Classical Electrodynamics - Jackson (Solucionario 1)

Solutions to Problems in Jackson, Classical Electrodynamics, Third Edition
Homer Reid December 8, 1999
Chapter 2
Problem 2.1
A point charge q is brought to a position a distance d away from an innite plane conductor held at zero potential. Using the method of images, nd: (a) the surface-charge density induced on the plane, and plot it; (b) the force between the plane and the charge by using Coulombs law for the force between the charge and its image; (c) the total force acting on the plane by integrating 2 /2 the whole plane;
0
over
(d) the work necessary to remove the charge q from its position to innity; (e) the potential energy between the charge q and its image (compare the answer to part d and discuss). (f ) Find the answer to part d in electron volts for an electron originally one angstrom from the surface.
(a) Well take d to be in the z direction, so the charge q is at (x, y, z ) = (0, 0, d). The image charge is q at (0, 0, d). The potential at a point r is (r) = q 4
0
The surface charge induced on the plane is found by dierentiating this: 1
1 1 |r dk| |r + dk|
Homer Reids Solutions to Jackson Problems: Chapter 2
d dz z=0 (z + d) q (z d) + = 4 |r + dk|3 |r + dk|3 qd = 2 (x2 + y 2 + d2 )3/2

0
z =0
(1)
We can check this by integrating this over the entire xy plane and verifying that the total charge is just the value q of the image charge:

(x, y )dxdy
qd 2
0 0
= qd
rddr 2 + d2 )3/2 ( r 0 rdr (r2 + d2 )3/2
qd 3/2 = u du 2 d2 qd 2u1/2 2 = 2 d = q (b) The point of this problem is that, for points above the z axis, it doesnt matter whether there is a charge q at (0, 0, d) or an innite grounded sheet at z = 0. Physics above the z axis is exactly the same whether we have the charge or the sheet. In particular, the force on the original charge is the same whether we have the charge or the sheet. That means that, if we assume the sheet is present instead of the charge, it will feel a reaction force equal to what the image charge would feel if it were present instead of the sheet. The force on the image charge would be just F = q 2 /16 0d2 , so this must be what the sheet feels. (c) Total force on sheet
2 1 2 dA 2 0 0 0 rdr q 2 d2 2 4 0 0 (r + d2 )3 q 2 d2 3 u du 8 0 d2 q 2 d2 1 u2 8 0 2 d2
= = = = =
q 2 d2 1 4 d 8 0 2
Homer Reids Solutions to Jackson Problems: Chapter 2 q2 16 0 d2
in accordance with the discussion and result of part b. (d) Work required to remove charge to innity
q2 dz 4 0 d (z + d)2 q2 u2 du 4 0 2d q2 1 4 0 2d q2 8 0 d
= = = =
(e) Potential energy between charge and its image = equal to the result in part d. (f ) q2 8 0 d = (1.6 1019 coulombs )2 8 (8.85 1012 coulombs V1 m1 )(1010 m ) q2 8 0 d
= 7.2 (1.6 1019 coulombs 1 V ) = 7.2 eV .
Problem 2.2
Using the method of images, discuss the problem of a point charge q inside a hollow, grounded, conducting sphere of inner radius a. Find (a) the potential inside the sphere; (b) the induced surface-charge density; (c) the magnitude and direction of the force acting on q . (d) Is there any change in the solution if the sphere is kept at a xed potential V ? If the sphere has a total charge Q on its inner and outer surfaces?
Problem 2.3
A straight-line charge with constant linear charge density is located perpendicular to the x y plane in the rst quadrant at (x0 , y0 ). The intersecting planes x = 0, y 0 and y = 0, x 0 are conducting boundary surfaces held at zero potential. Consider the potential, elds, and surface charges in the rst quadrant. (a) The well-known potential for an isolated line charge at (x0 , y0 ) is (x, y ) = (/4 0 ) ln(R2 /r2 ), where r2 = (x x0 )2 + (y y0 )2 and R is a constant. Determine the expression for the potential of the line charge in the presence of the intersecting planes. Verify explicitly that the potential and the tangential electric eld vanish on the boundary surface. (b) Determine the surface charge density on the plane y = 0, x 0. Plot / versus x for (x0 = 2, y0 = 1), (x0 = 1, y0 = 1), and (x0 = 1, y0 = 2). (c) Show that the total charge (per unit length in z ) on the plane y = 0, x 0 is 2 Qx = tan1 What is the total charge on the plane x = 0? (d) Show that far from the origin [ 0 , where = 2 2 x0 + y0 ] the leading term in the potential is asym = Interpret. 4 (x0 )(y0 )(xy ) . 0 4 x2 + y 2 and 0 = x0 y0
(a) The potential can be made to vanish on the specied boundary surfaces by pretending that we have three image line charges. Two image charges have charge density and exist at the locations obtained by reecting the original image charge across the x and y axes, respectively. The third image charge has charge density + and exists at the location obtained by reecting the original charge through the origin. The resulting potential in the rst quadrant is (x, y ) = = where
2 r1 = [(x x0 )2 + (y y0 )2 ] 2 r2 = [(x + x0 )2 + (y y0 )2 ]
R2 R2 R2 R2 ln 2 ln 2 ln 2 + ln 2 4 0 r1 r2 r3 r4 r2 r3 ln 2 0 r1 r4
(2)

2 r3 = [(x x0 )2 + (y + y0 )2 ] 2 r4 = [(x + x0 )2 + (y + y0 )2 ].
From this you can see that
when x = 0, r1 = r2 and r3 = r4 when y = 0, r1 = r3 and r2 = r4 and in both cases the argument of the logarithm in (2) is unity. (b) = d dy 1 dr2 1 dr3 1 dr1 1 dr4 + = 2 r2 dy r3 dy r1 dy r4 dy
0
y =0
We have dr1 /dy = (y y0 )/r1 and similarly for the other derivatives, so = y y0 y + y0 y y0 y + y0 + 2 2 2 2 2 r2 r3 r1 r4 1 1 y0 = 2 (x + x )2 + y 2 ) (x x0 )2 + y0 0 0
y =0
(c) Total charge per unit length in z
Qx
=
0
dx y0
0
dx 2 (x x0 )2 + y0
dx 2 (x + x0 )2 + y0
For the rst integral the appropriate substitution is (x x0 ) = y0 tan u, dx = y0 sec2 udu. A similar substitution works in the second integral. = =
/2 tan1 y 0
0 x
/2
du
du
tan1
x0 y0
x0 x0 tan1 + tan1 2 y0 2 y0 2 x 0 = tan1 . y0
(3)
The calculations are obviously symmetric with respect to x0 and y0 . The total charge on the plane x = 0 is (3) with x0 and y0 interchanged: Qy = 2 y0 tan1 x0
Since tan1 x tan1 (1/x) = /2 the total charge induced is Q =
which is, of course, also the sum of the charge per unit length of the three image charges. (d) We have = Far from the origin,
2 r1 2 r 2 r3 ln 2 2 r2 4 0 r1 4
= [(x x0 )2 + (y y0 )2 ] x0 y0 = x2 (1 )2 + y 2 (1 )2 x y y0 x 0 x2 (1 2 ) + y 2 (1 2 x y = x2 2x0 x + y 2 2y0 y ) xx0 + yy0 x2 + y 2
= (x2 + y 2 ) 1 2 Similarly,
2 r2 2 r3 2 r4
= (x2 + y 2 ) 1 2
xx0 + yy0 x2 + y 2 xx0 yy0 = (x2 + y 2 ) 1 2 2 x + y2 xx0 yy0 = (x2 + y 2 ) 1 2 x2 + y 2
Next,
2 2 r1 r4 2 2 r2 r3
= (x2 + y 2 )2 1 4 = (x2 + y 2 )2
(xx0 + yy0 )2 (x2 + y 2 )2 (xx0 yy0 )2 14 (x2 + y 2 )2
so
The (x2 + y 2 ) term in the denominator grows much more quickly than the (xx0 + yy0 ) term, so in the asymptotic limit we can use ln(1 + ) to nd = = (xx0 yy0 )2 (xx0 + yy0 )2 4 + 4 4 0 (x2 + y 2 )2 (x2 + y 2 )2 2 2 2 2 2 2 4(x2 x2 0 + y y0 2xyx0 y0 ) + 4(x x0 + y y0 + 2xyx0 y0 ) 2 2 2 4 0 (x + y )
2 0 yy0 ) 1 4 (xx 2 2 2 (x +y ) . = ln 2 0 +yy0 ) 4 0 1 4 (xx (x2 +y 2 )2
Homer Reids Solutions to Jackson Problems: Chapter 2 16xyx0 y0 4 0 (x2 + y 2 )2 4 (xy )(x0 y0 ) . 0 (x2 + y 2 )2
= =
Problem 2.4
A point charge is placed a distance d > R from the center of an equally charged, isolated, conducting sphere of radius R. (a) Inside of what distance from the surface of the sphere is the point charge attracted rather than repelled by the charged sphere? (b) What is the limiting value of the force of attraction when the point charge is located a distance a(= d R) from the surface of the sphere, if a R? (c) What are the results for parts a and b if the charge on the sphere is twice (half) as large as the point charge, but still the same sign?
Lets call the point charge q . The charged, isolated sphere may be replaced by two image charges. One image charge, of charge q1 = (R/d)q at radius r1 = R2 /d, is needed to make the potential equal at all points on the sphere. The second image charge, of charge q2 = q q1 at the center of the sphere, is necessary to recreate the eect of the additional charge on the sphere (the additional charge is the extra charge on the sphere left over after you subtract the surface charge density induced by the point charge q ). The force on the point charge is the sum of the forces from the two image charges: 1 4 0 qq1 d
R2 2 d
= =
qq2 d2
(4) (5)
q2 dR d2 + dR + 2 2 2 4 0 [d R ] d4
As d R the denominator of the rst term vanishes, so that term wins, and the overall force is attractive. As d , the denominator of both terms looks like d4 , so the dR terms in the numerator cancel and the overall force is repulsive. (a) The crossover distance is found by equating the two bracketed terms in (5):
[d2
dR R 2 ]2
d2 + dR d4
d4 R = (d + R)[d2 R2 ]2 0 = d5 2d3 R2 2d2 R3 + dR4 + R5
I used GnuPlot to solve this one graphically. The root is d/R=1.6178. (b) The idea here is to set d = R + a = R(1 + a/R) and nd the limit of (4) as a 0. F = q2 4 0 R2 (1 +
a 2 R) a R)
R2 (1 +
q 2 R2 aR (2R + 3a)(R 4a) + 4 0 4a2 R2 R4
R2
a 2 R2 (1 + R ) + (1 + a 4 4 ) R (1 + R
a R)
The second term in brackets approaches the constant 2/R 2 as a 0. The rst term becomes 1/4a2. So we have F q2 . 16 0 a2
Note that only the rst image charge (the one required to make the sphere an equipotential) contributes to the force as d a. The second image charge, the one which represents the dierence between the actual charge on the sphere and the charge induced by the rst image, makes no contribution in this limit. That means that the limiting value of the force will be as above regardless of the charge on the sphere. (c) If the charge on the sphere is twice the point charge, then q2 = 2q q1 = q (2 + R/d). Then (5) becomes dR 2d2 + dR q2 2 + 4 0 [d R2 ]2 d4 and the relevant equation becomes F = 0 = 2d5 4d3 R2 2d2 R3 + 2dR4 + R5 . Again I solved graphically to nd d/R = 1.43. If the charge on the sphere is half the point charge, then F = and the equation is dR d2 + 2dR q2 2 + 2 2 4 0 [d R ] 2d4
0 = d5 2d3 R2 4d2 R3 + dR4 + 2R5 . The root of this one is d/R=1.88.
Problem 2.5
(a) Show that the work done to remove the charge q from a distance r > a to innity against the force, Eq. (2.6), of a grounded conducting sphere is W = q2 a . 8 0 (r2 a2 )
Relate this result to the electrostatic potential, Eq. (2.3), and the energy discussion of Section 1.11. (b) Repeat the calculation of the work done to remove the charge q against the force, Eq. (2.9), of an isolated charged conducting sphere. Show that the work done is W = q2 a q 2 a qQ 1 . 4 0 2(r2 a2 ) 2r2 r
Relate the work to the electrostatic potential, Eq. (2.8), and the energy discussion of Section 1.11.
(a) The force is |F | = q2 a 1 3 4 0 y (1 a2 /y 2 )2
directed radially inward. The work is W = = = = = =
F dy
r
(6)
q2 a 4 0 q2 a 4 0 q2 a 4 0
dy 3 (1 a2 /y 2 )2 y r ydy 2 a 2 )2 ( y r du 2 u2 2 2 r a
r 2 a2
1 q2 a 4 0 2u
q2 a 8 0 (r2 a2 )
(7)
To relate this to earlier results, note that the image charge q = (a/r)q is located at radius r = a2 /r. The potential energy between the point charge and
10
its image is PE = = = 1 4 0 1 4 0 1 4 0 qq |r r | q 2 a r(r a2 /r) q 2 a r 2 a2
(8)
Result (7) is only half of (8). This would seem to violate energy conservation. It would seem that we could start with the point charge at innity and allow it to fall in to a distance r from the sphere, liberating a quantity of energy (8), which we could store in a battery or something. Then we could expend an energy equal to (7) to remove the charge back to innity, at which point we would be back where we started, but we would still have half of the energy saved in the battery. It would seem that we could keep doing this over and over again, storing up as much energy in the battery as we pleased. I think the problem is with equation (8). The traditional expression q1 q2 /4 0 r for the potential energy of two charges comes from calculating the work needed to bring one charge from innity to a distance r from the other charge, and it is assumed that the other charge does not move and keeps a constant charge during the process. But in this case one of the charges is a ctitious image charge, and as the point charge q is brought in from innity the image charge moves out from the center of the sphere, and its charge increases. So the simple expression doesnt work to calculate the potential energy of the conguration, and we should take (7) to be the correct result. (b) In this case there are two image charges: one of the same charge and location as in part a, and another of charge Q q at the origin. The work needed to remove the point charge q to innity is the work needed to remove the point charge from its image charge, plus the work needed to remove the point charge from the extra charge at the origin. We calculated the rst contribution above. The second contribution is
q (Q q )dy 4 0 y 2
1 4 0
qQ q 2 a + 3 dy y2 y
r
1 qQ q 2 a = 2 4 0 y 2y 1 qQ q 2 a = + 2 4 0 r 2r so the total work done is W = 1 q2 a q 2 a qQ . 4 0 2(r2 a2 ) 2r2 r
11
Review of Greens Functions

Some problems in this and other chapters use the Greens function technique. Its useful to review this technique, and also to establish my conventions since I dene the Greens function a little dierently than Jackson. The whole technique is based on the divergence theorem. Suppose A(x) is a vector valued function dened at each point x within a volume V . Then ( A(x )) dV = A(x ) dA (9)
where S is the (closed) surface bounding the volume V . If we take A(x) = (x) (x) where and are scalar functions, (9) becomes ((x )) ( (x )) + (x )2 (x ) dV = (x )
S
dA
x
where /n is the dot product of with the outward normal to the surface area element. If we write down this equation with and switched and subtract the two, we come up with 2 2 dV =
S
dA . n n
(10)
This statement doesnt appear to be very useful, since it seems to require that we know over the whole volume to compute the left side, and both and /n on the boundary to compute the right side. However, suppose we could choose (x) in a clever way such that 2 = (x x0 ) for some point x0 within the volume. (Since this is a function of x which also depends on x0 as a parameter, we might write it as x0 (x).) Then we could use the sifting property of the delta function to nd (x0 ) =
V
x0 (x )2 (x ) dV +
(x )
S
x0 n
x0 (x )
dA .
x
If is the scalar potential of electrostatics, we know that 2 (x ) = (x )/ 0 , so we have (x0 ) = 1

0 V
x0 (x )(x )dV +
(x )
S
x0 n
(11) Equation (11) allows us to nd the potential at an arbitrary point x0 as long as we know within the volume and both and /n on the boundary. boundary. Usually we do know within the volume, but we only know either or /n on the boundary. This lack of knowledge can be accommodated by choosing such that either its value or its normal derivative vanishes on the boundary surface, so that the term which we cant evaluate drops out of the surface integral. More specically,
x0 (x )
dA .
x
12
if we know but not /n on the boundary (Dirichlet boundary conditions), we choose such that = 0 on the boundary. Then (x0 ) = 1
0 V
x0 (x )(x )dV +
S
(x )
x0 n
dA .
x
(12)
if we know /n but not on the boundary (Neumann boundary conditions), we choose such that /n = 0 on the boundary. Then (x0 ) = 1
0 V
x0 (x )(x )dV +
S
x0 (x )
dA .
x
(13)
Again, in both cases the function x0 (x) has the property that 2 x0 (x) = (x x0 ).

Homer Reid December 8, 1999
Chapter 2: Problems 11-20

Problem 2.11
A line charge with linear charge density is placed parallel to, and a distance R away from, the axis of a conducting cylinder of radius b held at xed voltage such that the potential vanishes at innity. Find (a) the magnitude and position of the image charge(s); (b) the potential at any point (expressed in polar coordinates with the origin at the axis of the cylinder and the direction from the origin to the line charge as the x axis), including the asymptotic form far from the cylinder; (c) the induced surface-charge density, and plot it as a function of angle for R/b=2,4 in units of /2b; (d) the force on the charge.
(a) Drawing an analogy to the similar problem of the point charge outside the conducting sphere, we might expect that the potential on the cylinder can be made constant by placing an image charge within the cylinder on the line conducting the line charge with the center of the cylinder, i.e. on the x axis. Suppose we put the image charge a distance R < b from the center of the cylinder and give it a charge density . Using the expression quoted in Problem 2.3 for the potential of a line charge, the potential at a point x due to the line charge and its image is (x) = 4 ln
0
R2 |x R i|2 1
ln
0
R2 |x R i|2
Homer Reids Solutions to Jackson Problems: Chapter 2 |x R i|2 . |x R i|2
ln
0
We want to choose R such that the potential is constant when x is on the cylinder surface. This requires that the argument of the logarithm be equal to some constant at those points: |x R i|2 = |x R i|2 b2 + R 2 2R b cos = b2 + R2 2Rb cos . For this to be true everywhere on the cylinder, the term must drop out, which requires R = R. We can then rearrange the remaining terms to nd R = b2 . R
or
This is also analogous to the point-charge-and-sphere problem, but there are dierences: in this case the image charge has the same magnitude as the original line charge, and the potential on the cylinder is constant but not zero. (b) At a point (, ), we have = For large , this becomes 4 ln
0
ln
0
2 + R 2 2R cos . 2 + R2 2R cos 1 2R cos 1 2R cos
Using ln(1 x) = (x + x2 /2 + ), we have = (c) =

0
4 2
2(R R ) cos 0 R(1 b2 /R2 ) cos 0
r =b
2b 2R cos 2b 2R cos = 4 b2 + R 2 2bR cos b2 + R2 2bR cos = 2 b2 + b

b2 R
cos
3
b4 R2
2b R cos
b R cos b2 + R2 2bR cos
Homer Reids Solutions to Jackson Problems: Chapter 2 Multiplying the rst term by R2 /b2 on top and bottom yields = = 2
R2 b b b2 2bR cos
R2 +
R 2 b2 2 2b R + b2 2bR cos
with C some constant. We can dierentiate this to nd the electric eld due to the image charge: E(x) = (x) = ln |x R i|2 4 0 i) 2(x R = . 4 0 |x R i|2 2 1 i= RR 2 R i. R 2 b2
(d) To nd the force on the charge, we note that the potential of the image charge is C2 . (x) = ln 4 0 |x R i|2
The original line charge is at x = R, y = 0, and the eld there is E=
The force per unit width on the line charge is F = E = 2 R 2 0 R2 b2
tending to pull the original charge in toward the cylinder.
Problem 2.12
Starting with the series solution (2.71) for the two-dimensional potential problem with the potential specied on the surface of a cylinder of radius b, evaluate the coecients formally, substitute them into the series, and sum it to obtain the potential inside the cylinder in the form of Poissons integral: (, ) = 1 2
2
(b, )
0
b2 2 d b2 + 2 2b cos( )
What modication is necessary if the potential is desired in the region of space bounded by the cylinder and innity?
Referring to equation (2.71), we know the bn are all zero, because the ln term and the negative powers of are singular at the origin. We are left with
(, ) = a0 +
n=1
n {an sin(n) + bn cos(n)} .
(1)
Multiplying both sides successively by 1, sin n , and cos n and integrating at = b gives a0 an bn = = = 1 2 1 bn 1 bn
2
(b, )d
0 2
(2) (3) (4)
(b, ) sin(n)d
0 2
(b, ) cos(n)d.
0
Plugging back into (1), we nd (, ) = = 1 1

2
(b, )
0 2
1 + 2 n=1 b 1 + 2 n=1 b
[sin(n) sin(n ) + cos(n) cos(n )] d cos n( ) . (5)
(b, )
0
The bracketed term can be expressed in closed form. For simplicity dene x = (/b) and = ( ). Then 1 + xn cos(n) 2 n=1
= = = = = =
1 1 + xn ein + xn ein 2 2 n=1 1 2 1 2 1 2 1 2 1 2 1 1 1 + 2 2 1 xei 1 xei 1 1 xei xei + 1 + 2 2 1 xei xei + x2 1 x cos + 1 1 + x2 2x cos x cos x2 + 1 + x2 2x cos 1 x2 . 1 + x2 2x cos +
Plugging this back into (5) gives the advertised result.
Problem 2.13
(a) Two halves of a long hollow conducting cylinder of inner radius b are separated by small lengthwise gaps on each side, and are kept at dierent potentials V1 and V2 . Show that the potential inside is given by (, ) = V1 V 2 V1 + V 2 + tan1 2 b2 2b cos 2
where is measured from a plane perpendicular to the plane through the gap. (b) Calculate the surface-charge density on each half of the cylinder. This problem is just like the previous one. Since we are looking for an expression for the potential within the cylinder, the correct expansion is (1) with expansion coecients given by (2), (3) and (4): 1 2
2
a0
= = =
(b, )d
0
an
= = = =
bn
2 1 cos(n)d cos(n)d + V2 V1 n b 0 1 2 = V1 |sin n|0 + V2 |sin n| nbn = 0.
2 1 V1 d + V2 d 2 0 V1 + V 2 2 2 1 sin(n)d sin( n ) d + V V 2 1 bn 0 1 2 V1 |cos n|0 + V2 |cos n| nbn 1 [V1 (cos n 1) + V2 (1 cos n )] nbn 0 , n even 2(V1 V2 )/(nbn ) , n odd
With these coecients, the potential expansion becomes (, ) = V1 + V 2 2(V1 V2 ) + 2 1 n b

n
sin n.
(6)
n odd
Here we need an auxiliary result: 1 n x sin n = n = = 1 2i 1 (iy )n [ein ein ] n (x = iy )
n odd
n odd (1)n 1 (yei )2n+1 (yei )2n+1 2 n=0 2n + 1 1 tan1 (yei ) tan1 (yei ) 2 (7)
where in the last line we just identied the Taylor series for the inverse tangent function. Next we need an identity: tan1 1 tan1 2 = tan1 1 2 1 + 1 2 .
(I derived this one by drawing some triangles and doing some algebra.) With this, (7) becomes 1 n x sin n = n = Using this in (6) with x = /b gives (, b) = V1 V 2 V1 + V 2 + tan1 2b sin b2 2 . 1 tan1 2 1 tan1 2 2iy sin 1 + y2 2x sin 1 x2 .
n odd
(Evidently, Jackson and I dened the angle dierently).
Problem 2.15
(a) Show that the Green function G(x, y ; x , y ) appropriate for Dirichlet boundary conditions for a square two-dimensional region, 0 x 1, 0 y 1, has an expansion
G(x, y ; x , y ) = 2
n=1
gn (y, y ) sin(nx) sin(nx )
where gn (y, y ) satises 2 n2 2 gn (y, y ) = (y y ) y 2 and gn (y, 0) = gn (y, 1) = 0.
(b) Taking for gn (y, y ) appropriate linear combinations of sinh(ny ) and cosh(ny ) in the two regions y < y and y > y , in accord with the boundary conditions and the discontinuity in slope required by the source delta function, show that the explicit form of G is G(x, y ; x , y ) = 2 1 sin(nx) sin(nx ) sinh(ny< ) sinh[n (1 y> )] n sinh( n ) n=1
where y< (y> ) is the smaller (larger) of y and y . (I have taken out a factor 4 from the expressions for gn and G, in accordance with my convention for Greens functions; see the Greens functions review above.) (a) To use as a Greens function in a Dirichlet boundary value problem G must satisfy two conditions. The rst is that G vanish on the boundary of the region of interest. The suggested expansion of G clearly satises this. First, sin(nx ) is 0 when x is 0 or 1. Second, g (y, y ) vanishes when y is 0 or 1. So G(x, y ; x , y ) vanishes for points (x , y ) on the boundary. The second condition on G is 2 G = 2 2 + 2 x y 2 G = (x x ) (y y ). (8)
With the suggested expansion, we have 2 G = 2 gn (y, y ) sin(nx) n2 2 sin(nx ) x 2 n=1 2 2 G = 2 gn (y, y ) sin(nx) sin(nx ) 2 y y 2 n=1

We can add these together and use the dierential equation satised by gn to nd
2 G = (y y ) 2
sin(nx) sin(nx )
n=1
= (y y ) (x x ) since the innite sum is just a well-known representation of the function. (b) The suggestion is to take gn (y, y ) = An1 sinh(ny ) + Bn1 cosh(ny ), An2 sinh(ny ) + Bn2 cosh(ny ), y < y; y > y. (9)
The idea to use hyperbolic sines and cosines comes from the fact that sinh(ny ) and cosh(ny ) satisfy a homogeneous version of the dierential equation for g n (i.e. satisfy that dierential equation with the function replaced by zero). Thus gn as dened in (9) satises its dierential equation (at all points except y = y ) for any choice of the As and B s. This leaves us free to choose these coecients as required to satisfy the boundary conditions and the dierential equation at y = y . First lets consider the boundary conditions. Since y is somewhere between 0 and 1, the condition that gn vanish for y = 0 is only relevant to the top line of (9), where it requires taking Bn1 = 0 but leaves An1 undetermined for now. The condition that gn vanish for y = 1 only aects the lower line of (9), where it requires that 0 = An2 sinh(n ) + Bn2 cosh(n ) = (An2 + Bn2 )en + (An2 + Bn2 )en One way to make this work is to take An2 + Bn2 = en Then Bn2 = en + An2 so An2 = cosh(n ) 2An2 = en en and An2 + Bn2 = en .
(10)
and
Bn2 = sinh(n ).
With this choice of coecients, the lower line in (9) becomes gn (y, y ) = cosh(n ) sinh(ny )+sinh(n ) cosh(ny ) = sinh[n (1y )] (11) for (y > y ). Actually, we havent completely determined An2 and Bn2 ; we could multiply (11) by an arbitrary constant n and (10) would still be satised. Next we need to make sure that the two halves of (9) match up at y = y : An1 sinh(ny ) = n sinh[n (1 y )]. (12)

