Complex Analysis

Prepared by

Muhammad Shahnewaz Bhuyan

Lecturer

Department of Mathematics

Chittagong University of Engineering and Technology

Chattogram-4349

Started from : March 03, 2021

Last Updated : April 8, 2025

 Practice plays a role

All great men think alike

The only way to learn mathematics is to do

mathematics - Paul Richard Halmos

 Contents

Chapter 1. Prerequisites 1

1.1. Region 1

1.2. Limit, Continuity and Differentiability of Real-valued Functions 1

1.3. Series Expansion of Different Elementary Functions 1

Chapter 2. Introduction to Complex Number 2

2.1. Before the Notion of Complex Number 2

2.2. Complex Number 4

2.3. Polar Form of Complex Numbers 5

2.4. Properties of Complex Numbers 6

2.5. Geometry of Complex Equations 8

2.6. Curve and Its Classification 9

Chapter 3. Limit, Continuity and Derivative of Complex Functions 10

3.1. Limit 10

3.2. Continuity 11

3.3. Differentiability of Complex Functions 12

3.4. Necessary and Sufficient Condition of Differentiability 15

Chapter 4. Analytic and Harmonic Functions 29

4.1. Analytic Functions 29

4.2. Harmonic Functions 43

iv
 CONTENTS v

4.3. Construction of Analytic Function 45

4.4. Exact Differential Method 46

4.5. Milne Thomson Method 51

Chapter 5. Complex Integration 58

5.1. Complex Line Integral 58

5.2. Cauchy’s Theorem 68

5.3. Cauchy’s Integral Formula 79

5.4. Winding Number 94

5.5. Cauchy’s Inequality 97

Chapter 6. Power Series 98

6.1. Power Series 98

6.2. Convergence of Geometric Series 105

6.3. D’Alembert’s Ratio Test 107

6.4. Taylor’s Series 110

6.5. Laurent’s Series 115

6.6. Singularity and Its Types 119

Chapter 7. Conformal Mapping and Its Appliactions 123

7.1. Conformal Mapping 123

Appendix A. Necessary Theorems and Formulae 124

A.1. Green’s Theorem 124

A.2. Necessary Trigonometric Formulae 124

A.3. Tracing Polar Curves 124

 CONTENTS vi

Appendix. Bibliography 125

 CHAPTER 1

Prerequisites

1.1. Region

Definition 1.1.1 (Open interval)

Definition 1.1.2 (Neighbourhood and deleted neighbourhood)

Definition 1.1.3 (Open disk)

Definition 1.1.4 (Annulus region)

1.2. Limit, Continuity and Differentiability of Real-valued

Functions

1.3. Series Expansion of Different Elementary Functions

ex , sin x, cos x

1
 CHAPTER 2

Introduction to Complex Number

In this chapter, the reader is referred to study [3, 4].

2.1. Before the Notion of Complex Number

Mathematicians faced problems to find such numbers as those of the fol-

lowing problem.

Problem 2.1 Find two numbers sum of which is 6 and product of which is

18.

Some mathematicians thought that these type of problems can not be

solved in mathematics. Since this type of problems were arising frequently,

at a stage mathematicians tried to solve these. Historically Cardano was first

mathematician who tried to solve this problem. He tried to design a world of

imaginary number by using the simple equation x2 + 1 = 0.

2.1.1. Euler’s approach towards complex numbers. Let x be such

a number that rotates any number a through 90o , if we multiply 1 by x. So

√
a · x · x = −a ⇒ x2 = −1 ⇒ x = ± −1.

2.1.2. Argand diagram. Plot of complex numbers considering the x-axis

as real axis and y-axis as the imaginary axis is known as Argand diagram.

2
 2.1. BEFORE THE NOTION OF COMPLEX NUMBER 3

It is named after the Genevan amature mathematicians Jean-Robert Argand.

1

2.1.3. Complex numbers as vectors. A complex number can also as

2-D Euclidean vectors.

2.1.4. Powers of i. Observe that

i1 = i, i2 = −1, i3 = −i, i4 = 1

i5 = i, i6 = −1, i7 = −i, i8 = 1

i9 = i, i10 = −1, i11 = −i, i12 = 1

and so on.

2.1.5. Shortcut technique to find the large power of i. Divide the

power by 4 and find the remainder. If

(a) the remainder is 1, then the number is i.

(b) the remainder is 2, then the number is −1.

(c) the remainder is 3, then the number is −i.

(d) the remainder is 0, then the number is 1.

Problem 2.2 Find i435798776 .

1Historically, argand diagram was first described by Norwegian-Danish land surveyor

and mathematician Caspar Wessel [?, Complex Plane, Argand Diagram]. German math-

ematician Carl Friedrich Gauss has also contribution in the Geometric representation of

complex number.
 2.2. COMPLEX NUMBER 4

2.2. Complex Number

2.2.1. Modulus and argument of a complex number. Modulus of

p
a complex number z = x + iy is defined as |z| = x2 + y 2 and its amplitude
y
or argument is defined as θ = tan−1 .
x
p √
Example 2.1 Modulus of 2 − 3i is |2 − 3i| = 22 + (−3)2 13.
 2
2+i
Problem 2.3 Find the modulus and argument of .
3−i

Solution Given complex number is

 2
2+i 4 + 4i − 1
=
3−i 9 − 6i − 1
3 + 4i
=
8 − 6i
(3 + 4i)(8 + 6i)
=
(8 − 6i)(8 + 6i)
i
= .
2

s  2
1 1
So the required modulus is 02 + = and the required principal ar-
2 2
1
 
π π
gument is tan−1  2  = . Therefore general argument is 2nπ + , where
 
0 2 2
n = 0, ±1, ±2, · · · .

Exercise 2.1 Find the modulus and argument of

−2 2π 2π
(a) √ . Answer: Modulus 1; Argument , 2nπ + ,n∈R
1+i 3 3 3
1−i π π
(b) . Answer: Modulus 1; Argument − , 2nπ − , n ∈ R
1+i 2 2
 2.3. POLAR FORM OF COMPLEX NUMBERS 5

2.3. Polar Form of Complex Numbers

2.3.1. Polar form of complex numbers. Let us consider the complex

number (x, y) or x + iy. If its modulus and argument are r and θ respectively,

then

x = r cos θ and y = r sin θ.

So

z = r(cos θ + i sin θ),

which is the polar form of z. Sometimes, shortly is written z = cis θ to mean

z = r(cos θ + i sin θ).

Theorem 2.3.1 For two complex numbers z1 = x1 +iy1 = r1 (cos θ1 + i sin θ1 )

and z2 = x2 + iy2 = r2 (cos θ2 + i sin θ2 ), then

(a) z1 z2 = r1 r2 [cos (θ1 + θ2 ) + i sin (θ1 + θ2 )].

z1 r1
(b) = [cos (θ1 − θ2 ) + i sin (θ1 − θ2 )].
z2 r2

Proof. (a) Here

z1 z2 = [r1 (cos θ1 + i sin θ1 )] [r2 (cos θ2 + i sin θ2 )]

= r1 r2 [(cos θ1 cos θ2 − sin θ1 sin θ2 ) + i (sin θ1 cos θ2 + cos θ1 sin θ2 )]

= r1 r2 [cos (θ1 + θ2 ) + i sin (θ1 + θ2 )] .

 2.4. PROPERTIES OF COMPLEX NUMBERS 6

(b) Here

z1 r1 (cos θ1 + i sin θ1 )
=
z2 r2 (cos θ2 + i sin θ2 )
r1 (cos θ1 + i sin θ1 ) (cos θ2 − i sin θ2 )
= ·
r2 (cos θ2 + i sin θ2 ) (cos θ2 − i sin θ2 )
 
r1 (cos θ1 cos θ2 + sin θ1 sin θ2 ) + i (sin θ1 cos θ2 − cos θ1 sin θ2 )
=
r2 cos2 θ2 + sin2 θ2
r1
= [cos (θ1 − θ2 ) + i sin (θ1 − θ2 )] .
r2

Hence the proof is complete.

2.3.2. De Moivre’s Theorem. Theorem 2.3.1(a) can be extended for n

complex numbers as

z1 z2 z3 · · · zn

= r1 r2 r3 · · · rn [cos (θ1 + θ2 + θ3 + · · · θn ) + i sin (θ1 + θ2 + θ3 + · · · θn )]

When z1 = z2 = z3 = · · · = zn = z, then above equation becomes

z n = {r(cos θ + i sin θ)}n = rn (cos nθ + i sin nθ)

2.4. Properties of Complex Numbers

2.4.1. Triangular inequality. The following is known as the triangular

inequality of complex numbers.

Theorem 2.4.1 |z1 + z2 | ≤ |z1 | + |z2 | for any pair of complex numbers z1

and z2 .

Corollary 2.4.2 |z1 + z2 + z3 + · · · + zn | ≤ |z1 | + |z2 | + |z3 | + · · · + |zn | for

complex numbers z1 . z2 , z3 , · · · , zn .
 2.4. PROPERTIES OF COMPLEX NUMBERS 7

Theorem 2.4.3 |z1 − z2 | ≤ |z1 | + |z2 | for any pair of complex numbers z1

and z2 .

Theorem 2.4.4 For complex numbers z1 and z2 ,

|z1 − z2 | ≥ ||z1 | − |z2 || ≥ |z1 | − |z2 | .

Problem 2.4 For the complex numbers confirm that

z1 |z1 |
(a) ≤ .
z2 + z3 ||z2 | − |z3 ||

Problem 2.5 Show that

|z1 + z2 |2 + |z1 − z2 |2 = 2 |z1 |2 + 2 |z2 |2 .

Interpret this result geometrically and then using it, deduce that

p p
α+ α2 + β 2 + α − α2 − β 2 = |α + β| + |α − β| .

Proof. We can write that

|z1 + z2 |2 + |z1 − z2 |2 = (z1 + z2 ) (z1 + z2 ) + (z1 − z2 ) (z1 − z2 )

= (z1 + z2 ) (z1 + z2 ) + (z1 − z2 ) (z1 − z2 )

= z1 z1 + z1 z2 + z1 z2 + z2 z2 + z1 z1 − z1 z2 − z1 z2 + z2 z2

= |z1 |2 + |z2 |2 + |z1 |2 + |z2 |2

= 2 |z1 |2 + 2 |z2 |2 .

Hence the statement.

 2.5. GEOMETRY OF COMPLEX EQUATIONS 8

2.5. Geometry of Complex Equations

Theorem 2.5.1 The equation |z + z1 | + |z − z2 | = r represents

(a) an ellipse having z = z0 , z = −z0 as its foci and r as its length of

major axis, if r > |z1 − z2 |.

(b) a line segment, if r = |z1 − z2 |.

(c) an empty set, if r < |z1 − z2 |.

Theorem 2.5.2 The equation ||z + z1 | − |z − z2 || = r represents

(a) a hyperbola having z = z0 , z = −z0 as its foci and r as its length of

transversal axis, if r < |z1 − z2 |.

(b) two rays, if r = |z1 − z2 |.

(c) an empty set, if r > |z1 − z2 |.

Note 2.1 The equation |z − z0 | = r represents a circle centered at z = z0

with radius r unit.

Note 2.2 The equation |z + z0 | + |z − z0 | = 2a on the complex plane rep-

resents an ellipse having foci at z = z0 and z = −z0 with 2a as its length of

major axis.

Problem 2.6 Describe with geometrical sketch which type of regions are

represented by the following equations :

(a) |z − i| = |z + i|

(b) |z − 4| > |z|

(c) −π < argz < π, z ̸= 0

(d) 1 < |z + i| ≤ 2
 2.6. CURVE AND ITS CLASSIFICATION 9

 
1 1
(e) Re <
z  2
1 1
(f) Re ≤
z 2
(g) 1 < |z − 2i| ≤ 2

(h) |z − 1| + |z + 1| ≤ 3

(i) |z + i| + |z − i| ≤ 3

(j) |z + 21| + |z − 2i| = 6

Solution See [3, Example 35, Page 45].

2.6. Curve and Its Classification

Definition 2.6.1 (Curves) See [4, Page 3.7]

Definition 2.6.2 (Closed curves) See [4, Page 3.7]

Definition 2.6.3 (Simple closed curve) See [4, Page 3.7]

Definition 2.6.4 (Smooth curve or arc) See [4, Page 3.7]

Definition 2.6.5 (Piecewise or sectionally closed curve or contour) See [4,

Page 3.8]

Definition 2.6.6 (Jordan curve) See [4, Page 4.3]

 CHAPTER 3

Limit, Continuity and Derivative of Complex Functions

In this chapter, the reader is referred to study [3, 4, 1].

Problem 3.1 (Page 2.16, Problem 2.14 of [4]) Show that f (z) − ln z has a

branch point at z = 0.

Solution Let x = r cos θ and y = r sin θ. So

f (z) = ln z = ln(x + iy)

= ln(r(cos θ + i sin θ))

= ln(reiθ ) = ln r + iθ.

Suppose that

3.1. Limit

3.1.1. Limit. Let w = f (z) be defined and single-valued on a neighbour-

hood about z = z0 , except possibly at z = z0 itself. We say that L is the limit

of f (z) as z approaches z0 and we write

lim f (z) = L,
z→z0

if for every ϵ > 0, there exists a corresponding δ > 0 such that for all z

0 < |z − z0 | < δ ⇒ |f (z) − L| < ϵ.

10
 3.2. CONTINUITY 11

Note 3.1 (Page 2.6, [4]) If w = f (z) is multiple valued, the limit as z → z0

may depend on the particular branch.

3.2. Continuity

3.2.1. Definition of continuity. Let f (z) be defined and single-valued

in a neighbourhood of z = z0 as well as at z = z0 , then f (z) is continuous at

z = z0 whenever limz→z0 f (z) = f (z0 ). Formally, it is defined as follows.

Definition 3.2.1 (ϵ-δ definition of continuity) Let f (z) be defined and single-

valued on a neighbourhood of z = z0 . We say that f (z) is continuous at

z = z0 , if for every ϵ > 0 there exists a corresponding δ > 0 such that for all z

|z − z0 | < δ ⇒ |f (z) − f (z0 )| < ϵ.

3.2.2. Continuity test. In order to f (z) be continuous at z = z0 , the

following three conditions must be obeyed :

(i) f (z0 ) exists.

(ii) limz→z0 f (z) exists.

(iii) limz→z0 f (z) = f (z0 ).

Problem 3.2 [3, Example 15, Page 119] Show that f (z) = |z|2 is continuous

everywhere.

Solution Here f (z) = |z|2 = zz = x2 + y 2 . Clearly, f (z) is a function of

nonzero denominators for all x and y in the z-plane. Thus f (z) is continuous

everywhere.
 3.3. DIFFERENTIABILITY OF COMPLEX FUNCTIONS 12

3.3. Differentiability of Complex Functions

Definition 3.3.1 (Differentiable function and derivative) [2, Part 06] A func-

tion f is called differentiable at z = z0 , if

f (z) − f (z0 )
lim
z→z0 z − z0

exists. If the limit exists, then it is denoted by f ′ (z0 ) and called the derivative

of f at z = z0 .

Alternative approach of defining derivative of a complex function is as

follows.

Definition 3.3.2 (Derivative and differentiability) [4, Page 3.1] Let w =

f (z) be single valued in some region Γ of the z-plane. Then the derivative of

f (z) with respect to z is denoted by f ′ (z) and defined as

f (z + h) − f (z)
f ′ (z) = lim ,
h→0 h

provided the limit exists independent of the manner in which h → 0.

If the above limit exists, then we say that f (z) is differentiable at z.

3.3.1. Geometrical interpretation of complex derivative. f ′ (z0 ) =

reiθ means that near z0 stretching distances by the factor r and rotating by
i
the angle θ. For example, f ′ (z0 ) = − means that shrinki9ng by a factor of 2
2
3π
and rotating counterclockwise by radians.
2

Problem 3.3 What is physical or geometrical meaning of the derivative of

a complex valued function at a point?

 3.3. DIFFERENTIABILITY OF COMPLEX FUNCTIONS 13

Solution It can be interpreted as expressing the local or instantaneous stretch-

ing and rotation. ......

Problem 3.4 Using the notion of complex scale factor and complex de-

rivative, describe what happens to points close to 1 + i under the function

1
f (z) = .
z

Solution To a close approximation, a small disc D centered at z0 = 1 + i

1−i
is mapped to another small disc D′ centred at f (1 + i) = . In this
2
process, the disc D is scaled by the factor |f ′ (z0 )| and rotated through the
1 i 1
angle arg (f ′ (z0 )). Here f ′ (z) = −
2
. So f ′ (1 + i) = . Thus |f ′ (1 + i)| =
z 2 2
′ π 1
and arg (f (z0 )) = . Therefore f scales the disc D by the factor and rotates
2 2
π
it anticlockwise through the angle .
2

Exercise 3.1 Using the notion of complex scale factor, describe what hap-
4z + 3
pens to points close to i under the function f (z) = .
2z 2 + 1

Solution To a close approximation, a small disc centered at z0 = i is mapped

4i + 3
to another small disc D′ centred at f (i) = = −3 − 4i. In this process,
2i2 + 1
D is scaled by the factor |f ′ (z0 )| and rotated through the angle arg (f ′ (z0 )).
−8z 2 − 12z + 4 −8i2 − 12i + 4
Here f ′ (z) = . So f ′
(i) = = 12 − 12i. Thus
(2z 2 + 1)2 (2i2 + 1)2
√ π
|f ′ (i)| = 12 2 and arg (f ′ (i)) = − . Therefore f scales the disc D by the
4
√ π
factor 12 2 and rotates it clockwise through the angle .
4

Problem 3.5 Show that f (z) = |z|2 is differentiable at z = 0.

 3.3. DIFFERENTIABILITY OF COMPLEX FUNCTIONS 14

Solution At z = 0,

f (0 + ∆z) − f (0)
f ′ (0) = lim
∆z→0 ∆z
f (∆z) − |0|2
= lim
∆z→0 ∆z
|∆z|2 − 0
= lim
∆z→0 ∆z
∆z∆z
= lim = lim ∆z = 0.
∆z→0 ∆z ∆z→0

Thus f ′ (0) exists and therefore f (z) = |z|2 is differentiable at the origin.

Theorem 3.3.1 Every differentiable function is continuous.

f (z) − f (z0 )
Proof. Let f (z) be differentiable at z = z0 . That is f ′ (z0 ) = limz→z0
z − z0
exists.

Now

 
f (z) − f (z0 )
lim [f (z) − f (z0 )] = lim (z − z0 )
z→z0 z→z0 z − z0
f (z) − f (z0 )
⇒ lim f (z) − lim f (z0 ) = lim · lim (z − z0 )
z→z0 z→z0 z→z0 z − z0 z→z0

⇒ lim f (z) − f (z0 ) = f ′ (z0 ) · 0 = 0

z→z0

∴ lim f (z) = f (z0 ).

z→z0

Hence f (z) is continuous at z = z0 .

Note 3.2 Converse of the above theorem is not necessarily true. For example,

consider the function f (z) = z = x − iy. It is continuous at z = 0.

 3.4. NECESSARY AND SUFFICIENT CONDITION OF DIFFERENTIABILITY 15

Let h = ∆z = p + iq. So h = p − iq. Thus at z = z0 = 0,

f (z0 + h) − f (z0 )
f ′ (z0 ) = lim
h→0 h
f (0 + h) − f (0) f (h) − 0
⇒ f ′ (0) = lim = lim
h→0 h h→0 h
f (h) h
= lim = lim
h→0 h h→0 h

p − iq
= lim .
p→0,q→0 p + iq

Along real axis q = 0 and p → 0. So along real axis

p−0 p
f ′ (0) = lim = lim = lim 1 = 1
p→0 p + 0 p→0 p p→0

Again, along imaginary axis, p = 0 and q → 0. Thus along imaginary axis

0 − iq −iq
f ′ (0) = lim = lim = lim −1 = −1
q→0 0 + iq q→0 iq p→0

Here since the above two limits are not equal, f ′ (0) does not exist, though

f (z) is continuous at z = 0.

Note 3.3 f (z) = z̄ is continuous everywhere, but differentiable nowhere.

