3

Definite Integral

3.1

Definition, Necessary & sucient conditions

Let f : [a, b] R be a bounded real valued function on the closed, bounded interval [a, b].
Also let m, M be the infimum and supremum of f (x) on [a, b], respectively.
Definition 3.1.1. A partition P of [a, b] is an ordered set P = {a = x0 , x1 , x2 , ..., xn = b}
such that x0 < x1 < ... < xn .
Let mk and Mk be the infimum and supremum of f (x) on the subinterval [xk1 , xk ],
respectively.
Definition 3.1.2. Lower sum: The Lower sum, denoted with L(P, f ) of f (x) with
respect to the partition P is given by
L(P, f ) =

n
!
k=1

mk (xk xk1 ).

Definition 3.1.3. Upper sum: The Upper sum, denoted with U (P, f ) of f (x) with
respect to the partition P is given by
U (P, f ) =

n
!
k=1

Mk (xk xk1 ).

For a given partition P , U (P, f ) L(P, f ). In fact the same inequality holds for any two
partitions. (see Lemma (3.1.6) below.)
Definition 3.1.4. Refinement of a Partition: A partition Q is called a refinement
of the partition P if P Q.
The following is a simple observation.
Lemma 3.1.5. If Q is a refinement of P , then
L(P, f ) L(Q, f ) and U (P, f ) U (Q, f ).
Proof. Let P = {x0 , x1 , x2 , ..., xk1 , xk , ..., xn } and Q = {x0 , x1 , x2 , ..., xk1 , z, xk , ..., xn }.

Then
L(P, f ) = m0 (x1 x0 ) + ... + mk (xk xk1 ) + ... + mn1 (xn xn1 )

m0 (x1 x0 ) + ... + mk (xk z) + mk (z xk1 ) + ... + mn1 (xn xn1 )

= L(Q, f )

where mk = inf f (x) and mk = inf f (x).

[z,xk ]

[xk1 ,z]

Lemma 3.1.6. If P1 and P2 be any two partitions, then

L(P1 , f ) U (P2 , f ).
Proof. Let Q = P1 P2 . Then Q is a refinement of both P1 and P2 . So by Lemma (3.1.8),
L(P1 , f ) L(Q, f ) U (Q, f ) U (P2 , f ).
Definition 3.1.7. Let P be the collection of all possible partitions of [a, b]. Then the
upper integral of f is
" b
f = inf{U (P, f ) : P P}
a

and lower integral of f is

"

b
a

f = sup{L(P, f ) : P P}.

An immediate consequence of Lemma (3.1.6) is

Lemma 3.1.8. For a bounded function f : [a, b] IR,
"

b
a

"

f.
a

Definition 3.1.9. Riemann integrability: f : [a, b] R is said to be Riemann integrable if

" b
" b
f=
f
a

and the value of the limit is denoted with

"

b
a

f (x)dx. We say f R[a, b].

Example 1: Consider f (x) = x on [0, 1] and the sequence of partitions Pn = {0, n1 , n2 , ..., n1
, nn }.
n
2

Then
L(Pn , f ) = 0

1
1 1
n11
+ + ... +
n n n
n n

1
[1 + 2 + ... + (n 1)]
n2
n(n 1)
=
2n2
=

#1
1
Thus lim L(Pn , f ) = . Hence from the definition 0 f (x)dx 12 . Similarly
n
2
1 1
2 1
n1
+ + ... +
n n n n
nn
1
= 2 [1 + 2 + ... + n]
n
n(n + 1)
=
2n2

U (Pn , f ) =

#1
1
Hence lim U (Pn , f ) = . Again from the definition 0 f (x)dx 12 .
n
2

Example 2: Consider f (x) = x2 on [0, 1] and the sequence of partitions Pn = {0, n1 , n2 , ..., n1
, nn }.
n
Then
$ %2
& n '2 1
1 1
2
1
U (Pn , f ) = 2 +
+ ... +
n n
n
n
n n
1
= 3 [1 + 22 + ... + n2 ]
n
n(n + 1)(2n + 1)
=
6n3
1
Thus lim U (Pn , f ) = . Similarly
n
3
$ %2
$
%2
1
1
1
n1
1
L(Pn , f ) = 0 +
+ ... +
n
n
n
n
n
1
= 3 [1 + 22 + ... + (n 1)2 ]
n
n(n 1)(2n 1)
=
6n3
1
Therefore, lim L(Pn , f ) = .
n
3
3

Hence from the definition

#b
a

f 1/3 and

Remark 3.1. In the above two examples

#b

f 1/3.

#1
0

f=

#1
0

f thanks to Lemma 3.1.8

The following example illustrates the non-integrability.

1, x Q,
Example 3: On [0, 1], define f (x) =
0, x Q.
Let P be a partition of [0, 1]. In any sub interval [xk1 , xk ], there exists a rational number
and irrational number. Then the supremum in any subinterval is 1 and infimum is 0.
#1
#1
Therefore, L(P, f ) = 0 and U (P, f ) = 1. Hence 0 f = 0 f .
Necessary and sucient condition for integrability

Theorem 3.1.10. A bounded function f R[a, b] if and only if for every > 0, there
exists a partition P such that
U (P , f ) L(P , f ) < .
Proof. : Let > 0. Then from the definition of upper and lower integral we have
"

b
a

"

b
a

f U (P , f ) L(P , f ) < ( by hypothesis).

Thus the conclusion follows as > 0 is arbitrary.

#b
: Conversely, since a f is the infimum, for any > 0, there exists a partition P1 such
that
" b

U (P1 , f ) <
f+ .
2
a
Similarly there exists a partition P2 such that
L(P2 , f ) >

"

b
a

f .
2

Let P = P1 P2 . Then P is a refinement of P1 and P2 . Hence

U (P , f ) L(P , f ) U (P1 , f ) L(P2 , f )
" b
" b

f+
f+
2
2
a
a
" b
" b
= (as f is integrable,
f=
f)
a

This complete the theorem.

///
Now it is easy to see that the functions considered in Example 1 and Example 2 are
integrable. For any > 0, we can find n (large) and Pn such that n1 < . Then
U (Pn , f ) L(Pn , f ) =

1
1
(n(n + 1) n(n 1)) = < .
2
2n
n

Similarly one can choose n in Example 2.

Remark 3.2. f : [a, b] R is integrable if and only if there exists a sequence {Pn } of
partitions of [a, b] such that
lim U (Pn , f ) L(Pn , f ) = 0.

Remark 3.3. Let S(P, f ) =

n
!
i=1

ing

f (i )(xi xi1 ), i [xi1 , xi ]. Then we have the follow-

m(b a) L(P, f S(P, f ) U (P, f ) M (b a).

In fact, one has the following Darboux theorem:
Theorem 3.1.11. Let f f : [a, b] IR be a Riemann integrable function. Then for a
given > 0, there exists > 0 such that for any partition P with P := max |xi xi1 | <
1in

, we have

|S(P, f )

"

f (x)dx| < .
a

Corollary: If f R[a, b], then for any sequence of partitions {Pn } with Pn 0, we
#b
#b
have L(Pn , f ) a f (x)dx and U (Pn , f ) a f (x)dx.
Remark 3.4. From the above theorem, we note that if there exists a sequence of partition
{Pn } such that Pn 0 and U (Pn , f )L(Pn , f ) 0 as n , then f is not integrable.
5

Problem: Show that the function f : [0, 1] IR

1 + x x Q
f (x) =
1 x x Q

is not integrable.
Solution: Consider the sequence of partitions Pn = {0, n1 , n2 , ...., nn = 1}. Then
1 1
2 1
n 1
) + (1 + ) + .... + (1 + )
n n
n n
n n
1
= 1 + 2 (1 + 2 + ... + n)
n
3

as n
2

U (Pn , f ) = (1 +

Now using the fact that infimum of f on [0, n1 ] is 1 n1 , though it is not achieved, we get
L(Pn , f ) = (1

1 1
2 1
n 1
1
) + (1 ) + .... + (1 ) as n .
n n
n n
n n
2

Hence f is not integrable.

1
Problem: Consider f (x) = on [1, b]. Divide the interval in geometric progression and
x
compute U (Pn , f ) and L(Pn , f ) to show that f R[1, b].
Solution: Let Pn = {1, r, r2 , ..., rn = b} be a partition on [1, b]. Then
U (Pn , f ) = f (1)(r 1) + f (r)(r2 r) + .... + f (rn1 )(rn rn1 )
1
1
= (r 1) + r(r 1) + .. + n1 rn1 (r 1)
r
r
= n(r 1)
1

= n(b n 1)

Therefore lim U (Pn , f ) = lim

n

bn 1
1
n

= lim

b n ln b( 1
)
n2
1
n2

= ln b.

Similarly
L(Pn , f ) = f (r)(r 1) + f (r2 )(r2 r) + ... + f (rn )(rn rn1 )
1
1
= (r 1) + .. + n rn1 (r 1)
r
r
n 1
= (b n 1)
r
1
= n(1 1/n )
b
1/n
b 1
= 1/n 1 ln b as n .
b n
Theorem 3.1.12. Suppose f is a continuous function on [a, b]. Then f R[a, b].
Proof. Let > 0. By Theorem 3.1.10, we need to show the existence of a partition P
such that
U (P, f ) L(P, f ) < .
Since f is continuous on [a, b], this implies f is uniformly continuous on [a, b]. Therefore
there exists > 0 such that
|x y| < |f (x) f (y)| <

.
(b a)

(3.1)

Now choose a partition P such that

sup |xk xk1 | < .

