Lecture Notes 6: The Method of Images

UIUC Physics 435 EM Fields & Sources I
Fall Semester, 2007
Lect. Notes 6
Prof. Steven Errede
LECTURE NOTES 6
THE METHOD OF IMAGES
A useful technique for solving (i.e. finding) E ( r ) and / or V ( r ) for a certain class /
special classes of electrostatic (and magnetostatic) problems that have some (or high)
degree of mirror-reflection symmetry. Exploit awesome power of symmetry
intrinsic to the problem, if present.
Idea is to convert a (seemingly) difficult electrostatic problem involving spatiallyextended charged objects (e.g. charged conductors) and then replace them with a finite
number of carefully, intelligently chosen / well-placed discrete point charges!!
Solving the simpler point-charge problem is the solution for the original, more
complicated problem!!!
Can replace e.g. a charged surface of a conductor (which is at constant potential an

equipotential) by an equivalent / identical equipotential surface (at same potential) due to
one (or more) such / so-called image charges.
By doing this replacement, the original boundary conditions associated with original
problem are retained / conserved.
E -fields and potentials V of the original and surrogate problems must be the same /
identical!!
The method of images is best learned by example
Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois

2005 - 2008. All rights reserved.
Example 1:
Fall Semester, 2007
Lect. Notes 6
Prof. Steven Errede
Point Charge q Located Near An Infinite, Grounded Conducting Plane:
n.b. A grounded conductor is a special type of equipotential: infinite amounts of electrical

charge ( Q) can flow from / to ground to / from the conducting surface so as to maintain
electrostatic potential V = 0 (Volts) at all times. (In reality / real life, no such thing. e.g.
especially / particularly for AC / time-varying electromagnetic fields for frequencies f 1 MHz,
due to inductance effects (magnetic analog of capacitance)).
What are:
E ( r ) , V ( r ) and ( x, y, z = 0 )
( ( x, y, z = 0 ) = induced surface
Source
charge density on -plane)
r r r
Point, S
q
E ( r ) @ Field Point, P
r
r
-conducting plane
@ V ( x, y , z = 0 ) = 0
induced surface charge

density ( x, y, z = 0 )
wire
Earth ground
Symbol

Fall Semester, 2007
Lect. Notes 6
Prof. Steven Errede
Side View of Problem:
z
Lines of E :
* point radially outward @ +q
* are to surface of -conducting
plane @ z = 0 ( x, y )
+q
E (r )
( x, y , z = 0 )
-conducting plane
V ( x, y, 0) = 0
x
( x points outward from paper)
Now mirror-reflect above problem:

i.e. Let z z and simultaneously let + q ( z ) q ( z )
(more generally let r r for objects in problem)
z
+q
Grounded,
-conducting
plane
E (r )
( x, y , z = 0 )
x
V ( x, y, z = 0) = 0
Now (mentally) remove the grounded, -conducting plane:

Fall Semester, 2007
Lect. Notes 6
Prof. Steven Errede
Side View:
z
Original Charge
@ (0,0,+d)
+q
z = +d
Field / Observation
Point P
d
z = d
Image Charge
@ (0,0, d)
We have replaced -conducting grounded plane (equipotential, V ( x, y, z = 0) = 0 ) with an image

point charge q located at ( x, y, z ) = ( 0, 0, d ) .
The method of images is highly analogous to mirror-type optics problem!!
here we have point object +q at ( x, y, z ) = ( 0, 0, + d ) and plane mirror at ( x, y , z ) = ( x, y, 0 ) .
An image of point object is formed as a point image a distance d behind the mirror; point
image q is located at ( x, y, z ) = ( 0, 0, d ) .
Optical mirrors are essentially equipotentials optics works for electrostatic problems!!!
Mathematical constraint (boundary condition) on V: V ( x, y, z = 0 ) = 0 ( x, y, z = 0 ) .
(i.e. everywhere on -conducting grounded plane.)

