Graph the Regression Equation Over Your Scatter Plot
Go to Y=> [MENU] and clear equations. Press [VARS], then enter 5 and EQ and press [ENTER]
(This series of entries will copy the regression equation to your Y = screen.) Press [GRAPH] to view
the regression equation over your scatter plot
Plotting and Regression in Excel
You can also do regression in an Excel spreadsheet. To start, copy and paste the table of data into Excel.
With the two columns highlighted, including the column headings, click on the Chart icon and select
XY scatter. Accept the defaults until a graph appears. Select the graph, then click Chart, then Add
Trendline. From the choices of trendlines choose Linear.
Now lets begin our survey of the various modeling situations.
Linear Models
For these kinds of situations, the data will be modeled by the classic linear equation y = mx + b. Our task
will be to nd appropriate values of m and b for given data.
Example 1:
It is said that the height of a person is equal to his or her wingspan (the measurement from ngertip to
ngertip when your arms are stretched horizontally). If this is true, we should be able to take a table of
measurements, graph the measurements in an x y coordinate system, and verify this relationship. What
kind of graph would you expect to see? (Answer: You would expect to see the points on the line
y = x.)
Suppose you measure the height and wingspans of nine of your classmates and gather the following data.
Use your graphing calculator to see if the following measurements t this linear model (the line y = x).
Table 1.1:
Height (inches)

Wingspan (inches)

67

65

www.ck12.org

28

Table 1.1: (continued)

Height (inches)

Wingspan (inches)

64
56
60
62
71
72
68
65

63
57
61
63
70
69
67
65

We observe that only one of the measurements has the condition that they are equal. Why arent more
of the measurements equal to each other? (Answer: The data do not always conform to exact
specications of the model. For example, measurements tend to be loosely documented so
there may be an error arising in the way that measurements were taken.)
We enter the data in our calculator in L1 and L2. We then view a scatter plot. (Caution: note that
the data ranges exceed the viewing window range of [10, 10]. Change the window ranges accordingly to
include all of the data, say [40, 80].)
Here is the scatter plot:

Now let us compute the regression equation. Since we expect the data to be linear, we will choose the
linear regression option from the menu. We get the equation y = .76x + 14.
In general we will always wish to graph the regression equation over our data to see the goodness of t.
Doing so yields the following graph, which was drawn with Excel:

Since our calculator will also allow for a variety of non-linear functions to be used as models, we can
therefore examine quite a few real life situations. We will rst consider an example of quadratic modeling.
Quadratic Models
Example 2:

29

www.ck12.org

The following table lists the number of Food Stamp recipients (in millions) for each year after 1990.
Table 1.2:
years after 1990

Participants

1
2
3
4
5
6
7
8
9
10

22.6
25.4
27.0
27.5
26.6
2.55
22.5
19.8
18.2
17.2

We enter the data in our calculator in L3 and L4 (that enables us to save the last examples data). We
then will view a scatter plot. Change the window ranges accordingly to include all of the data. Use [2, 10]
for x and [2, 30] for y.
Here is the scatter plot:

Now let us compute the regression equation. Since our scatter plot suggests a quadratic model for the
data, we will choose Quadratic Regression from the menu. We get the equation:
y = 0.30x2 + 2.38x + 21.67.
Lets graph the equation over our data. We see the following graph:

Trigonometric Models
The following example shows how a trigonometric function can be used to model data.
Example 3:
With the skyrocketing cost of gasoline, more people have looked to mass transit as an option for getting
around. The following table uses data from the American Public Transportation Association to show the
number of mass transit trips (in billions) between 1992 and 2000.
Table 1.3:
year

Trips (billions)

1992
1993

8.5
8.2

www.ck12.org

30

Table 1.3: (continued)

year

Trips (billions)

1994
1995
1996
1997
1998
1999
2000

7.93
7.8
7.87
8.23
8.6
9.08
9.4

We enter the data in our calculator in L5 and L6. We then will view a scatter plot. Change the window
ranges accordingly to include all of the data. Use [2, 10] for both x and y ranges.
Here is the scatter plot:

Now let us compute the regression equation. Since our scatter plot suggests a sine model for the data, we
will choose Sine Regression from the menu. We get the equation:
y = .9327 sin(.4681x + 2.8734) + 8.7358.
Let us graph the equation over our data. We see the following graph:

This example suggests that the sine over time t is a function that is used in a variety of modeling situations.
Caution: Although the t to the data appears quite good, do we really expect the number of trips to
continue to go up and down in the future? Probably not. Here is what the graph looks like when projected
an additional ten years:

Exponential Models
Our last class of models involves exponential functions. Exponential models can be used to model growth
and decay situations. Consider the following data about the declining number of farms for the years 1980
- 2005.
Example 4:
The number of dairy farms has been declining over the past 20+ years. The following table charts the
decline:

31

www.ck12.org

Table 1.4:
Year

Farms (thousands)

1980
1985
1990
1995
2000
2005

334
269
193
140
105
67

We enter the data in our calculator in L5 (again entering the years as 1, 2, 3...) and L6. We then will view
a scatter plot. Change the window ranges accordingly to include all of the data. For the large yvalues,
choose the range [50, 350] with a scale of 25.
Here is the scatter plot:

Now let us compute the regression equation. Since our scatter plot suggests an exponential model for the
data, we will choose Exponential Regression from the menu. We get the equation: y = 490.6317 .7266 x
Lets graph the equation over our data. We see the following graph:

In the homework we will practice using our calculator extensively to model data.

Lesson Summary
1.
2.
3.
4.

Fit
Fit
Fit
Fit

data
data
data
data

to
to
to
to

linear models.
quadratic models.
trigonometric models.
exponential growth and decay models.

Review Questions
1. Consider the following table of measurements of circular objects:
(a)
(b)
(c)
(d)

Make a scatter plot of the data.

Based on your plot, which type of regression will you use?
Find the line of best t.
Comment on the values of m and b in the equation.

www.ck12.org

32

Table 1.5:
Object

Diameter (cm)

Circumference (cm)

Glass
Flashlight
Aztec calendar
Tylenol bottle
Popcorn can
Salt shaker
Coee canister
Cat food bucket
Dinner plate
Ritz cracker

8.3
5.2
20.2
3.4
13
6.3
11.3
33.5
27.3
4.9

26.5
16.7
61.6
11.6
41.4
20.1
35.8
106.5
85.6
15.5

2. Manatees are large, gentle sea creatures that live along the Florida coast. Many manatees are killed
or injured by power boats. Here are data on powerboat registrations (in thousands) and the number
of manatees killed by boats in Florida from 1987 - 1997.
(a) Make a scatter plot of the data.
(b) Use linear regression to nd the line of best t.
(c) Suppose in the year 2000, powerboat registrations increase to 700, 000. Predict how many
manatees will be killed. Assume a linear model and nd the line of best t.
Table 1.6:
Year

Boats

Manatees killed

1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997

447
460
480
497
512
513
526
557
585
614
645

13
21
24
16
24
21
15
33
34
34
39

3. A passage in Gullivers Travels states that the measurement of Twice around the wrist is once
around the neck. The table below contains the wrist and neck measurements of 10 people.
(a) Make a scatter plot of the data.
(b) Find the line of best t and comment on the accuracy of the quote from the book.
(c) Predict the distance around the neck of Gulliver if the distance around his wrist is found to be
52 cm.

33

www.ck12.org

Table 1.7:
Wrist (cm)

Neck (cm)

17.9
16
16.5
15.9
17
17.3
16.8
17.3
17.7
16.9

39.5
32.5
34.7
32
33.3
32.6
33
31.6
35
34

4. The following table gives womens average percentage of mens salaries for the same jobs for each
5-year period from 1960 - 2005.
(a)
(b)
(c)
(d)

Make a scatter plot of the data.

Based on your sketch, should you use a linear or quadratic model for the data?
Find a model for the data.
Can you explain why the data seems to dip at rst and then grow?
Table 1.8:

Year

Percentage

1960
1965
1970
1975
1980
1985
1990
1995
2000
2005

42
36
30
37
41
42
48
55
58
60

5. Based on the model for the previous problem, when will women make as much as men? Is your
answer a realistic prediction?
6. The average price of a gallon of gas for selected years from 1975 - 2008 is given in the following table:
(a)
(b)
(c)
(d)

Make a scatter plot of the data.

Based on your sketch, should you use a linear, quadratic, or cubic model for the data?
Find a model for the data.
If gas continues to rise at this rate, predict the price of gas in the year 2012.

www.ck12.org

34

Table 1.9:
Year

Cost

1975
1976
1981
1985
1995
2005
2008

1
1.75
2
2.57
2.45
2.75
3.45

7. For the previous problem, use a linear model to analyze the situation. Does the linear method provide
a better estimate for the predicted cost for the year 2011? Why or why not?
8. Suppose that you place $1, 000 in a bank account where it grows exponentially at a rate of 12%
continuously over the course of ve years. The table below shows the amount of money you have at
the end of each year.
(a) Find the exponential model.
(b) In what year will you triple your original amount?
Table 1.10:
Year

Amount

0
1
2
3
4
5
6

1000
1127.50
1271.24
1433.33
1616.07
1822.11
2054.43
9. Suppose that in the previous problem, you started with $3, 000 but maintained the same interest
rate.
(a) Give a formula for the exponential model. (Hint: note the coeicient and exponent in the
previous answer!)
(b) How long will it take for the initial amount, $3, 000, to triple? Explain your answer.

10. The following table gives the average daily temperature for Indianapolis, Indiana for each month of
the year:
(a) Construct a scatter plot of the data.
(b) Find the sine model for the data.
Table 1.11:
Month

Avg Temp (F)

Jan
Feb

22
26.3

35

www.ck12.org

Table 1.11: (continued)

Month

Avg Temp (F)

March
April
May
June
July
Aug
Sept
Oct
Nov
Dec

37.8
51
61.7
75.3
78.5
84.3
68.5
53.2
38.7
26.6

Review Answers
1.
2. (a)
(b)
(c)
(d)
3.
4. (a)
(b)
(c)
5.
6. (a)
(b)
(c)
7.
8. (a)
(b)
(c)
(d)

.
Linear.
y = 3.1334x + .3296.
m is an estimate of , and b should be zero but due to error in measurement it is not.
.
.
y = .120546x 39.0465; about 46 manatees will be killed in the year 2000. Note: there were
actually 81 manatees killed in the year 2000.
.
.
y = 2.0131x 0.2634; 104.42 cm.

.
Quadratic
y = .4848x2 2.4545x + 39.7333.
It might be because the rst wave of women into the workforce tended to take whatever jobs
they could nd without regard for salary.
9. The data suggest that women will reach 100% in 2009; this is unrealistic based on current reports
that women still lag far behind men in equal salaries for equal work.
10.
11. (a) .
(b) Cubic.
(c) y = .0277x3 0.3497x2 + 1.6203x 0.3157.
(d) $12.15.
12. Linear y = 0.35x + 0.88; Predicted cost in 2012 is $4.73; it is hard to say which model works best but
it seems that the use of a cubic model may overestimate the cost in the short term.
13.
14. (a) .
(b) A = 1000 2.7182.12t ; the amount will triple early in Year 9.
15.
www.ck12.org

36

16. (a) A = 3000 2.7182.12t .

(b) The amount will triple early in Year 9 as in the last problem because the exponential equations
3000 = 1000 2.7182.12t and
17. 9000 = 1000 2.7182.12t
18. both reduce to the same equation
19. 3 = 2.7182.12t
20. and hence have the same solution.
21.
22. (a) .
(b) y = 30.07 sin(.5196x 2.1503) + 51.46.

1.4 The Calculus

Learning Objectives
A student will be able to:

Use linear approximations to study the limit process.

Compute approximations for the slope of tangent lines to a graph.
Introduce applications of dierential calculus.

Introduction
In this lesson we will begin our discussion of the key concepts of calculus. They involve a couple of basic
situations that we will come back to time and again throughout the book. For each of these, we will make
use of some basic ideas about how we can use straight lines to help approximate functions.
Lets start with an example of a simple function to illustrate each of the situations.
Consider the quadratic function f (x) = x2 . We recall that its graph is a parabola. Lets look at the point
(1, 1) on the graph.

Suppose we magnify our picture and zoom in on the point (1, 1). The picture might look like this:

37

www.ck12.org

We note that the curve now looks very much like a straight line. If we were to overlay this view with a
straight line that intersects the curve at (1, 1), our picture would look like this:

We can make the following observations. First, this line would appear to provide a good estimate of the
value of f (x) for xvalues very close to x = 1. Second, the approximations appear to be getting closer
and closer to the actual vale of the function as we take points on the line closer and closer to the point
(1, 1). This line is called the tangent line to f (x) at (1, 1). This is one of the basic situations that we will
explore in calculus.
Tangent Line to a Graph
Continuing our discussion of the tangent line to f (x) at (1, 1), we next wish to nd the equation of the
tangent line. We know that it passes though (1, 1), but we do not yet have enough information to generate
its equation. What other information do we need? (Answer: The slope of the line.)
Yes, we need to nd the slope of the line. We would be able to nd the slope if we knew a second point on
the line. So lets choose a point P on the line, very close to (1, 1). We can approximate the coordinates of
P using the function f (x) = x2 ; hence P(x, x2 ). Recall that for points very close to (1, 1), the points on the
line are close approximate points of the function. Using this approximation, we can compute the slope of
the tangent as follows:
m = (x2 1)/(x 1) = x + 1 (Note: We choose points very close to (1, 1) but not the point itself, so x , 1).
In particular, for x = 1.25 we have P(1.25, 1.5625) and m = x + 1 = 2.25. Hence the equation of the tangent
line, in point slope form is y 1 = 2.25(x 1). We can keep getting closer to the actual value of the slope
by taking P closer to (1, 1), or x closer and closer to x = 1, as in the following table:

www.ck12.org

38

P(x, y)

(1.2, 1.44)

2.2

(1.15, 1.3225)

2.15

(1.1, 1.21)

2.1

(1.05, 1.1025)

2.05

(1.005, 1.010025)

2.005

(1.0001, 1.00020001)

2.0001

As we get closer to (1, 1), we get closer to the actual slope of the tangent line, the value 2. We call the
slope of the tangent line at the point (1, 1) the derivative of the function f (x) at the point (1, 1).
Lets make a couple of observations about this process. First, we can interpret the process graphically as
nding secant lines from (1, 1) to other points on the graph. From the diagram we see a sequence of these
secant lines and can observe how they begin to approximate the tangent line
to the graph at (1, 1). The
diagram shows a pair of secant lines, joining (1, 1) with points ( 2, 2) and ( 3, 3).

Second, in examining the sequence of slopes of these secants, we are systematically observing approximate
slopes of the function as point P gets closer to (1, 1). Finally, producing the table of slope values above
was an inductive process in which we generated some data and then looked to deduce from our data the
value to which the generated results tended. In this example, the slope values appear to approach the
value 2. This process of nding how function values behave as we systematically get closer and closer to
particular xvalues is the process of nding limits. In the next lesson we will formally dene this process
and develop some eicient ways for computing limits of functions.
Applications of Dierential Calculus
Maximizing and Minimizing Functions
Recall from Lesson 1.3 our example of modeling the number of Food Stamp recipients. The model was
found to be y = 0.5x2 + 4x + 19 with graph as follows: (Use viewing window ranges of [2, 14] on x and
[2, 30] on y)

39

www.ck12.org

We note that the function appears to attain a maximum value about an xvalue somewhere around x = 4.
Using the process from the previous example, what can we say about the tangent line to the graph for that
x value that yields the maximum y value (the point at the top of the parabola)? (Answer: the tangent
line will be horizontal, thus having a slope of 0.)

Hence we can use calculus to model situations where we wish to maximize or minimize a particular function.
This process will be particularly important for looking at situations from business and industry where
polynomial functions provide accurate models.
Velocity of a Falling Object
We can use dierential calculus to investigate the velocity of a falling object. Galileo found that the
distance traveled by a falling object was proportional to the square of the time it has been falling:
s(t) = 4.9t2 .
The average velocity of a falling object from t = a to t = b is given by (s(b) s(a))/(b a).
HW Problem #10 will give you an opportunity to explore this relationship. In our discussion, we saw
how the study of tangent lines to functions yields rich information about functions. We now consider the
second situation that arises in Calculus, the central problem of nding the area under the curve of a
function f (x).
Area Under a Curve
First lets describe what we mean when we refer to the area under a curve. Lets reconsider our basic
quadratic function f (x) = x2 . Suppose we are interested in nding the area under the curve from x = 0 to
x = 1.

We see the cross-hatched region that lies between the graph and the xaxis. That is the area we wish to
compute. As with approximating the slope of the tangent line to a function, we will use familiar linear
methods to approximate the area. Then we will repeat the iterative process of nding better and better
approximations.
www.ck12.org

40

Can you think of any ways that you would be able to approximate the area? (Answer: One ideas is that
we could compute the area of the square that has a corner at (1, 1) to be A = 1 and then take half to nd
an area A = 1/2. This is one estimate of the area and it is actually a pretty good rst approximation.)

We will use a variation of this covering of the region with quadrilaterals to get better approximations. We
will do so by dividing the xinterval from x = 0 to x = 1 into equal sub-intervals. Lets start by using four
such subintervals as indicated:

We now will construct four rectangles that will serve as the basis for our approximation of the area. The
subintervals will serve as the width of the rectangles. We will take the length of each rectangle to be the
maximum value of the function in the subinterval. Hence we get the following gure:

41

www.ck12.org

If we call the rectangles R1R4, from left to right, then we have the areas
(1)
1
1
R1 = f
= ,
4
4
64
(1)
1
1
R2 = f
= ,
4
2
16
(3)
1
9
R3 = f
= ,
4
4
64
1
1
R4 = f (1) = ,
4
4
and R1 + R2 + R3 + R4 =

30
64

15
32 .

Note that this approximation is very close to our initial approximation of 1/2. However, since we took the
maximum value of the function for a side of each rectangle, this process tends to overestimate the true
value. We could have used the minimum value of the function in each sub-interval. Or we could have used
the value of the function at the midpoint of each sub-interval.
Can you see how we are going to improve our approximation using successive iterations like we did to
approximate the slope of the tangent line? (Answer: we will sub-divide the interval from x = 0
to x = 1 into more and more sub-intervals, thus creating successively smaller and smaller
rectangles to rene our estimates.)
Example 1:
The following table shows the areas of the rectangles and their sum for rectangles having width w = 1/8.
Table 1.12:
Rectangle Ri

Area of Ri

R1
R2
R3
R4
R5
R6
R7
R8

1
512
4
512
9
512
16
512
25
512
36
512
49
512
64
512

A=
Ri = 195
512 . This value is approximately equal to .3803. Hence, the approximation is now quite a bit
less than .5. For sixteen rectangles, the value is 1432
4096 which is approximately equal to .34. Can you guess
what the true area will approach? (Answer: using our successive approximations, the area will
approach the value 1/3.)
We call this process of nding the area under a curve integration of f (x) over the interval [0, 1].
Applications of Integral Calculus
We have not yet developed any computational machinery for computing derivatives and integrals so we
will just state one popular application of integral calculus that relates the derivative and integrals of a
function.
Example 2:
There are quite a few applications of calculus in business. One of these is the cost function C(x) of
www.ck12.org

42

producing x items of a product. It can be shown that the derivative of the cost function that gives the
slope of the tangent line is another function that that gives the cost to produce an additional unit of the
product. This is called the marginal cost and is a very important piece of information for management
to have. Conversely, if one knows the marginal cost as a function of x, then nding the area under the
curve of the function will give back the cost function C(x).

Lesson Summary
1.
2.
3.
4.

We
We
We
We

used linear approximations to study the limit process.

computed approximations for the slope of tangent lines to a graph.
analyzed applications of dierential calculus.
analyzed applications of integral calculus.

Review Questions
1. For the function f (x) = x2 approximate the slope of the tangent line to the graph at the point (3, 9).

(a) Use the following set of xvalues to generate the sequence of secant line slopes: x = 2.9, 2.95, 2.975, 2.995, 2.999
(b) What value does the sequence of slopes approach?
2. Consider the function f (x) = x2 .
(a) For what values of x would you expect the slope of the tangent line to be negative?
(b) For what value of x would you expect the tangent line to have slope m = 0?
(c) Give an example of a function that has two dierent horizontal tangent lines?
3. Consider the function p(x) = x3 x. Generate the graph of p(x) using your calculator.
(a) Approximate the slope of the tangent line to the graph at the point (2, 6). Use the following set
of xvalues to generate the sequence of secant line slopes. x = 2.1, 2.05, 2.005, 2.001, 2.0001.
(b) For what values of x do the tangent lines appear to have slope of 0? (Hint: Use the calculate
function in your calculator to approximate the xvalues.)
(c) For what values of x do the tangent lines appear to have positive slope?
(d) For what values of x do the tangent lines appear to have negative slope?
4. The cost of producing x Hi Fi stereo receivers by Yamaha each week is modeled by the following
function:
C(x) = 850 + 200x .3x2 .
(a) Generate the graph of C(x) using your calculator. (Hint: Change your viewing window to reect
the high y values.)
(b) For what number of units will the function be maximized?
(c) Estimate the slope of the tangent line at x = 200, 300, 400.
(d) Where is marginal cost positive?
5. Find the area under the curve of f (x) = x2 from x = 1 to x = 3. Use a rectangle method that uses
the minimum value of the function within sub-intervals. Produce the approximation for each case of
the subinterval cases.
(a)
(b)
(c)
(d)

four sub-intervals.
eight sub-intervals.
Repeat part a. using a Mid-Point Value of the function within each sub-interval.
Which of the answers in a. - c. provide the best estimate of the actual area?

6. Consider the function p(x) = x3 + 4x.

43

www.ck12.org

(a) Find the area under the curve from x = 0 to x = 1.

(b) Can you nd the area under the curve from x = 1 to x = 0. Why or why not? What is
problematic for this computation?

7. Find the area under the curve of f (x) = x from x = 1 to x = 4. Use the Max Value rectangle
method with six sub-intervals to compute the area.
8. The Eiel Tower is 320 meters high. Suppose that you drop a ball o the top of the tower. The
distance that it falls is a function of time and is given by s(t) = 4.9t2 .

Find the velocity of the ball after 4 seconds. (Hint: the average velocity for a time interval is average
velocity = change in distance/change in time. Investigate the average velocity for t intervals close
to t = 4 such as 3.9 t 4 and closer and see if a pattern is evident.)

Review Answers
1.
2. (a)
(b)
3.
4. (a)
(b)
(c)
5.
6. (a)
(b)
(c)
(d)
7.
8. (a)
(b)
(c)
(d)
9.
10. (a)
(b)
(c)
(d)
11.
12. (a)
(b)

.
m=6
For x < 0
At x = 0 the tangent line is horizontal and thus has slope of 0.
Many dierent examples; for instance, a polynomial function such as p(x) = x3 4x.
Slope tends toward m = 11.
x = 57.
x < .57, x > .57.
.57 < x < .57.
.
x = 500.
.
x < 500.
.
.
.
.
.
The graph drops below the xaxis into the third quadrant. Hence we are not nding the area
below the curve but actually the area between the curve and the x-axis. But note that the curve
is symmetric about the origin. Hence the region from x = 1 to x = 0 will have the same area
as the region from x = 0 to x = 1.

13. .
m
.
14. 39.2 sec
www.ck12.org

44

1.5 Finding Limits

Learning Objectives
A student will be able to:

Find the limit of a function numerically.

Find the limit of a function using a graph.
Identify cases when limits do not exist.
Use the formal denition of a limit to solve limit problems.

Introduction
In this lesson we will continue our discussion of the limiting process we introduced in Lesson 1.4. We will
examine numerical and graphical techniques to nd limits where they exist and also to examine examples
where limits do not exist. We will conclude the lesson with a more precise denition of limits.
Lets start with the notation that we will use to denote limits. We indicate the limit of a function as the
x values approach a particular value of x, say a, as
lim f (x).
xa

So, in the example from Lesson 1.3 concerning the function f (x) = x2 , we took points that got closer to
the point on the graph (1, 1) and observed the sequence of slope values of the corresponding secant lines.
Using our limit notation here, we would write
lim

x1

x2 1
.
x1

Recall also that we found that the slope values tended to the value x = 2; hence using our notation we can
write
lim

x1

x2 1
= 2.
x1

Finding Limits Numerically

In our example in Lesson 1.3 we used this approach to nd that lim x1
to a more complicated function.
Consider the rational function f (x) =

x+3
.
x2 +x6

x2 1
x1

= 2. Lets apply this technique

Lets nd the following limit:

lim

x3 x2

x+3
.
+x6

Unlike our simple quadratic function, f (x) = x2 , it is tedious to compute the points manually. So lets use
the [TABLE] function of our calculator. Enter the equation in your calculator and examine the table of
points of the function. Do you notice anything unusual about the points? (Answer: There are error
readings indicated for x = 3, 2 because the function is not dened at these values.)
Even though the function is not dened at x = 3, we can still use the calculator to read the yvalues for
x values very close to x = 3. Press 2ND [TBLSET] and set Tblstart to 3.2 and to 0.1 (see screen
on left below). The resulting table appears in the middle below.

45

www.ck12.org

Can you guess the value of lim x3 x2x+3

? If you guessed .20 = (1/5) you would be correct. Before
+x6
we nalize our answer, lets get even closer to x = 3 and determine its function value using the [CALC
VALUE] tool.
Press 2ND [TBLSET] and change Indpnt from Auto to Ask. Now when you go to the table, enter
x = 2.99999. and press [ENTER] and you will see the screen on the right above. Press [ENTER] and
see that the function value is x = 0.2, which is the closest the calculator can display in the four decimal
= 15 .
places allotted in the table. So our guess is correct and lim x3 x2x+3
+x6
Finding Limits Graphically
Lets continue with the same problem but now lets focus on using the graph of the function to determine
its limit.
lim

x3

x2

x+3
+x6

We enter the function in the Y = menu and sketch the graph. Since we are interested in the value of the
function for x close to x = 3, we will look to [ZOOM] in on the graph at that point.

Our graph above is set to the normal viewing window [10, 10]. Hence the values of the function appear
to be very close to 0. But in our numerical example, we found that the function values approached
.20 = (1/5). To see this graphically, we can use the [ZOOM] and [TRACE] function of our calculator.
Begin by choosing [ZOOM] function and choose [BOX]. Using the directional arrows to move the cursor,
make a box around the x value 3. (See the screen on the left below Press [ENTER] and [TRACE] and
you will see the screen in the middle below.) In [TRACE] mode, type the number 2.99999 and press
[ENTER]. You will see a screen like the one on the right below.

The graphing calculator will allow us to calculate limits graphically, provided that we have the function
rule for the function so that we can enter its equation into the calculator. What if we have only a graph
given to us and we are asked to nd certain limits?
It turns out that we will need to have pretty accurate graphs that include suicient detail about the
location of data points. Consider the following example.
www.ck12.org

46

Example 1:
Find lim x3 f (x) for the function pictured here. Assume units of value 1 for each unit on the axes.

By inspection, we see that as we approach the value x = 3 from the left, we do so along what appears to
be a portion of the horizontal line y = 2. We see that as we approach the value x = 3 from the right, we
do so along a line segment having positive slope. In either case, the y values of f (x) approaches y = 2.
Nonexistent Limits
We sometimes have functions where lim xa f (x) does not exist. We have already seen an example of a
function where our a value was not in the domain of the function. In particular, the function was not
dened for x = 3, 2, but we could still nd the limit as x 3.
lim

x3 x2

x+3
1
=
5
+x6

What do you think the limit will be as we let x 2?

lim

x2 x2

x+3
+x6

Our inspection of the graph suggests that the function around x = 2 does not appear to approach a
particular value. For x > 2, the points all lie in the rst quadrant and appear to grow very quickly to large

47

www.ck12.org

positive numbers as we get close to x = 2. Alternatively, for x < 2 we see that the points all lie in the fourth
quadrant and decrease to large negative numbers. If we inspect actual values very close to x = 2 we can
see that the values of the function do not approach a particular value.
x

1.999

1000

1.9999

10000

ERROR

2.001

1000

2.0001

10000

For this example, we say that lim x2

x+3
x2 +x6

does not exist.

Formal Denition of a Limit

We conclude this lesson with a formal denition of a limit.
Denition:
We say that the limit of a function f (x) at a is L, written as lim xa f (x) = L , if for every open interval
D of L, there exists an open interval N of a, that does not include a, such that f (x) is in D for every
x in N.
This denition is somewhat intuitive to us given the examples we have covered. Geometrically, the
denition means that for any lines y = b1 , y = b2 below and above the line y = L, there exist vertical
lines x = a1 , x = a2 to the left and right of x = a so that the graph of f (x) between x = a1 and x = a2
lies between the lines y = b1 and y = b2 . The key phrase in the above statement is for every open
interval D, which means that even if D is very, very small (that is, f (x) is very, very close to L), it
still is possible to nd interval N where f (x) is dened for all values except possibly x = a.

Example 2:
Use the denition of a limit to prove that
lim (2x + 1) = 7.
x3

We need to show that for each open interval of 7, we can nd an open neighborhood of 3, that does not
include 3, so that all x in the open neighborhood map into the open interval of 7.
Equivalently, we must show that for every interval of 7, say (7 , 7 + ), we can nd an interval of 3, say
(3 , 3 + ), such that (7 < 2x + 1 < 7 + ) whenever (3 < x < 3 + ).
www.ck12.org

48

The rst inequality is equivalent to 6 < 2x < 6 + and solving for x, we have
3
Hence if we take =

< x<3+ .
2
2

, we will have 3 < x < 3 + 7 < 2x + 1 < 7 + .

Fortunately, we do not have to do this to evaluate limits. In Lesson 1.6 we will learn several rules that will
make the task manageable.

Lesson Summary
1.
2.
3.
4.

We
We
We
We

learned to nd the limit of a function numerically.

learned to nd the limit of a function using a graph.
identied cases when limits do not exist.
used the formal denition of a limit to solve limit problems.

Multimedia Links
For another look at the denition of a limit, the series of videos at Tutorials for the Calculus Phobe has
a nice intuitive introduction to this fundamental concept (despite the whimsical name). If you want to
experiment with limits yourself, follow the sequence of activities using a graphing applet at Informal Limits.
Directions for using the graphing applets at this very useful site are also available at Applet Intro.

Review Questions
1. Use a table of values to nd lim x2

x2 4
x+2 .

(a) Use xvalues of x = 1.9, 1.99, 1.999, 2.1, 2.099, 2.0099.

(b) What value does the sequence of values approach?
2. Use a table of values to nd lim x 1
2

2x1
2x2 +3x2

(a) Use xvalues of x = .49, .495, .49999, .51, .5099, .500001.

(b) What value does the sequence of values approach?
3. Consider the function p(x) = 3x3 3x. Generate the graph of p(x) using your calculator. Find each
of the following limits if they exist. Use tables with appropriate x values to determine the limits.
(a)
(b)
(c)
(d)

lim x4 (3x3 3x)

lim x4 (3x3 3x)
lim x0 (3x3 3x)
Find the values of the function corresponding to x = 4, 4, 0. How do these function values
compare to the limits you found in #a-c? Explain your answer.

4. Examine the graph of f (x) below to approximate each of the following limits if they exist.

49

www.ck12.org

(a)
(b)
(c)
(d)

lim x3
lim x2
lim x1
lim x4

f (x)
f (x)
f (x)
f (x)

5. Examine the graph of f (x) below to approximate each of the following limits if they exist.

(a)
(b)
(c)
(d)

lim x2 f (x)
lim x0 f (x)
lim x4 f (x)
lim x50 f (x)

In problems #6-8, determine if the indicated limit exists. Provide a numerical argument to justify your
answer.
6. lim x2 (x2 + 3)
7. lim x1 xx+1
2 1
8. lim x2 2x + 5
In problems #9-10, determine if the indicated limit exists. Provide a graphical argument to justify your
answer. (Hint: Make use of the [ZOOM] and [TABLE] functions of your calculator to view functions
values close to the indicated x value.
www.ck12.org

50

9. lim x4 (x2 + 3x)

10. lim x1 |x+1|
x+1

Review Answers
1.
2. (a) .
2 4
(b) lim x2 xx+2
= 4
3.
4. (a) .
(b) lim x 1 2x22x1
= 25
+3x2
2
5.
6. (a) lim x4 (3x3 3x) = 180
(b) lim x4 (3x3 3x) = 180
(c) lim x0 (3x3 3x) = 0
(d) They are the same values because the function is dened for each of these xvalues.
7.
8. (a) lim x3 f (x) = 1.5
(b) lim x2 f (x) = 0
(c) lim x1 f (x) = 2
(d) lim x4 f (x) does not exist.
9.
10. (a) lim x2 f (x) = 0
(b) lim x0 f (x) does not exist.
(c) lim x4 f (x) is some number close to 1 and less than 1, but not equal to 1.
(d) lim x50 f (x) is some number close to 1 and less than 1, but not equal to 1.
11. The limit does exist. This can be veried by using the [TRACE] or [TABLE] function of your
calculator, applied to x values very close to x = 2.
12. The limit does exist. This can be veried by using the [TRACE] or [TABLE] function of your
calculator, applied to x values very close to x = 1.
13. The limit does exist. This can be veried by using the [TRACE] or [TABLE] function of your
calculator, applied to x values very close to x = 2.
14. The limit does exist. This can be veried with either the [TRACE] or [TABLE] function of your
calculator.
15. The limit does not exist; [ZOOM] in on the graph around x = 1 and see that the yvalues approach
a dierent value when approached from the right and from the left.

1.6 Evaluating Limits

Learning Objectives
A student will be able to:

Find the limit of basic functions.

Use properties of limits to nd limits of polynomial, rational and radical functions.
Find limits of composite functions.
Find limits of trigonometric functions.
Use the Squeeze Theorem to nd limits.

51

www.ck12.org

Introduction
In this lesson we will continue our discussion of limits and focus on ways to evaluate limits. We will observe
the limits of a few basic functions and then introduce a set of laws for working with limits. We will conclude
the lesson with a theorem that will allow us to use an indirect method to nd the limit of a function.
Direct Substitution and Basic Limits
Lets begin with some observations about limits of basic functions. Consider the following limit problems:
lim 5,
x2

lim x.
x4

These are examples of limits of basic constant and linear functions, f (x) = c and f (x) = mx + b.
We note that each of these functions are dened for all real numbers. If we apply our techniques for nding
the limits we see that
lim 5 = 5,
x2

lim x = 4,
x4

and observe that for each the limit equals the value of the function at the xvalue of interest:
lim 5 = f (5) = 5,
x2

lim x = f (4) = 4.
x4

Hence lim xa f (x) = f (a). This will also be true for some of our other basic functions, in particular all
polynomial and radical functions, provided that the function is dened at x = a. For example, lim x3 x3 =

f (3) = 27 and lim x4 x = f (4) = 2. The properties of functions that make these facts true will
be discussed in Lesson 1.7. For now, we wish to use this idea for evaluating limits of basic functions.
However, in order to evaluate limits of more complex function we will need some properties of limits, just
as we needed laws for dealing with complex problems involving exponents. A simple example illustrates
the need we have for such laws.
Example 1:

Evaluate lim x2 (x3 + 2x). The problem here is that

while we know that the limit of each individual
function of the sum exists, lim x2 x2 = 8 and lim x2 2x = 2, our basic limits above do not tell us what
happens when we nd the limit of a sum of functions. We will state a set of properties for dealing with
such sophisticated functions.
Properties of Limits
Suppose that lim xa f (x) and lim xa g(x) both exist. Then
1.
2.
3.
4.

lim xa [c f (x)] = c lim xa f (x) where c is a real number,

lim xa [ f (x)]n = [lim xa f (x)]n where n is a real number,
lim xa [ f (x) g(x)] = lim xa f (x) lim xa g(x),
lim xa [[f (x)] g(x)] = lim xa f (x) lim xa g(x),

5. lim xa

f (x)
g(x)

www.ck12.org

lim xa f (x)
lim xa g(x)

provided that lim xa g(x) , 0.

52

With these properties we can evaluate a wide range of polynomial and radical functions. Recalling our
example above, we see that

lim (x3 + 2x) = lim (x3 ) + lim ( 2x) = 8 + 2 = 10.

x2

x2

x2

Find the following limit if it exists:

lim (2x2

x4

x).

Since the limit of each function within the parentheses exists, we can apply our properties and nd
lim (2x2

x4

x) = lim 2x2 lim

x4

x4

x.

Observe that the

second limit, lim x4 x,
is an application of Law #2 with n =

lim x4 (2x2 x) = lim x4 2x2 lim x4 x = 32 2 = 30.

1
2.

So we have

In most cases of sophisticated functions, we simplify the task by applying the Properties as indicated. We
want to examine a few exceptions to these rules that will require additional analysis.
Strategies for Evaluating Limits of Rational Functions
Lets recall our example
lim

x1

x2 1
.
x1

We saw that the function did not have to be dened at a particular value for the limit to exist. In this
example, the function was not dened for x = 1. However we were able to evaluate the limit numerically
2 1
by checking functional values around x = 1 and found lim x1 xx1
= 2.
Note that if we tried to evaluate by direct substitution, we would get the quantity 0/0, which we refer to as
an indeterminate form. In particular, Property #5 for nding limits does not apply since lim x1 (x1) =
0. Hence in order to evaluate the limit without using numerical or graphical techniques we make the
following observation. The numerator of the function can be factored, with one factor common to the
denominator, and the fraction simplied as follows:
(x + 1)(x 1)
x2 1
=
= x + 1.
x1
x1
In making this simplication, we are indicating that the original function can be viewed as a linear function
for x values close to but not equal to 1, that is,
x2 1
x1

= x + 1 for x , 1. In terms of our limits, we can say

lim

x1

x2 1
= lim (x + 1) = 1 + 1 = 2.
x1
x1

Example 2:
Find lim x0

x2 +5x
x .

This is another case where direct substitution to evaluate the limit gives the indeterminate form 0/0.
Reducing the fraction as before gives:
lim

x0

x2 + 5x
= lim (x + 5) = 5.
x0
x

53

www.ck12.org

Example 3:

lim

x9

x3
.
x9

In order to evaluate the limit, we need to recall that the dierence of squares of real numbers can be
factored as x2 y2 = (x + y)(x y).
We then rewrite and simplify the original function as follows:

x3
x3
1
=
=
.

x9
( x + 3)( x 3)
( x + 3)
Hence lim x9

x3
x9

= lim x9

1
x+3

= 16 .

You will solve similar examples in the homework where some clever applications of factoring to reduce
fractions will enable you to evaluate the limit.
Limits of Composite Functions
While we can use the Properties to nd limits of composite functions, composite functions will present some
diiculties that we will fully discuss in the next Lesson. We can illustrate with the following examples, one
where the limit exists and the other where the limit does not exist.
Example 4:
Consider f (x) =

1
x+1 ,

g(x) = x2 . Find lim x1 ( f g)(x).

We see that ( f g)(x) = x21+1 and note that property #5 does hold. Hence by direct substitution we have
lim x1 ( f g)(x) = (1)12 +1 = 12 .
Example 5:
1
Consider f (x) = x+1
, g(x) = 1. Then we have that f (g(x)) is undened and we get the indeterminate
form 1/0. Hence lim x1 ( f g)(x) does not exist.

Limits of Trigonometric Functions

In evaluating limits of trigonometric functions we will look to rely more on numerical and graphical techniques due to the unique behavior of these functions. Lets look at a couple of examples.
Example 6:
Find lim x0 sin(x).
We can nd this limit by observing the graph of the sine function and using the [CALC VALUE] function
of our calculator to show that lim x0 sin x = 0.
While we could have found the limit by direct substitution, in general, when dealing with trigonometric
functions, we will rely less on formal properties of limits for nding limits of trigonometric functions and
more on our graphing and numerical techniques.
The following theorem provides us a way to evaluate limits of complex trigonometric expressions.
Squeeze Theorem
Suppose that f (x) g(x) h(x) for x near a, and lim xa f (x) = lim xa h(x) = L.
Then lim xa g(x) = L.
In other words, if we can nd bounds for a function that have the same limit, then the limit of the function
that they bound must have the same limit.
Example 7:
www.ck12.org

54

Find lim x0 x2 cos(10x).

From the graph we note that:

1. The function is bounded by the graphs of x2 and x2
2. lim x0 x2 = lim x0 (x2 ) = 0.
Hence the Squeeze Theorem applies and we conclude that lim x0 x2 cos(10x) = 0.

Lesson Summary
1.
2.
3.
4.

We
We
We
We

learned to nd the limit of basic functions.

learned to nd the limit of polynomial, rational and radical functions.
learned how to nd limits of composite and trigonometric functions.
used the Squeeze Theorem to nd special limits.

Multimedia Links
For an introduction to nding limits (1.0), see Math Video Tutorials by James Sousa, Introduction to
Limits (8:46) .

Figure 1.1: Introduction to Limits (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/492
For more examples nding the limits of functions (1.1)(1.2), see Math Video Tutorials by James Sousa,
Determining the Limits of Functions (8:46) .

Figure 1.2: Determining Limits (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/493

55

www.ck12.org

For a brief, intuitive introduction to the Squeeze Theorem using everyday examples (1.1), see Khan
Academy Squeeze Theorem (7:36) . This video includes a graphical presentation of the Squeeze Theorem.

Figure 1.3: Squeeze Theorem (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/494
This video serves as an introduction to another (much longer) video, Khan Academy Proof of Limit x ->
0 sin(x)/x (18:04) .

Figure 1.4: Proof: lim (sin x)/x (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/495
For a video on determining limits of trigonometric functions (1.2)(1.3), see Math Video Tutorials by
James Sousa, Introduction to Limits (7:20) .

http://mathispower4u.wordpress.com/
http://mathispower4u.wordpress.com/
Figure 1.5: Determining Limits of Trigonometric Functions (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/496

Review Questions
Find each of the following limits if they exist.
1. lim x2 (x2 3x + 4)
2 16
2. lim x4 x x4
www.ck12.org

56

x2

3. lim x4 x4
x2
4. lim x1 x+1
5. lim x1 10x2
3x+1

lim x1 x+32
x1
2 25
lim x5 xx3 125
1
Consider f (x) = x+1
, g(x) = x2 . We found lim x1 ( f g)(x) = 12 Find lim x1 (g f )(x).
Consider function f (x) such that 5x 11 f (x) x2 4x + 9 for x 0. Use the Squeeze Theorem to
nd lim x5 f (x).
10. Use the Squeeze Theorem to show that lim x0 x4 sin( 1x ) = 0
6.
7.
8.
9.

Review Answers
1. lim x2 (x2 3x + 4) = 2.
2 16
=8
2. lim x4 x x4

x2

3. lim x4 x4 = 14
x2
4. lim x1 x+1
does not exist.
10x2
5. lim x1 3x+1 = 6
6.
7.
8.
9.
10.

lim x1 x+32
= 14
x1
2
25
2
lim x5 xx3 125
= 15
lim x1 (g f )(x) does not exist since g( f (x)) is undened.
lim x5 f (x) = 14 since lim x5 (5x 11) = lim x5 (x2 4x + 9) = 14
Note that x4 sin( 1x ) x4 , and since lim x0 x4 = lim x0 (x4 ) = 0, then by the Squeeze Theorem
must have lim x0 x4 sin( 1x ) = 0.

1.7 Continuity
Learning Objectives
A student will be able to:

Learn to examine continuity of functions.

Find one-sided limits.
Understand properties of continuous functions.
Solve problems using the Min-Max theorem.
Solve problems using the Intermediate Value Theorem.

Introduction
In this lesson we will discuss the property of continuity of functions and examine some very important
implications. Lets start with an example of a rational function and observe its graph. Consider the
following function:
f (x) = (x + 1)/(x2 1).
We know from our study of domains that in order for the function to be dened, we must use x , 1, 1.
Yet when we generate the graph of the function (using the standard viewing window), we get the following
picture that appears to be dened at x = 1:

57

www.ck12.org

The seeming contradiction is due to the fact that our original function had a common factor in the
numerator and denominator, x + 1, that cancelled out and gave us a picture that appears to be the graph
of f (x) = 1/(x 1).
But what we actually have is the original function, f (x) = (x + 1)/(x2 1), that we know is not dened at
x = 1. At x = 1, we have a hole in the graph, or a discontinuity of the function at x = 1. That is, the
function is dened for all other xvalues close to x = 1.
Loosely speaking, if we were to hand-draw the graph, we would need to take our pencil o the page when
we got to this hole, leaving a gap in the graph as indicated:

Now we will formalize the property of continuity of a function and provide a test for determining when we
have continuous functions.
Continuity of a Function
Denition:

www.ck12.org

58

The function f (x) is continuous at x = a if the following conditions all hold:

1. a is in the domain of f (x);

2. lim xa f (x) exists;
3. lim xa f (x) = f (a)

Note that it is possible to have functions where two of these conditions are satised but the third is not.
Consider the piecewise function

x, if x , 1
f (x) =

3, if x = 1

In this example we have lim x1 f (x) exists, x = 1 is in the domain of f (x), but lim x1 f (x) , f (1).

One-Sided Limits and Closed Intervals

Lets recall our basic square root function, f (x) =

x.

59

www.ck12.org

Since the domain of f (x) = x is x 0, we see that that lim x0 x does not exist. Specically, we cannot
nd open intervals around x = 0 that satisfy the limit denition. However we do note that as we approach
x = 0 from the right-hand side, we see the successive values tending towards x = 0. This example provides
some rationale for how we can dene one-sided limits.
Denition:

We say that the right-hand limit of a function f (x) at a is b, written as lim xa+ f (x) = b, if for every
open interval N of b, there exists an open interval (a, a + ) contained in the domain of f (x), such
that f (x) is in N for every x in (a, a + ).

For the example above, we write lim x0+ x = 0.

Similarly, we say that the left-hand limit of f (x) at a is b, written as lim xa f (x) = b, if for every open
interval N of b there exists an open interval (a , a) contained in the domain of f (x), such that f (x)
is in N for every x in (a , a).

Example 1:
Find lim x0+

x
|x| .

The graph has a discontinuity at x = 0 as indicated:

www.ck12.org

60

We see that lim x0+

x
|x|

= 1 and also that lim x0

x
|x|

= 1.

Properties of Continuous Functions

Lets recall our example of the limit of composite functions:

f (x) = 1/(x + 1), g(x) = 1.

We saw that f (g(x)) is undened and has the indeterminate form of 1/0. Hence lim x1 ( f g)(x) does not
exist.
In general, we will require that f be continuous at x = g(a) and x = g(a) must be in the domain of ( f g)
in order for lim xa ( f g)(x) to exist.
We will state the following theorem and delay its proof until Chapter 3 when we have learned more about
real numbers.
Min-Max Theorem: If a function f (x) is continuous in a closed interval I, then f (x) has both a maximum
value and a minimum value in I.
Example 2:
Consider f (x) = x3 + 1 and interval I = [2, 2].
The function has a minimum value at value at x = 2, f (2) = 7, and a maximum value at x = 2, where
f (2) = 9

61

www.ck12.org

We will conclude this lesson with a theorem that will enable us to solve many practical problems such as
nding zeros of functions and roots of equations.
Intermediate Value Theorem
If a function is continuous on a closed interval [a, b], then the function assumes every value between f (a)
and f (b).
The proof is left as an exercise with some hints provided. (Homework #10).
We can use the Intermediate Value Theorem to analyze and approximate zeros of functions.
Example 3:
Use the Intermediate Value Function to show that there is at least one zero of the function in the indicated
interval.

f (x) = 3x4 3x3 2x + 1, (1, 2)

We recall that the graph of this function is shaped somewhat like a parabola; viewing the graph in the
standard window, we get the following graph:
www.ck12.org

62

Of course we could zoom in on the graph to see that the lowest point on the graph lies within the fourth
quadrant, but lets use the [CALC VALUE] function of the calculator to verify that there is a zero in
the interval (1, 2). In order to apply the Intermediate Value Theorem, we need to nd a pair of xvalues
that have function values with dierent signs. Lets try some in the table below.
x

f (x)

1.1

.80

1.2

.36

1.3

.37

We see that the sign of the function values changes from negative to positive somewhere between 1.2 and
1.3. Hence, by the Intermediate Value theorem, there is some value c in the interval (1.2, 1.3) such that
f (c) = 0.

Lesson Summary
1.
2.
3.
4.
5.

We
We
We
We
We

learned to examine continuity of functions.

learned to nd one-sided limits.
observed properties of continuous functions.
solved problems using the Min-Max theorem.
solved problems using the Intermediate Value Theorem.

Multimedia Links
For a presentation of continuity using limits (2.0), see Math Video Tutorials by James Sousa, Continuity
Using Limits (5:44) .

63

www.ck12.org

Figure 1.6: Continuity using Limits (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/497
For a video presentation of the Intermediate Value Theorem (3.0), see Just Math Tutoring, Intermediate
Value Theorem (7:53) .

Figure 1.7: Intermediate Value Theorem (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/498

Review Questions
1. Generate the graph of f (x) = (|x + 1|)/(x + 1) using your calculator and discuss the continuity of the
function.
2. Generate the graph of f (x) = (3x 6)/(x2 4) using your calculator and discuss the continuity of the
function.
Compute the limits in #3 - 6.
3. lim x0+
4.
5.
6.

1+ x1
x3 8
lim x2 |x2|(x2)
lim x1+ 2x|x1|
x1
lim x2+ |x+2|+x+2
|x+2|x2

In problems 7 and 8, explain how you know that the function has a root in the given interval. (Hint: Use
the Intermediate Value Function to show that there is at least one zero of the function in the indicated
interval.):
7. f (x) = x3 + 2x2 x + 1, in the interval (2, 3)

8. f (x) = x 3 x 1, in the interval (9, 10)

www.ck12.org

64

9. State whether the indicated xvalues correspond to maximum or minimum values of the function
depicted below.

10. Prove the Intermediate Value Theorem: If a function is continuous on a closed interval [a, b], then
the function assumes every value between f (a) and f (b).

Review Answers
1. While graph of the function appears to be continuous everywhere, a check of the table values indicates
that the function is not continuous at x = 1.
2. While the function appears to be continuous for all x = 2, a check of the table values indicates that
the functionis not continuous at x = 2.
x
3. lim x0+
=2
1+ x1
x3 8
does
|x2|(x2)
2x|x1|
lim x1+ (x1) = 2
lim x2+ |x+2|+x+2
|x+2|x2 = 0

4. lim x2
5.

not exist

6.
7. f (2.5) = .375, f (2.9) = 3.669. By the Intermediate Value Theorem, there is an xvalue c with
f (c) = 0.
8. f (9.1) = .071, f (9.99) = .006. By the Intermediate Value Theorem, there is an xvalue c with
f (c) = 0.
9. x = a is a relative maximum, x = b is an absolute minimum, x = c is an absolute maximum and
x = d is not a maximum nor a minimum.
10. Here is an outline of the proof: we need to show that for every number d between f (a) and f (b),
there exists a number such that f (c) = d.
11. (a) Assume that f (a) < f (c) < f (b).
(b) Let S be the set of x [a, b] > for which f (x) < d. Note that a S , b S . so b is an upper bound
for set S . Hence by the completeness property of the reals, S has an upper bound, c.
(c) There are then three possibilities to explore: f (c) < d, f (c) = d, or f (c) > d. Explore these and
show why f (c) = d.

1.8 Innite Limits

Learning Objectives
A student will be able to:

Find innite limits of functions.

Analyze properties of innite limits.
Identify asymptotes of functions.
Analyze end behavior of functions.

65

www.ck12.org

Introduction
In this lesson we will discuss innite limits. In our discussion the notion of innity is discussed in two
contexts. First, we can discuss innite limits in terms of the value a function as we increase x without
bound. In this case we speak of the limit of f (x) as x approaches and write lim x f (x). We could
similarly refer to the limit of f (x) as x approaches - and write lim x f (x).
The second context in which we speak of innite limits involves situations where the function values increase
without bound. For example, in the case of a rational function such as f (x) = (x + 1)/(x2 + 1), a function
we discussed in previous lessons:

At x = 1, we have the situation where the graph grows without bound in both a positive and a negative
direction. We say that we have a vertical asymptote at x = 1, and this is indicated by the dotted line in
the graph above.
In this example we note that lim x1 f (x) does not exist. But we could compute both one-sided limits as
follows.
lim x1 f (x) = and lim x1 f (x) = +.
More formally, we dene these as follows:
Denition:
The right-hand limit of the function f (x) at x = a is innite, and we write lim xa+ f (x) = , if for every
positive number k, there exists an open interval (a, a + ) contained in the domain of f (x), such that
f (x) is in (k, ) for every x in (a, a + ).
The denition for negative innite limits is similar.
Suppose we look at the function f (x) = (x + 1)/(x2 1) and determine the innite limits lim x f (x) and
lim x f (x).
We observe that as x increases in the positive direction, the function values tend to get smaller. The same
www.ck12.org

66

is true if we decrease x in the negative direction. Some of these extreme values are indicated in the
following table.
x

f (x)

100

.0101

200

.0053

100

.0099

200

.005

We observe that the values are getting closer to f (x) = 0. Hence lim x f (x) = 0 and lim x f (x) = 0.
Since our original function was roughly of the form f (x) = 1x , this enables us to determine limits for
all other functions of the form f (x) = x1p with p > 0. Specically, we are able to conclude that
lim x x1p = 0. This shows how we can nd innite limits of functions by examining the end
behavior of the function f (x) = x1p , p > 0.
The following example shows how we can use this fact in evaluating limits of rational functions.
Example 1:
Find lim x

2x3 x2 +x1
.
x6 x5 +3x4 2x+1

Solution:
Note that we have the indeterminate form, so Limit Property #5 does not hold. However, if we rst divide
both numerator and denominator by the quantity x6 , we will then have a function of the form
f (x)
=
g(x)

2
2x3
xx6 + xx6 x16
x6
5
4
x6
xx6 + 3x
2x
+ x16
x6
x6
x6

2
x3

1
x

1
x4

+
3
x2

1
x5

2
x5

1
x6

1
x6

We observe that the limits lim x f (x) and lim x g(x) both exist. In particular, lim x f (x) = 0 and
2x3 x2 +x1
0
lim x g(x) = 1. Hence Property #5 now applies and we have lim x x6 x
5 +3x4 2x+1 = 1 = 0.

Lesson Summary
1.
2.
3.
4.

We
We
We
We

learned to nd innite limits of functions.

analyzed properties of innite limits.
identied asymptotes of functions.
analyzed end behavior of functions.

Multimedia Links
For more examples of limits at innity (1.0), see Math Video Tutorials by James Sousa, Limits at Innity
(9:42)

Review Questions
In problems 1 - 7, nd the limits if they exist.
1. lim x3+

(x+2)2
(x2)2 1

67

www.ck12.org

Figure 1.8: Limits At Innity (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/499
(x+2)2
(x2)2 1
(x+2)2
lim x1+ (x2)2 1
lim x 2x1
x+1
5
4 +1
lim x x +3x
x3 1
4
2 +3x+1
lim x 3x2x42x
2x2 +x3
2x2 x+3
lim x x5 2x
3 +2x3

2. lim x
3.
4.
5.
6.
7.

In problems 8 - 10, analyze the given function and identify all asymptotes and the end behavior of the
graph.
(x+4)2

8. f (x) = (x4)2 1
9. f (x) = 3x3 x2 + 2x + 2
2 8
10. f (x) = 2xx+2

Review Answers
(x+2)2
= +
(x2)2 1
2
(x+2)
lim x (x2)2 1 = 1
(x+2)2
lim x1+ (x2)2 1 =
lim x 2x1
x+1 = 2
5
4 +1
lim x x +3x
=
x3 1
3x4 2x2 +3x+1
lim x 2x4 2x2 +x3 = 32
2x2 x+3
lim x x5 2x
3 +2x3 = 0

1. lim x3+
2.
3.
4.
5.
6.

7.
8. Zero at x = 4; vertical asymptotes at x = 3, 5; f (x) 1 as x .
9. Zero at x = 1; no vertical asymptotes; f (x) as x ; f (x) as x .
10. Zero at x = 2; no vertical asymptotes but there is a discontinuity at x = 2; f (x) as x ;
f (x) as x .

Texas Instruments Resources

In the CK-12 Texas Instruments Calculus FlexBook, there are graphing calculator activities
designed to supplement the objectives for some of the lessons in this chapter. See http:
//www.ck12.org/flexr/chapter/9726.
www.ck12.org

68

Chapter 2
Derivatives
2.1 Tangent Lines and Rates of Change
Learning Objectives
A student will be able to:
Demonstrate an understanding of the slope of the tangent line to the graph.
Demonstrate an understanding of the instantaneous rate of change.
A car speeding down the street, the ination of currency, the number of bacteria in a culture, and the AC
voltage of an electric signal are all examples of quantities that change with time. In this section, we will
study the rate of change of a quantity and how is it related to the tangent lines on a curve.

The Tangent Line

If two points P(x0 , y0 ) and Q(x1 , y1 ) are two dierent points of the curve y = f (x) (Figure 1), then the slope
of the secant line connecting the two points is given by
m sec =

f (x1 ) f (x0 )
y1 y0
=
x1 x0
x1 x0

(1)

Now if we let the the point x1 approach x0 , Q will approach P along the graph f . Thus the slope of the
secant line will gradually approach the slope of the tangent line as x1 approaches x0 . Therefore (1) becomes
mtan = lim

x1 x0

f (x1 ) f (x0 )
.
x1 x0

(2)

If we let h = x1 x0 , then x1 = x0 + h and h 0 becomes equivalent to x1 x0 , so (2) becomes

mtan = lim

h0

f (x0 + h) f (x0 )
.
h

If the point P(x0 , y0 ) is on the curve f , then the tangent line at P has a slope that is given by

69

www.ck12.org

mtan = lim

h0

f (x0 + h) f (x0 )
h

provided that the limit exists.

Recall from algebra that the point-slope form for the tangent line is given by
y y0 = mtan (x x0 ).
Example 1:
Find the slope of the tangent line to the curve f (x) = x3 passing through point P(2, 8).
Solution:
Since P(x0 , y0 ) = (2, 8), using the slope of the tangent equation,
mtan = lim

h0

f (x0 + h) f (x0 )
h

we get
f (2 + h) f (2)
h0
h
(h3 + 6h2 + 12h + 8) 8
= lim
h0
h
3
2
h + 6h + 12h
= lim
h0
h
2
= lim (h + 6h + 12)

mtan = lim

h0

= 12.
Thus the slope of the tangent line is 12. Using the point-slope formula above,
y 8 = 12(x 2)
or
y = 12x 16
Next we are interested in nding a formula for the slope of the tangent line at any point on the curve f .
Such a formula would be the same formula that we are using except we replace the constant x0 by the
variable x. This yields
mtan = lim

f (x + h) f (x)
.
h

f (x) = lim

f (x + h) f (x)
,
h

h0

We denote this formula by

h0

where f (x) is read f prime of x. The next example illustrate its usefulness.
www.ck12.org

70

Example 2:
If f (x) = x2 3, nd f (x) and use the result to nd the slope of the tangent line at x = 2 and x = 1.
Solution:
Since
f (x) = lim

h0

f (x + h) f (x)
,
h

then

[(x+h)2 3][x2 3]

f(x) =limh0
h
2
2
2
= limh0 x +2xh+hh 3x +3
2
= limh0 2xh+h
h
= limh0 (2x + h)
= 2x

To nd the slope, we simply substitute x = 2 into the result f (x),

f (x) = 2x
f (2) = 2(2)
=4
and
f (x) = 2x
f (1) = 2(1)
= 2
Thus slopes of the tangent lines at x = 2 and x = 1 are 4 and 2, respectively.
Example 3:
Find the slope of the tangent line to the curve y =

1
x

that passes through the point (1, 1).

Solution:
Using the slope of the tangent formula
f (x) = lim

h0

f (x + h) f (x)
h

and substituting y = 1x ,

71

www.ck12.org

(
y = lim

1
x+h

1
x

h0

= lim

xxh
x(x+h)

h
xxh
= lim
h0 hx(x + h)
h
= lim
h0 hx(x + h)
1
= lim
h0 x(x + h)
1
= lim
h0 x(x + h)
1
= 2.
x
h0

Substituting x = 1,
1
1
= 1.

y =

Thus the slope of the tangent line at x = 1 for the curve y =

tangent line, we simply use the point-slope formula,

1
x

is m = 1. To nd the equation of the

y y0 = m(x x0 ),
where (x0 , y0 ) = (1, 1).
y 1 = 1(x 1)
= x + 1 + 1
= x + 2,
which is the equation of the tangent line.

Average Rates of Change

The primary concept of calculus involves calculating the rate of change of a quantity with respect to
another. For example, speed is dened as the rate of change of the distance travelled with respect to time.
If a person travels 120 miles in four hours, his speed is 120/4 = 30 mi/hr. This speed is called the average
speed or the average rate of change of distance with respect to time. Of course the person who travels
120 miles at a rate of 30 mi/hr for four hours does not do so continuously. He must have slowed down
or sped up during the four-hour period. But it does suice to say that he traveled for four hours at an
average rate of 30 miles per hour. However, if the driver strikes a tree, it would not be his average speed
that determines his survival but his speed at the instant of the collision. Similarly, when a bullet strikes a
target, it is not the average speed that is signicant but its instantaneous speed at the moment it strikes.
So here we have two distinct kinds of speeds, average speed and instantaneous speed.
www.ck12.org

72

The average speed of an object is dened as the objects displacement x divided by the time interval t
during which the displacement occurs:
v=

x
x1 x0
=
.
t
t1 t0

Notice that the points (t0 , x0 ) and (t1 , x1 ) lie on the position-versus-time curve, as Figure 1 shows. This
expression is also the expression for the slope of a secant line connecting the two points. Thus we conclude
that the average velocity of an object between time t0 and t1 is represented geometrically by the slope of
the secant line connecting the two points (t0 , x0 ) and (t1 , x1 ). If we choose t1 close to t0 , then the average
velocity will closely approximate the instantaneous velocity at time t0 .

