UNIT 17

Structure
Page Nos.
17.1 Introduction 5
Objectives
17.2 Integrability 6
Upper and Lower Sums
Upper and Lower Integrals
The De nite Integral
17.3 Fundamental Theorem of Calculus 21
17.4 Standard Integrals 27
17.5 Summary 32
17.6 Solutions/Answers 32

17.1 INTRODUCTION
As we have seen in the introduction to this block, from ancient times,
measuring areas and volumes was essential for construction purposes. We
have also seen that it was necessary to measure the lands for the purpose of
levying taxes. Formulae for the areas of plane gures like squares and
rectangles and squares were mentioned in ancient manuscripts, but no
justi cations were given.
In the third century B.C., the Greek mathematician Archimedes found a formula
for the area of a circle by what is now known as method of exhaustion. His
solution contained the seeds of the present day integral calculus. But it was
only later, in the seventeenth century, that Newton and Leibniz were able to
generalise Archimedes' method. They also established the link between
differential and integral calculus.
The French mathematician A. L. Cauchy gave a de nition of the integral under
which only the continuous functions are integrable. Later, the German
mathematician Riemann gave a de nition of the integral under which a larger
class of functions are integrable. A. L. Cauchy
1789-1857
Other mathematicians like Stieltjes and Lebesgue have de ned integrals which
improve upon the Riemann Integral. However, the most popular way of 5
Block 5 Integration
..........................................................................................................
introducing integration is through the Riemann Integral and we will follow this
practice. In our discussion, we will use an equivalent de nition of the integral
due to Darboux, a French Mathematician.

In Sec. 17.2, we discuss upper and lower sums introduced by Darboux. This
lead us to a de nition of the de nite integral and some of its basic properties. In
Sec. 17.3, we will state the Fundamental Theorem of Integral Calculus and
explain how to use it evaluate some de nite integrals. In Sec. 17.3 we will also
discuss the inde nite integral. In Sec. 17.4, we will nd the inde nite integrals
of some common functions like polynomial, rational functions, trigonometric
functions and functions like exponential, logarithm and inverse trigonometric
functions.
G.B. Riemann
1826-1866 Objectives
After studying this unit, you should be able to:
• de ne and calculate the lower and upper sums of some simple functions
de ned on [a, b] , corresponding to a partition of [a, b];

• de ne the upper and lower integrals of a function;

• de ne the de nite integral of a given function and check whether a given

function is integrable or not in simple cases;

• state and apply the Fundamental Theorem of Calculus;

• de ne the primitive or antiderivative of a function;

J. G. Darboux
• use the Fundamental Theorem to calculate the de nite integral of an
1842-1917
integrable function; and

• nd the inde nite integrals of some common functions.

17.2 INTEGRABILITY

In the introduction, we mentioned that Archimedes found the area of a circle

using the method of exhaustion. Let us brie y discuss this method with the
help of an example. Suppose that we want to nd the area of a semi-circle. We
Our discussion is only for bound the area from the above and below by regular polygons. See Fig. 1.
the purpose of
motivation. We don't
claim that we are being
faithful to the history of
Mathematics.

(a) Upper bound for the area of a semi- (b) Lower bound for the area of a
circle. semicircle.

Fig. 1: Bounds for the area of a semicircle.

In Fig. 1a, the outer polygon, coloured red, gives an upper bound for the area of
the semicircle. In Fig. 1b, the outer polygon, coloured blue gives a lower bound
6 for the area of the semicircle.
Unit 17 Introduction to Integration
..........................................................................................................

In Fig. 2, the yellow coloured region gives the difference between the area of
the semicircle and the area of the inner polygon. The red coloured region gives
the difference between the area of the outer polygon and the area of the
semicircle. You can see in the gure that as we increase the number of sides of
the inner and outer polygons, the red and yellow coloured regions become
smaller and smaller. In other words, the difference between the areas become
smaller and the areas of the outer and inner polygons become closer and
closer to the area of the semicircle. In the limiting case, they give the area of
the circle.

Fig. 2: Area of a semi-circle by method of exhaustion.

1 2

Fig. 3: Area under the curve f(x) = 4x3 12x2 + 9x + 1.

On the other hand suppose that we want to nd the area under the curve
f(x) = 4x3 12x2 + 9x + 1 from x = 0 to x = 2. You learnt how to nd the maxima
and minima of functions in the third block. You can use that knowledge to check
the following:

1. In the interval [0, 2], the function f(x) is increasing between x = 0 and 7
Block 5 Integration
..........................................................................................................

x = 12 and decreasing between x = 12 and x = 32 . It is increasing between

x = 32 and x = 2.
2. The function has a maxima at x = 21 and x = 2 in the interval [0, 2]; it has a
minima at x = 32 .

See Fig. 3. Unlike the semi-circle, the area under f(x) cannot be closely
approximated by regular polygons. So, instead of a single gure like a regular
polygon, we use several rectangles. See Fig. 4.

A1 A2 A3 A4 A5 A1 A2 A3 A4 A5
(a) Lower bound for the area. (b) Upper bound for the area.

A1 A2 A3 A4 A5
(c) Differences in area.

Fig. 4: Approximating the area under the curve using rectangles.

Here, we subdivide the interval [0, 2] into four parts at the points x = 14 , x = 34 ,
 
x = 45 . So, A1 denotes the point (0, 0), A2 denotes the point 14 , 0 , A3 denotes
   
the point 43 , 0 , A4 denotes the point 54 , 0 and A5 denotes the point (2, 0).
8
Unit 17 Introduction to Integration
..........................................................................................................

Consider the blue coloured rectangles with the line segments A1 A2 , A2 A3 ,

A3 A4 and A4 A5 as bases. You can see that the height of each rectangle is the
in mum value of the function f(x) in the interval given by the base. For example,
the height of the blue coloured rectangle with base A1 A2 is

1
  
inf f(x) x ∈ 0, .
4
You are probably wondering how we can nd the in mum of f(x). As we
mentioned earlier, f(x) goes on increasing from x = 0 all the way up to x = 14 .
Because of this the value
i of
i f(x) at the starting point x = 0 is less than the value
1
of f(x) at any point in 0, 4 . So, the in mum value is attained at the starting
point x = 0 and the in mum value is f(0) = 1.

As you can see in Fig. 4(a), the sum of the areas of these blue rectangles is
less than the area under the curve. So, the sum of the areas of the blue
rectangles give a lower bound for the area of the curve y = f(x).

Similarly, the height of each red coloured, outer, rectangle in Fig. 4(b) is the
supremum value of f(x) in the interval given by the base. For example, the
height of the outer rectangle with base A1 A2 is

1
  
sup f(x) x ∈ 0, .
4

Again, the fact that f(x) is increasing from x = 0 to x = 41 comes in useful in

nding the supremum. Because it is increasing, the value at x = 14 is greater
h h
than all the values of f(x) in the interval 0, 14 . So, the supremum value is
 
attained at x = 41 and this is f 41 = 41 9
16 = 2 16 units.

As you can see in Fig. 4(b), the sum of the areas of the red coloured rectangles
is greater than the area under the curve. So, the sum of the red coloured
rectangles give an upper bound for the area under the curve.

You may have noticed that, in Fig. 4((c),the yellow coloured portion gives the
difference between the area under the curve and the areas of the smaller, blue
coloured, rectangles. The red coloured portion gives the difference between the
area under the curve and the areas of the larger, red coloured, rectangles. As
in the case of the method of exhaustion, the sum of the areas of the outer
rectangles and the sum of the areas of the inner rectangles approach closer
and closer to the area under f(x).

In the next subsection, we will put our discussion on a formal footing.

17.2.1 Upper and Lower Sums

We begin this subsection with the de nition of a partition. By an ordered set, we

mean that the order of
De nition 1: Suppose n ≥ 1 and let listing the elements is
important. Thus,
a = x0 < x1 < x2 < · · · < xn 1 < b = xn {1, 2} 6 = {2, 1} as ordered
sets. The ordered sets
be points in an interval [a, b] ⊂ R. Then, the ordered set are special kind of sets
and should not be
{x0 , x1 , . . . , xn } ⊂ [a, b] confused with the usual
sets where the order is
is called a partition of [a, b]. We will denote the set of all partitions of [a, b] by
not important.
P([a, b]). 9
Block 5 Integration
..........................................................................................................

