FKP en

An Introduction to
Complex Analysis
JIŘÍ BOUCHALA AND M AREK L AMPART
Preface
Complex analysis is one of the most interesting of the fundamental topics in
the undergraduate mathematics course. Its importance to applications means that
it can be studied both from a very pure perspective and a very applied perspective.
This text book for students takes into account the varying needs and backgrounds
for students in mathematics, science, and engineering. It covers all topics likely to
feature in this course, including the subjects:
• complex numbers,
• differentiation,
• integration,
• Cauchy’s theorem and its consequences,
• Laurent and Taylor series,
• conformal maps and harmonic functions,
• the residue theorem.
Since the topics of complex analysis are not elementary subjects, there are some
reasonable assumptions about what readers should know. The reader should be fa-
miliar with relevant standard topics taught in the area of real analysis of real func-
tions of one and multiple variables, sequences, and series.
This text is mostly a translation from the Czech original [1].
The authors are grateful to their colleagues for their comments that improved
this text, to John Cawley who helped with the correction of many typos and En-
glish grammar, and also to RNDr. Alžběta Lampartová for her kind help with the
typesetting process.
prof. RNDr. Marek Lampart, Ph.D.
September 29, 2022
2
Contents
1 Complex numbers and Gauss plane 5

1.1 Complex numbers . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2 Geometric interpretation and argument of complex numbers . . . 6
1.3 Infinity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.4 Neighbourhood of a point . . . . . . . . . . . . . . . . . . . . . . 9
1.5 Sequence of complex numbers . . . . . . . . . . . . . . . . . . . 9
2 Complex functions of a real and a complex variable 11

2.1 Complex functions . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.2 Some important complex functions . . . . . . . . . . . . . . . . . 12
2.2.1 Exponential functions . . . . . . . . . . . . . . . . . . . 12
2.2.2 Trigonometric functions . . . . . . . . . . . . . . . . . . 12
2.2.3 Hyperbolic functions . . . . . . . . . . . . . . . . . . . . 13
2.2.4 Logarithmic functions . . . . . . . . . . . . . . . . . . . 14
2.2.5 Power functions . . . . . . . . . . . . . . . . . . . . . . . 15
2.2.6 𝑛-th roots . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.3 Real and imaginary parts of a function . . . . . . . . . . . . . . . 16
2.4 Limit of a function of a complex variable . . . . . . . . . . . . . 17
2.5 Continuity of a function of a complex variable . . . . . . . . . . . 18
2.6 Complex functions of a real variable and curves . . . . . . . . . . 18
3 Derivative of a complex function of a complex variable 23

3.1 Derivative of a function . . . . . . . . . . . . . . . . . . . . . . . 23
3.2 Harmonic functions, harmonic conjugate functions . . . . . . . . 25
3.3 Remark about geometric meaning of derivatives . . . . . . . . . . 28
4 Conformal function 29
4.1 Basic properties . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
4.2 Linear fractional functions . . . . . . . . . . . . . . . . . . . . . 30
5 Complex functions integral, Cauchy’s integral theorem and formulas 33

5.1 Integral of complex functions of a complex and real variable . . . 33
5.2 Cauchy’s integral theorem . . . . . . . . . . . . . . . . . . . . . 35
5.3 Cauchy’s integral formula . . . . . . . . . . . . . . . . . . . . . . 36
3
4 CONTENTS
5.4 Primitive functions and path independent integral . . . . . . . . . 38
6 Number series, sequences and function series 43

6.1 Number series . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
6.2 Sequence of functions, pointwise and uniform convergence . . . . 46
7 Power and Taylor series 49

7.1 Power series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
7.2 Taylor series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
8 Laurent series and classification of singularities 59

8.1 Laurent series . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
8.2 Isolated singularities and their classification . . . . . . . . . . . . 63
8.3 Laurent series with a centre ∞, classification of a point ∞ . . . . 64
9 Residue and the residue theorem 67

9.1 Residue of a function and it’s calculation . . . . . . . . . . . . . . 67
9.2 The residue theorem . . . . . . . . . . . . . . . . . . . . . . . . 69
9.3 Calculation integrals of real variable function with the aid of the
residue theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
10 Exercise 73
Chapter 1
Complex numbers and Gauss

plane
1.1 Complex numbers

This chapter is devoted to the fundamental structure of complex numbers and their
basic properties.
(i) A complex number 𝑧 is a number
𝑧 = 𝑥 + 𝑖𝑦 where 𝑥, 𝑦 ∈ ℝ and 𝑖2 = −1,
the numbers 𝑥 and 𝑦 are called the real and imaginary parts of the complex
number 𝑧 respectively, and are denoted by Re 𝑧 and Im 𝑧 respectively.1
(ii) It is easy to see that real and imaginary numbers are the special case of com-
plex numbers. Real numbers are those 𝑧 with Im 𝑧 = 0, and the imaginary
ones are characterized by the condition Re 𝑧 = 0.
(iii) The two complex numbers 𝑧1 and 𝑧2 are equal if and only if their real and
imaginary parts are equal, that is
[ ]
𝑧1 = 𝑧2 ⇔ Re 𝑧1 = Re 𝑧2 ∧ Im 𝑧1 = Im 𝑧2 .
(iv) Let 𝑧 = 𝑥+𝑖𝑦 be a complex number, then its absolute value is a non-negative
(real) number √ √
|𝑧| = 𝑥2 + 𝑦2 = (Re 𝑧)2 + (Im 𝑧)2
and the number complex adjoint is given by
𝑧 = 𝑥 − 𝑖𝑦 = Re 𝑧 − 𝑖 Im 𝑧.
1
Notation: by 𝑧 = 𝑥 + 𝑖𝑦 it is meant (unless otherwise stated) that 𝑥 = Re 𝑧 ∈ ℝ and 𝑦 = Im 𝑧 ∈
ℝ.
5
6 CHAPTER 1. COMPLEX NUMBERS AND GAUSS PLANE
(v) Let 𝑧1 = 𝑥1 + 𝑖𝑦1 and 𝑧2 = 𝑥2 + 𝑖𝑦2 be two complex numbers. Then it is

defined
𝑧1 + 𝑧2 = (𝑥1 + 𝑥2 ) + 𝑖(𝑦1 + 𝑦2 ),
𝑧1 − 𝑧2 = (𝑥1 − 𝑥2 ) + 𝑖(𝑦1 − 𝑦2 ),
𝑧1 𝑧2 = (𝑥1 𝑥2 − 𝑦1 𝑦2 ) + 𝑖(𝑥1 𝑦2 + 𝑥2 𝑦1 ).
Moreover, if 𝑧2 ≠ 0 = 0 + 0𝑖, it is also defined

𝑧1 1
= (𝑧1 𝑧2 ).
𝑧2 |𝑧2 |2
(vi) For every complex number 𝑧 = 𝑥 + 𝑖𝑦 it holds that
𝑧𝑧 = (𝑥 + 𝑖𝑦)(𝑥 − 𝑖𝑦) = 𝑥2 − (𝑖𝑦)2 = 𝑥2 + 𝑦2 = |𝑧|2 .
Remark 1.1 The most remarkable difference between real and complex numbers
is the fact that complex numbers are not ordered. The relation 𝑧1 < 𝑧2 between
complex numbers 𝑧1 and 𝑧2 is not defined unless they are both real.
Example 1.1 Find Re 𝑧0 and Im 𝑧0 for

2 + 3𝑖
𝑧0 = .
1 − 2𝑖
Solution:
2 + 3𝑖 1 + 2𝑖 −4 + 7𝑖 4 7
𝑧0 = ⋅ = = − + 𝑖,
1 − 2𝑖 1 + 2𝑖 5 5 5
hence
4 7
Re 𝑧0 = − and Im 𝑧0 = .
5 5
1.2 Geometric interpretation and argument of complex num-

bers
Given a complex number 𝑧 = 𝑥 + 𝑖𝑦, its real and imaginary parts uniquely define
an element (𝑥, 𝑦) ∈ ℝ2 . The set of all complex numbers is therefore naturally
identified with the plane ℝ2 ; it is called a Gauss plane and is denoted by ℂ.
Points in the plane ℝ2 can also be represented using polar coordinates, which
is a representation of the complex numbers. Let 𝑧 ∈ ℂ, 𝑧 ≠ 0, then there exists
𝜑 ∈ ℝ such that
𝑧 = |𝑧|(cos 𝜑 + 𝑖 sin 𝜑). (1.1)
We call (1.1) a polar form of a complex number 𝑧.
Since the functions sine and cosine are periodic, the number (angle) 𝜑 is not
uniquely determined by (1.1).
1.2. INTERPRETATION AND ARGUMENT 7
Definition 1.1 The set of all real numbers 𝜑 for which (1.1) holds is called an
argument of a complex number 𝑧 ∈ ℂ ⧵ {0} and is denoted by Arg 𝑧. Accordingly,
Arg 𝑧 = {𝜑 ∈ ℝ ∶ 𝑧 = |𝑧|(cos 𝜑 + 𝑖 sin 𝜑)}.
Remark 1.2 If 𝑧 = 0, then |𝑧| = 0, and (1.1) holds for arbitrary 𝜑 ∈ ℝ. This is
the reason why an argument of 0 is not defined.
Theorem 1.1 Let 𝑧 ∈ ℂ ⧵ {0} and 𝜑 ∈ Arg 𝑧. Then
Arg 𝑧 = {𝜑 + 2𝑘𝜋 ∶ 𝑘 ∈ ℤ}.
Proof Let 𝜑 ∈ Arg 𝑧. Obviously, sine and cosine are periodic functions, and there-
fore
{𝜑 + 2𝑘𝜋 ∶ 𝑘 ∈ ℤ} ⊂ Arg 𝑧.
On the other hand, let 𝜓 ∈ Arg 𝑧. We want to find 𝑘 ∈ ℤ such that 𝜓 = 𝜑 + 2𝑘𝜋.
𝜑, 𝜓 ∈ Arg 𝑧 ⇒
⇒ [𝑧 = |𝑧|(cos 𝜑 + 𝑖 sin 𝜑) = |𝑧|(cos 𝜓 + 𝑖 sin 𝜓) ∧ |𝑧| ≠ 0] ⇒
⎡ ⎤
⎢cos 𝜑 = cos 𝜓 ⎥
⎢ ⎥
⇒ cos 𝜑 + 𝑖 sin 𝜑 = cos 𝜓 + 𝑖 sin 𝜓 ⇒ ⎢ ∧ ⎥⇒
⎢ ⎥
⎢ sin 𝜑 = sin 𝜓 ⎥
⎣ ⎦
⎡ 2 ⎤
⎢cos 𝜑 = cos 𝜓 cos 𝜑⎥
⎢ ⎥ 2 2
⇒⎢ ∧ ⎥ ⇒ cos 𝜑 + sin 𝜑 = cos 𝜓 cos 𝜑 + sin 𝜓 sin 𝜑 ⇒
⎢ ⎥
⎢ sin2 𝜑 = sin 𝜓 sin 𝜑 ⎥
⎣ ⎦
⇒ 1 = cos(𝜓 − 𝜑) ⇒ [∃𝑘 ∈ ℤ ∶ 𝜓 − 𝜑 = 2𝑘𝜋] ⇒ [∃𝑘 ∈ ℤ ∶ 𝜓 = 𝜑 + 2𝑘𝜋] .
Definition 1.2 The argument 𝜑 ∈ Arg 𝑧 such that
−𝜋 < 𝜑 ≤ 𝜋
is called a principal value of an argument of a complex number 𝑧 ∈ ℂ ⧵ {0} and

we denote it arg 𝑧.
√
Example 1.2 Find Arg 𝑧0 and arg 𝑧0 for 𝑧0 = − 3 − 𝑖.
Solution: Obviously2 ,
1 𝜋 7𝜋
𝜋 + arcsin =𝜋+ = ∈ Arg 𝑧0 ,
2 6 6
therefore3 { }
7𝜋 5𝜋
Arg 𝑧0 = + 2𝑘𝜋 ∶ 𝑘 ∈ ℤ , arg 𝑧0 = − .
6 6
1.3 Infinity
As we enrich real numbers by the special symbols +∞ and −∞, we will also in-
troduce analogous symbols into the Gauss plane ℂ. It is more efficient to add only
one such symbol, ∞, infinity.
Now, the following is devoted to the another geometrical interpretation of com-
plex numbers, the so called stereographical projection, which clearly explains the
meaning of ∞.
Let us assume a unit sphere is located in such a way that it touches with its
"south pole" the plane of complex numbers at the point 0, and denote the "north
pole" by 𝑁. Now, join every complex number 𝑧 with 𝑧∗ ≠ 𝑁 belonging to the
unit sphere in such a way that 𝑧∗ is just an intersection of this sphere with a straight
line joining 𝑧 with 𝑁. In this way, we have generated one-to-one correspondence
between the set of complex numbers (zero corresponds to the "south pole") and
points of the unit sphere (without 𝑁).
We observe that the size of |𝑧| is inversely proportional to the distance between
𝑧 and 𝑁. This compels us to add only one special symbol, ∞, to ℂ and link it with
∗
the properties described by the stereographical projection of point 𝑁.

The set
ℂ ∪ {∞} = ℂ∞
is called an extended (sometimes also closed) Gauss plane.
For each 𝑧 ∈ ℂ we define:
1. 𝑧 ± ∞ = ∞ ± 𝑧 = ∞,
2. 𝑧 ⋅ ∞ = ∞ ⋅ 𝑧 = ∞ for 𝑧 ≠ 0,
𝑧
3. ∞
= 0,
𝑧
4. 0
= ∞ for 𝑧 ≠ 0,
∞
5. 𝑧
= ∞,
6. ∞𝑛 = ∞, ∞−𝑛 = 0, 0−𝑛 = ∞ if 𝑛 ∈ ℕ,
2
Try to draw a picture!
3
See Theorem 1.1 and Definition 1.2.
1.4. NEIGHBOURHOOD OF A POINT 9
7. |∞| = ∞, ∞ = ∞. 4
1.4 Neighbourhood of a point

Definition 1.3 By the neighbourhood of a point 𝑧0 ∈ ℂ with radius 𝜀 ∈ ℝ+ we
mean the set
𝑈 (𝑧0 , 𝜀) = {𝑧 ∈ ℂ ∶ |𝑧 − 𝑧0 | < 𝜀}.
By the neighbourhood of ∞ with radius 𝜀 ∈ ℝ+ we mean the set

{ }
1
𝑈 (∞, 𝜀) = 𝑧 ∈ ℂ ∶ |𝑧| > ∪ {∞}.
𝜀
By the ring neighbourhood of a point 𝑧 ∈ ℂ∞ with radius 𝜀 ∈ ℝ+ we mean the

set
𝑃 (𝑧, 𝜀) = 𝑈 (𝑧, 𝜀) ⧵ {𝑧}.
In case we do not care about the value of 𝜀, the neighbourhood 𝑈 (𝑧) or the ring
neighbourhood 𝑃 (𝑧) of a point 𝑧 is used.
Definition 1.4 We call the set 𝑀 ⊂ ℂ∞ open if and only if for all 𝑧 ∈ 𝑀 there
exists a neighbourhood 𝑈 (𝑧) such that
𝑈 (𝑧) ⊂ 𝑀.
Example 1.3
1. ∅, ℂ a ℂ∞ are open sets.
2. {𝑧 ∈ ℂ ∶ |𝑧 − 3| < |𝑧 + 2 − 𝑖|} and {𝑧 ∈ ℂ ∶ Im 𝑧 < 1} are open sets.

√
3. {2 + 3𝑖}, {𝑧 ∈ ℂ ∶ Re 𝑧 + 2 Im 𝑧 = 7}, and {𝑧 ∈ ℂ ∶ Im 𝑧 ≤ 1} are not
open sets.
1.5 Sequence of complex numbers

Definition 1.5 By a sequence in ℂ∞ we mean a function 𝑓 ∶ ℕ → ℂ∞ .
Definition 1.6 Let 𝑧 ∈ ℂ∞ and (𝑧𝑛 ) be a sequence in ℂ∞ . We say that a sequence

(𝑧𝑛 ) has a limit 𝑧 if for all 𝜀 ∈ ℝ+ there exists 𝑛0 ∈ ℕ such that for all 𝑛 ∈ ℕ, 𝑛 ≥ 𝑛0
it holds that 𝑧𝑛 ∈ 𝑈 (𝑧, 𝜀). We denote it lim 𝑧𝑛 = 𝑧 or 𝑧𝑛 → 𝑧. We call a sequence
(𝑧𝑛 ) of complex numbers convergent if there exists 𝑧 ∈ ℂ such that
lim 𝑧𝑛 = 𝑧.
4
Notice that the following operations are not defined: ∞ ± ∞, 0 ⋅ ∞, ∞ ⋅ 0, 00 , ∞
∞
, Arg ∞, arg ∞.
Remark 1.3
1. The definition of a limit in fact means that outside of any neighbourhood of

a point 𝑧 there is at most finite number of terms of sequence (𝑧𝑛 ).
2. Let (𝑧𝑛 ) be a sequence of complex numbers in the Gauss plane and 𝑧 ∈ ℂ∞ .

Denote by (𝑧∗𝑛 ) an image of this sequence by stereographic projection and 𝑧∗
a point on a unit sphere in ℝ3 . Then
𝑧𝑛 → 𝑧 (in ℂ∞ ) ⇔ 𝑧∗𝑛 → 𝑧∗ (in ℝ3 ).
Theorem 1.2 Let 𝑧𝑛 = 𝑥𝑛 + 𝑖𝑦𝑛 for all sufficiently large 𝑛 ∈ ℕ and 𝑧 = 𝑥 + 𝑖𝑦.
Then [ ]
lim 𝑧𝑛 = 𝑧 ⇔ lim 𝑥𝑛 = 𝑥 ∧ lim 𝑦𝑛 = 𝑦 .
Example 1.4 Find

(2𝑛 − 𝑖)𝑖
lim .
𝑛
Solution:
(2𝑛 − 𝑖)𝑖 ( )
1 1
lim = lim + 2𝑖 = lim + 𝑖 lim 2 = 0 + 2𝑖 = 2𝑖.
𝑛 𝑛 𝑛
Remark 1.4 The definition of a limit of a complex sequence is equivalent to the
definition of a limit of a real sequence. Therefore many theorems are analogous. In
what follows, we introduce only several of them.
Theorem 1.3 Any sequence of complex numbers has at most one limit.
Theorem 1.4 A sequence of complex numbers has a limit 𝑧 ∈ ℂ∞ if and only if

each subsequence also has a limit 𝑧.
Theorem 1.5 Let (𝑧𝑛 ) be a convergent sequence such that for all 𝑛 ∈ ℕ it holds
that 𝑧𝑛 ∈ ℂ. Then the sequence (𝑧𝑛 ) is bounded, which means that there exists
𝑘 ∈ ℝ+ such that for all 𝑛 ∈ ℕ it is |𝑧𝑛 | ≤ 𝑘.
Chapter 2
Complex functions of a real and a

complex variable
2.1 Complex functions

Definition 2.1 We call a complex function of a complex variable a function from
ℂ∞ to the set of all nonempty subsets of ℂ∞ . In other words, by a complex function
𝑓 we mean a formula which assigns to each 𝑧 ∈ 𝐷𝑓 ⊂ ℂ∞ one or more complex
numbers from ℂ∞ . The set 𝐷𝑓 is called a domain of a function 𝑓 . A complex num-
ber 𝑓 (𝑧) is called an 𝑓 – image of a point 𝑧. A function 𝑓 is called single-valued if
for all 𝑧 ∈ 𝐷𝑓 a set 𝑓 (𝑧) contains only one point. Otherwise, we call a function 𝑓
multi-valued, or accordingly double-valued, triple-valued, . . . , infinite-valued.
A function 𝑓 is called a complex function of a real variable if 𝐷𝑓 ⊂ ℝ .
Remark 2.1 A domain of a function defined only by a formula is a set of all num-
bers from ℂ∞ for which the formula makes sense. For example, the domain of the
function 𝑓 defined as 𝑓 (𝑧) = 1∕𝑧 is the set 𝐷𝑓 = ℂ∞ .
Example 2.1
1. 𝑓 (𝑧) = 𝑧2 is a single-valued function, 𝐷𝑓 = ℂ∞ .
2. 𝑓 (𝑧) = Arg 𝑧 is an infinite-valued function, 𝐷𝑓 = ℂ ⧵ {0}.
Remark 2.2 In the following, we use the notation Arg 𝑧 = arg 𝑧 + 2𝑘𝜋, 𝑘 ∈ ℤ,
instead of the correct notation Arg 𝑧 = {arg 𝑧 + 2𝑘𝜋 ∶ 𝑘 ∈ ℤ}.
Definition 2.2 Let 𝑓 be a multi-valued function. A single-valued function 𝜑 is

called a unique-branch of a function 𝑓 if:
(i) 𝐷𝜑 ⊂ 𝐷𝑓 ,
(ii) ∀𝑧 ∈ 𝐷𝜑 ∶ 𝜑(𝑧) ∈ 𝑓 (𝑧).
11
12 CHAPTER 2. COMPLEX FUNCTIONS
Example 2.2 Functions

𝜑1 (𝑧) = arg 𝑧,
𝜑2 (𝑧) = arg 𝑧 + 2𝜋
are a case of two different unique-branches of a function 𝑓 (𝑧) = Arg 𝑧.
2.2 Some important complex functions

2.2.1 Exponential functions
Definition 2.3 Let 𝑧 = 𝑥 + 𝑖𝑦 ∈ ℂ. We define an exponential function by 1
𝑒𝑧 = 𝑒𝑥+𝑖𝑦 = 𝑒𝑥 (cos 𝑦 + 𝑖 sin 𝑦).
Theorem 2.1 (Properties of exponential functions)
(i) 𝑒𝑧 is a single-valued function.
(ii) A codomain of 𝑒𝑧 is a set ℂ ⧵ {0}.
(iii) Function 𝑒𝑧 is periodic with a period of 2𝜋𝑖.
Proof Propositions (i) and (ii) follow from the definition and properties of the real
functions 𝑒𝑥 , sin 𝑥, cos 𝑥. Periodicity (iii) can be proved as follows:
𝑒𝑧+2𝜋𝑖 = 𝑒𝑥+𝑖𝑦+2𝜋𝑖 = 𝑒𝑥 (cos(𝑦 + 2𝜋) + 𝑖 sin(𝑦 + 2𝜋)) =
= 𝑒𝑥 (cos 𝑦 + 𝑖 sin 𝑦) = 𝑒𝑥+𝑖𝑦 = 𝑒𝑧 .
2.2.2 Trigonometric functions

Definition 2.4 Let 𝑧 = 𝑥 + 𝑖𝑦 ∈ ℂ. We define trigonometric functions as
𝑒𝑖𝑧 − 𝑒−𝑖𝑧
sin 𝑧 = ,
2𝑖
𝑒𝑖𝑧 + 𝑒−𝑖𝑧
cos 𝑧 = ,
2
sin 𝑧
tan 𝑧 = ,
cos 𝑧
cos 𝑧
cot 𝑧 = .
sin 𝑧
1
We use a notation 𝑒 for two different functions:
𝑒𝑧 ∶ ℂ → ℂ ⧵ {0} and 𝑒𝑥 ∶ ℝ → ℝ+ .
That is alright because for 𝑧 = 𝑥 + 0𝑖 = 𝑥 is
𝑒𝑧 = 𝑒𝑥+0𝑖 = 𝑒𝑥 (cos 0 + 𝑖 sin 0) = 𝑒𝑥 .
In other words, complex exponential function is an extension of real exponential function on ℂ. Sim-
ilarly for sin, cos, sinh, cosh, ln, ... .
2.2. SOME IMPORTANT COMPLEX FUNCTIONS 13
Theorem 2.2 (Properties of trigonometric functions)

(i) Each of trigonometric functions is single-valued.
(ii) The functions sin 𝑧 and cos 𝑧 are periodic with a period of 2𝜋,
the functions tan 𝑧 and cot 𝑧 are periodic with a period of 𝜋.
(iii) For all 𝑧 ∈ ℂ it holds that:

sin(−𝑧) = − sin 𝑧,
cos(−𝑧) = cos 𝑧,
tan(−𝑧) = − tan 𝑧,
cot(−𝑧) = − cot 𝑧.
(iv) For any 𝑧 ∈ ℂ Euler’s formula states that

𝑒𝑖𝑧 = cos 𝑧 + 𝑖 sin 𝑧.
(v)
sin 𝑧 = 0 ⇔ [∃𝑘 ∈ ℤ ∶ 𝑧 = 𝑘𝜋] ,
[ ]
𝜋
cos 𝑧 = 0 ⇔ ∃𝑘 ∈ ℤ ∶ 𝑧 = + 𝑘𝜋 .
2
Example 2.3 Let 𝑧0 = cos(4 + 𝑖). Find Re 𝑧0 and Im 𝑧0 .
Solution:
𝑒𝑖(4+𝑖) + 𝑒−𝑖(4+𝑖)
𝑧0 = cos(4 + 𝑖) = =
2
𝑒−1 (cos 4 + 𝑖 sin 4) + 𝑒 (cos(−4) + 𝑖 sin(−4))
= =
2
𝑒−1 + 𝑒 𝑒−1 − 𝑒
= cos 4 + 𝑖 sin 4.
2 2
Therefore
Re 𝑧0 = cosh 1 cos 4, Im 𝑧0 = − sinh 1 sin 4.
2.2.3 Hyperbolic functions

Definition 2.5 We define hyperbolic functions as
𝑒𝑧 − 𝑒−𝑧
sinh 𝑧 = ,
2
𝑧
𝑒 +𝑒 −𝑧
cosh 𝑧 = ,
2
sinh 𝑧
tanh 𝑧 = ,
cosh 𝑧
cosh 𝑧
coth 𝑧 = .
sinh 𝑧
Remark 2.3 Similarly as for real functions, we can define an inverse function as a
complex function. In contrast with real functions, we define inverse functions also
for functions which are not injective. In such cases the inverse function is multi-
valued like, e.g., logarithmic function.
2.2.4 Logarithmic functions

Definition 2.6 We define a logarithmic function as
Ln 𝑧 = {𝑤 ∈ ℂ ∶ 𝑒𝑤 = 𝑧}.
From properties of exponential functions (see Theorem 2.1) it follows that the do-
main of Ln 𝑧 is a set ℂ ⧵ {0}.
Let
𝑧 = |𝑧|(cos 𝜑 + 𝑖 sin 𝜑),
where |𝑧| > 0 and 𝜑 ∈ ℝ.
Ln 𝑧 = 𝑢 + 𝑖𝑣.
Then
𝑒𝑢+𝑖𝑣 = 𝑧,
that is
𝑒𝑢 (cos 𝑣 + 𝑖 sin 𝑣) = |𝑧| (cos 𝜑 + 𝑖 sin 𝜑) .
Therefore 2
𝑢 = ln |𝑧| ∧ [∃𝑘 ∈ ℤ ∶ 𝑣 = 𝜑 + 2𝑘𝜋] .
We proved that for all 𝑧 ∈ ℂ ⧵ {0} it holds that
Ln 𝑧 = ln |𝑧| + 𝑖(𝜑 + 2𝑘𝜋), 𝑘 ∈ ℤ,
or equivalently
Ln 𝑧 = ln |𝑧| + 𝑖 Arg 𝑧.
Example 2.4 Let 𝑧0 = Ln(−1 + 𝑖). Find Re 𝑧0 and Im 𝑧0 .

