II.

Integration by Substitution
A. Algebraic Substitution
Many integrals may be evaluated by introducing a new variable of integration, say z , in the
place of the original variable x , the two variables being connected by some suitable formula. The
cange of variable is usually brought about by means of an explicit substitution,

'
x=φ ( z ) , dx=φ ( z ) dz.

This process, called integration by substitution, is highly important. It is to be remembered that

not merely x , but dx as well, must be replaced by the proper expression in terms of the new
variable.
The substitution to be made must be determined by inspection of the integrand. No
general rules can be given; skill in the choice of substitutions comes only with practice. There is,
however, one rather crude rule of thumb which succeeds often enough to make it worth
consideration. Determine if possible, what quantity seems to be causing the trouble and then
introduce a new variable for that quantity.

Example 1 ∫ √ 3
x dx
1+ √ x |
Solution We let x=z 6; then dx=6 z5 dz . So

z 3 ( 6 z 5 dz )
∫ √ x 3dx =∫ 2
1+ √ x 1+ z
8
z dz
¿ 6∫ 2
z +1

Dividing the numerator by the denominator we have

√
∫ 1+ 3
x dx
x√
=6 ∫
( 6 4
z −z + z −1+
2
2
1
z +1 )
dz

¿6 ( 17 z − 15 z + 13 z −z+ arctan z)+C

7 5 3

Therefore

∫ √ x 3dx = 67 x7 / 6− 65 x 5 /6 +2 x1 /2 −6 x1 /6 + 6 arctan x 1/ 6 +C
1+ √ x

Example 2 ∫ x
5
√ x 2 + 4 dx|
Solution Let z=√ x 2+ 4 . Then z 2=x 2+ 4 , and 2 zdz=2 xdx . So
 ∫ x 5 √ x 2 +4 dx=∫ ( x 2 ) √ x 2 +4 ( x dx )
2

¿ ∫ ( z −4 ) z ( z dz )
2 2

¿ ∫ ( z 6−8 z 4 +16 z 2 ) dz
1 7 8 5 16 3
¿ z − z + z +C
7 5 3
1 3
z [ 15 z −16 z +560 ]+C
4 2
¿
105
1 2
( x +4 ) [ 15 ( x 2+ 4 ) −168 ( x 2+ 4 ) + 560 ] +C
3 /2 2
¿
105
1 2 3 /2
¿ ( x +4 ) ( 15 x 4−48 x 2 +128 ) +C
105

EXERCISES:
Evaluate the indefinite integral.
x ( 5 y +4 ) dy
1.∫ dx 10.∫
3+ √ x √5 y−1
dx 11. ∫ √ 1+ √ z dz
2.∫ 3
√ x−x ( 1+ ln x ) dx
dx 12.∫
3.∫ x
3

x √ 1+ 4 x
13.∫ √
y 5−1
4. ∫ x (1+ x ) dx
2 /3
dy
y
dx
5.∫ 3 14.∫ √ e z−9 dz
2 √ x+ √ x
( 12 x+ 1 ) dx 15.∫ x3 √ a2 + x 2 dx
6.∫ 16.∫ cos √ t dt
√ 4 x−3
7.∫
dx 17.∫ e2 x √ 1−e x dx
1+ √ x 18.∫ ln ( √ y+ 3 ) dy
( 4 x−1 ) dx
8.∫ 3 19.∫
y dy
( 2 x +1 ) 2 1+ √ y
2y
9. ∫ sin √ t dt e dy
20.∫ y
√ e −1

B. Trigonometric Substitution
You have already seen how some techniques of integration require a change-of-variable
substitution. In this section you will learn substitutions involving trigonometric functions that lead
to trigonometric integrals. There are three cases how changing a variable by a trigonometric
substitution often enables us to evaluate an integral containing an expression of one of the
following forms where a> 0 :

√ a2−x 2 √ a2 + x 2 √ x 2−a2
 Case I. The integrand contains an expression of the form √ a2−x 2, where a> 0.
Introduce a new variable θ by letting x=a sin θ , where
1 −1
0≤θ≤ π if x ≥ 0 and π ≤ θ< 0 if x <0
2 2

