14: DIFFERENTIATION

The gradient of a curve the exponent (power) by the coefficient of x, then

A curve has a variable gradient, unlike that of a subtract one from the exponent.
straight line which is fixed. So, the gradient of a
curve is defined at any given point on the curve. Differential of polynomials of the form 𝒚 = 𝒂𝒙𝒏
We use the following rules to differentiate
The gradient of a curve at a point is defined as the polynomials.
gradient of the tangent to the curve at that point.
y = ax n y = xn
To determine the gradient of the curve at a point, we dy dy
= nax n - 1 = nx n -1
may draw a straight line which ‘just touches’ the dx dx
curve at that point. This is the tangent to the curve at
that point. Then, we calculate the gradient of the As the examples below show, if the polynomial is not
tangent using any appropriate method. in the form y = x n , we may need to perform some
In the curve shown below, clearly, the gradient at A
algebraic manipulations so that the law can be
is different from the gradient at B or C or in fact at
applied. The rules of indices are often helpful here.
any other point.
Example 1
Differentiate each of the following functions.
dy
(i) y = x 4 = 4 x 4 - 1 = 4 x3
dx
dy
(ii) y = 3x 2 = 3 ( x2 - 1 ) = 6 x
dx
1 dy 1 3
(iii) y = x6 = ( 6 x 6 - 1 ) = x5
4 dx 4 2
1 dy 3
The method of finding the gradient of a curve by (iv) y = 3 = x -3 = -3x -3 - 1 = -3x -4 = - 4
x dx x
drawing the tangent and estimating the gradient of
the tangent is rather tedious. Furthermore, the result + 12
(v) 𝑦 = , - = 4𝑥 /0 = 4(−2𝑥 /0/6 ) = −8𝑥 /9
depends on the degree of accuracy in reading the 1,

scales on the graph, that is, the result is subjected to 8

=− 9
human error and errors due to the accuracy of the 𝑥
instrument in measuring.
1
In this chapter, we introduce a technique which dy 1 12 - 1 1
(vi) y = x = x 2 = x =
provides a shorter, simpler and an exact method for dx 2 2 x
finding the gradient of a curve at any point. This
; ; <
technique is called differentiation and once we know (vii) 𝑦 =
6
= 𝑥 /-
12 6 6
= − 0 𝑥 /-/6 = − 0 𝑥 /-/
√, 1,
the equation of the curve, we can apply the technique.
1
= >𝑥 9
2
The gradient function
If y is a function of x, then the derivative of y with 6 6 12 6 9
(viii) 𝑦 = = 𝑥 /9 = (−3𝑥 /9/6 ) = − 𝑥 /+
dy ?, < ? 1,? ?
respect to x, written as , is a new function that is 1
dx =− +
2𝑥
called the gradient function.
To obtain the gradient function from a polynomial
function of the form 𝑦 = 𝑎𝑥 % , we simply multiply
Differential of 𝒚 = 𝒌 and 𝒚 = 𝒌𝒙 Solution
dy dy First, ensure that each term is expressed in a form
If y = k then = 0. If y = kx, then =k
dx dx that is conformable to differentiation.
y = 6 x3 - 4 x -2
We can verify these results by recalling that the dy 8
= 6 ( 3x3 - 1 ) - 4 ( -2 x -2 - 1 ) = 18 x 2 + 3
gradient of any horizontal line, y = k is zero, and the dx x
gradient of the line y = kx is k.
Example 5
We may also derive these results by applying the rule
dy
for differentiation. Obtain an expression for if y = x2 (3 - x ) .
dx
dy
y = 4 º 4 x0 ,
dx
( )
= 4 0 x 0 - 1 = 0 , and
Solution
dy
y = 4 x, = 4 (1x1 - 1 ) = 4 We multiply before differentiating
dx y = x 2 ( 3 - x ) = 3x 2 - x3
dy
Example 2 = 3 ( 2 x 2 - 1 ) - 3x3 - 1 = 6 x - 3x 2
dx
Find the gradient of the curve y = x 3 at the point
1 Example 6
x=- .
4 dy x 4 - 3x + 2
Obtain an expression for if y = .
dx x2
Solution
y = x3 Solution
dy We divide before differentiating
= 3x3 - 1 = 3x
2
The gradient function,
dx x 4 - 3x + 2 3 2
y= = x 2 - + 2 = x 2 - 3x -1 + 2 x -2
1 x2 x x
\ The gradient of the curve at x = -
4 dy 3 4
2
= 2 x - 3 ( -1x -1 - 1 ) + 2 ( -2 x -2 - 1 ) = 2 x + 2 - 3
æ 1ö 3 dx x x
= 3ç - ÷ =
è 4 ø 16
Example 7
Differential of a sum or difference dy
y = ( 3 x - 1) .
2
Obtain an expression for if
When there is more than one term in the expression, dx
each term is differentiated separately.
Solution
Example 3 We may choose to expand before differentiating
y = ( 3x - 1) = ( 3x - 1)( 3x - 1) = 9 x 2 - 6 x + 1
2
dy
Find for y = 5 x 2 - 2 x 3
dx dy
= 9 ( 2 x 2 - 1 ) - 6 = 18 x - 6
dx
Solution
y = 5 x 2 - 2 x3 The Chain Rule
dy Sometimes the technique of expansion is not practical
= 5 ( 2 x 2 - 1 ) - 2 ( 3x3 - 1 ) = 10 x - 6 x
2

