AS/1013/A5/2018-2019 Sem 1

THE UNIVERSITY OF HONG KONG

DEPARTMENT OF MATHEMATICS

MATH1013 University Mathematics II

Assignment 5

Due Date: 30 Nov. 2018, before 1730.

Part I Drilling Exercises

You need not submit solutions to these exercises.

1. Find the area of the bounded region which is enclosed by the curves
y = x3 − 1 and y = x − 1.

2. Let F (x) = ∫0 cost dt.

x

(a) Find F (0), F (π/2) and F (π).

(b) Find the derivative of F at x = 0.
(c) If g(x) = (x − 1)2 , find (F ○ g)′ (1).

3. Use the method of substitution to evaluate the following integrals.

3 x
(a) ∫ √ dx
0 x+1
dx
(b) ∫ √ 3
x x −1
sin 3θ
(c) ∫ 1 + cos 3θ dθ
1 dx
(d) ∫0 √
4 − x2
√
1/ 2 s
(e) ∫0 √ ds.
1 − s4

1
 4. Evaluate the following integrals using integration by parts.

(a) ∫ x sec2 x dx

(b) ∫ sin(ln x) dx

5. Evaluate the following integrals.

4x − 1
(a) ∫
x2 +x−2
x2 + x − 4
(b) ∫
x3 + x2 + x + 1
2x2 − 19x + 54
(c) ∫
x3 − 7x2 + 8x + 16
∞
6. Determine the improper integral ∫ x−2 dx if exists.
1

Part II Problems to be Submitted

Submit your solutions of the following questions via the MOODLE assign-
ment drop box. Please follow the instructions there.

1
7. By rationalizing the denominator of the integrand, evaluate ∫ √ √ dx.
x−1+ x+1
x2
8. Let f (x) = 3 + ∫ sec(t − 1) dt. Find the first Taylor polynomial of f
1
with reference point a = −1. Hence estimate f (−0.9).

9. Let f be a continuous function on [a, b]. Show that there is c ∈ (a, b)

such that
1 b
f (c) = ∫ f (x) dx.
b−a a
10. Let f be continuous on [0, a]. Show that
a a
∫0 f (t) dt = ∫ f (a − t) dt.
0

π t sin t
Hence evaluate ∫ dt.
0 1 + cos2 t

2
11. Karl Weierstrass noticed that the substitution t = tan(x/2) can always
transform a rational function of sin x and cos x into a rational function
in t. To be precise, suppose that −π < x < π, then t = tan(x/2) has
inverse.

(a) Show that

1 t2
cos2 (x/2) = and sin2 (x/2) = ;
1 + t2 1 + t2
and hence deduce that
1 − t2 2t
cos x = and sin x = .
1 + t2 1 + t2
dx 2
(b) Show that = .
dt 1 + t2
(c) Use the substitution t = tan(x/2), show that

sin x + 2 cos x − 3 2 (−5t2 + 2t − 1)

I =∫ =
sin x − 2 cos x + 3 ∫
dt.
(t2 + 1) (5t2 + 2t + 1)

(d) (Not to be handed in) It can be proved that

2 (−5t2 + 2t − 1) 4(10t − 1) 2(4t + 3)

= − .
(t + 1) (5t + 2t + 1) 5 (5t + 2t + 1) 5 (t2 + 1)
2 2 2

Evaluate I as a function of x.

12. Evaluate the following improper integrals whenever exist. Please indi-
cate if they diverge to infinity.
π/2 1
(a) ∫ dx
0 1 − cos x
∞ 2
(b) ∫ xe−x dx
−∞

3
 Suggested Solution

1. We first find the intersection points of the curves. Solving

x3 − 1 = x − 1
x(x2 − 1) = x3 − x = 0.

It follows that the two curves intersect when x = 0 and x = ±1. When
0 ≤ x ≤ 1,
(x3 − 1) − (x − 1) = x(x2 − 1) ≤ 0;
and the inequality reverses when −1 ≤ x ≤ 0. So the area of the region
is
0 1 1 1 0 1 1 1 1
∫−1 x(x2 −1) dx+ ∫ x(1−x2 ) dx = [ x4 − x2 ] +[− x4 + x2 ] = .
0 4 2 −1 4 2 0 2

2. Note that F (x) = [sin t]0 = sin x.

x

(a) F (0) = sin 0 = 0, F (π/2) = sin(π/2) = 1 and F (π) = sin π = 0.

(b) F ′ (x) = cos x and F ′ (0) = 1.
(c) If g(x) = (x−1)2 , then (F ○g)(x) = sin [(x − 1)2 ]. By the chain rule,
(F ○ g)′ (x) = 2(x − 1) cos[(x − 1)2 ]. It follows that (F ○ g)′ (1) = 0.

