Chapter 4 THE VECTOR DIFFERENTIAL OPERATOR DEL, written V, is defined by 2 2i+3je 2ueir®ejsQoad Ve git gt ak tig tla tka, ‘This vector operator possesses properties anslogous to those of ordinary Vectors. It is useful in de- fining three quantities which arise in practical applications and are known as the gradient, the diver gence and the curl. The operator V is also known as nabla. ‘THE GRADIENT. Let (x,y, 2) be defined and differentiable at each point (x,y. 2) in a certain re- sion of space (.e. o defines a differentiable scalar field). ‘Then the gradient of written Vp or grad G, 1s defined by We = (2h +g +2ne Bb 4 28) 4 24, Bee ye eee Note that Ud defines a vector field. ‘he component of Vp inthe atrection of (GiE Vector as eiven sofas) fs called the die rectional derivative of in the direction a. Physically, this is the rate of change of @ at (x,y,z) in ealion a ‘THE DIVERGENCE. Let Viz.y,2) = i + 1,4 + 1k be defined and differentiable at each point (eoy,) in a certain region of space (L.e. V defines a differentiable vector field). Then the divergence of V, written V+V ot div V, is defined by vee (a4 Bi + Qeymie gir Vv = (Sa alt ght + bit bio ay, ae HMM Note the analogy with A-B = 4,2, + le + AgBy Also note that V-V 4 v-V. ‘THE CURL. If Vx,y, 2) is a differentiable vector field then the eur! ot rotation of V, written Vx V, curl V of rot V, Is defined by + Jat Roxas + ys + hw 7 ik ~/2 2 2 Be By Be ee Se58 GRADIENT, DIVERGENCE and CURL ae 22 22 = fey Bele — [ee Bly s [Be Oly mow ts a Wh _ h ay OV, ayy = BB + B-Boy + Be Be Note that in the expansion of the determinant the operators = , % +B must precede KM - FORMULAS INVOLVING V. If A and B are differentiable vectot functions, and @ and y are differen- tiable scalar functions of position (x,y, z), then LL Vb+y) = Vb + Vp or grade) = grad > + srad ye 2. Ve(A+B) = VeA+V-B ot div(A4B) = div A + divB UCAS 1 2 = Vxa+UxB or curl(A+B) = curl A + curlB 4. Vebay = (Wa)-A + d(V-ay 5. Vx(PAa) = (Vd)xA + (VX A) 6. VXAxB) = BeVx A) — A(VxB) 7. Vx(AxB) = (B-V)A — B(V-A) — (A-V)B + A(V-B) 8. WA-B) = (BV)A + (A-V)B + Bx(VxA) + AK(Vx 9p) = Vig « FH, He, HH 9. Vo) = Vip = SB + SS 4 SE woe = Ee B+ cats epacian opratr 10. Vx Woy ‘The curl of the gradient of ¢ is zero. H, Ve(Vx ay ‘The divergence of the curl of A is zero. 12. VxVxay = VV-a) — VA In Formulas 9-12, it is supposed that @ and A have continuous second partial derivatives, INVARIANCE, Consider two rectangular coordinate systems of frames of reference xyz and x'y's'(see figure below) having the same origin O but with axes rotated with respect to each other. ‘A point P in space has coordinates (x,y, 2)or (aly 2' relative to these coordinate systems. The equations of transformation between coordinates ‘or the coordinate transformations are given by ate lye + ley + ligt w y! = lage + lay + loot a= lt + lay + lent where Hp, j= 1,2,3, represent direction cosines of the x,y" and z/axes with respect tothe x,y, andGRADIENT, DIVERGENCE and CURL 59 -z axes (see Problem 38). In case the origins of the two coordinate systems are not coincident the equations of transformation become x! = lax + lay + bet + af @ y= lax + lay + ln + ab 2 = bax + ley + beat + ah where origin O of the xyz coordinate system is located at (aj, a3, a2) relative to the x'y'z' coordinate system. ‘The transformation equations (1) define a pare rotation, while equations (2) define a rotation plus translation. Any rigid body motion has the effect of a translation followed by a rotation. The trans- formation (1) 1s also called an orthogonal transformation. A general linear transformation is called an affine transformation. Physically a scalar point function or scalar field ¢(x,y,) evaluated at a particular point should be independent of the coordinates of the point. ‘Thus the temperature at a point is not dependent on whether coordinates (x,y,z) or (x!y/z') are used, ‘Then if (x,y,z) is the temperature at point P with coordinates (z,y,2) while ¢'(x!y'2') is the temperature at the Same point P with coordinates (x! y'2, we must have d(xy.z) =O Gty!2). I dleayre) = Sxtyie), where xy,2 and xy!" are related by the transformation equations (1) of (2), we call (x.y,2) an