CARL T. A. JOHNK Engineering Electromagnetic Fields and Waves JOHN WILEY & SONS New York Chichester Brisbane Toronto SingaporeCopyright © 1988, by John Wiley & Sons, Inc. All rights reserved. Published simultaneously in Canada. Reproduction or translation of any part of this work beyond that permitted by Sections 107 and 108 of the 1976 United States Copyright ‘Act without the permission of the copyright owner is unlawful. Requests for permission or further information should be addressed to the Permissions Department, John Wiley & Sons, Library of Congress Cataloging-in-Publication Data Jobnk, Carl ‘T.A. (Carl Theodore Adolf), 1919- Engincering electromagnetic ficlds and waves. Includes index 1, Electric engineering. 2. Electromagnetic fields. 3, Electromagnetic waves. I. Title TK145,J56 1988 6213 87-6159 ISBN 0-471-09879-5 Printed in the United States of America 10987654321 To Jeanette and the BoysCHAPTER 1 Vector Analysis and Electromagnetic Fields in Free Space ‘The introduction of vector analysis as an important branch of mathematics dates back to the midnineteenth century. Since then, it has developed into an essential tool for the physical scientist and engineer. The object of the treatment of vector analysis as given in the first two chapters is to serve the needs of the remainder of this book. In this chapter, attention is confined to the scalar and vector products as well as to certain integrals involving vectors. This provides a groundwork for the Lorentz. force effects defining the electric and magnetic fields and for the Maxwell integral relationships among these fields and their charge and current sources. The coordinate systems em- ployed are confined to the common rectangular, circular cylindrical, and_spherical systems. To unify their treatment, the generalized coordinate system is used. This time- saving approach permits developing the general rules for vector manipulations, to enable writing the desired vector operation in a given coordinate system by inspection This avoids the rederivation of the desired operation for each new coordinate system employed Next are postulated the Maxwell integral relations for the electric and magnetic fields produced by charge and current sources in free space. Applying the vector rules developed earlier, their solutions corresponding to simple classes of symmetric static charge and current distributions are considered. The chapter concludes with a discus- sion of transformations among the three common coordinate systems. 4-4 SCALAR AND VECTOR FIELDS elds can be at each instant in A field is taken to mean a mathematical function of space and time. classified as scalar or vector fields. A scalar field is a function having2 VECTOR ANALYSIS AND ELECTROMAGNETIC FIELDS IN FREE SPACE time, an assignable magnitude at every point of a region in space. Thus, the tem- perature field T(x, y, z, #) inside the block of material of Figure 1-1(a) is a scalar field. ‘To each point P(2y, z) there exists a corresponding temperature T(x, y, z, #) at any instant ¢ in time. The velocity of a fluid moving inside the pipe shown in Figure 1-1 (0) illustrates a vector field. A variable direction, as well as magnitude, of the fluid velocity occurs in the pipe where the cross-sectional area is changing. Other examples of scalar fields are mass, density, pressure, and gravitational potential. A force field, a velocity field, and an acceleration field are examples of vector fields. ‘The mathematical symbol for a scalar quantity is taken to be any letter: for example, 4, T, a, f. The symbol for a vector quantity is any letter set in boldface roman type, for example, A, H, a, g. Vector quantities are represented graphically by @ fen 6 em Gy) Temperature field ” atx = dem Heat source! (a) —S—.DPDAl-oé—’~—0 FIGURE |-I. Examples of scalar and vector fields. (a) Temperature field inside a block of material. (6) Fluid velocity field inside a pipe of changing cross-section.12 vector sums 3 ae i — E 4 a Unit Te” ae vector a 8s FIGURE 1.2. Graphic representations of a vector, equal vectors, a unit vector, and the representation of magnitude or length of a vector: means of arrows, or directed line segments, as shown in Figure 1-2. The magnitude or Iength of a vector A is written [AJ or simply 4, a positive real scalar. The negative of a vector is that vector taken in an opposing direction, with its arrowhead on the opposite end. A unit vector is any vector having a magnitude of unity. The symbol a is used to denote a unit vector, with a subscript employed to specify a special direction. For example, a, means a unit vector having the positive-x direction. Two vectors are said to be equal if they have the same direction and the same magnitude. (‘They need not be collinear, but only parallel to each other.) 4-2 VECTOR SUMS. The vector sum of A and B is defined in relation to the graphic sketch of the vectors, as in Figure 1-3. A physical illustration of the vector sum occurs in combining dis- placements in space. Thus, if a particle were displaced consecutively by the vector distance A and then by B, its final position would be denoted by the vector sum A +B = G shown in Figure 1-3(a). Reversing the order of these displacements pro- vides the same vector sum ©, so that A+B=Bt+A -1) the commutative law of the addition of vectors. If several vectors are to be added, an associative law (A+B) +D=A+ (B+D) (1-2) follows from the definition of vector sum and from Figure 1-3(b). : : 2 i A +B=C | ! ' : ; (a) (b) FIGURE 13. (@) The graphic definivion of he sum oftwo vector (6) The assoia- tive law of addition,4 VECTOR ANALYSIS AND ELECTROMAGNETIC FIELDS IN FREE SPACE 4-3 PRODUCT OF A VECTOR AND A SCALAR Ifa scalar quantity is denoted by u and if B denotes a vector quantity, their product uB means a vector having a magnitude u times the magnitude of B, and having the same direction as B if w is a positive scalar, or the opposite direction if w is negative. The following laws hold for the products of vectors and scalars. iB = Bu Commutative law u(vA) = (w)A Associative law (1-4) (w+ v)A=uA +A Distributive law (1-5) u(A +B) = uA +B Distributive law (1-6 4-4 COORDINATE SYSTEMS The solution of physical problems often requires that the framework of a coordinate system be introduced, particularly if explicit solutions are being sought. The system most familiar to engineers and scientists is the cartesian, or rectangular coordinate sys- tem, although two other frames of reference often used are the circular cylindrical and the spherical coordinate systems. The symbols employed for the independent coordinate variables of these orthogonal systems are listed as follows. 1. Rectangular coordinates: (x, 9, z 2. Gircular cylindrical coordinat 8. Spherical coordinates: (r, @, ) (p, 2) In Figure 1-4(a}, the point P in space, relative to the origin 0, is depicted in terms of the coordinate variables of the three common orthogonal coordinate systems: as P(x, y, 2) in the rectangular system, as P(p, @, z) in the circular cylindrical (or just “eylindrical”) system, and as P(r, 0,6) in the spherical coordinate system. In the cylindrical and spherical systems, it is seen that the rectangular coordinate axes, labeled (x), (9), and (z), are retained to establish proper angular references should observe that the coordinate variable (the azimuth angle) is common to both Rectangular Circular cylindricat Spherical (a) FIGURE 1-4. Notational conventions adopted in the three common coordinate systems. (a) Location of a point P in space. (6) The unit vectors at the typical point P. (c) The resolution of a vector A into its orthogonal components.5 1-4 COORDINATE SYSTEMS reouayds, (ereuds) quesungas uersus9 =¢) (ponupucs) $1 FANE 0 puljk> seinoutD ae inBueyoey @ ‘uapuliéa sejnoi9} UEIsuog = (2uerd) (ouejd) quersuog ==6 VECTOR ANALYSIS AND ELECTROMAGNETIC FIELDS IN FREE SPAGE the cylindrical and the spherical systems, with the x-axis taken as the @ = 0 reference, @ being generated in the positive sense from (x) toward (y). (By the “right-hand rule,” ifthe thumb of the right hand points in the positive z-direction, the fingers will indicate the positive- sense.) The radial distance in the cylindrical system is p, measured perpendicularly from the z-axis to the desired point P; in the spherical system, the Fadial distance is r, measured from the origin 0 to the point P, with @ denoting the desired declination angle measured positively from the reference z-axis to 7, as shown in Figure 1-4(a). The three coordinate systems shown are so-called “right-banded” systems, properly definable after first discussing the unit vectors at P. A. Unit Vectors and Coordinate Surfaces To enable expressing any vector A at the point P in a desired coordinate system, three orthogonal unit vectors, denoted by a and suitably subscripted, are defined at P in the positive-increasing sense of each of the coordinate variables of that system. Thus, as noted in Figure 1-4(6), a,, ay, a; are the mutually perpendicular unit vectors ofthe rectangular coordinate system, shown at P(x, J, 2) as dimensionless arrows of unit length originating at P and directed in the positive x,7, and z senses respectively. Note that the disposition of these unit vectors at the point P corresponds to a right-handed coordinate system, so-called because a rotation from the unit vector a, through the smaller angle toward a, and denoted by the fingers of the right hand, corresponds to the thumb pointing in the direction of a,. Similarly, in the cylindrical coordinate system of that figure, the unit vectors at P(p, #, 2) are ay, ay, a, as shown, pointing in the positive p, 6, and z senses; at P(r, 6, @) in the spherical system, the unit vectors a,, a9, ay are shown in the positive directions of the corresponding coordinates there. These ate also right-handed coordinate systems, since on rotating the fingers of the right hand from the first-mentioned unit vector to the second, the thumb points in the direction of the last unit vector of each triplet Notice from Figure 1-4) that the only constant unit vectors in