MSO202A: Introduction To Complex Analysis

Instructor:
Dr. G. P. Kapoor
FB 565, Department of Mathematics and Statistics
Tel. 7609, Email: gp@iitk.ac.in
 2

Lecture 1

Text Books:

 E. Kreyszig, Advanced Engineering Mathematics, 8th Ed., John

Wiley & Sons.

 Ruel V. Churchill, et al: Complex Variables and Applications,

McGraw Hill.

 John B. Conway: Functions of One Complex Variable, II Ed.,

Springer International Student Addition.
 3
Reference Books:

 Jan G. Krzyz: Problems in Complex Variable Theory, American

Elsevier Publishing Company.

 Lars V. Ahlfors: Complex Analysis, McGraw Hill.

 4
Supplementary Course Material

 Lecture Notes, Assignments and Course Plan will be available

on this course at the webpage http://home.iitk.ac.in/~gp
through the link MSO202A .

In the lecture notes, some proofs are marked (*). Such proofs
will not be asked in the exams.

Tutorial Classes

 The assignment problems marked (T) on the assignment

sheets will be discussed in the tutorial classes.

 The solutions/hints to the assignment problems marked (D)

will be made available on the course web‐site.

 The exercises given in the text books are usually not discussed
in the tutorial classes and the students are expected to solve
these problems on their own. However, the students can
approach the tutor if they have any difficulty in solving such
problems.
 5

Evaluation plan

 There will be 2 pre‐announced Quizzes of 40‐minutes

duration and a weightage of 20% marks for each.

 The End‐Course Examination will be of 2‐hours duration with

a weightage of 60% marks.
 6

Review of Complex Number System

Complex numbers were introduced to have solutions of

equations like x 2  1  0 which do not possess a solution in the
real number system.

A complex number z is an ordered pair ( x, y ) of real numbers.

If z1  ( x1 , y1 ), z2  ( x2 , y2 ) , the elementary operations are
defined as

z1  z2  ( x1  x2 , y1  y2 )
z1  z2 if x1  x2 , y1  y2
 z1  (  x1 ,  y1 )
z1 z2  ( x1 x2  y1 y2 , x1 y2  x2 y1 )
z  ( x,  y ) , z  x 2  y 2
 7

Notations:

 Throughout in the sequel, denote Complex

number ( a ,0)  a , i  (0,1) . With these notations, R  C,
where R is set of all real numbers and C is set of all
complex numbers.

 The Euclidean distance between any two points z1 , z2 C is

defined as z1  z2 and is sometimes denoted by d ( z1 , z2 ) .

Note that i  i,| i |  1, i 2  1. Thus, the complex number i is

the solution of the equation x 2  1  0 .

Further, writing x  ( x,0), y  ( y ,0), i  (0,1) , it is easily seen by

using the definitions of addition and product of complex
numbers that x + i y = z. This is called the cartesian
representation of the complex number z.
 8

2
Proposition 1. zz  z
2
Proof: ( x, y ).( x,  y )  ( x 2  y 2 ,0)  x 2  y 2  z

1 x y
Proposition 2. ( 2 , ), if z  0
z x y x y
2 2 2

1z z x y
Proof:  2 ( 2 , ).
zz z x y x y
2 2 2

zz zz
Proposition 3. Re z  , Im z 
2 2i
Proof. We give here the proof of the second part of the
proposition. The first part follows similarly.
 z  z  i (0,2 y )(0, 1) (2 y,0)
   y  Im z .
2 2 2
 9

Polar representation of Complex Numbers

With x  r cos  , y  r sin  , z  r (cos   i sin  )  ( r cos  , r sin  )

is called the polar representation of the complex number z.

r z ,
  angle between the line segment from 0 to z and
positive real axis
 arg z

It follows immediately that z1  z2  r1  r2 and 1   2  2k .

Further, if z1  r1 (cos 1  i sin 1 ), z2  r2 (cos  2  i sin  2 ) , it
follows that
z1 z2  r1r2 (cos(1   2 )  i sin(1   2 )) .
Thus, arg( z1 z2 )  arg( z1 )  arg( z2 ) .
 10
Similarly, using induction, if
z j  rj (cos  j  i sin  j ), j  1,2,..., n, then
z1 z2 ... zn  r1r2 ...rn (cos(1  ...   n )  i sin(1  ...   n ))
Thus, arg( z1 z2 ... zn )  arg z1  arg z2  ...  arg zn .
In particular, z n  r n (cos n  i sin n ), n  0.