3.4. Necessary and Sufficient Condition of Differentiability

3.4.1. Cauchy-Riemann equations. Cauchy-Riemann equations on

a complex-variable function w = f (z) = u(x, y) + iv(x, y) consist of the fol-

lowing two partial differential equations :


∂u ∂v
=



∂x ∂y 
(1)
∂u ∂v 
= − .


∂y ∂x
 3.4. NECESSARY AND SUFFICIENT CONDITION OF DIFFERENTIABILITY 16

This system of equations is named after two mathematicians Augustin Cauchy

and Bernhard Riemann.

Note 3.4 Cauchy-Riemann equations are also briefly referred as C-R equa-

tions. Shortly, C-R equations are written in the form: ux = vy and uy = −vx .

3.4.2. Polar form of Cauchy-Riemann equations. Cauchy-Riemann

equations in polar form are

∂u 1 ∂v ∂v 1 ∂u
= and =− .
∂r r ∂θ ∂r r ∂θ

Proof. See [3, Page 96-98].

3.4.3. Necessary condition of differentiability. The next theorem

states the necessary conditions for a function to be differentiable at a point.

Theorem 3.4.1 [2, Part 06] A necessary condition for a function

w = f (z) = u(x, y) + iv(x, y)

be differentiable at a point z0 = x0 + iy0 is that, at the point z0 = x0 + iy0 the

∂u ∂u ∂v ∂v
partial derivatives , , , exists and u, v satisfy the C-R equations
∂x ∂y ∂x ∂y
∂u ∂v ∂u ∂v
= and =− .
∂x ∂y ∂y ∂x

at the point z0 = x0 + iy0 .

Proof. Let a function w = f (z) = u(x, y) + iv(x, y) be differentiable at any

point z0 = x0 + iy0 . So f ′ (z0 ) exists. We have

f (z) − f (z0 )
f ′ (z0 ) = lim .
z→z0 z − z0
 3.4. NECESSARY AND SUFFICIENT CONDITION OF DIFFERENTIABILITY 17

When z → z0 along the real axis (that is along y = y0 ), then

∆z = z − z0 = (x + iy) − (x0 + iy0 )

= (x + iy0 ) − (x0 + iy0 ), since in this case y = y0

= x − x0 = ∆x.

So

f (z) − f (z0 )
f ′ (z0 ) = lim
z→z0 z − z0
f (z) − f (z0 )
= lim
∆z→0 ∆z
{u(x, y) + iv(x, y)} − {u(x0 , y0 ) + iv(x0 , y0 )}
= lim
∆x→0 ∆x
{u(x0 + ∆x, y0 ) + iv(x0 + ∆x, y0 )} − {u(x0 , y0 ) + iv(x0 , y0 )}
= lim ,
∆x→0 ∆x

since x − x0 = ∆x, y = y0

{u(x0 + ∆x, y0 ) − u(x0 , y0 )} + i{v(x0 + ∆x, y0 ) − v(x0 , y0 )}

= lim
∆x→0 ∆x
u(x0 + ∆x, y0 ) − u(x0 , y0 ) v(x0 + ∆x, y0 ) − v(x0 , y0 )
= lim + i lim
∆x→0 ∆x ∆x→0 ∆x
∂u ∂v
= (x0 , y0 ) + i (x0 , y0 ).
∂x ∂x

f (z) − f (z0 ) ∂u ∂v
Since limz→z0 exists, from above (x0 , y0 ) and (x0 , y0 ) must
z − z0 ∂x ∂x
exist.

Again, when z → z0 along the imaginary axis (that is along x = x0 ), then

∆z = z − z0 = (x + iy) − (x0 + iy0 )

= (x0 + iy0 ) − (x0 + iy0 ), since in this case x = x0

= i(y − y0 ) = i∆y.
 3.4. NECESSARY AND SUFFICIENT CONDITION OF DIFFERENTIABILITY 18

Moreover, when i∆y → 0, then ∆y → 0.

Here

f (z) − f (z0 )
f ′ (z0 ) = lim
z→z0 z − z0
f (z) − f (z0 )
= lim
∆z→0 ∆z
{u(x, y) + iv(x, y)} − {u(x0 , y0 ) + iv(x0 , y0 )}
= lim
∆y→0 i∆y
{u(x0 , y0 + ∆y) + iv(x0 , y0 + ∆y)} − {u(x0 , y0 ) + iv(x0 , y0 )}
= lim ,
∆y→0 i∆y

since y − y0 = ∆y, x = x0

{u(x0 , y0 + ∆y) − u(x0 , y0 )} + i{v(x0 , y0 + ∆y) − v(x0 , y0 )}

= lim
∆y→0 i∆y
u(x0 , y0 + ∆y) − u(x0 , y0 ) v(x0 , y0 + ∆y) − v(x0 , y0 )
= lim + i lim
∆y→0 i∆y ∆y→0 i∆y
1 ∂u ∂v
= (x0 , y0 ) + (x0 , y0 )
i ∂y ∂y
∂u ∂v
=−i (x0 , y0 ) + (x0 , y0 )
∂y ∂y
f (z) − f (z0 ) ∂u ∂v
Since limz→z0 exists, from above (x0 , y0 ) and (x0 , y0 ) must
z − z0 ∂y ∂y
exist.

By the uniqueness of f ′ (z0 ),

∂u ∂v ∂u ∂v
(x0 , y0 ) + i (x0 , y0 ) = −i (x0 , y0 ) + (x0 , y0 )
∂x ∂x ∂y ∂y
∂u ∂v ∂u ∂v
⇒ (x0 , y0 ) = (x0 , y0 ) and (x0 , y0 ) = − (x0 , y0 ).
∂x ∂y ∂y ∂x

Here z = z0 = x0 + iy0 is any point. Therefore

∂u ∂v ∂u ∂v
= and =− .
∂x ∂y ∂y ∂x

Hence the statement.

 3.4. NECESSARY AND SUFFICIENT CONDITION OF DIFFERENTIABILITY 19

Note 3.5 By the necessary condition stated in Theorem 3.4.1, it can be said

that

(i) if a function f fails to satisfy the C-R equations at z = z0 , then it is

not differentiable at z = z0 .

(ii) if a function f is differentiable at z = z0 , then f satisfies the C-R

equations, but the converse is not always true.

In a word, a function f satisfies the C-R equations at the point z = z0 does not

imply the differentiability of the function f at the point z = z0 . A function f

satisfying C-R equations at z = z0 may or may not be differentiable at z = z0 .

3.4.4. Sufficient condition of differentiability. [1] The next theorem

states the sufficient conditions for a function to be differentiable at a point.

Theorem 3.4.2 [2, Part 07] If the real functions u(x, y), v(x, y) are con-
∂u ∂u ∂v ∂v
tinuous and have continuous first order partial derivatives , , ,
∂x ∂y ∂x ∂y
in some neighbourhood of a point z0 = x0 + iy0 , and if u(x, y), v(x, y) sat-

isfies C-R equations at the point z0 = x0 + iy0 , then the complex function

f (z) = u(x, y) + iv(x, y) is differentiable at the point z0 = x0 + iy0 .

Note 3.6 Above theorem is employed to show a given complex-valued func-

tion differentiable at a point.

Note 3.7 The following (real and) complex functions are everywhere differ-

entiable and so everywhere continuous.

(i) Constant functions : f (z) = c.

(ii) Polynomial : f (z) = a0 z n + a1 z n−1 + a2 z n−2 + · · · + an−1 z + an

 3.4. NECESSARY AND SUFFICIENT CONDITION OF DIFFERENTIABILITY 20

(iii) Exponential : f (z) = ez

(iv) f (z) = sin z

(v) f (z) = cos z

(vi) f (z) = sinh z

(vii) f (z) = cosh z

Problem 3.6 Show that following complex-valued functions are everywhere

differentiable and everywhere continuous.

(a) f (z) = constant (e) f (z) = cos z

ez − e−z
(b) f (z) = polynomial (f) f (z) = sinh z =
2
ez + e−z
(c) f (z) = ez (g) f (z) = cosh z = .
2
(d) f (z) = sin z

Solution (b) Let f (z) = u(x, y) + iv(x, y) = z 2 + 1, when z = x + iy. Now

f (z) = z 2 + 1 = (x + iy)2 + 1

⇒ u(x, y) + iv(x, y) = x2 − y 2 + 1 + i2xy

⇒ u = x2 − y 2 + 1 and v = 2xy

∂u ∂v ∂u ∂v
⇒ = 2x, = 2y, = −2y, = 2x.
∂x ∂x ∂y ∂y

∂u ∂v ∂u ∂v
Here = and = − . Hence C-R equations are satisfied. Clearly
∂x ∂y ∂y ∂x
∂u ∂v ∂u ∂v
u, v, , , , are real-valued polynomial functions and so these are
∂x ∂x ∂y ∂y
continuous everywhere. Therefore the function f is differentiable everywhere.

Since every differentiable function is continuous, f is continuous everywhere.

 3.4. NECESSARY AND SUFFICIENT CONDITION OF DIFFERENTIABILITY 21

(c) Let f (z) = u(x, y) + iv(x, y) = ez , when z = x + iy. Now

f (z) = ez = ex .eiy = ex (cos y + i sin y)

⇒ u(x, y) + iv(x, y) = ex cos y + iex sin y

⇒ u = ex cos y and v = ex sin y

∂u ∂v ∂u ∂v
⇒ = ex cos y, = ex sin y, = −ex sin y, = ex cos y.
∂x ∂x ∂y ∂y

∂u ∂v ∂u ∂v
Here = and = − . Hence C-R equations are satisfied. We know
∂x ∂y ∂y ∂x
that real-valued functions ex , ey , sin x, sin y are continuous everywhere. Since

product of two continuous functions is continuous, the real-valued functions

∂u ∂v ∂u ∂v
u, v, , , , are continuous everywhere. Therefore the function f is
∂x ∂x ∂y ∂y
differentiable everywhere. Since every differentiable function is continuous, f

is continuous everywhere.

(d) Let f (z) = u(x, y) + iv(x, y) = sin z, when z = x + iy. Now

f (z) = sin z = sin(x + iy) = sin x cos(iy) + cos x sin(iy)

⇒ u(x, y) + iv(x, y) = sin x cosh y + i cos x sinh y,

since cos (iy) = cosh y and sin (iy) = i sinh y

⇒ u = sin x cosh y and v = cos x sinh y

∂u ∂v
⇒ = cos x cosh y, = − sin x sinh y,
∂x ∂x
∂u ∂v
= sin x sinh y, = cos x cosh y
∂y ∂y
d d
since (cosh y) = sinh y and (sinh y) = cosh y.
dy dy
 3.4. NECESSARY AND SUFFICIENT CONDITION OF DIFFERENTIABILITY 22

∂u ∂v ∂u ∂v
Here = and = − . Hence C-R equations are satisfied. We
∂x ∂y ∂y ∂x
know that real-valued functions sin x, sinh x, cos x, cosh x, sin, sinh y, cos y,

cosh y are continuous everywhere. Since product of two continuous functions

∂u ∂v ∂u ∂v
is continuous, the real-valued functions u, v, , , , are continuous
∂x ∂x ∂y ∂y
everywhere. Therefore the function f is differentiable everywhere. Since every

differentiable function is continuous, f is continuous everywhere.

Problem 3.7 Check the differentiability and continuity of the following

complex-valued functions.

(a) f (z) = tan z (e) f (z) = tanh z

(b) f (z) = sec z (f) f (z) = coth z

1
(c) f (z) = cot z (g) f (z) =
cosh z
1
(d) f (z) = csc z (h) f (z) = .
sinh z

sin z
Solution (a) We have f (z) = tan z = . We know that sin z and cos z
cos z
sin z
are differentiable everywhere. So f (z) = is differentiable everywhere,
cos z
except when
π
cos z = 0 ⇒ z = (2n + 1) , n ∈ Z.
2
n π o
Therefore f is differentiable at C \ z : z = (2n + 1) , n ∈ Z .
2
1
(b) We have f (z) = sec z = . We know that cos z are differentiable
cos z
1
everywhere. So f (z) = is differentiable everywhere, except when
cos z

π
cos z = 0 ⇒ z = (2n + 1) , n ∈ Z.
2
n π o
Therefore f is differentiable at C \ z : z = (2n + 1) , n ∈ Z .
2
 3.4. NECESSARY AND SUFFICIENT CONDITION OF DIFFERENTIABILITY 23

cos z
(c) We have f (z) = cot z = . We know that sin z is differentiable
sin z
1
everywhere. So f (z) = is differentiable everywhere, except when
sin z

sin z = 0 ⇒ z = nπ, n ∈ Z.

Therefore f is differentiable at C \ {z : z = nπ, n ∈ Z}.

1
(d) We have f (z) = csc z = . We know that sin z is differentiable
sin z
1
everywhere. So f (z) = is differentiable everywhere, except when
sin z

sin z = 0 ⇒ z = nπ, n ∈ Z.

Therefore f is differentiable at C \ {z : z = nπ, n ∈ Z}.

sinh z
(e) We have f (z) = tanh z = . We know that sinh z and cosh z
cosh z
sinh z
are differentiable everywhere. So f (z) = is differentiable everywhere,
cosh z
except when
π
cosh z = 0 ⇒ z = i(2n + 1) , n ∈ Z.
2
n π o
Therefore f is differentiable at C \ z : z = i(2n + 1) , n ∈ Z .
2
cosh z
(f) We have f (z) = coth z = . We know that cosh z and sinh z
sinh z
cosh z
are differentiable everywhere. So f (z) = is differentiable everywhere,
sinh z
except when

sinh z = 0 ⇒ z = inπ, n ∈ Z.

Therefore f is differentiable at C \ {z : z = inπ, n ∈ Z}.

1
(g) We have f (z) = . We know that cosh z is differentiable every-
cosh z
1
where. So f (z) = is differentiable everywhere, except when
cosh z
π
cosh z = 0 ⇒ z = i(2n + 1) , n ∈ Z.
2
 3.4. NECESSARY AND SUFFICIENT CONDITION OF DIFFERENTIABILITY 24

n π o
Therefore f is differentiable at C \ z : z = i(2n + 1) , n ∈ Z .
2

1
(h) We have f (z) = . We know that sinh z is differentiable every-
sinh z
1
where. So f (z) = is differentiable everywhere, except when
sinh z

sinh z = 0 ⇒ z = inπ, n ∈ Z.

Therefore f is differentiable at C \ {z : z = inπ, n ∈ Z}.

Problem 3.8 Check the differentiability of the following functions defined

over a suitable subset of C.

(a) f (z) = Re(z) (c) f (z) = z

(b) f (z) = Im(z) (d) f (z) = |z|.

Solution (a) Let f (z) = u(x, y) + iv(x, y) = Re(z), when z = x + iy. Now

f (z) = Re(z) = Re(x + iy)

⇒ u(x, y) + iv(x, y) = x + i.0,

⇒ u(x, y) = x and v(x, y) = 0

∂u ∂v ∂u ∂v
⇒ = 1, = 0, = 0, = 0.
∂x ∂x ∂y ∂y

∂u ∂v
Here ̸= . Hence C-R equations are not satisfied. Therefore the function
∂x ∂y
f is differentiable nowhere.
 3.4. NECESSARY AND SUFFICIENT CONDITION OF DIFFERENTIABILITY 25

(b) Let f (z) = u(x, y) + iv(x, y) = Im(z), when z = x + iy. Now

f (z) = Im(z) = Im(x + iy)

⇒ u(x, y) + iv(x, y) = y = y + i · 0,

⇒ u(x, y) = y and v(x, y) = 0

∂u ∂v ∂u ∂v
⇒ = 0, = 0, = 1, = 0.
∂x ∂x ∂y ∂y

∂u ∂v
Here ̸= − . Hence C-R equations are not satisfied. Therefore the func-
∂y ∂x
tion f is differentiable nowhere.

(c) Let f (z) = u(x, y) + iv(x, y) = z, when z = x + iy. Now

f (z) = Im(z)z = x + iy

⇒ u(x, y) + iv(x, y) = x − iy,

⇒ u(x, y) = x and v(x, y) = −y

∂u ∂v ∂u ∂v
⇒ = 1, = 0, = 0, = −1.
∂x ∂x ∂y ∂y

∂u ∂v
Here ̸= . Hence C-R equations are not satisfied. Therefore the function
∂x ∂y
f is differentiable nowhere.

(d) Let f (z) = u(x, y) + iv(x, y) = |z|, when z = x + iy. Now

f (z) = |z| = |x + iy|

p
⇒ u(x, y) + iv(x, y) = x2 + y 2 + i0,

p
⇒ u(x, y) = x2 + y 2 and v(x, y) = 0

∂u 2x x ∂v ∂u y ∂v
⇒ = p =p , = 0, =p , = 0.
∂x 2
2 x +y 2 x + y2
2 ∂x ∂y x + y2
2 ∂y
 3.4. NECESSARY AND SUFFICIENT CONDITION OF DIFFERENTIABILITY 26

Here
∂u ∂v x
= ⇒p =0 ⇒x=0
∂x ∂y x2 + y 2

and
∂u ∂v y
=− ⇒p =0 ⇒ y = 0.
∂y ∂x x2 + y 2
∂u x
But =p becomes undefined when (x, y) = (0, 0). Thus x = 0 and
∂x x + y2
2

y = 0 is impossible at a time. Hence C-R equations are satisfied nowhere by

f . Therefore the function f is differentiable nowhere.

Problem 3.9 [3, Example 15, Page 119] Show that f (z) = |z|2 is continuous

everywhere, but not differentiable except at the origin.

Solution For 1st part see Problem 3.2.

Proof of the second part. f (z) = u(x, y) + iv(x, y) = |z|2 , when z =

x + iy. Now

p
f (z) = |z|2 = |x + iy|2 = ( x2 + y 2 )2

⇒ u(x, y) + iv(x, y) = x2 + y 2 + i0,

⇒ u(x, y) = x2 + y 2 and v(x, y) = 0

∂u ∂v ∂u ∂v
⇒ = 2x, = 0, = 2y, = 0.
∂x ∂x ∂y ∂y

Here
∂u ∂v
= ⇒ 2x = 0 ⇒x=0
∂x ∂y

and
∂u ∂v
=− ⇒ 2y = −0 ⇒ y = 0.
∂y ∂x
 3.4. NECESSARY AND SUFFICIENT CONDITION OF DIFFERENTIABILITY 27

Hence for f C-R equations are satisfied only at the origin. Therefore the

function f is differentiable nowhere, except at the origin.

Alternative proof of the second part. To show f (z) = |z|2 is differen-

tiable at the origin, see Problem 3.5.

For z = z0 ̸= 0,

f (z0 + ∆z) − f (z0 )

f ′ (z0 ) = lim
∆z→0 ∆z
|z0 + ∆z|2 − |z0 |2
= lim
∆z→0 ∆z
(z0 + ∆z)(z0 + ∆z) − z0 z0
= lim
∆z→0 ∆z
(z0 + ∆z)(z0 + ∆z) − z0 z0
= lim
∆z→0 ∆z
z0 z0 + z0 ∆z + z0 ∆z + ∆z∆z − z0 z0
= lim
∆z→0 ∆z
z0 ∆z + z0 ∆z + ∆z∆z
= lim
∆z→0 ∆z
∆z
= z0 lim + lim z0 + lim ∆z
∆z→0 ∆z ∆z→0 ∆z→0

∆z
= z0 lim + z0 + 0
∆z→0 ∆z

∆x + i∆y
= z0 + z0 lim
(∆x,∆y)→(0,0) ∆x + i∆y

∆x − i∆y
= z0 + z0 lim
(∆x,∆y)→(0,0) ∆x + i∆y



z0 + z0 , along real axis

= .


z0 − z0 , along imaginary axis


Here along real and imaginary axes limits are different. So f ′ (z0 ) does not

exist at z = z0 ̸= 0.
 3.4. NECESSARY AND SUFFICIENT CONDITION OF DIFFERENTIABILITY 28

Therefore f (z) = |z|2 is continuous everywhere, but not differentiable ex-

cept at the origin.

r
p tt
 CHAPTER 4

Analytic and Harmonic Functions

For solving more problems the reader is referred to [4] and [3].