Solution:
√ ( )
3𝜋
𝑧0 = Ln(−1 + 𝑖) = ln 2 + + 2𝑘𝜋 𝑖, 𝑘 ∈ ℤ.
4
√ ( )
3
Therefore Re 𝑧0 = ln 2 and Im 𝑧0 = + 2𝑘 𝜋𝑖, 𝑘 ∈ ℤ.
4
Definition 2.7 We define a principal branch of a logarithm as a function on ℂ ⧵
{0} by
ln 𝑧 = ln |𝑧| + 𝑖 arg 𝑧.
2
The symbol ln denotes the natural logarithm, which is the function from ℝ+ to ℝ.
2.2. SOME IMPORTANT COMPLEX FUNCTIONS 15
Example 2.5 Let 𝑧0 = ln(−1 − 𝑖). Find Re 𝑧0 and Im 𝑧0 .

Solution:
√ 3𝜋
𝑧0 = ln(−1 − 𝑖) = ln 2− 𝑖.
4
Therefore
√ 3𝜋
Re 𝑧0 = ln 2, Im 𝑧0 = − .
4
2.2.5 Power functions

Recall that for any 𝑛 ∈ ℕ (resp. −𝑛 ∈ ℕ) is a function 𝑧 → 𝑧𝑛 , defined as
1
𝑧𝑛 = 𝑧𝑧𝑧 ⋯ 𝑧, (resp. 𝑧𝑛 = −𝑛 ).
⏟⏟⏟ 𝑧
𝑛-times
Definition 2.8 Let 𝑎 ∈ ℂ such that ±𝑎 ∉ ℕ. A power function is defined by
𝑧𝑎 = {𝑒𝑎𝑠 ∶ 𝑠 ∈ Ln 𝑧} = 𝑒𝑎 Ln 𝑧 .
Example 2.6 Find the real and imaginary parts of 2𝑖 .

Solution:
2𝑖 = 𝑒𝑖 Ln 2 = 𝑒𝑖(ln 2+2𝑘𝜋𝑖) = 𝑒−2𝑘𝜋+𝑖 ln 2 = 𝑒−2𝑘𝜋 (cos(ln 2) + 𝑖 sin(ln 2)) , 𝑘 ∈ ℤ.
Hence,
Re(2𝑖 ) = 𝑒−2𝑘𝜋 cos(ln 2) and Im(2𝑖 ) = 𝑒−2𝑘𝜋 sin(ln 2),
where 𝑘 ∈ ℤ.
2.2.6 𝑛-th roots

Definition 2.9 Let 𝑛 ∈ ℕ, 𝑛 ≠ 1. We define a function 𝑛-th root as
√𝑛
𝑧 = {𝑤 ∈ ℂ ∶ 𝑤𝑛 = 𝑧}.
Exercise 2.1
1. Prove that for all 𝑧 ∈ ℂ ⧵ {0} and 1 < 𝑛 ∈ ℕ it holds that

√
𝑛
1
𝑧 = 𝑧𝑛
1
and that a function 𝑧 → 𝑧 𝑛 is 𝑛-valued.
2. Prove that for 𝑎 = 𝑚∕𝑛, where 𝑚 ∈ ℤ ⧵ {0} and 𝑛 ∈ ℕ are not divisible
numbers, the function 𝑧 → 𝑧𝑎 is 𝑛-valued.
3. Prove that for 𝑎 ∈ ℂ ⧵ ℚ, the function 𝑧 → 𝑧𝑎 is infinite-valued.

√
Example 2.7 Find the real and imaginary parts of 4 𝑖.
Solution:
√ 1 1 1 𝜋 𝜋 𝜋
𝑖 = 𝑖 4 = 𝑒 4 Ln 𝑖 = 𝑒 4 ( 2 𝑖+2𝑘𝜋𝑖) = 𝑒 8 𝑖+𝑘 2 𝑖 =
4
( ) ( )
𝜋 𝜋 𝜋 𝜋
= cos +𝑘 + 𝑖 sin +𝑘 , 𝑘 ∈ {0, 1, 2, 3}.
8 2 8 2
Hence,
(√ ) ( ) (√ ) ( )
4 𝜋 𝜋 4 𝜋 𝜋
Re 𝑖 = cos +𝑘 and Im 𝑖 = sin +𝑘
8 2 8 2
where 𝑘 ∈ {0, 1, 2, 3}.
2.3 Real and imaginary parts of a function

In what follows, any complex function is meant to be a single-valued one.
Definition 2.10 Let 𝑓 ∶ ℂ → ℂ. Function
𝑢 ∶ ℝ2 → ℝ (resp. 𝑣 ∶ ℝ2 → ℝ)
defined on the set

{(𝑥, 𝑦) ∈ ℝ2 ∶ 𝑥 + 𝑖𝑦 ∈ 𝐷𝑓 }
by
𝑢(𝑥, 𝑦) = Re 𝑓 (𝑥 + 𝑖𝑦) (resp. 𝑣(𝑥, 𝑦) = Im 𝑓 (𝑥 + 𝑖𝑦))

be called the real part of a function 𝑓 (resp. imaginary part of a function 𝑓 ).
We denote function 𝑓 as
𝑓 = 𝑢 + 𝑖𝑣,
where 𝑢 is a real and 𝑣 an imaginary part of 𝑓 .
Example 2.8 Find real and imaginary parts of the function

𝑧
𝑓 (𝑧) = .
𝑧
Solution: Firstly,
𝑥 + 𝑖𝑦 𝑥2 − 𝑦2 2𝑥𝑦
𝑓 (𝑧) = 𝑓 (𝑥 + 𝑖𝑦) = = +𝑖 2 .
𝑥 − 𝑖𝑦 𝑥2 + 𝑦2 𝑥 + 𝑦2
Hence
𝑥2 − 𝑦2
Re(𝑓 ) = 𝑢(𝑥, 𝑦) = ,
𝑥2 + 𝑦2
2𝑥𝑦
Im(𝑓 ) = 𝑣(𝑥, 𝑦) = 2 .
𝑥 + 𝑦2
2.4. LIMIT OF A FUNCTION OF A COMPLEX VARIABLE 17
2.4 Limit of a function of a complex variable

Remark 2.4 By
𝑧0 ≠ 𝑧𝑛 → 𝑧0 ,
we mean that 𝑧𝑛 → 𝑧0 and that for all sufficiently large 𝑛 ∈ ℕ it is 𝑧𝑛 ∈ ℂ∞ ⧵ {𝑧0 }.
Definition 2.11 A function 𝑓 ∶ ℂ∞ → ℂ∞ has at the point 𝑧0 ∈ ℂ∞ a limit

𝑎 ∈ ℂ∞ if for each sequence (𝑧𝑛 ) such that 𝑧0 ≠ 𝑧𝑛 → 𝑧0 it holds 𝑓 (𝑧𝑛 ) → 𝑎.
It is denoted by
lim 𝑓 (𝑧) = 𝑎.
𝑧→𝑧0
Theorem 2.3 Let 𝑓 ∶ ℂ∞ → ℂ∞ and 𝑧0 , 𝑎 ∈ ℂ∞ . Then lim 𝑓 (𝑧) = 𝑎 if and only

𝑧→𝑧0
if ( )( )
(∀𝑈 (𝑎)) ∃𝑃 (𝑧0 ) ∀𝑧 ∈ 𝑃 (𝑧0 ) ∶ 𝑓 (𝑧) ∈ 𝑈 (𝑎).
Theorem 2.4 Let 𝑓 = 𝑢 + 𝑖𝑣 ∶ ℂ → ℂ, 𝑧0 = 𝑥0 + 𝑖𝑦0 , and 𝑎 = 𝛼 + 𝑖𝛽. Then

lim 𝑓 (𝑧) = 𝑎 if and only if
𝑧→𝑧0
lim 𝑢(𝑥, 𝑦) = 𝛼 ∧ lim 𝑣(𝑥, 𝑦) = 𝛽.

(𝑥,𝑦)→(𝑥0 ,𝑦0 ) (𝑥,𝑦)→(𝑥0 ,𝑦0 )
Example 2.9 Calculate ( )

𝑧−𝑖
lim .
𝑧→𝑖 𝑧2 + 1
Solution:
( ) ( ) ( )
𝑧−𝑖 1 1
lim = lim = lim =
𝑧→𝑖 𝑧2 + 1 𝑧→𝑖 𝑧 + 𝑖 𝑥+𝑖𝑦→𝑖 𝑥 + 𝑖(𝑦 + 1)
( )
𝑥 −(𝑦 + 1)
= lim +𝑖 2 =
𝑥+𝑖𝑦→𝑖 𝑥2 + (𝑦 + 1)2 𝑥 + (𝑦 + 1)2
( ) ( )
𝑥 −(𝑦 + 1)
= lim + 𝑖 lim =
(𝑥,𝑦)→(0,1) 𝑥2 + (𝑦 + 1)2 (𝑥,𝑦)→(0,1) 𝑥2 + (𝑦 + 1)2
1 1
= 0 − 𝑖 = − 𝑖.
2 2
Example 2.10 Calculate
lim arg 𝑧.
𝑧→−1
Solution: The latter limit doesn’t exist because
( 𝑛
) ( )
(−1)𝑛
1. −1 ≠ 𝑧𝑛 = cos 𝜋 + (−1) 𝑛
+ 𝑖 sin 𝜋 + 𝑛
→ −1,
( )
2. arg 𝑧2𝑛 → −𝜋,
( )
3. arg 𝑧2𝑛+1 → 𝜋.
2.5 Continuity of a function of a complex variable

Definition 2.12 Let 𝑓 ∶ ℂ∞ → ℂ∞ . The function 𝑓 is continuous at the point
𝑧0 ∈ ℂ∞ if
lim 𝑓 (𝑧) = 𝑓 (𝑧0 ).
𝑧→𝑧0
We say that a function 𝑓 is continuous on the set 𝑀 ⊂ ℂ∞ if for all 𝑧0 ∈ 𝑀 it

holds that
𝑧𝑛 → 𝑧0 ⎫
⎪
⎬ ⇒ 𝑓 (𝑧𝑛 ) → 𝑓 (𝑧0 ).
∀𝑛 ∈ ℕ ∶ 𝑧𝑛 ∈ 𝑀 ⎪⎭
We say that a function 𝑓 is continuous if it is continuous on its domain 𝐷𝑓 .
Theorem 2.5 Let 𝑓 ∶ ℂ∞ → ℂ∞ and 𝑧0 ∈ ℂ∞ . Then the following statements

are equivalent:
(i) 𝑓 is continuous at the point 𝑧0 ,
(ii) 𝑧𝑛 → 𝑧0 ⇒ 𝑓 (𝑧𝑛 ) → 𝑓 (𝑧0 ),

( ( )) ( )( ) ( )
(iii) ∀𝑈 𝑓 (𝑧0 ) ∃𝑈 (𝑧0 ) ∀𝑧 ∈ 𝑈 (𝑧0 ) ∶ 𝑓 (𝑧) ∈ 𝑈 𝑓 (𝑧0 ) .
Exercise 2.2 Think over the relation between the continuity of a function
𝑓 = 𝑢 + 𝑖𝑣 ∶ ℂ → ℂ and the continuity of the functions 𝑢, 𝑣 ∶ ℝ2 → ℝ.
Example 2.11
1. The function arg 𝑧 is not continuous since it is not continuous, e.g, at the
point −1 (see Example 2.9).
2. A function arg 𝑧 is continuous on the set

ℂ ⧵ (−∞, 0] = {𝑧 ∈ ℂ ∶ 𝑧 ∉ ℝ− ∧ 𝑧 ≠ 0}.
2.6 Complex functions of a real variable and curves

Let 𝑓 ∶ ℝ → ℂ∞ be a complex function of a real variable. We can define the limit
and continuity of such a function similarly as for complex functions of a complex
variable.
Definition 2.13 Let 𝑓 ∶ ℝ → ℂ∞ . We say that a function 𝑓 has a limit 𝑎 ∈ ℂ∞

at a point 𝑡0 ∈ ℝ if
𝑡0 ≠ 𝑡𝑛 → 𝑡0 (in ℝ) ⇒ 𝑓 (𝑡𝑛 ) → 𝑎.
It is denoted by
lim 𝑓 (𝑡) = 𝑎.
𝑡→𝑡0
2.6. COMPLEX FUNCTIONS OF A REAL VARIABLE AND CURVES 19
We say that a function 𝑓 is continuous at a point 𝑡0 ∈ ℝ if
lim 𝑓 (𝑡) = 𝑓 (𝑡0 ).

𝑡→𝑡0
We say that a function 𝑓 is continuous on the set 𝑀 ⊂ ℝ if for all 𝑡0 ∈ 𝑀 it holds

that
𝑡𝑛 → 𝑡0 ⎫
⎪
⎬ ⇒ 𝑓 (𝑡𝑛 ) → 𝑓 (𝑡0 ).
∀𝑛 ∈ ℕ ∶ 𝑡𝑛 ∈ 𝑀 ⎪ ⎭
We say that a function 𝑓 is continuous if it is continuous on its domain 𝐷𝑓 .
A highly important part of continuous functions are curves.
Definition 2.14 By a curve in ℂ∞ (resp. in ℂ) we mean any continuous complex

function of a real variable
𝛾 ∶ 𝐼 → ℂ∞ (resp. 𝛾 ∶ 𝐼 → ℂ),
where 𝐼 = 𝐷𝛾 ⊂ ℝ is an interval. The set
⟨𝛾⟩ = 𝛾(𝐼) = {𝛾(𝑡) ∶ 𝑡 ∈ 𝐼} ⊂ ℂ∞
is called an image of a curve 𝛾. Let 𝑀 = ⟨𝛾⟩. Then 𝛾 is called a parametrization

of a set 𝑀.
We have already noticed that there is one-to-one correspondence between points

from ℝ2 and points from ℂ:
(𝑥, 𝑦) ↔ 𝑥 + 𝑖𝑦.
Similarly, there is one-to-one correspondence between curves in ℝ2 and curves

in ℂ:
𝛾 = (𝛾1 , 𝛾2 ) ↔ 𝛾 = 𝛾1 + 𝑖𝛾2 .
That’s why all terms for curves in ℝ2 come up for curves in ℂ, (see [2]). For exam-
ple:
• a simple curve,
• a closed curve,
• a simple closed curve,
• a oppositely oriented curve,
• a smooth arc,
• a piecewise smooth arc,
• an initial and terminal point of the curve,
• a derivative of a curve at a point,
• a tangent vector to the curve, etc.
Exercise 2.3 Sketch an image of a curve 𝛾 in the Gauss plane if
1. 𝛾(𝑡) = 2 − 3𝑖 + 2𝑒−2𝑖𝑡 , 𝑡 ∈ [0, 3𝜋∕4],
2.
[ ]
⎧ 4𝑒𝑖𝑡 , 𝑡 ∈ 0,
𝜋
,
⎪
⎪ ( )2 [ ]
𝛾(𝑡) = ⎨ 𝜋 𝜋 𝜋
𝑖 4+ −𝑡 , 𝑡∈ ,4 + ,
⎪ 2 [ 2 2]
⎪ 𝜋
𝑡 − 4 − , 𝑡 ∈ 4 + ,8 +
𝜋 𝜋
.
⎩ 2 2 2
Definition 2.15 Let 𝑀 ⊂ ℂ∞ . A closure of the set 𝑀 is defined as a set 𝑀 of

all 𝑧 ∈ ℂ∞ for which there exists a sequence (𝑧𝑛 ) in 𝑀 such that 𝑧𝑛 → 𝑧.3 Sets
𝐴, 𝐵 ⊂ ℂ∞ are called disjoint if
𝐴 ∩ 𝐵 = 𝐴 ∩ 𝐵 = ∅.
A set 𝑀 ⊂ ℂ∞ is called connected if it cannot be written as a union of two

nonempty disjoint sets. A set Ω ⊂ ℂ∞ is called a domain if it at the same time
holds that:
(i) Ω is an open set,
(ii) Ω is a connected set.4
Definition 2.16 Let 𝑀 ⊂ ℂ∞ . A set 𝐾 ⊂ 𝑀 is called a component of the set 𝑀

if it fulfills the following two conditions:
(i) 𝐾 is a connected set,
(ii) for any connected set 𝐾 ∗ ⊂ 𝑀 such that 𝐾 ⊂ 𝐾 ∗ it holds 𝐾 = 𝐾 ∗ . 5
3
If we understand a closed set as a complement of an open set, then 𝑀 can be equivalently defined
as a smallest closed set containing 𝑀.
4
It means that for any two points 𝑧1 , 𝑧2 ∈ Ω there exists a curve 𝛾 ∶ [𝑎, 𝑏] → Ω such that
𝛾(𝑎) = 𝑧1 , 𝛾(𝑏) = 𝑧2 .
5
In other words, we call a component of a set any of its maximal connected subsets.
2.6. COMPLEX FUNCTIONS OF A REAL VARIABLE AND CURVES 21
Remark 2.5 It’s easy to show6 that any set 𝑀 ⊂ ℂ∞ is a union of a system of all
its components while this system is disjoint.
Definition 2.17 Let Ω ⊂ ℂ∞ denote a domain. Then its complement in ℂ∞ (i.e.,

the set ℂ∞ ⧵ Ω) which has 𝑛 different components is called an 𝑛 –times connected
domain. A one-time connected domain is called simply connected.
Exercise 2.4
1. ∅, ℂ, ℂ∞ , 𝑈 (𝑧), where 𝑧 ∈ ℂ∞ , are simple connected domains.
2. 𝑃 (𝑧), ℂ ⧵ {𝑧}, where 𝑧 ∈ ℂ, are two-times connected domains.
3. 𝑈 (1, 2020) ⧵ {2, 4, 5 + 𝑖} is a four-times connected domain.
4. 𝑈 (3, 2) ∪ 𝑈 (4𝑖, 3) is not a domain (it is not connected).
5. ℂ∞ ⧵ {𝑧 ∈ ℂ ∶ arg 𝑧 ∈ [0, 𝜋∕4]} is not a domain (it is not open).
6
See, e.g., [5].
Chapter 3
Derivative of a complex function

of a complex variable
3.1 Derivative of a function

Definition 3.1 Let 𝑓 ∶ ℂ → ℂ. A derivative of a function 𝑓 at a point 𝑧0 ∈ ℂ
is defined as
𝑓 (𝑧) − 𝑓 (𝑧0 )
𝑓 ′ (𝑧0 ) = lim
𝑧→𝑧0 𝑧 − 𝑧0
if the limit exists and is finite. A function 𝑓 is called holomorphic on a set Ω if
a set Ω ⊂ ℂ is open and for all 𝑧 ∈ Ω there exists 𝑓 ′ (𝑧). A function 𝑓 is called
holomorphic at a point 𝑧0 ∈ ℂ if 𝑓 is holomorphic on some neighbourhood of a
point 𝑧0 (i.e., it has a derivative at each point of some neighbourhood 𝑈 (𝑧0 )).
Remark 3.1 Notice that the above definition is formally identical to the definition
of a derivative of a real function of a real variable. Theorems and proofs concerning
computing derivatives are also formally identical,1 therefore we do not introduce
them.
Theorem 3.1 If a function 𝑓 ∶ ℂ → ℂ has a derivative at a point 𝑧0 ∈ ℂ, then it

is continuous at 𝑧0 .
Proof We assume
𝑓 (𝑧) − 𝑓 (𝑧0 )
𝑓 ′ (𝑧0 ) = lim ∈ ℂ.
𝑧→𝑧0 𝑧 − 𝑧0
Then there exists a ring neighbourhood 𝑃 (𝑧0 ) such that for every 𝑧 ∈ 𝑃 (𝑧0 ) it holds
that
| 𝑓 (𝑧) − 𝑓 (𝑧0 ) |
| | < |𝑓 ′ (𝑧 )| + 1.
| 𝑧−𝑧 | 0
| 0 |
1
For example, theorems concerning differentiation of a sum, product, quotient, composition of
functions, chain rule, etc.
23
24 CHAPTER 3. DERIVATIVE OF A COMPLEX FUNCTION
For any 𝑧 ∈ 𝑃 (𝑧0 ) it also holds that

( )
0 ≤ |𝑓 (𝑧) − 𝑓 (𝑧0 )| < |𝑓 ′ (𝑧0 )| + 1 |𝑧 − 𝑧0 |.
Now, let (𝑧𝑛 ) be a sequence such that 𝑧𝑛 → 𝑧0 . From above it follows that
|𝑓 (𝑧𝑛 ) − 𝑓 (𝑧0 )| → 0 therefore 𝑓 (𝑧𝑛 ) → 𝑓 (𝑧0 ).
We proved the continuity of the function 𝑓 at the point 𝑧0 (see Theorem 2.5).
Theorem 3.2 A function 𝑓 = 𝑢 + 𝑖𝑣 has a derivative at a point 𝑧0 = 𝑥0 + 𝑖𝑦0 if

and only if the following two conditions are fulfilled:
(i) functions 𝑢 and 𝑣 are differentiable2 at the point (𝑥0 , 𝑦0 ),
(ii) functions 𝑢 and 𝑣 fulfill at the point (𝑥0 , 𝑦0 ) the so called Cauchy–Riemann
equations
𝜕𝑢 𝜕𝑣
(𝑥 , 𝑦 ) = (𝑥 , 𝑦 ),
𝜕𝑥 0 0 𝜕𝑦 0 0
𝜕𝑢 𝜕𝑣
− (𝑥0 , 𝑦0 ) = (𝑥 , 𝑦 ).
𝜕𝑦 𝜕𝑥 0 0
Moreover, if 𝑓 ′ (𝑧0 ) exists then it holds that

𝜕𝑢 𝜕𝑣 𝜕𝑣 𝜕𝑢
𝑓 ′ (𝑧0 ) = (𝑥 , 𝑦 ) + 𝑖 (𝑥0 , 𝑦0 ) = (𝑥 , 𝑦 ) − 𝑖 (𝑥0 , 𝑦0 ).
𝜕𝑥 0 0 𝜕𝑥 𝜕𝑦 0 0 𝜕𝑦
Remark 3.2 Let us explain why it is possible to formulate 𝑓 ′ via partial derivatives
of 𝑢 and 𝑣. 3
Notice that if 𝑓 ′ (𝑧0 ) exists, then
𝑓 (𝑧) − 𝑓 (𝑧0 ) 𝑓 (𝑥0 + ℎ + 𝑖𝑦0 ) − 𝑓 (𝑥0 + 𝑖𝑦0 )
𝑓 ′ (𝑧0 ) = lim = lim =
𝑧→𝑧0 𝑧 − 𝑧0 ℎ→0 (𝑥0 + ℎ + 𝑖𝑦0 ) − (𝑥0 + 𝑖𝑦0 )
ℎ∈ℝ
𝑢(𝑥0 + ℎ, 𝑦0 ) + 𝑖𝑣(𝑥0 + ℎ, 𝑦0 ) − 𝑢(𝑥0 , 𝑦0 ) − 𝑖𝑣(𝑥0 , 𝑦0 )

= lim =
ℎ→0 (𝑥0 + ℎ − 𝑥0 ) + 𝑖(𝑦0 − 𝑦0 )
ℎ∈ℝ
𝑢(𝑥0 + ℎ, 𝑦0 ) − 𝑢(𝑥0 , 𝑦0 ) 𝑣(𝑥0 + ℎ, 𝑦0 ) − 𝑣(𝑥0 , 𝑦0 )
= lim + 𝑖 lim =
ℎ→0 ℎ ℎ→0 ℎ
𝜕𝑢 𝜕𝑣
= (𝑥 , 𝑦 ) + 𝑖 (𝑥0 , 𝑦0 ).
𝜕𝑥 0 0 𝜕𝑥
2
Recall a sufficient condition of differentiability:
𝜕𝜑 𝜕𝜑
Let 𝜑 ∶ ℝ2 → ℝ. If the functions and are continuous at the point (𝑥0 , 𝑦0 ),
𝜕𝑥 𝜕𝑦
then a function 𝜑 has a derivative at the point (𝑥0 , 𝑦0 ).
3
Think carefully over the meaning of formulas of the type lim ….
ℎ→0
ℎ∈ℝ
3.2. HARMONIC FUNCTIONS, HARMONIC CONJUGATE FUNCTIONS 25
Similarly,
𝑓 (𝑧) − 𝑓 (𝑧0 ) 𝑓 (𝑥0 + 𝑖(𝑦0 + 𝑠)) − 𝑓 (𝑥0 + 𝑖𝑦0 )

𝑓 ′ (𝑧0 ) = lim = lim =
𝑧→𝑧0 𝑧 − 𝑧0 𝑠→0 (𝑥0 + 𝑖(𝑦0 + 𝑠)) − (𝑥0 + 𝑖𝑦0 )
𝑠∈ℝ
𝑢(𝑥0 , 𝑦0 + 𝑠) + 𝑖𝑣(𝑥0 , 𝑦0 + 𝑠) − 𝑢(𝑥0 , 𝑦0 ) − 𝑖𝑣(𝑥0 , 𝑦0 )
= lim =
𝑠→0 (𝑥0 − 𝑥0 ) + 𝑖(𝑦0 + 𝑠 − 𝑦0 )
𝑠∈ℝ
𝑣(𝑥0 , 𝑦0 + 𝑠) − 𝑣(𝑥0 , 𝑦0 ) 1 𝑢(𝑥0 , 𝑦0 + 𝑠) − 𝑢(𝑥0 , 𝑦0 )
= lim + lim =
𝑠→0 𝑠 𝑖 𝑠→0 𝑠
𝜕𝑣 𝜕𝑢
= (𝑥 , 𝑦 ) − 𝑖 (𝑥0 , 𝑦0 ).
𝜕𝑦 0 0 𝜕𝑦
Example 3.1 Find points at which the function
𝑓 (𝑧) = 𝑒𝑧
has a derivative. Find 𝑓 ′ (𝑧).

