UNIT 6 CONFORMAL MAPPING AND ITS

APPLICATIONS
Structure Page No.
6.1 Introduction 131
Objectives
6.2 Mapping by Elementary Functions 132
Linear Transformations
The Transformation w = 1/ z
Linear Fractional Transfonnatlons
Mappings of the Upper Half Plane
The Transformation w = sin z
6.3 Conformal Mapping 137
6.4 Harmonic Conjugates 140
6.5 Applications of Conformal Mapping 141
Steady-State Temperatures
Steady Temperatures In a Haif Plane
Temperatures in a Quadrant
6.6 Summary 143
6.7 Solutions/Answers 143

6.1 INTRODUCTION
So far, in Unit 1-5, we were concerned with the analytical aspects such as the values of
the functions, differentiability, analyticity and integration of a function of a complex
variable. In this unit we shall concentrate on the geometrical interpretation of a
function of a complex variable as a mapping, or transformation. Here we shall discuss
the mapping of a region in a z-plane onto the w-plane under elementary functions and
conformal mappings.

In Sec. 6.2, we have started the discussion with different elementary transformation
viz. linear transformation, inverse transformation, bilinear transformation and
transformation w = sin z ,which map a given region in the z-plane onto a region in the
w-plane. One of the most beautiful geometrical aspect of an analytic function, viz., a
conformal map w = f (z) = u(x, y) + iv(x, y) which provides us with a powerful
solution technique for problems in two-dimensional potential theory by changing the
variables from x, y to u, v and simplifying the domain, while preserving the
governing Laplace equation, is discussed in Sec. 6.3. Conjugate property of harmonic
functions on a simply connected domain is discussed in Sec. 6.4. Applications of
conformal mappings to steady-state temperature problems are discussed in Sec. 6.5.

Objectives
After studying this unit you should be able to
obtain the image of a given region in the z-plane onto the w-plane under
elementary transformations viz., linear transformation, inverse transformation,
bilinear transformation and the transformation w = sin z ;
use the implicit form of linear fractional transformation to map distinct points in
the z-plane onto distinct points in the w-plane;
check if a given transformation is conformal at a given point ;
obtain harmonic conjugates of a given harmonic function on a simply connected
domain; I

use conformal mappings to solve certain physical problems in two-dimension.

Complex Analysis
6.2 MAPPING BY ELEMENTARY FUNCTIONS
In the case of a real-valued function y = f (x) of real variables x and y ,the
properties of the hnction are often exhibited by the graph of the function. But when
w = f (z) is a function, where z and w are complex variables, there is no such
convergent graphical representation available because each of the numbers z and w
is located in a plane viz., z and w planes respectively. In such cases the nature of the
function w = f(z) can be exhibited by studying geometric characteristics of certain
mappings between the two planes. In this section we shall see how various curves and
regions in the z-plane are mapped onto w-plane by elementary analytic functions.

6.2.1 Linear Transformations

You can start with reading the following:
-
Read Sec. 83, pages 299-301, Chapter 8 of the book. Go through the example
on page 300.

This section deals with the simplest geometrical mapping called linear
transformation. The general form of linear transformation is
w = A Z + B , (A#O),
where A and B are real or complex constants. The transformation is composed of
two components:
Z = Az
and
W=Z+B
which represents an expansion/contraction and a rotation, followed by a translation.
The mapping is illustrated by the Example on page 300 of the book. Here we are
giving one more example for your reference.

Example 1: Consider the rectangular region with vertices as

A(0, 0), B(1, 0), C(l, 2), D(0, 2) in the z-plane. Find the image of this region under
the transformation w = (1 - i) z - 2i .
Solution: We write the given transformation as

This can be considered as the composition of the transformations

i) ,
w = z + (1 - i) (translation)
ii) , ,
w = fiw (expansion)
iii) = e-ixJ4 w ,(rotation).
Mapping i) translates the rectangular region ABCD in the z-plane (see Fig. l(a)) onto
the rectangle A,B,C,D, ,under the mapping w, = z + (1 - i) as shown in Fig. l(b),
where,
A(0, 0) is mapped as A, (1, - I), B(l, 0) is mapped as B, (2, - I),
C(1, 2) is mapped as C, (2, l), D(0, 2) is mapped as Dl (1, 1) .

Mapping ii) expands the rectangular region A,B,C,D, ,under the mapping
w, = f i w, onto the rectangle A,B,C,D, as shown in Fig. l(c).

