Intuition What this page is for
The parent note taught you the recipe . Here we make sure you never meet a u-substitution problem that surprises you. We build a scenario matrix — a checklist of every "flavour" a problem can have — then solve one example per flavour, so every cell is filled. If you can do all of these, no exam sub can ambush you.
Reminder of the one tool we use (from the parent):
Here g ( x ) is the inner function we nickname u , and g ′ ( x ) is its derivative (its "slope"). See Chain Rule for why the derivative of the inside must appear.
Two antiderivative rules we lean on repeatedly (from Antiderivatives & Indefinite Integrals ):
Every u-sub problem falls into one of these cells. The last column names the example that covers it — one example per cell, no overlaps .
Cell
What makes it tricky
Covered by
A. Perfect match
derivative of inside sits right there
Ex 1
B1. Positive constant mismatch
you have x d x but need c x d x (c > 0 )
Ex 2
B2. Negative constant mismatch
g ′ carries a minus sign — you must drag a "− " through
Ex 3
C. Leftover variable
a stray x won't cancel — must solve x in terms of u
Ex 4
D. Definite, limits cross (decreasing inside)
inside function decreases → u ( b ) < u ( a ) , reversed limits
Ex 5
E. Degenerate limits (u ( a ) = u ( b ) )
new limits are equal → answer is 0 with no work
Ex 6
F1. Trig inside
inner function is a trig function whose derivative appears
Ex 7
F2. Logarithm g ′ / g form
numerator is exactly the derivative of the denominator
Ex 8
G. Real-world word problem
you must build the integral first
Ex 9
H. Exam twist (rewrite before you can see u )
no obvious inner function until you split/manipulate
Ex 10
We now walk them in order.
Evaluate ∫ ( 3 x 2 ) ( x 3 − 4 ) 7 d x .
Forecast: Look at x 3 − 4 . What is its derivative? If the answer already appears in front, this is the easiest possible case — guess the final power before reading on.
Choose u = x 3 − 4 .
Why this step? It is the inner function raised to a power; its derivative is 3 x 2 , which is sitting right there.
Differentiate: d u = 3 x 2 d x .
Why? This exact chunk 3 x 2 d x is in the integrand, so it becomes d u with nothing left over.
Replace: ∫ u 7 d u .
Why? Every x -piece (including d x ) is now gone — pure u .
Integrate: 8 u 8 + C .
Why? Power rule ∫ u n d u = n + 1 u n + 1 with n = 7 .
Back-substitute: 8 ( x 3 − 4 ) 8 + C .
Verify: Differentiate the answer with the Chain Rule : d x d 8 ( x 3 − 4 ) 8 = 8 8 ( x 3 − 4 ) 7 ⋅ 3 x 2 = 3 x 2 ( x 3 − 4 ) 7 . ✓ We recovered the integrand.
Evaluate ∫ cos ( 5 x ) d x .
Forecast: The inside is 5 x ; its derivative is 5 , a constant. We don't have a 5 in front. Since 5 is a constant, we are allowed to conjure it. Guess whether the answer has a 5 or a 5 1 .
Choose u = 5 x .
Why? It is the inner function of cos .
Differentiate: d u = 5 d x ⇒ d x = 5 1 d u .
Why? We need to replace d x , but only d x (not 5 d x ) appears, so solve for it.
Replace: ∫ cos ( u ) ⋅ 5 1 d u = 5 1 ∫ cos u d u .
Why? 5 1 is a constant , so it may leave the integral (linearity — see Antiderivatives & Indefinite Integrals ). A variable never could.
Integrate: 5 1 sin u + C .
Why? ∫ cos u d u = sin u (basic antiderivative — the reverse of d u d sin u = cos u ).
Back-substitute: 5 1 sin ( 5 x ) + C .
Verify: d x d 5 1 sin ( 5 x ) = 5 1 cos ( 5 x ) ⋅ 5 = cos ( 5 x ) . ✓
Evaluate ∫ x e − x 2 d x .
Forecast: Inside the exponential is − x 2 ; its derivative is − 2 x — a negative constant times x . We only have + x d x . Watch how a "− " sign must be dragged through. Guess whether the answer starts with + 2 1 or − 2 1 .