Fall Semester, 2007
Lect. Notes 6
Prof. Steven Errede
3-D View of Image Charge Problem for Grounded Infinite Conducting Plane:
Source Point, S1
+q (x,y,z) = (0,0,+d)
We
already
know how
to solve /
do this
problem!!
r1 = r r1
r
r1
Field / Observation
Point, P at (xp,yp,zp)
y
r2 = r r2
r2
d
x
Source Point, S2
(x,y,z) = (0,0,d)
Use Principle of Superposition to solve / determine total potential at observation / field point r :
VTOT ( r ) = ?
Potential @ point ( r )
Potential @ point ( r )
due to point charge

+q @ point ( r1)
due to point charge

q @ point ( r2 )
VTOT ( r ) = V1 ( r ) + V2 ( r )
=
=
1 +q
1 q
4 o r1 4 o r2
q 1 1

4 o r1 r2
separation distances
c
What are r1 & r2 ??
a
Use law of cosines c = a + b 2ab cos

to obtain expressions for r1 + r2 .
2

Fall Semester, 2007
Lect. Notes 6
Prof. Steven Errede
Here it is easier simply to use basic definition of separation distance, i.e.

Let: = 1
+q
d
0
r x 2 + y 2 + z 2
and: 2 = 1 =
r1 =
(x
(x
r1
r1
x1 ) + ( y p y1 ) + ( z p z1 )
0) + ( y p 0) + ( z p d )
r1 = x 2p + y 2p + ( z p d )
r2
(x
(x
x2 ) + ( y p y2 ) + ( z p z2 )
0) + ( y p 0) + ( z p + d )
Law of cosines:
r2
r12 = d 2 + r 2 2dr cos
r2 =
r22 = d 2 + r 2 2dr cos ( )
r2 = x 2p + y 2p + ( z p + d )
cos ( ) = cos cos + sin
=0
sin
= cos
r = d 2 + r 2 + 2dr cos
2
2
Then : VTOT ( r ) =
q 1 1

4 o r1 r2
Drop p on field point subscript:
q
1
VTOT ( r ) =
4 o x 2 + y 2 + ( z d )2
1
x2 + y 2 + ( z + d )
Note that:
1. VTOT ( r @ z = 0 ) = 0 i.e. VTOT 1 = 2 = = = 90o = 0

2
2. VTOT ( r ) = 0
q
1
VTOT ( r ) =
4 o x 2 + y 2 + ( z d )2
1
x2 + y 2 + ( z + d )
=
4 o
1 1
= V1 ( r ) + V2 ( r )
r1 r2
We can now determine ETOT ( r ) from either:

(a.) ETOT ( r ) = E1 ( r ) + E2 ( r ) =
or:
q r1 r2
(using the Principle of Superposition)
4 o r12 r22
What are r1 and r2 ?

(b.) ETOT ( r ) = VTOT ( r ) == x +
y + z VTOT ( r ) (e.g. in Cartesian coordinates)
y
z
x
6

Fall Semester, 2007
Lect. Notes 6
Prof. Steven Errede
Because we will see this same problem again (in the near future) from a different perspective, let
us rewrite the problem in spherical polar coordinates, using the law of cosine results:
{r =
r 2 + d 2 2rd cos
and
r2 = r 2 + d 2 + 2rd cos
q
1
1
4 o r + d 2 2rd cos
r 2 + d 2 + 2rd cos
1
1

ETOT ( r ) = VTOT ( r ) = r +
+
VTOT ( r )
r
r sin
r
VTOT ( r ) =
= ErTOT r + ETOT + ETOT

1
1
q
1
1
VTOT ( r ) =
2
2
2
2
4 o r r + d 2rd cos
r
r + d + 2rd cos
2r 2d cos )
2r + 2d cos )
(
(
q 1
1
=
+

3
3
4 o 2 r 2 + d 2 2rd cos 2 2 r 2 + d 2 + 2rd cos 2
ErTOT =
r d cos )
r + d cos )
(
(
q
3
3
4 o r 2 + d 2 2rd cos 2
r 2 + d 2 + 2rd cos 2
n.b. cos = sin