Geometrically, the average rate of change is represented by the slope of a secant line and the instantaneous
rate of change is represented by the slope of the tangent line (Figures 2 and 3).
Average Rate of Change (such as the average velocity)
The average rate of change of x = f (t) over the time interval [t0 , t1 ] is the slope m sec of the secant line to
the points (t0 , f (t0 )) and (t1 , f (t1 )) on the graph (Figure 2).

m sec =

f (t1 ) f (t0 )
t1 t0

73

www.ck12.org

Instantaneous Rate of Change

The instantaneous rate of change of x = f (t) at the time t0 is the slope mtan of the tangent line at the time
t0 on the graph.

mtan = f (t0 ) = lim

t1 t0

f (t1 ) f (t0 )
t1 t0

Example 4:
Suppose that y = x2 3.
1. Find the average rate of change of y with respect to x over the interval [0, 2].
2. Find the instantaneous rate of change of y with respect to x at the point x = 1.
Solution:
1. Applying the formula for Average Rate of Change with f (x) = x2 3 and x0 = 0 and x1 = 2 yields
f (x1 ) f (x0 )
x1 x0
f (2) f (0)
=
20
1 (3)
=
1
=4

m sec =

This means that the average rate of change of y is 4 units per unit increase in x over the interval [0, 2].
2. From the example above, we found that f (x) = 2x, so
mtan = f (x0 )
= f (1)
= 2(1)
= 2
This means that the instantaneous rate of change is negative. That is, y is decreasing at x = 1. It is
decreasing at a rate of 1 units per unit increase in x.
www.ck12.org

74

Multimedia Links
For a video explaining instantaneous rates of change (4.2), see Slopes of Tangents and Instantaneous Rates
of Change (9:26) .

Figure 2.1: Slopes of Tangents and Instantaneous Rate of Change (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/500
For a video with an application regarding velocity (4.2), see Calculus Help: Instantaneous Rates of Change
(9:03) .

Figure 2.2: Calculus Help: Instantaneous Rate of Change (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/501
The following applet illustrates how the slope of a secant line can become the slope of the tangent to a
curve at a point as h 0. Follow the directions on the page to explore how changing the distance between
the two points makes the slope of the secant approach the slope of the tangent Slope at a Point Applet.
Note that the function and the point of tangency can also be edited in this simulator.

Review Questions
1. Given the function y = 1/2 x2 and the values of x0 = 3 and x1 = 4, nd
(a)
(b)
(c)
(d)
(e)

The average rate of change of y with respect to x over the interval [x0 , x1 ].
The instantaneous rate of change of y with respect to x at x0 .
The slope of the tangent line at x1 .
The slope of the secant line between points x0 and x1 .
Make a sketch of y = 1/2 x2 and show the secant and tangent lines at their respective points.

2. Repeat problem #1 for f (x) = 1/x and the values x0 = 2 and x1 = 3.

3. Find the slope of the graph f (x) = x2 + 1 at a general point x. What is the slope of the tangent line
at x0 = 6?

4. Suppose that y = 1/ x.
(a) Find the average rate of change of y with respect to x over the interval [1, 3].
(b) Find the instantaneous rate of change of y with respect to x at point x = 1.

75

www.ck12.org

5. A rocket is propelled upward and reaches a height of h(t) = 4.9t2 in t seconds.

(a)
(b)
(c)
(d)

How high does it reach in 35 seconds?

What is the average velocity of the rocket during the rst 35 seconds?
What is the average velocity of the rocket during the rst 200 meters?
What is the instantaneous velocity of the rocket at the end of the 35 seconds?

6. A particle moves in the positive direction along a straight line so that after t nanoseconds, its traversed
distance is given by (t) = 9.9t3 nanometers.
(a) What is the average velocity of the particle during the rst 2 nanoseconds?
(b) What is the instantaneous velocity of the particle at t = 2 nanoseconds?

Review Answers
1.
2. (a) 72
(b) 3
(c) 4
(d) 72
3.
4. (a) 1
6
(b) 1
4
(c) 1
9
(d) 1
6
5. 2x, 12.
6.
(
)
7. (a) 63 12
(b) 12

8.
9. (a)
(b)
(c)
(d)
10.
11. (a)
(b)

6002.5 m
171.5 m/sec
31.3 m/sec
343 m/sec
39.6 m/sec
118.8 m/sec

2.2 The Derivative

Learning Objectives
A student will be able to:
Demonstrate an understanding of the derivative of a function as a slope of the tangent line.
Demonstrate an understanding of the derivative as an instantaneous rate of change.
Understand the relationship between continuity and dierentiability.
The function f (x) that we dened in the previous section is so important that it has its own name.
www.ck12.org

76

The Derivative
The function f is dened by the new function
f (x) = lim

h0

f (x + h) f (x)
h

where f is called the derivative of f with respect to x. The domain of f consists of all the values of
x for which the limit exists.
Based on the discussion in previous section, the derivative f represents the slope of the tangent line at
point x. Another way of interpreting it is to say that the function y = f (x) has a derivative f whose value
at x is the instantaneous rate of change of y with respect to point x.
Example 1:
Find the derivative of f (x) =

x
x+1 .

Solution:
We begin with the denition of the derivative,
f (x) = lim

h0

f (x + h) f (x)
1
= lim [ f (x + h) f (x)] ,
h0 h
h

where
x
x+1
x+h
f (x + h) =
x+h+1
f (x) =

Substituting into the derivative formula,

[
]
x+h
x
1

f (x) = lim
h0 h x + h + 1
x+1
]
[
1 (x + h)(x + 1) x(x + h + 1)
= lim
h0 h
(x + h + 1)(x + 1)
[ 2
]
1 x + x + hx + h x2 xh x
= lim
h0 h
(x + h + 1)(x + 1)
[
]
h
1
= lim
h0 h (x + h + 1)(x + 1)
1
= lim
h0 (x + h + 1)(x + 1)
1
=
.
(x + 1)2

Example 2:
Find the derivative of f (x) =

x and the equation of the tangent line at x0 = 1.

Solution:
Using the denition of the derivative,

77

www.ck12.org

f (x + h) f (x)
h0
h

x+h x
= lim
h0
h

x+h x x+h+ x
= lim

h0
h
x+h+ x
1 x+hx
= lim

h0 h
x+h+ x
1
= lim

h0
x+h+ x
1
= .
2 x

f (x) = lim

Thus the slope of the tangent line at x0 = 1 is

f (1) =

1
1
= .
2
2 1

For x0 = 1, we can nd y0 by simply substituting into f (x).

f (x0 ) y0

f (1) = 1 = 1
y0 = 1.
Thus the equation of the tangent line is
y y0 = m(x x0 )
1
y 1 = (x 1)
2
1
1
y= x+ .
2
2

Notation
Calculus, just like all branches of mathematics, is rich with notation. There are many ways to denote the
derivative of a function y = f (x) in addition to the most popular one, f (x). They are:
f (x)

dy
dx

df
dx

d f (x)
dx

In addition, when substituting the point x0 into the derivative we denote the substitution by one of the
following notations:

f (x0 )
www.ck12.org


d f 
 x x0
dx 

dy
| x x0
dx

78

d f (x0 )
dx

Existence and Dierentiability of a Function

If, at the point (x0 , f (x0 )), the limit of the slope of the secant line does not exist, then the derivative of
the function f (x) at this point does not exist either. That is,
if

m sec = lim

xx0

f (x) f (x0 )
= Does not exist
x x0

then the derivative f (x) also fails to exist as x x0 . The following examples show four cases where the
derivative fails to exist.

1. At a corner. For example f (x) = |x|, where the derivative on both sides of x = 0 dier (Figure 4).
2. At a cusp. For example f (x) = x2/3 , where the slopes of the secant lines approach + on the right
and on the left (Figure 5).
3. A vertical tangent. For example f (x) = x1/3 , where the slopes of the secant lines approach + on
the right and on the left (Figure 6).
4. A jump discontinuity. For example, the step function (Figure 7)

2,
f (x) =

2,

x<0
x0

where the limit from the left is 2 and the limit from the right is 2.

79

www.ck12.org

Many functions in mathematics do not have corners, cusps, vertical tangents, or jump discontinuities. We
call them dierentiable functions.
www.ck12.org

80

From what we have learned already about dierentiability, it will not be diicult to show that continuity is
an important condition for dierentiability. The following theorem is one of the most important theorems
in calculus:
Dierentiability and Continuity
If f is dierentiable at x0 , then f is also continuous at x0 .
The logically equivalent statement is quite useful: If f is not continuous at x0 , then f is not dierentiable
at x0 .
(The converse is not necessarily true.)
We have already seen that the converse is not true in some cases. The function can have a cusp, a corner,
or a vertical tangent and still be continuous, but it is not dierentiable.

Multimedia Links
For an introduction to the derivative (4.0)(4.1), see Math Video Tutorials by James Sousa, Introduction
to the Derivative (9:57) .

Figure 2.3: Introduction to the Derivative (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/502
The following simulator traces the instantaneous slope of a curve and graphs a qualitative form of derivative
function on an axis below the curve Surng the Derivative.
The following applet allows you to explore the relationship between a function and its derivative on a
graph. Notice that as you move x along the curve, the slope of the tangent line to f (x) is the height of the
derivative function, f (x) Derivative Applet. This applet is customizableafter doing the steps outlined on
the page, feel free to change the function denition and explore the derivative of many functions.
For a video presentation of dierentiability and continuity (4.3), see Dierentiability and Continuity (6:31)

Figure 2.4: Calculus 3.06a - Dierentiability and Continuity (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/503
.

81

www.ck12.org

Review Questions
In problems 16, use the denition of the derivative to nd f (x) and then nd the equation of the tangent
line at x = x0 .
1.
2.
3.
4.
5.
6.
7.
8.
9.

2
f (x) = 6x
; x0 = 3
f (x) = x + 2; x0 = 8
f (x) = 3x3 2; x0 = 1
1
f (x) = x+2
; x0 = 1
f (x) = ax2 b, (where a and b are constants); x0 = b
f (x) = x1/3 ; x0 = 1.
Find dy/dx| x=1 given that y = 5x2 2.

Show that f (x) = 3 x is continuous at x = 0 but

it is not dierentiable at x = 0. Sketch the graph.

Show that

x2 + 1 x 1
f (x) =

2x
x>1

is continuous and dierentiable at x = 1. Sketch the graph of f .

10. Suppose that f is a dierentiable function
property
f (x + y)and
= fhas
(x) the
+ f (y)
+ 3xy that
and
lim

h0

Find f (0) and

f (h)
= 4.
h

f (x).

Review Answers
1. f (x) = 12x, y = 36x (54
)
2. f (x) = 2 1x+2 , y = 1 12 x + 6
6

3.
4.
5.
6.
7.
8.
9.
10.

f (x) = 9x2 , y = 9x + 4
1
f (x) = (x+2)
2 , y = x

f (x) = 2ax, y = 2abx b(ab + 1)

f (x) = 3x12/3 , y = 13 x + 23
10
Hint: Take the limit from both sides.
Hint: Take the limit from both sides.
f (0) = 0, f (x) = 4 + 3x

2.3 Techniques of Dierentiation

Learning Objectives
A student will be able to:
Use various techniques of dierentiations to nd the derivatives of various functions.
Compute derivatives of higher orders.
Up to now, we have been calculating derivatives by using the denition. In this section, we will develop formulas and theorems that will calculate derivatives in more eicient and quick ways. It is highly
recommended that you become very familiar with all of these techniques.
www.ck12.org

82

The Derivative of a Constant

If f (x) = c where c is a constant, then f (x) = 0.
In other words, the derivative or slope of any constant function is zero.
Proof:
f (x) = lim

h0

f (x + h) f (x)
cc
= lim
=0
h0 h
h

Example 1:
If f (x) = 16 for all x, then f (x) = 0 for all x. We can also write d/dx(16) = 0.
The Power Rule
If n is a positive integer, then for all real values of x
d n
[x ] = nxn1
dx
The proof of the power rule is omitted in this text, but it is available at http://en.wikipedia.org/wiki/
Calculus_with_polynomials and also in video form at Khan Academy Proof of the Power Rule . Note

Figure 2.5: Proof: d/dx(xn) (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/504
that this proof depends on using the binomial theorem from precalculus.
Example 2:
If f (x) = x3 , then
f (x) = 3x31 = 3x2
and
d
[x] = 1 x11 = x0 = 1,
dx
d
d 1/2
1
1
1
1
[ x] =
[x ] = x1/21 = x1/2 = 1/2 = ,
dx
dx
2
2
2
x
2x
[
]
d 1
d 3
3
=
[x ] = 3x31 = 3x4 = 4 ,
3
dx x
dx
x
The Power Rule and a Constant
If c is a constant and f is dierentiable at all x, then
d
d
[c f (x)] = c [ f (x)].
dx
dx

83

www.ck12.org

In simpler notation,
(c f ) = c( f ) c f .
In other words, the derivative of a constant times a function is equal to the constant times the derivative
of the function.
Example 3:
d
d
[4x3 ] = 4 [x3 ] = 4[3x2 ] = 12x2 .
dx
dx
Example 4:
d [ 2 ]
d
d
8
=
[2x4 ] = 2 [x4 ] = 2[4x41 ] = 2[4x5 ] = 8x5 = 5 .
4
dx x
dx
dx
x
Derivatives of Sums and Dierences
If f and g are two dierentiable functions at x, then
d
d
d
[ f (x) + g(x)] =
[ f (x)] + [g(x)]
dx
dx
dx
and
d
d
d
[ f (x) g(x)] =
[ f (x)] [g(x)].
dx
dx
dx
In simpler notation,
( f g) = f g .
Example 5:
d
d
d
[3x2 + 2x] =
[3x2 ] + [2x]
dx
dx
dx
d 2
d
= 3 [x ] + 2 [x]
dx
dx
= 3[2x] + 2[1]
= 6x + 2.
Example 6:
d 3
d 3
d
[x 5x2 ] =
[x ] 5 [x2 ]
dx
dx
dx
2
= 3x 5[2x]
= 3x2 10x.
The Product Rule
If f and g are dierentiable at x, then
d
d
d
[ f (x) g(x)] = f (x) g(x) + g(x) f (x).
dx
dx
dx
www.ck12.org

84

In a simpler notation,
( f g) = f g + g f .
The derivative of the product of two functions is equal to the rst times the derivative of the second plus
the second times the derivative of the rst.
Keep in mind that
( f g) , f g .
Example 7:
Find

dy
dx

for y = (3x4 + 2)(7x3 1).

Solution:
There are two methods to solve this problem. One is to multiply the product and then use the derivative
of the sum rule. The second is to directly use the product rule. Either rule will produce the same answer.
We begin with the sum rule.
y = (3x4 + 2)(7x3 1)
= 21x7 3x4 + 14x3 2.
Taking the derivative of the sum yields
dy
= 147x6 12x3 + 42x2 + 0
dx
= 147x6 12x3 + 42x2 .
Now we use the product rule,
dy
= (3x4 + 2) (7x3 1) + (3x4 + 2) (7x3 1)
dx
= (3x4 + 2)(21x2 ) + (12x3 )(7x3 1)
= (63x6 + 42x2 ) + (84x6 12x3 )
= 147x6 12x3 + 42x2 ,
which is the same answer.
The Quotient Rule
If f and g are dierentiable functions at x and g(x) , 0, then
[
]
d
d
g(x) dx
[ f (x)] f (x) dx
[g(x)]
d f (x)
=
.
2
dx g(x)
[g(x)]
In simpler notation,
( )
f
g f f g
=
.
g
g2
The derivative of a quotient of two functions is the bottom times the derivative of the top minus the top
times the derivative of the bottom all over the bottom squared.

85

www.ck12.org

Keep in mind that the order of operations is important (because of the minus sign in the numerator) and
( )
f
f
, .
g
g
Example 8:
Find dy/dx for
y=
Solution:

x2 5
x3 + 2

[
]
dy
d x2 5
=
dx
dx x3 + 2
(x3 + 2) (x2 5) (x2 5) (x3 + 2)
=
(x3 + 2)2
3
(x + 2)(2x) (x2 5)(3x2 )
=
(x3 + 2)2
2x4 + 4x 3x4 + 15x2
=
(x3 + 2)2
x4 + 15x2 + 4x
=
(x3 + 2)2
x(x3 + 15x + 4)
=
.
(x3 + 2)2

Example 9:
At which point(s) does the graph of y =

x
x2 +9

have a horizontal tangent line?

Solution:
Since the slope of a horizontal line is zero, and since the derivative of a function signies the slope of the
tangent line, then taking the derivative and equating it to zero will enable us to nd the points at which
the slope of the tangent line equals to zero, i.e., the locations of the horizontal tangents.
x
,
+9
(x2 + 9)(1) x(2x)
y =
= 0.
(x2 + 9)2
y=

x2

Multiplying by the denominator and solving for x,

x2 + 9 2x2 = 0
x2 = 9
x = 3.
Therefore the tangent line is horizontal at x = 3, +3.
Higher Derivatives
If the derivative f of the function f is dierentiable, then the derivative of f , denoted by f , is
called the second derivative of f . We can continue the process of dierentiating derivatives and obtain
third, fourth, fth and higher derivatives of f . They are denoted by f , f , f , f (4) , f (5) , . . .
www.ck12.org

86

Example 10:
Find the fth derivative of f (x) = 2x4 3x3 + 5x2 x 1.
Solution:
f (x) = 8x3 9x2 + 5x x
f (x) = 24x2 18x + 5
f (x) = 48x 18
f (4) (x) = 48
f (5) (x) = 0
Example 11:
Show that y = x3 + 3x + 2 satises the dierential equation y + xy 2y = 0.
Solution:
We need to obtain the rst, second, and third derivatives and substitute them into the dierential equation.
y = x3 + 3x + 2
y = 3x2 + 3
y = 6x
y = 6.
Substituting,
y + xy 2y = 6 + x(6x) 2(3x2 + 3)
= 6 + 6x2 6x2 6
=0
which satises the equation.

Review Questions
Use the results of this section to nd the derivatives dy/dx.
1. y = 5x7
2. y = 12 (x3 2x2 + 1)

3. y = 2x3 1 x2 + 2x + 2
2

4.
5.
6.
7.
8.

y = a2 b2 + x2 a b + x (where a, b are constants)

y = x3 + x17
y = (x3 3x2 + x)(2x3 + 7x4 )
y = ( 1x + x12 )(3x4 7)

y = x + 1x

9. y =
10. y =

3
x+3
4x+1
2
x 9

87

www.ck12.org

11. Newtons Law of Universal Gravitation states that the gravitational force between two masses (say,
the earth and the moon), m and M, is equal to their product divided by the square of the distance r
between them. Mathematically,
mM
F =G 2 ,
r
(
2)
where G is the Universal Gravitational Constant 1.602 1011 Nm
. If the distance r between the
2
kg
two masses is changing, nd a formula for the instantaneous rate of change of F with respect to the
separation distance r.
[
]
12. Find
d 0 + 3
d 3 0
where 0 is a constant.
d3 y 
13. Find dx
, where y =
3
x=1

2
x3

Review Answers
(some answers simplify further than the given responses)
1.
2.
3.
4.
5.
6.
7.
8.

y
y
y
y
y
y
y
y

= 35x6
= 32 x2 2x

= 3 2x2 2x + 2
= 2x + 1
= 3
x78
x4
3
= (x 3x2 + x)(6x2 + 28x3 ) + (3x2 6x + 1)(2x3 + 7x4 )
= ( 1x + x12 )(12x3 ) + ( x12 x23 )(3x4 7)
= 2 1 x 2x13/2

9. y =
10.
11.

3
2 x( x+3)2
2
y = 4x(x22x36
9)2
dF
mM
dr = 2G r3
0 +32
30

12.
13. 120

2.4 Derivatives of Trigonometric Functions

Learning Objectives
A student will be able to:
Compute the derivatives of various trigonometric functions.
Recall from Chapter 1 that if the angle h is measured in radians,
limh0

sin h
h

= 1 and limh0

1cos h
h

= 0.

We now want to nd an expression for the derivative of the six trigonometric functions sin x, cos x, tan x, sec x, csc x,
and cot x. We rst consider the problem of dierentiating sin x, using the denition of the derivative.
www.ck12.org

88

sin(x + h) sin x
d
[sin x] = lim
h0
dx
h
Since
sin( + ) = sin cos + cos sin .
The derivative becomes
sin x cos h + cos x sin h sin x
d
[sin x] = lim
h0
dx
h
[
(
)
(
)]
cos h 1
sin h
= lim sin x
+ cos x
h0
h
h
(
)
)
(
1 cos h
sin h
= sin x lim
+ cos x lim
h0
h0
h
h
= sin x (0) + cos x (1)
= cos x.
Therefore,
d
[sin x] = cos x.
dx
It will be left as an exercise to prove that
d
[cos x] = sin x.
dx
The derivatives of the remaining trigonometric functions are shown in the table below.
Derivatives of Trigonometric Functions
d
[sin x] = cos x
dx
d
[cos x] = sin x
dx
d
[tan x] = sec2 x
dx
d
[sec x] = sec x tan x
dx
d
[csc x] = csc x cot x
dx
d
[sin x] = csc2 x
dx
Keep in mind that for all the derivative formulas for the trigonometric functions, the argument x is measured
in radians.
Example 1:
Show that

d
dx [tan x]

= sec2 x.

Solution:

89

www.ck12.org

It is possible to prove this relation by the denition of the derivative. However, we use a simpler method.
Since
tan x =

sin x
,
cos x

then
[
]
d
d sin x
[tan x] =
dx
dx cos x
Using the quotient rule,
(cos x)(cos x) (sin x)(sin x)
cos2 x
2
2
cos x + sin x
=
cos2 x
1
=
cos2 x
= sec2 x

Example 2:
Find f (x) if f (x) = x2 cos x + sin x.
Solution:
Using the product rule and the formulas above, we obtain
f (x) = x2 (sin x) + 2x cos x + cos x
= x2 sin x + 2x cos x + cos x.
Example 3:
Find dy/dx if y =

cos x
1tan x

. What is the slope of the tangent line at x = /3?

Solution:
Using the quotient rule and the formulas above, we obtain
(1 tan x)(sin x) (cos x)(sec2 x)
dy
=
dx
(1 tan x)2
sin x + tan x sin x + cos x sec2 x
=
(1 tan x)2
To calculate the slope of the tangent line, we simply substitute x = /3:

sin( 3 ) + tan( 3 ) sin( 3 ) + cos( 3 ) sec2 ( 3 )
dy 

=
.
dx 
(1 tan( 3 ))2
x= 3

We nally get the slope to be approximately


dy 

dx 
www.ck12.org

= 4.9.
x= 3

90

Example 4:
If y = sec x, nd y (/3).
Solution:
y = sec x tan x
y = sec x(sec2 x) + (sec x tan x) tan x
= sec3 x + sec x tan2 x.
Substituting for x = /3,

y = sec

()

()

+ sec
tan
3
3

= (2)3 + (2)( 3)2

()
3

= 8 + (2)(3)
= 14.
Thus y (/3) = 14.

Multimedia Links
For examples of nding the derivatives of trigonometric functions (4.4), see Math Video Tutorials by James
Sousa, The Derivative of Sine and Cosine (9:20) .

http://mathispower4u.wordpress.com/
http://mathispower4u.wordpress.com/
Figure 2.6: The Derivative of Sine and Cosine (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/505

Review Questions
Find the derivative y of the following functions:
1.
2.
3.
4.
5.

y = x sin x + 2
y = x2 cos x x tan x 1
y = sin2 x
sin x1
y = sin
x+1
x+sin x
y = cos
cos xsin x

6. y =

x
tan x

+2

91

www.ck12.org

7.
8.
9.
10.

y = csc x sin x + x
x
y = sec
csc x
If y = csc x, nd y (/6).
Use the denition of the derivative to prove that

d
dx [cos x]

= sin x.

Review Answers
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.

y
y
y
y

= x cos x + sin x
= 2x cos x x2 sin x tan x x sec2 x
= 2 cos x sin x
2 cos x
= (sin
x+1)2
(
)
x+1 2
y = 1 + tan
1tan
x

y = 2cot
x csc2 x
x
y = 1
y = sec2 x
y(/6) = 14
.

2.5 The Chain Rule

Learning Objectives
A student will be able to:
Know the chain rule and its proof.
Apply the chain rule to the calculation of the derivative of a variety of composite functions.
We want to derive a rule for the derivative of a composite function of the form f g in terms of the
derivatives of f and g. This rule allows us to dierentiate complicated functions in terms of known
derivatives of simpler functions.
The Chain Rule
If g is a dierentiable function at x and f is dierentiable at g(x), then the composition function f g =
f (g(x)) is dierentiable at x. The derivative of the composite function is:
( f g) (x) = f (g(x))g (x).
Another way of expressing, if u = u(x) and f = f (u), then
d
du
[ f (u)] = f (u) .
dx
dx
And a nal way of expressing the chain rule is the easiest form to remember: If y is a function of u and u
is a function of x, then
dy
dy du
=
. .
dx
du dx
Example 1:
www.ck12.org

92

Dierentiate f (x) = (2x3 4x2 + 5)2 .

Solution:
Using the chain rule, let u = 2x3 4x2 + 5. Then
d 2
d
[(2x2 4x2 + 5)2 ] =
[u ]
dx
dx
du
= 2u
dx
= 2(2x3 4x2 + 5)(6x2 8x).
The example above is one of the most common types of composite functions. It is a power function of the
type
y = [u(x)]n .
The rule for dierentiating such functions is called the General Power Rule. It is a special case of the
Chain Rule.
The General Power Rule
if
y = [u(x)]n
then
dy
= n[u(x)]n1 u (x).
dx
In simpler form, if
y = un
then
y = nun1 u .
Example 2:
What is the slope of the tangent line to the function y =

x2 3x + 2 that passes through point x = 3?

Solution:
We can write y = (x2 3x+2)1/2 . This example illustrates the point that n can be any real number including
fractions. Using the General Power Rule,
1
dy
1
= (x2 3x + 2) 2 1 (2x 3)
dx
2
1
= (x2 3x + 2)1/2 (2x 3)
2
(2x 3)
=
2 x2 3x + 2

To nd the slope of the tangent line, we simply substitute x = 3 into the derivative:

93

www.ck12.org


dy 

dx 

3
3 2
=
= =
.
4
2 2
2 32 3(3) + 2
2(3) 3

x=3

Example 3:
Find dy/dx for y = sin3 x.
Solution:
The function can be written as y = [sin x]3 . Thus
dy
= 3[sin x]2 [cos x]
dx
= 3 sin2 x cos x
Example 4:
Find dy/dx for y = 5 cos(3x2 1).
Solution:
Let u = 3x2 1. By the chain rule,
d
du
[ f (u)] = f (u)
dx
dx
where f (u) = 5 cos u. Thus
dy
= 5( sin u).(6x)
dx
= 5 sin u.(6x)
= 30x sin(3x2 1)
Example 5:
Find dy/dx for y = [cos(x2 )]3 .
Solution:
This example applies the chain rule twice because there are several functions embedded within each other.
Let u be the inner function and w be the innermost function.
y = (u(w))3
u(x) = cos x
w(x) = x2 .
Using the chain rule,
d
du
[ f (u)] = f (u)
dx
dx
d 3
d
[u ] =
[cos3 (x2 )]
dx
dx
d
=
[cos(x2 )]3
dx
= 3[cos(x2 )]2 [ sin(x2 )](2x)
= 6x[cos(x2 )]2 sin(x2 ).
Notice that we used the General Power Rule and, in the last step, we took the derivative of the argument.
www.ck12.org

94

Multimedia Links
For an introduction to the Chain Rule (5.0), see Khan Academy, Calculus: Derivatives 4: The Chain Rule
(9:11) .

Figure 2.7: The Chain Rule (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/506
For more examples of the Chain Rule (5.0), see [http://www.youtube.com/watch?v=PT11CSo-WkQ Math
Video Tutorials by James Sousa The Chain Rule: Part 1 of 2 (8:45) . and [http://www.youtube.com/watch?v=1x_-

Figure 2.8: The Chain Rule: Part 1 of 2 (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/507
wMYFOliQ Math Video Tutorials by James Sousa The Chain Rule: Part 2 of 2 (8:35) .

Figure 2.9: The Chain Rule: Part 2 of 2 (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/508

Review Questions
Find f (x).
1. f (x) = (2x2 3x)39
)3
(
2. f (x) = x3 x52

95

www.ck12.org

3. f (x) =
4.
5.
6.
7.
8.
9.
10.

1
3x2 6x+2
3

f (x) = sin x
f (x) = sin x
f (x) = sin3 x3
5)
f (x) = tan(4x

f (x) = 4x sin2 2x
sin x
f (x) = cos(3x2)
f (x) = (5x + 8)3 (x3 + 7x)13
(
)
x3 3
11. f (x) = 2x5

Review Answers
1. f (x) = 39(2x2 3x)38 (4x 3)
( 2 )4 ( 5 )
3x 10
2. f (x) = 3 x5x5
x3
3. f (x) =
4.
5.
6.
7.
8.

3(x1)

(3x2 6x+1)3
2

f (x) = 3 sin x cos x

f (x) = 3x2 cos x3
f (x) = 9x2 cos x3 sin2 x3
f (x) = 20x4 sec2 (4x5 )
2(1sin 2xcosx)
f (x) =
2
f (x)

9.
=

10. f (x) =
11. .

4xsin 2x
cos(3x2) cos x+3 sin(3x2) sin x
cos2 (3x2)
13(5x + 8)3 (x3 + 7x)12 (3x2

+ 7) + 15(x3 + 7x)13 (5x + 8)2

2.6 Implicit Dierentiation

Learning Objectives
A student will be able to:
Find the derivative of variety of functions by using the technique of implicit dierentiation.
Consider the equation
2xy = 1.
We want to obtain the derivative dy/dx. One way to do it is to rst solve for y,
y=

1
,
2x

and then project the derivative on both sides,

dy
d [1]
=
dx
dx 2x
1
= 2.
2x
www.ck12.org

96

There is another way of nding dy/dx. We can directly dierentiate both sides:
d
d
[2xy] =
[1].
dx
dx
Using the Product Rule on the left-hand side,
y

d
d
[2x] + 2x [y] = 0
dx
dx
dy
y[2] + 2x
= 0.
dx

Solving for dy/dx,

dy
2y
y
=
=
.
dx
2x
x
But since y =

1
2x

, substitution gives
dy
1
=
dx
x(2x)
1
= 2.
2x

which agrees with the previous calculations. This second method is called the implicit dierentiation
method. You may wonder and say that the rst method is easier and faster and there is no reason for the
second method. Thats probably true, but consider this function:
3y2 cos y = x3 .
How would you solve for y? That would be a diicult task. So the method of implicit dierentiation
sometimes is very useful, especially when it is inconvenient or impossible to solve for y in terms of x.
Explicitly dened functions may be written with a direct relationship between two variables with clear
independent and dependent variables. Implicitly dened functions or relations connect the variables in a
way that makes it impossible to separate the variables into a simple input output relationship. More notes
on explicit and implicit functions can be found at http://en.wikipedia.org/wiki/Implicit_function.
Example 1:
Find dy/dx if 3y2 cos y = x3 .
Solution:
Dierentiating both sides with respect to x and then solving for dy/dx,
d
d 3
[3y2 cos y] =
[x ]
dx
dx
d
d
3 [y2 ] [cos y] = 3x2
dx
dx
dy
dy
3(2y ) ( sin y)
= 3x2
dx
dx
dy
dy
6y + sin y
= 3x2
dx
dx
dy
[6y + sin y]
= 3x2 .
dx

97

www.ck12.org

Solving for dy/dx, we nally obtain

dy
3x2
=
.
dx
6y + sin y
Implicit dierentiation can be used to calculate the slope of the tangent line as the example below shows.
Example 2:
Find the equation of the tangent line that passes through point (1, 2) to the graph of 8y3 + x2 y x = 3.
Solution:
First we need to use implicit dierentiation to nd dy/dx and then substitute the point (1, 2) into the
derivative to nd slope. Then we will use the equation of the line (either the slope-intercept form or the
point-intercept form) to nd the equation of the tangent line. Using implicit dierentiation,
d
d
[8y3 + x2 y x] =
[3]
dx
dx
dy
dy
24y2
+ [(x2 )(1) + y(2x)] 1 = 0
dx
dx
dy
dy
24y2
+ x2
+ 2xy 1 = 0
dx
dx
dy
[24y2 + x2 ]
= 1 2xy
dx
dy
1 2xy
=
.
dx
24y2 + x2
Now, substituting point (1, 2) into the derivative to nd the slope,
1 2(1)(2)
dy
=
dx
24(2)2 + (1)2
3
=
.
97
So the slope of the tangent line is 3/97, which is a very small value. (What does this tell us about the
orientation of the tangent line?)
Next we need to nd the equation of the tangent line. The slope-intercept form is
y = mx + b,
where m = 3/97 and b is the yintercept. To nd it, simply substitute point (1, 2) into the line equation
and solve for b to nd the yintercept.
( 3 )
(1) + b
2=
97
197
b=
.
97
Thus the equation of the tangent line is
y=

3
197
x+
.
97
97

Remark: we could have used the point-slope form y y1 = m(x x1 ) and obtained the same equation.
www.ck12.org

98

Example 3:
Use implicit dierentiation to nd d2 y/dx2 if 5x2 4y2 = 9. Also nd

d2 y 
.
dx2 (x, y)=(2, 3)

What does the second

derivative represent?
Solution:
d
d
[5x2 4y2 ] =
[9]
dx
dx
dy
10x 8y
= 0.
dx
Solving for dy/dx,
dy
5x
= .
dx
4y
Dierentiating both sides implicitly again (and using the quotient rule),
(4y)(5) (5x)(4dy/dx)
d2 y
=
2
dx
(4y)2
20y
20x dy
=

2
16y
16y2 dx
5x dy
5

.
=
4y 4y dx
But since dy/dx = 5x/4y, we substitute it into the second derivative:
d2 y
5x 5x
5
.
=
2
4y 4y 4y
dx
d2 y
5
25x2
=

.
4y 16y2
dx2
This is the second derivative of y.

d2 y 
The next step is to nd: dx
2

(x, y)=(2, 3)


d2 y 

dx2 

(2, 3)

25(2)2
5

4(2) 16(3)2

5
.
72

Since the rst derivative of a function represents the rate of change of the function y = f (x) with respect to
x, the second derivative represents the rate of change of the rate of change of the function. For example, in
kinematics (the study of motion), the speed of an object (y ) signies the change of position with respect
to time but acceleration (y ) signies the rate of change of the speed with respect to time.

Multimedia Links
For more examples of implicit dierentiation (6.0), see Math Video Tutorials by James Sousa, Implicit
Dierentiation (8:10) .

99

www.ck12.org

Figure 2.10: Implicit Dierentiation (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/509

Figure 2.11: Related Rates Problem Using Implicit Dierentiation (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/510
For a video presentation of related rates using implicit dierentiation (6.0), see Just Math Tutoring,
Related Rates Using Implicit Dierentiation (9:56) .
For a presentation of related rates using cones (6.0), see Just Math Tutoring, Related Rates Using Implicit
Dierentiation (2:47) .

Figure 2.12: Related Rates using Cones (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/511

Review Questions
Find dy/dx by implicit dierentiation.
1. x2 + y2 = 500
2. x2 y + 3xy 2 = 1
3. 1x + 1y = 12

4. x y = 3
5. sin(25xy2 ) = x
6. tan3 (x2 y2 ) = tan(/4)
www.ck12.org

100

In problems #7 and 8, use implicit dierentiation to nd the slope of the tangent line to the given curve
at the specied point.
7.
8.
9.
10.

x2 y y2 x = 1 at (1, 1)
sin(xy) = y at (, 1)
Find y by implicit dierentiation for x3 y3 = 5.
Use implicit dierentiation to show that the tangent line to the curve y2 = kx at (x0 , y0 ) is given by
y0 y = 12 k(x + x0 ), where k is a constant.

Review Answers
1. y =
2. y =
3. y =
4. y =
5.
6.
7.
8.
9.
10.

y =
y =
1

x
y
y(2x+3)
x(x+3)
y2
x2

y
x
125y2 cos(25xy2 )
50xy cos(25xy2 )
x
y

1
1+
y = 2y
x2

2.7 Linearization and Newtons Method

Learning Objectives
A student will be able to:
Approximate a function by the method of linearization.
Know Newtons Method for approximating roots of a function.

Linearization: The Tangent Line Approximation

If f is a dierentiable function at x0 , then the tangent line, y = mx + b, to the curve y = f (x) at x0 is a
good approximation to the curve y = f (x) for values of x near x0 (Figure 8a). If you zoom in on the
two graphs, y = f (x) and the tangent line, at the point of tangency, (x0 , f (x0 )), or if you look at a table of
values near the point of tangency, you will notice that the values are very close (Figure 8b).
Since the tangent line passes through point (x0 , f (x0 )) and the slope is f (x0 ), we can write the equation
of the tangent line, in point-slope form, as
y y0 = m(x x0 )
y f (x0 ) = f (x0 )(x x0 )
Solving for y,

101

www.ck12.org

y = f (x0 ) + f (x0 )(x x0 )

So for values of x close to x0 , the values of y of this tangent line will closely approximate f (x). This gives
the approximation
f (x) = f (x0 ) + f (x0 )(x x0 ).
The Tangent Line Approximation (Linearization)
If f is a dierentiable function at x = x0 , then the approximation function
L(x) = f (x) f (x0 ) + f (x0 )(x x0 )
is a linearization of f near x0 .
Example 1:
www.ck12.org

102

Find the linearization of f (x) =

x + 3 at point x = 1.

Solution:
Taking the derivative of f (x),
1
f (x) = (x + 3)1/2 ,
2

we have f (1) =

4 = 2, f (1) = 1/4, and

f (x) f (x0 ) + f (x0 )(x x0 )
1
2 + (x 1)
4
1
7
x+ .
4
4

This tells us that near the point x = 1, the function f (x) =

As we move away from x = 1, we lose accuracy (Figure 9).

x + 3 approximates the line y = (x/4) + 7/4.

Example 2:
Find the linearization of y = sin x at x = /3.
Solution:
Since f (/3) = sin(/3) =

3/2, and f (x) = cos x, f (/3) = cos(/3) = 1/2, we have

3 1(
)
f (x)
+
x
2
2
3

3 x

+
2
2 6
x
+ 0.343.
2

103

www.ck12.org

Newtons Method
When faced with a mathematical problem that cannot be solved with simple algebraic means, such as
nding the roots of the polynomial x3 2x + 3 = 0, calculus sometimes provides a way of nding the
approximate solutions.

Lets say we are interested in computing 5 without using a calculator or a table. To do so, think about
this problem in a dierent way. Assume that we are interested in solving the quadratic equation
f (x) = x2 5 = 0

which leads to the roots x = 5.

The idea here is to nd the linearization of the above function, which is a straight-line equation, and then
solve the linear equation for x.
Since

4<

5<

or
2<

5 < 3,

We choose the linear approximation of f (x) to be near x0 = 2. Since f (x) = x2 5, f (x) = 2x and thus
f (2) = 1 and f (2) = 4. Using the linear approximation formula,
f (x) f (x0 ) + f (x0 )(x x0 )
1 + (4)(x 2)
1 + 4x 8
4x 9.
Notice that this equation is much easier to solve than f (x) = x2 5. Setting f (x) = 0 and solving for x, we
obtain,
4x 9 = 0
9
x=
4
= 2.25.
www.ck12.org

104

If you use a calculator, you will get x = 2.236 . . . As you can see, this is a fairly good approximation. To
be sure, calculate the percent dierence [% diff] between the actual value and the approximate value,

% di =

2|A X|
100%,
|A + X|

where A is the accepted value and X is the calculated value.

2|2.236 2.25|
100%
|2.236 + 2.25|
= 0.62%,

% di =

which is less than 1%.

We can actually make our approximation even better by repeating whatwe have just done not for x = 2, but
for x1 = 2.25 = 49 , a number that is even closer to the actual value of 5. Using the linear approximation
again,
f (x) f (x1 ) + f (x1 )(x x1 )
1
9(
9)

+
x
16 2
4
161
9
.
x
2
16
Solving for x by setting f (x) = 0, we obtain

x = x2 = 2.236111,

which is even a better

approximation than x1 = 9/4. We could continue this process generating a better
approximation to 5. This is the basic idea of Newtons Method.
Here is a summary of Newtons method.
Newtons Method

1. Guess the rst approximation to a solution of the equation f (x) = 0. A graph would be very helpful
in nding the rst approximation (see Figure below).
2. Use the rst approximation to nd the second, the second to nd the third and so on by using the
recursion relation

xn+1 = xn

105

f (xn )
.
f (xn )
www.ck12.org

Example 3:
Use Newtons method to nd the roots of the polynomial f (x) = x3 + x 1.
Solution:
f (x) = x3 + x 1
f (x) = 3x2 + 1.
Using the recursion relation,
f (xn )
f (xn )
x 3 + xn 1
= xn n 2
.
3xn + 1

xn+1 = xn

To help us nd the rst approximation, we make a graph of f (x). As Figure 11 suggests, set x1 = 0.6.
Then using the recursion relation, we can generate x2 , x3 , . . ..

www.ck12.org

106

xn3 + xn 1
3xn2 + 1
(0.6)3 + (0.6) 1
x2 = 0.6
3(0.6)2 + 1
= 0.6884615.

xn+1 = xn

Using the recursion relation again to nd x3 , we get

x3 = 0.6836403.
We conclude that the solution to the equation x3 + x 1 = 0 is about 0.6836403.

Multimedia Links
For a video presentation of Newtons method (10.0), see Math Video Tutorials by James Sousa, Newtons
method (9:48) .

Figure 2.13: Newtons Method (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/512

Review Questions
1. Find the linearization of
f (x) =

x2 + 1
x

at a = 1.
2. Find the linearization of f (x) = tan x at a = .
3. Use the linearization method to show that when x 1 (much less than 1), then (1 + x)n 1 + nx.
4. Use the result of problem #3, (1 + x)n 1 + nx , to nd the approximation for the following:
4
(a) f (x) = (1
x)
(b) f (x) = 1 x
5
(c) f (x) = 1+x
(
)2
3
3
(d) f (x) =
1 (x1)

(e) Without using a calculator, approximate (1.003)99 .

5. Use Newtons Method to nd the roots of x3 + 3 =
0.
6. Use Newtons Method to nd the roots of x + 3 1 + x = 0.

107

www.ck12.org

Review Answers
1.
2.
3.
4.
5.

f (x) 2
f (x) x
Hint: Let x = 0.

(a) 1 4x
(b) 1 12 x
(c) 5 + 5x
(d) 1 + 2x
(e) 1.297
6. x 1.442
7. x 1.146 and x 7.854

Texas Instruments Resources

In the CK-12 Texas Instruments Calculus FlexBook, there are graphing calculator activities
designed to supplement the objectives for some of the lessons in this chapter. See http:
//www.ck12.org/flexr/chapter/9727.

www.ck12.org

108

Chapter 3
Applications of Derivatives
3.1 Related Rates
Learning Objectives
A student will be able to:
Solve problems that involve related rates.

Introduction
In this lesson we will discuss how to solve problems that involve related rates. Related rate problems
involve equations where there is some relationship between two or more derivatives. We solved examples
of such equations when we studied implicit dierentiation in Lesson 2.6. In this lesson we will discuss some
real-life applications of these equations and illustrate the strategies one uses for solving such problems.
Lets start our discussion with some familiar geometric relationships.
Example 1: Pythagorean Theorem
x 2 + y2 = z2

109

www.ck12.org

We could easily attach some real-life situation to this geometric gure. Say for instance that x and y
represent the paths of two people starting at point p and walking North and West, respectively, for two
hours. The quantity z represents the distance between them at any time t. Lets now see some relationships
between the various rates of change that we get by implicitly dierentiating the original equation x2 +y2 = z2
with respect to time t.
x2 + y2 = z2
dx
dy
dz
2x + 2y = 2z .
dt
dt
dt
Simplifying, we have
dy
dz
Equation 1. x dx
dt + y dt = z dt

So we have relationships between the derivatives, and since the derivatives are rates, this is an example of
related rates. Lets say that person x is walking at 5 mph and that person y is walking at 3 mph. The rate
at which the distance between the two walkers is changing at any time is dependent on the rates at which
the two people are walking. Can you think of any problems you could pose based on this information?
One problem that we could pose is at what rate is the distance between x and y increasing after one hour.
That is, nd dz/dt.
Solution:
Assume that they have walked for one hour. So x = 5mi and y = 3. Using the Pythagorean Theorem, we
nd the distance between them after one hour is z = 34 = 5.83 miles.

If we substitute these values into Equation 1 along with the individual rates we get

dz
dt
dz
34 = 34
dt
34
dz
= .
dt
34

5(5) + 3(3) =

34

Hence after one hour the distance between the two people is increasing at a rate of

dz
dt

34
34

5.83 mph.

Our second example lists various formulas that are found in geometry.
As with the Pythagorean Theorem, we know of other formulas that relate various quantities associated
with geometric shapes. These present opportunities to pose and solve some interesting problems
www.ck12.org

110

Example 2: Perimeter and Area of a Rectangle

We are familiar with the formulas for Perimeter and Area:
P = 2 l + 2 w,
A = l w.
Suppose we know that at an instant of time, the length is changing at the rate of 8 ft/hour and the
perimeter is changing at a rate of 24 ft/hour. At what rate is the width changing at that instant?
Solution:
If we dierentiate the original equation, we have
Equation 2:

dp
dt

=2

dl
dt

+2

dw
dt .

Substituting our known information into Equation II, we have

24 = (2 8) + 2
8=2
4=

dw
dt

dw
dt

dw
.
dt

The width is changing at a rate of 4 ft/hour.

Okay, rather than providing a related rates problem involving the area of a rectangle, we will leave it to
you to make up and solve such a problem as part of the homework (HW #1).
Lets look at one more geometric measurement formula.
Example 3: Volume of a Right Circular Cone
1
V = r2 h
3

We have a water tank shaped as an inverted right circular cone. Suppose that water ows into the tank
at the rate of 5 ft3 /min. At what rate is the water level rising when the height of the water in the tank is
6 feet?
Solution:
We rst note that this problem presents some challenges that the other examples did not.

111

www.ck12.org

When we dierentiate the original equation, V = (1/3)r2 h, we get

dV
1
dr 1
dh
= (h)(2r) + r2 .
dt
3
dt
3
dt
The diiculty here is that we have no information about the radius when the water level is at 6 feet. So we
need to relate the radius a quantity that we do know something about. Starting with the original equation,
lets nd a relationship between h and r. Let r1 be the radius of the surface of the water as it ows out of
the tank.

Note that the two triangles are similar and thus corresponding parts are proportional. In particular,
r1
8
=
h
20
8h
2h
r1 =
= .
20
5
Now we can solve the problem by substituting r1 = (2h/5) into the original equation:
( )2
1 2h
4 3
V=
h=
h .
3
5
75
Hence

dV
dt

12 2 dh
75 h dt ,

and by substitution,
12
dh
(36)
75
dt
dh
375
ft
=
0.28
.
dt
432
min
5=

Lesson Summary
1. We learned to solve problems that involved related rates.
www.ck12.org

112

Multimedia Links
For a video presentation of related rates (12.0), see Math Video Tutorials by James Sousa, Related Rates
(10:34) .

Figure 3.1: Related Rates (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/513
In the following applet you can explore a problem about a melting snowball where the radius is decreasing
at a constant rate. Calculus Applets Snowball Problem. Experiment with changing the time to see how
the volume does not change at a constant rate in this problem. If youd like to see a video of another
example of a related rate problem worked out (12.0), see Khan Academy Rates of Change (Part 2) (5:38)

Figure 3.2: Rates-of-change (part 2) (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/514
.

Review Questions
1.
2. (a) Make up a related rates problem about the area of a rectangle.
(b) Illustrate the solution to your problem.
3. Suppose that a particle is moving along the curve 4x2 + 16y2 = 32. When it reaches the point (2, 1),
the xcoordinate is increasing at a rate of 3 ft/sec. At what rate is the ycoordinate changing at
that instant?
4. A regulation softball diamond is a square with each side of length 60 ft. Suppose a player is running
from rst base to second base at a speed of 18 ft/sec. At what rate is the distance between the runner
and home plate changing when the runner is 2/3 of the way from rst to second base?
5. At a recent Hot Air Balloon festival, a hot air balloon was released. Upon reaching a height of 300 ft,
it was rising at a rate of 20 ft/sec. Mr. Smith was 100 ft away from the launch site watching the
balloon. At what rate was the distance between Mr. Smith and the balloon changing at that instant?
6. Two trains left the St. Louis train station in the late morning. The rst train was traveling East at a
constant speed of 65 mph. The second train traveled South at a constant speed of 75 mph. At 3 PM,

113

www.ck12.org

7.
8.

9.

10.

the rst train had traveled a distance of 120 miles while the second train had traveled a distance of
130 miles. How fast was the distance between the two trains changing at that time?
Suppose that a 17 ft ladder is sliding down a wall at a rate of 6 ft/sec. At what rate is the bottom
of the ladder moving when the top is 8 ft from the ground?
Suppose that the length of a rectangle is increasing at the rate of 6 ft/min and the width is increasing
at a rate of 2 ft/min. At what rate is the area of the rectangle changing when its length is 25 ft and
its width is 15 ft?
Suppose that the quantity demand of new 40 plasma TVs is related to its unit price by the formula
p + x2 = 1200, where p is measured in dollars and x is measured in units of one thousand. How is
the quantity demand changing when x = 20, p = 1500, and the price per TV is decreasing at a rate
of $10?
The volume of a cube with side s is changing. At a certain instant, the sides of the cube are 6 inches
and increasing at the rate of 1/4 in/min. How fast is the volume of the cube increasing at that time?

11.
12. (a) Suppose that the area of a circle is increasing at a rate of 24 in2 / min. How fast is the radius
increasing when the area is 36 in2 ?
(b) How fast is the circumference changing at that instant?

Review Answers

1. Answers will vary.

3 ft
2. dy
dt = 2 sec
3. Using the following diagram,

dy
dt

720 ft
5200 sec

4. Using the following diagram,

dy
dt

ft
6000
100000 sec

www.ck12.org

ft
10.18 sec
.

ft
18.97 sec
.

114

17550
31300

99.20 mph.

5. Using the following diagram,

ds
dt

6. Using the following diagram,

dx
dt

ft
= 16
5 sec .

115

www.ck12.org

ft
7. dA
dt = 140 min
8. The demand is increasing at a rate of 1/4 per thousand units, or 250 units per week.
in3
9. dV
dt = 27 min
10.
2 in
11. (a) dr
dt = min
dp
in
(b) dt = 4 min

3.2 Extrema and the Mean Value Theorem

Learning Objectives
A student will be able to:
Solve problems that involve extrema.
Study Rolles Theorem.
Use the Mean Value Theorem to solve problems.

Introduction
In this lesson we will discuss a second application of derivatives, as a means to study extreme (maximum
and minimum) values of functions. We will learn how the maximum and minimum values of functions
relate to derivatives.
Lets start our discussion with some formal working denitions of the maximum and minimum values of a
function.
Denition A function f has a maximum at x = a if f (a) f (x) for all x in the domain of f. Similarly,
f has a minimum at x = a if f (a) f (x) for all x in the domain of f. The values of the function for
these xvalues are called extreme values or extrema.
Here is an example of a function that has a maximum at x = a and a minimum at x = d:

Observe the graph at x = b. While we do not have a minimum at x = b, we note that f (b) f (x) for all x
near b. We say that the function has a local minimum at x = b. Similarly, we say that the function
has a local maximum at x = c since f (c) f (x) for some x contained in open intervals of c.
Lets recall the Min-Max Theorem that we discussed in lesson on Continuity.
Min-Max Theorem: If a function f (x) is continuous in a closed interval I, then f (x) has both a maximum
value and a minimum value in I. In order to understand the proof for the Min-Max Theorem conceptually,
attempt to draw a function on a closed interval (including the endpoints) so that no point is at the highest
part of the graph. No matter how the function is sketched, there will be at least one point that is highest.
www.ck12.org

116

We can now relate extreme values to derivatives in the following Theorem by the French mathematician
Fermat.
Theorem: If f (c) is an extreme value of f for some open interval of c, and if f (c) exists, then f (c) = 0.
Proof: The theorem states that if we have a local max or local min, and if f (c) exists, then we must have
f (c) = 0.
Suppose that f has a local max at x = c. Then we have f (c) f (x) for some open interval (c h, c + h)
with h > 0.
So f (c + h) f (c) 0.
Consider limh0+

f (c+h) f (c)
.
h

Since f (c + h) f (c) 0, we have limh0+

Since f (c) exists, we have f (c) = limh0

f (c+h) f (c)
h
f (c+h) f (c)
h

If we take the left-hand limit, we get f (c) = limh0

Hence

f (c)

0 and

f (c)

0 it must be that

f (c)

limh0+ 0 = 0.
= limh0+
f (c+h) f (c)
h

f (c+h) f (c)
,
h

= limh0

and so f (c) 0.

f (c+h) f (c)
h

0.

= 0.

If x = c is a local minimum, the same argument follows.

Denition We will call x = c a critical value in [a, b] if f (c) = 0 or f (c) does not exist, or if x = c is
an endpoint of the interval.
We can now state the Extreme Value Theorem.
Extreme Value Theorem: If a function f (x) is continuous in a closed interval [a, b], with the maximum
of f at x = c1 and the minimum of f at x = c2 , then c1 and c2 are critical values of f.
Proof: The proof follows from Fermats theorem and is left as an exercise for the student.
Example 1:
Lets observe that the converse of the last theorem is not necessarily true: If we consider f (x) = x3 and its
graph, then we see that while f (0) = 0 at x = 0, x = 0 is not an extreme point of the function.

Rolles Theorem: If f is continuous and dierentiable on a closed interval [a, b] and if f (a) = f (b), then
f has at least one value c in the open interval (a, b) such that f (c) = 0.
The proof of Rolles Theorem can be found at http://en.wikipedia.org/wiki/Rolle's_theorem.

117

www.ck12.org

Mean Value Theorem: If f is a continuous function on a closed interval [a, b] and if f contains the
open interval (a, b) in its domain, then there exists a number c in the interval (a, b) such that f (b) f (a) =
(b a) f (c).
Proof: Consider the graph of f and secant line s as indicated in the gure.

By the Point-Slope form of line s we have

y f (a) = m(x a) and y = m(x a) + f (a).
For each x in the interval (a, b), let g(x) be the vertical distance from line S to the graph of f. Then we
have
g(x) = f (x) [m(x a) + f (a)] for every x in (a, b).
Note that g(a) = g(b) = 0. Since g is continuous in [a, b] and g exists in (a, b), then Rolles Theorem
applies. Hence there exists c in (a, b) with g (c) = 0.
So g (x) = f (x) m for every x in (a, b).
In particular,
g (c) = f (c) m = 0 and
f (c) = m
f (b) f (a)
f (c) =
ba
f (b) f (a) = (b a) f (c).
The proof is complete.
Example 2:
Verify that the Mean Value Theorem applies for the function f (x) = x3 + 3x2 24x on the interval [1, 4].
Solution:
We need to nd c in the interval (1, 4) such that f (4) f (1) = (4 1) f (c).
Note that f (x) = 3x2 + 6x 24, and f (4) = 16, f (1) = 20. Hence, we must solve the following equation:
36 = 3 f (c)
12 = f (c).
By substitution, we have
12 = 3c2 + 6c 24
3c2 + 6c 36 = 0
c2 + 2c 12 = 0

2 52
c=
4.61, 2.61.
2

www.ck12.org

118

Since we need to have c in the interval (1, 4), the positive root is the solution, c =

2+ 52
2

2.61.

Lesson Summary
1. We learned to solve problems that involve extrema.
2. We learned about Rolles Theorem.
3. We used the Mean Value Theorem to solve problems.

Multimedia Links
For a video presentation of Rolles Theorem (8.0), see Math Video Tutorials by James Sousa, Rolles
Theorem (7:54) .

Figure 3.3: Rolles Theorem (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/515
For more information about the Mean Value Theorem (8.0), see Math Video Tutorials by James Sousa,
Mean Value Theorem (9:52) .

Figure 3.4: The Mean Value Theorem (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/516
For an introduction to LHopitals Rule (8.0), see Khan Academy, LHopitals Rule (8:51) .
For a well-done, but unorthodox, student presentation of the Extreme Value Theorem and Related Rates
(3.0)(12.0), see Extreme Value Theorem (10:00) .

Review Questions
In problems #13, identify the absolute and local minimum and maximum values of the function (if they
exist); nd the extrema. (Units on the axes indicate 1 unit).

119

www.ck12.org

Figure 3.5: Introduction to LHopitals Rule (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/517

Figure 3.6: Calc video (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/518

1. Continuous on [0, 9]

2. Continuous on [0, 9]
www.ck12.org

120

3. Continuous on [0, 4] (4, 9)

In problems #46, nd the extrema and sketch the graph.

4.
5.
6.
7.
8.

f (x) = x2 6x + 4, [4, 1]
f (x) = x3 x4 , [0, 2]
f (x) = x2 + x42 , [2, 0]
Verify Rolles Theorem by nding values of x for which f (x) = 0 and f (x) = 0. f (x) = 3x3 12x
2
Verify Rolles Theorem for f (x) = x2 x1
.
(x+2)

9. Verify that the Mean Value Theorem works for f (x) = x , [1, 2].
10. Prove that the equation x3 + a1 x2 + a2 x = 0 has a positive root at x = r, and that the equation
3x2 + 2a1 x + a2 = 0 has a positive root less than r.

Review Answers
1. Absolute max at x = 7, absolute minimum at x = 4, relative maximum at x = 2, Note: there is
no relative minimum at x = 9 because there is no open interval around x = 9 since the function is
dened only on x = 9; the extreme values of f are f (7) = 7, f (4) = 0.

121

www.ck12.org

2. Absolute maximum at x = 7, absolute minimum at x = 9, relative minimum at x = 3, Note: there is

no relative minimum at x = 0 because there is no open interval around x = 0 since the function is
dened only on [0, 9]; the extreme values of f are f (7) = 9, f (9) = 0.

3. Absolute minimum at x = 0, f (0) = 1; there is no maximum since the function is not continuous on
a closed interval.

4. Absolute maximum at x = 3, f (3) = 13, absolute minimum at x = 1, f (1) = 3

www.ck12.org

122

5. Absolute maximum at x = 34 , f ( 43 ) .1055, absolute minimum at x = 2, f (2) = 8.

6. Absolute minimum at x = 2, f ( 2) = 4.

7. f (x) = 0 at x = 0, 2 f (x) = 0, x = 2 3 3 ; (by Rolles Theorem, there is a critical value in each of

the intervals (2, 0) and (2, 0) and we found those to be x = 2.

8. f (x) = 2 at x = 1, 0 f (x) = 0 at x 0.84 by Rolles Theorem, there is a critical value in the
interval (1, 0) and we found it to be x 0.48.

123

www.ck12.org


9. Need to nd c (1, 2) such that f (2) f (1) = (2 1) f (c); c = 2.
10. Let f (x) = x3 + a1 x2 + a2 x. Observe that f (0) = f (r) = 0. By Rolles Theorem, there exist c (0, r)
such that f (0) = 0.

3.3 The First Derivative Test

Learning Objectives
A student will be able to:
Find intervals where a function is increasing and decreasing.
Apply the First Derivative Test to nd extrema and sketch graphs.

Introduction
In this lesson we will discuss increasing and decreasing properties of functions, and introduce a method
with which to study these phenomena, the First Derivative Test. This method will enable us to identify
precisely the intervals where a function is either increasing or decreasing, and also help us to sketch the
graph. Note on notation: The symbol and are equivalent and denote that a particular element is
contained within a particular set.
Denition A function f is said to be increasing on [a, b] contained in the domain of f if f (x1 ) f (x2 )
whenever x1 x2 for all x1 , x2 [a, b]. A function f is said to be decreasing on [a, b] contained in
the domain of f if f (x1 ) f (x2 ) whenever x1 x2 for all x1 , x2 [a, b].
If f (x1 ) < f (x2 ) whenever x1 < x2 for all x1 , x2 [a, b], then we say that f is strictly increasing on [a, b].
If f (x1 ) > f (x2 ) whenever x1 > x2 for all x1 , x2 [a, b], then we say that f is strictly decreasing on
[a, b].
We saw several examples in the Lesson on Extreme and the Mean Value Theorem of functions that had
these properties.
Example 1:
The function f (x) = x3 is strictly increasing on (, +):

www.ck12.org

124

Example 2:
The function indicated here is strictly increasing on (0, a) and (b, c), and strictly decreasing on (a, b) and
(c, d).

