Calculus II (MATH 223)

Dr. Joseph K. Ansong

(Dept. of Mathematics, University of Ghana, Legon)

LECTURE NOTES
Math 223 : Calculus II J.K.A

2 Dr. Joseph K. Ansong

Contents

1 Introduction 1
1.1 Review of Calculus I . . . . . . . . . . . . . . . . . . . . . . . 1
1.1.1 Maximum & Minimum Values . . . . . . . . . . . . . . 1

2 The Mean Value Theorem and its Applications 9

2.1 The Mean Value Theorem (MVT) . . . . . . . . . . . . . . . . 9
2.2 Consequences of the MVT . . . . . . . . . . . . . . . . . . . . 15
2.2.1 Functions with Zero Derivatives . . . . . . . . . . . . . 15
2.2.2 Functions with Equal Derivatives . . . . . . . . . . . . 16
2.2.3 Increasing and Decreasing Functions . . . . . . . . . . 16
2.3 Applications of the MVT: Using the MVT to Establish an
Inequality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

3 Inverse Functions 25
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
3.2 Reflective Property of Inverse Functions . . . . . . . . . . . . 28
3.3 Existence of an Inverse Function . . . . . . . . . . . . . . . . . 29
3.3.1 Finding Inverse Functions . . . . . . . . . . . . . . . . 30
3.4 Continuity and Differentiability of Inverse Functions . . . . . . 31

4 Logarithmic and Exponential Functions 35

4.1 General Logarithmic Functions . . . . . . . . . . . . . . . . . 35
4.1.1 Laws of Logarithms . . . . . . . . . . . . . . . . . . . . 35
4.2 The Natural Logarithm . . . . . . . . . . . . . . . . . . . . . . 37
4.2.1 Laws of Natural Logarithms . . . . . . . . . . . . . . . 38
4.2.2 Derivatives of Logarithmic Functions . . . . . . . . . . 40
4.2.3 Logarithmic Differentiation . . . . . . . . . . . . . . . 42
4.2.4 The Natural Logarithm as a Limit . . . . . . . . . . . 43
4.3 General Exponential Functions . . . . . . . . . . . . . . . . . 44
4.3.1 Laws of Exponents . . . . . . . . . . . . . . . . . . . . 44
4.3.2 Derivative of Exponential functions . . . . . . . . . . . 46

3
Math 223 : Calculus II J.K.A

4.3.3 Natural Exponential functions . . . . . . . . . . . . . . 47

4.3.4 Derivatives of Exponential Functions . . . . . . . . . . 50
4.4 Logarithmic and Exponential Integration . . . . . . . . . . . . 51
4.5 Logarithmic Inequalities . . . . . . . . . . . . . . . . . . . . . 53

5 Indeterminate Forms and L’Hopital’s Rule 57

5.1 L’Hopital’s Rule . . . . . . . . . . . . . . . . . . . . . . . . . . 57
5.2 Indeterminate Forms ∞ − ∞ and 0 − ∞ . . . . . . . . . . . . 58
5.3 Indeterminate Forms of 00 , ∞0 , 1∞ . . . . . . . . . . . . . . . 59
5.4 Scales of Infinity(Order of Magnitude) . . . . . . . . . . . . . 59
5.4.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . 59
5.4.2 Relative Rates of Decay . . . . . . . . . . . . . . . . . 62

6 Hyperbolic Functions 65
6.1 Odd and Even Functions . . . . . . . . . . . . . . . . . . . . . 65
6.1.1 Basic Properties of Even and Odd Functions . . . . . . 66
6.1.2 Hyperbolic functions: Definitions . . . . . . . . . . . . 67
6.1.3 Graphs . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
6.2 Hyperbolic Identities and Osborn’s Rule . . . . . . . . . . . . 69
6.2.1 Relationship to the Exponential function . . . . . . . . 71
6.2.2 Osborn’s Rule . . . . . . . . . . . . . . . . . . . . . . . 72
6.3 Inverse Hyperbolic Functions . . . . . . . . . . . . . . . . . . . 72
6.3.1 Representation of Inverse Hyperbolic Functions as Log-
arithmic Functions . . . . . . . . . . . . . . . . . . . . 73
6.4 Derivatives of hyperbolic functions . . . . . . . . . . . . . . . 76
6.5 Derivatives of Inverse Hyperbolic Functions . . . . . . . . . . 77

7 Infinite Sequences 79
7.1 Definitions & Notation . . . . . . . . . . . . . . . . . . . . . . 79
7.1.1 Convergence & Divergence of a Sequence . . . . . . . . 81
7.1.2 Some Useful Theorems: . . . . . . . . . . . . . . . . . . 82
7.2 Bounded Monotonic Sequences . . . . . . . . . . . . . . . . . 87

8 Infinite Series 93
8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
8.1.1 Definitions & Notation . . . . . . . . . . . . . . . . . . 93
8.2 Convergence and Divergence of a Series . . . . . . . . . . . . . 94
8.3 Special Series . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
8.4 Some Useful Theorems . . . . . . . . . . . . . . . . . . . . . . 100
8.4.1 The Divergence Test . . . . . . . . . . . . . . . . . . . 100
8.5 Convergence of Positive-Term Series . . . . . . . . . . . . . . . 104

4 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

8.5.1 The Integral Test . . . . . . . . . . . . . . . . . . . . . 104

8.5.2 The Comparison Test . . . . . . . . . . . . . . . . . . . 108
8.5.3 The Limit Comparison Test . . . . . . . . . . . . . . . 109
8.6 Alternating Series . . . . . . . . . . . . . . . . . . . . . . . . . 111
8.7 Absolute Convergence, the Ratio & the Root Tests . . . . . . 113
8.7.1 The Ratio Test . . . . . . . . . . . . . . . . . . . . . . 114
8.7.2 The Root Test . . . . . . . . . . . . . . . . . . . . . . 116

9 Power Series 117

9.1 Power Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
9.1.1 Radius and Interval of Convergence . . . . . . . . . . . 117
9.2 Representation of Functions as Power Series . . . . . . . . . . 120
9.3 Taylor Series and the Taylor Polynomial . . . . . . . . . . . . 123
9.4 nth-degree Taylor Polynomial of f at a . . . . . . . . . . . . . 123

10 Integration 127
10.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
10.2 Fundamental Theorem of Calculus (FTC) . . . . . . . . . . . 128
10.2.1 Intuitive idea of the FTC . . . . . . . . . . . . . . . . . 128
10.2.2 The Mean Value Theorem (MVT) for integrals . . . . . 128
10.3 Riemann Sums & the Definite Integral: Integration as a Sum . 136
10.3.1 The Distance Problem . . . . . . . . . . . . . . . . . . 144
10.4 The Definite Integral . . . . . . . . . . . . . . . . . . . . . . . 145
10.5 Evaluating Definite Integrals . . . . . . . . . . . . . . . . . . . 151
10.6 Indefinite Integrals . . . . . . . . . . . . . . . . . . . . . . . . 154
10.6.1 Total change Theorem . . . . . . . . . . . . . . . . . . 155
10.7 Techniques of Integration . . . . . . . . . . . . . . . . . . . . . 156
10.7.1 Method of Substitution . . . . . . . . . . . . . . . . . . 156
10.7.2 Integrals of Symmetric Functions . . . . . . . . . . . . 161
10.8 Trigonometric Integrals . . . . . . . . . . . . . . . . . . . . . . 165
10.8.1 Integrating Powers of sin x and cos x . . . . . . . . . . 165
10.8.2 Integrals of Products of Trigonometric Functions . . . . 166
10.8.3 Integrals of Hyperbolic Functions . . . . . . . . . . . . 170
10.8.4 Trigonometric & Hyperbolic Substitutions . . . . . . . 171
10.9 Integrals of Rational Functions Using Partial Fractions . . . . 174
10.10The Reduction Formulae . . . . . . . . . . . . . . . . . . . . . 176
10.11Improper Integrals . . . . . . . . . . . . . . . . . . . . . . . . 178
10.11.1 Infinite Intervals of Integration . . . . . . . . . . . . . 178
10.11.2 Improper Integrals with Discontinuous Integrands . . . 180
10.12Applications of Integration . . . . . . . . . . . . . . . . . . . . 182
10.12.1 Area between curves . . . . . . . . . . . . . . . . . . . 182

5 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

10.12.2 Volumes . . . . . . . . . . . . . . . . . . . . . . . . . . 187

10.12.3 Volumes by Cylindrical Shells . . . . . . . . . . . . . . 189
10.12.4 Arc Length . . . . . . . . . . . . . . . . . . . . . . . . 191
10.12.5 Area of a Surface of Revolution . . . . . . . . . . . . . 193

11 Differential Equations 197

11.1 Separable equations . . . . . . . . . . . . . . . . . . . . . . . . 198
11.2 Linear Equations . . . . . . . . . . . . . . . . . . . . . . . . . 200

6 Dr. Joseph K. Ansong

Chapter 1

Introduction

1.1 Review of Calculus I

1.1.1 Maximum & Minimum Values
Definition 1. A function f has an absolute maximum (or global maximum)
at c if f (c) ≥ f (x) for all x in the domain D of f . Similarly, f has an absolute
minimum (global minimum) at c if f (c) ≤ f (x) for all x in D. The maximum
and minimum vlaues of f are called the extreme values of f .

Figure 1.1 illustrates these concepts.

Figure 1.1: Schematic plot showing the absolute maximum and minimum
positions.

1
Math 223 : Calculus II J.K.A

Definition 2. A function f has a local maximum (or relative maximum) at

c if f (c) ≥ f (x) for all x in some open interval containing c. Similarly, f has
a local minimum at c if f (c) ≤ f (x) for all x near c.

Figure 1.2: Schematic plot showing intervals of increase and decrease.

Example 1. See Figure 1.2. On the interval (a, c1 ), f (d1 ) is a local minimum
value; similarly f (c2 ) is a local minimum in the interval (d2 , d3 ). On (c1 , c2 ),
f (d2 ) is a local maximum.
Theorem 1 (The Extreme Value Theorem). If f is continuous on a
closed interval [a, b], then f attains an absolute maximum value f (c) for some
number c in [a, b] and an absolute minimum value f (d) for some number d
in [a, b].
NOTE: An extreme value can be taken on more than once, as illustrated
in Figure 1.3.

Figure 1.3: A schematic showing extreme values.

Theorem 2 (Fermat’s Theorem). If f has a local maximum or minimum

at c, and if f 0 (c) exists, then f 0 (c) = 0.

2 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

NOTE: When f 0 (c) = 0, f does not necessarily have a maximum or

minimum at c. For example, the function f (x) = x3 has f 0 (x) = 3x2 . Thus
f 0 (0) = 0, but f has no maximum or minimum value (see Figure 1.4).

Figure 1.4: Graph of y = x3 .

Definition 3 (Critical Number). A critical number of a function f is

a number c in the domain of f such that either f 0 (c) = 0 or f 0 (c) does not
exist.
Theorem 3 (Rolle’s Theorem). Let f be continuous on [a, b] and differ-
entiable on (a, b). If f (a) = f (b), then there exist at least one number c in
(a, b) such that f 0 (c) = 0.
An illustration is given in Figure 1.5.

Figure 1.5: Schematic illustrations of Rolle’s Theorem.

Proof. Figure 1.6 gives an illustration of the two cases involved in the proof
of the theorem.

3 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Figure 1.6: Schematic illustrations of Rolle’s Theorem.

Let f (a) = f (b) = d.

CASE 1:
f (x) = d for x in [a, b]. This implies that f 0 (x) = 0 for all x in [a, b]. Thus
f 0 (x) = 0 for some c in (a, b).
CASE 2:
f (x) 6= d for at least one x in [a, b]. So there must be a number x in (a, b)
where f (x) > d or f (x) < d. Supoose first that f (x) > d. Since f is
continuous on [a, b], then by the Extreme Value Theorem, f has an absolute
maximum value at some number c in [a, b]. The number c cannot be an end
point since f (a) = f (b) = d and we have assumed that f (x) > d for some x
in (a, b). Thus, c must lie in (a, b). Since f is differentiable on (a, b), f 0 (x)
exists and so by Fermat’s Theorem f 0 (c) = 0. The case f (x) < d can be
proved in a similar manner.
Example 2. Verify that the function satisfies the hypotheses of Rolle’s The-
orem (RT) on the given interval, and find all values of c that satisfy the
conclusion of the theorem
(1) f (x) = x2 − 4x + 3; [1, 3]
(2) f (x) = x3 − x; [−1, 1]
(3) f (x) = x3 − 9x; [−3, 3]
Solution. (1) f (x) = x2 − 4x + 3; [1, 3].
f (1) = 1 − 4 + 3 = 0
f (3) = 9 − 12 + 3 = 0
The polynomial f (x) is continuous and differentiable on [1, 3]. There-
fore the hypotheses of Rolle’s Theorem are satisfied.
f 0 (x) = 2x − 4

4 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

f 0 (c) = 0 =⇒ 2c − 4 = 0
=⇒ c = 2.
Sketch:

f (x) = x2 − x − 3x + 3 = x(x − 1) − 3(x − 1)

= (x − 1)(x − 3)
So the zeros are x = 1 and x = 3. See Figure 1.7a.

(2) f (x) = x3 − x; [−1, 1]

f (x) is a polynomial so it’s continuous and differentiable on [−1, 1].

f (−1) = −1 + 1 = 0
f (1) = 1 − 1 = 0
=⇒ f (−1) = f (1)

Thus, Rolle’s Theorem is satisfied.

f 0 (x) = 3x2 − 1
f 0 (c) = 0 =⇒ 3c2 − 1 = 0
1
=⇒ c2 =
√3
1 3
=⇒ c = ± √ = ± . (1.1)
3 3

A graph of f (x) = x(x2 − 1) = x(x − 1)(x + 1) is shown in Figure 1.7b.

Example 3 (Real life example). During a test dive of a prototype of a

submarine, the depth in meters of the submarine at time t in minutes is given
by
h(t) = t3 (t − 7)4
where 0 ≤ t ≤ 7.

(a) Use Rolle’s Theorem to show that there is some instant of time t = c
between 0 and 7 when h0 (c) = 0.

(b) Find the number c and interpret your results.

Solution.
h(t) = t3 (t − 7)4 ; 0 ≤ t ≤ 7.

5 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Figure 1.7: Plot of (a) f (x) = x2 − 4x + 3 and (b) f (x) = x3 − x.

(a) The polynomial function h(t) is continuous and differentiable on R or

(−∞, ∞).
h(0) = 0; h(7) = 0
=⇒ h(0) = h(7).
So the hypotheses of Rolle’s Theorem is satisfied.

(b)

h0 (t) = 3t2 (t − 7)4 + 4(t − 7)3 t3

= t2 (t − 7)3 [3(t − 7) + 4t]
= t2 (t − 7)3 (7t − 21)
= 7t2 (t − 7)3 (t − 3)

We find a number c in [0, 7] such that h0 (t) = 0.

h0 (c) = 0 =⇒ 7t2 (t − 7)3 (t − 3) = 0

=⇒ t = 0, 3, 7.
The number c = 3 is in (0, 7).
The submarine is on the surface at t = 0 and returns after 7 minutes,
h(7) = 0. The submarine reaches a maximum depth of h(3) = 6912 at
t = 3 minutes. See Figure 1.8.

6 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Figure 1.8: (a) Schematic of submarine dive (b) Plot of the submarine func-
tion h(t) = t3 (t − 7)4 .

7 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

8 Dr. Joseph K. Ansong

Chapter 2

The Mean Value Theorem and

its Applications

2.1 The Mean Value Theorem (MVT)

Theorem 4. Let f be continuous on [a, b] and differentiable on (a, b). Then
there exists at least one number c in (a, b) such that

f (b) − f (a)
f 0 (c) = ;
b−a

or f (b) − f (a) = f 0 (c)(b − a).

Interpretation:
Case (a): From Figure 2.1, the slope of the secant line is

f (b) − f (a)
S=
b−a
and the slope of the tangent line, T , to the function f at c is f 0 (c). MVT
tells us that there is at least one point (c, f (c)) in the interval (a, b) such that
T is parallel to S.
Case (b): Note that if f (a) = f (b) then we recover Rolle’s Theorem.

Proof. The equation of a line with slope m and passing through (x1 , y1 ) is
given by
y − y1 = m(x − x1 ).
Now
f (b) − f (a)
m=S= .
b−a

9
Math 223 : Calculus II J.K.A

Figure 2.1: Schematic illustration of the Mean Value Theorem.

Since (b, f (b)) is on S, the equation of the secant becomes

f (b) − f (a)
y − f (b) = (x − b)
b−a
f (b) − f (a)
=⇒ y = f (b) + (x − b)
b−a
Let D(x) be the function for the distance between f (x) and y such that

f (b) − f (a)
 
D(x) = f (x) − f (b) + (x − b)
b−a

Note that D(a) = 0 = D(b). Also, D(x) is continuous on [a, b] and differen-
tiable on (a, b), therefore we can apply Rolle’s Theorem that there is a c in
(a, b) such that D0 (c) = 0. Now,

f (b) − f (a)
D0 (x) = f 0 (x) −
b−a
So, D0 (x) = 0 implies that

f (b) − f (a)
f 0 (c) − =0
b−a
f (b) − f (a)
f 0 (c) =
b−a

10 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Example 4. Show that the function satisfies the hypotheses of the Mean
Value Theorem on the given interval, and find all values of c that satisfy the
conclusion of the theorem.

(a) f (x) = x2 + 1; [0, 2]

(b) f (x) = x3 ; [−1, 1]

Solution. (a) f (x) = x2 + 1; [0, 2]

f (x) is a polynomial and so it is continuous on [0, 2] and differentiable on
(0, 2). So the hypotheses of the MVT are satisfied.

f (b) − f (a)
c ∈ [0, 2] : f 0 (c) =
b−a
f (2) − f (0)
=⇒ f 0 (x) = 2x =
2−0
5−1
=⇒ 2x = =2
2
=⇒ x = 1
∴ c = 1 ∈ [0, 2].
See illustration of this in Figure 2.2a.
(b) f (x) = x3 ; [−1, 1]
f (x) is a polynomial so MVT is satisfied.

f (b) − f (a)
f 0 (c) =
b−a
f (1) − f (−1)
=⇒ f 0 (x) = 3x2 =
1 − (−1)
1 − (−1)
=⇒ 3x2 = =1
2
1
=⇒ x2 =
3
1
=⇒ x = ± √ .
3
This is illustrated in Figure 2.2b.

Exercise. Use the graph of f to estimate the values of c that satisfy the
conclusion of the MVT for the given interval shown in Figure 2.3.

11 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Figure 2.2: Plot of (a) f (x) = x2 + 1 and (b) f (x) = x3

Figure 2.3:

12 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Example 5 (Real Life Example). The position of a maglev (magnetic

levitation) moving along a straight, elevated monorail track is given by

S = f (t) = 4t2 , 0 ≤ t ≤ 30,

where S is measured in meters and t in seconds.

(a) Find the average velocity of the maglev during the first 4 seconds of
the run.

(b) Find the number c in (0, 4) that satisfy the conclusion of the MVT.

(c) Interpret your results.

Solution. S = f (t) = 4t2 , 0 ≤ t ≤ 30

f (b) − f (a)
(a) Average velocity = .
b−a
First 4 seconds implies the interval [0, 4], such that a = 0 and b = 4.
Thus
f (4) − f (0)
Average velocity =
4−0
64 − 0
=⇒ = 16 m/s
4
(b) Find c in (0, 4) such that

f (b) − f (a)
f 0 (c) =
b−a
Now, f 0 (t) = 8t. Thus 8c = 16 =⇒ c = 2.

(c) Since f 0 (t) measures the instantaneous velocity of the maglev at any
time t, the MVT tells us that at t = 2 s (between t = 0 and t = 4),
the maglev must attain an intantaneous velocity equal to the average
velocity over [0, 4].

Example 6. Let
n x2 , if x < 1
f (x) = (2.1)
2 − x, if x ≥ 1

Considering Figure 2.4(a), does f satisfy the hypotheses of the MVT on

[0, 2]? Explain.

13 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Figure 2.4: Plot of (a) a function, and (b) it’s derivative.

Solution.
n 2x, x<1
0
f (x) = (2.2)
−1, x≥1

Figure 2.4(b) shows the graph of f 0 (x). We note f does not satisfy the
hypothesis of the MVT since f 0 (x) does not exist at x = 1 which lies in [0, 2].
Example 7. Suppose that f (0) = −3 and f 0 (x) ≤ 5 for all values of x. The
inequality gives a restriction on the rate of growth of f , which then imposes
a restriction on the possible values of f . Use the MVT to determine how
large f (4) can possibly be.
Solution. Since we are given f 0 (x) ≤ 5, the function is differentiable and
therefore continuous on [0, 4].
f (b) − f (a)
f 0 (c) =
b−a
f (4) − f (0)
=⇒ f 0 (c) =
4−0
f (4) + 3
=⇒ f 0 (c) =
4
Since f 0 (x) ≤ 5, we have
f (4) + 3
≤5
4
=⇒ f (4) ≤ 17.
So the answer is 17.

14 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Example 8. At 2 : 00 P.M. a car’s speedometer reads 30 mi/h. At 2 : 10 P.M.

it reads 50 mi/h. Show that at some time between 2 : 00 and 2 : 10 the
acceleration is exactly 120 mi/h.
Solution. At 2 : 00 P.M. let t = 0 and speed be 30 mi/h. This implies that
at 2 : 10 P.M., t = 10 min, and speed is 50 mi/h.
By assumption, the car’s speed, say f (t), is continuous and differentiable
everywhere. Note that acceleration is the derivative of speed. Using the
MVT
f (b) − f (a) f (10) − f (0) 50 − 30 mi/h
f 0 (t) = = =
b−a 10 − 0 10 min
20
= = 120 mi/h2 .
10/60

2.2 Consequences of the MVT

In this section, we examine some consequences of the Mean Value Theorem.
We will especially consider
• Functions with zero derivatives,

• Functions with equal derivatives, and

• Increasing and decreasing functions.

2.2.1 Functions with Zero Derivatives

Theorem 5 (Zero Derivatives). If f 0 (x) = 0 for all x in an interval (a, b),
then f is constant on (a, b).
Proof. We show that for any pair of numbers in (a, b), f has the same value.
Let x1 and x2 be arbitrary numbers in (a, b) such that x1 < x2 . Since f (x) is
differentiable in (a, b), it is differentiable in (x1 , x2 ) and continuous in [x1 , x2 ].
Thus f satisfies the hypotheses of MVT on [x1 , x2 ]. This implies that there
is a c in (x1 , x2 ) such that

f (x2 ) − f (x1 )
f 0 (c) =
x2 − x1
=⇒ f (x2 ) − f (x1 ) = f 0 (c)(x2 − x1 )
But f 0 (x) = 0 for all x in (a, b).

=⇒ f 0 (c) = 0

15 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

=⇒ f (x2 ) − f (x1 ) = 0
=⇒ f (x2 ) = f (x1 ).

2.2.2 Functions with Equal Derivatives

Theorem 6 (Equal Derivatives). If f 0 (x) = g 0 (x) for all x in an interval
(a, b), then f and g differ by a constant on (a, b); that is, there exist a constant
c such that
f (x) = g(x) + c
for all x in (a, b).

Proof. Let h(x) = f (x) − g(x). This implies that

h0 (x) = f 0 (x) − g 0 (x) = 0

for every x in (a, b). By Theorem 5, h(x) is constant. Thus

f (x) − g(x) = c

=⇒ f (x) = g(x) + c.

2.2.3 Increasing and Decreasing Functions

Definition 4. A function f is called increasing on an interval I if f (x1 ) <
f (x2 ) whenever x1 < x2 in I.
It is called decreasing on I if f (x1 ) > f (x2 ) whenever x1 < x2 on I.

Figure 2.5 gives an illustration of increasing and decreasing functions.

Question: How can we use the derivative of a function to determine if it’s
increasing or decreasing? The following theorem provides the answer.

Theorem 7 (Increasing/Decreasing Test or I/D Test). (a) If f 0 (x) >

0 on an interval, then f is increasing on that interval.

(b) If f 0 (x) < 0 on an interval, then f is decreasing on that interval

The theorem is illustrated in Figure 2.6.

16 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Figure 2.5: (a) Increasing function (b) Decreasing function.

Figure 2.6: Positive and negative slopes of a function.

Proof. (a) Let x1 and x2 be any two numbers in the interval with x1 < x2 .
Given that f 0 (x) > 0, we know f is differentiable in [x1 , x2 ]. So by the
MVT there is a number c between x1 and x2 such that

f (x2 ) − f (x1 )
f 0 (c) =
x2 − x1

=⇒ f (x2 ) − f (x1 ) = f 0 (c)(x2 − x1 )

Now f 0 (c) > 0 since f 0 (x) > 0 for all x in the interval. Also x2 −x1 > 0,

17 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

therefore
f (x2 ) − f (x1 ) > 0
=⇒ f (x2 ) > f (x1 )
Therefore f is increasing.

(b) Try proving this.

Finding Intervals of Increase and Decrease

Knowing the intervals where a function is increasing or decreasing helps in
sketching a graph of the function. The procedure is outlined below:

1) Find all values x for which f 0 (x) = 0 or f 0 (x) does not exist (the critical
values). Use the critical values to partition the domain of f into open
intervals.

2) In each interval, select a test value c and determine the sign of f 0 (c).

(a) If f 0 (c) > 0, then f is increasing on that interval.

(b) If f 0 (c) < 0, then f is decreasing on that interval.
(c) If f 0 (c) = 0, then f is constant on that interval. That is, it has
no local maximum or minimum at c.

Example 9. Find the intervals on which f is increasing or decreasing:

(1) f (x) = −x3 + 3x2 + 1.

x2
(2) f (x) = .
x−1
(3) f (x) = x3 − 3x2 + 2.

Solution. (1) f (x) = −x3 + 3x2 + 1

f 0 (x) = −3x2 + 6x = −3x(x − 2)

This implies f 0 (x) is continuous everywhere and has zeros at x = 0

and x = 2. Using the zeros, we now partition the domain (−∞, ∞)
into open intervals as shown in Table 2.1. Figure 2.7 illustrates the
increasing and decreasing parts of the function.

18 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Interval −3x x − 2 f 0 (x) Description

(−∞, 0) + − − decreasing
(0, 2) − − + increasing
(2, ∞) − + − decreasing

Table 2.1: The intervals of increase and decrease for f (x) = −x3 + 3x2 + 1.

Figure 2.7: Increasing (+) and decreasing (−) parts of the function.

x2
(2) f (x) = .
x−1
(x − 1)(2x) − x2 · 1
f 0 (x) =
(x − 1)2
2x2 − 2x − x2
=
(x − 1)2
x2 − 2x x(x − 2)
=⇒ f 0 (x) = =
(x − 1)2 (x − 1)2
We note that f 0 (x) does not exist at x = 1. Now,

f 0 (x) = 0 =⇒ x = 0, 2.

So the critical values are 0, 1 and 2. Table 2.2 shows the intervals of
x2
increase and decrease. Figure 2.8a shows the graph of f (x) = ,
x−1

19 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Interval x x − 2 f 0 (x) Description

(−∞, 0) − − + increasing
(0, 1) + − − decreasing
(1, 2) + − − decreasing
(2, ∞) + + + increasing

x2
Table 2.2: The intervals of increase and decrease forf (x) = .
x−1

confirming the intervals where the function is increasing and decreasing.

x2
Figure 2.8: Plots of (a) f (x) = x−1
and (b) f (x) = x3 − 3x2 + 2.

(3) f (x) = x3 − 3x2 + 2. Try solving this before looking at the solution.

f 0 (x) = 3x2 − 6x

f 0 (x) = 3x(x − 2)
f 0 (x) = 0 =⇒ x = 0, 2
The intervals of increase and decrease are displayed in Table 2.3 The
graph of f (x) = x3 − 3x2 + 2 is displayed in Figure 2.8b, showing the
intervals where the function is increasing and decreasing.

20 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Interval 3x x − 2 f 0 (x) Description

(−∞, 0) − − + increasing
(0, 2) + − − decreasing
(2, ∞) + + + increasing

Table 2.3: The intervals of increase and decrease for f (x) = x3 − 3x2 + 2.

2.3 Applications of the MVT: Using the MVT

to Establish an Inequality
Example 10. Use the Mean Value Theorem to prove the following inequal-
ities

(a) cos x ≥ 1 − x for x ≥ 0

π
(b) tan x > x for 0 < x <
2
Solution. (a) cos x ≥ 1 − x for x ≥ 0. Let f (x) = cos x for [0, x]. f is
continuous and differentiable on the given interval, so by MVT

f (b) − f (a)
f 0 (c) =
b−a
where a = 0 and b = x.
f (x) − f (0)
=⇒ = − sin c
x−0
cos x − 1
=⇒ = − sin c
x
But sin c ≤ 1 =⇒ − sin c ≥ −1
cos x − 1
=⇒ ≥ −1
x
=⇒ cos x − 1 ≥ −x
∴ cos x ≥ 1 − x
π
(b) tan x > x for 0 < x < . Let f (x) = tan x on [0, x] where 0 < x < π2 .
2
By the MVT
f (b) − f (a)
f 0 (c) =
b−a

21 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

where a = 0 and b = x.
f (x) − f (0) tan x
=⇒ f 0 (c) = =
x−0 x
tan x
=⇒ sec2 c = , x>0
x
Note that
1
sec x = .
cos x
1
Thus, cos x < 1 =⇒ >1
cos x
=⇒ sec x > 1 =⇒ sec2 c > 1
tan x
=⇒ >1
x
π
∴ tan x > x, for 0 < x < .
2
Example 11. Suppose that 1 ≤ f 0 (x) ≤ 4 for all x in [2, 5]. Show that

3 ≤ f (5) − f (2) ≤ 12

Solution. 1 ≤ f 0 (x) ≤ 4 for all x in [2, 5]. Since f 0 (x) is finite, f is differen-
tiable and continuous on the given interval. By the MVT

f (b) − f (a)
f 0 (c) = , [2, 5]
b−a

f (5) − f (2) f (5) − f (2)

f 0 (c) = =
5−2 3
But 1 ≤ f 0 (c) ≤ 4. Thus

f (5) − f (2)
1≤ ≤4
3
∴ 3 ≤ f (5) − f (2) ≤ 12

Example 12. (a) Show that ex ≥ 1 + x for x ≥ 0.

(b) Deduce that

1
ex ≥ 1 + x + x2 for x ≥ 0
2

22 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

(c) Use mathematical induction to prove that for x ≥ 0 and any positive
integer n,
x2 xn
ex ≥ 1 + x + + ··· +
2! n!
Solution. (a) Let f (x) = ex − (1 + x) and show that f (x) ≥ 0. Now
f (0) = e0 − (1 + 0) = 1 − 1 = 0, and f 0 (x) = ex − 1. Note that if x > 0,
ex > 1 =⇒ f 0 (x) > 0. Since f 0 (x) > 0 for x > 0, it implies that f (x) is
increasing for x > 0.

