Differentiation · Beyond the First Derivative

Higher-Order Derivatives

Differentiate a derivative to get another derivative. Learn all the notation systems, the physical chain from position to jerk, how concavity connects to f″, and the beautiful patterns in the nth derivative of xn, sin x, cos x and ex. 10 worked examples · 10-question quiz.

Share this page

§ 01What Is a Higher-Order Derivative?

The derivative of a function is itself a function, and a function can be differentiated again. Repeating this process produces second, third, and in general nth derivatives.

When you differentiate f(x), you get f′(x), which tells you the rate of change of f. When you differentiate f′(x), you get f″(x), which tells you the rate of change of the rate of change, in other words, how quickly the slope itself is changing. This is the second derivative, and it governs crucial geometric properties like concavity and inflection points.

You can keep going: f‴(x) is the third derivative, and so on. In mechanics, if f(t) is position, each successive derivative has a vivid physical name: velocity, acceleration, jerk, snap, crackle, pop (yes, those are the real names).

The Differentiation Chain
f(x) → f′(x) → f″(x) → f‴(x) → f(4)(x) → ···

Each arrow means "differentiate with respect to x." Any step can be taken as many times as needed.

Why Do Higher-Order Derivatives Matter? The second derivative tells us whether a curve is concave up or concave down — essential for classifying turning points and sketching graphs. In physics it is acceleration. In economics it is the rate at which marginal cost changes. In differential equations it appears on equal footing with f itself and f′, making higher-order derivatives fundamental rather than optional.

§ 02All the Notation Systems

Higher-order derivatives come with three competing notation families. You need to recognise all of them, because different textbooks and fields favour different styles.

Lagrange / Prime Notation
f′, f″, f‴, f(n)
1st: f′(x)
2nd: f″(x)
3rd: f‴(x)
4th+: f(4)(x), f(5)(x), …
nth: f(n)(x)
Leibniz Notation
d²y/dx², d³y/dx³
1st: dy/dx
2nd: d²y/dx²
3rd: d³y/dx³
nth: dny/dxn
Newton / Dot Notation
ẏ, ÿ (physics)
Used when the variable is time t
1st derivative: ẏ (one dot)
2nd derivative: ÿ (two dots)
Rarely used beyond 2nd order
Operator Notation
D, D², Dⁿ
D = d/dx
D²f = d²f/dx²
Dnf = dnf/dxn
Useful in differential equations
Reading d²y/dx² Correctly The notation d²y/dx² is not a fraction. The 2 in the numerator sits with the operator d², and the 2 in the denominator squares the differential dx. It means "apply d/dx twice to y." Do not cancel the 2s or treat this as (dy/dx)².

Quick Translation Table

OrderLagrangeLeibnizNewton (time)Operator
1stf′(x)dy/dxDf
2ndf″(x)d²y/dx²ÿD²f
3rdf‴(x)d³y/dx³D³f
nthf(n)(x)dny/dxnrarely usedDnf

§ 03Physical Meaning — Position to Jerk

The most vivid illustration of higher-order derivatives comes from classical mechanics. If s(t) is the position of a particle at time t, each successive derivative has a concrete physical name and meaning.

0th derivative Position s(t)
1st derivative Velocity v = ds/dt
2nd derivative Acceleration a = d²s/dt²
3rd derivative Jerk j = d³s/dt³
4th derivative Snap d⁴s/dt⁴

Velocity is how fast you are moving. Acceleration is how fast your velocity is changing, what you feel pressed into a seat during take-off. Jerk (also called jolt) is how fast acceleration is changing: the lurch you feel when a lift starts or stops, or the snap of a whiplash. Engineers who design smooth roller coasters and elevator systems work hard to limit jerk.

Newton's Second Law is a Second-Order Relationship F = ma means F = m · d²s/dt². Newton's laws inherently involve the second derivative of position. This is why second-order differential equations are so prevalent in physics — they encode the relationship between force and the second derivative of position.

§ 04Concavity and the Second Derivative

The second derivative gives direct information about the shape (concavity) of a curve, and is the key tool for classifying stationary points.

Concavity Rules
f″(x) > 0  ⟹  concave up (∪ shape) at x f″(x) < 0  ⟹  concave down (∩ shape) at x f″(x) = 0  ⟹  possible inflection point

A curve is concave up where it curves like a bowl facing upward: the tangent lines lie below the curve. It is concave down where it curves like a hill: the tangent lines lie above the curve. An inflection point is where concavity changes sign.