1
q
1
1
VTOT ( r ) =
2
r
4 o r r + d 2 2rd cos
r 2 + d 2 + 2rd cos
+2rd sin )
2rd sin )
(
(
q 1 1
1
=
+
2
3
3
4 o r 2 r + d 2 2rd cos 2 2 r 2 + d 2 + 2rd cos 2
ETOT =
q 1
=+
r d sin
4 o r
1
1
+
3
3
2
2
2
r 2 + d 2 + 2rd cos 2
r + d 2rd cos
qd sin
1
1
=+
+
3
3
4 o r 2 + d 2 2rd cos 2
r 2 + d 2 + 2rd cos 2
3
i.e.
VTOT ( r ) = 0 because VTOT ( r ) has no -dependence.

r sin
VTOT ( r ) and ETOT ( r ) are azimuthally symmetric (invariant under rotations)
ETOT =
because original charge distribution is azimuthally symmetric no -dependence.

Fall Semester, 2007
Lect. Notes 6
Prof. Steven Errede
=0
Thus, ETOT ( r ) = ErTOT r + ETOT + ETOT , or:
r d cos )
r + d cos )
(
(
q
ETOT ( r ) = +
r
3
3
2
2
2
4 o r 2 + d 2 2rd cos 2
r + d + 2rd cos
+q
d
2d
d
with:

d sin )
d sin )
(
(

+
+
3
3
r 2 + d 2 2rd cos 2
r 2 + d 2 + 2rd cos 2
q
1
1
VTOT ( r ) =
2
4 o r + d 2 2rd cos
r 2 + d 2 + 2rd cos
The above expressions are potential and electric field associated with a spatially-extended
electric dipole, with electric dipole moment p + qr1 qr2 = qr21 (SI Units: Coulomb-meters)
with separation distance r21 r1 r2 = dz d ( z ) = dz + dz = 2dz. Here (i.e. in this problem),
the separation distance r21 = 2d .
z
+q
n.b. The vector r21 r1 r2 points from q to +q important convention!!!
p = 2qdz
2d
q

Fall Semester, 2007
Lect. Notes 6
Prof. Steven Errede
Electric Field E ( r ) and Equipotentials of an Electric Dipole with Electric Dipole Moment, p = pz :
z
electric field lines

equipotentials ( to E everywhere)
+q
We can now determine the surface free charge density free ( x, y, z = 0 ) on the infinite,
grounded conducting plane e.g. via two methods:
METHOD 1: free ( x, y, z = 0 ) = o
i.e. here, free ( x, y, z = 0 ) = o
VTOT ( r )
gradient normal to surface
=
where
= z ( here ) .
of
-conducting
plane
n
surface
VTOT ( r )
z
z =0
q
1
In Cartesian coordinates: VTOT ( x, y, z ) =
4 o x 2 + y 2 + ( z d )2
Thus: free ( x, y, z = 0 ) =
q
1
4 z x 2 + y 2 + ( z d )2
1
x2 + y 2 + ( z + d )
1
x2 + y2 + ( z + d )
z =0
z =0

Fall Semester, 2007
Lect. Notes 6
Prof. Steven Errede
2( z d )
2( z + d )
1

3
3
2 x 2 + y 2 + ( z d )2 2
x 2 + y 2 + ( z + d )2 2
z =0
q
d
d
=+
3
3
4 x 2 + y 2 + d 2 2
x 2 + y 2 + d 2 2
2qd
1
p
1
=
=
3
4 x 2 + y 2 + d 2 2
4 x 2 + y 2 + d 2 3 2
q
=
4
Electric dipole moment p = 2qd = q ( 2d ) where 2d = charge separation distance.

Note that the sign of the induced surface free charge on -conducting plane is opposite to
that of original charge q.
Note also that free ( x, y, z = 0 ) is greatest at ( x = 0, y = 0, z = 0 ) - directly underneath
free
=
original charge, q. i.e. max
2qd 1
2q 1
3=

4 d
4 d 2
See plot of free ( x, y, z = 0 ) below (on p. 12).