=⇒ f (x) > f (0) for x > 0

=⇒ f (x) > 0
=⇒ ex − (1 + x) > 0
=⇒ ex > 1 + x.
Thus,
ex ≥ 1 + x.

(b)
1
ex ≥ 1 + x + x2 for x ≥ 0
2
 
1
Let g(x) = ex − 1 + x + x2 , and show that g(x) ≥ 0.
2
For x = 0 :

g(0) = e0 − (1 + 0 + 0) = 0 =⇒ g(x) = 0
1
=⇒ ex = 1 + x + x2 when x = 0
2
For x > 0 :

g 0 (x) = ex − (0 + 1 + x)
=⇒ g 0 (x) = ex − (1 + x) = f (x)
from solution to (a). Since f (x) > 0 for x > 0, g 0 (x) > 0 for x > 0.
This implies that g(x) is increasing, and so

g(x) > g(0) for x > 0

 
x 1 2
=⇒ e − 1 + x + x > 0
2

23 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

1
=⇒ ex > 1 + x + x2
2
Thus,
1
ex ≥ 1 + x + x2 for x ≥ 0
2
(c) Let Sn be a statement about the positive integer n. Let

x2 xn
 
x
Sn (x) = e − 1 + x + + ··· +
2! n!

S1 (x) is true from the solution to problem (a).

S2 (x) is true from the solution to problem (b).
Assume Sk (x) is true. Then

x2 xk
 
x
Sk (x) = e − 1 + x + + ··· + ≥0
2! k!

By mathematical induction, we need to show that Sk+1 (x) is true. Now

x2 xk+1
 
x
Sk+1 (x) = e − 1 + x + + ··· +
2! (k + 1)!

Sk+1 (0) = 0 for x = 0

x2 xk
 
0 x
Sk+1 (x) = e − 1 + x + + ··· + = Sk (x) ≥ 0
2! k!
Thus, Sk+1 (x) is increasing, and so

Sk+1 (x) > Sk+1 (0) = 0

=⇒ Sk+1 (x) > 0

Hence,
Sk+1 (x) ≥ 0.
Thus, Sk+1 (x) is true. Therefore, by mathematical induction, Sn (x) is
true. That is
x2 xn
 
x
Sn = e − 1 + x + + ··· + ≥ 0.
2! n!

24 Dr. Joseph K. Ansong

Chapter 3

Inverse Functions

In this chapter, we examine inverse functions and some of their properties.

We will investigate the reflective property of an inverse function, examine
the existence of an inverse function, and find the derivative of an inverse
function.

3.1 Introduction
Consider the arrow diagrams in Figure 3.1.

Figure 3.1: Arrow diagrams. (a) one-to-one function (b) A function, but not
one-to-one (c) Not a function.

(1) f : A → B is a function, and it never takes on the same value twice.

That is, if x1 6= x2 then f (x1 ) 6= f (x2 ). It is a one-to-one function
(defined below).

25
Math 223 : Calculus II J.K.A

(2) g : A → B is a function. However, note that

g(2) = g(4) = c

Thus, g takes on the same value twice. So g is not a one-to-one function.

(3) h : A → B is not a function since h(2) takes on both a and c.

Definition 5. A function f is called a one-to-one function if it never takes
on the same value twice; that is

f (x1 ) 6= f (x2 ) whenever x1 6= x2

or if
f (x1 ) = f (x2 ) =⇒ x1 = x2 .

Horizontal Line Test

A function is one-to-one if and only if no horizontal line intersects its graph
more than once. Figure 3.2 illustrates the use of the horizontal line test.

Figure 3.2: The horizontal line test shows that (a) y = x3 is a one-to-one
function (b) y = x2 is not a one-to-one function.

Definition 6. Let f be a one-to-one function with domain A and range B.

Then its inverse function f −1 has domain B and range A and is defined
by
f −1 (y) = x ⇐⇒ f (x) = y

26 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

for any y in B. Equivalently, we may reverse x and y such that

f −1 (x) = y ⇐⇒ f (y) = x

This illustrated in Figure 3.3

Figure 3.3: Arrow diagram of a function and its inverse.

Remark. Domain of f −1 = range of f , and

Range of f −1 = domain of f .
1
Remark. Warning! f −1 (x) 6= . For reciprocals we write
f (x)
1
= [f (x)]−1
f (x)
Example 13. If f (2) = 7 and f (−10) = −17, find f −1 (7) and f −1 (−17).
Solution.
f −1 (7) = 2, f −1 (−17) = −10

Cancellation Equations
Let g(x) be the inverse of f . That is, g(x) = f −1 (x). Then

g [f (x)] = f −1 [f (x)] = x for all x ∈ A

f [g(x)] = f f −1 (x) = x for all x ∈ B

 

Example 14. Show that f (x) = x1/3 and g(x) = x3 are inverses of each
other.

27 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Solution.
1/3
f [g(x)] = f (x3 ) = x3 =x

3
g [f (x)] = g(x1/3 ) = x1/3 =x
This implies that f and g are inverses of each other. That is
f −1 (x) = g(x) and g −1 (x) = f (x).
The graphs of f (x) and g(x) are displayed in Figure 3.4. Note that the graph
of x3 is a reflection of the graph of x1/3 about the line y = x.

Figure 3.4: Plots of g = x3 and f = x1/3 .

3.2 Reflective Property of Inverse Functions

If (a, b) is on the graph of f (x), then b = f (a), and we have
f −1 (b) = f −1 [f (a)] = a
As shown in Figure 3.5, the graph of f −1 is the reflection of the graph of f
about the line y = x and vice versa.
Example 15. Examples of the reflective property of inverse functions are
illustrated in Figure 3.6.

28 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Figure 3.5: Schematic of the reflective property of inverse functions.

Figure 3.6: Illustration of the reflective property of some inverse functions.

3.3 Existence of an Inverse Function

Theorem 8. A function has an inverse if and only if it is one-to-one.

Proof. Suppose f : A → B has an inverse function y = g(x). Then

f (y) = f [g(x)] = x

If f (a) = f (b), then

g[f (a)] = g[f (b)] =⇒ a = b

29 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Therefore, f is one-to-one.
Now suppose f is one-to-one. For x ∈ B, there exist y ∈ A such that
f (y) = x. There is only one such y since f is one-to-one. Let

y = g(x)

=⇒ g(f (y)) = g(x) = y, (we used f (y) = x)

for any y, and
f [g(x)] = f (y) = x
for any x. So g is an inverse of f and vice versa.

3.3.1 Finding Inverse Functions

Below is a procedure for finding inverse functions:

(1) Write y = f (x)

(2) Solve for x in terms of y (if possible)

(3) Interchange x and y to obtain y = f −1 (x).

Example 16. Find the inverse of the following functions

1
(a) y = √
2x − 3
1 + 3x
(b) y =
5 − 2x
(c) y = ln (x + 3)
1
Solution. (a) y = √
2x − 3
1
=⇒ y 2 =
2x − 3

=⇒ y 2 (2x − 3) = 1 =⇒ 2y 2 x − 3y 2 = 1
1 + 3y 2
=⇒ x =
2y 2
1 + 3x2
∴ f −1 (x) =
2x2

30 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

1 + 3x
(b) y =
5 − 2x
=⇒ 5y − 2yx = 1 + 3x
=⇒ 3x + 2yx = 5y − 1
=⇒ x(3 + 2y) = 5y − 1
5y − 1
=⇒ x =
3 + 2y
5x − 1
∴ f −1 (x) =
3 + 2x
(c) y = ln (x + 3)
=⇒ x + 3 = ey
=⇒ x = ey − 3
∴ f −1 (x) = ex − 3

3.4 Continuity and Differentiability of Inverse

Functions
Theorem 9. Let f be one-to-one, so that it has an inverse f −1 .
(a) If f is continuous on its domain, then f −1 is continuous on its domain.
(b) If f is differentiable at c and f 0 (c) 6= 0, then f −1 is differentiable at
f (c).
Theorem 10. Let f be differentiable on its domain and have an inverse
function g = f −1 . Then the derivative of g is given by
1
g 0 (x) = ,
f0 [g(x)]
provided that f 0 [g(x)] 6= 0.
Proof. Since f is differentiable, g is also differentiable by Theorem 9. Since
g is the inverse function;
f [g(x)] = x
Differentiating (and using the Chain Rule), we get
f 0 [g(x)]g 0 (x) = 1
1
∴ g 0 (x) = .
f0 [g(x)]

31 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Example 17. Let g denote the inverse of the function f . Show that the
point (a, b) lies on the graph of f . Find g 0 (b).

(a) f (x) = 2x + 1; (2, 5)

(b) f (x) = x5 + 2x3 + x − 1; (0, −1)

Solution. (a) f (x) = 2x + 1; (2, 5)

=⇒ f (2) = 2(2) + 1 = 5

Therefore (2, 5) lies on the graph of f . Also f −1 (5) = g(5) = 2. Now

1 1
g 0 (b) = =⇒ g 0 (5) =
f 0 [g(b)] f 0 [g(5)]
1
=⇒ g 0 (5) =
f 0 (2)
Now, f 0 (x) = 2 =⇒ f 0 (2) = 2
1
∴ g 0 (5) =
2

(b) f (x) = x5 + 2x3 + x − 1; (0, −1).

f (0) = 0 + 0 + 0 − 1 = −1

Thus, (0, −1) lies on f . And f −1 (−1) = g(−1) = 0.

1 1
g 0 (−1) = =
f 0 [g(−1)] f 0 (0)

f 0 (x) = 5x4 + 6x + 1

=⇒ f 0 (0) = 0 + 0 + 1 = 1

∴ g 0 (−1) = 1

Exercise. (1) Suppose f (x) = x2 for x ∈ [0, ∞), and let g be the inverse
of f .
1
(a) Compute g 0 (x) using g 0 (x) =
f 0 [g(x)]
(b) Find g 0 (x) by first computing g(x).

32 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

(2) Suppose that g is the inverse of a function f . If f (2) = 4 and f 0 (2) = 3,

find g 0 (4).

(3) Find f −1 (a) for the function f and the real number a.

(a) f (x) = x3 + x − 1; a = −1
(b) f (x) = 2x5 + 3x3 + 2; a = 2.

(4) The graph of f is given in Figure 3.7

(a) Why is f one-to-one?

(b) State the domain and range of f −1 .
(c) Estimate the value of f −1 (1).

Figure 3.7: Plot of a function.

Solution to Selected Exercises

Solution. (1) f (x) = x2 for x ∈ [0, ∞).
1
(a) g 0 (x) =
f 0 [g(x)]
√
y = x2 =⇒ x = y, x≥0

33 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

√
∴ f −1 (x) = g(x) = x
0
√0
f (x) = 2x =⇒ f [g(x)] = 2 x
1
=⇒ g 0 (x) = √
2 x
√
(b) From g(x) = x
1 1
=⇒ g 0 (x) = x−1/2 = √
2 2 x
(2)
1
g 0 (4) =
f 0 [g(4)]
f (2) = 4 =⇒ f −1 (4) = g(4) = 2. From the first equation, we get
1 1
g 0 (4) = =
f 0 (2) 3

(3) f (x) = x3 + x − 1; a = −1.

(a) Let f −1 (−1) = x
=⇒ f (x) = −1
=⇒ x3 + x − 1 = −1
=⇒ x3 + x = 0 =⇒ x(x2 + 1) = 0
But (x2 + 1) > 0, thus x = 0.
∴ f −1 (−1) = 0
(b) f (x) = 2x5 + 3x3 + 2; a = 2. Let f −1 (2) = x
=⇒ f (x) = 2
2x5 + 3x3 + 2 = 2
=⇒ 2x5 + 3x3 = 0 =⇒ x3 (2x2 + 3) = 0
=⇒ x = 0
∴ f −1 (2) = 0.
(4) (a) Using the horizontal line test; any horizontal line through the
graph of f intersects it at only one point.
(b) Domain of f −1 = [−4, 2]. (The range of f −1 = [−3, 3]).
(c) Let f −1 (1) = x. Thus f (x) = 1. From the graph, we have
f −1 (1) ≈ −2.5

34 Dr. Joseph K. Ansong

Chapter 4

Logarithmic and Exponential

Functions

In this chapter, we will first study general exponential and general logarithmic
functions, and move on to consider special cases of these two functions: the
natural logarithmic function and the exponential function, and learn about
their graphs. We will delve into the proofs of some basic properties of these
two functions, and the behaviour of the functions at infinity and close to
the origin. The natural logarithmic and the exponential functions may be
represented as limits, this chapter addresses this aspect of the functions. The
chapter concludes by looking at logarithmic inequalities and limits, the scales
of infinity, that is, comparison of the magnitude of two functions, and finally,
how to differentiate logarithmic functions.

4.1 General Logarithmic Functions

The logarithmic ( or log) function with base a is denoted by loga , such that

loga x = y ⇐⇒ ay = x; a > 0, a 6= 1.

The log function is the inverse of the exponential function, for example f (x) =
ax . Using the cancellation equations with f (x) = ax , f −1 (x) = loga x,
loga ax = x for every x ∈ R
aloga x = x for every x > 0.

4.1.1 Laws of Logarithms

If x and y are positive numbers, then

35
Math 223 : Calculus II J.K.A

1. loga xy = loga x + loga y

x
2. loga y
= loga x − loga y

3. loga xr = r loga x
where r is any real number.

Proof. 1. loga xy = loga x + loga y. Let m = loga x and n = loga y

=⇒ x = am , y = an

=⇒ xy = an · am = am+n
The last line would be proved later from the properties of the general
exponential functions.

=⇒ loga xy = loga am+n = (m + n) loga a

Using the cancellation property, we get

∴ loga xy = m + n = loga x + loga y

2. loga xy = loga x − loga y.

Using the definition of m and n above, we have
x am
= n = am−n
y a
x
=⇒ loga = loga am−n = (m − n) loga a
y
x
∴ loga = loga x − loga y
g
3. loga xr = r loga x. Let m = loga x

=⇒ x = am

Raising both sides to the power r gives

xr = (am )r

=⇒ loga xr = loga amr = mr

∴ loga xr = r loga x

36 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

4.2 The Natural Logarithm

The natural logarithmic function is a special case of the logarithmic function;
it is the log function to the base e, where the number e ≈ 2.71828. By
replacing a with e in the general logarithm, we get the natural logarithm:
loge x = ln x

Properties
1. Here are a first set of properties for the natural log function:
ln x = y ⇐⇒ ey = x
ln ex = x, x∈R
ln x
e =x
Note that if x = 1, we get
ln e = 1

2. Below is another important property of the natural log function.

For a > 0(a 6= 1),
ln x
loga x =
ln a

Proof. Let m = loga x, then

x = am =⇒ ln x = ln am
=⇒ ln x = m ln a
ln x
=⇒ m =
ln a
ln x
∴ loga x =
ln a

The relation above is helpful for using calculators to compute the log-
arithm with any base. Another definition of the natural log is given by
using integrals, as shown below.

Definition 7. The natural logarithm function is defined by

Z x
1
ln x = dt for all x > 0.
1 t

37 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

If x > 0, ln x can be interpreted as the area of the region under the graph
1
of y = on the interval [1, x]. Figure 4.1 gives illustrations of the above
t
definition.

Figure 4.1: Schematic plot showing

Graph of ln x

Figure 4.2: Graph of ln x

4.2.1 Laws of Natural Logarithms

Theorem 11. Let x and y be positive numbers and let r be a rational
number. Then

(a) ln 1 = 0

38 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

(b) ln xy = ln x + ln y

(c) ln xy = ln x − ln y

(d) ln xr = r ln x

Proof. The proof in (a) is straightforward. For the proof of the expressions in
(b)-(d), you may use the approach used in the proof of the general logarithmic
functions; the proofs shown below rely on the derivative of the logarithmic
function which will be treated shortly.
Z x
1
(a) ln x = dt, =⇒ ln 1 = 0.
1 t

1 1
(b) Let F (x) = ln ax for a > 0. Then F 0 (x) = (a) = . But
ax x
d 1
(ln x) =
dx x
This implies that ln x and F (x) = ln ax differ by a constant, say c.
That is
ln ax = ln x + c
If x = 1 =⇒ c = ln a. Note that x = 1 is the intial value of ln x from
definition. Thus
ln ax = ln x + ln a
Since a can be any positive number, we can replace a by y to get

ln xy = ln x + ln y.

1
(c) Put x = in (b) to get
y

1
ln 1 = ln + ln y
y

1
=⇒ ln = − ln y
y
   
x 1 1
∴ ln = ln x · = ln x + ln = ln x − ln y.
y y y

39 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

(d) Let F = ln xr , and G = r ln x. Then

r r
F 0 (x) = , G0 (x) =
x x
=⇒ ln xr = r ln x + c
For x = 1, we get c = 0. Thus

ln xr = r ln x.

4.2.2 Derivatives of Logarithmic Functions

Prove the following identities

d x
(a ) = ax ln a (4.1)
dx
d 1
(loga x) = (4.2)
dx x ln a

Proof.
d x
(a ) = ax ln a
dx
Let a = eln a . Then
ax = e(ln a)x
d x x
(a ) = e(ln a)x · ln a = eln a · ln a
dx
d x
∴ (a ) = ax ln a
dx

Proof.
d 1
(loga x) =
dx x ln a
y
Let y = loga x. Then a = x.

d y
=⇒ (a ) = 1
dx
d y dy
=⇒ (a ) =1
dy dx

40 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Using equation (4.1), we get

dy
=⇒ ay (ln a) =1
dx
But ay = x. Thus
dy
x ln a =1
dx
dy 1
∴ = .
dx x ln a

Remark. Replacing a with e in equation (4.2), we get

d 1
ln x =
dx x
A more general form of this relation would be shown later.
Example 18. (1) Expand the following expressions using the laws of log-
arithms
√
2 3 x1/3 y 2/3
(a) ln (b) ln
5 z 1/2
Solution. (a) √
2 3 √
ln = ln 2 3 − ln 5
5
(b)
x1/3 y 2/3 1/3 2/3
− ln z 1/2


ln 1/2
= ln x y
z
= ln x1/3 + ln y 2/3 − ln z 1/2
1 2 1
= ln x + ln y − ln z
3 3 2
Example 19. Use the laws of logarithms to write the expression as the
logarithm of a single quantity.

(a) ln 4 + ln 6 − ln 12 (b) 3 ln 2 − 12 ln (x + 1)

24
Solution. (a) ln 4 + ln 6 − ln 12 = ln (4 × 6) − ln 12 = ln = ln 2
12
 
1 8
(b) 3 ln 2 − ln (x + 1) = ln 23 − ln (x + 1)1/2 = ln √
2 x+1

41 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Theorem 12. Let u be a differentiable function of x. Then

d 1
(a) ln |x| = , x 6= 0
dx x
d 1 du
(b) ln |u| = , u 6= 0
dx u dx
d g 0 (x)
Note: (b) is similar to the derivative ln[g(x)] = .
dx g(x)
Proof. (a) Already discussed in a remark above.

(b) Try proving it using using the chain rule.

Example 20. Find the derivatives of the following funtions

(a) f (x) = ln (2x + 3) (b) g(x) = ln x (c) y = ln | cos x|

x+1

Solution. (a)

1 d
f (x) = ln (2x + 3) =⇒ f 0 (x) = (2x + 3)
2x + 3 dx
1 2
= (2) =
2x + 3 2x + 3
ln x
(b) g(x) = x+1
(x + 1) x1 − ln x
g 0 (x) =
(x + 1)2
x+1
x
− ln x (x + 1) − x ln x
= =
(x + 1)2 x(x + 1)2

(c) y = ln | cos x|

dy 1 sin x
= (− sin x) = − = − tan x
dx cos x cos x

4.2.3 Logarithmic Differentiation

Used to simplify complicated expressions for differentiation.

Example 21.

42 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

√
(a) y = (2x + 1)2 (3x2 − 4)3 (c) y = x x

r
x−1
(b) y = 3 2
x +1
Solution. (a) y = (2x + 1)2 (3x2 − 4)3
=⇒ ln y = 2 ln (2x + 1) + 3 ln (3x2 − 4)
1 dy 2 3
=⇒ = (2) + 2 (6x)
y dx 2x + 1 3x − 4
4 18x
= + 2
2x + 1 3x − 4
 
dy 4 18x
· (2x + 1)2 (3x2 − 4)3
 
∴ = + 2
dx 2x + 1 3x − 4
r
3 x−1
(b) y =
x2 + 1
1/3
x−1

=⇒ y =
x2 + 1
1 1
=⇒ ln y = ln (x − 1) − ln (x2 + 1)
3 3
1 dy 1 1
=⇒ = − (2x)
y dx 3(x − 1) 3(x2 + 1)
1 2x
= −
3(x − 1) 3(x2 + 1)
√
(c) y = x x
Sorry, it’s an exercise! :)

4.2.4 The Natural Logarithm as a Limit

For any number y > 0
yx − 1 ax − 1
ln y = lim and ln a = lim
x→0 x x→0 x
Remark. We note the following important relations:
f (x + a) − f (a)
(a) f 0 (a) = lim
x→0 x−a
f (x) − f (0)
=⇒ f 0 (0) = lim (4.3)
x→0 x

43 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

(b) If f (x) = z = y x , then by logarithmic differentiation

ln z = x ln y
1 dz dz
= ln y =⇒ = z ln y
z dx dx
=⇒ f 0 (x) = y x ln y

f 0 (0) = ln y (4.4)
Thus, the derivative of f (x) = y x at x = 0 is ln y. Applying (4.4) in
(4.3) gives
yx − 1
ln y = lim
x→0 x
x
a −1
ln a = lim
x→0 x

4.3 General Exponential Functions

An exponential function is a function of the form
f (x) = ax , a>0
where a is a constant. The graph of ax with a = 2 is shown in Figure 4.3

Figure 4.3: Graph of y = 2x

4.3.1 Laws of Exponents

Let a and b be positive numbers. If x and y are real numbers, then

44 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

(a) ax+y = ax · ay ax
(d) = ax−y
ay
(b) (ax )y = axy
 a  x ax
(c) (ab)x = ax bx (e) = x
b b

Proof. (a)
ax ay = ax+y
loga (ax · ay ) = loga ax + loga ay
= x + y = loga ax+y

Since the logarithm function is one-to-one

ax · ay = ax+y

(b) (ax )y = axy . Let r = ax

=⇒ loga (ax )y = loga ry = y loga r

= y loga ax = xy loga a = xy
=⇒ loga (ax )y = loga axy
Since the logarithm is a one-to-one function, we get

(ax )y = axy

Remark. To prove the relation in (c), we first establish the following identity

loga xy = loga x + loga y

We have already proved the above relation under “general logarithms”. An

alternative proof is shown below, you may skip it if you want.
Proof. Let y = loga x. Then

dy 1
=
dx x ln a
Also let g = loga bx. Then

d g d dg
ag = bx =⇒ (a ) = b =⇒ (ag ) =b
dx dg dx

45 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

dg
=⇒ ag ln a ·=b
dx
dg b 1
=⇒ = =
dx bx ln a x ln a
Since dy/dx = dg/dx, y and g differ by a constant, say C. Thus

loga x = loga bx + C

x = 1 =⇒ loga 1 = loga b + C =⇒ C = − loga b

=⇒ loga bx = loga x + loga b
Since b is arbitrary, we get

loga xy = loga x + loga y

Proof. We’re now ready to prove the relation in (c); (ab)x = ax bx . Taking
the lagarithm of both sides, we get

=⇒ loga ax bx = loga ax + loga bx

= x loga a + x loga b
= x (loga a + loga b)
= x loga (ab)
=⇒ loga ax bx = loga (ab)x
∴ ax bx = (ab)x .

4.3.2 Derivative of Exponential functions

1). Let f (x) = ax by definition of the derivative:

0 f (x + h) − f (x) ax+h − ax
f (x) = lim = lim
h→0 h h→0 h
0 ah − 1
f (x) = ax lim .
h→0 h
0 ah − 1
Note that: f (0) = lim . Therefore
h→0 h
0 0
f (x) = f (0)ax

46 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Remember:
yx − 1 ax − 1 ah − 1
ln y = lim =⇒ ln a = lim = lim
x→0 x x→0 x h→0 h
0
where f (0) = ln a. Therefore,
0
f (x) = ax ln a
Alternatively:
x
f (x) = ax = eln a = ex ln a
0 d
=⇒ f (x) = ex ln a · (x ln a) = ex ln a · (ln a)
dx
0
∴ f (x) = ax · (ln a)
Theorem 13. Show that
d u du
a = (ln a)au ·
dx dx
where a is a positive number (a 6= 1) and u is the differentiable function of
x.
Proof. Let y = au , then
d u d u du
(a ) = (a )
dx du dx
d u du
(a ) = au ln a
dx dx

d 1 du
Question: Show that loga u = · where u is the differentiable,
dx u ln a dx
a > 0(a 6= 1)

4.3.3 Natural Exponential functions

The natural exponential function denoted by exp(x) or ex , is the function
satisfying
1. ln(ex ) = x, ∀x ∈ [−∞, ∞]
2. eln x = x, ∀x ∈ [0, ∞]
Equivalently, ex = y ⇔ ln y = x. It is the inverse of the natural log function:
Diagram below

47 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Figure 4.4: Graphs of y = ln x and y = ex

The number e
The number e is the number such that:
Z e
1
ln e = dt = 1
1 t

1
Figure 4.5: Graph of y =
t

e ≈ 2.718281828459

Theorem 14. a) ln ex = x ∀ x ∈ [−∞, ∞]. =⇒ ln e = 1

b) eln x = x ∀ x ∈ [0, ∞]

Example 22. 1) Solve ln(x + 3) = 6

=⇒ x + 3 = e6 =⇒ x + 3 ≈ 403.43
∴ x ≈ 400.43

48 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

2) Solve e1−2x = 5

=⇒ 1 − 2x = ln 5 =⇒ 2x = 1 − ln 5

=⇒ 2x = 1 − 1.609 = −0.609
∴ x ≈ −0.305

The number e as a Limit

0
Let f (x) = ln x and use the definition of derivative to compute f .

0 f (x + h) − f (x) ln(x + h) − ln(x)

f (x) = lim = lim
h→0 h h→0 h
Let x = 1, we get.

0 ln(1 + h) 1 1
=⇒ f (1) = lim = lim ln(1 + h) = lim ln(1 + h) h
h→0 h h→0 h h→0

0 1 0
But f (x) = ⇒ f (1) = 1
x
1
=⇒ lim ln(1 + h) h = 1,
h→0

since ln is a continuous function.

h 1
i
ln lim (1 + h) h = 1
h→0

1
∴ lim (1 + h) h = e
h→0

Note that a similar expression for ex as a limit may be obtained if the sub-
stitution x = 1 is not made in the above derivation.
1
Let n = ⇒ n → ∞ as h → 0
h
 n
1
∴ lim 1 + =e
n→∞ n
 x n
Exercise. 1) Show that ex = lim 1+
n→∞ n
eh − 1
2) Prove that lim =1
h→0 h

49 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Theorem 15 ( Laws of exponents). Let x and y be real numbers and r be

a rational number. Then we have the following:

1) ex ey = ex+y
ex
2) = ex−y
ey
3) (ex )r = exr

Proof. 1)
ex · ey = ex×y
=⇒ ln(ex · ey ) = ln ex + ln ey = x + y
=⇒ ln(ex ey ) = ln ex+y .
Since ln is one to one: ex ey = ex+y

2) Let y = −y in 1). Therefore we have:

ex
ex · e−y = ex−y =⇒ = ex−y
ey

4.3.4 Derivatives of Exponential Functions

Theorem 16. Let u be a differentiable function of x. Then we have the
following:
d x
(a) e = ex
dx

d u du
(b) e = eu
dx dx
d x
Proof. (a) e = ex Let y = ex
dx
1 dy dy
ln y = x =⇒ = 1 =⇒ =y
y dx dx

d x
∴ (e ) = ex
dx

50 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

d u du
(b) e = eu · Let y = eu
dx dx
ln y = u(x)

1 dy du dy du
· = ⇒ =y
y dx dx dx dx
d u du
(e ) = eu
dx dx

Exercise. (1) Differentiate the following:

e2x
a) f (x) =
1 + e−x
b) g(x) = ecos(x)
c) f (x) = x2 ln(e2x + 1)
d) g(t) = et ln t e). f (x) = 2x2

(2) Find the Limit of the following:

 2 
3t + 1 −0.1t
a) lim e
t→∞ 2t2 − 1
 x 
2e + 1
b) lim
x→∞ 3ex + 2
c) lim sin x1
x→∞

(3) Show that

log(3 + x) − log 3 1
lim =
x→0 x 3 ln 10

4.4 Logarithmic and Exponential Integration

(1) Let u be a differentiable function and u 6= 0. Then

1
Z
du = ln |u| + c.
u
NB:
d 1 1
Z
ln |x| = ⇒ ln |x| = dx
dx x x

51 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

(2) Let u be a differentiable function of x. Then:

Z
eu du = eu + c

ax
Z
(3) ax dx = + c, a > 0, a 6= 1. We use
ln a
y = ax ⇒ ln y = x ln a
0
y 0
⇒ = ln a ⇒ y = ax ln a
y
d x
⇒ (a ) = ax ln a
dx Z
ax + c1 = (ln a) ax dx

ax
Z
x
a dx = +c
ln a
Examples
Z √
1 ln x
Z Z
Integrate the following: a). dx b). dx c). e5x dx
2x + 1 x
Z 2
ex
d). 2
dx
Z 1 x
ex R3 x
e). x
dx f ). 0
2 dx
0 1+e

1
Z
Solution. a) I = dx, Let u = 2x + 1 ⇒ du = 2dx
  2x
Z +1
1 1 1 1 1
Z
I= du = du = ln |u| + c
u 2 2 u 2
1
∴ I= ln |2x + 1| + c
2
Z √
ln x
b) I = dx, Let u = ln x ⇒ du = x1 dx
x
√ 2 3
Z Z
1
I= udu = u 2 du = u 2 + c
3
2 3
∴ I= (ln x) 2 + c
3

52 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

c) I Z= e5x dx, Let uZ= 5x ⇒ du = 5dx

du 1 1
I = eu · = eu du ⇒ eu + c
5 5 5

1 5x
∴ e +c
5
2
ex
Z
d). I = 2
dx, Let u = x2 ⇒ du = − x22 dx ⇒ − 12 du = dx
x2
x
 
1 1 1
Z Z
u
I= e − du = − eu du = − eu + c
2 2 2

1 2
∴ I = − e x +c
2
1
ex
Z
e). I = x
dx Let u = 1 + ex ⇒ du = ex dx
Z 1 x0 1 + e Z
1
e du 1
I= · x
= du ⇒ ln |u| ⇒ ln |1 + ex | ⇒ ln |1 + e1 | − ln |1 + e0 |
0 u e 0 u
 
1+e
∴ I = ln ≈ 0.6201
2

3
2x 23 1 7
Z 3
f ). I = 2x dx = = − =
0 ln 2 0 ln 2 ln 2 ln 2

7
∴ I=
ln 2

4.5 Logarithmic Inequalities

a) For a number a > 1, y = loga x is an increasing function of x and it is
a decreasing function for 0 < a < 1.

b) If y = loga x, then x > 0. For example y = log3 (x − 1) ⇒ x − 1 > 0 ⇒

x > 1.

c) a > 1, loga x > loga y ⇒ x > y.

d) 0 < a < 1, loga x > loga y ⇒ x < y.

