[go: up one dir, main page]
Scribd
Upload
52 views75 pages

Chapter 10 Complex Function

Chapter 10 introduces functions of a complex variable, focusing on analytic functions and their properties. It discusses the complex number system, sets of points in the complex plane, and various types of regions, including open and closed sets, bounded and unbounded sets, and domains. The chapter also defines complex-valued functions, their domains, and the concept of single-valued versus multiple-valued functions.

Uploaded by

thakurgungun599
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
52 views75 pages

Chapter 10 Complex Function

Chapter 10 introduces functions of a complex variable, focusing on analytic functions and their properties. It discusses the complex number system, sets of points in the complex plane, and various types of regions, including open and closed sets, bounded and unbounded sets, and domains. The chapter also defines complex-valued functions, their domains, and the concept of single-valued versus multiple-valued functions.

Uploaded by

thakurgungun599
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
You are on page 1/ 75
Chapter 10 Functions of a Complex Variable: Analytic Functions 10.1 Introduction While studying the real number system IR, we have seen that there does not exist any real number whose square is a negative real number. Thus, the square root of a negative real number is not a valid arithmetic operation in IR and hence an equation of the form x? + 1 = 0 has no roots in IR. If these roots are to be determined, we need another number system called the complex number system defined on a complex plane. In a first course in Mathematics, you have studied the algebra of complex numbers, polar form of a complex number (defined in the Argand diagram), DeMoivre’s theorem, powers and roots of complex numbers. In this chapter, we shall introduce the complex sariable and functions of a complex variable. Further, we shall study the concept of limits, continuity and differentiability of the functions of a complex variable. In the process, we are led to the notion of an analytic function which is very important in any application of the complex variable theory. 10.2 Sets of Points in the Complex Plane Let S be a non-empty set of complex numbers. Let 2) € S be any complex number and 5be a positive real number. Then, we define the following: Circle The set of points which satisfies the equation Jz—2y)= 6 or (- a) + - yo) =F (10.1) defines a circle C of radius 5 with centre at zy = (Xo. Yo). This set consists of all points which lie on the boundary of the circle C (Fig, 10.1a). Any point z on this circle has the polar form z =z, + de®. As 6 varies form 0 to 2z, z traverses once over this circle in the counter clockwise direction. If % = 0, then the equation |z| = 6 defines a circle of radius 5 about the origin. 10.2. Engineering Mathematies n — 4) < ry is neither open nor closed Neighborhood of a point A.8- neiehhorhood of a point P(z,) in the complex plane is the set of all points = which he in the open disk | | < 6. Usually, the 6 = neighborhood about the point Ai) is denoted by NLP, 3) oF Vs(P). Ihwe exclude the point zy from the open disk |z ~ z9| < 6; then it is called the devered neighborhood of the point zy and is written as 0 < |z— zl < 6. Interior point A point = isan interior point of S, if all the points in some 5 - neighborhood of z eis S In Figs. 10.1 and 10.1, P is an interior point, Futerior point A point e outside S In Fig. 10.1a exterior point of S, if all the points in some 6 - neighborhood of id 10.1, 7 is an exterior point, oundary point A point z is a boundary point of S, if every 5- neighborhood of z contains at least point of Sand at least one point not in S. For example, for the set of points defined by Im (z) 2 1, wints on the Tine vy = 1 are the boundary points. The points on the circle |z — zo] = r are the points for the disk |z ~ z9| < r. The point Q in Fig. 10.1¢ is a boundary point. The totality | the boundary points define the boundary of S. Open set A set S is open, if every point of S is an interior point, For example, the sets zol 7} is an unbounded set. Connected set An open set S is connected, if any tw6 points 2, and z, belonging to S can be joined by @ polygonal line (a path which consists of finitely many straight line segments) which is totally contained in S (Fig. 10.2). Domain An open connected set is called a domain, Usually, a domain is denoted by D. Region A region is a domain together with all or some or none of its boundary points. Thus, a domain is always a region but a region may or may not be a domain, For example, an open disk is both a domain and a region but a closed disk is a region and not a domain, Usually, a region is denoted by R. 10.4 Engineering Mathematics (@) () Fig. 10.2, Connected sets. Extended complex plane The complex plane to which the