Mapping iii) rotates the rectangular region A, B ,C, D, by an angle - n 14 ,onto the
rectangle A3B3C3D3as shown in Fig. l(d).
 Conformal Mapping and Its
Applications

Fig. 1

Notice that rotation through an angle - x / 4 in the anticlockwise direction is a

rotation through an angle x 1 4 in the clockwise direction.
-x-

You may now try the following exercises.

Do the exercises 1,3,6 on page 301 of the book.

Let us now consider the inverse transformation.

L

6.2.2 The Transformation w = l/z

You can start with reading the following:

Read Secs. 84-85, pages 301-305, Chapter 8 of the book. Go through

Examples 1,2,3 on pages 304-305 of the book.

Afier going through Secs. 84-85 of the book you must have noticed the important
geometrical fact about w = 1/ z is that it transforms circles and lines into circles
. and lines.
1
We now show you through an example that under the transformation w = - ,a
z
hyperbola is transformed into a lemniscate.

Example 2: Show that the transformation w = 1/ z transforms the hyperbola

x2 - y 2 = 1, into a lemniscate p2 = cos241.
Complex Analysis z+z z-Z
Solution: We know that x = -and y=-.
2 2i
Thus, x2 - y 2 = 1 can be expressed as

1
Under the transformation w = - ,the above equation is transfonped into
Z

If we express w in polar form w = pei4,we get

You may now try the following exercises.

Do the exercises 1,2,4,7,14 on pages 305-306 of the book.

We shall now consider one of the most important class of elementary transformations.

6.2.3 Linear Fractional Transformations

You may read the following:

Read Secs. 86-87, on pages 307-311, Chapter 8 of the book. Go through

Examples 1,2 on page 309 and Examples 1,2 on pages 310-311.

Sec. 86 introduces you to linear fractional transformation which are also called
bilinear or Mobius transformation. Before you proceed fiuther, we would like you to
notice the following points about the bilinear transformation.
w=- a z + b (ad- bc $0).
cz + d
i) For each z for which cz + d + 0 , we obtain a unique value of w .
ii) w is not defined for any z if c = 0 = d .
iii) If a = c = 0 or b = d = 0 , then w =constant, for all z and whole z-plane is
mapped onto a point.
iv) If c = 0 and d + 0 , then we get w = (az + b) / d ,which is a linear mapping.
v) If a = 0, b # 0 and c # 0 , then Eqn.(l) can be written as

which can be written as a composition of the following mappings.

translationZ=z+d/c.
1
inversion W = -
z
 Conformal Mapping and Its

vi)
rotation and expansiodcontraction w =

If a # 0, c # 0 ,then Eqn.( 1) can be written as

(3W Applications

a
w=-+
C
which is a composition of the following mappings
rotation, expansiodcontraction and translation Z = cz + d
1
inversion: W = -
Z
rotation, expansiodcontraction and translation:
w=-+-a bc-ad W
C C
The bilinear transformation (1) is a composition of linear and inversion mappings and
therefore maps the circles and straight lines in the z-plane onto circles and
straight lines in the w-plane.

Further, these maps can as well be defined for the extended complex plane by
enlarging its domain of definition. We write
az+b
f (z) = - (ad -bc # 0)
cz + d

( 1
and define f(m) = a / c and f - = a, if c # 0 . Finally,
f(m)=a, if c=O
-cd

You may observe that bilinear transformation maps the complex plane excluding the
point {-d 1c) in one-to-one manner to the complex plane excluding the point
(3.
view of this observation we say that the bilinear transformation is a bijection of the
In

extended z-plane onto the extended w-plane.

Note that a fixed point of a linear fractional transformation (or any transformation)
w = f(z) is a point z, such that f(z,) = z, . It can be easily shown that a linear
az+b
fractional transformation w = -, (ad - bc # 0) which is not an identity mapping
cz + d
has at most two fixed points in the extended complex plane. If it has more than two
fixed points, then it must be the identity mapping w = z .

Let us assume that the linear transformation has three fixed points z, , z2, and z, .
Then by the implicit form of the transformation we get

3 w=z
which is an identity transformation and hence a contradiction.

Let us now consider the following example.

Example 3: Show that if the origin is a fixed point of a linear fractional

transformation, then the transformation can be written in the form w = -
cz+d
z
, d#.O.

Solution: Any linear fractional transformation can be written as

Complex Analysis

If 0 is a fixed point of the transformation then we have

A.0 + B
0=
C.0 + D
3 O=B/D*B=O (DtO).
The transformation then becomes
[A # o since D # 01

Now it is time for you to do some exercises on your own.