70000
60000
50000
g(yprime)
40000
30000
20000
10000
0 0 0.2 0.4 yprime 0.6 0.8 1
Figure 1: gn (y, y ) from Problem 2.15 with n=5, y=.41
This obviously happens when An1 = n sinh[n (1 y )] and n = n sinh(ny ) where n is any constant. In other words, we have gn (y, y ) = n sinh[n (1 y )] sinh(ny ), n sinh[n (1 y )] sinh(ny ), y < y; y > y. (13)
= n sinh[n (1 y> )] sinh(ny< )
with y< and y> dened as in the problem. Figure 1 shows a graph of this function n = 5, y = .41. The nal step is to choose the normalization constant n such that gn satises its dierential equation: 2 2y 2 n2 2 gn (y, y ) = (y y ). (14)
To say that the left-hand side equals the delta function requires two things: that the left-hand side vanish at all points y = y , and that its integral over any interval (y1 , y2 ) equal 1 if the interval contains the point y = y , and vanish otherwise. The rst condition is clearly satised regardless of the choice of n . The second condition may be satised by making gn continuous, which we have already done, but giving its rst derivative a nite jump of unit magnitude at y = y :
10
gn (y, y ) y
y =y +
= 1.
y =y
Dierentiating (13), we nd this condition to require nn [ cosh[n (1 y )] sinh(ny ) sinh[n (1 y )] cosh(ny )] = nn sinh(n ) = 1 so (14) is satised if n = Then (13) is gn (y, y ) = sinh[n (1 y> )] sinh(ny< ) n sinh(n ) 1 . n sinh(n )
and the composite Greens function is
G(x, y ; x , y )
= 2 = 2
gn (y, y ) sin(nx) sin(nx )

n=1
sinh[n (1 y> )] sinh(ny< ) sin(nx) sin(nx ) (15) . n sinh(n ) n=1
Problem 2.16
A two-dimensional potential exists on a unit square area (0 x 1, 0 y 1) bounded by surfaces held at zero potential. Over the entire square there is a uniform charge density of unit strength (per unit length in z ). Using the Green function of Problem 2.15, show that the solution can be written as (x, y ) = 4 3
0
sin[(2m + 1)x] (2m + 1)3 m=0
cosh[(2m + 1) (y (1/2))] cosh[(2m + 1)/2]
Referring to my Greens functions review above, the potential at a point x0 within the square is given by (x0 ) = 1
0 V
G(x0 ; x )(x )dV +

S
(x )
G n
(16) In this case the surface integral vanishes, because were given that vanishes on the boundary, and G vanishes there by construction. Were also given that
G(x0 ; x )
dA .
x
11
(x )dV = dx dy throughout the entire volume. Then we can plug in (15) to nd 2 (x0 ) = 0 1 n sinh( n ) n=1
1 0 0 1
sinh[n (1y> )] sinh(ny< ) sin(nx0 ) sin(nx )dx dy . (17)
The integrals can be done separately. The x integral is

1
sin(nx0 )
0
sin(nx )dx
= =
sin(nx0 ) [cos(n ) 1] n (2 sin(nx0 ))/n , n odd 0 , n even
(18)
The y integral is
y0 1
sinh[n (1 y0 )] = = =
sinh(ny )dy + sinh(ny0 )

0 y0
sinh[n (1 y )]dy
1 y0 1 sinh[n (1 y0 )] cosh(ny ) 0 sinh[ny0 ] cosh[n (1 y )] y0 n 1 {sinh[n (1 y0 )] cosh(ny0 ) + sinh(ny0 ) cosh[n (1 y0 )] sinh(ny0 ) sinh[n (1 y0 )]} n 1 {sinh[n ] sinh[n (1 y0 )] sinh(ny0 )}. (19) n
Inserting (18) and (19) in (17), we have (x0 ) = 4 3

0
n odd
sin(nx0 ) n3
sinh[n (1 y0 )] + sinh(ny0 ) sinh(n )
The thing in brackets is equal to what Jackson has, but this is tedious to show so Ill skip the proof.
12
Problem 2.17
(a) Construct the free-space Green function G(x, y ; x , y ) for twodimensional electrostatics by integrating 1/R with respect to z z between the limits Z , where Z is taken to be very large. Show that apart from an inessential constant, the Green function can be written alternately as G(x, y ; x , y ) = ln[(x x )2 + (y y )2 ] = ln[2 + 2 2 cos( )]. (b) Show explicitly by separation of variables in polar coordinates that the Green function can be expressed as a Fourier series in the azimuthal coordinate, 1 eim( ) gm (, ) G= 2 where the radial Green functions satisfy 1 gm m2 ( ) gm = . 2
Note that gm (, ) for xed is a dierent linear combination of the solutions of the homogeneous radial equation (2.68) for < and for > , with a discontinuity of slope at = determined by the source delta function. (c) Complete the solution and show that the free-space Green function has the expansion G(, ; , ) = 1 1 ln(2 >) 4 2 1 m m=1
< >
cos[m( )]
where < (> ) is the smaller (larger) of and . (As in Problem 2.15, I modied the text of the problem to match with my convention for Greens functions.) (a) R = [(x x )2 + (y y )2 + (z z )2 ]1/2 [a2 + u2 ]1/2 , a = [(x x )2 + (y y )2 ]1/2
Z Z
, u = (z z ).
+Z Z
Integrating, [a2 du + u2 ]1/2 = ln (a2 + u2 )1/2 + u
Homer Reids Solutions to Jackson Problems: Chapter 2 (Z 2 + a2 )1/2 + Z (Z 2 + a2 )1/2 Z
13
= ln = ln ln = ln
2+
(1 + (a2 /Z 2 ))1/2 + 1 (1 + (a2 /Z 2 ))1/2 1

a2 2Z 2 a2 2Z 2 2
4Z + a2 a2 2 = ln[4Z + a2 ] ln a2 . Since Z is much bigger than a, the rst term is essentially independent of a and is the nonessential constant Jackson is talking about. The remaining term is the 2D Greens function: G = ln a2 = ln[(x x )2 + (y y )2 ] in rectangular coordinates
= ln[2 + 2 2 cos( )] in cylindrical coordinates.
(b) The 2d Greens function is dened by 2 G(, ; , ) d d = 1 but 2 G = 0 at points other than (, ). These conditions are met if 2 G(, ; , ) = 1 ( ) ( ). (20)
You need the on the bottom there to cancel out the in the area element in the integral. The Laplacian in two-dimensional cylindrical coordinates is 2 = 1 1 . 2 2
Applying this to the suggested expansion for G gives 2 G(, ; , ) = 1 2
gm
m2 gm eim( ) . 2
If gm satises its dierential equation as specied in the problem, the term in brackets equals ( )/ for all m and may be removed from the sum, leaving 2 G(, ; , ) = = ( ) ( ) 1 2
eim( )
( ).
14
(c) As in Problem 2.15, well construct the functions gm by nding solutions of the homogenous radial dierential equation in the two regions and piecing them together at = such that the function is continuous but its derivative has a nite jump of magnitude 1/. For m 1, the solution to the homogenous equation 1 is f ( ) = Am m + Bm m . Thus we take gm = A1m m + B1m m A2m m + B2m m , < , > . m2 2 f ( ) = 0
In order that the rst solution be nite at the origin, and the second solution be nite at innity, we have to take B1m = A2m = 0. Then the condition that the two solutions match at = is A1m m = B2m m which requires A1m = m m for some constant m . Now we have m gm = m dgm d or mm so m = Then gm = 1 2m 1 2m 1 2m
m m m
B2m = m m
m m
, ,
< >
The nite-derivative step condition is dgm d =

=
=+
1 1 + 1 . 2m
, < , >
< >
15
Plugging this back into the expansion gives G = = 1 4 1 2
1 m 1 m
< > < >
eim( )
m
cos[m( )].
Jackson seems to be adding a ln term to this, which comes from the m = 0 solution of the radial equation, but I have left it out because it doesnt vanish as .
Problem 2.18
(a) By nding appropriate solutions of the radial equation in part b of Problem 2.17, nd the Green function for the interior Dirichlet problem of a cylinder of radius b [gm (, = b) = 0. See (1.40)]. First nd the series expansion akin to the free-space Green function of Problem 2.17. Then show that it can be written in closed form as G = ln or G = ln 2 2 + b4 2 b2 cos( ) b2 (2 + 2 2 cos( )) (b2 2 )(b2 2 ) + b2 | |2 . b2 | |2
(b) Show that the solution of the Laplace equation with the potential given as (b, ) on the cylinder can be expressed as Poissons integral of Problem 2.12. (c) What changes are necessary for the Green function for the exterior problem (b < < ), for both the Fourier expansion and the closed form? [Note that the exterior Green function is not rigorously correct because it does not vanish for or . For situations in which the potential falls of fast enough as , no mistake is made in its use.]
(a) As before, we write the general solution of the radial equation for gm in the two distinct regions: gm (, ) = A1m m + B1m m A2m m + B2m m , < , > . (21)
The rst boundary conditions are that gm remain nite at the origin and vanish on the cylinder boundary. This requires that B1m = 0
16
and A2m bm + B2m bm = 0 so A2m = m bm B2m = m bm for some constant m . Next, gm must be continuous at = : A1m m A1m With this we have gm (, ) = m = m b b
m
= m = m m
b b
b m b
b b
, ,
<
> .
Finally, dgm /d must have a nite jump of magnitude 1/ at = . 1 = dgm d dgm d b

m
=+ m1
= mm = 2mm so
bm b
+
m
m+1
mm
1 1 2m b
m m
m = and gm (, ) = = or gm (, ) = 1 2m 1 2m b2 b2 1 2m
b2
,
m
< > .
, < >
m
Plugging into the expansion for G gives G(, , , ) = 1 2 1 m n=1
b2
< >
cos m( ).
(22)
17
Here we need to work out an auxiliary result: 1 n x cos n( ) = n n=1 =

0 x 0 x
n=1 x
un1 du cos m( )
1 un cos n( ) du u n=1 cos( ) u 1 + u2 2u cos( ) du

x 0
=
0
1 ln(1 2u cos( ) + u2 ) 2 1 = ln[1 2x cos( ) + x2 ]. 2 =
(I summed the innite series here back in Problem 2.12. The integral in the second-to-last step can be done by partial fraction decomposition, although I cheated and looked it up on www.integrals.com). We can apply this result individually to the two terms in (22): G(, ; , ) = 1 1 + ( /b2 )2 2( /b2 ) cos( ) ln 4 1 + (< /> )2 2(< /> ) cos( ) 2 1 b4 + 2 2 2 b2 cos( ) > = ln 2 4 b4 2 > + < 2< > cos( ) 2 1 b4 + 2 2 2 b2 cos( ) > = ln 2 4 4 b 2 > + < 2< > cos( ) 2 1 > = ln 4 b2 b4 + 2 2 2 b2 cos( ) 1 ln 2 2 4 b ( + 2 2 cos( ))
(23)
This is Jacksons result, with an additional ln term thrown in for good measure. Im not sure why Jackson didnt quote this term as part of his answer; he did include it in his answer to problem 2.17 (c). Did I do something wrong? (b) Now we want to plug the expression for G above into (16) to compute the potential within the cylinder. If there is no charge inside the cylinder, the volume integral vanishes, and we are left with the surface integral: (, ) = (b, ) G dA .
=b
(24)
where the integral is over the surface of the cylinder. For this we need the normal derivative of (23) on the cylinder: G 1 = 4 2 2 cos( ) 22 2b2 cos( ) 2 2 2 2 + 2 b cos( ) + 2 2 cos( )
b4
18
Evaluated at = b this is G = 1 2 2 b 2 b(2 + b2 2b cos( )) .
=b
In the surface integral, the extra factor of b on the bottom is cancelled by the factor of b in the area element dA , and (24) becomes just the result of Problem 2.12. (c) For the exterior problem we again start with the solution (21). Now the boundary conditions are dierent; the condition at gives A2m = 0, while the condition at b gives A1m = m bm B1m = m bm .
m
From the continuity condition at = we nd A2m = m m b

m
The nite derivative jump condition gives mm or m = b

m
1 mm 1 2m
m
b
m
1 1 =
Putting it all together we have for the exterior problem gm = 1 2m b2 < >
m
This is the same gm we came up with before, but with b2 and terms ipped in rst term. But the closed-form expression was symmetrical in those two expressions (except for the mysterious ln term) so the closed-form expression for the exterior Greens function should be the same as the interior Greens function.

Homer Reid June 15, 2000
Problem 3.1
Two concentric spheres have radii a, b(b > a) and each is divided into two hemispheres by the same horizontal plane. The upper hemisphere of the inner sphere and the lower hemisphere of the outer sphere are maintained at potential V . The other hemispheres are at zero potential. Detemine the potential in the region a r b as a series in Legendre polynomials. Include terms at least up to l = 4. Check your solution against known results in the limiting cases b and a 0. The expansion of the electrostatic potential in spherical coordinates for problems with azimuthal symmetry is
(r, ) =
l=0
Al rl + Bl r(l+1) Pl (cos ).
(1)
We nd the coecients Al and Bl by applying the boundary conditions. Multiplying both sides by Pl (cos ) and integrating from -1 to 1 gives
1
(r, )Pl (cos )d(cos ) =

1
2 Al rl + Bl r(l+1) . 2l + 1
At r = a this yields
1
V
0
Pl (x)dx =
2 Al al + Bl a(l+1) , 2l + 1 1
and at r = b, V
Pl (x)dx =
1
2 Al bl + Bl b(l+1) . 2l + 1
The integral from 0 to 1 vanishes for l even, and is given in the text for l odd:
1 0
(l 2)!! 1 Pl (x)dx = ( )(l1)/2 l+1 . 2 2 2 !
The integral from -1 to 0 also vanishes for l even, and is just the above result inverted for l odd. This gives 1 (l 2)!! V ( )(l1)/2 l+1 2 2 2 ! (l 2)!! 1 V ( )(l1)/2 l+1 2 2 2 ! or l l with = Al al + Bl a(l+1) = Al bl + Bl b(l+1) = = 2 Al al + Bl a(l+1) 2l + 1 2 Al bl + Bl b(l+1) . 2l + 1
1 (2l + 1)(l 2)!! l = V ( )a(l1)/2 . 2 4 l+1 2 ! bl+1 + al+1 a2l+1 b2l+1 al+1 bl+1 (bl + al ) a2l+1 b2l+1
The solution is Al = l Bl = l
The rst few terms of (1) are (r, ) = 3 V 4 (a2 + b2 )r a2 b2 (a + b) a 4 b 4 (a 3 + b 3 ) 7 (a4 + b4 )r3 P (cos ) P3 (cos )+ 1 a3 b 3 r2 (a3 b3 ) 16 a7 b 7 r4 (a7 b7 )
In the limit as b , the problem reduces to the exterior problem treated in Section 2.7 of the text. In that limit, the above expression becomes (r, ) 3 V 4 a r
2
P1 (cos )
7 V 16
a r
P3 (cos ) +
in agreement with (2.27) with half the potential spacing. When a 0, the problem goes over to the interior version of the same problem, as treated in section 3.3 of the text. In that limit the above expression goes to 3 (r, ) V 4 7 r P1 (cos ) + V b 16 r b
3
P3 (cos ) +
This agrees with equation (3.36) in the text, with the sign of V ipped, because here the more positive potential is on the lower hemisphere.
Problem 3.2
A spherical surface of radius R has charge uniformly distributed over its surface with a density Q/4R2 , except for a spherical cap at the north pole, dened by the cone = . (a) Show that the potential inside the spherical surface can be expressed as = Q 8 0
l=0
1 rl [Pl+1 (cos ) Pl1 (cos )] l+1 Pl (cos ) 2l + 1 R
where, for l = 0, Pl1 (cos ) = 1. What is the potential outside? (b) Find the magnitude and direction of the electric eld at the origin. (c) Discuss the limiting forms of the potential (part a) and electric eld (part b) as the spherical cap becomes (1)very small, and (2) so large that the area with charge on it becomes a very small cap at the south pole.
(a) Lets denote the charge density on the sphere by (). At a point innitesimally close to the surface of the sphere, the electric eld is F = = so r =
r =R
(2)
The expression for the potential within the sphere must be nite at the origin, so the Bl in (1) are zero. Dierentiating that expansion, (2) becomes (r, ) = r
lAl rl1 Pl (cos )

l=1
Multiplying by Pl and integrating at r = R gives 1

0 1
()Pl (cos )d(cos ) =

1
2l Al Rl1 2l + 1 Pl (x)dx.
so Al =
2l + 1 2lRl1
Q 4R2
cos 0 1
To evaluate the integral we use the identity (eq. 3.28 in the text) Pl (x) = d 1 [Pl+1 (x) Pl1 (x)] (2l + 1) dx
so
cos
Pl (x)dx =
1
1 [Pl+1 (cos ) Pl1 (cos )] . 2l + 1
(We used the fact that Pl+1 (1) = Pl1 (1) for all l.) With this we have Al = Q [Pl+1 (cos ) Pl1 (cos )] 8 0 lRl+1
so the potential expansion is (r, ) = Q 8 0
l=1
1 rl [Pl+1 (cos ) Pl1 (cos )] l+1 Pl (cos ). l R
Within the body of the sum, I have an l where Jackson has a 2l + 1. Also, he includes the l = 0 term in the sum, corresponding to a constant term in the potential. I dont understand how he can determine that constant from the information contained in the problem; the information about the charge density only tells you the derivative of the potential. Theres nothing in this problem that xes the value of the potential on the surface beyond an arbitrary constant. (b) The eld at the origin comes from the l = 1 term in the potential: E(r = 0) = |r=0 1 r+ r r r=0 d Q [P2 (cos ) 1] P1 (cos ) r + P1 (cos ) = 8 0 R2 d Q 3 3 = cos r sin cos2 8 0 R2 2 2 = = 3Q sin2 k. 16 0 R2
The eld points in the positive z direction. That makes sense, since a positive test charge at the origin would sooner y up out through the uncharged cap than through any of the charged surface.
Problem 3.3
A thin, at, conducting, circular disk of radius R is located in the x y plane with its center at the origin, and is maintained at a xed potential V . With the information that the charge density on a disc at xed potential is proportional to (R2 2 )1/2 , where is the distance out from the center of the disc, (a) show that for r > R the potential is 2V R (r, , ) = r (b) nd the potential for r < R. (c) What is the capacitance of the disk? We are told that the surface charge density on the disk goes like (r) = K (R2 r2 )1/2 1 r 2 K 31 1+ = + R 2 R (2!)(2 2) = K R (2n 1)!! n! 2n n=0

l=0
(1)l 2l + 1
R 2l r
P2l (cos )
r R
r R
2n
531 (3!)(2 2 2)
r R
+ (3)
for some constant K . From the way the problem is worded, I take it were not supposed to try to gure out what K is explicitly, but rather to work the problem knowing only the form of (3). At a point innitesimally close to the surface of the disk (i.e., as /2), the component of in the direction normal to the surface of the disk must be proportional to the surface charge. At the surface of the disk, the normal direction. Hence direction is the negative 1 (r, ) r =
0
(4)
=(/2)
with the plus (minus) sign valid for above (below) the disc. For r < R the potential expansion is
(r, ) =
l=0
Al rl Pl (cos ).
(5)
Combining (3), (4), and (5) we have
Al rl1
l=0
d Pl (cos ) d
cos =0
K R 0
(2n 1)!! n! 2n n=0
r R
2n
(6)
For l even, dPl /dx vanishes at x = 0. For l odd, I used some of the Legendre polynomial identities to derive the formula d P2l+1 (x) dx = (1)l (2l + 1) (2l 1)!! . l ! 2l
x=0
This formula reminds one strongly of expansion (3). Plugging into (6) and equating coecents of powers of r, we nd A2l+1 = so (r, ) = A0 K
0
(1)l K (2l + 1)R2l+1 r R
l=1
(1)l 2l + 1
2l+1
P2l+1 (cos ).
I wrote A0 explicitly because we havent evaluated it yetthe derivative condition we used earlier gave no information about it. To nd A0 , observe that, on the surface of the disk (cos = 0), all the terms in the above sum vanish ( because Pl (0) is 0 for odd l) so = A0 on the disk. But = V on the disk. Therefore, A0 = V . We have (r, ) = V K
0
l=1
(1)l 2l + 1
r R
2l+1
P2l+1 (cos )
(7)
where the plus (minus) sign is good for less than (greater than)/2. Note that the presence of that sign preserves symmetry under reection through the z axis, a symmetry that is clearly present in the physical problem. (a) For r > R, there is no charge. Thus the potential and its derivative must be continuous everywherewe cant have anything like the derivative discontinuity that exists at = /2 for r < R. Since the physical problem is symmetric under a sign ip in cos , the potential expansion can only contain Pl terms for l even. The expansion is
(r, ) =
l=0
B2l r(2l+1) P2l (cos ).
At r = R, this must match up with (7): V K

0
l=1
(1)l P2l+1 (cos ) = 2l + 1
B2l R(2l+1) P2l (cos ).

l=0
Multiplying both sides by P2l (cos ) sin() and integrating gives B2l 2R(2l+1) 4l + 1
1
= V
1
Pl (x)dx + 2K
0
K
0
l=1
(1)l 2l + 1
1
P2l+1 (x)P2l (x)dx +

1 0
P2l+1 (x)Pl (x)dx
= 2V l,0 +
l=1
(1)l 2l + 1
P2l+1 (x)P2l (x)dx.

0
but I cant do this last integral.
Problem 3.4
The surface of a hollow conducting sphere of inner radius a is divided into an even number of equal segments by a set of planes; their common line of intersection is the z axis and they are distributed uniformly in the angle . (The segments are like the skin on wedges of an apple, or the earths surface between successive meridians of longitude.) The segments are kept at xed potentials V , alternately. (a) Set up a series representation for the potential inside the sphere for the general case of 2n segments, and carry the calculation of the coecients in the series far enough to determine exactly which coecients are dierent from zero. For the nonvanishing terms exhibit the coecients as an integral over cos . (b) For the special case of n = 1 (two hemispheres) determine explicitly the potential up to and including all terms with l = 3. By a coordinate transformation verify that this reduces to result (3.36) of Section 3.3.
(a) The general potential expansion is

l
(r, , ) =
l=0 m=l
Alm rl + Blm r(l+1) Ylm (, ).
(8)
For the solution within the sphere, niteness at the origin requires Blm = 0. Multiplying by Ylm and integrating over the surface of the sphere we nd Alm = = = 1 al V al
n (a, , ) Ylm (, ) d
k=1
(1)k
2k/n 2(k1)/n 1/2 Ylm (, ) sin d d 1 1 n 2k/n 2(k1)/n
V 2l + 1 (l m)! al 4 (l + m)!
Plm (x) dx
k=1
(1)k
eim d . (9)
The integral is easy:

2k/n 2(k1)/n
eim d =
1 e2imk/n e2im(k1)/n . im
This is to be summed from k = 1 to n with a factor of (1)k thrown in: = =
1 (e2mi(1/n) 1) (e2mi(2/n) e2mi(1/n) ) + (1 e2mi((n1)/n) ) im 2 1 e2mi/n + e2(2mi/n) e3(2mi/n) + + e(n1)(2mi/n) . (10) im
Putting x = exp(2mi/n), the thing in braces is 1 + x + x2 + x3 + + xn1 = 1 xn 1 e2mi = , 1x 1 + e2im/n
Note that the numerator vanishes. Thus the only way this thing can be nonzero is if the denominator also vanishes, which only happens if the exponent in the denominator equates to -1. This only happens if m/n = 1/2, 3/2, 5/2, . In that case, the 2mi/n term in the exponent of the terms in (10) equates to i, so all the terms with a plus sign in (10) come out to +1, while all the terms with a minus sign come out to -1, so all n terms add constructively, and (10) equates to 2n , m = n/2, 3n/2, 5n/2, = im 0, otherwise. Then the expression (9) for the coecients becomes Alm = 2nV 2l + 1 (l m)! imal 4 (l + m)!
1/2 1 1
Plm (x)dx,
m=
n 3n , , = 0, otherwise. 2 2
(b) As shown above, the only terms that contribute are those with m = n/2, m = 3n/2, et cetera. Of course there is also the constraint that m < l. Then, with n = 2, up to l = 3 the only nonzero terms in the series (9) are those with l = 1, m = 1, and l = 3, m = 1 or 3. We need to evaluate the integral for these terms. We have
1 1 1 1 1 1 1 P1 (x) dx 1 P3 (x) dx 3 P3 (x) dx 1
= =
1 1 1
(1 x2 )1/2 dx = (1 x2 )1/2
1
3 15 2 3 dx = x 2 2 8 15 . 4
= 15
(1 x2 )3/2 dx =
Using these results in (??), we have A11 A31 A33 = = = 3 4V i a 4 2 3V i 7 2 2a3 4 4! 5V i 7 a3 4 6!

