Integral Substitution Methods Explained
What Integral Substitution Actually Is
Substitution is the technique of reversing the chain rule. That's it. The chain rule tells you how to differentiate compositions of functions. Substitution lets you integrate them.
Most integrals you encounter in calculus courses follow recognizable patterns. Substitution is how you spot those patterns and exploit them. Without it, you're stuck trying to guess antiderivatives that you'll never find.
U-Substitution: The Workhorse
U-substitution is the first method you learn and the one you'll use most often. You replace a messy inner expression with a single variable.
The Core Idea
You look for a function and its derivative hiding inside the integral. If you see something like f(g(x)) · g'(x), you can substitute u = g(x) and the integral simplifies dramatically.
When to Use It
- The integrand contains a function and its derivative
- You spot a composite function structure
- A nested expression appears more than once (you can substitute one, solve, then back-substitute)
- The integral looks like it came from a product rule problem run backwards
Classic Examples
Example 1: ∫ 2x·cos(x²) dx
Notice x² inside cosine, and 2x sitting right there. That's the derivative of x². Set u = x², du = 2x dx. The integral becomes ∫ cos(u) du = sin(u) + C = sin(x²) + C.
Example 2: ∫ x·√(x+1) dx
Here u = x+1 works. Then x = u-1 and dx = du. You get ∫ (u-1)·√u du = ∫ (u^(3/2) - u^(1/2)) du. Integrate and back-substitute.
Trigonometric Substitution
Use this when u-substitution fails and you see square roots of quadratic expressions. It converts algebraic nightmare into trig integrals you can actually solve.
The Three Cases
- √(a² - x²) → substitute x = a·sin(θ)
- √(a² + x²) → substitute x = a·tan(θ)
- √(x² - a²) → substitute x = a·sec(θ)
After substituting, the radical disappears because trig identities handle it. You integrate in terms of θ, then convert back to x using right triangles.
When This Actually Works
Trigonometric substitution is overkill for simple rational functions. Save it for integrals with quadratic expressions under radicals, arc length problems, and situations where the geometry of a triangle actually helps you visualize the back-substitution.
Other Substitution Tricks Worth Knowing
Integration by Parts as Substitution
Sometimes treating a product uv' as a substitution works. Let u equal whatever looks hardest to integrate, and let dv equal the rest. This is backwards substitution—you're guessing which part should be the derivative.
Reverse Substitution
When an integral has bounds or expressions that suggest working in a different variable first, substitute to simplify the structure, solve, then reverse. For example, rational functions of sine and cosine often respond well to t = tan(x/2), converting everything to rational functions of t.
Method Comparison
| Method | Best For | Key Signal |
|---|---|---|
| U-substitution | Chain rule patterns, products with derivatives | Function + derivative present |
| Trig substitution | Square roots of quadratics | √(a² ± x²) forms |
| t = tan(x/2) | Trig rational integrals | Rational functions of sin, cos |
| Integration by parts | Products of unlike functions | Log, arctrig, or inverse trig |
Getting Started: A Practical Process
Follow these steps in order:
- Look for obvious derivatives. Check if any part of the integrand is the derivative of another part. U-substitution handles most of these instantly.
- Check for radicals with quadratics. If you see √(something² ± something), trigonometric substitution is your move.
- Try the hardest piece. If substitution doesn't work, integration by parts often succeeds by making the difficult part a derivative.
- Substitute and simplify. Replace variables, simplify the integrand, integrate in the new variable.
- Back-substitute. Replace your temporary variable with the original expression. Check by differentiating your answer.
Common Mistakes That Waste Time
- Choosing u based on what looks complicated rather than what has its derivative present
- Forgetting to convert dx when you substitute du = f'(x)dx
- Back-substituting incorrectly, especially with trig substitution
- Using trig substitution on expressions that don't need it
If your substitution makes the integral more complicated, try a different part to substitute. The right choice almost always simplifies the integral noticeably.
When Substitution Won't Save You
Some integrals don't yield to substitution at all. Elliptic integrals, for instance, have no elementary antiderivatives. Partial fractions work better for proper rational functions. Series expansion is the practical approach when closed forms don't exist.
Substitution is powerful, but it's one tool in a toolkit. Know when to reach for something else.