FIELD AND WAVE ELECTROMAGNETICS DAVID K. CHENG SYRACUSE UNIVERSITY ADDISON-WESLEY PUBLISHING COMPANY Roading, Massachusotts Menlo Park, California Amsterdam Don Mills, Ontario SydneyFIELD AND WAVE ELECTROMAGNETICS DAVID K. CHENG ‘SYRACUSE UNIVERSITY wy ‘ADOISON-WESLEY PUBLISHING COMPANY “Ruan, Mattchusos Meno Park, Catia London” Amsterdam "Co ts orace Senay‘This Book isin the APDISONWWESLEY SERIES IN ELECTRICAL ENGINEERING ‘SPONSORING EDITOR: Toms itobtins PRODUCTION EDITOR: Marilee Sorotskin > TEXT DESIGNER: Melinda Grosser ILLUSTRATOR: Dick Marton COVER DESIGNER AND ILLUSTRATOR: Richard Hangos ~ ART COORDINATOR: Dick Morton i PRODUCTION MANAGER: Herbert Nolan ‘The text of this book wes composed in Times Roman by Syntax Intemational. ‘Library of Congress Catalogog in Publication Data ‘Cheng, David 1 (David Ken, date Feld and wave eesromnagnets. 4 tony A39Sie fg EHOMMaHCIN. 2 Flt tesory Physio) A 1. ite. 5 ; ocraos ssors sizes : : SBN 0251.012394 ‘AACR ’ oven © 1989 by Adion-WateyPubiing Company, tne ‘lig rere No part of his palation maybe repoduand, ore in a even] steno tne Tithe acon Lome OF by any gana electronic, mechanical, phctacopyeg, eesrdoy ee piety, phot ERR be peer writen permisian athe pubisher. Pini in the Unites Seek of Ancona simulieneousy iw Canada, IESE son » fk 246] x | 8h1 tetiectromagnetic Modet 1 1 1 2 “1 -2 3 4 lh Introduction ‘The electromagnetic model ST-units and universal constants Review questions \2 ean 2-1 Introduction Vector addition and subtraction Products of vectors 1. Scalar or dot product 263.2 Vector oF cross product, 23.3 Product of three vectors ‘Orthogonal coordinate systems 2~4.1 Cartesian coordinates 2-42 Cylindrical coordinates 2-4.3 Spherical coordinates Gradient of a scalar field Divergence of a vector field Divergence theorein ‘Curl of @ vector field ‘Stokes's theorem ‘Two null identities 2-101 Identity 1 2-101 Identity IE Helmholtz’s theorem Review questions * Problems 10 u 3 B 15 16 18 uw m4 31 37 5 53 35 35 36 33-1 Intoduction 3-2 Fundamental postulates of electrostatics in free space Coulomb's law TPP S 2.2 Bloctric fila dige w a system of discrete charges 3-32 Bleetric fleld due to a continuous distribution of charge 3-4 Gauss's law and applications 3-5 Electric potential : 3-51 Electric potential due to a charge distribution 3-6 Conductors in static electric field 3-7 Dielectrics in static electric field 3-TA_ Equivalent charge distributions of polarized dielectrigs 3-8 Electric lux density and dielectric constant 3-811 Dielectric strength 3-9 Boundary conditions for electrostatic fields 3-10 Capacitance end capacitors 3-10.1 Series and parallel connections of capacitors 311 Electrostatic energy and forces, 3-114 Electrostatic energy in terms of Field quantities 3H112 Blectrostptic forees Review. questions Problems + . ~~ Aes toni etn : 4-1 Introduction i “4-2. Poisson's and Laplace's equations 4-3. Uniqueness of electrostatic solutions i a4 t 4-4.1 Point charge and conducting planes. 4-42 Line charge and parallel. ‘conducting cylinder: 4-4.3 Point charge and condueting sphere, v SR B 5 % 2 a 91 95 99 108 105 109 Ts uy 520 123 126 128 133 133 139 14 142 148 1474-5 4-6 47 So 52 ol 6-2 6-9 Boundary-value problems in Cartesian coordinates Boundary-vaiue problems in ‘ylindrical coordinates Boundary-value probleros in spherical coordinates Review questions Problems Introduction Current density and Ohm's law Electromotive force and Kirchho#'s voltage law Equation of continuity and Kirchhof's current law Power dissingtion and Joule’ law Boundary conditions for current density. Resistance calculations Review questions Problems i Introduction Fundamental postulates of magnetostaties in free space ‘Vector magnetic potential Biot Savar’s law and applications ‘The magnetic dipole Scalar magnetic potential zation and equivatent current densities | Magnetic field intensity and relative permeability; Magnetic circuits I Behavior of magnetic materials 6-10 Boundary conditions for magnetostatic fields 6-11 inductances and Inductors 6-12 Magnetic eneray 6-121 Magnetic energy in terms of field quantities 150 158 163 167 169 13 a 130 i82 83 (87 19 192 196 197 202252 235 257 259 268 269. mm m2 24 279 28 283 286, 287 288 290 291 292 293 294 298 301 302 307 843 87 8-8 8-9 8-21 Transverse eledtromugnetic waves 8-22 Polarization of plane waves, Plane waves in conducting media sod 8-3. Lowloss dietective ‘ 8-32 Good conductor 8-33 Group velocity Flow of electromagnetic power and the Poynting veetot 8-41 Instantaneous and average power densities “Notmmal incidence at a plane conducting boundary Oblique incidence at a plane conducting boundary 8-6.1 Perpendicular polarization 8-62 Parallel polarization Normal incidence at plane dicleceic boundary ‘Normal incidence at maltiple dielectric interfeces S-8.1 Wave impedance of total field 8-82 Impedance transformation With. multiple dielectries ‘Oblique incidence ata plane gieleczic boundary 8-91 Total reflection 8-9.2 Perpendicular polarization 8-9.3 Puralle! polarization Review questions Problems 9 Theory and Applications of Transmission Lines 94 9-2 9-3 o4 Introduction ‘Transverse electromagnetic wave along a parallel-plate transmission line 9-21 Lossy parallel-piate transmission tines General transmission-line equations 9-3. Wave characteristics of an infinite ‘transmission line 943.2 Trunsmission-line parameters 9-3.3 Attenuation constant from power relations ‘Wave characteristics on finite transmission’ fines 9-41 Transmission lines as circuit elements 9-42 Lines with resistive termination CONTENTS — xii! 32 3 317 318 319 322 326 329 32 336 30 340 342 347 349 350 352 353 356 358 36 363 370 372 315 379 38 385 388 390 3952 . xiv CONTENTS S fede 9-43. Lines with afbf a 04 TW oan 9-44 Transmission-ing circuits 407 9-5 The Smith chart att ur 9-5.1, Smith-chart calculations for lossy lines 420 ih 9-6 Transmission-ine impedance matching | ° 2 i 9-6.1 Impedance matching by quae ‘wave transformer 423 u 9-62 Single-stud matching 426 iw 9-63 Double-stub matching Bi Review questions | 435 un Problems i 47 : ~ ll 10 Waveguides and cavity Resonators ‘ 10-1 Introduction 443 10-2. General wave behaviors along uniform " guiding structures a4 10-2.1 Transverse plectfomagnetic waves 487 10-2.2 Transverse magietic waves “3 10-2.3 Transverse electric waves 452 1 10-3. Parallel~plate wavegijide : 456 10-31 TM waves between parallel plates : 457 10-32. TE waves between parallel plates . 461 10-333, Attenuation in Perales waveguides 463 10-4 Rectangular waveguides | 467 104.1 TM waves in rectangular waveguides 467 104.2 TE waves in rpetangular waveguides an , 10-4.3._Attemriogin rectangular waveguides 475 10-5 Dielectric waveguides 478 10-51 TM waves along dielectric sid . 479, 10-5.2 TE waves along a dielectric slab 483 . * 1026 Cavity resonators ‘ 486 . . £ TM up modes 1 i 407 2 TE yy modes 488 19-63 Quality factor dk cavity resonator 490 Review questions ©” : 493 Problems i 4956 au 20 * aan 4B 426 4a 435 437 aay 488 452 436 487 461 463 460 467 a 475 a8 a9 483, “97, ae 490 493 495, 11-1 Introduction 11-2 Radiation felds of elemental dipoles M-2.1 The elementat electric dipole 11-22 The elemental magnetic dipole 11-3. Antenna paviefff and actenna parameters” + tind This linear antennas 11-41 The halt-wave aipale 41-5 Antennaarrays * 15.1 Two-element arrays 11-S.2 General uniform linear arrays 11-6 Receiving aniennas 11~6.1 Internal impedance and directional patiern 11-62 Effective area A1-7 Some other antenna cypes L1-7.1 Traveling-wave antenna ~~. 11-72 Yagi-Uda antenna 11-73 Broadband antennas 18-8 Aperture Radiators References Review questions Problems Appendix A Symbols and Units ACL Fundamental SI (rationalized MKSA) units Ac Derived quantities A-3- Multiples and submuitiples of units Appendix Some Useful Material Constants B-1 Constants of free space B-2. Physical constants.of electron dnd proton 00 02 502 505 507 515 517 518 52 327 528 530 532 533 535 337 540 sas 545 547 352 552, 554 555 555xv -- CONTENTS BS Relative permittivities Gefviicconstanis)' B-4 Conductivities i Bo. Relative permeabilities“! Answers to Selected Problems | i ‘ : \ eB Index [s 4 Back Endnapere deft L Gradient, divergence, cur, and Laplacian operations Right. Cylindrical coordinates Spherical coordinates f } $56 $56 537 $59 1‘The many books on introductory electromagnetics can be roughly divided into two main groups. The frst group takes the traditional development; starting with the ‘experimental laws, generalizing them in steps, and finally synthesizing them in the form of Maxwell's equations. This is an inductive approach. The second group takes the axiomatic development: starting with Maxwell's equations, identifying each with the appropriate experimental law, and specializing the general equations to static and time-varying situations for analysis. Ths is a deduetive npprouch, A few books begin with a treatment of the special theory of relativity and develop all of electro- magnetic theory from Coulomb's iaw of force; but this approach requires the dis- cussion and understanding of the special theory of relativity frst and is perhaps best suited for a course at an advanced level. Proponents of the traditional development argue that itis the way electromag. netic theory was unzaveled historically (from special experimental laws to Maxwell's cquations), and that it is ensier for the students to follow than the other methods. 1 feel, however, that the way a body of knowledge was unraveled is not necessary the best way to teach the subject to students. The topics tend to be frngmented and cannot take full advantage of the conciseness of vector calculus, Students are puzzled at, and often form a mental block to, the subsequent introduction of gradient, di- Vergence. and carl operations, Asa process for formulating an electromagnetic model, this approach lacks cohesiveness and elegance. ‘The axiomatic development usually begins with the set of four Maxwells equa tions, either in differential or in integral form, as fundamental postulates, These are ‘equations of considerable complexity and are difficult to master. They are likely to cause consternation and resistance in students who are hit with all of them at the beginning of a book. Alert students will wonder about the meaning of the field vectors and about the nevessity and sufficiency of these geneval equations. At the initial stage students tend to be confused about the concepts of the electromagnet model, and they are not yet comfortable with the associated mathematical manip- ulations. In any case, the general Maxwel’s equations are soon simpliied to apply to stati fields, which allow the consideration of electrostatic fields and magneto- static fields separately. Why then should the entire set of four Maxwells equations ‘be introduced at the outset?”I i Mo PREFAC™ : E . : 5 It may be afgued that Coulomb's lay, though based on experimental evidence, Cheep path. Postule: Consider the sw stipulations of Coulomb's wr hag ae Gharzed bodies are very spall coinpared with their distance of sopro Gianee, The question arises repindng the st stpulatons How soll wet we CearBst bodes be in ardar to be considered "very small” compared wath ther dee crasces facies the-chasged bodies cannot be of vans ag sizes (ile! pod Sparies|. and there is difieulty in determining the!"true” distance between eho bea Ot fnite dimensions. Far given body sies the relative accuracy in dstenes meres ‘raat better when the zeparation is targér. Hotiever, practical consideratogs. (weakness of force, existence of extraneous charged bodies, etc.) restrict the usable sistance of separation in the laborstory, and experimental insecurrees aman es entirely avoided. This foads to a more importung question concerning the inverse: square relation of the second Stipulation, Even if the ch liged bodies were: of vanishing Sizes, experimental measurements could not be of at infinite accuracy no matter how skillful and careful an experimentor was. How then was i¢ possible for Coulomb fe,kmow that the force. was exactly inversely pepportional te the square (not the Zeon004th oF the 1.999999th power ofthe disince of separation?