Laplace and Poissons

Equations
Derivation
We have the differential form of Gauss law


Using D = E and E = - in the above equation, we get




This is the Poissons Equation



. V = D
V V
( )
2
D= V =- E V c c c V V = V V V =
2
V
or

c
V =
2
0 V V =
In free space , this equation becomes,


This equation is called the Laplace equation.

is called the Laplacian operator or simply Laplacian.
Note that the del operator is defined only in the
rectangular coordinates only, as
V
i j k
x y z
c c c
V = + +
c c c
2
V
The Laplacian in the three coordinate systems are
2 2 2
2
2 2 2
c c c
V + +
c c c
= ( ) cartesian
x y z
2 2
2
2 2 2
1

|
| | c c c c
V + +
|
c c c c
\ .
1
= ( ) Cylindrical
z
2
2 2
2 2 2 2 2
1 1
u
u u u u |
c c c c c
| | | |
V + +
| |
c c c c c
\ . \ .
1
= sin ( )
sin sin
r Spherical
r r r r r
Applications

one dimensional solution of Laplace equation
2 2 2
2
2 2 2
c c c
V + +
c c c
= = 0 ( )
V V V
V cartesian
x y z
--- (1)
V is a function of only one variable and is independent of the
other two variables. Under this condition, in rectangular
coordinate system the Laplace equation reduces to
One dimensional solution of Laplace Equation in rectangular
coordinate system
Let V be a function of z only. Then in Rectangular coordinate
system,the Laplaces Equation reduces to
2
2
2
0
V
V
z
c
V = =
c
Since V is a function of z only, it is independent
of x and y. Therefore,
Integrating both sides once we get
--- (2)
Equation (3) represents a family of equi potential surfaces
with z taking up constant values
dV
A
dz
=
Integrating both sides once again, we get
, A is an arbitrary constant --- (3)
V Az B = +
, B is an arbitrary constant --- (4)
A and B are arbitrary constants to be evaluated under suitable
boundary conditions.
Consider two such equi potential surfaces one at z = z
1
and the
other at z = z
2
. Let V= V
1
at z = z
1
and V = V
2
at z = z
2
.
We immediately recognize that this is the case with a parallel
plate capacitor with a plate separation of z
1
- z
2
= d and a
potential difference V
1
- V
2
.

Applying the above two conditions, called boundary conditions,
we get,
1 1
2 2
V V Az B
V V Az B
= = +
= = +
--- (5)
--- (6)
Solving equations (5) and (6) we get the values for A and B as
2 1
2 1
1 2 2 1
2 1
V V
A
z z
V z V z
B
z z

Substituting the values of A and B in equation (4) we get,

2 1 1 2 2 1
2 1 2 1
2 1 1 2
2 1
( ) ( )
V V Vz V z
V z
z z z z
V z z V z z
z z

= +


=

--- ( 7)
Further let, for simplicity, V
1
= 0 and z
1
= 0, V
2
= V
a

Then equation (7) reduces to
a
V
V z
d
=
We find that V is a linear function of z
V = V
2
= V
a
V = V
1
= 0

d
z = z
2

z = z
1

Fig 1 Parallel plate capacitor
--- (8)
1 Given V, Determine E using the formula E = - V V
2 Determine D using D = E
3 Find D on any one of the plates, D = D
S
= D
S
a
S
= D
N
a
N

on the chosen plate, and recognising that D
N
=
S

4 Determine Q by surface integration of
S
over the surface
area of the chosen plate using



5 Compute the capacitance using the formula
a
Q
C
V
=
S
S
Q dS =
}
Applying these five steps to the parallel plate capacitor we get,
0
;
a
a
z
a
z
a
S z S S N N N S
z
a a a
S S
z
V V
d
V
E V a
d
V
D E a
d
V
D D a D a D a D
d
V V V
Q dS dS S
d d d
c c
c
c c c
=
=
= V =
= =
= = = = =
= = =
} }
Therefore the capacitance of the parallel plate capacitor is
a
Q
S
C
V d
c
= =
|
We have, in cylindrical coordinates, the Laplace equation as
2 2
2
2 2 2
1
0
|
| | c c c c
V + + =
|
c c c c
\ .
1
V= ( )
V V V
Cylindrical
z
One dimensional solution of Laplace Equation in cylindrical
coordinate system
We consider that V is a function of only. In this case
the Laplace equation in cylindrical coordinates reduces to
0

| | c c
=
|
c c
\ .
V
Integrating this equation once we get,
1
0

| | c c
=
|
c c
\ .
V
V V A
A o r

c c
= =
c c
, , A an arbitrary constant
Case 1:
Integrating once again, we get
= + V Al n B, B an arbitrary constant
From this equation, we observe that equipotential surfaces
are given by = constant and are cylinders. Example of the
problem is that of a coaxial capacitor or coaxial cable.
Let us create the boundary conditions by choosing V = V
a
at
= a and V = 0 at = b, b > a.
Then we get from the above equation,
B b n l A V V
B a n l A V V
b
a
+ = =
+ = =
Solving these two equations for A and B we get
|
.
|

