1

OVERVIEW: DIFFERENTIAL CALCULUS

 2
Table of Contents

1. The Derivative

2. Limits

3. Derivatives of Elementary Functions

4. Rules of Differentiation

5. Monotonicity

6. The Second Derivative

7. Extrema

Download this chapter (without trainings and quizzes) as PDF.

If you think you've already mastered the content of this chapter, you can proceed directly to the Final
Exam.
 3
Learning Goals

▪ To be able to calculate the derivative of a function with the help of the limits of difference
quotients (Section 1).

▪ To understand the relationship between the derivative of a function and the slope of the
tangent of its graph. To understand the derivative of a function as the instantaneous rate of
change of the function (Section 1).

▪ To be able to determine the limit of a function at a point, as well as the left-handed and right-
handed limits at that point. To be able to understand the behaviour of a function at infinity as
well as calculate its asymptotes (Abschnitt 2).

▪ To be able to take the derivative of elementary functions (Section 3).

▪ To be able to use the rules of differentiation (constant rule, sum rule, product rule, chain rule)
(Section 4).

▪ To be able to determine the qualitative behaviour of the graph of the derivative of a function
by looking at the graph of the function, and vice versa (Section 5).

▪ To understand the concavity of a function with the help of the second derivative (Section 6).

▪ To be able to calculate the inflection points and saddle points of a function (Section 6).

▪ To become familiar with the terms extremum, maximum, and minimum. To use the
necessary criterium for a function to have an extremum, as well as the sufficient critieria for a
maximum or minimum with respect to the first and second derivatives. To be able to
calculate the extrema of a function, what type of extrema they are, and how to sketch the
graph of a function (Section 7).

Summary

In this chapter you will learn about the derivative of a function. You will learn to understand the
derivative at a point as the slope of the tangent of the graph at that point, as well as how to use
difference quotients. You will find out how to calculate the derivative of a function, and how to
deal with important features of functions such as monotonicity, concavity, extrema, inflection
points, and saddle points all with the help of the first and second derivative.
 4

General Notation

The following notation is used throughout the entire chapter:

The derivative of a function f at a point x is denoted by f ′(x). The derivative of f is the function
f . The second derivative of f is written f ′′ instead of (f ′)′.
′

The left- and right-handed limits of a function f at a point x0 are denoted respectively by

lim f (x) and lim f (x).

x→x0− x→x0+

The limit of a function f at a point x0 or as x goes to +∞ or −∞ is written respectively as

lim f (x), lim f (x) and lim f (x).

x→x0 x→∞ x→−∞

The derivative of a function f at a point x is given by

f (x + h) − f (x)
′
f (x) = lim .
h→0 h

The derivative of f at a point x is the limit of the so-called difference quotient

f (x + h) − f (x)

as h approaches 0.
 5

1. THE DERIVATIVE
 6
Table of Contents
1.1 The Derivative as the Slope of the Tangent
1.2 The Derivative as the Rate of Change
1.3 The Derivative as a Linear Approximation

Learning Goals
In this section you will learn about the idea of a derivative and the relationship between the
derivative of a function and the slope of its tangent on its graph. You will also learn to think of the
derivative as the instantaneous rate of change of a function.

1.1 The Derivative as the Slope of the Tangent

In calculus we are normally interested in determining the properties of curves, such as critical points,
slopes of tangents, or asymptotic behaviour. We can also use calculus to approximate the value of a
function at a specific point if an exact calculation is either unneeded, cumbersome or even impossible to
do. In all of these problems the derivative of a function is very useful.

The concept of the derivative of a function f (x) at a point x0 and what the general idea entails will be
discussed and then expanded upon in detail. Mathematic precision is unfortunately not a possibility in the
scope of this course since the exact idea of limits isn't available. In general, such exacting precision isn't
necessary in high school mathematics. The idea of a derivative can still be described in enough detail to be
used in very interesting and useful applications while still remaining in the scope of this course. Calculus
has been used in such ways for hundreds of years without needing such precision, like in determining the
orbits of planets, for example.

In order to visualize a function f (x) we normally consider its graph. Using graphs we can describe the idea
of a derivative as follows: The derivative of the function f (x) at a point x0 is the slope of the tangent at the
point on the graph with coordinates P = (x0, f (x0)) .

Warning: The tangent of a graph doesn't necessarily exist at all points on the graph. In order to understand
this better, we need to further analyze the idea of the tangent of a graph at a point P = (x0, f (x0)) . To do
so, we'll consider the graph of f (x), a fixed point P , a moveable point near P called Q:
Q = (x, f (x)) = (x0 + h, f (x0 + h)) , and the line through P and Q called a secant line. Now if we let
Q move in a free manner in the neighbourhood of P (letting h tend to 0) we'll consider the behaviour of
the secant line. In many cases, the position of the secant approximates the end position (when h is at 0)
independently of how we move the point Q. In such cases we call the result of this limiting process the
tangent line or simply the tangent of f (x) at a point P .
In the following applet the function being visualized is f (x)
P = (x0, f (x0)) = (1, 1)

value 1.
The tangent at the point P

−1

Red graph: f (x)

in order to vary x

If P

point P
= Q
−0.5

, i.e. h
1.5

0.5

−0.5

−1

−1.5

= x

= 0
1

2
− x

= x0 + h = 1+0.5

the line through the point P and Q

0.5

+ 1
1
P

x0

– o
1.5

, fixed point: P
Use the mouse to move around the black point along the x-axis

the secant approximates the black tangent line as h approaches 0.

Q

←
2

↓ ↑ →
= x
2

= (x0, f (x0)) = (1, 1)

. The blue secant is

= (x, f (x)) = (1.5, 1.75)

, the secant is not defined, however

.
−

∣
x + 1

. Change the x-coordinate of the point Q to move Q around P . Convince

The following second visualization deals with an exception: a point where there is no tangent to the graph
and hence no derivative. Here, again we'll consider the function f (x)
= (x0, f (x0)) = (0,1)
= x
2
and

yourself that there is a tangent of f at the point P and that the derivative (the slope of the tangent) has the

. Convince yourself that the secant is different depending on which

direction the point Q approaches the point x0
No tangent at the point P
= 0
x

(from both the negative and positive sides).

+ 1 but this time at the
7
 f
8
Q

1.5

P
1

0.5

−1.5 −1 −0.5 0.5 1 1.5

−0.5

−1

−1.5

– o + ← ↓ ↑ →

Red graph: f (x) = x

2
− x + 1 , fixed Punkt: P = (x0, f (x0)) = (0, 1) .
Using the mouse, move the black point along the x-axis
in order to vary x = x0 + h = h = 1.5 . The blue secant is
the line through the points P and Q = (x, f (x)) = (1.5, 1.75) .
If P = Q , i.e. h = 0, the secant is not defined.

Like in the first visualization, if the secant line gets closer to the tangent, and hence the slope of the secant
approaches the slope of the tangent, we call f differentiable at x0.

1.1 DEFINITION

f is called differentiable at x0 if there exists a unique c ∈ R such that the slope of the secant

f (x0 + h) − f (x0)
(h ≠ 0)
h

is c as h goes to 0. This number c is called the derivative of f at x0 and we write

(1.1)
′
c =: f (x0).

Note: It's important to note that c = +∞ and c = −∞ aren't allowed as slopes!

 9

1.2 THEOREM

Let f be differentiable at the point x0. Then c ′

= f (x0) from Equation (1.1) is the slope of the
tangent of the graph of f at the point (x0, f (x0)).
The tangent T (x) of the graph of f at the point (x0, f (x0)) has the form

′
T (x) = f (x0) ⋅ (x − x0) + f (x0).

Up until now we've understood differentiability at a point x0 via the slope of secant. In order to calculate
′
f (x0) this visualization isn't very necessary. Now we'll learn an analogous but more abstract way to define
differentiability.

1.3 DEFINITION

Alternative definition of the derivative via difference quotients:

f (x0+h)−f (x0)
The expression Q(h) :=
h
for h ≠ 0 is called the difference quotient. (This is just
the slope of the secant through the points (x0, f (x0)) and (x0 + h, f (x0 + h)).)
If Q(h) approaches a unique limit as we let h tend to 0, then for this limit we write

f (x0 + h) − f (x0)
lim .
h→0 h

As per Equation (1.1):

f (x0 + h) − f (x0)
′
f (x0) = lim .
h→0 h

Here the values of +∞ and −∞ for f ′(x0) aren't allowed, as before.

 10
More detail into the idea of limits will be done later on in your studies, however you can use the section on
limits to gain a first understanding of limits. Expand the tab below for a brief overview.
 11
On limits

Up until now we've spoken about a function approaching something. If we say that an expression
f (x) approaches a when we let x get closer to x0, then we write

lim f (x) = a.
x→x0

In such cases, f (x) can get arbitrarily close to a depending on how close x gets to x0.

lim f (x) = a mathematically means that up to any tolerance ε > 0 there exists a region
x→x0

(x0 − δ, x0 + δ) with δ ε
= δ( ) > 0 so that

a − ε < f (x) < a + ε

for all x in the interval (x0 − δ, x0 + δ) where x ≠ x0. This means that f (x) can't vary in value
around a more than ε, if x ≠ x0 doesn't vary by more than δ around x0. The number a is called the
limit.
12
 13

1.4 EXAMPLE

1. Let f (x) = 2x + 1 . At any point x0 ∈ R and for h ∈ R :

f (x0 + h) − f (x0) = 2(x0 + h) + 1 − (2x0 + 1) = 2h.

f (x0+h)−f (x0)
The difference quotient is then Q(h) =
h
= 2 . Since Q(h) = 2 is constant,
Q(h) goes to 2 as h goes to 0. This means that f is differentiable at any point x0 and
′
f (x0) = 2 .

2. Let f (x) = |x| be the absolute value function: f (x) = x for x ≥ 0 and f (x) = −x for
x < 0 . If h > 0 then f (h) = h, hence f (0 + h) − f (0) = h and Q(h) = 1 .
If we look back to (1.1), the idea of a number c comes into question. If we consider h < 0

then we can see that beacuse f (0 + h) − f (0) = −h and Q(h) = −1 , the value of c for
h < 0 is c = −1 ; for h > 0 c = 1. We can't find a unique c for Q(h) as h approaches 0,
and hence f is not differentiable at 0.
 14

1.5 DEFINITION

If a function f is differentiable at every point of its domain we call f differentiable. In this case,
since the function f has a derivative f ′(x) at every point in its domain, we can think of f ′ as a
function that gives us the derivative of the function f at any point x0.

1.6 REMARK

1. The derivative will be denoted as the first derivative in order to distinguish it from higher
derivatives (such as the second derivative, introduced later on).

2. Because f ′(x0) is also the limit of the difference quotient, instead of writing f ′(x0) we
df df
sometimes write dx
. Instead of f ′ we can also write
(x0)
dx
if f is dependent on x.

Δy
3. The derivative f ′(x0) is sometimes written lim
Δx
. In this case Δx = h and
Δx→0

Δy = f (x0 + h) − f (x0) . The Δ (capital delta) stands in general for a difference, Δx

stands for a distance on the x-axis, Δy for a distance on the y-axis.

1.2 The Derivative as the Rate of Change

The derivative of a function at a point can also be interpreted as the local or instantaneous rate of change
of the function at that point.
An example of this is an airplane whose altitude is changing constantly. Let the function f (x) be the height
of an airplane above sea-level and the distance travelled be x. The difference f (y) − f (x) gives us the
f (y)−f (x)
altitude between y and x. If we build the quotient y−x
, then this is the average altitude change of the
flight path between the distances y and x. If we let y get closer and closer to a fixed, chosen x, then this
value approximates the the average altitude change of the flight path, or the local rate of change of f at x.
Here, the slope of the flight path is the local rate of change of f at x. This is just the definition of the
derivative of f at the point x:

f (y) − f (x)
′
f (x) = lim .
y→x y − x
 15
One further example is the distance travelled by someone walking. If we start a clock at time t = 0 and let
the distance travelled up to the time t ≥ 0 be denoted by x(t).

x(t+h)−x(t)
In the time interval from the time point t to t + h the average velocity is h
. If we let h get
smaller and smaller (approaching zero), then this average velocity becomes a better approximation for the
instantaneous velocity at a time point t. The instantaneous velocity is often denoted by v(t), hence we
have:

x(t + h) − x(t)
′
v(t) = lim = x (t).
h→0 h

The instantaneous velocity of x(t) corresponds to the velocity v(t), in that v(t) ′
= x (t) .
A special notation is sometimes used for time-dependent variables: instead of writing a vertical dash for
the derivative, we write a dot above the function: ẋ(t) instead of x′(t).

