CHP 6
CHP 6
Chapter 6 Differentiation
1 / 50
Contents
1 Review
2 Techniques of Differentiation
3 Higher-order Derivatives
2 / 50
Review
Section 1
Review
3 / 50
Review
4 / 50
Review
f (a + h) − f (a)
lim ,
h→0 h
provided that this limit exists. The value of this limit, denoted by f ′ (a), is
called the derivative of f at x = a. In this case, f is said to be
differentiable at x = a.
Notes
(a) By putting x = a + h, we may equivalently express the definition of f ′ (a) as
f (x) − f (a)
f ′ (a) = lim .
x→a x−a
(b) The slope of the tangent line to the curve y = f (x) at P is f ′ (a).
5 / 50
Review
f (x + h) − f (x)
f ′ (x) = lim for all x ∈ I.
h→0 h
Notes
df dy ′
(a) Let y = f (x). Other notations for f ′ (x) include , , y and ẏ. Likewise,
dx dx
df dy
we may write f ′ (a) as , , y ′ (a) or ẏ(a).
dx x=a dx x=a
(b) Another way to obtain f ′ (a) is to evaluate f ′ (x) at x = a. This holds for a
large class of functions.
6 / 50
Review
7 / 50
Review
6. Formulae of differentiation
d
f (x) (f (x))
dx
c (c is a constant) 0
xr (r is a constant) rxr−1
ax (a is a positive constant and a ̸= 1) ax ln a
1
ln x
x
sin x cos x
cos x − sin x
tan x sec2 x
csc x − csc x cot x
sec x sec x tan x
cot x − csc2 x
d x
Note A particular case of the third formula is (e ) = ex .
dx
8 / 50
Review
7. Rules of differentiation
If both f and g are differentiable functions, then
d d
(a) [cf (x)] = c [f (x)] for every constant c.
dx dx
d d d
(b) [f (x) ± g(x)] = [f (x)] ± [g(x)].
dx dx dx
d d d
(c) [f (x)g(x)] = f (x) [g(x)] + g(x) [f (x)] (product rule).
dx dx dx
d d
−
d f (x) g(x) dx
[f (x)] f (x) dx
[g(x)]
(d) = (quotient rule).
dx g(x) [g(x)]2
9 / 50
Review
Example 1.1
dy
Find for each of the following.
dx
√ 5
y = 2x5 − 3x x − 6 + 4 ln x2 + 6
(a)
x
1
(b) y = x2 + cos x
x
3x − 1
(c) y= 2
x − 2x + 3
tan x
(d) y=
(6x + 1)2
10 / 50
Review
Solution
(a) Differentiating term by term, we obtain
√
dy d 5 5 2
= 2x − 3x x − 6 + 4 ln x + 6
dx dx x
d d 3
d d d
2x5 − 5x−6 +
= 3x 2 − (4 · 2 ln x) + (6)
dx dx dx dx dx
d d 3
d d
x5 − 3 x−6 + 8 (ln x) + 0
= 2 x2 − 5
dx dx dx dx
3 1 1
= 2 · 5x4 − 3 · x 2 − 5(−6)x−7 + 8 ·
2 x
9 1 30 8
= 10x4 − x 2 + 7 + .
2 x x
11 / 50
Review
12 / 50
Review
dy
dx
d 3x − 1
=
dx x2 − 2x + 3
d d
x2 − 2x + 3 dx (3x − 1) − (3x − 1) dx x2 − 2x + 3
= 2
(x2 − 2x + 3)
d d
d d d
x2 − 2x + 3 dx (3x) − dx (1) − (3x − 1) dx x2 − dx (2x) + dx (3)
= 2
(x2 − 2x + 3)
3 x2 − 2x + 3 − (3x − 1) (2x − 2)
= 2
(x2 − 2x + 3)
−3x2 + 2x + 7
= 2 .
(x2 − 2x + 3)
13 / 50
Review
14 / 50
Review
Example 1.2
15 / 50
Review
Solution
(a) Since
dy
= 12x3 − 12x,
dx
the slope of the tangent line at x = 2 is
dy
= 12(2)3 − 12 · 2 = 72.
dx x=2
16 / 50
Review
dy
= 0
dx
12x3 − 12x = 0
12x(x + 1)(x − 1) = 0
x = 0, −1 or 1,
17 / 50
Review
8. A useful rule of differentiation is the chain rule, which is applied to find the
derivatives of compositions of functions.
