WiSe 2012 15.11.

2012

Prof. Dr. A.-S. Smith

Dipl.-Phys. Ellen Fischermeier
Dipl.-Phys. Matthias Saba
am Lehrstuhl für Theoretische Physik I
Department für Physik
Friedrich-Alexander-Universität
Erlangen-Nürnberg

Theoretische Physik 2: Elektrodynamik

(Prof. A.-S. Smith)

Home assignment 5

Problem 5.1 Quadrupole moment

A nucleus with quadrupole moment Q finds itself in a cylindrically symmetric electric field with a
gradient (∂Ez /∂z)0 along the z axis at the position of the nucleus.
a) Show that the energy of quadrupole interaction is
 
e ∂Ez
W =− Q .
4 ∂z 0

b) If it is known that Q = 2 × 10−24 cm2 and that W/h is 10 MHz, where h is Planck’s constant,
calculate (∂Ez /∂z)0 in units of e/a30 , where a0 = }2 /me2 = 0.529 × 10−8 cm is the Bohr radius
in hydrogen.
c) Nuclear-charge distributions can be approximated by a constant charge-density throughout a
spheroidal volume of semimajor axis a and semiminor axis b. Calculate the quadrupole moment
of such a nucleus, assuming that the total charge is Ze. Given that Eu153 (Z = 63) has a
quadrupole moment Q = 2.5 × 10−24 cm2 and a mean radius
R = (a + b)/2 = 7 × 10−13 cm,
determine the fractional difference in radius (a − b)/R.

Problem 5.2 Hollow cylinder

a) 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 (φ, z). Using the appropriate separation of variables in cylindrical
coordinates, find a series solution for the potential anywhere inside the cylinder.
b) For the cylinder in part (a) the cylindrical surface is made of two equal half-cylinders, one at
potential V and the other at potential −V , so that
V ; − π2 < φ < π2

V (φ, z) = π 3π
−V ; 2 <φ< 2 .
(i) Find the potential inside the cylinder.
(ii) Assuming L  b, consider the potential at z = L/2 as a function of ρ and φ and compare
it with the two-dimensional case from Tutorial 5.3.
Problem 5.3 Spherical cap
A spherical surface of radius R has charge uniformly distributed over its surface with a density
Q/(4πR2 ), except for a spherical cap at the north pole, defined by the cone θ ≤ α.
a) Show that the potential inside the spherical surface can be expressed as
∞
QX 1 rl
Φ= [Pl+1 (cos α) − Pl−1 (cos α)] l+1 Pl (cos θ)
2 2l + 1 R
l=0

where, for l = 0, Pl−1 (cos α) = −1. What is the potential outside?

b) Find the magnitude and the direction of the electric field at the origin.
c) Discuss the limiting forms of the potential (from part a) and electric field (from 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.

Problem 5.4 Hemispherical boss

A large parallel plate capacitor is made up of two plane conducting sheets with separation D, one of
which has a small hemispherical boss of radius a on its inner surface (D  a). The conductor with the
boss is kept at zero potential, and the other conductor is at a potential such that far from the boss the
electric field between the plates is E0 .
a) Calculate the surface-charge densities at an arbitrary point on the plane and on the boss, and
sketch their behavior as a function of distance (or angle).
b) Show that the total charge on the boss has the magnitude 3E0 a2 /4.
c) If, instead of the other conducting sheet at a different potential, a point charge q is placed
directly above the hemispherical boss at a distance d from its center, show that the charge
induced on the boss is
d2 − a2
 
q 0 = −q 1 − √ .
d d 2 + a2

Due date: Tuesday, 20.11.12

WiSe 2012 15.11.2012

Prof. Dr. A.-S. Smith

Dipl.-Phys. Ellen Fischermeier
Dipl.-Phys. Matthias Saba
am Lehrstuhl für Theoretische Physik I
Department für Physik
Friedrich-Alexander-Universität
Erlangen-Nürnberg

Theoretische Physik 2: Elektrodynamik

(Prof. A.-S. Smith)