\
|

=
|
.
|

\
|

=
b
a
b V a V
B d n a
b
a
V V
A
a b b a
ln
ln ln
ln
Substituting these the values of A and B in the general
Expression for V , we get
|
.
|

\
|

+
|
.
|

\
|

=
b
a
b V a V
b
a
V V
V
a b b a
ln
ln ln
ln
ln

Letting V
b
= 0, we get
|
.
|

\
|
|
|
.
|

\
|
=
a
b
b
V V
a
ln
ln

a
b
a
b
V
a
b
a
b
V V
V E
a a
ln
ln
ln
ln
|
|
.
|

\
|
c
c
|
.
|

\
|
=
|
|
|
|
.
|

\
|
|
.
|

\
| c
c
=
c
c
= V =

Therefore
d n a a D a D a
a
b
a
V
D
Then
D a
a
b
V
E D
a
a
b
V
E
N N N S
a
S
S
a
a

ln
1

ln
1

ln
1
= =
|
.
|

\
|
=
=
|
.
|

\
|
= =
|
.
|

\
|
=

c c

i.e.,
Hence
|
.
|

\
|
= =
a
b
V
a
D
a S N
ln
1 c

Therefore,
1
ln
1
2
ln
S S
S of one of S of one of
the plates the plates
with a with a
a
a
Q dS dS
V S
b
a
a
V a L
b
a
a


c
c
t
= =
= =
=
=
} }
over a length L meters of the
coaxial cable
a
b
L
V
Q
C
a
ln
2 c t
= =
Therefore the capacitance per unit length of the line C
L
is
a
b
C
L
ln
2 c t
= F/m
Next we consider V as a function of only. In this case
The Laplace equation in cylindrical coordinate system
reduces to
|
2
2
2 2
1
0
|
c
V =
c
V=
V
Case 2:
i.e.,
2
2 2
1
0
|
=
d V
d
Excluding the value = 0 this equation becomes
2
2
0
|
=
d V
d
Integrating both sides we get
|
=
dV
A
d
Integrating once again, we get
| = + V A B
This is the general equation for V when V is a function of only.
|
From this equation, we observe that equipotential surfaces
are given by = constant and are planes To visualize this, choose two
such equipotential surfaces, V = V
a
at = and V = 0 at = 0.
Insulating gap
V = Va
at =
V = 0
at = 0

For the chosen boundary condition, we get
and
a
V
A
|
=
Thus the general expression for V becomes
0 0 0
a
V V A B
V A B B
o = = +
= = + =
a
V
V |
o
=
Once again we follow the 5 step procedure to determine
the capacitance of the system
1 1

a
V V
E V
o
c
= V = =
c
Note that E is a function of and not of . But the vector field E
Is a function of . Now,

a
V
D E
c
c
o
= =

o
c

a
N N S S S
V
a D a D D = = =

S
a
N
V
D
o
c
= =
The surface integration on
S
gives Q:
} } }
= = =
S
a
S
a
S
S
dS
V
ds
V
dS Q
o
c
o
c

and we get immediately the value for the capacitance of

the corner reflector system as
}
}
= = =
S
a
S
a
a
dS
V
dS
V
V
Q
C
o
c
o
c
Next we consider the Laplace Equation in spherical coordinates:
2
2 2
2 2 2 2 2
1 1 1
sin 0
sin sin
V V V
V r
r r r r
u
u u u u |
c c c c c
| | | |
V = + + =
| |
c c c c c
\ . \ .
In this system we consider that V is a function of r only.
Then the Laplace equation reduces to
0
1
2
2
2
=
|
.
|

\
|
c
c
c
c
= V
r
V
r
r
V
Again we exclude r = 0 from our solutions. Multiplying both
Sides by r
2
we get
0
2
=
|
.
|

