CIJAPThR

Functions of a Complex Variable

3.1.. INTRODUCTION

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

z=x+iy, wherei=Pi. The real numbers x and y are called the real and imagi nary parts ofz. In the Argands diagram, the complex numberz is represented by the point P(x, y). If (r, 0) are the polar coordi

(x, y)

nates of P, then r =

.Jx2 +

2 is called the modulus of z and is y

-

denotedby I z I.Also 0=tarr 1

iscalledtheargumentofz

and is denoted by arg. z. Every non-zero complex number z can be expressed as zr(cos0+i sin 9)_reiO If z = x + iy, then the complex number x iy is called the conjugate of the complex number z and is denoted by . Clearly, I I I = I z I , I z 12 = z ,

R(z)=, I(z)=-. 2 2i
3.2. FUNCTION OF A COMPLEX VARIABLE

z+I

zI

If x andy are real variables, thenz = x + iy is called a complex variable. If correspond ing to each value of a complex variable z(= x + iy) in a given region R, there correspond one or more values of another complex variable w (= u + iv), then w is called a function of the complex variable z and is denoted by
w = z) = u + iv For example, if w=z , 2 where z = x iy and w ftz) = u + iv then = u+iv=(x-i-iy) ) 2 +i(2xy) (x y = is = 2 and v = 2xy y Thus u and v, the real and imaginary parts of w, are functions of the real variablesx and

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

109

110

A TEXTBOOK OF ENGINEERING MATHEMATiCS

If to each value of z there corresponds one and only one value of w, then w

is

called a

8ingle-valued function of z. If to each value of z there correspond more than one values of w, then w is called a multi-valued function ofz. To represent w = flz) graphically, we take two Argand diagrams: one to represent the

pointz and the other to represent w. The former diagram is called the xOy-plane or the z-plane and the latter uOv-plane or the w-plane.

3.3. EXPONENTIAL FUNCTION OF A COMPLEX VARIABLE

Def. The exponential function of the complex variable z = x + iy, where x andy are real, is defined as Exp.(z)=eY=e(cosy+isiny)=eccisy Note. Wheu x =0, eY = cos y + i sin y = cis y

Y=cosyisiny=cis(y). 4 Changingitoi,e

3.4. PERIODICITY
& IS A PERIODIC FUNCTION, WHERE z IS A COMPLEX VARIABLE

ProoLLet then, bydefinition i.e.,

z=x+iy
ez=Y=ex(cosy + i siny)=ex[cos(2n1t+y)+i8in(2n+y)] = ex + i( +y) = e(X + y) + 2iix = eZ + 2nxi

e renlrnng unchanged when z is increased by any multiple of 2Ki. e is a periodic function with period 2ici.

ILLUSTRATIVE EXAMPLES
Eamp1e 1. Split up into real and
(i) e
2
imaginary parts:
1

53 e (ii)

(iii) e

SOL(i)

5 5 =e5[coi+isin.]e5[0+i. 1]=ie e Re[e 2 ] =o, Im[e 2 1=e . 5 25+9i = (5+3i) + 2 30i=16+30i

(cos30+isin30) 16 =e 6301 =e

(ii)

[..

=1] 2 i

Re [e] =e 16 sin 30. 16 cos 30, Im [e] =e (iii)

Irny = ecoyiem3) =e =ef( 1 4 etX0y [cos(ex sin y)+isin(ex amy)] =

Re(e) = eO6Y . cos (ex siny)

Irn(ee)=ec3I .sin(ex

FUNCTIONS OF A COMPLEX VARIABLE

111

Tmp1e 2. Find all values of z which sati8jv e t =1 L ez=eY=ex(cosy+isiny) SoLSince =1+i = ex(cosy+isiny)=1+i 1 e Equating real parts X cos y = 1 Equating imaginary parts ex sin y = 1 Squaring and adding (1) and (2), we get s or e=2 +sin )=1+1 e(co y y 2

or

2x=log2

:.

x= 1og2

Dividing(2)by(1), tany=1=tan y = nit + !, where n is an integer. l7t+ Y= where n=O,1,2 , o2+i(f z=x+Z 1 1 )
mpIe 3. Prove that [sin (a 0) + e sin 01 = sin a [sin (a nO) + e sin nO]. Sol. L.H.S. = [(sin cx cos 0-cos asin 0) + (cos ajam a) sin or =(sinczcosOi sinasinOY=[sina(cosOisinO)] sin a. = [sin ci e] nn0] cosa-isina)ai in [(sinacosn0-coeasinnO)+( =s a 1 R.H.S. sin no] no nO] sin = a [cos a [sin a cos nO i sin a sin

am a.

L.H.S.

R.H.S

Ezample4. Given i=j+CPi4wheIeL, P,Rare real, express pintheformM

giving the values of A and O
Sal. 1 1 .1

P R(-l)+LPi 2 R+L C LPRI LPRi CR) +Pi 2 CR LP

-

LPRi (RLPCR)LPi P R)--LPi (RL C P 2 R)+LPi (RL C 1

L + iLPR CR LP R P 2 CR) 2 P 2 L + R) C -LP 2 CR = A(cos o + i sin 0), say Equating real and imngin*ry parts, we get
-

112 R P 2 L P 2 R) + L C 2 LP (R
-

A TEXTBOOK OF ENGINEERING MATHEMATICS

A cos 0= A Sin

LPR (R LP CR) 2 P 2 CR) + L LP CR 2 Squaring and adding (1) and (2),

=
-

A2
=

L +L 2 R P 4 R (R 2 P 2 CR) LP CR) + L 2 1 P 2 [(R LP

-

CR) LP ] [L (R 2 R L 2 P ] P 2 CR) +L [(RLP 2

-

L R P 2 P 2 CR) + L LP (R 2

A=
-

LPR
CR) 2 LP
+

P 2 L

Dividing (2) by (1), tan 0R R-LP C 2 LP

R1_U,2c) 1 [

0=tan.j Hence

in0)=Ae 1 p=A(cos0+is

where

A=

LPR

I(R

and 0=taxr
+

1R(1_LP2C) 1
I

CR) 2 LP

P 2 L

U,

3.5. TRIGONOMETRIC FUNCTIONS OF A COMPLEX VARIABLE

1. Definitions. For all real values of x, we know that and e=cosxisinx e=cosx+isinx
Adding and subtracting, we get cos
= e + e = e

2z 2 These are called Eulers Exponential values of sin x and cos x, where x e R. If z = x + iy then trigonometric functions of z are defined as follows: e sin z e e e + = tanz= , smz= COsz= , cosz j(ez+e) 2i 2

cos

i (e
e

e) e

1
secz=

2
,

Sin Z

COS Z

1 cosecz= sm Z

2i
e

2. Eulers Theorem. For all values of 0, real or complex,

cos 0+ 1 sin 0. 1 e

FUNCTIONS OF A COMPLEX VARIABLE

=

113
e +e

For all values of 0, real or complex cos 0 cosO+isin6=

and sin 9

9 + ee e e 5 e_S =_=eiO. 2e + 2 Hence e 9 = COS 0 + i sin 9 for all values of 0. 3. Periodicity of Circular Functions. (a) To prove that sin z and cos z are periodic functions with period 2it.

We know that

sin z

iz_

iz

If n is any integer, then

sin (z

2nic)
= =

2 e

2 e_Z+

2i e
iz

2niu

e -iz e -2n,u 2i

e iz

e -iz

2i

u 2 e

9 2 e

= sin z sin z remains unchanged when z is increased by any multiple of 2it. sin z is a periodic function with period 2it. Similarly, cos z can be shown to be a periodic function with period 2it. (b) To prove that tan z is a periodic function with period it.

We know that Ifn is any integer, tan (z

tan z
+

e e i(e +e ) e (z +

e (z +
=

i[e + c_ (z+n)J Multiplying the numerator and denominator by e 1

= iz
-

e e e +e e 1 i[e e ] 1

2ni

iz
=

.e +e] i(e +e) i[e 2 tan z remains unchanged when z is increased by any multiple of it. tan z is a periodic function with period it.

tanz

e 2 -1

3.6. TRIGONOMETRIC IDENTITIES

If z is a complex variable, prove that (i) sin 2 z + cos 2z= 1 (ii) sin 2z
z 2z=cos z z =2cos sin --1 (iii) cos 2 (iv)tan2z= (vi) srn 3z
=
=

2 sin z cos z

1 2sin z 2 (v) sin (z)=sinz

2tanz 1tan z 3z 3 sin z 4 sin z 3 Stanztan

=

(vu) tan 3z

13tan 2

114 Proof. (i)L.H.S.

= +

A TEXTBOOK OF ENGINEERING MATHEMA1ICS

(e1x e

)2 +

(e +e

)2

(e2iz + e -2) +

2 (e

2 +2) = e

R.H.S.

(ii)
(ilL)

RH.s.=2SflZOBZ=2.e;.e+;=e_; (ez +

=sin2=L.H.S.

2z coe

z=

(e
,

e 2i
e 2iz +e -2Lz 2z

e_21Z+ 2iz+e .e _1(e l_ (e;e)l 2 1= ) 2cos z 2 e e i ( 2

)

n = 2 12sr z

=1+.(e2ze__2)=e

+e

=cos2z

Hence the result.

() R..H.S

2 tan z
1 tan z

2.

e e (ez + e) eiz e 12

e) (e + e) 2 + (e i [(eu + e)

2 (e

[i(e +e)j

e ) 4 i.2(e+e)
2 (e

(v)sin(z).
(vi) _e
&z

e e 2 tBfl2EL.H.8. i(e+e) 1 e eW 4 e ainz. =_____

-

e4dg

X 11

\\

(xy)+3xy(xy)1 [(e_e9+3.ek.e(Se)]

*[(2iajnz) +3(2isinz)]=[&sinz+6i.inzI3sia*4siaz.
tan 8 3tanz z

(vii) R.H8.

n 2 13ta z

1 f

i(e +e)j [i(e 3 [

+e)j

i(e +e)Ji1e 8 L

1 [e$_ +e

FUNCTIONS OF A COMPLEX VARIABLE

115

x 1(x 8 3--+--.i L3 iy
=

1+3

1
8 x

,wherex=ewe,y=e+e

2 3xy

x (3e + 3e + 6 + e 2 2 + e 2) 2 + 3x 3 iy iy (y ) iy (e + e +2 + 3e + 3e 6) 2 +j) 2 +4e +4) 2 x(4e (ee)(e +e -2iz 4)i(ez +e)(e +e iz 2 iy(4e 1) 2
+

y 2 2 2 + 3x y

x (3y 2 x ) 2

e e 8 = i(e +e) = tan 3z.

[. L
EXERCISE 3.1

2 + ab) = a (a bXa 8 2 +b i-b 8 3 22 (a+bXa ab)=a +b

1. 2.

Ifz=x+iy,findthereaandiaginarypartsof(i)e()exp(). Prove that

(i)ain(+nO)esmnO=esina I. 1 2 z

8.

0, where z = = 2 z+1 If z is a complex number, prove that (i)cos(z)=cosz

(zu)coe3z=4coaz3coaz

(ii)tan(z)=tanz sin2z (w)tanz= 1+coa2z

4.

If z ,z 1 2 are complex numbers, show that

+ (i)sin(z ) c + 2 s 1 ()coe ) c + s 1 =sinz =cosz cosz sinz inz inz osz osz z 2 z (z tan z 2 1 tan z 1 z 1+z z 3 +2sin 1 (w)smz = 2 smz + 1 (uz)tan(z ) 2 = z tanz 1 1tanz 2 2 2 z 2 z 1 +z 1 z 2 2 = 2am 1 coo z (u) coa z 2 2 5. Show that

(j)coe(cz+j)= 1(eP+eP)cosa+ (ee)sina ()sin(a+i)= (e4+e)sma+

6.

If a and

are the imaginary cube roots of unity, prove that

ae+pefr= _e2[coax+.Thsin4x).

Answers
(1) e cos y,

ex Sm y

(ii) e

coa 2xy, e Y sin 2xy.

-

116

A TEXTBOOK OF ENGINEERING MATHEMATICS

3.7. LOGARITHMIC FUNCTION OF A COMPLEX VARIABLE

Definition. If co = ez, where z and w are complex numbers, thenz is called a logarithm of 0 (0= z. w to the base e. Thus log 0 0 is a many-valued function. 1. Prove that log 2 = cos 2nit + i sin 2nic = 1 e We know that z.e e ez+ 1 =ez.1=w ez=w, then 2 Let By definition log 0 w = z + 2niti, where n is zero, or any + ye or ye integer. Thus if z be a logarithm of o, so isz + 2niti. Hence the logarithm of a complex number has infinite values and is thus a many-value function. 0 w. 0 w and is denoted by Log Note. The value z + 2n3ti is called the general value of log w=z+2niti=2niti+log Log w 0 Thus Ifw=x+iy,then Log(x+iy)=2nitilog(x+iy). 0 w. If we put n o, in the general value, we get the principal value of z, i.e., log 2. Prove that log ( N) = iti + log N, where N is positive. Proof. N = N( 1) = N(cos it + i sin it) = N log ( N) = log (N eim) = log N + log e = log N + iti. 3. Separate Log (cc + i13) into real and imaginary parts.
..

Proof. Let
..

2+132,O=tan_1 1 a+if3=r(cos9+isin9)sothatr= Jc

.L iO

Log (a +

i13)

= = =

2niti

2niti +

log (a + i13) = 2nni + log [r(cos 0 + i sin 0)] log (r e ) = 2nrci + log r + log e = 2nici + log r 5

2niti + log /a2 +132 + i tan

=

2niti

log (a 2

l32) + i tan 1

Re[Log (a Im{Log (a Note. Putting n

+ +

i1)] i13)]

log (a 2 2nn

+ 132)

+ tan
=

0, the principal value of log (a + i13)

- log (a 2 132) +

ILLUSTRATIVE EXAMPLES
Example 1. Prove thatlog (1 Deduce that log (1 +cos 0 Sol. log (1
+ + +

reL)

log (1

2r cos 9

1 ) + i tan 2 r

r SLfl 0 1+rcos0

isin 0)= log2cos-)+z-.

re)

log [1 + r(cos 0 + i sin 0)] = log [(1 + r cos 0) + i(r sin 0)] rsin0 . 1 = log [(1 + r cos 9)2 + (r sm Q)2] + i tan 1+rcosO 2
.

sin 9]+itair 2 os 2 0+r 2 log[12rcosO+r c

-

1 2

rsinO 1+rcos0

FUNCTIONS OF A COMPLEX VARIABLE

117

2 1+rcos6 Now log(1+cos0+zsm0)=log(1e) Puttmgr = 1 in(z),

log(1
+ cos

log[1+2rcos0r2]itan_1

rsin0

sm

0) 1

1
=

log(1

2 cos 0 1)

+ i

1 tan-

sinO l+cosO

-.

-.

log [2(1

cos 0)]

.0 0 2smcos 2 2 i tan- 20 2cos

-

--

log [2.2

cos2

..]+i tan 4 tan

J=

.. io[ cos

] 2 oj

Example 2. Find the general value of log ( 3). 3=3(1)=3cisit=3e 1 Sol. Log (3) = Log (3e9 = 2niti + log (3 e) = 2nici + log 3 + log e = 2niti + log 3 + ut = log 3 + i(2n + 1)ic. Example 3. Separate into real and imaginary parts Log (4 + 3i). (M.D.U. Dec. 2006) Sol.Let 43i=r(cosO+isinO) Equating real and imaginary parts r cos 0 = 4; r sin 0 = 3 Squaring and adding, r 2 = 16 + 9 = 25 r=5
..

Dividing, Log(4

tan 0=
+

..

0= tan
=

3i)

Log[r(cos 8 i sin 0)]

Log (re)

2nni

log

(10)

4 =2niu+logr+ loge=2niti+log5zO=log5+2rnti+itan j Re[Log(43i)]=log5 Im[Log (4

+

3i)1

(2n1c

4 tan

Example 4. Prove that tan (i log

+

b a ib) = 2
=

a+ib=r(cosO+isinO) Equating real and imaginary parts r cos 0 Dividing, Also tan 0= aib=r(cosOisinO)

SoLLet

a, r sin 0= b

118 L.H.S.=

A TEXTBOOK OF ENGINEERING MAThEMATICS

L = tan [i log 4 e tan [i( 2i0) log el ]

2

r(cosOzsmO)j

=tan2O=

[.. 2 a

of(i)]

2ab = a 2 b 2 lraniple 5. Express Log (Log i) in the form A

iB.

Sol.

i=c+isin=e
Logi=2nxi+loge=2ni+i =i(4n+1)

...

Log(Logi)=Log[i(4n+1)] =2mci+1og[i(4n+1]

=2miti+logilog(4n+l)
=2mici+loge+log(4n+1) =2mxi+i +log(4n+1)
=

log (4n

1)

i(4m

1)

3.8. THE GENERAL EXPONENTIAL FUNCTION

The general exponential function a is defined by the equation a = e z are any numbers, real or complex. Since Log a = 2n,ci + log a
nr+icga) z=( r 2

a,

where a and

Hence a is a many valued function and its principal value is obtained by putting n =0.

ILLUSTRATIVE EXAMPLES______________
lample 1. P?ove that jt is wholly real and find its principal value. Also show that the values of j jbrm aG.P. [Bydefinitioni =e 1 i SoL + log i) = ei[ flr4 + log (cci z/2 + i un I2)] 2 =
] 2 = e20

=e

+i/2] =e1(412 =

which is wholly real. The principal value of j = e the values of i are Putting n =0, 1,2 which form a G.P. whose common ratio is e.

(Putting n e

0)

2LV

IllS 1 +

V
(gpv)

J
+

a 1
= (gpLiX +

(gpLJ)

v)d

(,d)OI(5j

V)d =

(L UTS I + pL

soo) 2oj
+

(gi +

V)a =
=

(1uo

SOflIA

diuud unjL) 1
+
=

V) =

AWN !

w +v?

iS
lID] (2)

(S 9OO flJAS)

gy (q)
+

v= v=
,

g
GldwBxa
=

W] aaocI paJap2suo av saniva lvhI:13u1.Id uo puv g

_ags 113

II

j.

+(j+i) ;oj.Id{3j
=

a 01 [(x= ,j

[(

2oj

ujs

+(g 20{

//1L_g2old_

= 1 / 5 [

U 1

7+(+fl2O

I) jog

gffojp

+-j

soa

s p+r(J + 1)

Jo ]Jvd vai ai vip aaocf anvn 7vthouiid a C uo 2uiapisvo 2

+

adunx

= -d = + 7j
upp

[-[T

(I +

UJ) 3 U19 +

(I +
= 3 ,ci(I+111,)-

iLci(l +

pu 2uunbS

+ u7)u

ci

(I

u)Soo
&ruwt pu ea

sd
a 02 [OuWi+osoo=
...j

&iinb

?+ (I+
+ uI)I + +

ii+

U_a
UTS

= [g/u + flLUgJ(Jl +

D)a

+ iiLu3J(cii + D) 3 =

[(fU

+ pt

son) &j

+ ?2Lu}(T +

))a =

[1

ot

+ i1LuJ(+

)a

2oj (ci! +
+ Ufr).a

ji +

x? =
g

+ X)

05

(9oo rni)
61.1.

gj

+ gX) ]DZ/]

ci? +

= p

+.?Ji

1dun3xi

1VfldVA X31dV100 V dO SNOIIONfld

120

A TEXTBOOK OF ENGINEERING MATHEMATICS

Equating real and imaginary parts An 2 Ait ) sin 2 B = e_B 2 AnB = Dividing (ii) by (i), tan A 2
) 2 A = e_B

cos

...(ii)

2 Squaring and adding (i) and (ii), A

2 B

(co2 +sinI e.
-

An 2

Ani 2

...(II)

Example 5. If (a

ib)P

mY, then prove that y x

a) when only principal ) 2 log(a+b

2 tan 1

values are considered. (a+ib)P=mY Sol. both Taking log of sides, log (a + ib)P = log mx + Y or plog(a+ib)=(x+iy)logm or b1 1 p[log(a2+b2+itan _
aj

=xlogm+iylogm (Considering only the principal values)

Equating real and imaginary parts x log m

2 p log (a

) 2 b ...(ii)

1 ylogm=ptan

b a

ptan Dividing (ii) by (i), Example 6 If tan log (x

+

b a

) 2 2 b .-plog(a

2tan1 a 2 +b log(a ) 2

a+b 2 ib and 2 2a tan log (x+y )= 2 b 1a 2 Sol. tan log Cx + iy) = a + ib tanlog(xiy)_aib Now tan log (x ) = tan log (x + iy)(x iy) 2 2+y iy)
=

1, then prove that

...(ii) tan log(x + iy)+ tan log(x iy) 1tan iog(x+iy).tan log(xiy)

=tan[log(xiy).t-log(xiy)]= a+ib+aib
=

2a

b 2 1(a+ib)(aib)1_a

2 where a

2 b

1.

