AAU

2006
Complex Analysis
Instructor Hunduma Legesse
M.T
W W W. A A U . E D U . E T






Hunduma Legesse(PhD)





Addis Ababa University
Mathematics Department
Hunduma Legesse



CHAPTER 1: COMPLEX NUMBERS


Why complex numbers?
Why do we need new numbers?

The hardest thing about working with complex numbers is, understanding why you might
want to. Before introducing complex numbers, let us backup and look at simpler
examples of the need to deal with new numbers.

If you are like most people, initially number meant whole number, 0, 1, 2, 3 whole
number makes sense. They provide a way to answer questions of the form How many
? You are learned about the operations of addition and subtraction, and you found that
while subtraction is a perfectly good operation, some subtraction problems, like 3-5,
dont have answers if we only work with whole numbers. Then you find that if you are
willing to work with integers, ,-3, -2, -1, 0, 1, 2, 3, ,then all subtraction problems do
have answers! Furthermore, by considering examples such as temperature scales, you see
that negative numbers often make sense.

Now in integers all subtraction problems have got answers. But if we want to deal with
division, some, in fact most, division problems do not have answers in integers. For
example 12, 32, 53 and the like are not integers. So we need new numbers! Now we
have rational numbers to furnish such problems.

There is more to this story. There are problems with square roots and other operations,
but we will not get into that here. The point is that you have had to expand your idea of
number on several occasions, and now we are going to do that again.

The problem that leads to complex numbers concerns solutions of equations like,
(1) 0 1
2
= x (2) 0 1
2
= + x




Which equation has real roots? (Equation (1) or (2))
Can you explain?
To equation (1) -1 and 1 are the two real roots but equation (2) has no real root since
there is no real number whose square is negative. So if equation (2) is to be given
solutions then we must create a square root of -1.

In general for all quadratic equation of the form 0
2
= + + c bx ax to have a solution we
must need a number system in which ac b 4
2
is defined for all numbers a, b and c. The
number system which we are going to define is called the complex number system.

The Complex Number System

Activity 1

1. Consider the set {(x, y)/ x and y are real numbers} in the coordinate plane.
a. What is the difference between (2, 3) and (3, 2)?
b. When do you say that (a, b) = (c, d)?
c. What is the sum (2, 3) + (5, 2)?
d. Can you generalize the sum for (a, b) + (c, d)?

2. Identify whether the following lies on the X-axis or the Y-axis.
(2, 0) (-3, 0)
(1/2, 0) (0.234, 0)
( 2 , 0) (0, 2)
(0, -3) (0, 0.123)
(0, y) (x, 0) for any real numbers x and y.

3. Suppose the product of two ordered pairs (a, b) and (c, d) is defined as follows:
(a, b).(c, d) = (ac-bd, ad + bc)




Try to find the following:
a. (2, 3).(5, 2)
b. (3, 0).(4, 0)
c. (0, 3).(0, 4)
d. (1, 0).(1, 0)
e. (0, 1).(0, 1)
f. ( ) 0 ,
1
x ( ) 0 ,
2
x
g. ( ) , 0 )( , 0
2 1
y y for any real numbers ,
1
x ,
2
x
2 1
, , y y
Now we are in a position to define the system of complex numbers.

Definition:- The system of complex numbers denoted by C is the set of all ordered pairs
(x, y) of real numbers x and y which satisfies the following three conditions.
(i) (a, b) = (c, d) iff a = c and b = d
(ii) (a, b) + (c, d) = (a + b, c + d)
(iii) (a, b).(c, d) = (ac - bd, ad + bc)
Notation: For algebraic purposes we use the following notations.
- (x, 0) = x for any real number x.
- (0, 1) = i
- (0, y) = yi
Exercise
1. Use the above notation to show that i
2
= -1.
2. Any complex number (x, y) can be written us x + iy.
3. Simplify i
3
, i
4
, i
8
, i
11

Now using the above notations we can redefine complex numbers as:
Definition: A complex number z is one of the form z = x + iy for some real numbers x
and y.
(i) the number x is called the real part of z and is denoted by Re(z).
(ii) the number y is called the imaginary part of z and is denoted by Im(z).


Example: a) In z = 2-5i, Re(z) = 2 and Im(z) = -5
b) In z = 6+4i, Re(z) = 6 and Im(z) = 4
c) In z = 0 +2i = 2i, Re(z) = 0 and Im(z) = 2
d) In z = 0 + 0i = 0, Re(z) = 0 and Im(z) = 0
e) In z = 4 + 0i = 4, Re(z) = 4 and Im(z) = 0
Now we need to define the four arithmetic operations and equality on complex numbers
using the new notation. If z = x + iy and w = a + bi then
Equality: z = w if and only if x = a and y = b.

Addition and Subtraction
To add or subtract two complex numbers, you add or subtract the real parts and the
imaginary parts.
(i) z + w = (x + a) + (y + b)i
(ii) z w = (x a) + (y b)i

Example: a) (3 5i) + (6 + 7i) = (3 + 6) + (-5 + 7)i = 9 +2i
b) (3 5i) - (6 + 7i) = (3 - 6) + (-5 - 7)i = -3 -12i

Note: These operations are the same as combining similar terms in expressions that have
a variable. For example, if we were to simplify the expression (3 5x) + (6 + 7x) by
combining similar terms, then the constant 3 and 6 would be combined, and the terms
(-5x) and (7x) would be combined to yield (9 + 2x).

Multiplication:
The formula for multiplying two complex numbers is
zw = (xa yb) + (xb ay)i.
You do not need to memorize the formula, because you can arrive at the same
result by treating the complex numbers like expressions with a variable,
multiply them as usual, then simplify. The only difference is that powers of i
do simplify, while powers of x do not.

Example: (2 + 3i)(4 + 7i) = 2x4 + 2x7i +4x3i + 3ix7i
= 8 + 14i + 12i -21
= (8 21) + (14 + 12)i
= -13 + 26i
Exercise: Perform the following operations.
1) (-3 + 4i) + (2 2i)
2) 3i (2 4i)
3) (2 7i)(3 + 4i)
4) (1 + i)(2 3i)
5) (2 i) i(1 2i)

1.3. Complex Conjugate and Modulus

Activity: Given a complex number z = x + iy and w = x iy, find
a) The product zw.
b) The sum z + w
c) The difference z w
d) If you represent z = x + iy as (x, y) and w = x iy as (x, -y) on the
number plane, then can you tell any relation between z and w?

We call w = x iy the conjugate (or complex conjugate) of the complex number z = x
+yi. Actually if you look both of them by putting on the number plane you find that one
is the reflection of the other. The conjugate of a complex number z is simply the
reflection of z through the X-axis. Conjugates are important because of the fact that a
complex number times its conjugate is real. That is, (x + yi)(x yi) = x
2
+ y
2





DEFINITION: The complex conjugate or the conjugate of a complex number z = x+yi
is denoted by z is given by z x yi =
Example 1:
1. If , 6 5 i z = then i i z 6 5 ) 6 ( 5 + = =
2. If ,
2
1
1 i z + = then i z
2
1
1 =
3. If , 0 4 4 i z + = = then i z 4 =
4. If , 2i z = then i z 2 =

Example 2:
Number z Conjugate of z Product
2 + 3i 2 -3i 13
3 5i 3 +5i 34
4i -4i 16
5 5 25

From the above activities, definition and examples we can draw the following
conclusions as a theorem:

THEOREM: For any complex numbers z and w, we have
i) z z =
ii) ) Re( 2 z z z = +
iii) ) Im( 2 z i z z =
iv) w z w z = = +
v) w z w z . . =
vi)
z z
w
w
| |
=
|
\ .
, if w = 0



Note: (iv) and (v) of the above Theorem can be extended to any finite number of terms.
That is,
i)
n n
z z z z z z + + + = + + + ... ....
2 1 2 1

ii)
n n
z z z z z z z .... . . . ..... .
3 2 1 2 1
=

Division:
Division is the inverse process of multiplication; that is,

z
w
= u iff z = wu.provided w = 0
If z = x + yi , w = a +bi and u = c + di, then one can solve
x + yi = (a +bi)(c + di) to get
c =
2 2
ax by
a b
+
+
and d=
2 2
ay bx
a b

+

Thus u =
z
w
=
2 2
ax by
a b
+
+
+
2 2
ay bx
a b

+
i
Or we can use conjugates to simplify expressions of the form (x + yi) (a + bi) by
writing in the form of (x + yi)/ (a + bi) and multiplying both the numerator and
denominator by,
a bi which, is the conjugate of a + bi.
Example: if i z 3 2 + = and i w = 5 than
i
i
i
w
z
|
|
.
|

\
|
+

+
|
|
.
|

\
|
+
+
=

+
=
2 2 2 2
) 1 ( 5
) 1 ( 2 ) 3 ( 5
) 1 ( 5
) 1 ( 3 5 . 2
5
3 2

i
26
17
26
7
+ =

THEOREM: (C, +, ) is a field.
(Proof: exercise)




DEFINITION: The absolute value (or modulus) of a complex number , yi x z + =
denoted by z , is defined to be

2 2
y x z + =

Example:
1. If , 5 2 i z + = then 29 5 2
2 2
= + = z
2. If , 12 5 i z + = then 13 169 12 5
2 2
= = + = z
3. If , i z = then 1 1
2
= = z
4. If , 2 = z then 2 / 2 / ) 2 (
2
= = = z

Note: If z
1
= x
1
+y
1
i and z
2
= x
2
+y
2
i then
( ) ( ) ( ) ( )
2
1 2 1 2 1 2 1 2 1 2
z z x x y y i x x y y = + = +

2 1
z z is the distance between
1
z and
2
z
THEOREM: For any two complex numbers z and w
(i)
2
. z z z =
(ii) z z =
(iii) ( ) z z s Re
(iv) ( ) z z =s Im
(v) w z w z = .
(vi)
w
z
w
z
= if 0 = w
(vii) Triangle Inequality: w z w z + s +
(viii) w z w z >



Proof: Let yi x z + = and vi u w + = for some
9 e v u y x , , ,

(i) Since z = x - yi
( )( ) ( ) ( ) ( ) ( ) = + = + + + = + =
2
2 2 2 2
z y x i x y y x y x yi x yi x z z
( )
2
2 2 2 2
z x x oi y x = + = + = =

(ii) Since z x y = i we have
( )
2
2 2 2
z x y x y z = + = + =
(iii) Since
2 2 2
y x x + s , we have


(iv) Since
2 2 2
y x y + s , we have
( ) z y x y y z = + s = =
2 2 2
Im
(v) ( ) ( ) w z z w z = 2
2
bi (i)
( ) ( ) ( ) ( ) z w z z w z w z = = .
( )
2 2
w z w z
w
= =
w z w z =

(vi)
2
2
2
|
|
.
|

\
|
= =

= =
|
|
.
|

\
|
|
.
|

\
|
=
w
z
w
z
w w
z z
w
z
w
z
w
z
w
z
w
z
w

w
z
w
z
= Provided that 0 = w


( ) z y x x x z = + s = =
2 2 2
Re




(vii) ( ) ( ) ( )( ) w z w z w z w z w z + + = + + = +
2

w w z w w z z z + + + =
2 2
w w z w z z + + + =
( )
2 2
Re 2 w w z z + + =
| |
2 2 2
2 w z w w z z + = + + s
w z w z + s +
(viii) ( )( ) ( )
2 2 2
Re 2 w w z z w z w z w z + = =
( )
2 2 2
2 w z w w z z = + >
w z w z >

Note: that the Triangle Inequality can be extended to any finite sum by induction to give

n x n z
z z z z z z + + + s + + +
1 2

Example: If 1 = z , then show that 1 =
+
+
a bz
b az
, where a z b =
Solution:
( ) z b az
b az
z
z a b
b az
a z b
b az
+
+
= |
.
|

\
|
+
+
=
+
+ 1
as 1 = z z

( )
1 =
+
+
=
+
+
=
+
+
=
+
+

b az
b az
z b az
b az
z b az
b az
a z b
n az







1.4. Polar Coordinates of Complex Numbers

A) Geometric Representation of Complex Numbers
The complex number yi x z + = is uniquely determined by the ordered pair of real
numbers ( ) y x, . The same is true of the point ( ) y x p , in the plane with Cartesian
coordinates x and y. Hence it is possible to establish a 1-1 correspondence between
all complex numbers and all points of the given plane. We merely associate the
complex number z = x+iy with the point ( ) y x p , . The plane whose points represent
the complex numbers is called the complex plane or the z-plane. The real numbers
or points ( ) 0 , x x = are represented by points on the x-axis, hence the x-axis is called
the Real axis. And pure imaginary numbers or points ( ) y iy , 0 = are represented by
points on the y-axis, and hence we call the y-axis the imaginary axis.

Instead of considering the point ( ) y x p , as the representation of yi x z + = we may
equally well consider the directed segment or the vector extending from O to P as the
representation of the complex number yi x z + = . In this case, any parallel segment of
the same length and direction is taken as corresponding to the same complex number
0
1
2
3
1
2
x
y
P (2,-3)
z =2-3i
Real axis
Imaginary axis
z- Plane
p(-4,2)
z =4+2i



The representations of z z, , the sum and difference of complex numbers are as
follows:
z is the length of the vector
representing the complex number
z.

z is the reflection of z with respect
to the real axis.


The sum
2 1
z z + is represented
by the vector sum of the
vectors
1
z and
2
z





Real axis
z =z+yi
z =x-yi
z
Imaginary axis
Real axis
Imaginary axis
z
1
z
2
z
1
+z
2



The difference
2 1
z z is represented by a
vector from the point z
1
to the point z
2







Note that the product
2 1
z z of two complex numbers is a complex number. Therefore,
this product is neither the scalar product nor the vector product used in vector analysis.
Consequently, our complex numbers of z cannot be identified with the vectors of two
dimensional vector analyses.

Example:
a) If z
1
and z
2
are two given complex numbers as in the figure (a) below, construct
graphically
2 1
2 3 z z
b) If
3 2 1
, , z z z and
4
z are four given complex numbers as in the figure, construct
graphically
4 3 2 1
z z z z + + +
Solution:


Real axis
Imaginary axis
z
1
z
2
z
1
-z
2
-z
2
Real axis
Imaginary axis
z
1
z
2
3z
1
-2z
2
3z
1
-2z
2
(a)

Real axis
Imaginary axis
z
3
z
1
+z
2
+z
3
+z
4
z
4
z
3
z
4
z
2
z
1
z
2
(b)





B) POLAR FORM
L et ( ) u , r be the polar coordinates of the point representing the complex
number yi x z + = , . 0 > r
Then, u cos r x =
u sin r y =

2 2
y x z r + = =
|
.
|

\
|
=

x
y
1
tan u

r
z =x +yi
Real axis
Imaginary axis
u

Hence we can write the complex number yi x z + = as follows
u u sin cos r r z + = ( ) u u sin cos + = r
This form is called the polar form.

Notation: In short we can write, u u sin cos r r z + = ( ) u u sin cos + = r := rcisu .
The angle u is called the argument of z and we write as, ( ) z arg = u .
If ( ) z arg = u , then 2 , , n n Z u t + e is also argument of z.. Arg(z) is called the principal
argument of z, is the value of argument of z in the interval ( | t t, , that is,
( ) t t s < z arg .
If ( ) z Arg is the principal argument of z, then ( ) ( ) Z n n z Arg z e + = , 2 arg t describe all
possible values of ( ) z arg .

Example: Express the following complex number in polar form.
a) i z 3 2 2 + = b) i z 5 5 + =
c) i z 3 = d) 1 = z





Solution:
(a) ( ) 4 16 12 4 3 2 2
2
2
= = + = + r
( ) ( ) z Arg
x
y
= = =
|
|
.
|

\
|
= |
.
|

\
|
=

3
3 tan
2
3 2
tan tan
1 1 1
t
u
Therefore; ( )
3
4
3
sin
3
cos 4 3 2 2
t t t
cis i i z = + = + =
is the polar form of z.

(b) ( ) 2 5 50 25 25 5 5
2 2
= = + = + = r
( ) ( ) z Arg
x
y
= = = |
.
|

\
|

= |
.
|

\
|
=

4
3
1 tan
5
5
tan tan
1 1 1
t
u
Therefore;
4
3
2 5
t
cis is the polar form of z.

(c) ( ) Z n n r e |
.
|

\
|
+ = = = = + =

,
2
1
2 1 sin , 3 9 3 0
1 2 2
t u and
( ) Z n
n
e
+
= =

,
2
1 2
0 cos
1
t u . In particular if n = 0 then
.
2
t
u = In general, Z n
n
e
+
= ,
2
1 4
t u
The principal argument: ( ) .
2
t
= z Arg
Therefore, ( )
2
3
t
cis is the polar form of z.

(d) ( ) ( ) ( ) ( ) Z n n r e + = = . = = + =

, 1 2 1 cos 0 sin , 1 0 1
1 1 2 2
t u u u
The principal argument: ( ) . t = z Arg
Therefore, ( ) t cis is polar form of z.




DEFINITION: Let
1 1 1
u cis r z = and .
2 2 2
u cis r z = Then
. , 2
2 1 2 1 2 1
Z k k r r z z e + = . = = t u u
Remark: If
1 1 1
u cis r z = and
2 2 2
u cis r z = then, ( )
2 1 2 1 2 1
u u + = cis r r z z
Proof: ( ) ( ) i r r z z
2 2 2 1 1 1 2 1
sin cos sin cos u u u u + + =
( ) ( ) | | i r r
1 2 2 1 2 1 2 1 2 1
sin cos sin cos sin sin cos cos u u u u u u u u + + =
( ) ( ) ( ) i r r
2 1 2 1 2 1
sin cos u u u u + + + =
( )
2 1 2 1
u u + = cis r r
The geometric representation of
2 1
z z is best seen by use of the polar form of the product.
Given
1 1 1
u cis r z = and .
2 2 2
u cis r z = Choose the number 1 = z on the real axis and form
the triangle ( ). , 1 , 0
1
z A Then with vector
2
z as one side, construct triangle ( ) P z
, 2
, 0
similar to the first triangle, while keeping the orientation of the equal angles
1
u the same.
By similarity of triangles

2 1 2 1
r r z z OP = = , and ( )
2 1
arg u u + = P Hence
2 1
z z p =

real axis
Imaginary axis
u
1
u
2
u
3
z
2
P
z
1
1 0


( )
2 1
2
1
2
2
2 2
1 1
2 2
1 1
2
1
u u
u
u
u
u
u
=

= = cis
r
r
cis
cis
cis r
cis r
cis r
cis r
z
z
if 0
2
= z
Hence the quotient of two complex numbers
1
z and , 0
2
= z in polar form is
( )
2 1
2
1
2
1
u u = cis
r
r
z
z


The geometric representation of
2
1
z
z
is as follows:
By similarity of triangles
( ) ( ), , , 0 , 1 , 0
1 2
z Q z A ~ A

We have
OP
z
z
1
2
1
=
real axis
Imaginary axis
u
2
u
1
u
3
z
1
z
2
1 0
Q


1.5. Powers and Roots
THEOREM: Let
n
z z z , ,
2 1
be non-zero complex numbers such that
i i
r z = and
( )
i i
z u = arg for . , , 2 , 1 n i = Then ( ).
2 1 3 2 1 n n
cis z z z z u u u + + +
Proof: We prove by Mathematical Induction. For ( )
2 1 2 1 2 1
1 , 2 u u + = = cis r r z n shown
above. Assume that it is true for , k n = that is, ( ).
1 2 1 2 1 k k k
cis r r r z z z u u + + =
To show that it also holds true for , 1 + = k n
( ) ( ) ( )
1 1 1 2 1 1 2 1 = + +
+ + =
k k k k k k
cis r cis r r r z z z z u u u
( ) ( )
1 1 1 2 1 + +
+ + + =
k k k k
cis r r r r u u u
( )
1 2 1 1 2 1 + +
+ + + =
k k k
cis r r r r u u u
Hence it is true for all . 2 > n Therefore,
( ) 2
2 1 2 1 2 1
> + + + = n cis r r r z z z
n n n
u u u



COROLLARY: (De Moivre's Formula)
If , N n and cis r z e = u then
(i) ( ) u n cis r z
n n
= (ii) ( ) u n cis
r
z
z
n n
= =

1 1
1

Examples: 1. Calculate
(a) ( )
10
1 i + (b) ( )
16
3 i +
1
2
1
z
z
OQ
=
2
1
2
1
r
r
z
z
OQ = = and ( )
2 1
arg u u = Q
Hence
1
z
z
Q =
Solution: (a) since,
|
.
|

\
|
= +
4
3
2 1
t
cis i , by De Moivre's Formula we have
( ) ( )
|
.
|

\
|
=
|
|
.
|

\
|
|
.
|

\
|
= +
2
15
32
4
3
10 2 1
10
10
t t
cis cis i
( ) ( )
2
32
2
8 32
t t
t

= = cis cis
( ) ( ) i i 32
2
sin
2
cos 32 =

+ =
t t


(b) Since,
6
2 3
t
cis i = + , we have
( ) |
.
|

\
|
= = +
3
8
2
6
2 3
16 16
16
t
t
cis cis i
( )
3
2
2
3
2
2 2
16 16
t
t
t cis cis = |
.
|

\
|
+ =

|
.
|

\
|
+ = i
3
2
sin
3
2
cos 2
16
t t

( ) i i 3 1 2
2
3
2
1
2
15 16
+ =
|
|
.
|

\
|
+

=
Example: 2. Use De Moivre's Formula to show that
(a) u u u u
2 3
sin cos 3 cos 3 cos =
(b) u u u u
3 2
sin sin cos 3 3 sin =
(c) u u u u u
2 2 2 4
sin sin cos 6 cos 4 cos + =
(d) u u u u u
3 3
sin cos 4 sin cos 4 4 sin =


DEFINITION: A complex number z is called the n
th
root of a complex number w for
N e n if w z
n
= and we write .
n
w z =
Let ( ) i z | | sin cos + = be the n
th
root of u rcis w=
Then ( ) u | cis r n cis w z
n n
= =
r
n
= and . , 2 Z k k n e + = t u |

n
r = and . ,
2
Z k
n
k
e
+
=
t u
|
Hence Z k
n
k
n
e + = ,
2t u
| but these values of | yield the same value of z for any two
integers that differ by a multiple of n . Therefore, there are n-distinct roots when , 0 = z
namely
1 , , 1 , 0 ,
2
sin
2
cos =
(

|
.
|

\
| +
+ |
.
|

\
| +
= n k i
n
k
n
k
r z
n
k
t u t u

Thus, the n roots are
|
.
|

\
|
=
n
cis r z
n
u
0

|
.
|

\
| +
=
n
cis r z
n
t u 2
1

.
.
.
( )
|
.
|

\
| +
=

n
n
cis r z
n
n
t u 1 2
1







Geometrically the length of each of the n vectors
n
z
1
is the number .
n
r The argument
of one of those vectors is
n
u
and the other arguments are obtained by adding multiples
of .
2
n
to
n
u t

n
t 2
n
t 2
n
u
z
n-1
z
0
z
1
z
2
n
r


Example: Find the fourth roots of . 1 i +
Solution: Since ( ) , 4
4
sin
4
cos 2 1 = + = + n and i i
t t

3 , 2 , 1 , 0 ,
4
2
16
sin
4
2
16
cos 2 1
8
1
4
=
(

|
.
|

\
|
+ + |
.
|

\
|
+ = + k i
k k
i
t t t t


Therefore; the fourth roots of i + 1 are
(

+ =
16
sin
16
cos 2
8
1
0
t t
z
(

+ =
16
9
sin
16
9
cos 2
8
1
1
t t
i z
(

+ =
16
17
sin
16
17
cos 2
8
1
2
t t
i z
(

+ =
16
25
sin
16
25
cos 2
8
1
3
t t
i z



DEFINITION: A complex number z is called the n
th
root of unity if . 1 =
n
z
0 sin 0 cos 1 i z z
n n
+ = =
1 , , 1 , 0 ,
2
sin
2
cos = + = n k
n
k
i
n
k
z
k
t t
are n
th
roots of unity.
If
|
.
|

\
|
+ |
.
|

\
|
=
n
i
n
w
t t 2
sin
2
cos for 1 = k then by De Moivre's formula the roots of unity are
given by
1 2
, , , , 1


n
w w w . Since , 1 = z the n
th
root of unity are vertices of the n-sided
regular polygon inscribed in the unit circle.