Do the exercises 1,3,7(a), 8,12 on page 312, Chapter 8 of the book.

6.2.4 Mappings of the Upper Half Plane

This section introduces you to those linear fractional transformations that map the
upper half plane Im z > 0 onto the open disk I w 1 < 1 and the boundary Im z = 0 onto
the boundary I w I = 1. These transformation simply turns out to be of the form
. z-zo
w = - (Imz, >O)
- z-z,
Y

where, a is a real constant and z, (non-zero) is a complex constant.

Read Sec. 88, pages 313-315, Chapter 8 of the book. Go through Examples 1,
2 on page 315. Do the exercises 1,2,6 on pages 316-318.

6.2.5 The Transformation w = sin z

Read Sec. 89, pages 318-322, Chapter 8 of the book. Go through Examples 1-
4, pages 319-322 carefully.

This section introduces you to the transformation w = sin z which gives

u +iv =sinz=sinxcoshy +icosxsinhy
or,
u=sinxcoshy and v=cosxsinhy.
To find the image of certain region under this mapping, we begin the discussion by
studying images of vertical and horizontal lines.

Vertical line x = c, , (0 < c, < 11: / 2) is transformed into points on the curve

whichis the right hand branch of the hyperbola

Similarly horizontal line y = c,, c 2 > 0 , is mapped onto the curve

U = S ~ ~ X C Ov=cosxsinhcZ
S~C~, (-A<x<~)
which is an ellipse
 Conformal Mapping and Its
Applications

These two images are sufficient to give you an idea about the image of any region
under this map.

You may now try the following exercises.

Do the exercises 1,3,6,8 on pages 322-323 of the book.

In the next section we shall introduce you to the concept of conformal mapping which
provide us with a powerful solution technique for solving certain two-dimensional
boundary value problems. We shall discuss some of these problems in Sec. 6.5.

6.3
- -
CONFORMAL MAPPING
- - - - - - -

You can start with reading the following

Read Secs. 94-96, pages 343-351, Chapter 9 of the book. Read Examples 1-4,
page 345 and the example on page 346.

After going through Sec.94 of the book you would notice that conformal mapping
w = f (z) ,which maps a region D in the z-plane onto a region D* in the w plane,
preserves the angle between two curves both in magnitude and direction. A mapping
w = f(z) is conformal in D if f is analytic in D and its derivative f ' has no zeros
there. Let us once again consider Examples 1-3, on page 345 of the book. You may
find the following details useful for a better understanding of these examples.

Example 1 on page 345: Put w = u + iv and z = x + iy in the mapping w = ez .

Then, u + i v = w = e z =exeiY=ex(cosy+isiny)
3 u = e x cosy and v = e x siny
v
and 2 x = 1 n ( u 2 + v 2 ) a n d tany=-.
U
Now as x = c, we get,
u =eC1cosy and v = e c lsin y
3 u2 + v2 = e2cl 3 (u2 + v 2 )= ( e ' ~ ) ~

Fig. 2
Complex Analysls
which represents a circle centred at origin and of radius ecl.
For y = c 2 , weget

which is a straight line starting from the origin and inclined to u-axis at an angle c2
(see Fig. 2).
-x-

Example 2 on page 345: We illustrate this example by taking

Level curves of real and imaginary parts of f (z) are given by

and - = c 2 s x2 + y 2 + - y1= O
x 2 + y2 c2

For computational ease, let us consider c, = 113 and c2 = -1 12 . Then Eqns.(2) and
(3) reduce respectively to
x 2 + y 2 -3x=O and x2 + y 2 -2y=O.
Their point of intersection is
z, =(12/13,18113) [origin is not to be considered]

,
and f '(z, ) # 0 ,w = f (z,) = (113, - 112) . Mapping is illustrated in Fig. 3.

Fig. 3
-x-
Example 3 on page 345: Let L, and L, be two straight lines in the z-plane as shown Conformal Mapping and Its
Applications
in Fig. 4.

The transformation-
f(z)=Z=x+iy=x-iy
is a reflection across the x-axis.

You may notice that the angle measured from L, and L, is P (positive sense) and
between f (L, ) and f (L, ) it is P in magnitude but sense is negative.
-x-
Let us refer to Sec. 96, on page 348 of the book which introduces the concept of local
inverses. Here we would like to mention that if f(z) is analytic function in a domain
D containing a point z,, which is not a critical point for f(z) i.e., f1(z0);t 0 , then f
is locally one-to-one near zo (see Fig. 5). That is, an open neighbourhood Do of zo
exists such that Do c D and if z,, z, E Do and z, ;t 2, then f(z,) ;t f(z,) .