If both f and g are differentiable functions, then
d
[f (g(x))] = f ′ (g(x))g ′ (x).
dx
Note If we write y = f (u) and u = g(x), we may express the chain rule as
dy dy du
= · .
dx du dx
dy du
While this formula is easier to remember (if one regards both and as
du dx
dy du
‘quotients’), readers are reminded that du has not been defined and ,
du dx
are not actual quotients mathematically.
18 / 50
Review
Example 1.3
dy
Find for each of the following.
dx
5
(a) y = 2x2 + 1
4
(b) y = ecos(4x −9)
(c) y = ln sin2 x
(d) y = cos3 2x4 − x
r
x+1
(e) y =
x−1
19 / 50
Review
Solution
(a) Applying the chain rule, we obtain
h 5 i
2
d 2x + 1
dy d 2x2 + 1
= ·
dx d (2x2 + 1) dx
2
4
= 5 2x + 1 · 4x
4
= 20x 2x2 + 1 .
4
−16x3 ecos(4x −9)
sin 4x4 − 9 .
=
20 / 50
Review
21 / 50
Review
(e) Applying both the quotient rule and the chain rule, we obtain
1
x+1 2
d x−1 d x+1
dy x−1
= ·
dx x+1
d x−1 dx
− 12 d d
(x − 1) dx (x + 1) − (x + 1) dx (x − 1)
1 x+1
= · 2
2 x−1 (x − 1)
1 (x − 1) − (x + 1)
= ·
(x − 1)2
q
x+1
2 x−1
1
= − 1 3 .
(x + 1) (x − 1) 2
2
22 / 50
Techniques of Differentiation
Section 2
Techniques of Differentiation
23 / 50
Techniques of Differentiation
1. Implicit differentiation
(a) The functions we have encountered so far are expressed as
y = f (x),
F (x, y) = 0,
24 / 50
Techniques of Differentiation
Example 2.1
dy
Find if
dx
(a) x3 + 4xy 2 − y 3 = 7.
(b) y ln x = xexy − 1.
(c) x2 − xy 2 = tan(x − y).
6
(d) y = 3x2 + y 2 .
25 / 50
Techniques of Differentiation
Solution
(a) Differentiate both sides of the given function with respect to x:
d d
x3 + 4xy 2 − y 3
= (7)
dx dx
d d
3x2 + 4 xy 2 − y3
= 0
dx dx
d dx d
3x2 + 4 x y2 + y2 y3
− = 0
dx dx dx
" #
2
2 d y dy 2 d y 3 dy
3x + 4 x · +y − · = 0
dy dx dy dx
dy dy
3x2 + 4 2xy + y 2 − 3y 2 = 0
dx dx
dy
8xy − 3y 2 = −3x2 − 4y 2
dx
Hence
dy 3x2 + 4y 2
= 2 .
dx 3y − 8xy
26 / 50
Techniques of Differentiation
d d
(y ln x) = (xexy − 1)
dx dx
d dy d xy dx d
y (ln x) + ln x = x (e ) + exy − (1)
dx dx dx dx dx
xy
y dy d (e ) d(xy)
+ ln x = x · + exy
x dx d(xy) dx
y dy xy dy dx
+ ln x = xe x +y + exy
x dx dx dx
Thus,
dy y
x2 exy − ln x = − xyexy − exy ,
dx x
i.e.
dy y − x2 yexy − xexy
= .
dx x (x2 exy − ln x)
27 / 50
Techniques of Differentiation
d d d
x2 − xy 2
= [tan(x − y)]
" dx dx # dx
d y 2 dy dx d [tan(x − y)] d(x − y)
2x − x · + y2 = ·
dy dx dx d(x − y) dx
dy 2
2 dy
2x − 2xy +y = sec (x − y) 1 −
dx dx
Thus,
2 dy
sec (x − y) − 2xy = sec2 (x − y) + y 2 − 2x,
dx
i.e.
dy sec2 (x − y) + y 2 − 2x
= .