(3.2)

1kn

As f is continuous on [a, b] there exist xk , xk (xk1 , xk ) such that mk = f (xk ) and

Mk = f (xk ). By (3.2), |xk xk | < and hence by (3.1) |f (xk ) f (xk )| < (ba)
. Thus
U (P, f L(P, f ) =
=

n
!
k=1

n
!
k=1

(Mk mk )(xk xk1 )

(f (xk ) f (xk ))(xk xk1 )

n

(xk xk1 ) =
(b a) = .
(b a) k=1
(b a)
Therefore f R[a, b].
7

Integrability and discontinuous functions: We study the eect of discontinuity on

integrability of a function f (x).
Example: Consider the following function f : [0, 1] R.

1, x =
f (x) =
0, x =

1
2
1
2

Clearly U (P, f ) = 1 for any partition P . We notice that L(P, f ) will be less than 1. We
can try to isolate the point x = 12 in a subinterval of small length. Consider the partition
1
1
P = {0, 12 2 , 12 + 2 , 1}. Then L(P , f ) = ( ) + (1 ) = 1 . Therefore, for
2 2
2 2
given > 0 we have U (P , f ) L(P , f ) = . Hence f is integrable.
In fact we have the following theorem.
Theorem 3.1.13. Suppose f : [a, b] R be a bounded function which has finitely many
discontinuities. Then f R[a, b].
Proof follows by constructing suitable partition with sub-intervals of suciently small
length around the discontinuities as observed in the above example.

3.2

Properties of Definite Integral:

Property 1: For a constant c R,

"

cf (x)dx = c
a

Property 2: Let f1 , f2 R[a, b] . Then

"

(f1 + f2 )(x)dx =
a

"

"

f (x)dx.
a

f1 (x)dx +
a

"

f2 (x)dx.
a

Easy to show that for any partition P ,

U (P, f1 + f2 ) U (P, f1 ) + U (P, f2 )
L(P, f1 + f2 ) L(P, f1 ) + L(P, f2 )

Since f1 , f2 are integrable, for > 0 there exists P1 , P2 such that

U (P1 , f1 ) L(P1 , f1 ) <
U (P2 , f2 ) L(P2 , f2 ) <

(3.3)
(3.4)

Now taking P = P1 P2 , if necessary, we assume

U (P, f1 ) L(P, f1 ) < ,

U (P, f1 ) L(P, f2 ) <

(3.5)

Therefore, using (3.3)-(3.5) we get

U (P, f1 + f2 ) L(P, f1 + f2 ) U (P, f1 ) + U (P, f2 ) L(P, f2 ) L(P, f2 )
< + = 2.

Hence, f1 + f2 is integrable.
"

(f1 + f2 )(x)dx = lim S(Pn , f1 + f2 ) = lim

n

= lim

"

n
!
k=1

n
!
k=1

(f1 + f2 )(k )(xk xk1 )

f1 (k )(xk xk1 ) + lim

f1 (x)dx +
a

"

n
!
k=1

f2 (k )(xk xk1 )

f2 (x)dx
a

Property 3: If f (x) g(x) on [a, b]. Then

"

b
a

f (x)dx

"

g(x)dx.
a

"

First we note that m f (x) M implies m(b a)

f (x) M (b a). Then
a
" b
" b
" b
Property 1 and f (x) g(x) imply
(g f ) 0 or
g(x)dx
f (x)dx.
a
a"
a "
b
b
Property 4: If f R[a, b] then |f | R[a, b] and |
f (x)dx|
|f |(x)dx. Let

mk =

inf

[xk1 ,xk ]

|f |(x) and Mk =

sup |f |(x). Then we claim

[xk1 ,xk ]

Claim: Mk mk Mk mk
Proof of Claim: Note that for x, y [xi1 , xi ],
|f |(x) |f |(y) |f (x) f (y)| Mi (f ) mi (f ).
Now take supremum over x and infimum over y, to conclude the claim.

Now since f is integrable, there exists partitions {Pn } such that lim U (Pn , f )L(Pn , f ) =
n
0. i.e.,
n
!
lim
(Mk mk )(xk xk1 ) = 0.
n

This implies

lim

k=1

n
!

(Mk mk )(xk xk1 ) = 0.

k=1

Hence |f | is integrable. Note that |f | f |f |. Thus by Property 3 we get

"

b
a

|f |(x)dx

"

f (x)dx

"

b
a

|f |(x)dx = |

"

b
a

f (x)dx|

"

b
a

|f |(x)dx.

Property 5: Let f be bounded on [a, b] and let c (a, b). Then f is integrable on [a, b]
if and only if f is integrable on [a, c] and [c, b]. In this cases
"

f (x)dx =
a

"

f (x)dx +
a

"

f (x)dx.
c

Let f be integrable on [a, b]. For > 0, there exists partition P such that
U (P, f ) L(P, f ) < .

(3.6)

With out loss of generality we can assume that P contain c. (otherwise we can refine
P by adding c and the dierence will be closer than before) Let P1 = P [a, c] and
P2 = P [c, b]. Then P1 and P2 are partitions on [a, c] and [c, b] respectively. Also by
(3.6), U (P1 , f ) L(P1 , f ) < and U (P2 , f ) L(P2 , f ) < . This implies f is integrable
on [a, c] and [c, b]. Conversely, suppose f is integrable on [a, c] and [c, b]. Then for > 0,
there exists partitions P1 of [a, c] and P2 of [c, b] such that U (P1 , f ) L(P1 , f ) < 2 and
U (P2 , f ) L(P2 , f ) < 2 . Now take P = P1 P2 . Then U (P, f ) L(P, f ) < . So by
Remark 3.3, there exists {Pn } such that
"

f (x)dx = lim S(Pn , f ) = lim

n

Pn [a,c]

"

Pn

f (k )(xk1 xk )

f (k )(xk xk1 ) +

f (x)dx +
a

"

f (x)dx
c

10

Pn [c,b]

f (k )(xk xk1 )

Example: Consider the following function f : [0, 1] R.

1 x =
f (x) =
0 x =

1
n
1
n

for some n N, n 2

Then f is Riemann integrable.

< . Note that f (x) has only finitely many

N
2
discontinuities in [ N1 , 1] say 1 , 2 , ..., r . Define the partition P as
Solution: Let > 0. Choose N such that

P = {0,

, 1 , 1 + , ..., r , r + , 1}.
N
4r
4r
4r
4r

Since r is the last discontinuity, f = 0 in [r +

, 1].
4r

Now L(P , f ) = 0 and

+
+
+ ... +
+ 0 (1 r )
N
2r 2r
2r
4r
1

=
+ < .
N
2

U (P , f ) = 1

Alternatively, using the approach of isolating the discontinuous points, we may take the
Partition with length of 2k at each discontinuous points xk , k = 1, 2, 3, 4, ....... Then using
+ 1
the fact that
converges, one can prove that f is integrable.
2k
Example: Consider the following function f : [0, 1] R.
f (x) =

Then f is not Riemann integrable.

sin 1

xQ
x Q

Solution: We will show that f is not integrable on a sub interval of [0, 1]. Consider the
f on the subinterval I1 = [ 2 , 1]. Clearly L(P, f ) = 0 for any partition P of I1 because
f (x) 0 in the sub interval [ 2 , 1]. Let Mk be the Supremum of f on subintervals [xk1 , xk ]
of [ 2 , 1]. Also the minimum of Mk s is sin 1. Therefore,
U (P, f ) =

n
!
k=1

f (k )(xk xk1 ) > (1

11

2
) sin 1.

Hence U (P, f ) L(P, f ) can not be made less than for any < (1 2 ) sin 1.

///

Mean Value Theorem

Theorem 3.2.1. Let f (x) be a continuous function on [a, b]. Then there exists [a, b]
such that
" b
f (x)dx = f ()(b a).
a

Proof. Let m = min f (x) and M = max f (x). Then by Property 3, we have
x[a,b]

x[a,b]

m(b a)
i.e.

"

f M (b a),

1
m
(b a)

"

b
a

f M.

Now since f (x) is continuous, it attains all values between its maximum and minimum
" b
1
values. Therefore there exists [a, b] such that f () =
f.
(b a) a
Fundamental Theorem
Theorem 3.2.2. Let f (x) be a continuous function on [a, b] and let (x) =

"

f (s)ds.
a

Then is dierentiable and (x) = f (x).

" x+x
(x + x) (x)
1
Proof. As
=
f (s)ds, By Mean value theorem, there exists
x
x x
[x, x + x] such that
" x+x
f (s)ds = xf ().
x

(x + x) (x)
= lim f ().
x0
x0
x
f ( lim ) = f (x). Thus (x) = f (x).
Therefore lim

Since f is continuous, lim f () =

x0

Remark 3.5. It is always not true that

#b
a

x0

///

f (x)dx = f (b) f (a).

For example, take f (x) = x2 sin x1 for x = 0 and f (0) = 0. Then f is dierentiable on
[0, 1]. Here the derivatives at the end points are the left/right derivatives. It is easy to
check that f (x) = 2x sin x12 x2 cos x12 for x (0, 1) and f (0) = 0. Therefore f is not

12

bounded and so not integrable.