Example 1: Find the third roots of unity
Solution: The cube roots of unity are
1 0 sin 0 cos
0
= + = i z
i i z
2
3
2
1
3
2
sin
3
2
cos
1
+

= + =
t t

i i z
2
3
2
1
3
4
sin
3
4
cos
2

= + =
t t



z
0
z
1
z
2
Real zxis
-1
1
1
-1
Imaginary axis




Example 2: If ( ) u u sin cos
2
i r w z + = = then
1 , 0 ,
2
2
sin
2
2
cos =
(

|
.
|

\
| +
+ |
.
|

\
| +
= k
k
i
k
r z
t u t u

1 , 0 ,
2
sin
2
cos =
(

|
.
|

\
|
+ + |
.
|

\
|
+ = k k i k r t
u
t
u

(

+ =
2
sin
2
cos
u u
i r
But
2
cos 1
2
cos
u u +
= and ,
2
cos 1
2
sin
u u
= where the sign infront of the radical
depends on the quadrant in which
2
u
lies.
Example 3: Solve . 0 1
2
z for i z i z = + +
Solution: By the Quadratic Formula
( )
2
4 3
1 2
1 1 4
2
2 , 1
i i i i i
z

=

+
=
Since 6 . 0 cos ,
5
4
5
3
5 4 3 = |
.
|

\
|
= = u i i w and , 8 0 sin = u and hence u lies in the 4
th

quadrant so that
2
u
lies in the second quadrant.
;
5
2
2
5
3
1
2
cos 1
2
cos

=
+
=
+
=
u u
and

5
1
2
5
3
1
2
cos 1
2
sin =

+ =
u u

Hence
(

+ =
2
sin
2
cos 5 4 3
u u
i i
| | i i + =
(

= 2
5
1
5
2
5


Take only , 2 i + then
( )
1
2
2
2
2
1 2 , 1
=
+ +
=
+
=
i i
z
i i
z and i
i i
z =
+
= 1
2
2
2


Therefore; the solution set is{ } i 1 , 1

Note: If is any n
th
root of a complex number
0
z , then , , , , , , ,
1 2

n k
zw zw zw zw z where
w is the n
th
root of unity when k =1, that is, ,
2
sin
2
cos
n
i
n
w
t t
+ = are the n
th
roots of
0
z .
Multiplying z by
k
w corresponds to increasing the argument of z by .
2
n
k t


Now let us define
n
m
z for ( ) ( ), sin cos 1 , , u u m i r z If n m cd and n m + = = N e then
( ) u u m i m r z
m m
sin cos + =
and ( )
n
m
z
1
has n distinct roots
( ) 1 ,...., 1 , 0 ,
2
sin
2
cos
1
= |
.
|

\
|
+ +
(

|
.
|

\
|
+ = n h
n
h
n
m
i
n
h
n
m
r z
n
m
n
m
t u t u


h
n
m
w
n
m
i
n
m
r . sin cos |
.
|

\
|
+ =
u u

( ) 1 ...., 1 , 0 ... sin cos
1
=
)
`

|
.
|

\
|
+ = |
.
|

\
|
n k w
n
i
n
r z
m
k
m
n
m
n
u u

These two set of numbers are the same if the sets
h
w and
km
w coincides when both h
and k range the values . 1 ,..., 2 , 1 , 0 n

a) We first show
km
w has n distinct values. If some of its values correspond to two
distinct values k of k and k
11 1
are the same then the two part sets
m k
w
1
and
m k
w
11

coincide and there is N pe such that,



P
n
m k
n
m
n
m k
n
m
t
t u t u
2
2 2
11 1
=
|
|
.
|

\
|
+
|
|
.
|

\
|
+
and thus ( ) .
11 1
P
n
m
k k =
Since p and
11 1
k k must be divisible by n . Contradiction as n k k <
11 1
.
Hence
km
w has exactly n distinct values.
b) We next show that for each fixed value of k the number
km
w is one of the n distinct
numbers
h
w , so that the two set of numbers are identical.
Let { } 1 ,..., 2 , 1 , 0
0
e n k be fixed. Then either .
0 0
n mk or n mk > <
(i) If n mk <
0
, we choose
0
mk h = . Therefore, the values corresponding to
0
k in (**) is
the same as the value
0
mk (*).
(ii) If n mk >
0
, then by Division Algorithm
r qn mk + =
0
where { }. 1 ,...., 2 , 1 , 0 e n r So .
r r qn mk
w w w = =
+
In this case we choose
r h = .
Hence we conclude that the two set of numbers in (*) and (**) are the same.
Thus, the n-numbers in either set can be written as
( ) ( )
(

|
.
|

\
|
+ + |
.
|

\
|
+ = = k
n
m
i k
n
m
r z
n m
n
m
t u t u 2 sin 2 cos
. 1 ,..., 2 , 1 , 0 = n k
Similarly, ( )
n
m
m
n n
m
z z z
1 1

= |
.
|

\
|
=
Example: Find all values of ( )
2
3
3 1 i +
Solution: ( )
3
2
sin
3
2
cos 2 3 1
t t
i i + = +
( )
(

|
.
|

\
|
+ + |
.
|

\
|
+ = + k i k i t
t
t
t
2
3
2
2
3
sin 2
3
2
2
3
cos 8 3 1
2
3
1 , 0 = k




, o k if = then
( )
(

+ |
.
|

\
|
=
3
2
2
3
sin
3
2
2
3
cos 8
1
t
t
i z
| | . 2 2 8 sin cos 8 = = + = t t i

If 1 = k , then

(

|
.
|

\
|
+ |
.
|

\
|
+ |
.
|

\
|
+ = t
t
t
t
2
3
2
2
3
sin 2
3
2
2
3
cos 8
2
i z
| | t t 4 sin 4 cos 8 i + =
| | 2 2 8 0 1 8 = = + i

Regions in the Complex Plane

1. A collection of points in the z-plane is called set of points or a point set.
2. Let e.
0
z and . 0 > o The nbhd o of
0
z , denoted by ( ) { }
0 0
z z z z B = o
The circular disk centered at
0
z and radius o .

0
z
o
Real axis
Imaginary axis
( ) o
0
z B
0
z
o
Real axis
Imaginary axis
( ) o ,
0
*
z B








3. The deleted nbhd o of
0
z , denoted by ( ) o ,
0
z B
-
is the set of all e. z such that
. 0
0
o < < z z ( ) { } o o < < =
0 0
*
0 , z z z z B
4. A point
0
z is called a limit point or a cluster point or an accumulation point if
every deleted nbhd o of
0
z contains points of S different from
0
z .
Note that a limit point of S may or may not be element of S .
Example: 0 is the limit point of .
1
)
`

N e = n
n
S
i is a limit point of { }. 1 s = z z S
i 2 1+ is a limit point of { }. 3 < = z z S
5. If every limit point of a set S belongs to S , we say that the set is closed.
6. A limit point
0
z of S is said to be an interior point of S if ( ) o
0
z B nbhd a - such
that ( ) .
0
S z B _ o
A limit point
0
z of S such that every nbhd of
0
z contains both points of S and
points which don't belong to S is called a boundary point.


o
0
z
P
w
R


If every point of a set S is an interior point we say that S is an open set.

7. A set of points S is said to be bounded if and only if there is a positive number M
such that M z < for all . S z e Otherwise, S is called unbounded.


8. A set of points S is said to be connected if any two of its points can be joined by a
continuous curve all of whose points belongs to S .

-
0
z is interior point of S .
- . S P and S R e e Every nbhd of wcontains
both points of S and points not on S . Hence w
is a boundary point

S
1
S
3
P
Q
R O
S
X
S
2



1
S and
2
S are connected but
3
S is not.
9. An open connected set is called a domain or an open region.
10. If we add to any set S those limit points which are not already in S , we obtain
closed set called the closure of S , often denoted by S .

11. The closure of an open region is called a closed region.
12. If to an open region we add all, or some or none of its limit points we obtain a set
called REGION.

13. A region which is both closed and bounded is called compact set.


Example: describe the set of points such that
(a) ( ) { } 0 Re > z z
(b) ( ) ( ) { } 1 Im Re s + z z z
(c) { } 3 2 = i z z
(d) { } 3 2 < s i z z
(e) { } 1 2 1 < + v < i z i z z



Solution:
(a) ( ) . 0 0 Re iy x z if x z + = > >

Open
Unbounded
Connected
R.A
I.A


(b) ( ) ( ) yi x z if y x z z + = s + s + 1 1 Im Re

R.A
I.A
1
-1
i
-i
closed
bounded
connected
compact

(c) ( ) 3 2 3 2 3 2
2 2
= + = + = y x i yi x i z
R.A
I.A
closed
bounded
2
4
6
-2i
-4i
-2
-4
-6
2i
4i
6i


(d) ( ) iy x z if y x i z + = < + s < s 3 1 2 3 2
2 2


R.A
I.A
Neither open nor closed
connected
2
4
6
-2i
-4i
-2
-4
2i
4i
6i


(e) ( ) ( ) 1 2 1 1 1 2 1
2 2 2
< + + v < + < + v < y x y x i z i z



Open
bounded
1 2 3 4
-i
-2i
-3i
i
2i
3i
-1 -2 -3
Not connected








MaEd 461 Functions of Complex Variables
Worksheet I

1. Write each of the following in the form bi a + where a and b are real numbers.
a)
i
i
i +

1 2 3
13
3

b)
( ) ( ) ( ) i i i 3 2 1
5


c)
31 94 120
3 4 i i i +
d)
i
i
i
i
5
2
4 3
2 1
+

+

2. If i z i z 2 3 , 2
2 1
= + = and ,
2
3
2
1
3
i z +

= then find
a)
3
2
2
3
1
4 3 z z z +
b)
4
3
z
c)
3 2 1
4 3 z z z +
3. If
3 2 1
, , z z z are complex numbers prove that
a)
1 3 3
.z z z z
i
=
b) ( ) ( )
3 2 1 3 2 1
z z z z z z + + = + +
c) ( ) ( )
3 2 1 3 2 1
. . . . z z z z z z =
d) ( )
3 1 2 1 3 2 1
. . . z z z z z z z + = +
4. Prove that
a) | |
2 2 2 2
2 w z w z w z + = + +
b) ( ) ( ) z z z 2 Im Re s +


5. Express the following in polar form
a) i 3 2 2 +
b) i 2 2 2 2
c) i 2 6
d) i
2
3
2
3

6. Evaluate using polar form
a) ( )
10
2 2 i b) ( )
8
27 3 i
c) ( )
2
1
2 3 2 i d) ( )
5
1
4 4 i +
7. Solve
a) i z + = 1
3
d) 0 3 2
2
= + + z z
b) 8
4
= z e) ( ) ( ) 0 7 1 2
2
= + + + i z i z
c) i z 8
3
= f) 0 2
2 4
= + + iz z
8. Find the square roots of
a) i 12 5 b) i 5 4 8 +
9. Find the sixth roots of unity and show the roots graphically
10. Express u 6 cos and u 6 sin interms of u sin and u cos .
11. Find the locus of points in the z-plane satisfying
a) 2 < + i z d) 3 2 3 = + z z
b) ( ) 2 Im = iz e) 8 2 2 = + + z z
c) ( ) 0 2 Re > zi f) ( )
4
arg
4
t t
< s

z
12. Express each equation interms of conjugate coordinates
a) 5 2 = + y x b) 36
2 2
= + y x
13. Prove that the equation of any circle or straight line in z-plane can be written in the
form 0 . = + + + r z z z z x | |
Were x and r are real numbers, and | is a complex number.




CHAPTER 2: ANALYTIC FUNCTIONS

Introduction: In the previous section chapter 1, we defined complex
numbers and also considered the algebra of complex numbers. In this
section, we shall introduce the concepts of Limit and Contiunity which
underlie Analysis, and the notion of Differentiability applicable to the
complex valued functions of complex variables. It will be seen that the
notion of limit and continuity are development of the notion of
neighbourhood of a point, and the notion of differentiability constitutes a
departure from that of continuity in as much as the differentiability of a
function of a complex variable is not just equivalent to differentiability of
the real and imaginary parts thereof as is the case with continuity. This fact
leads to Cauchy-Riemann Equations and their important consequences. Far
reaching properties of analytic functions can only be arrived at with the help
of the Cauchys Integral Formula which is considered in the next chapter,
(Chapter 4).

2.1. Functions of a Complex Variable
A symbol such as z which can stand for any one of the numbers of a set S of complex
numbers is called a complex variable.

Let S be a set of points in the Z-plane. If to each value of z in S there corresponds a
unique value of a complex variable w we say that w is a function of a complex variable
z on S and write ) (z f w= or ) (z w = or ( ), z s w= etc. The set S is called the domain
of the function and the quantity ( ) z f is called the value of f at z or the image of z
under f .
Example: The following are functions of a complex variable z where . iy x z + =
a)
2
z z w + = d) ( )
2 2
y x i xy w + =
b)
z
z
w

=
1
e)
|
|
.
|

\
|

+
+
=
iy x
i y
y x
w
1
2
2 2

c) z i
z
z w + =
1

As in the real analysis, algebraic operations with functions of complex variables are
defined by corresponding operations on function values when functions are written in the
form f(z) and ( ) z } then the sum, the product, the quotient and composite are, respectively,
( ) ( ) ( ) ( )
( )
( ) z
z f
z z f z z f
}
} } + , . , and ( ) ( ) z f }
Example: if ( ) i z z z f + + = 1
2
and ( ) , 3 i z z + = } then
a) ( )( ) ( ) ( ) 1 4
2
+ + = } + = } + z z z z f z f
b) ( )( ) ( ) ( ) ( )( ) i z i z z z z f z f + + + = } = } 3 1 .
2

( ) ( ) ( ) i z i i z
z
+ + + + + = 1 2 3 3 3
2
3


c) ( ) ( ) ( ) ( )
0 f g z f g z =
( ) ( ) ( ) i i z i z i z f + + + + = + = 1 3 3 3
2

i i z z i gz + + + + = 2 3 1 6
2

( )i i gz 6 3
2
+ + =
By separating into real part and imaginary part any function ( ) z f can be expressed
interms of two real functions, that is in the form
( ) ( ) ( ) y x i y x u z f , , v + =
Similarly, in polar coordinates with ( ) u u sin cos i r z + =
( ) ( ) ( ) u u , , r i r u z f v + =

Example:
1. If ,
2
z w= then ( )
2 2
, y x y x u = and ( ) y y x = v 2 ,
( ) ( ) ( )i y y x iy x x w + = + = = 2
2 2 2 2

In polar coordinates
( ) ( ) u u u u 2 sin 2 cos 2 sin 2 cos
2 2 2 2
r i r i r z w + = + = = and thus
( ) ( ) u u u u 2 sin , 2 cos ,
2 2
r r and r r u = v =
2. If ,
1 z
z

then

( ) ( )
2 2 2 2
2 2
2 1 2 1 1 1 y x x
y
i
y x x
y x x
iy x
iy x
iy x
iy x
w
+ +
+
+ +

=

+
=
+

=
( ) ( )
2 2 2 2
2 2
2 1
,
2 1
,
y x x
y
y x and
y x x
y x x
y x u
+ +
= v
+ +

=

Graphical Representation of w = f(z)

By definition for each value of the independent variable iy x z + = ( in the domain of f )
the function yields a unique value iv u w + = of the dependent variable. Each of these
variables has two dimensions and their combined configuration is a four dimensional
entity that is rather difficult to graph. Because of this difficulty the graphical
representation of a complex function I effected by employing tow copies of the complex
plane labeled z plane and w plane. If ( ) z f w= is a function for every yi x z + = of its
domain in the z plane we calculate the corresponding iv u w + = and we locate if in the
w plane. Repetition of the same process of some set S in the domain of f which
yield "the image of S under f " in the w plane. The correspondence of points in z
plane into the w plane by is called mapping or transformation






Example:
1. Consider, 2 + = z w . This mapping is a translation by 2 units to the right.
Use two planes one for w and the other for z and hence we get each point in the z-plane
is translated by 2 units to the write. (Show on the plane: exercise)

2. If , ) ( z z f = then, f is a reflection of z with respect to the real axis. Thus the image of
every point in z-plane is the reflection of the point with respect to the real axis.
(Show on the plane: exercise)

2.2. Limits
At this level everybody knows how to find the limits and prove some
properties of limits of the case of real-valued functions of single variable and
real-valued functions of several variables. In this section we extend the
notion of limits to complex variable.

DEFINITION: Let f be a complex function with domain D. We say that the complex
number.
0
w written
( )
0
0
lim
z z
w z f

=

iff for all ( ) ( ) ( ). *
0 0
w B z f z B B
e
e e

Restatement
( )
0
0
0 lim
z z
w z f

> - = o
such that for any D z e
and , 0
0
o < < z z we have ( ) <e
0
w z f








Graphically


R.A
I.A
z
0
o
R.A
I.A
-
w
0
f(z)



Example 1: Show that
i z
i z
+
+ =
2
2 lim

Solution: Let 0 > - be given. We need to show that there exists 0 > o such that
( ) ( ) . 2 2 0 <e + < + < i z i Z o
Choose se o . Then
( ) ( ) ( ) . 2 2 2 0 <e + + < + < i i z i z o
i z
i z
+
+ =
2
2 lim

Example 2: Show that . 2
1
lim
2
i
i z
z
=

+

Solution: Let 0 e> . We observe that

( )( )
. 2 2 2
1
2
i z if i z i i z i
i z
i z i z
i
i z
z
= = + =

+
=

+

Choose se o . Then
e <

+
e < < < i
i z
z
i z i z 3
1
0
2
o


Example 3: Show that
( ) ( )
0
0
Im Re
lim

=
z
z
z z

Solution: Let 0 > e be given. We know that
( ) ( ) z z and z z s s Im Re

( ) ( )
0 1
Im
1
Re
= s s z if
z
z
and
z
z

So
( ) ( ) ( )
( ) ( ) e < s s = z z z
z
z
z
z z
Im Im .
Re Im . Re

Choose e s o . Then

( ) ( )
. 0
Im . Re
0 0 e < e < < <
z
z z
z z o

( ) ( )
0
0
Im . Re
lim

=
z
z
z z

Example 4 . Let show that
0
0
2 2
lim
z z
z z

=

Solution: Let 0 > e . Since ,
0 0
0
2 2
z z z z z z + = assume that 1
0
< z z . Then

0 0 0 0 0 0
2 1 2 2 z z z z z z z z z + < + s + = = +
( ) e < + < + =
0 0 0 0
0
2 2
2 1 z z z z z z z z z

0
0
2 1 z
z z
+
e
< .
Choose
)
`

+
e
=
0
2 1
, 1 min
z
o . Then
. 0 0
2 2
0
e < < < z z z z o
Exercise:
1. Show that
( )
i z
i yi x
2
6 3 2 lim

= +

2. Show that
( )
i z
i yi x
+
+ = +
1
2 3 2 3 lim
2

Theorem: (Uniqueness of Limit)

If a function f has a limit at C z e
0
then it is unique, that is, if
( )
0
0
lim
z z
w z f

=
and
( )
0
1
lim
z z
w z f

=
then
. 1 0
w w =
Proof: Since
( )
0
0
lim
z z
w z f

=
and
( )
0
1
lim
z z
w z f

=
, for 0 > e there exist
1
o and
2
o such
that
( ) ,
2
0
0 1 0
e
< < < w z f z z o and
( )
. 2
0
1 2 0
e
< < < w z f z z o

Choose { }
2 1
, min o o o = . Then if o < <
0
0 z z z then
( )
2
0
e
< w z f and ( ) .
2
2
e
< w z f
( ) ( )
1 0 1 0
w z f z f w w w + =
( ) ( )
1 0
w z f w z f + s
=e
e
+
e
<
2 2


Sinceeis arbitrary we have . 0
0 1 1 0
w w w w = = Hence the limit of a function is
unique.

Remark: The definition of limit does not specify the direction form which z must
approach
0
z . It requires that if a limit is to exist its value must be independent of the
direction of approach. This fact is useful in proving that a limit doesn't exist.