invariant with respect to the transfor mation, For example, «*+y"+2" is invariant under the transformation of rotation (J), since x*+y"+2" aay? eal, Similarly, a vector point function ot vector field A(zx,y,2) is called an invariant if A(x,y,2) = Acely\ 2). ‘This will be true if Agey2)i + Avaya + Agere = Acyl + Ayetyi2)i' + AGy.z)W! In Chap. 7 and 8, m general transformations are considered and the above concepts are extended, It can be shown (see Problem 41) that the gradient of an invariant scalar field is an invariant vector field with respect to the transformations (1) ot (2). Similarly, the divergence and curl of an in- Variant vector field are invariant under this transformation. SOLVED PROBLEMS ‘THE GRADIENT 1. If B(eays2) = Bay — y"2", find Vp (or grad ¢) at the point (1,—2,—1), 2 W Bie Bi Pneyr9 «Beep + Reta « eBory 2 ayy — ark = eee + Car-sceearh - 2-2i nk = oy + at60 GRADIENT, DIVERGENCE and CURL 2. Prove (a) WF+G) = VF+VG, (b) WFG) = FVG+G VF whete F and G ate differentiable sca- lar functions of x,y and 2. oo 2 Vere = Bir Ssr Swern 120 120 + eo Br , Be, 2 Gy OF y ye VBE Be Be BE Or ar | gr foc oc or 1BE RF + Qh «180 + 580 + 08 = G2 452 eaByr eG Bri BraBe = + OZ iF adore cages¥Z engoe = veeve = 21+ bye 2 mea = Bi Fs+ FZ wee 2 +3 +3 + Buon + Zoo + oon = ro + (FR vey + (2 GH ue est ae ea os ook 2614205420 u) «ore BEy Bay = ees = ri Bay Shay ois Sys Say = Ve + ove 3. Find Vp if (a) d= In |r|, 6) o= Fb (eo) eaateys tak, Then fr] =F yE and Gin [el = fingtey? 42, Vp = inate?) (eZ nerayeeey + sZinetayteet + 4B ina?ay? +2} os y 2 Mite | + Mga + ata + tating) - -t Wagne + aie wy) * Gaye * 2 Ve = Wy = = Vice ey? 2} = Botany + sBarrtrar + a Bereta etary Mad + barat a) 9 a datytnnMa) msinyinek _ rere 4. show that Vr" = nr" rn, Vets Vesta teeth = Veatey? ray? ash? deyeety/? 2 (aeeyeesy” = Bley) + Ble aye} 6 wh (atayteay*)GRADIENT, DIVERGENCE and CURL 61 Heya ae) 6 1 Bareyt eet yd + a tbaretestyh tas) is nGtayte sty" ak + yh + 2) = meee = Pe Note that sf r= re, where f, is a unit vector in the direction, then Vr"™= ar" ey 2) = where ¢ 1s a constant. B show tat Vo 1s a vector perpendicular to the surface. Let r+ x1tyj+2k be the position vector to any point P( yj #d2X Lies in the tangent plane to the surface at P. 2) on the surface, Then dr= dxi+ a tb = Bae Bay ee eo oe Be Bs Banca eas canny Le. Vpsdr= 0 sothat Vo is perpendicular to dr and therefore to the surface. 6. Pind a unit normal to the surface xy + 2ez= 4 at the point (2,~2,3). Voety +262) = (2ay + 2p + APJ + ek = D1 A) + aR at the point 2 mai + aj + ak 1 Viarrarray 8 4B having aston opposite to that move. ‘Then a unit normal to the surface = 6 2p Be aca voter unit normal is $i ~ 2 7. Find an equation for the tangent plane to the surface 2ez?— Sty —4x at the point (1, Veer ey 4x) = (2273-4) = Sed + Gerke ‘Then a normal to the surface at the point (1,-1.2) is Th~ 3) + 8k. ‘The equation of a plane passing through a point whose position vector is ry and which is perpendicular to the normal N is (r—r)-N= 0. (See Chap.2, Prob.18.) ‘Then the required equation is [et tys ee) - Us +am) + -ay ra = 0 o Tat) 341) + 8-2) = 0. 8. Let (x,y,z) and (atx, y+by, 2+Az) be the temperatures at two neighboring points P(x. and QcetAx, y+Ay, 2+A2) of a certain region. (@) Interpret physically the quantity A = State. yely, s¥B2) — Ote4.t) where As Is the distance between points P and Q.* waluate tim 42 - 42 and interpret () weatuate tim, F2 $2 and interpret phystealy. a ds” (c) show that 4 = 9 (@) Since A¢ 1s the change in temperature between points P and Q and As is the distance between these 5 unit distance in the direction from pans, 52 reeset th avert at ohne in tenet Pog.62 GRADIENT, DIVERGENCE and CURL () Prom the calculus, dG + Loe + Loy + Sas + sntntestnas ot oer Maher than Ady ant Qe A ds bs Tie ani aa PERS ee | ee eee o ds” Oe ds * By ds * Or ds a Bhar Bbay , Abus ab Oa ee ab, dz teat ie ae wee ote that since $f is a unit vector, Vob-Sf is the component of Vo in the direction of this unit vector. 