Example 1: Give a partition of the interval [0, 2].

Solution: The set of points

n we used too subdivide the interval [0, 2] in our
earlier discussion, namely 0, 41 , 34 , 45 , 2 , gives a partition of the interval [0, 2].
Note that, here x0 = 0, x1 = 41 , x2 = 34 , x3 = 54 and x4 = 2.

∗∗∗

Next, we introduce the concepts of upper and lower sums.

De nition 2: Let f : [a, b] → R be a bounded function and let

P = {a = x0 < x1 < . . . < xn = b} ⊂ [a, b]

be a partition. For 1 ≤ i ≤ n, let us write

Δi = xi xi 1 . . . (1)
mi = inf {f(x)|x ∈ [xi 1 , xi ]} . . . (2)
Mi = sup {f(x)|x ∈ [xi 1 , xi ]} . . . (3)

Then, we de ne the upper sum U(P, f) by

n
X
U(P, f) = Mi Δi . . . (4)
i=1

and the lower sum L(P, f) by

n
X
L(P, f) = mi Δi . . . (5)
i=1

These sums are called Darboux sums after the French mathematician, Jean
Gaston Darboux who de ned them.

Let us now look at an example to understand upper and lower sums.

Example 2: For the function f(x) = 4x3 12x2 + 9x + 1non the interval o [0, 2], nd
1 3 5
the upper and lower sums with respect to the partition 0, 4 , 4 , 4 , 2 .

Solution: Here, x0 = 0, x1 = 41 , x2 = 34 , x3 = 54 , and x4 = 2. Let us now

calculate the Δi 's.
1 1 3 1 1

Δ1 = x1 x0 = 0 = , Δ2 = x2 x1 = = 
4 4 4 4 2 . . . (6)
5 3 1 5 3
Δ3 = x3 x2 = = , Δ4 = x4 x3 = 2 = 
4 4 2 4 4
Let us now compute the values of Mi , mi , 1 ≤ i ≤ 4. Let us use Fig. 5 to nd
these values.
h i
We see that f(x) is increasing on 0, 14 . So, the in mum value is attained at
x = 0 and the supremum is attained at x = 41 . The in mum value is the
y-coordinate of P1 and the supremum value is the y-coordinate of P2 . So, we
get the following values for m1 and M1 .
1
   
m1 = inf f(x) x ∈ 0, = f(0) = 1 
4


. . . (7)
1 1 41 
    
M1 = sup f(x) x ∈ 0, =f = 
4 4 16

10
Unit 17 Introduction to Integration
..........................................................................................................
h i
You can see from Fig. 5 that, in the interval 14 , 34 , the function f(x) increases
from x = 14 to x = 21 and decreases from x = 12 to x = 34 . In this interval, f(x)
attains its in mum value at x = 41 and its supremum value at x = 21 . The in mum
 
value is the y-coordinate of P2 = 14 , 41 16 and the supremum value is the
 
y-coordinate of P3 = 21 , 3 . So, we get the following values for m2 and M2 .

1 3 1 41 
     
m2 = inf f(x) x ∈ , =f =
4 4 4 16 

. . . (8)
1 3 1
    
M2 = sup f(x) x ∈ , =f =3


4 4 2

h i
In the interval 43 , 45 , f(x) decreases from x = 34 to x = 45 . The function attains its
in mum value at x = 54 and the in mum value is the y-coordinate of the point
 
P5 = 45 , 21 3
16 . The function attains its supremum value at x = 4 and the
supremum value is the y-coordinate of the point P4 = 43 , 43
16 . So, we get the
following values for m3 and M3 .
3 5 5 21 
     
m3 = inf f(x) x ∈ , =f =
4 4 4 16 

. . . (9)
3 5 3 43 
    
M3 = sup f(x) x ∈ , =f = 
4 4 4 16


3 P3 P7
P4
P2

P5
1 P1 P6

A1 A2 A31 A4 2 A5

Fig. 5: In mum and Supremum values of the function f(x) = 4x3 12x2 + 9x + 1 in
the interval [0, 1].

h i
In the interval 54 , 2 , f(x) decreases from x = 54 to x = 52 and increases from
 
x = 52 to x = 2. The in mum value is the y-coordinate of P6 = 25 , 1 and the
supremum value is the y-coordinate of P7 = (2, 3). So, we get the following
values for m4 and M4 .
5 5
     
m4 = inf f(x) x ∈ ,2 =f = 1
4 2


. . . (10)
5
  
M4 = sup f(x) x ∈ ,2 = f(2) = 3 

4

11
Block 5 Integration
..........................................................................................................

From Eqn. (7), Eqn. (8), Eqn. (9) and Eqn. (10), we see that

U(P, f) = M1 Δ1 + M2 Δ2 + M3 Δ3 + M4 Δ4
41 1 1 43 1 3 511
= · +3· + · +3· =
16 4 2 16 2 2 64
L(P, f) = m1 Δ1 + m2 Δ2 + m2 Δ3 + m4 Δ4
1 41 1 21 1 3 59
=1· + · + · + ·1=
4 16 2 16 2 2 16

∗∗∗

You may have noticed that , m1 Δ1 , m2 Δ2 , m3 Δ3 and m4 Δ4 are respectively,

the areas of the blue coloured rectangles in Fig. 4(a) with bases A1 A2 , A2 A3 ,
A3 A4 and A4 A5 . Similarly, M1 Δ1 , M2 Δ2 , M3 Δ3 and M4 Δ4 are, respectively, the
areas of the red coloured rectangles in Fig. 4(b) with bases A1 A2 , A2 A3 , A3 A4
and A4 A5 . You may have noticed that the upper sum we have found is the sum
of the areas of the red coloured rectangles and the lower sum we have found is
the sum of the areas of the blue coloured rectangles. Thus, intuitively, the
upper sums provide an upper bound for the area below the curve

y = f(x) = 4x3 12x2 + 9x + 1

and the lower sums give a lower bound for the area.

Try the next exercise to check your understanding of upper and lower sums.

1 in the interval [0, 1]

E1) Find the U(P, f) and L(P, f) for the function f(x) = 1+x
n o
with respect to the partition 0, 14 , 21 , 35 , 1 .

In this subsection, we de ned what a partition of an interval is and de ned the

upper and lower sums of a function with respect to a partition. We conclude our
discussion of upper and lower sums and we will begin the discussion of upper
and lower integrals in the next subsection.

17.2.2 Upper and Lower Integrals

Let us now return to the problem of nding the area under the curve
4x3 12x2 + 9x + 1 from x = 0 to x = 2. As we have already noticed, the upper
sums provide an upper bound for the area and the lower sums provide a lower
bound for the area. What happens to these sums as we increase the number of
intervals in the partition? Let us investigate this now with the same example.

As we increase the number of subdivisions, the areas of the inner rectangles

become larger and larger and the areas of the outer rectangles become smaller
and smaller. The upper sums approach closer and closer to a value that we call
the upper integral. The lower sums approach closer and closer to a value that
we call the lower integral.

In the situation where the upper and lower integrals are equal, both the upper
and lower sums approach closer and closer to a common limiting value, which
is the area under the curve. The upper sums approach this value from above
and the lower sums approach the value from below. See Fig. 6. The purpose of
the gure is to give an intuitive feel. You will see a formal proof of this fact in
12 your Real Analysis course.
Unit 17 Introduction to Integration
..........................................................................................................

Upper Sum = 5.03 Upper Sum = 4.65

Lower Sum= 3.07 Lower Sum= 3.44
R2 R2
f(x)dx = 4 f(x)dx = 4
0
0

f
f

(a) Six subdivisions. (b) Ten subdivisions.

Uppersum = 4.32
Lowersum = 3.68
R2
f(x) dx = 4
0 f

(c) Twenty subdivisions.

Fig. 6: Upper and Lower Sums.

We will now proceed to formally de ne the upper and lower integrals of a real
valued function which is bounded in a closed interval. We start with a simple
theorem that provides the basic foundation for the de nition of these concepts.

Theorem 1: Let f : [a, b] → R be a bounded function and suppose that

m = inf{f(x)|x ∈ [a, b]} and M = sup{f(x)|x ∈ [a, b]}

Then, for any partition P of [a, b] we have Since f(x) is bounded, m

and M exist.
m(b a) ≤ L(P, f) ≤ U(P, f) ≤ M(b a) . . . (11)

Proof: Let

P = {a = x0 < x1 < · · · < xn 1 < xn = b} 13

Block 5 Integration
..........................................................................................................