Fig. 5

Remember that a real function can be one-to-one in the neighbourhood of a critical

point. For instance, 0 is the critical point of the function f(x) = x3 which is one-to-
one in the neighbourhood 0 . However, this is not the case with an analytic function.
 ..... .

Complex Analysis Further, the fact that an analytic function has no critical points in a domain D does
not guarantee that it is one-to-one in the whole domain. Consider the transformation
d
f (z) = ez . Since -ez = ez z 0 , this mapping has no critical points in the complex
dz
plane. If w # 0 then equation ez = w has infinite number of solutions. Thus, w = ez
is locally one-to-one but globally it is not. It is one-to-one in the disc D(0, r) where
r 5 7t . (remember ez+2n"i
= ez V n E Z )
You can now go through the example on page 349 of the book.

Read the Example on page 349 and do the exercises 2 , 3 , 5 , 7 , 9 on pages 350-
351 of the book.

The next section gives you the connection between conformal mappings and harmonic
functions.

6.4 HARMONIC CONJUGATES

You may recall that in unit-1, Sec. 1.6, we defined harmonic functions and discussed
its applications to study some of the 2-dimensional problems in heat flow,
electrostatics and fluid flow. In this section you will learn the conjugate property of
harmonic function that if u(x, y) is a harmonic function defined on a simply
connected domain D then u(x, y) always has a harmonic conjugate v(x, y) in D
Sec.97 of the book give you the details about it.

Read Secs. 97-99, pages 351-358, Chapter 9 of the book. Go through all the
examples given in these sections.

Consider the theorem on page 356 of the book. The result proved here is frequently
used in solving boundary value problems. Example on page 357 illustrates this
theorem. We are adding here one more example for your reference.

Example 4: Verify the theorem on page 356 of the book for the function
h(u, v) = 2v + 3 and the transformation
w = f ( z ) = i z 2 =-2xy+i(x2 - y 2 ) .
The contour C in the z-plane is the line
y=x, x > o .
Solution: The line y = x(x > 0) is mapped by f(z) = iz2 onto line r given by
v = 0 (u c 0) i.e., negative u-axis (see Fig. 6)

Fig. 6
 The value of h(u, v) on the negative u-axis is Conformal Mapping and Its
Applications
h(u, v) = 2.0 + 3 = 3
and H(x, y)=h[u(x, y), v(x, y)] =2(x2 - ~ ' ) + 3 ,
on the line C(y = x, x > 0) is
H(x, y) = 3 .
T h ~ verifies
s the theorem.
-x-

You may now try the following exercises.

Do the exercises 1,4,6 on pages 358-359 of the book.

You may recall that in Unit 1 (ref. Sec. 25 of the book), you were introduced to
harmonic functions. You saw that the h c t i o n T(x, y) defining the temperature
distribution in a thin plate lying in the xy plane is harmonic. In particular, it was
shown that the function T(x, y) taken in the form T(x, y) = e-Ysin x , is harmonic in
the semi-infinite vertical strip 0 < x < x , y > 0 and it satisfies all the boundary
1 conditions which describe steady-state temperature distribution in a thin homogeneous
plate in the xy plane having no sources or sinks and whose edges are insulated. You
tI may wonder why such a form of T(x, y) was taken. This section will help you find
l an answer to this and other related problems.

6.5 APPLICATIONS OF CONFORMAL MAPPING

6.5.1 Steady-State Temperatures
Let us start with the derivation of the equation governing the steady-state temperature
distribution T(x, y) in a thin, homogeneous semi-infinite plate (solid) in the xy plane
which is discussed in Sec. 100 of the book. It is shown there that the temperature
function T(x, y) satisfies Laplace's equation at each interior point of the solid.

Read Sec. 100, pages 361-363, Chapter 10 of the book.

Remember that you may have to read this section repeatedly (twice or thrice) for a
\
better grasp of the physics of the problem.

Having learnt that the function T(x, y) satisfies Laplace equation you may now
proceed to obtain the form of T(x, y) .

1 6.5.2 Steady Temperatures in a Half Plane

Expressions for the steady temperatures T(x, y) in a thin semi-infinite plate whose
faces are insulated and whose edges are kept at zero temperature are obtained in Secs.
101 and 102 under different boundary conditions. Read these sections carefully.

Read Secs. 101-102, pages 363-367, Chapter 10 of the book.

