Differentiation -An overview application to

Trigonometry, Logarithm and Exponents

HOW DO WE GO ABOUT DIFFERENTIATING THE BASIC TRIGONOMETRIC FUNCTIONS?

What sort of answers should you expect? Let’s start by taking a look back at the graph of
y  sin x and y  cos x

y  sin x y  cos x

Let’s take the graphical approach to this question. If we think about differentiating a function in
one variable, the derivatives represent the gradient of the function. So if we can find an expression
of the gradient of sin x , we have found the derivative of sin x . Looking carefully at the graph of
3   3
y  sin x , clearly we can see that there are turning points at ...  ,  , , ,... all these
2 2 2 2
points have a gradient of zero. We can also examine the gradient of y  sin x at other places.
Suppose we want to find the gradient at x  0 ,simply draw a line on top of the curve of y  sin x at
x0

From the graph , clearly at x  0 the gradient is equal to 1.

We do know that sin x is 2 periodic- that it repeats itself every 2 , then we know that the gradients at
the points ...  4 , 2 ,0, 2 , 4 ,... is equal to 1. We can use a similar visual argument at x   to see
that the line there is parallel to y   x , and so the gradient there is -1.

We can use the periodicity again to see that the gradient at the points ...  3 ,  ,  ,3 ,... is equal to -1.

If we then find the gradient of y  sin x at various other places, we can begin to build up a picture of
what derivative of sin x will look like. Although this “visual” method is a little crude (however we do
have more sophisticated methods), it does provide us with the correct graphical result. You will notice a
strangely familiar – it corresponds to our old friend cos x . Thus the derivatives of sin x is cos x

Similarly, you can do the same to y  cos x : this time, we find that the result is  sin x

CHAINS, PRODUCT AND QUOTIENT

Now that we have the basic results of

d
  sin x   cos x
dx
d
  cos x    sin x
dx
Now we can find the derivative of a whole range of trigonometric functions. Before we do that, lets have
a quick recap of the chain rule, product rule and quotient rules for differentiation
Let’s start with the chain rule. The formal definition is in the form

d
dx
 
f  g  x   f ' g  x   g ' x 
Don’t be daunted if this form of the rule isn’t familiar to you. If you’re used to the chain rule being
df df dg
written as something like   , not to worry the above definition is precisely equivalent.
dx dg dx
Example

Let’s take a look at the derivative of sin(2 x)

Solution

 
The first thing to do is to identify how this function relates to the standard form of f g  x  . Here we

 
can see that sin(2 x) is indeed in the form f g  x  where the f  x   sin x and g  x   2 x . Thus
we can easily find its respective derivatives, where f '  x   cos x and g '  x   2 . Putting everything
together yields the final result.

d    g ( x )  g '( x )
f g ( x) f'

sin ( 2 x )   cos  2 x   2
dx   
   
 2cos 2 x

Next let’s take a look at the product rule. This rule states that for two functions u and v in x :

d
 uv   u ' v  v ' u
dx

Let us jump straight in with an example.

Example

d
Evaluate  sin x cos x 
dx
Solution
Here we let

u  sin x and v  cos x

u '  cos x and v '   sin x
Putting them together

d
 uv   u ' v  v ' u
dx
  cos x  cos x     sin x  sin x 
 cos 2 x  sin 2 x
 d u
Finally, we look at the quotient rule. We use this rule to evaluate the derivative in the form of  ,
dx  v 
where u and v are functions of x . You will notice that this rule can be of product rule since
u 1
 u  . Hopefully you can spot this-meaning we can actually get by without knowing the quotient
v v
rule specifically. However, if you are more comfortable with an explicit formula, the quotient rule is:

d u ' v  v 'u
 uv  
dx v2

As before, we won’t go into derivation, just an example.

Example

d
Evaluate  tan x 
dx
Solution

sin x
We do know from our identity that tan x  .
cos x
Thus

u  sin x and v  cos x

u '  cos x and v '   sin x
Putting them together:

d u ' v  v 'u
 uv  
dx v2


 cos x  cos x     sin x  sin x 
 cos x 
2

cos 2 x  sin 2 x

cos 2 x
1

cos 2 x
 sec 2 x
Exercise
1. Use the product rule to evaluate the followings

a)
d 2
dx
 x cos x 
d  x6 
b)  tan x 
dx  2 
 2. Use the quotient rule to evaluate the followings
d
a)  cosec x 
dx
d
b)  cot x 
dx
3. Use the chain rule to evaluate
d 1 
a)  sin  4 x  
dx  2 
b)
d
dx
 cos 6 x 

d  2  x 
c)  sin   
dx   2 
d
4. Evaluate  x sec x 
dx

EXPONENTS AND LOGARITHMS

Let us consider looking at the gradient of these functions, y  x and y  x 2 . For y  x , the gradient
is uniform throughout – the gradient is 1, however for y  x 2 , the gradient is twice the x  coordinate.
What we see here is a functional relationship between the x  coordinate and the gradient of the line.