these coordinate systems area,, a,, anda,. The unit vectors a, and a, in the circular cylindrical system, for example, will change (in direction, not magnitude) as the angle @ rotates P to a new location. Thus, in certain differentiation or integration processes involving unit vector most unit vectors should not be treated as constants (see Example 1-1 in Section 1-6) In Figure 1-4(8), it is instructive to notice how the point P, in any of the co- ordinate systems, can be looked on as the intersection of three coordinate surfaces. A coordinate surface (not necessarily planar) is defined as that surface formed by simply setting the desired coordinate variable equal to a constant. Thus, the point P(x, y, 2) in the figure is the intersection of the three coordinate surfaces x= constant, y constant. z = constant (in this case planes), those constants depending on the desired location for P. (Any to such coordinate surfaces intersect orthogonally to define a line; while the perpendicular intersection of the line with the third surface pinpoints P.) ‘The unit vectors at P(x, 7, z) are thus perpendicular to their corresponding coordinate surfaces (¢.f.4 a4 is perpendicular to the surface x = constant). Because the coordinate surfaces are mutually perpendicular, so are the unit vectors, Similar observations at P(p, >, 2) in the cylindrical coordinate system are appli cable. P is the intersection of the three orthogonal coordinate surfaces p = constant {a right circular cylindrical surface), = constant (a semi-infinite plane), and z= constant (a plane), to cach of which the corresponding unit vectors are perpendicular, thus making a, ay, a, orthogonal as well. Equivalent comments apply to the unit vectors a,, ap, a at P(r, 0) in the spherical coordinate system of Figure 1-4(6),1-4 COORDINATE SYSTEMS 7 wherein the coordinate surfaces defining the intersection P in this instance become 7 = constant (a spherical surface), @ = constant (a conical surface), and g = constant (a semi-infinite plane) B. Representations in Terms of Vector Components A useful application of the product of a vector and a scalar as described in Section 1-3 occurs in the representation, at any point P in space, of the vector A in terms of its coordinate components. In the rectangular system of Figure 1-4(c) is shown the typical vector A at the point P(x, », z) in space. The perpendicular projections of A along the unit vectors a,, a, and a, yield the three vector components of A in rec- tangular coordinates, seen from the geometry to be the vectors a,A,, ayd,, and a,A that figure. Their vector sum, a,4, + a,d, + a,4, = A, thus provides the desired representation of A in the rectangular coordinate system. Similar manipulations into circular cylindrical and spherical coordinate components yield the other two corre- sponding diagrams depicted in Figure 1-4(c), whence the representations of A in terms of its components: A=a,A,+a,d,+a,4, Rectangular A=a,d,+aj4y+a,4, Circular cylindrical A=a,A,+ apd, +aydy Spherical (1-7) Because of the mutual perpendicularity of the components of any of these representa- tions, it is clear that the geometrical figure denoted by each dashed-line representation of Figure I-4(¢) is a parallelepiped (or box), with A appearing as a principal diagonal within each. The magnitude (or length) of each A in (1-7) thus becomes A=[42 + A? + A?}*? Rectangular A= [42 + 43 +.42]"? Circular cylindrical A=[A? + 43 + 43]? Spherical (1-8) C. Representation in Terms of Generalized Orthogonal Coordinates Noting the several similarities in the characterizations of the unit vectors and the vector A in the three common coordinate systems just described, and to permit unifying and shortening many discussions later on relative to scalar and vector fields, the system of generalized orthogonal coordinates is introduced. In this system, uy, #2, us denote the generalized coordinate variables, as suggested by Figure 1-5(a). The generalized ap- proach to developing properties of fields in terms of (w1, ua, 3) has the advantage of making it unnecessary to rederive certain desired expressions each time a new coordi- nate system is encountered. Just as for the three common coordinate systems already described relative to Figure 1-4, the poirit P(w,, wz, u3) in generalized coordinates, as seen in Figure 1-5(a), "Thus, the components of A in the rectangular coordinate system are the vectors a,4,, ay4,, and 2,4, Another usage is to refer to only the scalar multipliers (lengths) ,, ,, and A, as the components of A, although these are more properly the projections of A onto the unit vectors,| 8 VECTOR ANALYSIS AND ELECTROMAGNETIC FIELDS IN FREE SPACE fh inceaing wp «@ Rectangular Circular cylindrical ‘Spherical & o @ FIGURE 1-5. ‘The coordinate surfaces defining the typical point P and the unit vectors at P. is the intersection of three perpendicular coordinate surfaces, uy = constant, uy = constant, i, = constant. The intersections of pairs of such surfaces moreover define coordinate lines. The unit vectors, denoted a, a2, a3, are mutually perpendicular, tan- gent to the coordinate lines, and intersect the coordinate surfaces perpendicularly. The one-to-one correspondence of these generalized coordinate variables u,, 12, uy to their coordinate surfaces, and the generalized unit vectors 1, a2, a3 to the equivalent vee- tors of the three common coordinate systems, can be better appreciated on making a direct visual comparison of the generalized sketch of Figure 1-5(a) with (6), (c), and (d) of that figure. » Tf the vector A were located at the point P(u,, u2, us) in Figure 1-5(a), with the components a,A;, aj4y and ay4y in the directions of the unit vectors shown, the expression for A would be written A=a,4,+a,4, +434, Generalized (1-9) Its magnitude is deduced from the diagonal of the parallelepiped construction for (1-9). A= [At + Ad + 43]? (1-10) ‘The scalars 41, A>, and 43 are called the components of A, Examples of these expressions specialized to the three common coordinate systems have already been given in (1-7) and (1-8).1-5 DIFFERENTIAL ELEMENTS OF SPACE 9 4-5 DIFFERENTIAL ELEMENTS OF SPACE In the processes of integration in space to be considered shortly, the differential ele- ments of volume, surface, and line are frequently needed. A differential clement of volume dv is generated in the vicinity ofa point P(w,, w2, us) in space by means of the displacements d¢,, d¢,, and dé, on the coordinate surfaces, through the differential changes diy, diy, and dus in the coordinate variables. This situation is represented geometrically in Figure 1-6(a). Thus, a volume-clement do is represented in generalized orthogonal coordinates by means of the product of the differential length-elements as follows du = dl, dly de (1-11) The relation of the length-elements to differential changes in the coordinate variables 45 tla, and uy is provided by the relations Uy =hduy dey hyd, dl thy duty (1-12) ye tdu=Conta LE we ie . x=Constant 2 x+de=Constant s+ du, = Constant Constant Rectangular (oy Circular cylindrical Spherical ra @ FIGURE 1-6. The generation of a volume-clement do = dé,d¢,d’, at a typical point in space in orthogonal coordinate systems,10 > vecTor ANALYSIS AND ELECTROMAGNETIC FIELDS IN FREE SPACE so that (1-11) is written do = hyhghy dy dug dus (1-13) The coefficients fy, by, and hy are called metric coefficients, needed to give the elements a of (1-12) their required dimension of length (meter). From a consideration of the geometry of dv in each diagram of Figure 1-6(6), (¢), and (d), it is evident that the following length-elements and metric coefficients are applicable to the three common systems. dt,=dx dl hy = hy =hy = 1 Rectangular (1-14) by dled dy =dp d,=pdp — déy= dz shy =p.hg= 1 Ci b= cular cylindrical (1-15) dy=dr dtj=rdd dey hy =1hy er hy = sin 0a sin @ Spheri (1-16) “The substitution of these results into (1-13) therefore provides the volume-clement do in each system as follows do = dx dy dz Rectangular dv = pdpd dz Circular cylindrical dv =P? sin 8 dr d0.dgy Spherical (1-17) ‘An element ds of a surface S in space may be left in its scalar form ds, although for some purposes it may be given a vector characterization, ds, if desired. Suppose ds coincides with a coordinate surface x, = constant, as shown in Figure 1-7(a). Lh pment yo —— | += Constant 7) ds=apds= b apr? sino dade i : : ~y Be df as = ajdladly ae (x) =i ees) ~ St Zz ee @ oe FIGURE 1-7. ‘Typical surface element ds on a coordinate surface. Note the characterization of ds as a vector element through the multiplication with a normal unit vector. (a) A surface clement ds on the coordinate surface u, constant in the generalized orthogonal system. (8) A surface element ds on the coordinate surface r = constant in the spherical system.16 position vector IL Expressed as a scalar element, ds = dé dés = hyhs dug dus for that example. An illustra~ tion in spherical coordinates is shown in Figure 1-7(6); on the r = constant coordinate surface, ds = 1? sin Od0.d. A vector quality is given ds through multiplying it with cither the positive or the negative of the unit vector normal to ds. Thus, in Figure 1-7(b), the vector surface-element ds = a, ds is illustrated; ds = —a, ds is the other possible choice on the coordinate surface r = constant exemplified. These concepts are partic- ularly useful in the flux-integration techniques discussed in Section 1-9. Differential line-clements are frequently of interest in applications to vector integration. This subject is introduced in terms of the position vector r of spatial points treated in the next section. *4-6 POSITION VECTOR? In field theory, reference may be made to a point P(j,, tz, "3) in space by use of the position vector, denoted by the symbol x. The position vector of the point P in Figure 1-4, for example, is the vector x drawn from the origin 0 to the point P. Thus in rectangular coordinates, © is written reaxtaytaz (1-18) and in circular cylindrical coordinates r=aptag (1-19) while in spherical coordinates roa, (1-20) ‘A further application of the position