If A and B are subsets of be a partition of [a, b]. Recall the de nition of Mi in Eqn. (3) and the de nition of
R such that A ⊂ B, then U(P, f) in Eqn. (4). Since [xi 1 , xi ] ⊂ [a, b], we have Mi ≤ M. Since Δi > 0, we
inf A ≥ inf B and have Mi Δi ≤ MΔi . So, we have
sup A ≤ sup B. If a < b
and c > 0, then ac < bc. n
X n
X n
X
U(P, f) = Mi Δi ≤ M (xi xi 1 ) = M (xi xi 1 )
i=1 i=1 i=1

𝑏 = 𝑥𝑛 − 𝑥𝑛−1 The last sum in the previous line is what is called a telescopic sum because it
𝑥𝑛−1 − 𝑥𝑛−2
folds up like a telescope. Writing out the terms of the sum in reverse order, we
see that all the other terms except the rst term and last term cancel out. See
𝑥𝑛−2 − 𝑥𝑛−3 Fig. 7: (The arrows show the terms that are cancelled out.) So, we have
n
⋮ X
U(P, f) ≤ M (xi xi 1 ) ≤ M(b a)
𝑥2 − 𝑥1
i=1
𝑥1 − 𝑥0 = 𝑎 and this proves the rst part of Eqn. (11). We can prove the inequality for
m(b a) ≤ L(P, f) along similar lines. It follows from the fact that mi ≥ m. Let us
Fig. 7: Telescoping now prove the middle inequality, namely L(P, f) ≤ U(P, f). We have mi ≤ Mi ,
Sum. 1 ≤ i ≤ n. Since Δi > 0, it follows that mi Δi ≤ Mi Δi . Summing up both sides of
this inequality from i = 1 to n, we get L(P, f) ≤ U(P, f). 

Let us write

L(f) = {L(P, f)|P is a partition of [a, b]}

U(f) = {U(P, f)|P is a partition of [a, b]}

From Theorem 1 it follows that the U(f) is bounded below by m(b a). So, the
in mum of this set is nite. Similarly, L(f) is bounded above by M(b a) and
therefore the supremum of this set is nite. So, the following de nition of upper
and lower integrals is meaningful.

De nition 3: Let f : [a, b] → R be a bounded function. Then the upper integral

is de ned by

Rb
f(x)dx = inf(U(f)) = inf {U(P, f)|P is a partition of [a, b]} . . . (12)
a

and the lower integral is de ned by

Rb
f(x) dx = sup(L(f)) = sup {L(P, f)|P is a partition of [a, b]} . . . (13)
a
We will now look at some examples.

Example 3: Let us consider the function f(x) de ned by

(
0, if x is irrational.
f(x) =
1, if x is rational.

P. G. L. J. Dirichlet This function is called the Dirichlet function after the famous German
1805-1859 mathematician Lejeune Dirichlet who de ned this. Find the upper and lower
integrals of the function f(x).

Solution: Let

14 P = {0 = x0 < x1 < x2 < · · · < xn 1 < xn = 1}

Unit 17 Introduction to Integration
..........................................................................................................

be any partition of [0, 1]. Each sub-interval [xi , xi 1 ] contains both rational as
well has irrational numbers. On each interval [xi 1 , xi ], for rational values of x,
f(x) takes the value 1 and f(x) takes the value 0 for irrational values of x. So,
Mi = 1, mi = 0 for each i.
Thus
n
X n
X
U(P, f) = Mi Δxi = (1) (xi xi 1 ) = 1 0=1
i=1 i=1
and
n
X n
X
L(P, f) = mi Δxi = (0) (xi xi 1 ) = 0
i=1 i=1
Since P is an arbitrary partition of [0, 1], this means that U(P, f) = 1, L(P, f) = 0
for any partition of [0, 1]. It follows that
inf {U(P, f)|P is a partition of [0, 1]} = {1}
and
sup {L(P, f)|P is a partition of [0, 1]} = {0}.
So,

R1
f(x) dx = inf {U(P, f)|P is a partition of [0, 1]} = 1
0
and
R1
f(x) dx = sup {L(P, f)|P is a partition of [0, 1]} = 0
0

∗∗∗
Let us now look at another example, the upper and lower integrals of a step
function.
Example 4: Find the upper and lower integrals for the function f(x) de ned by
(
0, if 1 ≤ x < 0
f(x) =
1, if 0 ≤ x ≤ 1
in the interval [ 1, 1] .

··· 0
x0 = 1 x1 x2 xi0 xi0 +1 xn 1 xn = 1

Fig. 8: Graph of f(x).

15
Block 5 Integration
..........................................................................................................

Solution: Let

P = { 1 = x0 < x1 < · · · < xi 1 < xi < · · · < xn 1 < xn = 1}

be a partition of [ 1, 1]. Let i0 be such that xi0 < 0 ≤ xi0 +1 . Note that
(
0 if 0 ≤ k ≤ i0 + 1
mk =
1 if i0 + 1 < k ≤ n
h i
Can you see why this is so? In the subinterval 1, xi0 , f(x) = 0, so mk = 0 for
h h
all 1 ≤ k ≤ i0 . Also, f(x) = 0 in xi0 , 0 . So, mk = 0 for k = i0 + 1 also. Since
h i
f(x) = 1 in xi0 +1 , xn , mk = 1 for k > i0 + 1. So,

iX
0 +1 n
X n
X
L(P, f) = mk Δk + mk Δk = (xk xk 1 )
k=1 k=i0 +2 k=i0 +2

Since the last sum is a telescopic sum, we have

L(P, f) = xn xi0 +1 = 1 xi0 +1

If we write d = xi0 +1 , then d is the distance between 0 and the point xi0 +1 . Let us
write

Δ(P) = max {Δi |1 ≤ i ≤ n}

Then as Δ(P) approaches zero, d also approaches zero and L(P, f) approaches
R1
1 from below. So, f(x) dx = 1.
1

Again, as we did in the case of mk , check that

(
0, if 0 ≤ k ≤ i0
Mk =
1, if i0 + 1 ≤ k ≤ n

So, we have
i0
X n
X n
X
U(P, f) = Mk Δk + Mk Δk = (xk xk 1 ) = 1 x i0
k=1 k=i0 +1 k=i0 +1

Writing xi0 = e, e > 0 is the distance between xi0 and 0. So, U(P, f) = 1 + e ≥ 1
and 1 + e approaches closer and closer to 1 as Δ(P) approaches zero. So,

R1
f(x) dx = 1
1
. ∗∗∗

Here is an exercise to check your understanding of what we have discussed so

far.

R1 R1
E2) Find 0 f(x) dx and 0 f(x) dx for the function de ned by f(x) = 2 on [0, 1].
16
Unit 17 Introduction to Integration
..........................................................................................................

In this subsection, we de ned the concepts of upper and lower sums. We also
saw that, as we re ne a partition by adding more and more points the upper
sums and lower sums become closer and closer to certain values that we call
upper and lower integrals, respectively. In some cases, the upper and the lower
integral are equal, say 𝛼. In this case, the upper and lower sums converge to
this common value 𝛼. In the next subsection we will study the situations in
which the upper and lower integrals coincide.

17.2.3 The De nite Integral

In Exercise 2, we found the upper and lower integrals of the constant function
f(x) = 2 over the interval [0, 1]. You would have seen that the values coincide
and the common value is 2. Note that, the `area' under the function f(x) = 2 over
the interval [0, 1] is 2 and this is the area of the rectangle of sides 1 and 2 units.
(Here, we put `area' in quotes because this could be negative. You will see
more about this in the last unit of this block.) This agrees with our intuition that
the upper sums approach closer and closer to the `area' from the above and the
lower sums approach closer and closer to the `area' from below. So, the
common limit is the `area' of the region enclosed by the function. Motivated by
this, we de ne the integral of a real valued function over an interval as follows:

De nition 4: Let f : [a, b] → R be a bounded function. We say that f is

integrable if

Rb Rb
f(x) dx = f(x) dx
a a

Rb
We call the common value the integral of f and denote it by a f(x) dx. We call f
the integrand. We also call b and a, the upper and lower limits of
integration, respectively.
In some textbooks, you will come across Riemann's de nition of the integral
which is different from De nition 4. However, our de nition and Riemann's
de nition are equivalent de nitions of integrability in the sense that a function
that is integrable according to one de nition is also integrable according to the
other de nition. Further, the value of the integral will be the same.

