SCHAUM’S oullinés et a The perfect aid for better grades! Covers the complete course—basic theary, problems, and more é40 problems solved step by step Ideal for independent study! i Setants Aernicaa Leper Lamghes inate Ue aid GREE GPUS. |e teal beeerre ot Gongs if deglivitvenr. a! Wary rainOEP 199] . . wi + 3) S68 72 SE. ni SCHAUM’S OUTLINE OF THEORY AND PROBLEMS COMPLEX VARIABLES with an introduction. to CONFORMAL MAPPING and its application SI (METRIC) EDITION = | x Pg md - Seg Cy = a = = mej > o SCHAUM’S OUTLINE SERIES McGraw- Hill Book” Company Singapore —Schaum’s outline of theory and problems of COMPLEX VARIABLES SI (METRIC) EDITION Copyright © 1981 Exclusive rights by McGraw-Hill Book Co., Singapore for manufacture and export. This book cannot be re-exported from the country to which it is consigned by McGraw-Hill. PSeTAO- \G06| 6789KKP 89432109 Copyright © 1964, 1974 by McGraw-Hill, Inc. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or alactr A : transmitted, in any form or by any mcans ciectroni mechanical, photocopying, recording, or otherwise, c My Printed in SirigaporePreface The theory of functions of a complex variable, also calied for brevity complex variables or complex analysis, is one of the most beautiful as well as useful branches of mathematics. Although originating in an atmosphere of mystery, suspicion and distrust, as evidenced by the terms “imaginary” and “complex” present in the literature, it was finally placed on a sound foundation in the 19th century through the efforts of Cauchy, Riemann, Weierstrass, Gauss and other great mathematicians. Today the subject is recognized as an essential part of the mathematical background of engineers, physicists, mathematicians and other scientists. From the theoretical view- point this is because many mathematical concepts become clarified and unified when examined in the light of complex variable theory. From the applied viewpoint the theory is of tremendous value in the solution of problems of heat flow, potential theory, fluid mechanics, electromagnetic theory, aerodynamics, elasticity and many other fields of science and engineering. This book is designed for use as a supplement to all current standard texts or as a textbook for a formal course in complex variable theory and applications. It should also be of considerable value to those taking courses in mathematics, physies, aerodynamics, elas- ticity or any of the numerous other fields in which complex variable methods are employed. Each chapter begins with a clear statement of pertinent definitions, principles and theorems together with illustrative and other descriptive material. This is followed by graded sets of solved and supplementary problems. The solved problems serve to illustrate and amplify the theory, bring into sharp focus those fine points without which the student continually feels himself on unsafe ground, and provide the repetition of basic principles so vital to effective learning. Numerous proofs of theorems and derivations of formulae are included among the solved problems. The large number of supplementary problems with answers serve as a complete review of the material in each chapter. Topics covered include the algebra and geometry of complex numbers, complex differ- ential and integral calculus, infinite series including Taylor and Laurent series, the theory of residues with applications to the evaluation of integrals and series, and conformal mapping with applications drawn from various fields. An added feature is the chapter fall AMIS L2H0UNiS 25 Ulh on special topics which should prove useful as an introduction to some more advanced topics. Considerably more material has been included here than can be covered in most first courses. This has been done to make the book more flexible, to provide a more useful book of reference and to stimulate further interest in the topics. I wish to take this opportunity to thank the staff of the Schaum Publishing Company for their splendid cooperation. M. R. SPIEGEL Rensselaer Polytechnic Institute July, 1964CONTENTS Chapter I COMPLEX NUMBERS .................00.000c0cceceeeee scenes 1 The real number system. Graphical representation of real numbers. The com- plex number system. Fundamental operations with complex numbers. Absolute value. Axiomatic foundations of the complex number system. Graphical repre- sentation of complex numbers. Polar form of complex numbers. De Moivre’s theorem. Roots of complex numbers. Euler’s formula. Polynomial equations. The nth roots of unity. Vector interpretation of complex numbers. Spherical representation of complex numbers. Stereographic projection. Dot and cross product. Complex conjugate coordinates. Point sets. Chapter 2 FUNCTIONS, LIMITS AND CONTINUITY ................... 33 Variables and functions. Single-and multiple-valued functions. Inverse func- tions. Transformations. Curvilinear coordinates. The elementary functions. Branch points and branch lines. Riemann surfaces. Limits. Theorems on limits. Infinity. Continuity. Continuity in a region. Theorems on continuity. Uniform continuity. Sequences. Limit of a sequence. Theorems on limits of sequences, Infinite series. Chapter 3 | COMPLEX DIFFERENTIATION AND THE CAUCHY-RIEMANN EQUATIONS ...................... 63 Derivatives. Analytic functions. Cauchy-Riemann equations. Harmonic func- tions. Geometric interpretation of the derivative. Differentials. Rules for differentiation. Derivatives of elementary functions. Higher order derivatives. L’Hospital’s rule. Singular points. Orthogonal families. Curves. Applications to geometry and mechanics. Complex differential operators. Gradient, diver- gence, curl and Laplacian. Some identities involving gradient, divergence and curl. : Chapter 4 COMPLEX INTEGRATION AND CAUCHY’S THEOREM ..... 92 Complex line integrals. Real line integrals. Connection between real and com- plex line integrals. Properties of integrals. Change of variables. Simply-and multipiy-connected regions. Jordan curve theorem. Convention regarding traversal of a closed path. Green’s theorem in the plane. Complex form of Green’s theorem. Cauchy’s theorem. The Cauchy-Goursat theorem. Morera’s theorem. Indefinite integrals. Integrals of special functions. Some conse- quences of Cauchy’s theorem: Chapter § | CAUCHY’S INTEGRAL FORMULAE AND RELATED THEOREMS Cauchy's integral formulae. Some important theorems. Morera’s theorem. Cauchy’s inequality. Liouville’s theorem. Fundamental theorem of algebra. Gauss’ mean value theorem. Maximum modulus theorem. Minimum modulus theorem. The argument theorem. Rouché’s theorem. Poisson’s integral for- mulae for a circle. Poisson’s integral formulae for a half plane.Chapter 6 CONTENTS INFINITE SERIES. TAYLOR’S AND LAURENT SERIES.... Sequences of functions. Series of functions. Absolute convergence. Uniform convergence of sequences and series. Power series. Some important theorems. General theorems. Theorems on absolute convergence. Special tests for con- vergence. Theorems on uniform convergence. Theorems on power series. Taylor's theorem. Some special series. Laurent’s theorem, Classification of singularities. Entire functions. Meromorphic functions. Lagrange’s expan- sion. Analytic continuation. Page 139 ae Chapter 7 THE RESIDUE THEOREM. EVALUATION OF INTEGRALS AND SERIES ............... Residues. Calculation of residues. The residue theorem. Evaluation of definite integrals. Special theorems used in evaluating integrals. The Cauchy principal value of integrals. Differentiation under the integral sign. Leibnitz’s rule. Summation of series. Mittag-Leffler’s expansion theorem. Some special expansions. a Chapter 8 CONFORMAL MAPPING ...........0 cc cece tent e eet e ene eens Transformations or mappings. Jacobian of a transformation. Complex map- ping functions. Conformal mapping. Riemann’s mapping theorem. Fixed or invariant points of a transformation. Some general transformations. Trans- lation. Rotation. Stretching. Inversion. Successive transformations. The linear transformation. The bilinear or fractional transformation. Mapping of a half plane on to a circle. The Schwarz-Christoffel transformation. Trans- formations of boundaries in parametric form. Some special mappings. ee Chapter 9 PHYSICAL APPLICATIONS OF CONFORMAL MAPPING.... Boundary-value problems. Harmonic and conjugate functions, Dirichlet and Neumann problems. The Dirichlet problem for the unit circle. Poisson’s for- mula. The Dirichlet problem for the half plane. Solutions to Dirichlet and Neumann problems by conformal mapping. Applications te fluid flow. Basie assumptions. The complex potential. Equipotential lines and streamlines. Sources and sinks, Some special flows. Flow around obstacles. Bernoulli’s theorem. Theorems of Blasius. Applications to electrostatics. Coulomb’s law. Electric field intensity. Electrostatic potential. Gauss’ theorem. The complex electrostatic potential. Line charges. Conductors. Capacitance. Applications to heat flow. Heat flux. The complex temperature. 232 nn Chapter 1 0 SPECIAL TOPICS .......... 00. ccc cece eee eee e nen eens Analytic continuation. Schwarz’s reflection principle. Infinite products. Abso- lute, conditional and uniform convergence of infinite products. Some important theorems on infinite products. Weierstrass’ theorem for infinite . products. Some special infinite products. The gamma function. Properties of the gamma function. The beta function. Differential equations. Solution of differential equations by contour integrals. Bessel functions. Legendre functions. The hynergeometrie function. The zeta function. Asymptotic series. The method of steepest descents. Special asymptotic expansions. Elliptic functions.Chapter 1 THE REAL NUMBER SYSTEM The number system as we know it today is a result of gradual development. as indicate in the following list. fu i 1, Natural numbers 1,2,3,4,..., also called positive integers, were first used i counting. The symbols varied with the times, e.g. the Romans used I, II, III, IV, .... If a and 6 are natural numbers, the sum a+b and product a-b, (a)(b) or ab are also natural numbers. For this reason the set of natural numbers jis said to be closed under the operations of addition and multiplication or to satisfy the closure property with respect to these operations. 2. Negative integers and zero, denoted by —1,—2,—3,... and 0 respectively, arose to permit solutions of equations such as ~+b =a where a and b are any natural numbers. This leads to the operation of subtraction, or inverse of addition, and we write x = a—b. The set of positive and negative integers and zero is called the set of integers and is closed under the operations of addition, multiplication and subtraction. 3. Rational numbers or fractions such as 3,—%,... arose to permit solutions of equations such as bz=b or b0, to —a if a<0 and to 0ifa=0. The distance between two points a and b on the real axis is |a— 6]. THE COMPLEX NUMBER SYSTEM There is no real number x which satisfies the polynomial equation 22+1 = 0. To permit solutions of this and similar equations, the set of complex numbers is introduced. We can consider a complex number as having the form a+ bi where a and b are real numbers and i, which is called the imaginary unit, has the property that V=-], If z= a+bi, then a is called the real part of z and b is called the imaginary part of z and are denoted by Re {z} and Im {z} respectively. The symbol 2, which can stand for any of a set of complex numbers, is called a complex variable. Two complex numbers a + bi and ¢+di are equal if and only if a=c and b=d. We can consider real numbers as a subset of the set of complex numbers with b=0. Thus the complex numbers 0 + 0i and —3 + 0i represent the real numbers 0 and —3 respectively. If a=0, the complex number 0+ bi or bi is called a pure imaginary number. The complex conjugate, or briefly conjugate, of a complex number at+bi is a— bi. The complex conjugate of a complex number z is often indicated by 2 or z*. FUNDAMENTAL OPERATIONS WITH COMPLEX NUMBERS In performing operations with complex numbers we can proceed as in the algebra of real numbers, replacing 1? by —1 when it occurs. 