Section 5 4 Electrostatic Boundary Value Problems Package

11/8/2004 Section 5_4 Electrostatic Boundary Value Problems blank 1/2
5-4 Electrostatic Boundary

Value Problems
Reading Assignment: pp. 149-157
Q:
A:
We must solve differential equations, and apply

boundary conditions to find a unique solution.
In EE and CoE, we typically use a voltage source to

apply boundary conditions on electric potential
function V ( r ) .
This process is best demonstrated with a series of

examples:
Example: Dielectric Filled Parallel Plates
Jim Stiles The Univ. of Kansas Dept. of EECS

11/8/2004 Section 5_4 Electrostatic Boundary Value Problems blank 2/2
Example: Charge Filled Parallel Plates
Example: The Electrostatic Fields of a Coaxial Line

11/8/2004 Example Dielectric Filled Parallel Plates 1/8
Example: Dielectric Filled

Parallel Plates
Consider two infinite, parallel conducting plates, spaced a
distance d apart. The region between the plates is filled with a
dielectric ε . Say a voltage V0 is placed across these plates.
z=-d
+
_
V0
ε z=0
Q: What electric potential field V ( r ) , electric field E ( r )

and charge density ρs ( r ) is produced by this situation?
A: We must solve a boundary value problem ! We must

find solutions that:
a) Satisfy the differential equations of electrostatics

(e.g., Poisson’s, Gauss’s).
b) Satisfy the electrostatic boundary conditions.

Q: Yikes! Where do we even start ?
A: We might start with the electric potential field V ( r ) ,

since it is a scalar field.
a) The electric potential function must satisfy

Poisson’s equation:
− ρv ( r )
∇2V ( r ) =
ε
b) It must also satisfy the boundary conditions:
V ( z = −d ) = V0 V (z = 0 ) = 0
Consider first the dielectric region ( −d < z < 0 ). Since the

region is a dielectric, there is no free charge, and:
ρv ( r ) = 0
Therefore, Poisson’s equation reduces to Laplace’s

equation:
∇2V ( r ) = 0
This problem is greatly simplified, as it is evident that the

solution V ( r ) is independent of coordinates x and y . In
other words, the electric potential field will be a function
of coordinate z only:
V (r ) = V (z )

This make the problem much easier! Laplace’s equation

becomes:
∇2V ( r ) = 0
∇2V ( z ) = 0
∂2V ( z )
=0
∂z 2
Integrating both sides of Laplace’s equation, we get:
∂2V ( z )
∫ ∂z 2 dz = ∫ 0 dz
∂V ( z )
= C1
∂z
And integrating again we find:
∂V ( z )
∫ ∂z dz = ∫ C1 dz
V ( z ) = C1z + C 2
We find that the equation V ( z ) = C1z + C 2 will satisfy Laplace’s

equation (try it!). We must now apply the boundary conditions
to determine the value of constants C1 and C2.
We know that the value of the electrostatic potential at every

point on the top (z =-d) plate is V (-d)=V0, while the electric
potential on the bottom plate (z =0) is zero (V (0) =0 ).
Therefore:

V ( z = −d ) = −C1d + C 2 = V0
V ( z = 0 ) = C1 ( 0 ) + C2 = 0
Two equations and two unknowns (C1 and C2)!
Solving for C1 and C2 we get:
V0
C 2 = 0 and C1 = −
d
and therefore, the electric potential field within the dielectric

is found to be:
−V0z
V (r ) = ( −d ≤ z ≤ 0)
d
Before we proceed, let’s do a sanity check!
In other words, let’s evaluate our answer at z = 0 and z = -d,

to make sure our result is correct:
−V0 ( −d )
V ( z = −d ) = = V0
d
and
−V0 ( 0 )
V (z = 0 ) = = 0
d

Now, we can find the electric field within the dielectric by

taking the gradient of our result:
V0
E ( r ) = −∇V ( r ) = âz ( −d ≤ z ≤ 0)
d
And thus we can easily determine the electric flux density

by multiplying by the dielectric of the material:
εV0
D (r ) = ε E (r ) = âz ( −d ≤ z ≤ 0)
d
Finally, we need to determine the charge density that

actually created these fields!
Q: Charge density !?! I thought that we already

determined that the charge density ρv ( r ) is equal to zero?
A: We know that the free charge density within the

dielectric is zero—but there must be charge somewhere,
otherwise there would be no fields!