The next section discusses the steady-state temperature distribution in a thin plate
having the form of a quadrant.
--

Complex Analysls 6.5.3 Temperatures in a Quadrant

Having gone through Secs. 101 h d 102, we hope that you will not face difficulty in
understanding the following section.

Read. Sec. 103, pages 368-370, Chapter 10 of the book.

You may now try the following exercises.

El) Find the steady-state temperature of the semi-infinite strip (0 5 x 5 sc, y 2 0)

with the boundary conditions given in Fig. 7 below:

T = T,, sinx
lim T(x, y) = 0 (0 < x < 71)
Y-00
Flg. 7

E2) Suppose that two parallel planes are perpendicular to the z-plane and pass
through the horizontal lines y = a and y = b and that the temperature T(x, y)
is held constant at the values T(x, a) = T, and T(x, b) = T, ,respectively, on
these planes. Then show that T(x, y) is given by
T(x, y)=T, +-T2 - TI (y-a).
b-a
Also find the isotherms and lines of flow of T(x, y) .

E3) Find the steady-state temperature distribution of a semi-infinite slab shown in

Fig.8.

The segment [0, I] on the x-axis is insulated.

We now end this unit by giving a summary of what we have covered in it. Conformal Mapplng and Its
Applications

6.6 SUMMARY
In this unit you have learnt that

1. The general linear transformation

w = A z + B , (A#0)
is a composition of the transformations
Z = a z (A#0) and w = Z + B
and represents an expansion or contraction and a rotation, followed by a
translation.
1
2. The mapping w = - transforms circles and lines into circles and lines.
Z

3. The transformation
az+b
w=- (ad - bc z 0)
cz + d
where a, b, c, d are complex constants is called a linear fractional
transformation or Mobius transformation or bilinear transformation.

4. Bilinear transformation transforms circles and lines into circles and lines.

5. The transformation w = sin z can be written as

It transforms vertical lines x = c, (0 < c, < .n/ 2) onto the right hand branch of
the hyperbola

and the horizontal lines y = c2, c2 > 0 on to the ellipse

6. Transformation w = f (z) is conformal on a domain if f ( 2 ) is analytic and

f '(2) # 0 everywhere on the domain. Conformal maps preserve angles, both in
magnitude and sense.

7. A harmonic function defined on a simply connected domain always has a

harmonic conjugate in the domain.

6.7 SOLUTIONSIANSWERS
Exercises on Page 301 of the book.

1. w =iz, then
u + i v = i (x+iy)=ix-y
3 u=-y and v = x .
Now O < x < l a O < v < l . ImageisShowninFig.9
' Observethat IiI =1 and argi=.n/2.
Therefore, w = eiaz (a = n 1 2 ) .
N~~ put = rei9 nen = rei(e+a) = rei(e+%/2)
Complex Analysis Thus any radius vector represented by z in z -plane is rotated through an angle
n I 2 under the transformation w = iz .

Fig. 9

3. Given transformation w = (1 + i) z ,
(a) Using Polar coordinate we write z = reie .
Since z lies in the upper half plane, 0 c 8 c 7c .
= fieini4. Thus, w = rJZ ei(e+ni4) .
(1 + i) =(I+ i I eiad'+i)
Sketch of the mapping of the region y > 0 is shown in Fig. 10.

Fig. 10

Note that the result is obtained by rotating x-axis anticlockwise through

anangle n/4.

(b) u + i v = ( l + i ) ( x + i y ) a u = x - y and v = x + y
Now y > O s v - u = 2 y > O s v - u > O s v > u .

6. w = A( z + B) (A z 0) ,is composition of two transformation Z = z + B and

w = AZ . The first transformation is translation by means of the vector
representing B . Under this mapping, the image of any point (x, y) in the
z-plane is the point (X, Y) = (x + b,, y + b,) (B = b, + ib,) in the Z-plane. The Conformal Mapping and Its
Applications
second transformation w = AZ ,expands or contracts the radius vector representing Z
by the factor a = 1 A I and rotates it through an angle a = arg A about the origin of the
Z-plane.

Exercises on Pages 305-306 of the book.

U
1. We have, x =
u2 + v 2 '
U
Now x 2 c , (c, >0) e 2 2Cl
u2+ v

Thus inequality (7) holds.

U
2. x<c,a < c, (Clearly u < 0)
u2 + v
2 u <cl (u2 + v2)

a [II - ,I:[
+ v2 < which represents the

interior of the circle centred at and of radius =(1/ 2c, ( (See Fig. 11).