1/2
1/2
1/2
Now we can plug these coecients into (8) to piece together the solution. This involves some arithmetic in combining all the numerical factors in each
coecient, which I have skipped here. (r, , ) = V 3 r 7 r 3 sin (5 cos2 1) sin sin sin + a 16 a 7 r 3 3 + sin sin 3 + 144 a
Problem 3.6
Two point charges q and q are located on the z azis at z = +a and z = a, respectively. (a) Find the electrostatic potential as an expansion in spherical harmonics and powers of r for both r > a and r < a. (b) Keeping the product qa = p/2 constant, take the limit of a 0 and nd the potential for r = 0. This is by denition a dipole along the z azis and its potential. (c) Suppose now that the dipole of part b is surrounded by a grounded spherical shell of radius b concentric with the origin. By linear superposition nd the potential everwhere inside the shell. (a) First of all, for a point on the z axis the potential is (z ) = = = 1 q 1 4 0 |z a| z + a a q a a 2 a + 1 1+ + + 4 0 z z z z z q a a 3 + + 2 0 z z z
for z > a. Comparing this with the general expansion = at = 0 we can identify the Bl s and write (r, ) = q 2 0 r a a P1 (cos ) + r r
3
Bl r(l+1) Pl (cos )
P3 (cos ) +
for r > a. For r < a we can just swap a and r in this equation. (b) (r, ) = = qa a 2 P1 (cos ) + P3 (cos ) + 2 2 0 r r a 2 p P3 (cos ) + P1 (cos ) + 2 4 0 r r p cos as a 0. 4 0 r2
10
(c) When we put the grounded sphere around the two charges, a surface charge distribution forms on the sphere. Lets denote by s the potential due to this charge distribution alone (not including the potential of the dipole) and by d the potential due to the dipole. To calculate s , we pretend there are no charges within the sphere, in which case we have the general expansion (1), with Bl = 0 to keep us nite at the origin. The total potential is just the sum s + d : (r, ) = p 4 0 r2
cos +
l=0
Al rl Pl (cos ).
The condition that this vanish at r = b ensures, by the orthogonality of the Pl , that only the l = 1 term in the sum contribute, and that p . A1 = 4 0 b3 The total potential inside the sphere is then r p P1 (cos ). 1 (r, ) = 2 4 0 b b
Problem 3.7
Three point charges (q, 2q, q ) are located in a straight line with separation a and with the middle charge (2q ) at the origin of a grounded conducting spherical shell of radius b, as indicated in the gure. (a) Write down the potential of the three charges in the absence of the grounded sphere. Find the limiting form of the potential as a 0, but the product qa2 = Q remains nite. Write this latter answer in spherical coordinates. (b) The presence of the grounded sphere of radius b alters the potential for r < b. The added potential can be viewed as caused by the surface-charge density induced on the inner surface at r = b or by image charges located at r > b. Use linear superposition to satisfy the boundary conditions and nd the potential everywhere inside the sphere for r < a and r > a. Show that in the limit a 0, Q r5 (r, , ) 1 P2 (cos ). 2 0 r3 b5 (a) On the z axis, the potential is (z ) = = = q 2 1 1 + + 4 0 z |z a| z + a q a a 2 + 1 + + 4 0 r z z q a 2 a 4 + + . 2 0 z z z
+ 1
a a + z z
11
As before, from this result we can immediately infer the expression for the potential at all points: (r, ) = = a 4 a 2 q P2 (cos ) + P4 (cos ) + 2 0 r r r qa2 a 2 P4 (cos ) + P2 (cos ) + 3 2 0 r r Q P2 (cos ) as a 0 2 0 r3
(11)
(b) As in the previous problem, the surface charges on the sphere produce an extra contribution s to the potential within the sphere. Again we can express s with the expansion (1) (with Bl = 0), and we add s to (11) to get the full potential within the sphere: (r, ) = Q P2 (cos ) + 2 0 r3
Al rl Pl (cos )
l=0
From the condition that vanish at r = b, we determine that only the l = 2 term in the sum contributes, and that A2 = Then the potential within the sphere is (r, ) = Q r 1 3 2 0 r b
5
Q . 2 0 b5
P2 (cos ).
Problem 3.9
A hollow right circular cylinder of radius b has its axis coincident with the z axis and its ends at z = 0 and z = L. The potential on the end faces is zero, while the potential on the cylindrical surface is given as V (, z ). Using the appropriate separation of variables in cylindrical coordinates, nd a series solution for the potential anywhere inside the cylinder. The general solution of the Laplace equation for problems in cylindrical coordinates consists of a sum of terms of the form R()Q()Z (z ). The function is of the form Q() = A sin + B cos
12
with an integer. The z function is of the form Z (z ) = Cekz + Dekz . In this case, Z must vanish at z = 0 and z = L, which means we have to take k imaginary, i.e. n , n = 1, 2, 3, Z (z ) = C sin(kn z ) with kn = L With this form for Z , R must be taken to be of the form R() = EI (kn ) + F K (kn ). Since were looking for the potential on the inside of the cylinder and there is no charge at the origin, the solution must be nite as 0, which requires F = 0. Then the potential expansion becomes

(, , z ) =
n=1 =0
[An sin + Bn cos ] sin(kn z )I (kn ).
(12)
Multiplying by sin sin kn z and integrating at r = b, we nd

L 0 0 2
V (, z ) sin sin(kn z ) d dz = so An = Similarly, Bn = 2 LI (kn b)

L 0 0 2
L I (kn b)An 2 (13)
2 LI (kn b)
L 0 0
V (, z ) sin() sin(kn z ) d dz.
V (, z ) cos() sin(kn z ) d dz.
(14)
Problem 3.10
For the cylinder in Problem 3.9 the cylindrical surface is made of two equal halfcylinders, one at potential V and the other at potential V , so that V (, z ) = V V for /2 < < /2 for /2 < < 3/2
(a) Find the potential inside the cylinder. (b) Assuming L >> b, consider the potential at z = L/2 as a function of and and compare it with two-dimensional Problem 2.13. The potential expansion is (12) with coecients given by (13) and (14). The relevant integrals are
L 0 0 2
V (, z ) sin() sin(kn z ) d dz

L /2 3/2
13
= V
0
sin(kn z ) dz
/2
sin() d
sin() d
/2
= 0
L 0 L 0 /2 3/2 2
V (, z ) cos() sin(kn z ) d dz
= V
0
sin(kn z ) dz
/2 /2
cos() d
cos() d
/2
= =
2V kn An Bn
|sin |/2 |sin |/2
3/2
(n odd)
Hence, from (13) and (14),
0 , n or even 8V /kn , n odd, = 1, 5, 9, 8V /kn , n odd, = 3, 7, 11, = 0 = 0, = (1)( 1)/2 16V /(n 2 I (kn b)), 16V 2 n or even n and odd
The potential expansion is (, , z ) = (1)( 1)/2 cos() sin(kn z )I (kn ) nIv (kn b) (15)
n,
where the sum contains only terms with n and odd. (b) At z = L/2 we have (, , L/2) = 16V 2 (1)(n+ 2)/2 I (kn ) cos() . n I (kn b)
n,
As L , the arguments to the I functions become small. Using the limiting form for I quoted in the text as equation (3.102), we have (, ) = 16V 2 (1)(n+ 2)/2 cos() n b
n,
The sums over n and are now decoupled: (, ) = = = 16V 2 16V 2 (1)n 2n + 1 n=0 4

(1) cos() 2 + 1 b =0
(1) cos() 2 + 1 b =0 2b cos b2 2
4V tan1
14
This agrees with the result of Problem 2.13, with V1 = V2 = V . The rst series is just the Taylor series for tan1 (x) at x = 1, so it sums to /4. The second series can also be put into the form of the Taylor series for tan1 (x), using tricks exactly analogous to what I did in my solution for Problem 2.13.

Homer Reid June 15, 2000

Problem 3.11
A modied Bessel-Fourier series on the interval 0 a for an arbitrary function f () can be based on the homogenous boundary conditions: At = 0, J (k) d J (k ) = 0 d d ln[J (k)] = d a
At = a,
( real)
The rst condition restricts . The second condition yields eigenvalues k = yn /a, where yn is the nth positive root of x dJ (x)/dx + J (x) = 0. (a) Show that the Bessel functions of dierent eigenvalues are orthogonal in the usual way. (b) Find the normalization integral and show that an arbitrary function f () can be expanded on the interval in the modied Bessel-Fourier series
f () =
n=1
A n J
yn a
with the coecients An given by 2 An = 2 a 2 1 2 yn

2 J (yn )
dJ (yn ) dyn
2 1 0
f ()J
yn d. a
(a) The function J (k) satises the equation d 2 1 d J (k) + k 2 2 d d J (k) = 0. (1)
Multiplying both sides by J (k ) and integrating from 0 to a gives

a
J (k )
0
d 2 d J (k) + k 2 d d
J (k )J (k)
d = 0.
(2)
The rst term on the left can be integrated by parts:

a
J (k )
0
d d J (k) d d d
a
= J (k )
d J (k) d
a 0
d J (k ) d
d J (k) d. d
(3)
One of the conditions were given is that the thing in braces in the rst term here vanishes at = 0. At = a we can invoke the other condition: d ln[J (k)] d =
=a
1 d J (k) J (k) d
=a
d a J (ka) = J (ka). d Plugging this into (3), we have

a
J (k )
0
d d J (k) d d d
a
= J (k )J (k)
d J (k ) d
d J (k) . d
(4)
This is clearly symmetric in k and k , so when we write down (2) with k and k switched and subtract from (2), the rst integral (along with the 2 / term) vanishes, and we are left with (k 2 k 2 ) proving orthogonality. (b) If we multiply (1) by 2 J (k) and integrate, we nd
a 0 a
J (k )J (k) d = 0
0
J (k)
d [J (k)]d+k 2 d
a 0
2 J (k)J (k)d 2
a 0
J (k)J (k)d = 0. (5)
The rst and third integrals are of the form f (x)f (x)dx and can be done immediately. In the second integral we put f () = 2 J (k), g () = J (k) and integrate by parts:
a 0 2 2 J (k)J (k)d = 2 J (k) a a 0 a
2 J (k)d a 0
a 0
2 J (k)J (k)d
2 J (k)J (k)d =
1 2 2 a J (ka) 2
a 0
2 J (k)d.
Using this in (5), a2 2 (ak )2 2 J (ka) + aJ (ka) k 2 2 2 so

a 0 2 J (k)d 2 J (k)d
2 2 J (ka) = 0 2
= =
2 a2 2 2 2k a2 2 1
2 J (ka) +
a2 2 J (ka) 2k 2 d J (ka) d(ka)

2
2 (ka)2
2 J (ka) +
This agrees with what Jackson has if you note that k is chosen such that ka = ynm .
Problem 3.12
An innite, thin, plane sheet of conducting material has a circular hole of radius a cut in it. A thin, at, disc of the same material and slightly smaller radius lies in the plane, lling the hole, but separated from the sheet by a very narrow insulating ring. The disc is maintained at a xed potential V , whilc the innite sheet is kept at zero potential. (a) Using appropriate cylindrical coordinates, nd an integral expression involving Bessel functions for the potential at any point above the plane. (b) Show that the potential a perpendicular distance z above the center of the disc is z 0 (z ) = V 1 a2 + z 2 (c) Show that the potential a perpendicular distance z above the edge of the disc is kz V K (k ) 1 a (z ) = 2 a where k = 2a/(z 2 + 4a2 )1/2 , and K (k ) is the complete elliptic integral of the rst kind.
(a) As before, we can write the potential as a sum of terms R()Q()Z (z ). In this problem there is no dependence, so Q = 1. Also, the boundary conditions on Z are that it vanish at and be nite at 0, whence Z (z ) exp(kz ) for any k. Then the potential expansion becomes
(, z ) =
0
A(k )ekz J0 (k) dk.
(6)
To evaluate the coecients A(k ), we multiply both sides by J0 (k ) and integrate over at z = 0:

(, 0)J0 (k ) d
0
=
0
A(k )
0
J0 (k)J0 (k ) d dk
= so
A(k ) k
A(k )
= k
0
(, 0)J0 (k) d
a
= kV
0
J0 (k)d.
Plugging this back into (6),

a 0
(, z ) = V
0
k ekz J0 (k)J0 (k ) d dk.
(7)
The integral can be done right away. To do it, I appealed to the dierential equation for J0 : 1 J0 (u) + J0 (u) + J0 (u) = 0 u so
x 0 x x
uJ0 (u) du =
uJ0 du
x 0
0 x
J0 (u) du
x
= |uJ0 (u)|0 +
x
J0 (u) du
J0 (u) du
= |uJ0 (u)|0 = xJ0 (x) = xJ1 (x). (In going from the rst to second line, I integrated by parts.) Then (7) becomes
(, z ) = aV
0
J1 (ka)J0 (k)ekz dk.
(8)
(b) At = 0, (7) becomes

a
(0, z ) = V J0 (0)
0 a
kekz J0 (k )dk d
= V
0 a

0 a
z z
ekz J0 (k )dk d
0
= V = V
0 2
1 + z2
2
z d ( 2 + z 2 )3/2
Here we substitute u = + z 2 , du = 2 d: (0, z ) = V zJ0 (0) 2 1 u1/2

a2 +z 2 z2 a2 +z 2 z2
u3/2 du
= V z = Vz
1 1 z z2 + z2 z = V 1 a2 + z 2 (b) At = a, (8) becomes
(a, z ) = aV
0
J1 (ka)J0 (ka)ekz dk
Problem 3.13
Solve for the potential in Problem 3.1, using the appropriate Green function obtained in the text, and verify that the answer obtained in this way agrees with the direct solution from the dierential equation. For Dirichlet boundary value problems, the basic equation is (x) = 1
0 V
G(x; x )(x ) dV +
S
(x )
G(x; x ) n
dA .
x
(9)
Here there is no charge in the region of interest, so only the surface integral contributes. The Greens function for the two-sphere problem is
l
G(x; x ) =
l=0 m=l
Ylm ( , ) Ylm (, ) Rl (r; r ) 2l + 1
(10)
with Rl (r; r ) = 1
1
a b 2l+1
l r<
a2l+1 l+1 r<
1
l+1 r>
l r>
b2l+1
(11)
Actually in this case the potential cannot have any dependence, so all terms with m = 0 in (10) vanish, and we have G(x; x ) = 1 4
Pl (cos )Pl (cos )Rl (r; r ).

l=0
In this case the boundary surfaces are spherical, which means the normal to a surface element is always in the radial direction: 1 G(x; x ) = n 4
Pl (cos )Pl (cos )

l=0
Rl (r; r ). n
The surface integral in (9) has two parts: one integral S1 over the surface of the inner sphere, and a second integral S2 over the surface of the outer sphere: S1 = 1 4
Pl (cos )
l=0
Rl n
r =a 0 1 0
(a, )Pl (cos )a2 sin d d
V = 2 = where V 2
a2 Pl (cos )
l=0
Rl n
Pl (x) dx
r =a 0
l=0
a2 l Pl (cos )
1
Rl n
r =a
=
0
Pl (x) dx l odd l even.
(l 2)!! 1 , = ( )(l1)/2 2 2[(l + 1)/2]! = 0, A similar calculation gives S2 = = V 2 V 2
b2 Pl (cos ) b l Pl (cos )
l=0 2
Rl n Rl n
Pl (x) dx
r =b 1
l=0
r =b
because Pl is odd for l odd, so its integral from -1 to 0 is just the negative of the integral from 0 to 1. The nal potential is the sum of S1 and S2 : (r, ) = V 2
l Pl (cos ) r 2
l=0
Rl n
r =b
(12)
r =a
Since the point of interest is always between the two spheres, to nd the normal derivative at r = a we dierentiate with respect to r< , and at r = b with respect to r> . Also, at r = a the normal is in the +r direction, while at r = b the normal is in the negative r direction. a2 b2 Rl (r; r ) n Rl (r; r ) n = (2l + 1)a2
r =a
al1 1 b 1
a 2l+1 b (l+2) a b 2l+1
1 rl+1 rl
rl b2l+1
= (2l + 1)b2
r =b
a2l+1 rl+1
Combining these with some algebra gives (r, ) = V 2
(2l + 1)l Pl (cos )

l=0
(ab)l+1 (bl + al )r(l+1) (al+1 + bl+1 )rl b2l+1 a2l+1
in agreement with what we found in Problem 3.1.
Problem 3.14
A line charge of length 2d with a total charge Q has a linear charge density varying as (d2 z 2 ), where z is the distance from the midpoint. A grounded, conducting spherical shell of inner radius b > d is centered at the midpoint of the line charge. (a) Find the potential everywhere inside the spherical shell as an expansion in Legendre polynomials. (b) Calculate the surface-charge density induced on the shell. (c) Discuss your answers to parts a and b in the limit that d << b.
First of all, we are told that the charge density (z ) = (d2 z 2 ), and that the total charge is Q, whence
d
Q = 2
0
(d2 z 2 )dz = =
4 3 d 3
3Q . 4d3 In this case we have azimuthal symmetry, so the Greens function is G(x; x ) = 1 4
Pl (cos )Pl (cos )Rl (r; r )

l=0
(13)
with
l Rl (r; r ) = r<
1
l+1 r>
l r>
b2l+1
Since the potential vanishes on the boundary surface, the potential inside the sphere is given by (r, ) = 1
0 V
G(r, ; r , )(r , )dV.
In this case is only nonzero on the z axis, where r = z . Also, Pl (cos )=1 for z > 0, and (1)l for z < 0. This means that the contributions to the integral from the portions of the line charge for z > 0 and z < 0 cancel out for odd l, and add constructively for even l: (r, ) = We have
d d
1 4 0
Pl (cos ) 2
l=0,2,4,... 0
Rl (r; z )(z ) dz
Rl (r; z )(z ) dz =
0 0
l r<
1
l+1 r>
l r>
b2l+1
(d2 z 2 ) dz
This is best split up into two separate integrals:

d
=
0
l r< (d2 z 2 ) dz 2l+1 l+1 b r>
d 0
l l r< r> (d2 z 2 ) dz
The second integral is symmetric between r and r , so we may integrate it directly: b2l+1
0 d l l r< r> (d2 z 2 ) dz
d rl z l (d2 z 2 ) dz b2l+1 0 dl+3 rl dl+3 = 2l+1 b l+1 l+3
rl dl+3 (l + 1)(l + 3)b2l+1
(14)
The rst integral must be further split into two:

d
l r< (d2 z 2 ) dz l+1 r>

r 0 d r
= = = =
1 rl+1 1 rl+1
z l (d2 z 2 ) dz + rl
l+3
d2 z 2 dz z l+1
d r
r l 2 2 r2 d2 r2 d2 + + d l+1 l+3 d l(l + 2) l l+2 r2 r l 2 2 d2 d + (l + 2)(l + 3) l(l + 1) d l(l + 2)
d r r 1 d2 + rl l + l+1 l+3 lz (l 2)z l2
2 l+1
Combining this with (14), we have 2 rl dl+3 l(l + 2) (l + 1)(l + 3)b2l+1 0 (15) But something is wrong here, because with this result the nal potential will contain terms like r 0 Pl (cos ) and r2 Pl (cos ), which do not satisfy the Laplace equation. Rl (r; z )(z ) dz =
l d
r r2 d2 + (l + 2)(l + 3) l(l + 1) d
d2
10
Problem 3.15
Consider the following spherical cow model of a battery connected to an external circuit. A sphere of radius a and conductivity is embedded in a uniform medium of conductivity . Inside the sphere there is a uniform (chemical) force in the z direction acting on the charge carriers; its strength as an eective electric eld entering Ohms law is F . In the steady state, electric elds exist inside and outside the sphere and surface charge resides on its surface. (a) Find the electric eld (in addition to F ) and current density everywhere in space. Determine the surface-charge density and show that the electric dipole moment of the sphere is p = 4 0 a3 F/( + 2 ). (b) Show that the total current owing out through the upper hemisphere of the sphere is I= 2 a2 F + 2
Calculate the total power dissipation outside the sphere. Using the lumped circuit relations, P = I 2 Re = IVe , nd the eective external resistance Re and voltage Ve . (c) Find the power dissipated within the sphere and deduce the eective internal resistance Ri and voltage Vi . (d) Dene the total voltage through the relation Vt = (Re + Ri )I and show that Vt = 4aF/3, as well as Ve + Vi = Vt . Show that IVt is the power supplied by the chemical force.
(a) Whats going on in this problem is that the conductivity has a discontinuity going across the boundary of the sphere, but the current density must be constant there, which means there must an electric eld discontinuity in inverse proportion to the conductivity discontinuity. To create this electric eld discontinuity, there has to be some surface charge on the sphere, and this charge gives rise to extra elds both inside and outside the sphere. Since there is no charge inside or outside the sphere, the potential in those two regions satised the Laplace equation, and may be expanded in Legendre polynomials:
11
for r < a, for r > a,
(r, ) = in (r, ) = (r, ) = out (r, ) =
Al rl Pl (cos )
l=0
Bl r(l+1) Pl (cos )
l=0
Continuity at r = a requires that Al al = Bl al+1 so (r, ) = in (r, ) = out (r, ) =

l l=0 Al r Pl (cos ), 2l+1 (l+1) r Pl (cos ), l=0 Al a
Bl = a2l+1 Al
r<a r > a.
(16)
Now, in the steady state there can be no discontinuities in the current density, because if there were than there would be more current owing into some region of space than out of it, which means charge would pile up in that region, which would be a growing source of electric eld, which would mean we arent in steady state. So the current density is continuous everywhere. In particular, the radial component of the current density is continuous across the boundary of the sphere, i.e. Jr (r = a , ) = Jr (r = a+ , ). (17) Outside of the sphere, Ohms law says that J = E = out . Inside the sphere, there is an extra term coming from the chemical force: ) = ( + F k ). J = (E + F k in Applying (17) to these expressions, we have Using (16), this is
r in
+ F cos
r =a
r out
r =a
F P1 (cos )
lAl al1 Pl (cos ) =

l=0
(l + 1)Al al1 Pl (cos ).

l=0
Multiplying both sides by Pl (cos ) and integrating from to , we nd F A1 = 2A1 (18)
12
for l=1, and lAl = (l + 1)Al (19) (20) for l = 1. Since the conductivity ratio is positive, the second relation is impossible to satisfy unless Al = 0 for l = 1. The rst relation becomes A1 = F. + 2 Then the potential is (r, ) =
+2 +2
F r cos , F a3 r2 cos ,
r<a r>a
(21)
The dipole moment p is dened by (r, ) 1 pr 4 0 r3 as r . (22)
The external portion of (21) can be written as (r, ) = F a3 z + 2 r3
and comparing this with (22) we can read o . F a3 k p = 4 0 + 2 The electric eld is found by taking the gradient of (21): E(r, ) =
+2 F k, +2
r<a ), r > a (2 cos r + sin
a 3 r
The surface charge s () on the sphere is proportional to the discontinuity in the electric eld: s () = =
0 [Er (r
3 0 F cos . + 2
= a+ ) Er (r = a )]
(b) The current owing out of the upper hemisphere is just J dA = ) dA (Ein + F k + 2
/2 2 0
= 1
F
0
cos sin a2 d d (23)
a2 F =2 + 2
13
The Ohmic power dissipation in a volume dV is dP = E 2 dV (24)
To see this, suppose we have a rectangular volume element with sides dx, dy , and dz . Consider rst the current owing in the x direction. The current density there is Ex and the cross-sectional area is dydz , so I = Ex dydz. Also, the voltage drop in the direction of current ow is V = Ex dx. Hence the power 2 dissipation due to current in the x direction is IV = Ex dV . Adding in the contributions from the other two directions gives (24). For the power dissipated outside the sphere we use the expression for the electric eld we found earlier:
0 0 2 2
Pout =
E 2 (r, , )r2 sin d d dr

a
= 2 8 = 3
+ 2 + 2
F 2 a6
a 2 0
1 (4 cos2 + sin2 ) sin d dr r4
F 2 a3
Dividing by (23), we nd the eective external voltage Ve : Ve = Pout /I = and the eective external resistance: 2 . Re = Pout /I 2 = 3a (c) The power dissipated inside the sphere is Pin = )2 dV = (E + F k = 4 2 F2 ( + 2 )2 dV 4 aF 3 + 2
16 2 a3 F 2 3( + 2 )2
Since were in steady state, the current owing out through the upper hemisphere of the sphere must be replenished by an equal current owing in through the lower half of the sphere, so to nd the internal voltage and resistance we can just divide by (23): 8 Vi = Pin /I = aF 3 + 2 4 Ri = Pin /I 2 = . 3a
14
(c) (Re + Ri )I = 2 3a 1 2 + 2 4 a2 F = aF + 2 3
(Vi + Ve ) =
4aF 4 + 2 = aF 3( + 2 ) 3
Problem 3.17
The Dirichlet Green function for the unbounded space between the planes at z = 0 and z = L allows discussion of a point charge or a distribution of charge between parallel conducting planes held at zero potential. (a) Using cylindrical coordinates show that one form of the Green function is G(x, x ) =

1 L
Im n < n> Km . L L
eim( ) sin
n=1 m=
nz sin L
nz L
(b) Show that an alternative form of the Green function is G(x, x ) =

0
1 2 sinh(kz< ) sinh[k (L z> )] . sinh(kL)
dk eim( ) Jm (k)Jm (k )
m=
In cylindrical coordinates, the solutions of the Laplace equation look like linear combinations of terms of the form Tmk (, , z ) = eim Z (kz )Rm (k). (25)
There are two possibilities for the combination Z (kz )Rm (k), both of which solve the Laplace equation: Z (kz )Rm (k) = (Aekz + Bekz )[CJm (k) + DNm (k)] or Z (kz )Rm (k) = (Aeikz + Beikz )[CIm (k) + DKm (k)]. (27) (26)
The Greens function G(x; x ) must be a solution of the Laplace equation, and must thus take one of the above forms, at all points x = x. At x = x, G must be continuous, but have a nite discontinuity in its rst derivative.
15
Furthermore, G must vanish on the boundary surfaces. These conditions may be met by dividing space into two regions, one on either side of the source point x, and taking G to be dierent linear combinations of terms T (as in (25)) in the two regions. The question is, in which dimension (i.e., , z , or ) do we dene the two sides of the source point? (a) The rst option is to imagine a cylindrical boundary at = , i.e. at the radius of the source point, and take the inside and outside of the cylinder (i.e., < and > ) as the two distinct regions of space. Then, within each region, the entire range of z must be handled by one function, which means this one function must vanish at z = 0 and z = L. This cannot happen with terms of the form (26), so we are forced to take Z and R as in (27), with B = A and k restricted to the discrete values kn = n/L. Next considering the singularities of the functions in (27), we see that, to keep G nite everywhere, for the inner region ( < ) we can only keep the Im (k) term, while for the outer region we can only keep the Km (k) term. Then G(x; x ) will consist of linear combinations of terms T as in (25) subject to the restrictions discussed above: G(x; x ) = Amn (x)eim sin(kn z )Im (kn ), im sin(kn z )Km (kn ), mn Bmn (x)e
mn
< > .
Clearly, to establish continuity at = , we need to take Amk (x) = mk (z, )Km (k) and Bmk (x) = mk (z, )Im (k), where mk is any function of z and . Then we can write G as G(x; x ) =
mk
mk (z, )eim sin(kz )Im (k< )Km (k> ).
The obvious choice of mk needed to make this a delta function in z and is mk = (4/L)eim sin(kz ). Then we have G(x; x ) = 4 L eim( ) sin(kz ) sin(kz )Im (k< )Km (k> ).
mk
What I dont quite understand is that this expression already has the correct delta function behavior in , even though I never explicitly required this. To obtain this expression I rst demanded that it satisfy the Laplace equation for all points x = x, that it satisfy the boundary conditions of the geometry, and that it have the right delta function behavior in z and . But I never demanded that it have the correct delta function behavior in , and yet it does. I guess the combination of the requirements that I did impose on this thing is already enough to ensure that it meets the nal requirement. (b) The second option is to imagine a plane boundary at z = z , and take the two distinct regions to be the regions above and below the plane. In other words, the rst region is that for which 0 z z , and the second region that for which z z L. In this case, within each region the entire range of (from 0 to ) must be handled by one function. This requirement excludes terms of the form
16
(27), because Km is singular at the origin, while Im is singular at innity, and there is no linear combination of these functions that will be nite over the whole range of . Hence we must use terms of the form (26). To ensure niteness at the origin we must exlude the Nm term, so D = 0. To ensure vanishing at z = 0 we must take A = B , so the z function in the region 0 z z is proportional to sinh(kz ). To ensure vanishing at z = L we must take A = Be2kL , so the z function in the region z z L is proportional to sinh[k (z L)]. With these restrictions, the dierential equation and the boundary conditions are satised for all terms of the form (25) with no limitation on k . Hence the Greens function will be an integral, not a sum, over these terms: G(x ; x) =
m=0 0 m=0 0
Am (k, , , z )eim sinh(kz )Jm (k ) dk, Bm (k, , , z )eim sinh[k (z L)]Jm (k ) dk,
0z z zz L
Problem 3.18
The conguration of Problem 3.12 is modied by placing a conducting plane held at zero potential parallel to and a distance L away from the plane with the disc insert in it. For deniteness put the grounded plane at z = 0 and the other plane with the center of the disc on the z axis at z = L. (a) Show that the potential between the planes can be written in cylindrical coordinates (z, , ) as
(z, ) = V
0
dJ1 ()J0 (/a)
sinh(z/a) . sinh(L/a)
(b) Show that in the limit a with z, , L xed the solution of part a reduces to the expected result. Viewing your result as the lowest order answer in an expansion in powers of a1 , consider the question of corrections to the lowest order expression if a is large compared to and L, but not innite. Are there diculties? Can you obtain an explicit estimate of the corrections? (c) Consider the limit of L with (L z ), a and xed and show that the results of Problem 3.12 are recovered. What about corrections for L a, but not L ? (a) The general solution of the Laplace equation in cylindrical coordinates with angular symmetry that vanishes at z = 0 is
(, z ) =
0
A(k )J0 (k) sinh(kz ) dk.
(28)
17
Multiplying both sides by J0 (k ) and integrating at z = L yields

J0 (k )(, L) d =
0 0
A(k ) sinh(kL)
0
J0 (k )J0 (k) d 1 (k k ) k dk
dk
=
0
A(k ) sinh(kL) 1 A(k ) sinh(k L) k
= so A(k ) =
k sinh(kL) Vk = sinh(kL) V k sinh(kL)

x
J0 (k)(, L) d
0 a
J0 (k) d
0 ka
uJ0 (u) du.