We can now state the theorems that relate derivatives of functions to the increasing/decreasing properties
of functions.
Theorem: If f is continuous on interval [a, b], then:
1. If f (x) > 0 for every x [a, b], then f is strictly increasing in [a, b].
2. If f (x) < 0 for every x [a, b], then f is strictly decreasing in [a, b].
Proof : We will prove the rst statement. A similar method can be used to prove the second statement
and is left as an exercise to the student.
Consider x1 , x2 [a, b] with x1 < x2 . By the Mean Value Theorem, there exists c (x1 , x2 ) such that
f (x2 ) f (x1 ) = (x2 x1 ) f (c).
By assumption, f (x) > 0 for every x [a, b]; hence f (c) > 0. Also, note that x2 x1 > 0.
Hence f (x2 ) f (x1 ) > 0 and f (x2 ) > f (x1 ).
We can observe the consequences of this theorem by observing the tangent lines of the following graph.
Note the tangent lines to the graph, one in each of the intervals (0, a), (a, b), (b, +).

Note rst that we have a relative maximum at x = a and a relative minimum at x = b. The slopes of the
tangent lines change from positive for x (0, a) to negative for x (a, b) and then back to positive for
x (b, +). From this we example infer the following theorem:
First Derivative Test
Suppose that f is a continuous function and that x = c is a critical value of f. Then:
1. If f changes from positive to negative at x = c, then f has a local maximum at x = c.
2. If f changes from negative to positive at x = c, then f has a local minimum at x = c.
3. If f does not change sign at x = c, then f has neither a local maximum nor minimum at x = c.

125

www.ck12.org

Proof of these three conclusions is left to the reader.

Example 3:
Our previous example showed a graph that had both a local maximum and minimum. Lets reconsider
f (x) = x3 and observe the graph around x = 0. What happens to the rst derivative near this value?

Example 4:
Lets consider the function f (x) = x2 + 6x 9 and observe the graph around x = 3. What happens to the
rst derivative near this value?

We observe that the slopes of the tangent lines to the graph change from negative to positive at x = 3.
The rst derivative test veries this fact. Note that the slopes of the tangent lines to the graph are negative
for x (, 3) and positive for x (3, ).

Lesson Summary
1. We found intervals where a function is increasing and decreasing.
2. We applied the First Derivative Test to nd extrema and sketch graphs.

Multimedia Links
For more examples on determining whether a function is increasing or decreasing (9.0), see Math Video Tutorials by James Sousa, Determining where a function is increasing and decreasing using the rst derivative
www.ck12.org

126

(10:05) .

Figure 3.7: Determining where a function is increasing and decreasing using the rst deriviative (Watch
Youtube Video)
http://www.ck12.org/exbook/embed/view/519
For a video presentation of increasing and decreasing trigonometric functions and relative extrema (9.0),
see Math Video Tutorials by James Sousa, Increasing and decreasing trig functions, relative extrema (6:02)

http://mathispower4u.wordpress.com/
http://mathispower4u.wordpress.com/
Figure 3.8: Increasing and Decreasing Trig Functions / Relative Extrema (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/520
.
For more information on nding relative extrema using the rst derivative (9.0), see Math Video Tutorials
by James Sousa, Finding relative extrema using the rst derivative (6:18) .

Figure 3.9: Finding relative extrema using the rst derivative (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/521

Review Questions
In problems #12, identify the intervals where the function is increasing, decreasing, or is constant. (Units
on the axes indicate single units).

127

www.ck12.org

1.

2.

3. Give the sign of the following quantities for the graph in #2.
(a)
(b)
(c)
(d)

f (3)
f (1)
f (3)
f (4)

For problems #46, determine the intervals in which the function is increasing and those in which it is
decreasing. Sketch the graph.

4. f (x) = x2 1x
5. f (x) = (x2 1)5
6. f (x) = (x2 1)4
(a) Use the First Derivative Test to nd the intervals where the function increases and/or decreases
(b) Identify all max, mins, or relative max and mins
(c) Sketch the graph

For problems #710, nd the following:

7.
8.
9.
10.

f (x) = x2 4x 1
f (x) = x3 + 3x2 9x + 1
2
f (x) = x 3 (x
5)
f (x) = 2x x2 + 1

Review Answers
1.
2.
3.
4.

Increasing on (0, 3), decreasing on (3, 6), constant on (6, +).

Increasing on (, 0) and (3, 7), decreasing on (0, 3).
f (3) > 0, f (1) < 0, f (3) = 0, f (4) > 0

3
3
3
Relative minimum at x = 0.5; increasing on ( 0.5, 0) and (0, +), decreasing on (, 0.5).

www.ck12.org

128

5. Absolute minimum at x = 0; decreasing on (, 0), increasing on (0, +).

6. Absolute minimum at x = 1; relative maximum at x = 0; decreasing on (, 1), (0, 1), increasing

on (1, 0), (1, +).

7. Absolute maximum at x = 2; increasing on (, 2), decreasing on (2, +).

129

www.ck12.org

8. Relative maximum at x = 3, f (3) = 28; relative minimum at x = 1, f (1) = 4; increasing on

(, 3) and (1, +), decreasing on (3, 1).

2
3
9. Relative maximum at x = 0, f (0) = 0; relative minimum at x = 2, f (2) = 3 2 3 = 3 4, x =
1, f (1) = 4; increasing on (, 0) and (2, +) decreasing on (0, 2).

www.ck12.org

130

10. There are no maximums or minimums; no relative maximums or minimums.

3.4 The Second Derivative Test

Learning Objectives
A student will be able to:
Find intervals where a function is concave upward or downward.
Apply the Second Derivative Test to determine concavity and sketch graphs.

Introduction
In this lesson we will discuss a property about the shapes of graphs called concavity, and introduce a
method with which to study this phenomenon, the Second Derivative Test. This method will enable us
to identify precisely the intervals where a function is either increasing or decreasing, and also help us to
sketch the graph.
Denition A function f is said to be concave upward on [a, b] contained in the domain of f if f is an
increasing function on [a, b] and concave downward on [a, b] if f is a decreasing function on [a, b].
Here is an example that illustrates these properties.
Example 1:
Consider the function f (x) = x3 x:

The function has zeros at x = 1, 0 and has a relative maximum at x =

at x =

3
3 .

3
3

and a relative minimum

Note that the graph appears to be concave down for all intervals in (, 0) and concave up

131

www.ck12.org

for all intervals in (0, +). Where do you think the concavity of the graph changed from concave down
to concave up? If you answered at x = 0 you would be correct. In general, we wish to identify both the
extrema of a function and also points, the graph changes concavity. The following denition provides a
formal characterization of such points.
Denition A point on a graph of a function f where the concavity changes is called an inection point.
The example above had only one inection point. But we can easily come up with examples of functions
where there are more than one point of inection.
Example 2:
Consider the function f (x) = x4 3x3 + x 2.

We can see that the graph has two relative minimums, one relative maximum, and two inection points
(as indicated by arrows).
In general we can use the following two tests for concavity and determining where we have relative maximums, minimums, and inection points.
Test for Concavity
Suppose that f is continuous on [a, b] and that I is some open interval in the domain of f.
1. If f (x) > 0 for all x I, then the graph of f is concave upward on I.
2. If f (x) < 0 for all x I, then the graph of f is concave downward on I.
A consequence of this concavity test is the following test to identify extreme values of f.
Second Derivative Test for Extrema
Suppose that f is a continuous function near c and that c is a critical value of f. Then
1. If f (c) > 0, then f has a relative maximum at x = c.
2. If f (c) < 0, then f has a relative minimum at x = c.
3. If f (c) = 0, then the test is inconclusive and x = c may be a point of inection.
Recall the graph f (x) = x3 . We observed that x = 0, and that there was neither a maximum nor minimum.
The Second Derivative Test cautions us that this may be the case since at f (0) = 0 at x = 0.
So now we wish to use all that we have learned from the First and Second Derivative Tests to sketch graphs
of functions. The following table provides a summary of the tests and can be a useful guide in sketching
graphs.

www.ck12.org

132

Table 3.1:
Signs of rst and second derivatives

Information from applying First

and Second Derivative Tests

f (x) > 0
f (x) > 0

f is increasing
f is concave upward

f (x) > 0
f (x) < 0

f is increasing
f is concave downward

f (x) < 0
f (x) > 0

f is decreasing
f is concave upward

f (x) < 0
f (x) < 0

f is decreasing
f is concave downward

Shape of the graphs

Lets look at an example where we can use both the First and Second Derivative Tests to nd out information that will enable us to sketch the graph.
Example 3:
Lets examine the function f (x) = x5 5x + 2.
1. Find the critical values for which f (c) = 0.
f (x) = 5x4 5 = 0, or
x4 1 = 0 at x = 1.
Note that f (x) = 20x3 = 0 when x = 0.
2. Apply the First and Second Derivative Tests to determine extrema and points of inection.
We can note the signs of f and f in the intervals partioned by x = 1, 0.
Table 3.2:
Key intervals

f (x)

f (x)

Shape of graph

x < 1

1 < < 0

0<<1
x>1

+
+

Increasing,
concave
down
Decreasing,
concave
down
Decreasing, concave up
Increasing, concave up

133

www.ck12.org

Also note that f (1) = 20 < 0. By the Second Derivative Test we have a relative maximum at x = 1,
or the point (1, 6).
In addition, f (1) = 20 > 0. By the Second Derivative Test we have a relative minimum at x = 1, or the
point (1, 2). Now we can sketch the graph.

Lesson Summary
1. We learned to identify intervals where a function is concave upward or downward.
2. We applied the First and Second Derivative Tests to determine concavity and sketch graphs.

Multimedia Links
For a video presentation of the second derivative test to determine relative extrema (9.0), see Math Video
Tutorials by James Sousa, Introduction to Limits (8:46) .

Figure 3.10: The second derivative test to determine relative extrema (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/522

Review Questions
1. Find all extrema using the Second Derivative Test. f (x) =
2. Consider f (x) = x2 + ax + b, with f (1) = 3.

x2
4

4
x

(a) Determine a and b so that x = 1 is a critical value of the function f.

(b) Is the point (1, 3) a maximum, a minimum or neither?
In problems #36, nd all extrema and inection points. Sketch the graph.
3. f (x) = x3 + x2
2
4. f (x) = x x+3
www.ck12.org

134

5. f (x) = x3 12x
6. f (x) = 14 x4 + 2x2
7. Use your graphing calculator to examine the graph of f (x) = x(x 1)3 (Hint: you will need to change
the y range in the viewing window)

(a) Discuss the concavity of the graph in the interval (0, 12 ).

(b) Use your calculator to nd the minimum value of the function in the interval.

8. True or False: f (x) = x4 + 4x3 has a relative minimum at x = 2 and a relative maximum at x = 0?
9. If possible, provide an example of a non-polynomial function that has exactly one relative minimum.
10. If possible, provide an example of a non-polynomial function that is concave downward everywhere
in its domain.

Review Answers
1. There is a relative minimum at x = 2; the relative minimum is located at (2, 3).
2. f (1) = 3 suggests that a + b = 2 and f (1) = 0 = 2 + a; solving this system we have that a = 2,
b = 4; the point (1, 3) is an absolute max of f.
3. Relative maximum at x = 2
3 , relative minimum at x = 0; the relative maximum is located at
2
( 3 , 0.15); the relative minimum is located at (0, 0). There is a point of inection at ( 13 , 0.07).

4. Relative

maximum at x = 3, located at ( 3, 2 3); relative minimum at x = 3, located at

( 3, 2 3). There are no inection points.

135

www.ck12.org

5. Relative maximum at x = 2, relative minimum at x = 2; the relative maximum is located at (2, 16);
the relative minimum is located at (2, 16). There is a point of inection at (0, 0).

6. Relative maximums at x = 2, relative minimum at x = 0; the relative maximums are located at

(2, 4) and (2, 4); the relative minimum is located at (0, 0). There are two inection points, located
2 3 20
at ( 2 3 3 , 20
9 ) and ( 3 , 9 ).

www.ck12.org

136

7.
8. (a) The graph is concave up in the interval;
(b) There is a relative minimum at (0.25, 0.10)
9. False: there are inection points at x = 0 and x = 2. There is a relative minimum at x = 3.

1
10. f (x) = x2 + ( x1
)

11. f (x) =

x on (0, +)

137

www.ck12.org

Also, f (x) = x12

3.5 Limits at Innity

Learning Objectives
A student will be able to:

Examine end behavior of functions on innite intervals.

Determine horizontal asymptotes.
Examine indeterminate forms of limits of rational functions.
Apply LHospitals Rule to nd limits.
Examine innite limits at innity.

Introduction
In this lesson we will return to the topics of innite limits and end behavior of functions and introduce a
new method that we can use to determine limits that have indeterminate forms.
www.ck12.org

138

Examine End Behavior of Functions on Innite Intervals

Suppose we are trying to analyze the end behavior of rational functions. In Lesson on Innite Lim2 1
its we looked at some rational functions such as f (x) = 2x
and showed that lim x+ f (x) = 2 and
x2 1
lim x f (x) = 2. We required an analysis of the end behavior of f since computing the limit by direct
substitution yielded the indeterminate form
. Our approach to compute the innite limit was to look
at actual values of the function f (x) as x approached . We interpreted the result graphically as the
function having a horizontal asymptote at f (x) = 2.

+3x+1
=
We were then able to nd innite limits of more complicated rational functions such as lim x 3x2x42x
2x2 +x3
3
1
=
,
p
>
.
Similarly,
we
used
such
an
approach
to
compute
limits
whenever
using
the
fact
that
lim
x x p
2
2 1
direct substitution resulted in the indeterminate form , such as lim x1 xx1
= 2.
4

Now lets consider other functions of the form ( f (x)/g(x)) where we get the indeterminate forms
and determine an appropriate analytical method for computing the limits.

and

Example 1:
Consider the function f (x) =
note the following:

ln(x+1)
x

and suppose we wish to nd lim x0

ln(x+1)
x

and lim x

ln(x+1)
x

We

1. Direct substitution leads to the indeterminate forms 00 and

.
2. The function in the numerator is not a polynomial function, so we cannot use our previous methods
such as applying lim x x1p = .

Lets examine both the graph and values of the function for appropriate x values, to see if they cluster
around particular y values. Here is a sketch of the graph and a table of extreme values.
We rst note that domain of the function is (1, 0) (0, +) and is indicated in the graph as follows:

So, lim x0

ln(x+1)
x

appears to approach the value 1 as the following table suggests.

Note: Please see Dierentiation and Integration of Logarithmic and Exponential Functions in Chapter 6
for more on derivatives of Logarithmic functions.

139

www.ck12.org

So we infer that lim x0

ln(x + 1)/x

0.1

1.05361

0.001

1.0005

undef

0.001

0.9995

0.1

0.953102

ln(x+1)
x

= 1.

ln(x+1)

For the innite limit, lim x x

= 1, the inference of the limit is not as obvious. The function appears
to approach the value 0 but does so very slowly, as the following table suggests.
x

ln(x + 1)/x

10

0.23979

50

0.078637

100

0.046151

1000

0.006909

10000

0.000921

This unpredictable situation will apply to many other functions of the form. Hence we need another
f (x)
method that will provide a dierent tool for analyzing functions of the form g(x) .
LHospitals Rule: Let functions f and g be dierentiable at every number other than c in some interval,
with g (x) , 0 if x , c. If lim xc f (x) = lim xc g(x) = 0, or if lim xc f (x) = and lim xc g(x) = ,
then:
f (x)

f (x)

1. lim xc g(x) = lim xc g (x) as long as this latter limit exists or is innite.
2. If f and g are dierentiable at every number x greater than some number a, with g (x) , 0, then
f (x)
f (x)
lim x g(x) = lim x g (x) as long as this latter limit exists or is innite.
Lets look at applying the rule to some examples.
www.ck12.org

140

Example 2:
We will start by reconsidering the previous example, f (x) =
LHospitals Rule:

ln(x+1)
,
x

and verify the following limits using

ln(x + 1)
= 1.
x0
x
ln(x + 1)
= 0.
lim
x
x
lim

Solution:
Since lim x0 ln(x + 1) = lim x0 x = 0, LHospitals Rule applies and we have
1

ln(x + 1)
1
x+1
lim
= lim
= = 1.
x0
x0
x
1
1
Likewise,
1

ln(x + 1)
0
x+1
lim
= lim
= = 0.
x
x 1
x
1
Now lets look at some more examples.
Example 3:
Evaluate lim x0

e x 1
x .

Solution:
Since lim x0 (e x 1) = lim x0 x = 0, LHospitals Rule applies and we have
ex 1
ex
1
= lim
= = 1.
x0
x0 1
x
1

lim

Lets look at an example with trigonometric functions.

Example 4:
Evaluate lim x0

1cos x
.
x2

Solution:
Since lim x0 (1 cos x) = lim x0 x2 = 0, LHospitals Rule applies and we have
1 cos x
sin x
cos x
1
lim
= lim
= lim
= .
2
x0
x0 2x
x0 2
2
x
Example 5: Evaluate lim x+

x2
ex

Solution:
Since lim x+ x2 = lim x+ e x = +, LHospitals Rule applies and we have
lim

x+

x2
2x
= lim x .
x+ e
ex

Here we observe that we still have the indeterminate form

the limit as follows:
lim

x+

So we apply LHospitals Rule again to nd

x2
2x
2
= lim x = lim x = 0
x
x+ e
x+ e
e

LHospitals Rule can be used repeatedly on functions like this. It is often useful because polynomial
functions can be reduced to a constant.

141

www.ck12.org

Lesson Summary
1.
2.
3.
4.
5.

We
We
We
We
We

learned to examine end behavior of functions on innite intervals.

determined horizontal asymptotes of rational functions.
examined indeterminate forms of limits of rational functions.
applied LHospitals Rule to nd limits of rational functions.
examined innite limits at innity.

Review Questions
1. Use your graphing calculator to estimate lim x+ [x[ln(x + 3) ln(x)]] .
x
2. Use your graphing calculator to estimate lim x+ ln(1+2e
x) .
In problems #310, use LHospitals Rule to compute the limits, if they exist.
x2 9
x3

lim x0 1+xx 1x
ln(x)
lim x+ x
lim x+ x2 e2x

3. lim x3
4.
5.
6.
7.
8.
9.
10.

lim x0 (1 x) x
x
lim x0 e 1x
x2
e x 1x
lim x x2
1
lim x x 4 ln(x)

Review Answers
1. lim x+ [x[ln(x + 3) ln(x)]] = 3
x
2. lim x+ ln(1+2e
x) = 1
x2 9
x3 =6

lim x0 1+xx 1x = 1
ln(x)
lim x+ x = 0
lim x+ x2 e2x = 0
1
lim x0 (1 x) x = 1e :Hint:
x
lim x0 e 1x
= 12
x2
e x 1x
lim x x2 = 0

3. lim x3
4.
5.
6.

7.
8.
9.
1
10. lim x x 4 ln(x) = 0

Let (1 x) x = eln(1x) x , so lim x0 (1 x) x = lim x0 eln(1x) x = elimx0

1
x ln(1x)

3.6 Analyzing the Graph of a Function

Learning Objectives
A student will be able to:
Summarize the properties of function including intercepts, domain, range, continuity, asymptotes,
relative extreme, concavity, points of inection, limits at innity.
Apply the First and Second Derivative Tests to sketch graphs.
www.ck12.org

142

Introduction
In this lesson we summarize what we have learned about using derivatives to analyze the graphs of functions.
We will demonstrate how these various methods can be applied to help us examine a functions behavior
and sketch its graph. Since we have already discussed the various techniques, this lesson will provide
examples of using the techniques to analyze the examples of representative functions we introduced in the
Lesson on Relations and Functions, particularly rational, polynomial, radical, and trigonometric functions.
Before we begin our work on these examples, it may be useful to summarize the kind of information about
functions we now can generate based on our previous discussions. Lets summarize our results in a table
like the one shown because it provides a useful template with which to organize our ndings.
Table 3.3: Table Summary
Analysis

f (x)
Domain and Range
Intercepts and Zeros
Asymptotes and limits at innity
Dierentiability
Intervals where f is increasing
Intervals where f is decreasing
Relative extrema
Concavity
Inection points
Example 1: Analyzing Rational Functions
Consider the function f (x) =

x2 4
.
x2 2x8

General Properties: The function appears to have zeros at x = 2. However, once we factor the
expression we see
(x + 2)(x 2)
x2 4
x2
=
.
=
x4
(x 4)(x + 2)
x2 2x 8
Hence, the function has a zero at x = 2, there is a hole in the graph at x = 2, the domain is (, 2)
(2, 4) (4, +), and the yintercept is at (0, 12 ).
Asymptotes and Limits at Innity
Given the domain, we note that there is a vertical asymptote at x = 4. To determine other asymptotes, we
examine the limit of f as x and x . We have
x2 4
= lim
lim 2
x
x x 2x 8
Similarly, we see that lim x

x2 4
x2 2x8

x2
x2

x2
x2

4
x2

2x
x2

8
x2

= lim

1
1

2
x

4
x2

8
x2

= 1.

= 1. We also note that y , 1 since x , 2.

Hence we have a horizontal asymptote at y = 1.

Dierentiability
8x8
2
f (x) = 2x
= (x4)
2 < 0. Hence the function is dierentiable at every point of its domain, and since
(x2 2x8)

f (x) < 0 on its domain, then f is decreasing on its domain, (, 2) (2, 4) (4, +).
2

143

www.ck12.org

f (x) =

4
.
(x 4)3

f (x) , 0 in the domain of f. Hence there are no relative extrema and no inection points.
So f (x) > 0 when x > 4. Hence the graph is concave up for x > 4.
Similarly, f (x) < 0 when x < 4. Hence the graph is concave down for x < 4, x , 2.
Lets summarize our results in the table before we sketch the graph.
Table 3.4: Table Summary
f (x) =

x2 4
x2 2x8

Domain and Range

Intercepts and Zeros
Asymptotes and limits at innity
Dierentiability
Intervals where f is increasing
Intervals where f is decreasing
Relative extrema
Concavity
Inection points

Analysis

{
}
D = (, 2) (2, 4) (4, +) R = all reals , 1
)
(
zero at x = 2, yintercept at 0, 12
VA at x = 4, HA at y = 1, hole in the graph at
x = 2
dierentiable at every point of its domain
nowhere
(, 2) (2, 4) (4, +)
none
concave up in (4, +), concave down in (, 2)
(2, 4)
none

Finally, we sketch the graph as follows:

Lets look at examples of the other representative functions we introduced in Lesson 1.2.
Example 2:
Analyzing Polynomial Functions
Consider the function f (x) = x3 + 2x2 x 2.
General Properties
The domain of f is (, +) and the yintercept at (0, 2).
The function can be factored
f (x) = x3 + 2x2 x 2 = x2 (x + 2) 1(x + 2) = (x2 1)(x + 2) = (x 1)(x + 1)(x + 2)
www.ck12.org

144

and thus has zeros at x = 1, 2.

Asymptotes and limits at innity

Given the domain, we note that there are no vertical asymptotes. We note that lim x f (x) = + and
lim x f (x) = .
Dierentiability

f (x) = 3x2 + 4x 1 = 0 if x = 46 28 = 23 7 . These are the critical values. We note that the function
is dierentiable at every point of its domain.
(
(
)
)

f (x) > 0 on , 23 7 and 2+3 7 , + ; hence the function is increasing in these intervals.
(
)

2 7 2+ 7

and thus is f decreasing there.

Similarly, f (x) < 0 on
3 ,
3
f (x) = 6x + 4 = 0 if x = 23 , where there is an inection point.
(
)
In addition, f 23 7 < 0. Hence the graph has a relative maximum at x =

2 7
3

and located at the

2+ 7
3

and located at the

point (1.55, 0.63).

(
)
We note that f (x) < 0 for x < 23 . The graph is concave down in , 32 .
(
)
And we have f 2+3 7 > 0; hence the graph has a relative minimum at x =
point (0.22, 2.11).

(
)
We note that f (x) > 0 for x > 23 . The graph is concave up in 23 , + .
Table 3.5: Table Summary
f (x) = x3 + 2x2 x 2

Analysis

Domain and Range

Intercepts and Zeros
Asymptotes and limits at innity
Dierentiability

D = (, +), R = { all reals }

zeros at x = 1, 2, y, intercept at (0, 2)
no asymptotes
dierentiable
point of
(
) at every
(
) its domain

Intervals where f is increasing

Intervals where f is decreasing

, 23

and
)

2 7 2+ 7
3 ,
3

145

2+ 7
, +
3

www.ck12.org

Table 3.5: (continued)

f (x) = x3 + 2x2 x 2

Analysis

Relative extrema

relative maximum at x = 23 7 and located at the

point (1.55, 0.63);

relative minimum at x = 2+3 7 and located at the

point (0.22, 2.11).

Concavity

)
(
concave up in 23 , + .
(
)
concave down in , 32 .

Inection points

(
)
x = 23 , located at the point 23 , .74

Here is a sketch of the graph:

Example 3: Analyzing Radical Functions

Consider the function f (x) = 2x 1.

General Properties
(
)
The domain of f is 21 , + , and it has a zero at x = 21 .
Asymptotes and Limits at Innity
Given the domain, we note that there are no vertical asymptotes. We note that lim x f (x) = +.
Dierentiability
f (x) =

1
2x1

> 0 for the entire domain of f. Hence f is increasing everywhere in its domain. f (x) is not

dened at x = 21 , so x =
x = 21 , so

is a critical value.
(
)
(
)
< 0 everywhere in 12 , + . Hence f is concave down in 12 , + . f (x) is not dened at

1
(2x1)3
x = 12 is

f (x) =

1
2

an absolute minimum.
Table 3.6: Table Summary

f (x) =

2x 1

Domain and Range

Intercepts and Zeros
Asymptotes and limits at innity
Dierentiability
Intervals where f is increasing
www.ck12.org

Analysis
(
)
D = 12 , + , R = {y 0}
zeros at x = 21 , no yintercept
no asymptotes (
)
dierentiable in 12 , +
(
)
everywhere in D = 21 , +

146

Table 3.6: (continued)

f (x) =

2x 1

Analysis

Intervals where f is decreasing

Relative extrema

nowhere
none
(
)
absolute minimum at x = 12 , located at 12 , 0

Concavity
Inection points

concave down in
none

1
2 , +

Here is a sketch of the graph:

Example 4: Analyzing Trigonometric Functions

We will see that while trigonometric functions can be analyzed using what we know about derivatives, they
will provide some interesting challenges that we will need to address. Consider the function f (x) = x2 sin x
on the interval [, ].
General Properties
We note that f is a continuous function and thus attains an absolute maximum and minimum in [, ].
Its domain is [, ] and its range is R = { y } .
Dierentiability
f (x) = 1 2 cos x = 0 at x = 3 , 3 .
(
)
(
)
(
)
(
)
Note that f (x) > 0 on 3 , and , 3 ; therefore the function is increasing in 3 , and , 3 .
(
)
(
)
Note that f (x) < 0 on 3 , 3 ; therefore the function is decreasing in 3 , 3 .
f (x) = 2 sin x = 0 if x = 0, , . Hence the critical values are at x = , 3 , 0, 3 , .
( )
f 3 > 0; hence there is a relative minimum at x = 3 .
( )
f 3 < 0; hence there is a relative maximum at x = 3 .
f (x) > 0 on (0, ) and f (x) < 0 on (, 0). Hence the graph is concave up and decreasing on (0, ) and
concave down on (, 0). There is an inection point at x = 0, located at the point (0, 0).
Finally, there is absolute minimum at x = , located at (, ), and an absolute maximum at x = ,
located at (, ).

147

www.ck12.org

Table 3.7: Table Summary

f (x) = x 2 sin x

Analysis

Domain and Range

Intercepts and Zeros
Asymptotes and limits at innity
Dierentiability
Intervals where f is increasing
Intervals where f is decreasing
Relative extrema

D = [, ], R = { y }
x = 3 , 3
no asymptotes
dierentiable
in D = [, ]
(
)

, and (, 3 )
(3
)
3 , 3
relative maximum at x = /3
relative minimum at x = /3
absolute maximum at x =
(
)
absolute minimum at x = 12 , located at 12 , 0

Concavity
Inection points

concave up in (0, )
x = 0, located at the point (0, 0)

Lesson Summary
1. We summarized the properties of function, including intercepts, domain, range, continuity, asymptotes, relative extreme, concavity, points of inection, and limits at innity.
2. We applied the First and Second Derivative Tests to sketch graphs.

Multimedia Links
Each of the problems above started with a function and then we analyzed its zeros, derivative, and concavity.
Even without the function denition it is possible to sketch the graph if you know some key pieces of
information. In the following video the narrator illustrates how to use information about the derivative
of a function and given points on the function graph to sketch the function. Khan Academy Graphing
with Calculus (9:43) . Another approach to this analysis is to look at a function, its derivative, and its
second derivative on the same set of axes. This interactive applet called Curve Analysis allows you to trace
function points on a graph and its rst and second derivative. You can also enter new functions (including
the ones from the examples above) to analyze the functions and their derivatives.
For more information about computing derivatives of higher orders (7.0), see Math Video Tutorials by
James Sousa, Higher-Order Derivatives: Part 1 of 2 (7:34) and Math Video Tutorials by James Sousa,
Higher-Order Derivatives: Part 2 of 2 (5:21) .
www.ck12.org

148

Figure 3.11: Graphing with Calculus (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/523

Figure 3.12: Higher-Order Derivatives: Part 1 of 2 (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/524

Figure 3.13: Higher-Order Derivatives: Part 2 of 2 (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/525

149

www.ck12.org

For a presentation of higher order partial derivatives (7.0), see Calculus, Higher Order Derivatives (8:09)

Figure 3.14: Calculus: Higher Order Partial Derivatives (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/526
.

Review Questions
1. Summarize each of the following functions by lling out the table. Use the information to sketch a
graph of the function.
2. f (x) = x3 + 3x2 x 3
Table 3.8:
f (x) = x3 + 3x2 x 3

Analysis

Domain and Range

Intercepts and Zeros
Asymptotes and limits at innity
Dierentiability
Intervals where f is increasing
Intervals where f is decreasing
Relative extrema
Concavity
Inection points
2. f (x) = x4 + 4x3 4x2
Table 3.9:
f (x) = x4 + 4x3 4x2

Analysis

Domain and Range

Intercepts and Zeros
Asymptotes and limits at innity
Dierentiability
Intervals where f is increasing
Intervals where f is decreasing
Relative extrema
Concavity
Inection points
www.ck12.org

150

3. f (x) =

2x2
x2

Table 3.10:
f (x) =

2x2
x2

Analysis

Domain and Range

Intercepts and Zeros
Asymptotes and limits at innity
Dierentiability
Intervals where f is increasing
Intervals where f is decreasing
Relative extrema
Concavity
Inection points
1

4. f (x) = x x 3
Table 3.11:
1

f (x) = x x 3

Analysis

Domain and Range

Intercepts and Zeros
Asymptotes and limits at innity
Dierentiability
Intervals where f is increasing
Intervals where f is decreasing
Relative extrema
Concavity
Inection points

5. f (x) = 2x 6 + 3
Table 3.12:

f (x) = 2x 6 + 3

Analysis

Domain and Range

Intercepts and Zeros
Asymptotes and limits at innity
Dierentiability
Intervals where f is increasing
Intervals where f is decreasing
Relative extrema
Concavity
Inection points

151

www.ck12.org


6. f (x) = x2 2 x
Table 3.13:

f (x) = x2 2 x

Analysis

Domain and Range

Intercepts and Zeros
Asymptotes and limits at innity
Dierentiability
Intervals where f is increasing
Intervals where f is decreasing
Relative extrema
Concavity
Inection points
7. f (x) = 1 + cos x = 0 on [, ]
Table 3.14:
f (x) = 1 + cos x

Analysis

Domain and Range

Intercepts and Zeros
Asymptotes and limits at innity
Dierentiability
Intervals where f is increasing
Intervals where f is decreasing
Relative extrema
Concavity
Inection points

Review Answers
1.
Table 3.15:
f (x) = x3 + 3x2 x 3

Analysis

Domain and Range

Intercepts and Zeros
Asymptotes and limits at innity
Dierentiability

D = (, +), R = {all reals}

zeros at x = 1, 3, yintercept at (0, 3)
no asymptotes
dierentiable
point of )its domain
(
) at every
(

Intervals where f is increasing

Intervals where f is decreasing

www.ck12.org

, 32
3

and
)

32 3 3+2 3
,
3
3

152

3+2 3
, +
3

Table 3.15: (continued)

f (x) = x3 + 3x2 x 3

Analysis

Relative extrema

relative maximum at x =
point (2.15, 3.07);
relative minimum at x =
point (0.15, 3.07)

32 3
,
3

located at the

3+2 3
,
3

located at the

Concavity

concave up in (1, +)
concave down in (, 1)

Inection points

x = 1, located at the point (1, 0)

2.
2.
Table 3.16:
f (x) = x4 + 4x3 4x2

Analysis

Domain and Range

Intercepts and Zeros
Asymptotes and limits at innity
Dierentiability
Intervals where f is increasing
Intervals where f is decreasing
Relative extrema

D = (, +), R = {y 0}
zeros at x = 0, 2, yintercept at (0, 0)
no asymptotes
dierentiable at every point of its domain
(, 0) and (1, 2)
(0, 1) and (2, +)
relative maximum at x = 0, located at the point
(0, 0); and at at x = 2 located at the point (2, 0)
relative minimum at x = 1, located at the point
(1, 1)

Concavity

concave up in

)
(
)
concave down in , 32 and 53 , +

Inection points

x = 32 , 53 , located at the points

(
)
5
,
0.30
3

2 5
3 ,(3

2
3 , 0.79

and

3.
3.
Table 3.17:
f (x) =

2x2
x2

Domain and Range

Intercepts and Zeros
Asymptotes and limits at innity
Dierentiability

Analysis
D = (, 0) (0, +), R = {y , 0}
zeros at x = 1, no yintercept
HA y = 0
dierentiable at every point of its domain

153

www.ck12.org

Table 3.17: (continued)

f (x) =

2x2
x2

Analysis

Intervals where f is increasing

Intervals where f is decreasing
Relative extrema
Concavity

Inection points

(0, 2)
(, 0) and (2, +)
relative maximum at x = 2, located at the point
(2, 0.5)
concave up in (3, +)
concave down in (, 0) and (0, 3)
x = 3, located at the point (3, 94 )

4.
4.
Table 3.18:
1

f (x) = x x 3

Analysis

Domain and Range

Intercepts and Zeros
Asymptotes and limits at innity
Dierentiability

D = (, +), R = {all reals}

zeros at x = 1, 0, yintercept at (0, 0)
no asymptotes
dierentiable
in( (, 0) ) (0, +)
(
)

Intervals where f is increasing

Intervals where f is decreasing
Relative extrema

, 3 and
( 9 )
3
3
9 , 9

3
9 , +

relative
maximum
at x =
(
)
3
9 , 0.384
minimum
at x =
(relative
)

3
9 , 0.384

3
9 ,

located at the point

3
9 ,

located at the point

Concavity

concave up in (0, +)
concave down in (, 0)

Inection points

x = 0, located at the point (0, 0)

5.
5.
Table 3.19:

f (x) = 2x 6 + 3

Analysis

Domain and Range

Intercepts and Zeros
Asymptotes and limits at innity
Dierentiability

D = (3, +), R = {y 3}
zero at x = 15
2 , no yintercept
no asymptotes
dierentiable in (3, +)

www.ck12.org

154

Table 3.19: (continued)

f (x) = 2x 6 + 3

Analysis

Intervals where f is increasing

Intervals where f is decreasing
Relative extrema
Concavity
Inection points

nowhere
everywhere in D = (3, +)
none absolute maximum at x = 3, located at (3, 3)
concave up in (3, +)
none

6.
6.
Table 3.20:

f (x) = x2 2 x

Analysis

Domain and Range

Intercepts and Zeros
Asymptotes and limits at innity
Dierentiability

D = (0, +),R = {y 1.19}

3
zero at x = 4, no yintercept
no asymptotes
dierentiable
(
) in (0, +)

Intervals where f is increasing

Intervals where f is decreasing

3
16
, +
( 4 3 )
0, 416

Relative extrema

relative
( 3 minimum
) at x =
x = 416 , 1.19

Concavity
Inection points

concave up in (0, +)
none

3
16
4 ,

located at the point

7.
8. (g)
Table 3.21:
f (x) = 1 + cos x

Analysis

Domain and Range

Intercepts and Zeros
Asymptotes and limits at innity
Dierentiability
Intervals where f is increasing
Intervals where f is decreasing
Relative extrema

D = [, ], R = {0 y 2}
zeros at x = , , yintercept at (0, 2)
no asymptotes; lim x f (x) does not exist
dierentiable at every point of its domain
(, 0)
(0, )
absolute max at x = 0, located at the point (0, 2)
absolute minimums at x = , located at the points
(, 0) and (, 0)

Concavity
Inection points

concave down in (, )
(
)
(
)
x = 2 , located at the points 2 , 1 and 2 , 1

155

www.ck12.org

9.

3.7 Optimization
Learning Objectives
A student will be able to:

Use the First and Second Derivative Tests to nd absolute maximum and minimum values of a
function.
Use the First and Second Derivative Tests to solve optimization applications.

Introduction
In this lesson we wish to extend our discussion of extrema and look at the absolute maximum and minimum
values of functions. We will then solve some applications using these methods to maximize and minimize
functions.
Absolute Maximum and Minimum
We begin with an observation about nding absolute maximum and minimum values of functions that are
continuous on a closed interval. Suppose that f is continuous on a closed interval [a, b]. Recall that we can
nd relative minima and maxima by identifying the critical numbers of f in (a, b) and then applying the
Second Derivative Test. The absolute maximum and minimum must come from either the relative extrema
of f in (a, b) or the value of the function at the endpoints, f (a) or f (b). Hence the absolute maximum or
minimum values of a function f that is continuous on a closed interval [a, b] can be found as follows:

1. Find the values of f for each critical value in (a, b);

2. Find the values of the function f at the endpoints of [a, b];
3. The absolute maximum will be the largest value of the numbers found in 1 and 2; the absolute
minimum will be the smallest number.

The optimization problems we will solve will involve a process of maximizing and minimizing functions.
Since most problems will involve real applications that one nds in everyday life, we need to discuss how
the properties of everyday applications will aect the more theoretical methods we have developed in our
analysis. Lets start with the following example.
Example 1:
A company makes high-quality bicycle tires for both recreational and racing riders. The number of tires
that the company sells is a function of the price charged and can be modeled by the formula T (x) =
x3 + 36.5x2 + 50x + 250, where x is the priced charged for each tire in dollars. At what price is the
maximum number of tires sold? How many tires will be sold at that maximum price?
Solution:
Lets rst look at a graph and make some observations. Set the viewing window ranges on your graphing
calculator to [10, 50] for x and [500, 10000] for y. The graph should appear as follows:
www.ck12.org

156

We rst note that since this is a real-life application, we observe that both quantities, x and T (x), are
positive or else the problem makes no sense. These conditions, together with the fact that the zero of T (x)
is located at x = 37.9, suggest that the actual domain of this function is 0 < < 37. This domain, which we
refer to as a feasible domain, illustrates a common feature of optimization problems: that the real-life
conditions of the situation under study dictate the domain values. Once we make this observation, we can
use our First and Second Derivative Tests and the method for nding absolute maximums and minimums
on a closed interval (in this problem, [0, 37]), to see that the function attains an absolute maximum at
x = 25, at the point (25, 8687.5). So, charging a price of $25 will result in a total of 8687 tires being sold.
In addition to the feasible domain issue illustrated in the previous example, many optimization problems
involve other issues such as information from multiple sources that we will need to address in order to solve
these problems. The next section illustrates this fact.
Primary and Secondary Equations
We will often have information from at least two sources that will require us to make some transformations
in order to answer the questions we are faced with. To illustrate this, lets return to our Lesson on Related
Rates problems and recall the right circular cone volume problem.
1
V = r2 h.
3

We started with the general volume formula V = 13 r2 h , but quickly realized that we did not have suicient
information to nd dh
dt since we had no information about the radius when the water level was at a particular
height. So we needed to employ some indirect reasoning to nd a relationship between r and h, r(t) =

157

2h(t)
5 .

www.ck12.org

We then made an appropriate substitution in the original formula (V = 13

to nd the solution.

( )2
2h
5

h=

4 3
75 h )

and were able

We started with a primary equation, V = 13 r2 h, that involved two variables and provided a general model
of the situation. However, in order to solve the problem, we needed to generate a secondary equation,
2h(t)
r(t) = 5 , that we then substituted into the primary equation. We will face this same situation in most
optimization problems.
Lets illustrate the situation with an example.
Example 2:
Suppose that Mary wishes to make an outdoor rectangular pen for her pet chihuahua. She would like the
pen to enclose an area in her backyard with one of the sides of the rectangle made by the side of Marys
house as indicated in the following gure. If she has 90 ft of fencing to work with, what dimensions of the
pen will result in the maximum area?

Solution:
The primary equation is the function that models the area of the pen and that we wish to maximize,
A = xy.
The secondary equation comes from the information concerning the fencing Mary has to work with. In
particular,
2x + y = 90.
Solving for y we have
y = 90 2x.
We now substitute into the primary equation to get
A = xy = x(90 2x), or
A = 90x 2x2 .
It is always helpful to view the graph of the function to be optimized. Set the viewing window ranges on
your graphing calculator to [10, 100] for x and [500, 1200] for y. The graph should appear as follows:
www.ck12.org

158

The feasible domain of this function is 0 < < 45, which makes sense because if x is 45 feet, then the gure
will be two 45-foot-long fences going away from the house with 0 feet left for the width, y. Using our First
and Second Derivative Tests and the method for nding absolute maximums and minimums on a closed
interval (in this problem, [0, 45]), we see that the function attains an absolute maximum at x = 22.5, at
the point (22.5, 1012.5). So the dimensions of the pen should be x = 22.5, y = 45; with those dimensions,
the pen will enclose an area of 1012.5 ft2 .
Recall in the Lesson Related Rates that we solved problems that involved a variety of geometric shapes.
Lets consider a problem about surface areas of cylinders.
Example 3:
A certain brand of lemonade sells its product in 16ounce aluminum cans that hold 473 ml (1 ml = 1 cm3 ).
Find the dimensions of the cylindrical can that will use the least amount of aluminum.
Solution:
We need to develop the formula for the surface area of the can. This consists of the top and bottom
areas, each r2 , and the surface area of the side, 2rh (treating the side as a rectangle, the lateral area is
(circumference of the top) (height)). Hence the primary equation is
A = 2r2 + 2rh.
We observe that both our feasible domains require r, h > 0.
In order to generate the secondary equation, we note that the volume for a circular cylinder is given by
. We substitute
V = r2 h. Using the given information we can nd a relationship between r and h, h = 473
r2
( )
473
946
2
2
this value into the primary equation to get A = 2r + 2r r2 , or A = 2r + r .

3 946
d2 A
= 4r 946
=
0
when
r
=
4 9.06 cm. We note that dr2 > 0 since r > 0. Hence we have a minimum
r2

473

surface area when r = 3 946

3 946 2 = 8.44 cm.
40 4.22 cm and h =
dA
dr

159

www.ck12.org

Lesson Summary
1. We used the First and Second Derivative Tests to nd absolute maximum and minimum values of a
function.
2. We used the First and Second Derivative Tests to solve optimization applications.

Multimedia Links
For video presentations of maximum-minimum Business and Economics applications (11.0), see Math
Video Tutorials by James Sousa, Max & Min Apps. w/calculus, Part 1 (9:57) and Math Video Tutorials

Figure 3.15: Max and Min Apps. w/ Calculus - Part 1 of 2 (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/527
by James Sousa, Max & Min Apps. w/calculus, Part 2 (9:57) .

Figure 3.16: Max and Min Apps. w/ Calculus - Part 2 of 2 (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/528
To see more examples of worked out problems involving nding minima and maxima on an interval (11.0),
see the video at Khan Academy Minimum and Maximum Values on an Interval (11:41) . This video shows

Figure 3.17: Calculus: Maximum and minimum values on an interval (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/529
the process of applying the rst derivative test to problems with no context, just a given function and a
domain. A classic problem in calculus involves maximizing the volume of an open box made by cutting
www.ck12.org

160

squares from a rectangular sheet and folding up the edges. This very cool calculus applet shows one solution
to this problem and multiple representations of the problem as well. Calculus Applet on Optimization

Review Questions
In problems #14, nd the absolute maximum and absolute minimum values, if they exist.
f (x) = 2x2 6x + 6 on [0, 5]
f (x) = x3 + 3x2 on [2, 3]
2
f (x) = 3x 3 6x + x on [1, 8]
f (x) = x4 x3 on [2, 2]
Find the dimensions of a rectangle having area 2000 ft2 whose perimeter is as small as possible.
Find two numbers whose product is 50 and whose sum is a minimum.
John is shooting a basketball from half-court. It is approximately 45 ft from the half court line to
the hoop. The function s(t) = 0.025x2 + x + 15 models the basketballs height above the ground
s(t) in feet, when it is t feet from the hoop. How many feet from John will the ball reach its highest
height? What is that height?
8. The height of a model rocket t seconds into ight is given by the formula h(t) = 31 t3 + 4t2 + 25t + 4.
1.
2.
3.
4.
5.
6.
7.

(a) How long will it take for the rocket to attain its maximum height?
(b) What is the maximum height that the rocket will reach?
(c) How long will the ight last?
9. Show that of all rectangles of a given perimeter, the rectangle with the greatest area is a square.
10. Show that of all rectangles of a given area, the rectangle with the smallest perimeter is a square.

Review Answers
( )
Absolute minimum at x = 32 , f 23 = 32 . Absolute maximum at x = 5, f (5) = 26.
Absolute minimum at x = 0, f (0) = 0. Absolute maximum at x = 3, f (3) = 54.
Absolute minimum at x = 8, f (8) = 30. Absolute maximum at x = 1, f (1) = 3.
Absolute minimum
at x = 0.75, f (0.75) = 0.105. Absolute maximum at x = 2, f (2) = 24.

x = y = 20 5
x=y=5 2
At t = 20 ft, the basketball will reach a height of s(t) = 25 ft.
The rocket will take approximately t 10.4 sec to attain its maximum height of 321.7 ft. the rocket
will hit the ground at t 16.6 sec.
9. .
10. .
1.
2.
3.
4.
5.
6.
7.
8.

3.8 Approximation Errors

Learning Objectives
A student will be able to:

Extend the Mean Value Theorem to make linear approximations.

Analyze errors in linear approximations.
Extend the Mean Value Theorem to make quadratic approximations.
Analyze errors in quadratic approximations.

161

www.ck12.org

Introduction
In this lesson we will use the Mean Value Theorem to make approximations of functions. We will apply
the Theorem directly to make linear approximations and then extend the Theorem to make quadratic
approximations of functions.
Lets consider the tangent line to the graph of a function f at the point (a, f (a)). The equation of this line
is y = f (a) + f (a)(x a). We observe from the graph that as we consider x near a, the value of f (x) is very
close to f (a).

In other words, for x values close to a, the tangent line to the graph of a function f at the point (a, f (a))
provides an approximation of f (x) or f (x) f (a) + f (a)(x a). We call this the linear or tangent line
approximation of f at a and indicate it by the formula L(x) = f (a) + f (a)(x a).

The linear approximation can be used to approximate functional values that deviate slightly from known
values. The following example illustrates this process.
Example 1:
Use the linear approximation of the function f (x) =
Solution:
We know that f (6) =
x values close to 6.

x 2 at a = 6 to approximate

3.95.

6 2 = 2. So we will nd the linear approximation of the function and substitute

L(x) = f (6) + f (6)(x 6).

We note that f (x) =

1 ,
2 x2

f (6) = 14 .

We also know that f (6) = 2.

By substitution, we have
f (x) f (6) + f (6)(x 6) for x near 6.
Hence L(x) = 2 + 14 (x 6) =

+ 14 x.

We observe that to approximate 3.95 we need to evaluate the linear approximation at 5.95, and we have
www.ck12.org

1
2

162

L(5.95) = 12 + 14 (5.95) = 1.9875. If we were to compare this approximation to the actual value,
1.9874, we see that it is a very good approximation.

3.95

If we observe a table of x values close to 6, we see how the approximations compare to the actual value.
f (x) =

3.95

3.99

4.1

4.05

x2

1 1
+ x
2 4

Actual

L(x) =

5.95

1.9875

1.9874

5.99

1.9975

1.9974

6.01

2.0025

2.0024

6.05

2.0125

2.0124

Setting Error Estimates

We would like to have condence in the approximations we make. We therefore can choose the x values
close to a to ensure that the errors are within acceptable boundaries. For the previous example, we saw
that the values of L(x) close to a = 6, gave very good approximations, all within 0.0001 of the actual value.
Example 2:
Lets suppose that for the previous example, we did not require such precision. Rather, suppose we wanted
to nd the range of x values close to 6 that we could choose to ensure that our approximations lie within
0.01 of the actual value.
Solution:
The easiest way for us to nd the proper range ofx values( is to use
) the graphing calculator. We rst note
1
x
that our precision requirement can be stated as | x 2 2 + 4 | < 0.01.

If we enter the functions f (x) = x 2 and L(x) = 12 + 14 x into the Y = menu as Y1 and Y2 , respectively,
we will be able to view the function values of the functions using the [TABLE] feature of the calculator.
In order to view the dierences between the actual and approximate values, we can enter into the Y =
menu the dierence function Y3 = Y1 Y2 as follows:
1. Go to the Y = menu and place cursor on the Y3 line.
2. Press the following sequence of key strokes: [VARS] [FUNCTION] [Y1 ]. This will copy the function
Y1 onto the Y3 line of the Y = menu.
3. Press [-] to enter the subtraction operation onto the Y3 line of the Y = menu.
4. Repeat steps 1 - 2 and choose Y2 to copy Y2 onto the Y3 line of the Y = menu.
Your screen should now appear as follows:

Now lets setup the [TABLE] function so that we nd the required accuracy.
1. Press 2ND followed by [TBLSET] to access the Table Setup screen.

163

www.ck12.org

2. Set the [TBLStart] value to 5 and Tbl to 0.1.

Your screen should now appear as follows:

Now we are ready to nd the required accuracy.

Access the [TABLE] function, scroll through the table, and nd those x values that ensure
( x = 5.24
)
Y3 . At
we see that Y3 = 0.01115. At x = 6.84, we see that Y3 = 0.00911. Hence if 5.24 6.84, | x 2 21 + 4x | <
0.01.

Non-Linear Approximations
It turns out that the linear approximations we have discussed are not the only approximations that we
can derive using derivatives. We can use non-linear functions to make approximations. These are called
Taylor Polynomials and are dened as
T n (x) = f (a) + f (a)(x a) +

f n (a)
f (a)
(x a)2 + . . . +
(x a)n .
2!
n!

We call this the Taylor Polynomial of fcentered at a.

For our discussion, we will focus on the quadratic case. The Taylor Polynomial corresponding to n = 2 is
given by
T 2 (x) = f (a) + f (a)(x a) +

1
f (a)(x a)2 .
2

Note that this is just our linear approximation with an added term. Hence we can view it as an approximation of f for x values close to a.
Example 3:
Find the quadratic approximation of the function f (x) =
approximations from the rst example.

x 2 at a = 6 and compare them to the linear

Solution:
Recall that L(x) =

1
2

+ 14 x.

Hence T 2 (x) = L(x) + 12 f (6)(x 6)2 .

f (x) =

1
4 4(x2)3

; so f (6) =

Hence T 2 (x) = L(x)

www.ck12.org

1
64 (x

6)2 =

1
2

1
(62)2

1
= 32
.

+ 14 x

1
64 (x

1 2
6)2 = 64
x +

164

7
16 x

1
16 .

1 2
7
1
So T 2 (x) = 64
x + 16
x 16
. If we update our table from the rst example we can see how the quadratic
approximation compares with the linear approximation.

f (x) =

3.95

3.99

4.1

4.05

x2

1 1
+ x
2 4

T 2 (x) =

1 2
7
1
x + x
64
16
16

Actual

L(x) =

5.95

1.9875

1.9874

1.9874

5.99

1.9975

1.9974

1.9974

6.01

2.0025

2.0024

2.0024

6.05

2.0125

2.0124

2.0124

As you can see from the graph below, T (x) is an excellent approximation of f (x) near x = 6.

We get a slightly better approximation for the quadratic than for the linear. If we reect on this a bit, the
nding makes sense since the shape and properties of quadratic functions more closely approximate the
shape of radical functions.
Finally, as in the rst example, we wish to determine the range of x values that will ensure that our
approximations are within 0.01of the actual value. Using the [TABLE] feature of the calculator, we nd
that if 4.444 7.87, then | x 2 T (x)| < 0.01.

Lesson Summary
1.
2.
3.
4.

We
We
We
We

extended the Mean Value Theorem to make linear approximations.

analyzed errors in linear approximations.
extended the Mean Value Theorem to make quadratic approximations.
analyzed errors in quadratic approximations.

Review Questions
In problems #14, nd the linearization L(x) of the function at x = a.
1.
2.
3.
4.

f (x) = 2x4 6x3 near a = 2

2
f (x) = x 3 on a = 27

Find the linearization of the function f (x) = 5 x near a = 1 and use it to approximate 4.01.
Based on using linear approximations, is the following
approximation reasonable?
1.0014 = 1.004

165

www.ck12.org

5. Use a linear approximation to approximate the following:

3
16.08 4
6. Verify the the following linear approximation at a = 1. Determine the values of x for which the linear
approximation is accurate to 0.01.

x 2
3
2x +
3 3

7. Find the quadratic approximation for the function in #3, f (x) = 5 x near a = 1.
8. Determine the values of x for which the quadratic approximation found in #7 is accurate to 0.01.
9. Determine the quadratic approximation for f (x) = 2x4 6x3 near a = 2. Do you expect that the
quadratic approximation is better or worse than the linear approximation? Explain your answer.

Review Answers
1.
2.
3.
4.
5.
6.
7.
8.
9.

f (2) = 80, f (2) = 136; L(x) = 192 136x

f (27) = 9, f (27) = 2/9; L(x) = 45 +2x
f (1) = 2, f (1) = 14 ; L(x) = 94 14 x; 4.01 2.0025
Yes; using linear approximation on f (x) = x4 near a = 1 we nd that L(x) = 4x 3; L(1.001) = 1.004.
3
Using linear approximation on f (x) = x 4 near a = 16 we nd L(x) = 8 + 83 (x 16) = 2 + 38 x;
L(16.08) = 8.03.
0.84 < < 1.14
1 2
7
T (x) = 64
x 32
x + 143
64
0.87 2.56
T (x) = 48x2 +56x; we would expect it to be a better approximation since the graph of f (x) = 2x4 6x3
is similar to the graph of a quadratic function.
f (x) = 2x4 6x3
T (x) = 48x2 + 56x

Texas Instruments Resources

In the CK-12 Texas Instruments Calculus FlexBook, there are graphing calculator activities
designed to supplement the objectives for some of the lessons in this chapter. See http:
//www.ck12.org/flexr/chapter/9728.

www.ck12.org

166

Chapter 4
Integration
4.1 Indenite Integrals Calculus
Learning Objectives
A student will be able to:

Find antiderivatives of functions.

Represent antiderivatives.
Interpret the constant of integration graphically.
Solve dierential equations.
Use basic antidierentiation techniques.
Use basic integration rules.

Introduction
In this lesson we will introduce the idea of the antiderivative of a function and formalize as indenite
integrals. We will derive a set of rules that will aid our computations as we solve problems.
Antiderivatives

Denition A function F(x) is called an antiderivative of a function f if F (x) = f (x) for all x in the
domain of f.

Example 1:
Consider the function f (x) = 3x2 . Can you think of a function F(x) such that F (x) = f (x)? (Answer:
F(x) = x3 , F(x) = x3 6, many other examples.)
Since we dierentiate F(x) to get f (x), we see that F(x) = x3 +C will work for any constant C. Graphically,
we can think the set of all antiderivatives as vertical transformations of the graph of F(x) = x3 . The gure
shows two such transformations.

167

www.ck12.org

With our denition and initial example, we now look to formalize the denition and develop some useful
rules for computational purposes, and begin to see some applications.
Notation and Introduction to Indenite Integrals
The process of nding antiderivatives is called antidierentiation, more commonly referred to as integration. We have a particular sign and set of symbols we use to indicate integration:

f (x)dx = F(x) + C.
We refer to the left side of the equation as the indenite integral of f (x) with respect to x. The function
f (x) is called the integrand and the constant C is called the constant of integration. Finally the
symbol dx indicates that we are to integrate with respect to x.
Using this notation, we would summarize the last example as follows:

3x2 dx = x3 + C
Using Derivatives to Derive Basic Rules of Integration
As with dierentiation, there are several useful rules that we can derive to aid our computations as we
solve problems. The rst of these is a rule for integrating power functions, f (x) = xn [n , 1], and is stated
as follows:

1 n+1
x
+ C.
xn dx =
n+1
1 n+1
We can easily prove this rule. Let F(x) = n+1
x
+ C, n , 1 . We dierentiate with respect to x and we
have:
)
d ( 1 n+1
d ( 1 n+1 )
d

F (x) =
x
+C =
x
+ (C)
dx n + 1
dx n + 1
dx
( 1 ) d (
)
d
(C)
=
xn+1 +
n + 1 dx
dx
(n + 1)
xn + 0
=
n+1
= xn .

The rule holds for f (x) = xn [n , 1].

1 What happens in the case where we have a power function to
1
integrate with n = 1, say x dx = x dx. We can see that the rule does not work since it would result
www.ck12.org

168

in division by 0. However, if we pose the problem as nding F(x) such that F (x) = 1x , we recall that the
d
derivative of logarithm functions had this form. In particular, dx
ln x = 1x . Hence

1
dx = ln x + C.
x
In addition to logarithm functions, we recall that the basic exponentional function, f (x) = e x , was special
in that its derivative was equal to itself. Hence we have

e x dx = e x + C.
Again we could easily prove this result by dierentiating the right side of the equation above. The actual
proof is left as an exercise to the student.
As with dierentiation, we can develop several rules for dealing with a nite number of integrable functions.
They are stated as follows:
If f and g are integrable functions, and C is a constant, then

[ f (x) + g(x)] dx =
f (x)dx + g(x)dx,

[ f (x) g(x)]dx =
f (x)dx g(x)dx,

[C f (x)]dx = C
f (x)dx.
Example 2:
Compute the following indenite integral.
[
1]
3
2x3 + 2 dx.
x
x
Solution:
Using our rules we have
[

3
1]
1
1
2x3 + 2 dx = 2 x3 dx + 3
dx

dx
x
x
x
x2
( 4)
( 1 )
x
x
=2
+3
ln x + C
4
1
x4 3
=
ln x + C.
2
x
Sometimes our rules need to be modied slightly due to operations with constants as is the case in the
following example.
Example 3:
Compute the following indenite integral:

e3x dx.
Solution:

169

www.ck12.org

d 3x
We rst note that our rule for integrating exponential functions does not work here since dx
e = 3e3x . However, if we remember to divide the original function by the constant then we get the correct antiderivative
and have

e3x dx =

e3x
+ C.
3

We can now re-state the rule in a more general form as

ekx dx =

ekx
+ C.
k

Dierential Equations
We conclude this lesson with some observations about integration of functions. First, recall that the
integration process allows us to start with function f from which we nd another function F(x) such that
F (x) = f (x). This latter equation is called a dierential equation. This characterization of the basic
situation for which integration applies gives rise to a set of equations that will be the focus of the Lesson
on The Initial Value Problem.
Example 4:
Solve the general dierential equation f (x) = x 3 +
2

x.

Solution:
We solve the equation by integrating the right side of the equation and have

3
f (x) =
f (x)dx =
x dx +
xdx.

1
We can integrate both terms using the power rule, rst noting that x = x 2 , and have

1
2
3 5 2 3
x 2 dx = x 3 + x 2 + C.
f (x) =
x 3 dx +
5
3

Lesson Summary
1.
2.
3.
4.
5.
6.

We
We
We
We
We
We

learned to nd antiderivatives of functions.

learned to represent antiderivatives.
interpreted constant of integration graphically.
solved general dierential equations.
used basic antidierentiation techniques to nd integration rules.
used basic integration rules to solve problems.

Multimedia Link
The following applet shows a graph, f (x) and its derivative, f (x). This is similar to other applets weve
explored with a function and its derivative graphed side-by-side, but this time f (x) is on the right, and
f (x) is on the left. If you edit the denition of f (x), you will see the graph of f (x) change as well. The
c parameter adds a constant to f (x). Notice that you can change the value of c without aecting f (x).
Why is this? Antiderivative Applet.
www.ck12.org

170

Review Questions
In problems #13, nd an antiderivative of the function
1. f (x) = 1 3x2 6x
2
2. f (x) = x x 3
5
3. f (x) = 2x + 1
In #47, nd the indenite integral

4. (2 + 5)dx
5. 2(x 3)3 dx

6. (x2 3 x)dx
)
(
7.
x + x41 x dx
8. Solve the dierential equation

f (x) = 4x3 3x2 + x 3

2x
9. Find the antiderivative F(x) of the
function f (x) = 2e + x 2 that satises F(0) = 5.
10. Evaluate the indenite integral |x|dx (Hint: Examine the graph of f (x) = |x|.)