53 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Example 23. 1. Find the values of x that satisfy the ff inequalities:

a) log3 (2x + 3) > log3 (3x) b) log2 (x + 1) > log4 (x2 )

2. Find the solution set of 2 log49 (2x + 1) − 1 6 0

Solution. 1a)
log3 (2x + 3) > log3 (3x)
2x + 3 > 3x ⇒ x < 3
Also,
3
2x + 3 > 0 ⇒ x > −
2
And
3x > 0 ⇒ x > 0
Solution set: 0 < x < 3 . See Figure 4.6.

Figure 4.6:

(1b) log2 (x + 1) > log4 x2

NB: y = log2 (x + 1) ⇒ 2y = x + 1 ⇒ 4y = (x + 1)2
y = log4 (x+1)2 ∴ log4 (x+1)2 > log4 x2 ⇒ (x+1)2 > x2 ⇒ x2 +2x+1 >
x2
1
⇒ 2x + 1 > 0 ⇒ x > −
2

2) 2 log49 (2x + 1) − 1 6 0

1 1
log49 (2x + 1) 6 ⇒ log49 (2x + 1) 6 log49 49
2 2
⇒ log49 (2x + 1) 6 log49 7 ⇒ 2x + 1 6 7 ⇒ x 6 3
1
Also: 2x + 1 > 0 ⇒ x > − 21 ∴ − < x 6 3
2

54 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Example 24. 1. Use the definition of the natural log function to prove that,
for x > 0, x(1 + x) > (1 + x) ln(1 + x) > x. Hence show that ln(1 + x) >
x − x2 + x3 − x4 + · · · and that as
ln(1 + x)
x → 0, →1
x

1+x
1
Z
Solution. NB: ln(1 + x) = dt, for x > 0 ⇒ x + 1 > 1
1 t
1 1
Let 1 < t < 1 + x ⇒ < < 1.
1+x t
Integrating with respect to t gives
Z 1+x Z 1+x Z 1+x
1 1
dt < dt < dt
1 1+x 1 t 1

x
⇒ 1+x
< ln(1 + x) < x.......(1)

x < (1 + x) ln(1 + x) < x(1 + x) ∴ x(1 + x) > (1 + x) ln(1 + x) > x

From (1):,
x
ln(1 + x) > ⇒ ln(1 + x) > x(1 − x + x2 − x3 + · · · )
1+x

⇒ ln(1 + x) > x − x2 + x3 − x4 + · · ·
Also from (1):

1 ln(1 + x) 1 ln(1 + x)
< < 1 ⇒ lim < lim <1
1+x x x→0 1 + x x→0 x
ln(1 + x)
⇒ 1 < lim < 1.
x→0 x
ln(1+x) 1
∴ By the squeeze theorem: lim x
→ 1 ⇒ ln lim (1 + x) x ⇒ ln e = 1.
x→0 x→0

1
Exercise. 1. Show that for all x > 0, x − x2 < ln(1 + x), and for all x > 1,
2
x−1
we have < ln x < x − 1
x

55 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

56 Dr. Joseph K. Ansong

Chapter 5

Indeterminate Forms and

L’Hopital’s Rule

A function f (x) is said to be continuous at a point a, if lim f (x) = f (a).

x→a
However, some limits cannot be evaluated by a simple substitution. For
2x sin x
example, lim = lim 2 = 2 and lim = 1. In each example, a direct
x→0 x x→0 x→0 x
substitution results in the indeterminate form 00 . Simarlarly, using direct
10x 10
substitution for lim 2 = lim = 0 results in the indeterminate form
x→∞ x x→∞ x
∞
.
∞

f (x)
Definition 8. If lim f (x) = 0 and lim g(x) = 0, then the limit lim is
x→a x→a x→a g(x)
0
called an indeterminate form of the type .
0

f (x)
Definition 9. If lim f (x) = +∞ and lim g(x) = +∞, then lim is said
x→a x→a x→a g(x)
∞
to be an indeterminate form of the type .
∞

5.1 L’Hopital’s Rule

Theorem 17. (L’Hopital’s Rule) Suppose f and g are differentiable on an
open interval I that contains a, with the possible exception of a itself, and
0 f (x) 0
g (x) 6= 0 for all x in I. If lim is an interminate form of the type or
x→a g(x) 0

57
Math 223 : Calculus II J.K.A

∞
then
∞
0
f (x) f (x)
lim = lim 0 (5.1)
x→a g(x) x→a g (x)

provided the limit on right-hand side exists or is infinite.

d
[f (x)]
 
f (x) dx d f (x)
Remark. (1) lim = lim d 6= lim
x→a g(x) x→a [g(x)] x→a dx g(x)
dx
[Don’t use the Quotient Rule!!]
0 ∞
 
(2) Make sure the limit does have one of the indeterminate forms or
0 ∞
before applying L’Hopital’s Rule!
Example 25. 1) Use L’Hopital’s rule to evaluate the following:
√
x3 + x2 − 2x 9 − 3x − 3
(a) lim (b) lim
x→1 x−1 x→0 x
2) [Repeated application of L’Hopital’s rule]:
ex − x − 1 x3
(a) lim (c) lim
x→0 x2 x→∞ e2x
3 2
x − 3x + 4
(b) lim 4
x→2 x − 4x3 + 7x2 − 12x + 2

3. Evaluate the following:

ln x sin πx 1 + tan x
(a) lim (b) lim+ √ (c) lim
x→∞ x x→1 x−1 π−
x→ 2 sec x

5.2 Indeterminate Forms ∞ − ∞ and 0 − ∞

Definition 10. If lim f (x) = ∞ and lim g(x) = ∞, then the limit lim [f (x)−
x→a x→a x→a
g(x)] is said to be an indeterminate fom of the type ∞ − ∞.
Definition 11. If lim f (x) = 0 and lim g(x) = +∞, then lim f (x)g(x) is
x→a x→a x→a
said to be indeterminate form of the type 0 − ∞.
Remark. Indeterminate forms of the ∞ − ∞ and 0 − ∞ can be expressed
as one of the type 00 or ∞
∞
by algebraic manipulation.
Example 26. Evaluate the following:

58 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

1
 
2 1 1
(a) lim x sin 2 (c) lim+ − x
x→∞ 4x x→0 x e −1
√
(b) lim (x − x2 − 3x) (d) lim+ x ln x
x→∞ x→0

5.3 Indeterminate Forms of 00, ∞0, 1∞

The limit lim [f (x)]g(x) is said to be an indeterminate form of the type:
x→a

I. 00 if lim f (x) = 0 and lim g(x) = 0

x→a x→a

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

x→a x→a

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

x→a x→a

Remark. The above indeterminate forms can usually be converted to the

type 0 · ∞ by taking logarithm or using [f (x)]g(x) = eg(x) ln f (x) .
Procedure:

(i) Evaluate lim g(x) ln f (x) = L

x→a

(ii) Then lim [f (x)]g(x) = eL

x→a

Example 27. (1) Evaluate:

 x  
(a) lim+ X x 1 1
x→0 (b) lim 1 + (c) lim+
x→∞ x x→0 x

(2) Evaluate the following:

1 r
(a) lim (1 + 2h) h 1−x
h→0 (c) lim+ (sin x)
x→0 x
 ln x
ln(1 + x) 1
(b) lim (d) lim 1 +
x→0 x x→∞ x

5.4 Scales of Infinity(Order of Magnitude)

5.4.1 Motivation
Consider the solution to the following limits.

59 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

3x3 + 5
(1) lim
x→∞ 4x2 + x
3x + x52
=⇒ lim =∞
x→∞ 4 + 1
x

The limit going to infinity implies that the numerator, 3x3 + 5, is grow-
ing much faster than the denominator, 4x2 + x.
3x + 5
(2) lim
x→∞ 4x2 + 3x + 1
3/x + 5/x2
=⇒ lim =0
x→∞ 4 + 3/x + 1/x2

This implies that the denominator, in this case, is growing faster than
the numerator. Equivalently, we can say that the numerator is growing
more slowly than the denomintor.
3x3 + 5
(3) lim
x→∞ 4x3 + x
5
3+ x3 3
=⇒ lim 1 =
x→∞ 4 +
x2
4
Unlike the previous two examples, the numerator and denominator have
similar rates of growth, because the limit of their ratio is a constant.

In the examples, the limits are determined by the order of magnitude of the
polynomials in the numerator and denominator. In other words, the limits
could have been obtained if we considered only the term with highest order
of magnitude in the numerator and the denominator.

Motivational Questions
The main question to address is this: Apart from polynomials, is it possible
to determine the limit f (x)/g(x) of the two function f (x) and g(x) as x
approaches infinity or zero?
Why is this important? An answer to the question could tell us whether f (x)
is growing faster or slower than g(x). Think of the economies of two countries,
say Ghana and Nigeria, that are both growing very fast. Which economy is
growing more rapidly? Conversely, if both economies are decaying, can we
tell which one is decaying more slowly?
Consider the functions f (x) = x2 and g(x) = x. Notice that both func-
tions grow without bound as x becomes infinitely large (see Figure 5.1). But
f grows rapidly than g. In general, how can you tell which function is growing

60 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Figure 5.1: Graph of ln x

(decaying) more rapidly (slowly) than the other? The following definitions
and theorems will help in answering the question above.
Definition 12. (Order of Magnitude) A function f (x) is said to be of the
f (x)
k-th order of magnitude if k → L 6= 0 as x → ∞, where L is a constant.
x
Definition 13. (Relative Rates of Growth) Let f, g : R → R, a  R∗ (R∗ =
R ∪ {−∞, ∞}), and lim |f (x)| = lim |g(x)| = ∞.
x→a x→a
f (x)
a) We say f approaches ∞ on a higher order of magnitude than g if lim =
x→a g(x)
∞ or we can say that f (x) → ∞ faster than g(x), or g(x) → ∞ slower than
f.
b) We can say f approach ∞ on a lower order of magnitude than g(x) if
f (x)
lim = 0 or we can say that f (x) → ∞ slower than g(x) or g(x) → ∞
x→a g(x)
faster than f (x).
f (x)
c) We say f and g approaches ∞ on the same order of magnitude if lim =
x→a g(x)
C, C  R, C 6= 0.

Theorem: As x → ∞, the magnitude of the exponential, power and

logarithmic functions approach ∞. Moreover, the order of magnitude with
which they approach ∞ is given in the following order, with exponential
functions approaching the fastest, and logarithmic functions approaching the
slowest.
(1) Exponentials of the form ax , where a > 0, a 6= 1. If a > b then ax
approaches ∞ faster than bx as x → ∞.

61 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

(2) Power functions, xn , nN. If n > m, then xn approaches ∞ faster than

xm .

(3) Log functions, loga (x) for a > 0 and a 6= 1. All log functions approach
loga x ln(b)
∞ at the same rate, because = .
logb x ln(a)
If any of the above functions is multiplied by a non-zero constant , it
does not change the relative order of magnitude at which the function
approaches ∞. 5

Example
ex
1. lim = ∞, since ex grows faster than ln x as x → ∞.
x→∞ ln x
ln x
2. lim x = 0.
x→∞ e

NB: For a linear combination (sums and differences) of these functions, only
the dominant term matters.

Definition:Dominant term
Suppose f can be written as a linear combination of the functions {f1 , f2 , ..., fn }
and lim |f (x)| = ∞, where aR∗ . The dominant term of f is the function fi
x→a
which approaches ∞ on the highest order of magnitude. Denote this by f˜.

Theorem: Dominant term

Consider f (x) and g(x), with lim |f (x)| = lim |g(x)| = ∞. If f˜ and g̃ are
x→a x→a
f (x) f˜(x)
the dominant terms of f and g respectively, then lim = lim . eg:
x→a g(x) x→a g̃(x)
ex − 12x2 + x ex
lim = lim = ∞.
x→∞ x4 + 1 x→∞ x4

Theorem: Ranking growth rates as x → ∞

Let f << g mean that g grows faster than f as x → ∞ with positive real
numbers p, q, r and s and b > 1, then lnq x << xp << xp lnr x << xp+s <<
bx << xx .

5.4.2 Relative Rates of Decay

Definition: Relative rates of Decay
Let f, g : R → R, aR∗ , and lim |f (x)| = lim |g(x)| = 0.
x→a x→a

62 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

a. We say f approaches 0 on a higher order of magnitude than g if lim | fg(x)

(x)
|=
x→a
0.
b. We say f approaches 0 on a lower order of magnitude than g if lim | fg(x)
(x)
|=
x→a
∞.
c. We say f and g approach 0 on the same order of magnitude if lim | fg(x)(x)
|=
x→a
C, C  R, c 6= 0.

NB: Because ex grows faster than x2 as x → ∞, it also implies that e−x

decays faster than x−2 as x → ∞. So looking at the reciprocals of the ex-
ponential and power functions helps in determining their relative rates of
decay.
e−x + 1 e−x
 2 
1 x 2
Example 28. 1. lim = lim −2 + −2 = lim +x =∞
x→∞ x−2 x→∞ x x x→∞ ex
ex − 12x2 + x
2 Evaluate lim .
x→∞ x4 + 1
3. Show that the exponential function ex increases more rapidly as x → +∞
xn
than any fixed power of x. That is lim x = 0, n > 0 .....(1)
x→∞ e
ex
or lim n = +∞, n > 0.....(2)
x→∞ x

Solution
ex
  x 
x n
ln = ln e − ln x = x − n ln x = ln x − n
xn   ln x
ex  x 
∴ lim ln = lim ln x · lim −n =∞·∞=∞
x→∞ xn x→∞ x→∞ ln x
ex
∴ lim n = ∞
x→∞ x

2. Show that the function ln x increases more slowly than any positive inte-
ln x
gral power of x. That is: lim n = 0, x > 0, n > 1 (n is an integer).
x→∞ x
Solution
x
1 Rx 1 Rx 1 Rx 1
NB: If t > 1, 6 11 ⇒ ln x = 1 dt 6 1 1 dt = 1 t−1/2 dt = 2t 2 =
t t2 t t2 1
1
2x 2 − 2
1 ln x 2 2
∴ ln x 6 2x 2 − 2 ⇒ 6 1 −
x x2 x
ln x ln x 2 2 ln x
x>1:0< n 6 6 1 − . As x → ∞, 0 < lim n 6 0.
x x x2 x x→∞ x
ln x
∴ By squeeze(sandwich) theorem: lim n = 0. This implies: xn > ln x.
x→∞ x

63 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Examples
d k d
1. Prove the following: a). (ln x)k = b). (ln ln x)k =
dx x(ln x)1−k dx
k
x ln x(ln ln x)1−k
ln ln x ln2 x ln3 x
2. Find: a) lim = lim b) lim
x→∞ ln x x→∞ ln x x→∞ ln x
3.Determine
√ √which function has a higher order of magnitude: a). f (x) =
3
x, g(x) = √x √ √
b). f (x) = ln x, g(x) = 3 ln x c). f (x) = x, g(x) = ln3 x
x
4. Arrange the functions according to their order of magnitude: g = √ ,
√ ln x
x ln x x ln ln x x ln ln ln x
f= ,h= √ ,q= √
ln ln x ln x ln ln x
1 3−x 4x2 − 9x − 9
5. Find: a). lim b). lim c). lim
x→−4 (x + 4)4 x→3 (x − 3)4 x→3 x−3
√ √
6. Show that for large values of x, x > e ln x > ln3 x

2x3 − x2 + x
7. lim
x→0 x3 + 2x
2x3 − x2 + x + 1
8. lim
x→0 x3 + 2x + 2

64 Dr. Joseph K. Ansong

Chapter 6

Hyperbolic Functions

The analysis of certain problems such as the shape of a hanging cable and
pursuit curve of a missile involves combinations of exponential functions such
as e−cx and ecx , where c is a constant. Because they occur frequently in
maths, these combinations have been given the special name of hyperbolic
functions. They are related to the hyperbola, just as trigonometric functions
are related to the circle. Before defining hyperbolic functions, we first take
a look at odd and even functions.

6.1 Odd and Even Functions

Definition 14. A function is said to be even if f (−x) = f (x) and odd if
f (−x) = −f (x) for all x in its domain.
For example, f (x) = x2k is an even function for some kZ and f (x) =
x2k+1 is an odd function for the same kZ.

Symmetry
(i) Even functions are symmetric about the coordinate(vertical) axis. eg:
f (x) = x2 + 2

(ii) Odd functions are symmetric about the origin. eg: f (x) = x3 . See
Figure 6.1.

1
Theorem 18. Let f (x) be any given function. Then the functions (f (x) + f (−x))
2
1
and (f (x) − f (−x)) are even and odd functions respectively. That is, every
2
function can be written as the sum of an even and an odd function.

65
Math 223 : Calculus II J.K.A

Figure 6.1: Graph of (a) y = x2 + 2 (even), and (b) y = x3 (odd).

Proof.
1 1
Let fe (x) = (f (x) + f (−x)) and fo (x) = (f (x) − f (−x))
2 2
1 1
fe (−x) = (f (−x) + f (x)) = (f (x) + f (−x)) = fe (x)
2 2
1 1
fo (−x) = (f (−x) − f (x)) = − (f (x) − f (−x)) = −fo (x)
2 2
1 1
Now, fe (x) + fo (x) = [f (x) + f (−x)] + [f (x) − f (−x)]
2 2
1 1
= [f (x) + f (−x) + f (x) − f (−x)] = [2f (x)] = f (x)
2 2

6.1.1 Basic Properties of Even and Odd Functions

a. The sum or difference of two even (odd) functions is even (odd), and
any constant multiple of an even (odd) function is even (odd).

b. The product of two even functions is even, and the product of two odd
functions is even. The product of an even and odd function is odd.

c. The quotient of two even or two odd functions is an even function, but
the quotient of an even and an odd function is odd.

d. The derivative of an even function is odd and the derivative of an odd

function is even.

66 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Ra Ra Ra
e. For any aR, we have f (x) = 0 and
−a o
f (x) = 2
−a e 0
fe (x)

6.1.2 Hyperbolic functions: Definitions

1 1
Consider the function f (x) = ex . Then ex = (ex + e−x ) + (ex − e−x ) and
2 2
−x 1 x −x 1 x −x 1 x −x
e = (e +e )− (e −e ). Note that (e +e ) is even and 12 (ex −e−x )
2 2 2
is odd.

Definition 15. (DEFINITIONS OF THE HYPERBOLIC FUNCTIONS)

ex + e−x 1
cosh x = sech x =
2 cosh x

ex − e−x 1
sinh x = csch x =
2 sinh x
sinh x 1
tanh x = coth x =
cosh tanh x

Applications

Hyperbolic functions have several applications in science and engineering.

A well known application is the use of the hyperbolic cosine to describe
the shape of a hanging cable. If a heavy flexible wire (or cable; such as a
telephone or power line) is suspended between two points at the same height,
it takes the shape of a curve with equation y = A + B cosh (x/B) (see Figure
6.2a).
A second application occurs in the description of the velocity of ocean
waves. The velocity of a water wave with wave-length L moving across the
ocean (or a water body) with average depth d, is modeled by the function
s  
gL 2πd
v= tanh
2π L

where g is the acceleration due to gravity (see Figure 6.2b). Besides the two
physical applications mentioned above, hyperbolic functions also have several
mathematical applications. Most especially, they are often used to simplify
complicated expressions under the techniques of integration. We would see
some of these uses later in the course.

67 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Figure 6.2: (a) A hanging cable with equation of the form y = A +

B cosh (x/B). (b) Schematic of a water wave.

Hyperbolic cosine (cosh) and Hyperbolic sine (sinh)

Definition 16. The even function cosh : R → [1, ∞) is defined as follows:
ex + e−x 1 e2x + 1 1 1 + e−2x
   
cosh x = = = , xR
2 2 ex 2 e−x
The odd function sinh : R → (−∞, ∞) is defined by:
ex − e−x 1 e2x − 1 1 1 − e−2x
   
sinh x = = = , xR
2 2 ex 2 e−x

6.1.3 Graphs
1 1
(1) Use the additions of ex and e−x to graph the cosh and sinh functions.
2 2
(2) You may also use the approach of curve sketching in calculus I to sketch
the graphs of hyperbolic functions. See Figure 6.3.

sinh x
Example 29. Graph of tanh x. Let f (x) = tanh x =
cosh x
ex − e−x 1 − e−2x e2x − 1
=⇒ f (x) = = = 2x
ex + e−x 1 + e−2x e +1
1−1 e2x − 1
f (0) = = 0, f (x) = 0 ⇒ 2x = 0 ⇒ e2x = 1 ⇒ x = 0
2 e +1
∴ (0, 0) is an intercept on both axes.
1 − e−2x
As x → ∞, f (x) = = 1.
1 + e−2x
2x
e −1
As x → −∞, f (x) = 2x = −1.
e +1
So f (x) = ±1 are horizontal asymptotes.

68 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Figure 6.3: Sketch of (a) y = sinh x and (b) y = cosh x.

2ex
 
0 2
We can show that f (x) = sech x = > 0 ⇒ No turning points
e2x + 1
00
Also, f (x) = 0 ⇒ x = 0 ∴ (0, 0) is a point of inflection. See the graph in
Figure 6.4a. The graphs of other hyperbolic functions are displayed in Figure
6.4b-d.

6.2 Hyperbolic Identities and Osborn’s Rule

(a) sinh(−x) = − sinh x

(b) cosh(−x) = cosh x

(c) cosh2 x − sinh2 x = 1

(d) sech2 x = 1 − tanh2 x

(e) 1 + csch2 x = coth2 x

ex − e−x
Proof. (a) sinh x =
2
e−x − e+x ex − e−x
 
⇒ sinh(−x) = =−
2 2

= − sinh x

(b) Try it.

69 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Figure 6.4: Graph of (a) y = tanh x (b) y = csch x (c) y = sech x (d)
y = coth x.
2 2
ex + e−x ex − e−x
 
2 2
(c) cosh x − sinh x = −
2 2
e2x + 2 + e−2x e2x − 2 + e−2x 4
= − = =1
4 4 4
(d) sech2 x = 1 − tanh2 x
From (c), we have
cosh2 x sinh2 x 1
cosh2 − sinh2 = 1 ⇒ 2 − 2 =
cosh x cosh cosh2 x
=⇒ 1 − tanh2 x = sech2 x
Alternatively,
1 sinh2 x
sech2 x + tanh2 x = +
cosh2 x cosh2 x
 x −x 2
1 + sinh2 x 1 + e −e 2
= 2 =
2
cosh x x
e +e−x
2

70 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

4+e2x −2+e−2x
4
= e2x +2+e−2x
4

e2x + 2 + e−2x
∴ sech2 x + tanh2 x = =1
e2x + 2 + e−2x

(e) 1 + csch2 x = coth2 x

From c.: cosh2 x − sinh2 x = 1
cosh2 x sinh2 x 1
2 − 2 = ⇒ coth2 x − 1 = csch2 x
sinh x sinh x sinh2 x

∴ 1 + csch2 x = coth2 x

Exercise. Prove the following:

1. sinh(2x) = 2 sinh(x) cosh(x)

2. cosh(2x) = cosh2 x + sinh2 x

3. sinh(x + y) = sinh(x) cosh(x) + cosh(x) sinh(x)

3. cosh(x + y) = cosh(x) cosh(y) + sinh(x) sinh(y)

6.2.1 Relationship to the Exponential function

It can be shown that:

(i) ex = cosh(x) + sinh(x)

(ii) e−x = cosh(x) − sinh(x)

These are very similar to trigonometric expressions using Euler’s formula:

eix = cos(x) + i sin(x) (6.1)

e−ix = cos(x) − i sin(x) (6.2)

(6.1) + (6.2) gives

eix + e−ix
cos x = = cosh(ix) (6.3)
2

71 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

(6.1) - (6.2) results in

eix − e−ix eix − e−ix
 
sin x = = −i = −i sinh(ix) (6.4)
2i 2
Thus, (6.3) implies that
cos(iy) = cosh(−y) = cosh(y)
and (6.4) means that
sin(iy) = −i sinh(−y) = i sinh(y)
So
cosh(y) = cos(iy) and sinh(y) = −i sin(iy)
Thus, trigonometric formulae can be converted to formulas for hyperbolic
functions using Osborn’s rule:

6.2.2 Osborn’s Rule

cos can be converted to cosh and sin into sinh, except when there is a product
of two sines, then a sign change must be effected. That is
cos → cosh and sin → i sinh
Example 30. 1) cos(2x) = 1 − 2 sin2 x ⇒ cosh(2x) = 1 + 2 sinh2 x.
2) sin(2x) = 2 sin(x) cos(x) ⇒ sinh(2x) = 2 sinh(x) cosh(x)

6.3 Inverse Hyperbolic Functions

From the graphs of the hyperbolic functions, we notice that all of them are
one-to-one functions on (−∞, ∞) except cosh x and sech x. However, if we
restrict the domain of cosh x and sech x to [0, ∞). They become one-to -one
with inverses cosh−1 x and sech−1 x.

Definitions Domain
y = sinh−1 x → (−∞, ∞)
y = cosh−1 x → [1, ∞)
y = tanh−1 x → (−1, 1)
y = csch−1 x → (−∞, 0) ∪ (0, ∞)
y = sech−1 x → (0, 1]
y = coth−1 x → (−∞, −1) ∪ (1, ∞)

72 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Graphs
You may sketch the graph of the inverse hyperbolic functions by reflecting
the graph of the corresponding hyperbolic function about the line y = x (see
the section on Inverse Functions). The graphs of some inverse hyperbolic
functions are displayed in Figure 6.5.
Exercise. Sketch the graph of y = sech−1 x.

6.3.1 Representation of Inverse Hyperbolic Functions

as Logarithmic Functions
For some applications, it is more convenient to represent hyperbolic functions
as logarithmic functions. The following exercises illustrates some of these
representations and how to proof them.
Example 31. Prove that:
√
(1) sinh−1 x = ln(x + x2 + 1); (−∞, ∞)
√
(2) cosh−1 x = ln(x + x2 − 1); [1, ∞)
 
1 1+x
(3) tanh−1 x = ln ; (−1, 1)
2 1−x
√
Proof. (1) sinh−1 x = ln(x + x2 + 1). Let
ey − e−y
y = sinh−1 x ⇒ x = sinh y =
2
e2y − 1
=⇒ x = y
⇒ 2xey = e2y − 1 ⇒ e2y − 2xey − 1 = 0
2e
2
Let t = e , then t − 2xt − 1 = 0 is a quadratic in t = ey .
y

√
2x ± 4x2 + 4
⇒t=
2
√
=⇒ ey = x ± x2 + 1
√ √
. The 2 but x − x2 + 1 is not, since
√ root x + x + 1 is admissible
√
x − x2 + 1 < 0 because x < x2 + 1; however ey > 0.
√
∴ e y = x + x2 + 1
√
⇒ y = ln(x + x2 + 1)
√
∴ sinh−1 x = ln(x + x2 + 1)

73 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Figure 6.5: Graphs of hyperbolic functions (left panels) and their inverse
functions (right panels). The dashed curves (left panels) indicate the re-
flection of the hyperbolic function (i.e. the inverse function) about the line
y = x.

√
(2) cosh−1 x = ln(x + x2 − 1), [1, ∞)
Let y = cosh−1 x ⇒ x = cosh y

74 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

ey + e−y e2y + 1
=⇒ x = =
2 2ey
⇒ e2y − 2xey + 1 = 0
√
y 2x ± 4x2 − 4
⇒e =
2
y
√
∴ e = x ± x2 − 1
Note that

(x − 1) < x + 1 ⇒ (x − 1)(x − 1) < (x − 1)(x + 1) ⇒ (x − 1)2 < (x2 − 1)

√ √
x − 1 < x2 − 1 ⇒ x − x2 − 1 < 1
√
⇒ y = ln(x − x2 − 1)
√
⇒ y = cosh−1 x < 0, since x − x2 − 1 < 1. √
But cosh−1
√ x > 0 for all x > 1. So we discard x − x2 − 1 and retain
y 2
e = x + x − 1.
√
∴ y = cosh−1 x = ln(x + x2 − 1)
√
An alternative approach to discarding x − x2 − 1 is given as follows:
√
y
√ √ (x + x2 − 1)
2
e = x − x − 1 = (x − x − 1) 2 √
(x + x2 − 1)

x2 − (x2 − 1) 1 √
⇒ ey = √ = √ = (x + x2 − 1)−1
x + x2 − 1 x + x2 − 1
√
=⇒ y = cosh−1 x = − ln(x + x2 − 1) < 0
√
. But by definition, cosh−1 x > 0 ∀x > 1, so we discard x − x2 − 1.
 
−1 1 1+x
(3) tanh x = ln , (−1, 1)
2 1−x
Let y = tanh−1 x
ey − e−y
⇒ x = tanh y = y
e + ey
e2y − 1
⇒ x = 2y ⇒ xe2y + x = e2y − 1
e +1
⇒ e2y (x − 1) = −(x + 1)
−(x + 1)
 
2y 1+x
⇒e = =
(x − 1) 1−x

75 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A



1+x
⇒ 2y = ln
1−x
 
−1 1 1+x
⇒ y = tanh x = ln
2 1−x

Exercise. (a) Show that lim tanh(ln x) = 1.

x→∞

cosh(ln x) + sinh(ln x)
(b) Simplify: [Ans: x2 ]
cosh(ln x) − sinh(ln x)

6.4 Derivatives of hyperbolic functions

The derivatives of hyperbolic functions are easily computed because they are
defined in terms of ex and e−x .

Example 32. Show that

d d
(a) (sinh x) = cosh x (c) (tanh x) = sech2 x
dx dx
d d
(b) (cosh x) = sinh x (d) (coth x) = − csch2 x
dx dx
ex − e−x 1 1
Solution. (a) y = sinh x = = ex − e−x
2 2 2
dy 1 1 ex − e−x
= ex + e−x = = cosh x
dx 2 2 2

sinh x
(c) y = tanh x =
cosh x
dy cosh x(cosh x) − sinh x(sinh x)
⇒ =
dx cosh2 x
cosh2 x − sinh2 x
=
cosh2 x
1
= = sech2 x
cosh2 x
d
∴ (tanh x) = sech2 x
dx

76 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

1 cosh x
(d) y = coth x = =
tanh x sinh x
dy sinh2 x − cosh2 x
=⇒ =
dx sinh2 x
−1
=
sinh2 x
= − csch2 x

6.5 Derivatives of Inverse Hyperbolic Func-

tions
Procedure:
(a) Write the function as a log function and differentiate directly, or

(b) Use implicit differentiation.