FUNCTIONS OF A COMPLEX VARIABLE

121

EXERCISE 3.2
1. 2. Find the general value of (i) log ( i) Prove that (i)ilog (_-. =ic2tarrx x+i) (iii) i = e
3.
(4n+1) 2

(ii) log (1

i).

(ii)cos

[
(iv) Ig ji =

Laib)j 2 +b a
I

Show that (i)log(1+itancz)=logseca+ia (iii) log (6

+

(ii)Log (4]=2i(rnt_tan_1]. (M.D.U. Dec. 2005)

1 Si) = log 10 + i tan

(iv)

4.
5.

Prove that sin i log I(1+je is wholly real. 1-e) Prove that sin (log i ) = 1. 1
.

6. 7. 8.

1 (1ix Prove that tan 1 x = log 2z 1ix Ifloglog(x+iy)=p+iq,showthat y=xtan[tanqlog Prove that the principal value of
3 (azbY
C a + ib) -i q
(j)X+iY

.1x2+y2].

iscos2(pa+qlogr)+isin2(pct+qlogr),wherer

1 1 a2+b2

anda=tan .
a

9. 10.

If

(1

)XiY

=c

i13, prove that one of the values of tan 13 is

+

itx

y log 2.

Prove that

(1 + j) = sin (log 2) (1 j)l+L

i cos (log 2).

11.

Prove that the real part of the principal value of Prove that Log. =
,

jbog

(1+

is e

cos

log 2].

12.
13.

4m+1 where m and n are integers. 4n+1 If (x, y) is a point on the circle having its centre at the origin and radius a, prove that 1 x+iya ax it. . =log--+. log a+x 2 x+iya 2 Provethat log 1 = 1_eie (1

14.

e .( e

122 15.

A TEXTBOOK OF ENGINEERING MATHEMATICS

(a + ib) = A + iB, prove that ) 2 2 + ib ) (a 1 If(a + ib 1 B 2+2 )=A 2 (a + b 2 ) 2 2+ b ) (a 2 1 (i) (a, + b (ii)
tan
1 a

+ tan

--

tan

--

tan_1.

2 a

Answers 1.
(i)(4n

1)

iti

1
(ii)

log2 +z(8n

1)

7t

3.9. HYPERBOLIC FUNCTIONS

1. Definitions. For all values of x, real or complex (i) the quantity (ii) the quantity Thus srnh x
e

e
+e

is called hyperbolic

sine of x

and is written as sinh x

is called hyperbolic cosine of x and is written as cosh x.

eX+e_X

eX_e_ =

2 The other hyperbolic functions follows: tanhx= sechx= Note. cosh x

+
sinhx

cosh x
are

2 defined in terms of hyperbolic sine and cosine as coshx sinh x 1 sinhx cosh
= = eX _e_X e ex +eX eX

eX _e_X =

cosh x 1 coshx

cothx= cosechx

+
=

e_X

2
eX +e

2
+
0 e

sinh

= e

e
=

2
eX +e

.Z-4 = 0;
2
eX

2
eX +e 1

1+1 2
e

1
= e.

2 2 2 = functions 2. (a) Relations between hyperbolic and trigonometric 2 ix in these equations, we get

ainh x

ex;

coshxsinhx=

cosO=

eLO +e 6

sin0=

eLS e 0

2i

Putting 0

ei(+e_i

cos (ix)

2
1 e

e +eX 2
e_X

cosh x
(e

eX =

e_X)

sin(ix)=
=

2i ex_e_x) i ( 2 2i
sin(ix)

2i
.

2i

eX_e_X

=1.

2
.

iSlfl hx

tan(ix)=

cos(ix) cos(ix) = cot(ix)= sin(ix)

isinhx itanhx coshx i coshx coshx =zcothx = isinhx i sinhx 2

FUNCTIONS OF A COMPLEX VARIABLE

=
(ix)

123
=

sec

1 cos(zx) 1 sin (ix)

1 coshx 1 i sinh x

sech x 1 =itosechx. i sinh x

2

cosec(ix)=

0 e e 0 0 +e e 9 _e_e e ;tanh6= ;coshO= (b)ByDefinition,sinhO= 2 2 e+e Putting 0 = ix, we get sinh(ix)= cosh
= (ix)

e
=

=isinx;

e =

cos

eiX

tanh(zx)=

ix

ix =1.
e

2i
+
e

e_ix

SiflX =i. COS X

=itanx

2 3. Prove that hyperbolic functions are periodic and find their periods. (a) We know that sinh
=
x e _e_X

sinh

(x

2nni)

1 2 e 2 e

=
=

where n is any integer

e e

. e [ex.e e 2 ]= [e.1e. 1] 2

=sinhx

Thus sinh x remains unchanged when x is increased by any multiple of 2iti. Hence sinh x is a periodic function and its period is 27ti.
=

(b) cosh
(x

cosh x
+

+e

eX + 2niti
2nti)
= =

2niti)
,

2 [ex

where n is any integer

]=l[eX

n 2 +e_x .e

.1+e_x .11=

eX e_X

coshx

Thus cosh x remains unchanged when x is increased by any multiple of 2ici. Hence cosh x is a periodic function and its period is 27ci.
eX_eC

Cc)
tanh

tanhx=
eX +e

(x

n7ti)

eX+7
=
x

_e_1)
,

where

is any integer

e+hl +e_+7Z

nni_

flld

.e +e_x .e

124

A TEXTBOOK OF ENGINEERING MATHEMATICS

Multiplying the numerator and denominator by

e .e _e_X 2 eX _e_X eX +e_X

eX .e +e 2

=tanhx

{ e= cos 2nit + i sin 2nit = 1] Thus tanh x remains unchanged when x is increased by any multiple of ni. Hence tanh x is a periodic function and its period is iti. Note. cosech x, sech x and coth x being reciprocals of sinh x, cosh x and ta.nh x respectively, are also periodic functions with periods 2ici, 2ni and ni respectively.
3.. FORMULAE OF HYPERBOLIC FUNCTIONS 1. Prove that (a) cosh 2 x sinh 2 x = 1, (b) sech 2 x + tanh 2 x = 1, (c) coth 2 x cosech 2x= 1 Proof. (a) For all values of 0, cos 2 0 + sin 20= 1 Putting 8 = ix, we get cos 2 (ix) + sin 2 (ix) = 1 (cosh x) cos ix = cosh x; sin (ix) = I sinh xl 2 + (i sinh x) 2= 1 { 2 x sinh cosh 2x= 1 2 = 11 [... i 2 x sinh cosh (b) We know that 2x= 1 Dividing both sides by cosh 2 x, we have 1 tanh 2 x = sech 2 x = sech 2 x + tanh 2x = 1 2 x sinh 2x= 1 (c) We know that cosh Dividing both sides by sinh 2 x, we have 2 x 1 = cosech coth 2 x = coth 2 x cosech 2x= 1 2. Prove that (a) sinh (x y) = sinh x cosh y cosh x sinh y (b) cosh (x y) = cosh x cosh y sinh x sinh y tanh x tanh y (c) tanh (x y) = 1 tanh x tanh y

or or

Proof. (a) sinh(x y)=

=

sin i (x y)

[.

sinhx 4 sin

X]

(sin ix cos iy cos ix sin iy) (isinhxcoshycoshx. isinhy)

(b)

(c)

[... sini0=isinhe;cosi0=cosh8] sinh x cosh y cosh x sinh i cosh (x y) = cos i (x y) [. cosh x = cos ix] =cosixcosiy sinixsiniy=coshxcoshy p isinhx. isinhy =coshxcoshy (sinhx.sinhy) =1] 2 i = cosh x cosh y sinh x sinh y sinh(x y) sinh x cosh y cosh x sinhy tanh(xy)= = cosh(x y) cosh x cosh y sinh x sinhy
=

Dwithng the numerator and denominator by cosh x cosh

tanhxtanhy 1 h tanh y

FUNCTIONS OF A COMPLEX VARIABLE

125 2 sirth x cosh x 2 cosh 2x 2 tanh x 1tanh x 2

=

3. Prove that (a) sinh 2x

(b) cosh 2x
(c) tanh 2x

2x cosh

2x sinh

2x 2 sinh

x 2 1+ tanh 2x 1tanh

or

2 tanh x 2x 1 + tanh Proof. (a) We know that sin 20 = 2 sin 0 cos 0 Putting 0 = ix, we get sin (2ix) = 2 sin (ix) cos (ix) or i sinh 2x sinh2x=2sinhxcoshx 2 tan 9 Also sin 20 = 20 1+tan 2tanix 2.itanhx Putting 0 = ix, we get sin (2ix) = = ix 2 2 1+tan 1+(itanhx)
=

2 i sinh x. cosh x
.

or

or

or

1tanh x 2 2 1tan x h 28 2 8 sin (b) We know that cos 28 = cos 2 (ix) or cosh 2x = (cosh x) sin Putting 9 ix, we get cos (2ix) = cos 2 (ix) 2 2 (i sinh x) 2 x sinh 2x cosh 2x = cosh 20 1 cos 28 = 2 cos We know that 2x 1 2 (ix) 1 or cosh 2x = 2 cosh Putting 8 = ix, we get cos (2ix) = 2 cos cosh 2x +1 2x= cosh Cor. 2 cos2O=12sin O 2 Weknowthat 2 (ix) Putting 0 = ix, we get cos (2ix) = 1 2 sin 2x cosh 2x 1 2 (i sinh x) 2 = 1 + 2 sinh cosh 2x 1 Cor. 2 2 8 1tan cos 29 = We know that 2 8 1+ tan

isinh2x=

2itanhx

or

sinh

2tanhx

or

2 1tan (ix) 1(itanhx) 2 2 + tan 1 1 + (i tanh x) 2 (ix) 2 x 1+tanh cosh2x= 2x 1 tanh 2 tan 9 (c) We know that tan 20 = 2 0 1tan Putting 9 = ix, we get 2 tan (ix) tan (2ix) = 2 (ix) 1 tan 2itanhx 2itanhx itanh2x= = 2 1+tanh 1(itanhx) 2 x

Putting 0

ix, we get cos (2ix)

126

A TEXTBOOK OF ENGINEERING MATHEMATICS

tanh2x=

2 tanh x 1 + tanh 2x
+

4. Prove that (a) sinh 3x = 3 sinh x

x3coshx 3 (b)cosh3x=4cosh

3 x. 4 sinh (c)tanh3x= 3tanhx+tanh x 3 x 2 1+3tanh

or or or

or

Proof. (a) We know that sin 30 = 3 sin 04 sin 30 Putting 0 = ix, we get sin (3ix) = 3 sin (ix) 4 sin 3 (ix) i sinh 3x = 3i sinh x 4 (i sixth x) 3 i sinh 3x = 3i sinh x + 4 i sinh 3x x 3 sinh3x=3sinhx+4sinh (b) We know that cos 30 = 4 cos 3 0 3 cos 8 Putting 0 = ix, we get cos (3ix) = 4 cos 3 (ix) 3 cos (ix) cosh 3x = 4 cosh 3 x 3 cosh x

3= i

i]

or or

3tan0tan 0 3 20 13tan 3 (ix) Puttmg 8 = ix we get tan (3zx) = 3tan(ix)tan 2 (ix) 13tan 3 3.itanhx(itanhx) xtanhdx= 2 13(xtanhx) (c)Weknowthat tan30=
,

tanh 3x =

3. i tanh x + i tanh 3x 1+3tanh 2 x

or tanh 3x =

3x 3 tanh x + tanh 1+3tanh 2 x

or or or

5. Prove that Ci) 2 sixth A cosh B = sinh (A + B) + sinh (A B) (ii) 2 cosh A sinh B = sinh (A + B) sinh (A B) (iii) 2 cosh A cosh B = cosh (A + B) + cosh (A B) (iv) 2 sixth A sinh B = cosh (A + B) cosh (A B) Proof. We shall prove only the last result. The first three are left as an exercise for the student. We know that 2 sin x sin y = cos (x y) cos (x +y) Putting x = iA; y = iB, we get 2 sin (iA). sin (iB) = cos i (A B) cos i (A + B) 2. i sixth A. i sinh B = cosh (A B) cosh (A + B) 2 sinh A sinh B = cosh (A B) cosh (A + B) 2 sinh A sinh B = cosh (A + B) cosh (A B). 6. Prove that C+D C D cosh (i) sixth C + sinh D = 2 sixth

[:

2= i

1]

(ii) sixth C

sinh D =2 cosh C + D 2

C D 2

FUNCTIONS OF A COMPLEX VARIABLE

127

(iii)coshC+coshD=2cosh C+DcoshC_D 2 2 (iv)coshCcosh.D=2sinh C+DSjflhC_D 2 2 Proof. We shall prove only the last result. The first three are left as an exercise for the student. x+y . yx sin We know that cos x cos y = 2 sm 2 2 Putting x = iA andy = iB, we get (.A+B (.B-A cos CiA) cos (iB) = 2 sin 2 2 A+B.B_A coshAcoshB=2isinh

2 sinh

B-A A+B . sinh 2 2

2 sinh

A-B A+B . sinh 2 2 [.; sinh(x)=sinhx]

7. Prove that

tanh(x+y+z)=

tanh x + tanh y + tanh z + tanh x tanh y tanh z 1+tanhx tanhy+tanhytanhz+tanhztanhx tan a + tan f3 + tan y tan a tan I tan

Proof. We know that,

tan (a+13y)=

Putting a= ix; 3= iy; y= iz, we get

or

tan (ix) + tan (iy) + tan (iz) tan (ix) tan (iy) tan (iz) 1tan(ix)tan(iy)tan(iy)tan(iz)tan(iz)tan(ix) itanhx+itanhy+itanhzitanhx.itanhy .itanhz itanh(x+y+z)= 1itanhx.itanhyitanhy.itanhzitanhz.itanhx tanh x + tanh y + tanh z + tanh x tanh y tanh z tanh(x+y+z)= 1+taniixtaniy+taniytax1+taniiztathx
tani(x+y+z)=

ILLUSTRATIVE EXAMPLES
1. Separate into real and imaginary parts (b)cos(xiy) (a) sin (xiy) (d)cot(x+iy) (C) tan (x+iy) (f)cosec(x+iy). (e)sec(x+iy) SoL(a) sin(xiy)=sinxcosiy+cosxsiniy = sinx coshy + cosx. i sinhy = sinx coshy cos(xi-iy)=cosxcosiyainxsiniy (b) sin x. i sinh y = cos x cosh y = cos x cosh y
Example

i. i
.

coax sinhy

sin x

sinh y

128

A TEXTBOOK OF ENGINEERING MATHEMATICS

(c)

tan(x+zy)=

sin (x + iy) 2 sin (x + iy) cos (x iy) = cos(x+iy) 2cos(x+iy)cos(xiy)

(d)

sin2x+sin2iy 2sinAcosB=sin(A+B)+sin(AB) 2 cos A cos B = cos(A + B) + cos(A B) = cos 2x + cos 2iy sin2x+i.sinh2y sinh2y sin2x cos2x+cosh2y cos2x+cosh2y cos2x+cosh2y cos(x+iy)2cos(x+iy)sin(xiy) cot(x+iy)= sin(x+iy) 2sin(x+zy)srn(xiy)

[: [

2cosAsinB=sin(A+B)sin(AB) sin2xsin2iy 2 sin A Sin B = COS (A B) COS (A + B) cos 2iy cos 2x sin 2x i sinh 2y sin 2x sinh 2y = cosh 2y cos 2x 2x cosh 2y cos cosh 2y cos 2x 1 2cos(xiy) sec(xiy)= = (e) cos(x+zy) 2cos(x+zy)cos(xiy) 2 (cos x cos iy + sin x sin iy) = 2 (cos x cosh y + sin x. i sinh y) = COB 2x + cos 2iy cos 2x + cosh 2y 2 cos x cosh y 2 sin x sinh y cos2x+cosh2y cos2x+cosh2y 1 2sin(xiy) = . CD cosec Cx + zy) = sin(x+zy) 2sin(x+iy)sin(xiy) 2 (sin x cos iy cos x sin iy) = 2 (sin x cosh y + cos x . i sinh y) COB 2iy cos 2x cosh 2y cos 2x 2 sin x cosh y 2 COB x sinh y . cosh2ycos2x cosh2ycosh2x Example 2. Separate the following into real and imaginary parts: (a) sinh (x + iy) (b) cosh (x + iy) (c) tanh (x + iy) (d) coth (x + iy) sech + iy) (e) (x (f) cosech (x + iy). 1.. Sol. (a) srnh (x + zy) = sin z (x + iy) [. i srnh 9 = sin 19]

L [

= =

sin (ix

y) =

i (sin ix

cos y

cos

ix

sin y) COB y +

(b)

i (i sinh x cos y cosh(x+iy)=cosi(x+iy) = cos (ix y) = cos

cosh x sin y) = sinh x

COB y +

i cosh x sin y [.. cosh9=cosiOl = cosh sin ix sin y x cos y + i sirth x sin y

(c)

tani(x+iy) 4 tanh(xiy)= j tan(zxy)=z

[:

itanh9=taniO]

sin (ix y)
cos(zxy)
. .

.
=i.

2 sin (ix 9) COB (ix + y) 2Cos(lxy)cosCzx+y)

sin2ixsin2y . cos2ix+cos2y

isinh2xsin2y cosh2x+cos2y

FUNCTIONS OF A COMPLEX VARIABLE

129

(d)

sinh2x sin2y 1. cosh 2x + cos 2y cosh 2x + cos 2y cosh(x+iy) cosi(x+iy) .cos(ixy) coth(x+iy)= sinh(x+zy) -sini(x+iy) sm(ixy)

(e)

sech (x

2sin(ix+y)cos(ixy). sin2ix+sin2y 2sin(ix+y)sin(ixy) cos2ycos2ix . i sinh 2x + sin 2y sinh 2x sin 2y . 1. cos 2y cosh 2x cos 2y cosh 2x cos 2y cosh 2x sinh 2y sin 2y cosh2xcos2y cosh2xcos2y 1 1 zy) = = cosh (x + zy) cos z(x + zy) 1 2cos(ix+y) = 2(cosixcosysinixsiny) cos(ixy) 2cos(ix+y)cos(ixy) cos2ix+cos2y 2(cosh x cos y i sinh x sin y) cosh 2x + cos 2 cosh x COB y 2 sinh x sin y . cosh 2x + cos 2y cosh 2x + COB 2y

cosech(x+iy)=
.

sinh(x+iY)lSjni(X+iY)

sin(ixy)

2sin(ix+y) 2 sin (ix + y) sin (ix y) 2(sin ix cos y + cos ix siny) 2(i sinh x cos y + cosh x sin y) =. cos 2y cos 2ix cos 2y cosh 2x 2sinhxcosy 2coshxsiny cos 2y cosh 2x cos 2y cosh 2x 2 sinh x cos y 2 cosh x sin y . cosh2xcos2y cosh2xcos2y

Example 3.If u (i) tanh

=

log tan
+

then prove that (ii) cosh u

=

tan (P.T.U. 2006)

sec

e.