Solution: For each 𝑥 + 𝑖𝑦 ∈ ℂ is:
𝑓 (𝑥 + 𝑖𝑦) = 𝑒𝑥+𝑖𝑦 = 𝑒𝑥 cos 𝑦 +𝑖 𝑒𝑥 sin 𝑦,

⏟⏟⏟ ⏟⏟⏟
𝑢(𝑥,𝑦) 𝑣(𝑥,𝑦)
𝜕𝑢 𝜕𝑣
(𝑥, 𝑦) = 𝑒𝑥 cos 𝑦 = (𝑥, 𝑦),
𝜕𝑥 𝜕𝑦
𝜕𝑢 𝜕𝑣
− (𝑥, 𝑦) = 𝑒𝑥 sin 𝑦 = (𝑥, 𝑦).
𝜕𝑦 𝜕𝑥
Functions 𝑢 and 𝑣 have a derivative at any point (𝑥, 𝑦) ∈ ℝ2 . Therefore for each
𝑧 = 𝑥 + 𝑖𝑦 ∈ ℂ it holds that
𝜕𝑢 𝜕𝑣
𝑓 ′ (𝑧) = 𝑓 ′ (𝑥 + 𝑖𝑦) = (𝑥, 𝑦) + 𝑖 (𝑥, 𝑦) =
𝜕𝑥 𝜕𝑥
= 𝑒𝑥 cos 𝑦 + 𝑖𝑒𝑥 sin 𝑦 = 𝑒𝑥+𝑖𝑦 = 𝑓 (𝑥 + 𝑖𝑦) = 𝑓 (𝑧).
3.2 Harmonic functions, harmonic conjugate functions

Definition 3.2 Let 𝑀 ⊂ ℝ2 be an open set. We call a function 𝜑 ∶ ℝ2 → ℝ
harmonic on the set 𝑀 if for any (𝑥, 𝑦) ∈ 𝑀 it holds that
(i) function 𝜑 has continuous first and second partial derivatives at the point
(𝑥, 𝑦) (i.e., 𝜑 is of the class 𝐶 2 on 𝑀),
𝜕2𝜑 𝜕2𝜑
(ii) Δ𝜑(𝑥, 𝑦) = (𝑥, 𝑦) + (𝑥, 𝑦) = 0.
𝜕𝑥2 𝜕𝑦2
Example 3.2
1. Function 𝜑(𝑥, 𝑦) = 𝑥 + 𝑦 + 𝑒𝑥 cos 𝑦 is harmonic on ℝ2 .
2. Function 𝜑(𝑥, 𝑦) = Im (ln(𝑥 + 𝑖𝑦)) is not harmonic on ℝ2 ⧵ {(0, 0)}.4
Remark 3.3 In the following we use the formulation: "function 𝜑 is harmonic on

the set Ω ⊂ ℂ" instead of the correct form "function 𝜑 is harmonic on the set
{(𝑥, 𝑦) ∈ ℝ2 ∶ 𝑥 + 𝑖𝑦 ∈ Ω}".
Now, assume that a function 𝑓 = 𝑢 + 𝑖𝑣 has at any point of a domain Ω ⊂ ℂ a
second derivative5 and that functions 𝑢 and 𝑣 are of the class 𝐶 2 on the set {(𝑥, 𝑦) ∈
ℝ2 ∶ 𝑥 + 𝑖𝑦 ∈ Ω}. From Theorem 3.2 it follows that for any point 𝑥 + 𝑖𝑦 ∈ Ω we
have
𝜕𝑢 𝜕𝑣 𝜕𝑣 𝜕𝑢
𝑓 ′ (𝑥 + 𝑖𝑦) = (𝑥, 𝑦) + 𝑖 (𝑥, 𝑦) = (𝑥, 𝑦) − 𝑖 (𝑥, 𝑦),
𝜕𝑥 𝜕𝑥 𝜕𝑦 𝜕𝑦
2
𝜕 𝑢 2
𝜕 𝑣 2
𝜕 𝑢 𝜕2𝑣
𝑓 ′′ (𝑥 + 𝑖𝑦) = 2
(𝑥, 𝑦) + 𝑖 2 (𝑥, 𝑦) = − 2 (𝑥, 𝑦) − 𝑖 2 (𝑥, 𝑦).
𝜕𝑥 𝜕𝑥 𝜕𝑦 𝜕𝑦
If we compare real and imaginary parts in the latter equality, we find out that
for all 𝑥 + 𝑖𝑦 ∈ Ω ∶
Δ𝑢(𝑥, 𝑦) = 0 = Δ𝑣(𝑥, 𝑦).
In other words, functions 𝑢 and 𝑣 are harmonic on Ω.
The following theorem generalizes this contemplation.
Theorem 3.3 Let 𝑓 = 𝑢+𝑖𝑣 be holomorphic on the domain Ω ⊂ ℂ. Then functions

𝑢 and 𝑣 are harmonic on Ω.
Definition 3.3 We call functions 𝑢, 𝑣 ∶ ℝ2 → ℝ harmonic conjugate on the do-

main Ω ⊂ ℂ if at the same time it holds that
(i) functions 𝑢 and 𝑣 are harmonic on Ω,
(ii) functions 𝑢 and 𝑣 satisfy Cauchy–Riemann equations on Ω.
Remark 3.4 Notice that harmonic conjugate functions are real and imaginary parts
of holomorphic functions.
4
Why?
5
Let 𝑛 ∈ ℕ. The (𝑛 + 1)–th derivative of a function 𝑓 at a point 𝑧0 ∈ ℂ is defined by induction:
( )′
𝑓 (𝑛+1) (𝑧0 ) = 𝑓 (𝑛) (𝑧0 ),
i.e.,
𝑓 (𝑛) (𝑧) − 𝑓 (𝑛) (𝑧0 )
𝑓 (𝑛+1) (𝑧0 ) = lim
𝑧→𝑧0 𝑧 − 𝑧0
if the limit on the right exists and is finite.
3.2. HARMONIC FUNCTIONS, HARMONIC CONJUGATE FUNCTIONS 27
Example 3.3 Find holomorphic function 𝑓 = 𝑢 + 𝑖𝑣 if
𝑢(𝑥, 𝑦) = 𝑥2 − 𝑦2 + 2𝑥𝑦.
Solution: We search for a function 𝑣 ∶ ℝ2 → ℝ joined with a function 𝑢 by satisfying

Cauchy–Riemann equations:
𝜕𝑢 𝜕𝑣
(𝑥, 𝑦) = 2𝑥 + 2𝑦 = (𝑥, 𝑦) ⇒ 𝑣(𝑥, 𝑦) = 2𝑥𝑦 + 𝑦2 + 𝜑(𝑥),
𝜕𝑥 𝜕𝑦
where 𝜑 ∶ ℝ → ℝ. Now, instituting in the second Cauchy–Riemann equations we
get
𝜕𝑢 𝜕𝑣
− (𝑥, 𝑦) = 2𝑦 − 2𝑥 = (𝑥, 𝑦) = 2𝑦 + 𝜑′ (𝑥).
𝜕𝑦 𝜕𝑥
Therefore
𝜑(𝑥) = −𝑥2 + 𝑐, where 𝑐 ∈ ℝ,

𝑣(𝑥, 𝑦) = 2𝑥𝑦 + 𝑦2 − 𝑥2 + 𝑐.
One can easily make sure that 6 the function
𝑓 (𝑥 + 𝑖𝑦) = 𝑥2 − 𝑦2 + 2𝑥𝑦 + 𝑖(2𝑥𝑦 + 𝑦2 − 𝑥2 + 𝑐)
is holomorphic on ℂ for any 𝑐 ∈ ℝ.
Theorem 3.4 Let 𝑢 (resp. 𝑣) be a harmonic function on the simply connected do-
main
Ω ⊂ ℂ. Then there exists a function 𝑓 ∶ ℂ → ℂ uniquely determined up to strictly
imaginary (resp. strictly real) constant such that:
(i) 𝑓 is holomorphic on Ω,
(ii) for all 𝑥 + 𝑖𝑦 ∈ Ω ∶ 𝑢(𝑥, 𝑦) = Re 𝑓 (𝑥 + 𝑖𝑦) (resp. 𝑣(𝑥, 𝑦) = Im 𝑓 (𝑥 + 𝑖𝑦)).
Exercise 3.1
1. Find all on the domain ℂ ⧵ {0} holomorphic functions 𝑓 = 𝑢 + 𝑖𝑣, where

𝑦
𝑣(𝑥, 𝑦) = .
𝑥2 + 𝑦2
2. Prove that a function

𝑣(𝑥, 𝑦) = ln(𝑥2 + 𝑦2 )
is harmonic on the domain ℂ ⧵ {0} and that there does not exist a function
𝑢 ∶ ℝ2 → ℝ such that 𝑓 = 𝑢 + 𝑖𝑣 is holomorphic on ℂ ⧵ {0}.
6
It is necessary to check the conditions from Theorem 3.2.
3.3 Remark about geometric meaning of derivatives

Assume that a function 𝑓 ∶ ℂ → ℂ is holomorphic at the point 𝑧0 ∈ ℂ and that
′ (𝑧 )
0 ≠ 𝑓 ′ (𝑧0 ) = |𝑓 ′ (𝑧0 )| 𝑒𝑖 arg 𝑓 0 .
It follows from the definition of a derivative that

| 𝑓 (𝑧) − 𝑓 (𝑧0 ) |
lim || | = |𝑓 ′ (𝑧 )| ∈ ℝ+ .
𝑧→𝑧0 | 𝑧 − 𝑧0 || 0
Therefore, for every 𝑧 is close to the point 𝑧0 the number |𝑓 (𝑧) − 𝑓 (𝑧0 )| close
to the number |𝑓 ′ (𝑧0 )| ⋅ |𝑧 − 𝑧0 |. In other words, for a sufficiently small 𝛿 > 0
an 𝑓 –image of a circle {𝑧 ∈ ℂ ∶ |𝑧 − 𝑧0 | = 𝛿} differs a little from the circle
{𝑤 ∈ ℂ ∶ |𝑤 − 𝑓 (𝑧0 )| = |𝑓 ′ (𝑧0 )| ⋅ 𝛿}.
Now, let us demonstrate a geometrical interpretation of arg 𝑓 ′ (𝑧0 ). Let 𝛾 be
any smooth arc in ℂ such that 𝛾(𝑡0 ) = 𝑧0 . Then the number arg 𝛾 ′ (𝑡0 ) gives an
angle between tangent vector 𝛾 ′ (𝑡0 ) and a positive part of a real axis.7 Now, let
Γ(𝑡) = 𝑓 (𝛾(𝑡)) be a curve on a sufficiently small neighbourhood of a point 𝑡0 . Then,
explore a perturbation of a tangent vector Γ(′ (𝑡0 ) from
) a positive part of a real axis,
i.e., the argument Γ′ (𝑡0 ). Since Γ′ (𝑡0 ) = 𝑓 ′ 𝛾(𝑡0 ) 𝛾 ′ (𝑡0 ) = 𝑓 ′ (𝑧0 )𝛾 ′ (𝑡0 ), it holds
arg 𝑓 ′ (𝑧0 ) + arg 𝛾 ′ (𝑡0 ) ∈ Arg Γ′ (𝑡0 ).
In other words, the number arg 𝑓 ′ (𝑧0 ) gives an angle by which it is necessary to
rotate the directional vector of the tangent to the smooth arc 𝛾 at the point 𝛾(𝑡0 ) = 𝑧0
so that we get a directional vector of the tangent to the curve Γ = 𝑓 ◦𝛾 at the point
Γ(𝑡0 ) = 𝑓 (𝑧0 ) while the curve 𝛾 was arbitrarily picked.
These observations give us the following definition.
Definition 3.4 Let 𝑓 ∶ ℂ → ℂ be a function holomorphic at a point 𝑧0 and

𝑓 ′ (𝑧0 ) ≠ 0. The number |𝑓 ′ (𝑧0 )| is called the extensibility coefficient of the func-
tion 𝑓 at the point 𝑧0 .8 The number arg 𝑓 ′ (𝑧0 ) is called the rotational angle of
the function 𝑓 at the point 𝑧0 .
7
Draw a picture!
8
Moreover, if |𝑓 ′ (𝑧0 )| < 1, (resp. |𝑓 ′ (𝑧0 )| > 1) then a function 𝑓 is called a contraction, (resp.
a dilatation) of the function 𝑓 at the point 𝑧0 .
Chapter 4
Conformal function
4.1 Basic properties

Definition 4.1 A function 𝑓 ∶ ℂ∞ → ℂ∞ is called conformal in an open set
𝐺 ⊂ ℂ∞ if
(i) a function 𝑓 is continuous and injective in 𝐺,
(ii) a derivative 𝑓 ′ exists at any point of the set 𝐺 up to finitely many ones.
Exercise 4.1 Find out in which domains are the functions
𝑒𝑧 , ln 𝑧, sin 𝑧, 𝑧2 , 𝑧4
conformal.
Definition 4.2 Open sets 𝐺1 , 𝐺2 ⊂ ℂ∞ are called conformally equivalent if

there exists a function 𝑓 ∶ ℂ∞ → ℂ∞ such that
(i) a function 𝑓 is conformal in 𝐺1 ,
(ii) 𝑓 (𝐺1 ) = 𝐺2 .
Properties of conformal functions
(i) If a function 𝑓 is conformal on 𝐺, then 0 ≠ 𝑓 ′ (𝑧) ∈ ℂ for all 𝑧 ∈ 𝐺 up to

at most two points: the point ∞ (in case it belongs to 𝐺) and a point (in case
that such a point in 𝐺 exists) whose 𝑓 -image is ∞.1
(ii) An inverse function to the conformal function is also conformal.
(iii) An image of a domain under conformal function is also a domain.

1
Notice that conformal function preserves angles between curves pathing through 𝑧0 (𝑧0 ∈
𝐺, 𝑧0 ≠ ∞ ≠ 𝑓 (𝑧0 )) – see geometrical interpretation of arg 𝑓 ′ (𝑧0 ) on page 28. This property of
a function 𝑓 is called conformallity at the point 𝑧0 .
29
30 CHAPTER 4. CONFORMAL FUNCTION
(iv) All simply connected domains in ℂ∞ can be classified into four groups:
group 1. contains only empty-set,
group 2. contains only ℂ∞ ,
group 3. contains all domains of the form ℂ∞ ⧵ {𝑧0 }, where 𝑧0 ∈ ℂ∞ ,
group 4. contains all remaining simply connected domains.2
Then, any two simply connected domains Ω1 and Ω2 are conformally equiv-
alent if and only if they belong to the same group.
4.2 Linear fractional functions

Definition 4.3 A function 𝑓 ∶ ℂ∞ → ℂ∞ is called a linear fractional function if
there are numbers 𝑎, 𝑏, 𝑐, 𝑑 ∈ ℂ such that 𝑎𝑑 − 𝑏𝑐 ≠ 0 and
⎧ 𝑎𝑧 + 𝑏
⎪ if 𝑧 ∈ ℂ,
𝑓 (𝑧) = ⎨ 𝑐𝑧 + 𝑑
⎪ 𝑎
⎩ if 𝑧 = ∞.
𝑐
Properties of linear fractional functions
(i) Linear fractional function is the only conformal function from ℂ∞ on ℂ∞ .
(ii) Inverse function to the linear fractional function is a linear fractional function.
(iii) An image of a generalized circle under linear fractional function is a general-

ized circle. (Generalized circle is defined as a circle in ℂ or a line including
the point ∞.)
(iv) Let each of the sets {𝑧1 , 𝑧2 , 𝑧3 }, {𝑤1 , 𝑤2 , 𝑤3 } contains three different num-
bers from ℂ∞ . Then there exists exactly one linear fractional function 𝑓 such
that 𝑓 (𝑧1 ) = 𝑤1 , 𝑓 (𝑧2 ) = 𝑤2 and 𝑓 (𝑧3 ) = 𝑤3 .
(v) A special case of linear fractional function is a linear function given by

𝑓 (𝑧) = 𝑎𝑧 + 𝑏, where 𝑎, 𝑏 ∈ ℂ, 𝑎 ≠ 0. 3
Example 4.1 Find an image of a circle
𝐾 = {𝑧 ∈ ℂ ∶ |𝑧 − 1| = 1}
under the function

1
𝑓 (𝑧) = .
𝑧
2
I.e., all non-empty simply connected domains, whose complement contains at most two points.
3
Notice that one can get any linear function as a composition of three functions:
rotation (𝑧 → 𝑒𝑖 arg 𝑎 𝑧), homothety (𝑧 → |𝑎|𝑧), and shift (𝑧 → 𝑧 + 𝑏).
4.2. LINEAR FRACTIONAL FUNCTIONS 31
1 1 1
Solution: For 0, 2, 1 + 𝑖 ∈ 𝐾 it holds that 𝑓 (0) = ∞, 𝑓 (2) = , 𝑓 (1 + 𝑖) = − 𝑖,
2 2 2
hence an image of the circle 𝐾 is a line: 4
{ }
1
𝑓 (𝐾) = 𝑧 ∈ ℂ ∶ Re 𝑧 = ∪ {∞} .
2
4
See property (iii) of linear fractional functions.
32 CHAPTER 4. CONFORMAL FUNCTION
Chapter 5
Complex functions integral,

Cauchy’s integral theorem and
formulas
5.1 Integral of complex functions of a complex and real

variable
Theorem 5.1 (Jordan’s) Denote by 𝛾 a simple closed curve in ℂ. Then
ℂ∞ ⧵ ⟨𝛾⟩ = Ω1 ∪ Ω2 ,
where Ω1 and Ω2 are disjoint,1 non-empty, simply connected domains with common
boundary ⟨𝛾⟩.
Definition 5.1 Consider the situation from aforementioned theorem. That domain
of Ω1 , Ω2 , which does not contain ∞, is called an interior of the curve 𝛾 and is
denoted by int 𝛾. The domain containing ∞ is called an exterior of the curve 𝛾 and
is denoted by ext 𝛾.
Definition 5.2 Let 𝑓 = 𝑢 + 𝑖𝑣 ∶ ℝ → ℂ be a function continuous on the interval

[𝑎, 𝑏] where 𝑎, 𝑏 ∈ ℝ and 𝑎 < 𝑏.2 Then we define
𝑏 𝑏 𝑏 𝑏
𝑓 (𝑡) d𝑡 = 𝑢(𝑡) + 𝑖𝑣(𝑡) d𝑡 = 𝑢(𝑡) d𝑡 + 𝑖 𝑣(𝑡) d𝑡.
∫𝑎 ∫𝑎 ∫𝑎 ∫𝑎
Definition 5.3 Let 𝛾 ∶ [𝑎, 𝑏] → ℂ be a piece-wise smooth curve and let
𝑓 = 𝑢 + 𝑖𝑣 ∶ ℂ → ℂ
1
I.e., Ω1 ∩ Ω2 = ∅.
2
It means that functions 𝑢(𝑡) = Re 𝑓 (𝑡), 𝑣(𝑡) = Im 𝑓 (𝑡) ∶ ℝ → ℝ are continuous on [𝑎, 𝑏].
33
34 CHAPTER 5. COMPLEX FUNCTIONS INTEGRAL
be a continuous function on ⟨𝛾⟩. Then 3
𝑓 (𝑧) d𝑧 = 𝑢(𝑥, 𝑦) d𝑥 − 𝑣(𝑥, 𝑦) d𝑦 + 𝑖 𝑣(𝑥, 𝑦) d𝑥 + 𝑢(𝑥, 𝑦) d𝑦,

∫𝛾 ∫(𝛾) ∫(𝛾)
where integrals on the right are curve integrals of the second kind4 (𝛾 is meant to
be a curve in ℝ2 ).
Theorem 5.2 Let 𝛾 ∶ [𝑎, 𝑏] → ℂ be a smooth arc and let 𝑓 ∶ ℂ → ℂ be a

continuous function on ⟨𝛾⟩. Then
𝑏
𝑓 (𝑧) d𝑧 = 𝑓 (𝛾(𝑡))𝛾 ′ (𝑡) d𝑡.
∫𝛾 ∫𝑎
Proof Let 𝑓 = 𝑢 + 𝑖𝑣 and 𝛾 = 𝛾1 + 𝑖𝛾2 . Then
𝑓 (𝑧) d𝑧 = 𝑢(𝑥, 𝑦) d𝑥 − 𝑣(𝑥, 𝑦) d𝑦 + 𝑖 𝑣(𝑥, 𝑦) d𝑥 + 𝑢(𝑥, 𝑦) d𝑦 =

∫𝛾 ∫(𝛾) ∫(𝛾)
𝑏
= 𝑢(𝛾1 (𝑡), 𝛾2 (𝑡))𝛾1′ (𝑡) − 𝑣(𝛾1 (𝑡), 𝛾2 (𝑡))𝛾2′ (𝑡) d𝑡 +
∫𝑎
𝑏
+ 𝑖 𝑣(𝛾1 (𝑡), 𝛾2 (𝑡))𝛾1′ (𝑡) + 𝑢(𝛾1 (𝑡), 𝛾2 (𝑡))𝛾2′ (𝑡) d𝑡 =
∫𝑎
𝑏( )
= 𝑢(𝛾1 (𝑡), 𝛾2 (𝑡)) + 𝑖𝑣(𝛾1 (𝑡), 𝛾2 (𝑡)) 𝛾1′ (𝑡) +
∫𝑎
( )
+ 𝑖 𝑢(𝛾1 (𝑡), 𝛾2 (𝑡)) + 𝑖𝑣(𝛾1 (𝑡), 𝛾2 (𝑡)) 𝛾2′ (𝑡) d𝑡 =
𝑏 ( )( )
= 𝑓 𝛾1 (𝑡) + 𝑖𝛾2 (𝑡) 𝛾1′ (𝑡) + 𝑖𝛾2′ (𝑡) d𝑡
∫𝑎
𝑏
= 𝑓 (𝛾(𝑡))𝛾 ′ (𝑡) d𝑡.
∫𝑎

1
d𝑧
∫𝛾 𝑧
3
For easier recall:
𝑓 (𝑧) d𝑧 = (𝑢 + 𝑖𝑣)( d𝑥 + 𝑖 d𝑦) = 𝑢 d𝑥 − 𝑣 d𝑦 + 𝑖(𝑣 d𝑥 + 𝑢 d𝑦).
4
See [2].
5.2. CAUCHY’S INTEGRAL THEOREM 35
where 𝛾(𝑡) = 5𝑒𝑖𝑡 , 𝑡 ∈ [0, 2𝜋].