4.1. Analytic Functions

Definition 4.1.1 Analyticity at a point Let w = f (z) be a function and

z0 ∈ Domain(f ). Then f is said to be analytic at z = z0 , if there exists a

neighbourhood

V = {z : |z − z0 | < δ}

of z = z0 such that f ′ (p) exists for any p ∈ V .

Definition 4.1.2 (Analyticity at a point) [4, Page 3.1] A function w = f (z)

is called analytic at a point z = z0 , if there exists a neighbourhood |z −z0 | <

δ of z = z0 such that at all point of that neighbourhood f ′ (z) exists.

Definition 4.1.3 (Regular point) If a function w = f (z) is analytic at point

z = z0 , then z = z0 is called a ordinary or regular point of w = f (z).

Definition 4.1.4 (Analytic function) [4, Page 3.1] A function w = f (z) is

called analytic in a region Γ, if it is analytic at each point of Γ.

That is a function f is said to be analytic in a region Γ, if every point of Γ

is a regular point of f .

29
 4.1. ANALYTIC FUNCTIONS 30

Note 4.1 [4, Page 3.1] Analytic functions are also referred as regular func-

tions or holomorphic functions.

Theorem 4.1.1 (Necessary condition for w = f (z) to be analytic in a region)

If a function w = f (z) = u(x, y) + iv(x, y) is analytic in a region Γ, then

∂u ∂v ∂u ∂v
= and =− .
∂x ∂y ∂y ∂x

Proof. Let w = f (z) = u(x, y) + iv(x, y) when z = x + iy for x, y ∈ R. Now,

∂f ∂f ∂z ∂f ∂f
= · = ·1= (2)
∂x ∂z ∂x ∂z ∂z

and

∂f ∂f ∂z ∂f ∂f
= · = ·i=i (3)
∂y ∂z ∂y ∂z ∂z

Since f (z) = u(x, y) + iv(x, y),

∂f ∂u ∂v
= +i (4)
∂x ∂x ∂x

and

∂f ∂u ∂v
= +i (5)
∂y ∂y ∂y

A complex function has a derivative with respect to real variable if and only

if real and imaginary part do. Thus by Equation (4) and Equation (5), u

and v have also partial derivative with respect to x and y. Now comparing

Equation (2) and Equation (4),

∂f ∂u ∂v
= +i (6)
∂z ∂x ∂x
 4.1. ANALYTIC FUNCTIONS 31

Again, comparing Equation (3) and Equation (5),

∂f ∂u ∂v
i = +i
∂z ∂y ∂y
∂f 1 ∂u ∂v
⇒ = +
∂z i ∂y ∂y
∂u ∂v
= −i + (7)
∂y ∂y

Comparing real and imaginary part from Equations (6) and (7), we obtain

∂u ∂v ∂u ∂v
= and =− .
∂x ∂y ∂y ∂x

Hence our proof is complete.

Note 4.2 Above Theorem 4.1.1 says, dissatisfaction of the Cauchy-Riemann

equations by a function w = f (z) = u(x, y) + iv(x, y) in a region Γ implies

that w = f (z) is not analytic in that region. So for a function w = f (z) =

u(x, y) + iv(x, y) to be analytic in Γ, it is one of the preconditions that it must

satisfy the Cauchy-Riemann equations in Γ. Though the Cauchy-Riemann

equations hold for a function w = f (z) in Γ, it may not be analytic.

Problem 4.1 Check whether f (z) = 2xy + i(x2 − y 2 ) is analytic or not.

Solution Not analytic.

x − iy
Problem 4.2 Check whether f (z) = is analytic or not.
x2 + y 2
x −y
Solution Here u = and v = 2 . Analytic.
x2 +y 2 x + y2

Problem 4.3 Check whether f (z) = z 3 is analytic or not.

 4.1. ANALYTIC FUNCTIONS 32

Solution Here f (z) = z 3 = (x + iy)3 = (x3 − 3xy 2 ) + i(3x2 y − y 3 ). So

u = x3 − 3xy 2 and v = 3x2 y − y 3 . Analytic.

Problem 4.4 Show that f (z) = |z|2 is differentiable at z = 0, but not

analytic there.

Solution For the 1st part see Problem 3.9.

2nd Part : f (z) = u + iv = |z|2 = |x + iy|2 = x2 + y 2 . So

u = x2 + y 2 and v = 0.

Thus
∂u ∂u ∂v ∂v
= 2x, = 2y, = 0, = 0.
∂x ∂y ∂x ∂y

At z = 0 Cauchy-Riemann equations are not satisfied.

Only satisfaction of Cauchy-Riemann equations by w = f (z) is not the

guarantee of its becoming analytic in a region Γ. The extra condition with the

satisfaction of Cauchy-Riemann equations for becoming w = f (z) analytic in

a region Γ is described in the following theorem.

Theorem 4.1.2 (Sufficient condition for w = f (z) to be analytic in a region)

A function w = f (z) = u(x, y) + iv(x, y) is analytic in a region Γ, if the

following two conditions hold :

∂u ∂v ∂u ∂v
(i) , , are continuous in Γ.
∂x ∂y ∂y ∂x
∂u ∂v ∂u ∂v
(ii) = and =− in the region Γ.
∂x ∂y ∂y ∂x

Question 4.1 Write down the Cauhcy-Riemann partial differential equa-

tions. How are they related to analytic functions?

 4.1. ANALYTIC FUNCTIONS 33

Solution Cauchy-Riemann equations on a complex-variable function w =

f (z) = u(x, y) + iv(x, y) consist of the following two partial differential equa-

tions : 
∂u ∂v
=



∂x ∂y 
(8)
∂u ∂v 
= − .


∂y ∂x
Cauhcy-Riemann equations are related to analytic functions in the follow-

ing manner: A function w = f (z) = u(x, y) + iv(x, y) is analytic in a region Γ,

if and only if the following two conditions hold :

∂u ∂v ∂u ∂v
(i) , , are continuous in Γ.
∂x ∂y ∂y ∂x
∂u ∂v ∂u ∂v
(ii) = and =− in the region Γ.
∂x ∂y ∂y ∂x

Problem 4.5 Find a, b, c, and d so that the function f (z) = x2 + axy +

by 2 + i(cx2 + dxy + y 2 ) is analytic.

Solution See [4, Problem 3.16, Page 3.19].

Problem 4.6 Show that an analytic function with constant modulus is con-

stant.

Proof. Let f (z) = u(x, y) + iv(x, y) be an analytic function with constant

modulus c, that is,

|f (z)| = c

⇒ |u + iv| = c

√
⇒ u2 + v 2 = c

⇒ u2 + v 2 = c2 (9)
 4.1. ANALYTIC FUNCTIONS 34

Differentiating both sides of (9) with respect to x

∂u ∂v ∂u ∂v
2u + 2v =0 ⇒ u +v = 0. (10)
∂x ∂x ∂x ∂x

Similarly, differentiating both sides of (9) with respect to y

∂u ∂v
u +v =0
∂y ∂y
∂v ∂u
⇒ −u +v = 0, by Cauchy-Riemann equations. (11)
∂x ∂x

Squaring (10), (11) and then adding them

 2  2
∂u ∂v ∂v ∂u
u +v + −u +v =0
∂x ∂x ∂x ∂x
(   2 ) (   2 )
2 2
∂u ∂v ∂u ∂v
⇒ u2 + + v2 + =0
∂x ∂x ∂x ∂x
(   2 )
2
∂u ∂v
⇒ u2 + v 2


+ =0 (12)
∂x ∂x

Again,

∂u ∂v
f (z) = u + iv ⇒ f ′ (z) = +i
∂x ∂x
 2  2
∂u ∂v
⇒ |f ′ (z)|2 = + .
∂x ∂x

Thus from Equation (12)

c2 · |f ′ (z)|2 = 0 ⇒f ′ (z) = 0

∴f (z) = constant.

Hence the statement.

 4.1. ANALYTIC FUNCTIONS 35

Note 4.3 (Note 1, Page 90, [3]) The real functions of complex variables

are nowhere analytic unless these are constant valued. Hence f (z) = Re(z),

f (z) = Im(z), f (z) = |z|, f (z) = |z|2 etc are nowhere analytic, since these are

real valued but not constant on any domain in complex plane.

Problem 4.7 [3, Example 14, Page 116] or [4, Problem 3.18, Page 3.20]

Show that the function


 (1 + i)x3 − (1 − i)y 3
̸ 0
, if z =


 2 + y2
f (z) = u + iv = x


0,
 if z = 0.

is continuous and that the Cauchy-Riemann equations are satisfied at the

origin, yet f ′ (z) does not exist there.

Note 4.4 According to the question,

f (z) = 0 at z = 0

⇒u(x, y) + iv(x, y) = 0 + i0 at x + iy = 0 + i0

⇒u(x, y) = 0, v(x, y) = 0 at x = 0, y = 0

∴ u(0, 0) = 0, v(0, 0) = 0.

Solution When z ̸= 0, then

(1 + i)x3 − (1 − i)y 3 x3 − y 3 x3 + y 3
f (z) = u + iv = ⇒ u= 2 and v = 2 ,
x2 + y 2 x + y2 x + y2

which are rational and finite for all z ̸= 0 = (0, 0). So these u, v are continuous

at all those points where z ̸= 0. Hence f (z) is continuous at all those points

where z ̸= 0.
 4.1. ANALYTIC FUNCTIONS 36

When z = 0, then f (0) = 0 by definition of f (z). So u = 0 and v = 0 at

z = 0 = (0, 0). Again,

lim u(x, y) = 0 and lim v(x, y) = 0 ⇒ lim f (z) = 0.

(x,y)→(0,0) (x,y)→(0,0) z→0

Since limz→0 f (z) = f (0), function f (z) is continuous at z = 0.

At the origin z = 0 = 0 + 0i = (0, 0),

(∆x)3
∂u u(0 + ∆x, 0) − u(0, 0) u(∆x, 0) − 0 (∆x)2
= lim = lim = lim = 1,
∂x ∆x→0 ∆x ∆x→0 ∆x ∆x→0 ∆x

−(∆y)3
∂u u(0, 0 + ∆y) − u(0, 0) u(0, ∆y) − 0 (∆y)2
= lim = lim = lim = −1,
∂y ∆y→0 ∆y ∆y→0 ∆y ∆y→0 ∆y
(∆x)3
∂v v(0 + ∆x, 0) − v(0, 0) v(∆x, 0) − 0 (∆x)2
= lim = lim = lim = 1,
∂x ∆x→0 ∆x ∆x→0 ∆x ∆x→0 ∆x

(∆y)3
∂v v(0, 0 + ∆y) − v(0, 0) v(0, ∆y) − 0 (∆y)2
= lim = lim = lim = 1.
∂y ∆y→0 ∆y ∆y→0 ∆y ∆y→0 ∆y

∂u ∂v ∂u ∂v
Clearly, = and = − at z = 0. Hence Cauchy-Riemann
∂x ∂y ∂y ∂x
equations are satisfied by the given function at the origin.

Moreover, at the origin (when z = 0)

f (z) − f (0) f (z) − 0

f ′ (0) = lim = lim
z→0 z−0 z→0 z
(1 + i)x3 − (1 − i)y 3
 
1
= lim ·
(x,y)→(0,0) x2 + y 2 x + iy
 3
(x − y 3 ) + i(x3 + y 3 )

1
= lim ·
(x,y)→(0,0) x2 + y 2 x + iy
 4.1. ANALYTIC FUNCTIONS 37

When z → 0, that is (x, y) → (0, 0), along y = x, then

(x3 − x3 ) + i(x3 + x3 )
 
′ 1
f (0) = lim 2 2
·
(x,y)→(0,0) x +x x + ix
 3 
2ix 1 i
= lim 2
· = lim
x→0 2x x(1 + i) x→0 (1 + i)

i(1 − i) 1+i
= lim =
x→0 (12 − i2 ) 2

Again, when z → 0, that is (x, y) → (0, 0), along the imaginary axis x = 0,

then

(03 − y 3 ) + i(03 + y 3 )
 
′ 1
f (0) = lim 2 2
·
y→0 0 +y 0 + iy
 3 
y (i − 1) 1 i−1
= lim 2
· = lim
y→0 y iy y→0 i
i(i − 1) −1 − i
= lim 2
= = 1 + i.
y→0 i −1

Since f ′ (0) is not unique when z → 0 along different paths, f ′ (z) does not

exist at the origin.

Problem 4.8 [3, Example 13, Page 115] Show that the function

 x3 − 3xy 2 + i(y 3 − 3x2 y)
̸ 0
, if z =



f (z) = u + iv = x2 + y 2


0,
 if z = 0.

is continuous and satisfies the Cauchy-Riemann equations, but is not differen-

tiable at z = 0.

Solution When z ̸= 0, then

x3 − 3xy 2 + i(y 3 − 3x2 y) x3 − 3xy 2 y 3 − 3x2 y

f (z) = u+iv = ⇒ u = and v = ,
x2 + y 2 x2 + y 2 x2 + y 2
 4.1. ANALYTIC FUNCTIONS 38

which are rational and finite for all z ̸= 0 = (0, 0). So these u, v are continuous

at all those points where z ̸= 0. Hence f (z) is continuous at all those points

where z ̸= 0.

When z = 0, then f (0) = 0 by definition of f (z). So u = 0 and v = 0 at

z = 0 = (0, 0). Again,

lim u(x, y) = 0 and lim v(x, y) = 0 ⇒ lim f (z) = 0.

(x,y)→(0,0) (x,y)→(0,0) z→0

Since limz→0 f (z) = f (0), function f (z) is continuous at z = 0.

At the origin z = 0 = 0 + 0i = (0, 0),

(∆x)3
∂u u(0 + ∆x, 0) − u(0, 0) u(∆x, 0) − 0 (∆x)2
= lim = lim = lim = 1,
∂x ∆x→0 ∆x ∆x→0 ∆x ∆x→0 ∆x

0
∂u u(0, 0 + ∆y) − u(0, 0) u(0, ∆y) − 0 (∆y)2
= lim = lim = lim = 0,
∂y ∆y→0 ∆y ∆y→0 ∆y ∆y→0 ∆y

0
∂v v(0 + ∆x, 0) − v(0, 0) v(∆x, 0) − 0 (∆x)2
= lim = lim = lim = 0,
∂x ∆x→0 ∆x ∆x→0 ∆x ∆x→0 ∆x

(∆y)3
∂v v(0, 0 + ∆y) − v(0, 0) v(0, ∆y) − 0 (∆y)2
= lim = lim = lim = 1.
∂y ∆y→0 ∆y ∆y→0 ∆y ∆y→0 ∆y

∂u ∂v ∂u ∂v
Clearly, = and = − at z = 0. Hence Cauchy-Riemann
∂x ∂y ∂y ∂x
equations are satisfied by the given function at t at z = 0.

Moreover, at z = 0

f (z) − f (0) f (z) − 0

f ′ (0) = lim = lim
z→0 z−0 z→0 z
 3
x − 3xy + i(y − 3x2 y)
2 3

1
= lim · .
(x,y)→(0,0) x2 + y 2 x + iy
 4.1. ANALYTIC FUNCTIONS 39

When z → 0, that is (x, y) → (0, 0), along y = x, then

x3 − 3xx2 + i(x3 − 3x2 x)

 
′ 1
f (0) = lim 2 2
·
(x,y)→(0,0) x +x x + ix
−2x3 (1 + i)
 
1
= lim 2
· = −1
x→0 2x x(1 + i)

Again, when z → 0, that is (x, y) → (0, 0), along the imaginary axis x = 0,

then

03 − 3 · 0 · y 2 + i(y 3 − 3 · 02 · y)
 
′ 1
f (0) = lim ·
y→0 02 + y 2 0 + iy
 3 
iy 1
= lim · = 1.
y→0 y 2 iy

Since f ′ (0) is not unique when z → 0 along different paths, f ′ (z) does not

exist at the origin.

Exercise 4.1 [3, Example 33, Page 134] Examine the nature (continuity,

differentiability and analyticity) of the complex function


 x2 y 5 (x + iy)
, whenever z ̸= 0



f (z) = u + iv = x4 + y 10


0,
 otherwise.

at the origin.

Solution Continuous and differentiable. Not analytic.

Exercise 4.2 [3, Example 35, Page 139] Discuss the nature (continuity, dif-

ferentiability and analyticity) of the complex function


 xy 2 (x + iy)
, whenever z ̸= 0



f (z) = u + iv = x2 + y 4


0,
 otherwise.
 4.1. ANALYTIC FUNCTIONS 40

at the origin.

Solution Continuous and differentiable. Not analytic.

Exercise 4.3 [3, Example 32, Page 132] Show that the function

 x3 y 4 (x + iy)
, if z ̸= 0



f (z) = u + iv = x8 + y 8


0,
 otherwise.

is continuous. Also show that though the Cauchy-Riemann equations are sat-

isfied at the origin, f ′ (z) does not exist there.

Problem 4.9 [3, Example 18, Page 121] Show that f (z) = |z|2 is differen-

tiable at z = 0, but not analytic there.

Solution For 1st part see Problem 3.5. Here

f (z) = |z|2 ⇒ u + iv = x2 + y 2 ⇒ u = x2 + y 2 and v = 0.

So
∂u ∂u ∂v ∂v
= 2x, = 2y, = 0 and = 0.
∂x ∂y ∂x ∂y

From above it is clear that at z = 0, Cauchy-Riemann equations are satisfied.

But there is no neighbourhood of z = 0 at all points of which Cauchy-Riemann

equations are satisfied by f (z) = |z|2 . Therefore f (z) = |z|2 is not analytic at

the origin.

4.1.1. Difference between differentiability and analyticity of a

function at a point. See [2, Part 08, 13 min 19 sec] To become a func-

tion f is differentiable at a point z = z0 , it is must to exist f ′ (z) at z = z0

 4.1. ANALYTIC FUNCTIONS 41

only. But to become a function f analytic at z = z0 , it is not only must to

exist f ′ (z) at z = z0 , but also must to exist at all other points of (at least)

a neighbourhood |z − z0 | < δ of z = z0 . Thus a function f differentiable at

a point z = z0 may or may not be analytic at z = z0 . But if a function f is

analytic at a point z = z0 , then it must be differentiable at z = z0 .

4.1.2. Difference between differentiability and analyticity of a

function in a domain. See [2, Part 08, 17 min 39 sec] In case of domain,

both concepts are same.

Note 4.5 Concept of analyticity and differentiability of a function at a point

are not same.

Note 4.6 A differentiable function at a point may or may not be analytic at

that point. For example [2, Part 08, 19 min], f (z) = |z|2 is differentiable at the

origin, but is not analytic there (Problem 3.9). Because though at the origin
∂u ∂v ∂u ∂v
= and = − , there is no neighbourhood of z = z0 = (0, 0) = 0+0i
∂x ∂y ∂y ∂x
at all points of which C-R equations are satisfied.

Note 4.7 If a function analytic at a point, then it must be differentiable at

that point.

Note 4.8 [2, Part 08, 17 min & 22 min] The notions of analyticity and

differentiability of a function are same, when these are considered in a domain.

Theorem 4.1.3 [2, Part 08, 24 min] If a function is analytic in a domain Γ,

then it is infinitely many times differentiable (analytic) in Γ.

 4.1. ANALYTIC FUNCTIONS 42

That is, if a function is analytic in a domain Γ, then its any order derivative

is differentiable and also analytic in Γ.

Note 4.9 (Importance of the Theorem 4.1.3) According to [2, Part 08,

25 min] Theorem 4.1.3 is sometimes referred as the heart of complex analysis.

The concept of harmonic functions depends on this theorem. It is impossible

to develop the notion of harmonic function without this theorem. This type

of theorem was not there in case of real functions.

Theorem 4.1.4 [2, Part 08, 29 min] If a function is differentiable in a do-

main Γ, then it is infinitely many times differentiable in Γ.

Note 4.10 [2, Part 08, 31 min 30 sec] If a function is differentiable at only

finitely many points, then it is nowhere analytic.