1.3 The Derivative as a Linear Approximation

The idea of the derivative is to approximate a function locally by a linear function; approximate general
curves via straight lines. How we can use the derivative of a function to linearly approximate a function can
be read about in the tab below.
 16
Linear Approximations

According to the definition of the differentiability of f at x0

f (x0 + h) − f (x0)
lim = c.
h→0 h

Hence for the difference

f (x0 + h) − f (x0)
− c =: R(h)
h

we need to have that:

f (x0 + h) − f (x0)
= c + R(h), (1.2)
h

and

lim R(h) = 0.
h→0

In (1.2), if we multiply both sides by h we get

f (x0 + h) − f (x0) = c ⋅ h + R(h) ⋅ h

and hence

f (x0 + h) = f (x0) + c ⋅ h + R(h) ⋅ h.

 17
Function values of f near x0 can be approximated by f (x0) + c ⋅ h (here, x0 + h counts as near x0
since h is small). The error in the approximation that we get - the difference between the exact
value f (x0 + h) and the approximation f (x0) + c ⋅ h - is given via R(h) ⋅ h. For small h we also
get a good approximation of f (x0 + h), namely f (x0) + c ⋅ h, since R(h) ⋅ h goes to 0 as h → 0 .
This motivates the following definition of differentiability of a function f at x0, which is equivalent
to the earlier definition presented.

1.7 DEFINITION

f is said to be differentiable at x0 if there exists a unique c ∈ R such that

f (x0 + h) = f (x0) + c ⋅ h + R(h) ⋅ h, (1.3)

with

lim R(h) = 0.
h→0

This c is called the derivative of f at x0 and we write

′
c =: f (x0).

Important: Note that here c = +∞ and c = −∞ aren't allowed as values, as before.

Back to Chapter Overview

 18

EXERCISE 1
 19
Calculate the derivative of the function f (x) 2
= x + 2x + 1 at the point x = 1 by determining the
difference quotient.

Answer

′
f (1) = 4
 20
Solution

Step 1: By definition, we have

f (1 + h) − f (1)
′
f (1) = lim
h→0 h

We first calculate the difference quotient

f (1 + h) − f (1)

for h ,
∈ R h ≠ 0 .

Step 2:

2 2
f (1+h)−f (1) (1+h) +2(1+h)+1−(1 +2+1)
=
h h

2
1+2h+h +2+2h+1−4
=
h

2
h +4h
=
h

= h + 4

Step 3: We now calculate the limit of the difference quotient as h .

→ 0

lim h + 4 = 4
h→0

So f ′(1) = 4 .
 21

EXERCISE 2
 22
Let f (x) = x
3
+ 2 . Calculate the slope of the tangent T (x) on the graph of f at the point (−1, f (−1)),
and specify the equation of T (x).

Answer

T (x) is of the form T (x) = 3x + 4 and has a slope of 3.

Solution

The tangent T (x) is of the form T (x) ′

= f (−1) ⋅ (x − (−1)) + f (−1) , where f ′(−1)
corresponds to the slope of T (x).

We calculate f ′(−1) using difference quotients. For h ≠ 0

3 3 2 3
f (−1 + h) − f (−1) (−1 + h) + 2 − ((−1) + 2) −1 + 3h − 3h + h + 2 − (−1
= =
h h h

As h → 0 , we get that 3 − 3h + h2 goes to 3. So

f (−1 + h) − f (−1)
′
f (−1) = lim = 3
h→0 h

Since f (−1) = (−1)

3
+ 2 = −1 + 2 = 1 , it follows that

′
T (x) = f (−1) ⋅ (x − (−1)) + f (−1) = 3(x + 1) + 1 = 3x + 4
 23

EXERCISE 3
 24
The derivative of a function f is given by f ′(x) . Find the value of x in R such that the tangent of
= 2x + 1

the graph of f is parallel to the line x = y .

Answer

The tangent of the graph of f is parallel to the line x = y at x = 0.

Solution

The line x = y has a slope of 1. So the tangent of the graph of f is parallel to the line x = y

when it also has a slope of 1.

In addition, the slope of the tangent of the graph of f at x is calculated by f ′(x). So we are
looking for x such that f ′(x) = 1 .

We set f ′(x) = 1 , so 2x + 1 = 1 . This holds only when x = 0 . Therefore, the tangent of the
graph of f has a slope of 1 only at x = 0 .
 25

EXERCISE 4
 26
Let f (x) =
1

1+x
2
. Calculate f ′(x) using difference quotients. Find the value of x such that the tangent on
the graph of the function is horizontal.

Answer

′
f (x) = −
2x

(1+x2)2
. The tangent of the graph of the function is horizontal when x = 0 .

Solution

f (x+h)−f (x)
We calculate the difference quotient h
.

1 1
− 2 2
f (x + h) − f (x) 1+(x+h)
2
1+x
2
1 + x − (1 + (x + h) ) −2x − h
= = =
2 2 2
h h h(1 + (x + h) )(1 + x ) (1 + (x + h) )(1 +

As h tends to 0, the expression becomes − (1+x

2x

)
. Therefore 2 2

2x
′
f (x) = −
2 2
(1 + x )

The tangent on the graph of f at x has the slope f ′(x). When the tangent is horizontal it has a
slope of 0. So, to find x we set

′
f (x) = 0

′
f (x) = 0 only when x = 0 . So, the tangent on the graph of f is horizontal at the point x = 0 .
 27

2. LIMITS
 28
Table of Contents
2.1 The Behaviour of a Function Close to a Gap in the Domain
2.2 The Limit of a Function at Infinity
2.3 The Limit of a Sequence

Learning Goals

In this section we will learn different behaviours of functions near specific points, e.g. at gaps in the
domain, as well as their behaviour with large and small arguments. You will also know the idea of
limits with the help of examples and be able to understand them and calculate them with ease.

This section is about the idea of limits of a function and, more generally, about how functions behave close
to gaps in their domain. We will also investigate their behaviour for very large arguments (for example, at
infinity).

We already learned about the idea of a limit in the previous section on the derivative of a function. The
derivative of a function f (x) at the point x0 is:

f (x0 + h) − f (x0)
lim .
h→0 h

f (x0+h)−f (x0)
There, we used the limit to understand the behaviour of the difference quotient h
for smaller
and smaller values of h. Substituting h = 0 does not work, because division by 0 is not allowed.

In order to be fluent with the ideas of limits and behaviour at infinity, in most cases it suffices to have a
descriptive understanding of the concepts. We will limit ourselves to these cases. The exact mathematical
discussion is the topic of university mathematics.

2.1 The Behaviour of a Function Close to a Gap in the Domain

There are different behaviours of a function close to specific points. In four of the following five examples
the function has a limit at the gaps in the domain. For these four cases the limit is a real number. In the sixth
example, the value of the function close to the gaps becomes extremely large. In this case, we say the
function diverges or tends to infinity. (Warning: infinity is not a number!) In the last two examples in this
section there is no limit, since the value of the functions behave differently as the approach the left and right
sides of the gaps in the domain. In one case they become very large, and in the other very small (i.e. a large
negative value).
 29

2.1 EXAMPLE (LIMIT AT A GAP IN THE DOMAIN)

Consider the function

sin(x)
f or x ≠ 0
x
f (x) = {
1
f or x = 0
2

on its maximal domain R. It has a gap in its domain at x = 0 .

Study the function's graph and investigate the behaviour of the function close to the point x = 0 .

WARNING

If the limit lim g(x) of a function g exists then it does not necessarily have to be equal to
x→x0

g(x0) . Because in case of lim g(x) ≠ g(x0) we say g is not continuous at x0

x→x0

You should be able to see that for very small x the values of f (x) are very close to 1. Alternatively,
we say that as the argument x tends towards zero, the value of f (x) tends towards 1.
 30
We say this as follows:
As x goes to zero, f (x) goes to 1
or
The limit of f (x) as x tends to 0 is 1

and in both cases we write:

lim f (x) = 1.
x→0

2.2 EXAMPLE (PARAMETER-DEPENDENT LIMITS)

Consider the family of functions

sin(ax)
f (a, x) = , a ∈ R.
x

Take a look at how the limits depend on the values of a with the help of the visualization below.

−6 −4 −2 2 4 6

−2

−4
a = 3.00
1 8 – o + ← ↓ ↑ →
 31

2.3 EXAMPLE (DERIVATIVE AS THE LIMIT OF THE DIFFERENCE QUOTIENT)

We already encountered an example of this type in the last section on the derivative. The difference
quotient of the polynomial p(x) = x
2
+ x − 2 at the point x0 = 2 is the function of the variable h:

p(2 + h) − p(2)
Δ(h) = .
h

It has a maximal domain of D(Δ) = R\{0} . In this case the limit, i.e. the derivative, can be
calculated easily, since the difference quotient Δ(h) can be simplified: Δ(h) = h + 5 . This
expression is not only defined on D(Δ) = R\{0} ; it is also defined at the point h = 0. The result is
then:
As h goes to zero, Δ(h) goes to 5. The limit is the derivative of the function p(x) at the point x0 = 2

2.4 EXAMPLE (LIMIT AT A GAP IN THE DOMAIN, VERSION 2)

Consider the function

2
x − 1
f (x) =
x + 1

on its maximal domain D(f ) = R\{−1} . The expression is not defined at x = −1 because the
denominator disappears there. Now we will investigate the function close to this gap in the domain.
The expression can be simplified using the binomial formula to:

2
x − 1
f (x) = = x − 1.
x + 1

Because x − 1 is defined on R, the limit can be calculated easily: The limit of f (x) at the point
x = −1 is −2.
 32

2.5 EXAMPLE (f (x) TENDS TO INFINITY AT A GAP IN THE DOMAIN)

Study the graph of the function

1
f (x) =
2
x

on its maximal domain D(f ) = R\{0} .

Notice that for arguments x that approach zero, the value of the function gets larger and larger.
Using short form: As x goes to zero, f (x) goes to ∞ (infinity) or The function f (x) tends to ∞ as x
approaches 0, and we write:

lim f (x) = ∞.
x→0

The graph of the function approaches but never touches the y-axis. We say that the y-axis is a
vertical asymptote of the function f (x) as x goes to zero.
 33

2.6 EXAMPLE (BEHAVIOUR LEFT AND RIGHT OF GAPS IN THE DOMAIN, VERSION 1)

The function

1
f (x) =
x

on its maximal domain D(f ) = R\{0} has a gap in the domain at x = 0. In order to study the
behaviour of the function around 0 we need to look at its graph.

The graph behaves differently on the left and right side of the gap in the domain (i.e. for x > 0 and
x < 0 ):

Case If we let the argument x approach zero from the right side (positive x-values), then the value
1: of the function f (x) gets bigger and bigger. We say: For positive arguments x approaching
zero, the value of the function f (x) approaches positive infinity, +∞.

Case If we let the argument x approach zero from the left side (negative x-values), then the value
2: of the function f (x) gets smaller and smaller. We say: For negative arguments x
approaching zero, the value of the function f (x) approaches negative infinity, −∞.
 34
The y-axis is a vertical asymptote of the function f (x) as x approaches zero.
 35

2.7 EXAMPLE (BEHAVIOUR LEFT AND RIGHT OF GAPS IN THE DOMAIN, VERSION 2)

x
f (x) =
1 − x

on its maximal domain D(f ) = R\{1} . In order to study the behaviour of the function close to the
gap in the domain at x = 1 , we need to investigate the graph.

The graph behaves differently on the left (x < 1 ) and right (x > 1 ) side of the gap in the domain.
This is because the numerator x is positive around the point x = 1 , however the denominator 1 − x
is positive and negative around x = 1:

Case If we let the argument x approach 1 from the right side then the value of the function f (x)
1: gets smaller and smaller. We say: For arguments x > 1 the value of the function f (x)
approaches negative infinity, −∞.