Recall that we found that at a conductor/dielectric interface,

the surface charge density on the conductor is related to the
electric flux density in the dielectric as:
Dn = ân ⋅ D ( r ) = ρs ( r )
First, we find that the electric flux density on the bottom

surface of the top conductor (i.e., at z = −d ) is:
εV0 εV0
D (r ) = âz = âz
z =−d
d z =−d d
For every point on bottom surface of the top conductor, we

find that the unit vector normal to the conductor is:
ân = âz
Therefore, we find that the surface charge density on the

bottom surface of the top conductor is:
ρs + ( r ) = ân ⋅ D ( r )
z = −d
εV0
= âz ⋅ âz
d
εV0
= (z = −d )
d

Likewise, we find the unit vector normal to the top surface of

the bottom conductor is (do you see why):
ân = −âz
Therefore, evaluating the electric flux density on the top

surface of the bottom conductor (i.e., z = 0 ), we find:
ρs − ( r ) = ân ⋅ D ( r ) z = 0
εV0
= −âz ⋅ âz
d
−εV0
= (z = 0)
d
We should note several things about these solutions:
1) ∇xE ( r ) = 0
2) ∇ ⋅ D ( r ) = 0 and ∇2V ( r ) = 0
3) D ( r ) and E ( r ) are normal to the surface of the conductor

(i.e., their tangential components are equal to zero).
4) The electric field is precisely the same as that given by

using superposition and eq. 4.20 in section 4-5!

I.E.:
ρs + ρs − V0
E (r ) = âz − âz = âz ( −d < z < 0)
2ε 2ε d
In other words, the fields E ( r ) , D ( r ) , and V ( r ) are

attributable to charge densities ρs + ( r ) and ρs − ( r ) .
z=-d
ρs + ( r ) + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
V0
+
E (r )
_
ε
ρs − ( r ) _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
z=0

11/8/2004 Example Charge Filled Parallel Plates 1/4
Example: Charge Filled

Parallel Plates
Consider now a problem similar to the previous example (i.e.,
dielectric filled parallel plates), with the exception that the
space between the infinite, conducting parallel plates is filled
with free charge, with a density:
ρv ( r ) = −z ε 0 (-d < z < 0)
z=-d
+
_ V0
ρv (r )
z=0
z
Q: How do we determine the fields within the parallel plates
for this problem?
A: Same as before! However, since the charge density between

the plates is not equal to zero, we recognize that the electric
potential field must satisfy Poisson’s equation:
− ρv ( r )
∇2V ( r ) =
ε0

For the specific charge density ρv ( r ) = −z ε 0 :
− ρv ( r )
∇2V ( r ) = =z
ε0
Since both the charge density and the plate geometry are
independent of coordinates x and y, we know the electric
potential field will be a function of coordinate z only (i.e.,
V ( r ) = V ( z ) ).
Therefore, Poisson’s equation becomes:
∂2V ( z )
∇ V (z ) =
2
=z
∂z 2
We can solve this differential equation by first integrating both

sides:
∂2V ( z )
∫ ∂z 2 dz = ∫ z dz
∂V ( z ) z 2
= + C1
∂z 2
And then integrating a second time:
∂V ( r ) ⎛z2 ⎞
∫ ∂z dz = ∫ ⎜⎝ 2 1 ⎟⎠ dz
+ C
z3
V (r ) = + C1z + C 2
6

Note that this expression for V ( r ) satisfies Poisson’s equation

for this case. The question remains, however: what are the
values of constants C1 and C 2 ?
We find them in the same manner as before—boundary

conditions!
Note the boundary conditions for this problem are:
V ( z = −d ) = V0
V (z = 0 ) = 0
Therefore, we can construct two equations with two unknowns:
( −d )
3
V ( z = −d ) = V0 = + C1 ( −d ) + C 2
6
3
(0)
V (z = 0 ) = 0 = + C1 ( 0 ) + C 2
6
It is evident that C2 = 0, therefore constant C1 is:
⎛V0 d 2 ⎞
C1 = − ⎜ + ⎟
⎝d 6 ⎠
The electric potential field between the two plates is

therefore:

⎛V0 d 2 ⎞
z3
V (r ) = −⎜ + ⎟z ( −d < z < 0)
6 ⎝d 6 ⎠
Performing our sanity check, we find:
( −d )
3
⎛V0 d 2 ⎞
V ( z = -d ) = −⎜ + ⎟ ( -d )
6 ⎝ d 6 ⎠
−d3 d3
= +V0 +
6 6
= V0
and
3
(0)
⎛V0 d 2 ⎞
V (z = 0 ) = −⎜ + ⎟ (0)
6 ⎝ d 6 ⎠
=0+0+0
=0
From this result, we can determine the electric field E ( r ) , the

electric flux density D ( r ) , and the surface charge density
ρs ( r ) , as before.
Note, however, that the permittivity of the material between

the plates is ε 0 , as the “dielectric” between the plates is free-
space.

11/8/2004 Example The Electorostatic Fields of a Coaxial Line 1/10
Example: The Electrostatic

Fields of a Coaxial Line
A common form of a transmission line is the coaxial cable.
Outer
Conductor
a
+
ε V0
-
Inner
Conductor
Coax Cross-Section
The coax has an outer diameter b, and an inner diameter a.

The space between the conductors is filled with dielectric
material of permittivity ε .
Say a voltage V0 is placed across the conductors, such that the

electric potential of the outer conductor is zero, and the
electric potential of the inner conductor is V0.

The potential difference between the inner and outer

conductor is therefore V0 – 0 = V0 volts.
Q: What electric potential field V ( r ) , electric field E ( r )

and charge density ρs ( r ) is produced by this situation?
A: We must solve a boundary-value problem! We must

find solutions that:
a) Satisfy the differential equations of electrostatics

(e.g., Poisson’s, Gauss’s).
b) Satisfy the electrostatic boundary conditions.
Yikes! Where do we start ?
We might start with the electric potential field V ( r ) , since

it is a scalar field.
a) The electric potential function must satisfy

Poisson’s equation:
− ρv ( r )
∇2V ( r ) =
ε
b) It must also satisfy the boundary conditions:
V ( ρ = a ) = V0 V (ρ = b) = 0

Consider first the dielectric region ( a < ρ < b ). Since the

region is a dielectric, there is no free charge, and:
ρv ( r ) = 0
Therefore, Poisson’s equation reduces to Laplace’s equation:
∇2V ( r ) = 0
This particular problem (i.e., coaxial line) is directly solvable

because the structure is cylindrically symmetric. Rotating
the coax around the z-axis (i.e., in the âφ direction) does
not change the geometry at all. As a result, we know that
the electric potential field is a function of ρ only ! I.E.,:
V (r ) = V ( ρ )
This make the problem much easier. Laplace’s equation

becomes:
Be very careful during this

step! Make sure you
implement the gul durn ∇2V ( r ) = 0
Laplacian operator correctly.
∇2V ( ρ ) = 0
1 ∂ ⎛ ∂V ( ρ ) ⎞
⎜ρ ⎟+0+0 =0
ρ ∂ρ ⎝ ∂ρ ⎠
∂ ⎛ ∂V ( ρ ) ⎞
⎜ρ ⎟=0
∂ρ ⎝ ∂ρ ⎠

Integrating both sides of the resulting equation, we find:
∂ ⎛ ∂V ( ρ ) ⎞
∫ ∂ρ ⎜ρ ⎟ d ρ = ∫ 0d ρ
⎝ ∂ρ ⎠
∂V ( ρ )
ρ = C1
∂ρ
where C1 is some constant.
Rearranging the above equation, we find:
∂V ( ρ ) C1
=
∂ρ ρ
Integrating both sides again, we get:
∂V ( ρ ) C1
∫ dρ=∫ dρ
∂p ρ
V ( ρ ) = C1 ln [ ρ ] + C 2
We find that this final equation (V ( ρ ) = C 1 ln [ ρ ] + C 2 ) will

satisfy Laplace’s equation (try it!).
We must now apply the boundary conditions to determine

the value of constants C1 and C2.
* We know that on the outer surface of the inner

conductor (i.e., ρ = a ), the electric potential is equal to
V0 (i.e., V ( ρ = a ) = V0 ).