Sol. (i)

u=iogtan(+J
eu=tn(+)

130 1+tan 2 0 1tan 2 By componendo and dividendo e u/2 .e u12_

A TEXTBOOK OF ENGINEERING MATHEMATICS

e u12
u/2

1+tan 2 1tan 2

(
2 e eU 2 2 +e e 2

e (
tanh =tan

itanEJ:(i_tanE 1+tan 2 = 1_tanh2 2 1tan 2 2 1 = see 0. 1+tanh

(ii)

cosh u

[Using part (i)]

Example 4. If sin (A
X

iB)

x +

iy, prove that

2 A = 1. 2 cosec 2 A 9 sec = 1 (ii) x cosh B sinh B Sol. x + iy = sin (A + iB) = sin A cos iB + cos A sin iB = sin A cosh B + i cos A sinh B Equating real and imaginary parts on both sides x = sin A cosh B; y = cos A sinh B

From (z),

cosh B

sin A;
.

y sinh B
2
+

cos A
2

Squaringandadding, Also from

x

cosh B cosh B;

2 smh B =

=1 A+cos 2 =sin A

(t),

sinA

y cosA

smh B 2B sinh
=

Squaring and subtracting, or Example 5. If x

X +

= cosh 2B sin A cos A y =1. ec cosec x A s A 2 iy = cosh (u + iv) show that

Y + 2 v 9 cosec 2 v = 1. 2 see = 1 (ii) x cosh u stnh u iO] x + iy = cosh (u + iv) = cos i(u + iv) Sol. [. cosh e = i sinh sin v u = cosh cos v + u sin iu sin v = cos cos v + = cos (iu iu v) Equating the real and imaginary parts x=coshucosv;y=sinhusinv

From

(i),

COSzi U

cos u-, . y smh u

sin v

FUNCTIONS OF A COMPLEX ARIABLE

131

Squanng and adding, From

(i),

cosh 2 u

sinh 2 u

2v cos

2v sin

or or

x y = smh u = cosh u; cos U Sifl v Squaring and subtracting, x 2 u sinh 2 sec 2v y 2 v = cosh 2u 2 cosec Example 6. If x + iy = tan (A + iB); prove that 2 +y (i)x 2 +2xcot2A = .7 y =0. + (ii)x 2 2ycoth2B+1 x+iy=tan(AiB) Sol. x iy = tan (A iB) Changing i into i, we get tan 2A = tan [(A + iB) + (A iB)] Now tan(A+iB)+tan(AiB) = (x+iy)+(xiy) 1tan(A+iB)tan(AiB) 1(x+iy)(xiy) 1 2x or 1 (x 2 + y 2 ) 2x cot 2A = cot2A 1(x 2 +y 2 ) 2 + + 2x cot 2A = 1 x Again tan (2iB) = tan [(A + iB) (A iB)1 tan (A + iB) tan (A iB) = (x + iy) (x iy) 1+tan(A+iB)tan(AiB) 1(x+iy)(xiy)

1.

2x +y 1(x ) 2

. .

.(I)

2iy +y 1+x 2
+

or

1 = 2? or or coth2B 1+x +y 2 1+x +y y + x 2ycoth2B+1=0 Hence 2 If tan Example 7. (0 + i) = cos a + i sin a = eia, prove that

i tanh 2B

2 y

2y coth 2B ...(II)

(Delhi, 2002)
Sol. tan (9 + i4) = cos a + i sin a Changing i into i, we get tan (ei4)=cosaisina

..(ii)

Now

tan (9 + i) + tan (0 i) tan29=tan[(O+z)+(0z)]= 1tan(0+i)tan(9i) (cosa+isina)(cosaisina) 1(cosa+isincx)(cosaisina) 2cosa 2cosa 2 a + sin 2 a) 2 sin 2 a) 1 (cos = 1 (cos 2 a i It 2cosa 2cosa = oo = = = tan 2 1i 0

2:=nm:-

tanO=tana

0=nit+a]

132 Also tan 2i

=

A TEXTBOOK OF ENGINEERING MATHEMATICS

tan [(0 + i) (0 ip)] tan (0+ i) tan (0 i4) (cos a i sin a) (cos a i sin a) 1+tan(O+i)tan(Oi) 1+(cosa+isina)(coscisina) 2isina 2isina = = 2 2 1+1 1+(cos cx+sin a) i tanh 24 = i sin a or tanh = sin a

or

2 sin a e 2 0 e 2 e or 2= 0 1 20 + e e 2 e By componendo and dividendo

2e 20 2e -20 1 + sin a 1sina
.

e 2 2 e

1 sin a 1 + sin a 1sina

.

or e=

or

or

a 2 . a. a 2 a a a cos+sm 2 2 2 2= 2 2 = 4 e 0 a 2 a 2 . a a. a a cos +sin 2cossin cossin 2 2 2 2 2 2 a .a a cos+sin 1+tan a 2 2 =tanl+ 2 e= = a .a a 42 l_tanj

cos +sin +2cossin
.

Taking logarithms of both sides log e 20

log tan
+

or 2

log tan

=logtan + 2 4 2 Example 8. Separate into eal and imaginary parts log sin (x + iy). Sol. log sin (x + iy) = log (sin x cos iy + cos x sin iy) = log (sin x cosh y + i cos x sinh y) = log (a + iI), where a = sin x cosh y, f3 = cos x sinh
=

log (a 2

2)

i tan-
+

= -

log 2 cosh y (sin x 2

y) 2 2 x sinh cos
+

i tan-

smh

=logi 2 L
= =
.-

r 1 cos 2x
2

smxcoshy

cosh 2y + 1 2

1 + cos 2x cosh 2y 1 2 2 itan(cotxtanhy)

+

log [- (2 cosh 2y

2 cos 2x)]

i tan1 (cot x tanh y)

Example 9. If z = bounded. Sol. We know that sin z = sin (x + iy) = sin x cosh y + i cos x sinh y I sinz 12= I sinxcoshyicosxsinhy 12

log (cosh 2y cos 2x)] + i tan- (cot x tanh y). x + iy is a complex variable, then prove that sin z and cos z are not
.

[4-

FUNCTIONS OF A COMPLEX VARIABLE

= = =

133

sin x cosh 2 2 y + cos 2 x sinh 2y 2 x (1 + sinh sin 2 y) + (1 sin 2 x) sinh 2y 2 x + sinh sin 2y

Since x is real, 0

2x sin

1. But sinh

=
,

is not bounded.

and

I sin z I and hence sin z is not bounded. Similarly cos z 2 = cos 2 x + sinh 2y I cos z I and hence cos z is not bounded. Hence for a complex variable z, sin z and cos z can have any value. Remark. I sinz I Si and I cosz ionlywhenzis real. Example 10. Find all vzlues of z such that (ii)coshz=O. (i)sinhz=O Sol. We know that sinh z = sinh Cx + iy) = sinh x cos y + i cosh x sin y [See Example 2 (a) and (bXl cosh z = cosh (x + iy) = cosh x cos y + i sinh x sin y 2 2 = sinh x cos y + i cosh x sin y I sinh z 2y = sinh 2 x cos 2 y + cosh 2 x sin = sir}i 2 y) + (1 + sinh 2 x) sin 2y 2 x (1 sin = sinh 2 x + sin 2y =. lsinhzl=O sinhz=0 Now x+sin sinh y 2 =O and sin y = 0 sinhx=O and x=0 y = nt, where n is any integer. z=x+iy=n7ti sinh z = 0 only when z is purely imaginary and z = niti. cosh z 2 = sinh 2 x + cos 2y Similarly I coshz I =0 cosh z = 0 Now

x+cos sinh y 2 =0
=

sinh x

and
and
+

cos y
y
+

0
(2n
+

x=0

1)

-,

where n is any integer.

z
cosh

iy

(2n

1)

i
=

0 only when z is

purely

imaginary and z
=

(2n +

1)

Example 11. Find all values of z such that sin z Sol. Let z = x + iy, then sin z = 4 sin(x+iy)=4 sin x cosh y + i cos x sinh y = 4 sinxcoshy=4 cosxsinhy=0 and by comparing real and imaginary parts.
=-

4.

134 From (2), we have cos x x When y

= =

A TEXTBOOK OF ENGINEERING MATHEMATICS

0 (2n

or sinhy
+

0 0

1)

where n is any integer or y

0, from (1), we get sin x = 4 which is not possible since x is real. Whenx
=

(.

cosh 0

1)

(2n
+

1)

-,

from (1), we get or

sin
or

cosh y =4

C 1Y sin

cosh y =4

(1Ycoshy=4 When n is odd, cosh y = 4 which is not possible since cosh y> 0 for every y. .. n must be an even integer. coshy=4 or y=cosh4 From(3) Hence z=x+iy

(2n (4k

+ +

1) 1)

+ +

i cosh 4, i cosh 4,

where n is an even integer where k is any integer.

or

3.11. INVERSE HYPERBOLIC FUNCTIONS 1. Prove that sinir x = log (x + gx2 +1) Let sinhx=y, thenx=sinhy x= It is a quadratic in e 2x,J4x2+4 e= Rejecting the negative sign, e Takinglogarithms,
=

ee
2

1 2 e
2e

2 eY..2xeY..10

2.
x + ix2 +

1 or sinhx=log(x+
/x2+1)

y=log(x /x2+1)
Jx2 1
=

4 x = log (x + 2. Prove that cosh cosh x = y then x Let

or x=

1) cosh y

e y 2

y e2 2e

or

e2xe+1=O

It is a quadratic in e ey=2j4x2_4=x/x2_l 2 Rejectmg the negative sign e

=

+ ,Jx2

FUNCTIONS OF A COMPLEX VARIABLE

135
+ Jx2

log Cx
-

1)

or

1x cosh

log (x

+ Jx2

1)

3. Prove that tanh x

Let y
=

log
=

1+ X 1x tanh y e
=

tanh x, then x

or

ee
-

e+e

or

e 3 +e ee

By componendo and dividendo log

1+x 1x

2 e
1 log 1+x

1+x

i;;

or y

1 log 1+x 2 1x Example 1. Separate into real and imaginary parts tanh x
=

1 (i) 51w

(os

z sin 0), 0

<

<

1 (x + iy). (ii) tan 1 (cos 0 + i sin 8) = x + iy SoL (i) Let sin cos0+isinO=sin(x+ iy)=sinxcos iycosxsiniy = sin x cosh y + i cos x sinh y
Equating real and imaginary parts, we have cos 0 and
=

sin x cosh y

sin9=cosxsinhy Squaring (i) and (ii) and adding, we have 2y 2 y + cos 2 x sinh 2 x cosh 1 = sin
=

2 x(i sin

2 y) sinh
=

(1

2 x) sinh sin 2y
=

2x sin

2y sinh

or From (ii),

x=cos 1sin x sinh y 2 = 2 20 sin cos x Real part x

= = =

2y 2 x sinh cos

4x cos

2x cos

sin 0

1 cos sinG
CoSX
=

From (ii),

sirth y

sinO i- jsin0

y=sinh_1(J)

=log[l+/1+sin0] Imaginarypart y=1og(f+f1+sin0)

+

sinh_1x=iog(x+/2)

(ii) Let
then

1 (x tan tan (x

iy) iy)

= it + =

iv iv

136

A TEXTBOOK OF ENGINEERING MATHEMATICS

Adding (i) and (ii), we have 2u = tan (x

iy)

tan- (x

iy)

2x (x+iy)+(xiy) =tan 1(x+zyXxiy) 1x 2 2 y 2x 1 u=tan 2 1x 2 y 2 Subtracting (ii) from (i), we have 2iv = tan- (x + iy) tan- (x iy) 2iy (xiy)(xiy) = tan -1 = tan- 1 l+(x+iy)(xiy) +y 2 1+x

2iy

tan 2iv = 9 1+x 2 +y

2iy i tanh 2v = 1+x 2 + 2

v=

1 tanh 1+x 2 +y 2. 2

. 3 Example 2. Find all values of z such that sinh z = e Sol. Let z = x + iy, then 3 sinhz=e or

= sinh(xiy)=e 3

sinh x cos v

+ i

cosh x sin

= cos

It + i

sin

it

+ &

Equating real and imaginary parts, 1 sinhxcosy= 2 cosh x sin y = Using cosh 2x 3
2

= sinhx= 2cosy
cosh x = 2smy

2 2 x = 1, we have sinh 1

or or or

y cos 2 = 1 or 3 cos y sm 2 y = 4 sin 2 y 2 4sin y 4cos 2 y y = 4 sin 2 y) sin 2 y (1 sin 2 y) 2 3(1 sin 3=O y8sin 4 4sin y 2 in 3)(2sin 1)=O s ( y y 2

2 siny=

or

Rejecting sin y= 2

y 2 since sin
1 = sin y 2

1 for real y, we have

1 siny= 7

When sin y is negative, from (2), cosh x is also negative which is impossible. (: coshx>Oforallx)

__ ___

FUNCTIONS OF A COMPLEX VARIABLE

137

1 it ny=j=s 1 s n

n it + ( 1Y 1
=

where n is an integer, is the general value of y.

Case I. If n is even, y

it

Taking n
From(1),

2k,

cos y

cos (2klt +

cos

sinhx=-(i (1

11 (
it

z=xiy=log
CaselLlfnisodd, Taking n =2k
+

J+i

1, os y

cos [(2A +1) it

=cos From (1), sinhx=

=cos

1 sinh

i 7)

log

1
+

(i
=

log

z=xiy=log

f-i

(it

p
(K.U.K. 2005) 1

Example 3. Show that tank1 (cos 0) = coh (cosec 0). Sol. Let tanlr 1 (cos 0) = then cos 0 = tanh (p
(p

2 0 cos 1

tanh 2

( =

tanh (p 2 20 sin
= = =

sin 20
2 sech

( =

tanh (p 2 20 sin

sech(p=sin0

Taking reciprocals cosh (p

(p

cosec 0 cosh (cosec 0) 1 (cosec 0). cosh

tanlr (cos 0)

138

ICS A TEXTBOOK OF ENGINEERING MATHEMAT

EXERCISE 33
1.

Prove that

(i) (cosh x
(u) 2. 3. If y

sinh xYt
=

cosh

ix

sinh nx; a being a positive integer.

(1+tanhx 1 tanh x)

cosh 6A

sinh 6x. tx (tan t x). cot

+

log tan x, show that sinh ny

(a) If tan y = tan a tanh 13 and tan z (b) Prove that:

cot a tanh 13, prove that tan (y

z)

sinh 213 cosec 2a.

(i) z=sin (iii) 4.

5.
=

(ii)

=cos

tan
tan

(M.D. U. Dec. 2009)

sin 20 =tanhxtan v, provethat sin2 2 2 sec cii) x

Iftan 0=tanhxcoty and

If c cosh (0
+

cosh 2x + cos cosh2xcos2y

. 2 C

i)
+

iy, prove that

=

20 2 sech (i) x 6. 7. Iftan(x If sin (0

+ +

20 2 cosech y

cosec 4 22 y

A zy)=A+ tB, show that B

sin2x sinh2y

i)

p(cos a

i sin a), prove that (ii) tan a

=

(j) p 2
S. 9.

(cosh 20

cos 20)
+

tanh

p cot

0.

Ifsin (0

10.
11.

2 0 sin a. i sin a, prove that cos If cos (0 i) = cos a + i sin a, prove that (ii) cos 20 + cosh 20 2 0 = sin a (i) sin (Madras, 2003) If sin (0 + i) = tan a + i sec a, show that cos 20 cosh 20 = 3. a) = sin (0 2 prove that c (a) Ifcos (0 + iO) = R(cos a + i sin + sin (0 a)
+

i4)

cos a

2.

(KU.K 2005) (KU.K 2006)

(b) Ifu =logtan

12.

(--EJ
+

prove that 0=i logtan

(-+).

0=(n+)it+a. Iftan(0+i0)=tana+iseca,showthate2cot and2 (K. U.K Dec. 2009) If tan (x If tan (x

+ +

13. 14. 15.

iy) iy)

sin (u

2x = iv), prove that sinh 2y tanh v 2y. cosh (a + i13), prove that tanh a tan 13 = cosec 2x sinh

(S.V.T.U. 2006)

2CA IfCtan(xiy)=A+iB,provethattan2x= c B A 2 Prove that (1

+

16.

cosh x

sinh x)t = 2 cosh

(cosh

sinh

J.

FUNCTIONS OF A COMPLEX VARIABLE

139

17.

If cosh x

sec 0, prove that

(M.D.U. Dec., 2008)

2 2 =tan (i)tanh 18. If tan

=

(ii)x=logtan

tanh 1

prove that (ii)tan.i

inhu

(i)cosxcoshu (iii) u 19. 20. 21.

=

log tan

,y=2sinasinh13,provethatSeC(a+ i13+ sec(ai13)= Ifx=2cosacosh 13 x +y If sin [log (A srn Separate into real and imaginary parts. 2 (x (ii) sin (i) e (r+iy) 22 Iftanx+iy)9+ iP,proVett8
Prove that tan If x (I
+ +

-.

iB)]

iy, show that

2 1, where A

2 B

e u 2 .

COS

iy)
9

(iii) log cos Cx

iy).

22.
23. 24. 25.

cosh ycos .r cosir y sin x

,

u+iv 2
+

sinu+isinhv cosu+coshv

iy

cos Cu

iv), show that

. 2 +y =cosh v cos u) (ii)(1x) 2 2 + +x) ( cosh y v +cos u) 1 2 equation 2 13 are the roots of the 2 a and cosh Ii cos Cu + iv) = a + i13, prove that cos x2_(1+u2+v2 0 2 . )x+u Prove that tan (e)
=

26. 27. 28.

j!.

log tan

-J.

(M.D. U. Dec. 2008)

Find tanh x if 5 sinh x

cosh x

5.

2 x) and solve for tanh x.j 2 x by (1 tanh [Hint. Divide both sides by cosh x, square, replace sech If cos (x + iy) = a + i13, show that 2 13 = 1. 2 cosech 2 13 + y (ii) x 2 sech 2a= 1 2 cosec 2a y 2 sec (i) x

29.

Find all the roots of the equation:

2 (i)cosz= (tu) sin z
=

(ii)tanhz+2=O (iv) sinh z

cosh 4

30. 31. 32.

If I cos Cu

iv)

2u 1, show that sin

2 v. sinh

Iflogsin(Oi)=a+iP,Provethat

2 2 4e 20 + e (i) 2 cos 20 = e = Solve tan z e, where a is real.

(ii) C05 (9

13) = e cos CO 13).

(M.D. U. May 2009)

Answers

21.

(i) (ii)
-

CY

[cos (sinh x si4 y)

+

i sin (sinh x sin y,Ij

[(1 cos 2x cosh 2v)

i sin 2x sinh 2y]

140

A TEXTBOOK OF ENGINEERING MATHEMATICS

(iii) -log 27. (ii)

[.
+

[(cos 2x

cosh 2y)]

1 (tan x tanh i tan 29.(z)2nitilog(2+.)

log 3 1)
+

zn

.(

(iii)

flit +

( 1Y

(it

(iv) z (4n
32.

It

z=

-- log tan
+

[Hint. See Example 71

3.12. LIMIT OF A FUNCTION

(P.T.U. May 2006)

Let flz) be a single-valued function of z defined in a neighbourhood of z = z , then /z) is 0 said to have the limit 1 as z approaches z 0 (along any path, straight or curved) if given an arbitrary real number E> 0, however small, there exists a real number > 0 such that Z I f(z) 1 I <e whenever 0 < I z z 0 I and we write Jim f(z) = 1
z>zo

Remark 1. 6 usually depends on r. Remark 2. In real variables, x x 0 implies that x ap. proaches x 0 along the number line, either from left or from right. In complex variables, z 0 implies that z approaches z z 0 along any path, straight or curved. The limit must be independent of the manner in which z approaches z . If we get two different 0 limits as z --* z 0 along two different paths then limit does not exist.

3.13. THEOREMS ON LIMITS

1. If urn f(z) exists, then it is unique.
Z -4

2. If ftz) if and only if

u(x, y)

iv(x, y), where z

iy and z 0

, then lim f(z) 0 iy

Z 4 Z 0

0 u

0 iu

urn u(x, y) = u 0
0 X4X

and

Y-Yo

X4x 0 y_yo

urn

v(x, y)

. 0 v

3. If urn f(z) 4. If

1 and c is a constant, real or complex, then 11 and

urn
0 Z4Z

cf(z)

c lim f(z)
0 Z>Z

ci.

urn f(z)
0 Z4Z

urn z)
0 Z4Z

12, then
=

(i) urn [f(z) + g(z)1 = urn f(z) + urn g(z)

zO

11

(ii) lim [f(z) g(z)) = Jim f(z)

0 Z)2 0 ZZ

urn g(z)

11

FUNCTIONS OF A COMPLEX VARIABLE

141
.

