Physics 505 Electricity and Magnetism Fall 2003

Prof. G. Raithel

Problem Set 5

Problem 1 7 Points

A cylindrical surface with radius a and infinite length in the z-direction carries a surface potential V (z, φ).
Find a series solution for the potential anywhere outside the cylinder. Provide an integral solution for the
expansion coefficients. There are no charges in the volume of interest.

The Laplace equation ∆Φ(ρ, z, φ) = 0 is to be solved for the boundary condition V (z, φ) on a cylinder with
radius a and infinite length in both the +z- and −z-directions. We seek the solution in the exterior volume
∂2 1 ∂ 1 ∂2 ∂2
ρ > a. Using ∆ = ∂ρ 2 + ρ ∂ρ + ρ2 ∂φ2 + ∂z 2 and Φ(ρ, z, φ) = R(ρ)Q(φ)Z(z), we find

∂2 1 ∂ 1 ∂2 ∂2
Φ + Φ + Φ + Φ = 0
∂ρ2 ρ ∂ρ ρ2 ∂φ2 ∂z 2
1 00 1 0 1 Q00 Z 00
R + R + 2 + = 0 ,
R ρR ρ Q Z

where in the second line we have divided by RQZ and used 0 for derivatives. Since we are looking for
solutions that are orthogonal on the cylinder mantle, the z-dependence is chosen to be of the form

Z 00 = −k 2 Z ⇒ Z(z) = exp(ikz) with k real .

It follows

ρ2 00 ρ Q00
R + R0 + − k 2 ρ2 = 0 .
R R Q

Considering the fact that the potential must be a single-valued function of φ, and that there are no boundaries
in the φ-coordinate, the solution for Q is

Q00 = −m2 Q ⇒ Q(φ) = exp(imφ) with m integer .

The resultant equation for R,

µ ¶
1 m2
R00 + R0 − k 2 + 2 R = 0
ρ ρ
is solved by the modified Bessel functions Im (|k|ρ) and Km (|k|ρ). Since we are interested in the exterior
volume, we only use the solution that is regular for ρ → ∞; that solution is Km (|k|ρ) (see Eqs. 3.98-3.104
in the textbook). Note that Km (x) = K−m (x).
P∞ R∞
To summarize, Φ(ρ, z, φ) = m=−∞ k=−∞ Am (k) Km (|k|ρ) exp(ikz) exp(imφ) dk .

To find the coefficient functions Am (k), we consider this expression on the boundary,

∞
X Z ∞
V (z, φ) = Am (k) Km (|k|a) exp(ikz) exp(imφ) dk ,
m=−∞ k=−∞

multiply with exp(−ik 0 z) exp(−im0 φ) and integrate over the cylinder mantle,

Z ∞ Z 2π
V (z, φ) exp(−ik 0 z) exp(−im0 φ) dz dφ =
z=−∞ φ=0
X∞ Z ∞ Z ∞ Z 2π
= Am (k) Km (|k|a) exp(ikz) exp(imφ) exp(−ik 0 z) exp(−im0 φ) dz dφ dk ⇔
m=−∞ k=−∞ z=−∞ φ=0

Z ∞ Z 2π
V (z, φ) exp(−ik 0 z) exp(−im0 φ) dz dφ =
z=−∞ φ=0
X∞ Z ∞
Am (k) Km (|k|a) 4π 2 δm,m0 δ(k − k 0 ) dk ⇔
m=−∞ k=−∞

Z ∞ Z 2π
V (z, φ) exp(−ik 0 z) exp(−im0 φ) dz dφ = Am0 (k 0 ) Km0 (|k 0 |a) 4π 2 .
z=−∞ φ=0

1
R∞ R 2π
Thus, it is Am (k) = 4π 2 Km (|k|a) z=−∞ φ=0
V (z, φ) exp(−ikz) exp(−imφ) dz dφ .
Problem 3.17 9 Points

To demonstrate the general method, this problem is worked out in a verbose manner.