−1 1
In this case, with x=a sin θ , dx=a cos θ dθ , and cos θ ≥ 0 because π≤ θ ≤ π.
2 2
Moreover
√ a2−x 2= √ a2−a 2 sin2 θ
¿ √ a2 √ 1−sin2 θ
¿ a √ cos 2 θ
¿ a cos θ

Example 1 ∫ √
9−x 2
x
2
dx
|
Solution Observe that because the denominator is x 2, x ≠ 0 .
With the substitution indicated in Case I, let x=3 sin θ where
1 −1
0<θ ≤ π if x >0 and π ≤ ϑ < 0 if x <0. Then
2 2
dx=3cos θ dθ and
3 x √ 9− x2= √ 9−9 sin2 θ
θ ¿ 3 √ cos2 θ
0° √ 9− x2 ¿ 3 cos θ
x >0
Therefore

∫ √ 9−x 2
dx=∫
3 cos θ
( 3 cos θ dθ )
x2 9sin 2 θ
¿ ∫ cot θ dθ
2

¿ ∫ ( csc 2 θ−1 ) dθ
¿−cot θ−θ +C

1 x∧−1 1 −1 1
Because sin θ= π < θ< π ,θ=sin x . To find cot θ , refer to the figure
3 2 2 3
cot θ= √
9−x 2
x
Therefore

∫ √ 9−x 2
dx=
− √ 9−x 2 x
−arcsin +C
x2 x 3
Case II The integrand contains an expression of the form √ a2 + x 2, where a> 0. Introduce a new variable
1 −1
θ by letting x=a tan θ, where 0 ≤ θ< π if x ≥ 0 and π <θ<0 if x <0.
2 2
−1 1
With x=atan θ , dx=a sec 2 θ dθ , and because π <θ< π , sec θ ≥ 1. Furthermore
2 2

√ a2 + x 2=√ a2 + a2 tan2 θ
¿ √ a √ 1+ tan θ
2 2

¿ a √ sec 2 θ
¿ a secθ

Example 2 ∫ √ x +5 dx|
2

Solution Let x=√ 5 tan θ and dx= √ 5 sec 2 θ dθ

√ x 2+5=√ 5 tan2 θ+5

¿ √ 5 √ sec 2 θ
¿ √ 5 sec θ
Therefore
∫ √ x 2+ 5 dx=∫ √ 5 sec θ ( sec2 θ dθ )
¿ 5∫ sec θ
3

From trigonometric transformation

∫ sec3 θ dθ=∫ sec θ dθ+∫ tan2 θ sec θ dθ

∫ sec θ dθ=ln|sec θ+ tan θ|
And
√ x 2+5 x
∫ tan θ sec θ
2
θ

can be integrated by using integration by parts. √5

Let u=tan θ∧du=sec 2 θ dθ
dv =tan θ sec θ dθ∧v=sec θ

∫ tan2 θ sec θ dθ=tan θ sec θ−∫ sec3 θ dθ

Then
 ∫ sec3 θ dθ=ln|sec θ+tanθ|+ tan θ sec θ−∫ sec3 θ dθ
2∫ sec θ dθ=ln |sec θ+ tanθ|+ tan θ secθ +C
3

1 1
∫ sec3 θ dθ= 2 ln|sec θ+ tanθ|+ 2 tan θ sec θ+ C

sec θ=
√ x 2 +5 tan θ= x
√5 √5
Therefore

|
∫ √ x 2+ 5 dx= 52 ln √ x 5+5 + |
x 5 x √ x 2 +5
2
+ +C
√ √5 2 √ 5 √ 5
1 5 5
¿ x √ x + 5+ ln |√ x +5+ x|− √ 5+C
2 2
2 2 2
1 5
¿ x √ x + 5+ ln |√ x +5+ x|+ C
2 2
2 2