Example 5: Show that
0
lim
z
z
z
does not exist.
Solution: Let yi x z + = . Then iy x z =
i
y x
xy
y x
y x
yi x
yi x
z
z
2 2 2 2
2 2
2
+

=
+

=
If 0 z along the real axis, 0 = y ,

0 0
1 lim lim
2
2

= =
x z
x
x
z
z

If 0 z along the imaginary axis, 0 = y ,
0 0
. 1 lim lim
2
2

=

=
y z
y
y
z
z

Hence
0
lim
z
z
z
does not exist

Theorem: Suppose that
(a) ( ) ( ) ( ) y x iv y x u z f , , + = has domain D
(b)
0 0 0
iy x z + = is in D or on boundary of D
Then
( ) ( ) ( )
( ) ( ) ( ) ( )
0 0 0 0 0
0 0 0 0
, ,
, lim , lim lim
y x y x y x y x z z
v y x v and u y x u iff iv u z f

= = + =

Proof:
( ) Assume that
( )
0
0 0
. lim
z z
iv u z f

=
+

Let 0 > e be given. Then 0 > -o such that
( ) ( ) . 0
0 0 0
e < + < < iv u z f z z o


But since ( ) ( ) ( ) ( )
0 0 0 0 0 0
iv u z f v v and iv u z f u u + s + s
We have ( ) ( ) e < e <
0 0
, , v y x v and u y x u
Wherever ( ) ( ) . 0
2
0
2
2
0
2
o < + < y y x x

( ) ( )
( ) ( ) ( ) ( )
0 0 0 0
0 0
, , , ,
, lim , lim
y x y x y x y x
v y x v and u y x u

= =

( ) : Assume that
( ) ( )
( ) ( ) ( ) ( )
0 0 0 0
0 0
, , , ,
, lim , lim
y x y x y x y x
v y x v and u y x u

= =

Let 0 > e be given. Then 0 ,
2 1
> - o o such that
( ) ( ) ( )
, 2
, 0
0 1
2
0
2
0
e
< < + < u y x u y y x x o and
( ) ( ) ( )
, 2
, 0
0 2
2
0
2
0
e
< < + < v y x v y y x x o
Take { }
2 1
, min o o o = . Then
2 2
2 0
e
<
e
. < v v u u
Wherever ( ) ( ) . 0
2
0
2
0
o < + < y y x x
o < <
0
0 z z
( ) ( ) ( ) ( )
0 0 0 0
v v i u u iv u z f + = +
e =
e
+
e
< + s
2 2
0 0
v v u u

( )
0
0 0
. lim
z z
iv u z f

+ =

Example: 6. Evaluate
( ) | |
i z
i xy y x y x

+
1
5 2 lim
2 3 2

Solution: Since ( ) ( ) ( ) ( ) 1 1 5 , , 2 ,
0 0
2 3 2
= = = y x and xy y x y x v y x y x u
( ) ( )
( ) ( ) ( ) ( ) 1 , 1 , 1 , 1 ,
6 1 5 , lim 2 ,

= + = =
y x y x
y x v and y x u

( ) | |
( ) 1 1
6 2 5 2 lim
2 3 2

+ = +
z
i i xy y x y x


Theorem: If
( ) ( )
0 0
2 1
lim lim
z z z z
w z and w z f

= = o
then
(a)
( )( )
0
2 1
lim
z z
w w z f

+ = + o
. . . . Sum Rule
(b)
( )( )
0
2 1
lim
z z
w w z f

= o
. . . . Difference Rule
Proof: Let 0 > e be given. Then 0 ,
2 1
> - o o such that
( ) ,
2
0
1 1 0
e
< < < w z f z z o
( ) ,
2
0
2 2 0
e
< < < w z f z z o
Choose { }. , min
2 1
o o o =
( )
2
0
2 0
e
< < < w z f z z o
( ) ( ) ( )
2 1
w w z z f + + o
( ) ( )
2 1
w z w z f + s
=e
e
+
e
<
2 2


( ) ( ) | |
0
2 1
lim
z z
w w z z f

+ = + o

The proof of (b) is similar to (a)

Theorem: If
( ) ( )
0 0
2 1
lim lim
z z z z
w z and w z f

= =
then

( ) ( )
0
2 1
lim
z z
w w z z f

=
. . . . Product Rule
Proof: Assume that ( ) ( ) ( ) ( ) ( ) ( ) y x it y x s z y x iv y x u z f , , , , , + = + =
and
0 0 0
iy x z + = . Then
0 0 2 0 0 1
it s w and iv u w + = + =
Where
( ) ( ) ( )
( ) ( ) ( ) ( ) ( ) ( )
0 0 0 0 0 0
0 0 0
, , , , , ,
, lim , , lim , , lim
y x y x y x y x y x y x
y x s s y x v v y x u u

= = =


and to
( )
( ) ( )
0 0
, ,
, lim
y x y x
y x t

Since ( ) ( ) ( ) ( ), vt us i vt us z z f + = by product rule for limit of two variable functions
we have
( )( ) ( )( )
( ) ( ) ( ) ( )
0 0 0 0
0 0 0 0 0 0 0 0
, ,
, lim , lim
y x y x y x y x
v s t u y x vs ut and t v s u y x vt us

+ = + =

( ) ( ) ( ) ( )
0
0 0 0 0 0 0 0 0
lim
z z
s v t u i t v s u z z f

+ + = }

( ) ( )
2 1 0 0 0 0
. w w t i s i v u = +
Lemma: If
( )
0
2
, 0 lim
z z
w z

= =
then 0 > -o such that
( ) .
2
0
2
0
w
z z z > < < o
Proof: Since
( )
0
2
0 , lim
z z
w z

> - = o
such that
( ) .
2
0
2
2 0
w
w z z z < < < o
Writing ( ) ( ),
2 2
z z w w + = we have
( ) ( ) ( ) o < < + < + s
0
2
2 2
0
2
z z if z
w
z z w w
( ) .
2
0
2
0
w
z z z > < < o
Theorem: If
( )
0
, 0 lim
z z
w z

= =
then ( )
0
1 1
lim
z z
w z

=

Proof: Since
( )
0
1
0 , 0 lim
z z
w z

> - = = o
such that
( ) 0
2
1
0
2
1 0
e> e < < < for w w z z z o



by the lemma above 0
2
> -o such that
( )
2
0
2 0
w
z z z > < < o o
Taking { }, , min
2 1
o o o = we obtain

( )
( )
( )
=e
e
<

= < <
w
w
w
w z
w z
w z
z z
2
2
1
1 1
0
2
0

o

( ) w z
1 1
lim =

Theorem: (Quotient Rule). If
( ) ( )
0 0
2 1
0 lim lim
z z z z
w z and w z f

= = =

Then
( )
( )
0
2
1
lim
z z
w
w
z
z f

=

Proof:
( )
( )
( )
( )
( )
( )
0 0 0 0
1
lim . lim
1
. lim lim
z z z z z z z z
z
z f
z
z f
z
z f

=
(

=


2
1
2
1
1
.
w
w
w
w =

Remark: By induction it follows that
1.
( ) ( ) ( ) ( )
0
2 1 2 1
, . . . . . .. lim
z z
and w w w z f z f z f
n n

+ + + = + + +

2.
( ) ( ) ( )
0
2 1 2 1
, . . . . . . .. . lim
z z
w w w z f z f z f
n n

+ = +

Provided that
( )
0
lim
z z
w z f
i i

=





Exercise: Evaluate the following limits.
(a)
( )
i z
i z z i
2
10 5 lim
2 4

+

(b)
( ) ( ) ( )
i z
i z i z i z
3 1
3 . 5 3 lim
2 2

+

(c)
( )( )
( )
2
1
4 3 2
lim

+
z i
i z z

- Limit of a Polynomial Function
If ( ) , 0 , . . .
1
= + + + =
n b
n
n
a a z a z a z P is a polynomial of degree n, then
( ) ( )
0
0 0
lim
z z
z z P z P

e. =

- Limit of Rational Function
If
( )
( )
( )
( ) ( ) ( )
0
0 0
lim , 0
z z
z R z R then z Q and
z Q
z P
z R

= = =

2.3. CONTINUITY
Definition: Let ( ) z f be a function on . D Then ( ) z f is said to be continuous at
*
0
D z e (interior of D) iff 0 0 > - > e such that
( ) ( ) .
0 0
e < < z f z f z z o
Restatement: f is said to be continuous at
*
0
D z e iff the following conditions must be
satisfied.
(i) ( )
0
z f is defined ;
(ii)
( )
0
lim
z z
z f

exists; and
(iii)
( ) ( )
0
0
lim
z z
z f z f

=


Example: 1. Let, ( )

=
=
+

=
i if
i z if
z
i z
z f
2
1
1
2
show that f is continuous at i z =
0

Solution: (i) ( )
2
1
= i f It is defined at
0
z
(ii)
( )
i z i z i z
i
i i z z
i z
z f

= =
+
=
+

=
2 2
1 1
lim
1
lim lim
2

(iii)
( )
i z
i
z f

=
2
lim

f is continuous at .
0
i z =

Definition: 1. A function f is said to be continuous on a set S if it is continuous at
every point S .
2. f is said to be continuous if ti is continuous on its domain

Theorem: Suppose that
(i) ( ) ( ) ( ), , , y x v i y x u z f + =
(ii) ( ) z f is defined at every point of D,
(iii) bi a z + =
0
is an interior point of D.
Then f is continuous at
( ) ( )
( ) ( ) b a y x
b a u y x u iff z
, ,
, , lim
0

=
and
( ) ( )
( ) ( ) b a y x
b a v y x v
, ,
. , , lim

=


Example:
1. ( ) ( )
4 3 2
3 x xy i y x z f + = is continuous because the polynomials
3 2
y x and
4
3 x xy are continuous


2. ( )
y x
e i xy z
+
+ =
2
cos is continuous because xy cos and
y x
e
+ 2
are continuous
every where.

Theorem: If ( ) z f and ( ) z are continuous at
0
z , then
1. ( ) ( ) z z f is continuous at .
0
z
2. ( ) ( ) z z f . is continuous at .
0
z
3.
( )
( ) z
z f

is continuous at
0
z provided ( ) . 0
0
= z
4. ( ) ( ) z f is continuous at
0
z provided ( ) z f is continuous at ( )
0
z
Proof: The proof of this theorem is identical with the proof of the corresponding
theorem for real functions. It is left as exercise.

Example:
1. Every polynomial functions is continuous on . .
2. Every rational function is continuous on its domain.
3. ( )( ) y i y e z f
x
sin cos + = is continuous on the entire complex plane.
4. ( ) x hy i x hy x
e e
i x
e e
z f
y y y y
sin sin cos cos sin
2
cos
2
=

+
=

is continuous
on the entire plane.
5. ( ) , ln u i r z f + = where , u cis r z = is continuous on the entire plane except . 0 = z

Exercise:
Show that, the function ( ) , , sin cos cos . sin iy x z where y x i y hx z f + = + = is continuous
on the entire plane.






MaEd 461 Worksheet II

1. Decompose each of the given functions first in the form ) , ( ) , ( y x iv y x u + and then in
the form ( ) ( ) u u , , r iv r u +
a) ( ) z z z f 3
2
+ = c) ( )
|
.
|

\
|
+ =
z
i
z i z f Im
b) ( )
z
z
z f
+

=
1
1
d) ( )
3
z z f =
2. Describe the image of the rectangle with vertices 0 , 2 , 1 i and 1 2 + i under the
transformation
a) ( ) z w Re = c) iz w=
c) ( ) z w Im = d) z i z w + =
3. Using the definition of limit show that
a)
( ) | |
i z
i y i

+ = +
3
8 2 2 5 lim
2

b)
( ) ( ) | |
i z
i xy y i y x
2
8 2 2 lim
2 2

+ = + +

c)
( )
i z
i z z

= + 1 2 2 lim
2

d)
0
0 lim
2

=
z
z
x

4. Evaluate the following limits.
a)
( )
i z
i z z i
2
10 3 lim
2 4

+ +
d)
0
3
lim
2
2 3

z
z
z z

b)
( ) ( )
( )
2
1
1
4 3 2
lim
2

+
z
z i
i z z
e)
i z
z z
z
4
1
lim
4
1

+ +

c)
i z
z
z

|
|
.
|

\
|
+
+
1
1
lim
6
2
f)
i z
i z
i z

1 1
lim
5. Show that the following limits does not exist.
a)
0
2
lim
2
2
2 2

+
+
z
x
y
i
y x
xy
b)
0
lim
2 4
2

+
z
i
y x
y x




















2.4. DERIVATIVES

DEFINITION: Let ( ) z f w= be a complex function with domain D and
0
z is in the
interior of . If z z z A + =
0
be any other point in D , then f is said to be differentiable at
0
z if
( ) ( ) ( ) ( )
0 0
0 0
0
0
z z z z
z
z z z f
lim or
z z
z f z f
lim

+

exists. If the limit exists then we

denote it by ( ),
0
1
z f that is,

( )
( ) ( ) ( ) ( )
0 0
0 0
0
0
0
1
lim lim
z z z z
z
z f z z f
z z
z f z f
z f
A
A
A +
=

=

Notation: ( )
0 0 0
1
z z
dz
df
or z z
dz
df
or z f = =
Example 1. Let ( ) , k z f = where k is a complex constant. Find ( )
0
1
z f for e.
0
z .
Solution: Let e.
0
z . Then

( ) ( )
0 0 0
0 0
0
0 0 lim lim lim
z z z z z z
z z
k k
z z
z f z f

= =

( ) e. = z z f 0
1

Example 2. Let ( )
2
z z = . Show that ( ) . 2
0 0 0
1
e. = z z z
Solution:
( )
( ) ( )
0 0
0
2
0
2
0
0
0
1
lim lim
z z z z
z z
z z
z z
z z
z

( )( )
( )
0 0
0 0
0
0 0
2 lim lim
z z z z
z z z
z z
z z x z

= + =

+
=






Example: Given ( ) ,
n
z z f = where n is a non-negative integer.
Show that ( )
1
0 0
1
=
n
z n z f for all .
0
e. z
Solution:
( )
0
0
0
0
1
lim
z z
z z
z z
z f
n n

=


( )( )
0
0
1
0
2
0
2
0
3
0
2 1
0
. . . . .
lim
z z
z z
z z z z z z z z z z
n n n n n

+ + + + +
=
+


0
1
1
0
lim
z z
z z
n
k
k k n

=

=



0
1
0
1
1
0
1 lim lim
z z
n z z z
n
k
n
k
k k n

= =

= =



1
0

=
n
z n
Example: If ( ) ,
1
z
z f = then show that ( ) . 0
1
0
2
0
0
1
= = z
z
z f
Solution: Let . e
0
z and 0
0
= z . Then

( )
( )
0 0
0
0
0
0
0
1
lim
1 1
lim
z z z z
z z z z
z z
z z
z z
z f

=


0
2
0
0 0 0
1 1 1
lim
z z
z
z z z z

=





Example: If ( ) ,
2
z z f = show that ( ) 0 0
1
= f .
Solution:
( )
0 0 0
2 2
1
. 0 lim
.
lim
0
0
lim 0
z z z z z z
z
z
z z
z
z
f

= = =

=

Example: If ( ) ,
i z
i z
z f

+
= find ( ). 1
1
i f +
Solution:
( )
( )
i z
i z
i
i z
i z
i f
+
+
+

+
= +
1
1
1
2 1
lim 1
1


( )( )
( ) ( ) ( )
i z
i z i z
i z i i z
+
+
+ +
=
1
1
2 1
lim

( ) ( ) ( ) ( )( )
i z i z
i z i z
i iz
i z i z
i i iz
+ +

+
=
+
+ +
=
1 1
1
2 2 2
lim
1
2 2
lim


( )
( )( )
i z i z
i
i z i z i z
i z i
+ +
=

=


=
1 1
2
2
lim
1
1 2
lim

Example: If ( ) , z z f = then show that f is not different on the entire complex plane.
Solution: Consider the limit for e.
0
z

( ) ( ) ( ) ( )
0
lim lim
0
0 0
0
0
A
A
A +
=

z z z
z
z f z z f
z z
z f z f


0 0
lim lim
0
0
A A
A
=
A
A +
=
z z
z
z
z
z z z

Since
0
lim
z
z
z
does not exist,
0
lim
A
A
A
z
z
z
does not exist. Therefore,
f is not differentiable at .
0
z

Example: Show that ( )
2
z z f = is not differentiable at . 0
0
= z
Solution:
( ) ( ) ( ) ( )
0 0 0
0 0
0
0
0
0
0
lim lim lim
z z z z z z
z
z f z z f
z z
z z z z
z
z f z f

A
A =
=

( )( )
0 0
lim lim 0
0
0
0
0
0
A A
|
|
.
|

\
|
A + +
A
A
=
A
A + A +
=
z z
z z
z
z
z
z
z z z z z z

Along the real axis, 0
0
0
0
lim z z z z
z
z
z + =
(

A + +
A
A

Along the imaginary axis, , 0 = Ax

0
lim 0
0
0
0
A
+ =
|
|
.
|

\
|
A + +
A
A
z
z z z z
z
z
z

and . 0
0
0
0
0
0
= + = + z z z z z Hence f is not differentiable at
. 0
0
= z
Theorem: If f is differentiable at
0
z then f is continuous at
0
z .
Proof: Suppose
( ) ( ) | |
( ) ( )
( )
0
0
0
0
0
lim
z z
z z
z z
z f z f
z f z f

=

( ) ( )
( )
0 0
0
0
0
lim . lim
z z z z
z z
z z
z f z f

=

( ) 0 0 .
0
1
= = z f

( ) ( )
0
0
lim
z z
z f z f

=

Therefore f is continuous at
0
z



Remark: The converse of the above Theorem is not necessarily true. For instance
2
z f is continuous on . but the derivative of f exists at 0 = z only.

Note: The definition of the derivative ( ) z f
1
is identical in form to that of the derivative
of real valued function of one variable. But the significance difference is that the limit
involved in ( ) z f
1
is a two dimensional one. Many results from calculus of real variable
don't carry over to the calculus of complex variable. For example, the function ( )
2
x x f =
has the derivative 9 e x x 2 but the derivative of ( )
2
x x f = exists only at 0 = z .

Theorem: If f and are differentiable at
0
z are differentiable at
0
z and k is a
complex constant, then . , , , f f f kf + and

f
are differentiable at
0
z , and their
derivatives are given by:
1. ( ) ( ) ( ).
0
1
0
1
z f k z kf =
2. ( ) ( ) ( ) ( )
0
1
0
1
0
1
z z f z f + = +
3. ( )( ) ( ) ( )
0 0
1
0
z z f z f =
4. ( ) ( ) ( ) ( ) ( ) ( )
0
1
0 0 0
1
0
1
. . . z z f z z f z f + =
5. ( )
( ) ( ) ( ) ( )
( ) ( )
2
0
0
1
0 0 0
1
0
1
.
z
z z f z z f
z
f

=
|
|
.
|

\
|
Provided ( ) . 0
0
= z
Theorem: (Chain Rule). If f is differentiable at ( )
0
z and if is differentiable at
0
z ,
then ( ) ( ) z f is differentiable at
0
z and
( ) ( ) ( ) ( ) ( ). .
0
1
0
1
0
1
z z f z f o =
Exercise: Find the derivates of the following.
(a) ( ) n
z
z f
n
,
1
= is a non-negative integer
(b) ( ) i
z
i
z i z z f 6
2
3
2 3
+ + =


(c) ( ) ( ) i z
z
z i z 2
1
2
2
3
+ |
.
|

\
|
+ =
(d) ( )
i z
z i z
z h
+

=
2
3

(e) ( )
5
2
2
6
|
.
|

\
|
+ = iz
z
i
z h

Complex Differentials: The concept of differential of a complex functions is identical
with that of the differential of real function. Let ( ) z f w= be differentiable at the point z
And let ( ) ( ) z f z z f w A + = A so that

( )
0
lim
1
A
=
A
A
z
z f
z
w

( ) z z z f w A e + A = A .
1
where 0 0 A e z as ( ) z f f A
1
is called the differential w
and is denoted by ( ) z f f dw A =
1
. In particular if , z w= we have dz dw = and
z z dz A = A = . 1 . Hence, ( )dz f f dw
1
= .
Example: If ( ) , 2 1
3
2
z i w + = then
( ) ( ) ( ) . 2 1 12 4 . 2 1 3
2
2
2
2 1
df z i z i dz z i z i dz z f dw + = + = =

2.5. CAUCHY-RIMANN EQUATIONS
Suppose that a function f has a derivative at
0
z
Let i y x z
0 0 0
+ =
( ) ( ) ( ) y x v i y x u z f , , + =
( ) i b a z f + =
0
1

( ) ( )
0 0
z f z z f f A + = A
( ) ( )
0 0 0 0
, y x u y y x x u u + A + A + = A



( ) ( )
0 0 0 0
, , y x v y y x x v v A A + = A
Then
0 0
lim lim
A A
+ =
A + A
A + A
=
A
A
z z
i b a
y i x
v i u
z
f


0 0
Im lim , Re lim
A A
=
|
|
.
|

\
|
A + A
A + A
=
|
|
.
|

\
|
A + A
A + A

y y
b
y i x
v i u
a
y i x
v i u


In particular, when , 0 = Ay that is when , x z A = A we have
( ) ( )
( )
0 0 0
,
2
2 , ,
lim lim Re lim
0 0
0 0 0 0
A A A
=
A
A +
=
A
A
= |
.
|

\
|
A
A + A
=
x x x
y x
x
u
x
y x u y x x u
x
u
x
v i u
a

( ) ( )
( )
0 0 0
,
, ,
lim lim Im lim
0 0
0 0 0 0
A A A
c
c
=
A
A +
=
A
A
= |
.
|

\
|
A
A + A
=
x x x
y x
x
v
x
y x v y x x v
x
v
x
v i u
b

( ) ( ) ( ) 1 . . . . . . . . . . . , ,
0 0 0 0
b y x
x
v
and a y x
x
u
=
c
c
=
c
c

Similarly, when , 0 = Ax that is, when , y i z A = A we have
( ) ( )
( )
0 0 0
,
, ,
lim lim Re lim
0 0
0 0 0 0
A A A
c
c
=
A
A +
=
A
A
=
|
|
.
|

\
|
A
A + A
=
y y y
y x
y
v
x
y x v y y x v
y
v
y i
v i u
a

( ) ( )
( )
0 0 0
,
, ,
lim lim Im lim
0 0
0 0 0 0
A A A
c
c
=
A
A +
=
A
A
=
|
|
.
|

\
|
A
A + A
=
y y y
y x
y
u
x
y x u y y x u
y
u
y i
v i u
b






( ) ( ) ( ) 2 . . . . . . . . . . . , ,
0 0 0 0
b y x
y
u
and a y x
y
v
=
c
c
=
c
c

Hence, ) , ( ) , ( ) , ( ) , ( ) (
0 0 0 0 0 0 0 0 0
'
y x
y
u
i y x
y
v
y x
x
v
i y x
x
u
z f
c
c

c
c
=
c
c
+
c
c
=
If f is differentiable at
0
z then
( ) ( )
0 0 0 0
, , y x
y
v
y x
x
x
c
c
=
c
c
Cauchy Riemann Equations
( ) ( )
0 0 0 0
, , y x
y
u
y x
x
v
c
c
=
c
c


Thus we have proved the following theorem.

Theorem 1: Let ( ) ( ) ( ) y x iv y x u z f , , + = . If f is differentiable at
0 0 0
y i x z = = then the
partial derivatives
y
v
and
x
v
y
u
x
u
c
c
c
c
c
c
c
c
, , exist at ( )
0 0
, y x and satisfy the Cauchy-Riemann
Equations. The derivative is given by
( ) ( ) ( ) ( ) ( )
0 0 0 0 0 0 0 0 0
1
, , , , y x
y
u
i y x
y
v
y x
x
v
i y x
x
u
z f
c
c

c
c
=
c
c
+
c
c
=
Example: Consider ( ) ( ) xy i y x z z f 2
2 2 2
+ = = . We have seen that f is differentiable at
every point C z e . Hence Cauchy Riemann Equations must be satisfied every where
( ) ( ) xy y x v and y x y x u 2 , ,
2 2
= =
x
y
v
and y
x
v
y
y
u
x
x
u
2 2 , 2 , 2 =
c
c
= =
c
c
=
c
c
=
c
c

( ) , 2 2 , 2
y
v
y y
x
v
and
y
v
x
x
u
c
c
= = =
c
c
c
c
=
c
c
and thus
( ) z iy x y i x
x
v
i
x
u
z f 2 ) ( 2 2 2
1
= + = + =
c
c
+
c
c
=



Remark: Theorem1 is a necessary condition for the existence of ( ),
1
z f that is, the
Cauchy- Reiemann Equations are satisfied at ( )
0 0
y x does not imply differentiability of
f at
0
z .
For example
( )
0 , 0
0 ,
2 2
=
=

=
z if
z if
y x
xy
z f
is not differentiable at 0 = z but the Cauchy-
Riemann Equations are satisfied at (0,0).

Sufficient Conditions:
Conditions on v and u that ensure the existence of ( )
0
1
z f are given in the following
theorem.