(iii) urn [f(z)g(z)J = urn f(z) urn g(z)

zzo
0 Z)Z 0 Z4Z

1112 0

(iv) inn
-

g(z)

urn gtz)
Z 4 Z 0

= --

12

provided 12

urn f(z) =i providedl (v) limL=z0 0. 2 z - z, g(z) inn g(z) 12

Z -9

ILLUSTRATIVE EXAMPLES
does not exist. z Sol. If the limit exists, then it must be independent of the manner in which z approaches 0. Consider the path y 0 followed by x 0, we get
Example 1. Prove that urn
ziO

z-OZ

r. i,i =hm=1 hm=hmliim

.

II

x-O[y-Ox+iyj

x-40X

Now consider the path x

z-iOZ

0 followed by y

0, we get
.

4 y =O
Y
C II
)<

I
x

hm

lun I hm

r.

xiyl
I

y_OLx>Ox+iyJ
=

yIO

hm

iy
l)

1 As z 0 along two different paths, we get different limits. Hence the limit does not exist. Example 2. Show that urn
Sol. Let z
z-O

>

(RezImz) 2 does not exist. zO 2 lzI 0 along the path y = mx, then
=

y=O

bin

(RezImz) 2 2 1zI

(xy) 2 +y z-ox 2
inn
.

-----

2 (xmx) urn +rn x-ox x 2

V ___________

2 x (1rn) 2 (1m) urn )x 2 x-O(1+rn 2 1+m which depends on m. For different values of m, we have different paths and different limits. Hence the limit does not exist.
=
=

Example 3. Prove that

2 z*1+ _ 2z+2) iz
2
=

urn

4
,2

Sol.

urn

z1i 2z
+

-------

2 1+ijz

2)
=

lirn
Z-4

1+i

2 zl+i(z-_1+ j)

inn

z1i j) 2 _j 2 (z 1

z 9

z1i hm I z l+iL(z 1+ z)(z 1i) 1 urn (z1+i) 2

.

1+i

1 _1_ 2 4i (1+i1+i) 2

1 4

142

A TEXTBOOK OF ENGINEERING MATHEMATICS

EXERCISE 3.4
Show that the following limits do not exist: 1.
3 zOIzI

hm

2. jim
2 z,0z1 4

inn
z40

Re z
IZI

z,O

urn

2 Imz
2 Iz! (2z+3)(z1) z 2 2z+4
Z

Find the following limits:

5.
-

z-1i

inn (z 2

4z

7)

6. 8.
10. (P.T.U. May 2007)

z,2i

hm

7.
9. ii.

5y urn 3 ) (2x i
inn
z*1+i Z

iz 1 2 z1+
z,ii

urn

2 Z+i---i 2 2z+2

inn (z 2

2 z

+iz1 3 iz inn z -. (2z + 3i) (z 1) 2

Answers

:6:

:z!;i
10.4i

3.14.. CONTINUITY OF A FUNCTION

, then fz) is 0 Let flz) be a single-valued function of z defined in a neighbourhood of z = z small, there exists however 0, E> 0 if given an arbitrary real number said to continuous at z = z a real number 6> 0 such that 0 I <6 ) I <e whenever I zz 0 I fz)fz In other words, ftz) is continuous at z = if urn f(z) =
Z
--

20

Thus three conditions must be satisfied so that /tz) is continuous at z ) exists 0 0 i.e., flz 1. ftz) is defined at z 2. urn f(z) exists
Z -4

0 z

3. lim f(z)
Z,Zo

. flz ) 0

If any of the above conditions is not satisfied then y9z) is said to be discontinuous at
z
=

. 0 z

3.15. REMOVABLE DISCONTINUITY If f(z ) exists and urn f(z) 0

Z -420

) 0 1 exists but flz

1, then z

0 is called a point of remov z

able discontinzity.

FUNCTIONS OF A COMPLEX VARIABLE

143

) = 1, the function can be made continuous. If 0 0 such that 1(z By redefining 1(z) at z = z urn f(z) does not exist, discontinuity cannot be removed. Existence of limit is the necessary

0 z

condition for removable discontinuity.

3.16. CONTINUITY IN A REGION

A function 1(z) is said to be region.

continuous in a region

if it is citinuous at all points of the and examine the continuity

Remark 1. To examine the continuity of J(z) at z

we put z

off() at

0.
1(z)

Remark 2. If
1(z)
+

and

g(z)

are continuous at
) 0 where g(z

, then so also are the functions 0 z

g(z), 1(z)

g(z), 1(z) g(z) and

g(z)

0.

Remark 3. If 1(z) is continuous in a region, then the real and imaginary parts of 1(z) are also

continuous in the region. if 1(z) = u(x, y) + iv (x, y), then 1(z) is continuous if and only if u(x, y) and v(x, y) are separately continuous functions of x and y.

ILLUSTRATIVE EXAMPLES
Example 1. Show that the function f(z) defined by f(z)=
is not continuous at z =

Re(z ) 2 2 1z1
0,

zO

z=0

0.
= mx,

Sol. Given 1(0) = 0 0 along the path y Let z

then 2 x

z-0

urn

f(z) =

) 2 = lim urn Re(z 2 2 z-,0 z-,O Izi

+y 2

[...
=lim

y] (x )+2Lx =(x+iy) z = 2

2 1rn )x (1m 2 = =hm 2 1+m )x x-*0(1+m 2 which depends on m. For different values of m, we have different paths and different limits. Hence the limit does not exist and the function is not continuous at z = 0. ) so that the function 0 Example 2. Find the value of f(z

x 2 2 m x

+m x-ox x 2

f(z) Sol

2z+2 z 2
=

0 is continuous at z
=

z.

lim f(z) = urn

z41i

2 2 z 2 2 +2i z

z,1i z 2

lim

l zV + 2
2 (1-i)

144

A TEXTBOOK OF ENGINEERING MATHEMATICS

jim z,1i

i 2 (z1) 2 (1-i) Z 2

z,1i

lirn

(z1+i)(z1i) (z+ 1i)(z 1+i)

lim (z1i)

z1-iz+1j

urn

lim (z+1i) 2i

1iii 1i+1z Now flz) will be continuous at z 0 if

ftz ) 0 = urn

2(1i)

=x----=-

i 1i

1+i 1+i

1i 2

EXERCISE 3.5
Examine the continuity of the following functions:
z + 2 i

(i)fiz)=

z+ 0,

zi

2 3 z + zi iz

atz=z

(u)flz)=

..

zi

atz=i

zi

z=i

(ui)flz)=

2.

3.

z + 2 4 z2z atz=2z. z2i 2+3i, z=2i Show that the following functions are continuous for all z. (i) sin z . 2 (ii) e [Hint: Express fIz) as u(x, y) + iv(x, y) and show that u(x, y) and v(x, y) are continuous for all real values of x and y.j Show that the function

Im(z)
/z)=
IzI

z0

4.

0, z=0 is not continuous at z = 0. Find the value of flu) so that the function flz)
IZ

(M.D. U. Dec. 2005)

ai
zi
.

is continuous at z Answers

i.

1.

(i) Not continuous (ii) Continuous

(iii) Not continuous

4.

3i.

3.17. DIFFERENTIABILITY

Let w
Let

flz) be a single-valued function of the complex variable z(= x w+1w=f(z+iXz),then Ew = /z + tz) flz)

iy).

and

tw

f(z + Az)

1(z)

FUNCTIONS OF A COMPLEX VARIABLE

145
f(z+&)f(z) Az

AzO iSz

urn

liw

&O

urn

if it exists, is called the derivative of flz) and is denoted by f (z). (z+Af Thus lirn dz The function fiz) is said to be differentiable at z = z 0 if ) 0 f(z)f(z lim exists.
ZZr

0 ZZ

This limit is called the derivative of flz) at z f(z) f(z ) 0 Thus ) = lim 0 f(z

0 and is denoted by f (z z ). 0

z.,zo

ZZo

Equivalently, by putting z

0 z

Az, we have
Az

f(z)= hm

+zIz)f(z 0 f(z )

&-40

3.18. ANALYTIC FUNCTION (M.D.U. 2006, 2007, May 2008; P.T.U. May2007; U.P.T.U 2006)

If a single-valued function fl.z) possesses a unique derivative at every point of a region R, then flz) is called an analytic function or a regular function or a holomorphic function ofzinR. A point where the function ceases to be analytic is called a singular point.
NECESSARY AND SUFFICIENT CONDITIONS FOR f(z) TO BE ANALYTIC

The necessary and sufficient conditions for the function w = f(z) = u(x, y) + iv(x, y) to be analytic in a region R, are .uauvau (1) are continuous functions of x andy in the region I?. x Jy ax t3y aua au_ u
.
, -, ,

(Ii)

The conditions in (ii) are known as Cauchy-Riemaun equations or hiefly C-I? equations. (M.D.U. 2006, 2007, 2008; U.P.T.U. 2008) Proof. (a) Necessary Condition. Let w R, then
= =

ftz)

u(x, y)

iu(x, y) be analytic in a region

f (z)

exists uniquely at every point of that region.

Let ix and 6y be the increments in x and y respectively. Let au, & and z be the corre sponding increments in u, u and z respectively. Then, f(z-t-6z)f(z) (u+t5u)+i(v+&)(u+jv) f(z)= urn urn
6z,O
6z

6z.-,O

.3v urn i+zi &-OZ bzJ

(u

(1)

146

A TEXTBOOK OF ENGINEERING MATHEMATICS

must exist independ Since the function w flz) is analytic in the region R, the limit (1) and y > 0. 6 0, i.e., along whichever path 8x ent of the manner in which & = First, let & - 0 along a line parallel to x-axis so that y = 0 and y) [sincez=x+iy,z+&=(x8x)+i(j+6y)andz=&+iE .& u (& inn i+i-i=+i From(1), f (z)= &,O x !8x 8x) x Ox = 0 and & = i Oy. that so y-axis l to paralle line Now, let & > 0 along a u .&,l ia (8U inn i---+z---=--+ From(1), I (z)= 8y-4ok tOy) z Iy y i8y F .au

z ox.

av

From(2)and(3),wehave

+i--=----

au

.a

ax

ax

OX Y C-R equations must be Hence the necessary condition for f(z) to be analytic is that the satisfied.

Equating the real and imaginary parts,

u a . av ay au av ax ay

and

au av
=

(b) Sufficient Condition. Let /(z) = u tial derivatives i.e.,

, , ,

iv be a single-valued function possessing par

ax

ay ax ay
-

at each point of a region R and satisiying C-R equations. av

=

au

and

au av
=

of the region R. We shall show that f (z) is analytic, i.e., f (z) exists at every point omitting second and By Taylors theorem for functions of two variables, we have, on higher degree terms of and &y flz + &)= u(x+ Ox,y + Oy) iv(x + &c,y + 5y) E- dy)] y) = [U(x y)

Ox

= [u(x, y)

=f(z)+
or
(u

=I

(au .au (au .av (au .av a .au ax ax) ax ax;

+ +

iv(x, y)j

au

Ox + .av

(-b--

_+iJOX+L+i)

Sy

(au

+i

.av I Ox

+ i

I fry

[Using C-R equations]

(au

.av

(au

.dv

[..

j 2 ...J. j

FUNCTIONS OF A COMPLEX VARIABLE

147
.

(au
f(z +

(au

+i&y=&] 8 x

&) f(z)
r3z

au

at
f(z)= urn
z,O

ax
6z
exist.

ax

ax

Thusf (z) exists, because Hence ftz) is analytic.

a
,

av

ax at

Note 1. The real and imaginary parts of an analytic function are called conjugate functions. The Thus, if flz) = u(x, y) + iv (x, y) is an analytic function, then u(x, y) and v(x, y) are conjugate functions. equations. C-R given by is relation between two conjugate functions way Note 2. When a function ftz) is known to be analytic, it can be differentiated in the ordinary variable. as if z is a real 2 /z) = z f (z) = 2z Thus, = cos z etc. f(z) ftz) = sin z

3.20. CAUCHY-RIEMANN EQUATIONS IN POLAR COORDINATES (V. T. U. 2006; M.D. U. 2005; K U.K Dec. 200 U.P. T. U. 2008)

Let (r, 9) be the polar coordinates of the point whose cartesian coordinates are (x, y), then
= r cos 8, y = r sin 0, z=x iy=r(cos 0+ i sinO)= re ii + iv = f(z) = f(re) Differentiating (1) partially w.r.t. r, we have

au au

.av

_f(re).e
(u
.av

Differentiating (1) partially w.r.t. 0, we have

+i

.av
= =

ar ar Equating real and imaginary parts, we get

-=r

0 = ir f (re). ire av + ir
.

(-

+L

[Using (2))

au

av

ao
ar

or

au

ar lao and rae

and

au av =r

ae

av

ar

rae

ar lau

which is the polar form of C-R equations.

.

3.21. HARMONIC FUNCTIONS

(P.T.U. May 2007; U.P.T.U. 2007; M.D.U. May 2005)

+ i- = 0 is called a harmonic function. f ay ax
h a t 2 h a t 2

Any solution of the Laplaces equation,

Let fiz) equations.
=

iv be analytic in some region of the z-plane, then u and v satisfy C-R

148

A TEXTBOOK OF ENGINEERING MATHEMATICS

Jv

and Differentiating (1)

ay
u 2 a u 2 a

& and (2) w.r.t. y, we get

partially w.r.t. x

v 2
yix

and

2 y
Assuming

L xay

and adding (3) and (4), we get

Now, differentiating (1) partially w.r.t. y and (2) w.r.t. x, we get

...(6)
and

v 2 a
Assuming u 2 a
=

axy u 2 a v 2

2 x
and subtracting (7) from (6), we get

...(7)

v 2 a

Equations (5) and (8) show that the real and imaginary parts function satisfy the Laplaces equation. Hence

u and v

of an analytic

u and v

are known as harmonic functions.

322. ORTHOGONAL SYSTEM

Every analytic function f(z)

v(x, y)

= u + iv defines two families of curves u(x, y) = c 1 and , which form an orthogonal system. 2 c (U.P.T.U. 2009; M.D.U. May 2004)

Consider the two families of curves

and

u(x, y) = 2 v(x, y) = c
Differentiating (1) w.r.t. x, we get

u. u
+

audy
---.

=0

or

dy

1 (say) in

av
Similarly, from (2), we get

2 m
=

(say)

(x, y)

FUNCTIONS OF A COMPLEX VARIABLE

149

Since

/tz)

i.e.,

du dv m 1 m 2 = dx dx du dv dyy is analytic, u and v satisfy C-R equations dU dv du = dv and = dy dx ax dv dv

zom

2 1 m

dv dv

Thus the product of the slopes of the curves (1) and (2) is at right angles, i.e., they form an orthogonal system.

1. Hence the curves intersect

323. APPLICATION OF ANALYTIC FUNCTIONS TO FLOW PROBLEMS

Since the real and imaginary parts of an analytic function satisfy the Laplaces equa tion in two variables, these conjugate functions provide solutions to a number of field and flow problems. For example, consider the two dimensional irrotational motion of an incompressible fluid, in planes parallel to xy-plane. Let V be the velocity of a fluid particle, then it can be expressed as

V=vxI+vyj Since the motion is irrotational, there exists a scalar function (x, y), such that
V From (1) and (2), we have
=

V(x, y)
=

dx

dy

and v = dx The scalar function (x, y), which gives the velocity components, is called the velocity potential fiuiction or simply the velocity potential. Also the fluid being incompressible, div V = 0 (d d z+jI.(vi+vj)=0 dx dy) dv dv dx dy Substituting the values of v and v,, from (3) in (4), we get

or ---=0 dxkdx) dyy) 2 2 dy dx Thus the function is harmonic and can be treated as real part of an analytic function w = fiz) = p(x, y) + i i (x, ) For interpretation of conjugate function N (x, y), the slope at any point of the curve N (x, y) = c is given by

150

A TEXTBOOK OF ENGINEERING MATHEMATICS

dy

[By C-R equations]

ayax
V
=
-

[By (3)]

2 of the fluid particle is along the tangent This shows that the resultant velocity v2 + v curves are known to the curve , (x, y) = c i.e., the fluid particles move along this curve. Such nted by 4) (x, y) represe curves The n. functio stream the as stream lines and w (x, y) is called = c are called equipotential lines. the Since 4) (x, y) and i,(x, y) are conjugate functions of analytic function w = ftz), . onally orthog other ct each interse equipotential lines 4) (x, y) = c and the stream lines (x, y) = c,

Now,

dz
=

ax

ap i.a

ax

[By C-R equations]

[By(3)] dw
=

=v,iv,, The magnitude of resultant velocity

12
= i.JVx

the complex The function w = flz) which fully represents the flow pattern is called potential. arid (x, y) In the study of electrostatics and gravitational fields, the curves 4) (x, y) c of force respectively. In heat flow problems, the = c are called equipotential lines and lines respectively. curves 4) (x, y) = c and iji (x, y) = c are known as isothermals and heat flow lines

ILLUSTRATIVE EXAMPLES
f(z)
=

nates as Example 1. Find p such that the function f(z) expressed in polar coordi 2 sin p0 is analytic. 2 cos 20 + ir r 2 sinp0 2 cos20, v = r Sol. Let fiz) = u + iv, then u = r

ar au

2r cos 20,

ar

2r sinpO
=

2 sin 20, 2r

av

2 cos p0 pr

lau au lay and rae ar ar rae 2rcos2O=prcosp8 and 2rsinpO = 2r sin 20 Both these equations are satisfied if p = 2. ives of u and u must be Note. For a function fz) to be analytic, the first order partial derivat continuous in addition to C-R equations.

For f(z) to be analytic,

=---

FUNCTIONS OF A COMPLEX VARIABLE

151

Example 2. Show that the function f(z) = lT is not analytic at the origin, even though Cauchy-Riemann equations are satisfied, there at. (U.P.T.U. 2005) fiz)= u(x,y)+ iv(x,y)= JiJ then At the origin (0, 0), we have au u(x,0)u(0,0) 00 =hm Sol. Let
.
.

u(x,y)=

v(x,y)=0,

ax

x,O

x,O

a,

lim o
.

u(0, y)

u(0, 0)

lim o
.

v(x,0)v(0,0) av 00 =lim =limO -40 ax x-,O x X v(0,y)v(0,0) 00 =lim yO ayy-O y y

.

ax ay ay ax Hence C-R equations are satisfied at the origin. Now If z

-4

Clearly,

au av au

av

urn f (0) = z-,O

0 along the line y
=

f(0)

z mx, we get

zO

lim

/ii
x+iy

/1m21 X f(O)= urn =lirn----- x-*Ox(1-i-Lm) xO1+im Now this limit is not unique since it depends on m. Therefore, Hence the function 1(z) is not regular at the origin. s Example 3. Prove that the function f(z) defined by
flz)=

f (0) does not exist.

(1+i) 3 x ( 1iy ) ,zOandf(O)=O 2 2 x +y is continuous and the Cauchy-Riemann equations are satisfied at the origin, yet f (0) does not exist. (M.D.U. Dec. 2009) Sol. Here Let 1(z)
=

)+i(x +y 3 3 y (x ) 3 2 2 x +y

3 +y x 3 y 3 3 3 V x 3 + x then u 2 2 2 2 2 x+y x+y 2 x+y +y x 2 Since z0 x0,y0 u and v are rational functions of x and y with non-zero denominators. Thus, u, v and hence 1(z) are continuous functions when z 0. To test them for continuity at z = 0, on changing u, v to polar co-ordinates by putting x = r cos 0, y = r sin 0, we get 3 0 sin 3 0) and v = r (cos = r(cos 1 u 3 0) 3 0 + sin WhenziO,r 0 link u lim r (cos 3 0) = 0 3 0 sin
.

f(z)=u+iv

152

A TEXTBOOK OF ENGINEERING MATHEMATICS

Similarly,

urn v
zO

lim fiz)=0=/tO)

f(z) is continuous at z = 0. Hence fiz) is continuous for all values of z. At the origin (0, 0), we have x au u(x, 0) u(0, 0) =hm =lim
=
.

X
=

xO

x,O

=1
=

a)? av =lim
.

urn yo

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

X

lim yo
.

y xO
X

Jx

x0

x-*O
.

