5.7 · SLPaper 1Paper 2

Second derivative and concavity

In a minute

The , f′′(x)f''(x), is the derivative of the derivative. It tells you how the gradient is changing, and so which way the graph bends. Where f′′(x)>0f''(x) > 0 the graph is , and where f′′(x)<0f''(x) < 0 it is .

You'll be able to

  • Differentiate twice to find the , written f′′(x)f''(x) or d2ydx2\dfrac{\mathrm{d}^{2}y}{\mathrm{d}x^{2}}, and find its value at a point.
  • Use the sign of f′′(x)f''(x) to say where a graph is or .
  • Link the graphs of ff, f′f' and f′′f'', and read features of one from another.
  • Use your GDC to find where a graph is concave-up or concave-down when f′′(x)=0f''(x) = 0 can't be solved by hand.
Before you start3 quick questions on what this builds on
Check yourself
Let f(x)=x3−6x2+5xf(x) = x^{3} - 6x^{2} + 5x. Find f′(x)f'(x).
Check yourself
Let f(x)=e−3xf(x) = \mathrm{e}^{-3x}. Find f′(x)f'(x).
Check yourself
A function has derivative f′(x)=x2−4f'(x) = x^{2} - 4. For which values of xx is ff decreasing?

Revise first: 5.2 Increasing and decreasing functions and 5.6 Differentiation rules and standard functions

Differentiating twice

When you differentiate f(x)f(x), you get f′(x)f'(x), the gradient function. It is a function of xx too, so you can differentiate it again.

The result is the . It measures how fast the gradient changes as xx increases.

Learn this
f′′(x)=ddx(f′(x))d2ydx2=ddx(dydx)f''(x) = \frac{\mathrm{d}}{\mathrm{d}x}\big(f'(x)\big) \qquad \frac{\mathrm{d}^{2}y}{\mathrm{d}x^{2}} = \frac{\mathrm{d}}{\mathrm{d}x}\left(\frac{\mathrm{d}y}{\mathrm{d}x}\right)

Read f′′(x)f''(x) as "f double prime of x". d2ydx2\dfrac{\mathrm{d}^{2}y}{\mathrm{d}x^{2}} is said "d two y by d x squared", but nothing is squared: the 2s mean "differentiated twice".

Questions use both. Use f′′(x)f''(x) when the function has a name, and d2ydx2\dfrac{\mathrm{d}^{2}y}{\mathrm{d}x^{2}} when the curve is given as y=…y = \ldots

You need no new rules. You use the rules from 5.6 twice: the power rule, the chain rule, and the product and quotient rules.

Worked exampleLet f(x)=2x4−3x3+x−7f(x) = 2x^{4} - 3x^{3} + x - 7. Find f′′(x)f''(x).

Try it first, then reveal one step at a time.

  1. Differentiate each term once with the power rule: f′(x)=8x3−9x2+1f'(x) = 8x^{3} - 9x^{2} + 1. The term xx gives 11, and the constant −7-7 gives 00.
Worked exampleA curve has equation y=xe2xy = x\mathrm{e}^{2x}. Show that d2ydx2=4(x+1)e2x\dfrac{\mathrm{d}^{2}y}{\mathrm{d}x^{2}} = 4(x + 1)\mathrm{e}^{2x}.

Try it first, then reveal one step at a time.

  1. yy is a product, so use the product rule. Let u=xu = x and v=e2xv = \mathrm{e}^{2x}. Then dudx=1\dfrac{\mathrm{d}u}{\mathrm{d}x} = 1 and, by the chain rule, dvdx=2e2x\dfrac{\mathrm{d}v}{\mathrm{d}x} = 2\mathrm{e}^{2x}.
Check yourself
Let f(x)=x3+4xf(x) = x^{3} + \dfrac{4}{x}. Find f′′(x)f''(x).
Check yourself
Let f(x)=ln⁡(2x+1)f(x) = \ln(2x + 1), for x>−12x > -\tfrac{1}{2}. Find f′′(1)f''(1). Give your answer to 3 s.f.

Concavity: which way the graph bends

f′′(x)f''(x) is the rate of change of the gradient. If f′′(x)>0f''(x) > 0, the gradient increases as you move right. If f′′(x)<0f''(x) < 0, the gradient decreases.

Picture y=x2y = x^{2}. Its gradient is 2x2x, which goes −2-2, 00, 22, 44 as xx goes −1-1, 00, 11, 22. The gradient keeps increasing, and the curve bends upwards like a cup. Here f′′(x)=2>0f''(x) = 2 > 0.