point at z = o has been added is caleg the extended complex plane. The complex plane without the point at z= ©» is called the finite complex plane. Example 10.1 Describe the regions : Re(z) # 0}, (b) {z Im) > 1}, im (z) < Iz 1P}, (b) {z: 0 < Arg (z) $ 2/4}. State whether the region is a domain or not. Solution (a) The region is entire complex plane excluding the y-axis. It is open, it is not connected and it is unbounded. It is not a domain (Fig. 10.3a). Functions of a Complex Variable:Analytic Functions 10.5 (b) The region is the upper half-plane y > 1. It is open, connected and unbounded. It is a domain ig. 10.30). (©) The region is given by 1 Sl 2 > od y<(x-1P +, or wwe (y-F >a, y<(v- IP +y Pe ly-s] ag. or 5 i (ast : ( 2) which is the exterior of the circle with centre at (1, 1/2) and radius 1/2. It is open, connected and unbounded. It is a domain (Fig. 10.3c). (d) The region is given by 0 Stan" (ix) < m/4, or OSyS1, or OSySx. It is not open, connected and unbounded. It is not a domain (Fig. 10.3d). Exercise 10.1 Describe the following regions. Are they open/closed, connected and bounded/unbounded? Determine the regions which are domains. 1. Re (2) >0. 2, Re@ 0. 6. -2 [Re (2) 12. Im (2 - ) > Re (2 +2 +31). 13. Im (3/z) < 1/3. 14. 0S arg (2) < 703. 15, (w/4) arg (2) $ (m2). 16. ~(n/4) < arg (z) < (70/4), 17. Re (z) $ 1 and 0 < arg (2) < 7/4. 18. Iz-11S1z4 11. 19 122+ 11 2. 23, I2P +2 Rele1<3. . 24. |z-il>IzlandIzl< 1. 25. | +il4. 10.3 Functions of a Complex Variable We define the function of a complex variable in a similar way as the function of a real variable. Let S and S° be two non-empty sets of complex numbers. If there is a rule f, which assigns a complex number w in S” for each z in S, then fis said to be a complex valued function of a complex variable zand is written as. w=f(2) (10.6) The set S is called the domain of definition of f. When the set S, on which f is defined is not specified, we consider the so called natural domain of definition of the function f to be the set S, which can 10.6 Engineering Mathematics be the whole of the complex plane or some restricted part of the complex plane, for which fiz) is defined and is in S* c C. For example, the polynomial function Ne) = Pyle) = age + aye te ty z+ dy My #9 (10.7) where do, yy... a, are real oF complex constants, ean be computed for any z. Therefore, the natural domain of definition of P,(=) is the whole of the finite complex plane. However, the rational function ee) (10.8) A d Q{c) are some polynomials in 2, is defined for all z except for those values of 2 for which Q(=) = 0. Therefore, the natural domain of definition of the rational function defined in Eg, (10.8) is {2 sz € C and Qt) # 0}. fe) The value of w for a given rule f The inverse image or the pr that is (29) is called the image of the point z = zy under the image of point w in S* is a point z in S which has w as its image Let z = x + iv. The image w of the complex number z is also some complex number, say u + iv Therefore, we can write wok where u(x, v) and v(x, y) are real valued functions of two real variables x and y. Therefore, w= u + iv can be represented as a point (uw, v) in the w-v plane, also called the w-plane or the function plane (Fig. 10.4). = u(x, y) + iG, y) (10.9) z-plane (complex plane) w-plane (function plane) Fig. 10.4, A function w = fi . ue oF all images in S* is called the range of the function, Thus, the function w comet) transformation or a mapping which takes the points from a set $ in the into the points in the set S* in the w-plane, Therefor i a re, z) i Pars the function f(z) maps a region R in the = > Functions 10. Functions of a Complex Variable:Analytic Functions 7 an uniqi ihr Flee aaa be there exists mage in the w-plane, then the function f(@) is called a single-valued function. However, in the theory of complex variables, we come across functions which take more than one value far every = in D. Fore 7 ln fi h=zorw=z'",n intey the function a positive integer takes the» v pI ues , 04% wy = Il! [om *)vran( 2288) k=O, Myce n4) a n for a given =, Such a function is called a multiple-valued function. In such a case, we restrict the discussion to those parts of the domain in which the multiple-valued function behaves like a single- valued function. Fach one of these single-valued functions is called a branch of the multiple-valued function, Thus, the term function implies a single-valued function unless stated otherwise. Remark 1 (a) Given any two real valued functions wand v in two real variables x and y, we can write Ww 4 (%..