1. Addition oo _. .. . (a+ bi) + (e+di) = a+ bit+e+di = (ate) + (B+a)i 2. Subtraction (a + bi) — (e+ di) £ 7 t} a+bi—c—di = (a—c)+(b—d)i 8. Multiplication (a+ bi)(e+di) = ac+adi+t bei + bdi? = (ac— bd) + (ad + be)i 4. Division at+bi _ atbi c-di _ ac—adi+bei—bdi? c+di ~ ct+di c-di ~ ea? _ aet+bd+(be—adji _ actbd + be—ad , ~ of + a ~ 4+ @ +a The absolute value or modulus of a complex number a+ bi is defined as |a+bi| = Va? + b* Example: |—4+ 2i| = V(—4)? + (2)? = ¥20 = 2V5 2a, .--,2m are complex numbers, the following properties hold. leo] or fan ze s+ 2m] = fen feel ++ em 1 2 8. jzitze| = fel + leo] or = fazrtzet-+++2m| S [es] + [eo] + +--+ [em 4 \2itz2| 2 lail—|zo] or |21—22| 2 fes| — leaCHAP. 1] COMPLEX NUMBERS 3 AXIOMATIC FOUNDATIONS OF THE COMPLEX NUMBER SYSTEM From a strictly logical point of view it is desirable to define a complex number as an ordered pair (a,b) of real numbers a and b subject to certain operational definitions which turn out to be equivalent to those above. These definitions are as follows, where all letters represent real numbers. A. Equality (a,b) = (¢,d) if and only if a=c, b=d B. Sum (a,b) + (c,d) (a+e, b+d) C. Product (a,b)-(c,d) = (ac—bd, ad+be) m(a,b) = (ma, mb) From these we can show [Problem 14] that (a,b) = a(1,0)+ (0,1) and we associate this with a + bi where i is the symbol for (0,1) and has the property that 4? = (0,1)(0, 1) = (-1,0) [which can be considered equivalent to the real number —1] and (1,0) can be con- sidered equivalent to the real number 1. The ordered pair (0,0) corresponds to the real number 0. From the above we can prove that if 21, 22,23 belong to the set S of complex numbers, then 1. 2:+22 and 2:22 belong to S Closure law 2. t@tee = te+% Commutative law of addition 3. 21+ (22+ 2%) = (@1 +22) + 2s Associative law of addition 4. 2122 = 221 Commutative law of multiplication 5. 21(Z223) = (212%2)2s Associative law of multiplication ‘ 6. 21(%2 +23) = 2122 + 2123 Distributive law 7. 0 is called the identity with respect to at+O0=-O0+a =a, Lea = &:1 = %4, + addition, 1 is called the identity rth aman CaueG wie wv hk respect to multiplication. 8. For any complex number 2: there is a unique number z in S such that z+z, = 0; z is called the inverse of z: with respect to addition and is denoted by —2;. 9. For any z:~0 there is a unique number z in S such that 2:2 = 22; =1; 2 is called the inverse of 2: with respect to multiplication and is denoted by zz! or 1/z:. In general any set, such as S, whose members satisfy the above is called a field. GRAPHICAL REPRESENTATION OF COMPLEX NUMBERS If real scales are chosen on two mutually perpendicular axes X’OX and Y’OY [called the x and y axes respectively] as in Fig. 1-2, we can locate any point, in the plane de- termined by these lines, by the ordered pair of real numbers (x,y) called rectangular coordinates of the point. Examples of the location of such points are indicated by P, Q, R,S and T in Fig. 1-2. an Yy . Since a complex number x+iy can be *P(8, 4) considered as an ordered pair of real numbers, ”Q(-8, 3) 8 we can represent such numbers by points in 2 an xy plane called the complex plane or Argand diagram. The complex number represented 1 by P, for example, could then be read as either (8,4) or 3+4i. To each complex num- . - mendes mene ota ber there corresponds one and only one point X’ 4-3 x in the plane, and conversely to each point in R(—2.5,-1.5) ~2 . the plane there corresponds one and only one S(2, —2) complex number. Because of this we often y" refer to the complex number z as the point z. Fig.1-24 COMPLEX NUMBERS [CHAP. 1 Sometimes we refer to the x and y axes as the real and imaginary axes respectively and to the complex plane as the z plane. The distance between two points 2: = 2itiy: and Zz. = %2+iy2 in the complex plane is given by |z1— 2o| = V (a1 — £2)? + (yr — Y2)?. POLAR FORM OF COMPLEX NUMBERS If P is a point in the complex plane corre- sponding to the complex number (z, y) or z+ ty, then we see from Fig. 1-3 that xz=reosé, y=rsing where r = V2? +y? = |a+ty| is called the modulus or absolute value of z= x+iy {de- noted by mod z or |z|]; and 6, called the ampli- tude or argument of z = x+iy [denoted by arg z], is the angle which line OP makes with the positive x axis. It follows that z2=2+iy = r(cosé +isin 6) (1) which is called the polar form of the complex number, and r and 6 are called polar coordi- nates. It is sometimes convenient to write the Fig. 1-8 abbreviation cis 9 for cos@+isiné. y’ ar ean) For any complex number z~0 there corresponds only one value of @ in 020<2z. However, any other interval of length 27, for example —7<6@=7, can be used. Any particular choice, decided upon in advance, is called the principal range, and the value of ¢@ is called its principal value. DE MOIVRE’S THEOREM If 2: = aitty: = 71 (cosditising:) and 2% = %2+ty2 = 12 (cos 62+isin 62), we can show that [see Problem 19} 2122 = Tite {c08 (61+ 62) + isin (A + 42)} (2) 2 Tr soe a = 4 {cos{@1 — 92) + 7 sin (1 — 42)} (8) A generalization of (2) leads to Zi€e°* Sn = Miers tn {COS(G1+ 42+ +++ +n) + isin (Ait 62+ °°: + 6n)} (4) and if 2:=22=-°::=2,=2 this becomes 2” = {r(cosé +isiné)}* = r™(cosné +7 sinn#) (5) which is often called De Moivre’s theorem. ROOTS OF COMPLEX NUMBERS A number w is called an nth root of a complex number z if w*=z, and we write tn Wa terete thaneawr. w=zZ From De Moivre’s theorem we can show that if 2 is a positive integer, zn = {r(cosé + isin 6)}' ( favor /o+o9bn,\) OT ann os cr = rim Jdeog { = \ + isin | ——— k = 0,1,2,...,n-1 (6) n n from which it follows that there are » different values for 2'", ie. 1 different nth roots of z, provided z~0.CHAP, 1] COMPLEX NUMBERS 5 EULER’S FORMULA By assuming that the infinite series expansion e* = 1+2+22/2!+23/83!4+--. of elementary calculus holds when «=i9, we can arrive at the result e® = cosé+ising e = 2.71828... (7) which is called Fuler’s formula. It is more convenient, however, simply to take (7) as a definition of e®. In general, we define e = e*¥ = ere = e%(cosy +isiny) (8) | In the special case where y=0 this reduces to e*. Note that in terms of (7) De Moivre’s theorem essentially reduces to (e)" = eine, POLYNOMIAL EQUATIONS Often in practice we require solutions of polynomial equations having the form Goa" + Ae"! + az? + es + an-iz+an = 0 - (9) where do #0,41,...,@, are given complex numbers and n is a positive integer called the degree of the equation. Such solutions are also called zeros of the polynomial on the left of (9) or roots of the equation. A very important theorem called the fundamental theorem of algebra [to be proved in Chapter 5] states that every polynomial equation of the form (9) has at least one root which is complex. From this we can show that it has in fact n complex roots, some or all of which may be identical. If zZ1,22,..+,2n are the n roots, (9) can be written Go(% ~ 21)(2 — 22) +++ (2—2n) = which is called the factored form of the polynomial e (9) in the form (10), we can easily determine the roots. OQ (10) mf oe nat: we can write THE nth ROOTS OF UNITY The solutions of the equation z*=1 where n is a positive integer are called the nth roots of unity and are given by . 2 = cos2kr/n + isin2kr/n = etkrin k=0,1,2,...,n—-1 (11) If we let w = cos2r/n+isin2x/n = e/™ the n roots are 1,0, 0,...,0"7, Geo- metrically they represent the 7 vertices of a regular polygon of n sides inscribed in a circle of radius one with centre at the origin. This circle has the equation |z|=1 and is often called the wnit circle. VECTOR INTERPRETATION OF COMPLEX NUMBERS A complex number z = x+y can be con- y sidered as a vector OP whose initial point is the | origin O and whose terminal point P is the | eo point (z,y) as in Fig. 1-4. We sometimes call A _ OP =x+iy the position vector of P. Two vec- OW) tors having the same length or magnitude and direction but different initial points, such as 0 * OP and AB in Fig. 1-4, are considered equal. Hence we write OP = AB = x +iy. Fig. 1-4 ~6 COMPLEX NUMBERS [CHAP. 1 Addition of complex numbers corresponds to the parallelogram law for addition of vectors [see Fig. 1-5]. Thus to add the complex num- bers 2, and 22, we complete the parallelogram OABC whose sides OA and OC correspond to z, and Zz. The diagonal OB of this parallelo- gram corresponds to 21+. See Problem 5. SPHERICAL REPRESENTATION OF COMPLEX NUMBERS. STEREOGRAPHIC PROJECTION Let P [Fig. 1-6] be the complex plane and consider a unit sphere Jf [radius one] tangent to P atz=0. The diameter NS is perpendicular to P and we call points N and S the north and south poles of J- Corresponding to any ‘ point A on P we can construct line NA inter- secting Jf at point A’. Thus to each point of the complex plane ? there corresponds one and only one point of the sphere J, and we can represent any complex number by a point on the sphere. For completeness we say that the point N itself corresponds to the “point at infinity” of the plane. The set of all points of the complex plane including the point at- infinity is called the entire complex plane, the entire z plane, or the extended complex plane. Fig. 1-6 The above method for mapping the plane on to the sphere is called stereographic projection. The sphere is sometimes called the Riemann sphere. DOT AND CROSS PRODUCT Let 2: = mitiy: and z: = x2+iy2 be two complex numbers [vectors]. The dot product [also called the scalar product) of 21 and 22 is defined by z02%2 = |zi||zo|cos@ = tite + yry2 = Re (Zite} = 4{%1d2 + 21:22} (12) where @ is the angle between z, and 22 which lies between 0 and 7. The cross product of 2, and 22 is defined by 1 21X22 = |ail|4o| sin = Liy2— Yee = Im {2122} = py (212 — e142) (13) Clearly, Zz2 = (a1 ° 22) + (4% x 22) = || |zal e# (14) ” Tf 2: and z, are non-zero, then 1. A necessary and sufficient condition that 2: and 2e be perpendicular is that 41922 = 0. z. A necessary and sufficient condition that 2: and 22 be parallel is that z:Xz2 = 0. 3. The magnitude of the projection of z: on 22 is |#1 © 2|/|22]. 4. The area of a parallelogram having sides z: and 22 is |é x 2a].CHAP. 1] COMPLEX NUMBERS 7 COMPLEX CONJUGATE COORDINATES A point in the complex plane can be located by rectangular coordinates (2, y) or polar coordinates (r,@). Many other possibilities exist. One such possibility uses the fact that x= }(z+2),, y= 3-2) where z = x+iy. The coordinates (z, 2) which locate a point are called complex conjugate coordinates or ‘briefly conjugate coordinates of the point [see Problems 43 and 44). POINT SETS Any collection of points in the complex plane is called a (two-dimensional) point set, and each noint is called a momher ae clemon and each point is caued a member or element of the set. The following fundamenial definitions are given here for reference. 1, Neighbourhoods. A delta, or 5, neighbourhood of a point zo is the set of all points z such that |z—2z| < 8 where 8 is any given positive number. A deleted 8 neigh- bourhood of 2 is a neighbourhood of zo in which the point zo is omitted, i.e. 0 < |z—- 20] < 8. 2. Limit Points. A point zo is called a limit point, cluster point, or point of accumu- lation of a point set S if every deleted 6 neighbourhood of zp contains points of S. Since § can be any positive number, it follows that S must have infinitely many points. Note that z may or may not belong to the set S. 3. Closed Sets. A set S is said to be closed if every limit point of S belongs to S, i.e. if S contains all its limit points. For example, the set of all points z such that |z|=1 is a closed set. 4, Bounded Sets. A set S is called bounded if we can find a constant M such that |2| 1-7 ~ 1-7 2 2 2 Another method. By definition, (3 — 2i)/(-1+4) is that number a+ bi, where a and } are real, such that (—1+i)(a+ bi) = -a—b+(a—bji = 3—-2i. Then —a—6b = 3, a—b = —2 and solving simultaneously, a = —5/2, 6 = —1/2 or a+bi = —5/2 — 1/2, 5 + 5 20 5 + hi 8 +h 20 4-23 O sa tare > s-a ora + Tem ana o —~ dr 15 + 20: + 15i+ 201? 