Example 33. √
y = sinh−1 x = ln(x + x2 + 1)
 
dy 1 1 2 − 12
⇒ = √ 1 + (x + 1) · (2x)
dx x + x2 + 1 2
 
1 x
= √ · 1+ √
x + x2 + 1 x2 + 1
√
1 x + x2 + 1
= √ · √
x + x2 + 1 x2 + 1
1
=√
x2 + 1
Alternatively:
y = sinh−1 x ⇒ sinh y = x
Differentiate:
dy
cosh y ·=1
dx
dy 1
⇒ =
dx cosh y
But cosh2 y − sinh2 y = 1

⇒ cosh2 y = 1 + sinh2 y

77 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

q
⇒ cosh y = 1 + sinh2 y
√
= 1 + x2
dy 1
∴ =√
dx 1 + x2
Exercise. Show that
d 1
(a) cosh−1 x = √
dx 2
x −1
d 1
(b) tanh−1 x = [Note: sech2 y = 1 − tanh2 y]
dx 1 − x2
d 1
(c) coth−1 x = 2
dx x −1

78 Dr. Joseph K. Ansong

Chapter 7

Infinite Sequences

7.1 Definitions & Notation

Definition 17. A sequences is a list of numbers written in a definite order:

a1 , a2 , . . . , an , . . .

where a1 = first term, a2 = second term,. . . an = nth term.

A sequence {an } may also be defined as a function f (n) = an whose
domain is the set of positive integers.

Notation
The sequence a1 , a2 , . . . may be denoted by {an }∞
n=1 or simply {an }.
Some sequences have a simple definition with a formula for the nth term,
others are not so straightforward to define.
Example 34. 1. Simple definitions:
 ∞
2n 2n
(i) , an = .
n+2 1 n+2

∞
(−1)n+1 (n + 1) (−1)n+1 (n + 1)

(ii) , an =
2n 0 2n
The sequences above may be written as shown below, displaying the
first few terms:
 ∞  
2n 2 4 6 8
= , , , ,···
n+2 1 3 4 5 6

79
Math 223 : Calculus II J.K.A

∞
(−1)n+1 (n + 1)
  
2 3 4
= −1, , − , , · · ·
2n 0 2 4 8
The terms of this sequence show alternating signs of positive and neg-
ative terms. Such a sequence is called an alternating sequence.

2. Not so simple definitions:

The Fibonacci sequence {fn } is defined by

f1 = 1, f2 = 1
fn = fn−1 + fn−2 , n ≥ 3

The first few terms are given by {1, 1, 2, 3, 5, 8, 13, . . .}.

Graphs
The first few terms of an infinite sequence may be pictured by plotting the
points (a, an ), as displayed in Figure 7.1, for some of our previous examples.

Figure 7.1:

Exercise. (1) List the first 5 terms of each of the sequence {an }.
n+1 nπ
(a) an = (d) an = sin
3n − 1 2
3(−1)n
(b) an = (e) a1 = 3, an+1 = 2an − 1
n!
(−1)(n+1) 2n
(c) an = (f) a2 = 2, an+1 = 3an − 1
n+1
(2) Find a formula for the general term of the sequence, assuming that the
pattern of the first few terms continues.

80 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

   
1 2 3 4 5 1 1 1 1
(a) , , , , ,... (c) 1, , , , , . . .
2 3 4 5 6 3 5 7 9
 
1 1 1 1
(b) −1, , − , , − , . . . (d) {5, 1, 5, 1, 5, 1, . . .}
2 6 24 120

7.1.1 Convergence & Divergence of a Sequence

We observe from Figure 7.1a that the sequence an approaches the number
2 as n gets larger and larger. In order words, we can make the difference,
2 − an , as small as we like by taking n sufficiently large. This written as

2n
lim =2
n→∞ n + 2

Similarly, we notice that the sequence in Figure 7.1b approaches 0 as n get

larger. Thus, the limit as n → ∞ of that sequence is 0. A working or intuitive
(or less precise) definition of the limit of a sequence is given below.

Definition 18. A sequence {an } has the limit L, and written as

lim an = L,
n→∞

if we can make the terms an as close to L as we like by taking n sufficiently

large. If limn→∞ an exists, we say the sequence converges or is convergent.
Otherwise the sequence diverges or is divergent.

We now give a more precise definition of the limit of a sequence.

Definition 19. (Precise defintion) A sequence {an } converges and has

limit L, written as
lim an = L,
n→∞

if for every  > 0 there exists a positive integer N such that |an − L| < 
whenever n > N .

An illustration of this precise definition is displayed in Figure 7.2, showing

that for any , the sequence {an } must always lie between the horizontal lines
L −  and L +  for some n > N . Usually, the smaller the value of , the
larger the N .

81 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Figure 7.2:

7.1.2 Some Useful Theorems:

We notice that the precise definition of the limit of a sequence is similar to
that of the limit of a function at infinity. The main difference between a
function y = f (x) with domain (0, ∞) and that of a sequence {an } defined
as f (n) = an is that n is an integer. Theorem 19 below summarizes this idea.
Theorem 19. If limx→∞ f (x) = L and f (n) = an when n is an integer, then
limn→∞ an = L.
1
Example 35. Since we know that lim r = 0 for r > 0, it follows by
x→∞ x
theorem 19 that
1
lim r = 0 (r > 0)
n→∞ n

ln n
Example 36. Find lim .
n→∞ n

Solution. Notice that as n → ∞ both the numerator and denominator

approach ∞. We cannot apply L’Hopital’s rule directly since doesn’t appy
to sequences but to functions of real variables. But we can apply it to the
related function f (x) = ln x/x. Thus
ln x 1
lim = lim = 0
x→∞ x x→∞ x

By theorem 19, then implies that

ln n
lim = 0.
n→∞ n

Limit Laws for Sequences

If {an } and {bn } are convergent sequences and c is a constant, then
1. lim (an + bn ) = lim an + lim bn
n→∞ n→∞ n→∞

82 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

2. lim (an − bn ) = lim an − lim bn

n→∞ n→∞ n→∞

3. lim can = c lim an

n→∞ n→∞

4. lim (an · bn ) = lim an · lim bn

n→∞ n→∞ n→∞

an limn→∞ an
5. lim =
n→∞ bn limn→∞ bn
h ip
p
6. lim an = lim an if p > 0 and an > 0.
n→∞ n→∞

Theorem 20. (Squeeze theorem) If an ≤ bn ≤ cn for n ≥ N and

limn→∞ an = limn→∞ cn = L, then

lim bn = L
n→∞

The squeeze (or sandwich) theorem is illustrated in Figure 7.3.

Figure 7.3:

n!
Example 37. Find lim .
n→∞ nn
Solution. Let an = n!/nn . The first few terms of {an } are

1! 2! 2·1 3! 3·2·1
a1 = = 1, a2 = = , a3 = =
1 2 2·2 3 3·3·3
83 Dr. Joseph K. Ansong
Math 223 : Calculus II J.K.A

and the nth term is given by

1 · 2 · 3 · ··· · n 2 · 3 · ··· · n
 
n! 1
an = n = =
n n · n · n · ··· · n n n · n · ··· · n
Note that the expression in parentheses is at most 1. That is
2 · 3 · ··· · n
 
≤1
n · n · ··· · n
1 2 · 3 · ··· · n
 
1
=⇒ an = ≤ .
n n · n · ··· · n n
Thus, an ≤ 1/n, and because an > 0, we obtain the inequality
1
0 < an ≤
.
n
Since limn→∞ 1/n = 0, the Squeeze Theorem gives
n!
lim an = lim = 0.
n→∞ n→∞ nn

Some sequences cannot be written as functions that we can take the limit
of. This is especially true of the alternating sequence that we saw earlier.
The next theorem will help in finding the limit of such sequences.
Theorem 21. If limn→∞ |an | = 0, then limn→∞ an = 0.
Proof. We note that
−|an | ≤ an ≤ |an |.
Also, we have
lim (−|an |) = − lim |an | = 0.
n→∞ n→∞
Thus, we have lim (−|an |) = 0 = lim |an |, and so by the Squeeze Theorem,
n→∞ n→∞
we have
lim an = 0.
n→∞

(−1)n
Example 38. Find lim , if it exists.
n→∞ n
Solution. We have
(−1)n 1
lim = lim = 0
n→∞ n n→∞ n

Hence, by Theorem 21, we have

(−1)n
lim = 0.
n→∞ n

84 Dr. Joseph K. Ansong

 Math 223 : Calculus II J.K.A

The following theorem says that, given a convergent sequence {an }, if we

apply a continuous function to the terms of the sequence, f (an ), we again
obtain a convergent sequence.
Theorem 22. If lim an = L and the function f is continuous at L then
n→∞

lim f (an ) = f (L)

n→∞

Example 39. Find lim cos(π/n).

n→∞

Solution. Let an = π/n. Then cos(π/n) = f (an ) where f (x) = cos x. Now
lim π/n = 0, and since the cosine function is continuous at 0, we have
n→∞
 
lim cos(π/n) = cos lim π/n = cos(0) = 1.
n→∞ n→∞

Example 40. Find lim esin(π/n) .

n→∞

Solution. Let an = sin(π/n), then esin(π/n) = f (an ) where f (x) = ex . Now,

lim sin(π/n) = 0
n→∞

and since f is continuous at 0, we have

lim esin(π/n) = elimn→∞ sin(π/n) = e0 = 1
n→∞

Example 41. Show that the sequence {rn }∞ 0 converges if −1 < r ≤ 1 and
diverges for all other values of r. Also show that

n 0 if − 1 < r < 1
lim r = (7.1)
n→∞ 1 if r = 1
Solution. The proof/solution is given below as a series of different cases.
Note that, from Calculus I, we have
lim ax = ∞ if a > 1, and lim ax = 0 if 0 < a < 1.
n→∞ n→∞

Case 1: For r > 1

In this case, since limx→∞ rx = ∞ for r > 1, we apply Theorem 19 to
get lim rn = ∞. So the sequence diverges for r > 1.
n→∞

Case 2: For r = 1

lim rn = lim 1n = lim 1 = 1.

n→∞ n→∞ n→∞
So the sequence converges for r = 1 and the limit is equal to 1.

85 Dr. Joseph K. Ansong

 Math 223 : Calculus II J.K.A

Case 3: For 0 < r < 1

From Calculus I, limx→∞ rx = 0 if 0 < r < 1, so by Theorem 19, we
get lim rn = 0. So the sequence converges if 0 < r < 1 and the limit
n→∞
is equal to 0.

Case 4: For r = 0

lim rn = lim 0n = lim 0 = 0

n→∞ n→∞ n→∞

So the sequence converges for r = 0, and the limit is equal to 0.

Case 5: For −1 < r < 0

Note that if −1 < r < 0 then 0 < |r| < 1. Thus, we have
 n
lim |rn | = lim |r|n = lim |r| = 0n = 0.
n→∞ n→∞ n→∞

Hence, by Theorem 21, we have lim rn = 0. So the sequence converges

n→∞
and the limit is equal to 0.

Case 6: For r = −1
In this case we get the alternating sequence

{rn }∞ n ∞
0 = {(−1) }0 = 1, −1, 1, −1, · · ·

Thus, the limit lim (−1)n does not exist. So the sequence diverges for
n→∞
r = −1. Recall that Theorem 21 only applies if |an | = 0 as n → ∞.
We cannot use it if limn→∞ |an | 6= 0. For instance limn→∞ |(−1)n | =
limn→∞ | − 1|n = limn→∞ 1n = 1 does not imply that the sequence
(−1)n converges to 1.

Case 7: For r < −1

In this case we obtain an alternating sequence for every value of r as
in Case 6. For instance, if r = −3 we get the alternating sequence

{(−3)n }∞
0 = {1, −3, 9, −27, 81, · · · }

whose values keeps getting larger and larger as n increases. Thus, the
sequence diverges.
Example 42. Determine if the sequence converges or diverges, and find the
limit if it converges.
 2 ∞
2n + 3
(a)
3n − 7n2 n=2

86 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

∞
e2n

(b)
n2 n=1
∞
2n2 + 3

Solution. (a)
3n − 7n2 n=2

2n2 + 3 2 + n32 2
lim 2
= lim 3 =−
n→∞ 3n − 7n n→∞ −7 7
n

Therefore the sequence converges, and the limit is −2/7.

 2n ∞
e
(b)
n2 n=1
e2n
We need to find lim 2 , by applying L’Hopital’s rule. But, the rule
n→∞ n
can not be applied to sequences so we define the related function
e2x e2n
f (x) = , such that f (n) = .
x2 n2
Thus, using Theorem 19, we have
e2n e2x
lim = lim
n→∞ n2 x→∞ x2

Applying L’Hopital’s rule twice, we have

e2n 2e2x 2e2x
lim = lim = lim =∞
n→∞ n2 x→∞ 2x x→∞ 1

Therefore, the sequences diverges.

7.2 Bounded Monotonic Sequences

Definition 20. A sequence {an } is called increasing if an < an+1 for all
n ≥ 1, that is, a1 < a2 < a3 < · · · . It is called decreasing if an > an+1 for
all n ≥ 1. It is called monotonic if it is either increasing or decreasing.
 
n
Example 43. Show that the sequence is increasing.
n+1
n
Solution. Let an = . We need to show that an < an+1 for all n ≥ 1.
n+1
That is
n n+1
<
n+1 (n + 1) + 1

87 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

n n+1
=⇒ <
n+1 n+2
=⇒ n(n + 2) < (n + 1)2
=⇒ n2 + 2n < n2 + 2n + 1
=⇒ 0 < 1
which is true for n ≥ 1. Thus an < an+1 , and so an is increasing.
Alternatively, let an = f (n) = n/(n + 1), and consider the function
x
f (x) =
x+1

(x + 1) − x 1
=⇒ f 0 (x) = = >0
(x + 1)2 (x + 1)2
Thus, f 0 (x) > 0 if x > 0, and so f is increasing on (0, ∞). Therefore, the
sequence is increasing.

Definition 21. A sequence {an } is bounded above if there is a number M

such that
an ≤ M for all n ≥ 1.
It is bounded below if there is a number m such that

m ≤ an for all n ≥ 1.

If it is bounded above and below, then {an } is a bounded sequence.

Example 44. Determine if the following sequence is monotonic and/or

bounded
∞
(a) −n2 n=0


∞
(−1)n+1

(b) n=1

∞
Solution. (a) −n2 n=0


The sequence is decreasing, and therefore monotonic, since

−n2 > −(n + 1)2 for every n ≥ 0.

Note that the terms of the sequence will either be zero or negative.
Thus, the sequence is bounded above by 0.

88 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Remark. Any positive number or zero may be chosen as the upper

bound, say M , but it is standard to choose the smallest possible bound,
also called the supremum. That is why we chose M = 0, since
−n2 ≤ 0 for every n. Note that this sequence is not bounded be-
low since we can always choose another number below any potential
lower bound.
On the other hand, the sequence {n2 }∞ 0 is increasing from 0, and there-
fore bounded below by 0. Here, any negative number qualifies as a lower
bound, say m, but it’s standard to choose the largest possible bound,
also called the infimum. So the infimum or lower bound is m = 0.

∞
(−1)n+1 n=1

(b)
The sequence is an alternating sequence, with the terms alternating
between 1 and −1. Thus, the sequence is neither increasing nor de-
creasing, and so the sequence is not monotonic. However, the sequence
is bounded above by 1 and bounded below by −1.

The next theorem shows that if a sequence is both bounded and mono-
tonic, then it must be convergent. Note that not every bound sequence is
convergent. For example, the sequence {(−1)n }, is bounded as −1 ≤ an ≤ 1
but it is divergent. Also, not every monotonic sequence is bounded, for
example an = n is monotonic but an = n → ∞.

Theorem 23. (Monotonic Sequence Theorem) Every bounded, monotonic

sequence is convergent.
 n
2
Example 45. Show that is convergent and find its limit.
n!
2n
Solution. Let an = . The first few terms of the sequence are
n!
4 8 16
a1 = 2, a2 = = 2, a3 = = 1.3333, a4 = = 0.6667
2 6 24
32 64 1024
a5 = = 0.2667, a6 = = 0.0889, · · · , a10 = = 0.0003
120 720 3628800
Thus, the sequence appears to be decreasing from n = 2 onwards. To prove
an+1
this, we need to show that an ≥ an+1 . That is ≤ 1. Now
an

an+1 2n+1 n! 2n · 2 · n!
= · n =
an (n + 1)! 2 (n + 1) · n! · 2n

89 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

an+1 2
=⇒ =
an n+1
So we have
an+1
≤ 1 if n ≥ 1.
an
Therefore an ≥ an+1 if n ≥ 1. Since all the terms of the sequence are positive
and the sequence is decreasing, {an } is bounded below by 0. Hence, by
Theorem 23, the sequence is convergent.
We next find the limit of the sequence, using
an+1 2
=
an n+1
2
=⇒ an+1 = an
n+1
Let lim an = L then, obviously, lim an+1 = L. Thus, we have
n→∞ n→∞

2 2
L = lim aa+1 = lim an = lim · lim an = 0 · L = 0
n→∞ n→∞ n + 1 n→∞ n + 1 n→∞

2n
Hence lim = 0.
n→∞ n!

Exercise. 1. Find
3 + 5n2 ln n (−1)n
(a) lim (b) lim (c) lim
n→∞ n + n2 n→∞ n n→∞ n
2. Determine whether the given sequence converges or diverges. If it con-
verges then find the limit.
 2 ∞
(a)
3n − 5n + 3 (e) {sin(nπ)}∞n=0
1 + 8n − 3n2 n=3   ∞

(−1)n−2 n2
∞ n+1
(b) (f)
4 + n3 ln (5n) n=2
n=0
 4n ∞
e  
3
∞
(c) (g) cos
4 − e2n n=1 n+1
 ∞ n=1
ln (n + 2)
(d) (h) {ln (4n + 1) − ln (2 + 7n)}∞
ln (1 + 4n) n=1 n=0

(3) Consider the sequence {an } defined by the recurrence relation

1
a1 = 2 an+1 = (an + 6) for n = 1, 2, 3, · · ·
2

90 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

(i) Use mathematical induction to show that the sequence is increas-

ing.
(ii) Use mathematical induction to show that the sequence is bounded
above by 6, that is, an < 6 for all n.
(iii) Use the Monotonic Sequence Theorem to show that the sequence
converges, and that the limit is 6.

(4) Show that the sequence

∞
n3


n4 + 10000 n=0

is not monotonic.

91 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

92 Dr. Joseph K. Ansong

Chapter 8

Infinite Series

8.1 Introduction

8.1.1 Definitions & Notation

Definition 22. An infinite series (or a series) is the sum of the terms of
an infinite sequence of the form
∞
X X
a1 + a2 + . . . + an + · · · = an or an .
n=1

Each number ak is a term of the series, and an is the nth term.

Remark. Let’s start by removing some confusion between the concept of a

series and that of a sequence. A sequence is a list or collection of numbers
written in a particular order, whereas a series is an expression of an infinite
sum of numbers.

Since only finite sums may be added algebraically, what do we mean by an

“infinite sum”? As the following example illustrates, and as we shall explain
later on, the key to the definition is to consider the sequence of partial
sums {Sn }, where Sk is the sum of the first k numbers of the infinite
series. Consider the following infinite series

93
Math 223 : Calculus II J.K.A

Example 46. 0.6 + 0.06 + 0.006 + 0.0006 + 0.00006 + · · ·

S1 = 0.6
S2 = 0.6 + 0.06 = 0.66
S3 = 0.6 + 0.06 + 0.006 = 0.666
S4 = 0.6 + 0.06 + 0.006 + 0.0006 = 0.6666
S5 = 0.6 + 0.06 + 0.006 + 0.0006 + 0.00006 = 0.66666
and so on. Thus, the sequence of partial sums {Sn } may be written as
0.6, 0.66, 0.666, 0.6666, 0.66666, c . . . .
It follows that (we’ll appreciate this conclusion later under “Geometric se-
ries”)
2
Sn → as n → ∞.
3
We see that the more numbers of the infinite series that we add, the closer
the sum gets to 2/3. Thus, we write
2
= 0.6 + 0.06 + 0.006 + 0.0006 + · · ·
3
2
and call the sum of the infinite series.
3

8.2 Convergence and Divergence of a Series

Consider the partial sums of the series: a1 + a2 + . . . + an + . . .
s 1 = a1
s2 = a1 + a2
s3 = a1 + a2 + a3
..
.
n
X
sn = a1 + a2 + a3 + . . . + an = ai
i=1

Note that the partial sums {s1 , s2 , s3 , . . . , sn } now form a sequence.

P
Definition 23. (Convergence & Divergence of Series) A series an is
convergent (or converges) if its sequence of partial sums {sn } converges;
that is, if
lim sn = s for some real number s.
n→∞

94 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

X
The limit s is the sum of the series an , and we write

s = a1 + a2 + · · · + an + · · · .
X
The series an is divergent (or diverges) if {sn } diverges. A divergent
series has no sum.

Remark. The sum of an infinite series means that by adding sufficiently

many terms, we get as close as we like to the number s.

Example 47. Consider the series

1 1 1 1
+ + + ··· + + ··· ,
1·2 2·3 3·4 n(n + 1)

(a) find s1 , s2 , s3 , s4 , s5 , and s6

(b) find sn

(c) show that the series converges and find its sum.

Solution. (a) The first six partial sums are given by

1 1
s1 = =
1·2 2
1 1 2
s2 = + =
1·2 2·3 3
1 1 1 3
s3 = + + =
1·2 2·3 3·4 4
1 1 1 1 4
s4 = + + + =
1·2 2·3 3·4 4·5 5
4 1 5
s 5 = s 4 + a5 = + =
5 5·6 6
5 1 6
s 6 = s 5 + a6 = + =
6 6·7 7

(b) From the first six partial sums, it is straightforward to deduce that the
n
nth partial sum is given by the expression sn = . However, we
n+1
shall derive this in a different way. Using partial fractions, we can show
that the nth term is given by
1 1 1
an = = −
n(n + 1) n n+1

95 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Thus, the nth partial sum of the series may be written as

s n = a1 + a2 + a3 + · · · + an
       
1 1 1 1 1 1 1
= 1− + − + − + ··· + − .
2 2 3 3 4 n n+1
       
1 1 1 1 1 1 1
= 1 −  +  −  +  −  + ··· +  − .
2 2 3 3 4 n
 n+1

We see that all the numbers except the first and last terms cancel, and
so we get
1 n
sn = 1 − =
n+1 n+1
Remark. This series is an example of a telescoping series, because
of the cancellation nature of the terms of the series. That is, writing
the partial sums as shown above causes the terms to telescope to 1 −
1
. Note that the terms of different telescoping series cancel out
n(n + 1)
in different ways, not just the order shown in our example above.

(c) Using the expression for sn obtained above, we get

n
lim sn = lim = 1.
n→∞ n→∞ n + 1

Thus, the series converges and has a sum of 1. Hence, we can write
1 1 1 1
1= + + + ··· + + ··· .
1·2 2·3 3·4 n(n + 1)

Example 48. Consider the series

∞
X
(−1)n−1 = 1 + (−1) + 1 + (−1) + · · · + (−1)n−1 + · · · ,
n=1

(a) find s1 , s2 , s3 , s4 , s5 , and s6

(b) find sn

(c) show that the series diverges.

Solution. (a) The first six partial sums are given by

s1 = 1, s2 = 0, s3 = 1, s4 = 0, s5 = 1, and s6 = 0.

96 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

(b) The nth partial sum may be written as


1 if n is odd
sn = (8.1)
0 if n is even

(c) Since the sequence of partial sums {sn } oscillate between 1 and 0, it
follows that limn→∞ sn does not exist. Hence the series diverges.

8.3 Special Series

The Geometric Series

Theorem 24. The geometric series

∞
X
2 3 n−1
a + ar + ar + ar + . . . + ar + ··· = arn−1 a 6= 0
n=1

a
(a) converges and has the sum s = if |r| < 1
1−r
(b) diverges if |r| ≥ 1.

Proof. If r = 1, then we have

a + a + a + . . . + a = na
and limn→∞ na = ±∞, so the series will diverge.
If r 6= 1, we get
sn = a + ar + ar2 + . . . + arn−1 (8.2)
rsn = ar + ar2 + . . . + arn−1 + arn (8.3)

sn − rsn = a − arn [(8.2) − (8.3)]

Thus, we get
a(1 − rn ) a arn
sn = = +
1−r 1−r 1−r
a arn
=⇒ lim sn = lim + lim
n→∞ n→∞ 1 − r n→∞ 1 − r
a arn
= − lim
1 − r n→∞ 1 − r

97 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

If −1 < r < 1, limn→∞ rn = 0 and so

a
lim sn = .
n→∞ 1−r

If r ≤ −1 or r > 1, {rn } diverges and so limn→∞ sn does not exit.

In summary, the sum of the geometric series is:

∞
X a
arn−1 = if |r| < 1.
n=1
1−r
If |r| ≥ 1, the series diverges.

Example 49. Prove that the following series converges, and find its sum:

2 2 2
2+ + 2 + · · · + n−1 + · · ·
3 3 3
1
Solution. The series is geometric with a = 2 and r = < 1. Thus, the
3
series converges to the sum

a 2 2
s= = 1 = = 3.
1−r 1− 3
2/3

Example 50. Express the number 0.2̄ = 0.2222 . . . as a ratio of integers.

Solution.
2 2
0.2̄ = + 2 + ···
10 10
is a geometric series with a = 2/10, r = 1/10 and the sum is

a 2/10 2
s= = = .
1−r 1 − 1/10 9

4 8
Exercise. (1) Find the sum of the geometric series 3 + 2 + + + ...
3 9
(2) Prove that the following series converges, and find its sum

0.6 + 0.06 + 0.006 + 0.0006 + 0.00006 + · · ·

Note: this was the series we considered at the beginning of this chapter.

98 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

The harmonic series

Prove that the the harmonic series
∞
X 1 1 1 1
= 1 + + + + ···+
n=1
n 2 3 4
is divergent.
Solution.
s1 = 1
1
s2 = 1 +
2    
1 1 1 1 1 1 2
s4 = 1 + + + >1+ + + =1+
2 3 4 2 4 4 2
   
1 1 1 1 1 1 1
s8 = 1 + + + + + + +
2 3 4 5 6 7 8
   
1 1 1 1 1 1 1
>1+ + + + + + +
2 4 4 8 8 8 8
1 1 1 3
=1+ + + =1+
2 2 2 2
..
.
5
s32 > 1 +
2
6
s64 > 1 +
2
n
In general s2n > 1 + for every positive integer n. This implies that sn
2
can be made as large as desired by taking n sufficiently large. That is,
lim s2n = ∞, which implies that {sn } is divergent. The harmonic series is
n→∞
therefore divergent.
Exercise. (1) Determine whether the series is convergent or divergent:
∞  n−1
X 2
(a) 5 (b)
n=1
3

(2) Express the number as a ratio of integers

3.417 = 3.417417417 · · ·

99 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

8.4 Some Useful Theorems

P∞
Theorem 25. If the series n=1 an is convergent, then limn→∞ an = 0

Proof. Let sn = a1 + a2 + . . . + an−1 + an . Then

sn = sn−1 + an
⇒ an = sn − sn−1
⇒ lim an = lim sn − lim sn−1
n→∞ n→∞ n→∞

Let limn→∞ sn = s. Then limn→∞ sn−1 = s since limn→∞ n − 1 = ∞. Thus

lim an = s − s = 0
n→∞

Remark. The converse is not true in general; that is, if lim an = 0, it does
P n→∞
not necessarity follow that the series an is convergent. An example is the
harmonic series for which limn→∞ an = 0 even though the series is divergent.
Thus, to establish convergence of a series, it is not enough to prove
that limn→∞ an = 0, since that may be true for divergent as well as
for convergent series.

8.4.1 The Divergence Test

X
(i) If lim an 6= 0, then the series an is divergent.
n→∞

(ii) If lim an = 0, then further investigation is necessary to determine

n→∞ P
whether the series an is convergent or divergent.

The following examples demonstrate how to apply the divergence test.

Example 51. Determine if the following series is divergent

100 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

∞ ∞
X 2n X 1
(1) (3) √
n=1
5n + 1 n=1
n
∞ ∞
X en X 1
(2) (4)
n=1
n2 n=1
n3

2n 2 2
Solution. (1) lim = lim 1 = 6= 0.
n→∞ 5n + 1 n→∞ 5 + 5
n
Thus, by the test for divergence, the series is divergent.
en
(2) lim 2 = ∞.
n→∞ n
So the series diverges by the divergence test.

(3) limn→∞ √1n = 0. We cannot draw any conclusion, further investigation

is required. Later in the text, it will be simple to show that the series
is divergent.

(4) limn→∞ n13 = 0. Further investigation is required, by the test for di-
vergence. In a short while, it will be relatively easy to show that this
series is convergent.

P P
Theorem 26. If an and bn are convergent then so are
X X
(a) can = c an
X X X
(b) (an ± bn ) = an ± bn

∞  
X 5 2
Example 52. Find the sum of the series + n
n=1
n(n + 1) 3

∞ ∞ ∞
X 2 X 1 X 1
Solution. We can write the series n
= 2 n
. However, n
=
n=1
3 n=1
3 n=1
3
∞  n
X 1 1 1
is a geometric series with a = and r = so
n=1
3 3 3

∞ 1
X 1 3 1 3 1
n
= 1 = · =
n=1
3 1 − 3
3 2 2

101 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Thus,
∞
X 2 1
n
= 2 · = 1.
n=1
3 2
∞
X 1
In example 47, we showed that = 1. Thus
n=1
n(n + 1)

∞ ∞
X 5 X 1
=5 = 5 · 1 = 5.
n=1
n(n + 1) n=1
n(n + 1)

Therefore the sum of the series is given by

∞   ∞ ∞
X 5 2 X 5 X 2
+ n = + = 1 + 5 = 6.
n=1
n(n + 1) 3 n=1
n(n + 1) n=1 3n

X X
Theorem 27. If an is a convergent series and bn is divergent,
X
then (an + bn ) is divergent.

Example 53. Determine the convergence or divergence of the series

∞  
X 1 7
+
n=1
3n n

∞
X 1
Solution. The series n
is convergent, since it is a geometric series with
n=1
3
∞
X 1
r = 1/3. However, the series is the divergent harmonic series. There-
n=1
n
fore the given series is divergent by Theorem 27.

n2
Exercise. 1. Show that the series is divergent.
5n2 + 4
2. Prove that the following series converges, and find its sum:
∞  
X 7 2
+ n−1
n=1
n(n + 1) 3

102 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

∞ 
cosn x
X 
n n
3. Find the sum of the series: 4 x +
n=0
2n

The next theorem states that if corresponding terms of two series are
identical after a certain term, then both series converge or both series diverge.

X X
Theorem 28. If an and bn are series such that aj = bj for every
j > k, where k is a positive integer, then both series converge or both
series diverge.