¥) + i2(%, y), which defines a complex valued function. However, it may not always be possible to write w as a function of z. For example, if u(x, y) = 2x + y and v(x, y) = 6xy, we can write w = (2x + y) + i(6xy), but it is not possible to write w as a function of z alone and obtain / (b) If= = r(cos @ + i sin 8) = re’® = (r, 6) is taken in polar form, then the real and the imaginary Parts of f(z) can be expressed as real valued functions of the real variables r and @. Thus, we can also write w=f@) = f(re®) = u(r, 8) + iv(r, 8). (10.10) Example 10.2 Find the image of the point z = 2 + 3/ under the transformation w = f(e) = 2(z ~ 2/). Solution Substituting z = 2 + 3/ in w = f(c), we obtain the corresponding image in the w-plane as =f (2 + 3i) = (2+ 3) (2 + 31-21 w (2+3) (2+ i)=1+ 8. Example 10.3 Find all the pre-images in the z-plane under the transformation w = z (2 ~ 2i) = — 1. Solution Solving the equation (2 - 2) =— 1 or z?— 2iz +1 =0 for z, we get i+ 78 z= HEN Loa di 2 Hence, the pre-images of w =— 1 are z = (1 + y2)i. Example 10.4 For each of the following functions, express (2) in the form u(x, y) + iv (x, y) by separating into real and imaginary parts. State the natural domain of definition of the function in each case. @) fe) = |2P, Gi) fle) (iv) fe) —i fimeP, 10.8 () fe) = > Solution Let z= x + iy wi) i) (iii) (iv) (vy) Engineering Mathematics +22 and w = f(z) = u(x, ») + iv(x, y), or w= ut iv. fe) w= 4)? and v=0. xt yh Therefore, Since f(z) is defined for all z € C, the natural domain of definition of f(2) is the whole complex plane, 1 x-iy xtiy x+y z#0. x +y? Therefore, u=xw0? +) and v =~ yi(x? + y’) for all z #0. Since /(¢) is defined for all z € C, 2 #0, the natural domain of definition of f(2) is the whole of the complex plane excluding the point z = 0. 1 10° » ele Therefore, u= 1/1 - (+), and v= 0, for all z, [zl 41. Since f(z) is defined for all z, except for those Points which lie on the unite circle |z| = 1, the domain of definition of /(2) is the whole complex plane excluding the points z for which lz]=1. 1, 2 a @)= —— -ifim(@P=+- 57, xe0. fe) Re i [Im @} 37¥ Therefore, u= Wx and v=-y, x#0, Since the function f(2) is defined for all z exe ept when x = 0, the domain of definiton of f(z) is the whole complex plane exculding the po ints lying on the imaginary axis, that is, x = 0. f= (x-D+y _| (x-I)+iy (1+ 2x) ~ iy 1422 (1+2x)+2iy (1+2x) + 2iy (1+ 2x) - 2 = Ue= N+ 2x) +297] 4 3iy (1+ 2x)? + 4)? : z 2 Therefore, w= SAW 2x) 297 (1+ 2x)? +4)? Since the function f(z) is defined for all z except at z LG) is the whole complex plane except the point z and p= —_ 39 (1+ 2x)? + 4y? ~ 1/2, the domain of defi nition of ~ 1/2, that is (- 1/2, 0). oF 10.9 Functions of a Complex Variable:Analyic Function’ jive Example 10.5 Write the funetion w = 2 in the form u(x, y) + tx, y) and u(r, + 7010s 8) GE its graphical representation. Solution We have w = f(z) = 2°. Substituting z = x + iy, we get we (xt py =x = y+ 2ixy = ula, y) + HCL Y)- Therefore, ux, y= 37 =? and v(x, y) = 2ay Substituting z= (cos 6+ i sin 8) = re®, we get w= Pe? = 7 (cos 26 + i sin 26) = u(r, 8) + iv(r, 8). Therefore, u(r, 8)= P cos 26 and v(r, @) = 7 sin 20. Writing w = R(cos 9 + i sin 9), we get R =? and @ = 20. The graphical representation of the function w = z* is given in Fig. 10.5. A circle r = a in the z-plane is mapped onto the circle R = a” in the w-plane. The sector r < a, 0 < @< 2/2 in the z-plane is mapped onto the sector R < a’, 0 < @ S 7: in the w-plane. The image-of the hyperbola x” — y? = c, is the line w = c, and the image of the rectangular hyperbola 2xy = ¢) is the line v = c, where c, and c, are real constants. z-plane w-plane Fig. 10.5. The function w = 2. 2 Example 10.6 Find the image of the line Im(e) = 1 under the mapping w y and Solution For the function w = 2? = (x + iy)’ = u(x, y) + iv(e, y), we have u(x, y) = v(x, y) = 2xy. Substituting Im(z 1, we get y 1 and v(z, y) = 2x. (10.11) inating x from Eqs. (10.11), we obtain 2 u(x, y) = an Ey or v= 4 (u +1). } Therefore, the image of the line y right. It intersects the v-axis at (0, + 10.10 Engineering Mathematics = 1 is the parabola with vertex at (-1, 0) and opening tothe 2) (Fig. 10.6). Fig. 10.6, Example 10.6. Example 10.7 (a) Show that the equation w = Vz has more than one solution and hence defines a multiple-valued function. Write its branches and show that each branch is a single-valued function. (b) Discuss, whether w = 2 is a multiple-valued function, when (i) @ is an integer, (ii) @ is a non-integer. Solution (a) Let w= Vz =u + iv, We get wz or (ut inv =x+iy or wv? + imax tiy. Therefore, Wor =x and Qu = y. From the second equation, we find that if y # 0 then neither u, nor v can be zero. Hence, for z not real, we obtain w= yiQv) and [y/(Q2v)P — v Solving for v*, we get =x or vi +xv?— (7/4) = 0, [<3]. If we take negative sign, then becom: es ni ich i « Therefore, we have legative, which is not possible as v is real. : Pep and u=y20), where p= fre {Fay y?