Theorem 2: Let ( ) ( ) ( )
0 0 0
, , y i x z and y x v i y x u z f + = + = .If
y
v
and
y
v
x
v
x
u
v u
c
c
c
c
c
c
c
c
, , , , are all continuous at ( )
0 0
, y x and if the Cauchy-Riemann
Equations are satisfied at ( )
0 0
, y x , then f is differentiable at
0
z and the derivative is
given by
( ) ( ) ( ) ( )
0 0 0 0 0 0 0
1
, , , y x
y
u
i y x
y
v
x
v
i y x
x
u
z f
c
c

c
c
=
c
c
+
c
c
=
Proof: Since
y
u
and
x
u
c
c
c
c
are continuous, we have ( ) ( )
0 0 0 0
, y x u y y x u u A + + = A
( ) ( ) | | ( ) ( )
0 0 0 0 0 0 0 0
, , , , y x u y y x u y y x u y y x x u A + + A + A + A + =
y
y
u
x
x
u
A
|
|
.
|

\
|
e +
c
c
+ A |
.
|

\
|
=e
c
c
=
2 1

,
2 1
y x y
y
u
x
x
u
A e + A e + A
c
c
+ A
c
c
= where
0
0 0 ,
2 1
A
A e e
y
x as


Since the Cauchy Riemann Equations are satisfied at ( )
0 0
y x , we can replace
x
u
by
y
v
and
x
v
by
y
u
c
c
c
c
c
c
c
c
to obtain the equation in the form
v i u f A + A = A
( ) ( ) y x y i x
x
v
i y i x
x
u
A + A + A + A
c
c
+ A + A
c
c
=
2 1
o o
Where 0 0 0 0
4 2 2 3 1 1
A . A e + =e e = e y x as i and i o o

z
y
z
x
x
x
i
x
u
z
f
A
A
+
A
A
+
c
c
+
c
c
=
A
A

2 1
o o
Since , z y and z x A s A A s A we have
( )
1 1 s
A
A
s
A
A
z
y
and
z
x

So that
0 0
lim 0 lim
2 1
A A
A
A
= =
A
A
z z
z
y
z
x
o o
. Therefore,
=
A
A
A
z
f
z
lim
0

x
v
i
x
u
c
c
+
c
c

Hence f is differentiable at
0
z and
( ) ( ) ( ) ( ) ( ). , , , ,
0 0 0 0 0 0 0 0 0
1
y x
y
u
i y x
y
v
y x
x
v
i y x
x
u
z f
c
c

c
c
=
c
c
+
c
c
=
Example2. Let ( ) y e i y e z f
x
sin cos
2
+ = . Find ( ).
1
z f
Solution: The functions , sin , cos , cos sin y e
x
v
y e
x
u
y e
y
v
and y e u
x x x x
=
c
c
=
c
c
=
c
c
=
y e
y
v
and y e
y
u
x x
cos sin =
c
c
=
c
c
all continuous at all points. It is easily seen
that the partial derivatives satisfy the Cauchy Riemann Equations every where. So f is
differentiable every where and
( )
y
u
i
y
v
x
v
i
x
u
z f
c
c

c
c
=
c
c
+
c
c
=
1

( ) z f y e i y e
x x
= + = sin cos



Example: Show that ( ) z z z f = is not differentiable at any points, C z e .
Solution: ( ) ( ) ( ) ( ) y i y i x y i x y i x z z z f 2 = + + + = =
( ) ( ) 2 , 0 , 0 , 0 , 2 , , 0 , =
c
c
=
c
c
=
c
c
=
c
c
= =
y
v
x
v
y
u
x
u
y y x v y x u
(i)
y
v
x
v
y
u
x
u
v u
c
c
c
c
c
c
c
c
, , , , , are all continuous at every point.
(ii)
y
v
x
u
c
c
= = =
c
c
2 0
The Cauchy Riemann equations are not satisfied at any point ( ) y x, . Hence
f is not differentiable at any point . iy x z + =
Example: Identify the points where ( )
2 2
y i x z f + = is differential.
Solution: ( )
2
, x y x u = and ( )
2
, y y x v =
x
y
v
x
v
y
u
x
x
u
2 , 0 , 0 , 2 =
c
c
=
c
c
=
c
c
=
c
c

(i)
y
v
x
v
y
u
x
u
v u
c
c
c
c
c
c
c
c
, , , , , are all continuous
(ii) y x y x
y
v
x
u
= =
c
c
=
c
c
2 2 and
y
u
x
v
c
c
= = =
c
c
0 0
Hence the Cauchy Riemann Equations are satisfied iff . y x =
f is differentiable at points x i x z + = and at such points
( ) ( ) . 2 0 2
1 1
x i x
x
v
i
x
u
ix x f z f = + =
c
c
+
c
c
= + =
Remark: In polar form the Cauchy Riemann equations can be written

u c
c
=
c
c v
r r
u 1
and
u c
c
=
c
c u
r r
v 1





Proof: We know that u cos r x = and
|
.
|

\
|
= + = =

x
y x r r y
1
tan , , sin
1 2 2
u u By Chain
Rule,
|
|
.
|

\
|
+

c
c
+
+
c
c
=
c
c
c
c
+
c
c
c
c
=
c
c
2 2
2 2
.
y x
y u
y x
x
r
u
x
u
x
r
r
u
x
u
u
u
u


r
u
r
u u
u
u
sin
. cos .

c
c
+
c
c
=

|
|
.
|

\
|
+ c
c
+
+
c
c
=
c
c
c
c
+
c
c
c
c
=
c
c
2 2
2 2
.
y x
x u
y x
y
r
u
y
u
y
r
r
u
y
u
u
u
u


r
u
y
u u
u
u
cos
. sin .
c
c
+
c
c
=
Similarly,
u
u
u sin .
1
cos .
c
c

c
c
=
c
c v
r r
v
x
v

u
u
u cos .
1
sin .
c
c
+
c
c
=
c
c v
r r
v
y
v

If follows from the Cauchy Riemann equation that
u
u
u u
u
u cos
1
sin sin
1
cos
c
c
+
c
c
=
c
c
c
c

c
c
=
c
c v
r r
v u
r r
u
y
u
x
u

( ) 1 0 sin
1
cos
1
= |
.
|

\
|
c
c
+
c
c
|
.
|

\
|
c
c

c
c
u
u
u
u r
v u
r
v
r r
u

and
r
u
r
u v
r r
v
y
u
x
v u
u
u u
u
u
cos
sin sin
1
cos
c
c

c
c
=
c
c
c
c

c
c
=
c
c

( ) 2 0 sin
1
cos
1
= |
.
|

\
|
c
c

c
c
+ |
.
|

\
|
c
c

c
c
u
u
u
u
v
r r
u u
r r
v

Multiplying (1) by u cos and (2) by u sin and adding yields

u u c
c
=
c
c
=
c
c

c
c v
r r
u v
r r
u 1
0
1

Multiplying (1) by u sin and (2) by u cos , and adding yields

u u c
c
=
c
c
=
c
c

c
c u
r r
v u
r r
v 1
0
1


Theorem 3: Let ( ) u , r u and ( ) u , r v have a single real-valued is some nbhd of ( )
0 0
u r and
let
u u c
c
c
c
c
c
c
c v
r
v u
r
u
v u , , , , , be continuous at ( )
0 0
u r , and satisfy the Cauchy Riemann
equations in polar coordinate at ( )
0 0
u r where . 0
0
= r Then ( )
0
1
z f exists where
( )
0 0 0 0 0 0
sin cos u u u cis r r z = + = and
( ) ( ) |
.
|

\
|
c
c
+
c
c
=
r
v
i
r
u
z f
0 0 0
1
sin cos u u
Proof: ( )
x
v
i
x
u
z f
c
c
+
c
c
=
0
1

|
.
|

\
|
c
c

c
c
+
c
c
+
c
c
= u u u u sin cos sin cos
r
v
r
v
i
r
v
x
u

( ) u u sin cos i
r
v
i
r
u
|
.
|

\
|
c
c
+
c
c
=
Example 4: ( ) , 2 0 , 0 ,
2
2
1
t u
u
< < > = = r cis r z z f using Theorem 3 show that ( ) z f
1

exists everywhere except along the real axis and at the origin and ( )
( ) z f
z f
2
1
1
=
Solution: ( )
2
cos ,
u
u r r u = ( )
2
sin ,
u
u r r v =

2
cos
2
1 u
r
r
u
=
c
c

2
sin
2
1 u
r
r
v
=
c
c


2
sin
2
1 u
u
r
u
=
c
c

2
cos
2
1 u
u
r
v
=
c
c

(i)
u u c
c
c
c
c
c
c
c v
r
v u
r
u
v u , , , , , are continuous if 0 > r
(ii)
r
u
r
r
v
r c
c
= =

=
c
c
2
cos
2
1
2
cos
2
1 1 u u
u


r
v
r
r
v
r c
c
= =

=
c
c
2
sin
2
1
2
sin
2
1
.
2
1 1 u u
u





Therefore, the Cauchy Riemann equations are satisfied if . 0 > r
( ) ( ) |
.
|

\
|
c
c
+
c
c
=
r
v
i
r
u
i z f u u sin cos
1

( )
|
|
.
|

\
|
+ =
2
sin
2
1
2
cos
2
1
sin cos
u
u
u
u u
r
i
r
i
|
.
|

\
|

(

+ |
.
|

\
|
+ =
2
cos sin
2
sin cos
2
sin sin
2
cos cos
2
1 u
u
u
u
u
u
u
u i
r

|
|
.
|

\
|
|
.
|

\
|
+ = |
.
|

\
|
=
2
sin
2
cos
2
1 u
u
u
u i
r

|
.
|

\
|
= |
.
|

\
|
+ =
2
sin
2
cos
2
1
2
sin
2
cos
2
1 u u u u
i
r
i
r

( ) z f
cis r
2
1
2
2
1
= =
u


2.6. ANALYTIC FUNCTIONS

DEFINITION: A complex function ( ) z f is said to be analytic at C z e
0
iff f is
differentiable at every point of some nbhd of
0
z . A function f is said to be analytic on
a set D if it analytic at every point of D. A function f is said to be an entire function
iff it is analytic on the entire complex plane.

Remark: if f analytic at
0
z , then f is differentiable at
0
z but not viceversa.
Example 1 The function ( )
2
z z f = is differentiable at 0 = z but not analytic at 0 = z .
Note: . Re Monogenic c Holomorphi gular Analytic
Definition: if a function is analytic at some point in every nbhd of a point
0
z except at
0
z itself, then
0
z is called a singular point, or a singularity, of the function.


Example 2: If ( )
z
z f
1
= , then ( ) ( ) 0
1
2
1
=

= z
z
z f . Thus f is analytic at every point
except the point 0 = z , where it is not continuous, so that ( ) 0
1
f cannot exist.
The point If ( )
z
z f
1
= , then ( ) ( ) 0
1
2
1
=

= z
z
z f . Thus f is analytic at every
point except the point 0 = z is a singular point
Example3. Since the function ( )
2
z z f = is nowhere analytic, it has no singular points at
all points in C.
Example 4: ( ) { }. , , 0 /
1
3
i i
z z
z
z f .
+

=
i , 0 and i are singularities of . f
Theorem: Suppose that f and are analytic on S .
1. f is analytic on S .
2. . f is analytic on S .
3.

f
is analytic S , provided ( ) . 0 S z e =
4. ( ) ( ) z f is analytic S , if f is analytic at every ( ) z for all z in S
Theorem: Given ( ) ( ) ( ). , , y x iv y x u z f + = Suppose that
(i) the functions , , , , , ,
y
v
x
v
y
u
x
u
v u
c
c
c
c
c
c
c
c
are continuous in some N nbhd of
( )
0 0
, y x and
(ii) the Cauchy Riemann equation,
y
v
x
u
c
c
=
c
c
and
y
u
x
v
c
c
=
c
c
are satisfied at every
point of . N
Then f is analytic at
0 0 0
y i x z + =
Theorem: Suppose that the function ( ) iv u z f + = is analytic at
0
z then
y
v
x
u
c
c
=
c
c
and

y
u
x
v
c
c
=
c
c
at every point of some neighborhood of the point
0
z .


2.7. HARMONIC FUNCTIONS
Analytic functions possess the following property which will be established in the
nextchapter (Chapter 4).
f is analytic at
0
z
1
f is analytic at
0
z
11
f is analytic at
0
z
.
.
( ) n
f is analytic at
0
z
.
.
Let ( ) ( ) ( )
( )
. . . , , , , , ,
4 111 11 1
f f f f y x iv y x z f + = are also analytic at
0
z . Since
differentiability implies continuity it follows that . . . ,. , ,
111 11 1
f f f are continuous at
0
z .
We know that the derivatives of complex functions are given interms of the partial
derivatives of their component functions. Then since . . . ,. , ,
111 11 1
f f f are continuous at
0
z , it follows that the partial derivatives of all orders are continuous at ( )
0 0
, y x . In
particular, the second order mixed partial derivatives are continues and hence equal, i.e

y x
v
x y
v
and
y x
u
x y
u
c c
c
=
c c
c
c c
c
=
c c
c
2 2 2 2

But f is analytic at
0
z ,

y
u
x
v
and
y
v
x
u
c
c
=
c
c
c
c
=
c
c

Which upon differentiation yield
(i)
y x
v
x
u
c c
c
=
c
c
2
2
2
(iii)
2
2 2
y
v
x y
u
c
c
=
c c
c

(ii)
x y
v
y
u
c c
c
=
c
c
2
2
2
(iv)
2
2 2
x
v
y x
u
c
c
=
c c
c




Adding (i) and (ii), we obtain


. . . .(*)

Adding (iii) and (iv), we obtain

. . . .(**)

Either one of the two equations in (*) and (**) is called LAPLACE'S Equation.

Definition: any functions ( ) y x u , satisfies Laplace's equations throughout some nbhd
Of
0
z is said to be Harmonic at
0
z , provided that it has continuous second order partials.
If ( ) ( ) ( ) y x v i y x u z f , , + = is analytic at
0
z , then we have shown that u and v are
harmonic. Such pairs of Harmonic functions are called conjugate Harmonic functions.

Example 1: Let ( )
2 2
, y x y x u =
(a) Show that u is Harmonic
(b) Determine the conjugate Harmonic
Solution: (a) ( )
2 2
, y x y x u =
2 , 2 , 2 , 2
2
2
2
2
=
c
c
=
c
c
=
c
c
=
c
c
y
u
y
y
u
x
u
x
x
u

( ) 0 2 2
2
2
2
2
= + =
c
c
+
c
c
y
u
x
u

u is a harmonic function



0
2
2
2
2
=
c
c
+
c
c
y
u
x
u

0
2
2
2
2
=
c
c
+
c
c
y
v
x
v



(b) To find the conjugate Harmonic, let
( ) ( ) ( ) y x v i y x u z f , , + = be analytic
The Cauchy Riemann equations hold.
(i) ( ) ( ) x xy y x v x
y
v
y
v
x
u
| + = =
c
c

c
c
=
c
c
2 , 2 where ( ) x | is an arbitrary
functions of x .
(ii) ( ) ( ) y y x y
y
u
x
v
2 2 2
1
= = +
c
c
=
c
c
|
( ) 0
1
= x |
( ) , c x = | c a concstant
Hence ( ) c xy y x v + =2 , is the conjugate Harmonic, that is,
( ) ( ) ( ) c xy i y x z f + + = 2
2 2
is analytic function.
Example 2: Let ( ) x y e y x v
x
+ = cos ,
(a) Show that v is Harmonic
(b) Find u such that ( ) iv u z f + = is analytic.
Solution: a)
y e
y
v
x
cos
2
2
=
c
c

0 cos cos
2
2
2
2
= =
c
c
+
c
c
y e y e
y
v
x
v
x x

v is a Harmonic function
b) Let ( ) iv u z f + = be analytic.
( ) y y e u y e
x
u
y
v
x
u
x x
| + = =
c
c

c
c
=
c
c
sin sin
( ) y y e y e
y
u
x
v
x x 1
cos 1 cos
2
2
2
2
| + + =


( ) ( ) , 1
1
C y y y + = = | | where c is a complex constant.
( ) , sin , c y y e y x u
x
+ = and thus
( ) ( ) ( ) x y e i c y y e z f
x x
+ + + = cos sin
Note: The Laplace's Equations furnishes a necessary condition for a function to be the
real (imaginary part of analytic function).
Example 3: Show that ( ) ( )
2 2
, , y x y x u = cannot be the real part of an analytic function
Solution: Since , 0 4 2 2
2
2
2
2
= = + =
c
c
+
c
c
y
u
x
u
we have no analytic function such tat u the
real part of it.
Remark: The real and imaginary prts of an analytic function of a complex variable
when expressed in polary form satisfy the equations
0
1 1
2
2
2 2
2
=
c
c
+
c
c
+
c
c
u
u
r r
u
r r
u

and 0
1 1
2
2
2 2
2
=
c
c
+
c
c
+
c
c
u
v
r r
v
r r
v
Laplace's Equations in Polar Form
Proof: From Cauchy Riemann equations in polar form we have

2
2
r
v
r
r
v v
c
c
+
c
c
=
c
c
u
and
2
2 2
1
u u c
c
=
c c
c u
r r
v

To eliminate v differentiate (*) with respect to r and (**) with respect tou

2
2 2
r
v
r
r
v
r
v
c
c
+
c
c
=
c c
c
u
and
2
2 2
1
u u c
c
+ =
c c
c v
r
r r
v

Assuming the second partial derivatives are continuous we have
r
v
r
v
c c
c
=
c c
c
u u
2 2
. Hence

2
2
2
2
1
u c
c
=
c
c
+
c
c u
r r
u
r
r
u

0
1 1
2
2
2 2
2
=
c
c
+
c
c
+
c
c

u
u
r r
u
r r
u




To eliminate , u differentiate (*) with respect to u and (**) with respect to r we obtain:
u u u u c c
c

c
c
=
c
c
c c
c
=
c
c
r
u
r
u
r r
v
and
r
u
r
v
2
2 2
2 2
2
2
1 1

u u u u 2 2
2
2
2
2
2
2 2
2
2
2
2
2
2
2
2
2
2
r
u
r
u
r
v
r and
r
u
r
v
= =

u c c
c

c
c
=
r
u
r
r
v
r
2

Since ,
2 2
u u c c
c
=
c c
c
r
u
r
u
we obtain

r
v
r
r
v
r
v
c
c
=
c
c
+
c
c
2
2
2
2
2
u

0
1 1
2
2
2 2
2
=
c
c
+
c
c
+
c
c

u
v
r r
v
r r
v

Example 4: In the domain , 2 0 , 0 t u < < > r show that, the function, ( ) r r u ln , = u is
Harmonic and find its conjugate Harmonic.
Solution: =
c
c
=
c
c
=
c
c
=
c
c
2
2
2 2
2
,
1
, 0 ,
1
u u
u
r r
u u
r r
u

0
1 1
.
1 1 1
2 2 2
2
2
2
= + + =
c
c
=
c
c
+
c
c

r r r r
u
r
u
r r
u
u
and hence ( ) u , r u is Harmonic
function. To find its conjugate Harmonic, let ( ) v i u z f + = be analytic.
( ) r v
r
r
r
u
r
v
| u
u
= = =
c
c
=
c
c
1
1
.
( ) ( ) c r
r
r
u
r r
v
= = =
c
c
=
c
c
| |
u
0 .
1 1
1

( ) , , c r v + = u u where c is a complex constant .Therefore,
( ) ( ) c i r x f + + = u ln is analytic.





CHAPTER 3: ELEMENTARY FUNCTIONS

This chapter includes: elementary functions such as exponential functions, logarithmic
functions, trigonometric functions and hyperbolic functions of a complex variable. The
primarily objective of the chapter is to introduce elementary functions of complex
variables and to study its analyticity.

Now we begin the chapter by recalling that the power series representation of
x
e where x
is a real variable.

+ + = x e
x
1 +
! 2
2
x
...
! 3
3
+
x
=

=0
!
n
n
n
x

....
! 5 ! 3
sin
5 3
+ + =
x x
x x =

=
+
+

0
1 2
)! 1 2 (
) 1 (
n
n n
n
x


....
! 4 ! 2
1 cos
4 2
+ + =
x x
x =

0
2
)! 2 (
) 1 (
n
n n
n
x

Activity: in the above expressions replacing x by iy and using properties of the
imaginary unit i

a) find e
iy
.
b) show that e
iy
= cosy + isiny

The formula e
iy
= cosy + isiny is called Eulers Formula.

Exponential Function

In this section we want to extend the notion of real exponential functions to complex
exponential functions.
Let z = x + iy be a complex number then we try to extend the exponential rules for real to
complex and put,
e
z
= e
iy x iy x
e e =
+
= e
x
( cosy + isiny)

Definition:- Given a complex number z = x + iy we define the exponential function e
z
by
the equation e
z
= e
x
(cosy + isiny).

If z is given in polar form as z = r(
u cos u sin i +
) then using Eulers formula z can be
written as,
z = r
u i
e
and its conjugate z is given by,
z =
u
u u u u
i
re i r i r

= + = ) sin (cos ) sin (cos
.


Exercises

1. Given complex numbers z =
) sin (cos o o i r +
and w =
) sin (cos | | i +
show that
a) the product zw =
) ( | o

+ i
e r

b) the quotient z w =
) ( | o

i
e
r
(w= 0)

2. For any given complex number z and w show that
a)
w z w z
e e e
+
=

b)
w z w z
e e e

=

c)
nz n z
e e = ) (
for any integer n.
d)
x z
e e =
and arg(z) = y + 2k
t
for some integer k when z = x + iy.

3. Show that the exponential function of a complex variable is periodic with period 2i
t
.
4. Show that
) (
z z
e e =

5. Show that the exponential function is an entire function.
6. Find z such that e
z
= i.

Trigonometric Functions

From Eulers formula, we have e
iy
= cosy + isiny (1)
and
e
-iy
= cosy isiny (2)
Adding equation (1) and (2) and simplifying we get

2
cos
iy iy
e e
y

+
= (3)
Subtracting equation (2) from equation (1) and simplifying we get

i
e e
y
iy iy
2
sin

= (4)
Based on equations (3) and (4) we have the following definition.

Definition:- Given a complex variable z, the cosine of z denoted by z cos and the sine of
z are defined as


2
cos
iz iz
e e
z

+
=


i
e e
z
iz iz
2
sin

=



The other trigonometric functions are defined in terms of the sine and cosine functions.

z
z
z
cos
sin
tan =



z
z
z
sin
cos
cot =


z
z
cos
1
sec =


z
z
sin
1
csc =





Exercises

1. For any complex numbers z and w show that
a) sin(-z) = -sinz
b) cos(-z) = cosz
c) sin
2
z + cos
2
z =1
d) cos(z+w) = coszcosw sinzsinw
e) cos(z-w) = coszcosw +sinzsinw
f) sin(z+w) = coszsinw +sinzcosw
g) sin(z-w) = coszsinw sinzcosw
h)
z sin
=
z sin

i)
z cos
=
z cos

2. The function sinz is an entire function.
3. The function cosz is an entire function.

















Hyperbolic Functions

We know that the sine hyperbolic and cosine hyperbolic functions of real variable x
denoted by sinhx and coshx respectively are defined in terms of exponential function e
x

as

2
sinh
x x
e e
x

=
and
2
cosh
x x
e e
x

+
=


In a similar way we define hyperbolic functions of complex variable z as



2
sinh
z z
e e
z

=
and
2
cosh
z z
e e
z

+
=




The other hyperbolic functions are defined in terms of sinhz and coshz as

z
z
z
cosh
sinh
tanh =



z
z
z
sinh
cosh
coth =


z
hz
cosh
1
sec =


z
hz
sinh
1
csc =



Exercises

1. For any complex numbers z and w show that
a) sinh(-z) = -sinhz
b) cosh(-z) = coshz
c) cos
2
hz - sin
2
hz =1
d) cosh(z+w) = coshzcoshw +sinhzsinhw
e) cosh(z-w) = coshzcoshw -sinhzsinhw
f) sinh(z+w) = coshzsinhw +sinhzcoshw
g) sinh(z-w) = coshzsinhw sinhzcoshw
h)
z sinh
=
z sinh

i)
z cosh
=
z cosh

j) sin(iz) = isinhz
k) cosh(iz) = cosz

l) cos(iz) = coshz

3. Given z = x + iy for any real numbers x and y show that
a) sinhz = sinhxcosy + icoshxsiny
b) coshz = coshxcosy + isinhxsiny
c) sinz = sinxcoshy + icosxsinhy
d) cosz = cosxcoshy - isinxsinhy
e)
y x z
2 2
2
sin sinh sinh + =

f)
y x z
2 2
2
cos sinh cosh + =

4. The function sinhz is an entire function.
5. The function coshz is an entire function.
6. Find z such that
a) e
z
= i
b) sinz = 0
c) cosz = 0
d) tanz = 0
e) sinz = sinw
f) sinhz = 0
g) coshz = 0



























Logarithmic Functions

Consider w = e
z
, to each value of z corresponds a value of w. We shall now consider the
inverse process and examine the existence of z such that e
z
has a given value w.

Activities
1. Is there any complex number z such that e
z
= 0?
2. If there is a complex number z such that e
z
= w, then is z unique?