“ ie er ite Pesta everywhere. This, foF the evelopment of the cheneaens Fe ge spa iis nly necessary to define a sinale vector (namely the cleceta aM Tatensiy Uy spesiying its divergenge nd its Gul as postularg Ae relations in electrostaties for free space, ibeluding Coulomb's law and Gauss's Jaw, cam be derived trop th two rather simple postulates Relations & nena media can be dev ; Polarized dielectrics. ; m7 Similarly, for the magnetostatic model in, tre space it is necessary t© define Coir @ single magnetic fuxidensity vecior B by ypeiying its dnersevos aod Renan eestolates all other formulas can be derived fipm these es pears Relations in material media an be developed through ‘the concept of equivalent Fee ene ies OF cours! the validity of the postulates lies in thelr soitte re vield results that Spafonn with experimental evidtree rer mewwstvibg flds, the cleric andinagntc Held intensities are coupled. The curl E postulate for the electrostatic model mst be modified to conform with Bndey's law. In addition. the curl B postulate fgr the magnetostatic mee eee ‘so be modified inorder 0 be consisent withthe Lquation of continu We have,Imental evidence, Bb’s law: that the ration, and that Se square of tacit ¥ small must the x with their dis- ‘izes fideal point ween wo bodirs ‘stance measure- {i considerations strict the usable racies cannot be sing the inverse- vere of vanishing arsey no matter be For Coulomb square (not the TY \uestion not ndely that ey iplice, te and that itis 8 ofa limited ue approach in fields (Chapter ns (Chapter 7}, ch states that a ivergence and re electrostatic ly, the electric ates. All other v and Gauss's ons in material Iistributions of ‘sary 10 define teeipsied is wo f sulaies. OF tent hit wvsity to 5 ure coupled. conform with «© model must sity, We have, one aeecenie ce memneeenreturen im PREFACE vil then, the-four Maxwell's equations that constitute the electromagnetic model, 1 believe that this gradual development of the electromagnetic mode! based on Helmholtz’s theorem is novel, systematic, and more easily accepted by students, An the presentation of the material, strive for Jucidity and unity, and for smooth and logical How of kéeas. Many worked-out examples (2 total of 135 in the book) are included to emphasize fundamental concepts and to illustrute methods for solving typical problems. Review questions appear at the end of each chapter to test the students’ retension and wide-standing of the essential material inthe chapter, ‘The problems in cach chapter are desienest ©: reinforce a.tonts’ comprehension of the interrelationships baween the different quantities in the formulas, and to extend their ability-of applying the formulas to solve practical problems. I'do not believe in simple-minded drilltype problems thut accomplish little more than an exercise ‘00 a calculator ‘The subjects covered, besides the fundamentals of electromagnetic fields, include thesty and applications of (ransmission fines, waveguides and cesomators, and antennas and radiating systems. The fundamental concepts und the governing theory of electromagnetism do not change with the introduction of new electors netic deviets. Ample reasons and incentives for learning the furdamental principles of electromagnetics are given in Section 1-1.) hope that the contents of this book, strengthened by the novel approach, will proville students with a secure ond caf” ficient background for understanding and analyzing basic electromagneie phe. ‘nomena as well as prepare them for more advanced subjects in electromagnetic theory. ‘There is enough material in this book for a two-semester sequence of courses, Chapters 1 through 7 contain the material on fields, and Chapters § through )! on waves and applications. In schools where there is only a one-semester course on tlectromagnetics, Chapters 1 through 7, plus the fist four sections of Chapter § would provide a good foundation on fields and an introduction to waves n ue- bounded media. The remaining material could serve as a useful reference book on applications or as a textbook for a follow-up elective course. If one is pressed for time, some material, such as Example 2-2 in Section 2-2, Subsection 3-1!.2 01 clectrostatic forces, Subsection 6-5.1 on scalar magnetic potential, Seetion 6-8 fon maynetic circuits, and Subscetions 613.1 and 6-132 on maynclie forees and {orques, may be omiticd, Schools on a quarter system could adjust the muterial to be covered in accordance with the total number of hours assigned to the subject of electromagnetics. . ‘The book in its manuseript form was clas-tested several timcs in my classes on clectromagnetics at Syracuse University. I would like to thank all of the students in those classes who gave me feedback on the covered material, { would also Tike to thank all the reviewers ofthe manuscript who offered encouragement and valuable suggestions. Special thanks are due Mr. Chang-hong Liang and Mr. Baiclin Ma for their help in providing solutions to some of the problems. Syracuse, New York . DKG January 1983ia netic Model INTRODUCTION Stated ina simgie fisbion, eleceromayneticsis the study of the eects of electric charges at rest and in fotion. From elementary physics We know there are two kinds of charges: positive atid negative, Both positive and negative charges are sources of an slecirc fields Moving charges produce 2 current, which gives rise to 2 magnetic He'd. Here we tentatively speak of eleetcicfctd and magnetic field in a general ways more definitive meahings willbe attached to these terms later. A field is a spatial distribution of a quantify,.which may or may not hed fiinction of time. A time-varying electric file is accompanied by a magnetic Held, and vice versa, In other words, time-varying clectric and aiggndtic lds are coupled, resulting in an electromagnetic field. Under certain condicions, fime-depenident electrotaagnete fields produce waves tat radiate from the soufce. The concept of fields and waves is essential in the explanation of ction at a distance. In this book, Field anid Wate Electromagnetics, we study the Principles and applications of the laws of electromagnetism that govern electro magnetic Bhejomeha. Electrotnagnetits is of fundamental iniportince to physicists and electrical sagincers, Eléciromagnetic theory is indispentable in the understanding of the Principle of atom smashers, cathode-ray oscilloscopes, radar, satellite communicatiou, {slevision reception, remote sensing, radio astronomy, microwave devices, optical fiber communication, instrument-landing systems, electromechanical energy con version, and 9 on: Citcuit concepts represeat a restricted version, a special case, of ‘lectromagnetie coticepis. As we shall see in Chapide 7, when the source frequency 8 very low so that thé dimeasions of a conducting nétwork are much smaller than the ‘wavelengih, We have a quasi-static situation, which simpiifis an electromagnetic Problem toa circult problem, However, we hnsten.fo add that circuit theory is self 4 highly developed, sophisticated discipline. I applits to a different class of electrical ‘engineering problems, and it is certainly important in its own right, Two situalionsiflustrate the inadequacy 6f circult-theory concepts and the need of electromaBactiolield concepts. Figure 1-1 depicts a monopole antenna ofthe type we sor on a Wilkioralkie, Gi transmit, the source ai the busy feeds the antenna with ‘Amessage-carrying curtent ai un appropriate cattir frequency. From a eircuit-theory 12 THE ELECTROMAGNETIC MODEL /+ - ve Fig. 1-1 Amonopoleaitenna, Fig. 1-2 An electromagnetic problem, point of view, the source feeds into an open circuit because the Upper tip of theantenna utd ennected fo anything physically; hence no current would flow and nothing Could happen, This viewpoint, of course, cannot explain why communication san be established between waikieatkies at @ distance. ‘used. We shall se in Chapter 11 that when the length ofthe anton Past ofthe carier waveiength, a nonuniform eurtent will Now alone the enced antcnn, This current radiates a time-varying electromagnetic eld in specs, which an induce current in another antenna at a distance, in Fig. 1-2 we show a situation where an electromagnetic wave is incident from Fed 2r 8 lates conducting wall contzining a small hoe aperture) Electromagnets Selds wilt exist on the right side of the wall at points, such as Pin the gure thee ane Rot necessarily directly behind the aperture Circuit theory is obviously inadesoore hers for the determination (or even the explanation ofthe existence) a the hela oe ‘Die situation in Fi. 1~2, however, represents a problem of practical importance oo 's solutions relevant in evaluating the shielding ellectivenes ofthe conducting wal Generally speaking, circuit theory deals with lumpedcparameter systeme circuits consisting ‘of components charucietized by lumped parameter’ sec ay Teslstiniees, inductanets, and capacitances, Voltages and currents ote le she are vectors with both a magnitude and a direction, and their representation and manipulation require a knowledge of vector algebra and vector calcslus, Even ia Satie cases, the governing equations are, in general, partial diffeential equations. It "The prot ofthe wavelength andthe requncy of an AC sours isthe velocity of wave propagation 12 O oNB thentenna snd nothing nniog. gan be ps be appe- “able Peper aed space, whet cident from romagnetic ate, that are inadequate field atP, portance ag ‘ucting wall systems — ots such as the main ad the gov- cireuits are ordinates ‘hand, most ate any tation and wi in uations. It ropssation, git 1-2 ‘He eueernomatiuric wonen Loe ae one : = o ae he \ Poa ied 3 . | jae is essential that we ai¢ equipped to hafdlelvéctor quamtities and variables that ate both time- ahd sbace-dependent. The funbamentals of vector algebra and vecter calculus will.be déveloped in Chapter 3 Techniques for solving partial differenti! equations ard neolled in gealing with certain types of electromagnetic problems. These " techniques will ba'digeyssed in Chapter 4. The importance of acquiring a facility in the use of these mathematical tools'in the study of electromagnetics cannot be overemphasized. | ‘There are two approaches in the developuifént of @ scientific subject: the inductive ‘approach and thetiedactive approach, Using the inductive approact, one follows tke * Ristorical develoginent of the subject, startig with the observations of some simple experiraents and ihfering from them laws ad theorems. Its provess of seasoning from pacticuter phedowena to general prideipies. The deductive upprosch. on the other hand, postilatesia few funcimental! relations for an idealized model. The postulated reli see axioms, front whieh particu favs anal theorcms ean be derived. Phe validity of the model and the axioms is verified by their ability to prediet consequences thal check with experimental‘observations. In this book we prefer to tse the deductive.or axtomutieappoescn beewuse it is more elegant and enables the development oF Ue anbjpet of electmomegneties in an otcenly way Ie iatiaed model we adopt fr tung ene suds must cle 1 ‘ealworld situatighs and be able to explain physical phenomena: otherwise, we would be engaged in mortal excreses for no puryiose, For example,a theoretical mode could be built from.which one might obtain thay mathematical relations: but, if these ‘lations disagite with observed results, the lode is of no use. The mathematics may bbecorrect, bit the inderlying assumption of Wi8 model may be wrong or te implied approximations nfay not be justified. Three esskntial steps are involved in build a theory on an idealized model First, some basie quantities germane to the dubject of study are defined. Second. the rules of operatios,(the mathematic) of these quantities are specitied. Thin, some fundamental felatpns are postulated. These postulates or laws ace invariably based on numerous experiments) observations acduired under controlled conditions and synthesized by brilliant minds. A femiliar example is the ciccuit theory built on & citcuit model of ideal sources and pure resstnces, inductances, and capacitances 13 this case the basic quantities are voltages (),curcent (0), resisianoes R) inductances (Ly, and vapacitatces (C); the rules of operations are those of algcbra, ordinary differential equaticins, and Laplace transforntation; and the fundamental postulates are Kirchhof’s vollage and current iaws. Many relations and formulas