3. (a) Let u = x + 1. Then du = dx, u(0) = 1 and u(3) = 4. It follows that

x 3 4 u−1 4
√∫0 dx = ∫ du = ∫ (u1/2 − u−1/2 ) du
x+1 1 u1/2 1
4
2 2 2 8
= [ u3/2 − 2u1/2 ] = ⋅ 8 − 2 ⋅ 2 − + 2 = .
3 1 3 3 3
√
(b) Let u = x3 − 1. Then
3x2 3x3
du = √ dx = √ dx
2 x3 − 1 2x x3 − 1
2 1
du = √ dx.
3(1 + u )
2
x x3 − 1

4
 It follows that
dx 2 du 2 2 √
∫ √ 3 = ∫ =
1 + u2 3
arctan u+C = arctan x3 − 1+C.
x x −1 3 3

(c) Let u = 1 + cos(3θ). Then du = −3 sin(3θ) dθ and thus

sin 3θ 1 du 1 1
∫ dθ = − ∫ = − ln ∣u∣ + C = − (1 + cos 3θ) + C.
1 + cos 3θ 3 u 3 3
Note that 1 + cos 3θ ≥ 0 for all θ.
(d) Let x = 2 sin θ. Then sin 0 = 0, sin π/6 = 1/2, and dx = 2 cos θ dθ.
As a result,
1 dx π/6 2 cos θ π/6 2 cos θ π/6
∫0 √ =∫ √ dθ = ∫ dθ = ∫ dθ = π/6,
4 − x2 0 4 − 4 sin2 θ 0 2 cos θ 0
√
where we have used the fact that 4 − 4 sin2 θ = 2∣ cos θ∣ = 2 cos θ
as 0 ≤ θ ≤ π/6.
√
(e) Let t = s2 . Then t(0) = 0, t(1/ 2) = 1/2, and dt = 2s ds. It follows
that
√
1/ 2 s 1 1/2 dt 1 π
√ ds = ∫ √ = [arcsin t]0 = .
1/2
∫0
1 − s4 2 0 1 − t2 2 12

4. (a) Note that (tan x)′ = sec2 x. Integration–by–parts to get

∫ x sec2 x dx = ∫ x d tan x
sin x
= x tan x − ∫ tan x dx = x tan x − ∫ dx.
cos x
To evaluate the last integral, let u = cos x. Then du = sin x dx and
sin x du
∫ dx = ∫ = ln ∣ cos x∣ + C.
cos x u
To roundup,

∫ x sec2 x dx = x tan x − ln ∣ cos x∣ + C.

5
 (b) Let I = ∫ sin(ln x) dx. Integration–by–parts twice to get

I = x sin(ln x) − ∫ xd(sin(ln x)) = x sin(ln x) − ∫ cos(ln x) dx

= x sin(ln x) − (x cos(ln x) − ∫ x d(cos(ln x)))

= x sin(ln x) − (x cos(ln x) + ∫ sin(ln x) dx)

1
I = x (sin(ln x) − cos(ln x)) + C.
2

5. (a) Factorize the denominator to get

q(x) ∶= x2 + x − 2 = (x + 2)(x − 1).

Suppose the partial fraction decomposition of the integrand is

4x − 1 A B
f (x) ∶= = + .
x2+x−2 x−1 x+2
Multiply through by q(x) and group the like terms to get

4x − 1 = A(x + 2) + B(x − 1) = (A + B)x + (2A − B).

Comparing the coefficients to arrive at a system of linear equations

A + B = 4
{
2A − B = −1.

Solve the system to get A = 1 and B = 3. Thus,

1 3
∫ f (x) dx = ∫ dx + ∫ dx
x−1 x+2
= ln ∣x − 1∣ + 3 ln ∣x + 2∣ + C = ln ∣(x − 1)(x + 2)3 ∣ + C.

(b) Factorize the denominator to get

q(x) ∶= x3 + x2 + x + 1 = (x + 1)(x2 + 1).

6
 Suppose the partial fraction decomposition of the integrand is
x2 + x − 4 A Bx + C
f (x) ∶= = + 2 .
x +x +x+1 x+1
3 2 x +1
Multiply through by q(x) and group the like terms to get

x2 +x−4 = A(x2 +1)+(Bx+C)(x+1) = (A+B)x2 +(B+C)x+(A+C).

Comparing the coefficients to arrive at a system of linear equations

⎧
⎪ A + B = 1
⎪
⎪
⎨ B + C = 1
⎪
⎪
⎪
⎩ A + C = −4.

Solve the system to get A = C = −2 and B = 3. Thus,

−2 3x 2
∫ f (x) dx = ∫ dx + ∫ dx − ∫ dx
x−1 x2 + 1 x2 + 1
1 3
= ln + ln(x2 + 1) − 2 arctan x + C.
(x − 1)2 2

(c) Factorize the denominator to get

q(x) ∶= x3 − 7x2 + 8x + 16 = (x + 1)(x − 4)2 .