Review Answers
F(x) = x x3 3x2 + C
2
5
F(x) = x2 35 x 3 + C
6
5
F(x) = 12
(2x + 1) 5 + C
(
)

2 + 5 dx = 2x + 5x + C

(x3)4
5. 2(x 3)3 dx = 2 + C
(
)
3 10
6.
x2 3 x dx = 10
x 3 +C
)
(
2
7.
x + x41 x dx = x2 44 + C
1.
2.
3.
4.

8. f (x) = x4 x3 + x2 3x + C
2
9. F(x) = e2x + xx 2x + 4

2
10. |x|dx = x2 + C

4.2 The Initial Value Problem

Learning Objectives
Find general solutions of dierential equations
Use initial conditions to nd particular solutions of dierential equations

Introduction
In the Lesson on Indenite Integrals Calculus we discussed how nding antiderivatives can be thought of as
nding solutions to dierential equations: F (x) = f (x). We now look to extend this discussion by looking
at how we can designate and nd particular solutions to dierential equations.

171

www.ck12.org

Lets recall that a general dierential equation will have an innite number of solutions. We will look at
one such equation and see how we can impose conditions that will specify exactly one particular solution.
Example 1:
Suppose we wish to solve the following equation:

f (x) = e3x 6 x.
Solution:
We can solve the equation by integration and we have
3
1
f (x) = e3x 4x 2 + C.
3

We note that there are an innite number of solutions. In some applications, we would like to designate
exactly one solution. In order to do so, we need to impose a condition on the function f. We can do this by
specifying the value of f for a particular value of x. In this problem, suppose that add the condition that
f (0) = 1. This will specify exactly one value of C and thus one particular solution of the original equation:
3

Substituting f (0) = 1 into our general solution f (x) = 13 e3x 4x 2 + C gives 1 = 13 e3(0) 4(0) 2 + C or
3
C = 1 13 = 23 . Hence the solution f (x) = 13 e3x 4x 2 + 32 is the particular solution of the original

equation f (x) = e3x 6 x satisfying the initial condition f (0) = 1.

We now can think of other problems that can be stated as dierential equations with initial conditions.
Consider the following example.
Example 2:
Suppose the graph of f includes the point (2, 6) and that the slope of the tangent line to f at any point x
is given by the expression 3x + 4. Find f (2).
Solution:
We can re-state the problem in terms of a dierential equation that satises an initial condition.
f (x) = 3x + 4 with f (2) = 6.
By integrating the right side of the dierential equation we have
3 2
x + 4x + C as the general solution. Substituting the condition that f (2) = 6 gives
2
3
6 = (2)2 + 4(2) + C,
2
6 = 6 + 8 + C,

f (x) =

C = 8.
Hence f (x) = 32 x2 + 4x 8 is the particular solution of the original equation f (x) = 3x + 4 satisfying
the initial condition f (2) = 6.
Finally, since we are interested in the value f (2), we put 2 into our expression for f and obtain:
f (2) = 10

Lesson Summary
1. We found general solutions of dierential equations.
2. We used initial conditions to nd particular solutions of dierential equations.
www.ck12.org

172

Multimedia Link
The following applet allows you to set the initial equation for f (x) and then the slope eld for that equation
is displayed. In magenta youll see one possible solution for f (x). If you move the magenta point to the
initial value, then you will see the graph of the solution to the initial value problem. Follow the directions
on the page with the applet to explore this idea, and then try redoing the examples from this section on
the applet. Slope Fields Applet.

Review Questions
In problems #13, solve the dierential equation for f (x).

1. f (x) = 2e2x 2 x
2. f (x) = sin x e1x

3. f (x) = (2 + x) x
In problems #47, solve the dierential equation for f (x) given the initial condition.
4.
5.
6.
7.
8.

f (x) = 6x5 4x2 + 73 and f (1) = 4

2 + e2x and f (0) = 3.
f (x) = 3x

f (x) = x2 x12 and f (1) = 3

( )

f (x) = (2 cos x sin x), 2 x 2 , and f 3 = 3 + 12

Suppose the graph of f includes the point (2, 4) and that the slope of the tangent line to f at x is
2x + 4. Find f (5).

In problems #910, nd the function f that satises the given conditions.

9. f (x) = sin x e2x with f (0) = 52 and f (0) = 0
10. f (x) = 1x with f (4) = 7 and f (4) = 25

Review Answers
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.

f (x) = 22x 43 x 2 + C
f (x) = cos x + e1x + C
8 52
4 72
f (x) = 15
x + 35
x +C
4 3
6
f (x) = x 3 x + 73 x + 2
2x
f (x) = x3 + e2 + 52

3
f (x) = 35 x5 + 1x + 75
f (x) = 2 sin x + cos x
f (x) = x2 + 4x + 16; f (5) = 11
f (x) = sin x 14 e2x + 4x + 14
3
f (x) = 43 x 2 + 3x + 73

173

www.ck12.org

4.3 The Area Problem

Learning Objectives
Use sigma notation to evaluate sums of rectangular areas
Find limits of upper and lower sums
Use the limit denition of area to solve problems

Introduction
In The Lesson The Calculus we introduced the area problem that we consider in integral calculus. The
basic problem was this:
f (x) = x2 . Suppose we are interested in nding the area between the xaxis and the curve of f (x) = x2 ,
from x = 0 to x = 1.

We approximated the area by constructing four rectangles, with the height of each rectangle equal to the
maximum value of the function in the sub-interval.

We then summed the areas of the rectangles as follows:

www.ck12.org

174

1
4
1
R2 =
4
1
R3 =
4
1
R4 =
4
R1 =

and R1 + R2 + R3 + R4 =

30
64

15
32

(1)

1
,
4
64
(1)
1
f
= ,
2
16
(3)
9
f
= ,
4
64
1
f (1) = ,
4
f

0.46.

We call this the upper sum since it is based on taking the maximum value of the function within each
sub-interval. We noted that as we used more rectangles, our area approximation became more accurate.
We would like to formalize this approach for both upper and lower sums. First we note that the lower
sums of the area of the rectangles results in R1 + R2 + R3 + R4 = 13/64 0.20 Our intuition tells us that
the true area lies somewhere between these two sums, or 0.20 < Area < 0.46 and that we will get closer
to it by using more and more rectangles in our approximation scheme.
In order to formalize the use of sums to compute areas, we will need some additional notation and terminology.
Sigma Notation
In The Lesson The Calculus we used a notation to indicate the upper sum when we increased our rectangles

195
to N = 16 and found that our approximation A = 16
1 Ri = 512 0.38 . The notation we used to enabled
us to indicate the sum without the need to write out all of the individual terms. We will make use of this
notation as we develop more formal denitions of the area under the curve.

Lets be more precise with the notation. For example, the quantity A = Ri was found by summing the
areas of N = 16 rectangles. We want to indicate this process, and we can do so by providing indices to the
symbols used as follows:
A=

16

Ri = R1 + R2 + R3 + ... + R15 + R16 .

i=1

The sigma symbol with these indices tells us how the rectangles are labeled and how many terms are in
the sum.
Useful Summation Formulas
We can use the notation to indicate useful formulas that we will have occasion to use. For example, you
may recall that the sum of the rst n integers is n(n + 1)/2. We can indicate this formula using sigma
notation. The formula is given here along with two other formulas that will become useful to us.
n

i=

i=1

n(n + 1)
,
2

n(n + 1)(2n + 1)
,
6
i=1
[
]
n

n(n + 1) 2
i3 =
.
2
i=1
i2 =

We can show from associative, commutative, and distributive laws for real numbers that

175

www.ck12.org

n
i=1 (ai + bi ) = i=1 (ai )
n
n
i=1 (kai ) = k i=1 (ai ).

i=1 (bi )

and

Example 1:
Compute the following quantity using the summation formulas:
10

2i(i 6i).

i=1

Solution:
10

i=1

2i(i 6i) =

10
10
10

(2i2 12i) = 2
i2 12
i
i=1

i=1

i=1

)
)
(
(10)(10 + 1)(2 10 + 1)
(10)(11)
=2
12
6
2
= 770 660 = 110.

Another Look at Upper and Lower Sums

We are now ready to formalize our initial ideas about upper and lower sums.
Let f be a bounded function in a closed interval [a, b] and P = [x0 , . . . , xn ] the partition of [a, b] into n
subintervals.
We can then dene the lower and upper sums, respectively, over partition P, by
S (P) =
T (P) =

1
n

mi (xi xi1 ) = m1 (x1 x0 ) + m2 (x2 x1 ) + . . . + mn (xn xn1 ),

Mi (xi xi1 ) = M1 (x1 x0 ) + M2 (x2 x1 ) + . . . + Mn (xn xn1 ).

where mi is the minimum value of f in the interval of length xi xi1 and Mi is the maximum value of f
in the interval of length xi xi1 .
The following example shows how we can use these to nd the area.
Example 2:
Show that the upper and lower sums for the function f (x) = x2 , from x = 0 to x = 1, approach the value
A = 1/3.
Solution:
Let P be a partition of n equal sub intervals over [0, 1]. We will show the result for the upper sums. By
our denition we have
T (P) =

Mi (xi xi1 ) = M1 (x1 x0 ) + M2 (x2 x1 ) + . . . + Mn (xn xn1 ).

We note that each rectangle will have width 1n , and lengths

www.ck12.org

176

( )2 ( )2 ( )2
( )2
1
2
3
n
,
,
,
.
.
.
,
as indicated:
n
n
n
n

T (P) =

Mi (xi xi1 ) = M1 (x1 x0 ) + M2 (x2 x1 ) + . . . + Mn (xn xn1 )

1 ( 1 )2 1 ( 2 )2 1 ( 3 )2
1 ( n )2
=
+
+
+ ... +
n n
n n
n n
n n
1 ( 1 )2 2
=
(1 + 22 + 32 + . . . + n2 )
n n
(1)
( 1 ) ( n(n + 1)(2n + 1) ) ( (n + 1)(2n + 1) )
2
2
2
2
= 3 (1 + 2 + 3 + . . . + n ) = 3
=
.
6
n
n
6n2

We can re-write this result as:

(n + 1)(2n + 1)
1 ( n + 1 ) ( 2n + 1 ) 1 (
1) (
1)
=
=
1
+
2
+
.
6
n
n
6
n
n
6n2
We observe that as
x +,

1) (
1)
1
1(
1+
2+
.
6
n
n
3

We now are able to dene the area under a curve as a limit.

Denition Let f be a continuous function on a closed interval [a, b]. Let P be a partition of n equal sub
intervals over [a, b]. Then the area under the curve of f is the limit of the upper and lower sums, that
is

A = lim S (P) = lim T (P).

n+

n+

Example 3:
Use the limit denition of area to nd the area under the function f (x) = 4 x from 1 to x = 3.

177

www.ck12.org

Solution:
If we partition the interval [1, 3] into n equal sub-intervals, then each sub-interval will have length
and height 3 ix as i varies from 1 to n. So we have x = 2n and
S (P) =

2
n

n
n
n

(3 ix)x =
(3x)
i(x)2
1

= (3x)n
Since x =

31
n

n(n + 1)
(x)2 .
n

2
n

, we then have by substitution

(
)
n(n+1)
(3x)n n (x)2 = 6 2 + 2n = 4 + 2n 4 as n . Hence the area is A = 4.
This example may also be solved with simple geometry. It is left to the reader to conrm that the two
methods yield the same area.

Lesson Summary
1. We used sigma notation to evaluate sums of rectangular areas.
2. We found limits of upper and lower sums.
3. We used the limit denition of area to solve problems.

Review Questions
In problems #12 , nd the summations.

1. 10
i(2i 3)
i=1
n
2. i=1 (3 i)(2 + i)
In problems #35, nd S (P) and T (P) under the partition P.
}
{
3. f (x) = 1 x2 , P = 0, 12 , 1, 23 , 2
{
}
4. f (x) = 2x2 , P = 1, 12 , 0, 12 , 1
5. f (x) = 1x , P = {4, 3, 2, 1}
www.ck12.org

178

In problems #68, nd the area under the curve using the limit denition of area.
6. f (x) = 3x + 5 from x = 2 to x = 6.
7. f (x) = x2 from x = 1 to x = 3.
8. f (x) = 1x from x = 1 to x = 4.
In problems #910, state whether the function is integrable in the given interval. Give a reason for your
answer.
9. f (x) = |x
2| on the interval [1, 4]

1
if x is rational

on the interval [0, 1]

10. f (x) =

1 if x is irrational

Review Answers
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.

10
i(2i 3) = 605
i=1
n
1
2
i=1 (3 i)(2 + i) = 3 (19 n )
S (P) = 1.75; T (P) = 0.25(note that we have included areas under the xaxis as negative values.)
S (P) = 0.5; T (P) = 2.5
S (P) = 1.83; T (P) = 1.08
Area = 68
Area = 26
3
Area = 15
16
Yes, since f (x) = |x 2| is continuous on [1, 4]
No, since S (P) = 1; T (P) = 1

4.4 Denite Integrals

Learning Objectives
Use Riemann Sums to approximate areas under curves
Evaluate denite integrals as limits of Riemann Sums

Introduction
In the Lesson The Area Problem we dened the area under a curve in terms of a limit of sums.
A = lim S (P) = lim T (P)
n+

n+

where
S (P) =
T (P) =

1
n

mi (xi xi1 ) = m1 (x1 x0 ) + m2 (x2 x1 ) + . . . + mn (xn xn1 ),

Mi (xi xi1 ) = M1 (x1 x0 ) + M2 (x2 x1 ) + . . . + Mn (xn xn1 ),

179

www.ck12.org


S (P), and T (P) were examples of Riemann Sums. In general, Riemann Sums are of form ni=1 f (xi )x
where each xi is the value we use to nd the length of the rectangle in the ith sub-interval. For example, we
used the maximum function value in each sub-interval to nd the upper sums and the minimum function
in each sub-interval to nd the lower sums. But since the function is continuous, we could have used any
points within the sub-intervals to nd the limit. Hence we can dene the most general situation as follows:
Denition If f is continuous on [a, b], we divide the interval [a, b] into n sub-intervals of equal width with

x = ba
n . We let x0 = a, x1 , x2 , . . . , xn = b be the endpoints of these sub-intervals and let x1 , x2 , . . . , xn
be any sample points in these sub-intervals. Then the denite integral of f from x = a to x = b is

f (x)dx = lim

f (xi )x.

i=1

Example 1:
Evaluate the Riemann Sum for f (x) = x3 from x = 0 to x = 3 using n = 6 sub-intervals and taking the
sample points to be the midpoints of the sub-intervals.
Solution:
If we partition the interval [0, 3] into n = 6 equal sub-intervals, then each sub-interval will have length
1
1
30
6 = 2 . So we have x = 2 and

R6 =

f (xi )x = f (0.25)x + f (0.75)x + f (1.25)x + f (1.75)x + f (2.25)x + f (2.75)x

( 1 ) ( 1 ) ( 27 ) ( 1 ) ( 125 ) ( 1 ) ( 343 ) ( 1 ) ( 729 ) ( 1 ) ( 1331 ) ( 1 )

=
+
+
+
+
+
64 2
64 2
64 2
64 2
64 2
64
2
2556
=
= 39.93.
64
Now lets compute the denite integral using our denition and also some of our summation formulas.
Example 2:
Use the denition of the denite integral to evaluate

3
0

x3 dx.

Solution:
Applying our denition, we need to nd

x3 dx = lim
0

www.ck12.org

i=1

180

f (xi )x.

We will use right endpoints to compute the integral. We rst need to divide [0, 3] into n sub-intervals of
3
3
6
3i
length x = 30
n = n . Since we are using right endpoints, x0 = 0, x1 = n , x2 = n , . . . xi = n .
3

n 3
So 0 x3 dx = limn ni=1 f ( 3in )( 3n ) = limn 3n ni=1 ( 3in )3 = limn 3n ni=1 ( 27
)i3 = limn 81
i=1 i .
n3
n4
[
]
2

n(n+1)
Recall that n1 i3 =
. By substitution, we have
2

[
]2
81 n(n + 1)
81 [
1 ]2
81
x dx = lim = 4
= lim
1+

as n .
n
n 4
2
n
4
n
3

Hence

x3 dx =
0

81
.
4

Before we look to try some problems, lets make a couple of observations. First, we will soon not need to
rely on the summation formula and Riemann Sums for actual computation of denite integrals. We will
develop several computational strategies in order to solve a variety of problems that come up. Second,
the idea of denite integrals as approximating the area under a curve can be a bit confusing since we
may sometimes get results that
3 do not make sense when interpreted as areas. For example, if we were to
compute the denite integral 3 x3 dx, then due to the symmetry of f (x) = x3 about the origin, we would
3
nd that 3 x3 dx = 0. This is because for every sample point xj , we also have xj is also a sample point
3
with f (xj ) = f (xj ). Hence, it is more accurate to say that 3 x3 dx gives us the net area between x = 3
and
3 x = 3. If81 we wanted the total area bounded by the graph and the xaxis, then we would compute
2 0 x3 dx = 2 .

Lesson Summary
1. We used Riemann Sums to approximate areas under curves.
2. We evaluated denite integrals as limits of Riemann Sums.

Multimedia Link
For video presentations on calculating denite integrals using Riemann Sums (13.0), see Riemann Sums,
Part 1 (6:15) and Riemann Sums, Part 2 (8:32) .

Figure 4.1: Calculating a Denite Integral Using Riemann Sums - Part 1 (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/530
The following applet lets you explore Riemann Sums of any function. You can change the bounds and the
number of partitions. Follow the examples given on the page, and then use the applet to explore on your

181

www.ck12.org

Figure 4.2: Calculating a Denite Integral Using Riemann Sums - Part 2 (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/531
own. Riemann Sums Applet. Note: On this page the author uses Left- and Right- hand sums. These
are similar to the sums S (P) and T (P) that you have learned, particularly in the case of an increasing
(or decreasing) function. Left-hand and Right-hand sums are frequently used in calculations of numerical
integrals because it is easy to nd the left and right endpoints of each interval, and much more diicult
to nd the max/min of the function on each interval. The dierence is not always important from a
numerical approximation standpoint; ss you increase the number of partitions, you should see the Lefthand and Right-hand sums converging to the same value. Try this in the applet to see for yourself.

Review Questions
In problems #17 , use Riemann Sums to approximate the areas under the curves.
1. Consider f (x) = 2 x from x = 0 to x = 2. Use Riemann Sums with four subintervals of equal lengths.
Choose the midpoints of each subinterval as the sample points.
2. Repeat problem #1 using geometry to calculate the exact area of the region under the graph of
f (x) = 2 x from x = 0 to x = 2. (Hint: Sketch a graph of the region and see if you can compute its
area using area measurement formulas from geometry.)
3. Repeat problem #1 using the denition of the denite integral to calculate the exact area of the
region under the graph of f (x) = 2 x from x = 0 to x = 2.
4. f (x) = x2 x from x = 1 to x = 4. Use Riemann Sums with ve subintervals of equal lengths. Choose
the left endpoint of each subinterval as the sample points.
5. Repeat problem #4 using the denition of the denite intergal to calculate the exact area of the
region under the graph of f (x) = x2 x from x = 1 to x = 4.
6. Consider f (x) = 3x2 . Compute the Riemann Sum of f on [0, 1] under each of the following situations.
In each case, use the right endpoint as the sample points.
(a)
(b)
(c)
(d)

Two sub-intervals of equal length.

Five sub-intervals of equal length.
Ten sub-intervals of equal length.
Based on your answers above, try to guess the exact area under the graph of f on [0, 1].

7. Consider f (x) = e x . Compute the Riemann Sum of f on [0, 1] under each of the following situations.
In each case, use the right endpoint as the sample points.
(a)
(b)
(c)
(d)

Two sub-intervals of equal length.

Five sub-intervals of equal length.
Ten sub-intervals of equal length.
Based on your answers above, try to guess the exact area under the graph of f on [0, 1].

8. Find the net area under the graph of f (x) = x3 x; x = 1 to x = 1. (Hint: Sketch the graph and
check for symmetry.)
www.ck12.org

182

9. Find the total area bounded by the graph of f (x) = x3 x and the xaxis, from to x = 1 to x = 1.
10. Use your knowledge of geometry to evaluate the following denite integral:
3

9 x2 dx (Hint: set y =
0
geometry.)

9 x2 and square both sides to see if you can recognize the region from

Review Answers
1.
2.
3.
4.
5.
6.
7.

Area = 2
Area = 2
Area = 2
Area = 10.08
Area = 15.5
(a)
(b)
(c)
(d)

Area = 1.875
Area = 1.32
Area = 1.15
Area = 1

8.
9. (a) Area = 2.18
(b) Area = 1.89
(c) Area = 1.80
(d) Area = e1 1 1.71
10. The graph is symmetric about the origin; hence Area = 0.
11. Area = 12
12. The graph is that of a quarter circle of radius 3; hence Area =

9
4 .

4.5 Evaluating Denite Integrals

Learning Objectives
Use antiderivatives to evaluate denite integrals
Use the Mean Value Theorem for integrals to solve problems
Use general rules of integrals to solve problems

Introduction
In the Lesson on Denite Integrals, we evaluated denite integrals using the limit denition. This process
was long and tedious. In this lesson we will learn some practical ways to evaluate denite integrals. We
begin with a theorem that provides an easier method for evaluating denite integrals. Newton discovered
this method that uses antiderivatives to calculate denite integrals.
Theorem:
If f is continuous on the closed interval [a, b], then

f (x)dx = F(b) F(a),

183

www.ck12.org

where F is any antiderivative of f.

We sometimes use the following shorthand notation to indicate
b
]b
f (x)dx = F(x) .

b
a

f (x)dx = F(b) F(a) :

The proof of this theorem is included at the end of this lesson. Theorem 4.1 is usually stated as a part of
the Fundamental Theorem of Calculus, a theorem that we will present in the Lesson on the Fundamental
Theorem of Calculus. For now, the result provides a useful and eicient way to compute denite integrals.
We need only nd an antiderivative of the given function in order to compute its integral over the closed
interval. It also gives us a result with which we can now state and prove a version of the Mean Value
Theorem for integrals. But rst lets look at a couple of examples.
Example 1:
Compute the following denite integral:

x3 dx.
0

Solution:
Using the limit denition we found that

3
0

x3 dx =

81
4 .

We now can verify this using the theorem as follows:

We rst note that x4 /4 is an antiderivative of f (x) = x3 . Hence we have

]3
3
x4
81 0
81
3
x dx =
=
= .
4 0
4
4
4
0
We conclude the lesson by stating the rules for denite integrals, most of which parallel the rules we stated
for the general indenite integrals.
a
f (x)dx = 0
a
b
a
f (x)dx =
f (x)dx
a
b
b
b
k f (x)dx = k
f (x)dx
a
a
b
b
b
[ f (x) g(x)]dx =
f (x)dx
g(x)dx
a
a
a
b
c
b
f (x)dx =
f (x)dx +
f (x)dx where a < c < b.
a

Given these rules together with Theorem 4.1, we will be able to solve a great variety of denite integrals.
Example 2:
2

Compute 2 (x x)dx.
Solution:
4

(x

x)dx =
1

xdx
1

x2 ]4 2 3 ]4 (
1) 2
15 14
17
2
xdx =
= 8
(8 1) =

= .
x
2 1 3
2
3
2
3
6
1

Example 3:
www.ck12.org

184

Compute

(x + cos x)dx.

Solution:

(x + cos x)dx =

(x)dx =
0

2
2 + 4
x2 ] 2 sin x ] 2
+
=
+1=
.
2 0
1 0
4
4

(cos x)dx =
0

Lesson Summary
1. We used antiderivatives to evaluate denite integrals.
2. We used the Mean Value Theorem for integrals to solve problems.
3. We used general rules of integrals to solve problems.
Proof of Theorem 4.1
We rst need to divide [a, b] into n sub-intervals of length x =
the endpoints of these sub-intervals.

. We let x0 = a, x1 , x2 , . . . , xn = b be

ba
n

Let F be any antiderivative of f.

Consider F(b) F(a) = F(xn ) F(x0 ).
We will now employ a method that will express the right side of this equation as a Riemann Sum. In
particular,
F(b) F(a) = F(xn ) F(x0 )
= F(xn ) F(xn1 ) + F(xn+1 ) F(xn2 ) + F(xn2 ) . . . F(x1 ) + F(x1 )
n

=
[F(xi ) F(xi1 )].
1

Note that F is continuous. Hence, by the Mean Value Theorem, there exist ci [xi 1, xi ]
such that F(xi ) F(xi1 ) = F (ci )(xi xi1 ) = f (ci )x.
Hence
F(b) F(a) =

F (ci )(xi xi1 ) =

f (ci )x.

Taking the limit of each side as n we have

lim [F(b) F(a)] = lim

f (ci )x.

We note that the left side is a constant and the right side is our denition for

b
a

f (x)dx.

Hence
F(b) F(a) = lim

f (ci )x =

f (x)dx.
a

Proof of Theorem 4.2

x
Let F(x) = a f (x)dx.

185

www.ck12.org

By the Mean Value Theorem for derivatives, there exists c [a, b] such that
F (c) =

F(b) F(a)
.
ba

From Theorem 4.1 we have that F is an antiderivative of f. Hence, F (x) = f (x) and in particular,
F (c) = f (c). Hence, by substitution we have
f (c) =
Note that F(a) =

a
a

F(b) F(a)
.
ba

f (x)dx = 0. Hence we have

f (c) =

F(b) 0
F(b)
=
,
ba
ba

and by our denition of F(x) we have

1
1
f (c) =
F(b) =
ba
ba

f (x)dx.
a

This theorem allows us to nd for positive functions a rectangle that has base [a, b] and height f (c) such
b
that the area of the rectangle is the same as the area given by a f (x)dx. In other words, f (c) is the average
function value over [a, b].

Review Questions
In problems #18, use antiderivatives to compute the denite integral.
1.
2.
3.
4.
5.
6.

( 3x )dx
4
1
(t t2 )dt
05 1
( + 1 )dx
2
21 x
2
4(x 1)(x2 + 1)dx
08 4
( + x2 + x)dx
24 x
(e3x )dx
24 2
dx
1 x+3

7.

8. Find the average value of f (x) = x over [1, 9].

4
9. If f is continuous and 1 f (x)dx = 9, show that f takes on the value 3 at least once on the interval
[1, 4].
10. Your friend states that there is no area under the curve of f (x) = sin x on [0, 2] since he computed
2
sin xdx = 0. Is he correct? Explain your answer.
0

Review Answers
1.
2.
3.

( 3x )dx = 6
4
1
1
(t t2 )dt = 30
05 1
( x + 1 )dx =
2
2

www.ck12.org

2 52 2+

3 2
2

186

4.
5.
6.

1
08
24
24

4(x2 1)(x2 + 1)dx = 16

5
( 4x + x2 + x)dx =
(e3x )dx =

9417
48

196.19

e12 e6
3

2
7. 1 x+3
dx = 2 ln 7 2 ln 4
13
8. 6
9. Apply the Mean Value Theorem for integrals.
2
10. He is partially correct. The denite integral 0 sin x dx computes the net area under the curve.
the x-axis is given] by
However, the area between the curve and

A=2
sin x dx = cos x = 2.
0

4.6 The Fundamental Theorem of Calculus

Learning Objectives
Use the Fundamental Theorem of Calculus to evaluate denite integrals

Introduction
In the Lesson on Evaluating Denite Integrals, we evaluated denite integrals using antiderivatives. This
process was much more eicient than using the limit denition. In this lesson we will state the Fundamental
Theorem of Calculus and continue to work on methods for computing denite integrals.
Fundamental Theorem of Calculus:
Let f be continuous on the closed interval [a, b].
x
1. If function F is dened by F(x) = a f (t)dx, on [a, b] , then F (x) = f (x) on [a, b].
2. If g is any antiderivative of f on [a, b], then

f (t)dt = g(b) g(a).

We rst note that we have already proven part 2 as Theorem 4.1. The proof of part 1 appears at the end
of this lesson.
Think about this Theorem. Two of the major unsolved problems in science and mathematics turned
out to be solved by calculus which was invented in the seventeenth century. These are the ancient problems:
1. Find the areas dened by curves, such as circles or parabolas.
2. Determine an instantaneous rate of change or the slope of a curve at a point.
With the discovery of calculus, science and mathematics took huge leaps, and we can trace the advances
of the space age directly to this Theorem.
Lets continue to develop our strategies for computing denite integrals. We will illustrate how to solve
the problem of nding the area bounded by two or more curves.
Example 1:
Find the area between the curves of f (x) = x and g(x) = x3 .

187

www.ck12.org

Solution:
We rst observe that there are no limits of integration explicitly stated here. Hence we need to nd the
limits by analyzing the graph of the functions.

We observe that the regions of interest are in the rst and third quadrants from x = 1 to x = 1. We also
observe the symmetry of the graphs about the origin. From this we see that the total area enclosed is
1
1
[ 1
]
1
1
[1 1]
[1] 1

x2 
4 
x
(x x3 )dx = 2
xdx
x3 dx = 2   = = 2
2
=2
= .
2
4
2 4
4
2
0
0
0
0

Example 2:
Find the area between the curves of f (x) = |x 1| and the xaxis from x = 1 to x = 3.
Solution:
We observe from the graph that we will have to divide the interval [1, 3] into subintervals [1, 1] and [1, 3].

Hence the area is given by

( 2
) +1 ( 2
) +3
1
3


x
x
(x + 1)dx +
(x 1)dx = + x  +
x  = 2 + 2 = 4.
1
+1
2
2
1
1
Example 3:
Find the area enclosed by the curves of f (x) = x2 + 2x + 1 and
g(x) = x2 2x + 1.
www.ck12.org

188

Solution:
The graph indicates the area we need to focus on.

x3
(x 2x + 1)dx
(x + 2x + 1)dx = x2 + x
3
2
2
2

) 0
( 2
) 0


x
10 2
8
3
x + x  =
= .
 +
2
3
3
3
3
2

Before providing another example, lets look back at the rst part of the Fundamental Theorem. If function
x
F is dened by F(x) = a f (t)dt, on [a, b] then F (x) = f (x) on [a, b]. Observe that if we dierentiate the
integral with respect to x, we have
x
d
f (t)dt = F (x) = f (x).
dx a
This fact enables us to compute derivatives of integrals as in the following example.
Example 4:
Use the Fundamental Theorem to nd the derivative of the following function:
x

3
(1 + t)dt.
g(x) =
0

Solution:
While we could easily integrate the right side and then dierentiate, the Fundamental Theorem enables us
to nd the answer very routinely.
x

d
3
(1 + t)dt = 1 + 3 x.
g (x) =
dx 0
This application of the Fundamental Theorem becomes more important as we encounter functions that
may be more diicult to integrate such as the following example.
Example 5:
Use the Fundamental Theorem to nd the derivative of the following function:
x
g(x) =
(t2 cos t)dt.
2

Solution:
In this example, the integral is more diicult to evaluate. The Fundamental Theorem enables us to nd
the answer routinely.
x
d

g (x) =
(t2 cos t)dt = x2 cos x.
dx 2

189

www.ck12.org

Lesson Summary
1. We used the Fundamental Theorem of Calculus to evaluate denite integrals.
Fundamental Theorem of Calculus
Let f be continuous on the closed interval [a, b].
x
1. If function F is dened by F(x) = a f (t)dt, on [a, b], then F (x) = f (x), on [a, b].
2. If g is any antiderivative of f on [a, b], then
b
f (t)dt = g(b) g(a).
a

We rst note that we have already proven part 2 as Theorem 4.1.

Proof of Part 1.
1. Consider F(x) =

x
a

f (t)dt, on [a, b].

2. x, c [a, b], c < x.

x
c
x
Then a f (t)dt = a f (t)dt + c f (t)dt by our rules for denite integrals.
x
c
x
x
3. Then a f (t)dt a f (t)dt = c f (t)dt . Hence F(x) F(c) = c f (t)dt.
4. Since f is continuous on [a, b] and x, c [a, b], c < x then we can select u, v [c, v] such that f (u) is the
minimum value of and f (v) is the maximum value of f in [c, x]. Then we can consider f (u)(x c) as a lower
sum and f (v)(x c) as an upper sum of f from c to x. Hence
x
5. f (u)(x c) c f (t)dt f (v)(x c).
6. By substitution, we have:
f (u)(x c) F(x) F(c) f (v)(x c).
7. By division, we have
f (u)

F(x) F(c)
f (v).
xc

8. When x is close to c, then both f (u) and f (v) are close to f (c) by the continuity of f
9. Hence lim xc+
f (c).

F(x)F(c)
xc

= f (c). Similarly, if x < c, then lim xc

F(x)F(c)
xc

= f (c). Hence, lim xc

F(x)F(c)
xc

10. By the denition of the derivative, we have that

F (c) = lim xc

F(x)F(c)
xc

= f (c) for every c [a, b]. Thus, F is an antiderivative of f on [a, b].

Multimedia Link
For a video presentation of the Fundamental Theorem of Calculus (15.0), see Fundamental Theorem of
Calculus, Part 1 (9:26) .

Review Questions
In problems #14, sketch the graph of the function f (x) in the interval [a, b]. Then use the Fundamental
Theorem of Calculus to nd the area of the region bounded by the graph and the xaxis.
www.ck12.org

190

Figure 4.3: Fundamental Theorem of Calculus Part 1 (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/532
1.
2.
3.
4.

f (x) = 2x + 3, [0, 4]
f (x) = e x , [0, 2]
f (x) = x2 + x, [1, 3]
f (x) = x2 x, [0, 2]

(Hint: Examine the graph of the function and divide the interval accordingly.)
In problems #57 use antiderivatives to compute the denite integral.
5.
6.

+1
1
3
0

|x|dx

|x3 2|dx

(Hint: Examine the graph of the function and divide the interval accordingly.)
7.

+4
2

[|x 1| + |x + 1|] dx

(Hint: Examine the graph of the function and divide the interval accordingly.)
In problems #810, nd the area between the graphs of the functions.

8. f (x) = x, g(x) = x, [0, 2]

9. f (x) = x2 , g(x) = 4, [0, 2]
10. f (x) = x2 + 1, g(x) = 3 x, [0, 3]

Review Answers
1. Area = 28

191

www.ck12.org

2. Area = e2 1

3. Area =

38
3

4. Area = 1

5.

+1
1

|x|dx = 1

3 2
3

3
3 2|dx =
|x
(2 x3 )dx + 3 2 (x3 2)dx = 3 2 +
0 [
0
]
+4
7. 2 |x 1| + |x + 1| dx = 22
6.

8. Area =

www.ck12.org

74 2
3

192

57
4

9. Area =

16
3

10. Area =

10
3

4.7 Integration by Substitution

Learning Objectives
Integrate composite functions
Use change of variables to evaluate denite integrals
Use substitution to compute denite integrals

Introduction
In this lesson we will expand our methods for evaluating denite integrals. We rst look at a couple of
situations where nding antiderivatives requires special methods. These involve nding antiderivatives of

193

www.ck12.org

composite functions and nding antiderivatives of products of functions.

Antiderivatives of Composites
Suppose we needed to compute the following integral:

3x2 1 + x3 dx.
Our rules of integration are of no help here. We note that the integrand is of the form f (g(x)) g (x) where

g(x) = 1 + x3 and f (x) = x.

Since we are looking for an antiderivative F of f, and we know that F = f, we can re-write our integral as

3
2
1 + x3 3x2 dx = ( 1 + x3 ) 2 + C.
3
In practice, we use the following substitution scheme to verify that we can integrate in this way:
1. Let u = 1 + x3 .
2. Dierentiate both sides so du = 3x2 dx.
3.
Change the original
integral in x to an integral in u:
2
3
1 + x 3x dx =
udu, where u = 1 + x3 and du = 3x2 dx.
4. Integrate with respect to u:

1
2 3
udu =
u 2 du = u 2 + C.
3
5. Change the answer back to x:

3
2 3
2
udu = u 2 + C = ( 1 + x3 ) 2 + C.
3
3

While this method of substitution is a very powerful method for solving a variety of problems, we will
nd that we sometimes will need to modify the method slightly to address problems, as in the following
example.
Example 1:
Compute the following indenite integral:

x2 e x dx.
Solution:

We note that the derivative of x3 is 3x2 ; hence, the current problem is not of the form F (g(x)) g (x)dx.
But we notice that the derivative is o only by a constant of 3 and we know that constants are easy to
deal with when dierentiating and integrating. Hence
Let u = x3 .
Then du = 3x2 dx.
Then 31 du = x2 dx. and we are ready to change the original integral from x to an integral in u and integrate:

(1 ) 1
3
1
eu du = eu + C.
x2 e x dx =
eu du =
3
3
3
www.ck12.org

194

Changing back to x, we have

3
1 3
x2 e x dx = e x + C.
3

We can also use this substitution method to evaluate denite integrals. If we attach limits of integration
to our rst example, we could have a problem such as
4
1 + x3 3x2 dx.
1

The method still works. However, we have a choice to make once we are ready to use the Fundamental
Theorem to evaluate the integral.


Recall that we found that
1 + x3 3x2 dx =
udu for the indenite integral. At this point, we could
evaluate the integral by changing the answer back to x or we could evaluate the integral in u. But we need
to be careful. Since the original limits of integration were in x, we need to change the limits of integration
for the equivalent integral in u. Hence,
4
65
1 + x3 3x2 dx = u=2 udu, where u = 1 + x3
1
u=65
4
65

2 3 
2
2
3
1 + x 3x dx =
udu = u 2 
= ( 653 8).
3 u=2
3
1
u=2
Integrating Products of Functions

We are not able to state a rule for integrating products of functions,

f (x)g(x)dx but we can get a
relationship that is almost as eective. Recall how we dierentiated a product of functions:
d
f (x)g(x) = f (x)g (x) + g(x) f (x).
dx
So by integrating both sides we get

[ f (x)g (x) + g(x) f (x)]dx = f (x)g(x), or

f (x)g (x)dx = f (x)g(x) g(x) f (x).

In order to remember the formula, we usually write it as

udv = uv vdu.
We refer to this method as integration by parts. The following example illustrates its use.
Example 2:
Use integration by parts method to compute

xe x dx.
Solution:
We note that our other substitution method is not applicable here. But our integration by parts method
will enable us to reduce the integral down to one that we can easily evaluate.

195

www.ck12.org

Let u = x and dv = e x dx then du = dx and v = e x

By substitution, we have

xe dx = xe
x

e x dx.

We can easily evaluate the integral and have

x
x
xe dx = xe e x dx = xe x e x + C.
And should we wish to evaluate denite integrals, we need only to apply the Fundamental Theorem to the
antiderivative.

Lesson Summary
1. We integrated composite functions.
2. We used change of variables to evaluate denite integrals.
3. We used substitution to compute denite integrals.

Review Questions
Compute the integrals in problems #110.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.

x ln xdx
3
x ln xdx
1 x

dx
1 2x+1

x3 1 x2 dx
0
x cos xdx
1
x2 x3 + 9dx
0 ( 1
1)
x dx

e
2
x 2
3
x
xlnxe dx
5 dx
ex12
dx
1 x

Review Answers
1.
2.
3.
4.
5.
6.
7.

x2 (2 ln x1)

x ln xdx =
+C
4(
)
3
3
3
2
2
2
2
2 +
x
ln
xdx
=
ln
3

3
3
3
3
1 x
(x1) 2x+1

dx =
+c
3
1 2x+1

2
3
x 1 x2 dx = 15
0
x cos xdx = x sin x + cos x + c
[ 3
]
1
x2 x3 + 9dx = 29 10 2 27
0
)
(1
1
1
x dx = e x + c

e
2
x

www.ck12.org

196

8.
9.
10.

x3 e x dx = 12 e x (x2 1) + c
lnx
5

ex12

dx =

dx
1 x

4ln x
3

3x 2

4
3

+c

9x 2

=1

4.8 Numerical Integration

Learning Objectives

Use the Trapezoidal Rule to solve problems

Estimate errors for the Trapezoidal Rule
Use Simpsons Rule to solve problems
Estimate Errors for Simpsons Rule

Introduction
Recall that we used dierent ways to approximate the value of integrals. These included Riemann Sums
using left and right endpoints, as well as midpoints for nding the length of each rectangular tile. In this
lesson we will learn two other methods for approximating integrals. The rst of these, the Trapezoidal
Rule, uses areas of trapezoidal tiles to approximate the integral. The second method, Simpsons Rule, uses
parabolas to make the approximation.
Trapezoidal Rule
Lets recall how we would use the midpoint rule with n = 4 rectangles to approximate the area under the
graph of f (x) = x2 + 1 from x = 0 to x = 1.

If instead of using the midpoint value within each sub-interval to nd the length of the corresponding
rectangle, we could have instead formed trapezoids by joining the maximum and minimum values of the
function within each sub-interval:

197

www.ck12.org

h(b +b )

The area of a trapezoid is A = 12 2 , where b1 and b2 are the lengths of the parallel sides and h is the
height. In our trapezoids the height is x and b1 and b2 are the values of the function. Therefore in nding
the areas of the trapezoids we actually average the left and right endpoints of each sub-interval. Therefore
a typical trapezoid would have the area
A=
b

To approximate

a
b

x
( f (xi1 ) + f (xi )) .
2

f (x)dx with n of these trapezoids, we have

n

1
f (xi1 )x +
f (xi )x
f (x)dx
2 i1
i1
x
[ f (x0 ) + f (x1 ) + f (x1 ) + f (x2 ) + f (x2 ) + . . . + f (xn1 ) f (xn )]
=
2
ba
x
[ f (x0 ) + 2 f (x1 ) + 2 f (x2 ) + . . . + 2 f (xn1 ) f (xn )] , x =
.
=
2
n

Example 1:
Use the Trapezoidal Rule to approximate

3
0

x2 dx with n = 6.

Solution:
We nd x =

ba
n

1
= 30
6 = 2.
3
]
1[
f (0) + 2 f ( 12 ) + 2 f (1) + 2 f ( 32 ) + 2 f (2) + 2 f ( 52 ) + f (3)
x2 dx
4
0
]
1[
0 + (2 14 ) + (2 1) + (2 94 ) + (2 4) + (2 25
)+9
=
4
4
1 [ 73 ] 73
=
=
= 9.125.
4 2
8

Of course, this estimate is not nearly as accurate as we would like. For functions such as f (x) = x2 , we can
3
3

x3
2
easily nd an antiderivative with which we can apply the Fundamental Theorem that 0 x dx = 3 = 9.
0

But it is not always easy to nd an antiderivative. Indeed, for many integrals it is impossible to nd
an antiderivative. Another issue concerns the questions about the accuracy
3 of the approximation. In
particular, how large should we take n so that the Trapezoidal Estimate for 0 x2 dx is accurate to within
a given value, say 0.001? As with our Linear Approximations in the Lesson on Approximation Errors, we
can state a method that ensures our approximation to be within a specied value.
www.ck12.org

198

Error Estimates for Simpsons Rule

We would like to have condence in the approximations we make. Hence we can choose n to ensure that the
errors are within acceptable boundaries. The following method illustrates how we can choose a suiciently
large n.
Suppose | f (x)| k for a x b. Then the error estimate is given by
|ErrorT rapezoidal |
Example 2:
Find n so that the Trapezoidal Estimate for

3
0

k(b a)3
.
12n2

x2 dx is accurate to 0.001.

Solution:
We need to nd n such that |ErrorT rapezoidal | 0.001. We start by noting that | f (x)| = 2 for 0 x 3.
Hence we can take K = 2 to nd our error bound.
|ErrorT rapezoidal |

2(3 0)3
54
=
.
12n2
12n2

We need to solve the following inequality for n:

54
< 0.001,
12n2
54
n2 >
,
12(0.001)

54
n>
67.08.
12(0.001)
Hence we must take n = 68 to achieve the desired accuracy.
From the last example, we see one of the weaknesses of the Trapezoidal Ruleit is not very accurate for
functions where straight line segments (and trapezoid tiles) do not lead to a good estimate of area. It
is reasonable to think that other methods of approximating curves might be more applicable for some
functions. Simpsons Rule is a method that uses parabolas to approximate the curve.
Simpsons Rule:
As was true with the Trapezoidal Rule, we divide the interval [a, b] into n sub-intervals of length x = ba
n .
We then construct parabolas through each group of three consecutive points on the graph. The graph
below shows this process for the rst three such parabolas for the case of n = 6 sub-intervals. You can see
that every interval except the rst and last contains two estimates, one too high and one too low, so the
resulting estimate will be more accurate.

Using parabolas in this way produces the following estimate of the area from Simpsons Rule:

199

www.ck12.org

f (x)dx

x
[ f (x0 ) + 4 f (x1 ) + 2 f (x2 ) + 4 f (x3 ) + 2 f (x4 ) . . . + 2 f (xn2 ) + 4 f (xn1 ) + f (xn )] .
3

We note that it has a similar appearance to the Trapezoidal Rule. However, there is one distinction we
need to note. The process of using three consecutive xi to approximate parabolas will require that we
assume that n must always be an even number.
Error Estimates for the Trapezoidal Rule
As with the Trapezoidal Rule, we have a formula that suggests how we can choose n to ensure that the
errors are within acceptable boundaries. The following method illustrates how we can choose a suiciently
large n.
Suppose | f 4 (x)| k for a x b. Then the error estimate is given by
|Error simpson |
Example 3:
a. Use Simpsons Rule to approximate

1
dx
1 x

k(b a)5
.
180n4

with n = 6.

Solution:
We nd x =

ba
n

1
= 41
6 = 2.
4
]
1
1[
f (1) + 4 f ( 32 ) + 2 f (2) + 4 f ( 52 ) + 2 f (3) + 4 f ( 27 ) + f (4)
dx
6
1 x
]
1[
1 + (4 23 ) + (2 21 ) + (4 25 ) + (2 13 ) + (4 72 ) + 14
=
6
1 [ 3517 ]
=
= 1.3956.
6 420

This turns out to be a pretty good estimate, since we know that

1
dx = ln x
x

]4
= ln(4) ln(1) = 1.3863.
1

Therefore the error is less than 0.01.

b. Find n so that the Simpson Rule Estimate for

1
dx
1 x

is accurate to 0.001.

Solution:



 
 for 1 x 4.
We need to nd n such that |Error simpson | 0.001. We start by noting that  f 4 (x) =  24
x5
Hence we can take K = 24 to nd our error bound:
|Error simpson |

24(4 1)5
5832
=
.
180n4
180n4

Hence we need to solve the following inequality for n:

5832
< 0.001.
180n4
We nd that
www.ck12.org

200

5832
,
180(0.001)

5832
n>4
13.42.
180(0.001)

n4 >

Hence we must take n = 14 to achieve the desired accuracy.

Technology Note: Estimating a Denite Integral with a TI-83/84 Calculator
4
We will estimate the value of 1 1x dx .
Graph the function f (x) = 1x with the [WINDOW] setting shown below.
The graph is shown in the second screen.
Press 2nd [CALC] and choose option 7 (see menu below)
When the fourth screen appears, press [1] [ENTER] then [4] [ENTER] to enter the lower and
upper limits.
5. The nal screen gives the estimate, which is accurate to 7 decimal places.
1.
2.
3.
4.

Lesson Summary
1.
2.
3.
4.

We
We
We
We

used the Trapezoidal Rule to solve problems.

estimated errors for the Trapezoidal Rule.
used Simpsons Rule to solve problems.
estimated Errors for Simpsons Rule.

Multimedia Links
For video presentations of Simpsons Rule (21.0), see Simpsons Rule, Approximate Integration (7:21) and
Math Video Tutorials by James Sousa, Simpsons Rule of Numerical Integration (8:48) .
For a video presentation of Newtons Method (21.0), see Newtons Method (7:29) .

Review Question
1. Use the Trapezoidal Rule to approximate

1
0

x2 ex dx with n = 8.

201

www.ck12.org

Figure 4.4: Simpsons Rule - Approximate Integration (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/533

Figure 4.5: Simpsons Rule (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/534

Figure 4.6: Newtons Method (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/535

www.ck12.org

202

2. Use the Trapezoidal Rule to approximate

3. Use the Trapezoidal Rule to approximate
4. Use the Trapezoidal Rule to approximate

ln xdx with n = 6.

11
1 + x4 dx with
03 1
dx with n = 8.
1 x

n = 4.

3
5. How large should you take n so that the Trapezoidal Estimate for 1 1x dx is accurate to within 0.001?
1
6. Use Simpsons Rule to approximate 0 x2 ex dx with n = 8.
4
7. Use Simpsons Rule to approximate 1 x ln xdx with n = 6.
2
8. Use Simpsons Rule to approximate 0 14 dx with n = 6.
1 x +1
9. Use Simpsons Rule to approximate 0 1 + x4 dx with n = 4.
2
10. How large should you take n so that the Simpson Estimate for 0 edx is accurate to within 0.00001?

Review Answers
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.

x2 ex dx 0.16

ln xdx 1.26
11
1 + x4 dx 1.10
03 1
dx 1.10
1 x
Take
1 n = 19
x2 ex dx 0.16
04
xln xdx 4.28
12 1

dx 1.36
0
4
1 x +1
1 + x4 dx 1.09
0
Take n = 9
04

Texas Instruments Resources

In the CK-12 Texas Instruments Calculus FlexBook, there are graphing calculator activities
designed to supplement the objectives for some of the lessons in this chapter. See http:
//www.ck12.org/flexr/chapter/9729.

203

www.ck12.org

Chapter 5
Applications of Denite
Integrals
In this chapter, we will explore some of the many applications of the denite integral by using it to calculate
areas between two curves, volumes, length of curves, and several other applications from real life such as
calculating the work done by a force, the pressure a liquid exerts on an object, and basic statistical concepts.

5.1 Area Between Two Curves

Learning Objectives
A student will be able to:
Compute the area between two curves with respect to the x and yaxes.
In the last chapter, we introduced the denite integral to nd the area between a curve and the x axis
over an interval [a, b]. In this lesson, we will show how to calculate the area between two curves.
Consider the region bounded by the graphs f and g between x = a and x = b, as shown in the gures
below. If the two graphs lie above the xaxis, we can interpret the area that is sandwiched between them
as the area under the graph of g subtracted from the area under the graph f.
Therefore, as the graphs show, it makes sense to say that
[Area under f (Fig. 1a)] [Area under g (Fig. 1b)] = [Area between f and g (Fig. 1c)],
b
b
b
f (x)dx
g(x) =
[ f (x)| g(x)]dx.
a

This relation is valid as long as the two functions are continuous and the upper function f (x) g(x) on
the interval [a, b].
The Area Between Two Curves (With respect to the xaxis)
If f and g are two continuous functions on the interval [a, b] and f (x) g(x) for all values of x in the
interval, then the area of the region that is bounded by the two functions is given by
b
A=
[ f (x) g(x)]dx.
a

www.ck12.org

204

Figure 5.1

Figure 5.2

Figure 5.3

205

www.ck12.org

Example 1:
Find the area of the region enclosed between y = x2 and y = x + 6.

Figure 5.4
Solution:
We rst make a sketch of the region (Figure 2) and nd the end points of the region. To do so, we simply
equate the two functions,
x2 = x + 6,
and then solve for x.
x2 x 6 = 0
(x + 2)(x 3) = 0
from which we get x = 2 and x = 3.
So the upper and lower boundaries intersect at points (2, 4) and (3, 9).
As you can see from the graph, x + 6 x2 and hence f (x) = x + 6 and g(x) = x2 in the interval [2, 3].
Applying the area formula,
b
A=
[ f (x) g(x)]dx
a
3
=
[(x + 6) (x2 )]dx.
2

Integrating,
[

x3
x2
+ 6x
A=
2
3
125
=
.
6

]3
2

So the area between the two curves f (x) = x + 6 and g(x) = x2 is 125/6.
Sometimes it is possible to apply the area formula with respect to the ycoordinates instead of the
xcoordinates. In this case, the equations of the boundaries will be written in such a way that y is
expressed explicitly as a function of x (Figure 3).
The Area Between Two Curves (With respect to the yaxis)
www.ck12.org

206

Figure 5.5
If w and v are two continuous functions on the interval [c, d] and w(y) v(y) for all values of y in the
interval, then the area of the region that is bounded by x = v(y) on the left, x = w(y) on the right, below
by y = c, and above by y = d, is given by

A=

[w(y) v(y)]dy.

Example 2:
Find the area of the region enclosed by x = y2 and y = x 6.
Solution:

Figure 5.6
As you can see from Figure 4, the left boundary is x = y2 and the right boundary is y = x 6. The region
extends over the interval 2 y 3. However, we must express the equations in terms of y. We rewrite
x = y2
x=y+6
Thus

A=

2
[ 2
y

[y + 6 y2 ]dy

y3
+ 6y
=
2
3
125
=
.
6

207

]3
2

www.ck12.org

Multimedia Links
For a video presentation of the area between two graphs (14.0)(16.0), see Math Video Tutorials by James
Sousa, Area Between Two Graphs (6:12) .

Figure 5.7: The Area Between Two Graphs (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/536
For an additional video presentation of the area between two curves (14.0)(16.0), see Just Math Tutoring,
Finding Areas Between Curves (9:50) .

Figure 5.8: Finding Areas Between Curves (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/537

Review Questions
In problems #1 - 7, sketch the region enclosed by the curves and nd the area.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.

y = x2 , y = x, on the interval [0.25,

[ 1]]
y = 0, y = cos 2x, on the interval 4 , 2
y = | 1 + x| + 2, y = 1
5 x+7
y = cos x, y = sin x, x = 0, x = 2
x = y2 , y = x 2, integrate with respect to y
y2 4x = 4, 4x y = 16
y = 8 cos x, y = sec2 x, /3 x /3
Find the area enclosed by x = y3 and x = y.
If the area enclosed by the two functions y = k cos x and y = kx2 is 2, what is the value of k?
Find the horizontal line y = k that divides the region between y = x2 and y = 9 into two equal areas.

Review Answers
1. 49/192
www.ck12.org

208

2.
3.
4.
5.
6.
7.
8.
9.
10.

1/2
24
4 2
9/2
30 38

6 3
1
2

k 1.83
3
9/ 4

5.2 Volumes
Learning Objectives

Learn
Learn
Learn
Learn

the basic concepts of volume and how to compute it with a given cross-section
how to compute volume by the disk method
how to compute volume by the washer method
how to compute volume by cylindrical shells

In this section, we will use denite integrals to nd volumes of dierent solids.

The Volume Formula

A circular cylinder can be generated by translating a circular disk along a line that is perpendicular to the
disk (Figure 5). In other words, the cylinder can be generated by moving the cross-sectional area A (the
disk) through a distance h. The resulting volume is called the volume of solid and it is dened to be
V = Ah.

Figure 5.9
The volume of solid does not necessarily have to be circular. It can take any arbitrary shape. One useful
way to nd the volume is by a technique called slicing. To explain the idea, suppose a solid S is positioned
on the xaxis and extends from points x = a to x = b (Figure 6). Let A(x) be the cross-sectional area
of the solid at some arbitrary point x. Just like we did in calculating the denite integral in the previous
chapter, divide the interval [a, b] into n sub-intervals and with widths

209

www.ck12.org

Figure 5.10

x1 , x2 , x3 , . . . , xn .
Eventually, we get planes that cut the solid into n slices
S 1, S 2, S 3, . . . , S n.
Take one slice, S k . We can approximate slice S k to be a rectangular solid with thickness xk and crosssectional area A(xk ). Thus the volume Vk of the slice is approximately
Vk A(xk )xk .
Therefore the volume V of the entire solid is approximately
V = V1 + V2 + . . . + Vn
n

A(xk )xk .
k=1

If we use the same argument to derive a formula to calculate the area under the curve, let us increase the
number of slices in such a way that xk 0. In this case, the slices become thinner and thinner and, as a
result, our approximation will get better and better. That is,
V = lim =
x0

A(xk )xk .

k=1

Notice that the right-hand side is just the denition of the denite integral. Thus
V = lim =
x0

A(xk )xk

k=1

A(x)dx.
a

The Volume Formula (Cross-section perpendicular to the xaxis)

Let S be a solid bounded by two parallel planes perpendicular to the xaxis at x = a and x = b. If each of
the cross-sectional areas in [a, b] are perpendicular to the x axis, then the volume of the solid is given by
www.ck12.org

210

V=

A(x)dx.
a

where A(x) is the area of a cross section at the value of x on the xaxis.
The Volume Formula (Cross-section perpendicular to the yaxis)
Let S be a solid bounded by two parallel planes perpendicular to the yaxis at y = c and y = d. If each of
the cross-sectional areas in [c, d] are perpendicular to the yaxis, then the volume of the solid is given by

V=

A(y)dy.
c

where A(y) is the area of a cross section at the value of y on the yaxis.
Example 1:
Derive a formula for the volume of a pyramid whose base is a square of sides a and whose height (altitude)
is h.

Figure 5.11

Figure 5.12

211

www.ck12.org

Solution:
Let the yaxis pass through the apex of the pyramid, as shown in Figure (7a). At any point y in the
interval [0, h], the cross-sectional area is a square. If b is the length of the sides of any arbitrary square,
then, by similar triangles (Figure 7b),
1
2b
1
2a

hy
,
h
a
b = (h y).
h
=

Since the cross-sectional area at y is A(y) = b2 ,

A(y) = b2 =

a2
(h y)2 .
h2

Using the volume formula,

V=

A(y)dy

c
h

a2
(h y)2 dy
2
h
0

a2 h
= 2
(h y)2 dy.
h 0

Using usubstitution to integrate, we eventually get

]h
a2 [ 1
3
V = 2 (h y)
3
h
0
1 2
= a h.
3
Therefore the volume of the pyramid is V = 13 a2 h, which agrees with the standard formula.

Volumes of Solids of Revolution

The Method of Disks
Figure 8a
Suppose a function f is continuous and non-negative on the interval [a, b], and suppose that R is the region
between the curve f and the xaxis (Figure 8a). If this region is revolved about the xaxis, it will generate
a solid that will have circular cross-sections (Figure 8b) with radii of f (x) at each x. Each cross-sectional
area can be calculated by
A(x) = [ f (x)]2 .
Since the volume is dened as

V=

A(x)dx,
a

the volume of the solid is

www.ck12.org

212

Figure 5.13

Figure 5.14

213

www.ck12.org

V=

[ f (x)]2 dx.

Volumes by the Method of Disks (revolution about the xaxis)

b
V=
[ f (x)]2 dx.
a

Because the shapes of the cross-sections are circular or look like the shapes of disks, the application of this
method is commonly known as the method of disks.
Example 2
Calculate the volume of the solid that is obtained when the region under the curve
the xaxis over the interval [1, 7].

x is revolved about

Solution:
As Figures 9a and 9b show, the volume is

V=

[ f (x)]2 dx

2
[ x] dx

x2
=
2

]7
1

= 24.

Figure 5.15
Example 3:
Derive a formula for the volume of the sphere with radius r.
Solution:
One way to nd the formula is to use the disk method. From your algebra, a circle of radius r and center
at the origin is given by the formula
x2 + y2 = r2
www.ck12.org

214

Figure 5.16
If we revolve the circle about the xaxis, we will get a sphere. Using the disk method, we will obtain a
formula for the volume. From the equation of the circle above, we solve for y:

r 2 x2 ,

f (x) = y =
thus

V=

[ f (x)]2 dx

a
+r

= r2 x

]2
r2 x2 dx
]
3 r

x
3

4
= r3 .
3
This is the standard formula for the volume of the sphere.
The Method of Washers
To generalize our results, if f and g are non-negative and continuous functions and
f (x) g(x)
for
a x b,
Then let R be the region enclosed by the two graphs and bounded by x = a and x = b. When this region is
revolved about the xaxis, it will generate washer-like cross-sections (Figures 10a and 10b). In this case,
we will have two radii: an inner radius g(x) and an outer radius f (x). The volume can be given by:

V(x) =

(
)
[ f (x)]2 [g(x)]2 dx.

Volumes by the Method of Washers (revolution about the xaxis)

215

www.ck12.org

Figure 5.17

Figure 5.18

www.ck12.org

216

(
)
[ f (x)]2 [g(x)]2 dx.

V(x) =
a

Example 4:
Find the volume generated when the region between the graphs f (x) = x2 +1 and g(x) = x over the interval
[0, 3] is revolved about the xaxis.
Solution:

Figure 5.19
From the formula above,

(
)
[ f (x)]2 [g(x)]2 dx

(
)
(x2 + 1)2 (x)2 dx

(
)
x4 + x2 + 1 dx

V(x) =
a

303
=
.
5
The methods of disks and washers can also be used if the region is revolved about the yaxis. The analogous
formulas can be easily deduced from the above formulas or from the volumes of solids generated.
Disks:

V=

[u(y)]2 dy.

217

www.ck12.org

Figure 5.20

Washers:

V=

(
)
[w(y)]2 [v(y)]2 dy.

Example 5:
What is the volume of the solid generated when the region enclosed by y =
about the yaxis?

x, y = 3, and x = 0 is revolved

Solution:
Since the solid generated is revolved about the yaxis (Figure 12), we must rewrite y =
Thus u(y) = y2 . The volume is

V=

[u(y)]2 dy

[y2 ]2 dy

y4 dx

]3
y5
=
5
[ 5 0 ]
3
0
=
5
243
=
.
5

www.ck12.org

218

x as x = y2 .

Figure 5.21

Figure 5.22

219

www.ck12.org

Volume By Cylindrical Shells

The method of computing volumes so far depended upon computing the cross-sectional area of the solid
and then integrating it across the solid. What happens when the cross-sectional area cannot be found or
the integration is too diicult to solve? Here is where the shell method comes along.
To show how diicult it sometimes is to use the disk or the washer methods to compute volumes, consider
the region enclosed by the function f (x) = x x2 . Let us revolve it about the line x = 1 to generate the
shape of a doughnut-shaped cake. What is the volume of this solid?