Definition 3.2.3. A function F (x) is called anti-derivative of f (x), if F (x) = f (x).
Second Fundamental Theorem:
Theorem 3.2.4. Suppose F (x) is an anti- derivative of continuous function f (x). Then
#b
f (x)dx = F (b) F (a).
a
Proof. By First fundamental theorem, we have
d
dx

"

f (s)ds = f (x).
a

#x
Also F (x) = f (x). Hence a f (s)ds = F (x) + c for some constant c R. Taking x = a,
#b
we get c = F (a). Now taking x = b we get a f (x)dx = F (b) F (a).
///
Change of Variable formula

Theorem 3.2.5. Let u(t), u (t) be continuous on [a, b] and f is a continuous function on
the interval u([a, b]). Then
"

f (u(x))u (x)dx =
a

"

u(b)

f (y)dy.
u(a)

Proof. Note that f ([a, b]) is a closed and bounded interval. Since f is continuous, it has
#x
primitive F . i.e., F (x) = a f (t)dt. Then by chain rule of dierentiation, dtd F (u(t)) =
F (u(t))u (t). i.e., F (u(t)) is the primitive of f (u(t))u (t) and by Newton-Leibnitz formula,
we get
"
b

f (u(t))u (t)dt = F (u(b)) F (u(a)).

On the other hand, for any two points in u([a, b]), we have (by Newton-Leibnitz formula)
"

B
A

f (y)dy = F (B) F (A).

Hence B = u(b) and A = u(a).

#1
Example: Evaluate 0 x 1 + x2 dx.
13

Taking u = 1 + x2 , we get u = 2x and u(0) = 1, u(1) = 2. Then

"

3.3

x 1+

x2 dx

1
=
2

"

2
1

1 2
1 2
udu = u 3 2u=1 = (2 3 1).
3
3

Improper Integrals

In the previous section, we defined Riemann integral for functions defined on closed and
bounded interval [a, b]. In this section our aim is to extend the concept of integration to
the the following cases:
1. The function f (x) defined on unbounded interval [a, ) and f R[a, b] for all
b > a.
2. The function is not defined at some points on the interval [a, b].
We first consider
Improper integral of first kind: Suppose f is a bounded function defined on [a, )
and f R[a, b] for all b > a.
Definition 3.3.1. The improper integral of f on [a, ) is defined as
"

f (x)dx := lim

"

f (x)dx.
a

If the limit exists and is finite, we say that the improper integral converges. If the limit
goes to infinity or does not exist, then we say that the improper integral diverges.
Examples:
#
#b
1. (i) 1 x12 dx = limb 1 x12 dx = limb 1 1b = 1.
# dx
# b dx

b
2. (ii) 0 1+x
2 = limb 0 1+x2 = limb arctan x 0 = 2 .

Theorem 3.3.2. Comparison test: Suppose 0 f (x) (x) for all x a, then
#
#
1. a f (x)dx converges if a (x)dx converges.
#
#
2. a (x)dx diverges if a f (x)dx diverges.
#x
#x
Proof. Define F (x) = a f (t)dt and G(x) = a g(t)dt. Then by properties of Riemann
integral, 0 F (x) G(x) and we are given that lim G(x) exists. So G(x) is bounded.
x

F is monotonically increasing and bounded above. Therefore, lim F (x) exists.

x

Examples:
14

1.

#
1

dx
x2 (1+ex )

. Note that

x3
dx.
x+1

Note that

1
x2 (1+ex )

x3
x+1

x2
2

<

1
x2

and

#
1

dx
x2

converges.

x2 dx diverges.
#
Definition 3.3.3. Let f R[a, b] for all b > a. Then we say a f (x)dx converges
#
absolutely if a |f (x)|dx converges.
2.

>

on [1, ) and

In the following we show that absolutely convergence implies convergence of improper

integral.
#
#
Theorem 3.3.4. If the integral a |f (x)|dx converges, then the integral a f (x)dx converges.
#
Proof. Note that 0 f (x) + |f (x)| 2|f (x)|. So the improper integral a f (x) +
#
|f (x)|dx converges by comparison theorem above. Also a |f (x)| converges. Therefore,
#
#
#
f (x)dx = a f (x) + |f (x)|dx a |f (x)|dx also converges.
///
a
The converse of the above theorem is not true. For example take the integral
This integral does not converge absolutely. Indeed,
"

!
| sin x|
dx =
x
n=1

"

(n+1)
n

| sin x|
dx
x

" (n+1)

!
1

| sin x|
n
n
n=1
"

!
1
2!1
=
sin x =
.
n 0
n=1 n
n=0

On the other hand, by integration by parts,

lim

"

b
1

sin x
dx = lim
b
x

"

1
d(1 cos x)
1 x
$
%
" b
(1 cos b
1 cos x
= lim
+
dx
b
b
x2
1

It is not dicult to show that the limits on the right exist.

Examples:
# x
#
x
1. 1 sin
dx. Easy to see that | sin
| | x13 | and 1
x3
x3
15

dx
x3

converges.

sin x
dx.
x

2.

ex sin x
.
log(1+x)

Here first note that limx0

at x = 0. For x > 10(say):
0

ex sin x
log(1+x)

= 1. Therefore the integral is proper

ex
2
|f (x)|
< ex ex
log(1 + x)
Hence the integral

ex sin x
10 log(1+x)

converges by comparison theorem.

Theorem 3.3.5. Limit comparison test: Let f (x), g(x) are defined and positive for
f (x)
all x a and lim
= L.
x g(x)
#
#
1. If L (0, ), then the improper integrals a f (x)dx and a g(x)dx are either both
#
#
convergent or both divergent. i.e., a f (x)dx converges a g(x) converges.
#
#
#
2. If L = 0, then a f (x)dx converges if a g(x)dx converges. i.e, a g(x)dx con#
verges = a f (x)dx converges.
#
#
#
3. If L = , then a f (x)dx diverges if a g(x)dx diverges. i.e., a g(x)dx diverges
#
= a f (x)dx diverges.
Proof. From the definition of limits, for any > 0, there exists M > 0 such that
x M = L <

f (x)
< L + .
g(x)

Thus for x M , we have (L )g(x) < f (x) < (L + )g(x).

Now in case (1), since L > 0, we can find > 0 such that L > 0. Using the property 3,
it is enough to prove the convergence/divergence for x large, say x M . In this interval,
we have the comparison (L )g(x) < f (x) < (L + )g(x). Now integrating this, we get
the result.
In case (2), we have f (x) < (L + )g(x). Again, integrate on both sides.
In case (3) by the definition, for every M > 0, there exists, R such that f (x) > M g(x)
for all x > R. Now the result follows similar to (1) and (2).
Examples:
#
#
(x)
1
1. 1 dx
. Take f (x) = x+1
and g(x) = 1x . Then limx fg(x)
= 1 and 1 g(x)dx
x+1
#
diverges. So by above theorem, 1 f (x)dx diverges.
# dx
#
f (x)
1
1
2. 1 1+x
g(x)dx
2 . Take f (x) = 1+x2 and g(x) = x2 . Then limx g(x) = 1 and
1
#
converges. So by above theorem, 1 f (x)dx converges.
16

3.

x
dx.
cosh x
f (x)
limx g(x)
0

x
x
Let f (x) = cosh
= e2xe
. So choose g(x) = xex . Then
2x +1 xe
x
#
= 2 and 0 g(x) converges.

Improper integrals of second kind

Definition 3.3.6. Let f (x) be defined on [a, c) and f R[a, c ] for all > 0. Then
we define
" c
" c
f (x)dx = lim
f (x)dx.
0

#b

Then a f (x)dx is said to converge if the limit exists and is finite. Otherwise, we say
#b
improper integral a f (x)dx diverges.
" 1
" 1

dx
dx
= lim
= lim 2(1 ) = 2.
Example:
x 0
x 0
0
Suppose a1 , a2 , ....an are finitely many discontinuities of f (x) in [a, b]. Then
"

f (x)dx =
a

"

a1

f (x)dx +
a

"

a2

f (x)dx +
a1

"

a3

f (x)dx + .... +
a2

"

f (x)dx
an

If all the improper integrals on the right hand side converge, then we say the improper
integral of f over [a, b] converges. Otherwise, we say it diverges.
The following comparison and Limit comparison tests can be proved following similar
arguments:
Theorem 3.3.7. (Comparison Theorem:) Suppose 0 (x) f (x) for all x [a, c) and
are discontinuous at c.
#c
#c
1. If a f (x)dx converges then a (x)dx converges.
#c
#c
2. If a (x)dx diverges then a f (x)dx diverges.
Theorem 3.3.8. (Limit comparison theorem:) Suppose 0 < f (x), g(x) be continuous in
(x)
[a, c) and limxc fg(x)
= L. Then
#c
#c
1. If L (0, ). Then a f (x)dx and a g(x)dx both converge or diverge together.
#c
#c
2. If L = 0 and a g(x)dx converges then a f (x)dx converges.
#c
#c
3. If L = and a g(x)dx diverges then a f (x)dx diverges.
17

Proof. From the definition of limit, for each > 0, there exists > 0 such that
x (c , c) = (L )g(x) < f (x) < (L + )g(x)
Rest of the proof follows in the similar lines theorems on first kind, by choosing < L.
Transforming improper integrals:
Sometimes improper integrals may" be transformed into proper integrals. for example
3
dx
1
consider the improper integral I =
. Taking the transformation y =
,

3x
x 3x
1
"
dy

we get I =
. This is an improper integral of first kind. Instead, if we choose
1/2 y 3y 1
" 2
2udu

the transformation 3 x = u2 then I =

, which is a proper integral.
u 3 u2
0
Remark 3.6. It is important to note that the symmetric limit could be convergent but
the limit may not exist. For example,
"

1
1

dx
=
x3

"

0
1

dx
+
x3
" 1

"

1
0

dx
x3

dx
= lim
+ lim
1 0 1
x3 2 0
1
1
1
=
lim ( 2 2 ),
2 1 ,2 0 1 2

"