For convenience, we adopt the following convention: We let

Ra Rb
f(x) dx = f(x) dx . . . (14)
b a

There is a well known criterion for integrability of functions. We state this now
without proof.

Theorem 2: Let f : [a, b] → R be a bounded real valued function. Then, f is

integrable if and only if, for every 𝜖 > 0, there exists a partition P𝜖 such that
|U(P𝜖 , f) L(P𝜖 , f)| < 𝜖. 
While it is dif cult to show that the upper and lower integrals are equal in
particular cases, there are general results that prove that certain classes of
functions are integrable. You will see the proofs of such results in your Real
Analysis course. Here, we give one such result without proof.

Theorem 3(Integrability): Let f : [a, b] → R be any function. If f is bounded and

monotonic or if f is continuous, then f is integrable.  17
Block 5 Integration
..........................................................................................................

If a function is integrable, it is necessarily bounded. We can use this necessary

condition often to prove that certain functions are not integrable over certain
intervals.

On the other hand, Theorem 3 gives us a suf cient condition for a function to
be integrable and we use this to prove that certain functions are integrable
over an interval. It is worth noting here that any function which is continuous on
a closed interval is bounded on that interval.

Let us now look at some examples to understand how to apply the necessary
and suf cient conditions for integrability of a real valued function.

Example 5: Discuss the integrability of the following functions:

i) sin x ii) tan x iii) ln|x| iv) 3
2 2+x
v) 2+x vi) x2 1 vii) sin x
x2 1 x3 x2 +x+1 x

Solution:

i) The trigonometric function sin x is continuous in any closed and bounded

By a closed and bounded interval. We can apply Theorem 3 to conclude that sin x is integrable in
interval of R, we mean an every closed and bounded subinterval of R.
interval of the form [a, b],
a, b ∈ R. ii) The function tan x is not integrable in any closed interval containing an odd
multiple of 𝜋2 , i.e a real number of the form (2k + 1) 𝜋2 , k ∈ Z, because it is
unbounded in any interval that contains (2k + 1) 𝜋2 , k ∈ Z.
Of course, the rst dif culty is that the function itself is not de ned at
(2k + 1) 𝜋2 . Since cos(2k + 1) 𝜋2 is zero, and division by zero is not de ned,
the function tan x is not de ned at x = (2k + 1) 𝜋2 . We can try to rectify the
situation by assigning some value to the function at (2k + 1) 𝜋2 . As you will
see in more advanced courses, even if we change the value of an
integrable function at nitely many points in a closed and bounded interval
[a, b], the function remains integrable in that interval. Also, changing the
values of an integrable function at nitely many points of a closed and
bounded interval doesn't affect the value of the integral of the function over
that interval. So, we can de ne a function f(x) as follows:
(
0, if x = (2k + 1) 𝜋2 , k ∈ Z.
f(x) =
tan x, Otherwise.

sin x and cos x → 0 as x → (2k + 1) 𝜋 , but

When x 6 = (2k + 1) 𝜋2 , f(x) = tan x = cos x 2
sin x → 1. So, their ratio tends to ∞ as x → (2k + 1) 𝜋2 . Therefore, the
function f(x) is also unbounded in any interval that contains (2k + 1) 𝜋2 . So,
f(x) is not integrable in any interval containing (2k + 1) 𝜋2 . On the other hand,
tan x is integrable on every bounded interval not containing (2k + 1) 𝜋2 .

iii) The function ln|x| is not de ned at 0. We can de ne the value of ln|x| to be
1 at x = 0 as we did in the case of tan x. However, because ln|x| → ∞ as
x → 0 the function ln|x| is unbounded in any closed and bounded interval
of R that contains 0. Therefore, it is not integrable on any closed and
bounded interval of R that contains 0. The function ln|x| is continuous in
any closed and bounded interval of R not containing 0. So, it will be
integrable on all closed and bounded intervals of R that do not contain 0.

iv) The function 2 is a rational function, i.e. a ratio of two polynomials. All
3+x2
rational functions whose denominator doesn't vanish in an interval are
g(x)
integrable over that interval. Can you see why? If f(x) = h(x) is a ratio of
18
Unit 17 Introduction to Integration
..........................................................................................................

1 is also continuous in
two polynomials and h(x) 6 = 0 in an interval, then h(x)
the interval. Since g(x) is continuous because it is a polynomial, the
1 is also continuous. Since 3 + x2 ≥ 3 > 0, the denominator
product g(x) · h(x)
is always positive So, in particular, the rational function f(x) = 2 is
3+x2
integrable over any bounded interval in R.
v) In this case, solving the equation x2 1 = 0, we get x = ±1. Further, as
x → 1 and x → 1, the numerator approaches a nite limit other than 0 and
the denominator approaches 0. So the function is not bounded in any
closed and bounded interval containing ±1. So, it is not integrable on any
closed and bounded interval that contains ±1 It is integrable on any closed
and bounded interval that doesn't contain ±1.
 
vi) Here, the denominator x3 x2 + x + 1 = (x 1) x2 + 1 is zero at x = 1 and
non-zero at every other point on the real line. Both the numerator and the
denominator approaches zero as x → 1, but the numerator and the
denominator have a common factor x 1. We have
x+1
lim f(x) = lim 2 = 1.
x→1 x→1 x + 1

We can de ne a function f : R → R as follows:

 x2 1 , if x 6 = 1


f(x) = x3 x2 +x+1
1, if x = 1

The function f(x) is continuous in any interval containing 1. Further, the

denominator of f(x) doesn't vanish at any other point, so f(x) remains
remains continuous in any closed and bounded subinterval of R. Hence
f(x) is integrable on every closed and bounded subinterval of R.
vii) The denominator is 0 at x = 0. Both the numerator and the denominator
approaches 0 as x → 0. However, there is no common factor that we can
remove. We have, from Example 25, Unit 7,
sin x
lim = 1.
x→0 x

We de ne f : R → R by
(
1, if x = 0
f(x) = sin x
x if x 6 = 0

So, the function remains continuous at 0. At all other points, the

denominator neither vanishes nor tends to 0. So, this function is integrable
on every closed and bounded interval of R.

∗∗∗

Remark 1: You will nd that being unbounded is one of the most common
reasons for a function being non-integrable in most of the cases that we
encounter. One of the reasons is that the function tends to in nity at some
point. Again, a common reason for this could be that function is de ned as a
ratio of two functions, and the function in the denominator vanishes at some
point in the interval. But in these cases also, we have to carefully examine the
limit of the function as the value in the domain approaches the possibly
troublesome point. If the limit exists and is nite, the vanishing of the
denominator is not a problem. We can always rede ne the values at such
points provided that there are only nitely many points. 19
Block 5 Integration
..........................................................................................................

Here are some exercises for you to test your understanding of the discussion
so far.

E3) Which of the following functions are integrable? Justify your answer.
i) f(x) = x2 + 2x + 1 in the interval [1, 2].

0 if x = 1
ii) f(x) = x2 +x+1

2
if 1 < x ≤ 2
x 4x+3
in the interval [1, 2].
1 sin x if 0 ≤ x < 𝜋
(
iii) f(x) = cos x 2
0 if x = 𝜋2
in the interval 0, 𝜋2 .
 
(
0 if x = 0
iv) f(x) =
x ln x if x > 0
in the interval [0, 1].

E4) Is the Dirichlet function de ned in Example 3 bounded? Is it integrable?

Justify your answer.

In the next theorem, we list some of the properties of the integral without proof.
You will see proof of these results in your Real Analysis course.

Theorem 4: a) If f : [a, b] → R is a constant function f(x) = k, k ∈ R , then

Rb
f(x)dx = k(b a)
a

b) If f is a real valued function which is de ned and integrable on [a, c] and

[c, b], it is integrable on [a, b] and

Rb Rc Rb
f(x) dx = f(x) dx + f(x) dx
a a c

c) Algebra of Integrals: Let f1 : [a, b] → R and f2 : [a, b] → R be integrable

functions. Then, for any constants a1 and a2 ∈ R the function
(a1 f1 + a2 f2 ) : [a, b] → R de ned by

(a1 f1 + a2 f2 ) (x) = a1 f1 (x) + a2 f2 (x)

is also integrable and we have

Rb Rb Rb
(a1 f1 + a2 f2 ) (x) dx = a1 f1 (x) dx + a2 f2 (x) dx . . . (15)
a a a

The function (f1 f2 ) : [a, b] → R, de ned by (f1 f2 ) (x) = f1 (x)f2 (x) is integrable.