Choose u = − x 2 .
Why? It is the inner function of e ( ⋅ ) .
Differentiate: d u = − 2 x d x ⇒ x d x = − 2 1 d u .
Why? Solve for the chunk we actually have, x d x . Dividing by − 2 keeps the minus sign — this is the whole point of Cell B2.
Replace: ∫ e u ⋅ ( − 2 1 ) d u = − 2 1 ∫ e u d u .
Why? − 2 1 is a constant, so it leaves the integral, carrying its sign.
Integrate: − 2 1 e u + C .
Why? ∫ e u d u = e u (the exponential is its own antiderivative).
Back-substitute: − 2 1 e − x 2 + C .
Verify: d x d ( − 2 1 e − x 2 ) = − 2 1 e − x 2 ⋅ ( − 2 x ) = x e − x 2 . ✓ The two minus signs multiply to a plus — exactly recovering the integrand.
Evaluate ∫ x x + 1 d x .
Forecast: Inside the root is x + 1 ; its derivative is 1 . But there's an extra lone x multiplying — and x is not a constant, so we cannot pull it out. What do we do with it? Predict: we'll rewrite x using u .
Choose u = x + 1 .
Why? It's the inside of the root; its derivative 1 makes d u = d x clean.
Differentiate: d u = d x .
Handle the stray x : from u = x + 1 we get x = u − 1 .
Why this step? The leftover x can't be a constant lifted out (that's the classic mistake). Instead we express every x in terms of u , including this one.
Replace: ∫ ( u − 1 ) u d u = ∫ ( u 3/2 − u 1/2 ) d u .
Why? Distribute u = u 1/2 so we integrate powers of u .
Integrate: 5/2 u 5/2 − 3/2 u 3/2 = 5 2 u 5/2 − 3 2 u 3/2 .
Why? Power rule applied term-by-term with n = 2 3 and n = 2 1 .
Back-substitute: 5 2 ( x + 1 ) 5/2 − 3 2 ( x + 1 ) 3/2 + C .
Verify: Differentiate: 5 2 ⋅ 2 5 ( x + 1 ) 3/2 − 3 2 ⋅ 2 3 ( x + 1 ) 1/2 = ( x + 1 ) 3/2 − ( x + 1 ) 1/2 = ( x + 1 ) 1/2 [ ( x + 1 ) − 1 ] = x x + 1 . ✓
Evaluate ∫ 0 1 ( 4 − x 2 ) 2 x d x .
Forecast: Inside is u = 4 − x 2 , which decreases as x grows. So as x runs 0 → 1 , u runs 4 → 3 — the new upper limit (3 ) is below the new lower limit (4 ). The limits literally cross. Watch how we handle a reversed range. Guess a small positive answer.
The figure below plots the inner function u = 4 − x 2 (pale yellow) across x ∈ [ 0 , 1 ] . The blue dot marks the left endpoint x = 0 , which sits high at u = 4 ; the pink dot marks the right endpoint x = 1 , which has slid down to u = 3 . The white arrow along the curve shows the sweep downhill as x increases — that is exactly why the new u -limits come out reversed (running 4 → 3 ). Why include it? To make "the limits cross" a thing you can see : the endpoints swap their up/down order under the substitution.
Choose u = 4 − x 2 , d u = − 2 x d x ⇒ x d x = − 2 1 d u .
Why? Inside the squared bracket; its derivative − 2 x is proportional to the x in front, with a minus (Cell B2 flavour reused inside a definite integral).
Translate limits (the variable is now u ; the old numbers no longer describe the range — see Definite Integral & Fundamental Theorem of Calculus ) :
x = 0 ⇒ u = 4 − 0 = 4
x = 1 ⇒ u = 4 − 1 = 3
So u goes from 4 down to 3 — the picture shows this reversed sweep.
Replace (keep the limits in the order they came): ∫ 4 3 u 2 1 ⋅ ( − 2 1 ) d u = − 2 1 ∫ 4 3 u − 2 d u .
Why? Do not silently flip the limits; write 4 → 3 exactly as the substitution produced them.