1
2

dx
x3
(3.7)

where 1 , 2 0 as n . It is easy to see that if one takes 1 = 2 , then the limit exits
1
1
and is equal to 0. But if one takes 1 = (n+1)
2 , 2 = n2 , then the above limit in (3.7) does
not exist. So through dierent sequences, we are getting dierent limits. By now, by our
familiarity with existence of limits, we say integral diverges.
Gamma and Beta functions:
Consider the Gamma function defined as improper integral for p > 0,
(p) =

"

xp1 ex dx

This integral is improper of second kind in the neighbourhood of 0 as xp1 goes to infinity
as x 0 (when p < 1). Since the domain of integration is (0, ), the integral is improper
18

of first kind. To prove the convergence, we divide the integral into

(p) =

"

p1 x

e dx +

=I1 + I2

"

xp1 ex dx

f (x)
To see the convergence of I1 we take f (x) = xp1 ex and g(x) = xp1 , then lim
=1
x0 g(x)
" 1
and
xp1 dx converges. To see the convergence of I2 , take f (x) = xp1 ex and
0

#
f (x)
= lim x2+p1 ex = 0 and 1
x g(x)
x
(2) of limit comparison theorem, the integral converges.
g(x) =

1
.
x2

Then lim

1
dx
x2

converges. Hence by

Next we consider the Beta function defined as improper integral for p > 0, q > 0,
(p, q) =

"

1
0

xp1 (1 x)q1 dx

If p > 1 and q > 1, then the integral is definite integral. When p < 1 and/or q < 1, this
integral is improper of second kind at 0 and/or 1. To prove the convergence, we divide as
before
"

x
0

p1

(1 x)

q1

dx =

"

1/2

p1

(1 x)

q1

dx +

"

1
1/2

xp1 (1 x)q1 dx

= I1 + I2 .
To see the convergence of I1 , take f (x) = xp1 (1 x)q1 and g(x) = xp1 . Then
" 1/2
f (x)
q1
lim
= lim (1 x)
= 1 and
xp1 dx converges. Similarly, for convergence
x0 g(x)
x0
0
of I2 , we take f (x) = xp1 (1 x)q1 and g(x) = (1 x)q1 .
Some identities of beta and gamma functions:
#
1. (1) = 0 ex dx = 1.
2. ( + 1) = ().
Integration by parts formula implies,
( + 1) =

"

x e

= (x e

19

)|
0

"

x1 ex dx = ().

Therefore, (m + 1) = m! m IN .

3. ( 12 ) = .
$

1
( )
2

%2

"

=4

"

=4

"
"

eu ev dudv,

take u = r cos , v = r sin ,

0
/2

er rdrd =

4. (m, n) = (n, m). Substituting t = 1 x in the definition of (m, n), we get

(m, n) =

5. (m, n) = 2

"

"

tn1 (1 t)m1 dt = (n, m)

/2

sin2m1 cos2n1 d
0

Taking x = sin2 in (m, n), we get

(m, n) =
6. (m, n) =

"

cos

2m2

sin

2n2

cos sin d = 2

(m)(n)
.
(m+n)

"

"

2/3 x

Problem: Evaluate (i)

x e
dx (ii)
x 2 (1
0
0

For (i), take t = x, then the given integral becomes

"

4/3 t

e 2tdt = 2

For (ii), again take t =

2

"

t (1 t)

1/2

et t7/3 dt = 2(

/2

cos2m1 sin2n1 d.
0

x) 2 dx

10
56
) = (1/3).
3
27

x, then the integral becomes

1
3

"

"

tdt = 2

"

t4 (1 t)1/2 dt = 2(5, 3/2) = 2

(5)(3/2)
512
=
(13/2)
3465

Cauchy Principal Value:

#
Consider the improper integral I = 0 sin xdx. It is easy to see from the definition that
#0
I = lima (1 cos a) which does not exist.Similarily, sin xdx does not exist. But
lim

"

sin xdx
c

20

exists and is equal to 0. Though the improper integral does not exist, this symmetric
limit exists. This is called Cauchy Principal value of improper integral
Definition 3.3.9. The Cauchy Principal value of improper integral of first kind is defined
as
"
"

CPV

f (x)dx = lim

f (x)dx

For the improper integral of second kind, with c (a, b) as point of discontinuity of f (x)
as
" b
" c
" b
CPV
f (x)dx = lim
f (x)dx +
f (x)dx
0

c+

Examples:
#
First kind: x2n+1 dx for all n = 1, 2, 3, .... In this case it is easy to check that
" a
#
#0
lim
x2n+1 = 0. But the the improper integrals 0 x2n+1 and x2n+1 does not
a a
converge.
Second kind:

#1

x(2n+1) dx, for n = 1, 2, 3, .... Simply evaluate

"

(2n+1)

"

x(2n+1)

to see that the the limit is 0.

Integrals dependent on a Parameter
Cosider an integral
I() =

"

f (x, )dx
a

where the integrand is depend on the parameter . At times we can dierentiate under
the integral sign to evaluate the integral. It is sometimes not possible and leads to wrong
#
assertions. For example, we know that I = 0 sinx x = 2 . It is easy to notice with change
#
of variable formula, taking tx = y, that I = I(t) = 0 sin(tx)
= 2 . Now dierentiating
x
#

this, taking derivative inside integral, we get I (t) = 0 cos(tx)dx = 0, which doesnt
make sense.
Here we have a theorem, which explains under which conditions we can do the dierentiation under integral sign.
Theorem 3.3.10. Suppose,
21

d
1. Suppose f, d
f (x, ) are continuous functions for x [a, b] and in an interval of
containing 0 .
d
2. |f (x, )| a(x), | d
f (x, )| b(x) such that a, b are integrable on [a, b].

Then I is dierentiable, and

I () =

"

b
a

d
f (x, )dx.
d

Proof. .
d
I( + ) I()
I() = lim
0
d

," b
1
= lim
(f (x, + ) f (x, ))dx
a
d
d
Now by Taylors theorem, f (x, + ) f (x, ) = d
f (x, + ). Since d
f (x, )
d
d
is continuous, we have d
f (x, + ) = d
f (x, ) + , where 0 < < 1 and 0 as
0. Thus
" b
" b
d
d
d
I() = lim
f (x, ) + =
f (x, )dx
0 a d
d
a d

///
In fact the following holds.
Newton-Leibnitz Formula:
Let h(x) =

"

b(x)

f (x, t)dt. Then h (x) =

a(x)

Examples:

"

"

b(x)
a(x)

df
(x, t)dt+f (x, b(x))b (x)f (x, a(x))a (x)
dx

sin x
dx.
x #
0

By the above formula, I () = 0 ex cos xdx =

#
C. Also I(0) = 0 ex sin 0x = 0. Hence C = 0.
"
2. Test the convergence and evaluate the integral
1. Evaluate I() =

|I| e

ex

t2 /2

"

1
.
1+2

e 2 (t

Therefore, I() = arctan +

2 x2 )

cos(tx)dx.

|e

x2

cos(tx)|dx C
22

"

1 1
2
ex dx = ( ).
2 2

Hence the integral converges. By Newton Leibnitz formula,

Ia (t)

=
=

"
"

=e

e 2 (t
0

2 x2 )

(t cos tx x sin tx)dx

1 (t2 x2 )
e2
sin txdx
x

1 2
(t a2 )
2

sin at

Therefore, I (t) = lima Ia (t) = 0. Now note that I(0) =

.
Hence I(t) = 2 .

3.4

#
0

2 /2

ex

dx =

1 (1/2).
2

Approximation of definite integrals

Suppose we are given a function f : [a, b] IR that is integrable on [a, b] and has its Taylor
series in an interval containing [a, b]. Then we can approximate the definite integral using
+
n
the Taylor series. Let f (x) =
k=0 an t for |t| < R. Then we know from the termby-term integration theorem,
"

f (x)dx =
0

"
!
k=0

t
n

an x dx =
0

an

k=0

tn+1
|t| < R
n+1

#1
Example: Find the approximate value of the definite integral 0 x2 sin(x2 )dx with error
less than 104 .
We note that if the nt h term has k zeros in the first k decimals, then the approximation
is with error 10k1 .
"

1
2

x sin(x )dx =
0

=
=

"

1!

!
n=0

!
n=0

Hence

"

x2 sin(x2 ) =
0

n=0

(1)n

x4n+4
(2n + 1)!

(1)n x4n+5 1
(4n + 5)(2n + 1)! 0
(1)n
(4n + 5)(2n + 1)!

1
1
1
1

5 54 13 5! 17 7!

Since the 4th term has 4 zeros in the first 4 decimal places. So the approximation has
error less than 104 .
23

Application: A pendulum of mass m suspended by a inextensible string of length L is

released from initial position of angle (from equilibrium position). Let be the angular
coordinate. Then the quarter-period time /4 is given by

=
4

L
4g

"

sin2 ( ) sin2 ( )
2
2

By taking the transformation 2 defined as sin =

to
"
/2

% 12

sin(/2)
,
sin(/2)

d.
this integral is transformed

(1 k 2 sin2 ) 2 d, k = sin2 (/2).

