Chapter 2

Improper Integrals

We know that any continuous and bounded function on (a; b), with a; b 2 R, is
Riemann integrable.

In this chapter, we extend this concept to cases where the functions consid-
ered may be unbounded, as well as to cases where the interval of integration
may not be bounded.

2.1 Concept of Improper Integrals

De…nition 21 (Singularity of a Real-Valued Function) Let f be a real-
valued function de…ned on an interval (a; b). A point c 2 (a; b) is called a
singularity of f if one of the following conditions holds:

1. Unbounded Behavior: f (x) becomes unbounded (i.e., it tends to 1 or

1) as x approaches c from either side of the interval. This means that
the limit limx!c f (x) does not exist in a …nite sense.

2. Discontinuity: f (x) is not continuous at c, meaning the left-hand limit

limx!c f (x) and the right-hand limit limx!c+ f (x) do not exist or are
not equal to each other.

De…nition 22 We have the following two cases:

1. Let f : [a; b) ! R be a continuous function, where a; b 2 R and b is a

Rb
sigularity of the function f . Then, the improper integral a f (x)dx con-
verges, if
Zt
lim f (x)dx exists and is …nite
t!b
a

7
8 CHAPTER 2. IMPROPER INTEGRALS

we de…ne
Z b Zt
f (x)dx = lim f (x)dx
a t!b
a

2. Let f : [a; +1) ! R be a continuous function, where a 2 R. Then, the

R +1
improper integral a f (x)dx converges, if

Zt
lim f (x)dx exists and is …nite
t!+1
a

we de…ne
Z +1 Zt
f (x)dx = lim f (x)dx
a t!+1
a

Remark 23 When the improper integral does not converge, we say that it di-
verges.

Remark 24 The change of variables y = x, allows all the convergence results

we mentioned in cases 1) and 2) (see De…nition 1) to automatically adapt to
the case:

1. f : (a; b] ! R, be a continuous function, where a; b 2 R.

2. f : ( 1; b] ! R, be a continuous function, where b 2 R.

Example 25 We have

1. The following improper integral

Z
+1
1
dx converges
x2
1

Indeed, since

Zt
1 1
lim dx = lim 1 = 1 < +1
t!+1 x2 t!+1 t
1

2. The following improper integral

Z1
1
p dx converges
2
x
0
2.2. PROPERTIES OF IMPROPER INTEGRALS 9

Indeed, since
Z1
1 p
lim+ p dx = lim+ (2 2 s) = 2 < +1
s!0 2
x s!0
s

3. The following improper integral

Z1
1
dx diverges
x
0

Indeed, since
Z1
1
lim dx = lim (ln(1) ln(s)) = +1
s!0+ x s!0+
s

Proposition 26 (Riemann Integrals) For all …xed 2 R , we have

1.
Z
+1
1
dx converges () >1
x
1

2.
Z1
1
dx converges () <1
x
0

Proof. Complete it as a homework task.

2.2 Properties of Improper Integrals

1. Chasles’Relation: Let f be a real-valued and continuous function over
[a; +1), where a 2 R, and a0 2 (a; +1). Then, the following two im-
proper integrals
Z
+1 Z+1

f (x)dx and f (x)dx

a a0

have the same nature (they converge or diverge), and we have

0
Z
+1 Za Z
+1

f (x)dx = f (x)dx + f (x)dx

a a a0
10 CHAPTER 2. IMPROPER INTEGRALS

2. Linearity: Let f and g be two real-valued and continuous functions over

[a; +1), where a 2 R, and ; 2 R. Then, if
Z
+1 Z
+1

f (x)dx and g(x)dx converge

a a

Thus,
Z
+1 Z
+1 Z
+1

( f (x) + g(x))dx = f (x)dx + g(x)dx

a a a

3. Monotonicity:Let f and g be two real-valued and continuous functions

over [a; +1), where a 2 R, and their improper integrals over [a; +1)
converge. Moreover, if
f (x) g(x); 8x 2 [a; +1)

Thus
Z
+1 Z
+1

f (x)dx g(x)dx
a a

Remark 27 The previous properties remain true for functions de…ned on (a; b],
[a; b), ( 1; b] where a; b 2 R [ f 1; +1g and a < b.
De…nition 28 Let f be a real-valued function over (a; b) R and c 2 (a; b) is
a singularity of the function f . Moreover, if the following improper integrals
Zc Zb
f (x)dx and f (x)dx converge
a c

Thus
Zb Zt Zb
f (x)dx = lim f (x)dx + lim+ f (x)dx
t!c s!c
a a s

Example 29 The following improper integral

Z
+1
1
I= dx converges
1 + x2
1

Indeed, since
Z0 Zt
1 1
I = lim 2
dx + lim dx
s! 1 1+x t!+1 1 + x2
s 0
= lim [arctan(x)]0s + lim [arctan(x)]t0
s! 1 t!+1

= + = < +1
2 2
2.3. IMPROPER INTEGRALS FOR POSITIVE FUNCTIONS 11

2.3 Improper integrals for positive functions

Theorem 30 (Comparaison Criterion) Let f and g be two real-valued, con-
tinuous, and positive functions over [a; +1). Moreover, suppose that the func-
tion f is bounded above by the function g in the neighberhood of +1, i.e.
9A a, 8x > A, f (x) g(x)
Z
+1 Z
+1

1. If g(x)dx converges, then f (x)dx also converges.

a a

Z
+1 Z
+1

2. If f (x)dx diverges, then g(x)dx also diverges.

a a

Example 31 We have
1. The following improper integral
Z
+1
1
I1 = dx converges
x2 (1 + ex )
1

Indeed, since
1 + ex > 1, 8x 1
x2 (1 + ex ) > x2 , 8x 1
1 1
2 x
< , 8x 1
x (1 + e ) x2

Thus
Z
+1 Z
+1
1 1
dx < dx < +1
x2 (1 + ex ) x2
1 1

Z
+1
1
Because x2 dx converges (it is a special case of the Riemann integrals
1
= 2 > 1). Hence, based on the camparaison criterion, the improper
integral I1 is convergent.
De…nition 32 Let f and g be two real-valued functions de…ned over the interval
[a; b) R. Then, we have
1. We write f = Ob (g) if there exists a constant C > 0 and a neighborhood
of b (i.e., an interval [c; b) [a; b) for some c < b) such that:
jf (x)j Cjg(x)j
for all x 2 [c; b).
12 CHAPTER 2. IMPROPER INTEGRALS

2. We write f = ob (g) if:

f (x)
lim = 0:
x!b g(x)
Corollary 33 Let f and g be two real-valued, continuous, and positive functions
over [a; b), where a 2 R, b 2 R [ f+1g, and a < b. Then
Zb
1. If f = Ob (g), so the convergence of g(x)dx leads to the convergence of
a
Zb
f (x)dx
a

Zb
2. If f = ob (g), so the convergence of g(x)dx leads to the convergence of
a
Zb
f (x)dx
a

Corollary 34 (Equivalence Criterion) Let f and g be two real-valued, con-

tinuous, and positive functions over [a; b), where a 2 R, b 2 R [ f+1g, and
a < b. Moreover, the function f is equivalent to the function g (with g(x) > 0,
8x 2 [a; b)) in the neighberhood of b, i.e.
f (x)
lim =1
x!b g(x)
Zb Zb
Then, the improper integrals f (x)dx and g(x)dx have the same nature.
a a

Example 35 Study the following improper integral

Z
+1
x3 + 2x2 1
I2 = dx
x5 + 1
1

x3 +2x2 1
Solution 36 Since x3 + 2x2 1 x3 and x5 + 1 x5 , then x5 +1
+1 +1 +1
x3 1
x5 = x2 , i.e.
x3 +2x2 1
x5 +1
lim 1 =1
x!+1
x2
Z
+1 Z
+1
1 1
Thus, I2 and x2 dx have the same nature, however x2 dx converges
1 1
(because, it is a special case of Riemann integral = 2 > 1). Hence, the
improper integral I2 also converges.
2.4. IMPROPER INTEGRALS OF FUNCTIONS WITH ARBITRARY SIGN13

2.4 Improper Integrals of Functions with Arbi-

trary Sign
Proposition 37 Let f : [a; b) ! R be a continuous function, where a 2 R,
b 2 R [ f+1g and a < b.If the improper integral

Zb
jf (x)j dx converges
a

Then, the improper integral

Zb
f (x)dx is absolutely convergent, hence converges.
a

Example 38 Study the following improper integral

Z
+1
cos(x)
I3 = p dx
x2x
1

Solution 39 We know that jcos(x)j 1, 8x 1 (and generally 8x 2 R). Then,

we get
jcos(x)j 1 1
p p = 3 , 8x 2 [1; +1)
x x
2
x x
2
x2
Z
+1
1
Since 3 dx is convergent (because, it is a special case of Riemann integral
x2
1
3
= > 1). Thus, based on comparaison criterion we can conclude that the
2
Z
+1
jcos(x)j
improper integral p dx is convergent. Therfore, the improper integral I3
x2x
1
is absolutely convergent, hence converges.