Solutions to Home assignment 5

Solution of Problem 5.1 Quadrupole moment

a) In the nucleus the charge distribution has cylindrical symmetry (refer Classical Electrodynamics, by
J. D. Jackson, pages 142-143). If we choose the z-axis in the direction of the azimuthal symmetry, the
only non-vanishing terms of the quadrupole moment are the diagonal ones

Q33
Q11 = Q22 = −
2
To see that we will write the non-diagonal terms of quadrupole moment in cylindrical coordinates
Z
0 0 0 0 0 0 0 0
Qij = 3xi xj ρe (ρ , z )ρ dρ dϕ dz
0 0 0
ρe = ρe (ρ , z ) is charge density, and ρ is coordinate.
0
Every one of the quadrupole moments inviolves integration with respect to ρ from 0 to 2π and sine
or cosine are the functions inside the integral. That is the reason why we get 0 as solution. Diagonal
terms of quadrupole moments are:

Z
0 0 0 0 0 0 0 0 0 0
Q11 = Qxx = (3ρ 2 cos2 ϕ − (ρ 2 + z 2 ))ρe (ρ , z )ρ dρ dϕ dz
Z
0 0 0 0 0 0 0 0 0 0
= (3ρ 2 sin2 ρ − (ρ 2 + z 2 ))ρe (ρ , z )ρ dρ dϕ dz = Q22 = Qyy
Q11 + Q22 + Q33 = 0

The relation betwwen the quadrupole moment Q33 and the quadrupole moment of nucleus is given by
the relation (Jackson, page 143)
Q33 = eQ
Quadrupole part of potential energy of the nucleus is
 1X ∂Ej
W = − Qij ( )0
6 ∂xi
ij
1 ∂E1 ∂E2 ∂E3
= − {Q11 (( )0 + ( )0 ) + Q33 ( )0 }
6 ∂x1 ∂x2 ∂x3
1 ∂E3 ∂E3
= − {−Q11 ( )0 + Q33 ( )0 }
6 ∂x3 ∂x3
1 ∂E3
= − Q33 ( )0
4 ∂x3
e ∂Ez
= − Q( )0
4 ∂z
~ = 0.
where we have used the equation ∇ · E
b)

∂Ez |W | h a3 e
|( )0 | = 4
∂z h Q e2 a3o
6.626 × 10−27 (0.529 × 10−8 )3 e
= 4 × 107
2 × 10−24 (4.8 × 10−10 )2 a30
e
= 0.085 3
a0

c) We calculate quadrupole moment Q33 with the help of spherical coordinates

x = aη cos ϕ sin θ
y = aη sin ϕ sin θ
z = bη cos θ

The variable η takes values in the interval [0,1]. The Jacobian for spherical coordinates is
x, y, z
J( ) = a2 bη 2 sin θ
η, θ, ϕ
Quadrupole moment Q33 for constant charge density ρe is

Z
Q33 = ρe (3z 2 − r2 )dV
Z
= ρe (2b2 η 2 cos2 θ − a2 η 2 sin2 θ)a2 bη 2 sin θdηdϕdθ
Z 1 Z π
4 2
η dη cos2 θ sin θdθ =
0 0 15
Z 1 Z π
4
η 4 dη sin3 θdθ =
0 0 15
2 3
8πa b ρe 8πa bρe 4 8πa2 bρe 2
Q33 = − = (b − a2 )
15 15 15
Volume of spheroid is
4π 2
V = a b
3

2
so for charge density we get

Ze 3Ze
=ρe =
V 4πa2 b
For quadrupole moment and quadrupole moment of nucleus we get

2Ze
Q33 = (b − a)(b + a)
5
4Z
Q = R(b − a)
5
If we set a > b, the quadrupole moment of the nucleus is negative Q < 0

a−b 5 |Q| 5 × 2.5 × 10−24

= = = 0.101 ≈ 10%
R 4 ZR2 4 × 63 × (7 × 10−13 )2

Solution of Problem 5.2 Hollow cylinder

a) Boundary conditions for the potential (Figure 1) are

Φ(z = L) = Φ(z = 0) = 0

Φ(ρ = a) = V (ϕ, z) . (1)