Figure 5.23
If we wish to integrate with respect to the yaxis, we have to solve for x in terms of y. That would not be
easy (try it!). An easier way is to integrate with respect to the xaxis by using the shell method. Here is
how: A cylindrical shell is a solid enclosed by two concentric cylinders. If the inner radius is r1 and the
outer one is r2 , with both of height h, then the volume is (Figure 14)
V = [area of the cross-section] [height]
= (r22 r12 )h
= (r2 + r1 )(r2 r1 )h
]
[1
= 2 (r2 + r1 ) h (r2 r1 ).
2
Notice however that (r2 r1 ) is the thickness of the shell and 21 (r2 + r1 ) is the average radius of the shell.
Thus
V = 2 [average radius] [height] [thickness].
Replacing the average radius with a single variable r and using h for the height, we have
www.ck12.org

220

Figure 5.24

Figure 5.25

221

www.ck12.org

V = 2 r h [thickness].
In general the shells thickness will be dx or dy depending on the axis of revolution. This discussion leads
to the following formulas for rotation about an axis. We will then use this formula to compute the volume
V of the solid of revolution that is generated by revolving the region about the xaxis.
Volume By Cylindrical Shell about the yAxis
Suppose f is a continuous function in the interval [a, b] and the region R is bounded above by y = f (x)
and below by the xaxis, and on the sides by the lines x = a and x = b. If R is rotated around the yaxis,
then the cylinders are vertical, with r = x and h = f (x). The volume of the solid is given by
b
a
V=
2rhdx =
2x f (x)dx.
a

Volume By Cylindrical Shell about the xAxis

Equivalently, if the volume is generated by revolving the same region about the xaxis, then the cylinders
are horizontal with
d
v=
2rhdy,
c

where c = f 1 (a) and d = f 1 (b). The values of r and h are determined in context, as you will see in
Example 6.
Note: Example 7 shows what to do when the rotation is not about an axis.
Example 6:
A solid gure is created by rotating the region R (Figure 15) around the xaxis. R is bounded by the curve
y = x2 and the lines x = 0 and x = 2. Use the shell method to compute the volume of the solid.
Solution:
From Figure 15 we can identify the limits of integration: y runs from 0 to 4. A horizontal strip of this

region would generate a cylinder with height 2 y and radius y. Thus the volume of the solid will be
d
2rhdy
V=
c
4

=
2y(2 y)dy
0
4
= 2
(2y y3/2 )dy
0

[
2 5/2 ]4
2
= 2 y y
5
0
32
=
.
5
Note: The alert reader will have noticed that this example could be worked with a simpler integral using
disks. However, the following example can only be solved with shells.
Example 7:
Find the volume of the solid generated by revolving the region bounded by y = x3 + 21 x + 14 , y = 14 , and
x = 1, about x = 3.
www.ck12.org

222

Figure 5.26

Figure 5.27

223

www.ck12.org

Solution:
As you can see, the equation y = x3 + 12 x + 14 cannot be easily solved for x and therefore it will be necessary
to solve the problem by the shell method. We are revolving the region about a line parallel to the yaxis
and thus integrate with respect to x. Our formula is

V=

2rhdx.
a

In this case, the radius is 3 x and the height is x3 + 12 x +

1
4

41 . Substituting,

(
1
1 1)
(3 x) x3 + x + dx
2
4 4
0
1(
1
3 )
= 2
x4 + 3x3 x2 + x dx
2
2
0
]1
[ 1
3
1
3
= 2
x5 + x 4 x3 + x2
5
4
6
4 0
[ 1 3 1 3 ]
= 2
+ +
5
4 6 4
[ 17 ]
= 2
15
34
=
.
15
1

V = 2

Multimedia Links
The following applet allows you to try out solids of revolution about the x-axis for any two functions. You
can try inputting the examples above to test it out, and then experiment with new functions and changing
the bounds. Volumes of Revolution Applet. In the following video the narrator walks trough the steps of
setting up a volume integration (14.0)(16.0). Khan Academy Solids of Revolution (10:04) .

Figure 5.28: Solid of Revolution (part 1) (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/538
Sometimes the same volume problem can be solved in two dierent ways (14.0)(16.0). In these two
videos, the narrator rst nds a volume using shells Khan Academy Solid of Revolution (Part 5) (9:29) ,
and then he does the same volume problem using disks. Khan Academy Solid of Revolution (Part 6) (9:18)
. Together these videos show how both methods can be used to solve the same problem (though its not
always done this easily!).
www.ck12.org

224

Figure 5.29: Solid of Revolution (part 5) (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/539

Figure 5.30: Solid of Revolution (part 6) (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/540

Review Questions
In problems #1 - 4, nd the volume of the solid generated by revolving the region bounded by the curves
about the xaxis.
1.
2.
3.
4.

y = 9 x2 , y = 0
y = 3 + x, y =
1 + x2
y = sec x, y = 2, /4 x /4
y = 1, y = x, x = 0

In problems #58, nd the volume of the solid generated by revolving the region bounded by the curves
about the yaxis.
5.
6.
7.
8.

y = x3 , x = 0, y = 1
x = y2 , y = x 2
x = csc y, y = /4, y = 3/4, x = 0

y = 0, y = x, x = 4

In problems #912, use cylindrical shells to nd the volume generated when the region bounded by the
curves is revolved about the axis indicated.
9.
10.
11.
12.
13.

y = 1x , y = 0, x = 1, x = 3, about the yaxis

y = x2 , x = 1, y = 0, about the xaxis
y = 2x 1, y = 2x + 3, x = 2, about the yaxis
y2 = x, y = 1, x = 0, about the xaxis.
Use the cylindrical shells method to nd the volume generated when the region is bounded by
y = x3 , y = 1, x = 0 is revolved about the line y = 1.

225

www.ck12.org

Review Answers
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.

36
117
5
2

2
3
3
5
72
5

2
128
5

20/3

2
9
14

5.3 The Length of a Plane Curve

Learning Objectives
A student will be able to:
Learn how to nd the length of a plane curve for a given function.
In this section will consider the problem of nding the length of a plane curve. Formulas for nding the
arcs of circles appeared in early historical records and they were known to many civilizations. However,
very little was known about nding the lengths of general curves, such as the length of the curve y = x2 in
the interval [0, 2], until the discovery of calculus in the seventeenth century.
In calculus, we dene an arc length as the length of a plane curve y = f (x) over an interval [a, b] (Figure
17). When the curve f (x) has a continuous rst derivative f on [a, b], we say that f is a smooth function
(or smooth curve) on [a, b].

Figure 5.31
www.ck12.org

226

The Arc Length Problem

If y = f (x) is a smooth curve on the interval [a, b], then the arc length L of this curve is dened as

( )2
b
b
dy

2
L=
1 + [ f (x)] dx =
1+
dx.
dx
a
a
Example 1:
Find the arc length of the curve y = x3 /2 on [1, 3] (Figure 18).

Figure 5.32
Solution:
Since y = x3 /2,
dy
3
= x1/2 .
dx
2
Using the formula above, we get

1 + [ f (x)]2 dx =

]2
[3
1 + x1/2 dx
2

9
1 + xdx.
4

Using usubstitution by letting u = 1 + 94 x, then du = 94 dx. Substituting, and remembering to change the
limits of integration,

4 31/4
udu
L=
9 13/4
8 [ 3/2 ]31/4
=
u
13/4
27
4.65.

227

www.ck12.org

Multimedia Links
The formula you just used to nd the length of a curve was derived by using line segments to approximate
the curve. The derivation of that formula can be found at Wikipedia Entry on Arc Length. In the following
applet you can explore this further. Experiment with various curves and change the number of segments to
see how changing the number of segments is related to approximating the arc length. Arc Length Applet.
For video presentations showing how to obtain the arc length using parametric curves (16.0), see Just
Math Tutoring, Arc Length Using Parametric Curves, Example 1 (8:17) and Just Math Tutoring, Arc

Figure 5.33: Arc Length Using Parametric Curves (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/541
Length Using Parametric Curves, Example 2 (7:27) .

Figure 5.34: Arc Length Using Parametric Curves - Example 2 (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/542

Review Questions
1. Find the arc length of the curve
y=
on [0, 3].
2. Find the arc length of the curve

(x2 + 2)3/2
3

1
1
x = y3 +
6
2y

on y [1, 2].
3. Integrate

x=

sec4 t 1dt,

y .
4
4

4. Find the length of the curve shown in the gure below. The shape of the graph is called the astroid
because it looks like a star. The equation of its graph is x2/3 + y2/3 = 1.
www.ck12.org

228

5. The gure below shows a suspension bridge. The cable has the shape of a parabola with equation
kx2 = y. The suspension bridge has a total length of 2S and the height of the cable is h at each end.
Show that the total length of the cable is s
4h2
1 + 4 x2 dx.
L=2
S
0

Review Answers
1.
2.
3.
4.
5.

11
(17/12)
2
6
.

5.4 Area of a Surface of Revolution

Learning Objectives
A student will be able to:

229

www.ck12.org

 Learn how to nd the area of a surface that is generated by revolving a curve about an axis or a line.
In this section we will deal with the problem of nding the area of a surface that is generated by revolving
a curve about an axis or a line. For example, the surface of a sphere can be generated by revolving a
semicircle about its diameter (Figure 19) and the circular cylinder can be generated by revolving a line
segment about any axis that is parallel to it (Figure 20).

Figure 5.35

Figure 5.36
Area of a Surface of Revolution
If f is a smooth and non-negative function in the interval [a, b], then the surface area S generated by
revolving the curve y = f (x) between x = a and x = b about the xaxis is dened by

( )2
b
b

dy

2
S =
2 f (x) 1 + [ f (x)] dx =
2y 1 +
dx.
dx
a
a
www.ck12.org

230

Equivalently, if the surface is generated by revolving the curve about the yaxis between y = c and y = d,
then

( )2
d
d

dx

2
2x 1 +
S =
2g(y) 1 + [g (y)] dy =
dy.
dy
c
c
Example 1:
Find the surface area that is generated by revolving y = x3 on [0, 2] about the xaxis (Figure 21).
Solution:

Figure 5.37
The surface area S is

S =

2y

1+

)2
dy
dx
dx

2x3 1 + (3x2 )2 dx
0
2
= 2
x3 (1 + 9x4 )1/2 dx.
2

Using usubstitution by letting u = 1 + 9x4 ,

231

www.ck12.org

145

u1/2

S = 2
1

du
36

2 [ 2 3/2 ]145
=
u
36 3
1
]
2 2 [
=
(145)3/2 1
36 3
4

[1745]
108
203
Example 2:
Find the area of the surface generated by revolving the graph of f (x) = x2 on the interval [0,
the yaxis (Figure 22).

3] about

Solution:

Figure 5.38
Since the curve is revolved about the yaxis, we apply

( )2
d
dx
S =
2x 1 +
dy.
dy
c
So we write y = x2 as x =

y. In addition, the interval on the xaxis [0,

)2
(
3
1

S =
2 y 1 + dy.
2 y
0

Simplifying,

S =

4y + 1dy.

www.ck12.org

232

3] becomes [0, 3]. Thus

With the aid of usubstitution, let u = 4y + 1,

13 1/2
S =
u du
4 1
[
]

=
(13)3/2 1
6

= [46.88 1]
6
24

Multimedia Links
For video presentations of nding the surface area of revolution (16.0), see Math Video Tutorials by James
Sousa, Surface Area of Revolution, Part 1 (9:47) and Math Video Tutorials by James Sousa, Surface Area

http://mathispower4u.yolasite.com/
http://mathispower4u.yolasite.com/
Figure 5.39: Surface Area of Revolution - Part 1 of 2 (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/543
of Revolution, Part 2 (5:43) .

http://mathispower4u.yolasite.com
http://mathispower4u.yolasite.com
Figure 5.40: Surface Area of Revolution - Part 2 of 2 (about y-axis) (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/544

Review Questions
In problems #1 - 3 nd the area of the surface generated by revolving the curve about the xaxis.
1. y = 3x, 0 x 1

2. y = x, 1 x 9

233

www.ck12.org

3. y =

4 x2 , 1 x 1

In problems #46 nd the area of the surface generated by revolving the curve about the yaxis.
4.
5.
6.
7.
8.

x = 7y + 2, 0 y 3
x = y3 , 0 y 8
x = 9 y2 , 2 y 2
Show that the surface area of a sphere of radius r is 4r2 .
of height h and base radius r is
Show that the lateral area S of a right circular cone
S = r r2 + h2 .

Review Answers
1.
2.
3.
4.
5.
6.
7.
8.

3 10
112
8
75 50
823165.5
24
.
.

5.5 Applications from Physics, Engineering, and

Statistics
Learning Objectives
A student will be able to:
Learn how to apply denite integrals to several applications from physics, engineering, and applied
mathematics such as work, uids statics, and probability.
In this section we will show how the denite integral can be used in dierent applications. Some of the
concepts may sound new to the reader, but we will explain what you need to comprehend as we go along.
We will take three applications: The concepts, of work from physics, uid statics from engineering, and
the normal probability from statistics.

Work
Work in physics is dened as the product of the force and displacement. Force and displacement are vector
quantities, which means they have a direction and a magnitude. For example, we say the compressor
exerts a force of 200 Newtons (N) upward. The magnitude here is 200 N and the direction is upward.
Lowering a book from an upper shelf to a lower one by a distance of 0.5 meters away from its initial
position is another example of the vector nature of the displacement. Here, the magnitude is 0.5 m and the
direction is downward, usually indicated by a minus sign, i.e., a displacement of 0.5 m. The product of
those two vector quantities (called the inner product, see Chapter 10) gives the work done by the force.
Mathematically, we say
www.ck12.org

234

W = Fd,
where F is the force and d is the displacement. If the force is measured in Newtons and distance is in
meters, then work is measured in the units of energy which is in joules (J).
Example 1:
You push an empty grocery cart with a force of 44 N for a distance of 12 meters. How much work is done
by you (the force)?
Solution:
Using the formula above,
W = Fd
= (44)(12)
= 528 J.
Example 2:
A librarian displaces a book from an upper shelf to a lower one. If the vertical distance between the two
shelves is 0.5 meters and the weight of the book is 5 Newtons . How much work is done by the librarian?
Solution:
In order to be able to lift the book and move it to its new position, the librarian must exert a force that
is at least equal to the weight of the book. In addition, since the displacement is a vector quantity, then
the direction must be taken into account. So,
d = 0.5 meters.
Thus
W = Fd
= (5)(0.5)
= 2.5 J.
Here we say that the work is negative since there is a loss of gravitational potential energy rather than a
gain in energy. If the book is lifted to a higher shelf, then the work is positive, since there will be a gain
in the gravitational potential energy.
Example 3:
A bucket has an empty weight of 23 N. It is lled with sand of weight 80 N and attached to a rope of
weight 5.1 N/m. Then it is lifted from the oor at a constant rate to a height 32 meters above the oor.
While in ight, the bucket leaks sand grains at a constant rate, and by the time it reaches the top no sand
is left in the bucket. Find the work done:
1.
2.
3.
4.

by
by
by
by

lifting the
lifting the
lifting the
the lifting

empty bucket;
sand alone;
rope alone;
the bucket, the sand, and the rope together.

235

www.ck12.org

Solution:
1. The empty bucket. Since the buckets weight is constant, the worker must exert a force that is equal to
the weight of the empty bucket. Thus
W = Fd
= (23)(+32)
= 736 J.
2. The sand alone. The weight of the sand is decreasing at a constant rate from 80 N to 0 N over the
32 meter lift. When the bucket is at x meters above the oor, the sand weighs
F(x) = [original weight of sand][proportion left at elevation x]
( 32 x )
= 80
( 32 x )
= 80 1
32
= 80 2.5x N.
The graph of F(x) = 80 2.5x represents the variation of the force with height x (Figure 23). The work
done corresponds to computing the area under the force graph.

Figure 5.41
Thus the work done is

W=

F(x)dx
a

32

[80 2.5x]dx

[
2.5 2 ]32
x
= 80x
2
0
= 1280 J.
www.ck12.org

236

3. The rope alone. Since the weight of the rope is 5.1 N/m and the height is 32 meters , the total weight
of the rope from the oor to a height of 32 meters is
(5.1)(32) = 163.2 N.
But since the worker is constantly pulling the rope, the ropes length is decreasing at a constant rate and
thus its weight is also decreasing as the bucket being lifted. So at x meters, the (32 x) meters there
remain to be lifted of weight F(x) = (5.1)(32 x) N. Thus the work done to lift the weight of the rope is
35
32
W=
F(x)dx =
(5.1)(32 x)dx
0

W = (5.1) 32x

0
]32
2
x

= 2611.2 J.
4. The bucket, the sand, and the rope together. Here we are asked to sum all the work done on the empty
bucket, the sand, and the rope. Thus
Wtotal = 736 + 1280 + 2611.2 = 4627.2 J.

Fluid Statics: Pressure

You have probably studied that pressure is dened as the force per area
P=

F
,
A

which has the units of Pascals (Pa) or Newtons per meter squared, Pa = N/m2 . In the study of uids, such
as water pressure on a dam or water pressure in the ocean at a depth h, another equivalent formula can
be used. It is called the liquid pressure P at depth h:
P = wh.
where w is the weight density, which is the weight of the column of water per unit volume. For example,
if you are diving in a pool, the pressure of the water on your body can be measured by calculating the
total weight that the column of water is exerting on you times your depth. Another way to express this
formula, the weight density w, is dened as
w = g,
where is the density of the uid and g is the acceleration due to gravity (which is g = 9.8 m/sec2 on
Earth). The pressure then can be written as
P = wh = gh.
Example 4:
What is the total pressure experienced by a diver in a swimming pool at a depth of 2 meters ?
Solution

237

www.ck12.org

First we calculate the uid pressure the water exerts on the diver at a depth of 2 meters :
P = gh.
The density of water is = 1000 kg/m3 , thus
P = (1000)(9.8)(2)
= 19600 Pa.
The total pressure on the diver is the pressure due to the water plus the atmospheric pressure. If we
assume that the diver is located at sea-level, then the atmospheric pressure at sea level is about 105 Pa .
Thus the total pressure on the diver is
Ptotal = Pwater + Patm
= 19600 + 105
= 119600
= 1.196 105 Pa.
Example 5:
What is the uid pressure (excluding the air pressure) and force on the top of a at circular plate of radius
3 meters that is submerged horizontally at a depth of 10 meters ?
Solution:
The density of water is = 1000 kg/m3 . Then
P = gh
= (1000)(9.8)(10)
= 98000 Pa.
Since the force is F = PA, then
F = PA
= P r2
= (98000)()(3)2
= 2.77 106 N.
As you can see, it is easy to calculate the uid force on a horizontal surface because each point on the
surface is at the same depth. The problem becomes a little complicated when we want to calculate the
uid force or pressure if the surface is vertical. In this situation, the pressure is not constant at every point
because the depth is not constant at each point. To nd the uid force or pressure on a vertical surface
we must use calculus.
The Fluid Force on a Vertical Surface
Suppose a at surface is submerged vertically in a uid of weight density w and the submerged portion
of the surface extends from x = a to x = b along the vertical xaxis, whose positive direction is taken as
downward. If L(x) is the width of the surface and h(x) is the depth of point x, then the uid force F is
dened as
www.ck12.org

238

F=

wh(x)L(x)dx.
a

Example 6:
A perfect example of a vertical surface is the face of a dam. We can picture it as a rectangle of a certain
height and certain width. Let the height of the dam be 100 meters and of width of 300 meters. Find the
total uid force exerted on the face if the top of the dam is level with the water surface (Figure 24).

Figure 5.42
Solution:
Let x = the depth of the water. At an arbitrary point x on the dam, the width of the dam is L(x) = 300 m
and the depth is h(x) = xm . The weight density of water is
wwater = g
= (1000)(9.8)
= 9800 N/m2 .
Using the uid force formula above,

F=

wh(x)L(x)dx
a

100

(9800)(x)(300)dx
100
= 2.94 106
xdx
0

[
= 2.94 106
= 1.47 10

10

239

0
]100
x2

N.
www.ck12.org

Normal Probabilities
If you were told by the postal service that you will receive the package that you have been waiting for
sometime tomorrow, what is the probability that you will receive it sometime between 3:00 PM and 5:00
PM if you know that the postal services hours of operations are between 7:00 AM to 6:00 PM?
If the hours of operations are between 7 AM to 6 PM, this means they operate for a total of 11 hours. The
interval between 3 PM and 5 PM is 2 hours, and thus the probability that your package will arrive is
2 hours
= 0.182
11 hours
= 18.2%

P=

So there is a probability of 18.2% that the postal service will deliver your package sometime between the
hours of 3 PM and 5 PM (or during any 2 hour interval). That is easy enough. However, mathematically,
the situation is not that simple. The 11 hour interval and the 2 hour interval contain an innite number
of times. So how can one innity over another innity produce a probability of 18.2%? To resolve this
issue, we represent the total probability of the 11 hour interval as a rectangle of area 1 (Figure 25).
2
of the total rectangular area 1. This is
Looking at the 2 hour interval, we can see that it is equal to 11
why it is convenient to represent probabilities as areas. But since areas can be dened by denite integrals,
we can also dene the probability associated with an interval [a, b] by the denite integral

Figure 5.43

P=

f (x)dx,
a

where f (x) is called the probability density function (pdf). One of the most useful probability density
functions is the normal curve or the Gaussian curve (and sometimes the bell curve) (Figure 26). This
function enables us to describe an entire population based on statistical measurements taken from a small
sample of the population. The only measurements needed are the mean () and the standard deviation ().
Once those two numbers are known, we can easily nd the normal curve by using the following formula.
The Normal Probability Density Function
The Gaussian curve for a population with mean and standard deviation is given by
f (x) =
www.ck12.org

e(x)

240

2 /(22 )

Figure 5.44

where the factor 1/( 2) is called the normalization constant. It is needed to make the probability over
the entire space equal to 1. That is,
+
2
2
1
P( < x < ) =
e(x) /(2 ) = 1.
2
Example 7:
Suppose that boxes containing 100 tea bags have a mean weight of 10.2 ounces each and a standard
deviation of 0.1 ounce.
1. What percentage of all the boxes is expected to weigh between 10 and 10.5 ounces ?
2. What is the probability that a box weighs less than 10 ounces ?
3. What is the probability that a box will weigh exactly 10 ounces ?
Solution:
1. Using the normal probability density function,
f (x) =

e(x)

2 /(22 )

Substituting for = 10.2 and = 0.1, we get

f (x) =

2
2
1
e(x10.2) /(2(0.1) ) .
(0.1) 2

The percentage of all the tea boxes that are expected to weight between 10 and 10.5 ounces can be calculated
as
10.5
2
2
1
P(10 x 10.5) =
e(x10.2) /(2(0.1) ) dx.
(0.1) 2
10
2

The integral of e x does not have an elementary anti-derivative and therefore cannot be evaluated by
standard techniques. However, we can use numerical techniques, such as The Simpsons Rule or The
Trapeziod Rule, to nd an approximate (but very accurate) value. Using the programing feature of a
scientic calculator or, mathematical software, we eventually get
10.5
2
2
1
e(x10.2) /(2(0.1) ) dx 0.976.
(0.1) 2
10
That is,
P(10 x 10.2) 97.6%.
Technology Note: To make this computation with a graphing calculator of the TI-83/84 family, do the
following:

241

www.ck12.org

 From the [DISTR] menu (Figure 27) select option 2, which puts the phrase normalcdf in the home
screen. Add lower bound, upper bound, mean, standard deviation, separated by commas, close the
parentheses, and press [ENTER]. The result is shown in Figure 28.

Figure 5.45

Figure 5.46
2. For the probability that a box weighs less than 10.2 ounces, we use the area under the curve to the left
of x = 10.2. Since the value of f (9) is very small (less than a billionth),
2
2
1
e(910.2) /(2(0.1) ) dx
(0.1) 2
= 1.35 1032 ,

f (9) =

getting the area between 9 and 10 will yield a fairly good answer. Integrating numerically, we get

10

2
2
1
e(x10.2) /(2(0.1) ) dx
(0.1) 2
9
P(9 x 10.2) 0.02275

P(9 x 10) =

= 2.28%,
which says that we would expect 2.28% of the boxes to weigh less than 10 ounces.
3. Theoretically the probability here will be exactly zero because we will be integrating from 10 to 10,
which is zero. However, since all scales have some error (call it ), practically we would nd the probability
that the weight falls between 10 and 10 + .
Example 8:
www.ck12.org

242

An Intelligence Quotient or IQ is a score derived from dierent standardized tests attempting to measure
the level of intelligence of an adult human being. The average score of the test is 100 and the standard
deviation is 15.
1. What is the percentage of the population that has a score between 85 and 115?
2. What percentage of the population has a score above 140?
Solution:
1. Using the normal probability density function,
f (x) =

e(x)

2 /(22 )

and substituting = 100 and = 15,

f (x) =

15 2

e(x100)

2 /(2(15)2 )

The percentage of the population that has a score between 85 and 115 is
115
2
2
1
P(85 x 115) =
e(x100) /(2(15) ) .
15 2
85
Again, the integral of ex does not have an elementary anti-derivative and therefore cannot be evaluated.
Using the programing feature of a scientic calculator or a mathematical computer software, we get
115
2
2
1
e(x100) /(2(15) ) dx 0.68.
15 2
85
2

That is,
P(85 x 115) 68%.
Which says that 68% of the population has an IQ score between 85 and 115.
2. To measure the probability that a person selected randomly will have an IQ score above 140,

2
2
1
P(x 140) =
e(x100) /(2(15) ) dx.
140 15 2
This integral is even more diicult to integrate since it is an improper integral. To avoid the messy work, we
can argue that since it is extremely rare to meet someone with an IQ score of over 200, we can approximate
the integral from 140 to 200. Then
200
2
2
1
P(x 140)
e(x100) /(2(15) ) dx.
140 15 2
Integrating numerically, we get
P(x 140) 0.0039.
So the probability of selecting at random a person with an IQ score above 140 is 0.39%. Thats about one
person in every 250 individuals!

243

www.ck12.org

Multimedia Links
For a video presentation of an application of integration involving consumer and producer surplus (14.0),
see Math Video Tutorials by James Sousa, Consumer and Producer Surplus (10:22) .

Figure 5.47: Consumer and Producer Surplus (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/545
For video presentations of work and Hookes Law (14.0)(16.0), see Just Math Tutoring, Work and Hookes
Law, Example 1 (5:00) and Just Math Tutoring, Work and Hookes Law, Example 2 (6:52) .

Figure 5.48: Work and Hookes Law - Ex 1 (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/546

Figure 5.49: Work and Hookes Law - Ex 2 (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/547
For a video that uses calculus to explain centripetal acceleration (16.0), see Khan Academy, Centripetal
Acceleration (10:14) .
For an economics application involving equilibrium point (14.0), see Math Video Tutorials by James Sousa,
Equilibrium Point (4:58) .
For economics applications involving future and present values (14.0), see Math Video Tutorials by James
Sousa, Future and Present Value, Part 1 (6:51) ;Math Video Tutorials by James Sousa, Future and Present
Value, Part 2 (4:45) .
www.ck12.org

244

Figure 5.50: Calculus Proof that a=v2/r (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/548

Figure 5.51: Equilibrium Point (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/549

Figure 5.52: Future and Present Value - Part 1 of 2 (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/550

Figure 5.53: Future and Present Value - Part 2 of 2 (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/551

245

www.ck12.org

Review Questions
1. A particle moves along the xaxis by a force
F(x) =

x2

1
.
+1

If the particle has already moved a distance of 10 meters from the origin, what is the work done by
the force?
( )
2. A force of cos x
2 acts on an object when it is x meters away from the origin. How much work is
done by this force in moving the object from x = 1 to x = 5 meters ?
3. In physics, if the force on an object varies with distance then work done by the force is dened as
b
(see Example 5.15)
F(r)dr.
W=
a

That is, the work done corresponds to computing the area under the force graph. For example, the
strength of the gravitational eld varies with the distance r from the Earths center. If a satellite
of mass m is to be launched into space, then the force experienced by the satellite during and after
launch is
mM
F(r) = G 2 ,
r
2

is the Universal Gravitational

where M = 61024 kg is the mass of the Earth and G = 6.671011 Nm
kg2
Constant. If the mass of the satellite is 1000 kg and we wish to lift it to an altitude of 35, 780 km
above the Earths surface, how much work is needed to lift it? (Radius of Earth is 6370 km.)
4. Hooks Law states that when a spring is stretched x units beyond its natural length it pulls back with
a force
F(x) = kx,
where k is called the spring constant or the stiness constant. To calculate the work required to
stretch the spring a length x we use
b
W=
F(x)dx,
a

where a is the initial displacement of the spring (a = 0 if the spring is initially unstretched) and b is
the nal displacement. A force of 5 N is exerted on a spring and stretches it 1 m beyond its natural
length.
(a) Find the spring constant k.
(b) How much work is required to stretch the spring 1.8 m beyond its natural length?
5. When a force of 30 N is applied to a spring, it stretches it from a length of 12 cm to 15 cm. How
much work will be done in stretching the spring from 12 cm to 20 cm? (Hint: read the rst part of
problem #4 above.)
6. A at surface is submerged vertically in a uid of weight density w. If the weight density w is doubled,
is the force on the plate also doubled? Explain.
7. The bottom of a rectangular swimming pool, whose bottom is an inclined plane, is shown below.
Calculate the uid force on the bottom of the pool when it is lled completely with water.
8. Suppose f (x) is the probability density function for the lifetime of a manufacturers light bulb, where
x is measured in hours. Explain the meaning of each integral.
5000
(a) 1000 f (x)dx

(b) 3000 f (x)dx

9. The length of time a customer spends waiting until his/her entre is served at a certain restaurant is
modeled by an exponential density function with an average time of 8 minutes.
www.ck12.org

246

Figure 5.54

(a) What is the probability that a customer is served in the rst 3 minutes?
(b) What is the probability that a customer has to wait more than 10 minutes ?
10. The average height of an adult female in Los Angeles is 63.4 inches (5 feet 3.4 inches) with a standard
deviation of 3.2 inches .
(a)
(b)
(c)
(d)

What
What
What
What

is
is
is
is

the
the
the
the

probability
probability
probability
probability

that
that
that
that

a
a
a
a

females
females
females
females

height
height
height
height

is
is
is
is

less than 63.4 inches ?

between 63 and 65 inches ?
more than 6 feet ?
excatly 5 feet ?

Review Answers
1.
2.
3.
4.
5.

1.471 J
0J
5 1010 J

(a) k = 5 N/m
(b) 8.1 J
6. 3.2 J
7. Yes. To explain why, ask how w and F are mathematically related.
8. 63, 648 N
9.
10. (a) The probability that a randomly chosen light bulb will have a lifetime between 1000 and
5000 hours.
(b) The probability that a randomly chosen light bulb will have a lifetime of at least 3000 hours.
11.
12. (a) 31%
(b) 29%
13.
14. (a) 50%
(b) 24%
(c) 0.36%
(d) almost 0%

247

www.ck12.org

Image Sources
(1) .
(2) .
(3) .
(4) .
(5) .
(6) .
(7) .
(8) .
(9) .
(10) .
(11) .
(12) .
(13) .
(14) .
(15) .
(16) .
(17) .
(18) .
(19) .
(20) .
(21) .
(22) .
(23) .
(24) .
(25) .
(26) .
(27) .
(28) .
(29) .
www.ck12.org

248

(30) .
(31) .
(32) .
(33) .
(34) .
(35) .
(36) .
(37) .
(38) .

Texas Instruments Resources

In the CK-12 Texas Instruments Calculus FlexBook, there are graphing calculator activities
designed to supplement the objectives for some of the lessons in this chapter. See http:
//www.ck12.org/flexr/chapter/9730.

249

www.ck12.org

Chapter 6
Transcendental Functions
6.1 Inverse Functions
Functions such as logarithms, exponential functions, and trigonometric functions are examples of transcendental functions. If a function is transcendental, it cannot be expressed as a polynomial or rational
function. That is, it is not an algebraic function. In this chapter, we will begin by developing the concept
of an inverse of a function and how it is linked to its original numerically, algebraically, and graphically.
Later, we will take each type of elementary transcendental functionlogarithmic, exponential, and trigonometricindividually and see the connection between them and their respective inverses, derivatives, and
integrals.

Learning Objectives
A student will be able to:
Understand the basic properties of the inverse of a function and how to nd it.
Understand how a function and its inverse are represented graphically.
Know the conditions of invertabilty of a function.

One-to-One Functions
A function, as you know from your previous mathematics background, is a rule that assigns a single value
in its range to each point in its domain. In other words, for each output number, there is one or more
input numbers. However, a function never produces more than a single output for one input. A function
is said to be a one-to-one function if each output is associated with only one single input. For example,
f (x) = x2 assigns the output 9 for both 3 and 3, and thus it is not a one-to-one function.
One-to-One Function
The function f (x) is one-to-one in a domain D if f (a) , f (b), whenever a , b.
There is an easy method to check if a function is one-to-one: draw a horizontal line across the graph. If the
line intersects at only one point on the graph, then the function is one-to-one; otherwise, it is not. Notice
in the gure below that the graph of y = x2 is not one-to-one since the horizontal line intersects the graph
more than once. But the function y = x3 is a one-to-one function because the graph meets the horizontal
line only once.
www.ck12.org

250

Example 1:
Determine whether the functions are one-to-one: (a) f (x) = |x| (b) h(x) = x1/2 .
Solution:
It is best to graph both functions and draw on each a horizontal line. As you can see from the graphs,
f (x) = |x| is not one-to-one since the horizontal line intersects it at two points. The function h(x) = x1/2 ,
however, is indeed one-to-one since only one point is intersected by the horizontal line.

251

www.ck12.org

The Inverse of a Function

We discussed above the condition for a one-to-one function: for each output, there is only one input. A
one-to-one function can be reversed in such a way that the input of the function becomes the output and
the output becomes an input. This reverse of the original function is called the inverse of the function.
If f 1 is an inverse of a function f, then f 1 f = x. For example, the two functions f (x) = 2x + 3 and
h(x) = x3
2 are inverses of each other since
[ x 3]
f h = f (h(x)) = 2
+ 3 = x 3 + 3 = x,
2
(2x + 3) 3
2x
=
= x.
h f = h( f (x)) =
2
x
Thus
f h = h f = x,
and f and h are inverses of each other.
Note: In general, f 1 , 1f .
When is a function invertable?
It is interesting to note that if a function f (x) is always increasing or always decreasing over its domain,
then a horizontal line will cut through this graph at one point only. Then f in this case is a one-to-one
function and thus has an inverse. So if we can nd a way to prove that a function is constantly increasing
or decreasing, then it is invertable or monotonic. From previous chapters, you have learned that if
f (x) > 0 then f must be increasing and if f (x) < 0 then f must be decreasing.
To summarize, a function has an inverse if it is one-to-one in its domain or if its derivative is either f (x) > 0
or f (x) < 0.
Example 2:
Given the polynomial function f (x) = 3x5 + 2x + 1, show that it is invertable (has an inverse).
Solution:
Taking the derivative, we nd that f (x) = 15x4 + 2 > 0 for all x. We conclude that f (x) is one-to-one and
invertable. Keep in mind that it may not be easy to nd the inverse of f (x) = 3x5 + 2x + 1 (try it!), but
we still know that it is indeed invertable.
www.ck12.org

252

How to nd the inverse of a one-to-one function:

To nd the inverse of a one-to-one function, simply solve for x in terms of y and then interchange x and y.
The resulting formula is the inverse y = f 1 (x).
Example 3:
Find the inverse of f (x) =

4x + 1.

Solution:
From the discussion above, we can nd the inverse by rst solving for x in y =

y = 4x + 1,

4x + 1.

y2 = 4x + 1,
x=

y2 1
.
4

y=

x2 1
.
4

Interchanging x y,

Replacing y = f 1 (x),
f 1 (x) =
which is the inverse of the original function f (x) =

x2 1
4

4x + 1.

Graphs of Inverse Functions

What is the relationship between the graphs of f and f 1 ? If the point (a, b) is on the graph of f (x),
then from the denition of the inverse, the point (b, a) is on the graph of f 1 (x). In other words, when we
reverse the coordinates of a point on the graph of f (x) we automatically get a point on the graph of f 1 (x).
We conclude that f (x) and f 1 (x) are reections of one another about the line y = x. That is, each is a
mirror image of the other about the line y = x. The gure below shows an example of y = x2 and, when

the domain is restricted, its inverse y = x and how they are reected about y = x.

It is important to note that for the function f (x) = x2 to have an inverse, we must restrict its domain to
0 x < , since that is the domain in which the function is increasing.

253

www.ck12.org

Continuity and Dierentiability of Inverse Functions

Since the graph of a one-to-one function and its inverse are reections of one another about the line y = x,
it would be safe to say that if the function f has no breaks (no discontinuities) then f 1 will not have
breaks either. This implies that if f is continuous on the domain D, then its inverse f 1 is continuous on

the range R of f. For example, if f (x) = x, then its domain is x 0 and its range is y 0. This means
that f (x) is continuous for all x 0. The inverse of f (x) is f 1 (x) = x2 , where its domain is all x > 0 and
its range is y 0. We conclude that if f is a function with domain D and range R and it is continuous and
one-to-one on D, then its inverse f 1 is continuous and one-to-one on the range R of f.
Suppose that f has a domain D and a range R. If f is dierentiable and one-to-one on D, then its inverse
f 1 is dierentiable at any value x in R for which f ( f 1 (x)) , 0 and
d 1
1
[ f (x)] = 1
.
dx
f ( f (x))
The formula above can be written in a form that is easier to remember:
dy
1
=
.
dx
dx/dy
In addition, if f on its domain is either f (x) > 0 or f (x) < 0, then f has an inverse function f 1 and f 1
is dierentiable at all values of x in the range of f. In this case, f 1 is given by the formula above. The
example below illustrate this important theorem.
Example 4:
In Example 3, we were given the polynomial function f (x) = 3x5 +2x+1 and we showed that it is invertable.
Show that it is dierentiable and nd the derivative of its inverse.
Solution:
Since f (x) = 15x4 + 2 > 0 for all x R, f 1 (x) is dierentiable at all values of x. To nd the derivative of
f 1 , if we let x = f (y), then
x = f (y) = 3y5 + 2y + 1.
So
dx
= 15y4 + 2
dy
and
dy
1
1
=
=
.
4
dx
dx/dy
15y + 2
Since we are unable to solve for y in terms of x, we leave the answer above in terms of y. Another way of
solving the problem is to use Implicit Dierentiation:
Since
x = 3y5 + 2y + 1,
dierentiating implicitly,
www.ck12.org

254

d
d
[x] =
[3y5 + 2y + 1],
dx
dx
dy
1 = (15y4 + 2) .
dx
Solving for

dy
dx

we nally obtain
dy
1
=
,
4
dx
15y + 2

which is the same result.

Review Questions
In problems #1 - 3, nd the inverse function of f and verify that f f 1 = f 1 f = x.
1. f (x) = 3x + 1

2. 3 x
3. f (x) = x1
3
In problems #4 - 6, use the horizontal line test to verify whether the following functions have inverse.
4. h(x) = 4x
6
5. g(x) = |x +
4| |x 4|
6. f (x) = 2x 16 x2
In problems #7 - 8, use the functions f (x) = x + 4 and g(x) = 2x 5 to nd the specied functions.
7. g1 f 1
8. ( f g)1
In problems #9 - 10, show that f is monotonic (invertable) on the given interval (and therefore has an
inverse.)
9. f (x) = (x 5)[ 2 , [5,] )
10. f (x) = cos x, 0, 2

Review Answers
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.

f 1 (x) = x1
3
x3
3x + 1
Function has an inverse.
Function does not have an inverse.
Function does not have an inverse.
x+1
2
x+1
2
f (x)

f (x)

= 2(x 5) > 0 on x > 5.

= sin x, which is negative on the interval in question, so f (x) is monotonically decreasing.

255

www.ck12.org

6.2 Exponential and Logarithmic Functions

Learning Objectives
A student will be able to:
Understand and use the basic denitions of exponential and logarithmic functions and how they are
related algebraically.
Distinguish between an exponential and logarithmic functions graphically.

A Quick Algebraic Review of Exponential and Logarithmic Functions

Exponential Functions
Recall from algebra that an exponential function is a function that has a constant base and a variable
exponent. A function of the form f (x) = b x where b is a constant (and
) x b > 0 and b , 1 is called an
exponential function with base b. Some examples are f (x) = 2 x , f (x) = 12 , and f (x) = e x . All exponential
functions are continuous and their graph is one of the two basic shapes, depending on whether 0 < b < 1
or b > 1. The graph below shows the two basic shapes:

Logarithmic Functions
Recall from your previous courses in algebra that a logarithm is an exponent. If the base b > 0 and b , 1,
then for any value of x > 0, the logarithm to the base b of the value of x is denoted by
y = logb x.
This is equivalent to the exponential form
by = x.
For example, the following table shows the logarithmic forms in the rst row and the corresponding
exponential forms in the second row.
1
= 2
25
1
=
25

Logarithmic Form

log2 16 = 4

log5

log10 100 = 2

loge e = 1

Exponential Form

24 = 16

52

102 = 100

e1 = e

www.ck12.org

256

Historically, logarithms with base of 10 were very popular. They are called the common logarithms.
Recently the base 2 has been gaining popularity due to its considerable role in the eld of computer
science and the associated binary number system. However, the most widely used base in applications is
the natural logarithm, which has an irrational base denoted by e, in honor of the famous mathematician
Leonhard Euler. This irrational constant is e 2.718281. Formally, it is dened as the limit of (1 + x)1/x
as x approaches zero. That is,
lim (1 + x)1/x = e.
x0

We denote the natural logarithm of x by ln x rather than loge x. So keep in mind, that ln x is the power to
which e must be raised to produce x. That is, the following two expressions are equivalent:
y = ln x x = ey
The table below shows this operation.
Natural Logarithm ln

ln 2 = 0.693

ln 1 = 0

ln e = 1

ln e3 = 3

Equivalent Exponential Form

e0.693 = 2

e0 = 1

e1 = e

e3 = e3

A Comparison between Logarithmic Functions and Exponential Functions

Looking at the two graphs of exponential functions above, we notice that both pass the horizontal line
test. This means that an exponential function is a one-to-one function and thus has an inverse. To nd a
formula for this inverse, we start with the exponential function
y = bx .
Interchanging x and y,
x = by .
Projecting the logarithm to the base b on both sides,
logb x = logb by
= y logb b
= y(1)
= y.
Thus y = f 1 (x) = logb x is the inverse of y = f (x) = b x .
This implies that the graphs of f and f 1 are reections of one another about the line y = x. The gure
below shows this relationship.

257

www.ck12.org

Similarly, in the special case when the base b = e, the two equations above take the forms
y = f (x) = e x
and
f 1 (x) = ln x.
The graph below shows this relationship:

Before we move to the calculus of exponential and logarithmic functions, here is a summary of the two
important relationships that we have just discussed:
The function y = b x is equivalent to x = logb y if y > 0 and x R.
The function y = e x is equivalent to x = ln y if y > 0 and x R.
You should also recall the following important properties about logarithms:
logb vw = logb v + logb w
www.ck12.org

258

logb wv = logb v logb w

logb wn = n logb w
To express a logarithm with base in terms of the natural logarithm: logb w =
log w
To express a logarithm with base b in terms of another base a: logb w = loga b

ln w
ln b

Review Questions
Solve for x.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.

1
6 x = 216
x
e =3
log2 z = 3
ln x2 = 5
3e5x = 132
e2x 7e x + 10 = 0
4(3) x = 36
ln x ln 3 =( 2 )
2
y = 5 log10 2x
y = 3e2x/3

Review Answers
x = 3
ln 3
z=8
e5/2
x = 0.757
ln 2 and ln 5
x=2
x = 3e(2
)
x = 2 1 101y/5
( )
10. x = 32 ln 3y
1.
2.
3.
4.
5.
6.
7.
8.
9.

6.3 Dierentiation and Integration of Logarithmic and Exponential Functions

Learning Objectives
A student will be able to:
Understand and use the rules of dierentiation of logarithmic and exponential functions.
Understand and use the rules of integration of logarithmic and exponential functions.
In this section we will explore the derivatives of logarithmic and exponential functions. We will also see
how the derivative of a one-to-one function is related to its inverse.

259

www.ck12.org

The Derivative of a Logarithmic Function

Our goal at this point to nd an expression for the derivative of the logarithmic function y = logb x. Recall
that the exponential number e is dened as
e = lim (1 + a)1/a
a 0

(where we have substituted a for x for convenience). From the denition of the derivative of f (x) that you
already studied in Chapter 2,
f (x) = lim

h0

f (x + h) f (x)
f (w) f (x)
= lim
.
w x
h
wx

We want to apply this denition to get the derivative to our logarithmic function y = logb x. Using the
denition of the derivative and the rules of logarithms from the Lesson on Exponential and Logarithmic
Functions,
logb w logb x
d
[logb x] = lim
w x
dx
wx
logb (w/x)
= lim
w x
wx
( w )]
[ 1
logb
= lim
w x w x
(x
)]
[
x + (w x)
1
logb
= lim
w x w x
x
(
[ 1
w x )]
= lim
logb 1 +
w x w x
x
[
]
(
1
w x)
= lim
.
logb 1 +
w x x(w x)
x
At this stage, let a = (w x)/(x), the limit of w x then becomes a 0. Substituting, we get
[1 1

]
logb (1 + a)
a 0 x a
[1
]
1
= lim
logb (1 + a)
x a 0 a
[
]
1
= lim logb (1 + a)1/a .
x a 0
= lim

Inserting the limit,

=

[
]
1
logb lim (1 + a)1/a .
a 0
x

But by the denition e = lima 0 (1 + a)1/a ,

d
1
[logb x] = logb e.
dx
x
From the box above, we can express logb e in terms of natural logarithm by the using the formula logb w =
ln w/ ln b. Then
www.ck12.org

260

logb e =

ln e
1
=
.
ln b
ln b

Thus we conclude
1
d
[logb x] =
> 0,
dx
x ln b
and in the special case where b = e,
d
1
[lnx] = > 0.
dx
x
To generalize, if u is a dierentiable function of x and if u(x) > 0, then the above two equations, after the
Chain Rule is applied, will produce the generalized derivative rule for logarithmic functions.
Derivatives of Logarithmic Functions
d
1 du
[logb u] =
dx
u ln b dx
d
1 du
u
[ln u] =
=
dx
u dx
u
Remark: Students often wonder why the constant e is dened the way it is. The answer is in the derivative
1
of f (x) = ln x. With any other base the derivative of f (x) = logb x would be equal f (x) = x ln
b , a more
complicated expression than 1/x. Thinking back to another unexpected unit, radians, the derivative of

f (x) = sin(x) is the simple expression f (x) = cos(x) only if x is in radians. In degrees, f (x) = 180
cos(x),
which is more cumbersome and harder to remember.
Example 1:
Find the derivative of y = ln(2x2 4x + 3).
Solution:
Since

d
dx

[lnu] =

1 du
u dx

, for u = 2x2 4x + 3,
]
dy
1
d [ 2
= 2
2x 4x + 3
dx
2x 4x + 3 dx
1
= 2
(4x 4)
2x 4x + 3
4(x 1)
= 2
.
2x 4x + 3

Example 2:
Find

d
dx

[ln(sin x)].

Solution:
d
1
[ln(sin x)] =
[cos x]
dx
sin x
cos x
=
sin x
= cot x.

261

www.ck12.org

Example 3:
[
]
d
Find dx
ln(cos 5x)3 .
Solution:
Here we use the Chain Rule:
]
[
]
d [
1
ln(cos 5x)3 =
3(cos 5x)2 ( sin 5x) (5)
3
dx
cos 5x
[
]
1
2
=

15
cos
5x

sin
5x
cos3 5x
15 sin 5x
=
cos 5x
= 15 tan 5x.
Example 4:
Find the derivative of y = x3 log5 2x.
Solution:
Here we use the Product Rule along with

d
dx [logb u]

1 du
u ln b dx

d
d
d 3
[x log5 2x] = x3 [log5 2x] + [x3 ] log5 2x
dx
dx
dx
1
= x3
+ 3x2 log5 2x
2x ln 5
x2
=
+ 3x2 log5 2x.
2 ln 5
Example 5:
Find the derivative of y = ln

x
x+1

Solution:
We use the Quotient Rule and the natural logarithm rule:
(x + 1)(1) (1)(x)
(x + 1)2
1
x+1
=

x
(x + 1)2
1
=
.
x(x + 1)

y =

x
x+1

Integrals Involving Natural Logarithmic Function

In the last section, we have learned that the derivative of y = ln u(x) is y =
is

u (x)
dx = ln |u(x)| + C.
u(x)
d
If the argument of the natural logarithm is x, then dx
[ln x] = 1/x, thus

1
dx = ln |x| + C.
x

www.ck12.org

262

1
.u (x).
u(x)

The antiderivative

Example 6:
1
Evaluate x+1
dx.
Solution:
In general, whenever you encounter an integral with an integrand as a rational function, it might be possible
that it can be integrated with the rule of natural logarithm. To do so, determine the derivative of the
denominator. If it is the numerator itself, then the integration is simply the ln of the absolute value of the
denominator. Lets test this technique.

1
dx.
x+1
Notice that the derivative of the denominator is 1, which is equal to the numerator. Thus the solution is
simply the natural logarithm of the absolute value of the denominator:

1
dx = ln |x + 1| + C.
x+1
The formal way of solving such integrals is to use usubstitution by letting u equal the denominator. Here,
let u = x + 1, and du = dx. Substituting,

1
1
dx =
du
x+1
u
= ln |u| + C
= ln |x + 1| + C.
Remark: The integral must use the absolute value symbol because although x may have negative values,
the domain of ln(x) is restricted to x 0.
Example 7:

dx.
Evaluate 4x24x+1
+2x+1
Solution:
As you can see here, the derivative of the denominator is 8x + 2. Our numerator is 4x + 1. However, when
we multiply the numerator by 2, we get the derivative of the denominator. Hence

2(4x + 1)
4x + 1
1
dx =
dx
2
2
4x + 2x + 1
4x2 + 2x + 1

8x + 2
1
=
dx
2
2
4x + 2x + 1
1
= ln |4x2 + 2x + 1| + C.
2
Again, we could have used usubstitution.
Example 8:

Evaluate tan xdx.

Solution:
To solve, we rewrite the integrand as

tan xdx =

263

sin x
dx.
cos x
www.ck12.org

Looking at the denominator, its derivative is sin x. So we need to insert a minus sign in the numerator:

sin x
=
dx
cos x
= ln | cos x| + C.

Derivatives of Exponential Functions

We have discussed above that the exponential function is simply the inverse function of the logarithmic
function. To obtain a derivative formula for the exponential function with base b, we rewrite y = b x as
x = logb y.
Dierentiating implicitly,
1=
Solving for

dy
dx

1
dy
.
y ln b dx

and replacing y with b x ,

dy
= y ln b = b x ln b.
dx

Thus the derivative of an exponential function is

d x
[b ] = b x ln b.
dx
In the special case where the base is b x = e x , since ln e = 1 the derivative rule becomes
d x
[e ] = e x .
dx
To generalize, if u is a dierentiable function of x, with the use of the Chain Rule the above derivatives
take the general form
d u
du
[b ] = bu ln b .
dx
dx
And if b = e,
d u
du
[e ] = eu .
dx
dx
Derivatives of Exponential Functions
d u
du
[b ] = bu ln b
= u bu ln b
dx
dx
d u
du
[e ] = eu
= u eu
dx
dx
Example 9:
2

Find the derivative of y = 2 x .

www.ck12.org

264

Solution:
Applying the rule for dierentiating an exponential function,
y = (2x)2 x ln 2
2

= 2x

2 +1

x ln 2.

Example 10:
2

Find the derivative of y = e x .

Solution:
Since
d u
[e ] = u eu ,
dx
2
y = 2xe x .
Example 11:
Find f (x) if
2
1
f (x) = ek(xx0 ) .

where , , x0 , and k are constants and , 0.

Solution:
We apply the exponential derivative and the Chain Rule:
2
1
f (x) = (2k(x x0 ))ek(xx0 )

2k(x x0 ) k(xx0 )2
=
e
.

Integrals Involving Exponential Functions

Associated with the exponential derivatives in the box above are the two corresponding integration formulas:

1 u
b + C,
bu du =
ln b

eu du = eu + C.
The following examples illustrate how they can be used.
Example 12:

Evaluate 5 x dx.
Solution:

1 x
5 +C
ln 5
5x
=
+ C.
ln 5

5 x dx =

265

www.ck12.org

Example 13:

e x dx.
Solution:

e x dx = e x + C.

3
In the next chapter, we will learn how to integrate more complicated integrals, such as x2 e x dx, with
the use of usubstitution and integration by parts along with the logarithmic and exponential integration
formulas.

Multimedia Links
For a video presentation of the derivatives of exponential and logarithmic functions (4.4), see Math Video
Tutorials by James Sousa, The Derivatives of Exponential and Logarithmic Functions (8:26) .

http://mathispower4u.wordpress.com/
http://mathispower4u.wordpress.com/
Figure 6.1: The Derivatives of Exponential and Logarithmic Functions (Base not equal to e) (Watch
Youtube Video)
http://www.ck12.org/exbook/embed/view/552

Review Questions
1. Find dy/dx of y = e6x
3
2
2. Find dy/dx of y = e3x 2x (+6)
3. Find dy/dx of y = ex2 ln 1x
x

e
Find dy/dx of y = eex +e
x
Find dy/dx of y = cos(e x )
Find dy/dx
y = ln(sin(ln x))
of
1
Evaluate ex dx

Evaluate e x dx
Evaluate 4x24x3
dx
ex +e6x+7
x
10. Evaluate ex ex dx
e dx
11. Evaluate 0 x+e
ln 3 x
12. Evaluate ln 3 exe+4 dx

4.
5.
6.
7.
8.
9.

www.ck12.org

266

Review Answers
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.

y = 6e6x
3
2
y = (9x2 4x)e3x 2x +6
2
y = xe x [2 ln(x) + 1]
y = (ex +e4 x )2
y = e x sin e x
tan(ln x)
y =
x
ex + C
2e x/2 + C
1
2
2 ln |4x 6x + 7| + C
x
ln |e ex | + C
ln 2
ln 21
13

6.4 Exponential Growth and Decay

Learning Objectives
A student will be able to:
Apply the laws of exponential and logarithmic functions to a variety of applications.
Model situations of growth and decay in a variety of problems.
When the rate of change in a substance or population is proportional to the amount present at any time
t, we say that this substance or population is going through either a decay or a growth, depending on the
sign of the constant of proportionality.
This kind of growth is called exponential growth and is characterized by rapid growth or decay. For
example, a population of bacteria may increase exponentially with time because the rate of change of its
population is proportional to its population at a given instant of time (more bacteria make more bacteria
and fewer bacteria make fewer bacteria). The decomposition of a radioactive substance is another example
in which the rate of decay is proportional to the amount of the substance at a given time instant. In the
business world, the interest added to an investment each day, month, or year is proportional to the amount
present, so this is also an example of exponential growth.
Mathematically, the relationship between amount y and time t is a dierential equation:
dy
= ky.
dt
Separating variables,
dy
= kdt,
y
and integrating both sides,

dy
=
y

267

kdt,
www.ck12.org

gives us
ln y = kt + C,
y = ekt eC
= Cekt .
So the solution to the equation (dy/dt) = ky has the form y = Cekt . The box below summarizes the details
of this function.
The Law of Exponential Growth and Decay
The function y = Cekt is a model for exponential growth or decay, depending on the value of k.
If k > 0: The function represents exponential growth (increase).
If k < 0: The function represents exponential decay (decrease).
Where t is the time, C is the initial population at t = 0, and y is the population after time t.

Applications of Growth and Decay

Radioactive Decay
In physics, radioactive decay is a process in which an unstable atomic nucleus loses energy by emitting
radiation in the form of electromagnetic radiation (like gamma rays) or particles (such as beta and alpha
particles). During this process, the nucleus will continue to decay, in a chain of decays, until a new stable
nucleus is reached (called an isotope). Physicists measure the rate of decay by the time it takes a sample
to lose half of its nuclei due to radioactive decay. Initially, as the nuclei begins to decay, the rate starts
very fast and furious, but it slows down over time as more and more of the available nuclei have decayed.
The gure below shows a typical radioactive decay of a nucleus. As you can see, the graph has the shape
of an exponential function with k < 0.

The equation that is used for radioactive decay is y = Cekt . We want to nd an expression for the half-life
of an isotope. Since half-life is dened as the time it takes for a sample to lose half of its nuclei, then if we
starting with an initial mass C (measured in grams), then after some time t, y will become half the amount
that we started with, C/2. Substituting this into the exponential decay model,
www.ck12.org

268

y = Cekt
C
= Cekt .
2
Canceling C from both sides,
1
= ekt .
2
Solving for t, which is the half-life, by taking the natural logarithm on both sides,
1
= ln ekt
2
ln 2 = kt.
ln

Solving for t, and denoting it with new notation t1/2 for half-life (a standard notation in physics),
t1/2 =

0.693
ln 2
=
k
k

This is a famous expression in physics for measuring the half-life of a substance if the decay constant k is
known. It can also be used to compute k if the half-life t1/2 is known.
Example 1:
A radioactive sample contains 2 grams of nobelium. If you know that the half-life of nobelium is 25 seconds,
how much will remain after 3 minutes?
Solution:
Before we compute the mass of nobelium after 3 minutes, we need to rst know its decay rate k. Using the
half-life formula,
ln 2
k
ln 2
k=
t1/2
ln 2
=
25
= 0.028 sec1

t1/2 =

So the decay rate is k = 0.028/sec. The common unit for the decay rate is the Becquerel (Bq): 1 Bq is
equivalent to 1 decay per sec. Since we found k, we are now ready to calculate the mass after 3 minutes.
We use the radioactive decay formula. Remember, C represents the initial mass, C = 2 grams, and
t = 3 minutes = 180 seconds. Thus
y = Cekt
= 2e(0.028)(180)
= 0.013 grams.
So after 3 minutes, the mass of the isotope is approximately 0.013 grams.

269

www.ck12.org

Population Growth
The same formula y = Cekt can be used for population growth, except that k > 0, since it is an increasing
function.
Example 2:
A certain population of bacteria increases continuously at a rate that is proportional to its present number.
The initial population of the bacterial culture is 140 and jumped to 720 bacteria in 4 hours.
1. How many will be there in 10 hours?
2. How long will it take the population to double?
Solution:
From reading the rst sentence in the problem, we learn that the bacteria is increasing exponentially.
Therefore, the exponential growth formula is the correct model to use.
1. Just like we did in the previous example, we need to rst nd k, the growth rate. Notice that C =
140, t = 4, and y = 720. Substituting and solving for k.
y = Cekt
720 = 140ek(4) .
Dividing both sides by 140 and then projecting the natural logarithm on both sides,
720
= ln e4k
140
ln 5.143 = 4k
ln

k = 0.409.
Now that we have found k, we want to know how many will be there after 10 hours. Substituting,
y = Cekt
= 140e(0.409)(10)
= 8364 bacteria.
2. We are looking for the time required for the population to double. This means that we are looking for
the time at which y = 2C. Substituting,
y = Cekt
2C = Cekt
2 = ekt .
Solving for t requires taking the natural logarithm of both sides:
ln 2 = ln ekt
ln 2 = kt.
Solving for t,
www.ck12.org

270

ln 2
k
ln 2
=
0.409
t = 1.7 hours.
t=

This tells us that after about 1.7 hours (around 100 minutes) the population of the bacteria will double in
number.
Compound Interest
Investors and bankers depend on compound interest to increase their investment. Traditionally, banks
added interest after certain periods of time, such as a month or a year, and the phrase was the interest
is being compounded monthly or yearly. With the advent of computers, the compunding could be done
daily or even more often. Our exponential model represents continuous, or instantaneous, compounding,
and it is a good model of current banking practices. Our model states that
A = Pert ,
where P is the initial investment (present value) and A is the future value of the investment after time t at
an interest rate of r. The interest rate r is usually given in percentage per year. The rate must be converted
to a decimal number, and t must be expressed in years. The example below illustrates this model.
Example 3:
An investor invests an amount of $10, 000 and discovers that its value has doubled in 5 years. What is the
annual interest rate that this investment is earning?
Solution:
We use the exponential growth model for continuously compounded interest,
A = Pert
20, 000 = 10, 000er(5)
2 = e5r
ln 2 = 5r.
Thus
ln 2
5
= 0.139

r=

r = 13.9%
The investment has grown at a rate of 13.9% per year.
Example 4:
Going back to the previous example, how long will it take the invested money to triple?
Solution:

271

www.ck12.org

A = Pert
30, 000 = 10, 000e(0.139)(t)
3 = e0.139t
ln 3 = 0.139t
ln 3
t=
0.139
= 7.9 years.
Other Exponential Models and Examples
Not all exponential growths and decays are modeled in the natural base e or by y = Cekt . Actually, in
everyday life most are constructed from empirical data and regression techniques. For example, in the
business world the demand function for a product may be described by the formula
p = 12, 400

11, 000
,
2.2 + e0.003x

where p is the price per unit and x is the number of units produced. So if the business is interested in
basing the price of its unit on the number that it is projecting to sell, this formula becomes very helpful.
If a motorcycle factory is projecting to sell 7000 units in one month, what price should the factory set on
each motorcycle?
11, 000
2.2 + e0.003x
11, 000
= 12, 400
2.2 + e0.0003(7000)
11, 000
= 12, 400
2.2 + 0.122
= 7, 663.

p = 12, 400

Thus the factorys base price for each motorcycle should be set at $7663.
As another example, lets say a medical researcher is studying the spread of the u virus through a certain
campus during the winter months. Lets assume that the model for the spread is described by
P=

4500
, x 0,
1 + 4490e0.8x

where P represents the total number of infected students and x is the time, measured in days. Suppose
the researcher is interested in the number of students who will be infected in the next week (7 days).
Substituting x = 7 into the model,
4500
1 + 4490e0.8x
4500
=
1 + 4499e0.8(7)
4500
=
1 + 4499(0.004)
= 255.