X = C,
Fig. 11
Complex Analysis When c, =0, x c O

It can be observed that (0, y) when y starts increasing from 0 upwards in the
z-plane, (0, v) in the w-plane approaches (0, 0) from below and as (0, y)
starts decreasing from 0 downwards, (0, v) approaches (0, 0) from above.
So image in w-plane is half plane u c 0 as shown in Fig. 12.

4. Wehave,x=
u
and y =
-v
u2 + v uz + v z '
When O c y c 1 / 2 c , weget
-v 1
Oc <-
u 2 + v 2 2c
= v c O and - 2 c v c u 2 + v Z
= v c O and 0 c u 2 +(v+c)' - c Z
= v c O and cZc u Z+ ( v + c ) ~
u Z + (v + c ) >
~ cZ is the exterior of the circle (including boundary) with centre
(0, - c) and radius c . Therefore (4) represents lower half plane (v c 0)
excluding the interior as well as points on the circle u2 + (v + c)' = cZ (see
Fig. 13).

Fig. 13
 1
7. We can write the given map w = -
2-1
as composition of the two functions
Conformal Mapplng and It#
Applleatlons

Z = z - 1 (translation) and w = 11z (inversion).

Fig. 14 shows how a rectangular block in z-plane is transformed onto w-plane

1

under w = A .
2-1

1
We have chwsen a point z0 = - + i inside the rectangle. It is transformed as
2
follows:
Complex Analysis 14. (a) Use the hint given in the book. Simplify the resulting equation and get the
desired result.

(b) When w = 1/ z then z = 1/ w and Z = 1/ i? , given equation becomes

1 1 1 1
2A-=+(B-iC)-+(B+iC)=+2D=O
W W W W
a 2Dw i?+(B+Ci)w+(B-iC)i?+2A=O.
Put w = u + iv in the above equation and get the desired result.

Exercises on Page 312 of the book.

1. Using Implicit formula of Sec.87 of the book, we have

- (z - 2) (i + 2) a -I--- . w-1 - (3+4i) 2-2
(w - 1 (i + 1 -
(w+l)(i-) (z+2)(i-2) w+l 5 z+2

3. In this case we use modified form

(w -w1) - (z2 -z3)
(w2 -w1) (z-3)

z-1
7. (a) w=- = z a z-l=z 2 +z
z+l
a z2 = -1 a z = fi are the fixed points.

8. Ifz2=aandw2=a.Thenreplacez2byl/z2andw2byl/w2inthe
implicit form and let z, + 0 and w, +0. That is,
(w - w l ) (1/w2 - w 3 ) W
lim 2= lim (z-z1) (11~2-z3) - z2
w1+0 ( w - w ) ) (1/w2 - w , ) ' w , Zz+O (z-z)) (1/z2 - z 1 ) ' z 2

a lim ( w - w l ) (1-w2 - w 3 ) = lim ( z - z ~ )(l-z2z3)

W2+0 ( w - w ) ) (1-w1w2) z1+0 (z-z3) (1-z1z2)
W-W1 z-z1
a -- . Now if its fixed points are 0 and a then
W-W) z-Z)

zw - wz3 = Zw - w3z, (z3 +O and w 3 +0)

 az+b Conformal Mppping and Its
12. T(z)=- , where ad - bc ;t 0 , Applications
cz + d

3 c(a+d)=O, ( d + a ) (d-a)=O, b(a+d)=O

3 If a + d ;t 0 then c = 0, a = d and b = 0 and we get T(z) = z [which is not
possible].
Therefore, a + d = 0 3d = -a
az + b
Clearly, if d = -a then T(z) = -= w
CZ-a
3 az+b=czw-aw3 z(a-cw)=-aw-b

Exercises on Pages 316-317 of the book.

i-x
1. If z = x , t h e n w=-
i+x
(i-x) ( i ) -- l+ix+ix-x2 -
--1 - x 2 + i 2x
3 w=
x2 - j 2 (x2 +1) 1+x2 l+x

Clearly, 1 w l2 = ( I - x ~ )+~ 4x2 (I+x~)~

- =1
(1 + x ~ )(1 +~x 2 ) 2 (1 + x2)2

Thus points on real axis in Fig. 13, Appendix-2 of the book, are mapped onto
the boundary of the disc ( w I < 1 as shown in Fig. 15.

