Chapter 2

Problem 2.1
A point charge q is brought to a position a distance d away from an infinite
plane conductor held at zero potential. Using the method of images, find:

(a) the surface-charge density induced on the plane, and plot it;

(b) the force between the plane and the charge by using Coulomb’s law for
the force between the charge and its image;

(c) the total force acting on the plane by integrating σ 2 /20 over the whole
plane;

(d) the work necessary to remove the charge q from its position to infinity;

(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.

Part a
Figure 2.1 shows the arrangement of charges corresponding to a unit charge q
placed a distance d from a grounded plane conductor. Using the definition of
coordinates from the figure, q is placed at z = dẑ and −q at z0 = −dẑ. The
potential for z ≥ 0 is then given by the potential of the two point charges
 
q 1 1
Φ(x) = −
4π0 |x − dz| |x + dz|
 
q 1 1
= −
4π0 [x2 + y 2 + (z − d)2 ]1/2 [x2 + y 2 + (z + d)2 ]1/2

The expression for the induced surface charge density is basically given by
the normal derivative evaluated at the surface
!
− 21 · 2(z − d) − 21 · 2(z + d)



∂Φ  q 
= −

∂z z=0 4π0 [x2 + y 2 + (z − d)2 ]3/2 [x2 + y 2 + (z + d)2 ]3/2 

z=0

45
46 CHAPTER 2.

F=0

-q z’ O z q
z
x
x-z’

Figure 2.1: Problem 2.1. The geometry of the situation.

0.0

-0.2

-0.2 -0.2

-0.4
-0.4 -0.4
2pd s/q
2

-0.6 -0.6
-0.6

-0.8 -0.8

-0.8
-1.0 -1.0

-1.0 4 4
2 2
0 1 2 3 4 5 0 0
-2 -2
r/d -4 -4

Figure 2.2: Problem 2.1. 2D and 3D plots of the induced surface charge density.

so the induced surface charge density σ is


∂Φ  q d q
  r 2 −3/2
σ = −0 = − = − 1 +
∂z z=0 2π [x2 + y 2 + d2 ]3/2 2πd2 d

with r 2 = x2 + y 2 the radial distance from the origin in the xy-plane. Figure 2.2
shows the desired plot.

Part b
The force between the plane and the point charge is the attractive force between
the two point charges placed at a distance 2d and from Coulomb’s law this is
simply
q2
F =
4 × 4π0 d2
 47

Part c
The force per unit area on the plane is σ 2 /20 , as derived in Section 1.11. The
total force is
σ2
Z
F = da
S 20
Z ∞ −3
q2 r2

= 2πr dr 1 + 2
8π 2 0 d4 0 d
2
q
=
4 × 4π0 d2
as in part b.

Part d
The work W1 needed to move q to infinity is (1.18)
Z ∞
W1 = F (z) dz
d
Z ∞
q2 dz
=
d 4 × 4π 0 z2
q2
=
4 × 4π0 d

Part e
The potential energy W2 between q and its image is
W2 = qΦ(z = d)
q2
 
1
= −
4π0 2d
2
q
= −
2 × 4π0 d
= −2W1
The magnitudes in parts d and e are not equal because in d the image charge
moved with q. Put in another way, when q is moved to infinity, we do work only
on q. It is true that induced charge is moving in over the conductor, but this
costs nothing, since the whole conductor is at zero potential. By contrast, if we
simultaneously move in two point charges (with no conductor), we do work on
both of them, and the total is twice as great.

Part f
With d=1 Å and q=-e the work needed to move q to infinity becomes
e2
W = = 3.6 eV
16π0 d
This is a typical value for a work function of a metal. As Jackson remarks at
the end of Sec. 2.3, a work function of at metal is in large part due to attraction
from an image force when an electron is removed from the surface.
48 CHAPTER 2.

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 fixed potential
V ? If the sphere has a total charge Q on its inner and outer surfaces?

Solution
This situation is very similar to the ones in Secs. 2.2–2.4 so we can take over
most of the results from these sections. Jackson is even kind enough to tell us
the correct results in the last paragraph of Sec. 2.2. With q inside the sphere,
the image charge is luckily outside the region of interest. The situation is similar
to that of Jackson’s Fig. 2.2, with q and q 0 exchanged.

Part a
The potential is given by
q/4π0 q 0 /4π0
Φ(x) = +
|x − y| |x − y0 |
It is only possible to have Φ(x = a) = 0 if
a a2
q 0 = − q, y0 =
y y
and thus the potential becomes
 
q 1 1
Φ(x) = − 2
4π0 (x2 + y 2 − 2xycosγ)1/2 (a + (xy/a)2 − 2xycosγ)1/2
with γ the angle between x and y. In this form, it is obvious that Φ = 0 when
x = a.

