UIUC Physics 435 EM Fields & Sources I Fall Semester, 2007 Lect. Notes 5 Prof.

Steven Errede

LECTURE NOTES 5
Materials made up of normal matter (atoms, molecules, etc.) have some amazing
electromagnetic properties!

Simplest kinds of electromagnetic properties:

A.) conductor (of electricity) ⎫ Why materials conduct vs. do not conduct electricity
⎪
B.) partial conductor/insulator ⎬ depends on microscopic (i.e. quantum) structure of
C.) non-conductor ⇒ insulator ⎭⎪ materials & temperature (i.e. thermal/internal energy).

CONDUCTORS:

"normal" good conductors of electricity: ⎞ Have finite DC resistance, R = V/I (Ohm’s Law)
⎟
metals - gold, platinum, silver, copper... ⎠ @ finite temperatures, T > 0 K.

"superconductors" - low temperature SC's such ⎞

⎟ DC resistance vanishes below Tc (critical temp)
as lead (Tc ~4K) indium, niobium, ..... ⎟
Hi-Tc SC's (e.g. Tc ~77K): BSCO, YBCO ..... ⎟⎠

INSULATORS:
e.g. plastics, teflon, glass, rubber ….

PARTIAL CONDUCTORS:
e.g. doped plastics, semi-conductors (germanium, silicon, graphite….)

IONIC LIQUIDS:
e.g. salt water – can also conduct electricity
Acidic solutions – ions transport electrical charges – not electrons

An ideal/perfect conductor is a (hypothetical) material that would have an unlimited number of

completely free electrons/free charges. No such things truly exist in nature, but ∃ many
materials which do come (amazingly) close to an ideal/perfect conductor.

©Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois 1

2005 - 2008. All rights reserved.
 UIUC Physics 435 EM Fields & Sources I Fall Semester, 2007 Lect. Notes 5 Prof. Steven Errede

One important property of a conductor is that:

n.b. ENET ( r ) ≡ 0 is a different physics statement
1) ENET ( r ) ≡ 0 inside a conductor than E ( r ) ≡ 0 inside a conductor!

If ENET ( r ) = 0 inside a conductor, then free charges inside the conductor would move/be
accelerated by this Einside ( r ) = 0 , because: Finside ( r ) = qe Einside ( r ) = me ae ( r ) .

Indeed, this is precisely what happens inside a conductor e.g. when it is placed in a uniform
external electric field Eext ( r ) = E0 xˆ – the free charges inside the conductor (electrically neutral!)
re-distribute themselves to create/produce Einside ( r ) = 0 on extremely short timescales of ~ femto
→ pico-seconds ( ~ 10−15 − 10−12 sec ) .

The redistributed free charges pile up on the surface(s) of the conductor in such a way as to
produce Einside ( r ) = 0 . These induced charges produce an internal electric field of their own,
which exactly cancels the external field, Eext ( r ) !

e.g. consider a block of metal in a uniform external electric field, Eext ( r ) = E0 xˆ

Enet ( r ) = Eext ( r ) + Einduced ( r ) = 0

inside inside

⇒ Einduced ( r ) = − Eext ( r ) = − E0 xˆ
inside

Note that: Enet ( r ) = Eext ( r ) when a conductor is placed inside an initially uniform applied
outside

field, Eext ( r )

Induced surface charge, σ  Induced surface charge, σ ⊕

⎧  ← ⊕ ⎫
⎪  ← ⊕ ⎪
⎪ ⎪
⎪  ← ⊕ ⎪
⎪ ⎪
⎪  ← ⊕ ⎪
⎪  ← ⊕ ⎪
⎪ ⎪
Eext ( r ) = E0 xˆ ⎨  E (r ) ⊕ ⎬ Eext ( r ) = E0 xˆ
⎪  inside ⊕ ⎪
⎪ ⎪
⎪  ← ⊕ ⎪
⎪ ⎪
⎪  ← ⊕ ⎪
⎪  ← ⊕ ⎪
⎪ ⎪
⎩  ← ⊕ ⎭
E0 xˆ x̂

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

2005 - 2008. All rights reserved.
 UIUC Physics 435 EM Fields & Sources I Fall Semester, 2007 Lect. Notes 5 Prof. Steven Errede

Another important property of a conductor is that:

2) The volume free charge density, ρinside

free
( r ) = 0 inside a conductor.