This integral is an improper integral which is dicult to evaluate. We can use Taylor
series to find approximate value as described in the previous example.
Now suppose the function is not continuous but only integrable. Then we can still approximate the definite integral using Riemann sums. Recall from section 3.1, that if the
function is integrable then definite integral can be computed as
"

f (x)dx = lim
a

n
!

f (k )xk

k=1

where k [xk1 , xk ] and xk = xk xk1 . When we take equal partitions, we get

xk = ba
. We denote y0 = f (x0 ), y1 = f (x1 ), ....yn = f (xn ). When we take k = xk1
n
and k = xk we get the sums
((y0 + y1 + .... + yn1 ) x
(y1 + y2 + .... + yn ) x
Each of these sums is a Riemann sum and converges to definite integral
n . So we write Rectangular formula
"

b
a

"

f (x)dx (y0 + y1 + .... + yn1 )

b

f (x)dx (y1 + y2 + .... + yn )

24

ba
n

ba
.
n

#b
a

f (x)dx when

These approximations involve the area of rectangles with sides xk1 xk and height yk1 or
yk . Instead of taking the end points, we could take mid-points of the interval for better
approximation to obtain Trapezoidal Rule:
"

b
a

y0 + y1
y1 + y2
yn1 + yn
f (x)dx
x +
x + .... +
x
2
2
2
$
%
b a y0 + yn

+ y1 + y2 + ... + yn1
n
2

As the name suggests, the first two formulas involve the sum of areas of rectangles and
trapezoidal rule consists of areas of trapezoids. This is also seen as piecewise linear
approximation of the function y = f (x). (Explain Geometrically)!
We can also take quadratic approximation or with exact values at three points. Approximations like this and estimation of errors is part of Numerical integration theory which
is beyond the scope of this course.

25

3.5

Applications to Area, Arc length, Volume and Surface area

Suppose f (x) 0 on [a, b]. Then it is clear from the definition of Definite integral that
the area under the curve y = f (x) can be approximated by Riemann sums. i.e.,
A
=

n
!
k=1

f (i )(xi xi1 )

"

b
a

f (x)dx as n .

Similarly, the area bounded by the curves y = f (x) and y = g(x) where f (x) g(x) on
[a, b] is
" b
A=
(f (x) g(x))dx.
a

Example: Find the area bounded by y = x2 and y 2 = x

The curves interest at x = 0, 1. The upper curve is y 2 = x and lower curve is y 2 = x. So

by the above formula
" 1

2 1
1
A=
( x x2 )dx = =
3 3
3
0
" 1
1

One can also find by integrating along y: A =

( y y 2 )dy = .
3
0
Polar coordinates
A point (x, y) on the xy-plane is assigned polar coordinates (r, ) if the point is at a dis.
tance r = x2 + y 2 from the origin on the ray at an angle with positive x-axis. We allow
r negative with convention: (r, ) = (r, + ). Each point on the plane has infinitely
many representations in polar form, for example (1, 0) is at a distance of 1 units from the
origin on the x-axis. So it can be represented in polar form also as (r, ) = (1, 0). Also it is
same as (1, 2n), n N and (1, ). Each point (r, ) is same as (r, +2n) for all n N.
Example: The point (2, /6) can also be represented by (2, 5
) and (2, 7
)
6
6
Relation with cartesian coordinates:
We often use the following relations:
1. Given the polar coordinates (r), we can write the cartesian coordinates using x =
r cos , and y = r sin .
2. Given the cartesian coordinates (x, y), we can write polar coordinates using r =
.
x2 + y 2 , and = tan1 ( xy )
26

Circles and Straight lines:

1. The circle x2 +y 2 = a2 in cartesian coordinates, using (1) above, r2 cos2 +r2 sin2 =
a2 which is r = a.
2. The circle (x a)2 + y 2 = a2 = x2 + y 2 2ax = 0, again by (1) above we get
r2 2ar cos = 0 = r = 2a cos .
3. The circle x2 + (y a)2 = a2 = x2 + y 2 2ay = 0, again by (1) above we get
r2 2ar sin = 0 = r = 2a sin .
4. The straight line y = mx is = tan1 m
5. The straight line x = a is r = a sec and y = b is r = b csc .
Symmetry in polar coordinates: The symmetry of the graph of the function in polar coordinates helps one to plot/trace the graph. There are three types of symmetry
principles.
1. For (r, ) on the graph, suppose (r, ) is also on the graph. Then the graph is
symmetric about x axis.
2. For (r, ) on the graph, suppose (r, ) is also on the graph. Then the graph is
symmetric about y- axis.
3. For (r, ) on the graph, suppose (r, + ) is also on the graph. Then the graph is
symmetric about the origin.
Examples:
1. (leminiscate): Consider the function r2 = cos 2. If (r, ) is on the graph, then
r2 = cos 2() = cos 2 implies (r, ) is also on the graph. So the graph is
symmetric about x axis.
Again, r2 = cos 2( ) = cos(2 2) = cos 2 implies (r, + ) is also on the
graph. Therefore, graph is symmetric about y-axis.
We can also see that (r, + ) is also on the graph. So the graph is also symmetric
about the origin.
Hence it is enough to trace the curve in the first quadrant. Now since r2 0, the
domain of in the first quadrant is [ 4 , 4 ]. Also one can see by the derivative test
that = 0 is a point of local maxima(see figure 1).

27

r2 = cos 2

0 2

Figure 1: leminiscate
r = 1 cos

0 2

Figure 2: Cardioid

2. (Cardioid): Consider the function r = 1cos . Then if (r, ) graph = (r, )

graph. So the graph is symmetric with respect to x axis. So it is enough to trace
the curve for 0 . Again by derivative test we see that = 0 is a point of
minimum and = is point of maximum(see figure 2).

Area in polar coordinates: Let a region be bounded by the rays = and = and
the curve r = f (). We approximate the region with n non-overlapping circular sectors
based on the partition P of angle [, ]. The typical sector has radius rk = f (k ) and
k
central angle of radian measure k . Its area is
times the area of a circle rk . i.e.,
2
1
1
Ak = rk2 k = f (k )2 k
2
2
+
The area of the region is approximately nk=1 Ak . Taking n so that P 0, we

28

get
A=

"

1
1
f ()2 d =
2
2

"

r2 d.

Example: Find the area of the region enclosed by the cardioid r = 2(1 cos ).
Solution: From the graph discussed above, the range of is from 0 to 2. Therefore,
the area is
"
" 2
1 2 2
A=
r d =
(3 + cos 2 4 cos )d = 6.
2 0
0

Arc length
Consider a curve defined by y = f (x) between x = a and x = b. For example y = sin x
between x = 0 and . The length of this curve can be approximated by sum of lengths
of straight lines connecting (0, 0) (/4, sin(/4)) (/2, sin(/2)) (, 0). The arc
length s is approximately
/

1
( ) 2 + ( )2 +
4
2

2
1
) + (1 )2 +
4
2

( )2 + 1.
2

This approximation becomes better and better as we refine the partition P = {0, /4, /2, }.
For a given curve defined by function y = f (x) between x = a, b, we consider the partition P = {a = x0 , x1 , x2 , ....xk1 , xk , ...xn = b}. Then the length of this curve may be
approximated by the formula
n
!
.
s
(xi xi1 )2 + (f (xi ) + f (xi1 ))2
i=1

n
!

i=1
" b
a

1+

f (xi ) f (xi1 )
xi xi1

%2

(xi xi1 )

1 + (f (x))2 dx as n

The following two formulas are used for finding the Arc length or length of curve:
1. For a function y = f (x) between x = a and x = b
s=

"

b
a

1+

29

df
dx

%2

dx.

2. For a function x = f (y) between y = c and y = d

"

s=

1+

df
dy

%2

dy.

Parametric form: Suppose if an arc is defined in the parametric form x = x(t), y = y(t)
between t = T1 and t = T2 . Then we note from above approximation, that s may be
approximated by taking the partition P = {T1 = t0 , t1 , ...., tn = T2 } and
s

n
!

/$

i=1
" T2
T1

xi xi1
ti ti1

%2

yi yi1
ti ti1

%2

(ti ti1 )

.
(x (t))2 + (y (t))2 dt as n .

Example: Find the arc length of the curve defined by x = 2 cos2 , y = 2 cos sin ,
0 .
Solution: This curve is a circle with radius 1 at (1, 0). So the answer should be 2.
Applying formula
s=

"

=2

"

x ()2

y ()2

"

d = 2

1
(2 cos sin )2 + (cos2 sin2 )2 d

.
cos4 + 2 cos2 sin2 + sin4 d = 2
2

Example 2: Find the arc length of an arc of x 3 + y 3 = 1.

Solution: Take the parametrization: x = cos3 t, y = sin3 t, 0 t 2 . Then
$

dx
dt

%2

Hence the arc length is

l=3

"

dy
dt

%2

= 9 cos2 t sin2 t.

cos t sin t dt =

3
2

Problem: Find the approximate value of the length of ellipse x = a cos t, y = b sin t, 0
t 2 when a = 1, e = 1/2.
solution: By the arc length formula,
l = 4a

"

/2
0

1 e2 cos2 tdt

30

where e is ellipses eccentricity. This integral is non-elementary except when e = 0 or

1. The integrals in this form are called elliptic integrals. We can use Trapezoidal rule to
evaluate the value when a = 1 and e = 1/2. The answer with n = 10 is l = 5.870.
Suppose the curve is given in polar form r = f (), . Then by taking the
parametrization x = r cos = f () cos and y = r sin = f () sin with [, ], we
get
dx
dy
= f () cos f () sin ,
= f () sin + f () cos .
d
d
Hence the arc length is
" .
l=
f 2 () + (f )2 ()d.

Application to Work done

Suppose the force f (x) depends on position x is along a straight line from x = a to x = b.
Let n N, x = ba
and xi = a + ix for i = 1, 2, ..., n. Then the work done in moving a
n
particle under the force f (x) from xk1 to xk is approximately Wk = f (xk )x. The total
+
work done (in moving from a to b) approximately is W ni=1 f (xk )x, xk [xk1 , xk ].
#b
Taking n we get the total work done as W = a f (x)dx.