Fig. 1: Upright cylinder, the bases of which are on the zero potential, whereas its sheet is on the potential
V (ϕ, z)

We are going to find the solution of the Laplace equation using the method of separation of
variables
Φ(~x) = R(ρ)Q(ϕ)Z(z) . (2)
Upon insertion of the (2) in the Laplace equation we obtain
 2
1 d2 Q 1 1 d2 Z

d R 1 dR 1
+ + + =0 . (3)
dρ2 ρ dρ R ρ2 dϕ2 Q Z dz 2
Coordinate z is appearing only in the last term so we set
1 d2 Z
= −k 2 , (4)
Z dz 2

3
 where the constant k ≥ 0. The general solution of the (4) is

Z(z) = A sin kz + B cos kz . (5)

If we insert the boundary conditions Φ(z = L) = Φ(z = 0) = 0 in (5) we obtain

Z(z) = A sin kp z
pπ
kp = , p ≥ 1 because by definition kp ≥ 0 . (6)
L
We can find the solution for the function Q if we set

d2 Q 1
= −m2 , (7)
dϕ2 Q
where the constant m ≥ 0. Uniqueness of the function Q implies that m can assume only
discrete values
Q(ϕ) = C sin mϕ + D cos mϕ , m = 0, 1, 2, ... (8)
Thus, the variable ρ has to satisfy the following equation

d2 R 1 dR m2
 
2
+ − k + 2 R=0 . (9)
dρ2 ρ dρ ρ

The general solution contains modified Bessel functions of the first Im (kp ρ) and of the second
Km (kp ρ) kind
R(ρ) = EIm (kp ρ) + F Km (kp ρ) . (10)
When ρ → 0, the function Km (kp ρ) → ∞. To prevent the singularity of the potential inside of
the cylinder, we set F = 0. Solution for the R(ρ) is then

R(ρ) = EIm (kp ρ) . (11)

The general solution for the potential inside of the cylinder is

∞ X
X ∞
Φ(ρ, ϕ, z) = (Amp sin mϕ + Bmp cos mϕ)Im (kp ρ) sin kp z . (12)
m=0 p=1

Over the cylinder sheet (ρ = b) the potential is equal to V (ϕ, z). Using this condition we can
determine the coefficients Amp and Bmp .
∞ X
X ∞
Φ(ρ = b, ϕ, z) = V (ϕ, z) = (Amp sin mϕ + Bmp cos mϕ)Im (kp b) sin kp z . (13)
m=0 p=1

By multiplying the equation (13) with sin kp0 z cos m0 ϕ, then with sin kp0 z sin m0 ϕ, and by sub-
sequent integration we obtain
Z L Z 2π
1
B0p = dz dϕV (ϕ, z) sin kp z
πLI0 (kp b) 0 0
Z L Z 2π
2
Amp = dz dϕV (ϕ, z) sin kp z sin mϕ
πLIm (kp b) 0 0
Z L Z 2π
2
Bmp = dz dϕV (ϕ, z) sin kp z cos mϕ , m > 0 , (14)
πLIm (kp b) 0 0

4
 where we used the orthogonality relations
Z L
L
dz sin kp0 z sin kp z = δp0 p
0 2
Z 2π
dϕ sin m0 ϕ sin mϕ = πδm0 m
0
Z 2π
dϕ cos m0 ϕ cos mϕ = πδm0 m
0
Z 2π
dϕ sin m0 ϕ cos mϕ = 0 . (15)
0

By determining the coefficients in (14) the problem is solved.

b) (i) In the part a) of the problem we calculated the potential inside of the cylinder of the length
L, where both bases are grounded and the cylinder sheet is on the potential V (ϕ, z)
∞ X
X ∞
Φ(ρ, ϕ, z) = (Amp sin mϕ + Bmp cos mϕ)Im (kp ρ) sin kp z
m=0 p=1

Z L Z 2π
1
B0p = dz dϕV (ϕ, z) sin kp z
πLI0 (kp b) 0 0
Z L Z 2π
2
Amp = dz dϕV (ϕ, z) sin kp z sin mϕ
πLIm (kp b) 0 0
Z L Z 2π
2
Bmp = dz dϕV (ϕ, z) sin kp z cos mϕ , m > 0 , (16)
πLIm (kp b) 0 0

where kp = pπ/L, p = 1, 2, ...