METHOD 2: Use Gauss Law:
E idA =
free
Qencl
Use shrunken Gaussian Pillbox of height h centered on / around -conducting plane:

induced free charge ( x, y, 0 ) on
z, n1
surface
S1
E
n2
h
h
E = 0 inside conducting plane (z < 0)
S2
S3
n3 = z
On the conducting surface (@ z = 0), =
10
E 0 above conducting plane (z > 0)

sin = sin = 1 and cos = cos = 0 .
2
2
2

surf
TOT
Fall Semester, 2007
Lect. Notes 6
Prof. Steven Errede

q
r
r
d
d

r+
( x, y , z = 0 ) =
3
3
3
3
4 o r 2 + d 2 2 r 2 + d 2 2 r 2 + d 2 2 r 2 + d 2 2

q
2d
0r +
=
When = = 90o
3
4 o
2
r 2 + d 2 2
Now z = cos r sin
Please remember / derive this!!
Consider r, r
to lie in y z
plane:
z
z = r cos
z = cos

= =
2
2
z = sin
cos = cos = cos cos + sin sin = sin

2
2
2
Thus, when = 90o =

, z = cos r sin = cos r sin =
2
2
2
q ( 2d )
1
surf
ETOT
z
( x, y , z = 0 ) =
4 o r 2 + d 2 3 2
q ( 2d )
1
surf
ETOT
z
( x, y , z = 0 ) =
4 o x 2 + y 2 + d 2 3 2
On conducting plane z = 0
r 2 = x2 + y2 + z 2
r 2 = x 2 + y 2 on conducting plane

11
Gaussian Pillbox Surface:
Fall Semester, 2007
E idA = E1 idA1 +
S1
S2
Lect. Notes 6
=0
E2 idA2 +
E3 idA3
E3 = 0
inside
conductor
E just above surface
S1
free 1
=0
S3
E2 dA2
S2 shrinks 0
anyway
= E1 idA1
Prof. Steven Errede
r3
x, y
area of surface S1
Gauss Law:
q ( 2d )
1
z i A1 z
4 o x 2 + y 2 + d 2 3 2
free
( x, y, z = 0 ) A1
Q
= encl = free
E idA = E1 idA1 =
S1
free ( x, y, z = 0 ) =
q ( 2d )
1
p
1
=
3
4 x 2 + y 2 + d 2 2
4 x 2 + y 2 + d 2 3 2
Electric dipole moment p = q ( 2d ) ( Coulomb-meters ) where 2d = charge separation distance

Same answer as obtained in Method 1.
plane
TOT
NOTE: Q
p
= free ( x, y, 0 ) dA =
plane
4
1
qd
rdrd =
2
2
r2 + d 2
r + d
= q
0
The net force of attraction of charge +q to -conducting plane is just that of force of charge +q
attracted to its image charge, q a separation distance r21 r1 r2 = r = 2d away!!!
NET
+q
+q
( r ) = + qE q ( r = 2d ) =
+ q ( q )
4 o
( 2d )
z =
q2
4 o ( 2d )2
r = 2d
q
E q
12

Fall Semester, 2007
Lect. Notes 6
Prof. Steven Errede
We can also obtain the net force of attraction of the charge +q and grounded, infinite conducting
plane by adding up all of the individual contributions qdE ( r = ( 0, 0, d ) ) due to free ( x, y, z = 0 ) :
q ( 2d )
1
2qd
1
=
3
4 x 2 + y 2 + d 2 2
4 x 2 + y 2 + d 2 3 2
1
d
1
free ( x, y, 0 ) 2 r dA where dA = 2 rdr and r = cos z = z
4 o
r
r
free ( x, y, z = 0 ) =
F+NET
( r ) = q dETOT = q
q
S
plane
q 2 ( 2d ) 1
1
d
1
2
2 rdr z
0
3
2
4 4 o r 2 + d 2 2 r + d r 2 + d 2
2 2
qd
rdr
1
q2
=
=
z
z
4 o 0 r 2 + d 2 3
4 o ( 2d )2
F+NET
(r ) =
q
Integration over the conducting plane:

r
r 2 = x2 + y 2 + d 2
d
dr
The work done to assemble the Image Charge Problem (i.e. put +q first at (x,y,z = d) and then
bring in q at (x,y,z = d) from ) is:
1
2d
WICP = Fmech idl = Fmech ( 2d ) =
Also:
WICP =
o
2
all
space
E 2 d =
o
2
all
space
q2
4 o ( 2d )
E i E d =
o
2
2d =
2
( E iE
r
q2
(Joules)
4 o ( 2d )
1
+ E i E ) d
all
space
Note that this integral includes both the z > 0 and z < 0 regions for the image charge problem.

13
Fall Semester, 2007
Lect. Notes 6
Prof. Steven Errede
However, for the actual problem, i.e. the charge +q above grounded -conducting plane,
there is NO ELECTRIC FIELD in the z < 0 region!
1
Thus Wactual = WICP
2
1 1 q2
i.e. Wactual =
2 4 o ( 2d )
For the actual problem we can obtain Wactual by calculating the work required to bring +q in
from infinity to a distance d above the grounded -conducting plane. The mechanical force
required to oppose the electrical force of attraction is:
Fmech = FE =
q2
4 o ( 2 z )2
z (along z axis)
4 o
Wactual = Fmech idl =
q2
( 2z )
dz =
2
q2
16 o
dz
z2
q 2 1
q2
=
=
16 o z
16 o d
1 1 q2
Wactual =
2 4 o ( 2d )
Same answer as that obtained above!
IMPORTANT NOTES / COMMENTS ON IMAGE CHARGE PROBLEMS
1.) Image charges are always located outside of regions(s) where V ( r ) and E ( r ) are to be
calculated!!
Image charges cannot / must not be located inside region where V ( r ) and E ( r ) are to be
calculated (no longer the same problem!!)
2.) WICP (all space) = 2Wactual (half space).
In general, this is not true image charge problems. Be careful here! Depends on detailed
geometry of conducting surfaces.
3.) Depending on nature of problem, image charge(s) may or may not be opposite charge sign!!
4.) Depending on nature of problem, image charge(s) may or may not be same strength as
original charge Q.
14

Fall Semester, 2007
Lect. Notes 6
Prof. Steven Errede
Image Charge Problem
Example 2: (Griffiths Example 3.2 p. 124-126)

Point charge +q situated a distance a away from the center of a grounded conducting sphere of
radius R < a. Find the potential outside the sphere.
V(r = R) = 0
on sphere
(equipotential)
R
y
+q
a
Take origin of coordinates to be @ center of
sphere
x
From spherical and y axial symmetry (rotational invariance) of problem, if solution for image
charge q is to exist, it must be:
1.) inside spherical conductor (r < R)
2.) q image charge must lie along y axis (i.e. along line from charge +q to center of
sphere).
3.) because V (r = R) = 0 on sphere, q must be opposite charge sign of +q.
4.) want to replace grounded conducting sphere with equipotential V (r = R) = 0 by use
of image charge q at distance b away from center of sphere:
R
b
+q

15
Fall Semester, 2007
Lect. Notes 6
Prof. Steven Errede
NOTE: Two points on the surface of sphere where the potential VTOT (r = R) = 0 is easy to
calculate - is on the y axis at the field points P1 and P2:
r1 = r r1 and r2 = r r2
V (r = R) = 0
P3
P (General Field Point)