Theorem 4.1.5 [2, Part 08, 35 min] If a function is analytic at some points,

then it is analytic at uncountable number of points.

∂f ∂f ∂ 2 f
Theorem 4.1.6 (Young’s theorem) [2, Part 08, 39 min] If , ,
∂x ∂y ∂x∂y
∂ 2f ∂ 2f
exist at a point (a, b) and is continuous at (a, b), then is also
∂x∂y ∂y∂x
∂ 2f ∂ 2f
continuous at (a, b) and = .
∂x∂y ∂y∂x

Theorem 4.1.7 [2, Part 08, 42 min 50 sec] If a function f = u + iv is

analytic in a domain Γ, then all order partial derivatives of u and v exist and
∂ 2u ∂ 2u ∂ 2v ∂ 2v
continuous. Moreover, then = and = .
∂x∂y ∂y∂x ∂x∂y ∂y∂x

Note 4.11 [1, Section 24, Page 74] Composition of two analytic functions is

analytic.
 4.2. HARMONIC FUNCTIONS 43

4.2. Harmonic Functions

Definition 4.2.1 (Harmonic functions) [2, Part 09] A real-valued function

u of x and y which possesses continuous partial derivatives of first and second

∂ 2u ∂ 2u
order and satisfies Laplace’s equation + 2 = 0, is called a harmonic
∂x2 ∂y
function.

Note 4.12 It is to be noted that to define harmonic function, we have taken

a real-valued function u of x and y.

Example 4.1 u = x2 − y 2 is a harmonic function.

−y
Example 4.2 The real-valued function v = is harmonic.
x2+ y2

Problem 4.10 (Problem 3,7(a), Page 3.13 [4]) Show that the function

u = e−x (x sin y − y cos y)

is harmonic.

Theorem 4.2.1 [4, Problem -3.6, Page 3.12-3.13] If a function f (z) =

u(x, y)+iv(x, y) is analytic in a region Γ, then its component functions u and v

are harmonic in Γ provided u, v have continuous 2nd partial derivatives in Γ.

or, [3, Theorem 6, Page 99] The real and imaginary parts of an analytic

functions are harmonic.

 4.2. HARMONIC FUNCTIONS 44

Proof. Let f (z) = u(x, y) + iv(x, y) be analytic in a region Γ. So it satisfies

the Cauchy-Riemann equations


∂u ∂v
=



∂x ∂y 
(13)
∂u ∂v 
= − .


∂y ∂x
in Γ.

Differentiating partially both sides of each equation of (13) with respect

to x,
∂ 2u ∂ 2v ∂ 2u ∂ 2v
= and = − 2.
∂x2 ∂x∂y ∂x∂y ∂x

Again, differentiating partially both sides of each equation of (13) with

respect to y,
∂ 2u ∂ 2v ∂ 2u ∂ 2v
= 2 and = − .
∂y∂x ∂y ∂y 2 ∂y∂x

Since u, v have continuous second partial derivatives in Γ,

∂ 2u ∂ 2u ∂ 2v ∂ 2v
= and =
∂x∂y ∂y∂x ∂x∂y ∂y∂x
∂ 2v ∂ 2v ∂ 2u ∂ 2u
⇒ − = and = −
∂x2 ∂y 2 ∂x2 ∂y 2
∂ 2v ∂ 2v ∂ 2u ∂ 2u
⇒ + = 0 and + =0
∂x2 ∂y 2 ∂x2 ∂y 2
∂ 2u ∂ 2u ∂ 2v ∂ 2v
∴ + = 0 and + = 0.
∂x2 ∂y 2 ∂x2 ∂y 2

Therefore u and v are harmonic functions. Hence our proof is complete.

Contrapositive statement, used to show that a function is not analytic, of

the Theorem 4.2.1 is as follows.

Theorem 4.2.2 If at least one of u and v is not harmonic, then f = u + iv

is not analytic.
 4.3. CONSTRUCTION OF ANALYTIC FUNCTION 45

∂ 2u ∂ 2u
Note 4.13 By means of Theorem 4.1.7 when f is analytic, then , ,
∂x∂y ∂y∂x
∂ 2v ∂ 2v
, exists and continuous. So then
∂x∂y ∂y∂x

∂ 2u ∂ 2u ∂ 2v ∂ 2v
= and =
∂x∂y ∂y∂x ∂x∂y ∂y∂x

Thus it is not a must to explicitly write down the underlined assumption of

the Theorem 4.2.1 (like in its alternative statement).

Note 4.14 The converse of the Theorem 4.2.1 may not be true. For example,
y
both u(x, y) = x2 −y 2 and v(x, y) = − are harmonic. But f (z) = u+iv
x2 + y 2
is not analytic, as ux ̸= vy and uy ̸= −vx . For detailed calculation, see [4,

Problem 3.23, Page 3.22].

4.3. Construction of Analytic Function

4.3.1. Harmonic conjugate. [3, Page 99] The function v is said to be a

harmonic conjugate of u, if u, v are harmonic and u, v satisfy the Cauchy-

Riemann equations.

Definition 4.3.1 (Harmonic conjugate) [2, Part 09, 14 min] A function v is

said to be a harmonic conjugate of u, if f = u + iv is analytic.

4.3.2. Methods to construct an analytic function. Usually following

two methods are employed to construct an analytic function.

(i) Exact differential method

(ii) Milne Thomson method.

 4.4. EXACT DIFFERENTIAL METHOD 46

4.4. Exact Differential Method

4.4.1. Exact differential method to construct analytic function.

When one of u, v is given and another one is asked to find, then the exact

differential method is applied to construct the analytic function f = u + iv.

4.4.2. Real part given. Let u of the analytic function f (z) = u + iv be

given and we have to find v. From the definition of total differential,

∂v ∂v
dv = dx + dy
∂x ∂y
 
∂u ∂u
= − dx + dy, by Cauchy-Riemann equations (14)
∂y ∂x
 
∂u ∂u
= df (x, y), where df (x, y) = − dx + dy
∂y ∂x

∴ v = f (x, y) + c.

Alternatively. After (14) we solve it by

Z Z Z
∂u ∂u
v= dv = − dx + dy,
∂y ∂x

∂u ∂u
provided that should not contain x. If contains any terms in x, then
∂x ∂x
we assume it equals to 0.

Note 4.15 When one of u, v is given and another one is asked to find, then

at first check whether the given one is harmonic or not. If the given one, say u,

is not harmonic, then it is not necessary to find the other one v. Because, by

Theorem 4.2.2 if one of u, v are not harmonic, then f = u + iv is not analytic.

So if u is not harmonic, then f is not analytic.

 4.4. EXACT DIFFERENTIAL METHOD 47

Problem 4.11 Show that the function u = ex cos y is harmonic and find its

harmonic conjugate v.

Solution Given that

u = ex cos y. (15)

Differentiating (15) partially with respect to x and y respectively

∂u
= ex cos y (16)
∂x

and

∂u
= −ex sin y. (17)
∂y

Differentiating (16) partially with respect to x,

∂ 2u
= ex cos y (18)
∂x2

Differentiating (17) partially with respect to y,

∂ 2u
= −ex cos y (19)
∂y 2

Adding (18) and (19),

∂ 2u ∂ 2u
+ = 0.
∂x2 ∂y 2

Therefore u satisfies Laplace’s equation. Moreover, first and second derivatives

of u are continuous. Therefore u is harmonic.

 4.4. EXACT DIFFERENTIAL METHOD 48

Now

∂v ∂v
dv = dx + dy
∂x ∂y
 
∂u ∂u
= − dx + dy, by Cauchy-Riemann equations
∂y ∂x
∂u ∂u
= ex sin ydx + ex cos ydy, since = ex sin y, = ex cos ydy
∂y ∂x

= d(ex sin y)
Z Z
∴v= dv = d(ex sin y) = ex sin y + c.

Therefore

f (z) = u(x, y) + iv(x, y)

= ex cos y + i(ex sin y + c)

= ex (cos y + i sin y) + ic

= ex .eiy + ic

= ex+iy + ic = ex+iy + c′ , where c′ = ic.

Answer v = ex sin y + c.

Note 4.16 M dx + N dy = d(x, y). That means, M dx + N dy is expressed in

a form d(x, y) which can be integrated easily.

Problem 4.12 If the real part of an analytic function is given by u =

e−y cos x, then find its imaginary part.

Solution Here u = e−y cos x is harmonic.

 4.4. EXACT DIFFERENTIAL METHOD 49

We have

∂v ∂v
dv = dx + dy
∂x ∂y
 
∂u ∂u
= − dx + dy, by Cauchy-Riemann equations
∂y ∂x
∂u ∂u
= e−y cos xdx − e−y sin xdy, since = −e−y cos x, = −e−y sin x
∂y ∂x
Z Z Z
−y
∴v= dv = e cos x dx − e−y sin x dy.

∂u
But here = −e−y sin x contains terms in x. Thus we assume that
∂x
Z
− e−y sin x dy = 0.

e−y cos x dx = e−y sin x + c.

R
Therefore v =

4.4.3. Imaginary part given. Let v of the analytic function f (z) =

u + iv be given and we have to find u. From the definition of total differential,

∂u ∂u
du = dx + dy
∂x ∂y
   
∂v ∂v
= dx + − dy, by Cauchy-Riemann equations (20)
∂y ∂x
   
∂v ∂v
= df (x, y), where df (x, y) = dx + − dy
∂y ∂x

∴ u = f (x, y) + c.

Alternatively. After (20) we solve it by

Z Z Z
∂v ∂v
u= du = dx − dy,
∂y ∂x

∂v ∂v
provided that should not contain x. If contains any terms in x, then
∂x ∂x
we assume it equals to 0.
 4.4. EXACT DIFFERENTIAL METHOD 50

Problem 4.13 Find the harmonic conjugate of v = − sin x sinh y.

Solution Given that

v = − sin x sinh y. (21)

Differentiating (21) partially with respect to x and y respectively

∂v
= − cos x sinh y (22)
∂x

and

∂v
= − sin x cosh y. (23)
∂y

Differentiating (22) partially with respect to x,

∂ 2v
= sin x sinh y (24)
∂x2

Differentiating (23) partially with respect to y,

∂ 2v
= − sin x sinh y (25)
∂y 2

Adding (24) and (25),

∂ 2v ∂ 2v
+ = 0.
∂x2 ∂y 2

Therefore v satisfies Laplace’s equation. Moreover, first and second derivatives

of v are continuous. Therefore v is harmonic.

 4.5. MILNE THOMSON METHOD 51

Now

∂u ∂u
du = dx + dy
∂x ∂y
 
∂v ∂v
= dx + − dy, by Cauchy-Riemann equations
∂y ∂x

= − sin x cosh ydx + cos x sinh ydy,

∂v ∂v
since dx = − sin x cosh y, = − cos x sinh y
∂y ∂x

= d(cos x cosh y)
Z Z
∴u= du = d(cos x cosh y) = cos x cosh y + c.

Therefore

f (z) = u(x, y) + iv(x, y)

= (cos x cosh y + c) + i(− sin x sinh y).

Answer u = cos x cosh y + c.

Problem 4.14 If w(z) = φ(x, y) + iψ(x, y), represents the complex potential
x
for an electric field and ψ(x, y) = x2 − y 2 + , then determine the real-
x2 + y2
valued function φ.

Solution See [4, Problem 3.11, Page 3.18].

4.5. Milne Thomson Method

4.5.1. If the real part is given. Let the real part u of the analytic

function f (z) = u + iv be given. To construct v we follow the undermentioned

steps :
 4.5. MILNE THOMSON METHOD 52

∂u ∂u
(i) Find and .
∂x ∂y
(ii) Apply Cauchy-Riemann equations.

(iii) Find

∂u ∂v
f ′ (z) = +i
∂x ∂x
∂u ∂u
= −i (26)
∂x ∂y

(iv) Replace x by z and y by 0 in f ′ (z) of (26).

(v) Integrate the f ′ (z) obtained in the Step (iv) with respect to z.

Briefly, if u is given, then f can be constructed following the rule

Z
f (z) = [ux (z, 0) − iuy (z, 0)] dz + c.

Problem 4.15 Construct the analytic function f (z) = u(x, y)+iv(x, y) such

that its real part is given as u(x, y) = ex (x cos y − y sin y).

Solution Given that

u = ex (x cos y − y sin y). (27)

Differentiating (27) partially with respect to x and y respectively

∂u
= ex (x cos y − y sin x) + ex cos y (28)
∂x

and
∂u
= ex (−x sin y − y cos y − sin y). (29)
∂y

Differentiating (28) partially with respect to x,

∂ 2u
= ex (x cos y − y sin y) + 2ex cos y. (30)
∂x2
 4.5. MILNE THOMSON METHOD 53

Differentiating (29) partially with respect to y,

∂ 2u
= e(−x cos y + y sin y − 2 cos y). (31)
∂y 2

Adding (30) and (31),

∂ 2u ∂ 2u
+ = 0.
∂x2 ∂y 2

Therefore u satisfies Laplace’s equation. Moreover, first and second derivatives

of u are continuous. Therefore u is harmonic.

Hence

Z
f (z) = [(ez z + ez ) − i(0)] dz + c
Z Z
z z
= [e z + e ] dz + c = ez [z + 1] dz + c

= (z + 1)ez − ez + c, integration by parts

= zez + c.

Answer f (z) = zez + c.

Problem 4.16 Construct the analytic function f (z) = u(x, y)+iv(x, y) such

that its real part is given as u(x, y) = ex (x cos 2y − y sin 2y).

Solution Here u is harmonic.

Now

∂u
= ex (cos 2y) + (x cos 2y − y sin 2y)ex
∂x

and

∂u
= ex (−2x sin 2y − [2y cos 2y + sin 2y]).
∂y
 4.5. MILNE THOMSON METHOD 54

Now

∂u ∂v
f ′ (z) = +i
∂x ∂x
∂u ∂u
= −i
∂x ∂y

= ex (cos 2y + x cos 2y − y sin 2y) + iex (2x sin 2y + 2y cos 2y + sin 2y)

(32)

Replacing x by z and y by 0 in (32),

f ′ (z) = ez (z + 1)
Z Z
′
⇒ f (z)dz = ez (z + 1)dz

⇒ f (z) = (z + 1)ez − 1 · ex + c

∴ u + iv = zez + c = (x + iy)ex+iy + c

Therefore f (z) = zez + c.

4.5.2. If the imaginary part is given. Let the imaginary part v of

the analytic function f (z) = u + iv be given. To construct u we follow the

undermentioned steps :
∂v ∂v
(i) Find and .
∂x ∂y
(ii) Apply Cauchy-Riemann equations.

(iii) Find

∂u ∂v
f ′ (z) = +i
∂x ∂x
∂v ∂v
= +i (33)
∂y ∂x

(iv) Replace x by z and y by 0 in f ′ (z) of (33).

 4.5. MILNE THOMSON METHOD 55

(v) Integrate the obtained f ′ (z) obtained in the Step (iv) with respect

to z.

Briefly, if v is given, then f can be constructed following the rule

Z
f (z) = [vy (z, 0) + iux (z, 0)] dz + c.

Problem 4.17 Find the analytic function f (z) = u + iv, imaginary part of

which is given as v = ex (x sin y + y cos y).

Solution Here u is harmonic.

Now
∂v
= ex (sin y + 0) + (x sin y + y cos y)ex
∂x

and
∂v
= ex (cos x + y(− sin y) + cos y · 1) + 0.
∂y

Now

∂u ∂v
f ′ (z) = +i
∂x ∂x
∂v ∂v
= +i
∂y ∂x

= ex (cos x − y sin y + cos y) + iex (sin y + x sin y + y cos y) (34)

Replacing x by z and y by 0 in (34),

f ′ (z) = ez (z + 1)
Z Z
′
⇒ f (z)dz = ez (z + 1)dz

⇒ f (z) = (z + 1)ez − 1 · ex + c

∴ u + iv = zez + c = (x + iy)ex+iy + c.
 4.5. MILNE THOMSON METHOD 56

Therefore f (z) = zez + c.

Answer f (z) = zez + c.

Problem 4.18 If f (z) = u + iv is analytic and u − v = ex (cos y − sin y), then

find f (z).

Solution Given that f (z) = u + iv and u − v = ex (cos y − sin y). Now,

(u − v) + i(u + v) = u + i2 v + iu + iv

= (u + iv) + i(u + iv)

= f (z) + if (z)

= (1 + i)f (z)

⇒ U + iV = F (z), letting u − v = U, u + v = V, (1 + i)f (z) = F (z).

ey − cos x + sin x
Problem 4.19 If f (z) = u + iv is analytic and u − v = ,
cosh y − cos x
π  1
then find f (z) subject to the condition f = (3 − i).
2 2
∂2 ∂2 ∂2 ∂2
Theorem 4.5.1 + = 4 = 4 .
∂x2 ∂y 2 ∂z∂z ∂z∂z

Proof. See [3, Page 105–106].

Theorem 4.5.2 If f (z) is analytic, then

∂2 ∂2
 
2
+ |f (z)|p = p2 |f (z)|p−2 |f ′ (z)| .
∂x2 ∂y 2
 4.5. MILNE THOMSON METHOD 57

∂2 ∂2 ∂2 ∂2
Proof. We have + = 4 = 4 . So
∂x2 ∂y 2 ∂z∂z ∂z∂z
p p
∂2 ∂2 ∂2 ∂2
 
p 2

+ |f (z)| = 4 |f (z)| 2 = 4 (f (z)f (z)) 2
∂x2 ∂y 2 ∂z∂z ∂z∂z
p p
∂ ∂
= 4 (f (z)) 2 · (f (z)) 2
∂z ∂z
p p
p −1 p −1
= 4 · · (f (z)) 2 · f (z) · · (f (z)) 2 · f ′ (z)
′
2 2
p
−1
= p (f (z)f (z)) 2 · f ′ (z)f ′ (z)
2

p
2
−1 2
= p2 |f (z)| 2 |f ′ (z)|

2
= p2 |f (z)|p−2 |f ′ (z)| .

Hence the proof is complete.

Problem 4.20 If f (z) is analytic, then show that

 2
∂2

∂
(a) 2
+ |f (z)|2 = 4 |f ′ (z)|2 .
 ∂x2 ∂y 2 
∂ ∂2
(b) 2
+ 2
|f (z)|3 = 9 |f (z)| |f ′ (z)|2 .
 ∂x2 ∂y 
∂ ∂2
(c) + |f (z)|7 = 49 |f (z)|5 |f ′ (z)|2 .
∂x2 ∂y 2

Solution Immediate consequence of Theorem 4.5.2.

 CHAPTER 5

Complex Integration

For solving more problems the reader is referred to [4] and [3].

5.1. Complex Line Integral

5.1.1. Parameterization of some necessary curves. Let z0 and z1 be

two points and t is a parameter. Then parametric equation of

(i) a line joining two points z0 and z1 is z(t) = (1 − t)z0 + tz1 , when

−∞ < t < ∞.

(ii) the line segment between the points z0 and z1 is z(t) = (1 − t)z0 + tz1 ,

when 0 ≤ t ≤ 1.

Moreover, parametric equation of a circle with center at z0 and radius r is

z(θ) = z0 + r(cos θ + i sin θ), 0 ≤ θ ≤ 2π,

where θ is a parameter.

5.1.2. Complex line integral. See [4, Page 4.1]

5.1.3. Connection between complex and real line integrals. [4,

Page 4.2] Let f (z) = u(x, y) + iv(x, y) = u + iv. Then the complex line
R
integral c
f (z) dz can be expressed as real line integrals as below :

Z Z Z Z
f (z) dz = (u + iv)(dx + idy) = udx − vdy + i vdx + udy.
c c c c

58
 5.1. COMPLEX LINE INTEGRAL 59

R
Problem 5.1 Evaluate c
(z + 3)dz, when C is given by x = 2t, y = 4t − 1,

1 ≤ t ≤ 3.

Solution Here

z = x + iy = 2t + i(4t − 1)

⇒ dz = 2dt + 4idt = (2 + 4i)dt.