Solution: Using Definition 5.3:
1 𝑥 𝑦 −𝑦 𝑥
d𝑧 = d𝑥 + 2 d𝑦 + 𝑖 d𝑥 + 2 d𝑦 =
∫𝛾 𝑧 ∫(𝛾) 𝑥2 + 𝑦2 𝑥 + 𝑦2 ∫(𝛾) 𝑥2 + 𝑦2 𝑥 + 𝑦2
2𝜋 2𝜋
−25 sin 𝑡 cos 𝑡 25 sin 𝑡 cos 𝑡 25 sin2 𝑡 25 cos2 𝑡
= + d𝑡 + 𝑖 + d𝑡 =
∫0 25 25 ∫0 25 25
2𝜋
= 0+𝑖 1 d𝑡 = 2𝜋𝑖.
∫0
Using Theorem 5.2:
2𝜋 2𝜋
1 1
d𝑧 = 5𝑖𝑒𝑖𝑡 d𝑡 = 𝑖 d𝑡 = 2𝜋𝑖 .
∫𝛾 𝑧 ∫0 5𝑒𝑖𝑡 ∫0
5.2 Cauchy’s integral theorem

Theorem 5.3 (Cauchy’s integral) Let 𝑓 be a holomorphic function on simply con-
nected domain Ω ⊂ ℂ. Then for each closed piece-wise smooth curve 𝛾 in Ω (i.e.,
⟨𝛾⟩ ⊂ Ω) it holds that
𝑓 (𝑧) d𝑧 = 0.
∫𝛾
Proof We denote 𝑓 = 𝑢 + 𝑖𝑣 and define vector fields
𝑓1 (𝑥, 𝑦) = (𝑢(𝑥, 𝑦), −𝑣(𝑥, 𝑦)) ,

𝑓2 (𝑥, 𝑦) = (𝑣(𝑥, 𝑦), 𝑢(𝑥, 𝑦)) .
Then 𝑓1 and 𝑓2 are of the class 𝐶 2 on simply connected domain
Ω∗ = {(𝑥, 𝑦) ∈ ℝ2 ∶ 𝑥 + 𝑖𝑦 ∈ Ω}
(see Theorem 3.3). Futhermore, in Ω∗ it holds
𝜕𝑢 𝜕(−𝑣) 𝜕𝑣 𝜕𝑢
= and =
𝜕𝑦 𝜕𝑥 𝜕𝑦 𝜕𝑥
(see Theorem 3.2). Hence 𝑓1 and 𝑓2 are potentially on Ω∗ (see [1]). Therefore
𝑓 (𝑧) d𝑧 = 𝑓1 (𝑥, 𝑦)d𝑠 + 𝑖 𝑓2 (𝑥, 𝑦)d𝑠 = 0 + 𝑖0 = 0.

∫𝛾 ∫(𝛾) ∫(𝛾)
Theorem 5.4 (Cauchy’s generalized) Let Ω = int 𝛾, where 𝛾 is a simple closed

piece-wise smooth curve in ℂ. Then for any function 𝑓 ∶ ℂ → ℂ holomorphic on
Ω and continuous on Ω = Ω ∪ ⟨𝛾⟩ it holds that5
𝑓 (𝑧) d𝑧 = 0.
∫𝛾
Remark 5.1 Let

𝛾, 𝛾1 , 𝛾2 , … , 𝛾𝑛
be simple closed piece-wise smooth positively oriented curves in ℂ such that for
any 𝑖, 𝑗 ∈ {1, 2, … , 𝑛} it holds
⟨𝛾𝑖 ⟩ ⊂ ext 𝛾𝑗 for 𝑖 ≠ 𝑗,

⟨𝛾𝑖 ⟩ ⊂ int 𝛾.
Then a set
Ω = int 𝛾 ∩ ext 𝛾1 ∩ ext 𝛾2 ∩ ⋯ ∩ ext 𝛾𝑛
is an (𝑛 + 1)-times connected domain.6
Theorem 5.5 (Cauchy’s integral theorem for 𝑛 - times connected domains) Let
Ω be an (𝑛 + 1)-times connected domain described above and let 𝑓 ∶ ℂ → ℂ be a
holomorphic function in Ω and continuous on
Ω = Ω ∪ ⟨𝛾⟩ ∪ ⟨𝛾1 ⟩ ∪ ⟨𝛾2 ⟩ ∪ ⋯ ∪ ⟨𝛾𝑛 ⟩.
Then
∑
𝑛
𝑓 (𝑧) d𝑧 = 𝑓 (𝑧) d𝑧.
∫𝛾 ∫𝛾𝑖
𝑖=1
5.3 Cauchy’s integral formula

Theorem 5.6 Let 𝛾 be a simple closed piece-wise smooth positively oriented curve
in ℂ and let a function 𝑓 ∶ ℂ → ℂ be holomorphic on Ω = int 𝛾 and continuous
on Ω = Ω ∪ ⟨𝛾⟩. Then for any 𝑧0 ∈ Ω it holds that
1 𝑓 (𝑧)
𝑓 (𝑧0 ) = d𝑧. (5.1)
2𝜋𝑖 ∫𝛾 𝑧 − 𝑧0
Moreover, if 𝑛 ∈ ℕ, then for any 𝑧0 ∈ Ω exists 𝑓 (𝑛) (𝑧0 ) such that
𝑛! 𝑓 (𝑧)
𝑓 (𝑛) (𝑧0 ) = d𝑧. (5.2)
∫
2𝜋𝑖 𝛾 (𝑧 − 𝑧0 )𝑛+1
5
Notice the connection with Green’s theorem, see [1].
6
Draw a picture!
5.3. CAUCHY’S INTEGRAL FORMULA 37
Proof We will prove only assertion (5.1).

Let 𝑧0 ∈ Ω be an arbitrary point. For any 𝑟 > 0 we define a curve
𝛾𝑟 (𝑡) = 𝑧0 + 𝑟𝑒𝑖𝑡 , 𝑡 ∈ [0, 2𝜋].
From Theorem 5.5 it follows

[ ]
𝑓 (𝑧) 𝑓 (𝑧) 𝑓 (𝑧) − 𝑓 (𝑧0 ) 𝑓 (𝑧0 )
d𝑧 = lim d𝑧 = lim d𝑧 + d𝑧 .
∫𝛾 𝑧 − 𝑧0 𝑟→0+ ∫𝛾 𝑧 − 𝑧0
𝑟
𝑟→0+ ∫𝛾
𝑟
𝑧 − 𝑧0 ∫𝛾𝑟 𝑧 − 𝑧0
Now, from the assumption
𝑓 (𝑧) − 𝑓 (𝑧0 )
𝑓 ′ (𝑧0 ) = lim ∈ℂ
𝑧→𝑧0 𝑧 − 𝑧0
it follows that there exists 𝛿 > 0 and 𝑘 > 0 such that for all 𝑧 ∈ ℂ, 0 < |𝑧 − 𝑧0 | < 𝛿
it holds that
| 𝑓 (𝑧) − 𝑓 (𝑧0 ) |
| | ≤ 𝑘.
| 𝑧−𝑧 |
| 0 |
Therefore, for all sufficiently small 𝑟 > 0 we receive7
| 𝑓 (𝑧) − 𝑓 (𝑧 ) |
| 0 |
| d𝑧| ≤ 𝑘2𝜋𝑟,
|∫ 𝛾 𝑧 − 𝑧 |
| 𝑟 0 |
and thus
𝑓 (𝑧) − 𝑓 (𝑧0 )
lim d𝑧 = 0.
𝑟→0+ ∫𝛾 𝑧 − 𝑧0
𝑟
Moreover,
( )
2𝜋
𝑓 (𝑧0 ) 1 𝑖𝑡
lim d𝑧 = lim 𝑓 (𝑧0 ) 𝑟𝑖𝑒 d𝑡 = lim (𝑓 (𝑧0 )2𝜋𝑖) = 𝑓 (𝑧0 )2𝜋𝑖,
𝑟→0+ ∫𝛾 𝑧 − 𝑧0 𝑟→0+ ∫0 𝑟𝑒𝑖𝑡 𝑟→0+
𝑟
therefore
𝑓 (𝑧)
d𝑧 = 𝑓 (𝑧0 )2𝜋𝑖.
∫𝛾 𝑧 − 𝑧 0
7
We apply this estimation of the curve integral: Let 𝛾 ∶ [𝑎, 𝑏] → ℂ be a smooth arc and let
𝑓 ∶ ℂ → ℂ be a continuous function on ⟨𝛾⟩. Then
| | 𝑏
| |
| 𝑓 (𝑧) d𝑧| ≤ sup |𝑓 (𝑧)| ⋅ |𝛾 ′ (𝑡)| d𝑡 .
|∫𝛾 | 𝑧∈⟨𝛾⟩ ∫𝑎
| | ⏟⏞⏞⏞⏞⏟⏞⏞⏞⏞⏟
length of curve 𝛾
Remark 5.2
(i) It follows from Theorem 5.6 that a derivative of a holomorphic function is
also a holomorphic function. In other words, if a function 𝑓 is holomorphic
on an open set Ω and 𝑛 ∈ ℕ, then a function 𝑓 (𝑛) is holomorphic on Ω.
(ii) Let us consider Theorem 5.6. Then values of a function 𝑓 on Ω are uniquely
determined by values of a function 𝑓 on ⟨𝛾⟩.
(iii) We can obtain formula (5.2) if we formally differentiate 𝑛-times both sides of
formula (5.1) with respect to 𝑧0 .
Example 5.2 Find

𝑒𝑧
d𝑧
∫𝛾 𝑧(1 − 𝑧)3
3
where 𝛾(𝑡) = 𝑒𝑖𝑡 , 𝑡 ∈ [0, 2𝜋].
2
Solution:
It follows from Theorem 5.5 that:
𝑒𝑧 𝑒𝑧 𝑒𝑧
d𝑧 = d𝑧 + d𝑧,
∫𝛾 𝑧(1 − 𝑧)3 ∫𝛾1 𝑧(1 − 𝑧)3 ∫𝛾2 𝑧(1 − 𝑧)3
where
1 𝑖𝑡
𝛾1 (𝑡) = 𝑒 , 𝑡 ∈ [0, 2𝜋],
4
1
𝛾2 (𝑡) = 1 + 𝑒𝑖𝑡 , 𝑡 ∈ [0, 2𝜋].
4
Now, applying Theorem 5.6
𝑒𝑧 [ ]
𝑒𝑧 (1−𝑧)3 𝑒𝑧
d𝑧 = d𝑧 = 2𝜋𝑖 = 2𝜋𝑖,
∫𝛾1 𝑧(1 − 𝑧)3 ∫𝛾1 𝑧−0 (1 − 𝑧)3 𝑧=0
− 𝑒𝑧
𝑧
[( 𝑧 )′′ ]
𝑒𝑧 2𝜋𝑖 𝑒
d𝑧 = d𝑧 = − = 𝜋𝑖(−𝑒),
∫𝛾2 𝑧(1 − 𝑧)3 ∫𝛾2 (𝑧 − 1) 3 2! 𝑧 𝑧=1
hence
𝑒𝑧
d𝑧 = 𝜋𝑖(2 − 𝑒).
∫𝛾 𝑧(1 − 𝑧)3
5.4 Primitive functions and path independent integral

Definition 5.4 A function 𝐹 ∶ ℂ → ℂ is called a primitive function to a function
𝑓 ∶ ℂ → ℂ in a domain Ω ⊂ ℂ if for any 𝑧 ∈ Ω it holds that
𝐹 ′ (𝑧) = 𝑓 (𝑧).
5.4. PRIMITIVE FUNCTIONS 39
Theorem 5.7 Let 𝐹 be a primitive function to a function 𝑓 in a domain Ω. Then

every primitive function to 𝑓 in Ω is in the form 𝐹 + 𝑘 where 𝑘 ∈ ℂ.
Proof We have to prove that:
(i) for every 𝑘 ∈ ℂ a function 𝐹 + 𝑘 is primitive to 𝑓 in Ω,
(ii) if Φ is a primitive function to 𝑓 in Ω, then there exists 𝑘 ∈ ℂ such that

Φ = 𝐹 + 𝑘.
Ad i) (𝐹 + 𝑘)′ = 𝐹 ′ + 0 = 𝑓 in Ω.
Ad ii) Let us define a function 𝐺 = 𝑢 + 𝑖𝑣 ∶ ℂ → ℂ by a formula
𝐺(𝑧) = Φ(𝑧) − 𝐹 (𝑧).
Then for all 𝑧 ∈ Ω it holds that 𝐺′ (𝑧) = 0, therefore 8 for all 𝑥 + 𝑖𝑦 ∈ Ω one obtains
𝜕𝑢 𝜕𝑣
0 = 𝐺′ (𝑥 + 𝑖𝑦) = (𝑥, 𝑦) + 𝑖 (𝑥, 𝑦).
𝜕𝑥 𝜕𝑥
Therefore, for all 𝑥 + 𝑖𝑦 ∈ Ω it holds that
𝜕𝑢 𝜕𝑣 𝜕𝑣 𝜕𝑢
(𝑥, 𝑦) = (𝑥, 𝑦) = 0 = (𝑥, 𝑦) = − (𝑥, 𝑦) = 0.
𝜕𝑥 𝜕𝑦 𝜕𝑥 𝜕𝑦
Hence it follows that the functions 𝑢 and 𝑣 are constant on the set {(𝑥, 𝑦) ∈ ℝ2 ∶
𝑥 + 𝑖𝑦 ∈ Ω} and so does the function 𝐺 = 𝑢 + 𝑖𝑣 = Φ − 𝐹 in Ω.
Definition 5.5 An integral of a function 𝑓 ∶ ℂ → ℂ is called path independent

in a domain Ω ⊂ ℂ if for any two piece-wise smooth curves 𝛾1 and 𝛾2 are such that
(i) ⟨𝛾1 ⟩ ∪ ⟨𝛾2 ⟩ ⊂ Ω,
(ii) an initial point of 𝛾1 equals an initial point of 𝛾2 ,
(iii) a terminal point of 𝛾1 equals a terminal point of 𝛾2
and it holds that

𝑧2
𝑓 (𝑧) d𝑧 = 𝑓 (𝑧) d𝑧 = 𝑓 (𝑧) d𝑧
∫𝛾1 ∫𝛾2 ∫𝑧1
where 𝑧1 (resp. 𝑧2 ) denotes initial (resp. terminal) point of curves 𝛾1 , 𝛾2 .
Theorem 5.8 Let 𝑓 ∶ ℂ → ℂ be a holomorphic function on the simply connected

domain Ω ⊂ ℂ. Then an integral to a function 𝑓 is path independent in Ω.
Proof The assertion follows straightforwardly from Theorem 5.3.

8
See Theorem 3.2.
Theorem 5.9 (Morera’s) Let 𝑓 ∶ ℂ → ℂ be a continuous function on a domain

Ω ⊂ ℂ. Assume that for any simple closed piece-wise connected curve 𝛾 in Ω it
holds that
𝑓 (𝑧) d𝑧 = 0.
∫𝛾
Then 𝑓 is holomorphic in Ω.
Theorem 5.10 Assume that an integral of a continuous function 𝑓 ∶ ℂ → ℂ is

path independent in a domain Ω ⊂ ℂ. Then there exists a primitive function to 𝑓
in Ω. Moreover, for any point 𝑧0 ∈ Ω a primitive function 𝐹 to 𝑓 in Ω is given by 9
𝑧
𝐹 (𝑧) = 𝑓 (𝜉) d𝜉.
∫𝑧0
𝑧
Proof Let 𝑧0 ∈ Ω and 𝐹 (𝑧) = ∫𝑧 𝑓 (𝜉) d𝜉. We want to prove that for any 𝑧 ∈ Ω it
0
holds that
| 𝐹 (𝑧 + ℎ) − 𝐹 (𝑧) |
lim | − 𝑓 (𝑧)|| = 0.
ℎ→0 || ℎ |
Let 𝑧 ∈ Ω denote an arbitrary point. Let 𝑃 (0) be such that for all ℎ ∈ 𝑃 (0) it is
𝑧+ℎ ∈ Ω
and define for all ℎ ∈ 𝑃 (0) a curve 𝛾 given by
𝛾ℎ (𝑡) = 𝑧 + 𝑡ℎ, 𝑡 ∈ [0, 1].
Then for all ℎ ∈ 𝑃 (0) it holds that:
| 𝐹 (𝑧 + ℎ) − 𝐹 (𝑧) | 1 || |
𝑧+ℎ
|
0 ≤ || − 𝑓 (𝑧)|| = | 𝑓 (𝜉) d𝜉 − 𝑓 (𝑧)ℎ| =
| ℎ | |ℎ| ||∫𝑧 |
|
1 || | | |
| 1 | |
= | 𝑓 (𝜉) d𝜉 − 𝑓 (𝑧) 1 d𝜉 | = | 𝑓 (𝜉) − 𝑓 (𝑧) d𝜉 | ≤
|ℎ| ||∫𝛾ℎ ∫𝛾ℎ | |ℎ| |∫𝛾
| | ℎ
|
|
1
≤ sup |𝑓 (𝜉) − 𝑓 (𝑧)| |ℎ| = sup |𝑓 (𝜉) − 𝑓 (𝑧)| → 0 for ℎ → 0,
|ℎ| 𝜉∈⟨𝛾ℎ ⟩ 𝜉∈⟨𝛾ℎ ⟩
since 𝑓 is continuous in 𝑧.
Example 5.3 A function

1
𝑓 (𝑧) =
𝑧
9 𝑧
Formula ∫𝑧 𝑓 (𝜉) d𝜉 denotes an integral ∫𝛾 𝑓 (𝜉) d𝜉 where 𝛾 is any piece-wise smooth curve in Ω
0
whose initial point is 𝛾 = 𝑧0 and terminal point is 𝛾 = 𝑧.
5.4. PRIMITIVE FUNCTIONS 41
is holomorphic on a simply connected domain Ω = ℂ ⧵ {𝑧 ∈ ℝ ∶ 𝑧 ≤ 0}, therefore

a function (we integrate along curves in Ω)
𝑧 |𝑧| 𝑧 arg 𝑧
1 1
𝐹 (𝑧) = 𝑓 (𝜉) d𝜉 = d𝑥 + d𝜉 = [ln 𝑥]|𝑧| +𝑖 d𝑡 = ln 𝑧
∫1 ∫1 𝑥 ∫|𝑧| 𝜉 1 ∫0
is a primitive function to the function 𝑓 in Ω.10
Remark 5.3 Let 𝐹 denote a primitive function to a function 𝑓 on simply connected

domain Ω and let 𝑧1 , 𝑧2 ∈ Ω. Let us look at ∫𝑧 2 𝑓 (𝑧) d𝑧. 11
𝑧
1
Let 𝑧0 ∈ Ω be an arbitrary point. Then there exists a constant 𝑘 ∈ ℂ such that
for all 𝑧 ∈ Ω it holds that
𝑧
𝐹 (𝑧) = 𝑓 (𝜉) d𝜉 + 𝑘
∫𝑧0
(see Theorems 5.7, 5.8 and 5.10.)

𝑧2 𝑧0 𝑧2
𝑓 (𝑧) d𝑧 = 𝑓 (𝑧) d𝑧 + 𝑓 (𝑧) d𝑧 =
∫𝑧1 ∫𝑧1 ∫𝑧0
( 𝑧1 ) ( 𝑧2 )
= − 𝑓 (𝑧) d𝑧 + 𝑘 + 𝑓 (𝑧) d𝑧 + 𝑘 =
∫𝑧0 ∫𝑧0
𝑧
= 𝐹 (𝑧2 ) − 𝐹 (𝑧1 ) = [𝐹 (𝑧)]𝑧21 .
One can generalize this as follows:
Theorem 5.11 Let 𝑓 ∶ ℂ → ℂ be a function such that there exists a primitive

function to 𝑓 in a domain Ω ⊂ ℂ. Then an integral of the function 𝑓 is path
independent in Ω. Moreover, if 𝐹 is a primitive function to the function 𝑓 in Ω and
if 𝛾 is a piece-wise connected curve in Ω defined on [𝑖, 𝑗], then
𝑓 (𝑧) d𝑧 = 𝐹 (𝛾(𝑗)) − 𝐹 (𝛾(𝑖)).

∫𝛾
Exercise 5.1
( )′
1 1 1
1. Let 𝛾(𝑡) = 𝑒𝑖𝑡 , 𝑡 ∈ [0, 2𝜋]. Then ∫𝛾 2
d𝑧 = 0 since − = 2 in the
𝑧 𝑧 𝑧
domain ℂ ⧵ {0}.
1+𝑖 1+𝑖 1 1 1
2. ∫0 sin 𝑧 cos 𝑧 d𝑧 = ∫0 sin(2𝑧) d𝑧 = [− cos(2𝑧)]1+𝑖
0
= (1 − cos(2 + 2𝑖)) .
2 4 4
2𝜋𝑖 2𝜋𝑖 𝑧
3. ∫0 𝑧𝑒𝑧 d𝑧 = [𝑧𝑒𝑧 ]2𝜋𝑖
0
− ∫0 𝑒 d𝑧 = 2𝜋𝑖 − [𝑒𝑧 ]2𝜋𝑖
0
= 2𝜋𝑖 (the integration by
parts was used).
10
Think over it carefully!
11
Once again we integrate along piece-wise connected curves in Ω.
Chapter 6
Number series, sequences and

function series
6.1 Number series

Definition 6.1 A series of complex numbers is given by
∑
∞
𝑧1 + 𝑧2 + ⋯ + 𝑧𝑛 + ⋯ = 𝑧𝑛 , (6.1)
𝑛=1
where 𝑧𝑛 ∈ ℂ for all 𝑛 ∈ ℕ.

A number 𝑧𝑛 is called an 𝑛-th term of a series (6.1). A sequence (𝑠𝑛 ) defined
by
∑𝑛
𝑠𝑛 = 𝑧1 + 𝑧2 + ⋯ + 𝑧𝑛 = 𝑧𝑘
𝑘=1
is called a sequence of partial sum of a series (6.1). A series (6.1) called conver-
gent if there exists a finite limit lim 𝑠𝑛 ∈ ℂ. Then the number
𝑠 = lim 𝑠𝑛
is called a sum of the series (6.1) and 1

∑
∞
𝑠= 𝑧𝑛 .
𝑛=1
A series which is not convergent is called divergent.

∑
∞
Theorem 6.1 Let 𝑧𝑛 be a series. Then2
𝑛=1
1
∑
∞
Notice that in the following the notation 𝑧𝑛 is used for a series and also for its sum, which is
𝑛=1
a number.
2
See [3].
43
44 CHAPTER 6. SERIES AND SEQUENCES
(i) convergence neccesary condition
∑
∞
𝑧𝑛 is convergent ⇒ lim 𝑧𝑛 = 0.
𝑛=1
(ii) relation between convergent series and convergent real and imaginary parts
∑
∞
∑
∞
∑
∞
(𝑥𝑛 + 𝑖𝑦𝑛 ) is convergent ⇔ series 𝑥𝑛 and 𝑦𝑛 are convergent.
𝑛=1 𝑛=1 𝑛=1
∑
∞
Moreover, if a series (𝑥𝑛 + 𝑖𝑦𝑛 ) is convergent, then its sum is given as
𝑛=1
∑
∞
∑
∞
∑
∞
(𝑥𝑛 + 𝑖𝑦𝑛 ) = 𝑥𝑛 + 𝑖 𝑦𝑛 .
𝑛=1 𝑛=1 𝑛=1
(iii) Bolzano–Cauchy’s condition
∑
∞
( )( )( )
𝑧𝑛 is convergent ⇔ ∀𝜀 ∈ ℝ+ ∃𝑛0 ∈ ℕ ∀𝑛, 𝑚 ∈ ℕ; 𝑛, 𝑚 > 𝑛0 ∶
𝑛=1
|𝑠𝑛 − 𝑠𝑚 | < 𝜀
( )
∑ 𝑛
𝑠𝑛 = 𝑧𝑘 .
𝑘=1
(iv) absolute convergence test
∑
∞
∑
∞
|𝑧𝑛 | is convergent ⇒ 𝑧𝑛 is convergent .
𝑛=1 𝑛=1
∑
∞
∑
∞
(A series 𝑧𝑛 is called absolutely convergent if a series |𝑧𝑛 | is conver-
𝑛=1 𝑛=1
gent. A series which is convergent but not absolutely convergent is called a
conditionally convergent series.)
(v) comparison test
⎫
∀𝑛 ∈ ℕ ∶ |𝑧𝑛 | ≤ 𝑎𝑛 ⎪ ∑∞
⎪
∑∞
⎬ ⇒ 𝑧𝑛 is absolutely convergent.
𝑎𝑛 is convergent ⎪ 𝑛=1
𝑛=1 ⎪
⎭
6.1. NUMBER SERIES 45
(vi) d’Alembert’s ratio test
| 𝑧𝑛+1 | ∑∞
lim || |<1 ⇒
| 𝑧𝑛 is absolutely convergent,
| 𝑧𝑛 | 𝑛=1
| 𝑧𝑛+1 | ∑∞
lim || |>1 ⇒
| 𝑧𝑛 is divergent.
| 𝑧𝑛 | 𝑛=1
(vii) Cauchy’s criterion, also known as the 𝑛-th root test

√ ∑
∞
lim 𝑛
|𝑧𝑛 | < 1 ⇒ 𝑧𝑛 is absolutely convergent,
𝑛=1
√ ∑ ∞
lim 𝑛
|𝑧𝑛 | > 1 ⇒ 𝑧𝑛 is divergent.
𝑛=1
(viii) integral test

Let 𝑓 ∶ ℝ → ℝ be a non-negative, monotone, decreasing, and continuous
function defined on [1, +∞) and let |𝑧𝑛 | = 𝑓 (𝑛) for all 𝑛 ∈ ℕ. Then
∑
∞ +∞
|𝑧𝑛 | < +∞ ⇔ 𝑓 (𝑥) d𝑥 < +∞ .
∫1
𝑛=1
(ix) Leibniz criterion
⎫
∀𝑛 ∈ ℕ ∶ 0 ≤ 𝑧𝑛+1 ≤ 𝑧𝑛 ⎪ ∑∞
⎬ ⇒ (−1)𝑛+1 𝑧𝑛 is convergent.
lim 𝑧𝑛 = 0 ⎪ 𝑛=1
⎭
(x) convergence of a geometric series theorem