Case If we let the argument x approach 1 from the left side then the value of the function f (x)
2: gets larger and larger. We say: For arguments x < 1 the value of the function f (x)
approaches positive infinity, +∞.

The vertical line {(1, y) ∣ y ∈ R} is an asymptote of the function f (x) around the point x = 1 .
 36
With the above two examples we have seen the cases that arise in practice when describing the
behaviour of functions near gaps in their domains.

2.2 The Limit of a Function at Infinity

We are often interested in the behaviour of a function g at infinity, i.e. the behaviour of g(x) as x
approaches +∞ or −∞. In looking for the behaviour of g(x) as x approaches +∞, we substitute larger
and larger x values into g. If g(x) approaches a fixed value a ∈ R independently of the x value chosen,
then we say that a is the limit of g(x) as x approaches +∞, and we write lim g(x) . Analogously, we can
x→∞

analyze the behaviour as x approaches −∞ - here we substitute smaller and smaller x, and then we call the
limit lim , as long as the limit exists.
g(x)
x→−∞
 37

2.8 EXAMPLE (HORIZONTAL ASYMPTOTE, VERSION 1)

2
f (x) = , x → +∞
x + 2

The denominator (x + 2) of f (x) grows without limit as x goes to +∞, and hence f (x)
approaches zero, i.e. .
lim f (x) = 0
x→∞
 38

2.9 EXAMPLE (HORIZONTAL ASYMPTOTE, VERSION 2)

3x
f (x) = , x → +∞
x + 5

We write

3x 3x + 15 15 15
= − = 3 − .
x + 5 x + 5 x + 5 x + 5

The denominator (x + 5) of 15

x+5
grows without limit as x goes to +∞, and hence 15

x+5
approaches
zero, i.e. .
lim f (x) = 3
x→∞
 39

2.10 EXAMPLE (OPBLIQUE ASYMPTOTE)

We consider the function

5
x + 1
f (x) = , x → +∞.
4
x

5
x + 1 1
= x + .
4 4
x x

If we define fas(x) = x, then the difference f (x) − fas(x) =

1
4
x
approaches 0 as x approaches
infinity. Therefore, fas(x) = x is the asymptote of f (x) as x approaches infinity.

Click on the tab below for an example of when a limit does not exist.
 40
Example

The limits lim g(x) and lim g(x) of a function g do not necessarily have to exist, in general. For
x→+∞ x→−∞

example, sin(x) does not approach a fixed value as x approaches +∞. If we substitute the sequence
of all natural multiples of 2π into sin x, it is zero at all of those points. If we add to each of those
π

values and then substitute this new sequence into sin x, it is always equal to 1. There is no number
that sin x approaches as x goes to +∞; it oscillates forever.

2.3 The Limit of a Sequence

2.11 DEFINITION

Let g be a function defined on the natural numbers, i.e. g : N → R . Then g is called a sequence.

2.12 EXAMPLE

1. For example, the constant sequence g where g(n) = 1 for all n ∈ N is an example of a very
simple sequence.

2. g(n) = (−1) n
n
for n ∈ N is one further example of a somewhat more complicated sequence.
The first terms in the sequence are −1,2, −3,4, ….
 41
If g is a sequence, then we are often interested in the behaviour of g(n) for increasing n. If g(n)
approaches a unique value a ∈ R as n increases, then we say that g converges and we call a the
limit of g:

a = lim g(n).
n→∞

For an exact definition, click on the tab below.

Definition

The exact definition goes as follows: if there exists an a ∈ R such that for all ε > 0 there exists an
N ∈ N such that

a − ε < g(n) < a + ε

for all n ≥ N , then we say that

lim g(n)
n→+∞

exists and that

lim g(n) = a.
n→+∞
 42

2.13 EXAMPLE

1. A simple example is f (n) = 1 , i.e. f is constant. In this example, lim f (n) = a.

n→∞

1
2. f (n) = , n → ∞
n

Because the denominator grows without bound and the numerator is constant, f (n) tends to
zero as n goes to ∞.

WARNING

As with the limits of functions, the limit of a sequence does not necessarily have to exist!

2.14 EXAMPLE

n
f (n) = (−1) , n → ∞

If n is an even number, then (−1)n = 1 . If n is an odd number, then (−1)n = −1, thus f (n)
swaps between 1 and −1. There is thus no unique number in R (or +∞ or −∞) that f (n)
approaches as close as we desire, hence f (n) has no limit.

Back to Chapter Overview

 43

EXERCISE 1
 44
|x|
Verify whether lim
x
exists.
x→0

Answer

|x|
lim
x
does not exist.
x→0

Solution

|x|
Step 1: We check whether the right-hand and left-hand limits of x
at 0 exist and whether they
are equal.

|x| |x|
Step 2: For x > 0 , x
= 1 . So the right-hand limit of x
at 0 exists, and

|x|
lim = lim 1 = 1
x→0+ x x→0+

|x| |x|
Step 3: For x < 0 , x
= −1 . So the left-hand limit of x
at 0 exists, and

|x|
lim = lim − 1 = −1
x→0− x x→0−

|x| |x|
Step 4: Since the right-hand and left-hand limits of x
at 0 do not match, lim
x
does not exist.
x→0
 45

EXERCISE 2
 46
Verify whether lim
2x

x+2
exists and determine this limit if it does exist.
x→∞

Answer

lim
2x

x+2
exists and lim
2x

x+2
.
= 2
x→∞ x→∞
 47
Solution

Step 1: We first re-write the expression 2x

x+2
. For x > 0

1
2x 2x x
2
= ⋅ = .
1 2
x + 2 x + 2 1 +
x x

Step 2: The denominator 1 + 2

x
of the quotient 1+
2
2
tends to 1 as x , i.e.
→ ∞
x

2
lim (1 + ) = 1.
x→∞ x

Step 3: The numerator 2 of the quotient 1+

2
2
is constant, and so
x

lim 2 = 2.
x→∞
 48
Step 4: Since the limit of the denominator is 1 and not 0, we can calculate the limit of the quotient
as the quotient of the limits:

lim 2
2 x→∞
lim = = 2.
x→∞ 2 2
1 + lim (1 + )
x x
x→∞

Therefore it follows

2x
lim = 2.
x→∞ x + 2
 49

EXERCISE 3
 50
Verify whether lim
2x+1
2
−x −x+2
exists and determine this limit if it does exist. What can you say about the
x→−∞
2

limit of the reciprocal, i.e. ?

−x −x+2
lim
2x+1
x→−∞

Answer

3 2

The limit exists, and . In addition, exists in an

2x+1 x +2x+1 −x −x+2
lim 2
lim 2
= 0 lim 3
−x −x+2 −x −x+2 x +2x+1
x→−∞ x→−∞ x→−∞
2

improper sense and .

−x −x+2
lim 3
= ∞
x +2x+1
x→−∞
 51
Solution

The function f (x) :=

2x+1
2
−x −x+2
is the quotient of 2x + 1 and −x2 − x + 2.

We divide the numerator and denominator of f (x) by the highest power of x in the denominator,
i.e. x2.

We obtain:

1 2 1
2x + 1 2 2x + 1 + 2
x x x
f (x) = = ⋅ = .
2 1 2 1 2
−x − x + 2 −x − x + 2 −1 − +
2 2
x x x

As x tends to −∞ we get that 2

x
, 1
2
x
, − x1 and x
2
2
tend to zero.

So, as x tends to −∞, 2

x
+
1

x
2
tends to 0 and −1 − 1

x
+
2
2
x
tends to −1.

Since the limit of the new denominator is the real number −1, we can divide the numerator by
the denominator within the limit. We obtain:

2 1
2 1 lim +
2x + 1 + 2 x→−∞
x 2
x 0
x x
lim = lim = = = 0.
2 1 2 1 2
x→−∞ −x − x + 2 x→−∞
−1 − + lim − 1 − + −1
2 2
x x x x
x→−∞

Now we can discuss . Analogously to the first limit, we divide the numerator and
−x −x+2
lim
2x+1
x→−∞

denominator by x:

2 2
−x − x + 2 −x − 1 +
x
=
1
2x + 1 2 +
x

Since , , the quotient must tend to ∞.

2 1
lim − x − 1 + = ∞ lim 2 + = 2
x x
x→−∞ x→−∞

So this limit is an improper limit.

 52

EXERCISE 4
 53
Verify whether lim
x

x+1
exists and determine the limit if it does exist.
x→−1

Answer

lim
x

x+1
does not exist.
x→−1
 54
Solution

Step 1: We write

x x + 1 − 1 x + 1 1 1
= = − = 1 − . (1)
x + 1 x + 1 x + 1 x + 1 x + 1

Step 2: From the section on Limits, we are able to calculate that

1
lim = +∞
x→−1+ x + 1

and

1
lim = −∞.
x→−1− x + 1
 55
Step 3: Now with equation (1) we can conclude immediately that

x
lim = −∞
x→−1+ x + 1

and

x
lim = +∞.
x→−1− x + 1
 56

3. DERIVATIVES OF ELEMENTARY
FUNCTIONS
 57

Learning Goals

In this section we'll learn about the derivatives of certain simple elementary functions.

Now that we know in principle how we can take the derivative of a function f (by taking the limit of the
difference quotient) we can take the derivative of many different functions.

3.1 EXAMPLE

If we consider the function f (x) = c where c is a fixed real number, we can calculate:

f (x + h) − f (x) c − c
′
f (x) = lim = lim = lim 0 = 0.
h→0 h h→0 h h→0

If we take the derivative of a constant function, the result ends up being 0.

The next most complicated functions that we'd like to take the derivative of are monomials and
polynomials.
 58

3.2 EXAMPLE

1. First, we'll take the derivative of the function f (x) = x . We get

f (x + h) − f (x) x + h − x h
′
f (x) = lim = lim = lim = lim 1 = 1.
h→0 h h→0 h h→0 h h→0

2. Now let f (x) = x

2
, a quadratic monomial. We get:

′ f (x+h)−f (x)
f (x)= lim
h
h→0

2 2
(x+h) −x
= lim
h
h→0

2 2 2
x +2xh+h −x
= lim
h
h→0

2
2xh+h
= lim
h
h→0

= lim 2x + h
h→0

= 2x

For higher exponents, the following consideration can be useful: If f (x) = x

n
with n ≥ 3 , then (x + h)n
has the form xn + nxn−1h+remainder, where the remainder is the sum of the multiples of powers in h
of order 2 or higher. (This can be seen by using the Binomial Theorem, which is a generalization of the first
binomial formula. For n = 3 we get:
(x + h)
3
= (x + h) ⋅ (x + h)
2
= (x + h) ⋅ (x
2 2
+ 2xh + h ) = x
3 2
+ 3x h + 3xh
2
+ h
3
.) The
derivative can then be calculated to be nxn−1.
In the next section we'll learn about the product rule which will let us take the derivative of xn without
needing the Binomial Theorem; we just need the knowledge of the derivative of x.
 59

3.3 EXAMPLE

Now we'll look at the function , which is defined in x . If x and h is small enough,
1
∈ R ∖ {0} ≠ 0
x

then

1 1
f (x+h)−f (x) −
x+h x
=
h h

1 x−(x+h)
= ⋅
h x(x+h)

1 −h
= ⋅
h x(x+h)

1
= − ,
x(x+h)

and hence

1 1
′
f (x) = lim − = − .
2
h→0 x(x + h) x
 60

3.4 EXAMPLE

For the square root f (x) (

= √x x ≥ 0 ) for x > 0 and h small enough we get:

f (x + h) − f (x) = √ x + h − √x

(√ x+h−√x)⋅(√ x+h+√x)
=
√ x+h+√x

2 2
(√ x+h) −(√x)
=
√ x+h+√ x

x+h−x
=
√ x+h+√x

h
= ,
√ x+h+√x

i.e.

f (x + h) − f (x) 1
= .
h √ x + h + √x

Therefore

1 1
′
f (x) = lim = .
h→0 √ x + h + √x 2√x

We can see in this example that the square root is not differentiable at 0.