Therefore

Z Z 3
(z + 3)dz = [2t + i(4t − 1) + 3] (2 + 4i)dt
c 1
Z 3
= (4t + 8ti − 2i + 6 + 8ti − 16t + 4 + 12i) dt
1
Z 3
= (−12t + 16ti + 10i + 10) dt
1

3
= −6t2 + 8t2 i + 10it + 10t 1


= (−54 + 72i + 30i + 30) − (−6 + 8i + 10i + 10)

= −24 + 102i − 4 − 18i = 84i − 28.

Answer 84i − 28.

R
Problem 5.2 Evaluate c
(2z − z)dz, when C is given by x = −t, y = t2 + 2,

0 ≤ t ≤ 2.

Solution Here

z = x + iy = −t + i(t2 + 2)

⇒ dz = −dt + 2itdt = (2it − 1)dt.

 5.1. COMPLEX LINE INTEGRAL 60

Now, z = x − iy = −t − i(t2 + 2). Therefore

Z Z 2
2{−t − i(t2 + 2)} − {−t + i(t2 + 2)} (2it − 1)dt
 
(2z − z)dz =
c 0
Z 2
= (−t − 3it2 − 6i)(2it − 1)dt
0
Z 2
= (−12t + 16ti + 10i + 10) dt
0
Z 2
−2it2 + 6t3 + 12t + t + 3it2 + 6i dt
 
=
0
Z 2  2
2

1 3 3 4 13 2
3
= it + 6t + 13t + 6i dt = i t + t + t + 6it
0 3 2 2 0
 
8 44
= i + 24 + 26 + 12i − 0 = 50 + i .
3 3
44
Answer 50 + i .
3
R
Problem 5.3 Evaluate c
z 2 dz, when C is given by z(t) = 3t + i2t, −2 ≤

t ≤ 2.

Solution Here

z = 3t + i2t ⇒ dz = 3dt + 2idt = (3 + 2i)dt.

Therefore

Z Z 2 Z 2
2 2
z dz = (3t + i2t) (3 + 2i)dt = (9t2 + i12t2 − 4t2 )(3 + 2i)dt
c −2 −2
Z 2 Z 2
2 2
= (5t + i12t )(3 + 2i)dt = (15t2 + i36t2 + i10t2 − 24t2 )dt
−2 −2
2 2 2
t3 t3
Z  
2 2 3 46 3
= (−9t + i46t )dt = −9 + i46 = −3t + i t
−2 3 3 −2 3 −2
   
368 368
= −24 + i − 24 − i
3 3
736
= −48 + i
3
 5.1. COMPLEX LINE INTEGRAL 61

736
Answer −48 + i .
3
R
Problem 5.4 Evaluate C
(x2 + iy 3 ) dz, where C is the straight line from

z = −1 to z = i.

Solution Parametric equation of the line segment joining the points z = −1

and z = i is

z(t) = (1 − t)(−1, 0) + t(0, 1), 0 ≤ t ≤ 1

= 1(−1, 0) − t(−1, 0) + (0, t), 0 ≤ t ≤ 1

= (−1, 0) + (t, 0) + (0, t), 0 ≤ t ≤ 1

= (−1 + t, t), 0 ≤ t ≤ 1 = −1 + t + it, 0 ≤ t ≤ 1

⇒ dz = dt + idt = (1 + i)dt

Comparing z = −1 + t + it with z = x + iy, we obtain the parametric equation

of C as

x = −1 + t, y = t, 0 ≤ t ≤ 1.

Therefore

Z Z 1
2 3
(−1 + t)2 + it3 (1 + i)dt
 
(x + iy ) dz =
C 0
Z 1
1 − 2t + t2 + it3 (1 + i)dt
 
=
0
Z 1 2
t − 2t + 1 + it3 + it2 − 2it + i − t3 dt

=
0
1
t3 t4 t3 t4

= − t2 + t + i + i − it2 + it −
3 4 3 4 0

1
= (1 + 7i).
12
 5.1. COMPLEX LINE INTEGRAL 62

1
Answer (1 + 7i).
12
R
Problem 5.5 Evaluate C
ez dz, where C is the polynomial path consisting

of the line segments z = 0 to z = 2 and z = 2 to z = 1 + πi.

Solution Here

z = x + iy = 0 ⇒ x = 0, y = 0,

z = x + iy = 2 ⇒ x = 2, y = 0

and z = x + iy = 1 + πi ⇒ x = 1, y = π.

Let C1 denote the line segment from z = 0 to z = 2 and C2 denote the line

segment from z = 2 to z = 1 + πi. So C = C1 ∪ C2 . Thus the integral can be

written as
Z Z Z
z z
e dz = e dz + ez dz.
C C1 C2

Parametric equation of the line segment C1 is

z = (1 − t)0 + t · 2 ⇒ z = 2t, 0 ≤ t ≤ 1.

So dz = 2dt along C1 .

Parametric equation of the line segment C2 is

z = (1 − t)2 + t(1 + iπ), 0 ≤ t ≤ 1

= 2 − 2t + t + iπt, 0 ≤ t ≤ 1

= 2 − t + πit, 0 ≤ t ≤ 1

= 2 + (πi − 1)t, 0 ≤ t ≤ 1.
 5.1. COMPLEX LINE INTEGRAL 63

So dz = (πi − 1)dt along C2 .

Therefore

Z Z Z
z z
e dz = e dz + ez dz
C C1 C2
Z 1 Z 1
2t
= e 2dt + e[2+(πi−1)t] (πi − 1)dt
0 0
2t 1 1
  Z
e
=2 + (πi − 1) e2 e(πi−1)t dt
2 0 0
 1
2 2 (πi − 1)t
= (e − 1) + (πi − 1)e
πi − 1 0

= (e2 − 1) + e2 e(πi−1) − 1
 

= e2 − 1 + e(πi+1) − e2 = −1 + eπi e

= −1 + e(cos π + i sin π)

= −1 + e(−1 + 0.i) = −1 − e.

Answer −1 − e.

R z+1
Problem 5.6 Evaluate C
dz, where C is the right half circle |z| = 1
z
from z = −i to z = i.

Solution Here

z = x + iy = −i ⇒ x = 0, y = −1,

z = x + iy = i ⇒ x = 0, y = 1.

Equation of the curve C is

|z| = 1 ⇒ z = eiθ , 0 ≤ θ ≤ π.
 5.1. COMPLEX LINE INTEGRAL 64

So dz = ieiθ dθ along C. Therefore

Z Z  
z+1 1
dz = 1+ dz
C z C z
Z π Z π
−iθ

iθ
1 + eiθ dθ


= 1+e ie dθ = i
0 0
 iθ
π  iπ

e e 1
=i +θ =i +π−
i 0 i i
 
cos π + i sin π 1
=i +π−
i i
 
−1 1
=i +π−
i i
 
2
=i π− = πi − 2.
i

Answer πi − 2.
 
R 1 5
Problem 5.7 Evaluate C
− + 8 dz, where C is the circle
(z + i)3 z + i
|z + i| = 1.

Solution Here equation of the curve C is

|z + i| = 1 ⇒ z + i = eiθ , 0 ≤ θ ≤ 2π.
 5.1. COMPLEX LINE INTEGRAL 65

So dz = ieiθ dθ along C. Therefore

Z  
1 5
− + 8 dz
C (z + i)3 z + i
Z 2π  
1 5
= 3iθ
− iθ + 8 ieiθ dθ
0 e e
Z 2π
e−3iθ − 5e−iθ + 8 eiθ dθ


=i
0
Z 2π
e−2iθ − 5 + 8eiθ dθ


=i
0
2π
e−2iθ eiθ

=i − 5θ + 8
−2i i 0
 −4iπ
e2πi
  
e −1 1
=i − 10π + 8 − +0+8
−2i i 2i i
 
cos 4π − i sin 4π 8(cos 2π + i sin 2π) −1 + 16
=i − 10π + −
−2i i 2i
 
1 8 −15
=i − 10π + −
−2i i 2i
 
−1 + 16 − 15
= i −10π +
2i

= i [−10π + 0] = −10πi.

Answer −10πi.

R 1
Problem 5.8 Show that C
dz = 2πi, when C is the circle |z −z0 | = r.
z − z0

Proof. Here equation of the curve C is

|z − z0 | = r ⇒ z − z0 = reiθ , 0 ≤ θ ≤ 2π.

So dz = rieiθ dθ along C.
 5.1. COMPLEX LINE INTEGRAL 66

Therefore

Z Z
1 1
dz = dz
C (z − z0 )n C z − z0
Z 2π
1
= iθ
rieiθ dθ
0 re
Z 2π Z 2π
= idθ = i dθ
0 0

= i [θ]2π
0 = i [2π − 0] = 2πi.

Hence the statement.

R 1
Problem 5.9 Show that C
dz = 0, when C is the circle |z−z0 | = r.
(z − z0 )2

Proof. Here equation of the curve C is

|z − z0 | = r ⇒ z − z0 = reiθ , 0 ≤ θ ≤ 2π.

So dz = rieiθ dθ along C.

Therefore

Z Z
1 1
dz = dz
C (z − z0 )2 C (z − z0 )2
2π Z 2π
i 2π −iθ
Z Z
1 iθ 1
= rie dθ = idθ = e dθ
0 (reiθ )2 0 reiθ r 0
2π
i e−iθ

1 2π
= = − e−iθ 0
r −i 0 r
1  −i2π  1
=− e − 1 = − [cos 2π − i sin 2π − 1]
r r
1
= − [1 − 0 − 1] = 0.
r

Hence the statement.

 5.1. COMPLEX LINE INTEGRAL 67

The next theorem generalizes all problems like Problem 5.8 and Prob-

lem 5.9.

Theorem 5.1.1 [2, Part 12, 23 min] If z0 is an interior point of C and n ∈ N,

then



Z
1 2πi, for n = 1

dz =
C (z − z0 )n 

0,
 for n ̸= 1,

when C is the circle |z − z0 | = r described in the positive sense.

Proof. Here equation of the curve C is

|z − z0 | = r ⇒ z − z0 = reiθ , 0 ≤ θ ≤ 2π.

So dz = rieiθ dθ along C. Thus

Z Z 2π
1 1 iθ
dz = n rie dθ
C (z − z0 )n 0
iθ
(re )
Z 2π
i 1
= n−1 dθ
r 0 [e ]n−1
iθ

Z 2π
i
= n−1 e(1−n)iθ dθ.
r 0

Therefore when n = 1, then

Z Z 2π
1 i
n
dz = 0 dθ
C (z − z0 ) r 0

= i [θ]2π
0 = 2πi.
 5.2. CAUCHY’S THEOREM 68

and when n ̸= 1, then

Z Z 2π
1 i
n
dz = n−1 e(1−m)iθ dθ
C (z − z0 ) r 0
2π 2π
e(1−m)iθ e(1−m)iθ
 
i 1
= =
rn−1 i(1 − n) rn−1
0 0 1−n
 (1−m)i2π 2π
1 e 1
= −
rn−1 1−n 1−n 0
 2π
1 cos(1 − n)2π + i sin(1 − n)2π 1
= −
rn−1 1−n 1−n 0
 2π
1 1+0 1
= −
rn−1 1−n 1−n 0
 2π
1 1 1
= − = 0.
rn−1 1−n 1−n 0

Hence the statement.

5.2. Cauchy’s Theorem

5.2.1. Simply and multiply connected regions. [4, Page 4.3] A simply-

connected region is that one which has no hole in it. On the other hand, a

multiply-connected is that one which has one or more holes in it.

Definition 5.2.1 If any simple curve lying in a region Γ can be shrunk to a

point without leaving Γ, then that region is called simply connected. If a

region Γ is not simply connected is called a multiply-conneted region.

Example 5.1 See [4, Page 4.3].

5.2.2. Convention regarding traversal of a closed curve. See [4,

Page 4.4] The boundary C of a region Γ is said to be traversed in the positive

sense, if the observer traversing perpendicularly to the plane in that direction

 5.2. CAUCHY’S THEOREM 69

which has the region Γ to the left. When the boundary curve C of a region Γ

becomes a circle, then its positive direction is the counterclockwise direction.

Note 5.1 [4, Page 4.4] To mean integration of f (z) around the boundary C

in the positive sense we use the symbol

I
f (z) dz.
C

Definition 5.2.2 (Contour integral) [4, Page 4.4] The integral around the

boundary C of a region is called the contour integral.

Definition 5.2.3 (Singular point) If a function w = f (z) is not differentiable

at z = z0 , then the point z = z0 is called the singular point of that function.

7−z
Example 5.2 z = 4 is the singular point of the function f (z) = .
z−4

Definition 5.2.4 (Entire function) A function w = f (z) is called an entire

function, if it is differentiable everywhere in the complex plane.

Definition 5.2.5 (Alternative definition of entire function) [2, Part 08,

28 min] A function w = f (z) is called an entire function, if it is analytic

everywhere in the whole complex plane.

Example 5.3 Since all of the 7 functions of Problem 3.6 are differentiable

everywhere in the complex plane, these are entire functions.

The following theorem also known as Cauchy’s fundamental theorem

was first proved with the help of Green’s theorem (stated in Appendix A).
 5.2. CAUCHY’S THEOREM 70

Theorem 5.2.1 (Cauchy’s theorem) If f (z) is analytic in a region Γ and on

its boundary C with derivative f ′ (z) which is continuous everywhere in Γ and

on C, then

I
f (z) dz = 0.
C

Proof. Since f (z) is analytic in a region Γ and on its boundary C,

f (z + ∆z) − f (z)
f ′ (z) = lim
∆z→0 ∆z

exits and is unique independent of the manner ∆z → 0. So

f ′ (z) when ∆z → 0 along real axis

= f ′ (z) when ∆z → 0 along imaginary axis

∂ ∂x ∂ ∂y
⇒ (u + iv) = (u + iv)
∂x ∂z ∂y ∂z
   
∂u ∂v 1 ∂u ∂v 1
⇒ +i = +i
∂x ∂x ∂z ∂y ∂y ∂z
∂x ∂y
   
∂u ∂v 1 ∂u ∂v 1
⇒ +i = +i
∂x ∂x 1 ∂y ∂y i
∂u ∂v ∂v ∂u
⇒ +i = −i
∂x ∂x ∂y ∂y
∂u ∂v ∂u ∂v
⇒ = , and =− ,
∂x ∂y ∂y ∂x
 5.2. CAUCHY’S THEOREM 71

which are continuous inside and on C. Thus Green’s theorem can be applied.

Therefore

I I
f (z) dz = (u + iv)(dx + idy)
c C
I I
= (udx − vdy) + i (vdx + udy)
C C
ZZ   ZZ  
∂v ∂u ∂u ∂v
= − − dxdy + i − dxdy,
∂x ∂y ∂x ∂y
Γ Γ

by Green’s theorem
ZZ   ZZ  
∂v ∂v ∂v ∂v
= − + dxdy + i − dxdy
∂x ∂x ∂y ∂y
Γ Γ

= 0 + 0i = 0.

Hence the proof is complete.

Mathematician Goursat proved the above theorem removing the restriction

of existence continuous f ′ (z) inside Γ and on C. So this theorem is known as

Cauchy-Goursat theorem or Cauchy’s integral theorem. Briefly it is

referred as Cauchy’s theorem also.

Theorem 5.2.2 (Cauchy-Goursat theorem) If f (z) is analytic in a region Γ

and on its boundary C, then

I
f (z) dz = 0.
C

Proof. See [3, 4].

 5.2. CAUCHY’S THEOREM 72

Note 5.2 According to [2, Part 12], in case of Cauchy-Goursat theorem the

region Γ should be simply connected. But according to [4, Cauchy’s theo-

rem, Page 4.4], Cauchy’s theorem is valid, if the region Γ not only be simply-

connected but also be multiply-connected .

Note 5.3 From now, unless otherwise specified explicitly, by Cauchy’s theo-

rem we shall mean the Theorem 5.2.2.

Note 5.4 To apply Cauchy’s Theorem, the following two conditions must be

obeyed by the integrand f (z) :

(i) C is a closed curve and

(ii) f (z) is analytic inside and on C, that is, there is no singular point of

f (z) inside and on C.

H
Problem 5.10 Evaluate C
(z 5 − 2 + 5i) dz, where C is the circle |z| = 1.

Solution Clearly C : |z| = 1 is a closed contour. Here f (z) = z 5 − 2 + 5i is

a polynomial and so everywhere analytic. Thus f (z) is analytic inside and on

H
C. Thus by Cauchy’s theorem C
(z 5 − 2 + 5i) dz = 0.

Problem 5.11 Solve the Problem 4.24 of [4, Page 4.21] by applying Cauchy’s

theorem.

Solution Take help from [4, Problem 4.21].

Answer 0.
 5.2. CAUCHY’S THEOREM 73

Theorem 5.2.3 [2, Part 12, 15 min 44 sec] If C is closed contour and z0 is
H dz
lying outside C, then C
= 0.
(z − z0 )n

Proof. Hints (may be) : see [4, Page 4.20 & 4.21, Problem 4.21 & 4.22]

H z 2023 + z 2
Problem 5.12 Evaluate C
dz, where C is the circle |z − 1| = 1.
z − 10
z2
Solution Let f (z) = .
z − 10
Here equation of C is

|z − 1| = 1 ⇒ |x + iy − 1| = 1

⇒ |(x − 1) + i(y − 0)| = 1

⇒ (x − 1)2 + (y − 0)2 = 12 ,

which represents a circle with center (1, 0) and radius 1 unit.

Now, the integrand f (z) has singularities, when

z − 10 = 0 ⇒ z = 10.

Clearly, the singular point (10, 0) is lying outside the closed contour C. Thus

f (z) is analytic inside and on C. Therefore

z 2023 + z 2
I I
f (z) dz = dz = 0.
C C z − 10

Answer 0.

H 4z 2 + z + 4  x 2
Exercise 5.1 If f (a) = C dz, where C is the ellipse +
z−a 2
 y 2
= 1, then evaluate f (4.5).
3

Solution See [4, Problem 4.23, Page 4.21].

 5.2. CAUCHY’S THEOREM 74

H z+3
Problem 5.13 Evaluate C
dz, where C is the circle |z − 1| = 1.
z2 − 2z + 5
z+3
Solution Let f (z) = .
z2 − 2z + 5
Here equation of C is

|z − 1| = 1 ⇒ |x + iy − 1| = 1

⇒ |(x − 1) + i(y − 0)| = 1

⇒ (x − 1)2 + (y − 0)2 = 12 ,

which represents a circle with center (1, 0) and radius 1 unit.

Now, the integrand f (z) has singularities, when

p
−(−2) ± (−2)2 − 4 · 1 · 5
z 2 − 2z + 5 = 0 ⇒ z =
2·1

⇒ z = 1 ± 2i.

Clearly, the singular points (1, 2) and (1, −2) of f (z) are lying outside the

closed contour C. Thus f (z) is analytic inside and on C. Therefore

I I
z+3
f (z) dz = dz = 0.
C C z2 − 2z + 5

Answer 0.

H z−4
Problem 5.14 Find C
dz, where C is the circle |z| = 1.
z 2 + 2z + 2

Solution Point of singularities −1 ± i, which are lying outside the closed

z−4
contour C. Thus is analytic inside and on C. Therefore
z 2 + 2z + 2
z−4
I
2
dz = 0.
C z + 2z + 2

H z+4
Problem 5.15 Find C
dz, where C is the circle |z + 1| = 1.
z2 + 2z + 5
 5.2. CAUCHY’S THEOREM 75

Solution 0

H cos z
Problem 5.16 Find C
dz, where C is the circle |z| = 1.
(z 2 − 36)(z 2 + 16)

Solution Point of singularities z0 = ±6, ±4i, which are lying outside the
cos z
closed contour C. Thus is analytic inside and on C. There-
(z 2 − 36)(z 2 + 16)
H cos z
fore C
dz = 0.
(z 2 − 36)(z 2 + 16)
H 1
Problem 5.17 Evaluate C
tan z dz, where C is the circle |z| = .
2

Solution Let f (z) = tan z.