X
Proof. We may write the series an , by hypothesis, as
X
an = a1 + a2 + a3 + · · · + ak + ak+1 + · · · + an + · · ·
X X
Let Sn and Tn denote the nth partial sums of an and bn respectively.
X
The partial sums of an are given by

S1 = a1
S2 = a1 + a2
..
.
Sk = a1 + a2 + · · · + ak
Sn = Sk + ak+1 + · · · + an
=⇒ Sn − Sk = ak+1 + · · · + an

Similarly, we have

Tn = Tk + ak+1 + · · · + an
=⇒ Tn − Tk = ak+1 + · · · + an ,

since aj = bj for j > k. Thus,

Sn − Sk = Tn − Tk

=⇒ Sn = Tn + (Sk − Tk )
∴ lim Sn = lim Tn + (Sk − Tk )
n→∞ n→∞

and so either both of the limits exist or both do not exists. If both series
converge, then their sum differ by Sk − Tk .

103 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Theorem 28 shows that if we change a finite number of terms of a series, it

will have no effect on whether the series would converge or diverge. However,
it does change the value orX sum of a convergent series. For instance, if we
replace the first k terms of an by 0, it won’t affect the convergence of the
series. The above theorem is re-stated differently in the next theorem.

Theorem 29. For any positive integer k, the series

∞
X ∞
X
an = a1 + a2 + · · · and an = ak+1 + ak+2 + · · ·
n=1 n=k+1

either both converge or both diverge.

Example 54. Show that the following series converges

1 1 1
+ + ··· + + ··· (8.4)
3·4 4·5 (n + 2)(n + 3)

Solution. Consider the telescoping series of example 47, repeated below:

1 1 1 1 1
+ + + + ··· + + ···
1·2 2·3 3·4 4·5 n(n + 1)

By deleting the first two terms, we get the series in 8.4. Since the telescoping
series is convergent, by Theorem 29, the given series is also convergent.

8.5 Convergence of Positive-Term Series

In
Xthis section, we consider only positive-term series where the series
an is such that an > 0 for every n. The convergence or divergence of
an arbitrary series can often be determined from that of a related positive-
term series. Thus, positive-term series a the foundation for several test for
convergence of any given series.

8.5.1 The Integral Test

Consider a function f which is continuous, positive and decreasing on the
Z ∞
interval [1, ∞). The area under f is given by A = f (x)dx. If this
1

104 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

improper integral is convergent, then A is finite. If the integral is divergent

then A is infinite.
Now, if we divide the area under f into rectangular strips, each Zof which is
∞
taken to represent a term of the sequence an = f (n), then since f (x)dx
Z ∞1
is the sum of these strips, the series ∞
P
n=1 an is convergent if f (x)dx is
P∞ R∞ 1
convergent and n=1 an is divergent if 1 f (x)dx is divergent. Thus, we get
the integral test stated below:

Theorem 30. (Integral Test) Suppose f is a continuous, positive P de-

creasing function on [1, ∞) and let an = f (n). Then the series ∞ n=1 an
is R∞
convergent if and only if the improper integral 1 f (x)dx is convergent.
In other words:
Z ∞ ∞
X
(i) If f (x)dx is convergent, then an is convergent.
1 n=1
Z ∞ ∞
X
(ii) If f (x)dx is divergent, then an is divergent.
1 n=1

Remark. When using the integral test,

(a) it is important to verify the conditions of the test

(b) it is not necessary to start the series or the integral at n = 1.

(c) it is not necessary that f be always decreasing. What is important

is that f is ultimately or eventually decreasing (i.e. decreasing for x
larger than some number N ).

Example 55. Use the integral test to prove that the harmonic series
1 1 1
1+ + + ··· + + ···
2 3 n
is divergent.

Solution. Since an = 1/n, we let f (n) = 1/n and replace n by x to get

the function f (x) = 1/x. Because f is positive-valued, continuous, and

105 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

decreasing for x ≥ 1, we can apply the integral test to compute

Z ∞ Z t
1 1
dx = lim dx = lim [ln x]t1
1 x t→∞ 1 x t→∞

= lim [ln t − ln 1] = ∞
t→∞

Therefore the series diverges by the integral test.

2
X
Example 56. Determine whether the infinite series ne−n converges.
2 2
Solution. Now an = ne−n , so we consider the function f (x) = xe−x . If
x ≥ 1, f is positive-valued and continuous. We also need to show that the
function is eventually decreasing. Finding the first derivative, we have
2 2 2
f 0 (x) = e−x − 2x2 e−x = e−x (1 − 2x2 ) < 0

Therefore f is decreasing on [1, ∞]. Thus, we can apply the integral test to
compute
Z ∞ Z t   t
−x2 −x2 1 −x2
xe dx = lim xe dx = lim − e
1 t→∞ 1 t→∞ 2 1
   
1 1 1 1
= − lim t2 − = .
2 t→∞ e e 2e
Therefore the series converges by the integral test.
∞
X ln n
Example 57. Use the Integral Test to determine whether converges
n=1
n
or diverges.
Solution. We have an = (ln n)/n, so we consider f (x) = (ln x)/x. We note
that the function f is continuous and positive on the interval [1, ∞]. To show
that it’s decreasing, we compute
1
0 x· x
− ln x 1 − ln x
f (x) = 2
=
x x2
=⇒ f 0 (x) < 0 for ln x > 1 =⇒ x > e
Thus, the function is decreasing on the interval (e, ∞] or we can also use
[3, ∞] since e ≈ 2.7. Using the integral test we have
Z ∞ Z t
ln x ln x
I= dx = lim dx
3 x t→∞ 3 x

106 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Let u = ln x =⇒ du = (1/x)dx

ln x 1
Z Z
dx = udu = u2
x 2
 t
1 2 1
=⇒ I = lim (ln x) = lim (ln t)2 − (ln 3)2 = ∞
t→∞ 2 2 t→∞
3
Thus, the series diverges by the integral test.

Exercise. Test the series for convergence or divergence:

∞
X 1
n=1
n2 +1

The p-series
Example 58. For what values of p is the following series, called a p-series,
convergent?
∞
X 1
n=1
np

1
Solution. If p < 0, lim p = ∞.
n→∞ n
1
If p = 0, lim p = 1.
n→∞ n
1
In either case, lim p 6= 0 so the series diverges by the test for divergence.
n→∞ n
1
If p > 0, the function f (x) = p is continuous, positive and decreasing on
Z ∞ x
1
[1, ∞). But recall that dx converges if p > 1 and diverges if p ≤ 1.
1 xp
We just proved the following theorem

∞
X 1
Theorem 31. (p-series) The p-series p
is convergent if p > 1 and
n=1
n
divergent if p ≤ 1.

Example 59. Determine whether the series is convergent or divergent:

107 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

∞ ∞
X 1 X 1
(i) 3/2
(ii) 1/2
n=1
n n=1
n
∞
X 1
Solution. (i)
n=1
n3/2
This is a p-series with p = 3/2 > 1, therefore the series converges.
∞
X 1
(ii) 1/2
n=1
n
The series is a p-series with p = 1/2 < 1, hence the series is divergent.

Comparison Tests
P∞
The idea behind thisP∞ test is to compare a given series, say n=1 an , to a

P∞ n=1 bn (which is known to either

second series, say P∞ converge or diverge).
If thePterms of n=1 an are smaller than those of n=1 bn (and it converges),
then ∞ n=1 an also
P∞converges.
If thePterms of n=1 an are larger than those of ∞
P
n=1 bn (and it diverges)
then ∞ a
n=1 n also diverges. Thus, we formally have:

8.5.2 The Comparison Test

P∞ P∞
Theorem 32. Suppose that n=1 an and n=1 bn are series with positive
terms.
∞
X ∞
X
(i) If bn is convergent and an ≤ bn for all n, then an is also
n=1 n=1
convergent.
∞
X ∞
X
(ii) If bn is divergent and an ≥ bn for all n, then an is also
n=1 n=1
divergent.

Example 60. Determine whether the series converges or diverges

108 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

∞ ∞
X 1 X 5
(a) n
(b) √
n=1
7 +3 n=1
n−1
∞
X 1
Solution. (a) Comparing the given series to the series , we have
n=1
7n
 n
1 1 1
< = for all n ≥ 1.
7n + 3 7n 7
∞
X 1
The series is a geometric series with r = 1/7 < 1 therefore
n=1
7n
it is convergent. Hence, by the comparison test, the given series is
convergent.
X 1
(b) Comparing the given series to the p-series √ , we have
n
1 1
√ >√ for n ≥ 2
n−1 n
5 1
=⇒ √ >√ for n ≥ 2
n−1 n
X √
Since the series 1/ n is a divergent p-series, the given series is also
divergent.
Example 61. Use the comparison test to determine whether the series con-
verges or diverges
∞ ∞
X 1 X ln n
(a) 2
(b)
n=1
n +n+1 n=1
n

Remark. If the comparison test fails, a related test which may also be used
is the limit comparison test; stated below.

8.5.3 The Limit Comparison Test

P P
Theorem 33. (Limit Comparison Test) Suppose that an and bn
an
are series with positive terms. If lim = c where c is a finite number
n→∞ bn
and c > 0, then either both series converge or both diverge.

109 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Proof. This proof relies heavily on the Comparison Test, so go back and
read it if you haven’t yet. Let m and M be positive numbers such that
m < c < M . If limn→∞ (an /bn ) = c > 0, then this implies that an /bn is close
to c for large n. Hence there exists an integer N such that
an
m< <M whenever n > N
bn
=⇒ mbn < an < M bn whenever n > N
P P
If the series bn is convergent, then the seriesP M bn is also convergent
by Theorem 26. Since an < M bn , P the series an must converge P by the
Comparision Test. Also, if the
P series b n is divergent, so is the series mbn .
Since mbn < an , the series an is divergent by the Comparison Test.
Example 62. Determine whether the series converges or diverges
∞ ∞
X 1 X 5n2 + 3n
(a) √ (b)
n=1
3
n2 + 1 n=1
2n (n2 + 1)

Solution. (a) Let

1
an = √
3
,
n2 + 1
√3
and consider bn = 1/ n2 , which is a p-series with p = 2/3 < 1 and
therefore divergent. The term bn is obtained by deleting the 1 in the
denominator of an . Using the Limit Comparison Test, we compute
√
3
√
3
r
an 1 n2 n2 3 n2
lim = lim √ · = lim √ = lim =1>0
n→∞ bn n→∞ 3
n2 + 1 1 n→∞ 3
n2 + 1 n→∞ n2 + 1
P
Since bn is divergent, the given series is also divergent by the Limit
Comparison Test.
Remark. Note that we cannot use the basic Comparison Test in this
case because
1 1
√3
< √3
,
2
n +1 n2
P
and since bn diverges, we cannot draw any conclusion based on that
test.

(b) Let
∞
X 5n2 + 3n
an = n (n2 + 1)
.
n=1
2

110 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

By removing the terms of least magnitude in the numerator and de-

nominator of an , we get
5n2 5
n 2
= n
2 n 2
n
P
and so we choose bn = 1/2 . The series bn is a convergent geometric
series with r = 1/2 < 1. Applying the Limit Comparison Test, we have
an 5n2 + 3n 2n 5n2 + 3n
lim = lim n 2 · = lim =5>0
n→∞ bn n→∞ 2 (n + 1) 1 n→∞ n2 + 1
P P
Since bn is convergent, the series an is also convergent by the
Limit Comparison Test.
Exercise. Test the series for convergence or divergence
∞ ∞ √
X 1 X 8n + n
(a) n−1
(c)
n=1
2 n=1
5 + n2 + n7/2
∞
X 2n2 + 3n
(b) √
n=1
5 + n5

8.6 Alternating Series

An alternating series is a series whose terms are alternately positive and
negative. E.g.:
∞
X
(a) 1 − 2 + 3 − 4 + 5 + . . . = (−1)n−1 n
n=1
∞
1 1 1 X (−1)n−1
(b) 1 − + − + ... =
2 3 4 n=1
n

If the terms of an alternating series decrease toward 0 in absolute value, then

the series converges. Explicitly, we have

Theorem 34. (The Alternating Series Test) If the alternating series

∞
X
(−1)n−1 bn = b1 − b2 + b3 − b4 + . . . (bn > 0)
n=1

satisfies

111 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

(i) bn+1 ≤ bn for all n

(ii) lim bn = 0 then the series is convergent.

n→∞

Example 63. Determine whether the alternating series converges or diverges

∞ ∞
X 2n X 2n
(a) (−1)n−1 (b) (−1)n−1
n=1
4n2 − 3 n=1
4n − 3

Solution. (a) Let

2n
an =
4n2 − 3
By the Alternating Series Test, we first need to show that ak ≥ ak+1 for
every positive integer k. That is, we have to show that an is ultimately
decreasing. Secondly, we also need to show that limn→∞ an = 0. One
method of showing that an is decreasing is to define
2x
f (x) =
4x2−3
and prove that f is a decreasing function. Now,

(4x2 − 3)(2) − (2x)(8x) −8x2 − 6 −(8x2 + 6)

f 0 (x) = = = <0
(4x2 − 3)2 (4x2 − 3)2 (4x2 − 3)2

Thus, f (x) is decreasing and therefore f (k) ≥ f (k + 1); and hence

ak ≥ ak+1 for every k > 0.
Alternatively, we can also prove ak ≥ ak+1 by showing that ak − ak+1 ≥
0. That
2k 2(k + 1)
ak = ak+1 = 2 −
4k − 3 4(k + 1)2 − 3
8k 2 + 8k + 6
= ≥0
(4k 2 − 3)(4k 2 + 8k + 1)
for every positive integer k.A third method is to prove that ak+1 /ak ≤ 1.
Now
2
2n n
lim an = lim = lim =0
n→∞ n→∞ 4n2 − 3 n→∞ 4 − 32
n

Thus, the series converges by the Alternating Series Test.

112 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

(b) Let
2n
an =
4n − 3
In this case, there’s no need to show that ak ≥ ak+1 since
2n 2 1
lim an = lim = lim 3 = 6= 0
n→∞ n→∞ 4n − 3 n→∞ 4 − 2
n

implies that the second condition is not satisfied. This computation

also shows that
2n
lim (−1)n−1
n→∞ 4n − 3
does not exist, and so the series diverges, by the divergence test (see
section 8.4.1).

Exercise. Test the series for convergence or divergence

∞ ∞ ∞
X (−1)n−1 X (−1)n 3n X n2
(a) (b) (c) (−1)n+1
n=1
n n=1
4n − 1 n=1
n3 + 1

8.7 Absolute Convergence, the Ratio & the

Root Tests

P
Definition 24. A series P an is called absolutely convergent if the
series of absolute values |an | is convergent.

∞
X (−1)n−1
Example 64. The series is absolutely convergent. Why?
n=1
n3

X
Definition 25. A series an is called conditionally convergent if it
is convergent but not absolutely convergent.

∞
X (−1)n−1
Example 65. The series is convergent but not absolutely con-
n=1
n
vergent. Why?

113 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

X
Theorem 35. If a series an is absolutely convergent, then it is con-
vergent.

Proof. Note that the following inequality, which we have seen before, is true

−|an | ≤ an ≤ |an |.

=⇒ 0 ≤ an + |an | ≤ 2|an |,
obtained by adding |an | to both sides of the first inequality. Now let bn =
an + |an |, then
0 ≤ bn ≤ 2|an |
P P
If
P an is absolutely convergent, then |an | is convergent and therefore
P n2|a | is also convergent by Theorem 26. Since bn ≤ 2|an |, the series
P bn is Pconvergent
P by the Comparison
P Test. From bn = an + |an | we have
an = bn − |an |, and so an is convergent by Theorem 26.

Example 66. Determine whether the following series is convergent or diver-

gent
∞
X cos n cos 1 cos 2 cos 3
2
= 2 + 2 + 2 + ···
n=1
n 1 2 3

Solution. Note that the series is not an alternating series; the signs change
irregularly. Consider
∞ ∞
X cos n X | cos n|
2
=
n=1
n n=1
n2

Since | cos n| ≤ 1 for all n, we have

| cos n| 1
≤
n2 n2
But the series 1/n2 is a convergent p-series with p = 2, therefore | cos n|/n2
P P
is convergent by the Comparison Test. Thus, the given series is absolutely
convergent and therefore convergent by Theorem 35.

8.7.1 The Ratio Test

114 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Theorem 36. (Ratio Test)

∞
an+1 X
(i) If lim = L < 1, then the series an is absolutely conver-
n→∞ an n=1
gent (and therefore convergent).
∞
an+1 an+1 X
(ii) If lim = L > 1 or lim = ∞, then the series an
n→∞ an n→∞ an n=1
is divergent.
an+1
(iii) If lim = 1, the Ratio Test is inconclusive; that is, no con-
n→∞ an P
clusion can be drawn about the convergence or divergence of an .

Remark. Note that if the series has positive terms, the ratio test can still
be applied without using the absolute value symbols in the theorem.
Example 67. Determine whether the following series is absolutely conver-
gent, conditionally convergent, or divergent.
∞
X n2 + 4
(−1)n n
n=1
2

Solution. Applying the Ratio Test, we have

an+1 (n + 1)2 + 4 2n
lim = lim ·
n→∞ an n→∞ 2n+1 n2 + 4
1 n2 + 2n + 5
 
1 1
= lim 2
= (1) = < 1.
n→∞ 2 n +4 2 2
Therefore, by the Ratio Test, the series is absolutely convergent.
Exercise. 1. Test the series for absolute convergence
∞
X n3
(−1)n n
n=1
3

2. Test the convergence of the series

∞
X nn
n=1
n!

115 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

8.7.2 The Root Test

∞
X
p
n
Theorem 37. (i) If lim |an | = L < 1, then the series an is
n→∞
n=1
absolutely convergent (and therefore convergent).
∞
X
p
n
p
n
(ii) If lim |an | = L > 1 or lim |an | = ∞, then the series an is
n→∞ n→∞
n=1
divergent.
p
(iii) If lim n |an | = 1, the Root Test is inconclusive.
n→∞

Example 68. Test the convergence of the series

∞  n
X 2n + 3
n=1
3n + 2

Solution. Let  n
2n + 3
an =
3n + 2
3
p
n 2n + 3 2+ n 2
=⇒ lim |an | = lim = lim 2 = <1
n→∞ n→∞ 3n + 2 n→∞ 3 + 3
n
Hence, the series converges by the Root Test.

116 Dr. Joseph K. Ansong

Chapter 9

Power Series

9.1 Power Series

Definition 26. A power series centered at a is a series of the form:
∞
X
cn (x − a)n = co + c1 (x − a) + c2 (x − a)2 + . . . (9.1)
n=0

where the cn ’s are constants called the coefficients of the series.

Remark. If a = 0 and cn = 1 for all n, then we get the geometric series:
∞
X 1
xn = 1 + x + x2 + . . . + xn + . . . = (9.2)
n=0
1−x

which converges for −1 < x < 1, diverges for |x| ≥ 1 and has the sum
1/(1 − x).

9.1.1 Radius and Interval of Convergence

P∞ n
Theorem 38. For a given power series n=0 cn (x − a) there are only three
possibilities:
(i) The series converges only when x = a
(ii) The series converges for all x
(iii) There is a positive number R such that the series converges if |x−a| < R
and diverges if |x − a| > R.
R is called the radius of convergence and the interval of convergence
is the interval that consists of all values of x for which the series converges.

117
Math 223 : Calculus II J.K.A

Remark. 1. In general, the Ratio Test (or the Root Test) should be used
to determine the radius of convergence R.
2. Remember to always test the end points of the interval of convergence
for convergence (using a different test other than the Ratio or Root
Tests).
3. For a given power series ∞ n
P
n=0 cn (x − a) , the radius of convergence is
given by
cn 1
R = lim or R = lim √
n→∞ cn+1 n→∞ n cn

Alternatively, we have
1 cn+1 1 √
= lim or = lim n cn
R n→∞ cn R n→∞
A third method is to solve
an+1
L = lim .
n→∞ an
(i) If the series converges, then L < 1 and |x − a| < R, where R is
the radius of convergence.
(ii) If the series diverges, then L > 1.
We prove the first part of the third point.
∞
X cn+1
Proof. Consider the power series cn (x−a)n . Suppose the limit lim =
n=0
n→∞ cn
L where 0 ≤ L ≤ ∞. Applying the Ratio Test to the series, we have
an+1 cn+1
L = lim = lim · |x − a| = L|x − a|
n→∞ an n→∞ cn
where an = cn (x−a)n . By the Ratio Test, the series converges if L|x−a| < 1,
that is, if |x − a| < 1/L and diverges if |x − a| > 1/L. Thus, by the definition
of the radius of convergence, R, we have R = 1/L. From the equation above,
we also have
cn+1
L = lim
n→∞ cn
1 cn
∴R= = lim .
L n→∞ cn+1

Example 69. For what values of x is the series convergent?

118 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

∞ ∞ ∞
X (x − 3)n X xn X
(a) (b) √ (c) n!xn
n=1
n n=1
n n=0

(x − 3)n
Solution. Let an = . Then
n
an+1 (x − 3)n+1 n
= ·
an n+1 (x − 3)n

n n 1
= · (x − 3) = |(x − 3)| = |x − 3|
n+1 n+1 1 + n1
an+1 1
=⇒ lim = lim |x − 3| = |x − 3|
n→∞ an n→∞ 1 + 1
n

The series is convergent for |x − 3| < 1, which may be written as 2 < x < 4.

Remark. The radius of convergence is 1, obtained from |x − 3| < 1 by using

1
the third method stated above. Using the first approach, with cn = , we
n
may also find the radius of convergence from

cn 1 n+1 n+1 1
R = lim = lim · = lim = lim 1 + = 1.
n→∞ cn+1 n→∞ n 1 n→∞ n n→∞ n

Example 70. Find the radius and interval of convergence of

(a) The Bessel function of order 0:

∞
X (−1)n x2n
Jo (x) =
n=0
22n (n!)2

∞ ∞
X (−1)n−1 xn X (−1)n (x + 2)n
(b) (c)
n=1
n3 n=1
n.2n

(−1)n
Solution. (b) Let cn = . Then
n · 2n
cn 1 (n + 1) · 2n+1 1
R = lim = lim · = 2 lim 1 + =2
n→∞ cn+1 n→∞ n · 2n 1 n→∞ n

So the radius of convergence is 2.

119 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

9.2 Representation of Functions as Power Se-

ries
Our aim here is to learn how to represent functions as power series. This
helps in integrating functions that don’t have elementary antiderivatives, for
solving differential equations, and for approximating functions by polynomi-
als. The strategy is to manipulate geometric series or by differentiating or
integrating the given series. Consider
∞
1 X
= 1 + x + x2 + x3 + . . . = xn , |x| < 1 (9.3)
1−x n=0

Equation (9.3) is a representation of the function f (x) = 1/(1 − x) as the

sum of a power series (in this case the geometric series).

Example 71. Find a power series representation for the function and deter-
mine the radius of convergence:

1 1 1
(a) f (x) = (b) f (x) = (c) f (x) =
1+x 1 − x3 x+2

∞
1 X
Solution. (a) From the relation = xn , we replace x with −x to
1 − x n=0
get
∞ ∞
1 1 X X
= = (−x)n = (−1)n xn
1 − (−x) 1 + x n=0 n=0

∞
1 X
∴ = (−1)n xn
1 + x n=0

It is easy to show that the radius of convergence is 1.

(c)

∞ ∞
1 X  x n X (−1)n n
 
1 1 1 1
= = = − = x
x+2 2(1 + x2 ) 2 1 − (− x2 ) 2 n=0 2 n=0
2n+1

It is straightforward to show that the radius of convergence is 1.

120 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Differentiation and Integration of Power Series

Given the power series ∞ n
P
n=0 cn (x−a) , we can differentiate or integrate each
term of a power series. That is, if
∞
X
f (x) = cn (x − a)n = co + c1 (x − a) + c2 (x − a)2 + . . . (9.4)
n=0

with radius of convergence R > 0 and interval of convergence (a − R, a + R),

then
"∞ # ∞
df d X X d
(i) = cn (x − a)n = [cn (x − a)n ]
dx dx n=0 n=0
dx
∞
X
= ncn (x − a)n−1 = c1 + 2c2 (x − a) + . . .
n=1
Z "X∞ ∞ Z
Z #
X
n
(ii) f (x)dx = cn (x − a) = [cn (x − a)n ]
n=0 n=0
∞ n+1
X (x − a)
= cn + C = C + co (x − a) + . . .
n=0
n+1

NOTE:
(a) The radii of convergence of the power series in (i) and (ii) are both R.
(b) The interval of convergence in (i) and (ii) may be different.
Example 72. Express the function as a power series. What’s the radius of
convergence?
1
(a) f (x) = (differentiate equation (9.3))
(1 − x)2
(b) f (x) = ln(5 − x)
(c) f (x) = tan−1 x

Solution. (a) Differentiating

∞
1 X
= 1 + x + x2 + x3 + . . . = xn
1−x n=0

we get
1 2
X
= 1 + 2x + 3x + · · · = nxn−1
(1 − x)2 n=1

121 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

This may also be written as (by replacing n by n + 1)

1 X
2
= (n + 1)xn
(1 − x) n=0

The radius of convergence is the same as that of the original series,

R = 1.

1
(b) Notice that the derivate of ln(5 − x) is − . This implies that
5−x

1
Z
ln(5 − x) = − dx
5−x

Now, in terms of power series, we have

∞ ∞
1 X xn X xn
 
1 1 1
= = =
5−x 5 1 − x5 5 n=0 5n n=0
5n+1

Thus,
∞ ∞
xn xn+1
Z X X
ln(5 − x) = − dx = − +C
n=0
5n+1 n=0
5n+1 (n + 1)

where C is a constant of integration. To find C, we put x = 0 in the

equation to get C = ln 5. Hence, we have
∞ ∞
X xn+1 X xn
ln(5 − x) = ln 5 − = ln 5 −
n=0
5n+1 (n + 1) n=1
n5n

x
The radius of convergence of the original series is, from < 1 =⇒
5
|x| < 5, R = 5. Thus, the radius of convergence of the given series is
also R = 5.

Example 73. Use a power series to approximate the definite integral to six
decimal places
0.2
1
Z
dx
0 1 + x5

122 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

9.3 Taylor Series and the Taylor Polynomial

The goal here is to determine which functions can be represented as power
series and how to get these representations. We start by making use of the
following theorem:

If f has a power series representation at a, that is, if

X∞
f (x) = cn (x − a)n |x − a| < R,
n=0
f (n) (a)
then its coefficients are given by cn =
n!
Thus, we get the Taylor series about a:
∞ ∞
X X f (n) (a)
f (x) = cn (x − a)n = (x − a)n
n=0 n=0
n!
0 00 000
f (a) f (a) f (a)
⇒ f (x) = f (a) + (x − a) + (x − a)2 + (x − a)3 + . . .
1! 2! 3!
If a = 0, we get the Maclaurin series :
P∞ f n (0) n f 0 (0) f 00 (0) 2 f 000 (0) 3
f (x) = n=0 n!
x = f (0) + 1!
x + 2!
x + 3!
x + ...

Example 74. Find the Maclaurin series of the function f (x) = ex and its
radius of convergence.
Solution. If f (x) = ex , then the nth derivative of f is f (n) (x) = ex and
f (n) (0) = e0 = 1 for n = 0, 1, 2, . . .
Hence the Taylor series about x = 0 (i.e. the Maclaurin series) is given by
∞ ∞
x
X f (n) (0) n X 1 n 1 1
e = x = x = 1 + x + x2 + · · · + xn + · · ·
n=0
n! n=0
n! 2! n!

9.4 nth-degree Taylor Polynomial of f at a

Question: Under what circumstances can we say that a function f (x) is
equal to the sum of its Taylor series? That is, when is it true that
∞
X f (n) (a)
f (x) = (x − a)n (9.5)
n=0
n!

123 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Let Tn be the partial sums of the Taylor series of f (called the nth-degree
Taylor polynomial of f ) and given by
n
X f (i) (a)
Tn (x) = (x − a)i
i=0
i!
f 0 (a) f 00 (a) f (n) (a)
= f (a) + (x − a) + (x − a)2 + . . . + (x − a)n
1! 2! n!
For instance, the first (n = 1), second (n = 2) and third (n = 3) degree
Taylor polynomials are given by:
f 0 (a)
T1 (x) = f (a) + (x − a)
1!
f 0 (a) f 00 (a)
T2 (x) = f (a) + (x − a) + (x − a)2
1! 2!
f 0 (a) f 00 (a) f 000 (a)
T3 (x) = f (a) + (x − a) + (x − a)2 + (x − a)3 .
1! 2! 3!
Now, equation (9.5) is true (f is the sum of its Taylor series) if
lim Tn (x) = f (x) (9.6)
n→∞

However, if Rn (x) is the remainder of the Taylor series then we know that
f (x) = Tn (x) + Rn (x)
⇒ lim f (x) = lim Tn (x) + lim Rn (x)
n→∞ n→∞ n→∞
⇒ f (x) = lim Tn (x) + lim Rn (x)
n→∞ n→∞

This means that equation (9.6) is true if

lim Rn (x) = 0 (9.7)
n→∞

The following theorem therefore holds from the above argument:

Theorem 39. (Taylor Theorem) If f (x) = Tn (x) + Rn (x) where Tn is

the nth degree Taylor polynomial of f at a and

lim Rn (x) = 0 (9.8)

n→∞

for |x − a| < R, then f is equal to the sum of its Taylor series on the
interval |x − a| < R.

124 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Question: How can we prove or show equation (9.7) or (9.8) for a given
function? We employ the following two facts (or theorems):

Theorem 40. (Taylor Inequality) If |f (n+1) (x)| ≤ M for |x − a| ≤ R,

then the remainder Rn (x) of the Taylor series satisfies the inequality

M
|Rn (x)| ≤ |x − a|n+1 for |x − a| ≤ R
(n + 1)!

Theorem 41.
xn |x|n
lim = 0 = lim
n→∞ n! n→∞ n!

for every real number x.

Examples:

1. Show that the Maclaurin series for sin x is given by

∞
X x2n+1
(−1)n for all x
n=0
(2n + 1)!

and prove that it represents sin x for all x.

2. Prove that ex is equal to the sum of its Maclaurin series.

2
3. Evaluate e−x dx as an infinite series
R

Solution. 1. Let

f (x) = sin x =⇒ f (0) = 0

f 0 (x) = cos x =⇒ f 0 (0) = 1

f 00 (x) = − sin x =⇒ f 00 (0) = 0
f 000 (x) = − cos x =⇒ f 000 (0) = −1
f (4) (x) = sin x =⇒ f (4) (0) = 0
f (5) (x) = cos x =⇒ f (5) (0) = 1

125 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

We see that the derivatives repeat in a cycle of four, thus, we have

∞
X f (n) (0) n 0 f 00 (0) 2 f 000 (0) 3
f (x) = x = f (0) + f (0)x + x + x + ···
n=0
n! 2! 3!

x3 x5 x7 x2n+1
=⇒ sin x = x − + − + · · · + (−1)n + ···
3! 5! 7! (2n + 1)!
∞ 2n+1
n x
X
∴ sin x = (−1)
n=0
(2n + 1)!
We now need to show that lim Rn (x) = 0 in order to prove that
n→∞
the series actually represents sin x. Since f (n+1) (x) is either ± sin x or
± cos x, we have |f (n+1) (x)| ≤ 1 for all x. So letting M = 1 in the
Taylor Inequality, we have

M |x|n+1
|Rn (x)| ≤ |xn+1 | = .
(n + 1)! (n + 1)!

|x|n+1
lim = 0 =⇒ |Rn (x)| ≤ 0
n→∞ (n + 1)!