Now that we know some of the derivatives of elementary functions, here is a short table summarizing
some of the results we know.
 61

3.5 RULE

′
Function f (x) Derivative f (x) Conditions on (1)
¯
¯
¯
c (c ∈ R) 0 (2)

n n−1
x (n ∈ N0) nx (3)

n n−1
x (n ∈ Z, n < 0) nx x ≠ 0 (4)

1
√x x > 0 (5)
2√x

x x
e e (6)

1
ln(x) x > 0 (7)
x

x x
a (a > 0) ln a ⋅ a (8)

sin(x) cos(x) (9)

cos(x) − sin(x) (10)

Back to Chapter Overview

 62

EXERCISE 1
 63
Show with the help of difference quotients that the derivative of f (x) = x
3
is 3x2, as indicated in the
table.

Solution

Step 1: For fixed x ∈ R and h ,

∈ R h ≠ 0 we calculate the difference quotient

f (x + h) − f (x)

3 3 3 2 2 3 3
f (x + h) − f (x) (x + h) − x x + 3x h + 3xh + h − x
2 2
= = = 3x + 3xh + h
h h h

Step 2: Now we let h tend to 0

2 2 2
lim 3x + 3xh + h = 3x
h→0

So f ′(x) = 3x
2
.
 64

EXERCISE 2
 65
Show with the help of difference quotients that the derivative of the function f (x) =
1
2
x
is − x2 , as
3

indicated in the table.

Solution

Step 1: For a fixed x ∈ R and h ,

∈ R h ≠ 0 we calculate the difference quotient

f (x + h) − f (x)

1 1
− 2 2
f (x + h) − f (x) (x+h)
2
x
2
x − (x + h) 2x + h
= = = −
2 2 2 2
h h x (x + h) h x (x + h)

Step 2: Now we let h tend to 0.

2x + h 2x 2
lim − = − = −
2 2 4 3
h→0 x (x + h) x x

So f ′(x) .
2
= − 3
x
 66

EXERCISE 3
 67
Show that the function f (x) = √x is not differentiable in x . Note that the function f is not defined
= 0

for negative values of x. We therefore assume that x ≥ 0 .

 68
Solution

Step 1: We consider the differential quotient for x = 0 . In order for it to be defined at all, we
assume that h > 0.

f (0 + h) − f (0)

f (0 + h) − f (0) √0 + h − √0 √h 1
= = =
h h h √h

Step 2: We check the behaviour of the difference quotient as h tends to 0.

 69

1
lim = +∞
h→0 √h

Since we cannot have +∞ as a value for the slope, f (x) is not differentiable at x = 0.

As you can see in the picture, the y-axis is tangent to the graph of √x at the point (0,0).
 70

EXERCISE 4
 71
Let f (x) = sin(x) be the sine function. Draw the graph of f and draw the tangents of the graph at the
points x ,
= 0 x =
π

2
, π, and 3π

2
. Try to use these points to help derive the derivative of f (x) .
= sin(x)
 72
Solution

The graph of sine has the following form:

The tangents of the graph of sin(x) at the point x = 0 is the line y , as can be seen from the
= x

image. The slope of the line y = x is 1, hence f ′

(0) = 1 .

At the point x =
π

2
the tangent of the graph of sin(x) is parallel to the x-axis. This tangent has a
slope of 0, and hence f ′( π2 ) = 0 .
 73
At x = π the tangent of the graph of sin(x) is parallel to the line y = −x . The line y = −x has
a slope of −1, i.e. f . At x the tangent of the graph of sin(x) is parallel to the x-
′ 3π
(π) = −1 =
2

axis. Here, f ′
(
π

2
) = 0 . At x , ′
= 2π f (2π) = 1 .

The the values of the function f ′(x) at the respective points are: f ′(0) , ,
′ π
= 1 f ( ) = 0
2

, , and f .
′ ′ 3π ′
f (π) = −1 f ( ) = 0 (2π) = 1
2
 74
If we were to complete the values of the tangent of the graph of sin(x) at more values than the
ones considered here, we would see that the slopes of the tangents (and hence the value of f ′ at
these points) decrease to −1 as we go from 0 to π. If we move from π to 2π, then the the slopes
of the tangents of the graph increase (as does the value of f ′) up to 1. The following image shows
the approximate behaviour of the graph of f ′(x):

This is the graph of cos(x). As we know from formula sheets, cos(x) is the actually the derivative
of sin(x), i.e. f ′(x) .
= cos(x)
 75

4. RULES FOR DIFFERENTIATION

 76

Learning Goals

In this section we will learn step-by-step methods for taking the derivatives of sums, powers,
products, quotients, as well as compositions and constant multiples of functions.

From the previous sections we already know how to take the derivative of a function, and we've learned
the derivatives of the most important elementary functions. For the operations on combinations and
compositions of functions, there are a set of rules that break down taking the derivative into manageable
parts.
If f and g are differentiable functions, then

′ (f +g)(x+h)−(f +g)(x)
(f + g) (x) = lim
h
h→0

f (x+h)+g(x+h)−f (x)−g(x)
= lim
h
h→0

f (x+h)−f (x) g(x+h)−g(x)

= lim ( + )
h h
h→0

′ ′
= f (x) + g (x)

The sum of differentiable functions is also differentiable. The name of the rule comes from the operation
we just performed: the Sum Rule.

4.1 RULE

Sum Rule:
Given two differentiable functions f and g then

′ ′ ′
(f + g) (x) = f (x) + g (x).

Hence any finite sum of differentiable functions f1, f2, … , fn is also differentiable:

′ ′ ′ ′
(f1 + f2 + … fn) = f1 + f2 + … fn .
 77

4.2 EXAMPLE

We can calculate the derivative of the function f (x) = x

3
+ x + 1. This function is the sum of the
functions x3, x, and the constant function 1. According to the sum rule, this is just

′ 2
f (x) = 3x + 1

If we multiply a differentiable function f with a real number c, then the product is differentiable. The
derivative is then

c ⋅ f (x + h) − c ⋅ f (x) f (x + h) − f (x)
′ ′
(c ⋅ f ) (x) = lim = c ⋅ lim = c ⋅ f (x).
h→0 h h→0 h

Summarizing you get the so called constant rule.

4.3 RULE

The Constant Rule:

For any c ∈ R and differentiable f ,

′ ′
(c ⋅ f ) (x) = c ⋅ f (x).

How we take the derivative of the product of differentiable functions is described by the product rule.
 78

4.4 RULE

The Product Rule:

If f and g are two differentiable functions, then

′ ′ ′
(f ⋅ g) (x) = f (x) ⋅ g(x) + f (x) ⋅ g (x).

Derivation

(f ⋅g)(x+h)−(f ⋅g)(x)
′
(f ⋅ g) (x) = lim
h
h→0

f (x+h)⋅g(x+h)−f (x)⋅g(x)
= lim
h
h→0

f (x+h)⋅g(x+h)−f (x)⋅g(x+h)+f (x)⋅g(x+h)−f (x)⋅g(x)

= lim
h
h→0

(f (x+h)−f (x))⋅g(x+h)+f (x)⋅(g(x+h)−g(x))

= lim
h
h→0

f (x+h)−f (x) g(x+h)−g(x)

= lim ( ⋅ g(x + h) + f (x) ⋅ )
h h
h→0

′ ′
= f (x) ⋅ g(x) + f (x) ⋅ g (x)

4.5 EXAMPLE

As an example we can calculate the derivative of the function F (x) , which is the
1
= 2
⋅ sin x
x

product F (x) of the differentiable functions f (x) and g(x) . The

1
= f (x) ⋅ g(x) = 2
= sin x
x

derivative of f (x) is − x and the derivative of g(x) is cos x. Then

2
3

2 1
′ ′ ′
F (x) = f (x) ⋅ g(x) + f (x) ⋅ g (x) = − ⋅ sin x + ⋅ cos x.
3 2
x x
 79

4.6 REMARK

In the earlier section we stated that the derivative of every constant function is 0. According to the
product rule, if we treat g(x) = c as a function:

′ ′ ′ ′ ′ ′
(c ⋅ f ) (x) = (g ⋅ f ) (x) = g (x) ⋅ f (x) + g(x) ⋅ f (x) = 0 ⋅ f (x) + c ⋅ f (x) = c ⋅ f (x).

From this example we can see that the constant rule is simply a special case of the product rule.

Sometimes we would also like to take the derivative of composed functions of the form f (g(x)) (The
composition of functions is explained in Chapter VI Section 8). When dealing with compositions, we call f
the outer function and g the inner function. An example for such a composed function would be with
2

f (y) = e
y
,y = g(x) = x
2
: f (g(x)) = e
x
. We'll also often write (f ∘ g)(x) (spoken: f composed with g)
instead of f (g(x)) (spoken f of g of x).

The differentiation of the composition of two functions leads to the chain rule.

4.7 RULE

The Chain Rule

If f and g are differentiable functions, then

′ ′ ′
(f ∘ g) (x) = f (g(x)) ⋅ g (x)

We take the derivative of f , f ′, and evaluate it at g(x) and then take the derivative of the function g(x)
with respect to x and multiply both together.
 80

4.8 EXAMPLE

1. As an example, we can calculate the derivative of the function sin(x2). First we let the outer
function be f (y) = sin y and then use g(x) = x
2
as the inner function, and hence
f (g(x)) = sin(x )
2
. Since we have the composition of two functions, we can use the chain
rule here. First, f ′
(y) = cos y . For y = g(x) = x
2
we get that:

′ 2
f (g(x)) = cos(g(x)) = cos(x ).

Now we still need to calculate the derivative of g:

′
g (x) = 2x.

According to the chain rule, the derivative of f (g(x)) = sin(x )

2
is:

′ ′ ′ 2
(f ∘ g) (x) = f (g(x)) ⋅ g (x) = 2x cos(x ).

Once we have gained enough experience taking the derivative of functions, we do not need to
perform these small steps any longer. The calculation can be reduced and the last line can be
written down in one step.

2. If we know the derivative of a function f (x), then we also know the derivative of 1

f (x)
. If f is
differentiable and f (x) ≠ 0 , if we let h(y) =
1

y
then 1

f (x)
= (h ∘ f )(x) is differentiable as
per the chain rule, and

′
1 1 f (x)
′ ′ ′ ′
( ) = h (f (x)) ⋅ f (x) = − ⋅ f (x) = − .
2 2
f (x) (f (x)) (f (x))

f (x)
3. The previous example can be used to determine the derivative of a function of the form g(x)
.
f (x)
Let f and g be differentiable functions and g(x) ≠ 0 . Then interpret the quotient g(x)
as the
product f (x) ⋅ 1

g(x)
and apply the product rule
 81
f (x) 1 1 1
′ ′ ′ ′
( ) = (f (x) ⋅ ) = f (x) ⋅ + f (x) ⋅ ( ).
g(x) g(x) g(x) g(x)

We know the derivative of 1

g(x)
after 2. and obtain

′
1 1 g (x)
′ ′
(f (x) ⋅ ) = f (x) ⋅ + f (x) ⋅ (− ).
2
g(x) g(x) (g(x))

If you bring the right-hand side to the main denominator, you have obtained quotient rule of
differentiation

′ ′
f (x) f (x) ⋅ g(x) − f (x) ⋅ g (x)
′
( ) = .
2
g(x) (g(x))

Summarizing, we get the following rules for differentiation:

4.9 RULE

Function Derivative Conditions

¯
¯
¯ ′ ′
f (x) + g(x) f (x) + g (x)

′
c ⋅ f (x) c ⋅ f (x)

′ ′
f (x) ⋅ g(x) f (x) ⋅ g(x) + f (x) ⋅ g (x)

′ ′
(f ∘ g)(x) f (g(x)) ⋅ g (x)
′
1 f (x)
− 2
f (x) ≠ 0
f (x) (f (x))

′ ′
f (x) f (x)⋅g(x)−f (x)⋅g (x)
2
g(x) ≠ 0
g(x) (g(x))

Back to Chapter Overview

82
 83

EXERCISE 1
 84
Calculate the derivative of the function h(x) 2
= (1 − x )
9
using the chain rule.

Answer

′ 2 8
h (x) = −18x(1 − x )
 85
Solution

Step 1: We aim to express h as a composition of functions whose derivatives we know.