Let us try to solve for z, by putting
e
z
= w where z = x + iy

w e
iy x
=
+

w y i y e
x
= + ) sin (cos

w e
x
=
and y = arg(w)
Thus if
0 = w
that is if w
0 =
, then x =
w log
is a single value of x and stands for the
real logarithm of a the positive real number
w
. But y = arg(w) is infinite valued.
Hence z = x + iy =
w log
+ iarg(w), where arg(w) is infinite valued. We are now ready to
define logarithmic functions of a complex variable z.
Notation:- Instead of
log
z
we write the natural logarithm
z ln
and use
z log
for the
complex variable z. (Mostly instead of
z
we write r)
Definition:- Given a complex variable z =
z
u i
e
we define the logarithm of z by the
equation

z log
=
) log(
u i
e z
=
u i z + ln
----------- (1)
This definition is the natural one in the sense that it is written by formally using
properties of real logarithms.
Corresponding to the particular argument
u
of z such that -
t
<
t u s
, we may write
z =
) 2 ( t u n i
e z
+
for any integer n.
Thus equation (1) can be written as

z log
=
) 2 ( ln t u n i z + +
where n is an integer.
This shows that the function logz is multiple-valued with infinitely many values.
Now if n = 0 and -
t
<
t u s
we call the principal value of logz denoted by plogz and is
given by,
plogz =
u i z + ln
.




Example:- Find the logarithms of the following complex numbers.
1. z = 1+i
2. z = i
3. z = 2
4. z = -1
5. z = -ei
Solutions:- logz = ln
) 2 ( t u n i z + +
where -
t
<
t u s

1.
2 1 1 1
2 2
= + = + = i z
and
4
)
1
1
( tan
1
t
u = =


Hence logz = ln
2
) 2
4
( t
t
n i + +
=
) 2
4
( 2 ln
2
1
t
t
n i + +

and thus plogz =
4
2 ln
2
1 t
i +
which is obtained when n = 0

2.
1 1 1 0 0
2 2
= = + = + = i z
and
2
t
u =

Hence logz = ln1
) 2
2
( t
t
n i + +
=
) 2
2
( t
t
n i +

and thus plogz =
2
t
i
which is obtained when n = 0
3.
2 4 0 2 0 2
2 2
= = + = + = i z
and
0 = u

Hence logz = ln2
) 2 0 ( t n i + +
=
) 2 ( 2 ln t n i +

and thus plogz = 2 ln which is obtained when n = 0
4.
1 1 0 ) 1 ( 0 1
2 2
= = + = + = i z
and
t u =

Hence logz = ln1
) 2 ( t t n i + +
=
) 2 ( t t n i +

and thus plogz =
t i
which is obtained when n = 0
5.
e e e ei z = = + = =
2 2 2
) ( 0 0
and
2
t
u

=

Hence logz = lne
) 2
2
( t
t
n i +

+
=
) 2
2
( 1 t
t
n i +

+

and thus plogz =
2
1
t
i
which is obtained when n = 0








Branches of z log
Consider the principal logarithm plogz at z = x
o
+ 0i on the negative real axis (x
o
< 0).
At that point
xo
z

=
and
t u =
. Since
t u i i
as
xo
z
from the second quadrant,
and
t u i i
as
xo
z
from the third quadrant, plogz is not continuous on the
negative real axis.
Thus, its derivative does not exist at each point on the negative real axis.
The single valued-function Plogz =
u i z + ln
,
t u t < <
is defined at all points
u i
e z z =
except at the origin z = 0, and points on the negative real axis.
Plogz =
u i z + ln
,
t u t < <
has continuous components
u = lnr and v =
u
where
r z =
> 0, throughout its domain of definition.
Moreover, the derivatives


r r
u 1
=
c
c
,
0 =
c
c
u
u
and
0 =
c
c
r
v
,
1 =
c
c
u
v

are all continuous function of the point z in that domain, and satisfy Cauchy-Riemann
conditions in polar coordinates. Thus plogz is differentiable and

) )( sin (cos ) log (
r
v
i
r
u
i z p
dz
d
c
c
+
c
c
= u u

=
) 0
1
( i
r
e
i
+
u

=
u i
re
1

=
z
1


Therefore,
z
z p
dz
d 1
) log ( =
,
) ) arg( , 0 ( t t < < = z z


Activity

1. Using the complex plane identify the region in which plogz is analytic and
regions in which it is not analytic.
2. Consider
u i r z + = ln log
where r =
z
> 0 and
t
t
u
t
2
4 4
+ < <

a) Is
u i r z + = ln log
single-valued or not?




b) Find
) (log z
dz
d

c) Is logz analytic in the domain r =
z
> 0 and
t
t
u
t
2
4 4
+ < <
?
3. Replace
4
t
by
o
u
and consider
u i r z + = ln log
where r =
z
> 0 and
t u u u 2 + < <
o o


d) Is
u i r z + = ln log
single-valued or not?
e) Find
) (log z
dz
d



f) Is logz analytic in the domain r =
z
> 0 and
t u u u 2 + < <
o o
?

The principal values of logz represented by plogz is said to be the principal branch of
logz. Each point of the negative real axis
t u =
and the origin are singular points of
plogz. The ray
t u =
is the branch cut for the plogz. The ray
= u o
u
is the branch cut for
u i r z + = ln log
where r =
z
> 0 and
t u u u 2 + < <
o o
. The singular point z = 0 is
common to all branch cuts for logz is called a branch point.


Exercise: Show that

1.
z
e z
log
=

2. If w = e
z
and z = x+iy, then logw = z +
t k 2
for some integer k.
3. For complex numbers z and w in the domain of definition of log
a) log(zw) = logz + logw.
b)
w z
w
z
log log ) log( =

c)
z m z
m
log ) log( =

d)
. log
1
) log(
1
z
n
z
n
=

e)
z
n
m
z
n
m
log ) log( =

f)
n
z m
n
m
e z
) (log
=




Complex Exponents
If c is a complex constant,
c
z and
c
z

are defined as follows:

c
z =
z c
e
log
and
c
z

=
c
z
1
for z
0 =

Since logz is multiple-valued,
c
z is also multiple-valued. The same idea can be used for
the principal of power function.
The principal value of
c
z is
z c
e
log
where plogz is the principal logarithm of logz.
Hence
c
z =
z c
e
log
is single-valued for
u i
re z =
where r =
0 > z
and
t u t < <
.
Example:-
t t
t t
k
i k
i
i i i
e e e i
4
2
) 4 (
2
log 2 2 +
+

= = =
for some integer k.
The principal value of
i
i
2
is
t
e
.

Exercise

1. Find the domain in which
c
z is a single-valued
2. Find the domain in which
c
z is analytic.
3. In the domain where
c
z is analytic show that
1
) (

=
c c
cz z
dz
d

4. Use exercise number (3) to find the derivatives of the following:
a) f(z) =
3
2
z
b) f(z) =
i
z
2

c) f(z) =
i
z
4 3+

d) f(z) =
z
z
5. If c is a nonzero complex constant, then the exponential function with base c is
given by the equation,

c z z
e c
log
=
.

a) Show that the function f(z) =
z
c
is an entire function.
b) Find
) (
z
i
dz
d
.
c) Find
) ) 1 ((
z
i
dz
d
+
.









Inverse Of Trigonometric And Hyperbolic Functions

A complex function g(z) is said to be the inverse of a function f(z), if f(g(z)) = z = g(f(z)).
We denote the inverse of f(z) by
). (
1
z f


Inverses of Trigonometric and Hyperbolic functions can be described in terms of
logarithms.
Suppose w =
) ( sin
1
z

sinw = z

i
e e
z
iw iw
2

=


iw iw
e e iz

= 2


iw iw
e ize
2
2 =


0 1 2
2
=
iw iw
ize e


Solving and simplifying, we get
) 1 log( ) ( sin
2 1
z iz i z + =

,
which is a multiple-valued function.



Similarly, ) 1 log( ) ( cos
2 1
+ =

z z i z
) log(
2
) ( tan
1
z i
z i i
z

+
=

Exercise:

1. Find an expression for the following:
a) ) ( sinh
1
z

b) ) ( cosh
1
z

c) ) ( tanh
1
z

2. Find the derivatives for the following:

a) ) ( sin
1
z

b) ) ( cos
1
z

c) ) ( tan
1
z

d) ) ( sinh
1
z

e) ) ( cosh
1
z

f) ) ( tanh
1
z

CHAPTER 4: THE INTEGRAL

Introduction: This chapter will deal with the notion of integrability and integral of a
complex function along an oriented curve in the complex plane. We shall see that a
function is integrable along a curve if the curve is piecewise smooth and the function is
continuous along the curve. Some elementary properties of integration will also be
obtained. The following section of the chapter deals with the notion of oriented and
piecewise smooth curves.

In order to introduce the line integral of f(z) in a simple way, first it is better to define
the definite integral of a complex-valued function f of a real variable t.
We write,

f(t) = u(t) + iv(t) for
b t a s s


where u and v are real-valued piecewise continuous, functions of t on a bounded interval
(a, b). Here u is the real part of f and v is the imaginary part of f.
Example: f(t) =
it
e
is a complex-valued function defined on every interval. The real part
of f is cost and the imaginary part of is sint. In particular for
t 2 0 s s t
, f(t) =
it
e
is the
unit circle on the complex plane.

4.1. Complex valued Functions of a Real Variable

DEFINITION: A function f of the real variable t is said to be complex valued on an
interval, b t a s s if
) ( ) ( ) ( t iv t u t f + =
holds on this interval, where u and v are real valued functions. Here u is called the real
part of f and v is called the imaginary part.

Example 1:
it
e t f = ) ( is complex valued function defined on every Interval. The real
part of f is cost and the imaginary part is t sin . In particular, for
it
e t f t = s s ) ( , 2 0 t is
the unit circle on the complex plane.



The complex valued function ) (t | is said to be continuous at a integral
0
t if
0 ) ( ) ( lim
0
= t t | | as t tends to t
0

holds in the sense of real variable theory. This can be expressed in ) , ( o c terms also.
When f is continuous at every point of an interval, f is said to be continuous on the
interval, and a similar convention for passing from points to interval applies to other
properties introduced later.

If f is continuous, then so are u and v indeed,
Since, ) ( ) ( ) Re(
0
t f t f z s , we have
) ( ) ( ) ( ) (
0 0
t f t f t u t u s
and if the right side tends to zero as
0
t t , the left side must. Thus, u is continuous at
0
t if f is. A similar proof holds for v . Conversely, if u and v are continuous, then f must
be. This follows from the inequality
z z z Im Re + s
applied to )). ( ) ( ( )) ( ) ( ( ) ( ) (
0 0 0
t v t v i t v t u t f t f + =
The complex valued function f is said to be differentiable if u and v are, and in that
case, by definition ) ( ) ( ) (
1 1 1
t v i t u t f + = .

It is easily verified that the sum, difference, product and quotient of complex valued
functions can be differentiated as in real analysis. Also the Chain Rule holds in the
following form: If F is analytic at the Integral ) (t z = , and if ) (
1
t exists, then the
complex valued function ) ( ( ) ( t F t f = is at t and
) ( . ) ( ( ) (
1 1 1 1
t t F t f =
Where is a complex valued function of the real variablet .




A complex valued function ) ( ) ( ) ( t iv t u t f + = is said to be integrable over an interval
| | b a, if u and v are, and then, by definition.

} } }
+ =
b
a
b
a
b
a
dt t v i dt t u t f ) ( ) ( ) (
Many familiar rules of Integration for real functions can be carried over to the complex
case. In particular, the two forms of the fundamental Theorem of calculus for real
functions give corresponding results for complex functions, namely
) ( ) ( ) ( ) ( ), ( ) (
1
a f b f t f dt t f t f d f
dt
d
t
a
b
a
b
a
= = =
} }
u u
Where f is continuous or
1
f is continuous, respectively
Example 2: Let
it
e t f
t 2
3 ) ( = for 1 0 s st . Then
a) ( ) ( ) ( ) t i t i t t f t t t t 2 . cos 6 2 sin 3 2 cos 3
1 1
+ + =
( )
it
e i t i t i
t
t t t t
2
6 2 sin 2 cos 6 = + =
b) ( )
} } }
+ =
1
0
1
0
1
0
2 sin 3 2 cos 3 tdt i tdt dt t f t t

1
0
1
0
2 cos
2
3
2 sin
2
3
|
.
|

\
|
+ = t i t t
t
t
t

0 0
2
3
2
3
0
2
3
0 .
2
3
i i + =
|
|
.
|

\
|
|
.
|

\
|
+
|
|
.
|

\
|
=
t t t t

0 =
Exercise: find ( ) t f
1
and ( )
}
s s +
t
t
0
0 , 2 sin sin t t i t if dt t f

4.2. Contours

One of the main uses of complex valued functions is to represent curves in the complex
plane. As we recall from real analysis, a curve is given by parametric equations
), (t u x = ) (t v y =
Where u and v are real valued functions defined on some interval b t a s .


For instance the line 0 1= + x y is given by the parametric equations
, t x = , 1 = t y < < t
The unit circle 1 = z has parametric equations
, cos t x = , sin t y = t 2 0 s s t
NOTE: The set of point occupied by a curve is called the trace of the curve, and the
curve is said to le in a given region if the trace does.

A curve C, with parametric equations ) (t u x= and ) (t v y = , is said to be continuous if u
and v are continuous; differentiable if u and v are differentiable, and so forth. If we
define iy x z t v t u t f + = + = , ) ( ) ( ) ( then both equations ) ( ), ( t v y t u x = = are equivalent to
the single equation,
) (t f z =
The function ) (t f in this representation is, clearly, complex valued. It should perhaps
be mentioned that no logical distinction is made between the curve and the function f so
that, from a purely logical stand point, the curve is the function.

DEFINITION:
1. A continuously differentiable curve is called an arc.
2. A curve is said to be a smooth curve on | | b a, if it can be represented by two real-
valued functions ) (t u x = and 9 e s s = , ) ( b t a t v y such that their derivatives
) (
1
t u
dt
dx
= and ) (
1
t v
dt
dy
= exist and are continuous functions of t for b t a < < .

The above definition implies that the curve is continuous on b t a s s and has tangent at
every integral of b t a < < .
The length L of a smooth curve on | | b a , is given by the equation.

}
|
.
|

\
|
+ |
.
|

\
|
=
b
a
dt
dt
dy
dt
dx
L .
2 2





Example: find the length of the curve r z =
Solution: Parametric equation of the circle is given by,



ri
-r
r
R.A
ri
I.A
t r x cos = and t 2 0 , sin s s = t t r y
t r
dt
dx
sin = and t r
dt
dy
cos =
( ) ( )
} }
= + =
t t 2
0
2
0
2 2
cos sin dt dt t r t r L
r t 2 =
86


Let C be a curve with parametric equations ) ( , ), ( t y and t x | = = , and )) ( ), ( ( a a | is the
initial point and ( ) ( ) ( ) b b | , is the terminal point of C.

DEFINITION: A curve C is called a contour or path if it consists of a finite number of
smooth curves (arcs) C
1
, C
2
, , C
n
such that the terminal point of C
k
coincides with the
initial point of C
k+1
for k = 1,2, , n 1.

A contour is, therefore, a continuous chain of finite number of smooth curves. Hence C is
said to be piecewise smooth.



n
C C C C + + + = ...
2 1

The arrow indicates the
orientation of the path C.



If the initial point of path C concedes with its terminal point then C is said to be a closed
path, otherwise said to be open path.

A path C that does not interest itself (except that its initial point and terminal points may
coincide) is called a simple path, otherwise a multiple path.








87

DEFINITION: A simple closed piecewise smooth curve is called a JORDAN CURVE
(French Mathematician Camille Jordan)

Any Jordan Curve C divides the plane Integral o two pars each of which has C as their
boundary.

Theorem (Jordan Curve Theorem). Let C be a simple closed path in the plane. Then the plane
is divided by C in to three mutually dispoint sets: the curve C; the interior of C, Int(C) which
is open and bounded. The exterior of C, Ext (C), which is open and unbounded.

Furthermore, C is the boundary of both interior (C) and exterior (C).




DEFINITION: A simple open path is said to be Positively oriented it if is traced from the
initial to its terminal point. It is traced from its terminal to its initial point then it is said to be
negatively oriented. Similarly, a simple closed path is said to be positively oriented if when
traced its interior of C remains to our immediate left, otherwise negatively oriented.

For example the unit circle
t x cos =
and
t 2 0 , sin s s = t t y
is negatively oriented, but the
unit circle
t x cos =
and
t 2 0 , sin s s = t t y
is positively oriented.

A set S is said to be connected iff any two points is S can be joined by a piecewise smooth
curve that lies entirely in S . S is said to be simply connected iff for every simple closed
path C lying is S , Int( C ) also lies entirely in S . Otherwise we call S multiply connected.


88









The region bounded by two simple closed paths on of the other is called an ANNULUS or
ANNULAR REGION.






4.3. Contour Integrals

Consider a smooth curve C with parametric equations
) (t x | = and ) (t y = for b t a s s .
Let ) , ( y x M be a continuous function onC . Let { } n t t t P , . . . , ,
2 1 0
= . Let
1
= = A
k k k k
t t t t


k
t
1
t
2
t
t
1 t
2
....
t
k-1
t
k
...
t
n-1
n
t
b= t
n
a






89

On the curve C we get the coordinates ( ) ( ) ( )
k k k
t y x | = , so that ( ) ( ) ( )
n n
y x y x y x , . . . , ,
1 1 0 0

are sub n divisions of C . Further, ( )
k k
t | o = and ( )
k k
t o = for all . , . . . , 2 , 1 n k = Then the
value of the function for
k
t t = at ( )
k k
o o , ( )
k k
M o o , .

Since the projection of ) , (
1 1 k k
y x and ) , (
k k
y x on to the axis x is ,
k
x A we get the value
( )
k k
M o o , so that the sum of all becomes

y
z
x
(x
0
,y
0
)
(x
1
,y
1
) (x
k-1
,y
k-1
)
(x
k
,y
k
)
(x
n
,y
n
)
( )
k k
M o o ,

( )

=
A
n
k
k k k
x M
1
, o o

We define the line integral of ) , ( y x M along C with respect to x as

( )
0
, lim ) , (
1

A =
}

=
P
x M dk y x M
C
n
k
k k k
o o

Where { }
n
x x Max P A A = ., . . ,
2
provided the limit exists. The line integral of ) , ( y x M along
C with respect to y is defined analogously except that the arc must be projected on to
the axis y .
The integral of ) , ( y x M along C with respect to y is denoted by

}
C
dy y x M ) , (

90

Theorem: (Existence of line Integral)
Let C be smooth curve with parameterization ) (t x | = and ) (t y = for b t a s s and
) , ( y x M be a continuous real valued function onC . Then the line integral of M along
C with respect to x exists, and

( ) ( ) ( ) ( )
} }
=
b
a
C
dt t t t M dx y x M
1
, ) , ( | |
Proof: Let { }
n
t t t P ,..., ,
1 0
= be a partition of | | | | ( ) ( )
k k k k k k k
t t b a t o t | o t = = e

, , , , ,
1
.
Then

( )
0
, lim ) , (
1

A =
}

=
P
x M dk y x M
C
n
k
k k k
o o

Furthermore, ( ) t | is continuous and differentiable on ( ) b a, , and thus ( ) t | is continuous and
differentiable on ( )
k k
t t ,
1
for each . ,..., 2 , 1 n k = By the Mean Value Theorem, there exists
k
n in ( )
k k
t t ,
1
such that
( )
( ) ( )
k
k
k k
k k
k
t
x
t t
t t
A
A
=

1
1 1
| |
q |
( )
k k k
t x A = A q |
1

Thus ( ) ( ) ( )
k k
n
k
n
k
k k k k k
t M x M A = A

= =
q | o o o o
1
1 1
. , ,

( )
0
, lim ) , (
1

A =
}

=
P
x M dk y x M
C
n
k
k k k
o o


( ) ( ) ( ) ( ) ( )
0
, , lim
1
1

= A =
}

=
P
dt t t t M x M
b
a
n
k
k k k
| | o o


Reimann Integral
91

Remark: By the same reasoning
( ) ( ) ( )dt t t M dy y x M
b
a
C
1
, ) , ( |
} }
=
Example1. Evaluate the line integral w.r.t both variables of
xy y x M = ) , ( along 2 0 , , 4 :
2
s s = = t t y t x C .

Solution:
dt t t dt t t M dx y x M
C
4 . . 4 ) 4 ).( 2 , 4 ( ) , (
2
2
0
2
0
1 2
} } }
= =

}
= = = =
2
0
4 2
0
3
64 2 . 4 4 16 t dt t
dt t t t dt t t M dy y x M
C
2 . . 4 ) ).( 2 , 4 ( ) , (
2
2
0
2
0
1 2 2
} } }
= =

}
= = = =
2
0
2
0
5 4
5
256
5
32 . 8
5
8
8 t dt t

Theorem: Suppose that K is a complex constant and
2 1
c c C + = is a path consisting of
smooth curves
1
c and
2
c and suppose further that the line integral of the function ) , ( y x M and
) , ( y x N exists along
1
c and
2
c with respect to x . Then,
i)
} } }
+ =
1 2
) , ( ) , ( ) , (
C C C
dx y x M dx y x M dx y x M
ii)
} }
=
C C
dx y x M K dx y x KM ) , ( ) , (
iii)
} } }
+ = +
C C C
dx y x N dx y x M dx y x N y x M ) , ( ) , ( ) , ( ) , (
iv)
} }
=
C C
dx y x M dx y x M ) , ( ) , (




92



Example 2: Given the function x y y x M = ) , ( . Evaluate
i)
}
C
dx y x M ) , (
ii)
}
C
dy y x M ) , (
Where
2 1
c c C + = as the figure.
Solution: Parameterization C
t s s = . = t t y t x c 0 , sin cos :
1

1 0 , 1 :
2
s s = . = t t y x c
} } }
+ =
1 2
) , ( ) , ( ) , (
C C C
dx y x M dx y x M dx y x M
( )( ) ( )( )
} }
+ + =
1
0
1 1
0
1 1 cos sin cos dt t dt t t t
t

( ) 0 sin sin cos
0
2
+ =
}
t
dt t t t

} }
+ =
t t
0
2
0
sin sin cos tdt tdt t

t t
0 0
2
2 sin
2
1
2
1
2
sin
|
.
|

\
|
+ = t t
t

2 2
1
0
t
t =
=
+ =
Example 3: Evaluate
}

C
ydx xdy ) ( were C z : 2 = oriented in counterclockwise direction.
Solution: The parametric equations of C are:
t x cos 2 = t 2 0 s s t
t y sin 2 =
} }
=
t 2
0
) ) sin 2 .( sin 2 cos 2 . cos 2 ( ( dt t t dt t t ydx xdy
C


} }
= + =
t t 2
0
2
0
2 2
1 4 ) sin 2 (cos 4 dt dt t t
t
t
8 4
2
0
= = t

93

Note: Some times the path of integration is given in nonparametric form ) (x f y = or
) ( y g x = with initial point ) , ( b a and terminal point ) , ( d c . In such cases substitution yields
the following
} } } }
= =
a c
d
b
C
dy y y M dy y x M and dx x f x M dx y x M ) ), ( ( ) , ( ) ( , ( ) , (
If ) (x f y = then dx x f dy ) (
1
= and if ) ( y x = then dy y dx ) (
1
=
Therefore,
dy y y y M dx y x M
d
b
C
) ( . ) ), ( ( ) , (
1

} }
=
Example 4: Evaluate both line integral of y x y x M 2 ) , ( + = along
2 1
c c C + = where
2
1
: x y c = from ) 4 , 2 ( to ) 0 , 0 ( and
2
: 2 y x c = from ) 0 , 0 ( to ) 2 , 4 ( .
Solution: on xdx dy C 2 :
1
=
} } }
+ =
1 2
) , ( ) , ( ) , (
C C C
dx y x M dx y x M dx y x M


ydy dy C 2
2
= =

} }
+ + + =

2
0
2
0
2
2
2 ) 2 ( ) 2 ( dx y y dx x x

2
0
3 4
0
2
3
2
3
2
4
2
3
2
2
|
|
.
|

\
|
+ +
|
|
.
|

\
|
+ =

y y
x
x


|
.
|

\
|
+ + |
.
|

\
|
=
3
16
4
16
2
3
16
2
4


3
32
18
3
16
2 + + + =
22 16 6 = + =






94


} } }
+ =
1 2
) , ( ) , ( ) , (
C C C
dy y x M dy y x M dy y x M
} }
+ + + =

2
0
2
0
2
2
) 2 ( 2 ) 2 ( dy y y dx x x
} }
+ + + =

2
0
2
0
2
3 2
) 2 ( ) 4 2 ( dy y y dx x x
2
0
2
3
0
2
4 3
3 2
1
3
2
|
|
.
|

\
|
+ + |
.
|

\
|
+ =

y
y
x x
|
.
|

\
|
+ + |
.
|

\
|
+ = 4
3
8
2
16
3
16

4
3
8
2 . 8
3
16
+ + =
4 3 . 4
3
24
= =
4 12 8 = =

4.4. Anti- Derivatives

Let C be a smooth curve with parameterization | o s s + t t x t z ), ( ) ( . Suppose ) (z f w = is a
complex single valued function defined on.. Consider a partition
{ }
n
z z z z P ,... , ,
2 1 0
= induced from | | | o, where
0
z and
n
z coincides with the initial and
terminal points of curveC , respectively. Let
1
= A
k k k
z z z and in each are
k k
z z
1
choose
-
k
z (arbitrary) for . ,..., 2 , 1 n k =




Z



95


Denote { } n k t P
k
,..., 2 , 1 : max = A = .
0 0 0 A A
k k
t t P

We form a Riemann Sum

=
A
n
k
k k
t z f
1
*) ( . We define the complex curve ) (z f along C by
0
*) ( lim ) (
1

A =

}
=
P
t z f dz z f
n
k
k k
C


If the limit exist independent of the partition of | | | o, and independent of the choice *
k
z ,
then f is said to be integrable along C andC called the path of integration.