∑
∞
A series 𝑞 𝑛−1 , where 𝑞 ∈ ℂ, is convergent if and only if |𝑞| < 1. In that
𝑛=1
case it holds that
∑
∞
1
𝑞 𝑛−1 = .
𝑛=1
1−𝑞
Example 6.1
1. A series
∑∞
𝑛
𝑛
(1 + 𝑖)𝑛
𝑛=1
3
is absolutely convergent since
| 𝑛+1 𝑛+1 || √
|
| 3𝑛+1 (1 + 𝑖) | | 1 𝑛 + 1 | 2
| |=| |
| 𝑛 | | 3 𝑛 (1 + 𝑖)| → 3 < 1.
| 𝑛 | | |
| 3𝑛 (1 + 𝑖) |
| |
2. A series
∑∞ 𝜋
𝑒𝑖 𝑛
√
𝑛=1 𝑛
𝜋
𝑒𝑖 𝑛 1 𝜋 1 𝜋
is divergent because for all 𝑛 ∈ ℕ we have √ = √ cos + 𝑖 √ sin
𝑛 𝑛 𝑛 𝑛 𝑛
∑ 1
∞
𝜋
and at the same time a series √ cos is divergent. 3
𝑛=1 𝑛 𝑛
6.2 Sequence of functions, pointwise and uniform conver-

gence
Definition 6.2 We say that a sequence of functions (𝑓𝑛 ) converges pointwise on
the set Ω ⊂ ℂ∞ to the function 𝑓 if for all 𝑧 ∈ Ω
lim 𝑓𝑛 (𝑧) = 𝑓 (𝑧).
In other words, if
( )( )( )
(∀𝑧 ∈ Ω) ∀𝜀 ∈ ℝ+ ∃𝑛0 ∈ ℕ ∀𝑛 ∈ ℕ, 𝑛 ≥ 𝑛0 ∶ 𝑓𝑛 (𝑧) ∈ 𝑈 (𝑓 (𝑧), 𝜀).
It is denoted by 𝑓𝑛 → 𝑓 on Ω.
The natural number 𝑛0 mentioned above depends on the choice 𝑧 ∈ Ω and

𝜀 ∈ ℝ+ . If it is possible to choose 𝑛0 independently on 𝑧 ∈ Ω and functions 𝑓𝑛 and
𝑓 are finite then one observes uniform convergence on Ω.
Definition 6.3 Assume that 𝑓𝑛 and 𝑓 are finite functions defined on a set Ω ⊂ ℂ∞
for all 𝑛 ∈ ℕ. We say that a sequence of functions (𝑓𝑛 ) converges uniformly on
the set Ω to the function 𝑓 if
[ ]
lim sup |𝑓𝑛 (𝑧) − 𝑓 (𝑧)| = 0.
𝑧∈Ω
In other words,
( )( )( )
∀𝜀 ∈ ℝ+ ∃𝑛0 ∈ ℕ ∀𝑛 ∈ ℕ, 𝑛 ≥ 𝑛0 (∀𝑧 ∈ Ω) ∶ 𝑓𝑛 (𝑧) ∈ 𝑈 (𝑓 (𝑧), 𝜀).
It is denoted by 𝑓𝑛 →
→ 𝑓 on Ω.
Theorem 6.2 Let 𝑓𝑛 → → 𝑓 on Ω and let 𝑓𝑛 be a continuous function on Ω for all

𝑛 ∈ ℕ. Then a function 𝑓 is continuous on Ω.
3
6.2. SEQUENCE AND CONVERGENCY 47
Definition 6.4 Assume that functions 𝑓𝑛 and 𝑓 are finite and defined on a set Ω ⊂
ℂ∞ for all 𝑛 ∈ ℕ. We say that a function series
∑
∞
𝑓1 (𝑧) + 𝑓2 (𝑧) + ⋯ + 𝑓𝑛 (𝑧) + … = 𝑓𝑛 (𝑧) (6.2)
𝑛=1
converges pointwise (resp. converges uniformly) on a set Ω to the sum 𝑓 if a

sequence (𝑠𝑛 ) of partial sum of function series (6.2) converges 4 pointwise (resp.
uniformly) on Ω to the function 𝑓 .
Theorem 6.3 (Weierstrass) Assume that for all 𝑛 ∈ ℕ a function 𝑓𝑛 is holomor-

∑
∞
phic on Ω ⊂ ℂ and that a function series 𝑓𝑛 (𝑧) converges locally uniformly on
𝑛=1
Ω, i.e.,
∑
∞
(∀𝑧 ∈ Ω) (∃𝑈 (𝑧) ⊂ Ω) ∶ 𝑓𝑛 (𝑧) converges uniformly on 𝑈 (𝑧).
𝑛=1
Then a function 𝑓 defined by
∑
∞
𝑓 (𝑧) = 𝑓𝑛 (𝑧)
𝑛=1
is holomorphic in the domain Ω and for all 𝑝 ∈ ℕ and 𝑧 ∈ Ω it holds that
∑
∞
(𝑝)
𝑓 (𝑧) = 𝑓𝑛(𝑝) (𝑧).
𝑛=1
Moreover, if 𝛾 is a piece-wise smooth curve in Ω, then5
∑
∞
𝑓 (𝑧)d𝑧 = 𝑓𝑛 (𝑧)d𝑧.
∫𝛾 ∫𝛾
𝑛=1
4
∑
𝑛
𝑠𝑛 (𝑧) = 𝑓𝑘 (𝑧).
𝑘=1
5
Symbolically written:
(∑ )′ (∑ ) ( )
∑ ′
∑
… = (… ) , … = … .
∫ ∫
Chapter 7
Power and Taylor series
7.1 Power series

Definition 7.1 A power series with a centre 𝑧0 ∈ ℂ is defined as a functional
series
∑
∞
2
𝑎0 + 𝑎1 (𝑧 − 𝑧0 ) + 𝑎2 (𝑧 − 𝑧0 ) + ⋯ = 𝑎𝑛 (𝑧 − 𝑧0 )𝑛 (7.1)
𝑛=0
where 𝑎𝑛 ∈ ℂ for every 𝑛 ∈ ℕ ∪ {0}.
Let us focus on a convergence of the series (7.1). In other words, let us explore
for which 𝑧 ∈ ℂ that series is convergent. Obviously, series (7.1) is convergent for
𝑧 = 𝑧0 , i.e., in its centre, and the sum is 𝑎0 at this point. Now, assume that the
series (7.1) is convergent at the point 𝑧1 ≠ 𝑧0 and that 𝑧 ∈ ℂ is such a point that
|𝑧 − 𝑧0 | < |𝑧1 − 𝑧0 |. Then for all 𝑛 ∈ ℕ it holds that
| 𝑧 − 𝑧0 |𝑛
|𝑎 (𝑧 − 𝑧 )𝑛 | = |𝑎 (𝑧 − 𝑧 )𝑛 | | | (7.2)
| 𝑛 0 | | 𝑛 1 0 | |𝑧 − 𝑧 | .
| 1 0|
∑
∞
Now let us apply Theorem 6.1. From the assumption that the series 𝑎𝑛 (𝑧1 − 𝑧0 )𝑛
𝑛=0
is convergent, it follows that
( )
lim 𝑎𝑛 (𝑧1 − 𝑧0 )𝑛 = 0.
Therefore there exists 𝑘 ∈ ℝ+ such that ||𝑎𝑛 (𝑧1 − 𝑧0 )𝑛 || ≤ 𝑘 for all 𝑛 ∈ ℕ.
Moreover, since ||(𝑧 − 𝑧0 )(𝑧1 − 𝑧0 )|| < 1 it follows that a geometric series
∑∞
| 𝑧 − 𝑧0 |𝑛
𝑘 || |
|
𝑛=0 | 𝑧 1 − 𝑧 0 |
∑
∞
is convergent, and hence from (7.2) one can see that 𝑎𝑛 (𝑧 − 𝑧0 )𝑛 is absolutely
𝑛=0
convergent. This can be generalized as follows.
49
50 CHAPTER 7. POWER AND TAYLOR SERIES
∑
∞
Theorem 7.1 (Abel’s) Let 𝑎𝑛 (𝑧 − 𝑧0 )𝑛 be a series convergent at a point
𝑛=0
𝑧1 ≠ (𝑧0 . Then this )series is absolutely convergent and locally uniformly convergent
in 𝑈 𝑧0 , |𝑧1 − 𝑧0 | .
∑
∞
Corollary 1 If a power series 𝑎𝑛 (𝑧 − 𝑧0 )𝑛 is divergent at a point 𝑧2 ∈ ℂ, then it
𝑛=0
is divergent at each point of the set
{𝑧 ∈ ℂ ∶ |𝑧 − 𝑧0 | > |𝑧2 − 𝑧0 |}.
Theorem 7.2 For any power series (7.1) with a centre 𝑧0 there exists exactly one
number 𝑅 ∈ [0, +∞) ∪ {+∞}, which is called a radius of convergence of the
power series (7.1) , such that
(i) if |𝑧 − 𝑧0 | < 𝑅, then series (7.1) is absolutely convergent,
(ii) if |𝑧 − 𝑧0 | > 𝑅, then series (7.1) is divergent.
Proof Let us define

{ }
∑
∞
𝑅 = sup |𝑧 − 𝑧0 | ∶ 𝑧 ∈ ℂ ∧ 𝑎𝑛 (𝑧 − 𝑧0 ) is convergent
𝑛
.
𝑛=0
Now, one can easily prove the above assertions using Theorem 7.1.
Definition 7.2 Assume that a radius of convergence 𝑅 of a power series (7.1) satis-
fies 0 < 𝑅 < +∞. A set 𝑈 (𝑧0 , 𝑅) is called a disk of the convergence of the power
series (7.1). For 𝑅 = +∞ a disk of the convergence of the power series (7.1) is a
set 𝑈 (𝑧0 , +∞) = ℂ.
∑
∞
Remark 7.1 Assume that a radius of convergence 𝑅 of a power series 𝑎𝑛 (𝑧 −
𝑛=0
𝑧0 )𝑛 satisfies 0 < 𝑅 < +∞. Notice that in that case one can say nothing about the
convergence of the power series at the points of the circle
{𝑧 ∈ ℂ ∶ |𝑧 − 𝑧0 | = 𝑅}.
One can demonstrate this situation with the power series1
∑
∞
∑
∞ 𝑛
𝑧 ∑
∞ 𝑛
𝑧
𝑛
𝑧 , , .
𝑛=1 𝑛=1
𝑛 𝑛=1
𝑛2
1
∑
∞
Power series of the form 𝑎𝑛 (𝑧 − 𝑧0 )𝑛 where 𝑧0 = 0 and 𝑎0 = 0 are considered.
𝑛=0
7.1. POWER SERIES 51
Since
| 𝑧𝑛+1 | | 𝑧𝑛+1 |
| 𝑧𝑛+1 | | | | (𝑛+1)2 |
| | | 𝑛+1 | | |
| 𝑧𝑛 | → |𝑧|, | 𝑧𝑛 | → |𝑧|,
| |
| 𝑧𝑛 | → |𝑧|,
| |
| | | 𝑛 | | 𝑛2 |
| | | |
then (use d’Alembert’s ratio test) the radius of convergence of each of these power
series equals 1. Moreover,
∑
∞
(i) the series 𝑧𝑛 is divergent at every point of the circle {𝑧 ∈ ℂ ∶ |𝑧| = 1}
𝑛=1
since any 𝑧 ∈ ℂ, |𝑧| = 1 does not fulfill the necessary convergence condition
∑
∞
for 𝑧𝑛 , i.e., the condition lim 𝑧𝑛 = 0,
𝑛=1
∑
∞
𝑧𝑛
(ii) the series 𝑛
is uniformly convergent for 𝑧 = −1 (use the Leibnitz crite-
𝑛=1
rion) and is divergent for 𝑧 = 1 (use the integral test),
∑
∞
𝑧𝑛
(iii) the series 𝑛2
is absolutely convergent for all 𝑧 ∈ ℂ, |𝑧| = 1 (use the
𝑛=1
integral test).
Theorem 7.3 Assume that

| 𝑎𝑛+1 | √
lim || | = 𝐿, resp. lim 𝑛 |𝑎 | = 𝐾.
| 𝑛
| 𝑎𝑛 |
∑
∞
Then for a radius of convergence 𝑅 of a power series 𝑎𝑛 (𝑧 − 𝑧0 )𝑛 it holds
𝑛=0
⎧ ⎧
⎪ 1 if 𝐿 ∈ ℝ+ , ⎪ 1 if 𝐾 ∈ ℝ+ ,
⎪ 𝐿 ⎪ 𝐾
⎪ ⎪
𝑅 = ⎨ 0 if 𝐿 = +∞, resp. 𝑅 = ⎨ 0 if 𝐾 = +∞,
⎪ ⎪
⎪ ⎪
⎪ +∞ if 𝐿 = 0, ⎪ +∞ if 𝐾 = 0.
⎩ ⎩
Proof Notice that for 𝑧 ≠ 𝑧0 we have
| 𝑎 (𝑧 − 𝑧 )𝑛+1 | √
| 𝑛+1 |
| = 𝐿|𝑧 − 𝑧0 |, resp. lim ||𝑎𝑛 (𝑧 − 𝑧0 )𝑛 || = 𝐾|𝑧 − 𝑧0 |.
0
lim | 𝑛
| 𝑎𝑛 (𝑧 − 𝑧0 )𝑛 |
| |
Then it is enough to use d’Alembert’s ratio test, resp. Cauchy’s criterion.
Example 7.1 Find a domain of convergence of a power series 2

∑∞
𝑛 𝑛
𝑛
𝑧 .
𝑛=0
2
Solution:
√ √
𝑛
𝑛 𝑛 𝑛 1
lim = lim = ,
2𝑛 2 2
hence 𝑅 = 2. The series is absolutely convergent for every 𝑧 ∈ 𝑈 (0, 2) and is
divergent for every 𝑧 ∈ ℂ, |𝑧| > 2.
For |𝑧| = 2 it holds
|𝑛 |
lim || 𝑛 𝑧𝑛 || = lim 𝑛 = ∞ ≠ 0,
|2 |
∑
∞
𝑛
therefore the series 2𝑛
𝑧𝑛 is divergent (the necessary convergence condition is
𝑛=0
not fulfilled).
Example 7.2 Find a radius of convergence of a power series

∑
∞
(2𝑛)!
𝑧𝑛 .
𝑛=0
(𝑛!)2
Solution:
(2(𝑛 + 1))!
((𝑛 + 1)!)2 (2𝑛 + 2)(2𝑛 + 1)
= → 4,
(2𝑛)! (𝑛 + 1)(𝑛 + 1)
(𝑛!)2
hence 𝑅 = 1∕4.
∑
∞
Theorem 7.4 Assume that a power series 𝑎𝑛 (𝑧 − 𝑧0 )𝑛 has a radius of conver-
𝑛=0
gence 𝑅 > 0. Then a function 𝑓 defined by
∑
∞
𝑓 (𝑧) = 𝑎𝑛 (𝑧 − 𝑧0 )𝑛
𝑛=0
is holomorphic in the domain 𝑈 (𝑧0 , 𝑅). Moreover, for all 𝑝 ∈ ℕ and 𝑧 ∈ 𝑈 (𝑧0 , 𝑅)
the following equality holds
∑
∞
𝑓 (𝑝) (𝑧) = 𝑛 (𝑛 − 1) ⋯ (𝑛 − 𝑝 + 1) 𝑎𝑛 (𝑧 − 𝑧0 )𝑛−𝑝
𝑛=𝑝
2
It means, find a set of all 𝑧 ∈ ℂ for which the series converges.
7.1. POWER SERIES 53
∑
∞
and the power series 𝑛 (𝑛 − 1) ⋯ (𝑛 − 𝑝 + 1) 𝑎𝑛 (𝑧 − 𝑧0 )𝑛−𝑝 has the same radius
𝑛=𝑝
of convergence 𝑅.
Proof The above theorem follows immediately from Weierstrass (see Theorem 6.3)
and Abel’s theorems (see Theorem 7.1).
Example 7.3 Find a sum of a power series
∑∞
𝑧𝑛
(−1)𝑛−1
𝑛=1
𝑛
in a disk of convergence.
Solution: Since
1
𝑛+1
→ 1,
1
𝑛
a domain 𝑈 (0, 1) is a disk of convergence of the above power series. Let us define
a function 𝑓 by
∑∞
𝑧𝑛
𝑓 (𝑧) = (−1)𝑛−1 .
𝑛=1
𝑛
Then for all 𝑧 ∈ ℂ, |𝑧| < 1 it holds that
∑
∞
∑
∞
1 1
𝑓 ′ (𝑧) = (−1)𝑛−1 𝑧𝑛−1 = (−𝑧)𝑛−1 = = .
𝑛=1 𝑛=1
1 − (−𝑧) 1 + 𝑧
Therefore there exists 𝑐 ∈ ℂ such that for any 𝑧 ∈ 𝑈 (0, 1) we have
𝑓 (𝑧) = ln(1 + 𝑧) + 𝑐.
Obviously,
0 = 𝑓 (0) = ln 1 + 𝑐 = 𝑐
then for all 𝑧 ∈ 𝑈 (0, 1) holds
∑
∞
𝑧𝑛
𝑓 (𝑧) = (−1)𝑛−1 = ln(1 + 𝑧).
𝑛=1
𝑛
∑
∞
Theorem 7.5 (Abel’s) Assume that a power series 𝑎𝑛 (𝑧 − 𝑧0 )𝑛 has a radius of
𝑛=0
convergence 𝑅 ∈ (0, +∞) and that this power series is convergent at a point
𝑧1 = 𝑧0 + 𝑅𝑒𝑖𝜑 , where 𝜑 ∈ ℝ.
Then a function 𝑓 defined by
∑
∞
𝑓 (𝑧) = 𝑎𝑛 (𝑧 − 𝑧0 )𝑛
𝑛=0
is continuous on an interval [𝑧0 , 𝑧1 ], i.e., on a set

{ } { }
𝑧0 + 𝑟𝑒𝑖𝜑 ∶ 𝑟 ∈ [0, 𝑅] = 𝑧0 + (𝑧1 − 𝑧0 ) 𝑡 ∶ 𝑡 ∈ [0, 1] .
Specially:
( ) ( ) ( )
𝑓 (𝑧1 ) = 𝑓 𝑧0 + 𝑅𝑒𝑖𝜑 = lim 𝑓 𝑧0 + 𝑟𝑒𝑖𝜑 = lim 𝑓 𝑧0 + (𝑧1 − 𝑧0 ) 𝑡 .
𝑟→𝑅− 𝑡→1−
Example 7.4 Find a sum of a series
∑∞
1
(−1)𝑛−1 .
𝑛=1
𝑛
Solution: First of all, let us notice that the above series is convergent.3 Now, let 𝑓
be a function defined by
∑∞
𝑧𝑛
𝑓 (𝑧) = (−1)𝑛−1 .
𝑛=1
𝑛
From Theorem 7.5 and the above example it follows that
∑∞
1
(−1)𝑛−1 = 𝑓 (1) = lim 𝑓 (𝑧) = lim (ln(1 + 𝑧)) = ln 2.
𝑛=1
𝑛 𝑧→1− 𝑧→1−
7.2 Taylor series

As mentioned above, a sum of a power series (in a disk of convergence) is a holo-
morphic function. The following theorem outlines that every holomorphic function
is (at least locally) a sum of a certain power series.
Theorem 7.6 (Taylor’s expansion of holomorphic function) Let 𝑓 be a holomor-

phic function on 𝑈 (𝑧0 , 𝑅) where 𝑧0 ∈ ℂ and 𝑅 ∈ (0, +∞) ∪ {+∞}. Then there
∑
∞
exists exactly one power series 𝑎𝑛 (𝑧−𝑧0 )𝑛 such that for all 𝑧 ∈ 𝑈 (𝑧0 , 𝑅) it holds
𝑛=0
that
∑
∞
𝑓 (𝑧) = 𝑎𝑛 (𝑧 − 𝑧0 )𝑛 .
𝑛=0
3
Use the Leibniz criterion.
7.2. TAYLOR SERIES 55
Moreover, if 𝜚 denotes a real number such that 0 < 𝜚 < 𝑅, then the coefficients of
the above Taylor series satisfy
𝑓 (𝑛) (𝑧0 ) 1 𝑓 (𝑧)
𝑎𝑛 = = d𝑧,
𝑛! 2𝜋𝑖 ∫𝛾 (𝑧 − 𝑧0 )𝑛+1
where
𝛾(𝑡) = 𝑧0 + 𝜚 𝑒𝑖𝑡 , 𝑡 ∈ [0, 2𝜋].
Remark 7.2 If 𝑓 denotes a holomorphic function in ℂ, then a radius of conver-

gence of its Taylor series (with a centre at any point 𝑧0 ∈ ℂ) equals +∞. Examples
of such functions and their Taylor series with a centre at 0 are
∑
∞ 𝑛
𝑧 ∑
∞
𝑧2𝑛+1 ∑ ∞
𝑧2𝑛
𝑒𝑧 = , sin 𝑧 = (−1)𝑛 , cos 𝑧 = (−1)𝑛 .
𝑛=0
𝑛! 𝑛=0
(2𝑛 + 1)! 𝑛=0
(2𝑛)!
Example 7.5 Find a Taylor series of a function 𝑓 with a centre 𝑧0 if

1
a) 𝑓 (𝑧) = , 𝑧 = 0,
3−𝑧 0
1
b) 𝑓 (𝑧) = , 𝑧 = −1 + 3𝑖,
3−𝑧 0
c) 𝑓 (𝑧) = ln 𝑧, 𝑧0 = 2.
Solution: Ad a) Notice that a function 𝑓 is holomorphic on 𝑈 (0, 3). Using the

theorem about convergence of a geometric series is a crucial tool to find its Taylor
series: 4
∞ ( )
1 ∑ 𝑧 𝑛 ∑ 𝑧𝑛
∞
1 1 1
∀𝑧 ∈ 𝑈 (0, 3) ∶ 𝑓 (𝑧) = = = = .
3−𝑧 3 1− 𝑧 3 𝑛=0 3 𝑛=0
3𝑛+1
3
Ad b) Let us apply the same process as above. For all

𝑧 ∈ 𝑈 (−1 + 3𝑖, |3 − (−1 + 3𝑖)| ) = 𝑈 (−1 + 3𝑖, 5)
it holds that
1 1 1 1 ∑ (𝑧 + 1 − 3𝑖)𝑛 ∞
𝑓 (𝑧) = = = = .
3 − 𝑧 4 − 3𝑖 − (𝑧 − (−1 + 3𝑖)) 4 − 3𝑖 1 − 𝑧+1−3𝑖 𝑛=0
(4 − 3𝑖) 𝑛+1
4−3𝑖
Ad c) Obviously, the function 𝑓 is holomorphic on 𝑈 (2, 2). Then all 𝑧 ∈ 𝑈 (2, 2)

satisfy
( )𝑛
1∑ ∑
∞ ∞
1 1 1 𝑧−2 (−1)𝑛
𝑓 ′ (𝑧) = = = (−1)𝑛 = (𝑧 − 2)𝑛 .
𝑧 2 1 + 𝑧−2 2 𝑛=0 2 𝑛=0
2𝑛+1
2
4
See Theorem 6.1.
Therefore there exists 𝑐 ∈ ℂ such that for all 𝑧 ∈ 𝑈 (2, 2) it holds that
∑∞
(−1)𝑛 (𝑧 − 2)𝑛+1
𝑓 (𝑧) = + 𝑐.
𝑛=0
2𝑛+1 𝑛+1
Since
𝑓 (2) = ln 2 = 𝑐,
then for all 𝑧 ∈ 𝑈 (2, 2) it holds that
∑
∞
(−1)𝑛−1
𝑓 (𝑧) = ln 2 + (𝑧 − 2)𝑛 .
𝑛=1
2𝑛 𝑛
Theorem 7.7 (Liouville’s) Let 𝑓 be a holomorphic and bounded (i.e., there exists
𝑀 ∈ ℝ+ such that for every 𝑧 ∈ ℂ it holds |𝑓 (𝑧)| ≤ 𝑀) function on ℂ. Then 𝑓 is
constant on ℂ.
Proof It is known from Theorem 7.6 that for all 𝑧 ∈ ℂ we have

∑
∞
𝑓 (𝑧) = 𝑎 𝑛 𝑧𝑛 ,
𝑛=0
where for all 𝑛 ∈ ℕ ∪ {0} and all 𝜚 ∈ (0, +∞) it holds that
2𝜋
1 𝑓 (𝑧) 1 𝑓 (𝜚 𝑒𝑖𝑡 )
𝑎𝑛 = d𝑧 = 𝜚 𝑖𝑒𝑖𝑡 d𝑡,
2𝜋𝑖 ∫𝛾 𝑧𝑛+1 2𝜋𝑖 ∫0 (𝜚 𝑒𝑖𝑡 )𝑛+1
𝛾(𝑡) = 𝜚 𝑒𝑖𝑡 , 𝑡 ∈ [0, 2𝜋].
Using above given, for all 𝑛 ∈ ℕ ∪ {0} and all 𝜚 ∈ (0, +∞) it holds
| 2𝜋 𝑓 (𝜚 𝑒𝑖𝑡 ) | 2𝜋
1 | | 1 𝑀 𝑀
|𝑎𝑛 | = | 𝑖 d𝑡 | ≤ d𝑡 = 𝑛 .
2𝜋 |∫0 (𝜚 𝑒𝑖𝑡 )𝑛 | 2𝜋 ∫0 𝜚𝑛 𝜚
| |
Now, since one could choose a constant 𝜚 ∈ ℝ+ arbitrarily small, then it follows
from theestimation |𝑎𝑛 | ≤ 𝑀∕𝜚𝑛 that for any 𝑛 ∈ ℕ it holds 𝑎𝑛 = 0. Therefore for
every 𝑧 ∈ ℂ is 𝑓 (𝑧) = 𝑎0 and the function 𝑓 is constant.
Theorem 7.8 (Fundamental theorem of algebra) Every positive degree polyno-

mial has at least one root in ℂ. In other words, let 𝑓 ∶ ℂ → ℂ denote a function
defined by
𝑓 (𝑧) = 𝑎𝑛 𝑧𝑛 + 𝑎𝑛−1 𝑧𝑛−1 + ⋯ + 𝑎1 𝑧 + 𝑎0 ,
where
𝑛 ∈ ℕ, 𝑎0 , 𝑎1 , … , 𝑎𝑛 ∈ ℂ, 𝑎𝑛 ≠ 0.
Then there exists 𝑧 ∈ ℂ such that 𝑓 (𝑧) = 0.
7.2. TAYLOR SERIES 57
Proof Let assume that for all 𝑧 ∈ ℂ it holds
𝑓 (𝑧) ≠ 0
and let 𝐹 denote a function given by
1
𝐹 (𝑧) = .
𝑓 (𝑧)
Then obviously
1. 𝐹 is holomorphic on ℂ, i.e., for all 𝑧 ∈ ℂ it holds
𝑓 ′ (𝑧)
𝐹 ′ (𝑧) = − ,
𝑓 2 (𝑧)
2. 𝐹 is bounded on ℂ, i.e.,
1 1 1
lim 𝐹 (𝑧) = lim = = = 0.
𝑧→∞ 𝑧→∞
𝑧𝑛 (𝑎𝑛 + 𝑎𝑛−1 1𝑧 +⋯+ 𝑎0 𝑧1𝑛 ) ∞ (𝑎𝑛 ) ∞
Therefore the function 𝐹 on ℂ is constant (see Theorem 7.7). This is a contradiction

with a definition of 𝐹 .
Definition 7.3 Let 𝑓 denote a holomorphic function at a point 𝑧0 ∈ ℂ and let