This follows from Gauss’ Law (differential form):

∇ ⋅ Einside ( r ) = ρinside
free
(r ) εo
but if: Einside ( r ) = Eext ( r ) + Einduced ( r ) = 0, ∀r (i.e. everywhere inside conductor)
inside

then: ∇ ⋅ Einside ( r ) = 0, ∀rinside

⇒ ρinside
free
( r ) = 0, ∀rinside
3) Any induced charges on a conductor can ONLY reside on surface(s) of the conductor
– as surface charge distributions, σ free

(n.b. free surface charges minimize overall potential energy (i.e. maximize overall entropy of
system!))

If ∃ induced free charges, and ρinside

free
( r ) = 0 ∀r , the only place(s) such induced free charges can
reside is on the surface(s) of the conductor, as σ free ( r ) .

4) The entire volume & surface of a conductor is an equipotential.

If a & b are two arbitrary points, ra & rb on the surface of a conductor, the
must!!
potential difference, ΔVab ≡ Vb − Va = V ( rb ) − V ( ra ) = − ∫ E ( r )idl = 0 ⇒ Va = Vb
b

If ΔVab ≠ 0 then free charges will move!!

©Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois 3

2005 - 2008. All rights reserved.
 UIUC Physics 435 EM Fields & Sources I Fall Semester, 2007 Lect. Notes 5 Prof. Steven Errede

5) Just outside the surface of a conductor, Eoutside ( r @ surface ) is

perpendicular/normal to the surface, i.e. Eoutside ( r @ surface ) nˆsurface

If ∃ an E ( r ) = Etangential ( r ) @ surface, ⇒ free charge will move/flow!!!

This can’t happen in electrostatics!!

⎛ Coulombs ⎞
Example: A conducting surface of total surface area, Acond is charged with σ free = σ o ⎜ ⎟.
⎝ m2 ⎠
What is the electric field strength (i.e. electric field intensity) at the surface of the conductor,
Esurface ?

⇒ Use (the integral form of) Gauss’ Law – choose fully enclosing Gaussian surface S to be
infinitesimally above the conducting surface (e.g. δ h larger than conducting surface everywhere).
Then take limit as δ h → 0 :
Q σ free Acond
∫ E idA = encl = Esurface ( r ) nˆ everywhere.
S ε0 ε0

= Esurface Acond
σ free σ o
∴ Esurface = Esurface = = (Volts m )
εo εo

n.b. If we had instead used e.g. a “shrunken” Gaussian pillbox on surface of conductor:
Q
∫ E idA = ∫ Esurface idA1 + ∫ E tangential idA2 + ∫ Einside idA3 = encl
=0 =0 =0
S S1 S2 S3 εo
Qencl σ oπ R 2 σ free σ
π R 2 Esurface = = ⇒ Esurface = nˆ = o nˆ
ε0 εo εo εo

Get same answer! (We should!)

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

2005 - 2008. All rights reserved.
 UIUC Physics 435 EM Fields & Sources I Fall Semester, 2007 Lect. Notes 5 Prof. Steven Errede

The Free Surface Charge σ free

The Surface Electric field, E surface ( r )
The Surface Potential, V ( r ) ,
Electrostatic Force & Pressure
(Force per Unit Area) Acting on a Conductor

σ free
We have derived, using Gauss’ Law: Esurface ( r ) = = Esurface ( r )
εo

In vector notation: σ free Where n̂ is the outward pointing unit

Esurface ( r ) = nˆ
εo normal vector (outward = by convention).

From Griffiths Eqn’s 2.34-2.37, p. 89-90:

n̂
⎡⎣Vbabove ( rb ) − Vabelow ( ra ) ⎤⎦ lim
b b
Eabove = − ∫ E i d = ∫ ∇V i d = 0
a a
δ h→0

b ( surface of conductor is an equipotential )

½ δh

=0 σ free
½ δh Gave: ∇V above ( r ) − ∇V below ( r ) =− nˆ
ε0
a
∂V ( r ) σ
Ebelow OR: ≡ ∇V ( r )inˆ = − free
surface of conductor ∂n surface surface εo

∂V ( r )
OR: σ free = − ε o
∂n surface

If Esurface ( r ) or Vsurface ( r ) is known, then σ free can be obtained (and/or vice-versa).

©Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois 5

2005 - 2008. All rights reserved.
 UIUC Physics 435 EM Fields & Sources I Fall Semester, 2007 Lect. Notes 5 Prof. Steven Errede

FORCE & PRESSURE ON A CONDUCTOR

Consider an arbitrarily-shaped conductor, with σ free coulombs

m2 ( )
free charge residing on its
surface. Infinitesimally small surface “patch,” of surface area Apatch = π R 2 .

n̂

• R

Apatch = π R 2

Now Apatch also has surface charge σ free on it. The total charge of this patch is Qpatch = π R 2σ free .