\
|
c
c
c
c
r
V
r
--- ( a)
--- ( b)
One dimensional solution of Laplace Equation in spherical
coordinate system
Case 1:
Integrating once again, we get
2
V
r A
r
c
=
c
2
V A
or
r r
c
=
c
A
o r V B
r
= +
where A and B are arbitrary constants to be evaluated. This
equation represents a family of equi potential surfaces for
r = constant.
Let us choose two such equipotential surfaces at r = a and
R = b, b > a , such that at r = a, V = V
a
and at r = b, V = V
b

--- ( c)
--- ( d)
a
b
A
V B
a
A
V B
b
= +
= +
Solving these two equations we get
--- ( e)
--- ( f)
1 1
a b
V V
A
a b

=
| |

|
\ .
and
1 1
1 1
a b
V V
b a
B
b a

=
| |

|
\ .
We immediately recognise that this is the example of concentric
spheres or Spherical capacitor
Substituting the values of A and B in equation (d), we get,
1 1
1
1 1 1 1
a b
a b
V V
V V
b a
V
r
a b b a
| | | |

| |

| |
\ . \ .
= +
|
| | | |
\ .

| |
\ . \ .
1 1
1
1 1 1 1
b a
a b
V V
V V
a b
o r V
r
a b a b
| | | |

| |

| |
\ . \ .
= +
|
| | | |
\ .

| |
\ . \ .
Let V
b
= 0 Then equation (g) becomes
--- (g)
1 1
1 1
a
r b
V V
a b
| |

|
\ .
=
| |

|
\ .
--- (h)
2
1 1
1

1 1 1 1
a
a
V
r b
E V V r
r
a b a b
| |
| |

| |
\ .
|
= V = V =
| | | |
|

| |
|
\ . \ .
\ .
2

1 1
a
V
D E r
r
a b
c
c = ==
| |

|
\ .
2

1 1
a
S S S N N
V
D D D a D a r
r
a b
c
= = = =
| |

|
\ .
2
1 1
a
N S
V
D
a
a b
c
= =
| |

|
\ .
Therefore the charge Q on the capacitor plate is
2 2
1 1 1 1
a a
S
S S S
V V
Q dS dS dS
a a
a b a b
c c
= = =
| | | |

| |
\ . \ .
} } }
Therefore we get the expression for the spherical capacitor as
2
2
4 4
1 1 1 1
a
a a
Q
V a
C
V a V
a b a b
c t tc
= = =
| | | |

| |
\ . \ .
4
1 1
C
a b
tc
=
| |

|
\ .
For an isolated sphere , i.e., as b we get
4 C a t c =
Finally let us consider V as a function of only . In this case
The Laplaces equation reduces to
2
2
1
0 u
u u u
c c
| |
V =
|
c c
\ .
V= sin
sin
V
r
We exclude r = 0 and = n/2, n = 0, 1, 2, 3, . . .
Multiplying both sides of the above equation by r
2
sin, we get
0 u
u u
c c
| |
=
|
c c
\ .
sin
V
Case 2:
Integrating the equation with respect to we get,
or
u
u
c
=
c
sin
V
A
u u
c
=
c sin
V A
A is an arbitrary constant
Integrating once again, we get,
2 u = + ln(tan / ) V A B
This equation represents a family of equipotential surfaces for
constant . Let us consider two such equipotential surfaces at
=/2 and = and let V = 0 at =/2 and V = Va at =0.

--- (i)
The equipotential surfaces are cones as shown in figure below.
Such a system is called a conical antenna


Insulating gap
Ground plane
V = V
a
at =
V = 0 at = 0
Applying these two boundary conditions to the equation (i),
Solving for A and B and substituting these values in (i),we get,
2
2
u
o
=
ln(tan / )
ln(tan / )
a
V V
We have
2
2
u
o
=
ln(tan / )
ln(tan / )
a
V V
We use E = - V V to find the field strength, as
1
2
u u
u u o
c
= V = =
c

sin (tan / )
a
V V
E V a a
r r ln
Next we determine D using D =
0
E as
2
u
c
c
u o
= ==

sin (tan / )
a
V
D E a
r ln
2
c

o o
=
sin (tan / )
a
S
V
r ln
The total charge Q on the conical surface is therefore,
2
0
2
t
c
o |
o o

= =
} } }
sin
sin (tan / )
a
S
S o
V
Q ds r d dr
r ln
0
2
2
t c
o

=
}
(tan / )
a
V
Q dr
ln
This equation leads to an infinite value of charge and capacitance.
Therefore we have to consider a cone of finite size.
Our expression for Q is approximate, since, theoretically, the
potential surface = extends from r = 0 to r = . But our physical
conical surface extends from r = 0 to say, r = r
1
. The approximate
capacitance is
1
2
2
t c
o
=
(cot / )
a
r
C
ln