Suppose the partial fraction decomposition of the integrand is

2x2 − 19x + 54 A B C
f (x) ∶= = + + .
x − 7x + 8x + 16 x + 1 x − 4 (x − 4)2
3 2

Multiply through by q(x) and group the like terms to get

2x2 − 19x + 54 = A(x − 4)2 + B(x + 1)(x − 4) + C(x + 1)

= (A + B)x2 + (−8A − 3B + C)x + (16A − 4B + C).

Comparing the coefficients to arrive at a system of linear equations

⎧
⎪ A + B = 2 − − − − − (1)
⎪
⎪
⎨ −8A − 3B + C = −19 − − − − − (2)
⎪
⎪
⎪
⎩ 16A − 4B + C = 54 − − − − − (3)

7
 Subtract (2) from (3) to get

24A − B = 73 −−−−− (4)

From (1) and (4), we get A = 3 and B = −1; and then it follows
from (3) that C = 2. Thus,
3 1 2
∫ f (x) dx = ∫ dx − ∫ dx + ∫ dx
x+1 x−4 (x − 4)2
2
= ln ∣x + 1∣3 − ln ∣x − 4∣ − +C
x−4
(x + 1)3 2
= ln ∣ ∣− + C.
x−4 x−4

6. Note that f (x) = x−2 is continuous on [0, b] for any positive b. By the
fundamental theorem of calculus,
b b 1
∫1 x−2 dx = [−x−1 ]1 = 1 − .
b
It follows that
∞
∫1 x−2 dx = lim (1 − 1/b) = 1.
b→+∞

7. Rationalize the denominator of the integral to get

√ √
1 x+1− x−1
∫ √ √ dx = ∫
(x + 1) − (x − 1)
dx
x−1+ x+1
√ √
x+1− x−1 1 2 3 1 2 3
= ∫ dx = ⋅ (x + 1) 2 − ⋅ (x − 1) 2 + C
2 2 3 2 3
1 3 1 3
= (x + 1) 2 − (x − 1) 2 + C
3 3
(−1)2
8. Firstly, f (−1) = 3 + ∫1 sec(t − 1) dt = 3 + 0 = 3. Next, apply the
fundamental theorem of calculus and the chain rule to get

f ′ (x) = 0 + sec(x2 − 1)(x2 )′ = 2x sec(x2 − 1)

f ′ (−1) = 2(−1) sec((−1)2 − 1) = −2.

8
 It follows that
P1 (x) = 3 − 2(x + 1)

and hence f (−0.9) ≈ P1 (−0.9) = 3 − 2(.1) = 2.8.

9. This result is known as the mean value theorem for integration.

Consider the function
x
F (x) = ∫ f (t) dt, where a ≤ x ≤ b
a

which is continuous [a, b] and is differentiable on (a, b) by the funda-

mental theorem of calculus. Apply the mean value theorem to F , we
have
F (b) − F (a)
F ′ (c) = for some c ∈ (a, b).
b−a
Result follows immediately by the fundamental theorem of calculus
(F ′ (c) = f (c)).

10. Use the substitution x = a − t. Then

dx = −dt, x(0) = a, & x(a) = 0.

It follows that
a 0 a
∫0 f (t) dt = − ∫ f (a − x) dx = ∫ f (a − x) dx
a 0

as claimed. Apply this formula to the given integral with a = π and

t sin t
f (t) = , we have
1 + cos2 t
π π (π − x) sin(π − x) π (π − x) sin x
∫0 f (t) dt = ∫ dx = ∫ dx.
0 1 + cos2 (π − x) 0 1 + cos2 x

Rearrange the terms to get

π x sin x π π sin x
∫0 dx = ∫ dx.
1 + cos2 x 2 0 1 + cos2 x

9
 Finally, let u = cos x. Then

du = − sin x dx, u(0) = 1, & u(π) = −1;

and
π x sin x π −1 −du π π2
= = [arctan =
1
∫0 1 + cos2 x
dx ∫
2 1 1 + u2 2
u]−1
4
.

11. (a) Using the identity 1 + tan2 θ = sec2 θ with θ = x/2, we have
1
cos2 (x/2) = .
1 + t2
As cos2 (x/2) + sin2 (x/2) = 1, we get
t2
sin2 (x/2) = .
1 + t2
Next, as cos(2θ) = cos2 θ − sin2 θ, we have
1 − t2
cos x = .
1 + t2
It follows that
4t2
sin2 x = 1 − cos2 x =
(1 + t2 )2
2∣t∣
sin x = .
1 + t2
As −π < x < π, t = tan(x/2) has the same sign as sin x. Result
follows.
(b) Firstly, observe that
dt 1
= sec2 (x/2) > 0 whenever −π < x < π.
dx 2
It follows that t has a differentiable inverse and
dx 1 2
= = 2 cos2 (x/2) = .
dt dt/dx 1 + t2