Part b
The induced surface charge density can be calculated from the normal derivative
of Φ at the surface. The normal derivative out of the conductor is now radially
inward:  
∂Φ  ∂Φ 
σ = −0 = + 0
∂n x=a ∂x x=a
This is exactly the same derivative as the one leading to (2.5); the only difference
is a sign change and now a > y so we rearrange to a more suggestive form
y2
q 1− a2
σ=−
4πa2  y2
3/2
1+ a2 − 2 ay cosγ
 49

Note that now the total induced charge is equal to −q, independent of y, as can
easily be verified.

Part c
The force acting on q is most easily computed by writing up the force between
q and its image. The distance between q and q 0 is y 0 − y = y(a2 /y 2 − 1). q is
attracted towards the conductor with a force of magnitude
−2  3  2 −2
a2 q2

1 1 a a
|F| = qq 0 2 −1 = − 1
4π0 y y2 4π0 a2 y y2

Part d
If the sphere is kept at a fixed potential V :

• The potential inside is raised by a constant V

• No change to the inner surface charge density, but a charge 4π0 V a is

uniformly distributed on the outer surface

• No change to the force

If the sphere has a total charge Q:

• The potential inside is raised by a constant equal to the potential of the

sphere

• No change to the inner surface charge density, but a charge Q + q is

uniformly distributed on the outer surface

• No change to the force

50 CHAPTER 2.

Problem 2.3
A straight-line charge with constant linear charge density λ is located perpen-
dicular to the x-y plane in the first 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, fields, and surface charges in the first
quadrant.
(a) The well-known potential for an isolated line charge at (x0 , y0 ) is Φ(x, y) =
(λ/4π0 ) ln(R2 /r2 ), where r 2 = (x − x0 )2 + (y − y0 )2 and R is a constant.
Determine the expression for the potential of the line charge in the pres-
ence of the intersecting planes. Verify explicitly that the potential and the
tangential electric field vanish on the boundary surface.
(b) Determine the surface charge density σ on the plane y = 0, x ≥ 0. Plot
σ/λ 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 x0
Qx = − λ tan−1
π y0
What is the total charge on the plane x = 0?
p
(d) Show
p that far from the origin [ρ  ρ0 , where ρ = x2 + y 2 and ρ0 =
x20 + y02 ] the leading term in the potential is
4λ (x0 y0 )(xy)
Φ → Φasym =
π0 ρ4
Interpret.

Solution
The setup of conductors and line charges is that of Fig. 2.3.

Part a
The potential in the first quadrant is the sum of potentials from the four line
charges, e.g. the contribution from the line charge in the second quadrant is
λ R2
Φ2 (x, y) = − ln
4π0 (x + x0 )2 + (y − y0 )2
and the total potential is:
Φ(x, y) = Φ 1 + Φ2 + Φ3 + Φ4
(x − x0 )2 + (y + y0 )2 (x + x0 )2 + (y − y0 )2
 
λ
= ln ·
4π0 (x − x0 )2 + (y − y0 )2 (x + x0 )2 + (y + y0 )2
In this form, it is easy to see that Φ(0, y) = Φ(x, 0) = 0. To verify that the
tangential electric field vanishes at the conductor surfaces, we need (for the
horizontal plane) 
∂Φ 
Ek (y = 0) = −
∂x  y=0
 51

-l l
y0

x0
x

l -l

Figure 2.3: Problem 2.3. The setup of conductors and line charges.

This derivative is done more easily if the logarithm is split into four terms. I
just quote the result:

∂Φ λ n
= − (x−x2(x−x
2
0)
0 ) +(y−y0 )
2 +
2(x+x0 )
(x+x0 )2 +(y−y0 )2
∂x 4π0
o
− (x+x2(x+x
2
0)
0 ) +(y+y0 )
2 +
2(x−x0 )
(x−x0 )2 +(y+y0 )2

Inserting y = 0 it is easy to see that terms 1,4 and 2,3 cancel. We have thus
verified that the tangential component of the electric field vanishes.

Part b
To get the induced surface charge density, we now need the normal derivative
evaluated at the surface (∂Φ/∂y|y=0 for the horizontal conductor). Again, this
is more easily done with four terms and proceeds similarly:

∂Φ 2λ n
y−y0 y−y0
=− (x−x0 )2 +(y−y0 )2 − (x+x0 )2 +(y−y0 )2
∂y 4π0
o
+ (x+x0 )y+y
2 +(y+y )2 −
0
0
y+y0
(x−x0 )2 +(y+y0 )2

The induced surface charge density for the horizontal conductor is then:
 
λ y0 y0
σ(x) = − −
π (x − x0 )2 + y02 (x + x0 )+ y02

Figure 2.4 shows the desired plot of the induced surface charge density. When
the line charge is placed symmetrically (x0 = 1, y0 = 1), of course the dis-
tribution of charge on the two conductors is the same. With an asymmetric
placement (x0 = 1, y0 = 2) of the line charge, the charge distribution on the
other plane is the distribution for (x0 = 2, y0 = 1).
52 CHAPTER 2.

