University of the Philippines Diliman

MATHEMATICS 21
Elementary Analysis I
Course Module

Institute of Mathematics
MATHEMATICS 21
Elementary Analysis I
Course Module

Institute of Mathematics
University of the Philippines Diliman
iv

2018
c by the Institute of Mathematics, University of the Philippines Diliman.
All rights reserved.
No copies can be made in part or in whole without prior written permission from the
Institute of Mathematics, University of the Philippines Diliman.

Mathematics 21 Module Writers and Editors:

Carlo Francisco Adajar

Michael Baysauli
Katrina Burdeos
Lawrence Fabrero

s
Alip Oropeza

ic
at
h em
at
M
of
e
ut
s tit
In
UP
Contents

1 Limits and Continuity 1

1.1 Limit of a Function: An Intuitive Approach . . . . . . . . . . . . . . . . . . . . . . . 1
1.1.1 An Intuitive Approach to Limits . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1.2 Evaluating Limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.1.3 Other Techniques in Evaluating Limits . . . . . . . . . . . . . . . . . . . . . . 6
1.1.4 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

ic s
1.2 One-Sided Limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

at
1.2.1 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

em
1.3 Limits Involving Infinity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
h
1.3.1 Infinite Limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
at
1.3.2 Limits at Infinity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
M

1.3.3 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
of

1.4 Limit of a Function: The Formal Definition . . . . . . . . . . . . . . . . . . . . . . . 30

e

1.4.1 The Formal Definition of Limits . . . . . . . . . . . . . . . . . . . . . . . . . 30

ut
tit

1.4.2 Proving Limits using the Definition . . . . . . . . . . . . . . . . . . . . . . . . 31

1.4.3 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
s
In

1.5 Continuity of Functions; The Intermediate Value Theorem . . . . . . . . . . . . . . . 37

UP

1.5.1 Continuity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
1.5.2 The Intermediate Value Theorem . . . . . . . . . . . . . . . . . . . . . . . . . 44
1.5.3 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
1.6 Trigonometric Functions: Limits and Continuity;
The Squeeze Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
1.6.1 The Squeeze Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
1.6.2 Continuity of Trigonometric Functions . . . . . . . . . . . . . . . . . . . . . . 53
1.6.3 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
1.7 New Classes of Functions: Limits and Continuity . . . . . . . . . . . . . . . . . . . . 57
1.7.1 Inverse Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
1.7.2 Exponential and Logarithmic Functions . . . . . . . . . . . . . . . . . . . . . 58
1.7.3 Inverse Circular Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
1.7.4 Hyperbolic Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
1.7.5 Inverse Hyperbolic Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
1.7.6 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

v
vi CONTENTS

2 Derivatives and Differentiation 75

2.1 Slopes, the Derivative, and Basic Differentiation Rules . . . . . . . . . . . . . . . . . 75
2.1.1 The Tangent Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
2.1.2 Definition of the Derivative . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
2.1.3 Differentiability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
2.1.4 Techniques of Differentiation . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
2.1.5 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
2.2 The Chain Rule, and more on Differentiability . . . . . . . . . . . . . . . . . . . . . 84
2.2.1 The Chain Rule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
2.2.2 Derivatives from the Left and from the Right . . . . . . . . . . . . . . . . . . 86
2.2.3 Differentiability and Continuity . . . . . . . . . . . . . . . . . . . . . . . . . . 87
2.2.4 Graphical Consequences of Differentiability and Non-differentiability . . . . . 89
2.2.5 Higher Order Derivatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
2.2.6 Implicit Differentiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
2.2.7 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
2.3 Derivatives of Exponential and Logarithmic Functions . . . . . . . . . . . . . . . . . 96

ic s
2.3.1 Derivatives of Logarithmic Functions . . . . . . . . . . . . . . . . . . . . . . . 96

at
2.3.2 Logarithmic Differentiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

em
2.3.3 Derivatives of Exponential Functions . . . . . . . . . . . . . . . . . .
h . . . . . 99
2.3.4 Derivative of f (x)g(x) , where f (x) > 0 . . . . . . . . . . . . . . . . . . . . . . 101
at
2.3.5 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
M

2.4 Derivatives of Other New Classes of Functions . . . . . . . . . . . . . . . . . . . . . 103

of

2.4.1 Derivatives of Inverse Circular Functions . . . . . . . . . . . . . . . . . . . . . 103

e

2.4.2 Derivatives of Hyperbolic Functions . . . . . . . . . . . . . . . . . . . . . . . 104

ut

2.4.3 Derivatives of Inverse Hyperbolic Functions . . . . . . . . . . . . . . . . . . . 105

tit

2.4.4 Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106

s
In

2.5 More Indeterminate Forms and L’Hôpital’s Rule . . . . . . . . . . . . . . . . . . . . 107

0 ∞
UP

2.5.1 Indeterminate Forms of Type and . . . . . . . . . . . . . . . . . . . . . 107

0 ∞
2.5.2 L’Hôpital’s Rule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
2.5.3 Indeterminate Forms of Type 0 · ∞ and ∞ − ∞ . . . . . . . . . . . . . . . . . 111
2.5.4 Indeterminate Forms of Type 1∞ , 00 and ∞0 . . . . . . . . . . . . . . . . . . 113
2.5.5 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
2.6 The Mean Value Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
2.6.1 Rolle’s Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
2.6.2 The Mean Value Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
2.6.3 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
2.7 Relative Extrema of a Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
2.7.1 Relative Extrema . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
2.7.2 Critical Numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
2.7.3 Increasing/Decreasing Functions . . . . . . . . . . . . . . . . . . . . . . . . . 124
2.7.4 The First Derivative Test for Relative Extrema . . . . . . . . . . . . . . . . . 125
2.7.5 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
2.8 Concavity and the Second Derivative Test . . . . . . . . . . . . . . . . . . . . . . . . 130
CONTENTS vii

2.8.1 Concavity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130

2.8.2 Point of Inflection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
2.8.3 The Second Derivative Test for Relative Extrema . . . . . . . . . . . . . . . . 133
2.8.4 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
2.9 Graph Sketching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
2.9.1 Graphing Polynomial Funtions . . . . . . . . . . . . . . . . . . . . . . . . . . 137
2.9.2 Review of Asymptotes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
2.9.3 Graphing Rational Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
2.9.4 The Graph of f from the Graph of f 0 . . . . . . . . . . . . . . . . . . . . . . 142
2.9.5 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144

3 Applications of Differentiation 147

3.1 Rectilinear Motion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
3.1.1 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
3.2 Rates of Change and Related Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
3.2.1 Rates of Change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152

s
3.2.2 Related Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154

ic
at
3.2.3 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157

em
3.3 Local Linear Approximation and Differentials . . . . . . . . . . . . . . . . . . . . . . 159
3.3.1 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
h . . . . . 163
at
3.4 Absolute Extrema of a Function on an Interval . . . . . . . . . . . . . . . . . . . . . 165
M

3.4.1 Absolute Extrema on Closed and Bounded Intervals . . . . . . . . . . . . . . 166

3.4.2 Absolute Extrema On Open Intervals . . . . . . . . . . . . . . . . . . . . . . 167
of

3.4.3 Optimization: Application of Absolute Extrema on Word Problems . . . . . . 169

e
ut

3.4.4 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175

tit

4 Integration and Its Applications 179

s
In

4.1 Antidifferentiation and Indefinite Integrals . . . . . . . . . . . . . . . . . . . . . . . . 179

UP

4.1.1 Antiderivatives or Indefinite Integrals . . . . . . . . . . . . . . . . . . . . . . 179

4.1.2 Integration by Substitution . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
4.1.3 Particular Antiderivatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
4.1.4 Rectilinear Motion Revisited . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
4.1.5 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
4.2 The Definite Integral . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
4.2.1 Area of a Plane Region: The Rectangle Method . . . . . . . . . . . . . . . . . 196
4.2.2 The Definite Integral . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200
4.2.3 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
4.3 The Fundamental Theorem of the Calculus . . . . . . . . . . . . . . . . . . . . . . . 206
4.3.1 First Fundamental Theorem of the Calculus . . . . . . . . . . . . . . . . . . . 206
4.3.2 The Second Fundamental Theorem of Calculus . . . . . . . . . . . . . . . . . 209
4.3.3 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
4.4 Generalization of the Area of a Plane Region . . . . . . . . . . . . . . . . . . . . . . 214
4.4.1 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
4.5 Arc Length of Plane Curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
viii CONTENTS

4.5.1 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236

4.6 Volumes of Solids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
4.6.1 Volumes of Solids of Revolution . . . . . . . . . . . . . . . . . . . . . . . . . . 237
4.6.2 Volume of Solids by Slicing . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
4.6.3 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259
4.7 Mean Value Theorem for Integrals . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262
4.7.1 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265

ic s
at
h em
at
M
of
e
ut
s tit
In
UP
Chapter 2

Derivatives and Differentiation

2.1 Slopes, the Derivative, and Basic Differentiation Rules

In this section, we extend the notion of a tangent line to a circle to other curves at a point. Recall

ics
from Euclidean Geometry that a line tangent to a circle is a line that intersects the circle at exactly

at
one point. However, in a more general sense, a line tangent to a curve may intersect the curve at

em
points other than the point of tangency.
h
at
M

2.1.1 The Tangent Line

of
e

`
ut

y = f (x)
tit

Given a function f (x), we want to define the equation of P

s

the tangent line ` at a point, say P (x0 , f (x0 )) on the graph

In

of y = f (x).
UP

Figure 2.1.1

To do this, we need to find the slope of this tangent line. Consider another point Q(x1 , f (x1 )) on
←→
the graph of y = f (x). Form the secant line P Q.

←→
Q Then P Q has slope
f (x1 )
y = f (x)
)
∆y f (x1 ) − f (x0 )
P m←→ = .
f (x0 ) | {z }
PQ x1 − x0
∆x
Letting ∆x = x1 − x0 , the above expression is
equivalent to
x0 x1
Figure 2.1.2 f (x0 + ∆x) − f (x0 )
m←→ =
PQ
.
∆x

75
76 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

`
y = f (x)
Now, imagine the point Q moving along the curve y = f (x)
P
toward P .

As Q approaches P , the secant line approaches an object in

the plane. This object coincides with the line `.

Figure 2.1.3

We define the slope m of the line ` to be the limit of the slope m← → of the secant line as Q gets
PQ
closer and closer to P and call ` the tangent line to the graph of f at P . We have
m = lim m←→.
PQ
Q→P

Observe that as Q → P , x1 → x0 and so ∆x = x1 − x0 → 0. Hence,

f (x0 + ∆x) − f (x0 )
m = lim .
∆x→0 ∆x

ic s
Definition 2.1.1. If the function f is defined on an open interval containing x = x0 , then the

at
tangent line to the graph of f at the point P (x0 , f (x0 )) is the line
1. passing through P whose slope is given by h em
at
f (x0 + ∆x) − f (x0 )
m = lim ,
M

∆x→0 ∆x
of

provided that this limit exists.

e

2. with equation x = x0 if
ut
tit

f (x0 + ∆x) − f (x0 )

lim = +∞ or − ∞ and
s

∆x→0− ∆x
In

f (x0 + ∆x) − f (x0 )

lim = +∞ or − ∞.
∆x
UP

∆x→0+

Otherwise, there is no tangent line to the graph of f at P .

Remark 2.1.2.
1. The slope of the tangent line to the graph of f at P gives us an idea of the “flatness” or
“steepness” of the graph of f at P and whether the graph of f rises or falls at P . See Figure
2.1.4.

2. The tangent line to the graph of a function may intersect the graph at points other than the
point of tangency as shown in Figure 2.1.5.

P
` TL

NL

0
Figure 2.1.5 Figure 2.1.6
0 Figure 2.1.4
2.1. SLOPES, THE DERIVATIVE, AND BASIC DIFFERENTIATION RULES 77

Definition 2.1.3. The normal line to the graph of f at the point P is the line perpendicular to
the tangent line at P . (See Figure 2.1.6.)
1
Example 2.1.4. Give equations of the tangent line and the normal line to the graph of f (x) =
x
at x = 1.

Solution.
Let mT L denote the slope of the tangent line to the graph of f (x) at x = 1. We have
f (1 + ∆x) − f (1)
mT L = lim
∆x→0 ∆x
1
− 1
= lim 1+∆x
∆x→0 ∆x
1 − (1 + ∆x)
= lim
∆x→0 ∆x(1 + ∆x)  
−∆x 0
= lim
∆x→0 ∆x(1 + ∆x) 0
−1
= lim

s
∆x→0 1 + ∆x

ic
= −1

at
em
Thus, the equation of the tangent line to the graph of f (x) = x1 at (1, 1) is y − 1 = −(x − 1). Now,
h
since mT L = −1, the slope of the normal line is 1 and hence, the equation of the normal line to the
at
graph of f (x) = x1 at x = 1 is y − 1 = (x − 1), or simply y = x.
M

Example 2.1.5. Find the slope of the tangent line to the graph of g(x) = x2 + 2 at x = 1 and at
of

x = 2.
e
ut

Solution.
tit

At x = 1 the slope of the tangent line to the graph of g(x) is

s
In

g(1 + ∆x) − g(1)

mT L = lim
∆x
UP

∆x→0
[(1 + ∆x)2 + 2] − (12 + 2)
= lim
∆x→0 ∆x
[1 + 2∆x + (∆x)2 + 2] − 3
= lim
∆x→0 ∆x
2∆x + (∆x)2
= lim
∆x→0 ∆x
∆x(2 + ∆x)
= lim
∆x→0 ∆x
= lim (2 + ∆x)
∆x→0
= 2.

On the other hand, at x = 2 the slope of the tangent line to the graph of g(x) is
g(2 + ∆x) − g(2)
mT L = lim
∆x→0 ∆x
[(2 + ∆x)2 + 2] − (22 + 2)
= lim
∆x→0 ∆x
78 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

[4 + 4∆x + (∆x)2 + 2] − 6
= lim
∆x→0 ∆x
4∆x + (∆x)2
= lim
∆x→0 ∆x
= lim (4 + ∆x)
∆x→0
= 4.

In fact, we can compute for the slope of the tangent line to the graph of g(x) for any value of x.
We have
g(x + ∆x) − g(x)
mT L = lim
∆x→0 ∆x
[(x + ∆x)2 + 2] − (x2 + 2)
= lim
∆x→0 ∆x
[x + 2x∆x + (∆x)2 + 2] − (x2 + 2)
2
= lim
∆x→0 ∆x

s
2x∆x + (∆x) 2

ic
= lim

at
∆x→0 ∆x

em
= lim (2x + ∆x)
∆x→0 h
= 2x.
at
M

Using this formula, we see that the slope at x = 1 is 2 · 1 = 2 while the slope at x = 2 is 2 · 2 = 4,
of

which agree with the preceding calculations.

e
ut

2.1.2 Definition of the Derivative

tit

In the previous example, we obtained a function that gives the slope of the tangent line to the
s
In

graph of a function g(x) at any value of x. This function shall be called the derivative of g(x).
UP

Definition 2.1.6. The derivative of a function f (x), denoted f 0 (x), is the function

f (x + ∆x) − f (x)
f 0 (x) = lim .
∆x→0 ∆x
It is defined at all points x in the domain of f where the limit exists.

Remark 2.1.7.

1. Hence, from the definition, dom f 0 ⊆ dom f since there may be points x0 ∈ dom f at which
f 0 (x0 ) does not exist.

2. The definition also tells us that f 0 (x0 ) is the slope of the tangent line to the graph of the function
at the point P (x0 , f (x0 )).

3. To get the derivative of f at x = x0 , we use

f (x0 + ∆x) − f (x0 )

f 0 (x0 ) = lim .
∆x→0 ∆x
2.1. SLOPES, THE DERIVATIVE, AND BASIC DIFFERENTIATION RULES 79

Alternatively, by setting ∆x = x − x0 , we have

f (x) − f (x0 )
f 0 (x0 ) = lim .
x→x0 x − x0

dy d
4. Other notations: y 0 if y = f (x), , [f (x)] , Dx [f (x)].
dx dx
5. The process of computing the derivative is called differentiation.

Example 2.1.8. If g(x) = x2 + 2, then, following the computations in the previous example, we
have g 0 (x) = 2x.
√
Example 2.1.9. Find the derivative of f (x) = x.

Solution.
Using the definition of the derivative, we have
√ √ √ √ √ √
0 x + ∆x − x x + ∆x − x x + ∆x + x

s
f (x) = lim = lim ·√ √

ic
∆x→0 ∆x ∆x→0 ∆x x + ∆x + x

at
(x + ∆x) − x 1 1
√ √ = lim √ √ = √

em
= lim
∆x→0 ∆x( x + ∆x + x) ∆x→0 x + ∆x + x 2 x
h
at
1
Thus, f 0 (x) = √ . Note that dom f = [0, +∞) while dom f 0 = (0, +∞).
M

2 x
of

2.1.3 Differentiability
e
ut

Definition 2.1.10.
tit

1. A function f is said to be differentiable at x = x0 if the derivative f 0 (x0 ) of f at x = x0 exists.

s
In

That is, f is differentiable at x0 if x0 is in the domain of f 0 .

