Guide to MST124 Unit 8

Introduction – sections 1 and 2

I am starting by combining sections 1 and 2 and presenting the material in more or less
the reverse order. This means starting with the big idea, the Fundamental Theorem of
Calculus, then filling in some details afterwards. Of course having worked through this
version you can read through the sections in the right order later!
Fundamental Theorem of Calculus
If you have a continuous function f(x) and you know its indefinite integral F(x) + c then
you can find what’s called the definite integral between the limits a and b (where b > a)
by evaluating F(b) – F(a). Symbolically
𝑏
∫ 𝑓(𝑥) 𝑑𝑥 = 𝐹(𝑏) − 𝐹(𝑎)
𝑎

Here is an example – the way that it is written with square brackets is standard, so please
use this notation.
2
∫ 2𝑥 𝑑𝑥 = [𝑥 2 ]12 = 4 − 1 = 3
1

So once you have integrated you find the value of the answer for the upper value of x
then subtract the value for the lower value of x, and the answer is simply a number.
Notice that you don’t need to put in a ‘+ c’ anywhere, as the ‘c’s would just cancel out.
Also, this is called the integral from 1 to 2 rather than the integral from 2 to 1.
Here is another example, showing how the sum rule applies (the constant multiple rule
applies as well):
2 2
𝑥2
∫ (𝑥 + 𝑒 ) 𝑑𝑥 = [ + 𝑒 ] = (2 + 𝑒 2 ) − (0 + 1) = 1 + 𝑒 2
𝑥 𝑥
0 2 0

Try Activity 8 on p134 which gives practice using square brackets.

What does the answer mean?
The absolute value of the answer gives the area between the curve and the x-axis.
However, this is ‘signed area’: area above the x-axis is positive, while area below the x-
axis is negative. So if the curve is above the x-axis all the way between a and b the value
of the definite integral will be positive, while if the curve is below the x-axis all the way
then the value of the definite integral will be negative.
If the curve crosses the x-axis between a and b then the answer will be the balance of the
positive and negative parts.
Look at Example 2 on p112 and try Activity 2 on p113.
 The curve on the left is x2 – 2x, and its
x3
indefinite integral is − x 2 + c. Evaluating
3
this from 0 to 3 gives
3
x3
[ − x 2 ] = (9 − 9) − (0 − 0) = 0
3 0

so the positive and negative areas B and A

are the same! If you really want the actual
areas A and B added up then you need to
evaluate the integrals from 0 to 2 and 2 to 3
separately (or just find one of these integrals
and double the answer.) Read questions
carefully to see what’s wanted…
Try Activity 9 on p136, then read up to the end of section 2 (there’s not much new here)
and try some of Activities 10, 11 and 12. If you have time also try Activity 13 or 14.
Other points
If the limits of the integral are reversed then so is the sign of the answer.
𝑎 𝑏
∫ 𝑓(𝑥) 𝑑𝑥 = − ∫ 𝑓(𝑥) 𝑑𝑥
𝑏 𝑎

This point is the basis for Example 3 on p117 – an ‘extended definition’ of signed area is
given on the previous page. Some people find this point confusing but it’s all just an
illustration of the fact that –(b – a) = a – b.
Also you can approximate the area under a curve by dividing it by a series of vertical lines,
then treating the sections formed as rectangles. This helps to see that integration is a sort
of adding up. See Figure 2 on p107 of Book C.
Most of section 1 is about approximating areas: do read through it at some point and try
the activities, but move on if you are short of time and are safe with finding definite
integrals and areas.

3 Integration by substitution
3.1 Basic integration by substitution
This is probably the trickiest section of the unit, so take your time. The first 2 pages here
are a detailed explanation of how this integration technique is reversing the Chain Rule.
You know (or can work out) that differentiation sin(x2) gives you 2x cos(x2), therefore
integrating 2x cos(x2) must give you sin(x2) + c, but how would you approach this if you
were just given 2x cos(x2) to integrate? You need to spot that it is of the form
f(something)  the derivative of the something. So with 2x cos(x2) you can see that it’s
cos(x2)  the derivative of x2.
This is the basic thing you need to spot in order to use integration by substitution.
Read pp156 and 157 to find out more, but make sure you read p158 as well, where there
is a practical approach. Here is the practical approach with a parallel example, with colour-
coding:

Step Example  cos x sin2 x dx

1 Recognise the integrand as f(something)  This is (sin x)2  cos x, where cos
the derivative of the something x is the derivative of sin x

2 Call the something ‘u’ and find du/dx u = sin x, du/dx = cos x

3 Rearrange this result to get du = … du = cos x dx

4 Rewrite the remaining part of the integrand sin2 x = u2

in terms of u instead of x

5 Identify the expression for du in the integral  u2 du

and substitute it with du, so that the whole
integral is now in terms of u instead of x

6 Integrate in terms of u u3/3 + c

7 Rewrite the answer in terms of x sin3 x

+c
3

The point of the technique is to arrive at a simple integral. So if when you get to step 5
you are not looking at a simple integral, go back to stage 1 and try something different!
(Though there are variations which we will see soon.)