Hence, by the squeeze theorem, we have

lim |Rn (x)| = 0 ∴ lim Rn (x) = 0.

n→∞ n→∞

Therefore sin x is equal to the sum of its Maclaurin series.

126 Dr. Joseph K. Ansong

Chapter 10

Integration

10.1 Introduction
Definition 27. A function F is an antiderivative of a function f on an
interval I if F 0 (x) = f (x) for all x in I.

Comparison Properties of the Integration:

Z b
1. If f (x) ≥ 0 for a ≤ x ≤ b, then f (x)dx ≥ 0.
a
Z b Z b
2. If f (x) ≥ g(x) for a ≤ x ≤ b, then f (x)dx ≥ g(x)dx.
a a

3. If m ≤ f (x) ≤ M for a ≤ x ≤ b, then

Z b
m(b − a) ≤ f (x)dx ≤ M (b − a)
a

Property 1 says that areas are positive. Property 2 implies that a bigger
function has a bigger integral. Property 3 says that the area under the graph
of f is greater than the area of the rectangle with height m and less than the
area of the rectangle with height M .

127
Math 223 : Calculus II J.K.A

10.2 Fundamental Theorem of Calculus (FTC)

The Fundamental Theorem of Calculus (FTC) is the most important theorem
in calculus. It states that differentiation and integration are inverse processes.
In other words, if f is integrated and the result is then differentiated, we
arrive back at the original function f . Also, if we take a function F , first
differentiate it, and then integrate the result, we get back the original function
F in the form F (b) − F (a), where a and b are the lower and upper limits
of the integral. In what follows, we first give an intuitive idea of the FTC
before explicitly stating the theorem.

10.2.1 Intuitive idea of the FTC

Suppose f is a continuous function on [a, b], and define the function g by
Z x
g(x) = f (t)dt (10.1)
a
where a ≤ x ≤ b. Here, we want to show, intuitively, that the derivate of
g(x) is the function f (x). Note that g depends on only x, the variable in the
upper limit of the integral. If x is fixed number, the integral is just a number,
R x under the graph of f if f (x) ≥ 0. However, if x varies,
representing the area
then the number a f (t)dt also varies and defines the function g(x). Let’s
try to compute g 0 (x) from the definition of the derivative. If f (x) ≥ 0, then
g(x + h) − g(x) is obtained by subtracting areas, and it is the area under the
graph of f from x to x + h. If h is small, the area is approximately equal to
the area of the rectangle with height f (x) and width h:
g(x + h) − g(x) ≈ hf (x)
g(x + h) − g(x)
=⇒ ≈ f (x)
h
Thus, intuitively, we expect that
g(x + h) − g(x)
g 0 (x) = lim = f (x)
h→0 h

10.2.2 The Mean Value Theorem (MVT) for integrals

Definition 28. (Average Value of a Function) If f is integrable on [a, b],
then the average value of f over [a, b] is the number
Z b
1
fav = f (x)dx. (10.2)
b−a a

128 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Figure 10.1:

Geometrically, the average value of f is the height of the rectangle with

base on the interval [a, b] and having the same area as the area of the region
under the graph of f on [a, b]. Observe from equation (10.2) that
Z b
area of rectangle = (b − a)fav = f (x)dx = area of region under f
a

The following theorem, referred to as the Mean Value Theorem for Integrals
in some text books, guarantees that fav is always attained at (at least) one
number in an interval [a, b] if f is continuous.

Theorem 42. (The Mean Value Theorem for Integrals) If f is

continuous on [a, b], then there exists a number c in [a, b] such that
b
1
Z
f (c) = f (x)dx (10.3)
b−a a

Proof. Since f is continuous on the interval [a, b], the Extreme Value Theo-
rem tells us that f attains an absolute minimum value m at some number
in [a, b] and an absolute maximum value M at some number in [a, b]. So
m ≤ f (x) ≤ M for all x in [a, b]. By Property 3 of integrals, we have
Z b
m(b − a) ≤ f (x)dx ≤ M (b − a).
a

If b > a, then dividing by (b − a) results in

Z b
1
m≤ f (x)dx ≤ M.
b−a a

129 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Because the number

b
1
Z
f (x)dx
b−a a

lies between m and M , the Intermediate Value Theorem guarantees the ex-
istence of at least one number c in [a, b] such that
b
1
Z
f (c) = f (x)dx,
b−a a

as required.

Example 75. Find the number c which is guaranteed by the MVT for In-
tegrals for the function f (x) = 4 − 2x on the interval [0, 2].

Solution. The function f (x) = 4 − 2x is continuous on the interval [0, 2].

Therefore, the MVT for Integrals states that there is a number c in [0.2] such
that Z b
1
f (c) = f (x)dx
b−a a
where a = 0 and b = 2. Thus,
2
1 1
Z
4 − 2c = (4 − 2x)dx = (4) = 2
2−0 0 2

=⇒ 2c = 4 − 2 = 2
∴ c = 1.

We now state Part 1 of the Fundamental Theorem of Calculus; ab-

breviated as FTC1:

Theorem 43. (The Fundamental Theorem of Calculus, Part 1)

If f is continuous on [a, b], then the function g defined by
Z x
g(x) = f (t)dt a≤x≤b
a

is an antiderivative of f , that is,

x
d
Z
0
g (x) = f (t)dt = f (x).
dx a

130 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Proof. Let x and x + h be in the open interval (a, b), where h 6= 0. Then,
Z x+h Z x
g(x + h) − g(x) = f (t)dt − f (t)dt
a a
Z x Z x+h  Z x
= f (t)dt + f (t)dt − f (t)dt
a x a
Z x+h
= f (t)dt
x

By the Mean Value Theorem for Integrals, there exists a number c between
x and x + h such that Z x+h
f (t)dt = f (c) · h
x
Therefore, we have
x+h
g(x + h) − g(x) 1 f (c) · h
Z
= f (t)dt = = f (c).
h h x h

Now, observe that as h approaches 0 (from either the left 0− or the right 0+
side), the number c, which lies between x and x + h, approaches x, and by
continuity, f (c) approaches f (x). Therefore, we get
x+h
g(x + h) − g(x) 1
Z
0
g (x) = lim = lim f (t)dt = lim f (c) = f (x),
h→0 h h→0 h x h→0

as desired.
Z x √
Example 76. Find the derivative of the function g(x) = 2 + t3 dt
0
√
Solution. Since the function f (t) = 2 + t3 is continuous, the FTC (Part
1) gives √
g 0 (x) = 2 + x3
Z x4
d
Example 77. Find sec tdt
dx 1
Solution. Since the upper limit of the integral is not the simple variable, x,
we need to use the Chain Rule together with the FTC1. Let u = x4 . Then
du
= 4x3 . Thus
dx Z x4 Z u
d d
sec tdt = sec tdt
dx 1 dx 1

131 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Z u 
d du du
= sec tdt = sec u
du 1 dx dx
= sec(x4 ) · 4x3
Exercise. Find the derivative of the function:
Z x Z 3√
1
(a) F (x) = 2
dt (b) G(x) = 5 + t2 dt
−1 5 + t x
Z x3
dy
Exercise. If y = sin(t2 )dt, what is ?
0 dx

Theorem 44. (The Fundamental Theorem of Calculus, Part 2

(FTC2)) If f is continuous on [a, b],
Z b
f (x)dx = F (b) − F (a)
a

where F is any antiderivative of f , that is, F 0 = f .

Remark. • Part 2 of the FTC says that if a function F is first differen-

tiated, and the result is integrated, we get back the original function
F , but in the form F (b) − F (b).
• Part 1 and Part 2 of the FTC state that differentiation and integration
are inverse processes.
• The proof of FTC2 is an exercise for you to try.
Example 78. Evaluate the following integrals:
Z 3 Z π
2 (c) sin xdx
(a) (3x + 2x − 1)dx
0 0
Z 4 √
(b) 3 xdx
0
Z 3
Solution. (a) Let I = (3x2 + 2x − 1)dx
0
Z 3 3
(3x2 + 2x − 1)dx = x3 + x2 − x 0

I=
0

= (27 + 9 − 3) − 0 = 33

132 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Z 4 √
(b) Let I = 3 xdx
0

4
4 √
 
2 3/2
Z
I= 3 xdx = 3 x
0 3
0

4
= 2x3/2 = 2(8 − 0) = 16
0
Z π
(c) Let I = sin xdx
0

π
I = − cos x = − cos π + cos 0 = 1 + 1 = 2,
0

since cos π = −1.

Example 79. 1. Use FTC to find the derivative of the function.

Z u
1
(a) g(u) = 4
dt
π 1+t
Z 1
x
(b) h(x) = sin4 tdt
2
3x 3x 0 3x 
u−1
Z Z Z Z
(c) g(x) = du Hint: = +
2x u+1 2x 2x 0

2. Find a function f and a number a such that

Z x
f (t) √
6+ dt = 2 x
a t2
Z π
1 1
Solution. 1. (a) g(u) = 4
dt, g 0 (u) = .
u 1+t 1 + u4
Z 1
x 1 du 1
(b) h(x) = sin4 tdt. Let u = ⇒ =− 2
2 x dx x
1 Z u 
d d du
Z
x
4 4
⇒ sin tdt = sin tdt
dx 2 du 2 dx

sin4 ( x1 )
   
du
4 1 1
= (sin u) · = sin4 · − 2 =−
dx x x x2

133 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

3x Z 3x
u−1 d u−1
Z
0
(c) g(x) = du ⇒ g (x) = du
2x u + 1 dx 2x u + 1
Z 0 Z 3x
u−1 u−1

d
= du + du (10.4)
dx 2x u + 1 0 u+1
Z 0 Z 2x
u−1 u−1
Let I1 = du = − du
2x u + 1 0 u+1
ds
Let s = 2x ⇒ =2
dx
Z s
u−1

dI1 d ds
∴, =− du ·
dx ds 0 u+1 dx
s−1 2x − 1
   
=− · (2) = − (2) (10.5)
s+1 2x + 1
R 3x u − 1 ds
Let I2 = 0
du Let s = 3x ⇒ =3
u+1 dx
Z s
u−1

dI2 d ds
∴, = du ·
dx ds 0 u+1 dx
s−1 3x − 1
 
= (3) = (10.6)
s+1 3x + 1
Substitute (10.5) and (10.6) into (10.4) to get
2x − 1 3x − 1
   
0
g (x) = −2 +3
2x + 1 3x + 1

(2)
x √ x √
f (t) f (t)
Z Z
6+ 2
dt = 2 x ⇒ 2
dt = 2 x − 6 (10.7)
a t α t
Let x
f (t) √
Z
g(x) =dt = 2 x−6
α t2
f (x) 1 1 1 3
=⇒ g 0 (x) = 2 = 2 · x− 2 ⇒ f (x) = x2 · x− 2 = x 2
x 2
3
∴ f (x) = x 2
From (10.7), we get
x 3
t2 √
Z
2
dt = 2 x − 6,
α t

134 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

x √ x
√ √
Z
1
=⇒ t− 2 dt = 2 x − 6 ⇒ 2 t = 2 x − 6,
α α
√ √ √
=⇒ 2 x − 2 α = 2 x − 6,
√ √
=⇒ 2 α = 6 =⇒ α = 3,
∴ α = 9.

Example 80. (a) Find the average value of the function f (x) = 1 + x2 on
[−1, 2].

(b) Find a number c in [−1, 2] that satisfy the MVT for integrals for
f (x) = 1 + x2 .
Z b Z 2
1 1
Solution. (a) fave = f (x)dx = (1 + x2 )dx
b−a a 2 − (−1) −1
 2    
1 1 3 1 8 1
= x+ x = 2+ − −1 −
3 3 −1 3 3 3
 
1 14 4 18
∴, fave = + = =2
3 3 3 9

(b) By MVT for integrals,

fave = f (c) ⇒ 1 + c2 = 2 ⇒ c2 = 1 ⇒ c = ±1

There are two values of c in this case (see Figure 10.2).

Z 3
Example 81. 1. If f is continuous and f (x)dx = 8, show that f takes
1
on the value 4 at least once on the interval [1, 3].

2. Find the numbers b such that the average value of f (x) = 2 + 6x − 3x2
on the interval [0, b] is equal to 3.

Solution. 1. By MVT for integrals, there exists c in [1, 3] such that

3
1
Z
f (c) = f (x)dx
3−1 1

1
⇒ f (c) = (8) = 4.
2

135 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Figure 10.2:

b
1
Z
2. fave = (2 + 6x − 3x2 )dx
b−a 0

1 b 1
= 2x + 3x2 − x3 0 = [2b + 3b2 − b3 ]
b b
fave = 2 + 3b − b2
fave = 3 ⇒ 2 + 3b − b2 = 3 ⇒ b2 − 3b + 1 = 0
√ √
3± 9−4 3± 5
b= =
2 2
√ √
3+ 5 3− 5
∴, b = = 2.62 and = 0.38
2 2

10.3 Riemann Sums & the Definite Integral:

Integration as a Sum
The Riemann sum is a fundamental concept for the definition of the definite
integral. It is based on the idea that the area under the graph of a function
f can be obtained by dividing the area into smaller rectangular areas and
summing them up.

136 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Figure 10.3:

As shown in Figure 10.3, divide the area under the graph into subintervals,
or partitions:
[x0 , x1 ], [x1 , x2 ], · · · [xn−1 , xn ]
b−a
with x0 = a, xn = b, and the width of each strip is ∆x = .
n
The right-hand endpoints of each sub-interval is:
x1 = a + ∆x0 , x2 = x1 + ∆x = a + 2∆x,
x3 = a + 3∆x, · · · , xn = a + n∆x
That is,
xi = a + i∆x i = 1, 2, · · · , n. (10.8)
The left-hand endpoints of each sub-interval is:
x0 = a, x1 = a + ∆x, x2 = a + 2∆x, · · · , xn−1 = a + (n − 1)∆x
That is,
xi−1 = a + (i − 1)∆x i = 1, 2, · · · n. (10.9)

We may also use any sample point, x∗i in the ith subinterval [xi−1 , xi ] for the
height of the rectangle, f (x∗i ), as illustrated in Figure 10.4. For midpoints of
Riemann Sums, we have
 
∗ xi−1 + xi 1 1
xi = = [a + (i − 1)∆x + a + i∆x] = a + i − ∆x. (10.10)
2 2 2

137 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Figure 10.4:

Definition 29. The area A of the region S that lies under the graph of
the continuous function f is the limit of the sum of areas of approximating
rectangles (using the right endpoints):

A = lim Rn = lim [f (x1 )∆x + f (x2 )∆x + · · · + f (xn )∆x]

n→∞ n→∞
n
X
⇒ A = lim f (xi )∆x (10.11)
n→∞
i=1

OR (using the left endpoints)

A = lim Ln = lim [f (x0 )∆x + f (x1 )∆x + · · · + f (xn−1 )∆x]

n→∞ n→∞
n
X
⇒ A = lim Ln = lim f (xi−1 )∆x (10.12)
n→∞ n→∞
i=1

OR (using any sample points x∗i )

A = lim [f (x∗1 )∆x + f (x∗2 )∆x + · · · + f (x∗n )∆x]

n→∞
n
X
⇒ A = lim f (x∗i )∆x (10.13)
n→∞
i=1

138 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Summation Formulas:
To compute the Riemann sums, the following summation formulas are useful:
n
X
c = cn
i=1
n
X n(n + 1)
i=
i=1
2
n
X n(n + 1)(2n + 1)
i2 =
i=1
6
n  2
X
3 n(n + 1)
i =
i=1
2

Example 82. Evaluate the Riemann sum for f (x) = x2 with a = 0, b = 1

and n = 5, taking the sample points to be the

(a) right endpoints.

(b) left endpoints.

(c) midpoints.

Solution.
f (x) = x2 , a = 0, b = 1, n = 5

b−a 1−0 1
=⇒ ∆x = = =
n 5 5

(a) For the right endpoints, we have

i i
xi = a + i∆x = 0 + = ; i = 1, 2, · · · , 5.
5 5

So the endpoints are

1 2 3 4 5
x1 = , x 2 = , x 3 = , x 4 = , x 5 = = 1
5 5 5 5 5

Thus, the Riemann sum using the right endpoints with n = 5, R5 , is

139 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

given by
5
X
R5 = f (xi )∆x
i=1
= f (x1 )∆x + f (x2 )∆x + f (x3 )∆x + f (x4 )∆x + f (x5 )∆x
= [f (x1 ) + f (x2 ) + f (x3 ) + f (x4 ) + f (x5 )]∆x
         
1 2 3 4 1
= f +f +f +f + f (1) ·
5 5 5 5 5
"  #
2  2  2  2
1 2 3 4 1
= + + + + (1)2 ·
5 5 5 5 5
 
1
= [2.2]
5
∴ R5 = 0.44

(b) For the left endpoints, we have

i−1 i−1
xi−1 = a + (i − 1)∆x = 0 + = ; i = 1, 2, · · · , 5.
5 5
So the endpoints are
1 2 3 4
x0 = a = 0, x1 = , x2 = , x3 = , x4 =
5 5 5 5
Thus, the Riemann sum using the left endpoints with n = 5, L5 , is
given by
5
X
L5 = f (xi−1 )∆x
i=1
= f (x0 )∆x + f (x1 )∆x + f (x2 )∆x + f (x3 )∆x + f (x4 )∆x
= [f (x0 ) + f (x1 ) + f (x2 ) + f (x3 ) + f (x4 )]∆x
        
1 2 3 4 1
= f (0) + f +f +f +f ·
5 5 5 5 5
"  2  2  2  2 #
1 2 3 4 1
= (0)2 + + + + ·
5 5 5 5 5
 
1
= [1.2]
5
∴ L5 = 0.24

140 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

(c) For the midpoints, we have

i − 1/2 2i − 1
 
1
x∗i =a+ i− ∆x = 0 + = .
2 5 10

So the midpoints are

1 3 5 7 9
x∗1 = , x∗2 = , x∗3 = , x∗4 = , x∗5 =
10 10 10 10 10
Thus, the Riemann sum using the midpoints with n = 5, M5 , is given
by
5
X
M5 = f (x∗i )∆x
i=1
= f (x∗1 )∆x + f (x∗2 )∆x + f (x∗3 )∆x + f (x∗4 )∆x + f (x∗5 )∆x
= [f (x∗1 ) + f (x∗2 ) + f (x∗3 ) + f (x∗4 ) + f (x∗5 )]∆x
          
1 3 5 7 9 1
= f +f +f +f +f ·
10 10 10 10 10 5
"   2  2  2  2 #
2
1 3 5 7 9 1
= + + + + ·
10 10 10 10 10 5
 
1
= [1.65]
5
∴ M5 = 0.33

The next example evaluates the exact value of the Riemann sum in ex-
ample 82. And you should notice that the value of the Riemann sum using
the midpoints is much closer to the exact value.

Example 83. Find the area under the graph of y = x2 from 0 to 1 using
the mid-point of subintervals to approximate the height of the rectangles.

Solution.
b−a 1−0 1
∆x = = = .
n n n
The midpoints of the subintervals are given by (see equation (10.10)):
    
1 1 1 1
x̄i = (xi−1 + xi ) = a + i − ∆x = i −
2 2 2 n

141 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

n
X
A = lim f (x̄i )∆x, f (x) = x2
n→∞
i=1
n   2
X 1 1 1
= lim i− ·
n→∞
i=1
2 n n
n  2
1 X 1
= lim 3 i−
n→∞ n 2
i=1
n  
1 X 2 1
=⇒ A = lim 3 i −i+
n→∞ n 4
i=1

Applying the summation formulas, we get

 
1 n(n + 1)(2n + 1) n(n + 1) 1
∴ A = lim 3 − + n
n→∞ n 6 2 4
      
1 1 1 1 1 1 1
= lim 1+ 2+ − + + 2
n→∞ 6 n n 2 n n2 4n
1 1 2 1
= (1)(2) − (0 + 0) + 0 = = .
6 2 6 3

Exercise. Estimate the area under the graph of y = x2 from 0 to 1 by taking

the sample points to be midpoints and using four subintervals. Compare your
answers to L5 and R5 determinedpreviously.

Example 84. Use Riemann Sums to find the area of the region under the
graph of f (x) = 4 − x2 on the interval [−2, 1].

Solution. Partition the area under the graph (see Figure 10.5) into n subin-
tervals with
b−a 1 − (−2) 3
∆x = = = .
n n n

Using the right endpoints [you may choose left endpoints or midpoints] to
estimate the height of rectangles, we have (see equation (10.8)):

 
3
x∗i = −2 + i .
n

142 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Figure 10.5:

Thus,
n
X
A = lim f (x∗i )∆x, f (x) = 4 − x2
n→∞
i=1
n  
X 3i 3
= lim f −2 +
n→∞
i=1
n n
n
"  2 #
3X 3i
= lim 4 − −2 +
n→∞ n n
i=1
n   
3X 12 9 2
= lim 4 − 4 − i + 2i
n→∞ n n n
i=1
n  
3 X 12 9 2
= lim i − 2i
n→∞ n n n
i=1
" n n
#
36 X 27 X 2
= lim i− 3 i
n→∞ n2 n i=1
i=1
 
36 n(n + 1) 27 n(n + 1)(2n + 1)
= lim · − 3·
n→∞ n2 2 n 6
     
1 27 1 1
= lim 18 1 + − 1+ 2+
n→∞ n 6 n n
27
= 18 − (2) = 18 − 9 = 9.
6
Exercise. Repeat the example above using the midpoints of subintervals to

143 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

estimate the height of rectangles.

Solution.
   
1 1 3
x̄i = a + i − ∆x = −2 + i −
2 2 n
n n    
X X 1 3 3
A = lim f (x̄i )∆x = lim f −2 + i −
n→∞
i=1
n→∞
i=1
2 n n
n
(    2 )
3X 1 3
= lim 4 − −2 + i −
n→∞ n 2 n
i=1
n
(    2 )
3X 1 3 1 9
= lim 4 i− − i−
n→∞ n 2 n 2 n2
" i=1 n   n  2 #
36 X 1 27 X 1
= lim i− − 3 i−
n→∞ n2 2 n 2
i=1 i=1
  n  
36 n(n + 1) 1 27 X 2 1
= lim 2 − n − lim 3 i −i+
n→∞ n 2 2 n→∞ n 4
i=1
   
36 n(n + 1) 1 27 n(n + 1)(2n + 1) n(n + 1) 1
= lim 2 − n − lim 3 − + n
n→∞ n 2 2 n→∞ n 6 2 4
   
36 n(n + 1) 1 27 n(n + 1)(2n + 1) n(n + 1) 1
= lim 2 − n − lim 3 − + n
n→∞ n 2 2 n→∞ n 6 2 4
          
1 1 1 1 1 1 1 1 1 1
= lim 36 1+ − − lim 27 1+ 2+ − + + 2
n→∞ 2 n 2n n→∞ 6 n n 2 n n2 4n
   
1 1
= 36 − 27 (2) = 18 − 9 = 9
2 6

10.3.1 The Distance Problem

How do you find the distance travelled by an object during a certain time
period if the velocity is known at all times?
Remark. (1) If the velocity is constant, then the distance is given as:
distance = velocity × time
Eg: A car travelled along a straight line with a velocity of 60km/hr
over a two hour period (see Figure 10.6a). v = 60km/hr, 0 6 t 6 2.
Then the distance
 
km
d = 60 × 2 hr = 120 km.
hr

144 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Figure 10.6:

Remark. (2) If the velocity is not constant but a function ofP time, t, such
that v = f (t) during a time interval [a, b], then d = lim ni=1 f (t∗i )∆t
n→∞
b−a ∗
where ∆t = , ti = ti−1 for left endpoints, t∗i = ti for right end-
n
points or t∗ could be midpoint or any point in [ti−1 , ti ] (see Figure
10.6b).

10.4 The Definite Integral

Definition 30. If f is a continuous function defined for a 6 x 6 b, we divide
the integral [a, b] into n subintervals of equal width
b−a
∆x = .
n
We let x0 , x1 , x2 , · · · , xn be the endpoints of these subintervals where a = x0
and b = xn , and we choose sample points x∗1 , x∗2 , · · · , x∗n in these subintervals,
so x∗i lies in the ith subinterval [xi−1 , xi ]. Then the definite integral of f from
a to b is given as:
Z b Xn
f (x)dx = lim f (x∗i )∆x.
a n→∞
i=1
Z b
Remark. (a) f (x)dx is a number; it does not depend on x.
a

(b) f (x) is not necessarily a positive function. If f has both positive and
negative values, then a definite integral can be interpreted as a differ-

145 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

ence of areas above the x-axis and those below the x-axis (see Figure
10.7):
Z b
f (x)dx = A1 − A2
a

Figure 10.7:

n
X  3 
Example 85. 1. Express lim xi + xi sin(xi ) ∆x as an integral on
n→∞
i=1
the interval [0, π].
2. Use Riemann summation to compute the integral:
Z 5 Z 4
(a) 4xdx (b) (2x2 + 3)dx
0 0
n
X  3 
Solution. (1) Comparing lim xi + xi sin xi ∆x to the left of a defi-
n→∞
i=1
nite integral, we let f (x) = x3 + x sin x, a = 0, b = π.
Thus,
Xn Z π
 3
(x3 + x sin x)dx.

lim x + x sin x ∆x =
n→∞ 0
i=1
Z 5
Solution. 2(a) I = 4xdx.
0
5−0 5
Subdivide [0, 5] into n sub-intervals of equal length ∆x = = .
  n n
5 5
Using right endpoints: xi = a + i∆x = 0 + i = i
n n
Z 5 X n
∴ 4xdx = lim f (xi )∆x
0 n→∞
i=1
n n
X 5 20 X 5
= lim (4xi ) · = lim i
n→∞
i=1
n n→∞ n
i=1
n

146 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

 
100 Pn 100 n(n + 1) 1
= lim 2 lim 2 ·
i=1 i = n→∞ = lim 50 1 + = 50.
n→∞ n n 2 n→∞ n
(b)
4
4−0 4
Z
I= (2x2 + 3)dx, ∆x = =
0 n n
4i
xi = a + i∆x = , f (x) = 2x2 + 3
n
Z 4 n
X
2
I= (2x + 3)dx = lim f (xi )∆x
0 n→∞
i=1
n
"   # n 
2 
X 4i 4 4 X 32 2
= lim 2 +3 = lim i +3
n→∞
i=1
n n n→∞ n i=1 n2
   
4 32 n(n + 1)(2n + 1) 128 n(n + 1)(2n + 1)
= lim · + 3n = lim · + 12
n→∞ n n2 6 n→∞ n3 6
128
(2) + 12, =
6
128 164
= + 12 = .
3 3
R2
Example 86. Use Riemann Sums to evaluate 1 (x + 1)dx.
2−1 1 i
Solution. ∆x = = , xi = a + i∆x = 1 +
Z 2n n n 
n 
X i 1
f (x) = x + 1 ⇒ (x + 1)dx = lim 1+ +1 = 2.5
1 n→∞
i=1
n n

147 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Try the following exercises before looking up their solutions. Keep the
following expressions in mind:

Xn Z b
lim f (x̄i )∆x = f (x)dx




 n→∞ a
 i=1
n

 Z b
 X
lim f (a + i∆x)∆x = f (x)dx
n→∞ a
i=1


n
 Z b
b−a b−a

 X  
lim f a+i· · = f (x)dx



n→∞
i=1
n n a

Exercise. (1) By interpreting the limit as a definite integral, show that:

n
X 1
(a) lim = ln 2.
n→∞
i=1
n+i
1 + 2 p + 3 p + · · · + np
p
1
(b) lim p+1
= .
n→∞ n p+1
1 n n
(2) Find the limit as n → ∞ of the sum + 2
+ + ··· +
n (n + 1) (n + 2)2
n
.
(2n − 1)2
Z 3
(3) Use Riemann sums to evaluate (x3 − 6x)dx.
0

148 Dr. Joseph K. Ansong

 Math 223 : Calculus II J.K.A

n
X 1
Solution.(1)(a) We want to show that lim = ln 2.
n→∞
i=1
n+i

n n
X 1 X 1
lim = lim
n→∞
i=1
n+i n→∞
i=1
n(1 + ni )
n n
X 1 1 X
lim i · = lim f (x̄i ) · ∆x,
n→∞
i=1
1+ n
n n→∞
i=1

i (2 − 1)
where x̄i = 1 + =1+i ,
n n
1
a = 1, b = 2, f (x̄i ) =
x̄i
2−1 1
∆x = = .
n n
Thus n 2
1 1 1
X Z
lim i

· = dx
n→∞
i=1
1+ n n 1 x
2
= ln x 1
= ln(2) − ln(1) = ln 2.
Alternatively, we have
i 1−0
x̄i = =0+i ,
n n
1−0 1
a = 0, b = 1, ∆x = =
n n
1
f (x̄i ) = .
1 + x̄i
So
1
1 1 1
X Z
lim i

· = dx,
n→∞ 1+ n n 0 1+x
1
= ln(1 + x) = ln 2 − ln 1 = ln 2.
0

1p + 2p + · · · + np 1
(1)(b) lim p+1
= . Let
n→∞ n p+1

1p + 2p + · · · + np
 p
1 + 2 p + · · · + np

1
Sn = p+1
= p
·
n n n

149 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

 p  p  n p  1 X n  p n
1 2 i 1 X
⇒ + + ··· + · = · = f (x̄i )∆x,
n n n n i=1
n n i=1

where
i 1−0
x̄i = =0+i· , a = 0, b = 1
n n
b−a 1−0 1
∆x = = = , f (x̄i ) = x̄i p
n n n
n
X i p 1 Z 1
 
⇒ lim Sn = lim · = xp dx
n→∞ n→∞
i=1
n n 0

1
xp+1 1 1
= = −0=
p+1 0 p+1 p+1
1 n n n n n
(2) Sn = + 2
+···+ 2
= 2+ 2
+···+ .
n (n + 1) (2n − 1) n (n + 1) (2n − 1)2
n−1 n−1
X n X n
= =
.
(n + i)2 n 2 1+ i 2
i=0 i=0 n

i i(2 − 1)
Let x̄i = 1 + =1+ , a = 1, b = 2, and
n n
1 1
∆x = , f (x̄i ) = ¯2 .
n xi
2 2 2
1 1 1 1
Z Z
=⇒ lim Sn = dx = x−2 dx = − =− +1= .
n→∞ 1 x2 1 x 1 2 2