UP

2. A function f is differentiable on (a, b) if f is differentiable at every real number in (a, b).

3. A function f is differentiable everywhere if it is differentiable at every real number.

√ 1 √
Example 2.1.11. From the preceding example, Dx [ x ] = √ and so f (x) = x is differentiable
2 x
at any positive real number x. Also, f is not differentiable at x = 0.

Example 2.1.12. It will be shown later that if f (x) is a polynomial, a rational, or a trigonometric
function, then f (x) is differentiable on its domain.

2.1.4 Techniques of Differentiation

The previous section introduced a method of differentiation using Definition 2.1.6. The drawback
of such a method is that the process involved is tedious, even for the differentiation of relatively
simple functions such as in example 2.1.2. The theorems introduced in this chapter will greatly
simplify the process of finding the derivative of a function, by providing formulas for the derivatives
of algebraic and transcendental functions.
80 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

By the end of this section, the student will be able to:

• compute the derivatives of algebraic functions using power rule, sum/difference

rule, product rule, quotient rule;

• differentiate functions involving trigonometric functions;

• differentiate functions involving logarithmic functions;

• find the derivative using logarithmic differentiation;

• differentiate functions of the form f (x)g(x) ; and

• differentiate functions involving exponential functions.

s
ic
at
Differentiation Rules

h em
The following theorem introduces basic rules in finding the derivative of a function.
at
M
of

Theorem 2.1.13. Let f and g be functions and c ∈ R.

e

R, then f 0(x) = 0.
ut

1. If f (x) = c ∈
tit

Q, then f 0(x) = nxn−1.

s

2. (Power Rule) If f (x) = xn , where n ∈

In

3. If f (x) = c · g(x), then f 0 (x) = c · g 0 (x) if g 0 (x) exists.

UP

4. (Sum Rule) If h(x) = f (x) ± g(x), then h0 (x) = f 0 (x) ± g 0 (x), provided both f 0 (x) and
g 0 (x) exist.

5. (Product Rule) If h(x) = f (x)g(x), then h0 (x) = f 0 (x)g(x)+f (x)g 0 (x), provided f 0 (x)
and g 0 (x) both exist.
f (x) g(x)f 0 (x) − f (x)g 0 (x)
6. (Quotient Rule) If h(x) = , where g(x) 6= 0, then h0 (x) = ,
g(x) [g(x)]2
0 0
provided f (x) and g (x) both exist.

Proof. We show the proof of the first two statements only.

c−c 0
1. Dx [c] = lim = lim = lim 0 = 0.
∆x→0 ∆x ∆x→0 ∆x ∆x→0
2.1. SLOPES, THE DERIVATIVE, AND BASIC DIFFERENTIATION RULES 81

2. We make use of the Binomial Theorem1 in the proof of statement 2.

(x + ∆x)n − xn
Dx [xn ] = lim
∆x→0
 ∆x 
n(n−1) n−2
xn + nxn−1 ∆x + 2 x ∆x2 + . . . + nx∆xn−1 + ∆xn − xn
= lim
∆x→0
 ∆x 
n−1 n(n − 1) n−2 n−2 n−1
= lim nx + x ∆x + . . . + nx∆x + ∆x
∆x→0 2
= nxn−1

Example 2.1.14.

1. Dx [x5 ] = 5x5−1 = 5x4

ic s
2. Dx [(2x)5 ] = Dx [32x5 ] = 32 · 5x4 = 160x4 (Note that the derivative is not 5(2x)4 which is 80x4 .)

at
em
d √ d d 4√ d
3. [2x4 − 45 x + 7] = [2x4 ] − [ x] + [7] = 8x3 − 4
· 1
√ + 0 = 8x3 − 2
√ .
dx dx dx 5 dx h 5 2 x 5 x
at
4. If y = (x − 3)(2x2 − 3), then y = 2x3 − 6x2 − 3x + 9 and so y 0 = 6x2 − 12x − 3.
M
of

5. Alternatively, we may also use the Product Rule. Thus,

Dx [(x − 3)(2x2 − 3)] = (1)(2x2 − 3) + (x − 3)(4x) = 2x2 − 3 + 4x2 − 12x = 6x2 − 12x − 3.
e
ut

d √
tit

4
6. [(3x5 + 6x 3 − 2)(4x3 − 5 3 x)]
dx
s

1 √ 4 2
In

= (15x4 + 8x 3 )(4x3 − 5 3 x) + (3x5 + 6x 3 − 2)(12x2 − 5 · 31 x− 3 ).

UP

2x3 − x−3 + 4 (3x − 5)(6x2 + 3x−4 ) − (2x3 − x−3 + 4)(3)

7. If f (x) = , then f 0 (x) = .
3x − 5 (3x − 5)2
√  (2x3 − 4√x − 3x)(8x3 − √1 ) − (2x4 − √x + 2)(6x2 − √2 − 3)
2x4 − x + 2

d 2 x x
8. √ = √ .
dx 2x3 − 4 x − 3x (2x3 − 4 x − 3x)2

Remark 2.1.15.

1. Using items 1 to 4 of Theorem 2.1.13, one can show that the derivative of a polynomial function is
also a polynomial function. This means that a polynomial function is differentiable everywhere.

2. From the first item and the Quotient Rule, one can deduce that a rational function is differen-
tiable on its domain.
!
n n
1 n
xn−k y k
P
Binomial Theorem: If n is a positive integer, then (x + y) =
k=0 k
82 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

Derivatives of Trigonometric Functions

At this point, we know how to differentiate polynomial functions, rational functions, and functions
involving radicals. We now present formulas for the derivatives of trigonometric functions.

Theorem 2.1.16.

1. Dx [sin x] = cos x 4. Dx [cot x] = − csc2 x

2. Dx [cos x] = − sin x 5. Dx [sec x] = sec x tan x

3. Dx [tan x] = sec2 x 6. Dx [csc x] = − csc x cot x

Proof. We will prove statement 1 only. Statement 2 can be proved similarly, while items 3–6 can
be proved using items 1 and 2, and the product and quotient rules.
sin(x + ∆x) − sin x
Dx [sin x] = lim
∆x→0 ∆x

ic s
sin x cos(∆x) + cos x sin(∆x) − sin x
= lim

at
∆x→0 ∆x

em
 
sin(∆x) 1 − cos(∆x)
= lim cos x · − sin x ·
∆x ∆x
∆x→0
h
at
= (cos x)(1) − (sin x)(0)
M

= cos x
of
e
ut

Remark 2.1.17.
s tit

1. The formulas in the previous theorems consider trigonometric functions as real–valued functions.
In

Thus, whenever these formulas are applied to problems where trigonometric functions are viewed
UP

as functions on angles, the measure of an angle must be in radians.

2. Observe that the derivative of a trigonometric function is either another trigonometric function
or a product of trigonometric functions. That means that a trigonometric function is differen-
tiable where its derivative is defined. Moreover, observe that the domains of a trigonometric
function and its derivative are the same. Hence, a trigonometric function is differentiable on its
domain.

Example 2.1.18.
d
1. [3 sin x − 7 cos x] = 3 cos x + 7 sin x
dx
2. Dx [sec x csc x] = (sec x tan x)(csc x) + (sec x)(− csc x cot x)
cot x − x (1 + tan x)(− csc2 x − 1) − (cot x − x)(sec2 x)
3. If f (x) = , then f 0 (x) =
1 + tan x (1 + tan x)2
4. Dx [sin(2x)] = Dx [2 sin x cos x] = 2·(cos x·cos x+sin x·(− sin x)) = 2(cos2 x−sin2 x) = 2 cos(2x)
2.1. SLOPES, THE DERIVATIVE, AND BASIC DIFFERENTIATION RULES 83

2.1.5 Exercises
Exercises for Discussion
dy
A. Find . There is no need to simplify.
dx

1. y = 2x4 − 3x2 + x − 1 4 − x2
√ 6. y = √
1 8 3 x + tan x
2. y = 6 3 x − 2 + √
x x
√
3. y = x3 cos x cos x + π
7. y =
5 2x − x33
4. y = − sec x csc x
x
 
√ √
2 2 x x
5. y = x − 2 ( 3 x − cos x) 8. y = 5
x x + sin x

B. Miscellaneous Exercises.

1. Determine the equation of the normal line to the graph of g(x) = 2 sin x + tan x at the

ic s
point where x = π3 .

at
2. Find all the points on the graph of y = (x − 2)2 at which the tangent line is perpendicular
to the line with equation 2x − y + 2 = 0. h em
at
M

Supplementary Exercises
dy
of

A. Find . There is no need to simplify.

dx
e
ut

√
1. y = 5 sin x − 2 cot x
tit

2x3 csc x + x − 2
5. y =
2. y = (2x2 + 5x − 2)(3x − 7) (sin x + 1)(x cos x − 1)
s
In

2x + 1 √
3. y = 6. y = (x2 − 2x + 2)(sin x − x)(2 + cos x)
UP

csc x
√
 
x sin x − x 5 1
x3 − x4 + 5x5 (sin x + csc x)


7. y = 2
x − 3
4. y = 5
x cot x − 5 x
7 3
B. Given f (5) = 5 , f 0 (5) = − , g(5) = , and g 0 (5) = −8 , determine
2 2

1. (f − g)0 (5)
 f 0
3. (5)
g
 g 0
2. (f · g)0 (5) 4. (5)
f

C. Miscellaneous Exercises.

1. Find the equation of the tangent line to the graph of f (x) = 2x3 + 1 at x = −1.
2. Determine the values of a and b so that the line with equation 2x + y = b is tangent to the
graph of y = ax2 when x = 2.
84 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

2.2 The Chain Rule, and more on Differentiability

The previous section introduced formulas for differentiating a large scope of functions, ranging
from algebraic functions to inverse hyperbolic functions. But there are many types of functions
√
which do not neatly fall into one category. For instance, how can the function f (x) = x2 − 1 be
differentiated? The formulas provided in the previous section do not cover function compositions.
This section provides techniques for differentiation of compositions of functions. We also tackle
the geometric implications of differentiability and non-differentiability, and discuss the notion of a
higher-order derivative.

By the end of this section, the student will be able to

• differentiate using the chain rule;

• differentiate implicitly defined functions of one variable;

• discuss the differentiability and continuity of a function; and

ics
• find higher order derivatives of a function.

at
h em
at
2.2.1 The Chain Rule
M

Recall that in the previous section, to differentiate sin 2x, an identity was used to write it in terms
of

of sin x and cos x. The following theorem allows us to differentiate a given function, a composite
e

function in particular, without having to write it as a product or quotient of basic functions.

ut
s tit

Theorem 2.2.1 (Chain Rule). If the function g is differentiable at x = x0 and the function f
In

is differentiable at g(x0 ), then (f ◦ g)(x) is differentiable at x = x0 and

UP

(f ◦ g)0 (x0 ) = f 0 (g(x0 )) · g 0 (x0 ).

Remark 2.2.2. The chain rule can also be stated in the following manner:
dy dy du
If y = f (u) and u = g(x) , then = · or Dx [f (u)] = f 0 (u)Dx [u] .
dx du dx
Example 2.2.3.

1. Find Dx [(2x)5 ].

Solution.
Note that (2x)5 = (f ◦ g)(x) where f (x) = x5 and g(x) = 2x. We have f 0 (x) = 5x4 and
g 0 (x) = 2. Using chain the rule,

Dx [(2x)5 ] = f 0 (g(x)) · g 0 (x) = 5[g(x)]4 · 2 = 10(2x)4 .

√
2. Find h0 (x) if h(x) = x2 − 1.
2.2. THE CHAIN RULE, AND MORE ON DIFFERENTIABILITY 85

Solution.
√ √ 1
Note that x2 − 1 = (f ◦ g)(x) where f (x) = x and g(x) = x2 − 1. We have that f 0 (x) = √
2 x
and g 0 (x) = 2x. Using the chain rule,

1 2x
h0 (x) = f 0 (g(x)) · g 0 (x) = √ · 2x = √ .
2 x2 − 1 2 x2 − 1

3. Find Dx [sin(2x)].

Solution.
We have sin(2x) = (f ◦ g)(x) where f (x) = sin x and g(x) = 2x. So, f 0 (x) = cos x and g 0 (x) = 2
and by the chain rule,

Dx [sin(2x)] = f 0 (g(x)) · g 0 (x) = (cos[g(x)]) · (2) = 2 cos(2x).

4. Find h0 (x) if h(x) = 2 cos(x2 ).

ic s
Solution.

at
em
Note that h(x) = (f ◦ g)(x) where f (x) = 2 cos x and g(x) = x2 . Now, f 0 (x) = 2(− sin x) =
−2 sin x and g 0 (x) = 2x. Using chain the rule,
h
at
M

h0 (x) = f 0 (g(x)) · g 0 (x) = (−2 sin g(x)) · (2x) = −4x sin(x2 ).

of

dy √
5. Find if y = x3 − tan x.
e

dx
ut
tit

Solution.
√
s

u and u = x3 − tan x. Thus,

In

We let y =
UP

dy dy du 1 3x2 − sec2 x
= · = √ · (3x2 − sec2 x) = √ .
dx du dx 2 u 2 x3 − tan x

Remark 2.2.4.

1. The chain rule can also be extended to a finite composition of functions. To illustrate,

(f1 ◦ f2 ◦ f3 ◦ f4 )0 (x) = f10 ((f2 ◦ f3 ◦ f4 )(x)) · f20 ((f3 ◦ f4 )(x)) · f30 (f4 (x)) · f40 (x)

2. When computing derivatives using chain rule, we don’t actually write out the functions f and
g as in the previous examples, but we bear them in mind. It may help to keep in mind:
“The derivative of f (g(x)) is the derivative of the outside function evaluated at the inside func-
tion times the derivative of the inside function.“

Example 2.2.5.

1. Dx [(3x2 − x − 6)27 ] = 27(3x2 − x − 6)26 · Dx [3x2 − x − 6] = 27(3x2 − x − 6)26 · (6x − 1)

86 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

2. If y = (2x − 1)100 (2 − x)200 then

dy
= 100(2x − 1)99 · (2) · (2 − x)200 + (2x − 1)100 · 200(2 − x)199 · (−1)
dx
= (2x − 1)99 (2 − x)199 [200(2 − x) − 200(2x − 1)]
= (2x − 1)99 (2 − x)199 [400 − 200x − 400x + 200)]
= (2x − 1)99 (2 − x)199 (600 − 600x)
= 600(1 − x)(2x − 1)99 (2 − x)199

√ 5 3 5
 
− · 2 + 2)3 · (5x 2 ) − (2x 2 + 2)4 · √1 · 2 + csc x cot x
" 5
#
(2x +
2 2)4 ( 2x csc x) 4(2x 2 2x
3. Dx √ = √
2x − csc x ( 2x − csc x) 2

d d
4. [sec(cos x)] = sec(cos x) tan(cos x) · [cos x] = sec(cos x) tan(cos x) · (− sin x)
dx dx
3 3 3 3 1
5. Dx [cos2 (cot 4x 2 )] = 2 cos(cot 4x 2 ) · (− sin(cot 4x 2 )) · (− csc2 4x 2 ) · (6x 2 )

ic s
2.2.2 Derivatives from the Left and from the Right

at
em
Definition 2.2.6. Let the function f (x) be defined at x = x0 .
h
1. The derivative from the left of f (x) at x = x0 , denoted by f−0 (x0 ), is given by
at
M

f (x) − f (x0 )
f−0 (x0 ) = lim .
of

x→x−
0
x − x0
e
ut

2. The derivative from the right of f (x) at x = x0 , denoted by f+0 (x0 ), is given by
tit

f (x) − f (x0 )
s

f+0 (x0 ) = lim

In

.
x→x+
0
x − x0
UP

Remark 2.2.7.

1. In Definition 2.2.6, it is necessary that the function f is defined at x0 . Otherwise, the limit
expressions do not make sense.

2. The derivative from the left [right] is also referred to as the left-hand derivative [right-hand
derivative], or simply left derivative [right derivative].

3. The function f is differentiable at x = x0 if and only if f−0 (x0 ) and f+0 (x0 ) exist and

f−0 (x0 ) = f+0 (x0 ) = f 0 (x0 ).

Example 2.2.8. Determine if f (x) = |x| is differentiable at x = 0.