10
(c) Use the substitution t = tan(x/2), we get
2(1−t2 )
2t
+ −3 2 2 (−5t2 + 2t − 1)
I =∫ t2 +1 t2 +1
⋅ dt = ∫ dt.
2t
t2 +1 − 2(1−t2 )
t2 +1 +3 t +1
2 (t2 + 1) (5t2 + 2t + 1)

(d) On the one hand, note that

6 6 3
∫ dt = arctan t + C = x+C
5(t2 + 1) 5 5
8t 4 2t 4 4
∫ dt = ∫ = ln(t2 + 1) + C = ln (1 + tan2 (x/2)) + C.
5(t + 1)
2 5 t +1 5
2 5
On the other hand,
1 2 4 1 2 2 2
5t2 + 2t + 1 = 5 (t + ) + = 5 [(t + ) + ( ) ] .
5 5 5 5
It follows that
4(10t − 1) 40(t + 1/5) − 12 8 t + 1/5 12 1
= = ⋅ −
5(5t + 2t + 1) 52 [(t + 1 ) + ( 2 ) ] 5 (t + 1 ) + ( 2 ) 25 (t + 1 )2 + ( 2 )2
2 2 2 2 2
5 5 5 5 5 5

and hence
4(10t − 1) 8 t + 1/5 12 dt
∫ dt = ∫ dt −
5(5t + 2t + 1)
2 5 (t + 1 2
) + ( 25 )
2
25 ∫ 2 2
(t + 15 ) + ( 25 )
5
4 1 2 2 2 12 5 5 1
= ln ((t + ) + ( ) ) − ⋅ arctan ( (t + )) + C
5 5 5 25 2 2 5
4 1 x 2 2 2 6 1 x
= ln (( + tan ) + ( ) ) − arctan ( (1 + 5 tan )) + C
5 5 2 5 5 2 2
To round up,
2 2
4 ⎛ ( 5 + tan 2 ) + ( 5 ) ⎞ 6
1 x 2
1 x 3
I = ln − arctan ( (1 + 5 tan )) − x + C
5 ⎝ 1 + tan 2
2 x
⎠ 5 2 2 5
4 1 + 52 tan x2 + tan2 x2 6 1 x 3
= ln ( 2 x ) − arctan ( (1 + 5 tan )) − x + C
5 sec 2 5 2 2 5
4 1 6 1 x 3
= ln (1 + sin x) − arctan ( (1 + 5 tan )) − x + C.
5 5 5 2 2 5

11
 1
12. (a) Note that f (x) = is continuous on [a, π] for any small
1 − cos x
positive a. Note that
1 1
f (x) = = csc2 (x/2).
1 − (1 − 2 sin (x/2)) 2
2

Using the substitution u = x/2, it’s readily seen that

1
∫ f (x) dx = csc2 (x/2) dx = ∫ csc2 u du = − cot(x/2) + C.
2∫
By the fundamental theorem of calculus,
π/2
f (x) dx = [− cot(x/2)]a = cot(a/2) − 1.
π/2
∫a

It follows that
π/2 1
∫0 f (x) dx = lim+ ( − 1) = +∞.
a→0 tan(a/2)

(This improper integral diverges to positive infinity.)

(b) Note that f (x) = xe−x is continuous on [0, b] for any positive b. So
2

it is Riemann integrable on [0, b]. Let u = −x2 . Then du = −2xdx

and
1 1 1 2
xe−x dx = − ∫ eu du = − eu + C = − e−x + C.
2
∫ 2 2 2
By the fundamental theorem of calculus,
b 1 2 b 1
xe−x dx = [− e−x ] = (1 − e−b ) .
2 2
∫0 2 2
0

It follows that
∞
f (x) dx = lim (1 − e−b ) /2 = 1/2.
2
∫0 b→+∞

Similarly, for a < 0 we have

0 1 −x2 0 1
f (x) dx = [− e ] = (−1 + e−a )
2
∫−∞ 2 2
a

12
whereby
0
lim (−1 + e−a ) /2 = −1/2.
∫−∞ f (x) dx = a→−∞
2

As a result, the improper integral in the question exists and

∞
∫−∞ f (x) dx = −1/2 + 1/2 = 0.

Warning. It may be tempting to use the property that definite

integral of an odd function over [−b, b] must be zero, where b is
a positive real number (NOT +∞). To be precise, since xe−x is
2

odd,
b
dx = 0 for any b > 0.
2
∫−b xe
−x

Therefor,
+∞ b
xe−x dx = lim ∫ xe−x dx = 0.
2 2
∫−∞ b→+∞ −b

Though the computation is correct, it is not how we define im-

proper integral which is more demanding. The value obtained in
this special way is known as Cauchy principal value.

13