dx and may even fail to simplify an expression. For

example, examine the following expressions:
1
y = ( 2 x - 1) y = 3 x + 4 = ( x + 4) 3 and
46
Example 4
4 dy 1
y = 6 x3 - , obtain an expression for 1
= ( 3x - 1)
-
x 2
dx y= 2
3x - 1
If these expressions are expanded the numbers of The Product Rule
terms would be large. More so, those with negative The product rule is useful when differentiating a
and/or fractional indices would have an infinite product of two functions and in which a simple
number of terms. Therefore, we need to find a more multiplication such is not possible. The rule states:
efficient way of differentiating this type of
expression. The chain rule is used to differentiate dy du dv
such composite functions If y is of the form, y = uv , then = v +u .
dx dx dx
It is sometimes called ‘function of a function’ and the
method is often referred to as the ‘method of It is necessary in cases where the functions u and v
substitution’. cannot be combined with simple multiplication.

If we have a composite function such that

The Quotient Rule
y = f (t ) and t = g ( x)
We use the quotient rule when there is a quotient that
dy dy dt cannot be simplified using a simple division. The rule
then by the Chain rule = ´
dx dt dx states:

Example 8
du dv
Differentiate y = ( 2 x - 1) .
10 v -u
u dy dx dx .
If y = , then =
v dx v2
Solution
Applying the chain rule, we have: Example 10
y = ( 2 x - 1)
10
Differentiate y = ( 2 x + 1) 3x + 2 .

Let t = 2 x - 1, and y = t10

Solution
dt dy
=2 = 10t 9 Let u = 2 x + 1 and v = 3x + 2
dx dt
dy dy dt du
= ´ = 10t 9 ´ 2 = 20t 9 = 20 ( 2 x - 1)
9 =2
dx dt dx dx
dv
To determine , we use the chain rule
Example 9 dx
1 v = 3x + 2
Differentiate y = .
3x - 2 1
Let t = 3 x + 2 , so v = t 2
Solution dv 1 1
3
= ( 3x + 2 ) 2 ( 3) =
-