/VE- ‘Thus, for every complex z, w takes two values 2v or Funct ; o.11 Wnctions ofa Complex Variable:Analytic Functions | ‘Therefore, the function w ; zis a multi cach of which is a single. alued function, Geometrically, we can interpret this result . - te origin (Fig. 10.7) as follows: Let z = re, Consider a circle of radius r about ple- . = le-valued function, it has two branches w =; and W = Ws Fig. 10.7. The function w = vz. Now start at some point P(z = z = re) on this circle. Then w = /(e,) = Vre?, where = 0, = arg(z\). After one complete revolution @ = 6; + 2m, we get fp eli + 2a? Ir gio Fel w é which is not the same point P, Hence, the graph has two branches. However, after completing one more revolution 8 = 6, + 47, we get wa Vr eO+402 . Jy cil? , oni. i002 Which is the same point P. Therefore, if 0 < @< 2m, we are on one branch of the multiple-valued function w = Vz and if 2n< < 4z, Weare on the other branch of this function. Obviously, each one of the branches represents a single-valued function, The first interval 0 < 0 < 27 is usually called the principal range of 8 and the corresponding branch is called the principal branch of the multiple-valued function. In order to consider a single- Valued function representing we delete the line x > 0, y= 0, that is the line O@ in Fig. 10.7, \here the point Q is at infinity. This line O is called the branch line or the branch cut and the point O is called the branch point. It may be noted that we could have taken any other directed line through O as the branch line by suitably defining the principal range of 6. Since a complete revolution about any Point other than z = 0 does not produce different values of w, the point z= 0 is the only finite branch Point for this function. 10.12 Engineering Mathematics (b) Let z= re. The point z = 0 is the only possible branch point. Now, consider a circle of radius r about z = 0 (Fig. 10.7). Take a point P(z rel®l) on this circle. Then we fe) After one complete revolution @ = 0, + 27, we get re, where @ = 6, = arg(z))- wae ntl yi gE = GIMME f(z), Thus, the difference between the value of f after one complete revolution and its original value f(z)) is given by Asie) = [e°"* ~ 1] Ae). (i) If vis an integer, then A f(z) = 0 and we reach the starting point P. In this case, the given function is single-valued and therefore the function has no branch points or branch cuts Gi) If a isa non-integer, then A(z) # 0 and we do not reach the starting point P. Thus, in this case, the function is a multiple-valued function and the point z = 0 is a branch point for any non-integer @ We also note that there is no branch point other than the origin z = 0. Any directed line through the origin is a branch cut. For a= In, na positive integer, the function w = z™ has n branches, origin is the branch point of order (n - 1) and any directed line through the origin is the branch cut. Exercise 10.2 Find the domain of definition for the following functions. 22431 +(1-iz Find the image of the following functions at the indicated point. 6. Fat z=-1-2i, Tow 22+iF +Ime) at z=24 71 +241 at z=344i, 9. at 2=2+3) Mm 2e1-i, n. Write the following funct 12, 2 ons in the form u(x, ») + iv (x ¥). 13, 2. 14, 2. 10.13 Functions of a Complex Variable:Analytic Functions 17, 2, Rea Te Rew) #0. 19, Re(z) + Im(z + 1) - 3iz. 2 z+1 eu. z a 23. ree i. z4i Me + Q2- Iz #12 28. aie +1), 84 i 10.4 Elementary Functions In this section, we introduce a few elementary functions of a complex variable and analyse them. These functions, /(2), reduce to their counterparts f(x), the corresponding functions of real variable when z = x is real. We now discuss some complex elementary functions which are often used. 10.4.1 Exponential Function Ifa and b are real numbers, then we have (10.12) and We now formally write w (10.13) Substituting @ = iy in Eq. (10.12), we get Using the Maclaurin’s series expansion of sin y and cos y, we obtain the Euler's formula el =cosy+isiny. (10.14) Hence, we obtain from Eq. (10.13) weet =e [cosy tisiny]=u tiv. (10.15) 10.14 Engineering Mathematics Therefore, the image of the point z = (x, y) in the z-plane is the point w = (u, v) in the w-plane, where w= e" cosy and v= e" sin y. When y = 0, that is, z =x is real, it reduces to the exponential function of the real variable x Substituting x = z in the Maclaurin’s series for the real exponential function (10.16) Taking the limit of the magnitude of the ratio of successive terms in the complex series (10.16), we get Hence, the series expansion for the complex exponential function e converges for all z, The exponential function satisfies the following properties, which can be easily verified, () eis defined for all z. Therefore, its domain of definition is whole of the complex plane, Gi) e* is a single-valued function, (ili) |e‘| = Je) = Je'| |e} = &* and Arg(e*) GW) e° #0 for any 2. Therefore, its range is the whole of the complex plane except the origin. @ & emai (vi) e7! (vii) e71 e2 = e1*72, if and only if z = 2nai, n any integer. e? if and only if 2, = z, + 2nzi, n any integer, (ili) €7! fe? = 1-72, (i) e = Ve. 0) (e5)" =e and (e*)" = 6", n any positive imteger. Periodicity Since e”**! © 008 2ne.