𝑝 ∈ ℕ. The point 𝑧0 is called a 𝑝 – times root (or 𝑝 – times zero point) of a
function 𝑓 if
𝑓 (𝑧0 ) = 𝑓 ′ (𝑧0 ) = 𝑓 ′′ (𝑧0 ) = ⋯ = 𝑓 (𝑝−1) (𝑧0 ) = 0 ≠ 𝑓 (𝑝) (𝑧0 ).
Theorem 7.9 Let 𝑓 denote a holomorphic function at a point 𝑧0 ∈ ℂ and let

𝑓 (𝑧0 ) = 0. Then there exists 𝑈 (𝑧0 ) such that only exactly one of the following
properties holds
(i) 𝑓 is a zero function on 𝑈 (𝑧0 ),
(ii) 𝑓 (𝑧) ≠ 0 for all 𝑧 ∈ 𝑈 (𝑧0 ) ⧵ {𝑧0 }.
Proof Obviously, a function 𝑓 equals to a sum of its Taylor series (with a centre at
𝑧0 ) on some neighbourhood of 𝑧0 . In case that this series does not equal zero (i.e.,
𝑓 is not a zero function on any neighbourhood of 𝑧0 ), then there exists 𝑝 ∈ ℕ such
that 𝑧0 is a 𝑝 – times root of the function 𝑓 . In other words, on some neigbourhood
of the point 𝑧0 it holds that
∑
∞
𝑓 (𝑛) (𝑧0 ) ∑
∞
𝑓 (𝑛) (𝑧0 )
𝑓 (𝑧) = (𝑧 − 𝑧0 )𝑛 = (𝑧 − 𝑧0 )𝑝 (𝑧 − 𝑧0 )𝑛−𝑝 = (𝑧 − 𝑧0 )𝑝 𝜑(𝑧),
𝑛=𝑝
𝑛! 𝑛=𝑝
𝑛!
where a function
∑
∞
𝑓 (𝑛) (𝑧0 )
𝜑(𝑧) = (𝑧 − 𝑧0 )𝑛−𝑝
𝑛=𝑝
𝑛!
is holomorphic (and continuous) and non-zero at 𝑧0 . 5 Therefore, there exists 𝑈 (𝑧0 )

such that the function 𝜑 is non-zero in 𝑈 (𝑧0 ) and then for all 𝑧 ∈ 𝑈 (𝑧0 ) ⧵ {𝑧0 } we
have
𝑓 (𝑧) = (𝑧 − 𝑧0 )𝑝 𝜑(𝑧) ≠ 0.
Let us introduce an important corollary of Theorem 7.9.
Theorem 7.10 Let 𝑓 and 𝑔 denote holomorphic functions in a domain Ω ⊂ ℂ and

let 𝛾 denote a simple curve in Ω for which 𝑓 = 𝑔 on ⟨𝛾⟩. Then 𝑓 = 𝑔 on Ω.
Exercise 7.1 Using Theorem 7.10 prove that for all 𝑧 ∈ ℂ it holds that:
1. sin2 𝑧 + cos2 𝑧 = 1,
2. sin(2𝑧) = 2 sin 𝑧 cos 𝑧,

1 + cos(2𝑧) 1 − cos(2𝑧)
3. cos2 𝑧 = , sin2 𝑧 = ,
2 2
4. Re 𝑧 > 0 ⇒ ln (𝑧2 ) = 2 ln 𝑧.
5 𝑓 (𝑝) (𝑧0 )
𝜑(𝑧0 ) = 𝑝!
≠0
Chapter 8
Laurent series and classification

of singularities
8.1 Laurent series

Definition 8.1 A Laurent series with a centre at a point 𝑧0 ∈ ℂ is defined by
∑
∞
𝑎𝑛 (𝑧 − 𝑧0 )𝑛 , (8.1)
𝑛=−∞
where 𝑎𝑛 ∈ ℂ for all 𝑛 ∈ {… , −3, −2, −1, 0, 1, 2, 3, … }. A power series

∑
∞
𝑎𝑛 (𝑧 − 𝑧0 )𝑛
𝑛=0
is called a regular part of a Laurent series (8.1), a function series

∑
∞
∑
∞
1
−𝑛
𝑎−𝑛 (𝑧 − 𝑧0 ) = 𝑎−𝑛
𝑛=1 𝑛=1
(𝑧 − 𝑧0 )𝑛
is called a principal part of a Laurent series (8.1). We say that a Laurent se-
ries (8.1) is convergent on a set Ω ⊂ ℂ if their regular and principal part converge
on Ω. Then a function 𝑓 defined by 𝑓 (𝑧) = 𝑓1 (𝑧) + 𝑓2 (𝑧) on Ω, where 𝑓1 ( resp.
𝑓2 ) is a sum of the regular (resp. principal) part of Laurent series (8.1), is called a
sum of Laurent series (8.1).
Now, let us focus on convergence of Laurent series (8.1). First of all, let us
investigate a convergence of its principal part. If
1
𝜉=
𝑧 − 𝑧0
then
∑
∞
1 ∑ ∞
𝑎−𝑛 𝑛
= 𝑎−𝑛 𝜉 𝑛 ,
𝑛=1
(𝑧 − 𝑧0 ) 𝑛=1
59
60 CHAPTER 8. LAURENT SERIES AND SINGULARITIES
where on the right hand side there is a power series with a centre at 0 (in 𝜉). Let us
denote by 𝜌 its radius of convergence. Then it holds that 1
∑
∞
(i) if |𝜉| < 𝜌, then the series 𝑎−𝑛 𝜉 𝑛 is absolutely convergent,
𝑛=1
∑
∞
(ii) if |𝜉| > 𝜌, then the series 𝑎−𝑛 𝜉 𝑛 is divergent.
𝑛=1
If we define a number 𝑟 as
⎧ 1
⎪ if 0 < 𝜌 < +∞,
⎪ 𝜌
⎪
𝑟=⎨ 0 if 𝜌 = +∞,
⎪
⎪
⎪ +∞ if 𝜌 = 0,
⎩
then from the above it follows that
∑
∞
1
(i) if |𝑧 − 𝑧0 | > 𝑟, then the series 𝑎−𝑛 is absolutely convergent,
𝑛=1 (𝑧 − 𝑧0 )𝑛
∑
∞
1
(ii) if |𝑧 − 𝑧0 | < 𝑟, then the series 𝑎−𝑛 is divergent.
𝑛=1 (𝑧 − 𝑧0 )𝑛
Now, let us denote by 𝑅 a radius of convergence of the power series
∑
∞
𝑎𝑛 (𝑧 − 𝑧0 )𝑛 ,
𝑛=0
i.e., the regular part of Laurent series (8.1). Then exactly one possibility may be
fulfilled
𝑟 < 𝑅, 𝑟 = 𝑅, or 𝑟 > 𝑅.
1. If 𝑟 < 𝑅, then Laurent series (8.1) is absolutely convergent (and locally uni-
formly) on an annulus
𝑃 (𝑧0 , 𝑟, 𝑅) = {𝑧 ∈ ℂ ∶ 𝑟 < |𝑧 − 𝑧0 | < 𝑅}
and is divergent at every point of the set
{𝑧 ∈ ℂ ∶ |𝑧 − 𝑧0 | < 𝑟 or |𝑧 − 𝑧0 | > 𝑅}.
Moreover, one can demonstrate that a sum 𝑓 of Laurent series (8.1) is a holo-
morphic function on 𝑃 (𝑧0 , 𝑟, 𝑅) and that for all 𝑝 ∈ ℕ and 𝑧 ∈ 𝑃 (𝑧0 , 𝑟, 𝑅) it
holds that ( )
∑∞
d𝑝 (𝑧 − 𝑧0 )𝑛
(𝑝)
𝑓 (𝑧) = 𝑎𝑛 .
𝑛=−∞
d𝑧𝑝
1
See Theorem 7.2.
8.1. LAURENT SERIES 61
2. In the case that 𝑟 = 𝑅 Laurent series (8.1) is divergent in every point of the
set
{𝑧 ∈ ℂ ∶ |𝑧 − 𝑧0 | ≠ 𝑟 = 𝑅}.
3. Finally, if 𝑟 > 𝑅, then there does not exist any 𝑧 ∈ ℂ for which Laurent series
(8.1) is convergent.
Similarly as in Section 7.1 it has been shown that a sum of Laurent series (under
the assumption 𝑟 < 𝑅) is a holomorphic function on annulus 𝑃 (𝑧0 , 𝑟, 𝑅). Vice
versa, the following theorem states that every holomorphic function on annulus
𝑃 (𝑧0 , 𝑟, 𝑅) is a sum of some Laurent series.
Theorem 8.1 (holomorphic function expansion in Laurent series) Let 𝑓 be a holo-

morphic function on 𝑃 (𝑧0 , 𝑟, 𝑅), where 𝑧0 ∈ ℂ and 0 ≤ 𝑟 < 𝑅 ≤ +∞. Then there
∑
∞
exists exactly one Laurent series 𝑎𝑛 (𝑧−𝑧0 )𝑛 such that for every 𝑧 ∈ 𝑃 (𝑧0 , 𝑟, 𝑅)
𝑛=−∞
it holds that
∑
∞
𝑓 (𝑧) = 𝑎𝑛 (𝑧 − 𝑧0 )𝑛 .
𝑛=−∞
Moreover, if 𝜚 denotes some real number such that 𝑟 < 𝜚 < 𝑅, then the coefficients
of the above mentioned Laurent series (which is called a Laurent expansion of a
function 𝑓 ) are given by
1 𝑓 (𝑧)
𝑎𝑛 = d𝑧,
∫
2𝜋𝑖 𝛾 (𝑧 − 𝑧0 )𝑛+1
where
𝛾(𝑡) = 𝑧0 + 𝜚 𝑒𝑖𝑡 , 𝑡 ∈ [0, 2𝜋]

𝑛 ∈ {… , −3, −2, −1, 0, 1, 2, 3, … }.
Example 8.1 Find a Laurent expansion of a function

1
𝑓 (𝑧) =
(𝑧 − 1)(𝑧 − 2)
on every annulus with a centre at 𝑧0 = 0 on which 𝑓 is holomorphic.
Solution: Obviously, one has to find Laurent series of 𝑓 on the annuli
𝑃 (0, 0, 1), 𝑃 (0, 1, 2), 𝑃 (0, 2, +∞).
First of all, notice that for all 𝑧 ∈ ℂ ⧵ {1, 2} it holds

1 1
𝑓 (𝑧) = − .
𝑧−2 𝑧−1
Now, let us deal individually with each annulus.
1. From the implications

∞ ( )𝑛
1 1 1 1 ∑ 𝑧 ∑∞
1
|𝑧| < 2 ⇒ =− 𝑧 = − = − 𝑛+1 𝑧𝑛 ,
𝑧−2 21− 2 𝑛=0 2 𝑛=0 2
2
1 1 ∑
∞
|𝑧| < 1 ⇒ − = = 𝑧𝑛 ,
𝑧 − 1 1 − 𝑧 𝑛=0
one obtains 𝑧 ∈ 𝑃 (0, 0, 1) = {𝑧 ∈ ℂ ∶ 0 < |𝑧| < 1}
∞ ( )
∑ 1
𝑓 (𝑧) = 1 − 𝑛+1 𝑧𝑛 .
𝑛=0
2
Notice that we have found a Taylor series.
2. As we already know from the above part, for every 𝑧 ∈ ℂ such that
1 < |𝑧| < 2 it holds that
1 ∑ 1
∞
= − 𝑛+1 𝑧𝑛 .
𝑧 − 2 𝑛=0 2
Additionally, it holds that

∞ ( )
1∑ 1 𝑛 ∑
∞
1 1 1 1
|𝑧| > 1 ⇒ − =− 1
=− = − 𝑛+1
𝑧−1 𝑧 1− 𝑧 𝑛=0 𝑧 𝑛=0
𝑧
𝑧
therefore for every 𝑧 ∈ 𝑃 (0, 1, 2) = {𝑧 ∈ ℂ ∶ 1 < |𝑧| < 2} we get
∑
∞
1 ∑
∞
1
𝑓 (𝑧) = − 𝑧𝑛 + − .
𝑛=0
2𝑛+1 𝑛=1
𝑧𝑛
3. From the implication in the above part one obtains

∞ ( )
1 ∑ 2 𝑛 ∑ 2𝑛
∞
1 1 1
|𝑧| > 2 ⇒ = 2
= = .
𝑧−2 𝑧 1− 𝑧 𝑛=0 𝑧 𝑛=0
𝑧𝑛+1
𝑧
Therefore for all 𝑧 ∈ 𝑃 (0, 2, +∞) = {𝑧 ∈ ℂ ∶ 2 < |𝑧|} it holds that

∑∞
( 𝑛−1 ) 1
𝑓 (𝑧) = 2 − 1 𝑛.
𝑛=1
𝑧
Exercise 8.1 Find a Laurent expansion of a function

1
𝑓 (𝑧) =
(𝑧 − 1)(𝑧 − 2)
on each annulus with a centre at 𝑧0 = 1 on which 𝑓 is holomorphic.
8.2. ISOLATED SINGULARITIES AND THEIR CLASSIFICATION 63
8.2 Isolated singularities and their classification

Definition 8.2 A point 𝑧0 ∈ ℂ is called an isolated singularity of a function 𝑓 if
(i) function 𝑓 is not holomorphic at 𝑧0 ,
(ii) there exists a ring neighbourhood 𝑃 (𝑧0 ) on which 𝑓 is holomorphic.
If 𝑧0 is an isolated singularity of a function 𝑓 , then there exists a number 𝑅 ∈ ℝ+

such that 𝑓 is holomorphic on 𝑃 (𝑧0 , 𝑅) = 𝑃 (𝑧0 , 0, 𝑅) and therefore2 for all 𝑧 ∈
𝑃 (𝑧0 , 𝑅) it holds that
∑∞
𝑓 (𝑧) = 𝑎𝑛 (𝑧 − 𝑧0 )𝑛 .
𝑛=−∞
Depending on a number of non-zero coefficients of a principal part of this Lau-

rent series one can determine three cases:
(i) every coefficient of a principal part equals zero (i.e., 𝑎−𝑛 = 0 for all 𝑛 ∈ ℕ),
(ii) there exists at least one and at most finitely many non-zero coefficients of a
principal part (i.e., there exists 𝑛 ∈ ℕ such that 𝑎−𝑛 ≠ 0 and for all 𝑘 ∈
ℕ, 𝑘 > 𝑛 is 𝑎−𝑘 = 0),
(iii) there exist infinitely many non-zero coefficients of a principal part.
In the first case (i) we call a point 𝑧0 a removable singularity of a function 𝑓 ,

in the second case (ii) a point 𝑧0 is called a pole of order 𝑛 of a function 𝑓 ,3 and
in the third case (iii) a point 𝑧0 is called an essential singularity of a function 𝑓 .
Theorem 8.2 Let 𝑧0 ∈ ℂ be an isolated singularity of a function 𝑓 . Then
(i) 𝑧0 is a removable singularity of a function 𝑓 if and only if
lim 𝑓 (𝑧) ∈ ℂ,
𝑧→𝑧0
(ii) 𝑧0 is a pole of a fuction 𝑓 (resp. pole of order 𝑛 of a function 𝑓 ) if and only

if
lim 𝑓 (𝑧) = ∞
𝑧→𝑧0
( )
[ 𝑛
]
resp. if lim (𝑧 − 𝑧0 ) 𝑓 (𝑧) ∈ ℂ ⧵ {0} ,
𝑧→𝑧0
(iii) 𝑧0 is an essential singularity of a function 𝑓 if and only if lim 𝑓 (𝑧) does not
𝑧→𝑧0
exist.
2
See Theorem 8.1.
3
Pole of order 1 is also called a simple pole.
Theorem 8.3 (great Picard’s) Assume that 𝑧0 ∈ ℂ is an essential singularity of

a function 𝑓 . Then 𝑓 takes on any ring neighborhood of 𝑧0 all possible complex
values with at most a single exception. In other words, for all 𝑃 (𝑧0 ) there exists
𝑧 ∈ ℂ such that ( )
ℂ ⧵ {𝑧} ⊂ 𝑓 𝑃 (𝑧0 ) .
8.3 Laurent series with a centre ∞, classification of a point

∞
Definition 8.3 A Laurent series with a centre ∞ is given by
∑∞
𝑎𝑛
, (8.2)
𝑛=−∞
𝑧𝑛
where 𝑎𝑛 ∈ ℂ for every 𝑛 ∈ {… , −3, −2, −1, 0, 1, 2, 3, … }.

A power series
∑∞
𝑎−𝑛 𝑧𝑛
𝑛=1
is called a principal part of Laurent series (8.2), a function series
∑
∞
𝑎𝑛
𝑛=0
𝑧𝑛
is called a regular part of Laurent series (8.2). 4
Similarly as in Laurent series with a centre at 𝑧0 ∈ ℂ one can introduce a term of

convergence of a Laurent series with a centre ∞ and its sum. One can analogously
introduces an annulus of convergence, it this case of the form
{ }
1 1
𝑃 (∞, 𝑟, 𝑅) = 𝑧 ∈ ℂ ∶ < |𝑧| < .
𝑅 𝑟
Again, it holds that if Laurent series (8.2) is convergent on annulus 𝑃 (∞, 𝑟, 𝑅) ≠ ∅,
then a sum of the series on 𝑃 (∞, 𝑟, 𝑅) is a holomorphic function. It also holds as
an analogy of Theorem 8.1.
Theorem 8.4 Let 𝑓 be a holomorphic function on 𝑃 (∞, 𝑟, 𝑅) ≠ ∅. Then there

∑∞ 𝑎
𝑛
exists exactly one Laurent series 𝑛
such that for every 𝑧 ∈ 𝑃 (∞, 𝑟, 𝑅) it
𝑛=−∞ 𝑧
holds that
∑∞
𝑎𝑛
𝑓 (𝑧) = .
𝑛=−∞
𝑧𝑛
4
Notice that formally there is no difference between Laurent series with the centre 0 and Laurent
series with the centre ∞. If one needs to distinguish between these two cases it is necessary to mark
out the centre of the series or to establish its principal resp. regular part.
8.3. LAURENT SERIES WITH A CENTRE ∞ 65
Moreover, if 𝜚 is some real number such that

1 1
<𝜚< ,
𝑅 𝑟
it holds for coefficients that the above mentioned Laurent series
1 𝑓 (𝑧) 1
𝑎𝑛 = d𝑧 = 𝑓 (𝑧)𝑧𝑛−1 d𝑧,
∫
2𝜋𝑖 𝛾 𝑧−𝑛+1 2𝜋𝑖 ∫𝛾
where
𝛾(𝑡) = 𝜚 𝑒𝑖𝑡 , 𝑡 ∈ [0, 2𝜋],

𝑛 ∈ {… , −3, −2, −1, 0, 1, 2, 3, … }.
Definition 8.4 We call ∞ an isolated singularity of a function 𝑓 if there exists

𝑃 (∞) on which 𝑓 is holomorphic.
If ∞ is an isolated singularity of a function 𝑓 , then it is possible to expand 𝑓

on some 𝑃 (∞) in a Laurent series with a centre ∞, i.e., for all 𝑧 ∈ 𝑃 (∞)
∑∞
𝑎𝑛
𝑓 (𝑧) = .
𝑛=−∞
𝑧𝑛
Similarly as for finite isolated singularities, one can classify a point ∞ according
to the number of non-zero coefficients of a principal part of this Laurent series. An
analogy of Theorem 8.2 remains true.
Theorem 8.5 Let ∞ be an isolated singularity of a function 𝑓 . Then it holds that
(i) ∞ is a removable singularity of a function 𝑓 if and only if
lim 𝑓 (𝑧) ∈ ℂ,
𝑧→∞
(ii) ∞ is a pole of a function 𝑓 (resp. a pole of order 𝑛 of a function 𝑓 ) if and

only if
lim 𝑓 (𝑧) = ∞
𝑧→∞
( )
𝑓 (𝑧)
resp. if lim 𝑛 ∈ ℂ ⧵ {0} ,
𝑧→∞ 𝑧
(iii) ∞ is an essential singularity of a function 𝑓 if and only if lim 𝑓 (𝑧) does not
𝑧→∞
exist.
Chapter 9
Residue and the residue theorem
9.1 Residue of a function and it’s calculation

Definition 9.1 Let 𝑧0 ∈ ℂ (resp. ∞) be an isolated singularity of a function 𝑓
∑
∞
∑
∞
and let 𝑎𝑛 (𝑧 − 𝑧0 )𝑛 (resp. 𝑎𝑛 ∕𝑧𝑛 ) be a Laurent series of the function 𝑓
𝑛=−∞ 𝑛=−∞
on some ring neighbourhood of a point 𝑧0 (resp. ∞). A number 𝑎−1 (resp. −𝑎1 )
is called a residue of a function 𝑓 at the point 𝑧0 (resp. ∞), it is denoted by
res 𝑓 (𝑧0 ) (resp. res 𝑓 (∞)). 1
One could naturally ask why we call a number 𝑎−1 (resp. −𝑎1 ) a residue of
a function. To answer this question it is enough to realize that from the above
definition it follows that 2
( )
1 1 ∑
∞
res 𝑓 (𝑧0 ) = 𝑓 (𝑧)d𝑧 = 𝑎𝑛 (𝑧 − 𝑧0 )𝑛 d𝑧
2𝜋𝑖 ∫𝛾 2𝜋𝑖 ∫𝛾 𝑛=−∞
( )
1 1 ∑∞
𝑎𝑛
resp. res 𝑓 (∞) = − 𝑓 (𝑧)d𝑧 = − d𝑧.
2𝜋𝑖 ∫𝛾 2𝜋𝑖 ∫𝛾 𝑛=−∞
𝑧𝑛
Theorem 9.1 It holds
(i) if a point 𝑧0 ∈ ℂ is a removable singularity of a function 𝑓 , then3 res 𝑓 (𝑧0 ) = 0,
(ii) if 𝑓 is a holomorphic function at a point 𝑧0 ∈ ℂ and if a function 𝑔 has a

simple pole at 𝑧0 , then
res (𝑓 (𝑧)𝑔(𝑧)) = 𝑓 (𝑧0 ) res 𝑔(𝑧),

𝑧=𝑧0 𝑧=𝑧0
1
Notation res 𝑓 (𝑧) (resp. res 𝑓 (𝑧)) will be used sometimes.
𝑧=𝑧0 𝑧=∞
2
See Theorem 8.1 (resp. Theorem 8.4).
3
Warning! Let us assume a function 𝑓 (𝑧) = 1∕𝑧. Then ∞ is a removable singularity of a function
𝑓 but despite this it holds that res 𝑓 (∞) = −1 ≠ 0.
67
68 CHAPTER 9. RESIDUE AND THE RESIDUE THEOREM
(iii) if 𝑓 and 𝑔 are holomorphic functions at a point 𝑧0 ∈ ℂ and if the point 𝑧0 is

a simple root of a function 𝑔,4 then
( )
𝑓 (𝑧) 𝑓 (𝑧 )
res = ′ 0 ,
𝑧=𝑧0 𝑔(𝑧) 𝑔 (𝑧0 )
(iv) if 𝑧0 ∈ ℂ (resp. ∞) is a pole of 𝑘-order of a function 𝑓 , then

( 𝑘−1 )
1 d ( 𝑘
)
res 𝑓 (𝑧) = lim 𝑓 (𝑧)(𝑧 − 𝑧0 ) ,
𝑧=𝑧0 (𝑘 − 1)! 𝑧→𝑧0 d𝑧 𝑘−1
resp. ( )
(−1)𝑘 d 𝑘+1
res 𝑓 (𝑧) = lim 𝑧𝑘+2 𝑘+1 𝑓 (𝑧) ,
𝑧=∞ (𝑘 + 1)! 𝑧→∞ d𝑧
(v) if 𝑓 is a holomorphic function on ℂ⧵{𝑧1 , 𝑧2 , … , 𝑧𝑛 } where 𝑧1 , 𝑧2 , … , 𝑧𝑛 ∈ ℂ
are different isolated singularities of a function 𝑓 , then
∑
𝑛
res 𝑓 (∞) + res 𝑓 (𝑧𝑖 ) = 0.
𝑖=1
Exercise 9.1 Prove Theorem 9.1.
Example 9.1 Find