Now picture y=−x2y = -x^{2}. Its gradient, −2x-2x, keeps decreasing, and the curve bends downwards like an upside-down cup. Here f′′(x)=−2<0f''(x) = -2 < 0.

A curve that bends upwards is . A curve that bends downwards is .

Another way to see it: a concave-up curve lies above its tangents, and a concave-down curve lies below them.

Concavity is separate from increasing and decreasing. The sign of f′f' gives the direction, up or down. The sign of f′′f'' gives the bend. So there are four combinations.

Four curve pieces: increasing and concave-up, increasing and concave-down, decreasing and concave-up, decreasing and concave-down increasing, concave-up f′ > 0, f″ > 0 increasing, concave-down f′ > 0, f″ < 0 decreasing, concave-up f′ < 0, f″ > 0 decreasing, concave-down f′ < 0, f″ < 0
f′f' decides up or down. f′′f'' decides the bend.

Where a graph switches from one concavity to the other, it has a . You will find and test these in 5.8.

To find where a graph is concave-up or concave-down:

1. Find f′′(x)f''(x).

2. Solve f′′(x)=0f''(x) = 0 to find where its sign can change.

3. Find the sign of f′′(x)f''(x) on each side. Answer with inequalities in xx.

Worked exampleLet f(x)=x3−6x2+4f(x) = x^{3} - 6x^{2} + 4. Is the graph of ff concave-up or concave-down at x=1x = 1?

Try it first, then reveal one step at a time.

  1. Differentiate once: f′(x)=3x2−12xf'(x) = 3x^{2} - 12x.
Worked exampleLet f(x)=x4−6x2+2xf(x) = x^{4} - 6x^{2} + 2x. On which interval is the graph of ff concave-down?

Try it first, then reveal one step at a time.

  1. Differentiate twice: f′(x)=4x3−12x+2f'(x) = 4x^{3} - 12x + 2, so f′′(x)=12x2−12f''(x) = 12x^{2} - 12.
Check yourself
Let f(x)=(x−1)exf(x) = (x - 1)\mathrm{e}^{x}. Its derivative is f′(x)=xexf'(x) = x\mathrm{e}^{x}. Where is the graph of ff concave-up?
Check yourself
Let f(x)=x3−3x2−9x+1f(x) = x^{3} - 3x^{2} - 9x + 1. Which describes the graph of ff at x=2x = 2?

The graphs of f, f′ and f″

You can now link three graphs. Each one is the gradient graph of the one before it.

The sign of f′f' tells you where ff goes up or down. The sign of f′′f'' tells you where f′f' goes up or down. That is the same as how ff bends.

Graphs of f of x equals x cubed minus 3x, its derivative 3x squared minus 3, and its second derivative 6x, stacked with matching x-values y = f(x) x max min concave-down concave-up y = f′(x) x min of f′ y = f″(x) x f″(x) < 0 for x < 0 f″(x) > 0 for x > 0 x = −1 x = 0 x = 1
f(x)=x3−3xf(x) = x^{3} - 3x, f′(x)=3x2−3f'(x) = 3x^{2} - 3 and f′′(x)=6xf''(x) = 6x, lined up.

Follow the dashed lines. At x=−1x = -1 and x=1x = 1, ff has a maximum and a minimum. There f′f' crosses the xx-axis, because the gradient of ff is 00.

At x=0x = 0, f′f' has its lowest point, so f′′(0)=0f''(0) = 0. f′′f'' changes from negative to positive there. On ff, the bend switches from to , and ff is falling most steeply.

For a polynomial, each derivative is one degree lower. A cubic ff has a quadratic f′f' and a linear f′′f''. This helps you match graphs.

Explore it. Slide the point along f(x)=x3−3xf(x) = x^{3} - 3x from x=−2x = -2 to x=2x = 2, and watch the value of f′(x)f'(x). It falls until x=0x = 0, then rises. So the gradient decreases while the curve bends down, and increases while it bends up.

The sign of the derivativeDrag the slider
−2−112−4−3−2−11234increasingdecreasing○ stationary: x = ±1

f(x) = x³ − 3x, so f′(x) = 3x² − 3

−2.0

f′(−2.0) = 9.00 (increasing)

Where f′(x) > 0 the tangent slopes up and f is increasing; where f′(x) < 0 it is decreasing. The sign changes at the stationary points, where f′(x) = 0.

On your GDC you can do the same: graph ff, then find the derivative at several points and watch it rise or fall.