Preparation: We use the completeness (=closure) relations

∞
0 1 X
δ(φ − φ ) = exp(−imφ0 ) exp(imφ)
2π m=−∞
∞ ³ nπz ´ µ ¶
2 X nπz 0
δ(z − z 0 ) = sin sin
L n=1 L L
Z ∞
δ(ρ − ρ0 )
= kJm (kρ0 )Jm (kρ)dk ∀ m , (1)
ρ0 k=0

where the last one is obtained by treating Eq. 3.108 in the textbook with the replacements explained after
Eq. 3.112. In doubt, for any complete set of functions with given orthogonality relation the appropriate
closure relation can be obtained using the procedure outlined in class. The procedure is demonstrated again
in the following:

Completeness of the set {Jm (kρ), k ∈ [0, ∞[ , m fixed } ⇒ any f (ρ) can be expanded as

Z ∞
f (ρ) = A(k)Jm (kρ)dk (2)
k=0

Obtain expansion coefficient function A(k) using the orthogonality relation Eq. 3.108 of the textbook:

Z ∞ Z ∞ Z ∞
f (ρ)Jm (k 0 ρ)ρdρ = A(k) Jm (kρ)Jm (k 0 ρ)ρdρdk
ρ=0 k=0 ρ=0

Z ∞ Z ∞
0 1
f (ρ)Jm (k ρ)ρdρ = A(k) δ(k − k 0 )dk
ρ=0 k=0 k

Z ∞
A(k 0 ) = k 0 f (ρ)Jm (kρ)ρdρ
ρ=0

To simplify writing in what follows, swap primes from k to ρ:

Z ∞
A(k) = k f (ρ0 )Jm (kρ0 )ρ0 dρ0
ρ=0

Insert into Eq. 2:

Z ∞ Z ∞
f (ρ) = k f (ρ0 )Jm (kρ0 )ρ0 dρ0 Jm (kρ)dk
k=0 ρ=0
 Z ∞ ½Z ∞ ¾
f (ρ) = kJm (kρ )Jm (kρ)dk ρ0 f (ρ0 )dρ0
0
ρ=0 k=0

From the last line, it is obvious that

Z ∞
δ(ρ − ρ0 )
= kJm (kρ0 )Jm (kρ)dk ∀m
ρ0 k=0

a): Obtain given expansion of Green’s function.

Step 1: Write down Equation for Green’s function with δ-function in cylindrical coordinates,

δ(ρ − ρ0 )
∆G(x, x0 ) = −4πδ(x − x0 ) = −4π δ(φ − φ0 )δ(z − z 0 )
ρ0

Step 2: On right side, use completeness relations for two out of the three δ-functions. From the given result,
we suspect that we need to use the completeness relations for δ(z − z 0 ) and δ(φ − φ0 ):

∞ ∞ µ ¶ ³ nπz ´ δ(ρ − ρ0 )
4 X X nπz 0
∆G(x, x0 ) = − exp(−imφ0 ) exp(imφ) sin sin
L m=−∞ n=1 L L ρ0

∞
X X∞ ½ µ ¶ ¾ ³ nπz ´
0 4 0 nπz 0 δ(ρ − ρ0 )
∆G(x, x ) = − exp(−imφ ) sin exp(imφ) sin (3)
m=−∞ n=1
L L ρ0 L

Step 3: On left side, expand the Green’s function using the orthogonal function sets that have also been used
in Step 2. Note that x0 only enters as a parameter of the calculation; ∆ acts on x.