Solution.
We compute the derivatives from the left and from the right at x = 0.
2.2. THE CHAIN RULE, AND MORE ON DIFFERENTIABILITY 87

f (x) − f (0) f (x) − f (0)

f−0 (0) = lim f+0 (0) = lim
x→0− x x→0+ x
(−x) − 0 x−0
= lim = lim
x→0− x x→0+ x
= −1 = 1

Since f−0 (0) 6= f+0 (0), the function f (x) is not differentiable at x = 0.
(
1 2
Example 2.2.9. Determine if f (x) = 2 x +√24 , x < 4
is differentiable at x = 4.
16 x , x ≥ 4

Solution.
We compute the derivatives from the left and from the right at x = 4.

f (x) − f (4) f (x) − f (4)

f−0 (4) = lim f+0 (4) = lim
x→4− x−4 x→4+ √x − 4
( 12 x2 + 24) − 32 16 x − 32

s
= lim = lim

ic
x→4− x−4 x→4+ x√− 4

at
1 2
2 x − 8 16( x − 2)
= lim = lim

em
x→4− x−4 x→4+ x−4
1 (x + 4)(x − 4) h x−4
= lim = lim 16 √
at
x→4− 2 x−4 x→4+ (x − 4) x + 2
1 16
M

= lim (x + 4) = lim √
x→4− 2 x→4+ x+2
of

= 4 = 4
e

Since f−0 (4) = f+0 (4) = 4, the function f (x) is differentiable at x = 4. Moreover, f 0 (4) = 4.
ut
tit

2.2.3 Differentiability and Continuity

s
In

In applications that we will tackle later, functions that are used to model quantities and relationships
UP

are assumed to be continuous and/or differentiable on certain intervals. The following presents a
link between continuity and differentiability.

Theorem 2.2.10. If f is differentiable at x = x0 , then f is continuous at x = x0 .

Remark 2.2.11.

1. If f is discontinuous at x = x0 , then f is not differentiable at x = x0 .

2. If f is continuous at x = x0 , it does not mean that f is differentiable at x = x0 .

3. If f is not differentiable at x = x0 , it does not mean that f is not continuous at x = x0 .

Example 2.2.12. The function f (x) = |x| is continuous at x = 0 but is not differentiable at x = 0.

We present a theorem that is especially useful in determining whether a piecewise-defined function

f is differentiable at x = x0 .
88 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

Theorem 2.2.13.

1. If f is continuous at x = x0 from the left and lim f 0 (x) exists, then

x→x−
0

f−0 (x0 ) = lim f 0 (x).

x→x−
0

2. If f is continuous at x = x0 from the right and lim f 0 (x) exists, then

x→x+
0

f+0 (x0 ) = lim f 0 (x).

x→x+
0

Example 2.2.14. Determine if the following function is differentiable at x = −1:

(
x2 , x < −1
f (x) =

s
−1 − 2x , x ≥ −1

ic
at
em
Solution.
First, we check the continuity of f (x) at x = −1. Note that f (−1) = 1 and
h lim f (x) =
x→−1−
at
lim f (x) = 1, and thus, f (x) is continuous at x = −1. Next, we determine the derivative from
M

x→−1+
the left and derivative from the right at x = −1. We obtain
of

(
e

2x , x < −1
ut

f 0 (x) =
−2 , x > −1
s tit

and thus, f−0 (−1) = lim f 0 (x) = 2(−1) = −2, while f+0 (−1) = lim f 0 (x) = −2. Finally, since
In

x→−1− x→−1+
f−0 (−1) = f+0 (−1), f (x) is differentiable at x = −1.
UP

Example 2.2.15. Determine if the following function is differentiable at x = 1:

(
x2 + x + 2 , x ≤ 1
g(x) =
3x , x > 1

Solution.
It can be shown that g is discontinuous at x = 1. However, note that
(
2x + 1 , x < 1
g 0 (x) = ,
3 , x>1

and thus, lim g 0 (x) = 3 and lim g 0 (x) = 3. Although lim g 0 (x) = lim g 0 (x), we cannot
x→1− x→1+ x→1− x→1+
conclude that the function is differentiable at x = 1. (Why?) In fact, g is not differentiable at
x = 1. (Why?)
2.2. THE CHAIN RULE, AND MORE ON DIFFERENTIABILITY 89

2.2.4 Graphical Consequences of Differentiability and Non-differentiability

Geometrically, if f is differentiable at x = x0 , then the graph of f has a non-vertical tangent line
at x = x0 . The following remark gives the most commonly encountered circumstances for which f
fails to be differentiable at a value x = x0 .

Remark 2.2.16. A function f is not differentiable at x = x0 if one of the following is true:

1. f is discontinuous at x = x0 (see Figure 2.2.1)

2. the graph of f has a vertical tangent line at x = x0 (see Figure 2.2.2)

3. the graph of f has no well–defined tangent line at x = x0 , i.e., the graph of f has a corner, edge
or cusp at x = x0 (see Figure 2.2.3).

s
ic
at
em
x0 x0 x0
Figure 2.2.1 Figure 2.2.2
h Figure 2.2.3
at
M

2.2.5 Higher Order Derivatives

of

If the derivative f 0 of a function f is itself differentiable, then the derivative of f 0 is called the
e

second derivative of f and is denoted f 00 . We can continue to obtain the third derivative, f 000 ,
ut

the fourth derivative, f (4) , and even higher derivatives of f as long as we have differentiability.
stit

Definition 2.2.17. The n-th derivative of the function f , denoted by f (n) , is the derivative
In

of the (n − 1)-th derivative of f , that is,

UP

f (n−1) (x + ∆x) − f (n−1) (x)

f (n) (x) = lim
∆x→0 ∆x
Remarks.

1. The n in f (n) is called the order of the derivative.

2. The derivative of a function f is sometimes called the first derivative of f .

3. The function f is sometimes written as f (0) (x).

dn y dn
4. Other notations: Dx n [f (x)] , , [f (x)] , y (n)
dxn dxn
Example 2.2.18.

1. Find f (n) (x) for all n ∈ N where f (x) = x6 − x4 − 3x3 + 2x2 − 4.

Solution.
90 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

We differentiate repeatedly and obtain

f 0 (x) = 6x5 − 4x3 − 9x2 + 4x,

f 00 (x) = 30x4 − 12x2 − 18x + 4,
f 000 (x) = 120x3 − 24x − 18,
f (4) (x) = 360x2 − 24,
f (5) (x) = 720x,
f (6) (x) = 720,
f (n) (x) = 0 for all n ≥ 7.

√
2. Find f (4) (x) if f (x) = 2x − 3.

Solution.
We differentiate repeatedly and obtain
1 1 1
f 0 (x) = (2x − 3)− 2 · (2) = (2x − 3)− 2

s
2

ic
1 3 3
f 00 (x) = − (2x − 3)− 2 · (2) = − (2x − 2)− 2

at
2
3

em
5 5
000
f (x) = (2x − 3)− 2 · (2) = 3 (2x − 2)− 2
2 h
15 7 7
f (4) (x) = − (2x − 3)− 2 · (2) = −15 (2x − 2)− 2
at
2
M

2.2.6 Implicit Differentiation

of
e

Equations in two variables are used to define a function explicitly or implicitly. The equation
ut

y = x2 − 1 defines the function f (x) = x2 − 1 explicitly. On the other hand, the equation y 2 = x + 1
tit

defines two functions of x implicitly, namely:

s
In

√ √
f1 (x) = x + 1 and f2 (x) = − x + 1.
UP

Suppose y = f (x) is a function defined implicitly by the equation x3 − 2xy − 3y − 6 = 0, how do

dy
we find its derivative ?
dx
Solution.
x3 − 6
From x3 − 2xy − 3y − 6 = 0, we can define y explicitly in terms of x as y = . Thus,
2x + 3
dy (2x + 3)(3x2 ) − (x3 − 6)(2) 4x3 + 9x2 + 12
= = .
dx (2x + 3)2 (2x + 3)2

However, suppose a function is defined implictly by a function for which y cannot be isolated easily,
dy
such as x4 y 3 − 7xy = 7 or tan(x2 − 2xy) = y. How do we find ?
dx
dy
To obtain without solving for y explicitly in terms of x, we use the method called implicit
dx
dy
differentiation. To find using implicit differentiation, we
dx
1. think of the variable y as a differentiable function of the variable x,
2.2. THE CHAIN RULE, AND MORE ON DIFFERENTIABILITY 91

2. differentiate both sides of the equation, using the chain rule where necessary, and
dy
3. solve for .
dx
dy
Example 2.2.19. Find .
dx
1. x3 − 2xy − 3y − 6 = 0

Solution.
Dx [x3 − 2xy − 3y − 6] = Dx [0]
 dy  dy
3x2 − 2 1 · y + x · −3· =0
dx dx
dy dy
3x2 − 2y − 2x · −3· =0
dx dx
dy dy
3x2 − 2y = 2x · +3·
dx dx
dy
(2x + 3) = 3x2 − 2y
dx

ic s
dy 3x2 − 2y

at
=
dx 2x + 3

em
x3 −6
3x2 − 2 · 2x+3 2x + 3
= ·
h
2x + 3 2x + 3
at
3x · (2x + 3) − 2(x3 − 6)
2
M

=
(2x + 3)2
of

4x3 + 9x2 + 12
=
e

(2x + 3)2
ut
tit

2. x4 y 3 − 7xy = 7
s
In

Solution.
UP

Dx [x4 y 3 − 7xy] = Dx [7]

dy  dy 
4x3 y 3 + 3x4 y 2 · −7 y+x· =0
dx dx
dy −4x3 y 3 + 7y
=
dx 3x4 y 2 − 7x

3. tan(x2 − 2xy) = y

Solution.

Dx [tan(x2 − 2xy)] = Dx [y]

h  dy i dy
sec2 (x2 − 2xy) · 2x − 2 y + x · =
dx dx
dy 2x sec2 (x2 − 2xy) − 2y sec2 (x2 − 2xy)
=
dx 1 + 2x sec2 (x2 − 2xy)
92 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

d2 y
Example 2.2.20. Determine if xy 2 = y − 2.
dx2
Solution.
dy
We use implicit differentiation to get .
dx

Dx [xy 2 ] = Dx [y − 2]
dy dy
y 2 + x · 2y =
dx dx
dy y2
=
dx 1 − 2xy
Thus,
dy dy
d2 y (1 − 2xy)(2y dx ) − (y 2 )[−2(y + x dx )]
2
= 2
dx (1 − 2xy)
h i h  i
y2 y2
(1 − 2xy) 2y · 1−2xy − (y 2 ) −2 y + x · 1−2xy
= .

s
(1 − 2xy)2

ic
at
h em
at
M
of
e
ut
s tit
In
UP
2.2. THE CHAIN RULE, AND MORE ON DIFFERENTIABILITY 93

2.2.7 Exercises
Exercises for Discussion

A. The Chain Rule.

dy
1. Find . There is no need to simplify.
dx

a. y = cos 3x d. y = cot4 (5x) 1

2 √ sin x −
b. y = 6x 3h(x2 − x)5 i  
f. y = √+ 1 x2
4 5 cos x
c. y = csc 2x3 + 1

4
e. y = sec cot 3x − 2π
tan 2x

B. Differentiability.

1. Determine if ( 2
3x 3 − 1 , x ≤ 1
f (x) = 3
x −x+2 , x>1
is differentiable at x = 1.

ic s
2. Determine if

at
(
x2 − 4 , x < 3

em
g(x) = √
x−2 , x≥3
h
at
is differentiable at x = 3.
M

3. Determine if
(
of

(x − 1)2 , x ≤ 1
h(x) = √
x−1 , x>1
e
ut

is differentiable at x = 1.
s tit

C. Implicit Differentiation.
In

dy
UP

1. Use implicit differentiation to find . There is need to simplify.

dx
√
a. cos(x2 + y 2 ) = x − y 2 2
b. x3 y 2 = x 3 + y 3 + 1
ds √
2. Find if (3s + t)4 = 3 s − cos t.
dt
D. Higher Order Derivatives.

 √ 
N
1. Find f (n) (x) for all x ∈ if f (x) = 6x5 − 5x4 − 4x3 + 3x2 − 2x + 1.
2. Evaluate Dx2 x 4 − x2

E. Do as indicated.

1. If g(5) = 3, g 0 (5) = 4, f (3) = 5 and f 0 (3) = 6, find (f ◦ g)0 (5) and (g ◦ f )0 (3).
2. Determine the point/s on the graph of xy = (1 − x − y)2 where the tangent line/s is/are
parallel to the x−axis.
3. Determine Dx103 (cos 2x).
4. Find y 00 at the point with coordinate (2, 1) if 2x2 y − 4y 3 = 4.
94 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

Supplementary Exercises

A. The Chain Rule.

dy
1. Find . There is no need to simplify.
dx
√ √
e. y = cos3 [cot (csc x)]
r
a. y = 9 − x2 1 + cot x
3
c. y =
x cos x (1 − x2 )5 cos x
 
1
b. y = tan 3
d. y = sin 2x cos2 3x f. y = √
x 3x3 − tan x

2. Given the following table of values:

x f (x) f 0 (x) g(x) g 0 (x)
2 4 1 3 -2
3 -1 -3 4 -5
4 5 -4 4 1
Evaluate:

ic s
at
a. (f ◦ g)0 (2) b. (g ◦ f )0 (2) c. (g ◦ g)0 (2) d. (f ◦ g ◦ f )0 (2)

B. Differentiability. h em
at
1. Find the values of m and n so that
M

(
x2 , x < 1
of

f (x) =
mx + n , x ≥ 1
e
ut

is differentiable at x = 1.
tit

2. Find all the values of a and b such that

s
In

(
ax + b , x < 2
g(x) =
UP

2x2 − 1 , x ≥ 2
is differentiable at x = 2.

C. Implicit Differentiation.
dy
1. Use implicit differentiation to find . There is need to simplify.
dx
a. 2x sin y = (x + 2y)6 3 3
c. cot4 (xy) = 4x 2 − sin(x 5 + y 3 )
5

√ x4
b. sec(2x − y) − 3 = sin2 y −
4
2. Give the equation of the normal line to the graph of y 3 − xy 2 + cos(xy) = 2 at (0, 1).

D. Higher Order Derivatives.

1. Evaluate Dx3 sin 2x − x3 + cos x2

d2 y
2. Determine if (x + y)3 = xy 2 − 2y.
dx2
2.2. THE CHAIN RULE, AND MORE ON DIFFERENTIABILITY 95

E. Do as indicated.

1. If f (t) = at2 + bt + c , f (1) = 5, f 0 (1) = 3 and f 00 (1) = −4 , find f (3).

2. Determine the values of m and n for the curve x2 y + my 2 = n if the point (1, 1) is on its
graph and the tangent line at this point has equation 4x + 3y − 7 = 0.
d2 y 4
3. If x2 + 9y 2 = 36, show that 2
= − 3.
dx 9y
4. Find equations for the two tangent lines through the origin that are tangent to the curve
x2 − 4x + y 2 + 3 = 0.
5. Let
(
x2 , x ≤ 0
f (x) = .
x3 , x > 0
Show that f 0 (0) exists but f 00 (0) does not.

ic s
at
h em
at
M
of
e
ut
s tit
In
UP
96 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

2.3 Derivatives of Exponential and Logarithmic Functions

2.3.1 Derivatives of Logarithmic Functions
Our goal is to find the derivatives of logarithmic functions. We begin with the natural logarithmic
function and will find later that its derivative appears simplest among all logarithmic functions.
Let x > 0. Then
ln(x + ∆x) − ln x
Dx (ln x) = lim
∆x
∆x→0

1 x + ∆x
= lim ln
∆x→0 ∆x x
 
1 x ∆x
= lim ln 1 +
∆x→0 x ∆x x
" x #
1 ∆x ∆x
= lim ln 1 +
x ∆x→0 x
"  x #
1 ∆x ∆x
= ln lim 1 + , by continuity of ln .

ic s
x ∆x→0 x

at
∆x

em
Let h = . Then h → 0 if and only if ∆x → 0. Thus,
x h
at
 
1 1 1 1
Dx (ln x) = ln lim (1 + h) = ln e = .
h
M

x h→0 x x
Interestingly, the derivative of the ln function, one that cannot be expressed using a finite number
of

of operations (including extraction of nth roots) on polynomials, is a simple algebraic expression.

e
ut

In contrast, the derivatives of circular functions are products of their fellow circular functions.
tit

Now, since any logarithmic function can be expressed as the ln function times a constant, one easily
s
In

obtains its derivative. Indeed,

UP

 
ln x 1 1 1 1
Dx (loga x) = Dx = Dx (ln x) = · = .
ln a ln a ln a x x ln a
Note that the domain of the natural logarithmic function is (0, +∞), so it is an antiderivative of
1
f (x) = on (0, +∞). We wish to find a function F such that F 0 (x) = f (x) on the natural domain
x
R
of f , which is \ {0}. Using the chain rule, we can show that F (x) = ln |x| satisfies this property.

Let x 6= 0. Then |x| > 0, so ln |x| is defined. Moreover,

√ 1 1 2x 1
Dx (ln |x|) = Dx (ln x2 ) = √ · √ · 2x = 2 = .
x2 2 x2 2x x
We summarize the preceding computations into the following theorem.