dx sec2 (x − y) − 2xy
28 / 50
Techniques of Differentiation
Example 2.2
Find the slope of the tangent line to the curve with the equation
30 / 50
Techniques of Differentiation
Solution We first differentiate both sides of the given equation with respect to x:
d d
x2 + 2ey − 3
[ln(x + sin y)] =
dx dx
d [ln(x + sin y)] d(x + sin y) d d y d
x2 + 2
· = (e ) − (3)
d(x + sin y) dx dx dx dx
d (ey ) dy
1 dx d(sin y) dy
+ · = 2x + 2 ·
x + sin y dx dy dx dy dx
1 dy dy
1 + cos y = 2x + 2ey (1)
x + sin y dx dx
Putting x = 1 and y = 0 on both sides of (1) yields
1 dy dy
1 + cos 0 · = 2 · 1 + 2e0 ,
1 + sin 0 dx x=1 dx x=1
so that
dy
= −1 ,
dx x=1
which is the slope of the tangent line to the curve at the point (1, 0).
dy
Remark It is unnecessary to solve for in terms of x and y before calculating
dx
dy
.
dx x=1
31 / 50
Techniques of Differentiation
d 1 d 1
sin−1 x = √ cos−1 x = − √
and .
dx 1 − x2 dx 1 − x2
(ii) If x ∈ R, then
d 1
tan−1 x =
.
dx 1 + x2
32 / 50
Techniques of Differentiation
Proof.
(Optional) We apply implicit differentiation to prove the result for y = sin−1 x only: with
sin y = x,
we have
d(sin y) dx
=
dx dx
d(sin y) dy
· = 1
dy dx
dy
cos y = 1.
dx
π π
Since − ≤ y ≤ , it follows that
2 2
q p
cos y = 1 − sin2 y = 1 − x2
p
(the case that cos y = − 1 − sin2 y is rejected because cos y ≥ 0). Hence
d dy 1 1
sin−1 x =
= = √ .
dx dx cos y 1 − x2
33 / 50
Techniques of Differentiation
Example 2.3
dy
Find for each of the following.
dx
2x − 7
(a) y = sin−1
3
−1
(b) y = x cos (3x)
1
(c) y = tan−1 x + tan−1
x
34 / 50
Techniques of Differentiation
Solution
(a) Applying the chain rule, we obtain
35 / 50
Techniques of Differentiation
(b) Applying both the product rule and the chain rule, we obtain
dy d −1 dx
cos (3x) + cos−1 (3x)
= x
dx dx dx
d cos−1 (3x) d(3x)
= x · + cos−1 (3x)
d(3x) dx
" #
1
= x −p 3 + cos−1 (3x)
1 − (3x)2
3x
= cos−1 (3x) − √ .
1 − 9x2
36 / 50
Techniques of Differentiation
(c) Applying both the quotient rule and the chain rule, we obtain
dy d d 1
tan−1 x + tan−1
=
dx dx dx x
−1 1 1
1 d tan x d x
= + ·
d x1
1 + x2 dx
1 1 1
= + 2 − 2
1 + x2 1 + x1 x
1 1 1
= − x2 +1 · 2
1 + x2 x2
x
1 x2 1
= − ·
1 + x2 x2 + 1 x2
= 0.
37 / 50
Techniques of Differentiation
2. Logarithmic differentiation
dy
(a) Logarithmic differentiation is used to obtain the derivative of a function
dx
when the function y
(i) involves products and/or quotients of expressions of x, or
(ii) takes the form f (x)g(x) .
(b) Steps of logarithmic differentiation
(i) Take natural logarithm of the given function and differentiate both sides of the
new expression with respect to x
dy
(ii) Express in terms of x only
dx
38 / 50
Techniques of Differentiation
Example 2.4
dy
Find for each of the following.
dx
(x + 1)2
(a) y=
(x − 2)3 (3x + 4)5
√
4
x+1
(b) y=
x(x − 2)3
(c) y = xsin x
1
(d) y = 1 − x2 x
39 / 50
Techniques of Differentiation
Solution
(a) Taking natural logarithm of y gives
" #
(x + 1)2
ln y = ln
(x − 2)3 (3x + 4)5
h i h i
2 3 5
= ln (x + 1) − ln (x − 2) (3x + 4)
= 2 ln(x + 1) − 3 ln(x − 2) − 5 ln(3x + 4).
1 dy d d d
· = 2 [ln(x + 1)] − 3 [ln(x − 2)] − 5 [ln(3x + 4)]
y dx dx dx dx
d [ln(x + 1)] d(x + 1) d [ln(x − 2)] d(x − 2) d [ln(3x + 4)] d(3x + 4)
= 2 −3 −5
d(x + 1) dx d(x − 2) dx d(3x + 4) dx
2 3 5
= − − ·3
x+1 x−2 3x + 4
Hence
(x + 1)2
dy 2 3 15 2 3 15
=y − − = − − .
dx x+1 x−2 3x + 4 (x − 2)3 (3x + 4)5 x+1 x−2 3x + 4
40 / 50
Techniques of Differentiation
1
= ln(x + 1) − ln x − 3 ln(x − 2).