Second Derivative Test for Stationary Points
f′(c) = 0  and  f″(c) > 0  ⟹  local minimum at x = c f′(c) = 0  and  f″(c) < 0  ⟹  local maximum at x = c f′(c) = 0  and  f″(c) = 0  ⟹  test is inconclusive

When the test is inconclusive, fall back to the first derivative sign chart.

Why f″ > 0 Means a Minimum — an Intuitive Explanation If f′(c) = 0 (flat tangent) and f″(c) > 0 (the slope is increasing through zero), then just to the left of c the slope was negative (falling) and just to the right it is positive (rising). The function falls into c from the left and rises away to the right — a valley, i.e. a local minimum. The reasoning reverses for a maximum.

§ 05The nth Derivative — Patterns for Key Functions

For certain fundamental functions, differentiating repeatedly reveals a beautiful cyclic or algebraic pattern, allowing us to write down the nth derivative directly without performing n steps.

Polynomials — xn Terminates

Each differentiation reduces the power by one and multiplies by the old power. After n steps the result is a constant; one step further gives zero and all subsequent derivatives are zero.

nth Derivative of xⁿ
dk/dxk (xn) = n(n−1)(n−2)···(n−k+1) · xn−k dn/dxn (xn) = n!    (a constant) dk/dxk (xn) = 0   for k > n

Example: f(x) = x⁵. Then f‴(x) = 60x², f(4)(x) = 120x, f(5)(x) = 120, f(6)(x) = 0.

The Exponential — ex Never Changes

nth Derivative of eˣ
dn/dxn (ex) = ex   for all n ≥ 1

eˣ is its own derivative at every order. No matter how many times you differentiate, you always get eˣ back.

Sine and Cosine — A 4-Cycle

Differentiating sin x and cos x produces a repeating cycle of length 4. After four differentiations you return to the original function.

n mod 4dn/dxn (sin x)dn/dxn (cos x)
0 (n=0,4,8,…)sin xcos x
1 (n=1,5,9,…)cos x−sin x
2 (n=2,6,10,…)−sin x−cos x
3 (n=3,7,11,…)−cos xsin x
Compact Formula for nth Derivative of sin x
dn/dxn (sin x) = sin(x + nπ/2)

Similarly: dn/dxn(cos x) = cos(x + nπ/2). The derivative just phase-shifts the trig function by π/2 each time.

General Power Exponential — eax

nth Derivative of eax
dn/dxn (eax) = an eax

Each differentiation multiplies by the constant a. After n steps the factor is aⁿ.

Summary Reference Table

Function f(x)nth Derivative f(n)(x)Key property
xmm!/(m−n)! · xm−n  (n ≤ m)Terminates at n=m
exexSelf-replicating
eaxan eaxGrows by factor a each step
sin xsin(x + nπ/2)4-cycle
cos xcos(x + nπ/2)4-cycle
ln x(−1)n−1(n−1)!/xnFor n ≥ 1, x > 0
x−1 = 1/x(−1)n · n!/xn+1Alternating sign

§ 0610 Fully Worked Examples

Examples cover polynomial, trigonometric, exponential, and combined functions, progressing from straightforward repeated differentiation through concavity analysis to general nth derivatives.

Example 01 Find f″(x) for f(x) = x⁵ − 3x³ + 7x

Differentiate twice, term by term.

  • f′(x) = 5x⁴ − 9x² + 7 (power rule on each term)
  • f″(x) = 20x³ − 18x (differentiate f′)
f″(x) = 20x³ − 18x
Example 02 Find d³y/dx³ for y = sin(2x)

Apply the chain rule on each differentiation. The factor of 2 accumulates.

  • dy/dx = 2 cos(2x)
  • d²y/dx² = 2 · (−2 sin(2x)) = −4 sin(2x)
  • d³y/dx³ = −4 · 2 cos(2x) = −8 cos(2x)
d³y/dx³ = −8 cos(2x)

Pattern: each differentiation multiplies by 2. After n steps: 2ⁿ sin(2x + nπ/2).

Example 03 Find the second derivative of f(x) = x · eˣ

First derivative requires the product rule; then differentiate again.