Theorem 2.3.1.
1
1. Dx (ln x) = for all x > 0.
x
2.3. DERIVATIVES OF EXPONENTIAL AND LOGARITHMIC FUNCTIONS 97

1
2. Dx (loga x) = for all x > 0.
x ln a
1
3. Dx (ln |x|) = for all x 6= 0.
x

dy
Example 2.3.2. Find .
dx
1. y = log2 x

Solution.
dy 1
=
dx x ln 2
2. y = ln(3x2 + 2)

Solution.
dy 1 6x
= 2 · 6x = 2
dx 3x + 2 3x + 2

ic s
at
3. y = ln |7 − cos(2x)|

Solution.
dy 1 2 sin(2x)
h em
at
= · sin(2x) · 2 =
dx 7 − cos(2x) 7 − cos(2x)
M
of

4. y = ln [sin (log5 x)]

e
ut

Solution.
dy 1 1
tit

= · cos (log5 x) ·
dx sin (log5 x) x ln 5
s
In

5. y = x2 log x
UP

Solution.
dy 1 x
= 2x · log x + x2 · = 2x log x +
dx x ln 10 ln 10

2.3.2 Logarithmic Differentiation

Consider the function given by the equation
√
(x2 − 1)3 ( 3 1 + x + x3 )(sec4 x)
y= √ .
( csc x)( x45 )

dy
Suppose you are asked to find . One could imagine the excessive computations that would
dx
be brought about by repeated application of the product, quotient, power and chain rules. We
shall therefore introduce a process called logarithmic differentiation, a technique helpful in
differentiating an expression involving many products and quotients. In using this technique given
an equation in x and y:
98 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

1. Take the absolute value of both sides of the equation and apply properties of the absolute
value.

2. Take the natural logarithm of both sides and apply properties of logarithms to obtain a sum.
dy
3. Take the derivative of both sides implicitly with respect to x and solve for .
dx

Example 2.3.3.
√
3
dy x+1
1. Find if y = √ .
dx csc (x) 1 − x2
5

Solution.
We begin by taking the absolute value of both sides of the equation. Simplifying the result,
we get
 √3

 x+1 
|y| =  √

s
 
csc5 (x) 1 − x2 

ic
at
1
|x + 1| 3
=

em
1
| csc x|5 |1 − x2 | 2h
at
M

Next, we take the natural logarithm of both sides of the function. Using the properties of
of

logarithms, we rewrite the equation as a sum.

e
ut

1
!
|x + 1| 3
ln |y| = ln
tit

1
| csc x|5 |1 − x2 | 2
s
In

1 1
= ln |x + 1| 3 − ln | csc x|5 − ln |1 − x2 | 2
UP

1 1
= ln |x + 1| − 5 ln | csc x| − ln |1 − x2 |
3 2

Finally, we take the derivatives of both sides of the equation implicitly with respect to x and
dy
solve for ,
dx
 
1 1 2
Dx (ln |y|) = Dx ln |x + 1| − 5 ln | csc x| − ln |1 − x |
3 2
1 dy 1 1 5 1 1
· = · ·1− · (− csc x cot x) − · · (−2x)
y dx 3 x+1 csc x 2 1 − x2
 
dy 1 x
=y + 5 cot x +
dx 3(x + 1) 1 − x2
 √3
  
x+1 1 x
= √ + 5 cot x +
csc5 (x) 1 − x2 3(x + 1) 1 − x2
2.3. DERIVATIVES OF EXPONENTIAL AND LOGARITHMIC FUNCTIONS 99
√3
dy x2 tan4 x
2. Find if y = .
dx (2x2 + 1)3 log x
Solution.
Once again, we begin by taking the absolute value of both sides of the equation. This gives
us
2
|x| 3 | tan x|4
|y| = .
|2x2 + 1|3 | log x|

Then, we take the natural logarithm of both sides and rewrite the equation as a sum using the
properties of logarithms to obtain

2
ln |y| = ln |x| + 4 ln | tan x| − 3 ln |2x2 + 1| − ln | log x|.
3

s
dy

ic
Finally, we take the derivative of both sides implicitly with respect to x and solve for ,

at
dx

em
1 dy 2 1 1 1 1 1
· sec2 x − 3 · 2


= · +4· h · (4x) − ·
y dx 3 x tan x 2x + 1 log x x ln 10
√
at
3 2 4
 2

dy x tan x 2 4 sec x 12x 1
M

= + − 2 − .
dx (2x2 + 1)3 log x 3x tan x 2x + 1 (log x)(x ln 10)
of

2.3.3 Derivatives of Exponential Functions

e
ut

Next, we determine the derivatives of exponential functions, which we give in the following theorem.
s tit
In

Theorem 2.3.4.
UP

A. Dx (ax ) = ax ln a (a > 0 and a 6= 1)

B. Dx (ex ) = ex

Proof. Consider the exponential function of base a,

y = ax .

This is equivalent to the logarithmic equation

x = loga y,

and taking the derivative of both sides of the equation implicitly with respect to x, we obtain

Dx (x) = Dx (loga y)
1 dy
1= · .
y ln a dx
100 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

dy
Solving for , we obtain
dx

dy
= y ln a = ax ln a.
dx

dy
This proves the first statement of Theorem 2.3.4. Observe that if a = e, then = ex ln e = ex ,
dx
which proves the second statement of Theorem 2.3.4.

Thus, we obtain a function whose derivative is itself. (Is this the only function with this property?)
Note also that ln a > 0 when a > 1 and ln a < 0 when 0 < a < 1. Because ax > 0 for any x ∈ R,
Dx (ax ) = ax ln a > 0 when a > 1 and Dx (ax ) = ax ln a < 0 when 0 < a < 1. This explains why
f (x) = ax is an increasing function when a > 1 and decreasing when 0 < a < 1.

dy

s
Example 2.3.5. Find .

ic
dx

at
1. y = 4x

Solution.
h em
at
dy
M

= 4x ln 4
dx
of

3
2. y = ex
e
ut

Solution.
tit

dy
s

3
= ex · 3x2
In

dx
UP

3. y = 24x csc(ex )

Solution.
dy
= 24x ln 2 · 4 csc(ex ) + 24x [− csc (ex ) cot (ex ) · ex ]
 
dx

As a consequence also, if r ∈ R and x > 0, then

 
  1
r
Dx (x ) = Dx e r ln x r ln x
=e · r· = xr · rx−1 = rxr−1 .
x

Hence, the power rule holds even for irrational exponents. We state this as a theorem:

Theorem 2.3.6. (Power Rule) If f (x) = xr where r ∈ R, then f 0(x) = rxr−1.

√ √ √
For example, Dx (xπ ) = πxπ−1 and Dx ((cos x) 2) = 2(cos x) 2−1 (− sin x).
2.3. DERIVATIVES OF EXPONENTIAL AND LOGARITHMIC FUNCTIONS 101

2.3.4 Derivative of f (x)g(x) , where f (x) > 0

To differentiate expressions of the form f (x)g(x) , we either use logarithmic differentiation or rewrite
f (x)g(x) as eg(x) ln f (x) .
dy
Example 2.3.7. Find .
dx
1. y = xx , x > 0

Solution.
Using logarithmic differentiation, we have
ln y = ln (xx ) (no need to take absolute values since x > 0)
ln y = x ln x
1 dy 1
= 1 · ln x + x ·
y dx x
dy
= y(ln x + 1)
dx

ic s
dy
= xx (ln x + 1)

at
dx

em
Alternatively, we have
y = ex ln x
h
at
 
dy 1
M

x ln x
=e 1 · ln x + x ·
dx x
of

dy
= xx (ln x + 1)
e

dx
ut

2. y = (sin x)cos x , sin x > 0

s tit
In

Solution.
ln y = cos x ln(sin x)
UP

 
1 dy 1
= (− sin x) · ln(sin x) + cos x · · cos x
y dx sin x
cos2 x
 
dy cos x
= (sin x) − sin x ln(sin x) +
dx sin x
102 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

2.3.5 Exercises
Exercises for Discussion
dy
A. Find . Use logarithmic differentiation whenever appropriate.
dx
1. y = log5 (2x5 − 1) sec x 6. y = x2 · 35x−3
 

2. y = cos(ln |4 − x|)
√ 7. ln(x2 + y 2 ) = ex+y − ex−y
3. y = x5 (x2 − 1)4 sin x
xy )
(1 + x − x2 )3 8. ln 2 = (x2 − 1)π + 5(e
4. y = √ √
sec 4x(x − log x)6 9. y = (log4 x)e
x

x2 cot4 x
5. y = √ 2
3
5x − 2(log4 x)10 10. y = (tan 3x)1−x

B. Find the equation of the normal line to the graph of f (x) = 2x log2 x at the point where
x = 1.

ic s
Supplementary Exercises

at
dy

em
A. Find . Use logarithmic differentiation whenever appropriate.
dx h
at
csc(log3 x) 8. y = 2x + 3x − 5x
M

1. y = √
x − ln(5x) 
4 − 14x

of

  9. y = log8
2. sin x2 − y 2 = tan log 1 x 1 − x2


e

4
ut

3. y = logx2 +4 3 10. y = e4x−1 + ln x2 + e

√
tit

esin 2x + ln(1 + x)
4. y = 1 + x − log12 x
2log5 x 11. y =
s

log2 2 − x + log5 (2x + 1)

In

5. y = (x + 2x )sin 5x
2

12. ln |x − cos y| = y 2 + x − 2
UP

6. y x = xy
7. ln(x − y) = (tan x)y 13. x tan y − sin [(x − 1) ln y] = x
2.4. DERIVATIVES OF OTHER NEW CLASSES OF FUNCTIONS 103

2.4 Derivatives of Other New Classes of Functions

2.4.1 Derivatives of Inverse Circular Functions
We now do calculus on these functions. We begin with their derivatives, which we enumerate in
the following theorem.

Theorem 2.4.1.

1 1
1. Dx sin−1 x = √ 4. Dx cot−1 x = −



1 − x2 1 + x2
1 1
5. Dx sec−1 x = √


2. Dx cos−1 x = − √


1 − x2 x x2 − 1
1 1
6. Dx csc−1 x = − √


3. Dx tan−1 x =


1 + x2 x x2 − 1
Proof. We show the proof of statement 1 only. The rest can be proved similarly.

ic s
y = sin−1 x

at
x = sin y
Dx (x) = Dx (sin y) hem
at
dy
1 = cos y ·
M

dx
dy 1
of

=
dx cos y
e

p √ dy
ut

Note that cos y = ± 1 − sin2 y = ± 1 − x2 . Since y ∈ − π2 , π2 , cos y ≥ 0. Therefore,

 
=
dx
tit

1
√ .
s

1 − x2
In
UP

dy
Example 2.4.2. Find .
dx
1. y = sin−1 (3x)

Solution.
dy 1 3
=p · (3) = √
dx 1 − (3x)2 1 − 9x2

2. y = cot−1 (ln x)

Solution.
   
dy 1 1
= − 2
·
dx 1 + (ln x) x
 y
1 2
= log2 (x − y) − csc−1 (3x2 )

3.
5
Solution.
104 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

We perform implicit differentiation.

 y    
1 1 dy 1 dy
ln = · 1−
5 5 dx (x − y) ln 2 dx
 
−1 2

1
− 2 csc 3x · − √ · 6x
3x2 9x4 − 1
 
1 −1 2

6x
− 2 csc 3x · − √
dy (x − y) ln 2 3x 2 9x4 − 1
=  y  
dx 1 1 1
ln +
5 5 (x − y) ln 2

2.4.2 Derivatives of Hyperbolic Functions

Theorem 2.4.3.

1. Dx (sinh x) = cosh x 4. Dx (coth x) = − csch2 x

ic s
2. Dx (cosh x) = sinh x 5. Dx (sech x) = − sech x tanh x

at
em
3. Dx (tanh x) = sech2 x 6. Dx (csch x) = − csch x coth x
h
at
Proof. Again, we shall prove the first statement only.
M
of

ex − e−x dy ex − (e−x ) · (−1) ex + e−x

If y = sinh x = , then = = = cosh x.
e

2 dx 2 2
ut
tit

dy
Example 2.4.4. Find .
s

dx
In

1. y = csch x5 − 2


UP

Solution.
dy
= − csch x5 − 2 coth x5 − 2 · (5x4 )



dx
√
2. y = tanh3


x+3

Solution.
dy √ √ 1
= 3 tanh2 x + 3 · sech2



x+3 · √
dx 2 x+3

3. y = (cosh x)sinh x , cosh x > 0

Solution.
y = esinh x ln(cosh x)
 
dy sinh x ln(cosh x) 1
=e cosh x · ln(cosh x) + sinh x · · sinh x
dx cosh x
2.4. DERIVATIVES OF OTHER NEW CLASSES OF FUNCTIONS 105

2.4.3 Derivatives of Inverse Hyperbolic Functions

We will find that the derivative of each inverse hyperbolic function resembles that of the corre-
sponding inverse circular function. The proofs of the statements in the next theorem are left as
exercises.

Theorem 2.4.5.

1 1
1. Dx (sinh−1 x) = √ 4. Dx (coth−1 x) = , |x| > 1
x2 + 1 1 − x2
1 1
2. Dx (cosh−1 x) = √ 5. Dx (sech−1 x) = − √
x2 −1 x 1 − x2
1 1
3. Dx (tanh−1 x) = , |x| < 1 6. Dx (csch−1 x) = − √
1 − x2 |x| x2 + 1
dy
Example 2.4.6. Find .
dx

ic s
A. y = sinh−1 (1 − x)

at
Solution.
dy 1 −1 em
h
at
=p · (−1) = p
dx 2
(1 − x) + 1 (1 − x)2 + 1
M

 x 
1
of

−1 4 −1
B. y = coth (x ) − sech
4
e
ut

Solution.
tit

 x  
dy 1 3 +

1 1 1
· ·
s

= 4x ln
In

1 − x8
q
dx 1 x

1 2x

4 4
4 1− 4
UP

C. y = log10 cosh−1 (2x)

Solution.
dy 1 1
= −1 ·√ ·2
dx cosh (2x) ln 10 4x2 − 1
106 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

2.4.4 Exercise
Exercises for Discussion
dy
A. Find .
dx
√
1. y = 5cos
−1 x sec−1 (2x ) − e
4. y =
√ [log (3x)] tan−1 x
2. y = cos−1 6
cos x4 + ecot x − 4π 2 5. sin−1 (xy) = cos−1 (x − y)


tan−1 √x
3. y = sec ln csc−1 x 6. y = sin−1 x2



dy
B. Find .
dx

3
1. y = cosh(ln x) 4. y = 10csch(x ) sec−1 (cosh 4x)
√
2. y = tanh3 3 2x
5. cosh(x − y) = log3 x − tanh(x2 + y 2 )
log5 (sech(ex ))

s
√

ic
3. y = 6. sec−1 ( x) = tanh(x2 y) + y sinh 2
sinh (x − tan−1 x)

at
em
dy
C. Find . h
dx
at
M

1. y = cosh−1 x7 sin−1 (ex sinh 3x) + coth−1 (x2 )

4. y = √
of

2. y = tanh−1 (5x2 ) csch−1 (ln x) sech−1 x

√
5. sinh−1 (x + y) = cosh−1 (x − y) + x
e

√
ut

tanh−1 x
3x2 +1
3. y = 6. cosh x = tanh−1 y
tit

1 + sec x2
s
In

Supplementary Exercises
UP

dy
A. Find . There is no need to simplify.
dx

6. y = sinh−1 (ln x) + sin−1 (ln x)

 
−1 x+1
1. y = sin √
2 log5 x √
2. y = tan (sech x) + sech tan−1 x
−1 7. y = +sech(4x−2 −sec−1 ( x))


−1
coth (2x)
3. y = csch(cos−1 (x2 ))
  2
x −1
 8. sec−1 (2y cos x) = xy
2
4. y = tanh log7
x+1 9. 3y = y tanh(x) + 2
e tanh x
5. y = 2 10. 4y = y 2 − cos−1 (x) + 1
x − sech x
2.5. MORE INDETERMINATE FORMS AND L’HÔPITAL’S RULE 107

2.5 More Indeterminate Forms and L’Hôpital’s Rule

We began this course with the concept of the limit: the behavior of a function as the indepen-
dent variable approaches a certain value, or as it increases or decreases without bound. The new
techniques of differentiation we have learned so far will aid us in evaluating a wider range of lim-
its, including new types of indeterminate forms which we could not previously evaluate with the
techniques learned in Chapter 1.

By the end of this section, the student will be able to:

• evaluate limits with indeterminate form 0/0, ∞/∞ using L’Hôpital’s Rule; and

• evaluate limits with indeterminate forms 0 · ∞, ∞ − ∞, 00 , 1∞ , ∞0 .

0 ∞
2.5.1 Indeterminate Forms of Type and

s
0 ∞

ic
at
With the aid of derivatives, certain limits can be evaluated more conveniently. We first recall

em
some terminology defined in the early part of this course. We also recall here some techniques in
evaluating limits we have previously encountered. h
at
f (x)
M

Definition 2.5.1. The lim is an indeterminate form of type

x→a g(x)
of

0
1. if lim f (x) = lim g(x) = 0.
e
ut

0 x→a x→a
tit

∞
2. if lim f (x) and lim g(x) are both +∞ or −∞.
∞ x→a
s

x→a
In

Of course, “x → a” may be replaced by “x → a+ ”, “x → a− ”, “x → +∞” or “x → −∞”.