Fig. 2: Distribution of the potential over the cylinder.

Before we start to calculate the coefficients, observe that the distribution of the potential
on the cylinder is an even function for the transformation ϕ → −ϕ (Figure 2). Then the
potential also possesses the same symmetry, in other words, the potential is also the even
function in ϕ, so the coefficients Amp = 0. Coefficients Bmp are equal to
Z L  Z π/2 Z 3π/2 !
V
B0p = dz sin kp z dϕ − dϕ = 0 (17)
πLI0 (kp b) 0 −π/2 π/2

5
 !
Z L  Z π/2 Z 3π/2
2V
Bmp = dz sin kp z cos mϕdϕ − cos mϕdϕ
πLIm (kp b) 0 −π/2 π/2
      mπ 
2V 1 2  mπ  1 3mπ
= − (cos kp L − 1) sin − sin − sin
πLIm (kp b) kp m 2 m 2 2
  
2V 1  mπ  3mπ
= 2 [1 − (−1)p ] 3 sin − sin . (18)
π Im (kp b) mp 2 2

Using the relations for the transformation of the sum and difference of the sine function
into product we obtain
 mπ     mπ    mπ   
3mπ 3mπ
3 sin − sin = 2 sin + sin − sin
2 2 2 2 2
 mπ   mπ 
= 2 sin − 2 sin cos mπ
2 
 mπ 2
= 2 sin [1 − (−1)m ] . (19)
2
Coefficient Bmp is (for m > 0) equal to

4V 1  mπ 
Bmp = sin [1 − (−1)p ][1 − (−1)m ] . (20)
π 2 Im (kp b) mp 2

If either of the indices p and m is even, then Bmp = 0. Solution for the potential, using the
following substitutions

m → 2m − 1
p → 2p − 1 (21)

becomes
 
(2p − 1)π
∞ ∞ I2m−1 ρ
(−1)m−1
 
16V X X L (2p − 1)π
Φ(ρ, ϕ, z) = 2   sin z cos [(2m − 1)ϕ],
π (2m − 1)(2p − 1) (2p − 1)π L
m=1 p=1 I2m−1 b
L
(22)
where we used sin [(2m − 1)π/2] = (−1)m−1 .
(ii) Consider the sum over p in (22) for z = L/2 and for L  b > ρ
 
(2p − 1)π
∞ I2m−1 ρ
X 1 L
  (−1)p−1 . (23)
(2p − 1) (2p − 1)π
p=1 I2m−1 b
L
If
(2p − 1)π (2p − 1)π
ρ< b1 , (24)
L L
then (here Labinac refers to Jackson, page 108, equation 3.102, even though this depends
on the version of the Jackson he used)

(2p − 1)π 2m−1

   
(2p − 1)π 1
I2m−1 ρ ' ρ
L Γ(2m) L
(2p − 1)π 2m−1
   
(2p − 1)π 1
I2m−1 b ' b . (25)
L Γ(2m) L

6
 Sum over the index p becomes
∞
(−1)p−1  ρ 2m−1 
 ρ 2m−1 X 
1 1 π  ρ 2m−1
= 1 − + − ... = . (26)
b (2p − 1) b 3 5 4 b
p=1

In other words, sum over p gives the result (26) when L → ∞ because then for every finite
p the condition (24) is satisfied. In this limit the potential is
∞
4V X (−1)m−1  ρ 2m−1
Φ(ρ, ϕ, z) = cos [(2m − 1)ϕ] . (27)
π (2m − 1) b
m=1

If in the Tutorial 5.3 we put V2 = −V1 = −V , we obtain exactly this expression for the
potential.

Solution of Problem 5.3 Spherical cap

a) The configuration is shown in Figure 3.

