MAT 2110 Tutorials

Dr A. Banda
Solutions to selected questions from Tutorial sheet 2, 3, and 5

Tutorial sheet 2
2d) We find the nth derivative of f (x) = 1
a+bx = (a + bx)−1 .
Solution:
f 0 (x) = −(a + bx)−2 .b = −b(a + bx)−2
f 00 (x) = 2b2 (a + bx)−3
f 000 (x) = −2 · 3b3 (a + bx)−4
f (4) (x) = 2 · 3 · 4b4 (a + bx)−5

We can see from the derivatives above that

(−1)n n!bn
f (n) (x) = (−1)n n!bn (a + bx)−(n+1) =
(a + bx)(n+1)

2e) We predict the nth derivative of f (x) = cos(ax)

Solution:
f 0 (x) = −a sin(ax)
f 00 (x) = −a2 cos(ax) = −a2 f (x)
f 000 (x) = a3 sin(ax) = −a2 f 0 (x)
f (4) (x) = a4 cos(ax) = a4 f (x) = −a2 f 00 (x)
f (5) (x) = −a5 sin(ax) = a4 f 0 (x) = −a2 f 000 (x)

We can see from the pattern above that each derivative is −a2 times the
second previous one. Therefore, a formula that gives all derivatives is
(
(−1)k an cos(ax) if n = 2k
f (n) (x) = (k = 1, 2, . . . )
(−1)k an sin(ax) if n = 2k − 1

1
 3) Let y = sec(kx), then y 0 = k sec(kx) tan(kx) and

y 00 = k 2 (sec(kx) tan(kx) tan(kx) + sec(kx) sec2 (kx))

= k 2 sec(kx)(tan2 (kx) + sec2 (kx))
= k 2 sec(kx)(sec2 (kx) − 1 + sec2 (kx))
= k 2 sec(kx)(2 sec2 (kx) − 1)
= k 2 y(2y − 1)

4) If y = tan x, then y 0 = sec2 x = 1 + tan2 x which is a polynomial in tan x

of degree 2.
Suppose that
y k = Pk+1 (tan x)
where Pk+1 is a polynomial of degree k + 1, then

d 0
y k+1 = (Pk+1 (tan x)) = Pk+1 (tan x)(1 + tan2 x).
dx
0 0
Here the degree of Pk+1 is k and so Pk+1 (tan x)(1 + tan2 x) has degree
k + 2. This proves that the nth derivative of y = tan x is in the form
Pn+1 (tan x).
5) (f g)0 = f 0 g + f g 0 and (f g)00 = f 00 g + f 0 g 0 + f 0 g 0 + f g 00 = f 00 g + 2f 0 g 0 + f g 00 .
6b)

d x2
   
d x−y
= +1
dx x + y dx y
(1 − y 0 )(x + y) − (x − y)(1 + y 0 ) 2xy − x2 y 0
⇒ =
(x + y)2 y2
0 0 0 0
x + y − xy − yy − (x + xy − y − yy ) 2xy − x2 y 0
⇒ =
(x + y)2 y2
0 2 0
2y − 2xy 2xy − x y
⇒ =
(x + y)2 y2
 2 
x 2x 2xy 2y
⇒ − y0 = 2 −
y2 (x + y)2 y (x + y)2
2xy(x + y)2 − y 3
⇒ y0 = 2
x (x + y)2 − 2xy 2

2
 y2
7b) Differentiating x + 2y + 1 = (x−1) implicitly, we get

2yy 0 (x − 1) − y 2
1 + 2y 0 = .
(x − 1)2
At (2, −1) we get
−2y 0 − 1
1 + 2y 0 =
1
so that
−1
y0 =
2
Therefore, the equation of the tangent is
−1
y+1= (x − 2) or 2y + x = 0
2
√
7d) Differentiating 2x + y − 2 sin(xy) = π/2 implicitly, we get
√
2 + y 0 − 2 cos(xy)(y + xy 0 ) = 0.
At (π/4, 1) we get
√ 1 π
2 + y0 − 2 √ (1 + y 0 ) = 0
2 4
so that
π 0 π
2 + y0 − 1 − y = 0 or y 0 (1 − ) = −1.
4 4
Thus
4
y0 = −
4−π
Therefore, the equation of the tangent is
4 π
y−1=− (x − )
4−π 4
8c) Differentiating x3 − 3xy + y 3 = 1 implicitly, we get
3x2 − 3y − 3xy 0 + 3y 2 y 0 = 0
3x2 − 3y x2 − y
⇒ y0 = =
3x − 3y 2 x − y2
Differentiating again, we get
6x − 3y 0 − 3y 0 − 3xy 00 + 6y(y 0 )2 + 3y 2 y 00 = 0
(3y 2 − 3x)y 00 = 6y 0 − 6(y 0 )2 − 6x
2y 0 − 2(y 0 )2 − 2x
y 00 =
y2 − x
x2 −y
Substitute y 0 = x−y 2 and simplify to obtain
y − x2 (x2 − y)2
 
00
y =2 + +x
x − y2 (x − y 2 )2

3
 Tutorial sheet 3
1
3b) i) Let f (x) = x and x0 = 2. Then

1 2
f 0 (x) = − and f 00 (x) = 3 .
x2 x
Thus the linearization about x = 2 is
1 1
L(x) = − (x − 2).
2 4
So we have
1 1 1 1.997
≈ L(2.003) = − (0.003) = = 0.49925
2.003 2 4 4
ii) For 2 ≤ x ≤ 2.002, we have f 00 (x) > 0, i.e, f (x) > L(x). Hence, the
sign of the error is positive (by Corollary 2.3.5).
iii) Observe that
2 1
|f 00 (x)| = 3 < f 00 (2) = .
x 4
So with k = 14 , we have (by Corollary 2.3.6)

1 1
|E(2.003)| < · (0.003)2 = 0.000001125
4 2
1
So 2.003 lies in the interval

(0.49925, 0.49925 + 0.000001125) = (0.49925, 0.499251125).