r
R
P1
b q
r2
r2
r1
P2
r1
+q
In general:
At point P1:
VP1 ( r = R ) = 0
1 q q
+
4 o r1 r2
r1 = ay , r = R ( y ) = Ry , r1 = r r1 = Ry ay = ( R + a ) y
VTOT ( r ) = V1 ( r ) + V2 ( r ) =
r1 = r1 = ( R + a )
r2 = by , r = R ( y ) = Ry , r2 = r r2 = Ry by = ( R + b ) y
r2 = r2 = ( R + b )
1 q q
+
4 o r1 r2
1 q
q
=
+
=0
4 o ( R + a ) ( R + b )
VP1 ( r = R ) =
q
q
=
( R + a) ( R + b)
Relation #1
At point P2:
VP2 ( r = R ) = 0
r1 = ay , r = R ( + y ) = + Ry , r1 = r r1 = Ry ay = ( R a ) y
r1 = r1 = ( a R )
(a > R) !!
r2 = by , r = R ( + y ) = + Ry , r2 = r r2 = Ry by = ( R b ) y
r2 = r2 = ( R b )
VP2 ( r = R ) =
1 q
q
+
=0
4 o ( R a ) ( R b )
q
q
=
(a R) ( R b)
Relation #2
16

Fall Semester, 2007
Lect. Notes 6
Prof. Steven Errede
We now have two equations (Relations # 1 & 2), and we have two unknowns: q and b.
Solve equations simultaneously!
- First, we eliminate q :
R+b
From Relation #1 we have: q ' =
q
R + a
Rb
q
From Relation #2 we have: q ' =
a R
R+b Rb

=
OR: ( R + b )( a R ) = ( R + a )( R b )
R + a a R
R 2 + aR + ab bR = R 2 + aR bR ab
2 R 2 + 2ab = 0
OR: ab=R2
2
b=R
OR:
a
R+b
Then: q ' =
q
R + a
R + R2
1 + R
a 1 + R
a q
a q = R
a q = R
=
R+a
R+a
a [ R + a]
R
R [a + R]
R [ R + a]
q = q
q =
=
a
a [ R + a]
a [ R + a]
R
Thus: q ' = q
a

17
Fall Semester, 2007
Lect. Notes 6
Prof. Steven Errede
CHECK:
R2
R
Does q ' = q , located at r2 = by = y satisfy the B.C. that V ( r = R ) = 0 for any r = R?
a
a
VTOT ( r ) = V1 ( r ) + V2 ( r ) =
1 q q
+
4 o r1 r2
At an arbitrary field point P3 anywhere on the surface of sphere, r = R:
r1 = a 2 + R 2 2aR cos
and
r2 = b 2 + R 2 2bR cos
Note that we changed to z axis here in order
to define (& use) the polar angle, !!!
P3
R
r2
q
b
r1
+q
Then: VTOT ( r ) =
1 q q
+ with:
4 o r1 r2
R
q ' = q
a
R2
b=
a
q
1
VTOT ( r = R ) =
2
2
4 o a + R 2aR cos
18
q
4 o
q
4 o
( R a)
2
2
b + R 2bR cos
1
a
2
2
2
a + R 2aR cos
R2
+ R2 2 R
R cos
a
a
1
1
2
2
2
3
a + R 2aR cos
a
R2
+ R2 2 R
cos
R
a
a
( )
( ) (
( )
( )
( )

Fall Semester, 2007
q
1
VTOT ( r = R ) =
2
4 o a + R 2 2aR cos
q
1
=
2
4 o a + R 2 2aR cos
= 0 ,
(@ r = R )
Lect. Notes 6
Prof. Steven Errede
2
a
R 4 2 + a 2 2aR cos
R
a
a 2 + R 2 2aR cos
1
( )(
YES!!!
The scalar potential for an arbitrary point outside the grounded, conducting sphere (r > R) is:
1
q
VTOT ( r ) =
2 2
4 o a + r 2ar cos
Then: ETOT ( r ) = VTOT ( r ) and thus:

And thus: free ( r = R ) = o
2
2
2
R2
R
r cos
+r 2
a
a
( )
( R a)
( )
Er ( r ) =
VTOT ( r )
r
VTOT ( r )
V ( r )
= o TOT
= + o Er ( r )
n
r
r =R
r=R
total
Total charge on surface of the conducting sphere:
Q sphere
=
free
sphere
r=R
R
q
a
free dA =

19
Fall Semester, 2007
Image Charge Problem
Lect. Notes 6
+q
Prof. Steven Errede
a
Example #3:
Point charge +q near a charged conducting sphere of radius R.