1 1 dx 2 2 3x + 2
= ( 3x - 2 ) 2
-
y=
3x - 2 dy du dv 3
= v +u = 3x + 2 ´ 2 + ( 2 x + 1) ´
Let t = 3x - 2 and y = t 2
-1 dx dx dx 2 3x + 2
6x + 3 18 x + 11
dt dy 1 - 1 -1 = 2 3x + 2 + =
=3 =- t 2 2 3x + 2 2 3x + 2
dx dt 2
Example 11
dy dy dt 1 - 1 -1
= ´ = - t 2 ´3 4x + 1
dx dt dx 2 Differentiate y = .
2x - 4
-3 -3
= =
2 ( 3x - 2 )
3 3
2 t Solution
Let u = 4 x + 1 and v = 2 x - 4 .
 du dv Using the limit of a chord to find the gradient
=4 =2 function
dx dx
Consider the function, y = x 2 . To obtain the
Applying the quotient law,
du dv differential from first principles of y = x 2 , let us first
v -u look at its graph, or more precisely, a section of its
dy dx dx
= , graph.
dx v2
For the curve, y = x 2 let P be some arbitrary point
dy ( 2 x - 4 ) 4 - ( 4 x + 1) 2 18
= =- on the curve with coordinates ( x, y ) . So, the
( 2x - 4) ( 2x - 4)
2 2
dx
(
coordinates of P in terms of x will be x, x 2 . )
Example 12
The normal to the curve y = 2 x3 - 5 x 2 at the point
( 2, - 4) crosses the x-axis at A.
Find the coordinates of A.

Solution
y = 2 x3 - 5 x 2
The gradient function of the curve,
dy
dx
( ) ( )
= 2 3x3 - 1 - 5 2 x 2 - 1 = 6 x 2 - 10 x

The gradient of the tangent at (2, −4) is

= 6(2)0 − 10(2) = 4
1
Hence, the gradient of the normal is - (the
4 Some distance away, say h, to the right of P, we
product of the gradients of perpendicular lines choose another point, Q, on the curve, with the
= -1 ). x coordinate of (𝑥 + ℎ). The coordinates of Q will be
To find the equation of the normal, we use [(𝑥 + ℎ), (𝑥 + ℎ)0 ]. We join P to Q as shown in the
𝑦 − 𝑦6
=𝑚 diagram below.
𝑥 − 𝑥6
y - ( -4 ) 1
=-
x-2 4
4 y + 16 = - x + 2
4 y = - x - 14
When y = 0, x = -14 .
Therefore, A ( -14, 0) .

Differentiation as a limit
Earlier, we used differentiation to find the gradient of
a curve at a point. We did so by obtaining the
12
gradient function, which is itself a function that
1,
Now, we now find the gradient of the chord PQ of
gives us the gradient at any point on the curve. For
the curve. Recall the formula for the gradient of a
polynomial functions of the form 𝑦 = 𝑎𝑥 % , we
12 straight line joining the points (𝑥6 , 𝑦6) and (𝑥0 , 𝑦0) is
simply used the procedure, = 𝑛𝑎𝑥 %/6.
1,
2- /2;
𝑚= , where
,- /,;
We will now discover how these procedures were
derived through another method. (𝑥6 , 𝑦6 ) = (𝑥, 𝑥 0 ) and (𝑥0 , 𝑦0 ) = [(𝑥 + ℎ), (𝑥 + ℎ)0]
 ( x + h ) - x2
2

Gradient of PQ =
( x + h) - x
The gradient function of the curve, y = x 2 is 2𝑥.
x + 2hx + h - x
2 2 2
= d 2
h Hence,
dx
( x ) = 2x .
2hx + h 2
=
h In a similar fashion, we can repeat the procedure for
= 2x + h other simple polynomials such as 𝑦 = 𝑥 9 by drawing
the curve. We start with a chord PQ and move the
Imagine the chord PQ, hinged at P and being rotated
point Q until it approaches P. The gradient of the
clockwise as shown below. At each new position of
Q, along the curve, we will find the chord getting tangent PQ is the gradient of the curve at the point P.
shorter and shorter.
For example, PQ1 is shorter than PQ , PQ2 is
shorter than PQ1 , PQ3 is shorter than PQ2 and so
on. In each case, the length of h decreases and Q
approaches P.