9. Show that the greatest rate of change of , i.e. the maximum directional derivative, takes place {in the direction of, and has the magnitude of, the vector Ve. By Problem 8(c), 32 = Vob-&f is the projection of Vo in the direction 4! . This projection wil be 4 maximum when Vo and % have the same direction, Then the maximum value of 22 takes place inthe Girection of Vo and its magnitude is | Vp |. 10, Find the directional derivative of = x¥yz + 4x2? at (1,-2.1) in the direction 2 —j— 2k. Vb = Votye tte) = (aye 42h + Ped + Oy BRIM = 81 j— 10k at ae. ‘The unit vector in the direction of 21 — J — 2k is — Maik gd Jarra "oH $ ‘Then the required directional derivative is Your = @1-s~[Link]— Since this is positive, d is increasing in this direction. 11, (o) Inwhat tection from the potnt (2,1,~1) is the directional derivative of ¢ =2ys* a maximum? (6) What is the magnitude of this maximum? Vb = Vel) = ay 4 285 + aye = -H-Gk at ann, ‘Then by Problem 9,2. 14, GRADIENT, DIVERGENCE and CURL 63 (@) the directional derivative 1 a maximum in the direction Vd = a1 —4) + 12k, @) the magnitude of this maximum is |Vd| = VP + CaP GaP = VII = 4vi1. Find the angle between the surfaces 7+y?+2?=9 and 2 =22+y?—3 at the polnt (2,~1,2). ‘The angle between the surfaces at the point is the angle between the normals to the surfaces at the point Anormal to x? +y?#:?=9 at (2—1,2) 18 Vo, = Vietytst) = ot + yd tok = yt ak Anormal to r= x24y?-3 of x2497. at (2,-1.2) is Vo, = Vottyta2) = ait aso k = ok (Vhx)(Veba) = | Veal [Vebe| cos 0, where @ is the required angle. Then (Hi = 2 + 4m)-(44— 2) = | at apt aK! | 41-214 | cos 1 +4 ~ 4 = Varroa? Var Har HAI? cos 6 P25! 16 and cos 6 = 18 evan © cos 0.5819 8¥21 = 0.581; thus the acute angle ts. 6 |. Let R be the distance from a fixed point A(a,b,c) to any point P(x.y,z). Show that VR is a unit vector in the direction AP = R. Itty and Fp ate the position vectors ai +bj+ck and xi +yJ +k of A and P respectively, then pry Goi t GBM (eek, sothat Re ViemaF ry DF HEmeF . Then UR = WvenoP toabP er eaee) = Eq + Rb + emo « iF + bP + (ee eee Co ee Rg R {5 & unit vector in the direction Let P be any point on an ellipse whose foci are at points A and B, as shown in the figure below. Prove that lines AP and BP make equal angles with the tangent to the ellipse at P, Lat Ri= AP and Re=BP denote vectors drawn re- spectively from foci A and B to point P on the ellipe let T be a unit tangent to the ellipse at P. Since an ellipse is the locus of all points P the sum of whose distances from two fixed points 4 and B is a constant p, it 1s seen that the equation of the ellipse is RitRaep. By Problem 8, V(R+Rs) is a normal to the ellips hence [V(R,+R,)]+T=0 or (VR)+T = ~(VRy)-T. Since VR, and Viz are unit vectors in ditection Ry and Rp respectively (Problem 12), the cosine of the angle between VRz and T is equal to the cosine of the angle be- tween VR, and —T; hence the angles themselves are equal ‘The problem has a physical interpretation. Light rays (or sound waves) originating at focus A, for example, will be reflected from the ellipse to focus B.64 GRADIENT, DIVERGENCE and CURL ‘THE DIVERGENCE 1S. If A= x2c4 — 2y8274 + xy%ek, find V-A (or div A) at the point (1,~1,1), ao Va = Bie 85+ Bw ett yvs eat ... Buo + Zeman + Zero fF ober= GF tay? = aE) — GNP? + CMEAP = -3 ata, 16. Given = 20's". (a) Find VoVe (ot dW grad). @) Show that V-Vd = V°g, wnare V? = 2, + 25 + 25 donotes the Laplacian operator = 2a + 2 + wD cacy OV = LLabaty 6 Seat +e Sayieh = 6h + atts + wk tren VV = (Bat S44 Bane cocyiett + tty + acm) Beyer) s Query + Qe, Gove + Lar + Lest = ryt + at + ante? - + Say i ye B 6) VV bes Sas Sey + Be, + Bee, . Ze, Ze, Hy We * ae)" Be Bet Be Ee spe: Ve 11. Prove that V*(1) = 0. Ady. 2a, VO = ary Beret a naptaytesy . [a (aPty?4 29-7] = mtetaytesty lt tayte (Pa Pe aye? similarly,GRADIENT, DIVERGENCE and CURL 65 Fa, . ata y Fa _ OAR? rate OF Tare FF ey 1 af * oF ee? ane ‘The equation Vd =0 is called Laplace's equation. auaton ‘Then by addition, ¢ ys 0. It follows that = 1/r is @ solution of this 18. Prove: (a) V-(A+B) = Via + V-B ) VeGGA) = (Voy + GVA). (@) Let A= AL + Aod + dgk, B= Bul + Bol + Byk. ai +h + 20 + LAP Bp + (AH BDE + (ASB) ‘then V-(a+B) = = 2utnay + Batpay + Beas MMe, Ma, 2, Bee, Oh Sedo ae or = Se ys Bacar $b Sat Bit aa A + 21+ 2j+Buairege ie Ss + Ewe +a +a 2VA+ Ven (6) Veebay = Ve(hital + Gah + dak) afl +2 +2 Bean + Zeoy + 2eony 2a. gdh ete aoe + Ba, By ee = Ey cated e any + 21+ 2542 Ait St Bs aaa sdas say + oot + 2a + Sancta dy = (Vera + b(Veay 19. Prove Ve(% Let @=#-® and Ar in the result of Problem 18(b). Then Vern = (Veer + Ver = =9r°8rr + 37% + 0, using Problem 466 GRADIENT, DIVERGENCE and CURL 20. Prove VU VV —V UU) = UV — VU, From Problem 18(0), with @=U and A= VV, VU WV) © (Wy-CVH) + UT) = (WW) + VV Interehanging U and V yields V-c¥ Yu) = (Wv).