We will not discuss examples of applications of Theorem 4 now because we

will prove analogous results for inde nite integrals later and we will discuss
20 examples there.
Unit 17 Introduction to Integration
..........................................................................................................

80

v(t)
70

60

50

40

30

20

10

0
0 10 20 30 40

Fig. 9: Speed-time function.

You may have noticed that, while we discussed the existence of integrals of
various functions, we never actually found the integrals of anything but the
simplest of functions. It is generally dif cult to nd the integral of a function from
rst principles. In the next section we will discuss a method for actually nding
the integral in many cases.

17.3 FUNDAMENTAL THEOREM OF CALCULUS

So far in our discussion, we have not seen any connection between

differentiation and integration. For all we know, they may be unrelated. But, are
they really unrelated? Let us discuss a situation before we actually present the
relationship between integration and differentiation.

Let us suppose that you are making a long trip by car and you would like to
know the distance you have covered every half an hour. Although your
Odometer (the device for measuring the distance) is not working, your
speedometer is working. Suppose your friend is taking readings of the
speedometer from time to time and the values are as follows:

Time in minutes 5 15 25 30
Speed in km/hr 30 50 65 80

Can you nd the approximate distance you have travelled in rst half an hour?

Let us bring in Mathematics now. We have two functions, s(t) that gives the
distance travelled in t seconds and v(t) gives the speed in the tth second.

To simplify things, let us suppose that the acceleration was constant in the
given time intervals. In this case, the graph of the function is as in Fig. 9. We
have

Distance travelled = Speed × Time Taken 21

Block 5 Integration
..........................................................................................................

The distance travelled in rst 5 minutes is s(5) s(0). Since the minimum
speed in this interval is v(0), the distance travelled in this interval is at least
5 , i.e.
v(0) · 60

5
v(0) · ≤ s(5) s(0)
60
On the other hand, the maximum speed is v(5). So, the distance travelled in
5 , i.e.
this interval is at most v(5) · 60

5
s(5) s(0) ≤ v(5) ·
60
Combining the inequalities, we get the following upper bounds and lower
bounds for the distance travelled in each of the time intervals.
5 5
v(0) · ≤ s(5) s(0) ≤ v(5) ·
60 60
Similarly, for each of the remaining time intervals we will get bounds for the
distance travelled in that time interval. The following are the bounds:
5 ≤ 5

v(0) · 60 s(5) s(0) ≤ v(5) · 60 


10 10

v(5) · 60 ≤ s(15) s(5) ≤ v(15) · 60 


. . . (16)
v(15) · 10
60 ≤ s(25) s(15) ≤ v(25) · 10 
60 



5 5 

v(25) · 60 ≤ s(30) s(25) ≤ v(30) · 60

Adding up all the inequalities in Eqn. (16), we get

S1 ≤ s(30) s(0) ≤ S2 ,

where
5 5 10 5
S1 = v(0) · + v(5) · + v(15) · + v(25) ·
60 60 60 60
5 10 10 5
S2 = v(5) · + v(15) · + v(25) · + v(30) ·
60 60 60 60
Did you notice that S1 is the lower sum L(P, v) of the function v with respect to
the partition

P = {x0 = 0, x1 = 5, x2 = 15, x3 = 25, x4 = 30}

of the interval [0, 30]? Also, we have,

v(0) = inf {v(t) |[x0 , x1 ] } v(5) = inf {v(t) |[x1 , x2 ] }

v(15) = inf {v(t) |[x2 , x3 ] } v(25) = inf {v(t) |[x3 , x4 ] }

Similarly, check that S2 = U(P, v).

As we increase the number of partitions, i,e, increase value of n, S1

Rb
approaches closer and closer to a f(x) dx and S2 approaches closer and closer
Rb
to a f(x dx) and we have

Rb Rb
v(t) dt ≤ s(30) s(0) ≤ v(t) dt
22 a a
Unit 17 Introduction to Integration
..........................................................................................................

Since v(t) is continuous, it is integrable, so the upper integral and the lower
integral of v(t) are equal. Therefore,

Rb Rb R30
s(30) s(0) = v(t) dt = v(t) dt = v(t) dt . . . (17)
a a 0

looks plausible.

Now, recall the relationship between distance travelled and the speed. We
know that, speed is the rate of change of distance. In other words,
ds
= v(t).
dt
Rb
This suggests that, to nd the integral a f(x) dx, we should nd a function F
such that F0 (x) = f(x) for all x ∈ [a, b]. Then,

Rb
f(x)dx = F(b) F(a)
a

This last statement is true. We will now state the relation formally, clearly
mentioning the conditions on the functions f and F under which this relation is
true. Before we formally state this relationship we need some terminology.

De nition 5: Suppose that f and F are two real valued functions de ned on
(a, b) for some a, b ∈ R, a < b. We say that F is the antiderivative of f on (a, b)
if:
1) The function F is differentiable in (a, b) and continuous on [a, b].
2) The function f is continuous on [a, b].
3) F0 (x) = f(x) for all x ∈ (a, b).
We also call F a primitive of f.
In the next example, we give some examples of functions and their
antiderivatives

Example 6: Check the following:

2
i) The function F de ned by F(x) = x2 is the antiderivative of the function f
given by f(x) = x on any subinterval of R.

ii) The function F de ned by F(x) = x|x| is the antiderivative of f, de ned by

f(x) = 2|x| on every subinterval of R.
Solution:
i) We know from Unit 9 of Block 3 that F(x) = x2 /2 is a polynomial so, it is
differentiable and continuous on every subinterval of R. Also, f(x) = x is a
differentiable function since it is a polynomial. In particular, it is also
continuous. Further, F0 (x) = f(x). So, all the three conditions are satis ed
and F is the antiderivative of f.

ii) From example 3 in Unit 11, we know that F is differentiable and its
derivative is f. Further, we know that 2|x| is a continuous function on any
subinterval of R. So, it follows that F is the antiderivative of f.
∗∗∗

Here are some exercises for you to check your understanding of the concept of
antiderivative 23
Block 5 Integration
..........................................................................................................

E5) Check the following:

i) The function cos x + x sin x is the antiderivative of x cos x on any
subinterval of R.
ii) The function x ln|x| x is the antiderivative of ln|x| on any subinterval
of R not containing 0.

After going through the de nition of antiderivative and looking at the examples,
one natural question that you may have is the following: `In the case of part a)
2
of Example 6, the function F1 de ned by F1 (x) = x2 + 1 also satis es the three
conditions in the de nition. So, it is also an antiderivative of f. How many
antiderivatives can a function have and how are the different antiderivatives of
a function related to each other?'. The next theorem answers this question.

Theorem 5: Suppose F and G are two antiderivatives of f on [a, b], i.e. we have
F0 = f and G0 = f on (a, b). Then, F(x) = G(x) + C for some constant function C.
Conversely, if F is the antiderivative of a function f then F + C is also an
antiderivative of f where C is a constant function.

Proof: We have
d
(F(x) G(x)) = F0 (x) G0 (x) = f(x) f(x) = 0
dx
So, F G is a constant function, say C. Then, F(x) = G(x) + C. Conversely, if F
is an antiderivative of f on [a, b], for any constant function C we have
d
(F(x) + C) = F0 (x) = f(x).
dx
So, F + C is also an antiderivative of f(x) for any constant C. 

Let us take the set S to be set of all functions that are differentiable on [a, b]
with a continuous derivative. We de ne a relation on S by F ∼ G if F0 (x) = G0 (x).
Then, you can easily check that this is an equivalence relation. So, the set of all
antiderivatives of a continuous function forms an equivalence class. Let us
summarise the discussion in the form of a theorem.

Theorem 6: Let S to be set of all functions that are differentiable on [a, b] with a
continuous derivative. We de ne a relation on S by F ∼ G if F0 (x) = G0 (x).
Then, ∼ is an equivalence relation on S. 

