Example 5: Substitution: , , ,
Integrate the following functions w.r.t. :
(i) (ii) (iii) (iv)
(i) Let .
Put . Then .
.
(ii) Let .
Put . Then .
.
(iii) Let .
Put . Then .
Put . Then .
.
(iv) Let .
Put . Then .
.
Want the reason behind every step, the basics and the checks?
Substitution is the chain rule run backwards. If the integrand has the shape , a function of some inner expression multiplied by the derivative of that inner expression, then putting , turns it into the simpler . In each part we look for an inner expression whose derivative is also sitting in the integrand (up to a constant factor).
- Given
- Four integrands: (i) (with ); (ii) ; (iii) , ; (iv) .
- To find
- for each.
- Plan
- For each: choose ; write in terms of ; rewrite the whole integral in ; integrate; put back; differentiate to check.
Basics you need
If , then and . After integrating, replace by . Basics: substitution, the chain rule backwards
, , (), , . Basics: the power rule
and (Section 7.2, formulae (iii) and (i)). Basics: integration undoes differentiation
lies in , where the cosine is positive (Chapter 2). Used for the simplified form in part (iv). Basics: trigonometric identities
Part (i):
Choose the substitution
The inner expression is , with derivative , a constant. Assume and put
The missing factor is a constant, and a constant factor can always be supplied by dividing by it. This needs ; if the integrand is and the integral is just a constant, so that case is trivial.Substitution, Section 7.3.1. Basics: substitutionIntegrate in and return to
is formula (iii). The answer must be in the original variable, so is replaced by at the end.Standard formula (iii), Section 7.2. Lesson: integration by substitutionCheck: differentiate the answer
Valid for every real .
The chain rule produces the factor (derivative of the inner ), which cancels the : exactly the factor we divided by in Part (i), Step 1.
Part (ii):
Spot the inner function and its derivative
The sine is applied to , whose derivative is the other factor. Put
This is the ideal situation for substitution: the whole of is replaced by , and nothing is left over in .Integrate and return to
Formula (iii) again, then is put back.Standard formula (iii), Section 7.2. Lesson: integration by substitutionCheck: differentiate the answer
Valid for every real .
Outer: differentiates to ; inner: differentiates to . The product is the integrand.
Part (iii):
Where the integrand exists
is in a denominator, so ; and , need , that is
These are the conditions under which every symbol in the integrand has a value; the answer will be valid on each interval where they hold.First substitution:
. Put
Then
The in the integrand is, up to the factor 2, the derivative of the inner expression . So becomes , and every becomes . The book writes this as and then cancels the against the in the denominator; the result is the same.Power rule with . Basics: the power ruleSecond substitution:
Now is the derivative of . Put
Again an inner function () appears together with its derivative (), so the integral collapses to a power of . Formula (i) with finishes it.Standard formulae (iv) and (i), Section 7.2. Lesson: integration by substitutionReturn to
Undo the substitutions in reverse order: , then . The book also suggests ("Alternatively") the single substitution , without working it out. It does both jobs at once: then , and the integral is directly.Check: differentiate the answer
Three layers, three factors from the chain rule: the fifth power gives , the tangent gives , and the square root gives . The constants .
Part (iv):
Choose the substitution
, and the integrand contains exactly this factor. Put
with in , the principal range.
The sine is applied to , and the derivative of is the rest of the integrand. The integrand is defined for every real , since .Standard formula (x), Section 7.2. Basics: substitutionIntegrate and return to
Formula (iii), then is put back. This is the book's final form.A simpler form (optional)
If then and (because ). From :
so the answer can also be written .
The positive square root is correct only because the principal range keeps positive. This form is identical to the book's, not merely equal up to a constant.Principal value branch of , Chapter 2. Basics: trigonometric identitiesCheck: differentiate the answer
Valid for every real .
Outer gives ; inner gives .
(i) .
(ii) .
(iii) , .
(iv) , which equals .
These agree with the textbook's solution.
By Section 7.8, . The middle column is a computer's sum of very many thin strips, found without our answers (part (i) with ):
| Part | Interval | Sum of thin strips | |
|---|---|---|---|
| (i) | |||
| (ii) | |||
| (iii) | |||
| (iv) |
In (iii) the interval keeps between and 1, well below , so the integrand is defined throughout.
- Forgetting to convert . Writing loses the factor . Every substitution must replace as well.
- Leaving the answer in or . The question is in ; substitute back.
- Choosing or . Choose the inner expression whose derivative is also present (, , ), not the most complicated piece.
- Wrong sign in . It is , never negative, because of the principal range.
- Look for "a function of , times ". Then works.
- A missing constant factor is no problem: divide by it (part (i)). A missing factor involving is a problem: then try a different substitution.
- Substitutions can be chained (part (iii)), or combined into one.
Every fact used, and where it comes from
| Fact | Used in | Source |
|---|---|---|
| All parts | Section 7.3.1 (Lesson: integration by substitution) | |
| Parts (i), (ii), (iv) | Standard formula (iii), Section 7.2 | |
| Part (iii), Step 3 | Standard formula (i), Section 7.2 | |
| ; | Part (iii), Step 3; Part (iv), Step 1 | Standard formulae (iv), (x), Section 7.2 |
| Part (iii), Step 2 | Power rule, Section 7.2 formula (i) | |
| Chain rule | All checks | Theorem 4, Section 5.3.1 (Basics: the chain rule backwards) |
| ; | Part (iv), Step 3 | Chapter 2, Section 2.2; Class XI trigonometry |
Tell us if something on this page is wrong or unclear, or if you would like a fuller explanation, another method or a new feature. Send feedback on this problem →