∞
X ∞
X ³ nπz ´
∆G(x, x0 ) = ∆ exp(imφ) sin Amn (ρ|ρ0 , z 0 , φ0 )
m=−∞ n=1
L

Step 4: Write Laplacian in the proper coordinates and take derivatives of the orthogonal functions:

∞
X X∞ ½· ¸ ¾ ³ nπz ´
1 d d m2 n2 π 2
∆G(x, x0 ) = ρ − 2 − A mn (ρ|ρ 0 0
, z , φ 0
) exp(imφ) sin (4)
m=−∞ n=1
ρ dρ dρ ρ L2 L

Step 5: Expansions in orthogonal sets of functions are unique. Thus, we can separately equate the co-
¡ ¢ ³ ´
nπz 0
efficients of the exp(imφ) sin nπz in Eq. 3 and Eq. 4. Dividing by exp(−imφ 0
) sin and using
h 2 2 2
i L L

Dρ = ρ1 dρ d d
ρ dρ −m n π
ρ2 − L2 as abbreviation for the involved linear differential operator in ρ, we find

( )
Amn (ρ|ρ0 , z 0 , φ0 ) 4 δ(ρ − ρ0 )
Dρ ¡ 0¢ =−
exp(−imφ0 ) sin nπz L
L ρ0
Step 6: Noticing that the expression in the curly brackets of the last equation can only depend on the
parameters of Dρ (which are m and n), on ρ and on ρ0 , we define the reduced Green’s function

Amn (ρ|ρ0 , z 0 , φ0 )
gmn (ρ, ρ0 ) = ¡ 0¢
exp(−imφ0 ) sin nπz L

and proceed to solve

· ¸
1 ∂ ∂ m2 n2 π 2 0 4 δ(ρ − ρ0 )
ρ − 2 − gmn (ρ, ρ ) = − (5)
ρ ∂ρ ∂ρ ρ L2 L ρ0

Inspection of Eq. 3.98ff of the textbook shows that the differential equation is the modified Bessel differ-
ential equation (except the δ-function inhomogeneity), and that the linearly independent solutions of the
homogeneous equation are Im (kρ) and Km (kρ), with k = nπ 0
L . In the domain ρ < ρ only Im (kρ) is regular,
while in the domain ρ > ρ0 only Km (kρ) is regular. Thus, for the reduced Green’s function to be regular,
symmetric in ρ and ρ0 , and continuous at ρ = ρ0 , it must be of the form

gmn (ρ, ρ0 ) = C Im (kρ< ) Km (kρ> ) ,

with a constant C to be determined to match the inhomogeneity. Also, ρ< = min(ρ, ρ0 ) and ρ> = max(ρ, ρ0 ).

Step 7: Find C. Integrating Eq. 5 from ρ0 − ² to ρ0 + ² with ² → 0 and dropping vanishing terms we find

d d 4
gmn (ρ, ρ0 )|ρ=ρ0 +² − gmn (ρ, ρ0 )|ρ=ρ0 −² = − 0
dρ dρ ρL

The left side equals C k W [Im (x), Km (x)] with x = kρ0 and the Wronski determinant W =
¡d ¢ ¡d ¢
Im (x) dx Km (x) − dx Im (x) Km (x). The Wronsky determinant can be evaluated in the asymptotic re-
gion, where both Im (x) and Km (x) have simple forms given in Eqs. 3.102ff of the textbook, leading to
W [Im (x), Km (x)] = − x1 (see Eq. 3.147 in textbook). We thus find

1 1 4
−Ck = −Ck 0 = − 0
x kρ ρL

4
and therfore C = L.

Step 8:

Going backward through all steps, the Green’s function expansion is assembled into the final result:

4
gmn (ρ, ρ0 ) = Im (kρ< ) Km (kρ> ) ,
L

µ ¶
4 nπz 0
Amn (ρ|ρ , z , φ ) = Im (kρ< ) Km (kρ> ) exp(−imφ0 ) sin
0 0 0
L L
 nπ
and, using k = L ,

∞ ∞ µ ¶ ³ nπz ´
0 4 X X 0 nπz 0 nπ nπ
G(x, x ) = exp(−imφ ) exp(imφ) sin sin Im ( ρ< ) Km ( ρ> ) , q.e.d.
L m=−∞ n=1 L L L L

b:

Step 1: Write down Equation for Green’s function with δ-function in cylindrical coordinates,

δ(ρ − ρ0 )
∆G(x, x0 ) = −4πδ(x − x0 ) = −4π δ(φ − φ0 )δ(z − z 0 )
ρ0

δ(ρ−ρ0 )
Step 2: Use completeness relations for ρ0 and δ(φ − φ0 ):

∞
X Z ∞
0
∆G(x, x ) = −2 exp(−imφ0 ) exp(imφ) Jm (kρ) Jm (kρ0 )δ(z − z 0 ) k dk
m=−∞ k=0

∞
X Z ∞
0
∆G(x, x ) = {−2k exp(−imφ0 ) Jm (kρ0 )δ(z − z 0 )} exp(imφ) Jm (kρ) dk (6)
m=−∞ k=0

Step 3: Expand the Green’s function using the orthogonal function sets that have also been used in Step 2.

∞
X Z ∞
∆G(x, x0 ) = ∆ exp(imφ) Jm (kρ) Amk (z|ρ0 , z 0 , φ0 )dk
m=−∞ k=0

Step 4: Write Laplacian in the proper coordinates and take the obvious derivatives of the orthogonal func-
tions:

∞
X Z ∞ · ¸
∂2 m2 1 ∂ ∂
∆G(x, x0 ) = − + ρ exp(imφ) Jm (kρ) Amk (z|ρ0 , z 0 , φ0 )dk
m=−∞ k=0 ∂z 2 ρ2 ρ ∂ρ ∂ρ

h 2
i
The ordinary Bessel diff. equation reads − m
ρ2 +
1 ∂ ∂
ρ ∂ρ ρ ∂ρ Jm (kρ) = −k 2 Jm (kρ), and thus:

∞
X Z ∞ ½· ¸ ¾
d2
∆G(x, x0 ) = − k 2
A mk (z|ρ 0 0
, z , φ 0
) exp(imφ) Jm (kρ)dk (7)
m=−∞ k=0 dz 2

Step 5: We equate coefficients in Eqs. 6 and Eq. 7. Dividing by exp(−imφ0 ) Jm (kρ0 ) we find

· ¸½ ¾
d2 Amk (z|ρ0 , z 0 , φ0 )
− k2 = −2k δ(z − z 0 )
dz 2 exp(−imφ0 ) Jm (kρ0 )

Step 6: Noting that the expression in the curly brackets of the last equation can only depend on k, z and z 0 ,
we define the reduced Green’s function
 Amk (z|ρ0 , z 0 , φ0 )
gk (z, z 0 ) =
exp(−imφ0 ) Jm (kρ0 )

and proceed to solve

· ¸
d2 2
− k gk (z, z 0 ) = −2k δ(z − z 0 )
dz 2

A reduced Green’s function that matches the boundary conditions at z = 0 and z = L, that is symmetric in
z and z 0 and that is continuous at z = z 0 must be of the form

gk (z, z 0 ) = C sinh(k z< ) sinh(k (L − z> )) ,

where the constant C needs to be determined to match the inhomogeneity at z = z 0 . Also, z< = min(z, z 0 )
and z> = max(z, z 0 ).

Step 7: Find C. Integrating the differential equation for gk (z, z 0 ) from z 0 −² to z 0 +² with ² → 0 and dropping
vanishing terms we find

d d
gk (z, z 0 )|z=z0 +² − gk (z, z 0 )|z=z0 −² = −2k
dz dz

Direct calculation yields

d d
gk (z, z 0 )|z=z0 +² − gk (z, z 0 )|z=z0 −² = −kC [sinh(kz 0 ) cosh(k (L − z> )) + cosh(kz 0 ) sinh(k (L − z> ))]
dz dz
= −kC sinh(kL)

2
and thus, with the equation before, C = sinh(kL) .