Example: Find the work required to compress a spring from its natural length of 1 foot
to a length of 0.75 foot if the force constant is k = 16 kg/foot.
Solution: Hooks law ways that the force it takes to stretch or compress a spring x length
nits from its natural length is proportional to x. i.e.,F = kx, k is constant measured in
force units per unit length.
Suppose the given springs is placed on the x axis. It is fixed at x = 1 and movable end
at the origin. From the above formula, the force required to compress the spring from 0
to x with the formula F = 16x. To compress the spring from 0 to 0.25 ft, the force must
increase from F (0) = 0 to F (0.25) = 16 0.25 = 4 foot-kg. Therefore, the work done by
F over this interval is
"
0.25

W =

16xdx = 0.5 f t kg.

Volume of symmetrical objects

Method of Slicing:
Consider a solid lying alongside some interval [a, b] of the x-axis. For each x let A(x) be the
area of the cross section (of the solid) obtained by cutting it with a plane perpendicular
31

to the x-axis at x. We divide the interval into n subintervals [xi1 , xi ]. The planes that
are perpendicular to the x-axis at the points x0 , x1 , x2 , ..., xn divide the solid into n slices.
If the cross section between [xi1 , xi ] changes little bit along the that subinterval, then
it can be approximated by a cylinder of height xi xi1 with base A(xi ), xi [xi1 , xi ].
So the volume of the slice is Vi = A(xi )(xi xi1 ). Then the volume of the solid can be
approximated as
V

n
!

Vi =

i=1

n
!

A(xi )(xi

i=1

xi1 )

"

A(x)dx
a

as n . Now this can be done along any axis, say y-axis. In this case we get the
formula:
" b
V =
A(y)dy.
a

Solid of Revolution: Consider the area between the function y = f (x), x [a, b] and
x-axis. By revolving this area along x axis, we obtain a solid which is called solid of
revolution. It is easy to see that for this solid, the cross section is disc of radius f (x)
and the area of cross section A(x) is equal to [f (x)]2 . Hence the volume is
V =

"

A(x)dx =
a

"

f 2 (x)dx.
a

For example, take a cone of radius r and height h. Then this cone can be obtained by
revolving y = rx
about x axis between x = 0, h. Then the volume is
h
V =

"

h
0

r 2 x2
r2 h
dx
=
.
h2
3

Suppose the area revolved is bounded by two curves y = f (x) 0 and y = g(x) 0, with
f (x) g(x). Then each cross section looks like washer with outer radius r1 (x) = f (x)
and inner radius r2 (x) = g(x). The area of the cross section is (f 2 (x) g 2 (x)). The
volume of the solid is
" b
" b
2
2
V =
(r1 r2 )(x)dx =
(f 2 (x) g 2 (x))dx.
a

If the revolution is performed about y axis. Then

V =

"

b
a

(f 2 (y) g 2 (y))dy
32

Example: The volume of the solid obtained by revolving the area bounded by y = x2

and y = x about the x axis.

Solution: First we solve these two equation to find the interval of integration. Easy to

see that (real) solution of y = x2 , y = x is x = 0, 1. Next we can see that y = x is

above y = x2 in this interval. Hence by above formula, the required Volume is
V =

"

(x x4 )dx =

3
.
10

Volume by cylindrical shells:

A cylindrical shell is the region between two concentric cylinders of same height h. It is
something like top portion of Well above earth surface. Let r1 be the radius of outer
cylinder and r2 be that of inner cylinder. Then the volume of this shell is
V = (r12 r22 )h = 2ra th,
where ra is the average radius (r1 + r2 )/2 and t is thickness of shell.
Consider the solid generated by revolving y = f (x), a x b around the y-axis. We
divide the interval [a, b] into n subintervals [xi1 , xi ]. The volume V of the solid may be
approximated by the sum of the volumes Vi of the shells between [xi1 , xi ]. Each shell is
approximately cylindrical. Its height is f (xi ), where xi = (xi1 + xi )/2, the mid point.
Its thickness is (xi xi1 )/2. Its average radius is xi . Hence its volume is
V

n
!
i=1

Vi =

n
!

2xi f (xi )(xi

i=1

xi1 ) 2

"

b
a

xf (x)dx as n .

If the region is revolved about the x axis, then

V =

"

2yf (y)dy.
a

Suppose the solid is obtained by revolving(about y axis) the area between two curves
y = f (x) and y = g(x) with f (x) g(x). Then the shell height will be f (xi ) g(xi ) ).
Hence the volume will be given by
V = 2

"

b
a

x(f (x) g(x))dx.

Example: Find the volume obtained by revolving the area bounded by y = 2x2 x3 and
33

y = 0 about y axis.
Solution: The points of intersection of y = 0 and y = 2x x3 are x = 0, 2. Height of the
shell is f (x) = 2x2 x3 . So the volume is
V =

"

2xf (x) = 2
0

"

2
0

(2x3 x4 ) =

16
.
5

If we revolve the area about Arbitrary line parallel to axis , say x = c. Then the radius
of the shell be x c (or c x whichever is positive) instead of x. So the volume in this
case is
"
b

2(x c)(f (x) g(x))dx.

V =

Similarly, if the region is revolved about x = d, then

V =

"

b
a

2(y d)(f (y) g(y))dy.

Example: Find the volume of the solid obtained by rotating the region bounded by y = 0
and y = x x2 about x = 2.
Solution: The points of intersection are 0 and 1. So the radius is 2 x and height is
x x2 . Hence the volume is
V =

"

1
2

2(2 x)(x x )dx = 2

"

1
0

(x3 3x2 + 2x)dx =

.
2

Surface area of solids of revolution

Consider an object obtained by revolving a curve y = f (x), a x b about x-axis.
We assume that f is dierentiable and f is integrable. We find the surface area of
this by approximating the surface by cylinders having the radius r1 on one end and r2
2
at the other end with lateral height l. The surface area of such cylinder is 2 r1 +r
l.
2
Now we divided the interval into sub-intervals [xi1 , xi ]. Let L be the line segment
connecting f (xi1 ) and f (xi ). Consider the small cylinders with radii r1 = f (xi ) and
r2 = f (xi1 ). Then the surface area of this cylinder is Si = 2 f (xi )+f2 (xi1 ) |L|, where |L|
is the length of line segment touching f (xi1 ) and f (xi ) . We note as in arc length, |L| is

34

1+

&

f (xi1 )f (xi )
xi xi1
n
!

'2

(xi xi1 ). Hence applying mean value theorem,

f (xi ) + f (xi1 )
|L|
2
i=1
/
$
%2
n
!
f (xi ) + f (xi1 )
f (xi1 ) f (xi )
= 2
1+
(xi xi1 )
2
x

x
i
i1
i=1

= 2

n
!
f (xi ) + f (xi1 ) .
i=1

"

f (x)
a

1 + [f (xi )]2 (xi xi1 ), xi [xi1 , xi ]

.
1 + (f (x))2 dx.

Example: Find the surface area of the solid obtained by revolving the curve y =

4 x2 , 1 x 1 about x-axis.
Solution: This is the portion of the circle x2 + y 2 = 4 between [1, 1].
"

.
1 + (f (x))2 dx
a
0
" 1
x2
= 2
4 x2 1 +
dx
4 x2
1
" 1
= 2
dx = 4

S = 2

f (x)

If the curve is described as x = g(y), then we have:

S = 2

"

y
a

.
1 + g (y)2 dy

Also, if the rotation is about the y-axis, the formula becomes,

S = 2

"

y
a

1 + g (y)2 dy.

Problem: Find the surface area of the solid obtained by rotating y =

about y-axis.

35

x, 1 y 2

Solution: Given curve is x = y 3 , 1 y 2. By the given formula,

"

" 2 .
dx 2
S = 2
y 1 + ( ) dy = 2
y 3 1 + 9y 4 dy
dy
y=1
1
" 2
.
2

=
36y 3 1 + 9y 4 dy = (1 + 9y 4 )3/2 2y=1 = (1453/2 103/2 )
36 1
27
27
2

Problem: Find the surface area and volume of the solid generated by infinite curve
y = x1 , x 1. Interpret the result.
Solution: The surface area and volume are given by
S = 2

"

1
x

1
1 + 4 dx,
x

V =

"

1
dx
x2

It is easy to see that the integral in S diverges. Indeed,

"

b
1

1
x

1
1 + 4 dx >
x

"

b
1

1
dx.
x

However, the integral for V converges. This is sometimes described as a can that does not
hold enough paint to cover its own interior. Of course, a finite amount of paint cannot
cover infinite surface. But if we fill the can with finite amount of paint we will have
covered an infinite surface. This is known as Painters paradox.

3.6

Problems

1. Using the definition of definite integral, determine which of the following functions
defined on [0, 1] are integrable

1 x < 1
(a) f (x) =
2 x = 1
(d) f (x) =

1 + x x Q
(b) f (x) =
1 x x Q

cos

xQ
x Q

where [y] is integral value of y.

(c) f (x) =

sin x

x = n1 , n N

cos x otherwise

x[ 1 ] 0 < x 1
x
(e) f (x) =
0
x = 0.

2. Determine if the following statements are True/false? Give a proof if it is true and
give an example if it is false.
36

(a) If f is integrable then f 2 and |f | are also integrable.

(b) If f, g are integrable then so is f g and fg , (g = 0).

(c) If |f | is integrable then f is integrable.
(d) If f, g are integrable then f og is integrable.
3. Suppose f and g are continuous functions on [a, b] and
there exists x [a, b] such that f (x) = g(x).