80-601 _ 5+ 35i 80 — 607 3-i 9 — 1672 16—912 ~ 25 25 (on m) 3419 _ 3(é2)15 — (42994 L 3(—1)15 — (—1)% 2i-1 ,» a—-1 —-1+2: = -84i -1-2i _ stei—i-a? _ B45 _ L4i —1+2i -1-2i — 1— 472 ~ 5 If a =2+1, @=38-2i and z= -5 + vB, evaluate each of the following. (a) |82y~42.| = |8(2+4) — 4(8—-2i)| = [6+ 31-1248: | |-6+11i| = V(—6?+ (i? = Vi57 - (b) 28 — 822+ 42, -8 = (2+4)8 — 3(2+12 + 4(2+0 —8 1 1 = {(2)8 + 8(2)2(@) + 8(2)(a)? + 18} — 844+ 484+ 42) +84 4-8 = 8+12i-6-—i-12-1214+34+8+41-8 = -74+3i ———\ 4 [ 1 va \ yy. V2.4 r/ 1. fa 272 (ec) (&)* = \-2 2 ¢) = (~27 2 ¢) = (oz 7st) | - 1, ¥8,,38.7 _ (1, v3,\? _ 1 v8.,8, 1 VB. ~ [4 Tet Tah | = ete tt) = gr ett = -g--y in |2eta-5-if _— | 28-2)+@+)-5-E/? |Re=a_ts—i| ~ | 2@+)-@-29+3-4| _ {3-4 | 3 — 44 |2 — V@P?+-4?) 443i |4+ 3:2 (V4)? + (8)2 2 Find real numbers x and y such that Bu + Qty —in+5y = 7+ 5i. The given equation can be written as 3a + by +i(2y—a) = 7+65i. Then equating real and imaginary parts, 382+ By = 7, 2y—" = 5. Solving simultaneously, « — —1, ¥=2 Similvaneous: a4) Fre Prove: (a) #1 +22 = %: +22, (b) |zr2o] = |za| [eo]. Let 2, = 2, +%y,, 2 = ao+iyo. Then (@) 2yt% = a + ivy t+ ag + tyg = 2+ a+ i(yy+ Yo) = Bt tg — yy tye) = 2 — ty + ty — yg = B+ iy, + 22+ yg = 2, + % (6) zyzgl = | (ey + ty,)@et tye)| = | x20 ~ Vive + tarY2 + 1%) | = V(ey%_— Vy¥e)? + (eve + ita)? = V(ai+ y2)(ae+ y2) = Vad + y? Ve2 + y2 = |e, |z0l Another method. ler@ol® = (zy%q) (21%) = 2yez98pZo = (2 121)(zo%o) = fey[2 feel? or |zyzql = {ayl [zg where we have used the fact that the conjugate of a product of two complex numbers is equal to the product of their conjugates (see Problem 55).10 COMPLEX NUMBERS [CHAP. 1 GRAPHICAL REPRESENTATION OF COMPLEX NUMBERS. VECTORS 5. Perform the indicated operations both analytically and graphically: (a) (8 + 4%) + (5 +21), (b) (6—2i) — (2—5i), (c) (-8 +54) + (442i) + (5-84) + (—4 — 6i). (a) Analytically. (3+ 41) + (6+2) = 8+5+ 41+ 21 = 8+ 6: Graphically. Represent the two complex numbers by points P, and P, respectively as in Fig. 1-7 below. Complete the parallelogram with OP, and OP, as adjacent sides. Point P represents the sum, 8 + 6i, of the two given complex numbers. Note the similarity with the parallelogram law for addition of vectors OP, and OP, to obtain vector OP. For this reason it is often convenient to consider a complex number a+ bi as a vector having components a and b in the directions of the positive x and y axes respectively. Fig. 1-7 Fig. 1-8 (6) Analytically. (6—2i) —(2-5/) = 6-2-—21+ 51 = 4+ 3 : Graphically. (6—2i) —(2—5i) = 6—2i+(-2+5i). We now add 6-2: and (-2+57) as in part (a), The result is indicated by OP in Fig. 1-8 ubove. (c) Analytically. (-3 + 8) + 442) + G—3) + (4-6) = (-34445—4) + (51+ 21-31-61) = 2-2 Graphically. Represent the numbers to be added by 24, 22, 2 24 respectively. These are shown graphically in Fig. 1-9. To find the required sum proceed as shown in Fig. 1-10. At the terminal point of vector z, construct vector zp. At the terminal point of z, construct vector 25, and at the terminal point of z, construct vector z, The required sum, sometimes called the resultant, is obtained by constructing the vector OP from the initial point of 2; to the terminal point of 24, ie. OP = a, +a+agt aq = 2-24. y \ zy Ji ty Fig. 1-9 Fig. 1-10CHAP, 1] COMPLEX NUMBERS -_ = 6. If z, and 2: are two given complex numbers (vec- tors) as in Fig. 1-11, construct graphically (a) 32: — 222 (b) 422 + $21 (a) In Fig. 1-12 below, OA = 82, is a vector having length 4 8 times vecter z; and the same direction. OB = —2z, is a vector having length 2 times vector 22 2 and the opposite direction. Then vector OC = OA+ OB = 82; — 22. Fig. 1-11 Cc |” Fig. 1-12 Fig. 1-13 (b) The required vector (complex number) is represented by OP in Fig. 1-13 above. 7 Prove (a) |g: +22| = |a:| + fea|, (b) |e: + 22+ 2s| S fei] + zo] + es), (€) [zr — ze] 2 |z1| — |22] and give a graphical interpretation. (a) Analytically. Let z, = 2, + iy,, 22 = %_+ tyo, Then we must show that V (xy + ag)? + (yt)? = Yarty? + Vad +92 Zz Squaring both sides, this will be true if (ay + ag)? + (yy + ye)? S a2 + y2 + OVP yk +) + oF + ie. if aye, + yy, = Vat + ¥8)(x8 + 9) or if (squaring both sides again) wag + Qeywoyiye + yivg S stad a2 or Qe iesyit, S 2? 9 2 1 But this is equivalent to (x,y2—x,y,)2 2 0 which is true. Reversing the steps, which are reversible, proves the result. Graphically... The result follows graphically from the fact that {z,|, |zol, |z, +22] represent the lengths of the sides of a triangle (see Fig. 1-14) and that the sum of the lengths of two sides of a triangle is greater than or equal to the length of the third side. Or " Ry Se as \s1 + Ze +43 en ee ee ee Fig. 1-14 Fig. 1-15 (b) Analytically. By part (a), [ay+agteg] = fay t (egtas)] S leq] + [zgtzs| S leal + [zal + legl Graphically. The result is a consequence of the geometric fact that in a plane a straight line is the shortest distance between two points O and P (see Fig. 1-15). \12 8. 9. 11. COMPLEX NUMBERS (CHAP. 1 (c) Analytically, By part (a), |2y] = |21—22+22| S |2,—2e| + lal. Then [2-2] 2 lea] — lee. An equivalent result obtained on replacing 22 by ~2, is |2,+2%2| = || — |2ol. Graphically. The result is equivalent to the statement that a side of a triangle has length greater than or equal to the difference in lengths of the other two sides. Let the position vectors of points A(21,y:) and y B(«2, y2) be represented by 2: and 2: respec- tively. (a) Represent the vector AB as a com- teh Ln teee nn piex number. (0) Find the distance between Ae, ¥) points A and B. (a) From Fig. 1-16, OA + AB = OB or , : a AB = OB-0A = 4-% 0 (aq + ty) — (01 + ivy) = (%_— 2%) + i{ve— 41) Fig. 1-16 (b) The distance between points A and B is given by |AB| = | (#,- ty) + ya — wi) | = V(t2— 2)? + 2-01)? Il Let 2, = a,+iy: and 2: = %:+iye represent two non-collinear or non-parallel vectors. If a and b are real numbers (scalars) such that az, + bz. = 0, prove that a=0 and b=0. The given condition az, + bz, = 0 is equivalent to a(x, + ty) + b(zg+ iyo) = 0 or ax, + bagt+ i(ay, + bya) = = 0. Then ay + bi, = = 0 and ay, + by. 0. These equations have the simultaneous solution a= 0, 5=6 if ¥,/%1 7 wo/ao, ie. if the vectors are non-collincar or non-parallel vectors. Prove that the diagonais of a parallelogram bisect each other. Let GABC [Fig. 1-17] be the given parallelogram with diagonals intersecting at P. Since 2, + AC = 2,, AC = zg—2;, Then AP = m(Z_—%;) where OS m2=1. Since OB = 2, + %, OP = n(z,+2,) where’ O3n31. , But OA+AP = OP, ie. 2, + m(%q—%,) = n(zy +2) or (l—m—n)z,+ (m—n)z, = 0. Hence by Problem 9, 1—m—n = 0, m—n =0 or m=}, n=4 and so P is the midpoint of both diagonals. Fig. 1-17 Find an equation for the straight line which passes through two given points A(%, y1) and B{zxz, yz). Let 2; = 2, +t, and z = a2 + iy, be the posi- tion vectors of A and B respectively. Let z = «+ iy be the position vector of any point P on the line join- ing A and B. From Fig. 1-18, OA+ AP = OP or 4+ AP =2, ie. AP=2z-—2 OAS ARP = OF or 2, +t AB=2 ie, ARP — OA+ AB = OB or 2,+ AB =z, ic, AB = 2-2, @ Since AP and AB are collinear, AP = tAB or 0 2-2, = t(z29—2,) where ¢ is real, and the required equation is z= 2, + t(z,—2%,) or z= (1—2#)2, + te, Fig. 1-18CHAP. 1) COMPLEX NUMBERS ee eo Using z, = 2, + iy, 2 = 22+ iy, and z= 2+ iy, this can be written e-% yn ty 2, 7 Yew The first two are called parametric equations of the line and t is the parameter; the second is called the equation of the line in standard form. w— 2%, = tt,—2), y—y, = t(ye—y,) or Another method. Since AP and PB are collinear, we have for real numbers m and 7: mAP = nPB or mz — 21) = n(z_— 2) Salvi _ Mx, + Nay _ _ mat, + nx, _ My, + NY solving, ~ m+n “ men? Ym en which is called the symmetric form. 12 Let A(1,—2), B(—8, 4), C(2,2) be the three ver- tices of triangle ABC. Find the length of the median from C to the side AB. The position vectors of A,B and C are given by 4, = 1-21, z = ~8+4i and z3 = 2+2i respec- tively. Then from Fig. 1-19, AC = w-—% = 2+21-(1-2) = 144i BC = 2-2 = 24+ 2i-—(-34+4) = 5-2i AB = wm-% = -3 + 4i-— (1-2) = -4 + 6i Fig. 1-19 AD = }AB = 4(-4+ 6i) = -2+3i since D is the midpoint of AB. AC+OED= AD ot CD = AD-AC = -24+8i-(14+ 4 = -3-i. Then the length of median CD is |{CD| = |—3—i| = Vio. 13. Find an equation for (a) a circle of radius 4 with centre at (—2,1), (0) an ellipse with major axis of length 10 and foci at (—8,0) and (8, 8). (a) The centre can be represented by. the complex number —2+1%. If z is any point on the circle [Fig. 1-20], the distance from z to -2+7 is J2a~(-24+0| = 4 Then |z+2-—i| = 4 is the required equation. In rectangular form this is given by [\(w+2)+iy-1)| = 4, ie. (w@+2)2 + (y—1)2 = 16 y y Fig. 1-20 Fig. 1-21 (b) The sum of the distances from any point z on the ellipse (Fig. 1-21] to the foci must equal 10. Hence the required equation is jz+3j + j[z-3| = 10 In rectangular form this reduces to 22/25 + y2/16 = 1 (see Problem 74), AXIOMATIC FOUNDATIONS OF COMPLEX NUMBERS 14. Use the definition of a complex number as an ordered pair of real numbers and the definitions on Page 3 to prove that (a,b) = a(1, 0) + (0,1) where (0,1)(0,1) = (~1, 0).14 COMPLEX NUMBERS (CHAP. 1 From the definitions of sum and product on Page 8, we have (a,b) = (a,0) + (0,6) = a(1,0) + (0,1) where (0,1)(0,1) = (°0 —1+1, 0°14 1°0) = (—1,0) By identifying (1,0) with 1 and (0,1) with i, we see that (a,b) = a+ bi. 15. If 2: = (a1,b1), ze = (a2, b2) and zs = (as,bs), prove the distributive law: 21(%+%s) = 2122 + 212s. We have =. (zg -+23) = (44, by){(a2, ba) + (ag, b3)} = (ay, b1)(4g + ag, by + by) {a;(az + as) — b,(bg + bg), a,(bg + bg) + by(ag + ag)} (ag — bybg + Gyag — 016g, aybg + 634g + aybg + yas) (aydq— bybo, ayby + byag) + (a103 — 615g, abg + 0143) (a1, 6;)(@a, bg) + (ay, 5y)(ag, 53) = %4%_ + ees POLAR FORM OF COMPLEX NUMBERS 16. Express each of the following complex numbers in polar form. (a) 2+ 2V8i Modulus or absolute value, r = |2+2V3i| = V4+12 = 4. Amplitude or argument, ¢ = sin=! 2738/4 = sin-! 73/2 = 60° = 1/8 (radians). Then 2+ 273% ry(cos6 +isine) = 4{(cos 60° + 2 sin 60°) 4(cos 7/3 + 7 sin 7/3) The result can also be written as 4 cis 7/3 or, using Euler’s formula, as 4e74/8, (b) —5 + 5 2 = |-b+5i] = 25425 = bY! 6 = 180° — 45° = 185° = 37/4 (radians) Then —5 + 5i = 5V2 (cos 185° + isin 185°) = 5V2 cis 87/4 = BY2 edti4 (ce) -V¥6 — V2i r = |-Vé—V2i] = Ve+2 = 2V2 = 180° + 30° = 210° = 7n/6 (radians) Then *—J6— V2i = 2V2 (cos 210° + isin 210°) . = 2V2 cis 77/6 = 2V2 e7n/8 (d) —3i r = |-3] = [o—-3i| = Yo+9 = 3 @ = 270° = 37/2 (radians) Then —3i = 8(cos 37/2 + isin 37/2) a 8 cis 87/2 = S8e3ri/2 Fig. 1-25 17. Graph each of the following: (a) 6(cos240° +7sin 240°), (6) 4e°™, (c) 2e7™4, (a) 6(cos 240° + isin 240°) = 6 cis240° = 6cis 47/3 = 6 etti/8 can be represented graphicaily by OF in Fig. 1-26 beiow. If we start with vector OA, whose magnitude is 6 and whose direction is that of the positive axis, we can obtain OP by rotating OA counterclockwise through an angle of 240°. In general, rel® is equivalent to a vector obtained by rotating a vector of magnitude r and direction that of the positive # axis, counterclockwise through an angle 9.CHAP. 1] COMPLEX NUMBERS 18 Fig. 1-26 Fig. 1-27 Fig. 1-28 (b) 4 eti/5 = A(cos 87/5 + isin 87/5) = 4(cos 108° + i sin 108°) is represented by OP in Fig. 1-27 above. (c) Qe “V4 = {cos (—7/4) + isin (—7/4)} = 2{cos (—45°) + ¢ sin (—45°)} This complex number can be represented by vector OP in Fig. 1-28 above. This vector can be obtained by starting with vector OA, whose magnitude is 2 and whose direction is that of the positive x axis, and rotating it counterclockwise through an angle of —45° (which is the same as rotating it clockwise through an angle of 45°). B vy 18. A man travels 12 miles northeast, 20 miles 30° 2 west of north, and then 18 miles 60° south of west. Determine (a) analytically and (b) graphi- cally how far and in what direction he is from roe his starting point. . Cc (a) Analytically. Let O be the starting point (see Fig. A 1-29). Then the successive displacements are repre- Fig. 1-29 18 sented by vectors OA,AB and BC. The result of all 45° % three displacements is represented by the vector oO OC = OA+AB+ BC Now OA = 12(cos 45° + isin 45°) = 12 e7/4 4R = 20{cos {90° t 30° + isin (90° + 30°)} = 26 €@rt/2 BC = 18{cos (180° + 60°) + isin (180° + 60°)} = 18 efti/s Then OC = 12e/4 + 20 e2ti/3 4 18 Ariss {12 cos 45° + 20 cos 120° + 18 cos 240°} + i{12 sin 45° + 20 sin 120° + 18 sin 240°} {(12)(VB/2) + (20)(—1/2) + (18)(—1/2)} + é{(12)(8/2) + (20)(V3/2) + (18)(—-8/2) = (6V2— 19) + (672 + y3)i If r(cose +isine) = 6¥2—19 + (6V2+ V3), then r = a Gv2 — 19)? + 6v2 + V8)? = 14.7 approximately, and @ = cos7!