0
(29)
I worked out this integral earlier, in Problem 3.12: uJ0 (u) du = xJ1 (x).
0
Then (29) becomes A(k ) = and (28) is (, z ) = V sinh(kz ) dk sinh( kL) 0 sinh(z/a) d. =V J1 ()J0 (/a) sinh( L/a) 0 aJ1 (ka)J0 (k) 1 J0 (x) 1 x2 + 4 1,
V (ka)J1 (ka) k sinh(kL)
(30)
(b) For x
1,
and for x
1 and y
3 x+ 1 x sinh(x) 1 6x + = = 1 + (x2 y 2 ) + O(x4 ) 1 3 sinh(y ) y 6 y + 6y +
With these approximations we may expand the terms containing a in (30): J0 (/a) sinh(z/a) sinh(L/a) = 1 1 4 a a
2 2
z L
1+
1 6
(x2 y 2 ) +
(31) (32)
z 1 L
1 1 2 (L z 2 ) + 2 6 4
18
Then the potential expansion (30) becomes (, z ) = Vz L

0
J1 () d
1 1 1 2 (L z 2 ) + 2 a2 6 4
2 J1 () d +
The rst integral evaluates to 1, so for a innite the potential becomes simply (z ) = V z/L. This is just what we expect to get for the potential between two innite sheets, one grounded and the other at potential V. The second integral, unfortunately, has a bit of an innity problem. Its not hard to see where the problem comes: I derived the expansion above based on the premise that /a is small, but the integral goes over all up to , so for any nite a the expansions eventually become invalid in the integral. Im still trying to work out a better procedure for estimating corrections for nite a. (c) In this part were interested in taking L and looking at the potential a xed distance away from the plane with the circular insert. Calling the xed distance z , the z coordinate of the point were interested in is L z . We have sinh k (L z ) sinh(kL) cosh(kz ) + cosh(kL) sinh(kz ) = sinh kL sinh kL = cosh(kz ) coth(kL) sinh(kz )
(33)
Now, coth(kL) diers signicantly from 1 only for kLa 1, in which region kz z/L 1, so cosh(kz ) 1 and sinh(kz ) 0. By the time k gets big enough that kz is starting to get signicant, coth(kL) has long since started to look like 1, so the two terms in (33) add directly. The result is that, for all k , (33) can be approximated as exp(kz ). Then (30) becomes
(, z ) = aV
0
J1 (ka)J0 (k)ekz dk
as we found in Problem 3.12.

Homer Reid August 6, 2000

Problem 3.19
Consider a point charge q between two innite parallel conducting planes held at zero potential. Let the planes be located at z = 0 and z = L in a cylindrical coordinate system, with the charge on the z axis at z = z0 , 0 < z0 < L. Use Greens reciprocation theorem of Problem 1.12 with Problem 3.18 as the comparison problem. (a) Show that the amount of induced charge on the plate at z = L inside a circle of radius a whose center is on the z axis is given by QL (a) = q (z0 , 0) V
(b) Show that the induced charge density on the upper plate can be written as () = q 2
dk
0
sinh(kz0 ) kJ0 (k) sinh(kL)
(c) Show that the charge density at = 0 is (0) = q z0 sec2 8L2 2L
(a) Greens reciprocation theorem says that dV +

V S
dA =
V
dV +
S
dA.
(1)
Well use the unprimed symbols to refer to the quantities of Problem 3.18, and the primed symbols to refer to those of Problem 3.19. Then (r, z ) = 0 (r, z ) =? (r, z ) = 0, = 0, = V,
z=0 z = L and r > a z = L and r < a dk aJ1 (ak )J0 (rk )

0
=V
sinh(kz ) sinh(kL)
0<z<L
(r, z ) = q (r) (z z0 ) (r, z ) =? (r, z ) = 0, =?, Plugging into (1),
z = 0 or z = L 0zL sinh(kz0 ) +V sinh(kL)
qV
0
dk aJ1 (ak )
(r, z ) dA = 0
z =L,r<a
so (r, z ) dA = q dk aJ1 (ak )

0
z =L,r<a
sinh(kz0 ) q = (z0 , 0) sinh(kL) V
(2)
The integral on the left is just the total surface charge contained within a circle of radius a around the origin of the plane z = L. (b) The integrand on the left of (2) doesnt depend on , so we can do the angular part of the integral right away to give
a
2
0
(r, L)r dr = q
dk aJ1 (ak )
0
sinh(kz0 ) sinh(kL)
Dierentiating both sides with respect to a, we have
2a (a, L) = q
dk
0
sinh(kz0 ) [aJ1 (ak )] a sinh(kL)
(3)
where Ive blithely assumed that the partial derivative can be passed through the integral sign. The partial derivative is [aJ1 (ak )] = [xJ1 (x)] a x
x=ak
= |J1 (x) + xJ1 (x)|x=ak = |xJ0 (x)|x=ak = akJ0 (ak )
so (3) becomes (a, L) = (c) At a = 0, (4) becomes (0, L) = q 2
q 2
dk kJ0 (ak )
0
sinh(kz0 ) sinh(kL)
(4)
k
0
sinh(kz0 ) . sinh(kL)
I have no idea how to do this integral.
Problem 3.22
The geometry of a two-dimensional potential problem is dened in polar coordinates by the surfaces = 0, = , and = a, as indicated in the sketch. Using separation of variables in polar coordinates, show the the Green function can be written as
G(, ; , ) =
m=1
1 m/ m <
1
m/ >
> a2m/
m/
sin
sin
Problem 2.25 may be of use. As before, the procedure for determining the Greens function is to split the region of interest into two parts (one on each side of the observation point), nd separate solutions of the Laplace equation that satisfy the boundary conditions in each region, and then join the two solutions at the source point such that their values match up but the rst derivative (in whichever dimension we chose sides) has a nite discontinuity. Suppose the observation point is (, ). Lets break the region into two subregions, dened by 0 and a. The general solution of the Laplace equation in two-dimensional polar coordinates is ( , ) =A0 + B0 ln +
n
n [An sin n + Bn cos n ] + n [Cn sin n + Dn cos n ].
The solution in the rst region must be admissible down to = 0, which excludes the ln term and the negative powers of . However, these terms may be included in the solution for the second region. In both regions, the solution must vanish at = 0, which excludes the cos terms (i.e. Bn = Dn = 0). The solution must also vanish at = , which requires that n = m/ , m = 1, 2, . With these considerations we may write down the solutions for G in the two regions:
G(, ; , )
=
m=1
Am m/ sin
, m ,
0 a
(5) (6)
=
m=1
Bm m/ + Cm m/ sin
The solution in the second region must vanish at = a for all , i.e. Bm am/ + Cm am/ = 0 so Bm = m am/ and Cm = m am/ a
m/
where m can be anything. Then (6) becomes
G(, ; , ) =
m=1
m/
sin
a.
The solutions in the two regions must agree on the boundary between the two regions, i.e. at = . This determines Am and m : Am = m a
m/
m/
m = m m/
where m can be anything. Using these expressions for Am , Bm , and Cm we can write G(, ; , )
=
m=1
m m
m=1
a a
m/
a a
m/
m/ sin
m/
m m
0 a.
m/
m/ sin
This may be more succintly written as G(, ; , ) =

m
m fm (; ) sin
(7)
where fm (; ) = > a
m/
a >
m/
<
m/
The nal step is to choose the constant m in (7) such as to make 2 G(, ; , ) = The Laplacian of (7) is 2 G = 1 2 2 G= + 2 2 2 m
m
1 ( ) ( ).
2
(8)
d2 fm (; ) d 2 m fm (; ) =
fm (; ) sin
This is equal to (8) if m = m and m d2 fm (; ) d 2 m 1 sin

2
(9) 1 ( ).
At all points = , the latter condition is already satised by f as we constructed it earlier. At = , the condition is achieved by choosing m to satisfy =+ 1 d = . m fm (; ) (10) d = Referring to (7), we have d fm d d fm d =
++
m m
a a
m/
a a
m/
m/ 1
m/
(11) (12)
=
+
m/
m/ 1 .
Subtracting (12) from (11) we obtain dfm d Then from (10) we read o m = and plugging this into (9) gives m = 1 m/ a sin 2m m . m/ a 2m
=+
=
=
2m m/ 1 a .
Plugging this into (7) we obtain nally G(, ; , ) =

m
1 2m
< > a2
m/
< >
m/
sin
sin
I seem to be o by a factor of 2 here, but I cant nd where.
!#"$%'&()103254768#@9 ACBED8FGF'HPIGDB@QRBTSUIWVYXG`5a8b3cdD8efHPIgF 93hi!prq pst# u vxw@ysy yTywGyGxx edgfihkjmlonqpsrtl#udvqwjmx y{zs|g} ~C3m
U#ttx83EtYtYPgtW P!tYttE5kt38tkGY EEWPkkGkt888Eggt#EE 7tUtY8x'3gtYgPgYY mtxt5kEt3UEYtt3Y UtkPEt58#ttW5 YEtEE'tmx Ut8tEm'! WTt8#Pt7kg8 Gg3t$YtPP8tkgt$#tEU$ Pg x kt#P#83Etg 3 7 i 8 gE Ett@Ptg 5Yt!YU#T 3 3 8 e U eP q eP g U g # t! t
!#"%$#&' ($)1032547689'A@BC8D!EFHGPIQR2TSU$VW
UEq t X ! E Y ! 7 t 5 Y ` a bcbcb m 5 ` # Y bcbTb #x 'EUgETWYedtEEUt tf XEstE3$Pg5s338tg55gm7P RPYWtEtG 38t t h g 5 i g t i h ` g T i q k t p rt ) s k W8mtEC uvu s g7 txE! w X't5 tE xPUk YEUgE$Etc uyug$tEgkt5 E t 's$t P3 83ttxE gg gT! ttkg3'3TgYgY t#3P#83EtT t ! t) u 5 u u ! ) Pt # y t Y Y kET tgg ss$xW t)gts83tE!PY#Ps #xEY t7 f u YtEt UkT u e d s Et#kU5PT# ef u # e d e d
8tEGUvXYtEtt'! GT#UYG xEt# u t u # t u # g bTbcb e f u # e d tE3EUg! ~C3m #" xGxgtYxg88% $YE3tgx u tPEt xYtTg'&WgtkP@'Etxt gtEtGxmtt!xYxEmxYxE$ CdE#5YE 0 $ u c ) ( ( ( d b!132
#PT
Cd )( 0$ b4132 ( ( u %57698 2 ( ( u 357@A8 2 )( ( u B5DCE8 2 )H H u 8 8 8GF
kxE!Uk'$5tttkYEs ( uPIRQ C d ( S ( ( S 8UT 5 6 I Q 8 2 ( ( u S ( ( S 8UT bcbTbYX uWV % 8 ktt!$x#tEsc
U tq#E#'ExEGYt 2 GYt gPX E D 5 6 s u )( u V S u S bcbTb X $ ( ( ( ( u u S (b ( u S gtYtEt#xt78YP'$ ( u ~C3m YEeEEPt u F 5EEP7EtttEgY ttxEggPt mtPEtGtEg EGt#t$gE5tkF YPgYedtgtEt x gtPg5Wt t3st#tkYtYt WPtx83EtW8EkYtT3Et t$Y8 PYtxW88E! Y tgtkk 8 8@DCFE u !% $ 032 5 1 6 47 u @ # " ( 6 d u ( HG I bcbcb 1'& $ b #)( 1 p 89 @BA C P u )rRQ xEWtkW3tt' SUTVXW k8Et sUtE$8xEgt# T`Y a tgt3x 8 8ihp e 0 C C u ph 8 E 8i E u 8 8 A A Y & $cbd# u c g f g f C C bTbcb u x 5xkt5YtY83t t#EPx
8 ( ( u S( ( S 8 I 8 2 2 )( ( u S ( ( S I 2 ( ( u 8 S ( ( S 8 8 8 S ( ( S d 8 S ( g( u u S

!#"%$#&' ($)1032547689'A@BC8D!EFHGPIQR2TSU$VW `

"
# u gE 5WE8YtYt! Wtx # t$Pt X8 t#3xg'kT E u 8 @ C 8 E u @ GI 6 47 E u ! E u d u 8 @ C C bcbcb b71 P 8 @ C E 8@ C 8 C @ b E b 1 E S d u b bTbcb s t s YtP3 xttY E ' GPY3t E U8#Ex t5gtt g5ttW#Est E E S d u 1 S d u b $ 1 b CE CE C E 5 C 5 C 5 C ` # g8f XkP 1 b # gEttP P CFE C E CFE C E p C C C C E ` Tx#Egt CE CE C E C E E S $ 5 5 C 5 C C $ b d u C 1 b kttt) E$g5 tt)YgYt C E C E C E C E C E V C C C C C C C gEtYE tEgtkYEtYx#$ ( EP tt p xU88f X3ttkEmE P5k5YEt g$xEUt8 '#t5 YtT5xtUYY5 XtT C CE CE CE CE CE CE C V C C C C C C

C C
C X
X d u
P'3Tg3Xttt 1 b #!5Et T7ETEUgg C CE C E @ CFE C V C C C C E C E C E C C C X 8 8@ C E C p C 8@ A C @ YtEE gt ~C3m EgtB xG 8 3E A PEWY#88tEt Egt3Y C E C C F uk3 F Yg#xtt#P x Y tgt#E%YtxWEt$ u )r s s Xg t A b t x ggE G X P E C E C C F u dt u u gt u u ttUYYYt! u ` a bg GUtYGxTY3#PGtgYE ttY3tG8E3 Y 83Y T U5t xk88 mEEt33Ptk t P ! p g PYtx583E A ` b t5 5$P 38EY " y YBbg # v $ E8kktPYe dxEE#tYP #
C E C
x tT8 8 Ut#P0YP588 A 3Utt3x E7 A t A A A A A Y t UE5Yg gE8P Yx Etgt k$Y$& % XY! '! 8 8 C E p 8 A C d u

C E p C Y 1 b # YU5PT CE CFE C C Ps3gt E CE pA C A
Y 3E'kT
C E C
C E C
C A
C E
C E C
du
EgE3 ' YG8PE88UY UG8 s 'U888 5tYEEttY5 T 3 P tkY g t7t t Xm F Y3m$ Y 5EYtEtYE F

x A b bg b " F b b b u xt s TWPstktt#P u b Y u bgb b b b u u Y bg u b bg u ` b uE u

C E C
)r
Et 8P YtPk88EtggEs A I I u I u x F 8 8 8GF I I u F 8 8 F I F 8GF u I F F F 8GF I V F F X 8GF V ` X
!#"%$#&' ($)1032547689'A@BC8D!EFHGPIQR2TSU$VW a
~C3m
iYEeEE$YtY7WPg8tkYE p x X$txYdtxtgYsttPtsYE E8WPtx83EtEgttWYg8tWxt 5tgg#P !EPY5x8E r &t888xE ) Y#mx# !kE t#t!tE't u s 'ttY##t$P#g 9 YP888 A 'YE8 YWtEtt@EtGt3gP8 Wtk x5#E t # 5 7 E t m tYYt U'tYG3 u u ` a b$ u $ Y 'Eg YY8 G X t stYEtt##t qttE#'t 5 UdYYt!g xgt8stYE $$8EE

x #x#38tgkEs t I x 8UT tt8xEf XPtxY38tYG3tGtE$ xE T Etgt 'E! Xd YEe EE YtYtmPEt gt)E x'E ERP GYP5Wf dxtYtPttUC dE sPtxs'UmYPtWxYm'x kt EW YxPYmWt88Eg x# E5U#T ef uW p b b I 8 8 8 u Iu Iu e f g b V I a 8 X u 8 X bV I

tg# g g s 8 8 I I I $EW U g @#t Ytx8 %E8W xg Wg Ett!' TE T % Y X ! t C g8'gYtGgtY7 ` ttg x 4 6 7 g 8 I tg Yt'UEE5U EE P miEtUt s YtEs# 8 Iu m tG Y 3 tEPE s $d'E t@x f Y8Es$ tt 8TY3E X WEtGT 3P8 xEtkt#tt m! t tY8Y x'EEmU3PT3mPY #xY8Etktg 3 Ptt#YEmgtmYtEtP s ! Wse X# f XE8P8tE XtE Y8 'YE TY8#tEtU tUt a Y'#Yt uW p ` uvuW #tx#Y$#xt t ( u e f $#P#88EUP$P5g )( u p u g p
)r GY7YktUPtYtE5tiUtt! x ( u I S( ( S 8 ( 3 p u b V I e f 8 X b V I 8 8 8 u b V I e f 8 X b V I g 8 X gExUE s3P P UsP Y S ( ( S t t8EW5PxE5gg 5tt g3PT8YmY3 E!t8EkU PttxUkTg5t#PEt u ( u ` gE gtEt5tm txEkxstmU k'kT u 8 8 Iu I g 0 p & p k'kT I u 8 I 8 0 g &Y p p Y Y & t YU#Y u E@# x s @ E % XY! G P 8 CFE 8 p 8 A C d u
'st8PEUPT A A A A gfdP Y 8PttkEgP t C E C E C E p A C 6 C @ C C du C E C F A p 1 b # C du C E C F A

C u F A C E C C C F d u A C F d u
du
gEg$ t5t5E#UE PtYPd t Ttmt 8Y 5PYtTk E gEEU8EgWPYtEttE tYEE s ttt$'kT ( u a p u a p u F g F a p u F g &We dtttUkT C u C F Ps8t ` a p A u XtY3EtUkTW3s st u
#P &3t$Yt#xEY'kT A b A u u b u g b b u b Y b ` a b pHbg p T x gEPt8xkt58

Homer Reid October 8, 2000
Problem 4.8
A very long, right circular, cylindrical shell of dielectric constant / 0 and inner and outer radii a and b, respectively, is placed in a previously uniform electric eld E0 with its axis perpendicular to the eld. The medium inside and outside the cylinder has a dielectric constant of unity. (a) Determine the potential and electric elds in the three regions, neglecting end eects. (b) Sketch the lines of force for a typical case of b 2a. (c) Discuss the limiting forms of your solution appropriate for a solid dielectric cylinder in a uniform eld, and a cylindrical cavity in a uniform dielectric. We will take the axis of the cylinder to be the z axis and the electric eld to be aligned with the x axis: E0 = E0 i. Since the cylinder is very long and were told to neglect end eects, we can ignore the z direction altogether and treat this as a two-dimensional problem. (a) The general solution of the Laplace equation in two dimensional polar coordinates is (r, ) = [An rn + Bn rn ][Cn sin(n) + Dn cos(n)]
For the region inside the shell (r < a), the B coecients must vanish to keep the potential from blowing up at the origin. Also, in the region outside the shell 1
(r > b), the only positive power of r in the sum must be that which gives rise to the external electric eld, i.e. E0 r cos with An = 0 for n > 1. With these observations we may write expressions for the potential in the three regions: rn [An sin n + Bn cos n], rn [Cn sin n + Dn cos n] + rn [En sin n + Fn cos n], E0 r cos + rn [Gn sin n + Hn cos ], r<a a<rb
(r, ) =
The normal boundary condition at r = a is

0
=
x=a
x=a+
or
0
nan1 [An sin n + Bn cos n] = nan1 [Cn sin n + Dn cos n] na(n+1) [En sin n + Fn cos n]
From this we obtain two equations:

0 0
An = Cn En a2n Bn = Dn Fn a2n
(1) (2)
Next, the tangential boundary condition at r = a is or nan [An cos n Bn sin n] = nan [Cn cos n Dn sin n] + nan [En cos n Fn sin n] =
x=a+
x=a
from which we obtain two more equations: An = Cn + En a2n Bn = Dn + F n a

2n
(3) (4)
Similarly, from the normal boundary condition at r = b we obtain

0
E0 cos
nb(n+1) [Gn sin n + Hn cos ] = nbn1 [Cn sin n + Dn cos n] nb(n+1) [En sin n + Fn cos ]
which leads to
0 2 0 0
Gn = Cn b2n En Hn = Dn b2n Fn
(5) (6)
b E0 n1
Finally, we have the tangential boundary condition at r = b: bE0 sin + nbn [Gn cos n Hn sin n] = nbn [Cn cos n Dn sin n] + nbn [En cos n Fn sin n] giving Gn = Cn b2n + En b E0 n1 + Hn = Dn b
2 2n
(7) (8)
+ Fn .
The four equations (1), (3), (5), and (7) specify a degenerate system of linear equations, which can only be satised by taking An = Cn = En = Gn = 0 for all n. Next, for n = 1, the system of equations (2), (4), (6), and (8) specify the same degenerate system of equations, so Bn = Dn = Fn = Gn = 0 for n = 0. However, for n = 1, we have
0
B1 = D1 F1 a2 B1 = D1 + F1 a2
D1 =
1 1+ 2
B1
F1 =
1 2 a 1 2
B1 .
and H1 = b2 E0 +
0
D1 b 2
F1
H1 = b 2 E 0 + D 1 b 2 + F 1 0 = 2b2 E0 + b2 1 +
0
D1 + 1
F1
Substituting from above, 4b2 E0 = or B1 = b2 ( 1

0
b2 ( +
0)
a2 (
0)
B1
4 0b2 + 0 )2 a 2 (
0)
E0 .
Then 2 0 ( + 0 )b2 E0 + 0 )2 a 2 ( 0 )2 2 0 ( 0 )a2 b2 E0 F1 = 2 b ( + 0 )2 a 2 ( 0 )2 2 b2 (b2 a2 )( 2 ) 0 H1 = 2 E0 . 2 2 b ( + 0 ) a ( 0 )2 D1 = b2 ( The potential is 4 0 b2 E0 rcos , b 2 ( + 0 )2 a 2 ( 0 )2 a2 2 0 b2 (r, ) = E0 cos , ( + ) r + ( ) 0 0 b 2 ( + 0 )2 a 2 ( 0 )2 r 2 (b2 a2 )( 2 ) b2 0 E0 cos E0 rcos , 2 2 2 2 b ( + 0) a ( 0) r As
r<a a<r<b b < r.
(b) In Figure 4.1 Ive plotted the eld lines for b = 2a, = 5 0 . Also, as an appendix to this document Ive included the C program I wrote to generate this plot. (c) For a solid dielectric cylinder in a uniform eld, we would have a 0. In that case the eld would look like 2 0 rb ( + 0 )2 r On the other hand, a cylindrical cavity in a uniform dielectric corresponds to
0 , E0 r cos in all three regions, which is reassuring. The electric eld is 4 0 b2 E0 [cos r sin ] , r<a 2 b ( + 0 )2 a 2 ( 0 )2 a2 2 0 b2 ( + 0 ) ( 0 ) 2 E0 cos r 2 2 2 2 b ( + 0 ) a ( 0) r a2 E(r, ) = ( + 0 ) + ( 0 ) 2 E0 sin , a<r<b r 2 2 (b2 a2 )( 2 ) b 0 E0 [cos r + sin ] b 2 ( + 0 )2 a 2 ( 0 )2 r +E0 [cos r sin ] , b < r.
Figure 1: Field lines in Problem 4.8 for b = 2a,
= 5 0.
b , in which case the eld becomes 4 0 E0 i, ( + 0 )2 E(r, ) = 2 0 2 0( 0 ) a E0 i ( + 0) ( + 0 )2 r
r<a
2
E0 [cos r + sin ],
r > a.
Problem 4.9
A point charge q is located in free space a distance d away from the center of a dielectric sphere of radius a (a < d) and dielectric constant / 0 . (a) Find the potential at all points in space as an expansion in spherical harmonics. (b) Calculate the rectangular components of the electric eld near the center of the sphere. (c) Verify that, in the limit / conducting sphere.
0
, your result is the same as that for the
We will take the origin of coordinates at the center of the sphere, and put the point charge on the z axis at z = +h. Then the problem has azimuthal symmetry. (a) Since there is no free charge within the sphere, D = 0 there. But since the permittivity is uniform within the sphere, we may also write (D/ ) = E = 0 there. This means that polarization charge only exists on the surface of the sphere, so within the sphere the potential satises the normal Laplace equation, whence Al rl Pl (cos ) (r < a). (r, ) =
l
Now, in the region r > a, the potential may be written as the sum of two components 1 and 2 , where 1 comes from the polarization charge on the surface of the sphere, while 2 comes from the external point charge. Since 1 satises the Laplace equation for r > a, we may expand it in Legendre polynomials: 1 (r, ) =
l
Bl r(l+1) Pl (cos )
(r > a).
Putting this all together we may write the potential in the three regions as Al rl Pl (cos ), r<a l r q Bl r(l+1) + Pl (cos ), a<r<d (r, ) = l 4 0 d +1 qdl Bl + r(l+1) Pl (cos ), r > d. 4 0
On the other hand, 2 is just the potential due to a point charge at z = d: q rl Pl (cos ), r<d 4 0 dl+1 2 (r, ) = (9) dl q Pl (cos ), r > d. 4 0 rl+1
The normal boundary condition at r = a is r