  • f′(x) = eˣ + x eˣ = (1 + x)eˣ (product rule)
  • f″(x) = d/dx[(1+x)eˣ]. Apply the product rule again: (1+x)′·eˣ + (1+x)·eˣ = eˣ + (1+x)eˣ.
  • Simplify: eˣ[1 + 1 + x] = (2 + x)eˣ.
f″(x) = (x + 2)eˣ
Example 04 Classify the stationary points of f(x) = x³ − 3x using f″

Find where f′ = 0, then use the second derivative test.

  • f′(x) = 3x² − 3 = 3(x²−1) = 0 → x = ±1.
  • f″(x) = 6x.
  • At x = 1: f″(1) = 6 > 0 → local minimum. f(1) = 1 − 3 = −2.
  • At x = −1: f″(−1) = −6 < 0 → local maximum. f(−1) = −1 + 3 = 2.
Local max at (−1, 2); local min at (1, −2)
Example 05 Find the inflection points of f(x) = x⁴ − 6x²

Inflection points occur where f″ = 0 and changes sign.

  • f′(x) = 4x³ − 12x.
  • f″(x) = 12x² − 12 = 12(x²−1) = 0 → x = ±1.
  • For x < −1: f″ = 12(4−1) > 0 (concave up). For −1 < x < 1: f″ = 12(0−1) < 0 (concave down). For x > 1: concave up. Sign changes at both x = ±1. ✓
  • f(1) = 1 − 6 = −5. f(−1) = 1 − 6 = −5.
Inflection points at (1, −5) and (−1, −5)
Example 06 Find f(4)(x) for f(x) = cos x

Use the 4-cycle pattern: cos x → −sin x → −cos x → sin x → cos x.

  • f′(x) = −sin x.
  • f″(x) = −cos x.
  • f‴(x) = sin x.
  • f(4)(x) = cos x. The cycle has completed: we are back to the start.
f(4)(x) = cos x

This confirms: dn/dxn(cos x) = cos(x + nπ/2). At n=4: cos(x + 2π) = cos x. ✓

Example 07 Find the nth derivative of f(x) = e3x

The chain rule contributes a factor of 3 each time we differentiate.

  • f′(x) = 3e3x.
  • f″(x) = 3² e3x = 9e3x.
  • f‴(x) = 3³ e3x = 27e3x.
  • Pattern: each step multiplies by 3. After n steps the coefficient is 3ⁿ.
f(n)(x) = 3ⁿ e3x
Example 08 Find f(50)(x) for f(x) = sin x

Use the 4-cycle to avoid doing 50 differentiations.

  • 50 ÷ 4 = 12 remainder 2. So n mod 4 = 2.
  • From the table: when n mod 4 = 2, dn/dxn(sin x) = −sin x.
  • Alternatively: sin(x + 50π/2) = sin(x + 25π) = sin(x + π) = −sin x. ✓
f(50)(x) = −sin x
Example 09 Find the second derivative of f(x) = ln x
  • f′(x) = 1/x = x−1.
  • f″(x) = d/dx(x−1) = −x−2 = −1/x².
  • Since f″(x) = −1/x² < 0 for all x > 0, ln x is concave down everywhere on its domain.
f″(x) = −1/x²   (concave down for all x > 0)
Example 10 A particle has position s(t) = t⁴ − 4t³. Find velocity, acceleration, and jerk at t = 2.
  • Velocity: v(t) = s′(t) = 4t³ − 12t². At t=2: v(2) = 32 − 48 = −16.
  • Acceleration: a(t) = s″(t) = 12t² − 24t. At t=2: a(2) = 48 − 48 = 0.
  • Jerk: j(t) = s‴(t) = 24t − 24. At t=2: j(2) = 48 − 24 = 24.
v(2) = −16 units/s,  a(2) = 0,  jerk(2) = 24 units/s³

The zero acceleration at t = 2 means the particle is at a velocity turning point, its speed has momentarily stopped changing, even though jerk ≠ 0 means that will change immediately after.