That is all well and good, but why is it interesting to us? In both of those cases, we might consider the
gradient to be the by-product of the way that we define the line or a curve. Simply differentiating the
equation of the line and the curve gives us the result.

Imagine a curve whose gradient was always equal to the equation of the curve itself. That would mean
that for any value of x , the gradient would be equal to the value the function itself takes. Sound a little
crazy? Well the exponential function does exactly that. Let’s take some values.

For the function e x , its derivative is also e x .Lets take some x values
x- Function: Derivative:
coordinates ex
ex
0 e0  1 e0  1
1 e1 e1
2 e2 e2
3 e3 e3
4 e4 e4
But what is this elusive e ?

SOME BACKGROUND CHECK

e is like  . It’s a real number that has an amazing property. Whereas  helps us describe the relationship
between the radius and the circumference of a circle, e helps us describe the function whose gradient is
always equal to the value of the function itself. A nifty property. I’m sure you’ll agree, but e is a real
number-don’t you ever forget that!. You can use your calculator to find the value of e ,to five decimal places
e  2.71828 . Below is the graph of the curve y  e x

From the graph, we can note some interesting properties of the exponential function y  e x

 e0  1
 As x gets very negative, the value of e x approaches 0 , but from above (so e x never takes a value
less than 0)
 As x gets larger, e x gets larger incredibly quickly.

One crucial thing to remember about e x is the very property that defines it: The gradient is equal to the
value the function itself takes. Stated formally, we write

d x
dx
 e   ex

Not too hard to remember, is it? The derivative of e x is e x .

Remember that the standard rules with power apply to e x too, so we have:

 e a  b  e a  eb
 e a  b  e a  eb
 e 
a b
 eab

We simply use the chain rule when we find the derivative of e   .i.e
f x

dx

d f  x
e 
 e f  x  f ' x 

Let’s try an example:

Find :  
d x2
dx
e

Solution:

Here f  x   x 2 then f '( x)  2 x . Substitute into the above formula, we have

d x2
dx
e   ex  2x
2

 2 xe x
2

LOGARITHMIC FUNCTION: INVERSE OF THE EXPONENTIAL FUNCTION

As a function, just like many of the “old function”, e x has an inverse. Although it has many interesting
properties in its own right, we’re going to think of the natural logarithm of x , written as ln  x  , as
simply the inverse of e x . As such, we can immediately state that:

ln  e x   x
eln  x   x

Since ln  x  is the inverse function of e x , the graph of ln  x  is a mirror image of the graph of e x
along the line y  x .

From the graph, we can see that:

 As we increase the value of x , we find that the gradient of ln  x  falls rapidly

 As we get very large values of x , the gradients approaches 0. Hence its gradient will never be less
than 0

If we carefully draw a graph of these results, we would find that it looks like the graph below.
Again, hopefully you will recognize this graph.
1
Yes, it is the graph of y  .
x

Admittedly, it’s restricted to the positive values of x , but then again ln  x  is only defined for
these values of x . So we’re O.K !. Stating it formally:

d 1
 ln x  
dx x

We again call the chain rule to find the derivative of logarithmic functions in the form
ln  f  x   . The derivative then :

d 1
 ln( f ( x)    f ' x 
dx f  x

Example
d
Find the derivative of  ln(tan x) 
dx
Solution
sin x 1
Here f  x   tan x  then f '  x   sec 2 x  .
cos x cos 2 x
Substituting into the formula, we get
d 1
 ln(tan x)    sec 2 x
dx tan x
1 1
 
sin x cos 2 x
cos x
cos x 1
 
sin x cos 2 x
1

sin x cos x
1 1
 
sin x cos x
 sec x csc x

Exercise:
1. Simplify
a) e 4 x e3 x
e6 x
b) 2 x
e
c) ln  e4 x 
ln  x3 
d) e
2. Evaluate
a)
d 2x
dx
e 
d  1 x3 
b)  e 
dx  3 

c)
d x2
dx

e ln  4 x  
 ln  cot x  
d
d)
dx
3. Use the product rule to find
d x
dx
 e sin x 
For those of you with keen eye for details, we expect that our “visual” method of finding
derivatives was a little unsatisfactory. Don’t be despair: the results that we gave are indeed true,
but there are more rigorous approaches to achieve them than we presented. One of the more
accessible comes from power series, and this is discussed fully in this book: Real Analysis (J.M.
Howie, Springer-Verlag, New York, 2001.)
Finally, for those of you keen to see a derivation of the chain rule, product rule and quotient rule,
we’d recommend that you dive into Calculus 1 (J. Marsden and A. Weinstein, Springer-Verlag,
New York, 1999). The derivations rely on the idea of taking limit, but if you’re interested in
taking a look, then the text explains everything clearly.