Sometimes you get only the graph of y=f′(x)y = f'(x). Its sign shows where ff increases or decreases. Whether it rises or falls shows how ff bends.

Graph of y equals f prime of x, an upward parabola crossing the x-axis at minus 1 and 3 with its lowest point at x equals 1 x y O −1 3 1 (1, −4) y = f′(x)
This is the graph of y=f′(x)y = f'(x), not of ff.
Worked exampleThe diagram shows the graph of y=f′(x)y = f'(x). It crosses the xx-axis at x=−1x = -1 and x=3x = 3, and its lowest point is (1,−4)(1, -4). For which xx does the graph of ff bend upwards (concave-up)?

Try it first, then reveal one step at a time.

  1. Concave-up means f′′(x)>0f''(x) > 0. f′′f'' is the gradient of f′f', so look for where the graph of y=f′(x)y = f'(x) is rising.
Check yourself
Use the same graph of y=f′(x)y = f'(x). For which values of xx is ff decreasing and its graph concave-up?

On Paper 2, f′′(x)=0f''(x) = 0 may have no algebraic solution. Find f′′(x)f''(x) by hand, then graph y=f′′(x)y = f''(x) on your GDC and find where it crosses the xx-axis.

Worked exampleLet f(x)=ex−x3f(x) = \mathrm{e}^{x} - x^{3}, for x∈Rx \in \mathbb{R}. Use your calculator to find where the graph of ff is concave-down. Give the end points to 3 s.f.

Try it first, then reveal one step at a time.

  1. Differentiate twice by hand: f′(x)=ex−3x2f'(x) = \mathrm{e}^{x} - 3x^{2}, so f′′(x)=ex−6xf''(x) = \mathrm{e}^{x} - 6x.

Common mistakes

  • Solving f′(x)>0f'(x) > 0 when the question asks where the graph is concave-up. Concavity comes from the sign of f′′(x)f''(x), so differentiate twice first.
  • Stopping after one derivative. If the question asks for f′′(x)f''(x) or d2ydx2\dfrac{\mathrm{d}^{2}y}{\mathrm{d}x^{2}}, differentiate your first derivative again.
  • Losing a sign with negative powers. 4x=4x−1\dfrac{4}{x} = 4x^{-1} differentiates to −4x−2-4x^{-2}, and then to +8x−3+8x^{-3}.
  • Giving only the value where f′′(x)=0f''(x) = 0. That is where the concavity can change. The answer is a set of xx-values, such as −1<x<1-1 < x < 1.
  • Reading the graph of f′f' as if it were ff. For the concavity of ff, ask whether the graph of f′f' is rising or falling, not whether it is above the axis.

Exam tips

  • When a question asks where a graph is concave-up, write f′′(x)>0f''(x) > 0 as your first line, then solve. Answer with inequalities in xx.
  • "Show that f′′(x)=…f''(x) = \ldots": show the first derivative and every line of the second. Your last line must match the given result exactly.
  • On Paper 2, find f′′(x)f''(x) by hand, then write down the equation you put into the GDC. Give the end points to 3 s.f.
  • Use the notation the question uses. If it writes d2ydx2\dfrac{\mathrm{d}^{2}y}{\mathrm{d}x^{2}}, answer with d2ydx2\dfrac{\mathrm{d}^{2}y}{\mathrm{d}x^{2}}.

Key takeaways

Learn thisf′′(x)=ddx(f′(x))d2ydx2=ddx(dydx)f''(x) = \frac{\mathrm{d}}{\mathrm{d}x}\big(f'(x)\big) \qquad \frac{\mathrm{d}^{2}y}{\mathrm{d}x^{2}} = \frac{\mathrm{d}}{\mathrm{d}x}\left(\frac{\mathrm{d}y}{\mathrm{d}x}\right)
Learn thisf′′(x)>0f''(x) > 0: the gradient is increasing and the graph is concave-up.
Learn thisf′′(x)<0f''(x) < 0: the gradient is decreasing and the graph is concave-down.
Learn thisThe sign of f′f' gives up or down. The sign of f′′f'' gives the bend. The two are separate.
Learn thisWhere f′(x)=0f'(x) = 0, ff is stationary. Where f′′(x)=0f''(x) = 0, f′f' is stationary and ff may change concavity.
Learn thisTo find where ff is concave-up: find f′′(x)f''(x), solve f′′(x)=0f''(x) = 0, check the sign of f′′(x)f''(x) on each side, then answer with inequalities.

Practise this

2 past-paper questions on 5.7 in the AA HL bank, each with a mark scheme and a worked explanation.

Practise 5.7 →