(6V2—19)/r = cos—1(—.717) = 135°49’ approximately. Thus the man is 14.7 miles from his starting point in a direction 135°49’ — 90° = 45°49’ west of north. 1-29 which represents 2 miles BC. Then by determining t the axis, we obtain the approximate (6) Graphically. Using a convenient unit of length such as and a protractor to measure angles, construct vectors OA, AB number of units in OC and the angle which OC makes with the results of (a). PQ in PQ in Fie. -*5 and ey DE MOIVRE’S THEOREM 19. If 2: = ri(coséi+isiné:) and z = re(cosé2+isin6s), prove: (@) 122 = 1172 {cos (A: + 2) + isin (81 + 62)} z r at (b) z = rn {cos (91 — 62) + isin (4: — 62)}. (a) 2322 = {r,(cos 6, + i sin 6,)}{r(cos 92 + i sin 62)} 1y7o{(cos 6, cos 6 — sin @, sin @) + i(sin 6, cos 6, + cos 6, sin 6)} T1re{cos (64 + @») + isin (04 + 6o)}16 20. 21, 22. COMPLEX NUMBERS [CHAP. 1 ry(cos @, + isin 6,) (cos 6, — 4 sin 69) 4 _ (0) fy T2(cos 6g +4 8iN 62) (cos 6, — 1 sin 62) _ 71 {[ (cos 4; cos @g + sin 6; sin 62) + (sin 8, cos @, — cos 4, sin 4») Te cos® @, + sin? 6, = 4 {cos (6; — 62) + ¢ sin (6; — 62)} Yr In terms of Euler’s formula e = cos@ + isiné, the results state that if z,;=17,¢: and z, =r, els, 2 7, ei r. = 10; + 63) “1 _ i = 21 9-4), then 2% = ryr, e+ and zn = 7q0s re’ 1 8s Prove De Moivre’s theorem: (cos9+isin6)" = cosn@+isinn@ where n is any positive integer. We use the principle of mathematical induction. Assume that the result is true for the particular positive integer k, ie. assume (cose +isine)* = coske + isin ke. Then multiplying both sides by cos@ +isine, we find (cose +isine)kt! = (coske +isinke)(cose +isine) = cos(k+1)6 + isin (k+1)0 by Problem 19. Thus if the result is true for »=k, then it is also true for 1 = k+1. But since the result is clearly true for 1 =1, it must also be true for » = 1+1=2 and n= 2+1 = 8, ete, and so must be true for all positive integers. The result is equivalent to the statement (e')" = eni®, Prove the identities: (a) cos5¢ = 16 cos*@ — 20 cos*¢+5cosé@; (b) (sin5é)/(sing) = 16 cost@ — 12 cos?6 +1, if 640,+27,+22,.... We use the binomial formula (a+b)® = a® + (Pa"-1b + (B)a-2b2 + ++) + (Fam br + ++ + On nt! — Finn!’ The number n! or factorial n, is defined as the product 1*2°3--:2 and we define 0! = 1. where the coefficients (+) aiso denoted by ,C,, are called the binomial coefficients. From Problem 20, with n= 5, and the binomiai formuia, cos5@ + isinSe = (cose + isin o)§ = costs + (})(costa)(i sins) + (§)(cos* 4)(é sin 6)? + (3)(cos? 6)(i sin e)3 + (§)(cos 4)(i sin e)* + (isin o)s = costé + Sicosts sing — 10 costs sin®e — 10icos?e sine + Scosesinte + isin’e = ecos'¢ — 10 cos’esin?e + 6 cose sinte + (5 cost @ sing — 10 cos? ¢ sin’6 + sin’ 4) Hence (a) cos5@ = cos'¢ — 10 cos*@ sin?6 + 5 cose sin*@ = cosS@¢ — 10 cos? ¢ (1 — cos?¢) + 5 cose (1 — cos? 6)? = 116 cosde — 20 cosde@ + Beose ety s | and (b) sintds = 6costesine — 10 cos?ésinie + sind¢ . sin be = 5coste — 10 coste sin?e + sinte sin 6 = §&coste — 10 cos®¢ (1 — cos?) + (1 — cos 6)* = 16 coste — 12cos*e + 1 provided sing #0, ie. @ 0, ta, t27,.... 7) - - Show that (a) cos@ = ote (b) sing = oe We have (1) ef = cose+ ising, (2) e-#® = cose —itasineCHAP. 1] COMPLEX NUMBERS 17 18 ~0 (a) Adding (1) and (2), e+e = 2eos9 or cose = at ‘ 10 — 9—10 (b) Subtracting (2) from (1), e—e- = 2isine or sine = <= 23. 24. Prove the identities (a) sin'é = $sin@ —4sin39, » ©) cost? = $c0s40 + $ cos20 + a ie@ 6 8 18 — g~ié (a) sin’g = (SS) = ea = - 7 (em? — 3(0)%0-0) + B(0i\(6 2 e— 19)3} 4 \ _ 1G cis ear oti 3B /e0 — @-16\ 1 [esl — 9 80" = — gy ket? — 50” + Se # me 3i6) = a aero eee WH) 73 2 = 3 ine —14 singe 4 4 (b) cost 6 (“ + a _ (9 + 6 18)4 2 ~ 16 = + {(einy + 4(e!#)8(e~i8) + G(et9)2(e—18)2 + 4(¢0)(6-10)8 + (g—t8)4} 1 1 / ett + @~ 410 1 / e240 +. ¢— 240 8 = —(etie 246 -20 4 9-40) = 2/2 Te (are 3S 16 (@ + 46216 + 6 + de + e~ 40) 2( 5 ) + #( 2 )+8 _1 1 3 = g 008 40 + 9.008 26 + 8 Given a complex number (vector) z, interpret |’ geometrically ze* where « is real. A Let z=re'® be represented graphically by vector Sf OA in Fig. 1-30. Then By zeit = pelde gia = yeil@te) g z= rele zeta is the vector represented by OB. 6 x Hence multiplication of a vector z by ef amounts - oO to rotating z counterclockwise through angle a. We can consider e!@ ag an operator which acts on z to produce this rotation. Fig. 1-30 Prove: e® = elertz, f= 0,+1,+2,.... elOt2k7) = cos(o+ 2kz) + isin(e+2kr) = cose t+isine = ef? Evaluate each of the following. (a) [3(cos 40° + ¢sin 40°)][4(cos 80° + isin 80°)] = 3+ 4[cos (40° + 80°) + isin (40° + 80°)] = 12(cos 120° + i sin 120°) / RON = = 12 (-§+ 34) = -6 + 638i (by 2cis16°)7 _ 128 cis 105° = 29 ig (105° — 135°) (4cis45°)8 ~ 64 cis 136° 2[cos (—80°) + isin (—30°)] = 2[cos80° —isin30°] = V3 — 4 1+ VBi\" _ f 2cis(60°) |” - 19ne\10 ; ; 1, V3, = = = 1 o = Oo tz eet (c) G = =) if (60°) (cis 120°) cis 1200 cis 120 2 + 2 4 Another method. ». 10 10 (vs) = (255) = (e2r/3)l0 = 20/3 — V3i “lg 6% g27i/3 = (1)[cos (27/8) + i sin (22/3)) = — bie +18 COMPLEX NUMBERS [CHAP. 1 27. Prove that (a) arg(zize) = argzi+argzs, (b) arg(z:/z:) = arg2i—argze, stating 29. Find each of the indicated roots and locate them graphically. 28. (a) Find all values of z fo appropriate conditions of validity. Let 2, = 1, (cos6,+isiné,), 2. = r2(cos@+¢siné,), Then argz, = 61, arg zy = 9. (a) Since 2,2, = ryrz {cos (6, + 62) + isin (0, + 62)}, arg (z,22) = 6, +6, = arga, + arg zp. 2 r z (b) Since — = — {cos (61 — 62) + isin (0;—6,)}, arg (=) = 6,;— 62 = arg 2, — arg 2. 22 Te 22 Since there are many possible values for 9, = argz, and 6) = argz,, we can only say that the two sides in the above equalities are equal for some values of argz, and argz,. They may not hold even if principal values are used, Princip VSilies are used hich 2° = —32, and (b) locate these values in the com- plex plane. (a) In polar form, -32 = 32{cos (z+ 2kr) + isin (7 + 2kr)}, k = 0, +1, +2,---, Let z = r(cose + isine). Then by De Moivre’s theorem, 2 = r(cosbe + isin5e) = 382{cos (7 + 2kr) + i sin (x + 2kr)} and so) 5 = 82, 59 = 7+2kz, from which - r= 2, 6 = (r+2kr)/5. Hence z= 2 {cos (= a) + isin ( + ze) 5 5 Ifék=0, z = z = 2cos 2/5 + isin 7/5). Ifk=1, z = zw = 2(cos 87/5 + isin 37/5). Ifk=2, zz = 2g = 2(cos 62/6 + isin 57/5) = — Ifk=8, z = a = 2(cos Tr/5 + isin 77/5). . Ifk=4, 2 = 25; = 2(cos 97/5 + i sin 97/5). % ° By considering k = 5,6,... as well as " negative values, —1,—2,..., repetitions of the above five values of z are obtained. Hence these are the only solutions or roots of the given equation. These five roots are calied the fifth roote of —$2 and are collectively denoted by (—82)!/5, In general, a!/" represents the nth roots of a and there are n such roots. (b) The values of z are indicated in Fig. 1-31. Note that they are equally spaced along the circumference of a circle with centre at the origin and radius 2. Another way of saying this is that the roots are represented by the vertices of a regular polygon. Fig. 1-31 (a) (-1+at/3 ~1+i = V2 {eos (87/4 + Qke) + isin (82/4 + 2ke)} _aaate — eve! (Se/4 + ake oe (S24 + Bk | (Hi trye = “o%8 3 } + tsin( 3 If k=0, z, = 21/8(cos 7/4 + isin 7/4). If kK=1, z, = 2'/8(cog 117/12 + isin 117/12). If kK=2, 2, = 21/8(cos 197/12 + isin 197/12). These are represented graphically in Fig. 1-32. Fig. 1-82\ CHAP. 1] COMPLEX NUMBERS 19 / ¢ (b) (-278 — 2414 ~2V3— 24 = 4{cos (77/6 + 2k) + isin (7n/6 + 2kr)} (-273 ~— 2i)1/4 = aval cos ( 2/6 + tke + Zhe + isin ( 12/6 + Bkn | If k=0, 2, = V2 (cos 77/24 + isin 7/24). If kK=1, 2 = V2 (cos 197/24 + isin 197/24). If k=2, 23 = V2(cos 817/24 + isin 317/24). If k= 3, 2 = V2 (cos 430/24 + isin 437/24). These are represented graphically in Fig. 1-38. Fig. 1-33 30. Find the square roots of —15— 8i. Method 1. —15 — 8 = 17{cos (6+ 2kz) + isin(e@+2kr)} where cose = —15/17, sine = —8/17, Then the square roots of —15—8i are V17 (cos 6/2 + isin 6/2) (2) and V17 {cos (6/2 + 7) + isin (6/2+n)} = ~vi7 (cos 6/2 + i sin 9/2) (2) Now cos 6/2 = *V(1 + cose)/2 = +V(i— i6/in = +117 sin 6/2 = +V¥(1— cose)/2 = +V( + 1e/lt/2 = +4/V17 Since_@ is an angle in the third quadrant, 6/2 is an angle in the second quadrant. Hence cos 6/2 = -1/V17, sin 0/2 = 4/V17 and so from (1) and (2) the required square roots are —1+4i and 1 — 44. As a check note that (-1+4)2 = (i — 44) = —15 — 81. Method 2. Let p+iq, where p and q are real, represent the required square roots. Then (p+ iq)? = p?—q?+ 2pqi = —15 - Bi or (8) p?—g? = —16, (4) pq = —4, Substituting ¢ = —4/p from (4) into (8), it becomes p?—16/p2 = ~-15 or p4+15p2—16 0. ie. (p? + 16)(p?— 1) = 0 or p? = —16, p?=1. Since p is real, p= +1. From (4) if p=1, q=—4: if p=-l1,q=4. Thus the roots are —-1+4i and 1— 4i, POLYNOMIAL EQUATIONS 31. Solve the quadratic equation az*+bze+e= 0, a0. Transposing ¢ and dividing by a 0, ze 5, = - b\? b b\? e b\? ' 2 ; 24 9 —) = -£ 2 Adding (2) [completing the square], a+ a? + ( -) a + (3) b\? 2 — dace when (+m) = “at ; 2 6 _— +Vb2— doc Taking square roots, ct = 2a 2a —b + Ve? Hence z= avin a. 82. Solve the equation z? + (2i-—3)z2+5—i = 0. From Problem 31, a@=1, 6 = 2i—38, c= 5—4 and so the solutions are a = rot Vel=4ae — ~(2i- 38) + V2i— 3)? — AB — _ &-%+ ibs . 2a ~ 2(1) 3— 2+ (1-4) =? 2-3 or 1+i using the fact that the square roots of —15—8i are +(1—4i) [see Problem 30]. These are found to satisfy the given equation. bs20 33. 35. COMPLEX NUMBERS (CHAP. 1 If the real rational number p/g (where p and q have no common factor except +1, i.e. p/q is in lowest terms) satisfies the polynomial equation aoz"*+az""!'+--- +a, = 0 where do, 01,...,@n are integers, show that p and q must be factors of da and ao respectively. Substituting z= p/q in the given equation and multiplying by 9” yields app” + a, p"—1q Se G,—1 pg") + a,Q" = 0 (1) Dividing by p and transposing the last term, a,g" agp" + aypP—2@g toes + ayygh) = ot ? (2) Since the left side of (2) is an integer, so also is the right side. But since p has no factor in common with q, it cannot divide gq and so must divide a,. Similarly on dividing (1) by q and transposing the first term, we find that q must divide ap. Solve 6z* — 252° + 8222+ 82-10 = 0. The integer factors of 6 and —10 are respectively +1, +2,+8,+6 and +1,+2,+5,+10. Hence by Prob. 38 the possible rational solutions are +1, +1/2,+1/3, +1/6, +2, +2/3, +6, +5/2, +5/3, +6/6, +10, +10/3. By trial we find that z = —1/2 and 2 = 2/3 are solutions, and so the polynomial (2z + 1)(8¢ — 2) = 622—z—2 is a factor of 624 — 2522+ 8222+ 32-10, the other factor being 27— 42+ 5 as found by long division. Hence ' 0 6zt — 2623 + 8222 + 32—10 = (622—2—- 2)(22-4e +5) = The solutions of z2—42+5 = 0 are [see Problem 31] = AeVIe=80 _ 4tvr4 _ 48 _ pn; 2 ~ 2 2 Then the solutions are —1/2, 2/3, 2+%, 2—%. Prove that the sum and product of all the roots of ao2z*+ai2"-' + +++ tan = 0 where a *0, are —d:/do and (—1)"an/ao respectively. If 21,29) ---,%, are the n roots, the equation can be written in factored form as ag(z — %)(@ — %)+°*(@— 2%) = 9 Direct multiplication shows that ala” — (@y tag tices teqarmh toes + (1) eyzgQ° zn} = 0 It follows that —ag(z,+2zg+ °° +2) = a, and ao(—1)"212Q°+*%y = On from which ty tag tes +z, = Tay/ao, 2yhqh?'