0
=
r =a
r =a+
lAl al1 = (l + 1)Bl a(l+2) + Al =

0
lqal1 4 0 dl+1 (10)
(l + 1) q Bl a(2l+1) + l 4 0 dl+1
The tangential boundary condition at r = a is =

r =a
r =a+
Al al = Bl a(l+1) +
q al 4 0 d(l+1) q a2l+1 Bl = Al a2l+1 4 0 dl+1
(11)
Combining (10) and (11), we obtain Al =

0
1 + l+1 l 1 + l+1 l
2l + 1 l 1
q 4 0 dl+1 qa2l+1 4 0 dl+1
Bl =
0
In particular, as /
we have Al 0
as must happen, since the eld within a conducting sphere vanishes; and Bl qa2l+1 . 4 0 dl+1 (12)
With the coecients (12), the potential outside the sphere due to the polarization charge at the sphere boundary is 1 qa 1 (r, ) = 4 0 d a2 d
l
1 Pl (cos ). rl+1
Comparing with (9) we see that this is just the potential of a charge qa/d on the z axis at z = a2 /d. This is just the size and position of the image charge we found in Chapter 2 for a point charge outside a conducting sphere.
(b) Near the origin, we have (r, ) = A1 rP1 (cos ) + A2 r2 P2 (cos ) + = q 4

0
3 0 1 z+ d2 ( + 2 0 ) 2
5 0 d3 (2 + 3 0 )
(z 2 x2 y 2 ) +
so the eld components are Ex = 5 0 4 0 2 +3 0 q 5 0 Ey = 4 0 d2 2 + 3 0 q 3 0 Ez = 2 4 0 d +2 0 d2 q x + d y + d 5 0 + 2 +3 0
z + d
Problem 4.10
Two concentric conducting spheres of inner and outer radii a and b, respectively, carry charges Q. The empty space between the spheres is half-lled by a hemispherical shell of dielectric (of dielectric constant / 0 ), as shown in the gure. (a) Find the electric eld everywhere between the spheres. (b) Calculate the surface-charge distribution on the inner sphere. (c) Calculate the polarization-charge density induced on the surface of the dielectric at r = a. Well orient this problem such that the boundary between the dielectriclled space and the empty space is the xy plane. Then the region occupied by the dielectric is the region a < r < b, 0 < < /2, and the problem has azimuthal symmetry. (a) Since the dielectric has uniform permittivity, all the polarization charge exists on the boundary of the dielectric, so within its body we may take the potential to be a solution of the normal Laplace equation. The potential in the region between the spheres may then be written [Al rl + Bl r(l+1) ]Pl (cos ), 0<< 2 (r, ) = l (l+1) [Cl r + Dl r ]Pl (cos ), << 2 First lets apply the boundary conditions at the interface between the dielectric and free space. That region is described by = /2, a < r < b, and we
must have r which leads to Al Cl Pl (0)rl + Bl Dl Pl (0)rl+1 = 0 (13) (14) =

=/2+ 0
=/2
=/2+
= r
=/2
l [Al Cl ] P (0)rl1 (l + 1) [Bl Dl ] Pl (0)rl+2 = 0.
Since these equations must be satised for all r in the region a < r < b, the coecients of each power of r must vanish identically. In (13), this requirement is automatically satised for l even, since Pl (0) vanishes for even l. Similarly, (14) is automatically satised for l odd. For other cases the vanishing of the coecients must be brought about by taking
0
Al = C l Al = C l
Bl = Dl , Bl = Dl ,
l odd l even.
(15) (16)
Next lets consider the charge at the surface of the inner sphere. There are actually two components of this charge; one component comes from the surface distribution of the free charge +Q that exists on the sphere, and the other component comes from the bound polarization charge on the inner surface of the dielectric
Problem 4.13
Two long, coaxial, cylindrical conducting surfaces of radii a and b are lowered vertically into a liquid dielectric. If the liquid rises an average height h between the electrodes when a potential dierence V is established between them, show that the susceptibility of the liquid is e = (b2 a2 )gh ln(b/a) 2 0V
where is the density of the liquid, g is the acceleration due to gravity, and the susceptibility of air is neglected. First lets work out what happens when a battery of xed voltage V is connected between two coaxial conducting cylinders with simple vacuum between them. To begin, we can use Gauss law to determine the E eld between the
10
cylinders. For our Gaussian pillbox we take a disk of thickness dz and radius r, a < r < b centered on the axis of the cylinders. By symmetry there is no component of E normal to the top or bottom boundary surfaces, and the component normal to the side surfaces (the radial component) is uniform around the disc. Hence E dA = 2 r dzE = E () = q
0
1
0
(2 a dz )
a 0r
where is the surface charge on the inner conductor. This must integrate to give the correct potential dierence between the conductors:
b
V =
E ()d =
a
0
ln
b a
which tells us that, to establish a potential dierence V between the conductors, the battery has to ow enough charge to establish a surface charge of magnitude =
0V a ln(b/a)
(17)
on the cylinder faces (the surface charges are of opposite sign on the two cylinders). It is useful to gure out the energy per unit length stored in the electric eld between the cylinder plates here. This is just Wv = 1 2
0 b a b 0 2
E D d d
= = =
E 2 () d ln(b/a) (18)
a
0
a 2 2
0V 2 ln(b/a)
where the v subscript stands for vacuum, since (18) is the energy per unit length stored in the eld between the cylinders with just vacuum between them. Now suppose we introduce a dielectric material between the cylinders. If the voltage between the cylinders is kept at V , then the E eld must be just the same as it was in the no-dielectric case, because this eld integrated from a to b must still give the same potential dierence. However, in order to establish this same E eld in the presence of the retarding eects of the dielectric, the battery now has to establish a surface charge that is greater that it was before by a factor ( / 0 ). With this greater charge on the electrodes, the D eld will now be bigger by a factor ( / 0 ) than it was in our above calculation. So the
11
energy per unit length stored in the eld between the cylinders increases by a factor ( / 0 1) over the result (18): Wd = (
0)
V 2 . ln(b/a)
On the other hand, to get to this point the battery has had to ow enough charge to increase the surface charges to be of magnitude ( / 0 ) times greater than (17). In doing this the internal energy of the battery decreases by an amount equal to the work it had to do to ow the excess charge, namely Wb = V dQ = V (2 a d ) = (
0)
2V 2 ln(b/a)
(per unit length). The energy lost by the battery is twice that gained by the dielectric, so the system with dielectric between the cylinders has lower overall energy than the system with vacuum between the cylinders by a factor W = (
0)
V 2 ln(b/a)
(19)
(per unit length). Turning now to the situation in this problem, well take the axis of the cylinders as the z axis, so that the surface of the liquid is parallel to the xy plane. Well take the boundary between the liquid and the air above it to be at z = 0. With no potential between the cylinder plates, the liquid between the cylinders is at the same height as the liquid outside. Now suppose a battery of xed potential V is connected between the two cylinder plates. As we showed earlier, the combined system of battery and dielectric can lower its energy by having more of the dielectric rise up between the cylinders. However, at some point the energy win we get from this is balanced by the energy hit we take from the gravitational potential energy of having the excess liquid rise higher between the cylinders. The height at which we no longer gain by having more liquid between the cylinders is the height to which the system will settle. So suppose that, with a battery keeping a voltage V between the electrodes, the liquid between the electrodes rises to a height h above the surface of the liquid outside the electrodes. The decrease in electrostatic energy this aords over the case with just vacuum lling that space is just (19) times the height, i.e. V 2 Ee = h( 0 ) (20) ln(b/a) This must be balanced by the gravitational potential energy Eg of the excess liquid. Eg is easily calculated by noting that the area between the cylinders is (b2 a2 ), so the mass of liquid contained in a height dh between the cylinders is dm = (b2 a2 )dh, and if this mass is at a height h above the liquid surface its excess gravitational energy is dEg = (dm)gh = g(b2 a2 )hdh.
12
Integrating over the excess height of liquid between the cylinders, Eg = g(b2 a2 )
h
h dh =
0
1 g(b2 a2 )h2 . 2
(21)
Comparing (20) to (21), we nd that the gravitational penalty of the excess liquid just counterbalances the electrostatic energy reduction when h= 2( 0 )V 2 g (b2 a2 ) ln(b/a) 2e 0 V 2 = g (b2 a2 ) ln(b/a)
Solving for e , e = gh(b2 a2 ) ln(b/a) . 2 0V 2
So I seem to be o by a factor of 2 somewhere. Actually we should note one detail here. When the surface of the liquid between the cylinders rises, the surface of the liquid outside the cylinders must fall, since the total volume of the liquid is conserved. Hence there are really two other contributions to the energy shift, namely, the change in gravitational and electrostatic energies of the thin layer of liquid outside the cylinders that falls away when the liquid rises between the cylinders. But if the surface area of the vessel containing the liquid is suciently larger than the area between the cylinders, the dierence layer will be thin and its energy shifts negligible.
13
Appendix
Source code for eld line plotting program used in Problem 4.8.
/* * Program to draw field lines for Jackson problem 4.8. * Homer Reid October 2000 */ #include <stdio.h> #include <math.h> #include "/usr2/homer/include/GnuPlot.c" #define EZ 1.0 /* permittivity of free space #define EPS 5.0 /* permittivity of cylinder #define E0 1.0 */ */
/* external field (irrelevant here) */ */ */
#define A 4.0 /* radius of inner cylinder #define B 8.0 /* radius of outer cylinder
#define NUMLINES 25.0 /* number of field lines to draw */ #define NUMPOINTS 250.0 /* no. of pts to plot for each line */ #define DELTAX (4.0 * B) / NUMPOINTS #define DELTAY (4.0 * B) / NUMLINES /* horiz spacing of pts */ /* vert spacing of initial pts */
#define DENOM (B*B*(EPS+EZ)*(EPS+EZ) - A*A*(EPS-EZ)*(EPS-EZ)) /* * Return r component of electric field at position (r,phi). */ double Er(double r, double phi) { double Coeff; if ( r < A ) Coeff=(4.0*EPS*EZ*B*B)/DENOM; else if ( r < B ) Coeff=(2*EPS*B*B/DENOM)*( (EPS+EZ) - (EPS-EZ)*(A*A)/(r*r) ); else Coeff=1.0 - ((B*B - A*A)*(EZ*EZ-EPS*EPS)*(B*B)/(r*r*DENOM)); return Coeff*E0*cos(phi); }
14
/* * Return phi component of electric field at (r,phi). */ double Ephi(double r, double phi) { double Coeff; if ( r < A ) Coeff=(4.0*EPS*EZ*B*B)/DENOM; else if ( r < B ) Coeff=(2*EPS*B*B/DENOM)*( (EPS+EZ) + (EPS-EZ)*(A*A)/(r*r) ); else Coeff=1.0 + ((B*B - A*A)*(EZ*EZ-EPS*EPS)*(B*B)/(r*r*DENOM)); return -Coeff*E0*sin(phi); } void main() { double i,j,r,phi,x,y,dx,dy; double RComp,PhiComp; FILE *g; g=GnuPlot("Field lines"); /* * Send basic GnuPlot configuration commands. */ fprintf(g,"set terminal postscript portrait color\n"); fprintf(g,"set output fig4.1.eps\n"); fprintf(g,"set multiplot \n"); fprintf(g,"set size square\n"); fprintf(g,"set noxtics\n"); fprintf(g,"set noytics\n"); fprintf(g,"set xrange [%g:%g]\n",-2.0*B,2.0*B); fprintf(g,"set yrange [%g:%g]\n",-2.0*B,2.0*B); /* * Draw circles at r=a and r=b. */ fprintf(g,"plot - t , - t with lines, - t with lines\n"); fprintf(g,"e\n"); for(phi=0; phi<=2*M_PI; phi+=(2*M_PI/100)) fprintf(g,"%g %g\n",A*cos(phi),A*sin(phi));
15
fprintf(g,"e\n"); for(phi=0; phi<=2*M_PI; phi+=(2*M_PI/100)) fprintf(g,"%g %g\n",B*cos(phi),B*sin(phi)); fprintf(g,"e\n"); /* * Draw field lines. */ for (i=1.0; i<=NUMLINES; i+=1.0) { /* * Compute starting x and y coordinates and initiate plot. */ x=-2.0*B; y=2.0*B * ((NUMLINES - 2.0*i)/NUMLINES); fprintf(g,"plot - t with lines\n"); /* * Plot NUMPOINTS points for this field line. */ for (j=0.0; j<NUMPOINTS; j+=1.0) { /* * compute polar coordinates of present location */ r=sqrt(x*x + y*y); if (x==0.0) phi=(y>0.0) ? M_PI/2.0 : -M_PI/2.0; else phi=atan(y/x); /* * compute rise and run of electric field */ RComp=Er(r,phi); PhiComp=Ephi(r,phi); dx=cos(phi)*RComp - sin(phi)*PhiComp; dy=sin(phi)*RComp + cos(phi)*PhiComp; /* * bump x coordinate forward a fixed amount, and y * coordinate up or down by an amount depending on * the direction of the electric field at this point */ x+=DELTAX; y+=DELTAX * (dy/dx); fprintf(g,"%g %g\n",x,y);
16
}; fprintf(g,"e\n"); }; printf("Thank you for your support.\n"); }

Homer Reid November 8, 2000

Problem 5.1
Starting with the dierential expression dB = xx 0 I dl 4 |x x |3
for the magnetic induction at the point P with coordinate x produced by an increment of current I dl at x , show explicitly that for a closed loop carrying a current I the magnetic induction at P is B= 0 I 4
where is the solid angle subtended by the loop at the point P . This corresponds to a magnetic scalar potential, M = 0 I /4 . The sign convention for the solid angle is that is positive if the point P views the inner side of the surface spanning the loop, that is, if a unit normal n to the surface is dened by the direction of current ow via the right-hand rule, is positive if n points away from the point P , and negative otherwise. This is the same convention as in Section 1.6 for the electric dipole layer. I like to change the notation slightly: the observation point is r1 , the coordinate of a point on the current loop is r2 , and the displacement vector (pointing to the observation point) is r12 = r1 r2 . The solid angle subtended by the current loop at r1 is given by a surface integral over the loop: cos dA = 2 r12 S 1

Homer Reid February 11, 2001
Problem 5.10
A circular current loop of radius a carrying a current I lies in the x y plane with its center at the origin. (a) Show that the only nonvanishing component of the vector potential is A (, z ) = 0 Ia
dk cos kz I1 (k< )K1 (k> )

0
where < (> ) is the smaller (larger) of a and . (b) Show that an alternative expression for A is A (, z ) = 0 Ia 2
dkek|z| J1 (ka)J1 (k).

0
(c) Write down integral expressions for the components of magnetic induction, using the expressions of parts a and b. Evaluate explicitly the components of B on the z axis by performing the necessary integrations. (a) Translating Jacksons equation (5.33) into cylindrical coordinates, we have J = I (z ) ( a) (1)
Following Jackson, we take the observation point x on the x axis, so its coordinates are (, = 0, z ). Since there is no current in the z direction, and since the 1
current density is cylindrically symmetric, there is no vector potential in the or z directions. In the direction we have A = Ax sin + Ay cos = Ay = 0 4 0 = 4 Jy (x ) dx |x x | J (x ) cos dx |x x |
0 Re 4
J (x )ei dx |x x |
i
0 Re = 4
J (x )e
m=
eim( ) cos[k (z z )]Im (k< )Km (k> ) dk dx
where we substituted in Jacksons equation (3.148). Rearranging the order of integration and remembering that = 0, we have A = 0 Re 2 2
0
m=
J (x )ei(1m) cos[k (z z )]Im (k< )Km (k> )dx dk
If m = 1, the integral yields 2 ; otherwise it vanishes. Thus A = 0

0 0
J (r , z ) cos[k (z z )]I1 (k< )K1 (k> ) dz dr
dk
Substituting (1), we have A = Ia0
cos kz I1 (k< )K1 (k> ) dk.

0
(b) The procedure for obtaining this expression is identical to the one I just went through, but with the expression from Problem 3.16(b) used for the Greens function instead of equation (3.148). (c) Lets suppose that the observation point is in the interior region of the current loop, so < = , > = a. Then B = [ A] = = Ia0
0
A z
k sin kz I1 (k)K1 (ka) dk 1 A A + I1 (k) + kI1 (k) K1 (ka) dk cos kz 0
Bz = [ A]z = = Ia0
As = 0, I1 () 0, I1 ()/ 1/2, and I1 () 1/2, so B ( = 0) = 0 Ia0 k cos kzK1 (ka) dk Bz ( = 0) = 0 Ia0 = sin kzK1 (ka)dk z 0 The integral may be done by parts:
0
1 sin kzK1 (kz ) dk = sin kzK0 (ka) a
+
0
z a
cos kzK0 (ka) dk

0
K0 is nite at zero but sin vanishes there, and sin is nite at innity but K0 vanishes there, so the rst term vanishes. The integral in the second term is Jacksons equation (3.150). Plugging it in to the above, Bz ( = 0) = z I0 2 2 z (z + a2 )1/2 a2 I0 = . 2 (z 2 + a2 )3/2
Problem 5.11
A circular loop of wire carrying a current I is located with its center at the origin of coordinates and the normal to its plane having spherical angles 0 , 0 . There is an applied magnetic eld, Bx = B0 (1 + y ) and By = B0 (1 + x). (a) Calculate the force acting on the loop without making any approximations. Compare your result with the approximate result (5.69). Comment. (b) Calculate the torque in lowest order. Can you deduce anything about the higher order contributions? Do they vanish for the circular loop? What about for other shapes? (a) Basically were dealing with two dierent reference frames here. In the lab frame, R, the magnetic eld exists only in the xy plane, and the normal to the current loop has angles 0 , 0 . We dene the rotated frame R by aligning the z axis with the normal to the current loop, so that in R the current loop exists only in the x y plane, but the magnetic eld now has a z component. The force on the current loop is F= (J B)dV. (2)
Homer Reids Solutions to Jackson Problems: Chapter 5 PSfrag replacements z1 = z z y1 x 0 x1 y x1 x R R1 R1 R 0 y = y1 z1
Figure 1: Successive coordinate transformations in Problem 5.11.
The components of J are easy to express in R , but more complicated in R; the opposite is true for B. There are two ways to do the problem: we can work out the components of J in R and do the integral in R, or we can work out the components of B in R and do the integral in R , in which case we would have to transform the components of the force back to R to get the answer we desire. I think the former approach is easier. To derive the transformation matrix relating the coordinates of a point in R and R , I imagined that the transformation arose from two separate transformations, as depicted in gure (??). The rst transformation is a rotation through 0 around the z axis, which takes us from R to an intermediate frame R1 . Then we rotate through 0 around the y1 axis, which takes us to R . Evidently, the coordinates of a point in the various frames are related by x1 cos 0 sin 0 0 x y1 = sin 0 cos 0 0 y (3) z1 0 0 1 z x1 cos 0 0 sin 0 x y1 y = 0 1 0 (4) z1 sin 0 0 cos 0 z Multiplying matrices, cos 0 cos 0 x y = sin 0 sin 0 cos 0 z cos 0 sin 0 cos 0 sin 0 sin 0 x sin 0 y . 0 z cos 0
(5)
This matrix also gives us the transformation between unit vectors in the two
frames: i cos 0 cos 0 j = sin 0 sin 0 cos 0 k cos 0 sin 0 cos 0 sin 0 sin 0 sin 0 0 cos 0 i j . k (6)
We will also the inverse transformation, i.e. the expressions for coordinates in R in terms of coordinates in R : x cos 0 cos 0 sin 0 sin 0 cos 0 x y = cos 0 sin 0 cos 0 sin 0 sin 0 y . (7) z sin 0 0 cos 0 z
To do the integral in (2) its convenient to parameterize a point on the current loop by an angle reckoned from the x axis in R . If the loop radius is a, then the coordinates of a point on the loop are x = a cos , y = a sin , and the current density/volume element product is J dV = Id l = (Ia d ) = Ia d [ sin i + cos j]
= Ia d ( sin cos 0 cos 0 cos sin 0 ) i + (sin sin 0 + cos cos 0 ) j + (sin sin 0 )k
We also need the components of the B eld at a point on the current loop: B( ) = B0 [1 + y ( )] i + B0 [1 + x( )] = B0 [1 + a (cos cos 0 sin 0 + sin cos 0 )] i + B0 [1 + a (cos cos 0 cos 0 sin sin 0 )] j The components of the cross product are [J B]x dV = Jz By dV [J B]y dV = Jz Bx dV = ( )Ia2 B0 d sin2 sin 0 sin 0 = ( ) + Ia2 B0 d sin2 sin 0 cos 0
[J B]z dV = (Jx By Jy Bx ) dV = ( ) + 0
where we only wrote out terms containing a factor of cos2 or sin2 , since only these terms survive after the integral around the current loop (we grouped all the remaining terms into ( )). In the surviving terms, cos2 and sin2 turn into factors of after the integral around the loop. Then the force components are Fx = Ia2 B0 sin 0 sin 0 Fy = Ia2 B0 sin 0 cos 0 Fz = 0.
To compare this with the rst-order approximate result, note that the magnetic moment has magnitude a2 I and is oriented along the z axis: = a2 I sin 0 cos 0 m = a2 I k i + sin 0 sin 0 j + cos 0 k so B m = B0 (1 + y )mx + B0 (1 + x)my = B0 my i + mx j = Ia2 B0 sin 0 sin 0 i + sin 0 cos 0 j) in exact agreement with the result we calculated so laboriously above.
Problem 5.12
Two concentric circular loops of radii a, b and currents I, I , respectively (b < a), have an angle between their planes. Show that the torque on one of the loops is about the line of intersection of the two planes containing the loops and has the magnitude 0 II b2 N= 2a (n + 1) (n + 3/2) (2 n + 1) ( n + 2)(3/2) n=0
2
b a
2n 1 P2 n+1 (cos ).
The torque on the smaller loop is N= = r Jb (r) Ba (r) dr r Ba (r) Jb (r) r Jb (r) Ba (r) dr.
where Jb is the current density of the smaller loop and Ba is the magnetic eld of the larger loop. But r Jb vanishes, because the current ows in a circle around the originthere is no current owing toward or away from the origin. Thus N = rBr (r)Jb (r)dr (8) where Br is the radial component of the magnetic eld of the larger current loop. As in the last problem, its convenient to dene two reference frames for this situation. Let R be the frame in which the smaller loop (radius b, current I ) lies in the xy plane, and R the frame in which the larger loop lies in the x y plane. We might as well take the line of intersection of the two planes to be the y axis, so y = y . Then the z axis has spherical coordinates ( = , = 0) in
R, and for transforming back and forth between the two frames we may use the transformation matrices we derived in the last problem, with 0 = , 0 = 0. If we choose to evaluate the integral (8) in frame R, the current density is Jb (r) = I (r b) ( /2) sin i + cos j so the components of the torque are Nx = Ib2 Ny = Ib2
0 2
Br (r = b, = /2, ) sin d
0 2
(9) (10)
Br (r = b, = /2, ) cos d
To do the integral in (8), we need an expression for the radial component Br of the eld of the larger loop. Of course, we already have an expression for the eld in R : in that frame the eld is just that of a circular current loop in the x y plane, Jacksons equation (5.48): Br (r , ) = 0 I a 2r
2l+1 (1)l (2l + 1)!! r
l=0
We are interested in evaluating this eld at points along the smaller current loop, and for all such points r = b; then r< = b, r> = a and we have Br (r = b, ) = 0 I 2a
l=0
(1)l (2l + 1)!! 2l l !
b a
2l
P2l+1 (cos ).
(11)
To transform this to frame R, we rst note that, since the origins of R and R coincide, the unit vectors r and r coincide, so Br = Br . Next, (11) expresses the eld in terms of cos , the polar angle in frame R . How do we write this in terms of the angles and in frame R? Well, note that cos = z r x sin + z cos = r r sin cos sin + r cos cos = r = sin sin cos + cos cos
(12)
where in the second line we used the transformation matrix from Problem 5.11 to write down z in terms of x and z . Equation (12) is telling us what our coordinates in R are in terms of our coordinates in R; if a point has angular coordinates , in R, then (12) tells us what angle it has in R . (We could also work out what the azimuthal angle would be, but we dont need to, because (11) doesnt depend on .)
To express the Legendre function in (11) with the argument (12), we may make use of the addition theorem for associated Legendre polynomials: Pl (cos ) = Pl (cos cos + sin sin cos )
l
= Pl (cos )Pl (cos ) + 2

m=1
Plm (cos )Plm (cos ) cos m.
Of course, the smaller loop exists in the xy plane, so for all points on that loop we have = /2, whence
l
Pl (cos ) = Pl (0)Pl (cos ) + 2

m=1
Plm (0)Plm (cos ) cos m.
We may now write down an expression for the radial component of the magnetic eld of the larger loop, evaluated at points on the smaller loop, in terms of the angle that goes from 0 to 2 around that loop: Br () = 0 I 2a
l=0
(1)l (2l + 1)!! 2l l !