Here equation of C is

 2
1 p 1 1
|z| = ⇒ x2 + y 2 = ⇒ x2 + y 2 = ,
2 2 2

1
which represents a circle with center (0, 0) and radius unit. Clearly C is a
2
closed contour.
sin z
The integrand f (z) = has singularities, when
cos z

π
cos z = 0 ⇒ z = ± ≈ ±1.5708.
2

Observe that here is no singular point inside and on C. Thus f (z) is analytic

inside and on C. Therefore

I I
f (z) dz = tan z dz = 0.
C C

Answer 0

H z2 − z + 1 1
Problem 5.18 Obtain the value of C
dz, where C is |z| = .
z−1 2

Solution 0
 5.2. CAUCHY’S THEOREM 76

H e−z 1
Problem 5.19 Obtain the value of C
dz, where C is |z| = .
z+1 2

Solution See [4, Page 5.7, Problem 5.6].

Answer 0.

More general form of Theorem 5.1.1 for any closed contour C is as follows.

Theorem 5.2.4 [2, Part 12, 34 min] If C is a closed contour and z0 is an

interior point of C, then




Z
1 2πi, for n = 1

dz =
C (z − z0 )n 

0,
 for n ̸= 1.

whenever n ∈ N.

Proof. See Problem 4.21 and Problem 4.21 in [4, Page 4.20–4.21].

H dz
Note 5.5 From Cauchy-Goursat theorem C
= 0, when z0 is ly-
z − z0
ing outside the closed contour C. Again, from Theorem 5.2.4 we see that
H dz
C
= 0 for any natural number n > 1. Thus there is only one case,
(z − z0 )n
H dz
when C becomes nonzero for n ∈ N.
(z − z0 )n
 
H 5 1
Problem 5.20 Evaluate C z + dz, when c is |z| = 1.
z−6

Solution Clearly C is a closed contour. We have

I   I I
1 1
z5 + dz = 5
z dz + dz.
C z−6 C C z−6
 5.2. CAUCHY’S THEOREM 77

Now, z 2 is analytic inside and on C. Thus by Cauchy-Goursat theorem

I
z 5 dz = 0.
C

1
Again, singularity of is z = 4, which is lying outside C. Thus
z−6
I
1
dz = 0.
C z−6

Therefore

I   I I
1 1
z5 + dz = 5
z dz + dz = 0 + 0 = 0.
C z−6 C C z−6

Answer 0.

5.2.3. Consequences of Cauchy’s theorem. Some important conse-

quences of Cauchy’s theorem are as follows.

Theorem 5.2.5 If f (z) is analytic in a simply connected region Γ and a, b

Rb
are any two points in Γ, then a
f (z) dz is independent of the path in Γ from

a to b.

Proof. See [4, Problem 4.17, Page 4.19].

Theorem 5.2.6 If f (z) is analytic in a simply connected region Γ and a, b

are any two points in Γ, then the following statements hold :

Rz
(i) F (z) = a
f (ξ) dξ is analytic in Γ.

(ii) F ′ (z) = f (z).

Proof. See [4, Problem 4.18, Page 4.19].

 5.2. CAUCHY’S THEOREM 78

Definition 5.2.6 (Indefinite integral) [4, Page 4.20] A function F is called

f (z) dz, if F ′ (z) = f (z).

R
an indefinite integral of f and is denoted by

Theorem 5.2.7 If f (z) is analytic in a simply connected region Γ with

F ′ (z) = f (z) and a, b are any two points in Γ, then

Z b Z b
f (z) dz = F (b) − F (a), or F ′ (z) = F (z)|ba , or [F (z)]ba = F (b) − F (a).
a a

Theorem 5.2.8 If f (z) is analytic in a region bounded by two simple closed

curves C & C1 when C1 is lying inside C and f (z) is analytic on C & C1 also,

then
Z Z
f (z) dz = f (z) dz,
C C1

provided that C & C1 are traversed in the positive sense with respect to their

interiors.

Proof. See [4, Problem 4.20, Page 4.20].

Note 5.6 Above theorem reflects the fact that if we wish to integrate f (z)

along C, then C can be replaced by any other curve C1 provided that f (z) is

analytic in the region between C and C1 .

The following theorem is the generalized version of the Theorem 5.2.8.

Theorem 5.2.9 If f (z) is analytic in a region bounded by the non-overlapping

simple closed curves C, C1 , C2 , C3 , · · · , Cn when C1 , C2 , C3 , · · · , Cn are lying

inside C and f (z) is also analytic on all of these curves, then

Z Z Z Z Z
f (z) dz = f (z) dz + f (z) dz + f (z) dz + · · · + f (z) dz,
C C1 C2 C3 Cn
 5.3. CAUCHY’S INTEGRAL FORMULA 79

provided that all these curves are traversed in the positive sense with respect

to their interiors.

5.2.4. Converse of the Cauchy’s theorem. The converse of the The-

orem 5.2.2, due to Morera, stated below is known as Morera’s theorem.

Theorem 5.2.10 (Morera’s Theorem) If f (z) is continuous in a simply-

connected region Γ and

I
f (z) dz = 0
C

around every closed curve C in Γ, then f (z) is analytic in Γ.

5.3. Cauchy’s Integral Formula

The next theorem is known as Cauchy’s Integral Formula.

Note 5.7 [2, Part 13, 4 min 44 sec] Bounded rectifiable curve and closed

curves are same thing.

Theorem 5.3.1 (Cauchy’s Integral Formula) If f (z) be analytic inside and

on a simple closed curve C and z = z0 is lying inside C, then

I
1 f (z)
f (z0 ) = dz,
2πi c z − z0

where C is traversed in the positive (counterclockwise) sense.

Proof. Let C be a close curve and C1 be a circle inside C centered at z0

f (z)
with radius r. So is analytic inside and on C except at z = z0 (see
z − z0
Figure 5.1).
 5.3. CAUCHY’S INTEGRAL FORMULA 80

r
C1
z0

Figure 5.1.

f (z)
Since is analytic in the region bounded by the simple closed curves
z − z0
C and C1 , by Cauchy’s theorem for multiple regions

I I
f (z) f (z)
dz = dz.
C z − z0 C1 z − z0

Equation of C1 is

|z − z0 | = r ⇒ z − z0 = reiθ , 0 ≤ θ ≤ 2π ⇒ z = z0 + reiθ , 0 ≤ θ ≤ 2π.

So dz = iriθ dθ. Therefore

I I
f (z) f (z)
dz = dz
C z − z0 C1 z − z0
2π
f (z0 + reiθ ) iθ
Z
= ir dθ
0 reiθ
Z 2π
=i f (z0 + reiθ ) dθ.
0
 5.3. CAUCHY’S INTEGRAL FORMULA 81

When r → 0, then C1 → 0. Now taking limit r → 0 on both sides and making

use of the continuity of f (z),

I Z 2π
f (z)
lim dz = lim i f (z0 + reiθ ) dθ
r→0 C z − z0 r→0 0
I Z 2π
f (z)
⇒ dz = i lim f (z0 + reiθ ) dθ
C z − z0 0 r→0
Z 2π 

f lim z0 + reiθ dθ

=i
0 r→0
Z 2π Z 2π
=i f (z0 ) dθ = if (z0 ) dθ
0 0

= if (z0 ) [θ]2π
0 = if (z0 ) [2π − 0] = 2πif (z0 )
I
1 f (z)
∴ f (z0 ) = dz.
2πi c z − z0

Hence the proof is complete.

Note 5.8 Since Cauchy’s Integral Formula 5.3.1 can be written as

I
f (z)
dz = 2πif (z0 ),
c z − z0
H f (z)
it is employed to evaluate the value of c
dz.
z − z0
H cos z
Problem 5.21 Evaluate C
dz, when C is the ellipse 9x2 + 81y 2 = 1.
z

Solution The integrand has the singularities when

z = 0.

Given equation of C is

x2 y2
9x2 + 81y 2 = 1 ⇒  2 +  2 = 1,
1 1
3 9
 5.3. CAUCHY’S INTEGRAL FORMULA 82

which is an ellipse with center (0, 0).

Clearly, the singular point (0, 0) is lying inside the curve C.

Here f (z) = cos z and the integrand can be written as

I
f (z)
dz,
C z − z0

where z0 = 0.

Now, f (z0 ) = f (0) = cos 0 = 1. Therefore according to the Cauchy’s

integral formula,

I I
cos z f (z)
dz = dz = 2πi · f (z0 ) = (2πi) · (1) = 2πi.
C z C z−0

Answer 2πi.

H ez
Problem 5.22 Evaluate C
dz, when C is |z| = 5.
z − πi

Solution The integrand has the singularity when

z − πi = 0 ⇒ z = πi.

Given equation of C is

|z| = 5 ⇒ x2 + y 2 = 52 ,

which represents a circle with center (0, 0) and radius 5 unit.

 5.3. CAUCHY’S INTEGRAL FORMULA 83

Clearly, the point of singularity z = πi lies inside the closed curve C.

Agagin, here f (z) = ez . Thus by Cauchy’s integral formula

ez
I
dz = 2πif (πi) = 2πieπi
C z − πi

= 2πi(cos π + i sin π)

= 2πi(−1 + i.0) = −2πi.

Answer −2πi.

H 33z
Problem 5.23 Evaluate C
dz, when C is the ellipse |z−2|+|z+2| = 6.
z−i

Solution The integrand has the singularities when

z − i = 0 ⇒ z = i.

Given the equation of C is

|z − 2| + |z + 2| = 6 ⇒ |(x − 2) + iy| + |(x + 2) + iy| = 6

p p
⇒ (x − 2)2 + y 2 + (x + 2)2 + y 2 = 6

x2 y2
⇒ + √
2 = 1,
(3)2 5

which is an ellipse with center (0, 0).

Clearly, the singular point z0 = i = (0, 1) is lying inside the closed curve C.

Here f (z) = e3z . So the integrand can be written as

I
f (z)
dz,
C z − z0

where z0 = i.
 5.3. CAUCHY’S INTEGRAL FORMULA 84

Now, f (z0 ) = f (i) = e3i . Therefore according to the Cauchy’s integral

formula,

e3z
I I
f (z)
dz = dz = 2πi · f (z0 ) = (2πi) · (e3i ) = 2πie3i .
C z−i C z − z0

Answer 2πie3i . In polar form the answer is 2πi(cos3 + i sin 3).

H 3 + ez
Problem 5.24 Evaluate C
dz, when C is the circle |z| = 6.
z−i
H z 2 − 3z + 4i
Problem 5.25 Evaluate C dz, when C is the circle |z| = 3.
z + 2i

Solution Here f (z) = z 2 −3z +4i. z = −2i lying inside C. Cauchy’s integral

formula

z 2 − 3z + 4i
I
dz = 2πif (−2i) = 2πi(−4 + 6i + 4i)
C z + 2i

= −8πi − 12π − 8π

= −8πi − 20π.

Answer −8πi − 20π.

5.3.1. Technique to solve a problem having the integrand that

includes more than one singular point. To solve a problem with an inte-

grand including n number of singularities the following technique is employed.

(i) Using partial fraction express the given integrand as the sum of n

integrands each including one singularity.

(ii) Then depending on either corresponding singular point is lying inside

or outside of the closed curve, integrate each summand employing

either Cauchy’s theorem or Cauchy’s integral formula.

 5.3. CAUCHY’S INTEGRAL FORMULA 85

(iii) Finally add all the values obtained from the n integrations in Step-(ii).

5.3.1.1. All singular points lying inside the closed curve. After splitting the

given integrand partially employ only Cauchy’s integral formula.

H e2z
Problem 5.26 Evaluate C
dz where C is the circle |z| = 3.
(z − 1)(z − 2)

Solution The integrand has the singularities when

(z − 1)(z − 2) = 0 ⇒ z = 1, 2.

Now,

e2z
I I  
2z 1 1
dz = e + dz
C (z − 1)(z − 2) C (z − 1)(1 − 2) (2 − 1)(z − 2)
I  
2z 1 1
= e − dz
C z−2 z−1
e2z e2z
I I
= dz − dz.
C z −2 C z −1

Here the equation of C is

|z| = 3 ⇒ x2 + y 2 = 32 ,

which is a circle with center (0, 0) and radius 3 unit.

Clearly, the both singular points z0 = 2 = (2, 0) and z0 = 1 = (1, 0) are

lying inside the curve C. So by Cauchy’s integral formula

e2z e2z
I I
dz − dz = 2πi · e2(2) − 2πi · e2(1)
C z−2 C z−1

= 2πi(e4 − e2 ).

Answer 2πi(e4 − e2 ).
 5.3. CAUCHY’S INTEGRAL FORMULA 86

H sin πz 2 + cos πz 2
Problem 5.27 Evaluate C
dz where C is the circle |z| = 3.
(z − 1)(z − 2)

Solution See [4, Page 5.6, Problem 5.5(a)] or see [3, Page 232, Example 13].

Answer 4πi.

H eiz
Problem 5.28 Show that C
dz = 2πi sin t where C is the circle |z| =
z2 + 1
3 and t > 0.

Solution Hints: z 2 + 1 = (z + i)(z − i). See [3, Page 225, Example 8].

5.3.1.2. All singular points lying outside the closed curve. After splitting

the given integrand partially employ only Cauchy’s theorem.

H z+2
Problem 5.29 Evaluate C
dz where C is the circle |z| = 1.
(z − 3)(z − 4)

Solution The integrand has the singularities when

(z − 3)(z − 4) = 0 ⇒ z = 3, 4.

Now,
I I  
z+2 1 1
dz = (z + 2) + dz
C (z − 3)(z − 4) C (z − 3)(3 − 4) (4 − 3)(z − 4)
I  
1 1
= (z + 2) − dz
C z−4 z−3
I I
z+2 z+2
= dz − dz.
C z −4 C z −3

Here the equation of C is

|z| = 1 ⇒ x2 + y 2 = 12 ,

which is a circle with center (0, 0) and radius 1 unit.

 5.3. CAUCHY’S INTEGRAL FORMULA 87

Clearly, the both singular points z0 = 4 = (4, 0) and z0 = 3 = (3, 0) are

lying outside the curve C. Therefore by Cauchy’s theorem

I I I
z+2 z+2 z+2
dz = dz − dz = 0 − 0 = 0.
C (z − 3)(z − 4) C z−4 C z−3

Answer 0.

Problem 5.30 Show that



−2πi, if C is the circle |z − 1| = 4

e3z
I 
dz = .
C z − πi 

0,
 if C is the ellipse |z − 2| + |z + 2| = 6

Solution Hints: |z − 2| + |z + 2| = 6 represents an ellipse having foci at

z = 2, z = −2 and 6 as its length of major axis.

See [3, Page 224, Example 7].

5.3.1.3. Some singular points lying inside and some singular points lying

outside the closed curve. After splitting the given integrand partially employ

only Cauchy’s integral formula and Cauchy’s theorem according to the demand

of situation.

H z+4
Problem 5.31 Evaluate C
dz where C is the circle |z − 2| = 1.
z2 − 4

Solution Equation of the closed curve C is

|z − 2| = 1 ⇒ (x − 2)2 + y 2 = 12 ,

which is a circle with center (2, 0) and radius 1 unit.

 5.3. CAUCHY’S INTEGRAL FORMULA 88

Now,

I I
z+4 z+4
dz = dz
C z2 − 4 C (z + 2)(z − 2)
I  
1 1
= (z + 4) + dz
C (z + 2)(−2 − 2) (2 + 2)(z − 2)
1 1
 
I
= (z + 4)  4 − 4  dz
 
C z−2 z+2

I (z + 4) I (z + 4)
= 4 dz − 4 dz.
C z − 2 C z + 2

The integrand has the singularity when

z 2 − 4 = 0 ⇒ z = ± 2.

Clearly, the singular point z0 = 2 = (2, 0) is lying inside C and the singular

point z0 = −2 = (−2, 0) is lying outside the closed curve C. So applying the

Cauchy’s integral formula to the first integral and the Cauchy’s theorem to the

second integral

I I z+4 I z+4
z+4 4 dz − 4 dz
dz =
C z2 − 4 C z −2 C z +2

2+4 3
= 2πi · + 0 = 2πi · = 3πi.
4 2

Answer 3πi.

H z+6
Exercise 5.2 Evaluate C
dz where C is the circle |z − 2| = 1.
z2 − 4

Solution 4πi

H 3z 2 + z
Problem 5.32 Evaluate C
dz where C is the circle |z − 1| = 1.
z2 − 1
 5.3. CAUCHY’S INTEGRAL FORMULA 89

Solution See [4, Page 5.7, Problem 5.7].

Answer 4πi.

H z
Problem 5.33 Evaluate C
dz where C is the circle |z| = 2.
(9 − z 2 )(z + i)

Solution See [3, Page 230, Example 12].

π
Answer .
5

5.3.2. Cauchy’s Integral Formula for 1st order derivative. Cauchy’s

Integral Formula to get the 1st order derivative of an analytic function is as

follows.

Theorem 5.3.2 (Cauchy’s Integral Formula for 1st derivative) If f (z) be

analytic inside and on a simple closed curve C and z = z0 is any point inside

C, then
I
′ 1 f (z)
f (z0 ) = dz,
2πi c (z − z0 )2

where C is traversed in the positive (counterclockwise) sense.

Proof. See [4, Page 5.4, Problem 5.2].

5.3.3. Cauchy’s Integral Formula for n-th derivative. Theorem 5.3.1

can be extended for the n-th derivative of f (z) at z = z0 as follows.

Theorem 5.3.3 (Cauchy’s Integral Formula for n-th derivative) If f (z) be

analytic inside and on a simple closed curve C and z = z0 is any point inside

C, then
I
(n) n! f (z)
f (z0 ) = dz, n = 1, 2, 3, · · · .
2πi C (z − z0 )(n+1)
 5.3. CAUCHY’S INTEGRAL FORMULA 90

Proof. See [4, Page 5.5, Problem 5.3].

Note 5.9 Since Cauchy’s integral formula for n-th derivative 5.3.3 can be

written as

I
f (z) 2πi (n)
(n+1)
dz = f (z0 ), for n = 1, 2, 3, · · · ,
C (z − z0 ) n!

H f (z)
it is employed to evaluate the value of C
dz.
(z − z0 )(n+1)

H sin6 z
Problem 5.34 Evaluate dz where C is |z| = 1.
C  π 3
z−
6
Solution The equation of C is

|z| = 1 ⇒ x2 + y 2 = 12 ,

which is a circle with center (0, 0) and radius 1 unit.

The integrand has singularities when

 π 3 π π π
z− =0 ⇒ z= , , .
6 6 6 6

π
Thus singular point is z0 = = 0.5, which is lying inside C.
6
Let f (z) = sin6 z. Hence

sin6 z
I I
f (z)
dz = dz
C
 π 3 C (z − z0 )(2+1)
z−
6
2πi ′′
= f (z0 ),
2!

by Cauchy’s Integral formula for 2nd derivative

π 
=πi · f ′′
6
 5.3. CAUCHY’S INTEGRAL FORMULA 91

Here

f (z) = sin6 z ⇒ f ′ (z) = 6 sin5 z cos z

⇒ f ′′ (z) = −6 sin5 z sin z + 30 cos2 z sin4 z

Thus

π   5   √ !2  4
1 1 3 1 21
f ′′ = −6 + 30 = .
6 2 2 2 2 16

Therefore

sin6 z
I
21 21
dz = πi · = πi.
C
 π 3 16 16
z−
6
21
Answer πi.
16
H eiz
Problem 5.35 Evaluate C
dz where C is the curve given in [2,
(z 2 + 1)2
Part 514, 25 min].

Solution See [2, Part 14, 25 min].

π
Answer .
e
H e2z
Problem 5.36 Evaluate C
dz where C is the circle |z| = 3.
(z + 1)4

Solution See [4, Page 5.6, Problem 5.5(b)].

2
Answer 8πie− 3 .

H cos 2z
Problem 5.37 See [2, Part 14, 14 min]. Evaluate C
dz where C is
z5
the curve given as |z| = 1.
 5.3. CAUCHY’S INTEGRAL FORMULA 92

Solution Here f (z) = cos 2z. The integrand has the point of singularity

when

z5 = 0 ⇒ z = 0.