%y_ = (—1)"ay/dg as required. If p+qi is a root of az" ta,2""2+--+> +a, = 0 where ao%0, a1,...,dn, P and q are real, prove that p— qt is also a root. Let p+qi = re in polar form. Since this satisfies the equation, Gyr” ei + ayrr-l el(n-1)08 4 ++ + ayy rel’ +a, = 0 Taking the conjugate of both sides agr® e+ ayr—1e—t—8 4 +++ + dy re% + a, = 0 we see that re- = p—qi is also a root. The result does not hold if ao,...,@, are not all real (see Problem 32). The theorem is often expressed in the statement: The zeros of a polynomial with real coefficients occur in conjugate pairs. oceania one sememulaiiii in 5s ie adamant attain tenon iaaea wan imennsanicni —CHAP. 1) COMPLEX NUMBERS 21 THE rth ROOTS OF UNITY 37. Find all the 5th roots of unity. z= 1 = cos2kr + isin 2kr = etkti where k = 0, +1, +2, . Then zs cos 7 + isi n Se = etkri/5 where it is sufficient to use & = 0,1,2,3,4 since all other values of k lead to repetition. Thus the roots are 1, 27/5, e4mi/5, e8ri/5, e8ti/5, If we call e27i/5 = w, these can be denoted by 1, w, w, 3, wf, . 38 If n = 2,3,4,..., prove that (a) cos2 + cos <= + cos foes + cos "— Tn = -4 (4) sin = + sin + sin toes yb sin 2 — 1 = 0 Consider the equation z*—1 = 0 whose solutions are the nth roots of unity, 1, e2tt/n, eatin, ebtiin, ty E20—Dri/n By Problem 85 the sum of these roots is zero. Then 1 + e2mtin 4 gtmiin 4 g6ri/n 4... 4 gn-Dmiin = 9 ie, . 2(n — 1)2) ff. Qe . Se , An {: + cos 2= + cos #2 torte + copy mein + sin— + ++: + sin= n n n n n from which the required results follow. DOT AND CROSS PRODUCT 39. If 2: =3—4i and z = —4+81, find (a) 21022, (b) 21 X 2x. (a) 2° 2 = Re {z:2.} = Re {(3+ 47)(—4+34)} = Re {-24-—7i} = -—24 Another method. 2, °z) = (3)(—4) + (—4)(8) = —24 (b) 2) X 2 = Im {42.} = Im {(8 + 41)(-4+3)} = Im{-24-7} = —- Another method. %, X 2. = (8)(3) — (—4)(—4) = — 40. Find the acute angle between the vectors in Problem 39. From Problem 39(a), we have cose = 1-2 . ___~24 0 _ = 24 = —.96. lzi||ze] ~ [8 — 4a] (~4 + 83] 25 Then the acute angle is cos—1!.96 = 16°16’ approximately. Area of parallelogram [Fig. 1-34] = (base)(height) (lzz|)(12] sin 6) |24| |2q| sine = | 2; x zg| Fig. 1-34 ItCOMPLEX NUMBERS [CHAP. 1 ind the area of a triangle with vertices at A(x, 41), B(x2,y2) and C(2s, ys). The vectors from C to A and B [Fig. 1-35] are respectively given ° 42, F 7 2) + iy.— Vs) » _— = 440% ag = (aq— a} + ve — is) zy Since the area of a triangle with sides z; and 2; is half the area of the corresponding parallelogram, Fig. 1-95 we have by Problem 41: Area of triangle = $(|2,X%| = $|Im{{(e,—%s) — v1 —ys)][(72— 2s) + tly —ys)]} | t | (ay — %g)(Yo — Ys) — (Ys — ¥s%e — Zs) | $ | ve — vite + toys — Yot%s + H3V1 — Yo! wy yy l i in determinant form. COMPLEX CONJUGATE COORDINATES 48, Express each equation in terms of conjugate coordinates: (a) 2x +y = 5, (b) 27+ y’ = 86. (a) Since z= axe+iy, Z=a-~ty, # =i, y= =: Then 22 +y = 5 becomes 2 (2+) + (=) = 6 or (2i+1)z + (2i-1)E = 10: The equation represents a straight line in the z plane. {b) Method 1. The equation is (2+ ty\e—iy) = 36 or 2% = 36 Method 2. Substitute 2 = 223, y= + in 22+ y2 = 36 to obtain zz = 36. The equation represents a circle in the z plane of radius 6 with centre at the origin. AA TD 44. Prove that the equa y c. atk + Bz + BZ +y7 = 0 where a and y constant. The general equation of a circle in the zy plane can be written A(x? +y?) + Be +Cy+D = 0 which in conjugate coordinates becomes Aa +B(E 244) 4 0(25 2) 4p =0 or an+(2+8)e4(2-S)a+ = 0 Calling A =a, z + g = B and D=y, the required result follows. In the special case A = a =0, the circle degenerates into a line. ne in the z plane can be written as are real constants while e B may be a complex POINT SETS 45. Given the point set S: (i, 44, 41,3i,...} or briefly {i/n}. (a) Is S bounded? (b) What are its limit points, if any? (c) Is S closed? (d) What are its interior and boundary points? (e) Is S open? (f) Is S connected? (g) Is S an open region or domain? (h) What is the closure of S? (i) What is the complement of S? (j) Is S countable? (x) Is S compact? (0) Is the closure of S compact? (a) S is bounded since for every point z in S, |z| < 2 [for example], i.e. all points of S lie inside a circle of radius 2 with centre at the origin. (8) Since every deleted neighbourhood of z = 0 contains points of S, a limit point is z = 0. It is the only limit point.CHAP. 1) COMPLEX NUMBERS 23 Note that since S is bounded and infinite the Weierstrass-Bolzano theorem predicts at least one limit point. (c) S is not closed since the limit point z=0 does not belong to S. . (d). Every 5 neighbourhood of any point i/n [i.e. every circle of radius 5 with centre at i/n] contains points which belong to S and points which do not belong to S. Thus every point of S, as well as the point z= 0, is a boundary point, S has no interior points. (e) S does not consist of any interior points: Hence it cannot be open. Thus S is neither open nor closed. (f) If we join any two points of S by a polygonal path, there are points on this path which do not belong to S. Thus S is not connected. . (g) Since S is not an open connected set, it is not an open region or domain. (h) The closure of S consists of the set S together with the limit point zero, i.e. {0, 2, 444i, . 6}. (i) The complement of S is the set of all points not belonging to S, ie. all points z * 1, 7/2, 1/8, ... (3) There is a one to one correspondence between the elements of S and the natural numbers 1, 2, 3, ... as indicated below, iow wh k { ¢ 7g 1 2 38 4 Hence S is countable. (k) S is bounded but not closed. Hence it is not compact. (J) The closure of S is bounded and closed and so is compact. & Given the point sets A = (3, -7,4,24+47,5}, B= {-7, 0, -1,2+7}, C= (-V2i, 4, 3}. Find (2) A+B or AUB, (b) AB or ANB, (c) AC or ANC, (4) A(B+C) or AN(BUC), (e) AB+AC or (ANB)U(ANC), (f) A(BC) or AN(BNC). (eq) A+B = AUB consists of points belonging either to A or B or both and is given by {3, -i, 4,2 +4, 5, 0, -1}. (b) AB or ANB consists of points belonging to both A and B and is given by {-~i, 2+ 4%}. (ec) AC or ANC = {8}, consisting of only the member 3. : (d4) B+C or BUC = {-i, 0, —1, 2+7, —V2i, + 3}. Hence A(B+C) or AN(BUC) = {3, —1,2+%}, consisting of points belonging to both A and B+C. (e) AB = {-i1,2+%}, AC = {3} from parts (6) and (c). Hence AB+AC = {-i, 2+, 3}. From this and the result of (d) we sec that AB+tO) = AB+AC or An(BuC) = (ANB)U(ANC), which illustrates the fact that A,B,C satisfy the distributive law. We can show that sets exhibit many of the properties valid in the algebra of numbers. This is of great im- portance in theory and application. (fy) BC = BNC = @, the null set, since there are no points common to both B and C. Hence A(BC) = @ also. MISCELLANEOUS PROBLEMS 47. A number is called an algebraic number if it is a solution of a polynomial equation Goz" +a, 2"-!4+ --- + @n-12 +4, = 0 where Q,d1,...,@, are integers. Prove that (a) V8 + V2 and (b) ¥/4—2i are algebraic numbers. (a) Let 2 = V8 +V2 or z-y2 = 73, Squaring, z? — 2v2z+ 2=3 or 2-1= 2V22. Squaring again, z2*—2z2+1 = 822 or 2t—10z22+1 = 0, a polynomial equation with integer coefficients having V3 + V2 as a root. Hence ¥3+ v2 is an algebraic number. (6) Let z= V4 - 21 or z+2i = 4, Cubing, 23 + 322(2%) + 32(2i)2 + (21)8 = 4 or 23—-12z-4 = (8 — 622). Squaring, z6+ 1224 — 823 + 4822+ 962 +80 = 0, a polynomial equation with integer coefficients having 4 — 2i as a root. Hence 4 ~— 24 is an algebraic number. Numbers which are not algebraic, i.e. do not satisfy any polynomial equation with integer coefficients, are called transcendental numbers. It has been proved that the numbers 7 = 3.14159... and ¢ = 2.71828... are transcendental. However, it is still not yet known whether numbers such as ex or e+ 7, for example, are transcendental or not.24 COMPLEX NUMBERS [CHAP. 1 (a) The given equation is equivalent to |z—38{ = 2{[z2+3| or, “et z= att, [etiy-—3|] = 2|a+i+8|, ie, V(@~ 8)2 + y? = 2V(a+8)2 + y? Squaring and simplifying, this becomes et+y2+10e+9 = 0 or (@+5)?+y? = 16 ie; |z+5| = 4, a circle of radius 4 with centre at (—5,0) as shown in Fig. 1-36. Geometrically, any point P on this circle is such that the distance from P to point B(3,0) is twice the distance from P to point A(—3, 0). Another method. z-3 z+3 = 2 is equivalent to f2—3\(2#—3° _ . _ (3 (a =4 or ##+52+52+9 = 0 ie. (2+ 5)2+5) = 16 or [et5| =4. Fig. 1-36 (6) The given inequality is equivalent to [z—8| < 2[z+3] or V(x—3)?+y? < 2V(2+3)? + y%. Squaring and simplifying, this becomes x?+y2+102+9 > 0 or (x+5)?+y? > 16, ie. Jzt+5| > 4. The required set thus consists of all points external to the circle of Fig. 1-36. Sern @ aate A awl D = au. UIVEN Wie SEIS A ana vo Ir pres ar epresent geometrically (a) ANB or AB, (b) AUB or A aw Fig. 1-87 Fig. 1-38 50. Solve 27(1 —2*) = 16. Method Il. The equation can be written 24—2?7+16 = 0, ie. 24+822+16—922 = 0, (22+ 4)? — 922 = 0 or (22+ 4+ 8z)(z2+4-—382z) = 0. Then the required solutions are the solutions of 22+3z+4 = 0 3, vi, 3, v7. 2 = -2+¥! f+, and 22—8z2+4= 0, or 2 2 # and 3 Q* Method 2. Letting w= z?, the equation can be written w?-w+16 = 0 and w = To obtain solutions of z? = ; + 3 7i, the methods of Problem 30 can be used. Vii. + Nolet bo] 09CHAP. 1} COMPLEX NUMBERS 25 51. If 21, 22,23 represent vertices of an equilateral triangle, prove that a+ Bt 2k = 212, + e2ks + 2521 From Fig. 1-39 we see that %— a = 7/3 (zy — 24) % — 2g = eT/3 (29 ~ 25) Then by divisi —“1 _ 474 en by vision, 45 ~ 2 — zg or Bt 23+ 22 = ayzy t+ zoey + x92, Fig. 1-39 52. Prove that for m = 2,3,. (m—-1)r _ m m ~ Qm-1 The roots of z2*—1 are z= 1, e2ui/m, etrism, || ., etm—loti/m, Then we can write a — 1 = (2-1)(e— e2mi/my(z — etri/m)... (g — @2(m—- 1)mi/m) Dividing both sides by z—1 and then letting z=1 (realizing that (2™—1)/(z-1) = 1+2+22+ -+2™~-1] we find sin “sin 27 sin 3* mmm m = (1— e@ti/my(] — eftt/my ... (1 — e&(m—Dai/my (1) Taking the complex conjugate of both sides of (1) yields m = (1—¢-2ni/my(y — e~4ni/m).. «(1 — e-2m— 1) t/m) (2) Multiplying (7) by (2) using (1 — e2*ti/m)(1 — ¢—2kri/m) = 2 — 2 cos (2km/m), we have , mt = Qm-1 (1 - con2e))(1 — cos $#)-+-(1 ~ cos in =i) (3) m m Since 1 — cos(Zkz/m) = 2 sin? (kx/m), ($) becomes , _ 2 wine os ar (m—1)7 m= = 22m-2 sin? sin? sin a (4) Then taking the positive square root of both sides yields the required result. Supplementary Problems FUNDAMENTAL OPERATIONS WITH COMPLEX NUMBERS 53. Perform each of the indicated operations: _ _ 2— 3% -1)2 gai (a) (4— 3%) + (2¢— 8) (e) 42% .{h) (21-1) ts =~+ al () 8-1 +4) ~ 208) (A+ N3+A)1-) gq _ 4 +194 48 (c) (3 + 21)(2 — 1) 2— B+ as _ ) 3 — (2 + 7(3 — 22)(1 + 24) 1+i a) ) G-aaa+y—sG—y} Gy BENE—2942 Gy g (148)"_g (1-1) Ans. (a) —4—i (c) 8+7 (e) 11/17 — (10/17): (g) —15/2 + 5% (i) 2+ (b) —17+ 144 (dq) -94+7i (f) 21+4 {kh} —11/2 — (28/2) G) -8~ 23 54. If 2, =1-i, 2 = —-2+4i, 4 = V3 — 2i, evaluate each of the following: +241 (a) 28 +22,—3 (¢ | (h) lad t al? + [23 - 2 17 agar (b) | 2z_— 82, [2 is) (2) Re {2z3 + 323 — 523} (c) (23 ~ 25)8 ONS 3 (+ 2) (i) Im (2429/29) (@) | 2122 + 298, | (9) (@e ¥ &s)(21 — zg) Ans. (a) ~1—4i (c) 1024: (e) 3/5 (9) —7 + 3V3 + Y8i () -86 (8) 170 (d) 12 (fy -1/7 (A) 765 + 128/32 GQ) (6784+ 4/726 COMPLEX NUMBERS (CHAP. 1 55. Prove that (a) (Z1%) = 2%, (b) (%%9%3) = 2%_%s. Generalize these results. 57. Find real numbers x and y such that 2% — 3iy + 4ix —2y—5— 10% = (w@t+y+2) — (y—w+3)4. Ans. x=1, y=—2 58. Prove that (a) Re{z} = (z+ 2)/2, (6) Im{z} = (z2—%)/2i. 59. Prove that if the product of two complex numbers is zero then at least one of the numbers must be zero. 60. If w = 3iz—2? and z= x+y, find |w|? in terms of x and y. Ana. x4+ yt + Qa2y? — 6a2y — Gy? + 9a? + Dy? GRAPHICAL REPRESENTATION OF COMPLEX NUMBERS. VECTORS. 