(variation on image charge Example #2)
Use the superposition principle for image charges!!
( a ) q located at
Step 1: Replace the conducting sphere by an image charge q ' = R
( a ) y
2
rq = by = R
(same as in Example #2)
This makes surface of sphere an equipotential surface V (r = R) = 0.

Step 2: Add a second image charge q at center of sphere to raise potential on surface of
sphere to achieve required potential V (r = R) = V (positive or negative constant
potential on sphere)
Note: q is also on same axis ( y ) as q and q ' .
Then: TOT
free = free ( q ) + free ( q )
Surface free charge

Density due to q
surface free charge

density due to q
Then: VTOT ( r ) = Vq ( r ) + Vq ' ( r ) + Vq '' ( r ) =
( a) q
But: q ' = R
1 q q q
+ +
4 o rq rq' rq''
=r
and
Qsphere = q + q
Then: ETOT ( r ) = VTOT ( r ) and
TOT
free = o
VTOT ( r )
V ( r )
= o TOT
= + o Er ( r )
n
r
r =R
r=R
Since: Er ( r ) =
20
r =R
VTOT ( r )
r


Lecture Notes 6: The Method of Images

Uploaded by

Copyright:

Available Formats

Lecture Notes 6: The Method of Images

Uploaded by

Document Information

Original Description:

Original Title

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

Lecture Notes 6: The Method of Images

Uploaded by

Copyright:

Available Formats

UIUC Physics 435 EM Fields & Sources I

Fall Semester, 2007

Prof. Steven Errede

Can replace e.g. a charged surface of a conductor (which is at constant potential an

The method of images is best learned by example

Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois

UIUC Physics 435 EM Fields & Sources I

Fall Semester, 2007

Prof. Steven Errede

Point Charge q Located Near An Infinite, Grounded Conducting Plane:

n.b. A grounded conductor is a special type of equipotential: infinite amounts of electrical

charge density on -plane)

induced surface charge

Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois

UIUC Physics 435 EM Fields & Sources I

Fall Semester, 2007

Prof. Steven Errede

Side View of Problem:

Now mirror-reflect above problem:

Now (mentally) remove the grounded, -conducting plane:

UIUC Physics 435 EM Fields & Sources I

Fall Semester, 2007

Prof. Steven Errede

We have replaced -conducting grounded plane (equipotential, V ( x, y, z = 0) = 0 ) with an image

Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois

UIUC Physics 435 EM Fields & Sources I

Fall Semester, 2007

Prof. Steven Errede

due to point charge

due to point charge

Use law of cosines c = a + b 2ab cos

Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois

UIUC Physics 435 EM Fields & Sources I

Fall Semester, 2007

Prof. Steven Errede

Here it is easier simply to use basic definition of separation distance, i.e.

r12 = d 2 + r 2 2dr cos

r22 = d 2 + r 2 2dr cos ( )

cos ( ) = cos cos + sin

Drop p on field point subscript:

1. VTOT ( r @ z = 0 ) = 0 i.e. VTOT 1 = 2 = = = 90o = 0

We can now determine ETOT ( r ) from either:

What are r1 and r2 ?

Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois

UIUC Physics 435 EM Fields & Sources I

Fall Semester, 2007

Prof. Steven Errede

= ErTOT r + ETOT + ETOT

n.b. cos = sin

VTOT ( r ) = 0 because VTOT ( r ) has no -dependence.

because original charge distribution is azimuthally symmetric no -dependence.

Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois

UIUC Physics 435 EM Fields & Sources I

Fall Semester, 2007

Prof. Steven Errede

Thus, ETOT ( r ) = ErTOT r + ETOT + ETOT , or:

n.b. The vector r21 r1 r2 points from q to +q important convention!!!

Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois