4 JACOB LEWIS BOURJAILY

18. a) Using the results obtained in problem (2.17) above, we are to find a closed-form expression for
the Green’s function for the interior of a cylinder of radius b with Dirichlet boundary conditions.
From our general work with solutions to the Laplace equation with Green’s functions
and from our work about in problem (2.17), we know that the Green’s function for
the interior of the cylinder will be given by
∞
1 X im(φ−φ0 )
G(ρ, ρ0 ) = e gm (ρ, ρ0 ),
2π m=−∞

where the functions gm (ρ, ρ0 ) satisfy the Laplace equation and will be given in the
form ½
αm ρ0m ρ0 < ρ
gm = 0m 0−m ,
βm ρ + γm ρ ρ0 > ρ
where the coefficients αm , βm , and γm will be determined by the boundary conditions
and continuity requirements.
Specifically, because of Dirichlet boundary conditions are satisfied at ρ0 = b, we must
have that gm = 0, we must demand
βm bm + γm b−m = 0,
which implies
βm = ηm b−m and γm = −ηm bm ,
for some set of coefficients ηm . Similarly, continuity requires that the two solutions
agree when ρ = ρ0 and so
µ³ ´ µ ¶m ¶
m m −m ρ m b
αm ρ = βm ρ + γm ρ = ηm − ,
b ρ
which implies that
µ µ ¶m ¶
η m ³ ρ ´m b
αm = m − .
ρ b ρ
Risking redundancy, we have shown that
 ³ ´m ³¡ ¢ ³ ´m ´
 ηm ρ0 ρ m
ρ b − ρb ρ0 < ρ
gm = ³¡ ¢ ³ ´m ´
 ηm ρ m − b ρ0 > ρ
b ρ

Let us now proceed to determine the coefficients ηm . These can be found by noting the
discontinuity in the derivative of gm . Specifically, we have that
¯ ¯
4π dgm ¯¯ dgm ¯¯
− = − ,
ρ dρ0 ¯ρ0 ↓=ρ dρ0 ¯ρ0 ↑=ρ
µ µ ¶m ¶
ρm−1 bm ρm−1 ³ ρ ´m b
= mηm m + mηm m+1 − mηm m − ,
b ρ ρ b ρ
2π ³ ρ ´m
∴ ηm = − .
m b
Plugging this into our expression for gm , we see that
 ¡ ρ ¢m ³ ρ0 ´m ³¡ ρ ¢m ³ b ´m ´ ³³ 0 ´m ³ 0 ´m ´ 
 − 2π  µ µ ¶m ¶
2π ³ ρ+ ρ− ´m
2π ρρ ρ 0
m b ρ b − ρ = − m b2 − ρ ρ < ρ ρ−
gm = ³
¡ ρ ¢m ¡ ρ ¢m ³ ´ m ´ ³³ ´m ³ ´ m ´ = − − ,
 − 2π b 2π ρρ0 ρ 0  m b 2 ρ+
m b b − ρ =−
m b 2 ρ− 0 ρ >ρ

Let us now determine the m = 0 coefficients. We have demonstrated that, in general,

g0 = α0 + β0 log ρ0 . Because g0 (b) must vanish by the Dirichlet boundary conditions,
we see that α0 + β0 log b = 0] which immediately implies
α0 = β0 log(1/b).
Like we found in problem (2.17), we see that β0 = −4π. Therefore, we are now ready
to write the explicit Green’s function.
 PHYSICS 505: CLASSICAL ELECTRODYNAMICS HOMEWORK 3 5

Putting this into our expression for the Green’s function, we obtain

X∞ µµ ¶m µ 0 ¶m ¶
¡ ¢ 1 ρ− ρρ
G(ρ, ρ0 ) = log b2 − log(ρ2+ ) + 2 − 2
cos [m(φ − φ0 )] ,
m=1
m ρ + b
X∞ µ ¶m X∞ µ ¶m
¡ 2¢ 1 ρ− 1 ρρ0
= log b − log(ρ2+ ) + 2 cos [m(φ − φ0 )] − 2 2
cos [m(φ − φ0 )] .
m=1
m ρ + m=1
m b

We will evaluate the two sums separately, but using the same technique. With some
inspiration from Davis’ Fourier Series and Orthogonal Functions, we note that
Z x Z x
1 m 1 m
x = ξ m−1 dξ = ξ dξ.
m 0 0 ξ

Therefore, substituting (ρ− /ρ+ ) 7→ x in the above expressions, we can reexpress the
first series as
∞
X µ ¶m Z ρρ− Ã X ∞
!
1 ρ− 0 + 1
m 0
2 cos [m(φ − φ )] = 2 ξ cos [m(φ − φ )] dξ,
m=1
m ρ+ 0 ξ m=1

Noting the similar expression used during the solution of problem (2.12) and recalling
that |ξ| < 1, we see that

1 X m ³ im(φ−φ0 ) ´
∞
X ∞
0
m 0
ξ cos [m(φ − φ )] = ξ e + e−im(φ−φ ) ,
m=1
2 m=1
µ ¶
1 1 1
= + − 2 ,
2 1 − ξei(φ−φ0 ) 1 − ξe−i(φ−φ0 )
 ³ ´ ³ ´
i(φ−φ0 ) 0 0 0

1 2 − ξ e + e−i(φ−φ ) − 2 − 2ξ 2 + 2ξ ei(φ−φ ) + e−i(φ−φ )

= ¡ ¢ ,
2 1 + ξ 2 − ξ ei(φ−φ0 ) + e−i(φ−φ0 )

ξ cos(φ − φ0 ) − ξ 2
= .
1 + ξ 2 − 2ξ cos(φ − φ0 )

Inserting this relation into the integral above, we can determine the Green’s function
completely. Calling trivial substitutional integration, we see that

X∞ µ ¶m Z ρρ− Ã X ∞
!
1 ρ− 0 + 1
m 0
2 cos [m(φ − φ )] = 2 ξ cos [m(φ − φ )] dξ,
m=1
m ρ+ 0 ξ m=1
Z ρρ−
+ 1 ξ cos(φ − φ0 ) − ξ 2
=2 dξ,
0 ξ 1 + ξ 2 − 2ξ cos(φ − φ0 )
Z ρρ−
+ cos(φ − φ0 ) − ξ
=2 dξ,
0 1 + ξ 2 − 2ξ cos(φ − φ0 )
³ ´
Z 1+ ρρ− 2 −2 ρρ− cos(φ−φ0 )
+ + du
=− ,
1 u
à µ ¶2 !
ρ− ρ− 0
= − log 1 + −2 cos(φ − φ ) .
ρ+ ρ+

Similarly, we see that

∞ µ ¶m à µ 0 ¶2 !
X 1 ρρ0 ρρ ρρ0
0 0
2 cos [m(φ − φ )] = − log 1 + − 2 2 cos(φ − φ ) .
m=1
m b2 b2 b
6 JACOB LEWIS BOURJAILY

Therefore, we see that

µ ¶m
1 ³ ρ+ ρ− ´ m
X∞ X∞
¡ ¢ 1 ρ−
G(ρ, ρ0 ) = log b2 − log(ρ2+ ) + 2 cos [m(φ − φ0 )] − 2 2
cos [m(φ − φ0 )] ,
m=1
m ρ + m=1
m b
µ 2¶ Ã µ ¶2 ! µ ¶
b ρ− ρ− ³ ρ ρ ´2 ρρ0
0 + − 0
= log − log 1 + −2 cos(φ − φ ) + log 1 + − 2 2 cos(φ − φ ) ,
ρ2+ ρ+ ρ+ b2 b
 
µ 2¶ ¡ ρ+ ρ− ¢2
b  1 + b2 − 2 ρ+b2ρ− cos(φ − φ0 ) 
= log + log  ³ ´ ,
ρ2+ 2
1 + ρρ− +
− 2 ρ−
ρ+ cos(φ − φ 0)
 
µ 4 ¶ ¡ ρ+ ρ− ¢2 ρ+ ρ− 0
 b 1 + b2 − 2 b2 cos(φ − φ ) 
= log  2 2 ³ ´2 ,
b ρ+ ρ− ρ− 0
1 + ρ+ − 2 ρ+ cos(φ − φ )
µ ¶
b4 + ρ2 ρ02 − 2ρρ0 b2 cos(φ − φ0 )
∴ G(ρ, ρ0 ) = log .
b2 (ρ2 + ρ02 − 2ρρ0 cos(φ − φ0 ))
‘ ’
óπ²ρ ²́δ²ι δ²ιξαι

b) We are to show that the solution to the Laplace equation with the potential given as ϕ(bφ) on
the cylinder can be expressed as the Poisson’s integral of problem (2.12).
Because we consider a cylinder that is free of charges in the interior, the most general
solution to Poisson’s equation with Green’s functions is given by
Z ¯
1 ∂G(ρ, ρ0 ) ¯¯
ϕ(ρ, φ) = − ϕ(b, φ0 ) da0 .
4π ∂Ω ∂ρ0 ¯ρ0 =b
Let us begin by computing the derivative of the Green’s function in the direction
normal to the cylinder.
¯ ¯
∂G(ρ, ρ0 ) ¯¯ 2ρ2 ρ0 − 2ρb2 cos(φ − φ0 ) 2ρ0 − 2ρ cos(φ − φ0 ) ¯
¯
= 4 − ,
∂ρ0 ¯ 0 ρ =b b + ρ2 ρ02 − 2ρρ0 b2 cos(φ − φ0 ) ρ2 + ρ02 − 2ρρ0 cos(φ − φ0 ) ¯ 0 ρ =b
2 2 0 0
2ρ b − 2ρb cos(φ − φ ) 2b − 2ρ cos(φ − φ )
= − ,
b4 + ρ2 b2 − 2ρb3 cos(φ − φ0 ) ρ2 + b2 − 2ρb cos(φ − φ0 )
µ ¶
1 2ρ2 − 2ρb cos(φ − φ0 ) 2b2 − 2ρb cos(φ − φ0 )
= − ,
b b2 + ρ2 − 2ρb cos(φ − φ0 ) ρ2 + b2 − 2ρb cos(φ − φ0 )
µ ¶
1 2ρ2 − 2ρb cos(φ − φ0 ) − 2b2 + 2ρb cos(φ − φ0 )
= ,
b b2 + ρ2 − 2ρb cos(φ − φ0 )
µ ¶
2 b 2 − ρ2
=− .
b b2 + ρ2 − 2ρb cos(φ − φ0 )
Therefore, we see that
Z 2π 0 ¯
¯
1 0 ∂G(ρ, ρ ) ¯
ϕ(ρ, φ) = − ϕ(b, φ ) bdφ0 ,
4π 0 ∂ρ0 ¯ρ0 =b
Z 2π µ ¶
1 1 b 2 − ρ2
= ϕ(b, φ0 ) bdφ0 ,
2π 0 b b2 + ρ2 − 2ρb cos(φ − φ0 )
Z 2π µ ¶
1 0 b2 − ρ2
∴ ϕ(ρ, φ) = ϕ(b, φ ) 2 dφ0 .
2π 0 b + ρ2 − 2ρb cos(φ − φ0 )
‘ ’
óπ²ρ ²́δ²ι δ²ιξαι
 PHYSICS 505: CLASSICAL ELECTRODYNAMICS HOMEWORK 3 7

c) We are to determine what changes are necessary if the Green’s function is desired to describe
the exterior problem. Specifically, we are to show that the closed-form Green’s function remains
the same.
Following our now ‘standard’ technique of generating the terms for the Fourier series,
we see that the exterior Green’s function can be expressed as a Fourier expansion
over the azimuthal coordinate,
∞
1 X im(φ−φ0 )
G(ρ, ρ0 ) = e gm (ρ, ρ0 ),
2π m=−∞
where the functions gm (ρ, ρ0 ) satisfy the Laplace equation and will be given in the
form ½
αm ρ0m + βm ρ0−m ρ0 < ρ
gm = ,
γm ρ0−m ρ0 > ρ
where the coefficients αm , βm , and γm will be determined by the boundary conditions
and continuity requirements.
If we continue mindlessly applying the continuity condition and the condition that
gm (b) = 0 ∀ m as before, we conclude that
µ³ ´ µ ¶m ¶
−m m m ρ m b
αm = ηm b and βm = −ηm b and γm = ηm ρ − ,
b ρ
and we determine ηm by the ³ ´now-tedious procedure of derivative-discontinuity check-
m
2π b
ing and obtain ηm = m ρ . With these, we can reexpress the functions gm so
that they specifically refer to the current situation. We see that
 ³³ 0 ´m ³ ´m ´ ³³ 0 ´m ³ 2 ´m ´ 
 ηm ρ
− b
= 2π ρ
− b
ρ 0
< ρ  2π µµ ρ ¶m µ b2 ¶m ¶
b ρ0 m ρ ρρ0 −
gm = ³ ´ ³
m ¡ ¢m ³ ´ m ´ ³³ ´ m ³ ´ m ´ = − .
 ηm ρ0 ρ
− ρb
= m2π ρ
− ρρ0b2 0
ρ >ρ  m ρ + ρρ0
ρ b ρ0

2.21 We are to use Cauchy’s theorem to derive the Poisson integral solution.
Let us consider the function ϕ(z) which is analytic over a disc, C of radius b centered at
z = 0. Then, as ever middle schooler knows from trivial application of the Residue
theorem, I
1 dz 0
ϕ(z) = ϕ(z 0 ) 0 ∀z ∈ C,
2πi C z −z
because, by the analyticity of ϕ(z) in the region C, the only pole in C comes from
z 0 = z which has a simple Residue of ϕ(z).
Furthermore, if z ∈ C, then
I
1 dz 0
ϕ(z 0 ) 0 = 0,
2πi C z − b2 /z
because z 0 = b2 /z ∈ / C–as trivially seen by the fact that |z 0 − b2 /z| > b because
|z| < b. Therefore, we have that for z ∈ C,
I µ ¶
1 0 1 1
ϕ(z) = ϕ(z ) − 0 dz 0 .
2πi C z0 − z z − b2 /z
¡ ¢
Notice the trivial identity z 0 − b2 /z = z 0 /z z − z 0 , which can be seen by simple expan-
sion. Noting that |z 0 | = b on C, we have
I Ã !
1 0 1 z
ϕ(z) = ϕ(z ) − 0¡ ¢ dz 0 ,
2πi C z0 − z z z − z0
I Ã ! I µ ¶
1 0 z 0 z − b2 − z 0 z + |z|2 1 b2 − ρ2
= ϕ(z ) ¡ ¢ dz 0 = ϕ(z 0
≡ (b, φ 0
)) dz 0 ,
2πi C z 0 z 0 z − b2 − |z|2 + z 0 z 2πi C z 0 (b2 + ρ2 − 2bρ cos(φ − φ0 ))
Z 2π µ ¶
1 b2 − ρ2
∴ ϕ(ρ, φ) = ϕ(b, φ0 ) dφ0 .
2π 0 b + ρ − 2bρ cos(φ − φ0 )
2 2

‘ ’
óπ²ρ ²́δ²ι δ²ιξαι