61. Perform the indicated operations both analytically and graphically. (a) (2+ 8i) + (4-54) (c) 3(1+ 22) — 2(2 — 81) (e) 44-32 + $(5 + 2%) (bY (7 LN (A — OV fd\ 2/1 LN +L OA ON — OLE wy Ve 7 at; (ap Gla Tt 56. Prove that (a) (2/29) = 2,/%,, (b) | 21/zo| = |23|/|zo| if zp, 0. | (b) (74 + (4 — 22 — 2453) Ans. (a) 6—2i, (b) 3+38%, (c) —1 +121, (d) 9— Bi, (e) 19/2 + (8/2)t 62. If z;, 2 and z3 are the vectors indicated in Fig. 1-40, construct y graphically: 2 2 Zz (a) 22, +25 (c) 21 + (2, +25) (e) ha2— Bar + §2s * « (0) (2, + 2) + 2g (d) 32, — 2%, + 523 2 3 63. If z,;=4—3i and z, = —1+ 21, obtain graphically and analyti- cally (a) |2,+2|, (b) | %—22|, (€) #4, (d) | 2%, — 3%, — 2]. | Ans. (a) ¥10, (b) BY2, (c) B+5i, (d) 15 Fig. 1-40 64. The position vectors of points A, B and C of triangle ABC are given by z, = 1+ 2i, z= 4—2i and Zz = 1—6i% respectively. Prove that ABC is an isosceles triangle and find the lengths of the sides. Ans. 5, 5,8 65. Let 2,29,Z3,Z4 be the position vectors of the vertices for quadriiaterai ABCD. Prove that ABCD isa parallelogram if and only if 2,—%,—23;+2, = 0. 66. If the diagonals of a quadrilateral bisect each other, prove that the quadrilaterai is a paraiieiogram. 67. Prove that the medians of a triangle meet in a point. 68. Let ABCD be a quadrilateral and E,F,G,H the midpoints of the sides. Prove that EFGH is a parallelogram. 69. In parallelogram ABCD, point E bisects side AD. Prove that the point where BE meets AC trisects AC. 70. The position vectors of points A and B are 2+1i and 3—2i respectively. (a) Find an equation for line AB. (b) Find an equation for the line perpendicular to AB at its midpoint. Ans. (a) z— (277) = 1-31) or a-—2+t, y~i- si or Setry = 7 (b) z— (5/2 — #/2) = 4(8+%) or © = 86+ 5/2, y= t— 1/2 or 2#—-—8y=4 71. Describe and graph the locus represented by each of the following: (a) |z—i| = 2, (b) |2+2é|+ |z—2i| = 6, (ec) |z—-3|-|2+3| = 4, (d) 2(@+2) = 3, (e) Im {22} = 4. Ans. (a) circle, (5) ellipse, (ce) hyperbola, (d) circle, (e) hyperbola 72. Find an equation for (a) a circle of radius 2 with centre at (—3,4), (6) an ellipse with foci at (0, 2) and (0,--2) whose major axis has length 10. Ans. (a) |2+38-—4i| = 2 or (@+3)?+ (y—4% = 4, (b) [zt+2i| + | z-—2i| = 10CHAP. ij COMPLEX NUMBERS 27 Describe graphically the region represented by each of the following: (a) 1<|[zt+i| 52, (6) Re{22}>1, (ce) |2z+3i] >4, (@ |z+2—8i] + |2z-2+3%] < 10. 74, Show that the ellipse |z+3|+[z—8| = 10 can be expressed in rectangular form as 22/25 + y*/16 = 1 [see Problem 13(b)]. AXIOMATIC FOUNDATIONS OF COMPLEX NUMBERS 1. Use the definition of a complex number as an ordered pair of real numbers to prove that if the product of two complex numbers is zero then at least one of the numbers must be zero, 76. Prove the commutative laws with respect to (a) addition, (b) multiplication. 77. Prove the associative laws with respect to (a) addition, (b) multiplication. 78. (a) Find real numbers x and y such that (c,d)+(z,y) = (a,b) where (c,d) ¥ (0, 0). (b) How is («, y) related to the result for division of complex numbers given on Page 2? 79. Prove that $y, SiN 8y)--- {cos Gq, Sin @,) = (cos [@, + 62+ --- + 6q], sin [ey t+ 6g + --- + eq) 80. (a) How would you define (a, b)!/" where is a positive integer? (6) Determine (a, 5)!/2 in terms of a and b. POLAR FORM OF COMPLEX NUMBERS 81. Express each of the following complex numbers in polar form. (a) 2-24, (b) -1+ VBi, (c) 2V2+2V2%, (d) —i, (e) —4, (f) —2VB—2i, (g) V2i, (h) VB/2— 34/2. Ans. (a) 2V2 cis 315° or 2V2 e774, (b) 2 cis 120° or Qe2ri/3, (c) 4 cis 45° or 4e7/4, (d) cis 270° or eSt/2, (e) 4 cis 180° or 4e™, (f) 4 cis 210° or 4e77/6, (9) V/2 cis 90° or V2er/2, (h) V3 cis 300° or V3 e5tv/3, 82. Show that 2+%4 = VBeitan t/a, 83. Express in polar form: (a) —3— 4%, (6) 1—2i. Ans. (a) 5 ei(t + tan~! 4/3), (b) V5 e—itan7! 2 84. Graph each of the following and express in rectangular form. (a) 6 (cos 135° + i sin 135°), (6) 12 cis90°, (c) 4 cis 315°, (d) 2e57i/4, (e) Be77/6, (f) Ze—2ni/3, Ans. (a) ~8V2 + 3V2%, (b) 12%, (0) 2V2— 2V2i, (d) —V2 — VBi, (¢) —5V8/2 — (6/2)i, (f) 83/2 — (8/2)i 85. An airplane travels 150km southeast, 100km due west, 225km 30° north of east, and then 323km northeast. Determine (a) analytically and (b) graphically how far and in what direction it is from its starting point. Ans. 375km, 23° north of east (approx.) y 86. Three forces as shown in Fig. 1-41 act in a plane on an object placed at O. Determine (a) graphically and (3) ana- lytically what force is needed to prevent the object from moving. [This force is sometimes called the equilibrant.] 87. Prove that on the circle z = Rei, |el2| = ¢~Rsine, 88. (a) Prove that r,¢!: + rei: = rei where rs = Vr? + 2 + 2ryre cos (6, — 69) 6, = tan-! r,sin 6, + v2 sin 2 8 71 COB 0, + 12 CO8 Oy (b) Generalize the result in (a). Fig. 1-4128 COMPLEX NUMBERS (CHAP. 1 DE MOIVRE’S THEOREM 89. 91. 93. © = . Evaluate each of the following: (a) (5 cis 20°)(3 cis 40°) tc) (8 cis 40°)8 (a) (Be7#/8) (2e~S71/4)(GeSri/3) @ (45 a) (HH iy (6) (2 cis 50°)8 (2 cis 60°)* (40271/3)2 V3+i Ang. (a) 15/2 + (15¥3/2)i, (b) 82—3278%, (c) —16— 16/34, (d) 33/2 —(3V8/2)i, (e) —¥3/2 — (1/2)i Prove that (a) sin39 = 3sine — 4sin39, (b) cos3e = 4 cos*e — 3 cosé. Prove that the solutions of z4—8z2+1=0 are given by z = 2cos 36°, 2cos 72°, 2 cos 216°, 2 cos 252°. Show that (a) cos36° = (¥5+1)/4, (b) cos 72° = (¥5—1)/4. (Hint: Use Problem 91.] sin sin 46 Prove that (a) ——— sine 8cos8¢é —4 = 2cos3e + 6cose — 4 (b) cos4e = 8 sinte — 8 sin?9 + 1 Prove De Moivre’s theorem for (a) negative integers, (5) rational numbers. ROOTS OF COMPLEX NUMBERS Find each of the indicated roots and locate them graphically. (a) (2¥3 — 24/2, (b) (—4 + 40/5, (0) (2 + 2VB)1/3, (d) (—164)/4, (e) (64)1/8, (fF) (2. Ans. (a) 2cis165°, 2cis345°. (b) V2cis27°, V2cis99°, V2cis171°, V2 cis2438°, V2 cis 315°. (c) V4 cis20°, 4 cis140°, V4 cis260°. (d) 2cis67.5°, 2cis157.5°, 2 cis247.5°, 2 cis 337.5°. (e) 2 cis 0°, 2 cis 60°, 2 cis 120°, 2 cig 180°, 2 cis 240°, 2 cis 300°. (f) cis 60°, cia 180°, cis 300°. Find all the indicated roots and locate them in the complex plane. (a) cube roots of 8, (b) square roots of 42+ 472i, (c) fifth roots of —16+16V3i, (d) sixth roots of —27i. Ans, (a) 2 cis 0°, 2 cis 120°, 2 cis 240°. (b) V8 cis 22.5°, V8 cis 202.5°. (c) 2 cis 48°, 2cis 120°, 2cis 192°, 2 cis 264°, 2 cis 386°. (d) 73 cis 45°, V8 cis 105°, V3 cis 165°, V3 cis 225°, V3 cis 285°, V8 cis 345°. Solve the equations (a) 2‘ + 9, ® A¢i=ay¥y Ans. (a) 3 cis 45°, 8 cis 185°, 3 cis 225°, 3 cis 3 316° (b) V2 cis 40°, Y2 cis 100°, ¥2 cis 160°, V2 cis 220°, 4/2 cis 280°, V2 cis 340° Find the square roots of (a) 5—12i, (b) 8+ 4/53. Ans. (a) 83—2i, -8+ 2%, (b) ¥10 + V2i, —V10— V2: . Find the cube roots of —11—2i. Ans. 1+ 2i, 4 — V8 + (1+ 4V3)i, -$ — V8 + (4V8- 1% POLYNOMIAL EQUATIONS 100. 101. 102. 103. Solve the following equations, obtaining all roots: (a) 522+2z2+10 = 0, (6) 22+ (¢-—2)z+(8—%) = 0. Ans. (a) (~-1+ 7/5, (b) 1+4,1-2¢ Solve 25-—2z+—23+6z~—4 = 0. Ans. 1,1,2,-1+4 (a) Find all the roots of z#+22+1= 0 and (6b) locate them in the complex plane. Ans, $(1 + iv8), $(-1+tv3) Prove that the sum of the roots of az" + a,2°—-1+ agz®-2+-++- +a, = 0 where ag*0 taken rata time is (—1)"a,/ag where 00 is a lemniscate as shown in Fig. 1-48. M2. Let p, = a2 +05, n = 1,2,8,... where a, and 6, are positive integers. Prove that for every positive integer M we can always find positive integers A and B such that p:Po°+-py = A?+B2, (Example: If 6 = 22+12 and 25 = $?+ 42, then 5+25 = 22+ 112] Fig. 1-43 M48. Prove that sin d(n + Da fa\ ane an 1 naan fa tor 1 Pa \ - - 6 +t cope taj + T «cos (6 + na) a cos (@ + dna “oe sin fa (9 + gna) i: a+1 (b) sine + sin(e+e) + +++ + sin (e+ na) sing(nt De i (9 + $na) ~ “ sin $a 144, Prove that (a) Re{z}>0 and (b) |z-—1| <|z2+1] are equivalent statements. 145. A wheel of radius 1-2 metres (Fig. 1-44] is rotating counterclockwise about an axis through its centre at 30 revolutions per minute. (a) Show that the position and velocity of any point P on the wheel are given respectively by 4e'*t and 4rie!*t, where t is the time in seconds measured from the instant when P was on the Positive « axis. (b) Find the position and velocity when ¢ = 2/8 and ¢ = 15/4. 146. Prove that for any integer m > 1, 1 n- 147. If points P; and Pp», represented by 2, and z, respectively, are such that | 2+ 2%| = |2;—2|, prove that (a) z,/z, is a pure imaginary number, (6) LP,OP, = 90°. 148. Prove that for any integer m > 1, ot 27 cot 2%. cot (MaDe _ cot 2m cot am cot am cot am = 1 149. Prove and generalize: (a) csc? (x/7) + ese? (22/7) + csc? (4r/7) = 2 (0) tan? (7/16) + tan? (87/16) + tan? (57/16) + tan? (77/16) = 28 150. If masses ™m, mo, mg are located at points 21, 22) 23 respectively, prove that the centre of mass is given by a = Miz, + Mode + mazq mm, + Me + m3 Generalize to n masses. 151. Find that point on the line joining points , and z, which divides it in the ratio p: q. Ans. (gz, + pag)/(q + p) 152. Show that an equation for a circle passing through 3 points 21, 2,23 is given by VEY) - Be)32 153. 154. 156. 157. 158. 159. 160. 161. 162. 163. 164. 165. 166. 167. 168. COMPLEX NUMBERS (CHAP. 1 Prove that the medians of a triangle with vertices at 2,,%:,%3 intersect in the point. }(z, + 22+ 2). Prove that the rational numbers between 0 and 1 are countable. [Hint. Arrange the numbers as 0, 4, 4, 3,4, 2,4,%, 8... J 5. Prove that aii the reai rational numbers are couniabie. Prove that the irrational numbers between 0 and 1 are not countable. Represent graphically the set of values of z for which (a) |z| > |z—-1], (0) |z+2| >1+|z-2|. Show that (a) Ve + V3 and (b) 2— V2i are algebraic numbers. Prove that V2 + V3 is an irrational number. Let ABCD:--PQ represent a regular polygon of sides inscribed in a circle of unit radius. Prove that the product of the lengths of the diagonals AC,AD,...,AP is 4n esc? (7/n). Prove that if sing ~ 0, sin 26 _ {mot (¢) “ne gn-1 O {cos @ — cos (kx/n)} sin(2n+1)6 _ n { _ __sinte | (b) —sne - (2n + 1) i 3 sin? kn/(On + DY . Prove cosZng = (—i)" ll i- ___ costa) K=1 cos? (2k — 1x/4nf . If the product of two complex numbers z, and 2, is real and different from zero, prove that there exists a real number p such that 2, = po. If z is any point on the circle |z—1| = 1, prove that arg (z—1) = 2argz = 9 arg (22-2) and give a geometrical interpretation. Prove that under suitable restrictions (a) z™z* = 2™+", (b) (2m) = 2m, Re {z,} Re {zo} — Im {z,} Im {29} Re {z,} Im {z.} + Im {2,} Re {zg}. Prove (a) Re {2,25} (b) Im {2,29} Find the area of the polygon with vertices at 2+ 3i, 3+i, —2—4i, —4—7, -1+ 21. Ans, 47/2 Let @1,@g,...,@, and b,,by,...,6, be any complex numbers. Prove Schwarz’s inequality, "= (3 tot)( 3 tu) n bChapter 2 VARIABLES AND FUNCTIONS A symbol, such as z, which can stand for any one of a set of complex numbers is called a complex variable, i Ta. If to each value which a complex variable z can assume there corresponds one or more values of a complex variable w, we say that w is a function of z and write w= F(z) or w = G(z), ete. The variable z.is sometimes called an independent variable, while w is called a dependent variable. The value of a function at z=a is often written f(a). Thus if f(z) = 2, then f(2d) = (21)? = —4., SINGLE-AND MULTIPLE-VALUED FUNCTIONS If only one value of w corresponds to each value of z, we say that w is a single-valued function of z or that f(z) is single-valued. If more than one value of w corresponds to each value of z, we say that w is a multiple-valued or many-valued function of z. A multiple-valued function can be considered as a collection of single-valued functions, each member of which is called a branch of the function. It is customary to consider one particular member as a principal branch of the multiple-valued function and the value of the function corresponding to this branch as the principal value. Example 1: If w= 22, then to each value of z there is only one value of w. Hence w= f(z)=z2 is a single-valued function of 2. Example 2: If w =2z!/2, then to each value of z there are two values of w. Hence w = f(2) = 21/2 ig & multiple-valued.