Case III The integrand contains an expression of the form √ x 2−a2, where a> 0.
1 3
Introduce a new variable θ by letting x=a sec θ , where 0 ≤ θ< π if x ≥ a and π ≤θ< π
2 2
if x ≤−a .
1 3
With x=a sec θ, dx=a sec θ tan θ dθ , and tanθ ≥ 0 because either 0 ≤ θ< π∨π ≤θ< π .
2 2
Moreover
√ x 2−a2= √ a2 sec2 θ−a 2
¿ √ a √ sec θ−1
2 2

¿ a √ tan2 θ
¿ a tanθ

Example 3 ∫
dx
x √ x 2−9
3 |
Solution
Let x=3 secθ , dx=3 sec θ tan θ dθ

√ x 2−9= √ 9 sec2 θ−9

¿ 3 √ tan2 θ
¿ 3 tan θ

Therefore
dx 3 sec θ tan θ dθ 1
∫ =∫ ¿ ∫ ( 1+cos 2θ ) dθ
x 3
√ x −9
2 3
27 sec θ •3 tanθ 54
¿
1
27
∫ 2
cos θ dθ ¿
1
54( 1
)
θ+ sin 2 θ + C
2
 1
¿ ( θ+ sin θ cos θ ) +C
54 x
√ x −9
2

θ
3

From the triangle sin θ= √

x 2−9 cos θ= 3 x
, and θ=arcsec
x x 3

∫
dx
x 3 √ x 2−9
=
1
54(arcsec + √
x
3 x )
x 2−9 3
• +C
x

arcsec + √
1 x x 2−9
¿ +C
54 3 18 x 2

Exercises
Evaluate the indefinite integral.

dx
1.∫
x 2
√ 4− x2 11. ∫ √ 1−u2 du
2 x
2
2.∫ dt 12.∫ 2 dx
t √ t +25 4
√ x +6
dx dx
3.∫ 13.∫
x √ x 2 +4 ( 2+ x 2 )
3 /2

dx
4. ∫ 14.∫
dx
√ 4 x + x2 √ 4 x−x 2
x
5.∫ dx 15.∫ 2 2
dw
√ x −25 2
w √ w −7
x
6.∫
−x
dx e
√ x −25 2 16.∫ 3/ 2
dx
3 ( 9 e−2 x +1 )
x
7.∫ dx 17.∫ 2
dz
√ 16−x2 ( z −6 z +18 )
3/ 2
dx
8 .∫ ln w
3

( 4 x 2−9 )
3/ 2
18 .∫ dw
dx w √ ln 2 w−4
9. ∫ dx
( 5−4 x−x 2 )
3/ 2 19.∫
x √ 25−x 2
10.∫ √ 2 dx
4−x 2
x
2
sec x
20 .∫ 3 /2
dx
( 4−tan 2 x )
III. Integration by Rational Fraction
A. Introduction
We take up next the problem of integrating a rational algebraic fraction that is the quotient of two
polynomials. The first step in dealing with an integral of this type is to carry out the indicated division until
the numerator is of lower degree than the denominator. In developing the theory we shall suppose always
that this preliminary step has been taken. In this lesson, whenever the quantity a x 2 +bx +c occurs, it will
be assumed that b 2−4 ac< 0. If b 2−4 ac ≥ 0, the quantity a x 2 +bx +c can be factored into real linear
factors.
By methods already familiar, we can immediately integrate fractions of the forms
A A ( 2 ax+ b ) A
n
, n
, 2
( ax +b ) ( a x +bx +c ) a x + bx+ c
2

The first two lead to powers, if n>1 , to logarithm, if n=1; the third leads to an arctangent. We can also
integrate

A
n
, n>1
( a x 2+ bx+ C )
by a trigonometric substitution.