+ 1 sin 2am = 1, we can write ef c e720) any integer Therefore, th : 5 , ° complex uantty Dor eenPonential Function e* is periodic with period 2ni, Since the period is a 2m that is On ae values of e* are assumed in any infinite horizontal strip of width ~* In? x a and tany=1 or yoatm n=0,41,42,... Hence, we obtain zexthy Linz+i(Zenn), n=O41 420. 10.4.2 Trigonometric and Hyperbolic Functions Trigonometric functions For a real variable y, we get from the Euler's formula (Eq, 10.14) 10.16 Engineering Mathematics w 1 = cosy +isiny and e"=cosy—isiny, Adding and subtracting these equations, we obtain +e") and siny= 1 (eh 6 ) 2 cosy Using these equations as a guide, we write 3 +e") and sinz= x (e - e*) cos 2 Substituting 2 = x + iy, we obtain cos Og tet : Here ver en = tte * (cos x + i sin x) + e” (cos x ~ i sin x)] = + [cos x(e” + e°) +i sin x (0? - e)] } = cos x cosh y—i sin x sinh y. (10.17) Similarly, we obtain sin z = sin x cosh y + i cos x sinh y. (10.18) The other trigonometric functions are defined as 1 . cose z= 1, secz= 1, tanz= S82, oop z= S082, sinz cos z cos z sinz whenever the denominator is not zero. Complex trigonometric functions satisfy the same identities as real trigonometric functions. For example, it can be easily verified that (i) sin (~z)=~sinz and cos (- 2) = cos z. Giy sin? z+ cos? z= 1 (ili) sin (2, + 2) 21 €08 2, + cos 2, sin 25. (iv) cos (2, # 23) = cos 2, cos z, sin z, sin z». (¥) sin 22 = 2 sin z cos z and cos 2z = cos? z ~ sin® z, (vi) sin 7 = sinz. (vil) sin @ + 2nz) = sin 2, nm any integer. (ili) cos (z + 2nz) = cos z, nm any integer. The functions sin z and cos z are I Periodic with period 27 and the function tan z is periodic with period x. f . ; 17 Functions of a Complex Variable:Analytic Functions 10. We further have |sin 2/7 a |sin x cosh y + i cos x sinh yf in? Sin'x cosh?y + cos*x sinh?y 2 . in'x (1 + sinh?y) + (1 — sin’x) sinhy = sin?x + sinl Since sinh y= (e' ~ €°)/2 is not bounded, [sin z| and hence sin z is not a bounded function (unlike its real counterpart sin x for which |sin x| $1 for a real variable x). Similarly, we find that eos z|? = cos? x + sink? y. Therefore, |cos z| and hence cos z in not a bounded function. Now, sin z= 0 = = |sin z| = = sin? x + sinh? y, for all x, y. Therefore, sinx=0, and sinh y=0, or x 7m and y= 0, m any integer. Hence, sinz = 0, only when z is real and z = nz, n any integer. Similarly, we find that cos 2 = 0, _ only when z is real and z = ((2n + 1)m/2, n any integer. Hyperbolic functions Similar to complex trigonometric functions, we define the complex hyperbolic functions as a sinh z = » coshz= 242", cosechz=—!_, sechz=—!_, 2 z sinh z cosh z tanh z= SOM goth 2 = S8hz, cosh 2 sinh z whenever the denominator is not zero. Complex hyperbolic functions also satisfy the same identities as real hyperbolic functions, For, example, @ sinh (- z) i) cosh? z - sinh? z= 1. (lit) cosh (2, + 25) = cosh 2, cosh 2, + sinh 2, sinh zp. (iv) cosh 2z = cosh? z + sinh? z. (v) sinh 2z = 2 sinh z cosh 2. cosh z. =—sinh z “and cosh (- We further have the following results: p10 = Le 2 e] = Sle - e*] =i sinh , (@) sin iz = pe - 2i je + ey ze + e] = cosh z. 10.18 Engineering Mathemati sin iz = —i sin (i(x + i) isin (~ y + ix) = i sin (y - ix) (b) sinh 2 = = ifsin y cosh x ~ i cos y s nh x] = sinh x cos y + i cosh y sin Similarly, we obtain cosh z= cosh x cos y+ Z sinh x sin y. We also have [sinh 2|? = sinh’, cos'y + cosh? x sin’y = sinhx + sin?y, [eos y + sinh? x sin? y = sinh? x + cosy, Now, sinh z = 0 => |sinh z| = 0, that is sinh? x + sin?y = 0 for alll x, ». Therefore, sinh x= 0 and siny=0 or x=0 and y= nx Hence, inh z = 0, only when z is pure imaginary, and z = n7i, where n is any integer. Similatly, we id that cosh z = 0, when z is pure imaginary and z = [(2n + 1) miJ/2, where n is any integer. We further note that sinh z and cosh z are not bounded. Example 10.10 Express w = tan z in the form w = w + iv and show that sin2x+i hy tan 2 = STS ey 0s 2x + cosh 2y Solution We have sinz _ sin x cosh y+icos.x sinh y cosz — cosx cosh y—isin x sinh y tan z= [sin x cosh y +i cos.x sinh y}[cos x cosh y + © [eos x cosh y inx sinh y]_N xsinh y][cos x cosh y +i sinx sinh y] | D where, N sin x cos x(cosh’y — sinh?y) + i sinh y cosh y (cos*x + sin?x) v 1 = sin x cos x + i sinh y cosh y = — (sin 2x + i sinh 2y) D = cos? x coshy + sin? x sinh?y = cos*x (1 + sinh®y) + (1 = cos?x) sinh®y = cos? x + sinh?y = + (1 + cos 2x) + ; (cosh 2y - 1) = 3 (cos 2x + cosh 2y). Dividing, we obtain the required result, Example 10.11 Find all values of z such that sin z = 2. Solution Let z= x + iy. We have sin z = sin x cosh y + icos x sinh y = 2. ; ‘ons 10.19 Functions of a Complex Variable:Analytic Functions 1 Comparing the real and imaginary parts, we get @ sinxcoshy=2 and (ii) cos x sinh y=0. ee we Bnd that either cos x = 0 or sinh Y= 0. That is, either x = ((2n + I)xV/2 or y = % nany integer. For