Let P be ( x, y ) and let the x – coordinate of Q be

x + h , Hence,
P = ( x, x 3 )

(
Q ( x + h) , ( x + h)
3
)
The gradient of PQ
( x + h ) - x3
3
x3 + 3hx 2 + 3h 2 x + h3 - x3
= =
( x + h) - x h
3hx 2 + 3h2 x + h3
= = 3x 2 + 3hx + h 2
As the chord, PQ approaches tangent to the curve at h
P, the gradient of PQ = 2𝑥 + ℎ (derived above). As h ® 0 , the chord PQ ® the tangent to the
curve at P. Let h = 0 , the gradient of PQ
When h ® 0 , the chord PQ ® the tangent to the
= 3x 2 + 3(0) x + (0) 2 = 3x 2
curve at P. The gradient of PQ approaches 2𝑥 + 0 =
2𝑥. ∴ 3x 2 is the gradient function of the curve 𝑦 = 𝑥 9 .
 d 3 Example 14
Hence,
dx
( x ) = 3x2 OR when y = x3 , dy
dx
= 3x 2 .
d
Find ( cos 2 x ) .
dx
These results enable us to derive a general rule for
differentiation. Solution
Let y = cos 2 x and t = 2 x
In general, 1V
d 2
dx
( x ) = 2x ∴ 𝑦 = cos 𝑡 and 1,
=2

d n 12
d 3
( x ) = 3x 2 dx
( x ) = nx n - 1 . 1V
= −sin 𝑡 = − sin 2𝑥
dx 𝑡
d 4 By the Chain Rule:
dx
( x ) = 4 x3 𝑑𝑦 𝑑𝑦 𝑑𝑡
= ×
𝑑𝑥 𝑑𝑡 𝑑𝑥

𝑑𝑦
Differentiation of Trigonometric Functions = − sin 2𝑥 × 2
𝑑𝑥
So far, we have found the differential of functions
such as polynomials and we have used a general rule 𝑑
to obtain their derivatives. To differentiate (𝑐𝑜𝑠 2𝑥) = −2𝑠𝑖𝑛 2𝑥
𝑑𝑥
trigonometric functions we cannot use the same
procedure. The process is more complex and involves We can now state the following:
the study of more advanced calculus.
𝑑
𝑠𝑖𝑛 𝑎𝑥 = 𝑎 𝑐𝑜𝑠 𝑎𝑥
Only the sine and the cosine functions are to be 𝑑𝑥
considered at this level. Their derivatives are:
𝑑
𝑐𝑜𝑠 𝑎𝑥 = −𝑎 𝑠𝑖𝑛 𝑎𝑥
𝑑𝑥
𝑑 𝑑
(sin 𝑥) = cos 𝑥 (cos 𝑥) = −sin 𝑥 Example 15
𝑑𝑥 𝑑𝑥
p
Find the gradient of y = sin x at x = .
4
We can use these results to determine the derivative
of functions such as sin 𝑎𝑥 and cos 𝑎𝑥. Solution
dy
Example 13 y = sin x , = cos x
1
dx
Find 1, (sin 3𝑥). p p 1
\ The gradient of the curve at x = is cos =
4 4 2
Solution
\ The gradient =
1
Let 𝑦 = sin 3𝑥 and 𝑡 = 3𝑥 2
1V
∴ 𝑦 = sin 𝑡 and 1,
=3
12
1V
= cos 𝑡 = cos 3𝑥
By the Chain Rule:
𝑑𝑦 𝑑𝑦 𝑑𝑡
= ×
𝑑𝑥 𝑑𝑡 𝑑𝑥
𝑑𝑦
= 𝑐𝑜𝑠 3𝑥 × 3
𝑑𝑥
𝑑
(𝑠𝑖𝑛 3𝑥) = 3𝑐𝑜𝑠 3𝑥
𝑑𝑥