B. Distinct Linear Factors

The simplest case is that the numerator can be broken up into real linear factors, none of which is
repeated. In this case we may always rewrite the given fraction as a sum of fractions whose numerators
are constants and whose respective denominators are the factors of the original denominator.

|
3
x +2
Example 1 ∫ 3
dx
x −x
By division,
3
x +2 x +2
3
=1+ 3 .
x −x x −x

The factors of the denominator are x , x +1 , x −1. Assume

x+2 A B C
= + + ,
3
x −x x x +1 x−1

where A , B ,C are constants to be determined. Clearing the fractions, we find

x +2= A ( x 2−1 ) + Bx ( x−1 )+Cx (x +1)

This relation must hold for al values of x . Hence, assigning to x any three values whatever, we must obtain
three simultaneous equations to determine A , B ,C . but the most convenient values to use are 0 ,−1, 1
(the zeros of the denominator), for each of these causes two terms to drop out:
x=0 , A=−2
1
x=−1 , B=
2
3
x=1 , C=
2
Thus

( )
3
∫ (x 2+2 ) dx=∫ 1− 2x + 12 • x 1+2 + 32 • x−1
1
dx
x x −1
1 3
¿ x−2 ln x+ ln ( x+1 )+ ln ( x −1 )+ C
2 2

EXERCISES
Evaluate the following integrals.
( 2 x+ 11) dx
1.∫ 2
x + x −6 ( y 3+ 1 ) dy
11. ∫
( x−1 ) dx y ( y +2 )
2.∫ 2
x +5 x+6 ( y +13 ) dy
12.∫
dx ( y +1 ) ( y +3 ) ( y−2 )
3.∫ 2
x +4 x ( y 2 −14 y +5 ) dy
13.∫
( 3 x2 +8 x−12 ) dx y ( y−1 ) ( y−5 )
4. ∫ 3 2
x +7 x +12 x 21 dv
14.∫
( x 2−5 x +3 ) dx ( v−1 ) ( v +3 ) ( v −4 )
5.∫ 3 2
2 v dv
x −4 x +3 x 15.∫
( 17 x−6 ) dx ( v+ 1 )( v+ 2 )( v +3 )
6.∫ 3 2 2
6 y dy
x −x −6 x 16.∫ 4 2
( 5 x−12 ) dx y + 4 y +3
7.∫ 3 5 sin θ cos θ dθ
17.∫ 2
2
x −6 x +8 x
( 3 x 2−4 x−1 ) dx sin θ+3 sin θ−4
8.∫ 3x
e dx
18.∫ 2 x
2
x −x−2
( x 2−8 ) dx e −9
9. ∫ 2 19.∫ x
9 dx
x +5 x+ 6
e +9
( y 3 + 4 ) dy dy
10.∫ 20.∫ 2 y
y ( y +1 ) √e +4
C. Repeated Linear Factors.
If the denominator contains a factor ( x−a )r , the above method fails, since there would be r partial
fractions with denominator x−a , and these could be combined into a single fraction with denominator
x−a . In this case, corresponding to the factor ( x−a )r , we assumed r partial fractions of the form

A B D
+ + …+…+ .
x−a ( x −a ) 2
( x−a )r
 |
3
x −1
Example 2 ∫ 3
dx
x ( x +1 )
Assume
3
x −1 A B C D
3
= + + + ,
x ( x +1 ) x x+1 ( x+1 ) ( x +1 )3
2

then,
3 3 2
x −1= A ( x +1 ) + Bx ( x+ 1 ) +Cx ( x +1 ) + Dx
¿ A ( x 3 +3 x2 +3 x +1 ) + Bx ( x 2+ 2 x +1 ) +Cx ( x +1 ) + Dx
3 2 3 2 2
¿ A x + 3 A x +3 Ax + A+ B x +2 B x + Bx+C x +Cx+ Dx
3 3 2
x −1= ( A +B ) x + ( 3 A +2 B+C ) x + ( 3 A+ B+C + D ) x+ A .

The equations are

A+ B=1 ( 1 ) 3 A +2 B+C=0(2)
3 A + B+C+ D=0 ( 3 ) A=−1(4)

And the values of A , B ,C , D are −1 , 2,−1and 2 respectively.