y = 0, we obtain from (i) sin x = 2, which is not possible Forx = [Qn + l)n}/2, we obtain from (i) I" cosh y = 2 Since cosh y > O for every y, n must be even, Thus, y= cosh! 2 Hence, we obtain (4k + Iay2, y= cosh! 2 and 7 (4k +1) F +i cosh! 2, k= 041,42, Example 10.12 Find all values of z such that sinh z = "3, Solution Since, €™° = cos £4. j sin ¥ = ; +3, 7 WE have for z= x-+ iy 1). sinh z = sinh x c0s y + icosh xsiny= 4459, Comparing the real and imaginary parts, we obtain (@ sinh x cos y =4 and (ii) cosh x sin y = 8. . ot 3 or sinh x = Teosy and cosh x 2siny" Substituting in cosh?x — sinh?x = 1, we get 3 iy 4sin?y 4 cos? y or —3.cos* y— sin? y =4 sin® y cos” y, or 3(1~ sin? y) ~ sin? y = 4 sin? y(1 — sin? yy or 4 sin’ y-8 sin? y +3 =0. Solving for sin? y, we get sin y = 3/2 and sin” y = 1/2. Since, y is real sin? y = 3/2 is not possible. Thus, we have sin® y= 1/2 or sin yStinz. For sin y = -1/V2, we find from (ii) cosh x = - ¥3/2, which is not possible a . Hence, sin y = L = sin (3) or y=nn+(-D"F, many integer, When n is even, we get sin y = cos y = 1/2 and cosh x = 3/2 and sinh x = 1 N2, 10.20 Engineering Mathematics Hence, e vi+l -» (474) cosh x + sinh x = “h or x= When n is odd, we get sin y= 1/ cos y= -1/¥2 and cosh x = ¥3/2 and sinh x = -lV2, cosh x + sinh x » ae of xnin (43) Hence, Therefore, the solution: x + iy, where in (B+! xin ; v2 v3-1 z = i =nn-, n odd integer. and x nf ee Example 10.13 Find all values of z, such that na+ , meven integer 4 V2 sin z = cosh B + i sinh B, real. Solution We have cosh B +i sinh B= cos i + sin ip = sin ( +8] + sin iB =2 sin (E+0) cos F = V2 sin (E+i0). efore, the given equation becomes sin 2 = sin (E+). We obtain om car (Fin), n=0,+1,42, 104.3 Loga m Function ‘The real valued natural logarithm function of real variable x is defined as yolnx, if x= 1 x >0. Similarly, we define the logarithm of a complex variable Zax tivas welnz, if z=e", 2#0, (10.19) in 2 =e", we get rel? = evr, Writing = re and w= ut jy ions 10.21 ve Functions Functions ofa Complex Variable:Analytic Functe = glOvm) since + Many integer, we obtain pel? '20m = pte Therefore, le"|=e" =r or using which is the logarithm of a teal variable p= 2] and ang (e") = v= @ + Qnz, n any integer. Hence, for any complex z # 0, the solutions of the equation e” are given by w= Inz=Inr + (0+ 2nm), n any integer =Inr +i arg (@), 2 #0. (10.20) x2) in Eq.(10.20) can take infinite number of values, In z is a multiple-valued function. For agiven value of z in its domain of definition, the values of In z have the same real part, but their aginary parts differ by integral multiples of 2, For each n in Eq. (10.20), we obtain a different inch of the multiple-valued function In z. If we restrict arg (z) to the interval — a < arg (z) $ 7, which is called the principal argument of z, then the corresponding principal branch, Ln z of In z, is written as Ln z= In [z| + i Arg @) or Lnz=In|z| +i, -2<0< x (10.21a) or Lnz=In yx? +y? +i tan! (yx). (10.21b) The function Ln z is single-valued and is defined for all z # 0. Therefore, we have Inz=Lnz + 2nni. Ifz = x is real and positive, then |z| = x and Arg(z) = 0 and we obtain Ln z= In |z|=Inx, x>0 Which is the natural logarithm of real positive number x. If = x is real and negative, then |2| = |x| and Arg (2) = %, and we obtain Lnz= In |z| + ai = In |x| + mi, x <0. With this definition, we have Ln (1) =0, \ In (1) = 2nzi, Ln — = Ini + (- ), In t-p=mn v2 am The following properties of In z can be easily verified (i) (ii) In (e) = z + nai, n any integer, a 10.22. Engineering Mathematics (ili) In, (iv) In = In 2, + Inz) + 2nmi, m any integer, In =, — In zy + 2nmi, many integer. Often, In z is a Iso written as log =. Example 10.14 Find the general and the principal values of Glog + V3), (i) log = V3), Gi) log C1). Solution (i) 2 = 1+ 34, We get |=] = 2, Arg (2) = 2/3 Therefore, log (1+ V3) = log 2 +i (+200). n any integer and Log (1 + V¥3i) = log 2 + i/3. (ii) == 1 — V3i We get 2, Arg (2) = - 2/3. x ; Therefore, log (1 - V3i) = log 2 +i — 7 t2nm), n any integer and Log (1- V3i) = log 2 - ix3. iii) = ~ 140i. We get |z|= 1, Arg (@) = = Therefore. log (-1) = log 1 + i (4+ 2nz) = (Qn + 1) mi, n any integer and Log (-1) = ai. Example 10.15 Find the real and imaginary parts of Log [(1 + i) Log #]. Solution We have Log i = log 1 + (mi/2) = mil2. Now, (+ Log iz Lata mi=cisa Zé. 2 2 fore, Log{(1 + i) Log i] = Log ($43) ay 272 = log (3) +8) | + i tan! Cl) = Jog (4) += wers of a Complex Number Let © be a complex st ¢ definitions of e i logarithm functions of the o stant. Using the definit De al : hoa ions of exponential and log 10.4.4 General P Wests EE 2 #0, (10.22) , 23 Functions of a Complex Variable:Analytic Functions 10. ince log z has infini . umber of catuen When no les fo given 2, w as defined by Eq. (10.22), also has infinite ” Use cina * . 