The next theorem tells us that antiderivatives exist for all continuous functions.

Theorem 7: Let f : [a, b] → R be a continuous function.

Part 1: The function F : [a, b] → R de ned by
Rx
F(x) = f(t) dt . . . (18)
a

is an antiderivative of f(x), that is, F0 (x) = f(x) for all x ∈ [a, b].

Part 2: If F is an antiderivative of f in (a, b), then

Rb
f(x) dx = F(b) F(a) . . . (19)
a

24 
Unit 17 Introduction to Integration
..........................................................................................................

Theorem 7 is known as the Fundamental Theorem of Integral Calculus or

Fundamental Theorem of Calculus.

Let us now look at some applications of Theorem 7.

Example
 27: Findthe following: 2
  
d 
Rx d 
Rsin x  Rx
d  2
t2 dt
 
i) sin t dt ii) 1 iii) t + t dt
  
dx dx dt
  
0 1 x

Solution: We use chain rule to nd these derivatives.

R x2
i) We set v = x2 , and write F(x) = 0 sin t dt, f(t) = sin t. Then,

d d dv
(F(x)) = (F(v))
dx dv dx

Using Eqn. (18), we have dv d (F(v)) = f(v) = sin x2 . Also, dv = 2x. Therefore,
dx
2
d R x sin t dt = 2x sin x2 .
dx 0
sin x
R
ii) Setting v = sin x and F(x) = dx d (1 t2 ) dt, f(t) = 1 t2 we get
1
d d d
(F(x)) = (F(v)) (v) = f(v) cos x = (1 sin2 x) cos x = cos3 x.
dx dv dx
iii) From part b) of Theorem 4 we have
2 2
Rx  2  Rx  2  Rx  2 
t + t dt = t + t dt + t + t dt
1 1 x

So,
2 2
Rx  2  Rx  2  Rx  2 
t + t dt = t + t dt t + t dt = F1 (x) F2 (x) (say).
x 1 1

We have
d d
(F1 (x)) = x4 + x2 2x and (F (x)) = x2 + x.
 
dx dx 2
Therefore,
2
Rx  2
t + t dt = 2x5 + 2x3 x2

x.
x
∗∗∗

If you have understood the above example, you will have no dif culty in proving
the following general result:

Theorem 8: Let f : [a, b] → R be a continuous function and let

g2 (x)
R
F(x) = f(t) dt
g1 (x) 25
Block 5 Integration
..........................................................................................................

where g1 (x) : [c, d] → [a, b] and g2 (x) : [c, d] → [a, b] are differentiable functions
on [c, d]. Then, we have

F0 (x) = F0 (g2 (x))g20 (x) F0 (g1 (x))g10 (x). . . . (20)

g2 (x)
 
d 
R
i.e. f(t) dt = f (g2 (x)) g20 (x) f (g1 (x)) g10 (x) . . . (21)

dx

g1 (x)

Proof: Using Theorem 4, part b), we have

g2 (x) g1 (x) g2 (x)

R R R
f(x) dx = f(x) dx + f(x) dx . . . (22)
a a g1 (x)

We write
g2 (x) g2 (x) g1 (x)
R R R
F(x) = f(t) dt = f(x) dx f(x) dx using Eqn. (22)
g1 (x) a a

We have
g2 (x) g1 (x)
   
d d  d 
R R
F(x) = f(x) dx f(x) dx .
 
dx dx dx
 
a a

g2 (x)
 
d 
R
Let us consider the rst term f(x) dx. As before, we set v = g2 (x).

dx

a
Then,
g2 (x)
 
d   d d
R
f(x) dx = (F(v)) (g2 (x)) = F0 (g2 (x))g20 (x) . . . (23)
dx dv dx

a

Similarly, we get

g1 (x)
 
d 
R
f(t) dt = F0 (g1 (x))g10 (x). . . . (24)

dx

a

The result in Eqn. (20) follows from Eqn. (23) and Eqn. (24) and Eqn. (21) is a
rephrasing of Eqn. (20). 
Here are some exercises to check your understanding of the previous example.

E6) Find the

√
following:
x tan x Rx
d d dt
R R
1 + t4 dt 2
   
i) ii) 1+t dt iii) d
dx
dx dx √ 1 + t2
0 1 x

Note that any function in the set of antiderivatives of f(x) on the interval [a, b]
Rb
26 will give the same value a f(x) dx when we use Eqn. (19) for evaluating de nite
Unit 17 Introduction to Integration
..........................................................................................................

integrals. If F1 (x) is any other antiderivative of f(x), we saw that F1 (x) = F(x) + C
for some constant C. From Eqn. (19), we have, for any constant C,

Rb
f(x) dx = {F1 (b) F1 (a)} = {F(b) + C} {F(a) + C} = F(b) F(a)
a

which is independent of C.

So, in some sense, we can regard any two antiderivatives of a function over a
closed interval the `same' as far as nding the integral of the function over that
interval is concerned. More formally, let S be the set of all functions on [a, b]
whose derivative is continuous. De ne a relation ∼ by `F ∼ G if they have the
same derivative, i.e. F0 (x) = G0 (x) in the interval [a, b]'. Note that this means
that F and G are the antiderivatives of the same function. We leave it to you to
check that ∼ de nes an equivalence relation.

De nition 6: Let f : [a, b] → R be a continuous function. If F(x) is an

antiderivative of f(x), we write
R
f(x) dx = F(x) + C.

where C is an arbitrary constant, and call f(x) dx it the inde nite integral of
R

f(x).

We conclude this section with an exercise for you.

E7) Check that the relation ∼ that we de ned earlier is an equivalence

relation.

In the next section, we will discuss the inde nite integrals of some common
functions that we have encountered so far.

17.4 STANDARD INTEGRALS

You would have noticed that the functions that we commonly encounter are
either polynomials, rational functions or functions involving trigonometric
functions. Once we know how to integrate a certain set of functions, we can
integrate many other functions using the following result, which is analogous to
Theorem 4, part c). You will see that the results are a direct consequence of the
results on differentiation.

Theorem 9: a) Let F : [a, b] → R be the antiderivative of f : [a, b] → R. Then,

the antiderivative of (𝛼f)(x) exists for any 𝛼 ∈ R and equals 𝛼F(x). Also,
R R
(𝛼f)(x) dx = 𝛼 f(x) dx. . . . (25)

b) If G : [a, b] → R is the antiderivative of g : [a, b] → R, then antiderivative of

(f + g)(x) exists and equals (F + G)(x). Also,
R R R
(f + g)(x) dx = f(x) dx + g(x) dx . . . (26)
27
Block 5 Integration
..........................................................................................................

Proof: a) If 𝛼 = 0, there is nothing to prove since both the RHS and LHS of
Eqn. (25) are 0. So, let us assume that 𝛼 6 = 0. Since f(x) is continuous, 𝛼f(x)
is also continuous. Further, since F(x) is differentiable on (a, b), (𝛼F)(x) is
also differentiable on (a, b) and we have
d d
((𝛼F)(x)) = 𝛼 (F(x)) = (𝛼f)(x).
dx dx
Therefore, by de nition of inde nite integral, (𝛼F)(x) is an antiderivative of
𝛼(f(x)) and we have
R
(𝛼f)(x) dx = (𝛼F)(x) + C, C ∈ R . . . (27)

By the de nition of the inde nite integral, we have

R
f(x) dx = 𝛼 F(x) + C0 , C0 ∈ R


𝛼

= 𝛼F(x) + 𝛼C0 , C0 ∈ R

Since 𝛼 6 = 0, as C0 takes all possible values in R, 𝛼C0 also takes all possible
values in of R. If we write C for 𝛼C0 , C takes all the values in R. So,
R
𝛼 f(x) dx = 𝛼F(x) + C, C ∈ R
R
= (𝛼f)(x) dx from Eqn. (27).

b) The proof of part b) is along similar lines. We have

d d
{(F + G)(x)} = {F(x) + G(x)}
dx dx
d d
= (F(x)) + (G(x))
dx dx
= f(x) + g(x) = (f + g)(x).