§ 07Common Mistakes

Mistake 1 — Forgetting the chain rule on repeated differentiation When differentiating f(x) = sin(3x) twice, students sometimes write f″ = −sin(3x) instead of −9 sin(3x). Each differentiation of sin(3x) brings down a factor of 3 via the chain rule. After two steps the coefficient of x in the exponent/argument contributes 3² = 9.
Mistake 2 — Claiming f″(c) = 0 alone means an inflection point f″(c) = 0 is a necessary but not sufficient condition for an inflection point. You must also verify that f″ changes sign at c. Consider f(x) = x⁴: f″(0) = 0 but f″(x) = 12x² ≥ 0 everywhere — no sign change, so x = 0 is not an inflection point (it is actually a minimum).
Mistake 3 — Misreading d²y/dx² as (dy/dx)² These are completely different things. d²y/dx² is the second derivative (differentiate twice). (dy/dx)² is the square of the first derivative. Example: if y = x², then dy/dx = 2x, so (dy/dx)² = 4x², but d²y/dx² = 2.
Mistake 4 — Wrong cycle position for trig nth derivatives When finding f(n)(x) for sin x or cos x, always compute n mod 4 first. A common error is starting the cycle count at 1 instead of 0, or miscounting the remainder. Write out n ÷ 4 explicitly: for n = 17, 17 = 4×4 + 1, so n mod 4 = 1, giving cos x for sin's derivative.

§ 0810-Question Practice Quiz

Test your command of higher-order derivatives. A mix of computation, classification, and pattern recognition.

Score: 0 / 0
Question 1 · Fill in the blank

If f(x) = x⁴, what is f″(x)? Enter the coefficient: f″(x) = __ x²

First: f′(x) = 4x³. Then differentiate again.
  1. f′(x) = 4x³
  2. f″(x) = 12x²
Coefficient = 12
Question 2 · Multiple choice

What does f″(c) < 0 tell you about the curve at x = c?

  1. f″(c) < 0 means the slope is decreasing through c: the curve bends downward (∩), i.e. concave down.
  2. A local minimum requires f′(c)=0 AND f″(c)>0, two conditions together.
Question 3 · Fill in the blank

Find d²y/dx² for y = e5x. It equals __ · e5x.

Each differentiation of e5x multiplies by 5. Two differentiations give 5².
  1. dy/dx = 5e5x
  2. d²y/dx² = 25e5x
Coefficient = 25 = 5²
Question 4 · Multiple choice

What is f(3)(x) for f(x) = cos x?

  1. f′ = −sin x, f″ = −cos x, f‴ = sin x.
  2. Or: cos(x + 3π/2) = cos x cos(3π/2) − sin x sin(3π/2) = 0 + sin x = sin x. ✓
Question 5 · Fill in the blank

For f(x) = x⁶, what is f(6)(x)? (It's a constant: enter the value.)

The nth derivative of xn is n! (n factorial).
  1. dn/dxn(xn) = n! for the nth derivative.
  2. 6! = 6×5×4×3×2×1 = 720.
f(6)(x) = 720
Question 6 · Multiple choice

A particle's position is s(t) = t³ − 6t. What is its acceleration at t = 2?

  1. s′(t) = 3t² − 6 (velocity)
  2. s″(t) = 6t (acceleration)
  3. s″(2) = 12
Question 7 · Fill in the blank

Find f(100)(x) for f(x) = sin x. Enter the number k where f(100)(x) = k·sin x.

Compute 100 mod 4. What row of the cycle table does that correspond to?
  1. 100 ÷ 4 = 25 remainder 0. So n mod 4 = 0.
  2. When n mod 4 = 0, dn/dxn(sin x) = sin x.
  3. So f(100)(x) = sin x, meaning k = 1.
f(100)(x) = sin x  (k = 1)
Question 8 · Multiple choice

The second derivative test is inconclusive when:

  1. The second derivative test requires: f′(c) = 0 (critical point) AND the sign of f″(c) to classify it.
  2. If both are zero, the test gives no information, use the first derivative sign chart instead.
Question 9 · Fill in the blank

Find f‴(x) for f(x) = x⁵. Enter the coefficient of x²: f‴(x) = __ x²

f′ = 5x⁴, f″ = 20x³, then differentiate once more.
  1. f′(x) = 5x⁴
  2. f″(x) = 20x³
  3. f‴(x) = 60x²
Coefficient = 60
Question 10 · Multiple choice

Which of these is the correct Leibniz notation for the second derivative of y with respect to x?

  1. d²y/dx² is the Leibniz notation for "apply d/dx twice to y."
  2. (dy/dx)² is the square of the first derivative, a completely different quantity.

Quiz complete!

Cookie Settings