UP

Example 2.5.2. Evaluate the following limits.

x2 − 3x
 
0
1. lim 2
x→0 2x + x 0

Solution.

x2 − 3x x(x − 3) x−3
lim 2
= lim = lim = −3
x→0 2x + x x→0 x(2x + 1) x→0 2x + 1

 
sin 5x 0
2. lim
x→0 sin 3x 0

Solution.

   
sin 5x sin 5x 3x 5 5 5
lim = lim =1·1· =
x→0 sin 3x x→0 5x sin 3x 3 3 3
108 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

x2 + 3x − 10
 
0
3. lim 2
x→2− x − 4x + 4 0
Solution.

x2 + 3x − 10 (x + 5)(x − 2)
lim = lim
x→2− x2 − 4x + 4 x→2− (x − 2)2
 
x+5 7
= lim
x→2− x − 2 0−
= −∞
 
3x − 1 +∞
4. lim
x→+∞ 7 − 6x −∞
Solution.

1
3x − 1 3x − 1 3 − x1 1

s
x
· =−

ic
lim = lim 1 = lim 7
x→+∞ 7 − 6x x→+∞ 7 − 6x 2
x −6
x→+∞

at
x

2.5.2 L’Hôpital’s Rule

em
h
at
The following theorem tells us how derivatives can be used to evaluate limits that are indeterminate
0 ∞
M

of type or . It is usually referred to as L’Hôpital’s Rule, after the French mathematician

0 ∞
of

Guillaume François Marquis de L’Hôpital.

e
ut

Theorem 2.5.3. Let f and g be functions differentiable on an open interval I containing a

tit

f (x) 0
except possibly at a and g 0 (x) 6= 0 for all x ∈ I \ {a}. If lim is indeterminate of type
s

x→a g(x) 0
In

∞
or , then
UP

∞
f (x) f 0 (x)
lim = lim 0 ,
x→a g(x) x→a g (x)

f 0 (x) f 0 (x)
provided lim 0
exists or lim 0 = ±∞.
x→a g (x) x→a g (x)

Remark 2.5.4. L’Hôpital’s Rule, with suitable modifications, is valid if “x → a” is replaced by

“x → a+ ”, “x → a− ”, “x → +∞” or “x → −∞”.

Example 2.5.5. Evaluate the following limits.

x2 − 3x
 
0
1. lim 2
x→0 2x + x 0
Solution.

Dx x2 − 3x


x2 − 3x 2x − 3
lim 2 = lim 2

= lim = −3
x→0 2x + x x→0 Dx 2x + x x→0 4x + 1
2.5. MORE INDETERMINATE FORMS AND L’HÔPITAL’S RULE 109
 
sin 5x 0
2. lim
x→0 sin 3x 0

Solution.

sin 5x 5 cos 5x 5
lim = lim =
x→0 sin 3x x→0 3 cos 3x 3

x2 + 3x − 10
 
0
3. lim 2
x→2− x − 4x + 4 0

Solution.

x2 + 3x − 10
 
2x + 3 7
lim 2 = lim

s
x→2− x − 4x + 4 x→2 2x − 4
− 0−

ic
= −∞

at
4. lim
3x − 1

+∞
 hem
at
x→+∞ 7 − 6x −∞
M

Solution.
of
e
ut

3x − 1 3 1
tit

lim = lim =−
x→+∞ 7 − 6x x→+∞ −6 2
s
In

x3 − 3x + 2
 
0
UP

5. lim
x→1 1 − x + ln x 0

Solution.

x3 − 3x + 2 3x2 − 3
 
0
lim = lim
x→1 1 − x + ln x x→1 −1 + 1 0
x
6x
= lim 1
x→1 − 2
x
= −6

 
csc x −∞
6. lim
x→0− 1 − cot x +∞

Solution.
110 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

csc x − csc x cot x

lim = lim
x→0− 1 − cot x x→0− csc2 x
 
− cot x +∞
= lim
x→0− csc x −∞
csc2 x
= lim
x→0− − csc x cot x
 
csc x −∞
= lim
x→0− − cot x +∞
− cot x
= lim
x→0− csc x

Observe that the expression in the last line above is exactly the same as the expression in the
second line. Hence, continued application of L’Hôpital’s Rule here will just lead us to an infinite
string of equations and will not help us evaluate the limit. This example should make you realize

ic s
that L’Hôpital’s Rule is not always helpful. Sometimes, we must fall back to old-fashioned tricks.

at
em
For instance, we evaluate this limit by simply manipulating the given expression to obtain
h
a simpler expression:
at
M
of

1
csc x sin x
e

lim = lim cos x

ut

x→ 0− 1 − cot x x→ 0− 1 −
sin x
tit

1
= lim
s
In

x→ 0− sin x − cos x
= −1
UP

It is imperative to remember the behavior of each function introduced in this course; doing so
will help us in computing new limits. Recalling the graphs of our new functions will be helpful in
remembering their behavior. For instance, using the graph of f (x) = loga x, where 0 < a < 1, one
sees that lim loga x = +∞. Thus, if f (x) approaches 0 through positive values as x approaches
x→0+
k, then
lim loga [f (x)] = lim loga y = +∞.
x→k y→0+

x + sin x
Example 2.5.6. Evaluate: lim
x→+∞ x

Solution.
Note that since x + sin x ≥ x − 1 for any x ∈ R and x→+∞
lim x − 1 = +∞, then lim x + sin x = +∞
x→+∞
∞
as well. Thus, the limit is indeterminate of type .
∞
2.5. MORE INDETERMINATE FORMS AND L’HÔPITAL’S RULE 111

Dx (x + sin x) 1 + cos x
However, = andlim does not exist since cos x does not ap-
Dx (x) 1 1 + cos x
x→+∞
1
proach any particular value as x → +∞. Neither does 1 + cos x grow without bound. Thus,
L’Hôpital’s Rule does not apply.

We therefore need to employ other techniques. In particular, notice that for any x > 0, the following
hold:
x − 1 ≤ x + sin x ≤ x + 1
x−1 x + sin x x+1
⇔ ≤ ≤
x x x
x−1 x+1 x + sin x
Also, lim = 1 = lim , so by the Squeeze Theorem, lim = 1.
x→+∞ x x→+∞ x x→+∞ x

2.5.3 Indeterminate Forms of Type 0 · ∞ and ∞ − ∞

Definition 2.5.7.

s
1. The lim f (x)g(x) is an indeterminate form of type 0 · ∞ if either

ic
x→a

at
lim f (x) = 0 and lim g(x) = +∞ or − ∞, or

em
x→a x→a
h
lim f (x) = +∞ or − ∞ and lim g(x) = 0.
at
x→a x→a
M

2. The lim (f (x) + g(x)) is an indeterminate form of type ∞ − ∞ if either

x→a
of

lim f (x) = +∞ and lim g(x) = −∞, or

e

x→a x→a
ut
tit

lim f (x) = −∞ and lim g(x) = +∞.

x→a x→a
s
In

0 ∞
Remark 2.5.8. L’Hôpital’s Rule works only for indeterminate forms of type and . Any
0 ∞
UP

other indeterminate form must be expressed equivalently in one of these two forms if we wish to
apply L’Hôpital’s Rule. For the new indeterminate forms described above, these conversions can
be performed as described below.

1. If lim f (x) = 0 and lim g(x) = +∞ or −∞, write lim f (x)g(x) as:
x→a x→a x→a

f (x) 0
(a) lim , which is indeterminate of type , or
x→a 1 0
g(x)
g(x) ∞
(b) lim , which is indeterminate of type
x→a 1 ∞
f (x)

and apply L’Hôpital’s Rule.

2. If lim f (x) + g(x) is indeterminate of type ∞ − ∞, rewrite f (x) + g(x) as a single expression to
x→a
0 ∞
obtain an indeterminate form of type or and apply L’Hôpital’s Rule.
0 ∞
Example 2.5.9. Evaluate the following limits.
112 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

1. lim sin−1 (2x) csc x (0 · +∞)

x→0+

Solution.

sin−1 (2x)
 
−1 0
lim sin (2x) csc x = lim
x→0+ x→0+ sin x 0
1
√ · (2)
1 − 4x2
= lim
x→0+ cos x
2
=
1
= 2

2. lim tan θ ln (sin θ) (+∞ · 0)

θ→ π2 −

ics
Solution.

at
lim tan θ ln (sin θ) = lim
ln(sin θ) hem  
0
at
π−
θ→ 2 π−
θ→ 2 cot θ 0
M

1


sin θ cos θ
= lim
of

2
θ→ π2 − − csc θ
e

= lim − sin θ cos θ

ut

θ→ π2 −
tit

= −1(0)
s
In

= 0
UP

   
x 1 1 1
3. lim − −
x→1+ x − 1 ln x 0+ 0+

Solution.

   
x 1 x ln x − (x − 1) 0
lim − = lim
x→1+ x − 1 ln x x→1+ (x − 1) ln x 0
x · x1 + ln x − 1
= lim
x→1+ (x − 1) · 1 + ln x
x
 
ln x 0
= lim
x→1+ 1 − 1 + ln x 0
x
1
x
= lim
x→1+ 12 + 1
x x
1
=
2
2.5. MORE INDETERMINATE FORMS AND L’HÔPITAL’S RULE 113

2.5.4 Indeterminate Forms of Type 1∞ , 00 and ∞0

Definition 2.5.10. Let f be a nonconstant function. The lim f (x)g(x) is an indeterminate form
x→a
of type

1. 1∞ if lim f (x) = 1 and lim g(x) = +∞ or −∞.

x→a x→a

2. 00 if lim f (x) = 0, through positive values, and lim g(x) = 0.

x→a x→a

3. ∞0 if lim f (x) = +∞ and lim g(x) = 0.

x→a x→a

Remark 2.5.11. If lim f (x)g(x) is indeterminate of type 1∞ , 00 or ∞0 , we write

x→a

lim f (x)g(x) = lim eg(x) ln[f (x)]

x→a x→a

and evaluate lim g(x) ln[f (x)] first. Then, if

x→a

1. lim g(x) ln[f (x)] = L ∈ R, then x→a

lim f (x)g(x) = eL .

ic s
x→a

at
2. lim g(x) ln[f (x)] = +∞, then lim f (x)g(x) = +∞.

em
x→a x→a

3. lim g(x) ln[f (x)] = −∞, then lim f (x)g(x) = 0.

h
at
x→a x→a
M

Example 2.5.12. Evaluate the following limits.

of

1. lim xsin x 00


e

x→0+
ut

Solution.
tit

First, write xsin x = esin x ln x . Evaluate first lim sin x ln x.

s
In

x→0+
UP

 
ln x −∞
lim sin x ln x (0 · (−∞)) = lim
x→0+ x→0+ csc x +∞
1
x
= lim
− csc x cot x
x→0+
− sin2 x
 
0
= lim
x→0+ x cos x 0
−2 sin x cos x
= lim
x→0+ x(− sin x) + cos x
= 0

Hence, lim xsin x = e0 = 1.

x→0+

3 2x
 
2. lim 1− (1∞ )
x→+∞ x
Solution.
114 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

3
3 2x
= e2x ln(1− x )


1− x

 
3
  ln 1 −  
3 x 0
lim 2x ln 1 − (+∞ · 0) = lim
x→+∞ x x→+∞ 1 0
2x  
1 3
·
3 x2
1−
= lim x
x→+∞ 1
− 2
2x
−6
= lim
x→+∞ 3
1−
x
= −6

3 2x
 

s
Hence, lim 1− = e−6 .

ic
x→+∞ x

at
hem
at
M
of
e
ut
tit
s
In
UP
2.5. MORE INDETERMINATE FORMS AND L’HÔPITAL’S RULE 115

2.5.5 Exercises
Exercises for Discussion

A. Evaluate the following limits.

3−x
 p 
1. lim x 11. lim x− x2 − x + 2
x→3 2 − 8 x→+∞
ln x 12. lim (tan x)cos x
2. lim 1 x→ π2 −
+
x→0 e x
sec x
1
3. lim 13. lim x + e2x x
π − 1 + tan x x→0
x→ 2
x
1
4. lim √ 14. x + e2x x
lim
x→−∞ x2 + 1 x→+∞
√
ln (3 + ex ) 15. lim
x
cosh 3x
5. lim x→+∞
x→+∞ 7x  
sin x − x x−1
6. lim 16. lim x ln

s
x3 x→+∞ x+1

ic
x→0
1 − x + ln x

at
sin x + tan x
7. lim 17. lim

em
x→1 1 + cos πx x→0−ex + e−x − 2
1
8. lim x2 e x

cos x
18. lim x − π2
x→0+
h
x→ π2 +
at
−1
9. lim tan (x) ln x
M

x→0− 19. lim logx (x + 10)

  x→+∞
1
of

10. lim csc x − 20. lim ln(e3x + 1) − 3x

x→0− x x→+∞
e
ut
tit

B. Do as indicated.
s
In

1. If an electrostatic field E acts on a gaseous polar dielectric, the net dipole moment P per
1
UP

unit volume is P (E) = coth E − . Evaluate lim P (E).

E E→0+
 x
x+m
2. For what values of m is lim = e?
x→+∞ x − m

Supplementary Exercises

A. Evaluate the following limits.

2x3 + x2 + 1 1 − ex
2

1. lim 5. lim
x→−∞ x2 − x x→0 log3 (1 − x)

e2x − 4 2x3 + ln 5x
2. lim 6. lim
x→ln 2 x − ln 2 x→+∞ 7 + ex
cos x
sinh−1 x4


3. limπ
x→ 2 x − π 3


2
7. lim 1
x→+∞ 3x − 1
tanh x
4. lim 8. lim x (ln(−x))
x→0+ sech x − 1 x→0−
116 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

8 x
 
9. lim 1−
2
− 2 16. lim [1 + tan(11x)]cot(2x)
x→0+
x→−∞ 5x 5x
1 4 
2x − 3
2x+1
10. lim 2 − 17. lim
x→3+ x − 9 ln (2x − 5) x→+∞ 2x + 5
11. lim (ln(x + 1) − ln(x − 1))
x→+∞ x3
x3

12. lim cot(3x) ln [1 + tan(5x)] 18. lim
x→0+ x→+∞ x3 + 4
x csch x
13. lim e
x→0
19. lim (2x + 5)csch(2x+4)
14. lim x(ln 2)/(1+ln x) x→−2+
x→∞
√
1 3x 9
  
16 − 8x
15. lim 1+ 20. lim tanh
x→+∞ x x→2− x−2

ic s
at
h em
at
M
of
e
ut
s tit
In
UP
2.6. THE MEAN VALUE THEOREM 117

2.6 The Mean Value Theorem

In this section, we discuss one of the most important theorems in calculus, namely, the Mean
Value Theorem. It is used to prove several other theorems including many of the ones that will be
discussed in this course.
At the end of this section, the student will be able to:

• interpret Rolle’s Theorem and Mean Value Theorem using graphs

• solve problems by applying Rolle’s Theorem and Mean Value Theorem

2.6.1 Rolle’s Theorem

The following theorem is a specific case of the Mean Value Theorem and it was first proved by the
French Mathematician Michel Rolle in 1691.

ic s
at
Theorem 2.6.1 (Rolle’s Theorem). Let f be a function such that

(i) f is continuous on the closed interval [a, b] h em

at
(ii) f is differentiable on the open interval (a, b) and
M

(iii) f (a) = 0 = f (b).

of

Then there exists a number c in the open interval (a, b) such that f 0 (c) = 0.
e
ut
tit
s

y y y
In

c
UP

a b
x
a
x
a bx b1 b2

Figure 2.6.1 Figure 2.6.2 Figure 2.6.3

Remarks.
1. Note that condition (iii) of Rolle’s Theorem implies that the line passing through the points
(a, f (a)) and (b, f (b)) is horizontal. On the other hand, the conclusion implies that there is
a horizontal tangent line to the graph of f and the point of tangency has x-coordinate lying
between a and b. Refer to Figure 2.6.1.