From the given conditions we know that the

charge-density is azimuthaly symmetric and
satisfies

0 , 0≤θ≤α
σ(θ) =
Q/(4πR2 ) , α ≤ θ ≤ π

and we expect that the potential has the forms Fig. 3: Configuration in the problem.
∞
X
Φin (r, θ) = Al rl Pl (cos θ) inside the sphere
l=0
X∞
Φout (r, θ) = Bl r−(l+1) Pl (cos θ) outside the sphere .
l=0

The boundary conditions for the sphere r = R are

Φin (r, θ) = Φout (r, θ) (28)


~ out − E
(E ~ in )·~n = 4πσ(θ) (29)
r=R

From the boundary condition (28) we get

∞
X ∞
X
Al Rl Pl (cos θ) = Bl R−(l+1) Pl (cos θ)
l=0 l=0
⇒ Al Rl = Bl R−(l+1) .

The secondary boundary condition gives

~n = ~er


~ out − E
(E ~ in ) = 4πσ(θ) (30)
r=R

7
 For the L.H.S. in (30) we can write
  ∞ h
∂Φout ∂Φin  X
−(l+2) l−1
i
− + = (l + 1)B l R + lAl R Pl (cos θ)
∂r ∂r r=R
l=0

∞ h
- -Z
X i π
⇒ (l + 1)Bl R−(l+2) + lAl Rl−1 Pl (cos θ) = 4πσ(θ) × Pl0 (cos θ) sin θ dθ
l=0 0
∞ h
X iZ π Z π
−(l+2) l−1
(l + 1)Bl R + lAl R Pl (cos θ)Pl0 (cos θ) sin θ dθ = 4π σ(θ)Pl0 (cos θ) sin θ dθ
l=0 0 0

From the orthogonality relation we know

Z π
2
Pl0 (cos θ)Pl (cos θ) sin θ dθ = δl 0 l .
0 2l + 1
Second integral
Z π Z π
Q
σ(θ)Pl (cos θ) sin θ dθ = Pl (cos θ) sin θ dθ (31)
0 4πR2 α

Changing the variable cos θ = x and using the relation

dPl+1 dPl−1
− − (2l + 1)Pl = 0 , l≥1
dx dx
the integral (31) becomes
Z π Z cos α  
Q 1 dPl+1 (x) dPl−1 (x)
σ(θ)Pl (cos θ) sin θ dθ = − dx , l≥1
0 4πR2 2l + 1 −1 dx dx
Q 1
= 2
[Pl+1 (cos α) − Pl−1 (cos α)]
4πR 2l + 1

Q
⇒ (l + 1)Bl R−(l+2) + lAl Rl−1 = [Pl+1 (cos α) − Pl−1 (cos α)] (32)
2R2
For l = 0
Z π
Pl (cos θ) sin θ dθ = cos α + 1 . (33)
α

If we require that (32) is also valid for l = 0 and in the accordance with (33) we take as the
definition P−1 (cos α) = −1. The coefficients Al and Bl are then equal to
Q 1
Al = l+1
[Pl+1 (cos α) − Pl−1 (cos α)] (34)
2R 2l + 1
Q Rl
Bl = [Pl+1 (cos α) − Pl−1 (cos α)] . (35)
2 2l + 1
Finally, the potential solution reads
∞
QX 1 rl
Φin = [Pl+1 (cos α) − Pl−1 (cos α)] l+1 Pl (cos θ) (36)
2 2l + 1 R
l=0
∞
Q X 1 Rl
Φout = [Pl+1 (cos α) − Pl−1 (cos α)] l+1 Pl (cos θ) . (37)
2 2l + 1 R
l=0

8
b) The electric field inside the sphere is

~ in = − ∂Φin ~er − 1 ∂Φin ~eθ

E
∂r r ∂θ
∞
QX 1 n 0
o
= [Pl+1 (cos α) − Pl−1 (cos α)]R−(l+1) −Pl (cos θ)lrl−1~er + rl−1 sin θPl (cos θ)~eθ .
2 2l + 1
l=1
0
where Pl (cos θ) = dPl (cos θ)/d(cos θ). At the origin all terms vanish except the term l = 1

~ in (r = 0) = Q [P2 (cos α) − P0 (cos α)][−P1 (cos θ)~er + sin θP 0 (cos θ)~eθ ]