can be desived {rom this bisicatly fathoysiniple model, aad the retponscs of very elaborate networks can be determined, The validity and value of tite mode! have been amnply demonstrated Jn a lik manner, gn electromagnetic theory ean be built on a suitably chosen sfectromagnetic model. ta this section we shall take the first step of defining the basic4 THE ELECTROMAGNETIC MODEL /1 ‘Guantities of electromagnetics, The second step, the rules of operation, encompasses * Nector algebra, vector calculus, and partial differential equations. The fundamentals and the techniques for salving parti! diferentin! equations will be introxtuced when these aquationy arise ater nthe how The din! step, the fisdamnental poste lt beget in tes substsa in Chapters 3c, and 7s we deal with, respectively, staticelectric fields, steady magittic feldsand electromagnetic Relds, Je quantities in our eleciromagnctie model can he divided roughly imu two Galoeorivs: source Geld quantities. The souice of an electromagnetic hele canes saat charges a rest or in motion, However, an electromagnetic eld may SAN a edistribution of charges which will, in turo, change the feds hence, ake Separation between the cause and the effect is not always zo distinet. ease the symbol q (sometimes 0) to denote eletre charge. Electie charge is @ fndamental property of matter and it exists ony in postive oF mesativg integral ‘multiples of the charge on au eoetenn. v? : 1.60 10° (C), a-1) fhtere C is the abbreviation of the unit of charge, coulomb! It is named after the catah nltysicist Charles. de Coulomb, who formulated Coulomb's lor i 178s, Coulomb's iaw will be discussed in Chapter 3) A coulomb is avery large oni for Tate chases; il takes 1(1.60 x 10°") or 625 billion electrons to make apt In fact two L-C charges 1 m apart will exert a force of approximately | milion tov gist ether Some other physical constants for the electron ave listed in Appesel B-2, “The principle of conservation of electric charge, like the principle ofconservation of momentum, is a fondamentat postulate or law of physics. It states thet oecee charge is conserved; thats it can neither be created nor be destroyed. Electiecherecs Gah move from one place to another and can be redistributed under the influence of Sa.cketromagnetc Beld; but the algebraic sum of the positive and negative charges Be ose solace) system remains unchanged, The principle of carsereationof dee rl charge must be satisfied at al sines ad under any circamstancen is teprecceted Rathematically by the equation of continuity, which we will discuss In Section $4 Af formulation or solution of un clecromagnetic problem that violates the principle of Conservation of electric charge must be incorrect. We recall that the Kéchhate Satreet law in circuit theory, which maintains thet the sum ofall the currents leaving ‘Junction rust equal the sum of al the currents.entering the junction, is an assertion uh ,282 Muay Ge Mann byothesied queria he bese bung back of waver Quarts were Reg ari focus a feacton ofthe chars. ofan electron; it, to dal thes eaten een a Fe experiment * The system of une wil be aiscusse in Section 1-3 | | |compasses. ° Samentals Analysis), ced when, alates, will spectively, 1 into two le field [2 field may henge, the charge is 8 ve integral uy XY dat the in a. xe unit for pup 16. nillion vons Appendix ‘riccharges influence of sive charges ion of elec representeat an 5-4 be principle Kirchhoff’s: ents leaving an tion oars sere been wet i t t Leeee St aw OR oe “120 THE ELECTROMAGNETIC MODEL 3 se . of the coneriaiod roperty of elecirle chdge, lthplicit in the current law is the assumption tliat thkre iso cumulation| of effurge at the junction.) 4 Although, in almicroscopic sense, electriccharge either does or does not exist at point ina discrete these abrupt afiaijons on an atomniesealeare unimportant ‘when we consider fhe electromagnetic efiéeid of large aggregates of charges. [n com structing a mactosfopic or large-scale thteor) of electromagnetism, we find that ihe use of smogthed:dut average density fenetids yields very good results. (The same approach is used id mechanics where a gaoothed-out mass density function is defn in spite of the fact that sas is associated onl) wits elementary particles in a disorete manner on an atomic scale.) We define a voluhé charye densi, , asa source quantity asfollows: yt md * ‘ MG ea : om tin $2 (Cin), wa where ag isthe ambunt charge ina vely sult volume As. How small should de be? It should be small enough to represent an accurate variation of p, but large enough 16 contain avery inege numberof dseretepharats For exnmple, an slemental cue wits ‘sides as small as { micron (10-* m or 1 fem} has a volume of 107'* m3, which will still contain about 10" (100 dilon} atoms. A smoothed out fncion ofsoacec cued , defined with such 3 small Av is expected to yield accurate macroscopic results fat nearly all practical purposes, Jn some physigal situations, an amilunt bf charge Ay may be identified with an slement ofsurfuce As orsnclement of line An such eases it wil bemore appropriate 10 define a surface charge density, e078 lind charge densi), pr : a tin $4 jm; ws) bo imi SG a em fan SL Ci a4) Except for certain bpecat situations, charge densities vary from point to points hence P. Py 8nd py ate, itt gereral, point functions of space coordinates, ‘Current is the rate of change of charge with respect to time: that is, L da 5 i Tai (Chor ay tls) where /itselfmay Be time-dependent. The unit of current is coulomb per second (C/s), Which isthe safe &s ampere (A). A curzent rhust flow through a finite area (a cone ducting wire of & filtite etoss section, for instakicé); hence itis not a point function, It electromagnetics we define a vector point function volwne current density (or simply, current density) J, Which mmessures the amoustt of current flowing through a unit ares normal to the dirtdtion of current flow. The bold-faced J isa vector whose magnitudé is the current per ubit area (A/m) and whote diection is the direction of current Sow, We shall claboraté on the relation between I and J in Chapter 5. For very good6 THE ELECTROMAGNETIC MODEL/1 . conductors, high-frequency alternating curtents are confined in the surface layer, in- stead of flowing throughout the interior ofthe conductor. In such cases there is @ need to define a surface current density J,, which is the curvent per nit width on the con= ductor surface normal to the direction of current flow and has the unit of ampere per meter (Aji). , There ars for fundamental veetor field quantities in electromagnetics. electric Ji intensity Band magnetic field intensity Mh The definition and physical sigaiicaive of these {quantities will be explainied filly when they gre introduced later in the book, ALthis ime, we want only to establish the following. Electric field intensity E is the only vestor needed in discussing electrostatics (effects of stationary electric charges) in free space, and is defined as the electric force on a unit test charge. Blectric displacement vector D is useful in the study of electric iid in material medta, as we sball seein ‘Chapter 3. Similarly, magnetic flux density B is the only vector aeeded in discussing nagnetostatics (effects of steady electric currents) in free space, and is related to the magueti¢ force acting on a charge moving with a given velocity, The magnetic field intensity vector H is useful in the study of magnetic field in material media. The definition and significance of B and H will be discussed in Chapter 6 The four fundamental electromagnetic field quantities, cogether with their units, are tabulated in Table 1~1, In Table 1—1, V/s volt per meter, and T stands for tesa of volt-second per square meter, When there is no time variation (as in static, steady, ‘Table 1-1 Electromagnetic Field Quantities Symbols and Unite | for Field Quantities Field Quantity ‘Electro eld intensity Electric Electric Bux density (Elecirie dispiacemens) - ‘Magnets ux denaiyy 8 T Magnetic ‘Magnetic eld intensity 8 Alm or siationary cases) the electric field quantities E and D and the magnetic field ‘quantities B and HI form two separate vector pairs. In time-dependent cases, however, electric and magnetic fteld quantities are coupled; that ix, time-varying E and D will sive rise to Band H, and vice versa, All four quantities are point functions: they are defined at every point in space and, in general, are fonctions of space coordinates. Material (or medium) properties determine the relations between E and D and between Band H. These relations are calle the constitutive relations ofa medium and will be examined later : ectric ly density (or eleetriv displacement) Da mainetic fis density Cface layer, ine here isa need hon the cone of ampere per tics: electric fe fs density ance of these ‘book, At this Ess the only sargest in fie displacement w shall see in in discussing elated to the sagnetic field Vm The hue” nits, ands ws tesla statis steady, will yas Leng are coordinates. and D and ‘medium and og 1-2,,S1 UNITS AND UNIVERSAL CONSTANTS — 7 ‘The principal objective of studying elgctfomagnetism is to understand the inter action between charges:and currents at a. distance based on the electromagnetic model. Fieldsandt waves (lime- and spave-tependent elds)’ are basic conceptual Guantities ofthis nfodel. Fundamental postulates will relate E, D, B,H, and the soured quantities; and. dived relations will lead, t0 the explanation and prediction of electromagnetic phenomena. ‘ : 1-3 SI UNITS ANO™ UNIVERSAL CONSTANTS Amex: ny viysiaal quantity must de expressed as a number followed by ‘unit. Thus, we may talk about 2 length of thfee meters, a mass of two kilograms, and 2 time-period of ten seconds. To be useful, d unit system should be based on somé fandamental units of convenient (practical) sizes, in mechanics all quantities can be ‘expressed in terms Bf three b for length, mass and time}. In electromagnetics work a fourth hasie unit ov eucvent) is needed. The SU {Laternurionad System of Cats oF Le Spsiewe Anlerauonale dUnités} saa MESA syste built from the Four fondamental uits sted ia Table 1-2. All other units used in electromagnetics, including those appearing in Tablo 1-1, are derived units expressible in terms of 'm, kg.s,and A. For example, the unt for charge, coulomb (C)is ampere-second As); the unit for electric field intensity (V/mn] is kg!m/A-s*; and the wnit for magnetic Nox density, tesla (7), i ka/As#, More complete'tables of the units for various quantities are given in Appendix A, In our electromagnetic mode! there are three universal constants, in addition to the field quantities listed in Table 11. They felate to the properties ofthe ftee space (vacuum). They are as follows: velocity of electcomagnetie wave (including light) it fice space, &; permltisity of fice space, €95 and permeubilty of Iree space, jig, Matty ‘experiments have been performed for precise measurement of the velocity of light: to many décitnal places. For our purpose, itis sufficient to remember that ems x10" Gag U6) ‘Table 1-2 Fundamental St Units8 THE ELECTROMAGNETIC MODEL /1 ‘The other two constants, ¢9 and xo, pertain to electric and magnetic phenomena respectively: ep is the proportionality constant between the electric Sux density D and the electre field intensity Fin free space, such. that Dabs on) 4» is the proportionality constant between the magnetic field intensity H gad the magnetic flux density B in free space, such that [He 1-8 [Hed oe ‘The values of» and ig are determined by the choice ofthe unit system, and they are not independent, In the ST system (rationalized" MESA system), which is almost ‘universally adopted for electromagnetics work, the permeability of free space is chosen to be Ho = a x 1 Him), 9) where H/m stands for henry per meter, With the values of cand jp fixed in Eqs (1-6) and {1-9} the value ofthe permittivity of free space is then derived from the following relationships c= (nis) a-10) or Tod igs Com a Bag * 10 8854 10" F/m), corel) where F/m is the abbreviation for farad per meter, The three universal constants and their values are summarized in Table 1-3 ‘Now that we have defined the basic quantities and the universal constants of the electromagnetic model, we can develop the various subjects in