2 the principal value of 2°, © the principal branch Log z of log (z), we obtain from Eq. (10.22), Remark 2 =n is a positive integer er e ree Pe integer, then w = 2° has a unique value. When c= I/n, w = 2° has exactly 7 Example 10.16 Find the general and principal values of I te ” a Gi) + YBa". Solution (w= (ofa ame. clef] - Gr") The principal value of w (for n = 0) is e®!2, n=0,41,42,... Gi) w= nla elhoen allowe(-$-2m)] The principal value of w (for n = 0) is e*'?, Gi) w= (1+ iY) = gl tM) ivofonar(5 +2] Je (5-2 )]+[ooe-(E-2)] neothed The principal value of w (for n = 0) is given by tog2—Z] rifts 2] ago goal a? ge = 2e°" [cos (log 2) + i sin (log 2)} [1 + 312] = (1 + V3ie*? [cos (log 2) + # sin (log 2)]. Example 10.17 Using the principal branch of log 2, show that z, forRez>0 or Rez=0 and Imz>0, 212 = ae —z, elsewhere. Solution We have logis? +f Arg (22012 [22]? = elver"l? = wie)? +7 Are I) dog) z|#iARB(eIV2 = gle gfARECEY2 pay of Aratety/2 = dee 10.24 Engineering Mathematics Using the range of principal arguments for 2 Ary: for Rez>0 or Rez=0-and Imz>0, 2Argz—2a, forRez<0 and Imz20, 2Argz+2n, forRez<0 and Imz<0, Ara( we obtain the required result, 10.4.5 Inverse Trigonom rie and Hyperbolic Functions For a real variable x, we define the inverse sine function as yssin'x, when x=siny. Simila ly, we define the inverse sine function for a complex variable 2 as w=sin'z, when z= sin w. Writing sin w = (e™ — e)/(2i), we get oe =2iz or e™ — ice -1 Solving for e™, we obtain ev = 3 [eet Fava] = ice 1- Since + is covered by the double-valued function 1-2? , we can write eM =iz+ Vl-2? or iw= tog [i+ Ji 2? ]. Hence, in'z = — i log [ies 3]. Now, sin''z is defined for all z except when (10.23) t+ or iz=- JI which is not possible. Thus, sin"'z is defined for all z, inctions 10.25 ic Ful Functions of a Complex Variable:Analytic fi [Pa sin' (2) i tog | LEN? 240. cot 's = tan! (: = - 4 tog [Zt], cease. z 2 z- similarly, we define the inverse complex hyperbolic sine function as w= sinh'z, when z= sinh w. Writing sinh w = (e” - e")/2, we get e = 22 or &™-2ze"-1 Solving for e, we obtain Since = yz? +1 is covered by the double-valued function yz* +1, we can write or w= log [en +i]. log (2+ (#1). Other inverse complex hyperbolic functions are defined by the following: cosh™'z = log (+42 -1). Lez tanbr! z= 5 we ), eel. Hence, w = sinh” z I-z 1 L+yl+2? cosech”! z = sinh ‘(¢)- log Ee , 2 #0. sech™'z = 10.26 Engineering Mathematics tons" (2) = thog(Zt4), cee. 2) ain a Oe Since log = is a multiple-valued function, all the inverse trigonometric and hyperbolic functions are also multiple-valued. Using the principal values of both the square root and the logarithm functions we obtain the principal values of these functions. . Example 10.18) Write tan! in the form w + iv, Solution Let == + i and tan! z= w+ iv. We obtain u + iv = tan! (x + iy) (10.24) and u — iv = tan! (x - iy). (10.25) Adding Eqs. (10.24) and (10.25), we get tan! (x + iy) + tan! (@ - iy) = tan Subtracting Eq. (10.25) from (10.24), we get 2iv = tan”! (x + iy) — tan (x — iy) = tan” Hence, we obtain 1 + Gi) coth 2 = = log 2 8 Solution (G) Let w = tan 'z, We have z = tan w = sin w/cos w or Therefore, ; ions 10.27 Functions of a Complex Variable:Analytic Functions itz at ~£ we (22) +z) 2 i —2=0, that is, when z= 4%. We can also write This function is not defined, when it 0 or (ii) Let w = coth =, We have coth w = Therefore, we get This function is not defined when z + 1 = 0 or z~ 1 = 0, that is, when z = + 1. Example 10.20 Find all values of sin“! 2, treating 2 as complex number. (See Example 10.11). Solution We have sin! z =—i log [iz + y1-2"] For:=2,we get yl—z? =V3i and sin! 2 =~ i log (2+ ¥3)i] x Since 2 + ¥3 is positive, we get sin! 2= [inet2+.0)+{%-+200)) n=0,41,42, 0... © We obtain sin“! 2=-- i log (243) + (§ +200) = -i cosh! 2+ (E+20n), n=0,41,42,.... since cosh! 2= log (2+ V3). Exercise 10,3 Write the following functions in the form u + i. 4, Log (cos 2) a 1g (COS 2). 5. cos'(e%), 05 0S #. 10.28 Engineering Mathematics Find all the values of 2 which satisfy the following identities. 10. 12, 1M. Find the 16. log (ce $ef4l 9. ef =i. eral 11. sin (iz) = & tanz = 13. tanh z = +a 15, cosh z+ sinh z = @ a: complex constant =a + a2 5. general values of the following functions. 17. log (Log i). 18. sin [1-ee( Se: 19. tan wo] 20. DEE! a Cit. 22, 23. i*. 24. Prove the following identities. 25. 1-2? = tan! ~2— 26, cos z= sin" = tant ae 27. tan VE 2 cost et e +2? yie2? 28. sin (cos 2) = - iyz and cos (sin! z) = 1-2? 29. When is the function e (j) real, (ji) pure imaginary? 30. Show that 226°"? = (14 V3) + (1-¥3)j, 31. If cot (8+ id) = e, then show that @=(2n+1) = and eee Bry etc 4 2})* Where m is an integer. 32. If tanh (x + iy) = p + ig, then sho Pig then show that tanh 2x = 29/(1 + p+ @) and tan 2y = 2gi(l —pr- 4? 