Flip using the swap rule: − 2 1 ∫ 4 3 = + 2 1 ∫ 3 4 .
Why? ∫ a b = − ∫ b a : swapping limits flips the sign, which cancels the leading minus. This is the clean way to deal with crossed limits.
Integrate: 2 1 ⋅ − 1 u − 1 = − 2 u 1 , evaluated from 3 to 4 .
Why? Power rule with n = − 2 : ∫ u − 2 d u = − 1 u − 1 .
Evaluate: − 2 ⋅ 4 1 − ( − 2 ⋅ 3 1 ) = − 8 1 + 6 1 = 24 − 3 + 4 = 24 1 .
Verify: 24 1 ≈ 0.0417 — small and positive as forecast; numeric integration of the original from 0 to 1 gives 0.041 6 . ✓
Evaluate ∫ − 1 1 2 x ( x 2 + 3 ) 3 d x .
Forecast: Inner function u = x 2 + 3 . At x = − 1 , u = 4 ; at x = 1 , u = 4 again. The new limits are equal . Predict the answer before any algebra.
Choose u = x 2 + 3 , so d u = 2 x d x .
Why? 2 x is exactly the derivative of the inside — a perfect match on the u -piece.
Translate limits:
x = − 1 ⇒ u = ( − 1 ) 2 + 3 = 4
x = 1 ⇒ u = ( 1 ) 2 + 3 = 4
Replace: ∫ 4 4 u 3 d u .
Why? Top and bottom limits are identical.
Integrate & evaluate: an integral over a zero-width interval is 0 .
Why? ∫ c c ( anything ) = 0 — no width, no area.
Verify (two ways): (a) ∫ 4 4 = 0 by definition. (b) The integrand 2 x ( x 2 + 3 ) 3 is an odd function (flip x → − x and it changes sign), integrated over the symmetric interval [ − 1 , 1 ] , so the positive and negative areas cancel exactly. ✓
Evaluate ∫ sin 5 x cos x d x .
Forecast: Rewrite sin 5 x = ( sin x ) 5 . What is the derivative of sin x ? It's cos x — and there it is, multiplying. Perfect-match in disguise. Guess the power in the answer.
Choose u = sin x .
Why? It's the base being raised to a power, and its derivative cos x is present.
Differentiate: d u = cos x d x .
Why? The trailing cos x d x becomes exactly d u .
Replace: ∫ u 5 d u .
Why? Substituting u = sin x turns ( sin x ) 5 into u 5 , and the trailing cos x d x becomes d u — so every x -piece is gone and we have a pure u -integral.
Integrate: 6 u 6 + C .
Why? Power rule with n = 5 .
Back-substitute: 6 sin 6 x + C .
Verify: d x d 6 s i n 6 x = 6 6 s i n 5 x cos x = sin 5 x cos x . ✓
Evaluate ∫ x 2 + x + 7 2 x + 1 d x .
Forecast: Look at the bottom x 2 + x + 7 . Its derivative is 2 x + 1 — exactly the top ! Whenever the numerator is the derivative of the denominator, the answer is a natural log. Predict ln ∣ denominator ∣ .
Choose u = x 2 + x + 7 .
Why? Its derivative sits in the numerator — the special "∫ g g ′ = ln ∣ g ∣ " signature.
Differentiate: d u = ( 2 x + 1 ) d x .
Why? The whole numerator times d x becomes d u .
Replace: ∫ u 1 d u .
Why? The denominator x 2 + x + 7 becomes u , and the entire numerator ( 2 x + 1 ) d x becomes d u — leaving u d u , a pure u -integral with no x left.
Integrate: ln ∣ u ∣ + C .
Why? Log rule — this is the n = − 1 case the power rule can't touch, so ∫ u 1 d u = ln ∣ u ∣ .
Back-substitute: ln ∣ x 2 + x + 7∣ + C .
(The absolute value is optional here since x 2 + x + 7 > 0 always, but keep it as a habit.)
Verify: d x d ln ( x 2 + x + 7 ) = x 2 + x + 7 2 x + 1 . ✓
Water flows into a tank at a rate r ( t ) = ( t 2 + 1 ) 2 6 t litres per minute. How much water enters during the first 2 minutes?
