Electromagnetism 70006 Answers to Problem Set 9 Spring 2006

1. Jackson Prob. 5.1: Reformulate the Biot-Savart law in terms of the solid
angle subtended at the point of observation by the current-carrying circuit.
I
0 I dl0 (r r 0 )
B(r) =
2 |r r 0 |3
I
0 I 1
= dl0
2 |r r 0 |
I
0 I dl0
=
2 |r r 0 |
Let I
dl0
V =
|r r 0 |
then
0 I
B(r) = V
2
The ith component of V may be written
I
dl0 i
Vi =
|r r 0 |
where i is the unit vector along the ith axis. By virtue of Stokes theorem
this can be converted into a surface integral
Z " # Z
0 i 0 0 0 1
Vi = da n = da n i
S |r r 0 | S |r r 0 |

where S is a surface bounded by the circuit and where the direction of

the surface normal n0 is related to the sense of the current (l0 ) by the
right-hand rule. The above equation can be rewritten as
Z
n0
V = da
S |r r 0 |
Therefore
0 I
B(r) = [ W ]
4
with Z
n0
W = da
S |r r 0 |
Now
[ W ] = ( W ) 2 W
Z Z
n0 (r r 0 )
= da 0 |3
+ 4 da n0 (r r 0 )
S |r r S
= + 0.

1
 The second integral vanishes since r 0 is on a surface bounding the circuit,
which is away from the observation point r. The first integral is, as shown
on page 33 in Chap. 1 of the text, the solid angle subtended at the
observation point by the circuit that bounds S. Therefore,
0 I
B(r) =
4

Example: Consider a point of observation on the z axis above a circular

loop of radius a in the xy plane that carries current I. The loop subtends
a solid angle = 2(1 cos ), where is the angle between the z axis
and a line from the point of observation to the loop. Thus

z
= 2 1
a2 + z 2
and

1 z2 2a2
= 2 + 2 2 3/2
z = 2 z.
a2 + z 2 (a + z ) (a + z 2 )3/2
Therefore
0 I a2
B(z) = z,
2 (a2 + z 2 )3/2
confirming a result obtained in class directly from the Biot-Savart law.
2. Jackson Prob 5.3: Find Bz inside a uniformly wound solenoid. Use result
from Prob. 5.1 to write the contribution from a segment of the solenoid
of length dz as
0 N Idz d o N I
dBz = = d.
4 dz 4
where = 2(1 cos ) where is the angle that the line from the
observation point to the ring at z makes with the axis. Integrate from end
1 to end 2 to find
o N I 0 N I
Bz = [1 2 ] = [cos 2 cos 1 ] .
4 2
In terms of the angles shown in the figure in the text this becomes
0 N I
Bz = [cos 1 + cos 2 ]
2
3. Jackson Prob 5.7:
(a) As shown in class and in example with Prob. 5.1, the field of a single
loop in the xy plane at a distance z from its center on the axis is
0 I a2
Bz =
2 [a + z 2 ]3/2
2

2
(b) The field near the center of a Helmholtz pair is, therefore,

0 I a2 a2
Bz = +
2 [a2 + (z b/2)2 ]3/2 [a2 + (z b/2)2 ]3/2
" 2
0 Ia2 3 b a2 z 2 15 2a4 6b2 a2 + b4 z 4
= 1+ +
d3 2 d4 16 d8
#
7 5a6 30b2 a4 + 15b4 a2 b6 z 6
+ ,
16 d12

where d = b2 + 4a2 .
(c) dependence of of field. On the axis, we may write Bz = 0 +
2 z 2 + , where the coefficients k can be inferred from the above
equation. With the aid of B = 0 we find that near the origin,

(B ) Bz
= = 22 z +
z

Solving for B (taking into account that B = 0 on axis) we find that

near the axis,
B (z, ) 2 z
From B = 0, we may write
Bz B
= 2
z
Solving for Bz , we obtain

2 2 2
Bz (z, ) Bz (z, 0) 2 = 0 + 2 z +
2 2

(d) From mathematica the asymptotic series for Bz is

"
0 Ia2 3 b2 a2 15 2a4 6b2 a2 + b4
Bz == 1+ +
|z|3 2 z2 16 z 4
#
7 5a6 30b2 a4 + 15b4 a2 b6
+ ,
16 z 6

which can be obtained from the power series by the replacement

d |z|.
(e) For a Helmholtz coil one sets b = a. With this choice the terms in
the bracket for the small z expansion become

144z 4
[ ] 1
125a4

3
 Thus, to differ from uniformity by , the fractional distance must
satisfy z/a [(125/144) ]1/4 . For = 104 the limit is 0.097 and for
= 102 the limit is 0.305. Below is a figure showing the variation

Bz HzL for a Helmholtz coil

of Bz on the axis between two coils located at a = 1.

1
0.8
0.6
0.4
0.2
z
-1 -0.5 0.5 1
4. Jackson Prob 5.13: Find the vector potential and magnetic induction for a
uniformly charged sphere of radius a rotating about an axis with angular
momentum . We orient along the z axis and let r lie in the xz plane.
The vector r 0 is used to locate a point on the sphere. The surface current
density at r 0 is K(r 0 ) = [ r 0 ] . The corresponding vector potential is
Z Z Z
0 J (r 0 )d3 r0 0 K(r 0 )da0 0 [ r 0 ] da0
A(r) = 0
0
=
4 |r r | 4 |r r | 4 |r r 0 |
We write
[ r 0 ] = a sin 0 ( sin 0 x + cos 0 y)
As in the example worked out in Sec. 5.5 of the text, only the y component
can contribute to the integral. Therefore,
Z
0 a3 sin 0 cos 0 d0
Ay (r) = p ,
4 r2 + a2 2ar cos
where cos is the angle between r and r 0 . Expanding the denominator in
a series of spherical harmonics we obtain
Z
0 a3 X r< l
4
Ay (r) = Y (r) d0 sin 0 cos 0 Ylm (r0 )

l+1 2l + 1 lm
4 r
lm >

We first carry out the 0 integral to find

Z 2 s
0 0 0 0 (2l + 1)(l 1)! 1
d cos Ylm ( , ) = Pl (cos 0 ) (m1 m1 ).
0 4(l + 1)!

Noting that sin 0 = P11 (cos 0 ), we find

Z 1
4
sin 0 Pl1 (0 )d0 = l1
1 3

4
Putting the previous two results together, we find
Z r
0 0 0 0 4 3
d sin cos Ylm (r ) = (m1 m1 )l1
3 8
The sum over lm above becomes
r
X rl 16 2 3 r< 4 r<
<
l+1
= (Y11 (r) Y11 (r)) 2 = sin 2
r> 9 8 r> 3 r>
lm

Finally,

0 a4 1
A (r) = sin 2 r>a
3 r
0 a
= r sin r<a
3
In vector form, this becomes

0 a4 [ r]
A(r) = r>a
3 r3
0 a
= [ r] r<a
3
The corresponding formulas for the magnetic induction B = A are

0 a4 3( r)r
B(r) = r>a
3 r3
20 a
= r<a
3

Note: A somewhat different (and simpler) solution to this problem is found

in the text by Griffiths. He chooses coordinates with r be along the z axis
and in the xz plane at an angle with the z axis.