Then,
( x 3−1 ) dx
∫
x ( x +1 )
3
−1 2
=∫
x (+
x +1
−
1
+
2
( x+1 ) ( x +1 )3
2
dx
)
1 1
¿−ln x+ 2 ln ( x +1 ) + − +C .
x+1 ( x +1 )2

EXERCISES
Evaluate each of the following integrals.
1.∫
dx ( x−1 ) ( 2 x 3+ 4 x 2 +3 x+ 2 ) dx
2 8.∫
x ( x+2 ) 3
x ( x+1 )
( x +1 ) dx ( 5 v +3 ) dv
2.∫ 2 9. ∫ 2
x ( x −1 ) v ( v +1 ) ( v−3 )
( 5 y−4 ) dy ( 3 v 3−2 v 2−4 ) dv
3.∫ 3 2 10.∫ 2
y +4 y v ( v−1 ) ( v +2 )
dy
4. ∫ 3 9 ( x−1 ) dx
y −2 y
2 11. ∫ 2 2
x ( x −9 )
( 2 x 2+ 1 ) dx
5.∫ 3 ( x 3−12 x +4 ) dx
x −3 x+ 2 12.∫ 2
( x 2−7 ) dx ( x 2−3 x +2 )
6.∫ 3 ( x 2 +1 ) dx
x −12 x+ 16 13.∫ 3 2
( x 4 +1 ) dx x ( x−1 )
7.∫ 2 x dx
x ( x +1 )
2
14.∫
( x +2 )3
 y dy
15.∫
( y −3 )4

D. Quadratic Factors
Corresponding to a factor in the denominator of the form a x 2 +bx +c with b 2−4 ac< 0 , we assume the
A ( 2 ax+ b ) +B
partial fraction 2 , where A∧B are to be determined.
a x +bx +c

|
2
x +4 x+10
Example 3 ∫ 3 2
dx
x +2 x +5 x
Assume
x + 4 x +10 A B ( 2 x +2 )
2
C
3 2
= + 2 + 2 ,
x +2 x + 5 x x x +2 x +5 x +2 x+ 5
x 2+ 4 x +10=A ( x2 +2 x +5 ) + Bx ( 2 x+2 )+Cx .

Put x=0 : 5 A=10 , A=2

−1
Equate coefficients of x 2: A+2 b=1 , B= .
2
Equate coefficients of x : 2 A +2 B+C=4 , C=1
Therefore

( )
2
∫ xx3++24xx2+10
+5 x
2 1
dx=∫ − • 2
2 x+ 2
+ 2
x 2 x +2 x +5 x +2 x +5
1
dx

1 1 x +1
¿ 2 ln x− ln ( x + 2 x +5 ) + arctan
2
+C .
2 2 2
r
The case of repeated quadratic factors occurs less often. Corresponding to a factor ( a x 2 +bx +c ) , we
assume r partial fractions with linear numerators as above, and successive denominators building up step-
by-step just as in repeated linear factors.

EXERCISES
Evaluate the following integrals.
x dx y dy
1.∫ 2
7.∫ 3
x +6 x+ 13 ( y2+ 4)
( 4 x +5 ) dx y dy
2.∫ 2 8.∫
x +4 x +20 4
4 y +1
4 dx cos θ dθ
3.∫ 3 2 9. ∫
x −4 x + 3 x sinθ+ sin θ
3

10 dx 2
4. ∫ 3 2 10.∫
sec β dβ
4 x −4 x + 5 x 3
tan β + 4 tan β
dy 5 x dx
5.∫ 11. ∫ 3 2
( y −1 ) ( y 2 +1 ) x + x +4 x + 4
( 9 y +14 ) dy dx
6.∫ 12.∫ 2
( y −2 ) ( y 2+ 4 ) x √ 1−x
 ( x 3−4 ) dx
13.∫ 3 2
x +2 x +2 x
( x 3+10 ) dx
14.∫ 3 2
x −2 x +5 x
arctan y dy
15.∫ 3
y