( )
1
1. res 𝑧2 sin ,
𝑧=0 𝑧
𝑧3 sin 𝑧
2. res𝜋 ,
𝑧= 4 cos(2𝑧)
1
3. res .
𝑧=2𝜋𝑖 (𝑒𝑧 − 1)2
Solution:
Ad 1. For all 𝑧 ∈ ℂ ⧵ {0} it holds that
1 ∑∞
1 1 ∑∞
(−1)𝑛 1
𝑧2 sin = 𝑧2 (−1)𝑛 2𝑛+1
= 2𝑛−1
,
𝑧 𝑛=0
(2𝑛 + 1)! 𝑧 𝑛=0
(2𝑛 + 1)! 𝑧
therefore ( )
1 1 1
res 𝑧2 sin =− =− .
𝑧=0 𝑧 3! 6
Ad 2. Obviously, 𝜋∕4 is a simple root of a function
𝑔(𝑧) = cos(2𝑧),
4
I.e., 𝑔(𝑧0 ) = 0 ≠ 𝑔 ′ (𝑧0 ).
9.2. THE RESIDUE THEOREM 69
therefore 5 √
[ 3 ]
𝑧3 sin 𝑧 𝑧 sin 𝑧 2𝜋 3
res𝜋 = =− .
𝑧= 4 cos(2𝑧) −2 sin(2𝑧) 𝑧= 𝜋 256
4
Ad 3. In this case, a investigated function has a pole of the second order at 2𝜋𝑖.
Therefore 6
[ ]′
1 1 (𝑧 − 2𝜋𝑖)2
res = lim = ⋯ = −1.
𝑧=2𝜋𝑖 (𝑒𝑧 − 1)2 1! 𝑧→2𝜋𝑖 (𝑒𝑧 − 1)2
9.2 The residue theorem

Theorem 9.2 (Residue) Assume that Ω ⊂ ℂ denotes a simply connected domain
and 𝛾 is a simple closed piece-wise smooth positively oriented curve in Ω. Let 𝑓
be a holomorphic function on Ω ⧵ {𝑧1 , 𝑧2 , … , 𝑧𝑛 } where 𝑧1 , 𝑧2 , … , 𝑧𝑛 ∈ int 𝛾 are
different isolated singularities of a function 𝑓 . Hence it holds that
∑
𝑛
𝑓 (𝑧)d𝑧 = 2𝜋𝑖 res 𝑓 (𝑧𝑖 ) .
∫𝛾
𝑖=1
Proof The proof is a simple corollary of the definition of residue and Theorems 5.5
and 8.1.
Example 9.2 Find

1
𝑧2 sin d𝑧,
∫𝛾 𝑧+1
where
𝛾(𝑡) = 2𝑒𝑖𝑡 , 𝑡 ∈ [0, 2𝜋].
Solution: It follows from Theorem 9.2 that
1 1
𝑧2 sin d𝑧 = 2𝜋𝑖 res 𝑧2 sin .
∫𝛾 𝑧+1 𝑧=−1 𝑧+1
Since it holds that for every 𝑧 ∈ ℂ ⧵ {−1}
2 1 ( 2
)∑∞
(−1)𝑛 1
𝑧 sin = (𝑧 + 1) − 2(𝑧 + 1) + 1
𝑧+1 𝑛=0
(2𝑛 + 1)! (𝑧 + 1)2𝑛+1
is 7
( )
2 1 1 (−1)1 (−1)0 5
𝑧 sin d𝑧 = 2𝜋𝑖 res 𝑧2 sin = 2𝜋𝑖 1 +0+1 = 𝜋𝑖.
∫𝛾 𝑧+1 𝑧=−1 𝑧+1 3! 1! 3
5
See Theorem 9.1 – part(iii).
6
See Theorem 9.1 – part(iv).
7
9.3 Calculation integrals of real variable function with the

aid of the residue theorem
2𝜋
a) Integrals of the form ∫0 𝑅(sin 𝑥, cos 𝑥) d𝑥,
where 𝑅 ∶ ℝ2 → ℝ denotes a rational function of two variables and a function
𝑥 → 𝑅(sin 𝑥, cos 𝑥)) is continuous on [0, 2𝜋].
Let us use the substitution
𝑒𝑖𝑥 = 𝑧.
Then one obtains
1 1
𝑧− 𝑧
𝑧+ 𝑧 1
sin 𝑥 = , cos 𝑥 = , d𝑧 = 𝑒𝑖𝑥 𝑖 d𝑥, that is d𝑥 = d𝑧,
2𝑖 2 𝑖𝑧
and therefore,
( 1 1 )
2𝜋 𝑧− 𝑧
𝑧+ 𝑧 1
𝑅(sin 𝑥, cos 𝑥) d𝑥 = 𝑅 , d𝑧, (9.1)
∫0 ∫𝛾 2𝑖 2 𝑖𝑧
where
𝛾(𝑥) = 𝑒𝑖𝑥 , 𝑥 ∈ [0, 2𝜋].
Then equality (9.1) we derived by formal substitution directly follows from The-
orem 5.2. One can compute the integral on the right hand side of the above equality
using Theorem 9.2.
Example 9.3
2𝜋
d𝑥 1 1
d𝑥 = d𝑧 =
∫0 5
− cos 𝑥 ∫𝛾 5 𝑧+ 1𝑧 𝑖𝑧
4
4
− 2
( )
2 d𝑧 2 1
= − = − 2𝜋𝑖 res =
𝑖 ∫𝛾 (𝑧 − 2)(𝑧 − 1 ) 𝑖 𝑧= 12 (𝑧 − 2)(𝑧 − 12 )
2
1 8
= −4𝜋 1
= 𝜋
2
−2 3
𝛾(𝑥) = 𝑒𝑖𝑥 , 𝑥 ∈ [0, 2𝜋]
+∞
b) Integrals of the form ∫−∞ 𝑃 (𝑥)∕𝑄(𝑥) d𝑥, where 𝑃 , 𝑄 ∶ ℝ → ℝ are polyno-
mials satisfying
(i) 𝑄 does not have any real root,
(ii) the degree of the polynomial 𝑄 is at least 2 greater than the degree of the
polynomial 𝑃 .
9.3. RESIDUE THEOREM APPLICATION ON REAL VARIABLE FUNCTIONS71
From the above mentioned assumptions it coincides that

𝑘 +∞
𝑃 (𝑧) 𝑃 (𝑥) 𝑃 (𝑥)
lim d𝑧 = lim d𝑥 = d𝑥,
𝑘→+∞ ∫𝛼 𝑄(𝑧) 𝑘→+∞ ∫ −𝑘 𝑄(𝑥) ∫ −∞ 𝑄(𝑥)
𝑘
where
𝛼𝑘 (𝑡) = 𝑡, 𝑡 ∈ [−𝑘, 𝑘]
and that
𝑃 (𝑧)
lim d𝑧 = 0,
𝑘→+∞ ∫𝛽 𝑄(𝑧)
𝑘
where
𝛽𝑘 (𝑡) = 𝑘𝑒𝑖𝑡 , 𝑡 ∈ [0, 𝜋].
Therefore it holds that
+∞
𝑃 (𝑥) 𝑃 (𝑧)
d𝑥 = lim d𝑧,
∫−∞ 𝑄(𝑥) 𝑘→+∞ ∫𝛾 𝑄(𝑧)
𝑘
where8
⎧
⎪ 𝛼𝑘 (𝑡 + 𝑘), je-li 𝑡 ∈ [−2𝑘, 0),
𝛾𝑘 (𝑡) = ⎨
⎪ 𝛽𝑘 (𝑡), if 𝑡 ∈ [0, 𝜋].
⎩
Now, let 𝑈 (0, 𝑟) ⊂ ℂ denote a disk big enough to include all roots of the integral
𝑄 (one certainly exists!). Then it holds that for every real number 𝑘 > 𝑟
𝑃 (𝑧) 𝑃 (𝑧)
d𝑧 = d𝑧,
∫𝛾𝑘 𝑄(𝑧) ∫𝛾𝑟 𝑄(𝑧)
and therefore,
+∞
𝑃 (𝑥) 𝑃 (𝑧) 𝑃 (𝑧)
d𝑥 = lim d𝑧 = d𝑧.
∫−∞ 𝑄(𝑥) 𝑘→+∞ ∫𝛾 𝑄(𝑧)
𝑘
∫ 𝛾𝑟 𝑄(𝑧)
Now let us apply Theorem 9.2

( )
+∞
𝑃 (𝑥) 𝑃 (𝑧) ∑ 𝑃 (𝑧)
d𝑥 = d𝑧 = 2𝜋𝑖 res .
∫−∞ 𝑄(𝑥) ∫𝛾𝑟 𝑄(𝑧) 𝑧=𝑧𝑘 𝑄(𝑧)
𝑧𝑘 ∈ℂ∶
𝑄(𝑧𝑘 )=0,
Im 𝑧𝑘 >0

+∞
1
d𝑥.
∫−∞ (𝑥2 + 2𝑥 + 2)2
8
Draw geometrical images of curves 𝛼𝑘 , 𝛽𝑘 , 𝛾𝑘 .
Solution:
+∞ +∞
1 d𝑥
d𝑥 = d𝑥 =
∫−∞ 2
(𝑥 + 2𝑥 + 2) 2 ∫ −∞ ((𝑥 − (−1 + 𝑖)) (𝑥 − (−1 − 𝑖)))
2
1
= 2𝜋𝑖 res =
𝑧=−1+𝑖 ((𝑧 − (−1 + 𝑖)) (𝑧 − (−1 − 𝑖)))2
( )′
1
= 2𝜋𝑖 lim =
𝑧→−1+𝑖 (𝑧 − (−1 − 𝑖))2
[ ]
1 𝜋
= 2𝜋𝑖 −2 3
= .
(𝑧 − (−1 − 𝑖)) 𝑧=−1+𝑖 2
Chapter 10
Exercise
Exercise 10.1 Find the real and imaginary parts of a complex number
1. 𝑧 = (1 + 𝑖)(3 − 2𝑖),
2 − 3𝑖
2. 𝑧 = ,
3 + 4𝑖
1+𝑖
3. 𝑧 = ,
1−𝑖
2 − 4𝑖
4. 𝑧 = 2𝑖 − .
2
Exercise 10.2 Find a polar form of a complex number
√
1. 𝑧 = −1 + 3𝑖,
2. 𝑧 = 𝑖,
3. 𝑧 = −8,
√
4. 𝑧 = −1 − 3𝑖,
2+𝑖
5. 𝑧 = ,
3 − 2𝑖
3−𝑖
6. 𝑧 = .
2+𝑖
Exercise 10.3 Prove Moivre’s theorem:
( )𝑛
(∀𝑛 ∈ ℕ) (∀𝜑 ∈ ℝ) ∶ cos 𝜑 + 𝑖 sin 𝜑 = cos(𝑛𝜑) + 𝑖 sin(𝑛𝜑)
using mathematical induction.
Exercise 10.4 Let 𝜑 ∈ ℝ. Express sin(4𝜑) and cos(4𝜑) using sin 𝜑 a cos 𝜑.
73
74 CHAPTER 10. EXERCISE
( )24
1−𝑖
Exercise 10.5 Find Re 𝑧 and Im 𝑧 for 𝑧 = √ .
1 + 3𝑖
Exercise 10.6 Find Arg 𝑧 and arg 𝑧 for

(√ )126
1. 𝑧 = 3+𝑖 ,
2. 𝑧 = (1 + 𝑖)137 ,
3. 𝑧 = −1 − 5𝑖.
Exercise 10.7 Illustrate in the Gauss plane a set
1. {𝑧 ∈ ℂ∞ ∶ Re 𝑧 ≤ 1},
2. {𝑧 ∈ ℂ∞ ∶ Re(𝑧2 ) = 2},
3. {𝑧 ∈ ℂ∞ ∶ Im 1∕𝑧 = 1∕4},
4. {𝑧 ∈ ℂ∞ ∶ | Im 𝑧| < 1},
5. {𝑧 ∈ ℂ∞ ∶ |𝑧| = Re 𝑧 + 1},
6. {𝑧 ∈ ℂ∞ ∶ |𝑧 − 2| = |1 − 2𝑧|},
7. {𝑧 ∈ ℂ∞ ∶ |(𝑧 − 2)∕(𝑧 − 3)| = 1},
8. {𝑧 ∈ ℂ∞ ∶ |1 + 𝑧| < |1 − 𝑧|},
9. {𝑧 ∈ ℂ∞ ∶ |𝑧 + 1| = 2|𝑧 − 1|},
10. {𝑧 ∈ ℂ∞ ∶ 2 < |𝑧 + 2 − 3𝑖| < 4},
11. {𝑧 ∈ ℂ∞ ∶ 𝜋∕4 ≤ arg (𝑧 + 2𝑖) ≤ 𝜋∕2},
12. {𝑧 ∈ ℂ∞ ∶ |𝑧| + Re 𝑧 ≤ 1 ∧ 𝜋∕2 ≤ arg 𝑧 ≤ 𝜋∕4}.
Exercise 10.8 Let 𝑧1 , 𝑧2 ∈ ℂ ⧵ {0}. Prove the following implications:
⎫
𝜑1 ∈ Arg 𝑧1 ⎪
1. ⎬ ⇒ 𝜑1 + 𝜑2 ∈ Arg (𝑧1 𝑧2 ),
𝜑2 ∈ Arg 𝑧2 ⎪
⎭
⎫ ( )
𝜑1 ∈ Arg 𝑧1 ⎪ 𝑧1
2. ⎬ ⇒ 𝜑1 − 𝜑2 ∈ Arg 𝑧 .
𝜑2 ∈ Arg 𝑧2 ⎪ 2
⎭
75
Exercise 10.9 Calculate (if it exists) a limit for
1. lim (3 − 4𝑖)𝑛 ,
( )
𝑖
2. lim (−1)𝑛 + ,
𝑛
( )𝑛
1+𝑖
3. lim √ ,
2
( √ )6𝑛
1 − 3𝑖
4. lim .
2
Exercise 10.10 Let (𝑧𝑛 ) be a sequence of complex numbers. Prove the following
propositions:
1
1. 𝑧𝑛 → 0 ⇔ 𝑧𝑛
→ ∞,
⎫
|𝑧𝑛 | → 𝑟 ∈ ℝ ⎪ ( )
2. ⎬ ⇒ 𝑧𝑛 → 𝑟 cos 𝜑 + 𝑖 sin 𝜑 ,
arg 𝑧𝑛 → 𝜑 ∈ ℝ⎪
⎭
and prove that in a second case an opposite implication does not hold.
Exercise 10.11 Find all 𝑧 ∈ ℂ∞ for which it holds that
1. 𝑧3 = 1,
2. 𝑧2 = 𝑖,
3. 𝑧2 = 24𝑖 − 7,
( )
𝑧−1 2
4. = 2𝑖,
𝑧+1
5. 𝑧4 = −1,
6. 𝑧3 = 𝑖 − 1,
7. 𝑧5 = 1,
8. 𝑧2 = −11 + 60𝑖,
9. 𝑧2 = 3 + 4𝑖.
{ }
1
Exercise 10.12 Find a set 𝑀 = 𝑧
∶ 𝑧 ∈ Ω if
1. Ω = {𝑧 ∈ ℂ ∶ arg 𝑧 = 𝛼}, 𝛼 ∈ (−𝜋, 𝜋],

2. Ω = {𝑧 ∈ ℂ ∶ |𝑧 − 1| = 1},
3. Ω = {𝑧 ∈ ℂ ∶ Re 𝑧 = Im 𝑧},
4. Ω = {𝑥 + 𝑖𝑦 ∈ ℂ ∶ 𝑥 = 1},
5. Ω = {𝑥 + 𝑖𝑦 ∈ ℂ ∶ 𝑦 = 0}.
Exercise 10.13 Find a set 𝑀 = {𝑓 (𝑧) ∶ 𝑧 ∈ Ω} if

{ }
𝜋
1. Ω = 𝑧 ∈ ℂ ∶ |arg 𝑧| ≤ , 𝑓 (𝑧) = 𝑧2 ,
6
{ }
𝜋
2. Ω = 𝑧 ∈ ℂ ∶ | Im 𝑧| < , 𝑓 (𝑧) = 𝑒𝑧 ,
2
3. Ω = {𝑧 ∈ ℂ ∶ 0 < Re 𝑧 < 𝜋 ∧ Im 𝑧 > 0}, 𝑓 (𝑧) = 𝑒𝑖𝑧 ,
4. Ω = {𝑧 ∈ ℂ ∶ Im 𝑧 = 1∕2}, 𝑓 (𝑧) = 𝑧2 .
Exercise 10.14 Calculate
1. sin(2 − 3𝑖),
2. cos 𝑖,
3. cosh 𝑖,
4. Ln (−5 + 3𝑖) a ln(−5 + 3𝑖),

√ √
5. Ln (−4 − 3𝑖) a ln(−4 − 3𝑖),
6. Ln (𝑖𝑒2 ).
Exercise 10.15 Find all 𝑧 ∈ ℂ for which it holds that
1. sin 𝑧 = 3,
√
3
2. cos 𝑧 = ,
2
3. sin 𝑧 + cos 𝑧 = 2,
4. sin 𝑧 − cos 𝑧 = 3,
5. 𝑧2 + 2𝑧 + 9 + 6𝑖 = 0.
1. 2𝑖 ,
√
2. (−2) 2,
77
( )1+𝑖
1−𝑖
3. √ ,
2
3
4. 𝑖 4 ,
√
5. (−1) 3 ,
√
6. (− 3𝑖 + 1)−3 .
Exercise 10.17 Find the real and imaginary parts of a function 𝑓 ∶ ℂ → ℂ

defined by
1. 𝑓 (𝑧) = sin 𝑧,
2. 𝑓 (𝑧) = 𝑧2 cos 𝑧,
3. 𝑓 (𝑧) = 𝑧3 + 5𝑧 − 1,
4. 𝑓 (𝑧) = |𝑧| 𝑧,
5. 𝑓 (𝑧) = 𝑧2 𝑧,
1
6. 𝑓 (𝑧) = .
𝑧
Exercise 10.18 Find out if the function 𝑓 (𝑧) = 𝑧3 is injective on a set Ω in case
that
1. Ω = {𝑧 ∈ ℂ ∶ Re 𝑧 > 0},
{ }
𝜋
2. Ω = 𝑧 ∈ ℂ ∶ arg 𝑧 ∈ ⟨0, ) .
4
Exercise 10.19 Find a limit (if it exists):
Re 𝑧
1. lim ,
𝑧→0 𝑧
Im(𝑧2 )
2. lim ,
𝑧→0 𝑧𝑧
𝑧 Im 𝑧
3. lim ,
𝑧→0 |𝑧|
𝑧2
4. lim ,
𝑧→0 |𝑧|2
𝑧3
5. lim ,
𝑧→0 |𝑧|2
𝑧2 + 𝑧(2 − 𝑖) − 2𝑖
6. lim ,
𝑧→𝑖 𝑧2 + 1
Re 𝑧
7. lim .
𝑧→0 1 + |𝑧|
Exercise 10.20 Draw a set ⟨𝜑⟩ = {𝜑(𝑡) ∶ 𝑡 ∈ 𝐷𝜑} if
1. 𝜑(𝑡) = 1 − 𝑖𝑡, 𝐷𝜑 = [0, 2],
2. 𝜑(𝑡) = 𝑡 − 𝑖𝑡2 , 𝐷𝜑 = [−1, 2],
3. 𝜑(𝑡) = 1 + 𝑒−𝑖𝑡 , 𝐷𝜑 = [0, 2𝜋],
4. 𝜑(𝑡) = 𝑒2𝑖𝑡 − 1, 𝐷𝜑 = [0, 2𝜋],

{
𝑒𝑖𝜋𝑡 , 𝑡 ∈ [0, 1],
5. 𝜑(𝑡) =
𝑡 − 2, 𝑡 ∈ [1, 3],
{
𝑒𝑖𝑡 , 𝑡 ∈ [−𝜋∕2, 𝜋],
6. 𝜑(𝑡) =
3𝑡∕𝜋 − 4, 𝑡 ∈ [𝜋, 2𝜋].
Exercise 10.21 Find a curve 𝜑 ∶ ℝ → ℂ such that ⟨𝜑⟩ = Ω) if

1. Ω = {𝑧 ∈ ℂ ∶ |𝑧 − 2 + 3𝑖| = 2},
2. Ω is a line with endpoints 𝑎, 𝑏 ∈ ℂ, 𝑎 ≠ 𝑏,
3. Ω = {𝑧 ∈ ℂ ∶ Re 𝑧 = 2 Im 𝑧},
( )
4. Ω = {𝑧 ∈ ℂ ∶ Re 1𝑧 = 2}.
Exercise 10.22 Draw a set Ω and find out if it is a domain and an open set, where
1. Ω = {𝑧 ∈ ℂ ∶ |𝑧 − 𝑖| < 1 ∨ |𝑧 + 𝑖| < 1},
2. Ω = {𝑧 ∈ ℂ ∶ |𝑧 − 1| < 1 ∧ |𝑧 − 2| < 2},
3. Ω = {𝑧 ∈ ℂ ∶ |𝑧 − 1| < |𝑧 + 1|},
4. Ω = {𝑧 ∈ ℂ ∶ |𝑧 + 1| > 2|𝑧|},
5. Ω = {𝑧 ∈ ℂ ∶ 1 < |𝑧| < 2},

{ }
6. Ω = 𝑧 ∈ ℂ ∶ |𝑧| < 1 ∧ arg 𝑧 ∈ (−𝜋, 𝜋] ⧵ {0} ,
7. Ω = {𝑧 ∈ ℂ ∶ |2𝑧| < |1 + 𝑧2 |}.
Exercise 10.23 Find all points at which a function 𝑓 has a derivative and is holo-
morphic if
1. 𝑓 (𝑧) = Re 𝑧,
2. 𝑓 (𝑧) = |𝑧2 |,
79
3. 𝑓 (𝑧) = 𝑧𝑒𝑧 ,
4. 𝑓 (𝑧) = 𝑧|𝑧|,
Re 𝑧
5. 𝑓 (𝑧) = ,
𝑧
6. 𝑓 (𝑧) = 𝑧2 𝑧,
7. 𝑓 (𝑧) = 𝑧2 + 2𝑧 − 1.
Exercise 10.24 Find out if a function Φ is harmonic in a domain Ω, where
1. Φ(𝑥, 𝑦) = 𝑥2 − 𝑦2 + 2020, Ω = ℂ,
𝑥
2. Φ(𝑥, 𝑦) = 𝑥2 +𝑦2
+ 𝑥2 − 𝑦2 + 𝑥 − 𝑦, Ω = ℂ ⧵ {0}.
Exercise 10.25 Find (if it exists) a holomorphic function 𝑓 = 𝑢 + 𝑖𝑣 on Ω if
1. 𝑢(𝑥, 𝑦) = 𝑥3 − 3𝑥𝑦2 − 2𝑦, Ω = ℂ,

𝑥
2. 𝑢(𝑥, 𝑦) = , Ω = ℂ ⧵ {0},
𝑥2 + 𝑦2
3. 𝑢(𝑥, 𝑦) = 3𝑥2 − 𝑦2 + 3𝑥 + 𝑦, Ω = ℂ,
𝑦
4. 𝑢(𝑥, 𝑦) = 𝑥2 − 𝑦2 + 5𝑥 + 𝑦 − , Ω = ℂ ⧵ {0}.
𝑥2 + 𝑦2
Exercise 10.26 Let 𝑢(𝑥, 𝑦) = 𝑥3 − 3𝑥𝑦2 − 2𝑦 + 2. Find (if it exists) a holomorphic

function 𝑓 = 𝑢 + 𝑖𝑣 on ℂ for which it holds that
1. 𝑓 (0) = 𝑖,
2. 𝑓 (1) = 3 − 𝑖.
Exercise 10.27 Find (if it exists) a holomorphic function 𝑓 = 𝑢 + 𝑖𝑣 on Ω if
1. 𝑣(𝑥, 𝑦) = −3𝑥𝑦2 + 𝑥3 + 5, Ω = ℂ,
𝑦
2. 𝑣(𝑥, 𝑦) = arctan , Ω = {𝑧 ∈ ℂ ∶ Re 𝑧 > 0}.
𝑥
Exercise 10.28 Let 𝑣(𝑥, 𝑦) = 1+arctan 𝑥𝑦 . Find (if it exists) a holomorphic function
𝑓 = 𝑢 + 𝑖𝑣 in a domain {𝑧 ∈ ℂ ∶ Re 𝑧 > 0} for which it holds that
1. 𝑓 (3) = ln 3 + 6 + 𝑖,
2. 𝑓 (𝑒) = 1 − 𝑖.
Exercise 10.29 Prove that a function
𝑣(𝑥, 𝑦) = ln(𝑥2 + 𝑦2 )
is harmonic in two-times connected domain ℂ ⧵ {0} and that there does not exist a
function 𝑢 ∶ ℝ2 → ℝ such that a function 𝑓 = 𝑢 + 𝑖𝑣 is holomorphic on ℂ ⧵ {0}.
Exercise 10.30 Find a rotation angle and an explosion coefficient of a function 𝑓

at a point 𝑧0 if
𝜋
1. 𝑓 (𝑧) = 𝑒𝑧 , 𝑧0 = −1 − 𝑖,
2
2. 𝑓 (𝑧) = 𝑧3 , 𝑧0 = −3 + 4𝑖,
𝑧+𝑖
3. 𝑓 (𝑧) = , 𝑧 = 2𝑖.
𝑧−𝑖 0
Exercise 10.31 Find all points of the Gauss plane in which a function 𝑓 is a con-
traction
2
1. 𝑓 (𝑧) = ,
𝑧
2. 𝑓 (𝑧) = ln(𝑧 + 4).
Exercise 10.32 Draw sets Ω and 𝑓 (Ω) = {𝑓 (𝑧) ∶ 𝑧 ∈ Ω} if 1
1. Ω = 𝑈 (1, 2), 𝑓 (𝑧) = 1 − 2𝑖𝑧,
2. Ω = {𝑧 ∈ ℂ ∶ Re 𝑧 < 1}, 𝑓 (𝑧) = (1 + 𝑖)𝑧 + 1,