=lim

a), yO auav

yO v(O,y)v(0,O) =1 =hm y-4O y 3 av au and

Hence C-R equations are satisfied at the origin. Now, Let z

f(O)

zO

urn
=

f(z) f(0)
=

3 (x
z,O

)0 3 3 +y y + i(x ) 3 y (x+iy) + (x ) 2

0 along the line y

x, then
3 0+2ix

Also, let z

>

ii(1i)1-t-i 2 2 1+1 1+j) x-*02x ( 3 0 along the x-axis (i.e., y 0), then
f(O)=lim
=

=1+i x Since the limits (1) and (2) are different, f (0) does not exist. Example 4. Prove that the function sinh z is analytic and find its derivative. (P.T.U. 2007) fiz) u + iv sinh z sinh (x + iy) = sinh x cos y + i cosh x sin y Sol. Here u=sinhxcosyandv=coshxsiny f(0)=lirn
x*O
= =

3 x+ix

au
= =

cosh x cos y, sinh x sin y, and

au
=

sinh x sm y

cosh x cos y

auav
ox cry

du

au
ax

ay

Thus C-R equations are satisfied.

FUNCTIONS OF A COMPLEX VARIABLE

153
au au
,

Since smh x, cosh x, sin y and cos y are continuous functions, continuous functions satisfying C-R equations. Hence flz) is analytic every where. Now,
f(z)_--+i
=

axyar

a and &

are also

a .av

cosh x cos y + i sinh x sin y = cosh (x + iy) = cosh z. Example 5. Determine the analytic function whose real part is e(x cos 2y y sin 2y).

Sol. Let

fiz)

(Bombay, 2005; M.D.U. 2005, 2007) iv be the analytic function, where u = e(x cos 2y y sin 2y)

= =

2e(x cos 2 y y sin 2y)+ e cos 2y

e (2x cos 2y

2y sin 2y

cos 2y)

au

=e(2xsin2ysm2y2ycos y 2 )
=

sin 2y + 2y cos 2y) Since ftz) is analytic, u and v must satisf5r C-R equations au av a av and
e (2x sin 2y
+

axay
=

at

Now

2y sin 2y = e(2x cos 2y at Integrating w.r.t. y, treating x as constant, we get

ay

cos 2y)

eat [2x.

sin2y

{2Y

cOS2Y)_( ) 2 _ sin 2Y sin2y]

+ +

(x)

= e Cx sin 2y +y cos 2y) function of x. arbitrary where 0(x) is an

(x)

= =

2e (x sin 2y e (2x sin 2y

+ +

y cos 2y) sin 2y

+ +

(sin 2y) 2 e
+

(x)

2y cos 2y)
+

4(x)
[From (2)]

But

at

2 (2x sin 2y e

sin 2y

2y cos 2y)

(x) = c, an arbitrary constant. From(3), v=e(xsin2y+ycos2y)+c ftz) = u + iv = e (x cos 2y y sin 2y) + ie (x sin 2y = e [(x + iy) cos 2y + i(x + iy) sin 2y] + ic = (x + iy) e (cos 2y + i sin 2y) + ic + ic ze + ic. 2 = ze eY + ic = ze

y cos 2y) ic

154

A TEXTBOOK OF ENGINEERING MATHEMATICS

(Milne-Thomsons Method) This method determines the analytic function f(z) when u or v is given. Smce

z=x+y,z=xiy
-

sothatx=,y

+:

z 2i

Let

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

.

(z+. z (z+ z 1+ wI , I, 2i 2i} 2 2 Considering this as an identity in the two independent variables z,
= U

and putting

z,

weget

m iv(z, 0) which is the same as (1) if we replace x by z and y by, 0. Thus to express any function in terms of z, replace x by z and y by 0. Now ftz)=u+iv
ttz)
= u(z,

0)

f(z)

a + i.av au
e (2x cos 2y

.au
-

[C-R equations]
+

2y sin 2y

cos 2y)
+ +

ie(2x sin2y On replacing x by z andy by 0 on R.H.S., we get f (z) = e(2z + 1)

Integrating w.r.t. z, we have flz)
=

sin 2y+2y cos 2y)

(2z

1) !

2. !-_

ic

ze

ic,

taking the constant of integration as imaginary since u does not contain any constant. ) x 2y. 2 2 +y Example 6. Determine the analytic function w u + iv, if v = log (x y x y Sol.Here +x2 v=log( + ) 2 av ax av
Since 2x
2

2+1

2y

ayx2+y2

w=u+iv dw
=

au + i.au av + i.av

[C-R equations]

2x 2y ) 21-i-iI 2 2 2+1 2 x+y x+y

.1

Replacing x by z and y by 0, we get

dw hw

(2
+ j

Integrating w.r.t. z, we have

(i

2)z

lJ 2i log z

(i

2)

2i
+

+ c.

FUNCTIONS OF A COMPLEX VARIABLE

155

xY sin 2 2 y is harmonic. Find the conjugate (x ) Example 7. Show that the function u = e function v and express u + iv as an analytic function of z. ) 2 it = eY sin (x 2y Sol. Here

s(x y sin(x eco yeY 2x =2 ) 2 )

x y y os(x n( 4 yeYc si 4x = e ) 2 - )
3x

) 2 2 y 2eY cos (x 2 y ) 4xy e- cos (x 2

2 eY 4x

y 2 sin(x )

..(1)

n(x os(x Ysi y eYc xex _2y =_2 ) _ 2 ) u 2 a

(x y x (x cos y in ye 4 s +4x = e ) 2 )
2exY
+

) 2 (x y cos 2 Adding (1) and (2), we get

2 ) 4y 2 4xy e cos (x 2y

sin (x 2

) 2 y

...(2)

+ ,2
2 x

u 2 a

u 2 3

0 which proves that u

+

it

is harmomc.

Now, let then.

fiz) f(z)

iv i
=

i
) 2 sin 2 (x y + i [2x + 2x

[C-R equations] 2 e cos (x sin (x 2 9)

] y ) 2
+

[ 2y

eY

2y

cos (x 2 2 2z eiZ

] y ) 2

Replacing x by z and y by 0, we get 2z cos z 2 Integrating w.r.t. z, we have

f(z)
=

2iz sin z 2

2 2z (cos z

i sin z ) 2

fiz) which expresses Since

i e 2
+ ic
=

+ ic

u + iv as an analytic function of u + iv
= = =

z.
+

ie 2

ij(j2
+

ic
.

iei

+2ixy) 2 y

ic

ie

+ ic

ie [cos 2 )] + ic 2 ) + i sin (x 2 2 y (x y )] + 2 2y ) + i [ e cos (x 2 2y = e sin (x 2

ic

. y ecos(x c v_ 2 ) the xy-plane is given by the potential function in Example 8. An electrostatic field 2 y 9, find the stream function. 3x Sol. Let i(x, y) be a stream function. ByC-Requations
=_=_3x2+3y2

156

A TEXTBOOK OF ENGINEERING MATHEMATICS

Integrating (1) w.r.t. x, treating y as constant, we get 14! =.x 3 + 32 + F(y) sothat =6xy+F(y)
= =

From (2) and (3), 6xy + F(y) 6xy or F(y) 0 Hence +3xy 3 141=x + 2 c. &xamp1e 9. If u v (x y)(x 2 + 4xy + y ) and f(z) 2 x + iy, find f(z) in terms of z. Sol. We have u v (x y)(x 2 + 4xy +y ) 2 + =x y 2 3 3xy -y 3x au 3v 2 + 6xy 3y 3x 2

F(y)
=

iv is an analytic function of

and or

ax au

av

2 3x

6xy

2 3y

6xy3y =3x 2 Subtracting (2) from (1) 2=12xy Adding (1) and (2) 6y or 2=6x 2 Thus

av ax

au ax

au

or

=6xy

au av

av

3x =3y 2
+

I (z)

6x

2 i(3y

) 2 3x

Replacing x by z and y by 0, we get f(z) 3iz 2 Integrating flz) 3 + C. iz

=

Example 10. If u
z
=

e iy, find f(z) in terms of z.

=

2sin2x and f(z) 2 +e 2cos2x iv

iv is an analytic function of

Sol. Given
=

flz)

iflz)=iuv (1 i) flz) (u v) i(u v)

Adding (1) and (2), we get

+
=

or
where

F(z)=U+iV
F(z)=(1+i)flz),U=uv

and V=u+v
(given)

2sin2x V=u+v= e+e -2 2cos2x

FUNCTIONS OF A COMPLEX VARIABLE

157

sin2x 2 cosh 2y 2 cos 2x cosh 2y cos 2x Now, we proceed to find F(z) whose imaginary part is given. cos 2x)(2 cos 2x) sin 2x (2 sin 2x) = (cosh 2y
-

2sin2x

ax

2 (cosh2yco s2x) 2cos2xcosh2 2 2 2 X+8 X y_2(C08 m ) (cosh2yc0s2X) 2 2 cos 2x cosh 2y (cosh 2y cos

aV ay

2sin2xsinh2y 2 (cosh2yco s2x)

Since /tz) is analytic, so is F(z). F(z)=+i ax ax

=
-

au .av

av +.av
(cosh 2y

by C-R equations
+

2sin2xsinh2y

cos 2x) 2

2(cos2x cosh2y1) 2 (cosh 2y cos 2x)

By Mime-Thomsons method, replacing x by z and y by 0 on R.H.S., we get 2i 2i 2(cos 2z 1) = F(z) = i 1cos2z 2sin 2 z (1cos2z)

2z i cosec

Integrating w.r.t.z, we have F(z) = i cot z + c (1 + i)f(z)= icotz +c

ftz)
=

cot z + 1+i 1+i c i(1i) cotz+ 1+t 1i

fz)

1
=

(1

+ i)

cot z

C, where C
=

*xamp1e 11. Find the analytic function f(z)

iv, given v

(r

sin 0, r

0.

Sol. Using C-R equations in polar coordinates, we have au av ( r 1 COS 0 r = I = I ar ae r)

158

A TEXTBOOK OF ENGINEERING MAThEMATICS

lau rJO
Now(1)gives Integrating w.r.t. r

J 2 r i =I1lcos9 ar rJ

v ar

1 ( 1+i sinO

au(

r+

r)

cos

e + (9),

where 0(9) is an arbitrary function of 8.

au Also, from (2)

i. I r+-sm6+(9)
+

sin

r+- sin0+0(e)=_(r+?) sine

=

0 or (9)

u
Hence

= =

I r + fl
\

flz)

rj iv

cos 0

=(r+-)cose+c+ir_2J sinG. Example 12. If f(z) is an analytic function with constant modulus, show that f(z) is constant. (M.D.U. May 2009; Bombay 2005 S; P.T.U. 2005; U.P.T.U. 2008) Sol. Let fiz) = u + iv be an analytic function. Since I flz) I = constant = c (say), (c 0) I flz) 2 = u 2+v 2=c 2 Differentiating (1) partially w.r.t. x and y respectively, we have au av au av 2u 2v =0 or u +v =0 ax ax ax a,. au av au av 2u +2v =0 or u +v =0 ay
.

and

ay

ay

ay

Using Cauchy-Riemann equations

=

equation (3) becomes

au av au ax ay. and ay av au = 0 + v u ax ax

av

ax

Squaring and adding (2) and (4), we have

v II II 2 (u ) vax)

2 iaui

avi

axj

=0

FUNCTIONS OF A COMPLEX VARIABLE

2 t r iavi itau

159

or

[x)

II +1I
2 (au

vax)
iav
i

=0

or or

i +i =0 x) Jxj I f(z) 12

[: c0]

=0

r. [Since f

(z)

=a

.av

f (z) = 0

ftz)

constant.

Example 13. If f(z) is a regular function of z, prove that

I f(z)
12_4

. 2 I f(z) I

(J.N.T.U. 2006; Kottayam 2005; U.P.T.U. 2005, 2007) Sol.Let/tz)=u+ivsothat or I

fz) lflz)

=Ju2+v2

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

=2u+2v

a4

=21u+II +v+I
2 ax 2 ax ax) 2 ax ax

p 2 a

a a

av

ax

u 2 a

2 (au

v 2 a

2 (av

Similarly, Adding, we get

v 2 a u 2 a (au? (p 2 a =2 u+I----I +v+I

2 ay kay) 2 ay Lay

2 (aS

2 a +1 =2 uI+I+II +1I +vI+I+II ,2)

(a2u 2 Lax a2u 2 au 2 (au (a2v a2v 2 ax 2 ay ) 2 ay Lax)

(av?

2 (au1 ay

L)

2 Lax

Lax)

Since flz) equation.

= u + iv is a regular function of z, u and v satisfy C-R equations and Laplaces

au av au av

and

u 2 a

u 2 a

v 2 a

v 2 a

From (1), we get

2 a =210+ii +1I +0+1I +1 Lax Lax) 2 ay 2 ax [ Lax) Lax)

E
2 (au 2 iav (av?

p 2 a

(au?

1 2 iav r iaui =411I +1 vax) ax

,-

160

A TEXTBOOK OF ENGINEERING MATHEMATICS

Now

ftz)=u+iu

f (z)
From (2), we get
2 a
-

U
=

&)

and

f (z)
2

(au 12

(vN 2
+

2 a

3x

4 I f(z)

or

) 2 ay

[a 2 2 +

2 d
ay

ax

/tz)

12 =

4 I f(z) I

Example 14. If f(z) is a holomorphic function of z, show that

2

I--If(z)I +{lf(z)I}

lax

= If(z)

. 2 I

(M.D.U. May 2007; K.U.K. Dec. 2009; P.T.U. 2005; V.T.U. 2006; U.P.T.U. 2009)

Sol.L,et
then

f(z)=uiv
I flz) I ax
=

2 (u
2

v2)v2 =

4(x, y)

(say)

(u+v( 2u

au

+2uI

av

1
[u2+v2
ap

1 a av Iu+v I
.

ax)

Similarly

1 au av Iu+vI
iFIy ay)

av auj lu+v ax

ifu2+v2

ax

[Using C-R equations]

2 a

[]
a

2
=

i.

2 u

+ v 2

[
[
+1

U+VI

au

ax

ax)
-

au2l I avi +Iu+v a

ax

ax) J

2
+(U2+V2)(l

U 2 +U
/

vax)
2

=1_I
vax)

axJ 2

-If(z)I

lax

+{If(z)I1
a

2y

(av (au =1i +I)

vax)

Also

flz)=u+iv f(z)=
+1 ax

au

.av

ax

FUNCTIONS OF A COMPLEX VARIABLE

161
()2

I f(z) 12 From (1) and (2), we have

2 (au)

I 1(z)

2 i} +

I f(z)I} :21
+

12

Example 15. If z

iy, show that

:.

c:
a, ,

,ax

=4

2 a
azaz

SoLWehavez=x+iy

=xiy

axl axl az2a2az Now

ayi ayi
2
2 .a

a
a

a
a

ax ax

ay

i(a

and

1(a

.a

2 a
Hence

i(a

a i(a

1(a2

2 a

;i

2 a 2 a 2 (a i+i4-. azaz ia ,2) 2

Note. Remember the result of Example 15. Example 16. If f(z) is an analytic function of z, prove that
2 (a 2 a

log If(z) 1=0.

+

(a2

Sol. We know that Also

ax

ay }1=4 azaz
log I f(z) 12 log [f(z)
.

2 a

2 a

log I f(z) I

f()] =
2

[logf(z) logf()] [logf(z)

+

2 a

ax

J
ay

2 a

log I f(z) I

2 a

azaz

logf()]

az[f()

162

A TEXTBOOK OF ENGINEERING MATHEMATICS

EXERCISE 3.6
1. (a) Determine a, b, c, d so that the function f(z) (b) Determine p such that the function ftz) 2. (a) Show that z) (b) Show that flz)
= =
-

2 (x

+ axy +

) 2 by

2 + + i(cx

dxy

) is analytic. 2 y

xy +

1 be an analytic function. ) + i tan2 log (x 2 +y 2 y iy is everywhere continuous but not analytic.

2 is not analytic anywhere in the complex plane. (J.N. T. U. 2003; M.D. U. Dec. 2006)

(c) If w 3.

log z, find

and determine where w is non-analytic. u ar 1& v , raO ar

(J.N.T. U. 2005)

Show that the polar form of Cauchy-Riemann equation are Deduce that
+ + --

lau

rPiO

2 ar

r ar

2 r

0. f(z) f(0)
,

(K U.K Dec. 2005, 2009) 0 as z 0 along any radius vector but

4.

If/z)

, *

z *0, flO) = 0, prove that

not as z 5.

0 in any manner.
=

(a) Show that the function fz) defined by fiz)

,z o, flO) = 0, is not analytic at the x +y origin even though it satisfies Cauchy-Riemann equations at the origin. {xY2(x+iY)(x2 )4)

x x ( 3 Y 2

(b)Showthatfiz)=

is not analytic at z

0, although C

R equations are satisfied at the origin.

: ::

(c) Examine the nature of the function x+iy) x ( 5 y 2 f(z)= y x + 4 0

,

z=0

in a region including the origin. 6. Determine which of the following functions are analytic: (i) e (w) 7.

2
=

(ii) sin z

(iii) cosh or
2 -

(a) Show that f (z)

log(x 22 +y ) +itan . (vi) x IzI x+y 12 is continuous everywhere but not analytic at any point. (P. T. U. 2005)

()

x--iy 2 2

(b) Show that the function flz)

z I z I is not analytic anywhere.

FUNCTIONS OF A COMPLEX VARIABLE

163

8.

Show that u

iv

xzy+a

where a

0, is not an analytic fti.nction of z

iy, whereas u

iv

is such a function. 9. Determine the analytic function whose real part is 3 (i) x

2 3xy
/2

2 3x

2 3y

(M.D. U., Dec. 2008) (K U.K Dec. 2009) (iv) cos x cosh y (vi) sin2x cosh2ycos2x (J.N.T. U. 2005)

(ii) log

2 y

(iii) ex Cx cos y

y sin y)

(v) e_x (x sin y y cos y) (U.P.T.U. 2006, 2008) (vii) x sin x cosh y

y cos x sinh y

(viii)

ex [(x 2

) cos y 2 y

2xy sin y]. (M.D. U. May 2005)

(V. T. U. 2006) 10. Find the regular function whose imaginary part is (P. T. U. 2005)
+

(ii) cos x cosh y (M.D. U. Dec. 2006)

(iii) sinh x cos y

(iv) e (x sin y 11.

y cos y)

(v) e_x (x cos y

y sin y).
15

(U.P.T.U. 20O9)

Find the real part of the regular function whose imaginary part

2sinxsiny cos 2x + cosh 2y (Bombay 2006)

12.

13.

2 2xy 2x + 3y is harmonic. Find a function v such that f(z) = u + iv is y (a) Prove that u = analytic. Also express fiz) in terms of z. ) + x 2y is harmonic. Find its conjugate harmonic 2 2 +y (b) Show that the function v (x, y) = ln (x (M.D. U. May 2009) function u(x, y) and the corresponding analytic function /z). 2 y , find the stream 2 An electrostatic field in the xy-plane is given by the potential function 4) = x function.

14.

If w

4) + w represents the complex potential for an electric field and

+

15. 16. 17.

determine the function 4). 2 If the potential function is log (x

2 2 x+y (Bhopal, 2002 S; M.D. U. 2002)

i! =

2 x

2 y

, find the flux function and the complex potential function. y ) 2

In a two dimensional fluid flow, the stream function is i = tan If flz) (i) u
=

(),

find the velocity potential 4).

iv is an analytic function, find 1(z) if

v= e (cos y

sin y)

(M.D. U. 2006, 2007, May 2008)

164 x
x

A TEXTBOOK OF ENGINEERING MATHEMATICS

(u)u+v=

,whenj(1)=1 2

+y

(zzi)uv=

cosx+sinxe ,whenfii=O. 2(cosxcoshy) 2}

18.

If /(z) is an analytic function of z, prove that

(i)

I Rj9z)

2_2

f(z) 12

(ii) V 2 I f(z) 12 = 41 f(z) 12 19.

(Madras 2006)

20. 21.

Express 0 + ii as a function of z. Find the analytic function fiz) = u(r, 0) Prove that cojugates.
it

Prove that i = log [(x 1)2 + (y 2)21 is harmonic in every region which does not include the point (1, 2). Find a function such that 1 + iis is an analytic function of the complex variable z = x + iy.
+ iv(r, 2

0) such that

v(r,

0) =

2 r

cos 20 r cos 0 2.

=x 2

2 and y

x+y

are harmonic functions of (x, y) but are not harmonic

Answers
1.

(a)a=2,b=1,c=1,d=2 (c) 3 (i)z

(iv)

(b)p=1

2.
9.

at z = 0
+ 3z 2 + + ic z +

6. (i),
(ii) logz+

(iii),
ic

(iv)

except when z = 0,

(vi)

except when ic

=0

1+ ic ic

(iii)zez + (vi)

cos z sin
z

(v) i(ze + c) (viii)

..

cot

z + ic

(vii) z

ez + ic. 2 z
.

10.

0)

1+i

+ c

(ii) i

cos z

(iii)

z smh z

+ c

(zv)_ze_z+c

(v)lze_2+c

11.

2cosxcoshy +c. cos 2x + cosh 2y

12.

_ 2 (a)v=x + 2xy_2y_3x+c,f(z)(1+i)z2_(2+3j)z+jc y

(b)u(x,y)=2xy2tan+c;f(z)(i2)z+2i1ogz+c 13. 16. 19.

i

= 2xy + c
)+c 2 +y

14.

2xy +

2 +

c
+

15. 2 tan 1)

+ c,

2 log z

+ ic

- log(x 2
2 tan 1

17. (i)ezc
2i log

(ii) 1

(z___.J,

._(_
2)

(iii) -.(i_ cotjj

(z

1 2i)

20.

re + e 2 i(r

+ C.

3.24. COMPLEX INTEGRATION

Let flz) be a continuous function of the complex variable z = x + iy defined at all points of a curve C having end points A and B. Divide the curve C into n parts at the points
A=0 (z P ) (z P ) ,1
.,

P.(z)

P,(z,) = B.

FUNCTIONS OF A COMPLEX VARIABLE

165

Let

1 z

1 PL. Then the limit of the sum 1 and E be any point on the arc P_ z_

f() 6z
0, if it exists, is called as n > 00 and each z the line integral of f(z) along the curve C. It is denoted by

Y
1 Pi+

f(z) dz

(M.D. U. May 2005)

-1

0 and P, coincide so that In case the points P C is a closed curve, then this integral is called contour integral and is denoted by
If fz)
=

f(z) dz. 0 X

u(x, y)

iu(x, y), then since

dz=dxidy,wehave f(z) dz
=

(u
=

iv)(dx + idy)

(udx

vdy)

(vdx + udy)

which shows that the evaluation of the line integral of a complex function can be reduced to the evaluation of two line integrals of real functions. Moreover, the value of the integral depends on the path ofintegration unless the integrand is analytic. When the same path of integration is used in each integral, then
.0

f(z)dz=-J f(z)dz
a

If c is a point on the arc joining a and b, then

Lfdz=S f 0 dzSf1z
ILLUSTRATIVE EXAMPLES
Example 1. Evaluate ) dz. 2 (x y ix L the straight line from z 0 to z 1

1+i

= + i = (a) along (b) along the real axis from z = 0 to z = 1 and then along a line parallel to imaginary axis rom z = 1 to z = 1 + i. (c) along the imaginary axis from z = 0 to z = i and then along a line parallel to real axis rom z = i to z = 1 + i. Sot. (a) Along the straight line OP joining O(z = 0) and P(z = 1 + i), y = x, dy = dx and x aries from 0 to 1.

166

A TEXTBOOK OF ENGINEERING MATHEMATICS

1+i

(xy + ix2)dz=$

1+i

(x y

)(dx 2 ix

=jo(x_x+ix2)(dx+idx)=$:ix2(1+i)dx

B
(0,1)

P(1,1)

(-1 + i)

[1
=

/
1 0 A (1, 0) X

=(_1+i)()=_+i.

(b) Along the path OAP, where A is z

j
...
I

1+i

)dz 2 (xy+ix
=

=SOA(x_y+ix2+1x_y+jx21z
dx and x varies from 0 to 1.
r
(x+ix )dx=i+z----i =+i 2 X

...(

Now along OA, y

0, dz

(xy+ix )dz= I Jo JOA

=

1 3 .x

-Il

[2

1 2

1. 3

Also, along AP, x

1, dx

idy and y varies from 0 to 1

f,(x_Y+ix2)dx=(1_Y+i)idY=[c_1+i)Y_ij]

=1+ii=1+

Hence from (1),

I(x_y+ix2)dz Jo

1+i=-+i. =I--i( 2 6 3) .. 2)
=

Cc) Along the path OBP, where B is z

1+i

(x_y+ix2)dz=J(x-y+ix2)dz+f(x_y+jx2)cZz
=

Now along OB, x

0, dx

idy and y varies from 0 to 1

] 2 IOB(Y)1:CY)Y[
Also, along BP, y
=

=F

1, dx

dx and x varies from 0 to 1

391 r 2 11. Ix .xI .2 (x1+ix )dz= I )dx=IxI =+z Jo [2 2 3

JBP

(xy+zx

Hence from(2),

J:(x_)

) dz 2 ix

Note. The values of the integral are different along the three different paths.

Example 2. Evaluate I (x 2 Jo
(a)y=x

iy)dz along the paths

. 2 (b)y=x

FUNCTIONS OF A COMPLEX VARIABLE

167

Sol. (a) Along the line y

dy=dxsothatdz_dx+idx=(1+i)dx

iy)dz

2 f (x

ix)(l+ i)dx

=(1+i)
(1 + i)(2 3i) 5 1 6 6 6 , dy = 2x dx so that 2 (b) Along the parabola y = x dz=dx2ixdx=(1 +2ix)dx and x varies from 0 to 1.

P(l, 1)

J:2

iy)dz =$x2

)(1 2 ix

2ix) dx

(1i) 1x2(1+2ix)dx (i 3 1 .\ 2) (1i)(2+3i) 6 5 6 1. 6

is

dz, along Jo ( (a) the real axis to 2 and then vertically to 2 + i. (b) along the line 2y = x. ()2 2 = (x iy) ) 2ixy 2 Sol. = (x 2 y where A is (2, 0) and P (a) Along the path OAP, (2, 1). Example 3. Evaluate

p2+i

P (2, 1)

dz 2 ()
$OA

2 2ixy) dz y
=

+$(x2

2 y

2ixy) dz
0

Now, along OA, y 0 to 2

0, dz

...(1) dx and x varies from

M(2,O)

f
Also,

(x2_y2_2ixy)dz=fx2dx
JO

OA

along

AP, x

2, dz

idy and y varies from 0 to 1

1

= 0 =[j
3 1

2 rxsl

JAP
=

(x2y2_2ixy)dz=f(4_y2_4iy)idy

1 11 [4iY_i2Y2] =4ii+2=2+i 3 3 11 8 14 11 (z) d 2 z=+2+i=+i. 3 3 33

Hencefrom(1),wehave

JO

2+i

168 (b)AlongthelineOP, andy varies from 0 to 1.

A TEXTBOOK OF ENGINEERING MATHEMATICS

2y=x,
2+1

dx=2dy

sothat dz=2dy+idy=(2+i)dy

211

()2J

1 (x2_y2_2ixy)dz=J(4y2_y2_4iy2)(2+i)dy

(2

4i)

y2 1 f

dy

(:10

Example 4. Integrate 1(z) = x 2 + ixy from A(1, 1) to B(2, 4) along the curve x Sol. Equations of the path of integration are x = t, y = t 2 dx=dt, dy_2tdt t= 1 and atB(2, 4), t=2 AtA(1, 1),
iAz)

[-]
5i)

t, y

. 2 t

dz

f(x2 +

ixy)(dx + idy)=

:it 2 (t X dt f3

+2it dt)
2

1 Si (8 64 (1 2 .( 3 151 45. = I II 1+1112- I=+1. L 3 5) 3 5) 4) 4 5 Example 5. Prove that

=1

(t2_2t4)dt+if
1

3tdt=E.---.l r.i [3 [4J

=21c1
()

dz

0 In is an integer

1], where C is the circle I z a I

r.

Sol. The equation of the circle C is IzaI=r or za=re 0 where 0 varies from 0 to 2?t as z describes C once in the anti-clockwise direction. Also dz = ire do.
-

0 dO . c dz 2 ire r (z)q =1 =11 Jc za .30 0 re

(ii)

(z

dO=27u

aY dz

0 dO rr etU . ire 1 ir
ehl+1) 2 f

dO

re 8 12 4 ir71
I[i(fl+ .
1)J

- [.. n

1]

n+1 0.

12 1] [e

[.

cos 2(n

1)m

i sin 2(n

1)it

iO

1]

FUNCTIONS OF A COMPLEX VARIABLE

169
+

Example 6. Evaluate

2 (z

3z + 2) dz where C is thearc of the cycloid x = a(O sin 0),

V
A(G=t)
/4
4
-

cos ) between the points (0, 0) and (ia, 2a). Sol. The function f(z) = z 2 + 3z + 2 is a polynomial and therefore analytic in z-plane. Hence the line inte gral of fiz) between the points 0(0, 0) and A(ta, 2a) is independent of the path joining these points. Let us choose the path of integration as: (i) From 0(0, 0) to B(ita, 0) along the real axis; followed by (ii) From B(ica, 0) to A(ita, 2a) vertically
=

a(1

0 (0

0)

Now along OB,

Ly f(z)dz=Jf(z)dz+$f(z)dz
=

z=x+ iy=x, dz =dx

+ 3z +2 =x + 3x+ 2 2
no

and x varies from 0 to ira.

IOBSO
Also, along BA, x
=

1 2 3x 2 C 3x2)dx=[-.-++2xj x
=

a 3 ic

3
ira

a 2 3ir

+2ira=(2it 2 a 2 +9ira+12) 6

z=x+iy=ita+iy, dz = idy andy varies from 0 to 2a. fiz 4 + 2 -3z+2 3(lta )= +iy) -(ita+i z +2 y)

BA

f(z)dz=J
=

2 +3(rr.a+iy)+2jidy [Otaiy)
+ 3(ira+

3 {{(ira +iy)

2 iy)

2Y}
+

i:
2i
.

-(ica

3 i2a)

ira

2 i2a)

2a

-ica

(ir

3 2i)

(ir +

2 2i)

4ia

a 3 --it

a 2 it

..

From (1), we have

(z +3z+2)dz=(2ir2a2+9ma+ 12)+ (n+2i)+

2 (it+2i)

4ia

a 3 it

170
133 =it a
322 +lt a

A TEXTBOOK OF ENGINEERING MATHEMATICS

33 .3 +(it+2a) 2 +4ian a +2ita+Ot+2z) 3 2 3

a3

oa

o 022 ---it a

oL a3 .3 (it+2i) +(it+2) +2a(it+2i)

EXERCISE 3.7
1.

Evaluate

Jo

2 dz, along z
=
=

(a) the line y

(b) the real axis to 3 and then vertically to 3

. 2 3y +

(c) the parabola x 2. Evaluate

2+i

(2x + iy

1) dz, along

(a) the straight line joining (1

3.

i) to (2

i). (b) the curve x

=

t +

1, y

2t

1. (M.D. U. Dec. 2006)

Evaluate Evaluate

I Jo

.4+2i

dz, along the curve given by z

it.

4.

lz 12 dz, around the square with vertices at (0, 0), (1, 0), (1, 1) and (0, 1). M.D.U. 2002)

5.

Show that points z

=

0, z

(z
=

1) dz
=

0, where C is the boundary of the square whose vertices are at the i andz
=

1,z

i.

(M.D.U. Dec. 2005, May2008)

6. 7.

Evaluate

$(y_x_3x2idz,wherecisthestraight1inefromz=otoz=1+ i.

(a) Evaluate

this integral if C is the lower half of above circle? (b) Evaluate (z

(z

) dz, where C is the upper half of the circle I 2 z

1. What is the value of

z) dz where C is the upper half of the circle I z

2 I

3. What is the value (M.D. U. Dec. 2009)

of the integral if C is the lower half of the above given circle? 8. Prove that dz =

or xi according as C is the semi-circular arc I z I

1 from

1 to 1 above

9.

or below the real axis. Show that for every path between the limits, i ) z= (2+z d 2

(M.D. U. May 2005)

.12

10.

Evaluate

2+31

2 (z

z) dz along the line joining the points (1,

1) and (2, 3).

171
r, iai
11

iW

COMPLEX VARIABLE
z dz, where C is the contour

Evaluate

(a) straight line from z = i to z from z = i to z = 1. (b) left half of the unit circle I z I = 1 ed in the clockwise sense. (c) circle given by I z + 1 I = 1 describ

Answers
1.

(a).6+i Ca)4+81
1i

(b)6+ (b)4+i
2 2 7. Ca) j,

Cc)6+ 3.i0i (b) 30, Cc) 30 4.i i

10.

2.

6. 11.

( 103 64i)

Ca)

(b) 21

NS 3.25. SIMPLY AND MULTIPLY CONNECTED REGIO

(M.D.U. May 2006)

cross itself (Fig. 1). A curve which A curve is called simple closed curve if it does not crosses itself is called a multiple curve (Fig. 2).

Fig.1

Fig. 2

Fig. 3

region encloses points of A region is called simply connected if every closed curve in the ted indefinitely without pass the region only, i.e., every closed curve lying in it can be contrac a multiply connected region. In ing out of it. A region which is not simply connected is called Fig. 3 shows a multiply plain terms, a simply connected region is one which has no holes. . (There can be more than 2 C connected region R enclosed between two separate curves C and into a simply connected one, two separate curves). We can convert a multiply connected region by giving it one or more cuts (e.g., along the dotted line AB).

3.26. CAUCHYS INTEGRAL THEOREM

(U.P.T.U 2006)

point within Statement. If f(z) is an analytic function and f (z) is continuous at each and an a simple closed curve C, then

172

A TEXTBOOK OF ENGINEERING MAThEMATICS

Proof. Let R be the region bounded by the curve C. Let flz) = u (x, y) + iv(x, y), then

f(z)dz=(u+iv)(dx+idy)
=

C
+

.Jc

(udx

udy)

i (vdx
JC

udy)

Sincef(z) is continuous, the partial derivatives

a av

are also continuous in R. Hence by Greens Theorem, we have aU aU f(z)dz --dxdy+i -dxdy ...(2) ay) JJR y) Now fiz) being analytic at each point of the region R, by Cauchy-Riemann equations, we have

Q
f
A

HR L

and

ax

Thus, the two double integrals in (2) vanish. f(z)dz = 0. Hence Cor. 1.11 f(z) is analytic in a region R and P and Q are two points in R, then Q f(z)dz is JP independent of the path joining P and Q and lying entirely inR: Let PAQ and PBQ be any two paths joining P and Q. By Cauchys theorem,

I f(z)dz+ JQBP I f(z)dz =0 JPAQ I f(z)d.z_$PBQ f(z)dz =0 JPAQ

f(z)dz f(z)dz. $ $ Iff(z) is analytic in the region bounded by two sim
PAQ
=

PAQBP

f(z)dzo R

Hence

PBQ

Cor. 2. ple closed curves C and C , then 1

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

Let AB be a cross-cut joining the curves C and C , then the 1 doubly connected region becomes simply connected. By Cauchys theorem,
.APQA

APQABDEBA

f(z) dz

f(z)dz+f
JAB

f(z)dz+f
JBDEB

f(z)dz+f f(z)dz=0
JBA

f(z)dz
f(z) &

+f

f(z)dz
-

DEI(z)dz_f f(z)dz =0
E P

Jc ci (integrals around a closed curve are taken positive when the curve is traversed in counter clockwise direction)

f(z) dz

FUNCTIONS OF A COMPLEX VARIABLE

173

Hence

jf(z)d2=$f(z)dz

The theorem can be extended. If a closed curve C contains non-intersecting closed curves ducing cross-cuts, it can be shown that

, 1 C

C,

C,,, then by intro M1).U. May 2006)

Jf(z)dz=$f(z)dz+jf(z)dz+...-Jfzd.-.

3.27. CAUCHYS INTEGRAL FORMULA

(V I. U. 2007; U.P. T. U. 2006. 2007, 2008) Statement. If f(z) is analytic within and on a closed curve C and a is any point within C, then f /(a)= Proof. Consider the function 1 1(z) dz. 2rci .c z a

which is analytic at

every point within C except at z = a. 1 1 with a as centre and radius p such that C Draw a circle C lies entirely inside C. Thus is analytic in the region between C and

c.
...(1)

By Cauchys theorem, we have

f(zdf

X
z

Now, th equation of circle C is or I z a I =p dz = ipedO so that

9 pe

f() _f2n f(a +-J

Cza

pe dz
=

ipeLO do =i $ f(a 2
0

pe)d9

Hence by (1), we have

f(z)

if

2it

f(a

pe)dO

1 shrinks to the point a, i.e., In the limiting form, as the circle C

,.

0, then from (2),

()
----

,.2,t

dz

f(a)dO

if(a) J

p2Tc

dO = 2tif(a)

Hence

a)

dz

which is the required Cauchys integral formula.

Cor. By Cauchys integral formula, we have f(z) 1 /(a) = JC 2iu z a may be treated as a parameter. where a is any point within C and
---.

...(1)

174

A TEXTBOOK OF ENGINEERING MATHEMATICS

Differentiating both sides of (1) w.r.t. a,

f (a) = 2iu c aa [z a]
-

ic1

f(z)
=

2iti

(z

a) 2

Similarly,

I
=

f(a) and in general, f(a)

2iti (za) 3 3? f(z) dz Jc 27ti (z a) 4

n! f(z) dz. 2 (z aY Thus, if a function ofa complex variable has a first derivative in a simply connected region, all its higher derivatives.exist in that region. This property is not exhibited by the functions of real variables.
=

ILLUSTRATIVE EXAMPLES
Example 1. Evaluate 2 + 2ixy)dz, where C is the contour I z I = .1. (x y 2 Sol. fiz) = x 2 y 2 + 2ixy = (x + iy) 2=z 2 is analytic everywhere within and on I z I By Cauchys integral theorem, f(z)dz = 0.

1.

Example 2. Evaluate

2 (3z

4z +1) dz, where C

is the arc of the cycloid x = a(9 sin 8), y = a(1 cos 0) between (0, 0) and (2ica, 0). Sol. Here, fiz) = 3z 2 + 4z + 1 is analytic everywhere so that the integral is independent of the path of integra tion and depends only on the end points z 1 = 0 + iO and

1c(3z2+41)bo
r
=

2,ta

(3z + 2 4z+1)
+

O( )

A(z)

3 z
L

2 +z 2z Jo

a+ 3 8,t

a + 2na = 2ita (4n 2 Sit a + 47ta + 2

1).

Example 3. Evaluate (a)lzI=2 Sol. fiz)

=

-dz, where C is the circle z+1 (b)IzI=j-.

=

e is an analytic function. 2 (a) The pJint a = 1 lies inside the circle I z I By Cauchys integral formula,

2.

1 f(z) dz 2itz Jc z ( 1)

fi

1)

r
=

e dz z+1

2iue.

FUNCTIONS OF A COMPLEX VARIABLE

175
=

(b) The point a

1 lies outside the circle I z I

The function

is analytic within and on C. s-z+ 1

=

By Cauchys integral theorem, we have C z+1

Example 4. Evaluate i
I
COS1tZ

0. 3.

dz, where C is the circle I z I (z1)(z2) Sol. The integrand has singularities where (z1)(z2)=0 i.e., atz=1 and z=2. Both these points lie within the circle I z I = 3 2 is an analytic function. ftz) = cos itz 1 1 1 Also z1 z2 (z1)(z2)
JC

2 2 r cositz r cositz 2 cosicz c dz dz+cp dz=p JC z2 JC z1 JC(zl)(z2) = 2t i/U) + 2ti /2), where /(z) = cos = 2iti cos it + 2ici cos 4ic = 2ici + 2mi = 4iti.

Example 5. Evaluate

3z Z + 2 dz, where C is the circle I z c z 2 1

1 I

1.