We set f (y) = y
9
as the outer function and g(x) = 1 − x
2
as the inner function, so
f (g(x)) = (1 − x )
2 9
= h(x) . Using the chain rule, we can calculate the following:

′ ′ ′
h (x) = f (g(x)) ⋅ g (x).

First of all

′ 8
f (y) = 9y .

With y = g(x) = 1 − x
2
, we get

′ 8 2 8
f (g(x)) = 9(g(x)) = 9(1 − x ) .

Now we need to calculate the derivative of g. We obtain

′
g (x) = −2x.

Then, after the chain rule

′ ′ ′ 2 8 2 8
(f ∘ g) (x) = f (g(x)) ⋅ g (x) = 9(1 − x ) ⋅ (−2x) = −18x(1 − x )
 86

EXERCISE 2
 87
Calculate the derivative of the function h(x) 2x
= xe .

Answer

′ 2x
h (x) = (1 + 2x)e

Solution

Step 1: We aim to express h as the sum, product or composition of functions whose derivatives
we know.

h(x) is the product of x and e2x.

The derivative of x is 1. Moreover, we know the derivative of the function f (y) y

= e and the
derivative of the function g(x) = 2x. Therefore, by using the chain rule, we can calculate the
derivative of e2x.

′ ′ ′ g(x) ′ 2x
(f ∘ g) (x) = f (g(x)) ⋅ g (x) = e ⋅ g (x) = 2e

We can now determine the derivative of h using the product rule.

′ ′ 2x 2x 2x
h (x) = 1 ⋅ (f ∘ g)(x) + x ⋅ (f ∘ g) (x) = e + 2xe = (1 + 2x)e
 88

EXERCISE 3
 89
Calculate the derivative of the function f (x) = a
x
(for a > 0), by representing ax with the help of the
exponential function ex.

Answer

′ x
f (x) = ln a ⋅ a

Solution

First we write

x x⋅ln a
a = e

We see that we can differentiate f using the chain rule. The outer function is ey, the inner
function is x ln a.

Then

x ′ x⋅ln a ′ x⋅ln a x
(a ) = (e ) = ln a ⋅ e = ln a ⋅ a

This is the formula that you know from the previous section.
 90

EXERCISE 4
 91
Prove the quotient rule: f and g are differentiable functions and g(x) ≠ 0 such that

′ ′ ′
f (x) f (x) ⋅ g(x) − f (x) ⋅ g (x)
( ) =
2
g(x) (g(x))

Use the product rule and the rule for the derivative of a reciprocal function.

Solution

We write the quotient as a product

′ ′
f (x) 1
( ) = (f (x) ⋅ )
g(x) g(x)

It follows from the product rule that

′ ′
1 1 1
′
(f (x) ⋅ ) = f (x) ⋅ + f (x) ⋅ ( )
g(x) g(x) g(x)

From the section on Derivatives, we are able to calculate the derivative of 1

g(x)
. Thus, we obtain

′ ′ ′
1 g (x) f (x) ⋅ g(x) − f (x) ⋅ g (x)
′
f (x) ⋅ − f (x) ⋅ =
2 2
g(x) (g(x)) (g(x))
 92

5. MONOTONICITY OF FUNCTIONS
 93
Table of Contents
5.1 Monotonicity
5.2 Qualitative Behaviour of the Graphs of f and f ′

Learning Goals

In this section you will learn the relationship between the first deriative of a function and its
monotonicity and how to determine regions where functions increase or decrease.

5.1 Monotonicity
In calculus we are often interested in the regions in which a function f is increasing or decreasing. Finding
these regions is important when looking for the extrema of a given function. The monotonicity of a
function f together with its derivative f ′ will be brought to light here (the idea of monotonicity was
introduced already in Chapter VI, Section 1.4).

Let f = f (x) be a monotonically increasing function on an interval I . Then

f (x + h) − f (x) ≥ 0 f or all h > 0 and

f (x + h) − f (x) ≤ 0 f or all h < 0

where x + h ∈ I . Note that the difference quotient in both cases is:

f (x + h) − f (x)
≥ 0
h

Of course, the limit of the difference quotient as h → 0 greater than or equal to 0, i.e. f ′(x) .
≥ 0
 94
Given a differentiable function f that is monotonically increasing on an interval I , then the first
derivative of f is greater than or equal to 0, i.e.

′
f (x) ≥ 0, x ∈ I

In a similar fashion we can show that for monotonically decreasing functions, f ′(x) ≤ 0 for
x ∈ I .
Conversely, from the fact that f ′(x) ≥ 0 for x ∈ I , we can show that f is monotonically
increasing on I and from the fact that f ′(x) > 0 that it is strictly monotonically increasing.
Similar results hold for decreasing functions.
 95

5.1 EXAMPLE

The function f (x) = x

2
is monotonically decreasing on (−∞,0] and monotonically increasing on
[0, ∞) . The slope of the tangent at any point x0 with x0 ≤ 0 is f ′(x0) . The slope of the
≤ 0

tangent at any point x1 with x1 ≥ 0 is positive: f ′

(x1) ≥ 0.
Because the tangent at x0 < 0 has a negative slope, f ′(x) < 0 for x ∈ (−∞,0) , and hence f is
strictly monotonically decreasing on (−∞,0). Analogous results hold for f ′(x) > 0 on (0, ∞): f
is strictly monotonically increasing on (0, ∞).

The slope of a differentiable function is closely related to the sign of its first derivative. If the first derivative
of a function f vanishes at a point x0, then the tangent of the graph at that point is horizontal. Points on
the graph of a function where the the tangent line is horizontal are called critical or stationary points.

5.2 DEFINITION

Any point at which f ′(x) = 0 is called a critical point or a stationary point of f . We say that f has
a critical point at x or at the point (x, f (x)).
 96

5.3 EXAMPLE

The function f (x) = x

3
has a critical point at x , since f ′(x)
= 0 = 3x
2
and f ′(0) . The
= 0

function g(x) = x
3
− 3x has a first derivative of g′(x) = 3x
2
− 3 , and since
′ ′
g (1) = 0 = g (−1) g , has critical points at x = 1 and x = −1 .

To summarize:

5.4 RULE

′
f (x) on I Monotonicity of f on I (1)

¯
¯ ′
f (x) ≥ 0 ⇔f monotonically increasing (2)

′
f (x) ≤ 0 ⇔f monotonically decreasing (3)

and (4)

′
f (x) > 0 ⇒f strictly monotonically increasing (5)

′
f (x) < 0 ⇒f strictly monotonically decreasing (6)

5.2 Qualitative Behaviour of the Graphs of f and f ′

The previous section described how the monotonicity of a function corresponds to the properties of its
derivative. We will now take a look at two Mathlets in order to visualize this relationship. Both Mathlets
show a polynomial of the third degree (f ) and its first derivative f ′. In the first Mathlet you can make
changes to the function f , and in the second Mathlet you can make changes to the first derivative f ′.
A Function and its Derivative
 f 97

f 1

−5 −4 −3 −2 −1 1 2 3 4 5

−1

−2

−3

−4

−5

– o + ← ↓ ↑ →

In order to change the the blue graph of f = ax

3
+ bx
2
+ cx + d ,
choose the variables a, b, c and d:

a : 1 , b : 3 , c : 1 , d : −1

The red curve is the graph of the derivative f ′.

Observe how the monotonic intervals correspond to the sign and roots of the first derivative f ′.
Try values b and c where b2 < 3c b, 2
= 3c, and b2 > 3c !

In this Mathlet you can compare the monotonicity of the function f (blue graph) with the properties of f ′
(red graph) such as its sign and its roots. You also have the chance to define the roots of f ′.
A Derivative and its Corresponding Function
 98

f
3

′
f
2

−5 −4 −3 −2 −1 1 2 3 4 5

−1

−2

−3

−4

−5

– o + ← ↓ ↑ →

In order to change the blue graph of f with its derivative f ′(x) = a(x − b)(x − c) (displayed
in red), choose the variables a, b and c: (b and c are the roots of f !) ′

a : −1 ， b : −1 ， c : 2

Observe how the monotonic intervals correspond to the sign and roots of the derivative f ′.
Try values b and c with b ≠ c and b = c !
In contrast with the earlier Mathlet, f ′ has forced roots due to the required form of f above.

Back to Chapter Overview

 99

EXERCISE 1
 100
The derivative of a function f should have the following sign:

Interval (−∞, −1) −1 (−1,1) 1 (1,3) 3 (3, ∞)

′
f + 0 − 0 + 0 +

Sketch a possible graph of f .

 101
Solution

Since the slope is positive in the interval (−∞, −1), f is monotonically increasing. In addition,
′
f (−1) = 0 i.e. at x , the tangent on the graph has a slope of 0. Therefore, the graph of f
= −1

in the interval (−∞, −1] should be roughly of the following form:

 102
Since the slope is negative in the interval (−1,1), f is monotonically decreasing. The tangent is
horizontal again at x = 1 because f ′(1) .
= 0

The slope is positive again in the interval (1,3), so f increases. Since f ′(3) ,
= 0 x = 3 is a
stationary point.
 103
Since f ′(x) is positive for x ,
> 3 f continues to increase in the interval (3, ∞).

The derivative described above could be, for example, f ′(x) = (x + 1)(x − 1)(x − 3)
2
.
 104

EXERCISE 2
 105
Consider the graph of a function f . Determine the sign of the derivative of f in suitable intervals.

Solution

f is monotonically increasing in the interval (−∞, x0), so the sign of f ′ is positive there.

The tangent on the graph is horizontal at x0, so f ′(x0) = 0 .

f is monotonically decreasing in the interval (x0,0), so the sign of f ′ is negative there.

′
f (0) = 0 again at x = 0 , since the tangent on the graph has a slope of 0.

f increases monotonically in the interval (0, ∞). Therefore f ′ has a positive sign there.

Altogether we obtain the following table:

Interval (−∞, x0) x0 (x0,0) 0 (0, ∞)

′
f + 0 − 0 +
 106

EXERCISE 3
 107
Consider the function f (x) = x
4
+
4

3
x
3
− 4x
2
. Determine the intervals of monotonicity of f .
+ 2

Answer

f is strictly monotonically decreasing in the intervals (−∞, −2) and (0,1) and strictly
monotonically increasing in (−2,0) and (1, ∞).

Solution

First we calculate f ′(x).

′ 3 2
f (x) = 4x + 4x − 8x

Now we attempt to find the roots of f ′.

Clearly, 0 is a root. We factorise out x:

′ 2
f (x) = 4x(x + x − 2)

The other two roots −2 and 1 can be determined using the quadratic formula. With this we get

′
f (x) = 4x(x − 1)(x + 2)

For x ∈ (−∞, −2) or x ∈ (0,1) we have f ′(x) < 0 , and so f is strictly monotonically
decreasing there.

For x ∈ (−2,0) or x ∈ (1, ∞) we have f ′(x) > 0 , and so f is strictly monotonically increasing
there.
 108

EXERCISE 4
 109
Let f (x) = x ln x for x > 0 . Where is f monotonically increasing and where is f monotonically
decreasing? Where are the stationary points of f ?

Answer

f is monotonically decreasing in the interval (0, e−1) and monotonically increasing in the interval
(e
−1
, ∞) . There is a single stationary point which is at e−1.

Solution

We calculate f ′(x) with the product rule.

1
′
f (x) = ln x + x = ln x + 1
x

We set f ′(x) = 0 :

0 = ln x + 1

This is only true when x = e

−1
. Thus there exists only one stationary point which is at e−1.

Since ln x + 1 < 0 for x ∈ (0, e

−1
,
) f is monotonically decreasing there. In fact, f is strictly
monotonically decreasing.

Since ln x + 1 > 0 for x ∈ (e

−1
, ∞) f , is monotonically increasing there. In fact, f is strictly
monotonically increasing.
 110

6. THE SECOND DERIVATIVE

 111

Learning Goals

In this section you will learn how to find the concavity of a function with the help of the second
derivative, as well as what inflection points and saddle points of a function are and how to
calculate them.

The last section dealt with the growth of graphs with the help of the first derivative. One further aspect of
the behaviour of curves will now be investigated: concavity. In this section we will require that f is
differentiable on its domain, and that its derivative f ′ is a function defined on its domain.

The graph of the function x2 opens upwards, and the graph of the function −x2 opens
downwards.