150 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

10.5 Evaluating Definite Integrals

Theorem 45. (Evaluation Theorem:) If f is continuous on the integral

[a, b], then
Z b
f (x)dx = F (b) − F (a)
a

0 dF
where F is any anti-derivative of f, that is, F = = f.
dx

Proof. Divide [a, b] into n sub intervals with endpoints x0 , x1 , x2 , · · · , xn and

b−a
with width ∆x = , where a = x0 and b = xn . Now,
n
F (b) − F (a) = F (xn ) − F (x0 )
= F (xn ) − F (xn−1 ) + F (xn−1 ) − F (xn−2 ) + F (xn−2 ) + · · ·
−F (x2 ) + F (x2 ) − F (x1 ) + F (x1 ) − F (x0 )
= [F (xn ) − F (xn−1 )] + [F (xn−1 ) − F (xn−2 )] + · · ·
+[F (x1 ) − F (x0 )]
n
X
⇒ F (b) − F (a) = [F (xi ) − F (xi−1 )] (10.14)
i=1

Since F is continuous and differentiable on [xi−1 , xi ], we can apply MVT to

F. Thus, there exists a number x∗i [xi−1 , xi ] such that:
F (xi ) − F (xi−1 )
F 0 (x∗i ) =
xi − xi−1
⇒ F (xi ) − F (xi−1 ) = F (xi )[xi − xi−1 ] = f (x∗i )∆x
0 ∗
(10.15)
Substitute (10.15) into (10.14) to get
n
X
F (b) − F (a) = f (x∗i )∆x
i=1

Taking the limit as n → ∞ on both sides, we get

n
X
F (b) − F (a) = lim f (x∗i )∆x
n→∞
i=1

151 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Z b
∴ F (b) − F (a) = f (x)dx.
a

Example 87. Find the area under the cosine curve from 0 to b, where
π
06b6 .
2

Figure 10.8:

Solution.
Z b b
A= cos(x)dx = sin x = sin b − sin 0 = sin b
0 0

π π
Observe that if b = , then A = sin = 1 (see Figure 10.8).
2 2

Properties of Definite Integrals

Z b
(1) cdx = c(b − a)
a
Z b Z b Z b
(2) [f (x) ± g(x)] dx = f (x)dx ± g(x)dx
a a a
Z b Z b
(3) cf (x)dx = c f (x)dx, c = constant
a a
Z b Z b Z b
(4) [f (x) − g(x)]dx = f (x)dx − g(x)dx
a a a
Z b Z c Z b
(5) f (x)dx = f (x)dx + f (x)dx
a a c

152 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Z b
(6) If f (x) > 0 on [a, b], then f (x)dx > 0.
a
Z b Z b
(7) If f (x) > g(x) on [a, b], then f (x)dx > g(x)dx
a a
Z b
(8) If m 6 f (x) 6 M on [a, b], then m(b − a) 6 f (x)dx 6 M (b − a)
a
Z b Z a
(9) f (x)dx = − f (x)dx
a b
Z a
(10) f (x)dx = 0
a

Example 88. 1) Use the properties of definite integrals to evaluate:

Z 1 Z 2
2
(a) (5 + 2x )dx (b) (x3 + 1)dx
0 1
Z 2 Z 2
2) Given that f (x)dx = 3 and f (x)dx = 2, evaluate
−2 0
Z 0 Z 0 Z −2
(a) f (x)dx (c) 3f (x)dx − 2f (x)dx
2 2 0
Z 0
(b) (f (x) + 3)dx
−2
1 1  
2 2 17
Z
Solution. 1(a) (5 + 2x )dx = 5x + x3 2
= 5+ −0=
0 3 0 3 3
Z 2 Z 2 Z 2
3 3
(b) (x + 1)dx = x dx + 1dx
1 1 1

2 2
1
= x4 + x
4 1 1

1 15 19
= (16 − 1) + (2 − 1) = +1=
4 4 4
Z 2 Z 2
(2) We are given that f (x)dx = 3 and f (x)dx = 2
−2 0
Z 0 Z 2
(a) f (x)dx = − f (x)dx = −2
2 0

153 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Z 0 Z 0 Z 0
(b) (f (x) + 3)dx = f (x)dx + 3dx
−2 −2 −2
Note that:
Z 2 Z 0 Z 2
f (x)dx = f (x)dx + f (x)dx
−2 −2 0
Z 0 Z 0
⇒3= f (x)dx + 2 ⇒ f (x)dx = 3 − 2 = 1
−2 −2

From the equation in 2(b) above, we have

Z 0 Z 0
⇒ (f (x) + 3)dx = 1 + 3 dx
−2 −2
 0
=1+3 x = 1 + 3(0 + 2) = 1 + 6 = 7
−2
Z 0 Z −2 Z 0 Z −2
(c) 3f (x)dx − 2f (x)dx = 3 f (x)dx − 2 f (x)dx
2 0 2 0

= 3(−2) − 2(−1) = −6 + 2 = −4.

10.6 Indefinite Integrals

If F (x) is an anti-derivative Rof f (x), that is, F 0 (x) = f (x), then f (x)dx =
R

F (x). The indefinite integral f (x)dx is a function, while the definite integral
Rb
a
f (x)dx is a number.
Example 89. Evaluate:
Z
(a) xn dx (c) sin xdx

1
Z Z
x
(b) e dx (d) dx
x
xn+1
Z
Solution. (a) xn dx = +c
n+1
Z
ex dx = ex + c
R
(b) (c) sin xdx = − cos x + c

1
Z
(d) dx = ln |x| + c
x

154 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Exercise. Verify by differentiation that the following formula is correct:

x sin 2x
Z
sin2 xdx = − + C,
2 4

where C is a constant.

10.6.1 Total change Theorem

The integral of a rate of change is the total change:
Z b
F 0 (x)dx = F (b) − F (a).
a

Application
(a) If an object moves along a straight line with position function s(t), then
its velocity is v(t) = s0 (t). So
Z t2 Z t2
v(t)dt = s0 (t)dt = s(t2 ) − s(t1 )
t1 t1

is the change of position or displacement during the time from t1 to t2 .

Z t2
(b) The total distance travelled is given by |v(t)|dt where |v(t)| is the
t1
speed (see Figure 10.9).

Figure 10.9:

(c) The acceleration of the object is a(t) = v 0 (t), so

Z t2
a(t) = v(t2 ) − v(t1 )
t1

is the change in velocity from t1 to t2 .

155 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

10.7 Techniques of Integration

10.7.1 Method of Substitution
In general, this method works for integrals of the form f (g(x))g 0 (x)dx.

Theorem 46. If u = g(x) is a differentiable function whose range is an

interval I and f is continuous on I, then
Z Z
0
f (g(x))g (x)dx = f (u)du.

Proof. Let F be an anti-derivative of f. That is F 0 = f or F = f dx. By

R

chain rule,
d
F (g(x)) = F 0 (g(x))g 0 (x)
dx
d
Z Z
⇒ F (g(x))dx = F 0 (g(x))g 0 (x)dx = F (g(x)) + C
dx
Z Z
0 0
F (g(x))g (x)dx = F (g(x)) + C = F (u) + C = f (u)du

But F 0 = f and u = g(x)

Z Z
0
∴ f (g(x))g (x)dx = f (u)du.

Theorem 47. (For definite integrals) If g 0 is continuous on [a, b] and f

is continuous on the range of u = g(x), then:
Z b Z g(b)
0
f (g(x))g (x)dx = f (u)du.
a g(a)

Proof. Let F be an anti-derivative of f, that is F 0 = f.

d
F (g(x)) = F 0 (g(x)g 0 (x)) = f (g(x))g 0 (x))
dx
156 Dr. Joseph K. Ansong
Math 223 : Calculus II J.K.A

Z b
⇒ f (g(x))g 0 (x)dx = F [g(x)]ba
a

Z b
⇒ f (g(x))g 0 (x)dx = F [g(b)] − F [g(a)] (10.16)
a

By applying the Evaluation theorem again:

Z g(b) g(b)
f (u)du = F (u) = F [g(b)] − F [g(a)] (10.17)
g(a) g(a)

From (10.16) and (10.17) we get

Z b Z g(b)
0
f (g(x))g (x)dx = f (u)du.
a g(a)

Example 90. 1) Evaluate the following integrals:

Z Z
3 4
(a) x cos(x + 2)dx (f) cos3 xdx

x √
Z Z
(b) dx (g) sec x sec x + tan xdx
x2 + 1
√
Z
1
Z
x x
(c) e 1 + e dx (h) dx
x(x + 1)
x
Z
sin 2x − cos 2x
Z
(d) √ dx (i) dx
1 − 4x 2 sin 2x + cos 2x
sin x + cos x
Z Z
(e) tan xdx (j) dx
e−x + sin x

2) Evaluate the following:

π
√
Z 4 Z
2
(a) 2x + 1dx (d) esin x cos xdx
0 0
Z e
ln x Z e4
dx
(b) dx (e) √
1 x
Z 3 e x ln x
3x2 − 1 4
r
(c) dx 1 1
Z
3 2
2 (x − x) (f) 2
1 + dx
1 x x

157 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Z
Solution. 1(a) I = x3 cos(x4 + 2)dx , Let u = x4 + 2

1
⇒ du = 4x3 dx ⇒ x3 dx = du
4
1 1 1
Z
I= cos(u)du = sin u + c = sin(x4 + 2) + c
4 4 4
x
Z
1(b) I = dx , Let u = x2 + 1
x2 +1
1
⇒ du = 2xdx ⇒ xdx = du
2
Z 1
du 1 1 1 1
Z
I= 2
= du = ln |u| + c = ln |x2 + 1| + c
u 2 u 2 2
√
Z
1(c) I = ex 1 + ex dx , Let u = 1 + ex ⇒ du = ex dx

√
Z Z
1
⇒I= udu = u 2 du

2 3 2 3
u 2 + c = (1 + ex ) 2 + c
3 3
x
Z
1(d) I = √ dx, Let u = 1 − 4x2
1 − 4x2
1
⇒ du = −8xdx ⇒ xdx = − du
8
 
1 1
Z
∴I= √ · − du
u 8
1 1
Z
1 1
=− u− 2 du = − (2)u 2 + c
8 8
1 1
⇒ − (1 − 4x2 ) 2 + c
4
sin x
Z Z
1(e) I = tan(x)dx = dx. Let u = cos x
cos x
⇒ du = − sin xdx ⇒ sin xdx = −du
1
Z
I = − du = − ln |u| + c = − ln | cos x| + c
u
1
= ln + c = ln | sec x| + c
cos x

158 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Z Z
3 2
1(f) I = cos xdx = cos x(cos x)dx = cos x(1 − sin2 x)dx
Let u = sin x ⇒ du = cos xdx
1 1
Z
∴, I = (1 − u2 )du = u − u3 + c = sin x − sin3 x + c
3 3

√
Z
1(g) I = sec x sec x + tan xdx
Let u = sec x + tan x

du = (sec x tan x + sec2 x)dx = sec x(tan x + sec x)dx = sec x(u)dx
du
=⇒ sec xdx =
u
du √
Z Z
1
=⇒ I = · u = u−1 · u 2 du
u
Z
1 1 1
∴ I = u− 2 du = 2u 2 + c = 2(sec x + tan x) 2 + c

1 1
Z
1(h) I = dx, Let
x(x + 1) x(x + 1)
A B
= + ⇒ 1 = A(x + 1) + Bx
x x+1
If x = 0, A = 1 and if x = −1, −B = 1 ⇒ B = −1
1 1 1
= −
x(x + 1) x x+1
1 1
Z Z
I= dx − dx = ln |x| − ln |x + 1| + c
x x+1
x
= ln +c
x+1
sin 2x − cos 2x
Z
1(i) I = dx
sin 2x + cos 2x
Let u = sin 2x + cos 2x ⇒ du = (2 cos 2x − 2 sin 2x)dx
1
= 2(cos 2x − sin 2x)dx ⇒ du = (cos 2x − sin 2x)dx
2
1
=⇒ − du = (sin 2x − cos 2x)dx
2
159 Dr. Joseph K. Ansong
Math 223 : Calculus II J.K.A

1
u 1 1
Z Z
2
I=− =− du
u 2 u
1 1
= − ln |u| + c = − ln | sin 2x + cos 2x| + c.
2 2
sin x + cos x
Z
1(j) I = dx
e−x + sin x

sin x + cos x (sin x + cos x)ex

Z Z
= 1 dx = dx
ex
+ sin x 1 + ex sin x

Let u = 1 + ex sin x, du = (ex sin x + ex cos x)dx ⇒ du = (sin x +

cos x)ex dx
du
Z
I= = ln |u| + c
u
∴, I = ln |1 + ex sin x| + c
4 √
Z
2(a) I = 2x + 1dx
0
1
Let u = 2x + 1 ⇒ du = 2dx ⇒ du = dx
2
If x = 0, u = 1; x = 4, u = 9
9 √ 1 9 1
 
1
Z Z
I= u du = u 2 du
1 2 2 1
   9
1 2 3
= u2
2 3 1

1 3
 3
 1 √
= 9 2 − 1 2 = ( 9)3 − 1
3 3
1 26
= (27 − 1) = .
3 3
Z e Z e  
ln x 1
2(b) dx = ln x ·
1 x 1 x
1
Let u = ln x ⇒ du = dx If x = 1, u = 0; x = e, u = 1
x
Z 1 1
1 1 1
I= udu = u2 = (1 − 0) =
0 2 0 2 2

160 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

3
3x2 − 1
Z
2(c) I = 3 2
dx
2 (x − x)
Let u = x3 − x ⇒ du = (3x2 − 1)dx
If 2, u = 8 − 2 = 6 and x = 3, u = 27 − 3 = 24
24 24
du
Z
I= = ln u
6 u 6
= ln(24) − ln(6) = ln(4 · 6) − ln 6 = ln 4 + ln 6 − ln 6 = ln 4
Z π
2
2(d) I = esin x cos xdx.
0
Let u = sin x ⇒ du = cos xdx
π
If x = 0, u = 0 and x = , u = 1
2
Z 1 1
I= eu du = eu = e − 1.
0 0

10.7.2 Integrals of Symmetric Functions

Suppose f is continuous on [−a, a].
Z a Z a
(a) If f is even, that is f (−x) = f (x), then f (x)dx = 2 f (x)dx
−a 0
Z a
(b) If f is odd, that is f (−x) = −f (x), then f (x)dx = 0.
−a

Example 91. Evaluate the following integrals:

Z 2
(1) (x6 + 1)dx
−2

1
tan x
Z
(2)
−1 1 + x2 + x 4
Solution. (1) Since f (x) = x6 + 1, f (−x) = f (x). Thus, f is even.
Z 2  2
6 1 7
=⇒ I = 2 (x + 1)dx = 2 x + x
0 7 0
   
128 142
=2 +2 =2
7 7
284
= .
7

161 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

1
tan x
Z
(2) I = dx
−1 1 + x2 + x4

f (−x) = −f (x) ⇒ f (x) is odd ⇒ I = 0.

Exercise. Evaluate the following integrals:

Z a √
(1) I = x x2 + a2 dx
−a
π
x2 sin x
Z
2
(2) dx
− π2 1 + x6

Integrals of the form

√
r
p(x) ax + b
Z Z Z
n n
p(x) ax + b dx, √ dx, p(x) dx
n
ax + b cx + d
Solution Approach: For the first two cases, make use of the substitutions
√
un = ax + b or u = n ax+b

and let r
ax + b
n n ax + b
u = or u =
cx + d cx + d
for the last case.

Example 92. Evaluate

√
Z
(a) x2 x − 2dx

x
Z
(b) √
3
dx
x+1
Z r
1+x
(c) dx
1−x
√
Z
Solution. (a) I = x2 x − 2dx
√
Let u = x − 2 ⇒ u2 = x − 2 =⇒ 2udu = dx, and x2 = (u2 + 2)2
Z Z
⇒ I = (u + 2) · u · 2udu = 2u2 (u2 + 2)2 du
2 2

162 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Z
= 2u2 (u4 + 4u2 + 4)du

2 8 8
I = u7 + u5 + u3 + c
7 5 3
2 7 8 5 8 3
∴ I = (x − 2) 2 + (x − 2) 2 + (x − 2) 2 + c.
7 5 3
x
Z
(b) I = √
3
dx
x+1
1
Let u = (x + 1) 3 ⇒ u3 = x + 1 ⇒ 3u2 du = dx, and x = u3 − 1
Z 3
u −1
Z
∴I= · 3u du = 3 u(u3 − 1)du
2
u
3 3
Z
= 3 (u4 − u)du = u5 − u2 + c
5 2
3 5 3 2
I = (x + 1) 3 − (x + 1) 3 + c
5 2
Z r
1+x
(c) I = dx
1−x
Easier approach:
r s
1+x (1 + x) (1 + x)
= ·
1−x 1 − x (1 + x)
s
(1 + x)2 1+x
= 2
=√
1−x 1 − x2
Let x = sin θ ⇒ dx = cos θdθ
1 + sin θ
Z Z
⇒I= · cos θdθ = (1 + sin θ)dθ
cos θ

= θ − cos θ + c
√
cos2 θ = 1 − sin2 θ ⇒ cos θ = 1 − x2
√
I = sin−1 x − 1 − x2 + c

Long approach:
1+x
Let u2 = ⇒ u2 − xu2 = 1 + x ⇒ u2 − 1 = x(u2 + 1)
1−x

163 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

u2 − 1 4u
⇒x= dx = du [quotient rule]
u2 + 1 (u2
+ 1)2
4u
Z
∴I = u· 2 du
(u + 1)2
4u2
Z
= du
(u2 + 1)2
Let u = tan x ⇒ du = sec2 αdα
4 tan2 α
Z
I= · sec2 αdα
(tanα +1)2

Note that tan2 α + 1 = sec2 α

4 tan2 α
Z
I= dα
sec2 α

4 sin2 α
Z
⇒I= · cos2 αdα
cos2 α
Z
= 4 sin2 αdα

1
Z Z
4 · (1 − cos 2α)dα = 2 (1 − cos 2α)dα
2
 
1
= 2 α − sin 2α + c = 2α − sin 2α + c
2
Using a right angle triangle, we have
u 1
tan α = u, sin u = √ , cos α = √
1 + u2 1 + u2
u 1
sin 2α = 2 sin α cos α = 2 √ ·√
1 + u2 1 + u2
2u
=
1 + u2
2u
I = 2 tan−1 u − +c
1 + u2
1+x 2
Note: 1 + u2 = 1 + =
1−x 1−x
2u (1 − x)
⇒ 2
= 2u ·
1+u 2

164 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

r s
1+x (1 + x)
(1 − x) · u = (1 − x) = (1 − x)2
1−x (1 − x)
p √
= (1 − x)(1 + x) = 1 − x2
!
√
r
1 + x
∴ I = 2 tan−1 − 1 − x2 + c
1−x

10.8 Trigonometric Integrals

10.8.1 Integrating Powers of sin x and cos x
sinn xdx and cosn xdx.
R R
Here are some guidelines for integrating

(1) If n is an odd positive integer then write

Z Z
sin xdx = sinn−1 x sin xdx.
n

Because (n − 1) is an even integer, we can employ the identity sin2 x =

1 − cos2 x to get an expression that is easier to integrate.

(2) If n is an odd positive integer then write

Z Z
cos xdx = cosn−1 x cos xdx.
n

Since (n − 1) is even, we can use the trigonometric identity cos2 x =

1 − sin2 x to get a form that can be easily integrated.

(3) If n is even, use the half-angle formula (and its variants) to simplify the
integrand sinn x or cosn x. Here are some half-angle formulae to refresh
your memory:
1 1
cos2 x = (1 + cos 2x), sin2 x = (1 − cos 2x),
2 2
1 1
cos2 2x = (1 + cos 4x), sin2 2x = (1 − cos 4x).
2 2
We will use some examples to illustrate the procedure for integrating powers
trigonometric functions.

Example 93. Evaluate:

165 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Z
(a) cos5 xdx
Z
(b) sin4 xdx
Z Z Z
5 4
Solution. (a) I = cos xdx = cos x cos xdx = (1 − sin2 x)2 cos xdx
Let u = sin x ⇒ du = cos xdx
2 1
Z Z
I = (1 − u ) du = (1 − 2u2 + u4 )du = u − u3 + u5 x + c
2 2
3 5
2 1
= sin x − sin3 x + sin5 x + c
3 5
Z Z
(b) I = sin4 xdx = [sin2 x]2 dx

1 1
Z
2
sin x = (1 − cos 2x) ⇒ I = (1 − cos 2x)2 dx
2 4
1
Z
I= (1 − 2 cos 2x + cos2 2x)dx
4
1 1
But cos2 x = (1 + cos 2x) =⇒ cos2 2x = (1 + cos 4x)
2 2
Z  
1 1 1
=⇒ I = 1 − 2 cos 2x + + cos 4x dx
4 2 2
 
1 1 1
= x − sin 2x + x + sin 4x + c
4 2 8
x 1 1 1
= − sin 2x + x + sin 4x + c
4 4 8 32
3 1 1
∴ I = x − sin 2x + sin 4x + c
8 4 32

10.8.2 Integrals of Products of Trigonometric Func-

tions
Integrals of the form sinm x cosn dx
R
Z
Here are some guidelines for integrating integrals of the form sinm x cosn x dx.

166 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

(1) If m is an odd integer, re-write the integral in the form

Z Z
sin x cos x dx = sinm−1 x cosn x sin xdx
m n

and use the identity sin2 x = 1 − cos2 x to express sinm−1 x in terms of

cos x. You can then make the substitution u = cos x to simplify the
resulting integral.

(2) If n is an odd integer, re-write the integral in the form

Z Z
sin x cos x dx = sinm x cosn−1 x cos xdx
m n

and use the identity cos2 x = 1 − sin2 x to express cosn−1 x in terms

of sin x. You can then make the substitution u = sin x to simplify the
resulting integral.

(3) If both n and m are even, then use the half-angle formulas for cos2 x
and sin2 x to reduce the exponents by one-half.

Example 94. Evaluate

Z
(1) cos3 x sin4 xdx
Z
Solution. (1) cos3 x sin4 xdx

Z Z
3 4
cos x sin xdx = cos2 x sin4 x cos xdx

Z
= (1 − sin2 x) sin4 x cos xdx

Let u = sin x, then du = cos xdx, and the integral becomes

Z Z Z
cos x sin xdx = (1 − u )u du = (u4 − u6 )du
3 4 2 4

1 1 7
= u5 − u +C
5 7
1 1 7
= sin5 x − sin x + C
5 7

167 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

R
Integrals of the form tanm x secn dx
tanm x secn xdx.
R
Here are some guidelines for integrating integrals of the form

(1) If m is an odd integer, re-write the integral in the form

Z Z
m n
tan x sec x dx = tanm−1 x secn−1 x sec x tan x dx

and use the trigonometric identity tan2 x = sec2 x−1 to express tanm−1 x
in terms of sec x. You can then make the substitution u = sec x to sim-
plify the resulting integral.

(2) If n is even, re-write the integral in the form

Z Z
m n
tan x sec x dx = tanm x secn−2 x sec2 x dx

and use the identity sec2 x = 1 + tan2 x to express secn−2 x in terms of

tan x. You can then make the substitution u = tan x to simplify the
resulting integral.

(3) If n is odd and m is even, there is no standard procedure. You may try
using integration by parts.

Example 95. Evaluate

Z Z
3 5
(1) tan x sec xdx (3) cos 5x cos 3xdx
Z
(2) tan2 x sec4 xdx

Z
Solution. (2) tan3 x sec5 xdx

Z Z
3 5
tan x sec xdx = tan2 x sec4 x(sec x tan x)dx

Z
= (sec2 x − 1) sec4 x(sec x tan x)dx

168 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Let u = sec x and du = sec x tan xdx, then we get

Z Z
tan x sec xdx = (u2 − 1)u4 du
3 5

Z
= (u6 − u4 )du
1 1
= u7 − u5 + C
7 5
1 1
= sec7 x − sec5 x + C.
7 5
Z
(3) tan2 x sec4 xdx
Z Z
2 4
tan x sec xdx = tan2 x sec2 x sec2 xdx
Z
= tan2 x(tan2 x + 1) sec2 xdx

Let u = tan x, then du = sec2 xdx, and we get

Z Z
tan x sec xdx = u2 (u2 + 1)du
2 4

Z
= (u4 + u2 )du
1 1
= u5 + u3 + C
5 3
1 1
= tan2 x + tan3 x + C
5 3
(4) Here, we apply the Factor Formula (or Sum and Product For-
mula) in trigonometry:
C −D
   
C +D
cos C + cos D = 2 cos cos
2 2
Thus, we may write
1
cos 5x cos 3x = (cos 8x + cos 2x)
2
such that
1
Z Z
cos 5x cos 3x = (cos 8x + cos 2x) dx
2
1 1
= sin 8x + sin 2x + C
16 4

169 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

10.8.3 Integrals of Hyperbolic Functions

Remark. Go back to the chapter on hyperbolic functions and their deriva-
tives, do a review of them, and return here. From our lectures on the defi-
nitions of hyperbolic functions and their derivatives, we obtain the integrals
in Table 10.1:
Z Z
cosh udu = sinh u + C sinh udu = cosh u + C
Z Z
sech2 udu = tanh u + C csch u coth udu = − csch u + C
Z Z
sech u tanh udu = − sech u + C csch2 udu = − coth u + C

Table 10.1: Integrals of some hyperbolic functions.

Z
Example 96. Find cosh2 3x sinh 3xdx.

1
Solution. Let u = 3x =⇒ du = 3dx and dx = du. Thus
3
1
Z Z
I = cosh2 3x sinh 3xdx = cosh2 u sinh udu
3
Let t = cosh u =⇒ dt = sinh udu. Thus,
1 1
Z
I= t2 dt = t3 + C
3 9
1 1
∴ I = cosh3 u + C = cosh3 (3x) + C.
9 9
Z
Example 97. Find sinh2 xdx.

Solution. Note that cosh 2x = 1 + 2 sinh2 x. Thus

1
Z Z
2
I = sinh xdx = (cosh 2x − 1)dx
2
1 1
∴I= sinh 2x − x + C.
4 2
Alternatively, we could change to exponentials:
2
e − e−x
 x
2 1
sinh x = = (e2x − 2 + e2x )
2 4

170 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

1 2x 1 1 1
Z
=⇒ I = (e − 2 + e2x )dx = e2x − x − e−2x + C.
4 8 2 8

1 e2x − e−2x
 
1
= − x+C
4 2 2

1 1
∴I= sinh 2x − x + C,
4 2
which is the same answer we found earlier.

10.8.4 Trigonometric & Hyperbolic Substitutions

The following table gives substitutions that √are useful for√eliminating rad-
icals from certain types of integrands like a2 + x2 or x2 − a2 . Either
trigonometric or hyperbolic substitutions can be used to evaluate such in-
tegrals. In making trigonometric substitutions, we assume that the angle
θ is in the range of the corresponding inverse trig function. For √ instance,
for the substitution x = a sin θ, we have −π/2 ≤ θ ≤ π/2. If a2 − x2
occurs in a denominator, we add the restriction |x| = 6 a, or, equivalently,
−π/2 < θ < π/2.

Expression in Integrand Substitution Substitution

√ (Trigonometric) (Hyperbolic)
2 2
√a − x x = a sin θ
2 2
√a + x x = a tan θ x = a sinh θ
x 2 − a2 x = a sec θ x = a cosh θ

For hyperbolic substitution, the basic hyperbolic and inverse hyperbolic

identities are relevant. For example

cosh2 x − sinh2 x = 1; sech2 x = 1 − tanh2 x

√
1 + csch2 x = coth2 x; sinh−1 x = ln(x + x2 + 1)

√
 
−1 −1 1 1+x
cosh x = ln(x + x2 − 1); tanh x = ln
2 1−x

1
Z
Example 98. Evaluate √ dx
4 + x2

171 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Solution Method 1: Trigonometric Substitution

Solution. Let x = 2 tan θ =⇒ dx = 2 sec2 θdθ. Thus
√ p p √
4 + x2 = 4 + 4 tan2 θ = 2 1 + tan2 θ = 2 sec2 θ = 2 sec θ

Therefore,

1 1
Z Z Z
√ dx = 2 sec2 θdθ = sec θdθ
4 + x2 2 sec θ

= ln | sec θ + tan θ| + C,
(recall the integral of sec θ from Calculus I). We need to put back the original
x variable. Now, tan θ = x/2 implies that (using a right-angle triangle as in
Figure 10.10)

Figure 10.10:
√
2 4 + x2
cos θ = √ =⇒ sec θ =
4 + x2 2
Hence,
√
1 4 + x2 x
Z
I= √ dx = ln + + C.
4 + x2 2 2

This may be written in the form:

√ √ 
I = ln 4+ x2 + x − ln 2 + C = ln 4+ x2 +x +D

where D√= − ln 2 + C and where the absolute value sign has been removed
because 4 + x2 + x > 0 for every x.

172 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Solution Method 2: Hyperbolic Substitution

Solution. Let x = 2 sinh θ =⇒ dx = 2 cosh θdθ. Thus
√ p p p
4 + x = 4 + 4 sinh θ = 2 1 + sinh θ = 2 cosh2 θ = 2 cosh θ
2 2 2

Therefore
1 1
Z Z
√ dx = · 2 cosh θdθ
4 + x2 2 cosh θ
Z
= dθ

1
Z
=⇒ √ dx = θ + C
4 + x2
−1 x
 
Now x = 2 sinh θ =⇒ θ = sinh . Using the identity sinh−1 x = ln(x +
√ 2
x2 + 1), we get
" #
x
r 
x 2 1 √ 
θ = ln + + 1 + C = ln + ln 4 + x2 + x + C
2 2 2
√ 
= ln 4 + x2 + x − ln 2 + C
Z
1 √ 
∴ √ dx = ln 4 + x2 + x + D.
4 + x2

Using the trig substitution x = a tan θ, we can derive the formula:

dx 1 −1 x
Z  
= tan + C.
x 2 + a2 a a

Exercise. (1) Using a hyperbolic substitution, show that

Z √
x√ 2 √
x2 + 4dx = x + 4 + 2 ln(x + x2 + 4) + C.
2

(2) Use trigonometric substitutions to show that

Z √ 2 √
x −9 x
(a) dx = x2 − 9 − sec−1 + C.
x 3
(1 − x2 )3/2 (1 − x2 )5/2
Z
(b) dx = − + C.
x6 5x5

173 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

10.9 Integrals of Rational Functions Using Par-

tial Fractions
Using partial fractions, rational functions can be written as a sum of sim-
pler fractions that we already know how to integrate. If the degree of the
numerator is greater than or equal to the degree of the denominator, we first
perform the long division to simplify the rational function. Some times the
denominator may contain irreducible quadratic factors or repeated reducible
quadratic factors, in either case, you need to apply all the techniques of par-
tial fractions to simplify the rational function before integration. In other
words, if
P (x)
f (x) =
Q(x)
we try to factor Q(x) as far as possible. This may some times require “com-
pleting the square” of a quadratic expression before making a substitution
to integrate the given expression.