2. Continuity on [a, b] is important because there are functions that satisfy only conditions (ii) and
(iii) but do not satisfy the conclusion. Refer to Figure 2.6.2 and consider the function on the
interval [a, b1 ]. It may also be the case that the conclusion is satisfied even if one of premises is
not satisfied. Refer to Figure 2.6.2 and consider the function on the interval [a, b2 ].
118 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

3. Notice that f need not be differentiable at the endpoints a and b. Refer to Figure 2.6.1.

4. The number c ∈ (a, b) in the conclusion need not be unique. Refer to Figure 2.6.3.

2.6.2 The Mean Value Theorem

Theorem 2.6.2 (The Mean Value Theorem). Let f be a function such that

(i) f is continuous on the closed interval [a, b]

(ii) f is differentiable on the open interval (a, b).

f (b) − f (a)
Then there is a number c on the open interval (a, b) such that f 0 (c) = .
b−a

Remark 2.6.3. The Mean Value Theorem is the generalization of Rolle’s Theorem where the line
`, passing through (a, f (a)) and (b, f (b)) is not necessarily horizontal. The conclusion says that
there is a tangent line to the graph of f that is parallel to ` and whose point of tangency has

ic s
x-coordinate between a and b. Refer to Figure 2.6.4 and 2.6.5.

at
em
y
y h
at
`
M

x
a c b
of

c2 b
e

x
a c1
ut
tit

Figure 2.6.4 Figure 2.6.5

s
In

Example 2.6.4.
UP

1. Determine if Rolle’s Theorem is applicable to the given functions on the indicated intervals:

(a) f (x) = x3 − 4x2 + 5x − 2 on [1, 2]

Solution.
Note that f (x) is continuous on [1, 2] and differentiable on (1, 2) because it is a polynomial.
Moreover, f (1) = 0 = f (2). Hence Rolle’s Theorem can be applied. We therefore conclude
that there exists a c such that 1 < c < 2 and f 0 (c) = 0. In fact, we can solve for c. Since

f 0 (x) = 3x2 − 8x + 5 = (3x − 5)(x − 1),

then the number c = 53 is in the interval (1, 2) and satisfies f 0 (c) = 0.

(
x2 , x ≤ 21
(b) f (x) = on [0, 1]
x − 1 , x > 12

Solution.
2.6. THE MEAN VALUE THEOREM 119

Notice that the only possible point of discontinuity of f (x) is at x = 21 . Checking conditions
for continuity at x = 21 , we have
 
1 1 1 1
(i) f = (ii) lim f (x) = lim (x − 1) = − (iii) lim f (x) = lim x2 =
2 4 x→ 1+
x→ 1+ 2 x→ 1−
x→ 1− 4
2 2 2 2

1
Hence, f is not continuous at x = 2 and Rolle’s Theorem cannot be applied.

2. Apply the Mean Value Theorem to f (x) = 2x3 − 3x2 + x to show that 6x2 + 1 = 6x has at least
one real root between 0 and 1.

Solution.
Note that f 0 (x) = 6x2 − 6x + 1. Since f is a polynomial and f (0) = 0 = f (1), then f satisfies
the assumptions of Rolle’s Theorem on [0, 1]. Hence there is a c ∈ (0, 1) such that f 0 (c) = 0.
This c is a root of the equation 6x2 + 1 = 6x.
x+2 1
3. Let f (x) = . Show that there is a c ∈ (1, 2) such that if f 0 (c) = − .
x+1 6

ic s
at
Solution.

em
Since −1 ∈
/ [1, 2] and f is a rational function, f is continuous on [1, 2] and differentiable on (1, 2).
h f (2) − f (1) 4 3 1
By the Mean Value Theorem, there is a c ∈ (1, 2) such that f 0 (c) = = − =− .
at
2−1 3 2 6
M

4. Suppose that f (x) is continuous on [6, 15] and differentiable on the interval (6, 15) and f 0 (x) ≤ 10
of

for all x. If f (6) = −2, what is the largest possible value for f (15)?
e
ut

Solution.
tit

Note that f satisfies the hypotheses of the Mean Value Theorem. Hence there is a c ∈ (6, 15)
s

such that
In

f (15) − f (6) f (15) + 2

f 0 (c) = = ≤ 10.
UP

15 − 6 9
Therefore, f (15) ≤ 88.
120 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

2.6.3 Exercises
Exercises for Discussion

A. Determine if Rolle’s Theorem applies to the following functions on the given intervals, and if
so, find all values of c satisfying the conclusion of the theorem.

1. f (x) = x2 − x − 2 on [−1, 2] x2 − x − 12
4. f (x) = on [−3, 4]
2. f (x) = x3/4 − 2x1/4 on [0, 4] x−3
3. f (x) = 3 cos2 x on [ π2 , 3π
2 ] 5. f (x) = 1 − |x| on [−1, 1]

B. Determine if the Mean Value Theorem applies to the following functions on the given intervals,
and if so, find all values of c satisfying the conclusion of the theorem.

1. f (x) = x2 + 2x − 1 on [0, 1] 3. f (x) = 3(x − 4)2/3 on [−4, 5]

√
4. f (x) = 1 + cos x on [− π2 , π2 ]

ic s
4
2. f (x) = on [ 25 , 4]

at
9 − 2x 5. f (x) = (1 + sin x)2 on [−π, 0]

C. Do as indicated.
h em
at
1. Using the function f (x) = x4 − 2x2 + 4x, show that the Rolle’s Theorem confirms there
M

exist x ∈ [−2, 0] such that x is a root of the equation x3 − x + 1 = 0.

of

2. Prove that the equation x3 + 2x + 5 = 0 cannot have more than one real root.
e
ut

3. Let g be a function such that g 0 (x) ≥ −6 for all x ∈ [0, 2]. If g(0) = 0, determine the least
tit

possible value of g(2).

s
In

4. Show that sin b − sin a ≤ b − a whenever b ≥ a. In particular, show that sin x < x for all
x > 0.
UP

5. Let f (x) be differentiable at everywhere and suppose that f (1) = 1, f 0 (x) < 0 on (−∞, 1),
and f 0 (x) = 1 on (1, ∞).
a. Show that f (x) ≥ 1 for all x.
b. Must f 0 (1) = 0? Explain your answer.
6. Suppose two runners in a 100m dash finish a tie. Show that there is at least one instant
during the race where they had the same velocity.

Supplementary Exercises

A. Determine if Rolle’s Theorem applies to the following functions on the given intervals, and if
so, find all values of c satisfying the conclusion of the theorem.

1. f (x) = x2 − 4x + 3 on [1, 3] 4. f (x) = |x − 2| − 1 on [1, 3]

(
2. f (x) = sin πx − x2 + 2x on [0, 2] x2 − 4 if x < 1
5. f (x) = on [−2, 58 ]
3. f (x) = sin3 x cos x on [0, 2π] 5x − 8 if 1 ≤ x
2.6. THE MEAN VALUE THEOREM 121

B. Determine if the Mean Value Theorem applies to the following functions on the given intervals,
and if so, find all values of c satisfying the conclusion of the theorem.

−2 3. f (x) = ex + e4−x on [0, 4]

1. f (x) = on [−2, −1]
3x + 7
4. f (x) = ln 3x on [1, e]
2. f (x) = (2x − 11)−2 on [4, 6] 5. f (x) = csc−1 (x − 1) on [1, 3]

C. Do as indicated.

1. Suppose f (x) is continuous and differentiable everywhere. Suppose also that f (x) has at
least two distinct zeros. Show that f 0 (x) has at least one zero.
2. Use the Mean Value Theorem to show that the graph of f (x) = x5 − 3x3 − x + 5 has a
tangent line on (0, 2) which is parallel to 3x − y = 2.
18
3. Let f (x) = cos(2x) − 5 sin(5x). Show that there exists c ∈ (0, π6 ) such that f 0 (c) = − .
π

s
√ √ y−x

ic
4. Use the Mean Value Theorem to prove that if 0 < x < y, then y − x < √ . In
2 x

at
particular, show that the geometric mean of x and y is less than their arithmetic mean,

em
√
i.e. xy < 12 (x + y). h
at
5. Suppose f is continuous on an interval I. Use the Mean Value Theorem to show that if
f 0 (x) = 0 for all x ∈ I, then f is constant in I.
M

π
6. Use the previous item to prove the identity sin−1 x + cos−1 x = on [−1, 1].
of

2
e
ut
stit
In
UP
122 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

2.7 Relative Extrema of a Function

In this chapter, we wish to analyze a function’s behavior, which is best described using its graph.
We aim to sketch the graph of a function without the help of graphing software. To this end, we
will first develop tools to determine the turning points of the graph of a function.
At the end of this section, the student will be able to:

• interpret the notion of increasing functions, decreasing functions, and relative ex-
trema of a function graphically

• determine if a function is increasing or decreasing on an interval using the first

derivative

• find the relative extrema using the First Derivative Test

ic s
2.7.1 Relative Extrema

at
Definition 2.7.1.

em
1. A function f is said to have a relative maximum at x = c if there is an open interval I,
h
at
containing c, such that f (x) is defined for all x ∈ I and f (x) ≤ f (c) for all x ∈ I.
M

2. A function f is said to have a relative minimum at x = c if there is an open interval I,

of

containing c, such that f (x) is defined for all x ∈ I and f (x) ≥ f (c) for all x ∈ I.
e
ut

3. We say f has a relative extremum at x = c if f has either a relative maximum at x = c or a

tit

relative minimum at x = c.
s
In

4. If f has a relative extremum at x = c, then it is equivalent to saying that (c, f (c)) is a relative
extremum point of f , or f (c) is a relative extremum value of f .
UP

Example 2.7.2. Let f be a function whose graph is illustrated in Figure 2.7.1.

c1 c2 c3 c4 c5 c6
Figure 2.7.1

• f has a relative maximum at x = c3 and at x = c5 .

• f has a relative minimum at x = c4 .

• f has neither a relative maximum at x = c1 nor a relative minimum at x = c6 because there is

no open interval for which the point (c1 , f (c1 )) is the highest, or (c6 , f (c6 )) is the lowest.

• f has no relative minimum at x = c2 because f (c2 ) is not defined.

2.7. RELATIVE EXTREMA OF A FUNCTION 123

2.7.2 Critical Numbers

To make it easier to locate the relative extrema of a function, we narrow down our search to a few
candidates.

Definition 2.7.3. A number c ∈ domf is said to be a critical number of f if either f 0 (c) = 0 or

f 0 (x) is undefined at x = c.

Example 2.7.4. Find all the critical numbers of the following functions.
x3
1. f (x) = − 3x
9
Solution.
R. Differentiating f , we get f 0(x) = x3
2 1 1
Note that dom f = − 3 = (x2 − 9) = (x − 3)(x + 3).
3 3
The derivative is always defined and f 0 (x) = 0 when x = 3 or x = −3. Therefore the critical
numbers are 3 and −3.

2. f (x) = x4 + 2x3

ic s
at
Solution.
R and its derivative is f 0(x) = 4x3 + 6x2 = 2x2(2x + 3). Hence, the critical
em
The domain of f is
3 h
numbers of f are 0 and − .
at
2
M

x2
3. f (x) =
of

9 − x2
e

Solution.
ut

R \ {±3}. The derivative of f is f 0 (x) =

18x
. Note that f 0 (x) is
tit

Note that dom f =

(9 − x2 )2
s

not defined at x = ±3 but ±3 are not critical numbers of f because ±3 ∈ / domf . Meanwhile,
In

f 0 (x) = 0 when x = 0. Thus, the only critical number of f is 0.

UP

4. f (x) = −x4/3 + 4x1/3

Solution.
First, dom f = R. Differentiating f , we get f 0(x) = − 34 x1/3 + 43 x−2/3 = −4(x
3x
− 1)
2/3
. Note that
f 0 (x) = 0 if and only if x = 1 and that f 0 (x) is undefined at x = 0. Moreover, 0 ∈ domf . Thus,
the critical numbers of f are 0 and 1.

The following theorem tells us that finding the critical numbers of a function takes us one step
closer to finding its relative extremum points.

Theorem 2.7.5. If f has a relative extremum at x = c, then c is a critical number of f .

Illustration 2.7.6.

1. The graph of f (x) = x2 given in Figure 2.7.2 has a relative maximum point at its vertex
(0, 0). Note that f 0 (x) = 2x. Indeed f 0 (0) = 0.
124 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

2. Take f (x) = |x − 1|. From its graph shown in Figure 2.7.3, it is clear that f has a relative
minimum at x = 1. Since the graph has a corner at x = 1, we know that f 0 (1) is undefined.

y = −x2
y = |x − 1|

Figure 2.7.2 Figure 2.7.3

Remark 2.7.7. The preceding theorem says that relative extrema can only be attained at critical
numbers of f , that is, if a is not a critical number, then the point (a, f (a)) is not a relative extremum
point of f . However, these critical numbers are only candidates for relative extrema. If c is a critical
number, f may or may not have a relative extremum at x = c. This will be illustrated in the next
sections.

ic s
2.7.3 Increasing/Decreasing Functions

at
em
Our next goal is to determine when the graph of a function rises or falls. We will see then that this
h
is closely related to finding the relative extremum points of a function.
at
M

Definition 2.7.8. Let f be a function defined on an interval I.

of

1. f is said to be (strictly) increasing on I if f (a) < f (b) for all a, b ∈ I such that a < b.
e
ut

2. f is said to be (strictly) decreasing on I if f (a) > f (b) for all a, b ∈ I such that a < b.
tit

Example 2.7.9. Let f be a function whose graph is illustrated in Figure 2.7.1. Then
s
In

• f is decreasing on [c1 , c2 ), [c3 , c4 ] and [c5 , c6 ]. But note that it is incorrect to say f is decreasing
UP

on [c1 , c2 ) ∪ [c3 , c4 ] ∪ [c5 , c6 ].

• f is increasing on (c2 , c3 ] and [c4 , c5 ].

The following theorem gives us an analytical (as opposed to graphical) method of showing that the
function is increasing or decreasing. It should be noted that the Mean Value Theorem is a key in
proving this theorem.

Theorem 2.7.10. Let f be a function that is continuous on the closed interval [a, b] and
differentiable on the open interval (a, b).

1. If f 0 (x) > 0 for all x ∈ (a, b), then f is increasing on [a, b].

2. If f 0 (x) < 0 for all x ∈ (a, b), then f is decreasing on [a, b].

3. If f 0 (x) = 0 for all x ∈ (a, b), then f is constant on [a, b].

Example 2.7.11.
2.7. RELATIVE EXTREMA OF A FUNCTION 125

1. The derivative of f (x) = x2 is f 0 (x) = 2x. Note that f 0 (x) < 0 for all x ∈ (−∞, 0) and f 0 (x) > 0
for all x ∈ (0, ∞). Indeed, as seen in Figure 2.7.4, the graph of f is decreasing on the interval
(−∞, 0] and is increasing on the interval [0, ∞).

2. If f (x) = x3 , then f 0 (x) = 3x2 which is never negative. Observe the graph of f (x) = x3 in
Figure 2.7.5. Indeed it is always increasing.

y = x2 y = x3

Figure 2.7.4 Figure 2.7.5

2.7.4 The First Derivative Test for Relative Extrema

ic s
Note that if we want to reach the peak of a hill, we first have to climb up a slope and when we

at
finally reach the top, we would go downhill. Analogously, when a continuous function y = f (x)

em
attains a relative extremum, say a relative maximum, the behavior of the graph of f transitions
h
from increasing to decreasing as it proceeds from the left of the point to its right.
at
M

Theorem 2.7.12 (First Derivative Test for Relative Extrema). Let f be a function continuous
of

on the open interval (a, b) which contains the number c. Suppose that f is also differentiable
e

on the interval (a, b), except possibly at c.

ut

1. If f 0 (x) > 0 for all x ∈ (a, c) and f 0 (x) < 0 for all x ∈ (c, b), then f has a relative
s tit

maximum at x = c.
In

2. If f 0 (x) < 0 for all x ∈ (a, c) and f 0 (x) > 0 for all x ∈ (c, b), then f has a relative
UP

minimum at x = c.

3. If f 0 (x) does not change signs from (a, c) to (c, b), then there is no relative extremum at
x = c.

We will use Theorem 2.7.12 to verify whether a function has relative extrema at its critical numbers.
We enumerate below some suggested steps in finding the relative extrema of a given function.
Locating relative extremum values of a function

• Determine the critical numbers of f .

• Determine the sign of f 0 on the left and right of each critical number. (Note that in creating
the table of signs of f 0 , one should consider all values of x that make f 0 either zero or undefined,
regardless of whether they are critical numbers or not.)

• Conclude accordingly using the First Derivative Test.

126 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

Example 2.7.13. Determine the intervals where the given function is increasing or decreasing and
find all of its relative extrema.
x3
1. f (x) = − 3x
9
Solution.
x2 − 9
From Example 2.7.4.1, we know f 0 (x) = and the critical numbers of f are ±3.
3
−3 3
f 0 (x) + − +
Then f is increasing on (−∞, −3] and on [3, ∞), while it is decreasing on [−3, 3]. By the first
derivative test, (−3, 6) is a relative maximum point of the graph of f and (3, −6) is a relative
minimum point of the graph of f .

2. f (x) = x4 + 2x3

Solution.

s
ic
From Example 2.7.4.2, f 0 (x) = 4x3 + 6x2 = 2x2 (2x + 3) and the critical numbers of f are 0 and

at
− 23 .