Theorem: (Existence of Complex Integral)
Suppose that ) , ( ) , ( ) ( y x iv y x u z f + = is continuous at every point of a smooth curveC , then
the integral of f along C exists and
} }
+ + =
C C
idy dx iv u dz z f ) )( ( ) (
| |
}
+ + =
C
udy vdx i vdy udx ) (
} } } }
+ + =
C C C C
udy i vdx i vdy udx

Proof: Let
k k k k k k
y x z i z A + A = A + , * o o . Then
( ) ( ) ( ) ( )( )
k k
n
k
k x k x
n
k
k k
y i x iv u t z f A + A + = A

= = 1 1
* o o o o
( ) ( ) ( )( )
k k
n
k
k x k x
y i x v dx u A + A A =

=1
. o o o o

96


( ) ( )

= =
A A =
n
k
k k x
n
k
k k x
x v x u
1 1
o o o o
( ) ( )

= =
A + A +
n
k
k k x
n
k
k k x
y u i x u i
1 1
o o o o
( )
0
lim ) (
1
*

A =

}
=
P
z z f dz z f
n
k
k k
C



} } } }
+ + =
C C C C
vdx i vdx i vdy udx

Examples 1: Show that if C is any smooth curve from
0
z to
n
z then
}
=
C
n
z z dt
0
.
Proof: Here 1 ) ( , 1 ) (
*
= = zk f z f for all
*
zk with partition

{ }
n x
z z z z P ,... , ,
1 0
=


( )
0 0
) ( lim lim ) (
1
1
1
*

= A =

}
=

=
P P
z z z zk f dz z f
n
k
k k
n
k
k
C


0
) ( lim
0 0

= =
P
z z z z
n n


Note: In particular if C is a closed path then
0
z z
n
= and
}
=
C
dz 0
Exercise: Show that

}

=
C
n
z z
zdz
2
2
0
2
and if C is closed path, then
}
=
C
zdz 0 .
97


Theorem: Let
2 1
C C C + = be a path with
1
C and
2
C if ) (z f and ) (z are integrable
along
1
C and
2
C then

i)
} } }
+ =
1 2
) ( ) ( ) (
C C C
dz z f dz z f dz z f
ii)
} }
=
C C
dz z f k dz z kf ) ( ) (
iii) ( )
} } }
+ = +
C C C
dz z dz z f dz z z f ) ( ) ( ) ( ) (
iv)
} }

=
C C
dz z f dz z f ) ( ) (
v) If for some M , the function f satisfy the relation M z f s ) ( for all C z e and the
length of C is L then
ML dz z f
C
s
}
) (

Example 2: Evaluate
}
C
fdz where
3 2 1
C C C C + + = as in the figure.
The path of Integral egration can be represented as
follows.
1 0 , 0 :
1
s s = x y C hence 0 = dy
1 0 , 0 :
2
s s = y x C hence 0 = dx
x y C = :
3
from (1,1) to (0,0) hence dx dy =
Therefore,
} } } }
+ + =
3 2 1
C C C C
xdz xdz xdz xdz
} } }
+ + + + + =
3 2 1
) ( ) ( ) (
C C C
idy dx x idy dx x idy dx x



98


} } }
+ + + =
1
0
1
0
1
0
) 1 ( . 1 dx i x idy xdx

0
1
2
) 1 (
2
1
2
1
i x i + + + =
( )
|
.
|

\
|
+ + + = 1 2
2
1
2
1
(
2
1
i

2
2
2
1
2
1 i
i
i = + =

Exercises: Evaluate
1.
}
C
ydz along 3 2 , , 1 :
1
s s = =

t e y t x C
t

2. ( )
}
+
C
dz i z 1 3 2
2
along C as in Example 2.
3.
}
C
dz z , along the right half of the unit circle.
4. ( )
}

C
n
n dz z z ,
0
is an integer, along
it
re z z C =
0
: t 2 0 s st or r z z =
0

5. r i z C dz
i z
C
=

}
2 : ,
2
1

6. 3 6 :
6
5
=

}
C
i z C dz
i z
i

7.
( ) ( )
( ) 2 1 : ;
1
3
3
= +
+
}
C
i z C dz
i z

8. 1 3 2 : ,
4
= +
}
i z C dz z
C






99
Cauchys Theorem

Intuitive Version

In mathematics it is important to have an intuitive understanding and to be able to express
that intuition precisely. In this section Cauchys Theorem will be discussed rather
informally; the proofs will be simple and the definitions somewhat causal (although
entirely adequate for most of the applications). The next section provides more precise
definitions, sharper Theorems and more complete proofs.

One form of Cauchys Theorem states that if C is a simple closed curve and if f is analytic
on and inside C, then


0 ) ( =
}
C
dz z f

If the function f is not analytic on the whole region inside C, then the integral may or may
not be zero.
For example, given C to be the unit circle centered at the origin and
z
z f
1
) ( =
, then f is
analytic at all points except at the origin, and indeed the integral is not zero. In fact,



i dz
z
C
t 2
1
=
}

But
0
1
2
=
}
dz
z
C

In general one can show that
0
1
=
}
dz
z
C
n
if
1 = n
and
i dz
z
C
n
t 2
1
=
}
if n = 1
The proof of Cauchys Theorem in this section utilizes a Theorem from advanced calculus
called Greens Theorem.

Greens Theorem

Let R be a simple region in the xy-plane with a piecewise smooth boundary C oriented
counterclockwise. Let M(x, y) and N(x, y) be functions of two variables having
continuous partial derivatives on R. Then


}
= +
C
dy y x N dx y x M ) , ( ) , (
dxdy
y
M
x
N
R
) (
c
c

c
c
}}

Where C is oriented in a counterclockwise.


100


Theorem (preliminary Version of Cauchys Theorem)
Suppose that f is analytic with the derivative
dz
df
continuous on and inside a simple closed
curve C, then

0 ) ( =
}
C
dz z f


Consider a function ( ) ( ) ( ) y x iv y x u z f , , + = analytic at all points within and on the closed
contour C and is such that ) (
1
z f is continuous there. Then u and v and their partial
derivatives of the first order are continuous there, and consequently
( ) ,
2
2
2
2
}} } |
|
.
|

\
|
=
R C
dxdy
y
u
x
v
vdy udx
( ) .
2
2
2
2
}} } |
|
.
|

\
|
=
R C
dxdy
y
v
x
u
udy vdx
Since the Cauchy Riemann conditions are satisfied on R , the integrands of the two double
integrals vanish through out the region R . Since the line integrals on the left are the real and
imaginary coefficients of the complex number representing the line integral of ), (z f if
follows that
. 0 ) ( =
}
dz z f
C

Cauchy Goursat Theorem: If a function f is a analytic at all points interior to and on a
closed contour C , then
. 0 ) ( =
}
dz z f
C

(Proof Exercise)
Remark: The converse of the Cauchy Goursat Theorem is not necessarily true.
Give an example to verify it.




101

Theorem: (Independent of Path)
Suppose that
i) R is a simply connected region,
ii)
1
z and
2
z are points on R,
iii) ) (z f is analytic on R.
Then for everyC , joining
1
z and
2
z within R, the value of the integral
}
2
1
) (
z
z
dz z f is the
same. If
2 1
z z = then by Cauchy Goursat Theorem . 0 ) (
2
1
=
}
z
z
dz z f

Example 1. Evaluate ( )
}

C
dz z z 2 3
2
where
2 1
C C C + = as in the figure.
Let 0 , 1 1 0 :
3
= s s . = dy y C
( )
}

=
1
1
2
2 3 dx x x
( ) ( ) ( ) 2 1 1 1 1
1
1
2 3
= = =

x x
Theorem: Suppose that f is analytic on a simple connected region R . If
0
z is a point if R
then ( ) ( ).
0
z f d f
dz
d
z
z
= |
.
|

\
|
}
t t
Proof: Let ( ) ( )
}
=
z
z
d f z f
0
t t and z z A + be in R .

( ) ( ) ( )
( ) ( )
(
(

A
=
A
A +
=
A
A
} }
0
0
1
z
z
z
d f d f
z z
z F z z F
z
z f
t t t t
( ) ( ) ( ) ( )
} } }
A + A + A +
A
+
A
=
A
=
z z
z
z z
z
z z
z
d f
z
d z f f
z
d f
z
t t t t t t
1
) (
1 1

( ) ( ) ( ) z f d z f f
z
z z
z
+
A
=
}
A +
t t ) (
1





( ) ( )
} }
=
3
2 3 2 3
2 2
C C
dz z z dz z z

102



( )
( ) ( ) ( ) ( ) z f R z d z f f
z
z f
z
z F z z
z
e
A
=
A
A
=
}
A +
, ) (
1
t t is analytic f is
continuous. Thus 7 0 0 > - > e o for any t satisfying o t < = A z z we
have ( ) ( ) , e < z f f t and ( ) ( ) t t d z f f
z z
z
}
A +
) ( is independent of the path.

Hence with in the ( ) o , z B nbhd taking the path of integration the line segment joining
z and z z z z L z z A = A + = A + , with e = M we have

( )
( ) ( )
}
A +

A
=
A
A
=
z z
z
d z f f
z
z f
z
z F
t t ) (
1

e = e A
A
< .
1
z
z


( )
( ) e =
A
A
z f
z
z F
whenever . o t > z

( )
( ) ( ) ( )
o z
R z z f z f z f
z
z F
A
e =
A
A
, lim
1


Hence ( ) ( ) z f d f
dz
d
z
z
= |
.
|

\
|

}
0
t t

Definition: The function ( ) ( ) |
.
|

\
|
=
}
z
z
d f z f
0
t t is called an antiderivative or indefinite
integral of . f







103



Theorem: (Fundamental Theorem of Integration)

Let ( ) z f be analytic on a simple connected region R . If ( ) z f is an antiderivative of f
In R , then for any , ,
2 1
R z z e
( ) ( ) ( )
1 2
2
1
f F f F dz z f
z
z
=
}

Proof: Let
0
z be in R , Then by the above Theorem,
( ) ( ) . ,
0
R z d f z F
z
z
e =
}
t t
( ) ( )
}
=
2
0
2
z
z
d f z F t t
( ) ( )
} }
+ =
2
1
1
0
z
z
z
z
d f d f t t t t
( ) ( )
}
+ =
2
1
1
z
z
d f f F t t
( ) ( ) ( )
1 2
2
1
z F z F d f
z
z
=
}
t t

Examples:
1. ( ) ( ) i i i z dz z
i
i
i
i
2 1 1 2
2 2 1 2
1
+ = + = =
+

}

2. e e e e dz e
i i ii z
i
z
2
1
0
1
0
1
= = =
+ + +
}
t t
t

3. t
t
t
cos cos cos sin + = =
}
i z dz z
i
i


( )
|
|
.
|

\
|
+
+
=
|
|
.
|

\
| +
=
|
|
.
|

\
|
+
=

1
2
1
2
2
1
2
2
2 2
e
e e e e e
i i i





104

Theorem: (Cauchy Goursat for Multiple Connected Regions)
Let C be a closed path and n j C , ... , 2 , 1 ,
1
= be a finite number of closed paths in the
interior of C having no points common. Let ( ) | | C R int c = If ( ) z f is analytic in R
then ( ) ( )
} }
I
=
C
dz z f dz z f
Provided that C and Ihave the same orientation.
Proof: Orient S
C
R as follows.

Let smooth curves
0
r be connecting C
and 1 ... , 2 , 1 ,
1
= n k for r C
k
connecting
k
C to 1 ..., 2 , 1 ,
1
=
+
n fork C
k
and
n
r connecting
n
C toC

As shown in the figure. Denote n j Cj Cj Cj C C C ... , 2 , 1 , ,
' ' ' ' ' '
= + = + =
( ) ( )
2 1
'
1 0 1
' C r C r C + + + + = I
( )
n n
Cn r I + + + +

'
1
...
( ) ( )
0
' '
1 1
'
0
' '
2
... r C r C r C
n
+ + + + = I


Then ( )
n
C C C C C + + + + I + I ...
2 1
' ' '
2 1

1
I and
2
I are closed paths and f is analytic on
1
I and ) ( int ,
1 2
I I and ) ( int
2
I .

Hence by Cauchy Goursat Theorem
}
I
=
1
0 ) ( dz z f and
}
I
=
2
0 ) ( dz z f
} } }
I I + I
= = dz z f dz z f dz z f
C
) ( ) ( 0 ) (
2 1








105


Note: From the above Theorem if f is analytic on the shaded region shown below then

( ) ( )
} }
=
1
C C
dz z f dz z f



Example: Evaluate
}

C
z
dz
1
Where C is the following rectangular path.








Construct a circle
2
1
1 :
1
= z C
Let R be the multiply connected region bounded by C and
1
C . Clearly

1
1
) (

=
z
z f
Is analytic on R . Then
} }

=

1
1
1
1
C C
dz
z
dz
z
dz

t i
e z
2
1
1+ =

}
=
t 2
0
.
2
1
2
1
1
dt i e
e
t i
t i
dt i e z d
t i
2
1
2
1
= =

}
= =
t
t
2
0
2 i dt i



106

3. Evaluate dz
i z z
}
|
.
|

\
|
+
+
+ 2
3
1
4
were 4 : = z C








Singularities of ) (x f are 1 and i 2 4
Let 1 1 :
1
= + z C and 1 2 :
2
= + i z C
Then f is analytic on a closed multiply connected region determined by
1
C and
2
c
Hence dz
i z z
C
}
|
.
|

\
|
+
+
+ 2
3
1
4


} } } }
+
+
+
+
+
+
+
=
1 1
2
2
2
1
3
1
1
4
2
1
3
1
1
4
C C
C
C
dt
i z
z
dt
i z
dt
z
= ( ) ( ) i i i t t t 14 2 3 ) 0 ( 4 ) 0 ( 3 2 4 = + + +

4. Evaluate dz
i z z
z
C
}
+ 2
3
2
where C: 5 = z

2
3
2
+ + zi z
z
i z i z +
+

=
1
2
2
(by fartial fraction decomposition)

i z
B
i z
A
2
+
+
=



C
2
-2i
-1
C
1
C
4i
-4
-4i
4

107


The singularities of f are i and i 2


Let n i z C = 2 :
1

1 :
2
= + i z C


f is analytic on the closed multiply connected region R determined by
1
, C C and
2
C
} } } } }
+
+

+
+
+ +

=
+
2 1 2 1
1
2
1
2
2
1
2
3
2
C C C C C
dt
i z z i z
dz
dt
i z
dz
zi z
z

( ) i i i t t t 6 2 ) 0 ( 2 2 = + + =

4.7. The Cauchy Integral Formulas

Theorem: Let ( ) z f be analytic on positively oriented simple closed curve C and in Int ( ) C
( )
( )
}

=
C
dz
z z
z f
i
z f
0
0
2
1
t

This formula is called the Cauchy Integral formula

Proof: Consider the following multiply connected region.






1
-r
0
z
0
C

C
2
-2i
-1
C
1 C
108

Let 0 :
0 0 0
> = r z z C Int ( ) C and positively
Oriented.
Then
( ) ( )
} }

=

C C
dz
z z
z f
dz
z z
z f
0
0 0


( ) ( )
} }

+

=
C C
dz
z z
z f
dz
z z
z f z f
0
0
0
0
0
) (

But
( )
( ) ( ) i z f dz
z z
z f dz
z z
z f
C C
t 2
1
0
0
0
0
0
0 0
= =

} }
and f is analytic f is cont at
0
z
( ) ( ) e < - > > e
0
0 z f z f fo
Whenever o <
0
z z
Take,
2
, 0
o
o < < and o < I
0
: z z positive oriented
1
o C can be chosen go that I is in
Int ( )
0
C then z in I

( ) ( ) ( ) ( )
} }
I

dz
z z
z f z f
dz
z z
z f z f
C 0
0
0
0
0



( ) ( )
=
e
s

s
} }
I I
dz
d
dt dz
z z
z f z f
1 0
0

e = =
e
t t 2 2
1
i
d

Since 0 > e is oribitrary we have

( ) ( )
0
0
0
0
=

}
C
dz
z z
z f z f

Hence
( )
( ) i z f dz
z z
dz z f
C
t 2
0
0
=

}

( )
( )
}

= dz
z z
z f
i
z f
0
0
2
1
t


109
Example 1: Evaluate
}

C
dz
i z
z cas
when 2 : = z C positive oriented
Solution: ( ) z z f cos = is an in tire function i z =
0
is in Int ) (C
By the Cauchy Integral formula cos . 2
cos
i dz
i z
t
t =

}


( )
( )
( )
e
e
i e e i
ei ei
i
2
1
2
2
2
1
2 2
+
= + =

t t t
Example 2: Evaluate
}

dz
i z
z z
2
2
were 3 : = i z C
Solution ( ) z z z f =
2
is an in tire function and
0
z is i 2 is in Integral ) (C
By the Cauchy Integral
( ) ( ) i i i f i dz
i z
z z
=

}
4 2 2 2
2
2
t t
i t t 8 4 =
Example 3: Evaluate
( )
}
+
dz
z z 2
1
where 2 : + z C oriented negatively
Solution: Let
1 0
C C

= the
0
C is positively oriented

( ) ( )
} }
+
=
+
0
2
1
2
1
C C
dz
z z
dz
z z


}
+

=
0
2
1
1
C
dz
z
z

Hence ( )
2
1
= z f is analytical on C and in
Int ) (C and 2
0
= z is in integral ) (
0
C
( )
i i dz
z
z
dz
z z
C C
t t + = |
.
|

\
|

=
+
=
+

} }
2
1
2
2
1
2
1
0



-2i
C
C
0

110

4.8. Derivative of Analytic Function
Suppose that f is analytic on a positively oriented simple closed path C and integral of C
with
0
z in integral , then by Cauchy Integral formula ( )
( )
}

=
C
dz
z z
z f
i
z f
0
0
2
1
t
have need
0
t
show that f is diffble at
0
z and
( )
( )
( )
}

=
C
dz
z z
z f
i
z f
2
0
0
'
2
1
t


( )
( )dz z f
z z z zi z z i z
z z
f
|
|
.
|

\
|

A A
=
A
A +
}
0
0
1 1
2
1
t


( )
( )
|
|
.
|

\
|

A
=
}
dz
z z
z z z
z f
i
0
0
2
1
t


( )
( )
( )
( )( )
}
A
A
+

= dz
z z z z z
z f
i
z
dz
z z
z f
i
2
0 0
2
0
2 2
1
t t

Since f is analytic on C and it is bounded there. Hence ( ) 0 , > s M M z f on C (let L be
the length of C)
Consider a circle
0 0
: r z z I positively oriented and lying entirely in C . Choose z A so
small so that z z A +
0
is in interior of ( ) I and
0
r z < A
Then z r z z z z z z A = A > A
0 0 0

Thus
( )
( )( )
( )
( )
} }
A
A
s
A
A
s
A
A
C
r r
z r
m
z
dz
z z z z z
z f z
dz
z z z z z
z f
i
z
0
2
0
0
2
0 0 0
2
0 0 0
2 . .
2
.
2 2
t
t t t

As,t he limit 0 0 A z as
Hence ( )
( )
( )
}

=
C
dz
z z
z f
i
z f
2
0
0
'
2
1
t

Taking the derivative of (*) at
0
z we obtain ( )
( )
( )
}

=
C
dz
z z
z f
i
z f
3
0
0
' '
2
! 2
t


111

By induction if follows that
( )
( )
( )
} +

=
C
n
n
dz
z z
z f
i
n
z f
1
0
0
) (
2
!
t

Then If f is analytic at
0
z , then its derivatives
( )
( ) N e n z f
n
, , are also analytic at
0
z
If f in analytic on a region R , then
' ' '
, f f , ., are also analytic on R .

Generalized Cauchy Integral Formula
Suppose that
a) f is analytic on positively oriented simple closed path C and in the interior of C.
b)
0
z any point in the interior of C then,
( )
( )
0
z f
n
exists ,... 2 , 1 , 0 = n and
( )
( )
( )
( )
} +

= dz
z z
z f
i
n
z f
n
n
1
0
0
2
!
t

Example Evaluate
( )
}
+
+
dz
i z
e z
z
3
3
t
when t 2 0 5 : t e z C
it
s =
Solution: By the above theorem,
z
e z z f + =
3
) ( , t =
0
z and 2 3 1 = = + n n
Now since ( ) z f is an entire function and e = i z t
0
Int ) (C
the hypothesis of the above theorem are satisfied
thus using the theorem
( )
( )
( ) i
i
dz
i z
z f
C
t
t
t
=
+
}
! 2
2
3

( )
2 2 '
3 e z z f + =
( )
2 ' '
6 e z z f + =
( )
i
e i i f
t
t t

+ = 6
' '

( ) ( ) t t
t

e i e
i
sin cos
2 6 = i t
( ) i i i t t t t =
2
6 1 6

112

Example 2: Evaluate
( )
}

C
z z
z d
4 3
2
where 2 3 : = z C is positively oriented.
the only singularity of the integrand which lies in Int ) (C is 2
0
= z








Thus we take ( )
4
1
z
z f = is analytic on in Int ) (C , 2
0
= z is Int ) (C
2 3 1 = = + n n

( ) ( )
( )
} }
=

i
f i
dz
z
z
z z
t
n
C
2
2 2
2
1
2
3
4
4 3
t o


( ) ( )
5 ' 4
4
4 ,
1

= = = z z f z
z
z f
( )
6
6 ' '
20
20
z
z z f = =


( ) i
i
z f t
t
16
5
16
5
2
2
16
5
64
20
' '
= |
.
|

\
|
= =
4. Evaluate
( )
}

C
z z
t d
4 3
2
where 3 1 : = z C positive oriented.