electromagnetics. But, * This sytem of uote ssa to ba vaionalized because the factor dx doesnot appear inthe Maxell ‘equations (he fundamental postulates of eietromagnetsin} This faioe however, wil appear in many lensed relations. Js the woratonalled MKSA system, uy would be £0~" (Hm) and the Bucor 4x would uppest inthe Maxwells vation,cpheadmena ux density D ant) y H and the G8) fem, and they ich is almost ‘tee space is omy fn Has. (1-6) he following (1-10) aay onstants and he Maawelts| pase i may ‘Shor de would REVIEW QUESTIONS = @ Universal Constants >) nk ‘eos offigitin tae pace me Pemseabiliy fee baie Hm Permit ites sce Fim before we doatat, ve mst be equipped withthe appropri’ mathematical tool, In the following chapter, we discuss the base rues of opetation for veetor algebra and vector calculus. / - REVIEW QUESTIONS: RAT What is clecroragneties? R.A-2 Desertig two phenomena oF steations, ote thn those depicted in Figs tl and 2 thot cannot be idequttely explained by circuit tery. RA-3 What rg the three essential steps in building an iealizud model for the stady ofa scientific subject? t Ru-4 What are the four hindsmental SI units in lectromagnetios? Ru-S What ie the four fustdamental field quantiles inthe eletromagnetic modet? What are theie units? : eel i RA-6 What are the three ttiverst constants ia de electromagnetic model, and what are thie relations? . RA-T What are the fouree quastities in the elecerdmagnetic model?24 INTRODUCTION ‘As we noted in Chapter 1, some ofthe quantities in clectromagneties (such ns charge, current, energy} are scalars; and some others (seh as electric and magnetic fiekd intensities) are vectors, Both scalars und vectots can be functioiiSef time and position, ‘Ata given time and position, a scalar is completely speciied by its magnitude (positive OF negative together with its unit), Thus, we can specify for instance, a charge of ~ 1 uC at a certain location at t= 0, The specification of a vector ata given location and time, on the other hand, requires both « magnitude and a direction. How do'we specify the diection of a vector? tn a three-dimensional space three numbers are needed, und these numbers depend on the choice of a coordinate system Conversion of given vector from one coordinate system to another will change these mumbers, However, physteal laws and theorems relating various scalar and vector quantities certainly must hold irrespective of the coordinate system. The general expressions of the laws of electromagnetism, therefore, do not require the specification of a co ordinate system. A particular coordinate system is chosen only when a prabiem ofa given geometry is to be analyzed. For example, if we ure to determine the magnetic field at she center of a current-cacsying wire loop, itis more convenient to use Tee- angular coordinates if the Joop is rectangular, whereas polar coordinates (two. dimensional) will be more appropriate if the loop is cireular in shape. The basic cleetromagnetic relation governing the solution of such a problem is the same for both geomesies. ‘Three main topics will be dealt with inthis chapter on vector analysis: 1. Vector algebra —addition, subtraction, and multiplication of vectors 2 Orthogonal coordinate systems—Cartesian, cylindrical, and spherical coordi nates, 3. Vector calculus—differentiation and integration of vectors; line, surface, and volume integrals; “det” operator; gradient, divergence, and eur! operations. ‘Throughout the rest ofthis book, we will decompose, combine, differentiate, integrate, and otherwise manipulate vectors, It is imperative that one acquire a facility in vecor 10 2-2 ' VEC ‘AND SUE ¥sechaschatse, mages eld nd postion ius postive sartpe ot ot ke ni lowe 1. Conversion fone Or quantities expressions of ion of a co- problem of a the magnetic TE LO use ree dinates (to- pe. The basic the same for ors, icpcoordi- suf and rations, se, integrate, ity in vector Re yéctor apomon ano sustRactiOn 11 t i algebra and vecio§ claus. tn a thre‘ditaénsiona space a veior relation iin act {three scalar celatiohs Te use of wector-gnal sis techniques in electromagnetics leads 10 concise and! eldgent lormulations. A defiiency in vector analysis in the study oF lectromagnetiés ff similar to a deficieney Jn algebra and calculus in the study of physics; and itis govious that these deficiendies cannot yield fruitful results, In solving prdtticat probiems, we elway dédi with regions or objocts of a given shape, and iti névessary to express general otinglas ina coordinate system appro- priate for the piykn geometry. For example Nb familiar rectangvlar (x.y; 2) <0 orfinates are Bbiously, awkward 16 yse fof problems involving = circular ylinget of a sphere fecadse ihe boundaries ofa clfevlar Gylinder and ‘sphere cannot be describad by Eonstant values of x,y, and z fh this chapter we discuss the three moct commoniy-tsed dlthogénal (perpendicular) coordinate systems aad the represent. tion and opératioh of vectors in these systems. Familarity with these coordinate systems is essential in the solution ofclectzothagmnetie problems Vector. ctlcullis pertains to the ciferentiation and integration of vestors. By defining certdin differential operators we ean express the bsie his of electro: ‘magnetism i a concise way that is invariant with the choice ora coordinate system. In this chapter we.introduce the techniques for evafuating eilferemt types af iniesrals involving vectors, und define and discuss the various kinds of differential operators 2-2 VECTOR Appinion . Ane SuaTnaction, ‘Weknow that.a vector has a magnitude and a direction. A vector A can be written es asad, where A is the magnitude (and has the tinit and dimension) of A, i and ay is a dimenilonlesé unit vector’ with a unity magnitude having the dizection of A. Thus, Am lAl, od ‘The vector. can bo represented graphically by a diceted straight-line segment of a Jeneth |A| = 4 wih its arrowhead pointing ithe direction ofa,.asshown in Fig. 2. ‘Two veetors aré oqual iFthey have the saane hudgiude und the same direction, even though they muy be displaced in space. Sinde it is difficult to write boldfaced esters by hand, it is a-edmmon practice to use an arrow or a bar over a leticr (A or A) oF "In some books the ial vetr inthe eretion of ie vaio denoted by A, wy ot12° VECTOR ANALYSIS /2 Wei, Fig 2 Graphical representation oF vector A. goukny Ae wide 3 tater 10 distinguish vee oim'a alan, This distin. Buibing mark, once chosen, shoul never be omitied whenven so Wherever vectors ae weitten, Two vectors A and B, wi tions, such as given in Fig. 2 inthe same plane. C = ime direction nor i 5, determine a plane, Their sum A+ Bean be obt in opposite direc- is another vector tained graphically in two ways, 1. By the parallelogram rule: The resultant {eF4mm formedt by A snl B deuivn froin these 2 By the head-to-tail rule: The head of A co Us tector draw fom the til of Ato the head of Bora vectors A. B, ‘triangle, as shown in Fig. 2-240), is the dingonal vet ime poi as shown in the paraltelo- Tis obvious that veetor addition obeys the commutaive and associative iaws, Commutativelaw: AG B=B4A, Associative law: A+ (BLO = (A+B EC. Vector subtraction can be defined in {ertas of vector addition in the following w A-B=As(-m, (2-6) where ~B is the negative of vector ~ Phas the same magnitude us B, but that is, As direction is opposite to that of B. Thus, B= (—ay)p. on ~6)is illustrated in Fig, 2-3, ‘The operation represented by Eq, (2 (9 Two seco, * (©) Pacslietograr rte, Flg.2-2 Vector addition, C= * © Beadiiosait rte In fy beiwe i Aaisnot soteitt to say “the maltiplicatiot fone vector by another” or"thepprod. Uetofteo vectors” because thee are twa dite and very deren types of products oF two vectors. They are (1) scalar or dt. procs, and (2) vector or ezoss products {@) Two vestors: Eo #40) Subuaston of : This distins L mE Neier, A=, Fig 263° Vestor subtraction, sever vectors noe ny ee ee her vector C { Multiplication of f' vector A by a positive sealar & changes the magnitude of A by | ‘times without chilnging its direction (can be either greuter or less than 1) ie parattclo- i : kas agihay. O-8 I : These win! be defined in the following subsections, siative fas, x A ‘ . f 2-34 Scalar or Dot Product * . 2 : 4) ‘The scalar or dot ptoduct of two vectors A and B, denoted by A = B, isa scalar, which e-5) i equals the product of theimagnitudes of A and B and the cosine ofthe angle between owing way i them Thus, . . (2-6) F i A>B SAB 008 O49 2-9) Jeas B, but ° 8 A In Eq, 2-9), thesythbol & signifies “equst by definition” and Oy is the smaller angle en between A and B dnd is less than z radians (180°), as indicated in Fig. 2-4, The dot @ product of two vectors (1) is tess than or equal to the product of their magnitudes; (2) can be cither a positive or & negative qusntty, depending on whether the angle ' between them js snbller or larger than 1/2 radinns (90°); (3) is equal to the product of A ~~ ¢ Amn” Fig, 2-4 Musca the dt Fob tae product ofA and B.Equation (2~t1) enables us to find the’ mag on 14 VecroR anatysis/2.: the magnitude of one vector and the projection ofthe other vector u spon the first one; and (4) is zero when the vectors a ire perpendicular to each other. Itisevident that A aaa 2-10) A= JAR Q-uy ignitude of a vector when the exptession of ‘ector is given in any coordinate system: The dot product is commutative and distributive. “4 foo Mn tte Commutative [aw: A B= Be A, 2-1) Distributive laws A (B+ Ch= A BE Acc. 2-13) The commutative law is obvious from the definition of the dot product in Eq, 2-5) and the proof of Eq, 2~13) islet as an exercise. The associative law does not apply {o the dot product, since no more than two vectors cai be so multiplied and an ax. pression such as A» B+ Cis meaningless, ~~ Example 2-1 Prove the law of cosines for triangle. Sokuion: The law of cosines is a scalar relation: Side of @ triangle in terms of the l them. Refesring to Fig. 2-5, we fi ship that expresses the length of a lengths of the two other sides and the angle bevween ind the law of cosines states that Cu IFSP IRB sa, ‘We prove this by considering the sides as vectors; that is CHA+B. “Taking the dot product of Cwith tel, we have, from Eqs.(2-10)and (2-13). CaO CH (A+B A+B) SACALBBSIAB = AP BT + 24B C08 O gy Fig 2-8 Ulastrating Example 2-1, Not case and Vee Phe ery On oftthe first one! ontthat - 2-109) any expression of in Eg. 2-9), ves not apply and an ex- iY fea aft vale between 13, 2-3/ PRODUCTS OF VECTORS 15, ‘] e fay ped Note that @4p is, Uy definition, the smdlhiy angle between A and B and is equal 16 BBR (180° ~ 2); hence, Bos Ajy ~ cos (180° + jt) cos a. Therefore, pare fat ! a iL) Oma eet papcosa, | i * and the law of costhes follows directly. | } ‘| : ' 2-82 Vector or Cross Phoduet ‘The vector or eros r ipduct « f two vectgrs A and B, denoted by A x B, is a vectdr — perpendicular to tle plane containing A’ and Bits moenitude is AB sin O,,, where : } Gani "he sneer Ihals between # Ynd sin ts ciroction follows that of the > am of the right hand when the fingers rotate front,A to B through the angle 04» (the right i aod rule) |AB cin Oy 14) This sillustrated in Fig. 2~6, Since B sisi @4p is the height ofthe parallelogram formed by the vectors A andl B, we recognize thatthe magnitude of A x B, [AB sin Oa), which is ulways positive, Js numerically equal to’the area of the parallelogram, Using the defittion in £4, (2-14) and following the right-hand rule, we find that ‘A xB. @-15) ¥eoce the cross products nor comamutative, We can see that the eross product obeys the distibatve lew, Bea’ Ax B+ OSAKBLAXC. 46) : Can you show this in general without ésolving the vectors’ into rectangular { components?" f ‘The vector prdduct is obviously no# asbogiative; that is, Ax (Bx OQ # (Ax B)XC. 2-17) Fig.2-6 Cross product of A and B, A x B,2-38 48. VECTOR aNaLysiS /2 The vectot representing the triple product on the left side of the expression above is perpendicular to A and lies m the plane formed by Band C, whereas that on the right sise is perpendicular to Cand isin the plune formed by A’and B The order in witch the two vector products are porformed is therefore vital and int no eve shuld the parentheses be omitted, Product of Three VectJis There are two kinds of privincts of (hee veotary: anmely, (RE Neild iriple prodetet and the eecivr tiple product. The seakat ‘ciple product is much the simpler ofthe two and has the following property: AB x C= B+(Cx A)= CfA xB). + (2-18) Note the cyclic permutation of the order of the three vectors A, B, and C. Of course, AV(BxC)= ~A+(Cx B) ~ = HBHAxC) CB x A). 