vector © occurs in the symbolic designation of points in space. Instead of using the symbol P(u,, ua, #3) or P(x, 9, 2), you may employ the abbreviated notation P(r). By the same token, a scalar field F(u,, ta, ts; #) can be more compactly represented by the equivalent symbol F(x, ), if desired. ‘The differential element of length separating the points P(r) and P(r + dr) in space is denoted by the vector differential displacement de. The differential change dr does not in general occur in the same direction as the position vector x; this i exemplified in Figure 1-8(a). (The vector symbol df is sometimes used interchangeably with dr, particularly in line-integration applications.) ‘The differential displacement de (or dé) is written in terms of its generalized orthogonal components as follows. de = dl =a, dl, + a,dl, + agils (1-21) = ayhy duy + aghy dug + aghy dus (1-22) It is illustrated graphically in Figure 1-8(5) by means of the usual rectangular paral- lelepiped construction for a vector in terms of its components. Furthermore, the magnitude d/ of the vector dé is given by the diagonal of the rectangular parallelepiped; thus, dé = [ (hy du)? + (hp dua)? + (hg du)?] (1-23) 2Throughout the text, sections marked with an asterisk (*) may be omitted to conserve time if desiredcere tee eee eee ee eee 12 VECTOR ANALYSIS AND ELECTROMAGNETIC FIELDS IN FREE SPACE FIGURE 1-8. The position vector # used in defining points of space and its differential de. {¢) “The position vector x and a differential position change dr along an arbitrary path, (+) Showing the components of dr in generalized orthogonal coordinates. For example, in spherical coordinates hy = 1, # = 1 and fs =r sin 6, so that (1-22) and (1-23) are written i de = dl =a, dr + aged + agr sin Odp (1-24) with de = [(dr}? + (rd0)? + (r sin 8ag)?]¥? (1-25) The simplest expression for a differential vector displacement df occurs in the rectangular coordinate system, for which, from (1-14), with fy = hy = hy = 1 and with a, =a,, a, —a, and as =a,, the general form (1-22) becomes A dt =a,dx + aydy tad: (1-26) while its magnitude df is written, from the generalized (1-23), as de = [(dx)? + (b)? + (dzy?°J'? (1-27) Similarly, in the circular cylindrical coordinate system, the substitution of (1-15) into (1-22) and (1-23), and with a; =a,, a = ay and a3 = a,, the vector displace- ment dé and its magnitude become dl =a,dp + agpdd + a, dz (1-28) at = [(dp)? + (pag)? + (ae}*]'? (1-29) ‘The position vector # has useful applications in the dynamics of particles such as electrons and ions, for example. A study of Figure 1-8 reveals that if the vector dis- placement dr of a particle occurs in the time interval df, then the ratio dr/dt denotes \ the vector velocity of the particle at P(r). This particle velocity v is defined by the16 PosITION vector 13 derivative of the position vector #(0) de r(i+ Ad — r() v= = lim ——— dt aro at (1-30) A second such derivative of r(i) provides the vector acceleration @ = du/dt of the particle. Because the vector displacement dr of the particle is tangent to its path as shown in Figure 1-8, the velocity » = dr/dt will also be tangent at every point on ¢ = This property of tangency does not hold for acceleration, however, except in purely straight-line motion. The velocity at the point P(r) can be expressed systematically in terms of its generalized orthogonal coordinate velocity components by means of dr , Gm Bis tava + ayes (1-31) v For example, in a rectangular coordinate system, the notations 2, v2, and v3 mean Vy, Vy, and v, respectively. In all orthogonal coordinate systems except the rectangular system, some of or all the unit vectors may change direction as their location P moves in space. A graphical approach to obtaining the spatial derivatives of the unit vectors in an explicit coordi- nate system is described in the following example. EXAMPLE 4-4. Find the following partial derivatives of the unit vector a,: (a) éa,/ér; (b) 08,20; (¢) da,/0p (a) The partial deri ive da,/Or equals zero, since the unit vector a, does not vary in direction with r jor does it vary in magnitude, by the definition of a unit vector) (b) The partial derivative @a,/@0 can be found graphically from the accompanying figure. Ifa, is allowed only the differential change da, in the @ sense, then da, has @ ) EXAMPLE 1-1. (a} Differential da, generated by rotating a, O-vise. (0) Differential da, gen- erated by rotating a, 6-wise.14 VECTOR ANALYSIS AND ELECTROMAGNETIC FIELDS IN FREE SPACE. the direction of the unit vector ay. The length of da, is given precisely by the angle a0, from the definition of angle (arc divided by radius, and the radius is unity), to make da, become éa,], 5 constant whence the desired result is da, 9 |, ~ constant constant (6) The partial derivative éa,/@@ is found similarly from (5) of the figure. Allowing only the change dé in the position of a, generates the differential vector da,, having a direction specified by the unit vector ay and a magnitude given by d@ sin 0. This makes da, (for r = constant, = constant) become a, sin 6d as shown, whence a S = a, sind og By means of graphic techniques similar to those used in Example 1-1, one can show for spherical coordinates that all the spatial partial derivatives of the unit vectors in that system are zero except for a Beaten Gea 6 - a sind — a, cos 8 (1-32) a while in the circular cylindrical system, all are zero except for Bay _ ag eM ae (1-33) 4-7 SCALAR AND VECTOR PRODUCTS OF VECTORS Besides the simple product of a vector with a scalar quantity discussed in Section 1-3, two other kinds of products involving only vector quantities are now discussed. The first of these, called the scalar product or dot product), is defined as follows. A-B = AB cos 0 (1-34) in which @ signifies the angle between the vectors A and B. Noting from (1-34) that A+B may be written cither (4 cos 6)B or A(B cos 8) makes it evident that the scalar product A+B denotes the product of the scalar projection of either vector onto the other, times the magnitude of the other vector. The definition of A + B makes the scalar1.7 SCALAR AND VECTOR PRODUCTS OF vEcToRS 15, product useful, for example, in computing the work done by a constant force acting over a distance expressed as a vector. A generalization of this idea extended to the integral expression for work is taken up in the next section. Definition (1-34) permits the conclusion that if A and B are perpendicular, cos 0 is zero, making their scalar product zero. Again, if A and B happen to lie in the same direction, then A B denotes the product of their lengths. These observations lead to simple results involving the scalar products of the orthogonal unit vectors a,, a), and aj of the coordinate systems illustrated in Figure 1-5. For example, a, * a; = 2) * a3 = 3+ a, =0, while aja, =a,*a,=a;+a3=1. From the definition (1-34), and since B+ A means BA cos 0, the commutative law for the dot product follows. A-B=B-A (1-35) ‘The distributive law for the dot product of the sum of two vectors with a third vector A-(B+C)=A-B+A-G (1-36) can also be proved. EXAMPLE 4-2. Vector analysis can be used to shorten a number of proofs of geometry. Sup- pose one is to show that the diagonals of a rhombus are perpendicular, Represent its sides and diagonals by means of the vectors shown in the diagram, The diagonals are A+B=C and A—B=D. Form the dot product of G and D. (A+B)-(A-B)=A-A—B-B=A?— Bt which must equal zero because A = B fora rhombus. Therefore Gand D are perpendicular. If the vectors A and B are expressed in terms of their generalized orthogonal components in the manner of (1-9), their scalar product can be written A-B = (a,A; + 0,4, + 0545) * (ayBy + ayBy + agB5) ‘On expanding this expression by means of the distributive law (1-36) and applying the results obtained earlier for the dot products of the unit vectors, one obtains A+ B= A,B, + A,B, + A3B3 (1-37) B EXAMPLE 1-216 VECTOR ANALYSIS AND ELECTROMAGNETIC FIELDS IN FREE SPACE For example, the expansion of the dot product of two vectors in rectangular coordi- nates is A-B= A,B, + A,B, + A,B, (1-37b) and in circular cylindrical coordinates A-B= A,B, + AgBy + A.B, (1-37c) EXAMPLE 4-3. (a) At the point P(3, 5, 6), shown in (a) of the figure, are given the two vectors, D= —50a, + 60a, + 100a, and E = 12a, — 24a,. Find the vector magnitudes and the dot product D+ E. Use these to determine the projection D cos @ of D onto E, and the angle 0 between the vectors, (6) In (8) of the figure, at point P(5, 60°, 9) are given the two vectors F = 10a, + 8a, —4a, and G = —20a, + 80a, in cylindrical coordinates. Find the vector magnitudes and F+ G as well as the angle @ between the vectors. 50a, +60a,+ 10a, @ G=~20a,+80a, EXAMPLE 1-31-7 SCALAR AND VECTOR PRODUCTS OF vectors 17 , the vector magnitudes are D = (50? + 60? + 1007]? L 886, = [12? + 247)" = 26.833, while the dot product is found from expansion (1-37b) D-E=D,E, + D,E, + DE, (12) + 60(—24) = —2040 ‘The latter, by (1-34), also means DE cos ®, whence the projection D cos 0 becomes DE D cos =P -® - 2000 E ~ 26.833 76.03 This negative result shows that the projection D cos 0 along E is in the negative-E sense (meaning that @ exceeds 90°). The value of @ is found from the definition (1-34), yielding DE 0 = cos"! — cost DE 126.886(26.883) 26.82° (8) The magnitudes and dot product, from (1-7) and (1-37c) in circular cylindrical coordinates, are P= (F2+ P+ FI? = [107 +8? + 47] = 13.416 G = [207 + 807] = 96.462 F-G = 10(—20) — 4(8))} = —520 ‘The angle 0 between F and G is found from definition (1-34), obtai —920 a1 ee os aarecenaeny ~ 11793 From this result you may determine that the projection of F along G is in the negative-G sense ‘The second kind of product of one vector with another is called the vector product (or cross product), defined as follows Ax B=a,dBsin0 (1-38) in which 0 is the angle measured between A and B, and a, is a unit vector taken to be perpendicular to both A and B and having a direction determined from the right- hand rule provided that the rotation is taken from A to B through the angle 8. The vector product A x B is illustrated graphically in Figure 1-9. One may show from the diagram that AxB=—-BxA (1-39)18 VECTOR ANALYSIS AND ELECTROMAGNETIC FIELDS IN FREE SPACE, AxB AxB a \\ete Positive» P sense from A toB FIGURE 1-9, Illustrating the cross product which means that the vector product does not obey a commutative law. In forming the cross product, the ordering of the vectors, therefore, is an important consideration. If A and B are parallel vectors, sin 0 is zero to make their cross product zero. If A. and B happen to be perpendicular vectors, then A x B is a vector having a length ‘AB and a direction perpendicular to both A and B, with the ambiguity in the direction resolved by means of the right-hand rule. These observations applied to the cross products of the orthogonal unit vectors of Figure 1-5, for example, lead to the special results: a, X ay =a, Xa,=a; Xa; =0; a, Xa, = a3, a, Xa;—aj, and a; Xx a, = a). However, note that a, X a3 = —a). ‘A distributive law can be shown to hold for the cross product Ax (B+C)=AxB+Ax (1-40) Because of the noncommutativity of the cross product as expressed by (1-39), the order of the factors in (1-40) is important. If the vectors A and B are given in terms of their orthogonal components in the manner of (1-9), then their vector product is written AX B= (a4, + a,4, + 0543) x (a,B, + agB, + a3By) ‘The use of the distributive law (1-40) and the special results obtained for the cross products of the orthogonal unit vectors provides the following expansion Ax B=a,(4,B3 — A3Bp) + @:(4jB, — A,Bs) + a3(4,By — A2B,) which can alternatively be put into the compact determinental form AxB=|4, 4, 43 (1-41) By, By Bs1-7 SCALAR AND VECTOR PRODUCTS OF VECTORS 19 . Q Rr i. Pivot ¥ rotation EXAMPLE 4-4, ‘The definition of the-cross product can be used to express the moment of a force F about a point P in space. Suppose R is a vector connecting the point P with the point of application Q of the force vector F, as shown in the diagram. Then the vector moment M has the magnitude M = RF sin @ = |R x Fl. The turning direction of the moment, as well as its magnitude, are thus expressed by the vector product M=RxF 1-42) EXAMPLE 4-5. A force F = 10a, N is applied at a point Q(0, 3, 2) in space. Find the moment of F about the point P(2, 0, 0) The vector distance R between P and Q is R=a,(0—2) +a,(3—0) +a,(2-0) ~2a, + Sa, + 2a,m ‘The vector moment at P is found by means of (1-42) and the determinant (1-41) a, a, M=RxF 3 2) = -20a, — 20a, N-m 0 0 0 M, shown at P in the sketch, is a vector perpendicular to the plane formed by F and R. EXAMPLE 4-6. Given the two vectors F and G in (6) of the figure in Example 1-3, determine their vector cross product F x G, as well as the magnitude of the latter. Find the unit vector a, in the direction of the vector F x G. Verify that a, is perpendicular to F and wo. ‘sense of M EXAMPLE 1-520 VECTOR ANALYSIS AND ELECTROMAGNETIC FIELDS IN FREE SPACE From (1-41) expressed in circular cylindrical coordinates, F x G becomes J» a | | ae a a,| FxG=|F, Fy F, 10 8 —4 |G, Gy | |-20 0 80 = a,[8(80) — (—4)0] + a[—4(—20} — 10(80)] + a,[10(0) — 8(—20)] = 640a, — 720a, + 160a, The magnitude of F x G is [F x G| = (640? + 7207 + 1607]'? = 976.5, while the unit vector a, in the direction of the vector F x G is given by FxG _ 640a, ~ 720a, + 160a, [F x G| 976.5 0.655a, — 0.737a, + 0.1638a, The dot product a,*F becomes, from {1-37b), the zero result a, + F = (0.655)10 — 0.737(8) + 0.1638(—4) = verifying from the definition (1-34) that a, and F are perpendicular vectors. You may similarly show that a, and G are perpendicular. 4-8 VECTOR INTEGRATION Vector integration, for the purposes of field theory, encompasses integrals in space along lines, over surfaces, or throughout volume regions, as well as integrals in the time domain and the complex s domain. The subject of the present discussion concerns only integrations in space. ‘The vector notation embodies compactness as an important feature, s0 it is always worthwhile to examine the integrand of a vector integral carefully. The integrand may be either a scalar or a vector. Thus, the integrals f A-Bdf Line integral [x D)-ds Surface integral i F-Gd Volume integral possess scalar integrands, and so produce scalar results on integration. On the other hand, the integrals [ew Line integral ii Hx ds Surface integral J, * Kae Volume imegral contain vector integrands, and therefore yield vector results. In the last three examples, the integral process must take into account the different directions assumed by the integrand along the prescribed path /, on the surface S, or in the volume V definedL8 VECTOR INTEGRATION 21. df (Scalar Typical dé (Vector eae mde es placement) Py pl D Py ee a Py Pi (@) (6) EXAMPLE 1-7. (a) Integration of the scalar dé over a path ¢, (8) Integration of the vector df over the path Z. EXAMPLE 4-7. The different results provided by scalar and vector integrands is exemplified by simple integrals of scalar and vector displacements df or dé along some prescribed path in space. The integral da=fw summed over the path ¢ shown in (a) of the figure, provides its true scalar length d. On the other hand, the integral of the vector displacement dé on the same path R= [a produces quite a different answer, a vector result R determined only by the endpoints P, and Py of that path rather than by the form of the path between the endpoints. ‘This vector R is illustrated in (4) of the accompanying figure. So the line integral of d/ about a closed path is zero, whereas if dé is the integrand, the perimeter of the closed path is the result. An integral finding extensive utility in work or energy calculations is the scalar line integral F-dé= [i Fdf cos (1-43) frees, This integral sums the scalar product F - dé over the path ¢, as suggested by Figure 1-10. Only the projection of F along dé at each point on the path contributes to the integral result. The line integral (1-43) can be expressed in termas of the generalized orthogonal components of F and of dé in the following way, making use of (1-9), (1-21), and (1-37a) {eis [dt [aes + far, =f Pihade + [Fite dy + ff Fats des (1-44) In the rectangular coordinate system, in which hy = hy = hy = 1, (I-44) is written (eat [Pde t [Bat [Fae (1-45)22. VECTOR ANALYSIS AND ELECTROMAGNETIC FIELDS IN FREE SPACE Py Py ) © FIGURE 1-10. (a) A path ¢ and the field F in space. (b) Division of ¢ into vector elements df. {c) Scalar product F df {to be summed over the path) shown at the typical point P on the path, assuming (x4,3» £4) and (2, 92, ¢2) are the coordinates of the endpoints P, and P, of the path ¢. EXAMPLE 4-8, Evaluate the line integral (1-43) between the points P, (0, 0, 1) and P2(2, 4, 1) ‘over a path ¢ defined by the intersection of the two surfaces y = x? and z= 1, if F is the veetor field F= Ox — aSx2y + a,3927 i) ‘The path ¢ is illustrated in the figure. Inserting Fz = 10x, F, = —5x2y, and F, dz =0 from the definition of Z, it follows that 3yz? into (1-45) and since x? = y and 2 7 fre-f., loxds— f* Sa +0 = 20 — 106.7 = — 86.7 the desired result. EXAMPLE 1-81-9 ELECTRIC CHARGES, CURRENTS, AND THEIR DENSITIES 23 ‘This answer can also be obtained by expressing the differential displacement dx along the path in terms of dy, From the definition of /, dy = 2x dx and dz = 0. Thus —86.7 dl = fF Lords — J) 59? for 2 fixe EXAMPLE 4-9. A line integral such as (1-43) in general has a value depending on the shape of the path connecting the endpoints P, and P,. Evaluate the integral of Example 1-8 for the same function F and the same endpoints P,(0, 0, 1) and P2(2, 4, 1), but deform # into the straight-line path given by the intersection of the surfaces y = 2x and z= 1 Integral (1-43) now becomes few obviously different from the result obtained over the parabolic path in the last example. F is for this reason called a nonconservative field. A vector field for which the line integral (1-48) is independent of the shape of the path connecting a fixed pair of endpoints is said to be conservative. More is said later of such fields in connection with static electric charge distributions in Chapter 4. is 1Oxdx — f @y @ +0 = 60 4-9 ELECTRIC CHARGES, CURRENTS, AND THEIR DENSITIES ‘The physical and the chemical properties of matter are known to be governed by the electric and magnetic forces that act among the particles comprising all material sub- stances, whether inorganic or living cells. The fundamental electric particles of matter are of two varieties, commonly called positive and negative electric charges. Many experiments have provided the following conclusions concerning electric charge 1. The algebraic sum of the positive and negative electric charges in a closed system never changes; that is, the total electric charge of a defined aggregate of matter is conserved. 