P=

www.ck12.org

272

According to the model, 255 students will become infected with the u virus. Assume further that the
researcher wants to know how long it will take until 1000 students become infected with the u virus.
Solving for x,
P=

4500
.
1 + 4490e0.8x

Cross-multiplying,
P(1 + 4499e0.8x) = 4500
4500
1 + 4499e0.8x =
P
4500
0.8x
4499e
=
1
P
4500 P
=
P
4500 P
0.8x
e
=
.
4499P
Projecting ln on both sides,
0.8x = ln
x = ln

[ 4500 P ]
4499P
[ 4500 P ]
4499P

(0.8).

Substituting for P = 1000,

x = 9 days.
So the u virus will spread to 1000 students in 9 days.
Other applications are introduced in the exercises.

Multimedia Links
For a video presentation of exponential growth involving bacteria (some calculus in part c) (14.0), see
Khan Academy, Exponential Growth and Decay (16:00) .

Figure 6.2: Exponential Growth (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/318
For a video presentation of exponential decay (14.0), see Just Math Tutoring, Exponential Decay, Finding
Half-Life (6:08) .
For additional problems on exponential growth and decay (14.0), see Khan Academy, Word Problem
Solving, Exponential Growth and Decay (7:21) .

273

www.ck12.org

Figure 6.3: Exponential Decay / Finding Half Life (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/553

Figure 6.4: Word Problem Solving- Exponential Growth and Decay (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/554

Review Questions
1. In 1990, the population of the USA was 249 million. Assume that the annual growth rate is 1.8%.
(a) According to this model, what was the population in the year 2000?
(b) According to this model, in which year the population will reach 1 billion?
2. Prove that if a quantity A is exponentially growing and if A1 is the value at t1 and A2 at time t2 , then
( )
the growth rate will be given by
A1
1
ln
.
k=
t1 t2
A2
3. Newtons Law of Cooling states that the rate of cooling is proportional to the dierence in temperature
between the object and the surroundings. The law is expressed
by the formula
T (t) = (T 0 T r )ekt + T r
where T 0 is the initial temperature of the object at t = 0, T r is the room temperature (the surroundings), and k is a constant that is unique for the measuring instrument (the thermometer) called the
time constant. Suppose a liter of juice at 23C is placed in the refrigerator to cool. If the temperature
of the refrigerator is kept at 11C and k = 0.417, what is the temperature of the juice after 3 minutes?
4. Referring back to problem 3, if it takes an object 320 seconds to cool from 40C above room temperature to 22C above room temperature, how long will it take to cool another 10C?
5. Polonium210 is a radioactive isotope with half-life of 140 days. If a sample has a mass of 10 grams,
how much will remain after 10 weeks?
6. In the physics of acoustics, there is a relationship between the subjective sensation of loudness and
the physically measured intensity of sound. This relationship is called the sound level . It is specied
on a logarithmic scale and measured with units of decibels (dB). The sound level of any sound
is dened in terms of its intensity I (in the SI-mks unit system, it is measured in watts per meter
squared, W/m2 ) as
I
= 10 log 12 .
10
www.ck12.org

274

For example, the average decibel level of a busy street traic is 70 dB, normal conversation at a
dinner table is 55 dB, the sound of leaves rustling is 10 dB, the siren of a re truck at 30 meters is
100 dB, and a loud rock concert is 120 dB. The sound level 120 dB is considered the threshold of
pain for the human ear and 0 dB is the threshold of hearing (the minimum sound that can be heard
by humans.)
(a) If at a heavy metal rock concert a dB meter registered 130 dB, what is the intensity I of this
sound level?
(b) What is the sound level (in dB) of a sound whose intensity is 2.0 106 W/m2 ?
7. Referring to problem #6, a single mosquito 10 meters away from a person makes a sound that is
barely heard by the person (threshold 0 dB). What will be the sound level of 1000 mosquitoes at the
same distance?
8. Referring back to problem #6, a noisy machine at a factory produces a sound level of 90 dB. If an
identical machine is placed beside it, what is the combined sound level of the two machines?

Review Answers
1.
2. (a) 298.1 million
(b) 2077
3. Hint: use A1 = Cekt1
4. 14.4 C
5. 923 seconds, about 15 minutes
6. 7 grams
7.
8. (a) 10 W/m2
(b) 63 dB
9. 30 dB
10. 93 dB

6.5 Derivatives and Integrals Involving Inverse

Trigonometric Functions
Learning Objectives
A student will be able to:
Learn the basic properties inverse trigonometric functions.
Learn how to use the derivative formula to use them to nd derivatives of inverse trigonometric
functions.
Learn to solve certain integrals involving inverse trigonometric functions.

A Quick Algebraic Review of Inverse Trigonometric Functions

You already know what a trigonometric function is, but what is an inverse trigonometric function? If
we ask what is sin(/6) equal to, the answer is (1/2). That is simple enough. But what if we ask what
angle has a sine of (1/2)? That is an inverse trigonometric function. So we say sin(/6) = (1/2), but

275

www.ck12.org

sin1 (1/2) = (/6). The sin1 is the notation for the inverse of the sine function. For every one of the
six trigonometric functions there is an associated inverse function. They are denoted by
sin1 x, cos1 x, tan1 x, sec1 x, csc1 x, cot1 x
Alternatively, you may see the following notations for the above inverses, respectively,
arcsin x, arccos x, arctan x, arcsec x, arccsc x, arccot x
Since all trigonometric functions are periodic functions, they do not pass the horizontal line test. Therefore
they are not one-to-one functions. The table below provides a brief summary of their denitions and basic
properties. We will restrict our study to the rst four functions; the remaining two, csc1 and cot1 , are
of lesser importance (in most applications) and will be left for the exercises.
Table 6.1:
Inverse Function

Domain

Range

Basic Properties

sin1
cos1

1 x 1
1 x 1

tan1

all R

sec1

(, 1] [1, +)

sin(sin1 (x)) = x
cos1 (cos x)
cos(cos1 (x)) = x
tan1 (tan x)
tan(tan1 (x)) = x
sec1 (sec x)
sec(sec1 (x)) = x

y 2
0y
(
[

2 ,2

) (
]

0,
2 2,

=
=
=

The range is based on limiting the domain of the original function so that it is a one-to-one function.
Example 1:

What is the exact value of sin1 ( 3/2)?

Solution:
This is equivalent to sin x =
your scientic calculator.

3
2

. Thus sin1 ( 3/2) = /3. You can easily conrm this result by using

Example 2:
Most calculators do not provide a way to calculate the inverse of the secant function, sec1 x. A practical
trick however is to use the identity
sec1 x = cos1
(Recall that sec =

1
x

1
cos .)

For practice, use your calculator to nd sec1 (3.24).

Solution:
Since
1
1
=
= 0.3086,
x
3.24
sec1 3.24 = cos1 0.3086 = 72 .
www.ck12.org

276

Here are two other identities that you may need to enter into your calculator:
1
csc1 x = sin1 ,
x
1
1 1
cot x = tan
.
x

The Derivative Formulas of the Inverse Trigonometric Functions

If u is a dierentiable function of x then the generalized derivative formulas for the inverse trigonometric
functions are (we introduce them here without a proof):
d [ 1 ]
sin u =
dx
d [ 1 ]
cos u =
dx
d [ 1 ]
tan u =
dx
d [ 1 ]
sec u =
dx
d [ 1 ]
csc u =
dx
d [ 1 ]
cot u =
dx

du
1 < u < 1
1 u2 dx
1 du
1 < u < 1

1 u2 dx
1 du
< x <
1 + u2 dx
1
du  
u > 1
 
 x u2 1 dx
1
du  
u > 1
 
u u2 1 dx

1 du
< x <
1 + u2 dx

Example 3:
Dierentiate y = sin1 (2x4 )
Solution:
Let u = 2x4 , so
dy
1
=
(8x2 )
dx
1 (2x4 )2
8x3
=
.
1 4x8
Example 4:
Dierentiate tan1 (e3x ).
Solution:
Let u = e3x , so
dy
1
=
3e3x
dx
1 + (e3x )2
3e3x
.
=
1 + e6x
Example 5:
Find dy/dx if y = sec1 (sin x).

277

www.ck12.org

Solution:
Let u = sin x.
dy
1
cos x
= 

dx
sin x sin2 x 1
cos x

= 
sin x cos x
1
.
= 
sin x

The Integration Formulas of the Inverse Trigonometric Functions

The derivative formulas in the box above yield the following integrations formulas for inverse trigonometric
functions:

du
= sin1 u + c

2
1u
du
= tan1 u + c
1 + u2

 
du
= sec1 u + c

u u2 1
Example 6:
dx
Evaluate 1+4x
2.
Solution:
Before we integrate, we use usubstitution. Let u = 2x (the square root of 4x2 ). Then du = 2dx.
Substituting,

1/2
dx
=
du
2
1 + 4x
1 + u2

1
1
=
du
2
1 + u2
1
= tan1 u + c
2
1
= tan1 (2x) + c.
2
Example 7:

x
Evaluate e

1e2x

dx.

Solution:
We use usubstitution. Let u = e x , so du = e x dx. Substituting,

ex
ex
du
dx =

x
2x
2
1e
1u e

1
=
du

1 u2
= sin1 u + c
= sin1 (e x ) + c.
www.ck12.org

278

Example 8:

ln(2/ 3)

Evaluate the denite integral

ln2

x
e
dx.
1e2x

Solution:
Substituting u = ex , du = ex dx.
To change the limits,
1
x = ln 2 u = ex = e ln 2 = eln 1/2 = ,
2
(
)

2
3
x = ln u = ex = e ln 2/ 3 =
.
2
3
Thus our integral becomes

ln(2/ 3)

ln2

ex
1

e2x

3/2

dx =
1/2

3/2

u
1

u2

1/2

= sin u

du
u

1 u2

du

3/2

1/2

( 1 )

3

1
sin
= sin

2
2
[ ]
=
3 6

= .
3

Multimedia Links
For a video presentation of the derivatives of inverse trigonometric functions (4.4), see Math Video Tutorials by James Sousa, The Derivatives of Inverse Trigonometric Functions (8:55) .

Figure 6.5: The Derivatives of Inverse Trigonometric Functions (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/555
For three presentations of integration involving inverse trigonometric functions (18.0), see Math Video
Tutorials by James Sousa, Integration Involving Inverse Trigonometric Functions, Part 1 (7:39) ; Math
Video Tutorials by James Sousa, Integration Involving Inverse Trigonometric Functions, Part 2 (6:39) ;
This last video includes an example showing completing the square (19.0), Math Video Tutorials by James
Sousa, Integration Involving Inverse Trigonometric Functions, Part 3 (6:18) .

279

www.ck12.org

http://mathispower4u.yolasite.com
http://mathispower4u.yolasite.com
Figure 6.6: Integration Involving Inverse Trig Functions - Part 1 (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/556

Figure 6.7: Integration Involving Inverse Trig Functions: Part 2 (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/557

Figure 6.8: Integration Involving Inverse Trig Functions: Part 3 (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/558

www.ck12.org

280

Review Questions
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.

Find dy/dx of y = sec1 x2 .

Find dy/dx of y = tan11 x .
Find dy/dx of y = ln(cos1 x).
Find dy/dx of y = sin1 e4x .
1 2
Find dy/dx
ofdxy = sin (x ln x).
Evaluate 2 .
3 9xdx
Evaluate 1 x(x+1)
.
x3
Evaluate x2 +1 dx.
0
x
Evaluate 3 1+x
2 dx.
Given the points A(2, 1) and B(5, 4), nd a point Q in the interval [2, 5] on the xaxis that maximizes
angle ]AQB.

Review Answers
1.
2.
3.
4.
5.

2
x x4 1
1
(1+x2 )(tan1 x)2
1 1 2
(cos x) 1x
4x
4e
1e8x
x+2x4 ln x 2
1x (ln x)
( )
1
1 x
sin
3
3 +C

6.
7. /6
8. 12 ln(x2 + 1) 3 tan1 x + C
9. ln 2
10. 1 + 2 2

6.6 LHospitals Rule

Learning Objectives
A student will be able to:
Learn how to nd the limit of indeterminate form (0/0) by LHospitals rule.
If the two functions f (x) and g(x) are both equal to zero at x = a, then the limit
lim

x a

f (x)
g(x)

cannot be found by directly substituting x = a. The reason is because when we substitute x = a, the
substitution will produce (0/0), known as an indeterminate form, which is a meaningless expression.
To work around this problem, we use LHospitals rule, which enables us to evaluate limits of indeterminate
forms.
LHospitals Rule

281

www.ck12.org

If lim xa f (x) = lim xa g(x) = 0, and f (a) and g (a) exist, where g (a) , 0, then
lim

x a

f (x)
f (x)
= lim .
x a g (x)
g(x)

The essence of LHospitals rule is to be able to replace one limit problem with a simpler one. In each of
the examples below, we will employ the following three-step process:
f (x)

1. Check that lim xa g(x) is an indeterminate form 0/0. To do so, directly substitute x = a into f (x)
and g(x). If you get f (a) = g(a) = 0, then you can use LHospitals rule. Otherwise, it cannot be
used.
2. Dierentiate f (x) and g(x) separately.
f (x)
f (x)
3. Find lim xa g (x) . If the limit is nite, then it is equal to the original limit lim xa g(x) .
Example 1:
Find lim x0

2+x 2
.
x

Solution:
When x = 0 is substituted, you will get 0/0.
Therefore LHospitals rule applies:

lim

x0

2+x
x

d
dx ( 2 + x 2)

= lim
d
x0
dx (x)

1/(2 2 + x)
=

1
x=0
1
=
2 2

2
=
.
4

Example 2:
Find lim x0

1cos 2x
x2 +2x

Solution:
We can see that the limit is 0/0 when x = 0 is substituted.
Using LHospitals rule,

[
]
1 cos 2x
2 sin 2x
lim 2
=
x0 x + 2x
2x + 2 x=0
= 0/2
= 0.

Example 3:
Use LHospitals rule to evaluate lim x3

x2 9
x3

Solution:
lim =
x3

www.ck12.org

2x
x2 9
= lim
= 6.
x3 1
x3

282

Example 4:
Evaluate lim x0

sin 3x
x .

Solution:
3 cos 3x
sin 3x
= lim
= 3.
x0
x0
x
1

lim
Example 5:
Evaluate lim x/2

55 sin x
cos x

Solution:
We can use LHospitals rule since the limit produces the 0/0 once x = /2 is substituted. Hence
0 5 cos x
0
5 5 sin x
= lim
=
= 0.
cos x
1
x/2 sin x
x/2
lim

A broader application of LHospitals rule is when x = a is substituted into the derivatives of the numerator
and the denominator but both still equal zero. In this case, a second dierentiation is necessary.
Example 6:
Evaluate lim x0

e x cos x
.
x2

Solution:
e x cos x
e x + sin x
=
lim
.
x0
x0
2x
x2

lim

As you can see, if we apply the limit at this stage the limit is still indeterminate. So we apply LHospitals
rule again:
e x cos x
x0
2
11
0
=
= = 0.
2
2
= lim

Review Questions
Find the limits.
1.
2.
3.
4.
5.
6.

lim x0 tan
ln x
lim x1 tan
x
10x
6x
lim x0 e xe
x
lim x sin
xx
xe
lim x0 1e
x
If k is a nonzero constant and x > 0.
x 1
k
(a) Show that 1 t1k
dt = x k1 .
(b) Use LHospitals rule to nd limk0

xk 1
k .

7. Cauchys Mean Value Theorem states that if the functions f and g are continuous on the interval
(a, b) and g , 0, then there exists a number
c such that
f (c)
f (b) f (a)
=
.

g (c)
g(b) g(a)

283

www.ck12.org

Find all possible values of c in the interval (a, b) that satisfy this property for
f (x) = cos x
g(x) = sin x
on the interval

[ ]
[a, b] = 0, .
2

Review Answers
1.
2.
3.
4.
5.
6.
7.

1
1/
4
1
1
ln x

Image Sources
Texas Instruments Resources
In the CK-12 Texas Instruments Calculus FlexBook, there are graphing calculator activities
designed to supplement the objectives for some of the lessons in this chapter. See http:
//www.ck12.org/flexr/chapter/9731.

www.ck12.org

284

Chapter 7
Integration Techniques
7.1 Integration by Substitution
Each basic rule of integration that you have studied so far was derived from a corresponding dierentiation
rule. Even though you have learned all the necessary tools for dierentiating exponential, logarithmic,
trigonometric, and algebraic functions, your set of tools for integrating these functions is not yet complete.
In this chapter we will explore dierent ways of integrating functions and develop several integration
techniques that will greatly expand the set of integrals to which the basic integration formulas can be
applied. Before we do that, let us review the basic integration formulas that you are already familiar with
from previous chapters.
1. The Power Rule (n , 1):

xn + 1
+ C.
n+1

xn dx =
2. The General Power Rule (n , 1):

un

du
dx =
dx

un du =

un+1
+ C.
n+1

3. The Simple Exponential Rule:

e x dx = e x + C.
4. The General Exponential Rule:

u du

dx

dx =

eu du = eu + C.

5. The Simple Log Rule:

1
dx = ln |x| + C.
x

6. The General Log Rule:

285

www.ck12.org

du/dx
dx =
u

1
du = ln |u| + C.
u

It is important that you remember the above rules because we will be using them extensively to solve more
complicated integration problems. The skill that you need to develop is to determine which of these basic
rules is needed to solve an integration problem.

Learning Objectives
A student will be able to:
Compute by hand the integrals of a wide variety of functions by using the technique of usubstitution.
Apply the usubstitution technique to denite integrals.
Apply the usubstitution technique to trig functions.
Probably one of the most powerful techniques of integration is integration by substitution. In this technique,
you choose part of the integrand to be equal to a variable we will call u and then write the entire integrand
in terms of u. The diiculty of the technique is deciding which term in the integrand will be best for
substitution by u. However, with practice, you will develop a skill for choosing the right term.
Recall from Chapter 2 that if u is a dierentiable function of x and if n is a real number and n , 1, then
the Chain Rule tells us that
d n
du
[u ] = nun1 .
dx
dx
The reverse of this formula is the integration formula,

un du =

un+1
+ C, n , 1.
n+1

Sometimes it is not easy to integrate directly. For example, look at this integral:

(5x 2)2 dx.

One way to integrate is to rst expand the integrand and then integrate term by term.

2
(5x 2) dx = (25x2 20x + 4)dx

2
= 25 x dx 20 xdx + 4dx
=

25 3
x 10x2 + 4x + C.
3

That is easy enough. However, what if the integral was

(5x 2)15 dx?

www.ck12.org

286

Would you still expand the integrand and then integrate term by term? That would be impractical and
time-consuming. A better way of doing this is to change the variables. Changing variables can often turn
a diicult integral, such as the one above, into one that is easy to integrate. The method of doing this is
called integration by substitution, or for short, the u-substitution method. The examples below will
show you how the method is used.
Example 1:

Evaluate (x + 1)5 dx.

Solution:
Let u = x + 1. Then du = d(x + 1) = 1dx = dx. Substituting for u and du we get

5
(x + 1) dx =
u5 du.
Integrating using the power rule,
=

u6
+ C.
6

Since u = x + 1, substituting back,

(x + 1)6
+ C.
6

=
Example 2:

4x + 3dx.
Evaluate
Solution:
Let u = 4x + 3. Then du = 4dx. Solving for dx,

dx = du/4.
Substituting,

1
u1/2 dx
4

1
=
u1/2 dx
4
1 u3/2
+ C.
=
4 3/2

Simplifying,
1
= u3/2 + C
6
1
= (4x + 3)3/2 + C.
6

Trigonometric Integrands
We can apply the change of variable technique to trigonometric functions as long as u is a dierentiable
function of x. Before we show how, recall the basic trigonometric integrals:

287

www.ck12.org


cos udu = sin u + C,

sin udu = cos u + C,

sec2 udu = tan u + C,

csc2 udu = cot u + C,

(sec u)(tan u)du = sec u + C,

(csc u)(cot u)du = csc u + C.

Example 3:

Evaluate cos(3x + 2)dx.

Solution:
The argument of the cosine function is 3x + 2. So we let u = 3x + 2. Then du = 3dx, or dx = du/3.
Substituting,

1
cos u dx
3

1
=
cos udx.
3

cos(3x + 2)dx =

Integrating,
1
sin u + C
3
1
= sin(3x + 2) + C.
3
=

Example 4:
This example requires us to use trigonometric identities before we substitute. Evaluate

1
dx.
cos2 3x
Solution:
Since sec 3x =

1
cos 3x ,

the integral becomes

1
dx =
cos2 3x

sec2 3xdx.

Substituting for the argument of the secant, u = 3x, then du = 3dx, or dx = du/3. Thus our integral
becomes,

1
1
sec2 udu
sec2 u. du =
3
3
1
= tan u + C
3
1
= tan(3x) + C.
3
www.ck12.org

288

Some integrations of trigonometric functions involve the logarithmic functions as a solution, as shown in
the following example.
Example 5:

Evaluate tan xdx.

Solution:
As you may have guessed, this is not a straightforward integration. We need to make use of trigonometric
identities to simplify it. Since tan x = sin x/ cos x,

sin x
tan xdx =
dx.
cos x
Now make a change of variable x. Choose u = cos x. Then du = sin xdx, or dx = du/ sin x. Substituting,
(
)

sin x du
sin x
dx =
cos x
u sin x

du
=
.
u
This integral should look obvious to you. The integrand is the derivative of the natural logarithm ln u.
= ln |u| + C
= ln | cos x| + C.
1
, is
Another way of writing it, since ln |u| = ln |u|



1 

+C
= ln 
cos x 
= ln | sec x| + C.

Using Substitution on Denite Integrals

Example 6:
3
Evaluate 1

x
dx.
2x1

Solution:
Let u = 2x 1. Then du = 2dx, or dx = du/2. Before we substitute, we need to determine the new limits
of integration in terms of the u variable. To do so, we simply substitute the limits of integration into
u = 2x 1:
Lower limit: For x = 1, u = 2(1) 1 = 1.
Upper limit: For x = 3, u = 2(3) 1 = 5.
We now substitute u and the associated limits into the integral:

x du
.

u 2

As you may notice, the variable x is still hanging there. To write it in terms of u, since u = 2x 1, solving
for x, we get, x = (u + 1)/(2). Substituting back into the integral,

289

www.ck12.org


=
=
=
=
=

u + 1 du

1 2 u 2

1 5 u+1
du
4 1
u
5
1
(u + 1)u1/2 du
4 1

1 5 1/2
(u + u1/2 )du
4 1
[
]5
1 2u3/2 2u1/2
+
.
4
3
1 1

Applying the Fundamental Theorem of Calculus by inserting the limits of integration and calculating,

1 2(5)3/2 2(5)1/2 2(1)3/2 2(1)1/2

.
=
+
+

4
3
1
3
1
Calculating and simplifying, we get

4 52
.
=
3

We could have chosen u = 2x 1 instead. You may want to try to solve the integral with this substitution.
It might be easier and less tedious.
Example 7:
Lets try the substitution method of denite integrals with a trigonometric integrand.
/4
Evaluate 0 tan x sec2 xdx.
Solution:
Try u = tan x. Then du = sec2 xdx, dx = du/ sec2 x.
Lower limit: For x = 0, u = tan 0 = 0.
Upper limit: For x = /4, u = tan 4 = 1.
Thus

/4

tan x sec xdx =

0

udu
0
[ 2 ]1
u

1
1
= 0= .
2
2

Multimedia Links
For video presentations on integration by substitution (17.0), see Math Video Tutorials by James Sousa,
Integration by Substitution, Part 1 of 2 (9:42) and Math Video Tutorials by James Sousa, Integration by
Substitution, Part 2 of 2 (8:17) .
www.ck12.org

290

Figure 7.1: Integration Using Substitution - Part 1 of 2 (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/559

Figure 7.2: Integration Using Substitution - Part 2 of 2 (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/560

Review Questions
In the following exercises, evaluate the integrals.

3
2 dx
(x8)

2. 2 + xdx
1 dx
3.
x2+x
2
4. x+1 dx
ex
5. ex +2 dx

6. 3 tt+5 dt
2 dx
7.
3x1
8. sin x cos xdx

9. cos x 1 cos2 xdx

10. sin5 x cos xdx
11. x3 cos 4x4 dx
12. sec2 (2x + 4)dx
2
2
13. 0 xe x dx

14. 0 x sin x2 dx
1
15. 0 x(x + 5)4 dx

1.

Review Answers
3
1. x8
+C
2
2. 3 (2 + x)3/2 + C

3. 2 2 + x + C

291

www.ck12.org

4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.

x2
2

x 32 + ln |x + 1| + C

ln |e x + 2| + C
6 t + 5 ln |t| + C
4
3 3x 1 + C
1
2
2 sin x + C
1
2
2 cos x + C
1
6
6 sin x + C
1
4
16 sin 4x + C
1
2 tan(2x + 4) + C
1 4
2 (e 1)
1
520 56

7.2 Integration By Parts

Learning Objectives
A student will be able to:
Compute by hand the integrals of a wide variety of functions by using technique of Integration by
Parts.
Combine this technique with the usubstitution method to solve integrals.
Learn to tabulate the technique when it is repeated.
In this section we will study a technique of integration that involves the product of algebraic and exponential
or logarithmic functions, such as

x ln xdx
and

xe x dx.
Integration by parts is based on the product rule of dierentiation that you have already studied:
d
dv
du
[uv] = u + v .
dx
dx
dx
If we integrate each side,

dv
du
uv =
u dx + v dx
dx
dx

=
udv + vdu.
Solving for

udv,

udv = uv

www.ck12.org

292

vdu.

This is the formula for integration by parts. With the proper choice of u and dv, the second integral may
be easier to integrate. The following examples will show you how to properly choose u and dv.
Example 1:

Evaluate x sin xdx.

Solution:
We use the formula

udv = uv

vdu.

Choose
u=x
and
dv = sin xdx.
To complete the formula, we take the dierential of u and the simplest antiderivative of dv = sin xdx.
du = dx
v = cos x.
The formula becomes

x sin xdx = x cos x

( cos x)dx

= x cos x +

cos xdx

= x cos x + sin x + C.
A Guide to Integration by Parts
Which choices of u and dv lead to a successful evaluation of the original integral? In general, choose u to
be something that simplies when dierentiated, and dv to be something that remains manageable when
integrated. Looking at the example that we have just done, we chose u = x and dv = sin xdx. That led to
a successful evaluation of our integral. However, lets assume that we made the following choice,
u = sin x
dv = xdx.
Then
du = cos xdx
v = x2 /2.
Substituting back into the formula to integrate, we get

udv = uv vdu
2
x
x2
cos xdx
= sin x
2
2

293

www.ck12.org

As you can see, this integral is worse than what we started with! This tells us that we have made the
wrong choice and we must change (in this case switch) our choices of u and dv.

Remember, the goal of the integration by parts is to start with an integral in the form udv that is hard
to integrate directly and change it to an integral vdu that looks easier to evaluate. However, here is a
general guide that you may nd helpful:
1. Choose dv to be the more complicated portion of the integrand that ts a basic integration formula.
Choose u to be the remaining term in the integrand.
2. Choose u to be the portion of the integrand whose derivative is simpler than u. Choose dv to be the
remaining term.
Example 2:

Evaluate xe x dx.
Solution:
Again, we use the formula

udv = uv

vdu.

Let us choose
u=x
and
dv = e x dx.
We take the dierential of u and the simplest antiderivative of dv = e x dx:
du = dx
v = ex .
Substituting back into the formula,

udv = uv
= xe x

vdu

e x dx.

We have made the right choice because, as you can see, the new integral
than our original integral. Integrating, we nally obtain our solution

xe x dx = xe x e x + C.

vdu =

e x dx is denitely simpler

Example 3:

Evaluate ln xdx.
Solution:
Here, we only have one term, ln x. We can always assume that this term is multiplied by 1:

ln x1dx.
www.ck12.org

294

So let u = ln x, and dv = 1dx. Thus du = 1/xdx and v = x. Substituting,

udv = uv vdu

1
ln xdx = x ln x
x dx
x

= x ln x 1dx
= x ln x x + C.

Repeated Use of Integration by Parts

Oftentimes we use integration by parts more than once to evaluate the integral, as the example below
shows.
Example 4:

Evaluate x2 e x dx.
Solution:
With u = x2 , dv = e x dx, du = 2xdx, and v = e x , our integral becomes

2 x
2 x
x e dx = x e 2 xe x dx.
As you can see, the integral has become less complicated
than the original, x2 e x xe x . This tells us that

x
we have made the right choice. However, to evaluate xe dx we still need to integrate by parts with u = x
and dv = e x dx. Then du = dx and v = e x , and

2 x
2 x
x e dx = x e 2 xe x dx
[
]

2 x
= x e 2 uv udv
[
]

2 x
x
x
= x e 2 xe e dx
= x2 e x 2xe x + 2e x + C.

Actually, the method that we have just used works for any integral that has the form xn e x dx, where n
is a positive integer. The following section illustrates a systematic way of solving repeated integrations by
parts.

Tabular Integration by Parts

Sometimes, we need to integrate by parts several times. This leads to cumbersome calculations. In
situations like these it is best to organize our calculations to save us a great deal of tedious work and to
avoid making unpredictable mistakes. The example below illustrates the method of tabular integration.
Example 5:

Evaluate x2 sin 3xdx.

Solution:

295

www.ck12.org

Begin as usual by letting u = x2 and dv = sin 3xdx. Next, create a table that consists of three columns, as
shown below:

www.ck12.org

296

Table 7.1:
Alternate signs

u and its derivatives

dv and its antiderivatives

x2
2x
2
0

sin 3x
1
3 cos 3x
1
9 sin 3x
1
27 cos 3x

To nd the solution for the integral, pick the sign from the rst row (+), multiply it by u of the rst row
(x2 ) and then multiply by the dv of the second row, 1/3 cos 3x (watch the direction of the arrows.) This
is the rst term in the solution. Do the same thing to obtain the second term: Pick the sign from the
second row, multiply it by the u of the same row and then follow the arrow to multiply the product by the
dv in the third row. Eventually we obtain the solution

2
2
1 2
x2 sin 3xdx =
x cos 3x + x sin 3x +
cos 3x + C.
3
9
27

Solving for an Unknown Integral

There are some integrals that require us to evaluate two integrations by parts, followed by solving for the
unknown integral. These kinds of integrals crop up often in electrical engineering and other disciplines.
Example 6:

Evaluate e x cos xdx.

Solution:
Let u = e x , and dv = cos xdx. Then du = e x dx, v = sin x, and

e x cos xdx = e x sin x e x sin xdx.

Notice that the second integral looks the same as our original integral in form, except that it has a sin x
instead of cos x. To evaluate it, we again apply integration by parts to the second term with u = e x , dv =
sin xdx, v = cos x, and du = e x dx.
Then
[

]
( cos x)(e dx)

e cos xdx = e sin x e cos x

= e x sin x e x cos x e x cos xdx.

x

Notice that the unknown integral now appears on both sides of the equation. We can simply move the
unknown integral on the right to the left side of the equation, thus adding it to our original integral:

2 e x cos xdx = e x sin x + e x cos x + C.

Dividing both sides by 2, we obtain

1
1
1
e x cos xdx = e x sin x + e x cos x + C.
2
2
2

297

www.ck12.org

Since the constant of integration is just a dummy constant, let

C
2

C.

Finally, our solution is

1
1
e x cos xdx = e x sin x + e x cos x + C.
2
2

Multimedia Links
To see this same classic example worked out with narration 17.0, see Khan Academy Indenite Integration Series Part 7 (9:39) .

Figure 7.3: Indenite Integration (part 7) (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/561
For additional video presentations on integration by parts 17.0, see Math Video Tutorials by James Sousa,
Integration by Parts, Basic (7:08) ; Math Video Tutorials by James Sousa, Integration by Parts (10:03) ;

Figure 7.4: Integration by Parts: The Basics (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/562

Figure 7.5: Integration by Parts (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/563
Math Video Tutorials by James Sousa, Integration by Parts, Additional Examples (7:47) .
www.ck12.org

298

Figure 7.6: Integration by Parts - Additional Examples (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/564

Review Questions
Evaluate the following integrals. (Remark: Integration by parts is not necessarily a requirement to solve
the integrals. In some, you may need to use usubstitution along with integration by parts.)

1.
2.
3.
4.
5.
6.

3xe x dx
x2 ex dx
ln(3x + 2)dx
sin1 xdx
sec3 xdx
2x ln(3x)dx

(lnx)2
7.
x dx
8. Use both the method of usubstitution and the method of integration
by parts to integrate the

integral below. Both methods will produce equivalent answers.
x 5x 2dx

9. Use the method of tabular integration by parts to solve x2 e5x dx.

1
10. Evaluate the denite integral 0 x2 e x dx.
3
11. Evaluate the denite integral 1 ln(x + 1)dx.

Review Answers
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.

3xe x 3e x + C
ex (x2 + 2x + 2) + C
3x+2
2| 1] + C
3 [ln |3x +

1
x sin x + 1 x2 + C
1
1
2 (sec x)(tan x) + 2 ln | sec x + tan x| + C
x2 ln |3x| 12 x2 + C
1
3
3 (ln x) + C
2
4
2)5/2 + 75
(5x 2)3/2 + C
125 (5x
[ 2
]
2x
2
e5x x5 25
+ 125
+C
e2
6 ln 2 + 2

299

www.ck12.org

7.3 Integration by Partial Fractions

Learning Objectives
A student will be able to:
Compute by hand the integrals of a wide variety of functions by using technique of Integration by
Partial Fractions.
Combine the technique of partial fractions with usubstitution to solve various integrals.
This is the third technique that we will study. This technique involves decomposing a rational function
into a sum of two or more simple rational functions. For example, the rational function
x+4
x2 + x 2
can be decomposed into
x2

x+4
3
2
+
.
=
x+2 x1
+x2

The two partial sums on the right are called partial factions. Suppose that we wish to integrate the rational
function above. By decomposing it into two partial fractions, the integral becomes manageable:

(
x+4
2
3 )
dx
=
+
x+2 x1
x2 + x 2

1
1
dx + 3
dx
=2
x+1
x1
= 2 ln |x + 1| + 3 ln |x 1| + C.
To use this method, we must be able to factor the denominator of the original function and then decompose
the rational function into two or more partial fractions. The examples below illustrate the method.
Example 1:
Find the partial fraction decomposition of
2x 19
.
+x6

x2
Solution:

We begin by factoring the denominator as x2 + x 6 = (x + 3)(x 2). Then write the partial fraction
decomposition as
2x 19
A
B
=
+
.
x+3 x2
+x6

x2

Our goal at this point is to nd the values of A and B. To solve this equation, multiply both sides of the
equation by the factored denominator (x + 3)(x 2). This process will produce the basic equation.
2x 19 = A(x 2) + B(x + 3).
www.ck12.org

300

This equation is true for all values of x. The most convenient values are the ones that make a factor equal
to zero, namely, x = 2 and x = 3. Substituting x = 2,
2(2) 19 = A(2 2) + B(2 + 3)
15 = 0 + 5B
3 = B
Similarly, substituting for x = 3 into the basic equation we get
2(3) 19 = A(3 2) + B(3 + 3)
25 = 5A + 0
5=A
We have solved the basic equation by nding the values of A and B. Therefore, the partial fraction decomposition is
5
2x 19
3
=

.
x
+
3
x

2
+x6

x2

General Description of the Method

To be able to write a rational function f (x)/g(x) as a sum of partial fractions, must apply two conditions:
The degree of f (x) must be less than the degree of g(x). If so, the rational function is called proper.
If it is not, divide f (x) by g(x) (use long division) and work with the remainder term.
The factors of g(x) are known. If not, you need to nd a way to nd them. The guide below shows
how you can write f (x)/g(x) as a sum of partial fractions if the factors of g(x) are known.
A Guide to Finding Partial Fractions Decomposition of a Rational Function
1. To nd the partial fraction decomposition of a proper rational function, f (x)/g(x), factor the denominator g(x) and write an equation that has the form
f (x)
= (sum of partial fractions.)
g(x)
2. For each distinct factor ax + b, the right side must include a term of the form
A
.
ax + b
3. For each repeated factor (ax + b)n , the right side must include n terms of the form
A1
A2
A3
An
+
+
+ ... +
.
2
3
(ax + b) (ax + b)
(ax + b)n
(ax + b)
Example 2:
Use the method of partial fractions to evaluate

x+1
dx
(x+2)2

Solution:

301

www.ck12.org

According to the guide above (item #3), we must assign the sum of n = 2 partial sums:
x+1
A
B
=
+
.
2
(x + 2) (x + 2)2
(x + 2)
Multiply both sides by (x + 2)2 :
x + 1 = A(x + 2) + B
x + 1 = Ax + (2A + B).
Equating the coeicients of like terms from both sides,
1=A
1 = 2A + B.
Thus
A = 1.
B = 1.
Therefore the partial fraction decomposition is
x+1
1
1
=

.
2
x + 2 (x + 2)2
(x + 2)
The integral will become

)
1
1

x + 2 (x + 2)2

1
1
dx
dx
=
x+2
(x + 2)2


1
= ln  x + 2 +
+ c,
x+1

x+1
dx =
(x + 2)2

where we have used usubstitution for the second integral.

Example 3:
2 +3x+1
Evaluate 3x
dx .
x3 +2x2 +x
Solution:
We begin by factoring the denominator as x(x + 1)2 . Then the partial fraction decomposition is
3x2 + 3x + 1
A
B
C
= +
+
.
3
2
x
x + 1 (x + 1)2
x + 2x + x
Multiplying each side of the equation by x(x + 1)2 we get the basic equation
3x2 + 3x + 1 = A(x + 1)2 + Bx(x + 1) + Cx.
This equation is true for all values of x. The most convenient values are the ones that make a factor equal
to zero, namely, x = 1 and x = 0.
www.ck12.org

302

Substituting x = 1,
3(1)2 + 3(1) + 1 = A(1 + 1)2 + B(1)(1 + 1) + C(1)
1 = 0+0C
1 = C.
Substituting x = 0,
3(0)2 + 3(0) + 1 = A(0 + 1)2 + B(0)(0 + 1) + C(0)
1= A+0+0
1 = A.
To nd B we can simply substitute any value of x along with the values of A and C obtained.
Choose x = 1:
3(1)2 + 3(1) + 1 = A(1 + 1)2 + B(1)(1 + 1) + C(1)
7 = 4 + 2B 1
2 = B.
Now we have solved for A, B, and C. We use the partial fraction decomposition to integrate.
)

(
3x2 + 3x + 1
1
2
1
dx. =
+

dx
x x + 1 (x + 1)2
x3 + 2x2 + x
 


1
= ln  x + 2ln  x + 1 +
+ C.
x+1
Example 4:
This problem is an example of an improper rational function. Evaluate the denite integral
2 3
x 4x2 3x + 3
dx.
x2 3x
1
Solution:
This rational function is improper because its numerator has a degree that is higher than its denominator.
The rst step is to divide the denominator into the numerator by long division and obtain
x3 4x2 3x + 3
6x + 3
= (x 1) + 2
.
x2 3x
x 3x
Now apply partial function decomposition only on the remainder,
B
6x + 3
6x + 3
A
.
=
= +
2
x
x3
x(x 3)
x 3x
As we did in the previous examples, multiply both sides by x(x 3) and then set x = 0 and x = 3 to obtain
the basic equation
6x + 3 = A(x 3) + Bx

303

www.ck12.org

For x = 0,
3 = 3A + 0
1 = A.
For x = 3,
18 + 3 = 0 + 3B
15 = 3B
5 = B.
Thus our integral becomes
2[
2 3
2[
x 4x2 3x + 3
6x + 3 ]
1
5 ]
dx =
(x 1) + 2
dx =
x1
dx.
x x3
x2 3x
x 3x
1
1
1
Integrating and substituting the limits,
[

]
 

 2
x2
=
x ln  x 5ln  x 3
2
1
(4
) (1
)
=
2 ln2 5ln1
1 ln1 5ln2
2
2
1
= 4ln2 .
2

Multimedia Links
For a complete partial fractions problem (19.0), see Integration by Partial Fractions, Just Math Tutoring
(6:02) and for integration using partial fractions and a rationalizing substitution (19.0), see Integration

Figure 7.7: A Complete Partial Fractions Problem (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/565
using Partial Fractions and a rationalizing substitution, Just Math Tutoring (6:06) .
For an extensive presentation on integrating with partial fractions including the completing the square
technique (19.0) see Integration with partial fractions using various techniques, MIT Courseware (51:24)
.

Review Questions
Evaluate the following integrals.
www.ck12.org

304

Figure 7.8: Integration Using Partial Fractions / Rationalizing Sub (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/566

Instructor: David Jerison

View the complete course at: http://ocw.mit.edu/18-01F06
License: Creative Commons BY-NC-SA More information at http://ocw.mit.edu/terms More courses at
http://ocw.mit.edu
Instructor: David Jerison
View the complete course at: http://ocw.mit.edu/18-01F06
License: Creative Commons BY-NC-SA More information at http://ocw.mit.edu/terms More courses at
http://ocw.mit.edu
Figure 7.9: Lec 30 | MIT 18.01 Single Variable Calculus, Fall 2007 (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/567

305

www.ck12.org


1. x211 dx
x
2. x2 2x3
dx
3. x3 +x12 2x dx
3
4. x2x+4 dx
1
5. 0 1+
d
5 x1
6. 1 x2 (x+1) dx
7. Evaluate the integral by making the proper usubstitution to convert to a rational function:

cos
d.
sin2 +4 sin 5

3e
d.
e2 1

8. Evaluate the integral by making the proper usubstitution to convert to a rational function:
9. Find the area under the curve y = 1/(2+e x ), over the interval [ln3, ln4]. (Hint: make a usubstitution
to convert the
into
a rational function.)
 a+x
integrand

1
1

10. Show that a2 x
dx
=
ln
2
2a
ax + C.

Review Answers
 x1 
+C
1. 12 ln  x+1




1
2. 4 ln  x + 1 + 34 ln  x 3 + C  
3. 16 ln  x + 2 + 13 ln  x 1 12 ln  x + C
4. 12 x2 2ln(x2 + 4) + C
5. 1 ln2
5
6. 4
5 +
( 2ln 3 )
sin 1
7. 16 ln 5+sin
 + C 


3

8. 2 ln e 1 + 23 ln e + 1 + C
9. 12 ln 14
3
10. Hint: Decompose the integrand into partial fractions.

7.4 Trigonometric Integrals

Learning Objectives
A student will be able to:
Compute by hand the integrals of a wide variety of functions by using the Trigonometric Integrals.
Combine this technique with usubstitution.

Integrating Powers of Sines and Cosines

In this section we will study methods of integrating functions of the form

sinm xcosn xdx,

where m and n are nonnegative integers. The method that we will describe uses the famous trigonometric
identities
sin2 x = 12 (1 cos 2x) and
cos2 x = 12 (1 + cos 2x).
www.ck12.org

306

Example 1:

Evaluate sin2 xdx and cos2 xdx.

Solution:
Using the identities above, the rst integral can be

sin2 xdx =

written as

1
(1 cos 2x)dx
2
1
(1 cos 2x)dx
2
1
1
= (x sin 2x) + C
2
2
x 1
= sin 2x + C.
2 4

Similarly, the second integral can be written as

1
2
cos xdx =
(1 + cos 2x)dx
2

1
=
(1 + cos 2x)dx
2
)
1
1(
x + sin 2x + C
=
2
2
x 1
= + sin 2x + C.
2 4
Example 2:

Evaluate cos4 xdx.

Solution:

cos xdx =
=
=
=

(
)2
1
(cos x) dx =
(1 + cos 2x) dx
2

1
(1 + 2 cos 2x + cos2 2x)dx
4
(
)
1
1 1
1 + 2 cos 2x + + cos 4x dx
4
2 2
(
)
1
3
1
+ 2 cos 2x + cos 4x dx.
4
2
2
2

Integrating term by term,

]
1
1 [3
x + sin 2x + sin 4x + C
4 2
8
3
1
1
= x + sin 2x +
sin 4x + C.
8
4
32
=

Example 3:

Evaluate sin3 xdx.

Solution:

sin3 xdx =

sin2 x sin xdx

307

www.ck12.org

Recall that sin2 x + cos2 x = 1, so by substitution,

= (1 cos2 x) sin xdx

=
sin xdx cos2 x sin xdx.
The rst integral should be straightforward. The second can be done by the method of usubstitution by
letting u = cos x, so du = sin xdx. The integral becomes
]
[
du
= cos x
u2 sin x
sin x

= cos x + u2 du
u3
+C
3
1
= cos x + cos3 x + C.
3
= cos x +

If m and n are both positive integers, then an integral of the form

sinm x cosn xdx

can be evaluated by one of the procedures shown in the table below, depending on whether m and n are
odd or even.
Table 7.2:

sinm x cosn xdx

n odd
m odd
n and m even

Procedure

Identities

Let u = sin x
Let u = cos x
Use identities to reduce powers

cos2 x = 1 sin2 x
sin2 x = 1 cos2 x
sin2 x = (1/2)(1 cos 2x)
cos2 x = (1/2)(1 + cos 2x)

Example 4:
Evaluate sin3 x cos4 xdx.
Solution:
Here, m is odd. So according to the second procedure in the table above, let u = cos x, so du = sin x.
Substituting,

1
u4 sin3 x
du
sin x

= u4 sin2 xdu.

sin3 x cos4 xdx =

Referring to the table again, we can now substitute sin2 x = 1 cos2 x in the integral:
www.ck12.org

308

u4 (1 cos2 x)du

= u4 (1 u2 )du

= (u4 + u6 )du

1 5 1 7
u + u +C
5
7
1
1
= cos5 x + cos7 x + C.
5
7

Example 5:

Evaluate sin4 x cos4 xdx.

Solution:
Here, m = n = 4. We follow the third procedure in the table above:

sin4 x cos4 xdx = (sin2 x)2 (cos2 x)2 dx

[
]2 [ 1
]2
1
(1 cos 2x)
(1 + cos 2x) dx
=
2
2

1
=
(1 cos2 2x)2 dx
16

1
sin4 2xdx.
=
16
At this stage, it is best to use usubstitution to integrate. Let u = 2x, so du = 2dx.

1
4
4
sin x cos xdx =
sin4 udu
32

[
]2
1
1
1
2
2
(sin u) du =
(1 cos 2u) du
=
32
32
2
(
)
1 3
1
1
=
u sin 2u +
sin 4u + C
32 8
4
32
3
1
1
=
x
sin 4x +
sin 8x + C.
256
128
1024

Integrating Powers of Secants and Tangents

In this section we will study methods of integrating functions of the form

tanm x secn xdx,

where m and n are nonnegative integers. However, we will begin with the integrals

tan xdx
and

309

www.ck12.org


sec xdx.
The rst integral can be evaluated by writing

sin x
tan xdx =
dx.
cos x
Using usubstitution, let u = cos x, so du = sin xdx. The integral becomes

sin x 1
tan xdx =
u sin x

1
du =
du = ln |u| + C
u
= ln | cos x| + C
= ln(1/| cos x|) + C
= ln | sec x| + C.
The second integral

sec xdx, however, is not straightforwardit requires a trick. Let

sec x + tan x
sec xdx =
sec x
dx
sec x + tan x

sec2 x + sec x tan x

=
dx.
sec x + tan x

Use usubstitution. Let u = sec x + tan x, then du = (sec2 x + sec x tan x)dx, the integral becomes,

du
sec xdx =
u
= ln |u| + C
= ln | sec x + tan x| + C.
There are two reduction formulas that help evaluate higher powers of tangent and secant:

secn2 x tan x n 2
n
+
secn2 xdx,
sec xdx =
n1
n1

tanm1 x
m
tan xdx =
tanm2 xdx.
m1
Example 6:

Evaluate sec3 xdx.

Solution:
We use the formula above by substituting for n = 3.

sec x tan x 3 2
3
sec xdx =
+
sec xdx
31
31

1
1
= sec x tan x +
sec xdx
2
2
1
1
= sec x tan x + ln | sec x + tan x| + C.
2
2
www.ck12.org

310

Example 7:

Evaluate tan5 xdx.

Solution:
We use the formula above by substituting for m = 5.

tan4 x
tan3 xdx.
tan5 xdx =
4

We need to use the formula again to solve the integral tan3 xdx:

tan4 x
5
tan3 xdx
tan xdx =
4
[
]

tan4 x
tan2 x
=

tan xdx
4
2
1
1
= tan4 x tan2 x ln | cos x| + C.
4
2
If m and n are both positive integers, then an integral of the form

tanm x secn xdx

can be evaluated by one of the procedures shown in the table below, depending on whether m and n are
odd or even.
Table 7.3:

tanm x secn xdx

n even
m odd
m even
n odd

Procedure

Identities

Let u = tan x
Let u = sec x
Reduce powers of sec x

sec2 x = tan2 x + 1
tan2 x = sec2 x 1
tan2 x = sec2 x 1

Example 8:

Evaluate tan2 x sec4 xdx.

Solution:
Here n = 4 is even, and so we will follow the rst procedure in the table above. Let u = tan x, so
du = sec2 xdx. Before we substitute, split o a factor of sec2 x.

tan2 x sec4 xdx =

tan2 x sec2 x sec2 xdx.
Since sec2 x = tan2 x + 1,

tan2 x(tan2 x + 1) sec2 xdx.

Now we make the usubstitution:

311

www.ck12.org

u2 (u2 + 1)du

1
1
= u5 + u3 + C
5
3
1
1
5
= tan x + tan3 x + C.
5
3
Example 9:

Evaluate tan3 x sec3 xdx.

Solution:
Here m = 3 is odd. We follow the third procedure in the table. Make the substitution, u = sec x and
du = sec x tan xdx. Our integral becomes

tan3 x sec3 xdx =

tan2 x sec2 x(sec x tan x)dx

= (sec2 x 1) sec2 x(sec x tan x)dx

= (u2 1)u2 du
1
1
= u5 u3 + C
5
3
1
1
= sec5 x sec3 x + C.
5
3

Multimedia Links
For video presentations on computing the integrals of trigonometric functions (20.0), see Trigonometric
Integrals, Part 1 (5:57) ; see Trigonometric Integrals, Part 2 (6:01) ; see Trigonometric Integrals, Part

For more free math videos, visit http://PatrickJMT.com

For more free math videos, visit http://PatrickJMT.com
Figure 7.10: Trigonometric Integrals - Part 1 of 6 (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/568
3 (5:54) ; see Trigonometric Integrals, Part 4 (8:57) ; see Trigonometric Integrals, Part 5 (5:58) ; see
Trigonometric Integrals, Part 6 (8:42) .

Review Questions
Evaluate the integrals.
www.ck12.org

312

Figure 7.11: Trigonometric Integrals - Part 2 of 6 (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/569

Figure 7.12: Trigonometric Integrals - Part 3 of 6 (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/570

Figure 7.13: Trigonometric Integrals - Part 4 of 6 (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/571

For more free math videos, visit http://PatrickJMT.com

For more free math videos, visit http://PatrickJMT.com
Figure 7.14: Trigonometric Integrals - Part 5 of 6 (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/572

313

www.ck12.org

Figure 7.15: Trigonometric Integrals - Part 6 of 6 (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/573
1.
2.
3.
4.
5.
6.
7.

cos4 x sin xdx

sin2 5d
sin2 2z cos3 2zdz
sin x cos(x/2)dx
sec4 x tan3 xdx
tan4 x sec xdx

tan x sec4 xdx

/2
8. 0 tan5 2x dx
9. Graph and then nd the volume of the solid that results when the region enclosed by y = sin x, y =
cos x, x = 0, and x = /4 is revolved around the xaxis.
10.

11. (a) Prove that csc xdx = ln | csc x + cot x| + C

(b) Show that it can also be written in the following two forms: csc xdx = ln | tan 12 x|+C = ln | csc x
cot x| + C.

Review Answers
1. 15 cos5 x + C
1
2. 12 20
sin 10 + C
1
1
3
3. 6 sin 2z 10
sin5 2z + C
3x
x
4. 1
3 cos( 2 ) cos( 2 ) + C
1
1
7
5
5. 7 sec x 5 sec x + C
6. 14 sec3 x tan x 58 sec x tan x + 38 ln | sec x + tan x| + C
7. 23 tan3/2 x + 27 tan7/2 x + C
8. 1
2 + ln 2
9. V = /2
10. .

7.5 Trigonometric Substitutions

Learning Objectives
A student will be able to:
Compute by hand the integrals of a wide variety of functions by using technique of Trigonometric
Substitution.
www.ck12.org

314

 Combine this technique with other integration techniques to integrate.

When we are faced with integrals that involve radicals of the forms a2 x2 , x2 a2 , and x2 + a2 , we
may make substitutions that involve trigonometric functions to eliminate the radical. For example, to
eliminate the radical in the expression

a2 x2

we can make the substitution

x = a sin ,
/2 /2,
(Note: must be limited to the range of the inverse sine function.)
which yields,

a2

x2

a2

a2 sin2

a2 (1 sin2 )

= a cos2 = a cos .
The reason for the restriction /2 /2 is to guarantee that sin is a one-to-one function on this
interval and thus has an inverse.
The table below lists the proper trigonometric substitutions that will enable us to integrate functions with
radical expressions in the forms above.
Table 7.4:
Expression in Integrand

2
2
a x
2
2
a + x
x 2 a2

Substitution

Identity Needed

x = a sin
x = a tan
x = a sec

1 sin2 = cos2
1 + tan2 = sec2 x
sec2 1 = tan2

In the second column are listed the most common substitutions. They come from the reference right
triangles, as shown in the gure below. We want any of the substitutions we use in the integration to be
reversible so we can change back to the original variable afterward. The right triangles in the gure below
will help us reverse our substitutions.

Description: 3 triangles.
Example 1:

315

www.ck12.org

Evaluate

dx
.
x2 4x2

Solution:
Our goal rst is to eliminate the radical. To do so, look up the table above and make the substitution
x = 2 sin , /2 /2,
so that
dx
= 2 cos
d
Our integral becomes

2 cos d

(2 sin )2 4 4 sin2

2 cos d
=
(2 sin )2 (2 cos )

1
d
=
2
4
sin
1
csc2 d
=
4
1
= cot + C.
4

dx
=

2
x 4 x2

Up to this stage, we are done integrating. To complete the solution however, we need to express cot in
terms of x. Looking at the gure of triangles
above, we can see that the second triangle represents our case,
with a = 2. So x = 2 sin and 2 cos = 4 x2 , thus

cot =

4 x2
,
x

since
cot =

cos
.
sin

so that

dx
1
= cot + C

4
x 2 4 x2

1 4 x2
=
+ C.
4
x

Example 2:
x2 3
Evaluate
x dx.
Solution:
Again,we want to rst to eliminate the radical. Consult the table above and substitute x =
dx = 3 sec tan d. Substituting back into the integral,
www.ck12.org

316

3 sec . Then

x2 3
dx =
x

3 sec2 3
3 sec tan d

3
sec

tan2 d.

Using the integral identity from the section on Trigonometric Integrals,

tanm1 x
m
tan xdx =
tanm2 xdx.
m1
and letting m = 2 we obtain

x2 3
dx = 3 tan + C.
x

Looking at the
triangle represents our case, with a =
triangles above, the third
thus cos x = 3/x, which gives tan = x2 3/ 3. Substituting,
2

x 3
dx = 3 tan + C
x

x2 3
1
2
= x 3 tan
+ C.
3
Example 3:

dx
Evaluate
2
2
x

x +1

3. So x =

3 sec and

Solution:
From the table above, let x = tan then dx = sec2 d. Substituting into the integral,

dx
sec2 d
.
=

x2 x 2 + 1
tan2 tan2 + 1
But since tan2 + 1 = sec2 ,

sec2 d
2
tan sec
sec
d
=
tan2

1 cos2
=
d
cos sin2

=
cot csc d.
=

Since

d
d (csc )

= cot csc ,

dx
=
cot csc d

x2 x2 + 1
= csc + C.

317

www.ck12.org

Looking at the triangles above, therst triangle represents our case, with a = 1. So x = tan and thus
1+x2
sin x = x 2 , which gives csc =
. Substituting,
x
1+x

dx
= csc + C

2
x x2 + 1

1 + x2
=
+ C.
x

Multimedia Links
For video presentations on Trigonometric Substitutions (17.0), see Trigonometric Substitutions, Just Math
Tutoring (9:30) and Trigonometric Substitutions, Part 2, Just Math Tutoring (4:20) .

Figure 7.16: Trigonometric Substitution - Example 3 / Part 1 (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/574

Figure 7.17: Trigonometric Substitution - Example 3 / Part 2 (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/575

Review Questions
Evaluate the integrals.

1. 4 x2 dx
1 dx
2.
2
x9+x
3

3.
dx
2
1x
1

4.
dx
19x
2
3
5. x 4 x2 dx
1
6.
dx
x2 1x2 36
7. (x2 +25)2 dx
www.ck12.org

318

8.

x3 16 x2 dx
00
e x 1 e2x dx

9.
(Hint: First use usubstitution, letting u = e x )
10. Graph and then nd the area of the surface generated by the curve y = x2 from x = 1 to x = 0 and
revolved about the xaxis.

Review Answers
( )

1. 2 sin1 2x + 12 x 4 x2 + C

2. ln( 13 | 9 + x2 + x|) + C

2
2
3. x 31x 23 1 x2 + C
4. 13 sin1 3x + C
5. 43 (4 x2 )3/2 + 15 (4 x2 )5/2 + C
6.
7.
8.
9.
10.

1 x2 36
+C
36
x
1
x
1
1
50 x2 +25 + 250 tan (x/5) + C
2048
15
[
]

1
1
1 e2

sin
(e
)

e
2 2
[
Surface area is 3 18 5 ln(2 +

]
5)

7.6 Improper Integrals

Learning Objectives
A student will be able to:
Compute by hand the integrals of a wide variety of functions by using the technique of Improper
Integration.
Combine this technique with other integration techniques to integrate.
Distinguish between proper and improper integrals.
The concept of improper integrals is an extension to the concept of denite integrals. The reason for
the term improper is because those integrals either
include integration over innite limits or
the integrand may become innite within the limits of integration.
b
We will take each case separately. Recall that in the denition of denite integral a f (x)dx we assume
that the interval of integration [a, b] is nite and the function f is continuous on this interval.

Integration Over Innite Limits

If the integrand f is continuous over the interval [a, ), then the improper integral in this case is dened
as

l
f (x)dx = lim
f (x)dx.
a

319

www.ck12.org

If the integration of the improper integral exists, then we say that it converges. But if the limit of
integration fails to exist, then the improper integral is said to diverge. The integral above has an important
geometric interpretation that you need to keep in mind. Recall that, geometrically, the denite integral
b
l
f (x)dx represents the area under the curve. Similarly, the integral a f (x)dx is a denite integral that
a
represents the area under the curve f (x) over the interval [a, l], as the gure below shows. However, as l
approaches , this area will expand
to the area under the curve of f (x) and over the entire interval [a, ).
Therefore, the improper integral a f (x)dx can be thought of as the area under the function f (x) over the
interval [a, ).

Example 1:

Evaluate 1 dx
x .
Solution:
We notice immediately that the integral is an improper integral because the upper limit of integration
approaches innity. First, replace the innite upper limit by the nite limit l and take the limit of l to
approach innity:

l
dx
dx
= lim
l 1 x
x
= lim ln x]l1
l

= lim (ln l ln 1)
l

= lim ln l
l

= .

Thus the integral diverges.

Example 2:

Evaluate 2 dx
.
x2
Solution:
www.ck12.org

320

dx
= lim
l
x2
= lim

= lim

dx
2
2 x
[ 1 ]l
x
( 1

1)
2

1
= .
2
Thus the integration converges to 21 .
Example 3:

Evaluate +

dx
.
1+x2

Solution:
What we need to do rst is to split the integral into two intervals (, 0] and [0, +). So the integral
becomes
+
0
+
dx
dx
dx
=
+
.
2
2
1 + x2
1 + x
1 + x
0
Next, evaluate each improper integral separately. Evaluating the rst integral on the right,
0
0
dx
dx
= lim
2
l l 1 + x2
1 + x
[
]0
= lim tan1 x
l
l
[
]
1
= lim tan 0 tan1 l
l
[
( )]
= lim 0
= .
l
2
2
Evaluating the second integral on the right,

0

dx
= lim
1 + x2 l

dx
2
0 1+x
[
]l
= lim tan1 x
l

= 0= .
2
2

Adding the two results,

dx
= + = .
2
2 2
1+x

Remark: In the previous example, we split the integral at x = 0. However, we could have split the
integral at any value of x = c without aecting the convergence or divergence of the integral. The choice
is completely arbitrary. This is a famous thoerem that we will not prove here. That is,
+
c
+
f (x)dx =
f (x)dx +
f (x)dx.

321

www.ck12.org

Integrands with Innite Discontinuities

This is another type of integral that arises when the integrand has a vertical asymptote (an innite
discontinuity) at the limit of integration or at some point in the interval of integration. Recall from
Chapter 5 in the Lesson on Denite Integrals that in order for the function f to be integrable, it must
be bounded on the interval [a, b]. Otherwise, the function is not integrable and thus does not exist. For
example, the integral

dx
x1

develops an innite discontinuity at x = 1 because the integrand approaches innity at this point. However,
it is continuous on the two intervals [0, 1) and (1, 4]. Looking at the integral more carefully, we may split
the interval [0, 4] [0, 1) (1, 4] and integrate between those two intervals to see if the integral converges.

dx
=
x1

dx
+
x1

4
1

dx
.
x1

We next evaluate each improper integral. Integrating the rst integral on the right hand side,

l
dx
dx
= lim
x 1 l1 0 x 1
= lim [ln |x 1|]l0
l1

= lim [ln |l 1| ln | 1|]

l1

= .
The integral diverges because ln(1) and ln(0) are not dened, and thus there is no reason to evaluate the
second integral. We conclude that the original integral diverges and has no nite value.
Example 4:
3
Evaluate 1 dx .
x1

Solution:

dx
x1

= lim

dx

x1
l
[
]3
= lim 2 x 1
l
l1+
]
[

= lim 2 2 2 l 1
l1+

= 2 2.
l1+

So the integral converges to 2 2.

Example 5:
In Chapter 5 you learned to nd the volume of a solid by revolving a curve. Let the curve be y = xex , 0
x and revolving about the xaxis. What is the volume of revolution?
Solution:
www.ck12.org

322

From the gure above, the area of the region to be revolved is given by A = y2 = x2 e2x . Thus the
volume of the solid is

l
2 2x
V=
x e dx = lim
x2 e2x dx.
l

As you can see, we need to integrate by parts twice:

x2 2x
2 2x
xe2x dx
x e dx = e +
2
x2
x
1
= e2x e2x e2x + C.
2
2
4
Thus
[

x2
x
1
V = lim e2x e2x e2x
l
2
2
4
[ 2
]l
2x + 2x + 1
= lim
l
4e2x
0
[ 2
]
2l + 2l + 1
1
= lim

l
4e0
4e2l
[ 2
]
2l + 2l + 1 1
= lim
+ .
l
4
4e2l

]l
0

At this stage, we take the limit as l approaches innity. Notice that the when you substitute innity
2 +2l+1
into the function, the denominator of the expression 2l 4e
, being an exponential function, will approach
2l
innity at a much faster rate than will the numerator. Thus this expression will approach zero at innity.
Hence
[
1]
V = 0+
= ,
4
4
So the volume of the solid is /4.
Example 6:

323

www.ck12.org

Evaluate

dx
.
e x +ex

Solution:
This can be a tough integral! To simplify, rewrite the integrand as
1
ex
ex
1
=
=
=
.
e x + ex
ex (e2x + 1)
e2x + 1
1 + (e x )2
Substitute into the integral:

dx
=
x
e + ex

ex
dx.
1 + (e x )2

Using usubstitution, let u = e x , du = e x dx.

dx
du
=
e x + ex
1 + u2
= tan1 u + C
= tan1 e x + C.

Returning to our integral with innite limits, we split it into two regions. Choose as the split point the
convenient x = 0.
+
0
+
dx
dx
dx
=
+
.
x
x
x
x
x
e + ex
e + e
e + e
0
Taking each integral separately,

dx
= lim
x
e + ex l

dx
+ ex
l
[
]0
= lim tan1 e x
l
l
[
]
1 0
= lim tan e tan1 el
ex

0
4

= .
4
=

Similarly,

dx
= lim
x
e + ex l

dx
+ ex
0
[
]l
= lim tan1 e x
0
l
[
]
1 l
= lim tan e tan1 1
ex

= .
2 4
4

Thus the integral converges to

www.ck12.org

dx
= + = .
e x + ex
4 4
2

324

Multimedia Links
For a video presentation of Improper Integrals (22.0), see Improper Integrals, www.justmathtutoring.com
(6:23) .

Figure 7.18: Improper Integral - Basic Idea and Example (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/576
For a video presentation of Improper Integrals with Innity in the Upper and Lower Limits (22.0), see
Improper Integrals, www.justmathtutoring.com (7:55) .

Figure 7.19: Improper Integral - Innity in Upper and Lower Limits (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/577

Review Questions
1. Determine whether the following integrals are improper. If so, explain why.
7
(a) 1 x+2
x3 dx
7 x+2
(b) 1 x+3 dx
1
(c) 0 lnxdx

(d) 0 1 dx
/4 x2
(e) 0 tan xdx
Evaluate the integral or state that it diverges.
2.
3.
4.
5.
6.

12 x[

2.001

dx

1
1
x1 x+1

0 5x
e dx

5
1
dx
3 (x3)4
/2
tan xdx
/2

dx

325

www.ck12.org

1
7. 0 1 2 dx
1x
8. The region between the xaxis and the curve y = ex for x 0 is revolved about the xaxis.
(a) Find the volume of revolution, V.
(b) Find the surface area of the volume generated, S .

Review Answers
1.
2. (a) Improper; innite discontinuity at x = 3.
(b) Not improper.
(c) Improper; innite discontinuity at x = 0.
(d) Improper; innite interval of integration.
(e) Not improper.
1
3. 1.001
4. ln 3
5. 15
6. divergent
7. divergent
8. 2 .
9.
10. (a) V = /2
[
]
(b) S = 2 + ln(1 + 2) .

Homework
Evaluate the following integrals.

sin xcos xdx
1.

2. x tan2 (x2 ) sec2 (x2 )dx

ln3
3. 0
e2x 1dx

4. 0 x12 dx
8
5. 1 31x dx
x2 +x16
6. (x+1)(x3)
2 dx
7. Graph and nd the volume of the region enclosed by the xaxis, the yaxis, x = 2 and y = x2 /(9 x2 )
when revolved about the xaxis.
8. The Gamma Function, (x), is an improper integral that appears frequently in quantum physics. It is
dened as

(x) =
t x1 et dt.
0

The integral converges for all x 0.

a. Find (1).
b. Prove that (x + 1) = x(x), for all x 0.
( )

c. Prove that 12 = .
www.ck12.org

326

9. Refer to the Gamma Function dened in the previous exercise to prove that
)
(
n
(a) 0 ex dx = n+1
, n 0 [Hint: Let t = xn ]
n
1
(b) 0 (lnx)n dx = (1)n (n + 1), n 0 [Hint: Let t = ln x]
10. In wave mechanics, a sawtooth wave is described by the integral

+/

t sin(kt)dt,

where k is called the wave number, is the frequency, and t is the time variable. Evaluate the integral.

Answers
1.
2.

3
2
1
6

sin2/3 x + C
tan3 (x2 ) + C

3. ln 3
4. divergent
5.

9
8
(x3)2 

 x + 1

6. ln 
7.

19
5

1
x3

+C

)
94 ln5

8. a. (1) = 1
9. a. Hint: Let t = xn .
b. Hint: t = ln x
10.

2
(k)2

sin(k)

7.7 Ordinary Dierential Equations

General and Particular Solutions
Dierential equations appear in almost every area of daily life including science, business, and many others.
We will only consider ordinary dierential equations (ODE). An ODE is a relation on a function y of one
independent variable x and the derivatives of y with respect to x, i.e. y(n) = F(x, y, y , . . . ., y(n1) ). For
example, y + (y )2 + y = x.

An ODE is linear if F can be written as a linear combination of the derivatives of y, i.e. y(n) = ai (x)y(i) +
r(x). A linear ODE is homogeneous if r(x) = 0.
A general solution to a linear ODE is a solution containing a number (the order of the ODE) of arbitrary
variables corresponding to the constants of integration. A particular solution is derived from the general
solution by setting the constants to particular values. For example, for linear ODE of second degree
y + y = 0, a general solution have the form yg = A cos x + B sin x where A, B are real numbers. By setting
A = 1 and B = 0, y p = cos x
It is generally hard to nd the solution of dierential equations. Graphically and numerical methods are
often used. In some cases, analytical method works, and in the best case, y has an explicit formula in x.

327

www.ck12.org

Multimedia Links
For a video introduction to dierential equations (27.0), see Math Video Tutorials by James Sousa,
Introduction to Dierential Equations (8:12) .

Figure 7.20: Introduction to Dierential Equations (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/578

Slope Fields and Isoclines

dy
We now only consider linear ODE of the rst degree, i.e. dx
= F(x, y). In general, the solutions of a
dierential equation could be visualized before trying an analytic method. A solution curve is the curve
that represents a solution (in the xy plane).

The slope eld of the dierential |eq|uation is the set of all short line segments through each point (x, y)
and with slope F(x, y).

An isocline (for constant k) is the line along which the solution curves have the same gradient (k). By
calculating this gradient for each isocline, the slope eld can be visualized; making it relatively easy to
dy
sketch approxi- mate solution curves. For example, dx
= yx . The isoclines are y = kx .
Example 1 Consider

dy
dx

The solutions are y3 =

3
2

www.ck12.org

x
.
y2

We briey sketch the slope eld as above.

x2 + C.

328

Exercise
1. Sketch the slope eld of the dierential equation
2. Sketch the slope eld of the dierential equation
curve that passes through (1, 0).

dy
dx
dy
dx

= 1 y. Sketch the solution curves based on it.

= y x. Find the isoclines and sketch a solution

Dierential Equations and Integration

We begin the analytic solutions of dierential equations with a simple type where F(x, y) is a function of x
dy
only. dx
= f (x) is a function of x. Then any antiderivative of f is a solution by the Fundamental Theorem
of Calculus:
x
d
dx a f (t) dt = f (x).
dy
Example 1 Solve the dierential equation dx
= x with y(0) = 1.

2
Solution. y = x dx = x2 + C. Then y(0) = 1 gives 1 = 0 + C, i.e. C = 1 Therefore y =

329

x2
2

+ 1.
www.ck12.org

dy
Example 2 Solve the dierential equation dx
= 1 2.
1+x

1
Solution. We have y = 2 dx and a substitution u = sinh1 x gives y = cos h u du
=
2
1+x
1+sin h u

du = u = sin h1 x + C.

cos h u du
cos h u

Exercise
1. Solve the dierential equation
2. Solve the dierential equation

dy
dx
dy
dx

= 9 x2 with y(0) = 3.
= 1 2.
1+x+x

Hint: Let u = x + 21 .

Solving Separable First-Order Dierential Equations

The next type of dierential equation where analytic solution are rela- tively easy is when the dependence
dy
of F(x, y) on x and y are separable: dx
= F(x, y) where F(x, y) = f (x)g(y) is the product of a functions of x
and y respectively. The solution is in the form P(x) = Q(y). Here g(y) is never 0 or the values of y in the
solutions will be restricted by where g(y) = 0.
Example 1 Solve the dierential equation y = xy with the initial condition y(0) = 1.
Solution. Separating x and y turns the equation in dierential form
have ln |y| = 12 x2 + C.

dy
y

= xdx. Integrating both sides, we

Then y(0) = 1 gives ln |1| = 12 (0)2 + C, i.e. C = 0 and ln |y| = 12 x2 .

1 2

So |y| = e 2 x .
1 2

Therefore, the solutions are y = e 2 x .

Here Q(y) = y is 0 when y = 0 and the values of y in the solutions satisfy y > 0 or y < 0.
Example 2. Solve the dierential equation 2xy = 1 y2 .
Solution. Separating x and y turns the equation in dierential form

2
dy
1y2

dx
x .

2
A
B
Resolving the partial fraction 1y
2 = 1y + 1+y gives linear equations A + B = 2 and A B = 0.


(
)
1
1
 1+y  =
So 1y
+ 1+y
dy = dx
.
Integrating
both
sides,
we
have

ln
|1

y|
+
ln
|1
+
y|
=
ln
|x|
+
C
or
ln
x
1y


ln(eC |x|) = ln D|x| with D = eC > 0. Then  1+y  = D|x|, i.e. 1+y = Dx where D > 0.
1y

1y

Therefore, the solution has form y = Dx+1

Dx1 where D > 0.
Exercise
1. Solve the dierential equation
2. Solve the dierential equation
3. Solve the dierential equation

dy
dx
dy
dx
dy
dx

= e1y which satises the condition y(e) = 0.

= x(y2 + 1).
= x 2.
1y

Exponential and Logistic Growth

In some model, the population grows at a rate proportional to the current population without restrictions.
The population is given by the dierential equation dP
dt = kP, where k > 0 are growth. In a rened model,
www.ck12.org

330

(
)
the rate of growth is adjusted by another factor 1 KP where K is the carrier capacity. This is close to 1
when P is small compared with K but close to 0 when P is close to K.
Both dierential equations are separable and could be solved as in last section. The solutions are respectively:
P(t) = P(0)ekt and P(t) =

P0
1+Aekt

with A =

KP0
P0 .

Example 1 (Exponential Growth) The population of a group of immigrant increased from 10000 to 20000
from the end of rst year to the end of second year they came to an island. Assuming an exponential
growth model on the population, estimate the size of the group of initial immigrants.
Solution. The population of the group is given by P = P0 ekt where the initial population and relative
growth rate are to be determined.
At t = 1 (year), P = 10000, so 10000 = P0 ek1 = Pek
0 .
At t = 2 (year), P = 20000, so 20000 = P0 ek2 = P0 e2k .
Dividing both sides of the second equation by the rst, we have 2 = ek .
Then back in the rst equation, 10000 = P0 (2). So P0 = 5000. There are 5000 initial immigrants.
(
)
P0
Example 2 (Logistic Growth) The population on an island is given by the equation dP
dt = 0.05P 1 5000 , P0 =
1000. Find the population sizes P(20), P(30). At what time will the population rst exceed 4000?
Solution. The solution is given by P =

P0
1Ae0.05t

where A =

50001000
1000

= 4.

5000
5000
= 2023
=
1 + 4e1
1 + 4e0.05(20)
5000
5000
= 3785.
P(30) =
=
0.05(30)
1 + 4e1.5
1 + 4e
P(20) =

Solve for time, 4000 =

4000 in the 56th year.

5000
1+4e0.05(t)

gives e(0.05t) =

5000
4000 1

= 0.0625. So t = 56. The population rst exceed

Exercise
1. (Exponential Growth) The population of a suburban city increased from 10000 in 2005 to 30000 in
2007. Assuming an exponential growth model on the population, by which year will the population
rst exceeds 100000?
(
)
P0
2. (Logistic Growth) The population of a city is given by the equation dP
=
0.06P
1

dt
100000 , P0 =
25000. Find the population sizes P(10), P(25). At what time will the population rst exceed 90000?

Multimedia Links
For a video presentation of Dierential Equations including growth and decay (27.0), see Dierential
Equations, Growth and Decay (7:23) .

Numerical Methods (Eulers, Improved Euler, Runge-Kutta)

The Eulers method is a numerical approximation to a solution curve starting from the point (a, b) through
the algorithm:
yn+1 = yn + hF(xn , yn ) where x0 = a, y0 = b and h is the step size.
The shorter step size, the better is the approximation to the solution curve.

331

www.ck12.org

Figure 7.21: 6.2a Dierential Equations Growth and Decay - Calculus (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/579
Improved Euler (Heun) method adapts on Eulers method by using both end point values: yn+1 = yn +
h
2 [F(xn , yn ) + F(xn+1 , yn+1 )].
Since yn+1 also appears on the right side, we replace it by Eulers formula,
h
yn+1 = yn + [F(xn , yn ) + F(xn+1 , yn + hF(xn , yn ))].
2
The Runge-Kutta methods are an important family of implicit and explicit iterative methods for the
approximation of solutions of our ODE. On them, apply Simpsons rule:

yn+1 yn =

xn+1

xn +h

f (x)dx =
xn

f (x)dx

xn

{
(
)
}
h
h

y (xn ) + 4y xn +
+ y (xn + 1) .
6
2
Exercise 1. Apply the Eulers, improved Eulers and the Runge-Kutta methods on the ODE
dy
dx

= y to approximate the solution that satisfy y(0) = 1 from x = 0 to x = 1 with h = 0.2.

We know the exact solution is y = e x . Compare their relative accuracy against the exact solution.

Texas Instruments Resources

In the CK-12 Texas Instruments Calculus FlexBook, there are graphing calculator activities
designed to supplement the objectives for some of the lessons in this chapter. See http:
//www.ck12.org/flexr/chapter/9732.

www.ck12.org

332

Chapter 8
Innite Series
This chapter introduces the study of sequences and innite series. In calculus, we are interested in the
behavior of sequences and series, including nding whether a sequence approaches a number or whether
an innite series adds up to a number. The tests and properties in this chapter will help you describe the
behavior of a sequence or series.

8.1 Sequences
Learning Objectives
Demonstrate an understanding of sequences and their terms
Determine if the limit of a sequence exists and, if it exists, nd the limit
Apply rules, theorems, and Picards method to compute the limits of sequences

Sequences (rules, terms, indices)

The alphabet, the names in a phone book, the numbered instructions of a model airplane kit, and the
schedule in the local television guide are examples of sequences people may use. These examples are all
sets of ordered items. In mathematics, a sequence is a list of numbers. You can make nite sequences,
such as 2, 4, 6, 8. These sequences end. You can also make innite sequences, such as 3, 5, 7, 9, . . . , which
do not end but continue on as indicated by the three dots. In this chapter the word sequence refers to an
innite sequence.
Each term in a sequence is dened by its place of order in the list. Consider the sequence 3, 5, 7, 9, . . . The
rst term is 3 because it belongs to place 1 of the sequence. The second term is 5 because it belongs to
the second place of the sequence. Likewise, The third term is 7 because it is in the third place. Notice
that there is a natural relationship between the counting numbers, or the positive integers, and the terms
of the sequence. This leads us to the denition of a sequence.
Sequence
A sequence is a function from the domain of the set of counting numbers, or positive integers, to the
range which consists of the members of a sequence.
A sequence can be denoted by {an } or by a1 , a2 , a3 , a4 , . . . , an , . . .
The numbers a1 , a2 , a3 , a4 , . . . , an , . . . that belong to a sequence are called terms of the sequence. Each

333

www.ck12.org

subscript of 1, 2, 3, . . . on the terms a1 , a2 , a3 , a4 , . . . refers to the place of the terms in the sequence, or the
index. The subscripts are called the indices of the terms. We assume that n = 1, 2, 3, . . . , unless otherwise
noted.
Instead of listing the elements of a sequence, we can dene a sequence by a rule, or formula, in terms of
the indices.
Example 1
The formula an =

1
n

is a rule for a sequence.

We can generate the terms for this rule as follows:

1
1
=1
1

n
an =

1
n

2
1
2

3
1
3

...

4
1
3

...

Example 2
Consider the sequence rule an =

n2
n+1 .

The terms of the sequence are:

1

n
an =

n2

12

n+1

1+1

22

1
2

2+1

4
3

...

9
4

16
5

...

You can also nd rule for a sequence.

Example 3
Find the rule for the sequence below.
n
an =?

1
1
2

3
3
4

2
3

...

4
5

...

Look at each term in terms of its index. The numerator of each term matches the index. The denominator
n
is one more than the index. So far, we can write the formula an as n+1
. However, we are not done. Notice
that each even-indexed term has a negative sign. This means that all of terms of the sequence have a power
of 1. The powers of 1 alternate between odd and even. Usually, alternating powers of 1 can be denote
by (1)n or (1)n+1 . Since the terms are negative for even indices, we use (1)n+1 . Thus, the rule for the
(1)n n
(1)n n
sequence is an = n+1 . You can check the rule by nding the rst few terms of the sequence an = n+1 .

Limit of a Sequence
We are interested in the behavior of the sequence as the value of n gets very large. Many times a sequence
will get closer to a certain number, or limit, as n gets large. Finding the limit of a sequence is very similar
to nding the limit of a function. Lets look at some graphs of sequences.
Example 4
Find the limit of the sequence

1
2n+1

as n goes to innity.

Solution
1
We can graph the corresponding function y = 2n+1
for n = 1, 2, 3, . . . . The graph of is similar to the
1
continuous function y = 2x+1 for the domain of x 1.

www.ck12.org

334

To determine the limit, we look at the trend or behavior of the graph of sequence as n gets larger or travels
out to positive innity. This means we look at the points of sequence that correspond to the far right end
of the horizontal axis in the gure on the right. We see that the
of the sequence are getting closer
}
{ points
1
is 0 as n tends to innity. We write:
to the horizontal axis, y = 0. Thus, the limit of the sequence 2n+1
1
limn+ 2n+1
= 0.
Here is the precise denition of the limit of a sequence.
Limit of a Sequence
The limit of a sequence an is the number L if for each > 0, there exists an integer N such that
| an L | < for all n > N.
Recall that | an L | < means the values of an such that L < an < L + .
What does the denition of the limit of a sequence mean? Here is another example.
Example 5
Look at Figure 3.

335

www.ck12.org

{
Figure 3 shows the graph of the sequence

ln (n)
n

}
. Notice that from N on, the terms of

ln n
n

are between

L and L + . In other words, for this value of , there is a value N such that all terms of an are in the
ln (n)
interval from L and L + . Thus, limn+ n = 0.
Not every sequence has a limit.
Example 6
Here is a graph of the sequence {n + 1}.

Consider the sequence {n + 1} in Figure 4. As n gets larger and goes to innity, the terms of an = n + 1
become larger and larger. The sequence {n + 1} does not have a limit. We write limn+ (n + 1) = +.

Convergence and Divergence

We say that a sequence {an } converges to a limit L if sequence has a nite limit L. The sequence has
convergence. We describe the sequence as convergent. Likewise, a sequence {an } diverges to a limit L
if sequence does not have a nite limit. The sequence has divergence and we describe the sequence as
divergent.
Example 7
www.ck12.org

336

{
}
The sequence ln (n) grows without bound as n approaches innity. Note that the related function y =
ln(x) grows without bound. The sequence is divergent because it does not have a nite limit. We write
limn+ ln (n) = +.
Example 8

}
{
The sequence 4 8n converges to the limit L = 4 and hence is convergent. If you graph the function y = 4 8n
for n = 1, 2, 3, . . . , you will see that the graph approaches 4 as n gets larger. Algebraically,( as n) goes to
innity, the term 8n gets smaller and tends to 0 while 4 stays constant. We write limn+ 4 8n = 4.
Example 9
Does the sequence sn with terms 1, 1, 1, 1, 1, 1, . . . . have a limit?
Solution
This sequence oscillates, or goes back and forth, between the values 1 and 1. The sequence does not get
closer to 1 or 1 as n gets larger. We say that the sequence does not have a limit, or limn+ sn does not
exist.
Note: Each sequences limit falls under only one of the four possible cases:
1.
2.
3.
4.

A limit exists and the limit is L: limn+ sn = L.

There is no limit: limn+ sn does not exist.
The limit grows without bound in the positive direction and is divergent: limn+ sn = +.
The limit grows without bound in the negative direction and is divergent: limn+ sn = .

If a sequence has a nite limit, then it only has one value for that limit.
Theorem
If a sequence is convergent, then its limit is unique.
Keep in mind that being divergent is not the same as not having a limit.

LHpitals Rule
Realistically, we cannot graph every sequence to determine if it has a nite limit and the value of that
limit. Nor can we make an algebraic argument for the limit for every possible sequence. Just as there are
indeterminate forms when nding limits of functions, there are indeterminate forms of sequences, such as
0
0 , , 0 + . To nd the limit of such sequences, we can apply LHpitals rule.
Example 10
Find limn+

ln (n)
n .

Solution
We solved this limit by using a graph in Example 5. Lets solve this problem using LHpitals rule. The
numerator is ln (n)
{ and} the denominator is n. Both functions y = ln (n) and y = n do not have limits.
So, the sequence

ln (n)
n

is of the indeterminate form

Since the functions y = ln (n) and y = n are not

ln (x)
x , rst. Taking the
1
ln (x)
ln (x)
and denominator of y = x , we nd lim x+ x = lim x+ 1x = 0.
ln (n)
ln (x)
the points of y = n are a subset of the points of the function y = x

dierentiable, we apply LHpitals rule to the corresponding problem, lim x+

rst derivative of the numerator
ln (n)

Thus, limn+ n = 0 because

as x approaches innity. We also conrmed the limit of the sequence with its graph in Example 5.

337

www.ck12.org

Rules, Sandwich/Squeeze
Properties of function limits are also used with limits of sequences.
Theorem (Rules)
Let {an } and {bn } be sequences such that limn+ an = L1 and limn+ bn = L2 .
Let c be any constant. Then the following statements are true:
1. limn+ c = c
The limit of a constant is the same constant.
2. limn+ c an = c limn+ an = cL1
The limit of a constant times a sequence is the same as the constant times the limit of the sequence.
3. limn+ (an + bn ) = limn+ an + limn+ = L1 + L2
The limit of a sum of sequences is the same as the sum of the limits of the sequences.
4. limn+ (an bn ) = limn+ an limn+ bn = L1 L2
The limit of the product of sequences is the same as the product of the limits of the sequences.
( ) lim
an
5. If L2 , 0, then limn+ bann = limn+
= LL12 .
n+ bn
The limit of the quotient of two sequences is the same as the quotient of the limits of the sequences.
Lets apply these rules to help us nd limits.
Example 11
Find limn+

7n
9n+5 .

Solution
We could use LHpitals rule or we could use some of the rules in the preceding theorem. Lets use the rules
in the theorem. Divide both the numerator and denominator by the highest power of n in the expression
and using rules from the theorem, we nd the limit:

7n
= lim
n+
n+ 9n + 5
lim

7n
n
9n
n

+
7

5
n

Dividing both numerator and denominator by n

) Simplifying
= lim (
n+ 9 + 5
n
=

limn+ 7
(
) Applying the division rule for limits.
limn+ 9 + 5n

limn+ 7
Applying the rule for the limit of a sum to the denominator
limn+ 9 + limn+ 5n
7
7
=
= Evaluating the limits
9+0
9
=

Example 12
Find limn+

11
n

8
n2

)
.

Solution
www.ck12.org

338

lim

( 11

n+

8)
11
8
= lim
lim 2 Applying the rule for the dierence of two limits
2
n+
n+
n
n
n
1
1
= 11 lim
8 lim 2 Applying the rule for the limit of c times a limit
n+ n
n+ n
= 11 0 8 0 = 0 Evaluating the limits

As with limits of functions, there is a Sandwich/Squeeze Theorem for the limits of sequences.
Sandwich/Squeeze Theorem
Let {an } , {bn } and {cn } be sequences. Let N be a positive integer.
Suppose cn is a sequence such that an cn bn for all n N. Suppose also that
lim an = lim bn = L. Then lim cn = L.
You can see how the name of the theorem makes sense from the statement. After a certain point in the
sequences, the terms of a sequence cn are sandwiched or squeezed between the terms of two convergent
sequences with the same limit. Then the limit of the sequence cn is squeezed to become the same as the
limit of the two convergent sequences. Lets look at an example.
Example 13
Prove limn+

8n
n!

= 0.

Solution
Recall that n! is read as n factorial and is written as n! = n (n 1) (n 2) . . . 1.
We want to apply the Sandwich theorem by squeezing the sequence
to the same limit.

8n
n!

between two sequences that converge

First, we know that 0 8n! . Now we want to nd a sequence whose terms greater than or equal to the
n
terms of the sequence 8n! for some n.
We can write
8 8 8... 8
8n
=
n!
n (n 1) (n 2) . . . 1
8
8
8 8
=
...
n n1
2 1
(8) ( 8
8 8) (8 8 8
8)
...
...
=
n n1
9 8 7 6 5
1
8
8
Since each factor in the product n1
. . . 89 88 is less than or equal to 1, then the product n1
. . . 89 88 1.
Then we make an inequality:
(8 8 8
(8) ( 8
8 8) (8 8 8
8) (8)
8)
...
...

(1) . . .
n n1
9 8 7 6 5
1
7
7 6 5
1
(8) (8 8 8
8)
=
...
7 (7 ) 6 5
1
( 8 ) 87
=
n 7!

( )( 7)
n
Thus, limn+ 0 limn+ 8n! limn+ 8n 87! . By using the Rules Theorem, we have limn+ 0 = 0
( )( 7) ( 7)
( 7)
n
and limn+ 8n 87! = 87! limn+ 8n = 87! 0 = 0. Thus, 0 limn+ 8n! 0. By the Sandwich/Squeeze
n
Theorem, limn+ 8n! = 0.

339

www.ck12.org

Picards Method
The following method appeared in 1891 by Emile Picard, a famous French mathematician. It is a method
for solving initial value problems in dierential equations that produces a sequence of functions which
converge to the solution. Start with the initial value problem:
y = f (x, y) with y(x0 ) = y0
If f (x, y) and f x (x, y) are both continuous then a unique solution to the initial value problem exists by
Picards theory. Now if y(x) is the solution to the given problem, then a reformulation of the dierential
equation is possible:
x
x

y (t)dt =
f (t, y(t))dt
x0

x0

Now the Fundamental Theorem of Calculus is utilized to integrate the left hand side of the problem and
upon isolating , the following result is obtained:
x
f (t, y(t))dt
y(x) = y0 +
x0

The equation above is the starting point for the Picard iteration because it will be used to build the
sequence of functions which will describe the actual solution to the initial value problem. The Picard
sequence of functions is calculated as follows:
Step 1 - Dene Y0 (x) = y0
Step 2 - Substitute Y0 (t) = y0 for y(t) in f (t, y(t)):

x0
x

Y1 (x) = y0 +

f (t, Y0 (t))dt

Y1 (x) = y0 +

f (t, Y0 )dt
x0

Step 3 - Repeat step 2 with Y1 (t) for y(t) :

Y2 (x) = y0 +

f (t, Y1 (t))dt
x0

{
}
The substitution process is repeated n times and generates a sequence of functions Yn (x) which converges
to the initial value problem. To summarize this procedure mathematically,
Picards Method
Let {Yn (x)} be sequence dened successively by,
x
Yn (x) = y0 + x f (t, Yn1 (t))dt for n 0
0

The sequence of approximations converges to the solution y(x), i.e.

lim Yn (x) = y(x).

Now that we have dened Picards method, lets calculate a sequence of functions for an initial value
problem.
Example 1
www.ck12.org

340

{
}
Find the rst four functions Yn (x) 3n=0 dened by Picards method for the solution to the initial value
problem
y (x) = xy(x) with y(1) = 1.
Solution
We want to apply the Fundamental Theorem of Calculus to the dierential equations so that it is reformulated for use in the Picard method. Thus,
x
x

y (t)dt =
ty(t)dt
1
1
x
y(x) y(1) =
ty(t)dt
1
x
y(x) = 1 +
ty(t)dt
1

Now that the dierential equation has been rewritten for Picards method, we begin the calculations for
the sequence of functions. In all cases the rst function Y0 (x) is given by the initial condition:
Step 1 - Dene Y0 (x) = 1
Step 2 - Substitute Y0 (x) = 1 for y(t) in the integrand of y(x) = 1 +
x
Y1 (x) = 1 +
tdt
1
 x
t2 
Y1 (x) = 1 + 
2
Y1 (x) =
Step 3 - Substitute Y1 (x) =

1
2

x2
2

x
ty(t)dt:
1

1
2
x

1
+
2
2

for y(t) in the integrand as above:

)
x (
1 t2
+
dt
Y2 (x) = 1 +
t
2
2
1

( 2
) x
t
t4 

Y2 (x) = 1 +
+
4
8 
1

x4
5 x2
Y2 (x) = +
+
8
4
8
Step 4 - Substitute Y2 (x) =

5
8

x2
4

x4
8

for y(t) in the integrand as done previously:

)
x (
5 t2 t4
Y3 (x) = 1 +
t
+ +
dt
8
4
8
1

( 2
) x
5t
t4
t6 

Y3 (x) = 1 +
+
+
16
16 48 

29 5x2
x4
x6
Y3 (x) =
+
+
+
48
16
16 48
Thus, the initial four functions in the sequence dened by Picards method are:
{
}
1 x2
5 x2
x4
29 5x2
x4
x6
1, + ,
+
+ ,
+
+
+
2
2
8
4
8
48
16
16 48

341

www.ck12.org

The method also states that this sequence will converge to the solution y(x) of the initial value problem,
i.e.
lim Yn (x) = y(x)

n+

A pattern of the functions in the sequence Yn (x) is emerging but it is not an obvious one. We do know Yn (x)
will converge to the solution for this problem by Picards method. The exact solution for this problem can
be calculated and is given by:
y(x) = e

x2 1
2

Clearly this solution satises y(x) = xy(x) and y(1) = 1.

Review Questions
1. Find the rule for the sequence an .
n
1
an =?

...

...

Tell if each sequence is convergent, is divergent, or has no limit. If the sequence is convergent, nd its
limit.
2.

4
{n

3
n2 }
7
n

3. 6
4. {5, 5, 5,
} 5, 5, 5, . . .
6
5. 4n3n7
}
{
(1)n
6. 5n2
{
}
7. {(1)n n
}
4 2
8. (1)n 2n43n
2
+6n 4n
{ 2}
9. 6n
n
e
10. Let {an } be a sequence such that limn+ |an | = 0. Show that limn+ an = 0. (|an | is the absolute
value of an .)
{
}
11. Find the rst four functions Yn (x) 3n=0 dened by Picards method for the solution to the initial value
problem y (x) = 1 + y with y(0) = 0.
{
}
12. Find the rst four functions Yn (x) 3n=0 dened by Picards method for the solution to the initial value
problem y (x) = 1 + y2 with y(0) = 0.
{
}
13. Find the rst three functions Yn (x) 2n=0 dened by Picards method for the solution to the initial
value problem y (x) = y1/3 with y(0) = 18 .

Review Answers
1.
2.
3.
4.
5.

an = (1)n 2
convergent; Limit is 0
convergent; Limit is 6
No limit exists.
divergent

www.ck12.org

342

6.
7.
8.
9.
10.

convergent; Limit is 0
No limit exists.
No limit exists.
convergent; Limit is 0
By denition of absolute value, |an | an |an | . Then take limits of all three terms:
lim (1|an |) lim an lim |an |
n+

n+

n+

lim (|an |) lim an lim |an |

n+

n+

n+

0 lim an 0
n+

By the Sandwich/Squeeze Theorem, limn+ an = 0 also.

{
{
}
2
2
3}
11. Yn (x) 3n=0 = 0, x, x + x2 , x + x2 + x6
{
}
{
}
3
3
5
x7
12. Yn (x) 3n=0 = 0, x, x + x3 , x + x3 + 2x
+ 63
15
{
(
)4/3 }
{
}
1
+ 32 81 + 2x
13. Yn (x) 2n=0 = 18 , 18 + 2x , 32

Keywords
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.

sequence
rules
terms
index, indices
limit
convergence
divergence
LHpitals Rule
Sandwich/Squeeze Theorem
Picards Method

8.2 Innite Series

Learning Objectives

Demonstrate an understanding of series and the sequence of partial sums

Recognize geometric series and determine when they converge or diverge
Compute the sum of a convergent geometric series
Determine convergence or divergence of series using the nth-Term Test

Innite Series (series, sequence of partial sums, convergence, divergence)

Series
Another topic that involves an innite number of terms is the topic of innite series. We can represent
certain functions and numbers with an innite series. For example, any real number that can be written
as a non-terminating decimal can be represented as an innite series.

343

www.ck12.org

Example 1
The rational number
series:

4
9

can be written as 0.44444. We can expand the decimal notation as an innite

4
= 0.4 + 0.04 + 0.004 + 0.0004 + . . .
9
4
4
4
4
=
+
+
+
+ ...
10 100 1000 10, 000
4
4
4
4
=
+ 2 + 3 + 4 + ...
10 10
10
10

On the other hand, the number

nite series:

1
4

can be written as 0.25. If we expand the decimal notation, we get a

1
= 0.2 + 0.05
4
5
2
+
=
10 100
2
5
=
+ 2
10 10

Do you see the dierence between an innite series and a nite series? Lets dene what we mean by an
innite series.
Innite Series
An innite series is the sum of an innite number of terms, u1 , u2 , u3 , u4 , . . . , usually written as.
u1 + u2 + u3 + u4 + . . . .
A shorthand notation for an innite series is to use sigma notation:

k=1 uk , which can be read as the sum of the terms uk s for k equal to 1 to innity.
We can make nite sums from the terms of the innite series:
s1 = u1
s2 = u1 + u2
s3 = u1 + u2 + u3
The rst sum is the rst term of the sequence. The second sum is the sum of the rst two terms. The
third term is the sum of the rst three terms. Thus, the nth nite sum, sn is the sum of the rst n terms
of the innite series: sn = u1 + u2 + u3 + . . . + un .

Sequence of Partial Sums

As you can see, the sums sn = u1 + u2 + u3 + . . . + un form a sequence. The sequence is very important for
the study of the related innite series for it tells a lot about the innite series.
Partial Sums
For an innite series

sn = nk=1 uk .

k=1 uk

, the nth partial sum, sn is the sum of the rst n terms of the innite series:

The sequence {sn } formed from these sums is called the sequence of partial sums.
www.ck12.org

344

Example 2
Find the rst ve partial sums of the innite series 1 + 0.1 + 0.01 + 0.001 + .
Solution
s1 = u1 = 1
s2 = u1 + u2 = 1 + 0.1 = 1.1
s3 = 1 + 0.1 + 0.01 = 1.11
s4 = 1 + 0.1 + 0.01 + 0.001 + 0.0001 = 1.1111
s5 = 1 + 0.1 + 0.01 + 0.001 + 0.0001 = 1.1111
To further explore series, try experimenting with this applet. The applet shows the terms of a series as
well as selected partial sums of the series. Series Applet. As you see from this applet, for some series the
partial sums appear to approach a xed number, while for other series the partial sums do not. Exploring
this phenomenon is the topic of the next sections.

Convergence and Divergence

Just as with sequences, we can talk about convergence and divergence of innite series. It turns out that
the convergence or divergence of an innite series depends on the convergence or divergence of the sequence
of partial sums.
Convergence/Divergence of Series

sequence of partial sums for the series. If {sn } has a

Let
k=1 uk be an innite series and let {sn } be the
nite limit l , then the innite series converges and
k=1 uk = l.
If {sn } does not have a nite limit, then the innite series diverges. The innite series does not have a sum.
Example 3
Does the innite series 1 + 0.1 + 0.01 + 0.001 + . . . converge or diverge?
Solution
To make our work easier, write the innite series 1 + 0.1 + 0.01 + 0.001 + . . . as an innite series of fractions:
1+

1
1
1
+
+
+ ...
10 102 103

To solve for convergence or divergence of the innite series, write the formula for the nth partial sum

1
sn = nk=1 uk : sn = 1 + 10
+ 1012 + 1013 + . . . + 101n1 . Note that the nth partial sum ends with a power of
n 1 in the denominator because 1 is the rst term of the innite series.
1
It is rather diicult to nd limn+ sn = limn+ 1 + 10
+ 1012 + 1013 + . . . + 101n1 as it is written. We will
work the sum into a dierent form so that we can nd the limit of the sequence of partial sums.

First, multiply both sides of the equation sn = 1 +

1
10

1
102

1
103

+ ... +

1
10n1

by

1
10 :

1
1 (
1
1
1
1 )
sn =
1+
+ 2 + 3 + . . . + n1
10
10
10 10
10
10
1
1
1
1
1
1
sn =
+
+
+
+ ... + n
10
10 102 103 104
10
Now we have two equations:

345

www.ck12.org

1
1
1
1
+
+
+ . . . + n1
10 102 103
10
1
1
1
1
1
1
sn =
+ 2 + 3 + 4 + ... + n
10
10 10
10
10
10
sn = 1 +

Subtract the bottom equation from the top equation to cancel terms and simplifying:
1
1
1
1
1
+ 2 + 3 + 4 + . . . + n1
10 10
10
10
10
(1
1
1
1
1
1 )
sn =
+
+
+
+ ... + n
10
10 102 103 104
10

sn = 1 +

1
1
sn = 1 n
10
10
9
1
sn = 1 n
10
10

sn

Solve for sn by multiplying both sides of the last equation by

sn =

10
9

10
1
(1 n )
9
10

Now we nd the limit of both sides:

10 (
1 )
lim sn = lim
1 n
n 9
n+
10
)
(
10
10
1
10 ( 1 )
lim
1 n = lim
lim
n 9
n+ 9
n+ 9 10n
10
10
10
=
0=
9
9
The sum of the innite series is

10
9

and so the series converges.

Geometric Series
The geometric series is a special kind of innite series whose convergence or divergence is based on a
certain number associated with the series.
Geometric Series
A geometric series is an innite series written as
a + ar + ar2 + ar3 + . . . + ar{i1} + . . . .
In sigma notation, a geometric series is written as

k=1 ar

The number r is the ratio of the series.

Example 4
Here are some examples of geometric series.

www.ck12.org

346

k1 .

Table 8.1:
Geometric Series
1
+ ...
4k1
k
(1) 5
5
5
5
6 + 62 + 63 + . . . + 6k + . . .
1 + 3 + 32 + 33 + . . . + 3k1 + . . .

1+

1
4

1
42

1
43

+ ... +

1
4

5
1

16
3

The convergence or divergence of a geometric series depends on r.

Theorem
Suppose that the geometric series

k=1 ar

k1

has ratio r.

1. The geometric series converges if |r| < 1 and the sum of the series is.
2. The geometric series diverges if |r| 1.
Example 5
7
Determine if the series 7 + 78 + 872 + 873 + . . . + 8i1
+ . . . converges or diverges. If it converges, nd the sum
of the series.

Solution

( )k1

The series is a geometric series that can be written as

7 18
. Then a = 7 and the ratio r =
k=1
 
1
a
Because  8  < 1 , the series converges. The sum of the series is 1r = 17 1 = 77 = 8.
8

Example 6
Determine if the series

k=1 9

k1

1
8.

converges or diverges. If it converges, nd the sum of the series.

Solution
The series is a geometric series with a = 1 and the ratio r = 9. Because |9| > 1, the series diverges.
Example 7
Determine if 34 +
series.

3
42

3
43

+ ... +

3(1)k
4k

+ . . . converges or diverges. If it converges, nd the sum of the

Solution
If we rewrite the series in terms of powers of k, the series looks like this:
( 1 )1
( 1 )2
( 1 )k
3(1)1 3(1)2 3(1)3
3(1)k
+
+
+
.
.
.
+
+
.
.
.
=
3

+
3

+
.
.
.
+
3

+ ....
4
4
4
41
42
43
4k
 
It looks like a geometric series with a = 3 and r = 41 .Since  14  = 14 < 1, the series converges.
However, if we write the denition of a geometric series for a = 3 and r = 14 , the series looks like this:
)
+ (
( 1 )0
( 1 )1
( 1 )2

1 k1
3
=3
+3
+3
+ ...
4
4
4
4
k=1
( 1 )2
( 1 )1
+3
+ ...
=3+3
4
4
3(1)1

3(1)2

3(1)3

3(1)k

The original problem, 41 + 42 + 43 + . . . + 4k + . . ., does not have the leading term of 3. This
does not aect the convergence but will aect the sum of the series. We need to subtract 3 from the sum
( )1
( )2
3(1)1
3(1)2
3(1)3
3(1)k
of the series 3 + 3 14 + 3 14 + . . . to get the sum of 41 + 42 + 43 + . . . + 4k + . . ..

347

www.ck12.org

The sum of the series is:

a
1r

3=

3
1( 14 )

3=

3
5
4

3=

12
5

3=

12
5

15
5

= 35 .

Other Convergent Series

There are other innite series that will converge.
Example 8
Determine if

+ ( 2
n=1 4

2
k+1

converges or diverges. If it converges, nd the sum.

Solution
The nth partial sum sn is:
2 )
k k+1
k=1
(2
(2
2 ) (2
2 ) (2
2 )
2 )
=

+ ... +

1 1+1
2 2+1
3 3+1
n n+1
(2 2) (2 2) (2 2)
(2
2 )
=

+
+
+ ... +
1 2
2 3
3 4
n n+1

Sn =

n (

We can simplify sn further. Notice that the rst parentheses has 22 while the second parentheses has
These will add up to 0 and cancel out. Likewise, the 23 and 32

2
2

will cancel out. Continue in this way to cancel opposite terms. This sum is a telescoping sum, which is
a sum of terms that cancel each other out so that the sum will fold neatly like a folding telescope. Thus,
we can write the partial sum as
sn =
(
Then limn+ sn = limn+ 2

2
n+1

2
2
2

=2
.
1 n+1
n+1

= 2 and

+ ( 2
k=1 k

2
k+1

= 2.

Other Divergent Series (nth-Term Test)

Determining convergence by using the limit of the sequence of partial sums is not always feasible or
practical. For other series, it is more useful to apply tests to determine if an innite series converges or
diverges. Here are two theorems that help us determine convergence or divergence.
Theorem (The nth-Term Test)

If the innite series

k=1 uk converges, then limk+ uk=0
Theorem

If k+ uk , 0 or if k+ uk does not exist, then the innite series

k=1 uk diverges.
The rst theorem tells us that if an innite series converges, then the limit of the sequence of terms is 0.
The converse is not true: If the limit of the sequence of terms is 0, then the series converges. So, we cannot
use this theorem as a test of convergence.
The second theorem tells us that if limit of the sequence of terms is not zero, then the innites series
diverges. This gives us the rst test of divergence: the nth-Term Test or Divergence Test. Note that
if the test is applied and the limit of the sequence of terms is 0, we cannot conclude anything and must
use another test.
Example 9
Determine if

www.ck12.org

k
k=1 k+5

converges or diverges.

348

Solution
We can use the nth-Term Test to determine if the series diverges. Then we do not have to check for
convergence.
lim

k+

Because limk+
Example 10
Determine if

k
k+5

8
k=1 k3

k
= lim
k + 5 k+

, 0, the series

k
k
k+5
k

k
k=1 k+5

= lim

k+

1
1+

5
k

limk+ 1
limk+ 1 +

5
k

=1

diverges.

converges or diverges.

Solution
8
k3

Using the nth-Term Test, limk+

or divergence.

. Since the limit is 0, we cannot make a conclusion about convergence

Rules for Convergent Series, Reindexing

Rules
As with sequences, there are rules for convergent innite series that help make it easier to determine
convergence.
Theorem (Rules for Convergent Series)

1. Suppose
n=1 vk are convergent series with
k=1 uk = S 1 and
k=1 vk = S 2 .
k=1 uk and

Then k=1 (uk + vk ) and k=1 (uk vk ) are also convergent where

k=1 (uk + vk ) = k=1 (uk ) + k=1 (vk ) = S 1 + S 2 and

k=1 (uk vk ) = k=1 (uk ) k=1 (vk ) = S 1 S 2
(The sum or dierence of convergent series is also convergent.)
2. Let c , 0 be a constant.

Suppose
k=1 uk converges and
k=1 uk = S Then
k=1 cuk also converges where.

cui = c

k=1

If

k=1 uk

diverges, then

k=1 cuk

uk = cS

k=1

also diverges.

(Multiplying by a nonzero constant does not aect convergence or divergence.)

Example 10
Find the sum of

(
k=1

2
3k1

1
8k1

Solution

(
)

2
1
2
1
Using the Rules Theorem,
+
=
k1
k1
k=1 3
k=1 3k1 + k=1 8k1 .
8

2
1
k=1 3k1 is a convergent geometric series with a = 2 and r = 3 . Its sum is

1
1
k=1 8k1 is a convergent geometric series with a = 2 and r = 8 . Its sum is
)
(

1
29
1
2
8
2
Then
+
=
k=1 3k1
k=1 3k1 + i=1 8k1 = 3 + 7 = 7
8k1

2
1 13

2
1 18

2
3
7
8

=3.
=

8
7

Example 11

349

www.ck12.org

Find the sum of

k=1 2

6k1

Solution

)
(

5
5
By the Rules Theorem,
k=1 2 6k1 =
k=1 6k1 . The series 2 k=1

5
5
5
and r = 16 . The sum of is
k=1 6k1 1 1 = 5 = 6.
6
6
(
)

5
Then k=1 2 6k1 = 2 6 = 12.

5
6k1

is a geometric series with a = 5

Adding or subtracting a nite number of terms from an innite series does not aect convergence or
divergence.
Theorem

If
k=1 uk converges, then
k=1 uk + (u1 + u2 + . . . + um ) is also convergent.

If k=1 uk converges, then k=1 uk (u1 + u2 + . . . + um ) is also convergent.

Likewise, if
k=1 uk diverges, then
k=1 uk + (u1 + u2 + . . . + um ) and
k=1 uk (u1 + u2 + . . . + um ) are also
divergent.
For a convergent series, adding or removing a nite number of terms will not aect convergence, but it will
aect the sum.
Example 12
Find the sum of

3
k=1 5k1

(
)
3 + 35 .

Solution

is a geometric series with a = 3 and r = 51 . Its sum is

(
)

3
3
15
18
3
Then

3
+
k1
k=1 5
5 = 4 5 = 5
k=1 5k1

15
4

Reindexing
Another property of convergent series is that we can reindex a series without changing its convergence.
This means we can start the indices of the series with another number other than 1. Keep the terms in
order though for reindexing.
Example 13

4
k=1 3k1 is a convergent geometric series. It can be reindexed by changing the starting position of i and
the power of i. The new series is still convergent.

4
4
=
k1
k6
3
3
k=1
k=6

You can check that the series on the right is the same series as the one of the left by writing out the rst
few terms for each series. Notice that the terms are still in order.

Review Questions
1. Express the number

1
11

as an innite series.
(1)k
2. Find s1 , s2 , s3 and for k=1 2k .
3
3. Determine if the innite series 3 + 10
+ 1032 + 1033 converges or diverges.
4. What are the values of a and r for the geometric series 3 + 3(2)1 + 3(2)2 + 3(2)3 + . . . .?
www.ck12.org

350

Determine if each innite series converges or diverges. If a series converges, nd its sum.
5.

( 3 )k
k=1

( 5 )k1

2
6. +
k=1 3

k3
7.
k=1 k3 5

4k+2
8.
k=1 9k1

(( 2 )k1

9. Find the sum of k=2 3

+ 5k1 .

10. Suppose k=1 Uk is a convergent series and

k=1 (U k + Vk )
k=1 Vk is a divergent series. Explain why

and k=1 (Uk Vk ) both diverge.

11. Give an example of a geometric series whose sum is 3.
12. Give an example of a telescoping sum whose sum is 4.

Review Answers
1
= 92 + 94 + 96 + . . . . . . + 92k + . . .
= 0.09
1. 11
10
10
10
10
2. s1 = 12
s2 = 12 + 14 = 14
s3 = 12 + 14 18 = 38
1
s4 = 12 + 14 18 + 16
= 58
3
3
3
3. sn = 3 + 10 + 102 + 103 + . . . . . . + 103n1
(
)
1
1
3
3
3
3
10 sn = 10 3 + 10 + 102 + 103 + . . . + 10n1
3
3
3
3
3
3
10 sn = 10 + 102 + 103 + 104 + . . . . . . + 10n

Then sn = 3 +

3
10

3
102

+ . . . . . . + 103n1
(3
3
3
3
3 )
3
sn =
+ 2 + 3 + 4 + ...... + n
10
10 10
10
10
10

3
103

3
1
sn = 3 n
10
10
7
1
sn = 3 n
10
10
1 )
10 (
sn =
3 n
7
10
sn

Then limn sn = limn

10
7 (3

1
10n )

lim sn =

30
10
(3)
7
7

The series converges.

4. a = 3, r = 2
( )k1

3
5. The series
is a geometric series with a = 1 and r = 35 .|r| = 53 < 1 and so, the geometric
k=1 5
series converges. The sum is 11 3 = 52 .
5
+ ( 2 )k1
is a geometric series with a = 1 and r = 23 .|r| = 23 < 1 and so, the geometric
6. The series k=1 3
(
( )
)
1
8
3
23
34
series converges. The sum is
1 49 + 27
= 15 23
2
27 = 5 27 = 27 .
1( 3 )
3

351

www.ck12.org

k3
k=1 k3 3

diverges by the nth-Term Test: limk k3k3 = limk 11 3 = 1 , 0.

k3
( )k1
4k+2
3 4k1
4
3
. This is a geometric series with a = 64 and r = 49 .
8. k=1 9k1 = 4 k=1 4 k=1 9k1 = k=1 64 9
64
64
576
The series converges. The sum is
= 5 = 5 .
1( 49 )
9
(
)
(
(
)
)
k1
k1

2
1
2
1
9. k=2 3
+ 5k1 = k=2 3
+ k=2 5k1
7.

)k1

1
(1) = 35 1 = 52 and
k=2 5k1 =
(( )k1
(( )k1
)
)

2
1
2
1
Then

+
=

+
k=2
k=2
k=2 5k1 =
3
3
5k1
k=2

32

1
1( 23 )

1
1 15

1=

2
5

1
4

5
4

1 = 41 .

3
= 20

is divergent, then
+ or
Vk = . Since
k=1 Vk =
k=1
k=1 U k is convergent,

of something nite and

k=1 U k is nite. Then k=1 (U k + Vk ) =
k=1 U k +
k=1 Vk is the sum

something innite. Thus,

(U
+
V
)
is
innite
and
diverges.
Likewise,
(Uk Vk ) =
k
k
k=1
k=1
k=1 U k

k=1 Vk is the dierence of something nite and something innite and hence is still innite and
divergent.
( )k1

5
11. Sample answer:
k=1 2
3
)
( 4
4
12. Sample answer: k=1 k k+1
10. Since

k=1 Vk

Keywords
1.
2.
3.
4.
5.
6.
7.
8.

innite series
sequence of partial sums
convergence
divergence
geometric series
ratio of geometric series
nth-Term Test
reindexing

8.3 Series Without Negative Terms

Learning Objectives
Demonstrate an understanding of nondecreasing sequences
Recognize harmonic series, geometric series, and p series and determine convergence or divergence
Apply the Comparison Test, the Integral Test, and the Limit Comparison Test

Nondecreasing Sequences
In order to extend our study on innite series, we must rst take a look at a special type of sequence.
Nondecreasing Sequence
A nondecreasing sequence {S n } is a sequence of terms that do not decrease:
S 1 S 2 S 3 . . . S n . . . . Each term is greater than or equal to the previous term.
Example 1 5, 10, 15, 20, . . . is a nondecreasing sequence. Each term is greater than the previous term:
5 < 10 < 15 < 20 < . . . .
www.ck12.org

352

10, 000, 1000, 100, . . . is not a nondecreasing sequence. Each term is less than the previous term: 10, 000, >
1000, > 100 . . . .
3, 3, 4, 4, 5, 5, . . . is a nondecreasing sequence. Each term is less than or equal to the previous term: 3 3
4 4 5 5 ....
A discussion about sequences would not be complete without talking about limits. It turns out that certain
nondecreasing sequences are convergent.
Theorem
Let {S n } be a nondecreasing sequence: S 1 S 2 S 3 . . . S n . . . .
lim S n

1. If there is a constant B such that S n B for all n, then limn S n exists and limn S n = L where
L B.
2. If the constant B does not exist, then limn S n = +.
The theorem says that a bounded, convergent, nondecreasing sequence has a limit that is less than or equal
to the bound. If we cannot nd a bound, the sequence diverges.
}
{
n
converges or diverges. If it converges, nd its limit.
Example 2 Determine if the sequence 6n+5
Solution
1
2
3
4
Write the rst few terms: 11
, 17
, 23
, 29
, . . .. The sequence is nondecreasing. To determine convergence,
n
we see if we can nd a constant B such that 6n+5
B. If we cannot nd such a constant, then the sequence
diverges.

If two fractions have the same numerator but dierent denominators, the fraction with the smaller denomn
n
n
n
inator is the larger fraction. Thus, 6n+5
6n
= 16 . Then 6n+5
61 and, in fact, limn 6n+5
= 61 .

Series Without Negative Terms (harmonic, geometric, p-series)

There are several special kinds of series with nonnegative terms, i.e., terms that are either positive or zero.
We will study the convergence of such series by studying their corresponding sequences of partial sums.
Lets start with the harmonic series:

1
1 1 1
= 1 + + + + ....
K
2 3 4
k=1

The sequence of partial sums look like this:

S1 = 1
1
2
1 1
S1 = 1 + +
2 3
1 1 1
S1 = 1 + + +
2 3 4
S1 = 1 +

..
.

353

www.ck12.org

In order for the harmonic series to converge, the sequence of partial sums must converge. The sequence
of partial sums of the harmonic series is a nondecreasing sequence. By the previous theorem, if we nd
a bound on the sequence of partial sums, we can show that the sequence of partial sums converges and,
consequently, that the harmonic series converges.
It turns out that the sequence of partial sums cannot be made less than a set constant B. We will omit
the proof here, but the main idea is to show that the a selected innite subset of terms of the sequence
of partial sums are greater than a sequence that diverges, which implies that the sequence of partial sums
diverge. Hence, the harmonic series is not convergent.
We can also work with geometric series whose terms are all non-negative.
( )k1

3
Example 3 The geometric series
has all non-negative terms. The sequence of partial sums
k=1 2
looks like this:
S1 = 1
3
2
3 9
S1 = 1 + +
2 4
3 9 27
S1 = 1 + + +
2 4
8
S1 = 1 +

..
.
Intuitively, we can see that there is no bound on the sequence of partial sums and so the series diverges.
This is conrmed by the fact that the ratio of the series, r = 23 , tells us that the geometric series does not
converge.
Another example of an important series is the pseries:
1
1
1
1
k=1 K p = 1 + 2 p + 3 p + 4 p +, where p > 0.
The pseries may look like a harmonic series, but it will converge for certain values of p.
Theorem

converges for p > 1 and diverges for 0 < p 1.

1 converges or diverges.
Example 4 Determine if
k=1
The pseries

1
k=1 K p

Solution
Rewrite

1
K

as

1
1
K2

to get

k=1

1
K

+
k=1

the series diverges.

1
1

K2

. The value of p is 12 . This is less than 1, which tells us that

Comparison Test
Now that we have studied series without negative terms, we can apply convergence tests made for such
series. The rst test we will consider is the Comparison Test. The name of the test tells us that we will
compare series to determine convergence or divergence.
Theorem (The Comparison Test)

Let
k=1 uk and
k=1 vk be series without negative terms. Suppose that u1 v1 , u2 v2 , . . . , ui vi , . . . .
1. If

k=1 vk

www.ck12.org

converges, then

k=1 uk

converges.

354

2. If

k=1 uk

diverges, then

diverges.

k=1 vk

In order to use this test, we must check that for each index k, every uk is less than or equal to vk . This

is the comparison part of the test. If the series with the greater terms,
k=1 vk , converges, than the series

with the lesser terms , k=1 uk , converges. If the lesser series diverges, then the greater series will diverge.
You can only use the test in the orders given for convergence or divergence. You cannot use this test to
say, for example, that if the greater series diverges, than the lesser series also diverges.

1
Example 5 Determine whether
k=1 K 3 +3 converges or diverges.
Solution

1
looks similar to
k=1 K 3 , so we will try to apply the Comparison Test. Begin by comparing each

1
1
1
term. For each k, K 31+3 is less than or equal to K13 , so
k=1 K 3 +3 k=1 K 3 . Since
k=1 K 3 is a convergent

1
pseries, then, by the Comparison Test, k=1 K 3 +3 also converges.

1
Example 6 Determine whether
converges or diverges.
4
k=1
k=1 K 3 +3

k5

Solution

1
1
1
The series
is similar to
41 for all k. The series
4
4 . Using the Comparison Test,
4
k=1
k=1
k5
k
k5
1
k 1
1

diverges
since
it
is
a
pseries
with
p
=
.
By
the
Comparison
Test,
also diverges.
4
4
k=1
k=1
4
k

k5

The Integral Test

Another useful test for convergence or divergence of an innite series without negative terms is the Integral
Test. It involves taking the integral of the function related to the formula in the series. It makes sense to
use this kind of test for certain series because the integral is the limit of a certain series.
Theorem (The Integral Test)

Let
k=1 uk = k=1 f (k) be a series without negative terms. If f (x) is a decreasing, continuous, non-negative
function for x 1, then:
1.
2.

k=1 uk converges if and only if

k=1 uk

diverges if and only if

f (x) converges.

f (x) diverges.

In the statement of the Integral Test, we assumed that uk is a function f of k. We then changed that
function f to be a continuous function of x in order to evaluate the integral of f . If the integral is nite,
then the innite series converges. If the integral is innite, the innite series diverges. The convergence or
divergence of the innite series depends on the convergence or divergence of the corresponding integral.

1
Example 7 Determine if
3 converges or diverges.
k=1
(2k+1) 2

Solution
We can use the Integral Test to determine convergence. Write the integral form:

1
3

dx.

(2k + 1) 2

Next, evaluate the integral.

355

www.ck12.org

dx = lim (2k + 1) 2 dx
3

(2k + 1)

3
2

Use the following usubstitution to evaluate the integral:

u = 2x + 1
du = 2dx
Then

1
2

u 2 du =
3

1 u1/2
2 (1/2)

1
1

u2

= 1u .

(
)
t
3
1
1
Thus, limt (2x + 1) 2 dx = limt 2x+1
]41 = limt 1 + 2t+1
= 1 .
3

Since the integral is nite, the series

k=1

1
3

converges by the Integral Test.