3

1
1
C
1
C
2
1

113

Solution we will use the multiple an the
isolated singular 2 0
0 0
= = z and z which are
in Int ) (C

Let
2
1
1
= = z C and
2
1
2
2
= z C positive oriented
1
C and
2
C in Int ) (C
Then
( ) ( ) ( )
} } }

+

2 1
4 3 4 3 4 3
2 2 2
C C C
z z
t d
z z
t d
z z
t d



( ) ( )
} }
=

=
2 1
4
3
4
3
2
1
2
1
C C
dz
z
z
dz
z
z

( )
( )
2
2
1

=
z
z f ( )
4
1
z
z f =
2
0
= z in Int ) (C
0
0
= z in Int 3 ) (
1
= n C analytic is analytic on
1
C in Int ) (
1
C
( )
( )
i
g i
fm
i
i
t
t t
6
5
2
2 2
0
3
2
' '
+ =
( ) ( )
3
2

= f z f
( ) ( )
4 '
2 3

= z z f
( ) ( )
5 ' '
2 12

= z z f
( ) ( )
6 ' ' '
2 60

= z z f

16
15
64
60
0
' ' '

=

f
0
16
5
16
15
=

i i t t




114

Some Important Theorems
As a result of the Cauchy integral formula and other properties of analyticity
we have the following results:
Theorem 1: Moreras Theorem
(Partial converse of Cauchy-Goursat Theorem)
If ) (z f is continuous on a simple connected region R and ( )
}
=
C
dz z f 0 along any simple
closed curve R, then the function ) (z f is analytic in R
Proof: If ( )
}
=
C
dz z f 0 independent of C , then it follows that,
( )
}
z
z
d f
0
t t in independent of the path joining
0
z and z as long as
0
z and z are in R
Then ( ) z F is analytic and ( ) ( ) z f z F =
'

so ( ) z F
'
is analytic, if F is analytic on C
Hence F is analytic in R
Theorem 2: Cauchys Inequalities
If ) (z f is analytic on and inside a circle r z z C =
0
: , which is oriented in the positive
sense and M z f s ) ( , then
n
n
r
Mn
z f
!
) (
0
) (
s , for ,... 3 , 2 , 1 , 0 = n
Proof : By Cauchy integral formula we have
( )
( )
( )
}
=

=
+
, 2 , 1 , 0
2
!
1
0
0
n dz
z z
z f
i
n
z f
n
n
t

Since r z z =
0
and length r C of L t 2 =
( )
( )
( )
r
r
M n
dz
z z
z f n
z z
n n
n
t
t t
2
2
!
2
!
1 1
0
0
= s

=
+ + }

n
r
n M.
=


115


Theorem 3: (Liouvilles Theorem)
If for all z in a complex plane C if f(z) is analytic and ( ) z f is bounded that
is ( ) M z f f suchthat M s > - , , 0 , then ( ) z f must be constant function
Proof: Letting 1 = n in Cauchys inequities and rapacity
0
z by z we has
( ) s r and
r
M
z f , 0
'
,this implies that,
0 ) (
'
= z f
f(z) = 0
f(z) is a constant.

Theorem 4: Fundamental Theorem of Algebra
Every polynomial equation ( ) 0 = z p where
0 1
1
. 1 ) ( a z a n z an z an z P
n n
+ + + =

with is
are complete constants 0 + an has at least one root
Proof. If ( ) 0 = z p has no roots, then ( )
( ) z p
z f
1
= in analytic for all z and
( ) z an z f 0
f is analytic and bounded z
f must be a constant function by Liouvilles theorem
the reciprocal of any polynomial function is a constant, which is false.
Hence ( ) 0 = z p for at least one z

Theorem 5: Every polynomial equation ( ) 0 = z p (where the degree of P(z) is n) has
exactly n-roots
Proof: By fundamental then ( ) 0
0
= z p there
( ) ( ) | |
0 0 1
1
0 0 0
1
0
4 ... 1 tan ... 1 tan a z a z an z an a z z an z p z p
n n n z n n
+ + + + + =


( ) ( ) ( )
0
1
0
1
0 0
, ... z z a z z an z z an
n n n n
+ + + =


116

( ) ( ) z p z z = where ( ) z Q is a polynomial of degree ( ) ( ) ( ) z Q z z z p n
0
1 = applying
the fundamental than of again we see that ( ) 0 = z Q has at least one root, say
1
z and
( ) ( )( ) ( ) z R z z z z z p
1 0
= we here ( ) z F is a polynomial of degree n-2 containing in this
manner we see that 0 ) ( = z p has exactly n roots.

Theorem 6: Gausss Meon Value Theorem
If ( ) z f is analytic on and inside a circle r z z C =
0
: positively oriented then, ( ) z f is
meon of the values of ( ) z f on the circumference of C.
i.e.
( ) ( )dt re z f z f
t i
}
+ =
t
t
2
0
0 0
2
1

Proof. By Cauchy integral formula we have the following
( )
( )
( )
t i
re z z C
z z
dz z f
i
z f =

=
}
0
0
0
:
2
1
t

Or
t i
re z z + =
0

t 2 0 s s t
dt e r i dt
t i
=
( )
}
+
=
t
t
2
0
0
2
1
dt re i
re
re z f
i
t i
t i
t i

( ) dt re z f
i
t i
}
+ =
t
t
2
0
0
2
1

Hence ( ) ( ) dt re z f z f
t i
}
+ =
t
t
2
0
0 0
2
1

Solution: Hence ( )
6
, sin
0
2
t
= = z z z f

}
= |
.
|

\
|
= |
.
|

\
|
+
t
t t
t
2
0
2 2
4
1
6
sin 2
6
sin
2
1
dt e
t i



117

Theorem 7: (Maximum Modulus Theorem)
If ( ) z f is analytic on and inside a simple closed curveC , then the maximum value of
( ) z f occurs on C unless ( ) z f is a constant. It will further be shown that if for some point
z of the interior, ( ) z f = M, the function is necessarily a constant in the entire domain.

Proof: Let, if possible, ( ) z f attain the upper bound M at a point a of the interior so that
we have ( ) a f = M.
Let D be a circular domain with center a such that D along with its boundary circle is
contained in D. Then D is a neighbourhood of a. Then either ( ) z f = M at every point z of
D or there exists some point z
0
of Dsuch that
( ) M z f =
0
, so that ( ) c = M z f
0
, where c is some positive number.
By the virtue of the continuity of ( ) z f at z
0
, there exists a neighbourhood of z
0
for every
point z of which

2
) ( ) (
0
c
< z f z f

2 2
) ( ) (
0
c c
= + < M z f z f
Thus we can construct a circle with its center a and passing through z0 such that for points
z on a part of this circle

2
) (
c
< M z f
Let C
1
denote this part and C
2
the remainder of this circle,
By Gauss Mean value Theorem, we have


} } }
+ = =
1 2
) ) ( ) ( (
1
) (
1
) (
C C C
dz z f dz z f
l
dz z f
l
a f

l
l
M Ml l M
l
a f
2
] )
2
[(
1
) (
1
2 1
c c
= + s , where l
1
and l
2,
l are the lengths of
the parts C
1
and C
2
and C respectively

, 2
1
l
l
M M
c
s which is a contradiction.
Thus ( ) z f = M at every point z of the circular domain D which is a neighbourhood of the
point a.
It has been showen that if a is a point of D such that ( ) a f = M, then there exists a
neighbourhood of a such that for every point z of the same, we have ( ) z f = M

118
Thus the set D
1
of points z of D for which ( ) z f = M is an open set. Also we may easily see
that the set D
2
of point z of D for which ( ) z f = M is an open set.
Then if D2 is empty, we see that ( ) z f = M for every point of D and hence f(z) is a constant
in D. If however, D2 is nonempty, we have D = D1 U D2, where D1 and D2 are nonempty,
disjoint, open sets. However as D is a connected set, such decomposition is not possible
and we arrive at a contradiction. We have thus proved the theorem.


Theorem 8: (Minimum Modulus Theorem)
If ( ) z f is analytic on and inside a simple curve C and ( ) 0 = z f inside C , then the
minimum value of ( ) z f occurs on C
Proof: f(z) is analytic within and on C and ( ) 0 = z f inside C. Therefore,
) (
1
z f
is analytic
within C. Hence by maximum modulus principle
) (
1
z f
attains its maximum value on the
boundary of C, so that ( ) z f attains its minimum value on the boundary of C.























119


CHAPTER 5: INFINITE SERIES: POWER SERIES

Introduction: This chapter is devoted to the consideration of power series. It will also deal
with those of the parts of the subject of the convergence of sequences and infinite series
which are necessary for the purpose of this chapter.

Convergent Sequences

Activity
1. What is a sequence?
2. Define a real sequence.
3. Try to define a complex sequence in a similar way.

Recall that a sequence is a function defined on the set of nonnegative integers. For any
arbitrary sequence, the value associated to the nonnegative integer n is usually denoted by
the symbol z
n
and the sequence itself by the symbol {z
n
}.

Definition: A sequence {z
n
}is said to be convergent, if there exist a number L and to
every
c
> 0, there corresponds a positive integer m such that

, c < L z
n

m n >

In this case L is said to be the limit of the sequence, and we write
L z
n n
=

lim
. We may also
say that the sequence {z
n
}tends to L or converges to L.

A sequence which is not convergent will be referred to as Divergent. Analogous to the
case of real sequence, it may be easily shown that a convergent complex sequence cannot
tend to two different limits.

Suppose z
n
= x
n
+ iy
n
and
L z
n n
=

lim
where L = L
1
+ iL
2
, then we have

2 1 2 1 2 1
) ( ) ( ) ( ) ( L y L x iL iy L x iL L iy x L z
n n n n n n n
+ s + = + + =

Also we have

L z L x
n n
s
1
and
L z L y
n n
s
2


Thus we deduce that
L z
n n
=

lim
= L
1
+iL
2
implies that
1
lim
L x
n n
=

and
2
lim
L y
n n
=


Again
1
lim
L x
n n
=

and
2
lim
L y
n n
=

implies that
L z
n n
=

lim

Hence we have the following Theorem.

Theorem: A sequence {z
n
}is convergent iff the two real sequences {Re(z
n
)}and
{Im(z
n
)}are both convergent. Hence in the event of convergence,


)] Im( ) [Re(
lim lim
n n n n n
z i z z + =

=
+

) Re(
lim
n n
z ) Im(
lim
n n
z i


120


Theorem: (Cauchys General Principle of Convergence)
A sequence {z
n
}is convergent if and only if for every
c
> 0, there corresponds a positive
integer m such that

c <
+ n p n
z z

m n >
and
0 > p

Cauchy Sequence: A sequence {z
n
} is said to be a Cauchy (or Fundamental) sequence if
for every
c
> 0, there exists a positive n
o
such that


c <
n m
z z

0
, n m n >


Theorem: Every convergent sequence is a Cauchy sequence.
Proof: Suppose {
n
z
}converges to L, then for every
c
> 0, there exists a
positive n
o
such that

2
c
< L z
n

0
n n >

Hence if
0
n n >
and
0
n m >
, then
2
c
< L z
n
and
2
c
< L z
m

Therefore
c
c c
= + < + s + =
2 2
z
n m n m m n
z L L z z L L z z

Hence {
n
z
} is a Cauchy sequence.

Cauchys Criterion for Convergence of Sequence
A sequence in a complex plane C is convergent if it is a Cauchy sequence.

Infinite Series

Let {z
n
} denotes sequence of complex variables. The infinite sum z
1
+z
2
+z
3
+ . . .
denoted by

=1 n
n
z
is called an infinite series.
We now set up a sequence {s
n
} of partial sums where s
n
= z
1
+z
2
+z
3
++z
n

Then the series is said to be convergent if the sequence {s
n
} is convergent, and, in the
event of convergence,
n n
s
lim

is called the sum of the series

=1 n
n
z
.
In this case we write as

=1 n
n
z
=
n n
s
lim







121


Exercise

(i) Given convergent sequences{z
n
} and {t
n
}, show that the following are convergent:
1. {z
n
+t
n
}
2. {z
n
-t
n
}
3. {z
n
t
n
}
4. {az
n
}for any complex number a.
5.
} {
} {
n
n
t
z
provided that
0
lim
=
n n
t

(ii) Show that a series

=1 n
n
z
is convergent if and only if the two real terms (the real and
imaginary parts) of the series converges.
(iii) Given convergent series

=1 n
n
z
and

=1 n
n
w
, show that
1.
) (
1
n
n
n
w z +

= is convergent.
2.
) (
1

=

n
n n
w z
is convergent.
3.
0
lim
=
n n
z
, that means the general term of a convergent series tends to zero.

Cauchys General Principle of Convergence of a Series
Analogous to the principle of convergence for a sequence, we have the following for a
series.

The necessary and sufficient condition for

=1 n
n
z
to be convergent is that to a given positive
number
c
, there corresponds a positive integer m such that


c < + + + +
+ + + + p n n n n
z z z z ...
3 2 1

m n >
and
0 > p

The proof is left as exercise for students as it directly follows immediately from the
corresponding theorem for sequences.

Absolute Convergence of a Series
A series

=1 n
n
z
is said to be absolutely convergent if the real nonnegative term series

=1 n
n
z
is convergent.

122


We shall now prove that an absolutely convergent series is also convergent.
Let

=1 n
n
z
be an absolutely convergent series and let
0 > c
be given, then there exist a
positive integer m such that

c < + + + +
+ + + + p n n n n
z z z z ...
3 2 1

0 , > > p m n

But from triangle inequality we have

s + + + +
+ + + + p n n n n
z z z z ...
3 2 1 p n n n n
z z z z
+ + + +
+ + + + ...
3 2 1

p n ,

Thus

c < + + + +
+ + + + p n n n n
z z z z ...
3 2 1

0 , > > p m n

Hence

=1 n
n
z
is convergent.
Now we can state this result in the following form.
A sufficient condition for the convergence of the series

=1 n
n
z
is that the series

=1 n
n
z
is
converges.

Note:

=1 n
n
z
being a real nonnegative term series, its convergence can be examined by
means of the usual tests for a positive term series developed in real variables.
A series which is convergent without being absolutely convergent is called a conditionally
convergent series.
We can easily show by comparison test that if a series is absolutely convergent, then the
two corresponding real term series (the real and imaginary parts) are also absolutely
convergent.

Tests for the absolute convergence of a complex term series is analogous to tests for a
positive term series and can be restated as one for the absolute convergence of a complex
term series.
A series

=1 n
n
z
is absolutely convergent, if
. 1
1 lim
<
+

n
n
n
z
z

The geometric series

=0 n
n
z
converges if and only if
. 1 < z

Also the convergence is absolute, since

=0 n
n
z
being a real term geometric series with
common ratio
z
is convergent if and only if
1 < z
.
Thus the series is absolutely convergent when
. 1 < z

123
The geometric series divergent for
, 1 > z
as the general term does not tend to 0.

Power Series

So far we have been concerned with series with constant terms only. We shall now
consider a special type of series known as Power series, with variable terms.
A series of the type

=

0
0
) (
n
n
n
z z c
is called a power series; where z is a complex variable;
,... ,...., , ,
2 1 0 n
c c c c
are complex constants called coefficients of the series; and
0
z
a fixed
arbitrary complex number called the center of the series. Its convergence can be examined
by means of the usual tests for a positive term series developed in real variables, such as
ratio test, root test, comparison test and the like.

Examples: Test the following for convergence.

1.

=0
!
n
n
n
z
;
1 0
1
]
! )! 1 (
[
lim
1
lim
< =
+
=
+

+

n
z
n
z
n
z
n
n n
n

Hence by ratio test the series converges absolutely for every z.
2.

=0 n
n
z
; By similar test one can show that the geometric series

=0 n
n
z
converges
Absolutely, for
1 < z
and diverges for
1 > z

3.

=0
!
n
n
z n
; By the same test one can show that the series

=0
!
n
n
z n
converges
only for z = 0 and diverges for other values of z.

From these examples we can deduce that a power series

=

0
0
) (
n
n
n
z z c
is either
convergent for all values of z, or converges only for z =
0
z
or converge for z in some
regions of the complex plane.

Any power series of the form

=

0
0
) (
n
n
n
z z c

can be converted to the form

=0 n
n
n
w c
by putting w = z-
0
z
.


124


Theorem: (Abels Theorem)
If a power series

=0 n
n
n
w c
converges for a particular value
0
w
of w, then it converges for
all values of w for which
0
w w <
.
Proof: Suppose that

=0
0
n
n
n
w c
converges, then
0
0
lim
=

n
n n
w c


Hence we can find a real number M > 0 such that
M w c
n
n
s
0
for all values of n.
Thus
0
w w <
implies that
n
n
n
n
w
w
M w c
0
s

But as
0
w w <
the geometric series;
n
n
w
w

=0 0
is convergent.
Therefore by comparison test, the series

=0 n
n
n
w c
is absolutely convergent for all values w
for which
0
w w <
.

Theorem: (Cauchy-Hadamard Theorem).
For any power series

=0 n
n
n
w c
exactly one of the following holds true.
(i) The series diverges for every value of w other than 0.
(ii) The series converges for every value of w.
(iii) There exists a positive number r such that the series converges for every w such
that
, r w <
and diverges for every z such that
. r w >

Note: Associated with each power series,

=0 n
n
n
w c
there exists a unique number r called the
radius of convergence of the series having the following properties
a) If r = 0 the power series converges only at w = 0, and diverges for all other w.
(The case of Cauchy-Hadamard Theorem) (i).
b) If 0 < r < , then the power series converges absolutely for all w such that
r w <

and diverges for all w such that
r w >

(The case of Cauchy-Hadamard Theorem) (ii).
c) If r =, then the power series is convergent for all values of w in the complex plane.
(the case of Cauchy-Hadamard Theorem) (iii).


125



The circle
r w =
is called the circle of convergence.

For example the radius of convergence of the power series in the previous example,
1)

=0
!
n
n
n
z
is r = and its circle of convergence is
= z
.
2)

=0 n
n
z
is r = 1 and its circle of convergence is
1 = z

3)

=0
!
n
n
z n
is r = 0 and its circle of convergence is the point circle
0 = z


Exercise

Find the radius of convergence and circle of convergence for the following power series:
1.

=1
3
n
n
n
z
2.

=
+
0
) 6 ( !
n
n
i z n
3.

0
)! 2 (
) (
n
n
n
i z


4.

=1 n
n
n
n
z
5.

=1
!
n
n
n
n
z n
6.

=
+
1
2
) 1 (
n
n
n
z
n
n


7.

=
+

1
) 2 (
) 1 (
n
i
n
n
iz
8.

=
+ +
1
) 1 (
) 2 (
n
n
n
i z
i
9.

+
+
1
3
1
4 ) 1 (
) 2 (
n
n
n
n
z




Differentiation and Integration of a Power Series
Suppose

=0 n
n
n
z c
is a power series which converges to f(z) in its circle of convergence, that
is, f(z) =

=0 n
n
n
z c
, then term by term differentiation yields that = ) (
'
z f

0
1
n
n
n
z nc
and as
in the case of real series it can be shown that both of them have the same radius of
convergence. In a similar fashion one can find the higher derivatives of the power series
having the same radius of convergence. Hence we can state the following theorem.

126
Theorem: A power series

=0 n
n
n
z c
which converges to f(z) in its circle of convergence,
that is, f(z) =

=0 n
n
n
z c
represents an analytic function at every point inside the circle of
convergence of the power series.

Further, every power series possesses derivatives of all orders within its circle of
convergence and these derivatives are obtained through term by term differentiation of the
series.

In other words, every power series represents an analytic function inside its circle of
convergence.
Thus, through term by term differentiation and putting z = 0 we get
!
) 0 (
) (
n
f
c
n
n
= , if instead
of the series

=0 n
n
n
z c
, we consider the series

=

0
0
) (
n
n
n
z z c
with radius of convergence r,
then it converges in the domain .
0
r z z <
Moreover,
!
) (
0
) (
n
z f
c
n
n
= .

Note: The converse of the theorem above will be proved in the next section. There it will be
shown that if f(z) is analytic in a certain circle with its center at
0
z , then f(z) can be
expressed as the sum of power series in terms of ). (
0
z z



Development of Analytic Functions as Power Series
Taylors Theorem for Complex Functions

Theorem: If f(z) is analytic in a circular domain D with center
0
z , then for every z in D,

) (
!
) (
...
!
) (
) ( ...
! 2
) (
) ( ) ( ) ( ) ( ) (
0
) (
1
0 0
) (
0
0
) 2 (
2
0 0
'
0 0
z f
n
z z
n
z f
z z
z f
z z z f z z z f z f
n
n
n n
n

= + + + + + =
So that f(z) can be expressed as a power series about
0
z .

Proof: Let C denote the boundary of the circular domain D and let z be any point in D.
Again let
0
C denote any circle whose center is
0
z and whose radius is smaller than that of C
and which contains z in its interior. Assume
0
C is positively oriented. By Cauchys Integral
Formula, we have
127

t
d
z
f
i
z f
C
}

=
0
) (
2
1
) (

|
|
|
|
.
|

\
|

+ +

=
|
|
|
|
.
|

\
|

=

=

0
0
0
0
2
0
2
0
0
0
0
0
0
0 0 0
1
1
) (
) (
...
) (
) (
1
1
1
1 1
) ( ) (
1 1
z
z z
z
z z
z
z z
z
z z
z
z
z z
z z z z z
n
n

=
z z
z z
z
z z
z
z z
z
z z
z
n
n
n
n

+ +

1
) (
) (
) (
) (
...
) (
) (
) (
1
0
0
0
1
0
3
0
2
0
2
0
0
0

Therefore,

t
d
z
f
i
z f
C
}

=
0
) (
2
1
) ( =

t
d
z
f
i
C
}

0
0
) (
2
1
+

t
d
z
f
i
z z
C
}

0
2
0
0
) (
) (
2
) (
+

t
d
z
f
i
z z
C
}

0
3
0
2
0
) (
) (
2
) (
+
..
+

t
d
z
f
i
z z
C
n
n
}

0
) (
) (
2
) (
0
1
0
+

t
d
z
f
z
z z
i
C
n
n
}

0
) (
) (
) (
2
1
0
0

=
n
n
n
R z f
n
z z
z f z z z f z z z f +

+ + + +

) (
)! 1 (
) (
..... ) ( ) ( ) ( ) ( ) (
0
) 1 (
1
0
0
' ' 2
0 0
'
0 0

Where
n
R =

t
d
z
f
z
z z
i
C
n
n
}

0
) (
) (
) (
2
1
0
0

For every on
0
C ,
0
0
z
z z

is a constant and less than one. We denote it by p.

Also if
0
denote the radius of
0
C , we have
0 0 0 0 0 0
) ( ) ( z z z z z z z z z = > =
Thus we have

n n
n
p
z z
M
z z
M
p R
0 0
0
0 0
0
) ( 2
2
1

=

s

t
t

Where M f s ) (
0
C zc
As 0 < p < 1, we see that 0
lim
=
n n
R
Thus we obtain
) (
!
) (
...
!
) (
) ( ...
! 2
) (
) ( ) ( ) ( ) ( ) (
0
) (
1
0 0
) (
0
0
) 2 (
2
0 0
'
0 0
z f
n
z z
n
z f
z z
z f
z z z f z z z f z f
n
n
n n
n

= + + + + + =
Corollary: If f(z) is analytic in any domain D and
0
z is any point of D, then

) (
!
) (
...
!
) (
) ( ...
! 2
) (
) ( ) ( ) ( ) ( ) (
0
) (
1
0 0
) (
0
0
) 2 (
2
0 0
'
0 0
z f
n
z z
n
z f
z z
z f
z z z f z z z f z f
n
n
n n
n

= + + + + + =
128
Where the equality holds for every point z in the largest circular domain whose center is
0
z and which is contained in the domain D.
Proof: (Exercise) as it is the immediate consequence of the above theorem.