0.00

-0.05

-0.10

s/l
-0.15

(x0=2,y0=1)
-0.20 (x0=1,y0=1)
(x0=1,y0=2)

-0.25

-0.30
0 1 2 3 4 5
x

Figure 2.4: Problem 2.3. Plots of the surface charge density on the horizontal
plane for selected sets of (x0 , y0 )

Part c
The total charge on the horizontal plane y = 0, x ≥ 0 is obtained by integrating
σ dx:
Z ∞
Qx = σ dx
0
Z ∞ Z ∞ 
y0 λ dt dt
= − 2 2 − 2 2
π −x0 t + y0 x0 t + y 0
 x0
λ t 
= − tan−1
π y0 −x0
 
2 x0
= − λ tan−1
π y0

The total charge on the other plane is, as noted in part b, the same expression
with x0 and y0 exchanged:
   
2 y0 2 x0
Qy = − λ tan−1 = − λ cot−1
π x0 π y0

Since tan−1 x + cot−1 x = π/2 we see that the total induced surface charge per
unit length in z on the conductor surface is always −λ, as it should be in order to
prevent the electric field from the line charge from penetrating the conductors.

Part d
We shall find the leading term for ρ  ρ0 in several steps. First consider the
expressions under the logarithm, e.g.:

(x − x0 )2 + (y + y0 )2 = x2 + x20 − 2xx0 + y 2 + y02 + 2yy0 ' x2 − 2xx0 + y 2 + 2yy0

 53

Inserting this in the expression for Φ(x, y) from part a yields

 4
ρ − 4(xx0 − yy0 )2

λ
Φ(x, y) ' ln
4π0 ρ4 − 4(xx0 + yy0 )2
" (  2 ) (  2 )#
λ xx0 − yy0 xx0 + yy0
= ln 1 − 4 − ln 1 − 4
4π0 ρ2 ρ2
 
λ 4 4
' − 4 (xx0 − yy0 )2 + 4 (xx0 + yy0 )2
4π0 ρ ρ
4λ (x0 y0 )(xy)
=
π0 ρ4
This is the far field from a quadrupole.
54 CHAPTER 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?
[Answers: (a) d/R − 1 = 0.6180, (b) F = −q 2 /(16π0 a2 ), i.e., image force, (c)
for Q = 2q, d/R − 1 = 0.4276; for Q = q/2, d/R − 1 = 0.8823. The answer for
part b is the same.]

Solution
Now we return to the situation of Sec. 2.3. Refer to Jackson’s Fig. 2.5 (showing
the force on the point charge) to get an overview of the situation. NB: the
asymptotic (Coulomb-like) dependence has been divided out in the figure.

Part a
All we need to do here is find the root d/R > 1 of Eq. 2.9:
qR3 (2d − R2 )
 
1 1
|F| = Q − =0
4π0 d2 d(d2 − R2 )2
and the force is attractive for values of d/R less than the root. Defining x ≡ d/R
this reduces to
Q 5 Q Q
x − 2 x3 − 2x2 + x + 1 = 0
q q q
For Q/q = 1 and with the restraint that x > 1 my pocket calculator found
x = 1.618034, i.e. for positions closer than d/R − 1 = 0.6180 to the surface
the charge q is attracted towards the sphere. This also fits well with Jackson’s
Figure 2.5.

Part b
When the charge is a small distance a = d − R from the surface of the sphere,
we can perform a Taylor expansion of the force expression (x = d/R = 1+a/R):
qR3 (2d − R2 )
 
1 1
|F| = Q −
4π0 d2 d(d2 − R2 )2
q2 1 1 2x2 − 1
 
= 1 −
4π0 R2 x5 (1 − x12 )2
q2 1
= −
4π0 (2a)2
i.e. the image force! This limiting value is independent of the radius of the
sphere R and the charge on the sphere Q.
 55

Part c
For Q/q = 2, we find from the equation in part a d/R − 1 = 0.4276 while for
Q/q = 1/2 we find d/R − 1 = 0.8823. As mentioned in part b, the limiting value
for the attractive force when q is very close to the sphere is independent of Q
so the answers for the force are the same as before.
56 CHAPTER 2.

Problem 2.5
(a) Show that the work done to remove the charge q from a distance r > a to
infinity against the force, Eq. (2.6), of a grounded conducting sphere is

q2 a
W =
8π0 (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
q2 a q 2 a qQ
 
1
W = − −
4π0 2(r2 − a2 ) 2r2 r
Relate the work to the electrostatic potential, Eq. (2.8), and the energy
discussion of Section 1.11.