(in this case two-valued) function of z. Whenever we speak of function we shall, unless otherwise stated, assume single-valued function. INVERSE FUNCTIONS If w = f(z), then we can also consider z as a function of w, written z= g(w) = f-'(w). The function f-! is often called the inverse function corresponding to f. Thus w=— f(z) and a2 = f-'(g\ are inverse functions of each other. TRANSFORMATIONS If w= u+iv (where u and v are real) is a single-valued function of z = x + iy (where # and y are real), we can write u+iv = f(a +ty). By equating real and imaginary parts this is seen to be equivalent to Re len i u= u(z,y), v = u(x, y) (1) Thus given a point (2, y) in the z plane, such as P in Fig. 2-1 below, there corresponds a point (u,v) in the w plane, say P’ in Fig. 2-2 below. The set of equations (1) [or the equivalent, w = f(z)] is called a transformation. We say that point P is mapped or transformed into point P’ by means of the transformation and call P’ the image of P. 8384 FUNCTIONS, LIMITS AND CONTINUITY (CHAP. 2 i Example: If w= 22, then utiv = (e+iy)? = «2 —y?+ 2ixy and the transformation is u = t—y2, v = 2ay. The image of a point (1,2) in the z plane is the point (—3, 4) in the w plane. y » , | z plane P’ w plane /\wW @ P Q 2 u Fig. 2-1 Fig. 2-2 In general, under a transformation, a set of points such as those on curve PQ of Fig. 2-1 is mapped into a corresponding set of points, called the image, such as those on curve P’Q’ in Fig. 2-2. The particular characteristics of the image depend of course on the type of function f(z), which is sometimes called a mapping function. If f(z) is multiple- valued, a point (or curve) in the z plane is mapped in general into more than one point (or curve) in the w plane. CURVILINEAR COORDINATES Given the transformation w = f(z) or, equivalently, u = u(x,y), v = v(z,y), we call (x,y) the rectangular coordinates corresponding to a point P in the z plane and (u,v) the curvilinear coordinates of P. z plane w plane | | | Fig. 2-3 Fig. 2-4 The curves u(z,y) = ¢:, v(%,y) = ¢a, where ¢: and ce are constants, are called coordinate curves [see Fig. 2-3] and each pair of these curves intersects in a point. These curves map into mutually orthogonal lines in the w plane [see Fig. 2-4]. THE ELEMENTARY FUNCTIONS 1. Polynomial Functions are defined by Wo = Ak + A1A™- 14 6+ + On-12 + Gn = Plz) (2) where ao ~0,@i,...,@n are complex constants and n is a positive integer called the degree of the polynomial P(z). The transformation w = az+b is called a linear transformation.CHAP. 2] FUNCTIONS, LIMITS AND CONTINUITY 35 2. Kational Algebraic Functions are defined by P(2) w= 8 Qe) °) where P(z) and Q(z) are polynomials. We sometimes call (3) a rational transformation. The special case w = ant where ad—bc ~ 0 is often calied a bilinear or frac- tional linear transformation. 3. Exponential Functions are defined by w= e@ = ett = et(cosy tisiny) (4) where e = 2.71828... is the natural base of logarithms. If a is real and positive, we define at = eine (5) where Ina is the natural logarithm oF a. This reduces to Complex exponential functions ha imi functions. For example, e+e = oate, emi/e% = et *2, oN 4.8 WA eo Ur. afar thy tetennamateten nv eivenlar fuanettinnse cine _ j Trigonometric ru net ti iS. we aenne tne UPIZOnOMeLrIC Or CirCula TUNCLIONS SINnZ, cos z, etc., in terms of exponential functions as follows. sing = 2 e™ cos 2 eft+e™ 2i 2 secz = 1. 2 csc Zz = 1. 2t ~ ecosz ” e# + e-# sin z e# — e-# sin z ek — gt cos z i(e* + e~ 2) tanz = = Ta olin cotz = = nk COs z i(e* + e~*#) sin z e* —e Many of the pronerties familiar in the case of real trigon hold for the complex trigonometric functions. For example, we have sin’?z + cos’?’z = 1 1+ tan?z = sec?z 1 + cot?z2 = csc?z sin(—z) = —sinz cos(—z) = cosz tan(-—z) = —tanz sin (Zi+22) = sinz:cosz2 + cosz. sin 22 cos (41+ 22) = C084 coszz + sin 2: sin 22 tan 2, + tan Ze 1 = tan: tan Zz 3) Hyperbolic Functions are defined as follows: tan (21 + 22) . e* — e7? e+e* , sinh z = ——— cosh z = ———— 2 2 sech z = — 2 eschz = 1 -_2_ ~ coshz e% + e7? sinh z e—e* sinh z e—e* cosh z e+e* tanh z = = cothe = —— = -———= anh cosh z e+e7* sinh z e* —e-* The following properties hold: cosh?z — sinh?z = 1 1 — tanh?z = sech?z coth?z — 1 = csch?z sinh(—z) = —sinhz cosh (~z) = coshz tanh (-z) = —tanhz sinh (z:+22) = sinh: coshzz + cosh 2: sinh z2 cosh (#1#%) = coshz:coshz, + sinh 2; sinh 22 tanh 2; + tanh z2 tanh (a1 + 1 = tanh, tanh z nh (21 + 22) 1 + tanh z: tanh ze86 10. FUNCTIONS, LIMITS AND CONTINUITY (CHAP. 2 i The following relations exist between the trigonometric or circular functions and the hyperbolic functions: siniz = tsinhz cos iz cosh z tan iz i tanh z sinh iz = isin z cosh iz = cosz tanh iz = itanz ; Logarithmic Functions. If z =e”, then we write w=Inz, called the natural loga- rithm of z. Thus the naturai iogarithmic function is the inverse of the exponentiai | function and can be defined by w = Ing = Inr + i(6 + 2kx) k= 0,+1,+2,... where z = re® = re®+?k™), Note that Inz is a multiple-valued (in this case infinitely- many-valued) function. The principal-value or principal branch of Inz is sometimes defined as Inr+i@ where 0=26< 22. However, any other interval of length 27 can be used, e.g. —1r < 6 S7r, ete. The logarithmic function can be defined for real bases other than e. Thus if z=a”, then w=logaz where a>0 and a#0,1. In this case z=e”’™* and so w = (In z)/(In a). Inverse Trigonometric Functions. If z=sinw, then w=sin~‘z is called the inverse | sine of z or arc sine of z. Similarly we define other inverse trigonometric or circular | functions cos~'z, tan~1z, etc. These functions, which are multiple-valued, can be expressed in terms of natural logarithms as follows. In all cases we omit an additive constant 2krt, k=0,+1,+2,..., in the logarithm. sin-!z = *In (iz + V1 —2?) ese-1z = jin (v=?) cos"!z = Fin (@+V¥2—1) sec-!z = jin(LtviE# i=?) ton-te - Lin (1+ iz\ cotta = Lan (Zt!) 2 \1-tz/ — ai \ze-t/ Inverse Hyperbolic Functions. If z=sinhw then w=sinh-'z is called the inverse hyperbolic sine of z. Similarly we define other inverse hyperbolic functions cosh-'2z, tanh~!z, ete. These functions, which are multiple-valued, can be expressed in terms of natural iogarithms as follows. In ali cases we omit an additive constant Zizi, k=0,+1,+2,..., in the logarithm. sinh™*z = In (z+ V2? +1) esch~!z = In (4 et) cosh-*z = In(g+ yz?—1) sech-!z = in( 1 VI=F) z = 7% -1 1 3 in (324) coth-!z 5 in (3) The Function 2°, where a may be complex, is defined as e*™*, Similarly if f(z) and g(z) are two given functions of z, we can define f(z)? = e%# "f=, In general such func- tions are multiple-valued. tanh7! z Algebraic and Transcendental Functions. If w is a solution of the polynomial equation Po(Z)w" + Pi(zjyw'-) + --- + Pa-i(@)wt+ Pa(z) = 0 (6) where Po # 0, P:(z), ...,Pa(z) are polynomials in z and n is a positive integer, then w = f(z) is called an algebraic function of z. Example: w = z!/2 is a solution of the equation w?—z = 0 and so is an algebraic function of z.CHAP, 2} FUNCTIONS, LIMITS AND CONTINUITY 37 Any function which cannot be expressed as a solution of (6) is called a transcendental function. The logarithmic, trigonometric and hyperbolic functions and their correspond- ing inverses are examples of transcendental functions. The functions considered in 1-9 above, together with functions d by a finite number of operations involving addition, subtraction, multiplication, and roots are called elementary functions. @O 8. 3 Q. 2 a § gs @ 5 BRANCH POINTS AND BRANCH LINES Suppose that we are given the function w =z?, Suppose further that we allow z to make a complete circuit (counterclockwise) around the origin starting from point A [Fig. 2-5]. We have z=re®, w=vVre*2 so that at A, 6=6, and w=Vre%?, After a complete circuit back to A, 6 = A+2r and w= relate? = —\/p gil, Thus we have not achieved the same value of w with which we started. However, by making a second complete circuit back to A, ie. 6 = 6,4 An, w= Vr e%r+4n)/2 = \/r ei%/2 and we then do obtain ; the same value of w with which we started. Fig. 2-5 We can describe the above by stating that if 0 = 6 <2z2 we are on one branch of the multiple-valued function z'/, while if 224 <4 we are on the other branch of the function. z plane It is clear that each ich of the function is single-valued. In order to keep the function single-valued, we set up an artificial barrier such as OB where B is at infinity [although any other line from O can be used] which we agree not to cross. This barrier {drawn heavy in the figure] is called a branch line or branch cut, and point O is called a branch point. It should be noted that a circuit around any point other than z=0 does not lead to different values; thus z=0 is the only finite branch point. RIEMANN SURFACES There is another way to achieve the purpose of the branch line described above. To see this we imagine that the z plane consists of two sheets superimposed on each other. We now cut the sheets along OB and imagine that the lower edge of the bottom sheet is joined to the upper edge of the top sheet. Then starting in the bottom sheet and making one complete circuit about O we arrive in the top sheet. We must now imagine the other cut edges joined together so that by continuing the circuit we go from the top sheet back to the bottom sheet. . The collection of two sheets is called a Riemann surface corresponding to the function 2, Each sheet corresponds to a branch of the function and on each sheet the fu is single-valued, The concept of Riemann surfaces has the advantage in that the various values of multiple-valued functions are obtained in a continuous fashion. The ideas are easily extended. For example, for the function z'/? the Riemann surface has 3 sheets; for Inz the Riemann surface has infinitely many sheets. LIMITS Let f(z) be defined and single-valued in a neighbourhood of z=z) with the possible exception of z=z, itself (i.e. in a deleted neighbourhood § of zy). We say that the number /38 FUNCTIONS, LIMITS AND CONTINUITY [CHAP. 2 is the limit of f(z) as z approaches zo and write lim f(z) = 1 if for any positive num- amr ey ber « (however small) we can find some positive number 8 (usually depending on e) such that |f(z)-l| < « whenever 0 < |z-2| < 8. In such case we also say that f(z) approaches l as z approaches zo and write f(z) > I as z>2. The limit must be independent of the manner in which 2 approaches 2o. Geometrically, if zo is a point in the complex plane, then lim f(z) = / if the difference zmz in absolute value between f(z) and | can be made as small as we wish by choosing points z sufficiently close to zo (excluding z= 2o itself). a2 zeit 0 zat Then as z gets closer to i (i.e. z approaches 1), f(z) gets closer Example: Let f(z) = { to i2=—1. We thus suspect that lim f(z) = —1. To prove this we must see whether the above definition of limit is satisfied. For this proof see Problem 23. Note that lim f(z) * (i), ie. the limit of f(z) as z > iis not the same as the value of f(z) at z=1, ‘since fi) =0 by definition. The limit would in fact be —t even if f(z) were not defined at z=i. When the limit of a function exists it is unique, ie. it is the only one (see Problem 26). If f(z) is multiple-valued, the limit as z> 2 may depend on the particular branch. THEOREMS ON LIMITS If lim f(z) = A and limg(z) = B, then 1. lim {f(z)+g(z)} = limf(z) + limg(z) = A+B 2. lim {f(z)—g(z)} = lim f(z) — limg(z) = A-B z+ zy Brey z+ ky 8. lim fe)o(e) = {lims@)}im of} = AB lim f(z) . F(z) _ BF _ A . 