#b
a

#b

f=

4. Suppose f is continuous on [a, b] and f (x) 0. Show that if

f (x) = 0.

g. Then show that

#b
a

f (x)dx = 0, then

5. Prove that
(a) lim (
n

1
1
1
1

2
n
2
+
+. . .+ ) = loge 2 (b) lim [sin +sin +. . .+sin
] =
n n
n+1 n+2
2n
n
n
n

6. Discuss the convergence/divergence of improper integrals of first kind

(a)

"

e cosx dx,

(b)

"

dx
2
x (1 + ex )

(c)

"

"

x+1
dx (d)
x3

x2

dx

+ x

7. Discuss the convergence/divergence of improper integrals of second kind

(a)

"

2
1

x
dx
ln x

(b)

"

1
0

"

sin(x2 )
dx (c)
x

/2
1

tan x
dx
x3/2

(d)

"

3
0

8. Discuss the convergence/divergence of improper integrals

(a)

"

1
2

x2

e dx (b)

"

xp | ln x|q
dx, p, q > 1 (c)
(1 + x2 )2

"

log x
.
dx
|2 x|

dx
dx, p > 0
+ x2 )

|x|p (1

9. Show the following:

(a)

"

tx sin x

dx = arctan t,
x
2

(b)

"

1
0

xt 1
dx = log(1 + t)
log x

10. Using Beta and Gamma functions, evaluate the following:

(a)

"

x2

dx (b)

"

/2

tan xdx (c)

37

"

1
x (log( ))n dx (d)
x
m

"

/2

sin4 cos6 d
0

11. Compute the area:

(a) the region the lies inside the circle r = 1 and outside the cardioid r = 1 cos .

(b) The region inside the lemniscate r2 = 6 cos 2 and outside the circle r = 3.
12. Compute the arc length of
(a) x = cos t; y = sin t; t (0, 2) (b) x = a cos3 t; y = b sin3 t
13. Find the volume of solid of revolution by slicing method
(a) y = sin x about x-axis. (b) y = x3 , y = x, x 0, y 0 about y-axis.
14. Using the method of cylindrical shells find the Volume of the solids obtained by
(a) Rotating the area bounded by the following about y-axis
(a) y = 2x2 x3 , and y = 0 (b) y = x2 , and y = x

(c) y = x(1 x)2

(b) Rotating the area bounded by the following about x axis

(a) y =

x, from 0 to 1 (b) x = 1+y 2 , x = 0, y = 1, y = 2 (c) x = 4y 2 y 3 , x = 0

15. Prove the surface area of the solid obtained by revolving:

(a) the parabola, y 2 = 4ax between x = 0 and x = 3a, about y = 0 is

56
a2 .
3

(b) the circle, x2 + (y b)2 = a2 , (b > a) about y = 0 is 4 2 ab.

(c) one arc of the cycloid, x = a(t sint), y = a(1 cost) about x-axis is

64
a2 .
3

16. A spring with natural length of 10 inches. An 800 gram force stretches the spring
to 14 in. Find the force constant using Hooks law and the work done in stretching
the spring from 10 inches to 12 inches.
17. An electric elevator with a motor at the top has a multistrand cable weighing
4.5kg/ft. When the car is at the first floor, 180ft of cable are paid out, and effectively 0 ft are out when the car is at the top floor. How much work does the
motor do just lifting the cable when it takes the car from the first floor to top?

3.7

Hints

1. (a) Consider the partition P = {0, 1 , 1}. The upper and lower sum with this
partition is U (P , f ) = 1(1 ) + 2 = 1 + and L(P , f ) = 1(1 ) + 1 = 1.
Therefore, lim(U (P , f ) L(P , f )) = 0. Hence f is integrable.
0

38

(b) Consider the sequence of partitions {Pn }, where Pn = {0, n1 , n2 , ..... nn }. Then
+
Pn 0 as n 0 and U (Pn , f ) = ni=1 (1 + ni ) n1 = n1 (n + n+1
), L(Pn , f ) =
2
+n
i 1
1
n1
lim (U (Pn , f )L(Pn , f )) = 1. Hence by Darboux
i=1 (1 n ) n = n (n 2 ). So n0
Theorem f is not integrable
(c) Let > 0. Choose N such that 1/N < and 1/N < /4. Note that f has only
finite discontinuities in [ N1 , 1]. Hence integrable in [ N1 , 1]. Now for any partition
+
+
of [0, 1/N ], U (P, f ) L(P, f ) (Mi mi )xi
xi = 1/N < . Thus
f is integrable on [0, 1/N ] also.
Alternatively,
Note that f has finitely many discontinuities in [ 4 , 1]. Hence integrable in
[ 4 , 1]. Now for other half, 1 , 1 , 2 , .... be the infinitely many discontinuities of
f in [0, 4 ]. Then i = N1+i , for some N . Consider the partitions
P =

2
4

, 1

, 1 +
N +1

, ......, r
N +1

U (P , f ) L(P , f ) =

, r +
N +r

3
,
....,
0
. Then
N +r

(Mi mi )xi

i=1

!
i=1

Mi xi

C,
i
2
i=1

for some C > 0. Hence f is integrable.

(d) A similar problem is done in the class. One has to choose a sub-interval [a, b]
where cos x > 0 on that sub-interval [0, 1]. Then observe that min max f (x)
i

[xi1 ,xi ]

is strictly positive.
(e) f is continuous on [0, 1]. Use Uniform continuity of f to show that: for any
> 0, there exists P such that U (P, f ) L(P, f ) < .
(f) Note that f is continuous on [0, 1].
2. (a) Let M > 0 be such that |f (x)| M for all x [a, b]. Consider a partition

P = {a = x0 , x1 , x2 , , xn = b} such that U (P, f ) L(P, f ) < 2M

. Now we
2
2
claim that for this partition P we have U (P, f ) L(P, f ) < . For the proof
of the claim,let Mi and mi be the supremum and infimum of f and Mi and mi

39

be the supremum and infimum of f 2 in [xi1 , xi ] respectively, then

f 2 (x) f 2 (y) = (f (x) + f (y))(f (x) f (y)) 2M (Mi mi ).
Also, f 2 (x) 2M (Mi mi ) + f 2 (y) for all y [xi , xi+1 ]. This implies f 2 (x)

2M (Mi mi ) + mi and Mi 2M (Mi mi ) + mi . Therefore, Mi mi

2M (Mi mi ). Now the integrability of f 2 follows from
n
!
U (P, f ) L(P, f ) =
(Mi mi ).(xi+1 xi )
2

i=0

n
!
i=0

2M (Mi mi )(xi+1 x1 ) = 2M (U (P, f ) L(P, f )) .

Integrability of |f | is done in lectures.

(b) To check the Riemann integrability we need the function to be bounded. So
assume that fg is bounded on [a, b]. i.e., |g(x)| K for some K > 0. For any
x, y [0, 1] we have
f (x) f (y)
f (x)g(y) f (y)g(x)

=
g(x)
g(y)
g(x)g(y)
1
2 [|f (x)||g(y) g(x)| + |g(x)||f (x) f (y)|]
K

(3.8)

Now as f and g are integrable, P1 , P2 such that

U (P1 , f ) L(P1 , f ) <

K 2
K 2
, U (P2 , f ) L(P2 , f ) <
.
2 max |g|
2 max |f |

Now the integrability of fg follows from (3.8) by taking P = P1 P2 . Then

U (P, fg ) L(P, fg ) < K12 (K 2 ) = .
(c) No. Take a counter example as f (x) = 1 for x Q and f (x) = 1 for x Q.

0 x = 0
0 x Q
(d) No. Counter example is f (x) =
and g(x) = 1q x = pq
1 x (0, 1]

1 x = 0, 1.

1 x Q
Then f og(x) =
0 x Q

To show the integrability of g, let > 0 be given. Take N N such that

40

1
N

< 2 .

Now note that there are only finite number of rational numbers say 1 , 2 , ....k
[ N1 , 1] such that i = pqii and qi N . Let 1 be chosen such that 1 < and
1
1
i (j 4K
, j + 4K
), for i = j, i, j = 1, 2, ...k. Now consider
P = {0,

1
1
1
1
1
, 1 , 1 + , ...., k , k + , 1}
N
4k
4k
4k
4k

Then clearly L(P , g) = 0. Now

U (P , g) =

!
P

! 1
1
Mi xi
+
= .
N
2k
i=1

3. Take h = f g Then h is also continuous. Then by Mean Value theorem, [a, b]

#b
#b
#b
#b
such that a h(x)dx = h()(b a). But as a f (x)dx = a g(x)dx. So a h(x)dx = 0.
Hence h() = 0 f () = g()
4. We will prove it by contradiction. Suppose x0 [a, b] such that f (x0 ) = 0. Then
as f 0 we have f (x0 ) > 0. Also f is continuous so for a given > 0 > 0 such
that f (x) [f (x0 ) , f (x0 ) + ] for x [x0 , x0 + ]. So if we take = f (x2 0 )
then f (x) f (x2 0 ) x [x0 , x0 + ]. So by domain decomposition property of
#b
# x +
Riemann integral, we get a f (x)dx x00 f (x2 0 ) dx = f (xo ) > 0, Contradiction.
Hence f 0.
#2
1
5. (a) Take f (x) = x+1
then f is integrable on [0, 1] and 1 f (x)dx = ln 2. Take
+n
+n 1
i 1
Pn = {0, n1 , n2 , , nn }. Then S(Pn , f ) =
i=1 f ( n ) n =
i=1 i+n . Hence
lim S(Pn , f ) = ln 2.
n
#1
(b) Take f (x) = sin(x) then f is integrable on [0,1] and 0 f (x)dx = 2 . Take
+
+
Pn = {0, n1 , n2 , , nn }. Then S(Pn , f ) = ni=1 f ( ni ) n1 = ni=1 sin( i
) 1 . Hence
n n
+
limn S(Pn , f ) = i=1 sin( i
) 1 = 2
n n
" b
# x
6. (a) 0 e cos xdx = lim
ex cos xdx
b