2-19) ‘Ascan be seen from Fig. 2-7, each of the three expressions itt Eq. (2~18) has a magni- tude equal to the volume of the parallelepiped formed by the three vectors A, B, and. C: The pavallelepiped has a base with an area equal to [B:x C| = [BC sin 8} and a height equal to |A cos 8,|; hence the volume is |ABC sin @, cos 6, ‘The vector triple product A x (B x C) can be expanded as the difference of two simple vectors as follows: Ax (Bx) = BAC) — C{A-B) (2-20) Equation (2-20) is known as the “back-cab" rule and is @ useful veetor identity. (Note “BAC-CAB" on the right side of the equation!) Fig. 2-7 -llustrating salar triple product A (Bx C) i |sssion above is ‘at on the right ‘order in which case should the triple product pler of the two es 5. Ofcourse, sy {2-19} sy ha. = nage 4005 A, Band 2sin Of and erence of wo (2-20) identity. (Note 2-3/PROOUCTS OF VECTORS 17 Example 2-2 Prove ie back-cab rule of vector triple prodvet Solution: Tn order to prove Eq, (2-20), it is convenient to expand A into two ‘components ASASHAL ere Ay and A, are, respectively, paraliel and perpendiculac tothe plane containing B and C- Because the vector representing (B x C) is also perpendicular to the plane the ross product of A, and (B x C* vanishes, Let D = A x (Bx C) Since only Ay is effective hers, ws have Da Ay x Bx. Refering to Fig. 2-8, which shows the plane containing &, Cand Ay, we note ‘hat D lies in the same plane and is normal to Ay, The magnitude of (B x C) is BC sin (0, ~ 6) and that of Ay x (B x C) is AyBC sin (6, ~ 05, Henve, +p = AyBC sin (8, ~ 0) = B sin 0) AyC 098 05) — (€ sin 959/448 cos 94) = (Bay C)— Clay Bi] - ap. D ai ~Ciny. 8) Fig. 2-8 Mluscating the Dack-cab rule of veetr tiple as product, The expression above does not alone guarantee thatthe quantity inside the brackots to be D. since the former may contain & vector that wormal to D (parse fe Ay that is, Day = E+ a, does not guarantee E = D. in general, we can vite BlAy C)~ Clay B)= D+ kay, where k ig a scalar quantity, To deterenine k, we soalar-multiply both sides of the above equation by Ay, and obtain Ay BAY C)~ (Ay? OA BY = Ow AyD + ka? {Tie back-cab rut cn be veried in a srsightforward manner by expending the vetot inthe Cotesian «coordinate system (Problem P28. Only thot interested a genta poo aera to stay ta soe18 VECTOR ANALYsiss2-°. 9) 4" i Since Ay; D = 0,80 k =O and D = Bea — Clay! B), Which proves the back-cab rule inasmuch as Ay Co AE and Ay Be A-B, Diciion by a vector is not sfied, andl expressions suc as K/A and B/A are meaningless, : 2-4 ORTHOGONAL COORDINATE SYSTEMS Ni Mavs indicated before that although the laws ofiéctromagnetism ore invariant TaA Spordinate system, solution of practea! probes requies thatthe veka erived from these laws be exprasei! in a coordinate systean appropriate (9 the Fra aap Uk tiven problems. For espe ifwe are to determine the elestrs fed arenes ar described by ts = constant ay = comstan andar coos Where e's need not al be lengths (in the familar Cartesian gr entongulan ee Pate ostomy 4g, and us correspond to x, and ziespeetiey) When thee ee $Sufacs are mistually perpendicoar to bre another we have an vthogwal aches Frostean MeHoeOMA! coordinate systems are not used fecaune they comiane problems, : Some suices represented byw = constant = {2,03} in a edotdinate sytem recon, ey OaRes: they may be curved surlucn Lit ay ae and ae te ee Sian inthe three coordinate directions, They are caley he bus" hoea fot general igt-handed, orthogonal, curvilinear coordinate system the base sears e ‘azranged in such a way thatthe foliowing relations ses caren: AF as a \ Oa aa, i ata, =a, : 2-2) These three equations are not allindependent, ‘s the specification of one automatically jinplies the other two. We have, of coutse. ‘ WB aha, wa, a,j 0 (2-22) and i Bh RAG a a rale 1, 2-23) deer A can be written as the sun of ts gomponents in the three orthogonal Seay »sre invariant be relations siate to the sloctrie eld F the source ‘mal space a sat the three tant, wk orale 2 thes rhree des nae compen nate system be the unit ectors. In a vectors are 2-21a) ~ amb) (2-21e) momnatically \ ik i i i t TRE RAEN sete a 2-4/ ORTHOGONAL COORDINATE SYSTEMS 19 ‘where the magnitudes of the three components, Ayy 4, ind 4,,, may change with the location of A; that is, they may be functions of wy, 4p, and iy, From Eq. 225) the magnitude of Ais . A=|Alm (ai, +43, 4 4)", 2-25) Example 2-3. Given three vectors A, B, and C, obtain the expressions of (a) AB, (0) AB, ond (¢) C:(A x B) in the orthogonsl curvilinear coordinate system (ay gst) | Solucion: First we write A, Bai C inthe orthogonal coordinates (uy, uy. Ae ay + BA By Any BaBes * Baas Cm Gy, ale FAC 8) AB log day FB ada Bay) lB, + 4.By, AB) = Aa, Bar + AusBas + AnsBass & in view of Egs. (2-22) and (2-23), 26) BY A Be (aA, + ata t tia) X Bay + a, By, + Bf) Ages ~ A Bag) * ul AnD, ~ Ayla) + Ay ay ~ AggBa,) fe tay a, = A, Ae Av 2-27) I,m By Ezqutions (2-26) ind (2-29) express, respectively, the dot and cross products of two vectors in orthogonal curvilinear coordinates. They are important and should be remembered. ©) The expression for C- (A x B) can be written down immediately by combining the results in Eqs. (2-26) and (2-27) CHA B)= Caldas — AB) + CaltlgBay ~ Au,Bu)) + Cul AesBag — AyBed Jun Cu Cu eld An AL (228) Bay Bu 8,,| Bq. (2-28) caa be used to prove Eqs. (2-18) and (2-19) by observing that @ Permutation of the order ofthe vectors on the left side leads situply toa rearrange- ‘ment of the rows in the determinant on the right side, In vector calewlus (and in electromagnetics work), we are often required 10 Perform Hing, surfaoe, andvolume integenls. ta each ease we ncod to express the20 VECTOR ANALYSIS /2 Giferential length-changs corresponding to a differential ghange in one of the eo fetinates: However, some of the coordinates, say ui = |, 2 or 3), may not bow “gusth; and a conversion factor is needed to convert a diferential change du ino « changein length dé: dec inde, am Pete hs called a moire coeficlent and may itself be a fynction of uy uy, and. For example in the two-dimensianal polar coordinates (uy, s) = (7 #).a diferente Grants (dt) in (=) corsesponds to diferentio length-change df= rap (he = ru) in the aga, }limetion. & dice difleeenféalengtt change fy an arbitrary dtestion can be written s the vector sum of theeoniponent length cianges Wem 0 dls + ay dy bay ty (2-30) Em ats dts) + atts dts) + sits dus), (2-31) In view of Eq. (2-25), the magnitude of dé ig ‘ de = de? + (dey)? + (agp or hy dy Ug dag)? + hy day) B, (2-32). [The diferentia! volume do formed by giferentia! coordinate changes dy, dus, and us in dizections a, a,,, and a,, respectively i(dé, dey 42) ot do = Agltgits dy die diy 2-33) {ater we will have occasion to express the curcént or Sux flowing through a {iferential area, In such eases the cross-sectional area perpendicular te the carrone or ius How must be used, and itis convenient to consider the filferential ares a vector With @ direction normal to the surface; that i, Forinstuns ferent csi inn peroneal fertile oa mag autude ds, the current, di, flowing: through ds must be the. component of J normal to the area multipited by the area, Using the notation in Eq, (234), we can write: ‘simply, . dls Jods ley ta Jea, ds. (2-35) In general orthogonal curvilinear goordinates, the differential aroa ds, normal to the unit vector a, is : as, 8 des dt) "This Cis the symbol ofthe weer ¢, iy oe 1 3 Thene of the co- may not be a ge dy ito a (2-29) tha, and us, <2 diecentat Be des =rdo ehange in an eth changes: 2-39 ean oO 2-22) diy dita, an (2-33) vg through a 2 the eurrent sirea a vector (3) ves ofa mage J normal 19 ‘write simply (33) ‘onmal to the Biden dhs 2-4/ ORTHOGONAL COORDINATE SYSTEMS 21 = aks di dis) 236 Similarly, the differential area normal (o unit vectors a,, and ay, are, respectively, ds; = aby du, des) (37) and ds = ay (his dey dug). | © + 2-38) Many orthogonal coordinate systems exist; bot we shall only be concerned with the three that ure most common and most useful: 1, Cartesian (or rectangular) cooxdinates.* 2 Cylindrical coordinates, 3. Spherical coordinates. These will be discussed separately in the following subsections, Cartesian Coordinates, (ot, 35 My) = (x, 2) A point PCs, yt) in Cartesian coordinates is the intersection of three plones specified by =m, y= yy, and z= 7, as shown in Fig, 2-9. is a right-handed system with base vectors a,, a,, and a, Satislying the following relations: axaaa, (2-39) a, awa (2-390) a,x a,=a,. 2-399) ‘The position vector to the point Plx;, yi, 24) is OP = ax, + a,yy + 0.2, 2-40) ‘A véctor.A in Cartesian coordinates can be written as Ay $a, +a, e-4 3 Ths tera “Caren coordinates” is rfeced beware the erm “rectangular coordicaes” i customaiy ‘ssocated with (wo-dsvensione gomteny.22 VECTOR ANALYSIS /2 plane, Fig. 2-9 Cartesian coordinates, ‘The dot product of two vectors A and Bis, from Eq. (2-26), AVB=AB,+ A,B, + 4,3, and the cross product of A and Bis, from Eq, (2-29), Ax Be adsl ALBL+ 9ADy ~ A,B) + WA, ~ A,B.) a B elas A Be Bo ; (2-43) ayy Since % ¥, and 2 are lengths themselves, all hee thetic eoefictents ace unity; thats fy = he = iy = I. The expressions for the ciferemial length, didesentel ocr and differential volume a respectively, Gm ade +a,dy-+i,dz . \ [dse=a, dy dz ts, = 8, dx dz ds, =a, de dy; dos dx dy de and Fe from Eqs. (2-31), (2-36), (2-37), (2~38), and (2-33) — (2-44) {2-45a) (2-456) 2-450) (2-46) | i2-43) 5 are unity; vential area, 142-33)— * (44) : t "2-4/ ORTHOGONAL coohoWATE SYSTEMS 23 Example 2-4 A scalar line integral of a vector field of the type Pea js of considerable importance in both physics and clectromagnetics. {If F is a force, the integral is the work done by the force in moving from P, to P; along e specitieg bath; if Fis replaced by E, the electric field intensity, then the"ntegral represents en electromotive force) Assume Fs ary + a(x ~y2), Evaluate the scalar line integral from P,(5, 6) to (3, 3) in Fig. 2-10 fa) along the dicect path, P,P,; then (©)along path @) Pi4P,. : Fig. 2-10" Paths of integration (Exampie 2-4), ‘Solution: " First we must write the dot product F din Cartesian coordinates. Since this is a two-dimensional problem, we have, from Eq, (244), Fede (ary + 2,88 ~ yi] (ay de + a, dy) xy dy + Gx — y4ydy. (2-47) 10s important to remember that d@ in Certesin evordigats {6 aps gives by Eq, (2-44) irrespective of the path or the direction of integration, ‘The eireetion ot integration ip taken cave of by ising the proper limi on the meee 4) Along dizect path @)—The equation ofthe path P,2?, is —™ yaie-1) 2-48 ‘This is easily obtained by noting from Fig. 2-10 that the slope of the line PP, 1s. Hence y = (xis the equation ofthe dashed line passing through the ora and parallel to P,P,.Since line P,P; intersects the x-atis atx = ++. its equetion ‘is that of the dashed line shifted one unit in the positive x-direction; it can be: 24° VECTOR ANALYSIS/2 [ i £ ‘obtained by replacing = with (x ~ 1), We have, trom Es. 2-47) and (2-48), fired = Pitvartoe- ay] Pah® Par i . o [ice naxe Pay +3 era 4I= -10 In the integration with respect to y, the relation 3x 2y +3 derived feom Bq. (2-48) was used *¥dy. b) Along path @ — This path has two straight-line segments: From P, to 4 x=5,de«0. Be dé =(15 ~ y*)dv. dy =. Fide = 3: dx, From A to Py: y= Hence, [irk de =if' as ~ dys fh sede Pak@ = [)as—y ave [tarde = 18-24 = —6 ‘ ‘We see here that the value of the fine integral depends on the path of integration. In such a case, we say that the vector field F is not conservative, 2-42 Cylindrical Coordinates 2 ya) 4.2) “In cylindrical coordinates a point P(r, @,, 2.) the intersection of a circular eylin= Srical surface r= r,,a halfplane'containing the zaxis and making an angle $ = $, with the xz-plane, and a plane parallel to the xy-plage at = 2. As indicated in Fig. 2-i1, angle is measured from the positive x-axjs, and the base vector ay is tangential to the cylindrical qurface. The following right-hand relations apply axaea 1 (2-49a} Ae a, | (2-490) la xa