2l+1
b a
2l
P2l+1 (0)P2l+1 (cos )
+2
m=1
m m P2 l+1 (0)P2l+1 (cos ) cos m .
This looks ugly, but in fact when we plug it into the integrals (9) and (10) the sin and cos terms beat against the cos m term, integrating to 0 in the former case and m1 in the latter. The torque is Nx = 0 0 II b2 Ny = a
l=0
(1)l (2l + 1)!! 2l l !
b a
2l 1 1 P2 l+1 (0)P2l+1 (cos ).
To nish we just need to rewrite the numerical factor under the sum: (1)l (2l + 1)!! 1 (2l + 1)!! (l + 3/2) P2l+1 (0) = l l 2 l! 2 l! (l + 1)(3/2) (2l + 3 2)(2l + 3 4)(2l + 3 6) (5)(3) (l + 3/2) = l 2 (l + 1) (l + 1)(3/2) (l + 3/2) (l + 3/2 1)(l + 3/2 2) (5/2)(3/2) = (l + 1) (l + 1)(3/2) = (l + 3/2) (l + 1)(3/2)
2
= (l + 1)2
(l + 3/2) (l + 2)(3/2)
So my answer is Ny = 0 II b2 a
(l + 1)2
l=0
(l + 3/2) (l + 2)(3/2)
b a
2l 1 P2 l+1 (cos ).
Evidently Im o by a factor of 1/(l + 1)(2l + 1) under the sum, but I cant nd where. Can anybody help?
Problem 5.13
A sphere of radius a carries a uniform surface-charge distribution . The sphere is rotated about a diameter with constant angular velocity . Find the vector potential and magnetic-ux density both inside and outside the sphere.
Problem 5.14
A long, hollow, right circular cylinder of inner (outer) radius a (b), and of relative permeability r , is placed in a region of initially uniform magnetic-ux density B0 at right angles to the eld. Find the ux density at all points in space, and sketch the logarithm of the ratio of the magnitudes of B on the cylinder axis to B0 as a function of log10 r for a2 /b2 = 0.5, 0.1. Neglect end eects. Well take the cylinder axis as the z axis of our coordinate system, and well take B0 along the x axis: B0 = B0 i. To the extent that we ignore end eects, we may imagine the elds to have no z dependence, so we eectively have a two dimensional problem. There are two distinct current distributions in this problem. The rst is a current distribution Jfree giving rise to the uniform eld B0 far away from the cylinder; this current distribution is only nonvanishing at points outside the cylinder. The second is a current distribution Jbound = M existing only within the cylinder. Since there is no free current within the cylinder or in its inner region, the equations determining H in those regions are B = (H) = 0, H = Jfree = 0.
These imply that, within the cylinder and in its inner region, we may derive H from a scalar potential: H = m , with m satisfying the Laplace equation. In the external region, there is free current, so things are not so simple. To proceed we may separate the H eld in the external region into two components: one that arises from the free current, and one that arises from the bound currents within the cylinder. The former is just (1/0 )B0 and the second is again derivable from a scalar potential satisfying the Laplace equation. So, in the external region, H = (1/0 )B0 m .
10
So our task is to nd expressions for m in the three regions such that the boundary conditions on B and H are satised at the borders of the regions. Writing down the solutions of the 2-D Laplace equation in the three regions, and excluding terms which blow up as 0 or , we have n n=1 An cos n + Bn sin n m (, ) = Actually, we may argue on symmetry grounds that the sin terms must all vanish: otherwise, the elds would take dierent values on the positive and negative y axes, but there is nothing in the problem distinguishing these axes from each other. With this simplication we may write down expressions for the components of the H eld in the three regions: m = nAn n1 cos n, r n=1 m = n Cn n1 En (n+1) cos n, r n=1

r<a a<rb
n=1 n=1
n Cn cos n + Dn sin n + n En cos n + Fn sin n
n Gn cos n + Hn sin n
r<a a<r<b r < b.
Hr =
nGn (n+1) cos n , = (1 / ) B cos + (1 / ) B m 0 0 0 0 r r n=1 m = nAn n1 sin n, n=1 m = n Cn n1 + En (n+1) sin n, n=1

r<a a<r<b r < b.
H =
The boundary conditions at r = b are that H and H be continuous, where = 0 outside the cylinder and r 0 inside. With the above expressions for the components of H, we have 1 B0 cos + nGn b(n+1) cos n = r n Cn bn1 En b(n+1) cos n 0 n=1 n=1 1 B0 sin + nGn b(n+1) sin n = n Cn bn1 + En b(n+1) sin n. 0 n=1 n=1 We may multiply both sides of these by cos n and sin n and integrate from

(1 / ) B = (1 / ) B sin + nGn (n+1) sin n , 0 0 m 0 0 n=1
11
0 to 2 to nd 1 B0 + G1 b2 = r C1 + r E1 b2 0 1 B0 + G1 b2 = C1 + E1 b2 0 Gn b(n+1) = r Cn bn1 En b(n1) , n=1 (13) (14) (15) n=1 (16)
Gn b(n+1) = Cn bn1 + En b(n+1) , Similarly, at r = a we obtain A1 = r C1 r E1 a2 An a

n1
(17)
(n+1)
= r Cn a
n1
En a
n=1 (18) n = 1. (19)
A1 = C1 + E1 a2 An an1 = Cn an1 + En a(n+1) ,
For n = 1, the only solution turns out to be An = Cn = En = Gn = 0. For n = 1, multiplying (15) by r and adding and subtracting with (13) yields 2r C1 = (r + 1) 2r E1 = (1 r ) B0 + (r 1)G1 b2 0 (20) (21)
B0 2 b + (r + 1)G1 . 0
On the other hand, multiplying (18) by r and adding and subtracting with (17) yields 2r C1 = (r + 1)A1 2r E1 = (r 1)a A1 . Equating (20) with (22), we nd A1 = B0 (r 1) + G1 b2 0 (r + 1)
2
(22) (23)
while equating (21) with (23) yields A1 = B0 0 b2 a2 + (r + 1) G1 a2 (r 1) B0 0
and now equating these two equations gives G1 = 1 a b

2 2 (2 r 1)b 2 2 (r + 1) b (r 1)2 a2
b2 .
frag replacements
12
-0.5
(a/b) = 0.5 (a/b) = 0.1
-1
-1.5
log10 r
-2
-2.5
-3
-3.5
-4
-4.5 0 1 2
log10 r
Figure 2: Damping of eld inside cylindrical cylinder of permeability r .
The other coecients may be worked out from this one: A1 = 4r b2 B0 2 2 2 2 (r + 1) b (r 1) a 0 B0 2(r + 1)b2 C1 = 2 2 2 2 (r + 1) b (r 1) a 0 2(r 1)b2 B0 2 E1 = a . 2 2 2 2 (r + 1) b (r 1) a 0
The H eld is H= 4r b2 B0 i, (r + (r 1)2 a2 0 2b2 B0 a 2 a (r + 1) + (r 1) = i 2(r 1) 2 2 2 2 (r + 1) b (r 1) a 0 r r 2 2 2 2 (b a )(r 1) b B0 B0 , i+ i + 2 sin = (r + 1)2 b2 (r 1)2 a2 0 r2 1)2 b2 4r 2 . (r 1)2 a b2 r<a
2
cos r , a<r b.
The ratio r of the eld within the cylinder to the external eld is r= (r + 1)2
This relationship is graphed in Figure
13
Problem 5.16
A circular loop of wire of radius a and negligible thickness carries a current I . The loop is centered in a spherical cavity of radius b > a in a large block of soft iron. Assume that the relative permeability of the iron is eectively innite and that of the medium in the cavity, unity. (a) In the approximation of b a, show that the magnetic eld at the center of the loop is augmented by a factor (1 + a3 /2b3 ) by the presence of the iron. (b) What is the radius of the image current loop (carrying the same current) that simulates the eect of the iron for r < b? (a) There are two distinct current distributions in this problem: the free current density J1 owing in the loop, and the bound current density J2 owing in the iron. These give rise to two elds B1 and B2 , which must be summed at each point in space to get the observed eld. B1 is just the eld of a planar current loop, which Jackson has already worked out for us in his section 5.5: (1)n (2n + 1)!! r 2n 0 I P2n+1 (cos ), r < a 2a n=0 2n n! a (24) B1r = 0 Ia2 (1)n (2n + 1)!! a 2n P2n+1 (cos ), r > a. 2r3 2n n! r n=0 0 I 4a (1)n (2n 1)!! 2n1 n! n=0 (1)n (2n + 1)!! 2n (n + 1)! n=0

r a a r
2n
1 P2 n+1 (cos ), r < a
B1 =
On the other hand, since J2 vanishes for r < b, the eld B2 to which it gives rise has no divergence or curl in that region, which means that throughout the region it may be derived from a scalar potential satisfying the Laplace equation:
0 Ia2 4r3
2n
(25)
1 P2 n+1 (cos ),
r > a.
B2 = m =
An rn Pn (cos )
n=0
B2r = B2 =
nAn rn1 Pn (cos )

n=1 1 An rn1 Pn (cos ) n=1
(26) (27)
14
Since the iron lling the space r > b is assumed to have innite permeability, the H eld (and hence the B eld, since B = H for r < b) must be strictly radial at the boundary r = b. The An coecients are thus determined by the requirement that (27) and (25) sum to zero at r = b:
1 An bn1 Pn (cos ) = n=1
0 Ia2 4b3
(1)n (2n + 1)!! 2n (n + 1)! n=0
a b
2n
1 P2 n+1 (cos ).
The orthogonality of the associated Legendre polynomials requires that each term in the sum cancel individually, whence A2n = 0 A2n+1 = 0 Ia2 (1)n (2n + 1)!! 4b3 2n (n + 1)! a b2
2n
Then the eld of the bound current in the iron is determined everywhere in the region r < b: B2r = B2 0 Ia2 4b3 (1)n (2n + 1)(2n + 1)!! 2n (n + 1)! n=0 (1)n (2n + 1)!! 2n (n + 1)! n=0

ar b2
2n
P2n+1 (cos )
(28) (29)
0 Ia2 = 4b3
ar b2
2n
1 P2 n+1 (cos ).
As r 0, B2 0 and B2r 0 Ia2 /4b3 , while B1r 0 I/2a, so the total eld at r = 0 is 0 Ia2 0 I a3 0 I + = 1 + . Br (r = 0) = B1r (r = 0) + B2r (r = 0) = 2a 4b3 2a 2b3 (b) The B2 eld may be attributed to an image current ring outside r = b if, for suitable redenitions of I and a, the expressions (28) and (29) can be made to look like the r < a versions of (24) and (25).
Problem 5.18
A circular loop of wire having a radius a and carrying a current I is located in vacuum with its center a distance d away from a semi-innite slab of permeability . Find the force acting on the loop when (a) the plane of the loop is parallel to the face of the slab, (b) the plane of the loop is perpendicular to the face of the slab. (c) Determine the limiting form of your answer to parts a and b when d Can you obtain these limiting values in some simple and direct way? (a) Well take the loop to be at z = +d, and the slab of permeability to occupy the space z < 0, so that the boundary surface is z = 0. a.
15
In the region z < 0, there is no free current, so H = 0 everywhere; thus H may be obtained from a scalar potential, H = m , and since H = 0 as well we have 2 m = 0. The azimuthally symmetric solution of the Laplace equation in cylindrical coordinates that remains nite as z is
m (z < 0) =
0
dk A(k )ekz J0 (k),
(30)
and from this we obtain H (z < 0) = =

0
m =
dk kA(k )ekz J0 (k) (31) (32)
dk kA(k )ekz J1 (k) m = z
Hz (z < 0) =
dk kA(k )ekz J0 (k).

0
On the other hand, for z > 0 we may decompose the H eld into two components: one component H1 arising from the current loop, and a second component H2 arising from the bound currents running in the slab. H1 is just given by the curl of the vector potential we worked out in Problem 5.10: 0 Ia dk ek(zd) J1 (ka)J1 (k), 1 2 0 A, A = A , A = H1 = 0 Ia 0 dk ek(dz) J1 (ka)J1 (k), 2 0 so H1 = 1 A 0 z Ia 2 = Ia 2 Ia 2 = Ia 2
z>d z < d.
dk kek(zd) J1 (ka)J1 (k),

0
z>d z < d. (33)
dk kek(dz) J1 (ka)J1 (k),

0
H1z
1 1 = (A ) 0
1 J1 (k) J0 (k) z>d k 0 1 J1 (k) J0 (k) , z < d. dk kek(dz) J1 (ka) k 0 (34) dk kek(zd) J1 (ka)
In the last two equations we may use Jacksons identity (3.87), 1 1 J1 (k) = [J0 (k) + J2 (k)] k 2
16
Since the H2 eld arises entirely from bound currents, it may also be derived from a scalar potential m satisfying the Laplace equation. The azimuthally symmetric solution of the Laplace equation in cylindrical coordinates that remains nite for all and as z + is
to rewrite H1z as Ia 4 H1z = Ia 4
0 0
dk ek(zd) J1 (ka) [J2 (k) J0 (k)] , z > d dk ek(dz) J1 (ka) [J2 (k) J0 (k)] , z < d.
(35)
m (z > 0) =
0
dk B (k )ekz J0 (k)
and the components of H2 are
H2r (z > 0) = H2z (z > 0) =

0
dk kB (k )ekz J1 (k)
0
(36) (37)
dk kB (k )ekz J0 (k).
The required forms of the functions A(k ) and B (k ) are determined by the boundary conditions on H at the medium boundary, z = 0: H (z = 0 ) = H (z = 0+ ) H (z = 0 ) = 0 H (z = 0+ ).
Equating (32) with the sum of (??) and (??), we have
dk kA(k )J0 (k) =

0
0 Ia 2
dk kekd J1 (ka) (J2 (k) J0 (k)) +
dk kB (k )J0 (k)
0

Homer Reid April 20, 2001
Problem 5.19
A magnetically hard material is in the shape of a right circular cylinder of length L and radius a. The cylinder has a permanent magnetization M0 , uniform throughout its volume and parallel to its axis. (a) Determing the magnetic eld H and magnetic induction B at all points on the axis of the cylinder, both inside and outside. (b) Plot the ratios B/0 M0 and H/M0 at all points on the axis of the cylinder, both inside and outside. There is no free current in this problem, so H(, z ) may be derived from a scalar potential m (, z ) satisfying the Laplace equation. Dividing space into three regions
dk A(k )ekz J0 (k),

0
z > L/2 L/2 < z < L/2 z < L/2.
m =
dk B (k )ekz + C (k )ekz J0 (k),

0
dk D(k )ekz J0 (k),

0
The tangential boundary condition at z = +L/2 is m
=
z= L 2+
z= L 2
dk kA(k )ekL/2 J1 (k) =

0 0
dk k B (k )ekL/2 + C (k )ekL/2 J1 (k) (1)
This must hold for all . Multiplying both sides by J1 (k ), integrating from = 0 to = , and using the identity
d Jn (k)Jn (k ) =
0
1 (k k ) k
(2)
we obtain from (1) the relation A(k ) = B (k )ekL + C (k ). The perpendicular boundary condition at z = +L/2 is Bz (z = L/2+) = Bz (L/2) or 0 Hz (z = L/2+) = 0 Hz (z = L/2) + Mz (z = L/2) m z =
z= L 2+
(3)
m z
+ M ()
z= L 2
dk kA(k )ekL/2 J0 (k) =

0 0
dk k B (k )ekL/2 + C (k )ekL/2 J0 (k) + M () (4)
where M () = M1 , 0, <a > a.
Now we multiply both sides of (4) by J0 (k ) and integrate from = 0 to = to obtain A(k ) = B (k )ekL + C (k ) + M1 ekL/2 = B (k )ekL + C (k ) + (k ) where we dened (k ) = M1 ekL/2
0 a a
J0 (k)d
0
(5)
J0 (k)d =
aM1 kL/2 e J1 (ka). k
The solution of eqs. (3) and (5) is B (k ) = 1 kL e (k ) 2 1 C (k ) = A(k ) (k ). 2 (6)
From the boundary conditions at z = L/2 we may similarly obtain the relations B (k ) + C (k )ekL = D(k ) B (k ) C (k )ekL = D(k ) (k ) which may be solved to yield 1 B (k ) = D(k ) (k ) 2 Comparing (6) and (7) we nd A(k ) = D(k ) = kL M1 a cosh J1 (ka) k 2 M1 a kL/2 B (k ) = C (k ) = e J1 (ka). 2k C (k ) = 1 kL e (k ). 2 (7)
Then the components of the H eld are kL kz e J1 (ka)J1 (k), M1 a dk cosh 2 0 dk ekL/2 cosh(kz )J1 (ka)J1 (k), H = M 1 a 0 kL kz dk cosh e J1 (ka)J1 (k), M1 a 2 0
z > L/2 L/2 < z < L/2 z < L/2
M1 a
0
dk cosh
kL kz e J1 (ka)J0 (k), 2
z > L/2 L/2 < z < L/2 z < L/2.
Hz =
M1 a
dk ekL/2 sinh(kz )J1 (ka)J0 (k),

0
M1 a
dk cosh
0
kL kz e J1 (ka)J0 (k), 2
Problem 5.23
A right circular cylinder of length L and radius a has a uniform lengthwise magnetization M . (a) Show that, when it is placed with its at end against an innitely permeable plane surface, it adheres with a force F = 20 aLM 2 where k= K (k ) E (k ) K (k1 ) E (k1 ) k k1 k1 = a. a . + L2
2a , 4a2 + L2
a2
(b) Find the limiting form of the force if L
Well dene our coordinate system so that the z axis is the cylinder axis, and well take the surface of the permeable medium at z = 0. Our general strategy for this problem will be as follows. First, well nd the magnetic eld H0 that exists in all space when the cylinder is pressed up at against the innitely permeable medium. Then well calculate the shift dE in the energy of the magnetic eld incurred by moving the cylinder up a small distance dz o the surface of the medium. The force on the cylinder is then readily calculated as F = dE/dz . To calculate the energy shift incurred by moving the cylinder a distance dz away from the permeable medium, we wont have to go through and completely recalculate the elds and their energy in the new conguration. Instead, we can use the following little trick. When we move the cylinder up a distance dz , two things happen. First a gap of height dz opens between the surface and the face of the cylinder, where previously there had been a xed magnetization M, but now there is just free space. Second, between L and L + dz there is now a xed magnetization M where previously there was none. Moving the cylinder of xed M up a distance dz is thus formally equivalent to keeping the cylinder put and instead introducing a cylinder of the opposite magnetization M between 0 and dz , while also introducing a cylinder of magnetization +M between L and L + dz . The increase in eld energy in this latter case is fairly easily calculated by taking the integral of 0 Mc H0 over the regions in which the xed magnetization changes. So the rst task is to nd the eld that exists when the cylinder is pressed at against the surface. Since there are no free currents in the problem, we may derive H from a scalar potential satisfying the Laplace equation. To begin we write down the general solutions of the Laplace equation in cylindrical coordinates, observing rst that by symmetry we can only keep terms with no
azimuthal angle dependence: dk A(k )ekz J0 (k), 0 (m) = dk [B (k )ekz + C (k )ekz ]J0 (k), 0 dk D(k )e+kz J0 (k),
0
z>L 0<z<L z < 0. (8)
The boundary conditions at z = 0 are that H and Bz be continuous. Assuming rst of all that the medium existing in the region below z = 0 has nite permeability , the tangential boundary condition is m
0
=
z =0
m
0
z =0+
dk k D(k )J1 (k) =
dk k [B (k ) + C (k )]J1 (k).
(9)
Multiplying (9) by J1 (k ), integrating from 0 to , and using the identity (2), we nd D(k ) = B (k ) + C (k ). (10) The normal boundary condition at z = 0 is of a mixed type. Below the line we have simply Bz = Hz . Above the line we may write Bz = 0 [Hz + M ()], where M () represents the xed magnetic polarization of the cylinder: M () = M, 0, <a > a. (11)
The normal boundary condition at z = 0 is then 0

0
m z
z =0
= 0
0
m z
+ 0 M ()
z =0+
dk k D(k )J0 (k) =
dk k [B (k ) C (k )]J0 (k) + M ()
Now multiplying by J0 (k ), integrating from = 0 to , and using (2) yields D(k ) = B (k ) + C (k ) 0 Using (11), the integral on the RHS is
a
M ()J0 (k) d.
0
(12)
M
0
J0 (k) d =
Ma J1 (ka) (k ) k
where we dened a convenient shorthand. Then (12) is D(k ) = B (k ) + C (k ) (k ). 0
Now taking , we see that, to keep the B and C coecients from blowing up, we must have D 0. Then equation (??) tells us that B (k ) = C (k ), so the middle entry in (8) may be rewritten:
m (z, ) =
0
dk (k ) sinh(kz )J0 (k),
(0 < z < L).
The boundary conditions at z = L are m m z =

z =L+
z =L
z =L+
m z
+ M ()
z =L
with M () dened as above. Working through the same procedure as above yields the conditions A(k )ekL = (k ) sinh(kL) A(k )ekL = (k ) cosh(kL) + (k ) with (k ) dened as above. The solution is (k ) = (k )e+kL
A(k ) = (k ) sinh(kL). Plugging these back into (8) and dierentiating, we nd for the z component of the H eld M a dk ekz cosh(kL)J0 (k)J1 (ka), z>L 0 Hz (, z ) = (13) M a dk ekL cosh(kz )J0 (k)J1 (ka), 0 < z < L.
0
Now that we know the eld, we want to nd the change in energy density incurred by putting into this eld a short cylinder (radius a, height dz ) of between z = 0 and z = dz , and another cylinder of the magnetization M k between z = L and z = L + dz . The same size but with magnetization +M k change in eld energy is just the integral of 0 M H over the volume in which the magnetization density has changed:
dz a L+dz a
dU = 20 M = 20 M dz
Hz (z, ) d dz + 20 M
0 a 0 0 a L 0
Hz (z, ) d dz (14)
Hz (L, ) d
Hz (0, ) d
0
where in the last step we assumed that Hz remains essentially constant over a distance dz in the z direction, and may thus be taken out of the integral.
Inserting (13) into (), and exchanging the order of integration, we rst do the integral: a a J0 (k)d = J1 (ka). k 0 Then () becomes