(M.D.U. Dec. 2006) z 1 = 0 i.e., at z = 1 and z = 1. Sol. The integrand has singularities, where 2 The circle I z 1 I = 1 has centre at z = 1 and radius 1 and includes the point z = 1. 2 + z in an analytic function. fiz) = 3z 1 i( 1 1 1 = Also 1 (z1)(z+1) 2z1 z+1 2 z

z 3z + 2 Jc z 1 2

z 3z + 1 2 2Jc z1

z 3z 1 2 2Jc z+1

By Cauchys integral formula, 2 +Z 3z dz = 2iti/tl), where/tz) C z1 = 8iti By Cauchys integral theorem, +z 2 r 3z

JC

2 3z

z+1 From (1), we have +z 2 3z dz=4iu. p JC z 1 2

dz=0

176

A TEXTBOOK OF ENGINEERING MATHEMATICS

Example 6. Use Cauchys integral formula to evaluate

2z

where C is the circle I z I 2. (M.D.U. May 2004) (z+1Y Sol. The integrand has a singularity at z = 1 which lies within the circle I z I 2. f(z) n! Now f= (za) Herea=1,n+1=4 i.e., n=3,flz)=e e 3 f(z) = 2 e, f(a) = f( 1) = 2 z 2 e z 2 e 3? r 3 = _4 ------dz 2 From(1),wehavef(1)=-----4 dz => 8e itt JC (z + 2iu Jc (z + 1)
C

f dz,

2z

Hence C (z+1) Example 7 If f(V

dz 3z

2 3e

jc

+ 7z

+1

2 +2 y = 4, find the values of d-, where C is the circle x

f(3), f(l i) andf(l i). So!. The given circle C,is x 2 +y 2 = 4 or I z I = 2. Thepointz=3liesoutsidethecircle Iz I =2whilez=1iandz=1+ilieinsidethe circle.
Now, fl3)
..

+7z+1 2 3z f 3z+7z+1 dz is analytic within and on C. and C z3 z3 By Cauchys integral theorem, we have

---

I 3z+7z+1

-;: = 0

fi3)

Now let 4(z) = 3z 2 + 7z + 1 which is analytic everywhere. By Cauchys integral formula, we have
=

dZ ii 1

where

is a point within C.

Also

3z + 2 7z+1 fl) C z 2 + 7 + 1) fl) = 2iti(3 f(l i) = 2ici [6(1 i) + 71 = 2ni(13 f() = l2iti f(1i)=12iti.
2iti4)=

f(E) = 2iti(6 +
=

7)

6i)

2it(6

13i)

EXERCISE 3.8
1. Verify Cauchys theorem for the integral of z 3 taken over the boundary of the (i) rectangle with vertices 1, 1, 1 + i, 1 + i,

(ii) triangle with vertices (1, 2), (1, 4), (3, 2). 2. Evaluate

j.C (zaY t

2, 3, 4

where C is a closed curve containing the point z

a.

FUNCTIONS OF A COMPLEX VARIABLE

177

3.

Evaluate

z+ 2 dz, where C is the circle z3 (b)IzI=1

.-

(a)IzI=4 4. Evaluate

c z2

dz, where C is the circle. (b)IzI=1.

(a)IzI=3 5. Evaluate Jc

3Z + 2 7Z+1whereCisthe&cle z+1 (b)IziI=.1

(a)IzI=1.5 (c) I z I 6.
=

7. 8.

I sin3z dz, where C is the circle I z = 5. Evaluate It Jc z+ 2 +Z + 1 4x + 9y 2 dz, where C is the ellipse 2 Evaluate Cz 3z+2 Evaluate (a)2i,2i dz around a rectangle with vertices

1. (M.D. U. Dec. 2005, May 2008)

(b)i,2i,2+i,i. ciz, where C is I z I

=

9.

(i) Evaluate

1.

(M.D.U. May 2006, Dec. 2006)

,.

(ii) Evaluate 4,

sinGz
Iz. i, 6 3
it

Jc(

dz, where C is the circle

1.

(M.D. U. May 2005)

(iii) Evaluate 10. Evaluate (a) I z 11. 12. 13. 14. 15. 16. Evaluate
=

4 JC

(z
Z

2)

dz, 3)
,

where

C: I z I

c (z
3.5

1)(z

where C is the circle (b) I z I

=

3/2.

z+4 dz, where C is the circle I z I = 1. Cz 2 +2z+5 2 + cos dz, where C is the circle z I Evaluate Jc (z1)(z2) eX dz, where C is the circle I z I = 2. Evaluate 4 lXz 4) Jc (z 1 -1 e dz, where C is the circle I z I = Evaluate 4 Jcz(z+1)

3.

(U.P.T.U. 2008)

Evaluate
C Z7t

dr, where C is the circle I z

1 I

3.

2 and LetP(z)=a+bz+cz where C is the circle I z I

=

( P(z)

f P(z) f P(z) dzJ_dzj_r.dz =2iti

1. Evaluate P(z).

178

A TEXTBOOK OF ENGINEERING MATHEMATICS

z

17.
18.

Evaluate

4) Jc

dz, where C is the circle I z

z e

1.
2. 3

(a)

Evaluate

c (z +
C

dz , where C is the circle I z I 1) eZ 1)(z 5

2) 2

(b) Evaluate 19. If fl)

dz, where C is the circle I z I

(z

+
=

dz, where C is the ellipse

f +
4

1, find IU),

flu), f( 1) and f( i).

(J.N. T. U. 2005; M.D. U., Dec. 2007)

20.

Evaluate, using Cauchys integral formula:

Jc
Ui)

(z

e 2z 1)(z
Z

2)

dz,whereCisthecircle I z I =3.
=

J $
1

Cz

3z+2

dz, where C is I z 2 I
3.

(U.P.T.U. 2009; M.D.U. 2007) (M.D.U. May 2006) (U.P.T.U. 2008)

eZ dz ,where C is I z 11 (z + 1)

(iv) (v)

22 JC(z 2 +n)

clz,whereCis I z I =4.
2.

/2z+1 dz, where C is I z I (zz) 2

.

logz dz,whereCis I z1 (vi)f 3 C(z1)

C
p

2
=

f tanz d, where Cis the circle I z I (vu) I Z

1.5.

(viii)

4) JC

e 3z dz, where C is the square with vertices at (z1n2) Answers

1,

i.

(M.D. U. Dec. 2009)

2. 5. 8.
10. 13. 16. 18. 20.

0 (a)6d (b)0 (c)0 (a) 0 (b)

3. (a) 28ici (b) 0 6.2iti

9. (i) 11.0 14. 2ti
iti

2 (b) 0 4. (a) 2ti e 7.0 (ii)

12.4iti 15.

7ti

7ti

(iii) 12

,ti

(a)2ti (b)iti

ie

2ni

2 1+z+z 2 (a) 4iae 4 (i) 2ni(e

(iv)

17.2ti (b) 2wie ) 2 e (ii) (v)

4iti

19. 20ici, 2n(i

1),

14d, lfiiti

(iii) (vi)

l0iti

7ti

(vii) 0

(viii) 72iu.

FUNCTIONS OF A COMPLEX VARIABLE

179

CONFORMAL MAPPING*
(For KU.K Only) 3.28. TRANSFORMATION OR MAPPING We know that the real function y = fix) can be represented graphically by a curve in the xy-plane. Also, the real function z = fix, y) can be represented by a surface in three dimensional space. However, this method or graphical representation fails in the case of complex functions because a complex function w = flz) i.e., U + iv = fix + iy) involves four real variables, two independent variables x, y and two dependent variables u, v. Thus a four dimensional region is required to represent it graphically in the cartesian fashion. As it is not possible, we choose, two complex planes and call them z-plane and w-plane. In the z-plane, we plot the point z = x u + iv. Thus the function w = flz) + iy and in the w-plane, we plot the corresponding point w = defines a correspondence between points of these two planes. If the point z describes some curve C in the z-plane, the point w will move along a corresponding curve C in the w-plane, since to each (x, y) there corresponds a point (u, v). The function w = f(z) thus defines a map ping or transformation of the z-plane into the w-plane.

v=1

II

cJ
D

y=0 Z-plane

v=1 W-plane

For example, consider the transformation w = z + (1 i). Let us determine the region D of the w-plane corresponding to the rectangular region D in the z-plane bounded by x = 0, y = 0, x = 1 andy = 2. Since w = z + (1 i), we have u+iv=(x+iy)+(1i)=(x+1)+i(y1) u=x+1 and v=y1 Thus Hence the lines x = 0, y =0, x = 1 and y = 2 in the z-plane are mapped onto the lines u = 1, u = 1, u = 2 and u = 1 in the w-plane. The regions D and D are shown shaded in the figure.

*fl part is not included in the syllabus of MD.U., Rohtak.

180

A TEXTBOOK OF ENGINEERING IATHEMATICS

3.29. CONFORMAL TRANSFORMATION OR GEOMETRICAL REPRESENTATION OF w = f(z) Suppose two curves C ,C 1 2 in the z-plane intersect at the point P and the corresponding curves C , C 1 in the w-plane intersect at P under the transformation w = f(z). If the angle of 2 intersection of the curves at P is the same as the angle of intersection of the curves at F, both in magnitude and sense, then the transformation is said to be conformal at P.

Z-plane

W-plane

Definition. A transformation which preserves angles both in magnitude and sense be tween every pair of curves through a point is said to be conformal at the point. The conditions under which the transformation w = /tz) is conformal are given by the following theorem. 3.30. THEOREM If f(z) is analytic and f (z) 0 in a region R of the z-plane, then the mapping w = f(z) is conformal at all points of R. Proof. Let P(z) be a point in the region R of the z-plane and P(w) the corresponding in point the region R of the w-plane. Suppose P moves on a curve C and P moves on the corresponding curve C. Let Q(z + 6z) be a neighbouring point on C andQ (w + 3w) the corre sponthng point on C so that PQ
Y
=

bz and PQ =6w.
V
-

P(z)

P (w) Z-plane

W-plane

Then bz is a complex number whose modulus r is the length PQ and amplitude 0 is the angle which PQ makes with the x-axis. & = re

FUNCTIONS OF A COMPLEX VARIABLE

181

Similarly,

6w

9 re z

where r is the modulus and

e is the amplitude of 6w.

Let the tangent to C at P make an angle a with x-axis and the tangent to C at F make 0, 0 * a and 6 a. an angle a with u-axis, then as 6z

dz
Since f(z) From (1), Thus and 0, letf(z)
=

Lt
8zO

6z
=

Lt

&Or}

Irei(eo
=

pe, then p
(r

I f(z) I and

amplitude of f(z).

pei
p

Lt_Je
Lt
z,O

i(UO)

90)=aa =J ( 0

the corresponding curve 1 1 be another curve through P in the z-plane and C Now let C angle P f3 with x-axis and the an at makes to 1 C through P in the w-plane. If the tangent then as in (3), at P makes an angle fY with u-axis, 1 tangent to C

4=IY13
From(3)and4), aa=1313
Y

or

13cx=13a=y
V

Thus angle between the curves before and after the mapping is preserved in magnitude and sense. Hence the mapping by the analytic function w = fz) is conformal at each point where f(z) 0. Note 1 A point at which f(z) = 0 is called a critical point of the transformation. Note 2. From (2) p
zO r

4
0

P(z) R-plane X 0

P(w) w-plarie U

Lt

If follows that under the conformal transformation w = f(z), the lengths of arcs through P are magnified in the ratio p: 1, where p = f(z) I. Thus an infinitesimal length in the z-plane is magnified by the factor I f (z) I in the w-plane and consequently infinitesimal areas in the z-plane are magnified by the factor I f(z) i in the co-plane. Note 3. From (3), cc = a + shows that the tangent to the curve C at P is rotated through an angle 0 under the given transformation. Note 4. A harmonic function remains harmonic under a conformal transformation.

182
3.31.. COEFFICIENT OF MAGNIFICATION

A TEXTBOOK OF ENGINEERING MATHEMATICS

Coefficient of magnification for the conformal transformation w given by I f (a + i3) I where dash represents derivative.
3.32. ANGLE OF ROTATION

fiz) at z

i is

Angle of rotation for the conformal transformation w

fiz) at z

is given by amp.

[f(a + iJ3)].
Example 1. For the conformal transformation w (i) the coefficient of magnification at z (ii) The angle of rotation at z Sol. Here
= = =

, show that 2 z

i is

i is

f(z)=w=z 2 f(z)=2zandf(1+i)=2+2i.
=

Thecoefficientofmagnificationatz=1iis I f(l+i) I The angle of rotation at z

=

J4+4=2T2.

i is amp. [f(l

i)1

1 L=!. tan-

3.33. SOME STANDARD TRANSFORMATIONS c, where c is a complex constant z=x+iy, c=a+ibandw=u+iv Let then the transformation becomes u + it = (x + iy) + (a + ib) = (x + a) + i(y + b) u=x+aandu=y+b sothat Thus the transformation is a mere translation of the axes and preserves the shape and size. For example, the rectangle OMPN in z-plane is transformed to rectangle OMPN in the w-plane under the transformation w = z + (1 + 2i).
1. Translation: w
=

N(l

3i)

P(3 + 3i)

O(l N(O+i)
(2+1) 1

+ 21)

M(3 + 2i)

2
z-plane

(2+Oi)

X
w-plane

FUNCTIONS OF A COMPLEX VARIABLE

183

2. Rotation and Magnification: w = cz, where c is complex constant. and w = Pe 0 z = re, Let c = peia = p re 4 8 + X) then the transformation becomes Re R=prand=8a Thus the transformation maps a point (r, 8) in the z-plane into a point P (pr, 8 + a) in the w-plane. Hence the transformation consists of magnification of the radius vector of P by p = I c and its rotation through an angle a = amp (c). Thus under this transformation figure in w-plane is similar to the figure in z-plane (magnified by I c I) but rotated through an angle cx. Note 1. If a> 0 then rotation is anticlock wise and if a < 0 then rotation is clockwise. Note 2. In w = cz; if C is real then a = 0, then this transformation is only that of magnification (no rotation) in this case the two figures in z-plane and w-plane are similarly situated about origins but figure in w-plane is ci times figure in z plane. Such mapping is respective their called Magnification. For example the transformation w = (1 + i) z maps the square OMPN bounded by x = 0, y = 0, x = 1, y = 1 in z plane to the square OMPN zv-plane.
...

V
Y P(O + 2i)

N(0+i)

P(1

+i)

1)

o
Here
U
=

(1+01)

z-plane
u+iu=(1i)(x+iy)=(xy)+i(x+y)
X

y, V = X + y v=u i.e., v =y maps into u = y, x= 0 = u v i.e., = x v into u = x, maps y= 0 u v= 2 i.e., v = 1 +y maps into u = ly, x= 1 vu=2 i.e., v=x+1 u=x1, mapsinto y=l Square in z plane is mapped into square in the w plane bounded by u = v, u u+v= 2, vu =2 Verification: Here i
=

..

cI

Each side of the square in w-plane is I c I times (i.e., plane also amp ofc
=

..J) the

side of the square is z

tarr

and the sides of the square are rotated through an angle

184

A TEXTBOOK OF ENGINEERING MATHEMATICS

This transformation is rotation as well as magnification. Consider another example in which c is real:

Example: The transformation w = 2z maps the triangular region OAB bound by the lines x = 0, y = 0, x y = 1 into a similar triangle OAB in w- plane.

NA z-plane w-plane

u+iv=2(x+iy) Here w=2z v=2y u=2x u=O x=O mapsinto v=O y=O mapsinto
.. ..

xy=1mapsinto+=1

i.e.,

u+v=2

Varifiction: The two figures are similar but figure in w-plane is 2 times the figure is z-plane. This transformation is only Magnification

3. Inversion: w
Let

re and w 1 then the transformation becomes Re4 = and z

=

4 Re so that R

P(z) 1

Th: under the transformation W: in z-plane is mapped into the point P

a point P(r, 8) 8).

( o)
x
U

Consider the w-plane superposed on the z-plane. If P (1 1 is 8j, then is (r, 8) and P
,

(w)

1 is reverse of P w.r.t. the unit circle with i.e., OP OP 1 = 1 so that P OP = 1 = centre 0. [The inverse of a point P w.r.t. a circle having centre 0 and radius k is defined as the 1 2 point Q on OP such that OP. OQ = k
-

FUNCTiONS OF A COMPLEX VARIABLE

185 Thus the transformation w

1 in the real axis represents w The reflection P of P is an inversion of z w.r.t. the unit circle I into the real axis.
-

1 followed by reflection of the inverse

maps the interior of the unit circle I z I = 1 Obviously, the transformation w = into the exterior of the unit circle I w I = 1 and the exterior of I z I = 1 into the interior of I w I = 1. However, the origin z = 0 is mapped to the point w = 00, called the point at infinity. Note. This transformation w 1
=

maps a circle in z-plane to a circle in w-plane or to a straight

line if the circle in z-plane passes through the origin. The general equation of any circle in the z-plane is x + 2 2gx+2j5+e=O y

Let w=u+w=,thenz= w z so that

X
=

or x+zy= and
Y

1 U+w

U+v

Uil) 2 2

U 2 2 u+v

v 2 2 U+v

Substituting the values of x andy in (1), we get

u 2 ) 2 2 +v (u
+

v 2 +v (u ) 2 +v u 2

2gu 2 +v u 2

2fu +c=0 2 2 +v u

or or or

+v (u 2 ) + 1 +v u 2

2gu 2fb +c=0 +v 2 +v u u 2 0 2fb 2gu +v 2 u u +u

...(2) 2fv + 1 = 0 ) + 2gu 2 c(u + v 2 If c 0, the circle (1) does not pass through the origin and equation (2) represents a circle in the w-plane. If c = 0, the circle (1) passes through the origin and equation (2) reduces to 2gu 2fr + 1 = 0 which is a straight line in the w-plane. Regarding a straight line as a circle of infinite radius, we can say that the transformation 1 maps circles into circles. w=

(P. T. U., May 2005) az+b w= . .( 1) A transformation of the form cz + d where a, b, c, d are complex constants and ad be 0 is called a bilinear or Mobius transfor mation.

4. Bilinear Transformation:

186

A TEXTBOOK OF ENGINEERING MATHEMATICS

The transformation given by (1) is conformal, since dw ad-bc

=

(cz+d) 2 dw+b The inverse mapping of (1) is z = cw a which is also a bilinear transformation. The transformation (1) can be written as cwz + wd az b = 0 which is linear both in w and z and hence the name bilinear transformation

dz

From (1), we observe that each point in the z-plane except the point z

maps into

a unique point in the w-plane. Similarly, from (2), we observe that each point in the w-plane except the point w
=

maps into a unique point in the z-plane. Considering the two exceptional

points as points at infinity in respective planes, we can say that there is one to one correspondence between all points in the two plane. Every bilinear transformation w transformations (i) translation: w (iii) inversion: w
= =

az + b ad cz + d

bc

0 is the combination of basic

(ii) rotation and magnification: w

cz

By actual division, we have

bcad
C 2

1.
C

a d 1 bcad Taking w 1=z+ 2= w + 3= w , we get w = 2 w c c 1 w Thus, by these transformations, we successively pass from z-plane to w -plane, from 1 -plane to w 1 w -plane to w-plane. 3 -plane, from w 2 -plane to w 2 -plane and finally from w 3 Since each of these auxiliary transformations maps circles into circles, hence a bilinear transformation also maps circles into circles. Note 1. Cross Section: If four complex numbers z , z 1 , z 2 , z 3 , are taken in order 4
,
,

then

2 (z z 1 is called the cross-section of z . 4 ,z 1 ,z 2 ,z 3 2 (z ) 3 4 (z z z ) 1

Note 2. The cross-ratio is invariant under a bilinear transformation. , then 4 ,z 2 3 and z Thus if w ,w 1 ,w 2 3 and w 4 are the respective images of four distinct points z, z

1 (w ) 3 2 (w w w ) 4 (w w 1 ) (w 4 2 w ) 3

4 1 (z ) 3 2 ) (z z ) (z z 4 ) 3 1 z (z

a+b be 0 dividing the numerator and cz + d denominator of the right hand side by one of the four constants, we observe that there are only three
Note 3. In the bilinear transformation w
=

FUNCTIONS OF A COMPLEX VARIABLE

187

transformation.

ongtntg. Hence three independent conditions are required to determine a bilinear

ILLUSTRATIVE EXAMPLES

z-plane bounded by the lines x = 0, y = 0, x = 1, y = 2 is mapped under the transformation w=z+(2i)? Sol. The given transformation is i.e.,u--iv=x+iy+2i w=z+(2i) v=y1 u=x+2 u=2 mapsinto x=O mapsinto v=1 y=O mapsinto u=3 x= 1 mapsinto v=1 y=2 So the mapped region PQRS is also a rectangle bounded by u = 2, v = 1, u = 3, v = 1 (shown in the figure below)

Example 1. What is the region of w-plane into which the rectangular region is the

V C
U

SX=lR =2
sr_rs.r.rJ

x=O
Oy=OA

x1

0 pv=1 Q w-plane
Ut

z-plane

and determine the region in w-plane Example 2. Consider the transformation w = e x = 0, y = 0 and x + y = 1 in the lines the by corresponding to the triangular region bounded z-plane. Sol. The given transformation is

ut
= cos-+isin (x+iy) we z 4 i.e., u
+

iv

XL (1

i)(x+ iy)= v=-=(x+y)

[(xy)+ i(x+y)j

u=J=(x_y); x=Omapsinto y=Omapsinto

u=_j=(_y),v= -=y;
u=jx,v=--x

..

v=u
v=u

:.

x+y=1mapsintou=.1

..