#b

1
Now take I = 0 e cos xdx = 12 (1 cos beb + sin beb ). So lim I =

b
2
# x
e cos xdx is convergent.
0
# dx
#
(b) 1 x2 (1+ex ) 1 dx
which is convergent .Hence by comparison test given imx2
proper integral is convergent.
#
#1
#1
(c) 1 (x+1)dx
= 0 1x + 0 x3/2 dx. The second integral on the right side diverges.
3
x

x2

41

# 1 dx # dx
+

. Both the integral on the right side are convergent

0
0
1 x2
x
hence the given integral is convergent.
# ln 2 t/2
7. (a) Take ln x = t then x = et and the integral becomes 0 e t . It is easy to see
# ln 2
that integrand is 1t and the integral 0 1t diverges.
#1
2)
f (x)

(b) f (x) = sin(x

and Take g(x) = 1x . Then lim
= 0. Also 0 dxx is converx
x0 g(x)
# 1 sin(x2 )dx
gent so 0 x is convergent.
(d)

dx
x

x2 +

f (x)
(c) Take f (x) = tan(x)
and g(x) = tan x. Then lim
(0, ). Also as
x3 2
x 2 g(x)
#
# tan(x)dx
2
2
tan(x)dx
is
divergent
so
is divergent.
x3 2
1
1

# 1 ex2 dx

Now
is convergent Since if we
0
1
0
x
#
# x2
f (x)
1
take g(x) = 1x then lim
= 1 and 0 dxx is convergent But 1 e xdx is
x0 g(x)
#
f (x)
divergent. Since if we take g(x) = 1x then lim
= and 1 dxx is
x g(x)
divergent. Hence the given integral is divergent.
# p (| log x|)q
# p (log x)q # 1 xp (| log x|)q
(b) I = 0 x (1+x
= 1 x(1+x
= I1 + I2 . First consider I1 =
2 )2
2 )2 + 0
(1+x2 )2
# xp (log x)q
. Then we have the following cases.
(1+x2 )2
1
#
Case 1: p < 0 and q R. Take g(x) = (1+x1 2 )2 . Then 1 g(x) is convergent.
#
(x)
Also as limx fg(x)
= 0, 1 f (x) converges.
#
Case 2: p > 3, q > 0. Take g(x) = xp4 . Then 1 g(x) is divergent. Also as
#
(x)
limx fg(x)
= , 1 f (x) diverges.
#
Case 3: p > 3, q 0. Take g(x) = xp4 where > 0. Then 1 g(x) is
#
(x)
divergent. Also as limx fg(x)
= , 1 f (x) is divergent.
Case 4: 0 p < 3, q 0. Take g(x) = xp4+ where p 4 < 0. Then
#
#
(x)
g(x) is convergent. Also as limx fg(x)
= 0, 1 f (x) is convergent.
1
#
Case 5: 0 p < 3, q < 0. Take g(x) = xp4 . Then 1 g(x) is convergent.
#
(x)
Also as limx fg(x)
= 0, 1 f (x) is convergent.
# (log x)q
#
# q t
1
q
Case 6: p = 3. The I1 = 1 (1+
=
t e dt. Now if
1 2 4 1 (log x)
0
)
x2
#
q et

q > 1, take g(t) = tq . Then 0 tq is divergent. Also as limt tg(t)

= ,
# q t
#1
q
t e is divergent. For q 1, take g(x) = t . Then as 0 g(x) is divergent,
0
#1
# p (log x)q
q et
and limt0 tg(t)
= 1, 0 tq et is divergent. Thus 1 x(1+x
2 )2 is divergent for p = 4
and q R.
# 1 p (| log x|)q
Now consider I2 = 0 x (1+x
2 )2 . Putting log x = y one can see that I2 =
# e(p+1)y yq
dy. The following cases arise.
0 (1+e2y )2

8. (a)

1
2

ex dx =

#1

ex dx
0
x

ex dx
dx.
x

42

# z q
1
Case 1: p > 1, q > 1. In this case one can that I2 (p+1)
e z dz
q+1
0
and hence converges.
#1
e(p+1)y
Case 2: p R, q < 1. Then as 0 tq is divergent and limt0 (1+e
2y )2 = 1, I2
diverges.
#
e(p+1)y
Case 3: p < 1, q > 1. Then as 1 tq is divergent and limt0 (1+e
2y )2 = ,
I2 diverges.
Thus I is convergent i 1 p < 3 and q > 1.
#
# 1
#1
# dx
dx
dx
dx
(c) Note that |x|p (1+x
+ 1 |x|p (1+x
. Now
2) =
2) +
|x|p (1+x2 )
1 xp (1+x2 )
1
take g(x) = 1+x2 . Then by limit comparison test we have 1st and 3rd inte# 1 dx
# dx
gral on RHS convergent as (1+x
are convergent. Now for
2 ) and
1 (1+x2 )
# 1
# 1 dx
1
p
dx
1
,
take
g(x)
=
.
Then
lim

|x|
=
1
And
is
p
2
p
|x|
1 |x| (1+x )
1 xp
x0 |x|p (1 + x2 )
convergent for 0 < p < 1 Thus given integral is convergent for p < 1.
#
#
1
9. (a) Let f (t) = 0 etx sinxxdx . Then f
= 0 etx sin xdx = 1+t
2 f (t) =
t
#a
arctan t + c. By the second fundamental theorem, f (a) f (0) = 0 f (t)dt =
# a 1
dt, taking a , lima f (a) f (0) = /2. Also,
0 1+t2
|f (a)| = |

(b)

10. (a)
(b)
(c)

(d)
11. (a)
(b)

"

ax sinx

| C1

"

eax as a .

Therefore, lima f (a) = 0. Using this we get c = 2 and hence f (t) =

arctan t.
2
#1 t
#1
1
Let f (t) = 0 xln1
dx, then f
= 0 xt dx = t+1
f (t) = ln(t + 1) + c. Now
x
t
# 1 xt 1
f (0) = 0 c = 0 0 ln x dx = ln(t + 1)
#
#
1
2
I = 0 ex dx Put x2 = t 2xdx = dt I = 0 12 et t 2 dt I = 12 ( 12 )
#
#
1
1
I = 02 tan xdx = 02 sin 2 x cos 2 xdx = 12 ( 34 , 14 )
#1
#
Let I = 0 xm [ln x1 ]n dx. Put ln x1 = t I = 0 e(m+1)t tn dt. Now put
(m + 1)t = y we get I = (m+1)1(n+1) (n + 1)
#
Let I = 02 sin4 cos6 d = 12 ( 52 , 72 )
# /2
# /2
A = 2 12 0 ((r12 r22 )d = 0 (1 (1 cos )2 )d
# /2
= 0 ( cos2 + 2 cos )d = 2 /4

r2 = 6 cos 2 and r = 3 then 6 cos 2 = 3 = 6

#
#
A = 4 12 06 (r12 r22 )d = 2 06 (6 cos 2 3)d = [3 3 ]
43

12. (a) x = cos1t, y = sin t, z = t then by

.
# 2
# 2 .
L= 0
(( dx
)2 + ( dy
)2 + ( dz
)2 ) = 0
(2 + 2 2 ) = 2 (1 + 2 )
dt
dt
dt
3
(b) x = a cos1
t, y = b sin3 t then by
# 2
4
L= 0
(( dx
)2 + ( dy
)2 = a+b
(a2 + ab + b2 )
dt
dt
#
13. (a) Using V = 0 y 2 dx, y(x) = sin x and 0 x then
#
V = 0 sin2 xdx = 2 /2

(b) Given y = x3 , y = x, x 0, y 0 Then x3 = x x = 0, 1

#1
V = 0 ((x3 )2 x2 )dx = 4
21
#2
14. (a) (1)y = 2x2 x3 , y = 0 x = 0, 2 So V = 2 0 (x(2x2 x3 ))dx = 16
5
#1
(2)y = x2 , y = x x = 0, 1 Then V = 2 0 (x(x2 x))dx = 6
#1

(3)y = x(1 x)2 , y = 0 x = 0, 1 Then V = 2 0 (x.x(1 x)2 dx = 15

#1

(b) (1)y = x, y = 0, 1 Then V = 2 0 y.y 2 dy = 2

#2
(2)x = 1 + y 2 , x = 0, y = 1, y = 2 Then V = 2 1 y(1 + y 2 )dy = 21
2
#4
(3)x = 4y 2 y 3 , x = 0 y = 0, 4 Now V = 2 0 y(4y 2 y 3 )dy = 512
5
1

# 3a
.
dy 2
dy
15. (a) S = 2 0 2y (1 + ( dx
) )dy. As y = 4ax, dx
= xa . On integrating we
2

get S = 56a
3
# 3a 1
dy 2
(b) S = 2 0 y (1 + ( dx
) )dy
Changing into parametric form by taking x = a cos t,
y = b + a sin t, we have dx
= a sin t, dy
= a cos t
dt
dt
.
# 2
2
Then S = 2 0 (b + a sin t) (a2 sin t + a2 cos2 t)dt = 4 2 ab

16. F =
K=

80010
= 8N and
1000
8
= 80N/m Now
0.1

2.5
x = (14 10) 100
= 0.1m .Then using F = K x
# 0.05
W = 0 Kxdx = 1.1J

17. K = 4.5Kg/f t = 4.510

= 150N/m
0.3
In lifting the car x distance upward force needed is F = 150(180 x)N
# 180
Now W = 0 150(180 x)dx = 2.43 106 J

44