E 1
6R2
0 1
P1 (cos θ) = cos θ ; P1 (cos θ) = 1 ; P2 (cos α) = (3 cos2 α − 1)
2
⇒E ~ in (r = 0) = Q sin2 α(cos θ ~er − sin θ ~eθ ) = Q sin2 α ~ez .
4R2 4R2
c) When α → 0, the nonvanishing terms are those for which l = 0

Q Q
φin →
; φout →
R r
~ in (r = 0) → 0. In this limit the
In that case electrical field at the origin of the sphere is E
charge distribution mimics the uniform charge distibution in a sphere. For α → π we will expand
the expression

I(x) = Pl+1 (x) − Pl−1 (x)

in a Taylor series around x = cos π = −1

∂I(x)
I(x) = I(x)|x=−1 + |x=−1 (x + 1) + ...
∂x
∂Pl+1 (x) ∂Pl−1 (x)
= Pl+1 (−1) − Pl−1 (−1) + { |x=−1 − |x=−1 }(x + 1) + ...
∂x ∂x
= (2l + 1)Pl (x)|x=−1 (x + 1) + ...
' (2l + 1)(−1)l (x + 1)
For the potential we now get

∞
Q X rl
φin ' (cos α + 1) (−1)l l+1 Pl (cos θ)
2 R
l=0
∞
Q X Rl
φout ' (cos α + 1) (−1)l Pl (cos θ)
2 rl+1
l=0

And electrical field at the origin

Q Q Q
E~in (r = 0) = 2
sin2 αe~z = 2
(1 − cos α)(1 + cos α)e~z → (1 + cos α)e~z
4R 4R 2R2
The sum term in the expression for potentals can be summed
∞
X rl 1
(−1)l l+1
Pl (cos θ) = √
R r2 + R2 + 2rR cos θ
l=0

In this limit the charge density is like having one single charge Q/2(1 + cos α) at z = −R

9
Solution of Problem 5.4 Hemispherical boss
a) Our aim will be to find a potential which satisfies all the boundary conditions, as that will indeed
be the potential for the given problem. To find this potential, we will first find the potential for
a somewhat easier problem, and then tweak it until it satisfies all the boundary conditions of
the original problem.
So, let us look at the following auxiliary problem: a charge −q is positioned at a distance R
on the z-axis from the origin in front of the conducting sheet which has a small hemispherical
boss of radius a. Let us calculate the potential in the part of the space where r ≥ a and θ ≤ π/2
(Figure 1), where r and θ are as in the usual spherical coordinates.
Since such a sheet containing a spherical boss is suggestive of a hemi-sphere stuck on to
a plane, we can try to solve this problem by combining the approaches using the method of
images for the case of a plane and for that of a sphere. So let us first consider the plane, and
position the image charge q 0 = q at the point z 0 = −R. Then the plane is at zero potential.
However, the boss is not. So let us now position a second image charge q 00 = qa/R at the point
z 00 = a2 /R. Now the plane is not at zero potential anymore, so let us position another image
charge q 000 = −q 00 = −qa/R at the point z 000 = −z 00 = −a2 /R. The total potential is equal to
the sum of the potentials of the charge −q and the three image charges

Fig. 4: A point charge near a grounded, conducting sheet with a hemispherical boss.

q q
Φs (~r) = 1/2
−
(r2 + R2 + 2rR cos θ) (r2 + R2 − 2rR cos θ)1/2
−aq aq
+  1/2 +  1/2 .
a 4 2
2a r a 4 2a2 r
2
R r + 2+ cos θ 2
R r + 2− cos θ
R R R R
At last, the boundary coditions for this auxiliary problem are satisfied: for both the planar
and the boss part of the sheet, Φs (~r) = 0 (as can easily be verified).
This potential Φs (~r) is not the solution to the original problem, as it does not satisfy the
condition that far away from the sheet, the electric field is constant at E0 (in other words, the