3.2 Integration by substitution in practice

When you get the expression for du it could be that it’s a constant multiple of this which is
in the integral, so that you need adjust things slightly.
Look at Example 16 on p164 and see where the –½ came from in the answer. Also read
the paragraph underneath, and try a couple of sections from Activities 25 and 26 on p166.
Also try Activity 27 on this page.

3.3 Integrating functions of linear expressions

If the ‘something’ in the integral is just a linear expression then there are useful shortcuts
for doing the integration. Look at the green box on p169, where F(x) + c is the indefinite
integral of f(x). Here are 3 simple examples to illustrate this:

1 1 2𝑥+3
∫ 𝑒 𝑥+3 𝑑𝑥 = 𝑒 𝑥+3 + 𝑐 ∫ 𝑒 2𝑥 𝑑𝑥 = 𝑒 2𝑥 + 𝑐 ∫ 𝑒 2𝑥+3 𝑑𝑥 = 𝑒 +𝑐
2 2

The x + 3 acts in the same Check how this works by Check this one too – why is
way as x on its own differentiating the answer the ½ there in the answer?
The middle example is particularly useful, and I would recommend making a note of how
to integrate eax, sin ax and cos ax on p7 of your Handbook, where a is a constant.

Try a few of Activity 28 on p168, then look at Example 18 and try at least one of Activity
29. Also try a few of Activity 30 on p170. Example 19 and Activity 31 can be left out on a
first reading.

3.4 Integration by substitution for definite integrals

You can do definite integration by substitution simply by leaving out the limits, finding the
indefinite integral then putting the limits back again. However, there is an alternative
method: when you change the integral into a u integral you can change the limits to u
limits at the same time. Then you can complete the process without going back to x at all.

(Note that what you definitely need to avoid is mixing up the methods and writing a u
integral with x limits…)

These two approaches are illustrated in Examples 20 and 21 on pp172 and 173. See which
one of these you like and try a couple of Activity 32 on p174.

3.5 Finding more complicated integrals by substitution

A final twist in this section is the situation where you do the substitution but there is still
an x or a simple expression with x in left behind. In order to deal with this you need to
rearrange the relationship between u and x and substitute for this final expression as well.
This is illustrated in Example 22 on p175, though the fractions here do get a bit much, so
here is a simpler example:

u = x − 2,
∫ x(x − 2)3 dx
du
= 1, du = dx
= ∫(u + 2)u3 du = (u4 + 2u3 ) du dx
x=u+2
1 1 (x − 2)5 (x − 2)4
= u5 + u4 + c = + +c
5 2 5 2

Try Activity 33 on p176, then celebrate completing the trickiest section of the Unit.

4 Integration by parts
4.1 Basic integration by parts
This method is based on the formula for the Product Rule, but skipping the detailed
explanation the essential idea is that if you are trying to integrate a product which cannot
be directly tackled, then you can replace one part with its derivative and the other part
with its indefinite integral (without the + c) and see if that looks better.
Often one of the factors of the product will be x or a linear factor of x. Generally this is the
factor to choose to differentiate, as its derivative will just be a constant. An exception is
any integral involving ln x – see section 4.2 below.
The formula in Lagrange notation is given in the green box on p177, and Example 23 on
p178 is very standard, with the x being differentiated and the other factor, sin x, being
integrated. If you prefer Leibnitz notation, see the green box at the bottom of p188
instead and there is also an informal version on p179, though it’s probably best to write
out all the bits at first. Here is a simple example using each of the notations:

Lagrange notation
I = ∫(x + 1) e2x dx
∫ 𝑓(𝑥)𝑔(𝑥) 𝑑𝑥 =
f(x) = x + 1 so f’(x) = 1
g(x) = e2x so G(x) = ½e2x
𝑓(𝑥)𝐺(𝑥) − ∫ 𝑓 ′ (𝑥)𝐺(𝑥) 𝑑𝑥
1 1
I = 2(x + 1)e2x − ∫ 2 e2x dx
Note how the half is at the front
when writing f(x) G(x) here – it’s 1 1 1 3
always good to put numbers first I = 2(x + 1) e2x + 4 e2x + c = (2x + 4) ex + c