(Wo) + V VU. ‘Then subtracting, Ve VW) ~ Vw Vu) = Vw -Vu = WAT) + OV ~ [Tuy + VU] = uv -V0U 21. A fluid moves so that its velocity at any point is v(x,y,2). Show that the loss of fluid per unit Volume per unit time in a small parallelepiped having center at P(x,y,2) and edges parallel to the coordinate axes and having magnitude Ax, Ay, Az respectively, Is given approximately by div vv. Referring to the figure above, ‘x component of velocity v at P 4 12s, component of v at center of face Se ee ponent of vat center of face AFED = vy — 355 Ox ap 1 Be, component of vat center of face GHCB = % + $5! approx. ‘Then (1) volume of fluid crossing AFED per unit time = (2) volume of fuid crossing GHCB per unit time = Loss in volume per unit time in x direction >, Similars, 10ss in volume per unit tne iny dtection = SEF AcdyAe y Joss in volume per unit tne n = dretion = 288 Ady As ‘Then, total loss in volune per unt Volume per unit tne (3B + Be Bey anys oyeenion = divy = Vey Osby AeGRADIENT, DIVERGENCE and CURL. 67 ‘This 1s true exactly only in the limit as the paralfelepiped shrinks to P, i.e, as x,y and Az approach zero. If there is no Loss of fluid anywhere, then V-v = 0, This is called the continuity equation fot an in- compressible fluid. Since fluid is neither created nor destroyed at any point, it is sald to nave no sources for sinks. A vector such as v whose divergence is zero 1s sometimes called solenoidal. 22, Determine the constant a so thet the vector V = (x +8y)i + (y—22)4 * (#+a2)K is solenoidal, ‘A vector V is solenoidal if its divergence is zero (Problem 21. a a Vv = Zar + 2y-my + Bute erst ee 3 a ‘Then Vv=a+2=0 when 2 ‘THE CURL 23, If A = x24 — 26°ys) + 2yeth, find Voc A (or curl A) at the point (1,-1.1). a vx Bae Bye Buccs = aches + ayet A Ree Sas Zaye = ayes ety J k 12 2 a "|e % a Se oe a a a = (Zs - Zeal + (2a - Ze + (Zea - ee 5 y 2. wre - £0 = Gt rape oe?s — ayek = ae oe att, WI A= x’yi— Der) + 2yek, find curl curl A ewiewl a = Vx(Vxay 4 a oa 2 evef2 2 2] s velarsant- atezn lo 5. Lcae+zeyi ~ 42094] fy toe ‘ 1 * ~| 2 2 3 | = cre: a. (+21 ate 0 oe 25. Prove: (a) Vx(A+B) = Vxa + VxB (b) Vx(@ay = (Vb) x A + (VA).68 GRADIENT, DIVERGENCE and CURL (0) Let A= Ai +dyd + Agk, Be By +B,§+B,k. Then: xcasey = (24+ 24+ Bay x [eaeaye + agtBds + gt Vecnsmy = Bar Br + Bay x Ceyrage + rags + yr BO] 1 5 x ~[2 2 2 Be ey O AB, AgtBy gt By = (Beagray — aja + [2asay - Zaye = Qusray Zagat + Bang = Baraghs + Reseny - Zeayeahe » Pe Heyy + Be Bays Ble Sane ey 28g _ BBs 28; _ 285), , [Be _ 2 ates Wo | or oe = Vea + xB () Vxdday = Vx (bAb + O4,5 + 64,8) Bea - Zeal + Rea- Ze + em 2 soak . (o% + Bb ay - gee Bade ey Oz «(42 a oe Bala + [oe + Pte — oUt — Bade = plc2de_ Me, Shs Bho Bdg _ Qds, = ole Ban + Hy) 7 Mane ’ Be ~ Baas + Glan aon) ‘ ; = (Vay + # = ab A, Ae = pV) + (Vd) xaGRADIENT, DIVERGENCE and CURL 26, Evaluate V-(Axr) if VxA=0. Let A= Ai + Agi + Ak r= ek ty tek oak tren axe = [4 4p ds peo F (eAgmyden + (edo FADS + Arado wa Vines = Betearty + Botamety + Botinntn By, Be Bh yy 2 _ Ole ee er Or Bay gy (2h May 5 42 2M ae GE ae BEB on = Cent ys een): (Oe — Magy g (242 _ BA A a rC~S~—rsS—S—sSC = re(Vxay = recuna. if VxA=0 this reduces to zero. 21. Prove: (a) Vx(Vo) = 0 (curl grad P=0), () Ve(VxA) = 0 (div curl A =0). (Vx Wy = Vx Pr ys Hy [oi -|2 2 2 ole 9 Bb dt So 3 te te, hh. ye ke! Ge ee EE SRM Te provided we assume that ¢ has continuous second partial derivatives so that the order of differentiation is ‘immaterial. ck Waa = Ve [2 2 B (6) Ve(Vx ay a a & Ay Ag As 2dy _ Bde Bd _ PAs), y (24a _ Ody Gay on a ae Ok 3 Bde _ 2A, Bhs Bay, B/Oda _ ds ny 2 Bay y 2 ,BAe _ Bhs Co or uae) = are er70 GRADIENT, DIVERGENCE and CURL ety 7 Bede” Bee ~ Bae * Bede ~ By * assuming that A has continuous second partial derivatives. Note the similarity betweon the above results and the results (Cx€n) = (CxC)m=0, where m is a sealar and €+(CXA)=(CxC)-A=0 28. Find curl(rf(r)) where f(r) Is differentiable, eur iefon) = Vx efi = Vx foley fois sfow ‘ 4 x efo 8 cf Oe ey Or xfer) yf afer) a ot H_ ys gh-w MyM oO By so Boe he YB yee - Lz = C8, sinter, LLY aoa aL, Bee ae BT 4 i k -vx(2 2 2 Vex = Vel SS A de Bdy _ 24 Ody _ 24s Bly _ 24, = Vx [(S4e — She. Sai _ Shs. Sho _ Oda yy) uy Bele Goran Gn ata el 1 i x - 2 2 2 . ee ey a Bd _ By dy _ Ag Me 2dr en es SyGRADIENT, DIVERGENCE and CURL 7m ay _ Faery. My Ste Coe7 ae Bas yp 5 (ay Says g Shey Te ce doy oe doe Dod 0d, ete Okc i Om Cae at aM Cat” af” are Bay, Ft Boa, , oh te, Btayy , Shs Fag , Tao QR yas Sede * Gedy BP Bey * Bede * yo z + Dy cat + Aas + daly BP Ud 4 dad + Asko = Va + Vea) I desired, the labor of writing can be shortened in this as well as other derivations by writing only the f ‘components since the others can be obtained by symmetry ‘The result can also be established formally as follows. From Problem 47(a), Chapter 2, ” Ax (BxO) = BAC) = (ABC Placing A= B=V and C =F, Va Ver) = WV) — VE = UVR - VP ‘Note that the formula (1) must be written so that the operators A and B precede the operand C, etherwise the formalism fails to apply. 30. If ¥=@xr, prove w= fcurl¥ where w is a constant vector, ios ok culy = Vv = Vx(@xn = Vx a, os os soy ok Vx [Wor wey) + (ox —wy2)) + (Wry —@ax)k) i 4 k i: = 2 = Ash + wed + = 2@ 2 2 2 = 2a + ooh + ayk) = 20. (ar tyy Wax 042 yy —wex2 GRADIENT, DIVERGENCE and CURL ‘Then @ = 30x = deuly ‘This problem indicates that the curl of a vector feld has something to do with rotational properties of the fleld. ‘This 1s confirmed in Chapter 6. If the fleld F is that due to a moving fluid, for example, then a paddle wheel placed at various points In the fleld would tend to rotate in regions where curlF #0, while it curlF= 0 in the region there would be no rotation and the field F is then called irrational. A fleld which {is not irrotational is sometimes called a vortex field. Vea =0, VxE=— 28 ven BE, show that Band H satisty Vu= 2, aii VE 3. od x(Vxe) = Vx = ~ 200% 7 Vx (Vag) = Vx Sh aevxm = By Protien 29, Vx (Vxe) = -VeE+V(V-E) = -Ve. Sintuay, Vx (Vey = Vx 2) = 2VxE) put Vx (Van = -Vn + (Vem = -Vn. then Vn = 28, ‘The kiven equations ate related to Maxwell's equations of electromagnetic theory. The equatlon &e = 2 is called the wove equation. at : MISCELLANEOUS PROBLEMS. 82, (a) A vector V ts called irrotational if curl V=0 (see Problem 30). Pind constants a,b,c so that Vow (eH2y Haz) + (bx— By —2)) + (Ae Hey + 22)K Is irrotational. (6) Show that V can be expressed as the gradient of a scalar function. 1 , x @onvy = Ixy = | 2 _ 2 | + cents eo + ome tty tor bx Sy— ede ey 422 This equals zero when a=4, 52, ¢"—1 and Vo = (tay tan + Qe— ays + Gey Fe) () Assume v = Vb = 204 + By , 2, (as vave» Me ae BO. eta 6 ea. ee Then) semi @ Be watys, OE = beng tee, Integrating (1) partially with respect to, keeping y and « constent, cy bs Be ay eae + foe) where /(y,2) is an arbitrary function of y and z, similarly from (2) and (3),GRADIENT, DIVERGENCE and CURL 3 6 b= wy Pay + ey 6 bs ery tt hee). ‘Comparison of (4), (5) and (6) shows that there will be @ common value of ¢ if we choose ie eye 2 ay fos Ee, geeay- 548, ben = F-F so that 2 os 6-2 Bite msm Note that we can also add any constant tog. In general if Vx V =O, then we can find ¢ so that V= Vo. ‘A vector field V which can be derived ftom a scalar field $ so that V= Vo is called a conservative vector field and is called the scalar potential. Note that conversely if V= Ve, then Vx V = 0 (see Prob.27a). . Show that if (x,y,z) 18 any solution of Laplace's equation, then Vi is a vector which is both solenoidal and trrotational By hypothesis, ¢ satisties Laplace's equation V'¢ = 0, i.e. V-(Vd) = 0, Then Ve) is solenoidal (see Problems 21 and 22) From Problem 272, Vx (Vd) =0 so that Vo is also trrotatfonal. Give a possible definition of grad B. ‘Assume B= 8,1 + Boj + Byk. Formally, we can define grad as ve = (21+ 25+ 2m + ai + Bo a ec = Bigg 4 2Bayy , BB - Bus Bye Bw BBs yy, 28a