Clearly, the point of singularity is lying inside the closed contour C. Again, in
H cos 2z
this problem n = 5, as the integral can be written as C
dz. Hence by
(z − 0)5
Cauchy’s integral formula for higher order derivative,

I I
f (z) f (z)
dz = dz
C (z)5 C (z − 0)(4+1)
2πi ′v
= f (0),
4!
πi ′v
= · f (0).
12

Here

f (z) = cos 2z ⇒ f ′ (z) = −2 sin 2z

⇒ f ′′ (z) = −4 cos 2z

⇒ f ′′′ (z) = 8 sin 2z

⇒ f ′v (z) = 16 cos 2z.

Thus f ′v (0) = 16. Therefore

I
f (z) πi 4πi
5
dz = · 16 = .
C z 12 3
4πi
Answer .
3
H z+2
Problem 5.38 See [2, Part 14, 17 min 51 sec]. Evaluate C
dz
z 2 (z− 1 − i)
where C is the curve given as |z| = 1.
 5.3. CAUCHY’S INTEGRAL FORMULA 93

z+2
Solution Here f (z) = . The integrand has the point of singularity
z−1−i
when

z2 = 0 ⇒ z = 0.

Clearly, the point of singularity z = 0 is lying inside the closed contour C.

z+2
H z−1−i
Again, in this problem n = 2, as the integral can be written as C dz.
(z − 0)2
Hence by Cauchy’s integral formula for higher order derivative,

I I
f (z) f (z)
dz = dz
C z2 C (z − 0)(1+1)
2πi ′
= f (0),
1!

=2πi · f ′ (0).

Here

z+2 (z − 1 − i) · 1 − (z + 2) · 1
f (z) = ⇒ f ′ (z) = .
z−1−i (z − 1 − i)2

Thus

(−1 − i) − 2 −3 − i i(−3 − i) −1 + 3i
f ′ (0) = = = = .
(−1 − i)2 2i 2i · i 2

Therefore

−1 + 3i
I
f (z)
2
dz = 2πi · = πi · (−1 + 3i) = π(−3 − i).
C z 2

Answer π(−3 − i).

Caution 5.1 In Problem 5.38 z − 1 − i = 0 is not considered to get singular

point. Because if we consider this, then we get z = 1 + i as the singular point,

which does not lie inside C : |z| = 1. So then Cauchy’s integral formula for
 5.4. WINDING NUMBER 94

z+2
higher order derivative can not be applied. Thus f (z) = has been
z−1−i
considered.

Theorem 5.3.4 If f (z) be analytic in a region Γ bounded by two concentric

circles C1 and C2 and on the boundary, then

I I
1 f (z) 1 f (z)
f (z0 ) = dz − dz
2πi C1 z − z0 2πi C2 z − z0

for any point z0 in Γ.

Proof. See [4, Problem 5.32 of Page 5.19].

5.3.4. Importance of Cauchy’s integral formula. It says that if a

function f (z) is known on the simple closed curve c, then the values of the

function and all its derivatives can be found at all points inside c. Thus, if

a function of complex variable is analytic (first derivative exists) in a simple

connected region R, then all its higher derivatives exists in R. This is not

necessarily true for function of real variable.

5.4. Winding Number

5.4.1. Winding number. Winding number of a closed curve is defined

with respect to a point lying on the complex plane.

Definition 5.4.1 (Winding number) [2, Part 13, 28 min] Suppose that C is

a closed contour in C-plane and z0 is a given point in C \ C, that is, z0 does

not lie on C. Then the winding number of C about the point z0 is defined

to be the number of times C wraps around z0 in the positive direction and it

is denoted by n(C, z0 ).
 5.4. WINDING NUMBER 95

Note 5.10 Winding number is also known as index.

Note 5.11 Winding number is always an integer. It can be negative, zero

and positive. The number n(C, z0 ) becomes

(i) zero, when z0 lies outside the closed contour C.

(ii) positive, when C winds in the positive sense around z0 .

(iii) negative, when C winds in the negative sense around z0 .

Note 5.12 If z0 is winded by positive oriented simple closed contour C, then

n(C, z0 ) = 1.

5.4.2. Properties of winding number. [3, Page 216] Winding number

has following properties :

(i) Let C be a closed contour and C = C1 ∪ C2 , where C1 and C2 are

two closed contours in the complex plane having same initial points.

Then

n(C, z0 ) = n(C1 , z0 ) + n(C2 , z0 )

and

n(−C, z0 ) = −n(C, z0 )

for all z0 ∈
/ C.

(ii) If C1 and C2 are two homotopic closed curves in C \ {z0 }, then

n(C1 , z0 ) = n(C2 , z0 ) for all z0 ∈ C1 ∪ C2 .

5.4.3. Cauchy’s integral formula in terms of winding number.

Cauchy’s integral formula is stated in terms of winding number as follows.

 5.4. WINDING NUMBER 96

Theorem 5.4.1 [4, Page 13] Let f (z) be analytic inside and on a closed

contour C and z0 be an interior point of C. Then

I
f (z)
dz = 2πif (z0 ) · n(C, z0 ),
C z − z0

where n(C, z0 ) is the winding number of C around the point z0 .

Note 5.13 If it is not specified explicitly, always positive direction of traver-

sal should be considered.

Problem 5.39 Evaluate the following integrals

H ez H ez
(a) dz (b) dz,
C 1 C 7
z− z−
2 2

when closed curve C is given as C = 2 + 4 cos θ for 0 ≤ θ ≤ 2π.

Solution (a) Here f (z) = ez . The integrand has point of singularity when

1 1
z− =0 ⇒ z= .
2 2

From Figure [2, Part 13, 39 min] n(C, 1) = 2. By Cauchy’s integral formula

for winding number,

ez
I    
1 1
dz = 2πi · f · n C,
C
1 2 2
z−
2
1 1
= 2πi · e 2 · 2 = 4πie 2 .

1
Answer 4πie 2 .
 5.5. CAUCHY’S INEQUALITY 97

(b) Here f (z) = ez . The integrand has point of singularity when

1 1
z−
=0 ⇒ z= .
2 2
 
7
From Figure [2, Part 13, 39 min] n C, = 1. By Cauchy’s integral formula
2
for winding number,

ez
I    
7 7
dz = 2πi · f · n C,
C
7 2 2
z−
2
7 7
= 2πi · e 2 · 1 = 2πie 2 .

7
Answer 2πie 2 .

5.5. Cauchy’s Inequality

Theorem 5.5.1 (Cauchy’s inequality) [2, Part 15] If f (z) is analytic inside

and on a positively oriented circle C described by |z − z0 | = r and |f (z)| ≤ M

n!M
for all z ∈ C, then |f n (z0 )| ≤ when n ∈ N.
rn

Proof. See [2, Part 15].

 CHAPTER 6

Power Series

For solving more problems the reader is referred to [4], [3] and [2].

6.1. Power Series

Definition 6.1.1 (Power series) An infinite series having the form

∞
X
an (z − z0 )n = a0 + a1 (z − z0 ) + a2 (z − z0 )2 + a3 (z − z0 )3 + · · · .
n=0

is called power series. The coefficients an s are called complex constants

and the point z0 is called the center of the power series.

When z0 = 0, then the power series takes the form

∞
X
an z n = a0 + a1 z + a2 z 2 + az 3 + · · · .
n=0

P∞
Definition 6.1.2 (Absolute convergence) The power series n=0 an z n is
P∞
called absolutely convergent, if the series n=0 |an ||z|n converges.

P∞
Definition 6.1.3 (Conditionally convergent) The power series n=0 an z n is
P∞
called conditionally convergent, if the series n=0 an z n itself converges but
P∞
n=0 |an ||z|n diverges.

Definition 6.1.4 (Radius of convergence) A non-negative number R is called

P∞
radius of convergence of the power series n=0 an (z − z0 )n , if the power

series converges for |z − z0 | < R and diverges for |z − z0 | > R.

98
 6.1. POWER SERIES 99

Definition 6.1.5 (Circle of convergence) When R is the radius of conver-

P∞
gence of a power series n=0 an (z − z0 )n , then the circle |z − z0 | = R is called

the circle of convergence.

P∞
Note 6.1 The circle of convergence of a power series n=0 an (z − z0 )n is the

largest circle centered at z = z0 such that the series is convergent at all points

inside the circle.

1
1 an+1
Formula 6.1.1 = limn→∞ |an | n = limn→∞ , provided that the limit
R an
in the right hand side exists.

P∞ z n
Problem 6.1 Obtain the radius of convergence of the series n=1 n .
n
1
Solution Here an = . Now,
nn

1
1
= lim |an | n
R n→∞
1
n
1
= lim n
n→∞ n

1
 n
1
= lim , since n is positive
n→∞ nn
1
= lim = 0.
n→∞ n

Therefore R is infinite, that is, the given series converges everywhere in the

complex plane.

P 2−n z n
Problem 6.2 Obtain the radius of convergence of the series .
1 + in2
 6.1. POWER SERIES 100

2−n
Solution Here an = . Now,
1 + in2

1 an+1
= lim
R n→∞ an
2−(n+1)
1 + i(n + 1)2
= lim
n→∞ 2−n
1 + in2
2−(n+1) 1 + in2
= lim ×
n→∞ 1 + i(n + 1)2 2−n
2−n−1
= lim × 2n (1 + in2 )
n→∞ 1 + i(n2 + 2n + 1)

2−n−1+n
 
2 1
= lim   ×n i+ 2
n→∞
2
2 1 n
1 + in 1 + + 2
n n
 
−1 2 1
2 n i+ 2
n
= lim   
n→∞
2
1 2 1
n +i 1+ + 2
n2 n n
 
1
2−1 i + 2
n
= lim  
n→∞ 1 2 1
+i 1+ + 2
n2 n n
2−1 (i + 0)
=
0 + i (1 + 0 + 0)
2−1 i 1
= = .
i 2

Therefore R = 2.

P (n!)k n
Problem 6.3 Obtain the radius of convergence of the series z .
(nk)!
 6.1. POWER SERIES 101

(n!)k
Solution Here an = . Now,
(nk)!

1 an+1
= lim
R n→∞ an
[(n + 1)!]k
[(n + 1)k]!
= lim
n→∞ (n!)k
(nk)!
[(n + 1)!]k (nk)!
= lim ×
n→∞ [(n + 1)k]! (n!)k

(n + 1)k (n!)k (nk)!

= lim
n→∞ (kn + k)! (n!)k
(n + 1)k (nk)!
= lim
n→∞ (kn + k)!
(n + 1)k (nk)!
= lim
n→∞ (kn + k) (kn + k − 1) (kn + k − 2) . · · · . (kn + 2) (kn + 1) (kn)!
(n + 1)k
= lim
n→∞ (kn + k) (kn + k − 1) (kn + k − 2) . · · · . (kn + 2) (kn + 1)
 k
k 1
n 1+
n
= lim       
n→∞
k
k k−1 k−2 2 1
n k+ k+ k+ . ··· . k + k+
n n n n n
 k
1
1+
n
= lim       
n→∞ k k−1 k−2 2 1
k+ k+ k+ . ··· . k + k+
n n n n n

(1 + 0)k
=
(k + 0) (k + 0) (k + 0) . · · · . (k + 0) (k + 0)
| {z }
k−terms

1
= .
kk

Therefore R = k k .
 6.1. POWER SERIES 102

P (−1)n (z − 2i)n
Problem 6.4 Obtain the radius of convergence of the series .
n
(−1)n
Solution Here an = . Now,
n

1 an+1
= lim
R n→∞ an
(−1)n+1
= lim n+1
n→∞ (−1)n
n
(−1)n+1 n
= lim ×
n→∞ n+1 (−1)n

(−1) n
= lim
n→∞ n+1
n
= lim |−1| × lim
n→∞ n→∞ n+1

n
= lim (1) × lim  
n→∞ n→∞ 1
n 1+
n

1
= 1 × lim
n→∞ 1
1+
n

1
= lim
n→∞ 1
1+
n
 
1
= lim 
 
n→∞ 1
1+
n
1
=
1+0

= 1.

Therefore R = 1.
 6.1. POWER SERIES 103

P∞ n(n − 1)
Problem 6.5 Obtain the radius of convergence of the series 1 n
(z)n−2 .
2
n(n − 1)
Solution Here an = . Now,
2n

1 an+1
= lim
R n→∞ an
(n + 1){(n + 1) − 1}
= lim 2n+1
n→∞ n(n − 1)
2n
(n + 1)n 2n
= lim ×
n→∞ 2n+1 n(n − 1)
(n + 1)n 2n
= lim ×
n→∞ 2n × 21 n(n − 1)
n+1
= lim
n→∞ 2(n − 1)
 
1 n+1
= lim × lim
n→∞ 2 n→∞ n−1
 
1
n 1+
1 n
= × lim  
2 n→∞ 1
n 1−
n
1
1 1+
= × lim n
2 n→∞ 1
1−
n
1 1+0
= ×
2 1−0
1 1
= ×
2 1
1
= .
2

Therefore R = 2.

P∞
Problem 6.6 Find the radius of convergence of the series 1 (n log n) z n .
 6.1. POWER SERIES 104

Solution Here an = n log n. Now,

1 an+1
= lim
R n→∞ an
(n + 1) log(n + 1)
= lim
n→∞ n log n
n + 1 log(n + 1)
= lim ×
n→∞ n log n
n+1 log(n + 1)
= lim × lim
n→∞ n n→∞ log n

Now

n+1 1
lim = lim 1 +
n→∞ n n→∞ n

=1+0=1

and

log(n + 1) ∞
lim , form
n→∞ log n ∞
1
= lim n + 1 , by L’Hospital rule
n→∞ 1
n
n
= lim
n→∞ n + 1

1
= lim
n→∞ n+1
n

1
= lim
n→∞ 1
1+
n
1
= = 1.
1+0

1
Thus = 1 × 1 = 1. Therefore R = 1.
R
 6.2. CONVERGENCE OF GEOMETRIC SERIES 105

6.2. Convergence of Geometric Series

Definition 6.2.1 (Geometric Series) An infinite series of having the form

∞
X
arn = a + ar + ar2 + ar3 + · · · + arn + · · ·
n=0

is called a power series. In this case, r is called the common ratio of the

geometric series.

P∞
Theorem 6.2.1 A geometric series n=0 arn converges if and only if |r| < 1

and diverges |r| ≥ 1.

Note 6.2 If we can write the n-th term of a series can be written as the

power of some term (like in Problem 6.7), then the series is geometric.

P∞
Problem 6.7 Find the domain of convergence of the series n=0 3−n (z−1)2n

and if the domain of convergence is the disk find the radius of convergence.

Solution Given series is

∞ ∞
X
−n 2n
X  −1 n
3 (z − 1) = 3 (z − 1)2 .
n=0 n=0

So the given series is geometric with a = 1 and common ration r = 3−1 (z −1)2 .

Thus it will be convergent, if

3−1 (z − 1)2 < 1

(z − 1)2
⇒ <1
3

⇒ (z − 1)2 < 3

√
⇒ |z − 1| < 3.
 6.2. CONVERGENCE OF GEOMETRIC SERIES 106

Therefore the given series converges within the open disk centered at (1, 0)
√
with radius 3 unit. Hence the domain of convergence is

√
{z ∈ C : |z − 1| < 3},

√
which is an open disc with radius of convergence 3 unit.
 n
P∞ 2i
Problem 6.8 Find the domain of convergence of the series n=0
z+i+1
and if the domain of convergence is the disk find the radius of convergence.

Solution Given series is

∞  n
X 2i
,
n=0
z+i+1

2i
which one is a geometric series with a = 1 and common ration r = .
z+i+1
Thus it will be convergent, if

2i
<1
z+i+1
|2i|
⇒ <1
|z + i + 1|
2 √
⇒ < 1, since |2i| = 02 + 22 = 2
|z + i + 1|
|z + i + 1|
⇒ >1
2

⇒ |z + i + 1| > 2

⇒ |z − (−1 − i)| > 2.

Therefore the given series converges outside the open disk centered at (−1, −1)

with radius 2 unit. Hence the domain of convergence is

{z ∈ C : |z − (−1 − i)| > 2},

 6.3. D’ALEMBERT’S RATIO TEST 107

which is not an open disc. So there is no radius of convergence.

P∞
Problem 6.9 Find the domain of convergence of the series n=0 2−n z 2n and

if the domain of convergence is the disk find the radius of convergence.

P∞ −n 2n
P∞ n
Solution Given series is n=0 2 z = n=0 (2−1 z 2 ) . So it is a geometric

series with a = 1 and common ration r = 2−1 z 2 . Thus it will be convergent, if

2−1 z 2 < 1

|z 2 |
⇒ <1
|2|

⇒ z2 < 2

⇒ |z|2 < 2

√
⇒ |z| < 2.

Therefore the given series converges inside the open disk centered at (0, 0) with
√
radius 2 unit. Hence the domain of convergence is

√
{z ∈ C : |z| < 2},

√
which is an open disc with radius of convergence 2 unit.

6.3. D’Alembert’s Ratio Test

P∞
Theorem 6.3.1 (D’Alembert’s ratio test) If n=0 an be a series of complex
an+1
number and limn→∞ = l, then the series is convergent if l < 1 and
an
divergent if l > 1.

P∞
The series n=0 an is neither convergent nor divergent if l = 1.
 6.3. D’ALEMBERT’S RATIO TEST 108

 n
P∞ 1.3.5. · · · .(2n − 1) 1−z
Problem 6.10 For the series n=1 find the
n! z
domain of convergence of and if the domain of convergence is the disk, then

find the radius of convergence.

Solution Let
 n
1.3.5. · · · .(2n − 1) 1−z
un = .
n! z

Then

un+1
un
 n+1
1.3.5. · · · .(2n − 1)(2n + 1) 1 − z
(n + 1)! z
=  n
1.3.5. · · · .(2n − 1) 1 − z
n! z
 n+1  n
1.3.5. · · · .(2n − 1)(2n + 1) 1 − z n! z
= ×
(n + 1)! z 1.3.5. · · · .(2n − 1) 1 − z
(2n + 1)(n!) 1 − z
= ×
(n + 1)! z
2n + 1 1 − z
= × .
n+1 z

Therefore

an+1 2n + 1 1 − z
lim = lim ×
n→∞ an n→∞ n + 1 z
1
2+
= lim n × lim 1 − z
n→∞ 1 n→∞ z
1+
n
2+0 1−z
= × lim
1+0 n→∞ z
1−z
=2 .
z
 6.3. D’ALEMBERT’S RATIO TEST 109

The given series will be convergent, if

1−z
2 <1
z
1−z 1
⇒ <
z 2
1 − x − iy 1
⇒ <
x + iy 2
1 − x − iy 1
⇒ <
x + iy 2
|1 − x − iy| 1
⇒ <
|x + iy| 2
p
(1 − x)2 + (−y)2 1
⇒ p <
2
x +y 2 2
1 − 2x + x2 + y 2 1
⇒ 2 2
<
x +y 4

⇒ 4 − 8x + 4x2 + 4y 2 < x2 + y 2

⇒ 3x2 + 3y 2 − 8x + 4 < 0

8 4
⇒ x2 + y 2 − x + < 0.
3 3

 
4
Thus the given series will be convergent inside the open disk with center ,0
3
2
and radius unit. Hence the domain of convergence is
3

 
4 2
z∈C: z− < ,
3 3

2
and the radius of convergence is unit.
3
P∞ 2
Problem 6.11 Find the domain of convergence of the series n=1 z n and

if the domain of convergence is the disk find the radius of convergence.

 6.4. TAYLOR’S SERIES 110

2
Solution Here un = z n . Then

2
un+1 z (n+1)
lim = lim
n→∞ un n→∞ z n2
2 +2n+1
zn
= lim
n→∞ z n2

= lim z 2n+1
n→∞

2n+1
= lim z = l (say).
n→∞

Now, when

(i) |z| = 1, then l = 1.

(ii) |z|1, then l → 0.

(ii) |z| > 1, then l → ∞.

Thus the given series will be convergent when |z| < 1 and divergent when

|z| > 1. That is, the given series will be convergent inside the open disk with

center (0, 0) and radius 1 unit. Hence the domain of convergence is

{z ∈ C : |z| < 1} ,

and the radius of convergence is 1 unit.