Regions of a function that are concave up have increasing slope, in that the first derivative f ′ is
monotonically increasing. Regions of a function that are concave down have decreasing slope and
the derivative in that region is monotonically decreasing.
In many functions, the curvature of a function can change from being concave up to concave
down. Points at which the concavity changes are called inflection points, which need to be defined
precisely. In order to define an inflection point properly, we need the second derivative.
 112

6.1 DEFINITION

Let f be a differentiable function with first derivative f ′. If f ′ is differentiable at a point x0, i.e. the
limit

′ ′
f (x0 + h) − f (x0)
lim
h→0 h

exists, then (f ′)′(x0) exists as well. If the first derivative f ′ is differentiable at all of the points in
its domain, then the function f is called twice differentiable. Instead of writing (f ′)′, we write f ′′.
The second derivative of f at x0 is just the derivative of f ′:

′ ′
f (x0 + h) − f (x0)
′′
f (x0) = lim
h→0 h

6.2 EXAMPLE

1. Let f (x) = x
3
+ 2x
2
+ 5. Then f ′(x) = 3x
2
+ 4x and f ′′(x) = 6x + 4 .

2. Let f (x) = sin(x) . Then f ′(x) = cos(x) and f ′′(x) = − sin(x) .

3. Let f (x) = cos(x) . Taking the derivative we get f ′(x) = − sin(x) and f ′′(x) .
= − cos(x)

4. Let f (x) = e
2x
. By the chain rule, f ′(x) = 2e
2x
and f ′′(x) 2x
= 4e .

5. Let f (x) = ln(x) . Then f ′(x) =

1

x
and f ′′(x) = −
x
1
2
.
 113

6.3 EXAMPLE

The graph of the function f (x) = x

4
− 2x
3
+ 2x + 1 consists of a section that is concave up for
x < 0 , a concave down section for 0 < x < 1 , and a concave up section for x > 1 . The inflection
points are thus at the points x = 0 and x = 1 .

The first derivative can be calculated to be f ′(x) = 4x

3
− 6x
2
+ 2

From the image it can be seen that f ′ is increasing until the point x = 0 and then starts
decreasing. At this point the sign of f ( ) ′ ′
= f
′′
needs to change from positive to negative. At the
point x = 1 the sign of f ′′ needs to change from negative to positive.
 114

This can be recognized from the form of the second derivative:

′′
f (x) = 12x
2
− 12x = 12x(x − 1) f (x) . ′′
is positive for x < 0 and x > 1 (concave up) and
negative for 0 < x < 1 (concave down).
This observation can be used in order to cleanly define the mathematical idea of an inflection
point.

6.4 DEFINITION

Let f be twice differentiable. If f ′′(x) = 0 and the sign of f ′′ changes from one side to another at
x, then x is said to be an inflection point.

6.5 RULE

If f ′′(x) > 0 for x ∈ I then f is concave up on I . If f ′′(x) < 0 for x ∈ I then f is concave
down on I .
 115

6.6 REMARK

For a simple critierium for determining an inflection point, we need the second derivative: the
first derivative is not sufficient. The first derivative does however help in describing the inflection
point further. In Example 6.3 we can see that the type of inflection point at x = 0 and x = 1 is
different. At x = 1 the slope of the tangent is 0, and at x = 0 the slope of the tangent is 2.
An inflection point x that also satisfies f ′
(x) = 0 is called a saddle point.
Every saddle point is both an inflection point as well as a critical point.

6.7 EXAMPLE

In the previous example where f (x) = x

4 3
− 2x + 2x + 1 , it was shown that f has inflection
points at x = 0 and x = 1 . Because f ′
(x) = 4x
3
− 6x
2
+ 2, we get f ′(0) = 2 and f ′(1) = 0 . At
the point x = 1 the function has a saddle point, and at x = 0 , because f ′
(0) ≠ 0 , the function
does not have a saddle point.

Back to Chapter Overview

 116

EXERCISE 1
 117
Determine the second derivative of the function f (x) x
= e (x
2
− 4x + 5) and find where f ′′(x) is
positive, negative or 0. Where is the graph of f convex and where is it concave?

Answer

′′
f (x) = e (x
x 2
. ′′
− 1) f (x) = 0 for x = 1 and x .
= −1 f (x)
′′
is negative in the interval (−1,1)
and positive in the intervals (−∞, −1) and (1, ∞). So, the graph of f is concave in (−1,1) and
convex in (−∞, −1) and (1, ∞).

Solution

We calculate f ′(x) with the product rule.

′ x 2 x
f (x) = e (x − 4x + 5) + e (2x − 4)

x 2
= e (x − 4x + 5 + 2x − 4)

x 2
= e (x − 2x + 1)

We calculate f ′′(x) as the derivative of f ′(x) with the product rule.

′′ x 2 x
f (x) = e (x − 2x + 1) + e (2x − 2)

x 2
= e (x − 2x + 1 + 2x − 2)

x 2
= e (x − 1)

Since ex > 0 for all x ∈ R , we only need to consider the term x2 − 1.

x
2
− 1 = 0 if x = −1 or x = 1 . So, f ′′(x) = 0 for x = 1 and x = −1 .

x
2
− 1 > 0 for x ∈ (−∞, −1) or x ∈ (1, ∞) . So, f ′′(x) is positive in (−∞, −1) and (1, ∞).
Therefore, the graph of f in (−∞, −1) and (1, ∞) is convex.

x
2
− 1 < 0 for x ∈ (−1,1) . So, f ′′(x) is negative in (−1,1) and the graph of f in (−1,1) is
concave.
 118

EXERCISE 2
 119
Determine the second derivative and the inflection points of the function f (x) = 3x
5 4
− 5x + 2x + 1 .

Answer

′′
f (x) = 60x (x − 1) f
2
. has only one inflection point which is at 1.

Solution

We calculate f ′(x):

′ 4 3
f (x) = 15x − 20x + 2

We calculate f ′′(x):

′′ 3 2 2
f (x) = 60x − 60x = 60x (x − 1)

Now we determine the inflection points. Therefore, we set f ′′(x) = 0 .

f (x)
′′
has two roots 0 and 1.

Since x2 ≥ 0 for all x ∈ R and x − 1 < 0 for all x < 1 then f ′′(x) 2
= 60x (x − 1) ≤ 0 for all
x < 1 . Therefore, there is no change of sign of f ′′ at 0 and it follows that 0 is not an inflection
point of f .

Since 60x2(x − 1) < 0 for all 0 < x < 1 and 60x2(x − 1) > 0 for all x > 1 , there exists a
change of sign of f and, therefore, an inflection point of f .
′′
 120

EXERCISE 3
 121
2

Determine the second derivative and the inflection points of the function f (x) = xe
x
.

Answer

2
′′
f (x) = 2xe
x 2
(3 + 2x ) f . has only one inflection point which is at 0.

Solution

We calculate f ′(x) with the product and chain rule:

2 2
′ x 2 x
f (x) = e + 2x e

We calculate f ′′(x) as the derivative of f ′(x) with the sum, product and chain rule:

2 2 2 2
′′ x x 3 x x 2
f (x) = 2xe + 4xe + 4x e = 2xe (3 + 2x )

Now we determine the inflection points. Therefore, we set f ′′(x) = 0 .

2

Since 2ex > 0 and 3 + 2x2 > 0 for all x ∈ R 0 , is the only root of f ′′.
2

Since 2ex (3 + 2x2) is always positive, f ′′(x) has the same sign as x. Therefore, the sign of f ′′
changes at 0 and there exists an inflection point at 0.
 122

EXERCISE 4
 123
Determine the second derivative, the inflection points and the saddle points of the function
f (x) = x
4
− 4x
3
+ 1.

Answer

′′
f (x) = 12x
2
− 24x f. has inflection points at 0 and 2. Moreover, there exists a saddle point at 0.

Solution

We calculate f ′(x):

′ 3 2
f (x) = 4x − 12x

We calculate f ′′(x):

′′ 2
f (x) = 12x − 24x = 12x(x − 2)

Now we determine the inflection points. Therefore, we set f ′′(x) .

= 0

f
′′
has two roots at 0 and 2.

For x , ′′
< 0 f (x) = 12x(x − 2) is positive. For x ,
∈ (0,2) 12x(x − 2) is negative. For x ,
> 2

12x(x − 2) is positive again.

Therefore, f ′′ changes sign at 0 and 2 and these are inflection points of f .

Since f ′(0) = 0 , there also exists a saddle point at 0. Since f ′(2) = −16 2 , is not a saddle point
of f .
 124

7. EXTREMA
 125
Table of Contents
7.1 Global and Local Extrema
7.2 The Necessary Condition for Local Extrema
7.3 Sufficient Conditions for Local Extrema
7.4 Curve Sketching

Learning goals
In this section you will learn what extrema (minima/maxima) are and how to calculate the
extrema of a function with the help of the first and second derivative.

7.1 Global and Local Extrema

In applications, we are often interested in the largest possible or smallest possible values of a function.
Here, we differentiate whether the value is being compared to all of the values of the x variable (a global
extrema) or only neighbouring values of x (local extrema).

7.1 DEFINITION

Global Extrema (Global Maxima, Global Minima):

If

f (x1) ≥ f (x)

at a given point x1 for all x in the domain of f , then f is said to have a global maximum at x1.
The global maximum of f is the value of the function at that point, f (x1).

In a similar fashion, we say that f has a global minimum at x2 if

f (x2) ≤ f (x)

for all x in the domain of f . The value f (x2) is called the global minimum of f .
 126
Whether a function has a global maximum and/or minimum depends on the function. The function
2
f (x) = x , x ∈ R has a global minimum of zero at x = 0 , but does not have a global maximum. x2
reaches its global minimum at only one point. The function g(x) = cos(x), x ∈ R has a global minimum
of 1 and a global minimum of −1. The minima exist at points xl = π + l ⋅ 2π, l ∈ Z and the maxima
exist at points xk = k ⋅ 2π, k ∈ Z . Note as well that cos(x) has infinitely many minima and maxima. The
function h(x) = x, x ∈ R has no global extrema.

At local extrema, we only allow values of the variable x in the domain of f that are in a neighbourhood of
the local extrema. If f has a local extremum at x1, then we let δ > 0 be a small positive difference, and
examine the function in the region x1 − δ < x < x1 + δ .

7.2 DEFINITION

Local Extrema:
If at a given point x1 there is a δ > 0 such that for all x in the domain of f

f (x1) ≥ f (x) f or x1 − δ < x < x1 + δ,

then f has a local maximum of f (x1) at x1. Local minima can be defined in a similar fashion.
 127
Every global extremum is also a local extremum. The converse is in general not true, as the image above
illustrates.

If it is clear from the context whether local or global extrema are being referred to, the adjectives "local"
and "global" are often left out.

7.2 The Necessary Condition for Local Extrema

We require in advance that f is defined on an open interval I , (i.e. I ,

= R I = (a, b) ,
I = (a, ∞) or I = (−∞, b) ) and that f is differentiable on I .
Polynomials, sine, cosine, exponential functions, and many further functions fulfill these
requirements, and the examples given are even twice differentiable.
With the above conditions fulfilled, the derivative is a very convenient method for writing down
the necessary conditions for local extrema. For intervals with boundary points (e.g. [a, b]) see
Note at the end of the section below on the sufficient conditions for local extrema.

If f has a local maximum at x1 ∈ I , then f (x1 + h) − f (x1) ≤ 0 for all h where −δ < h < δ . From
this, it follows that for the slope of the secant to the left (h < 0) and the right (h > 0) of this point:

f (x1 + h) − f (x1)
≥ 0, h < 0, and
h

f (x1 + h) − f (x1)
≤ 0, h > 0
h

In addition:

f (x + h) − f (x) f (x + h) − f (x)
lim ≥ 0 and lim ≤ 0
h→0− h h→0+ h
 128
The limit f ′
(x1) satisfies both f ′
(x1) ≥ 0 and f ′
(x1) ≤ 0 , from which we can conclude that f ′
(x1) = 0 .
The point x1 is a critical point.
Geometrically, the secants to the left of a maximum are increasing or horizontal, and to the right of that
point are either decreasing or horizontal. Similarly, the secants to the left of a minimum are decreasing or
horizontal, and to the right of that point are either increasing or horizontal. Hence at minima and maxima
the tangent lines must be horizontal.
If a function f has an extremum at a point x, then those points are critical points of the function (see the
section on Monotonicity).