Homer Reid March 28, 2002
Problem 6.2
The charge and current densities for a single point charge q can be written formally as (x , t ) = q [x r(t )]; J(x , t ) = q v(t ) [x r(t )] where r(t ) is the charges position at time t and v(t ) is its velocity. In evaluating expressions involving the retarded time, one must put t = tret = t R(t )/c, where R = x r(t ). (a) As a preliminary to deriving the Heaviside-Feynman expressions for the electric and magnetic elds of a point charge, show that d3 x [x r(tret )] = 1
/c. Note that is evaluated at the retarded time. where = 1 v R (b) Starting with the Jemenko generalizations of the Coulomb and Biot-Savart laws, use the expressions for the charge and current densities for a point charge and the result of part a to obtain the Heaviside-Feynman expressions for the electric and magnetic elds of a point charge, E= and B= 0 q 4 q 4 R R2 + ret 1 R c t R 1 v c2 t R ret
ret
vR 2 R
+ ret
1 vR c t R
ret
(c) In our notation Feynmans expression for the electric eld is E= q 4 R R2 + ret [R]ret R c t R2 + ret 1 2 [R]ret c2 t2
while Heavisides expression for the magnetic eld is B= 0 q 4 vR 2 R2 + ret 1 vR c[R]ret t . ret
Show the equivalence of the two sets of expressions for the elds.
(a) Lets rst assume that the charge is traveling along the z axis, so that its position is given by . r(t) = (z0 + vz t)k
The retarded time tret (t, z ) at a given point z on the z axis is tret (t, z ) = t so Hence (x r[tret (x, t)]) = (x) (y ) {z [z0 + vz tret (t, z )]} z = (x) (y ) z [z0 + vz (t )] c vz = (x) (y ) z z0 vz t + z )] c vz z (z0 + vz t) = (x) (y ) 1 + c By the properties of the function we may write this as = 1 1 + vz /c (x) (y ) z z0 + v z t 1 + vz /c . z c
. r[tret (t, z )] = (z0 + vz tret (t, z ))k
The function is singling out the point in space from which originates the electromagnetic disturbance we feel at the origin at time t. Lets think about whats going on here in two limiting cases. First, as vz 0, the z delta function becomes (z (z0 + vz t)). This means that the source point for the eld we feel at the origin at time t is just z = z0 vz t, which is of course just the instantaneous location of the source particle at time t. In other words, the electromagnetic disturbance left behind by the particle at time t reaches the origin so quickly that the particle hasnt had time to move on. The electromagnetic disturbance seems to be coming from the instantaneous location of the particle itself. In the opposite limit vz c, the z delta function becomes (z (z0 vz t)/2). This says that the point from which we feel an electromagnetic disturbance at time t is half as far from the origin as the particle itself is at time t. This again makes sense. At each point in the particles motion, the electromagnetic disturbance it causes begins propagating toward the origin, while meanwhile the particle continues propagating away from the origin at the same speed. Hence when the electromagnetic disturbance has reached the origin, the particle has traveled as far as the electromagnetic disturbance did, but in the opposite direction, so it is now twice as far from the origin as it was when the disturbance we are just now feeling was generated.
Problem 6.5
A localized electric charge distribution produces an electrostatic eld, E = . Into this eld is placed a small localized time-independent current density J(x), which generates a magnetic eld H. (a) Show that the momentum of these electromagnetic elds, (6.117), can be transformed to 1 J d3 x Peld = 2 c provided the product H falls of rapidly enough at large distances. How rapidly is rapidly enough? (b) Assuming that the current distribution is localized to a region small compared to the scale of variation of the electric eld, expand the electrostatic potential in a Taylor series and show that Peld = 1 E(0) m, c2
where E(0) is the electric eld at the current distribution and m is the magnetic moment, (5.54), caused by the current. (c) Suupose the current distribution is placed instead in a uniform electric eld E0 (lling all space). Show that, no matter how complicated is the localized J, the result in part a is augmented by a surface integral contribution from innity equal to minus one-third of the result of part b, yielding Peld = 2 E0 m. 3c2
Compare this result with that obtained by working directly with (6.117) and the considerations at the end of Section 5.6.
(a) From the denition of electromagnetic eld momentum we have c2 Peld = = E H dV () H dV.
Focusing for now on the z component, we have c 2 Pz = ( Hy Hx ) dx dy dz x y (1)
Lets take our volume of integration to be a cube of side L, which we will eventually take to innity. Integrating the rst term by parts with respect to x, we have
L L L L L L
Hy dx x
dy dz =
L L
Hy
x=L x=L
Hy dx x
dy dz.
Similarly integrating the second term in (1) by parts with respect to y , we may write (1) as c 2 Pz = =
L L L L L
Hy
L L
x=L x=L x=L x=L
dy dz +
L L L L
Hx Hx
L L
y =L y =L y =L y =L
dx dz + dx dz +
( H)z dV Jz dV
Hy
L
dy dz +
where in going to the last line we used H = J since there is no timedependent E eld. This equation is just the z component of c2 P = H dA + J dV. (2)
If we now take L , the rst integral (which describes surface eects) vanishes providing the product (x)H(x) vanishes more quickly (i.e. like a higher power of x) than x2 . Then we are left with just the second term: c2 P = (b) We have (x) = (0) + x (0) + 1 2 xi xj 2 + xi xj J dV. (3)
We may arbitrarily choose (0) = 0. Also, we are told that the electric eld doesnt vary much in the region of nonvanishing J, in which case we may ignore the second derivatives of , to obtain (x) x (0) = x E(0). Plugging into (3), c2 P = We have x E(0) J = E(0) J x x E(0) J E(0) J x = E(0) x J E(0) J x x E(0) J dV. (4)
where in the rst line we added and subtracted a term, and in the second used the BAC-CAB identity of vector analysis. With this, (4) becomes c2 P = E(0) x + J dV E(0) J] x dV
= 2E(0) m
E(0) J] x dV
where in the rst term we have identied the denition of the dipole moment m. Evidently to get this to match up with what Jackson has we need to argue that second term is exactly half the rst, but I cant see how to do this for arbitrary J. Can anybody help? (c) From (2) we have c2 P = H dA + J dV.
The second term is just equal to (E m)/c2 , as computed in part b. For the rst term,
Problem 6.13
A parallel plate capacitor is formed of two at rectangular perfectly conducting sheets of dimensions a and b separated by a distance d small compared to a or b. Current is fed in and taken out uniformly along adjacent edges of length b. With the input current and voltage dened at this end of the capacitor, calculate the input impedance or admittance using the eld concepts of Section 6.9. (a) Calculate the electric and magnetic elds in the capacitor correct to second order in powers of the frequency, but neglecting fringing elds. (b) Show that the expansion of the reactance (6.140) in powers of the frequency to an appropriate order is the same as that obtained for a lumped circuit consisting of a capacitance C = 0 ab/d in series with an inductance L = 0 ad/3b.
(a) Well suppose the plates are oriented parallel to the xy plane, with the lower plate at z = 0 and the upper plate at z = d. Well take the edges of side a parallel to the x axis, and the edges of side b parallel to the y axis. Then the boundary condition on the current density is J(0, y, 0) = J(0, y, d) = J0 j for 0 < y < b.
With neglect of fringing elds, the electric eld between the plates exists only in the z direction, while the magnetic eld exists only in the x direction. We assume harmonic time dependence and write E(y ) = E (y )eit k B(y ) = B (y )eit x ; (5)
then time dierentiation becomes multiplication by i . The Maxwell equations are then E=0 B=0 B t 1 E B= 2 c t E= E z B x E y B y =0 =0 (6) = +iB =+ i E. c2
We postulate an expansion in powers of for E and B : E (y ) = E0 (y ) + E1 (y ) + 2 E2 (y ) + (7)
B (y ) = B0 (y ) + B1 (y ) + 2 B2 (y ) + Then the curl equations in (6) become E0 + E1 + 2 E2 + = i B0 + B1 + 2 B2 + y i B0 + B1 + 2 B2 + = 2 E0 + E1 + 2 E2 + y c
Equating equal powers of in these equations, we obtain E0 y B0 y E1 y B1 y E2 y B2 y E3 y B3 y E4 y B4 y =0 =0 = iB0 = i i i E0 = 2 2 c c = iB1 = 2 y c i = 2 E1 = 2 y c c i = iB2 = 2 y 2 2c i i = 2 E2 = 4 y 2 c 2c 3 = iB3 = y 24c4 i = 2 E3 = + 4 y 3 c 6c = E0 = B0 = E1 = iy B1 = E2 B2 E3 B3 E4 B4 i y c2 = 2 y2 2c = 2 y2 2c i = 2 y3 6c i = 4 y3 6c i 4 = y 24c2 4 = y 24c4
and so on. Plugging into (7), we obtain (ky )4 (ky )3 (ky )2 + + ) + ic(ky + ) 2 24 6 = cos ky + ic sin ky i B (y ) = cos ky + sin ky c E (y ) = (1 (8) (9) (10)
where k = /c, and where we simply wrote down what we guessed to be the sums of the full innite series from their rst few terms. To complete the problem we need to determine the constants and , for which purpose we appeal to the boundary conditions on the plates. We know that the discontinuities in the E and B eld are proportional to the surface charge and current densities on the plates. Since these conditions only give information on the dierences between the elds outside and between the plates, we ostensibly have to know what the elds are outside to get what they are inside. But for the purposes of this problem well just assume there are no elds outside, so the charge and current densities on the plates determine the elds inside. I know this is correct in the low-frequency limit, and in the highfrequency limit Im not yet sure how to compute the radiation elds in the region outside the plates, so I will ignore them.
The boundary conditions are Ez =

0
Bx = 0 Ky where and Ky are the charge density and y component of the surface current density on the top plate (assumed to be identical but with opposite sign on the bottom plate). Plugging in the solutions (9) and (??), we have = 0 ( cos ky + ic sin ky ) i 1 Ky = ( cos ky + sin ky ) 0 c
(11)
As a sanity check, we can verify the continuity relation between charge and current on the plates: Ky = = +i y t Plugging in (11), the left and right sides of this are 1 ik (k sin ky + cos ky ) 0 c RHS = i 0( cos ky + ic sin ky ) LHS = and the two are evidently equal. The forcing function in this problem is the surface current density specied on the edges of length b. If the total current fed into the y = 0 edge of the top plate is I (t) = I0 cos t (with an opposite current taken out of the y = 0 edge of the bottom plate) then the surface current boundary conditions are I0 cos t b Ky (y = a) = 0 Ky (y = 0) = Comparing with (11), we see that these boundary conditions we have to take 0 I 0 cos t b i0 I0 c cos t cot ka = b =
10
Plugging into (9) and (10), i0 I0 c cos t [cot ka cos ky + sin ky ] b i0 I0 c 1 = [cos ka cos ky + sin ka sin ky ] cos t b sin ka i0 I0 c cos[k (y a)] = cos t b sin ka 0 I 0 cos t [ cos ky + cot ka sin ky ] Bz = b 0 I 0 1 = cos t [ sin ka cos ky + cos ka sin ky ] b sin ka sin[k (y a)] 0 I 0 cos t = b sin ka Ez =
Problem 6.14
An ideal circular parallel plate capacitor of radius a and plate separation d a is connected to a current source by axial leads, as shown in the sketch. The current in the wire is I (t) = I0 cos t. (a) Calculate the electric and magnetic elds between the plates to second order in powers of the frequency (or wave number), neglecting the eects of fringing elds. (b) Calculate the volume integrals of we and wm that enter the denition of the reactance X, (6.140), to second order in . Show that in terms of the input current Ii , dened by Ii = iQ, where Q is the total charge on one plate, these energies are we d 3 x = 1 |Ii |2 d , 4 0 2 a2 wm d 3 x = 0 |Ii |2 d 4 8 1+ 2 a2 12c2
(c) Show that the equivalent series circuit has C 0 a2 /d, L 0 d/8 , and that an estimate for the resonant frequency of the system is res = 2 2c/a. Compare with the rst root of J0 (x).
(a) We work in cylindrical coordinates and assume harmonic time dependence ( eit ) for all quantities; then time dierentiation is replaced by multiplication by i. If we neglect the eects of fringing elds, everything is symmetric in , and the electric eld between the plates is entirely in the z direction, while
11
the magnetic eld is entirely in the direction: E(x, t) = E (r, z )eit z B = B (r, z )eit . (12)
The Maxwell equations for the elds between the plates are E=0 z B=0 B=0 (13) E B = iB E= t r 1 i 1 E (rB ) = 2 E. B= 2 c t r r c To proceed, lets propose an expansion of the elds in powers of the frequency: E=0 E (r) = E0 (r, z ) + E1 (r, z ) + 2 E2 (r, z ) +
2
(14) (15)
B (r) = B0 (r, z ) + B1 (r, z ) + B2 (r, z ) + Then the curl equations in (13) become
E0 + E1 + 2 E2 = i B0 + B1 + 2 B2 r i 1 rB0 + rB1 + 2 rB2 = 2 E0 + E1 + 2 E2 r r c Now we just have to go through and equate like powers of in these equations. For n = 0, we have E0 =0 r for some constant 1 , and E 0 = 1 (16)
1 (rB0 ) = 0 B0 = . (17) r r r But for nonzero this blows up at the origin. Hence we must take = 0, so B0 = 0. 2 For n = 1, we have E1 = iB0 = 0 r for some constant 2 , and i i1 1 (rB1 ) = 2 E0 = 2 r r c c Continuing, E2 1 = iB1 = 2 r r 2c 1 i i2 (rB2 ) = 2 E1 = 2 r r c c E 1 = 2 (18)
B1 =
i1 r. 2c2
(19)
E2 =
1 2 r 4c2 i2 B2 = 2 r 2c
(20) (21)
12
2 E3 = iB2 = 2 r r 2c 1 i i1 (rB3 ) = 2 E2 = 4 r2 r r c 4c
E3 =
2 2 r 4c2 i1 3 r B3 = 16c4
(22) (23)
Evidently E2n and E2n+1 have the same form but dier by the presence of 1 or 2 , as is true for B2n1 and B2n . Plugging in equations (16) through (23) into (14) and (15), we obtain 4 4 2 2 r + r + 2 4c 64c4 (kr)2 (kr)4 = (1 + 2 ) 1 + + 4 64 i (kr)2 kr + B (r) = (1 + 2 ) 1 c 2 8 E (r) = (1 + 2 ) 1 These look the rst few terms in certain Bessel functions: E (r) = (1 + 2 )J0 (kr) J0 (kr) i B (r) = J1 (kr) c where we can dene the constant = (1 + 2 ) since were dealing with a xed frequency. Inserting into (12) we obtain E(r, t) = J0 (kr)eit k i B(r, t) = J1 (kr)eit . c (24)
To work out the value of , we need to apply the boundary conditions at the capacitor plates. An easy way to do this is to consider what happens as 0. In that limit there is no magnetic eld, and the electric eld between the plates is just Ez (t) = 2 (t)/ 0 , where (t) is the instantaneous value of the surface charge induced on each plate (positive on the top plate, negative on the bottom). Now, the total charge on the top plate is just the integral of the current owing onto that plate: I0 sin t q = I (t) dt = and the surface charge is this divided by the plate area (since we are assuming a low frequency, any charge that ows onto the plate instantaneously equilibrates with the rest of the charge on the plate, yielding a constant surface charge density): I0 (t) = sin t. a2 Hence the electric eld in the low frequency limit is Ez ( 0) = 2I0 a2 sin .
0
13
Comparing this with(24) in the limit k 0, we obtain = Hence E(r, t) = 2I0 a2 J0 (kr) sin t k B(r, t) = 20 I0 c J1 (kr) cos t . a2 (25) 2I0 i . a2 0
(b) The average energy densities are we = wm =

0
E2 =
2 I0 2 (a )2
1 kr 2
kr + 4
2
1 2 B = 42 0
2 2 0 I 0 c 2 (a )2
(kr)2 + 8
We only have to keep the rst terms in the parentheses to get the energy right to second order in : Ue
2 I0 2 (a )2 2 I0 d = 2 a 2 0 2 2 0 I 0 c = 2 (a )2 a
(2d)(r dr)
0 0
Um
(2d)(rdr)
0
kr (kr)3 + Um 2 8
2 2 0 I 0 c 2 (a )2
(2d)(rdr)
0
kr (kr)3 + 2 8

Homer Reid May 24, 2002
Chapter 8: Waveguide Derivations

Before starting the problems, I thought it would be useful to run through my own derivations of some of the formulas from this chapter.
Waveguides and cavities: basic pedagogy

The unifying feature of waveguide and cavity problems is that we single out one spatial coordinate and announce from the start that the elds will have sinusoidal dependence on that coordinate. Taking the special coordinate to be z , this means that all components of all elds have the functional form f (x, y )eikz for some wavevector k . Assuming harmonic time dependence, we write explicitly E(x) = Ex (x, y )i + Ey (x, y )j + Ez (x, y )k ei(kzt) B(x) = Bx (x, y )i + By (x, y )j) + Bz (x, y )k ei(kzt) (1)
We have here a total of six functions f (x, y ) that we must nd to satisfy Maxwells equations with the relevant boundary conditions. At rst this would appear tough since the six elds are all coupled by Maxwells equations, but after a little algebra we obtain the following simplied situation: The z direction elds Ez (x, y ) and Bz (x, y ) turn out to satisfy (separately) simple onedimensional dierential equations, which may be readily solved upon specifying the boundary conditions for a particular situation. Meanwhile, the remaining elds (Ex , Ey , Bx , By ) can be expressed simply as linear combinations of Ez and Bz and their derivatives, so once we obtain the z elds we have everything. In what follows well derive the dierential equations satised by Ez and Bz and the equations giving the remaining elds in terms of them.
The dierential equations for Ez and Bz

The Maxwell curl equations are E= B , t B= 1 E c2 m t
where cm is the speed of light in the medium. We begin by applying the rst curl equation to our ansatz (1), obtaining y Ez ikEy = iBx x Ez + ikEx = iBy (2) (3) (4)
x Ey y Ex = iBz , i Ey = Bx y Ez . k k
and we pause to solve the rst two of these for Ex and Ey : Ex = i By x Ez , k k (5)
Next we apply the second curl equation to our ansatz, obtaining i Ex c2 m x Bz + ikBx = i 2 Ey cm x By y Bx = i 2 Ez . cm y Bz ikBy = (6) (7) (8)
But in (5) we solved for Ex and Ey , and if we then plug those solutions into (6) and (7) we can solve for Bx and By in terms of Bz and Ez : Bx = ikc2 m k 2 c2 m ikc2 m By = 2 k 2 c2 m 2 y Ez c2 mk y Bz 2 x Ez . cm k x Bz + (9) (10)
Finally, with the ansatz (1) the equation B = 0 reads Bx By + = ikBz . x y When we plug (9) and (10) into this, the terms involving Ez elds cancel, and we obtain an equation involving Bz alone: 2 2 + x2 y 2 or 2 2 + x2 y 2 Bz + 2 Bz = 0 (11) Bz + 2 k 2 Bz = 0 c2 m
where 2 k2. c2 If we had carried out this derivation in the reverse order we would have obtained the same equation for Ez : = 2 2 + 2 2 x y Ez + Ez = 0. (12)
We can think of equations (11) and (12) as eigenvalue equations that have solutions only for certain values of the parameter , which depend on the boundary conditions. Armed with equations (11) and (12) and the boundary conditions appropriate to our problem we can now solve for Bz and Ez and then use (9) and (10) to nd the remaining components of the B eld. The remaining components of the E eld are given by analogous equations: ikc2 m x Ez + 2 k 2 c2 m ikc2 Ey = 2 m y Ez k 2 c2 m y Bz k x Bz . k
Ex =
(13) (14)
Boundary Conditions; TE, TM, TEM Modes

The boundary conditions on the elds at the surfaces of the waveguide or cavity are that E and B be continuous, where denotes the component of the vector normal to the boundary surface and includes all other components of the vector. This means that the two eigenvalue equations (11) and (12) must be solved subject to dierent boundary conditions, which means in general their eigenvalues will be dierent. If we have a solution of (12) for some value of (i.e. for some combination of values of and k ), then there will be no nonzero solution of (11) for that value of , and hence we must have Bz = 0 identically for the mode at that frequency and wavevector. Since in this case the magnetic eld has nonzero components only transverse to the direction of propagation, this is called a transverse magnetic mode. Similarly, if (11) can be solved with nonzero Bz at some , then Ez = 0 and we have a transverse electric mode. A mode for which both Ez and Bz are zero is called a transverse electromagnetic mode, and can only exist in the region between two conducting surfaces, not within a single conductor as is possible for TE and TM modes. Since either Ez or Bz is zero, we can simplify some of the equations above and collect results appropriate to the two cases.
TM Modes
TE Modes
Bz 0
2 2 t Ez + Ez = 0,
Ez 0 En
S
=0
2 2 t B z + B z = 0,
Bn n
=0
S
Ex = Ey = Bx =
i y Ez 2 k 2 c2 m i x Ez By = 2 k 2 c2 m
ikc2 m x Ez 2 k 2 c2 m ikc2 m y Ez 2 k 2 c2 m
ic2 m y Bz 2 k 2 c2 m ic2 m Ey = 2 x Bz k 2 c2 m ikc2 Bx = 2 m x Bz k 2 c2 m ikc2 By = 2 m y Bz k 2 c2 m Ex =
(A factor of ei(kzt) is understood in all of these expressions.) For TM modes, the boundary condition is E = 0, and Ez is always perpendicular to the boundary surfaces, so the boundary condition for the eigenvalue equation is Ez = 0. For the case of TE modes, the boundary condition is B = 0. Suppose one boundary surface is the yz plane. The normal to this plane is the x direction, so Bx must vanish at this surface; but we just saw that in the TE case Bx x Bz , i.e. the derivative of Bz normal to the boundary surface must vanish. This is general: the boundary condition for the eigenvalue equation in the TM case is Bz /n = 0.
Power ow; Energy Loss

The ow of power down a waveguide is described by the z component of the 1 E B. Using the boxed expressions above, for Poynting vector S = E H = the two types of modes we obtain
TM Sz =
1 (Ex By Ey Bx ) kc2 m = (x Ez )2 + (y Ez )2 e2i(kzt) 2 2 ( k 2 c2 ) m kc2 m (x Ez )2 + (y Ez )2 2 k 2 c2 ) m (15)
or, in the time average, = = 2( 2
k (x Ez )2 + (y Ez )2 . 2 4
Similarly, for TE modes we obtain

TE Sz =
k (x Bz )2 + (y Bz )2 . 2 4
(16)
To address the issue of dissipation in the boundaries, we solve Maxwells equations within the boundary surfaces. The two curl equations are E = i B B = J i E (17)
= ( i ) E E
(18)
since in most cases. (For example, for a copper waveguide with air or vacuum interior we have we have 6 107 1 m1 , while 9 1012 1 m1 ( in rad/sec), so the approximation is good up to frequencies 1019 rad/sec.) Now we assume that the elds are only changing signicantly in the direction normal to the boundary surface (i.e., as we go deeper and deeper into the boundary surface the elds die out rapidly, whereas as we move along parallel to the boundary surface the elds dont change much) and keep only the normal derivative in the curl equations. If measures the depth of penetration into the surface, the curl equations become E = i B B = E Dierentiating the rst of these, taking the cross product with of both sides, and substituting in the second equation yields or, using the bac-cab rule, 2 E 2E = i E. 2 2 Evidently the component of the LHS vanishes here, so E = 0; the electric eld within the conducting boundary has no component normal to the surface. For the remaining components we obtain 2E + i E = 0 2 2E = i E 2
with solution E = e =e
i
E0
(1+i)
E0
where = 2/ is the skin depth and E0 is the eld just at the surface of the boundary. To keep the solution from blowing up as we penetrate into the conductor we take the negative sign in the exponent. From (17) we then obtain B=
i1 ( E0 )e(1+i) .
Evaluating this at the surface yields the modied boundary condition on the elds in the cavity or waveguide: B0 = i1 ( E0 ). (19)
From this equation we can work out the power loss per unit length in the cavity or waveguide. The power dissipated in a volume dV is (J E) dV = E 2 dV. We integrate over the volume occupied by the boundary surfaces in a length dz :
dP = dz
0
2 2(1+i) E0 e d dl e2i(kzt) 2 E0 dl e2i(kzt)
= dz
2(1 + i)
or, taking the time average, dP = dz 4 2

2 E0 dl
(20)
where the line integral is over the cross section of the surface boundary at a xed value of z .

Homer Reid May 24, 2002
Problem 8.2
A transmission line consisting of two concentric cylinders of metal with conductivity and skin depth , as shown, is lled with a uniform lossless dielectric (, ). A TEM mode is propagated along this line. (a) Show that the time-averaged power ow along the line is P = a2 |H0 |2 ln b a
where H0 is the peak value of the azimuthal magnetic eld at the surface of the inner conductor. (b) Show that the transmitted power is attenuated along the line as P (z ) = P0 e2z where = 1 2
1 a
+
b a
1 b
ln
(c) The characteristic impedance Z0 of the line is dened as the ratio of the voltage between the cylinders to the axial current owing in one of them at any position z . Show that for this line Z0 = 1 2 ln b a .
(d) Show that the series resistance and inductance per unit length of the line are R= ln 2 1 2 b a + 1 1 + a b c 4 1 1 + a b .
L=
(a) For the TEM mode, the electric eld in the waveguide may be written E(x, y, z, t) = Et (x, y )eikz eit where Et has only x and y components and may be derived from a scalar potential, i.e. Et = t . Since satises the Laplace equation, we may write its general form immediately (neglecting an arbitrary constant):
(, ) = 0 ln +
l=1
(l l + l l ) sin(l + l ).
In this part of the problem well neglect dissipation in the waveguide walls, so the boundary condition on Et is that its components transverse to the walls vanish, i.e. = = 0. r=b r=a This yields no condition on 0 , since the derivative of that term vanishes anyway, but on the terms in the summation we obtain the conditions l al + l al = l bl + l bl = 0 which can only be satised if l = l = 0 for l = 0. Hence we have () = 0 ln 1 . E = 0 (1)
The magnetic eld is found from Jacksons (8.28): 1 H= B= 0 = . The time-averaged Poynting vector is S= 1 1 (E H ) = 2 2 |0 |2 1
2
(z E) (2)
Integrating over the cross section of the waveguide, we obtain the power transfer:
b 2 0
P =
a
S dA |0 |2
b
= = =
1 2
2 d 2 a b |0 |2 ln a (a2 ) ln b a |0 |2 a2
(3)
Referring back to (2) to rewrite the term in brackets, we obtain P = (a2 ) ln b a |H (a)|2 (4)
(b) Without going back and completely re-solving for the elds in the waveguide for the case of nite conductivity, we can calculate the power loss per unit length
approximately using Jacksons equation (8.58): dP 1 = dz 2 1 = 2 =

2 0
| H|2 dl 2b 1 1 + b a 0 b .
2
+ 2a
0 a
Dividing by (3), we obtain = 1 dP 1 = 2P dz 2

1 a
+
b a
1 b
ln
(c) The elds inside the waveguide are E(, z, t) = H(, z, t) = 0 i(kzt) e 0 i(kzt) e
From the E eld we can compute the voltage dierence between the cylinders: V (z, t) = 0 ei(kzt)
b a
b d = 0 ei(kzt) ln a
(5)
while from the H eld we can compute the axial current owing in, say, the outer cylinder: I = 2b|Kb | = 2b| H( = b)| = 2 Dividing (5) by (6), we have Z= V 1 = I 2 b ln . a 0 ei(kzt) . (6)
Problem 8.4
Transverse electric and magnetic waves are propagated along a hollow, right circular cylinder of brass with inner radius R. (a) Find the cuto frequencies of the various TE and TM modes. Determine numerically the lowest cuto frequency (the dominant mode) in terms of the tube radius and the ratio of cuto frequencies of the next four higher modes to that of the dominant mode. For this part assume that hte conductivity of brass is innite. (b) Calculate the attenuation constant of the waveguide as a function of frequency for the lowest two modes and plot it as a function of frequency.
(a) The equation we have to solve is

2 (2 t + )(, ) = 0,
i.e. the Helmholtz equation. is Ez for the TM case and Hz for the TE case. The boundary conditions are ( = R) = 0 for the TM case, and ( /)( = R) = 0 for the TE case. The general solution of Helmholtz in 2D is
(, ) =
L=0
JL ()(AL eiL + BL eiL ) + NL ()(CL eiL + DL eiL ).
Since this solution must be valid everywhere in the interior of the waveguide, including at = 0, the part of the solution involving NL must vanish. Also, for a physical solution we must have L an integer. But otherwise I dont think there are any constraints on AL and BL . I guess these guys are determined by the eld conguration one forces into the waveguide at one of its ends. The allowed values of are determined by the boundary conditions. These are TM case : TE case : |=R = 0 =0
=R
= =
JL (R) = 0 JL (R) = 0
(7) (8) (9)
Hence the allowable eigenvalues are given by i = xi R
where the xi are the roots of JL (x) = 0 and JL (x) = 0. Referring to Jacksons tables on pages 114 and 370, we can write down the ve lowest-lying eigenvalues: 1 = 2 = 3 = 4a = 4b = 1.841 , R 2.405 , R 3.054 , R 3.832 , R 3.832 , R TE, L = 1 TM, L = 0 TE, L = 2 TE, L = 1 TM, L = 0.
The last two eigenvalues are degenerate. The lowest cuto frequency is 1 1.841 c = = . R (b) The lowest-lying mode is the TE mode with L = 1. For this mode we have Hz (, , z, t) = H0 J1 (1 )ei ei(kzt) (10)
2 with k 2 = 2 1 . The tangential component of the eld, from Jackson (8.33), is k (11) H (, , z, t) = 2 Hz
Using (10) and (11), we can nd the current induced in the wall of the conductor at = R: + k Ke = H( = R) = H0 J1 (1 R)ei ei(kzt ) 2z . R1 Then Jackson (8.58) is dP 1 = H 2 J 2 (1 R) 2R 1 + dz 2 0 1 k 2 R1
2
(12)
On the other hand, the transmitted power is given by Jackson (8.51): P = 1 2

2
1
2
2 1/2 A 2 1/2
Hz Hz dA R 0 2 J1 (1 ) d
2 = H0
The integral can be evaluated with Jackson (3.95):