yy 280 Ba yg 5 Bay y Poyy Chey 28s yy. Bens . OB + Bars eas Bo ‘The quantities 41, 14, ete A quantity of the form are called unit dyads. (Note that 45, for example, 1s not the same as j1.) ayit + oyhf + agtk + agit + onf5 + ogik + ogki + anki + aghk is called dyadic and the coetticients ass, a2 nents in the form fare {ts components. An array of these nine compo- oe a0 on Oe Oe ee oa tas fs called a 3 by 3 matrix. A dyadic 1s a generalization of a vector. till further generalization leads to ‘iadics which are quantities consisting of 27 terms of the form axu Lil + amu: JIL +... A study of how the components of a dyadic of tladic transform from one system of coordinates to another leads to the sub- Ject of censor analysis which ts taken up in Chapter 8.4 GRADIENT, DIVERGENCE and CURL 35. Let a vector A be defined by A= Axi + dol + ok and a dyadic @ by © = ay hi + ayehh + asgik + aacSi + aaahh + azdK + ak + ayokI + agkk Give a possible definition of A-@, Formally, assuming the distributive law to hold, AW = Ant + Aah hel @ = ANDY Aad + Aah ‘san example, consider {-@, This preduct is formed by taking the dot product of {with each term of Dani adding results. Typical examples are i+ ay lt, 1+ar2HJ, T+ aq51, A+ aogkJ, etc. If we give mean ing to these as follows Feaglt = aut = ot since et = elf = eua(t-i)$ = ayef since tet = feonsdt AeaeghS = opal and give analogous interpretation to the terms of 4-@ and K+, then anil DL = 0 sine I+} = =o since f-k = As@ = Aylaysht as05+ a59h) + Aalans t+ ana 5+ az) + Aalaay b+ 0g2 $+ oe) = (Anois + Azden + Agdar) t+ (Axaie + Aatae + Agtea) S + (Aneia + Azden + Ages) K which is @ vector. 36. (a) Interpret the symbol A-V. (b) Give @ possible meaning to (A-V)B. (c) Is it possible to write this as A-VB without ambiguity? (@) Let A = Ayi + 42j + Agk. Then, formally, + Age (244 2543 AV = tas + dow Ber ds+ uy 242+ 422+ 42 fe 2 oe san operator. Por example, Wb = et deD edd = mb + ate at UNS = ete shed = We + ee + oS ‘Note that this is the same as A-Vo. (b) Pormaly, using (0) with replaced by B= Bi + Bei + Bak, eee, oe ne adn = ho oud oddyn AB 6 AB + Oe 2B, | By, Bh on ao, oes 2. By, , Bh 2 GE het eG + AGE thet AGO + US eS + AS (c) Use the interpretation of VB as given in Problem 34. Then, according to the symbolism established {in Problem 35, AVe = (b+ Aas + Ag) VE = At-VB + Aof-VB + Ak VB 28, , 28, , 2 2B, , 2B, , 2B, 2B, , 280, , 2B Aya BBy 5 By 6 ag Rta + Bey 4 aay 6 dy Ete + 28g 5 BB ee ry eredGRADIENT, DIVERGENCE and CURL % Which gives the same result as that given in part (2). It follows that (A-V)R > A«V without ambi- ulty provided the concept of dyadics is Introduced with properties as indicated. BU. I A= ayzi yj + ath, B= 21 + ysj —ayk and p = 2eFys, find (@) (AV), (6) AVS, (0) BVA, (2) (AXV)S, (e) AX VO. AMS = lat— ys rey Ga+ s+ Sule = (aye = yD + 22 2) cata) Oey Ge Beaty) — ay Bearyaty + ak Qeatys? aot ~ Hy Zar fae (2yzy(deye®) — (yy (arte?) + Ce? (60%y2") = Sey2st — aye? + ays 0) AVE = (aye — yh + ety Bi + Bi ay (yeh my teh) «hey ht 272 + Ge'yP = ee = ays + yt ‘Comparison with (a) tlustrates the result (A-W)d = a-Vo. () BVA = [2+ yeh — aye (Ba + y+ 2 ©) WMA = [ott yi- ww (Sir Bs + Aula oo eo a en a Gon ey ote ae eo Oe Pens ty + pQrt— x) — OVI + 22K) = yt ay — Oy ey + OR? Beye For comparison of this with B-VA, see Problem 36(c). Opa s+2wle 1 ARDP = [eet = wyd Hann Ba eB + eB wd) + seed - 2,4 24 td STG) + ORS + Ay Sled En ny eB 6 oe Bo yy 6 oe ory Bn + ey —(6uty2a? + BPE + (aye — eV + Cares + aye De6 GRADIENT, DIVERGENCE and CURL 26, , 3, , 2%, CARVE = yet 2S eat SE Sh + vata Ms eke weg = ACG FE BNL F Caetye® mae # (actyet + Ay) Comparison with () ilustrates the result (axV)p = Ax Vd. INVARIANCE 38. Two rectangular xyz and x'y'2" coordinate systems having the same otigin are rotated with re- spect to each other. Derive the transformation equations between the coordinates of @ point in the two systems. Let r and 1! be the position vectors of any polnt P in the two systems (see figure on page 58). Then since r=, w Styl ea citys tak Now for any vector A we have (Problem 20, Chapter 2), Ae cath (a ® aka ‘Then letting A= 1,J,k in succession, @ Gor GAT © desbs + Uek RE & Lol! + bead! + leek! {; Bat + ess + ews atl + teas + bak! woe deta + aes s + ce Nk = gt! + int + kak! Substituting equations (2) in (1) and equating coefficients of 1.4K! we find @ * loge + tay + bea? hae + bay + hazy fore + boy * boot the required transformation equations. 