2. Electric charge exists only in positive or negative integral multiples of the mag- nitude of the electronic charge, ¢ = 1.60 x 107° C; this implies that electric charge is quantized. From the viewpoint of classical electromagnetic theory, an electric charge aggre- gate will be treated as though it were capable of being indefinitely divisible, such that a volume electric-charge density, denoted by the symbol p, is defined as follows? Ag A, om (1-46a) This limit of this ratio is taken such that the volume-clement in space does not be- come so small that it contains so few charged particles that the relatively smooth property of the density quantity p, is lost, although Ao is kept small enough that the integration of the quantities containing Av becomes a meaningful process. Figure 1-11 (a) illustrates the meaning of these quantities relative to a volume element Av. Ie is clear that the symbol p, for volume charge density should not be confused with the unsubseripted p, the radial variable of the circular cylindrical coordinates {p, ¢ 2)24 VECTOR ANALYSIS AND ELECTROMAGNETIC FIELDS IN FREE SPACE dv a= podv inside dv (a) (bo) ) FIGURE 1-1]. Geometries used in defining volume, surface, and line charge densities in space. (a) Quantities defining p,. (b) Quantities defining p,. (c) Quantities defining p¢. Because the charge Ag residing within any element Av may vary from point to point in a charge-bearing region, it is evident from (1-46a) that charge density is a function of space as well as possibly of time. Thus p, is a field, written in general Deltas Mas tg, {) OF Pls t) In some physical problems, the charge Aq is identified with an element of surface or line instead of a volume. The limiting ratio (1-46a) should then be defined as follows. p, = a C/m? (1-46b) Ag Pom g Cl (1-466) ‘The quantities associated with these definitions of volume, surface, and line charge densities are illustrated in Figure 1-11. In some systems of charge aggregates, two species of positive and negative charge densities may be present simultaneously. A net charge density p, (volume, surface, or line density) is in such an instance defined by* Po = pe +p, Ci? (1-47) in which p* and p; denote limiting ratios defined by (I-46a) due to the positive and negative charges Ag and Aq” respectively in Av. The dccurrence of both positive metallic ions and mobile electrons in a conductor is an example to which (1-47) may be applied. The densities, in this case being of equal magnitudes but opposite signs 4In some physical examples, such as in a plasma discharge, electrons and several kinds of ions may be pres- ent simultaneously with different densities g;. ‘Their net density at any point in the region may then be characterized by (47a) if'a total of m charge specics are to be found there.1-9 ELECTRIC CHARGES, CURRENTS, AND THEIR DENSITIES 25 (p3 = —p,), cancel, providing the net density p, = 0 in such a compensated charge system, ‘The total amount of charge contained by a volume, surface, or line region is obtained from the integral of the appropriate density function (1-46a), (1-46b), or (1-46c). Thus in some volume region, cach element di contains the charge dq = p, do, making the total charge in » the integral fa= J pwc Similar integral expressions may be constructed to yield the total charge on a given surface or a line in space. x EXAMPLE 4-40. (a) The radially dependent volume charge density p, = 507? C/m? exists within a sphere of radius r= 5 em. Find the total charge q contained by that sphere. (4) The same sphere of (a) is now covered with the angularly dependent surface charge density p, = 2 x 107? cos? 0 C/m?, Find the total charge on the spherical surface. (@) Making use of (1-47) and do of (1-17) obtains 1= fy pote = fff (60%)? sin Barada = 50 [ee fF sin 040 Roa pos = 50(2n)2 5] = 3.927 x 105 = 39.27 pC. lo Attention is called to the “product separability” of the integrand in this example, enabling the expression of the triple integrand as the product of three separate integrals in r, 0, and @. (6) Using ¢= fs pds in this case, along with the scalar surface element ds=»? sin 0 d0 db on this sphere of radius r= 0.05 m, as suggested by dr shown in Figure 1-7(4), yields on the complete sphere a= [raw = fle x 1073 cos? 8)? sin 8.0 de] --0.05 . on pe B cos? OT =2x 10 (0.05)? (° a f cos? @ sin 840 = 5 x 10 “ian 3 | 0.5 wc A vector field F(w, wg, 3, 4) at some given instant ¢, can be represented graphi- cally by use of a myriad of vectors of appropriate lengths and directions at many points in a region of space. A vector field plotted in this way is shown in Figure 1-12(a). This is, however, a cumbersome way to graph a vector field; usually a much more satisfactory representation is by use of a flax plot, a method replacing the vectors with a system of lines (called flux lines) drawn in accordance with the following rules. 1. The directions of the flux lines agree with the directions of the field vectors. 2. The transverse densities of the flux lines are the same as the magnitudes of the field vectors. The flux plot of the vector field of Figure 1-12(a), sketched in accordance with these rules, is noted in (6) of that figure. If a surface $ is, moreover, drawn in the region26 VECTOR ANALYSIS AND ELECTROMAGNETIC FIELDS IN FREE SPACE (a) () © FIGURE 1-12. A vector field F, its flux map, and the flux through typical surfaces (a) A vector field F, denoted by a system of arrows. (6) The flux map of the vector field F, showing. an open surface S through which a net flux passes. {e) A closed surface S, showing zero net flux emergent from it of space embracing that flux, then the net lines of flux y passing through § can be a measure of some physical quantity (such as charge, current, or power flow), depending on the physical meaning of F. The differential amount of flux dip passing through any surface-clement ds in space is defined by the scalar di = Fds cos 0 = F ° ds, a positive or negative result, depending on the angle between F and ds. The net (positive or negative) flux of F through § is therefore the integral of diy over S va [Feds (1-48) in which ds is taken to emerge from that side of S assumed positive, as shown in Figure 1-12(6). If S is a closed surface, the net flux through it is given by ye QF ds (1-49) as noted in Figure 1-12(¢). The latter will integrate to zero (an indication that just as,many flux lines leave Sas enter it) unless the interior volume of $ contains sources orsinks of flux lines. This view will be amplified later in the discussion of the divergence of a vector field. ‘The current flow through a surface embodies a good illustration of the flux con cept. Suppose that there are electric charges of density p,(u,, uo, us, ) in a region, and imagine that the charges have velocities averaging to the function »(w,, u2, Us, 0) within the elements dy with which the densities p, are identified. A current density func- tion J may then be defined at any point P in the region by J=pw Alm? or Cysec/m? (1.50a} This function is a measure, in the vicinity of any point P in space, of the instantaneous rate of flow of charge per unit cross-sectional area. If two species of charge density1-9 ELECTRIC CHARGES, CURRENTS, AND THEIR Densities 27 of opposite kinds, designated by pf and p;, exist simultancously in a region of space, then their total current density J at cach point is written J= pre + pe Alm? (1-50b) In general, for x species with densities p; and velocities », (e.g., electrons plus a mixture of ions) pre Am (1-50e) ‘The differential current flux di flowing through a surface element ds at which the current density J exists, is di = J+ ds amperes, to make the net current i (current flux) through S i= f,J-dsCpec or A (1-51) EXAMPLE 4-14. An electron beam of circular cross-section 1 mm in diameter in a cathode ray tube (CRT) has a measured current of 1 A, and a known average electron speed of 10° m/sec, Calculate the average current density, charge density, and rate of mass transport in the beam. Assuming a constant current density J=a,J, in the cross-section (1-51), yields the following current through any cross-section. i plat) (ad) =F. fpav= 7.4 e in which A denotes the cross-sectional arca of the beam. Thus the average current density is ‘The charge density in the beam, from (1-50a) in which J = a,4/m and »~ = —a,10°, becomes 4 == x 106 C/m? © Electron beam Cross-section A ————— EXAMPLE I-1128 VECTOR ANALYSIS AND E eCTROMAGNETIC FIELDS IN FREE SPACE The rate of mass ratio; this yield ‘ansport in the beam is the current times the electronic mass-to-charge 7 x 10718 kg/sec, assuming an electron mass of 9.1 x 1077! kg. 4-40 ELECTRIC AND MAGNETIC FIELDS IN TERMS OF THEIR FORCES Electric and magnetic fields are fundamentally fields of force that originate from clectric charges. Whether a force field may be termed electric, magnetic, or electromagnetic hinges on the motional state of the electric charges relative to the point at which the field observations are made. Electric charges at rest relative to an observation point give rise to an electrostatic (time-independent) field there. The relative motion of the charges provides an additional force field called magnetic. That added field is magneto- static if the charges are moving at constant velocities relative to the observation point. ‘Accelerated motions, on the other hand, produce both time-varying electric and magnetic fields termed electromagnetic fields. The connection of the electric and magnetic fields to their charge and current sources is provided by an elegant set of relations known as Maxwell’s equations, attributed historically to the work of many scientists and mathematicians well before Maxwell’s time, but to which he made significant contributions, They are introduced in the next section. Suppose that clectric and magnetic fields have been established in some region of space. The symbol for the electric field intensity (or just electric intensity) is the vector E; its units are force per unit charge (newtons per coulomb) ‘The magnetic field is represented by means of the vector B called magnetic flux density; it has the unit weber per square meter. If the fields E and B exist at a point P in space, their presence may be detected physically by means of a charge g placed at that point. The force F acting on that charge is given by the Lorentz force law F WE +e xB) F,+F,N in which q is the charge (coulomb) at the point P w is the velocity (meter per second) of the charge Eis the electric intensity (newton per coulomb) at P B is the magnetic flux density (weber per square meter or tesla) at P F, = gB, the electric field force acting on ¢ Fy = qv x B, the magnetic field force acting on In Figure 1-13, these quantities are illustrated typically in space. The force Fy has the same direction as the applied field E, whereas the magnetic field force Fp is at right angles to both the applied field B and the velocity v of the charged part The Lorentz, force expression (1-52) may be used for discussing the ballistics of charged particles traveling in a region of space on which the electric and magnetic fields E and B are imposed. The deflection or the focusing of an electron beam in a cathode ray tube are common examples. SJames Clerk Maxwell (1831—1879)I-11 MAXWELL’S INTEGRAL RELATIONS FOR FREE SPACE 29) @ (6) © FIGURE 1-13, Lorentz forces acting on a moving charge g in the presence of (a) only an E field, (6) only a B field, and (¢} both electromagnetic fields, EXAMPLE 4-42. An clectron at a given instant has the velocity v = (3)10°a, + (4)105a, m/sec at some position in empty space. At that point, the electric and magnetic fields are known to c E = 400a, V/m and B = 0.005a, Wh/m?, Find the total force acting on the electron. ne total force is found from the Lorentz reaction (1-52a) F g{E + v x B] = —1.6(10° 1°}[a,400 + (a,3- 10% + a,4- 105) x a,0.005] = (a,32 — a,6.4)10717N Although this is quite a small force, the very small mass of the electron charge provides a tremendous acceleration to the particle, namely @ = F/m = (a,3.51 — a,0.7)10"* mjsec?. 