Forecast: "Total amount from a rate" means integrate the rate over the time interval — that's the Definite Integral & Fundamental Theorem of Calculus . We need ∫ 0 2 r ( t ) d t . Inner function t 2 + 1 ; derivative 2 t ; we have 6 t — a constant fix. Guess an answer a bit under 3 litres.
Set up: V = ∫ 0 2 ( t 2 + 1 ) 2 6 t d t .
Why? Accumulated volume = integral of inflow rate over the time window (units: min L × min = L ).
Choose u = t 2 + 1 , d u = 2 t d t ⇒ 6 t d t = 3 d u .
Why? Inside of the squared bracket; fix the constant (6 t = 3 ⋅ 2 t ).
Translate limits:
Replace: ∫ 1 5 u 2 3 d u = 3 ∫ 1 5 u − 2 d u .
Why? The denominator ( t 2 + 1 ) 2 becomes u 2 , and the whole numerator 6 t d t becomes 3 d u from step 2. The 3 is a constant , so it may leave the integral (a variable never could) — leaving a clean power of u .
Integrate: 3 ⋅ − 1 u − 1 = − u 3 , from 1 to 5 .
Why? Power rule with n = − 2 .
Evaluate: − 5 3 − ( − 1 3 ) = − 5 3 + 3 = 5 − 3 + 15 = 5 12 = 2.4 litres .
Why? Fundamental Theorem: plug the upper u = 5 then subtract the lower u = 1 into − u 3 . The two minus signs on the lower term make it + 3 , and 3 − 5 3 over a common denominator of 5 gives 5 12 .
Verify: 2.4 L is positive, under 3 as forecast, and units check (L/min ⋅ min ). Numeric integration of r ( t ) on [ 0 , 2 ] gives 2.4 . ✓
Evaluate ∫ x ln x 1 d x (for x > 1 ).
Forecast: There's no obvious "inside function times its derivative"... until you notice that x 1 is the derivative of ln x . So the awkward-looking ln x is the real inner function. Predict a log-of-a-log.
Rewrite mentally: x ln x 1 = ln x 1 ⋅ x 1 .
Why this step? Splitting reveals the pair "function ln x and its derivative x 1 " — the chain-rule fingerprint that was hidden.
Choose u = ln x , d u = x 1 d x .
Why? ln x is the inner function; its derivative x 1 d x is present.
Replace: ∫ u 1 d u .
Why? The factor ln x 1 becomes u 1 , and the trailing x 1 d x becomes d u — so all x -pieces vanish, leaving u d u .
Integrate: ln ∣ u ∣ + C .
Why? Log rule — the n = − 1 case again.
Back-substitute: ln ∣ ln x ∣ + C .
Verify: d x d ln ( ln x ) = l n x 1 ⋅ x 1 = x l n x 1 . ✓
Recall Which example killed which cell? (one-to-one)
A perfect match ::: Ex 1
B1 positive constant mismatch ::: Ex 2
B2 negative constant mismatch (carry the minus) ::: Ex 3
C stray variable → solve x = u − 1 ::: Ex 4
D definite, limits cross (u ( b ) < u ( a ) ) → flip with ∫ a b = − ∫ b a ::: Ex 5
E degenerate equal limits give 0 ::: Ex 6
F1 trig inside ::: Ex 7
F2 logarithm g ′ / g ::: Ex 8
G real-world (build the integral) ::: Ex 9
H exam twist (rewrite first) ::: Ex 10
Common mistake The two traps that reappear across cells
Trap 1 (Cell C): pulling a variable out front. Only constants leave the integral; a stray x must be rewritten as x = u − 1 or similar.
Trap 2 (Cell D/E): forgetting to translate limits. After you switch to u , the old x -numbers describe the wrong range — always recompute u ( a ) and u ( b ) , keep them in the order they come out (even if they cross), or don't change limits at all and back-substitute. Never both.
"Match, +Const, −Const, Move-the-x, Cross-limits, Zero-limits, Trig-it, Log-it, Build-it, Rewrite-it." — the ten flavours in order.