INSTRUCTOR’S SOLUTIONS MANUAL INTRODUCTION to ELECTRODYNAMICS Third Edition David J. GriffithsChapter 1 Chapter 2 Chapter 3 Chapter 4 Chapter 5 Chapter 6 Chapter 7 Chapter 8 Chapter 9 Chapter 10 Chapter 11 Chapter 12 TABLE OF CONTENTS Vector Analysis Electrostatics Special Techniques Electrostatic Fields in Matter Magnetostatics Magnetostatic Fields in Matter Electrodynamics Conservation Laws Electromagnetic Waves Potentials and Fields Radiation Electrodynamics and Relativity 157 179 195 219Chapter 1 Vector Analysis roblem 1.1 (a) From the diagram, |B + C| cos; = |B| cos; + |C|cosé2. Multiply by |A|. |AIIB + C| cos 4s = |A||B|cos6; + |Al|C|cos6s. So: A.(B + C) = A-B+A-C. (Dot product is distributive.) Similarly: |B + C|sin 85 = [B|sin 8; + |C|sin8. Mulitply by ||. |AIIB + C]sin@s fi = |Al[B[sin#; A+ |Al|C|sin@, a. If fis the unit vector pointing out of the page, it follows that, Ax(B-+C) = (AxB) + (AXC). (Cross product is distributive.) TBleoses [Cleoobs (b) For the general case, see G. B. Hay’s Vector and Tensor Analysis, Chapter 1, Section 7 (dot product) and Section 8 (cross product). Problem 1.2 [Cin os ‘The triple cross-product is not in general associative. For example, suppose A = B and C is perpendicular to A, as in the diagram. ‘Then (BxC) points out-of-the-page, and Ax(BxC) points doun, and has magnitude ABC. But (AxB) = 0, so (AXB)xC = 0 ¢ i Ax(BxC). BxC yAx(BxC) Problem 1.3 a As +1k+19—1% A= V3, B=1R+19 +14 B cos8 = V3V3c0s8 = cos Problem 1.4 "The cross-product of any two vectors in the plane will give a vector perpendicular to the plane. For example, we might pick the base (A) and the left side (B): A= 18429408; B=-18+09 432.2 CHAPTER 1. VECTOR ANALYSIS ga AxB=]-1 20 -103 This has the right direction, but the wrong magnitude. To make a unit vector out of it, simply divide by its length: OX439 +22. |AxB| = V56F O74 fetish Problem 1.5 & y a Ax(BxC) Ae Ay A (By, — ByCy) (BsCz—-BzCz) (BsCy — ByCz) 8[Ay(BeCy — ByCz) — Az(B:Cz — BsC:)] + 9() + 2() (UM just check the x-component; the others go the same way.) R(AyBrCy ~ AyByCy — A,B,Oz + AzBzC,) + 9() + 2(). BUA-C) ~ C(A-B) = [Be(AzCe + AyCy + ACs) — On(AcBe + AyBy + AzB,)]R+()9 + (08 3(4yBsCy + AvBsC, ~ AyByCz ~ AzB:Cz) + 9() + 2(). They agree. Problem 1.6 Ax(BXxC)+Bx (Cx A)+Cx(A xB) = B(A-C)-C(A-B)+C(A-B)—A(C-B)+A(B-C)—B(C-A) So: Ax (BXxC) ~ (AxB)x€C = —Bx (Cx) = A(B-C) - C(A'B). If this is zero, then either A is parallel to C (including the case in which they point in opposite directions, or one is zero), or else B-C = B-A = 0, in which case B is perpendicular to A and C (including the case B = 0). Conclusion: [A x(BXC) = (AXB)xC <= either A is parallel to C, or B is perpendicular to A and C. Problem 1.7 AR 469 +82) — (28489472) = [PR-2H 4B Problem 1.8 {a) AyBy + ds B, = (cosy + sin 6A,)(c0s bBy + sin Bz) + (~ sin ¢Ay + cos 44,)(—sin 6B, + cos dB.) = cos? bAyBy + sin $008 ¢(Ay Be + AzBy) + sin? A,B, + sin? 6A, By — sin cos ¢(AyBs + AzBy) + cos? AsBe = (cos* @ + sin? g) Ay By + (sin? @ + cos? $) A, By = AyBy + AsBy. V (b) Aa)? + (Ay)? + As)? = Shy AAs = Der (Wer RipAy) (Bhy RAs) = Bye (URey Rat) Ay Ar ; ; Li This equals A? +A? + A? provided| 82_,Ry Rae = { of Vat } Moreover, if R is to preserve lengths for all vectors A, then this condition is not only sufficient but also necessary. For suppose A = (1,0,0). Then Sy (Us Ray Rie) Ay Ax = By Ry, Rn, and this must equal 1 (since we want A,+A,+Ay = 1). Likewise, D2, RyoPis = Ei, Ragas = 1. To check the case j # k, choose A = (1,1,0). Dya (Es RypRen) Ape = 3, Ra Ra + By Rae +3 Ru Ra +3, RgRu. But we already know that the first two sums are both 1; the third and fourth are equal, s0 3 Ru Riz = 3, RR, =0, and so on for other unequal combinations of j, k. V_In matrix notation: RR = 1, where R is the transpose of R.Problem 1.9 y “ty -2 Looking down the axis: we 2 “i 5 ‘A 120? rotation carries the 2 axis into the y (= Z) axis, y into x (=), and 2 into z (=). So As = Ay = As, Ae = A, oo1)\] r=(100 o10 Problem 1.10 (a) [No change] Ae Aas Ay = Ay 2 (b) [A> =A] the sense (Ae = —Az, Ay = ly» Az = —Az) (© (AxB) — (-A)x(-B) = (AxB). That is, if C = AxB, [C —) GC]. No minus sign, in contrast to behavior of an “ordinary” vector, as given by (b). If A and B are pseudovectors, then (AXB) —+ (A) x(B) = (AxB). So the cross-product of two pseudovectors is again a pseudovector. In the cross-product of a vector and a pseudovector, one changes sign, the other doesn’t, and therefore the cross-product is itself a vector. Angular momentum (L = rxp) and torque (N = rxF) are pseudovectors. (a) A(Bx@) — (-A)((-B)x(—C)) = -A(BX©). So, ifa = A(BXC), then changes sgn under inversion of coordinates. Problem 1.11 (AVF = 2WwK+ By? 9 + 4252 a pseudoscalar (OVS = 2aytt 4 Barytehy + 428g a OVE # sinyln 2&-+ €* cosy ln z¥ +e? siny(1/2)% Problem 1.12 (@) Vi = 10[(2y ~ 6x ~ 18) + (22 ~ By + 28) 97]. Vh =0 at summit, 60 0 Oo get-sauca PIE M88 2r — By +28 22y = 66 => y = 3 =p 2e- 2442809 = ‘Top is [3 miles north, 2 miles west, of South Hadley. (b) Putting in 2 = -2, y =3: f= 10(—12— 12-36 +36 + 84412) = [720 Re (6) Putting in 2 = 1, y =1: Vh = 10[(2 ~ 6 — 18) & + (2~ 8 + 28) 9] = 10(—22% + 22.9) = 220(- +9) [Vh| = 220V3 «(311 fe/mile} direction: [northwest.4 CHAPTER 1. VECTOR ANALYSIS. Problem 1.13 a= (e-2)8+W-V)94(2-2)% o= VE-2P FU -VP Fe (2) Ve?) = Ble— 2")? + yy + (2— 2718+ HOT+ HOH = 2e—2) 4 Ay—y")I +222 = 20 (0) VR) = £lle- 27 + v—vP + e204 BON + BOBS = -}0-92@- 2) - }0-}y-y!) ae - 22 = ON] 2+ VT (~ 28] = (1 = —(1/ 9. (© £0") =n = nahh 90,) = nae, 90[ VR) = nah Problem 1.14 +y coop +2 sing; multiply by sing: Jsing = 4y sing cos +2 sin? g =v sind + 2 cos 4; multiply by 008: ~y sin pcos b+ 2 008? 6. Add: Jsing + Zco0s¢ = 2(sin® 9 + cos? ¢) = 2. Likewise, Joos} — Zsing = y. So 8 = cos; 9 = —sing; 9 = sing; 34 = cosg. Therefore Fos AV fy + sin AV). —sin (WV fly + c08 (VF) } So Vf transforms as a vector. ged Problem 1.15 ()V-va = E@) + HGx2") + R222) = 2240-22 = 0. (V-v = Alay) + £Ov2) + LGze) =y + 22 +32. (QV -ve = Kv?) + Key + 2”) + F(2y2) = 0+ (22) + (2y) = Ac + y). Problem 1.16 Vv= £(S)+R(S4R(4) HK [et (2? +9? +248] $+ 2(-3/2)0-bae + 078 + ¥(- 3/0" by +0) = 8r-8(a? + y? + 22) = Br $ ‘This conclusion is surprising, because, from the —— this vector field is obviously diverging away from the origin. How, then, can V-v = 0? The answer is that V-v = 0 everywhere except at the origin, but at the origin our calculation is no good, since r = 0, and the expression for v blows up. In fact, V-v is infinite at [veer + vt + aty-F] +g [ate? +0? + 27)°3] 44 2(-3/2))7 422 jy = cos buy + sin dvs; Us SE = Gp cosd + Sy sin (Gr cos 6+ $e sind) cose + ($+ cond + Se sin) sin =. sin d+ 9 cos = — (S598 + BB) sing + (52 8 + BSB) cos =~ (-B sing + Be coed) sing + (~B ning + $e cond) co. So He 4 Me = M4 cost d + M8 sin deos.d + Se singcong + Be sin? + Ste sin? 4 — Se sin grcos ind. Use result in Prob. 1.14:— 9 singcosd + Be cos? = Sit (cos? ¢ + sin? 9) + 9 (sin? 6 + cos? #) = Gp + Gee ¥ Problem 1.18 2 (0) Vxve = (0 — 6x2) + 9(0 + 22) +.2(32" ~ 0) =[—GzzR + 229 +3272. (0 — 2y) + 9(0 — 32) + 2(0— 2) = [-2yR— 327-22. (22 — 22) + 9(0- 0) + 2(2y — 2y) = [B fo i (2ay + 27) (9) Vxve rs 2 (0) Vxv = | z ny % 2 3 y Problem 1.19 va yR+29j or v= yzk-taz9 + ayh; or v = (3x22 — 2) R+3H + (2? — Bee”) a or v = (sin z)(cosh y) & — (cosz)(sinh y) $; etc. Broblem 1.20 ©) Vio) = Bede + Mad y + Yea = (22+ ofl) 2+ (s98 +996) 9 + (198 + 0%) 2 =S(Bx+ I+ Ma) +9 (xs H+ Ha) = 10) +N. acd (iv) V-(AXB) = 2 (AyB, — AvBy) + & (AsBz — AzBz) + $ (AzBy — AyBr) = Ay ORs + Bote — A, 28s — By Ws + A, Be + BM — Ag BBs — BL Oe As Pe + B, x ~ Ayes ~ Bae Ba (hp — Sr) + By (Me — Ss) + Be (BE ~ Be) — Ae (4 - Bt) —Ay (BBs ~ 9B) ~ A, (Bt — Be) = B.(VxA) - As(V XB) qed (0) Vx (FA) = (fsa ~ 2G fad) w+ (fad — Sifted) 9 4 (Mipted — Apfel) g = (Fe + Aah ~ 98 ~ Ay) + (s8f + AaGl — 19 - Aad) + (s%fe + Ay BE - 19h — Ang) 2 = 5 [( Se) e+ Oe — 9+ (Be He) a] — [(AvBE— AsSE) 8+ (Ar BE - 4eGE) 9+ (4096 - ABE) 8] =S(VXA)—AXx (VS). ged Problem 1.2 (a) (AV) (Acts + Ay Me + Aad) + (Aa SEs + Ay Be + Aa SP) + (eSBs + Aye + ASR) a. = SRE, Let's jun do the # component (eh t0h +28) oe6 CHAPTER 1. VECTOR ANALYSIS = {2 [+ 2-BApe2e] + v2 [-bbye2a] + 22 [-4 Ape] } HE- AG +e tee)}=2(F- Bet +e )} = EER Same goes for the other components. Hence: [(F-V) * = 0 (0) (var¥) ve = (29 + Sea? f — 2ez fh) (cy R+ 2yzV + 3222) = 27 (yX +09 + 328) + 3x2? (2k + 22H +02) — nz (OK + 2H + 328) = (wy +3222?) 8+ (G24 — dayz) $+ (322 622) 2 i) [V(AB)], = F(AsBs + AyBy + AsBe) = Se Be + AaB + eB, + Ay Ge + Bab, + ABB [Ax(VxB)], sey — A.(VxB), Se 8 [Bx(VxA)], SB — Mn) — B, (Bx — Be) [(A-¥)B], (48 CE TABI = he athe + A,B (B-V)A], = Boe + 3, She + BoA So [Ax(VXB) + Bx(VxA) + (A-V)B + (B-V)Al, = Ay 3pe — Ay Bs — A, 8Bs + A, Bs + By Me — By Ma — Bs + BOs +A, Bs + Ay Be + ABs + Bs Be + By Be + BeBe = Bie + Aloe + B, (Se — hee) a ~ fe He) +B, (~2f + 9h + Me) + A,(— 9p + 9B + Oe) = [V(A-B}], (same for y and 2) (vi) [Vx(AxB)],, HlAxB). — F(AXB)y = §(AeBy ~ AyBz) ~ #(ArBe ~ Az Be) Shen, + Atha — 28a, — Ay Be ~ Aa, — As PBe + eB. + Ae MBs [(B-V)A - (A-V)B + (VB) — 2IV. “Ade Be Sf + By 9a + By Ms — A, Pa — Ay Be — A, Gs + Ay (Sfe + Ge + Be) — B (Spe + Sta + SA) = By She + Ae (Me fe 4 9B 4 BBs) + Be Bf — Hf 2A — Pa + Ay (882) + As (—e) + Bs (Ar) =[Vx(AxB)], (same for y and 2) Problem 1.23 Vis/9) = RU) x+ Esl) 9+ Rs/2 of set gy THE Ut fo (ges-+ 9 + Sa) — 5 (Ses + Seo + B42) = aE V-(A/9) = Bite 10) + SAvl9) + BelAs/) is — Aa ae 8s Aol 3 fo (Gte + Met Me) — (4cbe + Ay 4 A.Bs)] = ALAGA, ged i:[oe — )~ (4.8 48)] = MUKA HAXVe (game for y and z). ged [Vx(A/g)]e saa — B(Av/9) awe’ Problem 1.24 ae 5 8 dy Bz y —2e 0 (@) AxB = = 2(G22) +.9(02y) + 2(-22? ~ 692) V{AXB) = £(6e2) + $(Ozy) + £(-22" - 6p") = 62 +92 +0 = 152 VxA = 8 ($5 (82) ~ $20) +9 (H©) ~ HG) +2 (KOv)— H(2)) =0; BAVA) =0 VxB = 8(2(0)— £(-20)) +9 (Lu) - AO) +2(Z(-2z) — £66s)) = 5% ACVB) = 152 V-(AxB) 2 B.