1
3. Ω = 𝑈 (1, 2), 𝑓 (𝑧) = ,
𝑧
2𝑖𝑧
4. Ω = 𝑈 (1, 2), 𝑓 (𝑧) = ,
𝑧+3
𝑧−1
5. Ω = 𝑈 (1, 2), 𝑓 (𝑧) = ,
2𝑧 − 6
1
6. Ω = {𝑧 ∈ ℂ ∶ Re 𝑧 < 1}, 𝑓 (𝑧) = ,
𝑧
𝑧
7. Ω = {𝑧 ∈ ℂ ∶ Re 𝑧 < 1}, 𝑓 (𝑧) = ,
𝑧−1+𝑖
1
Use (and also prove) the following statement:
⎫
𝑓 is conformal in a domain Ω ⊂ ℂ∞ ,⎪
⎬ ⇒ 𝑓 (𝐴 ∩ 𝐵) = 𝑓 (𝐴) ∩ 𝑓 (𝐵).
𝐴, 𝐵 ⊂ Ω ⎪
⎭
81
𝑧
8. Ω = {𝑧 ∈ ℂ ∶ Re 𝑧 < 1}, 𝑓 (𝑧) = ,
𝑧−2
1
9. Ω = {𝑧 ∈ ℂ ∶ Re 𝑧 < 0 ∧ Im 𝑧 < 0}, 𝑓 (𝑧) = ,
𝑧
𝑧−1
10. Ω = {𝑧 ∈ ℂ ∶ Re 𝑧 > 0 ∧ Im 𝑧 > 0}, 𝑓 (𝑧) = ,
𝑧+1
𝑧−𝑖
11. Ω = {𝑧 ∈ ℂ ∶ −1 < Re 𝑧 < 0 ∧ Im 𝑧 < 0}, 𝑓 (𝑧) = ,
𝑧+𝑖
𝑧
12. Ω = {𝑧 ∈ ℂ ∶ |𝑧| < 1 ∧ Re 𝑧 < 0 ∧ Im 𝑧 > 0}, 𝑓 (𝑧) = .
𝑧−𝑖
Exercise 10.33 Find a linear fractional function 𝑓 such that
1. 𝑓 (−1) = 0, 𝑓 (𝑖) = 2𝑖, 𝑓 (1 + 𝑖) = 1 − 𝑖,
2. 𝑓 (𝑖) = ∞, 𝑓 (6) = 0, 𝑓 (∞) = 3,
3. 𝑓 (0) = 𝑖, 𝑓 (𝑖) = 0, 𝑓 (−1) = −𝑖.
Exercise 10.34 Find a linear function mapping the square with vertices 0, 1 −
𝑖, 2, 1 + 𝑖 on the square with vertices 1 + 𝑖, −1 + 𝑖, −1 − 𝑖, 1 − 𝑖.
Exercise 10.35 Let

Ω = {𝑧 ∈ ℂ ∶ Re 𝑧 > Im 𝑧}.
Find a linear fractional function 𝑓 such that 𝑓 (Ω) = 𝑈 (0, 1).
Exercise 10.36 Find a conformal function which maps the domain
{𝑧 ∈ ℂ ∶ |𝑧| < 1 ∧ Re 𝑧 > 0}
on the domain
{𝑧 ∈ ℂ ∶ Im 𝑧 > 0}.
Exercise 10.37 Let
Ω = {𝑧 ∈ ℂ ∶ Re 𝑧 > 0 ∧ Im 𝑧 < 0}.
Find a linear fractional function 𝑓 such that
𝑓 (Ω) = {𝑧 ∈ ℂ ∶ |𝑧| < 1 ∧ Re 𝑧 < 0}.
Exercise 10.38 Find a conformal function which maps the domain
{𝑧 ∈ ℂ ∶ Re 𝑧 > Im 𝑧 > 0}
on the domain 𝑈 (0, 1).

Exercise 10.39 Find images of lines parallel to a real, resp. imaginary axis under
1
a function 𝑓 (𝑧) = .
𝑧
Exercise 10.40 Find images of the sets
𝑀𝛼 = {𝑧 ∈ ℂ ∶ arg 𝑧 = 𝛼}, 𝑁𝑟 = {𝑧 ∈ ℂ ∶ |𝑧| = 𝑟},
where 𝛼 ∈ (−𝜋, 𝜋], 𝑟 ∈ ℝ+ under the function 𝑓 (𝑧) = ln 𝑧.
Exercise 10.41 Find

|𝑧| d𝑧
∫𝛾
if
[ ]
⎧3𝑒𝑖𝑡 , 𝑡 ∈ 0, 𝜋 ,
⎪ 2 [ ]
⎪( 𝜋 ) 𝜋 𝜋
𝛾(𝑡) = ⎨𝑖 3 + − 𝑡 , 𝑡 ∈ , +3 ,
⎪ 2 [ 2 2 ]
⎪𝑡 − − 3, 𝑡 ∈ 𝜋 + 3, 𝜋 + 6 .
𝜋
⎩ 2 2 2

𝑧3 d𝑧
∫𝛾
if [ ]
⎧𝑒𝑖𝑡 , 𝑡 ∈ − 𝜋 , 𝜋 ,
⎪ 2
⎪3
𝛾(𝑡) = ⎨ 𝑡 − 4, 𝑡 ∈ [𝜋, 2𝜋] ,
⎪𝜋
⎪− 2 + 𝑖 𝑡 + 6 + 2𝑖, 𝑡 ∈ [2𝜋, 3𝜋] .
⎩ 𝜋

|𝑧|𝑧 d𝑧,
∫𝛾
where 𝛾 denotes a simple piece-wise smooth positively oriented curve such that ⟨𝛾⟩
is a border of a set
{𝑧 ∈ ℂ ∶ |𝑧| < 2 ∧ Im 𝑧 > 0}.
Exercise 10.44 Using Cauchy’s integral formulas calculate the following integrals
2
1.
𝑧2 + 𝑖
d𝑧, where 𝑘 = {𝑧 ∈ ℂ ∶ |𝑧 − 2𝑖| = 1},
∫𝑘 𝑧
2
Notice that by a formula ∫𝑘 𝑓 (𝑧) d𝑧, where 𝑘 ⊂ ℂ, one means ∫𝛾 𝑓 (𝑧) d𝑧, where 𝛾 is a simple
closed piece-wise smooth positively oriented curve such that ⟨𝛾⟩ = 𝑘.
83
2.
sin 𝑧
d𝑧, where 𝑘 = {𝑧 ∈ ℂ ∶ |𝑧 + 𝑖| = 1},
∫𝑘 𝑧 + 𝑖
3.
sin 𝑧
d𝑧, where 𝑘 = {𝑧 ∈ ℂ ∶ |𝑧| = 3},
∫𝑘 𝑧2 − 7𝑧 + 10
4.
sin 𝑧
d𝑧, where 𝑘 = {𝑧 ∈ ℂ ∶ |𝑧| = 3},
∫𝑘 (𝑧 − 2𝑖)3
5.
cos 𝑧
d𝑧, where 𝑘 = {𝑧 ∈ ℂ ∶ |𝑧| = 4},
∫𝑘 𝑧2 − 𝜋 2
6.
1
𝑒𝑧
d𝑧, where 𝑘 = {𝑧 ∈ ℂ ∶ |𝑧 − 2| = 1},
∫𝑘 (𝑧 − 4)2
2
7.
𝑒𝑧 cos(𝜋𝑧) 3
d𝑧, where 𝛾(𝑡) = 𝑒𝑖𝑡 , 𝑡 ∈ [0, 2𝜋],
∫𝛾 𝑧 + 2𝑧
2 2
8.
d𝑧 −2 + 𝑒−4𝜋𝑖𝑡
, where 𝛾(𝑡) = , 𝑡 ∈ [0, 4],
∫𝛾 (𝑧2 − 1) 3 2
9.
d𝑧
,
∫𝛾 (1 − 𝑧)(𝑧 + 2)(𝑧 − 𝑖)2
where 𝛾 denotes a simple closed piece-wise smooth positively oriented curve
such that −2 ∈ int 𝛾, 𝑖 ∈ int 𝛾, 1 ∈ ext 𝛾.
Exercise 10.45 Find

1+𝑖 𝑧
1. ∫0 𝑒 d𝑧,
1+𝑖 3
2. ∫0 𝑧 d𝑧,
𝑖
3. ∫0 𝑧2 sin 𝑧 d𝑧,
𝑖
4. ∫0 𝑧 sin 𝑧 d𝑧.
Exercise 10.46 Find out if a series is convergent:

∑∞
𝑖𝑛
1. 𝑛
,
𝑛=1 𝑛2
∑∞
𝑛
2. 𝑛
(1 + 𝑖)𝑛 ,
𝑛=1 3
∑∞
(−𝑖)𝑛
3. .
𝑛=1 3𝑛 − 17
Exercise 10.47 Find a domain of convergence of a series 3

∑
∞ ( )
1 𝑧+1 𝑛
1. 2 𝑧−1
,
𝑛=1 𝑛
∞ ( 𝑛 )
∑ 𝑧 𝑛2
2. + 𝑛 .
𝑛=1 𝑛! 𝑧
Exercise 10.48 Find a radius of convergence of a power series

∑
∞
𝑧 𝑛
1. 𝑑 𝑛2011 ,
𝑛=1
∑
∞
2. 𝑛𝑛 (𝑧 − 1)𝑛 ,
𝑛=1
∑∞
3𝑛 (𝑧 − 1)𝑛
3. √ ,
𝑛=1 (3𝑛 − 2)2𝑛
∑∞
(𝑧 + 1 + 𝑖)𝑛
4. 𝑛
,
𝑛=0 3 (𝑛 − 𝑖)
∑∞ 𝑛
𝑛 𝑛
5. 𝑧 ,
𝑛=1 𝑛!
∞ ( )
∑
6. cos(𝑖𝑛) 𝑧𝑛 ,
𝑛=0
∑
∞
7. (𝑛2 − 𝑛 − 2)𝑧𝑛 ,
𝑛=0
∑
∞
𝑧𝑛
8. .
𝑛=0 (𝑛 + 8)!
Exercise 10.49 Find the sum of a power series in a disk of convergence

∑
∞
1. 𝑛𝑧𝑛 ,
𝑛=1
∑∞ 𝑛
𝑧
2. ,
𝑛=1 𝑛
3
It means, find all 𝑧 ∈ ℂ for which the series is convergent.
85
∑∞
𝑧2𝑛+1
3. ,
𝑛=0 2𝑛 + 1
∑
∞
𝑧𝑛
4. (−1)𝑛+1 ,
𝑛=1 𝑛+1
∑
∞
5. (𝑛2 − 𝑛 − 2)𝑧𝑛 .
𝑛=0
Exercise 10.50 Find the sum of a series

∑∞
1
1. 𝑛
,
𝑛=1 𝑛2
∑∞
(−1)𝑛
2. 𝑛
.
𝑛=1 𝑛2
Exercise 10.51 Find a Taylor series of a function 𝑓 with a centre 𝑧0 and find its
radius of convergence if
𝑧+1
1. 𝑓 (𝑧) = , 𝑧0 = −1,
𝑧2 + 4𝑧 − 5
𝑧
2. 𝑓 (𝑧) = 2 , 𝑧0 = 0,
𝑧 +𝑖
1+𝑧
3. 𝑓 (𝑧) = ln , 𝑧 = 0,
1−𝑧 0
4. 𝑓 (𝑧) = 𝑒3𝑧−2 , 𝑧0 = 1,
5. 𝑓 (𝑧) = sin(3𝑧2 + 2), 𝑧0 = 0,

1
6. 𝑓 (𝑧) = , 𝑧0 = 3,
(𝑧 − 1)3
7. 𝑓 (𝑧) = sin2 𝑧, 𝑧0 = 0.
Exercise 10.52 Find a domain of convergence of a Laurent series4

∑
∞
1. 2−|𝑛| 𝑧𝑛 ,
𝑛=−∞
∑∞
(𝑧 − 𝑖)𝑛
2. 2
.
𝑛=−∞ 𝑛 + 1
Exercise 10.53 Find a Laurent series of a function 𝑓 on an annulus

cos 𝑧
1. 𝑓 (𝑧) = , 0 < |𝑧| < 1,
𝑧2
4
That is, find all 𝑧 ∈ ℂ for which the series is convergent.
1
2. 𝑓 (𝑧) = , |𝑧| > 1,
𝑧2 + 1
𝑧2 + 1 1
3. 𝑓 (𝑧) = , 2
< |𝑧 − 𝑖| < 1,
𝑧(𝑧 − 𝑖)
1
4. 𝑓 (𝑧) = , |𝑧| > 52 ,
2𝑧 − 5
1
5. 𝑓 (𝑧) = , 1 < |𝑧 − 2| < 2,
𝑧(𝑧 − 2)
𝑧
6. 𝑓 (𝑧) = 2 , 0 < |𝑧 − 𝑖| < 2,
(𝑧 + 1)2
𝑧 − sin 𝑧
7. 𝑓 (𝑧) = , 0 < |𝑧| < ∞,
𝑧4
𝑧+2
8. 𝑓 (𝑧) = , 2 < |𝑧 − 1| < ∞,
𝑧2 − 4𝑧 + 3
1
9. 𝑓 (𝑧) = , 1 < |𝑧 − 1| < 2.
𝑧(𝑧 − 3)2
Exercise 10.54 Find a Laurent series of a function 𝑓 on all maximal annuli with a
centre at 𝑧0 on which 𝑓 is a holomorphic fucntion if
𝑧2 − 𝑧 + 3
1. 𝑓 (𝑧) = , 𝑧0 = 0,
𝑧3 − 3𝑧 + 2
𝑧+1
2. 𝑓 (𝑧) = , 𝑧0 = 1 + 𝑖.
𝑧2
Exercise 10.55 Determine the type of each isolated singularity of a function 𝑓 if
2 3
1. 𝑓 (𝑧) = 𝑧5 + 4𝑧3 − 2 + + 2,
𝑧 𝑧
𝑧2 − 4
2. 𝑓 (𝑧) = ,
𝑧−2
1
3. 𝑓 (𝑧) = ,
𝑧 − 𝑧3
𝑧4
4. 𝑓 (𝑧) = ,
𝑧4 + 1
𝑒𝑧
5. 𝑓 (𝑧) = ,
𝑧2 + 4
𝑧2 + 4
6. 𝑓 (𝑧) = ,
𝑒𝑧
1 − 𝑒𝑧
7. 𝑓 (𝑧) = ,
2 + 𝑒𝑧
87
1
8. 𝑓 (𝑧) = 𝑒 𝑧2 ,
1
9. 𝑓 (𝑧) = ,
(𝑧 − 3)2 (2 − cos 𝑧)
𝑧
10. 𝑓 (𝑧) = ,
sin 𝑧
𝑧
11. 𝑓 (𝑧) = 𝑧2 sin 𝑧+1 ,
1 − cos 𝑧
12. 𝑓 (𝑧) = .
sin2 𝑧
Exercise 10.56 Prove L’Hôpital’s rule:
Let 𝑓 and 𝑔 be holomorphic non-constant functions on some ring neighbourhood
of a point 𝑧0 ∈ ℂ and let lim 𝑓 (𝑧) = lim 𝑔(𝑧) = 0. Then it holds that
𝑧→𝑧0 𝑧→𝑧0
𝑓 (𝑧) 𝑓 ′ (𝑧)
lim = lim ′ .
𝑧→𝑧0 𝑔(𝑧) 𝑧→𝑧0 𝑔 (𝑧)
Exercise 10.57 Find a residue of a function 𝑓 at all its isolated singularities if

1
1. 𝑓 (𝑧) = ,
𝑧 + 𝑧3
𝑧2
2. 𝑓 (𝑧) = ,
(1 + 𝑧)3
1
3. 𝑓 (𝑧) = ,
(𝑧2 + 1)3
𝑧3 + 1
4. 𝑓 (𝑧) = ,
𝑧−2
1
5. 𝑓 (𝑧) = ,
𝑧6 (𝑧2 + 1)2
6. 𝑓 (𝑧) = tan 𝑧,
1
7. 𝑓 (𝑧) = ,
sin 𝑧
8. 𝑓 (𝑧) = cot 3 𝑧,
1
9. 𝑓 (𝑧) = sin 𝑧 sin ,
𝑧
sin(𝜋𝑧)
10. 𝑓 (𝑧) = .
(𝑧 − 1)3
Exercise 10.58 Using the residue theorem calculate an integral

1.
cos 𝑧
d𝑧, where 𝛾(𝑡) = 3𝑒𝑖𝑡 , 𝑡 ∈ [0, 2𝜋],
∫𝛾 𝑧3
2.
1 1
cos d𝑧, where 𝛾(𝑡) = 18𝑒𝑖𝑡 , 𝑡 ∈ [0, 2𝜋],
∫𝛾 𝑧 + 2 𝑧
3.
𝑧3
d𝑧, where 𝑘 = {𝑧 ∈ ℂ ∶ |𝑧| = 2},
∫𝑘 𝑧 4 − 1
4.
𝑧3 1𝑧
𝑒 d𝑧, where 𝑘 = {𝑧 ∈ ℂ ∶ |𝑧| = 2},
∫𝑘 𝑧 + 1
5.
𝑧+1
𝑧 sin d𝑧, where 𝛾(𝑡) = 2𝑒−𝑖𝑡 , 𝑡 ∈ [0, 6𝜋],
∫𝛾 𝑧−1
6.
𝑒𝜋𝑧
d𝑧,
∫𝛾 2𝑧2 − 𝑖
where 𝛾 is a simple closed piece-wise smooth positively oriented curve such
that { }
𝜋
int 𝛾 = 𝑧 ∈ ℂ ∶ |𝑧| < 1 ∧ 0 < arg 𝑧 < ,
2
7.
d𝑧
, where 𝑘 = {𝑧 ∈ ℂ ∶ |𝑧| = 2}.
∫𝑘 𝑧5 (𝑧10 − 2)
Exercise 10.59 Using the residue theorem calculate an integral 5
1. 𝜋
d𝑥
,
∫−𝜋 5 + 3 cos 𝑥
2. ∞
𝑥2 d𝑥
,
∫−∞ 𝑥4 + 6𝑥2 + 25
3. ∞
𝑥4 + 1
d𝑥,
∫0 𝑥6 + 1
4. ∞
𝑥2
d𝑥,
∫0 (𝑥2 + 1)3
5
Introduced integrals are meant to be integrals of a real variable.
89
5. 𝜋
cos 𝑥
d𝑥,
∫−𝜋 3 + 2 sin 𝑥
6.
2𝜋
cos2 (2𝑥)
d𝑥,
∫0 5 − 4 cos 𝑥
7. ∞
d𝑥
,
∫−∞ 1 + 𝑥6
8. ∞
d𝑥
.
∫−∞ 𝑥2 +𝑥+1
Bibliography
[1] J.Bouchala: Funkce komplexní proměnné, http://mi21.vsb.cz/, 2012.
[2] J. Bouchala, O. Vlach: Křivkový a plošný integrál, http://mi21.vsb.cz/,

2011.
[3] J. Bouchala, P. Vodstrčil: Řady, http://mi21.vsb.cz/, 2011.
[4] I. Černý: Analýza v komplexním oboru, Academia, Praha, 1983.
[5] I. Černý: Základy analysy v komplexním oboru, Academia, Praha, 1967.
[6] M. Dont, B. Opic: Matematická analýza III – úlohy, skripta ČVUT, Praha,
1989.
[7] J. Eliaš, J. Horváth, J. Kajan, R. Šulka: Zbierka úloh z vyššej matematiky 4,

Alfa, Bratislava, 1979.
[8] P. Galajda, Š. Schrötter: Funkcie komplexnej premennej a operátorový počet,

Alfa, Bratislava, 1991.
[9] J. M. Howie: Complex analysis, Springer-Verlag, London, 2003.
[10] I. Kluvánek, L. Mišík, M. Švec: Matematika II., SVTL, Bratislava, 1965.
[11] T. Needham: Visual complex analysis, Oxford University Press, New York,
1997.
[12] K. Rektorys a spolupracovníci: Přehled užité matematiky I a II, Prometheus,

Praha, 1995.
[13] J. Veselý: Komplexní analýza pro učitele, Karolinum, Univerzita Karlova,

Praha, 2001.
91

FKP en

Uploaded by

Copyright:

Available Formats

FKP en

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

FKP en

Uploaded by

Copyright:

Available Formats

An Introduction to

• Cauchy’s theorem and its consequences,

• Laurent and Taylor series,

• conformal maps and harmonic functions,

• the residue theorem.

prof. RNDr. Marek Lampart, Ph.D.

September 29, 2022

1 Complex numbers and Gauss plane 5

2 Complex functions of a real and a complex variable 11

3 Derivative of a complex function of a complex variable 23

5 Complex functions integral, Cauchy’s integral theorem and formulas 33

5.4 Primitive functions and path independent integral . . . . . . . . . 38

6 Number series, sequences and function series 43

7 Power and Taylor series 49

8 Laurent series and classification of singularities 59

9 Residue and the residue theorem 67

Complex numbers and Gauss

1.1 Complex numbers

(i) A complex number 𝑧 is a number

𝑧 = 𝑥 + 𝑖𝑦 where 𝑥, 𝑦 ∈ ℝ and 𝑖2 = −1,

(v) Let 𝑧1 = 𝑥1 + 𝑖𝑦1 and 𝑧2 = 𝑥2 + 𝑖𝑦2 be two complex numbers. Then it is

Moreover, if 𝑧2 ≠ 0 = 0 + 0𝑖, it is also defined

(vi) For every complex number 𝑧 = 𝑥 + 𝑖𝑦 it holds that

𝑧𝑧 = (𝑥 + 𝑖𝑦)(𝑥 − 𝑖𝑦) = 𝑥2 − (𝑖𝑦)2 = 𝑥2 + 𝑦2 = |𝑧|2 .

Example 1.1 Find Re 𝑧0 and Im 𝑧0 for

1.2 Geometric interpretation and argument of complex num-

Arg 𝑧 = {𝜑 ∈ ℝ ∶ 𝑧 = |𝑧|(cos 𝜑 + 𝑖 sin 𝜑)}.

Theorem 1.1 Let 𝑧 ∈ ℂ ⧵ {0} and 𝜑 ∈ Arg 𝑧. Then

Arg 𝑧 = {𝜑 + 2𝑘𝜋 ∶ 𝑘 ∈ ℤ}.

⇒ [𝑧 = |𝑧|(cos 𝜑 + 𝑖 sin 𝜑) = |𝑧|(cos 𝜓 + 𝑖 sin 𝜓) ∧ |𝑧| ≠ 0] ⇒

⇒ 1 = cos(𝜓 − 𝜑) ⇒ [∃𝑘 ∈ ℤ ∶ 𝜓 − 𝜑 = 2𝑘𝜋] ⇒ [∃𝑘 ∈ ℤ ∶ 𝜓 = 𝜑 + 2𝑘𝜋] .

Definition 1.2 The argument 𝜑 ∈ Arg 𝑧 such that

is called a principal value of an argument of a complex number 𝑧 ∈ ℂ ⧵ {0} and

the properties described by the stereographical projection of point 𝑁.

1.4 Neighbourhood of a point

By the neighbourhood of ∞ with radius 𝜀 ∈ ℝ+ we mean the set

By the ring neighbourhood of a point 𝑧 ∈ ℂ∞ with radius 𝜀 ∈ ℝ+ we mean the

2. {𝑧 ∈ ℂ ∶ |𝑧 − 3| < |𝑧 + 2 − 𝑖|} and {𝑧 ∈ ℂ ∶ Im 𝑧 < 1} are open sets.

1.5 Sequence of complex numbers

Definition 1.6 Let 𝑧 ∈ ℂ∞ and (𝑧𝑛 ) be a sequence in ℂ∞ . We say that a sequence

1. The definition of a limit in fact means that outside of any neighbourhood of

2. Let (𝑧𝑛 ) be a sequence of complex numbers in the Gauss plane and 𝑧 ∈ ℂ∞ .

𝑧𝑛 → 𝑧 (in ℂ∞ ) ⇔ 𝑧∗𝑛 → 𝑧∗ (in ℝ3 ).

Example 1.4 Find

Theorem 1.4 A sequence of complex numbers has a limit 𝑧 ∈ ℂ∞ if and only if

Complex functions of a real and a

2.1 Complex functions

1. 𝑓 (𝑧) = 𝑧2 is a single-valued function, 𝐷𝑓 = ℂ∞ .

2. 𝑓 (𝑧) = Arg 𝑧 is an infinite-valued function, 𝐷𝑓 = ℂ ⧵ {0}.

Definition 2.2 Let 𝑓 be a multi-valued function. A single-valued function 𝜑 is

(ii) ∀𝑧 ∈ 𝐷𝜑 ∶ 𝜑(𝑧) ∈ 𝑓 (𝑧).

Example 2.2 Functions

2.2 Some important complex functions

2.2.2 Trigonometric functions

Theorem 2.2 (Properties of trigonometric functions)

(iii) For all 𝑧 ∈ ℂ it holds that:

(iv) For any 𝑧 ∈ ℂ Euler’s formula states that

2.2.3 Hyperbolic functions

2.2.4 Logarithmic functions