To prove this identity for n < 0, we have

1 1 cos   i sin  1
   (cos   i sin(  )).
z r (cos   i sin  ) r r

So that, for n < 0,

1
z n  ( z 1 )  n  [ (cos   i sin(  )] n
r
 r n [cos n  i sin n ].
Thus, z n  r n [cos n  i sin n ] for all integers n. Taking r = 1 in
this identity,
(cos  i sin  )n  cos n  i sin n for all integers n,
which is called De‐Moivre’s Theorem.
 11

A special word about argument of a complex number

arg z is not a function, since for z  rei , arg z has all the values
 ,  2 ,  4 ,... so it is not single valued.

The identity
arg( z1 z2 )  arg z1  arg z2
has to be interpreted in the sense that for some value of arg on
LHS,  suitable values of arg z1 and arg z2 on RHS so that
equality holds. Conversely, for given values of arg z1 and arg z2
on RHS,  suitable values of arg( z1  z2 ) on RHS so that
equality holds.

For example, if z1  z2  i and the values of their arguments

3 3
are given as arg z1  , arg z2  , then z1 z2  1 and out of
2 2
all the values 3  2k , k  0,1,2,... . of arg( z1 z2 ) , we must
choose arg( z1 z2 )  3 , so that arg z1  arg z2  arg( z1 z2 ) holds.

1

arg(z1z2 )  3 3
i arg z  arg z 
1 2
2
 12

Conversely, if arg( z1 z2 )  5 is given, then we can take

 3 11 3
arg z1     2 , arg z2    4 .
2 2 2 2

To make arg z a function of z in the strict sense of the definition

of a function, we restrict the range of arg z as (  ,  ] (or with
another convention, some authors restrict this range as
[0,2 ) ). Once the range of arg z is so restricted, arg z is
denoted by Arg z.

Thus,

  Arg z  

(or, 0  Arg z  2 with the other convention).

 13

y
Remark: If z  x  iy , the principal value of tan 1 , denoed as
x
y  y 
Tan 1 , satisfies   Tan 1  , while   Arg z   . The
x 2 x 2
y
relation between Arg z and Tan 1 is therefore given by the
x
following:

 1 y
 Tan , if x  0
x

 y
Arg z  Tan 1   , if x  0, y  0
 x
 1 y
Tan   , if x  0, y  0
x

Graph of Tan 1 x
 14

Remark: Note that, in general,

Arg ( z1 z2 )  Arg z1  Arg z2 (1)

For example, with the convention   Arg z  


if z1  1, z2  i, then Arg z1   , Arg z2  , so that
2
3
Arg z1  Arg z2  .
2


But, Arg ( z1 z2 )   . This illustrates (1), when the convention
2
is   Arg z   .

Similarly, with the convention 0  Arg z  2 , if

3
z1  1, z2  i, then Arg z1   , Arg z2  , so that
2
5
Arg z1  Arg z2  .
2

But, Arg ( z1 z2 )  . This illustrates (1), when the convention is
2
0  Arg z  2 .
 15

Solution of the equation z n  c :

Let
c  r0 (cos 0  i sin 0 )
and
z  r (cos   i sin  )

The equation z n  c gives

r n (cos n  i sin n )  r0 (cos(0  2 k )  i sin(0  2 k ))

 r n  r0 and n  0  2k
 0  2 k
r r01/ n and   , k  0,1,..., n  1
n

Therefore, the n solutions of the equation z n  c are

 0  2 k  0  2 k
zk  r01/ n [cos( )  i sin( )], k  0,1,..., n  1.
n n

Since, zk  r01/ n , all the roots of z n  c lie on the circle

C (0, r01/ n )  {z : z  r01/ n }. Further, since the angles
 0  2 k
, k  0,1,..., n  1, divide this circle in to n equal sectors,
n
all these roots are equispaced on C (0, r01/ n ).
 16