Leibnitz notation
I = ∫(x + 1) e2x dx
𝑑𝑣 𝑑𝑢
∫𝑢 𝑑𝑥 = 𝑢𝑣 − ∫ 𝑣 𝑑𝑥 u = x + 1 so du/dx = 1
𝑑𝑥 𝑑𝑥
dv/dx = e2x so v = ½e2x
Rest is as above

Informal notation
Integral of product =
1st  antiderivative of 2nd – integral of (derivative of 1st  antiderivative of 2nd)
1st = x + 1, 2nd = e2x, derivative of 1st = 1, antiderivative of 2nd = ½e2x,…

Notice that when examples in this section mention swapping the expressions in the
integral they mean that the second expression is being chosen as the one to differentiate,
while the first one will be integrated. Also notice that integrals involving ex can usually be
simplified, while integrals with trig functions cannot.
Make sure you can follow Example 23, then try Activity 34 below. There are also examples
to look at in Activity 35 on p180.

4.2 Integration by parts in practice

There are (of course) some variations on the method.
1. The simple factor with x is a higher power of x, often a square. This means you
need to apply the technique twice, which is tedious but not difficult.
Look at Example 25 on p180 and try one section of Activity 36 on p181.
 2. The integral is of the form xn ln x. Here making ln x the factor which is
differentiated is the only possible approach.
Look at Example 26 on p182 and try a couple of sections from Activities 38 and 39
on p183. Notice in particular that Activity 38(b) gives you the integral of ln x, which
is not a standard integral.
3. Integrals of the form eax sin bx or eax cos bx. These are a bit special in that when
you have applied the technique twice you find that the original integral reappears.
You can then rearrange things to find the solution.
Look at Example 27 on p184, and try one section of Activity 40 if you have time.

4.3 Integration by parts for definite integrals

With definite integration by parts you can partially evaluate definite integrals as you go
along, as in Example 29 on p186, but it’s probably simplest to just find the indefinite
integral then apply the limits afterwards.
Try Activity 41, then Activity 42 or 43.

5 More integration
5.1 Trigonometric integrals
The 3 identities in the green box on p189 enable some particular trig integrals to be
found. The first two are particularly useful for integrating squares of sin and cos.
Look at Example 30 on p90 and try Activity 44 just below.
The rest of the section discusses alternative ways to integrate the product sin x cos x. This
is interesting but can be left out if you are pushed for time.

5.2 Choosing an integration method

This is a really important final section. The green box on p192 is a good summary, and
here is an example of each type listed:

Method Example

Standard integral 1
∫ 𝑑𝑥 = tan−1 𝑥 + 𝑐
1 + 𝑥2

Rearrange to get standard 2𝑥 2 + 3 3

integrals ∫ 𝑑𝑥 = ∫ (2𝑥 + ) 𝑑𝑥 = 𝑥 2 + 3 ln|𝑥| + 𝑐
𝑥 𝑥

Of the form f(ax) or f(x + b) or 𝜋 1 𝜋

∫ sin (3𝑥 + ) 𝑑𝑥 = − cos (3𝑥 + ) + 𝑐
f(ax + b) or a multiple of this (if in 4 3 4
doubt, use substitution)
 Of the form f()  the derivative 𝑥+1
of  (or a multiple of this) =
∫ (𝑥 2 +2𝑥−4)2 𝑑𝑥 Put  = x2 + 2x – 4
substitution

Of the form f(x) g(x) where f(x) ∫(𝑥 + 1)𝑒 2𝑥 𝑑𝑥 Put f(x) = x + 1, g(x) = e2x
becomes simpler when you
differentiate it = parts
∫ 𝑥 2 ln 𝑥 𝑑𝑥 Put f(x) = ln x, g(x) = x2
Note: this includes all integrals
with ‘ln’ in – differentiate the ln
part

A trig integral where a substitution 1

∫ 2 cos2 𝑥 𝑑𝑥 = ∫(1 + cos 2𝑥) 𝑑𝑥 = 𝑥 + sin 2𝑥 + 𝑐
would be useful 2

∫ tan2 𝑥 𝑑𝑥 = ∫(sec 2 𝑥 − 1)𝑑𝑥 = tan 𝑥 − 𝑥 + 𝑐

Try Activity 46 on p193.

5.3 Integration using a computer

Work through this section, noting in particular the difference between getting the exact
value of an area and a decimal approximation using quad_qags.