Step 8:

Going backward through all steps, the Greens function expansion is assembled into the final result

∞
X Z ∞
sinh(k z< ) sinh(k (L − z> ))
G(x, x0 ) = 2 exp(−imφ0 ) exp(imφ) Jm (kρ) Jm (kρ0 ) dk , q.e.d.
m=−∞ k=0 sinh(kL)
Problem 3.23 9 Points

Hint: Use Green’s function expansion techniques. For the second line, note Eq. 3.147.

We use the completeness relations

∞
1 X
δ(φ − φ0 ) = exp(−imφ0 ) exp(imφ)
2π m=−∞
∞ ³ nπz ´ µ ¶
2 X nπz 0
δ(z − z 0 ) = sin sin
L n=1 L L
∞
X
δ(ρ − ρ0 ) 2 ρ ρ0
= 2 J m (x mn )J m (xmn ) ∀m . (8)
ρ0 n=1
a2 Jm+1 (xmn ) a a

a):

Step 1: Write down Equation for Green’s function with δ-function in cylindrical coordinates,

δ(ρ − ρ0 )
∆G(x, x0 ) = −4πδ(x − x0 ) = −4π δ(φ − φ0 )δ(z − z 0 )
ρ0

δ(ρ−ρ0 )
Step 2: We use the completeness relations for δ(φ − φ0 ) and ρ0 to write:

∞
X ∞ ½
X ¾
4 ρ0 ρ
∆G(x, x0 ) = − 2 exp(−imφ0 ) Jm (xmn ) δ(z − z 0 ) exp(imφ) Jm (xmn ) (9)
m=−∞ n=1
a2 Jm+1 (xmn ) a a

Step 3: Expansion of the Green’s function in exp(imφ) Jm (xmn aρ ):

∞
X ∞
X xmn
∆G(x, x0 ) = ∆ exp(imφ) Jm ( ρ) Amn (z|ρ0 , z 0 , φ0 )
m=−∞ n=1
a

Step 4: Apply Laplacian:

∞
X X∞ · 2 ¸
∂ m2 1 ∂ ∂ xmn
∆G(x, x0 ) = 2
− 2
+ ρ exp(imφ) Jm ( ρ) Amn (z|ρ0 , z 0 , φ0 )
m=−∞ n=1
∂z ρ ρ ∂ρ ∂ρ a

h i
m2
The ordinary Bessel differential equation reads, in the present case, 1 ∂ ∂
ρ ∂ρ ρ ∂ρ − ρ2 Jm ( xmn
a ρ) =
¡ ¢2
− xmn
a Jm ( xmn
a ρ), and thus:

∞
X X∞ ½· 2 ³ x ´2 ¸ ¾
0 d mn 0 0 0 xmn
∆G(x, x ) = 2
− Amn (z|ρ , z , φ ) exp(imφ) Jm ( ρ) (10)
m=−∞ n=1
dz a a

4
Step 5: We equate coefficients in Eqs. 9 and Eq. 10. Dividing by − a2 J 2 (xmn )
exp(−imφ0 ) Jm (kρ0 ) yields
m+1
 · ³ x ´2 ¸
d2 mn
− gmn (z, z 0 ) = δ(z − z 0 )
dz 2 a

with the reduced Green’s function

a2 Jm+1
2
(xmn ) Amn (z|ρ0 , z 0 , φ0 )
gmn (z, z 0 ) = −
4 exp(−imφ0 ) Jm ( xmn
a ρ )
0

Step 6: A reduced Green’s function that matches the boundary conditions at z = 0 and z = L, that is
symmetric in z and z 0 and that is continuous at z = z 0 must be of the form

xmn xmn
gmn (z, z 0 ) = C sinh( z< ) sinh( (L − z> )) ,
a a

where the constant C needs to be determined to match the inhomogeneity at z = z 0 . Also, z< = min(z, z 0 )
and z> = max(z, z 0 ).