Therefore,
R
(f + g)(x) dx = (F + G)(x) + C, C ∈ R . . . (28)

We have
R R
f(x) dx + g(x) dx = F(x) + C1 + G(x) + C2 , C1 , C2 ∈ R

= F(x) + G(x) + C1 + C2 , C1 , C2 ∈ R

As C1 and C2 takes values over the whole of R, C1 + C2 also takes values

over the whole of R. Writing C for C1 + C2 , we have
R R
f(x) dx + g(x) dx = {(F + G)(x) + C|C ∈ R}
R
= (f + g)(x) dx from Eqn. (28).


Corollary 1: Let f1 , f2 ,. . ., fn be continuous functions on [a, b] and c1 , c2 , . . .,

cn ∈ R be constants. Then,
R n
X R
(c1 f1 + c2 f2 + · · · + cn fn ) (x) dx = ci fi (x) dx . . . (29)
i=1

28 
Unit 17 Introduction to Integration
..........................................................................................................

Table 1: Table of inde nite integrals

No. f(x) f(x) dx

R

1. xn , n 6 = 1, xn+1 +C
n+1

2. 1 ln|x| + C
x

3. sin x cos x + C

4. cos x sin x + C

5. tan x ln|sec x| + C

6. cot x ln|sin x| + C

7. sec x ln|sec x + tan x| + C

8. cosec x ln|cosec x cot x| + C

9. ex ex + C

10. √1 sin 1 x + C
1 x2

√
11. √1 ln x + x2 1 +C
x2 1

√
12. √1 ln x + x2 + 1 + C
x2 +1

13. √1 sec 1 x + C
x x2 1

14. 1 tan 1 x + C
1+x2

We now compile a table of functions and their inde nite integrals for future
reference in Table 1. We have not explicitly mentioned the intervals on which
the function in the second column is the inde nite integral of the rst column.
So, before using these results to evaluate de nite integrals over an interval,
check that the conditions for the antiderivatives are satis ed over the interval
you are integrating before you apply the formula!

You can check the table by differentiating the second column in the table and
see if you get the rst column. We will check some of the values in the next
example,

Example 8: Check that:

a) The inde nite integral of tan x is ln|sec x| + C in any subinterval of R not
containing (2k + 1) 𝜋2 , k ∈ Z. 29
Block 5 Integration
..........................................................................................................

b) The inde nite integral of sec x is ln|sec x + tan x| + C in any subinterval of R

not containing (2k + 1) 𝜋2 .
Solution: Indeed, we have dx d (ln|sec x|) = 1 · sec x · tan x = tan x. You can
sec x
check that tan x is continuous in any interval not containing (2k + 1) 𝜋2 . So,
ln|sec x| is the antiderivative of tan x. It follows that ln|sec x| + C is the inde nite
integral of tan x.

Let us now check that the antiderivative of sec x is ln|sec x + tan x|. Check that
sec x + tan x 6 = 0 whenever sec x and tan x are de ned. We have
d 1
· sec x tan x + sec2 x
 
(ln|sec x + tan x|) =
dx sec x + tan x
sec x(sec x + tan x)
= = sec x
(sec x + tan x)

∗∗∗

E8) Check the remaining values of Table 1.

Example 9: Calculate the following integrals:

R 3
R 3
R2 1
i) x dx ii) x 2 dx iii) dx
x5
R 1
R5 3
 R0 (x + 1)3
iv) dx v) x + 3x 2x dx vi) dx
x7/5 x2
Solution:
n+ 1
i) From the rst entry in Table 1, xn dx = xn+1 + C. So, we have
R
4
x dx = x4 + C.
R 3

ii) Again using rst entry in Table 1, we get

3
3 x 2 +1 2 5
R
x2 dx = 3 + C = x 2 dx + C
5
2 +1
iii) We have
1 x 5+1 x 4 1
R R
dx = x 5 dx = +C= +C= + C.
x5 5+1 4 4x4
iv) We have
7
1 7 x 5 +1 5
R R
7
dx = x 5 dx = 7
+C= 5
+C
x5 5 +1 2x 2
v) We now apply Eqn. (29) with f1 = x5 , c1 = 1, f2 = x3 , c2 = 3, f3 = x and
c3 = 2. We have
R R R R
x5 + 3x3 2x dx = x5 dx + 3 x3 dx 2 x dx
 

x6 x4 x2
= + C1 + 3 + C2 3 C3
6 4 2
x6 x4 x2
= +3 3 +C
6 4 2
where we write C for C1 + C2 C3 . In what follows, we will frequently
combine two or more constants into a single constant without explaining in
30 detail.
Unit 17 Introduction to Integration
..........................................................................................................

vi) We have (x + 1)3 = x3 + 3x2 + 3x + 1. Therefore,

(x + 1)3 3 1
2
= x + 3 + + 2.
x x x

Using Eqn. (29), we get

(x + 1)3 3 1
R R 
dx = x + 3 + + 2 dx
x2
R Rx x R 1 1
R
= x dx + 3 dx + 3 dx dx +
x x2
x2 x 2+1 1 x2
= + 3x + 3 ln|x| + +C= + 3x + + 3 ln|x| + C
2 2+1 x 2

∗∗∗

Here are some problems for you to try.

E9) Calculate the following integrals:

R 4
R 5
R2 1
i) x dx ii) x2 dx iii) dx
x7
R dx
R6 4 2
 R0 (x 1)4
iv) v) x + 3x 2x dx vi) dx
x5/3 x2

Let us now look at some examples that involve trigonometric functions.

Example 10: Calculate the following integrals:

R (2+3 sin x)2
i) (3 sin x + 4 cos x) dx ii) sec2 x dx iii) (x + sin x) dx iv) dx
R R R
2
cos x

Solution:

i) Using Eqn. (29), we get

R R R
(3 sin x + 4 cos x dx = 3 sin x dx + 4 cos x dx

= 3 cos x + 4 sin x + C

ii) We have

d d sin x cos x · cos x sin x( sin x)

 
tan x = =
dx dx cos x cos2 x
cos2 x + sin2 x 1
= 2
= = sec2 x
cos cos2 x
R
∴ sec2 x dx = tan x + C.

iii) Using Eqn. (29), we get

x2
R R R
(x + sin x) dx = x dx + sin x dx = cos x + C
2 31
Block 5 Integration
..........................................................................................................

iv) We have

(2 + 3 sin x)2 4 + 9 sin2 x + 12 sin x

=
cos2 x cos2 x
= 4 sec2 x + 9 tan2 x + 12 sec x tan x
= 4 sec2 x + 9(sec2 x 1) + 12 sec x tan x
= 13 sec2 x + 12 sec x tan x 9

Therefore, using Eqn. (29), we get

(2 + 3 sin x)2
R R R R
dx = 13 sec2 x dx + 12 sec x tan x dx 9 dx
cos2 x

We have sec2 x dx = tan x + C1 from part ii) of this example. Also

R

d
dx (sec x) = sec x tan x.. Therefore,
R
sec x tan x dx = sec x + C2 . Thus,

(2 + 3 sin x)2
R
dx = 13 tan x + 12 sec x 9x + C
cos2 x

∗∗∗

Try the following exercises to check your understanding of the above example.

E10) Calculate the following integrals:

R (3+2 cos x)2
i) (4 cos x + 3 tan x) dx ii) cosec2 x dx iii) dx
R R
2
sin x

We close this unit here. In the following units, we will develop methods, based
on Theorem 7, for evaluating integrals. You may like to go through a summary,
given in the next section, of the main topics discussed in this unit.

17.5 SUMMARY

In this Unit, we have:

1) de ned the lower and upper sums of functions de ned on [a, b]

corresponding to a partition of [a, b] and computed them for some simple
functions;
2) de ned the upper and lower integrals of a function and calculated them
for simple functions;
3) de ned the de nite integral of a given function and
4) stated the Fundamental Theorem of Calculus; and
5) used the Fundamental Theorem to calculate the de nite integral of some
integrable functions.