6.4. Taylor’s Series

Theorem 6.4.1 (Taylor’s theorem) [2, Part 20, 4 min] If f (z) is analytic

inside and on all points of a circle with center z0 and radius R, then

∞
X f n (z0 )
f (z) = an (z − z0 )n , whenever an =
n=0
n!

f ′ (z0 ) f ′′ (z0 ) f ′′′ (z0 )

= f (z0 ) + (z − z0 ) + (z − z0 )2 + (z − z0 )3 + · · · .
1! 2! 3!
 6.4. TAYLOR’S SERIES 111

for all z within C.

Proof. Let f (z) be analytic inside and on all points of a circle with center

z0 and radius R. Suppose that z be any point inside C such that |z − z0 | = r.

Construct a circle C1 with radius r1 that encloses z. Again, let w be a point

such that |w − z0 | = r1 < R.

C1 w
r1
z0
r
R z

Since f (z) is analytic inside and on the closed curve C1 , by Cauchy’s inte-

gral formula
I
1 f (w)
f (z) = dw.
2πi C1 w−z

Now

1 1
=
w−z (w − z0 ) − (z − z0 )
 −1
1 1 z − z0
=  = 1−
z − z0 w − z0 w − z0
(w − z0 ) 1 −
w − z0
"  2  3 #
1 z − z0 z − z0 z − z0
= 1+ + + + ···
w − z0 w − z0 w − z0 w − z0

1 z − z0 (z − z0 )2 (z − z0 )3
= + + + + ···
w − z0 (w − z0 )2 (w − z0 )3 (w − z0 )4
 6.4. TAYLOR’S SERIES 112

Thus

(z − z0 )2
 
z − z0
I
1 1
f (z) = f (w) + + + · · · dw
2πi C1 w − z0 (w − z0 )2 (w − z0 )3
f (w)(z − z0 )
I I
1 f (w) 1
= dw + dw
2πi C1 w − z0 2πi C1 (w − z0 )2
f (w)(z − z0 )2
I
1
+ dw + · · ·
2πi C1 (w − z0 )3
z − z0
I I
1 f (w) f (w)
= dw + dw
2πi C1 w − z0 2πi C1 (w − z0 )2
(z − z0 )2
I
f (w)
+ 3
dw + · · · .
2πi C1 (w − z0 )

By Cauchy’s integral formula for higher order derivative

I
n n! f (w)
f = for n = 0, 1, 2, 3, · · · ,
2πi C1 (w − z0 )(n+1)

we get

f ′ (a) f ′′ (a) f ′′′ (a)

f (z) =f (a) + (z − z0 ) + (z − z0 )2 + (z − z0 )3 + ··· .
1! 2! 3!

Hence the proof is complete.

Note 6.3 When z0 = 0, then the Taylor’s series of f (z) about z0 ,

∞
X f n (z0 )
f (z) = an (z − z0 )n , whenever an =
n=0
n!

f ′ (z0 ) f ′′ (z0 ) f ′′′ (z0 )

= f (z0 ) + (z − z0 ) + (z − z0 )2 + (z − z0 )3 + · · · .
1! 2! 3!

takes the form

∞
X f n (0)
f (z) = an z n , whenever an =
n=0
n!

f ′ (0) f ′′ (0) 2 f ′′′ (0) 3

= f (0) + z+ z + z + ··· ,
1! 2! 3!
 6.4. TAYLOR’S SERIES 113

which is often called a Maclaurin series.

Problem 6.12 Expand f (z) = sin z in Taylor’s series about z = 0.

Solution Here

f (z) = sin z ∴ f (0) = 0

⇒ f ′ (z) = cos z ∴ f ′ (0) = 1

⇒ f ′′ (z) = − sin z ∴ f ′′ (0) = 0

⇒ f ′′′ (z) = − cos z ∴ f ′′′ (0) = −1

⇒ f iv (z) = sin z ∴ f iv (0) = 0

⇒ f v (z) = cos z ∴ f v (0) = 1

..
.

Therefore by Taylor’s theorem about z = 0

f ′ (0) f ′′ (0) f ′′′ (0)

f (z) = sin z =f (0) + (z − 0) + (z − 0)2 + (z − 0)3
1! 2! 3!
f iv (0) f v (0)
+ (z − 0)4 + (z − 0)5 + · · ·
4! 5!
z2 z3 z4 z5
=0 + z · 1 + ·0+ · (−1) + ·0+ · 1 + ···
2! 3! 4! 5!
z3 z5
=z − + − ··· .
3! 5!
z3 z5
Answer sin z = z − + − ···.
3! 5!
π
Problem 6.13 Expand f (z) = sin z in Taylor’s series about z = .
4
 6.4. TAYLOR’S SERIES 114

Solution Here

π  1
f (z) = sin z ∴ f =√
4 2
π  1
⇒ f ′ (z) = cos z ∴ f ′ =√
4 2
π  1
⇒ f ′′ (z) = − sin z ∴ f ′′ = −√
4 2
π  1
⇒ f ′′′ (z) = − cos z ∴ f ′′′ = −√
4 2
π  1
⇒ f iv (z) = sin z ∴ f iv =√
4 2
π  1
⇒ f v (z) = cos z ∴ f v =√
4 2

and so on.
π
Therefore by Taylor’s theorem about z =
4
π  π 
′ ′′
π  f  π f 4  π 2
f (z) = sin z =f + 4 z− + z−
4 1! 4 2! 4
π  π  π 
′′′ iv v
f  π 3 f  π 4 f 4  π 5
+ 4 z− + 4 z− + z−
3! 4 4! 4 5! 4

+ ···

1 1 1
√  −√  −√ 
1 2 π 
2 π  2
2 π 3
=√ + z− + z− + z−
2 1! 4 2! 4 3! 4
1 1
√  √
2 π 4

2
 π 5
+ z− + z− + ···
4! 4 5! 4
π 2  π 3
  
1

  π  z − z − 

=√ 1+ z− − 4 − 4 + ··· .
2 4 2! 3! 


π 2  π 3
  
1

  π  z − z − 

Answer sin z = √ 1+ z− − 4 − 4 + ··· .
2
 4 2! 3! 

 6.5. LAURENT’S SERIES 115

Problem 6.14 Expand f (z) = ln(1 + z) in Taylor’s series about z = 0.

Solution See [4, Problem 6.23(a) of Page 6.19].

6.5. Laurent’s Series

Theorem 6.5.1 If f (z) be analytic inside and on an annular region Γ de-

scribed by r1 < |z − z0 | < r2 , bounded by two concentric circles C1 and C2

centered at z0 with radii r1 and r2 respectively, then for all z ∈ Γ

∞ ∞
X
n
X bn
f (z) = an (z − z0 ) + ,
n=0 n=1
(z − z0 )n

where
Z
1 f (z)
an = dz, n = 0, 1, 2, 3, · · ·
2πi C1 (z − z0 )n+1

and
Z
1 f (z)
bn = dz, n = 1, 2, 3, · · · .
2πi C2 (z − z0 )−n+1

C1
r2
z0
r1 z

Proof. According to the Cauchy’s integral formula 5.3.4

Z Z
1 f (z) 1 f (z)
f (a) = dz − dz.
2πi C1 z−a 2πi C2 z−a
 6.5. LAURENT’S SERIES 116

we have,

Z Z
1 f (w) 1 f (w)
f (z) = dw − dw
2πi C1 w−z 2πi C2 w−z
Z Z
1 f (w) 1 f (w)
= dw + dw
2πi C1 w−z 2πi C2 z − w
Z Z
1 f (w) 1 f (w)
= dw + dw
2πi C1 w − z0 + z0 − z 2πi C2 z − z0 + z0 − w
Z Z
1 f (w) 1 f (w)
= dw + dw
2πi C1 (w − z0 ) − (z − z0 ) 2πi C2 (z − z0 ) − (w − z0 )
Z Z
1 f (w) 1 f (w)
=   dw +   dw
2πi C1 z − z0 2πi C2 w − z0
(w − z0 ) 1 − (z − z0 ) 1 −
w − z0 z − z0
  −1  −1
z − z0 w − z0
Z Z
1 f (w) 1 f (w)
= 1− dw + 1− dw
2πi C1 w − z0 w − z0 2πi C2 z − z0 z − z0
"  2  3 #
z − z0 z − z0 z − z0
Z
1 f (w)
= 1+ + ++ + · · · dw
2πi C1 w − z0 w − z0 w − z0 w − z0
"  2  3 #
w − z0 w − z0 w − z0
Z
1 f (w)
+ 1+ + + + · · · dw
2πi C2 z − z0 z − z0 z − z0 z − z0

(z − z0 )2 (z − z0 )3
 
z−a
Z
1 1
= f (w) + + ++ + · · · dw
2πi C1 w − z0 (w − z0 )2 (w − z0 )3 (w − z0 )4
(w − z0 )2 (w − z0 )3
 
w − z0
Z
1 1
+ f (w) + + + + · · · dw
2πi C2 z − z0 (z − z0 )2 (z − z0 )3 (z − z0 )4
z − z0 (z − z0 )2
Z Z Z
1 f (w) f (w) f (w)
= dw + 2
dw + 3
dw
2πi C1 w − z0 2πi C1 (w − z0 ) 2πi C1 (w − z0 )

(z − z0 )3
Z Z
f (w) 1
+ 4
dw + · · · + f (w) dw
2πi C1 (w − z0 ) 2πi(z − z0 ) C2
Z Z
1 1
+ (w − z0 )f (w) dw + (w − z0 )2 f (w) dw
2πi(z − z0 )2 C2 2πi(z − z0 )3 C2
Z
1
+ (w − z0 )3 f (w) dw + · · ·
2πi(z − z0 )4 C2
∞ Z
X
n 1 f (w)
= (z − z0 ) · dw
n=0
2πi C1 (w − z0 )n+1
∞ Z
X 1 1
+ · f (w)(w − z0 )n−1 dw.
n=1
(z − z0 )n 2πi C2
 6.5. LAURENT’S SERIES 117

But

∞ Z
X 1 1
n
· f (w)(w − z0 )n−1 dw
n=1
(z − z0 ) 2πi C2

∞ Z
X
−n 1 f (w)
= (z − z0 ) · dw.
n=1
2πi C2 (w − z0 )−n+1

Therefore

∞ Z
X
n 1 f (w)
f (z) = (z − z0 ) · n+1
dw
n=0
2πi C 1
(w − z0 )
∞ Z
X
−n 1 f (w)
+ (z − z0 ) · dw
n=1
2πi C2 (w − z0 )−n+1
∞ ∞
X X bn
= an (z − z0 )n + ,
n=0 n=1
(z − z0 )n

where
Z
1 f (w)
an = dw, n = 0, 1, 2, 3, · · ·
2πi C1 (w − z0 )n+1

and
Z
1 f (w)
bn = dw, n = 1, 2, 3, · · · .
2πi C2 (w − z0 )−n+1

Hence the proof is complete.

P∞
Note 6.4 [2, Part 21, 13 min] In the Laurent’s expansion f (z) = n=0 an (z−
P∞ P∞
z0 )n + n=1 bn (z − z0 )−n of f (z), the part n=1 bn (z − z0 )−n is referred as the
P∞
principal part and the part n=0 an (z −z0 )n is referred as the regular part.

Note 6.5 [2, Part 21, 14:23 min] Many authors replace bn by a−n in their

texts. So, the Laurent’s expansion of f (z) takes the form

∞ Z
X
n 1 f (z)
f (z) = an (z − z0 ) , where an = dz,
n=−∞
2πi C (z − z0 )n+1
 6.5. LAURENT’S SERIES 118

for n = 0, ±1, ±2, ±3, · · · and C is a curve between C1 and C2 (of figure

of Theorem 6.5.1). Because, we know that in case of deformation, integration

along C1 and C2 are equal. So, C1 and C2 are not indicated separately.

Note 6.6 [2, Part 21, 17 min 20 sec] In light of the above discussion of

Note 6.5, many authors also write the Laurent’s expansion of f (z) in the form:
P∞ n
P−1
f (z) = n=0 an (z − z0 ) + n=−∞ an (z − z0 )n .

6.5.1. Difference between Taylor’s theorem and Laurent’s theo-

rem. [2, Part 21, 8 min] In case Taylor’s theorem f (z) needs to be analytic

on an open disk, when f (z) needs to analytic in an annulus region in Laurents

theorem. The power of (z − a) is positive in Taylor’s theorem. On the other

hand, power of (z − a) can be both positive and negative in case Laurent’s

theorem.

Problem 6.15 Obtain the Laurent’s series about the stated singularity for

each of the following functions.

e2z 1
(a) about z = 1 (d) about z = 3
(z − 1)3 z 2 (z
− 3)2
z z − sin z
(b) about z = −2 (e) about z = 0
(z + 1)(z + 2) z3
1
(c) (z−3) sin about z = −2
z+2

z2
Problem 6.16 Expand f (z) = in Laurent’s series for the
(z − 1)(z − 2)
region (a) 1 < |z| < 2 and (b) 0 < |z| < 1.
 6.6. SINGULARITY AND ITS TYPES 119

Solution Let

z2 A B
≡1+ +
(z − 1)(z − 2) z−1 z−2

⇒z 2 ≡ 1(z − 1)(z − 2) + A(z − 2) + B(z − 1) (35)

Putting x = 2 in (35), we obtain that B = 4. Again plugging x = 1 in (35) we

z2 −1 4
obtain that A = −1. So f (z) = =1+ + .
(z − 1)(z − 2) z−1 z−2
(a) We expand f (z) in such a way that the binomial expansion remains

valid in the region

1 z
1 < |z| < 2 ⇒ < 1 and < 1.
|z| 2

Therefore

−1 4 1 4
f (z) = 1 + + =1− −
z−1 z−2 z−1 2−z
 −1
1 4 1 1  z −1
=1−   −  z =1− z 1− z
 −2 1−
1 2 1 − 2
z 1− 2
z
   
1 1 1 1 z  z 2  z 3
=1− 1 + + 2 + 3 + ··· − 2 1 + + + + ···
z z z z 2 2 2
z2 z3
   
1 1 1 1
=1− + + + + ··· − 2 + z + + 2 + ···
z z2 z3 z4 2 2
1 1 1 1 z2 z3
= ··· − 4 − 3 − 2 − − 1 − z − − 2 − ···
z z z z 2 2

6.6. Singularity and Its Types

6.6.1. Zeros and its order. There are some points where the function

value of complex functions is zero. Those points are zeros of that complex

function.

Definition 6.6.1 (Zero) A point z0 is called a zero of f (z), if f (z0 ) = 0.

 6.6. SINGULARITY AND ITS TYPES 120

Example 6.1 z = 3 is a zero of f (z) = (z − 3)2 .

z3
Example 6.2 z = 0 is a zero of f (z) = .
z − 3ez

Definition 6.6.2 (Order of zero) A zero z = z0 of f (z) is said to be order n+

1, if f (z0 ) = f ′ (z0 ) = f ′′ (z0 ) = f ′′′ (z0 ) = · · · = f (n) (z0 ) = 0, but f (n+1) (z0 ) ̸= 0.

Example 6.3 z = 1 is a zero of f (z) = (z − 1)2 of order 2.

Example 6.4 Order of zeros of f (z) = z 4 is 4.

Definition 6.6.3 (Simple zero) A zero z = z0 of f (z) is said to be simple,

if its order is 1.

6.6.2. Singularity. There may exist some points where a function may

not be analytic. Study of singularity deals with such types of points.

Definition 6.6.4 (Singularity) [3, Page 245] [4, Page 3.5] A point at which

f (z) fails to be analytic is called a singular point or singularity of f (z).

Note 6.7 Sometimes, singular points are referred as critical points, see [3,

Page 245].

6.6.3. Classification of singularities. Mainly, there two types of sin-

gularities: isolated and non-isolated.

Definition 6.6.5 (Isolated and non-isolated singularity) A singularity z = z0

of f (z) is called an isolated singular point or isolated singularity, if there

exists a neighbourhood |z − z0 | < δ of z = z0 such that there is no singularity

 6.6. SINGULARITY AND ITS TYPES 121

within 0 < |z − z0 | < δ. A singularity which is not isolated, is called a non-

isolated singularity.

1
Example 6.5 z = 0 is an isolated singularity of f (z) = .
z
 
z−2 1
Example 6.6 z = 0, 1 are isolated singularities of f (z) = sin .
z2 z−1

Note 6.8 Isolated singularity of f (z) can be interpreted as the limit point
 
1
of zeros of f (z). For example, consider the function f (z) = sin . Now
z
1 1
f (z) = 0 implies that = nπ. So z = . Thus, the set of all zeros of f (z)
  z nπ
1
is : n ∈ Z , limit point of which is 0. So z = 0 is an isolated singularity
nz
of f (z).

Isolated singularities are of three types: removable singularity, pole, and

essential singularity.

Definition 6.6.6 (Removable singularity) Let z = z0 be an isolated singu-

larity of f (z). Then z = z0 is called a removable singularity, if limz→z0 f (z)

exists.

Redefining f (z) at z = z0 , by setting f (z0 ) = limz→z0 f (z), it can be shown

that f (z) becomes continuous and analytic at z = z0 . So, an isolated singu-

larity z = z0 of a complex function f (z) is said to be a removable singularity,

if the function f (z) can be redefined at z = z0 in such a way that it becomes

analytic there.
 6.6. SINGULARITY AND ITS TYPES 122

sin z
Example 6.7 Clearly, z = 0 is an isolated singularity of f (z) = . Again,
z
limz→z0 f (z) = 1. So, f (z) can redefined at z = 0 in the following way:

 sin z
, if z ̸= 0



f (z) = z ,


1,
 if z = 0

which is analytic at z = 0. Thus, z = 0 is a removable singularity of f (z).

Mnemonic The concept of removable singularity is similar to the concept

of continuous extension.

Note 6.9 If the principal part of the Laurent’s series expansion of f (z) about

z = z0 does not contain any part (that is, principal part does not exist), then

z = z0 is a removable singularity.
 CHAPTER 7

Conformal Mapping and Its Appliactions

For solving more problems the reader is referred to [4], [3].

7.1. Conformal Mapping

Definition 7.1.1 Let w = f (z) be a function and z0 ∈ Domain(f ). Then f

is said to be analytic at z = z0 , if there exists a neighbourhood

Vδ (z0 ) = {z : |z − z0 | < δ}

of z = z0 such that f ′ (z) exists at p for any p ∈ V .

123
 APPENDIX A

Necessary Theorems and Formulae

A.1. Green’s Theorem

Theorem A.1.1 (Green’s theorem) [4, 3] If P (x, y) and Q(x, y) are contin-

uous and have partial derivatives in a region Γ and on its boundary curve C,

then
I ZZ  
∂Q ∂P
P dx + Q dy = − dxdy.
C ∂x ∂y
Γ

Note A.1 [4, 3]Green’s theorem is valid for both simply-connected and

multiply-connected regions.

A.2. Necessary Trigonometric Formulae

Formula A.2.1 [2, Part 11, 36 min, 25 sec] For n ∈ N,




1,
 if n is even
cos nπ =


−1, if n is odd


Briefly, cos nπ = (−1)n for n ∈ N.

Formula A.2.2 [2, Part 11, 36 min, 44 sec] sin nπ = 0 for any n ∈ Z.

A.3. Tracing Polar Curves

r = 1 + a cos θ represents a cardioid.

124
 Bibliography

[1] J. W. Brown and R. V. Churchill, Complex Variables and Applications, Eight ed.,

McGraw-Hill, 2014-2015.

[2] S. Nowaj, Complex Analysis in Bengali, S. N. Math Tutorials, Youtube Channel,

Last Updated in 2023, URL: https://www.youtube.com/playlist?list=PLPgzE_

2cFfOrE7d0vtvT5ieg9x-sdON-1.

[3] M. F. Rahman, Complex Analysis, Titas Publicatins, 2023.

[4] M. R. Spiegel, S. Lipschutz, J. J. Schiller and D. Spellman, Complex Variables, Second

(Special Indian) ed., Schuam’s Outline Series, Tata McGraw Hill Eduacation Private

Limited, New Delhi, 2010.

125