7.3 THEOREM

Necessary Condition:
Let f be differentiable on (a, b). In order for f to have a local extremum at a point x ∈ (a, b) , at
that point:

′
f (x) = 0.

7.4 EXAMPLE

1. Let f (x) = sin(x) . Because sin(x) ≤ 1 for all x ∈ R and sin( π2 ) , the point x
= 1 =
π

2
is
a local maximum. The derivative of f is f ′(x) = cos(x) . At the local extremum x =
π

2
,
f (
′ π

2
) = cos(
π

2
) = 0 .

2. If f (x) = x
3
, then f ′(x) = 3x
2
and x = 0 is a critical point, but there is no extremum
there, since f (x) > 0 for all x > 0 and f (x) < 0 for all x < 0. The critical points - the roots
of the derivative - are only candidates for points at which extrema can exist. Whether or not a
critical point is a point at which the function has an extremum, including whether it is a
minimum or maximum, needs to be investigated further.

7.3 Sufficient Conditions for Local Extrema

 129
In the section on monotonicity we saw how the sign of f influences the monotonicity of f . Assume f
′ ′

changes sign at x1 from positive to negative. Then f increases to the left of x1 and decreases to the right
of x1. It follows that f needs to have a local maximum at x1. If the derivative to the left of x2 is negative
and positive to the right of x2, then f has a minimum at x2. When doing this analysis, it suffices to only
look at neighbouring x-values, i.e. x1 − δ < x < x1 + δ for some small δ > 0 .

7.5 THEOREM

A Sufficient Condition for Local Minima and Maxima:

Let f be differentiable on an open interval I , let f ′(x0) = 0 , and let δ > 0 . If f ′(x) > 0 for
x0 − δ < x < x0 and f ′(x) < 0 for x0 < x < x0 + δ , then f has a maximum at x0.
If f ′
(x) < 0 for x0 − δ < x < x0 and f ′
(x) > 0 for x0 < x < x0 + δ then f has a minimum at
x0 .

7.6 EXAMPLE

2 2 2

Let f (x) = e
−x
, and hence f ′(x) = −2xe
−x
. Because e−x > 0 for all x ,
∈ R x = 0 is the only
root of f ′
. At x
(x) = 0 the sign of f changes from positive to negative, hence f has a maximum
′

at x .
= 0

A change in the derivative near a critical point from positive values to negative values means that f ′(x) is
strictly monotonically decreasing there. That is certainly the case when the derivative of the derivative (the
second derivative) is negative, as we saw in the previous section on monotonicity.
We now require that f is not only differentiable on an interval I , but twice differentiable. For almost all of
the example functions, this condition is satisfied.

7.7 THEOREM

A Sufficient Condition for Local Minima and Maxima:

Let f be twice differentiable on the open interval I and let f ′(x0) = 0 .
If f ′′
(x0) < 0 then f has a local maximum at x0.
If f ′′
(x0) > 0 then f has a local minimum at x0.
 130
Note

The conditions in theorems 7.5 and 7.7 are sufficient conditions for a function to have a local
maximum or minimum, i.e. the existence of local extrema follows from these requirements. These
conditions are however not necessary: there are functions that have local extrema that do not
satisfy these conditions. For example, the function f (x) = x
4
has a minimum at x = 0 , even
though f ′′
(0) = 0 is not positive.
On the other hand, it is necessary for a function to have an extremum at a given point x0 in an open
interval that the derivative of the function has a root there (Theorem 7.3). If f ′(x0) ≠ 0 , then the
function definitely has no extremum at x0. This result is not true for closed intervals!

7.8 EXAMPLE

1. Let f (x) = −2x

3
+ 3x
2
, and hence f ′(x) = −6x
2
+ 6x = −6x(x − 1) . Note that f has
critical points at x = 0 and x = 1 f. ′
changes sign at x = 0 from negative to positive, and
changes sign at x = 1 from positive to negative. According to Theorem 7.5, f has a local
minimum at x = 0 and a local maximum at x = 1.

2. Let f (x) = −2x

3
+ 3x
2
again, where f ′(x) = −6x
2
+ 6x = −6x(x − 1) with critical
points x = 0 and x = 1 . This time, Theorem 7.7 will be used to investigate the points at
which f has an extrema. Note that f ′′(x) = −12x + 6 , and with that: f ′′(0) = 6 > 0 and
′′
f (1) = −6 < 0 . As per Theorem 7.7, f has a local minimum at x = 0 and a local
maximum at x = 1 .

3. Let f (x) = 1 − (x + 2)
4
and hence f ′(x) = −4(x + 2)
3
. The only critical point is at
x = −2 . Obviously f ′ changes sign at x = −2 from positive to negative, hence f has a local
maximum at x = −2 .
Because f ′′(x) = −12(x + 2)
2
, f ′′(−2) = 0 . According to Theorem 7.7, nothing can be
said about this function's extrema based off of this information.
One way of finding the extrema of this function without using differential calculus is to
consider the values of the function near the point x = −2 f (−2) = 1. and f (x) ≤ 1 for all
x ∈ R , hence f has a global maximum at x = −2 .

The following table has a summarized list of sufficient conditions for local extrema:
 131

′
f
= changes sign at x1 f rom positive to negative ⇒ local maximum at x1

′
f
= changes sign at x2 f rom negative to positive ⇒ local minimum at x2

or alternatively

′ ′′
f (x1) = 0 and f (x1) < 0 ⇒ local maximum at x1
=

′ ′′
f (x2) = 0 and f (x2) > 0 ⇒ local minimum at x2
=
 132
Expanded Note on Extrema on Intervals with Boundary Points

Sometimes the extrema of a function f need to be determined on an interval that includes its
boundary points. How to proceed in such a case will be examined below.
Let f (x) = x
2
on the interval [−1,2]. In order to find all the local and global extrema, we first need
to find all of the local extrema. A local extremum can be at −1, 2, or on the interval (−1,2).
Because the interval (−1,2) is open, the local extrema on that interval can be found with the help
of the first derivative. From the fact that f ′(x) = 2x , the only critical point is x = 0 . Because
′′
f (x) = 2 , it follows that f ′′
(0) = 2 > 0 , and hence there is a local minimum of f at x = 0 . Now,
we need to determine whether f has local extrema at the points x = −1 and x = 2 . Because f is
strictly monotonically decreasing on [−1,0), there is a local maximum at −1. Because f is strictly
monotonically increasing on (0,2], f has a local maximum at x = 2 . Here, we can also use the
monotonicity of f to prove the existence of local extrema - because x = −1 and x = 2 are the
boundary points of the domain of f , we cannot argue using the above method.
If we now want to find what the global maximum and global minimum are, then we compare the
value of the function at each of the corresponding local extrema. This method is successful only
because on closed intervals, the global extrema will always exist; every global extremum is also a
local extremum.
The points at x = −1 and x = 2 are both local maxima; f takes on the values f (−1) = 1 and
f (2) = 4 there. Because f (2) = 4 > 1 = f (−1) f , has its global maximum of 4 at x = 2 .
Because there is only one local minimum, namely at x , this is also the point where f has its
= 0

global minimum, and the global minimum is f (0) = 0 .

If we change the original problem slightly by considering the function f (x) = x
2
on the interval
(−∞,2] , then we need to argue these points differently. The only critical point of f on (−∞,2) is
x = 0 , and there is a local minimum there. It can easily be seen that this is also the global
minimum, because f (x) is always ≥ 0, and f (0) = 0 .
At x = 2 there is a local maximum, because the function is strictly monotonically increasing on
(0,2] . This point is not the global maximum; take for example the value of the function at x = −3 :
f (−3) = 9 > 4 = f (2) . The function has no global maximum on this new domain, since
lim f (x) = ∞ .
x→−∞

7.4 Curve Sketching

When looking to sketch the graph of a function f , its important characteristics are sought after. This
includes its domain, roots, monotonicity, asymptotic behaviour, inflection points, saddle points, extrema,
etc.
In this section we restrict the discussion to a function's inflection and saddle points, as well as its extrema.
 133

7.9 EXAMPLE

Let f (x) = x
4
− 4x
3
+ 4x , and hence
2

′
f (x) = 4x
3
− 12x
2
+ 8x = 4x(x
2
− 3x + 2) = 4x(x − 1)(x − 2) f . ′
has roots x0 = 0 ,
x1 = 1 , and x2 = 2. Because f ′
(x) changes sign at all three of its roots, there are extrema at x0,
x1 and x2.
Because the sign changes from positive to negative at x0, there is a local minimum at x0 = 0 .
Analogously, we can show that f has a local maximum at x1 = 1 and a local minimum at x2 = 2 .
Here, f ′′(x) and hence the roots of f ′′ can be
2 2 2
= 12x − 24x + 8 = 12(x − 2x + )
3

calculated with the p − q formula to be x3 and x4 . For this reason, f ′′ can

1 1
= 1 + = 1 −
√3 √3

be rewritten:

1 1
′′
f (x) = 12(x − (1 + ))(x − (1 − )).
√3 √3

f
′′
changes sign at x3 and x4 and hence x3 and x4 are inflection points. Because neither x3 nor x4
are roots of f ′, neither of these inflection points can be saddle points, i.e. f has no saddle points.

Now the existence of global extrema need to be investigated. Because f (x) 2

= x (x − 2)
2
, f is
always greater than or equal to zero. Because f (x0) = f (0) = 0 and f (x2) = f (2) = 0 f , has
global minima at x0 and x2. For the local maximum at x1, because f (1) = 1 however for example
f (−1) = 9 > 1 = f (1) f , cannot have a global maximum at x1. The fact that f has no global
maximum can also be seen by taking either the limit lim f (x) = ∞ or lim f (x) = ∞ .
x→∞ x→−∞
 134

Curve sketching is a valuable tool for analyzing functions. You can use it to get a rough picture of how the
graph of a function f looks. In the following applet you can change the properties of f and watch how the
curves of f , f ′, and f ′′ relate to each other.

A Function and its First and Second Derivatives

 135
′′
f f

2
′
f

−5 −4 −3 −2 −1 1 2 3 4 5

−1

−2

−3

−4

−5

– o + ← ↓ ↑ →

To change the blue graph of the function f (x) = (x − a)(x − b)(x − c)(x − d) ,
choose the variables a, b, c, and d: (these are the roots of f !)

a : 0 , b : 0 , c : 2 , d : 2

The red curve is the graph of the derivative, f ′.

The green curve is the graph of the 2nd derivative, f ′′.
Observe:
- how the monotonic intervals and critical points of f
correspond to the sign and roots of the derivative, f ′
- the existence of critical points of f (the roots of f ′)
where f has extrema,
- how the monotonic intervals and critical points of f
correspond to the sign and roots of the second derivative, f ′′,
- how the extrema of f ′ correspond to the inflection points of f ,
- how the extrema of f ′′ correspond to the inflection points of f ′,
- how saddle points of f relate to roots of both f ′ and f ′′.
Try using three or four of the same values for a, b, c, and d!
The graph of the initial function is the graph of the function in the example above.

Back to Chapter Overview

 136

EXERCISE 1
 137
Calculate the local extrema of the function f (x) = xe
−0.5x
, using only the first derivative of f .

Answer

There exists a local maximum at x = 2 . There are no more local extrema.

Solution

First, we find the critical points. Therefore, we need to calculate f ′(x):

′ −0.5x −0.5x −0.5x

f (x) = e − 0.5xe = e (1 − 0.5x).

We set f ′(x) = 0 , so

−0.5x
e (1 − 0.5x) = 0.

Since e−0.5x ≠ 0 for all values of x, we must have 1 − 0.5x = 0 and so x = 2 . Therefore, a local
extremum can only exist at x0 := 2 . Now we must verify whether a local extremum does in fact
exist at x = 2 and, if this is the case, which type it is.

Since e−0.5x > 0 for all values of x ,

∈ R f (x)
′
has the same sign as 1 − 0.5x. Then f ′(x) > 0

for x < 2 and f ′

(x) < 0 for x > 2. Therefore, there is a local maximum at x = 2 .
 138

EXERCISE 2
 139
Calculate the local extrema of the function f (x) 2
= x e . Use the second derivative to determine
−2x

whether the stationary points are local extrema.