Example: All the roots of z n  1, called nth roots of unity, can be

written as

2 2 2( n  1) 2( n  1)
cos0  i sin 0, cos  i sin ,...,cos  i sin
n n n n
or

2 2
1, n , wn2 ,..., wnn 1; where wn  cos  i sin .
n n

Note that if z0 is any root of the equation z n  c , then all the

roots of this equation are given by

z0 , z0n , z0 wn2 ,..., z0 wnn 1

1/ n  0  2 k  0  2 k
since, zk  c [cos( )  i sin( )], k  0,1,..., n  1
n n
gives that
0  2k  2l 0  2k  2l
zk nl  c
1/ n
[cos( )  i sin( )],
n n ,
0  l  n 1

whose distinct values are obtained for k  k  l  k  n  1.

 17

Vector Representation of Complex Numbers

Any complex number z  ( x, y ) can be represented as the

vector
  
z  x i  y j  r , ( say ) .

This representation helps in geometrically visualizing addition

and subtraction of complex numbers as vectors.

However, it does not help in visualizing the product of complex

numbers as this is different from the vector product of
corresponding vectors.
     
( since, if z1  x1i  y1 j  r1 and z2  x2 i  y2 j  r2 , then
z1 z2  ( x1 x2  y1 y2 , x1 y2  x2 y1 ) is in xy  plane itself
 
while, for the corresponding vectors r1 , r2 ,
i j k
   
r1  r2  x1 y1 0  ( x1 y2  y1 x2 )k  ( z1 z2 )k
x2 y 2 0
is perpendicular to xy  plane.)
 18

Representation of Points, Curves and Regions by Complex

Numbers

Representation of Points
z
z

z
-z

z z z
1 2
iz
2

90 z
z
1

z
r

1/r -
1/z
 19
Equation of a circle and disk in terms of Complex Numbers

Equation of a circle with center z0 and radius r :

z  z0  r
Equation of an open disk with center z0 and radius r :
z  z0  r
Equation of a closed disk with center z0 and radius r :
z  z0  r

Equation of a Line in terms of Complex Numbers

 
Equation of a line L passing through a and parallel to vector b
is
  
r  a  t b ,   t  

or, in terms of notation of a complex variables z, a and b, this

equation is

z  a tb
za za
t   Im( )  0.
b b

Thus, equation of the line L is given by

 za 
L   z :Im( )  0.
 b 
 20

* Algebraic Structure of Complex Numbers

Field: ( X , , .) is a field if
(i) (X, +) is an abelian group.
(ii) (X – {0}, .) is an abelian group.
(iii) ‘.’ is distributive over ‘+’.

It is easily verified that (C, +, .) is a field that contains the field

(R, +, .).

Ordered Set: (X, <), where, ‘<’ is a relation, is called an ordered

set if
(i) One and only one of the statements x < y, x = y, y < x
holds for any x and y.
(ii) ‘<’ is transitive.

Ordered Field: An ordered set X is called an ordered field if

(i) X is a field
(ii) X is an ordered set
(iii) If y < z, then x + y < x + z for all x, y and zX
(iv) If x > 0, y > 0, then xy > 0.

It is easily verified that (C, +, .) is a field as well as an ordered

set with respect to dictionary ordering (dictionary order on 
is defined by
( x1 , y1 )  ( x2 , y2 ) if either x1  x2 or if x1  x2 then y1  y2 .
 21
However, (C, +, .) is not an ordered field with any order,
since in every ordered field 1 is always positive (for, either 1 is
positive or ‐1 is positive and, if ‐1 is positive, then (‐1)(‐1) = 1
is positive, which is a contraction), so that ‐1= (‐1, 0) is always
negative. Now, either (0, 1) > 0 or ‐ (0, 1) > 0
If (0,1)  0 then (0,1).(0,1)  ( 1,0)  (0,0) ,
which implies (R, +, .) can not be an ordered field.

If (0,1)  0 then  (0,1).  (0,1)  ( 1,0)  (0,0) , which

implies (  , +, .) can not be an ordered field.

Alternatively, in every ordered field, square of every element is

positive. This gives ‐1 is positive being square of (0,1), a
contradiction since ‐1 is always negative as in the above
arguments.