Edge-on view of infinitesimally small patch:

σ free σ
nˆ , E patch
above
= ( + nˆ ) = + free nˆ
2ε o 2ε o

+++++++ σ free on Apatch ( = π R 2 )

σ free σ
patch =
E below ( −nˆ ) = − free nˆ
2ε o 2ε o

We have discussed before (Griffiths Ex. 2.4, p. 73-74 and/or P435 Lect. Notes 2 p. 9-12) that a
surface charge σ has a (net) E -field ⊥ to surface on which σ resides, both above and below
the surface.

The transverse/tangential E -field components (on a flat surface, from symmetry) were shown to
cancel. (n.b. an infinitesimally small surface patch is FLAT.)

But we also said that Einside ( r ) = 0 inside a conductor! How do we reconcile these two
statements?
Consider an edge-on view of the (arbitrarily shaped conductor, but with the (infinitesimally)
small surface patch (of area Apatch) removed:
σ free on surface
of conductor

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

2005 - 2008. All rights reserved.
 UIUC Physics 435 EM Fields & Sources I Fall Semester, 2007 Lect. Notes 5 Prof. Steven Errede

Now consider just the electrostatic surface charge itself – i.e. mentally “erase” the presence of
the conductor altogether: Enet ( r ) circular “hole” where infinitesimal
n̂ patch of area Apatch = π R 2 was
(now removed)

The surface charge distribution is a (very!) special configuration that is an equipotential surface
– but it is still a collection of individual point charges (@ the microscopic level). Each individual
point charge q has its own radial electric field, Epoint ( r )
E point

We ask: what is the net electric field Enet ( r ) on the conductor’s surface? (e.g. at the location of
the “hole” where the patch of infinitesimal area Apatch = π R 2 was (which is now removed)).
It’s hard to rigorously prove the following, for an arbitrarily-shaped charged conductor, but it
can be rigorously proved (analytically) for symmetrically-shaped conductors – e.g. a charged
sphere, with an infinitesimally small spherical cap removed, e.g. at its north pole. (see e.g. J.D.
Jackson’s book Electrodynamics for this proof…)
In general, for arbitrarily-shaped conductors, because their charged surfaces are equipotential
surfaces, the net electric field, Enet ( rpatch ) on the conductor’s surface, right at the “hole” location
of the (missing/removed) patch, associated with the free charge everywhere else on the
conductor’s surface, is:
σ free
Enet ( rpatch ) = + nˆ (!!!)
2ε o

infinitesimally above/below surface, @ the patch “hole” location:

σ free
Enet ( rpatch ) = Enet
above
( rpatch ) = Enetbelow ( rpatch ) = + nˆ
2ε 0

Thus, we see that: above outside

EToT ( rpatch ) = E patch
above
( rpatch ) + Enetabove ( rpatch )
σ free σ σ
above
EToT ( rp ) = 2ε o
nˆ + free nˆ = free nˆ
2ε o εo
And we also see that:
below inside
EToT ( rpatch ) = E patch
below
( rpatch ) + Enetbelow ( rpatch )
σ free σ
below
EToT ( rpatch ) = − 2ε o
nˆ + free nˆ = 0
2ε o

©Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois 7

2005 - 2008. All rights reserved.
 UIUC Physics 435 EM Fields & Sources I Fall Semester, 2007 Lect. Notes 5 Prof. Steven Errede

Thus, the net electric field just above the surface of a conductor arises from two equal
contributions – the E patch from σ free on the infinitesimal patch of area Apatch = π R 2 itself, and the
net contribution from the free charge on the remainder of the conducting surface! Likewise, the
internal E field inside the conductor = 0, because these two contributions cancel each other!!!
Thus, VECTORIALLY, in the region of the infinitesimally small patch of area Apatch = π R on/at
the surface of the conductor, we have:
n̂

σ free
above outside
EToT ( rpatch ) = εo
nˆ

Apatch
•

below inside
EToT (r ) = 0

above outside
EToT ( rpatch ) = E patch
above
( rpatch ) + Enetabove ( rpatch ) below inside
EToT ( rpatch ) = E patch
below
( rpatch ) + Enetbelow ( rpatch )
σ free σ σ σ free σ
above
EToT ( rp ) = 2ε o
nˆ + free nˆ = free nˆ
2ε o εo
below
EToT ( rpatch ) = − 2ε o
nˆ + free nˆ = 0
2ε o

