Miscellaneous · Q17:
Integrate the function:
Let .
Put . Then .
(for , ).
Want the reason behind every step, the basics and the checks?
The integrand is a power of the inner expression multiplied by (almost) its derivative. By the chain rule the derivative of is , so the substitution turns into and leaves just .
- Given
- A differentiable function with continuous derivative , constants and , and a real exponent . (For the integrand is a constant; see Step 6.)
- To find
- .
- Plan
- (1) Name the substitution and find by the chain rule. (2) Integrate (). (3) Return to . (4) Differentiate back. (5) Treat . (6) Conditions and the key. (7) A numerical example.
Basics you need
: derivative of the outer function at the inner one, times the derivative of the inner function . Basics: substitution (the chain rule backwards)
for ; (standard formulae (i) and (xii), Section 7.2). Basics: powers of x
Solution
Choose the substitution
The integrand is "something to a power" times "the derivative of that something" (apart from the constant ). This is the pattern substitution is made for. Dividing by is allowed because .Chain rule (Chapter 5); Section 7.3.1. Basics: substitutionIntegrate in (for )
The constant comes outside (Property (IV), Section 7.2.1), and the power formula needs .Standard formula (i), Section 7.2. Basics: powers of xReturn to
Replace by .Check: differentiate the answer
Chain rule twice: outer , middle , inner :
The factors and produced by differentiating cancel the denominator exactly, which is why the answer carries underneath.The excluded case
Then the integrand is and the same substitution gives
valid wherever .
The formula of Step 3 would divide by , so it cannot be used; is a logarithm instead (standard formula (xii)). The question does not mention this case, but a complete answer states it.Conditions, and the textbook's answer
- (we divided by ). If , the integrand is the constant and the integral is that constant times , plus .
- for the formula of Step 3 (Step 5 covers ).
- The power must make sense: for a positive integer , any ; for a negative integer , only where ; for a fractional , usually only where .
The key gives , exactly our answer (the key leaves these conditions unstated).
An answer is valid on any interval where both the integrand and the answer are defined and the derivative check of Step 4 holds; these bullet points describe exactly those intervals.A numerical example
Take , , , . The question becomes , and the formula gives
From 0 to 1 this predicts , and numerical integration of from 0 to 1 also gives .
A general formula is easiest to trust after one concrete case. The negative value is fine: the integrand is negative on most of , because runs from 1 to 3 radians and is negative beyond .
For , , on any interval where the power is defined:
(and when ).
(Textbook answer: .)
Second example: , , , . Then and the integrand is ; the formula gives .
From 0 to 1: . Expanding, , whose integral from 0 to 1 is .
- Forgetting the . , so the answer has in the denominator.
- Using the power formula for . It divides by zero; the answer is a logarithm.
- Writing as . These differ by the factor : means the derivative of evaluated at .
- (); here and .
- A linear inside always brings a factor into the integral.
Every fact used, and where it comes from
| Fact | Used in | Source |
|---|---|---|
| Chain rule | Steps 1, 4 | Chapter 5 (Basics) |
| Integration by substitution | Steps 1, 2, 5 | Section 7.3.1 (Lesson: substitution) |
| , | Step 2 | Standard formula (i), Section 7.2 |
| Constants come outside | Step 2 | Property (IV), Section 7.2.1 |
| Step 5 | Standard formula (xii), Section 7.2 |
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 →