39. Prove! = Lisi + led + bok = Lah + Lnad + bag Kl © Lyi + Lael + beak Pet any vector Awe have A = (Avi + (AS) + (ACKDI. ‘Then letting A = 14k" in suecesston, = dhne + ats + = Git + dead + = MDE + MDs + hat + ligS + gk fey + lop + teak fay + tao + toa ‘ y KGRADIENT, DIVERGENCE and CURL 7 40. Prove that = 1 Af m=n, and 0 if mén, where m and n can assume any of the values ove that 3 lon lin Os mA cat of the 1,2,3. From equations (2) of Problem 38, Het = 1 = (hall + bead! + lash) = Chaat! + load! + laak) heed haat! + laa + Leak!) «Cant! + bed! + hol) = taal + laalon + besten © = Gaal! + Lagi! + gl) + hat! + band! * took’) fal + tailea + laloo ‘hese establish th rquzed result where met, Ry considering :4JJ, kek and Wo the result can te proved form-2 and m3 Litmen ; br tne fgg = {180 oan tn pp By ‘The symbol fg, 18 called Kronecker's symbol. 41. if d(x,y,2) 1s a scalar invariant with respect to a rotation of axes, prove that grad¢ is a vector Anvariant under this transformation, By hypothesis (2.7.2) = b'(e'sy',2). To establish the desired result we must prove that , , , , , Cs 4 1 koe Sls Se tt * yt * OF Using the chain rule and the transformation equations (8) of Problem 38, we have % we, we ww, Oe Oy Oe” Or! Ox o oy! ae Gee oe % 2% % lee | ee! a wa, wee By | alon 337 i Multiplying these equations by 4,4,k respectively, adding and using Problem 39, the required result fol- lows.7B 2 3. 82. Bt. 61. IF ). Find @(r) such that Up = . IEF is a differentiable function of 7.2 GRADIENT, DIVERGENCE and CURL SUPPLEMENTARY PROBLEMS It p= at —2¥y, tind Vo and | Vo at the point (2,-2,-1). Ans. 101—4J— 16k, 2/93 WA = 28h — yf + x27 and G2 22 ay, find AV and AXVE at the point (1.1.1) Ans. 5, = j= AIk OF aoa G = 22%y — ay", tind (2) VoPG) and (6) V(FG) at the point (1,0,~2) dns, (0) “41 +95 ¢k, (8) 8) Pind Viel? dns, are Prove Vy = Lee, Ans. (6— 278 — 9) svatunte Vor ave + & ur Vu = ate, find U. Ans. 1/3 + constant Sand $u)=0. dns. ges ha ~ Eee Pind Vy) where Y= Gy? #27) € Ans. Q— WV = By 2G + BP h, fd Plegsd MH OULADA =A Ans. at Vi = Gy? 2ays2yt + (8 + aay 7a )4 + (G28 — 8%)", find Ans. ip = xy? —x¥y2° + 3y + (9/2) + constant IfU is a differentiable function of xy.2 , prove WUsde = dU where x.7.2 are differentiable functions of ¢, prove that aE gpd oe J. IEA ts a constant vector, prove Vir-A) = A. AE Atwaya) = As Aas + Aah, show that dA = (WAzedeyt + (WAgedng + (WAgedtdk. Prove WE) = GWE=FVE ip Gyo, ¢ é 3. Find a unit vector which is perpendicular to the surface of the paraboloid of revolution z » x+y" at the arya cin (1,2.6). Ane, t= sie tena vai Find the unit outward drawn normal to the surface (x — 1)? ry? +(= #2)? = 9 at the point (3,1.—4). Ans. (2t + $= 2ky/8 Find an equation forthe tangent plane to the surface x3? +x"y = 2 —1 at the point (1,-3.2). Ans, —y—32 +120 Pind equations for the tangent plane and notmal line to the surface z= 2+y” at the point (2,~1,5). eet Ans, = ty 225, 3 ) ind the directional derivative of = 4x2" Ans. 316/7 or x e4tt2, y=], 2 e045 eye at (21,2) in the direction 21— 2) + 6k Find the directional derivative of P = 4e%"2*# at the point (1.1 (-3,5,6). dns. —20/9 1) in 8 direction toward the polat0. a. 2. 2. ™ 1%. 6. n 8. 2. 80. a1 GRADIENT, DIVERGENCE and CURL 9 In what direction from the point (1,9,2) 1s the directional derivative of = 2xz—y? a maximum? What is the magnitude of this maximum? Ans. In the direction of the vector 4i— 6} + 2k, 27/14 Find the values of the constants 0,b,¢ so that the directional derivative of gb = axy+ bys +02%° at (1,2,=1) has @ maximum of magnitude 64 in e direction parallel to the : axis. Ans, a*6, be24, c= —8 FPind the acute angle between the surfaces xy"2=ax+2° and ae°=y+22 =1 at the potnt (12,1), 3. arc cos “= 19%ss! Ans, aro cos FS cE = 10°58 ind the constants « and b so that the surface ax” — by? ey +29 4 atthe point (1,-1.2). Ans. a= 5/2, B=1 (a+2)x will be orthogonal to the surface (@) Let w and v