We are now in a position to ask: what is the net/total force, Fpatch

total
( r ) acting on the
infinitesimally small patch, of area Apatch = π R 2 ?
1) There can be no contribution(s) to the net/total force, Fpatch
total
( r ) acting on the patch due to the
free surface charge, σ free (or associated E − fields E patch
above
( r ) + E below
patch ( r ) ) acting on the patch

itself (analogous to trying to lift yourself up by pulling on your shoes!)

total
Fpatch ( rpatch ) = Qpatch * Eat patch ( rpatch ) = Qpatch * Enet ( rpatch ) net E -field @ patch from σ free
elsewhere on conducting surface
⎛σ ⎞
total
Fpatch ( rpatch ) = σ free Apatch * Enet ( rpatch ) = σ free Apatch * ⎜ 2εfree nˆ ⎟
⎝ o ⎠

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

2005 - 2008. All rights reserved.
 UIUC Physics 435 EM Fields & Sources I Fall Semester, 2007 Lect. Notes 5 Prof. Steven Errede

σ 2free ( rpatch )
F total
patch (r ) =
patch
2ε o
Apatch nˆ NOTE: this is a force which points in the outward
normal (⊥) direction!

If we sum up the infinitesimal force contributions from all the “patches” associated with the
entirety of the conducting surface, the magnitude of the force is:
σ 2free ( rpatch ) σ 2free
Fconductor ( rsurface ) = ∑ Fpatch ( rpatch ) = ∫
N
n.b. Assumption made here
total
dA = Aconductor
i =1
conductor 2ε o 2ε o is that σ free ( r ) = constant

Now, pressure ≡ force/unit area, i.e. P = F

A
Thus:
1 ⎛ σ free ( rsurface ) ⎞ ε o 2
2

Then, electrostatic pressure, P@ surface ( rsurface ) = ε o ⎜ ⎟ = E@ surface ( rsurface )

2 ⎜ ε 2
⎟ 2
⎝ o ⎠

Since: σ free ( rpatch ) ⎛ σ free ( rpatch ) ⎞

Esurface ( rpatch ) = 2 Enet ( rpatch ) = 2 nˆ = ⎜ ⎟ nˆ
2ε o ⎜ ε ⎟
⎝ o ⎠

1 ⎛ σ free ( rsurface ) ⎞ ε o 2
2

Then, electrostatic pressure, P@ surface ( rsurface ) = ε o ⎜ ⎟ = E@ surface ( rsurface )

2 ⎜ ε 2
⎟ 2
⎝ o ⎠

n.b. If electrostatic forces are able/capable of performing mechanical work W, then must do so at
the expense of electrostatic energy (recall that total energy must always be conserved!)
n.b. metals are elastic solids!!
⇒ Suppose a small area, a of the conductor is pulled into the electric field region by an
infinitesimal distance, Δx . Then work done by field on area, a is: F
area, a

Δx

⎛ F (r ) ⎞
ΔW ( r ) = F ( r )iΔx = ⎜ ⎟ ( aΔx ) = P@ surface ( r ) * aΔx = P@ surface ( r ) ΔV
F Δx
⎝ a ⎠ = volume, ΔV

OR: ΔW ( rsurface )
∴ ΔW ( rsurface ) = P@ surface ( rsurface ) ΔV P@ surface ( rsurface ) = = Energy Density, uE ( rsurface )
ΔV

Electrostatic Pressure, P (@ surface) = Volume Energy Density of Electrostatic Field, uE (@surface)

1
P@ surface ( r @ surface ) = U E @ surface ( r @ surface ) = ε o E@2 surface ( r @ surface )
2
Joules 3
m ( )
n.b. Electrostatic field exerts a negative pressure on conductor – i.e. electrostatic force on the
conductor pulls conductor into electrostatic field !!!

©Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois 9

2005 - 2008. All rights reserved.
 UIUC Physics 435 EM Fields & Sources I Fall Semester, 2007 Lect. Notes 5 Prof. Steven Errede

CAPACITORS
A capacitor is a device that enables the storage (long and/or short term) of electric charge, Q.
Since there are electric fields associated with electric charge, a capacitor is also a device that
enables the storage (long and/or short term) of electrical energy.