8) Let f (x) = sin x, then f (a) = sin a, f (b) = sin b and f 0 (x) = cos x. By the
Mean Value Theorem, there is c ∈ (a, b) such that

f (b) − f (a) sin b − sin a

cos c = f 0 (c) = =
b−a b−a
But | cos c| ≤ 1, so

| sin b − sin a| sin b − sin a

= ≤1
|b − a| b−a

from which the result follows.

4
 Tutorial Sheet 5
1e) To evaluate x(ln x)3 dx, we let u = ln x. Then eu = x and dx = eu du.
R

Thus we now evaluate Z

e2u u3 du

by parts.
R √ √
1f) To evaluate xe x dx, we let u = x. Then u2 = x and dx = 2udu. Thus
we now evaluate Z
2 u3 eu du

by parts.
1g) Let I = x2 tan−1 xdx, U = x2 and V = tan−1 xdx. Then dU = 2xdx
R

and V = x tan−1 x − 21 ln(1 + x2 ) Thus

Z
1 1
I = x2 (x tan−1 x − ln(1 + x2 )) − 2 x(x tan−1 x − ln(1 + x2 ))dx
2 2
Z Z
−1 1 −1
= x (x tan x − ln(1 + x )) − 2 x tan xdx − x ln(1 + x2 )dx
2 2 2
2
Thus
Z
1
I = x2 (x tan−1 x − ln(1 + x2 )) − 2I − x ln(1 + x2 )dx
2

Let u = 1 + x2 , then du = 2xdx and

Z Z
x ln(1 + x2 )dx = 2 ln udu

= 2(u ln u − u + C)
= 2((1 + x2 ) ln(1 + x2 ) − (1 + x2 )) + C

So we have
1
I + 2I = x2 (x tan−1 x − ln(1 + x2 )) − 2((1 + x2 ) ln(1 + x2 ) − (1 + x2 )) + C
2
and solving for I and simplifying, we get
  2  
1 3 −1 5x 2 2
I= x tan x − + 2 ln(1 + x ) − 2(1 + x ) + C
3 2

2e) The integral should be

sin3 (ln x) cos3 (ln x)

Z
dx.
x

5
 Let u = ln x, then du = x1 dx and

sin3 (ln x) cos3 (ln x)

Z Z
dx = sin3 u cos3 udu
x

Thus you can now use the identity sin2 u + cos2 u = 1.

3d) In = secn xdx
R

Z
I1 = sec xdx = ln | sec x + tan x| + C
Z
I2 = sec2 xdx = tan x + C

For n ≥ 3 we have
Z Z
n
In = sec xdx = secn−2 x sec2 xdx.

Let U = secn−2 x and dV = sec2 xdx, then

dU = (n − 2) secn−3 x sec x tan xdx = (n − 2) secn−2 x tan xdx

and V = tan x. Thus

Z
n−2
In = sec x tan x − (n − 2) tan x secn−2 x tan xdx
Z
= secn−2 x tan x − (n − 2) tan2 x secn−2 xdx
Z
n−2
= sec x tan x − (n − 2) (sec2 x − 1) secn−2 xdx
Z
= secn−2 x tan x − (n − 2) secn x − secn−2 xdx
Z Z
n−2
= sec x tan x − (n − 2) sec xdx − (n − 2) secn−2 xdx
n

= secn−2 x tan x − (n − 2)In − (n − 2)In−2

Thus we have

(1 + n − 2)In = secn−2 x tan x − (n − 2)In−2

so that
secn−2 x tan x (n − 2)
In = − In−2
n−1 n−1
To find I6 we apply the above formula:
sec x tan x 1 sec x tan x 1
I3 = − I1 = − ln | sec x + tan x| + C
2 2 2 2

6
 sec2 x tan x 2 sec2 x tan x 2
I4 = − I2 = − tan x + C
3 3 3 3
sec3 x tan x 3
I5 = − I3
4 4
3

sec x tan x 3 sec x tan x 1
= − − ln | sec x + tan x| + C
4 4 2 2
sec3 x tan x 3
= − (sec x tan x − ln | sec x + tan x|) + C
4 8
sec4 x tan x 4
I6 = − I4
5 5
sec4 x tan x 4 sec2 x tan x 2

= − − tan x + C
5 5 3 3
4
sec x tan x 4
sec2 x tan x − 2 tan x + C


= −
5 15
sec5 tan x 5
I7 = − I5
6 6
sec5 tan x 5 sec3 x tan x 3

= − − (sec x tan x − ln | sec x + tan x|) + C
6 6 4 8
5
 
sec tan x 5 3 3
= − sec3 x tan x − sec x tan x + ln | sec x + tan x| + C
6 24 2 2

xdx
, we complete the square of 3 − 2x − x2 :
R
4i) For 3
(3−2x−x2 ) 2

3 − 2x − x2 = 3 + 12 − 12 − 2x − x2 = 4 − (x + 1)2

so Z Z
xdx xdx
3 = 3
(3 − 2x − x2 ) 2 (4 − (x + 1)2 ) 2
Let u = x + 1, then du = dx and x = u − 1. Substituting we have

(u − 1)
Z Z
xdx
3 = 3 du
2
(4 − (x + 1) ) 2 (4 − u2 ) 2
Z Z
u du
= 3 du − 3
2
(4 − u ) 2 (4 − u2 ) 2

You can now use the substitution u = 2 sin θ to get

Z Z Z
xdx 1 1
3 = sec θ tan θdθ − sec2 θdθ
(4 − (x + 1)2 ) 2 2 8