* And, we know that on the inner surface of the outer

conductor (i.e., ρ = b ) the electric potential is equal to
zero (i.e., V ( ρ = b ) = 0 ).
Therefore, we can write:
V ( ρ = a ) = C1 ln [a ] + C 2 = V0
V ( ρ = b ) = C1 ln [b ] + C 2 = 0
Two equations and two unknowns (C1 and C2)!
Solving for C1 and C2 we get:
−V0 −V0
C1 = =
ln [b] − ln [ a ] ln ⎡⎣b/a ⎤⎦
V0 ln [b]
C2 =
ln ⎡⎣b/a ⎤⎦
and therefore, the electric potential field within the

dielectric is found to be:
−V0 ln [ ρ ] V0 ln [b]
V (r ) = + (b > ρ > a)
ln ⎡⎣b/a ⎤⎦ ln ⎡⎣b/a ⎤⎦

Before we move on, we should do a sanity check to make

sure we have done everything correctly. Evaluating our
result at ρ = a , we get:
−V0 ln [ a ] V0 ln [b]
V ( ρ = a) = +
ln ⎡⎣b/a ⎤⎦ ln ⎡⎣b/a ⎤⎦
V0 (ln [b] − ln [ a ])
=
ln ⎡⎣b/a ⎤⎦
V0 (ln ⎡⎣b/a ⎤⎦)
=
ln ⎡⎣b/a ⎤⎦
= V0
Likewise, we evaluate our result at ρ = b :
−V0 ln [b] V0 ln [b]

V ( ρ = b) = +
ln ⎡⎣b/a ⎤⎦ ln ⎡⎣b/a ⎤⎦
V0 (ln [b] − ln [b])
=
ln ⎡⎣b/a ⎤⎦
=0
Our result is correct!
Now, we can determine the electric field within the

dielectric by taking the gradient of the electric potential
field:
V0 1
E ( r ) = −∇V ( r ) = âρ (b > ρ > a)
ln ⎡⎣b/a ⎤⎦ ρ

Note that electric flux density is therefore:
εV0 1
D (r ) = ε E (r ) = âρ (b > ρ > a)
ln ⎡⎣b/a ⎤⎦ ρ
Finally, we need to determine the charge density that

actually created these fields!
Q1: Just where is this charge? After all, the

dielectric (if it is perfect) will contain no free charge.
A1: The free charge, as we might expect, is in the

conductors. Specifically, the charge is located at the
surface of the conductor.
Q2: Just how do we determine this surface

charge ρs ( r ) ?
A2: Apply the boundary conditions!
Recall that we found that at a conductor/dielectric interface,

the surface charge density on the conductor is related to the
electric flux density in the dielectric as:
Dn = ân ⋅ D ( r ) = ρs ( r )

First, we find that the electric flux density on the surface of

the inner conductor (i.e., at ρ = a ) is:
εV0 1
D (r ) = âρ
ρ =a
ln ⎡⎣b/a ⎤⎦ ρ ρ =a
εV0 1
= âρ
ln ⎡⎣b/a ⎤⎦ a
For every point on outer surface of the inner conductor, we

find that the unit vector normal to the conductor is:
ân = âρ
Therefore, we find that the surface charge density on the

outer surface of the inner conductor is:
ρsa ( r ) = ân ⋅ D ( r ) ρ =a
εV0 1
= âρ ⋅ âρ
ln ⎡⎣b/a ⎤⎦ a
εV0 1
= ( ρ = a)
ln ⎡⎣b/a ⎤⎦ a

Likewise, we find the unit vector normal to the inner surface of

the outer conductor is (do you see why?):
ân = −âρ
Therefore, evaluating the electric flux density on the inner

surface of the outer conductor (i.e., ρ = b ), we find:
ρsb ( r ) = ân ⋅ D ( r ) ρ =b
εV0 1
= −âρ ⋅ âρ
ln ⎡⎣b/a ⎤⎦ b
−εV0 1
= (ρ = b)
ln ⎡⎣b/a ⎤⎦ b
Note the charge on the outer conductor is negative, while that

of the inner conductor is positive. Hence, the electric field
points from the inner conductor to the outer.
_ _ _
_ _ E (r )
_ _
+
+ + ++
+ +
_ + + _
+ +
+ +
_ + ++ _
_ _
_ _
_