may, (2-480)‘integration reutar eyline angle } = dy indicated in vera, is Wha 4 da) 2-490) (2-890) Fig. 2-11 CViindsical ovordinates Cylindrical coordinates ate important for problems with longline charges or eutrents, and in places where cylindrical or circular boundaries exist. The two-dimensional olur coordinates are a special case at = = 0. ‘A vector in cylindrica! coordinates is written as A at, bag, bay 2 50) ‘The expressions for the dot and cross products of two vectors in cylindrical evordi- ‘ates follow from Eqs, (2-26) and (2-27) directly. Two of the three coordinates, r and = (and us are themselves lengths; hence ‘fy = hy o 1. However, is an angle requiring a metric coefficient h, = r 10 convert do to dfs. The general expression fora differential length in eylindrical coordinates is then, from Eq. (2-31): dem a, dr + age dd + ade es ‘The expressions for diferential areas and differential volume are 5, = ar dé de 2-52ay ~ ds, = ay dr dz 2-528) a3, = ay dr dp (2-520) (erwra) do=rdrdp dz,lee 1 ! : , Fip.2-12_ A diereatal EDP sy volume element in cylindrical eT onto i A wpical dilfeentil volume clement at a point (rg, 2) resulting from differential changes dr, d6, and dyin the three orthogonal coordinate directions is shown in Fig. 2-12, ‘ ‘A sector given in cylindrical coordinates can be trangformed into one in Cartesian coordinates, and vice yersa. Syppose we want to expresg A= 3,4, + aj, + acd, in Cartesian coordinates: that ig, We, want to write Aas aydy ¥ 4,4, + acd, and deter mine A., 4,-and 4,,Firstofell, we note that 4,,the =-component ofA, is not changed by the transformation from eybindrical to Cartesian coorginajes. To find 4, ve equate the dot products of both expressions of A with a,. Thus, Ars Ara k Ade et AQily Ayo The term containing 4, disappears here because a. a, = 0, Referring to Fig. 2-13, ‘hich shows the relative positions ofthe base vectors a,, a, 4, and ay, we sve that “a per eosd @sa) and sin 6 (2-35) acnaton(Ere) =! Hence, wy i A 45086 ~ desing, (2-36) Fig-2-13 Relais between Bint ty and ay |» differential is shown in. pin COS ‘emu > Fig 2-13, wwe see that 2-84) (2-55) (2-56) om i Sa eh es 2-4 ORTHOGONAL COORDINATE SYSTEMS 27 Similarly, to find 4,, we take the dot products of both expressions of A with a,: A= Ara, FA ay t Asay ay, From Fig. 2-13, we find (2-37) and (esd) 1 follows that Ay= Asin 6+ Ay cos 2-59) tis convenient to write the relations between the components ofa vector ia Cartesian and cylindrical coordinates in a matrix form: fA) fewse sine" O)Pa] 4,[~ [= o bos i [-| al lo o La. Our problem is now solved except that the cos @ and sin $ in Eq, (2-60) should be converted into Cartesian coordinates, Moreover, dy, Ay, and A, may themselves be fanctions of r $; and'z, In that ease, they too should be converted into functions of %: J and 2 in the final answer, The following conversion formulas aze obvious from Fig 2-13. From cylindrical to Cartesian coordinates xe res ¢ aang yarsing (61d) pez (2-616) ‘The inverse relations (from Cartesian to cylindrical coordinates) ate 2-620) (02) (2.620)28” VECTOR ANALYSIS 2 | M ad Example 2-5 Express the yeetér A= 4(3 008 6) — adr +355 in Cartesian coordinates, Solution: ‘ne 2-6) pay, have : ~ [Fores sing } 07/3 cos g Ge Sa ap 9 oO us or . A= 2,(3 cost) + Br sins) + a4(3 sin 6 cos ~ 2 cos 4) + 9,5" But, ftom Eqs. 2-61) and (2-62), and Therefore, Fi 2-14 Poth for koe ie o a ** (Exainpie 2-4), er : )2-4/ ORTHOGONAL COORDINATE SYSTEMS 29 Solutian: We shall solve this problem in two ways: first in Cartesian coordinates, then in cylindrical coordinates. a) dn Cartesian coordinates, From the given F and the expression for dé in Eq. (2-44), we have Fe dem xy de — 2x dy ‘The equation of the quarter-cirecle it x? + y? = 9(0 = x, y < 3). Therefore, [le de = fo ORK dx —2 ivory o-2y -Lwr +9sinrt “(3 ¥) In cylindrical coordinates. Here we frst transform F into cylindrical coordinates. mY Tnverting Bq. (2-55), we have fe [is ee | {a AL lo 4 Ay cose sing = 1 TaL =j-sing coed 0 | 4, (2-63) 0 o La, With the given F, Eq. (2-63) gives F, cord sing OV xy al-fes se o)/-3] Fy, 0 0 dio de = 9,3 dp i which feads to t t F = aay cos 6 ~ 2x sin 6) — aylxy sin p + 2x 008 ¢) co é For the-present problem the path of integration is along a quarter-circle of a } radius 3. There is no change in r or z along the path {dr = 0 and dz =O); tence é Eq, (2-31) simplifies to : and = lay sin @ + 2x cos 6} dp.30 VECTOR ANALYSIS /2 Becouse ofthe circular path, F, is immaterial to the present integration. Atong (he path, x= 3 cos guand y= 3 sin g. Therefore, J7P de = ff? ~219 sa? # os 6 + cos #40 = Msn 8 + 6 +n gong + -o(i93) - ‘which is the same as before, In this particular example, ¥ is given in Cartesian coordinates and the path is Sireular. There is ao compelling ressori to soive the proiblem in one oF the other coordinates. We huve showin the eonversion of wctors the procediee of soto in both coordinates Example 2-7 Given iky/r + acky2, evaluate the scalar gueface Tmegcal . Gras over the surface of closed cylinder about the -asis specified by z= 43 and r= 2, 8 shown in Fig, 2-15, Solution: | In connection with Eq, (234) we noted that the direction of de ig normal to the surface, This statement is actuplly imprecise because a narmal to a surface can Point in either of two directions. No ambiguity would arise im Fa, (235), since the Choice ofa, simply determines the reference direction of curtent flow. fn the present sase, where F + ds is to be integrated over: a closed surface (denoted by the circle on the integral sign), the direction of ds is always to be taken ap that of the outward normal, Our problem isto carty out the surface integral fleas fia Fig. 2-15 1A cylindrical surface (Example 2-7. FL » 8 The 24.3 “Sph Ap. foestation. Along J the pach is oF the other cof solution ‘is Normal suriace ed Si since the the present he ciscie on fhe outward 2-4/ ORTHOGONAL COORDINATE SYSTEMS 31 ‘over the entire specified surface. This integeal gives the net oupward flux of the vector F through the enclosed surface + The oylinder in Fig. 2~15 has three surfaces: the top face, the bottom face, and. the side wall. So, GF idsm fh eon dee [Reads + feo agas We evaluate the thee integra on he igh side separately @) Top fae z=3,a,ma, Fea be nti ds = ¢ dr do (from Eq, 2-52), fh, Feade= [Pf har drdp « 12s, by Bottom face, indy = 2, ‘which is exuctly the same as the integral over the top face. 9 Side wall. ds = rd@ dz = 2d6 da {irom Eq 2~52a); rac Fred = [8 Pb, dé de 128k, Therefore, GF sds = 12eky + 12ehs + 12Rk, = Talk, + 283), 2-43 Spherical Coordinates (a. tay uy) = (R, 8,8) A point P(R,, 0, $,) in spherical coordinates is specified as the intersection of the following three surfaces: a spherical surface centered at the origin with a radivs R Ry} aright circular cone with its apex at the origia, its axis coinciding with th is32 VECTOR ANALYSIS /2 Y . Fig. 2-16. Spherial coordinates and having aballangle@ © Oy; and a half-plane containing the z-axis and making an anele dw 6, with the xz-planc, The base cector fy at P is radial fom the orgin wel {uite diferent from a, in eslinéical coordinates, the latter being perpendicaiar to the Savis. The base vector a lies in the @ = 6, plane and is tangential to the spherical surface, whereas the base vector ay isthe same a that in cylindrical cookies These are illustrated in Fig. 2-16, For a rightcanded system we have gh ay = ay 2-642) 1X 2g = a 2-640) BEX Besa, 2-64) Spherical coordinates are important for problems jnvotving point sources and ‘eaions with spherical boundaries, Waen an observers very far frain the source regron of finite extent, the iattcr could be considered as the origin gfa spherical coordinate system and, as 2 result, suitable simplifying approximations cow he male, This i {he reason that spherical Coordinates are ined in salving antenna probiews inthe fur field - A sector in spherical coordinates is written as As anda t ayy + agdy 2-65) ‘The expressions for the dot and crqs products of wo vectors in spherigal coordinates can be obtained irom Ege. (2-26) end (2-27), i Jn spherical coordinates, only R(u3) is a length, The other ‘two coordinates, § and 6 (us and wy), are angles. Refers}ag to Fig, 2-17, where atypical differential voluame ‘lementis shown, we ste that metrie coecients hy » Rand iy = R sin 9 are required The aning an ay her? dinates, 2-64a) 2-648) 640) 'S and region inate This is he far 2-4/ ORTHOGONAL CooRDINATE sysreMs 33 37 A diercatil volume element in sphesica! coordinates. eeonNert dO and dg into dé and de, ‘iffrential length is rom Eq. (2-31), [ees an dR + aR ud bape sin ag. } 265) GTS expressions for diferemil areas and difrental volume resulting from ditfer- ‘ntial changes AR, dd, and di in the thiee coortins directions are Respectively. The general expression for a aR? sin 6 dB ap ds, = aR sin 8 aR ag Ay =a, R AR ab do = R sin aR dap. (2-68) For convenience the base vectors, mettic Coefficients, and expressions for the differ- ential volume are tabulated in Table 2-1 A vector given in spherical in be transformed into one in Cartesian ©F cylindrical coordinates, and vice verse is and . == R sind cos 6 (2-69) y= Rin Osing i (2-696) 2= Revs 8. (2-650)34 VECTOR ANALYSIS/2--2 4.) Table 2=-1 The Basie Onhogonal Cooiate Sytess + eile scree Coneien Cyl Spherical Coordinates Coordinates | Coordinates Coordnatesystem Relations | a8 69) mae) ™ a * * Pa Vector a 2 % 2 “ we 8 a ie i 1 . Moris Coeicents fs re , k ' uv Asano een een é aay révdble —— RsinoeRaneo — Conversely, measurements in Cartesian coordinates can be transformed into those in spherical coordinates: Ra JPaP EE (2-708) ‘ (2-700) (2-102) Example 2-8 The position of 2 pgint Fin spherical coordinates is (8, 120°, 330°) Specify its location (a) im Cartesiath coordinates, and (b) in, cylindrical coordinates, Solution: The spherical coordinates gf the given point are R= 8, 8 = 120°, and $= 330, “ an 9) fx Cartesian coordinates, We use Eas (2-682, b x= 8sin IQ" p05 330° =6 y= 8 sin 129° si 380° w 2/5" * Fa B cos 1M ed : ‘Hence, the location of the point is A(6,2./3, — 4), and the position vector {the ‘vector going irom the origin to the pointhis . WP aag ania ‘ ee a & gon Yee enord Exam odin: Solut This & poir all po. Inet: efi 3 gen produ * Recal vector Oe. ©phericat vordinates Kae) Rsing dR a de into "nose 2 my (2-705) (2-106) 120°, 330%, vordingtes. 120", and nN vector (the i coy en 2-4/ ORTHOGONAL COORDINATE SYSTEMS 35 « >) Ineylindrical coordinates. The cylindrical coordinates of point P ean be obtained by applying Eqs. (2~62a, b,c) to the results in part (a), but they can be calculated directly from the given spherical coordinates by the following relations, which can be verified by comparing Figs. 2~11 and 2-16: : re Rsind a-7ia) ono en) z= Reosd, (79 ‘We have P(4,/3, 330°, ~4); and its position vector in cylindrical coordinates is DP aad j3-a4. It is interesting to note here that the “position veetor” of a point in eytindrical coordinates, unlike that in Cartesian coordinates, does not specify the position of the point exactly. Can you write down the position vector of the point P im sphetical coordinates? Example 2-9 Convert the vector A= agdg + Syily + tyy into Cartesian oo ordinates, Suluian: Un this problesn we want (a serite A i the fori fA aya, ay ah ‘This is very differemt from the preceding problem of converting the coordinates of a point. First ofall, we assume that the expression of the given vector A holds for ail poings of interest and that all three given components ay, cp and ay may be fanctions of coordinate variables. Second, at a given point, Ag, Ay, and A, will have dlefinite numerical values, but these values that determine tie disection of A will, in general, be entirely different from the coordinate values of the point, Taking dot product of 4 with a, we have Ags Aa, Aga: a, + Andy! 8, + Agag ay, Recalling that aq” ayy ay a,, and a, a, yield, respectively, the component of unit Wectors ag, ty, and ay io the direction of a,, we find, from Fig. 2-16 and Eqs. (2-694, bo}: x ee 2-73 terme 8 = Fah 7 fa 24 = 605 8008 § en ey) ¢ HOt OF + PEP) ee 2-74)f 86 VECTOR ANALYSIS /2 : Thus, +f - ‘Aco 'An sin 9.608 6 + A, 605 8 005 } — Ay sin Agt Aaxt :” = 485 fo (2-79 PES ATE EES ) Similarly, . y= An sii 9sin d + Ap cos 0 sin + Ag con 2-79 and ‘ Ag Ass Agcos 6 ~ Ay sip @ = —— — “. (2-77) ee, Nt ye + 2 & I Ag, dg.and A, are themselves functions of R, U,and @, they tgo need to be converted into functions of x, y, and z by the use of Egs. (2~70a, 6, ¢} Equations (2-75), (2-76), and (2-77) disclose the fact that when a vector has a simple form in one coordinate system, its conversion into apother coordinate system usually results in a more complicated expression, : Example 2-10 Assuming that g cloud of electrons copfined in a region between two spheres of radii Zand 5 cm hap.q charge density of Saree Cm), find the total charge contained in the region Solution: We have “The given conditions ofthe prgblent obviously point to the use of spherical coordi- tussion for dein Ey.{2~68), we perfor a triple integration. = JPL eae oR? sin Ga a9 de. ‘Two things are of importance here. Fist, since p js givep in units of coulombs per cubic meter, the iimits of integration for R must be converted to meters, Second, the Tull range of integration for 0 ig from Oto w radians.not from 0 to 2x radians. A litle teflection will convince ug that a hulf-ircle (aot a fal-cirele} rotated about the zaxis or 2-5 GRA! pain Os ee We ata are tior. poi rep: dep + dine chi the2-S/GRADIENT OF A SCALAR FIELD 37 through 2n-tadians (¢ from 0 to 2n) generates a sphere, We have ay Qa —3 1078 ff [00° cost sn aR a op ke io Ef 2 = n310 Lf (apg aha) sn 0 cos oy & = =09 x 10°* [2 (con cos! 6 cp 2-76) : sin 26 |= : : mons (SAE te HO, 77) 2-5. GRADIENT OF A SCALAR FIELD omer Ta electromagnetics we have to deal with quantities thot depend on both sme and Mo position, Since thre coordinate vaiabls are involved ina thrce-cmensional space core ‘We expect to encounter scalar and vector fields that are functions of fous variables: fh sade (Gy, ta) Un general, the felds may change as uny one of the four variables changes, ‘We now address the method for describing the space rate of change of a sealar Beld at a given time, Partial derivatives with respect to the three space-coordinate variables are involves ancl inanmieh ax the rate pf chime bay he diferent in cient diese between tions, a vectos is needed to define the space rate of ehange ofa scalar field av piven point and at a given time, Let us consider 2 scalar fonction of space coordinates Vu, us, as), which may Fepresent, say, the temperature distribution in a building, the altitude of a mountain. ‘ous terrain, or the electric potential in @ region. The magnitude of V, in genera, depends on the position of the point in space, but it may.be constant along certain lines or surfaces, Figure 2-18 shows two surfaces om which the magnitude of ¥ is constant and has the values V; and ¥, + dV, respectively, where dV indicates a smal change in V. We should note that constant-V’ surfaces need not coincide with eny of the surfaces that define a particular coordinate system. Point P; is on sueiace 0. Py sambs per cond, the Alittle fae axis & f Fig.2-18 Concerning gradient of sealar.38 VECTOR ANALYSIS /3 isthe corcesponding point on surface ¥; + d¥ along the normal vector dn; and Py is & point close to P; along anpther vector dé x4 dn. For the same change dV in V, the space rate of change, dV, fs obviously greatest along da because dn is the shortest Glistance between the two surfaces." Since the miagnitude of dVde depends om the direction of dé, dV ide isa directional derivative. We define the vector that represents both the magnitude and the direction of the maximum spp rate of increase of sealer 4 the gradient of that scalar, We write . oe a | ray 80,5 0-78) For brevity itis customary ta employ the operator del, represented by the symbol V ‘and write VV in place of grad Thus, | vr a — 2-79) We have assumed that dV is positive (an increase in ); it d¥ is negative (a dectease in V from P, to P;) VV will Be negative in them, civection, + The directional desivative along die Wav dav eee WWF} a, (2-80) Eauation (2-80) states that the space rate of increase of Yin thea, direction is equal to the projection (the component) ofthe gradient of V in that ditection, We can tice write Pa. A) a8 : © vewryae (2-81) where dé = ad. Now, dV.in Eq, (2-81) fs the total differential of ¥ as a result of a change in position {irom P, tp F; In Fig, 2-18}; it can be expressed in terms of the differential changes in coordinates: i Bae OY 2 ein, an pM ay dl a (2-82) i where dé,, dé, and dé are the components of the vector differential displacement <€ in a chosen coordinate system: fh terms of general orthogonal curvilineas coordi. We [BARS otal teateent, changes AVsnd 47 would be used tnd the ratio AY/A/ would become the sevauve d7/d a9 ¢ ypprchel sre, We avai hax in vr of Soap ns Ii Co or the Ins eresind Py is in V, the shortest son the spresents Fa sealar cers (2-80) vis equal ‘eu ako aren sult ofa as oF the oO avanti coordi ecoma the 2-5/GRADIENT OF A SCALAR FIELD 39 ates (i, uz, wy) dé is (ftom Eq. 2-31), dem ade, aa dls +, dy = ahs dy) + a,,(hty diez) + a,,(hy dirs). (2-83) ks instructive to write dV in Eq. 2-82) 28 the dot product of two vectors, a follows: Ww a ew Ba Fe tet El dls de viv oa (nF) ae 84 Comparing Eq. (2~84) with Eq, (2-81), we obtain - vo vw Ween, Ft ta Spt ha ae 2-85) or . (2-46) Equation (2-86) ic a useful formua for computing the gradient of a sealar, when the sealar is given as a function of space coordinates, Jn Cartesian coordinates, (4, sayy) = (3.2) nd hy = hy * hy = 1, we have (2-87) or (2-89) In view of Eq, (2-88), it is convenient to consider V ir Cartesian coordinates as ‘vector differential operator. —— ao, 28 | vaar valde 2-29) ~~, ay as) ‘Looking at By, (2-86) ane is tempted to deline Was “ Ve (a oe pane, BAe irda ty dag * ™ i dig, in general orthogonal coordinates, but one must refrain from doing so. True, this definition would yield a correct answer for the gradient of a scalar. However, the40 VECTOR ANALYSIS /2 1 i same syimbol V has been used canventionally to signify some diferenta! operations of a vector (divergence and curl which we will conser lafer in this chaplen), whore ‘anextension of Vas an operator in ghnéral orthogonal coordinates would be ingornect Example 2-11 Thelectrostatic field intensity Kis derivable as the negative gradient ofa scalar electric potential V; that = ~V¥. Determine Eat the point (11, O)if 2 vate rin : 8 Vm Ros Solution: We use Eq. (2-86) to evaluate E = ~WV in Cartesian coordinates for Part (a) and in spherical coordinates for part (0) ~ % 1,1.0)=(a,+0,2) 4 bay, Thus, B¢ y= (: De Be we Ry Pray 4) it 1 “= ets ( * T+ (7/16) é é a ee adeengeewgatrg] utes (an cos 0 ay sin Ke | In view of Eq, (2-77), the rpsult-above converts very simply to E = —a.¥ in Cartesian coordinates. Thi i nt surprising since'a carefl examination of the siven V reveals that VoR cos 6 is, fact, equal to Ve, In Cartesian coordinates, where ~#| poy i B= Wie. a We) ai 2-6 DIVERGENCE OF A VECTOR Figud { i In the preceding section we considéjed the spatinl derivatives of a scalar field, which {ed to the definition of the gradient. We now turn our attention to the spatial deriv. atives ofa vector field. This will lead to the definitions of he divergence and the curl nit2-6/ DIVERGENCE OF A VECTOR FIELD 41 OE of a vector, We discuss the mening of divergence in this section and that of cur! in erations é Section 2-8, Both are very important in the study of electromagnetism, 2 where £ nthe study of vector fields itis convenient to represent field variations graphically reorrec B by directed feld fines, which are called fx lines or sereamtines, These ate dicected gradient e lines or curves that indicate at each point the direction ofthe vetor fled. The nag. Feo nitude of the field at point is depicted by the density ofthe lines in the vicinity of aa the point. In other words, the number of flux lines that pass through a unit sucfsce : sormal (0 a vector is a measure of the magnitude of the vector. The flux ofa vector k the flow of an iner:mpressible fluid such as water. For a volume e pitas enclosed surice there willbe an excess of outward or inward ow heoUgh i the surFice only when ne volume contains, respectively, a ouree or tai ro pnabssfor 8 4 met positive civésgenee indicates the presence Of a source of Rud inside the volume, Audl « net negative divergence indicates the presence of a sink. The net outward Raw: of the fluid per unit volume is therefore a measure of the strength of the enclosed source, tyne We define the divergence of a vector field A at a point, abbreviated div A, as the net outward flux of A. per unit colume as the volume about the point tends ta aero: diva ig BA ono BD (2-90) ‘Toe numerator in Eq, (2~90), representing the net outward Bus, isan integral over the entire surface $ that bounds the volume. We have been exposed to this type of surface integral in Example 2-7, Equation (2-90) is the general definition of div A Which is a seclar quantity whose magnitude may vary from point t0 point 28 A itself varies, This definition holds for any coordinate system; the expression for div A. like that for A, will, of course, depend on the choice ofthe coordinate system At the beginning of this section we intimated thatthe divergence of a vector is 4 {ype of spatial derivative, The reader may perhaps wonsder about the presence of an integral in ths expression given by Eq. (2-9); bus a two-dimensional surface integral walyin divided by a three-dimensional yolume wil ead to spatial derivatives as the voluine ee approaches zero. We shull now derive the expression for div A in Cartesian cor mote ordinates. nates, Consider a differential volume of sides Ax, Ay, and dr centered about a point (os ¥os Zo) i the eld of a vector A, as shown in Fig. 2~19, in Cartesian coordinates, A=a,4,"ha,d, + 2,4... We wish (6 find div A at the point (xg, yp. 29) Sines the dlflerervial volume has si faces, the surface integral in the aumerator of Eq, (2-90) can be desomposed into six parts, Ny wdc Garis [oe thet ag he he Lan] nel Thee\ 42° VECTOR ANALYSIS /2 Fig, 2519 A diferentis! volume in Chrtesfan econ r+ (2-92) The quantity A,([ R,) . centered at the origin, as shown in Fig. 2-22. Fig. 2-22 A spherical shell ‘region (Example 2-15),48 VECTOR ANALYSIS /2 Solution: Here the speciied region has two gutfaces at Rw Ry and R= Ry At outer surface: R = Ra, is = agR3 sin O.d0 ag; Gon, Feds 2" [ORR sin 8 di m sek Atinner suriace: R = Ry, ds =apR? sin Od8a6s, me Fras — J [SUR ORE sin 9d de 483, Actually, singe the integrand is independent of 0 or:¢ in both cases, the integral ofa constant over a spherical surface is simply the constant multiplied by the area of {he surface 6783 for the outer surico and 48? for the inne surface), and no integer tion is necessary, Adding the two results, we have ius ack — itp, a ~ To find the volume integral we first determine’ V - F for an F that has only an Fy component: : : 1éo le ~ vem ips ae Bone aR Fo ae gO) Since V- F ia constant, its volume integral equals:the product ofthe constant end the volume, The volume of the sill region hetweet, the two spherical surfaces with radii Ry apd R, is 4n(R} RY. Therefore, [9 oF =i- FY m aniche? — BD, as before 4 ® ‘This example shows that the divergence theorem holds even when the volume ‘bas holes—that is, even when the volume is enclosed bya rultiply connected surface, ea 2-8 CURL OF A VECTOR FIELD : In Section 2-6 we stated that a net outward flux ef a vector A through a surface bounding a volume indicates the preéence of sone, This source may be called a + flow source and div A is a measure ofthe strength of the flow source. There is another kind of source, called vortex source, which causes aciteulation ofa vector field around At The ner circulation (or simply circulation) of a vector field around a closed path is defined as the scalar line integral ofthe vector over the, pajh. We have Girevlation of arbund contour € 8 4 A ae, 2-1) Equation (2-112) is a mathematical definition, The physigal meaning of civculation depends on what kind of field ghe vector A represents. IFA is a force acting on an object, its circulation will be the wotf done by the fore in moving the object once th t aro will oe of ar ins the the dice eon ‘ths nie As, "eceR. rilofa oincegra tot ad ses with, « volume d surface, a surlace 2ealled a another aes eu colton Ag on an vject once i t 2-8/ CURL OF A VECTOR FIELD 49 Fig. 2-23 Rotation between a, nd dé in defining car around the contour; if A represen s an citric ild intensity, then the circulation Dil be an electromotive force sound the dlosed path, as we shall see luter in tne book, The familiar phenomenon of water whirling down a sink driin isan example of & vortex sink causing a circulation of fuid velocity. A circulation of A inay enn even when div A = 0 (when there is to flow source) Since circulation as defined im Eq. (2-112) isa line integral of a dot product. is ‘alue obviously depends on the orientation of the contour C eeative to the vector AI order to define point funetion, whieh isa measure ofthe strength of vortex souree, we must make C very small and orient it in such a way that the circulation ie a maximum. We define’ CUA SV eA sina. | In words, Eq, (2~113) states that the cil of a vector field A, denoted by Cut! A or Y x A. 8 a sector whose magnitude is the maximun net circulation of A per wait aren 2s the area tends t0 zero and whase direction isthe normal direction of the area when the area Is oriented to make the net cireulation maximum. Because the normal te aa area can point in two opposite directions, we adhere to the right-hand rule that when {fe fingers of the right hand follow the direction of dé, the thumb points to the a direction. This is ilustraied in Fig. 2-23. Curl A is a veetor point function and Conventionally weisten us V x A (del cross A) although V is not to be considered a Neftor operator except in Cartesian coordinates. The component of ¥ x A in any Other direction a, is a, (V x A}, whieh can be determined from the circulation pet Unit area narzal to a, as the area approaches zero, (03) ots (x A= a, (Vx 4) = tim 1 (G.4-«). ano As, where the direction of the tine integration around the eomtonr C, bounding asea 4s, and the ditection a, follow the right-hand rule "te books pubinbed in Europe the cr ofA ison called ine rottan ofA and wren as rot A50 Fig. 2-24 Determining (¥ x A), We now use Eq, (2-114) 10 find the three components of ¥ x A in Cartesian coordinates. Refer to Fig, 2-24 where a differential vectangular are parallel ts the serplane and having sides Av and Az is drawn about s typical point Pty, ty Wehavea, = a,and As, = Ay As and the contour C, consists of the four sides 1,23, and 4 Thus, ree i ars Fay, 4) ans apaero By Be (Feed 4n Cartesian coordinates A= a,4.-+ a,4, + 2,4 The conjributions of the four sides to the line integral are Side 1: de maf see dz, As dé Also Pi) de where 4, (coro +42, gan beevpanded a Tafor seis: 4 ayaa. A(soro4 Ma re) Alen rods Zia a HOT, a6 where H.O.T. contains the factors (Ay), (App, ete Thus, Finca Ade [abou +norhs eit deter x Side3: dé = a,ds, A+ dex where : Als 50 HOT, 2-118) 7s +Horh-an (2-119) Wo. 38.20 ous As dl= {at Yor 20) ~ 2 4Damtesian lel to the Jor Fah 855,23, a ony Hee four 2-116) (e117) a 2-118) 2-119) egies 2-8/ CURL OF A VECTOR FieLD: 51 Note that dé is the same for sides 1 and 3, but that the integration on side 1 is going upward (a Az change in 2), while thet on side 3 is soing downward (a = Az change in 2). Combining Eqs. 2-117) and (2-119), we have Jagd #e=( H+Hon ie &y ‘The FLO. in a. (2-120) sil contain powers of dy Sima it may be showe at [ge-(Beonon)| svar itn ee oz exe oe 02 Ay Ac. (2-120) Sabsutus'ng Eqs, (2-120) and (2-121) in Eq, (2-115) and ‘noting that the higher. ‘order terms tend to zero as Ay +0, we obtain the x-component of V x A eA, ay (Wx A= ‘ A close examination of Eq. (2. enable us to write down the = and for the cust of A in Caetes 22) will reveal a cystic order in s, 9. and sand miponents of Vx A. The entire expression ins coordinates is ; ] ‘Compared to the expression for V+ A in Bq, (2-104), that for Vx A in Eq. (2-123) is more complicated, a8 itis expected io be, because it is a vector with three vompor sents, whereas VA is a scalar, Fortunately Eq. (2123) can be remembered tattier sasily by arranging it in a determinantal form in the manner of the cross product exhibited in Eq. 2-43), ‘The derivation of ¥ x A in other coordinate systems follows the seme procedure, However, ts more invoived because in curvilinear coordinates not only A but alse 4 changes if iagnitude asthe integration of A «dé iscartied out on opposite sides of curvitinear rectangle, The expression for V x A in general orthogonal curv linear coordinates fay. 13.1 is given below, Jaa» 1ijé ean Ba AL yas$2. VECTOR ANALYSIS/2 |. It is apparent from Eq, (2-125) that an operator form gannot be found here for the symboi V in order 10 consider ¥'% A a cross prpduc, The expressions of VX A am india] and sphenest ordinates ean be cap abtaned row Lr ©1259 be using the appropriate wy, and ig and their metre cbefients hy, ae Example 2-16 Show that ¥ x 420 if 8) A = o4b/r) in eylindrieal coordinates, where kis a epnstamt, or cpordinates, where f{R) is any. function of the radial woe Solution : 8) In eylindricat coordinates the’ fotiowing apply: uy. uss ta) =(r.,2)5 hy = fig = rand ty = L. We have. Rom Fa 2 13 fear Lye waanile (2-126) YA ry hich yes or the given A. : ar aia aye 28 ay Veans 8 lo kg 1) In spherical coordinates the ‘following apply: ty. fig = Rand hy = R Sin Hone, Hy) ROB): hy =U xAeminlS 2 2 ot TAs ROR ig |: een ‘An at Say sn oy Rosin GAs! and, for the given A, i a.+ for the & AVA ‘ 135) by. is a g Fr he cadial a L i i 2-3/STOKES'S THEORES. 53 A cur-free vector tied is called an irrortional or a conservative fied. We wit See in the next chapter that an electrostatic fild is ireotational or conservative, ‘The Skbretsions for V x A given in Eqs (2-126) and (2-127) for cylindrical and spherical Coordinates, respectively, will be useful for ater reference, 2-9 STOKES'S THEOREM For a very srnall differential Vx Ain Eq. (2-113) leads to Wee Ae Aspe acae a2 area ds, bounded by a comour C;, the definition of dn ehtaivung Ka, (2-128), we have taken the dot product of both sides of Ey, (2-113) With a, As; of As; For an arbitrary surface S, we can subdiy smatl differential areas. Figure 2-25 shows such a schem ‘ifferential element, The lft side of Eq. 2128) is the us oft the area As,, Adding the contributions of all the differential ide it into maay, say N, we with As, a6 a typical he vector V x A through areas to the Mux, we have tim Sv x Ayjsidsy = Lav areas 1H Now we sum up the fine integrals around the contour Foprosonted by the right side of Fy, (2128), Since oF two #8 ofall the differential elements the comman frit of the contours lions by ao eotours, the net 10 Ibe Lata Like integra is zero, C bounding the entire area S aadveent elements is traversed in opposite di teonstéhustion ofall he conten parts ie tle ier 1 ‘and only the contribution from the external contour remains afier the summation, dnd (Late) = [acee e Combining Eqs. (2-129) and (2-130), we obtain the Stokes's thenren ivan ea anu enn 30) Fig. 2-25 Subcivided area for proof of Stoke's theorei.54 VECTOR ANALYSIS /2 Which states that the surfuce integral of the curl of.4 vector field over an open surface ‘is equal to the closed line integeat of the vector along the contour bounding the surface As with the divergence theorem the validity ofthe limiting processes leading to the Stoke's theorem requires that the vector field A, as well as its frst derivatives, ‘exist and be continuous both on $ and along C. Stokes's theorem converts a surface integral of the curl of a vector tq'eiline integrai of the vector, and vice verse. Like the divergence theorem, Stokes's {orem is an importarit identity in vector analysis, and we will use it frequently in establishing other theurems and relations in chs. aagnetics st If the suriaee integral of ¥ x Ais cattied over a dlosed surface, here will be no surfuce-bounding extemal contour, and Eq (2-131) tells ws that GB xasraseo a-sy for any closed surfece S, The goumetay in Fig, 2-25 is ehisen deliberately to em- Phusize the fact that a nontrivial application of Stokes’s theowun always implies un ‘pen surface with a rim, The simplest open surface would be Stwordimensional plane or disk with its citcumference.as the contout. We remind ourselves here that the directions of dé and ds (a, follow the ight hand rule Example 2-17 Given F = a,xy ~a.2x, verify Stokes's theorem over a quarter cirewlar disk with a radius 3.n the first quadraat, as was shown in Fig. 2-14 (Example 2-6), : SSolition: Let ws ist find the gurfage integral of Vx F From Eq (2-130), a Sa, vxraia 12 RW xp 2x Therefore 4 : x Pyedsm [° Py xb) ota, day) Len rran fen a “LEP ~aenele ip ePilay th th 2-410 TW Ts w 210.4 te cn O visurface surfiuce, ading to ivatives, + surface st. Like analysis, Bons, in ill bene 2-23 to em plies an quarters plu : 2440/TWO NULL IDENTITIES 55. It is importang to we she proper tnt for the two variables of integration, We can interchange the order of integration as Given a= BL f and get the same result. But it would be quite wrong if the 0 to 3 range were used as ‘the kange of integration for both x and y. (Do you Know why?) For the line integral around ABO.A we have already evaluated the part around the arc from to B in Example 26, ~ From 6 0 O:x eO,and Fede Fo(aydybu 2a dy 20, From 0 to Qand Fs dé = F(a, dx) = xy dx = 0, Hence, Jao. ¥ae= [fF ate 9(043), from Example 2-6, and Stokes's theorem is verified +94] és OF course, Stokes's theorem has been established in Eq, (2-134) as 2 general identity: there is mo need to use @ particular example to prove it, We worked out the ‘example above for practice on surface and line integrals. (We note here that both the veetor fickt and its lest spatial derivatives ave Gnite and continuous on the surface as well ss on the contour of interest) . 2-10 TWO NULL IDENTITIES 210.1 ‘Two identities involving repeated del operations are of considerable importance in the study of electromagnetism, especially when we introduce potential functions We shall discuss them separately below. . Identity 1 co Vxi¥yy=0] (2-133) Las Jn words, the curl of te gradient of any scalar field is identically of ¥ and its Brst derivatives everywhere is implied hero) Equation (2~133) can be proveil readily in Cartesian coordinates by using $9) for ¥ ond performing the imicated uperitions. tn general if we take the Surface integral of ¥ x (VV) over wny surface, the result i equal tothe line integral of VV around the closed pati bounding th surface, as asserted by Stokes’ theorein: Liv xen wag wry ae 2-134) 0. (The existence2102 56 VECTOR ANALYSIS /2 However, from Eq, (2-81), (2-135) ‘The combination of Eqs. (2134) ani'(2-135) states thatthe surface integral of Vx (YH?) over any surface is zoro. The integrand itself must therefore vanish, which leads to the identity in Eq. (2-133), Since a eaordinate system is not specified in the deriva- ion. the identity is general ane dal is invariantwith the choices of coordinate ssstems. . e A converse statement ¢f Idgnity:1 can be madé as fallows., 1 ¢ »otor field Is corre. tien it com be express asthe gradient of @ sealer fed Le a vector eld be, Then, if Vx E = 0, we can define a seaiar eld V such nat eenvy, | 2-139 ‘The negative sign here s unimportant as far as Identity [is concerned. Its inghuded in Eq. (2-136) becouse this relation cdoforms with a base elaion beowcen electie field intensity E and electric scalar potential Fin electrostatics, which we wil take up ‘n the next chapter, At this stage its immaterial what E and ¥ repens) We know from Sextion 2-8 that a curifes vector fed iw conservative fells ence an rote, onal (a canseraative) sector eld ca usays be expressed 5 the graders ofa seclur shel Identity : View also]. 139) Wx ay 139) In words, the divergence of the el of ny vector fe is idenically sro Equation (2-137), too, can be préved easily in Cartesian coordinates by using Eq. (2-89) for Vand performing the sidicated operations. We can prove it in general without regard to a coordinate systém by taking the'volume integral of V» (V > Aj om the left side. Applying the diverzgnee theorem, wichave GPW iA do mf way a, e-09 Let us choose, for example, the arbitrary volume V enclosed’by a surface Sin Fig. 2~26. » The closed surface 5 can be spiif into, two open surfaces. §, and S,, connected by a ‘common boundary which has been drawn twice as C,!and Gy. We then apply Stokes’. theorem to surce $, bounded by C,” and surface Sf bound by Cy. and vite the Fight side of Eq. (2-138) as aot QW Ay dom fe Ayan dst L(V x Alaa ds mG aiuerga “& (2-139)