One can make a simple capacitor using e.g. two parallel, very thin, conducting plates, separated
by a distance d and initially uncharged. Then, we put charge +Q e.g. on the upper plate and put
charge –Q on the lower plate. A potential difference, ΔVab = Vb ( z = d ) − Va ( z = 0 ) equipotential
surface now exists between the two plates, as shown in the figure below:
Vb ( z = d ) point b +Q on upper plate
+++++++++++++++++
+ + z=d
Potential +++++++++++++++++
Difference E (r )
d = plate separation

ΔVab = Vb ( z = d ) − Va ( z = 0 ) −−−−−−−−−−−−−−−−−−−−
− − z=0
Equipotential Surface −−−−−−−−−−−−−−−−−−−−
Va ( z = 0 ) point a −Q on lower plate

A static electric field E ( r ) exists between the parallel plates (n.b. If the length × width (L × W)
dimensions are both large in comparison to the plate separation, d i.e. L d and W d then
the electric field E ( r ) between the parallel plates will be nearly uniform inside the gap region –
i.e. E ( r ) E0 ( − zˆ ) = − E0 zˆ ← constant
We know that ΔVab = Vb ( z = d ) − Va ( z = 0 ) = − ∫ E ( r )id and that ∫ E ( r )i d
b b
is independent
a a
of the path taken from a → b .

So let’s simply take a path straight up along the ẑ axis − i.e. one that just barely touches the
insides of the plates:
Vb b +Q z=d we know:
+++++++++++++++ E = E0 ( − zˆ ) = − E0 zˆ
zˆ d d =d ( + zˆ ) = + d zˆ
−−−−−−−−−−−−−−−
Va a −Q z=0 E = constant, ≠ fcn of ( z )
both

ΔVAb = Vb ( z = d ) − Va ( z = 0 ) = − ∫ E id = + Ed , for L, W
b
d
a

both

E=
ΔVab Volts
d m( )
= constant for L, W d.

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

2005 - 2008. All rights reserved.
 UIUC Physics 435 EM Fields & Sources I Fall Semester, 2007 Lect. Notes 5 Prof. Steven Errede

Then one can also see that if E = constant between conducting

⎛ both
⎞
plates ⎜ for L, W d ⎟ then ΔV ( z ) = Ez, where z is measured from the bottom plate (z = 0).
⎜ ⎟
⎝ ⎠

From basic E&M principles, we know that E is linearly proportional to charge, Q

⎛ 1 ⎛Q⎞ ⎞
⎜ cf : E ( r ) = ⎜ 2 ⎟ rˆ for a point charge ⎟ .
⎝ 4πε o ⎝ r ⎠ ⎠

(
Similarly, the potential, V is also linearly proportional to the charge, Q cf : ΔV = − ∫ E id
c
)
The constant of proportionality is known as the capacitance, C of the system, i.e. we define
capacitance as the ratio of charge, Q to potential difference, ΔV associated with this system:
S.I. units of capacitance is FARADS, F
Q 1 Coulomb
C≡ 1 FARAD =
ΔV per volt
pot'l difference

both

Thus, for the case of the parallel plate capacitor, for L, W d we see that:
Q Q
capacitance of parallel plate capacitor ( L, W d) C≡ =
ΔVAB Ed

Using Gauss’ Law (integral form) on (any) one of the parallel plates - e.g. the upper plate – use
shrunken Gaussian pillbox as shown in figure below:

©Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois 11

2005 - 2008. All rights reserved.
 UIUC Physics 435 EM Fields & Sources I Fall Semester, 2007 Lect. Notes 5 Prof. Steven Errede

Then:
Qencl
∫ s
E idA =
εo
dA1 = dA1 zˆ, E zˆ
=0
= ∫ E idA1 + ∫ E idA2 + ∫ E idA3 dA2 = dA2 ρˆ ( in x-y plane )
S1 S2 S3

E ⊥dA2
everywhere
on S2

dA3 = − dA3 zˆ
=0
= ( 0 zˆ )i( Adisk zˆ ) + ( − Ezˆ )i( − Adisk zˆ ) AS1 = AS3 = Adisk = π R 2
E = 0 inside upper plate (z > d )

Qencl = +σ Adisk

Qencl σ Adisk
= + EAdisk = =+
εo εo
σ
∴ E=
εo
Q Q Q
∴ C= = = εo but Q = σ Aplate
ΔVab Ed σd

ε o Aplate both

∴ C= for parallel plate capacitor, with L,W d

d

⇒ Note that capacitance, C depends only on ε o (electric permittivity of free space, 8.85×10−12
Farads/m) and (purely) geometrical factors (Aplate & d).