We should note several things about these solutions:
1) ∇xE ( r ) = 0
2) ∇ ⋅ D ( r ) = 0 and ∇2V ( r ) = 0
3) D ( r ) and E ( r ) are normal to the surface of the

conductor (i.e., their tangential components are equal to
zero).
4) The electric field is precisely the same as that given by

eq. 4.31 in section 4-5!
a ρsa V0 1
E (r ) = âρ = âρ (b > ρ > a)
ερ ln ⎡⎣b/a ⎤⎦ ρ
In other words, the fields E ( r ) , D ( r ) , and V ( r ) are

attributable to free charge densities ρsa ( r ) and ρsb ( r ) .

Section 5 4 Electrostatic Boundary Value Problems Package

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Section 5 4 Electrostatic Boundary Value Problems Package

Uploaded by

Copyright:

Available Formats

11/8/2004 Section 5_4 Electrostatic Boundary Value Problems blank 1/2

5-4 Electrostatic Boundary

We must solve differential equations, and apply

In EE and CoE, we typically use a voltage source to

This process is best demonstrated with a series of

Example: Dielectric Filled Parallel Plates

Jim Stiles The Univ. of Kansas Dept. of EECS

Example: Charge Filled Parallel Plates

Example: The Electrostatic Fields of a Coaxial Line

Jim Stiles The Univ. of Kansas Dept. of EECS

Example: Dielectric Filled

Q: What electric potential field V ( r ) , electric field E ( r )

A: We must solve a boundary value problem ! We must

a) Satisfy the differential equations of electrostatics

b) Satisfy the electrostatic boundary conditions.

Jim Stiles The Univ. of Kansas Dept. of EECS

Q: Yikes! Where do we even start ?

A: We might start with the electric potential field V ( r ) ,

a) The electric potential function must satisfy

b) It must also satisfy the boundary conditions:

Consider first the dielectric region ( −d < z < 0 ). Since the

Therefore, Poisson’s equation reduces to Laplace’s

This problem is greatly simplified, as it is evident that the

Jim Stiles The Univ. of Kansas Dept. of EECS

This make the problem much easier! Laplace’s equation

Integrating both sides of Laplace’s equation, we get:

And integrating again we find:

We find that the equation V ( z ) = C1z + C 2 will satisfy Laplace’s

We know that the value of the electrostatic potential at every

Jim Stiles The Univ. of Kansas Dept. of EECS

Two equations and two unknowns (C1 and C2)!

Solving for C1 and C2 we get:

and therefore, the electric potential field within the dielectric

Before we proceed, let’s do a sanity check!

In other words, let’s evaluate our answer at z = 0 and z = -d,

Jim Stiles The Univ. of Kansas Dept. of EECS

Now, we can find the electric field within the dielectric by

And thus we can easily determine the electric flux density

Finally, we need to determine the charge density that

Q: Charge density !?! I thought that we already

A: We know that the free charge density within the

Jim Stiles The Univ. of Kansas Dept. of EECS

Recall that we found that at a conductor/dielectric interface,

First, we find that the electric flux density on the bottom

For every point on bottom surface of the top conductor, we

Therefore, we find that the surface charge density on the

Jim Stiles The Univ. of Kansas Dept. of EECS

Likewise, we find the unit vector normal to the top surface of

Therefore, evaluating the electric flux density on the top

We should note several things about these solutions:

3) D ( r ) and E ( r ) are normal to the surface of the conductor

4) The electric field is precisely the same as that given by

Jim Stiles The Univ. of Kansas Dept. of EECS

In other words, the fields E ( r ) , D ( r ) , and V ( r ) are

Jim Stiles The Univ. of Kansas Dept. of EECS

Example: Charge Filled

ρv ( r ) = −z ε 0 (-d < z < 0)

A: Same as before! However, since the charge density between

Jim Stiles The Univ. of Kansas Dept. of EECS

For the specific charge density ρv ( r ) = −z ε 0 :

Therefore, Poisson’s equation becomes:

We can solve this differential equation by first integrating both

And then integrating a second time:

Jim Stiles The Univ. of Kansas Dept. of EECS