17.6 SOLUTIONS/ANSWERS

1 . Then, f(x) is monotonic decreasing in [0, 1]. So,

E1) Let f(x) = 1+x

32 inf {f(x) |x ∈ [xi , xi+1 ] } = f (xi+1 )

Unit 17 Introduction to Integration
..........................................................................................................

and
sup {f(x) |x ∈ [xi , xi+1 ] } = f (xi )
for any partition {0 = x0 < x1 < x2 < · · · < xn = 1} of [0, 1]. We have Δ1 = 41 ,
 
Δ2 = 41 , Δ3 = 10
1 , Δ = 2 . Further, we have m = f 1 =
4 5  1 4
4,
5
  
1 3 3 5
m2 = f 2 = 2 , m3 = f 5 = 8 and m4 = f(1) = 2 . So, 1

1 4 1 2 1 5 2 1 151
L(P, f) = · + · + · + · =
4 5 4 3 10 8 5 2 240
     
Also, M1 = f(0) = 1, M2 = f 14 = 4 , M = f 1 = 2 and M = f 3 = 5 .
5 3 2 3 4 5 8
So, U(P, f) = 1 · 14 + 45 · 41 + 23 · 10
1 + 5 · 2 = 23
8 5 30

E2) Let P = {0 = x0 < x1 < x2 < · · · < xn 1 < xn = 1} be any partition of [0, 1].
Since f(x) is a constant function on [0, 1] it follows that Mi = 2, mi = 2 for
any interval [xi , xi 1 ]. So, we have
n
X n
X
U(P, f) = Mi (xi xi 1 ) = 2 (xi xi 1 ) = 2(1 0) = 2
i=1 i=1
and
n
X n
X
L(P, f) = mi (xi xi 1 ) = 2 (xi xi 1 ) = 2(1 0) = 2
i=1 i=1
Therefore,
inf {U(P, f)|P is a partition of [0, 1]} = {2}
and
sup {L(P, f)|P is a partition of [0, 1]} = {2}
So,
R1
f(x) dx = sup{U(P, f)|P is a partition of [0, 1]} = sup{2} = 2.
0
R1
Similarly 0 f(x) dx = 2

E3) i) In this case, since f(x) = x2 + 2x + 1 is a polynomial, it is continuous.

So, it is integrable.
ii) In this case, the denominator factors as (x 1)(x 3), so we examine
the points 1 and 3. The numerator has no real roots. So, there is no
possibility of removing common factors. The domain doesn't contain
3, so we need not bother about it. However, 1 is in the domain and the
function becomes unbounded as x → 1. So, the function is not
integrable.
iii) Here, the numerator and the denominator tends to 0 as x → 𝜋2 .
Applying L'hospital's rule, we get
1 sin x cos x
lim𝜋 = lim𝜋 = 0.
x→ 2 cos x x→ 2 sin x
At all other points, 1 sin x and cos x are also continuous. Further,
cos x 6 = 0 at all the points other than 𝜋2 . Therefore, the function is
continuous at all the points other than x = 𝜋2 . So, the function is
integrable over the interval 0, 𝜋2 .
 
33
Block 5 Integration
..........................................................................................................

iv) Here ln x tends to ∞ as x → 0. But

 
1 x 1
ln x x2
lim x ln x = lim 1 = lim 1
= lim x = 0.
x→0 x→0 x→0 x→0
x x2
So, the function is continuous at x = 0. Since ln x and x are continuous
at all other points, their product is also continuous at all the points
other than 0. So, this function is continuous in the interval [0, 1] and
hence integrable.
E4) The Dirichlet function bounded because the only values it takes are 0 and
1 and so |f(x)| ≤ 1. However, as we saw in Example 3, the upper and
lower integrals are not equal and therefore it is not integrable.

E5) i) The functions x and sin x are differentiable in every subinterval of R.

So, x sin x is also differentiable in every subinterval of R. Since cos x
is also differentiable in every subinterval of R, it follows that
cos x + x sin x is differentiable in every subinterval of R. Further,
d
dx (cos x + x sin x) = sin x sin x + x cos x = x cos x. Since the function
cos x is continuous it follows that cos x + x sin x is the antiderivative of
x cos x.
The functions f(x) = x ln|x| is differentiable in any interval of R not
containing 0 and g(x) = x is differentiable in any interval of R. So,
x ln|x| x is differentiable on every interval of R that doesn't contain 0.
Further,
d 1
(x ln|x| x) = ln|x| + x 1 = ln|x|
dx x
The function ln|x| is continuous in every subinterval of R not
containing 0. It follows that x ln|x| x is the antiderivative of ln|x| in any
interval of R not containing 0.
√
Rt 2    √
E6) i) Here F(x) = t + 1 dt, f(t) = 1 + t2 . We set v = x.
0

d d dv 1 x2 + 1
(F(x)) = (F(v)) = f(v) √ = √
dx dv dx 2 x 2 x
tan x
R
1 + t2 dt, f(x) = 1 + t2 . We set v = tan x.
   
ii) Here F(x) =
1

d d dv
(F(x)) = (F(v)) = f(v) sec2 x
dx dv dx
= 1 + tan2 x sec2 x = sec4 x.
 

iii) We have
√
Rx dt
Rx dt
Rx dt
= +
1 + t2 1 + t2 √ 1 + t2
1 1 x
√
Rx Rx x
dt dt dt
R
=
√ 1 + t2 1 + t2 1 + t2
34 x 1 1
Unit 17 Introduction to Integration
..........................................................................................................
√
Rx dt
Rx dt
Rx dt
Let us write F(x) = , F1 (x) = , F2 (x) = . We
√ 1 + t2 1 + t2 1 + t2
x 1 1
have
d d d d √
d √

(F(x)) = (F1 (x)) (F2 (x)) = f(x) (x) f x x
dx dx dx dx dx
1 1
= √
1 + x2 2 x(1 + x)

E7) We have F0 (x) = F0 (x), so F ≡ F and the relation is re exive. If F ≡ G, we

have F0 (x) = G0 (x), G0 (x) = F0 (x). So, G0 (x) = F0 (x), that is G ≡ F and the
relation is symmetric. IF F ≡ G and G ≡ H, then F0 (x) = G0 (x) and
G0 (x) = H0 (x), F0 (x) = H0 (x), that is F ≡ H. So, the relation is transitive.
Thus, the relation is an equivalence relation.

E8) 1. When n 6 = 1 have

 n 
d x 1 d  n+1  1
= x = {(n + 1)xn+1 1 } = xn .
dx n + 1 n + 1 dx n+1

2. We have
d 1
(ln|x|) = .
dx x
3. We have
d d
( cos x) = (cos x) = ( sin x) = sin x.
dx dx
d (sin x) = cos x.
4. We have dx
d (ln|cosec x|) =
6. We have dx 1
cosec x cosec x cot x = cot x.
8. We have
d
(ln|cosec x cot x|)
dx
1
cosec x cot x + cosec2 x = cosec x
 
=
cosec x cot x
d (ex ) = ex .
9. We have dx
d sin 1 x = √ 1
 
10. We have dx . See Unit 9.
1 x2
11. We have
d  1 1 1
p   
ln x + x2 1 = √ 1+ √ (2x)
dx x + x2 1 2 x2 1
1
=√
x2 1
12. Similar to 11.
 
13. We have dx d sec 1 x = √ 1 . See Unit 9.
x x2 1
 
14. We have dx d tan 1 x = 1 . See Unit 9.
2 1+x

E9) i) Using the rst entry in Table 1, we get

x4+1 x5
R
x4 dx = dx = +C
4+1 5 35
Block 5 Integration
..........................................................................................................

5
x 2 +1 2 7
R
ii) We have 5
dx = x 2 + C.
+1 7
2
1 x 7+1 1
R 7
R
iii) We have dx = x dx = = .
x7 7+1 6x6
iv) Using Eqn. (29), we have
R6 R R R
x + 3x4 2x2 dx = x6 dx + 3 x4 dx x2 dx

2

x7 x5 x3
= +3 2 +C
7 5 3

vi) Expanding (x 1)4 using binomial theorem, we get

1)4
R (x
R x4 4x3 + 6x2 4x + 1
dx = dx
x2 x2
4 1

R 3 
= x 4x + 6 + 2 dx
x x

E10) i) We have
R R R
(4 cos x + 3 tan x) dx = 4 cosx + 3 tan x

= 4 sin x + ln|sec x| + C

d (cot x) =
ii) We have dx cosec2 x. Therefore, cosec2 x dx =
R
cot x + C.
iii) Expanding the numerator, we get

9 + 12 cos x + 4 cos2 x
R R 2
R
dx = 9 cosec x dx + 12 cosec x cot x dx
sin2 x
R
+4 cot2 x dx

= 9 cot x
R 12 cosec
2
x

+4 cosec x 1 dx

= 9 cot x 12 cosec x 4 cot x 4x + C

= 13 cot x 12 cosec x 4x + C

36