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

2005 - 2008. All rights reserved.
 UIUC Physics 435 EM Fields & Sources I Fall Semester, 2007 Lect. Notes 5 Prof. Steven Errede

Forces on a Parallel Plate Capacitor

Suppose two parallel conducting plates, each of area A = L × W are separated by small
distance d L,W .
The potential difference between plates is initially ΔV . The top plate has surface charge density
+σ , the bottom plate has surface charge density −σ .

Top plate plate

+σ d separation
distance

b ẑ
potential Ο ŷ
ΔV ΔV
difference ΔV −σ E= ( − zˆ ) = − zˆ x̂
d d
a Bottom plate

ΔV = ΔVab = Vbtop − Vabottom

Neglecting fringe fields effects @ edges of parallel plate capacitor (valid approximation if
d L, W ):

force ⎛ Joules ⎞
Pressure = = field energy density uE ⎜ 3 ⎟
per unit area ⎝ m ⎠
2
F 1 1 ⎛ ΔV ⎞
P= = uE = ε o E 2 = ε o ⎜ ⎟
A 2 2 ⎝ d ⎠

2
1 1 ⎛ ΔV ⎞
Attractive force on (each) plate: F = P ⋅ A = uE ⋅ A = ε o E 2 A = ε o ⎜ ⎟ A (Newtons)
2 2 ⎝ d ⎠

Side view:
ẑ
+σ
F top = − Fzˆ E
F bot = + Fzˆ x̂ ŷ
−σ

⎛ -plate ⎞
F ToT ⎜ ⎟ =F +F = − Fzˆ + Fzˆ = 0
TOP BOT

⎝ capacitor ⎠

No net force acts on parallel plate capacitor − Newton’s 1st Law: for every action, ∃ equal and
opposite reaction.

©Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois 13

2005 - 2008. All rights reserved.
 UIUC Physics 435 EM Fields & Sources I Fall Semester, 2007 Lect. Notes 5 Prof. Steven Errede

Griffiths EXAMPLE 2.11

Find the capacitance, C of two concentric spherical metal shells, with radii a & b, b > a.

Place +Q on inner sphere and place –Q on outer sphere. Use Gauss’ Law and show that
1 ⎛Q⎞
for a ≤ r ≤ b : E ( r ) = ⎜ ⎟ rˆ
4πε o ⎝ r 2 ⎠

⎧ b
⎪ΔVab = − ∫ E id
⎪ a
⎪ b
Then: ⎨ = − ∫ E idr
⎪
a

⎪ Q b 1 Q ⎛1 1⎞
⎪
⎩
=− ∫
4πε o ra 2
dr = ⎜ − ⎟ b>a
4πε o ⎝ a b ⎠

Then: Q 1 ab
Cconcentric ≡ = 4πε o = 4πε o
ΔVab ⎛1 1⎞ (b − a )
⎜ − ⎟
spheres

⎝a b⎠

Again, note that C depends only on ε o and (purely) geometrical factors (radii a & b)

How much work, W is done in charging up a capacitor − e.q. (if it is initially uncharged)?

Charging an initially uncharged capacitor means individually removing electrons from the upper plate
of the parallel-plate capacitor (inner sphere of concentric spherical capacitor) and transporting them
to the lower plate of the parallel-plate capacitor (outer sphere of concentric spherical capacitor),
respectively.
W
If ΔVab = (Griffiths 2.38, p.9) then W = QΔVab
Q
The infinitesimal amount of work dW needed to transport an infinitesimal amount of charge, dQ is
dW = ΔVdQ
Q ⎛Q⎞
But: ΔV = Therefore: dW = ⎜ ⎟ dQ
C ⎝C ⎠

The total work done in charging a capacitor from Q = 0 to Q = QToT is:

Q = QToT ⎛Q⎞ 1 Q =QToT 1 ⎛ QToT

2
⎞ 1 ⎛ QToT
2
⎞
WToT = ∫ dW = ∫ ⎜ ⎟ = ∫ = ⎜ ⎟ Thus: WToT = ⎜ ⎟ but: QToT = C ΔV
dQ QdQ
Q =0
⎝C ⎠ C Q = 0 2⎝ C ⎠ 2⎝ C ⎠
1 1 ⎛ Q2 ⎞ 1
constant ∴ WToT = C ΔV 2 = ⎜ ⎟ = Q ΔV
2 2⎝ C ⎠ 2

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

2005 - 2008. All rights reserved.