188

A TEXTBOOK OF ENGINEERING MATHEMATICS

z-plane
7t

w-plane

The transformation w

4 z transforms the triangle OAB in z-plane into a triangle e

.

0KB in w-plane rotated through an angle

Example 3. Under the transformation w

(i) I z 2i = 2. (ii) y x + 1 = 0. Sol. The given transformation is

1 w=
Z

find the image of the following curves (Kerala 2005) (P.T.U., May 2007)

1 or z= w

or x+iy=
=

1 u+zv

uiv 2 2 u-i-v

so that (i) The given curve is

u u +v 2

I z 2i I = 2 or I x + i(y 2) I = 2 or +(y2) x = 2 2 or + x 2 4y=0 y which is a circle in the z-plane with centre (0, 2) and radius 2. Substituting the valus of x and y from (1) in (2), we get 2 Cu or or 1
+ 4v = +

and y

v +v 2

) 2 v

2 -i-v (u ) 2 1

4v

2 u

2 v

=0

or

+v 2 u 4v 22+2 2+v ) (u u +

=0

4v 0

u2+v2+u2+v2

(ii)yx+ V 2 2 u+v

0, a straight line which is the required image of the given curve. 1=Omaps into
U 2 2 u+v

+1=Oi.e.,uv+u 2 +v 2 =0
.

or

2 + v u u 2

v =

0 which is a circle with centre at

(-,)

and radius

FUNCTIONS OF A COMPLEX VARIABLE

189

Example 4. Find the image of infinite strip Also show the regions graphically. Sol. The given transformation is 1 1 w= or z-- w z so that If or
JIy
=

under the transformation w

1 or x+iy= u+tv
=

uiv 2 2 u+v

2 2 u+v

and y

2 2 u+v

1 then y= , 7

O or v 4v= + 2 u + 2 then
2 2 u+v

1 v 2 4 u+u 2) 4 +(v u = 2
2 +

2 2 v or a

2v

2 (v 1)2 0 or u

V
y
=

1/2

////////// //////////4
y=

1/4

0
W-plane

ZpIane

Hence the infinite strip 2 Cu u

+

SyS
=

is transformed into the region between the two circles

and

4, centre (0, 2), radius 2 2 + (v + 1)2 = 1, centre (0, 1), radius 1. u 2 x y Example 5. Show that the image of the hyperbola 2
2)2

1 under the transformation

2 is the Lemniscate p

cos 20.
=

(Bombay 2005; J.N.T.U. 2005)

So!. Given transformation is w

w= pe 1

Letz= re l-j
...

1 p=,=e

190 Equation of the hyperbola is 2 9 = 1 i.e., x cos2O=1 2 r

A TEXTBOOK OF ENGINEERING MATHEMATICS

or or

2 cos r 20

29 sin

-4 p

COB

2 C )

or p 2=

COS

2p.
=

Example 6. Show that the transformation w onto the straight line 4u

+

2z + maps the circle x 2 z4

+9 4x

0. w z
=

So!. The given transformation is The inverse transformation is

2z+3 z4 4w+3 w-2

Now the equation x 2 +9 4x = 0 can be written as zi Substituting for z and from (1), we get 4w+3 4iZ+3 w2 2 or or (4w+3 2 w2
+

2(z + ) =0

4i+3 2

l6w+l2w+12i+92(4w+3jj_8w_6+4wj1+3w_8Y_6)O 22(w + ) + 33 = 0 or 22(2u) + 33 = 0 or 4u + 3 = 0. Example 7. Show that w

=

1f-

maps the real axis of the z-plane into the circle I wi

1 and the halfplane y

>

0 into the interior of the unit circle w I

1 in the w-plane. (P.T.U. May 2007)

Sol. Given transformation is w

+z
=

IwI
or
or

= =
=

1=

iz i+z

br or

Iizi

li+zI
Ix+i(1+y)I

Iixiyj
Jx2

iix+iyI

Ix+i(1y)I

/x2 2 or +(1+y) 2 = (1 +y) (1 y) 2 or 4y=O or y=O which is the equation of real axis in z-plane Hence the real axis of z-plane is mapped into I w = 1 Now the interior of I w = 1 means 2 +(1y) wI <1 or or i.e., i.e., Iiz)<Ii+zI

x < + 2 (1 (1-t x -y) y) or 0 y< 4 y>O Hence the half plane y > 0 is mapped into the interior of the circle

IwI

1.

FUNCTIONS OF A COMPLEX VARIABLE

191
=

Example 8. Show that the transformation w

2 maps the circle z 1 I z

1 into the

cardioid p

2(1 + cos 4)), where w pe in the w-plane. e 2 2 = pe = r Sol. Let z = rezS, then w = z 2 and 4) = 28 so that p = r Now the equation of circle in the z-plane is Iz1 =1 or Ix+iy1 1=1 1) y 1 (x = 2 I(x1)+iyl=1 or or or
or or w-plane.

)] =re +isinO +iy r(cosO z=x = 2rcosO=O [: 0 2 r y or 2x=O + 2 x =4cos r O or 2 r=2cos8 [Using (1)] 2 = 2(1 + cos 28) or p = 2(1 + cos 4)) r The circle z 1 I = 1 in z-plane transforms into the cardioid p = 2(1 + COS 4)) in

Example 9. Determine the region of the u-plane into which the first quadrant ofz-plane . 2 is mapped by the transformation w = z Sol. Proceeding as in example 5, 4) = 28
For the first quadrant in z-plane, 0
<

<

4)

iv

Hence the first quadrant of z-plane is mapped into the upper half of u-plane.
Example 10. Determine the region of the w-plane into which the region

1 and

y
so that

1 is mapped by the transformation w

Sol. The given transformation is 1 we have -i-,

. 2 z
2 = (x

2 or u + iv = (x + iy) 2 w =z 2 and v = 2y u = y u
1
=

) 2 y

2ixy

When x

y 2 and v

so

that v 2

1 ( (u

[a left-handed parabola with vertex

(-_,

0) and latus rectum 1]

1) atv 4(u 2 soth haveu=1y and v=2y = 2 Whenx=1,we [a left-handed parabola with vertex (1, 0) and latus rectum 4] 1 21 andv=x Wheny= -,wehaveu=x

sothatv

=u+

[a right-handed parabola with vertex (

0) and latus rectum 1]

) atv 4(u-i-1 2 soth x = aveu=x v=2 =1,weh land 2 Wheny [a right-handed parabola with vertex ( 1, 0) and latus rectum 4]

192

A TEXTBOOK OF ENGINEERING MATHEMATICS V

NV

( 1,0)

ui N N
\1I4.O

>1<

1(1, 0)

i___
0
Z-plane

1 y=

y = 1/2
-,

W-plane
=

Thus the rectangular region bounded by the lines x into the region bounded by the parabolas
v2=(u_),v2=.4(u_1)

1 and y

1 maps

and v2=u,v2=4(u+1).
=

Example 11. Show that the transformation iv

ellipse u
= (C +

z+

maps the circle I z I

=

c into the

cos 0, v

= (C

sin 0. Discuss the case when c z I

=

1.

Sol. A point on the circle

z

c can be written as z
C

0 ce

w=z+ becomesu+iv =cei8+e_i0 or

u + iv
=

c(cos

+ i

sin 0)

(cos
C

i sin 0)

..

i .( ( c+ cos 9 + z C
c J

1. sm 0

( i u=Ic+IcosO
.

and

C)

1. ( v=Ic!smo

C)

which are the parametric equations of an ellipse. Whenc=1,wehaveu=2cos9andv=0 Since I cosO I 1,weget2u2andv=0 The transformation gives a segment of the u-axis of length 4.
Example 12. Show that the transformation w
=

z+

converts the straight line arg z

(1 a I

into a branch of the hyperbola of eccentricity sec ci.

<

(Bombay 2005 S)

Sol. Let z

re and w
U +

u + iv re
+

iv

1 e
r

FUNCTIONS OF A COMPLEX VARIABLE

193

=r(cos8+isinG) (cosOisinO)

( i =Ir+IcosO+ilr__Isin9 r)

( i sine ( i u= Ir+I cosO,v= IrI 1)

S

arg

rj a=9=a
i r)

u=Ir+Icosa,v=IrISma
\

i. r)

Eliminate r; 1 r+=
r
U

cosa

V 1 ,r= sina r

i i ( ( Ir+i IrI

r)

r)

u 2 cos 2 a
cos a
=

sin 2 a
-

sm 2 a a
it

this equation represents a branch :. 2 2 2 4sina 4cosa of hyperbola in w-plane 2 a 2 = 4 sin a, b 2 2 = 4 cos a Here Eccentricity e of the hyperbola is given by 2 Ce 2=a b - 1) 2 1=tan e a 2 1) (e a=4cos 4sin a 2 2a 2 a = sec = 1+ tan e=seca. Eccentricity of the hyperbola = sec a 2 and show that it transforms the region Example 13. Discuss the transformation w = e axis at y = iz into the upper half of the i-plane. between the real axis and a line parallel to real Sol. Let w = Re, then the given transformation becomes or 1 where
<

...

R=e and=y The real axis i.e., y = 0 maps into the positive u-axis p = 0 in the w-plane. = it in the w-plane. The line y = it maps into the negative u-axis
Rej4)=eY=eX.eiYsothat

Thus the region between the lines y = 0 and y = it maps into the upper half of w-plane. Note. The region between the lines y = 0 and y = it maps into the lower half of w-plane. ez The region between the lines y = c and y = c + 2it maps into the whole of the w-plane, since is periodic with period 2iti. 0 = 1 in the w-plane. The imaginary axis x = 0 maps into a unit circle R = e

194

A TEXTBOOK OF ENGINEERING MATHEMATICS

=

Example 14. Show that the transformation w ez is always conformal. Under the mapping, find the images of the regions. (i) the line segment 0 <y <A, A <2; x < 0 (ii) the rectangle bounded by the lines x 0, y 0, x 1 andy 7C
= = = =

(iii) The rectangular region bounded by the lines a x <b ; c y d. (P.T.U., Dec. 2005) w=fiz)=ez Sol. f(w) ez 0 for anyz The transformation is conformable in the region of w i.e., for all values of z Let z x + iy and w-Re
= =

&ieX+Y=eX(eiY)

and

(i)For0<y<A,A<2it,x<0wehaveO<p<A i.e., x<0 R<1=0<IwI<1 The image curves are given in Fig. (i).

0<argw<A<2it

y=A 3

titt=tttttttt

Z-plane

T
Fig. (i)
=

W-plane

1 whose angle of rotation is Image is the interior of the portion of the circle I R I A <2,t. (ii) The rectangle formed by x 0, y 0, x 1 and y R=1 x=0 4=0 y=O R=e x=1 y=ic Ifwe take w u + iv, then u + iv e X+iY ex (cosy + i siny) u ex cos y, v ex srn y u=1,v=0 x=0,y=0 u=e,v=0 x=1,y=0 u=e,v=0 u=1,v=0. x=0,y=it =
= = = = = = = = = = =
.

FUNCTIONS OF A COMPLEX VARIABLE

195

The image curves are given in Fig. (ii),

Y B(1,
it)

(e, 0) (1,0)

0(1,0) (e,0)

-U

z-plane

w-plane Fig. (ii)

=

Image is the region included between two semicircies I R I = 1 and I R I y <d c x (iii) The rectangle bounded by the lines a eaexeb = e(<R<eb axb
cd cyd = The image is shown in Fig. (iii) V
y=d
C

i-plane

w-plane

Fig. (iii) The image is the region PQRS. Example 15. Discuss the transformation w = sin z. Sol. The given transformation is w=u +iv =sin(x+ iy) iv = sin x cosh y+icos x sinh y u+ or u = sin x cosh y so that v = cos x sinh y and (1), we get equation from y Eliminating
2 smx Thus the straight lines x

2 y = 11 = 1 2 y sjp,h {. cosh 2 cosx = c in the z-plane are mapped into confocal hyperbolas in the

196

A TEXTBOOK OF ENGINEERING MATHEMATICS

2 v u 2 = + 2 y sinh y cosh Thus the straight lines y = c in the z-plane are mapped into confocal ellipses. The lines x = 0 and y = 0 map into the lines u = 0 and v = 0 respectively in the w-plane. Example 16. Show that transformation w = cosh z maps the lines parallel to x-axis and lines parallel to y-axis into confocal central conics. SoL Given transformation is w = cosh z i.e., u + iv = cosh (x iy) = cosh x cos y + i sinh x sin y

Eliminating x from equation (1), we get

u=coshxcosy v=sinhxsiny Lines parallel to x-axis are given by y u = cosh x cos k u = sinh x sin k coshx= sinhx=
U

cos k

sin k Squaring and subtracting, we get

= 1, which is a hyperbola in w-plane. sin 2 k Here 2 k, b 2 = cos a 2 = sin 2k Eccentricity of the curve is given by b 2=a 2 Ce 2 1) 2 k = cos sin 2 k (e k = sec 2 2 1) or e 2k 2 = 1 + tan e=seck Fociof(2) are(ae, 0)=(cosk .seck, 0)=( 1,0) Centre of(2) is (0,0) Now lines paralled toy- axis are x = From (1), u=coshkcosy v = sinh k sin y

cos 2 k

coshk

cosx

v =srnx sinh k
= 1 which is an ellipse cosh k smh k Here a k, b 2 k 2 2 = cosh 2 = sinh Eccentricity of ellipse is given by b 2 Cl e k (1 2 k = cosh 2 2=a ) i.e., sinh 2 2 k = 1 k = sech 2 tanh esechk Foci of(3) are ( ae, 0) = ( cosh k sech k, 0) = ( 1, 0)

Squanng and adding

U 2 2

V 2 2

) or 2 e

FUNCTIONS OF A COMPLEX VARIABLE

197

Centre of(3) is (0,0) The centres and foci of both the conics are same. Hence lines to x-axis and y-axis maps into confocal central conies. Example 17. The bilinear transformation w = into similar circle of arg
W 1 W
=

az+b zz transforms the circle arg +d 2 z z

w w

2 respectively. ,z 1 2 corresponds to z ,w 1 constant where w

Sol. Here

az + b cz +d b az + 1 ;w= +d 2 cz +d 1 cz

So that

w 1 2 ww
W

1 +b az + b az 1 +d cz d cz +b 2 az+baz +d 2 cz+cl cz

1 cz +d z z 2 +dzz 1 cz 2 d cz + 2 +a 1 cz

ww 1 2 ww
arg

1 ZZ
=

2 ZZ

where

.t

is complex constant and

.

1 (zz ww 1 =argLLj 2 ww
=

arg
arg

.t + t +

arg A

z z

1 z

= =

k (say) where k is real because argument of a complex number is always real.

Now, arg

W W

1 and k; represents a circle in w-plane passing through the points w

2 W

2 in the z-plane. 1 and z 2 corresponding to z w

Example 18. Find the bilinear transformation which maps the points z = 1, i, 1 into the points w = i, 0, i. (Bombay 2006) Hence find the image of I z I <1.

Sol. Let the required bilinear transformation be

az+b cz + d

198

A TEXTBOOK OF ENGINEERING MATHEMATICS

Substituting the corresponding values of w and z in (1), we get a+b ai+b a+b 1 = 0= z= c+d ci+d c+d
,

These equations can be written as (a

+

b)

i(c +

b and
From (3), Adding (2) and (4), Subtracting (4) from (2),

d) = 0 ia = 0

( a

b) i( c d) = 0 b=ia
2b 2ic 2a 2id
=

0 0
w

or c or d
=

a ia
or

[Using

(3)1

Substituting for b, c, d in (1), we get

az
azia

z+z

which is the required bilinear transformation. Now, from (5), I

z

1-w 1+w

I <1 is mapped into the region

i

1w <1 1+w

or

lullwi <1 i1+wi I

or 2 (1u) < + (1+u) v
.

or
or or

1w I <I 1+w I 1uk 1<1 1+u+iv I u>0.

I i 1=1

Hence the interior of the circle x 2 +y 2 = 1 in the z-plane is mapped into the entire half of the w-plane to the right of the imaginary axis.

EXERCISE 3.9
1. 2. Find the image of the circle I z I = 2 under the transformation w = z 3 + 2i. Find the image of the triangle with vertices at i, 1 + i, 1 i in the z-pane under the transforma tion w = 3z + 4 2i. Determine the region in the w-plane in which the rectangle bounded by the lines x = 0, y = 0, x = 2
and y = 3 is mapped under the transformation w = Je4 z Consider the transformation w = 2z and determine the region 1) of the w-plane into which the triangular region D enclosed by the lines x = 0, y = 0, x + y = 1 in the z- plane is mapped under this transformation Find the image of the semi-infinite strip x> 0, 0 < y <2, under the transformation w = iz + 1. Show the region graphically.
.

3.

4.

5.

FUNCTIONS OF A COMPLEX VARIABLE

199

6. 7.

Find the image of the circle f z

3 I

5 under the mapping iv

Show that the map of the real axis of the z-plane on the w-plane by the transformation w is a circle and find its centre and radius.

z+i

8.

Prove that w

maps the upper half of the z-plane into the upper half of the u-plane.
=
--

9.
10.

Under the transformation w

f--, find the map of the circle I

=

z I = 1 in the w- plane.

Show that the transformation w

---,

1+z

transforms the circle I z I

1 into the real axis of

w-plane and the interior of the circle I z I < 1 into the upper half of the w-plane.
11.

Show that under the transformation w

real axis in the z-plane is mapped into the circle

12.

13. 14.
15.

I w I = 1, What portion of the z-plane corresponds to the interior of the circle? , obtain the map in the u-plane of the square with vertices (0, 0), 2 Under the transformation w z z-plane. the in 2) (2, 0), (2, 2), (0, . 2 Find the images of the straight lines x = 0, y = 0, x = 1 andy = 1 under the transformation w = z Determine the region of the w-plane in to which the triangle formed by x = 1, y = 1 and x + y = 1 . 2 is mapped under the transformation iv = z maps the domain in the z-plane to the right of the line Show that the transformation iv = x = a into the interior of a hyperbola in the u-plane. Show that the map of the circle Iz I find its axes and centre.
=

16.

2 under the transformation

iv +

2i

is an ellipse,

17.

18.

1 b 2 a (a + b) in the transforms the circle I z I = 4z z-plane into an ellipse of semi axes a, b in the w-plane. Find the bilinear transformation which maps: (a) the points z = 1, i, 1 onto the point w = i, 0, i. (V.T.U. 2006) (b) the points z = 1, i, 1 onto the point w = 0, 1, under this mx = line the of y image the Find 1, (c) the points z = 0, i, 1 into the points w = i, 0. transformation. Show that the transformation
iv =

Answers
1. 2. 8. 4. ) 3) 4 i-(u2 (u 2 = A triangle with vertices (4, 1), (7, 1) and (7, 3). Rectangular region bounded by the lines v = u, v = u, it + u = 4, u A triangular region bounded by the lines u = 0, v = 0, u + v = 2.

6.

200

A TEXTBOOK OF ENGINEERING MATHEMATICS

5.

1<u<1,v>0 ( i 1

6.

W+j

=j

7 11. 12. 13. 14. 16. 18.

9.v=0

y>O
(0, 0), (4, 0), (0,8), ( 4, 0); and isosceles triangle. Negative real axis, positive real axis, v 2 = 4 (1 u), v 2 = 4 (1 Region bounded by the parabolas v 2 = 4(1 u), u 2 = 1 2u 53
+

u).

z+i (a)w= zz (c)w= i(1+z) + ;m(u ) 2 +2um=0. u 1z

(b)w=

i(1z) 1+z