Fig. 15
Complex Analysis Consider segment CD : 0 I x < 1. Since u 2 0, v 2 0 segment CD is mapped
onto arc C'D' . Ray DE :1I x < a, is mapped onto arc D'E' (u I 0, v 2 0) and
SO on.
X
2. Z = iz 3 Rotation by angle - in anticlockwise sense.
2
w=-i - Z 3 is already discussed in Exercise 1.
i+Z
w = -W 3 It is contraction by factor - 1 followed by rotation through an angle
arg(-1) = 7c about the origin.

6. To be added.

Exercises on Page 322 of the book.

1. Consider Fig. 16.

Fig. 16

We know that for w = sin z , u = sin x cosh y and v = cos x sinh y

Imageof A(c,, 0) isgivenby (sinc,, 0).
e Y+ e-Y
Clearly, if y 2 0, cosh y = 2 1 and sinh y 2 0 . As y +a,cosh y and
z

u2 v2
x=c,,gives ---- - 1 . Thus as point (c, , y) moves upward along
sin2c1 cos2c1
the vertical line x = c, ,it traces left-hand branch of hyperbola in one-to-one
manner from A' to E' as. shown in the Fig. 16.

X 7C
3. It is clear that each point of the interior of the strip - - I x I - and y 2 0 lies
2 2
IC X
on one of the horizontal line segment y = constant, - - <- x I - and y r 0
2 2
(see Fig. 17). Note that images of these segments are distinct and cover the
1
entire 'half w-plane. For example, image of the horizontal line segment

Consider general horizontal segment y = c ,

(c > 0)
 It is mapped in one-to-one fashion to the upper-half of ellipse A'B'C' given by Conformal Mapping and Its
Applications

. As y + m ,whole of v > 0 is covered.

Fig. 17

Exercises on pages 350-351 of the book.

2. Given transformation is f (z) = w = 11z . This function is analytic everywhere

except at z = 0 . Now f '(2) = -1 1z2 . At z = 1, f '(1) = -1 and arg (f '(1)) = n:
and at z = i, f '(i) = 1 and arg (f '(i)) = 0 .

For transformation w = 1/ z . u = xz
X
3. + y2 " = x 2- +Y Y2
and

u -v
X = Now for y = x - 1
u2 + v2 y = u 2 + v z'
-v - -
U
-1
u2 + v 2 u2 + v2
=3 - v = u - u 2 - v 2 3 u 2 + v2 - u - v = o

yA vA

$. (a)

Fig. 18

It is a circle with centre (1 12, 1/ 2) and radius 1/ f i (see Fig. 18).

-v
For y=O,
u -I- v
=
= 0 v = 0 , a straight line.

Observe that arg fl(l) = arg (-1) = ~r.

Complex Analysis 7. Sincew=f(z)isconformalatzO=(x,,yO),~fl(zO))~O.
Throughout N , we have
1 1 1 1
Xu=-vy, Yu = - JV,. Y, -
--Jux.
J X v =-JUy,
We also have
U, = v y and U, =-v,. (a: w = f (2) is analytic)

This gives J # O and x u = y , and x, =-y,.

Thus real and imaginary part of g(w) = x(u, v) + i y(u, v) have continuous
partial derivative of first order and Cauchy-Riemann equations are satisfies in
the neighbourhood N . Therefore, g is analytic in N .

9. An arc z = z(t) (a 5 t I b) is said to be smooth if the derivative zl(t) is

continuous on the closed interval a < t I b and non-zero on the open interval
a<t<b.

Fig. 19

Let C be the smooth arc defined by z = z(t) (a 5 t Ib) (see Fig. 19).
By definition zl(t) is continuous on the closed interval a I t I b and
zf(t)# 0 , a < t < b . We have w(t) = f(z(t)) be the function defined at all
points of C . The image of C is l- under this transformation and
w '(t) = f '(z(t)) zf(t) ( a ~ t ~ b ) .
Since zf(t)+ 0 and f is conformal fl(z(t)) # 0 , therefore, wf(t)# 0, a < t < b
and wf(t) is continuous. Hence, arc r is smooth.

Exercises on page 358 of the book.

Given harmonic function u(x, y) = x 3 - 3xy2 . According to Eqn.(5) Sec.97,

the function

V(X, Y)=
(0.0)
I
(x3 Y)

{-ut(s,t)ds+uS(s,t)dtj [u(s, t) = s
3
- 3st 1

is a harmonic conjugate of u(x, y) .