Part a
The force attracting q towards the grounded sphere, i.e. opposite to the direction
of the position vector of q, is given by Eq. (2.6)
 3  −2
1 q2 a a2
F = 1− 2
4π0 a2 y y
Then the work required to move q to infinity is
Z ∞
W = + F dy
r
∞  3  −2
1 q2 a a2
Z
= dy 1− 2
4π0 a2 r y y

This integral is obviously performed by substituting x = 1 − a2 /y 2 and is oth-

erwise elementary. We get
q2 a
W =
8π0 (r2 − a2 )
We can compare this result with the energy required to disassemble the charges
by using Eq. (1.51) for the potential energy of a distribution of discrete charges
(which is the negative value of the energy required to disassemble the charges)
1 X X qi qj
W0 =
8π0 i j |xi − xj |

It is understood that i = j terms (infinite “self-energy” terms) are omitted in

the double sum. This gives two (equal) terms as follows
 2
q2 a

0 q aq 1
−W = = = 2W
4π0 r r(1 − a2 /r2 ) 4π0 (r2 − a2 )
 57

This is the same effect found in Problem 2.1. The reason for the unequal 1:2
magnitudes is again that in the former case, work is only done on one charge
(the “real” one) and the image charge is changing position and size continually.
In the latter case, we do work on both (constant magnitude) charges.

Part b
Now the force is (2.9)

qa3 (2 − y 2 − a2 )
 
1 q
F = Q−
4π0 y 2 y(y 2 − a2 )2

One could attack this expression head-on but it is easier to deconvolute the
force expression into its two constituents from the image charge q 0 = −aq/y
and the charge (Q − q 0 ) sitting at the center of the sphere. Then the work is
the expression from part a plus
Z ∞ 
q Q a
WQ−q0 = − + q dr
4π0 r r2 r3

This is very simple and we get

q2 a q 2 a qQ
 
1
W = − −
4π0 2(r2 − a2 ) 2r2 r

The first term is the one from part a while the two latter terms is the work
done against the attraction from the charge (Q − q 0 ) at the center of the sphere.
Comparing again to the work needed to disassemble the discrete (constant mag-
nitude) charges through Eq. (1.51) we get

q(Q + ar q)
 2  2
q 2 a qQ
 
1 q a 1 q a
−W 0 = − = − −
4π0 r2 − a2 r 4π0 r2 − a2 r2 r

For the same reasons as in part a, the two former terms are double the corre-
sponding ones in part a, but the last term is the same because Q on the sphere
is fixed in both cases.
58 CHAPTER 2.

Problem 2.6
The electrostatic problem of a point charge outside an isolated, charged con-
ducting sphere is equivalent to that of three charges, the original and two others,
one located at the center of the sphere and another (“the image charge”) inside
the now imaginary sphere, on the line joining the center and the original charge.
If the point charge and sphere are replaced by two conducting spheres of
radii ra and rb , carrying total charges Qa and Qb , respectively, with centers
separated by a distance d > ra + rb , there is an equivalence with an infinite
set of charges within each sphere, one at the center and a set of images along
the line joining the centers. The charges and their locations can be determined
iteratively, starting with a charge qa (1) at the center of the first sphere and
qb (1) correspondingly for the second sphere. The charge qb (1) has its image
qa (2) within the first sphere and vice versa. Then the image charge within the
first sphere induces another image within the second sphere, and so on. the sum
of all the charges within each sphere must be scaled to be equal to Qa and Qb .
The electrostatic potential outside the spheres, the force between the spheres,
etc. can be found by summing the contributions from all the charges.

(a) Show that the charges and their positions are determined iteratively by
the relations,

qa (j) = −ra qb (j − 1)/db (j − 1), xa (j) = ra2 /db (j − 1), da (j) = d − xa (j)
qb (j) = −rb qa (j − 1)/da (j − 1), xb (j) = rb2 /da (j − 1), db (j) = d − xb (j)

for j = 2, 3, 4, . . ., with da (1) = db (1) = d, and xa (1) = xb (1) = 0.

(b) Find the image charges and their locations as well as the potentials on the
spheres and force between them by means of a suitable computer program.
[In computing the potential on each sphere, evaluate it in different places:
e.g., in the equatorial plane and at the pole opposite the other sphere.
This permits a check on the equipotential of the conductor and on the
accuracy of computation.]

(c) As an example, show that for two equally charged spheres of the same
radius R, the force between them when almost in contact is 0.6189 times
the value that would be obtained if all the charge on each sphere were
concentrated at its center. Show numerically and by explicit summation
of the series that the capacitance of two identical conducting spheres in
contact is C/4π0 R = 1.3863 · · · [= ln 4]