4. limogy = Tmow ~ Bt B*° INFINITY By means of the transformation w=1/z the point 2=0 (ie. the origin) is mapped into w=, called the point at infinity in the w plane. Similarly we denote by z= the point at infinity in the z plane. To consider the behaviour of f(z) at z=, it suffices to let z=1/w and examine the behaviour of f(1/w) at w=0. We say that lim f(z) = l or f(z) approaches | as z approaches infinity, if for any «> 0 we can find M>0 such that f(z) -l| < « whenever |z| > M. We say that lim f(z) = © or f(z) approaches infinity as z approaches 20, if for any N> 0 we can find 3>0 such that |f(z)|>N whenever 0 < |z—20| < 8. CONTINUITY ; Let f(z) be defined and single-valued in a neighbourhood of z=z, as well as at z=Z% (i.e. in a 5 neighbourhood of zo). The function f(z) is said to be continuous at z=Zo if lim f(z) = f(zo). Note that this implies three conditions which must be met in order that f(z) be continuous at z= Zo:CHAP. 2] FUNCTIONS, LIMITS AND CONTINUITY 89 1. lim f(z) = 1 must exist zy 2. (Zo) must exist, ie, f(z) is defined at zo 3. l= f(z) if f(z) is continuous at z. we can write this in the suggestive form by = 5 » < oe — a 5 ot & - 2G Example 1: If f(z) = =; then from the Example on Page 38, lim f(z) = —-1. But fj =0. = zo 0 Hence lim f(z) ¥ f(t) and the function is not continuous at z =i. zai Example 2: = If f(z) =z? for all z, then lim f(z) = f@) = —1 and f(z) is continuous at z = 14. Points in the z plane where f(z) fails to be continuous are called discontinuities of f(z), and f(z) is said to be discontinuous at these points. If lim f(z) exists but is not equal to f(zo), we call 2 a removable discontinuity since by redefining f(%o) to be the same as lim f(z) the function becomes continuous. amv ag Alternative to the above definition of continuity, we can define f(z) as continuous at z= if for any «>0 we can find §>0 such that |f(z) —f(z)| < « whenever |z—2| < 8. Note that this is simply the definition of limit with 1 = f(zo) and removal of the restriction that z¥ 2. To examine the continuity of f(z) at z=, we place z= 1/w and examine the continuity of f(1/w) at w=0. CONTINUITY IN A REGION A function f(z) is said to be continuous in a region if it is continuous at all points of the region. THEOREMS ON CONTINUITY Theorem 1. If f(z) and g 4) are continuous at z=2., so also are the functions f(z) + 9(2), F(z) — g(2), fle) g(z) and A 1 the last only if g(zo)~0. Similar results hold for continuity in a region. Theorem 2. Among the functions continuous in every finite region are (a) all poly- nomials, (0) e?, (c) sinz and cosz. Theorem 3. If w=/(z) is continuous at z=z and z=g(¢) is continuous at ¢= Ly and if ¢,=f(z0), then the function w = g/f(z)], called a function of a function or composite function, is continuous at z= 2%. This is sometimes briefly stated as: A continuous function of a continuous function is continuous. Theorem 4, If f(z) is continuous in a closed region, it is bounded in the region; i.e. there exists a constant M such that \f(2)| 0, we can find 8 > 0 (which will in general depend on both « and the particular point zo) such that | f(z) — f(zo) | < « whenever |z — 20] < 8. If we can find § depending on « but not on the particular point zo, we say that f(z) is uniformly continuous in the region.40 FUNCTIONS, LIMITS AND CONTINUITY (CHAP. 2 Alternatively, f(z) is uniformly continuous in a region if for any «>0 we can find 38> 0 such that | f(z1) — f(z2)|<« whenever |z:— | <8 where z: and 22 are any two points of the region. Theorem. If f(z) is continuous in a closed region, it is uniformly continuous there. SEQUENCES A function of a positive integral variable, designated by f(n) or un, where x = 1,2,38,..., is called a sequence. Thus a sequence is a set of numbers w1, ue, us, ... in a definite order of arrangement and formed according to a definite rule. Each number in the sequence is called a term and u, is called the nth term. The sequence 21, U2,us,... is also designated briefly by {un}. The sequence is called finite or infinite according as there are a finite number of terms or not. Unless otherwise specified, we shall consider infinite sequences only. Example i: The set of numbers i,72,i3,...,7199 ig a finite sequence; the nth term is given by Uy = im, n= 1,2,...,100. ate G+ 7? 2)’ 8p is given by u, = (1+ 1)"/n!, n = 1,2,3,.... Example 2: The set of numbers 1+, is an infinite sequence; the nth term LIMIT OF A SEQUENCE A number | is called the limit of an infinite sequence 1, u2,us,... if for any positive number « we can find a positive number N depending on « such that .|u,—l| < « for all n>N. In such case we write lim wu, = l. If the limit of a sequence exists, the sequence now oO is called convergent; otherwise it is called divergent. A sequence can converge to only one limit, i. é. if a limit exists it is unique. A more intuitive but unrigorous way of expressing this concept of limit is to say that @ S€Quence 1, tz, us, ... has a limit 1 if the successive terms get “closer and closer” to 1. This is often used to provide a “guess” as to the value of the limit, after which the definition is applied to see if the guess is really correct. THEOREMS ON LIMITS OF SEQUENCES If lima, = A and limb, = B, then 1. lim(ant+b,) = lima, + limb, = A+B 2, lim(a,—b,) = lima, — limb, = A—B 3. lim (anbs) = (Jim ax) (im bs) = AB lim @n A 4. limp = Tp, = g if Bxo nN co Further discussion of sequences is given in Chapter 6. INFINITE SERIES Let w1,te,us,... be a given sequence. Form a new sequence S:, S2,S3,... defined by Si =m, S2.= Uitue, Ss = wituetus, °°, Sn = urtuet-e+++tn where S,, called the nth partial sum, is the sum of the first n terms of the sequence {tn}.CHAP. 2] FUNCTIONS, LIMITS AND CONTINUITY 4l The sequence S1,S2,83,... is symbolized by Wtwtwte = Day n=1 which is called an infinite series. If lim S, = S exists, the series is called convergent N= 0 and S is its sum; otherwise the series is called divergent. A necessary condition that a series converges is lim uw, = 0; however, this is not sufficient (see Problems 40 and 150). nwo Further discussion of infinite series is given in Chapter 6. Solved Problems FUNCTIONS AND TRANSFORMATIONS 1. Let w= f(2)=2. Find the values of w which correspond to (a) 2 = —2+i% and (b) 2 = 1-37, and show how the correspondence can be represented graphically. (a) w = f(-24+) = (242 = A4-4i422 = 9-4; (os) w = f(l—3) = (1-312 = 1-614 912 = —3~ 6; z plane w plane y v q.. t 03-4 C36: Fig. 2-6 Fig. 2-7 The point z = —2+4, represented by point P in the z plane of Fig. 2-6, has the image point w=3-—4i represented by P’ in the w plane of Fig. 2-7. We say that P is mapped into P’ by means of the mapping function or transformation w = 22, Similarly, 2 = 1-38: [point Q of Fig. 2-6] is mapped into w = —8—6i [point Q’ of Fig. 2-7]. To each point in the z plane there corresponds one and only one point (image) in the w plane, so that w is a single-valued function of 2. 2. Show that the line joining the points P and Q in the z plane of Problem 1 [Fig. 2-6] is mapped by w =z" into a curve joining points P’Q’ [Fig. 2-7] and determine the equation of this curve. Points P and Q have coordinates (—2,1) and (1,—-3). Then the parametric equations of the line joining these points are given by e2~(-2) _ y-1l _ = a¢_ _ I-(-2) ~ “g=7 = ¢ or a= 3t-2, y= 1-4¢ The equation of the line PQ can be represented by z = 8t-2+4(1— 4t). The curve in the w plane into which this line is mapped has the equation {3t-2+ i(1—4H}? = (8¢-2)2~(1- 4t)? + 2(8t — 2)(1 — 4t)i 8 — 4t — 702 + (—4 + 22¢ — 2402)1 Then since w = u+iv, the parametric equations of the image curve are given by u= 3-—4¢—7t, v = —4+ 22t— 2422 By assigning various values to the parameter ?, this curve may be graphed. w= 242 FUNCTIONS, LIMITS AND CONTINUITY (CHAP. 2 | 3. A point P moves in a counterclockwise direction around a circle in the z plane having centre at the origin and radius 1. If the mapping function is w=2*, show that when P makes one complete revolution the image P’ of P in the w plane makes three complete revolutions in a counterclockwise direction on a circle having centre at the origin and radius 1. Let g—vei@, Then on the cir ate Fe 2nen on © oire. lo | owe Letting (p;¢) denote polar coordinates 2\ *I =1 [Fig 281, r=1 and z=. Hence w = 2 = (ei6)3 = 9310, in the w plane, we have w = pel = e319 go that p=1, ¢= 39, z plane w plane v v Fig. 2-8 Fig. 2-9 Since p=1, it follows that the image point P’ moves on a circle in the w plane of radius 1 and centre at the origin [Fig. 2-9]. Also, when P moves counterclockwise through an angle 6, P’ moves counterclockwise through an angle 39. Thus when P makes one complete revolution, P’ makes three complete revolutions. In terms of vectors it means that vector O’P’ is rotating three times as fast as vector OP. 4. If ¢, and cz are any real constants, determine the set of all points in the z plane which map into the lines (a) u = ci, (b) v = ce in the w plane by means of the mapping func- tion w =z. Illustrate by considering the cases ¢: = 2,4,—2,—4 and c2=2,4,—2,—4. We have w= utiv = 2 = (xt+iy)? = x2-y?+2izy so that u = 2?—y?, v = Qay. Then lines usc, and v=cg in the w plane correspond respectively to hyperbolae e2—y2 = «, and Quy = cg in the z plane as indicated in Figures 2-10 and 2-11. ~~» z piane a” y ya z , / 7 gt—y? = —2 Fig. 2-10 5. Referring to Problem 4, determine: (a) the image of the region in the first quadrant bounded by 2?—y? = —2, zy = 1, 2?-—y? = —4 and wy = 2; (b) the image of the region in the z plane bounded by all the branches of 2?-—y? = 2, zy =1, 2?-y = —2 and zy = —1; (c) the curvilinear coordinates of that point in the zy plane whose rectangular coordinates are (2, —1).(b) (e) . 2) FUNCTIONS, LIMITS AND CONTINUITY 43 The region in the z plane is indicated by the shaded portion PQRS of Fig. 2-10. This region maps into the required image region P’Q’R'S'’ shown shaded in Fig. 2-11. It should be noted that curve PQRSP is traversed in a counterclockwise direction and the image curve P’Q’R’S’P’ is also traversed in a counterclockwise direction. The region in the z plane is indicated by the shaded portion PTUVWXYZ of Fig. 2-10. This region maps into the required image region P’7”U'V’ shown shaded in Fig. 2-11. It is of interest to note that when the boundary of the region PTUVWXYZ is traversed only once, the boundary of the image region P’T’U'V’ is traversed twice. This is due to the fact that the eight points P and W, T and X, U and Y, V and Z of the z plane map into the four points P’ or W’, T’ or X’, U' or Y’, V’ or Z’ respectively. However, when the boundary of region PQRS is traversed only once, the boundary of the image region is also traversed only once. The difference is due to the fact that in traversing the curve PTUVWXYZP we are encircling the origin z= 0, whereas when we are traversing the curve PQRSP we are not encircling the origin. uw = 2 —y? = (2)2?—(—1)2 = 8, v = Qey = 2(2)(—1) = —4. Then the curvilinear coordinates are u=8, v=—4, , MULTIPLE-VALUED FUNCTIONS 6. Let w> =z and suppose that corresponding to the particular value z = z; we have w = w. (a) If we start at the point z: in the z plane [see Fig. 2-12] and make one complete circuit counterclockwise around the origin, show that the value of w on returning to 2: is wie?/5_ (b) What are the values of w on returning to 2, after 2,3,... complete circuits around the origin? (ec) Discuss parts (a) and (b) if the paths do not enclose the origin. (a) (6) (0) z plane w plane v w errs / | \ 1 aes, cA Tis 5 wye q | u wye8ti/5 w,eStt/5 i | Fig. 2-12 Fig. 2-13 We have z= rei®, so that w= 21/5 = rl/seie/s, If r=, and @ = 6,, then Wy = 7i/5 eir/5, As @ increases from 6, to ¢,+ 27, which is what happens when one complete circuit counter- clockwise around the origin is made, we find Ww = TVS gil +2m/5 = pl/5 gi0/5 e2ri/S = yw, e2ni/d After 2 complete circuits around the origin, we find w= PS er t4m/5 = pl/5 G19/5 gtti/S = wy, etti/d Similarly after 3 and 4 complete circuits around the origin, we find w= wy, ebti/d and w= wy, e8ti/5 After 5 complete circuits the value of w is w,elri/5 = w,, so that the original value of w is obtained after 5 revolutions about the origin. Thereafter the cycle is repeated {see Fig. 2-18]. Another method. Since w5 =z, we have argz = 6argw from which Change in argw = 4 (Change in arg 2) Then if argz increases by 27,47,67,87,107,..., argw increases by 27/5, 41/5, 67/5, 87/5, 2v,... leading to the same results obtained in (a) and (5). If the path does not enclose the origin then the increase in arg z is zero and so the increase in arg w is also zero. In this case the value of w is w,, regardless of the number of circuits made.