Now v(x, y)=

(x9

5 - 6st ds + (3s' -3t2) dt

Y)

(0,0)
 Evaluating the integral by first integrating along the horizontal path from the Conformal Mapping and Its
Applications
point (0, 0) to the point (x, 0) and then along the vertical path from (x, 0) to
the point (x, y) ,we get
v(x, y) = 3x2y- y 3 .
The corresponding analytic function is given by
f(z) = x3 - 3xy2 + i ( 3 - y3)
~ = x3~ + 3ixly
~ - 3xy2 - iy3
= z3 .

4. Consider the transformation w = exp z .

O I y I x and x = O a u = c o s y , v=siny
a u 2 + v 2 =1 and v 2 0 .

u
=2-u+
u2 + v 2 '
We find that the partial derivatives h ,, h ,, h ,, h ,exist and are continuous
everywhere in the w-plane except at point (0, 0) where h(u, v) is not even
defined. You can easily verify that h(u, v) satisfies Laplace equation, i.e.,
h,+h,=O.
1
Function h(u, v) being the real part of the function g(w) = 2 - w + - , which
W
- is analytic everywhere in the w-plane except at the origin, is harmonic
everywhere except at the origin. r in this case is the semicircle u2 + v2 = 1
(see Fig. 20).

Fig. 20

On the semicircle u 2 + v2 = 1, v 2 0, h assumes the value

u
h=2-u+-=2.
1
Now H(x, y) given by H(x, y) = h[u(x, y), v(x, y)] is
H(x, y) = h[ex cos y, e x sin y]
ex cosy
= 2 - e x cosy+
e2'
It can be seen easily that along the segment 0 1y I x of y-axis.
H(x, y ) = 2 - e 0 cosy+--cos y - 2
eO
which illustrates the theorem of Sec. 99, Chapter 8.
Complex Analysls
6. Consider h(u, v) = Re(-2iw + e-" ) = 2v + e-" cos v .
Now h, =2-e-"sinv. Alongu-axis (v=O), h, = 2 .

Fig. 21

Also, H(x, Y)= h[u(x, y), v(x, y)l = h[x2 - y ', 2xyl
= 4xy + ey2-" cos (2xy).
.: H, (x, y) = 4y - 2x e Y'-"' cos (2xy) - 2ysin (2xy) e Y'-"'
and H, (x, y) = 4x + 2y ey'-x2cos (2xy) - 2x sin (2xy) e Y'-"

Thus, normal derivative H, along positive y-axis = 4y and normal derivative

,
H along positive x-axis = 4x (see Fig. 21).

Exercise on page 142 of the unit.

El) Consider the function f(z) = e" it is an entire function. Its real and imaginary
parts are u(x, y) = e-Ycosx and v(x, y) = e-Ysin x.. Both u and v are
harmonic functions satisfying the Laplace equation. Consider
v(x, y) = e-Ysinx .
Observe that v(x, y) satisfies all the boundary conditions viz.,
v(0, y) = 0, V(R,y) = 0, V(X,0) = sin x
and lim v(x, y) = 0 .
Y+*

Thus the required solution is given by T(x, y) = Toe-Ysin x .

E2) Consider Fig. 22. A reasonable assumption is that the temperature at all points
on the plane passing through the line y = yo is constant. Hence T(x, y) = t(y) ,
where t(y) is a function of y alone. Now Laplace's equation implies that
tW(y)=0,V x ( a s y s b ) .

a t(y) = my + c ,where m and c are constants. The stated boundary

conditions T(x, a) = t(a) = T, and T(x, b) = t(b) = T2 leads to the solution
T, =ma-tc and T2 = m b + c

T - T2 aT2 - bT,
:. T(x, y) = -
+

"T, +-T2 - TI (Y-a).

a-b a-b b-a
 Conformal Mapping and Its
Applications

T (x, a) = T, < s CX, Y) =?

Heat flow llnes

Fig. 22

Now isotherms T(x, y) = a give

'r2- T,
a = T , +-(y-a)* (a-TI) (b -a) + a = y
b-a (T2 -TI)
They are horizontal lines. The conjugate harmonic function-ic
rn m

S(x, Y)=-
11 - 12
b-a
and heat flow lines are vertical line segments between the horizontal lines. If
TI > T2 ,then the heat flows along these segments from the plane through y = a
to the plane though y = b .

No heat transfer takes place along the insulated segment. Therefore, the
temperature distribution problem in E3) is equivalent to one shown in Fig. 23.
The temperature between two parallel planes kept at constant temperature T,
and T2,respectively is Fig. 23
T(x, Y)= T1 + (T2 - TI)x [ref E2)]
which is also the required solution of the given problem.