0
f (x) −
− 32
+
0
+
h em
at
M

Hence f is decreasing on (−∞, − 23 ] and increasing on [− 32 , ∞). Therefore, f has a relative

minimum at x = − 23 . It is good to observe that 0 is a critical number of f yet f does not have
of

a relative extremum at 0.
e
ut

x2
3. f (x) =
tit

9 − x2
s
In

Solution.
18x
From Example 2.7.4.3, f 0 (x) = and the only critical number of f is 0.
UP

(9 − x2 )2

−3 0 3
999

999

f 0 (x) − − + +

The graph of f is decreasing on (−∞, −3) and on (−3, 0]. It is increasing on [0, 3) and on (3, ∞).
Therefore, f has a relative minimum at x = 0.

4. f (x) = −x4/3 + 4x1/3

Solution.
−4(x − 1)
From Example 2.7.4.4, f 0 (x) = and the critical numbers of f are 0 and 1.
3x2/3
0 1
f 0 (x) + + −
Then the graph of f is increasing on (−∞, 1] and decreasing on [1, ∞). We conclude that f has
a relative maximum at x = 1.
2.7. RELATIVE EXTREMA OF A FUNCTION 127

x2 + 4x + 3
5. f (x) =
2(x − 1)2
Solution.
−(3x + 5)
Differentiating and simplifying, we get f 0 (x) = . Note that the only critical number
(x − 1)3
5
of f is − . We form the following table:
3

− 53 1

999
f 0 (x) − + −

We conclude that f is increasing on [− 35 , 1); decreasing on the intervals (−∞, − 53 ] and (1, ∞);
and has a relative minimum at x = − 35 . Notice that even if there was a change of sign at x = 1,
we don’t say that f has a relative maximum at x = 1 since f is undefined when x = 1. In the
first place, 1 is not a critical number of f .

ic s
at
h em
at
M
of
e
ut
tit
s
In
UP
128 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

2.7.5 Exercises
Exercises for Disccussion

A. Determine the intervals on which the graph of the following functions is increasing or decreas-
ing. Use the first derivative test to find its relative extremum points.

√
1. f (x) = 1 − 4x − x2 6. f (x) = x+2 3

2. f (x) = x2 (x − 1)3 √
7. f (x) = 2x 3 − x
3. f (x) = x2/3 (x − 1)2 x
8. f (x) = 2 − sin x
x2
4. f (x) = e2x + 16e−x
x2 + 2 9. f (x) =
√ 2
1+ x
5. f (x) = √ 10. f (x) = ln(x2 + 1)
1− x

B. Given the graph of the derivative of f and assuming f is continuous everywhere, find the

ic s
intervals where f is increasing, decreasiing and the x-coordinates of the relative extrema of

at
f.
h em
at
1. 2.
M

2
of

3
e

(0, 2) 1
ut

2
(−2, 0) (2, 0)
tit

−2 −1 1 2 3
s

1
In

−1
UP

−2 −1(−1, 0) 1 2 −2

−1 −3

C. Do as indicated.

R


1
1. Suppose f (x) in increasing on (0, 1]. Show that f is decreasing on .
x2 + 1
2. The population P of a new species of frogs in a conservation facility is modelled by

2500t2
P (t) = 50 + ,
25 + t2
where t is the number of years after the introduction of the new species.
a. How many frogs will populate the area in the long run?
b. Describe the behavior of the population. Is this a good model? Why or why not?
2.7. RELATIVE EXTREMA OF A FUNCTION 129

Supplementary Exercises

A. Determine the intervals on which the graph of the following functions is increasing or decreas-
ing. Use the first derivative test to find its relative extremum points.

1. f (x) = 4x3 + 3x2 − 18x

(
x2 − x − 2, if x < 2
6. f (x) =
2. f (x) = x3 − 9x2 + 24x − 16 (x − 2)3 , if x ≥ 2
3. f (x) = x3
− 12x2
+ 21x
2 7. f (x) = sin2 x − 2 cos x
x +4 √
4. f (x) =
x 8. f (x) = 2x − 2 cos x
x
5. f (x) = 9. f (x) = ln cosh x
(3x + 1)2

B. Given the graph of the derivative of f and assuming f is continuous everywhere, find the
intervals where f is increasing, decreasiing and the x-coordinates of the relative extrema of
f.

ic s
at
1 2

2 em
h 2 (1, 32 )
at
(0, 1)
M

1 (− 21 , 34 ) 1
of

(−3, 0) (−1, 0) (2, 0)

e

−3 −2 −1 1 2 3
ut

(0, −1) −4 −3 −2 −1 (0, 0)1 2

−1
tit

−1
(−2, −1)
s
In

−2
(2, −2) −2
UP

C. Do as indicated.
 
1
1. Show that if f is decreasing on (0, +∞), the g(x) = f is increasing on (0, +∞).
x3
2. Use a first derivative to show that sin x < x for all x > 0.
3. Find a polynomial with critical numbers 0 and 3 with relative maximum at x = 0 and no
relative extremum at x = 3.
4. Find the value of c such that f (x) = x + ln(x2 + 1) + c tan−1 x has a relative maximum at
x = −5 and a relative minimum at x = 3.
130 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

2.8 Concavity and the Second Derivative Test

The shape of the graph of a function provides many useful insights on the physical quantity that
it represents. Thus, we study the concavity of a graph.

At the end of this section, the student will be able to:

• interpret the notion of concavity, and points of inflection of a function graphically

• determine if a function is concave up or concave down on an interval using the

second derivative

• find the relative extrema using the Second Derivative Test

2.8.1 Concavity

ic s
Definition 2.8.1.

at
1. The graph of a function f is said to be concave up at the point P (c, f (c)) if f 0 (c) exists and
em
there is an open interval I containing c such that for all x ∈ I \ {c}, the point (x, f (x)) is
h
at
above the tangent line to the graph of f at P . We say that the graph of f is concave up on
M

an interval I if it is concave up at (c, f (c)) for all c ∈ I.

of

2. The graph of a function f is said to be concave down at the point P (c, f (c)) if f 0 (c) exists
e

and there is an open interval I containing c such that for all x ∈ I \ {c}, the point (x, f (x))
ut

is below the tangent line to the graph of f at P . We say that the graph of f is concave down
tit

on an interval I if it is concave down at (c, f (c)) for all c ∈ I.

s
In

Example 2.8.2. Let f be a function whose graph is given below.

UP

c1 c2 c3 c4 c5 c6 c7

Figure 2.8.1

• The graph of f is concave up at the point P .

• The graph of f is concave up on (c1 , c2 ) ∪ (c2 , c3 ) ∪ (c3 , c5 ) and concave down on (c5 , c7 ).

• The graph of f is neither concave up nor concave down at the point (c5 , f (c5 )).
2.8. CONCAVITY AND THE SECOND DERIVATIVE TEST 131

• Concavity of the graph of f at x = c1 and x = c3 cannot be determined because f 0 (c1 ) and

f 0 (c3 ) are not defined.

• Concavity of the graph of f at x = c2 and x = c7 cannot be determined because f is not defined

at these values of x.

We wish to find another characterization of concavity aside from its graphical definition. Observe
the behavior of the tangent lines to the concave up curve in Figure 2.8.2a. As a point goes from
left to right, the slope increases. Meanwhile, the slope of the tangent line decreases as a point goes
from left to right along the concave down curve in Figure 2.8.2b. We shall state this as a theorem.

Figure 2.8.2a Figure 2.8.2b

s
Recall that f 00 (x) = Dx (f 0 (x)). Hence if f 00 (x) > 0 for all x ∈ (a, b), then f 0 (x) is increasing on

ic
at
(a, b) and the graph of f is concave up on (a, b). If f 00 (x) < 0, then f 0 (x) is decreasing on (a, b) and

em
the graph of f is concave down on (a, b).
h
at
Theorem 2.8.3 (Test for Concavity). Let f be a function such that f 00 exists on (a, b).
M

1. If f 00 (x) > 0 for all x ∈ (a, b), then the graph of f is concave up on (a, b).
of
e

2. If f 00 (x) < 0 for all x ∈ (a, b), then the graph of f is concave down on (a, b).
ut
tit

2.8.2 Point of Inflection

s
In

Example 2.8.4. If f (x) = x3 , then f 00 (x) = 6x. Note that f 00 (x) > 0 when x > 0 and f 00 (x) < 0
UP

when x < 0. Hence, the graph of f is concave up on (0, ∞) and concave down on (−∞, 0) (refer to
Figure 2.8.3a). Observe that that the graph of f changed concavity at x = 0. In this case, we call
the point (0, f (0)) = (0, 0) a point of inflection.
g(x) = x2/3
f (x) = x3 √
3
f (x) = x
f (x) = x4

Figure 2.8.3a Figure 2.8.3b Figure 2.8.3c Figure 2.8.3d

Definition 2.8.5. The graph of f (x) has a point of inflection at P (c, f (c)) if f is continuous at
x = c and the graph of f changes concavity at P , i.e. there is an open interval (a, b) containing c
such that
1. f 00 (x) > 0 for all x ∈ (a, c) and f 00 (x) < 0 for all x ∈ (c, b) or;

2. f 00 (x) < 0 for all x ∈ (a, c) and f 00 (x) > 0 for all x ∈ (c, b).
132 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

Theorem 2.8.6. If P (c, f (c)) is a point of inflection of the graph of the function f , then
f 00 (c) = 0 or f 00 (c) does not exist.

Example 2.8.7.

1. The graph of f (x) = x3 has a point of inflection when x = 0. Indeed, f 00 (0) = 0.

√
2. The graph of f (x) = 3 x (refer to Figure 2.8.3b) has a point of inflection at x = 0. Note that
2
f 00 (x) = − 5/3 . Indeed, f 00 (0) is undefined.
9x
Remark 2.8.8. The converse of the preceding theorem is not true. That is, if f 00 (c) = 0 or f 00 (c)
dne, the graph of f may or may not have a point of inflection at x = c. Take for example the
graphs of f (x) = x4 and g(x) = x2/3 illustrated in Figures 2.8.3c and 2.8.3d. The graph of both
functions have no points of inflection but f 00 (x) = 0 and g 00 (x) does not exist when x = 0. From
now on, we will refer to the values of x in the domain of f that satisfy either f 00 (x) = 0 or f 00 (x)

s
does not exist, as possible points of inflection.

ic
at
Example 2.8.9. Determine all points of inflection of the graph of the following functions and
discuss their concavity. hem
at
x3
1. f (x) = − 3x
M

9
of

Solution.
2x
e

Note that f 00 (x) = = 0 when x = 0.

ut

3
tit

0
s
In

f 00 (x) − +
UP

Then the graph of f has a point of inflection at (0, 0) and the graph of f is concave down on
(−∞, 0) and concave up on (0, ∞).

2. f (x) = x4 + 2x3

Solution.
Observe that f 00 (x) = 12x2 + 12x = 12x(x + 1) = 0 when x = 0 or x = −1.

−1 0
f 00 (x) + − +

Then (−1, −1) and (0, 0) are points of inflection of the graph f . The graph of f is concave up
on (−∞, −1) and (0, ∞). The graph of f is concave down on (−1, 0).

x2
3. f (x) =
9 − x2
Solution.
2.8. CONCAVITY AND THE SECOND DERIVATIVE TEST 133

(9 − x2 )2 18 − 18x(2)(9 − x2 )(−2x) 54(3 + x2 )

Note that f 00 (x) = = which is never zero and is
(9 − x2 )4 (9 − x2 )3
undefined when x = ±3.
−3 3

999

999
f 00 (x) − + −

Note that the graph of f cannot have points of inflection at x = ±3 because ±3 ∈

/ dom f . Also,
the graph of f is concave up on (−∞, −3) and (3, ∞), while f is concave down on (−3, 3).

4. f (x) = −x4/3 + 4x1/3

Solution.
−4(x + 2)
Verify that f 00 (x) = = 0 when x = −2 and f 00 (x) is undefined when x = 0.
9x5/3

−2 0
f 00 (x) − + −
√

ic s
Then (−2, 2 3 2) and (0, 0) are points of inflection of the graph of f . The graph of f is concave

at
down on (−∞, −2) and (0, ∞) and concave up on (−2, 0).

2.8.3
h
The Second Derivative Test for Relative Extremaem
at
M

The following is another test for relative extrema that uses the second derivative of a function.
Compare this test with the first derivative test for relative extrema.
of
e
ut

Theorem 2.8.10 (Second Derivative Test for Relative Extrema). Let f be a function such
tit

that f 0 and f 00 exist for all values of x on some open interval containing x = c and f 0 (c) = 0.
s
In

1. If f 00 (c) < 0, then f has a relative maximum value at x = c.

UP

2. If f 00 (c) > 0, then f has a relative minimum value at x = c.

3. If f 00 (x) = 0, we have no conclusion (f may or may not have a relative extrema at x = c).

Remark 2.8.11. Note that the second derivative test is only applicable to the first type of critical
numbers: those that make the first derivative zero (x-coordinates of so-called stationary points).

Example 2.8.12. Given the following functions, determine if the second derivative test is appli-
cable to their critical numbers. If yes, what can you conclude using the second derivative test?

x3
1. f (x) = − 3x with C.N. : −3, 3
9
Solution.
x2 − 9 2x
Recall that f 0 (x) = and f 00 (x) = . The second derivative test is applicable to both
3 00 3
−3 and 3. Observe that f (−3) = −2 < 0 and thus, f has a relative maximum at x = −3.
Similarly, f 00 (3) = 2 > 0 and thus, f has a relative minimum at x = 3.
134 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

2. f (x) = x4 + 2x3 with C.N. : 0, − 32

Solution.
Recall that f 0 (x) = 4x3 + 6x2 and f 00 (x) = 12x2 + 12x. The second derivative test is applicable
to both 0 and − 23 . Note that f 00 (0) = 0 and this gives us an inconclusive case. On the other
hand, f 00 − 32 = 9 > 0 and so f has a relative minimum at x = − 32 .

x2
3. f (x) = with C.N. : 0
9 − x2
Solution.
18x 2
We have seen that f 0 (x) = and f 00 (x) = 54(3 + x ) . The second derivative test is
(9 − x2 )2 (9 − x2 )3
54(3)
applicable to 0. Since f 00 (0) = 93 > 0, f has a relative minimum at x = 0.

4. f (x) = −x4/3 + 4x1/3 with C.N. : 0, 1

Solution.

ic s
−4(x − 1) 00 (x) = −4(x + 2) . The second derivative test is applicable to 1 but
f 0 (x) = and f

at
3x2/3 9x5/3

em
not to 0. Because f 00 (1) = − 43 , then f has a relative maximum at x = 1.
b
h
5. f (x) = ax2 + bx + c, where a 6= 0. C.N. : − 2a .
at
M

Solution.
of

Note that f 0 (x) = 2ax+b and f 00 (x) = 2a. The second derivative test is applicable to − 2a
b
. From
e

00 b b
f (− 2a ) = 2a, we conclude that if a > 0, f has a relative minimum at x = − 2a . Meanwhile,
ut

b
a < 0, then f has a relative maximum at x = − 2a .
s tit
In
UP
2.8. CONCAVITY AND THE SECOND DERIVATIVE TEST 135

2.8.4 Exercises
Exercises for Discussion

A. Identify the possible points of inflection of the graph of each function. Determine which of
these are indeed points of inflection using a table of signs. Determine the intervals on which
the graph of the function is concave up or concave down.

1. f (x) = (x + 2)3 x
5. f (x) = √
x2
+1
2. f (x) = x2/3 √ 1
x 6. f (x) = x + √
3. f (x) = 2 x
x +2 7. f (x) = sin x cos x
x2 − 3 sin x
4. f (x) = 8. f (x) =
x2 + 1 2 − cos x

B. Determine which critical numbers satisfy the hypotheses of the second derivative test and

ic s
apply the test to these numbers.

at
em
1. f (x) = (x2 − 1)4 h 4. f (x) = 2x + cot x
2. f (x) = x1/3 (x + 4) 5. f (x) = cos2 x − 2 sin x
at
√
M

3. f (x) = x4/3 − 6x1/3 6. f (x) = 3x + 2 sin x

of

C. Suppose that water is flowing at a constant rate into the container below. If f (t) is the water
e
ut

level (height) in the container after time t, how would you describe the graph of y = f (t)?
tit
s
In

1. 2.
UP

Supplementary Exercises

A. Identify the possible points of inflection of the graph of each function. Determine which of
these are indeed points of inflection using a table of signs. Determine the intervals on which
the graph of the function is concave up or concave down.

1. f (x) = −x4 + 10x3 − 36x2 − 90x + 200

(
9−x , if x ≤ 3
4. f (x) =
2. f (x) = x2 (2x2 − 4x + 3) x2 − 3 , if x > 3
√
3. f (x) = (x + 2)5 + x4 + 8x3 + 24x2 5. f (x) = x2 + 2x + x
136 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

1 − 2x2 e2x 7. f (x) = ln(x3 − 1)

6. f (x) =
e2x 8. f (x) = 2x2 + ln x

B. Determine which critical numbers satisfy the hypotheses of the second derivative test and
apply the test to these numbers.