4
Differentiate both sides of the above expression with respect to x:
1 dy 1 d d d
· = [ln(x + 1)] − (ln x) − 3 [ln(x − 2)]
y dx 4 dx dx dx
1 d [ln(x + 1)] d(x + 1) d d [ln(x − 2)] d(x − 2)
= · − (ln x) − 3 ·
4 d(x + 1) dx dx d(x − 2) dx
1 1 1 1
= · − −3·
4 x+1 x x−2
Hence
√
4
dy 1 1 3 x+1 1 1 3
=y − − = − − .
dx 4(x + 1) x x − 2 x(x − 2)3 4(x + 1) x x − 2
41 / 50
Techniques of Differentiation
42 / 50
Techniques of Differentiation
43 / 50
Techniques of Differentiation
44 / 50
Techniques of Differentiation
Example 2.5
45 / 50
Techniques of Differentiation
Solution
(a) Note that f (0) = 2, i.e.
f −1 (2) = 0.
Since
f ′ (x) = 3x2 + 4,
it follows from the inverse function theorem that
′ 1 1 1 1
f −1 (2) = = ′ = = .
f ′ (f −1 (2)) f (0) 3(0)2 + 4 4
π
(b) Note that f = −π, i.e.
2
π
f −1 (−π) = .
2
Since
f ′ (x) = − sin x − 2,
it follows from the inverse function theorem that
′ 1 1 1 1
f −1 (−π) = = ′ π = = − .
f ′ (f −1 (−π)) f 2 − sin π2 − 2 3
46 / 50
Higher-order Derivatives
Section 3
Higher-order Derivatives
47 / 50
Higher-order Derivatives
Let y = f (x).
dy
1. The derivative of f , (or y ′ or f ′ ), is itself a function of x. If we
′
dx
differentiate f with respect to x, the resulting function is known as the
d2 y ′′
second-order derivative of f and is denoted by , y or f ′′ . Repeating this
dx2
process gives the third-order , fourth-order derivatives, etc. of f . These are
called the higher-order derivatives of f .
dn y (n)
2. In general, the nth derivative of f , denoted by ,y or f (n) , is obtained
dxn
by differentiating f with respect to x successively for n times, i.e.
d2 y
d dy
=
dx2 dx dx
and inductively,
dn y dn−1 y
d
n
= for n = 2, 3, . . . .
dx dx dxn−1
2 3
d2 y d3 y
dy dy
Note In general, ̸= , ̸= , etc.
dx2 dx dx3 dx
48 / 50
Higher-order Derivatives
Example 3.1
x3 − 1
(a) Let f (x) = . Find f ′ (x), f ′′ (x) and f ′′′ (x).
x
(b) Let g(x) = 5x3 − 2x2 + 6x + 1. Find all higher-order derivatives of g.
(c) Let h(x) = xex . Find all higher-order derivatives of h.
Solution
(a) Note that f (x) = x2 − x−1 . Differentiating f with respect to x successively,
we obtain
1
f ′ (x) = 2x − (−1)x−2 = 2x + x−2 = 2x + ,
x2
2
f ′′ (x) = 2 + (−2)x−3 = 2 − 2x−3 = 2 − ,
x3
6
f ′′′ (x) = (−2)(−3)x−4 = .
x4
49 / 50
Higher-order Derivatives
g ′ (x) = 15x2 − 4x + 6 ,
g ′′ (x) = 30x − 4 ,
g ′′′ (x) = 30 ,
(4)
g (x) = 0.
From this, we see that
g (n) (x) = 0 for n = 4, 5, . . . .
(c) Applying the product rule to differentiate h with respect to x thrice, we
obtain
d x dx
h′ (x) = x (e ) + ex = xex + ex = (x + 1)ex ,
dx dx
d d
h′′ (x) = (x + 1) (ex ) + ex (x + 1) = (x + 1)ex + ex = (x + 2)ex ,
dx dx
d d
h′′′ (x) = (x + 2) (ex ) + ex (x + 2) = (x + 2)ex + ex = (x + 3)ex .
dx dx
Inductively,
h(n) (x) = (x + n)ex for n = 1, 2, . . . .
50 / 50