10
condition that lim Φ(~r) = −E0 z = −E0 r cos θ). So now we try to tweak Φs (~r) a bit. So let
r→∞
R → ∞. Then, Φs (~r) can be expanded as

r2 r2
 
q r r
Φs (~r) = 1− − cos θ − 1 + − cos θ + ...
R 2R2 R 2R2 R
a4 a2 a4 a2
 
qa
− 1− 2 2 − cos θ − 1 + 2 2 − cos θ + ...
rR 2r R rR 2r R rR

2qa3 a3 a3
   
2qr 2q r
Φs (~r) ' − 2 cos θ + 2 2 cos θ = − 1− 3 cos θ = −E0 r − 2 cos θ
R r R R r R r
2q
where E0 := = constant.
R2
a3
 
Hence, as r → ∞, Φs (~r) → −E0 r cos θ, as desired. Therefore, Φs (~r) = Φ(~r) = −E0 r − 2 cos θ
r
is the desired potential.
2q
Note that, physically, E0 = 2 is the magnitude of an electric field between two infinite hemi-
R
q
spherical (of radius R → ∞) sheets over which a surface charge σ = has been smeared
2πR2
(because, as R → ∞, these hemi-spherical sheets together form a parallel plate capacitor,
2q
between the plates of which we know the electric field to be E0 = 4πσ = 2 ). Thus, having
R
found the potential, we can now calculate the electric field

a3 a3
   
~ ~ ∂Φ 1 ∂Φ
E = −∇Φ = − ~er − ~eθ = E0 cos θ 1 + 2 3 ~er − E0 sin θ 1 − 3 ~eθ .
∂r r ∂θ r r

Now, finally, we can calculate the surface charge density on the planar and the boss part

a3 a3
     
1 ~  E0  E0
σp ≡ σplanar = E · ~ez 
 =− 1 − 3 sin θ~eθ · ~ez 
 = 1− 3
4π θ=π/2 4π r θ=π/2 4π r

because in the planar part of the surface of interest ~er · ~ez = 0 and ~eθ · ~ez = −1.

1 ~  3
σb ≡ σboss = E · ~er  = E0 cos θ .
4π r=a 4π

The surface charge densities on the boss and the planar part are shown in the figures 3 and 4,
respectively.
Note: The surface charge density on the boss part is the same as the one found for the sphere
in the uniform field (see homework, Problem 4.2 b).

Fig. 5: Surface charge density on the boss. Fig. 6: Surface charge density on the planar part.

11
 b) The total charge on the boss is
2π π/2 π/2
3E0 a2 3E0 a2
Z Z Z
Qb = dϕ dθ sin θa2 σb = dθ sin θ cos θ = .
0 0 2 0 4

c) We have already found the potential for this configuration in the ‘auxiliary problem’ of part a),
with the only difference being that here −q and R are replaced by q and d, respectively. So,
making the following changes in Φs (~r) from part a),

q → −q
R → d,

we get
q q
Φs (~r) = − 1/2
+
(r2 + d2 + 2rd cos θ) (r2 + d2 − 2rd cos θ)1/2
aq aq
+ 1/2 − 1/2 .
a4 2a2 r a4 2a2 r
 
d r2 + 2+ cos θ d r2 + 2− cos θ
d d d d

The surface charge density on the boss is

 
1 ~  1 ∂Φ 
σ= E · ~er 
 =−
4π r=a 4π ∂r r=a
" #
q d2 − a2


1 1
= − .
4πa (a2 + d2 + 2ad cos θ)3/2 (a2 + d2 − 2ad cos θ)3/2

In the calculation of the induced charge on the boss, integrals of the following form appear
Z π/2 !
sin θdθ 1 1 1
=∓ − .
0 (a2 + d2 ± 2ad cos θ)3/2 ad d ± a (a2 + d2 )1/2

The induced charge on the boss is then

Z 2π Z π/2
q0 = dϕdθ sin θa2 σ(θ)
0 0

" ! !#
qa d2 − a2 1 1 1 1 1 1
= − − + −
2 ad d + a (a2 + d2 )1/2 ad d − a (a2 + d2 )1/2

" # " #
q d2 − a2 2d 2 d2 − a2
= − = −q 1 − .
2d d2 − a2 (a2 + d2 )1/2 d (d2 + a2 )1/2

12