Note: If the center of the circle C is chosen at the origin then writing
0
z = 0, we get
) 0 (
!
...
!
) 0 (
...
! 2
) 0 (
) 0 ( ) 0 ( ) (
) (
1
) ( ) 2 (
2 ' n
n
n n
n
f
n
z
n
f
z
f
z zf f z f

=
= + + + + + =
This series is known as Maclaurins series.
Remark: The power series expansion of analytic function is unique. (The proof is left as an
exercise).
Observations and examples; When it is known that f is analytic at all points within the
circle C, the convergence of Taylors series to f(z) is assured; no test for the convergence of
the series is required. The maximum radius of C is the distance from the point
0
z to the
singular point of f that is nearest to
0
z , since the function is to be analytic at all points
inside C.
As our first example of Maclaurins series expansion let
z
e z f = ) ( , then
z n
e z f = ) (
) (
and
1 ) 0 (
) (
=
n
f .since
z
e z f = ) ( is analytic for every value of z, then
1.
z
e z f = ) ( =

=0
!
n
n
n
z
when < z
Similarly, one can show that
2.
1 2
0
)! 1 2 (
) 1 (
sin
+

=
n
n
n
z
n
z when < z
3.
n
n
n
z
n
z
2
0
)! 2 (
) 1 (
cos

= when < z
4.
1 2
0
)! 1 2 (
1
sinh
+

+
=
n
n
z
n
z when < z
5.
n
n
z
n
z
2
0
)! 1 2 (
1
cosh

=
+
= when < z
6.
n
n
z
z

=
=

0
1
1
when 1 < z



Principle of Substitution
Suppose that

=
=
0
) (
n
n
n
z c z f and that the series converges < z z : and g(z) is a
function analytic through out the disk < z , then the function

=
=
0
)) ( ( )) ( (
n
n
n
z g c z g f and
the series converges < ) ( : z g z .
129




Example:
1. To express
2
z
e as a power series we use the expansion
z
e =

=0
!
n
n
n
z
when < z and
replace z by
2
z .
Thus we get

2
z
e =

=0
2
!
) (
n
n
n
z
=

=0
2
!
n
n
n
z
.
2.
To expand
z
e z f = ) ( in power series of (z-2i), first we write
z
e z f = ) ( =
i z i i i z
e e e e
2 2 2 2
=


and thus we replace z by z-2i in the expansion of
z
e =

=0
!
n
n
n
z
.
Hence we get
=
z
e
i z i i i z
e e e e
2 2 2 2
= =

0
2
!
) 2 (
n
n
i
n
i z
e .
3. The Maclaurins series for
z 9 3
1

can be obtained from

n
n
z
z

=
=

0
1
1
,
here we write
z 9 3
1

=
) 3 1 (
1
3
1
z
and replacing z by 3z we get,

z 9 3
1

=
3
1

=0
) 3 (
n
n
z =
n
n
n
z

0
1
3 .
4. We can expand also expressions like
3
z
e
z
in the region in which it is analytic.
Thus
3
z
e
z
=

=0
3
!
1
n
n
n
z
z
which can be written as

0
3
!
n
n
n
z
which has a negative exponent in the expansion.
Laurents Series

In example (4) above we get series with negative and nonnegative exponents such series
leads as to the introduction of a new series called Laurents series. Laurents series are
generalization of Taylors series in that they have a finite or infinite numbers of negative
exponents in addition to the nonnegative exponents of ( )
0
z z .

Theorem: Let denote any point on either of two concentric circles C
1
:
1 0
= z and
C
2
:
2 0
= z about a point z
0
, where
1 2
< . If f is analytic on C
1
and C
2
and
130
throughout the region between those two circles, then at each point z between them f(z) is
represented by a convergent series of negative and nonnegative exponents of ) (
0
z z as,







=

=

+ =
1
0
0
0
) (
) ( ) (
n
n
n n
n
n
z z
b
z z a z f
Where
} +

=
1
1
0
) (
) (
2
1
C
n
n
d
z
f
i
a

t
(n = 0, 1, 2, 3, )

t
d
z
f
i
b
C
n
n
} +

=
2
1
0
) (
) (
2
1
(n= 1, 2, 3, )
Each integral being taken counterclockwise; And this series is called Laurents series.
In case f is analytic at every point on and inside C
1
except the point z
0
itself, the radius
2

may be taken arbitrarily small. The expansion is then valid when
1 0
0 < < z z .
Remark:
(i) The open annulus
2

1 0
< < z is called the annulus of convergence of the
series.
(ii) If C is any positively oriented simple closed curve that lies in the annulus of
convergence we write the series in the form of

n
n
n
z z c ) (
0

=
where

t
d
z
f
i
c
C
n
n
} +

=
1
0
) (
) (
2
1
and n = 0,
,..... 3 , 2 , 1
(iv) The Laurents series of f(z) converges to f at any point on any closed set in the
annulus of convergence.
(v) Laurents series can be differentiated and integrated term by term within the
annulus of convergence.
(vi) If a Laurents series of a function on a given annulus exists, then it is unique.
(vii) If f is analytic at all points on and inside C
1
, then b
n
is zero for each n, since the
function is analytic inside and on C
2
; the integral is therefore has the value zero
and the series reduced to Taylors series.

The coefficients are usually found by other means than the above formula. For
example in the expansion of

3
z
e
z
=

=0
3
!
1
n
n
n
z
z
=

0
3
!
n
n
n
z
= ....
! 7 ! 6 ! 5 ! 4 ! 3
1
! 2
1 1 1
4 3 2
2 3
+ + + + + + + +
z z z z
z z z

b
n
= 0 for n > 3 and b
3
= 1, b
2
= 1, b
1
=
! 2
1
and an =
)! 3 (
1
+ n
for each n = 0, 1,
2, 3,
Examples: Find the Laurents series expansion for the following:
1.
z
z f
1
) ( = , with 1
0
= z and annulus of convergence < < 1 1 z
131
Solution: f is analytic in the given annulus, and thus we seek a series of (z-1) that
converges to f for all z in the annulus;
that is for 1 1 > z .

1 1 > z 1
1
1
<

z

Hence the
1
1
z
can be substituted for z in any series that converges for all z such that
. 1 < z
Therefore,

)
1
1
1 )( 1 (
1
) 1 ( 1
1 1
) (

+
=
+
= =
z
z
z z
z f
=
n
n
n
z z
z
z ) 1 (
1
) 1 (
1
1
)
1
1
1 (
1
1
1
0

=

=
1
0
) 1 (
1
) 1 (
+

n
n
n
z
is the Laurents series of the
function
z
z f
1
) ( = , with 1
0
= z .
2.
1
1
) (
2

=
z
z f in the annulus,
(a) 2 1 0 < < z
(b) < + < 1 2 z
Solution: (a)
1
1
) (
2

=
z
z f = )
1
1
)(
1
1
(
+ z z

= )
2 1
1
(
1
1
+ z z

= )
2
1
1
1
(
2
1
)
1
1
(

z
z

=
1
0
2
) 1 (
) 1 (
1
1
+

n
n
n
n
z
z

=

=
+

0
1
1
2
) 1 (
) 1 (
n
n
n
n
z

is the Laurents series of the function
1
1
) (
2

=
z
z f in the annulus
2 1 0 < < z
132
(b) In a similar way one can show that the Laurents series of
1
1
) (
2

=
z
z f
in the annulus of < + < 1 2 z is

=
+
+
0
2
) 1 (
2
n
n
n
z



Exercise: Find the series expansion of the following in the given region.
1. = ) (z f
9
z
e
z
, on < < z 0
2. = ) (z f )
1
cos(
z
, on < < z 0
3. = ) (z f
iz z
i z
+
+
2
2 5
, on 2 1 < < i z
4. f(z) =
) 1 (
1
2
z z
on (a) 1 0 < < z and (b) 1 > z





























133



CHAPTER 6: RESIDUE AND POLES

Singularities and Zeros of Analytic Functions

Recall that a point z
0
is a singularity point of the function f(z) provided that f fails to be
analytic at z
0
while every neighborhood of z
0
contains at least one point at which f is
analytic.

Definition: If a function is analytic at some point in every neighborhood of a point z
0

except at z
0
itself, then z
0
is called a singularity point, or a singularity of the function.

Example:
1. f(z) =
z
1
, z
0
= 0 is the only singularity point.
2. f(z) =
) 1 (
1
2 3
+
+
z z
z
, has three singular points: 0, i, -i.
3. f(z) =
) sin(
1
z
t
, has infinite number of singular points: ,....
3
1
,
2
1
, 1
4. f(z) = plog(z) has infinite number of singular points: 0 and any point on the negative real
axis.
5. f(z) =
2
z has no singular point since it is nowhere analytic.

Types of Singularities

From the above examples we observe that for some singular point z
0
every neighborhood of
it contains other singular points and for some other singular point z
0
we can find a deleted
neighborhood of z
0
through which f is analytic. The first type of singularity is said to be
non-isolated singularity and the latter one is isolated singularity. Thus we can make the
following distinctions.





Definition: Suppose z
0
is a singular point of the function f.
(i) If every neighborhood of z
0
contains at least one singular point of f other than
z
0
, then we z
0
is said to be non-isolated singular point.
(ii) If there exists a deleted neighborhood of z
0
throughout which f is analytic, then
z
0
is said to be isolated singularity.

134
Activity: Identify the singular points of the functions given in the above examples as
isolated or non-isolated.

Further isolated singularities are classified as: Removable singularities, Poles and Essential
singularities.
When z
0
is an isolated singular point of f, a positive number r exists such that the function
is analytic at each point z for which 0 <
0
z z < r. Thus in this domain the function is
represented by the Laurents series as;



=

=

+ =
1
0
0
0
) (
) ( ) (
n
n
n n
n
n
z z
b
z z a z f
Where
} +

=
C
n
n
d
z
f
i
a

t
1
0
) (
) (
2
1
(n = 0, 1, 2, 3, )

t
d
z
f
i
b
C
n
n
} +

=
1
0
) (
) (
2
1
(n= 1, 2, 3, ) (each integral being taken
counterclockwise).
In particular;
}
=
C
d f
i
b
t
) (
2
1
1
, where C is any closed contour around z
0
, oriented in
the positive sense and f is analytic on C and inside C except at z
0
itself.
Now if no negative exponents appear in the Laurents series, z
0
is called removable
singularity. If only a finite number of negative exponents of z-z
0
occurs with nonzero
coefficients appears in the Laurents series expansion, then z
0
is called a pole of order N.
(If N=1 it is called a simple pole). In this case Laurents Series is of the form,


0 ,
) (
......
) (
) ( ) (
0
2
0
2
0
1
0
0
=

+ +

+ =

=
N
N
N
n
n
n
b
z z
b
z z
b
z z
b
z z a z f
If there are an infinite number of negative exponents of z-z
0
with nonzero coefficients, then
z
0
is called an essential singularity point of f.









Examples:
1.
z
z
z f
sin
) ( = has a removable singularity at 0; since its Laurents series has no negative
exponent of z as given below.

135
......
! 7 ! 5 ! 3
1
sin
6 4 2
+ + =
z z z
z
z

2.
4
) (
z
e
z f
z
= has pole of order 4 at 0; since its Laurents series is given as:

4
) (
z
e
z f
z
= =

0
4
!
n
n
n
z

3. )
1
sin( ) (
z
z f = has an essential singularity at 0; as there are infinitely many negative
exponents of z in its Laurents series expansion as given below.
)
1
sin( ) (
z
z f = =
1 2
1
)! 1 2 (
) 1 (
+

n
n
z n


Suppose that f(z) has a pole of order N at z
0
, then its series expansion is given as;

0 ,
) (
......
) (
) ( ) (
0
2
0
2
0
1
0
0
=

+ +

+ =

=
N
N
N
n
n
n
b
z z
b
z z
b
z z
b
z z a z f
Hence for
0
z z = ,

=
+

+ + + + + =
0
0
1
0 1
2
0 2 0 1 0
) ( ) ( ..... ) ( ) ( ) ( ) (
n
N n
n
N
N N N
N
z z a z z b z z b z z b b z f z z w
hich is Taylors series. Therefore, the function g(z) = ) ( ) (
0
z f z z
N
is analytic and the pole
of has become a removable singularity of the function g, and 0 ) (
lim
0
= =
N z z
b z g . Thus we
have the following Theorem:

Theorem: Suppose that f(z) is analytic in a deleted neighborhood of z
0
. Then f has pole of
order N at z
0
if and only if
g(z) = ) ( ) (
0
z f z z
N
has removable singularity at z
0

and 0 ) (
lim
0
= =
N z z
b z g .
Definition: In the Laurents series expansion the coefficient
1
b of
1
0
) (

z z is called the
residue of at the isolated singular point z
0
.
Note that every function has a residue at each of its isolated singular points, because
Laurents series about the point represents the function throughout a neighborhood of the
point except at the singular point itself. (Actually the value of the residue may be zero).

We now have a powerful method of evaluating certain integrals around a closed contour.

Example: Evaluate the following Integrals over the given curve C which is oriented in the
positive sense.
1. dz
z
e
C
z
}

2
) 1 (
, where C: 2 = z
136
The only singular point is z = 1 inside C and the residue at z = 1 is
e
1
, hence
dz
z
e
C
z
}

2
) 1 (
=
e
i t 2
.
2. dz e
C
z
}
2
1
, where C: 2 = z
The only singular point is at z = 0 inside C and the residue at z = 0 is 0, hence
dz
z
C
}
1
1
2
= 0.
3. dz
z
C
}
1
1
2
, where C: 2 1 0 < < z
The only singular point is at z = 1 inside C and the residue at z = 1 is
2
1
, hence dz
z
C
}
1
1
2
=
i t .
Theorem: If a function has only a finite number of singular points in some domain, then
those singular points are necessarily isolated
Residue Theorem: Let C be a closed contour within and on which a function f is analytic
except for a finite number of singular points say, z
1
, z
2
, z
3
, .., z
n
interior to C. If k
1
, k
2
, k
3
,
., k
n
denote the residue of f at those singular points respectively, then

) .... ( 2 ) (
3 2 1 n
C
k k k k i dz z f + + + + =
}
t
(where C is oriented in the
positive sense).
Proof: Consider a contour C with singular points z
1
, z
2
, z
3
, .., z
n
and with the
corresponding residues k
2
, k
3
, ., k
n
. Since the singular points are isolated, we can make a
small circles C
i
centered at each z
i
for i = 1, 2, 3, ,n oriented as C. Now f is analytic on
and inside C except inside the C
i
s. Hence by Cauchy-Goursat Theorem, we have

=
}
C
dz z f ) ( +
}
1
) (
C
dz z f +
}
2
) (
C
dz z f +
}
3
) (
C
dz z f ..+
}
n
C
dz z f ) (
But
}
=
i
C
i
dz z f
i
k ) (
2
1
t
, from Laurents series.
Therefore, ) .... ( 2 ) (
3 2 1 n
C
k k k k i dz z f + + + + =
}
t
Some times finding residues using the expansion of Laurents series becomes very long and
takes time, so we look for other methods.

When f has a pole of order N at z
0
, then we know that the Laurents series expansion is
given by,
137
0 ,
) (
......
) (
) ( ) (
0
2
0
2
0
1
0
0
=

+ +

+ =

=
N
N
N
n
n
n
b
z z
b
z z
b
z z
b
z z a z f
Hence for
0
z z = ,

=
+

+ + + + + =
0
0
1
0 1
2
0 2 0 1 0
) ( ) ( ..... ) ( ) ( ) ( ) (
n
N n
n
N
N N N
N
z z a z z b z z b z z b b z f z z
Now differentiating term-by-term to get b
1
, yields

+
+ =
N n
n N
N
N
z z
N n
n
b N z f z z
dz
d
) (
)! 1 (
!
)! 1 ( )] ( ) [(
0 1 0
) 1 (
) 1 (


Hence taking the limit as z tends to z
0
, we get

lim
0
z z 1 0
) 1 (
) 1 (
)! 1 ( )] ( ) [( b N z f z z
dz
d
N
N
N
=

Since the residue at z = z
0
is b1 we can solve for b
1
from this expression and use it
wherever necessary.

Example: Find the residue of the following at each of its poles.
1.
) 1 (
2 5
) (

=
z z
z
z f , f has two poles at 0 and 1and each has order one thus using the above
equation

lim
0
z z 1 0
) 1 (
) 1 (
)! 1 ( )] ( ) [( b N z f z z
dz
d
N
N
N
=

We get, b
1
at 0 is 2, and b
1
at 3.

2.
) 4 )( 9 (
) (
2 2
2
+ +
=
z z
z
z f , f has four poles -3i, 3i, -2i and 2i and each of pole order one,
thus as in example (1), we get

b
1
at -3i is
10
3i
, and b
1
at 3i is
10
3i
b
1
at -2i is
5
i
, and b
1
at 2i is
5
i
.








Exercise: (I) Find the residue of the following at each of is poles and determine the order of
the pole.
138
1.
6
) (
z
e
z f
z
=
2.
2 2 2
2
) 4 )( 9 (
) (
+ +
=
z z
z
z f
3.
3
2
) 1 (
) 3 (
) (

+
=
z
i z
z f
4.
) 1 )( 9 (
2 3
) (
2
2
+
+
=
z z
z
z f
5.
) 4 (
1
) (
3
+
=
z z
z f

(II) Find the value of the contour integral of the following assuming that C is oriented in
the positive sense.
1. dz
z
e
C
z
} 6
C: 4 = z
2. dz
z z
z
C
}
+ +
2 2 2
2
) 4 )( 9 (
a) C: 4 = z
b) C: 5 . 1 = i z
c) C: 1 = z
3. dz
z z
z
C
}
+
+
) 1 )( 9 (
2 3
2
2
a) C: 4 = z
b) C: 2 2 = z
c) C: 1 = z
4. dz
z z
C
}
+ ) 4 (
1
3
a) C: 4 = z
b) C: 2 = z
c) 3 2 = + z











Application of Residues
139

Evaluation of Improper Real Integrals

One of the important applications of the theory of residues consists in the evaluation of
certain types of real definite integrals. These integrals often arise in physical problems,
especially in the solution of boundary value problems in in partial differential equations.
Sometimes they appear as integrals in the complex plan, even though they are reducible to
real definite integrals. Thus this section is devoted to the techniques of evaluating certain
types of real definite integrals with the help of the notions of complex integration and of
residue at a point. In each given case, the choice of a suitable curve along which integration
is to be effected will play an important part. This curve is usually known as a contour and
integration along this contour is called contour integration. The examples treated here and
in the following sections illustrate some useful method.
Example: 1. To evaluate the improper integral
} }

+
=
+ 1 2
1
1
2
0
2
x
dx
x
dx


}

+1 2
1
2
x
dx
-this integral represents integration, along the entire length of the real axis,
of the function,

)
1
)(
1
(
1
1
) (
2
i z i z z
z f
+
=
+
= a function with a simple pole at the point z = i and z = -i.
Let C
R
denotes the upper half of a circle R z = , where R > 1. Integrating f along the
boundary of the semicircular region in the positive sense, we have

ik dz z f dx x f
R
C
R
R
t 2 ) ( ) ( = +
} }

, where k is the residue of f at the pole z = i. (k =

i 2
1
)

Therefore, when R > 1,
t + =
} }

R
C
R
R
dz z f dx x f ) ( ) (

Now R z = when z is on C
R
, and 1 1 1
2 2 2
= > + R z z








Consequently,

140

1 1 1
2 2 2

s
+
} }
R
R
R
dz
z
dz
R R
C C
t
, and this quantity approaches zero as R tends to infinity. It
follows that,
t =
+
}

R
R
R
x
dx
1
2
lim

Therefore,
} }

+
=
+ 1 2
1
1
2
0
2
x
dx
x
dx
=
2
t


2. To evaluate
} }

+ +
=
+ +
2 2 2
2
0
2 2 2
2
) 4 )( 9 ( 2
1
) 4 )( 9 ( x x
dx x
x x
dx x

Following the same method as the first example, we note that the singular points of the
function

2 2 2
2
) 4 )( 9 (
) (
+ +
=
z z
z
z f
Consists of simple poles at the point 3i and -3i; and poles of the second order at 2i and -2i
The residues, at 3i is
i 50
3
and at 2i is
200
13i
.
Let C
R
denotes the upper half of a circle R z = , where R > 3. Integrating f along the
boundary of the semicircular region in the positive sense, we have

100
)
200
13
50
3
( 2 ) ( ) (
t
t =

= +
} }

i
i
i dz z f dx x f
R
C
R
R


Therefore, when R > 3,

100
) ( ) (
t
+ =
} }

R
C
R
R
dz z f dx x f
Now R z = when z is on C
R
, and
2 2 2
2
) 4 )( 9 (
) (

s
R R
R
z f
Consequently,

2 2 2 2 2 2 2 2 2
2
) 4 )( 9 ( ) 4 )( 9 ( ) 4 )( 9 (
=

s
+ +
} }
R R
R
R R
dz
z z
dz z
R R
C C
t
, and this quantity
approaches zero as R tends to infinity.







141
It follows that,

100 ) 4 )( 9 (
2 2 2
lim
t
=
+ +
}

R
R
R
x x
dx

Therefore,
} }

+ +
=
+ +
2 2 2
0
2 2 2
) 4 )( 9 ( 2
1
) 4 )( 1 ( x x
dx
x x
dx
=
200
t


3. By similar argument one can show that:
16
3
1 2
1
) 1 (
2
0
2 2
t
=
+
=
+
} }

x
dx
x
dx


4. To evaluate,
} }

+
=
+ 1
cos
2
1
1
cos
2
0
2
x
dx
x
xdx
the method that is employed in the above examples
does not apply, since z cos increases and tends to infinity as y tends to infinity, thus
here we use some modifications.
Now x cos is the real part of
ix
e ; therefore,

} }

+
=
+
)
1
Re(
2
1
1
cos
2
0
2
x
dx e
x
xdx
ix

Writing f(z) as
1
2
+ z
e
iz
and note that ) 0 ( , 1 > s =

y e e
y iz


The singularities of f are simple poles at i and I; and the residue of at i is
ei 2
1

Thus
e z
dz e
x
dx e
R
C
iz R
R
ix
t
=
+
+
+
} }

1 1
2 2

Where C
R
denotes the upper half of a circle, R z = , where R > 1, hence the second
integral over C
R
approaches zero as R tends to infinity. Integrating f along the boundary of
the semicircular region in the positive sense, we have

} }

+
=
+ 1
cos
2
1
1
cos
2
0
2
x
dx
x
xdx
=
e 2
t









142
Exercise: Use Residues to find the values of the following integrals:
1.
}

+ + 2 2
2
x x
dx

2.
}

+ + + ) 2 2 )( 1 (
2 2
x x x
xdx

3.
}

+ + 5 4
sin
2
x x
xdx

4.
}

+ +
0
2 2
) 4 )( 1 (
sin
x x
xdx x

5.
}

+ +
0
2 2
2
) 4 )( 1 ( x x
dx x

6.
}

+
0
4
) 1 (x
dx

7.
}

+
0
6
2
) 1 (x
dx x

8.
}

+
0
2 2
) 1 (x
dx