4-14 MAXWELL’S INTEGRAL RELATIONS FOR FREE SPACE The relationships among the electric and magnetic force fields and their associated charge and current distributions in space are provided by Maxwell’s equations, postu- lated here in integral form for the fields E and B in free space. , (ecB) ds = i p,do C (1-53) ,B ds =0Wb (1-54) fE-a= S [Bray B ? Ho d sd [E45 fh (cob ds A (1-56)30> VECTOR ANALYSIS AND ELEGIROMAGNETIC FIELDS IN FREE SPACE in which E = E(u,, up, us, ¢) is the electric intensity field B= Bly, 2, us, t) fy ic J- ds = i(t) is the net current flowing through any open surface bounded by the closed line ¢ s the magnetic flux density field q{0) is the net charge inside any closed surface S €q is the permittivity of free space (¥ 107°/36z F/m) Ho is the permeability of free space (= 4x x 1077 H/m) ‘The Maxwell equations® (1-53) through (1-56) must be simultaneously by. the field solutions E and B for all possible closed paths ¢ and surfaces Sin the region of space occupied by these fields. This strict requirement might appear to limit severely the number of practical problems that can be solved by means of these in- tegrals. Indeed, their application to the discovery of field solutions E(w, u2, ws, !) and B(u,, uz, us, £) is restricted, in the present treatment, to problems in which the charge or current distributions have particular symmetries that serve to simplify the solutions. ‘The equivalent differential forms of Maxwell’s equations, developed in the next chapter, have a somewhat wider range of application in problem solving at the introductory level. ‘The reader is to be assured that only a low-level introduction to methods for obtaining electric and magnetic field solutions of Maxwell’s integral relations (1-53) through (1-56) is attempted here. For the purposes of this introductory treatment, the Maxwell relations are simplified by considering only the field solutions of a few simple, symmetrical geometries of static charge or current distributions. In Examples 1-13 through 1-17 that follow, these simplifications are shown to enable, in one or two steps, solving for the electric or magnetic field of a given charge or current dlis- tribution. The symmetry of the distribution will be seen to be the key to providing quick solutions for the desired field. Symmetries about a point, a line, or a plane are considered. A. Gauss’s Law for Electric Fields in Free Space Maxwell’s integral law (1-53) §, (eoB) ds =f) pote =a [1-53] is also known as Gauss’s law for electric fields in free space. The meanings of the quantities are illustrated in Figure 1-14. Thus, suppose that there is in free space an electric field E(w, wo, ug, #) (denoted by the E-field flux line distribution in that figure), plus some related electric charge distribution of density p(t, u2, us, £) as shown. Con- struct in this region a closed surface S, with S having any desired shape and enclosing Although given the collective name Maxwell’s equations, historically they were in a gradual process of evolution over many years before Maxwell’ time. For an enjoyable and first-rate account of the details, you are encouraged to read the historical surveys at the beginning of each chapter in R, S. Elliot Electromagnetics. New York: McGraw-Hill, 1966.L-Il MAXWELL’S INTEGRAL RELATIONS FOR FREE SpacE 31. ic (usu, us) 8 (Positive on outer surface) Electric charges: Bo (1, Ua) C/m’ FIGURE 1-14, Typical closed surface Sin a region containing an electric ficld and a related clectric charge. Gauss’s law must hold for all closed surfaces constructed in the region, whether charges are contained or not. all or any part of the electric charge in the region, or no charge at all, as desired. ‘Then the Maxwell-Gauss law (1-53) means that the integral of the quantity (é9) * ds over that closed surface S (the net, outward flux of €7E emanating from 8) sure of the amount of electric charge J, p,do = q that is contained only within the volume V bounded by that surface S. The closed-surface integral of (¢pE) * ds thus automatically excludes any charge that happens to lie outside §. (The surface element ds on S is by convention taken as positively outward from S, as shown in Figure 1-14, or away from its interior volume V,) The constant €9 in this Maxwell-Gauss law, called the permittivity of free space, is approximately 10~°/36x F/m in the mks system of units.” To evaluate the amount of electric charge g within some volume V’ surrounded by the closed surface S, Gauss’s law (1-53) can be employed to do this two ways: (1) from the right side of (1-53), by use of the volume integral of the change density p, con tained within the volume V;; or (2) from the left side of (1-53), by integrating (gE) + ds over the closed surface S that bounds the volume V of interest. Ifa known charge distribution is static (motionless) and happens to possess a particular symmetry in free space, then Gauss’s integral law (1-53) can even be used to evaluate the electric field E produced by that charge. The small class of symmetric, static charge problems that can casily be solved by use of Gauss’s law are illustrated in the following example. EXAMPLE 4-43. Find the electric ficld intensity E of the following static charge distributions in free space: (a) a point charge Q; (4) a spherical cloud of radius ro containing a uniform volume density p,; (c) a very long line charge of uniform linear density p,; (d) a very large planar (surface) charge of density py. ‘These charge distributions are illustrated in Figure 1-15. Closed surfaces $ are shown, appropriately chosen to permit solving for E by the usé of Gauss’s law (1-53 (a) Field of a point charge (symmetry about a point). To evaluate the field E of the static point charge Q, choose § in Gauss’s law (1-53) to be the sphere with Q at its center, as in Figure 1-15(a). To show that E has only a radial component about 7 The significance of the units of €, is clarified in Chapter 4 in the discussion of capacitance. A correct interpretation of the factor €y in (1-53) is that it is a proportionality factor accounting for the proper units (mks) of the equation,32 VECTOR ANALYSIS AND ELECTROMAGNETIC FIELDS IN FREE SPACE Sphere Sy Spherical charged cloud Z& Spherical ‘closed surface $ (a) (b) @ = . Crear einai sues OB (ce) qd) FIGURE 1-15, Static charge distributions having symmetries such that Gauss’s law applied to appropriate closed surfaces will lead to solutions for E. (a) Static point charge; spherical surface S constructed to evaluate E(7). (6) Charged cloud of uniform density, showing S, and Sy used to evaluate E(7). (¢) Uniform line charge. (d) Uniform surface charge. the charge, observe that for this time-static problem (d/d? = 0, for all fields), (1-55) reduces to $E- dé =0 for all closed lines ¢. Then integrating E- dé about any circumferential path of radius r over the sphere in Figure 1-15(a) yields the con- clusion that Ey and Ez are zero. Furthermore, assuming Q positive, E must be directed radially outward if the integral of €oE over § is to yield a positive answer. ‘Thus (1-53) yields ,cola,B) ads = Q Since a, a, = 1 and from the symmetry E, is constant on S, E, may be extracted from the integral to obtain Qa E,= N, yr V/m 1-57: Freq? NIC" Vi (1-57a) or, in vector form E=af,=a,—%, (1-57b) ‘4tnegrL-Il MAXWELL’S INTEGRAL RELATIONS FOR FREE SPACE 33 Coulomb’s law for the force acting on another point charge Q! in the presence of Qs deduced by combining (1-57b) with the Lorentz force relation (1-52a), In the absence of a B field, the force on Q! when immersed in the E field (1-57b) of the charge Qis gQ Feo Fy = QE=a, (1-58) Field of a charged cloud (symmetry about a point). For the spherical cloud containing a uniform charge density p, G/m®, two cases arise. The field outside the cloud (r > ro) can be obtained from Gauss’s law (1-53) applied to a concentric sphere 5, of radius 1, as shown in Figure 1-15(b). That E has only an £, component is shown as in part (a). Then the charge g enclosed by Sj is obtained by integrating p, dv through out the sphere, so (1-53) becomes fi, col.) ads = pode Solving for E, (constant on $) yields pol(4/3)mr§) _ Peri Ane? Ber r>I9 (1-59) an inverse-square result, Itis of the form of the point-charge result (1-57a), assuming the field point outside the charge cloud (r > ro). Inside the cloud (r < ro), applying (1-53) to the closed surface S, of Figure 1-15(8) yields fy, 6018) ads = fy, poo in which the volume integration is carried out only throughout the interior of S, obtaining p,(4/3)ar*. With E, constant on S3, Po E, 3€ r0 «<0 (1-62) Ttis evident that the electric field to either side of a uniform, infinite planar charge is everywhere constant. Flux plots of the electric fields of the four charge distributions covered in this example are shown in Figure 1-16. B. Ampére’s Circuital Law in Free Space Maxwell’s integral law (1-56) B ? Mo ae dt d . . dt = f.yods +5 I (ec) sds i+ [1-56] is often called Ampére’s circuital law for free space. Figure 1-17 illustrates the meanings of the field quantities relative to any closed line ¢ that bounds a two-sided surface 5. The positive direction of the typical element ds may be taken to either side of S, but the positive integration sense about ¢ must agree with the right-hand rule relative to ds. The relation (1-56) means that the line integral of the B field (modified by 15 *) around any arbitrary closed path ¢ must, at any time f, equal the sum of the net electric current i plus the time rate of change of the net electric flux , passing through the surface 8 bounded by ¢ Displacement current flux: Time-varying JeB-as oF flux. by i and aye/at te (b) FIGURE 1-17, Induced magnetic fields and Ampére’s law. Any closed line ¢ such as that of (a) may be superposed anywhere on the example of (6); Ampére’s law must be true for it, (a) Typical closed line ¢ hounding a surface S, relative to the fields in Ampére’s law. (6) A sym metric example showing the B field induced by electric currents and displacement currents36 ve TOR ANALYSIS AND FLEGTROMAGNETIC FIELDS IN FREE SPACE ‘The two terms on the right side of (1-56) denote the two kinds of electric currents that occur physically in free space. The first, i, has already been discussed in relation to (1-50) and is given the name convection current when it is comprised of one or more species of moving charges in free space; it is also called conduction current if it pertains to electric charges drifting or transported within a solid, liquid, or gas. The second term, dip_/dt, is called displacement current and denotes the time rate of change of the net, instantaneous electric flux , that passes through the surface §. The displacement current term is the history-making contribution of Maxwell, who provided that missing link to unify the theories of electricity and magnetism and predicted the propagation of electromagnetic waves in empty space in the absence of charges and currents. The quantity fig is called the permeability of free space; it has the value 47 x 1077 Hm in the mks system of units. ‘A comparison of (1-56) with Gauss’s law (1-53) shows that Ampére’s circuital law is more comprehensive; it involves both the magnetic field B and the time-varying electric field E, as well as electric currents that might be flowing in a region. Indeed, it specifies that either electric currents or time-varying electric fields in a region, or both, will give rise to a magnetic field B such that (1-56) must be satisfied for all possible closed lines constructed in the region. ‘The direct application of Ampére’s circuital law (1-56) to obtaining time-varying field solutions E(x, ‘) and B(x, ¢) whenever, for instance, a current distribution J(xr, !) is somehow specified is not, in general, feasible. The difficulty lies in part in not knowing how to specify the current distribution without more information about the accompanying fields; the intricacies may be appreciated more fully on recognizing that the field solutions must satisfy simultaneously all four of Maxwell’s integral relations, (1-53) through (1-56) ‘No field solutions of the complete Ampére’s law (1-56) are attempted at this time. Instead, consider for a moment only the magnetic B field of a static (direct) current distribution i, in which event the Maxwell-Ampére law (1-56) reduces to the form + dé = [J+ds =i Ampére’s law for static fields (1-68 Lae i Pk ) To evaluate the amount of static electric current i passing through some surface S, the static Ampére law (1-63) can be employed to do this two ways: (1) from the right side of (1-63), using the surface integral of J+ ds over any desired surface S (a flux integral); or (2) from the left side of (1-63) by integrating (B/uo) * dé about the closed path ¢ that bounds the surface S (a closed line integral). If'a known static current distribution also happens to possess a simple symmetry in free space, then the static Ampere integral law (1-63) can even be employed to evaluate the magnetic B field produced by that current, Simple illustrations of these uses of Ampére’s law (1-63) are described in the following examples. EXAMPLE 4-44. Find the net static electric current j that flows through each of the surfaces $ bounded by the paths ¢ chosen for the three direct current systems of Figure 1-18. "Phe significance of the units of fg is clarified in Chapter 5, in the discussion of inductance. In (1-56), to is the factor that properly adjusts the units of the term in which it appears, to yield an equality that is dimensionally correct.[-ll MAXWELL’S INTEGRAL RELATIONS FOR FREF SPACE 37 FIGURE 1-18, ‘Three examples of direct current systems showing closed paths ¢ that bound surfaces § through which net currents flow. (a) Infinitely long wire carrying I A. (b) A two-mesh de circuit. (¢) A four-turn coil carrying 1A. (a) For the long, straight wire of Figure 1-18(a), the path ¢, shown yields a net current j= 0 through S,; while ¢, embraces i= 1A, a positive result if ds on S; is taken to be positive in the upward direction. (0) Assume a positive ds in the upward direction on S. Then, by inspection of Figure 1-18(4), the net conduction current i through S becomes i= [\J-ds= 2141-31 0 For the path ¢ constructed about the coil in Figure 1-18(c) such that the coil pierces ‘S four times, the net current becomes [pdras= 40a if ds is assumed positive in the direction shown. EXAMPLE 4-45. Field of « long, rowed wire (symmetry about a line). Use Ampére’s circuital law (1-63) to find the B field of the static current J in an infinite, straight, round wire of radius a, shown in Figure 1-19(a). Find B both inside and outside the wire. 'As in Figure 1-19(a), assume a symmetric, closed integration path ¢, having the radius p shown. From (1-63), the B field must be @-directed if the Tine integration counter ‘clockwise (looking from above) is to yield the positive current J emerging from S;. With B= a,b, and df = asp dé on the closed contour, (1-63) yields ae By roe hab! but B, is of constant magnitude on ¢;, obtaining exterior to the wire Hol Hol SF pdb 2xp pra If p op Pe ne ce (1-64) ° Ona® EXAMPLE 4-46. Field of a flat current sheet (symmetry about a plane). Use Ampére’s circuital law to find B on both sides of a thin, infinite current sheet in the x = 0 plane and carrying the constant, static surface current density J, = a,Jg. A/m. ‘The infinite sheet can be viewed as pairs of thin current filaments located aty, —y, carrying the differential current di = J,, dy as in Figure 1-20(a). On recalling the exteriorIL-1] MAXWELL'S INTEGRAL RELATIONS FOR FREE space 39) (0) Current sheet (top view) (@) 0) FIGURE 1-20. An infinite, thin current sheet carrying @ constant surface current of density Jo=8Je Alm. (a) Showing paired current filaments, and resultant dB-feld. (#) Symmetric closed path about which Ampére’s circuital law is taken. field result (1-64) of Example 1-15, one may conclude that the paired currents produce a net, j-directed, differential field dB at any point on the x-axis as shown. The superposed effect of the whole current sheet is therefore a y-directed B field on the positive x side of the sheet, and a negative y-directed field on the other side. Then Ampére’s law (1-63) becomes, for the symmetric, rectangular path shown in Figure 1-20(8) Peerage Pr with the surface integral on the right side reducing to a line integral over any yo width of the sheet. Because both B, and J,, are constants over the indicated paths, this becomes 2B, ¥ = MoJuzo t0 yield By = HqJu,/2. In vector form, therefore —a,B,) + ay dy= to fe, Jad B- Hale x>0 tole 9 <0 (1-65) 2 EXAMPLE 4-47. Find the magnetic fields of the following coil configurations, each carrying a static current I: (a) an n turn, closely wound toroid of circular cross-section; (b) an infinitely long, closely wound solenoid having n turns in every length d. The coils are illustrated in Figure 1-21 (a) The magnetic flux developed by / in the toroid is g-directed as in Figure 1-21(a), a result following from symmetry and the application of the right-hand rule to the positive current sense shown. ‘Thus, inside the toroid, B = ayBg, exact if the winding is idealized into a current sheet, The application of the time-static Ampére’s circuital law (1-63) to the symmetric closed line ¢ of radius p shown therefore yields 4, (agBq) “apd? = onl, in which By, from the symmetry, is constant around ¢, and 71is the met current passing through $ bounded by ¢. ‘Thus sont Pe op (1-66)40 VECTOR ANALYSIS AND ELECTROMAGNETIC FIELDS IN FREE SPACE, Closed path Closed path B=0(p>6) @ (6) FIGURE 1-21, Two coil configurations, the magnetic fields of which can be found using Ampére's circuital law. (a) Toroidal winding of» turns, showing symmetric path ¢. (6) Infinitely Jong solenoid, showing a typical rectangular closed path 2, an inverse p-dependent field inside the region bounded by the current sheet. If the radius p of ¢ in Figure 1-21(a) were chosen to cause ¢ to fall outside the torus (with p Gnegr? It is easily shown that this electric field obeys Faraday’s law (1-68) for a time-static field. If any closed path, such as ¢ = ¢, + ¢, shown in Figure 1-23, is chosen in the42 FIELDS IN FREE SPACE Direction ‘ot integration! (@) (o) FIGURE 1-23. Closed paths constructed about a point charge and a charge distribution, rela- tive to Faraday’s law for static charges. (a) Point charge Q.. {6) Charge distribution p,, space about a point charge, the integral of E-dé from any point P, to any other point P, along the path ¢, is (1-69) This result? is seen to be independent of the choice of the path connecting P, and P); it is a function only of the radial distances r, and ra to the respective endpoints P, and P,. Therefore, if the integration is taken around the complete path ¢ = ¢, + ¢, shown in Figure 1-23, the two integrals from P, to P, via ¢, and thence from P, back to P, via 4, will cancel, and (1-68) follows. Static charge distributions like those depicted in Figure 1-23(4) are, in general, just collections of differential charge-clements dq = p.dv; whereas their static electric fields are just superpositions (vector sums) of the conservative differential electric fields dE produced by each of those static charge- elements. One may thereby agree that Faraday’s law (1-68) for static electric fields is true in general Valid field solutions B(x, #) and B(x, t) satisfying Faraday’s law (1-55) must also satisfy the remaining Maxwell’s integral relations of (1-52) through (1-56); however, if the time variations of the fields are not too fast, in some cases a static solution for The physical interpretation of the result (1-69) is of interest. [t implies that the ne! work done in moving ‘unit test charge around a closed path is zero; such an electric field has already been termed conserative, “Thus (1-69) forms the basis of the theary of the scalar potential field of static electric charges to be discussed in Chapter 4L-Il MAXWELL’S INTEGRAL RELATIONS FOR FREE space 43 B satisfying the static form of Ampére’s circuital law. (1-63) [1-63) can be assumed to be known field (e.g., the field solutions of Examples 1-15 through 1-17. Ir the current densities J are slowly time-varying, one can assume they will giv rise to a slowly time-varying B field. Such a static field on which time variations are imposed is called quasinslatic. On inserting the quasi-static field B(, #) into Faraday’ law (1-55), a first-order approximation to the E field can then be obtained, assuming that the field symmetry permits the extraction of the solution for E from (1-55). An iterative process can sometimes then be employed to improve the accuracy of the quasi-static solution!” although if the time variations of the fields are not excessively rapid, the first-order solution will often suffice. EMAMPLE 4-48. ‘The long solenoid of Figure 1-21(8) carries a suitably slowly time-varying current i= [g sin cot. Determine from Ampére’s law the quasi-static magnetic flux density developed inside the coil of radius a, and then use Faraday’s law to find the induced electric intensity field both inside and outside the coil From Example 1-17(4), the magnetic flux density inside the long solenoid carrying a static current 1 was found to be (1-67). Thus, the solenoid current Jy sin wt will to a first-order approximation provide the quasi-static magnetic flux density Be ol () sin wt = a,Bg sin oo (1-70) in which Bo = ftono/d, the amplitude of B. This assumption is reasonably accurate for an angular frequency @'which is not too large. The electric field E induced by this time- varying B field is found by means of Faraday’s law (1-53), the line integral of which is first taken around the symmetric path ¢ of radius p inside the coil, as shown in Figure 1-24. Faraday’s law becomes (aglg) agde 4 fcaBo sin wt ds P (aw 6 a ds (22Bo sin wt) + a in which, from the circular symmetry, Fy must be a constant on ¢, Thus Ey $ de = -wy cos ot [ds but §, dé = 2xp and fx ds = np, so that P coset p ro) (Sketch a diagram patterned after Figure 1-15(b), showing the labeled Gaussian surfaces employed.) Show from your solution that the exterior E field (r> 79) is identical with that expected if the total charge in the sphere were concentrated entirely at the origin. (c) If fo = 1073 Gjm? and ro = 10cm, find q in the sphere and sketch F, versus 7. [Answer: (a) 7 = npor (b) E, = por*/tEora for 1 <1, Ey = por§/4€ or? for > 79] 1-28. A spherical shell of charge possesses the constant volume charge density p, between its inner and outer radii a and 6. Use Gauss’s law to prove that the E field for r < a is zero; that for a <1 6) itis p,(b* — a*)/3eor*. (Show an appropriately labeled sketch along with the details of your proof.) 1-29. Let the volume charge density within a spherical region of radius r= a be given by P, = pol) + fr), in which po denotes the density at the origin, Determine the & that will make the total charge in the sphere zcro. For this £, why is the E field external to the sphere zero? Find E as a function of r within the sphere, making use of Gauss's law. [Answer: f = —4/3a] //1-30, An infinitely long, cylindrical cloud of radius p = a in free space contains the static, uniform volume charge density p,. With a suitably labeled sketch, make use of the symmetry and Gauss’s law (1-53) to obtain the following. (a) The electric field outside the cloud (p > a). (b) The interior electric field (p <4). (c) Show that the exterior E field is the same as that expected if the same total charge per length ¢ were concentrated as a line charge along the z axis, as in Figure 1-15(c}. [Answer: (a) E = a,p,a7/2€op (b) a,pop/2€o] 1-31. Let an infinitely long, cylindrical charged cloud of radius @ contain the static charge density p, = po(p/a)?, varying parabolicaily to the density po at the cloud surface. (a) Make use of (1-17) to determine the total charge g in any length / of this cloud. (b) Sketch a diagram as suggested by Figure 1-15(c), making use of the symmetry and Gauss’s law (1-53) to find the E field outside the cloud (p > a), and then inside it (p a, Pop? /teoa? for p ) is zero. (Do not use Gauss’s law.) (b) Repeat (a) if both surface charge densities are positive SECTION 1-11B 1-33, A hollow, circular cylindrical conductor in free space, assumed infinitely long to avoid end effects, and having the inner and outer radii b and ¢, respectively, carries the direct current 1. (a) Assuming a constant, z-directed current density in the conductor cross section, show that the vector current density at any point therein is J = a,l/n(c? — 67). (b) Use Ampére’s law to show that the exterior magnetic field is the same as that of the solid conductor of Figure 1-19 carrying the same total current [, Show that B inside the hollow interior (p <6) is zero, whereas that within the conductor (b

c. (c) Sketch a graph of By versus p over the (0, ¢} range, assuming @= 3mm, 6 = 6mm, c= 8mm, J= 100A. Find the current density in each conductor, expressed in Aem?, 1-35, Show that the static B fields of the coaxial line of Problem 1-34 are the superposition of the fields of the hollow conductor of Problem 1-33 and those of the isolated conductor of Example 1-15. 1-36. ‘Two parallel, indefinitely thin current sheets of infinite extent in free space are located at x= Fd, possessing the uniform but oppositely directed surface-current densities + Jses respectively. (The currents are assumed charge-compensated, making electric fields absent in this problem.) (a) Employ results of Example 1-16 and superposition (not Ampére’s law} to show that B between the sheets (—dd) is zero, (b) Sketch a fiux plot of the net B, field. Show that if J,, = 100 A/m, then By = 1.257 mWb/m?. (c) Comment on the changes in the fields if both current densities were assumed in the same (+2) direction, 1-37. An infinite, plane conducting slab of thickness d in free space bas its sides coincident with the x = —d/2 and d/2 planes. Assume the constant volume current density J = a7, A/m? within the conductor. {a) In the manner of Example 1-16, use Ampére’s law to show that the B field outside the conducting sab ({x| > a2) is taysleJ.d. (b) Make use of Ampére’s law to find B inside the slab. [Hint: Choose a rectangular closed path 2 with one side parallel to the known field of (a), and its other side aligned with the unknown field.) 1-38. ‘Two parallel, round conductors, infinitely long and carrying the currents 1, 1, are spaced 2d m apart. Assume them parallel to the z-axis and centered about the origin on the x-axis. (a) Sketch a top view of the conductors in the x-y plane, with the current in conductor | atx = d assumed + z-directed. Show its vector field contribution B, at the normal distance p; from I to the typical location P(x, 9), making use of (1-64). Showing B, decomposed into its By, and Bys components, use the geometry to develop the expression for By solely in terms of x and. (b) Doing the same for conductor 2, express the total B at P, due to both conductors, entirely in rectangular coordinate form. {c) If = 10 A and 2d = 5 cm, find B at the origin. Find also the vector B at the following {x, 3) locations expressed in centimeters: (1.25, 0), (3.75, 0), (0, 1.25), (1.25, 1.25), (2.5, 1.25), (3.75, 1.25), (0, 2.5), (2.5, 2.5}. [Answer (b): Hol yy» S =| pat), (2-2 )+a,(5-—S el G 3) ene60 VECTOR ANALYSIS AND ELECTROMAGNETIC FIELDS IN FREE SPACE PROBLEM 1-39 in which pale d tr? p= [ee d)? +971" 1.39. Find, by superposition, the magnetic fields of a pair of coaxial, ideally closely wound toroids of circular cross section as shown, assuming the same number of turns and the identical currents /, Assume the currents first in the same direction; then, in opposing directions. SECTION 1-12 1-40, Introducing the unit vector ay at P on Figure 1-26(a), from the geometry verify (1-73b). Similarly verify (1-73c). AAI. (a) From the geometry of Figure 1-26(5), verify that the projections of the unit vector a, onto a,, ay, and a, yield a, +a, = sin 0 cos , a,* a, = sin 6 sin g, a, +a, = cos 0. (b) Modify Figure 1-26(d) to enable deducing the following projections: ay- a, = cos 0 cos, ap* ay = cos 0 sin $, aya, = —sin 0, Show similarly from the geometry that ay -a, = —sin g, ay* a, = cos $,a4° a, = 0. (c) Expressing A in rectangular coordinate form; A= aA, + ad, + a,A,, Use the foregoing results and methods discussed in Section 1-12 to deduce the expressions for the spherical coordinate components of A in terms of its rectangular components, that is, A, =a,+ A= Ax sin 0 cos 6 + A, sin 8 sin 6 + A, cos 0 dg = A, €08 0 cos + A, cos O sin § — A, sin 6 Ag = —Agsin § + Ay cos @ [1-79a,b,<] 1-42, (a) Asphere of radius a and centered at the origin is expressed in rectangular coordinates by x? +9? + 2? =a’. Use the appropriate coordinate translormations to reduce this to the simpler spherical coordinate expression. (b) Center the sphere at x= a to make its expression (x— a)? +9? + 2? =a’. Transform this to its spherical coordinate form, r= 2a sin 0 cosh. 143, ‘Transform the following vector fields to the circular cylindrical coordinate system. (a) A= 10a, (b) B= 10ya,, (c) D = 3(1 — ja, + a,4y?. [Answer: A= a,10 cos ¢ — ay10 sing, B=a,l0psin cos —agl0p sin? , | D=a,3(1 — p? cos? $) sin @ + ag3(1 — p? cos* d) cos $ + a,4p? sin® 6] A-44, Transform the given vector fields to the spherical coordinate system. (a) A = 10a, (b} E=a,100 [Answer A= a,10 sin O cos $+ agl0 cos cos p —a,l0 sing, B= ,100r sin? 8 sin $ cos @ + ay1007 sin 8 cos 8 sin $ cos @ + ay100r sin 4 cos? G] V