Answer

There exists a local maximum at x = 1 and a local minimum at x = 0.

 140
Solution

First, we find the critical points. Therefore, we calculate f ′(x):

′ −2x 2 −2x −2x

f (x) = 2xe − 2x e = 2xe (1 − x)

We set f ′(x) = 0 , so

−2x
2xe (1 − x) = 0

Since e−2x ≠ 0 for all values of x, either x = 0 or x = 1 . So, local extrema can only exist at
x0 := 0 and x1 := 1 . Now we must verify whether local extrema do in fact exist at these points
and, if so, which type they are.

The second derivative is obtained using the product rule:

′′ −2x −2x −2x −2x 2

f (x) = 2e (1 − x) − 4xe (1 − x) − 2xe = 2e (2x − 4x + 1)

We substitute x0 = 0 into f ′′ and get

′′
f (0) = 2 > 0

Therefore, there is a local minimum at x = 0 .

With x1 = 1 , we get

′′ −2
f (1) = −2e < 0

Therefore, there is a local maximum at x = 1 .

141
 142

EXERCISE 3
 143
Maximize the perimeter, U , of a rectangle that lies on the x-axis with a corner at the origin and the
opposite corner on the parabola −0.25x2 + 4. The base should be on the positive x-axis.

Answer

The maximum perimeter is U = 10 .

 144
Solution

First, we need to find the function that we want to maximize. It should be determined by the
maximum value of 2x + 2y, where x is the width of the rectangle (plotted along the x-axis) and y
is the height (plotted along the y-axis). Since y = −0.25x
2
+ 4, we can replace y in the equation
for the [Link], we can determine the maximum value of
2x + 2 ⋅ (−0.25x
2
+ 4) = −0.5x
2
+ 2x + 8 .

Of course, we should continue to apply x ≥ 0 , and since the parabola cuts the x-axis at the point
(4,0) , we can also apply x ≤ 4 . So we set f (x) = −0.5x
2
+ 2x + 8 for x ∈ [0,4] and
determine the maximum value of f .

First, we look for a maximum in the interval (0,4).

At such points, we need f ′(x) = 0. So we calculate f ′(x):

′
f (x) = −x + 2

Then f ′(2) = 0 , and there are no more roots of f ′. Therefore, in the interval (0,4), f has only the
critical point x = 2 .

Since f ′′(x) = −1 is constant and f ′′(2) = −1 < 0 , there is a local maximum at x = 2 .

Since f (0) = 8 and f (4) = 8 , but f (2) = 10 , the maximum value of f in [0,4] is at x = 2 . So
the maximum perimeter is U = 10 .
 145

EXERCISE 4
 146
Find the minimum value of the function f (x) = x
3
−
3

2
2
x + 1 in the interval [−1,2]. Where does this
occur?

Answer

The smallest value of the function is − 32 and it occurs when x = −1 .

 147
Solution

To determine the smallest value of the function, we need to find the global minimum. First, we
look for the critical points in the interval (−1,2). Therefore, we calculate f ′(x):

′ 2
f (x) = 3x − 3x = 3x(x − 1)

We set f ′(x) = 0 , and so

3x(x − 1) = 0

′
f (x) has two roots, x = 0 and x = 1 , which both lie in the interval (−1,2). So, minima can only
exist at x = 0 and x = 1 in the interval (−1,2). Now, we calculate the second derivative:

′′
f (x) = 6x − 3

We calculate f ′′(0) = −3 < 0 . Therefore, there is a local maximum at x = 0 (not a local

minimum).

Furthermore, f ′′(1) = 3 > 0 . Therefore, f has a local minimum at x = 1 . The value of the
function at that point is f (1) =
1

2
.

To verify whether this is in fact the smallest value of the function in [−1,2], we still need to
compare it with f (−1) and f (2).

On a f (−1) = −
3

2
<
1

2
, f (2) = 3 >
1

2
. Therefore, in the interval [−1,2], the smallest value of
the function f occurs at x = −1 , and the smallest value is − 32 .
 148

As you can see from the picture, the global minimum occurs at x = −1 and a local maximum
occurs at x .
= 0
 149

EXERCISE 1
 150
Let f (x) 2
= x + 2x + 3. Calculate, with the help of difference quotients, the derivative of f at x .
= 1

Give the tangent, T (x), of the graph of f at the point (1, f (1))!

Answer

The derivative at x = 1 is f ′(1) , and the tangent has the form:

= 4

T (x) = 4x + 2
 151
Solution

For h ≠ 0 , the difference quotient of f at x = 1 is the following:

f (1 + h) − f (1)

This is calculated as follows:

2 2 2 2
(1 + h) + 2(1 + h) + 3 − (1 + 2 + 3) 1 + 2h + h + 2 + 2h + 3 − 6 4h + h
= = =
h h h

As h tends towards zero, the expression tends towards 4

lim 4 + h = 4
h→0

Therefore, f ′(1) = 4 .

The tangent, T (x), has the form:

′
T (x) = f (1) ⋅ (x − 1) + f (1) = 4(x − 1) + 6 = 4x + 2
 152

EXERCISE 2
 153
Let f (x) =
1

x−2
− 2
5

x +x−6
. Determine the maximum domain, D, in R for which f is defined, and then
determine the shape of the curve of f . Use the knowledge of monotonicity and asymptotic behavior that
you have learned in this section.

Answer

The maximum domain for which f is defined is D = R ∖ {−3,2} . The graph has the form:
 154
Solution

The polynomial x2 + x − 6 has the roots x0 := 2 and x1 := −3 , so f is not defined at these

points. Since x − 2 also has the root x0 = 2 , the maximum domain, D, for which f is defined is
D = R ∖ {−3,2} .

First, we simplify f (x) by combining 1

x−2
and − x +x−6
5
2
over a common denominator:

1 5 1 5 x + 3 5
f (x) = − = − = −
2
x − 2 x + x − 6 x − 2 (x − 2)(x + 3) (x − 2)(x + 3) (x − 2)(

Now, we determine the monotonicity behavior. The first derivative of f is:

1
′
f (x) = −
2
(x + 3)

Since f ′(x) < 0 for all values of x in D, f is strictly monotonically decreasing over all intervals in
D . Hence, f is strictly monotonically decreasing over (−∞, −3), (−3,2) and (2, ∞).

Now, we examine the behavior of f as x tends towards x0 and x1. As x tends towards 2,
f (x) =
1

x+3
tends towards 1

5
, i.e.

1
lim f (x) = lim f (x) = lim f (x) =
x→2− x→2+ x→2 5

For x ,
< −3 f (x) is always negative. As we approach x = −3 from the left, the term x + 3
tends towards 0. Therefore, the reciprocal, f (x), tends towards −∞:

lim f (x) = −∞
x→−3−
 155
For x ,
> −3 f (x) is always positive. Again, x + 3 tends to 0, as we approach x = −3 from the
right. However, since x + 3 is positive, the reciprocal, f (x), tends towards +∞:

lim f (x) = ∞
x→−3+

Since the right and left-hand limits at x = −3 are different, the limit lim f (x) does not exist.
x→−3

Now, we will determine the behaviour of f (x) as x tends to +∞ and −∞. Since x + 3 increases
unbounded towards +∞ and decreases unbounded towards −∞, then f (x), as the reciprocal
of x + 3, tends towards zero in both cases. I.e.

lim f (x) = lim f (x) = 0

x→−∞ x→+∞

Overall, f has the following shape:

 156

EXERCISE 3
 157
Let f (x) x
= e (x − 1)
3
. Sketch the graph of f (x) by investigating the aspects of functions learned in this
chapter, i.e. look for local and global extrema of f , as well as inflection and saddle points.

Answer

There is a local and global minimum at x = −2 . There are no other local or global extrema. f has
inflection points at x ,
= 1 x = −2 + √ 3 , and x = −2 − √ 3 . The point x = 1 is also a saddle
point. There are no other inflection or saddle points.
 158
Solution

First of all, take the derivative of f with the help of the product rule:

′ x 3 x 2 x 2
f (x) = e (x − 1) + 3e (x − 1) = e (x + 2)(x − 1) .

Critical points occur at the roots of f ′(x). Since ex > 0 for all real x, x = −2 and x = 1 are the
only roots of f ′(x). So, local extrema can only exist at x = −2 or x = 1 .

By looking at the conditions on the first derivative, we can then investigate these critical points for
the existence of local extrema.

At x = −2 f (x) , ′
changes sign from negative to positive, hence there exists at local minimum at
x = −2 .

At x ,
= 1 f (x)
′
does not change sign: f ′(x) > 0 for −2 < x < 1 and f ′(x) > 0 for x > 1, i.e
f is strictly monotonically increasing in (−2,1) as well as (1, ∞). Therefore, there is no local
extremum at x = 1.

Since lim f (x) = ∞ , there is no point at which there is a global maximum.

x→∞

Since f ′(x) is negative in the interval (−∞, −2), this means f is decreasing in this interval. Since
f is monotonically increasing in the interval [−2, ∞), there must exist a global minimum at
x = −2 .

To determine the inflection points, we need the second derivative f ′′(x):

′′ x 2 x 2 x x 2
f (x) = e (x + 2)(x − 1) + e (x − 1) + 2e (x + 2)(x − 1) = e (x − 1)(x + 4x + 1)

′′
f (x) has the roots at x = 1 as well as x = −2 + √ 3 and −2 − √3. So, these points are the
only options for an inflection point.

Since f ′′(x) changes sign at the points x ,

= 1 x = −2 + √ 3 and −2 − √3, these points are
inflection points.
 159
A saddle point is an inflection point that is at the same time a critical point. Therefore x = 1 is
the only saddle point.
 160

EXERCISE 4
 161
A lamp hangs over the center of a round table. The diameter is one meter, so that the radius of the
tabletop is half a meter. The height of the lamp above the table will be denoted by h. The brightness, I, at
the edge of the table is then calculated as:

sin φ
I = k ⋅ ,
2
a

where k is a fixed constant.

At which height, h, should the lamp be fixed above the table in order to get the optimum illumination at
the edge of the table?

Answer

The lamp should be fixed at the height h =

1
meters above the table in order to get the
2√ 2

optimum illumination at the edge of the table.

 162
Solution

First of all, we must construct a function to maximize. I is a function depending on two variables (
φ and a), one of which we have to eliminate from I.

Since sin φ =
h

a
, we obtain:

h
I = k ⋅
3
a

Using Pythagoras' theorem, a2 = h

2
+
1

4
, and so a = √
1

4
2
+ h . Hence, we obtain:

h
I = k ⋅
3
1 2
( + h ) 2
4

So, for all h in the interval [0, ∞), we define:

h
f (h) = k ⋅
3
1 2
( + h ) 2
4

and look for the global maximum of this function.

For h = 0, we have f (0) = 0 . Obviously the global maximum does not occur at this point since f
takes positive values for all h > 0. So, we look for the global maximum in the interval (0, ∞).
Since every global maximum of f in (0, ∞) is also a local maximum, we first look for local
extrema.
 163
We calculate the derivative:

2
1 h
′
f (h) = k ⋅ − 3k ⋅
3 5
1 2 1 2
( + h ) 2 ( + h ) 2
4 4

This simplifies to the following:

1 2
− 2h
′ 4
f (h) = k ⋅
5
1 2
( + h ) 2
4

The point where the global extremum occurs, must be obtained from f ′(h) = 0 . So, we set

1 2
− 2h
′ 4
0 = f (h) = k ⋅
5
1 2
( + h ) 2
4

and see that this is only true when , or h .

1 2 1
− 2h = 0 =
4 2√ 2

We must now justify why there is in fact a local maximum when h =

1
. It is true because f ′
2√ 2

changes sign from positive to negative when h =

1
.
2√ 2

Since f ′ only changes sign (from positive to negative) at h in the interval (0, ∞), it also
1
=
2√ 2

follows that f is strictly monotonically increasing in (0, and strictly monotonically

1
)
2√ 2

decreasing in ( 1
. Therefore, the global maximum occurs at h
, ∞) =
1
.
2√ 2 2√ 2