(Vx A) — A-(VxB) = 0 — (—152) = 152. ¥ (b) A-B = Szy ~ day = —2y ; V(A-B) = V(—ay) = 8 (—2y) + 7H (-sy) = -yR-2F zo % Ax(VxB)=| 2 2y 32 | =&(—10y) +9(52); Bx(VxA) =0 00 -5 (A-V)B = (29 + vg +3282) (Buk — 229) = K(6y) + 9(-22) (B-V)A = (3% = 2p) (a@&+ 29 + 322) = X(3y) + 9(—42) AX(VXB) + Bx(VXA) + (A‘V)B-+ (BVA lly & + Sa ¥ + Gy% — 2H + ByX — Ae} = -yR—2P = V-(A-B). Vv (6) Vx(AxB) = & (J (22? ~ 6y?) ~ (020) +9 ($(Cxz) ~ Z(-22* ~ 6y*)) +2 (P(Oxy) - £ (622) -12y — Sy) + 9(6x + 4x) + 2(0) = —2ly + 102g &e) + Fv) + FGx) =14+24+3=6; VB= L(y) + £(-22) =0 (B-V)A ~ (A-V)B + A(V-B) ~ B(V-A) = By — 429 — Gy + 225 — 18yK 4 122H = —2yR+ Oey = Vx(AxB). ¥ Problem 1.25 v. (0) Ge = 2; Fe = 5% =0 [VT Viv = 2% + G29.8 CHAPTER 1. VECTOR ANALYSIS Problem 1.26 VV xv) = & (He - Ft) +h (Ge - Se) +B (St - Fe) = (255 - Bis) + (Se — Ba) + (See - Spe) =o. by equality ofcroderivatves. &(-20) + H(-32) + F(-2) = 0.4 yk- 327-22 + VAY Xv From Prob. 1.18: VV» Problem 127 53 2 " wx(We) =| ae ae | = Robbe ~ aba) +9 aE — ahs) + 80g ~ ads g Be by be = 0, by equality of cross-derivatives. In Prob. 1.11(b), VF = 2ey8et& + 32°yP24 9 4+ 42%y82? 8, 20 x y a VX(V=| Om aeyiet Satytet datytet = R(B- dr? y?z5 — 4- 307y?24) + 9(4- BzyS28 ~2- dayS2*) + 2(2- Bay?24 — 3 Qzy?z4) = 0. V Problem 1.28 (@) (0,0,0) —+ (1,0,0). 2:04 1,y vedi = fiz? de = (2°/3)\} = 1/3. (1,0,0) —> (1,1,0). s fvedl=0. (1,0) 9 (1,0). svedl = y8dz = do; fv-dl Total: fv -dl= (1/3) +041 (b) (0,0,0) —+ (0,0,1). 2 =y = 0,220 1jdl=deav- dl fv-d=o. (0,0,1) —+ (0,1,1). 2 = 0,y 20-9 1,2 = yd = dy $yv-dl = 2yzdy = By dy; f v-dl = 3 2ydy = yh (0,11) 9 (1,1). 2:09 Ly =z = bl = dexv-dl= 2? dz; fv-dl= fp 2? de = (29/3) = 1/3. Total: fv-dl=0+14 (1/3) =[4/3. (0) t= y= 2:04 lids = dy = dziv-dl =o? de + 2yz dy + y? dz = 2" de + 227 dz + 22 de = 40*de; Sve dls ff 4c" dz = (42°/3))3 = [473.] (@ §v-a= (4/3) - (4/3) = Problem 1.29 ay 10 > Lz = O;da = dedyij;v-da = y(2? ~ 3)dedy = —3ydedy;fv-da = -3f2 dz eydy = ~3(218) (G1) = -3(2)(2) = [22.] In Ex. 1.7 we got 20, for the same boundary line (the square in the zy- plane), so the answer is [16:] the surface integral does not depend only on the boundary line. The total flux for the cube is 20 + 12 = [32.] Problem JT dr = [ 2 dzdy dz. You can do the integrals in any order—here it is simplest to save z for last: 2 ff Ya)al« ‘The sloping surface is 2+-y-+z= 1, so the x integral is [(°~""”) dz = 1—-y—z. For a given 2, y ranges from 0 to 1—2, so the y integral is f{°-")(1—y—z) dy = [(1—z)y—(y?/2)I$- = (1-2)? - (1 2)?/2] = (1 2)?/2=(1/2)~ = + (22/2). Finally, the = integral is {2 2%(b 2+ 8)ae = (3-284 )ae = (8-449 1/60. Problem 1.31 T(b) =14442=7; Ta) =0. =[FO)-Te) UT = (22 + Ay)R-+ (4a + 229)9 + (Gy2")8) VT-dl (20 + dy)de + (42 + 20°)dy + (Gyz")de dy = de = 0. [VT ual = f2(22) de = 0, dz 0. fVT-dl = fp (4) dy SVT-dl = f} (62?) dz |. [VT -dl = f} (0) dz =0. 1, de = dz =0.[VTedl = f2(2) dy » dy = dz =0. [VT = f} (22 +4) de +4=5, (a) Segment 1: 2:0-91, y Segment 2: a1 Segment 3: 2:01, 225|) (b) Segment 1: 2:01, Segment 2: y:0-¥1, Segment 3: 2:01, = (2? +42)|, = 2015 SP VT =7.0 (©) 2:041, y=2, 2=2?, dy =drdz=2ndr. VIudl = (22 + 4a)de + (4 + 22°)de + (6x2")2x de = (10 + 142°)dx. SLUT dt = JP (102 + 1428)de = (5a? + 20") =542=7.0 Problem 1.32 Vv sy t22+30 Sevvjar = fly +224 8x) drdyde = [f {ely +224 3x) de} dyde SP [yt 222 + $27]? = 2(y 4.22) +6 S{Bey+42+6)dy} ae SD lv? + Get Oy]§ = 4 + 2042 + 6) =82 +16 = flee 16) Numbering the surfaces as in Fig. 1.20: (i) da = dy dz, (42? + 162)fp = 16 +32 = [48 dzdedz. [veda = [J4zdedz = 16. = -drdz9,y =0. veda =0. fveda = 0. (v) da = drdy3,2 =2. vida = Gr dr dy. fv-da = 24 (vi) da = ~dr dy’, 2 =0. veda =0. [veda > veda =8+16+24 = 48 V Problem 1.33 Vxv = &(0 —2y) +9(0— 32) + 2(0—2) = -2yk- 329-22. da = dy dz, if we agree that the path integral shall run counterclockwise. So (Vxv)-da = —2y dy dz.10 CHAPTER 1. VECTOR ANALYSIS Se xv)eda = f{ 0" (-2u)ay} dz e Sv =-e-2" , = ~ {24-424 2%)dz gz — 22? + ah = ~(8-8+8)=[-# Meanwhile, v-dl = (sy)dx + (2yz)dy + (3e)dz. There are three segments. . 7! =0; de=de=0.y:0-42. fudl=0. ; r= 2—y; do = 0, de = dy, y 1240. ved = Yyzdy. JP 2u(2— v)dy = ~ fo (Ay ~ 2y2)ay = ~ (2y? d= “+0. vedl=0. fred Problem 1.34 By Corollary 1, f(Vxv)-da should equal 4. Vxv = (42? — 22)% +228. (i) da = dydz%, 2 =1; 2:01. (Vxv)-da = (42? — 2)dydz; f(Vxv)da = fj (42" — 2)dz (iv) da = -dede, () das dedy2, 221; 20-41. (Vxv)eda = Dd dy, [C0 xv}da 2, 3 UV xv)-da = -F +2 Problem 1.35 (a) Use the product rule Vx (fA) = f(WxA)—Ax (Vf): [fovxa)-da= [vxa)-das [fax (vp}-de= f tad [Ia x0) aa, aed. (Lused Stokes’ theorem in the last step.) (b) Use the product rule V-(A x B) = B-(VXA)—A-(V XB): [? (wxaydr= [ v(AxB)ar+ [A (VxB)ar = f(A 2m. Seda = flrsind)(r dr dg) = fos? dr 2" dp = 2nE. Total: [vida = wR° + 20K? = Suh. ¥ vv) ArB (r(cos0 + sind cosd)) + tg & (sin 6(— sind + cos6 cos ¢)) + =a &(—sin d) 4 2r(cos0 + sin 8 cos 4) + =; (—2sin 8 cos@ + cos? B.cos 4 — sin? cos 4) — —b cos sarr2sin@fos8 + 2sin? @cos 4 — 2sin8 fos 6 + cos? @ cos — sin? @ eos } — coe 4] = _ Fahy [(sin? 6 + cos* 0) cos $ ~ cos] = 0. = lWt=0 Check: ros = 2, rsin cos = x => in Cartesian coordinates t = x + 2. Obviously, Laplacian is zero. i Gradient Theorem: [° Vtedl = t(b) — t(a) Segment 1: 0 = ¥. 6=0, 7:02. dl=drf Ved [Vedi = [2dr =2 Segment 2 0= 5, r=2, 6:05. di=rsinddsg = 200d. Veedl = (—sing)(2d6) = -2sinddg. [Vtdl = — [¥ 2sinpdg = 2eos¢|f = -2. (cos + sin cos #)dr = (0+ 1)dr = dr.sing + cont ef) (208) = 2sind dB Meanwhile, (by = la) = 2G +0) - Ol =2. v sing R + c08 69; @=% Problem 1.41 From Fig. 1.42,[8 = cos@& + sin #9; Multiply frst by cos , second by sin , and subtract: Scosg — Gsing = cos? bX + cos gsin g§ + sin’ dx — sing cos ¢: So|% = cos $8 — sing 6. Multiply frst by sin, second by cos, and add: Ssind + cos = sin pcos px + sin® dF — sin pos p+ cos* d: So|¥ =sin d8 +cosod. Problem 1.42 (@) Vv LE (ss(2 + sin? #)) +h A(esindcosd) + £(32) 2o(2 + sin? ) +} s(cos"6 ~ sn? ) +3 442sin® d + cos? } — sin? 6 +3 44sin? + cos? 6+3=[B. (b) f(V-v)dr = f(8)sdsdodz =8 fj sds fF dd fp dz = 8(2) (5) (5) = [40m Meanwhile, the surface integral has five parts: . top: 7 =5, da = sdsdp%; v-da = 3z8dsd@ = 15sdsdd. [v-da=15 f° sds [2 dd = 15x. bottom: 2=0, da =~sdsd2; veda = -32zsdedp=0. Jv-da =0. 5 é sin pcos ds dz =0. fv-da =0. ssingcos dds dz =0. fv-da=0. (2 + sin? )s dbdz = 4(2 + sin” ¢)dgdz. SPQ + sin? o)d@ [8 de = (4)(m + $)(6) = 25x. So fveda = 15m +25 = 40m. ¥ (©) Vxv = (1% G2) = &(ssin pcos) 8+ (& (s+ sin? $)) - £(82)) 6 +2 (£(6*sin pos) - & ( )e = _}(2ssin pcos — #2sing cos) = Problem 1.43, (@) 3(8*) ~ 28) ~1 = 27-6 -1=[20. (b) cose © (@) (24-3) = int = [ears] Problem 1.44 (a) 2,22 +3)45(2) de = 2043) = (b) By Eq. 1.94, 6(1- 2) = 6(@ - 1), 901+34+2=[6) Tory CHAPTER 1. VECTOR ANALYSIS Se + 3) de = 9 (-3) 0(ifa <5). 5 (a) J. F@) [2H Ale)] de = 2 F(2)6(2)|N, — Sq de (# (2) 6(2) ae. ‘The first term is zero, since 6(2) = 0 at +00; # (2 f(z)) = 2 +f = 2H + f. So the integral is — f°, (ez + f) 5(2) de = 0— f(0) = —f(0) = — f%, fl@)6(2) dr. So, e£6(e) = -6(2). ged (b) JR S@) de = F(2)0@)IZ, — (0) = J f(z)5(a) de. So Problem 1.46 (a) [o(e) = @5( —¥').] Check: fo(r)ér =a fH -r)ar =a. V Pe) = a —F) — 0). (c) Bvidently p(r) = Ad(r — R). To determine the constant A, we require Q= Jodr = [Ad(r ~ R)4nr? dr = A4nR®. SoA= 73x. | alr) = znd(r - R). Problem 1.47 ~ ———— (a) a? + aa +0? =[30%.] (b) fle)? feS%(8) dr = gh = He +99) = [F] (0) 2 = 25 +944 = 38 > 36 = 6°, s0 € is outside V, so the integral is [2276.] (@) (e- (28 +29 +28)? =(1R+09-+ (-1)4)? =14 and hence the integral is e-(d — e) = (3,2,1) Problem 1.48 First method: use Eq. 1.99 to write J = fe" (4nd%(x)) dr = Ane Second method: integrating by parts (use Eq. 1.59). 2, La(e)de = f(co) ~ fg" Las = F(c0) - (H(ce) - (0) (z). ged Ge ~ fz venars fe Fda But Vier) = (Fer) peeve =f perinet ars fer5 Poinoandge =n femrarye-® fsinadoag 4n(-e- +e") = dev (Here R= 00, 60 e-* = 0.) =n (0) + de Problem 1.49 (a) VF: = £(0) + £00) + & (2! “ETE z [O} vre= SR Son VxXFi -| =[-225} oxr.-| ose ogee Sygje0 8 few eae15 Fy iva gradient; Fi isa curl] [U2=}(@?+y* +2)| would do (F2 = Vl). Ay = Fs An = = 0 would do it. For Ay, we want (22 — 9dr) = (2-H) =o, Se — Me ‘Ai = 429] (Fi = VxAu). (But these are not unique.) Ry ) VFs = £02)+£e2)+ Rly) =0; VXFs=| ew yz ze ay So Fy can be written as the gradient of a scalar (Fs = VUs) and as the curl of a vector (Fs = VAs). In fact, [Uy = sy2 | does the job. For the vector potential, we have (@-2)+9 (y—y) +8(2~2) ye, which suggests Ay = dy2 + fle); Ay fs Go mas, sugzoning Ay = Hote + Meus A he aay, 20 Ay = dey + R(y2)s A fea? + (v2) ey? +1(e,y) 4y2? + o(2.9) | Putting this all together: Fe (=v) Ry (@ — 2) 9 +2 (v*—2*) 8} | (eeain, not unique). Problem 1.50 (@ > (a): VxF = Vx(-VU)=0 (Bq. 1.44 - curl of gradient is always zero). (a) > (©): §F-dl = J(VXxF) “da = 0 (Eq. 1.57-Stokes’ theorem). (09 O) SP,B d= PF d= LF -a+ fp Fd $F d= 0,80 : . [Fa-[ Pa ler fer (©) > (c): same as (c) + (b), only in reverse; (c) = (a): same as (a)= (©). Problem 1.51 (@ > (a): VF = V-(VXW) =0 (Eq 1.46—divergence of curl is always zero). (a) => (): $F «da = [(V-F) dr = 0 (Eq, 1.56—divergence theorem) (©) > (b): fF -da— f,,F da = $F -da=0, 50 [Pedee [Ped fi hs (Note: sign change because for $ F-da, da is outward, whereas for surface IT it is inward.) (0) = (c): same as (c) + (b), in rovorse; (c)=> (a): same as (a)=> (c) - Problem 1.52 ~ In Prob. 1.15 we found that V-vg = 0; in Prob. 1.18 we found that Vxve = 0. So Ve can be written as the gradient of a scalar; va can be written as the curl of a vector. (a) To find t: () =v sta ste+ Soy) @) & = (Qay+ 2") (3) 3 = ve16 CHAPTER 1. VECTOR ANALYSIS From (1) & (3) we get $f = 2yz => f = yz? + gly) t= ye ty2* + 9(y), 90 ft = Qry +2? + B= 2xy + 2? (from (2)) > $f =0. We may as well pick g = 0; then [t= ay? + yz" (b) To find W: Se — Sil = 22; le — Ba = gee; Ola Oe Pick W. = 0; then a2. ow, 2 322 oe W802? = We = 2072? + Hly,2) ea = -2ee = Wy = 222 + 9(y,2). |. May as well pick f = 9 = 0. & (28) +9 (S24) +.2(-202).7 You can add any gradient (Vt) to W without changing its curl, so this answer is far from unique. Some other solutions: W = 22-229; W = (2zyz + 22%) 842792; W = zyz— fate g + $a? (yet) &. Probelm 1.53 overs 1 Fy (77 #? 0086) + ind vv = ind? cos) = = Lase 1 Pee 1 att? cos 0 + —— c0s8r? cos $+ 3 a —? cos sin) (-#? cost cos4) rcos@ 6 [4sin@ + cos 6 — cos 4] = 4r-cosé. a sp “p [omer 5 [dr contyy sin ce dap = [2a [ontense [oo = (2) @) = Surface consists of four parts: (1) Curved: da = R? sind di dot; r=R. v ka (R? cosd) (FR? sind daa) . 1p 7p wn footie f ase m (3) (2) =28. a eaedl(2) Left: da = rar a9 (3) Back: da = r dr d8 }; |. v-da = (rcosdsin g) (rdrd#)=0. fv-da=0. /2. v+da = (-r? cos@ sin ¢) (rdr dd) = —r3 cos dr a. [vas [re Fova-- (ie) on= (4) Bottom: da = rsin dr dp; 0 = 7/2. v-da = (r? cos¢) (r dr dg) . Romp Ie dam [Pdr f conpae = ie ® Problem a 5 & vxv=| 2 % &/|=2(b-a). So f(Vxv)-da=(b—a)nR?. ay be 0 = yk +b29) (Gok dy + de8) = ayde + body; 24 y? = 10 = Dede + 2ydy = 0, s0 dy = ~(z/y) de. 80. v-dl = ayde +b2(—2/y) dx = } (ay? —b2?) dx For the “upper” semicircle, y = VRE = 2%, so v-dl = ~~ de, Fak (a+ d)2" 4 (oneein® 2 re in (2) foa= f[ dem {or sin R) -@+9[- SRP + sin @LL, R a = fRO-9 soe = one = 8 (ain“¥(-1) —sin"(4) = Fe (-2- 4) line = 5rR*(—a) ‘And the same for the lower semicircle (y changes sign, but the limits on the integral are reversed) 80 fv-dl=2R%(b—a). ¥ Problem 1.55 ()2=2=0; de=dz=0; y:091. ved (y +32) dy = yal 1 1 [va a (2) 2=0; 252-24; de = —2dy; ys1-¥0, v-dl= (y + 32) dy + 6de = ydy ~ 12dy = (y ~ 12) dy. [var fu-ma--(}-x) @)2=y=0; de=dy=0; 2:240. v-dl=6dz, [ran [oan 1 yay = 5. 1 =-7+218 CHAPTER 1. VECTOR ANALYSIS Total: fv-d=}-}412-12=[ Meanwhile, Stokes’ thereom says § v-dl = (xv) -da, Here da = dy dz, so all we need is (Vxv)e= £6) - 2(y +32) =0. Therefore f(V xv) -da=0. ¥ Problem 1.56 Start at the origin. () @=§, 6=0; 720-91. veal = (ros? 8) (dr) fved=0. (2) r=1, = 5; 6:0->4/2. v-dl = (3r)(rsinOdg) =3de. fv a=s p= ¥. (8) 6= $F rsind =y=1, sor = gly, dr = zahy cosas, 0: FF. cos? 8 (- £058 ) d theta ~ 2289508 yy in? @ sin?6 cos ay) @=-e® see) 8 (seeeeete sind ‘Therefore Total: on fracoete} Stokes’ theorem says this should equal (WV xv) - da 1 a = [Fyne — Fc- ineos)] t xv Reva] 6 Fain 1fa a. an) st +2 [Pe-rrcosesina) ~ F (roos'a)] 6 1 raind = 3cot0#-66. (1) Back face: da = -r dr d9 $; (Vxv)-da=0. f(Vxv)-da=0. (2) Bottom: da = —rsin@ dr db6; (Vxv) -da = Grsin 8 dr dd. @ = ¥, 80 (Wxv) - de [sreosa}é + 46rd + 2 -arcosdsind + 2reos0 sind] 6rdrdd Lon _ gr DIT19 Problem 1.57 vedl=ydz. (1) Left side: = (2) Bottom: dz=0. ‘Therefore f'v-dl = a-#; dz=—-ds; y=0. Therefore fv-dl=0. (3) Back: By de = —1/2dy; y:2040. fv-dl= fy (—}ay) =~: Meanwhile, Vxv = &, so [((V xv) da is the projection of this surface on the zy plane = } -a-2a =a. ¥ Problem 1.58 ~ _ 12a 18 ga? 1 dy, w= a5 Fant pg (684? cos) + a ae (7? tan) 1 1 j 4r (ent 29 — sin? = spar sin + —— Ar? (cos? — sin? ) = <7 (sin? @ + cos* 6 sin? 6) gp to = and” (0 [ova J (2) (P? sind de dd di) = [ote Pcotow fao~ (Rt) (2m) & + =| 2 a =" (+94) 16 i ff (r+ 33). Surface consists of two parts: () The ice cream: 10-4 2x; 8:0-+ n/6; da = R?sinO d0d0f; v-da = (R°sin@) (R? sin 0d0d9) = Risin? 4.d0 a6. [vda= (2) The cone: 0 = 3; :0-+ 2m; 7:04 R; da=rsinOdpdr6 = BrdrdpG; v-da= Br drdp =R wea foo RY) (2) [i- Jina)” = 2rRt (% ~ jsneo") = me (--») v-da= V3 [ra fae= va ans Dee omens ‘Therefore f v-da = 28° (¢— 8 + v3) = ff (or +3V3). Vv Problem 1.59 (2) Corollary 2 says §(V7)-dl = 0. Stokes’ theorem says #(VP)-dl = [Vx (VT)]-da. $0 [[Vx(VT)}-da = 0, and since this is true for any surface, the integrand must vanish: Vx(VT) = 0, confirming Eq. 1.44.20 CHAPTER 1. VECTOR ANALYSIS (b) Corollary 2says $('V xv) -da = 0. Divergence theorem says f(V xv)-da = ['V-(W xv) dr. So f V+(V xv) dr 0, and since this is true for any volume, the integrand must vanish: V(V Xv) = 0, confirming Eq. 1.46. (a) Divergence theorem: $v-da = [(W-v) dr. Let v = eT, where ¢ is a constant vector. Using product rule #5 in front cover: V-v = V-(eT) = T(V-c) + ¢- (VT). But ¢ is constant so V-e = 0. Therefore we have: Je-(VT)dr = fTe- da. Since ¢ is constant, take it outside the integrals: ¢-fVI'dr = c+ fT da. But ¢ js any constant vector—in particular, it could be be &, or $, or 80 each component of the integral on left equals corresponding component on the right, and hence [vrar= fran aoa (b) Let v -+ (v xe) in divergence theorem. Then f V-(v x e)dr = f(v xe) da, Product rule #6 => V.(v x €) = (Vv) ~v-(VXe) =e: (VXv). (Note: VXxe = 0, since ¢ is constant.) Meanwhile vector identity (1) says da (v x e) = €- (da xv) = —c+ (v x da). Thus fe (Vxv) dr =~ fe- (v x da). Take ¢ outside, and again let © be &, , # then: [ovxar=- fv xda. aed (c) Let v = TVU in divergence theorem: ['V-(TVU) dr = [T'VU-da. Product rule #£(5) = V(TWU) = TV(VU) + (VU) (VT) =TV2U + (VU) (VT). Therefore) / (rV°U + (VU) (VT) dr= f (PVU)-da. ged (4) Rewrite (c) with PU: f (UV?T + (VT)-(VU)) dr= [(UVT)-da. Subtract this from (c), noting that the (VU) -(W7) terms cancel: J (9°U - UV?) dr = I (IVU -UVT)-da. ged (e) Stoke's theorem: [(V xv) -da = fv-dl. Let ?. By Product Rule #(7): Vx(eT) = T(Vxe) — x (VT) = ex (VT) {since © is constant). Therefore, ~ f(c x (WT))-da = f Te- dl. Use vector indentity #1 to rewrite the first term (e x (VI) -da = ¢-(VT xda). So — fe-(VI x da) = fe-Tl. Pull e outside, and let + %, 7, and @ to prove: [vrxdn=- fra oot Problem 161 (a) da = R? sin dd db. Let the surface be the northern hemisphere. The & and ¥ components clearly integrate to-zero, and the & component off is e038, £0 af in? 9 0/2 = | wsindcosoda dia = ann? 2 [sin deowe dd = anf a8? = [nie a, (b) Let T= 1 in Prob. 1.60(a). Then VT = 0,80 fda=0. qed. (c) This follows from (b). For suppose a; # api then if you put them together to make a closed surface, fda =a; — ag £0. (a) For one such triangle, da = }(r x dl) (since r x dl is the area of the parallelogram, and the direction is perpendicular to the surface), so for the entire conical surface, a = 3 fr x dl.a1 (@) Let T = ¢-r, and use product rule #4: VT = V(c-r) = € x (Vxr) + (c- V)r. But Vxr = 0, and (-V)r= (eek + eyZ teaR Nek +9 = 28) = cok +0y9 +6,2 =e. So Prob. 1.60(¢) says fta= fie-na= Problem 1.62 @ [oxina=-ex [tanexazane oat 2)-kgo- For a sphere of radius R: J (hf) (Rsin@ d9d6e) J (4s) (7? sin ard da) Sveda S(V-v) dr So divergence theorem checks. Rf sinddddé = 4nR. (f«) ([sin@ d0 dé) = 4nR. | 3 Evidently there is no delta function at the origin. n 9 (a,n) 19 1 fe = Vx 8) = Sg (Pe) = a5 OM) = Gln + De = [nt 2)E (except for n = 2, for which we already know (Eq. 1.99) that the divergence is 4n8%(r)). (2) Geometrically, it should be zero. Likewise, the curl in the spherical coordinates obviously gives To be certain there is no lurking delta function here, we integrate over a sphere of radius J, using Prob. 1.60(b): If Vx(e"#) = 0, then f(Vxv)dr = 02 -fvxda. But v = r°f and da = Fi'sin 949 dé are both in the f directions, s0 v x da = 0. ¥Chapter 2 Electrostatics Froviem 7: () [Eas ()|F ZS. where r is the distance from center to each numeral. F points toward the missing 4. Explanation: by superposition, this is equivalent to (a), with an extra —9 at 6 o'clock—since the force of all twelve is zero, the net force is that of ~g only. (©) [Zero 1 9@ ‘i issit 7 (€)| <= 4. [pointing toward the missing 9. Same reason as (b). Note, however, that if you explained (b) as ine F cancellation in pairs of opposite charges (1 o'clock against 7 o'clock; 2 against 8, ete.), with one unpaired g doing the job, then you'll need a different explanation for (A). Problem 2.2 = (a) “Horizontal” components cancel, Net vertical field is: B, = 1=2cos8. 2 T ey Hore s? = 2? + (§)° ; cos@ = j, so] B= 2g, a Anco (28+ (§)")"” ‘When z >> d you're so far away it just looks like a single charge 2g; the field agT ga should reduce to B = z4-22. And it does (just set d -» 0 in the formula). (b) This time the “vertical” components cancel, leaving 2 sin 8%, or 7 a SBsoerOT From far away, (z > d), the field goes like B ~ qf &, which, as we shall se, isthe field of a dipole. (If we set d +0, we got E = 0, as is appropriate; to the extent that this configuration looks like a single point charge from far away, the net charge is zero, so E -+ 0.) 223 Problem 2.3 BE, = yt Mt cos8; (0? = abate ceils = oa [otra = dato Be = mate Band = ah cats = aa [- atell 5 5 ; leew? )§+ (zen) 4: For z > L you expect it to look like a point charge q = AL: B-> term 0, and the 2 term —> 72-342, Problem 2.4 From Ex. 2.1, with L + $ and z ~> y/2? + (8)? (distance from center of edge to P), field of one edge is: Aba, Te checks, for with 2 > 1 the & 1 da There are 4 sides, and we want vertical components only, so multiply by 4 cos@ Ey Daz se +8) (+e “Horizontal” components cancel, leaving: E = gL- { f4f! cos 6} 2. Here, o? = 12 +24, cos@ = # (both constants), while fll = 2nr. So 1 _AQar)z Ee tra 24 ay Problem 2.6 Break it into rings of radius r, and thickness dr, and use Prob. 2.5 to express the field of each ring. Total charge of a ring is 0 2ar- dr = X-2nr, 80 = cdr is the “line charge” of each ring. 1 (odr)2arz Fre (2 4 2)?” Ea 1 BR r meh a 1 1 1 we = Dron |= — eens 8. Bowe = reg "|e IEE =z24 CHAPTER 2. ELECTROSTATICS For >» 2 the second term -> 0, 60 Epiane = qig2no% =| 8. Por 2 R, oghew = tare)” wi (1-38), s01]6 and B= ji tie = 1G, where Q=nRo. Problem 2.7 Fis clearly in the 2 direction. From the diagram, dg = oda = oF? sin0 d0 49, a? = R? + 2? ~ 2Rzcosd, cosy = =Aee, So 1 f aR? sin@ dd dé(z ~ Roosd) “tro | +e WRewsge | S462 2 a {2 Reos§) sind cost, p 6=0>u=41 = glare) f wo Integral can be done by partial fractions—or look it up. [=a et = aagtnPa) [5 For z > R (outside the sphere), E. For 2 < R (inside), Be Problem 2.8 According to Prob. 2.7, all shells interior to the point (i.e. at smaller r) contribute as though their charge were concentrated at the center, while all exterior shells contribute nothing. Therefore whore Qin is the total charge interior to the point. Outside the sphere, all the charge is interior, so 1@Q ter Inside the sphere, only that fraction of the total which is interior to the point. counts: Qn = ola ie Problem 2.9 (8) p= 0 V-B = coXeh (0? a). Problem 2.18 From Prob. 2.12, the field inside the positive sphere is E,. = sfr., where ry. is the vector from the positive is —zf;r-. So the total field is center to the point in question. Likewise, the field of the negative sphere ix 2 Een) ze! [rx (3) pdr (since p depends on r’, not r) , from Prob. 1.62). yf QVxE=kF F | = B[R(2z — 22) + 9(0 — 0) + 2(2y — 2y)] jy) ay ta? ys] 0 Ey is a possible electrostatic field. e Let's go by the indicated path: Ev dl = (y?de + (22y + 2?)dy + 2yzde)k (205 90520) Step I: y= 2 = 0; dy = dz =0. E- dl = ky* de =0. Step Il: x = 29, y :0 + yo, 2 = 0. dz = dz = 0. E-dl=k(2zy + 2*)dy = 2kroy dy. Jy Bo dl = Ake fo ydy = kaoyg. T Step III: x = x0, y = yo, z : 0 ~> 29; da = dy = 0. ad 1028 CHAPTER 2, ELECTROSTATICS B. dl = 2kyzdz = 2hyos de. Sip B+ dl = 2yok f° zdz = kyo23. (co.yp.z0) V(20,¥0, 20) = — [ E+ dl = —k(tou3 + vo2s), or [V(x ys 2) = —K(ay? + v2’). (ao-402) 94 (ev tye) Smale? e+ Oeyts) 94 2ye tee Check: —wv mul tes 492") 24, Problem 2.21 vo=-cea { Outside the sphere (r > R) : Inside the sphere (r < R) : So for r > R: V(r) oa (ataih) a= ate and for r < R: V(r) = JE (aoe Joo Harn (* When r > R, WV = eB (LE When r < R, WV = goad (3~ Problem 2.22 E = ,L:28 (Prob. 2.13). In this case we cannot set the reference point at 00, since the charge itself extends to'o0. Let’s set it at s =a. Then clear why a = 00 would be no good—likewise the other “natural” point, a = 0.) aig 20g (In (3) 8 = ~ G2 = EV Problem 2.28 VO) = ~ {QB d= — (Ear ef SEP )dr — [2O)dr = £OG9 — F (In (Z) +a (2-3) -149}5/2in Problem 2.24 Using - 2.22 and the fields from Prob. 2.16: V(Q)-V(0)=- (E-dl=—fPE-dl- fPE-di= = Sl + se malt = Problem 2.25 (@)29 OV= aot, Age = = Ag inet VFAE, : i\ =| Ain | EAE | A on (eevee a = oe | wc : = dhe i RM = alg one (VA FHE = |Z (Va =2). In each case, by symmetry $Y = $Y =0. a |- ieee age meesgy 2a (-4) wan? 4 (agrees with Prob. 2.2a) B= ~a8s (agro tale - cores} * = oats voter (GE) 0 (B= mae {bomb - 12 =] b 5 rel | (agrees with Prob. 2.6). Ifthe right-hand charge in (a) is ~a, then [V = 0}, which, naively, suggests B = —VV = 0, in contradiction with the answer to Prob. 2.25. The point is that we only know V on the z axis, and from this we cannot hope to compute B, = —9Y or By = 9. That was OK in part (a), because we knew from symmetry that E, = E, = 0. But now E points in the x direction, so knowing V on the z axis is insufficient to determine E. Seen ae vos ef (2) a= BE gvin= (where r = 4/V3) - NEA v= * (2) a whore # = fi? +32 — Vita. XA oe Grea J2Jo Vit 4 a a ~ vin a ane, : E “Wie [i+ J tansavn— Vth) — _ 24v2\_ oh, ((2+v2)? -2e(258) eo (e) 2 V¢_) - V(b) = $* [1-4 v3] 2| (agrees with Ex. 2.1).30 CHAPTER 2. ELECTROSTATICS Problem 2.27 L Cut the cylinder into slabs, as shown in the figure, and use result of Prob. 2.25c, with z+ 2 and 0 ~» pdz: 4 v= ue a ° (VRE EaE 2) de SS ch = ovR + Rina + VRE a) — 29/2 * Problem 2.28 Orient axes so P is on 2 axis. Here p is constant, dr = r? sin @dr db dp, V = he Satis 5 "dd = 2x. Bree) |0 = 2 (Vee HEI - VP FF) nf Heese} So Tere areca 8 = fe (VF? =2ete-h31 13 ° But p= qqu, 80 V(z) = ah aff (RP ) Problem 2.29 = - VV = ALV?f(S)ar = zs Sole’)(V*E)dr (since p is a function of r', not r) ate Jole!)[An6°(0— v)] dr = — 2 ole). ¥ Problem 2.30. (@) Bx. 24: Expove = 3258 Evetow = 36,8 (A always pointing up); Esnove — Evetow = Za. ¥ Ex. 2.5: At each surface, E = 0 one side and E = £ other side, so AB = £.¥ Prob. 2.11: Boye = 285# = £7; Bia = 0350 A= £8. ¥ ©) SSF) Outside: $B + da = E(2n8)l = LQme = LQnR)l +E £6 (at surface). Inside: Qune = 0, 80 B= (© Vous = Be = Bf (at surface); Vs Mas = Bg = 2 (at surface); 28a = 0; 90 Mas — On = Problem 2.31 ©, xO) # ot me | (0) W, = 0, Wa = hy (=); Wa = ay (ee - S)s We = (eee (a))- @ 3) | ig L 1 2g? 1 Waa deh (A -t-2+ de} =| ae (298)32 CHAPTER 2. ELECTROSTATICS Problem 2.32 (a) W =} JoVdr. From Prob. 2.21 (or Prob. 2.28): 1a /f® ad atk 0-m = ye 4 ~ = Fe Te SY =3 (0) W = 4 fE%dr. Outnde (r > R) E else 2 sy) -49{(Df+s @}- ak (k+ ae) mike (0) W = ${ f,VEB-da+ f,B%dr}, where V is large enough to enclose all the charge, but otherwise arbitrary. Let’s use a sphere of radius a > R. Here V w # (a!) (gph) rameanass [" Bars f° (Gi data} = $(gope 2 Grey 1 @fi 1 =aei(atsn-a As a+ co, the contribution from the surface integral (;-$:) goes to zero, while the volume integral (a5 ECG - 1) Picks up the slack. Problem 2.58 aw = av = a(e ys (q= charge on sphere of radius r). neo) 1 4 gatnipsaty = © on sphere) a= pro apy (q= total charge on sphere) ¥ d= amrtdr p = AT gar = Shree ‘ = dartde p= pede = Fe _ 1 (a) 1 (349 1 3g 4 =a (+ (BP) = meee Lae (® 4 1 BPR Fre FE Jy 7 Gre ROS Ineo33 Problem 2.34 ()W= fk B= Ao (a 8), and hence f By «Ep dr Woe = Wit Wa + ¢0f Ba «Badr = ggg? (b+ Problem 2.35 (@lon= Gi w= onal rat! © = abe (0) VO) =~ JB dl = ~ f(g B)dr — [Oar — SE (ahs) ar ~ ae =| (©) [av 0] (the charge “drains off”); V(0) = — 2 (Oar — "(25 &)dr — [2(0)dr Problem 2.36 @ fat @ Fait ||°° = “Gere where r = vector from center of large sphere. where ra (ts) is the vector from center of cavity a (6). y= Se, Freer} Problem 2.37 Betwomn the plates, Z = 0; outside the plates Z = o/¢9 = Q/eoA. So poop 2G |e Problem 2.38 a Inde, B = 0; outside, Bue babe ets fe ooisee Se as Fe = Sfeda = f(qSex)}(qhg Re) e088 R? sin 8 dd dep aoa = aby (afa)"2n Jo” sin cos9.d0 = 2-(H)* (sin? IE” = ah)? = dea3 CHAPTER 2. ELECTROSTATICS Problem 2.39 Say the charge on the inner cylinder is @ for a length L. The field is given by Gauss's law: [B+ da = B-2n5-L= LQune = 2,18. Potential difference between the cylinders is elt pear Ls ) Qf rail ca ‘As set up here, a is at the higher potential, so V = V (a) ~ V(®) = z,%7 In (2). Bre mG) V(b) -V(a) = c= HEEB: 80 capacitance per wnit length is ‘oblem 2.40 (a) W = (force) x(distance) = (pressure) x (area) x (distance) =| £° 5" Ae, (b) W = (energy per unit volume)x(decrease in volume) = (0¥) (Ae). Same as (a), confirming that the energy lost is equal to the work done. Problem 2.41 From Prob. 2.4, the field at height 2 above the center of a square loop (side a) is esas os eee 4reo (224 8) ay E= Here \ + 0 4 (s0e figure), and we integrate over a from 0 to a 1 @ Let u= ©, s0.ada = 2du. am Coit «age fhe ~ Treo ee write “() 2+ co (infinite plane): B 2% a (point charge): Let f(a) g -' VIF — §, and expand as a Taylor series: Sle) = FO Fes 0+ s+35 Here f(0) = tan“*(1) - $= ¥- $ =0; (2) = ey bs = aa TO) Se) = fet (a? + Oat + ‘Thus (since #4 =2 <1), Bx 2e av Problem 2.42 = 1 @ (Bsin@cosg + anaes PEF*)} SHA — Being) Problem 2.43 - From Prob. 2.12, the field inside a uniformly charged sphere is: E = z2- fr. So the force per unit volume is f= p= (7955) (aeSqs)F = 2 (aths)’s, and the force in the z direction on dr is: =hedr= 2( a ? cosd(e? z 9) 10086 (r? sin 8 dr di dé). ‘The total force on the “northern” hemisphere is: Fr = [tare & (ate) ar ["cososinoen "a6 3 (_Q \*(RY) (sito 3Q7 co inns) \4 2 Io GaneoR® Problem 24a 1 o lo Voom = Fes J = Teg R L o “ da = 2nR? sind do, You ta etn 1R2(1 — 08 8). 1 ofaeR®) p7/? _sinde =a EL eet ig oR oR oR (1 * Vpote ~ Veenter =| 5—~(v72 — 1). Fo 9 Fieg _Yoe 340/7-) Problem 2.45 : First let's determine the electric field inside and outside the sphere, using Gauss’s law: foar= [ony sinaaranag = snk [ar = ( R)36 CHAPTER 2. ELECTROSTATICS SoB= rrt(r< Rs B= fist (r>h). Method I: wae [oar (@a.248) = 2 [ “(ey savers 2 [” (BES) sorte 4c, nrg(A) {freee du) (Eee ()E}-8 Eoe) | TERT daa We 5 [over (ea. 243) Bor ), where s_ is distance from \—37 Total Now s, = Va) +2, and s_ = Vy tay +2, 50 i —(b) Equipotentials are given by 44;4% = e(teo¥e/) = = constant. That is: P+ 2ay +02 +2? = ky? — 2ay +a? +2) = y2(k— 1) + 2%(k 1) +.0%(b — 1) —2ay(k +1) =0, or poeta? —2ay (#4) ‘The equation for a circle, with center at (yo,0) and radius R, is (yo)? +27 = RE ory? +2" + (xB - R?) — 2yy0 = 0. Bvidently the equipotentials are circles, with yo = a(f4}) and 2 pene < Rb Roa yQ a? wa? (B81)? — a = ohana Rb RO = ya? =a? (HY) eiabbemtaED) 5 or, in terms of Vo: Sie oF ebco¥e)® 4.1 @2ReoV0/® 4 gr Dtea¥a/ gircoVe]X — | ~ SeineaVe]h — em BweaValR R= awit GT = Sam RE) ) o\ ae | TE Problem 2.48 ~ 7 an () VV =—£ (Eq. 2.24), 80 (0) gv = mv? + |0 ‘Apu = 1] (constant). (Note: p, hence also I, is negative.)38 CHAPTER 2, ELECTROSTATICS av. ~avn ae = (Note: 1 is negative, so 8 is positive; q is positive) (6) Multiply by V’ = 4: dv" dv ao Vw gy [vaya [vay 37 =290°P + oni But V(0) = V'(0) =0 (cathode is at potential zero, and field at cathode is zero), so the constant is zero, and V2 a4pvi? > x 2VBV4 > v-Mav = 2B de; [uw =2vB [ar > wes = 2VBe + constant. But V(0) = 0, s0 this constant is also zero. vi = Va, 50 V(e) = (v8) “28, or V(2) = (f0)""s" = (Seman. a Interms of Vo (instead of I): | V(e) a (3)"" (see graph). Without space-charge, V would increase linearly: V(x) = Vo (3)- vt ay 1 41 -as _|__4¢0¥o Vol ~a ge = ona 37 =| aaa snthout Vive -[ vac)" fm mga, = at Ve 19 \'P . (0) = Yo= (BER)? V8 = BBE = Mucayiysr — ya, 4e0A T= Siatve?? = Kye”, where| K = So Problem 2.49 en (b) [Wes.] The field of a point charge at the origin is radial and symmetric, s0 VxE = 0, and hence this is also true (by superposition) for any collection of charges. © Vee [= a= al a(+)e = mil 3 Ble Seam gf [pears “herar} © B= aa [a0 "hap39 Now fret/Mdr = 22" — 4 pe" dp ¢— exactly right to kill the last term. Therefore — feca= ade (148) ernsen =£ (148) emo on vont { i (©) Does the result in (d) hold for a nonspherical surface? Suppose we R make a “dent” in the sphere—pushing a patch (area R® sin@ dé dp) from radius R out to radius $ (area $? sin 6 d9 dé). q a frnan= {2% (v4 8) 2r4sranoanay — (148) 7 ce ano anay} = [04 Seem (48m) snore 1@ f es, sind ,drd0dp = 3," sinoaoag [ re-t!¥ar b= Fines 4 7 =~ ij sinodods (7 (14+ 5)) : =-f(14 8) esa (142 =f (1+§)¢ 148 So the change in zr JV dr exactly compensates for the change in fda, and we got 1g for the total using the dented sphere, just as we did with the perfect sphere. Any closed surface can be built up by successive distortions of the sphere, so the result holds for all shapes. By superposition, if there are many charges inside, the total is LQme. Charges outside do not contribute (in the argument above we found that © for this volume $B -da-+ 3s JV dr = 0—and, again, the sum is not changed by distortions of the surface, as long as q remains outside). So the new “Gauss's Law” holds for any charge configuration. (0 In differential form, “Gauss's law” reads: or, putting it all in terms of E: 1 a ee ands vE~ 55 [Bea Zp. Since B= —WV, this als yields “Poison’s equation”: -V°V + 35¥40 CHAPTER 2. ELECTROSTATICS Problem 2.50 p= 60 V-E = 2 (az) = [Ga] (constant everywhere) The same charge density would be compatible (as far as Gauss’s law is concerned) with E = ay, for instance, or E = ($)r, etc. The point is that Gauss’s law (and VxE = 0) by themselves do not determine the field—like any differential equations, they must be supplemented by appropriate boundary conditions. Ordinarily, these are so “obvious” that we impose them almost subconsciously (“E must go to zero far from the source charges”) —or we appeal to symmetry to resolve the ambiguity (“the field must be the same—in ‘magnitude—on both sides of an infinite plane of surface charge”). But in this case there are no natural boundary conditions, and no persuasive symmetry conditions, to fix the answer. The question “What is the electric field produced by a uniform charge density filling all of space?” is simply ill-posed: it does not give us sufficient information to determine the answer. (Incidentally, it won’t help to appeal to Coulomb's law (B =e Segedr)—the integral is hopelessly indefinite, in this case.) Problem 2.51 Compare Newton's law of universal gravitation to Coulomb's law: Now, G = 6.67 x 10-1! N m?/kg?, and for the sun M = 1.99 x 108 kg, R = 6.96 x 10 m, so the sun’s gravitational energy is W = 2.28 x 10" J. At the current rate, this energy would be dissipated in a time W _ 2.28 x10" FSS [2.87 x 107 years] 5.90 x 10"al Problem 2.52 First eliminate z, using the formula for the ellipsoid: 2) = arab THETA FOTO TT OTD Now (for parts (a) and (b)) set c+ 0, “squashing” the ellipsoid down to an ellipse in the zy plane: —— 2 Saab T= TOF = OT (I multiplied by 2 to count both surfaces.) Randers Vere, (@) For the circular disk, set a= (b) For the ribbon, let Q/b = A, and then take the limit b -> oo: (©) Let b= 0,7 = Vy? +2, making an ellipsoid of revolution: ‘The charge on a ting of width de is ; dg =o2nrds, where ds = de® + dr? = dey/1 + (ar/day. : ard gg H 2 SE pads = defi 4 OE mde VE] FTA. Thos aor ane, Ma) = 2 =2ar. 2 Plt + Te (Constant!) of we | : ® G o . i) R (else q would be inside), so V7? +0? — 2Ra = 12 1 aan Le = 9,0 -\[eta- em = gq la~ R) ~ (ot RY = 3(-2R) - [atry44 CHAPTER 3. SPECIAL TECHNIQUES (c) The force on q, due to the sphere, is the same as the force of the image charge q', to wit: eer ee ee 4xeg (ab)? ~ Srey To bring q in from infinity to a, then, we do work aR f ire R treo Problem 3.8 Place a second image charge, q", at the center of the sphere; this will mot alter the fact that the sphere is an epupotential, i aed but merely mereuse that potential from zero to Vy = oS a = dreqVoR at center of 5 For a neutral sphere, g! +4 = 0. ag ( 1 Fe w(t tne at * GH = ). _ (Rafa) (R?/a)(2a~ R?/a) Pay? (Drop the minus sign, because the problem asks for the force of attraction.) Problem 3.9 ~ (a) Image problem: \ above, —A below. Potential was found in Prob. 2.47: * = = ings? ot vs Vins) = Ge inte fou) = GA Into fot) * -[2o({Bee | area" Lye (e—d? ov av _ av = (0) 0 = a5. Here = evaluated at 2 = 0. a 1 L a) = FR (eer o~ areca} S -2{ a --S} 44 _. ae Pah Pee wey Check: Total charge induced on a strip of width I parallel tothe y axis: 28 [Feo I A, as it should be. Dd For ma = 8 | ape = =A. Therefore Aina45 Problem 3.10 ‘The image configuration is as shown. ven = {7 Pt For this to work, { must be and integer divisor of 180°.] Thus 180°, 90°, 60°, 45°, etc., are OK, but no ethers. It works for 45, say, with the charges as show 48" he (Note the strategy: to make the x axis an equipotential (V = 0), you place the image charge (1) in the reflection point. ‘To make the 45 line an equipotential, you place charge (2) at the image point. But that screws up the 2 axis, so you must now insert image (3) to balance (2). Moreover, to make the 45° line V = 0 you also need (4), to balance (1). But now, to restore the 2 axis to V = 0 you need (5) te balance (2), [abd eo on why work for @ = 45° 5 fo z00d ‘The reason this doesn’t work for arbitrary angles is that you areeven- 198" tne 80 tually forced to place an image charge within the original region of 2 interest, and that’s not allowed—all images must go outside the re- ® sion, or you're no longer dealing with the same problem at all.) ia 2) Problem 3.1 - eral From Prob. 2.47 (with yo +d): |V = ae 2 [eee where o? = yo? — RY => VF, and acoth(2neo¥o/) = \ cays eos Brave { SREY Sh} = (aivcing) 5 = cose (779%8), or Ja = ITT. Problem 3.12 - V(2,y) = Do Cye"™"*/* sin(nny/a) (Bq. 3.30), where Cy Voly)sin(nzy/a) dy (Eq, 3.34). ; 4M, for 0 Im S72", where Z = e~™(#-™)/4, Now 135. =o In (Ret) =F anr+io),ar where Re! = 142, ‘Therefore penne 1 1 14Z Te tm {Fann +s0)} = to but (23> jis" tine rare/a 14 2ie™**/9 sin(ay/a) — [ra e-ne-innial Lt en7#/* (etrv/a — emerv/e) © _ 2e772/*sin(xy/a) __2sin(my/a)___sin(my/a) tan = aeeja = Gases ~ sinha) Therefore (sey) an ¥ hy ea = 0, with boundary conditions bh @ V(@,0) =0, Volv) Gi) V(@a) =0, vio 2 Gi) V@,y) =0, (iv) V(b,u) = Volo). As in Ex. 34, separation of variables yields V(z,u) = (Act + Be“) (Csin ky + Dos ky) Here (i)=+ D = 0, (ii)=+ B = ~A, (ii) ka is an integer multiple of mz V(z,y) = AC (aaa = ernie) sin(nay/a) = (2AC) sinh(nmz/a) sin(nzy /a). But (2AC) is a constant, and the most general linear combination of separable solutions consistent with (i), fi, i) is V¢ew) = 3° Casinh(nnz/a) sin(ary/a, a Itremains to determine the coefficients C,, s0 as to fit boundary condition (iv): © 6x sinh(nab/a) sin(nmy/a) = Vo(y). Fourier’s trick = G, sinh(nnb/a) = = f Vo(y) sin(nay/a) dy. ‘Therefore 2 On = TSatoenb [a Vou) sin(ery/a) dy.48 CHAPTER 3. SPECIAL TECHNIQUES 2 ; 2M 0, if nis even, () Cn = aaaeas” | sored or amr *{ 8, tenes: } aN sinh(naz/a) sin(nay/a) COO 2 errr a lem 3.15 Same format as Ex. 3.5, only the boundary conditions are: @ V=0 when (i) V=0 when Gi) _V=0 when (iv) V=0 when () V=0 when (vi) V=Ve when ‘This time we want sinusoidal functions in x and y, exponential in 2: X (x) = Asin(kx) + Beos(kx), Y(y) =C'sin(ly) + Deos(ly), Z(z) = BeVP FP: + GeV, ()> B= 0; (ii) k= nx/a; (iii) D = 0; (iv) 1 = mma; (v)=> E+G =0. Therefore 2(2) = 2Bsinh(n Vn? + sa) Putting this all together, and combining the constants, we have: Vienne) = >> =f Crum Sin(nxa:/a) sin(mry/a) sinh( n/n? + m2 /a). Me 4 It remains to evaluate the constants Crm, by imposing boundary condition (vi): w= ey [Cam sinh(w Vn? + m)) sin(naz/a) sin(mny/a), According to Eqs. 3.50 and 3.51: if n or mis even, Crm sink. (-viF +m?) = (2) vf feotorsortssiaracay={ Sieve, aan } oo nm’ Therefore Veena) = “He yy SY i tnons loping la) Oe) a9 Problem 3.16 Pala) recs @e-n= a) (2-1)? =4 ioe -1)? = Gx [lev se “0 = gh lle? 0)? -1442")] = EE le -1) (2-0) = § fee 62" - 1) + (2* -1) 102] =} (Ge — 24 50° ~ 52) =F (102* ~ 62) =| 32° - Bo, ‘We need to show that Pa(cos 8) satisfies me (x00 1(1+1)P, with = 3, where Fy(cosd) = } cos (50s? @~ 3) . Bi =F [-sin0 (Se0s*@ ~ 3) + cosd(10c0s6(—sind}] = ~Fsind (Seos* 9 — 3 + 10.0s" 4) 5 ~ sind (Scos?# — 1). ) = 3 [sino (508? # —1)] = -3 [2sind cose (5<08% 9 1) + sin? #(—-10¢0s0sin)] = Seer ae (sno) = —3cos6 [5.c05? —1 — 5 (1 — cos? #)] = —3 e088 (10.cos? # — 6) a 8-4-3 e050 (5c0s?@ ~ 3) = -1(L+ IDPs. ged 1 [Perc -2)[|,= 20-11-9074 eI Problem 3.17, (@) Inside: Vir,8) = JA! P(c0s8) (Bq, 3.66) where (+1) oR [ Vo(0)P;(cos6) sin 8.40 (Ea. 3.69) a In this case Vo(0) = Vo comes outside the integral, 50 tM | Picoat) sino, 250 CHAPTER 3. SPECIAL TECHNIQUES But Po(cos@) = 1, so the integral can be written, [ Remnntcansneas={ 5 wine } (Eq. 3.68) ‘Therefore _f% ifl#o a={%, meat Plugging this into the general form: V(r,0) = Aor? Fo(cos 0) = [Vo ‘The potentials constany thoughout the sphere Outside: V(r.) -E z P-Pi(cos) (Eq. 3.72), where B = AED pie [ Vo(8)Pi(cos6)sinbd (Eq. 3.73) (2141) pest , 0, leo = BED vo f aicseo anos = { ee ee } j ‘Therefore Vir,8) = V0= (ie. equals Vo at r= R, then falls off tke 2), (b) Sane eee V(r,8) =) 1D , ), forr > R (Bq, 3.79) where Bi = RA, (Eq. 381) and A= Ee 20(0)Pi(cos6)sin@d# (Eq, 3.84) _ i 0, ifl 40 = agai | Plco00) sind { Rooleo, if1=0 }. a Therefore Foo torr R voy=4 ° Reel torr >R oF51 ‘Note: in terms of the total charge Q = 4x Rao, V(r) = Problem Vo(8) = keos(38) = k [4.os* 8 ~ 3 cos] = k aPs(cos®) + AP, (cos 8)] (I know that any 3°? order polynomial can be expressed as a linear combination of the first four Legendre polynomials; in this case, since the polynomial is odd, I only need P, and Ps.) 4.0086 — sono = al} (5c0s* @ - 3cos6) + Acosd = "2 ex?a + (9 - 3a) cond, 8 3 2 § -=6-3a=8- Therefore ; Vo(0) = § [BPs (e088) ~ SF; (c0s0)] Now wo LaAr'Fi(cose), forr< R (Eq, 3.66) vere =4 ie B , Di gavPi(cosd), forr > R (Ba. 3.71) where A= 'Vo(6)Fi(cos6) sind (Bq. 3.69) = pt {a [meanncunsnsar-s /ripnicansesa| = E@+) I, 2 nee = Sa Soren Saray} = Fae os 86a) 8k/5R*, ifl=3 a FE { aera, itl=1 } erro otherwise) ‘Therefore V(r,8) = fa E{a(Z)°} [scos?@ seme] -3 coso {1 (E)’ [scost o— 3) - s}52 CHAPTER 3. SPECIAL TECHNIQUES (for r < R). Meanwhile, By = A:R%*" (Iq, 3.81—this follows from the continuity of V at R). Therefore _f skRt/s, itt=3 . mua { SEI, EI=2) rote So SHE A poo) 4 HA Vira) ; P,(cos 4) [ (2) sco-a(2)' rien), v(r,8) ~4 (2) coos (2) ooo ea] -} (for r > R). Finally, using Ba. 3.83: or (0) = eS (2+ 1)AURM1P (cos 0) = 69 [9A,P, + 7Aa REPS] r= cos (- 3) » +7( SE) ner] =| 2k 07; (c0s0) + 5674(cos8)] 7 BR, BR bri * = SE [-acon0 + 52 (Scan a 3c80)] = £086[-9 + 28- 5.08? @ — 28-3] os 6 [140 cos? 6 - 93] Use Bq. 3.88: 0(0) = S(2L41) ARF (e080). But Ba. 3.60says: Ay = & Problem 3.19 Fr | Yo(OP (0086) sina Putting them together: 210) = $14 PC AComs), with Gm [ YD AIome) sin. aed Problem 3.20 Set V =0 on the equatorial plane, far from the sphere. ‘Then the potential is the same as Ex. 3.8 plus the potential of a uniformly charged spherical shell:Problem 3.21 (0) Vir) = 3° PEA (cos0) (r > R), 80 V(0,0) = m PQ) = Pe = [va] Since r > Rin this region, VP 7B =r VIF RF =r [ee FR)? — Lenny. ] fe Zora De [ta ee ae te Coaparng tke powers ofr, Tee that By = =, By = 0, By = 2... terete 7 eo voy = ELLE rpesey...], os (for r > R). = 2[-48) (Bcos?@— 1) +. |: (b) V(r,0) = $5 Air! Pi(cos) (r < R). In the northern hemispere, 0 < @ < x/2, ad Veo) =D Aut =a Vet]. Since r < Rin this region, Vr? +R? = RYT + (F/RP =R ph + S/R) = Flr/R +. | . Therefore Sat aig [ns Rares Comparing ike powers: Ag = 57-R, A= —9-, Aa = 5 op... +80 Yon = [RP (co + so Pa(cons) +.) (for r < R, northern hemisphere). = Ele (Remesd (F) Coote) +...], In the southern hemisphere we'll have to go for = x, using P;(—1) = (~1)! Vinn) = Een = i [ a Re ]34 CHAPTER 3. SPECIAL TECHNIQUES se = +(0/20), Ao = Ao, 4; put an overbar on A; to distinguish it from the northern A,). The only difference is the sign of Ay: (for r < R, southern hemisphere). Problem 3.22 Sarr A(c080), (F< R) (Ba. V(r) = where B, = AR™+1 (Eq. 3.81) and ane) Tan [ MAC sngaw (Bq. 3.84) 3.78), S SPE Alcose), (+> R) (Ba. 3.79), aS = oF siro{]n (e056) sind dB — fr (cso) ava} (let = = cos) EA = mie {fame [ne}. Now i( | Plz)d and hence 6, (-1)'F,(z), since F(z) is even, for even I, and odd, for odd 1. Therefore [rt-aa-2) = cat f nies, : : 1 if Lis even a , a= gdint-con froee| ais [ Rlerds, tts oad ;55 So Ay = Az = Ag = Ag =0, and all we need are Ai, As, and As 5 7008 = 3 (63% — 72 4 15%)| [row = 2S (632° — 702" +18) de = 3 (00% 0% +6F)|, 213515) 1 a = 3-F+3) 7605-35) = 36 ‘Therefore and Thos Vir.) = Problem 3.23 19: (,ov) 107 305 \" os) * ae Look for solutions of the form V(s,4) = S(s) (4): 1gd@ (4s 1.@e led (otf) 4st ao Multiply by s? and divide by V = Sé: $f (,85) , Lee 37a as) * ag Since the first term involves s only, and the second ¢ only, each is a constant: sd/ ds ee peace =C,, with +0 = age56 CHAPTER 3. SPECIAL TECHNIQUES Now Cz must be negative (else we get exponentials for &, which do not return to their original value—as ‘geometrically they must— when ¢ is increased by 2m). es =k, Then 2S = i? = Acos sin kg. CG; GF = HES = Acoskd + Brink. Moreover, since ®( + 2x) = (4), must be an integer: k= 0,1,2,3,... (negative integers are just repeats, but & = 0 must be included, since # = A (a constant) is OK). ot (3) = K°S can be solved by S = 4”, provided n is chosen right: ngs! dws dn ok (sng!) = ne (9%) Evidently the general solution is S(s) = Cs + Ds-*, unless k = 0, in which case we have only one solution to a second-order equation—namely, $ = constant. So we must treat k = 0 separately. One solution is a constant—but what’s the other? Go back to the differential equation for $, and put in k = 0: as _o ds a (4s as 65h (#98) <0 08 = constant = 0-9 as = 0% + $= CIms + D (another constant) So the second solution in this case is In s. [How about ®7 That too reduces to a single solution, & = A, in the case k = 0, What's the second solution here? Well, putting k = 0 into the & equation: ee ae SS =0 S = constant = B+ = BS+ A we cH But a term of the form B¢ is unacceptable, since it does not return to its initial value when is augmented by 2x] Conclusion: ‘The general solution with cylindrical symmetry is [V(s,6) = do + boln s + > [9 (ax coskp + by sin kd) + 8° *(cy coskd + dy sin kd)]. | mi Yes: the potential of a line 1s included. Problem 8.24 Picking V = on the yz plane, with Bg in the x direction, we have (Bq. 3.74): 5 wv when s = R, — ay v ~Foscos¢, for $> R. —>~ Evidently a = by = by = dk =0, and ay = cy = 0 except for k = 1: V(0,) = (a8 + 2) cos. (= ey = —a,R*; (i) 9 a1 = Bp. Therefore V(s.4) (#08 BE) cose, or= [Real 008 4.] R Problem 3.25 ~ Inside: (3,8) = a9 + D> s* (ax coskd + de sin kd). (In this region Ing and s* are no good—they blow wpats=0,) “ Outside: V(s,) =o + S~ 3 (cx.coskd + dy sin kg). (Here Ins and s* are no good at s -> 00). Vous _ Via sy ( Os ) oS {rs (ox cost +d sino) kRE (ax cosk6 + by sin ko) } (Eq. 2.36). ‘Thus asin 5g = Evidently ay = cy = 0; by = dy = 0 except k = 5; a = Sey (ws + try). Also, V is continuous at s = R: at Bsn 59 = d+ yd sin 5p. So ag = dp (might as well choose both zero); Sj = R-¥ds, oF dy = Ry. Canbinng thos results: @ = Sea (Rds + Ry) = 10eoRDy; bs = sap: ds = SE. Therefore _ asin5d [ °/R, fors f 20a. 29\(F oe {[creosoy(ecos0)R? sin a ab = 267 [ os? asin do = arte (. oe) eee 4nFk payne Garett — Cys (b) we _1_ 4 R°k 0080 _[ RR? cosd Vie 3 | Bey | (Dipole)39. ‘This is also the exact potential. Conclusion: all multiple moments of this distribution (except the dipole) are exactly zero, Problem 3.29 Using Eq. 3.94 with r! = d/2: os as Dy (2) races: fora. we let 8 -» 180° +8, 50 cos ~+ ~ cos8: (-1)"Pa(z), 80° 1) 118 ay" -2)- = #3(5) [Pa(cos@) — P,(—cos8)] = es PS (3) P,(c0s8). Vag = Zt} Ap oos0) = SEE? wile [Vana =O eS (2) Fy(cos6) o- St (6.cos" 9 ~ 3c088) = we Fes (6cos? 0 —3c0s8) Probiem 3.30 @) () @=[2a] Gi) p = [Sqaa, >) @ Q=([2¢] Gi) p=[ga2, OMQ= = [Baad] (ayy (rom Eq, 1.64, 9-8 = sin sin ¢). ‘Problem 3.31 ~ (a) This point is at r= a,0=§, ¢=0,s0E= F=qE=|-—% 2. Treva? () Herer =a, 9=0,s0 B= = aw OV =41¥(0,0,0) ~ V(a,0,0)) = 5 Problem 3.32 — = gad, =p ES. Therefore Q==0, 80 Vinone = Guts P= and, 80 Vip = GCP”. Theret 1, acosd : faa |v (r,0) = a (-3 > ). E(r,6) = ro (2sser rained} |60 CHAPTER 3. SPECIAL TECHNIQUES Problem 3.33 =(p-f)F+(p-6)6 2pcosF + psind@. So Eq. 3.104 Problem 3.34 cos 0# — psin86 (Fig. 3.36). So 3(p-#)F ~ a. 3.108. ¥ 3pcos0# ~ pcos # + psind A night # above th pantheon qian by Be. F= gto mE, BE aye te ee ee eres ere whee A ga Moki by = Ge = BG & (50) =a But » =O when = =o comtat = ~A/d and tee v= 242-2 ie ‘This integral can also be integrated directly. Let x = u?; dz = 2udu, Therefore Problem 3.35 ‘The image configuration is shown in the figure; the positive image charge forces cancel in pairs. The net force of the negative image charges is: ~ q 1 1 oP * Pata@—a * far %e-ay TH ¥ (same as for only one plane: Ieee 1 ¢ Frey 2x)? F When a + 00 (ie. a > 2) only the 4 term survives: F = Eq, 3.12). When 2 = @/2, ass (5a/2)? |- lege sor ie (a/2)? © (3a/2)% (5a/2)? ae {lear * marél Problem 3.36 Following Prob. 2.47, we place image line charges —) at y = and +A at y = —b (here y is the horizontal ais, z vertical). In the solution to Prob. 2.47 substitute: atb a-b\?_(atb\? .»_,_ R hae tba (25) 2 (OH) rere, v= deb)om(-som(@) [tay + 27I[(y - 7 +2] = Ao (fee et aig y= oad 2 and — [Ay (Fe Bas cos Iles] R)? + RE — Pas cond] = |daeg {G+ a — 2ascos ¢)[(as/R)? + K+ 208.089] J” Problem 3.37 Sinoe the configuration is azimuthally symmetric, V(r,@) = > (ae! + BL) r(cose) {a)r>b: A, =0 for all L, since V +0 at oo. Therefore V(r,6) =D FEA (cose) acr (c# + #) P,(cos 6); De : ar=0, itt #0, oo (ce +a)A ae Coa? + =P = Vo, if = 0; Pating (2) into (1) gives By = B8G, —a2#"1C, 1540, By = 60p + a aC, Thoreore pa (= a)Co + a% B= BC, + Dy] (0) Oia! + ~a™1G, 1F0,. ao — aCo, @) Gi) => DBC + Wl ges cose) - (ca + GED GD a, (cas + n=GED) 20, wr 4;62 CHAPTER 3. SPECIAL TECHNIQUES (41) Br ~ 10 + (14 1D, = 0; (1-4 1)(B ~ Dy) = - 1010), Bae + (a +z ) = 4 fort=1; G+ BB -Di)= ‘Therefore (4 {B= Di) + WC = O,for #1, Cr+ BUD) = fe) Plug (2) and (1') into (3): For 14 0. or 1: (tt) [0 = 2) C4 aC) 41 HC, Therefore (1') and (2) + [B= Gr for l>1 (41) HG, For b= G+ 2 (PAG +8G) = C420, = > [GaMSa] i = -0'G, > —a%k/3eo;| Br = (6° - a) Cy | By = (0° - a5) k/3e0. Por = 0: Bo~Do = 0 = By = Do = (= a)0p-+a¥o = aVs 00s, $0 BC = 7 (#-08)k avo, ke Conclusion: | V(r8) = Fag 8 |r 2d. |VUe8) = The ge ov ave $ (ou) = ~c0 SC] =~ [- + K (142 ) coe] -o(-B+ coed) (= f ovda = "24na? = [aracoVe = Qi] At larger: Vw Problem 3.38 Use multipole expansion (Iq. 3.95): pdr + Adz = dz, and r+ 2: vr) a = [nrcmog me ® 2a ‘The integral is 2m antl Zrateos @) for n even, zero for n odd. Use separation of variables in cylindrical coordinates (Prob. 3,23) V(s,d) = ao + bolns + D> [s* (ax cos kd + by sin kd) + s~*(cy cos ke + dy sin kd)] a63 8R: V(s,d)= De, s*(ce coskd + dy sinks) (Ins and st blow up as § > 00). (We may as well pick constants so V + 0 as s + oo, and hence a = 0.) Continuity at s = R => YRMencoske + desinkp) = S>R-K(cy cosh + disinkd), 80 ce = Ra, de = RP*by, Eq. 2.36 says: > Fate coskd-+ dysin kd) ~ S> RR (ay coskd + by sin kp) = ze Ture “anconk + sinks) = { ore eit Fourier’s trick: multiply by (cos/4) dé and integrate from 0 to 2n, using a ao , =; sigdea{ FAL i kp cost gd = 0; es kpcoslbdd { } m kal "I Then "sind 7 aut-tsa= 28 [ fostoao- Joma -2{s4 Multiply by (sing) dp and integrate, using J sin kdsinlbdd = { RAL } m kel ny = 2 frinigas frintgag] = 22 {S| PY 0 oes UR nh = af todo f wa] -2{ sto" , oo b= ge 2eosIn) 9, if Lis even _f 0, if Lis even 2 { Soa /ten if Lis odd }on-{ 2oo/nel@R'*, if Lis odd }. Conclusion: 2ooR 1 (s/R)* (8< R) \ | E went faye Gm} Prt 340 1_ SA Palcosd), Use Eq, 3.95, in the form V(r) = 28) 5, == (#) + (Monopole.)64 CHAPTER 3. SPECIAL TECHNIQUES ® an ao tf esincns/a) de = { (2)" sin (22) ~ % cos (#2) }]" we) { (£)" int ~ sin'-n) = crt) = Fen) } = PE; V(r) = & (-) ty (3eos*9—1).| (Quadrupole) Problem (a) The average field due to a point charge g at r is 1 4 Ex (roms / where E= a 8 fkec a 80 Base = Tico) ane (Here r is the source point, dr is the field point, so 4 goes from r to dr.) The field at r due to uniform 1 charge p over the sphere is E, = 7 fp dr. This time dr is the source point and r is the field point, 50. goes from dr tor, and hence carrie the opposite sign. So with p = ~g/ ($x*), the two expressions agree: Eyye = Ey. (b) From Prob. 2.12: —P Fregh® (¢) If there are many charges inside the sphere, Faye is the sum of the individual averages, and prot is the sum of the individual dipole moments. So Eae = —7—Pos. ged FreoR™ (d) The same argument, only with g placed at r outside the sphere, gives 1 (4nRp) eo Ew £ (Geld at dve to uniformly charged sphere) = 31-286 But this is precisely the field produced by q (at x) at the center of the sphere. So the average field (over the sphere) due to a point charge outside the sphere is the same as the field that same charge produces at the center. And by superposition, this holds for any collection of exterior charges. Problem 3.42 @) } Eup = Frapys P0808 + sind 8) P i eae : Fraps Peosslsin eos AR + sinBsin dy + 00808) + sin O(cosBcosgX + cos@sin ¥ — sin 8)] = phy |ssi K+ Bi i 29 — sin? = paseoreoe +3sinOcosOsin dg + (2cos* @ — sin? @) (ta) [a Bat [cosa = [sin dap =0, vo the & and J terms drop out, and Fag = 2n, s0 3sin 8 cos 8(cos OR + sind $) + (Bcos? @ — 1) #]r? sind dr dO dé. (&)™ fra J (costa) sina. {_____, (cor? 0008 )g=1—1 Evidently [Eye = 0,] which contradicts the result of Prob. 3.41. [Note, however, that the r integral, [ bar, blows up, since Inr -> —oo as r > 0. If, a8 suggested, we truncate the r integral at r = ¢, then itis finite, and tho # integral gives Baye = 0) (0) We want E within the e-sphere to be a delta function: E = AS%(x), with A. selected so that the average field is consistent with the general theorem in Prob. 3.41: Eve wy i AB (x) dr 2, and hence Problem (ay ‘Teor + (VVe) dr, But V «(Vi VV2) = (VV) (WV2) + Vi(V7¥A), so t= [v-cavuyar— fucorm)= f views) -dar 2 [ vied. But the surface integral is over a huge sphere “at infinity”, where Vi and Vs +0. So T= 2 J Vipadr. By ‘he sume argument, with 1 and 2 reversed, J = 2 [vinar. 80 [ Vindr = f Yon. ar. aed66 CHAPTER 3. SPECIAL TECHNIQUES {a (1): Qa= fyardr = f, pr dr =0; Vis = Vas } (b) Situation (2): Qa =f, p2dr Qu = J,padr =Q; Veo = Vie SVipa dr = Vie f, pr dr + Vis fy o2 dr = Vas. UNewwteteehee. a. } Green's reciprocity theorem says QVab = QVie» 80 Vab = Vie-_aed Problem 3.44 (a) Situation (1): actual. Situation (2): right plate at Vo, let plate at V =0, no charge at 2. yoo v= sta: [Viera = VQ + VosQ00 + Var i But Vi, = Vp, = 0 and Qz, =0,s0 fVipodr = 0. J Yea = Ve + Vases + UnQn- But Vig =0 Qe, = 4) Veo = Vos Qry = Qz, and Vey = Vo(z/d). So 0 = Va(2/d)q + VoQ2, and hence =a]. Situation (1): actual. Situation (2): left plate at Vo, right plate at V = 0, no charge at 2. [Vindr=o= [vide But Van = Yo (1-9), 80 faQts + Vea Qer + ViaQry = VoQi + WVer +0. Qi = —a1-2/d). (b) Situation (1): actual. Situation (2): inner sphere at Vo, outer sphere at zero, no charge at r. J Yred 1. So fVipadr = 0. = Vax Qaa + Vix Qra + Vir Qo But Va Qn= [Vee de = Vines + Vir + Vas, = Qao + an +0. But V;, is the potential at r in configuration 2: V(r) = A+ B/r, with V(a) = Vo = A+ B/a = Vo, or aA +B = aVo, and V(b) = 0 > A+B/b = 0, or bA+ B = 0. Subtract: (6-a)A = —avy > A= —aVo/(b~ a); B (2 — }) = Vo = BYS® = B = abVo/(b— a). So V(r) = 7fl4; (£ - 1). Therefore Oat rags (-2) = BE (3-1):67 Now let Situation (2) be: inner sphere at zero, outer at Vo, no charge at r. J[Yierdr=0= [ Vopr dr =VonQn + ¥isQns + Vinh, = 0+ Wen + QiNo This time jsplaystyleV(r) = A+ & with V(a) =0 = A+ B/a=0; V(b) = Vo > A+ B/b = Vo, s0 % @ ie @ @ Vey= is (1-8). tharetre os (1-2 4). Problem 3.45 ~ 1s 1 Sal Se. ae ‘ ® i> 04-3 | [sree Sena} rw i pst rey 2 + Ee = fer! =r! cos6! = Le 5 1 Youd = Fe ali (Paste —r9) per (b) Because 2? = y? = (a/2)? for all four charges, a Because z = O for all four charges, Qez = —(/2a/2)( This leaves only Gey = Que =8[(5) (3) + (5) (+3) 0+ (-5) G) -0 (-5) (9) ~ Be © Dba = ff =f-f=1. So 1 ga [1° Palcos0)p ar’ (the m = 2 term in Ba. 3.95). Quy = [3(a/2)? - (v2a/2)"] (9-4-9 +9) a+) =O and Qsz = Qyz = Quz = Qey a = f [Bins —ah)(ry — ds) — (r —4)°%j] pdr (11 drop the primes, for simplicity.) f [aren ~ 196,)] pdr ~ 34k [rode 5 aad; [ ripdr + 3dd, foa +24- roars, = #6, f par = Oy — Beans + dyn) +9dd,0 + 2848p - P54 Soifp=0 and Q=0thenD, =Qy. aed (@) Bq. 3.95 with n = 3: Voce = Fee [Oo Pa coserpar' Pa(cos6) = ; (5c0s* @ - 3c0s0) Define the “octopole moment” as Os ree68 CHAPTER 3. SPECIAL TECHNIQUES Problem 3.48 a = vr +a? = 2racosd, a = vr +a? + 2racosd, 3 = Vet +i Brbcos8, oT Vr? + 0 + Prbcos, tava asi Bx 310: (— 3) m Men (ne wane ot 3 fy (Bq. 3.15), s0 ea) [bf sod— 2s 2 2 ona] = 2 (22) (Bont eee Set B= Tat | field in the vicinity of the sphere produced by +g): V6.0) = ~Ba(: E) owe (agrees with Eq. 3.76) Problem 8.47 ‘The boundary conditions are (i) V=0when (i), V=Vo when y Gil) V =O when x (wy) Vv Go back to Eq. 3.26 and examine the case k = 0: @?X/dz? = d?¥/dy* But this configuration is symmetric in 2, so A = 0, and hence the k D=0,C =Vo/a, and subtract off this part: V(z,y) = Vol + V(2,9). when ), 0 X(z) = Ar+B, ¥(y) = Cy+D. solution is V(z,y) = Cy + D. Pick The remainder (V(x, y)) satisfies boundary conditions similar to Ex. 3.4: () V=0wheny=0, (i) ¥=0 when y=a, (ii) 7 = —Vo(y/a) when (iv) 0 = —Vo(y/a) when =(The point of peeling off Vo(y/a) was to recover (fi), on which the constraint k = nx/a depends.) ‘The solution (following Ex. 3.4) is V(x,y) = D> Cu cosh(nxz/a) sin(nzy/a), and it remains to ft condition (ii): V(b,u) = 0 Cn cosh(nnb/a) sin(nny/a) = —Vo(y/) Invoke Fourier’s trick: Y ca cosh(nnb/a) i Cn cosh(nnb/a) = “2 “fp ysin(nxy/a) dy. n(ny/a) sin(n'ny/a) dy = -* f * yoin(n'ny/) dy, c west lee) sin(nry/a) — (2 %) cosier/e)]| 2M 2M ~ arcatary (fe) 0" = Fe cae Vey) = 1)" cosh(nz/a) sn Problem 3.48 (@) Using Prob. 3.14b (with b = a): Vee) = % 3 sabe sing) proud ‘nsinh(na) avy) _ Ne ‘nm cosh(nz/a) sin(nxy/a) ot) = 05g =e De, (FT) pata) Lene ata sin(nny/a) - Lae sin(nty/a) sinh(nm) 460g 1_ fs, ie ene 2 Sao f sin(nry/a) dy. But fe sin(nay/2) dy = ~© cos(nay/a)lg = [1 ~ cox(n)} = 22 ince m is oda) BeaVo 1 stb ing, [have not found a way to sum this series analytically. Mathematica gives the numerical value 0.0866434, ‘which agrees precisely with In 2/8.)70 CHAPTER 3. SPECIAL TECHNIQUES Using Prob. 3.47 (with b = a/2): _yfu.2 1)" cosh(nn2z/a) sin(nxy/a) V(e.u) =Vo [t + z pr eosh(nn /2) | fer eee ee eee ote) = BE mats b+ 25 () remmeselerereen _ 1)" cosh(nmz/a)] __ €oM% * cosh (nnz/a) a or! ia ty cite ~fe 2 ee IF an a ey ph he [2 -2 [erate "conos(oa] aa en But f” “nmbors|a) de = 2 shnrz/0) - 5 2 sintina/2). -2% [er ie : (aushierte) a [: + ty {cociatien “ah ne (Again, T have not found a way to sum this series analytically. The numerical value is -0.612111, which agrees with the expected value (In2— n)/4.] (b) From Prob. 3.23: V(s,¢) = a0 +boIns+ D> (cust + nz) [ce cos(k) + dy sin(k¢)]. In the interior (s < R) bo and by must be zero (Ins and 1/s blow up at the origin). Symmetry => dy = 0. So V (6,0) = a9 + Sans" cos(kg). At the surface: V(R#) = Loa coho) = { 9 Hyer # < 2/4 ko . * Fourier’s trick: multiply by cos(k’d) and integrate from —m to m: Vosin(k’¢)/# ae = (Vo/k')sin(k'x/4), if k #0, 3 -RE i ' Vo cos(k') dg xe J cost) cost’) a6 = c (Odo = { wenn: But 0, itkeR cataentide={ On, ik m, fk=8 20.n So Bray = Vor/2 => a9 = Vo/4; nay = (2Vo/k) sin(kx/4) = ax = (2Vo/mkR*) sin(kn/4) (k # 0); hence V(sd) = Ve [i+ > Jom] Using Bq. 249, and noting that in this ease & (6) = 0 =r ‘eos(ke)| Bath Sst )cos(kd). f ‘We want the net (line) charge on the segment opposite to Vo (—m <@ < -Bn/4 and 3x/4< @ R until the end: z in(kr/4) kot aa) = ae = Mae Esso 18) be cost) 20k §* = kL? * sin(kx/4) cos(k@) (where x = s/R— 1 at the end). Mz) = = [2 ty) (Pee ~ cy Qn 3 5 a\2 6 10 Problem 8.49 Now consider the pendulum: F = -mga —T'#, where T — mgcos = mo?/L and (by conservation of energy) mgl cos = (1/2)mu? = v? = 2glcosg (assuming it started from rest at ¢ = 90°, as stipulated). But cos 4 = — cos, 80 T = mg(—cos6) + (m/l)(~2gl cos) = ~3mg.cosé, and hence —mg(cos0# — sin 9 6) + 3mgcosot ng(2cos0F + sind 6). This total force is such as to keep the pendulum on a circular arc, and it is identical to the force on q in the field of a dipole, with mg + gp/4reol?. Evidently q also executes semicircular motion, as though it were on a tether of fixed length 1.Chapter 4 Electrostatic Fields in Matter | i Problem 4.1 [2 = 500/108 = 5x 108. Table 4.1: a/4meo = 0.66 x 10%, so a = 4 (8.85 x 10-!9)(0.66 x 10-89) = 3 |. p= ab =ed > d=aB/e= (7.34 x 10-)(5 x 108)/(1.6 x 10-1) = 2.29 x 10-8 m, a/R = (2.29 x 10-*)/(0.5 x 10 4.6 x 10-®.] To ionize, say d= R. Then R= aB/e=aV/er > V = Rez[a= (0.5 x 10°19)(1.6 x 10-19)(10-*)/(7.34 x 10-") = [10*V. Problem @2 First find the field, at radius r, using Gauss’ law: [E-da= 1 Qene, or B= Giz dsQene- Lem (e--¥) Jerre" (8) [Note: Qune(r -? 00) = q.] So the field of the electron cloud is Be = 2% [h rete (1425+ 28)]. ‘The proton will be shifted from r = 0 to the point d where E, = E (the external field): eal aay 2 e = gaa [Pett (1425 +255) Expanding in powers of (d/a): ct a EAH) arta) 4) remo) = EY AS ote fo + of eG ah — og BAe : _ + higher order terms. — a B74 CHAPTER 4. ELECTROSTATIC FIELDS IN MATTER 3) = ara saa” [2S deat [Not s0 different from the uniform sphere model of Bx. 4.1 (see Bq, 4.2). Note that this result predits =a = 30? = 3 (0.5 x 10-")® = 0,09 x 10-°m?, compared with an experimental value (Table 4.1) of 0°68 x 10-2 m?. Ironically the “classical” formula (Bq, 4.2) is slightly closer to the empirical value] Problem 4.3 — alr) = Ar. Electric field (by Gauss’s Law): fE-da = B (tar?) = Que = 3 ff AFAnt* dr, or B= 1 drArt Ar Gir? eT te a: ad? /4ey = E => d = V4eqEJA. So the induced dipole moment is p = ed = 2e/éofAVE. Evidently pis proportional to BY? For Eq. £1 to hold im the weak-field limit, E must be proportional to r, for small r, which means that p must go to a constant (not zero) at the origin: [p(0) # 0} (nor infinite). ‘This “internal” field balances the external field E when nucleus is “off-center” an amount Problem 4.4 Field of q: giz f. Induced dipole moment of atom: p = aE = | poy q ag regr? wh) (to the right). Force on q due to this field: Problem 4.5 Field of pi at po (0 = 1/2 in Eq. 3.103): Ey 6 (points down). a ‘Torque on pa: Na = po x Bs = paBi sin90° = abi =| PP | (pointe into the page). | Field of pa at ps (0 = in Eq. 3.103); Ba = 7? (28) (points to the right). ‘Torque on pr: Ny = pi X Ea 2 (points into the page). image dipole as shown in Fig. (a). Redraw, placing p, at the origin, Fig. (b. Problem 4.6 (a) Freeney 2080 + sin); p=pcosd + psindd. « N = pxB= aacagys [0008 + sind) x o0n08 + sin0d)] __P ee re é = Freq [0088 sind} + 2sindeos6(— 8)] _ Psindcosd oft = ates (8) (ont ofthe page)Psin20 Freo( i625) For 0 <0 < 1/2, N tends to rotate p counterclockwise; for x/2 <0 < x, N rotates p clockwise. Thus the stable orientation is perpendicular to the surface—either T or |. But sin @ cos @ = (1/2)sin26, so| N= (out of the page). U = -prEs, but Ba = Problem 4.9 () P= (p- V)E (Eq. 4.5); B= Problem 4.7 Say the field is uniform and points in the y direction. First slide p in from infinity along the x axis—this takes no work, since F is 1 dl. ff (if E is not uniform, slide p in along a trajectory . the field.) Now | ly rotate (counterclockwise) into final position. The torque exerted by E is N = pxE = pEsin@%. The torque we exert is N = pEsin@ | fi 2 clockwise, and dé is counterclockwise, so the net work done by us is | negative: U= [0, pEsin di = pE (-cos8)|- , qed | Problem 4.8. 4s [3 (po-f) # — po]. So U = 53 [pr-p2 — 3(px-f) (Pa-f))- aed a _aktyy +28 treo Fy FHA B= (ogg tegy +g) BaP TATE a ita eran argtoae| + 2 ara vn desman) -ateSeennona] a 28). Fe = ga e- 3(p -#)#] () B= G23 lp: (-8](-#) ~P} = gy [P-#) 8p]. (This i from Ba, 3.104; the minus signs ‘are because r points toward p, in this problem.) 5 (3(p-#)# - p] a ca (Note that the forces are equal and opposite, as you would expect from Newton's third law. Problem 4.10 1a ER] p= -V-P = - 55 Belt? kr) hate! = [SE (0) Forr < R, E= gi prt (Prob. 2.12), so E = [=(k/) F. Forr > R, same as if all charge at center; but Qtor = (KR)(4nR?) + (-3k)($R*) = 0, so[B=0.76 CHAPTER 4. ELECTROSTATIC FIELDS IN MATTER Problem 4.11 ps = 0; 4 = P-fil= +P (plus sign at one end—the one P points toward; minus sign at the other—the one P points away from). (i) L> a. Then the ends look like point charges, and the whole thing is like a physical dipole, of length L and charge Pxa®. See Fig. (a). (i) L , Seo? i ino ea alae ; ° So V(r,0) = neg J 2! 1 (4/3), con es, ce. ppe=| Pree) ce) Problem 4.13 ‘Think of it as two cylinders of opposite uniform charge density -tp. Inside, the field at a distance s from the axis of a uniformly charge cylinder is given by Gauss’s law: B2nst = Lpns?€ = E = (p/2e0)s. For two such cylinders, one plus and one minus, the net field (inside) is E = B, + B_ = (p/2c0) (sy —s-). But 14. — 8. = =d, so E = [—pd/(2eo),] where d is the vector from the negative axis to positive axis. In this case the total dipole moment of a chunk of length £ is P (nae) = (gna) d. So pd = P, and {E = —P/(2¢0),] for sa. Problem 4.14 Total charge on the dielectric is Quot = f, onda + f, podr = $,P-da— fy V-Pdr. But the divergence theorem says f, Pda = fy V-P dr, $0 Qen« aed Problem 4.15, . 2 (ok) ke Jo (ar =a), @a=-vP 35 (P2)=-4: { ka (atr =a) } Gouss’s law => B= gh. Sue #. For r b, Qenc =!) (4a?) + J (gh) Antdy = —4nka — 4nk(r — a) = 0 (Prob. 4.14), s0 [ Amr; s0 [B= —(/eor)®. (0) $Deda = Qjag, = 0 D = 0 everywhere, D = @B +P = 0 E = (-1/e)P, s0 [E=0(forr 6);| [B=—(k/er)# (ora metal plate.) This is true in both slabs; D points down. es () D=a8> By = 20/3e0. E=o/e: in slab 1, E (c) P = coxeE, 80 P = coxed/(€otr) ‘/€2 in slab 2. But € = cor, 80 €) = Deo. (Note: D = 0 inside the 205 €2 = Seo. [Fi = 0/240] (d) V = Bya + Bya = (0a/6e0)(3 +4) = (©) = 05 —P, at top of slab (1) a» = +P; at bottom of slab (1) (Xe/6r)0; Xe = 6-1 SP =(1-G2)0. [R= 0/2)[P=o/8, Toa] 600. /2, P; at bottom of slab (2) = 0/3, Na) =P, at top of slab (2) = —a/3. (2 tasted { (iat ute care no total surface charge above: oe { total surface charge below: —- total surface charge above: o- (0/2) =0/2, (@/2) - (@/3) + (@/3)-0 = @ ~ (o/2) + (2/2) ~ (@/3) = 20/3, (0/3) - ¢ = -20/3, a of? Problem 4.19 With no dielectric, Cy = Aco/d (Eq. 2-54). In configuration (a), with + on upper plate, E =o/eo (in air) and E = o/¢ (in dielectric). SoV = £4 + £4 “(nte)=|@ In configuration (b), with potential difference V: E = V/d, 80 o = egE = eoV/d (in air). -o on lower, D =o between the plates. 85 (1+)79 P= eXcE = coxeV/d (in dielectric), 80 oy = ~eoxeV/d (at top surface of dielectric). Oo = OV /d = ay + ay = a — EOXeV/d, 80 oy = egV (1 + Xe)/d = eo€-V/d (on top plate above dielectric) =@_1(,4, AVA VV.) _ Ato (Lter) [Go _ lee aa B= 5 (of 40/9) = duet?) macau sesoee) al| Cotmuata 2 Sb nw Mite te = seelPtee = Mee ntee = Oe [Which is greater? Gb — Op = bite — te = Gatelimtee = Mateesad—tee — Peal > 0, So Ch > Ca] Ifthe x axis points down: a» (top surface) [| o7 (top plate) war 3 wane (@) dielectric TIES = (b) air } 0 ‘F (left) ( dielectne —E=DR | SF Gieht) Fabien 230 JDeda = Qj, => DAnr? = phnr? = D = Lor = B = (pr/30)f, for r < Rj Darr? = p$xR? > D = PR [3r? => B= (pRP /3eor?) 8, for r > R. 7 ean eed rar = OR, oe a ( z) ve-[e Oe ye 7 [re= Jey * 322 ~| eq 14 3e,) Problem 4.21 Let @ be the charge on a length £ of the inner conductor. fod = 2 ~ Bregat io (e a), subject to the boundary conditions : () Vin = Vout at s=a, 6 (i) em = cohen at s= a, Bf (iii) Vout + —Eoscos@ for s > a. From Prob. 3.23 (invoking boundary condition Vals.) = D> s*(ax cos kd +b, sinkd), Vour(86) = B,8cos-+ 9s" coskp + de sin kd) f