Step 7: Find C. Integrating the differential equation for gmn (z, z 0 ) from z 0 − ² to z 0 + ² with ² → 0 and
dropping vanishing terms we find

d d
gmn (z, z 0 )|z=z0 +² − gmn (z, z 0 )|z=z0 −² = 1
dz dz

Direct calculation similar to Problem 3.17 b) yields

a
C=− .
xmn sinh( xmn L
a )

Step 8:

Going backward through all steps, the Greens function expansion is assembled into

∞ ∞
04 X X xmn xmn 0 sinh( xmn xmn
a z< ) sinh( a (L − z> ))
G(x, x ) = exp(−imφ0 ) exp(imφ) Jm ( ρ) Jm ( ρ) .
a m=−∞ n=1 a a 2
xmn Jm+1 (xmn ) sinh( xmn L
a )

The potential of a point charge q at x0 follows from its charge density, ρ(x00 ) = qδ(x00 − x0 ), and

Z Z
0 1 00 00q 3 00 q
Φ(x, x ) = ρ(x )G(x, x )d x = δ(x00 − x0 )G(x, x00 )d3 x00 = G(x, x0 )
4π²0 V 4π²0 V 4π²0

Thus, with the above obtained G(x, x0 ) it is

∞
X X∞
q xmn xmn 0 sinh( xmn xmn
a z< ) sinh( a (L − z> ))
Φ(x, x0 ) = exp(−imφ0 ) exp(imφ) Jm ( ρ) Jm ( ρ) , q.e.d.
aπ²0 m=−∞ n=1 a a 2
xmn Jm+1 (xmn ) sinh( xmn L
a )
b):

Step 1: Write down Equation for Green’s function with δ-function in cylindrical coordinates,

δ(ρ − ρ0 )
∆G(x, x0 ) = −4πδ(x − x0 ) = −4π δ(φ − φ0 )δ(z − z 0 )
ρ0

Step 2: We use the completeness relations for δ(φ − φ0 ) and δ(z − z 0 ) to write:

∞
X X∞ ½ µ ¶ ¾ ³ nπz ´
0 4 0 nπz 0 δ(ρ − ρ0 )
∆G(x, x ) = − exp(−imφ ) sin exp(imφ) sin (11)
m=−∞ n=1
L L ρ0 L

¡ nπz ¢
Step 3: Expansion of the Green’s function in exp(imφ) sin L :

∞
X ∞
X ³ nπz ´
∆G(x, x0 ) = ∆ exp(imφ) sin Amn (ρ|ρ0 , z 0 , φ0 )
m=−∞ n=1
L

Step 4: Apply Laplacian:

∞
X X∞ ½· ³ nπ ´2 m2 ¸ ¾ ³ nπz ´
1 d d
∆G(x, x0 ) = ρ − − 2 Amn (ρ|ρ0 , z 0 , φ0 ) exp(imφ) sin (12)
m=−∞ n=1
ρ dρ dρ L ρ L

³ ´
nπz 0
Step 5: We equate coefficients in Eqs. 11 and Eq. 12, and divide by − L4 exp(−imφ0 ) sin L . We find

· ³ nπ ´2 m2 ¸
1 d d δ(ρ − ρ0 )
ρ − − 2 gmn (ρ, ρ0 ) =
ρ dρ dρ L ρ ρ0

for the reduced Green’s function

L Amn (ρ|ρ0 , z 0 , φ0 )
gmn (ρ, ρ0 ) = − ¡ 0¢
4 exp(−imφ0 ) sin nπz L

Step 6: The differential equation for gmn is the modified Bessel equation (except the inhomogeneity). Thus,
a reduced Green’s function that solves the homogeneous equation, is symmetric in ρ and ρ0 , is continuous at
ρ = ρ0 , is regular at ρ = 0 and is vanishing at ρ = a must be of the form

µ ¶
nπ nπ Im ( nπa
L ) nπ
gmn (ρ, ρ0 ) = C Im ( ρ< ) I m ( ρ> ) − K m ( ρ > )
L L Km ( nπa
L ) L

where the constant C needs to be determined to match the inhomogeneity at ρ = ρ0 . Also, ρ< = min(ρ, ρ0 )
and ρ> = max(ρ, ρ0 ).

Step 7: Find C. Integrating the differential equation for gmn (ρ, ρ0 ) from ρ0 − ² to ρ0 + ² with ² → 0 and
dropping vanishing terms we find
 d d 1
gmn (ρ, ρ0 )|ρ=ρ0 +² − gmn (ρ, ρ0 )|ρ=ρ0 −² = 0
dρ dρ ρ

Straightforward calculation yields

nπ Im ( nπa
L )
1 = −Cρ0 0 0
nπa [Im (x)Km (x) − Im (x)Km (x)|x= nπ 0
L Km ( L ) L ρ

Km ( nπa
L )
C =
Im ( nπa
L )

where in the second line Eq. 3.147 has been used.

Step 8:

Going backward through all steps, the Greens function expansion is assembled into

∞ ∞ ½ ³ nπz ´ µ ¶
4 X X Im ( nπρnπz 0
L )
<

G(x, x0 ) = exp(imφ) exp(−imφ0 ) sin sin nπa

L m=−∞ n=1 L Im ( L )L
h nπa nπρ> nπa nπρ> oi
Im ( )Km ( ) − Km ( )Im ( ) .
L L L L

q
The potential of a point charge q at x0 follows from Φ(x, x0 ) = 0
4π²0 G(x, x ), leading to

∞
X X∞ ½ ³ nπz ´ µ ¶
0 q 0 Im ( nπρ
nπz 0
L )
<

Φ(x, x ) = exp(imφ) exp(−imφ ) sin sin

Lπ²0 m=−∞ n=1 L Im ( nπa
L )
L
h nπa nπρ> nπa nπρ> i o
Im ( )Km ( ) − Km ( )Im ( ) q.e.d. .
L L L L

c):

The equation (∆+λ)Φ = 0 for the given boundary conditions has the following complete set of orthonormal
eigenfunctions (∗ see addendum below)

r µ ¶ √ ³x ρ´
1 2 kπz 2 mn
Φkmn = √ exp(imφ) sin Jm .
2π L L aJm+1 (xmn ) a

Using the Bessel differential equation, it is seen that

µ ¶
1 ∂ ∂ 1 ∂2 ∂2
∆Φkmn = ρ + + 2 Φkmn
ρ ∂ρ ∂ρ ρ2 ∂φ2 ∂z
à µ ¶2 !
1 ∂ ∂ kπ m2
= ρ − − 2 Φkmn
ρ ∂ρ ∂ρ L ρ
 Ã µ ¶2 !
m2 x2mn kπ m2
= − 2 − − 2 Φkmn
ρ2 a L ρ
à µ ¶2 !
x2 kπ
= − mn − Φkmn = −λkmn Φkmn , .
a2 L

x2mn ¡ kπ ¢2
The eigenvalue λkmn of the eigenfunction Φkmn is λkmn = a2 + L . Increasing the number of indices
in Eq. 3.160 of the textbook and setting λ = 0, we find

¡ ¢ ³ 0´ ¡ xmn ρ ¢ ³ ´
∞ exp(im(φ − φ0 )) sin kπz sin kπz xmn ρ0
8
∞
X X∞ X
L L Jm a Jm a
G(x, x0 ) = ³¡ ¢ ¡ kπ ¢2 ´ 2
La2 m=−∞ n=1 xmn 2
+ L Jm+1 (xmn )
k=1 a

q
and for the potential Φ(x, x0 ) of a point charge q at x0 it follows from Φ(x, x0 ) = 0
4π²0 G(x, x ) that

¡ ¢ ³ 0´ ¡ xmn ρ ¢ ³ ´
∞ exp(im(φ − φ0 )) sin kπz sin kπz xmn ρ0
2q
∞
X X∞ X
L L Jm a J m a
Φ(x, x0 ) = ³¡ ¢ ¡ kπ ¢2 ´ 2 q.e.d.
La2 π²0 m=−∞ n=1 xmn 2
+ L Jm+1 (xmn )
k=1 a

Discussion. By the uniqueness theorem, all solutions must be identical. Therefore, some hard-to-come-by
sum rules can be extracted. The comparison of a) with c) yields a discrete Fourier series expansion of the
reduced Green’s function in z:

a sinh( xmn xmn

a z< ) sinh( a (L − z> ))
gmn (z, z 0 ) = −
xmn sinh( xmn L
a )
¡ ¢ ³ 0
´
X∞ sin kπz sin kπz
2 L L
= − ¡ x ¢2 ¡ kπ ¢2
L mn
+ L
k=1 a
for 0 < z, z 0 < L and m = 0, ±1, ±2, ... and n = 1, 2, 3, ...

The comparison of b) with c) yields a discrete Bessel series expansion of the radial reduced Green’s function:

Im ( nπρ h nπρ> i
L ) nπa nπρ> nπa
<

gmn (ρ, ρ0 ) = − I m ( )K m ( ) − K m ( )I m ( )
Im ( nπa
L ) L L L L
¡ ¢ ³ ´
ρ0
2 X
∞ Jm xmk a
ρ
Jm xmk a
= − 2 ³¡ ¢ ¡ nπ ¢2 ´ 2
a xmk 2
+ Jm+1 (xmk )
k=1 a L

for 0 < ρ, ρ0 < a and m = 0, ±1, ±2, ... and n = 1, 2, 3, ...

Addendum. If not known or guessed from the statement of the problem, the eigenfunctions Φ (and
corresponding eigenvalues λ) of ∆Φ + λ = 0 for the given boundary conditions are obtained as follows.
Writing Φ = R(ρ)Q(φ)Z(z) it is
 ∂2 1 ∂ 1 ∂2 ∂2
Φ + Φ + Φ + Φ + λΦ = 0
∂ρ2 ρ ∂ρ ρ2 ∂φ2 ∂z 2
1 00 1 0 1 Q00 Z 00
R + R + 2 + +λ = 0 ,
R ρR ρ Q Z

0
where in the second line we have divided by RQZ and used for derivatives. To match the boundary
conditions at z = 0 and L, for the z-dependence we choose

µ ¶
kπ
Z(z) = sin z with k = 1, 2, ... .
L

¡ kπ ¢2
Then, Z 00 /Z = − L , and it follows

" µ ¶2 #
ρ2 00 ρ 0 Q00 kπ
R + R + + λ− ρ2 = 0 .
R R Q L

Considering the fact that the eigenfunction must be single-valued functions of φ, the solution for Q is

Q(φ) = exp(imφ) with m integer .

Then, Q00 /Q = −m2 , and the remaining equation for R,

" µ ¶2 #
2
1 kπ m
R00 + R0 + λ − − 2 R=0
ρ L ρ

¡ kπ ¢2 q
is the standard Bessel equation (Eq. 3.75 in textbook) with k replaced by L . Since the eigenfunc- λ−
µ q ¶
¡ ¢2
tions must be regular at ρ = 0 and zero at ρ = a, the radial dependence is Jm ρ λ − kπ L with an
eigenvalue λ such that

s µ ¶2
kπ
a λ− = xmn .
L

x2mn ¡ kπ ¢2
Thus, λkmn = a2 + L , and the corresponding eigenfunction is

µ ¶ ³x ´
kπz mn ρ
Φkmn = constant × exp(imφ) sin Jm .
L a

R
The normalization constant follow from Φ∗kmn (ρ, φ, z)Φkmn (ρ, φ, z) ρdρdφdz = 1 and Eq. 3.95 of the text-
book.