be differentiable functions of x,y and z. Show that a necessary and sufficient condition that u and v are functionally related by the equation F(u,c) = 0 is that Vax Vo = 0, FAX are functionally related. (b) Determine whether = ate tanx + are tany and © Ans. (b) Yes (v = tan u) ). (a) Show that a necessary and sufficient condition that u(x.y.2), v(ey.2) and w(x.y,2) be functionally re- lated through the equation F(u,vw) = 0 is Vu-Vox Vw = 0. (Express Vu-VoxVu in determinant form. ‘This determinant is called the Jacobian of u,v,w with re 3u,v,w) 4, ae sega a (e) Determine whether w axty +2, 2 ety?e2? and w=xy4ye-tee ate fonctiomtly related. mm Be Ox dy Oz mm (B® Bo rearemee be Be Be i MA = Saye? 6 Dy f= Pye k and d= a7 —y2, find VEAL GATS, (VGA, VM, at the point (1.1). Ans. (@) 4, (®) = 18, (6) 1, 6 Byaluate div (20°24 — ay’ej + 3y27k). Ans. axe — Day? +672 p= Bhp —ys® +aey 42x — By —5, tind Vp. Ans, G2 + Bbry — 22° ~ oye Evaluate V'(ine). Ans. 1/7? Prove Vi nr (attyr®-? where n is a constant. (Bay — 2)t + (xe? +y*)§ — 2°27, find V(V-F) at the point (2,-1,0). Ans. ~6i + 24) — 32k feo i a constant vector and v= @xr, prove that div v~ Prove Vebyy = bV'y + 2Ve-Vy + uV'o. i U=ax’y, -? —2y ovatunte grad [grad U)(erad V9]. Ans. (6y2°—125)1 + 6274 + ye Evaluate V-(Pr). dns. 67? Bratuate Ve [+Vave?y]. Ans. arm Braluste V°LV-/r?)]. Ans. 2r°* 1 Ast/r, find grad div A, Ans. —2°* 6 2. (Find fo suon tmat V'7(r)= 0. Ans. f@)= A+ Bh where A and B ate arbitrary constants. cyrrove ye = Lf8, 100. 101. 102. |. Graph the vector fields A=x GRADIENT, DIVERGENCE and CURL Prove that the vector A = ay*271 + 4x%?j — 30°)? Is solenotdal. Show that A= (2x7 + Bxy"2yt + (Bey — Sey) — (4y72? + 2°2)K Is not solenoidal but B= xy27A Is solenoidal. Find the most general differentiable function f(r) s0 that (Hr is solenoidal Ans. f(r) = C/P where C is an arbitrary constant. 23 isa "sink fold". Plot and give a physical interpretation. Py IfU and ¥ ate ditterentiable scalar fields, prove that Vx VV is solenoidal. ‘Show that the vector field V = WA = des? f= yz j + oe29K_and_ = wpe, tind Vx A, (cul A), (€)Vx(Vua), (2) V[A-curt A], (e) curl grad (PAY at the point (1,1,1), Ans, (a)i +4, (b)S1~ 3} 4k, (e) 51+ 9k, (A) —2-+4 48K, (6) 0 It F = xyz, G = ay 322, find (a) V[VF) such that a= Veh Ans. p = 3x°y + x27 — yz + constant10s. 108. 105. 106. 107. 108. 109. 10, an a. us. 14. us. GRADIENT, DIVERGENCE and CURL 81 Show that E = r/*® is itrotational. Find G such that B= —V and such that (e)= 0 where a>0. Ans. = In(a/r) WA and Bare notational, prove that AX 1s solenolda. Xf f() Is altferentable, prove that (ee i imoationa Istheres siteentable vector function V such that (@) cul V=r, (b) eu V-= 21+ #347 If 0, find W. Ans. (a) No, (b) V = 3xj + (2y—x)k + Vp, where dis an arbitrary twice differentiable function. Sion testo tae netins xo 2 2B oye - 22 oy. Vxn=e Qe. Vee eRe a where p 1s a function of s..2 and ¢ ts the velocity of eh, assumed constant, are iven by B+ -V6- 22 Be vKa where A and d, called the vector and scalar potentials respectively, satisfy the ‘equations, Vat Seo, MVo-1 Sheep, Va 23, TA eu & af Or (ey olven the dyadic = 11644 ¢kR, evaluate F-(@-1) and (r-8)-t. (6) there any ambiguity in ling t-asr? (c) What does -a-r= 1 represent geometially? dns. (2)t-(aet)=oayer=atey2ea", (No, (e) Sphere of radius one with center atthe oxen. Vee = amp ei — 9°) + ys? and B= 274 ~ xyJ + 9° k, pive a possible significance to (Ax V)B at the point .—1.1) (2) st possbie to write the result as Ax (VR) by use of dyadies ? Ans, (@)~444~ 15 + 2K ~ Sh = 440 + OKK (6) Yes, ifthe operations are suiteby performed Prove that (e.y.2) = 2% +y+27 is @ scalar invariant under a rotation of axes If AGz.y.2) ts an invartant differentiable vector eld with respect to @ rotation of axes, prove that (a) div A and (6) cur A are Invariant scalar and vector flelds respectively under the transformation, Solve equations (3) of Solved Problem 38 for x.y.2 in terms of ='y/,2" Ans. x = basa! bay’ + one's y= hee! + bey! + beg2', 2 = he’ + bay’ + en! If A and B are invariant under rotation show that A+B and AX B ate also invariant Show that the Laplacian operator {s Invariant under a rotation.