33. If sin! (uw + iv) = a+ iB, then show t 2 w that sin? @ and cosh? Bare the roots of the equation F-test ea ye 34. If sinh” (x + iy) = w+ im, then show that ions 10.29 Functions of a Complex Variable:Analytic Function a sinh” w cosh’ cos? vp sin? 0 Zap + then show that one of the values of y/x is 2 tan | (hiay/log(a® + 6°). 35. If (a + iby? 10.5 Limit and Continuity 10.5.1 Limit of a Function alued function of z defined on S which includes the 6-neighborhood of a point may or may not be defined at z= zo. Then, the function fiz) is said to have given any arbitrary small real number € > 0, there exists a real number Let f=) be a single The function alimit Le © a > 0, such that UG) - Li) zo. When the limit then for every z in the d-neighborhood of zp, the value of f(z) lies in the €-neighborhood of 10.9). Fig. 10.9. Limit of a function, Theorem 10.1 If lim f(z) exists, then it is unique. Proof Let the function f(z) have two different limits L, and L) as z — zp, that is =L, and lim fi)=Ly, Ly # Ly, lim ft) Hence, given an arbitrary real number € > 0, there exist real numbers 5, > 0 and 6 > 0, such that IPG) ~ Ly] < €2, whenever 0<|7~ 201 <4 and | f(c) - Ly] < ©, whenever 0 < [2 ~ ol < & 10.30 Engineering Mathematics If 8 = min (6), 65}, then for 0 < |z ~ zal < 6, we have [Ly — Lal = (fle) — by - WE) ~ ED ele 7 < =. y= bal 2 ts Sey Ly + Uf This implies |) — l= 0 or Ly- Ly =O or Ly > by ark 3 e ), the it jon along any straight line or a curved path (Fig. 10.10 i limit as ame along all these paths, If we obtain two different limits as z — ;, nt paths, we conclude that the limit does not exist. > sp from any dire along two dite Fig, 10.10. Limit of a function, Theorem 10.2 Let f(z) = ulx, y) + O(a, y), z= x + iy, 29 = xp + yy and L = up + ivy Then. m fiz) = L, if and only if Him wt») = uy and tim vx, y) = vp, mY sony Proof Let S(z) = L. Then, | f(2) - L| < e, whenever O<|z-21<65 or 0) +7 ~ ¥)| <8, we have 9 < [ee x0) + iY — yo < 8. Hence, for 0 < |(x lux, ¥) + Cx, ¥) ~ (uy + iv, ol 0, there exists The function s2) has a limit Las = 9 6, if for any arbitrary sma areal number 3 > 0 such that lz] > 1/8. [Ae) = L| < whenever Alternately, we substitute z= VE Since § > 0 as z > ©», we obtain dim f@) = i F (ve). Fxample 10.21 Using the definition of limits, show that (i) lim ( - 2) Gi) Bx + iv") = 4i. Solution We want to determine a real number 5 > 0 such that for a given real number € > 0 \f@)-L| 0, we need to determine 5> 0, such that . ions 10.33 Functions of a Complex Variable:Analytie Functions 96+ V5) 0, we want to determine a real number 6 > 0 such that 1 Ife) L1 <6 whenever |z| > =. Therefore, we have <& whenever [z| > Zz Now, |1/z2| < € implies || > 1/Vé . Thus, we can take < Ve. With this choice of & we find that 1 1 1 | 1) g whenever |z| > 3 md sin{ 4] a sole Alternative Substituting z = W/E, we get /(1/8) = 6 Since 6 > 0 as z + =, we need to determine a real number 6 > 0 such that for a given real number € > 0, we have [f(/é) — 01 < & whenever 0< 1g1<6 Now, ipie) - 01= |B <& oF I< Ve. Thus, we can take 5< Vé. With this choice of 6, we find that 10.34 Engineering Mathematics [|< whenever 0<|§| <5 Therefore, fin? = 0 or lim [V2}=0. Example 10.23 Find the following limits ii) im (Bx +iy?], (iii) f= 0 tn f= Solution -)(z+) z-1 =lim(@+)=2. (ii) 2 = x + iy = 23. Therefore, x = 0, y = 2. lim Bx + 7] = lim Bx + y?] = 47. zu 0092 = lim ve = lim e50|2— ie |” go] 2E [EH Fx tin LEH 2 +H] Vz-t+yz-i (ii) im [ . ~iyé 7 vlogs Trellis Example 10.24 Show that the following limits do not exist. 2 © lim = (i) tim (Rez=tmep se - 1 2 in| ae + | Solution Here, we shall show that as z—> 0 alon i , s 2 two differe of ifferent lints and therefore the limit does not exist. . nipats we obtain two dierent i (i) When we choose the path y —> 0 followed by x > 0, we get ions 10.35 ; ions Functions of a Complex Variable:Analyic Funct lim 290 = lim Le = lim 2 =41- a ee x40 |x| jim in x90] 90) Hence, the limit does not exist. (i) Consider any straight line path y = mx (for different values of m we get different paths). Hence, __ [Rez—Imz} -y? _m? xt _(I-m) tim BRO2= HME] ig, =D? jy Came? em) m0 iP Poe y? (em) Lem? Since the limit depends on m and is not unique, the limit does not exist. (iii) Consider the path y > 0 followed by x > 0. We get ' a [ere Be] sn[ teen oa Since the limit is not unique, the limit does not exist. lo, x>0, 1, x<0. 10.5.2 Continuity of a Function Lat fle) bea single-valued function of z defined in some neighborhood of the point zp including the point z,, Then, f(2) is said to be continuous at the point zp if fora given real number € > 0, we can find a real number 6 > 0, such that le) —fe)| < & wherever [2 ~ 21 < 5 (10.29) Thus, a function f(z) is continuous at 29 if the following three conditions are satisfied: i) f(z) exists, (i) fim 1G) exists, and (iii) lim fle) = f@0)- A function f(z) which is not continuous at Z is said to be discontinuous at zp. A function f(z) is said to be continuous in a domain D if it is continuous at every point in D. We note the following:

You might also like