√ √
16000 3
4. f (x) = 2 3 x − x2
1. f (x) = x2 +
x
  5. f (x) = cos(π sin x)
200
2. f (x) = (x − 4) −2 2
x 6. f (x) = e−x
√
3. f (x) = x − 2 x + 1 + 1 7. f (x) = x − ln sech x

C. Given the graph of the derivative of f and assuming f is continuous everywhere, find the
intervals where f is concave up, concave down and the x-coordinates of the points of inflection
of f .

ic s
at
1 2

2
h em (0, 0) (3, 0)
at
−2 −1 1 2 3
M

(0, 1)
1 −1
of

(−1, 0) −2
e

−3
ut

−3 −2 −1 1 2 3
(0, −1)
tit

−1 −4
(−2, −4)
s

−5
In

−2
(2, −2) −6
UP

D. Do as indicated.

1. If f is a function concave down everywhere, show that e−f (x) is concave up everywhere.
2. Suppose n is a positive integer such that n ≥ 2. Let g(x) = xn+1 − (n + 1)xn .
a. Prove that g has a relative minimum at x = n.
b. Explain why the Second Derivative Test cannot be used to show that g has a relative
maximum at x = 0 when n is even.
c. Show that g has a point of inflection at x = n − 1 and at x = 0 when n is odd.
2.9. GRAPH SKETCHING 137

2.9 Graph Sketching

We have laid out the tools we need to analyze the behavior of a function. We are now ready to
sketch the graph of functions using its derivatives.

At the end of this section, the student will be able to:

• find the vertical, horizontal, and oblique asymptotes of a function

• graph polynomial functions (of degree at most 5) and rational functions

Here are some guidelines we can follow.

Guidelines for Graphing a Function f Analytically:

1. Find the domain and intercepts of f .

ic s
2. Determine vertical, horizontal and oblique asymptotes of f .

at
em
3. Compute f 0 (x) and f 00 (x), and determine the critical numbers.
h
4. Divide the real number line using the numbers that make f 0 zero or undefined and the
at
numbers that make f 00 zero or undefined. Determine the sign of f 0 (x) and f 00 (x) on each
M

interval. It would be easier if you make a table(s) of signs.

of
e

5. Determine intervals on which the function is increasing,decreasing, concave up, concave

ut

down. Determine the coordinates of the relative extremum points and points of inflection.
stit

6. Plot important points (intercepts, holes, extrema, points of inflection) and asymptotes
In

first.
UP

7. Use the table of signs of f 0 and f 00 to graph the rest of the function.

2.9.1 Graphing Polynomial Funtions

x3
Example 2.9.1. Sketch the graph of f (x) = − 3x.
9

Solution.

domf : R √
interval f f0 f 00 conclusion
x-intercept: x = 0, ±3 3 (−∞, −3) + - inc., cd
−3 6 0 - rel. max.
y-intercept: y = 0
(−3, 0) - - dec, cd
x2 − 9
f 0 (x) = CN: −3, 3 0 0 - 0 POI
3 (0, 3) - + dec, cu
2x
f 00 (x) = PPOI : 0 3 −6 0 + rel. min.
3
(3, ∞) + + inc, cu
138 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

(−3, 6)

√ √
(−3 3, 0) (3 3, 0)
(0, 0)

(3, −6)

Example 2.9.2. Sketch the graph of f (x) = x4 + 2x3 using derivatives.

Solution.

s
R

ic
domf : f 0 (x) = 4x3 + 6x2 CN: 0, − 23

at
x-intercept: x = 0, −2 f 00 (x) = 12x2 + 12x PPOI: 0, −1
y-intercept: y = 0 h em
at
interval f f0 f 00 conclusion
M

(−∞, − 32 ) - + dec., cu
of

−27
− 32 16 0 + rel. min. (0, 0)
e

(− 32 , −1) + + inc, cu
ut

−1 −1 + 0 POI
tit

(−1, −1)
(−1, 0) + - inc, cd
s
In

0 0 0 0 POI
( −3 , −27
)
UP

(0, ∞) + + inc, cu 2 16

2.9.2 Review of Asymptotes

In graphing functions accurately, it is also important to take note of its asymptotes. In this course,
we will discuss linear asymptotes of a function. The formal definition of an asymptote involves
limits.

Definition 2.9.3. The line x = a is a vertical asymptote of the graph of f (x) if at least one of the
following is true:

(a) lim f (x) = ∞ x=a (b) lim f (x) = −∞ x=a

x→a− x→a−
2.9. GRAPH SKETCHING 139

(c) lim f (x) = ∞ x=a (d) lim f (x) = −∞ x=a

x→a+ x→a+

Remark 2.9.4. To look for the vertical asymptotes of the graph of a function, we have to think
of a real number a such that the limit of f as x approaches a is infinite. For a rational polynomial
P (x)
, x = a is a vertical asymptote if (x − a) is a factor of Q but not of P (or the multiplicity of
Q(x)
x − a as a factor of Q is greater than the multiplicity of x − a as a factor of P ).
Example 2.9.5.
x2 − 4 (x − 2)(x + 2)
1. f (x) = = has a vertical asymptote at x = 3 but has none at x = 2.
x2 − 5x + 6 (x − 3)(x − 2)

s
To verify this using limits,

ic
at
(x − 2)(x + 2) (x + 2)

em
lim = lim = −4
x→2 (x − 3)(x − 2) x→2 (x − 3)
h
at
   
(x − 2)(x + 2) 1·5 (x − 2)(x + 2) 1·5
M

lim = −∞ lim =∞
x→3− (x − 3)(x − 2) (0− ) · 1 x→3+ (x − 3)(x − 2) (0+ ) · 1
of
e

4 − x2 (2 − x)(2 + x)
ut

2. f (x) = = has a vertical asymptote at x = 2.

x2 − 4x + 4 (x − 2)2
tit

 
(2 − x)(2 + x) −(x + 2) −4
s
In

lim = lim = ±∞
x→2∓ (x − 2)2 x→2∓ (x − 2) 0∓
UP

Definition 2.9.6. The line y = b is a horizontal asymptote of the graph of f if at least one of the
following is true:

y=b
(a) lim f (x) = b
x→∞
y=b

y=b
(b) lim f (x) = b
x→−∞
y=b

Remark 2.9.7. To find all horizontal asymptotes of a polynomial or rational function, we only
need to evaluate the limit of the function as x → ∞ and as x → −∞. If the limit is finite, then there
is a horizontal asymptote. Notice that in a rational function, whenever the degree of the numerator
is less than or equal to the degree of the denominator, then there is a horizontal asymptote.
140 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

Example 2.9.8.
x+3
1. f (x) = has a horizontal asymptote at y = 0.
x2 − 1
1 1 3 1 1 3
x+3 x2 x + x2 x+3 x2 x + x2
lim · 1 = lim 1 =0 lim · 1 = lim 1 =0
x→∞ x2 − 1 x→∞ 1− x→∞ x2 − 1 x→∞ 1−
x2 x2 x2 x2

2x + 1 2 2
2. f (x) = has a horizontal asymptote at y = and y = − .
|5x − 2| 5 5
2x + 1 2x + 1 2 2x + 1 2x + 1 2
lim = lim = lim = lim =−
x→∞ |5x − 2| x→∞ 5x − 2 5 x→−∞ |5x − 2| x→−∞ −5x + 2 5

Definition 2.9.9. The graph of f has the line y = mx + b, m 6= 0, as an oblique asymptote if at

least one of the following is true:

(a) lim f (x) − (mx + b) = 0

x→∞

s
y = mx + b

ic
y = mx + b y = mx + b y = mx + b

at
hem
at
M
of

(b) lim f (x) − (mx + b) = 0.

x→−∞
e
ut
stit
In
UP

y = mx + b
y = mx + b y = mx + b y = mx + b

P (x)
Remark 2.9.10. If P (x) and Q(x) are polynomial functions with deg(P ) = deg(Q) + 1, then
Q(x)
has an oblique asymptote.
x2 + 3
Example 2.9.11. The rational function f (x) = has an oblique asymptote because the
x−1
degree of the numerator is 2, which is one degree greater than that of the denominator. To find
the equation of the oblique asymptote, we use long division.

x+1 Thus
4
x2


x−1 +3 f (x) = (x + 1) + .
x−1
− x2 + x
From here it can be shown that y = x + 1 is an oblique asymptote
x+3
since
−x+1 4
lim f (x) − (x + 1) = lim = 0.
x→±∞ x→±∞ x + 1
4
2.9. GRAPH SKETCHING 141

2.9.3 Graphing Rational Functions

x2
Example 2.9.12. Sketch the graph of f (x) = using derivatives.
9 − x2

R
Solution.
domf : \ {±3} interval f f0 f 00 conclusion
x-int: x = 0 y-int: y = 0 (−∞, −3) - - dec., cd
−3 und und und V.A.
Vertical Asymptotes: x =  −3and x = 3 (−3, 0) - + dec, cu
9
lim f (x) = −∞ − 0 0 0 + rel. min.
 0
x→−3 −
(0, 3) + + inc, cu
9
lim f (x) = ∞ +
3 und und und V.A.
x→−3+  0 (3, ∞) + - inc, cd
9
lim f (x) = ∞
x→3− 0+ 
x2

9 f (x) =
lim f (x) = −∞ 9 − x2
x→3+ 0−
Horizontal Asymptotes: y = −1 x = −3 x=3
x2 1 (0, 0)

s
lim = lim 9 = −1

ic
x→±∞ 9 − x2 x→±∞ 2 − 1 y = −1
x

at
Oblique Asymptotes: none

em
18x
f 0 (x) = CN: 0
(9 − x2 )2 h
54(3 + x2 )
at
f 00 (x) = PPOI: none
(9 − x2 )3
M

4x
of

Example 2.9.13. Sketch the graph of f (x) = using derivatives.

x2 +1
e
ut

Solution.
domf : R
tit

x-int: x = 0 interval f f0 f 00 conclusion

s

√
In

(−∞, − 3) - - dec., cd
y-int: y = 0 √ √
− 3 − 3 - 0 POI
√
UP

Vertical Asymptote.: none (− 3, −1) - + dec, cu

Horizontal Asymptote: y = 0 −1 −2 0 + rel. min.
4
x 0 (−1, 0) + + inc, cu
lim = =0
1
x→±∞ 1 + x12 0 0 + 0 POI
Oblique Asymptote: none (0, 1) + - inc, cd
4 − 4x2 1
√
2 0 - rel. max.
f 0 (x) = 2 CN: −1, 1 (1, 3) - - dec, cd
(x + 1)2 √ √
8x(x2 − 3) √ 3 3 - 0 POI
√
f 00 (x) = 2 3
PPOI: 0, ± 3 ( 3, ∞) - + dec, cu
(x + 1)

(1, 2) √ √
( 3, 3)

(0, 0)
y=0

4x
√ √ f (x) =
(− 3, − 3) x2 + 1
(−1, −2)
142 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

x−1
Example 2.9.14. Sketch the graph of f (x) = 4 + x + using derivatives.
x2 − 1

Solution.

R
domf : \ {−1, 1} √ x = −1

x-intercept: x = −5±2 5
1 x2 + 5x + 5
f (x) = 4 + x + = , x 6= 1 x−1
x+1 x+1 f (x) = 4 + x +
y-intercept: y = 5   x2 − 1
11 (1, 11
)
V.A.: x = −1 hole: 1, 2
2
H.A.: none y =x+4
(0, 5)
lim f (x) = ∞ lim f (x) = −∞
x→∞ x→−∞
O.A.: y = x + 4
x2 + 2x
f 0 (x) = CN: 0, −2
(x + 1)2 (−2,1)
2 √

s
f 00 (x) = PPOI: none ( −5− 5

ic
2
, 0)
(x + 1)3 √

at
( −5+
2
5
, 0)

em
0 00
interval f f f conclusion
(−∞, −2) + - inc, cd
h
at
−2 1 0 - rel. max.
M

(−2, −1) - - dec, cd

−1 und und und V.A.
of

(−1, 0) - + dec, cu
e

0 5 0 + rel. min.
ut

(0, ∞) + + inc, cu
s tit

2.9.4 The Graph of f from the Graph of f 0

In

We know that we can gain insight on the shape of a function f if we know the behavior of its
UP

derivative. From the graph of the derivative of f , what can we deduce about the graph of f ?
Example 2.9.15. Given the graph of f 0 below and assuming that f is continuous everywhere,
sketch a possible graph of f .

−1 1 2

−1

Solution.
2.9. GRAPH SKETCHING 143

(a) First, we need to determine the critical numbers of f . Remember that these are the numbers
that make f 0 zero or undefined. From the graph of f 0 , we see that the critical numbers are −1,
0 and 2.

(b) Next, we need to determine where the graph of f has points of inflection. These are the points
where the graph of f 0 changes from increasing to decreasing, or vice versa. The possible points
of inflection of the graph of f are attained at values of x where the graph of f 0 has a horizontal
tangent line or a vertical tangent line or has no tangent line. Thus, the x-coordinates of the
possible points of inflection of f are 0 and 1.

(c) Recall that the graph of f is increasing on [a, b] if f 0 (x) > 0 for all x ∈ (a, b). This means that
the graph of f 0 is above the x-axis on the interval (a, b). Similarly, the graph of f is decreasing
on [a, b] if the graph of f 0 is below the x-axis on the interval (a, b).

(d) Meanwhile, the graph of f is concave up on an interval (a, b) if the graph of f 0 is increasing on
(a, b). It is concave down on (a, b) if the graph of f 0 is decreasing on (a, b).

f0 f 00

s
interval conclusions on the graph of f

ic
(−∞, −1) + − increasing, concave down

at
−1 0 − relative maximum
(−1, 0) − − h em
decreasing, concave down
at
0 und und relative minimum, POI
M

(0, 1) + + increasing, concave up

1 + 0 POI
of

(1, 2) + − increasing, concave down

e
ut

2 0 − relative maximum
tit

(2, ∞) − − decreasing, concave down

s
In

From this table, we can construct a possible graph of f .

UP

−1 0 1 2
144 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

2.9.5 Exercises
Exercises for Discussion

A. Explain why the following functions do not have a linear asymptote and sketch their graphs.

1. y = x3 − 3x2 + 1 3. y = x1/3 (x + 4)
2. y = −x4 + 8x3 − 18x2 + 10 4. y = sin x cos x

B. Find all vertical, horizontal and oblique asymptotes of the following functions using limits.

√
x2 + 1 x+1−1
1. f (x) = 4. f (x) =
x+1 x
sin x 2
x − 4x + 4
2. f (x) = 5. f (x) = 2
x x +x−6

s
x2 − 1 4

ic
3. f (x) = 2 6. f (x) = x + 1 +

at
5x + 1 x−4

C. Sketch the graph of the following functions analytically. h em

at
M

x x2 + 1
1. f (x) =
of

(2x + 1)2 4. f (x) =

x+1
e

12x2 x3 − 1
ut

2. f (x) = 5. f (x) = 2
(x − 2)2 x −1
tit

2x3 (x + 1)3
s

3. f (x) = 6. f (x) =
In

x2 − 4 x2
UP

D. Sketch a possible graph of the function f given the graph of f 0 . Assume that f is continuous
everywhere.

1. 2.

(−1, 2) 2 3
(0, 2.6)
2
1
(−2, 1) (0, 1)
√ 1
(−2, 0) (3 − 2, 0)
(−1, −0.6) (3, 0)
−3 −2 −1(0, 0) 1 2
−4 −3 −2 −1 1 2
−1
−1

−2 (1, −2) −2
2.9. GRAPH SKETCHING 145

Supplementary Exercises

A. Explain why the following functions do not have a linear asymptote and sketch their graphs.

1. y = 3x4 − 4x3 3. y = x4/3 − x1/3

2. y = x4 − 8x2 + 16 4. y = cosh x

B. Find all vertical, horizontal and oblique asymptotes of the following functions using limits.

2x + 3 1
1. f (x) = 3. f (x) =
x+4 |x − 3|
1
4. f (x) = ln(1 + )
x−9 x2
2. f (x) = √ 5. f (x) = coth x
x−3

s
C. Sketch the graph of the following functions analytically.

ic
at
em
1. f (x) = 4x3 + 12x2 (x + 1)3
5. f (x) =
h x2
(x − 1)2
at
2. f (x) = (x − 2)(3x + 2)
x2 6. f (x) =
M

x+1
4(2x + 3)(x + 3) (2x − 1)2
of

3. f (x) = 7. f (x) =
(x + 2)2 (x − 2)2
e
ut

x2 + 3x + 2 x3
4. f (x) = 8. f (x) =
tit

3x + 2 x2 − 4
s
In

D. Sketch a possible graph of the function f given the table of signs of f 0 and f 00 . Assume that
UP

f is continuous everywhere

1. 2.

x f 0 (x) f 00 (x) x f 0 (x) f 00 (x)

− − + −
x = −1 und und x=0 0 0
− + + +
x=0 0 + x=1 und und
+ + − +
x=1 + 0 x=4 0 +
+ − + +
146 CHAPTER 2. DERIVATIVES AND DIFFERENTIATION

s
ic
at
hem
at
M
of
e
ut
stit
In
UP