(2k+1) 2

Limit Comparison Test, Simplied Limit Comparison Test

Another test we can use to determine convergence of series without negative terms is the Limit Comparison Test. It is easier to use than the Comparison Test.
Theorem (The Limit Comparison Test)

Suppose
k=1 uk is a series without negative terms. Then one of the following will hold.

vk is a convergent series without negative terms and limk uvkk is nite, then
1. If
k=1
k=1 uk converges.

vk
2. If k=1 wk is a divergent series without negative terms and limk wk is positive, then
k=1 uk diverges.
The Limit Comparison Test says to make a ratio of the terms of two series and compute the limit. This
test is most useful for series with rational expressions.

k4 +6k3 1
Example 8 Determine if
k=1 7k5 +K 2 converges or diverges.
Solution
Just as with rational functions, the behavior of the series

k4 +6k3 1
k=1 7k5 +K 2

when k goes to innity behaves

k4
like the series with only the highest powers of k in the numerator and denominator:
k=1 7k5 . We will use

k4
k4
the series
5 to apply the Limit Comparison Test. First, when we simplify the series
k=1 7k5 , we
k=1
7k
1
1
1 1
get the series k=1 7k . This is a harmonic series because k=1 7k = 7 k=1 k and the multiplier 17 does not

1
aect the convergence or divergence. Thus,
k=1 7k diverges. So, we will next check that the limit of the
ratio of the terms of the two series is positive:
k2 +6k3 1
7k5 +k2
lim
1
k
7k

Using the Limit Comparison Test, because

diverges.

1
7k

7k4 + 42k3 7
= 1 > 0.
k
7k4 + K

= lim

diverges and the limit of the ratio is positive, then

k4 +6k3 1
k=1 7k5 +K 2

Unlike the Comparison Test, you do not have to compare the terms of both series. You may just make a
ratio of the terms.
www.ck12.org

356

There is a Simplied Limit Comparison Test, which may be easier for you to use.
Theorem ( The Simplied Limit Comparison Test)

Suppose
uk and
v are series without negative terms. If limk uvkk is nite and positive, then
k=1
k=1

either k=1 uk and both k=1 vk converge or

k=1 uk and
k=1 vk both diverge.

Example 9 Determine if k=1 8k2+5 converges or diverges.

Solution

is a series without negative terms. To apply the Simplied Limit Comparison Test, we can
2

2
8k +5
2
compare
=
with
the
series
,
which
is
a
convergent
geometric
series.
Then
lim
k
2
k
k
k=1 8 +5
k=1 8
8k

2
2
limk 8k8+5 = 1 > 0. Thus, since
k=1 8k +5 also converges.
k=1 8k converges, then
k=1 8k +5

Review Question
1. Write an example of a nondecreasing sequence.
2. Write an example of a sequence that is not nondecreasing.
3. Suppose {S n } is a nondecreasing sequence such that for each M > 0, there is an N, such that S n > M
for all n > N. {Does }the sequence converge? Explain.
2
4. Determine if 2n5n2 +7 converges or diverges. If it converges, nd its limit.
( )k1

1
5. Determine if
converges or diverges. If it converges, nd its sum.
k=3 4
Determine if each series converges or diverges.
6.

k=1

7.
k=1
8.
k=1

9. k=1

10.
k=1
11.
k=1

1
1

(4k+1) 2
2
3k5 4
5
(k+1)(k+3)
7

5 2
k
k3 +4k2 +1
3k6 +2k4
1
5

(3k1) 2

+
3

3
12. Maria uses the integral test to determine if
converges.
She
nds
that
= 3. She then
2
k=1 k
x2
1

3
states that
k=1 k2 converges and the sum is 3. What error did she make?

Review Answers
1. Sample answer: 4, 4, 4, 5, 5, 5, 6, 6, 7, . . .
2. Sample answer: 200, 20, 2, . . .
3. No, the nondecreasing sequence is not bounded. For any number M, there is a point in the sequence
such that the rest of the sequence is greater than M. The terms of the sequence are not bounded by
any
{ 2value
} of M.
5n
4. 2n2 +7 is bounded and nonincreasing. It converges to 52
(
)
( )k1

1
5.
is a geometric series with a = 1 and r = 14 . The sum is 11 1 1 + 14 = 13 45 = 43 54 =
k=3 4
6.

1615
12

k=1

1
12 .
1

1
(4k+1) 2

diverges by Comparison test (using

357

k=1

1
1

, a divergent pseries).

(4k) 2

www.ck12.org


2
2
7.
pseries).
5 4 converges by the Limit Comparison Test (using
k=1
k=1 3k5 , which is a convergent
3k

5
5
5
8.
=
converges
by
the
Limit
Comparison
Test
(using
,
which
is a
k=1 (k1)(k3)
k=1 k2 +4k+4
k=1 k2
convergent pseries).

9. k=1 57 2 is a divergent pseries.

k
k3 +4k
1
2
10. k=1 3k6 +2k+1
4 converges by the Limit Comparison Test (using
k=1 3k3 , which is a convergent pseries).

1
11.
converges
by
the
Integral
Test.
5
k=1
(3k1) 2

12. Maria is correct that the series converges. She made an error by saying that the value of the related

3
3
3
integral gives the sum of the innite series. However,
k=1 (k2 ) = 3 + 2 + 9 + . . . is greater than 3.

Keywords
1.
2.
3.
4.
5.
6.
7.
8.

nondecreasing sequence
harmonic series
geometric series
pseries
Comparison Test
Integral Test
Limit Comparison Test
Simplied Limit Comparison Test

8.4 Series With Odd or Even Negative Terms

Learning Objectives

Demonstrate an understanding of alternating series

Apply the Alternating Series Test to an appropriate series
Explain the dierence between absolute and conditional convergence
Determine absolute and/or conditional convergence of series

Alternating Series (harmonic, geometric, p-series)

Alternating series are series whose terms alternate between positive and negative signs. Generally,
alternating series look like one of these expressions:
u1 u2 + u3 u4 + . . . or u1 + u2 u3 + u4 . . . .
Either the terms with the even indices can have the negative sign or the terms with the odd indices can
have the negative sign. The actual numbers represented by the ui s are positive.
There are several types of alternating series. One type is the alternating harmonic series:

k=1

(1)k+1

1 1 1
1
= 1 + + ....
k
2 3 4

This series has terms that look like the harmonic series but the terms with even indices have a negative
sign.
Another kind is the alternating geometric series. Here is one example:
www.ck12.org

358

(1)k

k=1

( 2 )k1
3

= 1 +

2 ( 2 )2 ( 2 )2

+
....
3
3
3

The odd-indexed terms of this series have the negative sign.

The alternating pseries is another type of alternating series. An example could look like this:

1
1
1
1
(1)k1
=1
+

....
3
3
3
3
2
3
4
k
k=1
From all of these examples, we can see that the alternating signs depend on the expression in the power of
1 in the innite series.

The Alternating Series Test

As its name implies, the Alternating Series Test is a test for convergence for series who have alternating
signs in its terms.
Theorem (The Alternating Series Test)
The alternating series u1 u2 + u3 u4 + . . . or u1 + u1 + u2 u3 + u4 . . . converge if:
1. u1 u2 u3 . . . uk . . .
and
2. limk+ uk = 0.
Take the terms of the series and drop their signs. Then the theorem tells us that the terms of the series
must be nonincreasing and the limit of the terms is 0 in order for the test to work. Here is an example of
how to use The Alternating Series Test.
Example 1
Determine how if

k+1 k+5
k=1 (1)
k3 +k

converges or diverges.

Solution

k+1 k+5 is an alternating series. We must rst check that the terms of the series are
The series
k=1 (1)
k3 +k
u
uk+1
nonincreasing. Note that in order for uk uuk + 1, then 1 k+1
uk , or uk 1.
So we can check that the ratio of the (k + 1)st term to the k th term is less than or equal to one.
uk+1
=
uk

(k+1)+5
(k+1)3 +(k+1)
k+5
k3 +k

(k + 1) + 5
k3 + k

(k + 1)3 + (k + 1) k + 5

Expanding the last expression, we get:

(k + 6)(k3 + k)
uk+1
k4 + 6k3 + k2 + 6k
= 3
=
.
uk
(k + 3k2 + 4k + 2)(k + 5)
k4 + 8k3 + 19k2 + 20k + 10
Since k is positive and all the sum of the numerator are part of the denominators sum, the numerator is
u
less than the denominator and so, k+1
uk < 1. Thus, uk uk + 1 for all k. By the Alternating Series Test, the

k+5
k+1
series k=1 (1) k3 +k converges.
Keep in mind that both conditions have to be satised for the test to prove convergence. However, if the
limit condition is not satised, the innite series diverges.

359

www.ck12.org

Alternating Series Remainder

We nd the sequence of partial sums for an alternating series. A partial sum can be used to approximate
the sum of the series. If the alternating series converges, we can actually nd a bound on the dierence
between the partial sum and the actual sum. This dierence, or remainder, is called the error.
Theorem (Alternating Series Remainder)
Suppose an alternating series satises the conditions of the Alternating Series Test and has the sum S . Let
sn be the nth partial sum of the series. Then |S sn | un+1 .
The main idea of the theorem is that the remainder |S sn | cannot get larger than the n + 1st term in
the series, un+1 . This is the term whose index is one more than the index of the partial sum used in the
dierence.
Example 2
Compute s3 for the series

k=1 (1)

k+1 k+5
k3 +k

and determine the bound on the remainder.

Solution
First we compute the third partial sum to approximate the sum S of the series:
1+5
2+5
3+5
+ (1)3 3
+ (1)4 3
3
1 +1
2 +2
3 +3
6
7
8
=
+
4 10 30
90 42 + 16
64
=
=
60
60

s3 = (1)2

The theorem tells us to use the next term in the series, u4 , to calculate the bound on the dierence or
remainder. Remember that the part (1)k+1 just gives the sign of the term and, so we just use the part
k+5
to calculate u4 .
k3 +k


Thus u = 4+5 = 9 . Then |S s | = S 64  < 9 0.13. This tells us that the absolute value of the
4

43 +4

68

error or remainder is less than 0.13.

60

68

Absolute and Conditional Convergence

There are other types of convergence for innite series: absolute convergence and conditional convergence.
Absolute Convergence

Let
u = u1 + u2 + u3 . . . + uk + . . . be an innite series. Then the series is absolutely convergent if
k=1 k
k=1 |uk | = |u1 | + |u2 | + |u3 | + . . . + |uk | + . . . converges.

The innite series

1 | + |u2 | + |u3 | + . . . + |uk | + . . . is the series made by taking the absolute
k=1 |uk | = |u
values of the terms of the series
k=1 |uk | = |u1 | + |u2 | + |u3 | + . . . + |uk | + . . . .
The convergence of the series of absolute values tells us something about the convergence of the series.
Theorem

If
k=1 |uk | = |u1 | + |u2 | + |u3 | + . . . + |uk | + . . ., then
k=1 |uk | = |u1 | + |u2 | + |u3 | + . . . + |uk | + . . . also converges.
This tells us that if you can show absolute convergence, then the series converges.
If the series of absolute value diverges, we cannot conclude anything about the series.
Example 3
www.ck12.org

360

Determine if the series

(1)k+1
k=1 3k4 k

converges absolutely.

Solution

 (1)k+1 
1
k=1  3k4 k  =
k=1 3k4 k , which behaves like the series
k=1
(1)k+1
This is a pseries with p = 4. The series k=1 3k4 k converges absolutely and hence converges.
We nd the series of absolute values:

1
.
3k4

Example 4
Determine if the series

k=1

(1)k+1
2k+1

converges absolutely.

Solution

 (1)k+1  1 1 1
The series made up of the absolute values of the terms is k=1  2k+1  = 3 + 5 + 7 +. . . . This series behaves

1 1
1
k+1 1
like
k=1 2k = 2 k=1 k , which diverges. The series
k=1 (1)
2k+1 does not converge absolutely.
It is possible to have a series that is convergent, but not a absolutely convergent.
Conditional Convergence
An innite series that converges, but does not converge absolutely, is called a conditionally convergent
series.
Example 5
Determine if

k=1

(1)k+1
k

converges absolutely, converges conditionally, or diverges.

Solution

1
The series of absolute values is
k=1 k . This is the harmonic series, which does not converge. So, the series
k+1
(1)

(1)k+1
does not converge absolutely. The next step is to check the convergence
. This
k=1
k=1
k
k
will tell us if the series converges conditionally. Applying the Alternating Series Test:
The sequence 11 > 12 > 13 > . . . is nondecreasing and limk 1k = 0.

(1)k+1
The series
converges. Hence, the series converges conditionally, but not absolutely.
k=1
k

Rearrangement
Making a rearrangement of terms of a series means writing all of the terms of a series in a dierent
order. The following theorem explains how rearrangement aects convergence.
Theorem

If
k=1 uk is an absolutely convergent series, then the new series formed by a rearrangement of the terms
of the series also converges absolutely.
This tells us that rearrangement does not aect absolute convergence.

Review Questions
Determine if the series converges or diverges.

1.
(1)k+1 3k2k+k
k=1
k+1 3k
2.
k=1 (1)
2k
k+1 4 .
3. Compute s3 for
k=1 (1)
k3

k 5
4. The series k=1 (1)k k52 converges according to the Alternating Series Test. Let
k=1 (1) k2 = S .

Compute s3 for k=1 (1)k k52 and determine the bound on |s3 S |.

361

www.ck12.org

5. The series

k=1

Compute s4 for

(1)k+1
converges according to
k!
(1)k+1
and determine the
k=1
k!

the Alternating Series Test. Let

(1)k+1
k
(1)k+1
k
(1)k+1
k


S  < 0.05

S  < 0.005

S  < 0.0001

Determine if each series converges absolutely, converges conditionally, or diverges.

9.
10.
11.
12.

k+1 3k
k=1 (1)
2k
k+1
(1) k
k=1 2k2 +2
(4)k+1
k=1 7k2
(1)k+1
7
k=1
k2

Review Answers
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.

(1)k+1 3k2k+k converges by the Alternating Series Test.

k=1

k+1 3k converges by the Alternating Series Test.

k=1 (1)
2k
197
s3 = 54 ;
5
s3 = 151
36 ; |s3 S | < 16
15
1
s3 = 24 ; |s3 S | < 120
19
199
9, 999
converges absolutely
converges conditionally
divergent
converges absolutely

Keywords
1.
2.
3.
4.
5.
6.
7.
8.
9.

alternating series
alternating harmonic series
alternating geometric series
alternating pseries
Alternating Series Test
Alternating Series Remainder
conditional convergence
absolute convergence
rearrangement

www.ck12.org

k=1

(1)k+1
k!

= S.

bound on |s4 S |.

(1)k+1
The series
converges according to the Alternating Series Test. Let
k=1
k=1
k
least value of n such that:

6.  nk=1

7.  nk=1

8.  nk=1

362

(1)k+1
k

= S . Find the

8.5 Ratio Test, Root Test, and Summary of Tests

Ratio Test
We have seen the Integral Test,(Limit) Comparison Test and Alternating Series Test which impose conditions on the sign of an . We are going to introduce two tests for a stronger version of convergence that do
not.
Denition (Absolute convergence)

A series
an is absolutely convergent if the series of the absolute values
|an | is convergent.
To this end, we need to distinguish the other type of convergence.
Denition (Conditional convergence)

A series
an is conditionally convergent if the series is convergent but not absolutely convergent.
Theorem If

an is absolutely convergent, then it is convergent.

The proof is quite straightforward and is left as an exercise. The converse of Theorem 5.1.1 is not true. The
(1)n1

series
is convergent by the Al- ternating Series Test, but its absolute series, 1n (the harmonic
n
series), is divergent.



cos n
is absolutely convergent since  cosn2 n  n12 for any 1 n, , and n12 is convergent
Example 1
n2

(e.g. by the p test). Indeed, by the Integral and Comparison tests, cosn p n is absolutely convergent for
any and p > 1.



an +1 

The limit of the ratio  an  gives us a comparison of the tail part (i.e n Large an ) of the series an with
a geometric series.

Theorem (The Ratio Test) Let an be a series of non-zero numbers .



 = < 1, then the series is absolutely convergent.
(A) If limn  ana+1
n




 = > 1 or limn  an +1  = then the series is absolutely divergent.
(B) If limn  ana+1
an
n


 = = 1 then the test is inconclusive.
(C) If limn  ana+1
n
: we could ignore the zero-valued an s as far as the sum is concerned.
Proof. (A) The proof is by comparison with a geometric series.



 <
< 1 It follows from the denition of limit that there is an integer N,  ana+1
n
for all n N. Let = +1
|a
| < |aN |, |aN+2 | < |aN+1 | < 2 |aN |, . . . and recursively we have
2 Then
N+1

nN+1
nN which is nite. Combining with the nitely
|an+1 | <
|aN | for n N, and n=N |an | < |aN |
n=N

many terms, n=0 |an | is still nite.

If < 1, then <

+1
2

(B) A similar argument concludes limn an , 0. So the series is divergent.

n
Example 2 Test the series An! for absolute convergence where A is a constant.


n
a
An+1
n!
A

Solution. Let an = An! . Then  n+1
an  = (n+1)! An = n+1 0 as n . So by the Ratio Test, the series is
absolutely convergent for any constant A. Indeed, the sum is eA which is very large for large A, but still
nite.


a
We see the limitation of the Ratio Test is when lim  n+1  does not exist (not ) or is 1.
an

363

www.ck12.org

1


1
an+1 
(n+1)2
n

Example 3 (Ratio Test inconclusive) limn n+1
=
lim
lim
=
1,
and
lim
n n+1
n  aN  = 1
n
1
1
n
n2
1
1
for both n and n2 The former (harmonic series) diverges while the latter converges (by, say, the ptest).

Questions (related to the Ratio Test) What if







a
 exist separately for n odd and n even, i.e limn  a2n+1  , limn  a2n  exist but are
1. limit of  n+1
an
a2 n
a2 n1
dierent?




a
 , limn  a2n+2  exist but are dierent?
2. limn  a2n+1
a
2n1
2n
Exercise
Determine whether the following series is absolutely convergent, conditionally convergent, or divergent
with the Ratio Test and other tests if necessary:

n3
1.
n=1
2n

2.
e2n n!
n=1

n1 1
3.
n=1 (1)
n
4.

2
1.3

3
2.4

4
3.5

5
4.6

+ ...

nth - root Test

If the general term an resembles an exponential expression, the following test is handy.
Theorem (The Root Test)
 
(A) If limn  n an  = < 1, then the series is absolutely convergent.
 
 
(B) If limn  n an  = > 1 or limn  n an  = , then the series is absolutely divergent.
 n 
 n = = 1, then the test is inconclusive.
(C) If lim
n

The proof is similar to that of the Ratio Test and is left as an exercise.

Example 1 Consider an = n1p where p > 0. We already know it is convergent exactly when p > 1.
n
To apply
 the Root Test, weneed limn n which is 1 after some work. Alternatively, we could check
a
n

limn  n+1
an  = 1 so limn n = 1 by the argument similar to the proof of the Ratio Test. The Root Test
is also inconclusive.
( n+1 )n
Example 2 Test the series
for convergence.
2n+3
(
)n

n+1
n+1
Solution. Let an = 2n+3
. Then limn n |an | = limn 2n+3
= 21 . So the series is absolutely absolutely
convergent.
What if we have applied the Ratio Test?
]n


(
)
(
)
(
) [
a
 = limn n+2 n+1 / n+1 n = limn n+2 . (n+2)(2n+3)
limn  n+1
an
2n+5
2n+3
2n+5
(n+1)(2n+5)
We could still argue the limit is

1
2

with some work. So we should learn to apply the right test.

Exercise
Determine whether the following series is absolutely convergent, conditionally convergent, or divergent
with the Root Test and other tests if necessary:
1.

)n
n
n=1 n2 +1

www.ck12.org

364

2.
3.

( (1)n (ln
n=1

n)

)n

nn

n=1 53+2n


a

If limn  n+1
an 

 
= , then limn  n an  = .
( 2 )n

n 1
convergent?
5. (Hard) Is the series
n=1 n2 +1
4.

Hint:

)
n2 1 n
2
n +1

[
(
n

)
n2 1 n2
2
n +1

and limh0 (1 + h)1/h exists and equals e.

Summary of Procedures for Determining Convergence

We have seen various test for convergence of
the form of an .

an in action. To summarize, the key phrase is recognize

Some series are hard for any test even if it works. However, if a specic test works wells for a series, we
should develop the insight to pick it. So applying the list of tests in a x order is not advised. Nonetheless,
some tests are relatively simple to apply if we have no idea towards a series at all. The following is a
summary:
Table 8.2:

Test

Form of of an

Comments

No/little test

geometric,
harmonic,
ptest
limn an = 0
corresp. integral in nice
closed form

(1)n bn
(or

(1)n1 bn )
need companion known
series
recognize good form

clear answer

2
3

Test of divergence
Integral Test

Alternating Series

(Limit) Comparison

Ratio Test
Root Test
Combination

7
8

an+1
an

an resembles cn n
composite of forms

inexpensive test
easy integration
check conditions on bn
compare
evaluate ratio
evaluate nth root
combined methods

: the inconclusive cases need other tests.

Example 1

1
diverges by the Integral Test since
n(ln n)

Example 2 sin

1
n

1
dx
x(ln x)

= ln |ln |x|| + C diverges.

n diverges by Limit Comparison Test (against the harmonic series) since limn

limh0 sinh h

= 1 and the harmonic series diverges.

( )

Example 3 1n sin 1n converges with Limit Comparison Test (against n12 ) since limn
limh0

sin h
h

1
n

sin
1
n2

1
n

sin

1
n

1
n

limn

=
sin
1
n

= 1 and the latter series converges.

n
Example
4 Convergence of (cn )nis determined with lim
( 3
n |cn | by the Root Test. For example
n n
n
n
n
2) , converges since limn ( 3 2) = 11 = 0, but ( 21) (no exponent n) diverges by rationalizing

365

www.ck12.org

1
n

the numerator:
n
2 1 = n1 21
n2

n
1
= 12 n2 and applying Limit Comparison test with 2n
.
2 n +2 n +...+1

The Root Test is inconclusive on (1 + 1n )n , but the simpler Test for Divergence conrms its divergence
since (1 + 1n )n > 1 always.
(1)n1
Example 5
is convergent for q > 0 by the Alternating Series Test since limn n1q = 0 and
nq
1
1
nq (n+1)q . It is absolutely convergent for q > 1 by the ptest. So it is conditionally convergent for
0 < q 1.

2.2n1
2n
2n+1 n!
n.2n
Example 6
n=1 n! = n=2 (n1)! = 2 n=1 n! is absolutely convergent by the Ratio Test since limn (n+1)! . 2n =

limn

2
n+1

21
n1
n.2 n

= 0.

n+1

. Notice n2 10n + 1 is never 0 and is positive for n 10, we

n 1
could ignore the terms before n = 10. Dropping the lower powers of n leads to the candidate
=
3
n2

Example 7 Consider the series

n2 10n+1

for applying Limit Comparison Test since limn

by the ptest. A combination of tests is applied.

n+1
n2
.
n2 10n+1
n

n2

= 1. So the series is (absolutely) convergent

Multimedia Links
For video presentations of the varying tests for convergence (23.0), see Just Math Tutoring, Using the
Ratio Test to Determine if a Series Converges (7:38) ; Just Math Tutoring, Root Tests for Series (10:07)

Austin Math Tutor, Austin Math Tutoring, Austin Algebra Tutor, Austin Calculus Tutor
Austin Math Tutor, Austin Math Tutoring, Austin Algebra Tutor, Austin Calculus Tutor
Figure 8.1: Using the Ratio Test to Determine if a Series Converges #1 (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/580
;Just Math Tutoring, Limit Comparison Test and Direct Comparison Test (7:35) ; Just Math Tutoring,
Alternate Series Test (7:35) .
Exercises
1. For what values of p is the series

1
n=2 n(ln n) p

convergent?

Test the following series for convergence or divergence:

2 1
2.
n=1 sin n

(1)n1 n
3.
n=1
n+1
n
n
4. n=1 ( 5 3)n
n
n

5.
n=1 ( 5 3)

3n
6. For what values of c is the series
n=1 2n +cn convergent?
www.ck12.org

366

austinmathtutor, UT math tutor, UT math tutoring, justmathtutoring.com, austin calculus tutor, austin
math tutor
austinmathtutor, UT math tutor, UT math tutoring, justmathtutoring.com, austin calculus tutor, austin
math tutor
Figure 8.2: Root Test for Series (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/581

Austin Math Tutor, Austin Math Tutoring, Austin Algebra Tutor, Austin Calculus Tutor
Austin Math Tutor, Austin Math Tutoring, Austin Algebra Tutor, Austin Calculus Tutor
Figure 8.3: Limit Comparison Test and Direct Comparison Test (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/582

Figure 8.4: More Alternating Series Examples (Watch Youtube Video)

http://www.ck12.org/exbook/embed/view/583

367

www.ck12.org

8.6 Power Series

Power Series and Convergence
Denition (Power Series)
A Power Series is a series of the form

(PS1)

n=0 an x

= a0 + a1 x + a2 x2 + a3 x3 + . . .

where x is a variable and the an s are constants (in our case, real numbers) called the coeicients of the
series.

The summation sign

is a compact and convenient shorthand notation. Reader unfamiliar with the
notation might want to write out the detail a few times to get used to it.
Power series are a generalization of polynomials, potentially with innitely many terms. As observed, the
indices n of an are non-negative, so no negative integral exponents of x, e.g. 1x appears in a power series.
More generally, a series of the form

n
2
3
(PS2)
n=0 an (x x0 ) = a0 + a1 (x x0 ) + a2 (x x0 ) + a3 (x x0 ) + . . .
is called a power series in (x x0 )(1 = (x x0 )0 ) or a power series centered at x0 ((PS1) represents series
centered at x = 0).
Given any value of x, a power series ((PS1) and (PS2)) is a series of numbers. The rst question is:
Is the power series (as in (PS1) or (PS2)) a function of x?
Since the series is always dened at x = 0 (resp. x = x0 ), the question becomes:
For what value of x is a power series convergent?
The answers are known for some series. Convergence tests could be applied on some others.
Example 1 Let r , 0 and x0 be real.

n
n=0 (r(x x0 ))

= 1 + r(x x0 ) + r2 (x x0 )2 + r3 (x x0 )3 + . . . is absolutely convergent and equals

 
for |r(x x0 )| < 1, i.e. |x x0 | <  1r , and diverges otherwise.

1
1r(xx0 )

1
Let r = 1; x0 = 0. Then xn is the power series for 1x
on (1, 1). Let r = 1; x0 = 2. Then (1)n (x 2)n

1
1
is the power series for 1(1)(x2)
= x1
on (1, 3). So xn and (1)n1 (x 2)n are the power series for the
1
same function 1x
but on dierent intervals. There will be more detailed discussion in 8.7.
2n

n
Example 2 n=0 x is absolutely convergent for |x| < 1 by Comparison Test (against
n=0 x ) and diverges
for x = 1 by the Test for Divergence.

Exercise

n
1. Write a power series
n=0 an (x + 2) centered at x = 2 for the same function
what interval does equality hold?
Hint: Substitute y = x + 2 in

1
1x .

2. Discuss the convergence of the series

n 2n
n=0 2 x .

Hint: Apply a combination of tests in 8.5.

www.ck12.org

368

1
1x

in Example 6.1.1. On

Interval and Radius of Convergence

The following theorem characterizes the values of x where a power series is convergent.
Theorem (Interval of convergence)

n
Given a power series
n=0 an (x x0 ) . Exactly one of the following three describes all the values where the
series is convergent:
(A) The series converges exactly at x = x0 only.
(B) The series converges for all x.
(C) There is a real number Rc > 0 that the series converges if |x x0 | < Rc and diverges if |x x0 | > Rc .
This Rc is unique for a power series, called the radius of convergence. By convention Rc = 0 for case
(A) and Rc = for case (B). The only two values of x the Theorem cannot conrm are the endpoints
x = x0 Rc . In any case, the values x where the series converges is an interval, called the interval of
convergence. It is the singleton x0 for case (A) and (, ) for case (B). For case (C), it is one of the
four possible intervals: (x0 Rc , x0 + Rc ), (x0 Rc x0 + Rc ], [x0 Rc x0 + Rc ), and [x0 Rc x0 + Rc ]. Here, the
endpoints must be checked separately for convergence.

xn
Example 1 Find the radius of convergence and interval of convergence of the series
n=0 n2 .
)2 

  n+1
 (

bn +1 
xn
n2 
x
1


Solution. Let bn = n2 . Then  bn  =  (n+1)2 xn  =  1+ 1 x |x| as n . So the series is absolutely
n
convergent for |x| < 1 (and divergent for |x| > 1) by the Ratio Test. Rc = 1. This leaves the endpoint values
to check.
If x = 1, then the series is absolutely convergent by the ptest. Hence the series is absolutely convergent
for|x| 1. The interval of convergence is [1, 1].

n
n
inside the
Example 2 If the series
n=0 an x isconvergent at x = 2, i.e. 2 is
n=0 an 2 converges, then

n
interval of convergence. So Rc 2. Conversely, if the series n=0 an (3) diverges, then n=0 an xn is
divergent at x = 3, i.e. 3 is outside the interval of convergence. So Rc | 3| = 3.
Exercise
Find the radius of convergence and interval of convergence of the following series.

n
1.
n=1 nx

(1)n xn
2.
n=1
xn/3n
3. n=1 n!

4.
n(x x0 )n
n=1
n
5. Given
n=0 an x converges at x = 5 and diverges at x = 7. Deduce where possible, the convergence
or divergence of these series:

(a)
a
n=0 n n
(b)
a 3
n=0 n
(c)
an (8)n
n=0
(d)
an (9)n
n=0

(e) n=0 an (6)n

Term-by-Term Dierentiation of Power Series

The goal of the next 3 sections is to nd power series representations of certain classes of functions, namely
derivatives, integrals and products.

369

www.ck12.org

In the study of dierentiation (resp. integration), we have found the derivatives (resp. integrals) of better
known functions, many with known power series representations. The power series representations of
the derivatives (resp. integrals) can be found by term-by-term dierentiation (resp. integration) by the
following theorem.
Theorem (Term-by-term dierentiation and Integration)

n
Suppose
n=0 an (x x0 ) has radius of convergence Rc . Then the function f dened by f (x) = n=0 an (x
x0 )n is dierentiable on (x0 Rc , x0 + Rc ) and

n1 ,
(A) f (x) =
n=1 nan (x x0 )

an (xx0 )n+1
(B) f (x)dx =
+ C, and these power series have same radius of convergence Rc .
n=0
n+1
(A) means (dropping x0 the derivative of a power series is the same as the term-by-term dierentiation of
the power series:
d
d
n
n
dx n=0 an x = n=0 dx (an x ) and
(B) means the integral of a power series is the same as the term-by-term integration of the power series:

(an xn )dx =
(an xn )dx.
n=0

Example 1 Find a power series for g(x) =

n=0
1
(1x)2

and its radius of convergence.

1
n
Solution. We recognize g(x) as the derivative of 1x
whose power series representation is
n=0 x with

d
n1 and has radius of convergence 1.
n
radius of convergence Rc = 1. By (A), g(x) = dx
n=0 x = n=1 nx
Exercise
Find a power series and the radius of convergence for the following functions:
1.
2.
3.

x
2x
x2
(1x)3
2x
(1x)3

3x2
(1x)4

Multimedia Links
For a video presentation of nding the interval of convergence of a power series (24.0), see Just Math
Tutoring, Power Series, nding the interval of convergence (9:46) .

Austin Math Tutor, Austin Math Tutoring, Austin Algebra Tutor, Austin Calculus Tutor
Austin Math Tutor, Austin Math Tutoring, Austin Algebra Tutor, Austin Calculus Tutor
Figure 8.5: Power Series - Finding the Interval of Convergence (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/584

www.ck12.org

370

Term-by-Term Integration of Power Series

Example 1 Find a power series for h(x) = tan1 x and its radius of convergence.

1
n 2n
Solution. We recognize h(x) as the antiderivative of 1+x
2 =
n=0 (1) x .

(1)n x2n+1
n 2n
By Term-by-Term Theorem (B), h(x) =
+ C and has radius of convern=0 (1) x dx = n=0
2n+1
gence 1.

(1)n x2n+1
Then C = tan1 0 = 0 and h(x) =
n=0
2n+1 .
Exercise
1. Find a power
series for ln(1 + x2 ) and nd the radius of convergence.

1
2. Express tan xdx as a power series and nd the radius of convergence.
3. Find a power series for ln(1 + x + x2 ) in x and in x + 21 , and nd the radius of convergence.

Multimedia Links
For a video presentation showing how to dierentiate and integrate power series (25.0), see Just Math
Tutoring, Dierentiating and Integrating a Power Series (10:09) .

Austin Math Tutor, Austin Math Tutoring, Austin Algebra Tutor, Austin Calculus Tutor
Austin Math Tutor, Austin Math Tutoring, Austin Algebra Tutor, Austin Calculus Tutor
Figure 8.6: Dierentiating and Integrating Power Series (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/585

Series Multiplication of Power Series

n
an xn and
Denition (Series Multiplication) The power series product of two power series
n=0
n=0 bn x

is a power series n=0 cn xn dened by cn = ni=0 ai bni = a0 bn + a1 bn1 + . . . + an b0 (like polynomials).

A result is: the product of power series is the power series of the product.

n
If f (x) =
xn and g(x) =
n=0 an
n=0 bn x converges on a common interval |x| < Ra b, then their product

n
power series n=0 cn x also converges on Ra b and is the power series for the product function f (x)g(x).
1
.
Example 1 Find a power series for (1x)(12x)

1
n
Solution. 1x
=
n=0 x with radius of convergence 1 and

1
12x

n
n=0 (2x)

with radius of convergence 12 .

So for |x| < 12 ,

1 (1x)(12x)=(1+x+x2 +...)(1+2x+4x2 +...)=1+3x+7x2 +15x3 +...by the above result=

n=0 (2

371

n+1 1)xn .

www.ck12.org

Exercise
1. Find the rst 4 terms of a power series for
2. Find a power series for

1
(1rx)(1sx)

1
.
(1+x)(1+2x)

where r, s > 0 are real numbers and

(a) r , s
(b) r = s.

8.7 Taylor and Maclaurin Series

Taylor and Maclaurin Polynomials
We know the linear approximation function L1 (x) to a (smooth) function f (x) at x = x0 is given by the
tangent line at the point. L1 (x0 ) = f (x0 ) and L1 (x0 ) = f (x0 ). Indeed, this is the only linear function with
these 2 properties.
Theorem (nth degree Taylor polynomial) Given a function f with continuous nth derivative in an open
interval containing x0 . There exists unique nth degree polynomial p(x) with p( j) (x0 ) = f ( j) (x0 ), for
0 j n.
: the functions in this text have continuous derivatives at the center x0 unless otherwise stated.
This polynomial
f (n) (x0)

T n (x) = f (x0 ) + f (x0 )(x x0 ) + 2!0 (x x0 )2 + . . . +

(x x0 )n ) is called the nth degree Taylor
n!
th
polynomial of f at x0 . If x0 = 0, it is called the n degree Maclaurin polynomial of f and denoted by
Mn (x). Rn (x) = f (x) T n (x) is the remainder of the Taylor polynomial.
f (x )

Example 1 Let f (x) = x6 , x0 = 1. Then f (x) = x2 and f (x) = x. So f (1) =

T 1 (x) = 61 + 12 (x 1), T 2 (x) = 16 + 12 (x 1) + (x 1)2 , and T 3 (x) = f (x) itself.
3

1
2

and f (1) = 1. Hence

Example 2 Let f (x) = sin x, x0 = 0 and take n = 3. Then f (x) = cos x, f (x) = sin x, f (x) = cos x.
So f (0) = 1, f (0) = 0, f (0) = 1. M3 (x) = x 3!1 x3 = x 16 x3 is the third-degree Maclaurin polynomial
of f .
www.ck12.org

372


Example 3 Find the second-degree Taylor polynomial of f (x) = (tan
f (x) = sec2 x
) x at x0 = 4 . Solution.
( )
and f (x) = 2 sec x sec x tan x = 2 sec2 x tan x. So f 4 = 2 and f 4 = 2 2 = 4. Then
(
)
(
)2
(
)
(
)2
T 2 (x) = 1 + 2 x 4 + 2!4 x 4 = 1 + 2 x 4 + 2 x 4 .

Exercise Find the Taylor series of the following functions at the given x0 with given degree n.
1. f (x) = e x at x = 0, n = 3
2. f (x) = ln x at x = 1, n = 4
3. f (x) = 1 + x + x2 + x3 + x4 at x = 1, n = 4

Taylor and Maclaurin Series

Denition (Taylor Series of f )
The Taylor series of a function f at x = x0 is the power series

f (n) (x0 )
T (x) =
(x x0 )n
n!
n=0

= f (x0 ) + f (x0 )(x x0 ) +

f (x0 )
f (x0 )
(x x0 )2 +
(x x0 )3 + . . .
2!
3!

taking all the terms of the Taylor polynomials. The Maclaurin series M(x) of f is the Taylor series at
x = 0.
Example 1 Find the Maclaurin series of f (x) = cos x.
Solution. f (x) = sin x, f (x) = cos x, f (x) = sin x, f (x) = cos x, f (4) (x) = sin x, f (5) (x) =
cos x, . . ..
Notice the pattern repeats every 4 terms. So f (0) = 0, f (0) = 1, f (0) = 0, f (0) = 1, f (4) (0) =
0, x, f (5) (0) = 1 . . ..
The Maclaurin series of f (x) = cos x is
M(x) = 1

1
1
1
1 2
x + x4 x 6 + x8 . . . =
n=0
2!
4!
6!
8!

(1)n 2n
x .
(2n)!

Exercise Find the Taylor series of following functions at the given x0 .

1
1. f (x) = 1x
at x = 0
x
2. f (x) = e at x = 1
3. f (x) = 1x at x = 2

Convergence of Taylor and Maclaurin Series

Since f (n) (x) is dened for all functions f in this text, the Taylor series T (x) of f is always dened. As for
power series in general, the rst question is:
Is T (x) convergent at x = a ? There is no guarantee except at a = x0 . The second question is:
If T (x) converges at x = a, does it equal f (a)? The answer is negative as show by the function:

373

www.ck12.org

12
f1 (x) =
e x x,0

0 x = 0
Then

( 2 ) 12

f1 (x) =
e x x,0

x3

0 x = 0
It can be veried that f1 (0) = 0, f1 (0) = 0, f1 (0) = 0, . . ..
So the Maclaurin series is 0, clearly dierent from f except at x = 0.
Nevertheless, here is a positive result.
Theorem If f has a power series representation at x = x0 , i.e.

n
f (x) =
n=0 an (x s0 ) for |x x0 | < Rc , then the coeicients are given by an =

f (n) (x0 )
.
n!

So any power series representation at x = x0 has the form:

f (x) =

f (n) (x0 )
(x x0 )n
n!
n=0

Exercise
(n)

1. Find the higher order derivatives of the function f1 (x) above thus recursively showing f1 (0) = 0 for
n0
2. Verify the Theorem using term-by-term dierentiation.

Taylors Formula with Remainder, Remainder Estimation, Truncation Error

Recall the remainder Rn (x) of the nth degree Taylor polynomial at x = x0 is given by Rn (x) = f (x) T n (x).
Theorem (Convergence of Taylor series)
If limn Rn (x) = 0 for |x x0 | < Rc , then f is equal to its Taylor Series on the interval |x x0 | < Rc .
The above condition limn Rn (x) = 0 could be achieved through the following bound.
Theorem (Remainder Estimation)
If | f (n+1) (x)| M for |x x0 | r, then we have the following bound for Rn (x):
|Rn (x)|

M
|x
(n+1)!

x0 |n+1 for |x x0 | r.

Example 1 The function e x is equal to its Maclaurin series for all x. Proof. Let f (x) = e x . We need to
nd the above bound on Rn (x).
If |x| r, f (n) (x) = e x er for n 0 and the remainder estimation gives |Rn (x)|
Since limn

er
|x|n+1
(n+1)!

0. Hence e x is equal to its Taylor

www.ck12.org

|x|n+1
= 0, limn |Rn (x)|
(n+1)!
xn
series n=0 n! for all x.

= er limn

374

er
|x|n+1
(n+1)!

for |x| r.

= 0 by the squeeze Theorem. So limn Rn (x) =

Example 2 (Truncation Error) What is the truncation error of approximating f (x) =

third-degree Maclaurin polynomial in for |x| 0.1.

1 + x by its

Solution.
1
3
3
1 ( 1)
1
1
f (x) = (1 + x) 2 , f (x) =
(1 + x) 2 = (1 + x) 2 ,
2
2 2
4
( 1) ( 3)
(3) ( 5)
7
7
3
15

25
52
(4)
f (x) =
(1 + x) = (1 + x) , f (x) =
(1 + x) 2 = (1 + x) 2 .
4
2
8
8
2
16

For |x| 0.1, | f (4) (x)| =

15
16

1
7

(1+x) 2

15
16 .

1
5
4
4
6
So |R4 (x)| 15
16 4! |x| 128 (0.1) 4 10 . This is the truncation error of approximating by the
third-degree Maclaurin polynomial.

The following applet illustrates approximating functions with Taylor Series. You can change the center of
the series and also observe how the error changes for the estimation at a particular value of x where the
error is f (x) T n (x). Taylor Series and Polynomials Applet.
Exercise
1. Find the power series representation of f (x) = sin x at x = 0 for all x. Why is it the Maclaurin series?
2. Find the power series representation of f (x) = cos x at x = 3 for all x. Why is it the Taylor series at
x = 3 ?

3. What is the truncation error of approximating f (x) = 1 + x by its fourth-degree Maclaurin series
in for |x| 0.1.

Combining Series, Eulers Formula

In many cases, we could nd Taylor (Maclaurin) series of functions from existing series without going
through the proof that limn Rn (x) = 0 Examples are products, quotients and some sine and cosine
functions.
Example 1 Find the Maclaurin series of f (x) = x sin x for all x.

(1)n x2n+1
Solution. The Maclaurin series of sin x is sin x =
n=0 (2n+1)! . So the Maclaurin series of x sin x is

(1)n x2n+2
(1)n x2n+1
x sin x = x
n=0 (2n+1)! = n=0 (2n+1)!
Example 2 Find the Maclaurin series of f (x) = cos2 x for all x.
Solution. We could avoid multiplying the Maclaurin series of cos x with itself, by applying: cos2 x =
1
1
2 + 2 cos 2x

(1)n (2x)2n
on the Maclaurin series of cos 2x =
, giving
n=0
(2n)!

(1)n (2x)2n
1 1 (1)n (2x)2n
=1+
cos x = +
2 2 n=0
(2n)!
2 (2n)!
n=0
2

For any real number , ei = cos +i sin where i =

Formula.

1 is the imaginary unit. This is the Eulers

Eulers formula combines the complementary sine and cosine functions into the simpler exponential function
and heavily applies the separation of real and imaginary parts of complex numbers.
Example 3 Find the Maclaurin series of cos x and sin x for all x through eix .

375

www.ck12.org

Solution. cos x + i sin x = eix which has a Maclaurin series

eix =

(ix)n
n=0

n!

(ix)2m (ix)2m+1
+
(2m)! m=0 (2m + 1)!
m=0

by dividing into sum of odd and even indices. So cos x + i sin x =

2m
m (x)
m=0 (1) (2m)!

+i

2m+1
m (x)
m=0 (1) (2m+1)!

The Maclaurin series of cos x and sin x follow by separately taking the real and imaginary parts.
Exercise
1. Find and compare the Maclaurin series for sin x cos x and sin 2x.
2. Find the Maclaurin series of

x
ex

for all x.

Hint: would you divide x by e x ?

3. Find the Maclaurin series of cos 3x and sin 3x for all x using Eulers formula.

n
n
4. Find expressions for the series
n=0 cos n x and
n=0 sin n x for all and |x| < 1 using Eulers
formula.

Binomial Series
We have learned the Binomial Theorem for positive integer exponents:
(a + b)n = an + nan1 b +
n ( )

n nk k
=
a b
k
k=0

n(n 1) n2 2
a b + . . . + nabn1 + bn (BE)
2!

where the Binomial coeicients are denoted by

(n)
(n) n(n1)...(nk+1)
for k 1
=
1
and
k!
0
k =
AS a simple Binomial function, take a = 1 and b = x. Then (1 + x)n =

()

n
k=0 0

xk

Let r be a real number and f (x) = (1 + x)r . Is f (x) equal to a series in the form of (BE) except that there
may be an innite series? The answer is yes.
Theorem (Binomial Series) Let r be a real number and |x| < 1. Then
(1 + x)r = 1 + rex +
n ( )

r k
=
x
k
k=0

r(r 1) 2 r(r 1)(r 2) 3

x +
x
2!
3!

where the Binomial coeicients are denoted by

(n)
(n) n(n1)...(nk+1)
for k 1
0 = 1 and k =
2!
As a simple Binomial function, take a = 1 and b = x. Then
n

(1 + x) =

n ( )

1
k=0

www.ck12.org

376

(K)xk .

Let r be a real number and f (x) = (1 + x)r . Is f (x) equal to a series in the form of (BE) except that there
may be an innite series? The answer is yes.
Theorem (Binomial Series) Let r be a real number and |x| < 1. Then
(1 + x)r = 1 + rex +
n ( )

r k
x
=
k
k=0
where

(r )
0

= 1 and

(r )
k

r(r1)...(rk+1)
2!

r(r 1) 2 r(r 1)(r 2) 3

x +
x ...
2!
3!

for k 1

Example 1 Find a power series representation of

1 + x.

Solution. We need to compute the Binomial coeicients for r = 12

( 1 ) ( 1 ) ( 1 ) ( 3 ) . . . ( 2k3 )
(1)k1 1.3.5 . . . (2k 3)
2
2
2
2 = 2
=
k!
k
2k k!
(1)k1 (2k 2)!
(1)k1 (2k 2)!
(1)k1 (2k 2)!
=
=
=
2.4.6 . . . (2k 2))2k k!
2k1 k!(k 1)2k k!
22k1 k!(k 1)!
So if |x| < 1,

(1)

k
2
1+x=
k=0 k x = 1 + k=0

(1)k1 (2k2)! k
x
22k1 k!(k1)!

Example 2 Find a power series representation of

1
(1x)m

where m is a positive integer.

Solution. We need to compute the Binomial coeicients for r = m (and will replace x by x)
( )
(m)(m 1)(m 2) . . . (m k + 1)
m
=
k
k!
(
)
k
(1) m(m + 1)(m + 2) . . . (m + k 1)
k m+k1
=
= (1)
k!
k
(
)

1
m+k1
So
=
(1)k
(x)k
m
k
(1 x)
k=0
(
)

k m+k1
=
(1)
(1)k xk
k
k=0
(
)

2 m+k1 k
=
(1) k
x
k
k=0
)
(

m+k1 k
=
x
k
k=0
Exercise
1. Find a power series representation of

1
1+x
1
(2x)2

at x = 0.

2. Find a power series representation of

at x = 0.
(
)2

3. Notice 1 + x + x2 = x + 34 + x + 21 . Find a power series representation of (1 + x + x2 ) at x = 21 .

In what interval is the equality true?
Choosing Centers
Taylor Series (indeed Taylor polynomials of lower degrees) often provide good approximation of functions.
However, the choice of center could determine

377

www.ck12.org

1. whether the intended value of x is inside the interval of convergence

2. rate of convergence, i.e. how many terms to take to achieve prescribed degree of accuracy
For frequently used functions, the rst choice may be the standard center (see the list at the end of this
section).
Example 1 Approximate ln 0 : 99
Solution. Since .99 is close to the center x = 1, we use the standard Taylor series for ln(1 x).
ln 0.99 = ln(1 0.01) 0.01

(0.01)2
0.01005
2

Then we may be able to deduce a useful Taylor Series centered close to the given x.
Example 2 Approximate sin(1.1) to 4 decimal places.
Since 1.1 is close to 3 , we would try to nd a Taylor Series of sin x at x0 = 3 . Let f (x) = sin x. Then

( )
( )
( )
f (x) = sin x, f (x) = cos x, f (x) = sin x, f (x) = cos x and f 3 = 23 , f 3 = 12 , f 3 =

( )
23 , f 3 = 12 .
This pattern repeats and limn Rn (x) = 0 can be checked as in the case x0 = 0. So the Taylor Series is

(1)n 3 (
)2n (1)n (
)2n+1
sin x =
x
+
x
3
3
2(2n)!
2(2n + 1)!
n=0
n=0
Hence

(1)n 3
(1)n
2n
sin(1.1)
(0.0528) +
(0.0528)2n+1
2(2n)!
2(2n
+
1)!
n=0
n=0
0.86481823 + 0.02638773 taking 2 terms from each sum
0.8912

We may also apply algebraic manipulation to standard Taylor Series.

Example 3 Approximate

1
1.92

to 4 decimal places.

Solution. There is standard Taylor Series:

1
n1 for |x| < 1 through term-by-term dierentiation of the series for
=
k=1 nx
(1x)2
at x = 0.9).

1
1x

1
( (1x)
2 is inadequate

Since 1.9 is close to 2, we consider

( )n1
1
1
1
1
x
=
n
for |x| < 2.
x 2 = 4
2
k=1
4
2
(2x)
(1 )
2

So we take x = 0.1 and then

1
1
1 ( x )n1
n
. (1 + 2(0.05) + 3(0.05)2 + 4(0.05)3 ) 0.277.
=
2
4
2
4
1.9
n=1

Exercise
1. Approximate ln 0.9 to 4 decimal places.
2. Approximate sin(0.8) to 6 decimal places.
Hint: consider center
3. Approximate
www.ck12.org

1
93

to 6 decimal places.

378

Evaluating Nonelementary Integrals

There are many simple-looking functions that have no explicit formula for their integral in the form of
elementary functions. We could write their in-tegrals as Taylor Series in their interval of convergence.
x
Example 1 Find a power series representation of ex dx
x

Solution. Since ex is not dened at x = 0, we apply the Taylor Series of e x at, say, x = 1 by writing
e x = e.e x1 with a change of variable u = x 1.

ex
e.eu
1 (u)n
=
= e.
x
1+u
1 + u n=0 n!
)
(
)(
u2 u3 u4
2
3
+
+
+ ...
= e 1 u + u u + ... 1 + u +
2!
3!
2!
(
)
1
1
3
11
= e 1 + u2 u3 + u4 u5 + . . . for |u| < 1
2
3
8
30
(
)
x
1 4
3 5
11 6
so ex dx = e u + 61 u3 12
u + 40
u 180
u + . . . where u = x 1

2
Example 2 Find the power series representation of sinxx dx

Solution. Direct substitution of x2 in the Maclaurin Series of sin x gives sin x2 =

(1)n 4n+1
sin x2
=
n=0 (2n+1)! x
2

(1)n
2 2n+1
n=0 (2n+1)! (x )

and

So

(1)n
sin x2
dx =
x4n+1 dx
x
(2n
+
1)!
n=0

(1)n
(1)n x4n+2
x4n+2
=
=
(2n + 1)! 4n + 2 n=0 (4n + 2) (2n + 1)!
n=0

Exercise
1. Find the power series representation (Maclaurin Series) of

e x dx and approximate

decimal places.

2. Find the power series representation (Maclaurin Series) of sin x2 dx.

ex dx to 6
2

Frequently Used Maclaurin Series

Some frequently used Maclaurin Series are listed below
n
1
x
1x = n=1

1
n
=
n=1 nx
(1x)2

n
x
ex =
n=1
n!
xn
ln 1 x =
n,
n=1
n+1 xn ,
ln 1 + x =
(1)
n=1
n

(1)n x2n+1
sin x =
n=1 (2n+1)!
(1)n x2n
cos x = n=1 (2n)!

(1)n x2n+1
tan1 x =
n=1 (2n+1)

379

in(1, 1)
in(1, 1)
in(1, 1)
in(1, 1)
in(1, 1)
in(1, 1)
in(1, 1)
in(1, 1)
www.ck12.org

Multimedia Links
For video presentations on the Taylor and Maclaurin Series (26.0), see Just Math Tutoring, Taylor and
Maclaurin Series, Example 1 (6:29) and Just Math Tutoring, Taylor and Maclaurin Series, Example 2

Austin Math Tutor, Austin Math Tutoring, Austin Algebra Tutor, Austin Calculus Tutor
Austin Math Tutor, Austin Math Tutoring, Austin Algebra Tutor, Austin Calculus Tutor
Figure 8.7: Taylor and Maclaurin Series - Example 1 (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/586
(9:45) .

Austin Math Tutor, Austin Math Tutoring, Austin Algebra Tutor, Austin Calculus Tutor
Austin Math Tutor, Austin Math Tutoring, Austin Algebra Tutor, Austin Calculus Tutor
Figure 8.8: Taylor and Maclaurin Series - Example 2 (Watch Youtube Video)
http://www.ck12.org/exbook/embed/view/587

8.8 Calculations with Series

Binomial Series
We have learned the Binomial Theorem for positive integer exponents:
(a + b)n = an + nan1 b +
n ( )

n nk k
=
a b
k
k=0

n(n 1) n2 2
a b + . . . + nabn1 + bn (BE)
2!

where the Binomial coeicients are denoted by

(n)
(n) n(n1)...(nk+1)
for k 1
0 = 1 and k =
k!
AS a simple Binomial function, take a = 1 and b = x. Then (1 + x)n =
www.ck12.org

380

(n)

k=0 0

xk

Let r be a real number and f (x) = (1 + x)r . Is f (x) equal to a series in the form of (BE) except that there
may be an innite series? The answer is yes.
Theorem (Binomial Series) Let r be a real number and |x| < 1. Then
(1 + x)r = 1 + rex +
n ( )

r k
=
x
k
k=0

r(r 1) 2 r(r 1)(r 2) 3

x +
x
2!
3!

where the Binomial coeicients are denoted by

(n)
(n) n(n1)...(nk+1)
=
1
and
for k 1
0
k =
2!
As a simple Binomial function, take a = 1 and b = x. Then
n

(1 + x) =

n ( )

1
k=0

(K)xk .

Let r be a real number and f (x) = (1 + x)r . Is f (x) equal to a series in the form of (BE) except that there
may be an innite series? The answer is yes.
Theorem (Binomial Series) Let r be a real number and |x| < 1. Then
(1 + x)r = 1 + rex +
n ( )

r k
=
x
k
k=0
where

(r )
0

= 1 and

(r )
k

r(r1)...(rk+1)
2!

r(r 1) 2 r(r 1)(r 2) 3

x +
x ...
2!
3!

for k 1

Example 1 Find a power series representation of

1 + x.

Solution. We need to compute the Binomial coeicients for r = 12

( 1 ) ( 1 ) ( 1 ) ( 3 ) . . . ( 2k3 )
(1)k1 1.3.5 . . . (2k 3)
2
2
2
2 = 2
=
k!
k
2k k!
k1
k1
(1) (2k 2)!
(1) (2k 2)!
(1)k1 (2k 2)!
=
=
=
2.4.6 . . . (2k 2))2k k!
2k1 k!(k 1)2k k!
22k1 k!(k 1)!
So if |x| < 1,

(1)

k
2
1+x=
k=0 k x = 1 + k=0

(1)k1 (2k2)! k
x
22k1 k!(k1)!

Example 2 Find a power series representation of

1
(1x)m

where m is a positive integer.

Solution. We need to compute the Binomial coeicients for r = m (and will replace x by x)
( )
(m)(m 1)(m 2) . . . (m k + 1)
m
=
k
k!
(
)
k
(1) m(m + 1)(m + 2) . . . (m + k 1)
k m+k1
=
= (1)
k!
k
So

381

www.ck12.org

(
)

1
k m+k1
=
(1)
(x)k
k
(1 x)m k=0
(
)

k m+k1
(1)
=
(1)k xk
k
k=0
(
)

2 m+k1 k
=
(1) k
x
k
k=0
)
(

m+k1 k
=
x
k
k=0
Exercise
1. Find a power series representation of

1
1+x
1
(2x)2

at x = 0.

2. Find a power series representation of

at x = 0.
(
)2

3. Notice 1 + x + x2 = x + 34 + x + 21 . Find a power series representation of (1 + x + x2 ) at x = 21 .

In what interval is the equality true?

Choosing Centers
Taylor Series (indeed Taylor polynomials of lower degrees) often provide good approximation of functions.
However, the choice of center could determine
1. whether the intended value of x is inside the interval of convergence
2. rate of convergence, i.e. how many terms to take to achieve prescribed degree of accuracy
For frequently used functions, the rst choice may be the standard center (see the list at the end of this
section).
Example 1 Approximate ln 0 : 99
Solution. Since .99 is close to the center x = 1, we use the standard Taylor series for ln(1 x).
ln 0.99 = ln(1 0.01) 0.01

(0.01)2
0.01005
2

Then we may be able to deduce a useful Taylor Series centered close to the given x.
Example 2 Approximate sin(1.1) to 4 decimal places.
Since 1.1 is close to 3 , we would try to nd a Taylor Series of sin x at x0 = 3 . Let f (x) = sin x. Then

( )
( )
( )
f (x) = sin x, f (x) = cos x, f (x) = sin x, f (x) = cos x and f 3 = 23 , f 3 = 12 , f 3 =

( )
23 , f 3 = 12 .
This pattern repeats and limn Rn (x) = 0 can be checked as in the case x0 = 0. So the Taylor Series is

(1)n 3 (
)2n (1)n (
)2n+1
sin x =
x
+
x
3
3
2(2n)!
2(2n + 1)!
n=0
n=0
www.ck12.org

382

Hence

(1)n 3
(1)n
2n
sin(1.1)
(0.0528) +
(0.0528)2n+1
2(2n)!
2(2n
+
1)!
n=0
n=0
0.86481823 + 0.02638773 taking 2 terms from each sum
0.8912
We may also apply algebraic manipulation to standard Taylor Series.
Example 3 Approximate

1
1.92

to 4 decimal places.

Solution. There is standard Taylor Series:

1
n1 for |x| < 1 through term-by-term dierentiation of the series for
=
k=1 nx
(1x)2
at x = 0.9).

1
1
1x (1x)2

is inadequate

Since 1.9 is close to 2, we consider

( )n1

1
x
= 14 (11x )2 = 41
for |x| < 2.
k=1 n 2
(2x)2
2

So we take x = 0.1 and then

1
1 ( x )n1
1
=
n
. (1 + 2(0.05) + 3(0.05)2 + 4(0.05)3 ) 0.277.
2
4 n=1 2
4
1.9

Exercise
1. Approximate ln 0.9 to 4 decimal places.
2. Approximate sin(0.8) to 6 decimal places.
Hint: consider center
3. Approximate

1
93

to 6 decimal places.

Evaluating Non-Elementary Integrals

There are many simple-looking functions that have no explicit formula for their integral in the form of
elementary functions. We could write their in- tegrals as Taylor Series in their interval of convergence.
x
Example 1 Find a power series representation of ex dx
x

Solution. Since ex is not dened at x = 0, we apply the Taylor Series of e x at, say, x = 1 by writing
e x = e.e x1 with a change of variable u = x 1.

ex
e.eu
1 (u)n
=
= e.
x
1+u
1 + u n=0 n!
(
)
u2 u3 u4
2
3
= e(1 u + u u + . . .) 1 + u +
+
+
+ ...
2!
3!
2!
)
(
1
3
11
1
= e 1 + u2 u3 + u4 u5 + . . . for |u| < 1
2
3
8
30

so
ex

)
+ . . . where u = x 1

2
Example 2 Find the power series representation of sinxx dx
x

(
dx = e u + 16 u3

1 4
12 u

3 5
40 u

11 6
180 u

383

www.ck12.org

Solution. Direct substitution of x2 in the Maclaurin Series of sin x gives

(1)n
(1)n 4n+1
2 2n+1 and sin x2 =
sin x2 =
n=0 (2n+1)! (x )
n=0 (2n+1)! x
2
So

(1)n
sin x2
dx =
x4n+1 dx
x
(2n
+
1)!
n=0

(1)n
(1)n x4n+2
x4n+2
=
=
(2n + 1)! 4n + 2 n=0 (4n + 2) (2n + 1)!
n=0

Exercise
1. Find the power series representation (Maclaurin Series) of

e x dx and approximate

decimal places.

2. Find the power series representation (Maclaurin Series) of sin x2 dx.

ex dx to 6
2

Frequently Used Maclaurin Series

Some frequently used Maclaurin Series are listed below
n
1
x
1x = n=1

1
n
=
n=1 nx
(1x)2

xn
ex =
n=1
n!
xn
ln 1 x =
n=1
n,

n+1 xn ,
(1)
ln 1 + x =
n=1
n
n
2n+1
(1) x
sin x = n=1 (2n+1)!

(1)n x2n
cos x =
n=1 (2n)!

(1)n x2n+1
tan1 x =
n=1 (2n+1)

in(1, 1)
in(1, 1)
in(1, 1)
in(1, 1)
in(1, 1)
in(1, 1)
in(1, 1)
in(1, 1)

Texas Instruments Resources

In the CK-12 Texas Instruments Calculus FlexBook, there are graphing calculator activities
designed to supplement the objectives for some of the lessons in this chapter. See http:
//www.ck12.org/flexr/chapter/9733.

www.ck12.org

384