Review of partial fraction techniques

The following examples give a review of some of the partial fraction tech-
niques needed to simplify integrals of rational functions.

(1) Q(x) is a product of linear factors, some of which are repeated.

3x3 − 18x2 + 29x − 4 A B C D

3
= + + 2
+
(x + 1)(x − 2) x + 1 x − 2 (x − 2) (x − 2)3

Multiplying both sides of the equation by the denominator on the left-

hand side, we can show that

A = 2, B = 1, C = −3, D = 2.

We can then integrate each partial fraction on the right-hand side.

(2) Q(x) contains repeated quadratic factors

5x3 − 3x2 + 7x − 3 Ax + B Cx + D
2 2
= 2 + 2
(x + 1) x +1 (x + 1)2

It can be shown that

A = 5, B = −3, C = 2, D = 0.

174 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

(3) Q(x) has irreducible quadratic factors. For example, find:

2x − 1
Z
2
dx.
x − 6x + 13
Completing the square, we can write

x2 − 6x + 13 = (x − 3)2 + 4.

Thus,
2x − 1 2x − 1
Z Z
2
dx = dx
x − 6x + 13 (x − 3)2 + 4
Let u = x − 3 =⇒ x = u + 3 and dx = du. Thus
2x − 1 2(u + 3) − 1 2u + 5
Z Z
2
dx = 2
du = du
x − 6x + 13 u +4 u2 + 4
2u 1
Z Z
= 2
du + 5 2
du
u +4 u +4
5 u
= ln(u2 + 4) + tan−1 + C
2 2
5 x−3
= ln(x2 − 6x + 13) + tan−1 + C.
2 2
x+5
Z
Example 99. Find dx
x2 + x − 2
Solution. We can write the rational fraction as
x+5 x+5 A B
= = +
x2 +x−2 (x − 1)(x + 2) x−1 x+2

=⇒ x + 5 = A(x + 2) + B(x − 1)
If x = 1, we get 6 = 3A =⇒ A = 2
If x = −2, we get 3 = −3B =⇒ B = −1. Thus, we can write
x+5 2 1
= − .
x2 +x−2 x−1 x+2
Hence Z  
x+5 2 1
Z
2
dx = − dx
x +x−2 x−1 x+2
= 2 ln |x − 1| − ln |x + 2| + C.

175 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

x3 + x
Z
Example 100. Find dx.
x−1
Solution. Using long division, it can be shown that

x3 + x 2
= x2 + x + 2 +
x−1 x−1
Thus,
x3 + x
Z  
2
Z
dx = x2 + x + 2 + dx
x−1 x−1
x3 x2
= + + 2x + 2 ln |x − 1| + C.
3 2

10.10 The Reduction Formulae

Integration by parts can be used to derive reduction formulas for evalu-
ating certain integrals. The formulas enable us to express some integrals in
terms of integrals whose integrands involve lower powers. Recall the inte-
gration by part formulas:

Z Z
udv = uv − vdu.
Z b Z b
udv = [uv]ba − vdu.
a a

Z
Example 101. Find a reduction formula for sinn xdx, where n ≥ 2 is an
integer.

Solution. Let Z Z
n
I= sin xdx = sinn−1 x sin xdx

Integrating by parts, we let

u = sinn−1 x and dv = sin xdx

=⇒ du = (n − 1) sinn−2 x cos xdx and v = − cos x

Z Z
n
=⇒ sin xdx = uv − vdu

176 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Z
n−1
= − sin x cos x + (n − 1) sinn−2 x cos2 xdx

But cos2 x = 1 − sin2 x, so that

Z Z Z
n n−1 n−2
sin xdx = − sin x cos x + (n − 1) sin xdx − (n − 1) sinn xdx

Moving the last term on the right to the left-hand side gives
Z Z
n n−1
n sin dx = − sin x cos x + (n − 1) sinn−2 xdx

or

1 n−1
Z Z
sin dx = − sinn−1 x cos x +
n
sinn−2 xdx (10.18)
n n

Z
Example 102. Use the reduction formula in (10.18) to find sin4 dx.

Solution. Here, n = 4. So we have

1 3 3
Z Z
4
sin dx = − sin x cos x + sin2 xdx (10.19)
4 4
Z
Applying the reduction formula again to the integral sin2 xdx, we get

1 1
Z Z
2
sin xdx = − sin x cos x + dx
2 2
1 1
= − sin x cos x + x + C1
2 2
Substituting into equation (10.19), we obtain
 
1 3 3 1 1
Z
4
sin xdx = − sin x cos x + − sin x cos x + x + C1
4 4 2 2
1 3 3
Z
∴ sin4 xdx = − sin3 x cos x − sin x cos x + x + C
4 8 8
where C = 34 C1 .

177 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

10.11 Improper Integrals

10.11.1 Infinite Intervals of Integration
Improper Integrals are integrals that have infinite intervals of integration
or unbounded integrands. The following definition provides a way of com-
puting integrals over infinite intervals as integrals over finite intervals.

Definition 31. (1) If f is continuous on [a, ∞], then

Z ∞ Z b
f (x)dx = lim f (x)dx (10.20)
a b→∞ a

provided that the limit exists.

(2) If f is continuous on [−∞, b], then

Z b Z b
f (x)dx = lim f (x)dx (10.21)
−∞ a→−∞ a

provided that the limit exists. Each improper integral in (10.20)

and (10.21) is convergent if the limit exists anddivergent if the
limit does not exist.

(3) If f is continuous on (−∞, ∞), then

Z ∞ Z c Z ∞
f (x)dx = f (x)dx + f (x)dx (10.22)
−∞ −∞ c

where c is any real number, provided that both improper integrals

on the right-hand side exist. The improper integral is convergent if
both improper integrals on the right are convergent, and divergent
if one or both are divergent.

∞
1
Z
Example 103. Find dx
1 x2
Solution. ∞ Z t
1 1
Z
I= 2
dx = lim dx
1 x t→∞ 1 x2
 t  
1 1
∴ I = lim − = lim − + 1 = 1.
t→∞ x 1 t→∞ t

178 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

The improper integral is convergent in this case.

Example 104. Determine if the following integral is convergent or divergent.
If it is convergent, find its value.
Z ∞ Z ∞
1 2
(a) dx (c) xe−x dx
1 x −∞
Z 0 Z ∞
1
(b) √ dx (d) sin xdx
−∞ 3−x −2
Z ∞
1
Solution. (a) I = dx
1 x
Z t
1 t
I = lim dx = lim ln(x)
t→∞ 1 x t→∞ 1

∴ I = lim [ln t − ln 1] = ∞
t→∞

Hence, the integral is divergent.

Z 0
1
(b) I = √ dx
−∞ 3−x
Z 0 Z 0
1
I = lim √ dx = lim (3 − x)−1/2 dx
t→−∞ t 3−x t→−∞ t

√ 0 √ √
= lim −2 3 − x = lim (−2 3 + 2 3 − t)
t→−∞ t t→−∞
√
∴ I = −2 3 + ∞ = ∞
Thus, the limit is infinite and hence the integral is divergent.
Z ∞
2
(c) I = xe−x dx
−∞
Z 0 Z ∞
−x2 2
I= xe dx + xe−x dx
−∞ 0

Now, let Z 0 Z 0
−x2 2
I1 = xe dx = lim xe−x dx
−∞ t→−∞ t
The integral is easily integrated by using substitution to get
 
1 −x2 0
= lim − e
t→−∞ 2 t

179 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

 
1 1 −t2
= lim − + e
t→−∞ 2 2
1
=⇒ I1 = − .
2
Now let Z ∞ Z t
−x2 2
I2 = xe dx = lim xe−x dx
0y t→∞ 0

The integral is easily integrated by using substitution to get

 
1 −x2 t
= lim − e
t→∞ 2 0
 
1 −t2 1
= lim − e +
t→∞ 2 2
1
=⇒ I2 =
2
Thus,
1 1
I = I1 + I2 = − + = 0.
2 2
Thus, the improper integral is convergent and converges to 0.
Z ∞
(d) I = sin xdx
−2
Z t
I = lim sin xdx = lim (− cos x)|t−2
t→∞ −2 t→∞

∴ I = lim (cos 2 − cos t)

t→∞
Thus, the limit does not exist and so the integral is divergent.
Exercise. (1) Determine whether the integral converges or diverges, and
if it converges, find its value.
Z ∞ Z ∞
1 1
(a) dx (b) dx
2 (x − 1)2 2 x−1
Z ∞
1
(2) Evaluate 2
dx
−∞ 1 + x

10.11.2 Improper Integrals with Discontinuous Inte-

grands

180 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Definition 32. (a) If f is continuous on [a, b) and discontinuous at b,

the Z b Z t
f (x)dx = lim− f (x)dx,
a t→b a
provided the limit exists.

(b) If f is continuous on (a, b] and discontinuous at a, then

Z b Z b
f (x)dx = lim+ f (x)dx,
a t→a t

provided the limit exists.

(c) If f has a discontinuity at a number c in the open interval (a, b)

but is continuous elsewhere on [a, b], then
Z b Z c Z b
I= f (x)dx = f (x)dx + f (x)dx,
a a c

provided both of the improper integrals on the right converge. If

both converge, then the value of the improper integral I is the sum
of the two values.

3
1
Z
Example 105. Evaluate I = √ dx.
0 3−x
Solution. Note that the integrand has an infinite discontinuity at x = 3.
Thus,
 √
Z t
1 t
I = lim− √ dx = lim− −2 3 − x 0
t→3 0 3−x t→3
√ √
= lim− −2 3 − t + 2 3
t→3
√ √
= 0 + 2 3 = 2 3.
Z 3
1
Example 106. Evaluate I = dx if possible.
0 x−1

Solution. Observe that the integrand is not continuos at x = 1, which occurs

in the middle of the interval [0, 3]. So we integrate as
Z 1 Z 3
1 1
I= dx + dx
0 x−1 1 x−1

181 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Now,
1 t
1 1
Z Z
I1 = dx = lim− dx = lim− ln |x − 1|t0
0 x−1 t→1 0 x−1 t→1

= lim− (ln |t − 1| − ln | − 1|)

t→1

= lim− ln(1 − t) = −∞
t→1

Thus, I1 is divergent. ZThus the whole integral I is also divergent and we

3
1
don’t need to evaluate dx.
1 x−1

Remark. WARNING!!!
If we had not noticed the discontinuity at x = 1 in the previous example and
evaluated the integral as an ordinary integral, we would have gotten
3
1
Z
dx = ln |x − 1|30 = ln 2 − ln 1 = ln 2
0 x−1

which is wrong! Therefore, it is important to always check whether a given

integral is an ordinary integral or an improper integral, by looking at the
integrand, f , on the interval [a, b].

10.12 Applications of Integration

10.12.1 Area between curves

y=f (x)
y
1. Recall that the area bounded by
the curve y = f (x) and y = 0 in the
interval x ∈ [a, b] (figure 10.11) is A
given by
a b x
Z b
A= f (x)dx.
a Figure 10.11: Area under a curve

2. Now consider a region S which lies between the two curves y = f (x)
and y = g(x) in the interval [a, b] such that f (x) ≥ g(x) for all x in [a, b]. To
find the area between the curves:

182 Dr. Joseph K. Ansong

 Math 223 : Calculus II J.K.A

y
y=f (x)
y
y=f (x)

f (x∗i )

S
a −g(x∗i ) b x
a b x x∗i
y=g(x) y=g(x)

Figure 10.13: Approximating rectan-

Figure 10.12: Area between two curves gle

Step 1 : Divide S into n rectangular strips of equal width ∆x. The height of
the ith strip is f (x∗i ) − g(x∗i )

Step 2 : The area of the ith strip is Ai = [f (x∗i ) − g(x∗i )]∆x. Summing all the
strips we get the Riemann sum, Rn .
n
X
Rn = [f (x∗i ) − g(x∗i )]∆x.
i=1

Step 3 : Find the limit as n → ∞ since the approximation gets better with
more strips:
n
X
A = lim [f (x∗i ) − g(x∗i )]∆x
n→∞
i=1

Step 4 : Finally we get the area between the curves

Z b
A= [f (x) − g(x)]dx (10.23)
a

NOTE:

(a) In general, it is important to use a sketch to identify the top and bottom
curves and to also find an approximating rectangle. You may also have
to determine the interval [a, b].

(b) If f (x) ≥ g(x) in some parts of the region and g(x) ≥ f (x) for other
parts, then split the region into several parts (see figure 10.14):

183 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

y
y=f (x)

A1

A = A1 + A2 A2
Z c Z b y=g(x)
= [f (x) − g(x)]dx + [g(x) − f (x)]dx x
a c b
a c
Z b
A= |f (x) − g(x)|dx Figure 10.14: Divided region
a

(c) If x = f (y) and x = g(y) for y in the interval [c, d] where f and g are
continuous and f (y) ≥ g(y) for c ≤ y ≤ d then

Z d
A= [f (y) − g(y)]dy (10.24)
c

Example 107. Find the area bounded by the curves y = x2 and y = x.

Solution. Let f (x) = x, g(x) = x2 . We need to find the intersection of

the two curves by equating them:

x2 = x =⇒ x2 − x = 0 =⇒ x(x − 1) = 0,

so they intersect at x = 0 and x = 1. Thus, the area between the curves (see
Figure 10.15a) is given by
Z 1
A = [f (x) − g(x)]dx
0
Z 1  1
2 1 2 1 3
= (x − x )dx = x − x
0 2 3 0
 
1 1 1
= − −0= .
2 3 6

Alternatively, we could integrate with respect to y (see Figure 10.15b). For

√
y = x2 , we solve for x to get x = ± y. Here, we need only the right-hand
√
branch of the curve so we choose f (y) = x = y. From y = x, we define
g(y) = y. When x = 0 we have y = 0 and when x = 1 we get y = 1; so the

184 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

interval of integration remain the same. Thus,

Z 1
A = [f (y) − g(y)]dy
0
 1
√ 2 3/2 1 1/2
= ( y − y)dy = y − y
3 2 0
 
2 1 1
= − −0= .
3 2 6

Figure 10.15:

Example 108. Find the area of the region enclosed by f (x) = sin x, g(x) =
cos x, x = π2 and the y-axis.

Solution. The graph of the required region is displayed in Figure 10.16.

The intersection point is given by

sin x = cos x =⇒ tan x = 1

π
∴x=
4
Thus, the required area, A, is given by the sum of the area A1 and that of
A2 such that A = A1 + A2 .
Z π/4 Z π/2
A1 = (cos x − sin x)dx, A2 = (sin x − cos x)dx
0 π/4

185 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Figure 10.16:

π/4 π/2
=⇒ A = (sin x + cos x) + (− cos x − sin x)
0 π/4
" √ √ ! # " √ √ !#
2 2 2 2
= + − (0 + 1) + (0 − 1) − − −
2 2 2 2
√ √
= 2−1+ 2−1
√
∴A=2 2−2
Example 109. Find the area bounded by the curves

y = x + 1, y = 9 − x2 ,

and the lines x = −1 and x = 2.

Solution. Here, we are given the interval [a, b] = [−1, 2]. By sketching the
functions we see that the upper curve is f (x) = 9 − x2 and the lower curve is
g(x) = x + 1 (see Figure 10.17). So the area between the curves is given by
Z 2
(9 − x2 ) − (x + 1) dx
 
A =
−1
2  2
1 2 1 3
Z
2
= (8 − x − x )dx = 8x − x − x
−1 2 3 −1
   
8 1 1
= 16 − 2 − − −8 − +
3 2 3
34 49 117
= + = .
3 6 6

186 Dr. Joseph K. Ansong

 Math 223 : Calculus II J.K.A

Figure 10.17: The blue shaded region is the required area, and the smaller
brown region is a representative rectangle.

10.12.2 Volumes
Cylinders
The volume, V , of a cylinder of height, h, is given by

V = (Area of cross-section) × height = Ah

Solids of revolution
Let S be a general solid which lies in the interval [a, b] and let A(x) be the
area of a cross-section perpendicular to the x−axis at the point x (see Figures
10.19 and 10.20). To find the volume of S:

Step 1 : Divide S into n “slabs” of equal width ∆x

Step 2 : At the point x∗i ∈ [xi−1 , xi ], the area of the slab is A(x∗i ) and its height
is ∆x. Thus, the volume of the slab Si is V (Si ) ≈ A(x∗i )∆x. Summing

187 Dr. Joseph K. Ansong

 Math 223 : Calculus II J.K.A

Figure 10.18: circular cylinder

y
∆x
y

S
S A(x)

a xi−1 xi b x
a x∗i
x b x
Figure 10.19: Figure 10.20:

the volume of all the slabs we get the Riemann sum:

n
X
V ≈ A(x∗i )∆x
i=1

Step 3 Find the limit as n → ∞ since the approximation gets better with more
slabs:
n
X
V = lim A(x∗i )∆x
n→∞
i=1

Step 4 Finally we get the volume of S:

Z b
V = A(x)dx (10.25)
a

NOTE:

188 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

(a) The volume may also be calculated in terms of y:

Z d
V = A(y)dy
c

(b) If the cross-section is a disk, then we find A as

A = π(radius)2

(c) If the cross-section is a washer, we find A as

A = π(outer radius)2 − π(inner radius)2
2 2
= π(rout − rin )

Examples:
Find the volume of the solid obtained by rotating the region bounded by the
given curves about the specified line
(a) y = x2 , 0 ≤ x ≤ 1, y = 0; about x-axis
(b) y = ex , y = 0, x = 0, x = 1; about x-axis
(c) y = x2 , 0 ≤ x ≤ 2, y = 4, x = 0; about y-axis
(d) y = x2 , y 2 = x; about x-axis
(e) y = x4 , y = 1; about y = 2
(f ) y = x2 , x = y 2 ; about x = −1

Volumes of other solids

Examples:
1. Find the volume of a pyramid whose base is a square with side L and
whose height is h.
2. Find the volume of the frustrum with base radius R and top radius r.

10.12.3 Volumes by Cylindrical Shells

Consider a cylindrical shell of inner radius r1 , outer radius r2 and height h.
Let V1 = πr12 h and V2 = πr22 h be the volume of the inner and outer cylinders
respectively. Then the volume of the shell is
V = V2 − V1 = πr22 h − πr12 h
= π(r22 − r12 )h
= π(r2 + r1 )(r2 − r1 )h
(r2 + r1 )
= 2π (r2 − r1 )h.
2

189 Dr. Joseph K. Ansong

 Math 223 : Calculus II J.K.A

Let r = r2 +r
2
1
(the average radius of the cylinder)
and ∆r = r2 − r1 (the thickness of the shell). Then

V = 2πrh∆r. (10.26)
That is,
V = [circumference(2πr)] × [height(h)] × [thickness(∆r)].
In general, let S be the solid obtained by rotating about the y-axis the region
bounded by y = f (x), y = 0 and x ∈ [a, b].

To obtain the volume V :

Step 1 : Divide [a,b] into n rectangular strips in [xi−1 , xi ] of equal width ∆x
and let x̄i be the midpoint of the ith strip.
Step 2 : The height of the ith strip is f (x̄i ), radius of the shell is x̄i and the
thickness of the shell is ∆x. Thus, the volume of the ith shell is
V = (2π x̄i )f (x̄i )∆x.
Summing the volume of all shells in [a, b], we get
n
X
V ≈ (2π x̄i )f (x̄i )∆x.
i=1

Step 3 : Find the limit as n → ∞ since the approximation gets better with
more strips:
n
X
V = lim (2π x̄i )f (x̄i )∆x
n→∞
i=1

Step 4 : Finally we get the volume of the shell as:

Z b
V = 2πxf (x)dx (10.27)
a

NOTE:
(a) In equation (10.27),
2πx := Circumference
f (x) := Height
dx := Thickness

190 Dr. Joseph K. Ansong

 Math 223 : Calculus II J.K.A

(b) It’s important to always make a sketch to identify the radius and height
of a shell.

Example 110. 1. Use both slicing and the method of cylindrical shells
to obtain the volume of the solid obtained by rotating about the y-axis
the region bounded by
√
y = x, and y = x2 .

2. Use the method of cylindrical shells to find the volume of the solid
obtained by rotating the region bounded by the given curves about the
specified axis:

(a) x = 1 + y 2 , x = 0, y = 1, y = 2; x-axis
√
(b) x = y, x = 0, y = 1; x-axis
(c) y = x2 , y = 0, x = −2, x = −1; y-axis
(d) y = x2 , y = 0, x = 1, x = 2; about x = 1

10.12.4 Arc Length

Our aim here is to find the length of any general curve which is defined by
a continuous function in a given interval. We do that by approximating the
curve using a series of polygons. Let C be such a curve and y = f (x) be the
continuous function in the interval a ≤ x ≤ b. To obtain the length, L, of
the curve, we

Step 1: divide [a, b] into n subintervals with end points xo , x1 , . . . xn and equal
width ∆x.
If yi = f (xi ), then the point Pi (xi , yi ) lies on C and the polygon with
vertices P0 , P1 , . . . Pn is an approximation to C.

Step 2: The length of the arc in [xi−1 , xi ] is approximately given by

p
Li = |Pi−1 Pi | = (xi − xi−1 )2 + (yi − yi−1 )2
p
= (∆x)2 + (∆yi )2

Summing all polygons, we get

n
X p
L= (∆x)2 + (∆yi )2
i=1

191 Dr. Joseph K. Ansong

 Math 223 : Calculus II J.K.A

Step 3: Find the limit as n → ∞ since the approximation gets better with more
polygons:
n
X p
L = lim (∆x)2 + (∆yi )2
n→∞
i=1

By the Mean Value Theorem, there is a number x∗i [xi−1 , xi ] such that

yi − yi−1 = f 0 (x∗i )(xi − xi−1 )

⇒ ∆yi = f 0 (x∗i )∆x

n
X p
⇒ L = lim (∆x)2 + (f 0 (x∗i )∆x)2
n→∞
i=1
n
X p
= lim 1 + [f 0 (x∗i )]2 ∆x
n→∞
i=1

Step 4: Finally we get the length of the curve:

s 2
Z bp b 
dy
Z
L= 0 2
1 + [f (x)] dx = 1+ dx (10.28)
a a dx

NOTE:
If the curve has an equation of the form x = g(y), c ≤ y ≤ d and g 0 (y) is
continuous, then
s 2
d d 
dx
Z p Z
L= 0 2
1 + [g (y)] dy = 1+ dy (10.29)
c c dy

Example 111. Find the Length of the arc in the given interval:

(a) y = 1 + 6x3/2 , 0 ≤ x ≤ 1.

1√
(b) x = y(y − 3), 1 ≤ y ≤ 9.
3

(c) y 2 = 4(x + 4)3 , 0 ≤ x ≤ 2, y > 0.

192 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

The arc length function

The distance, S(x), along the curve C from the initial point [a, f (a)] to any
point [x, f (x)] on the curve is the arc length function and is given by
Z x p
S(x) = 1 + [f 0 (t)]2 dt. (10.30)
a

Differentiating the equation gives

s  2
ds dy
= 1+
dx dx
s  2
dy
=⇒ ds = 1+ dx
dx
=⇒ (ds)2 = (dx)2 + (dy)2 .

Example 112. Find the arc length function for the curve y = 2x3/2 with
the starting point Po (1, 2).

10.12.5 Area of a Surface of Revolution

This is the surface obtained by rotating a curve about a line. Let’s start
from finding the area of the surface of a simple object and progress to the
complex case of a general surface area.

1. The lateral surface area of a cone of base radius r and slant height h is
given by

A = πrl

2. Consider a frustum of a cone with slant height l and top radius r1 and
base radius r2 . Then the area is given by

A = πr2 (l1 + l2 ) − πr1 l1 = π[(r2 − r1 )l1 + r2 l]

Using similar triangles we get

l1 l1 + l
=
r1 r2
⇒ r2 l1 = r1 l1 + r1 l
⇒ (r2 − r1 )l1 = r1 l

193 Dr. Joseph K. Ansong

 Math 223 : Calculus II J.K.A

Thus,
(r1 + r2 )
A = π[r1 l + r2 l] = 2π l
2
⇒ A = 2πrl (10.31)

where r = (r1 + r2 )/2 is the average radius.

3. In general, consider the surface obtained by rotating the curve y = f (x),
a ≤ x ≤ b about the x-axis, where f has continuous derivatives.
To obtain the area of the surface of revolution:
Step 1: divide [a, b] into n subintervals with end points x0 , x1 , . . . xn and equal
width ∆x.
If yi = f (xi ), then the point Pi (xi , yi ) lies on the curve.
Step 2: The area of the surface obtained by rotating the polygon Pi−1 Pi about
the x-axis is given by (using (10.31))
yi−1 + yi
Si = 2π |Pi−1 Pi |.
2
If x∗i [xi−1 , xi ], then the length of the arc is
p
|Pi−1 Pi | = 1 + [f 0 (x∗i )]2 ∆x.

Thus,
yi−1 + yi p
Si = 2π 1 + [f 0 (x∗i )]2 ∆x
2
Now, if ∆x is small, then

yi = f (xi ) ≈ f (x∗i ),
yi−1 = f (xi ) ≈ f (x∗i ),
yi−1 + yi 2f (x∗i )
⇒ ≈ = f (x∗i ).
2 2
Thus,
p
Si ≈ 2πf (x∗i ) 1 + [f 0 (x∗i )]2 ∆x

Summing over all polygons, we get

n
X p
S≈ 2πf (x∗i ) 1 + [f 0 (x∗i )]2 ∆x.
i=1

194 Dr. Joseph K. Ansong

 Math 223 : Calculus II J.K.A

Step 3: Finding the limit as n → ∞, we get

n
X p
S = lim 2πf (x∗i ) 1 + [f 0 (x∗i )]2 ∆x
n→∞
i=1

Step 4: Finally we get the area of the surface of revolution

s 2
b Z b 
dy
Z p
S= 2πf (x) 1 + [f 0 (x)]2 dx = 2πy 1+ dx (10.32)
a a dx

NOTE:
If the curve is given by x = g(y), c ≤ y ≤ d, then
s 2
d 
dx
Z
S= 2πx 1 + dy (10.33)
c dy

SUMMARY:
Equations (10.32) and (10.33) may also be written respectively as:
Z Z
S = 2πyds and S = 2πxds,

where
s  2 s  2
dy dx
ds = 1+ dx or ds = 1+ dy
dx dy

Example 113. 1. Find the area of the surface obtained by rotating the
curve about the x-axis
√
y = x, 4 ≤ x ≤ 9

2. Find the area of the surface obtained by rotating about the y-axis (use
2 formulas)
√
y = 3 x, 1 ≤ y ≤ 2

195 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

196 Dr. Joseph K. Ansong

Chapter 11

Differential Equations

A differential equation is an equation that contains an unknown function and

some of its derivatives.
E.g.

dy
= −kxy or y 0 = −kxy (11.1)
dx

where k is a constant.
The order of a differential equation is the order of the highest derivative
that occurs in the equation. E.g.:

y 0 = −kxy [First order]

y 00 + kx = 0 [Second order]
000 00
5y − ky + 2y = 0 [Third order]

We will consider only first order differential equations.

A function f is called a solution of a differential equation if the equation is
satisfied when y = f (x) and its derivatives are substituted into the equation.
For instance, y = f (x) is a solution of (11.1) if

f 0 (x) = −kxf (x)

for x in some interval.

1
Example: Show that y = x − x
is a solution to

xy 0 + y = 2x.

197
Math 223 : Calculus II J.K.A

Solution:
1
y =x−
x
1
⇒ y0 = 1 + 2
x 
0 1 1
⇒ xy + y = x 1 + 2 + x −
x x
1 1
=x+ +x−
x x
= 2x.

Let’s consider two types of differential equations and how to solve them.

11.1 Separable equations

These are equations of the form

dy
= g(x)f (x)
dx
Solution: Separate the variables and integrate the resulting equation if
possible:

dy
= g(x)f (x)
Z dx Z
dy
⇒ = g(x)dx
f (y)

Examples:
1. Solve

dy e2x
= 3
dx 4y

2. Find the solution of the differential equation that satisfies the given initial
condition:
dy
(a) = y 2 + 1, y(1) = 0
dx
dy y cos x
(b) = , y(0) = 1
dx 1 + y2

198 Dr. Joseph K. Ansong

Math 223 : Calculus II J.K.A

Mixing problems
Mixing problems are an application of first order differential equations. If
y(t) is the amount of substance in a tank at time t, then y 0 (t) is the rate
at which the substance is being added minus the rate at which it is being
removed.
Example:
A tank contains 20 Kg of salt dissolved in 5000L of water. Brine that contains
0.03 Kg of salt per liter of water enters the tank at a rate of 25L/min. The
solution is kept thoroughly mixed and drains from the tank at the same rate.
How much salt remains in the tank after half an hour?
Solution:
Let y(t) be the amount of salt (in kilograms) after t minutes.
y(0) = 20, y(30) =?
The rate of change of the amount of salt is:

dy
= (rate in)-(rate out)
dx   
kg L kg
rate in = 0.03 25 = 0.75 .
L min min

The tank always contains 5000 L of liquid so the concentration at time t is

y(t)/5000 [Kg/L]. Thus
  
y(t) kg L y(t) kg
rate out = 25 = .
5000 L min 200 min

Thus,

dy y(t) 150 − y(t)

= 0.75 − =
dt 200 200
Solving the equation we get

|150 − y| = 130e−t/200

Since y(t) is a continuous function and y(0) = 20 and the right hand side is
never 0, we deduce that 150 − y(t) is always positive. Thus we get

y(t) = 150 − 130e−t/200

⇒ y(30) = 150 − 130e−30/200 ≈ 38.1kg

199 Dr. Joseph K. Ansong

 Math 223 : Calculus II J.K.A

11.2 Linear Equations

A first order linear differential equation is an equation of the form
dy
= P (x)y = Q(x) (11.2)
dx
where P and Q are continuous functions on a given interval.
Solution to equation (11.2):

Step 1: Find the integrating factor

R
P (x)dx
µ=e (11.3)

Step 2: Multiply equation (11.3) by (11.2) to get

P (x)dx dy
R R R
P (x)dx P (x)dx
e + P (x)e y=e Q(x)
dx
dy
⇒µ + µP (x)y = µQ(x)
| dx {z }
d
dx
[µy]

d
⇒ [µy] = µQ(x) (11.4)
dx

Step 3: Integrate equation (11.4) to get the solution

Z 
1
y= µQ(x)dx + C (11.5)
µ
where C is a constant.
Examples:
1. Which of the following differential equations is linear?

(a) y 0 + ex y = x2 y 2 (b) y + sin x = x3 y 0

2. Solve the differential equation

(a) y 0 − 2y = 2ex (b) xy 0 − 2y = x2

3. Solve the initial-value problem (IVP):

y 0 = x + y, y(0) = 2

200 Dr. Joseph K. Ansong