2 = H0
2 1/2
R2 [J2 (1 R)]2 2
(13)
Dividing (12) by (13), we calculate the attenuation coecient: = = = 2 1 dP = 2P dz R 1 dP 2 = 2P dz R 1 dP 2 = 2P dz R J1 (1.841) J2 (1.841) J1 (1.841) J2 (1.841) J1 (1.841) J2 (1.841)
2
1+
2
k 2 R1
2 2 2
2
1/2
1+
2
Rk (1.841)2
2 2 2
1/2
1/2
R2 2 (1.841)2
2 2 2
Problem 8.5
A waveguide is constructed so that the cross section of the guide forms a right triangle with sides of length a, a, 2a, as shown. The medium inside has r = r = 1. (a) Assuming innite conductivity for the walls, determine the possible modes of propagation and their cuto frequencies. (b) For the lowest mode of each type calculate the attenuation constant, assuming that the walls have large, but nite, conductivity. Compare the result with that for a square guide of side a made from the same material.
(a) Well take the origin of coordinates at the lower left corner of the triangle. Then the boundary surfaces are dened by x = 0, y = a, and x = y . The 2 task is to solve (2 t + ) = 0 subject to the vanishing of , or its normal derivative, at the walls. In the text, Jackson nds the form of the solutions for a rectangular waveguide. A little bit of staring at the triangular waveguide reveals that appropriate solutions for this geometry can be assembled from linear combinations of the solutions for the rectangular case. For example, a term like sin kx x sin ky y , for suitable choices of kx and ky , already vanishes on the two legs of the triangle. To get it to vanish on the third boundary surface, i.e. the hypotenuse (x = y ), we can simply subtract the same term with kx and ky swapped. In other words, we take Ez (x, y ) = Hz (x, y ) = Amn sin Bmn mx ny nx my sin sin sin a a a a mx ny nx my cos cos + cos cos a a a a (TM) (TE)
The TE case involves the plus sign because in the normal derivative on the diagonal boundary surface the x derivative comes in with the opposite sign as the y derivative. 2 These satisfy (2 t + mn ) = 0, where 2 2 (nx + n2 y ). a In contrast to the rectangular case, TM modes with m = n vanish identically. For both TM and TE modes, mode (m, n) is the same mode as (n, m). As in the case of the rectangular waveguide, the smallest value of is to be had for the TE1 , 0 mode, in which case 10 = a and the cuto frequency is c(1,0) = /(a ). For the TM case the lowest propagating mode is (m, n) = (2, 1), for which 21 = 5/a and c(2,1) = 5c(1,0) .
2 = mn
(b) The lowest-frequency TE mode has Hz = H0 cos Ht = The power loss is ik 2 H0 a10 y x + cos a a x y sin i + sin j . a a |n H|2 dl x +1 a
2
1 dP = dz 2 On the lower surface (y = 0) we have

2 2 2 |n H|2 = |Hy + Hz | = H0
(14)
cos
k2 2 2 x 4 sin a2 10 a
The contribution of the lower surface to the integral in (14) is thus lower
2 = aH0
k2 2 3 + 2 4 2 a 10
(15)
The contribution of the right (vertical) boundary surface is the same. On the diagonal boundary surface, we have 1 n = ( i + j) 2 with magnitude |n H|2 = 1 2 2 2 [H + Hy + Hz + 2Hx Hy ] 2 x k H2 +4 = 0 4 cos2 2 a a10 = 1 ] n H = [Hz i + Hz j (Hy + Hx )k 2
sin2
where = x = y is the common coordinate as we move from (0, 0) to (a, a). In the integral in (14) we can put dl = 2d and integrate over from 0 to a to obtain k2 2 2 = 2aH0 1+ 2 4 . a 10 diagonal Adding this to two times (15) and inserting into (14), we have
2 k2 2 aH0 dP = 3 + 2 + (2 + 2) 2 4 . dz 2 a 10
(16)
On the other hand, from Jackson (8.51) we have P =

2 H0 2

a 0
y 0
2 2 x y x y cos2 + cos2 + 2 cos cos a a a a
1/2
dx dy
By symmetry, the integral is just half of what we would get from integrating the integrand over a square of side a, which, by inspection, is a2 . Hence P =
2 a 2 H0 4
2 2
.1/2
We could at this point proceed to write out the explicit form of the attenuation constant, but whats the point?
Problem 8.6
A resonant cavity of copper consists of a hollow, right circular cylinder of inner radius R and length L, with at end faces. (a) Determine the resonant frequencies of the cavity for all types of waves. With (1/ R) as a unit of frequency, plot the lowest four resonant frequencies of each type as a function of R/L for 0 < R/L < 2. Does the same mode have the lowest frequency for all R/L ? (b) If R=2 cm, L=3 cm, and the cavity is made of pure copper, what is the numerical value of Q for the lowest resonant mode?
(a) Taking the origin at the center of the cavity, the and components of the elds must vanish at z = L/2. Since the z dependence of all eld components is eikz , the allowed values of k are k = n/L, with E sin kz for k even and E cos kz for k odd.
10
The equation characterizing T M modes is

2 (2 t + )Ez = 0,
Ez |S = 0.
Expanding this in cylindrical coordinates, we obtain 2 Ez 1 Ez 1 2 Ez + + + 2 Ez = 0 2 2 2 We put Ez (, ) = R()P () to obtain 2P + P = 0 2 2 2 R 1 R 2 = 0. + + 2 2 The solutions are P () = ei R() = J (). For single-valuedness we require , and to ensure Ez ( = R) = 0 we require = xm /R where xm is the mth root of J (x) = 0. Hence
Ez = AJ xm
i it e e R
sin kz, cos kz,
k even k odd
(TM modes).
For TE modes the relevant equation is

2 (2 t + )Bz = 0,
Bz n
= 0.
S
The general solution to the dierential equation is the same as above, but now the boundary condition requires J (R) = 0, so = ym /R where ym is the mth root of J (y ) = 0. Then the solutions look like Bz = AJ ym i it e e R sin kz, cos kz, k even k odd (TE modes).
As we saw above, the allowed wavevectors are k = n/L. The frequency is related to the wavenumber according to
2 + k2 mn = cm m n cm 2 2 R xm + = c 2 + 2 m ym R
R L R L
n2 ,
2
(TM) (TE)
n2 ,
11
PSfrag replacements
9 TE modes TM modes 8
(units of c/R)
1 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
R/L Figure 1: TM and TE mode frequencies for the resonant cavity of Problem 8.6.
12
(The m subscript on cm is not related to the m subscripts on and ). The lowest four TM and TE mode frequencies are shown in Figure 1. (b) The lowest resonant mode is the TE1,1,1 mode.info

Homer Reid January 13, 2003
Problem 11.4
A possible clock is shown in the gure. It consists of a ashtube F and a photocell P shielded so that each views only the mirror M, located a distance d away, and mounted rigidly with respect to the ashtube-photocell assembly. The electronic innards of the box are such that when the photocell responds to a light ash from the mirror, the ashtube is triggered with a negligible delay and emits a short ash toward the mirror. The clock thus ticks once every (2d/c) seconds when at rest. (a) Suppose that the clock moves with a uniform velocity v, perpendicular to the line from P F to M, relative to an observer. Using the second postulate of relativity, show by explicit geometrical or algebraic construction that the observer sees the relativistic time dilatation as the clock moves by. (b) Suppose that the clock moves with a velocity v parallel to the line from P F to M. Verify that here, too, the clock is observed to tick more slowly, by the same time dilatation factor.
(a) Suppose that, relative to an observer, the clock is moving with speed v perpendicular to the direction in which the light travels back and forth. Let dt be the time measured by the observer for the light to travel from the ashtube to the mirror. The vertical distance traveled by the light is d, butas far as the observer is concernedduring the time dt the mirror has also translated a horizontal distance vdt. Hence the total distance the observer sees the light
travel (one-way) is d2 + (vdt)2 . But this distance must equal cdt, since the light must have the universal speed c in any reference frame. Hence we have cdt = or dt = 1 2 1+ v c2 d2 + (vdt)2
1/2
d . c
This is the time (as measured by the observer) in which the light makes the trip from ashtube to mirror. The light takes the same amount of time to travel back to the photocell, so the total period of the clock is just twice this, or period = 1 2 1 v c2
1/2
2d c
which is greater than the period of the clock at rest by the normal Lorentz dilatation factor. (b) Now we suppose that the clock is moving relative to an observer with speed v parallel to the direction of motion of the light. We align the z axis with the direction of motion. Then we may write down the space-time coordinates (in the clocks rest frame) of the three relevant events in the operation of the clock (taking x4 = ct): xa = (0, 0, 0, 0), xb = (0, 0, d, d), xc = (0, 0, 0, 2d), (light leaves ashtube) (light reaches mirror) (light reaches photocell).
and using this we may write down the spacetime coordinates of the three events above in the rest frame of the observer: xa = (0, 0, 0, 0) xb = 0, 0, (1 + )d, (1 + )d xc = 0, 0, 2d, 2d
The transformation matrix from the clocks reference frame to the observers reference frame is 1 0 0 0 0 1 0 0 = 0 0 0 0
Evidently the dierence in the time coordinates of events a and c (which is just the observed period of the clock) is ct = 2d, so again the observer observes
the clock to have a period of 2d/c, longer by the factor than the period of the clock in its rest frame.
Problem 11.5
A coordinate system K moves with a velocity v relative to another system K . In K a particle has a velocity u and an acceleration a . Find the Lorentz transformation law for acceleration, and show that in the system K the components of acceleration parallel and perpendicular to v are 1 1+
v2 c2 vu c2 3/2 3
a =
v2 c2 3
a =
1 1+
vu c2
a +
v (a u ) . c2
The initial components of the velocity in the moving frame (frame K ) are u and u . Using Jacksons equations 11.31 to transform these to frame K , we obtain u (0) = u (0) = u +v 1+ v
u v c2
(1) (2)
u u v . 1 + c2
After a time dt has passed in frame K , a time dt = dt/ has passed in frame K , and the velocity has increased by the amount a dt = a dt/. Then we can write down the new components of the velocity in K and again transform to K: u (dt) = u (dt) = v 1 + Subtracting (1) from (3), we have u + a dt + v 1+
(u +a dt )v c2
(3) (4)
u (u +a dt )v c2
u =
u + a dt + v 1+
(u +a dt )v c2
u +v 1+
v2 c2 u v c2 u v c2
a dt = 1+
u v c2 2
+ 1+
a dt
Using the relation dt = dt/ we can rewrite this as a dt 1 + 1+

v2 c2 3/2
= 1+
u v c2
u v c2
1+
v 2 1/2 c2
a dt
Dividing by dt and taking the limit as dt 0, we obtain a u a = lim = dt0 dt 1 1+

v2 c2 3/2 2
u v c2
So evidently Im o by 1 in the exponent of the denominator. What am I doing wrong? Next, subtracting (2) from (4), we have 1 u + a dt u u = u v v 1 + u v + a dt v 1 + c2 2 2 c c a dt v u v u a dt 1 + 2 2 c c 1 = u v 2 u v a dt v v 1 + c2 + 1 + c2 2 c v a dt + u a u a dt 2 c 1 = u v 2 u v a dt v v + 1 + c2 1 + c2 2 c Again putting in dt = dt , we have v u a u a dt a dt + 2 c 1 = 2 u v a dtv u v 2 v + 1 + c2 1 + c2 v c2
As before, when we divide by dt and take the limit as dt 0 the second term in the denominator becomes irrelevant and we obtain a = lim u = dt0 dt 1
2 1+ v u v c2
Jackson writes the second term in the brackets a little dierently. To show that his expression is equivalent to mine, we use the BAC-CAB rule: v (a u ) = (v u )a (v a )u = vu a + a va u + u
1+
2 v c2 u v
c2
a +
v u a u a c2
a +
v u a u a c2
The parallel components cancel, and we are left with = v (u a a u ) which is the way I wrote it above. But Im still o by 1 in the exponent of the term in the denominator! What am I missing?
Problem 11.6
Assume that a rocket ship leaves the earth in the year 2010. One of a set of twins born in 2080 remains on earth; the other rides in the rocket. The rocket ship is so constructed that it has an acceleration g in its own rest frame (this makes the occupants feel at home). It accelerates in a straight-line path for 5 years (by its own clocks), decelerates at the same rate for 5 more years, turns around, accelerates for 5 years, decelerates for 5 years, and lands on earth. The twin in the rocket is 40 years old. (a) What year is it on earth? (b) How far away from the earth did the rocket ship travel? Let v (t) be the speed of the rocket, as observed on earth, at a time t as measured on earth. Using the rst result of problem 11.5, we see that the acceleration of the rocket as measured from earth is a= dv v (t)2 = 1 2 dt c
3/2
g,
d 3/2 = 1 2 dt where = g/c and the plus or minus sign depends on whether we are accelerating or decelerating. Manipulating this a little, we obtain d = dt. (1 2 )3/2 Integrating, we obtain (1 2 )1/2
2 1
or
= (t2 t1 ).
(5)
For the rst leg of the rockets journey, we have t1 = 1 = 0 and we take the plus sign in (5). Then we nd = t (1 2 )1/2 or (t) = and (t) = 1 1 2 (t) = 1 + (t)2 (7) t [1 + (t)2 ]1/2 (6)
Next lets work out the relation between time as measured on the rocket and time as measured on earth. With primed (unprimed) quantities referring to the rocket (to earth), the innitesimal relation is dt = (t)dt or
t2
t2 t 1 = =
t1 t2 t1
dt (t) dt 1 + (t)2
du 1 t2 t1 1 + u2 1 = sinh1 (t2 ) sinh1 (t1 )
For the rst leg of the journey this becomes t2 = 1 sinh(t2 ). (8)
Now we know that the rst leg of the journey lasts until t2 = 5 years, and we have 9.8 m s2 g = 3.27 108 s1 . = = c 3 108 m s1 Then the time on earth at the end of the rst leg of the journey is, from (8), t2 = 1 3.27 108 sinh 3.27 108 s1 (5 yr) 3.153 107 s 1 yr s
= 2.65 109 s 84 yr.
Finally, the distance the rocket travels during the rst leg of its journey is
t
d=c
0 t
(t)dt t 1 + (t)2 dt
=c
0
c t u du 0 1 + u2 c 1/2 1 + (t)2 1 = 15 = 3.0 10 meters. = The behavior of the rocket on the subsequent three legs of the journey is similar to that in the rst leg. In particular, the total distance traveled away from earth is twice that covered in the rst leg, or 6.0 1015 meters, and the total time elapsed on earth during the rockets journey is four times that elapsed during the rst leg, or 484=336 years. So it should be the year 2436 on earth by the time the rocket returns home.
Problem 11.13
An innitely long straight wire of negligible cross-sectional area is at rest and has a uniform linear charge density q0 in the inertial frame K . The frame K (and the wire) move with a velocity v parallel to the direction of the wire with respect to the laboratory frame. (a) Write down the electric and magnetic elds in cylindrical coordinates in the rest frame of the wire. Using the Lorentz transformation properties of the elds, nd the components of the electric and magnetic elds in the laboratory. (b) What are the charge and current densities associated with the wire in its rest frame? In the laboratory? (c) From the laboratory charge and current densities, calculate directly the electric and magnetic elds in the laboratory. Compare with the results of part (a). I dont like q0 as a symbol for charge density, because it appears to have the wrong units. Ill use instead. Well take our z axis to coincide with the wire and take v in the positive z direction. (a) In the rest frame there is no current and the E eld is static; hence B = 0. The electric eld is found by considering a Gaussian pillbox in the shape of a right circular cylinder coaxial with the wire and of radius r and length dz . There is no electric eld normal to the upper and lower surfaces, and the eld normal to the radial bounding surface is uniform across the circumference. On the other hand, the charge enclosed in the cylinder is dz . Then Gauss law is E dA = 4Q = 2rdzEr = 4dz 2 = Er = . r In terms of cartesian components we have Ex = 2 x x2+y2 , Ey = 2 y x2 +y2 .
(9)
The eld-strength tensor in the laboratory frame is F = F 0 0 0 x 0 1 0 0 x 1 0 = 2 2 0 x +y 2 0 0 1 0 y 0 0 0 0 0 x y 0 x 2 0 0 x . = 0 0 y (x 2 + y 2 ) y 0 x y 0 y 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0
Reading o from this the transformed elds, and dropping the primes on x and y since the z boost leaves these coordinates unchanged, we have 2 (xi + yj) x2 + y 2 2 = r r B = (y i + xj) 2 2(x + y 2 ) 2 = . r E= (10) (11) (12) (13)
(b) In the rest frame there is no current and the charge density is = (x) (y ), so J = c( (x) (y ), 0, 0, 0). The transformed current density is J = J
0 0 0 1 0 = c 0 0 1 0 0 (x) (y ) 0 = c 0 (x) (y )
(x) (y ) 0 0 0 0 0 .
(c) Computing the electric eld in the laboratory frame is easy, since the charge density is the same as we had before but with a factor of thrown in. Then the E eld is just (9) but with that factor of thrown in, i.e. E= 2 r r
10
which agrees with (10). For the magnetic eld, we note that the current density in the lab frame is J = c (x) (y ) z. Then the current piercing a disc of radius r is I = c. On the other hand, by symmetry the magnetic eld in the azimuthal direction around the circumference of this disc is constant, so we may use Amperes law, B = (4/c)J to write 2rB = or B= which agrees with (13). 2 r 4 (c) c
Problem 11.15
In a certain reference frame a static, uniform, electric eld E0 is parallel to the x axis, and a static, uniform, magnetic induction B0 = 2E0 lies in the x y plane, making an angle with the axis. Determine the relative velocity of a reference frame in which the electric and magnetic elds are parallel. What are the elds in that frame for 1 and (/2)? The untransformed elds are E = E0 i, B = 2E0 (cos i + sin j).
Lets suppose we boost along the z axis. Then the elds transform according to Jackson equation (11.149): E = E0 (1 2 sin )i + 2E0 cos j B = 2E0 cos i + E0 (2 sin + )j. The angle between the elds is given by cos = = E B |E ||B |
2 (1 2 E0
2 2 2 E0 cos (1 + 2 ) 4 sin + 4 2 )1/2 (4 + 4 sin + 2 )1/2
Setting this equal to unity, we obtain
11
Problem 11.18
The electric and magnetic elds of a particle of charge q moving in a straight line with speed v = c, given by (11.152), become more and more concentrated as 1, as indicated in Fig. 11.9. Choose axes so that the charge moves along the z axis in the positive direction, passing the origin at t = 0. Let the spatial coordinates of the observation point be (x, y, z ) and dene the transverse vector r , with components x and y. Consider the elds and the source in the limit of = 1. (a) Show that the elds can be written as E = 2q r 2 (ct z ); r B = 2q v r (ct z ). 2 r
(b) Show by substitution into the Maxwell equations that these elds are consistent with a 4-vector source density, J = qcv (2) (r ) (ct z ) where the 4-vector v = (1, v ). (c) Show that the elds of part a are derivable from either of the following 4-vector potentials, A0 = Az = 2q (ct z ) ln(r ); A = 0 or A0 = Az = 0; A = 2q (ct z ) ln(r ) where is an irrelevant parameter setting the scale of the logarithm. Show that the two potentials dier by a gauge transformation and nd the gauge function, . (a) In the reference frame in which the particle is at rest at the origin, the elds are q E = (x i + y j + z k), B = 0. (x 2 + y 2 + z 2 )3/2 Transforming back to the laboratory frame according to Jackson 11.148, the electric eld is q E= (x i + y j + z k) 2 2 (x + y + z 2 )3/2 where in this expression the coordinates are still those of the observation point in the moving frame. The transformation of these to the lab frame is x = x, y = y = y , z = (z ct). Then the correct expression for the transformed eld is q E= 2 (xi + y j + (z ct)k). 2 [x + y + 2 (z ct)2 ]3/2
12
In the limit 1, we have . For z = ct the 2 factor in the denominator then ensures that all eld components are zero. For z = ct, however, although the z component of the eld clearly vanishes, the behavior of the other components is not as immediately clear. To elucidate the behavior of, say, the x component of the eld at z = ct we integrate it from z = ct to z = ct + :
ct+ ct+
Ex dz = qx
ct
qx 2 r 2qx = 2 r =
ct /r
2 [r
2 (z
/r
du [1 + u2 ]3/2
2
dz ct)2 ]3/2
2 + 2 r
Taking the limit for any nite , we nd

ct+
lim
Ex dz =
ct
2qx 2 . r
(14)
On the other hand, integrating between two points on the same side of z = ct, say from z = ct + to z = ct + 2 , we nd
ct+2
Ex dz =
ct+
2qx 2 r
2 2 + 4 2 r
2 + 2 r
which vanishes as . Since Ex vanishes at any point z = ct but yields something nonzero when integrated across that point, we conclude that it is just a function in (z ct) with coecient given by (14): Ex = and, similarly, Ey = Combining these, we can write E = 2q r 2 (ct z ). r 2qx 2 (z ct). r 2qy 2 (z ct). r
The B eld is given by Jackson (11.150) with, in the ultrarelativistic limit, : =k v r (ct z ). B = 2q 2 r

Classical Electrodynamics - Jackson (Solucionario 1)

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Classical Electrodynamics - Jackson (Solucionario 1)

Uploaded by

Copyright:

Available Formats

Solutions to Problems in Jackson, Classical Electrodynamics, Third Edition

Homer Reid December 8, 1999

The surface charge induced on the plane is found by dierentiating this: 1

Homer Reids Solutions to Jackson Problems: Chapter 2

d dz z=0 (z + d) q (z d) + = 4 |r + dk|3 |r + dk|3 qd = 2 (x2 + y 2 + d2 )3/2

rddr 2 + d2 )3/2 ( r 0 rdr (r2 + d2 )3/2

Homer Reids Solutions to Jackson Problems: Chapter 2 q2 16 0 d2

= 7.2 (1.6 1019 coulombs 1 V ) = 7.2 eV .

Homer Reids Solutions to Jackson Problems: Chapter 2

Homer Reids Solutions to Jackson Problems: Chapter 2

From this you can see that

(c) Total charge per unit length in z

x0 x0 tan1 + tan1 2 y0 2 y0 2 x 0 = tan1 . y0

Since tan1 x tan1 (1/x) = /2 the total charge induced is Q =

Homer Reids Solutions to Jackson Problems: Chapter 2

= [(x x0 )2 + (y y0 )2 ] x0 y0 = x2 (1 )2 + y 2 (1 )2 x y y0 x 0 x2 (1 2 ) + y 2 (1 2 x y = x2 2x0 x + y 2 2y0 y ) xx0 + yy0 x2 + y 2

xx0 + yy0 x2 + y 2 xx0 yy0 = (x2 + y 2 ) 1 2 2 x + y2 xx0 yy0 = (x2 + y 2 ) 1 2 x2 + y 2

(xx0 + yy0 )2 (x2 + y 2 )2 (xx0 yy0 )2 14 (x2 + y 2 )2

2 0 yy0 ) 1 4 (xx 2 2 2 (x +y ) . = ln 2 0 +yy0 ) 4 0 1 4 (xx (x2 +y 2 )2

Homer Reids Solutions to Jackson Problems: Chapter 2

d4 R = (d + R)[d2 R2 ]2 0 = d5 2d3 R2 2d2 R3 + dR4 + R5

q 2 R2 aR (2R + 3a)(R 4a) + 4 0 4a2 R2 R4

0 = d5 2d3 R2 4d2 R3 + dR4 + 2R5 . The root of this one is d/R=1.88.

Homer Reids Solutions to Jackson Problems: Chapter 2

(a) The force is |F | = q2 a 1 3 4 0 y (1 a2 /y 2 )2

directed radially inward. The work is W = = = = = =

Homer Reids Solutions to Jackson Problems: Chapter 2

its image is PE = = = 1 4 0 1 4 0 1 4 0 qq |r r | q 2 a r(r a2 /r) q 2 a r 2 a2

1 qQ q 2 a = 2 4 0 y 2y 1 qQ q 2 a = + 2 4 0 r 2r so the total work done is W = 1 q2 a q 2 a qQ . 4 0 2(r2 a2 ) 2r2 r

Homer Reids Solutions to Jackson Problems: Chapter 2

Review of Greens Functions

If is the scalar potential of electrostatics, we know that 2 (x ) = (x )/ 0 , so we have (x0 ) = 1

Homer Reids Solutions to Jackson Problems: Chapter 2

Solutions to Problems in Jackson, Classical Electrodynamics, Third Edition

Chapter 2: Problems 11-20

Homer Reids Solutions to Jackson Problems: Chapter 2 |x R i|2 . |x R i|2

2 + R 2 2R cos . 2 + R2 2R cos 1 2R cos 1 2R cos

Using ln(1 x) = (x + x2 /2 + ), we have = (c) =

2(R R ) cos 0 R(1 b2 /R2 ) cos 0

2b 2R cos 2b 2R cos = 4 b2 + R 2 2bR cos b2 + R2 2bR cos = 2 b2 + b

b R cos b2 + R2 2bR cos

The original line charge is at x = R, y = 0, and the eld there is E=

The force per unit width on the line charge is F = E = 2 R 2 0 R2 b2

tending to pull the original charge in toward the cylinder.

Homer Reids Solutions to Jackson Problems: Chapter 2

n {an sin(n) + bn cos(n)} .

(2) (3) (4)

Plugging back into (1), we nd (, ) = = 1 1

[sin(n) sin(n ) + cos(n) cos(n )] d cos n( ) . (5)

Plugging this back into (5) gives the advertised result.

Homer Reids Solutions to Jackson Problems: Chapter 2

2 1 cos(n)d cos(n)d + V2 V1 n b 0 1 2 = V1 |sin n|0 + V2 |sin n| nbn = 0.

With these coecients, the potential expansion becomes (, ) = V1 + V 2 2(V1 V2 ) + 2 1 n b

Homer Reids Solutions to Jackson Problems: Chapter 2

Here we need an auxiliary result: 1 n x sin n = n = = 1 2i 1 (iy )n [ein ein ] n (x = iy )

(Evidently, Jackson and I dened the angle dierently).

Homer Reids Solutions to Jackson Problems: Chapter 2

gn (y, y ) sin(nx) sin(nx )

where gn (y, y ) satises 2 n2 2 gn (y, y ) = (y y ) y 2 and gn (y, 0) = gn (y, 1) = 0.

Homer Reids Solutions to Jackson Problems: Chapter 2

Homer Reids Solutions to Jackson Problems: Chapter 2

0 0 0.2 0.4 yprime 0.6 0.8 1

Figure 1: gn (y, y ) from Problem 2.15 with n=5, y=.41

= n sinh[n (1 y> )] sinh(ny< )